From 4dfbb9d8c6cbfc32faa5c71145bd2a43e1f8237c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Oct 2006 01:45:14 -0400 Subject: Lockdep: add lockdep_set_class_and_subclass() and lockdep_set_subclass() This annotation makes it possible to assign a subclass on lock init. This annotation is meant to reduce the _nested() annotations by assigning a default subclass. One could do without this annotation and rely on lockdep_set_class() exclusively, but that would require a manual stack of struct lock_class_key objects. Signed-off-by: Peter Zijlstra Signed-off-by: Dmitry Torokhov --- include/linux/lockdep.h | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 1314ca0f29be..14fec2a23b2e 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -202,7 +202,7 @@ extern int lockdep_internal(void); */ extern void lockdep_init_map(struct lockdep_map *lock, const char *name, - struct lock_class_key *key); + struct lock_class_key *key, int subclass); /* * Reinitialize a lock key - for cases where there is special locking or @@ -211,9 +211,14 @@ extern void lockdep_init_map(struct lockdep_map *lock, const char *name, * or they are too narrow (they suffer from a false class-split): */ #define lockdep_set_class(lock, key) \ - lockdep_init_map(&(lock)->dep_map, #key, key) + lockdep_init_map(&(lock)->dep_map, #key, key, 0) #define lockdep_set_class_and_name(lock, key, name) \ - lockdep_init_map(&(lock)->dep_map, name, key) + lockdep_init_map(&(lock)->dep_map, name, key, 0) +#define lockdep_set_class_and_subclass(lock, key, sub) \ + lockdep_init_map(&(lock)->dep_map, #key, key, sub) +#define lockdep_set_subclass(lock, sub) \ + lockdep_init_map(&(lock)->dep_map, #lock, \ + (lock)->dep_map.key, sub) /* * Acquire a lock. @@ -257,10 +262,12 @@ static inline int lockdep_internal(void) # define lock_release(l, n, i) do { } while (0) # define lockdep_init() do { } while (0) # define lockdep_info() do { } while (0) -# define lockdep_init_map(lock, name, key) do { (void)(key); } while (0) +# define lockdep_init_map(lock, name, key, sub) do { (void)(key); } while (0) # define lockdep_set_class(lock, key) do { (void)(key); } while (0) # define lockdep_set_class_and_name(lock, key, name) \ do { (void)(key); } while (0) +#define lockdep_set_class_and_subclass(lock, key, sub) \ + do { (void)(key); } while (0) # define INIT_LOCKDEP # define lockdep_reset() do { debug_locks = 1; } while (0) # define lockdep_free_key_range(start, size) do { } while (0) -- cgit v1.2.3 From 88aa0103e408616e433c209e80169ab8d6eda99e Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Wed, 11 Oct 2006 01:45:31 -0400 Subject: Input: serio - add lockdep annotations Signed-off-by: Jiri Kosina Acked-by: Peter Zijlstra Signed-off-by: Dmitry Torokhov --- drivers/input/serio/libps2.c | 3 ++- drivers/input/serio/serio.c | 6 +++++- include/linux/serio.h | 1 + 3 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index dcb16b5cbec0..e5b1b60757bb 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -189,7 +189,7 @@ int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) return -1; } - mutex_lock_nested(&ps2dev->cmd_mutex, SINGLE_DEPTH_NESTING); + mutex_lock(&ps2dev->cmd_mutex); serio_pause_rx(ps2dev->serio); ps2dev->flags = command == PS2_CMD_GETID ? PS2_FLAG_WAITID : 0; @@ -296,6 +296,7 @@ EXPORT_SYMBOL(ps2_schedule_command); void ps2_init(struct ps2dev *ps2dev, struct serio *serio) { mutex_init(&ps2dev->cmd_mutex); + lockdep_set_subclass(&ps2dev->cmd_mutex, serio->depth); init_waitqueue_head(&ps2dev->wait); ps2dev->serio = serio; } diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 960fae3c3cea..480fdc5d20b3 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -538,8 +538,12 @@ static void serio_init_port(struct serio *serio) "serio%ld", (long)atomic_inc_return(&serio_no) - 1); serio->dev.bus = &serio_bus; serio->dev.release = serio_release_port; - if (serio->parent) + if (serio->parent) { serio->dev.parent = &serio->parent->dev; + serio->depth = serio->parent->depth + 1; + } else + serio->depth = 0; + lockdep_set_subclass(&serio->lock, serio->depth); } /* diff --git a/include/linux/serio.h b/include/linux/serio.h index 3a697cc6ecae..b99c5ca9708d 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -41,6 +41,7 @@ struct serio { void (*stop)(struct serio *); struct serio *parent, *child; + unsigned int depth; /* level of nesting in serio hierarchy */ struct serio_driver *drv; /* accessed from interrupt, must be protected by serio->lock and serio->sem */ struct mutex drv_mutex; /* protects serio->drv so attributes can pin driver */ -- cgit v1.2.3 From 07646e217f473a3e6213f8228336a9046833d6aa Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 11 Oct 2006 23:45:23 -0400 Subject: Lockdep: fix compile error in drivers/input/serio/serio.c lockdep_set_subclass() was missing in !LOCKDEP case Signed-off-by: Andrew Morton Signed-off-by: Dmitry Torokhov --- include/linux/lockdep.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 14fec2a23b2e..819f08f1310d 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -268,6 +268,8 @@ static inline int lockdep_internal(void) do { (void)(key); } while (0) #define lockdep_set_class_and_subclass(lock, key, sub) \ do { (void)(key); } while (0) +#define lockdep_set_subclass(lock, sub) do { } while (0) + # define INIT_LOCKDEP # define lockdep_reset() do { debug_locks = 1; } while (0) # define lockdep_free_key_range(start, size) do { } while (0) -- cgit v1.2.3 From aaa248f6c9c81b2683db7dbb0689cd5ed1c86d88 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 17 Oct 2006 00:09:42 -0700 Subject: [PATCH] rename net_random to random32 Make net_random() more widely available by calling it random32 akpm: hopefully this will permit the removal of carta_random32. That needs confirmation from Stephane - this code looks somewhat more computationally expensive, and has a different (ie: callee-stateful) interface. [akpm@osdl.org: lots of build fixes, cleanups] Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller Cc: Stephane Eranian Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/net.h | 7 +-- include/linux/random.h | 3 ++ lib/Makefile | 2 +- lib/random32.c | 142 +++++++++++++++++++++++++++++++++++++++++++++++++ net/core/dev.c | 2 - net/core/utils.c | 116 ---------------------------------------- 6 files changed, 150 insertions(+), 122 deletions(-) create mode 100644 lib/random32.c (limited to 'include/linux') diff --git a/include/linux/net.h b/include/linux/net.h index c257f716e00f..15c733b816f0 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -19,6 +19,7 @@ #define _LINUX_NET_H #include +#include #include struct poll_table_struct; @@ -193,9 +194,9 @@ extern int sock_map_fd(struct socket *sock); extern struct socket *sockfd_lookup(int fd, int *err); #define sockfd_put(sock) fput(sock->file) extern int net_ratelimit(void); -extern unsigned long net_random(void); -extern void net_srandom(unsigned long); -extern void net_random_init(void); + +#define net_random() random32() +#define net_srandom(seed) srandom32(seed) extern int kernel_sendmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, size_t num, size_t len); diff --git a/include/linux/random.h b/include/linux/random.h index 5d6456bcdeba..0248b30e306d 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -69,6 +69,9 @@ extern struct file_operations random_fops, urandom_fops; unsigned int get_random_int(void); unsigned long randomize_range(unsigned long start, unsigned long end, unsigned long len); +u32 random32(void); +void srandom32(u32 seed); + #endif /* __KERNEL___ */ #endif /* _LINUX_RANDOM_H */ diff --git a/lib/Makefile b/lib/Makefile index 59070dbfbeb4..4b8052f6a7b0 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -12,7 +12,7 @@ lib-$(CONFIG_SMP) += cpumask.o lib-y += kobject.o kref.o kobject_uevent.o klist.o -obj-y += sort.o parser.o halfmd4.o iomap_copy.o debug_locks.o +obj-y += sort.o parser.o halfmd4.o iomap_copy.o debug_locks.o random32.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) CFLAGS_kobject.o += -DDEBUG diff --git a/lib/random32.c b/lib/random32.c new file mode 100644 index 000000000000..4a15ce51cea7 --- /dev/null +++ b/lib/random32.c @@ -0,0 +1,142 @@ +/* + This is a maximally equidistributed combined Tausworthe generator + based on code from GNU Scientific Library 1.5 (30 Jun 2004) + + x_n = (s1_n ^ s2_n ^ s3_n) + + s1_{n+1} = (((s1_n & 4294967294) <<12) ^ (((s1_n <<13) ^ s1_n) >>19)) + s2_{n+1} = (((s2_n & 4294967288) << 4) ^ (((s2_n << 2) ^ s2_n) >>25)) + s3_{n+1} = (((s3_n & 4294967280) <<17) ^ (((s3_n << 3) ^ s3_n) >>11)) + + The period of this generator is about 2^88. + + From: P. L'Ecuyer, "Maximally Equidistributed Combined Tausworthe + Generators", Mathematics of Computation, 65, 213 (1996), 203--213. + + This is available on the net from L'Ecuyer's home page, + + http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme.ps + ftp://ftp.iro.umontreal.ca/pub/simulation/lecuyer/papers/tausme.ps + + There is an erratum in the paper "Tables of Maximally + Equidistributed Combined LFSR Generators", Mathematics of + Computation, 68, 225 (1999), 261--269: + http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme2.ps + + ... the k_j most significant bits of z_j must be non- + zero, for each j. (Note: this restriction also applies to the + computer code given in [4], but was mistakenly not mentioned in + that paper.) + + This affects the seeding procedure by imposing the requirement + s1 > 1, s2 > 7, s3 > 15. + +*/ + +#include +#include +#include +#include + +struct rnd_state { + u32 s1, s2, s3; +}; + +static DEFINE_PER_CPU(struct rnd_state, net_rand_state); + +static u32 __random32(struct rnd_state *state) +{ +#define TAUSWORTHE(s,a,b,c,d) ((s&c)<>b) + + state->s1 = TAUSWORTHE(state->s1, 13, 19, 4294967294UL, 12); + state->s2 = TAUSWORTHE(state->s2, 2, 25, 4294967288UL, 4); + state->s3 = TAUSWORTHE(state->s3, 3, 11, 4294967280UL, 17); + + return (state->s1 ^ state->s2 ^ state->s3); +} + +static void __set_random32(struct rnd_state *state, unsigned long s) +{ + if (s == 0) + s = 1; /* default seed is 1 */ + +#define LCG(n) (69069 * n) + state->s1 = LCG(s); + state->s2 = LCG(state->s1); + state->s3 = LCG(state->s2); + + /* "warm it up" */ + __random32(state); + __random32(state); + __random32(state); + __random32(state); + __random32(state); + __random32(state); +} + +/** + * random32 - pseudo random number generator + * + * A 32 bit pseudo-random number is generated using a fast + * algorithm suitable for simulation. This algorithm is NOT + * considered safe for cryptographic use. + */ +u32 random32(void) +{ + unsigned long r; + struct rnd_state *state = &get_cpu_var(net_rand_state); + r = __random32(state); + put_cpu_var(state); + return r; +} +EXPORT_SYMBOL(random32); + +/** + * srandom32 - add entropy to pseudo random number generator + * @seed: seed value + * + * Add some additional seeding to the random32() pool. + * Note: this pool is per cpu so it only affects current CPU. + */ +void srandom32(u32 entropy) +{ + struct rnd_state *state = &get_cpu_var(net_rand_state); + __set_random32(state, state->s1 ^ entropy); + put_cpu_var(state); +} +EXPORT_SYMBOL(srandom32); + +/* + * Generate some initially weak seeding values to allow + * to start the random32() engine. + */ +static int __init random32_init(void) +{ + int i; + + for_each_possible_cpu(i) { + struct rnd_state *state = &per_cpu(net_rand_state,i); + __set_random32(state, i + jiffies); + } + return 0; +} +core_initcall(random32_init); + +/* + * Generate better values after random number generator + * is fully initalized. + */ +static int __init random32_reseed(void) +{ + int i; + unsigned long seed; + + for_each_possible_cpu(i) { + struct rnd_state *state = &per_cpu(net_rand_state,i); + + get_random_bytes(&seed, sizeof(seed)); + __set_random32(state, seed); + } + return 0; +} +late_initcall(random32_reseed); diff --git a/net/core/dev.c b/net/core/dev.c index 4d891beab138..81c426adcd1e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3502,8 +3502,6 @@ static int __init net_dev_init(void) BUG_ON(!dev_boot_phase); - net_random_init(); - if (dev_proc_init()) goto out; diff --git a/net/core/utils.c b/net/core/utils.c index 94c5d761c830..d93fe64f6693 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -30,119 +30,6 @@ #include #include -/* - This is a maximally equidistributed combined Tausworthe generator - based on code from GNU Scientific Library 1.5 (30 Jun 2004) - - x_n = (s1_n ^ s2_n ^ s3_n) - - s1_{n+1} = (((s1_n & 4294967294) <<12) ^ (((s1_n <<13) ^ s1_n) >>19)) - s2_{n+1} = (((s2_n & 4294967288) << 4) ^ (((s2_n << 2) ^ s2_n) >>25)) - s3_{n+1} = (((s3_n & 4294967280) <<17) ^ (((s3_n << 3) ^ s3_n) >>11)) - - The period of this generator is about 2^88. - - From: P. L'Ecuyer, "Maximally Equidistributed Combined Tausworthe - Generators", Mathematics of Computation, 65, 213 (1996), 203--213. - - This is available on the net from L'Ecuyer's home page, - - http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme.ps - ftp://ftp.iro.umontreal.ca/pub/simulation/lecuyer/papers/tausme.ps - - There is an erratum in the paper "Tables of Maximally - Equidistributed Combined LFSR Generators", Mathematics of - Computation, 68, 225 (1999), 261--269: - http://www.iro.umontreal.ca/~lecuyer/myftp/papers/tausme2.ps - - ... the k_j most significant bits of z_j must be non- - zero, for each j. (Note: this restriction also applies to the - computer code given in [4], but was mistakenly not mentioned in - that paper.) - - This affects the seeding procedure by imposing the requirement - s1 > 1, s2 > 7, s3 > 15. - -*/ -struct nrnd_state { - u32 s1, s2, s3; -}; - -static DEFINE_PER_CPU(struct nrnd_state, net_rand_state); - -static u32 __net_random(struct nrnd_state *state) -{ -#define TAUSWORTHE(s,a,b,c,d) ((s&c)<>b) - - state->s1 = TAUSWORTHE(state->s1, 13, 19, 4294967294UL, 12); - state->s2 = TAUSWORTHE(state->s2, 2, 25, 4294967288UL, 4); - state->s3 = TAUSWORTHE(state->s3, 3, 11, 4294967280UL, 17); - - return (state->s1 ^ state->s2 ^ state->s3); -} - -static void __net_srandom(struct nrnd_state *state, unsigned long s) -{ - if (s == 0) - s = 1; /* default seed is 1 */ - -#define LCG(n) (69069 * n) - state->s1 = LCG(s); - state->s2 = LCG(state->s1); - state->s3 = LCG(state->s2); - - /* "warm it up" */ - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); - __net_random(state); -} - - -unsigned long net_random(void) -{ - unsigned long r; - struct nrnd_state *state = &get_cpu_var(net_rand_state); - r = __net_random(state); - put_cpu_var(state); - return r; -} - - -void net_srandom(unsigned long entropy) -{ - struct nrnd_state *state = &get_cpu_var(net_rand_state); - __net_srandom(state, state->s1^entropy); - put_cpu_var(state); -} - -void __init net_random_init(void) -{ - int i; - - for_each_possible_cpu(i) { - struct nrnd_state *state = &per_cpu(net_rand_state,i); - __net_srandom(state, i+jiffies); - } -} - -static int net_random_reseed(void) -{ - int i; - unsigned long seed; - - for_each_possible_cpu(i) { - struct nrnd_state *state = &per_cpu(net_rand_state,i); - - get_random_bytes(&seed, sizeof(seed)); - __net_srandom(state, seed); - } - return 0; -} -late_initcall(net_random_reseed); - int net_msg_cost = 5*HZ; int net_msg_burst = 10; @@ -153,10 +40,7 @@ int net_ratelimit(void) { return __printk_ratelimit(net_msg_cost, net_msg_burst); } - -EXPORT_SYMBOL(net_random); EXPORT_SYMBOL(net_ratelimit); -EXPORT_SYMBOL(net_srandom); /* * Convert an ASCII string to binary IP. -- cgit v1.2.3 From a460e745e8f9c75a0525ff94154a0629f9d3e05d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 17 Oct 2006 00:10:03 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming Introduce desc->name and eliminate the handle_irq_name() hack. Add set_irq_chip_and_handler_name() to set the flow type and name at once. Signed-off-by: Ingo Molnar Acked-by: Thomas Gleixner Cc: "Eric W. Biederman" Cc: Matthew Wilcox Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/i8259.c | 7 ++++--- arch/i386/kernel/io_apic.c | 17 ++++++++++------- arch/i386/kernel/irq.c | 2 +- arch/x86_64/kernel/i8259.c | 7 ++++--- arch/x86_64/kernel/io_apic.c | 15 ++++++++------- arch/x86_64/kernel/irq.c | 2 +- include/linux/irq.h | 22 ++++++++++------------ kernel/irq/chip.c | 33 +++++++++------------------------ 8 files changed, 47 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c index d53eafb6daa7..62996cd17084 100644 --- a/arch/i386/kernel/i8259.c +++ b/arch/i386/kernel/i8259.c @@ -113,7 +113,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c index 0612a33bb896..c4ef801b765b 100644 --- a/arch/x86_64/kernel/i8259.c +++ b/arch/x86_64/kernel/i8259.c @@ -178,7 +178,8 @@ void make_8259A_irq(unsigned int irq) { disable_irq_nosync(irq); io_apic_irqs &= ~(1<name); - seq_printf(p, "-%s", handle_irq_name(irq_desc[i].handle_irq)); + seq_printf(p, "-%-8s", irq_desc[i].name); seq_printf(p, " %s", action->name); for (action=action->next; action; action = action->next) diff --git a/include/linux/irq.h b/include/linux/irq.h index c64f3cc7e870..775f5a7da493 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -141,6 +141,7 @@ struct irq_chip { * @pending_mask: pending rebalanced interrupts * @dir: /proc/irq/ procfs entry * @affinity_entry: /proc/irq/smp_affinity procfs entry on SMP + * @name: flow handler name for /proc/interrupts output * * Pad this out to 32 bytes for cache and indexing reasons. */ @@ -165,8 +166,9 @@ struct irq_desc { cpumask_t pending_mask; #endif #ifdef CONFIG_PROC_FS - struct proc_dir_entry *dir; + struct proc_dir_entry *dir; #endif + const char *name; } ____cacheline_aligned; extern struct irq_desc irq_desc[NR_IRQS]; @@ -271,12 +273,6 @@ extern void fastcall handle_simple_irq(unsigned int irq, struct irq_desc *desc); extern void fastcall handle_percpu_irq(unsigned int irq, struct irq_desc *desc); extern void fastcall handle_bad_irq(unsigned int irq, struct irq_desc *desc); -/* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -extern const char *handle_irq_name(irq_flow_handler_t handle); - /* * Monolithic do_IRQ implementation. * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly) @@ -326,10 +322,12 @@ extern struct irq_chip no_irq_chip; extern struct irq_chip dummy_irq_chip; extern void -set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, - irq_flow_handler_t handle); +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name); + extern void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name); /* * Set a highlevel flow handler for a given IRQ: @@ -337,7 +335,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained); static inline void set_irq_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } /* @@ -349,7 +347,7 @@ static inline void set_irq_chained_handler(unsigned int irq, irq_flow_handler_t handle) { - __set_irq_handler(irq, handle, 1); + __set_irq_handler(irq, handle, 1, NULL); } /* Handle dynamic irq creation and destruction */ diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index 11c99697acfe..2d0dc3efe813 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -499,7 +499,8 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) #endif /* CONFIG_SMP */ void -__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) +__set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, + const char *name) { struct irq_desc *desc; unsigned long flags; @@ -540,6 +541,7 @@ __set_irq_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained) desc->depth = 1; } desc->handle_irq = handle; + desc->name = name; if (handle != handle_bad_irq && is_chained) { desc->status &= ~IRQ_DISABLED; @@ -555,30 +557,13 @@ set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle) { set_irq_chip(irq, chip); - __set_irq_handler(irq, handle, 0); + __set_irq_handler(irq, handle, 0, NULL); } -/* - * Get a descriptive string for the highlevel handler, for - * /proc/interrupts output: - */ -const char * -handle_irq_name(irq_flow_handler_t handle) +void +set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle, const char *name) { - if (handle == handle_level_irq) - return "level "; - if (handle == handle_fasteoi_irq) - return "fasteoi"; - if (handle == handle_edge_irq) - return "edge "; - if (handle == handle_simple_irq) - return "simple "; -#ifdef CONFIG_SMP - if (handle == handle_percpu_irq) - return "percpu "; -#endif - if (handle == handle_bad_irq) - return "bad "; - - return NULL; + set_irq_chip(irq, chip); + __set_irq_handler(irq, handle, 0, name); } -- cgit v1.2.3 From 7762f5a0b709b415fda132258ad37b9f2a1db994 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 17 Oct 2006 00:10:07 -0700 Subject: [PATCH] document i_size_write locking rules Unless someone reads the documentation for write_seqcount_{begin,end} it is not obvious, that i_size_write() needs locking. Especially, that lack of such locking can result in a system hang. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 34406ed467c3..661c7c572149 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -656,7 +656,11 @@ static inline loff_t i_size_read(struct inode *inode) #endif } - +/* + * NOTE: unlike i_size_read(), i_size_write() does need locking around it + * (normally i_mutex), otherwise on 32bit/SMP an update of i_size_seqcount + * can be lost, resulting in subsequent i_size_read() calls spinning forever. + */ static inline void i_size_write(struct inode *inode, loff_t i_size) { #if BITS_PER_LONG==32 && defined(CONFIG_SMP) -- cgit v1.2.3 From d343fce148a4eee24a907a05c4101d3268045aae Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 17 Oct 2006 00:10:18 -0700 Subject: [PATCH] knfsd: Allow lockd to drop replies as appropriate It is possible for the ->fopen callback from lockd into nfsd to find that an answer cannot be given straight away (an upcall is needed) and so the request has to be 'dropped', to be retried later. That error status is not currently propagated back. So: Change nlm_fopen to return nlm error codes (rather than a private protocol) and define a new nlm_drop_reply code. Cause nlm_drop_reply to cause the rpc request to get rpc_drop_reply when this error comes back. Cause svc_process to drop a request which returns a status of rpc_drop_reply. [akpm@osdl.org: fix warning storm] Cc: Marc Eshel Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/svc4proc.c | 12 ++++++------ fs/lockd/svcproc.c | 16 +++++++++------- fs/lockd/svcsubs.c | 6 ------ fs/nfsd/lockd.c | 14 ++++++++------ include/linux/lockd/bind.h | 5 +++++ include/linux/lockd/xdr.h | 4 ++++ include/linux/sunrpc/msg_prot.h | 4 +++- include/linux/sunrpc/xdr.h | 1 + net/sunrpc/svc.c | 5 +++++ 9 files changed, 41 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index fa370f6eb07b..399ad11b97be 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -96,7 +96,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = nlmsvc_testlock(file, &argp->lock, &resp->lock); @@ -126,7 +126,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -169,7 +169,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = nlmsvc_cancel_blocked(file, &argp->lock); @@ -202,7 +202,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = nlmsvc_unlock(file, &argp->lock); @@ -339,7 +339,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = nlmsvc_share_file(host, file, argp); @@ -372,7 +372,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to lock the file */ resp->status = nlmsvc_unshare_file(host, file, argp); diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 75b2c81bcb93..6a931f4ab75c 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -59,7 +59,7 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error; + u32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) @@ -88,6 +88,8 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, no_locks: if (host) nlm_release_host(host); + if (error) + return error; return nlm_lck_denied_nolocks; } @@ -122,7 +124,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = cast_status(nlmsvc_testlock(file, &argp->lock, &resp->lock)); @@ -153,7 +155,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -196,7 +198,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = cast_status(nlmsvc_cancel_blocked(file, &argp->lock)); @@ -229,7 +231,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = cast_status(nlmsvc_unlock(file, &argp->lock)); @@ -368,7 +370,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = cast_status(nlmsvc_share_file(host, file, argp)); @@ -401,7 +403,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to unshare the file */ resp->status = cast_status(nlmsvc_unshare_file(host, file, argp)); diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index c5f9113cdc70..7dac96e6c82c 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -135,12 +135,6 @@ out_unlock: out_free: kfree(file); -#ifdef CONFIG_LOCKD_V4 - if (nfserr == 1) - nfserr = nlm4_stale_fh; - else -#endif - nfserr = nlm_lck_denied; goto out_unlock; } diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 7b889ff15ae6..9b9e7e127c03 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -39,18 +39,20 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) fh_put(&fh); rqstp->rq_client = NULL; exp_readunlock(); - /* nlm and nfsd don't share error codes. - * we invent: 0 = no error - * 1 = stale file handle - * 2 = other error + /* We return nlm error codes as nlm doesn't know + * about nfsd, but nfsd does know about nlm.. */ switch (nfserr) { case nfs_ok: return 0; + case nfserr_dropit: + return nlm_drop_reply; +#ifdef CONFIG_LOCKD_V4 case nfserr_stale: - return 1; + return nlm4_stale_fh; +#endif default: - return 2; + return nlm_lck_denied; } } diff --git a/include/linux/lockd/bind.h b/include/linux/lockd/bind.h index 81e3a185f951..aa50d89eacd7 100644 --- a/include/linux/lockd/bind.h +++ b/include/linux/lockd/bind.h @@ -10,6 +10,11 @@ #define LINUX_LOCKD_BIND_H #include +/* need xdr-encoded error codes too, so... */ +#include +#ifdef CONFIG_LOCKD_V4 +#include +#endif /* Dummy declarations */ struct svc_rqst; diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h index bb0a0f1caa91..66fdae3b490c 100644 --- a/include/linux/lockd/xdr.h +++ b/include/linux/lockd/xdr.h @@ -13,6 +13,8 @@ #include #include +struct svc_rqst; + #define NLM_MAXCOOKIELEN 32 #define NLM_MAXSTRLEN 1024 @@ -22,6 +24,8 @@ #define nlm_lck_blocked __constant_htonl(NLM_LCK_BLOCKED) #define nlm_lck_denied_grace_period __constant_htonl(NLM_LCK_DENIED_GRACE_PERIOD) +#define nlm_drop_reply __constant_htonl(30000) + /* Lock info passed via NLM */ struct nlm_lock { char * caller; diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h index 1e65f2dd80e5..606cb2165232 100644 --- a/include/linux/sunrpc/msg_prot.h +++ b/include/linux/sunrpc/msg_prot.h @@ -56,7 +56,9 @@ enum rpc_accept_stat { RPC_PROG_MISMATCH = 2, RPC_PROC_UNAVAIL = 3, RPC_GARBAGE_ARGS = 4, - RPC_SYSTEM_ERR = 5 + RPC_SYSTEM_ERR = 5, + /* internal use only */ + RPC_DROP_REPLY = 60000, }; enum rpc_reject_stat { diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 953723b09bc6..ac69e5511606 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -74,6 +74,7 @@ struct xdr_buf { #define rpc_proc_unavail __constant_htonl(RPC_PROC_UNAVAIL) #define rpc_garbage_args __constant_htonl(RPC_GARBAGE_ARGS) #define rpc_system_err __constant_htonl(RPC_SYSTEM_ERR) +#define rpc_drop_reply __constant_htonl(RPC_DROP_REPLY) #define rpc_auth_ok __constant_htonl(RPC_AUTH_OK) #define rpc_autherr_badcred __constant_htonl(RPC_AUTH_BADCRED) diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 2807fa0eab40..eb44ec929ca1 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -828,6 +828,11 @@ svc_process(struct svc_rqst *rqstp) *statp = procp->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); /* Encode reply */ + if (*statp == rpc_drop_reply) { + if (procp->pc_release) + procp->pc_release(rqstp, NULL, rqstp->rq_resp); + goto dropit; + } if (*statp == rpc_success && (xdr = procp->pc_encode) && !xdr(rqstp, resv->iov_base+resv->iov_len, rqstp->rq_resp)) { dprintk("svc: failed to encode reply\n"); -- cgit v1.2.3 From 58ff407bee5a55f9c1188a3f9d70ffc79485183c Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 17 Oct 2006 00:10:19 -0700 Subject: [PATCH] Fix IO error reporting on fsync() When IO error happens on metadata buffer, buffer is freed from memory and later fsync() is called, filesystems like ext2 fail to report EIO. We solve the problem by introducing a pointer to associated address space into the buffer_head. When a buffer is removed from a list of metadata buffers associated with an address space, IO error is transferred from the buffer to the address space, so that fsync can later report it. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 11 +++++++++-- include/linux/buffer_head.h | 2 ++ 2 files changed, 11 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/buffer.c b/fs/buffer.c index f65ef8821c73..35527dca1dbc 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -452,6 +452,7 @@ static void end_buffer_async_write(struct buffer_head *bh, int uptodate) bdevname(bh->b_bdev, b)); } set_bit(AS_EIO, &page->mapping->flags); + set_buffer_write_io_error(bh); clear_buffer_uptodate(bh); SetPageError(page); } @@ -571,6 +572,10 @@ EXPORT_SYMBOL(mark_buffer_async_write); static inline void __remove_assoc_queue(struct buffer_head *bh) { list_del_init(&bh->b_assoc_buffers); + WARN_ON(!bh->b_assoc_map); + if (buffer_write_io_error(bh)) + set_bit(AS_EIO, &bh->b_assoc_map->flags); + bh->b_assoc_map = NULL; } int inode_has_buffers(struct inode *inode) @@ -669,6 +674,7 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) spin_lock(&buffer_mapping->private_lock); list_move_tail(&bh->b_assoc_buffers, &mapping->private_list); + bh->b_assoc_map = mapping; spin_unlock(&buffer_mapping->private_lock); } } @@ -765,7 +771,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) spin_lock(lock); while (!list_empty(list)) { bh = BH_ENTRY(list->next); - list_del_init(&bh->b_assoc_buffers); + __remove_assoc_queue(bh); if (buffer_dirty(bh) || buffer_locked(bh)) { list_add(&bh->b_assoc_buffers, &tmp); if (buffer_dirty(bh)) { @@ -786,7 +792,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) while (!list_empty(&tmp)) { bh = BH_ENTRY(tmp.prev); - __remove_assoc_queue(bh); + list_del_init(&bh->b_assoc_buffers); get_bh(bh); spin_unlock(lock); wait_on_buffer(bh); @@ -1167,6 +1173,7 @@ void __bforget(struct buffer_head *bh) spin_lock(&buffer_mapping->private_lock); list_del_init(&bh->b_assoc_buffers); + bh->b_assoc_map = NULL; spin_unlock(&buffer_mapping->private_lock); } __brelse(bh); diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 131ffd37e716..5d9fb0e94156 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -69,6 +69,8 @@ struct buffer_head { bh_end_io_t *b_end_io; /* I/O completion */ void *b_private; /* reserved for b_end_io */ struct list_head b_assoc_buffers; /* associated with another mapping */ + struct address_space *b_assoc_map; /* mapping this buffer is + associated with */ atomic_t b_count; /* users using this buffer_head */ }; -- cgit v1.2.3 From 74d919465a93b6c2b928b29a8ed3e5e41adbfa93 Mon Sep 17 00:00:00 2001 From: Ben Collins Date: Wed, 18 Oct 2006 08:55:54 -0400 Subject: [pci_ids] Add Quicknet XJ vendor/device ID's. Signed-off-by: Ben Collins --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f069df245469..f3a168f3c9df 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2351,3 +2351,5 @@ #define PCI_DEVICE_ID_RME_DIGI32_PRO 0x9897 #define PCI_DEVICE_ID_RME_DIGI32_8 0x9898 +#define PCI_VENDOR_ID_QUICKNET 0x15E2 +#define PCI_DEVICE_ID_QUICKNET_XJ 0x0500 -- cgit v1.2.3 From 29f3eb64634cf96903a3cdb56b1f9a80bebad17d Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 16 Oct 2006 16:20:21 -0700 Subject: pci: Additional search functions In order to finish converting to pci_get_* interfaces we need to add a couple of bits of missing functionaility pci_get_bus_and_slot() provides the equivalent to pci_find_slot() (pci_get_slot is already taken as a name for something similar but not the same) pci_get_device_reverse() is the equivalent of pci_find_device_reverse but refcounting Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- drivers/pci/search.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++++++-- include/linux/pci.h | 3 ++- 2 files changed, 72 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/search.c b/drivers/pci/search.c index d529462d1b53..2f13eba5d5ae 100644 --- a/drivers/pci/search.c +++ b/drivers/pci/search.c @@ -139,6 +139,31 @@ struct pci_dev * pci_get_slot(struct pci_bus *bus, unsigned int devfn) return dev; } +/** + * pci_get_bus_and_slot - locate PCI device from a given PCI slot + * @bus: number of PCI bus on which desired PCI device resides + * @devfn: encodes number of PCI slot in which the desired PCI + * device resides and the logical device number within that slot + * in case of multi-function devices. + * + * Given a PCI bus and slot/function number, the desired PCI device + * is located in system global list of PCI devices. If the device + * is found, a pointer to its data structure is returned. If no + * device is found, %NULL is returned. The returned device has its + * reference count bumped by one. + */ + +struct pci_dev * pci_get_bus_and_slot(unsigned int bus, unsigned int devfn) +{ + struct pci_dev *dev = NULL; + + while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { + if (dev->bus->number == bus && dev->devfn == devfn) + return dev; + } + return NULL; +} + /** * pci_find_subsys - begin or continue searching for a PCI device by vendor/subvendor/device/subdevice id * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids @@ -274,6 +299,45 @@ pci_get_device(unsigned int vendor, unsigned int device, struct pci_dev *from) return pci_get_subsys(vendor, device, PCI_ANY_ID, PCI_ANY_ID, from); } +/** + * pci_get_device_reverse - begin or continue searching for a PCI device by vendor/device id + * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids + * @device: PCI device id to match, or %PCI_ANY_ID to match all device ids + * @from: Previous PCI device found in search, or %NULL for new search. + * + * Iterates through the list of known PCI devices in the reverse order of + * pci_get_device. + * If a PCI device is found with a matching @vendor and @device, the reference + * count to the device is incremented and a pointer to its device structure + * is returned Otherwise, %NULL is returned. A new search is initiated by + * passing %NULL as the @from argument. Otherwise if @from is not %NULL, + * searches continue from next device on the global list. The reference + * count for @from is always decremented if it is not %NULL. + */ +struct pci_dev * +pci_get_device_reverse(unsigned int vendor, unsigned int device, struct pci_dev *from) +{ + struct list_head *n; + struct pci_dev *dev; + + WARN_ON(in_interrupt()); + down_read(&pci_bus_sem); + n = from ? from->global_list.prev : pci_devices.prev; + + while (n && (n != &pci_devices)) { + dev = pci_dev_g(n); + if ((vendor == PCI_ANY_ID || dev->vendor == vendor) && + (device == PCI_ANY_ID || dev->device == device)) + goto exit; + n = n->prev; + } + dev = NULL; +exit: + dev = pci_dev_get(dev); + up_read(&pci_bus_sem); + pci_dev_put(from); + return dev; +} /** * pci_find_device_reverse - begin or continue searching for a PCI device by vendor/device id @@ -382,12 +446,16 @@ exit: } EXPORT_SYMBOL(pci_dev_present); -EXPORT_SYMBOL(pci_find_bus); -EXPORT_SYMBOL(pci_find_next_bus); EXPORT_SYMBOL(pci_find_device); EXPORT_SYMBOL(pci_find_device_reverse); EXPORT_SYMBOL(pci_find_slot); +/* For boot time work */ +EXPORT_SYMBOL(pci_find_bus); +EXPORT_SYMBOL(pci_find_next_bus); +/* For everyone */ EXPORT_SYMBOL(pci_get_device); +EXPORT_SYMBOL(pci_get_device_reverse); EXPORT_SYMBOL(pci_get_subsys); EXPORT_SYMBOL(pci_get_slot); +EXPORT_SYMBOL(pci_get_bus_and_slot); EXPORT_SYMBOL(pci_get_class); diff --git a/include/linux/pci.h b/include/linux/pci.h index 5c604f5fad67..09bf88fc80c5 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -452,13 +452,14 @@ struct pci_dev *pci_find_slot (unsigned int bus, unsigned int devfn); int pci_find_capability (struct pci_dev *dev, int cap); int pci_find_next_capability (struct pci_dev *dev, u8 pos, int cap); int pci_find_ext_capability (struct pci_dev *dev, int cap); -struct pci_bus * pci_find_next_bus(const struct pci_bus *from); +struct pci_bus *pci_find_next_bus(const struct pci_bus *from); struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from); struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, unsigned int ss_vendor, unsigned int ss_device, struct pci_dev *from); struct pci_dev *pci_get_slot (struct pci_bus *bus, unsigned int devfn); +struct pci_dev *pci_get_bus_and_slot (unsigned int bus, unsigned int devfn); struct pci_dev *pci_get_class (unsigned int class, struct pci_dev *from); int pci_dev_present(const struct pci_device_id *ids); -- cgit v1.2.3 From 6b4b78fed47e7380dfe9280b154e8b9bfcd4c86c Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Fri, 29 Sep 2006 15:23:23 -0500 Subject: PCI: optionally sort device lists breadth-first Problem: New Dell PowerEdge servers have 2 embedded ethernet ports, which are labeled NIC1 and NIC2 on the chassis, in the BIOS setup screens, and in the printed documentation. Assuming no other add-in ethernet ports in the system, Linux 2.4 kernels name these eth0 and eth1 respectively. Many people have come to expect this naming. Linux 2.6 kernels name these eth1 and eth0 respectively (backwards from expectations). I also have reports that various Sun and HP servers have similar behavior. Root cause: Linux 2.4 kernels walk the pci_devices list, which happens to be sorted in breadth-first order (or pcbios_find_device order on i386, which most often is breadth-first also). 2.6 kernels have both the pci_devices list and the pci_bus_type.klist_devices list, the latter is what is walked at driver load time to match the pci_id tables; this klist happens to be in depth-first order. On systems where, for physical routing reasons, NIC1 appears on a lower bus number than NIC2, but NIC2's bridge is discovered first in the depth-first ordering, NIC2 will be discovered before NIC1. If the list were sorted breadth-first, NIC1 would be discovered before NIC2. A PowerEdge 1955 system has the following topology which easily exhibits the difference between depth-first and breadth-first device lists. -[0000:00]-+-00.0 Intel Corporation 5000P Chipset Memory Controller Hub +-02.0-[0000:03-08]--+-00.0-[0000:04-07]--+-00.0-[0000:05-06]----00.0-[0000:06]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC2, 2.4 kernel name eth1, 2.6 kernel name eth0) +-1c.0-[0000:01-02]----00.0-[0000:02]----00.0 Broadcom Corporation NetXtreme II BCM5708S Gigabit Ethernet (labeled NIC1, 2.4 kernel name eth0, 2.6 kernel name eth1) Other factors, such as device driver load order and the presence of PCI slots at various points in the bus hierarchy further complicate this problem; I'm not trying to solve those here, just restore the device order, and thus basic behavior, that 2.4 kernels had. Solution: The solution can come in multiple steps. Suggested fix #1: kernel Patch below optionally sorts the two device lists into breadth-first ordering to maintain compatibility with 2.4 kernels. It adds two new command line options: pci=bfsort pci=nobfsort to force the sort order, or not, as you wish. It also adds DMI checks for the specific Dell systems which exhibit "backwards" ordering, to make them "right". Suggested fix #2: udev rules from userland Many people also have the expectation that embedded NICs are always discovered before add-in NICs (which this patch does not try to do). Using the PCI IRQ Routing Table provided by system BIOS, it's easy to determine which PCI devices are embedded, or if add-in, which PCI slot they're in. I'm working on a tool that would allow udev to name ethernet devices in ascending embedded, slot 1 .. slot N order, subsort by PCI bus/dev/fn breadth-first. It'll be possible to use it independent of udev as well for those distributions that don't use udev in their installers. Suggested fix #3: system board routing rules One can constrain the system board layout to put NIC1 ahead of NIC2 regardless of breadth-first or depth-first discovery order. This adds a significant level of complexity to board routing, and may not be possible in all instances (witness the above systems from several major manufacturers). I don't want to encourage this particular train of thought too far, at the expense of not doing #1 or #2 above. Feedback appreciated. Patch tested on a Dell PowerEdge 1955 blade with 2.6.18. You'll also note I took some liberty and temporarily break the klist abstraction to simplify and speed up the sort algorithm. I think that's both safe and appropriate in this instance. Signed-off-by: Matt Domsch Signed-off-by: Greg Kroah-Hartman --- Documentation/kernel-parameters.txt | 5 ++ arch/i386/pci/common.c | 59 +++++++++++++++++++++++- arch/i386/pci/pci.h | 7 +++ drivers/pci/probe.c | 92 +++++++++++++++++++++++++++++++++++++ include/linux/pci.h | 1 + 5 files changed, 162 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index ff571f9298e0..dd00fd556a60 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1231,6 +1231,11 @@ and is between 256 and 4096 characters. It is defined in the file machine check when some devices' config space is read. But various workarounds are disabled and some IOMMU drivers will not work. + bfsort Sort PCI devices into breadth-first order. + This sorting is done to get a device + order compatible with older (<= 2.4) kernels. + nobfsort Don't sort PCI devices into breadth-first order. + pcmv= [HW,PCMCIA] BadgePAD 4 pd. [PARIDE] diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c index 68bce194e688..6d5ace845e44 100644 --- a/arch/i386/pci/common.c +++ b/arch/i386/pci/common.c @@ -20,6 +20,7 @@ unsigned int pci_probe = PCI_PROBE_BIOS | PCI_PROBE_CONF1 | PCI_PROBE_CONF2 | PCI_PROBE_MMCONF; +int pci_bf_sort; int pci_routeirq; int pcibios_last_bus = -1; unsigned long pirq_table_addr; @@ -117,6 +118,20 @@ void __devinit pcibios_fixup_bus(struct pci_bus *b) pci_read_bridge_bases(b); } +/* + * Only use DMI information to set this if nothing was passed + * on the kernel command line (which was parsed earlier). + */ + +static int __devinit set_bf_sort(struct dmi_system_id *d) +{ + if (pci_bf_sort == pci_bf_sort_default) { + pci_bf_sort = pci_dmi_bf; + printk(KERN_INFO "PCI: %s detected, enabling pci=bfsort.\n", d->ident); + } + return 0; +} + /* * Enable renumbering of PCI bus# ranges to reach all PCI busses (Cardbus) */ @@ -130,11 +145,11 @@ static int __devinit assign_all_busses(struct dmi_system_id *d) } #endif +static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { +#ifdef __i386__ /* * Laptops which need pci=assign-busses to see Cardbus cards */ -static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { -#ifdef __i386__ { .callback = assign_all_busses, .ident = "Samsung X20 Laptop", @@ -144,6 +159,38 @@ static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = { }, }, #endif /* __i386__ */ + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1950"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 1955", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 1955"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2900", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2900"), + }, + }, + { + .callback = set_bf_sort, + .ident = "Dell PowerEdge 2950", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell"), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge 2950"), + }, + }, {} }; @@ -189,6 +236,8 @@ static int __init pcibios_init(void) pcibios_resource_survey(); + if (pci_bf_sort >= pci_force_bf) + pci_sort_breadthfirst(); #ifdef CONFIG_PCI_BIOS if ((pci_probe & PCI_BIOS_SORT) && !(pci_probe & PCI_NO_SORT)) pcibios_sort(); @@ -203,6 +252,12 @@ char * __devinit pcibios_setup(char *str) if (!strcmp(str, "off")) { pci_probe = 0; return NULL; + } else if (!strcmp(str, "bfsort")) { + pci_bf_sort = pci_force_bf; + return NULL; + } else if (!strcmp(str, "nobfsort")) { + pci_bf_sort = pci_force_nobf; + return NULL; } #ifdef CONFIG_PCI_BIOS else if (!strcmp(str, "bios")) { diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h index 1814f74569c6..ad065cebd7b9 100644 --- a/arch/i386/pci/pci.h +++ b/arch/i386/pci/pci.h @@ -30,6 +30,13 @@ extern unsigned int pci_probe; extern unsigned long pirq_table_addr; +enum pci_bf_sort_state { + pci_bf_sort_default, + pci_force_nobf, + pci_force_bf, + pci_dmi_bf, +}; + /* pci-i386.c */ extern unsigned int pcibios_max_latency; diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index a3b0a5eb5054..e159d6604494 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1067,3 +1067,95 @@ EXPORT_SYMBOL(pci_scan_bridge); EXPORT_SYMBOL(pci_scan_single_device); EXPORT_SYMBOL_GPL(pci_scan_child_bus); #endif + +static int __init pci_sort_bf_cmp(const struct pci_dev *a, const struct pci_dev *b) +{ + if (pci_domain_nr(a->bus) < pci_domain_nr(b->bus)) return -1; + else if (pci_domain_nr(a->bus) > pci_domain_nr(b->bus)) return 1; + + if (a->bus->number < b->bus->number) return -1; + else if (a->bus->number > b->bus->number) return 1; + + if (a->devfn < b->devfn) return -1; + else if (a->devfn > b->devfn) return 1; + + return 0; +} + +/* + * Yes, this forcably breaks the klist abstraction temporarily. It + * just wants to sort the klist, not change reference counts and + * take/drop locks rapidly in the process. It does all this while + * holding the lock for the list, so objects can't otherwise be + * added/removed while we're swizzling. + */ +static void __init pci_insertion_sort_klist(struct pci_dev *a, struct list_head *list) +{ + struct list_head *pos; + struct klist_node *n; + struct device *dev; + struct pci_dev *b; + + list_for_each(pos, list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + b = to_pci_dev(dev); + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->dev.knode_bus.n_node, &b->dev.knode_bus.n_node); + return; + } + } + list_move_tail(&a->dev.knode_bus.n_node, list); +} + +static void __init pci_sort_breadthfirst_klist(void) +{ + LIST_HEAD(sorted_devices); + struct list_head *pos, *tmp; + struct klist_node *n; + struct device *dev; + struct pci_dev *pdev; + + spin_lock(&pci_bus_type.klist_devices.k_lock); + list_for_each_safe(pos, tmp, &pci_bus_type.klist_devices.k_list) { + n = container_of(pos, struct klist_node, n_node); + dev = container_of(n, struct device, knode_bus); + pdev = to_pci_dev(dev); + pci_insertion_sort_klist(pdev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_bus_type.klist_devices.k_list); + spin_unlock(&pci_bus_type.klist_devices.k_lock); +} + +static void __init pci_insertion_sort_devices(struct pci_dev *a, struct list_head *list) +{ + struct pci_dev *b; + + list_for_each_entry(b, list, global_list) { + if (pci_sort_bf_cmp(a, b) <= 0) { + list_move_tail(&a->global_list, &b->global_list); + return; + } + } + list_move_tail(&a->global_list, list); +} + +static void __init pci_sort_breadthfirst_devices(void) +{ + LIST_HEAD(sorted_devices); + struct pci_dev *dev, *tmp; + + down_write(&pci_bus_sem); + list_for_each_entry_safe(dev, tmp, &pci_devices, global_list) { + pci_insertion_sort_devices(dev, &sorted_devices); + } + list_splice(&sorted_devices, &pci_devices); + up_write(&pci_bus_sem); +} + +void __init pci_sort_breadthfirst(void) +{ + pci_sort_breadthfirst_devices(); + pci_sort_breadthfirst_klist(); +} + diff --git a/include/linux/pci.h b/include/linux/pci.h index 09bf88fc80c5..4689e2a699c0 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -443,6 +443,7 @@ extern void pci_remove_bus(struct pci_bus *b); extern void pci_remove_bus_device(struct pci_dev *dev); extern void pci_stop_bus_device(struct pci_dev *dev); void pci_setup_cardbus(struct pci_bus *bus); +extern void pci_sort_breadthfirst(void); /* Generic PCI functions exported to card drivers */ -- cgit v1.2.3 From 7a54f25cef6c763f16c9fd49ae382de162147873 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 13 Oct 2006 20:05:19 -0700 Subject: PCI Hotplug: move pci_hotplug.h to include/linux/ This makes it possible to build pci hotplug drivers outside of the main kernel tree, and Sam keeps telling me to move local header files to their proper places... Signed-off-by: Greg Kroah-Hartman --- drivers/pci/hotplug/acpi_pcihp.c | 2 +- drivers/pci/hotplug/acpiphp.h | 2 +- drivers/pci/hotplug/acpiphp_core.c | 2 +- drivers/pci/hotplug/acpiphp_glue.c | 2 +- drivers/pci/hotplug/acpiphp_ibm.c | 1 - drivers/pci/hotplug/cpci_hotplug_core.c | 2 +- drivers/pci/hotplug/cpci_hotplug_pci.c | 2 +- drivers/pci/hotplug/cpqphp.h | 1 - drivers/pci/hotplug/cpqphp_core.c | 1 + drivers/pci/hotplug/cpqphp_ctrl.c | 1 + drivers/pci/hotplug/cpqphp_nvram.c | 1 + drivers/pci/hotplug/cpqphp_pci.c | 1 + drivers/pci/hotplug/cpqphp_sysfs.c | 1 + drivers/pci/hotplug/fakephp.c | 2 +- drivers/pci/hotplug/ibmphp.h | 2 +- drivers/pci/hotplug/pci_hotplug.h | 236 -------------------------------- drivers/pci/hotplug/pci_hotplug_core.c | 7 +- drivers/pci/hotplug/pciehp.h | 2 +- drivers/pci/hotplug/pcihp_skeleton.c | 2 +- drivers/pci/hotplug/rpadlpar_sysfs.c | 2 +- drivers/pci/hotplug/rpaphp_core.c | 2 +- drivers/pci/hotplug/sgi_hotplug.c | 2 +- drivers/pci/hotplug/shpchp.h | 3 +- include/linux/pci_hotplug.h | 236 ++++++++++++++++++++++++++++++++ 24 files changed, 258 insertions(+), 257 deletions(-) delete mode 100644 drivers/pci/hotplug/pci_hotplug.h create mode 100644 include/linux/pci_hotplug.h (limited to 'include/linux') diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index 51cb9f817c22..270a33cc08f6 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -29,10 +29,10 @@ #include #include #include +#include #include #include #include -#include "pci_hotplug.h" #define MY_NAME "acpi_pcihp" diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h index 7fff07e877c7..59c5b242d86d 100644 --- a/drivers/pci/hotplug/acpiphp.h +++ b/drivers/pci/hotplug/acpiphp.h @@ -38,7 +38,7 @@ #include #include /* for KOBJ_NAME_LEN */ #include -#include "pci_hotplug.h" +#include #define dbg(format, arg...) \ do { \ diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c index e2fef60c2d06..c57d9d5ce84e 100644 --- a/drivers/pci/hotplug/acpiphp_core.c +++ b/drivers/pci/hotplug/acpiphp_core.c @@ -37,10 +37,10 @@ #include #include +#include #include #include #include -#include "pci_hotplug.h" #include "acpiphp.h" #define MY_NAME "acpiphp" diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 83e8e4412de5..c44311ac2fd3 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -45,11 +45,11 @@ #include #include +#include #include #include #include "../pci.h" -#include "pci_hotplug.h" #include "acpiphp.h" static LIST_HEAD(bridge_list); diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c index d0a07d9ab30c..bd40aee10e16 100644 --- a/drivers/pci/hotplug/acpiphp_ibm.c +++ b/drivers/pci/hotplug/acpiphp_ibm.c @@ -35,7 +35,6 @@ #include #include "acpiphp.h" -#include "pci_hotplug.h" #define DRIVER_VERSION "1.0.1" #define DRIVER_AUTHOR "Irene Zubarev , Vernon Mauery " diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c index d06ab4045134..684551559d44 100644 --- a/drivers/pci/hotplug/cpci_hotplug_core.c +++ b/drivers/pci/hotplug/cpci_hotplug_core.c @@ -29,12 +29,12 @@ #include #include #include +#include #include #include #include #include #include -#include "pci_hotplug.h" #include "cpci_hotplug.h" #define DRIVER_AUTHOR "Scott Murray " diff --git a/drivers/pci/hotplug/cpci_hotplug_pci.c b/drivers/pci/hotplug/cpci_hotplug_pci.c index 4afcaffd031c..7b1beaad2752 100644 --- a/drivers/pci/hotplug/cpci_hotplug_pci.c +++ b/drivers/pci/hotplug/cpci_hotplug_pci.c @@ -26,9 +26,9 @@ #include #include #include +#include #include #include "../pci.h" -#include "pci_hotplug.h" #include "cpci_hotplug.h" #define MY_NAME "cpci_hotplug" diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h index ea040c32f47d..298ad7f3f4f4 100644 --- a/drivers/pci/hotplug/cpqphp.h +++ b/drivers/pci/hotplug/cpqphp.h @@ -28,7 +28,6 @@ #ifndef _CPQPHP_H #define _CPQPHP_H -#include "pci_hotplug.h" #include #include /* for read? and write? functions */ #include /* for delays */ diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c index 1fc259913b68..5617cfdadc5c 100644 --- a/drivers/pci/hotplug/cpqphp_core.c +++ b/drivers/pci/hotplug/cpqphp_core.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c index 3ec2ad7db49a..79ff6b4de3a6 100644 --- a/drivers/pci/hotplug/cpqphp_ctrl.c +++ b/drivers/pci/hotplug/cpqphp_ctrl.c @@ -36,6 +36,7 @@ #include #include #include +#include #include "cpqphp.h" static u32 configure_new_device(struct controller* ctrl, struct pci_func *func, diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c index cf0878917537..298a6cfd8406 100644 --- a/drivers/pci/hotplug/cpqphp_nvram.c +++ b/drivers/pci/hotplug/cpqphp_nvram.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include "cpqphp.h" diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c index 0d9688952f4a..fc7c74d72595 100644 --- a/drivers/pci/hotplug/cpqphp_pci.c +++ b/drivers/pci/hotplug/cpqphp_pci.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "../pci.h" #include "cpqphp.h" #include "cpqphp_nvram.h" diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c index 5bab666cd67e..634f74d919d3 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include "cpqphp.h" diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c index aaeb1129132e..e27907c91d92 100644 --- a/drivers/pci/hotplug/fakephp.c +++ b/drivers/pci/hotplug/fakephp.c @@ -35,10 +35,10 @@ #include #include #include +#include #include #include #include -#include "pci_hotplug.h" #include "../pci.h" #if !defined(MODULE) diff --git a/drivers/pci/hotplug/ibmphp.h b/drivers/pci/hotplug/ibmphp.h index dba6d8ca9bda..612d96301509 100644 --- a/drivers/pci/hotplug/ibmphp.h +++ b/drivers/pci/hotplug/ibmphp.h @@ -30,7 +30,7 @@ * */ -#include "pci_hotplug.h" +#include extern int ibmphp_debug; diff --git a/drivers/pci/hotplug/pci_hotplug.h b/drivers/pci/hotplug/pci_hotplug.h deleted file mode 100644 index a675a05c4091..000000000000 --- a/drivers/pci/hotplug/pci_hotplug.h +++ /dev/null @@ -1,236 +0,0 @@ -/* - * PCI HotPlug Core Functions - * - * Copyright (C) 1995,2001 Compaq Computer Corporation - * Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com) - * Copyright (C) 2001 IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to - * - */ -#ifndef _PCI_HOTPLUG_H -#define _PCI_HOTPLUG_H - - -/* These values come from the PCI Hotplug Spec */ -enum pci_bus_speed { - PCI_SPEED_33MHz = 0x00, - PCI_SPEED_66MHz = 0x01, - PCI_SPEED_66MHz_PCIX = 0x02, - PCI_SPEED_100MHz_PCIX = 0x03, - PCI_SPEED_133MHz_PCIX = 0x04, - PCI_SPEED_66MHz_PCIX_ECC = 0x05, - PCI_SPEED_100MHz_PCIX_ECC = 0x06, - PCI_SPEED_133MHz_PCIX_ECC = 0x07, - PCI_SPEED_66MHz_PCIX_266 = 0x09, - PCI_SPEED_100MHz_PCIX_266 = 0x0a, - PCI_SPEED_133MHz_PCIX_266 = 0x0b, - PCI_SPEED_66MHz_PCIX_533 = 0x11, - PCI_SPEED_100MHz_PCIX_533 = 0x12, - PCI_SPEED_133MHz_PCIX_533 = 0x13, - PCI_SPEED_UNKNOWN = 0xff, -}; - -/* These values come from the PCI Express Spec */ -enum pcie_link_width { - PCIE_LNK_WIDTH_RESRV = 0x00, - PCIE_LNK_X1 = 0x01, - PCIE_LNK_X2 = 0x02, - PCIE_LNK_X4 = 0x04, - PCIE_LNK_X8 = 0x08, - PCIE_LNK_X12 = 0x0C, - PCIE_LNK_X16 = 0x10, - PCIE_LNK_X32 = 0x20, - PCIE_LNK_WIDTH_UNKNOWN = 0xFF, -}; - -enum pcie_link_speed { - PCIE_2PT5GB = 0x14, - PCIE_LNK_SPEED_UNKNOWN = 0xFF, -}; - -struct hotplug_slot; -struct hotplug_slot_attribute { - struct attribute attr; - ssize_t (*show)(struct hotplug_slot *, char *); - ssize_t (*store)(struct hotplug_slot *, const char *, size_t); -}; -#define to_hotplug_attr(n) container_of(n, struct hotplug_slot_attribute, attr); - -/** - * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use - * @owner: The module owner of this structure - * @enable_slot: Called when the user wants to enable a specific pci slot - * @disable_slot: Called when the user wants to disable a specific pci slot - * @set_attention_status: Called to set the specific slot's attention LED to - * the specified value - * @hardware_test: Called to run a specified hardware test on the specified - * slot. - * @get_power_status: Called to get the current power status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_attention_status: Called to get the current attention status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_latch_status: Called to get the current latch status of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_adapter_status: Called to get see if an adapter is present in the slot or not. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_address: Called to get pci address of a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_max_bus_speed: Called to get the max bus speed for a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * @get_cur_bus_speed: Called to get the current bus speed for a slot. - * If this field is NULL, the value passed in the struct hotplug_slot_info - * will be used when this value is requested by a user. - * - * The table of function pointers that is passed to the hotplug pci core by a - * hotplug pci driver. These functions are called by the hotplug pci core when - * the user wants to do something to a specific slot (query it for information, - * set an LED, enable / disable power, etc.) - */ -struct hotplug_slot_ops { - struct module *owner; - int (*enable_slot) (struct hotplug_slot *slot); - int (*disable_slot) (struct hotplug_slot *slot); - int (*set_attention_status) (struct hotplug_slot *slot, u8 value); - int (*hardware_test) (struct hotplug_slot *slot, u32 value); - int (*get_power_status) (struct hotplug_slot *slot, u8 *value); - int (*get_attention_status) (struct hotplug_slot *slot, u8 *value); - int (*get_latch_status) (struct hotplug_slot *slot, u8 *value); - int (*get_adapter_status) (struct hotplug_slot *slot, u8 *value); - int (*get_address) (struct hotplug_slot *slot, u32 *value); - int (*get_max_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); - int (*get_cur_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); -}; - -/** - * struct hotplug_slot_info - used to notify the hotplug pci core of the state of the slot - * @power: if power is enabled or not (1/0) - * @attention_status: if the attention light is enabled or not (1/0) - * @latch_status: if the latch (if any) is open or closed (1/0) - * @adapter_present: if there is a pci board present in the slot or not (1/0) - * @address: (domain << 16 | bus << 8 | dev) - * - * Used to notify the hotplug pci core of the status of a specific slot. - */ -struct hotplug_slot_info { - u8 power_status; - u8 attention_status; - u8 latch_status; - u8 adapter_status; - u32 address; - enum pci_bus_speed max_bus_speed; - enum pci_bus_speed cur_bus_speed; -}; - -/** - * struct hotplug_slot - used to register a physical slot with the hotplug pci core - * @name: the name of the slot being registered. This string must - * be unique amoung slots registered on this system. - * @ops: pointer to the &struct hotplug_slot_ops to be used for this slot - * @info: pointer to the &struct hotplug_slot_info for the initial values for - * this slot. - * @release: called during pci_hp_deregister to free memory allocated in a - * hotplug_slot structure. - * @private: used by the hotplug pci controller driver to store whatever it - * needs. - */ -struct hotplug_slot { - char *name; - struct hotplug_slot_ops *ops; - struct hotplug_slot_info *info; - void (*release) (struct hotplug_slot *slot); - void *private; - - /* Variables below this are for use only by the hotplug pci core. */ - struct list_head slot_list; - struct kobject kobj; -}; -#define to_hotplug_slot(n) container_of(n, struct hotplug_slot, kobj) - -extern int pci_hp_register (struct hotplug_slot *slot); -extern int pci_hp_deregister (struct hotplug_slot *slot); -extern int __must_check pci_hp_change_slot_info (struct hotplug_slot *slot, - struct hotplug_slot_info *info); -extern struct subsystem pci_hotplug_slots_subsys; - -/* PCI Setting Record (Type 0) */ -struct hpp_type0 { - u32 revision; - u8 cache_line_size; - u8 latency_timer; - u8 enable_serr; - u8 enable_perr; -}; - -/* PCI-X Setting Record (Type 1) */ -struct hpp_type1 { - u32 revision; - u8 max_mem_read; - u8 avg_max_split; - u16 tot_max_split; -}; - -/* PCI Express Setting Record (Type 2) */ -struct hpp_type2 { - u32 revision; - u32 unc_err_mask_and; - u32 unc_err_mask_or; - u32 unc_err_sever_and; - u32 unc_err_sever_or; - u32 cor_err_mask_and; - u32 cor_err_mask_or; - u32 adv_err_cap_and; - u32 adv_err_cap_or; - u16 pci_exp_devctl_and; - u16 pci_exp_devctl_or; - u16 pci_exp_lnkctl_and; - u16 pci_exp_lnkctl_or; - u32 sec_unc_err_sever_and; - u32 sec_unc_err_sever_or; - u32 sec_unc_err_mask_and; - u32 sec_unc_err_mask_or; -}; - -struct hotplug_params { - struct hpp_type0 *t0; /* Type0: NULL if not available */ - struct hpp_type1 *t1; /* Type1: NULL if not available */ - struct hpp_type2 *t2; /* Type2: NULL if not available */ - struct hpp_type0 type0_data; - struct hpp_type1 type1_data; - struct hpp_type2 type2_data; -}; - -#ifdef CONFIG_ACPI -#include -#include -#include -extern acpi_status acpi_run_oshp(acpi_handle handle); -extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, - struct hotplug_params *hpp); -int acpi_root_bridge(acpi_handle handle); -#endif -#endif - diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c index fa666d0cc489..f5d632e72323 100644 --- a/drivers/pci/hotplug/pci_hotplug_core.c +++ b/drivers/pci/hotplug/pci_hotplug_core.c @@ -30,6 +30,8 @@ #include #include #include +#include +#include #include #include #include @@ -37,11 +39,8 @@ #include #include #include +#include #include -#include -#include -#include "pci_hotplug.h" - #define MY_NAME "pci_hotplug" diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 30f021c55fe5..4fb12fcda563 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -31,11 +31,11 @@ #include #include +#include #include #include /* signal_pending() */ #include #include -#include "pci_hotplug.h" #define MY_NAME "pciehp" diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c index 2b9e10e38613..50bcd3fe61da 100644 --- a/drivers/pci/hotplug/pcihp_skeleton.c +++ b/drivers/pci/hotplug/pcihp_skeleton.c @@ -33,8 +33,8 @@ #include #include #include +#include #include -#include "pci_hotplug.h" #define SLOT_NAME_SIZE 10 struct slot { diff --git a/drivers/pci/hotplug/rpadlpar_sysfs.c b/drivers/pci/hotplug/rpadlpar_sysfs.c index db69be85b458..6c5be3ff578c 100644 --- a/drivers/pci/hotplug/rpadlpar_sysfs.c +++ b/drivers/pci/hotplug/rpadlpar_sysfs.c @@ -14,7 +14,7 @@ */ #include #include -#include "pci_hotplug.h" +#include #include "rpadlpar.h" #define DLPAR_KOBJ_NAME "control" diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c index 7288a3eccfb3..141486df235b 100644 --- a/drivers/pci/hotplug/rpaphp_core.c +++ b/drivers/pci/hotplug/rpaphp_core.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -36,7 +37,6 @@ #include "../pci.h" /* for pci_add_new_bus */ /* and pci_do_scan_bus */ #include "rpaphp.h" -#include "pci_hotplug.h" int debug; static struct semaphore rpaphp_sem; diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c index f31d83c2c633..b62ad31a9739 100644 --- a/drivers/pci/hotplug/sgi_hotplug.c +++ b/drivers/pci/hotplug/sgi_hotplug.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -29,7 +30,6 @@ #include #include "../pci.h" -#include "pci_hotplug.h" MODULE_LICENSE("GPL"); MODULE_AUTHOR("SGI (prarit@sgi.com, dickie@sgi.com, habeck@sgi.com)"); diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index 7e7d490622e1..ea2087c34149 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -31,12 +31,11 @@ #include #include +#include #include #include /* signal_pending(), struct timer_list */ #include -#include "pci_hotplug.h" - #if !defined(MODULE) #define MY_NAME "shpchp" #else diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h new file mode 100644 index 000000000000..a675a05c4091 --- /dev/null +++ b/include/linux/pci_hotplug.h @@ -0,0 +1,236 @@ +/* + * PCI HotPlug Core Functions + * + * Copyright (C) 1995,2001 Compaq Computer Corporation + * Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com) + * Copyright (C) 2001 IBM Corp. + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or (at + * your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Send feedback to + * + */ +#ifndef _PCI_HOTPLUG_H +#define _PCI_HOTPLUG_H + + +/* These values come from the PCI Hotplug Spec */ +enum pci_bus_speed { + PCI_SPEED_33MHz = 0x00, + PCI_SPEED_66MHz = 0x01, + PCI_SPEED_66MHz_PCIX = 0x02, + PCI_SPEED_100MHz_PCIX = 0x03, + PCI_SPEED_133MHz_PCIX = 0x04, + PCI_SPEED_66MHz_PCIX_ECC = 0x05, + PCI_SPEED_100MHz_PCIX_ECC = 0x06, + PCI_SPEED_133MHz_PCIX_ECC = 0x07, + PCI_SPEED_66MHz_PCIX_266 = 0x09, + PCI_SPEED_100MHz_PCIX_266 = 0x0a, + PCI_SPEED_133MHz_PCIX_266 = 0x0b, + PCI_SPEED_66MHz_PCIX_533 = 0x11, + PCI_SPEED_100MHz_PCIX_533 = 0x12, + PCI_SPEED_133MHz_PCIX_533 = 0x13, + PCI_SPEED_UNKNOWN = 0xff, +}; + +/* These values come from the PCI Express Spec */ +enum pcie_link_width { + PCIE_LNK_WIDTH_RESRV = 0x00, + PCIE_LNK_X1 = 0x01, + PCIE_LNK_X2 = 0x02, + PCIE_LNK_X4 = 0x04, + PCIE_LNK_X8 = 0x08, + PCIE_LNK_X12 = 0x0C, + PCIE_LNK_X16 = 0x10, + PCIE_LNK_X32 = 0x20, + PCIE_LNK_WIDTH_UNKNOWN = 0xFF, +}; + +enum pcie_link_speed { + PCIE_2PT5GB = 0x14, + PCIE_LNK_SPEED_UNKNOWN = 0xFF, +}; + +struct hotplug_slot; +struct hotplug_slot_attribute { + struct attribute attr; + ssize_t (*show)(struct hotplug_slot *, char *); + ssize_t (*store)(struct hotplug_slot *, const char *, size_t); +}; +#define to_hotplug_attr(n) container_of(n, struct hotplug_slot_attribute, attr); + +/** + * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use + * @owner: The module owner of this structure + * @enable_slot: Called when the user wants to enable a specific pci slot + * @disable_slot: Called when the user wants to disable a specific pci slot + * @set_attention_status: Called to set the specific slot's attention LED to + * the specified value + * @hardware_test: Called to run a specified hardware test on the specified + * slot. + * @get_power_status: Called to get the current power status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_attention_status: Called to get the current attention status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_latch_status: Called to get the current latch status of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_adapter_status: Called to get see if an adapter is present in the slot or not. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_address: Called to get pci address of a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_max_bus_speed: Called to get the max bus speed for a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * @get_cur_bus_speed: Called to get the current bus speed for a slot. + * If this field is NULL, the value passed in the struct hotplug_slot_info + * will be used when this value is requested by a user. + * + * The table of function pointers that is passed to the hotplug pci core by a + * hotplug pci driver. These functions are called by the hotplug pci core when + * the user wants to do something to a specific slot (query it for information, + * set an LED, enable / disable power, etc.) + */ +struct hotplug_slot_ops { + struct module *owner; + int (*enable_slot) (struct hotplug_slot *slot); + int (*disable_slot) (struct hotplug_slot *slot); + int (*set_attention_status) (struct hotplug_slot *slot, u8 value); + int (*hardware_test) (struct hotplug_slot *slot, u32 value); + int (*get_power_status) (struct hotplug_slot *slot, u8 *value); + int (*get_attention_status) (struct hotplug_slot *slot, u8 *value); + int (*get_latch_status) (struct hotplug_slot *slot, u8 *value); + int (*get_adapter_status) (struct hotplug_slot *slot, u8 *value); + int (*get_address) (struct hotplug_slot *slot, u32 *value); + int (*get_max_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); + int (*get_cur_bus_speed) (struct hotplug_slot *slot, enum pci_bus_speed *value); +}; + +/** + * struct hotplug_slot_info - used to notify the hotplug pci core of the state of the slot + * @power: if power is enabled or not (1/0) + * @attention_status: if the attention light is enabled or not (1/0) + * @latch_status: if the latch (if any) is open or closed (1/0) + * @adapter_present: if there is a pci board present in the slot or not (1/0) + * @address: (domain << 16 | bus << 8 | dev) + * + * Used to notify the hotplug pci core of the status of a specific slot. + */ +struct hotplug_slot_info { + u8 power_status; + u8 attention_status; + u8 latch_status; + u8 adapter_status; + u32 address; + enum pci_bus_speed max_bus_speed; + enum pci_bus_speed cur_bus_speed; +}; + +/** + * struct hotplug_slot - used to register a physical slot with the hotplug pci core + * @name: the name of the slot being registered. This string must + * be unique amoung slots registered on this system. + * @ops: pointer to the &struct hotplug_slot_ops to be used for this slot + * @info: pointer to the &struct hotplug_slot_info for the initial values for + * this slot. + * @release: called during pci_hp_deregister to free memory allocated in a + * hotplug_slot structure. + * @private: used by the hotplug pci controller driver to store whatever it + * needs. + */ +struct hotplug_slot { + char *name; + struct hotplug_slot_ops *ops; + struct hotplug_slot_info *info; + void (*release) (struct hotplug_slot *slot); + void *private; + + /* Variables below this are for use only by the hotplug pci core. */ + struct list_head slot_list; + struct kobject kobj; +}; +#define to_hotplug_slot(n) container_of(n, struct hotplug_slot, kobj) + +extern int pci_hp_register (struct hotplug_slot *slot); +extern int pci_hp_deregister (struct hotplug_slot *slot); +extern int __must_check pci_hp_change_slot_info (struct hotplug_slot *slot, + struct hotplug_slot_info *info); +extern struct subsystem pci_hotplug_slots_subsys; + +/* PCI Setting Record (Type 0) */ +struct hpp_type0 { + u32 revision; + u8 cache_line_size; + u8 latency_timer; + u8 enable_serr; + u8 enable_perr; +}; + +/* PCI-X Setting Record (Type 1) */ +struct hpp_type1 { + u32 revision; + u8 max_mem_read; + u8 avg_max_split; + u16 tot_max_split; +}; + +/* PCI Express Setting Record (Type 2) */ +struct hpp_type2 { + u32 revision; + u32 unc_err_mask_and; + u32 unc_err_mask_or; + u32 unc_err_sever_and; + u32 unc_err_sever_or; + u32 cor_err_mask_and; + u32 cor_err_mask_or; + u32 adv_err_cap_and; + u32 adv_err_cap_or; + u16 pci_exp_devctl_and; + u16 pci_exp_devctl_or; + u16 pci_exp_lnkctl_and; + u16 pci_exp_lnkctl_or; + u32 sec_unc_err_sever_and; + u32 sec_unc_err_sever_or; + u32 sec_unc_err_mask_and; + u32 sec_unc_err_mask_or; +}; + +struct hotplug_params { + struct hpp_type0 *t0; /* Type0: NULL if not available */ + struct hpp_type1 *t1; /* Type1: NULL if not available */ + struct hpp_type2 *t2; /* Type2: NULL if not available */ + struct hpp_type0 type0_data; + struct hpp_type1 type1_data; + struct hpp_type2 type2_data; +}; + +#ifdef CONFIG_ACPI +#include +#include +#include +extern acpi_status acpi_run_oshp(acpi_handle handle); +extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, + struct hotplug_params *hpp); +int acpi_root_bridge(acpi_handle handle); +#endif +#endif + -- cgit v1.2.3 From eb409460b1abec0e2a1f9c9d07019f4157a6d6bc Mon Sep 17 00:00:00 2001 From: Lijun Chen Date: Mon, 16 Oct 2006 21:59:42 -0700 Subject: [TIPC]: Added subscription cancellation capability This patch allows a TIPC application to cancel an existing topology service subscription by re-requesting the subscription with the TIPC_SUB_CANCEL filter bit set. (All other bits of the cancel request must match the original subscription request.) Signed-off-by: Allan Stephens Signed-off-by: Per Liden Signed-off-by: David S. Miller --- include/linux/tipc.h | 1 + net/tipc/subscr.c | 99 ++++++++++++++++++++++++++++++++++++++++++++-------- 2 files changed, 85 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tipc.h b/include/linux/tipc.h index 243a15f54002..bea469455a0c 100644 --- a/include/linux/tipc.h +++ b/include/linux/tipc.h @@ -129,6 +129,7 @@ static inline unsigned int tipc_node(__u32 addr) #define TIPC_SUB_PORTS 0x01 /* filter for port availability */ #define TIPC_SUB_SERVICE 0x02 /* filter for service availability */ +#define TIPC_SUB_CANCEL 0x04 /* cancel a subscription */ #if 0 /* The following filter options are not currently implemented */ #define TIPC_SUB_NO_BIND_EVTS 0x04 /* filter out "publish" events */ diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index c51600ba5f4a..7a918f12a5df 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -155,7 +155,7 @@ void tipc_subscr_report_overlap(struct subscription *sub, sub->seq.upper, found_lower, found_upper); if (!tipc_subscr_overlap(sub, found_lower, found_upper)) return; - if (!must && (sub->filter != TIPC_SUB_PORTS)) + if (!must && !(sub->filter & TIPC_SUB_PORTS)) return; subscr_send_event(sub, found_lower, found_upper, event, port_ref, node); } @@ -176,6 +176,13 @@ static void subscr_timeout(struct subscription *sub) if (subscriber == NULL) return; + /* Validate timeout (in case subscription is being cancelled) */ + + if (sub->timeout == TIPC_WAIT_FOREVER) { + tipc_ref_unlock(subscriber_ref); + return; + } + /* Unlink subscription from name table */ tipc_nametbl_unsubscribe(sub); @@ -198,6 +205,20 @@ static void subscr_timeout(struct subscription *sub) atomic_dec(&topsrv.subscription_count); } +/** + * subscr_del - delete a subscription within a subscription list + * + * Called with subscriber locked. + */ + +static void subscr_del(struct subscription *sub) +{ + tipc_nametbl_unsubscribe(sub); + list_del(&sub->subscription_list); + kfree(sub); + atomic_dec(&topsrv.subscription_count); +} + /** * subscr_terminate - terminate communication with a subscriber * @@ -227,12 +248,9 @@ static void subscr_terminate(struct subscriber *subscriber) k_cancel_timer(&sub->timer); k_term_timer(&sub->timer); } - tipc_nametbl_unsubscribe(sub); - list_del(&sub->subscription_list); - dbg("Term: Removed sub %u,%u,%u from subscriber %x list\n", + dbg("Term: Removing sub %u,%u,%u from subscriber %x list\n", sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); - kfree(sub); - atomic_dec(&topsrv.subscription_count); + subscr_del(sub); } /* Sever connection to subscriber */ @@ -252,6 +270,49 @@ static void subscr_terminate(struct subscriber *subscriber) kfree(subscriber); } +/** + * subscr_cancel - handle subscription cancellation request + * + * Called with subscriber locked. Routine must temporarily release this lock + * to enable the subscription timeout routine to finish without deadlocking; + * the lock is then reclaimed to allow caller to release it upon return. + * + * Note that fields of 's' use subscriber's endianness! + */ + +static void subscr_cancel(struct tipc_subscr *s, + struct subscriber *subscriber) +{ + struct subscription *sub; + struct subscription *sub_temp; + int found = 0; + + /* Find first matching subscription, exit if not found */ + + list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, + subscription_list) { + if (!memcmp(s, &sub->evt.s, sizeof(struct tipc_subscr))) { + found = 1; + break; + } + } + if (!found) + return; + + /* Cancel subscription timer (if used), then delete subscription */ + + if (sub->timeout != TIPC_WAIT_FOREVER) { + sub->timeout = TIPC_WAIT_FOREVER; + spin_unlock_bh(subscriber->lock); + k_cancel_timer(&sub->timer); + k_term_timer(&sub->timer); + spin_lock_bh(subscriber->lock); + } + dbg("Cancel: removing sub %u,%u,%u from subscriber %x list\n", + sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); + subscr_del(sub); +} + /** * subscr_subscribe - create subscription for subscriber * @@ -263,6 +324,21 @@ static void subscr_subscribe(struct tipc_subscr *s, { struct subscription *sub; + /* Determine/update subscriber's endianness */ + + if (s->filter & (TIPC_SUB_PORTS | TIPC_SUB_SERVICE)) + subscriber->swap = 0; + else + subscriber->swap = 1; + + /* Detect & process a subscription cancellation request */ + + if (s->filter & htohl(TIPC_SUB_CANCEL, subscriber->swap)) { + s->filter &= ~htohl(TIPC_SUB_CANCEL, subscriber->swap); + subscr_cancel(s, subscriber); + return; + } + /* Refuse subscription if global limit exceeded */ if (atomic_read(&topsrv.subscription_count) >= tipc_max_subscriptions) { @@ -281,13 +357,6 @@ static void subscr_subscribe(struct tipc_subscr *s, return; } - /* Determine/update subscriber's endianness */ - - if ((s->filter == TIPC_SUB_PORTS) || (s->filter == TIPC_SUB_SERVICE)) - subscriber->swap = 0; - else - subscriber->swap = 1; - /* Initialize subscription object */ memset(sub, 0, sizeof(*sub)); @@ -296,8 +365,8 @@ static void subscr_subscribe(struct tipc_subscr *s, sub->seq.upper = htohl(s->seq.upper, subscriber->swap); sub->timeout = htohl(s->timeout, subscriber->swap); sub->filter = htohl(s->filter, subscriber->swap); - if ((((sub->filter != TIPC_SUB_PORTS) - && (sub->filter != TIPC_SUB_SERVICE))) + if ((!(sub->filter & TIPC_SUB_PORTS) + == !(sub->filter & TIPC_SUB_SERVICE)) || (sub->seq.lower > sub->seq.upper)) { warn("Subscription rejected, illegal request\n"); kfree(sub); -- cgit v1.2.3 From ae8064ac32d07f609114d73928cdef803be87134 Mon Sep 17 00:00:00 2001 From: John Heffner Date: Wed, 18 Oct 2006 20:36:48 -0700 Subject: [TCP]: Bound TSO defer time This patch limits the amount of time you will defer sending a TSO segment to less than two clock ticks, or the time between two acks, whichever is longer. On slow links, deferring causes significant bursts. See attached plots, which show RTT through a 1 Mbps link with a 100 ms RTT and ~100 ms queue for (a) non-TSO, (b) currnet TSO, and (c) patched TSO. This burstiness causes significant jitter, tends to overflow queues early (bad for short queues), and makes delay-based congestion control more difficult. Deferring by a couple clock ticks I believe will have a relatively small impact on performance. Signed-off-by: John Heffner Signed-off-by: David S. Miller --- include/linux/tcp.h | 2 ++ net/ipv4/tcp_output.c | 20 +++++++++++++++----- 2 files changed, 17 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 0e058a2d1c6d..2d36f6db3706 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -342,6 +342,8 @@ struct tcp_sock { unsigned long last_synq_overflow; + __u32 tso_deferred; + /* Receiver side RTT estimation */ struct { __u32 rtt; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f22536e32cb1..ca406157724c 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1096,10 +1096,14 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ u32 send_win, cong_win, limit, in_flight; if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) - return 0; + goto send_now; if (icsk->icsk_ca_state != TCP_CA_Open) - return 0; + goto send_now; + + /* Defer for less than two clock ticks. */ + if (!tp->tso_deferred && ((jiffies<<1)>>1) - (tp->tso_deferred>>1) > 1) + goto send_now; in_flight = tcp_packets_in_flight(tp); @@ -1115,7 +1119,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ /* If a full-sized TSO skb can be sent, do it. */ if (limit >= 65536) - return 0; + goto send_now; if (sysctl_tcp_tso_win_divisor) { u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache); @@ -1125,7 +1129,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ */ chunk /= sysctl_tcp_tso_win_divisor; if (limit >= chunk) - return 0; + goto send_now; } else { /* Different approach, try not to defer past a single * ACK. Receiver should ACK every other full sized @@ -1133,11 +1137,17 @@ static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_ * then send now. */ if (limit > tcp_max_burst(tp) * tp->mss_cache) - return 0; + goto send_now; } /* Ok, it looks like it is advisable to defer. */ + tp->tso_deferred = 1 | (jiffies<<1); + return 1; + +send_now: + tp->tso_deferred = 0; + return 0; } /* Create a new MTU probe if we are ready. -- cgit v1.2.3 From 62752ee198dca9209b7dee504763e51b11e9e0ca Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 17 Oct 2006 10:31:38 +0200 Subject: [PATCH] Take i_mutex in splice_from_pipe() The splice_actor may be calling ->prepare_write() and ->commit_write(). We want i_mutex on the inode being written to before calling those so that we don't race i_size changes. The double locking behavior is done elsewhere in splice.c, and if we eventually want _nolock variants of generic_file_splice_write(), fs modules might have to replicate the nasty locking code. We introduce inode_double_lock() and inode_double_unlock() to consolidate the locking rules into one set of functions. Signed-off-by: Mark Fasheh Signed-off-by: Jens Axboe --- fs/inode.c | 36 ++++++++++++++++++++++++++++++++++++ fs/splice.c | 24 +++++++++++------------- include/linux/fs.h | 3 +++ 3 files changed, 50 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/fs/inode.c b/fs/inode.c index d9a21d122926..26cdb115ce67 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1306,6 +1306,42 @@ void wake_up_inode(struct inode *inode) wake_up_bit(&inode->i_state, __I_LOCK); } +/* + * We rarely want to lock two inodes that do not have a parent/child + * relationship (such as directory, child inode) simultaneously. The + * vast majority of file systems should be able to get along fine + * without this. Do not use these functions except as a last resort. + */ +void inode_double_lock(struct inode *inode1, struct inode *inode2) +{ + if (inode1 == NULL || inode2 == NULL || inode1 == inode2) { + if (inode1) + mutex_lock(&inode1->i_mutex); + else if (inode2) + mutex_lock(&inode2->i_mutex); + return; + } + + if (inode1 < inode2) { + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD); + } else { + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD); + } +} +EXPORT_SYMBOL(inode_double_lock); + +void inode_double_unlock(struct inode *inode1, struct inode *inode2) +{ + if (inode1) + mutex_unlock(&inode1->i_mutex); + + if (inode2 && inode2 != inode1) + mutex_unlock(&inode2->i_mutex); +} +EXPORT_SYMBOL(inode_double_unlock); + static __initdata unsigned long ihash_entries; static int __init set_ihash_entries(char *str) { diff --git a/fs/splice.c b/fs/splice.c index a567010b62ac..c1072b6940c3 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -713,6 +713,7 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, { int ret, do_wakeup, err; struct splice_desc sd; + struct inode *inode = out->f_mapping->host; ret = 0; do_wakeup = 0; @@ -722,8 +723,13 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, sd.file = out; sd.pos = *ppos; - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); + /* + * The actor worker might be calling ->prepare_write and + * ->commit_write. Most of the time, these expect i_mutex to + * be held. Since this may result in an ABBA deadlock with + * pipe->inode, we have to order lock acquiry here. + */ + inode_double_lock(inode, pipe->inode); for (;;) { if (pipe->nrbufs) { @@ -797,8 +803,7 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, pipe_wait(pipe); } - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); + inode_double_unlock(inode, pipe->inode); if (do_wakeup) { smp_mb(); @@ -1400,13 +1405,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, * grabbing by inode address. Otherwise two different processes * could deadlock (one doing tee from A -> B, the other from B -> A). */ - if (ipipe->inode < opipe->inode) { - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_CHILD); - } else { - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_CHILD); - } + inode_double_lock(ipipe->inode, opipe->inode); do { if (!opipe->readers) { @@ -1450,8 +1449,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, i++; } while (len); - mutex_unlock(&ipipe->inode->i_mutex); - mutex_unlock(&opipe->inode->i_mutex); + inode_double_unlock(ipipe->inode, opipe->inode); /* * If we put data in the output pipe, wakeup any potential readers. diff --git a/include/linux/fs.h b/include/linux/fs.h index 661c7c572149..853a02f23936 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -623,6 +623,9 @@ enum inode_i_mutex_lock_class I_MUTEX_QUOTA }; +extern void inode_double_lock(struct inode *inode1, struct inode *inode2); +extern void inode_double_unlock(struct inode *inode1, struct inode *inode2); + /* * NOTE: in a 32bit arch with a preemptable kernel and * an UP compile the i_size_read/write must be atomic -- cgit v1.2.3 From 6da61809822c22634a3de2dcb3c60283b836a88a Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 17 Oct 2006 18:43:07 +0200 Subject: [PATCH] Introduce generic_file_splice_write_nolock() This allows file systems to manage their own i_mutex locking while still re-using the generic_file_splice_write() logic. OCFS2 in particular wants this so that it can order cluster locks within i_mutex. Signed-off-by: Mark Fasheh Signed-off-by: Jens Axboe --- fs/splice.c | 80 ++++++++++++++++++++++++++++++++++++++++++++---------- include/linux/fs.h | 2 ++ 2 files changed, 68 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/fs/splice.c b/fs/splice.c index c1072b6940c3..68e20e65c6e1 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -707,13 +707,12 @@ out_ret: * key here is the 'actor' worker passed in that actually moves the data * to the wanted destination. See pipe_to_file/pipe_to_sendpage above. */ -ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags, - splice_actor *actor) +static ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, + struct file *out, loff_t *ppos, size_t len, + unsigned int flags, splice_actor *actor) { int ret, do_wakeup, err; struct splice_desc sd; - struct inode *inode = out->f_mapping->host; ret = 0; do_wakeup = 0; @@ -723,14 +722,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, sd.file = out; sd.pos = *ppos; - /* - * The actor worker might be calling ->prepare_write and - * ->commit_write. Most of the time, these expect i_mutex to - * be held. Since this may result in an ABBA deadlock with - * pipe->inode, we have to order lock acquiry here. - */ - inode_double_lock(inode, pipe->inode); - for (;;) { if (pipe->nrbufs) { struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; @@ -803,8 +794,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, pipe_wait(pipe); } - inode_double_unlock(inode, pipe->inode); - if (do_wakeup) { smp_mb(); if (waitqueue_active(&pipe->wait)) @@ -815,6 +804,69 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, return ret; } +ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags, + splice_actor *actor) +{ + ssize_t ret; + struct inode *inode = out->f_mapping->host; + + /* + * The actor worker might be calling ->prepare_write and + * ->commit_write. Most of the time, these expect i_mutex to + * be held. Since this may result in an ABBA deadlock with + * pipe->inode, we have to order lock acquiry here. + */ + inode_double_lock(inode, pipe->inode); + ret = __splice_from_pipe(pipe, out, ppos, len, flags, actor); + inode_double_unlock(inode, pipe->inode); + + return ret; +} + +/** + * generic_file_splice_write_nolock - generic_file_splice_write without mutexes + * @pipe: pipe info + * @out: file to write to + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Will either move or copy pages (determined by @flags options) from + * the given pipe inode to the given file. The caller is responsible + * for acquiring i_mutex on both inodes. + * + */ +ssize_t +generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + int err; + + ret = __splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); + if (ret > 0) { + *ppos += ret; + + /* + * If file or inode is SYNC and we actually wrote some data, + * sync it. + */ + if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { + err = generic_osync_inode(inode, mapping, + OSYNC_METADATA|OSYNC_DATA); + + if (err) + ret = err; + } + } + + return ret; +} + +EXPORT_SYMBOL(generic_file_splice_write_nolock); + /** * generic_file_splice_write - splice data from a pipe to a file * @pipe: pipe info diff --git a/include/linux/fs.h b/include/linux/fs.h index 853a02f23936..d695ba2346a3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1758,6 +1758,8 @@ extern ssize_t generic_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); extern ssize_t generic_file_splice_write(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); +extern ssize_t generic_file_splice_write_nolock(struct pipe_inode_info *, + struct file *, loff_t *, size_t, unsigned int); extern ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe, struct file *out, loff_t *, size_t len, unsigned int flags); extern long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, -- cgit v1.2.3 From 01de85e057328ecbef36e108673b1e81059d54c1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 17 Oct 2006 19:50:36 +0200 Subject: [PATCH] Add lockless helpers for remove_suid() Right now users have to grab i_mutex before calling remove_suid(), in the unlikely event that a call to ->setattr() may be needed. Split up the function in two parts: - One to check if we need to remove suid - One to actually remove it The first we can call lockless. Signed-off-by: Jens Axboe --- include/linux/fs.h | 2 ++ mm/filemap.c | 30 ++++++++++++++++++++++-------- 2 files changed, 24 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index d695ba2346a3..2fe6e3f900ba 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1712,6 +1712,8 @@ extern void __iget(struct inode * inode); extern void clear_inode(struct inode *); extern void destroy_inode(struct inode *); extern struct inode *new_inode(struct super_block *); +extern int __remove_suid(struct dentry *, int); +extern int should_remove_suid(struct dentry *); extern int remove_suid(struct dentry *); extern void remove_dquot_ref(struct super_block *, int, struct list_head *); diff --git a/mm/filemap.c b/mm/filemap.c index 3464b681f844..7c7addb9333c 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1884,11 +1884,10 @@ repeat: * if suid or (sgid and xgrp) * remove privs */ -int remove_suid(struct dentry *dentry) +int should_remove_suid(struct dentry *dentry) { mode_t mode = dentry->d_inode->i_mode; int kill = 0; - int result = 0; /* suid always must be killed */ if (unlikely(mode & S_ISUID)) @@ -1901,13 +1900,28 @@ int remove_suid(struct dentry *dentry) if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) kill |= ATTR_KILL_SGID; - if (unlikely(kill && !capable(CAP_FSETID))) { - struct iattr newattrs; + if (unlikely(kill && !capable(CAP_FSETID))) + return kill; - newattrs.ia_valid = ATTR_FORCE | kill; - result = notify_change(dentry, &newattrs); - } - return result; + return 0; +} + +int __remove_suid(struct dentry *dentry, int kill) +{ + struct iattr newattrs; + + newattrs.ia_valid = ATTR_FORCE | kill; + return notify_change(dentry, &newattrs); +} + +int remove_suid(struct dentry *dentry) +{ + int kill = should_remove_suid(dentry); + + if (unlikely(kill)) + return __remove_suid(dentry, kill); + + return 0; } EXPORT_SYMBOL(remove_suid); -- cgit v1.2.3 From 79e2de4bc53d7ca2a8eedee49e4a92479b4b530e Mon Sep 17 00:00:00 2001 From: Thomas Maier Date: Thu, 19 Oct 2006 23:28:15 -0700 Subject: [PATCH] export clear_queue_congested and set_queue_congested Export the clear_queue_congested() and set_queue_congested() functions located in ll_rw_blk.c The functions are renamed to blk_clear_queue_congested() and blk_set_queue_congested(). (needed in the pktcdvd driver's bio write congestion control) Signed-off-by: Thomas Maier Cc: Peter Osterlund Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- block/ll_rw_blk.c | 20 ++++++++++---------- include/linux/blkdev.h | 2 ++ 2 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index c847e17e5caa..132a858ce2c5 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -117,7 +117,7 @@ static void blk_queue_congestion_threshold(struct request_queue *q) * congested queues, and wake up anyone who was waiting for requests to be * put back. */ -static void clear_queue_congested(request_queue_t *q, int rw) +void blk_clear_queue_congested(request_queue_t *q, int rw) { enum bdi_state bit; wait_queue_head_t *wqh = &congestion_wqh[rw]; @@ -128,18 +128,20 @@ static void clear_queue_congested(request_queue_t *q, int rw) if (waitqueue_active(wqh)) wake_up(wqh); } +EXPORT_SYMBOL(blk_clear_queue_congested); /* * A queue has just entered congestion. Flag that in the queue's VM-visible * state flags and increment the global gounter of congested queues. */ -static void set_queue_congested(request_queue_t *q, int rw) +void blk_set_queue_congested(request_queue_t *q, int rw) { enum bdi_state bit; bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; set_bit(bit, &q->backing_dev_info.state); } +EXPORT_SYMBOL(blk_set_queue_congested); /** * blk_get_backing_dev_info - get the address of a queue's backing_dev_info @@ -159,7 +161,6 @@ struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev) ret = &q->backing_dev_info; return ret; } - EXPORT_SYMBOL(blk_get_backing_dev_info); void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data) @@ -167,7 +168,6 @@ void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data) q->activity_fn = fn; q->activity_data = data; } - EXPORT_SYMBOL(blk_queue_activity_fn); /** @@ -2067,7 +2067,7 @@ static void __freed_request(request_queue_t *q, int rw) struct request_list *rl = &q->rq; if (rl->count[rw] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, rw); + blk_clear_queue_congested(q, rw); if (rl->count[rw] + 1 <= q->nr_requests) { if (waitqueue_active(&rl->wait[rw])) @@ -2137,7 +2137,7 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio, } } } - set_queue_congested(q, rw); + blk_set_queue_congested(q, rw); } /* @@ -3765,14 +3765,14 @@ queue_requests_store(struct request_queue *q, const char *page, size_t count) blk_queue_congestion_threshold(q); if (rl->count[READ] >= queue_congestion_on_threshold(q)) - set_queue_congested(q, READ); + blk_set_queue_congested(q, READ); else if (rl->count[READ] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, READ); + blk_clear_queue_congested(q, READ); if (rl->count[WRITE] >= queue_congestion_on_threshold(q)) - set_queue_congested(q, WRITE); + blk_set_queue_congested(q, WRITE); else if (rl->count[WRITE] < queue_congestion_off_threshold(q)) - clear_queue_congested(q, WRITE); + blk_clear_queue_congested(q, WRITE); if (rl->count[READ] >= q->nr_requests) { blk_set_queue_full(q, READ); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d370d2cfe138..9575e3a5ff2a 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -651,6 +651,8 @@ extern void blk_recount_segments(request_queue_t *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern int sg_scsi_ioctl(struct file *, struct request_queue *, struct gendisk *, struct scsi_ioctl_command __user *); +extern void blk_clear_queue_congested(request_queue_t *q, int rw); +extern void blk_set_queue_congested(request_queue_t *q, int rw); extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); -- cgit v1.2.3 From 3fcfab16c5b86eaa3db3a9a31adba550c5b67141 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 19 Oct 2006 23:28:16 -0700 Subject: [PATCH] separate bdi congestion functions from queue congestion functions Separate out the concept of "queue congestion" from "backing-dev congestion". Congestion is a backing-dev concept, not a queue concept. The blk_* congestion functions are retained, as wrappers around the core backing-dev congestion functions. This proper layering is needed so that NFS can cleanly use the congestion functions, and so that CONFIG_BLOCK=n actually links. Cc: "Thomas Maier" Cc: "Jens Axboe" Cc: Trond Myklebust Cc: David Howells Cc: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/lib/usercopy.c | 3 +- block/ll_rw_blk.c | 71 --------------------------------------------- drivers/md/dm-crypt.c | 3 +- fs/fat/file.c | 3 +- fs/nfs/write.c | 4 ++- fs/reiserfs/journal.c | 3 +- fs/xfs/linux-2.6/kmem.c | 5 ++-- fs/xfs/linux-2.6/xfs_buf.c | 3 +- include/linux/backing-dev.h | 7 +++++ include/linux/blkdev.h | 24 ++++++++++++--- include/linux/writeback.h | 1 - mm/Makefile | 3 +- mm/backing-dev.c | 69 +++++++++++++++++++++++++++++++++++++++++++ mm/page-writeback.c | 17 +++-------- mm/page_alloc.c | 5 ++-- mm/shmem.c | 3 +- mm/vmscan.c | 6 ++-- 17 files changed, 126 insertions(+), 104 deletions(-) create mode 100644 mm/backing-dev.c (limited to 'include/linux') diff --git a/arch/i386/lib/usercopy.c b/arch/i386/lib/usercopy.c index 258df6b4d7d7..d22cfc9d656c 100644 --- a/arch/i386/lib/usercopy.c +++ b/arch/i386/lib/usercopy.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -741,7 +742,7 @@ survive: if (retval == -ENOMEM && is_init(current)) { up_read(¤t->mm->mmap_sem); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto survive; } diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 132a858ce2c5..136066583c68 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -56,11 +56,6 @@ static kmem_cache_t *requestq_cachep; */ static kmem_cache_t *iocontext_cachep; -static wait_queue_head_t congestion_wqh[2] = { - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), - __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) - }; - /* * Controlling structure to kblockd */ @@ -112,37 +107,6 @@ static void blk_queue_congestion_threshold(struct request_queue *q) q->nr_congestion_off = nr; } -/* - * A queue has just exitted congestion. Note this in the global counter of - * congested queues, and wake up anyone who was waiting for requests to be - * put back. - */ -void blk_clear_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - clear_bit(bit, &q->backing_dev_info.state); - smp_mb__after_clear_bit(); - if (waitqueue_active(wqh)) - wake_up(wqh); -} -EXPORT_SYMBOL(blk_clear_queue_congested); - -/* - * A queue has just entered congestion. Flag that in the queue's VM-visible - * state flags and increment the global gounter of congested queues. - */ -void blk_set_queue_congested(request_queue_t *q, int rw) -{ - enum bdi_state bit; - - bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; - set_bit(bit, &q->backing_dev_info.state); -} -EXPORT_SYMBOL(blk_set_queue_congested); - /** * blk_get_backing_dev_info - get the address of a queue's backing_dev_info * @bdev: device @@ -2755,41 +2719,6 @@ void blk_end_sync_rq(struct request *rq, int error) } EXPORT_SYMBOL(blk_end_sync_rq); -/** - * blk_congestion_wait - wait for a queue to become uncongested - * @rw: READ or WRITE - * @timeout: timeout in jiffies - * - * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion. - * If no queues are congested then just wait for the next request to be - * returned. - */ -long blk_congestion_wait(int rw, long timeout) -{ - long ret; - DEFINE_WAIT(wait); - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); - ret = io_schedule_timeout(timeout); - finish_wait(wqh, &wait); - return ret; -} - -EXPORT_SYMBOL(blk_congestion_wait); - -/** - * blk_congestion_end - wake up sleepers on a congestion queue - * @rw: READ or WRITE - */ -void blk_congestion_end(int rw) -{ - wait_queue_head_t *wqh = &congestion_wqh[rw]; - - if (waitqueue_active(wqh)) - wake_up(wqh); -} - /* * Has to be called with the request spinlock acquired */ diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 655d816760e5..a625576fdeeb 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -602,7 +603,7 @@ static void process_write(struct crypt_io *io) /* out of memory -> run queues */ if (remaining) - blk_congestion_wait(bio_data_dir(clone), HZ/100); + congestion_wait(bio_data_dir(clone), HZ/100); } } diff --git a/fs/fat/file.c b/fs/fat/file.c index f4b8f8b3fbdd..8337451e7897 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -13,6 +13,7 @@ #include #include #include +#include #include int fat_generic_ioctl(struct inode *inode, struct file *filp, @@ -118,7 +119,7 @@ static int fat_file_release(struct inode *inode, struct file *filp) if ((filp->f_mode & FMODE_WRITE) && MSDOS_SB(inode->i_sb)->options.flush) { fat_flush_inodes(inode->i_sb, inode, NULL); - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } return 0; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f6675d2c386c..ca92ac36fe9d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -57,6 +57,8 @@ #include #include #include +#include + #include #include @@ -395,7 +397,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); - writeback_congestion_end(); + congestion_end(WRITE); return err; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index ad8cbc49883a..85ce23268302 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -53,6 +53,7 @@ #include #include #include +#include /* gets a struct reiserfs_journal_list * from a list head */ #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \ @@ -970,7 +971,7 @@ int reiserfs_async_progress_wait(struct super_block *s) DEFINE_WAIT(wait); struct reiserfs_journal *j = SB_JOURNAL(s); if (atomic_read(&j->j_async_throttle)) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); return 0; } diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index d59737589815..004baf600611 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "time.h" #include "kmem.h" @@ -53,7 +54,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } @@ -131,7 +132,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 9bbadafdcb00..db5f5a3608ca 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "xfs_linux.h" STATIC kmem_zone_t *xfs_buf_zone; @@ -395,7 +396,7 @@ _xfs_buf_lookup_pages( XFS_STATS_INC(xb_page_retries); xfsbufd_wakeup(0, gfp_mask); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto retry; } diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index f7a1390d67f5..7011d6255593 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -10,6 +10,8 @@ #include +struct page; + /* * Bits in backing_dev_info.state */ @@ -88,6 +90,11 @@ static inline int bdi_rw_congested(struct backing_dev_info *bdi) (1 << BDI_write_congested)); } +void clear_bdi_congested(struct backing_dev_info *bdi, int rw); +void set_bdi_congested(struct backing_dev_info *bdi, int rw); +long congestion_wait(int rw, long timeout); +void congestion_end(int rw); + #define bdi_cap_writeback_dirty(bdi) \ (!((bdi)->capabilities & BDI_CAP_NO_WRITEBACK)) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 9575e3a5ff2a..7bfcde2d5578 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -651,8 +651,26 @@ extern void blk_recount_segments(request_queue_t *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern int sg_scsi_ioctl(struct file *, struct request_queue *, struct gendisk *, struct scsi_ioctl_command __user *); -extern void blk_clear_queue_congested(request_queue_t *q, int rw); -extern void blk_set_queue_congested(request_queue_t *q, int rw); + +/* + * A queue has just exitted congestion. Note this in the global counter of + * congested queues, and wake up anyone who was waiting for requests to be + * put back. + */ +static inline void blk_clear_queue_congested(request_queue_t *q, int rw) +{ + clear_bdi_congested(&q->backing_dev_info, rw); +} + +/* + * A queue has just entered congestion. Flag that in the queue's VM-visible + * state flags and increment the global gounter of congested queues. + */ +static inline void blk_set_queue_congested(request_queue_t *q, int rw) +{ + set_bdi_congested(&q->backing_dev_info, rw); +} + extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); extern void blk_sync_queue(struct request_queue *q); @@ -767,10 +785,8 @@ extern int blk_queue_init_tags(request_queue_t *, int, struct blk_queue_tag *); extern void blk_queue_free_tags(request_queue_t *); extern int blk_queue_resize_tags(request_queue_t *, int); extern void blk_queue_invalidate_tags(request_queue_t *); -extern long blk_congestion_wait(int rw, long timeout); extern struct blk_queue_tag *blk_init_tags(int); extern void blk_free_tags(struct blk_queue_tag *); -extern void blk_congestion_end(int rw); static inline struct request *blk_map_queue_find_tag(struct blk_queue_tag *bqt, int tag) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index a341c8032866..fc35e6bdfb93 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -85,7 +85,6 @@ int wakeup_pdflush(long nr_pages); void laptop_io_completion(void); void laptop_sync_completion(void); void throttle_vm_writeout(void); -void writeback_congestion_end(void); /* These are exported to sysctl. */ extern int dirty_background_ratio; diff --git a/mm/Makefile b/mm/Makefile index 12b3a4eee88d..f3c077eb0b8e 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -10,7 +10,8 @@ mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \ obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \ page_alloc.o page-writeback.o pdflush.o \ readahead.o swap.o truncate.o vmscan.o \ - prio_tree.o util.o mmzone.o vmstat.o $(mmu-y) + prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \ + $(mmu-y) ifeq ($(CONFIG_MMU)$(CONFIG_BLOCK),yy) obj-y += bounce.o diff --git a/mm/backing-dev.c b/mm/backing-dev.c new file mode 100644 index 000000000000..f50a2811f9dc --- /dev/null +++ b/mm/backing-dev.c @@ -0,0 +1,69 @@ + +#include +#include +#include +#include +#include + +static wait_queue_head_t congestion_wqh[2] = { + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]), + __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1]) + }; + + +void clear_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + clear_bit(bit, &bdi->state); + smp_mb__after_clear_bit(); + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(clear_bdi_congested); + +void set_bdi_congested(struct backing_dev_info *bdi, int rw) +{ + enum bdi_state bit; + + bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested; + set_bit(bit, &bdi->state); +} +EXPORT_SYMBOL(set_bdi_congested); + +/** + * congestion_wait - wait for a backing_dev to become uncongested + * @rw: READ or WRITE + * @timeout: timeout in jiffies + * + * Waits for up to @timeout jiffies for a backing_dev (any backing_dev) to exit + * write congestion. If no backing_devs are congested then just wait for the + * next write to be completed. + */ +long congestion_wait(int rw, long timeout) +{ + long ret; + DEFINE_WAIT(wait); + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE); + ret = io_schedule_timeout(timeout); + finish_wait(wqh, &wait); + return ret; +} +EXPORT_SYMBOL(congestion_wait); + +/** + * congestion_end - wake up sleepers on a congested backing_dev_info + * @rw: READ or WRITE + */ +void congestion_end(int rw) +{ + wait_queue_head_t *wqh = &congestion_wqh[rw]; + + if (waitqueue_active(wqh)) + wake_up(wqh); +} +EXPORT_SYMBOL(congestion_end); diff --git a/mm/page-writeback.c b/mm/page-writeback.c index a0f339057449..8d9b19f239c3 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -222,7 +222,7 @@ static void balance_dirty_pages(struct address_space *mapping) if (pages_written >= write_chunk) break; /* We've done our duty */ } - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } if (nr_reclaimable + global_page_state(NR_WRITEBACK) @@ -314,7 +314,7 @@ void throttle_vm_writeout(void) if (global_page_state(NR_UNSTABLE_NFS) + global_page_state(NR_WRITEBACK) <= dirty_thresh) break; - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } } @@ -351,7 +351,7 @@ static void background_writeout(unsigned long _min_pages) min_pages -= MAX_WRITEBACK_PAGES - wbc.nr_to_write; if (wbc.nr_to_write > 0 || wbc.pages_skipped > 0) { /* Wrote less than expected */ - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); if (!wbc.encountered_congestion) break; } @@ -422,7 +422,7 @@ static void wb_kupdate(unsigned long arg) writeback_inodes(&wbc); if (wbc.nr_to_write > 0) { if (wbc.encountered_congestion) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); else break; /* All the old data is written */ } @@ -955,15 +955,6 @@ int test_set_page_writeback(struct page *page) } EXPORT_SYMBOL(test_set_page_writeback); -/* - * Wakes up tasks that are being throttled due to writeback congestion - */ -void writeback_congestion_end(void) -{ - blk_congestion_end(WRITE); -} -EXPORT_SYMBOL(writeback_congestion_end); - /* * Return true if any of the pages in the mapping are marged with the * passed tag. diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 40db96a655d0..afee38f04d84 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -1050,7 +1051,7 @@ nofail_alloc: if (page) goto got_pg; if (gfp_mask & __GFP_NOFAIL) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto nofail_alloc; } } @@ -1113,7 +1114,7 @@ rebalance: do_retry = 1; } if (do_retry) { - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto rebalance; } diff --git a/mm/shmem.c b/mm/shmem.c index b378f66cf2f9..4959535fc14c 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include @@ -1131,7 +1132,7 @@ repeat: page_cache_release(swappage); if (error == -ENOMEM) { /* let kswapd refresh zone for GFP_ATOMICs */ - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } goto repeat; } diff --git a/mm/vmscan.c b/mm/vmscan.c index af73c14f9d88..f05527bf792b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1059,7 +1059,7 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) /* Take a nap, wait for some writeback to complete */ if (sc.nr_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } /* top priority shrink_caches still had more to do? don't OOM, then */ if (!sc.all_unreclaimable) @@ -1214,7 +1214,7 @@ scan: * another pass across the zones. */ if (total_scanned && priority < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); /* * We do this so kswapd doesn't build up large priorities for @@ -1458,7 +1458,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages) goto out; if (sc.nr_scanned && prio < DEF_PRIORITY - 2) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); } lru_pages = 0; -- cgit v1.2.3 From 34e856e6a522a8fc0feba7497f5b05aeaa13d473 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 19 Oct 2006 23:28:17 -0700 Subject: [PATCH] Make userspace proof contains the constants for personality(2) but also some defintions that are useless or even harmful in userspace such as the personality() macro. Signed-off-by: Ralf Baechle Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/personality.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/personality.h b/include/linux/personality.h index 80d780e5a8f5..bf4cf2080e5c 100644 --- a/include/linux/personality.h +++ b/include/linux/personality.h @@ -1,6 +1,8 @@ #ifndef _LINUX_PERSONALITY_H #define _LINUX_PERSONALITY_H +#ifdef __KERNEL__ + /* * Handling of different ABIs (personalities). */ @@ -12,6 +14,8 @@ extern int register_exec_domain(struct exec_domain *); extern int unregister_exec_domain(struct exec_domain *); extern int __set_personality(unsigned long); +#endif /* __KERNEL__ */ + /* * Flags for bug emulation. * @@ -71,6 +75,7 @@ enum { PER_MASK = 0x00ff, }; +#ifdef __KERNEL__ /* * Description of an execution domain. @@ -111,4 +116,6 @@ struct exec_domain { #define set_personality(pers) \ ((current->personality == pers) ? 0 : __set_personality(pers)) +#endif /* __KERNEL__ */ + #endif /* _LINUX_PERSONALITY_H */ -- cgit v1.2.3 From 145fc655a1ceabda76cf2ad74f7cf96863c65b65 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 19 Oct 2006 23:28:28 -0700 Subject: [PATCH] genirq: clean up irq-flow-type naming, fix Re-add the set_irq_chip_and_handler() prototype, it's still widely used. Signed-off-by: Ingo Molnar Cc: Olaf Hering Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irq.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/irq.h b/include/linux/irq.h index 775f5a7da493..52fc4052a0ae 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -321,6 +321,9 @@ extern int can_request_irq(unsigned int irq, unsigned long irqflags); extern struct irq_chip no_irq_chip; extern struct irq_chip dummy_irq_chip; +extern void +set_irq_chip_and_handler(unsigned int irq, struct irq_chip *chip, + irq_flow_handler_t handle); extern void set_irq_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, irq_flow_handler_t handle, const char *name); -- cgit v1.2.3 From 8ac773b4f73afa6fd66695131103944b975d5d5c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 19 Oct 2006 23:28:32 -0700 Subject: [PATCH] OOM killer meets userspace headers Despite mm.h is not being exported header, it does contain one thing which is part of userspace ABI -- value disabling OOM killer for given process. So, a) create and export include/linux/oom.h b) move OOM_DISABLE define there. c) turn bounding values of /proc/$PID/oom_adj into defines and export them too. Note: mass __KERNEL__ removal will be done later. Signed-off-by: Alexey Dobriyan Cc: Nick Piggin Cc: David Woodhouse Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 4 +++- include/linux/Kbuild | 1 + include/linux/mm.h | 3 --- include/linux/oom.h | 10 ++++++++++ mm/oom_kill.c | 1 + 5 files changed, 15 insertions(+), 4 deletions(-) create mode 100644 include/linux/oom.h (limited to 'include/linux') diff --git a/fs/proc/base.c b/fs/proc/base.c index 26a8f8416b79..8df27401d292 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -72,6 +72,7 @@ #include #include #include +#include #include "internal.h" /* NOTE: @@ -689,7 +690,8 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, if (copy_from_user(buffer, buf, count)) return -EFAULT; oom_adjust = simple_strtol(buffer, &end, 0); - if ((oom_adjust < -16 || oom_adjust > 15) && oom_adjust != OOM_DISABLE) + if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) && + oom_adjust != OOM_DISABLE) return -EINVAL; if (*end == '\n') end++; diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 5114ff18101d..a1155a2beb32 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -120,6 +120,7 @@ header-y += netrom.h header-y += nfs2.h header-y += nfs4_mount.h header-y += nfs_mount.h +header-y += oom.h header-y += param.h header-y += pci_ids.h header-y += pci_regs.h diff --git a/include/linux/mm.h b/include/linux/mm.h index 5a6068ff5556..d538de901965 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1115,9 +1115,6 @@ int in_gate_area_no_task(unsigned long addr); #define in_gate_area(task, addr) ({(void)task; in_gate_area_no_task(addr);}) #endif /* __HAVE_ARCH_GATE_AREA */ -/* /proc//oom_adj set to -17 protects from the oom-killer */ -#define OOM_DISABLE -17 - int drop_caches_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask, diff --git a/include/linux/oom.h b/include/linux/oom.h new file mode 100644 index 000000000000..ad76463629a0 --- /dev/null +++ b/include/linux/oom.h @@ -0,0 +1,10 @@ +#ifndef __INCLUDE_LINUX_OOM_H +#define __INCLUDE_LINUX_OOM_H + +/* /proc//oom_adj set to -17 protects from the oom-killer */ +#define OOM_DISABLE (-17) +/* inclusive */ +#define OOM_ADJUST_MIN (-16) +#define OOM_ADJUST_MAX 15 + +#endif diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 20f41b082e16..2e3ce3a928b9 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -15,6 +15,7 @@ * kernel subsystems and hints as to where to find out what things do. */ +#include #include #include #include -- cgit v1.2.3 From cd9ae2b6a75bb1fa0d370929c2d7a7da1ed719d9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 19 Oct 2006 23:28:40 -0700 Subject: [PATCH] NFS: Deal with failure of invalidate_inode_pages2() If invalidate_inode_pages2() fails, then it should in principle just be because the current process was signalled. In that case, we just want to ensure that the inode's page cache remains marked as invalid. Also add a helper to allow the O_DIRECT code to simply mark the page cache as invalid once it is finished writing, instead of calling invalidate_inode_pages2() itself. Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 6 ++++-- fs/nfs/direct.c | 13 ++----------- fs/nfs/inode.c | 28 +++++++++++++++++++++++----- include/linux/nfs_fs.h | 1 + 4 files changed, 30 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 481f8892a919..58d44057813e 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -203,8 +203,10 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) * Note: assumes we have exclusive access to this mapping either * through inode->i_mutex or some other mechanism. */ - if (page->index == 0) - invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1); + if (page->index == 0 && invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1) < 0) { + /* Should never happen */ + nfs_zap_mapping(inode, inode->i_mapping); + } unlock_page(page); return 0; error: diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 1e873fcab947..bdfabf854a51 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -497,6 +497,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode if (dreq->commit_data != NULL) nfs_commit_free(dreq->commit_data); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } } @@ -517,6 +518,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode { nfs_end_data_update(inode); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } #endif @@ -830,17 +832,6 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov, retval = nfs_direct_write(iocb, (unsigned long) buf, count, pos); - /* - * XXX: nfs_end_data_update() already ensures this file's - * cached data is subsequently invalidated. Do we really - * need to call invalidate_inode_pages2() again here? - * - * For aio writes, this invalidation will almost certainly - * occur before the writes complete. Kind of racey. - */ - if (mapping->nrpages) - invalidate_inode_pages2(mapping); - if (retval > 0) iocb->ki_pos = pos + retval; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index bc9376ca86cd..9979ad1cf8eb 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -131,6 +131,15 @@ void nfs_zap_caches(struct inode *inode) spin_unlock(&inode->i_lock); } +void nfs_zap_mapping(struct inode *inode, struct address_space *mapping) +{ + if (mapping->nrpages != 0) { + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; + spin_unlock(&inode->i_lock); + } +} + static void nfs_zap_acl_cache(struct inode *inode) { void (*clear_acl_cache)(struct inode *); @@ -671,13 +680,20 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode)) ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + goto out; if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { - nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); - if (S_ISREG(inode->i_mode)) - nfs_sync_mapping(mapping); - invalidate_inode_pages2(mapping); - + if (mapping->nrpages != 0) { + if (S_ISREG(inode->i_mode)) { + ret = nfs_sync_mapping(mapping); + if (ret < 0) + goto out; + } + ret = invalidate_inode_pages2(mapping); + if (ret < 0) + goto out; + } spin_lock(&inode->i_lock); nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; if (S_ISDIR(inode->i_mode)) { @@ -687,10 +703,12 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) } spin_unlock(&inode->i_lock); + nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); } +out: return ret; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 76ff54846ada..6b2de1be5815 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -290,6 +290,7 @@ static inline int nfs_verify_change_attribute(struct inode *inode, unsigned long * linux/fs/nfs/inode.c */ extern int nfs_sync_mapping(struct address_space *mapping); +extern void nfs_zap_mapping(struct inode *inode, struct address_space *mapping); extern void nfs_zap_caches(struct inode *); extern struct inode *nfs_fhget(struct super_block *, struct nfs_fh *, struct nfs_fattr *); -- cgit v1.2.3 From 7111c66e4e70588c9602035a4996c9cdc2087d2d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:45 -0700 Subject: [PATCH] fix svc_procfunc declaration svc_procfunc instances return __be32, not int Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/svc4proc.c | 40 ++++++++++++++++++++-------------------- fs/lockd/svcproc.c | 40 ++++++++++++++++++++-------------------- fs/nfs/callback_xdr.c | 4 ++-- fs/nfsd/nfs2acl.c | 10 +++++----- fs/nfsd/nfs3acl.c | 6 +++--- fs/nfsd/nfs3proc.c | 44 ++++++++++++++++++++++---------------------- fs/nfsd/nfs4proc.c | 4 ++-- fs/nfsd/nfsproc.c | 32 ++++++++++++++++---------------- include/linux/sunrpc/svc.h | 2 +- 9 files changed, 91 insertions(+), 91 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 399ad11b97be..4e719860b4bf 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -68,7 +68,7 @@ no_locks: /* * NULL: Test for presence of service */ -static int +static __be32 nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -78,7 +78,7 @@ nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -107,7 +107,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -150,7 +150,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -183,7 +183,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -217,7 +217,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlm4svc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -253,12 +253,12 @@ static const struct rpc_call_ops nlm4svc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -282,35 +282,35 @@ static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *a return rpc_success; } -static int nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_TEST_RES, argp, nlm4svc_proc_test); } -static int nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlm4svc_proc_lock); } -static int nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlm4svc_proc_cancel); } -static int nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: UNLOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlm4svc_proc_unlock); } -static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: GRANTED_MSG called\n"); @@ -320,7 +320,7 @@ static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *arg /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -353,7 +353,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -386,7 +386,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -399,7 +399,7 @@ nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -417,7 +417,7 @@ nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -446,7 +446,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlm4svc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 6a931f4ab75c..db8d85c32d29 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -96,7 +96,7 @@ no_locks: /* * NULL: Test for presence of service */ -static int +static __be32 nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -106,7 +106,7 @@ nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -136,7 +136,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -179,7 +179,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -212,7 +212,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -246,7 +246,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlmsvc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -282,12 +282,12 @@ static const struct rpc_call_ops nlmsvc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -311,28 +311,28 @@ static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *ar return rpc_success; } -static int nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_TEST_RES, argp, nlmsvc_proc_test); } -static int nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlmsvc_proc_lock); } -static int nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlmsvc_proc_cancel); } -static int +static __be32 nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -340,7 +340,7 @@ nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, return nlmsvc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlmsvc_proc_unlock); } -static int +static __be32 nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -351,7 +351,7 @@ nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -384,7 +384,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -417,7 +417,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -430,7 +430,7 @@ nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -448,7 +448,7 @@ nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -477,7 +477,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlmsvc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 29f932192054..5998d0c71757 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -36,7 +36,7 @@ struct callback_op { static struct callback_op callback_ops[]; -static int nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) { return htonl(NFS4_OK); } @@ -399,7 +399,7 @@ static unsigned process_op(struct svc_rqst *rqstp, /* * Decode, process and encode a COMPOUND */ -static int nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) { struct cb_compound_hdr_arg hdr_arg; struct cb_compound_hdr_res hdr_res; diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 9187755661df..8d48616882c1 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -21,7 +21,7 @@ /* * NULL call. */ -static int +static __be32 nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -30,7 +30,7 @@ nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsacld_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; @@ -97,7 +97,7 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsacld_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd_attrstat *resp) { @@ -128,7 +128,7 @@ static int nfsacld_proc_setacl(struct svc_rqst * rqstp, /* * Check file attributes */ -static int nfsacld_proc_getattr(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getattr(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); @@ -140,7 +140,7 @@ static int nfsacld_proc_getattr(struct svc_rqst * rqstp, /* * Check file access */ -static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, +static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { int nfserr; diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index d4bdc00c1169..ed6e2c27b5e8 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -19,7 +19,7 @@ /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -28,7 +28,7 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsd3_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; @@ -93,7 +93,7 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsd3_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd3_attrstat *resp) { diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index a5ebc7dbb384..a12663fdfe16 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -43,7 +43,7 @@ static int nfs3_ftypes[] = { /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -52,7 +52,7 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get a file's attributes */ -static int +static __be32 nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd3_attrstat *resp) { @@ -76,7 +76,7 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, /* * Set a file's attributes */ -static int +static __be32 nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, struct nfsd3_attrstat *resp) { @@ -94,7 +94,7 @@ nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, /* * Look up a path name component */ -static int +static __be32 nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_diropres *resp) { @@ -118,7 +118,7 @@ nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Check file access */ -static int +static __be32 nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { @@ -137,7 +137,7 @@ nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, struct nfsd3_readlinkres *resp) { @@ -155,7 +155,7 @@ nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, /* * Read a portion of a file. */ -static int +static __be32 nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, struct nfsd3_readres *resp) { @@ -195,7 +195,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, /* * Write data to a file */ -static int +static __be32 nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, struct nfsd3_writeres *resp) { @@ -223,7 +223,7 @@ nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, * At least in theory; we'll see how it fares in practice when the * first reports about SunOS compatibility problems start to pour in... */ -static int +static __be32 nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { @@ -265,7 +265,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* * Make directory. This operation is not idempotent. */ -static int +static __be32 nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { @@ -285,7 +285,7 @@ nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, struct nfsd3_diropres *resp) { @@ -307,7 +307,7 @@ nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, /* * Make socket/fifo/device. */ -static int +static __be32 nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, struct nfsd3_diropres *resp) { @@ -343,7 +343,7 @@ nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, /* * Remove file/fifo/socket etc. */ -static int +static __be32 nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { @@ -363,7 +363,7 @@ nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { @@ -379,7 +379,7 @@ nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, struct nfsd3_renameres *resp) { @@ -401,7 +401,7 @@ nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, struct nfsd3_linkres *resp) { @@ -424,7 +424,7 @@ nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { @@ -459,7 +459,7 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, * Read a portion of a directory, including file handles and attrs. * For now, we choose to ignore the dircount parameter. */ -static int +static __be32 nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { @@ -517,7 +517,7 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, /* * Get file system stats */ -static int +static __be32 nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsstatres *resp) { @@ -534,7 +534,7 @@ nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get file system info */ -static int +static __be32 nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsinfores *resp) { @@ -576,7 +576,7 @@ nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get pathconf info for the specified file */ -static int +static __be32 nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_pathconfres *resp) { @@ -619,7 +619,7 @@ nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Commit a file (range) to stable storage. */ -static int +static __be32 nfsd3_proc_commit(struct svc_rqst * rqstp, struct nfsd3_commitargs *argp, struct nfsd3_commitres *resp) { diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index d1fac6872c44..795ad6c5cb2c 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -715,7 +715,7 @@ out_kfree: /* * NULL call. */ -static int +static __be32 nfsd4_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -731,7 +731,7 @@ static inline void nfsd4_increment_op_stats(u32 opnum) /* * COMPOUND call. */ -static int +static __be32 nfsd4_proc_compound(struct svc_rqst *rqstp, struct nfsd4_compoundargs *args, struct nfsd4_compoundres *resp) diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 9ee1dab5d44a..09030afd7249 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -30,7 +30,7 @@ typedef struct svc_buf svc_buf; #define NFSDDBG_FACILITY NFSDDBG_PROC -static int +static __be32 nfsd_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -56,7 +56,7 @@ nfsd_return_dirop(int err, struct nfsd_diropres *resp) * Get a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { @@ -72,7 +72,7 @@ nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, * Set a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, struct nfsd_attrstat *resp) { @@ -92,7 +92,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, * doesn't exist yet. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, struct nfsd_diropres *resp) { @@ -112,7 +112,7 @@ nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, struct nfsd_readlinkres *resp) { @@ -132,7 +132,7 @@ nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, * Read a portion of a file. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, struct nfsd_readres *resp) { @@ -172,7 +172,7 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, * Write data to a file * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, struct nfsd_attrstat *resp) { @@ -197,7 +197,7 @@ nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, * and the actual create() call in compliance with VFS protocols. * N.B. After this call _both_ argp->fh and resp->fh need an fh_put */ -static int +static __be32 nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { @@ -348,7 +348,7 @@ done: return nfsd_return_dirop(nfserr, resp); } -static int +static __be32 nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { @@ -363,7 +363,7 @@ nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, void *resp) { @@ -381,7 +381,7 @@ nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, void *resp) { @@ -401,7 +401,7 @@ nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, void *resp) { @@ -430,7 +430,7 @@ nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, * Make directory. This operation is not idempotent. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { @@ -454,7 +454,7 @@ nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { @@ -470,7 +470,7 @@ nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, struct nfsd_readdirres *resp) { @@ -509,7 +509,7 @@ nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, /* * Get file system info */ -static int +static __be32 nfsd_proc_statfs(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_statfsres *resp) { diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 9c9a8ad92477..965d6c20086e 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -335,7 +335,7 @@ struct svc_version { /* * RPC procedure info */ -typedef int (*svc_procfunc)(struct svc_rqst *, void *argp, void *resp); +typedef __be32 (*svc_procfunc)(struct svc_rqst *, void *argp, void *resp); struct svc_procedure { svc_procfunc pc_func; /* process the request */ kxdrproc_t pc_decode; /* XDR decode args */ -- cgit v1.2.3 From 52921e02a4f4163a7b1f4b5dde71e1debc71de4a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:46 -0700 Subject: [PATCH] lockd endianness annotations Signed-off-by: Alexey Dobriyan Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/clntlock.c | 4 +-- fs/lockd/mon.c | 12 +++---- fs/lockd/svc4proc.c | 4 +-- fs/lockd/svclock.c | 10 +++--- fs/lockd/svcproc.c | 8 ++--- fs/lockd/svcshare.c | 4 +-- fs/lockd/svcsubs.c | 4 +-- fs/lockd/xdr.c | 76 +++++++++++++++++++++--------------------- fs/lockd/xdr4.c | 80 ++++++++++++++++++++++----------------------- include/linux/lockd/lockd.h | 12 +++---- include/linux/lockd/share.h | 4 +-- include/linux/lockd/xdr.h | 26 +++++++-------- include/linux/lockd/xdr4.h | 26 +++++++-------- 13 files changed, 135 insertions(+), 135 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index e8c7765419e8..b85a0ad2cfb6 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -100,12 +100,12 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) /* * The server lockd has called us back to tell us the lock was granted */ -u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) +__be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) { const struct file_lock *fl = &lock->fl; const struct nfs_fh *fh = &lock->fh; struct nlm_wait *block; - u32 res = nlm_lck_denied; + __be32 res = nlm_lck_denied; /* * Look up blocked request based on arguments. diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index e0179f8c327f..eb243edf8932 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -148,8 +148,8 @@ nsm_create(void) * XDR functions for NSM. */ -static u32 * -xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +static __be32 * +xdr_encode_common(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { char buffer[20], *name; @@ -176,7 +176,7 @@ xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_mon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -192,7 +192,7 @@ xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_unmon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -202,7 +202,7 @@ xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat_res(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->status = ntohl(*p++); resp->state = ntohl(*p++); @@ -212,7 +212,7 @@ xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) } static int -xdr_decode_stat(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->state = ntohl(*p++); return 0; diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 4e719860b4bf..0ce5c81ff507 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -24,14 +24,14 @@ /* * Obtain client and file from arguments */ -static u32 +static __be32 nlm4svc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error = 0; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 814c6064c9e0..7e219b938552 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -334,13 +334,13 @@ static void nlmsvc_freegrantargs(struct nlm_rqst *call) * Attempt to establish a lock, and if it can't be granted, block it * if required. */ -u32 +__be32 nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_lock *lock, int wait, struct nlm_cookie *cookie) { struct nlm_block *block, *newblock = NULL; int error; - u32 ret; + __be32 ret; dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n", file->f_file->f_dentry->d_inode->i_sb->s_id, @@ -415,7 +415,7 @@ out: /* * Test for presence of a conflicting lock. */ -u32 +__be32 nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, struct nlm_lock *conflock) { @@ -448,7 +448,7 @@ nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, * afterwards. In this case the block will still be there, and hence * must be removed. */ -u32 +__be32 nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) { int error; @@ -476,7 +476,7 @@ nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) * be in progress. * The calling procedure must check whether the file can be closed. */ -u32 +__be32 nlmsvc_cancel_blocked(struct nlm_file *file, struct nlm_lock *lock) { struct nlm_block *block; diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index db8d85c32d29..32e99a6e8dca 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -22,8 +22,8 @@ #define NLMDBG_FACILITY NLMDBG_CLIENT #ifdef CONFIG_LOCKD_V4 -static u32 -cast_to_nlm(u32 status, u32 vers) +static __be32 +cast_to_nlm(__be32 status, u32 vers) { /* Note: status is assumed to be in network byte order !!! */ if (vers != 4){ @@ -52,14 +52,14 @@ cast_to_nlm(u32 status, u32 vers) /* * Obtain client and file from arguments */ -static u32 +static __be32 nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error = 0; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) diff --git a/fs/lockd/svcshare.c b/fs/lockd/svcshare.c index b9926ce8782e..6220dc2a3f2c 100644 --- a/fs/lockd/svcshare.c +++ b/fs/lockd/svcshare.c @@ -23,7 +23,7 @@ nlm_cmp_owner(struct nlm_share *share, struct xdr_netobj *oh) && !memcmp(share->s_owner.data, oh->data, oh->len); } -u32 +__be32 nlmsvc_share_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { @@ -64,7 +64,7 @@ update: /* * Delete a share. */ -u32 +__be32 nlmsvc_unshare_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 7dac96e6c82c..e83024e16042 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -78,14 +78,14 @@ static inline unsigned int file_hash(struct nfs_fh *f) * This is not quite right, but for now, we assume the client performs * the proper R/W checking. */ -u32 +__be32 nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, struct nfs_fh *f) { struct hlist_node *pos; struct nlm_file *file; unsigned int hash; - u32 nfserr; + __be32 nfserr; nlm_debug_print_fh("nlm_file_lookup", f); diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c index 61c46facf257..b7c949256e5a 100644 --- a/fs/lockd/xdr.c +++ b/fs/lockd/xdr.c @@ -43,7 +43,7 @@ loff_t_to_s32(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 *nlm_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -69,8 +69,8 @@ static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static inline u32 * -nlm_encode_cookie(u32 *p, struct nlm_cookie *c) +static inline __be32 * +nlm_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -78,8 +78,8 @@ nlm_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm_decode_fh(__be32 *p, struct nfs_fh *f) { unsigned int len; @@ -95,8 +95,8 @@ nlm_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32 * -nlm_encode_fh(u32 *p, struct nfs_fh *f) +static inline __be32 * +nlm_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(NFS2_FHSIZE); memcpy(p, f->data, NFS2_FHSIZE); @@ -106,20 +106,20 @@ nlm_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static inline u32 * -nlm_decode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static inline u32 * -nlm_encode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; s32 start, len, end; @@ -153,8 +153,8 @@ nlm_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s32 start, len; @@ -184,8 +184,8 @@ nlm_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm_encode_testres(__be32 *p, struct nlm_res *resp) { s32 start, len; @@ -221,7 +221,7 @@ nlm_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -238,7 +238,7 @@ nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return 0; @@ -246,7 +246,7 @@ nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -266,7 +266,7 @@ nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -282,7 +282,7 @@ nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm_decode_cookie(p, &argp->cookie)) || !(p = nlm_decode_lock(p, &argp->lock))) @@ -292,7 +292,7 @@ nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -313,7 +313,7 @@ nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -323,7 +323,7 @@ nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -332,7 +332,7 @@ nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlmsvc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -344,7 +344,7 @@ nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlmsvc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -357,7 +357,7 @@ nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return 0; @@ -366,13 +366,13 @@ nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlmsvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -389,7 +389,7 @@ nlmclt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) #endif static int -nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -403,7 +403,7 @@ nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; @@ -438,7 +438,7 @@ nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -455,7 +455,7 @@ nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -470,7 +470,7 @@ nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -483,7 +483,7 @@ nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return -EIO; @@ -493,7 +493,7 @@ nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return -EIO; @@ -502,7 +502,7 @@ nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c index 36eb175ec335..f4c0b2b9f75a 100644 --- a/fs/lockd/xdr4.c +++ b/fs/lockd/xdr4.c @@ -44,8 +44,8 @@ loff_t_to_s64(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 * -nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -71,8 +71,8 @@ nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -80,8 +80,8 @@ nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_decode_fh(__be32 *p, struct nfs_fh *f) { memset(f->data, 0, sizeof(f->data)); f->size = ntohl(*p++); @@ -95,8 +95,8 @@ nlm4_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(f->size); } -static u32 * -nlm4_encode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(f->size); if (f->size) p[XDR_QUADLEN(f->size)-1] = 0; /* don't leak anything */ @@ -107,20 +107,20 @@ nlm4_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static u32 * -nlm4_decode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static u32 * -nlm4_encode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm4_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 len, start, end; @@ -153,8 +153,8 @@ nlm4_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm4_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 start, len; @@ -185,8 +185,8 @@ nlm4_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm4_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm4_encode_testres(__be32 *p, struct nlm_res *resp) { s64 start, len; @@ -227,7 +227,7 @@ nlm4_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -244,7 +244,7 @@ nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return 0; @@ -252,7 +252,7 @@ nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -272,7 +272,7 @@ nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -288,7 +288,7 @@ nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm4_decode_cookie(p, &argp->cookie)) || !(p = nlm4_decode_lock(p, &argp->lock))) @@ -298,7 +298,7 @@ nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -319,7 +319,7 @@ nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -329,7 +329,7 @@ nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -338,7 +338,7 @@ nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlm4svc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -350,7 +350,7 @@ nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlm4svc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -363,7 +363,7 @@ nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return 0; @@ -372,13 +372,13 @@ nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -388,14 +388,14 @@ nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) */ #ifdef NLMCLNT_SUPPORT_SHARES static int -nlm4clt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) +nlm4clt_decode_void(struct rpc_rqst *req, __be32 *p, void *ptr) { return 0; } #endif static int -nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -409,7 +409,7 @@ nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; @@ -444,7 +444,7 @@ nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -461,7 +461,7 @@ nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -476,7 +476,7 @@ nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -489,7 +489,7 @@ nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return -EIO; @@ -499,7 +499,7 @@ nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return -EIO; @@ -508,7 +508,7 @@ nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 2909619c0295..862d9730a60d 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -154,7 +154,7 @@ int nlm_async_reply(struct nlm_rqst *, u32, const struct rpc_call_ops *); struct nlm_wait * nlmclnt_prepare_block(struct nlm_host *host, struct file_lock *fl); void nlmclnt_finish_block(struct nlm_wait *block); int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout); -u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *); +__be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *); void nlmclnt_recovery(struct nlm_host *); int nlmclnt_reclaim(struct nlm_host *, struct file_lock *); void nlmclnt_next_cookie(struct nlm_cookie *); @@ -184,12 +184,12 @@ typedef int (*nlm_host_match_fn_t)(struct nlm_host *cur, struct nlm_host *ref) /* * Server-side lock handling */ -u32 nlmsvc_lock(struct svc_rqst *, struct nlm_file *, +__be32 nlmsvc_lock(struct svc_rqst *, struct nlm_file *, struct nlm_lock *, int, struct nlm_cookie *); -u32 nlmsvc_unlock(struct nlm_file *, struct nlm_lock *); -u32 nlmsvc_testlock(struct nlm_file *, struct nlm_lock *, +__be32 nlmsvc_unlock(struct nlm_file *, struct nlm_lock *); +__be32 nlmsvc_testlock(struct nlm_file *, struct nlm_lock *, struct nlm_lock *); -u32 nlmsvc_cancel_blocked(struct nlm_file *, struct nlm_lock *); +__be32 nlmsvc_cancel_blocked(struct nlm_file *, struct nlm_lock *); unsigned long nlmsvc_retry_blocked(void); void nlmsvc_traverse_blocks(struct nlm_host *, struct nlm_file *, nlm_host_match_fn_t match); @@ -198,7 +198,7 @@ void nlmsvc_grant_reply(struct nlm_cookie *, u32); /* * File handling for the server personality */ -u32 nlm_lookup_file(struct svc_rqst *, struct nlm_file **, +__be32 nlm_lookup_file(struct svc_rqst *, struct nlm_file **, struct nfs_fh *); void nlm_release_file(struct nlm_file *); void nlmsvc_mark_resources(void); diff --git a/include/linux/lockd/share.h b/include/linux/lockd/share.h index cd7816e74c05..630c5bf69b07 100644 --- a/include/linux/lockd/share.h +++ b/include/linux/lockd/share.h @@ -21,9 +21,9 @@ struct nlm_share { u32 s_mode; /* deny mode */ }; -u32 nlmsvc_share_file(struct nlm_host *, struct nlm_file *, +__be32 nlmsvc_share_file(struct nlm_host *, struct nlm_file *, struct nlm_args *); -u32 nlmsvc_unshare_file(struct nlm_host *, struct nlm_file *, +__be32 nlmsvc_unshare_file(struct nlm_host *, struct nlm_file *, struct nlm_args *); void nlmsvc_traverse_shares(struct nlm_host *, struct nlm_file *, nlm_host_match_fn_t); diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h index 66fdae3b490c..29e7d9fc9dad 100644 --- a/include/linux/lockd/xdr.h +++ b/include/linux/lockd/xdr.h @@ -90,19 +90,19 @@ struct nlm_reboot { */ #define NLMSVC_XDRSIZE sizeof(struct nlm_args) -int nlmsvc_decode_testargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_testres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_lockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_cancargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_unlockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_encode_void(struct svc_rqst *, u32 *, void *); -int nlmsvc_decode_void(struct svc_rqst *, u32 *, void *); -int nlmsvc_decode_shareargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_encode_shareres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlmsvc_decode_notify(struct svc_rqst *, u32 *, struct nlm_args *); -int nlmsvc_decode_reboot(struct svc_rqst *, u32 *, struct nlm_reboot *); +int nlmsvc_decode_testargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_testres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_lockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_cancargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_unlockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_encode_void(struct svc_rqst *, __be32 *, void *); +int nlmsvc_decode_void(struct svc_rqst *, __be32 *, void *); +int nlmsvc_decode_shareargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_encode_shareres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlmsvc_decode_notify(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlmsvc_decode_reboot(struct svc_rqst *, __be32 *, struct nlm_reboot *); /* int nlmclt_encode_testargs(struct rpc_rqst *, u32 *, struct nlm_args *); int nlmclt_encode_lockargs(struct rpc_rqst *, u32 *, struct nlm_args *); diff --git a/include/linux/lockd/xdr4.h b/include/linux/lockd/xdr4.h index 3cc1ae25009b..dd12b4c9e613 100644 --- a/include/linux/lockd/xdr4.h +++ b/include/linux/lockd/xdr4.h @@ -23,19 +23,19 @@ -int nlm4svc_decode_testargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_testres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_lockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_cancargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_unlockargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_res(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_encode_void(struct svc_rqst *, u32 *, void *); -int nlm4svc_decode_void(struct svc_rqst *, u32 *, void *); -int nlm4svc_decode_shareargs(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_encode_shareres(struct svc_rqst *, u32 *, struct nlm_res *); -int nlm4svc_decode_notify(struct svc_rqst *, u32 *, struct nlm_args *); -int nlm4svc_decode_reboot(struct svc_rqst *, u32 *, struct nlm_reboot *); +int nlm4svc_decode_testargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_testres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_lockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_cancargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_unlockargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_res(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_encode_void(struct svc_rqst *, __be32 *, void *); +int nlm4svc_decode_void(struct svc_rqst *, __be32 *, void *); +int nlm4svc_decode_shareargs(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_encode_shareres(struct svc_rqst *, __be32 *, struct nlm_res *); +int nlm4svc_decode_notify(struct svc_rqst *, __be32 *, struct nlm_args *); +int nlm4svc_decode_reboot(struct svc_rqst *, __be32 *, struct nlm_reboot *); /* int nlmclt_encode_testargs(struct rpc_rqst *, u32 *, struct nlm_args *); int nlmclt_encode_lockargs(struct rpc_rqst *, u32 *, struct nlm_args *); -- cgit v1.2.3 From 0dbb4c6799cf8fa8c5ba1926153a30960117477d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:49 -0700 Subject: [PATCH] xdr annotations: NFS readdir entries on-the-wire data is big-endian [in large part pulled from Alexey's patch] Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 6 +++--- fs/nfs/internal.h | 6 +++--- fs/nfs/nfs2xdr.c | 4 ++-- fs/nfs/nfs3xdr.c | 4 ++-- fs/nfs/nfs4_fs.h | 2 +- fs/nfs/nfs4proc.c | 4 ++-- fs/nfs/nfs4xdr.c | 2 +- include/linux/nfs_xdr.h | 2 +- 8 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index c86a1ead4772..4133ef5264e5 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -142,12 +142,12 @@ nfs_opendir(struct inode *inode, struct file *filp) return res; } -typedef u32 * (*decode_dirent_t)(u32 *, struct nfs_entry *, int); +typedef __be32 * (*decode_dirent_t)(__be32 *, struct nfs_entry *, int); typedef struct { struct file *file; struct page *page; unsigned long page_index; - u32 *ptr; + __be32 *ptr; u64 *dir_cookie; loff_t current_index; struct nfs_entry *entry; @@ -220,7 +220,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) static inline int dir_decode(nfs_readdir_descriptor_t *desc) { - u32 *p = desc->ptr; + __be32 *p = desc->ptr; p = desc->decode(p, desc->entry, desc->plus); if (IS_ERR(p)) return PTR_ERR(p); diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index bea0b016bd70..d205466233f6 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -93,15 +93,15 @@ extern void nfs_destroy_directcache(void); /* nfs2xdr.c */ extern int nfs_stat_to_errno(int); extern struct rpc_procinfo nfs_procedures[]; -extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs3xdr.c */ extern struct rpc_procinfo nfs3_procedures[]; -extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs4xdr.c */ #ifdef CONFIG_NFS_V4 -extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); #endif /* nfs4proc.c */ diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 1d801e30c40e..3be4e72a0227 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -468,8 +468,8 @@ err_unmap: goto out; } -u32 * -nfs_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { if (!*p++) { if (!*p) diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index b4e740e4494a..0ace092d126f 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -583,8 +583,8 @@ err_unmap: goto out; } -u32 * -nfs3_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs3_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { struct nfs_entry old = *entry; diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 61095fe4b5ca..6f346677332d 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -212,7 +212,7 @@ extern void nfs_free_seqid(struct nfs_seqid *seqid); extern const nfs4_stateid zero_stateid; /* nfs4xdr.c */ -extern uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; struct nfs4_mount_data; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 7421bcb3b728..8d09b47c91b9 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -141,7 +141,7 @@ const u32 nfs4_fs_locations_bitmap[2] = { static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { - u32 *start, *p; + __be32 *start, *p; BUG_ON(readdir->count < 80); if (cookie > 2) { @@ -162,7 +162,7 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, * when talking to the server, we always send cookie 0 * instead of 1 or 2. */ - start = p = (u32 *)kmap_atomic(*readdir->pages, KM_USER0); + start = p = kmap_atomic(*readdir->pages, KM_USER0); if (cookie == 0) { *p++ = xdr_one; /* next */ diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index e284123b9774..0cf3fa312a33 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4421,7 +4421,7 @@ out: return status; } -uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus) +__be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { uint32_t bitmap[2] = {0}; uint32_t len; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index dc5397d9d23c..ac8058bc8f94 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -811,7 +811,7 @@ struct nfs_rpc_ops { int (*pathconf) (struct nfs_server *, struct nfs_fh *, struct nfs_pathconf *); int (*set_capabilities)(struct nfs_server *, struct nfs_fh *); - u32 * (*decode_dirent)(u32 *, struct nfs_entry *, int plus); + __be32 *(*decode_dirent)(__be32 *, struct nfs_entry *, int plus); void (*read_setup) (struct nfs_read_data *); int (*read_done) (struct rpc_task *, struct nfs_read_data *); void (*write_setup) (struct nfs_write_data *, int how); -- cgit v1.2.3 From bc4785cd475a11ba125df7af674e16c6ea1cfc30 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:51 -0700 Subject: [PATCH] nfs: verifier is network-endian Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/nfs3proc.c | 2 +- fs/nfs/nfs4proc.c | 6 +++--- include/linux/nfs_fs.h | 2 +- include/linux/nfs_xdr.h | 8 ++++---- 4 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 3b234d4601e7..e5f128ffc32d 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -668,7 +668,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, { struct inode *dir = dentry->d_inode; struct nfs_fattr dir_attr; - u32 *verf = NFS_COOKIEVERF(dir); + __be32 *verf = NFS_COOKIEVERF(dir); struct nfs3_readdirargs arg = { .fh = NFS_FH(dir), .cookie = cookie, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8d09b47c91b9..8118036cc449 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -138,7 +138,7 @@ const u32 nfs4_fs_locations_bitmap[2] = { | FATTR4_WORD1_MOUNTED_ON_FILEID }; -static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, +static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { __be32 *start, *p; @@ -2915,11 +2915,11 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po .rpc_resp = clp, .rpc_cred = cred, }; - u32 *p; + __be32 *p; int loop = 0; int status; - p = (u32*)sc_verifier.data; + p = (__be32*)sc_verifier.data; *p++ = htonl((u32)clp->cl_boot_time.tv_sec); *p = htonl((u32)clp->cl_boot_time.tv_nsec); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 6b2de1be5815..45228c1a1195 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -157,7 +157,7 @@ struct nfs_inode { * This is the cookie verifier used for NFSv3 readdir * operations */ - __u32 cookieverf[2]; + __be32 cookieverf[2]; /* * This is the list of dirty unwritten pages. diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index ac8058bc8f94..768c1ad5ff6f 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -266,7 +266,7 @@ struct nfs_writeargs { struct nfs_writeverf { enum nfs3_stable_how committed; - __u32 verifier[2]; + __be32 verifier[2]; }; struct nfs_writeres { @@ -420,7 +420,7 @@ struct nfs3_createargs { unsigned int len; struct iattr * sattr; enum nfs3_createmode createmode; - __u32 verifier[2]; + __be32 verifier[2]; }; struct nfs3_mkdirargs { @@ -467,7 +467,7 @@ struct nfs3_linkargs { struct nfs3_readdirargs { struct nfs_fh * fh; __u64 cookie; - __u32 verf[2]; + __be32 verf[2]; int plus; unsigned int count; struct page ** pages; @@ -503,7 +503,7 @@ struct nfs3_linkres { struct nfs3_readdirres { struct nfs_fattr * dir_attr; - __u32 * verf; + __be32 * verf; int plus; }; -- cgit v1.2.3 From 63f103111fdfc3cba00e4c94921d32362f375d93 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:54 -0700 Subject: [PATCH] nfsd: nfserrno() endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsproc.c | 7 +++---- include/linux/nfsd/export.h | 2 +- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 09030afd7249..03ab6822291f 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -579,11 +579,11 @@ struct svc_version nfsd_version2 = { /* * Map errnos to NFS errnos. */ -int +__be32 nfserrno (int errno) { static struct { - int nfserr; + __be32 nfserr; int syserr; } nfs_errtbl[] = { { nfs_ok, 0 }, @@ -615,11 +615,10 @@ nfserrno (int errno) { nfserr_badname, -ESRCH }, { nfserr_io, -ETXTBSY }, { nfserr_notsupp, -EOPNOTSUPP }, - { -1, -EIO } }; int i; - for (i = 0; nfs_errtbl[i].nfserr != -1; i++) { + for (i = 0; i < ARRAY_SIZE(nfs_errtbl); i++) { if (nfs_errtbl[i].syserr == errno) return nfs_errtbl[i].nfserr; } diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 6e78ea969f49..27666f5b8b53 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -118,7 +118,7 @@ struct svc_export * exp_parent(struct auth_domain *clp, int exp_rootfh(struct auth_domain *, char *path, struct knfsd_fh *, int maxsize); int exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); -int nfserrno(int errno); +__be32 nfserrno(int errno); extern struct cache_detail svc_export_cache; -- cgit v1.2.3 From 83b11340d683a67a77e35a5ffb5ad4afbf0be4e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:55 -0700 Subject: [PATCH] nfsfh simple endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfsfh.c | 10 +++++----- include/linux/nfsd/nfsfh.h | 6 +++--- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 501d83884530..727ab3bd450d 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -76,7 +76,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) * comment in the NFSv3 spec says this is incorrect (implementation notes for * the write call). */ -static inline int +static inline __be32 nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) { /* Type can be negative when creating hardlinks - not to a dir */ @@ -110,13 +110,13 @@ nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) * This is only called at the start of an nfsproc call, so fhp points to * a svc_fh which is all 0 except for the over-the-wire file handle. */ -u32 +__be32 fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) { struct knfsd_fh *fh = &fhp->fh_handle; struct svc_export *exp = NULL; struct dentry *dentry; - u32 error = 0; + __be32 error = 0; dprintk("nfsd: fh_verify(%s)\n", SVCFH_fmt(fhp)); @@ -315,7 +315,7 @@ static inline void _fh_update_old(struct dentry *dentry, fh->ofh_dirino = 0; } -int +__be32 fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, struct svc_fh *ref_fh) { /* ref_fh is a reference file handle. @@ -451,7 +451,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, st * Update file handle information after changing a dentry. * This is only called by nfsd_create, nfsd_create_v3 and nfsd_proc_create */ -int +__be32 fh_update(struct svc_fh *fhp) { struct dentry *dentry; diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index 069257ea99a0..749bad1ca16f 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -209,9 +209,9 @@ extern char * SVCFH_fmt(struct svc_fh *fhp); /* * Function prototypes */ -u32 fh_verify(struct svc_rqst *, struct svc_fh *, int, int); -int fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *); -int fh_update(struct svc_fh *); +__be32 fh_verify(struct svc_rqst *, struct svc_fh *, int, int); +__be32 fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *); +__be32 fh_update(struct svc_fh *); void fh_put(struct svc_fh *); static __inline__ struct svc_fh * -- cgit v1.2.3 From 131a21c2177c267ab259fcd06947c6f593a7de8e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:56 -0700 Subject: [PATCH] xdr annotations: NFSv2 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs2acl.c | 18 ++++++------ fs/nfsd/nfsxdr.c | 72 +++++++++++++++++++++++------------------------ include/linux/nfsd/nfsd.h | 2 +- include/linux/nfsd/xdr.h | 50 ++++++++++++++++---------------- 4 files changed, 71 insertions(+), 71 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 8d48616882c1..fd5397d8c62a 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -158,7 +158,7 @@ static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessarg /* * XDR decode functions */ -static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -169,7 +169,7 @@ static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *argp) { struct kvec *head = rqstp->rq_arg.head; @@ -194,7 +194,7 @@ static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, return (n > 0); } -static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -202,7 +202,7 @@ static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } -static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -217,7 +217,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -259,7 +259,7 @@ static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -267,7 +267,7 @@ static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, } /* ACCESS */ -static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -278,7 +278,7 @@ static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); @@ -287,7 +287,7 @@ static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 1135c0d14557..56ebb1443e0e 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -37,8 +37,8 @@ static u32 nfs_ftypes[] = { /* * XDR functions for basic NFS types */ -static u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { fh_init(fhp, NFS_FHSIZE); memcpy(&fhp->fh_handle.fh_base, p, NFS_FHSIZE); @@ -50,13 +50,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { memcpy(p, &fhp->fh_handle.fh_base, NFS_FHSIZE); return p + (NFS_FHSIZE>> 2); @@ -66,8 +66,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -82,8 +82,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_pathname(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_pathname(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -98,8 +98,8 @@ decode_pathname(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr(__be32 *p, struct iattr *iap) { u32 tmp, tmp1; @@ -151,8 +151,8 @@ decode_sattr(u32 *p, struct iattr *iap) return p; } -static u32 * -encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static __be32 * +encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -195,7 +195,7 @@ encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct kstat stat; vfs_getattr(fhp->fh_export->ex_mnt, fhp->fh_dentry, &stat); @@ -206,13 +206,13 @@ u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfssvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfssvc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -220,7 +220,7 @@ nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) } int -nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -231,7 +231,7 @@ nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -242,7 +242,7 @@ nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readargs *args) { unsigned int len; @@ -273,7 +273,7 @@ nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_writeargs *args) { unsigned int len; @@ -303,7 +303,7 @@ nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -315,7 +315,7 @@ nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -328,7 +328,7 @@ nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinkargs *args) +nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -338,7 +338,7 @@ nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinka } int -nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -350,7 +350,7 @@ nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_symlinkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -363,7 +363,7 @@ nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -382,13 +382,13 @@ nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, * XDR encode functions */ int -nfssvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } int -nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -396,7 +396,7 @@ nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropres *resp) { p = encode_fh(p, &resp->fh); @@ -405,7 +405,7 @@ nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkres *resp) { *p++ = htonl(resp->len); @@ -421,7 +421,7 @@ nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readres *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -440,7 +440,7 @@ nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirres *resp) { xdr_ressize_check(rqstp, p); @@ -453,7 +453,7 @@ nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_statfsres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_statfsres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_statfsres *resp) { struct kstatfs *stat = &resp->stats; @@ -471,7 +471,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int d_type) { struct nfsd_readdirres *cd = container_of(ccd, struct nfsd_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; int buflen, slen; /* @@ -497,7 +497,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, *p++ = htonl((u32) ino); /* file id */ p = xdr_encode_array(p, name, namlen);/* name length & name */ cd->offset = p; /* remember pointer */ - *p++ = ~(u32) 0; /* offset of next entry */ + *p++ = htonl(~0U); /* offset of next entry */ cd->buflen = buflen; cd->buffer = p; @@ -509,7 +509,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, * XDR release functions */ int -nfssvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfssvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index d0d4aae7085f..2f75160a5824 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -50,7 +50,7 @@ * Callback function for readdir */ struct readdir_cd { - int err; /* 0, nfserr, or nfserr_eof */ + __be32 err; /* 0, nfserr, or nfserr_eof */ }; typedef int (*encode_dent_fn)(struct readdir_cd *, const char *, int, loff_t, ino_t, unsigned int); diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h index 0e53de87d886..877192d3ae79 100644 --- a/include/linux/nfsd/xdr.h +++ b/include/linux/nfsd/xdr.h @@ -81,7 +81,7 @@ struct nfsd_readdirargs { struct svc_fh fh; __u32 cookie; __u32 count; - u32 * buffer; + __be32 * buffer; }; struct nfsd_attrstat { @@ -108,9 +108,9 @@ struct nfsd_readdirres { int count; struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; + __be32 * offset; }; struct nfsd_statfsres { @@ -135,43 +135,43 @@ union nfsd_xdrstore { #define NFS2_SVC_XDRSIZE sizeof(union nfsd_xdrstore) -int nfssvc_decode_void(struct svc_rqst *, u32 *, void *); -int nfssvc_decode_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); -int nfssvc_decode_sattrargs(struct svc_rqst *, u32 *, +int nfssvc_decode_void(struct svc_rqst *, __be32 *, void *); +int nfssvc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); +int nfssvc_decode_sattrargs(struct svc_rqst *, __be32 *, struct nfsd_sattrargs *); -int nfssvc_decode_diropargs(struct svc_rqst *, u32 *, +int nfssvc_decode_diropargs(struct svc_rqst *, __be32 *, struct nfsd_diropargs *); -int nfssvc_decode_readargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readargs(struct svc_rqst *, __be32 *, struct nfsd_readargs *); -int nfssvc_decode_writeargs(struct svc_rqst *, u32 *, +int nfssvc_decode_writeargs(struct svc_rqst *, __be32 *, struct nfsd_writeargs *); -int nfssvc_decode_createargs(struct svc_rqst *, u32 *, +int nfssvc_decode_createargs(struct svc_rqst *, __be32 *, struct nfsd_createargs *); -int nfssvc_decode_renameargs(struct svc_rqst *, u32 *, +int nfssvc_decode_renameargs(struct svc_rqst *, __be32 *, struct nfsd_renameargs *); -int nfssvc_decode_readlinkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readlinkargs(struct svc_rqst *, __be32 *, struct nfsd_readlinkargs *); -int nfssvc_decode_linkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_linkargs(struct svc_rqst *, __be32 *, struct nfsd_linkargs *); -int nfssvc_decode_symlinkargs(struct svc_rqst *, u32 *, +int nfssvc_decode_symlinkargs(struct svc_rqst *, __be32 *, struct nfsd_symlinkargs *); -int nfssvc_decode_readdirargs(struct svc_rqst *, u32 *, +int nfssvc_decode_readdirargs(struct svc_rqst *, __be32 *, struct nfsd_readdirargs *); -int nfssvc_encode_void(struct svc_rqst *, u32 *, void *); -int nfssvc_encode_attrstat(struct svc_rqst *, u32 *, struct nfsd_attrstat *); -int nfssvc_encode_diropres(struct svc_rqst *, u32 *, struct nfsd_diropres *); -int nfssvc_encode_readlinkres(struct svc_rqst *, u32 *, struct nfsd_readlinkres *); -int nfssvc_encode_readres(struct svc_rqst *, u32 *, struct nfsd_readres *); -int nfssvc_encode_statfsres(struct svc_rqst *, u32 *, struct nfsd_statfsres *); -int nfssvc_encode_readdirres(struct svc_rqst *, u32 *, struct nfsd_readdirres *); +int nfssvc_encode_void(struct svc_rqst *, __be32 *, void *); +int nfssvc_encode_attrstat(struct svc_rqst *, __be32 *, struct nfsd_attrstat *); +int nfssvc_encode_diropres(struct svc_rqst *, __be32 *, struct nfsd_diropres *); +int nfssvc_encode_readlinkres(struct svc_rqst *, __be32 *, struct nfsd_readlinkres *); +int nfssvc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd_readres *); +int nfssvc_encode_statfsres(struct svc_rqst *, __be32 *, struct nfsd_statfsres *); +int nfssvc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd_readdirres *); int nfssvc_encode_entry(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); -int nfssvc_release_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); +int nfssvc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); /* Helper functions for NFSv2 ACL code */ -u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp); -u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp); +__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp); +__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp); #endif /* LINUX_NFSD_H */ -- cgit v1.2.3 From 91f07168cef8e99dd16f608fbc703e7a5af0237f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:57 -0700 Subject: [PATCH] xdr annotations: NFSv3 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs3acl.c | 10 ++-- fs/nfsd/nfs3xdr.c | 126 +++++++++++++++++++++++----------------------- include/linux/nfsd/xdr3.h | 114 ++++++++++++++++++++--------------------- 3 files changed, 125 insertions(+), 125 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index ed6e2c27b5e8..78b2c83d00c5 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -122,7 +122,7 @@ static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, /* * XDR decode functions */ -static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *args) { if (!(p = nfs3svc_decode_fh(p, &args->fh))) @@ -133,7 +133,7 @@ static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *args) { struct kvec *head = rqstp->rq_arg.head; @@ -163,7 +163,7 @@ static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -208,7 +208,7 @@ static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, } /* SETACL */ -static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh); @@ -219,7 +219,7 @@ static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfs3svc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 247d518248bf..b4baca3053c3 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -42,23 +42,23 @@ static u32 nfs3_ftypes[] = { /* * XDR functions for basic NFS types */ -static inline u32 * -encode_time3(u32 *p, struct timespec *time) +static inline __be32 * +encode_time3(__be32 *p, struct timespec *time) { *p++ = htonl((u32) time->tv_sec); *p++ = htonl(time->tv_nsec); return p; } -static inline u32 * -decode_time3(u32 *p, struct timespec *time) +static inline __be32 * +decode_time3(__be32 *p, struct timespec *time) { time->tv_sec = ntohl(*p++); time->tv_nsec = ntohl(*p++); return p; } -static inline u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size; fh_init(fhp, NFS3_FHSIZE); @@ -72,13 +72,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv3 ACL code */ -u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size = fhp->fh_handle.fh_size; *p++ = htonl(size); @@ -91,8 +91,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -107,8 +107,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr3(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr3(__be32 *p, struct iattr *iap) { u32 tmp; @@ -153,8 +153,8 @@ decode_sattr3(u32 *p, struct iattr *iap) return p; } -static inline u32 * -encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static inline __be32 * +encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -186,8 +186,8 @@ encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, return p; } -static inline u32 * -encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_saved_post_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct inode *inode = fhp->fh_dentry->d_inode; @@ -224,8 +224,8 @@ encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * The inode may be NULL if the call failed because of a stale file * handle. In this case, no attributes are returned. */ -static u32 * -encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; if (dentry && dentry->d_inode != NULL) { @@ -243,8 +243,8 @@ encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) } /* Helper for NFSv3 ACLs */ -u32 * -nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 * +nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { return encode_post_op_attr(rqstp, p, fhp); } @@ -252,8 +252,8 @@ nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) /* * Enocde weak cache consistency data */ -static u32 * -encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_wcc_data(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; @@ -278,7 +278,7 @@ encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfs3svc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -286,7 +286,7 @@ nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args } int -nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -303,7 +303,7 @@ nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -314,7 +314,7 @@ nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -325,7 +325,7 @@ nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readargs *args) { unsigned int len; @@ -355,7 +355,7 @@ nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeargs *args) { unsigned int len, v, hdr; @@ -393,7 +393,7 @@ nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -417,7 +417,7 @@ nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } int -nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -429,7 +429,7 @@ nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_symlinkargs *args) { unsigned int len; @@ -481,7 +481,7 @@ nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_mknodargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -505,7 +505,7 @@ nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -518,7 +518,7 @@ nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -530,7 +530,7 @@ nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -542,7 +542,7 @@ nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -562,7 +562,7 @@ nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { int len, pn; @@ -590,7 +590,7 @@ nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_commitargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -609,14 +609,14 @@ nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, * will work properly. */ int -nfs3svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs3svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } /* GETATTR */ int -nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { if (resp->status == 0) @@ -626,7 +626,7 @@ nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, /* SETATTR, REMOVE, RMDIR */ int -nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_wccstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -635,7 +635,7 @@ nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, /* LOOKUP */ int -nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -648,7 +648,7 @@ nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, /* ACCESS */ int -nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -659,7 +659,7 @@ nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* READLINK */ int -nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -680,7 +680,7 @@ nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, /* READ */ int -nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -704,7 +704,7 @@ nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, /* WRITE */ int -nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_writeres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -719,7 +719,7 @@ nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, /* CREATE, MKDIR, SYMLINK, MKNOD */ int -nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_createres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -733,7 +733,7 @@ nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, /* RENAME */ int -nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_renameres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameres *resp) { p = encode_wcc_data(rqstp, p, &resp->ffh); @@ -743,7 +743,7 @@ nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, /* LINK */ int -nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_linkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -753,7 +753,7 @@ nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, /* READDIR */ int -nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -776,8 +776,8 @@ nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, return xdr_ressize_check(rqstp, p); } -static inline u32 * -encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, +static inline __be32 * +encode_entry_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen, ino_t ino) { *p++ = xdr_one; /* mark entry present */ @@ -790,8 +790,8 @@ encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, return p; } -static inline u32 * -encode_entryplus_baggage(struct nfsd3_readdirres *cd, u32 *p, +static inline __be32 * +encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, struct svc_fh *fhp) { p = encode_post_op_attr(cd->rqstp, p, fhp); @@ -853,7 +853,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, { struct nfsd3_readdirres *cd = container_of(ccd, struct nfsd3_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; caddr_t curr_page_addr = NULL; int pn; /* current page number */ int slen; /* string (name) length */ @@ -919,7 +919,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, } else if (cd->rqstp->rq_respages[pn+1] != NULL) { /* temporarily encode entry into next page, then move back to * current and next page in rq_respages[] */ - u32 *p1, *tmp; + __be32 *p1, *tmp; int len1, len2; /* grab next page for temporary storage of entry */ @@ -1009,7 +1009,7 @@ nfs3svc_encode_entry_plus(struct readdir_cd *cd, const char *name, /* FSSTAT */ int -nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsstatres *resp) { struct kstatfs *s = &resp->stats; @@ -1031,7 +1031,7 @@ nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, /* FSINFO */ int -nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsinfores *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1055,7 +1055,7 @@ nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, /* PATHCONF */ int -nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_pathconfres *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1074,7 +1074,7 @@ nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, /* COMMIT */ int -nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_commitres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -1090,7 +1090,7 @@ nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, * XDR release functions */ int -nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { fh_put(&resp->fh); @@ -1098,7 +1098,7 @@ nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_release_fhandle2(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle2(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fhandle_pair *resp) { fh_put(&resp->fh1); diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h index 474d882dc2f3..79963867b0d7 100644 --- a/include/linux/nfsd/xdr3.h +++ b/include/linux/nfsd/xdr3.h @@ -51,7 +51,7 @@ struct nfsd3_createargs { int len; int createmode; struct iattr attrs; - __u32 * verf; + __be32 * verf; }; struct nfsd3_mknodargs { @@ -98,8 +98,8 @@ struct nfsd3_readdirargs { __u64 cookie; __u32 dircount; __u32 count; - __u32 * verf; - u32 * buffer; + __be32 * verf; + __be32 * buffer; }; struct nfsd3_commitargs { @@ -122,79 +122,79 @@ struct nfsd3_setaclargs { }; struct nfsd3_attrstat { - __u32 status; + __be32 status; struct svc_fh fh; struct kstat stat; }; /* LOOKUP, CREATE, MKDIR, SYMLINK, MKNOD */ struct nfsd3_diropres { - __u32 status; + __be32 status; struct svc_fh dirfh; struct svc_fh fh; }; struct nfsd3_accessres { - __u32 status; + __be32 status; struct svc_fh fh; __u32 access; }; struct nfsd3_readlinkres { - __u32 status; + __be32 status; struct svc_fh fh; __u32 len; }; struct nfsd3_readres { - __u32 status; + __be32 status; struct svc_fh fh; unsigned long count; int eof; }; struct nfsd3_writeres { - __u32 status; + __be32 status; struct svc_fh fh; unsigned long count; int committed; }; struct nfsd3_renameres { - __u32 status; + __be32 status; struct svc_fh ffh; struct svc_fh tfh; }; struct nfsd3_linkres { - __u32 status; + __be32 status; struct svc_fh tfh; struct svc_fh fh; }; struct nfsd3_readdirres { - __u32 status; + __be32 status; struct svc_fh fh; int count; - __u32 verf[2]; + __be32 verf[2]; struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; - u32 * offset1; + __be32 * offset; + __be32 * offset1; struct svc_rqst * rqstp; }; struct nfsd3_fsstatres { - __u32 status; + __be32 status; struct kstatfs stats; __u32 invarsec; }; struct nfsd3_fsinfores { - __u32 status; + __be32 status; __u32 f_rtmax; __u32 f_rtpref; __u32 f_rtmult; @@ -207,7 +207,7 @@ struct nfsd3_fsinfores { }; struct nfsd3_pathconfres { - __u32 status; + __be32 status; __u32 p_link_max; __u32 p_name_max; __u32 p_no_trunc; @@ -217,12 +217,12 @@ struct nfsd3_pathconfres { }; struct nfsd3_commitres { - __u32 status; + __be32 status; struct svc_fh fh; }; struct nfsd3_getaclres { - __u32 status; + __be32 status; struct svc_fh fh; int mask; struct posix_acl *acl_access; @@ -266,70 +266,70 @@ union nfsd3_xdrstore { #define NFS3_SVC_XDRSIZE sizeof(union nfsd3_xdrstore) -int nfs3svc_decode_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); -int nfs3svc_decode_sattrargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); +int nfs3svc_decode_sattrargs(struct svc_rqst *, __be32 *, struct nfsd3_sattrargs *); -int nfs3svc_decode_diropargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_diropargs(struct svc_rqst *, __be32 *, struct nfsd3_diropargs *); -int nfs3svc_decode_accessargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_accessargs(struct svc_rqst *, __be32 *, struct nfsd3_accessargs *); -int nfs3svc_decode_readargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readargs(struct svc_rqst *, __be32 *, struct nfsd3_readargs *); -int nfs3svc_decode_writeargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_writeargs(struct svc_rqst *, __be32 *, struct nfsd3_writeargs *); -int nfs3svc_decode_createargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_createargs(struct svc_rqst *, __be32 *, struct nfsd3_createargs *); -int nfs3svc_decode_mkdirargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_mkdirargs(struct svc_rqst *, __be32 *, struct nfsd3_createargs *); -int nfs3svc_decode_mknodargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_mknodargs(struct svc_rqst *, __be32 *, struct nfsd3_mknodargs *); -int nfs3svc_decode_renameargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_renameargs(struct svc_rqst *, __be32 *, struct nfsd3_renameargs *); -int nfs3svc_decode_readlinkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readlinkargs(struct svc_rqst *, __be32 *, struct nfsd3_readlinkargs *); -int nfs3svc_decode_linkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_linkargs(struct svc_rqst *, __be32 *, struct nfsd3_linkargs *); -int nfs3svc_decode_symlinkargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_symlinkargs(struct svc_rqst *, __be32 *, struct nfsd3_symlinkargs *); -int nfs3svc_decode_readdirargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readdirargs(struct svc_rqst *, __be32 *, struct nfsd3_readdirargs *); -int nfs3svc_decode_readdirplusargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_readdirplusargs(struct svc_rqst *, __be32 *, struct nfsd3_readdirargs *); -int nfs3svc_decode_commitargs(struct svc_rqst *, u32 *, +int nfs3svc_decode_commitargs(struct svc_rqst *, __be32 *, struct nfsd3_commitargs *); -int nfs3svc_encode_voidres(struct svc_rqst *, u32 *, void *); -int nfs3svc_encode_attrstat(struct svc_rqst *, u32 *, +int nfs3svc_encode_voidres(struct svc_rqst *, __be32 *, void *); +int nfs3svc_encode_attrstat(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_encode_wccstat(struct svc_rqst *, u32 *, +int nfs3svc_encode_wccstat(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_encode_diropres(struct svc_rqst *, u32 *, +int nfs3svc_encode_diropres(struct svc_rqst *, __be32 *, struct nfsd3_diropres *); -int nfs3svc_encode_accessres(struct svc_rqst *, u32 *, +int nfs3svc_encode_accessres(struct svc_rqst *, __be32 *, struct nfsd3_accessres *); -int nfs3svc_encode_readlinkres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readlinkres(struct svc_rqst *, __be32 *, struct nfsd3_readlinkres *); -int nfs3svc_encode_readres(struct svc_rqst *, u32 *, struct nfsd3_readres *); -int nfs3svc_encode_writeres(struct svc_rqst *, u32 *, struct nfsd3_writeres *); -int nfs3svc_encode_createres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd3_readres *); +int nfs3svc_encode_writeres(struct svc_rqst *, __be32 *, struct nfsd3_writeres *); +int nfs3svc_encode_createres(struct svc_rqst *, __be32 *, struct nfsd3_diropres *); -int nfs3svc_encode_renameres(struct svc_rqst *, u32 *, +int nfs3svc_encode_renameres(struct svc_rqst *, __be32 *, struct nfsd3_renameres *); -int nfs3svc_encode_linkres(struct svc_rqst *, u32 *, +int nfs3svc_encode_linkres(struct svc_rqst *, __be32 *, struct nfsd3_linkres *); -int nfs3svc_encode_readdirres(struct svc_rqst *, u32 *, +int nfs3svc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd3_readdirres *); -int nfs3svc_encode_fsstatres(struct svc_rqst *, u32 *, +int nfs3svc_encode_fsstatres(struct svc_rqst *, __be32 *, struct nfsd3_fsstatres *); -int nfs3svc_encode_fsinfores(struct svc_rqst *, u32 *, +int nfs3svc_encode_fsinfores(struct svc_rqst *, __be32 *, struct nfsd3_fsinfores *); -int nfs3svc_encode_pathconfres(struct svc_rqst *, u32 *, +int nfs3svc_encode_pathconfres(struct svc_rqst *, __be32 *, struct nfsd3_pathconfres *); -int nfs3svc_encode_commitres(struct svc_rqst *, u32 *, +int nfs3svc_encode_commitres(struct svc_rqst *, __be32 *, struct nfsd3_commitres *); -int nfs3svc_release_fhandle(struct svc_rqst *, u32 *, +int nfs3svc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); -int nfs3svc_release_fhandle2(struct svc_rqst *, u32 *, +int nfs3svc_release_fhandle2(struct svc_rqst *, __be32 *, struct nfsd3_fhandle_pair *); int nfs3svc_encode_entry(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, @@ -338,9 +338,9 @@ int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); /* Helper functions for NFSv3 ACL code */ -u32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, +__be32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp); -u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp); +__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp); #endif /* _LINUX_NFSD_XDR3_H */ -- cgit v1.2.3 From 2ebbc012a9433a252be7ab4ce54e94bf7b21e506 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:58 -0700 Subject: [PATCH] xdr annotations: NFSv4 server Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 2 +- fs/nfsd/nfs4xdr.c | 66 +++++++++++++++++++++++------------------------ include/linux/nfsd/xdr4.h | 24 ++++++++--------- 3 files changed, 46 insertions(+), 46 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 795ad6c5cb2c..ca6414248527 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -664,7 +664,7 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ static int nfsd4_verify(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_verify *verify) { - u32 *buf, *p; + __be32 *buf, *p; int count; int status; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 77be0c4785e6..3419d99aeb1a 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -94,7 +94,7 @@ check_filename(char *str, int len, int err) * consistent with the style used in NFSv2/v3... */ #define DECODE_HEAD \ - u32 *p; \ + __be32 *p; \ int status #define DECODE_TAIL \ status = 0; \ @@ -144,13 +144,13 @@ xdr_error: \ } \ } while (0) -static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) +static __be32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) { /* We want more bytes than seem to be available. * Maybe we need a new page, maybe we have just run out */ int avail = (char*)argp->end - (char*)argp->p; - u32 *p; + __be32 *p; if (avail + argp->pagelen < nbytes) return NULL; if (avail + PAGE_SIZE < nbytes) /* need more than a page !! */ @@ -197,7 +197,7 @@ defer_free(struct nfsd4_compoundargs *argp, return 0; } -static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) +static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) { void *new = NULL; if (p == argp->tmp) { @@ -951,8 +951,8 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) argp->pagelen -= len; } } - argp->end = (u32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); - argp->p = (u32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); + argp->end = (__be32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); + argp->p = (__be32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); argp->rqstp->rq_vec[v].iov_len = len; write->wr_vlen = v+1; @@ -1179,7 +1179,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * task to translate them into Linux-specific versions which are more * consistent with the style used in NFSv2/v3... */ -#define ENCODE_HEAD u32 *p +#define ENCODE_HEAD __be32 *p #define WRITE32(n) *p++ = htonl(n) #define WRITE64(n) do { \ @@ -1209,8 +1209,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * Header routine to setup seqid operation replay cache */ #define ENCODE_SEQID_OP_HEAD \ - u32 *p; \ - u32 *save; \ + __be32 *p; \ + __be32 *save; \ \ save = resp->p; @@ -1235,10 +1235,10 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * seperated @sep. */ static int nfsd4_encode_components(char sep, char *components, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { - u32 *p = *pp; - u32 *countp = p; + __be32 *p = *pp; + __be32 *countp = p; int strlen, count=0; char *str, *end; @@ -1272,10 +1272,10 @@ static int nfsd4_encode_components(char sep, char *components, * encode a location element of a fs_locations structure */ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { int status; - u32 *p = *pp; + __be32 *p = *pp; status = nfsd4_encode_components(':', location->hosts, &p, buflen); if (status) @@ -1320,11 +1320,11 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *sta */ static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { u32 status; int i; - u32 *p = *pp; + __be32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; char *root = nfsd4_path(rqstp, exp, &status); @@ -1355,7 +1355,7 @@ static u32 nfs4_ftypes[16] = { static int nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { int status; @@ -1376,20 +1376,20 @@ nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, } static inline int -nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, u32 **p, int *buflen) +nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen); } static inline int -nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, u32 **p, int *buflen) +nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen); } static inline int nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, whotype, id, group, p, buflen); } @@ -1423,7 +1423,7 @@ static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) */ int nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, - struct dentry *dentry, u32 *buffer, int *countp, u32 *bmval, + struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *rqstp) { u32 bmval0 = bmval[0]; @@ -1432,11 +1432,11 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct svc_fh tempfh; struct kstatfs statfs; int buflen = *countp << 2; - u32 *attrlenp; + __be32 *attrlenp; u32 dummy; u64 dummy64; u32 rdattr_err = 0; - u32 *p = buffer; + __be32 *p = buffer; int status; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1831,7 +1831,7 @@ out_serverfault: static int nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, - const char *name, int namlen, u32 *p, int *buflen) + const char *name, int namlen, __be32 *p, int *buflen) { struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; @@ -1864,10 +1864,10 @@ out_put: return nfserr; } -static u32 * -nfsd4_encode_rdattr_error(u32 *p, int buflen, int nfserr) +static __be32 * +nfsd4_encode_rdattr_error(__be32 *p, int buflen, int nfserr) { - u32 *attrlenp; + __be32 *attrlenp; if (buflen < 6) return NULL; @@ -1887,7 +1887,7 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, { struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; - u32 *p = cd->buffer; + __be32 *p = cd->buffer; int nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ @@ -2321,7 +2321,7 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_re { int maxcount; loff_t offset; - u32 *page, *savep, *tailbase; + __be32 *page, *savep, *tailbase; ENCODE_HEAD; if (nfserr) @@ -2479,7 +2479,7 @@ nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_writ void nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) { - u32 *statp; + __be32 *statp; ENCODE_HEAD; RESERVE_SPACE(8); @@ -2617,7 +2617,7 @@ nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op) */ int -nfs4svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs4svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -2639,7 +2639,7 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) } int -nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundargs *args) +nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundargs *args) { int status; @@ -2660,7 +2660,7 @@ nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoun } int -nfs4svc_encode_compoundres(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundres *resp) +nfs4svc_encode_compoundres(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundres *resp) { /* * All that remains is to write the tag and operation count... diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index 66e642762a07..003193fe6fc6 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -258,9 +258,9 @@ struct nfsd4_readdir { struct svc_fh * rd_fhp; /* response */ struct readdir_cd common; - u32 * buffer; + __be32 * buffer; int buflen; - u32 * offset; + __be32 * offset; }; struct nfsd4_release_lockowner { @@ -371,12 +371,12 @@ struct nfsd4_op { struct nfsd4_compoundargs { /* scratch variables for XDR decode */ - u32 * p; - u32 * end; + __be32 * p; + __be32 * end; struct page ** pagelist; int pagelen; - u32 tmp[8]; - u32 * tmpp; + __be32 tmp[8]; + __be32 * tmpp; struct tmpbuf { struct tmpbuf *next; void (*release)(const void *); @@ -395,15 +395,15 @@ struct nfsd4_compoundargs { struct nfsd4_compoundres { /* scratch variables for XDR encode */ - u32 * p; - u32 * end; + __be32 * p; + __be32 * end; struct xdr_buf * xbuf; struct svc_rqst * rqstp; u32 taglen; char * tag; u32 opcnt; - u32 * tagp; /* where to encode tag and opcount */ + __be32 * tagp; /* where to encode tag and opcount */ }; #define NFS4_SVC_XDRSIZE sizeof(struct nfsd4_compoundargs) @@ -419,10 +419,10 @@ set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp) cinfo->after_ctime_nsec = fhp->fh_post_ctime.tv_nsec; } -int nfs4svc_encode_voidres(struct svc_rqst *, u32 *, void *); -int nfs4svc_decode_compoundargs(struct svc_rqst *, u32 *, +int nfs4svc_encode_voidres(struct svc_rqst *, __be32 *, void *); +int nfs4svc_decode_compoundargs(struct svc_rqst *, __be32 *, struct nfsd4_compoundargs *); -int nfs4svc_encode_compoundres(struct svc_rqst *, u32 *, +int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *, struct nfsd4_compoundres *); void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *); void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op); -- cgit v1.2.3 From 6264d69d7df654ca64f625e9409189a0e50734e9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:58 -0700 Subject: [PATCH] nfsd: vfs.c endianness annotations don't use the same variable to store NFS and host error values Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 299 ++++++++++++++++++++++++---------------------- include/linux/nfsd/nfsd.h | 38 +++--- 2 files changed, 176 insertions(+), 161 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 1141bd29e4e3..f21e917bb8ed 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -110,7 +110,7 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct dentry *dentry = *dpp; struct vfsmount *mnt = mntget(exp->ex_mnt); struct dentry *mounts = dget(dentry); - int err = nfs_ok; + int err = 0; while (follow_down(&mnt,&mounts)&&d_mountpoint(mounts)); @@ -148,14 +148,15 @@ out: * clients and is explicitly disallowed for NFSv3 * NeilBrown */ -int +__be32 nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, int len, struct svc_fh *resfh) { struct svc_export *exp; struct dentry *dparent; struct dentry *dentry; - int err; + __be32 err; + int host_err; dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name); @@ -193,7 +194,7 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, exp2 = exp_parent(exp->ex_client, mnt, dentry, &rqstp->rq_chandle); if (IS_ERR(exp2)) { - err = PTR_ERR(exp2); + host_err = PTR_ERR(exp2); dput(dentry); mntput(mnt); goto out_nfserr; @@ -210,14 +211,14 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, } else { fh_lock(fhp); dentry = lookup_one_len(name, dparent, len); - err = PTR_ERR(dentry); + host_err = PTR_ERR(dentry); if (IS_ERR(dentry)) goto out_nfserr; /* * check if we have crossed a mount point ... */ if (d_mountpoint(dentry)) { - if ((err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { + if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { dput(dentry); goto out_nfserr; } @@ -236,7 +237,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -244,7 +245,7 @@ out_nfserr: * Set various file attributes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int check_guard, time_t guardtime) { @@ -253,7 +254,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int accmode = MAY_SATTR; int ftype = 0; int imode; - int err; + __be32 err; + int host_err; int size_change = 0; if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE)) @@ -319,19 +321,19 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, * If we are changing the size of the file, then * we need to break all leases. */ - err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* ENOMEM or EWOULDBLOCK */ + host_err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* ENOMEM or EWOULDBLOCK */ goto out_nfserr; - err = get_write_access(inode); - if (err) + host_err = get_write_access(inode); + if (host_err) goto out_nfserr; size_change = 1; - err = locks_verify_truncate(inode, NULL, iap->ia_size); - if (err) { + host_err = locks_verify_truncate(inode, NULL, iap->ia_size); + if (host_err) { put_write_access(inode); goto out_nfserr; } @@ -357,8 +359,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, err = nfserr_notsync; if (!check_guard || guardtime == inode->i_ctime.tv_sec) { fh_lock(fhp); - err = notify_change(dentry, iap); - err = nfserrno(err); + host_err = notify_change(dentry, iap); + err = nfserrno(host_err); fh_unlock(fhp); } if (size_change) @@ -370,7 +372,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -420,11 +422,12 @@ out: return error; } -int +__be32 nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfs4_acl *acl) { - int error; + __be32 error; + int host_error; struct dentry *dentry; struct inode *inode; struct posix_acl *pacl = NULL, *dpacl = NULL; @@ -440,20 +443,20 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, if (S_ISDIR(inode->i_mode)) flags = NFS4_ACL_DIR; - error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); - if (error == -EINVAL) { + host_error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); + if (host_error == -EINVAL) { error = nfserr_attrnotsupp; goto out; - } else if (error < 0) + } else if (host_error < 0) goto out_nfserr; - error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); + if (host_error < 0) goto out_nfserr; if (S_ISDIR(inode->i_mode)) { - error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); + if (host_error < 0) goto out_nfserr; } @@ -464,7 +467,7 @@ out: posix_acl_release(dpacl); return (error); out_nfserr: - error = nfserrno(error); + error = nfserrno(host_error); goto out; } @@ -571,14 +574,14 @@ static struct accessmap nfs3_anyaccess[] = { { 0, 0 } }; -int +__be32 nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *supported) { struct accessmap *map; struct svc_export *export; struct dentry *dentry; u32 query, result = 0, sresult = 0; - unsigned int error; + __be32 error; error = fh_verify(rqstp, fhp, 0, MAY_NOP); if (error) @@ -598,7 +601,7 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor query = *access; for (; map->access; map++) { if (map->access & query) { - unsigned int err2; + __be32 err2; sresult |= map->access; @@ -637,13 +640,15 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor * The access argument indicates the type of open (read/write/lock) * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access, struct file **filp) { struct dentry *dentry; struct inode *inode; - int flags = O_RDONLY|O_LARGEFILE, err; + int flags = O_RDONLY|O_LARGEFILE; + __be32 err; + int host_err; /* * If we get here, then the client has already done an "open", @@ -673,10 +678,10 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * Check to see if there are any leases on this file. * This may block while leases are broken. */ - err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* NOMEM or WOULDBLOCK */ + host_err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* NOMEM or WOULDBLOCK */ goto out_nfserr; if (access & MAY_WRITE) { @@ -689,10 +694,9 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, } *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_mnt), flags); if (IS_ERR(*filp)) - err = PTR_ERR(*filp); + host_err = PTR_ERR(*filp); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -830,14 +834,15 @@ nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset return size; } -static int +static __be32 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { struct inode *inode; struct raparms *ra; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = nfserr_perm; inode = file->f_dentry->d_inode; @@ -855,12 +860,12 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (file->f_op->sendfile && rqstp->rq_sendfile_ok) { rqstp->rq_resused = 1; - err = file->f_op->sendfile(file, &offset, *count, + host_err = file->f_op->sendfile(file, &offset, *count, nfsd_read_actor, rqstp); } else { oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); } @@ -874,13 +879,13 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, spin_unlock(&rab->pb_lock); } - if (err >= 0) { - nfsdstats.io_read += err; - *count = err; + if (host_err >= 0) { + nfsdstats.io_read += host_err; + *count = host_err; err = 0; fsnotify_access(file->f_dentry); } else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -895,7 +900,7 @@ static void kill_suid(struct dentry *dentry) mutex_unlock(&dentry->d_inode->i_mutex); } -static int +static __be32 nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) @@ -904,7 +909,8 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err = 0; + __be32 err = 0; + int host_err; int stable = *stablep; #ifdef MSNFS @@ -940,18 +946,18 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, /* Write the data. */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); - if (err >= 0) { + if (host_err >= 0) { nfsdstats.io_write += cnt; fsnotify_modify(file->f_dentry); } /* clear setuid/setgid flag after write */ - if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) + if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) kill_suid(dentry); - if (err >= 0 && stable) { + if (host_err >= 0 && stable) { static ino_t last_ino; static dev_t last_dev; @@ -977,7 +983,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (inode->i_state & I_DIRTY) { dprintk("nfsd: write sync %d\n", current->pid); - err=nfsd_sync(file); + host_err=nfsd_sync(file); } #if 0 wake_up(&inode->i_wait); @@ -987,11 +993,11 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, last_dev = inode->i_sb->s_dev; } - dprintk("nfsd: write complete err=%d\n", err); - if (err >= 0) + dprintk("nfsd: write complete host_err=%d\n", host_err); + if (host_err >= 0) err = 0; else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1001,12 +1007,12 @@ out: * on entry. On return, *count contains the number of bytes actually read. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { - int err; + __be32 err; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1030,12 +1036,12 @@ out: * The stable flag requests synchronous writes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) { - int err = 0; + __be32 err = 0; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1067,12 +1073,12 @@ out: * Unfortunately we cannot lock the file to make sure we return full WCC * data to the client, as locking happens lower down in the filesystem. */ -int +__be32 nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, unsigned long count) { struct file *file; - int err; + __be32 err; if ((u64)count > ~(u64)offset) return nfserr_inval; @@ -1100,14 +1106,15 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, * * N.B. Every call to nfsd_create needs an fh_put for _both_ fhp and resfhp */ -int +__be32 nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, int type, dev_t rdev, struct svc_fh *resfhp) { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_perm; if (!flen) @@ -1134,7 +1141,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, /* called from nfsd_proc_mkdir, or possibly nfsd3_proc_create */ fh_lock_nested(fhp, I_MUTEX_PARENT); dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; err = fh_compose(resfhp, fhp->fh_export, dchild, fhp); @@ -1173,22 +1180,22 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, err = nfserr_perm; switch (type) { case S_IFREG: - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); break; case S_IFDIR: - err = vfs_mkdir(dirp, dchild, iap->ia_mode); + host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); break; case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK: - err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); + host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); break; default: printk("nfsd: bad file type %o in nfsd_create\n", type); - err = -EINVAL; + host_err = -EINVAL; } - if (err < 0) + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1203,7 +1210,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, * directories via NFS. */ if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1218,7 +1225,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1226,7 +1233,7 @@ out_nfserr: /* * NFSv3 version of nfsd_create */ -int +__be32 nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, @@ -1234,7 +1241,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; __u32 v_mtime=0, v_atime=0; int v_mode=0; @@ -1264,7 +1272,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * Compose the response file handle. */ dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; @@ -1320,8 +1328,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; } - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); - if (err < 0) + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1350,7 +1358,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, */ set_attr: if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1368,7 +1376,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } #endif /* CONFIG_NFSD_V3 */ @@ -1378,13 +1386,14 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * fits into the buffer. On return, it contains the true length. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) { struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, fhp, S_IFLNK, MAY_NOP); if (err) @@ -1403,18 +1412,18 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = inode->i_op->readlink(dentry, buf, *lenp); + host_err = inode->i_op->readlink(dentry, buf, *lenp); set_fs(oldfs); - if (err < 0) + if (host_err < 0) goto out_nfserr; - *lenp = err; + *lenp = host_err; err = 0; out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1422,7 +1431,7 @@ out_nfserr: * Create a symlink and look up its inode * N.B. After this call _both_ fhp and resfhp need an fh_put */ -int +__be32 nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, char *path, int plen, @@ -1430,7 +1439,8 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap) { struct dentry *dentry, *dnew; - int err, cerr; + __be32 err, cerr; + int host_err; umode_t mode; err = nfserr_noent; @@ -1446,7 +1456,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, fh_lock(fhp); dentry = fhp->fh_dentry; dnew = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; @@ -1458,21 +1468,21 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (unlikely(path[plen] != 0)) { char *path_alloced = kmalloc(plen+1, GFP_KERNEL); if (path_alloced == NULL) - err = -ENOMEM; + host_err = -ENOMEM; else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); kfree(path_alloced); } } else - err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); - if (!err) + if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); - if (err) - err = nfserrno(err); + host_err = nfsd_sync_dir(dentry); + } + err = nfserrno(host_err); fh_unlock(fhp); cerr = fh_compose(resfhp, fhp->fh_export, dnew, fhp); @@ -1482,7 +1492,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1490,13 +1500,14 @@ out_nfserr: * Create a hardlink * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *name, int len, struct svc_fh *tfhp) { struct dentry *ddir, *dnew, *dold; struct inode *dirp, *dest; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE); if (err) @@ -1517,24 +1528,25 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, dirp = ddir->d_inode; dnew = lookup_one_len(name, ddir, len); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; dold = tfhp->fh_dentry; dest = dold->d_inode; - err = vfs_link(dold, dirp, dnew); - if (!err) { + host_err = vfs_link(dold, dirp, dnew); + if (!host_err) { if (EX_ISSYNC(ffhp->fh_export)) { err = nfserrno(nfsd_sync_dir(ddir)); write_inode_now(dest, 1); } + err = 0; } else { - if (err == -EXDEV && rqstp->rq_vers == 2) + if (host_err == -EXDEV && rqstp->rq_vers == 2) err = nfserr_acces; else - err = nfserrno(err); + err = nfserrno(host_err); } dput(dnew); @@ -1544,7 +1556,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out_unlock; } @@ -1552,13 +1564,14 @@ out_nfserr: * Rename a file * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, struct svc_fh *tfhp, char *tname, int tlen) { struct dentry *fdentry, *tdentry, *odentry, *ndentry, *trap; struct inode *fdir, *tdir; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE); if (err) @@ -1589,22 +1602,22 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, fill_pre_wcc(tfhp); odentry = lookup_one_len(fname, fdentry, flen); - err = PTR_ERR(odentry); + host_err = PTR_ERR(odentry); if (IS_ERR(odentry)) goto out_nfserr; - err = -ENOENT; + host_err = -ENOENT; if (!odentry->d_inode) goto out_dput_old; - err = -EINVAL; + host_err = -EINVAL; if (odentry == trap) goto out_dput_old; ndentry = lookup_one_len(tname, tdentry, tlen); - err = PTR_ERR(ndentry); + host_err = PTR_ERR(ndentry); if (IS_ERR(ndentry)) goto out_dput_old; - err = -ENOTEMPTY; + host_err = -ENOTEMPTY; if (ndentry == trap) goto out_dput_new; @@ -1612,14 +1625,14 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if ((ffhp->fh_export->ex_flags & NFSEXP_MSNFS) && ((atomic_read(&odentry->d_count) > 1) || (atomic_read(&ndentry->d_count) > 1))) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_rename(fdir, odentry, tdir, ndentry); - if (!err && EX_ISSYNC(tfhp->fh_export)) { - err = nfsd_sync_dir(tdentry); - if (!err) - err = nfsd_sync_dir(fdentry); + host_err = vfs_rename(fdir, odentry, tdir, ndentry); + if (!host_err && EX_ISSYNC(tfhp->fh_export)) { + host_err = nfsd_sync_dir(tdentry); + if (!host_err) + host_err = nfsd_sync_dir(fdentry); } out_dput_new: @@ -1627,8 +1640,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, out_dput_old: dput(odentry); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); /* we cannot reply on fh_unlock on the two filehandles, * as that would do the wrong thing if the two directories @@ -1647,13 +1659,14 @@ out: * Unlink a file or directory * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, char *fname, int flen) { struct dentry *dentry, *rdentry; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_acces; if (!flen || isdotent(fname, flen)) @@ -1667,7 +1680,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, dirp = dentry->d_inode; rdentry = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(rdentry); + host_err = PTR_ERR(rdentry); if (IS_ERR(rdentry)) goto out_nfserr; @@ -1684,22 +1697,23 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, #ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && (atomic_read(&rdentry->d_count) > 1)) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_unlink(dirp, rdentry); + host_err = vfs_unlink(dirp, rdentry); } else { /* It's RMDIR */ - err = vfs_rmdir(dirp, rdentry); + host_err = vfs_rmdir(dirp, rdentry); } dput(rdentry); - if (err == 0 && - EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); + if (host_err) + goto out_nfserr; + if (EX_ISSYNC(fhp->fh_export)) + host_err = nfsd_sync_dir(dentry); out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1708,11 +1722,12 @@ out: * Read entries from a directory. * The NFSv3/4 verifier we ignore for now. */ -int +__be32 nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, struct readdir_cd *cdp, encode_dent_fn func) { - int err; + __be32 err; + int host_err; struct file *file; loff_t offset = *offsetp; @@ -1734,10 +1749,10 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, do { cdp->err = nfserr_eof; /* will be cleared on successful read */ - err = vfs_readdir(file, (filldir_t) func, cdp); - } while (err >=0 && cdp->err == nfs_ok); - if (err) - err = nfserrno(err); + host_err = vfs_readdir(file, (filldir_t) func, cdp); + } while (host_err >=0 && cdp->err == nfs_ok); + if (host_err) + err = nfserrno(host_err); else err = cdp->err; *offsetp = vfs_llseek(file, 0, 1); @@ -1754,10 +1769,10 @@ out: * Get file system stats * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) { - int err = fh_verify(rqstp, fhp, 0, MAY_NOP); + __be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP); if (!err && vfs_statfs(fhp->fh_dentry,stat)) err = nfserr_io; return err; @@ -1766,7 +1781,7 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) /* * Check for a user's access permissions to this inode. */ -int +__be32 nfsd_permission(struct svc_export *exp, struct dentry *dentry, int acc) { struct inode *inode = dentry->d_inode; diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 2f75160a5824..19a3c83d496e 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -72,57 +72,57 @@ int nfsd_racache_init(int); void nfsd_racache_shutdown(void); int nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct svc_export **expp); -int nfsd_lookup(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_lookup(struct svc_rqst *, struct svc_fh *, const char *, int, struct svc_fh *); -int nfsd_setattr(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_setattr(struct svc_rqst *, struct svc_fh *, struct iattr *, int, time_t); #ifdef CONFIG_NFSD_V4 -int nfsd4_set_nfs4_acl(struct svc_rqst *, struct svc_fh *, +__be32 nfsd4_set_nfs4_acl(struct svc_rqst *, struct svc_fh *, struct nfs4_acl *); int nfsd4_get_nfs4_acl(struct svc_rqst *, struct dentry *, struct nfs4_acl **); #endif /* CONFIG_NFSD_V4 */ -int nfsd_create(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_create(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, int type, dev_t rdev, struct svc_fh *res); #ifdef CONFIG_NFSD_V3 -int nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); -int nfsd_create_v3(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); +__be32 nfsd_create_v3(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, struct svc_fh *res, int createmode, u32 *verifier, int *truncp); -int nfsd_commit(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_commit(struct svc_rqst *, struct svc_fh *, loff_t, unsigned long); #endif /* CONFIG_NFSD_V3 */ -int nfsd_open(struct svc_rqst *, struct svc_fh *, int, +__be32 nfsd_open(struct svc_rqst *, struct svc_fh *, int, int, struct file **); void nfsd_close(struct file *); -int nfsd_read(struct svc_rqst *, struct svc_fh *, struct file *, +__be32 nfsd_read(struct svc_rqst *, struct svc_fh *, struct file *, loff_t, struct kvec *, int, unsigned long *); -int nfsd_write(struct svc_rqst *, struct svc_fh *,struct file *, +__be32 nfsd_write(struct svc_rqst *, struct svc_fh *,struct file *, loff_t, struct kvec *,int, unsigned long, int *); -int nfsd_readlink(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_readlink(struct svc_rqst *, struct svc_fh *, char *, int *); -int nfsd_symlink(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_symlink(struct svc_rqst *, struct svc_fh *, char *name, int len, char *path, int plen, struct svc_fh *res, struct iattr *); -int nfsd_link(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_link(struct svc_rqst *, struct svc_fh *, char *, int, struct svc_fh *); -int nfsd_rename(struct svc_rqst *, +__be32 nfsd_rename(struct svc_rqst *, struct svc_fh *, char *, int, struct svc_fh *, char *, int); -int nfsd_remove(struct svc_rqst *, +__be32 nfsd_remove(struct svc_rqst *, struct svc_fh *, char *, int); -int nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type, +__be32 nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type, char *name, int len); int nfsd_truncate(struct svc_rqst *, struct svc_fh *, unsigned long size); -int nfsd_readdir(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_readdir(struct svc_rqst *, struct svc_fh *, loff_t *, struct readdir_cd *, encode_dent_fn); -int nfsd_statfs(struct svc_rqst *, struct svc_fh *, +__be32 nfsd_statfs(struct svc_rqst *, struct svc_fh *, struct kstatfs *); int nfsd_notify_change(struct inode *, struct iattr *); -int nfsd_permission(struct svc_export *, struct dentry *, int); +__be32 nfsd_permission(struct svc_export *, struct dentry *, int); int nfsd_sync_dir(struct dentry *dp); #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) -- cgit v1.2.3 From b37ad28bcaa7c486a4ff0fb6c3bdaaacd67b86ce Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:28:59 -0700 Subject: [PATCH] nfsd: nfs4 code returns error values in net-endian Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 82 ++++++++++++------------- fs/nfsd/nfs4recover.c | 14 ++--- fs/nfsd/nfs4state.c | 96 ++++++++++++++--------------- fs/nfsd/nfs4xdr.c | 150 ++++++++++++++++++++++----------------------- include/linux/nfsd/state.h | 10 +-- include/linux/nfsd/xdr4.h | 30 ++++----- 6 files changed, 191 insertions(+), 191 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index ca6414248527..63823945f972 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -67,10 +67,10 @@ fh_dup2(struct svc_fh *dst, struct svc_fh *src) *dst = *src; } -static int +static __be32 do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode) { - int status; + __be32 status; if (open->op_truncate && !(open->op_share_access & NFS4_SHARE_ACCESS_WRITE)) @@ -88,11 +88,11 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs return status; } -static int +static __be32 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct svc_fh resfh; - int status; + __be32 status; fh_init(&resfh, NFS4_FHSIZE); open->op_truncate = 0; @@ -131,10 +131,10 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o return status; } -static int +static __be32 do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { - int status; + __be32 status; /* Only reclaims from previously confirmed clients are valid */ if ((status = nfs4_check_open_reclaim(&open->op_clientid))) @@ -161,10 +161,10 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ } -static inline int +static inline __be32 nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; dprintk("NFSD: nfsd4_open filename %.*s op_stateowner %p\n", (int)open->op_fname.len, open->op_fname.data, open->op_stateowner); @@ -261,7 +261,7 @@ out: /* * filehandle-manipulating ops. */ -static inline int +static inline __be32 nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) { if (!current_fh->fh_dentry) @@ -271,7 +271,7 @@ nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putfh *putfh) { fh_put(current_fh); @@ -280,10 +280,10 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putf return fh_verify(rqstp, current_fh, 0, MAY_NOP); } -static inline int +static inline __be32 nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) { - int status; + __be32 status; fh_put(current_fh); status = exp_pseudoroot(rqstp->rq_client, current_fh, @@ -291,7 +291,7 @@ nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) return status; } -static inline int +static inline __be32 nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!save_fh->fh_dentry) @@ -301,7 +301,7 @@ nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!current_fh->fh_dentry) @@ -314,7 +314,7 @@ nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) /* * misc nfsv4 ops */ -static inline int +static inline __be32 nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_access *access) { if (access->ac_req_access & ~NFS3_ACCESS_FULL) @@ -324,10 +324,10 @@ nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_acc return nfsd_access(rqstp, current_fh, &access->ac_resp_access, &access->ac_supported); } -static inline int +static inline __be32 nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_commit *commit) { - int status; + __be32 status; u32 *p = (u32 *)commit->co_verf.data; *p++ = nfssvc_boot.tv_sec; @@ -339,11 +339,11 @@ nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_com return status; } -static int +static __be32 nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_create *create) { struct svc_fh resfh; - int status; + __be32 status; dev_t rdev; fh_init(&resfh, NFS4_FHSIZE); @@ -423,10 +423,10 @@ nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_cre return status; } -static inline int +static inline __be32 nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_getattr *getattr) { - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -442,11 +442,11 @@ nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ge return nfs_ok; } -static inline int +static inline __be32 nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_link *link) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -456,11 +456,11 @@ nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static int +static __be32 nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) { struct svc_fh tmp_fh; - int ret; + __be32 ret; fh_init(&tmp_fh, NFS4_FHSIZE); if((ret = exp_pseudoroot(rqstp->rq_client, &tmp_fh, @@ -474,16 +474,16 @@ nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) return nfsd_lookup(rqstp, current_fh, "..", 2, current_fh); } -static inline int +static inline __be32 nfsd4_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lookup *lookup) { return nfsd_lookup(rqstp, current_fh, lookup->lo_name, lookup->lo_len, current_fh); } -static inline int +static inline __be32 nfsd4_read(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_read *read) { - int status; + __be32 status; /* no need to check permission - this will be done in nfsd_read() */ @@ -508,7 +508,7 @@ out: return status; } -static inline int +static inline __be32 nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readdir *readdir) { u64 cookie = readdir->rd_cookie; @@ -531,7 +531,7 @@ nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_re return nfs_ok; } -static inline int +static inline __be32 nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readlink *readlink) { readlink->rl_rqstp = rqstp; @@ -539,10 +539,10 @@ nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_r return nfs_ok; } -static inline int +static inline __be32 nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_remove *remove) { - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -556,11 +556,11 @@ nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_rem return status; } -static inline int +static inline __be32 nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_rename *rename) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -589,10 +589,10 @@ nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static inline int +static inline __be32 nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_setattr *setattr) { - int status = nfs_ok; + __be32 status = nfs_ok; if (setattr->sa_iattr.ia_valid & ATTR_SIZE) { nfs4_lock_state(); @@ -614,13 +614,13 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_se return status; } -static inline int +static inline __be32 nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_write *write) { stateid_t *stateid = &write->wr_stateid; struct file *filp = NULL; u32 *p; - int status = nfs_ok; + __be32 status = nfs_ok; /* no need to check permission - this will be done in nfsd_write() */ @@ -661,12 +661,12 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ * attributes matched. VERIFY is implemented by mapping NFSERR_SAME * to NFS_OK after the call; NVERIFY by mapping NFSERR_NOT_SAME to NFS_OK. */ -static int +static __be32 nfsd4_verify(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_verify *verify) { __be32 *buf, *p; int count; - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -741,7 +741,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, struct svc_fh *save_fh = NULL; struct nfs4_stateowner *replay_owner = NULL; int slack_space; /* in words, not bytes! */ - int status; + __be32 status; status = nfserr_resource; current_fh = kmalloc(sizeof(*current_fh), GFP_KERNEL); diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 1cbd2e4ee122..e9d07704680e 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -83,13 +83,13 @@ md5_to_hex(char *out, char *md5) *out = '\0'; } -int +__be32 nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname) { struct xdr_netobj cksum; struct hash_desc desc; struct scatterlist sg[1]; - int status = nfserr_resource; + __be32 status = nfserr_resource; dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n", clname->len, clname->data); @@ -193,7 +193,7 @@ nfsd4_build_dentrylist(void *arg, const char *name, int namlen, struct dentry_list *child; if (name && isdotent(name, namlen)) - return nfs_ok; + return 0; dentry = lookup_one_len(name, parent, namlen); if (IS_ERR(dentry)) return PTR_ERR(dentry); @@ -333,14 +333,14 @@ purge_old(struct dentry *parent, struct dentry *child) int status; if (nfs4_has_reclaimed_state(child->d_name.name)) - return nfs_ok; + return 0; status = nfsd4_clear_clid_dir(parent, child); if (status) printk("failed to remove client recovery directory %s\n", child->d_name.name); /* Keep trying, success or failure: */ - return nfs_ok; + return 0; } void @@ -365,10 +365,10 @@ load_recdir(struct dentry *parent, struct dentry *child) printk("nfsd4: illegal name %s in recovery directory\n", child->d_name.name); /* Keep trying; maybe the others are OK: */ - return nfs_ok; + return 0; } nfs4_client_to_reclaim(child->d_name.name); - return nfs_ok; + return 0; } int diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ebcf226a9e4a..e5ca6d7028df 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -710,7 +710,7 @@ out_err: * as described above. * */ -int +__be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) { u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; @@ -721,7 +721,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) nfs4_verifier clverifier = setclid->se_verf; unsigned int strhashval; struct nfs4_client *conf, *unconf, *new; - int status; + __be32 status; char dname[HEXDIR_LEN]; if (!check_name(clname)) @@ -875,14 +875,14 @@ out: * * NOTE: callback information will be processed here in a future patch */ -int +__be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm) { u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; - int status; + __be32 status; if (STALE_CLIENTID(clid)) return nfserr_stale_clientid; @@ -1280,13 +1280,13 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) { * Called to check deny when READ with all zero stateid or * WRITE with all zero or all one stateid */ -static int +static __be32 nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) { struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_file *fp; struct nfs4_stateid *stp; - int ret; + __be32 ret; dprintk("NFSD: nfs4_share_conflict\n"); @@ -1444,7 +1444,7 @@ static struct lock_manager_operations nfsd_lease_mng_ops = { }; -int +__be32 nfsd4_process_open1(struct nfsd4_open *open) { clientid_t *clientid = &open->op_clientid; @@ -1501,7 +1501,7 @@ renew: return nfs_ok; } -static inline int +static inline __be32 nfs4_check_delegmode(struct nfs4_delegation *dp, int flags) { if ((flags & WR_STATE) && (dp->dl_type == NFS4_OPEN_DELEGATE_READ)) @@ -1522,12 +1522,12 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) return NULL; } -static int +static __be32 nfs4_check_deleg(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_delegation **dp) { int flags; - int status = nfserr_bad_stateid; + __be32 status = nfserr_bad_stateid; *dp = find_delegation_file(fp, &open->op_delegate_stateid); if (*dp == NULL) @@ -1546,11 +1546,11 @@ out: return nfs_ok; } -static int +static __be32 nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_stateid **stpp) { struct nfs4_stateid *local; - int status = nfserr_share_denied; + __be32 status = nfserr_share_denied; struct nfs4_stateowner *sop = open->op_stateowner; list_for_each_entry(local, &fp->fi_stateids, st_perfile) { @@ -1575,7 +1575,7 @@ nfs4_alloc_stateid(void) return kmem_cache_alloc(stateid_slab, GFP_KERNEL); } -static int +static __be32 nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, struct nfs4_delegation *dp, struct svc_fh *cur_fh, int flags) @@ -1590,7 +1590,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, get_file(dp->dl_vfs_file); stp->st_vfs_file = dp->dl_vfs_file; } else { - int status; + __be32 status; status = nfsd_open(rqstp, cur_fh, S_IFREG, flags, &stp->st_vfs_file); if (status) { @@ -1604,7 +1604,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, return 0; } -static inline int +static inline __be32 nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, struct nfsd4_open *open) { @@ -1619,22 +1619,22 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, return nfsd_setattr(rqstp, fh, &iattr, 0, (time_t)0); } -static int +static __be32 nfs4_upgrade_open(struct svc_rqst *rqstp, struct svc_fh *cur_fh, struct nfs4_stateid *stp, struct nfsd4_open *open) { struct file *filp = stp->st_vfs_file; struct inode *inode = filp->f_dentry->d_inode; unsigned int share_access, new_writer; - int status; + __be32 status; set_access(&share_access, stp->st_access_bmap); new_writer = (~share_access) & open->op_share_access & NFS4_SHARE_ACCESS_WRITE; if (new_writer) { - status = get_write_access(inode); - if (status) - return nfserrno(status); + int err = get_write_access(inode); + if (err) + return nfserrno(err); } status = nfsd4_truncate(rqstp, cur_fh, open); if (status) { @@ -1738,14 +1738,14 @@ out: /* * called with nfs4_lock_state() held. */ -int +__be32 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct nfs4_file *fp = NULL; struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; - int status; + __be32 status; status = nfserr_inval; if (!access_valid(open->op_share_access) @@ -1833,11 +1833,11 @@ static struct work_struct laundromat_work; static void laundromat_main(void *); static DECLARE_WORK(laundromat_work, laundromat_main, NULL); -int +__be32 nfsd4_renew(clientid_t *clid) { struct nfs4_client *clp; - int status; + __be32 status; nfs4_lock_state(); dprintk("process_renew(%08x/%08x): starting\n", @@ -1996,9 +1996,9 @@ access_permit_write(unsigned long access_bmap) } static -int nfs4_check_openmode(struct nfs4_stateid *stp, int flags) +__be32 nfs4_check_openmode(struct nfs4_stateid *stp, int flags) { - int status = nfserr_openmode; + __be32 status = nfserr_openmode; if ((flags & WR_STATE) && (!access_permit_write(stp->st_access_bmap))) goto out; @@ -2009,7 +2009,7 @@ out: return status; } -static inline int +static inline __be32 check_special_stateids(svc_fh *current_fh, stateid_t *stateid, int flags) { /* Trying to call delegreturn with a special stateid? Yuch: */ @@ -2043,14 +2043,14 @@ io_during_grace_disallowed(struct inode *inode, int flags) /* * Checks for stateid operations */ -int +__be32 nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filpp) { struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; stateid_t *stidp; struct inode *ino = current_fh->fh_dentry->d_inode; - int status; + __be32 status; dprintk("NFSD: preprocess_stateid_op: stateid = (%08x/%08x/%08x/%08x)\n", stateid->si_boot, stateid->si_stateownerid, @@ -2125,7 +2125,7 @@ setlkflg (int type) /* * Checks for sequence id mutating operations. */ -static int +static __be32 nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, struct nfsd4_lock *lock) { struct nfs4_stateid *stp; @@ -2169,7 +2169,7 @@ nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *statei clientid_t *lockclid = &lock->v.new.clientid; struct nfs4_client *clp = sop->so_client; int lkflg = 0; - int status; + __be32 status; lkflg = setlkflg(lock->lk_type); @@ -2241,10 +2241,10 @@ check_replay: return nfserr_bad_seqid; } -int +__be32 nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateowner *sop; struct nfs4_stateid *stp; @@ -2310,10 +2310,10 @@ reset_union_bmap_deny(unsigned long deny, unsigned long *bmap) } } -int +__be32 nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; unsigned int share_access; @@ -2365,10 +2365,10 @@ out: /* * nfs4_unlock_state() called after encode */ -int +__be32 nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; dprintk("NFSD: nfsd4_close on file %.*s\n", @@ -2404,10 +2404,10 @@ out: return status; } -int +__be32 nfsd4_delegreturn(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_delegreturn *dr) { - int status; + __be32 status; if ((status = fh_verify(rqstp, current_fh, S_IFREG, 0))) goto out; @@ -2635,7 +2635,7 @@ check_lock_length(u64 offset, u64 length) /* * LOCK operation */ -int +__be32 nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner) { struct nfs4_stateowner *open_sop = NULL; @@ -2644,7 +2644,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock struct file *filp; struct file_lock file_lock; struct file_lock conflock; - int status = 0; + __be32 status = 0; unsigned int strhashval; dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n", @@ -2793,14 +2793,14 @@ out: /* * LOCKT operation */ -int +__be32 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt) { struct inode *inode; struct file file; struct file_lock file_lock; struct file_lock conflock; - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -2873,13 +2873,13 @@ out: return status; } -int +__be32 nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner) { struct nfs4_stateid *stp; struct file *filp = NULL; struct file_lock file_lock; - int status; + __be32 status; dprintk("NFSD: nfsd4_locku: start=%Ld length=%Ld\n", (long long) locku->lu_offset, @@ -2965,7 +2965,7 @@ out: return status; } -int +__be32 nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner) { clientid_t *clid = &rlockowner->rl_clientid; @@ -2974,7 +2974,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner * struct xdr_netobj *owner = &rlockowner->rl_owner; struct list_head matches; int i; - int status; + __be32 status; dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n", clid->cl_boot, clid->cl_id); @@ -3111,7 +3111,7 @@ nfs4_find_reclaim_client(clientid_t *clid) /* * Called from OPEN. Look for clientid in reclaim list. */ -int +__be32 nfs4_check_open_reclaim(clientid_t *clid) { return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 3419d99aeb1a..d7b630f1a9ae 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -68,8 +68,8 @@ #define NFS4_REFERRAL_FSID_MAJOR 0x8000000ULL #define NFS4_REFERRAL_FSID_MINOR 0x8000000ULL -static int -check_filename(char *str, int len, int err) +static __be32 +check_filename(char *str, int len, __be32 err) { int i; @@ -95,7 +95,7 @@ check_filename(char *str, int len, int err) */ #define DECODE_HEAD \ __be32 *p; \ - int status + __be32 status #define DECODE_TAIL \ status = 0; \ out: \ @@ -217,7 +217,7 @@ static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) } -static int +static __be32 nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) { u32 bmlen; @@ -240,7 +240,7 @@ nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *iattr, struct nfs4_acl **acl) { @@ -418,7 +418,7 @@ out_nfserr: goto out; } -static int +static __be32 nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access) { DECODE_HEAD; @@ -429,7 +429,7 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access DECODE_TAIL; } -static int +static __be32 nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) { DECODE_HEAD; @@ -444,7 +444,7 @@ nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) } -static int +static __be32 nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit) { DECODE_HEAD; @@ -456,7 +456,7 @@ nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit DECODE_TAIL; } -static int +static __be32 nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create) { DECODE_HEAD; @@ -496,7 +496,7 @@ nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegreturn *dr) { DECODE_HEAD; @@ -508,13 +508,13 @@ nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegretu DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_getattr(struct nfsd4_compoundargs *argp, struct nfsd4_getattr *getattr) { return nfsd4_decode_bitmap(argp, getattr->ga_bmval); } -static int +static __be32 nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) { DECODE_HEAD; @@ -529,7 +529,7 @@ nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) { DECODE_HEAD; @@ -568,7 +568,7 @@ nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) { DECODE_HEAD; @@ -587,7 +587,7 @@ nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) { DECODE_HEAD; @@ -606,7 +606,7 @@ nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup) { DECODE_HEAD; @@ -621,7 +621,7 @@ nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) { DECODE_HEAD; @@ -699,7 +699,7 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf) { DECODE_HEAD; @@ -713,7 +713,7 @@ nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_con DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_downgrade *open_down) { DECODE_HEAD; @@ -729,7 +729,7 @@ nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_d DECODE_TAIL; } -static int +static __be32 nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) { DECODE_HEAD; @@ -744,7 +744,7 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) { DECODE_HEAD; @@ -758,7 +758,7 @@ nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *readdir) { DECODE_HEAD; @@ -774,7 +774,7 @@ nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *read DECODE_TAIL; } -static int +static __be32 nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove) { DECODE_HEAD; @@ -789,7 +789,7 @@ nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove DECODE_TAIL; } -static int +static __be32 nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename) { DECODE_HEAD; @@ -809,7 +809,7 @@ nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename DECODE_TAIL; } -static int +static __be32 nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) { DECODE_HEAD; @@ -820,7 +820,7 @@ nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr) { DECODE_HEAD; @@ -834,7 +834,7 @@ nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *seta DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid *setclientid) { DECODE_HEAD; @@ -859,7 +859,7 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid_confirm *scd_c) { DECODE_HEAD; @@ -872,7 +872,7 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s } /* Also used for NVERIFY */ -static int +static __be32 nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify) { #if 0 @@ -908,7 +908,7 @@ nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify DECODE_TAIL; } -static int +static __be32 nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) { int avail; @@ -959,7 +959,7 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_release_lockowner *rlockowner) { DECODE_HEAD; @@ -973,7 +973,7 @@ nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_rel DECODE_TAIL; } -static int +static __be32 nfsd4_decode_compound(struct nfsd4_compoundargs *argp) { DECODE_HEAD; @@ -1234,7 +1234,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) /* Encode as an array of strings the string given with components * seperated @sep. */ -static int nfsd4_encode_components(char sep, char *components, +static __be32 nfsd4_encode_components(char sep, char *components, __be32 **pp, int *buflen) { __be32 *p = *pp; @@ -1271,10 +1271,10 @@ static int nfsd4_encode_components(char sep, char *components, /* * encode a location element of a fs_locations structure */ -static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, +static __be32 nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, __be32 **pp, int *buflen) { - int status; + __be32 status; __be32 *p = *pp; status = nfsd4_encode_components(':', location->hosts, &p, buflen); @@ -1292,7 +1292,7 @@ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, * Returned string is safe to use as long as the caller holds a reference * to @exp. */ -static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *stat) +static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *stat) { struct svc_fh tmp_fh; char *path, *rootpath; @@ -1318,11 +1318,11 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, u32 *sta /* * encode a fs_locations structure */ -static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, +static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, __be32 **pp, int *buflen) { - u32 status; + __be32 status; int i; __be32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; @@ -1353,7 +1353,7 @@ static u32 nfs4_ftypes[16] = { NF4SOCK, NF4BAD, NF4LNK, NF4BAD, }; -static int +static __be32 nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, __be32 **p, int *buflen) { @@ -1375,19 +1375,19 @@ nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, return 0; } -static inline int +static inline __be32 nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen); } -static inline int +static inline __be32 nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen); } -static inline int +static inline __be32 nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, __be32 **p, int *buflen) { @@ -1398,7 +1398,7 @@ nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, FATTR4_WORD0_RDATTR_ERROR) #define WORD1_ABSENT_FS_ATTRS FATTR4_WORD1_MOUNTED_ON_FILEID -static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) +static __be32 fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) { /* As per referral draft: */ if (*bmval0 & ~WORD0_ABSENT_FS_ATTRS || @@ -1421,7 +1421,7 @@ static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) * @countp is the buffer size in _words_; upon successful return this becomes * replaced with the number of words written. */ -int +__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *rqstp) @@ -1437,7 +1437,7 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, u64 dummy64; u32 rdattr_err = 0; __be32 *p = buffer; - int status; + __be32 status; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1829,13 +1829,13 @@ out_serverfault: goto out; } -static int +static __be32 nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, const char *name, int namlen, __be32 *p, int *buflen) { struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; - int nfserr; + __be32 nfserr; dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) @@ -1865,7 +1865,7 @@ out_put: } static __be32 * -nfsd4_encode_rdattr_error(__be32 *p, int buflen, int nfserr) +nfsd4_encode_rdattr_error(__be32 *p, int buflen, __be32 nfserr) { __be32 *attrlenp; @@ -1888,7 +1888,7 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; __be32 *p = cd->buffer; - int nfserr = nfserr_toosmall; + __be32 nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ if (name && isdotent(name, namlen)) { @@ -1944,7 +1944,7 @@ fail: } static void -nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_access *access) +nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_access *access) { ENCODE_HEAD; @@ -1957,7 +1957,7 @@ nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_acc } static void -nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_close *close) +nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close) { ENCODE_SEQID_OP_HEAD; @@ -1972,7 +1972,7 @@ nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_clos static void -nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_commit *commit) +nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_commit *commit) { ENCODE_HEAD; @@ -1984,7 +1984,7 @@ nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_com } static void -nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_create *create) +nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_create *create) { ENCODE_HEAD; @@ -1998,8 +1998,8 @@ nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_cre } } -static int -nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_getattr *getattr) +static __be32 +nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_getattr *getattr) { struct svc_fh *fhp = getattr->ga_fhp; int buflen; @@ -2017,7 +2017,7 @@ nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ge } static void -nfsd4_encode_getfh(struct nfsd4_compoundres *resp, int nfserr, struct svc_fh *fhp) +nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh *fhp) { unsigned int len; ENCODE_HEAD; @@ -2057,7 +2057,7 @@ nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denie } static void -nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock *lock) +nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lock *lock) { ENCODE_SEQID_OP_HEAD; @@ -2073,14 +2073,14 @@ nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock } static void -nfsd4_encode_lockt(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lockt *lockt) +nfsd4_encode_lockt(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lockt *lockt) { if (nfserr == nfserr_denied) nfsd4_encode_lock_denied(resp, &lockt->lt_denied); } static void -nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_locku *locku) +nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_locku *locku) { ENCODE_SEQID_OP_HEAD; @@ -2096,7 +2096,7 @@ nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock static void -nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link *link) +nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_link *link) { ENCODE_HEAD; @@ -2109,7 +2109,7 @@ nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link static void -nfsd4_encode_open(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open *open) +nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open *open) { ENCODE_SEQID_OP_HEAD; @@ -2174,7 +2174,7 @@ out: } static void -nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_confirm *oc) +nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_confirm *oc) { ENCODE_SEQID_OP_HEAD; @@ -2189,7 +2189,7 @@ nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfs } static void -nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_downgrade *od) +nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_downgrade *od) { ENCODE_SEQID_OP_HEAD; @@ -2203,8 +2203,8 @@ nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct n ENCODE_SEQID_OP_TAIL(od->od_stateowner); } -static int -nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, +static __be32 +nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_read *read) { u32 eof; @@ -2268,8 +2268,8 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, return 0; } -static int -nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readlink *readlink) +static __be32 +nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readlink *readlink) { int maxcount; char *page; @@ -2316,8 +2316,8 @@ nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_r return 0; } -static int -nfsd4_encode_readdir(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readdir *readdir) +static __be32 +nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readdir *readdir) { int maxcount; loff_t offset; @@ -2396,7 +2396,7 @@ err_no_verf: } static void -nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_remove *remove) +nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_remove *remove) { ENCODE_HEAD; @@ -2408,7 +2408,7 @@ nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rem } static void -nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rename *rename) +nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_rename *rename) { ENCODE_HEAD; @@ -2425,7 +2425,7 @@ nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ren * regardless of the error status. */ static void -nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setattr *setattr) +nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setattr *setattr) { ENCODE_HEAD; @@ -2444,7 +2444,7 @@ nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_se } static void -nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setclientid *scd) +nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setclientid *scd) { ENCODE_HEAD; @@ -2463,7 +2463,7 @@ nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd } static void -nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_write *write) +nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_write *write) { ENCODE_HEAD; @@ -2641,7 +2641,7 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) int nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundargs *args) { - int status; + __be32 status; args->p = p; args->end = rqstp->rq_arg.head[0].iov_base + rqstp->rq_arg.head[0].iov_len; diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 8bf23cf8b603..a597e2e72469 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -164,7 +164,7 @@ update_stateid(stateid_t *stateid) * is cached. */ struct nfs4_replay { - u32 rp_status; + __be32 rp_status; unsigned int rp_buflen; char *rp_buf; unsigned intrp_allocated; @@ -273,19 +273,19 @@ struct nfs4_stateid { ((err) != nfserr_stale_stateid) && \ ((err) != nfserr_bad_stateid)) -extern int nfsd4_renew(clientid_t *clid); -extern int nfs4_preprocess_stateid_op(struct svc_fh *current_fh, +extern __be32 nfsd4_renew(clientid_t *clid); +extern __be32 nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filp); extern void nfs4_lock_state(void); extern void nfs4_unlock_state(void); extern int nfs4_in_grace(void); -extern int nfs4_check_open_reclaim(clientid_t *clid); +extern __be32 nfs4_check_open_reclaim(clientid_t *clid); extern void put_nfs4_client(struct nfs4_client *clp); extern void nfs4_free_stateowner(struct kref *kref); extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); -extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); +extern __be32 nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); extern void nfsd4_init_recdir(char *recdir_name); extern int nfsd4_recdir_load(void); extern void nfsd4_shutdown_recdir(void); diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index 003193fe6fc6..45ca01b5f844 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -334,7 +334,7 @@ struct nfsd4_write { struct nfsd4_op { int opnum; - int status; + __be32 status; union { struct nfsd4_access access; struct nfsd4_close close; @@ -426,38 +426,38 @@ int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *, struct nfsd4_compoundres *); void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *); void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op); -int nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, - struct dentry *dentry, u32 *buffer, int *countp, +__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, + struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *); -extern int nfsd4_setclientid(struct svc_rqst *rqstp, +extern __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid); -extern int nfsd4_setclientid_confirm(struct svc_rqst *rqstp, +extern __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm); -extern int nfsd4_process_open1(struct nfsd4_open *open); -extern int nfsd4_process_open2(struct svc_rqst *rqstp, +extern __be32 nfsd4_process_open1(struct nfsd4_open *open); +extern __be32 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open); -extern int nfsd4_open_confirm(struct svc_rqst *rqstp, +extern __be32 nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **); -extern int nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner); -extern int nfsd4_open_downgrade(struct svc_rqst *rqstp, +extern __be32 nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner); -extern int nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner); -extern int nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt); -extern int nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, +extern __be32 nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner); -extern int +extern __be32 nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner); extern void nfsd4_release_compoundargs(struct nfsd4_compoundargs *); -extern int nfsd4_delegreturn(struct svc_rqst *rqstp, +extern __be32 nfsd4_delegreturn(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_delegreturn *dr); #endif -- cgit v1.2.3 From c7afef1f963bec198b186cc34b9e8c9b9ce2e266 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:02 -0700 Subject: [PATCH] nfsd: misc endianness annotations Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/export.c | 4 ++-- fs/nfsd/lockd.c | 2 +- fs/nfsd/nfs4callback.c | 2 +- fs/nfsd/nfs4state.c | 4 ++-- fs/nfsd/nfscache.c | 8 ++++---- fs/nfsd/nfssvc.c | 2 +- include/linux/nfsd/cache.h | 6 +++--- include/linux/nfsd/export.h | 2 +- include/linux/nfsd/nfsd.h | 2 +- include/linux/nfsd/nfsfh.h | 2 +- include/linux/nfsd/state.h | 2 +- 11 files changed, 18 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index e13fa23bd108..f37df46d2eaa 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1148,12 +1148,12 @@ exp_find(struct auth_domain *clp, int fsid_type, u32 *fsidv, * for a given NFSv4 client. The root is defined to be the * export point with fsid==0 */ -int +__be32 exp_pseudoroot(struct auth_domain *clp, struct svc_fh *fhp, struct cache_req *creq) { struct svc_export *exp; - int rv; + __be32 rv; u32 fsidv[2]; mk_fsid_v1(fsidv, 0); diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 9b9e7e127c03..11fdaf7721b4 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -25,7 +25,7 @@ static u32 nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) { - u32 nfserr; + __be32 nfserr; struct svc_fh fh; /* must initialize before using! but maxsize doesn't matter */ diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 8497ed4862b2..f57655a7a2b6 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -461,7 +461,7 @@ nfs4_cb_null(struct rpc_task *task, void *dummy) { struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp; struct nfs4_callback *cb = &clp->cl_callback; - u32 addr = htonl(cb->cb_addr); + __be32 addr = htonl(cb->cb_addr); dprintk("NFSD: nfs4_cb_null task->tk_status %d\n", task->tk_status); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ae1d47715b90..2e468c9e64d9 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -713,7 +713,7 @@ out_err: __be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct xdr_netobj clname = { .len = setclid->se_namelen, .data = setclid->se_name, @@ -878,7 +878,7 @@ out: __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index fdf7cf3dfadc..6100bbe27432 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -29,7 +29,7 @@ */ #define CACHESIZE 1024 #define HASHSIZE 64 -#define REQHASH(xid) ((((xid) >> 24) ^ (xid)) & (HASHSIZE-1)) +#define REQHASH(xid) (((((__force __u32)xid) >> 24) ^ ((__force __u32)xid)) & (HASHSIZE-1)) static struct hlist_head * hash_list; static struct list_head lru_head; @@ -127,8 +127,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp, int type) struct hlist_node *hn; struct hlist_head *rh; struct svc_cacherep *rp; - u32 xid = rqstp->rq_xid, - proto = rqstp->rq_prot, + __be32 xid = rqstp->rq_xid; + u32 proto = rqstp->rq_prot, vers = rqstp->rq_vers, proc = rqstp->rq_proc; unsigned long age; @@ -258,7 +258,7 @@ found_entry: * In this case, nfsd_cache_update is called with statp == NULL. */ void -nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, u32 *statp) +nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp) { struct svc_cacherep *rp; struct kvec *resv = &rqstp->rq_res.head[0], *cachv; diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 8067118b1c0c..0aaccb03bf76 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -491,7 +491,7 @@ out: } int -nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) +nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp) { struct svc_procedure *proc; kxdrproc_t xdr; diff --git a/include/linux/nfsd/cache.h b/include/linux/nfsd/cache.h index c3a3557c2a5b..007480cd6a60 100644 --- a/include/linux/nfsd/cache.h +++ b/include/linux/nfsd/cache.h @@ -26,14 +26,14 @@ struct svc_cacherep { c_type, /* status, buffer */ c_secure : 1; /* req came from port < 1024 */ struct sockaddr_in c_addr; - u32 c_xid; + __be32 c_xid; u32 c_prot; u32 c_proc; u32 c_vers; unsigned long c_timestamp; union { struct kvec u_vec; - u32 u_status; + __be32 u_status; } c_u; }; @@ -75,7 +75,7 @@ enum { void nfsd_cache_init(void); void nfsd_cache_shutdown(void); int nfsd_cache_lookup(struct svc_rqst *, int); -void nfsd_cache_update(struct svc_rqst *, int, u32 *); +void nfsd_cache_update(struct svc_rqst *, int, __be32 *); #endif /* __KERNEL__ */ #endif /* NFSCACHE_H */ diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 27666f5b8b53..045e38cdbe64 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -117,7 +117,7 @@ struct svc_export * exp_parent(struct auth_domain *clp, struct cache_req *reqp); int exp_rootfh(struct auth_domain *, char *path, struct knfsd_fh *, int maxsize); -int exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); +__be32 exp_pseudoroot(struct auth_domain *, struct svc_fh *fhp, struct cache_req *creq); __be32 nfserrno(int errno); extern struct cache_detail svc_export_cache; diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 19a3c83d496e..68d29b66c6e2 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -64,7 +64,7 @@ extern struct svc_serv *nfsd_serv; * Function prototypes. */ int nfsd_svc(unsigned short port, int nrservs); -int nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp); +int nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp); /* nfsd/vfs.c */ int fh_lock_parent(struct svc_fh *, struct dentry *); diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index 749bad1ca16f..f3b51d62ec7d 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -157,7 +157,7 @@ typedef struct svc_fh { __u64 fh_post_size; /* i_size */ unsigned long fh_post_blocks; /* i_blocks */ unsigned long fh_post_blksize;/* i_blksize */ - __u32 fh_post_rdev[2];/* i_rdev */ + __be32 fh_post_rdev[2];/* i_rdev */ struct timespec fh_post_atime; /* i_atime */ struct timespec fh_post_mtime; /* i_mtime */ struct timespec fh_post_ctime; /* i_ctime */ diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index a597e2e72469..c3673f487e84 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -125,7 +125,7 @@ struct nfs4_client { char cl_recdir[HEXDIR_LEN]; /* recovery dir */ nfs4_verifier cl_verifier; /* generated by client */ time_t cl_time; /* time of last lease renewal */ - u32 cl_addr; /* client ipaddress */ + __be32 cl_addr; /* client ipaddress */ struct svc_cred cl_cred; /* setclientid principal */ clientid_t cl_clientid; /* generated by server */ nfs4_verifier cl_confirm; /* generated by server */ -- cgit v1.2.3 From a90b061c0bf712961cea40d9c916b300073d12e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 23:29:03 -0700 Subject: [PATCH] nfsd: nfs_replay_me We are using NFS_REPLAY_ME as a special error value that is never leaked to clients. That works fine; the only problem is mixing host- and network- endian values in the same objects. Network-endian equivalent would work just as fine; switch to it. Signed-off-by: Al Viro Acked-by: Trond Myklebust Acked-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs4proc.c | 6 +++--- fs/nfsd/nfs4state.c | 4 ++-- include/linux/nfsd/nfsd.h | 1 + 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 63823945f972..0a7bbdc4a10a 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -177,7 +177,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open /* check seqid for replay. set nfs4_owner */ status = nfsd4_process_open1(open); - if (status == NFSERR_REPLAY_ME) { + if (status == nfserr_replay_me) { struct nfs4_replay *rp = &open->op_stateowner->so_replay; fh_put(current_fh); current_fh->fh_handle.fh_size = rp->rp_openfh_len; @@ -188,7 +188,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open dprintk("nfsd4_open: replay failed" " restoring previous filehandle\n"); else - status = NFSERR_REPLAY_ME; + status = nfserr_replay_me; } if (status) goto out; @@ -937,7 +937,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, } encode_op: - if (op->status == NFSERR_REPLAY_ME) { + if (op->status == nfserr_replay_me) { op->replay = &replay_owner->so_replay; nfsd4_encode_replay(resp, op); status = op->status = op->replay->rp_status; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 2e468c9e64d9..293b6495829f 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1477,7 +1477,7 @@ nfsd4_process_open1(struct nfsd4_open *open) } if (open->op_seqid == sop->so_seqid - 1) { if (sop->so_replay.rp_buflen) - return NFSERR_REPLAY_ME; + return nfserr_replay_me; /* The original OPEN failed so spectacularly * that we don't even have replay data saved! * Therefore, we have no choice but to continue @@ -2233,7 +2233,7 @@ check_replay: if (seqid == sop->so_seqid - 1) { dprintk("NFSD: preprocess_seqid_op: retransmission?\n"); /* indicate replay to calling function */ - return NFSERR_REPLAY_ME; + return nfserr_replay_me; } printk("NFSD: preprocess_seqid_op: bad seqid (expected %d, got %d)\n", sop->so_seqid, seqid); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 68d29b66c6e2..eb231143d579 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -238,6 +238,7 @@ void nfsd_lockd_shutdown(void); #define nfserr_badname __constant_htonl(NFSERR_BADNAME) #define nfserr_cb_path_down __constant_htonl(NFSERR_CB_PATH_DOWN) #define nfserr_locked __constant_htonl(NFSERR_LOCKED) +#define nfserr_replay_me __constant_htonl(NFSERR_REPLAY_ME) /* error codes for internal use */ /* if a request fails due to kmalloc failure, it gets dropped. -- cgit v1.2.3 From 690a973f48b6ba2954465992c08e65059c8374fe Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Sat, 21 Oct 2006 18:37:01 +0200 Subject: [PATCH] x86-64: Speed up dwarf2 unwinder This changes the dwarf2 unwinder to do a binary search for CIEs instead of a linear work. The linker is unfortunately not able to build a proper lookup table at link time, instead it creates one at runtime as soon as the bootmem allocator is usable (so you'll continue using the linear lookup for the first [hopefully] few calls). The code should be ready to utilize a build-time created table once a fixed linker becomes available. Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- Makefile | 1 + include/asm-generic/vmlinux.lds.h | 16 ++ include/linux/unwind.h | 2 + init/main.c | 1 + kernel/unwind.c | 318 +++++++++++++++++++++++++++++++++----- 5 files changed, 299 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/Makefile b/Makefile index 62a1343cf327..389ff0cca9a7 100644 --- a/Makefile +++ b/Makefile @@ -499,6 +499,7 @@ endif ifdef CONFIG_UNWIND_INFO CFLAGS += -fasynchronous-unwind-tables +LDFLAGS_vmlinux += --eh-frame-hdr endif ifdef CONFIG_DEBUG_INFO diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 69240b52f8e1..9d0d11c180d9 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -125,6 +125,10 @@ *(__param) \ VMLINUX_SYMBOL(__stop___param) = .; \ } \ + \ + /* Unwind data binary search table */ \ + EH_FRAME_HDR \ + \ __end_rodata = .; \ . = ALIGN(4096); @@ -157,6 +161,18 @@ *(.kprobes.text) \ VMLINUX_SYMBOL(__kprobes_text_end) = .; +#ifdef CONFIG_STACK_UNWIND + /* Unwind data binary search table */ +#define EH_FRAME_HDR \ + .eh_frame_hdr : AT(ADDR(.eh_frame_hdr) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_unwind_hdr) = .; \ + *(.eh_frame_hdr) \ + VMLINUX_SYMBOL(__end_unwind_hdr) = .; \ + } +#else +#define EH_FRAME_HDR +#endif + /* DWARF debug sections. Symbols in the DWARF debugging sections are relative to the beginning of the section so we begin them at 0. */ diff --git a/include/linux/unwind.h b/include/linux/unwind.h index 73e1751d03dd..749928c161fb 100644 --- a/include/linux/unwind.h +++ b/include/linux/unwind.h @@ -26,6 +26,7 @@ struct module; * Initialize unwind support. */ extern void unwind_init(void); +extern void unwind_setup(void); #ifdef CONFIG_MODULES @@ -73,6 +74,7 @@ extern int unwind_to_user(struct unwind_frame_info *); struct unwind_frame_info {}; static inline void unwind_init(void) {} +static inline void unwind_setup(void) {} #ifdef CONFIG_MODULES diff --git a/init/main.c b/init/main.c index ee123243fb53..36f608a7cfba 100644 --- a/init/main.c +++ b/init/main.c @@ -503,6 +503,7 @@ asmlinkage void __init start_kernel(void) printk(KERN_NOTICE); printk(linux_banner); setup_arch(&command_line); + unwind_setup(); setup_per_cpu_areas(); smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */ diff --git a/kernel/unwind.c b/kernel/unwind.c index 2e2368607aab..f7e50d16dbf6 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -11,13 +11,15 @@ #include #include -#include +#include +#include #include #include #include #include extern char __start_unwind[], __end_unwind[]; +extern const u8 __start_unwind_hdr[], __end_unwind_hdr[]; #define MAX_STACK_DEPTH 8 @@ -100,6 +102,8 @@ static struct unwind_table { } core, init; const void *address; unsigned long size; + const unsigned char *header; + unsigned long hdrsz; struct unwind_table *link; const char *name; } root_table; @@ -145,6 +149,10 @@ static struct unwind_table *find_table(unsigned long pc) return table; } +static unsigned long read_pointer(const u8 **pLoc, + const void *end, + signed ptrType); + static void init_unwind_table(struct unwind_table *table, const char *name, const void *core_start, @@ -152,14 +160,30 @@ static void init_unwind_table(struct unwind_table *table, const void *init_start, unsigned long init_size, const void *table_start, - unsigned long table_size) + unsigned long table_size, + const u8 *header_start, + unsigned long header_size) { + const u8 *ptr = header_start + 4; + const u8 *end = header_start + header_size; + table->core.pc = (unsigned long)core_start; table->core.range = core_size; table->init.pc = (unsigned long)init_start; table->init.range = init_size; table->address = table_start; table->size = table_size; + /* See if the linker provided table looks valid. */ + if (header_size <= 4 + || header_start[0] != 1 + || (void *)read_pointer(&ptr, end, header_start[1]) != table_start + || header_start[2] == DW_EH_PE_omit + || read_pointer(&ptr, end, header_start[2]) <= 0 + || header_start[3] == DW_EH_PE_omit) + header_start = NULL; + table->hdrsz = header_size; + smp_wmb(); + table->header = header_start; table->link = NULL; table->name = name; } @@ -169,7 +193,143 @@ void __init unwind_init(void) init_unwind_table(&root_table, "kernel", _text, _end - _text, NULL, 0, - __start_unwind, __end_unwind - __start_unwind); + __start_unwind, __end_unwind - __start_unwind, + __start_unwind_hdr, __end_unwind_hdr - __start_unwind_hdr); +} + +static const u32 bad_cie, not_fde; +static const u32 *cie_for_fde(const u32 *fde, const struct unwind_table *); +static signed fde_pointer_type(const u32 *cie); + +struct eh_frame_hdr_table_entry { + unsigned long start, fde; +}; + +static int cmp_eh_frame_hdr_table_entries(const void *p1, const void *p2) +{ + const struct eh_frame_hdr_table_entry *e1 = p1; + const struct eh_frame_hdr_table_entry *e2 = p2; + + return (e1->start > e2->start) - (e1->start < e2->start); +} + +static void swap_eh_frame_hdr_table_entries(void *p1, void *p2, int size) +{ + struct eh_frame_hdr_table_entry *e1 = p1; + struct eh_frame_hdr_table_entry *e2 = p2; + unsigned long v; + + v = e1->start; + e1->start = e2->start; + e2->start = v; + v = e1->fde; + e1->fde = e2->fde; + e2->fde = v; +} + +static void __init setup_unwind_table(struct unwind_table *table, + void *(*alloc)(unsigned long)) +{ + const u8 *ptr; + unsigned long tableSize = table->size, hdrSize; + unsigned n; + const u32 *fde; + struct { + u8 version; + u8 eh_frame_ptr_enc; + u8 fde_count_enc; + u8 table_enc; + unsigned long eh_frame_ptr; + unsigned int fde_count; + struct eh_frame_hdr_table_entry table[]; + } __attribute__((__packed__)) *header; + + if (table->header) + return; + + if (table->hdrsz) + printk(KERN_WARNING ".eh_frame_hdr for '%s' present but unusable\n", + table->name); + + if (tableSize & (sizeof(*fde) - 1)) + return; + + for (fde = table->address, n = 0; + tableSize > sizeof(*fde) && tableSize - sizeof(*fde) >= *fde; + tableSize -= sizeof(*fde) + *fde, fde += 1 + *fde / sizeof(*fde)) { + const u32 *cie = cie_for_fde(fde, table); + signed ptrType; + + if (cie == ¬_fde) + continue; + if (cie == NULL + || cie == &bad_cie + || (ptrType = fde_pointer_type(cie)) < 0) + return; + ptr = (const u8 *)(fde + 2); + if (!read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType)) + return; + ++n; + } + + if (tableSize || !n) + return; + + hdrSize = 4 + sizeof(unsigned long) + sizeof(unsigned int) + + 2 * n * sizeof(unsigned long); + header = alloc(hdrSize); + if (!header) + return; + header->version = 1; + header->eh_frame_ptr_enc = DW_EH_PE_abs|DW_EH_PE_native; + header->fde_count_enc = DW_EH_PE_abs|DW_EH_PE_data4; + header->table_enc = DW_EH_PE_abs|DW_EH_PE_native; + put_unaligned((unsigned long)table->address, &header->eh_frame_ptr); + BUILD_BUG_ON(offsetof(typeof(*header), fde_count) + % __alignof(typeof(header->fde_count))); + header->fde_count = n; + + BUILD_BUG_ON(offsetof(typeof(*header), table) + % __alignof(typeof(*header->table))); + for (fde = table->address, tableSize = table->size, n = 0; + tableSize; + tableSize -= sizeof(*fde) + *fde, fde += 1 + *fde / sizeof(*fde)) { + const u32 *cie = fde + 1 - fde[1] / sizeof(*fde); + + if (!fde[1]) + continue; /* this is a CIE */ + ptr = (const u8 *)(fde + 2); + header->table[n].start = read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + fde_pointer_type(cie)); + header->table[n].fde = (unsigned long)fde; + ++n; + } + WARN_ON(n != header->fde_count); + + sort(header->table, + n, + sizeof(*header->table), + cmp_eh_frame_hdr_table_entries, + swap_eh_frame_hdr_table_entries); + + table->hdrsz = hdrSize; + smp_wmb(); + table->header = (const void *)header; +} + +static void *__init balloc(unsigned long sz) +{ + return __alloc_bootmem_nopanic(sz, + sizeof(unsigned int), + __pa(MAX_DMA_ADDRESS)); +} + +void __init unwind_setup(void) +{ + setup_unwind_table(&root_table, balloc); } #ifdef CONFIG_MODULES @@ -193,7 +353,8 @@ void *unwind_add_table(struct module *module, init_unwind_table(table, module->name, module->module_core, module->core_size, module->module_init, module->init_size, - table_start, table_size); + table_start, table_size, + NULL, 0); if (last_table) last_table->link = table; @@ -303,6 +464,26 @@ static sleb128_t get_sleb128(const u8 **pcur, const u8 *end) return value; } +static const u32 *cie_for_fde(const u32 *fde, const struct unwind_table *table) +{ + const u32 *cie; + + if (!*fde || (*fde & (sizeof(*fde) - 1))) + return &bad_cie; + if (!fde[1]) + return ¬_fde; /* this is a CIE */ + if ((fde[1] & (sizeof(*fde) - 1)) + || fde[1] > (unsigned long)(fde + 1) - (unsigned long)table->address) + return NULL; /* this is not a valid FDE */ + cie = fde + 1 - fde[1] / sizeof(*fde); + if (*cie <= sizeof(*cie) + 4 + || *cie >= fde[1] - sizeof(*fde) + || (*cie & (sizeof(*cie) - 1)) + || cie[1]) + return NULL; /* this is not a (valid) CIE */ + return cie; +} + static unsigned long read_pointer(const u8 **pLoc, const void *end, signed ptrType) @@ -610,49 +791,108 @@ int unwind(struct unwind_frame_info *frame) unsigned i; signed ptrType = -1; uleb128_t retAddrReg = 0; - struct unwind_table *table; + const struct unwind_table *table; struct unwind_state state; if (UNW_PC(frame) == 0) return -EINVAL; if ((table = find_table(pc)) != NULL && !(table->size & (sizeof(*fde) - 1))) { - unsigned long tableSize = table->size; - - for (fde = table->address; - tableSize > sizeof(*fde) && tableSize - sizeof(*fde) >= *fde; - tableSize -= sizeof(*fde) + *fde, - fde += 1 + *fde / sizeof(*fde)) { - if (!*fde || (*fde & (sizeof(*fde) - 1))) - break; - if (!fde[1]) - continue; /* this is a CIE */ - if ((fde[1] & (sizeof(*fde) - 1)) - || fde[1] > (unsigned long)(fde + 1) - - (unsigned long)table->address) - continue; /* this is not a valid FDE */ - cie = fde + 1 - fde[1] / sizeof(*fde); - if (*cie <= sizeof(*cie) + 4 - || *cie >= fde[1] - sizeof(*fde) - || (*cie & (sizeof(*cie) - 1)) - || cie[1] - || (ptrType = fde_pointer_type(cie)) < 0) { - cie = NULL; /* this is not a (valid) CIE */ - continue; + const u8 *hdr = table->header; + unsigned long tableSize; + + smp_rmb(); + if (hdr && hdr[0] == 1) { + switch(hdr[3] & DW_EH_PE_FORM) { + case DW_EH_PE_native: tableSize = sizeof(unsigned long); break; + case DW_EH_PE_data2: tableSize = 2; break; + case DW_EH_PE_data4: tableSize = 4; break; + case DW_EH_PE_data8: tableSize = 8; break; + default: tableSize = 0; break; + } + ptr = hdr + 4; + end = hdr + table->hdrsz; + if (tableSize + && read_pointer(&ptr, end, hdr[1]) + == (unsigned long)table->address + && (i = read_pointer(&ptr, end, hdr[2])) > 0 + && i == (end - ptr) / (2 * tableSize) + && !((end - ptr) % (2 * tableSize))) { + do { + const u8 *cur = ptr + (i / 2) * (2 * tableSize); + + startLoc = read_pointer(&cur, + cur + tableSize, + hdr[3]); + if (pc < startLoc) + i /= 2; + else { + ptr = cur - tableSize; + i = (i + 1) / 2; + } + } while (startLoc && i > 1); + if (i == 1 + && (startLoc = read_pointer(&ptr, + ptr + tableSize, + hdr[3])) != 0 + && pc >= startLoc) + fde = (void *)read_pointer(&ptr, + ptr + tableSize, + hdr[3]); } + } + + if (fde != NULL) { + cie = cie_for_fde(fde, table); ptr = (const u8 *)(fde + 2); - startLoc = read_pointer(&ptr, - (const u8 *)(fde + 1) + *fde, - ptrType); - endLoc = startLoc - + read_pointer(&ptr, - (const u8 *)(fde + 1) + *fde, - ptrType & DW_EH_PE_indirect - ? ptrType - : ptrType & (DW_EH_PE_FORM|DW_EH_PE_signed)); - if (pc >= startLoc && pc < endLoc) - break; - cie = NULL; + if(cie != NULL + && cie != &bad_cie + && cie != ¬_fde + && (ptrType = fde_pointer_type(cie)) >= 0 + && read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType) == startLoc) { + if (!(ptrType & DW_EH_PE_indirect)) + ptrType &= DW_EH_PE_FORM|DW_EH_PE_signed; + endLoc = startLoc + + read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if(pc >= endLoc) + fde = NULL; + } else + fde = NULL; + } + if (fde == NULL) { + for (fde = table->address, tableSize = table->size; + cie = NULL, tableSize > sizeof(*fde) + && tableSize - sizeof(*fde) >= *fde; + tableSize -= sizeof(*fde) + *fde, + fde += 1 + *fde / sizeof(*fde)) { + cie = cie_for_fde(fde, table); + if (cie == &bad_cie) { + cie = NULL; + break; + } + if (cie == NULL + || cie == ¬_fde + || (ptrType = fde_pointer_type(cie)) < 0) + continue; + ptr = (const u8 *)(fde + 2); + startLoc = read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if (!startLoc) + continue; + if (!(ptrType & DW_EH_PE_indirect)) + ptrType &= DW_EH_PE_FORM|DW_EH_PE_signed; + endLoc = startLoc + + read_pointer(&ptr, + (const u8 *)(fde + 1) + *fde, + ptrType); + if (pc >= startLoc && pc < endLoc) + break; + } } } if (cie != NULL) { -- cgit v1.2.3 From a1bae67243512ca30ceda48e3e24e25b543f8ab7 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 21 Oct 2006 18:37:02 +0200 Subject: [PATCH] i386: Disable nmi watchdog on all ThinkPads Even newer Thinkpads have bugs in SMM code that causes hangs with NMI watchdog. Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 10 +++++----- drivers/firmware/dmi_scan.c | 20 ++++++++++++++++++++ include/linux/dmi.h | 2 ++ 3 files changed, 27 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 3e8e3adb0489..eaafe233a5da 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -219,11 +219,11 @@ static int __init check_nmi_watchdog(void) int cpu; /* Enable NMI watchdog for newer systems. - Actually it should be safe for most systems before 2004 too except - for some IBM systems that corrupt registers when NMI happens - during SMM. Unfortunately we don't have more exact information - on these and use this coarse check. */ - if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004) + Probably safe on most older systems too, but let's be careful. + IBM ThinkPads use INT10 inside SMM and that allows early NMI inside SMM + which hangs the system. Disable watchdog for all thinkpads */ + if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004 && + !dmi_name_in_vendors("ThinkPad")) nmi_watchdog = NMI_LOCAL_APIC; if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT)) diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c index b8b596d5778d..37deee6c0c1c 100644 --- a/drivers/firmware/dmi_scan.c +++ b/drivers/firmware/dmi_scan.c @@ -326,6 +326,26 @@ char *dmi_get_system_info(int field) } EXPORT_SYMBOL(dmi_get_system_info); + +/** + * dmi_name_in_vendors - Check if string is anywhere in the DMI vendor information. + * @str: Case sensitive Name + */ +int dmi_name_in_vendors(char *str) +{ + static int fields[] = { DMI_BIOS_VENDOR, DMI_BIOS_VERSION, DMI_SYS_VENDOR, + DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, DMI_BOARD_VENDOR, + DMI_BOARD_NAME, DMI_BOARD_VERSION, DMI_NONE }; + int i; + for (i = 0; fields[i] != DMI_NONE; i++) { + int f = fields[i]; + if (dmi_ident[f] && strstr(dmi_ident[f], str)) + return 1; + } + return 0; +} +EXPORT_SYMBOL(dmi_name_in_vendors); + /** * dmi_find_device - find onboard device by type/name * @type: device type or %DMI_DEV_TYPE_ANY to match all device types diff --git a/include/linux/dmi.h b/include/linux/dmi.h index 38dc403be70b..904bf3d2d90b 100644 --- a/include/linux/dmi.h +++ b/include/linux/dmi.h @@ -69,6 +69,7 @@ extern struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from); extern void dmi_scan_machine(void); extern int dmi_get_year(int field); +extern int dmi_name_in_vendors(char *str); #else @@ -77,6 +78,7 @@ static inline char * dmi_get_system_info(int field) { return NULL; } static inline struct dmi_device * dmi_find_device(int type, const char *name, struct dmi_device *from) { return NULL; } static inline int dmi_get_year(int year) { return 0; } +static inline int dmi_name_in_vendors(char *s) { return 0; } #endif -- cgit v1.2.3 From 3343571d9f88a0de542d33aea9ab881f00ff866d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 19 Oct 2006 14:44:53 +0900 Subject: [PATCH] libata: typo fix Typo fix in commment. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/libata.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/libata.h b/include/linux/libata.h index d0a7ad5ed518..b03d5a340dc8 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -143,7 +143,7 @@ enum { ATA_DFLAG_CFG_MASK = (1 << 8) - 1, ATA_DFLAG_PIO = (1 << 8), /* device limited to PIO mode */ - ATA_DFLAG_NCQ_OFF = (1 << 9), /* devied limited to non-NCQ mode */ + ATA_DFLAG_NCQ_OFF = (1 << 9), /* device limited to non-NCQ mode */ ATA_DFLAG_SUSPENDED = (1 << 10), /* device suspended */ ATA_DFLAG_INIT_MASK = (1 << 16) - 1, -- cgit v1.2.3 From da3ed32fe568148ede256975d40825ffcdac767b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:08 -0700 Subject: [PATCH] md: add another COMPAT_IOCTL for md .. so that you can use bitmaps with 32bit userspace on a 64 bit kernel. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat_ioctl.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h index cfdb4f6a89d4..c26c3adcfacf 100644 --- a/include/linux/compat_ioctl.h +++ b/include/linux/compat_ioctl.h @@ -131,6 +131,7 @@ COMPATIBLE_IOCTL(RUN_ARRAY) COMPATIBLE_IOCTL(STOP_ARRAY) COMPATIBLE_IOCTL(STOP_ARRAY_RO) COMPATIBLE_IOCTL(RESTART_ARRAY_RW) +COMPATIBLE_IOCTL(GET_BITMAP_FILE) ULONG_IOCTL(SET_BITMAP_FILE) /* DM */ COMPATIBLE_IOCTL(DM_VERSION_32) -- cgit v1.2.3 From 1c05b4bc22cd640d3a534bd2851a8413d5df3709 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:08 -0700 Subject: [PATCH] md: endian annotation for v1 superblock access Includes a couple of bugfixes found by sparse. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 13 ++++++----- include/linux/raid/md_p.h | 56 +++++++++++++++++++++++------------------------ 2 files changed, 35 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index f7f19088f3be..7daa7b1e145f 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -974,12 +974,13 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev) * version 1 superblock */ -static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb) +static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb) { - unsigned int disk_csum, csum; + __le32 disk_csum; + u32 csum; unsigned long long newcsum; int size = 256 + le32_to_cpu(sb->max_dev)*2; - unsigned int *isuper = (unsigned int*)sb; + __le32 *isuper = (__le32*)sb; int i; disk_csum = sb->sb_csum; @@ -989,7 +990,7 @@ static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb) newcsum += le32_to_cpu(*isuper++); if (size == 2) - newcsum += le16_to_cpu(*(unsigned short*) isuper); + newcsum += le16_to_cpu(*(__le16*) isuper); csum = (newcsum & 0xffffffff) + (newcsum >> 32); sb->sb_csum = disk_csum; @@ -1106,7 +1107,7 @@ static int super_1_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version) if (le32_to_cpu(sb->chunksize)) rdev->size &= ~((sector_t)le32_to_cpu(sb->chunksize)/2 - 1); - if (le32_to_cpu(sb->size) > rdev->size*2) + if (le64_to_cpu(sb->size) > rdev->size*2) return -EINVAL; return ret; } @@ -1228,7 +1229,7 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev) else sb->resync_offset = cpu_to_le64(0); - sb->cnt_corrected_read = atomic_read(&rdev->corrected_errors); + sb->cnt_corrected_read = cpu_to_le32(atomic_read(&rdev->corrected_errors)); sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->size<<1); diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h index b6ebc69bae54..3f2cd98c508b 100644 --- a/include/linux/raid/md_p.h +++ b/include/linux/raid/md_p.h @@ -206,52 +206,52 @@ static inline __u64 md_event(mdp_super_t *sb) { */ struct mdp_superblock_1 { /* constant array information - 128 bytes */ - __u32 magic; /* MD_SB_MAGIC: 0xa92b4efc - little endian */ - __u32 major_version; /* 1 */ - __u32 feature_map; /* bit 0 set if 'bitmap_offset' is meaningful */ - __u32 pad0; /* always set to 0 when writing */ + __le32 magic; /* MD_SB_MAGIC: 0xa92b4efc - little endian */ + __le32 major_version; /* 1 */ + __le32 feature_map; /* bit 0 set if 'bitmap_offset' is meaningful */ + __le32 pad0; /* always set to 0 when writing */ __u8 set_uuid[16]; /* user-space generated. */ char set_name[32]; /* set and interpreted by user-space */ - __u64 ctime; /* lo 40 bits are seconds, top 24 are microseconds or 0*/ - __u32 level; /* -4 (multipath), -1 (linear), 0,1,4,5 */ - __u32 layout; /* only for raid5 and raid10 currently */ - __u64 size; /* used size of component devices, in 512byte sectors */ + __le64 ctime; /* lo 40 bits are seconds, top 24 are microseconds or 0*/ + __le32 level; /* -4 (multipath), -1 (linear), 0,1,4,5 */ + __le32 layout; /* only for raid5 and raid10 currently */ + __le64 size; /* used size of component devices, in 512byte sectors */ - __u32 chunksize; /* in 512byte sectors */ - __u32 raid_disks; - __u32 bitmap_offset; /* sectors after start of superblock that bitmap starts + __le32 chunksize; /* in 512byte sectors */ + __le32 raid_disks; + __le32 bitmap_offset; /* sectors after start of superblock that bitmap starts * NOTE: signed, so bitmap can be before superblock * only meaningful of feature_map[0] is set. */ /* These are only valid with feature bit '4' */ - __u32 new_level; /* new level we are reshaping to */ - __u64 reshape_position; /* next address in array-space for reshape */ - __u32 delta_disks; /* change in number of raid_disks */ - __u32 new_layout; /* new layout */ - __u32 new_chunk; /* new chunk size (bytes) */ + __le32 new_level; /* new level we are reshaping to */ + __le64 reshape_position; /* next address in array-space for reshape */ + __le32 delta_disks; /* change in number of raid_disks */ + __le32 new_layout; /* new layout */ + __le32 new_chunk; /* new chunk size (bytes) */ __u8 pad1[128-124]; /* set to 0 when written */ /* constant this-device information - 64 bytes */ - __u64 data_offset; /* sector start of data, often 0 */ - __u64 data_size; /* sectors in this device that can be used for data */ - __u64 super_offset; /* sector start of this superblock */ - __u64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */ - __u32 dev_number; /* permanent identifier of this device - not role in raid */ - __u32 cnt_corrected_read; /* number of read errors that were corrected by re-writing */ + __le64 data_offset; /* sector start of data, often 0 */ + __le64 data_size; /* sectors in this device that can be used for data */ + __le64 super_offset; /* sector start of this superblock */ + __le64 recovery_offset;/* sectors before this offset (from data_offset) have been recovered */ + __le32 dev_number; /* permanent identifier of this device - not role in raid */ + __le32 cnt_corrected_read; /* number of read errors that were corrected by re-writing */ __u8 device_uuid[16]; /* user-space setable, ignored by kernel */ __u8 devflags; /* per-device flags. Only one defined...*/ #define WriteMostly1 1 /* mask for writemostly flag in above */ __u8 pad2[64-57]; /* set to 0 when writing */ /* array state information - 64 bytes */ - __u64 utime; /* 40 bits second, 24 btes microseconds */ - __u64 events; /* incremented when superblock updated */ - __u64 resync_offset; /* data before this offset (from data_offset) known to be in sync */ - __u32 sb_csum; /* checksum upto devs[max_dev] */ - __u32 max_dev; /* size of devs[] array to consider */ + __le64 utime; /* 40 bits second, 24 btes microseconds */ + __le64 events; /* incremented when superblock updated */ + __le64 resync_offset; /* data before this offset (from data_offset) known to be in sync */ + __le32 sb_csum; /* checksum upto devs[max_dev] */ + __le32 max_dev; /* size of devs[] array to consider */ __u8 pad3[64-32]; /* set to 0 when writing */ /* device state information. Indexed by dev_number. @@ -260,7 +260,7 @@ struct mdp_superblock_1 { * into the 'roles' value. If a device is spare or faulty, then it doesn't * have a meaningful role. */ - __u16 dev_roles[0]; /* role in array, or 0xffff for a spare, or 0xfffe for faulty */ + __le16 dev_roles[0]; /* role in array, or 0xffff for a spare, or 0xfffe for faulty */ }; /* feature_map bits */ -- cgit v1.2.3 From 4f2e639af4bd5e152fc79256e333643d3dd6c10f Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Sat, 21 Oct 2006 10:24:09 -0700 Subject: [PATCH] md: endian annotations for the bitmap superblock And a couple of bug fixes found by sparse. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/bitmap.c | 10 +++++----- include/linux/raid/bitmap.h | 20 ++++++++++---------- 2 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index d47d38ac71b1..d6f614738bbd 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -536,7 +536,7 @@ static int bitmap_read_sb(struct bitmap *bitmap) printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) " "-- forcing full recovery\n", bmname(bitmap), events, (unsigned long long) bitmap->mddev->events); - sb->state |= BITMAP_STALE; + sb->state |= cpu_to_le32(BITMAP_STALE); } success: /* assign fields using values from superblock */ @@ -544,11 +544,11 @@ success: bitmap->daemon_sleep = daemon_sleep; bitmap->daemon_lastrun = jiffies; bitmap->max_write_behind = write_behind; - bitmap->flags |= sb->state; + bitmap->flags |= le32_to_cpu(sb->state); if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN) bitmap->flags |= BITMAP_HOSTENDIAN; bitmap->events_cleared = le64_to_cpu(sb->events_cleared); - if (sb->state & BITMAP_STALE) + if (sb->state & cpu_to_le32(BITMAP_STALE)) bitmap->events_cleared = bitmap->mddev->events; err = 0; out: @@ -578,9 +578,9 @@ static void bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits, spin_unlock_irqrestore(&bitmap->lock, flags); sb = (bitmap_super_t *)kmap_atomic(bitmap->sb_page, KM_USER0); switch (op) { - case MASK_SET: sb->state |= bits; + case MASK_SET: sb->state |= cpu_to_le32(bits); break; - case MASK_UNSET: sb->state &= ~bits; + case MASK_UNSET: sb->state &= cpu_to_le32(~bits); break; default: BUG(); } diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index 84d887751855..ebd42a3710b4 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -146,16 +146,16 @@ enum bitmap_state { /* the superblock at the front of the bitmap file -- little endian */ typedef struct bitmap_super_s { - __u32 magic; /* 0 BITMAP_MAGIC */ - __u32 version; /* 4 the bitmap major for now, could change... */ - __u8 uuid[16]; /* 8 128 bit uuid - must match md device uuid */ - __u64 events; /* 24 event counter for the bitmap (1)*/ - __u64 events_cleared;/*32 event counter when last bit cleared (2) */ - __u64 sync_size; /* 40 the size of the md device's sync range(3) */ - __u32 state; /* 48 bitmap state information */ - __u32 chunksize; /* 52 the bitmap chunk size in bytes */ - __u32 daemon_sleep; /* 56 seconds between disk flushes */ - __u32 write_behind; /* 60 number of outstanding write-behind writes */ + __le32 magic; /* 0 BITMAP_MAGIC */ + __le32 version; /* 4 the bitmap major for now, could change... */ + __u8 uuid[16]; /* 8 128 bit uuid - must match md device uuid */ + __le64 events; /* 24 event counter for the bitmap (1)*/ + __le64 events_cleared;/*32 event counter when last bit cleared (2) */ + __le64 sync_size; /* 40 the size of the md device's sync range(3) */ + __le32 state; /* 48 bitmap state information */ + __le32 chunksize; /* 52 the bitmap chunk size in bytes */ + __le32 daemon_sleep; /* 56 seconds between disk flushes */ + __le32 write_behind; /* 60 number of outstanding write-behind writes */ __u8 pad[256 - 64]; /* set to zero */ } bitmap_super_t; -- cgit v1.2.3 From d42552c3ace1fa1f16ae02ce642f4c733cec40ca Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 21 Oct 2006 10:24:12 -0700 Subject: [PATCH] pci: declare pci_get_device_reverse() We seem to have lost the declaration of pci_get_device_reverse(), if we ever had one. Add a CONFIG_PCI=0 stub too. Acked-by: Alan Cox Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci.h | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 4689e2a699c0..09be0f81b27b 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -455,7 +455,11 @@ int pci_find_next_capability (struct pci_dev *dev, u8 pos, int cap); int pci_find_ext_capability (struct pci_dev *dev, int cap); struct pci_bus *pci_find_next_bus(const struct pci_bus *from); -struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from); +struct pci_dev *pci_get_device(unsigned int vendor, unsigned int device, + struct pci_dev *from); +struct pci_dev *pci_get_device_reverse(unsigned int vendor, unsigned int device, + struct pci_dev *from); + struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, unsigned int ss_vendor, unsigned int ss_device, struct pci_dev *from); @@ -660,7 +664,12 @@ static inline struct pci_dev *pci_find_device(unsigned int vendor, unsigned int static inline struct pci_dev *pci_find_slot(unsigned int bus, unsigned int devfn) { return NULL; } -static inline struct pci_dev *pci_get_device (unsigned int vendor, unsigned int device, struct pci_dev *from) +static inline struct pci_dev *pci_get_device(unsigned int vendor, + unsigned int device, struct pci_dev *from) +{ return NULL; } + +static inline struct pci_dev *pci_get_device_reverse(unsigned int vendor, + unsigned int device, struct pci_dev *from) { return NULL; } static inline struct pci_dev *pci_get_subsys (unsigned int vendor, unsigned int device, -- cgit v1.2.3 From 7516795739bd53175629b90fab0ad488d7a6a9f7 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Sat, 21 Oct 2006 10:24:14 -0700 Subject: [PATCH] Reintroduce NODES_SPAN_OTHER_NODES for powerpc Reintroduce NODES_SPAN_OTHER_NODES for powerpc Revert "[PATCH] Remove SPAN_OTHER_NODES config definition" This reverts commit f62859bb6871c5e4a8e591c60befc8caaf54db8c. Revert "[PATCH] mm: remove arch independent NODES_SPAN_OTHER_NODES" This reverts commit a94b3ab7eab4edcc9b2cb474b188f774c331adf7. Also update the comments to indicate that this is still required and where its used. Signed-off-by: Andy Whitcroft Cc: Paul Mackerras Cc: Mike Kravetz Cc: Benjamin Herrenschmidt Acked-by: Mel Gorman Acked-by: Will Schmidt Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/Kconfig | 9 +++++++++ arch/powerpc/configs/pseries_defconfig | 1 + include/linux/mmzone.h | 6 ++++++ mm/page_alloc.c | 2 ++ 4 files changed, 18 insertions(+) (limited to 'include/linux') diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 8b6910465578..2bd9b7fb0f6c 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -751,6 +751,15 @@ config ARCH_MEMORY_PROBE def_bool y depends on MEMORY_HOTPLUG +# Some NUMA nodes have memory ranges that span +# other nodes. Even though a pfn is valid and +# between a node's start and end pfns, it may not +# reside on that node. See memmap_init_zone() +# for details. +config NODES_SPAN_OTHER_NODES + def_bool y + depends on NEED_MULTIPLE_NODES + config PPC_64K_PAGES bool "64k page size" depends on PPC64 diff --git a/arch/powerpc/configs/pseries_defconfig b/arch/powerpc/configs/pseries_defconfig index 9828663652e9..d2833c1a1f3d 100644 --- a/arch/powerpc/configs/pseries_defconfig +++ b/arch/powerpc/configs/pseries_defconfig @@ -184,6 +184,7 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 CONFIG_MIGRATION=y CONFIG_RESOURCES_64BIT=y CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y +CONFIG_NODES_SPAN_OTHER_NODES=y # CONFIG_PPC_64K_PAGES is not set CONFIG_SCHED_SMT=y CONFIG_PROC_DEVICETREE=y diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 59855b8718a0..ed0762b283a9 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -674,6 +674,12 @@ void sparse_init(void); #define sparse_index_init(_sec, _nid) do {} while (0) #endif /* CONFIG_SPARSEMEM */ +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +#else +#define early_pfn_in_nid(pfn, nid) (1) +#endif + #ifndef early_pfn_valid #define early_pfn_valid(pfn) (1) #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ebd425c2e2a7..f5fc45472d5c 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1689,6 +1689,8 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, for (pfn = start_pfn; pfn < end_pfn; pfn++) { if (!early_pfn_valid(pfn)) continue; + if (!early_pfn_in_nid(pfn, nid)) + continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); init_page_count(page); -- cgit v1.2.3 From faf6bbcf94caee10ba34adb86db4ecca96bfd3bf Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Sat, 21 Oct 2006 10:24:17 -0700 Subject: [PATCH] cpuset: mempolicy migration typo fix Mistyped an ifdef CONFIG_CPUSETS - fixed. I doubt that anyone ever noticed. The impact of this typo was that if someone: 1) was using MPOL_BIND to force off node allocations 2) while using cpusets to constrain memory placement 3) when that cpuset was migrating that jobs memory 4) while the tasks in that job were actively forking then there was a rare chance that future allocations using that MPOL_BIND policy would be node local, not off node. Signed-off-by: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mempolicy.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h index 09f0f575ddff..daabb3aa1ec6 100644 --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -150,7 +150,7 @@ extern void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new); extern void mpol_fix_fork_child_flag(struct task_struct *p); #define set_cpuset_being_rebound(x) (cpuset_being_rebound = (x)) -#ifdef CONFIG_CPUSET +#ifdef CONFIG_CPUSETS #define current_cpuset_is_being_rebound() \ (cpuset_being_rebound == current->cpuset) #else -- cgit v1.2.3 From 04fed361dadb7921507a470947ac23d2f26352cf Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 22 Oct 2006 15:57:18 +0100 Subject: [PATCH] Remove __must_check for device_for_each_child() Eliminate more __must_check madness. The return code from device_for_each_child() depends on the values which the helper function returns. If the helper function always returns zero, it's utterly pointless to check the return code from device_for_each_child(). The only code which knows if the return value should be checked is the caller itself, so forcing the return code to always be checked is silly. Hence, remove the __must_check annotation. Signed-off-by: Russell King Signed-off-by: Linus Torvalds --- include/linux/device.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/device.h b/include/linux/device.h index 662e6a10144e..9d4f6a963936 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -393,7 +393,7 @@ extern void device_unregister(struct device * dev); extern void device_initialize(struct device * dev); extern int __must_check device_add(struct device * dev); extern void device_del(struct device * dev); -extern int __must_check device_for_each_child(struct device *, void *, +extern int device_for_each_child(struct device *, void *, int (*fn)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); -- cgit v1.2.3 From 0344c6c5387ba335bba5a66fd44714b94c98573f Mon Sep 17 00:00:00 2001 From: Christian Krafft Date: Tue, 24 Oct 2006 18:31:24 +0200 Subject: [POWERPC] sysfs: add support for adding/removing spu sysfs attributes This patch adds two functions to create and remove sysfs attributes and attribute_group to all cpus. That allows to register sysfs attributes in a subdirectory like: /sys/devices/system/cpu/cpuX/group_name/what_ever This will be used by cbe_thermal to group all attributes dealing with thermal support in one directory. Signed-of-by: Christian Krafft Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/sysfs.c | 66 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/cpu.h | 8 ++++++ 2 files changed, 74 insertions(+) (limited to 'include/linux') diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index d45a168bdaca..cda21a27490a 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -299,6 +299,72 @@ static struct notifier_block __cpuinitdata sysfs_cpu_nb = { .notifier_call = sysfs_cpu_notify, }; +static DEFINE_MUTEX(cpu_mutex); + +int cpu_add_sysdev_attr(struct sysdev_attribute *attr) +{ + int cpu; + + mutex_lock(&cpu_mutex); + + for_each_possible_cpu(cpu) { + sysdev_create_file(get_cpu_sysdev(cpu), attr); + } + + mutex_unlock(&cpu_mutex); + return 0; +} +EXPORT_SYMBOL_GPL(cpu_add_sysdev_attr); + +int cpu_add_sysdev_attr_group(struct attribute_group *attrs) +{ + int cpu; + struct sys_device *sysdev; + + mutex_lock(&cpu_mutex); + + for_each_possible_cpu(cpu) { + sysdev = get_cpu_sysdev(cpu); + sysfs_create_group(&sysdev->kobj, attrs); + } + + mutex_unlock(&cpu_mutex); + return 0; +} +EXPORT_SYMBOL_GPL(cpu_add_sysdev_attr_group); + + +void cpu_remove_sysdev_attr(struct sysdev_attribute *attr) +{ + int cpu; + + mutex_lock(&cpu_mutex); + + for_each_possible_cpu(cpu) { + sysdev_remove_file(get_cpu_sysdev(cpu), attr); + } + + mutex_unlock(&cpu_mutex); +} +EXPORT_SYMBOL_GPL(cpu_remove_sysdev_attr); + +void cpu_remove_sysdev_attr_group(struct attribute_group *attrs) +{ + int cpu; + struct sys_device *sysdev; + + mutex_lock(&cpu_mutex); + + for_each_possible_cpu(cpu) { + sysdev = get_cpu_sysdev(cpu); + sysfs_remove_group(&sysdev->kobj, attrs); + } + + mutex_unlock(&cpu_mutex); +} +EXPORT_SYMBOL_GPL(cpu_remove_sysdev_attr_group); + + /* NUMA stuff */ #ifdef CONFIG_NUMA diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 3fef7d67aedc..f02d71bf6894 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -33,6 +33,14 @@ struct cpu { extern int register_cpu(struct cpu *cpu, int num); extern struct sys_device *get_cpu_sysdev(unsigned cpu); + +extern int cpu_add_sysdev_attr(struct sysdev_attribute *attr); +extern void cpu_remove_sysdev_attr(struct sysdev_attribute *attr); + +extern int cpu_add_sysdev_attr_group(struct attribute_group *attrs); +extern void cpu_remove_sysdev_attr_group(struct attribute_group *attrs); + + #ifdef CONFIG_HOTPLUG_CPU extern void unregister_cpu(struct cpu *cpu); #endif -- cgit v1.2.3 From 735a7ffb739b6efeaeb1e720306ba308eaaeb20e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Fri, 27 Oct 2006 11:42:37 -0700 Subject: [PATCH] drivers: wait for threaded probes between initcall levels The multithreaded-probing code has a problem: after one initcall level (eg, core_initcall) has been processed, we will then start processing the next level (postcore_initcall) while the kernel threads which are handling core_initcall are still executing. This breaks the guarantees which the layered initcalls previously gave us. IOW, we want to be multithreaded _within_ an initcall level, but not between different levels. Fix that up by causing the probing code to wait for all outstanding probes at one level to complete before we start processing the next level. Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/dd.c | 30 ++++++++++++++++++++++++++++++ include/asm-generic/vmlinux.lds.h | 9 ++++++++- include/linux/init.h | 28 +++++++++++++++++++--------- 3 files changed, 57 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index db01b95a47a5..c5d6bb4290ad 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -70,6 +71,8 @@ struct stupid_thread_structure { }; static atomic_t probe_count = ATOMIC_INIT(0); +static DECLARE_WAIT_QUEUE_HEAD(probe_waitqueue); + static int really_probe(void *void_data) { struct stupid_thread_structure *data = void_data; @@ -121,6 +124,7 @@ probe_failed: done: kfree(data); atomic_dec(&probe_count); + wake_up(&probe_waitqueue); return ret; } @@ -337,6 +341,32 @@ void driver_detach(struct device_driver * drv) } } +#ifdef CONFIG_PCI_MULTITHREAD_PROBE +static int __init wait_for_probes(void) +{ + DEFINE_WAIT(wait); + + printk(KERN_INFO "%s: waiting for %d threads\n", __FUNCTION__, + atomic_read(&probe_count)); + if (!atomic_read(&probe_count)) + return 0; + while (atomic_read(&probe_count)) { + prepare_to_wait(&probe_waitqueue, &wait, TASK_UNINTERRUPTIBLE); + if (atomic_read(&probe_count)) + schedule(); + } + finish_wait(&probe_waitqueue, &wait); + return 0; +} + +core_initcall_sync(wait_for_probes); +postcore_initcall_sync(wait_for_probes); +arch_initcall_sync(wait_for_probes); +subsys_initcall_sync(wait_for_probes); +fs_initcall_sync(wait_for_probes); +device_initcall_sync(wait_for_probes); +late_initcall_sync(wait_for_probes); +#endif EXPORT_SYMBOL_GPL(device_bind_driver); EXPORT_SYMBOL_GPL(device_release_driver); diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index e3e83bcaf710..9d873163a7ab 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -216,10 +216,17 @@ #define INITCALLS \ *(.initcall1.init) \ + *(.initcall1s.init) \ *(.initcall2.init) \ + *(.initcall2s.init) \ *(.initcall3.init) \ + *(.initcall3s.init) \ *(.initcall4.init) \ + *(.initcall4s.init) \ *(.initcall5.init) \ + *(.initcall5s.init) \ *(.initcall6.init) \ - *(.initcall7.init) + *(.initcall6s.init) \ + *(.initcall7.init) \ + *(.initcall7s.init) diff --git a/include/linux/init.h b/include/linux/init.h index e92b1455d7af..ff40ea118e3a 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -84,19 +84,29 @@ extern void setup_arch(char **); * by link order. * For backwards compatibility, initcall() puts the call in * the device init subsection. + * + * The `id' arg to __define_initcall() is needed so that multiple initcalls + * can point at the same handler without causing duplicate-symbol build errors. */ -#define __define_initcall(level,fn) \ - static initcall_t __initcall_##fn __attribute_used__ \ +#define __define_initcall(level,fn,id) \ + static initcall_t __initcall_##fn##id __attribute_used__ \ __attribute__((__section__(".initcall" level ".init"))) = fn -#define core_initcall(fn) __define_initcall("1",fn) -#define postcore_initcall(fn) __define_initcall("2",fn) -#define arch_initcall(fn) __define_initcall("3",fn) -#define subsys_initcall(fn) __define_initcall("4",fn) -#define fs_initcall(fn) __define_initcall("5",fn) -#define device_initcall(fn) __define_initcall("6",fn) -#define late_initcall(fn) __define_initcall("7",fn) +#define core_initcall(fn) __define_initcall("1",fn,1) +#define core_initcall_sync(fn) __define_initcall("1s",fn,1s) +#define postcore_initcall(fn) __define_initcall("2",fn,2) +#define postcore_initcall_sync(fn) __define_initcall("2s",fn,2s) +#define arch_initcall(fn) __define_initcall("3",fn,3) +#define arch_initcall_sync(fn) __define_initcall("3s",fn,3s) +#define subsys_initcall(fn) __define_initcall("4",fn,4) +#define subsys_initcall_sync(fn) __define_initcall("4s",fn,4s) +#define fs_initcall(fn) __define_initcall("5",fn,5) +#define fs_initcall_sync(fn) __define_initcall("5s",fn,5s) +#define device_initcall(fn) __define_initcall("6",fn,6) +#define device_initcall_sync(fn) __define_initcall("6s",fn,6s) +#define late_initcall(fn) __define_initcall("7",fn,7) +#define late_initcall_sync(fn) __define_initcall("7s",fn,7s) #define __initcall(fn) device_initcall(fn) -- cgit v1.2.3 From 2ae88149a27cadf2840e0ab8155bef13be285c03 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 28 Oct 2006 10:38:23 -0700 Subject: [PATCH] mm: clean up pagecache allocation - Consolidate page_cache_alloc - Fix splice: only the pagecache pages and filesystem data need to use mapping_gfp_mask. - Fix grab_cache_page_nowait: same as splice, also honour NUMA placement. Signed-off-by: Nick Piggin Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/splice.c | 9 ++++----- include/linux/pagemap.h | 14 +++++++++----- mm/filemap.c | 24 ++++++------------------ 3 files changed, 19 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/fs/splice.c b/fs/splice.c index 49fb9f129938..8d705954d294 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -74,7 +74,7 @@ static int page_cache_pipe_buf_steal(struct pipe_inode_info *pipe, wait_on_page_writeback(page); if (PagePrivate(page)) - try_to_release_page(page, mapping_gfp_mask(mapping)); + try_to_release_page(page, GFP_KERNEL); /* * If we succeeded in removing the mapping, set LRU flag @@ -333,7 +333,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, break; error = add_to_page_cache_lru(page, mapping, index, - mapping_gfp_mask(mapping)); + GFP_KERNEL); if (unlikely(error)) { page_cache_release(page); if (error == -EEXIST) @@ -557,7 +557,6 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, { struct file *file = sd->file; struct address_space *mapping = file->f_mapping; - gfp_t gfp_mask = mapping_gfp_mask(mapping); unsigned int offset, this_len; struct page *page; pgoff_t index; @@ -591,7 +590,7 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, goto find_page; page = buf->page; - if (add_to_page_cache(page, mapping, index, gfp_mask)) { + if (add_to_page_cache(page, mapping, index, GFP_KERNEL)) { unlock_page(page); goto find_page; } @@ -613,7 +612,7 @@ find_page: * This will also lock the page */ ret = add_to_page_cache_lru(page, mapping, index, - gfp_mask); + GFP_KERNEL); if (unlikely(ret)) goto out; } diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 64f950925151..c3e255bf8594 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -52,19 +52,23 @@ static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask) void release_pages(struct page **pages, int nr, int cold); #ifdef CONFIG_NUMA -extern struct page *page_cache_alloc(struct address_space *x); -extern struct page *page_cache_alloc_cold(struct address_space *x); +extern struct page *__page_cache_alloc(gfp_t gfp); #else +static inline struct page *__page_cache_alloc(gfp_t gfp) +{ + return alloc_pages(gfp, 0); +} +#endif + static inline struct page *page_cache_alloc(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x), 0); + return __page_cache_alloc(mapping_gfp_mask(x)); } static inline struct page *page_cache_alloc_cold(struct address_space *x) { - return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0); + return __page_cache_alloc(mapping_gfp_mask(x)|__GFP_COLD); } -#endif typedef int filler_t(void *, struct page *); diff --git a/mm/filemap.c b/mm/filemap.c index cb26e33fd0ff..7b84dc814347 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -467,25 +467,15 @@ int add_to_page_cache_lru(struct page *page, struct address_space *mapping, } #ifdef CONFIG_NUMA -struct page *page_cache_alloc(struct address_space *x) +struct page *__page_cache_alloc(gfp_t gfp) { if (cpuset_do_page_mem_spread()) { int n = cpuset_mem_spread_node(); - return alloc_pages_node(n, mapping_gfp_mask(x), 0); + return alloc_pages_node(n, gfp, 0); } - return alloc_pages(mapping_gfp_mask(x), 0); + return alloc_pages(gfp, 0); } -EXPORT_SYMBOL(page_cache_alloc); - -struct page *page_cache_alloc_cold(struct address_space *x) -{ - if (cpuset_do_page_mem_spread()) { - int n = cpuset_mem_spread_node(); - return alloc_pages_node(n, mapping_gfp_mask(x)|__GFP_COLD, 0); - } - return alloc_pages(mapping_gfp_mask(x)|__GFP_COLD, 0); -} -EXPORT_SYMBOL(page_cache_alloc_cold); +EXPORT_SYMBOL(__page_cache_alloc); #endif static int __sleep_on_page_lock(void *word) @@ -826,7 +816,6 @@ struct page * grab_cache_page_nowait(struct address_space *mapping, unsigned long index) { struct page *page = find_get_page(mapping, index); - gfp_t gfp_mask; if (page) { if (!TestSetPageLocked(page)) @@ -834,9 +823,8 @@ grab_cache_page_nowait(struct address_space *mapping, unsigned long index) page_cache_release(page); return NULL; } - gfp_mask = mapping_gfp_mask(mapping) & ~__GFP_FS; - page = alloc_pages(gfp_mask, 0); - if (page && add_to_page_cache_lru(page, mapping, index, gfp_mask)) { + page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~__GFP_FS); + if (page && add_to_page_cache_lru(page, mapping, index, GFP_KERNEL)) { page_cache_release(page); page = NULL; } -- cgit v1.2.3 From 3bb1a852ab6c9cdf211a2f4a2f502340c8c38eca Mon Sep 17 00:00:00 2001 From: Martin Bligh Date: Sat, 28 Oct 2006 10:38:24 -0700 Subject: [PATCH] vmscan: Fix temp_priority race The temp_priority field in zone is racy, as we can walk through a reclaim path, and just before we copy it into prev_priority, it can be overwritten (say with DEF_PRIORITY) by another reclaimer. The same bug is contained in both try_to_free_pages and balance_pgdat, but it is fixed slightly differently. In balance_pgdat, we keep a separate priority record per zone in a local array. In try_to_free_pages there is no need to do this, as the priority level is the same for all zones that we reclaim from. Impact of this bug is that temp_priority is copied into prev_priority, and setting this artificially high causes reclaimers to set distress artificially low. They then fail to reclaim mapped pages, when they are, in fact, under severe memory pressure (their priority may be as low as 0). This causes the OOM killer to fire incorrectly. From: Andrew Morton __zone_reclaim() isn't modifying zone->prev_priority. But zone->prev_priority is used in the decision whether or not to bring mapped pages onto the inactive list. Hence there's a risk here that __zone_reclaim() will fail because zone->prev_priority ir large (ie: low urgency) and lots of mapped pages end up stuck on the active list. Fix that up by decreasing (ie making more urgent) zone->prev_priority as __zone_reclaim() scans the zone's pages. This bug perhaps explains why ZONE_RECLAIM_PRIORITY was created. It should be possible to remove that now, and to just start out at DEF_PRIORITY? Cc: Nick Piggin Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 6 +----- mm/page_alloc.c | 2 +- mm/vmscan.c | 55 +++++++++++++++++++++++++++++++++++++------------- mm/vmstat.c | 2 -- 4 files changed, 43 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index ed0762b283a9..e06683e2bea3 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -218,13 +218,9 @@ struct zone { * under - it drives the swappiness decision: whether to unmap mapped * pages. * - * temp_priority is used to remember the scanning priority at which - * this zone was successfully refilled to free_pages == pages_high. - * - * Access to both these fields is quite racy even on uniprocessor. But + * Access to both this field is quite racy even on uniprocessor. But * it is expected to average out OK. */ - int temp_priority; int prev_priority; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index f5fc45472d5c..ecf853b5e30e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2407,7 +2407,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat, zone->zone_pgdat = pgdat; zone->free_pages = 0; - zone->temp_priority = zone->prev_priority = DEF_PRIORITY; + zone->prev_priority = DEF_PRIORITY; zone_pcp_init(zone); INIT_LIST_HEAD(&zone->active_list); diff --git a/mm/vmscan.c b/mm/vmscan.c index f05527bf792b..b32560ead5c0 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -723,6 +723,20 @@ done: return nr_reclaimed; } +/* + * We are about to scan this zone at a certain priority level. If that priority + * level is smaller (ie: more urgent) than the previous priority, then note + * that priority level within the zone. This is done so that when the next + * process comes in to scan this zone, it will immediately start out at this + * priority level rather than having to build up its own scanning priority. + * Here, this priority affects only the reclaim-mapped threshold. + */ +static inline void note_zone_scanning_priority(struct zone *zone, int priority) +{ + if (priority < zone->prev_priority) + zone->prev_priority = priority; +} + static inline int zone_is_near_oom(struct zone *zone) { return zone->pages_scanned >= (zone->nr_active + zone->nr_inactive)*3; @@ -972,9 +986,7 @@ static unsigned long shrink_zones(int priority, struct zone **zones, if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->temp_priority = priority; - if (zone->prev_priority > priority) - zone->prev_priority = priority; + note_zone_scanning_priority(zone, priority); if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ @@ -1024,7 +1036,6 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->temp_priority = DEF_PRIORITY; lru_pages += zone->nr_active + zone->nr_inactive; } @@ -1065,13 +1076,22 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) if (!sc.all_unreclaimable) ret = 1; out: + /* + * Now that we've scanned all the zones at this priority level, note + * that level within the zone so that the next thread which performs + * scanning of this zone will immediately start out at this priority + * level. This affects only the decision whether or not to bring + * mapped pages onto the inactive list. + */ + if (priority < 0) + priority = 0; for (i = 0; zones[i] != 0; i++) { struct zone *zone = zones[i]; if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) continue; - zone->prev_priority = zone->temp_priority; + zone->prev_priority = priority; } return ret; } @@ -1111,6 +1131,11 @@ static unsigned long balance_pgdat(pg_data_t *pgdat, int order) .swap_cluster_max = SWAP_CLUSTER_MAX, .swappiness = vm_swappiness, }; + /* + * temp_priority is used to remember the scanning priority at which + * this zone was successfully refilled to free_pages == pages_high. + */ + int temp_priority[MAX_NR_ZONES]; loop_again: total_scanned = 0; @@ -1118,11 +1143,8 @@ loop_again: sc.may_writepage = !laptop_mode; count_vm_event(PAGEOUTRUN); - for (i = 0; i < pgdat->nr_zones; i++) { - struct zone *zone = pgdat->node_zones + i; - - zone->temp_priority = DEF_PRIORITY; - } + for (i = 0; i < pgdat->nr_zones; i++) + temp_priority[i] = DEF_PRIORITY; for (priority = DEF_PRIORITY; priority >= 0; priority--) { int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ @@ -1183,10 +1205,9 @@ scan: if (!zone_watermark_ok(zone, order, zone->pages_high, end_zone, 0)) all_zones_ok = 0; - zone->temp_priority = priority; - if (zone->prev_priority > priority) - zone->prev_priority = priority; + temp_priority[i] = priority; sc.nr_scanned = 0; + note_zone_scanning_priority(zone, priority); nr_reclaimed += shrink_zone(priority, zone, &sc); reclaim_state->reclaimed_slab = 0; nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL, @@ -1226,10 +1247,15 @@ scan: break; } out: + /* + * Note within each zone the priority level at which this zone was + * brought into a happy state. So that the next thread which scans this + * zone will start out at that priority level. + */ for (i = 0; i < pgdat->nr_zones; i++) { struct zone *zone = pgdat->node_zones + i; - zone->prev_priority = zone->temp_priority; + zone->prev_priority = temp_priority[i]; } if (!all_zones_ok) { cond_resched(); @@ -1614,6 +1640,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) */ priority = ZONE_RECLAIM_PRIORITY; do { + note_zone_scanning_priority(zone, priority); nr_reclaimed += shrink_zone(priority, zone, &sc); priority--; } while (priority >= 0 && nr_reclaimed < nr_pages); diff --git a/mm/vmstat.c b/mm/vmstat.c index 45b124e012f5..8614e8f6743b 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -587,11 +587,9 @@ static int zoneinfo_show(struct seq_file *m, void *arg) seq_printf(m, "\n all_unreclaimable: %u" "\n prev_priority: %i" - "\n temp_priority: %i" "\n start_pfn: %lu", zone->all_unreclaimable, zone->prev_priority, - zone->temp_priority, zone->zone_start_pfn); spin_unlock_irqrestore(&zone->lock, flags); seq_putc(m, '\n'); -- cgit v1.2.3 From 52fd24ca1db3a741f144bbc229beefe044202cac Mon Sep 17 00:00:00 2001 From: Giridhar Pemmasani Date: Sat, 28 Oct 2006 10:38:34 -0700 Subject: [PATCH] __vmalloc with GFP_ATOMIC causes 'sleeping from invalid context' If __vmalloc is called to allocate memory with GFP_ATOMIC in atomic context, the chain of calls results in __get_vm_area_node allocating memory for vm_struct with GFP_KERNEL, causing the 'sleeping from invalid context' warning. This patch fixes it by passing the gfp flags along so __get_vm_area_node allocates memory for vm_struct with the same flags. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmalloc.h | 3 ++- mm/vmalloc.c | 18 +++++++++++------- 2 files changed, 13 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index ce5f1482e6be..dc9a29d84abc 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -60,7 +60,8 @@ extern struct vm_struct *get_vm_area(unsigned long size, unsigned long flags); extern struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end); extern struct vm_struct *get_vm_area_node(unsigned long size, - unsigned long flags, int node); + unsigned long flags, int node, + gfp_t gfp_mask); extern struct vm_struct *remove_vm_area(void *addr); extern int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages); diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 1133dd3aafcf..6d381df7c9b3 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -160,13 +160,15 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages) return err; } -struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, - unsigned long start, unsigned long end, int node) +static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, + unsigned long start, unsigned long end, + int node, gfp_t gfp_mask) { struct vm_struct **p, *tmp, *area; unsigned long align = 1; unsigned long addr; + BUG_ON(in_interrupt()); if (flags & VM_IOREMAP) { int bit = fls(size); @@ -180,7 +182,7 @@ struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long flags, addr = ALIGN(start, align); size = PAGE_ALIGN(size); - area = kmalloc_node(sizeof(*area), GFP_KERNEL, node); + area = kmalloc_node(sizeof(*area), gfp_mask, node); if (unlikely(!area)) return NULL; @@ -236,7 +238,7 @@ out: struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags, unsigned long start, unsigned long end) { - return __get_vm_area_node(size, flags, start, end, -1); + return __get_vm_area_node(size, flags, start, end, -1, GFP_KERNEL); } /** @@ -253,9 +255,11 @@ struct vm_struct *get_vm_area(unsigned long size, unsigned long flags) return __get_vm_area(size, flags, VMALLOC_START, VMALLOC_END); } -struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, int node) +struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, + int node, gfp_t gfp_mask) { - return __get_vm_area_node(size, flags, VMALLOC_START, VMALLOC_END, node); + return __get_vm_area_node(size, flags, VMALLOC_START, VMALLOC_END, node, + gfp_mask); } /* Caller must hold vmlist_lock */ @@ -487,7 +491,7 @@ static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot, if (!size || (size >> PAGE_SHIFT) > num_physpages) return NULL; - area = get_vm_area_node(size, VM_ALLOC, node); + area = get_vm_area_node(size, VM_ALLOC, node, gfp_mask); if (!area) return NULL; -- cgit v1.2.3 From 5fa3839a64203b2ab727dcb37da9b2d7079fca28 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sat, 28 Oct 2006 10:38:46 -0700 Subject: [PATCH] Constify compat_get_bitmap argument This means we can call it when the bitmap we want to fetch is declared const. Signed-off-by: Stephen Rothwell Cc: Christoph Lameter Cc: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 2 +- kernel/compat.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compat.h b/include/linux/compat.h index f4ebf96f5308..f1553196826f 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -196,7 +196,7 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp, #define BITS_TO_COMPAT_LONGS(bits) \ (((bits)+BITS_PER_COMPAT_LONG-1)/BITS_PER_COMPAT_LONG) -long compat_get_bitmap(unsigned long *mask, compat_ulong_t __user *umask, +long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, unsigned long bitmap_size); long compat_put_bitmap(compat_ulong_t __user *umask, unsigned long *mask, unsigned long bitmap_size); diff --git a/kernel/compat.c b/kernel/compat.c index 75573e5d27b0..d4898aad6cfa 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -678,7 +678,7 @@ int get_compat_sigevent(struct sigevent *event, ? -EFAULT : 0; } -long compat_get_bitmap(unsigned long *mask, compat_ulong_t __user *umask, +long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, unsigned long bitmap_size) { int i, j; -- cgit v1.2.3 From 093a8e8aecd77b2799934996a55a6838e1e2b8f3 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:51 -0700 Subject: [PATCH] taskstats_tgid_free: fix usage taskstats_tgid_free() is called on copy_process's error path. This is wrong. IF (clone_flags & CLONE_THREAD) We should not clear ->signal->taskstats, current uses it, it probably has a valid accumulated info. ELSE taskstats_tgid_init() set ->signal->taskstats = NULL, there is nothing to free. Move the callsite to __exit_signal(). We don't need any locking, entire thread group is exiting, nobody should have a reference to soon to be released ->signal. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats_kern.h | 13 ++----------- kernel/exit.c | 1 + kernel/fork.c | 1 - 3 files changed, 3 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 16894b7edcc8..a437ca0d226b 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -49,17 +49,8 @@ static inline void taskstats_tgid_alloc(struct signal_struct *sig) static inline void taskstats_tgid_free(struct signal_struct *sig) { - struct taskstats *stats = NULL; - unsigned long flags; - - spin_lock_irqsave(&sig->stats_lock, flags); - if (sig->stats) { - stats = sig->stats; - sig->stats = NULL; - } - spin_unlock_irqrestore(&sig->stats_lock, flags); - if (stats) - kmem_cache_free(taskstats_cache, stats); + if (sig->stats) + kmem_cache_free(taskstats_cache, sig->stats); } extern void taskstats_exit_alloc(struct taskstats **, unsigned int *); diff --git a/kernel/exit.c b/kernel/exit.c index f250a5e3e281..06de6c4e8ca3 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -128,6 +128,7 @@ static void __exit_signal(struct task_struct *tsk) flush_sigqueue(&tsk->pending); if (sig) { flush_sigqueue(&sig->shared_pending); + taskstats_tgid_free(sig); __cleanup_signal(sig); } } diff --git a/kernel/fork.c b/kernel/fork.c index 29ebb30850ed..213326609bac 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -897,7 +897,6 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts void __cleanup_signal(struct signal_struct *sig) { exit_thread_group_keys(sig); - taskstats_tgid_free(sig); kmem_cache_free(signal_cachep, sig); } -- cgit v1.2.3 From 17b02695b254aa2ef0e53df9c8e6548f86e66a9d Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:52 -0700 Subject: [PATCH] taskstats_tgid_alloc: optimization Every subthread (except first) does unneeded kmem_cache_alloc/kmem_cache_free. Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats_kern.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index a437ca0d226b..664224008fb2 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -32,6 +32,9 @@ static inline void taskstats_tgid_alloc(struct signal_struct *sig) struct taskstats *stats; unsigned long flags; + if (sig->stats != NULL) + return; + stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); if (!stats) return; -- cgit v1.2.3 From b8534d7bd89df0cd41cd47bcd6733a05ea9a691a Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 28 Oct 2006 10:38:53 -0700 Subject: [PATCH] taskstats: kill ->taskstats_lock in favor of ->siglock signal_struct is (mostly) protected by ->sighand->siglock, I think we don't need ->taskstats_lock to protect ->stats. This also allows us to simplify the locking in fill_tgid(). Signed-off-by: Oleg Nesterov Cc: Shailabh Nagar Cc: Balbir Singh Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 - include/linux/taskstats_kern.h | 15 ++++++--------- kernel/fork.c | 2 +- kernel/taskstats.c | 16 ++++++---------- 4 files changed, 13 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6735c1cf334c..eafe4a7b8237 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -466,7 +466,6 @@ struct signal_struct { struct pacct_struct pacct; /* per-process accounting information */ #endif #ifdef CONFIG_TASKSTATS - spinlock_t stats_lock; struct taskstats *stats; #endif }; diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 664224008fb2..6562a2050a25 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -23,28 +23,26 @@ static inline void taskstats_exit_free(struct taskstats *tidstats) static inline void taskstats_tgid_init(struct signal_struct *sig) { - spin_lock_init(&sig->stats_lock); sig->stats = NULL; } -static inline void taskstats_tgid_alloc(struct signal_struct *sig) +static inline void taskstats_tgid_alloc(struct task_struct *tsk) { + struct signal_struct *sig = tsk->signal; struct taskstats *stats; - unsigned long flags; if (sig->stats != NULL) return; + /* No problem if kmem_cache_zalloc() fails */ stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); - if (!stats) - return; - spin_lock_irqsave(&sig->stats_lock, flags); + spin_lock_irq(&tsk->sighand->siglock); if (!sig->stats) { sig->stats = stats; stats = NULL; } - spin_unlock_irqrestore(&sig->stats_lock, flags); + spin_unlock_irq(&tsk->sighand->siglock); if (stats) kmem_cache_free(taskstats_cache, stats); @@ -59,7 +57,6 @@ static inline void taskstats_tgid_free(struct signal_struct *sig) extern void taskstats_exit_alloc(struct taskstats **, unsigned int *); extern void taskstats_exit_send(struct task_struct *, struct taskstats *, int, unsigned int); extern void taskstats_init_early(void); -extern void taskstats_tgid_alloc(struct signal_struct *); #else static inline void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) {} @@ -71,7 +68,7 @@ static inline void taskstats_exit_send(struct task_struct *tsk, {} static inline void taskstats_tgid_init(struct signal_struct *sig) {} -static inline void taskstats_tgid_alloc(struct signal_struct *sig) +static inline void taskstats_tgid_alloc(struct task_struct *tsk) {} static inline void taskstats_tgid_free(struct signal_struct *sig) {} diff --git a/kernel/fork.c b/kernel/fork.c index 213326609bac..3da978eec791 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -830,7 +830,7 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts if (clone_flags & CLONE_THREAD) { atomic_inc(¤t->signal->count); atomic_inc(¤t->signal->live); - taskstats_tgid_alloc(current->signal); + taskstats_tgid_alloc(current); return 0; } sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 9aeee511a463..b2efda94615a 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -241,11 +241,11 @@ static int fill_tgid(pid_t tgid, struct task_struct *tgidtsk, tsk = first; read_lock(&tasklist_lock); /* Start with stats from dead tasks */ - if (first->signal) { - spin_lock_irqsave(&first->signal->stats_lock, flags); + if (first->sighand) { + spin_lock_irqsave(&first->sighand->siglock, flags); if (first->signal->stats) memcpy(stats, first->signal->stats, sizeof(*stats)); - spin_unlock_irqrestore(&first->signal->stats_lock, flags); + spin_unlock_irqrestore(&first->sighand->siglock, flags); } do { @@ -276,7 +276,7 @@ static void fill_tgid_exit(struct task_struct *tsk) { unsigned long flags; - spin_lock_irqsave(&tsk->signal->stats_lock, flags); + spin_lock_irqsave(&tsk->sighand->siglock, flags); if (!tsk->signal->stats) goto ret; @@ -288,7 +288,7 @@ static void fill_tgid_exit(struct task_struct *tsk) */ delayacct_add_tsk(tsk->signal->stats, tsk); ret: - spin_unlock_irqrestore(&tsk->signal->stats_lock, flags); + spin_unlock_irqrestore(&tsk->sighand->siglock, flags); return; } @@ -464,15 +464,10 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size_t size; int is_thread_group; struct nlattr *na; - unsigned long flags; if (!family_registered || !tidstats) return; - spin_lock_irqsave(&tsk->signal->stats_lock, flags); - is_thread_group = tsk->signal->stats ? 1 : 0; - spin_unlock_irqrestore(&tsk->signal->stats_lock, flags); - rc = 0; /* * Size includes space for nested attributes @@ -480,6 +475,7 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); + is_thread_group = (tsk->signal->stats != NULL); if (is_thread_group) size = 2 * size; /* PID + STATS + TGID + STATS */ -- cgit v1.2.3 From 7259f0d05d595b73ef312a082e628627c6414969 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sun, 29 Oct 2006 22:46:36 -0800 Subject: [PATCH] lockdep: annotate DECLARE_WAIT_QUEUE_HEAD kernel: INFO: trying to register non-static key. kernel: the code is fine but needs lockdep annotation. kernel: turning off the locking correctness validator. kernel: [] show_trace_log_lvl+0x58/0x16a kernel: [] show_trace+0xd/0x10 kernel: [] dump_stack+0x19/0x1b kernel: [] __lock_acquire+0xf0/0x90d kernel: [] lock_acquire+0x4b/0x6b kernel: [] _spin_lock_irqsave+0x22/0x32 kernel: [] prepare_to_wait+0x17/0x4b kernel: [] lpfc_do_work+0xdd/0xcc2 [lpfc] kernel: [] kthread+0xc3/0xf2 kernel: [] kernel_thread_helper+0x5/0xb Another case of non-static lockdep keys; duplicate the paradigm set by DECLARE_COMPLETION_ONSTACK and introduce DECLARE_WAIT_QUEUE_HEAD_ONSTACK. Signed-off-by: Peter Zijlstra Cc: Greg KH Cc: Markus Lidel Acked-by: Ingo Molnar Cc: Arjan van de Ven Cc: James Bottomley Cc: Marcel Holtmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/bluetooth/bluecard_cs.c | 2 +- drivers/message/i2o/exec-osm.c | 2 +- drivers/scsi/dpt/dpti_i2o.h | 2 +- drivers/scsi/imm.c | 2 +- drivers/scsi/lpfc/lpfc_hbadisc.c | 2 +- drivers/scsi/lpfc/lpfc_sli.c | 4 ++-- drivers/scsi/ppa.c | 2 +- drivers/usb/net/usbnet.c | 2 +- include/linux/wait.h | 9 +++++++++ 9 files changed, 18 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/bluetooth/bluecard_cs.c b/drivers/bluetooth/bluecard_cs.c index 845b8680032a..cbc07250b898 100644 --- a/drivers/bluetooth/bluecard_cs.c +++ b/drivers/bluetooth/bluecard_cs.c @@ -282,7 +282,7 @@ static void bluecard_write_wakeup(bluecard_info_t *info) clear_bit(ready_bit, &(info->tx_state)); if (bt_cb(skb)->pkt_type & 0x80) { - DECLARE_WAIT_QUEUE_HEAD(wq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); DEFINE_WAIT(wait); unsigned char baud_reg; diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 01a5a702b037..a2350640384b 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -124,7 +124,7 @@ static void i2o_exec_wait_free(struct i2o_exec_wait *wait) int i2o_msg_post_wait_mem(struct i2o_controller *c, struct i2o_message *msg, unsigned long timeout, struct i2o_dma *dma) { - DECLARE_WAIT_QUEUE_HEAD(wq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); struct i2o_exec_wait *wait; static u32 tcntxt = 0x80000000; unsigned long flags; diff --git a/drivers/scsi/dpt/dpti_i2o.h b/drivers/scsi/dpt/dpti_i2o.h index b3fa7ed71faf..5a49216fe4cf 100644 --- a/drivers/scsi/dpt/dpti_i2o.h +++ b/drivers/scsi/dpt/dpti_i2o.h @@ -49,7 +49,7 @@ #include typedef wait_queue_head_t adpt_wait_queue_head_t; -#define ADPT_DECLARE_WAIT_QUEUE_HEAD(wait) DECLARE_WAIT_QUEUE_HEAD(wait) +#define ADPT_DECLARE_WAIT_QUEUE_HEAD(wait) DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait) typedef wait_queue_t adpt_wait_queue_t; /* diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c index 2d95ac9c32c1..e31f6122106f 100644 --- a/drivers/scsi/imm.c +++ b/drivers/scsi/imm.c @@ -1153,7 +1153,7 @@ static int __imm_attach(struct parport *pb) { struct Scsi_Host *host; imm_struct *dev; - DECLARE_WAIT_QUEUE_HEAD(waiting); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waiting); DEFINE_WAIT(wait); int ports; int modes, ppb; diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index d586c3d3b0d0..19c79a0549a7 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -305,7 +305,7 @@ lpfc_do_work(void *p) { struct lpfc_hba *phba = p; int rc; - DECLARE_WAIT_QUEUE_HEAD(work_waitq); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(work_waitq); set_user_nice(current, -20); phba->work_wait = &work_waitq; diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 24a1779b9af4..582f5ea4e84e 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -2983,7 +2983,7 @@ lpfc_sli_issue_iocb_wait(struct lpfc_hba * phba, struct lpfc_iocbq * prspiocbq, uint32_t timeout) { - DECLARE_WAIT_QUEUE_HEAD(done_q); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(done_q); long timeleft, timeout_req = 0; int retval = IOCB_SUCCESS; uint32_t creg_val; @@ -3061,7 +3061,7 @@ int lpfc_sli_issue_mbox_wait(struct lpfc_hba * phba, LPFC_MBOXQ_t * pmboxq, uint32_t timeout) { - DECLARE_WAIT_QUEUE_HEAD(done_q); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(done_q); DECLARE_WAITQUEUE(wq_entry, current); uint32_t timeleft = 0; int retval; diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index b0eba39f208a..89a2a9f11e41 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -1012,7 +1012,7 @@ static LIST_HEAD(ppa_hosts); static int __ppa_attach(struct parport *pb) { struct Scsi_Host *host; - DECLARE_WAIT_QUEUE_HEAD(waiting); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waiting); DEFINE_WAIT(wait); ppa_struct *dev; int ports; diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index cf3d20eb781c..40873635d80e 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -554,7 +554,7 @@ static int usbnet_stop (struct net_device *net) { struct usbnet *dev = netdev_priv(net); int temp; - DECLARE_WAIT_QUEUE_HEAD (unlink_wakeup); + DECLARE_WAIT_QUEUE_HEAD_ONSTACK (unlink_wakeup); DECLARE_WAITQUEUE (wait, current); netif_stop_queue (net); diff --git a/include/linux/wait.h b/include/linux/wait.h index b3b9048421d8..e820d00e1383 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -79,6 +79,15 @@ struct task_struct; extern void init_waitqueue_head(wait_queue_head_t *q); +#ifdef CONFIG_LOCKDEP +# define __WAIT_QUEUE_HEAD_INIT_ONSTACK(name) \ + ({ init_waitqueue_head(&name); name; }) +# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) \ + wait_queue_head_t name = __WAIT_QUEUE_HEAD_INIT_ONSTACK(name) +#else +# define DECLARE_WAIT_QUEUE_HEAD_ONSTACK(name) DECLARE_WAIT_QUEUE_HEAD(name) +#endif + static inline void init_waitqueue_entry(wait_queue_t *q, struct task_struct *p) { q->flags = 0; -- cgit v1.2.3 From 351edd240d0ba8620789ca9e24f5a38b62157f23 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 29 Oct 2006 22:46:40 -0800 Subject: [PATCH] MTD: fix last kernel-doc warning Fix the last current kernel-doc warning: Warning(/var/linsrc/linux-2619-rc3g5//include/linux/mtd/nand.h:416): No description found for parameter 'write_page' Signed-off-by: Randy Dunlap Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mtd/nand.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 70420bbae82b..8b3ef4187219 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -355,7 +355,7 @@ struct nand_buffers { * @priv: [OPTIONAL] pointer to private chip date * @errstat: [OPTIONAL] hardware specific function to perform additional error status checks * (determine if errors are correctable) - * @write_page [REPLACEABLE] High-level page write function + * @write_page: [REPLACEABLE] High-level page write function */ struct nand_chip { -- cgit v1.2.3 From 6e42acc4115bc376b8523acbcba2b2b7cc27d016 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 27 Oct 2006 19:08:42 -0700 Subject: [PATCH] libata: unexport ata_dev_revalidate() ata_dev_revalidate() isn't used outside of libata core. Unexport it. Signed-off-by: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 1 - drivers/ata/libata.h | 1 + include/linux/libata.h | 1 - 3 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 83728a9457ad..a8fd0c3e59b3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -6122,7 +6122,6 @@ EXPORT_SYMBOL_GPL(ata_std_prereset); EXPORT_SYMBOL_GPL(ata_std_softreset); EXPORT_SYMBOL_GPL(sata_std_hardreset); EXPORT_SYMBOL_GPL(ata_std_postreset); -EXPORT_SYMBOL_GPL(ata_dev_revalidate); EXPORT_SYMBOL_GPL(ata_dev_classify); EXPORT_SYMBOL_GPL(ata_dev_pair); EXPORT_SYMBOL_GPL(ata_port_disable); diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index a5ecb71390a9..0ed263be652a 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -53,6 +53,7 @@ extern unsigned ata_exec_internal(struct ata_device *dev, extern unsigned int ata_do_simple_cmd(struct ata_device *dev, u8 cmd); extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class, int post_reset, u16 *id); +extern int ata_dev_revalidate(struct ata_device *dev, int post_reset); extern int ata_dev_configure(struct ata_device *dev, int print_info); extern int sata_down_spd_limit(struct ata_port *ap); extern int sata_set_spd_needed(struct ata_port *ap); diff --git a/include/linux/libata.h b/include/linux/libata.h index b03d5a340dc8..abd2debebca2 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -702,7 +702,6 @@ extern int ata_std_prereset(struct ata_port *ap); extern int ata_std_softreset(struct ata_port *ap, unsigned int *classes); extern int sata_std_hardreset(struct ata_port *ap, unsigned int *class); extern void ata_std_postreset(struct ata_port *ap, unsigned int *classes); -extern int ata_dev_revalidate(struct ata_device *dev, int post_reset); extern void ata_port_disable(struct ata_port *); extern void ata_std_ports(struct ata_ioports *ioaddr); #ifdef CONFIG_PCI -- cgit v1.2.3 From 4fa2eeeac5e13a8579ee45bc172eed690d28fbb7 Mon Sep 17 00:00:00 2001 From: Peer Chen Date: Thu, 2 Nov 2006 18:55:48 -0500 Subject: pci_ids.h: Add NVIDIA PCI ID Signed-off-by: Jeff Garzik --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f3a168f3c9df..fa4e1d799782 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1213,6 +1213,7 @@ #define PCI_DEVICE_ID_NVIDIA_NVENET_21 0x0451 #define PCI_DEVICE_ID_NVIDIA_NVENET_22 0x0452 #define PCI_DEVICE_ID_NVIDIA_NVENET_23 0x0453 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE 0x0560 #define PCI_VENDOR_ID_IMS 0x10e0 #define PCI_DEVICE_ID_IMS_TT128 0x9128 -- cgit v1.2.3 From 86f4f0f9ba6e35fbbc409dfc3d8615c1a9822482 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 2 Nov 2006 22:07:05 -0800 Subject: [PATCH] fix UFS superblock alignment issues ufs2 fails to mount on x86_64, claiming bad magic. This is because ufs_super_block_third's fs_un1 member is padded out by 4 bytes for 8-byte alignment, pushing down the rest of the struct. Forcing this to be packed solves it. I took a quick look over other on-disk structures and didn't immediately find other problems. I was able to mount & ls a populated ufs2 filesystem w/ this change. Signed-off-by: Eric Sandeen Cc: Evgeniy Dushistov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ufs_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ufs_fs.h b/include/linux/ufs_fs.h index 61eef508b041..28967eda9d7b 100644 --- a/include/linux/ufs_fs.h +++ b/include/linux/ufs_fs.h @@ -908,7 +908,7 @@ struct ufs_super_block_third { __fs64 fs_csaddr; /* blk addr of cyl grp summary area */ __fs64 fs_pendingblocks;/* blocks in process of being freed */ __fs32 fs_pendinginodes;/*inodes in process of being freed */ - } fs_u2; + } __attribute__ ((packed)) fs_u2; } fs_un1; union { struct { -- cgit v1.2.3 From f46c483357c2d87606bbefb511321e3efd4baae0 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 2 Nov 2006 22:07:16 -0800 Subject: [PATCH] Add printk_timed_ratelimit() printk_ratelimit() has global state which makes it not useful for callers which wish to perform ratelimiting at a particular frequency. Add a printk_timed_ratelimit() which utilises caller-provided state storage to permit more flexibility. This function can in fact be used for things other than printk ratelimiting and is perhaps poorly named. Cc: Ulrich Drepper Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 2 ++ kernel/printk.c | 21 +++++++++++++++++++++ 2 files changed, 23 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 80f39cab470a..24b611147adb 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -171,6 +171,8 @@ __attribute_const__ roundup_pow_of_two(unsigned long x) extern int printk_ratelimit(void); extern int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst); +extern bool printk_timed_ratelimit(unsigned long *caller_jiffies, + unsigned int interval_msec); static inline void console_silent(void) { diff --git a/kernel/printk.c b/kernel/printk.c index f7d427ef5038..66426552fbfe 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -31,6 +31,7 @@ #include #include #include +#include #include @@ -1101,3 +1102,23 @@ int printk_ratelimit(void) printk_ratelimit_burst); } EXPORT_SYMBOL(printk_ratelimit); + +/** + * printk_timed_ratelimit - caller-controlled printk ratelimiting + * @caller_jiffies: pointer to caller's state + * @interval_msecs: minimum interval between prints + * + * printk_timed_ratelimit() returns true if more than @interval_msecs + * milliseconds have elapsed since the last time printk_timed_ratelimit() + * returned true. + */ +bool printk_timed_ratelimit(unsigned long *caller_jiffies, + unsigned int interval_msecs) +{ + if (*caller_jiffies == 0 || time_after(jiffies, *caller_jiffies)) { + *caller_jiffies = jiffies + msecs_to_jiffies(interval_msecs); + return true; + } + return false; +} +EXPORT_SYMBOL(printk_timed_ratelimit); -- cgit v1.2.3 From b918f6e62cd46774f9fc0a3fbba6bd10ad85ee14 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Thu, 2 Nov 2006 22:07:19 -0800 Subject: [PATCH] swsusp: debugging Add a swsusp debugging mode. This does everything that's needed for a suspend except for actually suspending. So we can look in the log messages and work out a) what code is being slow and b) which drivers are misbehaving. (1) # echo testproc > /sys/power/disk # echo disk > /sys/power/state This should turn off the non-boot CPU, freeze all processes, wait for 5 seconds and then thaw the processes and the CPU. (2) # echo test > /sys/power/disk # echo disk > /sys/power/state This should turn off the non-boot CPU, freeze all processes, shrink memory, suspend all devices, wait for 5 seconds, resume the devices etc. Cc: Pavel Machek Cc: Stefan Seyfried Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/ABI/testing/sysfs-power | 17 +++++++++++++++- Documentation/power/interface.txt | 13 ++++++++++++ include/linux/pm.h | 4 +++- kernel/power/disk.c | 37 ++++++++++++++++++++++++++--------- 4 files changed, 60 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index d882f8093871..dcff4d0623ad 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -21,7 +21,7 @@ Description: these states. What: /sys/power/disk -Date: August 2006 +Date: September 2006 Contact: Rafael J. Wysocki Description: The /sys/power/disk file controls the operating mode of the @@ -39,6 +39,19 @@ Description: 'reboot' - the memory image will be saved by the kernel and the system will be rebooted. + Additionally, /sys/power/disk can be used to turn on one of the + two testing modes of the suspend-to-disk mechanism: 'testproc' + or 'test'. If the suspend-to-disk mechanism is in the + 'testproc' mode, writing 'disk' to /sys/power/state will cause + the kernel to disable nonboot CPUs and freeze tasks, wait for 5 + seconds, unfreeze tasks and enable nonboot CPUs. If it is in + the 'test' mode, writing 'disk' to /sys/power/state will cause + the kernel to disable nonboot CPUs and freeze tasks, shrink + memory, suspend devices, wait for 5 seconds, resume devices, + unfreeze tasks and enable nonboot CPUs. Then, we are able to + look in the log messages and work out, for example, which code + is being slow and which device drivers are misbehaving. + The suspend-to-disk method may be chosen by writing to this file one of the accepted strings: @@ -46,6 +59,8 @@ Description: 'platform' 'shutdown' 'reboot' + 'testproc' + 'test' It will only change to 'firmware' or 'platform' if the system supports that. diff --git a/Documentation/power/interface.txt b/Documentation/power/interface.txt index a66bec222b16..74311d7e0f3c 100644 --- a/Documentation/power/interface.txt +++ b/Documentation/power/interface.txt @@ -30,6 +30,17 @@ testing). The system will support either 'firmware' or 'platform', and that is known a priori. But, the user may choose 'shutdown' or 'reboot' as alternatives. +Additionally, /sys/power/disk can be used to turn on one of the two testing +modes of the suspend-to-disk mechanism: 'testproc' or 'test'. If the +suspend-to-disk mechanism is in the 'testproc' mode, writing 'disk' to +/sys/power/state will cause the kernel to disable nonboot CPUs and freeze +tasks, wait for 5 seconds, unfreeze tasks and enable nonboot CPUs. If it is +in the 'test' mode, writing 'disk' to /sys/power/state will cause the kernel +to disable nonboot CPUs and freeze tasks, shrink memory, suspend devices, wait +for 5 seconds, resume devices, unfreeze tasks and enable nonboot CPUs. Then, +we are able to look in the log messages and work out, for example, which code +is being slow and which device drivers are misbehaving. + Reading from this file will display what the mode is currently set to. Writing to this file will accept one of @@ -37,6 +48,8 @@ to. Writing to this file will accept one of 'platform' 'shutdown' 'reboot' + 'testproc' + 'test' It will only change to 'firmware' or 'platform' if the system supports it. diff --git a/include/linux/pm.h b/include/linux/pm.h index 6b27e07aef19..070394e846d0 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -116,7 +116,9 @@ typedef int __bitwise suspend_disk_method_t; #define PM_DISK_PLATFORM ((__force suspend_disk_method_t) 2) #define PM_DISK_SHUTDOWN ((__force suspend_disk_method_t) 3) #define PM_DISK_REBOOT ((__force suspend_disk_method_t) 4) -#define PM_DISK_MAX ((__force suspend_disk_method_t) 5) +#define PM_DISK_TEST ((__force suspend_disk_method_t) 5) +#define PM_DISK_TESTPROC ((__force suspend_disk_method_t) 6) +#define PM_DISK_MAX ((__force suspend_disk_method_t) 7) struct pm_ops { suspend_disk_method_t pm_disk_mode; diff --git a/kernel/power/disk.c b/kernel/power/disk.c index d3a158a60312..b1fb7866b0b3 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -71,7 +71,7 @@ static inline void platform_finish(void) static int prepare_processes(void) { - int error; + int error = 0; pm_prepare_console(); @@ -84,6 +84,12 @@ static int prepare_processes(void) goto thaw; } + if (pm_disk_mode == PM_DISK_TESTPROC) { + printk("swsusp debug: Waiting for 5 seconds.\n"); + mdelay(5000); + goto thaw; + } + /* Free memory before shutting down devices. */ if (!(error = swsusp_shrink_memory())) return 0; @@ -120,13 +126,21 @@ int pm_suspend_disk(void) if (error) return error; + if (pm_disk_mode == PM_DISK_TESTPROC) + goto Thaw; + suspend_console(); error = device_suspend(PMSG_FREEZE); if (error) { resume_console(); printk("Some devices failed to suspend\n"); - unprepare_processes(); - return error; + goto Thaw; + } + + if (pm_disk_mode == PM_DISK_TEST) { + printk("swsusp debug: Waiting for 5 seconds.\n"); + mdelay(5000); + goto Done; } pr_debug("PM: snapshotting memory.\n"); @@ -143,16 +157,17 @@ int pm_suspend_disk(void) power_down(pm_disk_mode); else { swsusp_free(); - unprepare_processes(); - return error; + goto Thaw; } - } else + } else { pr_debug("PM: Image restored successfully.\n"); + } swsusp_free(); Done: device_resume(); resume_console(); + Thaw: unprepare_processes(); return error; } @@ -249,6 +264,8 @@ static const char * const pm_disk_modes[] = { [PM_DISK_PLATFORM] = "platform", [PM_DISK_SHUTDOWN] = "shutdown", [PM_DISK_REBOOT] = "reboot", + [PM_DISK_TEST] = "test", + [PM_DISK_TESTPROC] = "testproc", }; /** @@ -303,17 +320,19 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n) } } if (mode) { - if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT) + if (mode == PM_DISK_SHUTDOWN || mode == PM_DISK_REBOOT || + mode == PM_DISK_TEST || mode == PM_DISK_TESTPROC) { pm_disk_mode = mode; - else { + } else { if (pm_ops && pm_ops->enter && (mode == pm_ops->pm_disk_mode)) pm_disk_mode = mode; else error = -EINVAL; } - } else + } else { error = -EINVAL; + } pr_debug("PM: suspend-to-disk mode set to '%s'\n", pm_disk_modes[mode]); -- cgit v1.2.3 From 3fd593979802f81ff6452596ac61e3840f917589 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Thu, 2 Nov 2006 22:07:24 -0800 Subject: [PATCH] Create compat_sys_migrate_pages This is needed on bigendian 64bit architectures. Signed-off-by: Stephen Rothwell Acked-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 4 ++++ kernel/compat.c | 33 +++++++++++++++++++++++++++++++++ kernel/sys_ni.c | 1 + 3 files changed, 38 insertions(+) (limited to 'include/linux') diff --git a/include/linux/compat.h b/include/linux/compat.h index f1553196826f..80b17f440ec1 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -230,5 +230,9 @@ asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp); extern int compat_printk(const char *fmt, ...); extern void sigset_from_compat(sigset_t *set, compat_sigset_t *compat); +asmlinkage long compat_sys_migrate_pages(compat_pid_t pid, + compat_ulong_t maxnode, const compat_ulong_t __user *old_nodes, + const compat_ulong_t __user *new_nodes); + #endif /* CONFIG_COMPAT */ #endif /* _LINUX_COMPAT_H */ diff --git a/kernel/compat.c b/kernel/compat.c index d4898aad6cfa..6952dd057300 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -982,4 +982,37 @@ asmlinkage long compat_sys_move_pages(pid_t pid, unsigned long nr_pages, } return sys_move_pages(pid, nr_pages, pages, nodes, status, flags); } + +asmlinkage long compat_sys_migrate_pages(compat_pid_t pid, + compat_ulong_t maxnode, + const compat_ulong_t __user *old_nodes, + const compat_ulong_t __user *new_nodes) +{ + unsigned long __user *old = NULL; + unsigned long __user *new = NULL; + nodemask_t tmp_mask; + unsigned long nr_bits; + unsigned long size; + + nr_bits = min_t(unsigned long, maxnode - 1, MAX_NUMNODES); + size = ALIGN(nr_bits, BITS_PER_LONG) / 8; + if (old_nodes) { + if (compat_get_bitmap(nodes_addr(tmp_mask), old_nodes, nr_bits)) + return -EFAULT; + old = compat_alloc_user_space(new_nodes ? size * 2 : size); + if (new_nodes) + new = old + size / sizeof(unsigned long); + if (copy_to_user(old, nodes_addr(tmp_mask), size)) + return -EFAULT; + } + if (new_nodes) { + if (compat_get_bitmap(nodes_addr(tmp_mask), new_nodes, nr_bits)) + return -EFAULT; + if (new == NULL) + new = compat_alloc_user_space(size); + if (copy_to_user(new, nodes_addr(tmp_mask), size)) + return -EFAULT; + } + return sys_migrate_pages(pid, nr_bits + 1, old, new); +} #endif diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 0e53314b14de..d7306d0f3dfc 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -135,6 +135,7 @@ cond_syscall(sys_madvise); cond_syscall(sys_mremap); cond_syscall(sys_remap_file_pages); cond_syscall(compat_sys_move_pages); +cond_syscall(compat_sys_migrate_pages); /* block-layer dependent */ cond_syscall(sys_bdflush); -- cgit v1.2.3 From 4833ed094097323f5f219820f6ebdc8dd66f501f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 3 Nov 2006 00:27:06 -0800 Subject: [IPX]: Trivial parts of endianness annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/ipx.h | 14 +++++++------- include/net/ipx.h | 22 +++++++++++----------- net/ipx/af_ipx.c | 54 ++++++++++++++++++++++++++--------------------------- net/ipx/ipx_proc.c | 12 ++++++------ net/ipx/ipx_route.c | 10 +++++----- 5 files changed, 56 insertions(+), 56 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ipx.h b/include/linux/ipx.h index 4f29c60964c4..eb19b4ea84f4 100644 --- a/include/linux/ipx.h +++ b/include/linux/ipx.h @@ -7,8 +7,8 @@ struct sockaddr_ipx { sa_family_t sipx_family; - __u16 sipx_port; - __u32 sipx_network; + __be16 sipx_port; + __be32 sipx_network; unsigned char sipx_node[IPX_NODE_LEN]; __u8 sipx_type; unsigned char sipx_zero; /* 16 byte fill */ @@ -23,13 +23,13 @@ struct sockaddr_ipx { #define IPX_CRTITF 1 struct ipx_route_definition { - __u32 ipx_network; - __u32 ipx_router_network; + __be32 ipx_network; + __be32 ipx_router_network; unsigned char ipx_router_node[IPX_NODE_LEN]; }; struct ipx_interface_definition { - __u32 ipx_network; + __be32 ipx_network; unsigned char ipx_device[16]; unsigned char ipx_dlink_type; #define IPX_FRAME_NONE 0 @@ -55,8 +55,8 @@ struct ipx_config_data { */ struct ipx_route_def { - __u32 ipx_network; - __u32 ipx_router_network; + __be32 ipx_network; + __be32 ipx_router_network; #define IPX_ROUTE_NO_ROUTER 0 unsigned char ipx_router_node[IPX_NODE_LEN]; unsigned char ipx_device[16]; diff --git a/include/net/ipx.h b/include/net/ipx.h index 5c0cf33826c5..4a423d2695ba 100644 --- a/include/net/ipx.h +++ b/include/net/ipx.h @@ -15,9 +15,9 @@ #include struct ipx_address { - __u32 net; + __be32 net; __u8 node[IPX_NODE_LEN]; - __u16 sock; + __be16 sock; }; #define ipx_broadcast_node "\377\377\377\377\377\377" @@ -28,7 +28,7 @@ struct ipx_address { struct ipxhdr { __u16 ipx_checksum __attribute__ ((packed)); #define IPX_NO_CHECKSUM 0xFFFF - __u16 ipx_pktsize __attribute__ ((packed)); + __be16 ipx_pktsize __attribute__ ((packed)); __u8 ipx_tctrl; __u8 ipx_type; #define IPX_TYPE_UNKNOWN 0x00 @@ -48,14 +48,14 @@ static __inline__ struct ipxhdr *ipx_hdr(struct sk_buff *skb) struct ipx_interface { /* IPX address */ - __u32 if_netnum; + __be32 if_netnum; unsigned char if_node[IPX_NODE_LEN]; atomic_t refcnt; /* physical device info */ struct net_device *if_dev; struct datalink_proto *if_dlink; - unsigned short if_dlink_type; + __be16 if_dlink_type; /* socket support */ unsigned short if_sknum; @@ -71,7 +71,7 @@ struct ipx_interface { }; struct ipx_route { - __u32 ir_net; + __be32 ir_net; struct ipx_interface *ir_intrfc; unsigned char ir_routed; unsigned char ir_router_node[IPX_NODE_LEN]; @@ -82,10 +82,10 @@ struct ipx_route { #ifdef __KERNEL__ struct ipx_cb { u8 ipx_tctrl; - u32 ipx_dest_net; - u32 ipx_source_net; + __be32 ipx_dest_net; + __be32 ipx_source_net; struct { - u32 netnum; + __be32 netnum; int index; } last_hop; }; @@ -97,7 +97,7 @@ struct ipx_sock { struct sock sk; struct ipx_address dest_addr; struct ipx_interface *intrfc; - unsigned short port; + __be16 port; #ifdef CONFIG_IPX_INTERN unsigned char node[IPX_NODE_LEN]; #endif @@ -132,7 +132,7 @@ extern struct ipx_interface *ipx_primary_net; extern int ipx_proc_init(void); extern void ipx_proc_exit(void); -extern const char *ipx_frame_name(unsigned short); +extern const char *ipx_frame_name(__be16); extern const char *ipx_device_name(struct ipx_interface *intrfc); static __inline__ void ipxitf_hold(struct ipx_interface *intrfc) diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index bef3f61569f7..c272a38af325 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -83,13 +83,13 @@ DEFINE_SPINLOCK(ipx_interfaces_lock); struct ipx_interface *ipx_primary_net; struct ipx_interface *ipx_internal_net; -extern int ipxrtr_add_route(__u32 network, struct ipx_interface *intrfc, +extern int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, unsigned char *node); extern void ipxrtr_del_routes(struct ipx_interface *intrfc); extern int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, struct iovec *iov, int len, int noblock); extern int ipxrtr_route_skb(struct sk_buff *skb); -extern struct ipx_route *ipxrtr_lookup(__u32 net); +extern struct ipx_route *ipxrtr_lookup(__be32 net); extern int ipxrtr_ioctl(unsigned int cmd, void __user *arg); #undef IPX_REFCNT_DEBUG @@ -177,7 +177,7 @@ static void ipxitf_clear_primary_net(void) } static struct ipx_interface *__ipxitf_find_using_phys(struct net_device *dev, - unsigned short datalink) + __be16 datalink) { struct ipx_interface *i; @@ -190,7 +190,7 @@ out: } static struct ipx_interface *ipxitf_find_using_phys(struct net_device *dev, - unsigned short datalink) + __be16 datalink) { struct ipx_interface *i; @@ -202,7 +202,7 @@ static struct ipx_interface *ipxitf_find_using_phys(struct net_device *dev, return i; } -struct ipx_interface *ipxitf_find_using_net(__u32 net) +struct ipx_interface *ipxitf_find_using_net(__be32 net) { struct ipx_interface *i; @@ -237,7 +237,7 @@ static void ipxitf_insert_socket(struct ipx_interface *intrfc, struct sock *sk) /* caller must hold intrfc->if_sklist_lock */ static struct sock *__ipxitf_find_socket(struct ipx_interface *intrfc, - unsigned short port) + __be16 port) { struct sock *s; struct hlist_node *node; @@ -252,7 +252,7 @@ found: /* caller must hold a reference to intrfc */ static struct sock *ipxitf_find_socket(struct ipx_interface *intrfc, - unsigned short port) + __be16 port) { struct sock *s; @@ -268,7 +268,7 @@ static struct sock *ipxitf_find_socket(struct ipx_interface *intrfc, #ifdef CONFIG_IPX_INTERN static struct sock *ipxitf_find_internal_socket(struct ipx_interface *intrfc, unsigned char *ipx_node, - unsigned short port) + __be16 port) { struct sock *s; struct hlist_node *node; @@ -600,10 +600,10 @@ int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node) /* see if we need to include the netnum in the route list */ if (IPX_SKB_CB(skb)->last_hop.index >= 0) { - u32 *last_hop = (u32 *)(((u8 *) skb->data) + + __be32 *last_hop = (__be32 *)(((u8 *) skb->data) + sizeof(struct ipxhdr) + IPX_SKB_CB(skb)->last_hop.index * - sizeof(u32)); + sizeof(__be32)); *last_hop = IPX_SKB_CB(skb)->last_hop.netnum; IPX_SKB_CB(skb)->last_hop.index = -1; } @@ -772,7 +772,7 @@ static void ipxitf_discover_netnum(struct ipx_interface *intrfc, } else { printk(KERN_WARNING "IPX: Network number collision " "%lx\n %s %s and %s %s\n", - (unsigned long) htonl(cb->ipx_source_net), + (unsigned long) ntohl(cb->ipx_source_net), ipx_device_name(i), ipx_frame_name(i->if_dlink_type), ipx_device_name(intrfc), @@ -812,7 +812,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) int i, rc = -EINVAL; struct ipx_interface *ifcs; char *c; - u32 *l; + __be32 *l; /* Illegal packet - too many hops or too short */ /* We decide to throw it away: no broadcasting, no local processing. @@ -833,7 +833,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) goto out; c = ((u8 *) ipx) + sizeof(struct ipxhdr); - l = (u32 *) c; + l = (__be32 *) c; /* Don't broadcast packet if already seen this net */ for (i = 0; i < IPX_SKB_CB(skb)->ipx_tctrl; i++) @@ -855,7 +855,7 @@ static int ipxitf_pprop(struct ipx_interface *intrfc, struct sk_buff *skb) /* That aren't in the list */ if (ifcs == intrfc) continue; - l = (__u32 *) c; + l = (__be32 *) c; /* don't consider the last entry in the packet list, * it is our netnum, and it is not there yet */ for (i = 0; i < IPX_SKB_CB(skb)->ipx_tctrl; i++) @@ -885,8 +885,8 @@ static void ipxitf_insert(struct ipx_interface *intrfc) ipx_primary_net = intrfc; } -static struct ipx_interface *ipxitf_alloc(struct net_device *dev, __u32 netnum, - unsigned short dlink_type, +static struct ipx_interface *ipxitf_alloc(struct net_device *dev, __be32 netnum, + __be16 dlink_type, struct datalink_proto *dlink, unsigned char internal, int ipx_offset) @@ -960,7 +960,7 @@ static __be16 ipx_map_frame_type(unsigned char type) static int ipxitf_create(struct ipx_interface_definition *idef) { struct net_device *dev; - unsigned short dlink_type = 0; + __be16 dlink_type = 0; struct datalink_proto *datalink = NULL; struct ipx_interface *intrfc; int rc; @@ -1073,7 +1073,7 @@ out: static int ipxitf_delete(struct ipx_interface_definition *idef) { struct net_device *dev = NULL; - unsigned short dlink_type = 0; + __be16 dlink_type = 0; struct ipx_interface *intrfc; int rc = 0; @@ -1110,7 +1110,7 @@ out: } static struct ipx_interface *ipxitf_auto_create(struct net_device *dev, - unsigned short dlink_type) + __be16 dlink_type) { struct ipx_interface *intrfc = NULL; struct datalink_proto *datalink; @@ -1122,7 +1122,7 @@ static struct ipx_interface *ipxitf_auto_create(struct net_device *dev, if (dev->addr_len > IPX_NODE_LEN) goto out; - switch (htons(dlink_type)) { + switch (ntohs(dlink_type)) { case ETH_P_IPX: datalink = pEII_datalink; break; case ETH_P_802_2: datalink = p8022_datalink; break; case ETH_P_SNAP: datalink = pSNAP_datalink; break; @@ -1266,7 +1266,7 @@ __u16 ipx_cksum(struct ipxhdr *packet, int length) return ~sum; } -const char *ipx_frame_name(unsigned short frame) +const char *ipx_frame_name(__be16 frame) { char* rc = "None"; @@ -1401,7 +1401,7 @@ out: /* caller must hold a reference to intrfc */ -static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) +static __be16 ipx_first_free_socketnum(struct ipx_interface *intrfc) { unsigned short socketNum = intrfc->if_sknum; @@ -1410,7 +1410,7 @@ static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) if (socketNum < IPX_MIN_EPHEMERAL_SOCKET) socketNum = IPX_MIN_EPHEMERAL_SOCKET; - while (__ipxitf_find_socket(intrfc, ntohs(socketNum))) + while (__ipxitf_find_socket(intrfc, htons(socketNum))) if (socketNum > IPX_MAX_EPHEMERAL_SOCKET) socketNum = IPX_MIN_EPHEMERAL_SOCKET; else @@ -1419,7 +1419,7 @@ static unsigned short ipx_first_free_socketnum(struct ipx_interface *intrfc) spin_unlock_bh(&intrfc->if_sklist_lock); intrfc->if_sknum = socketNum; - return ntohs(socketNum); + return htons(socketNum); } static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) @@ -1473,7 +1473,7 @@ static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) ipxs->port)) { SOCK_DEBUG(sk, "IPX: bind failed because port %X in use.\n", - ntohs((int)addr->sipx_port)); + ntohs(addr->sipx_port)); goto out_put; } } else { @@ -1488,7 +1488,7 @@ static int ipx_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (ipxitf_find_socket(intrfc, addr->sipx_port)) { SOCK_DEBUG(sk, "IPX: bind failed because port %X in use.\n", - ntohs((int)addr->sipx_port)); + ntohs(addr->sipx_port)); goto out_put; } } @@ -1665,7 +1665,7 @@ static int ipx_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty intrfc = ipxitf_find_using_phys(dev, pt->type); if (!intrfc) { if (ipxcfg_auto_create_interfaces && - ntohl(IPX_SKB_CB(skb)->ipx_dest_net)) { + IPX_SKB_CB(skb)->ipx_dest_net) { intrfc = ipxitf_auto_create(dev, pt->type); if (intrfc) ipxitf_hold(intrfc); diff --git a/net/ipx/ipx_proc.c b/net/ipx/ipx_proc.c index 4c0c71206e54..b7463dfca63e 100644 --- a/net/ipx/ipx_proc.c +++ b/net/ipx/ipx_proc.c @@ -260,22 +260,22 @@ static int ipx_seq_socket_show(struct seq_file *seq, void *v) ipxs = ipx_sk(s); #ifdef CONFIG_IPX_INTERN seq_printf(seq, "%08lX:%02X%02X%02X%02X%02X%02X:%04X ", - (unsigned long)htonl(ipxs->intrfc->if_netnum), + (unsigned long)ntohl(ipxs->intrfc->if_netnum), ipxs->node[0], ipxs->node[1], ipxs->node[2], ipxs->node[3], - ipxs->node[4], ipxs->node[5], htons(ipxs->port)); + ipxs->node[4], ipxs->node[5], ntohs(ipxs->port)); #else - seq_printf(seq, "%08lX:%04X ", (unsigned long) htonl(ipxs->intrfc->if_netnum), - htons(ipxs->port)); + seq_printf(seq, "%08lX:%04X ", (unsigned long) ntohl(ipxs->intrfc->if_netnum), + ntohs(ipxs->port)); #endif /* CONFIG_IPX_INTERN */ if (s->sk_state != TCP_ESTABLISHED) seq_printf(seq, "%-28s", "Not_Connected"); else { seq_printf(seq, "%08lX:%02X%02X%02X%02X%02X%02X:%04X ", - (unsigned long)htonl(ipxs->dest_addr.net), + (unsigned long)ntohl(ipxs->dest_addr.net), ipxs->dest_addr.node[0], ipxs->dest_addr.node[1], ipxs->dest_addr.node[2], ipxs->dest_addr.node[3], ipxs->dest_addr.node[4], ipxs->dest_addr.node[5], - htons(ipxs->dest_addr.sock)); + ntohs(ipxs->dest_addr.sock)); } seq_printf(seq, "%08X %08X %02X %03d\n", diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index a30dbb1e08fb..8bfaefaf8841 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -20,16 +20,16 @@ DEFINE_RWLOCK(ipx_routes_lock); extern struct ipx_interface *ipx_internal_net; extern __u16 ipx_cksum(struct ipxhdr *packet, int length); -extern struct ipx_interface *ipxitf_find_using_net(__u32 net); +extern struct ipx_interface *ipxitf_find_using_net(__be32 net); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); extern int ipxitf_demux_socket(struct ipx_interface *intrfc, struct sk_buff *skb, int copy); extern int ipxitf_send(struct ipx_interface *intrfc, struct sk_buff *skb, char *node); -extern struct ipx_interface *ipxitf_find_using_net(__u32 net); +extern struct ipx_interface *ipxitf_find_using_net(__be32 net); -struct ipx_route *ipxrtr_lookup(__u32 net) +struct ipx_route *ipxrtr_lookup(__be32 net) { struct ipx_route *r; @@ -48,7 +48,7 @@ unlock: /* * Caller must hold a reference to intrfc */ -int ipxrtr_add_route(__u32 network, struct ipx_interface *intrfc, +int ipxrtr_add_route(__be32 network, struct ipx_interface *intrfc, unsigned char *node) { struct ipx_route *rt; @@ -118,7 +118,7 @@ out: return rc; } -static int ipxrtr_delete(__u32 net) +static int ipxrtr_delete(__be32 net) { struct ipx_route *r, *tmp; int rc; -- cgit v1.2.3 From d99f160ac53e51090f015a8f0617cea25f81a191 Mon Sep 17 00:00:00 2001 From: Eric W. Biederman Date: Sun, 5 Nov 2006 23:52:12 -0800 Subject: [PATCH] sysctl: allow a zero ctl_name in the middle of a sysctl table Since it is becoming clear that there are just enough users of the binary sysctl interface that completely removing the binary interface from the kernel will not be an option for foreseeable future, we need to find a way to address the sysctl maintenance issues. The basic problem is that sysctl requires one central authority to allocate sysctl numbers, or else conflicts and ABI breakage occur. The proc interface to sysctl does not have that problem, as names are not densely allocated. By not terminating a sysctl table until I have neither a ctl_name nor a procname, it becomes simple to add sysctl entries that don't show up in the binary sysctl interface. Which allows people to avoid allocating a binary sysctl value when not needed. I have audited the kernel code and in my reading I have not found a single sysctl table that wasn't terminated by a completely zero filled entry. So this change in behavior should not affect anything. I think this mechanism eases the pain enough that combined with a little disciple we can solve the reoccurring sysctl ABI breakage. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sysctl.h | 9 ++++++--- kernel/sysctl.c | 8 +++++--- 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 1b24bd45e080..c184732a70fc 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -961,8 +961,8 @@ extern ctl_handler sysctl_ms_jiffies; /* * Register a set of sysctl names by calling register_sysctl_table * with an initialised array of ctl_table's. An entry with zero - * ctl_name terminates the table. table->de will be set up by the - * registration and need not be initialised in advance. + * ctl_name and NULL procname terminates the table. table->de will be + * set up by the registration and need not be initialised in advance. * * sysctl names can be mirrored automatically under /proc/sys. The * procname supplied controls /proc naming. @@ -973,7 +973,10 @@ extern ctl_handler sysctl_ms_jiffies; * Leaf nodes in the sysctl tree will be represented by a single file * under /proc; non-leaf nodes will be represented by directories. A * null procname disables /proc mirroring at this node. - * + * + * sysctl entries with a zero ctl_name will not be available through + * the binary sysctl interface. + * * sysctl(2) can automatically manage read and write requests through * the sysctl table. The data and maxlen fields of the ctl_table * struct enable minimal validation of the values being written to be diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 0c8e805bbd6f..09e569f4792b 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1315,7 +1315,9 @@ repeat: return -ENOTDIR; if (get_user(n, name)) return -EFAULT; - for ( ; table->ctl_name; table++) { + for ( ; table->ctl_name || table->procname; table++) { + if (!table->ctl_name) + continue; if (n == table->ctl_name || table->ctl_name == CTL_ANY) { int error; if (table->child) { @@ -1532,7 +1534,7 @@ static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, int len; mode_t mode; - for (; table->ctl_name; table++) { + for (; table->ctl_name || table->procname; table++) { /* Can't do anything without a proc name. */ if (!table->procname) continue; @@ -1579,7 +1581,7 @@ static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, static void unregister_proc_table(ctl_table * table, struct proc_dir_entry *root) { struct proc_dir_entry *de; - for (; table->ctl_name; table++) { + for (; table->ctl_name || table->procname; table++) { if (!(de = table->de)) continue; if (de->mode & S_IFDIR) { -- cgit v1.2.3 From 7cc13edc139108bb527b692f0548dce6bc648572 Mon Sep 17 00:00:00 2001 From: Eric W. Biederman Date: Sun, 5 Nov 2006 23:52:13 -0800 Subject: [PATCH] sysctl: implement CTL_UNNUMBERED This patch takes the CTL_UNNUMBERD concept from NFS and makes it available to all new sysctl users. At the same time the sysctl binary interface maintenance documentation is updated to mention and to describe what is needed to successfully maintain the sysctl binary interface. Signed-off-by: Eric W. Biederman Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/svc.c | 3 --- fs/nfs/sysctl.c | 5 ----- include/linux/sysctl.h | 14 +++++++++++--- 3 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 634139232aaf..8ca18085e68d 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -353,9 +353,6 @@ EXPORT_SYMBOL(lockd_down); * Sysctl parameters (same as module parameters, different interface). */ -/* Something that isn't CTL_ANY, CTL_NONE or a value that may clash. */ -#define CTL_UNNUMBERED -2 - static ctl_table nlm_sysctls[] = { { .ctl_name = CTL_UNNUMBERED, diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c index 2fe3403c2409..3ea50ac64820 100644 --- a/fs/nfs/sysctl.c +++ b/fs/nfs/sysctl.c @@ -18,11 +18,6 @@ static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; static struct ctl_table_header *nfs_callback_sysctl_table; -/* - * Something that isn't CTL_ANY, CTL_NONE or a value that may clash. - * Use the same values as fs/lockd/svc.c - */ -#define CTL_UNNUMBERED -2 static ctl_table nfs_cb_sysctls[] = { #ifdef CONFIG_NFS_V4 diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index c184732a70fc..d98562f1df76 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -6,10 +6,17 @@ **************************************************************** **************************************************************** ** + ** WARNING: ** The values in this file are exported to user space via - ** the sysctl() binary interface. However this interface - ** is unstable and deprecated and will be removed in the future. - ** For a stable interface use /proc/sys. + ** the sysctl() binary interface. Do *NOT* change the + ** numbering of any existing values here, and do not change + ** any numbers within any one set of values. If you have to + ** have to redefine an existing interface, use a new number for it. + ** The kernel will then return -ENOTDIR to any application using + ** the old binary interface. + ** + ** For new interfaces unless you really need a binary number + ** please use CTL_UNNUMBERED. ** **************************************************************** **************************************************************** @@ -48,6 +55,7 @@ struct __sysctl_args { #ifdef __KERNEL__ #define CTL_ANY -1 /* Matches any name */ #define CTL_NONE 0 +#define CTL_UNNUMBERED CTL_NONE /* sysctl without a binary number */ #endif enum -- cgit v1.2.3 From 81ac95c5569d7a60ab5db6c1ccec56c12b3ebcb5 Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Wed, 8 Nov 2006 17:44:40 -0800 Subject: [PATCH] nfsd4: fix open-create permissions In the case where an open creates the file, we shouldn't be rechecking permissions to open the file; the open succeeds regardless of what the new file's mode bits say. This patch fixes the problem, but only by introducing yet another parameter to nfsd_create_v3. This is ugly. This will be fixed by later patches. Signed-off-by: J. Bruce Fields Acked-by: Neil Brown Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs3proc.c | 2 +- fs/nfsd/nfs4proc.c | 6 ++++-- fs/nfsd/vfs.c | 4 +++- include/linux/nfsd/nfsd.h | 2 +- 4 files changed, 9 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 64db601c2bd2..7f5bad0393b1 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -258,7 +258,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* Now create the file and set attributes */ nfserr = nfsd_create_v3(rqstp, dirfhp, argp->name, argp->len, attr, newfhp, - argp->createmode, argp->verf, NULL); + argp->createmode, argp->verf, NULL, NULL); RETURN_STATUS(nfserr); } diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 4a73f5b2546f..50bc94243ca1 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -93,6 +93,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o { struct svc_fh resfh; __be32 status; + int created = 0; fh_init(&resfh, NFS4_FHSIZE); open->op_truncate = 0; @@ -105,7 +106,7 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o status = nfsd_create_v3(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &open->op_iattr, &resfh, open->op_createmode, - (u32 *)open->op_verf.data, &open->op_truncate); + (u32 *)open->op_verf.data, &open->op_truncate, &created); } else { status = nfsd_lookup(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &resfh); @@ -122,7 +123,8 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o memcpy(open->op_stateowner->so_replay.rp_openfh, &resfh.fh_handle.fh_base, resfh.fh_handle.fh_size); - status = do_open_permission(rqstp, current_fh, open, MAY_NOP); + if (!created) + status = do_open_permission(rqstp, current_fh, open, MAY_NOP); out: fh_put(&resfh); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index f21e917bb8ed..1a7ad8c983d1 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1237,7 +1237,7 @@ __be32 nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, - int *truncp) + int *truncp, int *created) { struct dentry *dentry, *dchild = NULL; struct inode *dirp; @@ -1331,6 +1331,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); if (host_err < 0) goto out_nfserr; + if (created) + *created = 1; if (EX_ISSYNC(fhp->fh_export)) { err = nfserrno(nfsd_sync_dir(dentry)); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index eb231143d579..edb54c3171b3 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -89,7 +89,7 @@ __be32 nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); __be32 nfsd_create_v3(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, struct svc_fh *res, int createmode, - u32 *verifier, int *truncp); + u32 *verifier, int *truncp, int *created); __be32 nfsd_commit(struct svc_rqst *, struct svc_fh *, loff_t, unsigned long); #endif /* CONFIG_NFSD_V3 */ -- cgit v1.2.3 From 46d52b09fa6a2d1e313cb75ca352d6f466e67bd1 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 8 Nov 2006 17:44:55 -0800 Subject: [PATCH] IPMI: retry messages on certain error returns Some more errors from the IPMI send message command are retryable, but are not being retried by the IPMI code. Make sure they get retried. Signed-off-by: Corey Minyard Cc: Frederic Lelievre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 4 +++- include/linux/ipmi_msgdefs.h | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index e55a0d276729..0b07ca1b71fa 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3242,7 +3242,9 @@ void ipmi_smi_msg_received(ipmi_smi_t intf, report the error immediately. */ if ((msg->rsp_size >= 3) && (msg->rsp[2] != 0) && (msg->rsp[2] != IPMI_NODE_BUSY_ERR) - && (msg->rsp[2] != IPMI_LOST_ARBITRATION_ERR)) + && (msg->rsp[2] != IPMI_LOST_ARBITRATION_ERR) + && (msg->rsp[2] != IPMI_BUS_ERR) + && (msg->rsp[2] != IPMI_NAK_ON_WRITE_ERR)) { int chan = msg->rsp[3] & 0xf; diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index 22f5e2afda4f..4d04d8b58a0a 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -75,6 +75,8 @@ #define IPMI_INVALID_COMMAND_ERR 0xc1 #define IPMI_ERR_MSG_TRUNCATED 0xc6 #define IPMI_LOST_ARBITRATION_ERR 0x81 +#define IPMI_BUS_ERR 0x82 +#define IPMI_NAK_ON_WRITE_ERR 0x83 #define IPMI_ERR_UNSPECIFIED 0xff #define IPMI_CHANNEL_PROTOCOL_IPMB 1 -- cgit v1.2.3 From ec68307cc5a8dc499e48693843bb42f6b6028458 Mon Sep 17 00:00:00 2001 From: Eric W. Biederman Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: refactor so we only have one function that writes to the chip This refactoring actually optimizes the code a little by caching the value that we think the device is programmed with instead of reading it back from the hardware. Which simplifies the code a little and should speed things up a bit. This patch introduces the concept of a ht_irq_msg and modifies the architecture read/write routines to update this code. There is a minor consistency fix here as well as x86_64 forgot to initialize the htirq as masked. Signed-off-by: Eric W. Biederman Cc: Andi Kleen Acked-by: Bryan O'Sullivan Cc: Cc: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 26 ++++++++-------- arch/x86_64/kernel/io_apic.c | 31 +++++++++---------- drivers/pci/htirq.c | 72 +++++++++++++++----------------------------- include/linux/htirq.h | 11 ++++--- 4 files changed, 58 insertions(+), 82 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 507983c513c3..ad84bc2802a6 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2624,18 +2624,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -2673,7 +2671,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; cpumask_t tmp; @@ -2681,9 +2679,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) cpu_set(vector >> 8, tmp); dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2695,8 +2694,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_MT_ARBITRATED) | HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 3b8f9c68ad3c..41bfc49301ad 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1955,18 +1955,16 @@ void arch_teardown_msi_irq(unsigned int irq) static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) { - u32 low, high; - low = read_ht_irq_low(irq); - high = read_ht_irq_high(irq); + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); - low &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); - high &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); - low |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); - high |= HT_IRQ_HIGH_DEST_ID(dest); + msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); } static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) @@ -1987,7 +1985,7 @@ static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask) dest = cpu_mask_to_apicid(tmp); - target_ht_irq(irq, dest, vector & 0xff); + target_ht_irq(irq, dest, vector); set_native_irq_info(irq, mask); } #endif @@ -2010,14 +2008,15 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) vector = assign_irq_vector(irq, TARGET_CPUS, &tmp); if (vector >= 0) { - u32 low, high; + struct ht_irq_msg msg; unsigned dest; dest = cpu_mask_to_apicid(tmp); - high = HT_IRQ_HIGH_DEST_ID(dest); + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); - low = HT_IRQ_LOW_BASE | + msg.address_lo = + HT_IRQ_LOW_BASE | HT_IRQ_LOW_DEST_ID(dest) | HT_IRQ_LOW_VECTOR(vector) | ((INT_DEST_MODE == 0) ? @@ -2026,10 +2025,10 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) HT_IRQ_LOW_RQEOI_EDGE | ((INT_DELIVERY_MODE != dest_LowestPrio) ? HT_IRQ_LOW_MT_FIXED : - HT_IRQ_LOW_MT_ARBITRATED); + HT_IRQ_LOW_MT_ARBITRATED) | + HT_IRQ_LOW_IRQ_MASKED; - write_ht_irq_low(irq, low); - write_ht_irq_high(irq, high); + write_ht_irq_msg(irq, &msg); set_irq_chip_and_handler_name(irq, &ht_irq_chip, handle_edge_irq, "edge"); diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index 0e27f2404a83..e346fe31f97a 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -27,82 +27,55 @@ struct ht_irq_cfg { struct pci_dev *dev; unsigned pos; unsigned idx; + struct ht_irq_msg msg; }; -void write_ht_irq_low(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} - -void write_ht_irq_high(unsigned int irq, u32 data) -{ - struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); -} -u32 read_ht_irq_low(unsigned int irq) +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); unsigned long flags; - u32 data; spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); + if (cfg->msg.address_lo != msg->address_lo) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_lo); + } + if (cfg->msg.address_hi != msg->address_hi) { + pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); + pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_hi); + } spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + cfg->msg = *msg; } -u32 read_ht_irq_high(unsigned int irq) +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) { struct ht_irq_cfg *cfg = get_irq_data(irq); - unsigned long flags; - u32 data; - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - spin_unlock_irqrestore(&ht_irq_lock, flags); - return data; + *msg = cfg->msg; } void mask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data |= 1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo |= 1; + write_ht_irq_msg(irq, &msg); } void unmask_ht_irq(unsigned int irq) { struct ht_irq_cfg *cfg; - unsigned long flags; - u32 data; + struct ht_irq_msg msg; cfg = get_irq_data(irq); - spin_lock_irqsave(&ht_irq_lock, flags); - pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx); - pci_read_config_dword(cfg->dev, cfg->pos + 4, &data); - data &= ~1; - pci_write_config_dword(cfg->dev, cfg->pos + 4, data); - spin_unlock_irqrestore(&ht_irq_lock, flags); + msg = cfg->msg; + msg.address_lo &= ~1; + write_ht_irq_msg(irq, &msg); } /** @@ -152,6 +125,9 @@ int ht_create_irq(struct pci_dev *dev, int idx) cfg->dev = dev; cfg->pos = pos; cfg->idx = 0x10 + (idx * 2); + /* Initialize msg to a value that will never match the first write. */ + cfg->msg.address_lo = 0xffffffff; + cfg->msg.address_hi = 0xffffffff; irq = create_irq(); if (irq < 0) { diff --git a/include/linux/htirq.h b/include/linux/htirq.h index 1f15ce279a23..108f0d91e11e 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -1,11 +1,14 @@ #ifndef LINUX_HTIRQ_H #define LINUX_HTIRQ_H +struct ht_irq_msg { + u32 address_lo; /* low 32 bits of the ht irq message */ + u32 address_hi; /* high 32 bits of the it irq message */ +}; + /* Helper functions.. */ -void write_ht_irq_low(unsigned int irq, u32 data); -void write_ht_irq_high(unsigned int irq, u32 data); -u32 read_ht_irq_low(unsigned int irq); -u32 read_ht_irq_high(unsigned int irq); +void fetch_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); +void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg); void mask_ht_irq(unsigned int irq); void unmask_ht_irq(unsigned int irq); -- cgit v1.2.3 From 43539c38cd8edb915d1f0e1f55dcb70638b4cc8e Mon Sep 17 00:00:00 2001 From: Eric W. Biederman Date: Wed, 8 Nov 2006 17:44:57 -0800 Subject: [PATCH] htirq: allow buggy drivers of buggy hardware to write the registers This patch adds a variant of ht_create_irq __ht_create_irq that takes an aditional parameter update that is a function that is called whenever we want to write to a drivers htirq configuration registers. This is needed to support the ipath_iba6110 because it's registers in the proper location are not actually conected to the hardware that controlls interrupt delivery. [bos@serpentine.com: fixes] Signed-off-by: Eric W. Biederman Cc: Andi Kleen Cc: Cc: Roland Dreier Signed-off-by: Bryan O'Sullivan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/htirq.c | 29 ++++++++++++++++++++++++----- include/linux/htirq.h | 5 +++++ 2 files changed, 29 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index e346fe31f97a..0a8d1cce9fa0 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -25,6 +25,8 @@ static DEFINE_SPINLOCK(ht_irq_lock); struct ht_irq_cfg { struct pci_dev *dev; + /* Update callback used to cope with buggy hardware */ + ht_irq_update_t *update; unsigned pos; unsigned idx; struct ht_irq_msg msg; @@ -44,6 +46,8 @@ void write_ht_irq_msg(unsigned int irq, struct ht_irq_msg *msg) pci_write_config_byte(cfg->dev, cfg->pos + 2, cfg->idx + 1); pci_write_config_dword(cfg->dev, cfg->pos + 4, msg->address_hi); } + if (cfg->update) + cfg->update(cfg->dev, irq, msg); spin_unlock_irqrestore(&ht_irq_lock, flags); cfg->msg = *msg; } @@ -79,16 +83,14 @@ void unmask_ht_irq(unsigned int irq) } /** - * ht_create_irq - create an irq and attach it to a device. + * __ht_create_irq - create an irq and attach it to a device. * @dev: The hypertransport device to find the irq capability on. * @idx: Which of the possible irqs to attach to. - * - * ht_create_irq is needs to be called for all hypertransport devices - * that generate irqs. + * @update: Function to be called when changing the htirq message * * The irq number of the new irq or a negative error value is returned. */ -int ht_create_irq(struct pci_dev *dev, int idx) +int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update) { struct ht_irq_cfg *cfg; unsigned long flags; @@ -123,6 +125,7 @@ int ht_create_irq(struct pci_dev *dev, int idx) return -ENOMEM; cfg->dev = dev; + cfg->update = update; cfg->pos = pos; cfg->idx = 0x10 + (idx * 2); /* Initialize msg to a value that will never match the first write. */ @@ -144,6 +147,21 @@ int ht_create_irq(struct pci_dev *dev, int idx) return irq; } +/** + * ht_create_irq - create an irq and attach it to a device. + * @dev: The hypertransport device to find the irq capability on. + * @idx: Which of the possible irqs to attach to. + * + * ht_create_irq needs to be called for all hypertransport devices + * that generate irqs. + * + * The irq number of the new irq or a negative error value is returned. + */ +int ht_create_irq(struct pci_dev *dev, int idx) +{ + return __ht_create_irq(dev, idx, NULL); +} + /** * ht_destroy_irq - destroy an irq created with ht_create_irq * @@ -162,5 +180,6 @@ void ht_destroy_irq(unsigned int irq) kfree(cfg); } +EXPORT_SYMBOL(__ht_create_irq); EXPORT_SYMBOL(ht_create_irq); EXPORT_SYMBOL(ht_destroy_irq); diff --git a/include/linux/htirq.h b/include/linux/htirq.h index 108f0d91e11e..c96ea46737d0 100644 --- a/include/linux/htirq.h +++ b/include/linux/htirq.h @@ -15,4 +15,9 @@ void unmask_ht_irq(unsigned int irq); /* The arch hook for getting things started */ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev); +/* For drivers of buggy hardware */ +typedef void (ht_irq_update_t)(struct pci_dev *dev, int irq, + struct ht_irq_msg *msg); +int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update); + #endif /* LINUX_HTIRQ_H */ -- cgit v1.2.3 From a9b14973a8c42b2aecc968851372203c6567e196 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Thu, 19 Oct 2006 19:52:26 -0500 Subject: [PATCH] Slight refactor of interrupt mapping for FSL parts * Cleaned up interrupt mapping a little by adding a helper function which parses the irq out of the device-tree, and puts it into a resource. * Changed the arch/ppc platform files to specify PHY_POLL, instead of -1 * Changed the fixed phy to use PHY_IGNORE_INTERRUPT * Added ethtool.h and mii.h to phy.h includes Signed-off-by: Paul Mackerras --- arch/powerpc/sysdev/fsl_soc.c | 28 +++++++++++----------------- arch/ppc/platforms/83xx/mpc834x_sys.c | 4 ++-- arch/ppc/platforms/85xx/mpc8540_ads.c | 4 ++-- arch/ppc/platforms/85xx/mpc8560_ads.c | 4 ++-- arch/ppc/platforms/85xx/mpc85xx_cds_common.c | 6 +++--- arch/ppc/platforms/85xx/sbc8560.c | 2 +- arch/ppc/platforms/85xx/stx_gp3.c | 2 +- arch/ppc/platforms/85xx/tqm85xx.c | 4 ++-- arch/ppc/platforms/mpc8272ads_setup.c | 6 +++--- arch/ppc/platforms/mpc866ads_setup.c | 4 ++-- drivers/net/phy/fixed.c | 2 +- include/asm-powerpc/prom.h | 7 +++++++ include/linux/phy.h | 3 +++ 13 files changed, 40 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c index dbe92ae20333..ad31e56e892b 100644 --- a/arch/powerpc/sysdev/fsl_soc.c +++ b/arch/powerpc/sysdev/fsl_soc.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -146,7 +147,7 @@ static int __init gfar_mdio_of_init(void) } for (k = 0; k < 32; k++) - mdio_data.irq[k] = -1; + mdio_data.irq[k] = PHY_POLL; while ((child = of_get_next_child(np, child)) != NULL) { int irq = irq_of_parse_and_map(child, 0); @@ -177,6 +178,7 @@ static const char *gfar_tx_intr = "tx"; static const char *gfar_rx_intr = "rx"; static const char *gfar_err_intr = "error"; + static int __init gfar_of_init(void) { struct device_node *np; @@ -204,8 +206,7 @@ static int __init gfar_of_init(void) if (ret) goto err; - r[1].start = r[1].end = irq_of_parse_and_map(np, 0); - r[1].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[1]); model = get_property(np, "model", NULL); @@ -214,12 +215,10 @@ static int __init gfar_of_init(void) r[1].name = gfar_tx_intr; r[2].name = gfar_rx_intr; - r[2].start = r[2].end = irq_of_parse_and_map(np, 1); - r[2].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 1, &r[2]); r[3].name = gfar_err_intr; - r[3].start = r[3].end = irq_of_parse_and_map(np, 2); - r[3].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 2, &r[3]); n_res += 2; } @@ -323,8 +322,7 @@ static int __init fsl_i2c_of_init(void) if (ret) goto err; - r[1].start = r[1].end = irq_of_parse_and_map(np, 0); - r[1].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[1]); i2c_dev = platform_device_register_simple("fsl-i2c", i, r, 2); if (IS_ERR(i2c_dev)) { @@ -459,8 +457,7 @@ static int __init fsl_usb_of_init(void) if (ret) goto err; - r[1].start = r[1].end = irq_of_parse_and_map(np, 0); - r[1].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[1]); usb_dev_mph = platform_device_register_simple("fsl-ehci", i, r, 2); @@ -507,8 +504,7 @@ static int __init fsl_usb_of_init(void) if (ret) goto unreg_mph; - r[1].start = r[1].end = irq_of_parse_and_map(np, 0); - r[1].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[1]); usb_dev_dr = platform_device_register_simple("fsl-ehci", i, r, 2); @@ -591,8 +587,7 @@ static int __init fs_enet_of_init(void) r[2].name = fcc_regs_c; fs_enet_data.fcc_regs_c = r[2].start; - r[3].start = r[3].end = irq_of_parse_and_map(np, 0); - r[3].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[3]); fs_enet_dev = platform_device_register_simple("fsl-cpm-fcc", i, &r[0], 4); @@ -754,8 +749,7 @@ static int __init cpm_uart_of_init(void) goto err; r[1].name = scc_pram; - r[2].start = r[2].end = irq_of_parse_and_map(np, 0); - r[2].flags = IORESOURCE_IRQ; + of_irq_to_resource(np, 0, &r[2]); cpm_uart_dev = platform_device_register_simple("fsl-cpm-scc:uart", i, &r[0], 3); diff --git a/arch/ppc/platforms/83xx/mpc834x_sys.c b/arch/ppc/platforms/83xx/mpc834x_sys.c index 3397f0de1592..b84f8df325c4 100644 --- a/arch/ppc/platforms/83xx/mpc834x_sys.c +++ b/arch/ppc/platforms/83xx/mpc834x_sys.c @@ -121,8 +121,8 @@ mpc834x_sys_setup_arch(void) mdata->irq[0] = MPC83xx_IRQ_EXT1; mdata->irq[1] = MPC83xx_IRQ_EXT2; - mdata->irq[2] = -1; - mdata->irq[31] = -1; + mdata->irq[2] = PHY_POLL; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC83xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/mpc8540_ads.c b/arch/ppc/platforms/85xx/mpc8540_ads.c index 4f839da6782f..00a3ba57063f 100644 --- a/arch/ppc/platforms/85xx/mpc8540_ads.c +++ b/arch/ppc/platforms/85xx/mpc8540_ads.c @@ -92,9 +92,9 @@ mpc8540ads_setup_arch(void) mdata->irq[0] = MPC85xx_IRQ_EXT5; mdata->irq[1] = MPC85xx_IRQ_EXT5; - mdata->irq[2] = -1; + mdata->irq[2] = PHY_POLL; mdata->irq[3] = MPC85xx_IRQ_EXT5; - mdata->irq[31] = -1; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/mpc8560_ads.c b/arch/ppc/platforms/85xx/mpc8560_ads.c index 14ecec7bbed7..3a060468dd95 100644 --- a/arch/ppc/platforms/85xx/mpc8560_ads.c +++ b/arch/ppc/platforms/85xx/mpc8560_ads.c @@ -156,9 +156,9 @@ mpc8560ads_setup_arch(void) mdata->irq[0] = MPC85xx_IRQ_EXT5; mdata->irq[1] = MPC85xx_IRQ_EXT5; - mdata->irq[2] = -1; + mdata->irq[2] = PHY_POLL; mdata->irq[3] = MPC85xx_IRQ_EXT5; - mdata->irq[31] = -1; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c index 5ce0f69c1db6..2d59eb776c95 100644 --- a/arch/ppc/platforms/85xx/mpc85xx_cds_common.c +++ b/arch/ppc/platforms/85xx/mpc85xx_cds_common.c @@ -451,9 +451,9 @@ mpc85xx_cds_setup_arch(void) mdata->irq[0] = MPC85xx_IRQ_EXT5; mdata->irq[1] = MPC85xx_IRQ_EXT5; - mdata->irq[2] = -1; - mdata->irq[3] = -1; - mdata->irq[31] = -1; + mdata->irq[2] = PHY_POLL; + mdata->irq[3] = PHY_POLL; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/sbc8560.c b/arch/ppc/platforms/85xx/sbc8560.c index 764d580ff535..1d10ab98f66d 100644 --- a/arch/ppc/platforms/85xx/sbc8560.c +++ b/arch/ppc/platforms/85xx/sbc8560.c @@ -129,7 +129,7 @@ sbc8560_setup_arch(void) mdata->irq[25] = MPC85xx_IRQ_EXT6; mdata->irq[26] = MPC85xx_IRQ_EXT7; - mdata->irq[31] = -1; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/stx_gp3.c b/arch/ppc/platforms/85xx/stx_gp3.c index 4bb18ab27672..b1f5b737c70d 100644 --- a/arch/ppc/platforms/85xx/stx_gp3.c +++ b/arch/ppc/platforms/85xx/stx_gp3.c @@ -123,7 +123,7 @@ gp3_setup_arch(void) mdata->irq[2] = MPC85xx_IRQ_EXT5; mdata->irq[4] = MPC85xx_IRQ_EXT5; - mdata->irq[31] = -1; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/85xx/tqm85xx.c b/arch/ppc/platforms/85xx/tqm85xx.c index dd45f2e18449..4ee2bd156dc5 100644 --- a/arch/ppc/platforms/85xx/tqm85xx.c +++ b/arch/ppc/platforms/85xx/tqm85xx.c @@ -137,9 +137,9 @@ tqm85xx_setup_arch(void) mdata->irq[0] = MPC85xx_IRQ_EXT8; mdata->irq[1] = MPC85xx_IRQ_EXT8; - mdata->irq[2] = -1; + mdata->irq[2] = PHY_POLL; mdata->irq[3] = MPC85xx_IRQ_EXT8; - mdata->irq[31] = -1; + mdata->irq[31] = PHY_POLL; /* setup the board related information for the enet controllers */ pdata = (struct gianfar_platform_data *) ppc_sys_get_pdata(MPC85xx_TSEC1); diff --git a/arch/ppc/platforms/mpc8272ads_setup.c b/arch/ppc/platforms/mpc8272ads_setup.c index 1f9ea36837b1..0bc06768cf24 100644 --- a/arch/ppc/platforms/mpc8272ads_setup.c +++ b/arch/ppc/platforms/mpc8272ads_setup.c @@ -266,10 +266,10 @@ static void __init mpc8272ads_fixup_mdio_pdata(struct platform_device *pdev, int idx) { m82xx_mii_bb_pdata.irq[0] = PHY_INTERRUPT; - m82xx_mii_bb_pdata.irq[1] = -1; - m82xx_mii_bb_pdata.irq[2] = -1; + m82xx_mii_bb_pdata.irq[1] = PHY_POLL; + m82xx_mii_bb_pdata.irq[2] = PHY_POLL; m82xx_mii_bb_pdata.irq[3] = PHY_INTERRUPT; - m82xx_mii_bb_pdata.irq[31] = -1; + m82xx_mii_bb_pdata.irq[31] = PHY_POLL; m82xx_mii_bb_pdata.mdio_dat.offset = diff --git a/arch/ppc/platforms/mpc866ads_setup.c b/arch/ppc/platforms/mpc866ads_setup.c index e95d2c111747..8a0c07eb4449 100644 --- a/arch/ppc/platforms/mpc866ads_setup.c +++ b/arch/ppc/platforms/mpc866ads_setup.c @@ -361,7 +361,7 @@ int __init mpc866ads_init(void) fmpi->mii_speed = ((((bd->bi_intfreq + 4999999) / 2500000) / 2) & 0x3F) << 1; /* No PHY interrupt line here */ - fmpi->irq[0xf] = -1; + fmpi->irq[0xf] = PHY_POLL; /* Since either of the uarts could be used as console, they need to ready */ #ifdef CONFIG_SERIAL_CPM_SMC1 @@ -380,7 +380,7 @@ int __init mpc866ads_init(void) fmpi->mii_speed = ((((bd->bi_intfreq + 4999999) / 2500000) / 2) & 0x3F) << 1; /* No PHY interrupt line here */ - fmpi->irq[0xf] = -1; + fmpi->irq[0xf] = PHY_POLL; return 0; } diff --git a/drivers/net/phy/fixed.c b/drivers/net/phy/fixed.c index f14e99276dba..096d4a100bf2 100644 --- a/drivers/net/phy/fixed.c +++ b/drivers/net/phy/fixed.c @@ -254,7 +254,7 @@ static int fixed_mdio_register_device(int number, int speed, int duplex) goto device_create_fail; } - phydev->irq = -1; + phydev->irq = PHY_IGNORE_INTERRUPT; phydev->dev.bus = &mdio_bus_type; if(number) diff --git a/include/asm-powerpc/prom.h b/include/asm-powerpc/prom.h index 524629769336..39f04ef96f38 100644 --- a/include/asm-powerpc/prom.h +++ b/include/asm-powerpc/prom.h @@ -17,6 +17,7 @@ */ #include #include +#include #include /* Definitions used by the flattened device tree */ @@ -331,6 +332,12 @@ extern int of_irq_map_one(struct device_node *device, int index, struct pci_dev; extern int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_irq); +static inline void of_irq_to_resource(struct device_node *dev, int index, struct resource *r) +{ + r->start = r->end = irq_of_parse_and_map(dev, index); + r->flags = IORESOURCE_IRQ; +} + #endif /* __KERNEL__ */ #endif /* _POWERPC_PROM_H */ diff --git a/include/linux/phy.h b/include/linux/phy.h index 9447a57ee8a9..892d6abe57e5 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -20,6 +20,9 @@ #include #include +#include +#include +#include #define PHY_BASIC_FEATURES (SUPPORTED_10baseT_Half | \ SUPPORTED_10baseT_Full | \ -- cgit v1.2.3 From 2b4ac44e7c7e16cf9411b81693ff3e604f332bf1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Nov 2006 12:27:48 -0800 Subject: [PATCH] vmalloc: optimization, cleanup, bugfixes - reorder 'struct vm_struct' to speedup lookups on CPUS with small cache lines. The fields 'next,addr,size' should be now in the same cache line, to speedup lookups. - One minor cleanup in __get_vm_area_node() - Bugfixes in vmalloc_user() and vmalloc_32_user() NULL returns from __vmalloc() and __find_vm_area() were not tested. [akpm@osdl.org: remove redundant BUG_ONs] Signed-off-by: Eric Dumazet Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmalloc.h | 3 ++- mm/vmalloc.c | 26 +++++++++++++------------- 2 files changed, 15 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index dc9a29d84abc..924e502905d4 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -23,13 +23,14 @@ struct vm_area_struct; #endif struct vm_struct { + /* keep next,addr,size together to speedup lookups */ + struct vm_struct *next; void *addr; unsigned long size; unsigned long flags; struct page **pages; unsigned int nr_pages; unsigned long phys_addr; - struct vm_struct *next; }; /* diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 46606c133e82..7dc6aa745166 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -186,10 +186,8 @@ static struct vm_struct *__get_vm_area_node(unsigned long size, unsigned long fl if (unlikely(!area)) return NULL; - if (unlikely(!size)) { - kfree (area); + if (unlikely(!size)) return NULL; - } /* * We always allocate a guard page. @@ -532,11 +530,12 @@ void *vmalloc_user(unsigned long size) void *ret; ret = __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL); - write_lock(&vmlist_lock); - area = __find_vm_area(ret); - area->flags |= VM_USERMAP; - write_unlock(&vmlist_lock); - + if (ret) { + write_lock(&vmlist_lock); + area = __find_vm_area(ret); + area->flags |= VM_USERMAP; + write_unlock(&vmlist_lock); + } return ret; } EXPORT_SYMBOL(vmalloc_user); @@ -605,11 +604,12 @@ void *vmalloc_32_user(unsigned long size) void *ret; ret = __vmalloc(size, GFP_KERNEL | __GFP_ZERO, PAGE_KERNEL); - write_lock(&vmlist_lock); - area = __find_vm_area(ret); - area->flags |= VM_USERMAP; - write_unlock(&vmlist_lock); - + if (ret) { + write_lock(&vmlist_lock); + area = __find_vm_area(ret); + area->flags |= VM_USERMAP; + write_unlock(&vmlist_lock); + } return ret; } EXPORT_SYMBOL(vmalloc_32_user); -- cgit v1.2.3 From d8b295f29091310d746509bb6d5828aaf4907a18 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 10 Nov 2006 12:27:53 -0800 Subject: [PATCH] Fix missing parens in set_personality() If you call set_personality() with an expression such as: set_personality(foo ? PERS_FOO1 : PERS_FOO2); then this evaluates to: ((current->personality == foo ? PERS_FOO1 : PERS_FOO2) ? ... which is obviously not the intended result. Add the missing parents to ensure this gets evaluated as expected: ((current->personality == (foo ? PERS_FOO1 : PERS_FOO2)) ? ... Signed-off-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/personality.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/personality.h b/include/linux/personality.h index bf4cf2080e5c..012cd558189b 100644 --- a/include/linux/personality.h +++ b/include/linux/personality.h @@ -114,7 +114,7 @@ struct exec_domain { * Change personality of the currently running process. */ #define set_personality(pers) \ - ((current->personality == pers) ? 0 : __set_personality(pers)) + ((current->personality == (pers)) ? 0 : __set_personality(pers)) #endif /* __KERNEL__ */ -- cgit v1.2.3 From 68589bc353037f233fe510ad9ff432338c95db66 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 14 Nov 2006 02:03:32 -0800 Subject: [PATCH] hugetlb: prepare_hugepage_range check offset too (David:) If hugetlbfs_file_mmap() returns a failure to do_mmap_pgoff() - for example, because the given file offset is not hugepage aligned - then do_mmap_pgoff will go to the unmap_and_free_vma backout path. But at this stage the vma hasn't been marked as hugepage, and the backout path will call unmap_region() on it. That will eventually call down to the non-hugepage version of unmap_page_range(). On ppc64, at least, that will cause serious problems if there are any existing hugepage pagetable entries in the vicinity - for example if there are any other hugepage mappings under the same PUD. unmap_page_range() will trigger a bad_pud() on the hugepage pud entries. I suspect this will also cause bad problems on ia64, though I don't have a machine to test it on. (Hugh:) prepare_hugepage_range() should check file offset alignment when it checks virtual address and length, to stop MAP_FIXED with a bad huge offset from unmapping before it fails further down. PowerPC should apply the same prepare_hugepage_range alignment checks as ia64 and all the others do. Then none of the alignment checks in hugetlbfs_file_mmap are required (nor is the check for too small a mapping); but even so, move up setting of VM_HUGETLB and add a comment to warn of what David Gibson discovered - if hugetlbfs_file_mmap fails before setting it, do_mmap_pgoff's unmap_region when unwinding from error will go the non-huge way, which may cause bad behaviour on architectures (powerpc and ia64) which segregate their huge mappings into a separate region of the address space. Signed-off-by: Hugh Dickins Cc: "Luck, Tony" Cc: "David S. Miller" Acked-by: Adam Litke Acked-by: David Gibson Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/hugetlbpage.c | 4 +++- arch/powerpc/mm/hugetlbpage.c | 8 ++++++-- fs/hugetlbfs/inode.c | 21 ++++++++------------- include/linux/hugetlb.h | 10 +++++++--- mm/mmap.c | 2 +- 5 files changed, 25 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index eee5c1cfbe32..f3a9585e98a8 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -70,8 +70,10 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) * Don't actually need to do any preparation, but need to make sure * the address is in the right region. */ -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index fd68b74c07c3..506d89768d45 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -491,11 +491,15 @@ static int open_high_hpage_areas(struct mm_struct *mm, u16 newareas) return 0; } -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len, pgoff_t pgoff) { int err = 0; - if ( (addr+len) < addr ) + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) return -EINVAL; if (addr < 0x100000000UL) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 0bea6a619e10..7f4756963d05 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -62,24 +62,19 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) loff_t len, vma_len; int ret; - if (vma->vm_pgoff & (HPAGE_SIZE / PAGE_SIZE - 1)) - return -EINVAL; - - if (vma->vm_start & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end & ~HPAGE_MASK) - return -EINVAL; - - if (vma->vm_end - vma->vm_start < HPAGE_SIZE) - return -EINVAL; + /* + * vma alignment has already been checked by prepare_hugepage_range. + * If you add any error returns here, do so after setting VM_HUGETLB, + * so is_vm_hugetlb_page tests below unmap_region go the right way + * when do_mmap_pgoff unwinds (may be important on powerpc and ia64). + */ + vma->vm_flags |= VM_HUGETLB | VM_RESERVED; + vma->vm_ops = &hugetlb_vm_ops; vma_len = (loff_t)(vma->vm_end - vma->vm_start); mutex_lock(&inode->i_mutex); file_accessed(file); - vma->vm_flags |= VM_HUGETLB | VM_RESERVED; - vma->vm_ops = &hugetlb_vm_ops; ret = -ENOMEM; len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 5081d27bfa27..ace64e57e17f 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -60,8 +60,11 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff) { + if (pgoff & (~HPAGE_MASK >> PAGE_SHIFT)) + return -EINVAL; if (len & ~HPAGE_MASK) return -EINVAL; if (addr & ~HPAGE_MASK) @@ -69,7 +72,8 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) return 0; } #else -int prepare_hugepage_range(unsigned long addr, unsigned long len); +int prepare_hugepage_range(unsigned long addr, unsigned long len, + pgoff_t pgoff); #endif #ifndef ARCH_HAS_SETCLEAR_HUGE_PTE @@ -107,7 +111,7 @@ static inline unsigned long hugetlb_total_pages(void) #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL -#define prepare_hugepage_range(addr, len) (-EINVAL) +#define prepare_hugepage_range(addr,len,pgoff) (-EINVAL) #define pmd_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; }) diff --git a/mm/mmap.c b/mm/mmap.c index 497e502dfd6b..bdace87d7c01 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1379,7 +1379,7 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, * Check if the given range is hugepage aligned, and * can be made suitable for hugepages. */ - ret = prepare_hugepage_range(addr, len); + ret = prepare_hugepage_range(addr, len, pgoff); } else { /* * Ensure that a normal request is not falling in a -- cgit v1.2.3 From b96e7ecbd052a0916b6078e7600604d7e276a336 Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Tue, 14 Nov 2006 19:48:48 -0800 Subject: [NETFILTER]: ip6_tables: fixed conflicted optname for getsockopt 66 and 67 for getsockopt on IPv6 socket is doubly used for IPv6 Advanced API and ip6tables. This moves numbers for ip6tables to 68 and 69. This also kills XT_SO_* because {ip,ip6,arp}_tables doesn't have so much common numbers now. The old userland tools keep to behave as ever, because old kernel always calls functions of IPv6 Advanced API for their numbers. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/in6.h | 12 +++++++++++- include/linux/netfilter/x_tables.h | 16 ---------------- include/linux/netfilter_arp/arp_tables.h | 25 +++++++++++++------------ include/linux/netfilter_ipv4/ip_tables.h | 27 +++++++++++++++------------ include/linux/netfilter_ipv6/ip6_tables.h | 27 +++++++++++++++------------ 5 files changed, 54 insertions(+), 53 deletions(-) (limited to 'include/linux') diff --git a/include/linux/in6.h b/include/linux/in6.h index 9be6a4756f0b..f28621f638e0 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -225,7 +225,7 @@ struct in6_flowlabel_req #endif /* - * Netfilter + * Netfilter (1) * * Following socket options are used in ip6_tables; * see include/linux/netfilter_ipv6/ip6_tables.h. @@ -240,4 +240,14 @@ struct in6_flowlabel_req #define IPV6_RECVTCLASS 66 #define IPV6_TCLASS 67 +/* + * Netfilter (2) + * + * Following socket options are used in ip6_tables; + * see include/linux/netfilter_ipv6/ip6_tables.h. + * + * IP6T_SO_GET_REVISION_MATCH 68 + * IP6T_SO_GET_REVISION_TARGET 69 + */ + #endif diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 04319a76103a..022edfa97ed9 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -96,22 +96,6 @@ struct _xt_align /* Error verdict. */ #define XT_ERROR_TARGET "ERROR" -/* - * New IP firewall options for [gs]etsockopt at the RAW IP level. - * Unlike BSD Linux inherits IP options so you don't have to use a raw - * socket for this. Instead we check rights in the calls. */ -#define XT_BASE_CTL 64 /* base for firewall socket options */ - -#define XT_SO_SET_REPLACE (XT_BASE_CTL) -#define XT_SO_SET_ADD_COUNTERS (XT_BASE_CTL + 1) -#define XT_SO_SET_MAX XT_SO_SET_ADD_COUNTERS - -#define XT_SO_GET_INFO (XT_BASE_CTL) -#define XT_SO_GET_ENTRIES (XT_BASE_CTL + 1) -#define XT_SO_GET_REVISION_MATCH (XT_BASE_CTL + 2) -#define XT_SO_GET_REVISION_TARGET (XT_BASE_CTL + 3) -#define XT_SO_GET_MAX XT_SO_GET_REVISION_TARGET - #define SET_COUNTER(c,b,p) do { (c).bcnt = (b); (c).pcnt = (p); } while(0) #define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0) diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index 44e39b61d9e7..0be235418a2f 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -112,19 +112,20 @@ struct arpt_entry * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use a raw * socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in.h before adding new number here. */ -#define ARPT_CTL_OFFSET 32 -#define ARPT_BASE_CTL (XT_BASE_CTL+ARPT_CTL_OFFSET) - -#define ARPT_SO_SET_REPLACE (XT_SO_SET_REPLACE+ARPT_CTL_OFFSET) -#define ARPT_SO_SET_ADD_COUNTERS (XT_SO_SET_ADD_COUNTERS+ARPT_CTL_OFFSET) -#define ARPT_SO_SET_MAX (XT_SO_SET_MAX+ARPT_CTL_OFFSET) - -#define ARPT_SO_GET_INFO (XT_SO_GET_INFO+ARPT_CTL_OFFSET) -#define ARPT_SO_GET_ENTRIES (XT_SO_GET_ENTRIES+ARPT_CTL_OFFSET) -/* #define ARPT_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH */ -#define ARPT_SO_GET_REVISION_TARGET (XT_SO_GET_REVISION_TARGET+ARPT_CTL_OFFSET) -#define ARPT_SO_GET_MAX (XT_SO_GET_REVISION_TARGET+ARPT_CTL_OFFSET) +#define ARPT_BASE_CTL 96 + +#define ARPT_SO_SET_REPLACE (ARPT_BASE_CTL) +#define ARPT_SO_SET_ADD_COUNTERS (ARPT_BASE_CTL + 1) +#define ARPT_SO_SET_MAX ARPT_SO_SET_ADD_COUNTERS + +#define ARPT_SO_GET_INFO (ARPT_BASE_CTL) +#define ARPT_SO_GET_ENTRIES (ARPT_BASE_CTL + 1) +/* #define ARPT_SO_GET_REVISION_MATCH (APRT_BASE_CTL + 2) */ +#define ARPT_SO_GET_REVISION_TARGET (ARPT_BASE_CTL + 3) +#define ARPT_SO_GET_MAX (ARPT_SO_GET_REVISION_TARGET) /* CONTINUE verdict for targets */ #define ARPT_CONTINUE XT_CONTINUE diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index a536bbdef145..4f06dad0bde9 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -101,18 +101,21 @@ struct ipt_entry /* * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use a raw - * socket for this. Instead we check rights in the calls. */ -#define IPT_BASE_CTL XT_BASE_CTL - -#define IPT_SO_SET_REPLACE XT_SO_SET_REPLACE -#define IPT_SO_SET_ADD_COUNTERS XT_SO_SET_ADD_COUNTERS -#define IPT_SO_SET_MAX XT_SO_SET_MAX - -#define IPT_SO_GET_INFO XT_SO_GET_INFO -#define IPT_SO_GET_ENTRIES XT_SO_GET_ENTRIES -#define IPT_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH -#define IPT_SO_GET_REVISION_TARGET XT_SO_GET_REVISION_TARGET -#define IPT_SO_GET_MAX XT_SO_GET_REVISION_TARGET + * socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in.h before adding new number here. + */ +#define IPT_BASE_CTL 64 + +#define IPT_SO_SET_REPLACE (IPT_BASE_CTL) +#define IPT_SO_SET_ADD_COUNTERS (IPT_BASE_CTL + 1) +#define IPT_SO_SET_MAX IPT_SO_SET_ADD_COUNTERS + +#define IPT_SO_GET_INFO (IPT_BASE_CTL) +#define IPT_SO_GET_ENTRIES (IPT_BASE_CTL + 1) +#define IPT_SO_GET_REVISION_MATCH (IPT_BASE_CTL + 2) +#define IPT_SO_GET_REVISION_TARGET (IPT_BASE_CTL + 3) +#define IPT_SO_GET_MAX IPT_SO_GET_REVISION_TARGET #define IPT_CONTINUE XT_CONTINUE #define IPT_RETURN XT_RETURN diff --git a/include/linux/netfilter_ipv6/ip6_tables.h b/include/linux/netfilter_ipv6/ip6_tables.h index d7a8e9c0dad0..4aed340401db 100644 --- a/include/linux/netfilter_ipv6/ip6_tables.h +++ b/include/linux/netfilter_ipv6/ip6_tables.h @@ -107,18 +107,21 @@ struct ip6t_entry /* * New IP firewall options for [gs]etsockopt at the RAW IP level. * Unlike BSD Linux inherits IP options so you don't have to use - * a raw socket for this. Instead we check rights in the calls. */ -#define IP6T_BASE_CTL XT_BASE_CTL - -#define IP6T_SO_SET_REPLACE XT_SO_SET_REPLACE -#define IP6T_SO_SET_ADD_COUNTERS XT_SO_SET_ADD_COUNTERS -#define IP6T_SO_SET_MAX XT_SO_SET_MAX - -#define IP6T_SO_GET_INFO XT_SO_GET_INFO -#define IP6T_SO_GET_ENTRIES XT_SO_GET_ENTRIES -#define IP6T_SO_GET_REVISION_MATCH XT_SO_GET_REVISION_MATCH -#define IP6T_SO_GET_REVISION_TARGET XT_SO_GET_REVISION_TARGET -#define IP6T_SO_GET_MAX XT_SO_GET_REVISION_TARGET + * a raw socket for this. Instead we check rights in the calls. + * + * ATTENTION: check linux/in6.h before adding new number here. + */ +#define IP6T_BASE_CTL 64 + +#define IP6T_SO_SET_REPLACE (IP6T_BASE_CTL) +#define IP6T_SO_SET_ADD_COUNTERS (IP6T_BASE_CTL + 1) +#define IP6T_SO_SET_MAX IP6T_SO_SET_ADD_COUNTERS + +#define IP6T_SO_GET_INFO (IP6T_BASE_CTL) +#define IP6T_SO_GET_ENTRIES (IP6T_BASE_CTL + 1) +#define IP6T_SO_GET_REVISION_MATCH (IP6T_BASE_CTL + 4) +#define IP6T_SO_GET_REVISION_TARGET (IP6T_BASE_CTL + 5) +#define IP6T_SO_GET_MAX IP6T_SO_GET_REVISION_TARGET /* CONTINUE verdict for targets */ #define IP6T_CONTINUE XT_CONTINUE -- cgit v1.2.3 From c7835a77c86422d276b0d1a4c70924d933014c13 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 15 Nov 2006 21:14:42 -0800 Subject: [TG3]: Disable TSO on 5906 if CLKREQ is enabled. Due to hardware errata, TSO must be disabled if the PCI Express clock request is enabled on 5906. The chip may hang when transmitting TSO frames if CLKREQ is enabled. Update version to 3.69. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 20 ++++++++++++++++---- include/linux/pci_regs.h | 1 + 2 files changed, 17 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 6e86866bd3fa..1dbdd6bb587b 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -68,8 +68,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.68" -#define DRV_MODULE_RELDATE "November 02, 2006" +#define DRV_MODULE_VERSION "3.69" +#define DRV_MODULE_RELDATE "November 15, 2006" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -10366,7 +10366,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) u32 pci_state_reg, grc_misc_cfg; u32 val; u16 pci_cmd; - int err; + int err, pcie_cap; /* Force memory write invalidate off. If we leave it on, * then on 5700_BX chips we have to enable a workaround. @@ -10541,8 +10541,19 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) GET_ASIC_REV(tp->pci_chip_rev_id) != ASIC_REV_5906) tp->tg3_flags2 |= TG3_FLG2_JUMBO_CAPABLE; - if (pci_find_capability(tp->pdev, PCI_CAP_ID_EXP) != 0) + pcie_cap = pci_find_capability(tp->pdev, PCI_CAP_ID_EXP); + if (pcie_cap != 0) { tp->tg3_flags2 |= TG3_FLG2_PCI_EXPRESS; + if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) { + u16 lnkctl; + + pci_read_config_word(tp->pdev, + pcie_cap + PCI_EXP_LNKCTL, + &lnkctl); + if (lnkctl & PCI_EXP_LNKCTL_CLKREQ_EN) + tp->tg3_flags2 &= ~TG3_FLG2_HW_TSO_2; + } + } /* If we have an AMD 762 or VIA K8T800 chipset, write * reordering to the mailbox registers done by the host @@ -11809,6 +11820,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, else if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5700 || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5701 || tp->pci_chip_rev_id == CHIPREV_ID_5705_A0 || + GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906 || (tp->tg3_flags & TG3_FLAG_ENABLE_ASF) != 0) { tp->tg3_flags2 &= ~TG3_FLG2_TSO_CAPABLE; } else { diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index c312a12ad2d6..c321316f1bc7 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -371,6 +371,7 @@ #define PCI_EXP_DEVSTA_TRPND 0x20 /* Transactions Pending */ #define PCI_EXP_LNKCAP 12 /* Link Capabilities */ #define PCI_EXP_LNKCTL 16 /* Link Control */ +#define PCI_EXP_LNKCTL_CLKREQ_EN 0x100 /* Enable clkreq */ #define PCI_EXP_LNKSTA 18 /* Link Status */ #define PCI_EXP_SLTCAP 20 /* Slot Capabilities */ #define PCI_EXP_SLTCTL 24 /* Slot Control */ -- cgit v1.2.3 From da63fc7ce63b43426dc3c69c05e28de2872c159a Mon Sep 17 00:00:00 2001 From: OGAWA Hirofumi Date: Thu, 16 Nov 2006 01:19:28 -0800 Subject: [PATCH] fat: add fat_getattr() This adds fat_getattr() for setting stat->blksize. (FAT uses the size of cluster for proper I/O) Signed-off-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fat/file.c | 10 ++++++++++ fs/msdos/namei.c | 1 + fs/vfat/namei.c | 1 + include/linux/msdos_fs.h | 2 ++ 4 files changed, 14 insertions(+) (limited to 'include/linux') diff --git a/fs/fat/file.c b/fs/fat/file.c index 8337451e7897..0aa813d944a6 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -303,7 +303,17 @@ void fat_truncate(struct inode *inode) fat_flush_inodes(inode->i_sb, inode, NULL); } +int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + generic_fillattr(inode, stat); + stat->blksize = MSDOS_SB(inode->i_sb)->cluster_size; + return 0; +} +EXPORT_SYMBOL_GPL(fat_getattr); + struct inode_operations fat_file_inode_operations = { .truncate = fat_truncate, .setattr = fat_notify_change, + .getattr = fat_getattr, }; diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c index b0f01b3b0536..452461955cbd 100644 --- a/fs/msdos/namei.c +++ b/fs/msdos/namei.c @@ -654,6 +654,7 @@ static struct inode_operations msdos_dir_inode_operations = { .rmdir = msdos_rmdir, .rename = msdos_rename, .setattr = fat_notify_change, + .getattr = fat_getattr, }; static int msdos_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index edb711ff7b05..0afd745a37cd 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -1004,6 +1004,7 @@ static struct inode_operations vfat_dir_inode_operations = { .rmdir = vfat_rmdir, .rename = vfat_rename, .setattr = fat_notify_change, + .getattr = fat_getattr, }; static int vfat_fill_super(struct super_block *sb, void *data, int silent) diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h index ce6c85815cbd..24a9ef1506b6 100644 --- a/include/linux/msdos_fs.h +++ b/include/linux/msdos_fs.h @@ -402,6 +402,8 @@ extern const struct file_operations fat_file_operations; extern struct inode_operations fat_file_inode_operations; extern int fat_notify_change(struct dentry * dentry, struct iattr * attr); extern void fat_truncate(struct inode *inode); +extern int fat_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat); /* fat/inode.c */ extern void fat_attach(struct inode *inode, loff_t i_pos); -- cgit v1.2.3 From 9d92fe17b652f5496c97bc83fdfe925f3182f602 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Fri, 17 Nov 2006 01:07:39 -0500 Subject: Input: serio - remove serio_unregister_port_delayed() Now that i8042 reserves IRQs early there are no more users of this function and it should be removed. Signed-off-by: Dmitry Torokhov --- drivers/input/serio/serio.c | 18 ------------------ include/linux/serio.h | 5 ----- 2 files changed, 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 3cb99d454ecd..cd55ddb7df4f 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -44,7 +44,6 @@ EXPORT_SYMBOL(serio_interrupt); EXPORT_SYMBOL(__serio_register_port); EXPORT_SYMBOL(serio_unregister_port); EXPORT_SYMBOL(serio_unregister_child_port); -EXPORT_SYMBOL(__serio_unregister_port_delayed); EXPORT_SYMBOL(__serio_register_driver); EXPORT_SYMBOL(serio_unregister_driver); EXPORT_SYMBOL(serio_open); @@ -64,7 +63,6 @@ static struct bus_type serio_bus; static void serio_add_driver(struct serio_driver *drv); static void serio_add_port(struct serio *serio); -static void serio_destroy_port(struct serio *serio); static void serio_reconnect_port(struct serio *serio); static void serio_disconnect_port(struct serio *serio); @@ -173,7 +171,6 @@ enum serio_event_type { SERIO_RESCAN, SERIO_RECONNECT, SERIO_REGISTER_PORT, - SERIO_UNREGISTER_PORT, SERIO_REGISTER_DRIVER, }; @@ -307,11 +304,6 @@ static void serio_handle_event(void) serio_add_port(event->object); break; - case SERIO_UNREGISTER_PORT: - serio_disconnect_port(event->object); - serio_destroy_port(event->object); - break; - case SERIO_RECONNECT: serio_reconnect_port(event->object); break; @@ -716,16 +708,6 @@ void serio_unregister_child_port(struct serio *serio) mutex_unlock(&serio_mutex); } -/* - * Submits register request to kseriod for subsequent execution. - * Can be used when it is not obvious whether the serio_mutex is - * taken or not and when delayed execution is feasible. - */ -void __serio_unregister_port_delayed(struct serio *serio, struct module *owner) -{ - serio_queue_event(serio, owner, SERIO_UNREGISTER_PORT); -} - /* * Serio driver operations diff --git a/include/linux/serio.h b/include/linux/serio.h index b99c5ca9708d..8f52228390ab 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -85,11 +85,6 @@ static inline void serio_register_port(struct serio *serio) void serio_unregister_port(struct serio *serio); void serio_unregister_child_port(struct serio *serio); -void __serio_unregister_port_delayed(struct serio *serio, struct module *owner); -static inline void serio_unregister_port_delayed(struct serio *serio) -{ - __serio_unregister_port_delayed(serio, THIS_MODULE); -} void __serio_register_driver(struct serio_driver *drv, struct module *owner); static inline void serio_register_driver(struct serio_driver *drv) -- cgit v1.2.3 From 610a5b742e9df4e59047f22d13d8bd83cafce388 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Fri, 17 Nov 2006 11:51:41 +1100 Subject: [CRYPTO] api: Remove one too many semicolon This patch has removed one too many semicolon in crypto.h. Signed-off-by: Yoichi Yuasa Signed-off-by: Herbert Xu --- include/linux/crypto.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 8f2ffa4caabf..6485e9716b36 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -245,7 +245,7 @@ int crypto_alg_available(const char *name, u32 flags) __deprecated_for_modules; int crypto_has_alg(const char *name, u32 type, u32 mask); #else -static int crypto_alg_available(const char *name, u32 flags); +static int crypto_alg_available(const char *name, u32 flags) __deprecated_for_modules; static inline int crypto_alg_available(const char *name, u32 flags) { -- cgit v1.2.3 From b3438f8266cb1f5010085ac47d7ad6a36a212164 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 20 Nov 2006 11:47:18 -0800 Subject: Add "pure_initcall" for static variable initialization This is a quick hack to overcome the fact that SRCU currently does not allow static initializers, and we need to sometimes initialize those things before any other initializers (even "core" ones) can do so. Currently we don't allow this at all for modules, and the only user that needs is right now is cpufreq. As reported by Thomas Gleixner: "Commit b4dfdbb3c707474a2254c5b4d7e62be31a4b7da9 ("[PATCH] cpufreq: make the transition_notifier chain use SRCU breaks cpu frequency notification users, which register the callback > on core_init level." Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Arjan van de Ven Cc: Andrew Morton , Signed-off-by: Linus Torvalds --- drivers/cpufreq/cpufreq.c | 2 +- include/asm-generic/vmlinux.lds.h | 2 ++ include/linux/init.h | 8 ++++++++ 3 files changed, 11 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 86e69b7f9122..dd0c2623e27b 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -59,7 +59,7 @@ static int __init init_cpufreq_transition_notifier_list(void) srcu_init_notifier_head(&cpufreq_transition_notifier_list); return 0; } -core_initcall(init_cpufreq_transition_notifier_list); +pure_initcall(init_cpufreq_transition_notifier_list); static LIST_HEAD(cpufreq_governor_list); static DEFINE_MUTEX (cpufreq_governor_mutex); diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 9d873163a7ab..e60d6f21fa62 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -215,6 +215,8 @@ .notes : { *(.note.*) } :note #define INITCALLS \ + *(.initcall0.init) \ + *(.initcall0s.init) \ *(.initcall1.init) \ *(.initcall1s.init) \ *(.initcall2.init) \ diff --git a/include/linux/init.h b/include/linux/init.h index ff40ea118e3a..5eb5d24b7680 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -93,6 +93,14 @@ extern void setup_arch(char **); static initcall_t __initcall_##fn##id __attribute_used__ \ __attribute__((__section__(".initcall" level ".init"))) = fn +/* + * A "pure" initcall has no dependencies on anything else, and purely + * initializes variables that couldn't be statically initialized. + * + * This only exists for built-in code, not for modules. + */ +#define pure_initcall(fn) __define_initcall("0",fn,1) + #define core_initcall(fn) __define_initcall("1",fn,1) #define core_initcall_sync(fn) __define_initcall("1s",fn,1s) #define postcore_initcall(fn) __define_initcall("2",fn,2) -- cgit v1.2.3 From fb47ddb2db9c18664bd7b06c201a2398885b64fc Mon Sep 17 00:00:00 2001 From: David L Stevens Date: Sun, 19 Nov 2006 10:38:39 -0800 Subject: [IGMP]: Fix IGMPV3_EXP() normalization bit shift value. The IGMPV3_EXP() macro doesn't correctly shift the normalization bit, so time-out values are longer than they should be. Thanks to Dirk Ooms for finding the problem in IGMPv3 - MLDv2 had a similar problem that was already fixed a year ago. :-( Signed-off-by: David L Stevens Signed-off-by: David S. Miller --- include/linux/igmp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 03f43e2893a4..21dd56905271 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -191,7 +191,7 @@ struct ip_mc_list #define IGMPV3_MASK(value, nb) ((nb)>=32 ? (value) : ((1<<(nb))-1) & (value)) #define IGMPV3_EXP(thresh, nbmant, nbexp, value) \ ((value) < (thresh) ? (value) : \ - ((IGMPV3_MASK(value, nbmant) | (1<<(nbmant+nbexp))) << \ + ((IGMPV3_MASK(value, nbmant) | (1<<(nbmant))) << \ (IGMPV3_MASK((value) >> (nbmant), nbexp) + (nbexp)))) #define IGMPV3_QQIC(value) IGMPV3_EXP(0x80, 4, 3, value) -- cgit v1.2.3 From 52bad64d95bd89e08c49ec5a071fa6dcbe5a1a9c Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:54:01 +0000 Subject: WorkStruct: Separate delayable and non-delayable events. Separate delayable work items from non-delayable work items be splitting them into a separate structure (delayed_work), which incorporates a work_struct and the timer_list removed from work_struct. The work_struct struct is huge, and this limits it's usefulness. On a 64-bit architecture it's nearly 100 bytes in size. This reduces that by half for the non-delayable type of event. Signed-Off-By: David Howells --- arch/x86_64/kernel/mce.c | 2 +- drivers/ata/libata-core.c | 11 +++----- drivers/ata/libata-eh.c | 2 +- drivers/char/random.c | 2 +- drivers/char/tty_io.c | 2 +- fs/aio.c | 4 +-- fs/nfs/client.c | 2 +- fs/nfs/namespace.c | 3 ++- include/linux/aio.h | 2 +- include/linux/kbd_kern.h | 2 +- include/linux/libata.h | 4 +-- include/linux/nfs_fs_sb.h | 2 +- include/linux/sunrpc/rpc_pipe_fs.h | 2 +- include/linux/sunrpc/xprt.h | 2 +- include/linux/tty.h | 2 +- include/linux/workqueue.h | 44 +++++++++++++++++++++++--------- kernel/workqueue.c | 51 +++++++++++++++++++++----------------- mm/slab.c | 8 +++--- net/core/link_watch.c | 9 +++---- net/sunrpc/cache.c | 4 +-- net/sunrpc/rpc_pipe.c | 3 ++- net/sunrpc/xprtsock.c | 6 ++--- 22 files changed, 96 insertions(+), 73 deletions(-) (limited to 'include/linux') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index bbea88801d88..5306f2630905 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -307,7 +307,7 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) static int check_interval = 5 * 60; /* 5 minutes */ static void mcheck_timer(void *data); -static DECLARE_WORK(mcheck_work, mcheck_timer, NULL); +static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer, NULL); static void mcheck_check_cpu(void *info) { diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 915a55a6cc14..0bb4b4dced76 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -937,12 +937,9 @@ void ata_port_queue_task(struct ata_port *ap, void (*fn)(void *), void *data, if (ap->pflags & ATA_PFLAG_FLUSH_PORT_TASK) return; - PREPARE_WORK(&ap->port_task, fn, data); + PREPARE_DELAYED_WORK(&ap->port_task, fn, data); - if (!delay) - rc = queue_work(ata_wq, &ap->port_task); - else - rc = queue_delayed_work(ata_wq, &ap->port_task, delay); + rc = queue_delayed_work(ata_wq, &ap->port_task, delay); /* rc == 0 means that another user is using port task */ WARN_ON(rc == 0); @@ -5320,8 +5317,8 @@ void ata_port_init(struct ata_port *ap, struct ata_host *host, ap->msg_enable = ATA_MSG_DRV | ATA_MSG_ERR | ATA_MSG_WARN; #endif - INIT_WORK(&ap->port_task, NULL, NULL); - INIT_WORK(&ap->hotplug_task, ata_scsi_hotplug, ap); + INIT_DELAYED_WORK(&ap->port_task, NULL, NULL); + INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug, ap); INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan, ap); INIT_LIST_HEAD(&ap->eh_done_q); init_waitqueue_head(&ap->eh_wait_q); diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 02b2b2787d9b..9f6b7cc74fd9 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -332,7 +332,7 @@ void ata_scsi_error(struct Scsi_Host *host) if (ap->pflags & ATA_PFLAG_LOADING) ap->pflags &= ~ATA_PFLAG_LOADING; else if (ap->pflags & ATA_PFLAG_SCSI_HOTPLUG) - queue_work(ata_aux_wq, &ap->hotplug_task); + queue_delayed_work(ata_aux_wq, &ap->hotplug_task, 0); if (ap->pflags & ATA_PFLAG_RECOVERED) ata_port_printk(ap, KERN_INFO, "EH complete\n"); diff --git a/drivers/char/random.c b/drivers/char/random.c index eb6b13f4211a..f2ab61f3e8ae 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1424,7 +1424,7 @@ static unsigned int ip_cnt; static void rekey_seq_generator(void *private_); -static DECLARE_WORK(rekey_work, rekey_seq_generator, NULL); +static DECLARE_DELAYED_WORK(rekey_work, rekey_seq_generator, NULL); /* * Lock avoidance: diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index e90ea39c7c4b..7297acfe520c 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -3580,7 +3580,7 @@ static void initialize_tty_struct(struct tty_struct *tty) tty->overrun_time = jiffies; tty->buf.head = tty->buf.tail = NULL; tty_buffer_init(tty); - INIT_WORK(&tty->buf.work, flush_to_ldisc, tty); + INIT_DELAYED_WORK(&tty->buf.work, flush_to_ldisc, tty); init_MUTEX(&tty->buf.pty_sem); mutex_init(&tty->termios_mutex); init_waitqueue_head(&tty->write_wait); diff --git a/fs/aio.c b/fs/aio.c index 94766599db00..11a1a7100ad6 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -227,7 +227,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) INIT_LIST_HEAD(&ctx->active_reqs); INIT_LIST_HEAD(&ctx->run_list); - INIT_WORK(&ctx->wq, aio_kick_handler, ctx); + INIT_DELAYED_WORK(&ctx->wq, aio_kick_handler, ctx); if (aio_setup_ring(ctx) < 0) goto out_freectx; @@ -876,7 +876,7 @@ static void aio_kick_handler(void *data) * we're in a worker thread already, don't use queue_delayed_work, */ if (requeue) - queue_work(aio_wq, &ctx->wq); + queue_delayed_work(aio_wq, &ctx->wq, 0); } diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 5fea638743e4..6f0487d6f44a 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -143,7 +143,7 @@ static struct nfs_client *nfs_alloc_client(const char *hostname, INIT_LIST_HEAD(&clp->cl_state_owners); INIT_LIST_HEAD(&clp->cl_unused); spin_lock_init(&clp->cl_lock); - INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp); + INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state, clp); rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); clp->cl_boot_time = CURRENT_TIME; clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index ec1114b33d89..5ed798bc1cf7 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -21,7 +21,8 @@ static void nfs_expire_automounts(void *list); LIST_HEAD(nfs_automount_list); -static DECLARE_WORK(nfs_automount_task, nfs_expire_automounts, &nfs_automount_list); +static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts, + &nfs_automount_list); int nfs_mountpoint_expiry_timeout = 500 * HZ; static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent, diff --git a/include/linux/aio.h b/include/linux/aio.h index 0d71c0041f13..9e350fd44d77 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -194,7 +194,7 @@ struct kioctx { struct aio_ring_info ring_info; - struct work_struct wq; + struct delayed_work wq; }; /* prototypes */ diff --git a/include/linux/kbd_kern.h b/include/linux/kbd_kern.h index efe0ee4cc80b..06c58c423fe1 100644 --- a/include/linux/kbd_kern.h +++ b/include/linux/kbd_kern.h @@ -158,7 +158,7 @@ static inline void con_schedule_flip(struct tty_struct *t) if (t->buf.tail != NULL) t->buf.tail->commit = t->buf.tail->used; spin_unlock_irqrestore(&t->buf.lock, flags); - schedule_work(&t->buf.work); + schedule_delayed_work(&t->buf.work, 0); } #endif diff --git a/include/linux/libata.h b/include/linux/libata.h index abd2debebca2..5f04006e8dd2 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -568,8 +568,8 @@ struct ata_port { struct ata_host *host; struct device *dev; - struct work_struct port_task; - struct work_struct hotplug_task; + struct delayed_work port_task; + struct delayed_work hotplug_task; struct work_struct scsi_rescan_task; unsigned int hsm_task_state; diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 7ccfc7ef0a83..95796e6924f1 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -51,7 +51,7 @@ struct nfs_client { unsigned long cl_lease_time; unsigned long cl_last_renewal; - struct work_struct cl_renewd; + struct delayed_work cl_renewd; struct rpc_wait_queue cl_rpcwaitq; diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h index a2eb9b4a9de3..4a68125b6de6 100644 --- a/include/linux/sunrpc/rpc_pipe_fs.h +++ b/include/linux/sunrpc/rpc_pipe_fs.h @@ -30,7 +30,7 @@ struct rpc_inode { #define RPC_PIPE_WAIT_FOR_OPEN 1 int flags; struct rpc_pipe_ops *ops; - struct work_struct queue_timeout; + struct delayed_work queue_timeout; }; static inline struct rpc_inode * diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 60394fbc4c70..3e04c1512fc4 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -177,7 +177,7 @@ struct rpc_xprt { unsigned long connect_timeout, bind_timeout, reestablish_timeout; - struct work_struct connect_worker; + struct delayed_work connect_worker; unsigned short port; /* diff --git a/include/linux/tty.h b/include/linux/tty.h index 44091c0db0b4..c1f716446161 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -53,7 +53,7 @@ struct tty_buffer { }; struct tty_bufhead { - struct work_struct work; + struct delayed_work work; struct semaphore pty_sem; spinlock_t lock; struct tty_buffer *head; /* Queue head */ diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 9bca3539a1e5..9faaccae570e 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -17,6 +17,10 @@ struct work_struct { void (*func)(void *); void *data; void *wq_data; +}; + +struct delayed_work { + struct work_struct work; struct timer_list timer; }; @@ -28,32 +32,48 @@ struct execute_work { .entry = { &(n).entry, &(n).entry }, \ .func = (f), \ .data = (d), \ + } + +#define __DELAYED_WORK_INITIALIZER(n, f, d) { \ + .work = __WORK_INITIALIZER((n).work, (f), (d)), \ .timer = TIMER_INITIALIZER(NULL, 0, 0), \ } #define DECLARE_WORK(n, f, d) \ struct work_struct n = __WORK_INITIALIZER(n, f, d) +#define DECLARE_DELAYED_WORK(n, f, d) \ + struct delayed_work n = __DELAYED_WORK_INITIALIZER(n, f, d) + /* - * initialize a work-struct's func and data pointers: + * initialize a work item's function and data pointers */ #define PREPARE_WORK(_work, _func, _data) \ do { \ - (_work)->func = _func; \ - (_work)->data = _data; \ + (_work)->func = (_func); \ + (_work)->data = (_data); \ } while (0) +#define PREPARE_DELAYED_WORK(_work, _func, _data) \ + PREPARE_WORK(&(_work)->work, (_func), (_data)) + /* - * initialize all of a work-struct: + * initialize all of a work item in one go */ #define INIT_WORK(_work, _func, _data) \ do { \ INIT_LIST_HEAD(&(_work)->entry); \ (_work)->pending = 0; \ PREPARE_WORK((_work), (_func), (_data)); \ + } while (0) + +#define INIT_DELAYED_WORK(_work, _func, _data) \ + do { \ + INIT_WORK(&(_work)->work, (_func), (_data)); \ init_timer(&(_work)->timer); \ } while (0) + extern struct workqueue_struct *__create_workqueue(const char *name, int singlethread); #define create_workqueue(name) __create_workqueue((name), 0) @@ -62,24 +82,24 @@ extern struct workqueue_struct *__create_workqueue(const char *name, extern void destroy_workqueue(struct workqueue_struct *wq); extern int FASTCALL(queue_work(struct workqueue_struct *wq, struct work_struct *work)); -extern int FASTCALL(queue_delayed_work(struct workqueue_struct *wq, struct work_struct *work, unsigned long delay)); +extern int FASTCALL(queue_delayed_work(struct workqueue_struct *wq, struct delayed_work *work, unsigned long delay)); extern int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, - struct work_struct *work, unsigned long delay); + struct delayed_work *work, unsigned long delay); extern void FASTCALL(flush_workqueue(struct workqueue_struct *wq)); extern int FASTCALL(schedule_work(struct work_struct *work)); -extern int FASTCALL(schedule_delayed_work(struct work_struct *work, unsigned long delay)); +extern int FASTCALL(schedule_delayed_work(struct delayed_work *work, unsigned long delay)); -extern int schedule_delayed_work_on(int cpu, struct work_struct *work, unsigned long delay); +extern int schedule_delayed_work_on(int cpu, struct delayed_work *work, unsigned long delay); extern int schedule_on_each_cpu(void (*func)(void *info), void *info); extern void flush_scheduled_work(void); extern int current_is_keventd(void); extern int keventd_up(void); extern void init_workqueues(void); -void cancel_rearming_delayed_work(struct work_struct *work); +void cancel_rearming_delayed_work(struct delayed_work *work); void cancel_rearming_delayed_workqueue(struct workqueue_struct *, - struct work_struct *); + struct delayed_work *); int execute_in_process_context(void (*fn)(void *), void *, struct execute_work *); @@ -88,13 +108,13 @@ int execute_in_process_context(void (*fn)(void *), void *, * function may still be running on return from cancel_delayed_work(). Run * flush_scheduled_work() to wait on it. */ -static inline int cancel_delayed_work(struct work_struct *work) +static inline int cancel_delayed_work(struct delayed_work *work) { int ret; ret = del_timer_sync(&work->timer); if (ret) - clear_bit(0, &work->pending); + clear_bit(0, &work->work.pending); return ret; } diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 17c2f03d2c27..44fc54b7decf 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -122,29 +122,33 @@ EXPORT_SYMBOL_GPL(queue_work); static void delayed_work_timer_fn(unsigned long __data) { - struct work_struct *work = (struct work_struct *)__data; - struct workqueue_struct *wq = work->wq_data; + struct delayed_work *dwork = (struct delayed_work *)__data; + struct workqueue_struct *wq = dwork->work.wq_data; int cpu = smp_processor_id(); if (unlikely(is_single_threaded(wq))) cpu = singlethread_cpu; - __queue_work(per_cpu_ptr(wq->cpu_wq, cpu), work); + __queue_work(per_cpu_ptr(wq->cpu_wq, cpu), &dwork->work); } /** * queue_delayed_work - queue work on a workqueue after delay * @wq: workqueue to use - * @work: work to queue + * @work: delayable work to queue * @delay: number of jiffies to wait before queueing * * Returns 0 if @work was already on a queue, non-zero otherwise. */ int fastcall queue_delayed_work(struct workqueue_struct *wq, - struct work_struct *work, unsigned long delay) + struct delayed_work *dwork, unsigned long delay) { int ret = 0; - struct timer_list *timer = &work->timer; + struct timer_list *timer = &dwork->timer; + struct work_struct *work = &dwork->work; + + if (delay == 0) + return queue_work(wq, work); if (!test_and_set_bit(0, &work->pending)) { BUG_ON(timer_pending(timer)); @@ -153,7 +157,7 @@ int fastcall queue_delayed_work(struct workqueue_struct *wq, /* This stores wq for the moment, for the timer_fn */ work->wq_data = wq; timer->expires = jiffies + delay; - timer->data = (unsigned long)work; + timer->data = (unsigned long)dwork; timer->function = delayed_work_timer_fn; add_timer(timer); ret = 1; @@ -172,10 +176,11 @@ EXPORT_SYMBOL_GPL(queue_delayed_work); * Returns 0 if @work was already on a queue, non-zero otherwise. */ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, - struct work_struct *work, unsigned long delay) + struct delayed_work *dwork, unsigned long delay) { int ret = 0; - struct timer_list *timer = &work->timer; + struct timer_list *timer = &dwork->timer; + struct work_struct *work = &dwork->work; if (!test_and_set_bit(0, &work->pending)) { BUG_ON(timer_pending(timer)); @@ -184,7 +189,7 @@ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, /* This stores wq for the moment, for the timer_fn */ work->wq_data = wq; timer->expires = jiffies + delay; - timer->data = (unsigned long)work; + timer->data = (unsigned long)dwork; timer->function = delayed_work_timer_fn; add_timer_on(timer, cpu); ret = 1; @@ -468,31 +473,31 @@ EXPORT_SYMBOL(schedule_work); /** * schedule_delayed_work - put work task in global workqueue after delay - * @work: job to be done - * @delay: number of jiffies to wait + * @dwork: job to be done + * @delay: number of jiffies to wait or 0 for immediate execution * * After waiting for a given time this puts a job in the kernel-global * workqueue. */ -int fastcall schedule_delayed_work(struct work_struct *work, unsigned long delay) +int fastcall schedule_delayed_work(struct delayed_work *dwork, unsigned long delay) { - return queue_delayed_work(keventd_wq, work, delay); + return queue_delayed_work(keventd_wq, dwork, delay); } EXPORT_SYMBOL(schedule_delayed_work); /** * schedule_delayed_work_on - queue work in global workqueue on CPU after delay * @cpu: cpu to use - * @work: job to be done + * @dwork: job to be done * @delay: number of jiffies to wait * * After waiting for a given time this puts a job in the kernel-global * workqueue on the specified CPU. */ int schedule_delayed_work_on(int cpu, - struct work_struct *work, unsigned long delay) + struct delayed_work *dwork, unsigned long delay) { - return queue_delayed_work_on(cpu, keventd_wq, work, delay); + return queue_delayed_work_on(cpu, keventd_wq, dwork, delay); } EXPORT_SYMBOL(schedule_delayed_work_on); @@ -539,12 +544,12 @@ EXPORT_SYMBOL(flush_scheduled_work); * cancel_rearming_delayed_workqueue - reliably kill off a delayed * work whose handler rearms the delayed work. * @wq: the controlling workqueue structure - * @work: the delayed work struct + * @dwork: the delayed work struct */ void cancel_rearming_delayed_workqueue(struct workqueue_struct *wq, - struct work_struct *work) + struct delayed_work *dwork) { - while (!cancel_delayed_work(work)) + while (!cancel_delayed_work(dwork)) flush_workqueue(wq); } EXPORT_SYMBOL(cancel_rearming_delayed_workqueue); @@ -552,11 +557,11 @@ EXPORT_SYMBOL(cancel_rearming_delayed_workqueue); /** * cancel_rearming_delayed_work - reliably kill off a delayed keventd * work whose handler rearms the delayed work. - * @work: the delayed work struct + * @dwork: the delayed work struct */ -void cancel_rearming_delayed_work(struct work_struct *work) +void cancel_rearming_delayed_work(struct delayed_work *dwork) { - cancel_rearming_delayed_workqueue(keventd_wq, work); + cancel_rearming_delayed_workqueue(keventd_wq, dwork); } EXPORT_SYMBOL(cancel_rearming_delayed_work); diff --git a/mm/slab.c b/mm/slab.c index 3c4a7e34eddc..a65bc5e992c3 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -753,7 +753,7 @@ int slab_is_available(void) return g_cpucache_up == FULL; } -static DEFINE_PER_CPU(struct work_struct, reap_work); +static DEFINE_PER_CPU(struct delayed_work, reap_work); static inline struct array_cache *cpu_cache_get(struct kmem_cache *cachep) { @@ -916,16 +916,16 @@ static void next_reap_node(void) */ static void __devinit start_cpu_timer(int cpu) { - struct work_struct *reap_work = &per_cpu(reap_work, cpu); + struct delayed_work *reap_work = &per_cpu(reap_work, cpu); /* * When this gets called from do_initcalls via cpucache_init(), * init_workqueues() has already run, so keventd will be setup * at that time. */ - if (keventd_up() && reap_work->func == NULL) { + if (keventd_up() && reap_work->work.func == NULL) { init_reap_node(cpu); - INIT_WORK(reap_work, cache_reap, NULL); + INIT_DELAYED_WORK(reap_work, cache_reap, NULL); schedule_delayed_work_on(cpu, reap_work, HZ + 3 * cpu); } } diff --git a/net/core/link_watch.c b/net/core/link_watch.c index 4b36114744c5..f2ed09e25dfd 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -35,7 +35,7 @@ static unsigned long linkwatch_flags; static unsigned long linkwatch_nextevent; static void linkwatch_event(void *dummy); -static DECLARE_WORK(linkwatch_work, linkwatch_event, NULL); +static DECLARE_DELAYED_WORK(linkwatch_work, linkwatch_event, NULL); static LIST_HEAD(lweventlist); static DEFINE_SPINLOCK(lweventlist_lock); @@ -171,10 +171,9 @@ void linkwatch_fire_event(struct net_device *dev) unsigned long delay = linkwatch_nextevent - jiffies; /* If we wrap around we'll delay it by at most HZ. */ - if (!delay || delay > HZ) - schedule_work(&linkwatch_work); - else - schedule_delayed_work(&linkwatch_work, delay); + if (delay > HZ) + delay = 0; + schedule_delayed_work(&linkwatch_work, delay); } } } diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 00cb388ece03..d5725cb1491e 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -285,7 +285,7 @@ static struct file_operations content_file_operations; static struct file_operations cache_flush_operations; static void do_cache_clean(void *data); -static DECLARE_WORK(cache_cleaner, do_cache_clean, NULL); +static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean, NULL); void cache_register(struct cache_detail *cd) { @@ -337,7 +337,7 @@ void cache_register(struct cache_detail *cd) spin_unlock(&cache_list_lock); /* start the cleaning process */ - schedule_work(&cache_cleaner); + schedule_delayed_work(&cache_cleaner, 0); } int cache_unregister(struct cache_detail *cd) diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 9a0b41a97f90..97be3f7fed44 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -837,7 +837,8 @@ init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) INIT_LIST_HEAD(&rpci->pipe); rpci->pipelen = 0; init_waitqueue_head(&rpci->waitq); - INIT_WORK(&rpci->queue_timeout, rpc_timeout_upcall_queue, rpci); + INIT_DELAYED_WORK(&rpci->queue_timeout, + rpc_timeout_upcall_queue, rpci); rpci->ops = NULL; } } diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 757fc91ef25d..3c7532cd009e 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1262,7 +1262,7 @@ static void xs_connect(struct rpc_task *task) xprt->reestablish_timeout = XS_TCP_MAX_REEST_TO; } else { dprintk("RPC: xs_connect scheduled xprt %p\n", xprt); - schedule_work(&xprt->connect_worker); + schedule_delayed_work(&xprt->connect_worker, 0); /* flush_scheduled_work can sleep... */ if (!RPC_IS_ASYNC(task)) @@ -1375,7 +1375,7 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to) /* XXX: header size can vary due to auth type, IPv6, etc. */ xprt->max_payload = (1U << 16) - (MAX_HEADER << 3); - INIT_WORK(&xprt->connect_worker, xs_udp_connect_worker, xprt); + INIT_DELAYED_WORK(&xprt->connect_worker, xs_udp_connect_worker, xprt); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_UDP_CONN_TO; xprt->reestablish_timeout = XS_UDP_REEST_TO; @@ -1420,7 +1420,7 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to) xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; - INIT_WORK(&xprt->connect_worker, xs_tcp_connect_worker, xprt); + INIT_DELAYED_WORK(&xprt->connect_worker, xs_tcp_connect_worker, xprt); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_TCP_CONN_TO; xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; -- cgit v1.2.3 From 6bb49e5965c1fc399b4d3cd2b5cf2da535b330c0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:54:45 +0000 Subject: WorkStruct: Typedef the work function prototype Define a type for the work function prototype. It's not only kept in the work_struct struct, it's also passed as an argument to several functions. This makes it easier to change it. Signed-Off-By: David Howells --- drivers/block/floppy.c | 4 ++-- include/linux/workqueue.h | 9 +++++---- kernel/workqueue.c | 6 +++--- 3 files changed, 10 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 9e6d3a87cbe3..5a14fac13b12 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -996,7 +996,7 @@ static DECLARE_WORK(floppy_work, NULL, NULL); static void schedule_bh(void (*handler) (void)) { - PREPARE_WORK(&floppy_work, (void (*)(void *))handler, NULL); + PREPARE_WORK(&floppy_work, (work_func_t)handler, NULL); schedule_work(&floppy_work); } @@ -1008,7 +1008,7 @@ static void cancel_activity(void) spin_lock_irqsave(&floppy_lock, flags); do_floppy = NULL; - PREPARE_WORK(&floppy_work, (void *)empty, NULL); + PREPARE_WORK(&floppy_work, (work_func_t)empty, NULL); del_timer(&fd_timer); spin_unlock_irqrestore(&floppy_lock, flags); } diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 9faaccae570e..cef40b22ff9a 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -11,10 +11,12 @@ struct workqueue_struct; +typedef void (*work_func_t)(void *data); + struct work_struct { unsigned long pending; struct list_head entry; - void (*func)(void *); + work_func_t func; void *data; void *wq_data; }; @@ -91,7 +93,7 @@ extern int FASTCALL(schedule_work(struct work_struct *work)); extern int FASTCALL(schedule_delayed_work(struct delayed_work *work, unsigned long delay)); extern int schedule_delayed_work_on(int cpu, struct delayed_work *work, unsigned long delay); -extern int schedule_on_each_cpu(void (*func)(void *info), void *info); +extern int schedule_on_each_cpu(work_func_t func, void *info); extern void flush_scheduled_work(void); extern int current_is_keventd(void); extern int keventd_up(void); @@ -100,8 +102,7 @@ extern void init_workqueues(void); void cancel_rearming_delayed_work(struct delayed_work *work); void cancel_rearming_delayed_workqueue(struct workqueue_struct *, struct delayed_work *); -int execute_in_process_context(void (*fn)(void *), void *, - struct execute_work *); +int execute_in_process_context(work_func_t fn, void *, struct execute_work *); /* * Kill off a pending schedule_delayed_work(). Note that the work callback diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 44fc54b7decf..1e9d61ecf762 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -217,7 +217,7 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) while (!list_empty(&cwq->worklist)) { struct work_struct *work = list_entry(cwq->worklist.next, struct work_struct, entry); - void (*f) (void *) = work->func; + work_func_t f = work->func; void *data = work->data; list_del_init(cwq->worklist.next); @@ -513,7 +513,7 @@ EXPORT_SYMBOL(schedule_delayed_work_on); * * schedule_on_each_cpu() is very slow. */ -int schedule_on_each_cpu(void (*func)(void *info), void *info) +int schedule_on_each_cpu(work_func_t func, void *info) { int cpu; struct work_struct *works; @@ -578,7 +578,7 @@ EXPORT_SYMBOL(cancel_rearming_delayed_work); * Returns: 0 - function was executed * 1 - function was scheduled for execution */ -int execute_in_process_context(void (*fn)(void *data), void *data, +int execute_in_process_context(work_func_t fn, void *data, struct execute_work *ew) { if (!in_interrupt()) { -- cgit v1.2.3 From 365970a1ea76d81cb1ad2f652acb605f06dae256 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:54:49 +0000 Subject: WorkStruct: Merge the pending bit into the wq_data pointer Reclaim a word from the size of the work_struct by folding the pending bit and the wq_data pointer together. This shouldn't cause misalignment problems as all pointers should be at least 4-byte aligned. Signed-Off-By: David Howells --- drivers/block/floppy.c | 4 ++-- include/linux/workqueue.h | 27 +++++++++++++++++++++++---- kernel/workqueue.c | 41 ++++++++++++++++++++++++++++++++--------- 3 files changed, 57 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index 5a14fac13b12..aa1eb4466f9d 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -1868,7 +1868,7 @@ static void show_floppy(void) printk("fdc_busy=%lu\n", fdc_busy); if (do_floppy) printk("do_floppy=%p\n", do_floppy); - if (floppy_work.pending) + if (work_pending(&floppy_work)) printk("floppy_work.func=%p\n", floppy_work.func); if (timer_pending(&fd_timer)) printk("fd_timer.function=%p\n", fd_timer.function); @@ -4498,7 +4498,7 @@ static void floppy_release_irq_and_dma(void) printk("floppy timer still active:%s\n", timeout_message); if (timer_pending(&fd_timer)) printk("auxiliary floppy timer still active\n"); - if (floppy_work.pending) + if (work_pending(&floppy_work)) printk("work still pending\n"); #endif old_fdc = fdc; diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index cef40b22ff9a..ecc017d24cf3 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -14,11 +14,15 @@ struct workqueue_struct; typedef void (*work_func_t)(void *data); struct work_struct { - unsigned long pending; + /* the first word is the work queue pointer and the pending flag + * rolled into one */ + unsigned long management; +#define WORK_STRUCT_PENDING 0 /* T if work item pending execution */ +#define WORK_STRUCT_FLAG_MASK (3UL) +#define WORK_STRUCT_WQ_DATA_MASK (~WORK_STRUCT_FLAG_MASK) struct list_head entry; work_func_t func; void *data; - void *wq_data; }; struct delayed_work { @@ -65,7 +69,7 @@ struct execute_work { #define INIT_WORK(_work, _func, _data) \ do { \ INIT_LIST_HEAD(&(_work)->entry); \ - (_work)->pending = 0; \ + (_work)->management = 0; \ PREPARE_WORK((_work), (_func), (_data)); \ } while (0) @@ -75,6 +79,21 @@ struct execute_work { init_timer(&(_work)->timer); \ } while (0) +/** + * work_pending - Find out whether a work item is currently pending + * @work: The work item in question + */ +#define work_pending(work) \ + test_bit(WORK_STRUCT_PENDING, &(work)->management) + +/** + * delayed_work_pending - Find out whether a delayable work item is currently + * pending + * @work: The work item in question + */ +#define delayed_work_pending(work) \ + test_bit(WORK_STRUCT_PENDING, &(work)->work.management) + extern struct workqueue_struct *__create_workqueue(const char *name, int singlethread); @@ -115,7 +134,7 @@ static inline int cancel_delayed_work(struct delayed_work *work) ret = del_timer_sync(&work->timer); if (ret) - clear_bit(0, &work->work.pending); + clear_bit(WORK_STRUCT_PENDING, &work->work.management); return ret; } diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1e9d61ecf762..967479756511 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -80,6 +80,29 @@ static inline int is_single_threaded(struct workqueue_struct *wq) return list_empty(&wq->list); } +static inline void set_wq_data(struct work_struct *work, void *wq) +{ + unsigned long new, old, res; + + /* assume the pending flag is already set and that the task has already + * been queued on this workqueue */ + new = (unsigned long) wq | (1UL << WORK_STRUCT_PENDING); + res = work->management; + if (res != new) { + do { + old = res; + new = (unsigned long) wq; + new |= (old & WORK_STRUCT_FLAG_MASK); + res = cmpxchg(&work->management, old, new); + } while (res != old); + } +} + +static inline void *get_wq_data(struct work_struct *work) +{ + return (void *) (work->management & WORK_STRUCT_WQ_DATA_MASK); +} + /* Preempt must be disabled. */ static void __queue_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) @@ -87,7 +110,7 @@ static void __queue_work(struct cpu_workqueue_struct *cwq, unsigned long flags; spin_lock_irqsave(&cwq->lock, flags); - work->wq_data = cwq; + set_wq_data(work, cwq); list_add_tail(&work->entry, &cwq->worklist); cwq->insert_sequence++; wake_up(&cwq->more_work); @@ -108,7 +131,7 @@ int fastcall queue_work(struct workqueue_struct *wq, struct work_struct *work) { int ret = 0, cpu = get_cpu(); - if (!test_and_set_bit(0, &work->pending)) { + if (!test_and_set_bit(WORK_STRUCT_PENDING, &work->management)) { if (unlikely(is_single_threaded(wq))) cpu = singlethread_cpu; BUG_ON(!list_empty(&work->entry)); @@ -123,7 +146,7 @@ EXPORT_SYMBOL_GPL(queue_work); static void delayed_work_timer_fn(unsigned long __data) { struct delayed_work *dwork = (struct delayed_work *)__data; - struct workqueue_struct *wq = dwork->work.wq_data; + struct workqueue_struct *wq = get_wq_data(&dwork->work); int cpu = smp_processor_id(); if (unlikely(is_single_threaded(wq))) @@ -150,12 +173,12 @@ int fastcall queue_delayed_work(struct workqueue_struct *wq, if (delay == 0) return queue_work(wq, work); - if (!test_and_set_bit(0, &work->pending)) { + if (!test_and_set_bit(WORK_STRUCT_PENDING, &work->management)) { BUG_ON(timer_pending(timer)); BUG_ON(!list_empty(&work->entry)); /* This stores wq for the moment, for the timer_fn */ - work->wq_data = wq; + set_wq_data(work, wq); timer->expires = jiffies + delay; timer->data = (unsigned long)dwork; timer->function = delayed_work_timer_fn; @@ -182,12 +205,12 @@ int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, struct timer_list *timer = &dwork->timer; struct work_struct *work = &dwork->work; - if (!test_and_set_bit(0, &work->pending)) { + if (!test_and_set_bit(WORK_STRUCT_PENDING, &work->management)) { BUG_ON(timer_pending(timer)); BUG_ON(!list_empty(&work->entry)); /* This stores wq for the moment, for the timer_fn */ - work->wq_data = wq; + set_wq_data(work, wq); timer->expires = jiffies + delay; timer->data = (unsigned long)dwork; timer->function = delayed_work_timer_fn; @@ -223,8 +246,8 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) list_del_init(cwq->worklist.next); spin_unlock_irqrestore(&cwq->lock, flags); - BUG_ON(work->wq_data != cwq); - clear_bit(0, &work->pending); + BUG_ON(get_wq_data(work) != cwq); + clear_bit(WORK_STRUCT_PENDING, &work->management); f(data); spin_lock_irqsave(&cwq->lock, flags); -- cgit v1.2.3 From 65f27f38446e1976cc98fd3004b110fedcddd189 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:55:48 +0000 Subject: WorkStruct: Pass the work_struct pointer instead of context data Pass the work_struct pointer to the work function rather than context data. The work function can use container_of() to work out the data. For the cases where the container of the work_struct may go away the moment the pending bit is cleared, it is made possible to defer the release of the structure by deferring the clearing of the pending bit. To make this work, an extra flag is introduced into the management side of the work_struct. This governs auto-release of the structure upon execution. Ordinarily, the work queue executor would release the work_struct for further scheduling or deallocation by clearing the pending bit prior to jumping to the work function. This means that, unless the driver makes some guarantee itself that the work_struct won't go away, the work function may not access anything else in the work_struct or its container lest they be deallocated.. This is a problem if the auxiliary data is taken away (as done by the last patch). However, if the pending bit is *not* cleared before jumping to the work function, then the work function *may* access the work_struct and its container with no problems. But then the work function must itself release the work_struct by calling work_release(). In most cases, automatic release is fine, so this is the default. Special initiators exist for the non-auto-release case (ending in _NAR). Signed-Off-By: David Howells --- arch/x86_64/kernel/mce.c | 6 +-- arch/x86_64/kernel/smpboot.c | 12 +++-- arch/x86_64/kernel/time.c | 4 +- block/as-iosched.c | 7 +-- block/cfq-iosched.c | 8 +-- block/ll_rw_blk.c | 8 +-- crypto/cryptomgr.c | 7 +-- drivers/acpi/osl.c | 25 +++------- drivers/ata/libata-core.c | 20 ++++---- drivers/ata/libata-scsi.c | 14 +++--- drivers/ata/libata.h | 4 +- drivers/block/floppy.c | 6 +-- drivers/char/random.c | 6 +-- drivers/char/sysrq.c | 4 +- drivers/char/tty_io.c | 31 ++++++------ drivers/char/vt.c | 6 +-- drivers/cpufreq/cpufreq.c | 10 ++-- drivers/input/keyboard/atkbd.c | 6 +-- drivers/input/serio/libps2.c | 6 +-- drivers/net/e1000/e1000_main.c | 10 ++-- drivers/pci/pcie/aer/aerdrv.c | 2 +- drivers/pci/pcie/aer/aerdrv.h | 2 +- drivers/pci/pcie/aer/aerdrv_core.c | 8 +-- drivers/scsi/scsi_scan.c | 7 +-- drivers/scsi/scsi_sysfs.c | 10 ++-- fs/aio.c | 14 +++--- fs/bio.c | 6 +-- fs/file.c | 6 ++- fs/nfs/client.c | 2 +- fs/nfs/namespace.c | 9 ++-- fs/nfs/nfs4_fs.h | 2 +- fs/nfs/nfs4renewd.c | 5 +- include/linux/libata.h | 3 +- include/linux/workqueue.h | 99 +++++++++++++++++++++++++++++--------- include/net/inet_timewait_sock.h | 2 +- ipc/util.c | 7 ++- kernel/kmod.c | 16 +++--- kernel/kthread.c | 13 +++-- kernel/power/poweroff.c | 4 +- kernel/sys.c | 4 +- kernel/workqueue.c | 19 +++----- mm/slab.c | 6 +-- net/core/link_watch.c | 6 +-- net/ipv4/inet_timewait_sock.c | 5 +- net/ipv4/tcp_minisocks.c | 3 +- net/sunrpc/cache.c | 6 +-- net/sunrpc/rpc_pipe.c | 7 +-- net/sunrpc/sched.c | 8 +-- net/sunrpc/xprt.c | 7 +-- net/sunrpc/xprtsock.c | 18 ++++--- security/keys/key.c | 6 +-- 51 files changed, 293 insertions(+), 219 deletions(-) (limited to 'include/linux') diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index 5306f2630905..c7587fc39015 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -306,8 +306,8 @@ void mce_log_therm_throt_event(unsigned int cpu, __u64 status) */ static int check_interval = 5 * 60; /* 5 minutes */ -static void mcheck_timer(void *data); -static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer, NULL); +static void mcheck_timer(struct work_struct *work); +static DECLARE_DELAYED_WORK(mcheck_work, mcheck_timer); static void mcheck_check_cpu(void *info) { @@ -315,7 +315,7 @@ static void mcheck_check_cpu(void *info) do_machine_check(NULL, 0); } -static void mcheck_timer(void *data) +static void mcheck_timer(struct work_struct *work) { on_each_cpu(mcheck_check_cpu, NULL, 1, 1); schedule_delayed_work(&mcheck_work, check_interval * HZ); diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c index 62c2e747af58..9800147c4c68 100644 --- a/arch/x86_64/kernel/smpboot.c +++ b/arch/x86_64/kernel/smpboot.c @@ -753,14 +753,16 @@ static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int sta } struct create_idle { + struct work_struct work; struct task_struct *idle; struct completion done; int cpu; }; -void do_fork_idle(void *_c_idle) +void do_fork_idle(struct work_struct *work) { - struct create_idle *c_idle = _c_idle; + struct create_idle *c_idle = + container_of(work, struct create_idle, work); c_idle->idle = fork_idle(c_idle->cpu); complete(&c_idle->done); @@ -775,10 +777,10 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) int timeout; unsigned long start_rip; struct create_idle c_idle = { + .work = __WORK_INITIALIZER(c_idle.work, do_fork_idle), .cpu = cpu, .done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done), }; - DECLARE_WORK(work, do_fork_idle, &c_idle); /* allocate memory for gdts of secondary cpus. Hotplug is considered */ if (!cpu_gdt_descr[cpu].address && @@ -825,9 +827,9 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid) * thread. */ if (!keventd_up() || current_is_keventd()) - work.func(work.data); + c_idle.work.func(&c_idle.work); else { - schedule_work(&work); + schedule_work(&c_idle.work); wait_for_completion(&c_idle.done); } diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c index e3ef544d2cfb..9f05bc9b2dad 100644 --- a/arch/x86_64/kernel/time.c +++ b/arch/x86_64/kernel/time.c @@ -563,7 +563,7 @@ static unsigned int cpufreq_delayed_issched = 0; static unsigned int cpufreq_init = 0; static struct work_struct cpufreq_delayed_get_work; -static void handle_cpufreq_delayed_get(void *v) +static void handle_cpufreq_delayed_get(struct work_struct *v) { unsigned int cpu; for_each_online_cpu(cpu) { @@ -639,7 +639,7 @@ static struct notifier_block time_cpufreq_notifier_block = { static int __init cpufreq_tsc(void) { - INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); + INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get); if (!cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER)) cpufreq_init = 1; diff --git a/block/as-iosched.c b/block/as-iosched.c index 50b95e4c1425..f371c9359999 100644 --- a/block/as-iosched.c +++ b/block/as-iosched.c @@ -1274,9 +1274,10 @@ static void as_merged_requests(request_queue_t *q, struct request *req, * * FIXME! dispatch queue is not a queue at all! */ -static void as_work_handler(void *data) +static void as_work_handler(struct work_struct *work) { - struct request_queue *q = data; + struct as_data *ad = container_of(work, struct as_data, antic_work); + struct request_queue *q = ad->q; unsigned long flags; spin_lock_irqsave(q->queue_lock, flags); @@ -1332,7 +1333,7 @@ static void *as_init_queue(request_queue_t *q, elevator_t *e) ad->antic_timer.function = as_antic_timeout; ad->antic_timer.data = (unsigned long)q; init_timer(&ad->antic_timer); - INIT_WORK(&ad->antic_work, as_work_handler, q); + INIT_WORK(&ad->antic_work, as_work_handler); INIT_LIST_HEAD(&ad->fifo_list[REQ_SYNC]); INIT_LIST_HEAD(&ad->fifo_list[REQ_ASYNC]); diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 1d9c3c70a9a0..6cec3a1dccb8 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1841,9 +1841,11 @@ queue_fail: return 1; } -static void cfq_kick_queue(void *data) +static void cfq_kick_queue(struct work_struct *work) { - request_queue_t *q = data; + struct cfq_data *cfqd = + container_of(work, struct cfq_data, unplug_work); + request_queue_t *q = cfqd->queue; unsigned long flags; spin_lock_irqsave(q->queue_lock, flags); @@ -1987,7 +1989,7 @@ static void *cfq_init_queue(request_queue_t *q, elevator_t *e) cfqd->idle_class_timer.function = cfq_idle_class_timer; cfqd->idle_class_timer.data = (unsigned long) cfqd; - INIT_WORK(&cfqd->unplug_work, cfq_kick_queue, q); + INIT_WORK(&cfqd->unplug_work, cfq_kick_queue); cfqd->cfq_quantum = cfq_quantum; cfqd->cfq_fifo_expire[0] = cfq_fifo_expire[0]; diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 9eaee6640535..eb4cf6df7374 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -34,7 +34,7 @@ */ #include -static void blk_unplug_work(void *data); +static void blk_unplug_work(struct work_struct *work); static void blk_unplug_timeout(unsigned long data); static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io); static void init_request_from_bio(struct request *req, struct bio *bio); @@ -227,7 +227,7 @@ void blk_queue_make_request(request_queue_t * q, make_request_fn * mfn) if (q->unplug_delay == 0) q->unplug_delay = 1; - INIT_WORK(&q->unplug_work, blk_unplug_work, q); + INIT_WORK(&q->unplug_work, blk_unplug_work); q->unplug_timer.function = blk_unplug_timeout; q->unplug_timer.data = (unsigned long)q; @@ -1631,9 +1631,9 @@ static void blk_backing_dev_unplug(struct backing_dev_info *bdi, } } -static void blk_unplug_work(void *data) +static void blk_unplug_work(struct work_struct *work) { - request_queue_t *q = data; + request_queue_t *q = container_of(work, request_queue_t, unplug_work); blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL, q->rq.count[READ] + q->rq.count[WRITE]); diff --git a/crypto/cryptomgr.c b/crypto/cryptomgr.c index 9b5b15601068..2ebffb84f1d9 100644 --- a/crypto/cryptomgr.c +++ b/crypto/cryptomgr.c @@ -40,9 +40,10 @@ struct cryptomgr_param { char template[CRYPTO_MAX_ALG_NAME]; }; -static void cryptomgr_probe(void *data) +static void cryptomgr_probe(struct work_struct *work) { - struct cryptomgr_param *param = data; + struct cryptomgr_param *param = + container_of(work, struct cryptomgr_param, work); struct crypto_template *tmpl; struct crypto_instance *inst; int err; @@ -112,7 +113,7 @@ static int cryptomgr_schedule_probe(struct crypto_larval *larval) param->larval.type = larval->alg.cra_flags; param->larval.mask = larval->mask; - INIT_WORK(¶m->work, cryptomgr_probe, param); + INIT_WORK(¶m->work, cryptomgr_probe); schedule_work(¶m->work); return NOTIFY_STOP; diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 068fe4f100b0..02b30ae6a68e 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -50,6 +50,7 @@ ACPI_MODULE_NAME("osl") struct acpi_os_dpc { acpi_osd_exec_callback function; void *context; + struct work_struct work; }; #ifdef CONFIG_ACPI_CUSTOM_DSDT @@ -564,12 +565,9 @@ void acpi_os_derive_pci_id(acpi_handle rhandle, /* upper bound */ acpi_os_derive_pci_id_2(rhandle, chandle, id, &is_bridge, &bus_number); } -static void acpi_os_execute_deferred(void *context) +static void acpi_os_execute_deferred(struct work_struct *work) { - struct acpi_os_dpc *dpc = NULL; - - - dpc = (struct acpi_os_dpc *)context; + struct acpi_os_dpc *dpc = container_of(work, struct acpi_os_dpc, work); if (!dpc) { printk(KERN_ERR PREFIX "Invalid (NULL) context\n"); return; @@ -602,7 +600,6 @@ acpi_status acpi_os_execute(acpi_execute_type type, { acpi_status status = AE_OK; struct acpi_os_dpc *dpc; - struct work_struct *task; ACPI_FUNCTION_TRACE("os_queue_for_execution"); @@ -615,28 +612,22 @@ acpi_status acpi_os_execute(acpi_execute_type type, /* * Allocate/initialize DPC structure. Note that this memory will be - * freed by the callee. The kernel handles the tq_struct list in a + * freed by the callee. The kernel handles the work_struct list in a * way that allows us to also free its memory inside the callee. * Because we may want to schedule several tasks with different * parameters we can't use the approach some kernel code uses of - * having a static tq_struct. - * We can save time and code by allocating the DPC and tq_structs - * from the same memory. + * having a static work_struct. */ - dpc = - kmalloc(sizeof(struct acpi_os_dpc) + sizeof(struct work_struct), - GFP_ATOMIC); + dpc = kmalloc(sizeof(struct acpi_os_dpc), GFP_ATOMIC); if (!dpc) return_ACPI_STATUS(AE_NO_MEMORY); dpc->function = function; dpc->context = context; - task = (void *)(dpc + 1); - INIT_WORK(task, acpi_os_execute_deferred, (void *)dpc); - - if (!queue_work(kacpid_wq, task)) { + INIT_WORK(&dpc->work, acpi_os_execute_deferred); + if (!queue_work(kacpid_wq, &dpc->work)) { ACPI_DEBUG_PRINT((ACPI_DB_ERROR, "Call to queue_work() failed.\n")); kfree(dpc); diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 0bb4b4dced76..b5f2da6ac80e 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -914,7 +914,7 @@ static unsigned int ata_id_xfermask(const u16 *id) * ata_port_queue_task - Queue port_task * @ap: The ata_port to queue port_task for * @fn: workqueue function to be scheduled - * @data: data value to pass to workqueue function + * @data: data for @fn to use * @delay: delay time for workqueue function * * Schedule @fn(@data) for execution after @delay jiffies using @@ -929,7 +929,7 @@ static unsigned int ata_id_xfermask(const u16 *id) * LOCKING: * Inherited from caller. */ -void ata_port_queue_task(struct ata_port *ap, void (*fn)(void *), void *data, +void ata_port_queue_task(struct ata_port *ap, work_func_t fn, void *data, unsigned long delay) { int rc; @@ -937,7 +937,8 @@ void ata_port_queue_task(struct ata_port *ap, void (*fn)(void *), void *data, if (ap->pflags & ATA_PFLAG_FLUSH_PORT_TASK) return; - PREPARE_DELAYED_WORK(&ap->port_task, fn, data); + PREPARE_DELAYED_WORK(&ap->port_task, fn); + ap->port_task_data = data; rc = queue_delayed_work(ata_wq, &ap->port_task, delay); @@ -4292,10 +4293,11 @@ fsm_start: return poll_next; } -static void ata_pio_task(void *_data) +static void ata_pio_task(struct work_struct *work) { - struct ata_queued_cmd *qc = _data; - struct ata_port *ap = qc->ap; + struct ata_port *ap = + container_of(work, struct ata_port, port_task.work); + struct ata_queued_cmd *qc = ap->port_task_data; u8 status; int poll_next; @@ -5317,9 +5319,9 @@ void ata_port_init(struct ata_port *ap, struct ata_host *host, ap->msg_enable = ATA_MSG_DRV | ATA_MSG_ERR | ATA_MSG_WARN; #endif - INIT_DELAYED_WORK(&ap->port_task, NULL, NULL); - INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug, ap); - INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan, ap); + INIT_DELAYED_WORK(&ap->port_task, NULL); + INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug); + INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan); INIT_LIST_HEAD(&ap->eh_done_q); init_waitqueue_head(&ap->eh_wait_q); diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 5c1fc467fc7f..c872b324dbd3 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -3079,7 +3079,7 @@ static void ata_scsi_remove_dev(struct ata_device *dev) /** * ata_scsi_hotplug - SCSI part of hotplug - * @data: Pointer to ATA port to perform SCSI hotplug on + * @work: Pointer to ATA port to perform SCSI hotplug on * * Perform SCSI part of hotplug. It's executed from a separate * workqueue after EH completes. This is necessary because SCSI @@ -3089,9 +3089,10 @@ static void ata_scsi_remove_dev(struct ata_device *dev) * LOCKING: * Kernel thread context (may sleep). */ -void ata_scsi_hotplug(void *data) +void ata_scsi_hotplug(struct work_struct *work) { - struct ata_port *ap = data; + struct ata_port *ap = + container_of(work, struct ata_port, hotplug_task.work); int i; if (ap->pflags & ATA_PFLAG_UNLOADING) { @@ -3190,7 +3191,7 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel, /** * ata_scsi_dev_rescan - initiate scsi_rescan_device() - * @data: Pointer to ATA port to perform scsi_rescan_device() + * @work: Pointer to ATA port to perform scsi_rescan_device() * * After ATA pass thru (SAT) commands are executed successfully, * libata need to propagate the changes to SCSI layer. This @@ -3200,9 +3201,10 @@ static int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel, * LOCKING: * Kernel thread context (may sleep). */ -void ata_scsi_dev_rescan(void *data) +void ata_scsi_dev_rescan(struct work_struct *work) { - struct ata_port *ap = data; + struct ata_port *ap = + container_of(work, struct ata_port, scsi_rescan_task); struct ata_device *dev; unsigned int i; diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index 0ed263be652a..7e0f3aff873d 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -81,7 +81,7 @@ extern struct scsi_transport_template ata_scsi_transport_template; extern void ata_scsi_scan_host(struct ata_port *ap); extern int ata_scsi_offline_dev(struct ata_device *dev); -extern void ata_scsi_hotplug(void *data); +extern void ata_scsi_hotplug(struct work_struct *work); extern unsigned int ata_scsiop_inq_std(struct ata_scsi_args *args, u8 *rbuf, unsigned int buflen); @@ -111,7 +111,7 @@ extern void ata_scsi_rbuf_fill(struct ata_scsi_args *args, unsigned int (*actor) (struct ata_scsi_args *args, u8 *rbuf, unsigned int buflen)); extern void ata_schedule_scsi_eh(struct Scsi_Host *shost); -extern void ata_scsi_dev_rescan(void *data); +extern void ata_scsi_dev_rescan(struct work_struct *work); extern int ata_bus_probe(struct ata_port *ap); /* libata-eh.c */ diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c index aa1eb4466f9d..3f1b38276e96 100644 --- a/drivers/block/floppy.c +++ b/drivers/block/floppy.c @@ -992,11 +992,11 @@ static void empty(void) { } -static DECLARE_WORK(floppy_work, NULL, NULL); +static DECLARE_WORK(floppy_work, NULL); static void schedule_bh(void (*handler) (void)) { - PREPARE_WORK(&floppy_work, (work_func_t)handler, NULL); + PREPARE_WORK(&floppy_work, (work_func_t)handler); schedule_work(&floppy_work); } @@ -1008,7 +1008,7 @@ static void cancel_activity(void) spin_lock_irqsave(&floppy_lock, flags); do_floppy = NULL; - PREPARE_WORK(&floppy_work, (work_func_t)empty, NULL); + PREPARE_WORK(&floppy_work, (work_func_t)empty); del_timer(&fd_timer); spin_unlock_irqrestore(&floppy_lock, flags); } diff --git a/drivers/char/random.c b/drivers/char/random.c index f2ab61f3e8ae..fa764688cad1 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1422,9 +1422,9 @@ static struct keydata { static unsigned int ip_cnt; -static void rekey_seq_generator(void *private_); +static void rekey_seq_generator(struct work_struct *work); -static DECLARE_DELAYED_WORK(rekey_work, rekey_seq_generator, NULL); +static DECLARE_DELAYED_WORK(rekey_work, rekey_seq_generator); /* * Lock avoidance: @@ -1438,7 +1438,7 @@ static DECLARE_DELAYED_WORK(rekey_work, rekey_seq_generator, NULL); * happen, and even if that happens only a not perfectly compliant * ISN is generated, nothing fatal. */ -static void rekey_seq_generator(void *private_) +static void rekey_seq_generator(struct work_struct *work) { struct keydata *keyptr = &ip_keydata[1 ^ (ip_cnt & 1)]; diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index 5f49280779fb..c64f5bcff947 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -219,13 +219,13 @@ static struct sysrq_key_op sysrq_term_op = { .enable_mask = SYSRQ_ENABLE_SIGNAL, }; -static void moom_callback(void *ignored) +static void moom_callback(struct work_struct *ignored) { out_of_memory(&NODE_DATA(0)->node_zonelists[ZONE_NORMAL], GFP_KERNEL, 0); } -static DECLARE_WORK(moom_work, moom_callback, NULL); +static DECLARE_WORK(moom_work, moom_callback); static void sysrq_handle_moom(int key, struct tty_struct *tty) { diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 7297acfe520c..83e9e7d9b58c 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -1254,7 +1254,7 @@ EXPORT_SYMBOL_GPL(tty_ldisc_flush); /** * do_tty_hangup - actual handler for hangup events - * @data: tty device + * @work: tty device * * This can be called by the "eventd" kernel thread. That is process * synchronous but doesn't hold any locks, so we need to make sure we @@ -1274,9 +1274,10 @@ EXPORT_SYMBOL_GPL(tty_ldisc_flush); * tasklist_lock to walk task list for hangup event * */ -static void do_tty_hangup(void *data) +static void do_tty_hangup(struct work_struct *work) { - struct tty_struct *tty = (struct tty_struct *) data; + struct tty_struct *tty = + container_of(work, struct tty_struct, hangup_work); struct file * cons_filp = NULL; struct file *filp, *f = NULL; struct task_struct *p; @@ -1433,7 +1434,7 @@ void tty_vhangup(struct tty_struct * tty) printk(KERN_DEBUG "%s vhangup...\n", tty_name(tty, buf)); #endif - do_tty_hangup((void *) tty); + do_tty_hangup(&tty->hangup_work); } EXPORT_SYMBOL(tty_vhangup); @@ -3304,12 +3305,13 @@ int tty_ioctl(struct inode * inode, struct file * file, * Nasty bug: do_SAK is being called in interrupt context. This can * deadlock. We punt it up to process context. AKPM - 16Mar2001 */ -static void __do_SAK(void *arg) +static void __do_SAK(struct work_struct *work) { + struct tty_struct *tty = + container_of(work, struct tty_struct, SAK_work); #ifdef TTY_SOFT_SAK tty_hangup(tty); #else - struct tty_struct *tty = arg; struct task_struct *g, *p; int session; int i; @@ -3388,7 +3390,7 @@ void do_SAK(struct tty_struct *tty) { if (!tty) return; - PREPARE_WORK(&tty->SAK_work, __do_SAK, tty); + PREPARE_WORK(&tty->SAK_work, __do_SAK); schedule_work(&tty->SAK_work); } @@ -3396,7 +3398,7 @@ EXPORT_SYMBOL(do_SAK); /** * flush_to_ldisc - * @private_: tty structure passed from work queue. + * @work: tty structure passed from work queue. * * This routine is called out of the software interrupt to flush data * from the buffer chain to the line discipline. @@ -3406,9 +3408,10 @@ EXPORT_SYMBOL(do_SAK); * receive_buf method is single threaded for each tty instance. */ -static void flush_to_ldisc(void *private_) +static void flush_to_ldisc(struct work_struct *work) { - struct tty_struct *tty = (struct tty_struct *) private_; + struct tty_struct *tty = + container_of(work, struct tty_struct, buf.work.work); unsigned long flags; struct tty_ldisc *disc; struct tty_buffer *tbuf, *head; @@ -3553,7 +3556,7 @@ void tty_flip_buffer_push(struct tty_struct *tty) spin_unlock_irqrestore(&tty->buf.lock, flags); if (tty->low_latency) - flush_to_ldisc((void *) tty); + flush_to_ldisc(&tty->buf.work.work); else schedule_delayed_work(&tty->buf.work, 1); } @@ -3580,17 +3583,17 @@ static void initialize_tty_struct(struct tty_struct *tty) tty->overrun_time = jiffies; tty->buf.head = tty->buf.tail = NULL; tty_buffer_init(tty); - INIT_DELAYED_WORK(&tty->buf.work, flush_to_ldisc, tty); + INIT_DELAYED_WORK(&tty->buf.work, flush_to_ldisc); init_MUTEX(&tty->buf.pty_sem); mutex_init(&tty->termios_mutex); init_waitqueue_head(&tty->write_wait); init_waitqueue_head(&tty->read_wait); - INIT_WORK(&tty->hangup_work, do_tty_hangup, tty); + INIT_WORK(&tty->hangup_work, do_tty_hangup); mutex_init(&tty->atomic_read_lock); mutex_init(&tty->atomic_write_lock); spin_lock_init(&tty->read_lock); INIT_LIST_HEAD(&tty->tty_files); - INIT_WORK(&tty->SAK_work, NULL, NULL); + INIT_WORK(&tty->SAK_work, NULL); } /* diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 8e4413f6fbaf..8ee04adc37f0 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -155,7 +155,7 @@ static void con_flush_chars(struct tty_struct *tty); static void set_vesa_blanking(char __user *p); static void set_cursor(struct vc_data *vc); static void hide_cursor(struct vc_data *vc); -static void console_callback(void *ignored); +static void console_callback(struct work_struct *ignored); static void blank_screen_t(unsigned long dummy); static void set_palette(struct vc_data *vc); @@ -174,7 +174,7 @@ static int vesa_blank_mode; /* 0:none 1:suspendV 2:suspendH 3:powerdown */ static int blankinterval = 10*60*HZ; static int vesa_off_interval; -static DECLARE_WORK(console_work, console_callback, NULL); +static DECLARE_WORK(console_work, console_callback); /* * fg_console is the current virtual console, @@ -2154,7 +2154,7 @@ out: * with other console code and prevention of re-entrancy is * ensured with console_sem. */ -static void console_callback(void *ignored) +static void console_callback(struct work_struct *ignored) { acquire_console_sem(); diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index dd0c2623e27b..7a7c6e6dfe4f 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -42,7 +42,7 @@ static DEFINE_SPINLOCK(cpufreq_driver_lock); /* internal prototypes */ static int __cpufreq_governor(struct cpufreq_policy *policy, unsigned int event); -static void handle_update(void *data); +static void handle_update(struct work_struct *work); /** * Two notifier lists: the "policy" list is involved in the @@ -665,7 +665,7 @@ static int cpufreq_add_dev (struct sys_device * sys_dev) mutex_init(&policy->lock); mutex_lock(&policy->lock); init_completion(&policy->kobj_unregister); - INIT_WORK(&policy->update, handle_update, (void *)(long)cpu); + INIT_WORK(&policy->update, handle_update); /* call driver. From then on the cpufreq must be able * to accept all calls to ->verify and ->setpolicy for this CPU @@ -895,9 +895,11 @@ static int cpufreq_remove_dev (struct sys_device * sys_dev) } -static void handle_update(void *data) +static void handle_update(struct work_struct *work) { - unsigned int cpu = (unsigned int)(long)data; + struct cpufreq_policy *policy = + container_of(work, struct cpufreq_policy, update); + unsigned int cpu = policy->cpu; dprintk("handle_update for cpu %u called\n", cpu); cpufreq_update_policy(cpu); } diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index cbb93669d1ce..8451b29a3db5 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -567,9 +567,9 @@ static int atkbd_set_leds(struct atkbd *atkbd) * interrupt context. */ -static void atkbd_event_work(void *data) +static void atkbd_event_work(struct work_struct *work) { - struct atkbd *atkbd = data; + struct atkbd *atkbd = container_of(work, struct atkbd, event_work); mutex_lock(&atkbd->event_mutex); @@ -943,7 +943,7 @@ static int atkbd_connect(struct serio *serio, struct serio_driver *drv) atkbd->dev = dev; ps2_init(&atkbd->ps2dev, serio); - INIT_WORK(&atkbd->event_work, atkbd_event_work, atkbd); + INIT_WORK(&atkbd->event_work, atkbd_event_work); mutex_init(&atkbd->event_mutex); switch (serio->id.type) { diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index e5b1b60757bb..b3e84d3bb7f7 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -251,9 +251,9 @@ EXPORT_SYMBOL(ps2_command); * ps2_schedule_command(), to a PS/2 device (keyboard, mouse, etc.) */ -static void ps2_execute_scheduled_command(void *data) +static void ps2_execute_scheduled_command(struct work_struct *work) { - struct ps2work *ps2work = data; + struct ps2work *ps2work = container_of(work, struct ps2work, work); ps2_command(ps2work->ps2dev, ps2work->param, ps2work->command); kfree(ps2work); @@ -278,7 +278,7 @@ int ps2_schedule_command(struct ps2dev *ps2dev, unsigned char *param, int comman ps2work->ps2dev = ps2dev; ps2work->command = command; memcpy(ps2work->param, param, send); - INIT_WORK(&ps2work->work, ps2_execute_scheduled_command, ps2work); + INIT_WORK(&ps2work->work, ps2_execute_scheduled_command); if (!schedule_work(&ps2work->work)) { kfree(ps2work); diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 726ec5e88ab2..03294400bc90 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -183,7 +183,7 @@ void e1000_set_ethtool_ops(struct net_device *netdev); static void e1000_enter_82542_rst(struct e1000_adapter *adapter); static void e1000_leave_82542_rst(struct e1000_adapter *adapter); static void e1000_tx_timeout(struct net_device *dev); -static void e1000_reset_task(struct net_device *dev); +static void e1000_reset_task(struct work_struct *work); static void e1000_smartspeed(struct e1000_adapter *adapter); static int e1000_82547_fifo_workaround(struct e1000_adapter *adapter, struct sk_buff *skb); @@ -908,8 +908,7 @@ e1000_probe(struct pci_dev *pdev, adapter->phy_info_timer.function = &e1000_update_phy_info; adapter->phy_info_timer.data = (unsigned long) adapter; - INIT_WORK(&adapter->reset_task, - (void (*)(void *))e1000_reset_task, netdev); + INIT_WORK(&adapter->reset_task, e1000_reset_task); e1000_check_options(adapter); @@ -3154,9 +3153,10 @@ e1000_tx_timeout(struct net_device *netdev) } static void -e1000_reset_task(struct net_device *netdev) +e1000_reset_task(struct work_struct *work) { - struct e1000_adapter *adapter = netdev_priv(netdev); + struct e1000_adapter *adapter = + container_of(work, struct e1000_adapter, reset_task); e1000_reinit_locked(adapter); } diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c index 04c43ef529ac..55866b6b26fa 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -160,7 +160,7 @@ static struct aer_rpc* aer_alloc_rpc(struct pcie_device *dev) rpc->e_lock = SPIN_LOCK_UNLOCKED; rpc->rpd = dev; - INIT_WORK(&rpc->dpc_handler, aer_isr, (void *)dev); + INIT_WORK(&rpc->dpc_handler, aer_isr); rpc->prod_idx = rpc->cons_idx = 0; mutex_init(&rpc->rpc_mutex); init_waitqueue_head(&rpc->wait_release); diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h index daf0cad88fc8..3c0a58f64dd8 100644 --- a/drivers/pci/pcie/aer/aerdrv.h +++ b/drivers/pci/pcie/aer/aerdrv.h @@ -118,7 +118,7 @@ extern struct bus_type pcie_port_bus_type; extern void aer_enable_rootport(struct aer_rpc *rpc); extern void aer_delete_rootport(struct aer_rpc *rpc); extern int aer_init(struct pcie_device *dev); -extern void aer_isr(void *context); +extern void aer_isr(struct work_struct *work); extern void aer_print_error(struct pci_dev *dev, struct aer_err_info *info); extern int aer_osc_setup(struct pci_dev *dev); diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c index 1c7e660d6535..08e13033ced8 100644 --- a/drivers/pci/pcie/aer/aerdrv_core.c +++ b/drivers/pci/pcie/aer/aerdrv_core.c @@ -690,14 +690,14 @@ static void aer_isr_one_error(struct pcie_device *p_device, /** * aer_isr - consume errors detected by root port - * @context: pointer to a private data of pcie device + * @work: definition of this work item * * Invoked, as DPC, when root port records new detected error **/ -void aer_isr(void *context) +void aer_isr(struct work_struct *work) { - struct pcie_device *p_device = (struct pcie_device *) context; - struct aer_rpc *rpc = get_service_data(p_device); + struct aer_rpc *rpc = container_of(work, struct aer_rpc, dpc_handler); + struct pcie_device *p_device = rpc->rpd; struct aer_err_source *e_src; mutex_lock(&rpc->rpc_mutex); diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 94a274645f6f..d3c5e964c964 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -362,9 +362,10 @@ static struct scsi_target *scsi_alloc_target(struct device *parent, goto retry; } -static void scsi_target_reap_usercontext(void *data) +static void scsi_target_reap_usercontext(struct work_struct *work) { - struct scsi_target *starget = data; + struct scsi_target *starget = + container_of(work, struct scsi_target, ew.work); struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); unsigned long flags; @@ -400,7 +401,7 @@ void scsi_target_reap(struct scsi_target *starget) starget->state = STARGET_DEL; spin_unlock_irqrestore(shost->host_lock, flags); execute_in_process_context(scsi_target_reap_usercontext, - starget, &starget->ew); + &starget->ew); return; } diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index e1a91665d1c2..259c90cfa367 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -218,16 +218,16 @@ static void scsi_device_cls_release(struct class_device *class_dev) put_device(&sdev->sdev_gendev); } -static void scsi_device_dev_release_usercontext(void *data) +static void scsi_device_dev_release_usercontext(struct work_struct *work) { - struct device *dev = data; struct scsi_device *sdev; struct device *parent; struct scsi_target *starget; unsigned long flags; - parent = dev->parent; - sdev = to_scsi_device(dev); + sdev = container_of(work, struct scsi_device, ew.work); + + parent = sdev->sdev_gendev.parent; starget = to_scsi_target(parent); spin_lock_irqsave(sdev->host->host_lock, flags); @@ -258,7 +258,7 @@ static void scsi_device_dev_release_usercontext(void *data) static void scsi_device_dev_release(struct device *dev) { struct scsi_device *sdp = to_scsi_device(dev); - execute_in_process_context(scsi_device_dev_release_usercontext, dev, + execute_in_process_context(scsi_device_dev_release_usercontext, &sdp->ew); } diff --git a/fs/aio.c b/fs/aio.c index 11a1a7100ad6..ca1c5180a17f 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -53,13 +53,13 @@ static kmem_cache_t *kioctx_cachep; static struct workqueue_struct *aio_wq; /* Used for rare fput completion. */ -static void aio_fput_routine(void *); -static DECLARE_WORK(fput_work, aio_fput_routine, NULL); +static void aio_fput_routine(struct work_struct *); +static DECLARE_WORK(fput_work, aio_fput_routine); static DEFINE_SPINLOCK(fput_lock); static LIST_HEAD(fput_head); -static void aio_kick_handler(void *); +static void aio_kick_handler(struct work_struct *); static void aio_queue_work(struct kioctx *); /* aio_setup @@ -227,7 +227,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) INIT_LIST_HEAD(&ctx->active_reqs); INIT_LIST_HEAD(&ctx->run_list); - INIT_DELAYED_WORK(&ctx->wq, aio_kick_handler, ctx); + INIT_DELAYED_WORK(&ctx->wq, aio_kick_handler); if (aio_setup_ring(ctx) < 0) goto out_freectx; @@ -470,7 +470,7 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req) wake_up(&ctx->wait); } -static void aio_fput_routine(void *data) +static void aio_fput_routine(struct work_struct *data) { spin_lock_irq(&fput_lock); while (likely(!list_empty(&fput_head))) { @@ -859,9 +859,9 @@ static inline void aio_run_all_iocbs(struct kioctx *ctx) * space. * Run on aiod's context. */ -static void aio_kick_handler(void *data) +static void aio_kick_handler(struct work_struct *work) { - struct kioctx *ctx = data; + struct kioctx *ctx = container_of(work, struct kioctx, wq.work); mm_segment_t oldfs = get_fs(); int requeue; diff --git a/fs/bio.c b/fs/bio.c index f95c8749499f..c6c07ca5b5a9 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -955,16 +955,16 @@ static void bio_release_pages(struct bio *bio) * run one bio_put() against the BIO. */ -static void bio_dirty_fn(void *data); +static void bio_dirty_fn(struct work_struct *work); -static DECLARE_WORK(bio_dirty_work, bio_dirty_fn, NULL); +static DECLARE_WORK(bio_dirty_work, bio_dirty_fn); static DEFINE_SPINLOCK(bio_dirty_lock); static struct bio *bio_dirty_list; /* * This runs in process context */ -static void bio_dirty_fn(void *data) +static void bio_dirty_fn(struct work_struct *work) { unsigned long flags; struct bio *bio; diff --git a/fs/file.c b/fs/file.c index 8e81775c5dc8..3787e82f54c1 100644 --- a/fs/file.c +++ b/fs/file.c @@ -91,8 +91,10 @@ out: spin_unlock(&fddef->lock); } -static void free_fdtable_work(struct fdtable_defer *f) +static void free_fdtable_work(struct work_struct *work) { + struct fdtable_defer *f = + container_of(work, struct fdtable_defer, wq); struct fdtable *fdt; spin_lock_bh(&f->lock); @@ -351,7 +353,7 @@ static void __devinit fdtable_defer_list_init(int cpu) { struct fdtable_defer *fddef = &per_cpu(fdtable_defer_list, cpu); spin_lock_init(&fddef->lock); - INIT_WORK(&fddef->wq, (void (*)(void *))free_fdtable_work, fddef); + INIT_WORK(&fddef->wq, free_fdtable_work); init_timer(&fddef->timer); fddef->timer.data = (unsigned long)fddef; fddef->timer.function = fdtable_timer; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 6f0487d6f44a..23ab145daa2d 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -143,7 +143,7 @@ static struct nfs_client *nfs_alloc_client(const char *hostname, INIT_LIST_HEAD(&clp->cl_state_owners); INIT_LIST_HEAD(&clp->cl_unused); spin_lock_init(&clp->cl_lock); - INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state, clp); + INIT_DELAYED_WORK(&clp->cl_renewd, nfs4_renew_state); rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); clp->cl_boot_time = CURRENT_TIME; clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 5ed798bc1cf7..371b804e7cc8 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -18,11 +18,10 @@ #define NFSDBG_FACILITY NFSDBG_VFS -static void nfs_expire_automounts(void *list); +static void nfs_expire_automounts(struct work_struct *work); LIST_HEAD(nfs_automount_list); -static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts, - &nfs_automount_list); +static DECLARE_DELAYED_WORK(nfs_automount_task, nfs_expire_automounts); int nfs_mountpoint_expiry_timeout = 500 * HZ; static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent, @@ -165,9 +164,9 @@ struct inode_operations nfs_referral_inode_operations = { .follow_link = nfs_follow_mountpoint, }; -static void nfs_expire_automounts(void *data) +static void nfs_expire_automounts(struct work_struct *work) { - struct list_head *list = (struct list_head *)data; + struct list_head *list = &nfs_automount_list; mark_mounts_for_expiry(list); if (!list_empty(list)) diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 6f346677332d..c26cd978c7cc 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -185,7 +185,7 @@ extern const u32 nfs4_fs_locations_bitmap[2]; extern void nfs4_schedule_state_renewal(struct nfs_client *); extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); extern void nfs4_kill_renewd(struct nfs_client *); -extern void nfs4_renew_state(void *); +extern void nfs4_renew_state(struct work_struct *); /* nfs4state.c */ struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp); diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c index 7b6df1852e75..823298561c0a 100644 --- a/fs/nfs/nfs4renewd.c +++ b/fs/nfs/nfs4renewd.c @@ -59,9 +59,10 @@ #define NFSDBG_FACILITY NFSDBG_PROC void -nfs4_renew_state(void *data) +nfs4_renew_state(struct work_struct *work) { - struct nfs_client *clp = (struct nfs_client *)data; + struct nfs_client *clp = + container_of(work, struct nfs_client, cl_renewd.work); struct rpc_cred *cred; long lease, timeout; unsigned long last, now; diff --git a/include/linux/libata.h b/include/linux/libata.h index 5f04006e8dd2..b3f32eadbef5 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -568,6 +568,7 @@ struct ata_port { struct ata_host *host; struct device *dev; + void *port_task_data; struct delayed_work port_task; struct delayed_work hotplug_task; struct work_struct scsi_rescan_task; @@ -747,7 +748,7 @@ extern int ata_ratelimit(void); extern unsigned int ata_busy_sleep(struct ata_port *ap, unsigned long timeout_pat, unsigned long timeout); -extern void ata_port_queue_task(struct ata_port *ap, void (*fn)(void *), +extern void ata_port_queue_task(struct ata_port *ap, work_func_t fn, void *data, unsigned long delay); extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val, unsigned long interval_msec, diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index ecc017d24cf3..4a3ea83c6d16 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -11,18 +11,19 @@ struct workqueue_struct; -typedef void (*work_func_t)(void *data); +struct work_struct; +typedef void (*work_func_t)(struct work_struct *work); struct work_struct { - /* the first word is the work queue pointer and the pending flag - * rolled into one */ + /* the first word is the work queue pointer and the flags rolled into + * one */ unsigned long management; #define WORK_STRUCT_PENDING 0 /* T if work item pending execution */ +#define WORK_STRUCT_NOAUTOREL 1 /* F if work item automatically released on exec */ #define WORK_STRUCT_FLAG_MASK (3UL) #define WORK_STRUCT_WQ_DATA_MASK (~WORK_STRUCT_FLAG_MASK) struct list_head entry; work_func_t func; - void *data; }; struct delayed_work { @@ -34,48 +35,77 @@ struct execute_work { struct work_struct work; }; -#define __WORK_INITIALIZER(n, f, d) { \ +#define __WORK_INITIALIZER(n, f) { \ + .management = 0, \ .entry = { &(n).entry, &(n).entry }, \ .func = (f), \ - .data = (d), \ } -#define __DELAYED_WORK_INITIALIZER(n, f, d) { \ - .work = __WORK_INITIALIZER((n).work, (f), (d)), \ +#define __WORK_INITIALIZER_NAR(n, f) { \ + .management = (1 << WORK_STRUCT_NOAUTOREL), \ + .entry = { &(n).entry, &(n).entry }, \ + .func = (f), \ + } + +#define __DELAYED_WORK_INITIALIZER(n, f) { \ + .work = __WORK_INITIALIZER((n).work, (f)), \ + .timer = TIMER_INITIALIZER(NULL, 0, 0), \ + } + +#define __DELAYED_WORK_INITIALIZER_NAR(n, f) { \ + .work = __WORK_INITIALIZER_NAR((n).work, (f)), \ .timer = TIMER_INITIALIZER(NULL, 0, 0), \ } -#define DECLARE_WORK(n, f, d) \ - struct work_struct n = __WORK_INITIALIZER(n, f, d) +#define DECLARE_WORK(n, f) \ + struct work_struct n = __WORK_INITIALIZER(n, f) + +#define DECLARE_WORK_NAR(n, f) \ + struct work_struct n = __WORK_INITIALIZER_NAR(n, f) -#define DECLARE_DELAYED_WORK(n, f, d) \ - struct delayed_work n = __DELAYED_WORK_INITIALIZER(n, f, d) +#define DECLARE_DELAYED_WORK(n, f) \ + struct delayed_work n = __DELAYED_WORK_INITIALIZER(n, f) + +#define DECLARE_DELAYED_WORK_NAR(n, f) \ + struct dwork_struct n = __DELAYED_WORK_INITIALIZER_NAR(n, f) /* - * initialize a work item's function and data pointers + * initialize a work item's function pointer */ -#define PREPARE_WORK(_work, _func, _data) \ +#define PREPARE_WORK(_work, _func) \ do { \ (_work)->func = (_func); \ - (_work)->data = (_data); \ } while (0) -#define PREPARE_DELAYED_WORK(_work, _func, _data) \ - PREPARE_WORK(&(_work)->work, (_func), (_data)) +#define PREPARE_DELAYED_WORK(_work, _func) \ + PREPARE_WORK(&(_work)->work, (_func)) /* * initialize all of a work item in one go */ -#define INIT_WORK(_work, _func, _data) \ +#define INIT_WORK(_work, _func) \ do { \ - INIT_LIST_HEAD(&(_work)->entry); \ (_work)->management = 0; \ - PREPARE_WORK((_work), (_func), (_data)); \ + INIT_LIST_HEAD(&(_work)->entry); \ + PREPARE_WORK((_work), (_func)); \ + } while (0) + +#define INIT_WORK_NAR(_work, _func) \ + do { \ + (_work)->management = (1 << WORK_STRUCT_NOAUTOREL); \ + INIT_LIST_HEAD(&(_work)->entry); \ + PREPARE_WORK((_work), (_func)); \ + } while (0) + +#define INIT_DELAYED_WORK(_work, _func) \ + do { \ + INIT_WORK(&(_work)->work, (_func)); \ + init_timer(&(_work)->timer); \ } while (0) -#define INIT_DELAYED_WORK(_work, _func, _data) \ +#define INIT_DELAYED_WORK_NAR(_work, _func) \ do { \ - INIT_WORK(&(_work)->work, (_func), (_data)); \ + INIT_WORK_NAR(&(_work)->work, (_func)); \ init_timer(&(_work)->timer); \ } while (0) @@ -94,6 +124,27 @@ struct execute_work { #define delayed_work_pending(work) \ test_bit(WORK_STRUCT_PENDING, &(work)->work.management) +/** + * work_release - Release a work item under execution + * @work: The work item to release + * + * This is used to release a work item that has been initialised with automatic + * release mode disabled (WORK_STRUCT_NOAUTOREL is set). This gives the work + * function the opportunity to grab auxiliary data from the container of the + * work_struct before clearing the pending bit as the work_struct may be + * subject to deallocation the moment the pending bit is cleared. + * + * In such a case, this should be called in the work function after it has + * fetched any data it may require from the containter of the work_struct. + * After this function has been called, the work_struct may be scheduled for + * further execution or it may be deallocated unless other precautions are + * taken. + * + * This should also be used to release a delayed work item. + */ +#define work_release(work) \ + clear_bit(WORK_STRUCT_PENDING, &(work)->management) + extern struct workqueue_struct *__create_workqueue(const char *name, int singlethread); @@ -112,7 +163,7 @@ extern int FASTCALL(schedule_work(struct work_struct *work)); extern int FASTCALL(schedule_delayed_work(struct delayed_work *work, unsigned long delay)); extern int schedule_delayed_work_on(int cpu, struct delayed_work *work, unsigned long delay); -extern int schedule_on_each_cpu(work_func_t func, void *info); +extern int schedule_on_each_cpu(work_func_t func); extern void flush_scheduled_work(void); extern int current_is_keventd(void); extern int keventd_up(void); @@ -121,7 +172,7 @@ extern void init_workqueues(void); void cancel_rearming_delayed_work(struct delayed_work *work); void cancel_rearming_delayed_workqueue(struct workqueue_struct *, struct delayed_work *); -int execute_in_process_context(work_func_t fn, void *, struct execute_work *); +int execute_in_process_context(work_func_t fn, struct execute_work *); /* * Kill off a pending schedule_delayed_work(). Note that the work callback diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 5f48748fe017..f7be1ac73601 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -84,7 +84,7 @@ struct inet_timewait_death_row { }; extern void inet_twdr_hangman(unsigned long data); -extern void inet_twdr_twkill_work(void *data); +extern void inet_twdr_twkill_work(struct work_struct *work); extern void inet_twdr_twcal_tick(unsigned long data); #if (BITS_PER_LONG == 64) diff --git a/ipc/util.c b/ipc/util.c index cd8bb14a431f..a9b7a227b8d4 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -514,6 +514,11 @@ void ipc_rcu_getref(void *ptr) container_of(ptr, struct ipc_rcu_hdr, data)->refcount++; } +static void ipc_do_vfree(struct work_struct *work) +{ + vfree(container_of(work, struct ipc_rcu_sched, work)); +} + /** * ipc_schedule_free - free ipc + rcu space * @head: RCU callback structure for queued work @@ -528,7 +533,7 @@ static void ipc_schedule_free(struct rcu_head *head) struct ipc_rcu_sched *sched = container_of(&(grace->data[0]), struct ipc_rcu_sched, data[0]); - INIT_WORK(&sched->work, vfree, sched); + INIT_WORK(&sched->work, ipc_do_vfree); schedule_work(&sched->work); } diff --git a/kernel/kmod.c b/kernel/kmod.c index bb4e29d924e4..7dc7a9dad6ac 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -114,6 +114,7 @@ EXPORT_SYMBOL(request_module); #endif /* CONFIG_KMOD */ struct subprocess_info { + struct work_struct work; struct completion *complete; char *path; char **argv; @@ -221,9 +222,10 @@ static int wait_for_helper(void *data) } /* This is run by khelper thread */ -static void __call_usermodehelper(void *data) +static void __call_usermodehelper(struct work_struct *work) { - struct subprocess_info *sub_info = data; + struct subprocess_info *sub_info = + container_of(work, struct subprocess_info, work); pid_t pid; int wait = sub_info->wait; @@ -264,6 +266,8 @@ int call_usermodehelper_keys(char *path, char **argv, char **envp, { DECLARE_COMPLETION_ONSTACK(done); struct subprocess_info sub_info = { + .work = __WORK_INITIALIZER(sub_info.work, + __call_usermodehelper), .complete = &done, .path = path, .argv = argv, @@ -272,7 +276,6 @@ int call_usermodehelper_keys(char *path, char **argv, char **envp, .wait = wait, .retval = 0, }; - DECLARE_WORK(work, __call_usermodehelper, &sub_info); if (!khelper_wq) return -EBUSY; @@ -280,7 +283,7 @@ int call_usermodehelper_keys(char *path, char **argv, char **envp, if (path[0] == '\0') return 0; - queue_work(khelper_wq, &work); + queue_work(khelper_wq, &sub_info.work); wait_for_completion(&done); return sub_info.retval; } @@ -291,6 +294,8 @@ int call_usermodehelper_pipe(char *path, char **argv, char **envp, { DECLARE_COMPLETION(done); struct subprocess_info sub_info = { + .work = __WORK_INITIALIZER(sub_info.work, + __call_usermodehelper), .complete = &done, .path = path, .argv = argv, @@ -298,7 +303,6 @@ int call_usermodehelper_pipe(char *path, char **argv, char **envp, .retval = 0, }; struct file *f; - DECLARE_WORK(work, __call_usermodehelper, &sub_info); if (!khelper_wq) return -EBUSY; @@ -318,7 +322,7 @@ int call_usermodehelper_pipe(char *path, char **argv, char **envp, } sub_info.stdin = f; - queue_work(khelper_wq, &work); + queue_work(khelper_wq, &sub_info.work); wait_for_completion(&done); return sub_info.retval; } diff --git a/kernel/kthread.c b/kernel/kthread.c index 4f9c60ef95e8..1db8c72d0d38 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -31,6 +31,8 @@ struct kthread_create_info /* Result passed back to kthread_create() from keventd. */ struct task_struct *result; struct completion done; + + struct work_struct work; }; struct kthread_stop_info @@ -111,9 +113,10 @@ static int kthread(void *_create) } /* We are keventd: create a thread. */ -static void keventd_create_kthread(void *_create) +static void keventd_create_kthread(struct work_struct *work) { - struct kthread_create_info *create = _create; + struct kthread_create_info *create = + container_of(work, struct kthread_create_info, work); int pid; /* We want our own signal handler (we take no signals by default). */ @@ -154,20 +157,20 @@ struct task_struct *kthread_create(int (*threadfn)(void *data), ...) { struct kthread_create_info create; - DECLARE_WORK(work, keventd_create_kthread, &create); create.threadfn = threadfn; create.data = data; init_completion(&create.started); init_completion(&create.done); + INIT_WORK(&create.work, keventd_create_kthread); /* * The workqueue needs to start up first: */ if (!helper_wq) - work.func(work.data); + create.work.func(&create.work); else { - queue_work(helper_wq, &work); + queue_work(helper_wq, &create.work); wait_for_completion(&create.done); } if (!IS_ERR(create.result)) { diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c index f1f900ac3164..678ec736076b 100644 --- a/kernel/power/poweroff.c +++ b/kernel/power/poweroff.c @@ -16,12 +16,12 @@ * callback we use. */ -static void do_poweroff(void *dummy) +static void do_poweroff(struct work_struct *dummy) { kernel_power_off(); } -static DECLARE_WORK(poweroff_work, do_poweroff, NULL); +static DECLARE_WORK(poweroff_work, do_poweroff); static void handle_poweroff(int key, struct tty_struct *tty) { diff --git a/kernel/sys.c b/kernel/sys.c index 98489d82801b..c87b461de38d 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -880,7 +880,7 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user return 0; } -static void deferred_cad(void *dummy) +static void deferred_cad(struct work_struct *dummy) { kernel_restart(NULL); } @@ -892,7 +892,7 @@ static void deferred_cad(void *dummy) */ void ctrl_alt_del(void) { - static DECLARE_WORK(cad_work, deferred_cad, NULL); + static DECLARE_WORK(cad_work, deferred_cad); if (C_A_D) schedule_work(&cad_work); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 967479756511..8d1e7cb8a51a 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -241,14 +241,14 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) struct work_struct *work = list_entry(cwq->worklist.next, struct work_struct, entry); work_func_t f = work->func; - void *data = work->data; list_del_init(cwq->worklist.next); spin_unlock_irqrestore(&cwq->lock, flags); BUG_ON(get_wq_data(work) != cwq); - clear_bit(WORK_STRUCT_PENDING, &work->management); - f(data); + if (!test_bit(WORK_STRUCT_NOAUTOREL, &work->management)) + work_release(work); + f(work); spin_lock_irqsave(&cwq->lock, flags); cwq->remove_sequence++; @@ -527,7 +527,6 @@ EXPORT_SYMBOL(schedule_delayed_work_on); /** * schedule_on_each_cpu - call a function on each online CPU from keventd * @func: the function to call - * @info: a pointer to pass to func() * * Returns zero on success. * Returns -ve errno on failure. @@ -536,7 +535,7 @@ EXPORT_SYMBOL(schedule_delayed_work_on); * * schedule_on_each_cpu() is very slow. */ -int schedule_on_each_cpu(work_func_t func, void *info) +int schedule_on_each_cpu(work_func_t func) { int cpu; struct work_struct *works; @@ -547,7 +546,7 @@ int schedule_on_each_cpu(work_func_t func, void *info) mutex_lock(&workqueue_mutex); for_each_online_cpu(cpu) { - INIT_WORK(per_cpu_ptr(works, cpu), func, info); + INIT_WORK(per_cpu_ptr(works, cpu), func); __queue_work(per_cpu_ptr(keventd_wq->cpu_wq, cpu), per_cpu_ptr(works, cpu)); } @@ -591,7 +590,6 @@ EXPORT_SYMBOL(cancel_rearming_delayed_work); /** * execute_in_process_context - reliably execute the routine with user context * @fn: the function to execute - * @data: data to pass to the function * @ew: guaranteed storage for the execute work structure (must * be available when the work executes) * @@ -601,15 +599,14 @@ EXPORT_SYMBOL(cancel_rearming_delayed_work); * Returns: 0 - function was executed * 1 - function was scheduled for execution */ -int execute_in_process_context(work_func_t fn, void *data, - struct execute_work *ew) +int execute_in_process_context(work_func_t fn, struct execute_work *ew) { if (!in_interrupt()) { - fn(data); + fn(&ew->work); return 0; } - INIT_WORK(&ew->work, fn, data); + INIT_WORK(&ew->work, fn); schedule_work(&ew->work); return 1; diff --git a/mm/slab.c b/mm/slab.c index a65bc5e992c3..5de81473df34 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -313,7 +313,7 @@ static int drain_freelist(struct kmem_cache *cache, static void free_block(struct kmem_cache *cachep, void **objpp, int len, int node); static int enable_cpucache(struct kmem_cache *cachep); -static void cache_reap(void *unused); +static void cache_reap(struct work_struct *unused); /* * This function must be completely optimized away if a constant is passed to @@ -925,7 +925,7 @@ static void __devinit start_cpu_timer(int cpu) */ if (keventd_up() && reap_work->work.func == NULL) { init_reap_node(cpu); - INIT_DELAYED_WORK(reap_work, cache_reap, NULL); + INIT_DELAYED_WORK(reap_work, cache_reap); schedule_delayed_work_on(cpu, reap_work, HZ + 3 * cpu); } } @@ -3815,7 +3815,7 @@ void drain_array(struct kmem_cache *cachep, struct kmem_list3 *l3, * If we cannot acquire the cache chain mutex then just give up - we'll try * again on the next iteration. */ -static void cache_reap(void *unused) +static void cache_reap(struct work_struct *unused) { struct kmem_cache *searchp; struct kmem_list3 *l3; diff --git a/net/core/link_watch.c b/net/core/link_watch.c index f2ed09e25dfd..549a2ce951b0 100644 --- a/net/core/link_watch.c +++ b/net/core/link_watch.c @@ -34,8 +34,8 @@ enum lw_bits { static unsigned long linkwatch_flags; static unsigned long linkwatch_nextevent; -static void linkwatch_event(void *dummy); -static DECLARE_DELAYED_WORK(linkwatch_work, linkwatch_event, NULL); +static void linkwatch_event(struct work_struct *dummy); +static DECLARE_DELAYED_WORK(linkwatch_work, linkwatch_event); static LIST_HEAD(lweventlist); static DEFINE_SPINLOCK(lweventlist_lock); @@ -127,7 +127,7 @@ void linkwatch_run_queue(void) } -static void linkwatch_event(void *dummy) +static void linkwatch_event(struct work_struct *dummy) { /* Limit the number of linkwatch events to one * per second so that a runaway driver does not diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index cdd805344c61..8c74f9168b7d 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -197,9 +197,10 @@ EXPORT_SYMBOL_GPL(inet_twdr_hangman); extern void twkill_slots_invalid(void); -void inet_twdr_twkill_work(void *data) +void inet_twdr_twkill_work(struct work_struct *work) { - struct inet_timewait_death_row *twdr = data; + struct inet_timewait_death_row *twdr = + container_of(work, struct inet_timewait_death_row, twkill_work); int i; if ((INET_TWDR_TWKILL_SLOTS - 1) > (sizeof(twdr->thread_slots) * 8)) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0163d9826907..af7b2c986b1f 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -45,8 +45,7 @@ struct inet_timewait_death_row tcp_death_row = { .tw_timer = TIMER_INITIALIZER(inet_twdr_hangman, 0, (unsigned long)&tcp_death_row), .twkill_work = __WORK_INITIALIZER(tcp_death_row.twkill_work, - inet_twdr_twkill_work, - &tcp_death_row), + inet_twdr_twkill_work), /* Short-time timewait calendar */ .twcal_hand = -1, diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index d5725cb1491e..d96fd466a9a4 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -284,8 +284,8 @@ static struct file_operations cache_file_operations; static struct file_operations content_file_operations; static struct file_operations cache_flush_operations; -static void do_cache_clean(void *data); -static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean, NULL); +static void do_cache_clean(struct work_struct *work); +static DECLARE_DELAYED_WORK(cache_cleaner, do_cache_clean); void cache_register(struct cache_detail *cd) { @@ -461,7 +461,7 @@ static int cache_clean(void) /* * We want to regularly clean the cache, so we need to schedule some work ... */ -static void do_cache_clean(void *data) +static void do_cache_clean(struct work_struct *work) { int delay = 5; if (cache_clean() == -1) diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 97be3f7fed44..49dba5febbbd 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -54,10 +54,11 @@ static void rpc_purge_list(struct rpc_inode *rpci, struct list_head *head, } static void -rpc_timeout_upcall_queue(void *data) +rpc_timeout_upcall_queue(struct work_struct *work) { LIST_HEAD(free_list); - struct rpc_inode *rpci = (struct rpc_inode *)data; + struct rpc_inode *rpci = + container_of(work, struct rpc_inode, queue_timeout.work); struct inode *inode = &rpci->vfs_inode; void (*destroy_msg)(struct rpc_pipe_msg *); @@ -838,7 +839,7 @@ init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) rpci->pipelen = 0; init_waitqueue_head(&rpci->waitq); INIT_DELAYED_WORK(&rpci->queue_timeout, - rpc_timeout_upcall_queue, rpci); + rpc_timeout_upcall_queue); rpci->ops = NULL; } } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index a1ab4eed41f4..eff44bcdc95a 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -41,7 +41,7 @@ static mempool_t *rpc_buffer_mempool __read_mostly; static void __rpc_default_timer(struct rpc_task *task); static void rpciod_killall(void); -static void rpc_async_schedule(void *); +static void rpc_async_schedule(struct work_struct *); /* * RPC tasks sit here while waiting for conditions to improve. @@ -305,7 +305,7 @@ static void rpc_make_runnable(struct rpc_task *task) if (RPC_IS_ASYNC(task)) { int status; - INIT_WORK(&task->u.tk_work, rpc_async_schedule, (void *)task); + INIT_WORK(&task->u.tk_work, rpc_async_schedule); status = queue_work(task->tk_workqueue, &task->u.tk_work); if (status < 0) { printk(KERN_WARNING "RPC: failed to add task to queue: error: %d!\n", status); @@ -695,9 +695,9 @@ rpc_execute(struct rpc_task *task) return __rpc_execute(task); } -static void rpc_async_schedule(void *arg) +static void rpc_async_schedule(struct work_struct *work) { - __rpc_execute((struct rpc_task *)arg); + __rpc_execute(container_of(work, struct rpc_task, u.tk_work)); } /** diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 80857470dc11..4f9a5d9791fb 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -479,9 +479,10 @@ int xprt_adjust_timeout(struct rpc_rqst *req) return status; } -static void xprt_autoclose(void *args) +static void xprt_autoclose(struct work_struct *work) { - struct rpc_xprt *xprt = (struct rpc_xprt *)args; + struct rpc_xprt *xprt = + container_of(work, struct rpc_xprt, task_cleanup); xprt_disconnect(xprt); xprt->ops->close(xprt); @@ -932,7 +933,7 @@ struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t si INIT_LIST_HEAD(&xprt->free); INIT_LIST_HEAD(&xprt->recv); - INIT_WORK(&xprt->task_cleanup, xprt_autoclose, xprt); + INIT_WORK(&xprt->task_cleanup, xprt_autoclose); init_timer(&xprt->timer); xprt->timer.function = xprt_init_autodisconnect; xprt->timer.data = (unsigned long) xprt; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 3c7532cd009e..cfe3c15be948 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1060,13 +1060,14 @@ static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock) /** * xs_udp_connect_worker - set up a UDP socket - * @args: RPC transport to connect + * @work: RPC transport to connect * * Invoked by a work queue tasklet. */ -static void xs_udp_connect_worker(void *args) +static void xs_udp_connect_worker(struct work_struct *work) { - struct rpc_xprt *xprt = (struct rpc_xprt *) args; + struct rpc_xprt *xprt = + container_of(work, struct rpc_xprt, connect_worker.work); struct socket *sock = xprt->sock; int err, status = -EIO; @@ -1144,13 +1145,14 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) /** * xs_tcp_connect_worker - connect a TCP socket to a remote endpoint - * @args: RPC transport to connect + * @work: RPC transport to connect * * Invoked by a work queue tasklet. */ -static void xs_tcp_connect_worker(void *args) +static void xs_tcp_connect_worker(struct work_struct *work) { - struct rpc_xprt *xprt = (struct rpc_xprt *)args; + struct rpc_xprt *xprt = + container_of(work, struct rpc_xprt, connect_worker.work); struct socket *sock = xprt->sock; int err, status = -EIO; @@ -1375,7 +1377,7 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to) /* XXX: header size can vary due to auth type, IPv6, etc. */ xprt->max_payload = (1U << 16) - (MAX_HEADER << 3); - INIT_DELAYED_WORK(&xprt->connect_worker, xs_udp_connect_worker, xprt); + INIT_DELAYED_WORK(&xprt->connect_worker, xs_udp_connect_worker); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_UDP_CONN_TO; xprt->reestablish_timeout = XS_UDP_REEST_TO; @@ -1420,7 +1422,7 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to) xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; - INIT_DELAYED_WORK(&xprt->connect_worker, xs_tcp_connect_worker, xprt); + INIT_DELAYED_WORK(&xprt->connect_worker, xs_tcp_connect_worker); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_TCP_CONN_TO; xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; diff --git a/security/keys/key.c b/security/keys/key.c index 80de8c3e9cc3..70eacbe5abde 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -30,8 +30,8 @@ DEFINE_SPINLOCK(key_user_lock); static LIST_HEAD(key_types_list); static DECLARE_RWSEM(key_types_sem); -static void key_cleanup(void *data); -static DECLARE_WORK(key_cleanup_task, key_cleanup, NULL); +static void key_cleanup(struct work_struct *work); +static DECLARE_WORK(key_cleanup_task, key_cleanup); /* we serialise key instantiation and link */ DECLARE_RWSEM(key_construction_sem); @@ -552,7 +552,7 @@ EXPORT_SYMBOL(key_negate_and_link); * do cleaning up in process context so that we don't have to disable * interrupts all over the place */ -static void key_cleanup(void *data) +static void key_cleanup(struct work_struct *work) { struct rb_node *_n; struct key *key; -- cgit v1.2.3 From c4028958b6ecad064b1a6303a6a5906d4fe48d73 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 22 Nov 2006 14:57:56 +0000 Subject: WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells --- arch/i386/kernel/cpu/mcheck/non-fatal.c | 6 +- arch/i386/kernel/smpboot.c | 11 +- arch/i386/kernel/tsc.c | 4 +- arch/powerpc/platforms/pseries/eeh_event.c | 6 +- drivers/atm/idt77252.c | 9 +- drivers/block/aoe/aoe.h | 2 +- drivers/block/aoe/aoecmd.c | 4 +- drivers/block/aoe/aoedev.c | 2 +- drivers/block/paride/pd.c | 8 +- drivers/block/paride/pseudo.h | 10 +- drivers/block/sx8.c | 7 +- drivers/block/ub.c | 8 +- drivers/bluetooth/bcm203x.c | 7 +- drivers/char/cyclades.c | 9 +- drivers/char/drm/via_dmablit.c | 6 +- drivers/char/epca.c | 8 +- drivers/char/esp.c | 14 +- drivers/char/genrtc.c | 4 +- drivers/char/hvsi.c | 16 +- drivers/char/ip2/i2lib.c | 12 +- drivers/char/ip2/ip2main.c | 23 +-- drivers/char/isicom.c | 12 +- drivers/char/moxa.c | 8 +- drivers/char/mxser.c | 9 +- drivers/char/pcmcia/synclink_cs.c | 8 +- drivers/char/sonypi.c | 4 +- drivers/char/specialix.c | 14 +- drivers/char/synclink.c | 9 +- drivers/char/synclink_gt.c | 10 +- drivers/char/synclinkmp.c | 8 +- drivers/char/tpm/tpm.c | 6 +- drivers/connector/cn_queue.c | 8 +- drivers/connector/connector.c | 31 ++-- drivers/cpufreq/cpufreq_conservative.c | 7 +- drivers/cpufreq/cpufreq_ondemand.c | 28 +-- drivers/i2c/chips/ds1374.c | 12 +- drivers/ieee1394/hosts.c | 9 +- drivers/ieee1394/hosts.h | 2 +- drivers/ieee1394/sbp2.c | 28 +-- drivers/ieee1394/sbp2.h | 2 +- drivers/infiniband/core/addr.c | 6 +- drivers/infiniband/core/cache.c | 7 +- drivers/infiniband/core/cm.c | 19 +- drivers/infiniband/core/cma.c | 10 +- drivers/infiniband/core/iwcm.c | 7 +- drivers/infiniband/core/mad.c | 25 +-- drivers/infiniband/core/mad_priv.h | 2 +- drivers/infiniband/core/mad_rmpp.c | 18 +- drivers/infiniband/core/sa_query.c | 10 +- drivers/infiniband/core/uverbs_mem.c | 7 +- drivers/infiniband/hw/ipath/ipath_user_pages.c | 7 +- drivers/infiniband/hw/mthca/mthca_catas.c | 4 +- drivers/infiniband/ulp/ipoib/ipoib.h | 16 +- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 25 +-- drivers/infiniband/ulp/ipoib/ipoib_main.c | 10 +- drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 22 ++- drivers/infiniband/ulp/iser/iser_verbs.c | 10 +- drivers/infiniband/ulp/srp/ib_srp.c | 7 +- drivers/input/keyboard/lkkbd.c | 6 +- drivers/input/keyboard/sunkbd.c | 6 +- drivers/input/mouse/psmouse-base.c | 7 +- drivers/isdn/act2000/capi.c | 4 +- drivers/isdn/act2000/capi.h | 2 +- drivers/isdn/act2000/module.c | 18 +- drivers/isdn/capi/kcapi.c | 14 +- drivers/isdn/hisax/amd7930_fn.c | 7 +- drivers/isdn/hisax/config.c | 9 +- drivers/isdn/hisax/hfc4s8s_l1.c | 5 +- drivers/isdn/hisax/hfc_2bds0.c | 9 +- drivers/isdn/hisax/hfc_pci.c | 6 +- drivers/isdn/hisax/hfc_sx.c | 6 +- drivers/isdn/hisax/icc.c | 6 +- drivers/isdn/hisax/isac.c | 6 +- drivers/isdn/hisax/isar.c | 6 +- drivers/isdn/hisax/isdnl1.c | 6 +- drivers/isdn/hisax/w6692.c | 6 +- drivers/isdn/i4l/isdn_net.c | 6 +- drivers/isdn/pcbit/drv.c | 4 +- drivers/isdn/pcbit/layer2.c | 6 +- drivers/isdn/pcbit/pcbit.h | 2 + drivers/macintosh/smu.c | 4 +- drivers/md/dm-crypt.c | 8 +- drivers/md/dm-mpath.c | 18 +- drivers/md/dm-raid1.c | 4 +- drivers/md/dm-snap.c | 9 +- drivers/md/kcopyd.c | 4 +- drivers/media/dvb/b2c2/flexcop-pci.c | 9 +- drivers/media/dvb/cinergyT2/cinergyT2.c | 18 +- drivers/media/dvb/dvb-core/dvb_net.c | 19 +- drivers/media/dvb/dvb-usb/dvb-usb-remote.c | 7 +- drivers/media/dvb/dvb-usb/dvb-usb.h | 2 +- drivers/media/video/cpia_pp.c | 20 ++- drivers/media/video/cx88/cx88-input.c | 6 +- drivers/media/video/ir-kbd-i2c.c | 6 +- drivers/media/video/pvrusb2/pvrusb2-context.c | 13 +- drivers/media/video/saa6588.c | 6 +- drivers/media/video/saa7134/saa7134-empress.c | 9 +- drivers/message/fusion/mptfc.c | 14 +- drivers/message/fusion/mptlan.c | 29 +-- drivers/message/fusion/mptsas.c | 25 +-- drivers/message/fusion/mptspi.c | 14 +- drivers/message/i2o/driver.c | 2 +- drivers/message/i2o/exec-osm.c | 13 +- drivers/message/i2o/i2o_block.c | 15 +- drivers/message/i2o/i2o_block.h | 2 +- drivers/misc/tifm_7xx1.c | 18 +- drivers/mmc/mmc.c | 14 +- drivers/mmc/mmc.h | 2 +- drivers/mmc/mmc_sysfs.c | 10 +- drivers/mmc/tifm_sd.c | 28 +-- drivers/net/8139too.c | 26 +-- drivers/net/bnx2.c | 6 +- drivers/net/cassini.c | 6 +- drivers/net/chelsio/common.h | 2 +- drivers/net/chelsio/cxgb2.c | 16 +- drivers/net/e100.c | 8 +- drivers/net/ehea/ehea_main.c | 9 +- drivers/net/hamradio/baycom_epp.c | 14 +- drivers/net/irda/mcs7780.c | 6 +- drivers/net/irda/sir-dev.h | 2 +- drivers/net/irda/sir_dev.c | 8 +- drivers/net/iseries_veth.c | 12 +- drivers/net/ixgb/ixgb_main.c | 10 +- drivers/net/myri10ge/myri10ge.c | 7 +- drivers/net/ns83820.c | 10 +- drivers/net/pcmcia/xirc2ps_cs.c | 12 +- drivers/net/phy/phy.c | 9 +- drivers/net/plip.c | 38 ++-- drivers/net/qla3xxx.c | 20 ++- drivers/net/qla3xxx.h | 4 +- drivers/net/r8169.c | 23 ++- drivers/net/s2io.c | 16 +- drivers/net/s2io.h | 2 +- drivers/net/sis190.c | 13 +- drivers/net/skge.c | 15 +- drivers/net/skge.h | 2 +- drivers/net/spider_net.c | 9 +- drivers/net/sungem.c | 6 +- drivers/net/tg3.c | 6 +- drivers/net/tlan.c | 23 ++- drivers/net/tlan.h | 1 + drivers/net/tulip/21142.c | 7 +- drivers/net/tulip/timer.c | 7 +- drivers/net/tulip/tulip.h | 7 +- drivers/net/tulip/tulip_core.c | 3 +- drivers/net/wan/pc300_tty.c | 23 +-- drivers/net/wireless/bcm43xx/bcm43xx.h | 2 +- drivers/net/wireless/bcm43xx/bcm43xx_main.c | 20 ++- drivers/net/wireless/hostap/hostap.h | 2 +- drivers/net/wireless/hostap/hostap_ap.c | 19 +- drivers/net/wireless/hostap/hostap_hw.c | 21 +-- drivers/net/wireless/hostap/hostap_info.c | 6 +- drivers/net/wireless/hostap/hostap_main.c | 8 +- drivers/net/wireless/ipw2100.c | 47 ++--- drivers/net/wireless/ipw2100.h | 10 +- drivers/net/wireless/ipw2200.c | 227 +++++++++++++----------- drivers/net/wireless/ipw2200.h | 16 +- drivers/net/wireless/orinoco.c | 28 +-- drivers/net/wireless/prism54/isl_ioctl.c | 8 +- drivers/net/wireless/prism54/isl_ioctl.h | 4 +- drivers/net/wireless/prism54/islpci_dev.c | 5 +- drivers/net/wireless/prism54/islpci_eth.c | 4 +- drivers/net/wireless/prism54/islpci_eth.h | 2 +- drivers/net/wireless/prism54/islpci_mgt.c | 2 +- drivers/net/wireless/zd1211rw/zd_mac.c | 7 +- drivers/net/wireless/zd1211rw/zd_mac.h | 2 +- drivers/oprofile/cpu_buffer.c | 9 +- drivers/oprofile/cpu_buffer.h | 2 +- drivers/pci/hotplug/shpchp.h | 4 +- drivers/pci/hotplug/shpchp_core.c | 2 +- drivers/pci/hotplug/shpchp_ctrl.c | 19 +- drivers/pcmcia/ds.c | 7 +- drivers/rtc/rtc-dev.c | 7 +- drivers/scsi/NCR5380.c | 11 +- drivers/scsi/NCR5380.h | 4 +- drivers/scsi/aha152x.c | 4 +- drivers/scsi/imm.c | 12 +- drivers/scsi/ipr.c | 9 +- drivers/scsi/libiscsi.c | 7 +- drivers/scsi/libsas/sas_discover.c | 22 ++- drivers/scsi/libsas/sas_event.c | 14 +- drivers/scsi/libsas/sas_init.c | 6 +- drivers/scsi/libsas/sas_internal.h | 12 +- drivers/scsi/libsas/sas_phy.c | 45 +++-- drivers/scsi/libsas/sas_port.c | 30 ++-- drivers/scsi/ppa.c | 12 +- drivers/scsi/qla4xxx/ql4_os.c | 7 +- drivers/scsi/scsi_transport_fc.c | 60 ++++--- drivers/scsi/scsi_transport_iscsi.c | 8 +- drivers/scsi/scsi_transport_spi.c | 7 +- drivers/spi/spi_bitbang.c | 7 +- drivers/usb/atm/cxacru.c | 12 +- drivers/usb/atm/speedtch.c | 15 +- drivers/usb/atm/ueagle-atm.c | 6 +- drivers/usb/class/cdc-acm.c | 6 +- drivers/usb/core/hub.c | 18 +- drivers/usb/core/hub.h | 2 +- drivers/usb/core/message.c | 7 +- drivers/usb/core/usb.c | 9 +- drivers/usb/gadget/ether.c | 6 +- drivers/usb/host/u132-hcd.c | 62 +++---- drivers/usb/input/hid-core.c | 7 +- drivers/usb/misc/ftdi-elan.c | 86 +++------ drivers/usb/misc/phidgetkit.c | 21 ++- drivers/usb/misc/phidgetmotorcontrol.c | 11 +- drivers/usb/net/kaweth.c | 9 +- drivers/usb/net/pegasus.c | 6 +- drivers/usb/net/pegasus.h | 2 +- drivers/usb/net/usbnet.c | 7 +- drivers/usb/serial/aircable.c | 13 +- drivers/usb/serial/digi_acceleport.c | 14 +- drivers/usb/serial/ftdi_sio.c | 19 +- drivers/usb/serial/keyspan_pda.c | 22 ++- drivers/usb/serial/usb-serial.c | 7 +- drivers/usb/serial/whiteheat.c | 15 +- drivers/video/console/fbcon.c | 6 +- fs/9p/mux.c | 16 +- fs/gfs2/glock.c | 8 +- fs/ncpfs/inode.c | 8 +- fs/ncpfs/sock.c | 20 ++- fs/nfsd/nfs4state.c | 7 +- fs/ocfs2/alloc.c | 9 +- fs/ocfs2/cluster/heartbeat.c | 10 +- fs/ocfs2/cluster/quorum.c | 4 +- fs/ocfs2/cluster/tcp.c | 78 ++++---- fs/ocfs2/cluster/tcp_internal.h | 8 +- fs/ocfs2/dlm/dlmcommon.h | 2 +- fs/ocfs2/dlm/dlmdomain.c | 2 +- fs/ocfs2/dlm/dlmrecovery.c | 5 +- fs/ocfs2/dlm/userdlm.c | 10 +- fs/ocfs2/journal.c | 7 +- fs/ocfs2/journal.h | 2 +- fs/ocfs2/ocfs2.h | 2 +- fs/ocfs2/super.c | 2 +- fs/reiserfs/journal.c | 12 +- fs/xfs/linux-2.6/xfs_aops.c | 21 ++- fs/xfs/linux-2.6/xfs_buf.c | 9 +- include/linux/connector.h | 4 +- include/linux/i2o.h | 2 +- include/linux/mmc/host.h | 2 +- include/linux/ncp_fs_sb.h | 8 +- include/linux/reiserfs_fs_sb.h | 3 +- include/linux/relay.h | 2 +- include/linux/usb.h | 2 +- include/net/ieee80211softmac.h | 4 +- include/net/sctp/structs.h | 2 +- include/scsi/libsas.h | 23 ++- include/scsi/scsi_transport_fc.h | 4 +- include/scsi/scsi_transport_iscsi.h | 2 +- include/sound/ac97_codec.h | 2 +- include/sound/ak4114.h | 2 +- kernel/relay.c | 10 +- mm/swap.c | 4 +- net/atm/lec.c | 9 +- net/atm/lec.h | 2 +- net/bluetooth/hci_sysfs.c | 12 +- net/bridge/br_if.c | 10 +- net/bridge/br_private.h | 2 +- net/core/netpoll.c | 4 +- net/dccp/minisocks.c | 3 +- net/ieee80211/softmac/ieee80211softmac_assoc.c | 18 +- net/ieee80211/softmac/ieee80211softmac_auth.c | 23 +-- net/ieee80211/softmac/ieee80211softmac_event.c | 12 +- net/ieee80211/softmac/ieee80211softmac_module.c | 4 +- net/ieee80211/softmac/ieee80211softmac_priv.h | 13 +- net/ieee80211/softmac/ieee80211softmac_scan.c | 13 +- net/ieee80211/softmac/ieee80211softmac_wx.c | 6 +- net/ipv4/ipvs/ip_vs_ctl.c | 6 +- net/irda/ircomm/ircomm_tty.c | 11 +- net/sctp/associola.c | 11 +- net/sctp/endpointola.c | 10 +- net/sctp/inqueue.c | 9 +- net/xfrm/xfrm_policy.c | 8 +- net/xfrm/xfrm_state.c | 8 +- sound/aoa/aoa-gpio.h | 2 +- sound/aoa/core/snd-aoa-gpio-feature.c | 16 +- sound/aoa/core/snd-aoa-gpio-pmf.c | 16 +- sound/i2c/other/ak4114.c | 8 +- sound/pci/ac97/ac97_codec.c | 7 +- sound/pci/hda/hda_codec.c | 10 +- sound/pci/hda/hda_local.h | 1 + sound/ppc/tumbler.c | 8 +- 282 files changed, 1771 insertions(+), 1450 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/cpu/mcheck/non-fatal.c b/arch/i386/kernel/cpu/mcheck/non-fatal.c index 1f9153ae5b03..6b5d3518a1c0 100644 --- a/arch/i386/kernel/cpu/mcheck/non-fatal.c +++ b/arch/i386/kernel/cpu/mcheck/non-fatal.c @@ -51,10 +51,10 @@ static void mce_checkregs (void *info) } } -static void mce_work_fn(void *data); -static DECLARE_WORK(mce_work, mce_work_fn, NULL); +static void mce_work_fn(struct work_struct *work); +static DECLARE_DELAYED_WORK(mce_work, mce_work_fn); -static void mce_work_fn(void *data) +static void mce_work_fn(struct work_struct *work) { on_each_cpu(mce_checkregs, NULL, 1, 1); schedule_delayed_work(&mce_work, MCE_RATE); diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index 4bb8b77cd65b..02a9b66b6ac3 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1049,13 +1049,15 @@ void cpu_exit_clear(void) struct warm_boot_cpu_info { struct completion *complete; + struct work_struct task; int apicid; int cpu; }; -static void __cpuinit do_warm_boot_cpu(void *p) +static void __cpuinit do_warm_boot_cpu(struct work_struct *work) { - struct warm_boot_cpu_info *info = p; + struct warm_boot_cpu_info *info = + container_of(work, struct warm_boot_cpu_info, task); do_boot_cpu(info->apicid, info->cpu); complete(info->complete); } @@ -1064,7 +1066,6 @@ static int __cpuinit __smp_prepare_cpu(int cpu) { DECLARE_COMPLETION_ONSTACK(done); struct warm_boot_cpu_info info; - struct work_struct task; int apicid, ret; struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu); @@ -1089,7 +1090,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) info.complete = &done; info.apicid = apicid; info.cpu = cpu; - INIT_WORK(&task, do_warm_boot_cpu, &info); + INIT_WORK(&info.task, do_warm_boot_cpu); tsc_sync_disabled = 1; @@ -1097,7 +1098,7 @@ static int __cpuinit __smp_prepare_cpu(int cpu) clone_pgd_range(swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS, KERNEL_PGD_PTRS); flush_tlb_all(); - schedule_work(&task); + schedule_work(&info.task); wait_for_completion(&done); tsc_sync_disabled = 0; diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c index fbc95828cd74..9810c8c90750 100644 --- a/arch/i386/kernel/tsc.c +++ b/arch/i386/kernel/tsc.c @@ -217,7 +217,7 @@ static unsigned int cpufreq_delayed_issched = 0; static unsigned int cpufreq_init = 0; static struct work_struct cpufreq_delayed_get_work; -static void handle_cpufreq_delayed_get(void *v) +static void handle_cpufreq_delayed_get(struct work_struct *work) { unsigned int cpu; @@ -306,7 +306,7 @@ static int __init cpufreq_tsc(void) { int ret; - INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get, NULL); + INIT_WORK(&cpufreq_delayed_get_work, handle_cpufreq_delayed_get); ret = cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); if (!ret) diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c index 137077451316..49037edf7d39 100644 --- a/arch/powerpc/platforms/pseries/eeh_event.c +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -37,8 +37,8 @@ /* EEH event workqueue setup. */ static DEFINE_SPINLOCK(eeh_eventlist_lock); LIST_HEAD(eeh_eventlist); -static void eeh_thread_launcher(void *); -DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); +static void eeh_thread_launcher(struct work_struct *); +DECLARE_WORK(eeh_event_wq, eeh_thread_launcher); /* Serialize reset sequences for a given pci device */ DEFINE_MUTEX(eeh_event_mutex); @@ -103,7 +103,7 @@ static int eeh_event_handler(void * dummy) * eeh_thread_launcher * @dummy - unused */ -static void eeh_thread_launcher(void *dummy) +static void eeh_thread_launcher(struct work_struct *dummy) { if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0) printk(KERN_ERR "Failed to start EEH daemon\n"); diff --git a/drivers/atm/idt77252.c b/drivers/atm/idt77252.c index 87b17c33b3f9..f40786121948 100644 --- a/drivers/atm/idt77252.c +++ b/drivers/atm/idt77252.c @@ -135,7 +135,7 @@ static int idt77252_change_qos(struct atm_vcc *vcc, struct atm_qos *qos, int flags); static int idt77252_proc_read(struct atm_dev *dev, loff_t * pos, char *page); -static void idt77252_softint(void *dev_id); +static void idt77252_softint(struct work_struct *work); static struct atmdev_ops idt77252_ops = @@ -2866,9 +2866,10 @@ out: } static void -idt77252_softint(void *dev_id) +idt77252_softint(struct work_struct *work) { - struct idt77252_dev *card = dev_id; + struct idt77252_dev *card = + container_of(work, struct idt77252_dev, tqueue); u32 stat; int done; @@ -3697,7 +3698,7 @@ idt77252_init_one(struct pci_dev *pcidev, const struct pci_device_id *id) card->pcidev = pcidev; sprintf(card->name, "idt77252-%d", card->index); - INIT_WORK(&card->tqueue, idt77252_softint, (void *)card); + INIT_WORK(&card->tqueue, idt77252_softint); membase = pci_resource_start(pcidev, 1); srambase = pci_resource_start(pcidev, 2); diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h index 6d111228cfac..2308e83e5f33 100644 --- a/drivers/block/aoe/aoe.h +++ b/drivers/block/aoe/aoe.h @@ -159,7 +159,7 @@ void aoecmd_work(struct aoedev *d); void aoecmd_cfg(ushort aoemajor, unsigned char aoeminor); void aoecmd_ata_rsp(struct sk_buff *); void aoecmd_cfg_rsp(struct sk_buff *); -void aoecmd_sleepwork(void *vp); +void aoecmd_sleepwork(struct work_struct *); struct sk_buff *new_skb(ulong); int aoedev_init(void); diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 8a13b1af8bab..97f7f535f412 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -408,9 +408,9 @@ rexmit_timer(ulong vp) /* this function performs work that has been deferred until sleeping is OK */ void -aoecmd_sleepwork(void *vp) +aoecmd_sleepwork(struct work_struct *work) { - struct aoedev *d = (struct aoedev *) vp; + struct aoedev *d = container_of(work, struct aoedev, work); if (d->flags & DEVFL_GDALLOC) aoeblk_gdalloc(d); diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c index 6125921bbec4..05a97197c918 100644 --- a/drivers/block/aoe/aoedev.c +++ b/drivers/block/aoe/aoedev.c @@ -88,7 +88,7 @@ aoedev_newdev(ulong nframes) kfree(d); return NULL; } - INIT_WORK(&d->work, aoecmd_sleepwork, d); + INIT_WORK(&d->work, aoecmd_sleepwork); spin_lock_init(&d->lock); init_timer(&d->timer); d->timer.data = (ulong) d; diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c index 40a11e567970..9d9bff23f426 100644 --- a/drivers/block/paride/pd.c +++ b/drivers/block/paride/pd.c @@ -352,19 +352,19 @@ static enum action (*phase)(void); static void run_fsm(void); -static void ps_tq_int( void *data); +static void ps_tq_int(struct work_struct *work); -static DECLARE_WORK(fsm_tq, ps_tq_int, NULL); +static DECLARE_DELAYED_WORK(fsm_tq, ps_tq_int); static void schedule_fsm(void) { if (!nice) - schedule_work(&fsm_tq); + schedule_delayed_work(&fsm_tq, 0); else schedule_delayed_work(&fsm_tq, nice-1); } -static void ps_tq_int(void *data) +static void ps_tq_int(struct work_struct *work) { run_fsm(); } diff --git a/drivers/block/paride/pseudo.h b/drivers/block/paride/pseudo.h index 932342d7a8eb..bc3703294143 100644 --- a/drivers/block/paride/pseudo.h +++ b/drivers/block/paride/pseudo.h @@ -35,7 +35,7 @@ #include #include -static void ps_tq_int( void *data); +static void ps_tq_int(struct work_struct *work); static void (* ps_continuation)(void); static int (* ps_ready)(void); @@ -45,7 +45,7 @@ static int ps_nice = 0; static DEFINE_SPINLOCK(ps_spinlock __attribute__((unused))); -static DECLARE_WORK(ps_tq, ps_tq_int, NULL); +static DECLARE_DELAYED_WORK(ps_tq, ps_tq_int); static void ps_set_intr(void (*continuation)(void), int (*ready)(void), @@ -63,14 +63,14 @@ static void ps_set_intr(void (*continuation)(void), if (!ps_tq_active) { ps_tq_active = 1; if (!ps_nice) - schedule_work(&ps_tq); + schedule_delayed_work(&ps_tq, 0); else schedule_delayed_work(&ps_tq, ps_nice-1); } spin_unlock_irqrestore(&ps_spinlock,flags); } -static void ps_tq_int(void *data) +static void ps_tq_int(struct work_struct *work) { void (*con)(void); unsigned long flags; @@ -92,7 +92,7 @@ static void ps_tq_int(void *data) } ps_tq_active = 1; if (!ps_nice) - schedule_work(&ps_tq); + schedule_delayed_work(&ps_tq, 0); else schedule_delayed_work(&ps_tq, ps_nice-1); spin_unlock_irqrestore(&ps_spinlock,flags); diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 47d6975268ff..54509eb3391b 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -1244,9 +1244,10 @@ out: return IRQ_RETVAL(handled); } -static void carm_fsm_task (void *_data) +static void carm_fsm_task (struct work_struct *work) { - struct carm_host *host = _data; + struct carm_host *host = + container_of(work, struct carm_host, fsm_task); unsigned long flags; unsigned int state; int rc, i, next_dev; @@ -1619,7 +1620,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) host->pdev = pdev; host->flags = pci_dac ? FL_DAC : 0; spin_lock_init(&host->lock); - INIT_WORK(&host->fsm_task, carm_fsm_task, host); + INIT_WORK(&host->fsm_task, carm_fsm_task); init_completion(&host->probe_comp); for (i = 0; i < ARRAY_SIZE(host->req); i++) diff --git a/drivers/block/ub.c b/drivers/block/ub.c index 0d5c73f07265..2098eff91e14 100644 --- a/drivers/block/ub.c +++ b/drivers/block/ub.c @@ -376,7 +376,7 @@ static int ub_submit_clear_stall(struct ub_dev *sc, struct ub_scsi_cmd *cmd, int stalled_pipe); static void ub_top_sense_done(struct ub_dev *sc, struct ub_scsi_cmd *scmd); static void ub_reset_enter(struct ub_dev *sc, int try); -static void ub_reset_task(void *arg); +static void ub_reset_task(struct work_struct *work); static int ub_sync_tur(struct ub_dev *sc, struct ub_lun *lun); static int ub_sync_read_cap(struct ub_dev *sc, struct ub_lun *lun, struct ub_capacity *ret); @@ -1558,9 +1558,9 @@ static void ub_reset_enter(struct ub_dev *sc, int try) schedule_work(&sc->reset_work); } -static void ub_reset_task(void *arg) +static void ub_reset_task(struct work_struct *work) { - struct ub_dev *sc = arg; + struct ub_dev *sc = container_of(work, struct ub_dev, reset_work); unsigned long flags; struct list_head *p; struct ub_lun *lun; @@ -2179,7 +2179,7 @@ static int ub_probe(struct usb_interface *intf, usb_init_urb(&sc->work_urb); tasklet_init(&sc->tasklet, ub_scsi_action, (unsigned long)sc); atomic_set(&sc->poison, 0); - INIT_WORK(&sc->reset_work, ub_reset_task, sc); + INIT_WORK(&sc->reset_work, ub_reset_task); init_waitqueue_head(&sc->reset_wait); init_timer(&sc->work_timer); diff --git a/drivers/bluetooth/bcm203x.c b/drivers/bluetooth/bcm203x.c index 516751754aa9..9256985cbe36 100644 --- a/drivers/bluetooth/bcm203x.c +++ b/drivers/bluetooth/bcm203x.c @@ -157,9 +157,10 @@ static void bcm203x_complete(struct urb *urb) } } -static void bcm203x_work(void *user_data) +static void bcm203x_work(struct work_struct *work) { - struct bcm203x_data *data = user_data; + struct bcm203x_data *data = + container_of(work, struct bcm203x_data, work); if (usb_submit_urb(data->urb, GFP_ATOMIC) < 0) BT_ERR("Can't submit URB"); @@ -246,7 +247,7 @@ static int bcm203x_probe(struct usb_interface *intf, const struct usb_device_id release_firmware(firmware); - INIT_WORK(&data->work, bcm203x_work, (void *) data); + INIT_WORK(&data->work, bcm203x_work); usb_set_intfdata(intf, data); diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c index e608dadece2f..acb2de5e3a98 100644 --- a/drivers/char/cyclades.c +++ b/drivers/char/cyclades.c @@ -926,9 +926,10 @@ cy_sched_event(struct cyclades_port *info, int event) * had to poll every port to see if that port needed servicing. */ static void -do_softint(void *private_) +do_softint(struct work_struct *work) { - struct cyclades_port *info = (struct cyclades_port *) private_; + struct cyclades_port *info = + container_of(work, struct cyclades_port, tqueue); struct tty_struct *tty; tty = info->tty; @@ -5328,7 +5329,7 @@ cy_init(void) info->blocked_open = 0; info->default_threshold = 0; info->default_timeout = 0; - INIT_WORK(&info->tqueue, do_softint, info); + INIT_WORK(&info->tqueue, do_softint); init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); init_waitqueue_head(&info->shutdown_wait); @@ -5403,7 +5404,7 @@ cy_init(void) info->blocked_open = 0; info->default_threshold = 0; info->default_timeout = 0; - INIT_WORK(&info->tqueue, do_softint, info); + INIT_WORK(&info->tqueue, do_softint); init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); init_waitqueue_head(&info->shutdown_wait); diff --git a/drivers/char/drm/via_dmablit.c b/drivers/char/drm/via_dmablit.c index 60c1695db300..806f9ce5f47b 100644 --- a/drivers/char/drm/via_dmablit.c +++ b/drivers/char/drm/via_dmablit.c @@ -500,9 +500,9 @@ via_dmablit_timer(unsigned long data) static void -via_dmablit_workqueue(void *data) +via_dmablit_workqueue(struct work_struct *work) { - drm_via_blitq_t *blitq = (drm_via_blitq_t *) data; + drm_via_blitq_t *blitq = container_of(work, drm_via_blitq_t, wq); drm_device_t *dev = blitq->dev; unsigned long irqsave; drm_via_sg_info_t *cur_sg; @@ -571,7 +571,7 @@ via_init_dmablit(drm_device_t *dev) DRM_INIT_WAITQUEUE(blitq->blit_queue + j); } DRM_INIT_WAITQUEUE(&blitq->busy_queue); - INIT_WORK(&blitq->wq, via_dmablit_workqueue, blitq); + INIT_WORK(&blitq->wq, via_dmablit_workqueue); init_timer(&blitq->poll_timer); blitq->poll_timer.function = &via_dmablit_timer; blitq->poll_timer.data = (unsigned long) blitq; diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 706733c0b36a..7c71eb779802 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -200,7 +200,7 @@ static int pc_ioctl(struct tty_struct *, struct file *, static int info_ioctl(struct tty_struct *, struct file *, unsigned int, unsigned long); static void pc_set_termios(struct tty_struct *, struct termios *); -static void do_softint(void *); +static void do_softint(struct work_struct *work); static void pc_stop(struct tty_struct *); static void pc_start(struct tty_struct *); static void pc_throttle(struct tty_struct * tty); @@ -1505,7 +1505,7 @@ static void post_fep_init(unsigned int crd) ch->brdchan = bc; ch->mailbox = gd; - INIT_WORK(&ch->tqueue, do_softint, ch); + INIT_WORK(&ch->tqueue, do_softint); ch->board = &boards[crd]; spin_lock_irqsave(&epca_lock, flags); @@ -2566,9 +2566,9 @@ static void pc_set_termios(struct tty_struct *tty, struct termios *old_termios) /* --------------------- Begin do_softint ----------------------- */ -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { /* Begin do_softint */ - struct channel *ch = (struct channel *) private_; + struct channel *ch = container_of(work, struct channel, tqueue); /* Called in response to a modem change event */ if (ch && ch->magic == EPCA_MAGIC) { /* Begin EPCA_MAGIC */ struct tty_struct *tty = ch->tty; diff --git a/drivers/char/esp.c b/drivers/char/esp.c index 15a4ea896328..93b551962513 100644 --- a/drivers/char/esp.c +++ b/drivers/char/esp.c @@ -723,9 +723,10 @@ static irqreturn_t rs_interrupt_single(int irq, void *dev_id) * ------------------------------------------------------------------- */ -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { - struct esp_struct *info = (struct esp_struct *) private_; + struct esp_struct *info = + container_of(work, struct esp_struct, tqueue); struct tty_struct *tty; tty = info->tty; @@ -746,9 +747,10 @@ static void do_softint(void *private_) * do_serial_hangup() -> tty->hangup() -> esp_hangup() * */ -static void do_serial_hangup(void *private_) +static void do_serial_hangup(struct work_struct *work) { - struct esp_struct *info = (struct esp_struct *) private_; + struct esp_struct *info = + container_of(work, struct esp_struct, tqueue_hangup); struct tty_struct *tty; tty = info->tty; @@ -2501,8 +2503,8 @@ static int __init espserial_init(void) info->magic = ESP_MAGIC; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; - INIT_WORK(&info->tqueue, do_softint, info); - INIT_WORK(&info->tqueue_hangup, do_serial_hangup, info); + INIT_WORK(&info->tqueue, do_softint); + INIT_WORK(&info->tqueue_hangup, do_serial_hangup); info->config.rx_timeout = rx_timeout; info->config.flow_on = flow_on; info->config.flow_off = flow_off; diff --git a/drivers/char/genrtc.c b/drivers/char/genrtc.c index 817dc409ac20..23b25ada65ea 100644 --- a/drivers/char/genrtc.c +++ b/drivers/char/genrtc.c @@ -102,7 +102,7 @@ static void gen_rtc_interrupt(unsigned long arg); * Routine to poll RTC seconds field for change as often as possible, * after first RTC_UIE use timer to reduce polling */ -static void genrtc_troutine(void *data) +static void genrtc_troutine(struct work_struct *work) { unsigned int tmp = get_rtc_ss(); @@ -255,7 +255,7 @@ static inline int gen_set_rtc_irq_bit(unsigned char bit) irq_active = 1; stop_rtc_timers = 0; lostint = 0; - INIT_WORK(&genrtc_task, genrtc_troutine, NULL); + INIT_WORK(&genrtc_task, genrtc_troutine); oldsecs = get_rtc_ss(); init_timer(&timer_task); diff --git a/drivers/char/hvsi.c b/drivers/char/hvsi.c index 2cf63e7305a3..82a41d5b4ed0 100644 --- a/drivers/char/hvsi.c +++ b/drivers/char/hvsi.c @@ -69,7 +69,7 @@ #define __ALIGNED__ __attribute__((__aligned__(sizeof(long)))) struct hvsi_struct { - struct work_struct writer; + struct delayed_work writer; struct work_struct handshaker; wait_queue_head_t emptyq; /* woken when outbuf is emptied */ wait_queue_head_t stateq; /* woken when HVSI state changes */ @@ -744,9 +744,10 @@ static int hvsi_handshake(struct hvsi_struct *hp) return 0; } -static void hvsi_handshaker(void *arg) +static void hvsi_handshaker(struct work_struct *work) { - struct hvsi_struct *hp = (struct hvsi_struct *)arg; + struct hvsi_struct *hp = + container_of(work, struct hvsi_struct, handshaker); if (hvsi_handshake(hp) >= 0) return; @@ -951,9 +952,10 @@ static void hvsi_push(struct hvsi_struct *hp) } /* hvsi_write_worker will keep rescheduling itself until outbuf is empty */ -static void hvsi_write_worker(void *arg) +static void hvsi_write_worker(struct work_struct *work) { - struct hvsi_struct *hp = (struct hvsi_struct *)arg; + struct hvsi_struct *hp = + container_of(work, struct hvsi_struct, writer.work); unsigned long flags; #ifdef DEBUG static long start_j = 0; @@ -1287,8 +1289,8 @@ static int __init hvsi_console_init(void) } hp = &hvsi_ports[hvsi_count]; - INIT_WORK(&hp->writer, hvsi_write_worker, hp); - INIT_WORK(&hp->handshaker, hvsi_handshaker, hp); + INIT_DELAYED_WORK(&hp->writer, hvsi_write_worker); + INIT_WORK(&hp->handshaker, hvsi_handshaker); init_waitqueue_head(&hp->emptyq); init_waitqueue_head(&hp->stateq); spin_lock_init(&hp->lock); diff --git a/drivers/char/ip2/i2lib.c b/drivers/char/ip2/i2lib.c index 54d93f0345e8..c213fdbdb2b0 100644 --- a/drivers/char/ip2/i2lib.c +++ b/drivers/char/ip2/i2lib.c @@ -84,8 +84,8 @@ static void iiSendPendingMail(i2eBordStrPtr); static void serviceOutgoingFifo(i2eBordStrPtr); // Functions defined in ip2.c as part of interrupt handling -static void do_input(void *); -static void do_status(void *); +static void do_input(struct work_struct *); +static void do_status(struct work_struct *); //*************** //* Debug Data * @@ -331,8 +331,8 @@ i2InitChannels ( i2eBordStrPtr pB, int nChannels, i2ChanStrPtr pCh) pCh->ClosingWaitTime = 30*HZ; // Initialize task queue objects - INIT_WORK(&pCh->tqueue_input, do_input, pCh); - INIT_WORK(&pCh->tqueue_status, do_status, pCh); + INIT_WORK(&pCh->tqueue_input, do_input); + INIT_WORK(&pCh->tqueue_status, do_status); #ifdef IP2DEBUG_TRACE pCh->trace = ip2trace; @@ -1573,7 +1573,7 @@ i2StripFifo(i2eBordStrPtr pB) #ifdef USE_IQ schedule_work(&pCh->tqueue_input); #else - do_input(pCh); + do_input(&pCh->tqueue_input); #endif // Note we do not need to maintain any flow-control credits at this @@ -1810,7 +1810,7 @@ i2StripFifo(i2eBordStrPtr pB) #ifdef USE_IQ schedule_work(&pCh->tqueue_status); #else - do_status(pCh); + do_status(&pCh->tqueue_status); #endif } } diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index a3f32d46d2f8..cda2459c1d60 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -189,12 +189,12 @@ static int ip2_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear); static void set_irq(int, int); -static void ip2_interrupt_bh(i2eBordStrPtr pB); +static void ip2_interrupt_bh(struct work_struct *work); static irqreturn_t ip2_interrupt(int irq, void *dev_id); static void ip2_poll(unsigned long arg); static inline void service_all_boards(void); -static void do_input(void *p); -static void do_status(void *p); +static void do_input(struct work_struct *); +static void do_status(struct work_struct *); static void ip2_wait_until_sent(PTTY,int); @@ -918,7 +918,7 @@ ip2_init_board( int boardnum ) pCh++; } ex_exit: - INIT_WORK(&pB->tqueue_interrupt, (void(*)(void*)) ip2_interrupt_bh, pB); + INIT_WORK(&pB->tqueue_interrupt, ip2_interrupt_bh); return; err_release_region: @@ -1125,8 +1125,8 @@ service_all_boards(void) /******************************************************************************/ -/* Function: ip2_interrupt_bh(pB) */ -/* Parameters: pB - pointer to the board structure */ +/* Function: ip2_interrupt_bh(work) */ +/* Parameters: work - pointer to the board structure */ /* Returns: Nothing */ /* */ /* Description: */ @@ -1135,8 +1135,9 @@ service_all_boards(void) /* */ /******************************************************************************/ static void -ip2_interrupt_bh(i2eBordStrPtr pB) +ip2_interrupt_bh(struct work_struct *work) { + i2eBordStrPtr pB = container_of(work, i2eBordStr, tqueue_interrupt); // pB better well be set or we have a problem! We can only get // here from the IMMEDIATE queue. Here, we process the boards. // Checking pB doesn't cost much and it saves us from the sanity checkers. @@ -1245,9 +1246,9 @@ ip2_poll(unsigned long arg) ip2trace (ITRC_NO_PORT, ITRC_INTR, ITRC_RETURN, 0 ); } -static void do_input(void *p) +static void do_input(struct work_struct *work) { - i2ChanStrPtr pCh = p; + i2ChanStrPtr pCh = container_of(work, i2ChanStr, tqueue_input); unsigned long flags; ip2trace(CHANN, ITRC_INPUT, 21, 0 ); @@ -1279,9 +1280,9 @@ static inline void isig(int sig, struct tty_struct *tty, int flush) } } -static void do_status(void *p) +static void do_status(struct work_struct *work) { - i2ChanStrPtr pCh = p; + i2ChanStrPtr pCh = container_of(work, i2ChanStr, tqueue_status); int status; status = i2GetStatus( pCh, (I2_BRK|I2_PAR|I2_FRA|I2_OVR) ); diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index 58c955e390b3..1637c1d9a4ba 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -530,9 +530,9 @@ sched_again: /* Interrupt handlers */ -static void isicom_bottomhalf(void *data) +static void isicom_bottomhalf(struct work_struct *work) { - struct isi_port *port = (struct isi_port *) data; + struct isi_port *port = container_of(work, struct isi_port, bh_tqueue); struct tty_struct *tty = port->tty; if (!tty) @@ -1474,9 +1474,9 @@ static void isicom_start(struct tty_struct *tty) } /* hangup et all */ -static void do_isicom_hangup(void *data) +static void do_isicom_hangup(struct work_struct *work) { - struct isi_port *port = data; + struct isi_port *port = container_of(work, struct isi_port, hangup_tq); struct tty_struct *tty; tty = port->tty; @@ -1966,8 +1966,8 @@ static int __devinit isicom_setup(void) port->channel = channel; port->close_delay = 50 * HZ/100; port->closing_wait = 3000 * HZ/100; - INIT_WORK(&port->hangup_tq, do_isicom_hangup, port); - INIT_WORK(&port->bh_tqueue, isicom_bottomhalf, port); + INIT_WORK(&port->hangup_tq, do_isicom_hangup); + INIT_WORK(&port->bh_tqueue, isicom_bottomhalf); port->status = 0; init_waitqueue_head(&port->open_wait); init_waitqueue_head(&port->close_wait); diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index 96cb1f07332b..2d025a9fd14d 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -222,7 +222,7 @@ static struct semaphore moxaBuffSem; /* * static functions: */ -static void do_moxa_softint(void *); +static void do_moxa_softint(struct work_struct *); static int moxa_open(struct tty_struct *, struct file *); static void moxa_close(struct tty_struct *, struct file *); static int moxa_write(struct tty_struct *, const unsigned char *, int); @@ -363,7 +363,7 @@ static int __init moxa_init(void) for (i = 0, ch = moxaChannels; i < MAX_PORTS; i++, ch++) { ch->type = PORT_16550A; ch->port = i; - INIT_WORK(&ch->tqueue, do_moxa_softint, ch); + INIT_WORK(&ch->tqueue, do_moxa_softint); ch->tty = NULL; ch->close_delay = 5 * HZ / 10; ch->closing_wait = 30 * HZ; @@ -509,9 +509,9 @@ static void __exit moxa_exit(void) module_init(moxa_init); module_exit(moxa_exit); -static void do_moxa_softint(void *private_) +static void do_moxa_softint(struct work_struct *work) { - struct moxa_str *ch = (struct moxa_str *) private_; + struct moxa_str *ch = container_of(work, struct moxa_str, tqueue); struct tty_struct *tty; if (ch && (tty = ch->tty)) { diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c index 048d91142c17..5ed2486b7581 100644 --- a/drivers/char/mxser.c +++ b/drivers/char/mxser.c @@ -389,7 +389,7 @@ static int mxser_init(void); /* static void mxser_poll(unsigned long); */ static int mxser_get_ISA_conf(int, struct mxser_hwconf *); static int mxser_get_PCI_conf(int, int, int, struct mxser_hwconf *); -static void mxser_do_softint(void *); +static void mxser_do_softint(struct work_struct *); static int mxser_open(struct tty_struct *, struct file *); static void mxser_close(struct tty_struct *, struct file *); static int mxser_write(struct tty_struct *, const unsigned char *, int); @@ -590,7 +590,7 @@ static int mxser_initbrd(int board, struct mxser_hwconf *hwconf) info->custom_divisor = hwconf->baud_base[i] * 16; info->close_delay = 5 * HZ / 10; info->closing_wait = 30 * HZ; - INIT_WORK(&info->tqueue, mxser_do_softint, info); + INIT_WORK(&info->tqueue, mxser_do_softint); info->normal_termios = mxvar_sdriver->init_termios; init_waitqueue_head(&info->open_wait); init_waitqueue_head(&info->close_wait); @@ -917,9 +917,10 @@ static int mxser_init(void) return 0; } -static void mxser_do_softint(void *private_) +static void mxser_do_softint(struct work_struct *work) { - struct mxser_struct *info = private_; + struct mxser_struct *info = + container_of(work, struct mxser_struct, tqueue); struct tty_struct *tty; tty = info->tty; diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c index 1a0bc30b79d1..e4d950072b5a 100644 --- a/drivers/char/pcmcia/synclink_cs.c +++ b/drivers/char/pcmcia/synclink_cs.c @@ -421,7 +421,7 @@ static irqreturn_t mgslpc_isr(int irq, void *dev_id); /* * Bottom half interrupt handlers */ -static void bh_handler(void* Context); +static void bh_handler(struct work_struct *work); static void bh_transmit(MGSLPC_INFO *info); static void bh_status(MGSLPC_INFO *info); @@ -547,7 +547,7 @@ static int mgslpc_probe(struct pcmcia_device *link) memset(info, 0, sizeof(MGSLPC_INFO)); info->magic = MGSLPC_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; @@ -842,9 +842,9 @@ static int bh_action(MGSLPC_INFO *info) return rc; } -static void bh_handler(void* Context) +static void bh_handler(struct work_struct *work) { - MGSLPC_INFO *info = (MGSLPC_INFO*)Context; + MGSLPC_INFO *info = container_of(work, MGSLPC_INFO, task); int action; if (!info) diff --git a/drivers/char/sonypi.c b/drivers/char/sonypi.c index c084149153de..fc87070f1866 100644 --- a/drivers/char/sonypi.c +++ b/drivers/char/sonypi.c @@ -765,7 +765,7 @@ static void sonypi_setbluetoothpower(u8 state) sonypi_device.bluetooth_power = state; } -static void input_keyrelease(void *data) +static void input_keyrelease(struct work_struct *work) { struct sonypi_keypress kp; @@ -1412,7 +1412,7 @@ static int __devinit sonypi_probe(struct platform_device *dev) goto err_inpdev_unregister; } - INIT_WORK(&sonypi_device.input_work, input_keyrelease, NULL); + INIT_WORK(&sonypi_device.input_work, input_keyrelease); } sonypi_enable(0); diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index 7e1bd9562c2a..99137ab66b62 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -2261,9 +2261,10 @@ static void sx_start(struct tty_struct * tty) * do_sx_hangup() -> tty->hangup() -> sx_hangup() * */ -static void do_sx_hangup(void *private_) +static void do_sx_hangup(struct work_struct *work) { - struct specialix_port *port = (struct specialix_port *) private_; + struct specialix_port *port = + container_of(work, struct specialix_port, tqueue_hangup); struct tty_struct *tty; func_enter(); @@ -2336,9 +2337,10 @@ static void sx_set_termios(struct tty_struct * tty, struct termios * old_termios } -static void do_softint(void *private_) +static void do_softint(struct work_struct *work) { - struct specialix_port *port = (struct specialix_port *) private_; + struct specialix_port *port = + container_of(work, struct specialix_port, tqueue); struct tty_struct *tty; func_enter(); @@ -2411,8 +2413,8 @@ static int sx_init_drivers(void) memset(sx_port, 0, sizeof(sx_port)); for (i = 0; i < SX_NPORT * SX_NBOARD; i++) { sx_port[i].magic = SPECIALIX_MAGIC; - INIT_WORK(&sx_port[i].tqueue, do_softint, &sx_port[i]); - INIT_WORK(&sx_port[i].tqueue_hangup, do_sx_hangup, &sx_port[i]); + INIT_WORK(&sx_port[i].tqueue, do_softint); + INIT_WORK(&sx_port[i].tqueue_hangup, do_sx_hangup); sx_port[i].close_delay = 50 * HZ/100; sx_port[i].closing_wait = 3000 * HZ/100; init_waitqueue_head(&sx_port[i].open_wait); diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index 06784adcc35c..147c30da81ea 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -802,7 +802,7 @@ static int save_tx_buffer_request(struct mgsl_struct *info,const char *Buffer, u /* * Bottom half interrupt handlers */ -static void mgsl_bh_handler(void* Context); +static void mgsl_bh_handler(struct work_struct *work); static void mgsl_bh_receive(struct mgsl_struct *info); static void mgsl_bh_transmit(struct mgsl_struct *info); static void mgsl_bh_status(struct mgsl_struct *info); @@ -1071,9 +1071,10 @@ static int mgsl_bh_action(struct mgsl_struct *info) /* * Perform bottom half processing of work items queued by ISR. */ -static void mgsl_bh_handler(void* Context) +static void mgsl_bh_handler(struct work_struct *work) { - struct mgsl_struct *info = (struct mgsl_struct*)Context; + struct mgsl_struct *info = + container_of(work, struct mgsl_struct, task); int action; if (!info) @@ -4337,7 +4338,7 @@ static struct mgsl_struct* mgsl_allocate_device(void) } else { memset(info, 0, sizeof(struct mgsl_struct)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, mgsl_bh_handler, info); + INIT_WORK(&info->task, mgsl_bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index d4334c79f8d4..07f34d43dc7f 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -485,7 +485,7 @@ static void enable_loopback(struct slgt_info *info); static void set_rate(struct slgt_info *info, u32 data_rate); static int bh_action(struct slgt_info *info); -static void bh_handler(void* context); +static void bh_handler(struct work_struct *work); static void bh_transmit(struct slgt_info *info); static void isr_serial(struct slgt_info *info); static void isr_rdma(struct slgt_info *info); @@ -1878,9 +1878,9 @@ static int bh_action(struct slgt_info *info) /* * perform bottom half processing */ -static void bh_handler(void* context) +static void bh_handler(struct work_struct *work) { - struct slgt_info *info = context; + struct slgt_info *info = container_of(work, struct slgt_info, task); int action; if (!info) @@ -3326,7 +3326,7 @@ static struct slgt_info *alloc_dev(int adapter_num, int port_num, struct pci_dev } else { memset(info, 0, sizeof(struct slgt_info)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->raw_rx_size = DMABUFSIZE; info->close_delay = 5*HZ/10; @@ -4799,6 +4799,6 @@ static void rx_timeout(unsigned long context) spin_lock_irqsave(&info->lock, flags); info->pending_bh |= BH_RECEIVE; spin_unlock_irqrestore(&info->lock, flags); - bh_handler(info); + bh_handler(&info->task); } diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index 3e932b681371..13a57245cf2e 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -602,7 +602,7 @@ static void enable_loopback(SLMP_INFO *info, int enable); static void set_rate(SLMP_INFO *info, u32 data_rate); static int bh_action(SLMP_INFO *info); -static void bh_handler(void* Context); +static void bh_handler(struct work_struct *work); static void bh_receive(SLMP_INFO *info); static void bh_transmit(SLMP_INFO *info); static void bh_status(SLMP_INFO *info); @@ -2063,9 +2063,9 @@ int bh_action(SLMP_INFO *info) /* Perform bottom half processing of work items queued by ISR. */ -void bh_handler(void* Context) +void bh_handler(struct work_struct *work) { - SLMP_INFO *info = (SLMP_INFO*)Context; + SLMP_INFO *info = container_of(work, SLMP_INFO, task); int action; if (!info) @@ -3805,7 +3805,7 @@ static SLMP_INFO *alloc_dev(int adapter_num, int port_num, struct pci_dev *pdev) } else { memset(info, 0, sizeof(SLMP_INFO)); info->magic = MGSL_MAGIC; - INIT_WORK(&info->task, bh_handler, info); + INIT_WORK(&info->task, bh_handler); info->max_frame_size = 4096; info->close_delay = 5*HZ/10; info->closing_wait = 30*HZ; diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 6ad2d3bb945c..36f91a655275 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -325,9 +325,9 @@ static void user_reader_timeout(unsigned long ptr) schedule_work(&chip->work); } -static void timeout_work(void *ptr) +static void timeout_work(struct work_struct *work) { - struct tpm_chip *chip = ptr; + struct tpm_chip *chip = container_of(work, struct tpm_chip, work); down(&chip->buffer_mutex); atomic_set(&chip->data_pending, 0); @@ -1105,7 +1105,7 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend init_MUTEX(&chip->tpm_mutex); INIT_LIST_HEAD(&chip->list); - INIT_WORK(&chip->work, timeout_work, chip); + INIT_WORK(&chip->work, timeout_work); init_timer(&chip->user_read_timer); chip->user_read_timer.function = user_reader_timeout; diff --git a/drivers/connector/cn_queue.c b/drivers/connector/cn_queue.c index 05f8ce2cfb4a..b418b16e910e 100644 --- a/drivers/connector/cn_queue.c +++ b/drivers/connector/cn_queue.c @@ -31,9 +31,11 @@ #include #include -void cn_queue_wrapper(void *data) +void cn_queue_wrapper(struct work_struct *work) { - struct cn_callback_data *d = data; + struct cn_callback_entry *cbq = + container_of(work, struct cn_callback_entry, work.work); + struct cn_callback_data *d = &cbq->data; d->callback(d->callback_priv); @@ -57,7 +59,7 @@ static struct cn_callback_entry *cn_queue_alloc_callback_entry(char *name, struc memcpy(&cbq->id.id, id, sizeof(struct cb_id)); cbq->data.callback = callback; - INIT_WORK(&cbq->work, &cn_queue_wrapper, &cbq->data); + INIT_DELAYED_WORK(&cbq->work, &cn_queue_wrapper); return cbq; } diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c index b49bacfd8de8..5e7cd45d10ee 100644 --- a/drivers/connector/connector.c +++ b/drivers/connector/connector.c @@ -135,40 +135,39 @@ static int cn_call_callback(struct cn_msg *msg, void (*destruct_data)(void *), v spin_lock_bh(&dev->cbdev->queue_lock); list_for_each_entry(__cbq, &dev->cbdev->queue_list, callback_entry) { if (cn_cb_equal(&__cbq->id.id, &msg->id)) { - if (likely(!test_bit(0, &__cbq->work.pending) && + if (likely(!test_bit(WORK_STRUCT_PENDING, + &__cbq->work.work.management) && __cbq->data.ddata == NULL)) { __cbq->data.callback_priv = msg; __cbq->data.ddata = data; __cbq->data.destruct_data = destruct_data; - if (queue_work(dev->cbdev->cn_queue, - &__cbq->work)) + if (queue_delayed_work( + dev->cbdev->cn_queue, + &__cbq->work, 0)) err = 0; } else { - struct work_struct *w; struct cn_callback_data *d; - w = kzalloc(sizeof(*w) + sizeof(*d), GFP_ATOMIC); - if (w) { - d = (struct cn_callback_data *)(w+1); - + __cbq = kzalloc(sizeof(*__cbq), GFP_ATOMIC); + if (__cbq) { + d = &__cbq->data; d->callback_priv = msg; d->callback = __cbq->data.callback; d->ddata = data; d->destruct_data = destruct_data; - d->free = w; + d->free = __cbq; - INIT_LIST_HEAD(&w->entry); - w->pending = 0; - w->func = &cn_queue_wrapper; - w->data = d; - init_timer(&w->timer); + INIT_DELAYED_WORK(&__cbq->work, + &cn_queue_wrapper); - if (queue_work(dev->cbdev->cn_queue, w)) + if (queue_delayed_work( + dev->cbdev->cn_queue, + &__cbq->work, 0)) err = 0; else { - kfree(w); + kfree(__cbq); err = -EINVAL; } } else diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index c4c578defabf..5ef5ede5b884 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -59,7 +59,7 @@ static unsigned int def_sampling_rate; #define MAX_SAMPLING_DOWN_FACTOR (10) #define TRANSITION_LATENCY_LIMIT (10 * 1000) -static void do_dbs_timer(void *data); +static void do_dbs_timer(struct work_struct *work); struct cpu_dbs_info_s { struct cpufreq_policy *cur_policy; @@ -82,7 +82,7 @@ static unsigned int dbs_enable; /* number of CPUs using this policy */ * is recursive for the same process. -Venki */ static DEFINE_MUTEX (dbs_mutex); -static DECLARE_WORK (dbs_work, do_dbs_timer, NULL); +static DECLARE_DELAYED_WORK(dbs_work, do_dbs_timer); struct dbs_tuners { unsigned int sampling_rate; @@ -420,7 +420,7 @@ static void dbs_check_cpu(int cpu) } } -static void do_dbs_timer(void *data) +static void do_dbs_timer(struct work_struct *work) { int i; lock_cpu_hotplug(); @@ -435,7 +435,6 @@ static void do_dbs_timer(void *data) static inline void dbs_timer_init(void) { - INIT_WORK(&dbs_work, do_dbs_timer, NULL); schedule_delayed_work(&dbs_work, usecs_to_jiffies(dbs_tuners_ins.sampling_rate)); return; diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index bf8aa45d4f01..e1cc5113c2ae 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -47,13 +47,17 @@ static unsigned int def_sampling_rate; #define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000) #define TRANSITION_LATENCY_LIMIT (10 * 1000) -static void do_dbs_timer(void *data); +static void do_dbs_timer(struct work_struct *work); + +/* Sampling types */ +enum dbs_sample {DBS_NORMAL_SAMPLE, DBS_SUB_SAMPLE}; struct cpu_dbs_info_s { cputime64_t prev_cpu_idle; cputime64_t prev_cpu_wall; struct cpufreq_policy *cur_policy; - struct work_struct work; + struct delayed_work work; + enum dbs_sample sample_type; unsigned int enable; struct cpufreq_frequency_table *freq_table; unsigned int freq_lo; @@ -407,30 +411,31 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info) } } -/* Sampling types */ -enum {DBS_NORMAL_SAMPLE, DBS_SUB_SAMPLE}; - -static void do_dbs_timer(void *data) +static void do_dbs_timer(struct work_struct *work) { unsigned int cpu = smp_processor_id(); struct cpu_dbs_info_s *dbs_info = &per_cpu(cpu_dbs_info, cpu); + enum dbs_sample sample_type = dbs_info->sample_type; /* We want all CPUs to do sampling nearly on same jiffy */ int delay = usecs_to_jiffies(dbs_tuners_ins.sampling_rate); + + /* Permit rescheduling of this work item */ + work_release(work); + delay -= jiffies % delay; if (!dbs_info->enable) return; /* Common NORMAL_SAMPLE setup */ - INIT_WORK(&dbs_info->work, do_dbs_timer, (void *)DBS_NORMAL_SAMPLE); + dbs_info->sample_type = DBS_NORMAL_SAMPLE; if (!dbs_tuners_ins.powersave_bias || - (unsigned long) data == DBS_NORMAL_SAMPLE) { + sample_type == DBS_NORMAL_SAMPLE) { lock_cpu_hotplug(); dbs_check_cpu(dbs_info); unlock_cpu_hotplug(); if (dbs_info->freq_lo) { /* Setup timer for SUB_SAMPLE */ - INIT_WORK(&dbs_info->work, do_dbs_timer, - (void *)DBS_SUB_SAMPLE); + dbs_info->sample_type = DBS_SUB_SAMPLE; delay = dbs_info->freq_hi_jiffies; } } else { @@ -449,7 +454,8 @@ static inline void dbs_timer_init(unsigned int cpu) delay -= jiffies % delay; ondemand_powersave_bias_init(); - INIT_WORK(&dbs_info->work, do_dbs_timer, NULL); + INIT_DELAYED_WORK_NAR(&dbs_info->work, do_dbs_timer); + dbs_info->sample_type = DBS_NORMAL_SAMPLE; queue_delayed_work_on(cpu, kondemand_wq, &dbs_info->work, delay); } diff --git a/drivers/i2c/chips/ds1374.c b/drivers/i2c/chips/ds1374.c index 4630f1969a09..15edf40828b4 100644 --- a/drivers/i2c/chips/ds1374.c +++ b/drivers/i2c/chips/ds1374.c @@ -140,12 +140,14 @@ ulong ds1374_get_rtc_time(void) return t1; } -static void ds1374_set_work(void *arg) +static ulong new_time; + +static void ds1374_set_work(struct work_struct *work) { ulong t1, t2; int limit = 10; /* arbitrary retry limit */ - t1 = *(ulong *) arg; + t1 = new_time; mutex_lock(&ds1374_mutex); @@ -167,11 +169,9 @@ static void ds1374_set_work(void *arg) "can't confirm time set from rtc chip\n"); } -static ulong new_time; - static struct workqueue_struct *ds1374_workqueue; -static DECLARE_WORK(ds1374_work, ds1374_set_work, &new_time); +static DECLARE_WORK(ds1374_work, ds1374_set_work); int ds1374_set_rtc_time(ulong nowtime) { @@ -180,7 +180,7 @@ int ds1374_set_rtc_time(ulong nowtime) if (in_interrupt()) queue_work(ds1374_workqueue, &ds1374_work); else - ds1374_set_work(&new_time); + ds1374_set_work(NULL); return 0; } diff --git a/drivers/ieee1394/hosts.c b/drivers/ieee1394/hosts.c index d90a3a1898c0..8f4378a1631c 100644 --- a/drivers/ieee1394/hosts.c +++ b/drivers/ieee1394/hosts.c @@ -31,9 +31,10 @@ #include "config_roms.h" -static void delayed_reset_bus(void * __reset_info) +static void delayed_reset_bus(struct work_struct *work) { - struct hpsb_host *host = (struct hpsb_host*)__reset_info; + struct hpsb_host *host = + container_of(work, struct hpsb_host, delayed_reset.work); int generation = host->csr.generation + 1; /* The generation field rolls over to 2 rather than 0 per IEEE @@ -145,7 +146,7 @@ struct hpsb_host *hpsb_alloc_host(struct hpsb_host_driver *drv, size_t extra, atomic_set(&h->generation, 0); - INIT_WORK(&h->delayed_reset, delayed_reset_bus, h); + INIT_DELAYED_WORK(&h->delayed_reset, delayed_reset_bus); init_timer(&h->timeout); h->timeout.data = (unsigned long) h; @@ -234,7 +235,7 @@ int hpsb_update_config_rom_image(struct hpsb_host *host) * Config ROM in the near future. */ reset_delay = HZ; - PREPARE_WORK(&host->delayed_reset, delayed_reset_bus, host); + PREPARE_DELAYED_WORK(&host->delayed_reset, delayed_reset_bus); schedule_delayed_work(&host->delayed_reset, reset_delay); return 0; diff --git a/drivers/ieee1394/hosts.h b/drivers/ieee1394/hosts.h index bc6dbfadb891..d553e38c9543 100644 --- a/drivers/ieee1394/hosts.h +++ b/drivers/ieee1394/hosts.h @@ -62,7 +62,7 @@ struct hpsb_host { struct class_device class_dev; int update_config_rom; - struct work_struct delayed_reset; + struct delayed_work delayed_reset; unsigned int config_roms; struct list_head addr_space; diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c index 6986ac188281..cd156d4e779e 100644 --- a/drivers/ieee1394/sbp2.c +++ b/drivers/ieee1394/sbp2.c @@ -493,20 +493,25 @@ static void sbp2util_notify_fetch_agent(struct scsi_id_instance_data *scsi_id, scsi_unblock_requests(scsi_id->scsi_host); } -static void sbp2util_write_orb_pointer(void *p) +static void sbp2util_write_orb_pointer(struct work_struct *work) { + struct scsi_id_instance_data *scsi_id = + container_of(work, struct scsi_id_instance_data, + protocol_work.work); quadlet_t data[2]; - data[0] = ORB_SET_NODE_ID( - ((struct scsi_id_instance_data *)p)->hi->host->node_id); - data[1] = ((struct scsi_id_instance_data *)p)->last_orb_dma; + data[0] = ORB_SET_NODE_ID(scsi_id->hi->host->node_id); + data[1] = scsi_id->last_orb_dma; sbp2util_cpu_to_be32_buffer(data, 8); - sbp2util_notify_fetch_agent(p, SBP2_ORB_POINTER_OFFSET, data, 8); + sbp2util_notify_fetch_agent(scsi_id, SBP2_ORB_POINTER_OFFSET, data, 8); } -static void sbp2util_write_doorbell(void *p) +static void sbp2util_write_doorbell(struct work_struct *work) { - sbp2util_notify_fetch_agent(p, SBP2_DOORBELL_OFFSET, NULL, 4); + struct scsi_id_instance_data *scsi_id = + container_of(work, struct scsi_id_instance_data, + protocol_work.work); + sbp2util_notify_fetch_agent(scsi_id, SBP2_DOORBELL_OFFSET, NULL, 4); } /* @@ -843,7 +848,7 @@ static struct scsi_id_instance_data *sbp2_alloc_device(struct unit_directory *ud INIT_LIST_HEAD(&scsi_id->scsi_list); spin_lock_init(&scsi_id->sbp2_command_orb_lock); atomic_set(&scsi_id->state, SBP2LU_STATE_RUNNING); - INIT_WORK(&scsi_id->protocol_work, NULL, NULL); + INIT_DELAYED_WORK(&scsi_id->protocol_work, NULL); ud->device.driver_data = scsi_id; @@ -2047,11 +2052,10 @@ static void sbp2_link_orb_command(struct scsi_id_instance_data *scsi_id, * We do not accept new commands until the job is over. */ scsi_block_requests(scsi_id->scsi_host); - PREPARE_WORK(&scsi_id->protocol_work, + PREPARE_DELAYED_WORK(&scsi_id->protocol_work, last_orb ? sbp2util_write_doorbell: - sbp2util_write_orb_pointer, - scsi_id); - schedule_work(&scsi_id->protocol_work); + sbp2util_write_orb_pointer); + schedule_delayed_work(&scsi_id->protocol_work, 0); } } diff --git a/drivers/ieee1394/sbp2.h b/drivers/ieee1394/sbp2.h index abbe48e646c3..1b16d6b9cf11 100644 --- a/drivers/ieee1394/sbp2.h +++ b/drivers/ieee1394/sbp2.h @@ -348,7 +348,7 @@ struct scsi_id_instance_data { unsigned workarounds; atomic_t state; - struct work_struct protocol_work; + struct delayed_work protocol_work; }; /* For use in scsi_id_instance_data.state */ diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index e11187ecc931..84b2f5cb3722 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -55,11 +55,11 @@ struct addr_req { int status; }; -static void process_req(void *data); +static void process_req(struct work_struct *work); static DEFINE_MUTEX(lock); static LIST_HEAD(req_list); -static DECLARE_WORK(work, process_req, NULL); +static DECLARE_DELAYED_WORK(work, process_req); static struct workqueue_struct *addr_wq; void rdma_addr_register_client(struct rdma_addr_client *client) @@ -215,7 +215,7 @@ out: return ret; } -static void process_req(void *data) +static void process_req(struct work_struct *work) { struct addr_req *req, *temp_req; struct sockaddr_in *src_in, *dst_in; diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index 20e9f64e67a6..98272fbbfb31 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -285,9 +285,10 @@ err: kfree(tprops); } -static void ib_cache_task(void *work_ptr) +static void ib_cache_task(struct work_struct *_work) { - struct ib_update_work *work = work_ptr; + struct ib_update_work *work = + container_of(_work, struct ib_update_work, work); ib_cache_update(work->device, work->port_num); kfree(work); @@ -306,7 +307,7 @@ static void ib_cache_event(struct ib_event_handler *handler, event->event == IB_EVENT_CLIENT_REREGISTER) { work = kmalloc(sizeof *work, GFP_ATOMIC); if (work) { - INIT_WORK(&work->work, ib_cache_task, work); + INIT_WORK(&work->work, ib_cache_task); work->device = event->device; work->port_num = event->element.port_num; schedule_work(&work->work); diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 25b1018a476c..e1990f531d0a 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -101,7 +101,7 @@ struct cm_av { }; struct cm_work { - struct work_struct work; + struct delayed_work work; struct list_head list; struct cm_port *port; struct ib_mad_recv_wc *mad_recv_wc; /* Received MADs */ @@ -161,7 +161,7 @@ struct cm_id_private { atomic_t work_count; }; -static void cm_work_handler(void *data); +static void cm_work_handler(struct work_struct *work); static inline void cm_deref_id(struct cm_id_private *cm_id_priv) { @@ -669,8 +669,7 @@ static struct cm_timewait_info * cm_create_timewait_info(__be32 local_id) return ERR_PTR(-ENOMEM); timewait_info->work.local_id = local_id; - INIT_WORK(&timewait_info->work.work, cm_work_handler, - &timewait_info->work); + INIT_DELAYED_WORK(&timewait_info->work.work, cm_work_handler); timewait_info->work.cm_event.event = IB_CM_TIMEWAIT_EXIT; return timewait_info; } @@ -2987,9 +2986,9 @@ static void cm_send_handler(struct ib_mad_agent *mad_agent, } } -static void cm_work_handler(void *data) +static void cm_work_handler(struct work_struct *_work) { - struct cm_work *work = data; + struct cm_work *work = container_of(_work, struct cm_work, work.work); int ret; switch (work->cm_event.event) { @@ -3079,12 +3078,12 @@ int ib_cm_establish(struct ib_cm_id *cm_id) * we need to find the cm_id once we're in the context of the * worker thread, rather than holding a reference on it. */ - INIT_WORK(&work->work, cm_work_handler, work); + INIT_DELAYED_WORK(&work->work, cm_work_handler); work->local_id = cm_id->local_id; work->remote_id = cm_id->remote_id; work->mad_recv_wc = NULL; work->cm_event.event = IB_CM_USER_ESTABLISHED; - queue_work(cm.wq, &work->work); + queue_delayed_work(cm.wq, &work->work, 0); out: return ret; } @@ -3146,11 +3145,11 @@ static void cm_recv_handler(struct ib_mad_agent *mad_agent, return; } - INIT_WORK(&work->work, cm_work_handler, work); + INIT_DELAYED_WORK(&work->work, cm_work_handler); work->cm_event.event = event; work->mad_recv_wc = mad_recv_wc; work->port = (struct cm_port *)mad_agent->context; - queue_work(cm.wq, &work->work); + queue_delayed_work(cm.wq, &work->work, 0); } static int cm_init_qp_init_attr(struct cm_id_private *cm_id_priv, diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 845090b0859c..189f73f3f721 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -1341,9 +1341,9 @@ static int cma_query_ib_route(struct rdma_id_private *id_priv, int timeout_ms, return (id_priv->query_id < 0) ? id_priv->query_id : 0; } -static void cma_work_handler(void *data) +static void cma_work_handler(struct work_struct *_work) { - struct cma_work *work = data; + struct cma_work *work = container_of(_work, struct cma_work, work); struct rdma_id_private *id_priv = work->id; int destroy = 0; @@ -1374,7 +1374,7 @@ static int cma_resolve_ib_route(struct rdma_id_private *id_priv, int timeout_ms) return -ENOMEM; work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ROUTE_QUERY; work->new_state = CMA_ROUTE_RESOLVED; work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED; @@ -1431,7 +1431,7 @@ static int cma_resolve_iw_route(struct rdma_id_private *id_priv, int timeout_ms) return -ENOMEM; work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ROUTE_QUERY; work->new_state = CMA_ROUTE_RESOLVED; work->event.event = RDMA_CM_EVENT_ROUTE_RESOLVED; @@ -1585,7 +1585,7 @@ static int cma_resolve_loopback(struct rdma_id_private *id_priv) } work->id = id_priv; - INIT_WORK(&work->work, cma_work_handler, work); + INIT_WORK(&work->work, cma_work_handler); work->old_state = CMA_ADDR_QUERY; work->new_state = CMA_ADDR_RESOLVED; work->event.event = RDMA_CM_EVENT_ADDR_RESOLVED; diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index c3fb304a4e86..9bfa785252dc 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -828,9 +828,10 @@ static int process_event(struct iwcm_id_private *cm_id_priv, * thread asleep on the destroy_comp list vs. an object destroyed * here synchronously when the last reference is removed. */ -static void cm_work_handler(void *arg) +static void cm_work_handler(struct work_struct *_work) { - struct iwcm_work *work = arg, lwork; + struct iwcm_work lwork, *work = + container_of(_work, struct iwcm_work, work); struct iwcm_id_private *cm_id_priv = work->cm_id; unsigned long flags; int empty; @@ -899,7 +900,7 @@ static int cm_event_handler(struct iw_cm_id *cm_id, goto out; } - INIT_WORK(&work->work, cm_work_handler, work); + INIT_WORK(&work->work, cm_work_handler); work->cm_id = cm_id_priv; work->event = *iw_event; diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index a72bcea46ff6..5a54ac35e961 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -65,8 +65,8 @@ static struct ib_mad_agent_private *find_mad_agent( static int ib_mad_post_receive_mads(struct ib_mad_qp_info *qp_info, struct ib_mad_private *mad); static void cancel_mads(struct ib_mad_agent_private *mad_agent_priv); -static void timeout_sends(void *data); -static void local_completions(void *data); +static void timeout_sends(struct work_struct *work); +static void local_completions(struct work_struct *work); static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, struct ib_mad_agent_private *agent_priv, u8 mgmt_class); @@ -356,10 +356,9 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, INIT_LIST_HEAD(&mad_agent_priv->wait_list); INIT_LIST_HEAD(&mad_agent_priv->done_list); INIT_LIST_HEAD(&mad_agent_priv->rmpp_list); - INIT_WORK(&mad_agent_priv->timed_work, timeout_sends, mad_agent_priv); + INIT_DELAYED_WORK(&mad_agent_priv->timed_work, timeout_sends); INIT_LIST_HEAD(&mad_agent_priv->local_list); - INIT_WORK(&mad_agent_priv->local_work, local_completions, - mad_agent_priv); + INIT_WORK(&mad_agent_priv->local_work, local_completions); atomic_set(&mad_agent_priv->refcount, 1); init_completion(&mad_agent_priv->comp); @@ -2198,12 +2197,12 @@ static void mad_error_handler(struct ib_mad_port_private *port_priv, /* * IB MAD completion callback */ -static void ib_mad_completion_handler(void *data) +static void ib_mad_completion_handler(struct work_struct *work) { struct ib_mad_port_private *port_priv; struct ib_wc wc; - port_priv = (struct ib_mad_port_private *)data; + port_priv = container_of(work, struct ib_mad_port_private, work); ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP); while (ib_poll_cq(port_priv->cq, 1, &wc) == 1) { @@ -2324,7 +2323,7 @@ void ib_cancel_mad(struct ib_mad_agent *mad_agent, } EXPORT_SYMBOL(ib_cancel_mad); -static void local_completions(void *data) +static void local_completions(struct work_struct *work) { struct ib_mad_agent_private *mad_agent_priv; struct ib_mad_local_private *local; @@ -2334,7 +2333,8 @@ static void local_completions(void *data) struct ib_wc wc; struct ib_mad_send_wc mad_send_wc; - mad_agent_priv = (struct ib_mad_agent_private *)data; + mad_agent_priv = + container_of(work, struct ib_mad_agent_private, local_work); spin_lock_irqsave(&mad_agent_priv->lock, flags); while (!list_empty(&mad_agent_priv->local_list)) { @@ -2434,14 +2434,15 @@ static int retry_send(struct ib_mad_send_wr_private *mad_send_wr) return ret; } -static void timeout_sends(void *data) +static void timeout_sends(struct work_struct *work) { struct ib_mad_agent_private *mad_agent_priv; struct ib_mad_send_wr_private *mad_send_wr; struct ib_mad_send_wc mad_send_wc; unsigned long flags, delay; - mad_agent_priv = (struct ib_mad_agent_private *)data; + mad_agent_priv = container_of(work, struct ib_mad_agent_private, + timed_work.work); mad_send_wc.vendor_err = 0; spin_lock_irqsave(&mad_agent_priv->lock, flags); @@ -2799,7 +2800,7 @@ static int ib_mad_port_open(struct ib_device *device, ret = -ENOMEM; goto error8; } - INIT_WORK(&port_priv->work, ib_mad_completion_handler, port_priv); + INIT_WORK(&port_priv->work, ib_mad_completion_handler); spin_lock_irqsave(&ib_mad_port_list_lock, flags); list_add_tail(&port_priv->port_list, &ib_mad_port_list); diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h index d06b59083f6e..d5548e73e068 100644 --- a/drivers/infiniband/core/mad_priv.h +++ b/drivers/infiniband/core/mad_priv.h @@ -102,7 +102,7 @@ struct ib_mad_agent_private { struct list_head send_list; struct list_head wait_list; struct list_head done_list; - struct work_struct timed_work; + struct delayed_work timed_work; unsigned long timeout; struct list_head local_list; struct work_struct local_work; diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index 1ef79d015a1e..3663fd7022be 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -45,8 +45,8 @@ enum rmpp_state { struct mad_rmpp_recv { struct ib_mad_agent_private *agent; struct list_head list; - struct work_struct timeout_work; - struct work_struct cleanup_work; + struct delayed_work timeout_work; + struct delayed_work cleanup_work; struct completion comp; enum rmpp_state state; spinlock_t lock; @@ -233,9 +233,10 @@ static void nack_recv(struct ib_mad_agent_private *agent, } } -static void recv_timeout_handler(void *data) +static void recv_timeout_handler(struct work_struct *work) { - struct mad_rmpp_recv *rmpp_recv = data; + struct mad_rmpp_recv *rmpp_recv = + container_of(work, struct mad_rmpp_recv, timeout_work.work); struct ib_mad_recv_wc *rmpp_wc; unsigned long flags; @@ -254,9 +255,10 @@ static void recv_timeout_handler(void *data) ib_free_recv_mad(rmpp_wc); } -static void recv_cleanup_handler(void *data) +static void recv_cleanup_handler(struct work_struct *work) { - struct mad_rmpp_recv *rmpp_recv = data; + struct mad_rmpp_recv *rmpp_recv = + container_of(work, struct mad_rmpp_recv, cleanup_work.work); unsigned long flags; spin_lock_irqsave(&rmpp_recv->agent->lock, flags); @@ -285,8 +287,8 @@ create_rmpp_recv(struct ib_mad_agent_private *agent, rmpp_recv->agent = agent; init_completion(&rmpp_recv->comp); - INIT_WORK(&rmpp_recv->timeout_work, recv_timeout_handler, rmpp_recv); - INIT_WORK(&rmpp_recv->cleanup_work, recv_cleanup_handler, rmpp_recv); + INIT_DELAYED_WORK(&rmpp_recv->timeout_work, recv_timeout_handler); + INIT_DELAYED_WORK(&rmpp_recv->cleanup_work, recv_cleanup_handler); spin_lock_init(&rmpp_recv->lock); rmpp_recv->state = RMPP_STATE_ACTIVE; atomic_set(&rmpp_recv->refcount, 1); diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index 1706d3c7e95e..e45afba75341 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -360,9 +360,10 @@ static void free_sm_ah(struct kref *kref) kfree(sm_ah); } -static void update_sm_ah(void *port_ptr) +static void update_sm_ah(struct work_struct *work) { - struct ib_sa_port *port = port_ptr; + struct ib_sa_port *port = + container_of(work, struct ib_sa_port, update_task); struct ib_sa_sm_ah *new_ah, *old_ah; struct ib_port_attr port_attr; struct ib_ah_attr ah_attr; @@ -992,8 +993,7 @@ static void ib_sa_add_one(struct ib_device *device) if (IS_ERR(sa_dev->port[i].agent)) goto err; - INIT_WORK(&sa_dev->port[i].update_task, - update_sm_ah, &sa_dev->port[i]); + INIT_WORK(&sa_dev->port[i].update_task, update_sm_ah); } ib_set_client_data(device, &sa_client, sa_dev); @@ -1010,7 +1010,7 @@ static void ib_sa_add_one(struct ib_device *device) goto err; for (i = 0; i <= e - s; ++i) - update_sm_ah(&sa_dev->port[i]); + update_sm_ah(&sa_dev->port[i].update_task); return; diff --git a/drivers/infiniband/core/uverbs_mem.c b/drivers/infiniband/core/uverbs_mem.c index efe147dbeb42..db12cc0841df 100644 --- a/drivers/infiniband/core/uverbs_mem.c +++ b/drivers/infiniband/core/uverbs_mem.c @@ -179,9 +179,10 @@ void ib_umem_release(struct ib_device *dev, struct ib_umem *umem) up_write(¤t->mm->mmap_sem); } -static void ib_umem_account(void *work_ptr) +static void ib_umem_account(struct work_struct *_work) { - struct ib_umem_account_work *work = work_ptr; + struct ib_umem_account_work *work = + container_of(_work, struct ib_umem_account_work, work); down_write(&work->mm->mmap_sem); work->mm->locked_vm -= work->diff; @@ -216,7 +217,7 @@ void ib_umem_release_on_close(struct ib_device *dev, struct ib_umem *umem) return; } - INIT_WORK(&work->work, ib_umem_account, work); + INIT_WORK(&work->work, ib_umem_account); work->mm = mm; work->diff = PAGE_ALIGN(umem->length + umem->offset) >> PAGE_SHIFT; diff --git a/drivers/infiniband/hw/ipath/ipath_user_pages.c b/drivers/infiniband/hw/ipath/ipath_user_pages.c index 413754b1d8a2..8536aeb96af8 100644 --- a/drivers/infiniband/hw/ipath/ipath_user_pages.c +++ b/drivers/infiniband/hw/ipath/ipath_user_pages.c @@ -214,9 +214,10 @@ struct ipath_user_pages_work { unsigned long num_pages; }; -static void user_pages_account(void *ptr) +static void user_pages_account(struct work_struct *_work) { - struct ipath_user_pages_work *work = ptr; + struct ipath_user_pages_work *work = + container_of(_work, struct ipath_user_pages_work, work); down_write(&work->mm->mmap_sem); work->mm->locked_vm -= work->num_pages; @@ -242,7 +243,7 @@ void ipath_release_user_pages_on_close(struct page **p, size_t num_pages) goto bail; - INIT_WORK(&work->work, user_pages_account, work); + INIT_WORK(&work->work, user_pages_account); work->mm = mm; work->num_pages = num_pages; diff --git a/drivers/infiniband/hw/mthca/mthca_catas.c b/drivers/infiniband/hw/mthca/mthca_catas.c index cd044ea2dfa4..e948158a28d9 100644 --- a/drivers/infiniband/hw/mthca/mthca_catas.c +++ b/drivers/infiniband/hw/mthca/mthca_catas.c @@ -57,7 +57,7 @@ static int catas_reset_disable; module_param_named(catas_reset_disable, catas_reset_disable, int, 0644); MODULE_PARM_DESC(catas_reset_disable, "disable reset on catastrophic event if nonzero"); -static void catas_reset(void *work_ptr) +static void catas_reset(struct work_struct *work) { struct mthca_dev *dev, *tmpdev; LIST_HEAD(tlist); @@ -203,7 +203,7 @@ void mthca_stop_catas_poll(struct mthca_dev *dev) int __init mthca_catas_init(void) { - INIT_WORK(&catas_work, catas_reset, NULL); + INIT_WORK(&catas_work, catas_reset); catas_wq = create_singlethread_workqueue("mthca_catas"); if (!catas_wq) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 0b8a79d53a00..3b2ee0eba05e 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -136,11 +136,11 @@ struct ipoib_dev_priv { struct list_head multicast_list; struct rb_root multicast_tree; - struct work_struct pkey_task; - struct work_struct mcast_task; + struct delayed_work pkey_task; + struct delayed_work mcast_task; struct work_struct flush_task; struct work_struct restart_task; - struct work_struct ah_reap_task; + struct delayed_work ah_reap_task; struct ib_device *ca; u8 port; @@ -254,13 +254,13 @@ int ipoib_add_pkey_attr(struct net_device *dev); void ipoib_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_ah *address, u32 qpn); -void ipoib_reap_ah(void *dev_ptr); +void ipoib_reap_ah(struct work_struct *work); void ipoib_flush_paths(struct net_device *dev); struct ipoib_dev_priv *ipoib_intf_alloc(const char *format); int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port); -void ipoib_ib_dev_flush(void *dev); +void ipoib_ib_dev_flush(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); int ipoib_ib_dev_open(struct net_device *dev); @@ -271,10 +271,10 @@ int ipoib_ib_dev_stop(struct net_device *dev); int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port); void ipoib_dev_cleanup(struct net_device *dev); -void ipoib_mcast_join_task(void *dev_ptr); +void ipoib_mcast_join_task(struct work_struct *work); void ipoib_mcast_send(struct net_device *dev, void *mgid, struct sk_buff *skb); -void ipoib_mcast_restart_task(void *dev_ptr); +void ipoib_mcast_restart_task(struct work_struct *work); int ipoib_mcast_start_thread(struct net_device *dev); int ipoib_mcast_stop_thread(struct net_device *dev, int flush); @@ -312,7 +312,7 @@ void ipoib_event(struct ib_event_handler *handler, int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey); int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey); -void ipoib_pkey_poll(void *dev); +void ipoib_pkey_poll(struct work_struct *work); int ipoib_pkey_dev_delay_open(struct net_device *dev); #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 8bf5e9ec7c95..f10fba5d3265 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -400,10 +400,11 @@ static void __ipoib_reap_ah(struct net_device *dev) spin_unlock_irq(&priv->tx_lock); } -void ipoib_reap_ah(void *dev_ptr) +void ipoib_reap_ah(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, ah_reap_task.work); + struct net_device *dev = priv->dev; __ipoib_reap_ah(dev); @@ -613,10 +614,11 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) return 0; } -void ipoib_ib_dev_flush(void *_dev) +void ipoib_ib_dev_flush(struct work_struct *work) { - struct net_device *dev = (struct net_device *)_dev; - struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv; + struct ipoib_dev_priv *cpriv, *priv = + container_of(work, struct ipoib_dev_priv, flush_task); + struct net_device *dev = priv->dev; if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags) ) { ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n"); @@ -638,14 +640,14 @@ void ipoib_ib_dev_flush(void *_dev) */ if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) { ipoib_ib_dev_up(dev); - ipoib_mcast_restart_task(dev); + ipoib_mcast_restart_task(&priv->restart_task); } mutex_lock(&priv->vlan_mutex); /* Flush any child interfaces too */ list_for_each_entry(cpriv, &priv->child_intfs, list) - ipoib_ib_dev_flush(cpriv->dev); + ipoib_ib_dev_flush(&cpriv->flush_task); mutex_unlock(&priv->vlan_mutex); } @@ -672,10 +674,11 @@ void ipoib_ib_dev_cleanup(struct net_device *dev) * change async notification is available. */ -void ipoib_pkey_poll(void *dev_ptr) +void ipoib_pkey_poll(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, pkey_task.work); + struct net_device *dev = priv->dev; ipoib_pkey_dev_check_presence(dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 85522daeb946..71114a8c12b9 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -933,11 +933,11 @@ static void ipoib_setup(struct net_device *dev) INIT_LIST_HEAD(&priv->dead_ahs); INIT_LIST_HEAD(&priv->multicast_list); - INIT_WORK(&priv->pkey_task, ipoib_pkey_poll, priv->dev); - INIT_WORK(&priv->mcast_task, ipoib_mcast_join_task, priv->dev); - INIT_WORK(&priv->flush_task, ipoib_ib_dev_flush, priv->dev); - INIT_WORK(&priv->restart_task, ipoib_mcast_restart_task, priv->dev); - INIT_WORK(&priv->ah_reap_task, ipoib_reap_ah, priv->dev); + INIT_DELAYED_WORK(&priv->pkey_task, ipoib_pkey_poll); + INIT_DELAYED_WORK(&priv->mcast_task, ipoib_mcast_join_task); + INIT_WORK(&priv->flush_task, ipoib_ib_dev_flush); + INIT_WORK(&priv->restart_task, ipoib_mcast_restart_task); + INIT_DELAYED_WORK(&priv->ah_reap_task, ipoib_reap_ah); } struct ipoib_dev_priv *ipoib_intf_alloc(const char *name) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 3faa1820f0e9..f0a4fac1a215 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -399,7 +399,8 @@ static void ipoib_mcast_join_complete(int status, mcast->backoff = 1; mutex_lock(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, + &priv->mcast_task, 0); mutex_unlock(&mcast_mutex); complete(&mcast->done); return; @@ -435,7 +436,8 @@ static void ipoib_mcast_join_complete(int status, if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) { if (status == -ETIMEDOUT) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, &priv->mcast_task, + 0); else queue_delayed_work(ipoib_workqueue, &priv->mcast_task, mcast->backoff * HZ); @@ -517,10 +519,11 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, mcast->query_id = ret; } -void ipoib_mcast_join_task(void *dev_ptr) +void ipoib_mcast_join_task(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, mcast_task.work); + struct net_device *dev = priv->dev; if (!test_bit(IPOIB_MCAST_RUN, &priv->flags)) return; @@ -610,7 +613,7 @@ int ipoib_mcast_start_thread(struct net_device *dev) mutex_lock(&mcast_mutex); if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags)) - queue_work(ipoib_workqueue, &priv->mcast_task); + queue_delayed_work(ipoib_workqueue, &priv->mcast_task, 0); mutex_unlock(&mcast_mutex); spin_lock_irq(&priv->lock); @@ -818,10 +821,11 @@ void ipoib_mcast_dev_flush(struct net_device *dev) } } -void ipoib_mcast_restart_task(void *dev_ptr) +void ipoib_mcast_restart_task(struct work_struct *work) { - struct net_device *dev = dev_ptr; - struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, restart_task); + struct net_device *dev = priv->dev; struct dev_mc_list *mclist; struct ipoib_mcast *mcast, *tmcast; LIST_HEAD(remove_list); diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index 18a000034996..693b77002897 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -48,7 +48,7 @@ static void iser_cq_tasklet_fn(unsigned long data); static void iser_cq_callback(struct ib_cq *cq, void *cq_context); -static void iser_comp_error_worker(void *data); +static void iser_comp_error_worker(struct work_struct *work); static void iser_cq_event_callback(struct ib_event *cause, void *context) { @@ -480,8 +480,7 @@ int iser_conn_init(struct iser_conn **ibconn) init_waitqueue_head(&ib_conn->wait); atomic_set(&ib_conn->post_recv_buf_count, 0); atomic_set(&ib_conn->post_send_buf_count, 0); - INIT_WORK(&ib_conn->comperror_work, iser_comp_error_worker, - ib_conn); + INIT_WORK(&ib_conn->comperror_work, iser_comp_error_worker); INIT_LIST_HEAD(&ib_conn->conn_list); spin_lock_init(&ib_conn->lock); @@ -754,9 +753,10 @@ int iser_post_send(struct iser_desc *tx_desc) return ret_val; } -static void iser_comp_error_worker(void *data) +static void iser_comp_error_worker(struct work_struct *work) { - struct iser_conn *ib_conn = data; + struct iser_conn *ib_conn = + container_of(work, struct iser_conn, comperror_work); /* getting here when the state is UP means that the conn is being * * terminated asynchronously from the iSCSI layer's perspective. */ diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index 4b09147f438f..214f66195af2 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -390,9 +390,10 @@ static void srp_disconnect_target(struct srp_target_port *target) wait_for_completion(&target->done); } -static void srp_remove_work(void *target_ptr) +static void srp_remove_work(struct work_struct *work) { - struct srp_target_port *target = target_ptr; + struct srp_target_port *target = + container_of(work, struct srp_target_port, work); spin_lock_irq(target->scsi_host->host_lock); if (target->state != SRP_TARGET_DEAD) { @@ -575,7 +576,7 @@ err: spin_lock_irq(target->scsi_host->host_lock); if (target->state == SRP_TARGET_CONNECTING) { target->state = SRP_TARGET_DEAD; - INIT_WORK(&target->work, srp_remove_work, target); + INIT_WORK(&target->work, srp_remove_work); schedule_work(&target->work); } spin_unlock_irq(target->scsi_host->host_lock); diff --git a/drivers/input/keyboard/lkkbd.c b/drivers/input/keyboard/lkkbd.c index 979b93e33da7..b7f049b45b6b 100644 --- a/drivers/input/keyboard/lkkbd.c +++ b/drivers/input/keyboard/lkkbd.c @@ -572,9 +572,9 @@ lkkbd_event (struct input_dev *dev, unsigned int type, unsigned int code, * were in. */ static void -lkkbd_reinit (void *data) +lkkbd_reinit (struct work_struct *work) { - struct lkkbd *lk = data; + struct lkkbd *lk = container_of(work, struct lkkbd, tq); int division; unsigned char leds_on = 0; unsigned char leds_off = 0; @@ -651,7 +651,7 @@ lkkbd_connect (struct serio *serio, struct serio_driver *drv) lk->serio = serio; lk->dev = input_dev; - INIT_WORK (&lk->tq, lkkbd_reinit, lk); + INIT_WORK (&lk->tq, lkkbd_reinit); lk->bell_volume = bell_volume; lk->keyclick_volume = keyclick_volume; lk->ctrlclick_volume = ctrlclick_volume; diff --git a/drivers/input/keyboard/sunkbd.c b/drivers/input/keyboard/sunkbd.c index cac4781103c3..6cd887c5eb0a 100644 --- a/drivers/input/keyboard/sunkbd.c +++ b/drivers/input/keyboard/sunkbd.c @@ -208,9 +208,9 @@ static int sunkbd_initialize(struct sunkbd *sunkbd) * were in. */ -static void sunkbd_reinit(void *data) +static void sunkbd_reinit(struct work_struct *work) { - struct sunkbd *sunkbd = data; + struct sunkbd *sunkbd = container_of(work, struct sunkbd, tq); wait_event_interruptible_timeout(sunkbd->wait, sunkbd->reset >= 0, HZ); @@ -248,7 +248,7 @@ static int sunkbd_connect(struct serio *serio, struct serio_driver *drv) sunkbd->serio = serio; sunkbd->dev = input_dev; init_waitqueue_head(&sunkbd->wait); - INIT_WORK(&sunkbd->tq, sunkbd_reinit, sunkbd); + INIT_WORK(&sunkbd->tq, sunkbd_reinit); snprintf(sunkbd->phys, sizeof(sunkbd->phys), "%s/input0", serio->phys); serio_set_drvdata(serio, sunkbd); diff --git a/drivers/input/mouse/psmouse-base.c b/drivers/input/mouse/psmouse-base.c index 6f9b2c7cc9c2..52bb2226ce2f 100644 --- a/drivers/input/mouse/psmouse-base.c +++ b/drivers/input/mouse/psmouse-base.c @@ -888,9 +888,10 @@ static int psmouse_poll(struct psmouse *psmouse) * psmouse_resync() attempts to re-validate current protocol. */ -static void psmouse_resync(void *p) +static void psmouse_resync(struct work_struct *work) { - struct psmouse *psmouse = p, *parent = NULL; + struct psmouse *parent = NULL, *psmouse = + container_of(work, struct psmouse, resync_work); struct serio *serio = psmouse->ps2dev.serio; psmouse_ret_t rc = PSMOUSE_GOOD_DATA; int failed = 0, enabled = 0; @@ -1121,7 +1122,7 @@ static int psmouse_connect(struct serio *serio, struct serio_driver *drv) goto out; ps2_init(&psmouse->ps2dev, serio); - INIT_WORK(&psmouse->resync_work, psmouse_resync, psmouse); + INIT_WORK(&psmouse->resync_work, psmouse_resync); psmouse->dev = input_dev; snprintf(psmouse->phys, sizeof(psmouse->phys), "%s/input0", serio->phys); diff --git a/drivers/isdn/act2000/capi.c b/drivers/isdn/act2000/capi.c index 6ae6eb322111..946c38cf6f8a 100644 --- a/drivers/isdn/act2000/capi.c +++ b/drivers/isdn/act2000/capi.c @@ -627,8 +627,10 @@ handle_ack(act2000_card *card, act2000_chan *chan, __u8 blocknr) { } void -actcapi_dispatch(act2000_card *card) +actcapi_dispatch(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, rcv_tq); struct sk_buff *skb; actcapi_msg *msg; __u16 ccmd; diff --git a/drivers/isdn/act2000/capi.h b/drivers/isdn/act2000/capi.h index 49f453c53c64..e55f6a931f66 100644 --- a/drivers/isdn/act2000/capi.h +++ b/drivers/isdn/act2000/capi.h @@ -356,7 +356,7 @@ extern int actcapi_connect_req(act2000_card *, act2000_chan *, char *, char, int extern void actcapi_select_b2_protocol_req(act2000_card *, act2000_chan *); extern void actcapi_disconnect_b3_req(act2000_card *, act2000_chan *); extern void actcapi_connect_resp(act2000_card *, act2000_chan *, __u8); -extern void actcapi_dispatch(act2000_card *); +extern void actcapi_dispatch(struct work_struct *); #ifdef DEBUG_MSG extern void actcapi_debug_msg(struct sk_buff *skb, int); #else diff --git a/drivers/isdn/act2000/module.c b/drivers/isdn/act2000/module.c index d89dcde4eade..90593e2ef872 100644 --- a/drivers/isdn/act2000/module.c +++ b/drivers/isdn/act2000/module.c @@ -192,8 +192,11 @@ act2000_set_msn(act2000_card *card, char *eazmsn) } static void -act2000_transmit(struct act2000_card *card) +act2000_transmit(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, snd_tq); + switch (card->bus) { case ACT2000_BUS_ISA: act2000_isa_send(card); @@ -207,8 +210,11 @@ act2000_transmit(struct act2000_card *card) } static void -act2000_receive(struct act2000_card *card) +act2000_receive(struct work_struct *work) { + struct act2000_card *card = + container_of(work, struct act2000_card, poll_tq); + switch (card->bus) { case ACT2000_BUS_ISA: act2000_isa_receive(card); @@ -227,7 +233,7 @@ act2000_poll(unsigned long data) act2000_card * card = (act2000_card *)data; unsigned long flags; - act2000_receive(card); + act2000_receive(&card->poll_tq); spin_lock_irqsave(&card->lock, flags); mod_timer(&card->ptimer, jiffies+3); spin_unlock_irqrestore(&card->lock, flags); @@ -578,9 +584,9 @@ act2000_alloccard(int bus, int port, int irq, char *id) skb_queue_head_init(&card->sndq); skb_queue_head_init(&card->rcvq); skb_queue_head_init(&card->ackq); - INIT_WORK(&card->snd_tq, (void *) (void *) act2000_transmit, card); - INIT_WORK(&card->rcv_tq, (void *) (void *) actcapi_dispatch, card); - INIT_WORK(&card->poll_tq, (void *) (void *) act2000_receive, card); + INIT_WORK(&card->snd_tq, act2000_transmit); + INIT_WORK(&card->rcv_tq, actcapi_dispatch); + INIT_WORK(&card->poll_tq, act2000_receive); init_timer(&card->ptimer); card->interface.owner = THIS_MODULE; card->interface.channels = ACT2000_BCH; diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c index 8c4fcb9027b3..783a25526315 100644 --- a/drivers/isdn/capi/kcapi.c +++ b/drivers/isdn/capi/kcapi.c @@ -208,9 +208,10 @@ static void notify_down(u32 contr) } } -static void notify_handler(void *data) +static void notify_handler(struct work_struct *work) { - struct capi_notifier *np = data; + struct capi_notifier *np = + container_of(work, struct capi_notifier, work); switch (np->cmd) { case KCI_CONTRUP: @@ -235,7 +236,7 @@ static int notify_push(unsigned int cmd, u32 controller, u16 applid, u32 ncci) if (!np) return -ENOMEM; - INIT_WORK(&np->work, notify_handler, np); + INIT_WORK(&np->work, notify_handler); np->cmd = cmd; np->controller = controller; np->applid = applid; @@ -248,10 +249,11 @@ static int notify_push(unsigned int cmd, u32 controller, u16 applid, u32 ncci) /* -------- Receiver ------------------------------------------ */ -static void recv_handler(void *_ap) +static void recv_handler(struct work_struct *work) { struct sk_buff *skb; - struct capi20_appl *ap = (struct capi20_appl *) _ap; + struct capi20_appl *ap = + container_of(work, struct capi20_appl, recv_work); if ((!ap) || (ap->release_in_progress)) return; @@ -527,7 +529,7 @@ u16 capi20_register(struct capi20_appl *ap) ap->callback = NULL; init_MUTEX(&ap->recv_sem); skb_queue_head_init(&ap->recv_queue); - INIT_WORK(&ap->recv_work, recv_handler, (void *)ap); + INIT_WORK(&ap->recv_work, recv_handler); ap->release_in_progress = 0; write_unlock_irqrestore(&application_lock, flags); diff --git a/drivers/isdn/hisax/amd7930_fn.c b/drivers/isdn/hisax/amd7930_fn.c index bec59010bc66..3b19caeba258 100644 --- a/drivers/isdn/hisax/amd7930_fn.c +++ b/drivers/isdn/hisax/amd7930_fn.c @@ -232,9 +232,10 @@ Amd7930_new_ph(struct IsdnCardState *cs) static void -Amd7930_bh(struct IsdnCardState *cs) +Amd7930_bh(struct work_struct *work) { - + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -789,7 +790,7 @@ Amd7930_init(struct IsdnCardState *cs) void __devinit setup_Amd7930(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) Amd7930_bh, cs); + INIT_WORK(&cs->tqueue, Amd7930_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c index 785b08554fca..cede72cdbb31 100644 --- a/drivers/isdn/hisax/config.c +++ b/drivers/isdn/hisax/config.c @@ -1137,7 +1137,6 @@ static int checkcard(int cardnr, char *id, int *busy_flag, struct module *lockow cs->tx_skb = NULL; cs->tx_cnt = 0; cs->event = 0; - cs->tqueue.data = cs; skb_queue_head_init(&cs->rq); skb_queue_head_init(&cs->sq); @@ -1554,7 +1553,7 @@ static void hisax_b_l2l1(struct PStack *st, int pr, void *arg); static int hisax_cardmsg(struct IsdnCardState *cs, int mt, void *arg); static int hisax_bc_setstack(struct PStack *st, struct BCState *bcs); static void hisax_bc_close(struct BCState *bcs); -static void hisax_bh(struct IsdnCardState *cs); +static void hisax_bh(struct work_struct *work); static void EChannel_proc_rcv(struct hisax_d_if *d_if); int hisax_register(struct hisax_d_if *hisax_d_if, struct hisax_b_if *b_if[], @@ -1586,7 +1585,7 @@ int hisax_register(struct hisax_d_if *hisax_d_if, struct hisax_b_if *b_if[], hisax_d_if->cs = cs; cs->hw.hisax_d_if = hisax_d_if; cs->cardmsg = hisax_cardmsg; - INIT_WORK(&cs->tqueue, (void *)(void *)hisax_bh, cs); + INIT_WORK(&cs->tqueue, hisax_bh); cs->channel[0].d_st->l2.l2l1 = hisax_d_l2l1; for (i = 0; i < 2; i++) { cs->bcs[i].BC_SetStack = hisax_bc_setstack; @@ -1618,8 +1617,10 @@ static void hisax_sched_event(struct IsdnCardState *cs, int event) schedule_work(&cs->tqueue); } -static void hisax_bh(struct IsdnCardState *cs) +static void hisax_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *st; int pr; diff --git a/drivers/isdn/hisax/hfc4s8s_l1.c b/drivers/isdn/hisax/hfc4s8s_l1.c index d852c9d998b2..de9b1a4d6bac 100644 --- a/drivers/isdn/hisax/hfc4s8s_l1.c +++ b/drivers/isdn/hisax/hfc4s8s_l1.c @@ -1083,8 +1083,9 @@ tx_b_frame(struct hfc4s8s_btype *bch) /* bottom half handler for interrupt */ /*************************************/ static void -hfc4s8s_bh(hfc4s8s_hw * hw) +hfc4s8s_bh(struct work_struct *work) { + hfc4s8s_hw *hw = container_of(work, hfc4s8s_hw, tqueue); u_char b; struct hfc4s8s_l1 *l1p; volatile u_char *fifo_stat; @@ -1550,7 +1551,7 @@ setup_instance(hfc4s8s_hw * hw) goto out; } - INIT_WORK(&hw->tqueue, (void *) (void *) hfc4s8s_bh, hw); + INIT_WORK(&hw->tqueue, hfc4s8s_bh); if (request_irq (hw->irq, hfc4s8s_interrupt, IRQF_SHARED, hw->card_name, hw)) { diff --git a/drivers/isdn/hisax/hfc_2bds0.c b/drivers/isdn/hisax/hfc_2bds0.c index 6360e8214720..8d9864453a23 100644 --- a/drivers/isdn/hisax/hfc_2bds0.c +++ b/drivers/isdn/hisax/hfc_2bds0.c @@ -549,10 +549,11 @@ setstack_2b(struct PStack *st, struct BCState *bcs) } static void -hfcd_bh(struct IsdnCardState *cs) +hfcd_bh(struct work_struct *work) { - if (!cs) - return; + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); + if (test_and_clear_bit(D_L1STATECHANGE, &cs->event)) { switch (cs->dc.hfcd.ph_state) { case (0): @@ -1072,5 +1073,5 @@ set_cs_func(struct IsdnCardState *cs) cs->dbusytimer.function = (void *) hfc_dbusy_timer; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); - INIT_WORK(&cs->tqueue, (void *)(void *) hfcd_bh, cs); + INIT_WORK(&cs->tqueue, hfcd_bh); } diff --git a/drivers/isdn/hisax/hfc_pci.c b/drivers/isdn/hisax/hfc_pci.c index 93f60b563515..5db0a85b827f 100644 --- a/drivers/isdn/hisax/hfc_pci.c +++ b/drivers/isdn/hisax/hfc_pci.c @@ -1506,8 +1506,10 @@ setstack_2b(struct PStack *st, struct BCState *bcs) /* handle L1 state changes */ /***************************/ static void -hfcpci_bh(struct IsdnCardState *cs) +hfcpci_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); u_long flags; // struct PStack *stptr; @@ -1722,7 +1724,7 @@ setup_hfcpci(struct IsdnCard *card) Write_hfc(cs, HFCPCI_INT_M2, cs->hw.hfcpci.int_m2); /* At this point the needed PCI config is done */ /* fifos are still not enabled */ - INIT_WORK(&cs->tqueue, (void *)(void *) hfcpci_bh, cs); + INIT_WORK(&cs->tqueue, hfcpci_bh); cs->setstack_d = setstack_hfcpci; cs->BC_Send_Data = &hfcpci_send_data; cs->readisac = NULL; diff --git a/drivers/isdn/hisax/hfc_sx.c b/drivers/isdn/hisax/hfc_sx.c index 954d1536db1f..4fd09d21a27f 100644 --- a/drivers/isdn/hisax/hfc_sx.c +++ b/drivers/isdn/hisax/hfc_sx.c @@ -1251,8 +1251,10 @@ setstack_2b(struct PStack *st, struct BCState *bcs) /* handle L1 state changes */ /***************************/ static void -hfcsx_bh(struct IsdnCardState *cs) +hfcsx_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); u_long flags; if (!cs) @@ -1499,7 +1501,7 @@ setup_hfcsx(struct IsdnCard *card) cs->dbusytimer.function = (void *) hfcsx_dbusy_timer; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); - INIT_WORK(&cs->tqueue, (void *)(void *) hfcsx_bh, cs); + INIT_WORK(&cs->tqueue, hfcsx_bh); cs->readisac = NULL; cs->writeisac = NULL; cs->readisacfifo = NULL; diff --git a/drivers/isdn/hisax/icc.c b/drivers/isdn/hisax/icc.c index da706925d54d..682cac32f259 100644 --- a/drivers/isdn/hisax/icc.c +++ b/drivers/isdn/hisax/icc.c @@ -77,8 +77,10 @@ icc_new_ph(struct IsdnCardState *cs) } static void -icc_bh(struct IsdnCardState *cs) +icc_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -674,7 +676,7 @@ clear_pending_icc_ints(struct IsdnCardState *cs) void __devinit setup_icc(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) icc_bh, cs); + INIT_WORK(&cs->tqueue, icc_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/isac.c b/drivers/isdn/hisax/isac.c index 282f349408bc..4e9f23803dae 100644 --- a/drivers/isdn/hisax/isac.c +++ b/drivers/isdn/hisax/isac.c @@ -81,8 +81,10 @@ isac_new_ph(struct IsdnCardState *cs) } static void -isac_bh(struct IsdnCardState *cs) +isac_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -674,7 +676,7 @@ clear_pending_isac_ints(struct IsdnCardState *cs) void __devinit setup_isac(struct IsdnCardState *cs) { - INIT_WORK(&cs->tqueue, (void *)(void *) isac_bh, cs); + INIT_WORK(&cs->tqueue, isac_bh); cs->dbusytimer.function = (void *) dbusy_timer_handler; cs->dbusytimer.data = (long) cs; init_timer(&cs->dbusytimer); diff --git a/drivers/isdn/hisax/isar.c b/drivers/isdn/hisax/isar.c index 674af673ff96..6f1a6583b17d 100644 --- a/drivers/isdn/hisax/isar.c +++ b/drivers/isdn/hisax/isar.c @@ -437,8 +437,10 @@ extern void BChannel_bh(struct BCState *); #define B_LL_OK 10 static void -isar_bh(struct BCState *bcs) +isar_bh(struct work_struct *work) { + struct BCState *bcs = container_of(work, struct BCState, tqueue); + BChannel_bh(bcs); if (test_and_clear_bit(B_LL_NOCARRIER, &bcs->event)) ll_deliver_faxstat(bcs, ISDN_FAX_CLASS1_NOCARR); @@ -1580,7 +1582,7 @@ isar_setup(struct IsdnCardState *cs) cs->bcs[i].mode = 0; cs->bcs[i].hw.isar.dpath = i + 1; modeisar(&cs->bcs[i], 0, 0); - INIT_WORK(&cs->bcs[i].tqueue, (void *)(void *) isar_bh, &cs->bcs[i]); + INIT_WORK(&cs->bcs[i].tqueue, isar_bh); } } diff --git a/drivers/isdn/hisax/isdnl1.c b/drivers/isdn/hisax/isdnl1.c index bab356886483..a14204ec88ee 100644 --- a/drivers/isdn/hisax/isdnl1.c +++ b/drivers/isdn/hisax/isdnl1.c @@ -315,8 +315,10 @@ BChannel_proc_ack(struct BCState *bcs) } void -BChannel_bh(struct BCState *bcs) +BChannel_bh(struct work_struct *work) { + struct BCState *bcs = container_of(work, struct BCState, tqueue); + if (!bcs) return; if (test_and_clear_bit(B_RCVBUFREADY, &bcs->event)) @@ -362,7 +364,7 @@ init_bcstate(struct IsdnCardState *cs, int bc) bcs->cs = cs; bcs->channel = bc; - INIT_WORK(&bcs->tqueue, (void *)(void *) BChannel_bh, bcs); + INIT_WORK(&bcs->tqueue, BChannel_bh); spin_lock_init(&bcs->aclock); bcs->BC_SetStack = NULL; bcs->BC_Close = NULL; diff --git a/drivers/isdn/hisax/w6692.c b/drivers/isdn/hisax/w6692.c index 1655341797a9..3aeceaf9769e 100644 --- a/drivers/isdn/hisax/w6692.c +++ b/drivers/isdn/hisax/w6692.c @@ -101,8 +101,10 @@ W6692_new_ph(struct IsdnCardState *cs) } static void -W6692_bh(struct IsdnCardState *cs) +W6692_bh(struct work_struct *work) { + struct IsdnCardState *cs = + container_of(work, struct IsdnCardState, tqueue); struct PStack *stptr; if (!cs) @@ -1070,7 +1072,7 @@ setup_w6692(struct IsdnCard *card) id_list[cs->subtyp].card_name, cs->irq, cs->hw.w6692.iobase); - INIT_WORK(&cs->tqueue, (void *)(void *) W6692_bh, cs); + INIT_WORK(&cs->tqueue, W6692_bh); cs->readW6692 = &ReadW6692; cs->writeW6692 = &WriteW6692; cs->readisacfifo = &ReadISACfifo; diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c index 1f8d6ae66b41..2e4daebfb7e0 100644 --- a/drivers/isdn/i4l/isdn_net.c +++ b/drivers/isdn/i4l/isdn_net.c @@ -984,9 +984,9 @@ void isdn_net_write_super(isdn_net_local *lp, struct sk_buff *skb) /* * called from tq_immediate */ -static void isdn_net_softint(void *private) +static void isdn_net_softint(struct work_struct *work) { - isdn_net_local *lp = private; + isdn_net_local *lp = container_of(work, isdn_net_local, tqueue); struct sk_buff *skb; spin_lock_bh(&lp->xmit_lock); @@ -2596,7 +2596,7 @@ isdn_net_new(char *name, struct net_device *master) netdev->local->netdev = netdev; netdev->local->next = netdev->local; - INIT_WORK(&netdev->local->tqueue, (void *)(void *) isdn_net_softint, netdev->local); + INIT_WORK(&netdev->local->tqueue, isdn_net_softint); spin_lock_init(&netdev->local->xmit_lock); netdev->local->isdn_device = -1; diff --git a/drivers/isdn/pcbit/drv.c b/drivers/isdn/pcbit/drv.c index 6ead5e1508b7..1966f3410a13 100644 --- a/drivers/isdn/pcbit/drv.c +++ b/drivers/isdn/pcbit/drv.c @@ -68,8 +68,6 @@ static void pcbit_set_msn(struct pcbit_dev *dev, char *list); static int pcbit_check_msn(struct pcbit_dev *dev, char *msn); -extern void pcbit_deliver(void * data); - int pcbit_init_dev(int board, int mem_base, int irq) { struct pcbit_dev *dev; @@ -129,7 +127,7 @@ int pcbit_init_dev(int board, int mem_base, int irq) memset(dev->b2, 0, sizeof(struct pcbit_chan)); dev->b2->id = 1; - INIT_WORK(&dev->qdelivery, pcbit_deliver, dev); + INIT_WORK(&dev->qdelivery, pcbit_deliver); /* * interrupts diff --git a/drivers/isdn/pcbit/layer2.c b/drivers/isdn/pcbit/layer2.c index 937fd2120381..0c9f6df873fc 100644 --- a/drivers/isdn/pcbit/layer2.c +++ b/drivers/isdn/pcbit/layer2.c @@ -67,7 +67,6 @@ extern void pcbit_l3_receive(struct pcbit_dev *dev, ulong msg, * Prototypes */ -void pcbit_deliver(void *data); static void pcbit_transmit(struct pcbit_dev *dev); static void pcbit_recv_ack(struct pcbit_dev *dev, unsigned char ack); @@ -299,11 +298,12 @@ pcbit_transmit(struct pcbit_dev *dev) */ void -pcbit_deliver(void *data) +pcbit_deliver(struct work_struct *work) { struct frame_buf *frame; unsigned long flags, msg; - struct pcbit_dev *dev = (struct pcbit_dev *) data; + struct pcbit_dev *dev = + container_of(work, struct pcbit_dev, qdelivery); spin_lock_irqsave(&dev->lock, flags); diff --git a/drivers/isdn/pcbit/pcbit.h b/drivers/isdn/pcbit/pcbit.h index 388bacefd23a..19c18e88ff16 100644 --- a/drivers/isdn/pcbit/pcbit.h +++ b/drivers/isdn/pcbit/pcbit.h @@ -166,4 +166,6 @@ struct pcbit_ioctl { #define L2_RUNNING 5 #define L2_ERROR 6 +extern void pcbit_deliver(struct work_struct *work); + #endif diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index ade25b3fbb35..4871158aca3e 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -600,7 +600,7 @@ core_initcall(smu_late_init); * sysfs visibility */ -static void smu_expose_childs(void *unused) +static void smu_expose_childs(struct work_struct *unused) { struct device_node *np; @@ -610,7 +610,7 @@ static void smu_expose_childs(void *unused) &smu->of_dev->dev); } -static DECLARE_WORK(smu_expose_childs_work, smu_expose_childs, NULL); +static DECLARE_WORK(smu_expose_childs_work, smu_expose_childs); static int smu_platform_probe(struct of_device* dev, const struct of_device_id *match) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 08a40f4e4f60..ed2d4ef27fd8 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -458,11 +458,11 @@ static void dec_pending(struct crypt_io *io, int error) * interrupt context. */ static struct workqueue_struct *_kcryptd_workqueue; -static void kcryptd_do_work(void *data); +static void kcryptd_do_work(struct work_struct *work); static void kcryptd_queue_io(struct crypt_io *io) { - INIT_WORK(&io->work, kcryptd_do_work, io); + INIT_WORK(&io->work, kcryptd_do_work); queue_work(_kcryptd_workqueue, &io->work); } @@ -618,9 +618,9 @@ static void process_read_endio(struct crypt_io *io) dec_pending(io, crypt_convert(cc, &ctx)); } -static void kcryptd_do_work(void *data) +static void kcryptd_do_work(struct work_struct *work) { - struct crypt_io *io = data; + struct crypt_io *io = container_of(work, struct crypt_io, work); if (io->post_process) process_read_endio(io); diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index d754e0bc6e90..e77ee6fd1044 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -104,8 +104,8 @@ typedef int (*action_fn) (struct pgpath *pgpath); static kmem_cache_t *_mpio_cache; struct workqueue_struct *kmultipathd; -static void process_queued_ios(void *data); -static void trigger_event(void *data); +static void process_queued_ios(struct work_struct *work); +static void trigger_event(struct work_struct *work); /*----------------------------------------------- @@ -173,8 +173,8 @@ static struct multipath *alloc_multipath(struct dm_target *ti) INIT_LIST_HEAD(&m->priority_groups); spin_lock_init(&m->lock); m->queue_io = 1; - INIT_WORK(&m->process_queued_ios, process_queued_ios, m); - INIT_WORK(&m->trigger_event, trigger_event, m); + INIT_WORK(&m->process_queued_ios, process_queued_ios); + INIT_WORK(&m->trigger_event, trigger_event); m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache); if (!m->mpio_pool) { kfree(m); @@ -379,9 +379,10 @@ static void dispatch_queued_ios(struct multipath *m) } } -static void process_queued_ios(void *data) +static void process_queued_ios(struct work_struct *work) { - struct multipath *m = (struct multipath *) data; + struct multipath *m = + container_of(work, struct multipath, process_queued_ios); struct hw_handler *hwh = &m->hw_handler; struct pgpath *pgpath = NULL; unsigned init_required = 0, must_queue = 1; @@ -421,9 +422,10 @@ out: * An event is triggered whenever a path is taken out of use. * Includes path failure and PG bypass. */ -static void trigger_event(void *data) +static void trigger_event(struct work_struct *work) { - struct multipath *m = (struct multipath *) data; + struct multipath *m = + container_of(work, struct multipath, trigger_event); dm_table_event(m->ti->table); } diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 48a653b3f518..fc8cbb168e3e 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -883,7 +883,7 @@ static void do_mirror(struct mirror_set *ms) do_writes(ms, &writes); } -static void do_work(void *ignored) +static void do_work(struct work_struct *ignored) { struct mirror_set *ms; @@ -1269,7 +1269,7 @@ static int __init dm_mirror_init(void) dm_dirty_log_exit(); return r; } - INIT_WORK(&_kmirrord_work, do_work, NULL); + INIT_WORK(&_kmirrord_work, do_work); r = dm_register_target(&mirror_target); if (r < 0) { diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 5281e0094072..91c7aa1fed0e 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -40,7 +40,7 @@ #define SNAPSHOT_PAGES 256 struct workqueue_struct *ksnapd; -static void flush_queued_bios(void *data); +static void flush_queued_bios(struct work_struct *work); struct pending_exception { struct exception e; @@ -528,7 +528,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) } bio_list_init(&s->queued_bios); - INIT_WORK(&s->queued_bios_work, flush_queued_bios, s); + INIT_WORK(&s->queued_bios_work, flush_queued_bios); /* Add snapshot to the list of snapshots for this origin */ /* Exceptions aren't triggered till snapshot_resume() is called */ @@ -603,9 +603,10 @@ static void flush_bios(struct bio *bio) } } -static void flush_queued_bios(void *data) +static void flush_queued_bios(struct work_struct *work) { - struct dm_snapshot *s = (struct dm_snapshot *) data; + struct dm_snapshot *s = + container_of(work, struct dm_snapshot, queued_bios_work); struct bio *queued_bios; unsigned long flags; diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c index f1db6eff4857..b3c01496c737 100644 --- a/drivers/md/kcopyd.c +++ b/drivers/md/kcopyd.c @@ -417,7 +417,7 @@ static int process_jobs(struct list_head *jobs, int (*fn) (struct kcopyd_job *)) /* * kcopyd does this every time it's woken up. */ -static void do_work(void *ignored) +static void do_work(struct work_struct *ignored) { /* * The order that these are called is *very* important. @@ -628,7 +628,7 @@ static int kcopyd_init(void) } kcopyd_clients++; - INIT_WORK(&_kcopyd_work, do_work, NULL); + INIT_WORK(&_kcopyd_work, do_work); mutex_unlock(&kcopyd_init_lock); return 0; } diff --git a/drivers/media/dvb/b2c2/flexcop-pci.c b/drivers/media/dvb/b2c2/flexcop-pci.c index 06893243f3d4..6e166801505d 100644 --- a/drivers/media/dvb/b2c2/flexcop-pci.c +++ b/drivers/media/dvb/b2c2/flexcop-pci.c @@ -63,7 +63,7 @@ struct flexcop_pci { unsigned long last_irq; - struct work_struct irq_check_work; + struct delayed_work irq_check_work; struct flexcop_device *fc_dev; }; @@ -97,9 +97,10 @@ static int flexcop_pci_write_ibi_reg(struct flexcop_device *fc, flexcop_ibi_regi return 0; } -static void flexcop_pci_irq_check_work(void *data) +static void flexcop_pci_irq_check_work(struct work_struct *work) { - struct flexcop_pci *fc_pci = data; + struct flexcop_pci *fc_pci = + container_of(work, struct flexcop_pci, irq_check_work.work); struct flexcop_device *fc = fc_pci->fc_dev; flexcop_ibi_value v = fc->read_ibi_reg(fc,sram_dest_reg_714); @@ -371,7 +372,7 @@ static int flexcop_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e if ((ret = flexcop_pci_dma_init(fc_pci)) != 0) goto err_fc_exit; - INIT_WORK(&fc_pci->irq_check_work, flexcop_pci_irq_check_work, fc_pci); + INIT_DELAYED_WORK(&fc_pci->irq_check_work, flexcop_pci_irq_check_work); return ret; diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index ff7d4f56ced3..dd0bcbe140bd 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -127,7 +127,7 @@ struct cinergyt2 { struct dvbt_set_parameters_msg param; struct dvbt_get_status_msg status; - struct work_struct query_work; + struct delayed_work query_work; wait_queue_head_t poll_wq; int pending_fe_events; @@ -141,7 +141,7 @@ struct cinergyt2 { #ifdef ENABLE_RC struct input_dev *rc_input_dev; char phys[64]; - struct work_struct rc_query_work; + struct delayed_work rc_query_work; int rc_input_event; u32 rc_last_code; unsigned long last_event_jiffies; @@ -724,9 +724,10 @@ static struct dvb_device cinergyt2_fe_template = { #ifdef ENABLE_RC -static void cinergyt2_query_rc (void *data) +static void cinergyt2_query_rc (struct work_struct *work) { - struct cinergyt2 *cinergyt2 = data; + struct cinergyt2 *cinergyt2 = + container_of(work, struct cinergyt2, rc_query_work.work); char buf[1] = { CINERGYT2_EP1_GET_RC_EVENTS }; struct cinergyt2_rc_event rc_events[12]; int n, len, i; @@ -807,7 +808,7 @@ static int cinergyt2_register_rc(struct cinergyt2 *cinergyt2) strlcat(cinergyt2->phys, "/input0", sizeof(cinergyt2->phys)); cinergyt2->rc_input_event = KEY_MAX; cinergyt2->rc_last_code = ~0; - INIT_WORK(&cinergyt2->rc_query_work, cinergyt2_query_rc, cinergyt2); + INIT_DELAYED_WORK(&cinergyt2->rc_query_work, cinergyt2_query_rc); input_dev->name = DRIVER_NAME " remote control"; input_dev->phys = cinergyt2->phys; @@ -848,9 +849,10 @@ static inline void cinergyt2_resume_rc(struct cinergyt2 *cinergyt2) { } #endif /* ENABLE_RC */ -static void cinergyt2_query (void *data) +static void cinergyt2_query (struct work_struct *work) { - struct cinergyt2 *cinergyt2 = (struct cinergyt2 *) data; + struct cinergyt2 *cinergyt2 = + container_of(work, struct cinergyt2, query_work.work); char cmd [] = { CINERGYT2_EP1_GET_TUNER_STATUS }; struct dvbt_get_status_msg *s = &cinergyt2->status; uint8_t lock_bits; @@ -894,7 +896,7 @@ static int cinergyt2_probe (struct usb_interface *intf, mutex_init(&cinergyt2->sem); init_waitqueue_head (&cinergyt2->poll_wq); - INIT_WORK(&cinergyt2->query_work, cinergyt2_query, cinergyt2); + INIT_DELAYED_WORK(&cinergyt2->query_work, cinergyt2_query); cinergyt2->udev = interface_to_usbdev(intf); cinergyt2->param.cmd = CINERGYT2_EP1_SET_TUNER_PARAMETERS; diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c index 8859ab74f0fe..ebf4dc5190f6 100644 --- a/drivers/media/dvb/dvb-core/dvb_net.c +++ b/drivers/media/dvb/dvb-core/dvb_net.c @@ -127,6 +127,7 @@ struct dvb_net_priv { int in_use; struct net_device_stats stats; u16 pid; + struct net_device *net; struct dvb_net *host; struct dmx_demux *demux; struct dmx_section_feed *secfeed; @@ -1123,10 +1124,11 @@ static int dvb_set_mc_filter (struct net_device *dev, struct dev_mc_list *mc) } -static void wq_set_multicast_list (void *data) +static void wq_set_multicast_list (struct work_struct *work) { - struct net_device *dev = data; - struct dvb_net_priv *priv = dev->priv; + struct dvb_net_priv *priv = + container_of(work, struct dvb_net_priv, set_multicast_list_wq); + struct net_device *dev = priv->net; dvb_net_feed_stop(dev); priv->rx_mode = RX_MODE_UNI; @@ -1167,9 +1169,11 @@ static void dvb_net_set_multicast_list (struct net_device *dev) } -static void wq_restart_net_feed (void *data) +static void wq_restart_net_feed (struct work_struct *work) { - struct net_device *dev = data; + struct dvb_net_priv *priv = + container_of(work, struct dvb_net_priv, restart_net_feed_wq); + struct net_device *dev = priv->net; if (netif_running(dev)) { dvb_net_feed_stop(dev); @@ -1276,6 +1280,7 @@ static int dvb_net_add_if(struct dvb_net *dvbnet, u16 pid, u8 feedtype) dvbnet->device[if_num] = net; priv = net->priv; + priv->net = net; priv->demux = dvbnet->demux; priv->pid = pid; priv->rx_mode = RX_MODE_UNI; @@ -1284,8 +1289,8 @@ static int dvb_net_add_if(struct dvb_net *dvbnet, u16 pid, u8 feedtype) priv->feedtype = feedtype; reset_ule(priv); - INIT_WORK(&priv->set_multicast_list_wq, wq_set_multicast_list, net); - INIT_WORK(&priv->restart_net_feed_wq, wq_restart_net_feed, net); + INIT_WORK(&priv->set_multicast_list_wq, wq_set_multicast_list); + INIT_WORK(&priv->restart_net_feed_wq, wq_restart_net_feed); mutex_init(&priv->mutex); net->base_addr = pid; diff --git a/drivers/media/dvb/dvb-usb/dvb-usb-remote.c b/drivers/media/dvb/dvb-usb/dvb-usb-remote.c index 0a3a0b6c2350..794e4471561c 100644 --- a/drivers/media/dvb/dvb-usb/dvb-usb-remote.c +++ b/drivers/media/dvb/dvb-usb/dvb-usb-remote.c @@ -13,9 +13,10 @@ * * TODO: Fix the repeat rate of the input device. */ -static void dvb_usb_read_remote_control(void *data) +static void dvb_usb_read_remote_control(struct work_struct *work) { - struct dvb_usb_device *d = data; + struct dvb_usb_device *d = + container_of(work, struct dvb_usb_device, rc_query_work.work); u32 event; int state; @@ -128,7 +129,7 @@ int dvb_usb_remote_init(struct dvb_usb_device *d) input_register_device(d->rc_input_dev); - INIT_WORK(&d->rc_query_work, dvb_usb_read_remote_control, d); + INIT_DELAYED_WORK(&d->rc_query_work, dvb_usb_read_remote_control); info("schedule remote query interval to %d msecs.", d->props.rc_interval); schedule_delayed_work(&d->rc_query_work,msecs_to_jiffies(d->props.rc_interval)); diff --git a/drivers/media/dvb/dvb-usb/dvb-usb.h b/drivers/media/dvb/dvb-usb/dvb-usb.h index 376c45a8e779..0d721731a524 100644 --- a/drivers/media/dvb/dvb-usb/dvb-usb.h +++ b/drivers/media/dvb/dvb-usb/dvb-usb.h @@ -369,7 +369,7 @@ struct dvb_usb_device { /* remote control */ struct input_dev *rc_input_dev; char rc_phys[64]; - struct work_struct rc_query_work; + struct delayed_work rc_query_work; u32 last_event; int last_state; diff --git a/drivers/media/video/cpia_pp.c b/drivers/media/video/cpia_pp.c index 41f4b8d17559..b12cec94f4cc 100644 --- a/drivers/media/video/cpia_pp.c +++ b/drivers/media/video/cpia_pp.c @@ -82,6 +82,8 @@ struct pp_cam_entry { struct pardevice *pdev; struct parport *port; struct work_struct cb_task; + void (*cb_func)(void *cbdata); + void *cb_data; int open_count; wait_queue_head_t wq_stream; /* image state flags */ @@ -130,6 +132,20 @@ static void cpia_parport_disable_irq( struct parport *port ) { #define PARPORT_CHUNK_SIZE PAGE_SIZE +static void cpia_pp_run_callback(struct work_struct *work) +{ + void (*cb_func)(void *cbdata); + void *cb_data; + struct pp_cam_entry *cam; + + cam = container_of(work, struct pp_cam_entry, cb_task); + cb_func = cam->cb_func; + cb_data = cam->cb_data; + work_release(work); + + cb_func(cb_data); +} + /**************************************************************************** * * CPiA-specific low-level parport functions for nibble uploads @@ -664,7 +680,9 @@ static int cpia_pp_registerCallback(void *privdata, void (*cb)(void *cbdata), vo int retval = 0; if(cam->port->irq != PARPORT_IRQ_NONE) { - INIT_WORK(&cam->cb_task, cb, cbdata); + cam->cb_func = cb; + cam->cb_data = cbdata; + INIT_WORK_NAR(&cam->cb_task, cpia_pp_run_callback); } else { retval = -1; } diff --git a/drivers/media/video/cx88/cx88-input.c b/drivers/media/video/cx88/cx88-input.c index 57e1c024a547..e60a0a52e4b2 100644 --- a/drivers/media/video/cx88/cx88-input.c +++ b/drivers/media/video/cx88/cx88-input.c @@ -145,9 +145,9 @@ static void ir_timer(unsigned long data) schedule_work(&ir->work); } -static void cx88_ir_work(void *data) +static void cx88_ir_work(struct work_struct *work) { - struct cx88_IR *ir = data; + struct cx88_IR *ir = container_of(work, struct cx88_IR, work); unsigned long timeout; cx88_ir_handle_key(ir); @@ -308,7 +308,7 @@ int cx88_ir_init(struct cx88_core *core, struct pci_dev *pci) core->ir = ir; if (ir->polling) { - INIT_WORK(&ir->work, cx88_ir_work, ir); + INIT_WORK(&ir->work, cx88_ir_work); init_timer(&ir->timer); ir->timer.function = ir_timer; ir->timer.data = (unsigned long)ir; diff --git a/drivers/media/video/ir-kbd-i2c.c b/drivers/media/video/ir-kbd-i2c.c index 1457b1602221..ab87e7bfe84f 100644 --- a/drivers/media/video/ir-kbd-i2c.c +++ b/drivers/media/video/ir-kbd-i2c.c @@ -268,9 +268,9 @@ static void ir_timer(unsigned long data) schedule_work(&ir->work); } -static void ir_work(void *data) +static void ir_work(struct work_struct *work) { - struct IR_i2c *ir = data; + struct IR_i2c *ir = container_of(work, struct IR_i2c, work); ir_key_poll(ir); mod_timer(&ir->timer, jiffies+HZ/10); } @@ -400,7 +400,7 @@ static int ir_attach(struct i2c_adapter *adap, int addr, ir->input->name,ir->input->phys,adap->name); /* start polling via eventd */ - INIT_WORK(&ir->work, ir_work, ir); + INIT_WORK(&ir->work, ir_work); init_timer(&ir->timer); ir->timer.function = ir_timer; ir->timer.data = (unsigned long)ir; diff --git a/drivers/media/video/pvrusb2/pvrusb2-context.c b/drivers/media/video/pvrusb2/pvrusb2-context.c index f129f316d20e..cf129746205d 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-context.c +++ b/drivers/media/video/pvrusb2/pvrusb2-context.c @@ -45,16 +45,21 @@ static void pvr2_context_trigger_poll(struct pvr2_context *mp) } -static void pvr2_context_poll(struct pvr2_context *mp) +static void pvr2_context_poll(struct work_struct *work) { + struct pvr2_context *mp = + container_of(work, struct pvr2_context, workpoll); pvr2_context_enter(mp); do { pvr2_hdw_poll(mp->hdw); } while (0); pvr2_context_exit(mp); } -static void pvr2_context_setup(struct pvr2_context *mp) +static void pvr2_context_setup(struct work_struct *work) { + struct pvr2_context *mp = + container_of(work, struct pvr2_context, workinit); + pvr2_context_enter(mp); do { if (!pvr2_hdw_dev_ok(mp->hdw)) break; pvr2_hdw_setup(mp->hdw); @@ -92,8 +97,8 @@ struct pvr2_context *pvr2_context_create( } mp->workqueue = create_singlethread_workqueue("pvrusb2"); - INIT_WORK(&mp->workinit,(void (*)(void*))pvr2_context_setup,mp); - INIT_WORK(&mp->workpoll,(void (*)(void*))pvr2_context_poll,mp); + INIT_WORK(&mp->workinit, pvr2_context_setup); + INIT_WORK(&mp->workpoll, pvr2_context_poll); queue_work(mp->workqueue,&mp->workinit); done: return mp; diff --git a/drivers/media/video/saa6588.c b/drivers/media/video/saa6588.c index a81285ca7d5b..8ba05c214ca7 100644 --- a/drivers/media/video/saa6588.c +++ b/drivers/media/video/saa6588.c @@ -322,9 +322,9 @@ static void saa6588_timer(unsigned long data) schedule_work(&s->work); } -static void saa6588_work(void *data) +static void saa6588_work(struct work_struct *work) { - struct saa6588 *s = (struct saa6588 *)data; + struct saa6588 *s = container_of(work, struct saa6588, work); saa6588_i2c_poll(s); mod_timer(&s->timer, jiffies + msecs_to_jiffies(20)); @@ -417,7 +417,7 @@ static int saa6588_attach(struct i2c_adapter *adap, int addr, int kind) saa6588_configure(s); /* start polling via eventd */ - INIT_WORK(&s->work, saa6588_work, s); + INIT_WORK(&s->work, saa6588_work); init_timer(&s->timer); s->timer.function = saa6588_timer; s->timer.data = (unsigned long)s; diff --git a/drivers/media/video/saa7134/saa7134-empress.c b/drivers/media/video/saa7134/saa7134-empress.c index 65d044086ce9..daaae870a2c4 100644 --- a/drivers/media/video/saa7134/saa7134-empress.c +++ b/drivers/media/video/saa7134/saa7134-empress.c @@ -343,9 +343,10 @@ static struct video_device saa7134_empress_template = .minor = -1, }; -static void empress_signal_update(void* data) +static void empress_signal_update(struct work_struct *work) { - struct saa7134_dev* dev = (struct saa7134_dev*) data; + struct saa7134_dev* dev = + container_of(work, struct saa7134_dev, empress_workqueue); if (dev->nosignal) { dprintk("no video signal\n"); @@ -378,7 +379,7 @@ static int empress_init(struct saa7134_dev *dev) "%s empress (%s)", dev->name, saa7134_boards[dev->board].name); - INIT_WORK(&dev->empress_workqueue, empress_signal_update, (void*) dev); + INIT_WORK(&dev->empress_workqueue, empress_signal_update); err = video_register_device(dev->empress_dev,VFL_TYPE_GRABBER, empress_nr[dev->nr]); @@ -399,7 +400,7 @@ static int empress_init(struct saa7134_dev *dev) sizeof(struct saa7134_buf), dev); - empress_signal_update(dev); + empress_signal_update(&dev->empress_workqueue); return 0; } diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c index 1dd491773150..ef2b55e19910 100644 --- a/drivers/message/fusion/mptfc.c +++ b/drivers/message/fusion/mptfc.c @@ -1018,9 +1018,10 @@ mptfc_init_host_attr(MPT_ADAPTER *ioc,int portnum) } static void -mptfc_setup_reset(void *arg) +mptfc_setup_reset(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = + container_of(work, MPT_ADAPTER, fc_setup_reset_work); u64 pn; struct mptfc_rport_info *ri; @@ -1043,9 +1044,10 @@ mptfc_setup_reset(void *arg) } static void -mptfc_rescan_devices(void *arg) +mptfc_rescan_devices(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = + container_of(work, MPT_ADAPTER, fc_rescan_work); int ii; u64 pn; struct mptfc_rport_info *ri; @@ -1154,8 +1156,8 @@ mptfc_probe(struct pci_dev *pdev, const struct pci_device_id *id) } spin_lock_init(&ioc->fc_rescan_work_lock); - INIT_WORK(&ioc->fc_rescan_work, mptfc_rescan_devices,(void *)ioc); - INIT_WORK(&ioc->fc_setup_reset_work, mptfc_setup_reset, (void *)ioc); + INIT_WORK(&ioc->fc_rescan_work, mptfc_rescan_devices); + INIT_WORK(&ioc->fc_setup_reset_work, mptfc_setup_reset); spin_lock_irqsave(&ioc->FreeQlock, flags); diff --git a/drivers/message/fusion/mptlan.c b/drivers/message/fusion/mptlan.c index 314c3a27585d..b7c4407c5e3f 100644 --- a/drivers/message/fusion/mptlan.c +++ b/drivers/message/fusion/mptlan.c @@ -111,7 +111,8 @@ struct mpt_lan_priv { u32 total_received; struct net_device_stats stats; /* Per device statistics */ - struct work_struct post_buckets_task; + struct delayed_work post_buckets_task; + struct net_device *dev; unsigned long post_buckets_active; }; @@ -132,7 +133,7 @@ static int lan_reply (MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, static int mpt_lan_open(struct net_device *dev); static int mpt_lan_reset(struct net_device *dev); static int mpt_lan_close(struct net_device *dev); -static void mpt_lan_post_receive_buckets(void *dev_id); +static void mpt_lan_post_receive_buckets(struct mpt_lan_priv *priv); static void mpt_lan_wake_post_buckets_task(struct net_device *dev, int priority); static int mpt_lan_receive_post_turbo(struct net_device *dev, u32 tmsg); @@ -345,7 +346,7 @@ mpt_lan_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) priv->mpt_rxfidx[++priv->mpt_rxfidx_tail] = i; spin_unlock_irqrestore(&priv->rxfidx_lock, flags); } else { - mpt_lan_post_receive_buckets(dev); + mpt_lan_post_receive_buckets(priv); netif_wake_queue(dev); } @@ -441,7 +442,7 @@ mpt_lan_open(struct net_device *dev) dlprintk((KERN_INFO MYNAM "/lo: Finished initializing RcvCtl\n")); - mpt_lan_post_receive_buckets(dev); + mpt_lan_post_receive_buckets(priv); printk(KERN_INFO MYNAM ": %s/%s: interface up & active\n", IOC_AND_NETDEV_NAMES_s_s(dev)); @@ -854,7 +855,7 @@ mpt_lan_wake_post_buckets_task(struct net_device *dev, int priority) if (test_and_set_bit(0, &priv->post_buckets_active) == 0) { if (priority) { - schedule_work(&priv->post_buckets_task); + schedule_delayed_work(&priv->post_buckets_task, 0); } else { schedule_delayed_work(&priv->post_buckets_task, 1); dioprintk((KERN_INFO MYNAM ": post_buckets queued on " @@ -1188,10 +1189,9 @@ mpt_lan_receive_post_reply(struct net_device *dev, /* Simple SGE's only at the moment */ static void -mpt_lan_post_receive_buckets(void *dev_id) +mpt_lan_post_receive_buckets(struct mpt_lan_priv *priv) { - struct net_device *dev = dev_id; - struct mpt_lan_priv *priv = dev->priv; + struct net_device *dev = priv->dev; MPT_ADAPTER *mpt_dev = priv->mpt_dev; MPT_FRAME_HDR *mf; LANReceivePostRequest_t *pRecvReq; @@ -1335,6 +1335,13 @@ out: clear_bit(0, &priv->post_buckets_active); } +static void +mpt_lan_post_receive_buckets_work(struct work_struct *work) +{ + mpt_lan_post_receive_buckets(container_of(work, struct mpt_lan_priv, + post_buckets_task.work)); +} + /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ static struct net_device * mpt_register_lan_device (MPT_ADAPTER *mpt_dev, int pnum) @@ -1350,11 +1357,13 @@ mpt_register_lan_device (MPT_ADAPTER *mpt_dev, int pnum) priv = netdev_priv(dev); + priv->dev = dev; priv->mpt_dev = mpt_dev; priv->pnum = pnum; - memset(&priv->post_buckets_task, 0, sizeof(struct work_struct)); - INIT_WORK(&priv->post_buckets_task, mpt_lan_post_receive_buckets, dev); + memset(&priv->post_buckets_task, 0, sizeof(priv->post_buckets_task)); + INIT_DELAYED_WORK(&priv->post_buckets_task, + mpt_lan_post_receive_buckets_work); priv->post_buckets_active = 0; dlprintk((KERN_INFO MYNAM "@%d: bucketlen = %d\n", diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c index b752a479f6db..4f0c530e47b0 100644 --- a/drivers/message/fusion/mptsas.c +++ b/drivers/message/fusion/mptsas.c @@ -2006,9 +2006,10 @@ __mptsas_discovery_work(MPT_ADAPTER *ioc) *(Mutex LOCKED) */ static void -mptsas_discovery_work(void * arg) +mptsas_discovery_work(struct work_struct *work) { - struct mptsas_discovery_event *ev = arg; + struct mptsas_discovery_event *ev = + container_of(work, struct mptsas_discovery_event, work); MPT_ADAPTER *ioc = ev->ioc; mutex_lock(&ioc->sas_discovery_mutex); @@ -2068,9 +2069,9 @@ mptsas_find_phyinfo_by_target(MPT_ADAPTER *ioc, u32 id) * Work queue thread to clear the persitency table */ static void -mptsas_persist_clear_table(void * arg) +mptsas_persist_clear_table(struct work_struct *work) { - MPT_ADAPTER *ioc = (MPT_ADAPTER *)arg; + MPT_ADAPTER *ioc = container_of(work, MPT_ADAPTER, sas_persist_task); mptbase_sas_persist_operation(ioc, MPI_SAS_OP_CLEAR_NOT_PRESENT); } @@ -2093,9 +2094,10 @@ mptsas_reprobe_target(struct scsi_target *starget, int uld_attach) * Work queue thread to handle SAS hotplug events */ static void -mptsas_hotplug_work(void *arg) +mptsas_hotplug_work(struct work_struct *work) { - struct mptsas_hotplug_event *ev = arg; + struct mptsas_hotplug_event *ev = + container_of(work, struct mptsas_hotplug_event, work); MPT_ADAPTER *ioc = ev->ioc; struct mptsas_phyinfo *phy_info; struct sas_rphy *rphy; @@ -2341,7 +2343,7 @@ mptsas_send_sas_event(MPT_ADAPTER *ioc, break; } - INIT_WORK(&ev->work, mptsas_hotplug_work, ev); + INIT_WORK(&ev->work, mptsas_hotplug_work); ev->ioc = ioc; ev->handle = le16_to_cpu(sas_event_data->DevHandle); ev->parent_handle = @@ -2366,7 +2368,7 @@ mptsas_send_sas_event(MPT_ADAPTER *ioc, * Persistent table is full. */ INIT_WORK(&ioc->sas_persist_task, - mptsas_persist_clear_table, (void *)ioc); + mptsas_persist_clear_table); schedule_work(&ioc->sas_persist_task); break; case MPI_EVENT_SAS_DEV_STAT_RC_SMART_DATA: @@ -2395,7 +2397,7 @@ mptsas_send_raid_event(MPT_ADAPTER *ioc, return; } - INIT_WORK(&ev->work, mptsas_hotplug_work, ev); + INIT_WORK(&ev->work, mptsas_hotplug_work); ev->ioc = ioc; ev->id = raid_event_data->VolumeID; ev->event_type = MPTSAS_IGNORE_EVENT; @@ -2474,7 +2476,7 @@ mptsas_send_discovery_event(MPT_ADAPTER *ioc, ev = kzalloc(sizeof(*ev), GFP_ATOMIC); if (!ev) return; - INIT_WORK(&ev->work, mptsas_discovery_work, ev); + INIT_WORK(&ev->work, mptsas_discovery_work); ev->ioc = ioc; schedule_work(&ev->work); }; @@ -2511,8 +2513,7 @@ mptsas_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *reply) break; case MPI_EVENT_PERSISTENT_TABLE_FULL: INIT_WORK(&ioc->sas_persist_task, - mptsas_persist_clear_table, - (void *)ioc); + mptsas_persist_clear_table); schedule_work(&ioc->sas_persist_task); break; case MPI_EVENT_SAS_DISCOVERY: diff --git a/drivers/message/fusion/mptspi.c b/drivers/message/fusion/mptspi.c index e4cc3dd5fc9f..f422c0d0621c 100644 --- a/drivers/message/fusion/mptspi.c +++ b/drivers/message/fusion/mptspi.c @@ -646,9 +646,10 @@ struct work_queue_wrapper { int disk; }; -static void mpt_work_wrapper(void *data) +static void mpt_work_wrapper(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct _MPT_SCSI_HOST *hd = wqw->hd; struct Scsi_Host *shost = hd->ioc->sh; struct scsi_device *sdev; @@ -695,7 +696,7 @@ static void mpt_dv_raid(struct _MPT_SCSI_HOST *hd, int disk) disk); return; } - INIT_WORK(&wqw->work, mpt_work_wrapper, wqw); + INIT_WORK(&wqw->work, mpt_work_wrapper); wqw->hd = hd; wqw->disk = disk; @@ -784,9 +785,10 @@ MODULE_DEVICE_TABLE(pci, mptspi_pci_table); * renegotiate for a given target */ static void -mptspi_dv_renegotiate_work(void *data) +mptspi_dv_renegotiate_work(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct _MPT_SCSI_HOST *hd = wqw->hd; struct scsi_device *sdev; @@ -804,7 +806,7 @@ mptspi_dv_renegotiate(struct _MPT_SCSI_HOST *hd) if (!wqw) return; - INIT_WORK(&wqw->work, mptspi_dv_renegotiate_work, wqw); + INIT_WORK(&wqw->work, mptspi_dv_renegotiate_work); wqw->hd = hd; schedule_work(&wqw->work); diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index 64130227574f..7fc7399bd2ec 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -232,7 +232,7 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) break; } - INIT_WORK(&evt->work, (void (*)(void *))drv->event, evt); + INIT_WORK(&evt->work, drv->event); queue_work(drv->event_queue, &evt->work); return 1; } diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index a2350640384b..9e529d8dd5cb 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -371,8 +371,10 @@ static int i2o_exec_remove(struct device *dev) * new LCT and if the buffer for the LCT was to small sends a LCT NOTIFY * again, otherwise send LCT NOTIFY to get informed on next LCT change. */ -static void i2o_exec_lct_modified(struct i2o_exec_lct_notify_work *work) +static void i2o_exec_lct_modified(struct work_struct *_work) { + struct i2o_exec_lct_notify_work *work = + container_of(_work, struct i2o_exec_lct_notify_work, work); u32 change_ind = 0; struct i2o_controller *c = work->c; @@ -439,8 +441,7 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, work->c = c; - INIT_WORK(&work->work, (void (*)(void *))i2o_exec_lct_modified, - work); + INIT_WORK(&work->work, i2o_exec_lct_modified); queue_work(i2o_exec_driver.event_queue, &work->work); return 1; } @@ -460,13 +461,15 @@ static int i2o_exec_reply(struct i2o_controller *c, u32 m, /** * i2o_exec_event - Event handling function - * @evt: Event which occurs + * @work: Work item in occurring event * * Handles events send by the Executive device. At the moment does not do * anything useful. */ -static void i2o_exec_event(struct i2o_event *evt) +static void i2o_exec_event(struct work_struct *work) { + struct i2o_event *evt = container_of(work, struct i2o_event, work); + if (likely(evt->i2o_dev)) osm_debug("Event received from device: %d\n", evt->i2o_dev->lct_data.tid); diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index eaba81bf2eca..70ae00253321 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -419,16 +419,18 @@ static int i2o_block_prep_req_fn(struct request_queue *q, struct request *req) /** * i2o_block_delayed_request_fn - delayed request queue function - * delayed_request: the delayed request with the queue to start + * @work: the delayed request with the queue to start * * If the request queue is stopped for a disk, and there is no open * request, a new event is created, which calls this function to start * the queue after I2O_BLOCK_REQUEST_TIME. Otherwise the queue will never * be started again. */ -static void i2o_block_delayed_request_fn(void *delayed_request) +static void i2o_block_delayed_request_fn(struct work_struct *work) { - struct i2o_block_delayed_request *dreq = delayed_request; + struct i2o_block_delayed_request *dreq = + container_of(work, struct i2o_block_delayed_request, + work.work); struct request_queue *q = dreq->queue; unsigned long flags; @@ -538,8 +540,9 @@ static int i2o_block_reply(struct i2o_controller *c, u32 m, return 1; }; -static void i2o_block_event(struct i2o_event *evt) +static void i2o_block_event(struct work_struct *work) { + struct i2o_event *evt = container_of(work, struct i2o_event, work); osm_debug("event received\n"); kfree(evt); }; @@ -938,8 +941,8 @@ static void i2o_block_request_fn(struct request_queue *q) continue; dreq->queue = q; - INIT_WORK(&dreq->work, i2o_block_delayed_request_fn, - dreq); + INIT_DELAYED_WORK(&dreq->work, + i2o_block_delayed_request_fn); if (!queue_delayed_work(i2o_block_driver.event_queue, &dreq->work, diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index 4fdaa5bda412..d9fdc95b440d 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -96,7 +96,7 @@ struct i2o_block_request { /* I2O Block device delayed request */ struct i2o_block_delayed_request { - struct work_struct work; + struct delayed_work work; struct request_queue *queue; }; diff --git a/drivers/misc/tifm_7xx1.c b/drivers/misc/tifm_7xx1.c index 1ba8754e9383..2ab7add78f94 100644 --- a/drivers/misc/tifm_7xx1.c +++ b/drivers/misc/tifm_7xx1.c @@ -33,9 +33,10 @@ static void tifm_7xx1_eject(struct tifm_adapter *fm, struct tifm_dev *sock) spin_unlock_irqrestore(&fm->lock, flags); } -static void tifm_7xx1_remove_media(void *adapter) +static void tifm_7xx1_remove_media(struct work_struct *work) { - struct tifm_adapter *fm = adapter; + struct tifm_adapter *fm = + container_of(work, struct tifm_adapter, media_remover); unsigned long flags; int cnt; struct tifm_dev *sock; @@ -169,9 +170,10 @@ tifm_7xx1_sock_addr(char __iomem *base_addr, unsigned int sock_num) return base_addr + ((sock_num + 1) << 10); } -static void tifm_7xx1_insert_media(void *adapter) +static void tifm_7xx1_insert_media(struct work_struct *work) { - struct tifm_adapter *fm = adapter; + struct tifm_adapter *fm = + container_of(work, struct tifm_adapter, media_inserter); unsigned long flags; tifm_media_id media_id; char *card_name = "xx"; @@ -261,7 +263,7 @@ static int tifm_7xx1_suspend(struct pci_dev *dev, pm_message_t state) spin_unlock_irqrestore(&fm->lock, flags); flush_workqueue(fm->wq); - tifm_7xx1_remove_media(fm); + tifm_7xx1_remove_media(&fm->media_remover); pci_set_power_state(dev, PCI_D3hot); pci_disable_device(dev); @@ -328,8 +330,8 @@ static int tifm_7xx1_probe(struct pci_dev *dev, if (!fm->sockets) goto err_out_free; - INIT_WORK(&fm->media_inserter, tifm_7xx1_insert_media, fm); - INIT_WORK(&fm->media_remover, tifm_7xx1_remove_media, fm); + INIT_WORK(&fm->media_inserter, tifm_7xx1_insert_media); + INIT_WORK(&fm->media_remover, tifm_7xx1_remove_media); fm->eject = tifm_7xx1_eject; pci_set_drvdata(dev, fm); @@ -384,7 +386,7 @@ static void tifm_7xx1_remove(struct pci_dev *dev) flush_workqueue(fm->wq); - tifm_7xx1_remove_media(fm); + tifm_7xx1_remove_media(&fm->media_remover); writel(TIFM_IRQ_SETALL, fm->addr + FM_CLEAR_INTERRUPT_ENABLE); free_irq(dev->irq, fm); diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 766bc54406e5..21fd39e4a20f 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -1165,18 +1165,16 @@ static void mmc_setup(struct mmc_host *host) */ void mmc_detect_change(struct mmc_host *host, unsigned long delay) { - if (delay) - mmc_schedule_delayed_work(&host->detect, delay); - else - mmc_schedule_work(&host->detect); + mmc_schedule_delayed_work(&host->detect, delay); } EXPORT_SYMBOL(mmc_detect_change); -static void mmc_rescan(void *data) +static void mmc_rescan(struct work_struct *work) { - struct mmc_host *host = data; + struct mmc_host *host = + container_of(work, struct mmc_host, detect.work); struct list_head *l, *n; unsigned char power_mode; @@ -1259,7 +1257,7 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev) spin_lock_init(&host->lock); init_waitqueue_head(&host->wq); INIT_LIST_HEAD(&host->cards); - INIT_WORK(&host->detect, mmc_rescan, host); + INIT_DELAYED_WORK(&host->detect, mmc_rescan); /* * By default, hosts do not support SGIO or large requests. @@ -1357,7 +1355,7 @@ EXPORT_SYMBOL(mmc_suspend_host); */ int mmc_resume_host(struct mmc_host *host) { - mmc_rescan(host); + mmc_rescan(&host->detect.work); return 0; } diff --git a/drivers/mmc/mmc.h b/drivers/mmc/mmc.h index cd5e0ab3d84b..149affe0b686 100644 --- a/drivers/mmc/mmc.h +++ b/drivers/mmc/mmc.h @@ -20,6 +20,6 @@ void mmc_remove_host_sysfs(struct mmc_host *host); void mmc_free_host_sysfs(struct mmc_host *host); int mmc_schedule_work(struct work_struct *work); -int mmc_schedule_delayed_work(struct work_struct *work, unsigned long delay); +int mmc_schedule_delayed_work(struct delayed_work *work, unsigned long delay); void mmc_flush_scheduled_work(void); #endif diff --git a/drivers/mmc/mmc_sysfs.c b/drivers/mmc/mmc_sysfs.c index 10cc9734eaa0..fd9a5fc6db7b 100644 --- a/drivers/mmc/mmc_sysfs.c +++ b/drivers/mmc/mmc_sysfs.c @@ -320,18 +320,10 @@ void mmc_free_host_sysfs(struct mmc_host *host) static struct workqueue_struct *workqueue; -/* - * Internal function. Schedule work in the MMC work queue. - */ -int mmc_schedule_work(struct work_struct *work) -{ - return queue_work(workqueue, work); -} - /* * Internal function. Schedule delayed work in the MMC work queue. */ -int mmc_schedule_delayed_work(struct work_struct *work, unsigned long delay) +int mmc_schedule_delayed_work(struct delayed_work *work, unsigned long delay) { return queue_delayed_work(workqueue, work, delay); } diff --git a/drivers/mmc/tifm_sd.c b/drivers/mmc/tifm_sd.c index 0fdc55b08a6d..e846499a004c 100644 --- a/drivers/mmc/tifm_sd.c +++ b/drivers/mmc/tifm_sd.c @@ -99,7 +99,7 @@ struct tifm_sd { struct mmc_request *req; struct work_struct cmd_handler; - struct work_struct abort_handler; + struct delayed_work abort_handler; wait_queue_head_t can_eject; size_t written_blocks; @@ -496,9 +496,9 @@ err_out: mmc_request_done(mmc, mrq); } -static void tifm_sd_end_cmd(void *data) +static void tifm_sd_end_cmd(struct work_struct *work) { - struct tifm_sd *host = data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); struct mmc_request *mrq; @@ -608,9 +608,9 @@ err_out: mmc_request_done(mmc, mrq); } -static void tifm_sd_end_cmd_nodma(void *data) +static void tifm_sd_end_cmd_nodma(struct work_struct *work) { - struct tifm_sd *host = (struct tifm_sd*)data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); struct mmc_request *mrq; @@ -661,11 +661,14 @@ static void tifm_sd_end_cmd_nodma(void *data) mmc_request_done(mmc, mrq); } -static void tifm_sd_abort(void *data) +static void tifm_sd_abort(struct work_struct *work) { + struct tifm_sd *host = + container_of(work, struct tifm_sd, abort_handler.work); + printk(KERN_ERR DRIVER_NAME ": card failed to respond for a long period of time"); - tifm_eject(((struct tifm_sd*)data)->dev); + tifm_eject(host->dev); } static void tifm_sd_ios(struct mmc_host *mmc, struct mmc_ios *ios) @@ -762,9 +765,9 @@ static struct mmc_host_ops tifm_sd_ops = { .get_ro = tifm_sd_ro }; -static void tifm_sd_register_host(void *data) +static void tifm_sd_register_host(struct work_struct *work) { - struct tifm_sd *host = (struct tifm_sd*)data; + struct tifm_sd *host = container_of(work, struct tifm_sd, cmd_handler); struct tifm_dev *sock = host->dev; struct mmc_host *mmc = tifm_get_drvdata(sock); unsigned long flags; @@ -772,8 +775,7 @@ static void tifm_sd_register_host(void *data) spin_lock_irqsave(&sock->lock, flags); host->flags |= HOST_REG; PREPARE_WORK(&host->cmd_handler, - no_dma ? tifm_sd_end_cmd_nodma : tifm_sd_end_cmd, - data); + no_dma ? tifm_sd_end_cmd_nodma : tifm_sd_end_cmd); spin_unlock_irqrestore(&sock->lock, flags); dev_dbg(&sock->dev, "adding host\n"); mmc_add_host(mmc); @@ -799,8 +801,8 @@ static int tifm_sd_probe(struct tifm_dev *sock) host->dev = sock; host->clk_div = 61; init_waitqueue_head(&host->can_eject); - INIT_WORK(&host->cmd_handler, tifm_sd_register_host, host); - INIT_WORK(&host->abort_handler, tifm_sd_abort, host); + INIT_WORK(&host->cmd_handler, tifm_sd_register_host); + INIT_DELAYED_WORK(&host->abort_handler, tifm_sd_abort); tifm_set_drvdata(sock, mmc); sock->signal_irq = tifm_sd_signal_irq; diff --git a/drivers/net/8139too.c b/drivers/net/8139too.c index d02ed51abfcc..931028f672de 100644 --- a/drivers/net/8139too.c +++ b/drivers/net/8139too.c @@ -594,7 +594,7 @@ struct rtl8139_private { u32 rx_config; struct rtl_extra_stats xstats; - struct work_struct thread; + struct delayed_work thread; struct mii_if_info mii; unsigned int regs_len; @@ -636,8 +636,8 @@ static struct net_device_stats *rtl8139_get_stats (struct net_device *dev); static void rtl8139_set_rx_mode (struct net_device *dev); static void __set_rx_mode (struct net_device *dev); static void rtl8139_hw_start (struct net_device *dev); -static void rtl8139_thread (void *_data); -static void rtl8139_tx_timeout_task(void *_data); +static void rtl8139_thread (struct work_struct *work); +static void rtl8139_tx_timeout_task(struct work_struct *work); static const struct ethtool_ops rtl8139_ethtool_ops; /* write MMIO register, with flush */ @@ -1010,7 +1010,7 @@ static int __devinit rtl8139_init_one (struct pci_dev *pdev, (debug < 0 ? RTL8139_DEF_MSG_ENABLE : ((1 << debug) - 1)); spin_lock_init (&tp->lock); spin_lock_init (&tp->rx_lock); - INIT_WORK(&tp->thread, rtl8139_thread, dev); + INIT_DELAYED_WORK(&tp->thread, rtl8139_thread); tp->mii.dev = dev; tp->mii.mdio_read = mdio_read; tp->mii.mdio_write = mdio_write; @@ -1596,15 +1596,16 @@ static inline void rtl8139_thread_iter (struct net_device *dev, RTL_R8 (Config1)); } -static void rtl8139_thread (void *_data) +static void rtl8139_thread (struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8139_private *tp = netdev_priv(dev); + struct rtl8139_private *tp = + container_of(work, struct rtl8139_private, thread.work); + struct net_device *dev = tp->mii.dev; unsigned long thr_delay = next_tick; if (tp->watchdog_fired) { tp->watchdog_fired = 0; - rtl8139_tx_timeout_task(_data); + rtl8139_tx_timeout_task(work); } else if (rtnl_trylock()) { rtl8139_thread_iter (dev, tp, tp->mmio_addr); rtnl_unlock (); @@ -1646,10 +1647,11 @@ static inline void rtl8139_tx_clear (struct rtl8139_private *tp) /* XXX account for unsent Tx packets in tp->stats.tx_dropped */ } -static void rtl8139_tx_timeout_task (void *_data) +static void rtl8139_tx_timeout_task (struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8139_private *tp = netdev_priv(dev); + struct rtl8139_private *tp = + container_of(work, struct rtl8139_private, thread.work); + struct net_device *dev = tp->mii.dev; void __iomem *ioaddr = tp->mmio_addr; int i; u8 tmp8; @@ -1695,7 +1697,7 @@ static void rtl8139_tx_timeout (struct net_device *dev) struct rtl8139_private *tp = netdev_priv(dev); if (!tp->have_thread) { - INIT_WORK(&tp->thread, rtl8139_tx_timeout_task, dev); + INIT_DELAYED_WORK(&tp->thread, rtl8139_tx_timeout_task); schedule_delayed_work(&tp->thread, next_tick); } else tp->watchdog_fired = 1; diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 01b76d3aa42f..b12cc4596b8d 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -4339,9 +4339,9 @@ bnx2_open(struct net_device *dev) } static void -bnx2_reset_task(void *data) +bnx2_reset_task(struct work_struct *work) { - struct bnx2 *bp = data; + struct bnx2 *bp = container_of(work, struct bnx2, reset_task); if (!netif_running(bp->dev)) return; @@ -5630,7 +5630,7 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->pdev = pdev; spin_lock_init(&bp->phy_lock); - INIT_WORK(&bp->reset_task, bnx2_reset_task, bp); + INIT_WORK(&bp->reset_task, bnx2_reset_task); dev->base_addr = dev->mem_start = pci_resource_start(pdev, 0); mem_len = MB_GET_CID_ADDR(17); diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index 521c5b71023c..fe08f3845491 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -4066,9 +4066,9 @@ static int cas_alloc_rxds(struct cas *cp) return 0; } -static void cas_reset_task(void *data) +static void cas_reset_task(struct work_struct *work) { - struct cas *cp = (struct cas *) data; + struct cas *cp = container_of(work, struct cas, reset_task); #if 0 int pending = atomic_read(&cp->reset_task_pending); #else @@ -5006,7 +5006,7 @@ static int __devinit cas_init_one(struct pci_dev *pdev, atomic_set(&cp->reset_task_pending_spare, 0); atomic_set(&cp->reset_task_pending_mtu, 0); #endif - INIT_WORK(&cp->reset_task, cas_reset_task, cp); + INIT_WORK(&cp->reset_task, cas_reset_task); /* Default link parameters */ if (link_mode >= 0 && link_mode <= 6) diff --git a/drivers/net/chelsio/common.h b/drivers/net/chelsio/common.h index 5d9dd14427c5..8b1bedbce0d5 100644 --- a/drivers/net/chelsio/common.h +++ b/drivers/net/chelsio/common.h @@ -209,7 +209,7 @@ struct adapter { struct peespi *espi; struct port_info port[MAX_NPORTS]; - struct work_struct stats_update_task; + struct delayed_work stats_update_task; struct timer_list stats_update_timer; struct semaphore mib_mutex; diff --git a/drivers/net/chelsio/cxgb2.c b/drivers/net/chelsio/cxgb2.c index ad7ff9641a7e..f607cc6a276b 100644 --- a/drivers/net/chelsio/cxgb2.c +++ b/drivers/net/chelsio/cxgb2.c @@ -927,10 +927,11 @@ static void t1_netpoll(struct net_device *dev) * Periodic accumulation of MAC statistics. This is used only if the MAC * does not have any other way to prevent stats counter overflow. */ -static void mac_stats_task(void *data) +static void mac_stats_task(struct work_struct *work) { int i; - struct adapter *adapter = data; + struct adapter *adapter = + container_of(work, struct adapter, stats_update_task.work); for_each_port(adapter, i) { struct port_info *p = &adapter->port[i]; @@ -951,9 +952,10 @@ static void mac_stats_task(void *data) /* * Processes elmer0 external interrupts in process context. */ -static void ext_intr_task(void *data) +static void ext_intr_task(struct work_struct *work) { - struct adapter *adapter = data; + struct adapter *adapter = + container_of(work, struct adapter, ext_intr_handler_task); elmer0_ext_intr_handler(adapter); @@ -1087,9 +1089,9 @@ static int __devinit init_one(struct pci_dev *pdev, spin_lock_init(&adapter->async_lock); INIT_WORK(&adapter->ext_intr_handler_task, - ext_intr_task, adapter); - INIT_WORK(&adapter->stats_update_task, mac_stats_task, - adapter); + ext_intr_task); + INIT_DELAYED_WORK(&adapter->stats_update_task, + mac_stats_task); #ifdef work_struct init_timer(&adapter->stats_update_timer); adapter->stats_update_timer.function = mac_stats_timer; diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 19ab3441269c..e7737d02bb05 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2102,9 +2102,10 @@ static void e100_tx_timeout(struct net_device *netdev) schedule_work(&nic->tx_timeout_task); } -static void e100_tx_timeout_task(struct net_device *netdev) +static void e100_tx_timeout_task(struct work_struct *work) { - struct nic *nic = netdev_priv(netdev); + struct nic *nic = container_of(work, struct nic, tx_timeout_task); + struct net_device *netdev = nic->netdev; DPRINTK(TX_ERR, DEBUG, "scb.status=0x%02X\n", readb(&nic->csr->scb.status)); @@ -2637,8 +2638,7 @@ static int __devinit e100_probe(struct pci_dev *pdev, nic->blink_timer.function = e100_blink_led; nic->blink_timer.data = (unsigned long)nic; - INIT_WORK(&nic->tx_timeout_task, - (void (*)(void *))e100_tx_timeout_task, netdev); + INIT_WORK(&nic->tx_timeout_task, e100_tx_timeout_task); if((err = e100_alloc(nic))) { DPRINTK(PROBE, ERR, "Cannot alloc driver memory, aborting.\n"); diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 6ad696101418..83fa32f72398 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2224,11 +2224,12 @@ static int ehea_stop(struct net_device *dev) return ret; } -static void ehea_reset_port(void *data) +static void ehea_reset_port(struct work_struct *work) { int ret; - struct net_device *dev = data; - struct ehea_port *port = netdev_priv(dev); + struct ehea_port *port = + container_of(work, struct ehea_port, reset_task); + struct net_device *dev = port->netdev; port->resets++; down(&port->port_lock); @@ -2379,7 +2380,7 @@ static int ehea_setup_single_port(struct ehea_port *port, dev->tx_timeout = &ehea_tx_watchdog; dev->watchdog_timeo = EHEA_WATCH_DOG_TIMEOUT; - INIT_WORK(&port->reset_task, ehea_reset_port, dev); + INIT_WORK(&port->reset_task, ehea_reset_port); ehea_set_ethtool_ops(dev); diff --git a/drivers/net/hamradio/baycom_epp.c b/drivers/net/hamradio/baycom_epp.c index 1ed9cccd3c11..3c33d6f6a6a6 100644 --- a/drivers/net/hamradio/baycom_epp.c +++ b/drivers/net/hamradio/baycom_epp.c @@ -168,8 +168,9 @@ struct baycom_state { int magic; struct pardevice *pdev; + struct net_device *dev; unsigned int work_running; - struct work_struct run_work; + struct delayed_work run_work; unsigned int modem; unsigned int bitrate; unsigned char stat; @@ -659,16 +660,18 @@ static int receive(struct net_device *dev, int cnt) #define GETTICK(x) #endif /* __i386__ */ -static void epp_bh(struct net_device *dev) +static void epp_bh(struct work_struct *work) { + struct net_device *dev; struct baycom_state *bc; struct parport *pp; unsigned char stat; unsigned char tmp[2]; unsigned int time1 = 0, time2 = 0, time3 = 0; int cnt, cnt2; - - bc = netdev_priv(dev); + + bc = container_of(work, struct baycom_state, run_work.work); + dev = bc->dev; if (!bc->work_running) return; baycom_int_freq(bc); @@ -889,7 +892,7 @@ static int epp_open(struct net_device *dev) return -EBUSY; } dev->irq = /*pp->irq*/ 0; - INIT_WORK(&bc->run_work, (void *)(void *)epp_bh, dev); + INIT_DELAYED_WORK(&bc->run_work, epp_bh); bc->work_running = 1; bc->modem = EPP_CONVENTIONAL; if (eppconfig(bc)) @@ -1213,6 +1216,7 @@ static void __init baycom_epp_dev_setup(struct net_device *dev) /* * initialize part of the baycom_state struct */ + bc->dev = dev; bc->magic = BAYCOM_MAGIC; bc->cfg.fclk = 19666600; bc->cfg.bps = 9600; diff --git a/drivers/net/irda/mcs7780.c b/drivers/net/irda/mcs7780.c index b32c52ed19d7..f0c61f3b2a82 100644 --- a/drivers/net/irda/mcs7780.c +++ b/drivers/net/irda/mcs7780.c @@ -560,9 +560,9 @@ static inline int mcs_find_endpoints(struct mcs_cb *mcs, return ret; } -static void mcs_speed_work(void *arg) +static void mcs_speed_work(struct work_struct *work) { - struct mcs_cb *mcs = arg; + struct mcs_cb *mcs = container_of(work, struct mcs_cb, work); struct net_device *netdev = mcs->netdev; mcs_speed_change(mcs); @@ -927,7 +927,7 @@ static int mcs_probe(struct usb_interface *intf, irda_qos_bits_to_value(&mcs->qos); /* Speed change work initialisation*/ - INIT_WORK(&mcs->work, mcs_speed_work, mcs); + INIT_WORK(&mcs->work, mcs_speed_work); /* Override the network functions we need to use */ ndev->hard_start_xmit = mcs_hard_xmit; diff --git a/drivers/net/irda/sir-dev.h b/drivers/net/irda/sir-dev.h index 9fa294a546d6..2a57bc67ce35 100644 --- a/drivers/net/irda/sir-dev.h +++ b/drivers/net/irda/sir-dev.h @@ -22,7 +22,7 @@ struct sir_fsm { struct semaphore sem; - struct work_struct work; + struct delayed_work work; unsigned state, substate; int param; int result; diff --git a/drivers/net/irda/sir_dev.c b/drivers/net/irda/sir_dev.c index 3b5854d10c17..17b0c3ab6201 100644 --- a/drivers/net/irda/sir_dev.c +++ b/drivers/net/irda/sir_dev.c @@ -100,9 +100,9 @@ static int sirdev_tx_complete_fsm(struct sir_dev *dev) * Both must be unlocked/restarted on completion - but only on final exit. */ -static void sirdev_config_fsm(void *data) +static void sirdev_config_fsm(struct work_struct *work) { - struct sir_dev *dev = data; + struct sir_dev *dev = container_of(work, struct sir_dev, fsm.work.work); struct sir_fsm *fsm = &dev->fsm; int next_state; int ret = -1; @@ -309,8 +309,8 @@ int sirdev_schedule_request(struct sir_dev *dev, int initial_state, unsigned par fsm->param = param; fsm->result = 0; - INIT_WORK(&fsm->work, sirdev_config_fsm, dev); - queue_work(irda_sir_wq, &fsm->work); + INIT_DELAYED_WORK(&fsm->work, sirdev_config_fsm); + queue_delayed_work(irda_sir_wq, &fsm->work, 0); return 0; } diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c index 2284e2ce1692..d6f4f185bf37 100644 --- a/drivers/net/iseries_veth.c +++ b/drivers/net/iseries_veth.c @@ -166,7 +166,7 @@ struct veth_msg { struct veth_lpar_connection { HvLpIndex remote_lp; - struct work_struct statemachine_wq; + struct delayed_work statemachine_wq; struct veth_msg *msgs; int num_events; struct veth_cap_data local_caps; @@ -456,7 +456,7 @@ static struct kobj_type veth_port_ktype = { static inline void veth_kick_statemachine(struct veth_lpar_connection *cnx) { - schedule_work(&cnx->statemachine_wq); + schedule_delayed_work(&cnx->statemachine_wq, 0); } static void veth_take_cap(struct veth_lpar_connection *cnx, @@ -638,9 +638,11 @@ static int veth_process_caps(struct veth_lpar_connection *cnx) } /* FIXME: The gotos here are a bit dubious */ -static void veth_statemachine(void *p) +static void veth_statemachine(struct work_struct *work) { - struct veth_lpar_connection *cnx = (struct veth_lpar_connection *)p; + struct veth_lpar_connection *cnx = + container_of(work, struct veth_lpar_connection, + statemachine_wq.work); int rlp = cnx->remote_lp; int rc; @@ -827,7 +829,7 @@ static int veth_init_connection(u8 rlp) cnx->remote_lp = rlp; spin_lock_init(&cnx->lock); - INIT_WORK(&cnx->statemachine_wq, veth_statemachine, cnx); + INIT_DELAYED_WORK(&cnx->statemachine_wq, veth_statemachine); init_timer(&cnx->ack_timer); cnx->ack_timer.function = veth_timed_ack; diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index e09f575a3a38..d1ebb91ed278 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -106,7 +106,7 @@ static boolean_t ixgb_clean_rx_irq(struct ixgb_adapter *adapter); static void ixgb_alloc_rx_buffers(struct ixgb_adapter *adapter); void ixgb_set_ethtool_ops(struct net_device *netdev); static void ixgb_tx_timeout(struct net_device *dev); -static void ixgb_tx_timeout_task(struct net_device *dev); +static void ixgb_tx_timeout_task(struct work_struct *work); static void ixgb_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp); static void ixgb_vlan_rx_add_vid(struct net_device *netdev, uint16_t vid); @@ -489,8 +489,7 @@ ixgb_probe(struct pci_dev *pdev, adapter->watchdog_timer.function = &ixgb_watchdog; adapter->watchdog_timer.data = (unsigned long)adapter; - INIT_WORK(&adapter->tx_timeout_task, - (void (*)(void *))ixgb_tx_timeout_task, netdev); + INIT_WORK(&adapter->tx_timeout_task, ixgb_tx_timeout_task); strcpy(netdev->name, "eth%d"); if((err = register_netdev(netdev))) @@ -1493,9 +1492,10 @@ ixgb_tx_timeout(struct net_device *netdev) } static void -ixgb_tx_timeout_task(struct net_device *netdev) +ixgb_tx_timeout_task(struct work_struct *work) { - struct ixgb_adapter *adapter = netdev_priv(netdev); + struct ixgb_adapter *adapter = + container_of(work, struct ixgb_adapter, tx_timeout_task); adapter->tx_timeout_count++; ixgb_down(adapter, TRUE); diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 806081b59733..98703e086ee7 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -2615,9 +2615,10 @@ static u32 myri10ge_read_reboot(struct myri10ge_priv *mgp) * This watchdog is used to check whether the board has suffered * from a parity error and needs to be recovered. */ -static void myri10ge_watchdog(void *arg) +static void myri10ge_watchdog(struct work_struct *work) { - struct myri10ge_priv *mgp = arg; + struct myri10ge_priv *mgp = + container_of(work, struct myri10ge_priv, watchdog_work); u32 reboot; int status; u16 cmd, vendor; @@ -2887,7 +2888,7 @@ static int myri10ge_probe(struct pci_dev *pdev, const struct pci_device_id *ent) (unsigned long)mgp); SET_ETHTOOL_OPS(netdev, &myri10ge_ethtool_ops); - INIT_WORK(&mgp->watchdog_work, myri10ge_watchdog, mgp); + INIT_WORK(&mgp->watchdog_work, myri10ge_watchdog); status = register_netdev(netdev); if (status != 0) { dev_err(&pdev->dev, "register_netdev failed: %d\n", status); diff --git a/drivers/net/ns83820.c b/drivers/net/ns83820.c index b0127c71a5b6..312e0e331712 100644 --- a/drivers/net/ns83820.c +++ b/drivers/net/ns83820.c @@ -427,6 +427,7 @@ struct ns83820 { u8 __iomem *base; struct pci_dev *pci_dev; + struct net_device *ndev; #ifdef NS83820_VLAN_ACCEL_SUPPORT struct vlan_group *vlgrp; @@ -631,10 +632,10 @@ static void fastcall rx_refill_atomic(struct net_device *ndev) } /* REFILL */ -static inline void queue_refill(void *_dev) +static inline void queue_refill(struct work_struct *work) { - struct net_device *ndev = _dev; - struct ns83820 *dev = PRIV(ndev); + struct ns83820 *dev = container_of(work, struct ns83820, tq_refill); + struct net_device *ndev = dev->ndev; rx_refill(ndev, GFP_KERNEL); if (dev->rx_info.up) @@ -1841,6 +1842,7 @@ static int __devinit ns83820_init_one(struct pci_dev *pci_dev, const struct pci_ ndev = alloc_etherdev(sizeof(struct ns83820)); dev = PRIV(ndev); + dev->ndev = ndev; err = -ENOMEM; if (!dev) goto out; @@ -1853,7 +1855,7 @@ static int __devinit ns83820_init_one(struct pci_dev *pci_dev, const struct pci_ SET_MODULE_OWNER(ndev); SET_NETDEV_DEV(ndev, &pci_dev->dev); - INIT_WORK(&dev->tq_refill, queue_refill, ndev); + INIT_WORK(&dev->tq_refill, queue_refill); tasklet_init(&dev->rx_tasklet, rx_action, (unsigned long)ndev); err = pci_enable_device(pci_dev); diff --git a/drivers/net/pcmcia/xirc2ps_cs.c b/drivers/net/pcmcia/xirc2ps_cs.c index f3914f58d67f..5de8850f2323 100644 --- a/drivers/net/pcmcia/xirc2ps_cs.c +++ b/drivers/net/pcmcia/xirc2ps_cs.c @@ -332,6 +332,7 @@ static irqreturn_t xirc2ps_interrupt(int irq, void *dev_id); */ typedef struct local_info_t { + struct net_device *dev; struct pcmcia_device *p_dev; dev_node_t node; struct net_device_stats stats; @@ -353,7 +354,7 @@ typedef struct local_info_t { */ static int do_start_xmit(struct sk_buff *skb, struct net_device *dev); static void do_tx_timeout(struct net_device *dev); -static void xirc2ps_tx_timeout_task(void *data); +static void xirc2ps_tx_timeout_task(struct work_struct *work); static struct net_device_stats *do_get_stats(struct net_device *dev); static void set_addresses(struct net_device *dev); static void set_multicast_list(struct net_device *dev); @@ -567,6 +568,7 @@ xirc2ps_probe(struct pcmcia_device *link) if (!dev) return -ENOMEM; local = netdev_priv(dev); + local->dev = dev; local->p_dev = link; link->priv = dev; @@ -591,7 +593,7 @@ xirc2ps_probe(struct pcmcia_device *link) #ifdef HAVE_TX_TIMEOUT dev->tx_timeout = do_tx_timeout; dev->watchdog_timeo = TX_TIMEOUT; - INIT_WORK(&local->tx_timeout_task, xirc2ps_tx_timeout_task, dev); + INIT_WORK(&local->tx_timeout_task, xirc2ps_tx_timeout_task); #endif return xirc2ps_config(link); @@ -1344,9 +1346,11 @@ xirc2ps_interrupt(int irq, void *dev_id) /*====================================================================*/ static void -xirc2ps_tx_timeout_task(void *data) +xirc2ps_tx_timeout_task(struct work_struct *work) { - struct net_device *dev = data; + local_info_t *local = + container_of(work, local_info_t, tx_timeout_task); + struct net_device *dev = local->dev; /* reset the card */ do_reset(dev,1); dev->trans_start = jiffies; diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 3af9fcf76c81..a443976d5dcf 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -394,7 +394,7 @@ out_unlock: EXPORT_SYMBOL(phy_start_aneg); -static void phy_change(void *data); +static void phy_change(struct work_struct *work); static void phy_timer(unsigned long data); /* phy_start_machine: @@ -549,7 +549,7 @@ int phy_start_interrupts(struct phy_device *phydev) { int err = 0; - INIT_WORK(&phydev->phy_queue, phy_change, phydev); + INIT_WORK(&phydev->phy_queue, phy_change); if (request_irq(phydev->irq, phy_interrupt, IRQF_SHARED, @@ -585,10 +585,11 @@ EXPORT_SYMBOL(phy_stop_interrupts); /* Scheduled by the phy_interrupt/timer to handle PHY changes */ -static void phy_change(void *data) +static void phy_change(struct work_struct *work) { int err; - struct phy_device *phydev = data; + struct phy_device *phydev = + container_of(work, struct phy_device, phy_queue); err = phy_disable_interrupts(phydev); diff --git a/drivers/net/plip.c b/drivers/net/plip.c index 71afb274498f..6bb085f54437 100644 --- a/drivers/net/plip.c +++ b/drivers/net/plip.c @@ -138,9 +138,9 @@ static const unsigned int net_debug = NET_DEBUG; #define PLIP_NIBBLE_WAIT 3000 /* Bottom halves */ -static void plip_kick_bh(struct net_device *dev); -static void plip_bh(struct net_device *dev); -static void plip_timer_bh(struct net_device *dev); +static void plip_kick_bh(struct work_struct *work); +static void plip_bh(struct work_struct *work); +static void plip_timer_bh(struct work_struct *work); /* Interrupt handler */ static void plip_interrupt(int irq, void *dev_id); @@ -207,9 +207,10 @@ struct plip_local { struct net_local { struct net_device_stats enet_stats; + struct net_device *dev; struct work_struct immediate; - struct work_struct deferred; - struct work_struct timer; + struct delayed_work deferred; + struct delayed_work timer; struct plip_local snd_data; struct plip_local rcv_data; struct pardevice *pardev; @@ -306,11 +307,11 @@ plip_init_netdev(struct net_device *dev) nl->nibble = PLIP_NIBBLE_WAIT; /* Initialize task queue structures */ - INIT_WORK(&nl->immediate, (void (*)(void *))plip_bh, dev); - INIT_WORK(&nl->deferred, (void (*)(void *))plip_kick_bh, dev); + INIT_WORK(&nl->immediate, plip_bh); + INIT_DELAYED_WORK(&nl->deferred, plip_kick_bh); if (dev->irq == -1) - INIT_WORK(&nl->timer, (void (*)(void *))plip_timer_bh, dev); + INIT_DELAYED_WORK(&nl->timer, plip_timer_bh); spin_lock_init(&nl->lock); } @@ -319,9 +320,10 @@ plip_init_netdev(struct net_device *dev) This routine is kicked by do_timer(). Request `plip_bh' to be invoked. */ static void -plip_kick_bh(struct net_device *dev) +plip_kick_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = + container_of(work, struct net_local, deferred.work); if (nl->is_deferred) schedule_work(&nl->immediate); @@ -362,9 +364,9 @@ static const plip_func connection_state_table[] = /* Bottom half handler of PLIP. */ static void -plip_bh(struct net_device *dev) +plip_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = container_of(work, struct net_local, immediate); struct plip_local *snd = &nl->snd_data; struct plip_local *rcv = &nl->rcv_data; plip_func f; @@ -372,20 +374,21 @@ plip_bh(struct net_device *dev) nl->is_deferred = 0; f = connection_state_table[nl->connection]; - if ((r = (*f)(dev, nl, snd, rcv)) != OK - && (r = plip_bh_timeout_error(dev, nl, snd, rcv, r)) != OK) { + if ((r = (*f)(nl->dev, nl, snd, rcv)) != OK + && (r = plip_bh_timeout_error(nl->dev, nl, snd, rcv, r)) != OK) { nl->is_deferred = 1; schedule_delayed_work(&nl->deferred, 1); } } static void -plip_timer_bh(struct net_device *dev) +plip_timer_bh(struct work_struct *work) { - struct net_local *nl = netdev_priv(dev); + struct net_local *nl = + container_of(work, struct net_local, timer.work); if (!(atomic_read (&nl->kill_timer))) { - plip_interrupt (-1, dev); + plip_interrupt (-1, nl->dev); schedule_delayed_work(&nl->timer, 1); } @@ -1284,6 +1287,7 @@ static void plip_attach (struct parport *port) } nl = netdev_priv(dev); + nl->dev = dev; nl->pardev = parport_register_device(port, name, plip_preempt, plip_wakeup, plip_interrupt, 0, dev); diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c index ec640f6229ae..d79d141a601d 100644 --- a/drivers/net/qla3xxx.c +++ b/drivers/net/qla3xxx.c @@ -2008,7 +2008,7 @@ static irqreturn_t ql3xxx_isr(int irq, void *dev_id) "%s: Another function issued a reset to the " "chip. ISR value = %x.\n", ndev->name, value); } - queue_work(qdev->workqueue, &qdev->reset_work); + queue_delayed_work(qdev->workqueue, &qdev->reset_work, 0); spin_unlock(&qdev->adapter_lock); } else if (value & ISP_IMR_DISABLE_CMPL_INT) { ql_disable_interrupts(qdev); @@ -3182,11 +3182,13 @@ static void ql3xxx_tx_timeout(struct net_device *ndev) /* * Wake up the worker to process this event. */ - queue_work(qdev->workqueue, &qdev->tx_timeout_work); + queue_delayed_work(qdev->workqueue, &qdev->tx_timeout_work, 0); } -static void ql_reset_work(struct ql3_adapter *qdev) +static void ql_reset_work(struct work_struct *work) { + struct ql3_adapter *qdev = + container_of(work, struct ql3_adapter, reset_work.work); struct net_device *ndev = qdev->ndev; u32 value; struct ql_tx_buf_cb *tx_cb; @@ -3278,9 +3280,12 @@ static void ql_reset_work(struct ql3_adapter *qdev) } } -static void ql_tx_timeout_work(struct ql3_adapter *qdev) +static void ql_tx_timeout_work(struct work_struct *work) { - ql_cycle_adapter(qdev,QL_DO_RESET); + struct ql3_adapter *qdev = + container_of(work, struct ql3_adapter, tx_timeout_work.work); + + ql_cycle_adapter(qdev, QL_DO_RESET); } static void ql_get_board_info(struct ql3_adapter *qdev) @@ -3459,9 +3464,8 @@ static int __devinit ql3xxx_probe(struct pci_dev *pdev, netif_stop_queue(ndev); qdev->workqueue = create_singlethread_workqueue(ndev->name); - INIT_WORK(&qdev->reset_work, (void (*)(void *))ql_reset_work, qdev); - INIT_WORK(&qdev->tx_timeout_work, - (void (*)(void *))ql_tx_timeout_work, qdev); + INIT_DELAYED_WORK(&qdev->reset_work, ql_reset_work); + INIT_DELAYED_WORK(&qdev->tx_timeout_work, ql_tx_timeout_work); init_timer(&qdev->adapter_timer); qdev->adapter_timer.function = ql3xxx_timer; diff --git a/drivers/net/qla3xxx.h b/drivers/net/qla3xxx.h index 65da2c0bfda6..ea94de7fd071 100644 --- a/drivers/net/qla3xxx.h +++ b/drivers/net/qla3xxx.h @@ -1186,8 +1186,8 @@ struct ql3_adapter { u32 numPorts; struct net_device_stats stats; struct workqueue_struct *workqueue; - struct work_struct reset_work; - struct work_struct tx_timeout_work; + struct delayed_work reset_work; + struct delayed_work tx_timeout_work; u32 max_frame_size; }; diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 27f90b2139c0..1f9663a70823 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -424,6 +424,7 @@ struct ring_info { struct rtl8169_private { void __iomem *mmio_addr; /* memory map physical address */ struct pci_dev *pci_dev; /* Index of PCI device */ + struct net_device *dev; struct net_device_stats stats; /* statistics of net device */ spinlock_t lock; /* spin lock flag */ u32 msg_enable; @@ -455,7 +456,7 @@ struct rtl8169_private { void (*phy_reset_enable)(void __iomem *); unsigned int (*phy_reset_pending)(void __iomem *); unsigned int (*link_ok)(void __iomem *); - struct work_struct task; + struct delayed_work task; unsigned wol_enabled : 1; }; @@ -1492,6 +1493,7 @@ rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) SET_MODULE_OWNER(dev); SET_NETDEV_DEV(dev, &pdev->dev); tp = netdev_priv(dev); + tp->dev = dev; tp->msg_enable = netif_msg_init(debug.msg_enable, R8169_MSG_DEFAULT); /* enable device (incl. PCI PM wakeup and hotplug setup) */ @@ -1764,7 +1766,7 @@ static int rtl8169_open(struct net_device *dev) if (retval < 0) goto err_free_rx; - INIT_WORK(&tp->task, NULL, dev); + INIT_DELAYED_WORK(&tp->task, NULL); rtl8169_hw_start(dev); @@ -2087,11 +2089,11 @@ static void rtl8169_tx_clear(struct rtl8169_private *tp) tp->cur_tx = tp->dirty_tx = 0; } -static void rtl8169_schedule_work(struct net_device *dev, void (*task)(void *)) +static void rtl8169_schedule_work(struct net_device *dev, work_func_t task) { struct rtl8169_private *tp = netdev_priv(dev); - PREPARE_WORK(&tp->task, task, dev); + PREPARE_DELAYED_WORK(&tp->task, task); schedule_delayed_work(&tp->task, 4); } @@ -2110,9 +2112,11 @@ static void rtl8169_wait_for_quiescence(struct net_device *dev) netif_poll_enable(dev); } -static void rtl8169_reinit_task(void *_data) +static void rtl8169_reinit_task(struct work_struct *work) { - struct net_device *dev = _data; + struct rtl8169_private *tp = + container_of(work, struct rtl8169_private, task.work); + struct net_device *dev = tp->dev; int ret; if (netif_running(dev)) { @@ -2135,10 +2139,11 @@ static void rtl8169_reinit_task(void *_data) } } -static void rtl8169_reset_task(void *_data) +static void rtl8169_reset_task(struct work_struct *work) { - struct net_device *dev = _data; - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = + container_of(work, struct rtl8169_private, task.work); + struct net_device *dev = tp->dev; if (!netif_running(dev)) return; diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 33569ec9dbfc..250cdbeefdfd 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -5872,9 +5872,9 @@ static void s2io_tasklet(unsigned long dev_addr) * Description: Sets the link status for the adapter */ -static void s2io_set_link(unsigned long data) +static void s2io_set_link(struct work_struct *work) { - nic_t *nic = (nic_t *) data; + nic_t *nic = container_of(work, nic_t, set_link_task); struct net_device *dev = nic->dev; XENA_dev_config_t __iomem *bar0 = nic->bar0; register u64 val64; @@ -6379,10 +6379,10 @@ static int s2io_card_up(nic_t * sp) * spin lock. */ -static void s2io_restart_nic(unsigned long data) +static void s2io_restart_nic(struct work_struct *work) { - struct net_device *dev = (struct net_device *) data; - nic_t *sp = dev->priv; + nic_t *sp = container_of(work, nic_t, rst_timer_task); + struct net_device *dev = sp->dev; s2io_card_down(sp); if (s2io_card_up(sp)) { @@ -6992,10 +6992,8 @@ s2io_init_nic(struct pci_dev *pdev, const struct pci_device_id *pre) dev->tx_timeout = &s2io_tx_watchdog; dev->watchdog_timeo = WATCH_DOG_TIMEOUT; - INIT_WORK(&sp->rst_timer_task, - (void (*)(void *)) s2io_restart_nic, dev); - INIT_WORK(&sp->set_link_task, - (void (*)(void *)) s2io_set_link, sp); + INIT_WORK(&sp->rst_timer_task, s2io_restart_nic); + INIT_WORK(&sp->set_link_task, s2io_set_link); pci_save_state(sp->pdev); diff --git a/drivers/net/s2io.h b/drivers/net/s2io.h index 12b719f4d00f..3b0bafd273c8 100644 --- a/drivers/net/s2io.h +++ b/drivers/net/s2io.h @@ -1000,7 +1000,7 @@ s2io_msix_fifo_handle(int irq, void *dev_id); static irqreturn_t s2io_isr(int irq, void *dev_id); static int verify_xena_quiescence(nic_t *sp, u64 val64, int flag); static const struct ethtool_ops netdev_ethtool_ops; -static void s2io_set_link(unsigned long data); +static void s2io_set_link(struct work_struct *work); static int s2io_set_swapper(nic_t * sp); static void s2io_card_down(nic_t *nic); static int s2io_card_up(nic_t *nic); diff --git a/drivers/net/sis190.c b/drivers/net/sis190.c index aaba458584fb..b70ed79d4121 100644 --- a/drivers/net/sis190.c +++ b/drivers/net/sis190.c @@ -280,6 +280,7 @@ enum sis190_feature { struct sis190_private { void __iomem *mmio_addr; struct pci_dev *pci_dev; + struct net_device *dev; struct net_device_stats stats; spinlock_t lock; u32 rx_buf_sz; @@ -897,10 +898,11 @@ static void sis190_hw_start(struct net_device *dev) netif_start_queue(dev); } -static void sis190_phy_task(void * data) +static void sis190_phy_task(struct work_struct *work) { - struct net_device *dev = data; - struct sis190_private *tp = netdev_priv(dev); + struct sis190_private *tp = + container_of(work, struct sis190_private, phy_task); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->mmio_addr; int phy_id = tp->mii_if.phy_id; u16 val; @@ -1047,7 +1049,7 @@ static int sis190_open(struct net_device *dev) if (rc < 0) goto err_free_rx_1; - INIT_WORK(&tp->phy_task, sis190_phy_task, dev); + INIT_WORK(&tp->phy_task, sis190_phy_task); sis190_request_timer(dev); @@ -1436,6 +1438,7 @@ static struct net_device * __devinit sis190_init_board(struct pci_dev *pdev) SET_NETDEV_DEV(dev, &pdev->dev); tp = netdev_priv(dev); + tp->dev = dev; tp->msg_enable = netif_msg_init(debug.msg_enable, SIS190_MSG_DEFAULT); rc = pci_enable_device(pdev); @@ -1798,7 +1801,7 @@ static int __devinit sis190_init_one(struct pci_dev *pdev, sis190_init_rxfilter(dev); - INIT_WORK(&tp->phy_task, sis190_phy_task, dev); + INIT_WORK(&tp->phy_task, sis190_phy_task); dev->open = sis190_open; dev->stop = sis190_close; diff --git a/drivers/net/skge.c b/drivers/net/skge.c index b2949035f66a..3b67614372a7 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -1327,10 +1327,11 @@ static void xm_check_link(struct net_device *dev) * Since internal PHY is wired to a level triggered pin, can't * get an interrupt when carrier is detected. */ -static void xm_link_timer(void *arg) +static void xm_link_timer(struct work_struct *work) { - struct net_device *dev = arg; - struct skge_port *skge = netdev_priv(arg); + struct skge_port *skge = + container_of(work, struct skge_port, link_thread.work); + struct net_device *dev = skge->netdev; struct skge_hw *hw = skge->hw; int port = skge->port; @@ -3072,9 +3073,9 @@ static void skge_error_irq(struct skge_hw *hw) * because accessing phy registers requires spin wait which might * cause excess interrupt latency. */ -static void skge_extirq(void *arg) +static void skge_extirq(struct work_struct *work) { - struct skge_hw *hw = arg; + struct skge_hw *hw = container_of(work, struct skge_hw, phy_work); int port; mutex_lock(&hw->phy_mutex); @@ -3456,7 +3457,7 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port, skge->port = port; /* Only used for Genesis XMAC */ - INIT_WORK(&skge->link_thread, xm_link_timer, dev); + INIT_DELAYED_WORK(&skge->link_thread, xm_link_timer); if (hw->chip_id != CHIP_ID_GENESIS) { dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG; @@ -3543,7 +3544,7 @@ static int __devinit skge_probe(struct pci_dev *pdev, hw->pdev = pdev; mutex_init(&hw->phy_mutex); - INIT_WORK(&hw->phy_work, skge_extirq, hw); + INIT_WORK(&hw->phy_work, skge_extirq); spin_lock_init(&hw->hw_lock); hw->regs = ioremap_nocache(pci_resource_start(pdev, 0), 0x4000); diff --git a/drivers/net/skge.h b/drivers/net/skge.h index 537c0aaa1db8..23e5275d920c 100644 --- a/drivers/net/skge.h +++ b/drivers/net/skge.h @@ -2456,7 +2456,7 @@ struct skge_port { struct net_device_stats net_stats; - struct work_struct link_thread; + struct delayed_work link_thread; enum pause_control flow_control; enum pause_status flow_status; u8 rx_csum; diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index 418138dd6c68..f88fcac0e46a 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1945,10 +1945,11 @@ spider_net_stop(struct net_device *netdev) * called as task when tx hangs, resets interface (if interface is up) */ static void -spider_net_tx_timeout_task(void *data) +spider_net_tx_timeout_task(struct work_struct *work) { - struct net_device *netdev = data; - struct spider_net_card *card = netdev_priv(netdev); + struct spider_net_card *card = + container_of(work, struct spider_net_card, tx_timeout_task); + struct net_device *netdev = card->netdev; if (!(netdev->flags & IFF_UP)) goto out; @@ -2122,7 +2123,7 @@ spider_net_alloc_card(void) card = netdev_priv(netdev); card->netdev = netdev; card->msg_enable = SPIDER_NET_DEFAULT_MSG; - INIT_WORK(&card->tx_timeout_task, spider_net_tx_timeout_task, netdev); + INIT_WORK(&card->tx_timeout_task, spider_net_tx_timeout_task); init_waitqueue_head(&card->waitq); atomic_set(&card->tx_timeout_task_counter, 0); diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 253e96e7ad20..004d651681ad 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -2281,9 +2281,9 @@ static void gem_do_stop(struct net_device *dev, int wol) } } -static void gem_reset_task(void *data) +static void gem_reset_task(struct work_struct *work) { - struct gem *gp = (struct gem *) data; + struct gem *gp = container_of(work, struct gem, reset_task); mutex_lock(&gp->pm_mutex); @@ -3043,7 +3043,7 @@ static int __devinit gem_init_one(struct pci_dev *pdev, gp->link_timer.function = gem_link_timer; gp->link_timer.data = (unsigned long) gp; - INIT_WORK(&gp->reset_task, gem_reset_task, gp); + INIT_WORK(&gp->reset_task, gem_reset_task); gp->lstate = link_down; gp->timer_ticks = 0; diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index c20bb998e0e5..d9123c9adc1e 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -3654,9 +3654,9 @@ static void tg3_poll_controller(struct net_device *dev) } #endif -static void tg3_reset_task(void *_data) +static void tg3_reset_task(struct work_struct *work) { - struct tg3 *tp = _data; + struct tg3 *tp = container_of(work, struct tg3, reset_task); unsigned int restart_timer; tg3_full_lock(tp, 0); @@ -11734,7 +11734,7 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, #endif spin_lock_init(&tp->lock); spin_lock_init(&tp->indirect_lock); - INIT_WORK(&tp->reset_task, tg3_reset_task, tp); + INIT_WORK(&tp->reset_task, tg3_reset_task); tp->regs = ioremap_nocache(tg3reg_base, tg3reg_len); if (tp->regs == 0UL) { diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c index e14f5a00f65a..f85f00251123 100644 --- a/drivers/net/tlan.c +++ b/drivers/net/tlan.c @@ -296,6 +296,7 @@ static void TLan_SetMulticastList( struct net_device *); static int TLan_ioctl( struct net_device *dev, struct ifreq *rq, int cmd); static int TLan_probe1( struct pci_dev *pdev, long ioaddr, int irq, int rev, const struct pci_device_id *ent); static void TLan_tx_timeout( struct net_device *dev); +static void TLan_tx_timeout_work(struct work_struct *work); static int tlan_init_one( struct pci_dev *pdev, const struct pci_device_id *ent); static u32 TLan_HandleInvalid( struct net_device *, u16 ); @@ -562,6 +563,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev, priv = netdev_priv(dev); priv->pciDev = pdev; + priv->dev = dev; /* Is this a PCI device? */ if (pdev) { @@ -634,7 +636,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev, /* This will be used when we get an adapter error from * within our irq handler */ - INIT_WORK(&priv->tlan_tqueue, (void *)(void*)TLan_tx_timeout, dev); + INIT_WORK(&priv->tlan_tqueue, TLan_tx_timeout_work); spin_lock_init(&priv->lock); @@ -1040,6 +1042,25 @@ static void TLan_tx_timeout(struct net_device *dev) } + /*************************************************************** + * TLan_tx_timeout_work + * + * Returns: nothing + * + * Params: + * work work item of device which timed out + * + **************************************************************/ + +static void TLan_tx_timeout_work(struct work_struct *work) +{ + TLanPrivateInfo *priv = + container_of(work, TLanPrivateInfo, tlan_tqueue); + + TLan_tx_timeout(priv->dev); +} + + /*************************************************************** * TLan_StartTx diff --git a/drivers/net/tlan.h b/drivers/net/tlan.h index a44e2f2ef62a..41ce0b665937 100644 --- a/drivers/net/tlan.h +++ b/drivers/net/tlan.h @@ -170,6 +170,7 @@ typedef u8 TLanBuffer[TLAN_MAX_FRAME_SIZE]; typedef struct tlan_private_tag { struct net_device *nextDevice; struct pci_dev *pciDev; + struct net_device *dev; void *dmaStorage; dma_addr_t dmaStorageDMA; unsigned int dmaSize; diff --git a/drivers/net/tulip/21142.c b/drivers/net/tulip/21142.c index fa3a2bb105ad..942b839ccc5b 100644 --- a/drivers/net/tulip/21142.c +++ b/drivers/net/tulip/21142.c @@ -26,10 +26,11 @@ static u16 t21142_csr15[] = { 0x0008, 0x0006, 0x000E, 0x0008, 0x0008, }; /* Handle the 21143 uniquely: do autoselect with NWay, not the EEPROM list of available transceivers. */ -void t21142_media_task(void *data) +void t21142_media_task(struct work_struct *work) { - struct net_device *dev = data; - struct tulip_private *tp = netdev_priv(dev); + struct tulip_private *tp = + container_of(work, struct tulip_private, media_work); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->base_addr; int csr12 = ioread32(ioaddr + CSR12); int next_tick = 60*HZ; diff --git a/drivers/net/tulip/timer.c b/drivers/net/tulip/timer.c index 066e5d6bcbd8..df326fe1cc8f 100644 --- a/drivers/net/tulip/timer.c +++ b/drivers/net/tulip/timer.c @@ -18,10 +18,11 @@ #include "tulip.h" -void tulip_media_task(void *data) +void tulip_media_task(struct work_struct *work) { - struct net_device *dev = data; - struct tulip_private *tp = netdev_priv(dev); + struct tulip_private *tp = + container_of(work, struct tulip_private, media_work); + struct net_device *dev = tp->dev; void __iomem *ioaddr = tp->base_addr; u32 csr12 = ioread32(ioaddr + CSR12); int next_tick = 2*HZ; diff --git a/drivers/net/tulip/tulip.h b/drivers/net/tulip/tulip.h index ad107f45c7b1..25f25da76917 100644 --- a/drivers/net/tulip/tulip.h +++ b/drivers/net/tulip/tulip.h @@ -44,7 +44,7 @@ struct tulip_chip_table { int valid_intrs; /* CSR7 interrupt enable settings */ int flags; void (*media_timer) (unsigned long); - void (*media_task) (void *); + work_func_t media_task; }; @@ -392,6 +392,7 @@ struct tulip_private { int csr12_shadow; int pad0; /* Used for 8-byte alignment */ struct work_struct media_work; + struct net_device *dev; }; @@ -406,7 +407,7 @@ struct eeprom_fixup { /* 21142.c */ extern u16 t21142_csr14[]; -void t21142_media_task(void *data); +void t21142_media_task(struct work_struct *work); void t21142_start_nway(struct net_device *dev); void t21142_lnk_change(struct net_device *dev, int csr5); @@ -444,7 +445,7 @@ void pnic_lnk_change(struct net_device *dev, int csr5); void pnic_timer(unsigned long data); /* timer.c */ -void tulip_media_task(void *data); +void tulip_media_task(struct work_struct *work); void mxic_timer(unsigned long data); void comet_timer(unsigned long data); diff --git a/drivers/net/tulip/tulip_core.c b/drivers/net/tulip/tulip_core.c index 0aee618f883c..5a35354aa523 100644 --- a/drivers/net/tulip/tulip_core.c +++ b/drivers/net/tulip/tulip_core.c @@ -1367,6 +1367,7 @@ static int __devinit tulip_init_one (struct pci_dev *pdev, * it is zeroed and aligned in alloc_etherdev */ tp = netdev_priv(dev); + tp->dev = dev; tp->rx_ring = pci_alloc_consistent(pdev, sizeof(struct tulip_rx_desc) * RX_RING_SIZE + @@ -1389,7 +1390,7 @@ static int __devinit tulip_init_one (struct pci_dev *pdev, tp->timer.data = (unsigned long)dev; tp->timer.function = tulip_tbl[tp->chip_id].media_timer; - INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task, dev); + INIT_WORK(&tp->media_work, tulip_tbl[tp->chip_id].media_task); dev->base_addr = (unsigned long)ioaddr; diff --git a/drivers/net/wan/pc300_tty.c b/drivers/net/wan/pc300_tty.c index 931cbdf6d791..b2a23aed4428 100644 --- a/drivers/net/wan/pc300_tty.c +++ b/drivers/net/wan/pc300_tty.c @@ -125,8 +125,8 @@ static int cpc_tty_write_room(struct tty_struct *tty); static int cpc_tty_chars_in_buffer(struct tty_struct *tty); static void cpc_tty_flush_buffer(struct tty_struct *tty); static void cpc_tty_hangup(struct tty_struct *tty); -static void cpc_tty_rx_work(void *data); -static void cpc_tty_tx_work(void *data); +static void cpc_tty_rx_work(struct work_struct *work); +static void cpc_tty_tx_work(struct work_struct *work); static int cpc_tty_send_to_card(pc300dev_t *dev,void *buf, int len); static void cpc_tty_trace(pc300dev_t *dev, char* buf, int len, char rxtx); static void cpc_tty_signal_off(pc300dev_t *pc300dev, unsigned char); @@ -261,8 +261,8 @@ void cpc_tty_init(pc300dev_t *pc300dev) cpc_tty->tty_minor = port + CPC_TTY_MINOR_START; cpc_tty->pc300dev = pc300dev; - INIT_WORK(&cpc_tty->tty_tx_work, cpc_tty_tx_work, (void *)cpc_tty); - INIT_WORK(&cpc_tty->tty_rx_work, cpc_tty_rx_work, (void *)port); + INIT_WORK(&cpc_tty->tty_tx_work, cpc_tty_tx_work); + INIT_WORK(&cpc_tty->tty_rx_work, cpc_tty_rx_work); cpc_tty->buf_rx.first = cpc_tty->buf_rx.last = NULL; @@ -659,21 +659,23 @@ static void cpc_tty_hangup(struct tty_struct *tty) * o call the line disc. read * o free memory */ -static void cpc_tty_rx_work(void * data) +static void cpc_tty_rx_work(struct work_struct *work) { + st_cpc_tty_area *cpc_tty; unsigned long port; int i, j; - st_cpc_tty_area *cpc_tty; volatile st_cpc_rx_buf *buf; char flags=0,flg_rx=1; struct tty_ldisc *ld; if (cpc_tty_cnt == 0) return; - for (i=0; (i < 4) && flg_rx ; i++) { flg_rx = 0; - port = (unsigned long)data; + + cpc_tty = container_of(work, st_cpc_tty_area, tty_rx_work); + port = cpc_tty - cpc_tty_area; + for (j=0; j < CPC_TTY_NPORTS; j++) { cpc_tty = &cpc_tty_area[port]; @@ -882,9 +884,10 @@ void cpc_tty_receive(pc300dev_t *pc300dev) * o if need call line discipline wakeup * o call wake_up_interruptible */ -static void cpc_tty_tx_work(void *data) +static void cpc_tty_tx_work(struct work_struct *work) { - st_cpc_tty_area *cpc_tty = (st_cpc_tty_area *) data; + st_cpc_tty_area *cpc_tty = + container_of(work, st_cpc_tty_area, tty_tx_work); struct tty_struct *tty; CPC_TTY_DBG("%s: cpc_tty_tx_work init\n",cpc_tty->name); diff --git a/drivers/net/wireless/bcm43xx/bcm43xx.h b/drivers/net/wireless/bcm43xx/bcm43xx.h index d6a8bf09878e..fbc0c087f53c 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx.h +++ b/drivers/net/wireless/bcm43xx/bcm43xx.h @@ -787,7 +787,7 @@ struct bcm43xx_private { struct tasklet_struct isr_tasklet; /* Periodic tasks */ - struct work_struct periodic_work; + struct delayed_work periodic_work; unsigned int periodic_state; struct work_struct restart_work; diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c index a1b783813d8e..728a9b789fdf 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c @@ -3177,9 +3177,10 @@ static int estimate_periodic_work_badness(unsigned int state) return badness; } -static void bcm43xx_periodic_work_handler(void *d) +static void bcm43xx_periodic_work_handler(struct work_struct *work) { - struct bcm43xx_private *bcm = d; + struct bcm43xx_private *bcm = + container_of(work, struct bcm43xx_private, periodic_work.work); struct net_device *net_dev = bcm->net_dev; unsigned long flags; u32 savedirqs = 0; @@ -3242,11 +3243,11 @@ void bcm43xx_periodic_tasks_delete(struct bcm43xx_private *bcm) void bcm43xx_periodic_tasks_setup(struct bcm43xx_private *bcm) { - struct work_struct *work = &(bcm->periodic_work); + struct delayed_work *work = &bcm->periodic_work; assert(bcm43xx_status(bcm) == BCM43xx_STAT_INITIALIZED); - INIT_WORK(work, bcm43xx_periodic_work_handler, bcm); - schedule_work(work); + INIT_DELAYED_WORK(work, bcm43xx_periodic_work_handler); + schedule_delayed_work(work, 0); } static void bcm43xx_security_init(struct bcm43xx_private *bcm) @@ -3598,7 +3599,7 @@ static int bcm43xx_init_board(struct bcm43xx_private *bcm) bcm43xx_periodic_tasks_setup(bcm); /*FIXME: This should be handled by softmac instead. */ - schedule_work(&bcm->softmac->associnfo.work); + schedule_delayed_work(&bcm->softmac->associnfo.work, 0); out: mutex_unlock(&(bcm)->mutex); @@ -4149,9 +4150,10 @@ static void __devexit bcm43xx_remove_one(struct pci_dev *pdev) /* Hard-reset the chip. Do not call this directly. * Use bcm43xx_controller_restart() */ -static void bcm43xx_chip_reset(void *_bcm) +static void bcm43xx_chip_reset(struct work_struct *work) { - struct bcm43xx_private *bcm = _bcm; + struct bcm43xx_private *bcm = + container_of(work, struct bcm43xx_private, restart_work); struct bcm43xx_phyinfo *phy; int err = -ENODEV; @@ -4178,7 +4180,7 @@ void bcm43xx_controller_restart(struct bcm43xx_private *bcm, const char *reason) if (bcm43xx_status(bcm) != BCM43xx_STAT_INITIALIZED) return; printk(KERN_ERR PFX "Controller RESET (%s) ...\n", reason); - INIT_WORK(&bcm->restart_work, bcm43xx_chip_reset, bcm); + INIT_WORK(&bcm->restart_work, bcm43xx_chip_reset); schedule_work(&bcm->restart_work); } diff --git a/drivers/net/wireless/hostap/hostap.h b/drivers/net/wireless/hostap/hostap.h index e663518bd570..e89c890d16fd 100644 --- a/drivers/net/wireless/hostap/hostap.h +++ b/drivers/net/wireless/hostap/hostap.h @@ -35,7 +35,7 @@ int hostap_80211_get_hdrlen(u16 fc); struct net_device_stats *hostap_get_stats(struct net_device *dev); void hostap_setup_dev(struct net_device *dev, local_info_t *local, int main_dev); -void hostap_set_multicast_list_queue(void *data); +void hostap_set_multicast_list_queue(struct work_struct *work); int hostap_set_hostapd(local_info_t *local, int val, int rtnl_locked); int hostap_set_hostapd_sta(local_info_t *local, int val, int rtnl_locked); void hostap_cleanup(local_info_t *local); diff --git a/drivers/net/wireless/hostap/hostap_ap.c b/drivers/net/wireless/hostap/hostap_ap.c index ba13125024cb..08bc57a4b895 100644 --- a/drivers/net/wireless/hostap/hostap_ap.c +++ b/drivers/net/wireless/hostap/hostap_ap.c @@ -49,10 +49,10 @@ MODULE_PARM_DESC(autom_ap_wds, "Add WDS connections to other APs " static struct sta_info* ap_get_sta(struct ap_data *ap, u8 *sta); static void hostap_event_expired_sta(struct net_device *dev, struct sta_info *sta); -static void handle_add_proc_queue(void *data); +static void handle_add_proc_queue(struct work_struct *work); #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT -static void handle_wds_oper_queue(void *data); +static void handle_wds_oper_queue(struct work_struct *work); static void prism2_send_mgmt(struct net_device *dev, u16 type_subtype, char *body, int body_len, u8 *addr, u16 tx_cb_idx); @@ -807,7 +807,7 @@ void hostap_init_data(local_info_t *local) INIT_LIST_HEAD(&ap->sta_list); /* Initialize task queue structure for AP management */ - INIT_WORK(&local->ap->add_sta_proc_queue, handle_add_proc_queue, ap); + INIT_WORK(&local->ap->add_sta_proc_queue, handle_add_proc_queue); ap->tx_callback_idx = hostap_tx_callback_register(local, hostap_ap_tx_cb, ap); @@ -815,7 +815,7 @@ void hostap_init_data(local_info_t *local) printk(KERN_WARNING "%s: failed to register TX callback for " "AP\n", local->dev->name); #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT - INIT_WORK(&local->ap->wds_oper_queue, handle_wds_oper_queue, local); + INIT_WORK(&local->ap->wds_oper_queue, handle_wds_oper_queue); ap->tx_callback_auth = hostap_tx_callback_register(local, hostap_ap_tx_cb_auth, ap); @@ -1062,9 +1062,10 @@ static int prism2_sta_proc_read(char *page, char **start, off_t off, } -static void handle_add_proc_queue(void *data) +static void handle_add_proc_queue(struct work_struct *work) { - struct ap_data *ap = (struct ap_data *) data; + struct ap_data *ap = container_of(work, struct ap_data, + add_sta_proc_queue); struct sta_info *sta; char name[20]; struct add_sta_proc_data *entry, *prev; @@ -1952,9 +1953,11 @@ static void handle_pspoll(local_info_t *local, #ifndef PRISM2_NO_KERNEL_IEEE80211_MGMT -static void handle_wds_oper_queue(void *data) +static void handle_wds_oper_queue(struct work_struct *work) { - local_info_t *local = data; + struct ap_data *ap = container_of(work, struct ap_data, + wds_oper_queue); + local_info_t *local = ap->local; struct wds_oper_data *entry, *prev; spin_lock_bh(&local->lock); diff --git a/drivers/net/wireless/hostap/hostap_hw.c b/drivers/net/wireless/hostap/hostap_hw.c index ed00ebb6e7f4..c19e68636a1c 100644 --- a/drivers/net/wireless/hostap/hostap_hw.c +++ b/drivers/net/wireless/hostap/hostap_hw.c @@ -1645,9 +1645,9 @@ static void prism2_schedule_reset(local_info_t *local) /* Called only as scheduled task after noticing card timeout in interrupt * context */ -static void handle_reset_queue(void *data) +static void handle_reset_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, reset_queue); printk(KERN_DEBUG "%s: scheduled card reset\n", local->dev->name); prism2_hw_reset(local->dev); @@ -2896,9 +2896,10 @@ static void hostap_passive_scan(unsigned long data) /* Called only as a scheduled task when communications quality values should * be updated. */ -static void handle_comms_qual_update(void *data) +static void handle_comms_qual_update(struct work_struct *work) { - local_info_t *local = data; + local_info_t *local = + container_of(work, local_info_t, comms_qual_update); prism2_update_comms_qual(local->dev); } @@ -3050,9 +3051,9 @@ static int prism2_set_tim(struct net_device *dev, int aid, int set) } -static void handle_set_tim_queue(void *data) +static void handle_set_tim_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, set_tim_queue); struct set_tim_data *entry; u16 val; @@ -3209,15 +3210,15 @@ prism2_init_local_data(struct prism2_helper_functions *funcs, int card_idx, local->scan_channel_mask = 0xffff; /* Initialize task queue structures */ - INIT_WORK(&local->reset_queue, handle_reset_queue, local); + INIT_WORK(&local->reset_queue, handle_reset_queue); INIT_WORK(&local->set_multicast_list_queue, - hostap_set_multicast_list_queue, local->dev); + hostap_set_multicast_list_queue); - INIT_WORK(&local->set_tim_queue, handle_set_tim_queue, local); + INIT_WORK(&local->set_tim_queue, handle_set_tim_queue); INIT_LIST_HEAD(&local->set_tim_list); spin_lock_init(&local->set_tim_lock); - INIT_WORK(&local->comms_qual_update, handle_comms_qual_update, local); + INIT_WORK(&local->comms_qual_update, handle_comms_qual_update); /* Initialize tasklets for handling hardware IRQ related operations * outside hw IRQ handler */ diff --git a/drivers/net/wireless/hostap/hostap_info.c b/drivers/net/wireless/hostap/hostap_info.c index 50f72d831cf4..5fd2b1ad7f5e 100644 --- a/drivers/net/wireless/hostap/hostap_info.c +++ b/drivers/net/wireless/hostap/hostap_info.c @@ -474,9 +474,9 @@ static void handle_info_queue_scanresults(local_info_t *local) /* Called only as scheduled task after receiving info frames (used to avoid * pending too much time in HW IRQ handler). */ -static void handle_info_queue(void *data) +static void handle_info_queue(struct work_struct *work) { - local_info_t *local = (local_info_t *) data; + local_info_t *local = container_of(work, local_info_t, info_queue); if (test_and_clear_bit(PRISM2_INFO_PENDING_LINKSTATUS, &local->pending_info)) @@ -493,7 +493,7 @@ void hostap_info_init(local_info_t *local) { skb_queue_head_init(&local->info_list); #ifndef PRISM2_NO_STATION_MODES - INIT_WORK(&local->info_queue, handle_info_queue, local); + INIT_WORK(&local->info_queue, handle_info_queue); #endif /* PRISM2_NO_STATION_MODES */ } diff --git a/drivers/net/wireless/hostap/hostap_main.c b/drivers/net/wireless/hostap/hostap_main.c index 53374fcba77e..0796be9d9e77 100644 --- a/drivers/net/wireless/hostap/hostap_main.c +++ b/drivers/net/wireless/hostap/hostap_main.c @@ -767,14 +767,14 @@ static int prism2_set_mac_address(struct net_device *dev, void *p) /* TODO: to be further implemented as soon as Prism2 fully supports * GroupAddresses and correct documentation is available */ -void hostap_set_multicast_list_queue(void *data) +void hostap_set_multicast_list_queue(struct work_struct *work) { - struct net_device *dev = (struct net_device *) data; + local_info_t *local = + container_of(work, local_info_t, set_multicast_list_queue); + struct net_device *dev = local->dev; struct hostap_interface *iface; - local_info_t *local; iface = netdev_priv(dev); - local = iface->local; if (hostap_set_word(dev, HFA384X_RID_PROMISCUOUSMODE, local->is_promisc)) { printk(KERN_INFO "%s: %sabling promiscuous mode failed\n", diff --git a/drivers/net/wireless/ipw2100.c b/drivers/net/wireless/ipw2100.c index 4e4eaa2a99ca..0f554373a60d 100644 --- a/drivers/net/wireless/ipw2100.c +++ b/drivers/net/wireless/ipw2100.c @@ -316,7 +316,7 @@ static void ipw2100_release_firmware(struct ipw2100_priv *priv, struct ipw2100_fw *fw); static int ipw2100_ucode_download(struct ipw2100_priv *priv, struct ipw2100_fw *fw); -static void ipw2100_wx_event_work(struct ipw2100_priv *priv); +static void ipw2100_wx_event_work(struct work_struct *work); static struct iw_statistics *ipw2100_wx_wireless_stats(struct net_device *dev); static struct iw_handler_def ipw2100_wx_handler_def; @@ -679,7 +679,8 @@ static void schedule_reset(struct ipw2100_priv *priv) queue_delayed_work(priv->workqueue, &priv->reset_work, priv->reset_backoff * HZ); else - queue_work(priv->workqueue, &priv->reset_work); + queue_delayed_work(priv->workqueue, &priv->reset_work, + 0); if (priv->reset_backoff < MAX_RESET_BACKOFF) priv->reset_backoff++; @@ -1873,8 +1874,10 @@ static void ipw2100_down(struct ipw2100_priv *priv) netif_stop_queue(priv->net_dev); } -static void ipw2100_reset_adapter(struct ipw2100_priv *priv) +static void ipw2100_reset_adapter(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, reset_work.work); unsigned long flags; union iwreq_data wrqu = { .ap_addr = { @@ -2071,9 +2074,9 @@ static void isr_indicate_association_lost(struct ipw2100_priv *priv, u32 status) return; if (priv->status & STATUS_SECURITY_UPDATED) - queue_work(priv->workqueue, &priv->security_work); + queue_delayed_work(priv->workqueue, &priv->security_work, 0); - queue_work(priv->workqueue, &priv->wx_event_work); + queue_delayed_work(priv->workqueue, &priv->wx_event_work, 0); } static void isr_indicate_rf_kill(struct ipw2100_priv *priv, u32 status) @@ -5524,8 +5527,11 @@ static int ipw2100_configure_security(struct ipw2100_priv *priv, int batch_mode) return err; } -static void ipw2100_security_work(struct ipw2100_priv *priv) +static void ipw2100_security_work(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, security_work.work); + /* If we happen to have reconnected before we get a chance to * process this, then update the security settings--which causes * a disassociation to occur */ @@ -5748,7 +5754,7 @@ static int ipw2100_set_address(struct net_device *dev, void *p) priv->reset_backoff = 0; mutex_unlock(&priv->action_mutex); - ipw2100_reset_adapter(priv); + ipw2100_reset_adapter(&priv->reset_work.work); return 0; done: @@ -5923,9 +5929,10 @@ static const struct ethtool_ops ipw2100_ethtool_ops = { .get_drvinfo = ipw_ethtool_get_drvinfo, }; -static void ipw2100_hang_check(void *adapter) +static void ipw2100_hang_check(struct work_struct *work) { - struct ipw2100_priv *priv = adapter; + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, hang_check.work); unsigned long flags; u32 rtc = 0xa5a5a5a5; u32 len = sizeof(rtc); @@ -5965,9 +5972,10 @@ static void ipw2100_hang_check(void *adapter) spin_unlock_irqrestore(&priv->low_lock, flags); } -static void ipw2100_rf_kill(void *adapter) +static void ipw2100_rf_kill(struct work_struct *work) { - struct ipw2100_priv *priv = adapter; + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, rf_kill.work); unsigned long flags; spin_lock_irqsave(&priv->low_lock, flags); @@ -6117,14 +6125,11 @@ static struct net_device *ipw2100_alloc_device(struct pci_dev *pci_dev, priv->workqueue = create_workqueue(DRV_NAME); - INIT_WORK(&priv->reset_work, - (void (*)(void *))ipw2100_reset_adapter, priv); - INIT_WORK(&priv->security_work, - (void (*)(void *))ipw2100_security_work, priv); - INIT_WORK(&priv->wx_event_work, - (void (*)(void *))ipw2100_wx_event_work, priv); - INIT_WORK(&priv->hang_check, ipw2100_hang_check, priv); - INIT_WORK(&priv->rf_kill, ipw2100_rf_kill, priv); + INIT_DELAYED_WORK(&priv->reset_work, ipw2100_reset_adapter); + INIT_DELAYED_WORK(&priv->security_work, ipw2100_security_work); + INIT_DELAYED_WORK(&priv->wx_event_work, ipw2100_wx_event_work); + INIT_DELAYED_WORK(&priv->hang_check, ipw2100_hang_check); + INIT_DELAYED_WORK(&priv->rf_kill, ipw2100_rf_kill); tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long)) ipw2100_irq_tasklet, (unsigned long)priv); @@ -8290,8 +8295,10 @@ static struct iw_handler_def ipw2100_wx_handler_def = { .get_wireless_stats = ipw2100_wx_wireless_stats, }; -static void ipw2100_wx_event_work(struct ipw2100_priv *priv) +static void ipw2100_wx_event_work(struct work_struct *work) { + struct ipw2100_priv *priv = + container_of(work, struct ipw2100_priv, wx_event_work.work); union iwreq_data wrqu; int len = ETH_ALEN; diff --git a/drivers/net/wireless/ipw2100.h b/drivers/net/wireless/ipw2100.h index 55b7227198df..de7d384d38af 100644 --- a/drivers/net/wireless/ipw2100.h +++ b/drivers/net/wireless/ipw2100.h @@ -583,11 +583,11 @@ struct ipw2100_priv { struct tasklet_struct irq_tasklet; struct workqueue_struct *workqueue; - struct work_struct reset_work; - struct work_struct security_work; - struct work_struct wx_event_work; - struct work_struct hang_check; - struct work_struct rf_kill; + struct delayed_work reset_work; + struct delayed_work security_work; + struct delayed_work wx_event_work; + struct delayed_work hang_check; + struct delayed_work rf_kill; u32 interrupts; int tx_interrupts; diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c index 1f742814a01c..587a0918fa52 100644 --- a/drivers/net/wireless/ipw2200.c +++ b/drivers/net/wireless/ipw2200.c @@ -187,9 +187,9 @@ static struct ipw_rx_queue *ipw_rx_queue_alloc(struct ipw_priv *); static void ipw_rx_queue_free(struct ipw_priv *, struct ipw_rx_queue *); static void ipw_rx_queue_replenish(void *); static int ipw_up(struct ipw_priv *); -static void ipw_bg_up(void *); +static void ipw_bg_up(struct work_struct *work); static void ipw_down(struct ipw_priv *); -static void ipw_bg_down(void *); +static void ipw_bg_down(struct work_struct *work); static int ipw_config(struct ipw_priv *); static int init_supported_rates(struct ipw_priv *priv, struct ipw_supported_rates *prates); @@ -862,11 +862,12 @@ static void ipw_led_link_on(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_link_on(void *data) +static void ipw_bg_led_link_on(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_link_on.work); mutex_lock(&priv->mutex); - ipw_led_link_on(data); + ipw_led_link_on(priv); mutex_unlock(&priv->mutex); } @@ -906,11 +907,12 @@ static void ipw_led_link_off(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_link_off(void *data) +static void ipw_bg_led_link_off(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_link_off.work); mutex_lock(&priv->mutex); - ipw_led_link_off(data); + ipw_led_link_off(priv); mutex_unlock(&priv->mutex); } @@ -985,11 +987,12 @@ static void ipw_led_activity_off(struct ipw_priv *priv) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_led_activity_off(void *data) +static void ipw_bg_led_activity_off(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, led_act_off.work); mutex_lock(&priv->mutex); - ipw_led_activity_off(data); + ipw_led_activity_off(priv); mutex_unlock(&priv->mutex); } @@ -2228,11 +2231,12 @@ static void ipw_adapter_restart(void *adapter) } } -static void ipw_bg_adapter_restart(void *data) +static void ipw_bg_adapter_restart(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, adapter_restart); mutex_lock(&priv->mutex); - ipw_adapter_restart(data); + ipw_adapter_restart(priv); mutex_unlock(&priv->mutex); } @@ -2249,11 +2253,12 @@ static void ipw_scan_check(void *data) } } -static void ipw_bg_scan_check(void *data) +static void ipw_bg_scan_check(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, scan_check.work); mutex_lock(&priv->mutex); - ipw_scan_check(data); + ipw_scan_check(priv); mutex_unlock(&priv->mutex); } @@ -3831,17 +3836,19 @@ static int ipw_disassociate(void *data) return 1; } -static void ipw_bg_disassociate(void *data) +static void ipw_bg_disassociate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, disassociate); mutex_lock(&priv->mutex); - ipw_disassociate(data); + ipw_disassociate(priv); mutex_unlock(&priv->mutex); } -static void ipw_system_config(void *data) +static void ipw_system_config(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, system_config); #ifdef CONFIG_IPW2200_PROMISCUOUS if (priv->prom_net_dev && netif_running(priv->prom_net_dev)) { @@ -4208,11 +4215,12 @@ static void ipw_gather_stats(struct ipw_priv *priv) IPW_STATS_INTERVAL); } -static void ipw_bg_gather_stats(void *data) +static void ipw_bg_gather_stats(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, gather_stats.work); mutex_lock(&priv->mutex); - ipw_gather_stats(data); + ipw_gather_stats(priv); mutex_unlock(&priv->mutex); } @@ -4268,8 +4276,8 @@ static void ipw_handle_missed_beacon(struct ipw_priv *priv, if (!(priv->status & STATUS_ROAMING)) { priv->status |= STATUS_ROAMING; if (!(priv->status & STATUS_SCANNING)) - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); } return; } @@ -4607,8 +4615,8 @@ static void ipw_rx_notification(struct ipw_priv *priv, #ifdef CONFIG_IPW2200_MONITOR if (priv->ieee->iw_mode == IW_MODE_MONITOR) { priv->status |= STATUS_SCAN_FORCED; - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); break; } priv->status &= ~STATUS_SCAN_FORCED; @@ -4631,8 +4639,8 @@ static void ipw_rx_notification(struct ipw_priv *priv, /* Don't schedule if we aborted the scan */ priv->status &= ~STATUS_ROAMING; } else if (priv->status & STATUS_SCAN_PENDING) - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); else if (priv->config & CFG_BACKGROUND_SCAN && priv->status & STATUS_ASSOCIATED) queue_delayed_work(priv->workqueue, @@ -5055,11 +5063,12 @@ static void ipw_rx_queue_replenish(void *data) ipw_rx_queue_restock(priv); } -static void ipw_bg_rx_queue_replenish(void *data) +static void ipw_bg_rx_queue_replenish(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, rx_replenish); mutex_lock(&priv->mutex); - ipw_rx_queue_replenish(data); + ipw_rx_queue_replenish(priv); mutex_unlock(&priv->mutex); } @@ -5489,9 +5498,10 @@ static int ipw_find_adhoc_network(struct ipw_priv *priv, return 1; } -static void ipw_merge_adhoc_network(void *data) +static void ipw_merge_adhoc_network(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, merge_networks); struct ieee80211_network *network = NULL; struct ipw_network_match match = { .network = priv->assoc_network @@ -5948,11 +5958,12 @@ static void ipw_adhoc_check(void *data) priv->assoc_request.beacon_interval); } -static void ipw_bg_adhoc_check(void *data) +static void ipw_bg_adhoc_check(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, adhoc_check.work); mutex_lock(&priv->mutex); - ipw_adhoc_check(data); + ipw_adhoc_check(priv); mutex_unlock(&priv->mutex); } @@ -6299,19 +6310,26 @@ done: return err; } -static int ipw_request_passive_scan(struct ipw_priv *priv) { - return ipw_request_scan_helper(priv, IW_SCAN_TYPE_PASSIVE); +static void ipw_request_passive_scan(struct work_struct *work) +{ + struct ipw_priv *priv = + container_of(work, struct ipw_priv, request_passive_scan); + ipw_request_scan_helper(priv, IW_SCAN_TYPE_PASSIVE); } -static int ipw_request_scan(struct ipw_priv *priv) { - return ipw_request_scan_helper(priv, IW_SCAN_TYPE_ACTIVE); +static void ipw_request_scan(struct work_struct *work) +{ + struct ipw_priv *priv = + container_of(work, struct ipw_priv, request_scan.work); + ipw_request_scan_helper(priv, IW_SCAN_TYPE_ACTIVE); } -static void ipw_bg_abort_scan(void *data) +static void ipw_bg_abort_scan(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, abort_scan); mutex_lock(&priv->mutex); - ipw_abort_scan(data); + ipw_abort_scan(priv); mutex_unlock(&priv->mutex); } @@ -7084,9 +7102,10 @@ static int ipw_qos_set_tx_queue_command(struct ipw_priv *priv, /* * background support to run QoS activate functionality */ -static void ipw_bg_qos_activate(void *data) +static void ipw_bg_qos_activate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, qos_activate); if (priv == NULL) return; @@ -7394,11 +7413,12 @@ static void ipw_roam(void *data) priv->status &= ~STATUS_ROAMING; } -static void ipw_bg_roam(void *data) +static void ipw_bg_roam(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, roam); mutex_lock(&priv->mutex); - ipw_roam(data); + ipw_roam(priv); mutex_unlock(&priv->mutex); } @@ -7479,8 +7499,8 @@ static int ipw_associate(void *data) &priv->request_scan, SCAN_INTERVAL); else - queue_work(priv->workqueue, - &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); } return 0; @@ -7491,11 +7511,12 @@ static int ipw_associate(void *data) return 1; } -static void ipw_bg_associate(void *data) +static void ipw_bg_associate(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, associate); mutex_lock(&priv->mutex); - ipw_associate(data); + ipw_associate(priv); mutex_unlock(&priv->mutex); } @@ -9410,7 +9431,7 @@ static int ipw_wx_set_scan(struct net_device *dev, IPW_DEBUG_WX("Start scan\n"); - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, &priv->request_scan, 0); return 0; } @@ -10547,11 +10568,12 @@ static void ipw_rf_kill(void *adapter) spin_unlock_irqrestore(&priv->lock, flags); } -static void ipw_bg_rf_kill(void *data) +static void ipw_bg_rf_kill(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, rf_kill.work); mutex_lock(&priv->mutex); - ipw_rf_kill(data); + ipw_rf_kill(priv); mutex_unlock(&priv->mutex); } @@ -10582,11 +10604,12 @@ static void ipw_link_up(struct ipw_priv *priv) queue_delayed_work(priv->workqueue, &priv->request_scan, HZ); } -static void ipw_bg_link_up(void *data) +static void ipw_bg_link_up(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, link_up); mutex_lock(&priv->mutex); - ipw_link_up(data); + ipw_link_up(priv); mutex_unlock(&priv->mutex); } @@ -10606,15 +10629,16 @@ static void ipw_link_down(struct ipw_priv *priv) if (!(priv->status & STATUS_EXIT_PENDING)) { /* Queue up another scan... */ - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, &priv->request_scan, 0); } } -static void ipw_bg_link_down(void *data) +static void ipw_bg_link_down(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, link_down); mutex_lock(&priv->mutex); - ipw_link_down(data); + ipw_link_down(priv); mutex_unlock(&priv->mutex); } @@ -10626,38 +10650,30 @@ static int ipw_setup_deferred_work(struct ipw_priv *priv) init_waitqueue_head(&priv->wait_command_queue); init_waitqueue_head(&priv->wait_state); - INIT_WORK(&priv->adhoc_check, ipw_bg_adhoc_check, priv); - INIT_WORK(&priv->associate, ipw_bg_associate, priv); - INIT_WORK(&priv->disassociate, ipw_bg_disassociate, priv); - INIT_WORK(&priv->system_config, ipw_system_config, priv); - INIT_WORK(&priv->rx_replenish, ipw_bg_rx_queue_replenish, priv); - INIT_WORK(&priv->adapter_restart, ipw_bg_adapter_restart, priv); - INIT_WORK(&priv->rf_kill, ipw_bg_rf_kill, priv); - INIT_WORK(&priv->up, (void (*)(void *))ipw_bg_up, priv); - INIT_WORK(&priv->down, (void (*)(void *))ipw_bg_down, priv); - INIT_WORK(&priv->request_scan, - (void (*)(void *))ipw_request_scan, priv); - INIT_WORK(&priv->request_passive_scan, - (void (*)(void *))ipw_request_passive_scan, priv); - INIT_WORK(&priv->gather_stats, - (void (*)(void *))ipw_bg_gather_stats, priv); - INIT_WORK(&priv->abort_scan, (void (*)(void *))ipw_bg_abort_scan, priv); - INIT_WORK(&priv->roam, ipw_bg_roam, priv); - INIT_WORK(&priv->scan_check, ipw_bg_scan_check, priv); - INIT_WORK(&priv->link_up, (void (*)(void *))ipw_bg_link_up, priv); - INIT_WORK(&priv->link_down, (void (*)(void *))ipw_bg_link_down, priv); - INIT_WORK(&priv->led_link_on, (void (*)(void *))ipw_bg_led_link_on, - priv); - INIT_WORK(&priv->led_link_off, (void (*)(void *))ipw_bg_led_link_off, - priv); - INIT_WORK(&priv->led_act_off, (void (*)(void *))ipw_bg_led_activity_off, - priv); - INIT_WORK(&priv->merge_networks, - (void (*)(void *))ipw_merge_adhoc_network, priv); + INIT_DELAYED_WORK(&priv->adhoc_check, ipw_bg_adhoc_check); + INIT_WORK(&priv->associate, ipw_bg_associate); + INIT_WORK(&priv->disassociate, ipw_bg_disassociate); + INIT_WORK(&priv->system_config, ipw_system_config); + INIT_WORK(&priv->rx_replenish, ipw_bg_rx_queue_replenish); + INIT_WORK(&priv->adapter_restart, ipw_bg_adapter_restart); + INIT_DELAYED_WORK(&priv->rf_kill, ipw_bg_rf_kill); + INIT_WORK(&priv->up, ipw_bg_up); + INIT_WORK(&priv->down, ipw_bg_down); + INIT_DELAYED_WORK(&priv->request_scan, ipw_request_scan); + INIT_WORK(&priv->request_passive_scan, ipw_request_passive_scan); + INIT_DELAYED_WORK(&priv->gather_stats, ipw_bg_gather_stats); + INIT_WORK(&priv->abort_scan, ipw_bg_abort_scan); + INIT_WORK(&priv->roam, ipw_bg_roam); + INIT_DELAYED_WORK(&priv->scan_check, ipw_bg_scan_check); + INIT_WORK(&priv->link_up, ipw_bg_link_up); + INIT_WORK(&priv->link_down, ipw_bg_link_down); + INIT_DELAYED_WORK(&priv->led_link_on, ipw_bg_led_link_on); + INIT_DELAYED_WORK(&priv->led_link_off, ipw_bg_led_link_off); + INIT_DELAYED_WORK(&priv->led_act_off, ipw_bg_led_activity_off); + INIT_WORK(&priv->merge_networks, ipw_merge_adhoc_network); #ifdef CONFIG_IPW2200_QOS - INIT_WORK(&priv->qos_activate, (void (*)(void *))ipw_bg_qos_activate, - priv); + INIT_WORK(&priv->qos_activate, ipw_bg_qos_activate); #endif /* CONFIG_IPW2200_QOS */ tasklet_init(&priv->irq_tasklet, (void (*)(unsigned long)) @@ -11190,7 +11206,8 @@ static int ipw_up(struct ipw_priv *priv) /* If configure to try and auto-associate, kick * off a scan. */ - queue_work(priv->workqueue, &priv->request_scan); + queue_delayed_work(priv->workqueue, + &priv->request_scan, 0); return 0; } @@ -11211,11 +11228,12 @@ static int ipw_up(struct ipw_priv *priv) return -EIO; } -static void ipw_bg_up(void *data) +static void ipw_bg_up(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, up); mutex_lock(&priv->mutex); - ipw_up(data); + ipw_up(priv); mutex_unlock(&priv->mutex); } @@ -11282,11 +11300,12 @@ static void ipw_down(struct ipw_priv *priv) ipw_led_radio_off(priv); } -static void ipw_bg_down(void *data) +static void ipw_bg_down(struct work_struct *work) { - struct ipw_priv *priv = data; + struct ipw_priv *priv = + container_of(work, struct ipw_priv, down); mutex_lock(&priv->mutex); - ipw_down(data); + ipw_down(priv); mutex_unlock(&priv->mutex); } diff --git a/drivers/net/wireless/ipw2200.h b/drivers/net/wireless/ipw2200.h index dad5eedefbf1..626a240a87d8 100644 --- a/drivers/net/wireless/ipw2200.h +++ b/drivers/net/wireless/ipw2200.h @@ -1290,21 +1290,21 @@ struct ipw_priv { struct workqueue_struct *workqueue; - struct work_struct adhoc_check; + struct delayed_work adhoc_check; struct work_struct associate; struct work_struct disassociate; struct work_struct system_config; struct work_struct rx_replenish; - struct work_struct request_scan; + struct delayed_work request_scan; struct work_struct request_passive_scan; struct work_struct adapter_restart; - struct work_struct rf_kill; + struct delayed_work rf_kill; struct work_struct up; struct work_struct down; - struct work_struct gather_stats; + struct delayed_work gather_stats; struct work_struct abort_scan; struct work_struct roam; - struct work_struct scan_check; + struct delayed_work scan_check; struct work_struct link_up; struct work_struct link_down; @@ -1319,9 +1319,9 @@ struct ipw_priv { u32 led_ofdm_on; u32 led_ofdm_off; - struct work_struct led_link_on; - struct work_struct led_link_off; - struct work_struct led_act_off; + struct delayed_work led_link_on; + struct delayed_work led_link_off; + struct delayed_work led_act_off; struct work_struct merge_networks; struct ipw_cmd_log *cmdlog; diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c index 336cabac13b3..936c888e03e1 100644 --- a/drivers/net/wireless/orinoco.c +++ b/drivers/net/wireless/orinoco.c @@ -980,9 +980,11 @@ static void print_linkstatus(struct net_device *dev, u16 status) } /* Search scan results for requested BSSID, join it if found */ -static void orinoco_join_ap(struct net_device *dev) +static void orinoco_join_ap(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, join_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; int err; unsigned long flags; @@ -1055,9 +1057,11 @@ static void orinoco_join_ap(struct net_device *dev) } /* Send new BSSID to userspace */ -static void orinoco_send_wevents(struct net_device *dev) +static void orinoco_send_wevents(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, wevent_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; union iwreq_data wrqu; int err; @@ -1864,9 +1868,11 @@ __orinoco_set_multicast_list(struct net_device *dev) /* This must be called from user context, without locks held - use * schedule_work() */ -static void orinoco_reset(struct net_device *dev) +static void orinoco_reset(struct work_struct *work) { - struct orinoco_private *priv = netdev_priv(dev); + struct orinoco_private *priv = + container_of(work, struct orinoco_private, reset_work); + struct net_device *dev = priv->ndev; struct hermes *hw = &priv->hw; int err; unsigned long flags; @@ -2434,9 +2440,9 @@ struct net_device *alloc_orinocodev(int sizeof_card, priv->hw_unavailable = 1; /* orinoco_init() must clear this * before anything else touches the * hardware */ - INIT_WORK(&priv->reset_work, (void (*)(void *))orinoco_reset, dev); - INIT_WORK(&priv->join_work, (void (*)(void *))orinoco_join_ap, dev); - INIT_WORK(&priv->wevent_work, (void (*)(void *))orinoco_send_wevents, dev); + INIT_WORK(&priv->reset_work, orinoco_reset); + INIT_WORK(&priv->join_work, orinoco_join_ap); + INIT_WORK(&priv->wevent_work, orinoco_send_wevents); netif_carrier_off(dev); priv->last_linkstatus = 0xffff; @@ -3608,7 +3614,7 @@ static int orinoco_ioctl_reset(struct net_device *dev, printk(KERN_DEBUG "%s: Forcing reset!\n", dev->name); /* Firmware reset */ - orinoco_reset(dev); + orinoco_reset(&priv->reset_work); } else { printk(KERN_DEBUG "%s: Force scheduling reset!\n", dev->name); @@ -4154,7 +4160,7 @@ static int orinoco_ioctl_commit(struct net_device *dev, return 0; if (priv->broken_disableport) { - orinoco_reset(dev); + orinoco_reset(&priv->reset_work); return 0; } diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c index 286325ca3293..e7700b4257eb 100644 --- a/drivers/net/wireless/prism54/isl_ioctl.c +++ b/drivers/net/wireless/prism54/isl_ioctl.c @@ -158,8 +158,9 @@ prism54_mib_init(islpci_private *priv) * schedule_work(), thus we can as well use sleeping semaphore * locking */ void -prism54_update_stats(islpci_private *priv) +prism54_update_stats(struct work_struct *work) { + islpci_private *priv = container_of(work, islpci_private, stats_work); char *data; int j; struct obj_bss bss, *bss2; @@ -2494,9 +2495,10 @@ prism54_process_trap_helper(islpci_private *priv, enum oid_num_t oid, * interrupt context, no locks held. */ void -prism54_process_trap(void *data) +prism54_process_trap(struct work_struct *work) { - struct islpci_mgmtframe *frame = data; + struct islpci_mgmtframe *frame = + container_of(work, struct islpci_mgmtframe, ws); struct net_device *ndev = frame->ndev; enum oid_num_t n = mgt_oidtonum(frame->header->oid); diff --git a/drivers/net/wireless/prism54/isl_ioctl.h b/drivers/net/wireless/prism54/isl_ioctl.h index 65f33acd0a42..0802fa64996f 100644 --- a/drivers/net/wireless/prism54/isl_ioctl.h +++ b/drivers/net/wireless/prism54/isl_ioctl.h @@ -32,12 +32,12 @@ void prism54_mib_init(islpci_private *); struct iw_statistics *prism54_get_wireless_stats(struct net_device *); -void prism54_update_stats(islpci_private *); +void prism54_update_stats(struct work_struct *); void prism54_acl_init(struct islpci_acl *); void prism54_acl_clean(struct islpci_acl *); -void prism54_process_trap(void *); +void prism54_process_trap(struct work_struct *); void prism54_wpa_bss_ie_init(islpci_private *priv); void prism54_wpa_bss_ie_clean(islpci_private *priv); diff --git a/drivers/net/wireless/prism54/islpci_dev.c b/drivers/net/wireless/prism54/islpci_dev.c index ec1c00f19eb3..e35fcb2543c4 100644 --- a/drivers/net/wireless/prism54/islpci_dev.c +++ b/drivers/net/wireless/prism54/islpci_dev.c @@ -861,11 +861,10 @@ islpci_setup(struct pci_dev *pdev) priv->state_off = 1; /* initialize workqueue's */ - INIT_WORK(&priv->stats_work, - (void (*)(void *)) prism54_update_stats, priv); + INIT_WORK(&priv->stats_work, prism54_update_stats); priv->stats_timestamp = 0; - INIT_WORK(&priv->reset_task, islpci_do_reset_and_wake, priv); + INIT_WORK(&priv->reset_task, islpci_do_reset_and_wake); priv->reset_task_pending = 0; /* allocate various memory areas */ diff --git a/drivers/net/wireless/prism54/islpci_eth.c b/drivers/net/wireless/prism54/islpci_eth.c index a8261d8454dd..103a37877733 100644 --- a/drivers/net/wireless/prism54/islpci_eth.c +++ b/drivers/net/wireless/prism54/islpci_eth.c @@ -482,9 +482,9 @@ islpci_eth_receive(islpci_private *priv) } void -islpci_do_reset_and_wake(void *data) +islpci_do_reset_and_wake(struct work_struct *work) { - islpci_private *priv = (islpci_private *) data; + islpci_private *priv = container_of(work, islpci_private, reset_task); islpci_reset(priv, 1); netif_wake_queue(priv->ndev); priv->reset_task_pending = 0; diff --git a/drivers/net/wireless/prism54/islpci_eth.h b/drivers/net/wireless/prism54/islpci_eth.h index bc9d7a60b8d6..99d37eda9f01 100644 --- a/drivers/net/wireless/prism54/islpci_eth.h +++ b/drivers/net/wireless/prism54/islpci_eth.h @@ -68,6 +68,6 @@ void islpci_eth_cleanup_transmit(islpci_private *, isl38xx_control_block *); int islpci_eth_transmit(struct sk_buff *, struct net_device *); int islpci_eth_receive(islpci_private *); void islpci_eth_tx_timeout(struct net_device *); -void islpci_do_reset_and_wake(void *data); +void islpci_do_reset_and_wake(struct work_struct *); #endif /* _ISL_GEN_H */ diff --git a/drivers/net/wireless/prism54/islpci_mgt.c b/drivers/net/wireless/prism54/islpci_mgt.c index 2e061a80b294..656ec9fa7128 100644 --- a/drivers/net/wireless/prism54/islpci_mgt.c +++ b/drivers/net/wireless/prism54/islpci_mgt.c @@ -387,7 +387,7 @@ islpci_mgt_receive(struct net_device *ndev) /* Create work to handle trap out of interrupt * context. */ - INIT_WORK(&frame->ws, prism54_process_trap, frame); + INIT_WORK(&frame->ws, prism54_process_trap); schedule_work(&frame->ws); } else { diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index a7d29bddb298..5e4f4b707375 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -1090,9 +1090,10 @@ void zd_dump_rx_status(const struct rx_status *status) #define LINK_LED_WORK_DELAY HZ -static void link_led_handler(void *p) +static void link_led_handler(struct work_struct *work) { - struct zd_mac *mac = p; + struct zd_mac *mac = + container_of(work, struct zd_mac, housekeeping.link_led_work.work); struct zd_chip *chip = &mac->chip; struct ieee80211softmac_device *sm = ieee80211_priv(mac->netdev); int is_associated; @@ -1113,7 +1114,7 @@ static void link_led_handler(void *p) static void housekeeping_init(struct zd_mac *mac) { - INIT_WORK(&mac->housekeeping.link_led_work, link_led_handler, mac); + INIT_DELAYED_WORK(&mac->housekeeping.link_led_work, link_led_handler); } static void housekeeping_enable(struct zd_mac *mac) diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h index b8ea3de7924a..7957cac3de25 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.h +++ b/drivers/net/wireless/zd1211rw/zd_mac.h @@ -121,7 +121,7 @@ enum mac_flags { }; struct housekeeping { - struct work_struct link_led_work; + struct delayed_work link_led_work; }; #define ZD_MAC_STATS_BUFFER_SIZE 16 diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index fc4bc9b94c74..a83c3db7d18f 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -29,7 +29,7 @@ struct oprofile_cpu_buffer cpu_buffer[NR_CPUS] __cacheline_aligned; -static void wq_sync_buffer(void *); +static void wq_sync_buffer(struct work_struct *work); #define DEFAULT_TIMER_EXPIRE (HZ / 10) static int work_enabled; @@ -65,7 +65,7 @@ int alloc_cpu_buffers(void) b->sample_received = 0; b->sample_lost_overflow = 0; b->cpu = i; - INIT_WORK(&b->work, wq_sync_buffer, b); + INIT_DELAYED_WORK(&b->work, wq_sync_buffer); } return 0; @@ -282,9 +282,10 @@ void oprofile_add_trace(unsigned long pc) * By using schedule_delayed_work_on and then schedule_delayed_work * we guarantee this will stay on the correct cpu */ -static void wq_sync_buffer(void * data) +static void wq_sync_buffer(struct work_struct *work) { - struct oprofile_cpu_buffer * b = data; + struct oprofile_cpu_buffer * b = + container_of(work, struct oprofile_cpu_buffer, work.work); if (b->cpu != smp_processor_id()) { printk("WQ on CPU%d, prefer CPU%d\n", smp_processor_id(), b->cpu); diff --git a/drivers/oprofile/cpu_buffer.h b/drivers/oprofile/cpu_buffer.h index 09abb80e0570..49900d9e3235 100644 --- a/drivers/oprofile/cpu_buffer.h +++ b/drivers/oprofile/cpu_buffer.h @@ -43,7 +43,7 @@ struct oprofile_cpu_buffer { unsigned long sample_lost_overflow; unsigned long backtrace_aborted; int cpu; - struct work_struct work; + struct delayed_work work; } ____cacheline_aligned; extern struct oprofile_cpu_buffer cpu_buffer[]; diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h index ea2087c34149..50757695844f 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -70,7 +70,7 @@ struct slot { struct hotplug_slot *hotplug_slot; struct list_head slot_list; char name[SLOT_NAME_SIZE]; - struct work_struct work; /* work for button event */ + struct delayed_work work; /* work for button event */ struct mutex lock; }; @@ -187,7 +187,7 @@ extern int shpchp_configure_device(struct slot *p_slot); extern int shpchp_unconfigure_device(struct slot *p_slot); extern void shpchp_remove_ctrl_files(struct controller *ctrl); extern void cleanup_slots(struct controller *ctrl); -extern void queue_pushbutton_work(void *data); +extern void queue_pushbutton_work(struct work_struct *work); #ifdef CONFIG_ACPI diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c index 235c18a22393..4eac85b3d90e 100644 --- a/drivers/pci/hotplug/shpchp_core.c +++ b/drivers/pci/hotplug/shpchp_core.c @@ -159,7 +159,7 @@ static int init_slots(struct controller *ctrl) goto error_info; slot->number = sun; - INIT_WORK(&slot->work, queue_pushbutton_work, slot); + INIT_DELAYED_WORK(&slot->work, queue_pushbutton_work); /* register this slot with the hotplug pci core */ hotplug_slot->private = slot; diff --git a/drivers/pci/hotplug/shpchp_ctrl.c b/drivers/pci/hotplug/shpchp_ctrl.c index c39901dbff20..158ac7836096 100644 --- a/drivers/pci/hotplug/shpchp_ctrl.c +++ b/drivers/pci/hotplug/shpchp_ctrl.c @@ -36,7 +36,7 @@ #include "../pci.h" #include "shpchp.h" -static void interrupt_event_handler(void *data); +static void interrupt_event_handler(struct work_struct *work); static int shpchp_enable_slot(struct slot *p_slot); static int shpchp_disable_slot(struct slot *p_slot); @@ -50,7 +50,7 @@ static int queue_interrupt_event(struct slot *p_slot, u32 event_type) info->event_type = event_type; info->p_slot = p_slot; - INIT_WORK(&info->work, interrupt_event_handler, info); + INIT_WORK(&info->work, interrupt_event_handler); schedule_work(&info->work); @@ -408,9 +408,10 @@ struct pushbutton_work_info { * Handles all pending events and exits. * */ -static void shpchp_pushbutton_thread(void *data) +static void shpchp_pushbutton_thread(struct work_struct *work) { - struct pushbutton_work_info *info = data; + struct pushbutton_work_info *info = + container_of(work, struct pushbutton_work_info, work); struct slot *p_slot = info->p_slot; mutex_lock(&p_slot->lock); @@ -436,9 +437,9 @@ static void shpchp_pushbutton_thread(void *data) kfree(info); } -void queue_pushbutton_work(void *data) +void queue_pushbutton_work(struct work_struct *work) { - struct slot *p_slot = data; + struct slot *p_slot = container_of(work, struct slot, work.work); struct pushbutton_work_info *info; info = kmalloc(sizeof(*info), GFP_KERNEL); @@ -447,7 +448,7 @@ void queue_pushbutton_work(void *data) return; } info->p_slot = p_slot; - INIT_WORK(&info->work, shpchp_pushbutton_thread, info); + INIT_WORK(&info->work, shpchp_pushbutton_thread); mutex_lock(&p_slot->lock); switch (p_slot->state) { @@ -541,9 +542,9 @@ static void handle_button_press_event(struct slot *p_slot) } } -static void interrupt_event_handler(void *data) +static void interrupt_event_handler(struct work_struct *work) { - struct event_info *info = data; + struct event_info *info = container_of(work, struct event_info, work); struct slot *p_slot = info->p_slot; mutex_lock(&p_slot->lock); diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index a20d84d707d9..e469a46a388b 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -698,9 +698,10 @@ static int pcmcia_card_add(struct pcmcia_socket *s) } -static void pcmcia_delayed_add_pseudo_device(void *data) +static void pcmcia_delayed_add_pseudo_device(struct work_struct *work) { - struct pcmcia_socket *s = data; + struct pcmcia_socket *s = + container_of(work, struct pcmcia_socket, device_add); pcmcia_device_add(s, 0); s->pcmcia_state.device_add_pending = 0; } @@ -1246,7 +1247,7 @@ static int __devinit pcmcia_bus_add_socket(struct class_device *class_dev, init_waitqueue_head(&socket->queue); #endif INIT_LIST_HEAD(&socket->devices_list); - INIT_WORK(&socket->device_add, pcmcia_delayed_add_pseudo_device, socket); + INIT_WORK(&socket->device_add, pcmcia_delayed_add_pseudo_device); memset(&socket->pcmcia_state, 0, sizeof(u8)); socket->device_count = 0; diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c index 583789c66cdb..dcf5f86461f7 100644 --- a/drivers/rtc/rtc-dev.c +++ b/drivers/rtc/rtc-dev.c @@ -53,9 +53,10 @@ static int rtc_dev_open(struct inode *inode, struct file *file) * Routine to poll RTC seconds field for change as often as possible, * after first RTC_UIE use timer to reduce polling */ -static void rtc_uie_task(void *data) +static void rtc_uie_task(struct work_struct *work) { - struct rtc_device *rtc = data; + struct rtc_device *rtc = + container_of(work, struct rtc_device, uie_task); struct rtc_time tm; int num = 0; int err; @@ -398,7 +399,7 @@ static int rtc_dev_add_device(struct class_device *class_dev, spin_lock_init(&rtc->irq_lock); init_waitqueue_head(&rtc->irq_queue); #ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL - INIT_WORK(&rtc->uie_task, rtc_uie_task, rtc); + INIT_WORK(&rtc->uie_task, rtc_uie_task); setup_timer(&rtc->uie_timer, rtc_uie_timer, (unsigned long)rtc); #endif diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c index a6aa91072880..bb3cb3360541 100644 --- a/drivers/scsi/NCR5380.c +++ b/drivers/scsi/NCR5380.c @@ -849,7 +849,7 @@ static int __devinit NCR5380_init(struct Scsi_Host *instance, int flags) hostdata->issue_queue = NULL; hostdata->disconnected_queue = NULL; - INIT_WORK(&hostdata->coroutine, NCR5380_main, hostdata); + INIT_DELAYED_WORK(&hostdata->coroutine, NCR5380_main); #ifdef NCR5380_STATS for (i = 0; i < 8; ++i) { @@ -1016,7 +1016,7 @@ static int NCR5380_queue_command(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) /* Run the coroutine if it isn't already running. */ /* Kick off command processing */ - schedule_work(&hostdata->coroutine); + schedule_delayed_work(&hostdata->coroutine, 0); return 0; } @@ -1033,9 +1033,10 @@ static int NCR5380_queue_command(Scsi_Cmnd * cmd, void (*done) (Scsi_Cmnd *)) * host lock and called routines may take the isa dma lock. */ -static void NCR5380_main(void *p) +static void NCR5380_main(struct work_struct *work) { - struct NCR5380_hostdata *hostdata = p; + struct NCR5380_hostdata *hostdata = + container_of(work, struct NCR5380_hostdata, coroutine.work); struct Scsi_Host *instance = hostdata->host; Scsi_Cmnd *tmp, *prev; int done; @@ -1221,7 +1222,7 @@ static irqreturn_t NCR5380_intr(int irq, void *dev_id) } /* if BASR_IRQ */ spin_unlock_irqrestore(instance->host_lock, flags); if(!done) - schedule_work(&hostdata->coroutine); + schedule_delayed_work(&hostdata->coroutine, 0); } while (!done); return IRQ_HANDLED; } diff --git a/drivers/scsi/NCR5380.h b/drivers/scsi/NCR5380.h index 1bc73de496b0..713a108c02ef 100644 --- a/drivers/scsi/NCR5380.h +++ b/drivers/scsi/NCR5380.h @@ -271,7 +271,7 @@ struct NCR5380_hostdata { unsigned long time_expires; /* in jiffies, set prior to sleeping */ int select_time; /* timer in select for target response */ volatile Scsi_Cmnd *selecting; - struct work_struct coroutine; /* our co-routine */ + struct delayed_work coroutine; /* our co-routine */ #ifdef NCR5380_STATS unsigned timebase; /* Base for time calcs */ long time_read[8]; /* time to do reads */ @@ -298,7 +298,7 @@ static void NCR5380_information_transfer(struct Scsi_Host *instance); #ifndef DONT_USE_INTR static irqreturn_t NCR5380_intr(int irq, void *dev_id); #endif -static void NCR5380_main(void *ptr); +static void NCR5380_main(struct work_struct *work); static void NCR5380_print_options(struct Scsi_Host *instance); #ifdef NDEBUG static void NCR5380_print_phase(struct Scsi_Host *instance); diff --git a/drivers/scsi/aha152x.c b/drivers/scsi/aha152x.c index 306f46b85a55..0cec742d12e9 100644 --- a/drivers/scsi/aha152x.c +++ b/drivers/scsi/aha152x.c @@ -1443,7 +1443,7 @@ static struct work_struct aha152x_tq; * Run service completions on the card with interrupts enabled. * */ -static void run(void) +static void run(struct work_struct *work) { struct aha152x_hostdata *hd; @@ -1499,7 +1499,7 @@ static irqreturn_t intr(int irqno, void *dev_id) HOSTDATA(shpnt)->service=1; /* Poke the BH handler */ - INIT_WORK(&aha152x_tq, (void *) run, NULL); + INIT_WORK(&aha152x_tq, run); schedule_work(&aha152x_tq); } DO_UNLOCK(flags); diff --git a/drivers/scsi/imm.c b/drivers/scsi/imm.c index e31f6122106f..0464c182c577 100644 --- a/drivers/scsi/imm.c +++ b/drivers/scsi/imm.c @@ -36,7 +36,7 @@ typedef struct { int base_hi; /* Hi Base address for ECP-ISA chipset */ int mode; /* Transfer mode */ struct scsi_cmnd *cur_cmd; /* Current queued command */ - struct work_struct imm_tq; /* Polling interrupt stuff */ + struct delayed_work imm_tq; /* Polling interrupt stuff */ unsigned long jstart; /* Jiffies at start */ unsigned failed:1; /* Failure flag */ unsigned dp:1; /* Data phase present */ @@ -733,9 +733,9 @@ static int imm_completion(struct scsi_cmnd *cmd) * the scheduler's task queue to generate a stream of call-backs and * complete the request when the drive is ready. */ -static void imm_interrupt(void *data) +static void imm_interrupt(struct work_struct *work) { - imm_struct *dev = (imm_struct *) data; + imm_struct *dev = container_of(work, imm_struct, imm_tq.work); struct scsi_cmnd *cmd = dev->cur_cmd; struct Scsi_Host *host = cmd->device->host; unsigned long flags; @@ -745,7 +745,6 @@ static void imm_interrupt(void *data) return; } if (imm_engine(dev, cmd)) { - INIT_WORK(&dev->imm_tq, imm_interrupt, (void *) dev); schedule_delayed_work(&dev->imm_tq, 1); return; } @@ -953,8 +952,7 @@ static int imm_queuecommand(struct scsi_cmnd *cmd, cmd->result = DID_ERROR << 16; /* default return code */ cmd->SCp.phase = 0; /* bus free */ - INIT_WORK(&dev->imm_tq, imm_interrupt, dev); - schedule_work(&dev->imm_tq); + schedule_delayed_work(&dev->imm_tq, 0); imm_pb_claim(dev); @@ -1225,7 +1223,7 @@ static int __imm_attach(struct parport *pb) else ports = 8; - INIT_WORK(&dev->imm_tq, imm_interrupt, dev); + INIT_DELAYED_WORK(&dev->imm_tq, imm_interrupt); err = -ENOMEM; host = scsi_host_alloc(&imm_template, sizeof(imm_struct *)); diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 2dde821025f3..d51c3e764bb0 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -2093,7 +2093,7 @@ static void ipr_release_dump(struct kref *kref) /** * ipr_worker_thread - Worker thread - * @data: ioa config struct + * @work: ioa config struct * * Called at task level from a work thread. This function takes care * of adding and removing device from the mid-layer as configuration @@ -2102,13 +2102,14 @@ static void ipr_release_dump(struct kref *kref) * Return value: * nothing **/ -static void ipr_worker_thread(void *data) +static void ipr_worker_thread(struct work_struct *work) { unsigned long lock_flags; struct ipr_resource_entry *res; struct scsi_device *sdev; struct ipr_dump *dump; - struct ipr_ioa_cfg *ioa_cfg = data; + struct ipr_ioa_cfg *ioa_cfg = + container_of(work, struct ipr_ioa_cfg, work_q); u8 bus, target, lun; int did_work; @@ -6926,7 +6927,7 @@ static void __devinit ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg, INIT_LIST_HEAD(&ioa_cfg->hostrcb_pending_q); INIT_LIST_HEAD(&ioa_cfg->free_res_q); INIT_LIST_HEAD(&ioa_cfg->used_res_q); - INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread, ioa_cfg); + INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread); init_waitqueue_head(&ioa_cfg->reset_wait_q); ioa_cfg->sdt_state = INACTIVE; if (ipr_enable_cache) diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 5d8862189485..e11b23c641e2 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -719,9 +719,10 @@ again: return rc; } -static void iscsi_xmitworker(void *data) +static void iscsi_xmitworker(struct work_struct *work) { - struct iscsi_conn *conn = data; + struct iscsi_conn *conn = + container_of(work, struct iscsi_conn, xmitwork); int rc; /* * serialize Xmit worker on a per-connection basis. @@ -1512,7 +1513,7 @@ iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx) if (conn->mgmtqueue == ERR_PTR(-ENOMEM)) goto mgmtqueue_alloc_fail; - INIT_WORK(&conn->xmitwork, iscsi_xmitworker, conn); + INIT_WORK(&conn->xmitwork, iscsi_xmitworker); /* allocate login_mtask used for the login/text sequences */ spin_lock_bh(&session->lock); diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c index d977bd492d8d..fb7df7b75811 100644 --- a/drivers/scsi/libsas/sas_discover.c +++ b/drivers/scsi/libsas/sas_discover.c @@ -647,10 +647,12 @@ void sas_unregister_domain_devices(struct asd_sas_port *port) * Discover process only interrogates devices in order to discover the * domain. */ -static void sas_discover_domain(void *data) +static void sas_discover_domain(struct work_struct *work) { int error = 0; - struct asd_sas_port *port = data; + struct sas_discovery_event *ev = + container_of(work, struct sas_discovery_event, work); + struct asd_sas_port *port = ev->port; sas_begin_event(DISCE_DISCOVER_DOMAIN, &port->disc.disc_event_lock, &port->disc.pending); @@ -692,10 +694,12 @@ static void sas_discover_domain(void *data) current->pid, error); } -static void sas_revalidate_domain(void *data) +static void sas_revalidate_domain(struct work_struct *work) { int res = 0; - struct asd_sas_port *port = data; + struct sas_discovery_event *ev = + container_of(work, struct sas_discovery_event, work); + struct asd_sas_port *port = ev->port; sas_begin_event(DISCE_REVALIDATE_DOMAIN, &port->disc.disc_event_lock, &port->disc.pending); @@ -722,7 +726,7 @@ int sas_discover_event(struct asd_sas_port *port, enum discover_event ev) BUG_ON(ev >= DISC_NUM_EVENTS); sas_queue_event(ev, &disc->disc_event_lock, &disc->pending, - &disc->disc_work[ev], port->ha->core.shost); + &disc->disc_work[ev].work, port->ha->core.shost); return 0; } @@ -737,13 +741,15 @@ void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *port) { int i; - static void (*sas_event_fns[DISC_NUM_EVENTS])(void *) = { + static const work_func_t sas_event_fns[DISC_NUM_EVENTS] = { [DISCE_DISCOVER_DOMAIN] = sas_discover_domain, [DISCE_REVALIDATE_DOMAIN] = sas_revalidate_domain, }; spin_lock_init(&disc->disc_event_lock); disc->pending = 0; - for (i = 0; i < DISC_NUM_EVENTS; i++) - INIT_WORK(&disc->disc_work[i], sas_event_fns[i], port); + for (i = 0; i < DISC_NUM_EVENTS; i++) { + INIT_WORK(&disc->disc_work[i].work, sas_event_fns[i]); + disc->disc_work[i].port = port; + } } diff --git a/drivers/scsi/libsas/sas_event.c b/drivers/scsi/libsas/sas_event.c index 19110ed1c89c..d83392ee6823 100644 --- a/drivers/scsi/libsas/sas_event.c +++ b/drivers/scsi/libsas/sas_event.c @@ -31,7 +31,7 @@ static void notify_ha_event(struct sas_ha_struct *sas_ha, enum ha_event event) BUG_ON(event >= HA_NUM_EVENTS); sas_queue_event(event, &sas_ha->event_lock, &sas_ha->pending, - &sas_ha->ha_events[event], sas_ha->core.shost); + &sas_ha->ha_events[event].work, sas_ha->core.shost); } static void notify_port_event(struct asd_sas_phy *phy, enum port_event event) @@ -41,7 +41,7 @@ static void notify_port_event(struct asd_sas_phy *phy, enum port_event event) BUG_ON(event >= PORT_NUM_EVENTS); sas_queue_event(event, &ha->event_lock, &phy->port_events_pending, - &phy->port_events[event], ha->core.shost); + &phy->port_events[event].work, ha->core.shost); } static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) @@ -51,12 +51,12 @@ static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) BUG_ON(event >= PHY_NUM_EVENTS); sas_queue_event(event, &ha->event_lock, &phy->phy_events_pending, - &phy->phy_events[event], ha->core.shost); + &phy->phy_events[event].work, ha->core.shost); } int sas_init_events(struct sas_ha_struct *sas_ha) { - static void (*sas_ha_event_fns[HA_NUM_EVENTS])(void *) = { + static const work_func_t sas_ha_event_fns[HA_NUM_EVENTS] = { [HAE_RESET] = sas_hae_reset, }; @@ -64,8 +64,10 @@ int sas_init_events(struct sas_ha_struct *sas_ha) spin_lock_init(&sas_ha->event_lock); - for (i = 0; i < HA_NUM_EVENTS; i++) - INIT_WORK(&sas_ha->ha_events[i], sas_ha_event_fns[i], sas_ha); + for (i = 0; i < HA_NUM_EVENTS; i++) { + INIT_WORK(&sas_ha->ha_events[i].work, sas_ha_event_fns[i]); + sas_ha->ha_events[i].ha = sas_ha; + } sas_ha->notify_ha_event = notify_ha_event; sas_ha->notify_port_event = notify_port_event; diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c index c836a237fb79..7b4e9169f44d 100644 --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -65,9 +65,11 @@ void sas_hash_addr(u8 *hashed, const u8 *sas_addr) /* ---------- HA events ---------- */ -void sas_hae_reset(void *data) +void sas_hae_reset(struct work_struct *work) { - struct sas_ha_struct *ha = data; + struct sas_ha_event *ev = + container_of(work, struct sas_ha_event, work); + struct sas_ha_struct *ha = ev->ha; sas_begin_event(HAE_RESET, &ha->event_lock, &ha->pending); diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h index bffcee474921..137d7e496b6d 100644 --- a/drivers/scsi/libsas/sas_internal.h +++ b/drivers/scsi/libsas/sas_internal.h @@ -60,11 +60,11 @@ void sas_shutdown_queue(struct sas_ha_struct *sas_ha); void sas_deform_port(struct asd_sas_phy *phy); -void sas_porte_bytes_dmaed(void *); -void sas_porte_broadcast_rcvd(void *); -void sas_porte_link_reset_err(void *); -void sas_porte_timer_event(void *); -void sas_porte_hard_reset(void *); +void sas_porte_bytes_dmaed(struct work_struct *work); +void sas_porte_broadcast_rcvd(struct work_struct *work); +void sas_porte_link_reset_err(struct work_struct *work); +void sas_porte_timer_event(struct work_struct *work); +void sas_porte_hard_reset(struct work_struct *work); int sas_notify_lldd_dev_found(struct domain_device *); void sas_notify_lldd_dev_gone(struct domain_device *); @@ -75,7 +75,7 @@ int sas_smp_get_phy_events(struct sas_phy *phy); struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy); -void sas_hae_reset(void *); +void sas_hae_reset(struct work_struct *work); static inline void sas_queue_event(int event, spinlock_t *lock, unsigned long *pending, diff --git a/drivers/scsi/libsas/sas_phy.c b/drivers/scsi/libsas/sas_phy.c index 9340cdbae4a3..b459c4b635b1 100644 --- a/drivers/scsi/libsas/sas_phy.c +++ b/drivers/scsi/libsas/sas_phy.c @@ -30,9 +30,11 @@ /* ---------- Phy events ---------- */ -static void sas_phye_loss_of_signal(void *data) +static void sas_phye_loss_of_signal(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PHYE_LOSS_OF_SIGNAL, &phy->ha->event_lock, &phy->phy_events_pending); @@ -40,18 +42,22 @@ static void sas_phye_loss_of_signal(void *data) sas_deform_port(phy); } -static void sas_phye_oob_done(void *data) +static void sas_phye_oob_done(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PHYE_OOB_DONE, &phy->ha->event_lock, &phy->phy_events_pending); phy->error = 0; } -static void sas_phye_oob_error(void *data) +static void sas_phye_oob_error(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; struct sas_ha_struct *sas_ha = phy->ha; struct asd_sas_port *port = phy->port; struct sas_internal *i = @@ -80,9 +86,11 @@ static void sas_phye_oob_error(void *data) } } -static void sas_phye_spinup_hold(void *data) +static void sas_phye_spinup_hold(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; struct sas_ha_struct *sas_ha = phy->ha; struct sas_internal *i = to_sas_internal(sas_ha->core.shost->transportt); @@ -100,14 +108,14 @@ int sas_register_phys(struct sas_ha_struct *sas_ha) { int i; - static void (*sas_phy_event_fns[PHY_NUM_EVENTS])(void *) = { + static const work_func_t sas_phy_event_fns[PHY_NUM_EVENTS] = { [PHYE_LOSS_OF_SIGNAL] = sas_phye_loss_of_signal, [PHYE_OOB_DONE] = sas_phye_oob_done, [PHYE_OOB_ERROR] = sas_phye_oob_error, [PHYE_SPINUP_HOLD] = sas_phye_spinup_hold, }; - static void (*sas_port_event_fns[PORT_NUM_EVENTS])(void *) = { + static const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = { [PORTE_BYTES_DMAED] = sas_porte_bytes_dmaed, [PORTE_BROADCAST_RCVD] = sas_porte_broadcast_rcvd, [PORTE_LINK_RESET_ERR] = sas_porte_link_reset_err, @@ -122,13 +130,18 @@ int sas_register_phys(struct sas_ha_struct *sas_ha) phy->error = 0; INIT_LIST_HEAD(&phy->port_phy_el); - for (k = 0; k < PORT_NUM_EVENTS; k++) - INIT_WORK(&phy->port_events[k], sas_port_event_fns[k], - phy); + for (k = 0; k < PORT_NUM_EVENTS; k++) { + INIT_WORK(&phy->port_events[k].work, + sas_port_event_fns[k]); + phy->port_events[k].phy = phy; + } + + for (k = 0; k < PHY_NUM_EVENTS; k++) { + INIT_WORK(&phy->phy_events[k].work, + sas_phy_event_fns[k]); + phy->phy_events[k].phy = phy; + } - for (k = 0; k < PHY_NUM_EVENTS; k++) - INIT_WORK(&phy->phy_events[k], sas_phy_event_fns[k], - phy); phy->port = NULL; phy->ha = sas_ha; spin_lock_init(&phy->frame_rcvd_lock); diff --git a/drivers/scsi/libsas/sas_port.c b/drivers/scsi/libsas/sas_port.c index 253cdcf306a2..971c37ceecb4 100644 --- a/drivers/scsi/libsas/sas_port.c +++ b/drivers/scsi/libsas/sas_port.c @@ -181,9 +181,11 @@ void sas_deform_port(struct asd_sas_phy *phy) /* ---------- SAS port events ---------- */ -void sas_porte_bytes_dmaed(void *data) +void sas_porte_bytes_dmaed(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_BYTES_DMAED, &phy->ha->event_lock, &phy->port_events_pending); @@ -191,11 +193,13 @@ void sas_porte_bytes_dmaed(void *data) sas_form_port(phy); } -void sas_porte_broadcast_rcvd(void *data) +void sas_porte_broadcast_rcvd(struct work_struct *work) { + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; unsigned long flags; u32 prim; - struct asd_sas_phy *phy = data; sas_begin_event(PORTE_BROADCAST_RCVD, &phy->ha->event_lock, &phy->port_events_pending); @@ -208,9 +212,11 @@ void sas_porte_broadcast_rcvd(void *data) sas_discover_event(phy->port, DISCE_REVALIDATE_DOMAIN); } -void sas_porte_link_reset_err(void *data) +void sas_porte_link_reset_err(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_LINK_RESET_ERR, &phy->ha->event_lock, &phy->port_events_pending); @@ -218,9 +224,11 @@ void sas_porte_link_reset_err(void *data) sas_deform_port(phy); } -void sas_porte_timer_event(void *data) +void sas_porte_timer_event(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_TIMER_EVENT, &phy->ha->event_lock, &phy->port_events_pending); @@ -228,9 +236,11 @@ void sas_porte_timer_event(void *data) sas_deform_port(phy); } -void sas_porte_hard_reset(void *data) +void sas_porte_hard_reset(struct work_struct *work) { - struct asd_sas_phy *phy = data; + struct asd_sas_event *ev = + container_of(work, struct asd_sas_event, work); + struct asd_sas_phy *phy = ev->phy; sas_begin_event(PORTE_HARD_RESET, &phy->ha->event_lock, &phy->port_events_pending); diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index 89a2a9f11e41..584ba4d6e038 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -31,7 +31,7 @@ typedef struct { int base; /* Actual port address */ int mode; /* Transfer mode */ struct scsi_cmnd *cur_cmd; /* Current queued command */ - struct work_struct ppa_tq; /* Polling interrupt stuff */ + struct delayed_work ppa_tq; /* Polling interrupt stuff */ unsigned long jstart; /* Jiffies at start */ unsigned long recon_tmo; /* How many usecs to wait for reconnection (6th bit) */ unsigned int failed:1; /* Failure flag */ @@ -627,9 +627,9 @@ static int ppa_completion(struct scsi_cmnd *cmd) * the scheduler's task queue to generate a stream of call-backs and * complete the request when the drive is ready. */ -static void ppa_interrupt(void *data) +static void ppa_interrupt(struct work_struct *work) { - ppa_struct *dev = (ppa_struct *) data; + ppa_struct *dev = container_of(work, ppa_struct, ppa_tq.work); struct scsi_cmnd *cmd = dev->cur_cmd; if (!cmd) { @@ -637,7 +637,6 @@ static void ppa_interrupt(void *data) return; } if (ppa_engine(dev, cmd)) { - dev->ppa_tq.data = (void *) dev; schedule_delayed_work(&dev->ppa_tq, 1); return; } @@ -822,8 +821,7 @@ static int ppa_queuecommand(struct scsi_cmnd *cmd, cmd->result = DID_ERROR << 16; /* default return code */ cmd->SCp.phase = 0; /* bus free */ - dev->ppa_tq.data = dev; - schedule_work(&dev->ppa_tq); + schedule_delayed_work(&dev->ppa_tq, 0); ppa_pb_claim(dev); @@ -1086,7 +1084,7 @@ static int __ppa_attach(struct parport *pb) else ports = 8; - INIT_WORK(&dev->ppa_tq, ppa_interrupt, dev); + INIT_DELAYED_WORK(&dev->ppa_tq, ppa_interrupt); err = -ENOMEM; host = scsi_host_alloc(&ppa_template, sizeof(ppa_struct *)); diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 5b8db6109536..bbbc9d039baa 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -1011,9 +1011,10 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha, * the mid-level tries to sleep when it reaches the driver threshold * "host->can_queue". This can cause a panic if we were in our interrupt code. **/ -static void qla4xxx_do_dpc(void *data) +static void qla4xxx_do_dpc(struct work_struct *work) { - struct scsi_qla_host *ha = (struct scsi_qla_host *) data; + struct scsi_qla_host *ha = + container_of(work, struct scsi_qla_host, dpc_work); struct ddb_entry *ddb_entry, *dtemp; DEBUG2(printk("scsi%ld: %s: DPC handler waking up.\n", @@ -1315,7 +1316,7 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, ret = -ENODEV; goto probe_failed; } - INIT_WORK(&ha->dpc_work, qla4xxx_do_dpc, ha); + INIT_WORK(&ha->dpc_work, qla4xxx_do_dpc); ret = request_irq(pdev->irq, qla4xxx_intr_handler, SA_INTERRUPT|SA_SHIRQ, "qla4xxx", ha); diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index 38c215a78f69..3571ce8934e7 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -241,9 +241,9 @@ fc_bitfield_name_search(remote_port_roles, fc_remote_port_role_names) #define FC_MGMTSRVR_PORTID 0x00000a -static void fc_timeout_deleted_rport(void *data); -static void fc_timeout_fail_rport_io(void *data); -static void fc_scsi_scan_rport(void *data); +static void fc_timeout_deleted_rport(struct work_struct *work); +static void fc_timeout_fail_rport_io(struct work_struct *work); +static void fc_scsi_scan_rport(struct work_struct *work); /* * Attribute counts pre object type... @@ -1613,7 +1613,7 @@ fc_flush_work(struct Scsi_Host *shost) * 1 on success / 0 already queued / < 0 for error **/ static int -fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work, +fc_queue_devloss_work(struct Scsi_Host *shost, struct delayed_work *work, unsigned long delay) { if (unlikely(!fc_host_devloss_work_q(shost))) { @@ -1625,9 +1625,6 @@ fc_queue_devloss_work(struct Scsi_Host *shost, struct work_struct *work, return -EINVAL; } - if (delay == 0) - return queue_work(fc_host_devloss_work_q(shost), work); - return queue_delayed_work(fc_host_devloss_work_q(shost), work, delay); } @@ -1712,12 +1709,13 @@ EXPORT_SYMBOL(fc_remove_host); * fc_starget_delete - called to delete the scsi decendents of an rport * (target and all sdevs) * - * @data: remote port to be operated on. + * @work: remote port to be operated on. **/ static void -fc_starget_delete(void *data) +fc_starget_delete(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, stgt_delete_work); struct Scsi_Host *shost = rport_to_shost(rport); unsigned long flags; struct fc_internal *i = to_fc_internal(shost->transportt); @@ -1751,12 +1749,13 @@ fc_starget_delete(void *data) /** * fc_rport_final_delete - finish rport termination and delete it. * - * @data: remote port to be deleted. + * @work: remote port to be deleted. **/ static void -fc_rport_final_delete(void *data) +fc_rport_final_delete(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, rport_delete_work); struct device *dev = &rport->dev; struct Scsi_Host *shost = rport_to_shost(rport); struct fc_internal *i = to_fc_internal(shost->transportt); @@ -1770,7 +1769,7 @@ fc_rport_final_delete(void *data) /* Delete SCSI target and sdevs */ if (rport->scsi_target_id != -1) - fc_starget_delete(data); + fc_starget_delete(&rport->stgt_delete_work); else if (i->f->dev_loss_tmo_callbk) i->f->dev_loss_tmo_callbk(rport); else if (i->f->terminate_rport_io) @@ -1829,11 +1828,11 @@ fc_rport_create(struct Scsi_Host *shost, int channel, rport->channel = channel; rport->fast_io_fail_tmo = -1; - INIT_WORK(&rport->dev_loss_work, fc_timeout_deleted_rport, rport); - INIT_WORK(&rport->fail_io_work, fc_timeout_fail_rport_io, rport); - INIT_WORK(&rport->scan_work, fc_scsi_scan_rport, rport); - INIT_WORK(&rport->stgt_delete_work, fc_starget_delete, rport); - INIT_WORK(&rport->rport_delete_work, fc_rport_final_delete, rport); + INIT_DELAYED_WORK(&rport->dev_loss_work, fc_timeout_deleted_rport); + INIT_DELAYED_WORK(&rport->fail_io_work, fc_timeout_fail_rport_io); + INIT_WORK(&rport->scan_work, fc_scsi_scan_rport); + INIT_WORK(&rport->stgt_delete_work, fc_starget_delete); + INIT_WORK(&rport->rport_delete_work, fc_rport_final_delete); spin_lock_irqsave(shost->host_lock, flags); @@ -1963,7 +1962,7 @@ fc_remote_port_add(struct Scsi_Host *shost, int channel, } if (match) { - struct work_struct *work = + struct delayed_work *work = &rport->dev_loss_work; memcpy(&rport->node_name, &ids->node_name, @@ -2267,12 +2266,13 @@ EXPORT_SYMBOL(fc_remote_port_rolechg); * was a SCSI target (thus was blocked), and failed * to return in the alloted time. * - * @data: rport target that failed to reappear in the alloted time. + * @work: rport target that failed to reappear in the alloted time. **/ static void -fc_timeout_deleted_rport(void *data) +fc_timeout_deleted_rport(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, dev_loss_work.work); struct Scsi_Host *shost = rport_to_shost(rport); struct fc_host_attrs *fc_host = shost_to_fc_host(shost); unsigned long flags; @@ -2366,15 +2366,16 @@ fc_timeout_deleted_rport(void *data) * fc_timeout_fail_rport_io - Timeout handler for a fast io failing on a * disconnected SCSI target. * - * @data: rport to terminate io on. + * @work: rport to terminate io on. * * Notes: Only requests the failure of the io, not that all are flushed * prior to returning. **/ static void -fc_timeout_fail_rport_io(void *data) +fc_timeout_fail_rport_io(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, fail_io_work.work); struct Scsi_Host *shost = rport_to_shost(rport); struct fc_internal *i = to_fc_internal(shost->transportt); @@ -2387,12 +2388,13 @@ fc_timeout_fail_rport_io(void *data) /** * fc_scsi_scan_rport - called to perform a scsi scan on a remote port. * - * @data: remote port to be scanned. + * @work: remote port to be scanned. **/ static void -fc_scsi_scan_rport(void *data) +fc_scsi_scan_rport(struct work_struct *work) { - struct fc_rport *rport = (struct fc_rport *)data; + struct fc_rport *rport = + container_of(work, struct fc_rport, scan_work); struct Scsi_Host *shost = rport_to_shost(rport); unsigned long flags; diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 9b25124a989e..9c22f1342715 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -234,9 +234,11 @@ static int iscsi_user_scan(struct Scsi_Host *shost, uint channel, return 0; } -static void session_recovery_timedout(void *data) +static void session_recovery_timedout(struct work_struct *work) { - struct iscsi_cls_session *session = data; + struct iscsi_cls_session *session = + container_of(work, struct iscsi_cls_session, + recovery_work.work); dev_printk(KERN_INFO, &session->dev, "iscsi: session recovery timed " "out after %d secs\n", session->recovery_tmo); @@ -276,7 +278,7 @@ iscsi_alloc_session(struct Scsi_Host *shost, session->transport = transport; session->recovery_tmo = 120; - INIT_WORK(&session->recovery_work, session_recovery_timedout, session); + INIT_DELAYED_WORK(&session->recovery_work, session_recovery_timedout); INIT_LIST_HEAD(&session->host_list); INIT_LIST_HEAD(&session->sess_list); diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c index 9f070f0d0f2b..3fded4831460 100644 --- a/drivers/scsi/scsi_transport_spi.c +++ b/drivers/scsi/scsi_transport_spi.c @@ -964,9 +964,10 @@ struct work_queue_wrapper { }; static void -spi_dv_device_work_wrapper(void *data) +spi_dv_device_work_wrapper(struct work_struct *work) { - struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data; + struct work_queue_wrapper *wqw = + container_of(work, struct work_queue_wrapper, work); struct scsi_device *sdev = wqw->sdev; kfree(wqw); @@ -1006,7 +1007,7 @@ spi_schedule_dv_device(struct scsi_device *sdev) return; } - INIT_WORK(&wqw->work, spi_dv_device_work_wrapper, wqw); + INIT_WORK(&wqw->work, spi_dv_device_work_wrapper); wqw->sdev = sdev; schedule_work(&wqw->work); diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c index a23862ef72b2..08c1c57c6128 100644 --- a/drivers/spi/spi_bitbang.c +++ b/drivers/spi/spi_bitbang.c @@ -265,9 +265,10 @@ static int spi_bitbang_bufs(struct spi_device *spi, struct spi_transfer *t) * Drivers can provide word-at-a-time i/o primitives, or provide * transfer-at-a-time ones to leverage dma or fifo hardware. */ -static void bitbang_work(void *_bitbang) +static void bitbang_work(struct work_struct *work) { - struct spi_bitbang *bitbang = _bitbang; + struct spi_bitbang *bitbang = + container_of(work, struct spi_bitbang, work); unsigned long flags; spin_lock_irqsave(&bitbang->lock, flags); @@ -456,7 +457,7 @@ int spi_bitbang_start(struct spi_bitbang *bitbang) if (!bitbang->master || !bitbang->chipselect) return -EINVAL; - INIT_WORK(&bitbang->work, bitbang_work, bitbang); + INIT_WORK(&bitbang->work, bitbang_work); spin_lock_init(&bitbang->lock); INIT_LIST_HEAD(&bitbang->queue); diff --git a/drivers/usb/atm/cxacru.c b/drivers/usb/atm/cxacru.c index e6565633ba0f..3dfa3e40e148 100644 --- a/drivers/usb/atm/cxacru.c +++ b/drivers/usb/atm/cxacru.c @@ -158,7 +158,7 @@ struct cxacru_data { const struct cxacru_modem_type *modem_type; int line_status; - struct work_struct poll_work; + struct delayed_work poll_work; /* contol handles */ struct mutex cm_serialize; @@ -347,7 +347,7 @@ static int cxacru_card_status(struct cxacru_data *instance) return 0; } -static void cxacru_poll_status(struct cxacru_data *instance); +static void cxacru_poll_status(struct work_struct *work); static int cxacru_atm_start(struct usbatm_data *usbatm_instance, struct atm_dev *atm_dev) @@ -376,12 +376,14 @@ static int cxacru_atm_start(struct usbatm_data *usbatm_instance, } /* Start status polling */ - cxacru_poll_status(instance); + cxacru_poll_status(&instance->poll_work.work); return 0; } -static void cxacru_poll_status(struct cxacru_data *instance) +static void cxacru_poll_status(struct work_struct *work) { + struct cxacru_data *instance = + container_of(work, struct cxacru_data, poll_work.work); u32 buf[CXINF_MAX] = {}; struct usbatm_data *usbatm = instance->usbatm; struct atm_dev *atm_dev = usbatm->atm_dev; @@ -720,7 +722,7 @@ static int cxacru_bind(struct usbatm_data *usbatm_instance, mutex_init(&instance->cm_serialize); - INIT_WORK(&instance->poll_work, (void *)cxacru_poll_status, instance); + INIT_DELAYED_WORK(&instance->poll_work, cxacru_poll_status); usbatm_instance->driver_data = instance; diff --git a/drivers/usb/atm/speedtch.c b/drivers/usb/atm/speedtch.c index c870c804470f..7ed34bb1c50f 100644 --- a/drivers/usb/atm/speedtch.c +++ b/drivers/usb/atm/speedtch.c @@ -142,7 +142,7 @@ struct speedtch_instance_data { struct speedtch_params params; /* set in probe, constant afterwards */ - struct work_struct status_checker; + struct delayed_work status_checker; unsigned char last_status; @@ -498,8 +498,11 @@ static int speedtch_start_synchro(struct speedtch_instance_data *instance) return ret; } -static void speedtch_check_status(struct speedtch_instance_data *instance) +static void speedtch_check_status(struct work_struct *work) { + struct speedtch_instance_data *instance = + container_of(work, struct speedtch_instance_data, + status_checker.work); struct usbatm_data *usbatm = instance->usbatm; struct atm_dev *atm_dev = usbatm->atm_dev; unsigned char *buf = instance->scratch_buffer; @@ -576,7 +579,7 @@ static void speedtch_status_poll(unsigned long data) { struct speedtch_instance_data *instance = (void *)data; - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); /* The following check is racy, but the race is harmless */ if (instance->poll_delay < MAX_POLL_DELAY) @@ -596,7 +599,7 @@ static void speedtch_resubmit_int(unsigned long data) if (int_urb) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); if (!ret) - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); else { atm_dbg(instance->usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); mod_timer(&instance->resubmit_timer, jiffies + msecs_to_jiffies(RESUBMIT_DELAY)); @@ -640,7 +643,7 @@ static void speedtch_handle_int(struct urb *int_urb) if ((int_urb = instance->int_urb)) { ret = usb_submit_urb(int_urb, GFP_ATOMIC); - schedule_work(&instance->status_checker); + schedule_delayed_work(&instance->status_checker, 0); if (ret < 0) { atm_dbg(usbatm, "%s: usb_submit_urb failed with result %d\n", __func__, ret); goto fail; @@ -855,7 +858,7 @@ static int speedtch_bind(struct usbatm_data *usbatm, usbatm->flags |= (use_isoc ? UDSL_USE_ISOC : 0); - INIT_WORK(&instance->status_checker, (void *)speedtch_check_status, instance); + INIT_DELAYED_WORK(&instance->status_checker, speedtch_check_status); instance->status_checker.timer.function = speedtch_status_poll; instance->status_checker.timer.data = (unsigned long)instance; diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index f6b9f7e1f716..e39bb09f5af9 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -658,9 +658,9 @@ static int request_dsp(struct uea_softc *sc) /* * The uea_load_page() function must be called within a process context */ -static void uea_load_page(void *xsc) +static void uea_load_page(struct work_struct *work) { - struct uea_softc *sc = xsc; + struct uea_softc *sc = container_of(work, struct uea_softc, task); u16 pageno = sc->pageno; u16 ovl = sc->ovl; struct block_info bi; @@ -1352,7 +1352,7 @@ static int uea_boot(struct uea_softc *sc) uea_enters(INS_TO_USBDEV(sc)); - INIT_WORK(&sc->task, uea_load_page, sc); + INIT_WORK(&sc->task, uea_load_page); init_waitqueue_head(&sc->sync_q); init_waitqueue_head(&sc->cmv_ack_wait); diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 9a9012fd284b..6408e10fdbf8 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -421,9 +421,9 @@ static void acm_write_bulk(struct urb *urb) schedule_work(&acm->work); } -static void acm_softint(void *private) +static void acm_softint(struct work_struct *work) { - struct acm *acm = private; + struct acm *acm = container_of(work, struct acm, work); dbg("Entering acm_softint."); if (!ACM_READY(acm)) @@ -927,7 +927,7 @@ skip_normal_probe: acm->rx_buflimit = num_rx_buf; acm->urb_task.func = acm_rx_tasklet; acm->urb_task.data = (unsigned long) acm; - INIT_WORK(&acm->work, acm_softint, acm); + INIT_WORK(&acm->work, acm_softint); spin_lock_init(&acm->throttle_lock); spin_lock_init(&acm->write_lock); spin_lock_init(&acm->read_lock); diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index ba165aff9ea4..ad0ffbe8f7d7 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -167,9 +167,10 @@ static void set_port_led( #define LED_CYCLE_PERIOD ((2*HZ)/3) -static void led_work (void *__hub) +static void led_work (struct work_struct *work) { - struct usb_hub *hub = __hub; + struct usb_hub *hub = + container_of(work, struct usb_hub, leds.work); struct usb_device *hdev = hub->hdev; unsigned i; unsigned changed = 0; @@ -351,9 +352,10 @@ hub_clear_tt_buffer (struct usb_device *hdev, u16 devinfo, u16 tt) * talking to TTs must queue control transfers (not just bulk and iso), so * both can talk to the same hub concurrently. */ -static void hub_tt_kevent (void *arg) +static void hub_tt_kevent (struct work_struct *work) { - struct usb_hub *hub = arg; + struct usb_hub *hub = + container_of(work, struct usb_hub, tt.kevent); unsigned long flags; spin_lock_irqsave (&hub->tt.lock, flags); @@ -641,7 +643,7 @@ static int hub_configure(struct usb_hub *hub, spin_lock_init (&hub->tt.lock); INIT_LIST_HEAD (&hub->tt.clear_list); - INIT_WORK (&hub->tt.kevent, hub_tt_kevent, hub); + INIT_WORK (&hub->tt.kevent, hub_tt_kevent); switch (hdev->descriptor.bDeviceProtocol) { case 0: break; @@ -880,7 +882,7 @@ descriptor_error: INIT_LIST_HEAD(&hub->event_list); hub->intfdev = &intf->dev; hub->hdev = hdev; - INIT_WORK(&hub->leds, led_work, hub); + INIT_DELAYED_WORK(&hub->leds, led_work); usb_set_intfdata (intf, hub); @@ -2281,7 +2283,7 @@ check_highspeed (struct usb_hub *hub, struct usb_device *udev, int port1) /* hub LEDs are probably harder to miss than syslog */ if (hub->has_indicators) { hub->indicator[port1-1] = INDICATOR_GREEN_BLINK; - schedule_work (&hub->leds); + schedule_delayed_work (&hub->leds, 0); } } kfree(qual); @@ -2455,7 +2457,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1, if (hub->has_indicators) { hub->indicator[port1-1] = INDICATOR_AMBER_BLINK; - schedule_work (&hub->leds); + schedule_delayed_work (&hub->leds, 0); } status = -ENOTCONN; /* Don't retry */ goto loop_disable; diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h index 0f8e82a4d480..035d3ef35888 100644 --- a/drivers/usb/core/hub.h +++ b/drivers/usb/core/hub.h @@ -230,7 +230,7 @@ struct usb_hub { unsigned has_indicators:1; enum hub_led_mode indicator[USB_MAXCHILDREN]; - struct work_struct leds; + struct delayed_work leds; }; #endif /* __LINUX_HUB_H */ diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 7729c0744886..89572bc021b1 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -1501,9 +1501,10 @@ struct set_config_request { }; /* Worker routine for usb_driver_set_configuration() */ -static void driver_set_config_work(void *_req) +static void driver_set_config_work(struct work_struct *work) { - struct set_config_request *req = _req; + struct set_config_request *req = + container_of(work, struct set_config_request, work); usb_lock_device(req->udev); usb_set_configuration(req->udev, req->config); @@ -1541,7 +1542,7 @@ int usb_driver_set_configuration(struct usb_device *udev, int config) return -ENOMEM; req->udev = udev; req->config = config; - INIT_WORK(&req->work, driver_set_config_work, req); + INIT_WORK(&req->work, driver_set_config_work); usb_get_dev(udev); if (!schedule_work(&req->work)) { diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 467cb02832f3..ab2f68fc7d2d 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -210,9 +210,10 @@ static void ksuspend_usb_cleanup(void) #ifdef CONFIG_USB_SUSPEND /* usb_autosuspend_work - callback routine to autosuspend a USB device */ -static void usb_autosuspend_work(void *_udev) +static void usb_autosuspend_work(struct work_struct *work) { - struct usb_device *udev = _udev; + struct usb_device *udev = + container_of(work, struct usb_device, autosuspend.work); usb_pm_lock(udev); udev->auto_pm = 1; @@ -222,7 +223,7 @@ static void usb_autosuspend_work(void *_udev) #else -static void usb_autosuspend_work(void *_udev) +static void usb_autosuspend_work(struct work_struct *work) {} #endif @@ -304,7 +305,7 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1) #ifdef CONFIG_PM mutex_init(&dev->pm_mutex); - INIT_WORK(&dev->autosuspend, usb_autosuspend_work, dev); + INIT_DELAYED_WORK(&dev->autosuspend, usb_autosuspend_work); #endif return dev; } diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 1c17d26d03b8..107119c54301 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -1833,9 +1833,9 @@ static void rx_fill (struct eth_dev *dev, gfp_t gfp_flags) spin_unlock_irqrestore(&dev->req_lock, flags); } -static void eth_work (void *_dev) +static void eth_work (struct work_struct *work) { - struct eth_dev *dev = _dev; + struct eth_dev *dev = container_of(work, struct eth_dev, work); if (test_and_clear_bit (WORK_RX_MEMORY, &dev->todo)) { if (netif_running (dev->net)) @@ -2398,7 +2398,7 @@ autoconf_fail: dev = netdev_priv(net); spin_lock_init (&dev->lock); spin_lock_init (&dev->req_lock); - INIT_WORK (&dev->work, eth_work, dev); + INIT_WORK (&dev->work, eth_work); INIT_LIST_HEAD (&dev->tx_reqs); INIT_LIST_HEAD (&dev->rx_reqs); diff --git a/drivers/usb/host/u132-hcd.c b/drivers/usb/host/u132-hcd.c index 32c635ecbf31..4f95a249c913 100644 --- a/drivers/usb/host/u132-hcd.c +++ b/drivers/usb/host/u132-hcd.c @@ -163,7 +163,7 @@ struct u132_endp { u16 queue_next; struct urb *urb_list[ENDP_QUEUE_SIZE]; struct list_head urb_more; - struct work_struct scheduler; + struct delayed_work scheduler; }; struct u132_ring { unsigned in_use:1; @@ -171,7 +171,7 @@ struct u132_ring { u8 number; struct u132 *u132; struct u132_endp *curr_endp; - struct work_struct scheduler; + struct delayed_work scheduler; }; #define OHCI_QUIRK_AMD756 0x01 #define OHCI_QUIRK_SUPERIO 0x02 @@ -198,7 +198,7 @@ struct u132 { u32 hc_roothub_portstatus[MAX_ROOT_PORTS]; int flags; unsigned long next_statechange; - struct work_struct monitor; + struct delayed_work monitor; int num_endpoints; struct u132_addr addr[MAX_U132_ADDRS]; struct u132_udev udev[MAX_U132_UDEVS]; @@ -314,7 +314,7 @@ static void u132_ring_requeue_work(struct u132 *u132, struct u132_ring *ring, if (delta > 0) { if (queue_delayed_work(workqueue, &ring->scheduler, delta)) return; - } else if (queue_work(workqueue, &ring->scheduler)) + } else if (queue_delayed_work(workqueue, &ring->scheduler, 0)) return; kref_put(&u132->kref, u132_hcd_delete); return; @@ -393,12 +393,8 @@ static inline void u132_endp_init_kref(struct u132 *u132, static void u132_endp_queue_work(struct u132 *u132, struct u132_endp *endp, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &endp->scheduler, delta)) - kref_get(&endp->kref); - } else if (queue_work(workqueue, &endp->scheduler)) - kref_get(&endp->kref); - return; + if (queue_delayed_work(workqueue, &endp->scheduler, delta)) + kref_get(&endp->kref); } static void u132_endp_cancel_work(struct u132 *u132, struct u132_endp *endp) @@ -414,24 +410,14 @@ static inline void u132_monitor_put_kref(struct u132 *u132) static void u132_monitor_queue_work(struct u132 *u132, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &u132->monitor, delta)) { - kref_get(&u132->kref); - } - } else if (queue_work(workqueue, &u132->monitor)) - kref_get(&u132->kref); - return; + if (queue_delayed_work(workqueue, &u132->monitor, delta)) + kref_get(&u132->kref); } static void u132_monitor_requeue_work(struct u132 *u132, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(workqueue, &u132->monitor, delta)) - return; - } else if (queue_work(workqueue, &u132->monitor)) - return; - kref_put(&u132->kref, u132_hcd_delete); - return; + if (!queue_delayed_work(workqueue, &u132->monitor, delta)) + kref_put(&u132->kref, u132_hcd_delete); } static void u132_monitor_cancel_work(struct u132 *u132) @@ -493,9 +479,9 @@ static int read_roothub_info(struct u132 *u132) return 0; } -static void u132_hcd_monitor_work(void *data) +static void u132_hcd_monitor_work(struct work_struct *work) { - struct u132 *u132 = data; + struct u132 *u132 = container_of(work, struct u132, monitor.work); if (u132->going > 1) { dev_err(&u132->platform_dev->dev, "device has been removed %d\n" , u132->going); @@ -1319,15 +1305,14 @@ static void u132_hcd_initial_setup_sent(void *data, struct urb *urb, u8 *buf, } } -static void u132_hcd_ring_work_scheduler(void *data); -static void u132_hcd_endp_work_scheduler(void *data); /* * this work function is only executed from the work queue * */ -static void u132_hcd_ring_work_scheduler(void *data) +static void u132_hcd_ring_work_scheduler(struct work_struct *work) { - struct u132_ring *ring = data; + struct u132_ring *ring = + container_of(work, struct u132_ring, scheduler.work); struct u132 *u132 = ring->u132; down(&u132->scheduler_lock); if (ring->in_use) { @@ -1386,10 +1371,11 @@ static void u132_hcd_ring_work_scheduler(void *data) } } -static void u132_hcd_endp_work_scheduler(void *data) +static void u132_hcd_endp_work_scheduler(struct work_struct *work) { struct u132_ring *ring; - struct u132_endp *endp = data; + struct u132_endp *endp = + container_of(work, struct u132_endp, scheduler.work); struct u132 *u132 = endp->u132; down(&u132->scheduler_lock); ring = endp->ring; @@ -1947,7 +1933,7 @@ static int create_endpoint_and_queue_int(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); ring = endp->ring = &u132->ring[0]; @@ -2036,7 +2022,7 @@ static int create_endpoint_and_queue_bulk(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); endp->dequeueing = 0; @@ -2121,7 +2107,7 @@ static int create_endpoint_and_queue_control(struct u132 *u132, if (!endp) { return -ENOMEM; } - INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp); + INIT_DELAYED_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler); spin_lock_init(&endp->queue_lock.slock); INIT_LIST_HEAD(&endp->urb_more); ring = endp->ring = &u132->ring[0]; @@ -3100,10 +3086,10 @@ static void u132_initialise(struct u132 *u132, struct platform_device *pdev) ring->number = rings + 1; ring->length = 0; ring->curr_endp = NULL; - INIT_WORK(&ring->scheduler, u132_hcd_ring_work_scheduler, - (void *)ring); + INIT_DELAYED_WORK(&ring->scheduler, + u132_hcd_ring_work_scheduler); } down(&u132->sw_lock); - INIT_WORK(&u132->monitor, u132_hcd_monitor_work, (void *)u132); + INIT_DELAYED_WORK(&u132->monitor, u132_hcd_monitor_work); while (ports-- > 0) { struct u132_port *port = &u132->port[ports]; port->u132 = u132; diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 6d08a3bcc952..ebc9e823a46e 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -969,9 +969,10 @@ static void hid_retry_timeout(unsigned long _hid) } /* Workqueue routine to reset the device */ -static void hid_reset(void *_hid) +static void hid_reset(struct work_struct *work) { - struct hid_device *hid = (struct hid_device *) _hid; + struct hid_device *hid = + container_of(work, struct hid_device, reset_work); int rc_lock, rc; dev_dbg(&hid->intf->dev, "resetting device\n"); @@ -2015,7 +2016,7 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) init_waitqueue_head(&hid->wait); - INIT_WORK(&hid->reset_work, hid_reset, hid); + INIT_WORK(&hid->reset_work, hid_reset); setup_timer(&hid->io_retry, hid_retry_timeout, (unsigned long) hid); spin_lock_init(&hid->inlock); diff --git a/drivers/usb/misc/ftdi-elan.c b/drivers/usb/misc/ftdi-elan.c index 9b591b8b9369..e4e2cf2ba915 100644 --- a/drivers/usb/misc/ftdi-elan.c +++ b/drivers/usb/misc/ftdi-elan.c @@ -156,9 +156,9 @@ struct usb_ftdi { struct usb_device *udev; struct usb_interface *interface; struct usb_class_driver *class; - struct work_struct status_work; - struct work_struct command_work; - struct work_struct respond_work; + struct delayed_work status_work; + struct delayed_work command_work; + struct delayed_work respond_work; struct u132_platform_data platform_data; struct resource resources[0]; struct platform_device platform_dev; @@ -210,23 +210,14 @@ static void ftdi_elan_init_kref(struct usb_ftdi *ftdi) static void ftdi_status_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) - return; - } else if (queue_work(status_queue, &ftdi->status_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(status_queue, &ftdi->status_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_status_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) - kref_get(&ftdi->kref); - } else if (queue_work(status_queue, &ftdi->status_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(status_queue, &ftdi->status_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_status_cancel_work(struct usb_ftdi *ftdi) @@ -237,25 +228,14 @@ static void ftdi_status_cancel_work(struct usb_ftdi *ftdi) static void ftdi_command_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(command_queue, &ftdi->command_work, - delta)) - return; - } else if (queue_work(command_queue, &ftdi->command_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(command_queue, &ftdi->command_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_command_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(command_queue, &ftdi->command_work, - delta)) - kref_get(&ftdi->kref); - } else if (queue_work(command_queue, &ftdi->command_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(command_queue, &ftdi->command_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_command_cancel_work(struct usb_ftdi *ftdi) @@ -267,25 +247,14 @@ static void ftdi_command_cancel_work(struct usb_ftdi *ftdi) static void ftdi_response_requeue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(respond_queue, &ftdi->respond_work, - delta)) - return; - } else if (queue_work(respond_queue, &ftdi->respond_work)) - return; - kref_put(&ftdi->kref, ftdi_elan_delete); - return; + if (!queue_delayed_work(respond_queue, &ftdi->respond_work, delta)) + kref_put(&ftdi->kref, ftdi_elan_delete); } static void ftdi_respond_queue_work(struct usb_ftdi *ftdi, unsigned int delta) { - if (delta > 0) { - if (queue_delayed_work(respond_queue, &ftdi->respond_work, - delta)) - kref_get(&ftdi->kref); - } else if (queue_work(respond_queue, &ftdi->respond_work)) - kref_get(&ftdi->kref); - return; + if (queue_delayed_work(respond_queue, &ftdi->respond_work, delta)) + kref_get(&ftdi->kref); } static void ftdi_response_cancel_work(struct usb_ftdi *ftdi) @@ -475,9 +444,11 @@ static void ftdi_elan_kick_command_queue(struct usb_ftdi *ftdi) return; } -static void ftdi_elan_command_work(void *data) +static void ftdi_elan_command_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, command_work.work); + if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); return; @@ -500,9 +471,10 @@ static void ftdi_elan_kick_respond_queue(struct usb_ftdi *ftdi) return; } -static void ftdi_elan_respond_work(void *data) +static void ftdi_elan_respond_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, respond_work.work); if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); return; @@ -534,9 +506,10 @@ static void ftdi_elan_respond_work(void *data) * after the FTDI has been synchronized * */ -static void ftdi_elan_status_work(void *data) +static void ftdi_elan_status_work(struct work_struct *work) { - struct usb_ftdi *ftdi = data; + struct usb_ftdi *ftdi = + container_of(work, struct usb_ftdi, status_work.work); int work_delay_in_msec = 0; if (ftdi->disconnected > 0) { ftdi_elan_put_kref(ftdi); @@ -2691,12 +2664,9 @@ static int ftdi_elan_probe(struct usb_interface *interface, ftdi->class = NULL; dev_info(&ftdi->udev->dev, "USB FDTI=%p ELAN interface %d now a" "ctivated\n", ftdi, iface_desc->desc.bInterfaceNumber); - INIT_WORK(&ftdi->status_work, ftdi_elan_status_work, - (void *)ftdi); - INIT_WORK(&ftdi->command_work, ftdi_elan_command_work, - (void *)ftdi); - INIT_WORK(&ftdi->respond_work, ftdi_elan_respond_work, - (void *)ftdi); + INIT_DELAYED_WORK(&ftdi->status_work, ftdi_elan_status_work); + INIT_DELAYED_WORK(&ftdi->command_work, ftdi_elan_command_work); + INIT_DELAYED_WORK(&ftdi->respond_work, ftdi_elan_respond_work); ftdi_status_queue_work(ftdi, msecs_to_jiffies(3 *1000)); return 0; } else { diff --git a/drivers/usb/misc/phidgetkit.c b/drivers/usb/misc/phidgetkit.c index abb4dcd811ac..33e716c6a79b 100644 --- a/drivers/usb/misc/phidgetkit.c +++ b/drivers/usb/misc/phidgetkit.c @@ -81,8 +81,8 @@ struct interfacekit { unsigned char *data; dma_addr_t data_dma; - struct work_struct do_notify; - struct work_struct do_resubmit; + struct delayed_work do_notify; + struct delayed_work do_resubmit; unsigned long input_events; unsigned long sensor_events; }; @@ -374,7 +374,7 @@ static void interfacekit_irq(struct urb *urb) } if (kit->input_events || kit->sensor_events) - schedule_work(&kit->do_notify); + schedule_delayed_work(&kit->do_notify, 0); resubmit: status = usb_submit_urb(urb, SLAB_ATOMIC); @@ -384,9 +384,10 @@ resubmit: kit->udev->devpath, status); } -static void do_notify(void *data) +static void do_notify(struct work_struct *work) { - struct interfacekit *kit = data; + struct interfacekit *kit = + container_of(work, struct interfacekit, do_notify.work); int i; char sysfs_file[8]; @@ -405,9 +406,11 @@ static void do_notify(void *data) } } -static void do_resubmit(void *data) +static void do_resubmit(struct work_struct *work) { - set_outputs(data); + struct interfacekit *kit = + container_of(work, struct interfacekit, do_resubmit.work); + set_outputs(kit); } #define show_set_output(value) \ @@ -575,8 +578,8 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic kit->udev = usb_get_dev(dev); kit->intf = intf; - INIT_WORK(&kit->do_notify, do_notify, kit); - INIT_WORK(&kit->do_resubmit, do_resubmit, kit); + INIT_DELAYED_WORK(&kit->do_notify, do_notify); + INIT_DELAYED_WORK(&kit->do_resubmit, do_resubmit); usb_fill_int_urb(kit->irq, kit->udev, pipe, kit->data, maxp > URB_INT_SIZE ? URB_INT_SIZE : maxp, interfacekit_irq, kit, endpoint->bInterval); diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c index 5c780cab92e0..0385ffcc7419 100644 --- a/drivers/usb/misc/phidgetmotorcontrol.c +++ b/drivers/usb/misc/phidgetmotorcontrol.c @@ -41,7 +41,7 @@ struct motorcontrol { unsigned char *data; dma_addr_t data_dma; - struct work_struct do_notify; + struct delayed_work do_notify; unsigned long input_events; unsigned long speed_events; unsigned long exceed_events; @@ -148,7 +148,7 @@ static void motorcontrol_irq(struct urb *urb) set_bit(1, &mc->exceed_events); if (mc->input_events || mc->exceed_events || mc->speed_events) - schedule_work(&mc->do_notify); + schedule_delayed_work(&mc->do_notify, 0); resubmit: status = usb_submit_urb(urb, SLAB_ATOMIC); @@ -159,9 +159,10 @@ resubmit: mc->udev->devpath, status); } -static void do_notify(void *data) +static void do_notify(struct work_struct *work) { - struct motorcontrol *mc = data; + struct motorcontrol *mc = + container_of(work, struct motorcontrol, do_notify.work); int i; char sysfs_file[8]; @@ -348,7 +349,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic mc->udev = usb_get_dev(dev); mc->intf = intf; mc->acceleration[0] = mc->acceleration[1] = 10; - INIT_WORK(&mc->do_notify, do_notify, mc); + INIT_DELAYED_WORK(&mc->do_notify, do_notify); usb_fill_int_urb(mc->irq, mc->udev, pipe, mc->data, maxp > URB_INT_SIZE ? URB_INT_SIZE : maxp, motorcontrol_irq, mc, endpoint->bInterval); diff --git a/drivers/usb/net/kaweth.c b/drivers/usb/net/kaweth.c index 7c906a43e497..fa78326d0bf0 100644 --- a/drivers/usb/net/kaweth.c +++ b/drivers/usb/net/kaweth.c @@ -222,7 +222,7 @@ struct kaweth_device int suspend_lowmem_ctrl; int linkstate; int opened; - struct work_struct lowmem_work; + struct delayed_work lowmem_work; struct usb_device *dev; struct net_device *net; @@ -530,9 +530,10 @@ resubmit: kaweth_resubmit_int_urb(kaweth, GFP_ATOMIC); } -static void kaweth_resubmit_tl(void *d) +static void kaweth_resubmit_tl(struct work_struct *work) { - struct kaweth_device *kaweth = (struct kaweth_device *)d; + struct kaweth_device *kaweth = + container_of(work, struct kaweth_device, lowmem_work.work); if (IS_BLOCKED(kaweth->status)) return; @@ -1126,7 +1127,7 @@ err_fw: /* kaweth is zeroed as part of alloc_netdev */ - INIT_WORK(&kaweth->lowmem_work, kaweth_resubmit_tl, (void *)kaweth); + INIT_DELAYED_WORK(&kaweth->lowmem_work, kaweth_resubmit_tl); SET_MODULE_OWNER(netdev); diff --git a/drivers/usb/net/pegasus.c b/drivers/usb/net/pegasus.c index 33abbd2176b6..78cf6f091285 100644 --- a/drivers/usb/net/pegasus.c +++ b/drivers/usb/net/pegasus.c @@ -1280,9 +1280,9 @@ static inline void setup_pegasus_II(pegasus_t * pegasus) static struct workqueue_struct *pegasus_workqueue = NULL; #define CARRIER_CHECK_DELAY (2 * HZ) -static void check_carrier(void *data) +static void check_carrier(struct work_struct *work) { - pegasus_t *pegasus = data; + pegasus_t *pegasus = container_of(work, pegasus_t, carrier_check.work); set_carrier(pegasus->net); if (!(pegasus->flags & PEGASUS_UNPLUG)) { queue_delayed_work(pegasus_workqueue, &pegasus->carrier_check, @@ -1318,7 +1318,7 @@ static int pegasus_probe(struct usb_interface *intf, tasklet_init(&pegasus->rx_tl, rx_fixup, (unsigned long) pegasus); - INIT_WORK(&pegasus->carrier_check, check_carrier, pegasus); + INIT_DELAYED_WORK(&pegasus->carrier_check, check_carrier); pegasus->intf = intf; pegasus->usb = dev; diff --git a/drivers/usb/net/pegasus.h b/drivers/usb/net/pegasus.h index 006438069b66..98f6898cae1f 100644 --- a/drivers/usb/net/pegasus.h +++ b/drivers/usb/net/pegasus.h @@ -95,7 +95,7 @@ typedef struct pegasus { int dev_index; int intr_interval; struct tasklet_struct rx_tl; - struct work_struct carrier_check; + struct delayed_work carrier_check; struct urb *ctrl_urb, *rx_urb, *tx_urb, *intr_urb; struct sk_buff *rx_pool[RX_SKBS]; struct sk_buff *rx_skb; diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 760b5327b81b..79b5474fe234 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -782,9 +782,10 @@ static struct ethtool_ops usbnet_ethtool_ops = { * especially now that control transfers can be queued. */ static void -kevent (void *data) +kevent (struct work_struct *work) { - struct usbnet *dev = data; + struct usbnet *dev = + container_of(work, struct usbnet, kevent); int status; /* usb_clear_halt() needs a thread context */ @@ -1146,7 +1147,7 @@ usbnet_probe (struct usb_interface *udev, const struct usb_device_id *prod) skb_queue_head_init (&dev->done); dev->bh.func = usbnet_bh; dev->bh.data = (unsigned long) dev; - INIT_WORK (&dev->kevent, kevent, dev); + INIT_WORK (&dev->kevent, kevent); dev->delay.function = usbnet_bh; dev->delay.data = (unsigned long) dev; init_timer (&dev->delay); diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c index 812275509137..2a4ac9bd6a3a 100644 --- a/drivers/usb/serial/aircable.c +++ b/drivers/usb/serial/aircable.c @@ -92,6 +92,7 @@ struct aircable_private { struct circ_buf *rx_buf; /* read buffer */ int rx_flags; /* for throttilng */ struct work_struct rx_work; /* work cue for the receiving line */ + struct usb_serial_port *port; /* USB port with which associated */ }; /* Private methods */ @@ -251,10 +252,11 @@ static void aircable_send(struct usb_serial_port *port) schedule_work(&port->work); } -static void aircable_read(void *params) +static void aircable_read(struct work_struct *work) { - struct usb_serial_port *port = params; - struct aircable_private *priv = usb_get_serial_port_data(port); + struct aircable_private *priv = + container_of(work, struct aircable_private, rx_work); + struct usb_serial_port *port = priv->port; struct tty_struct *tty; unsigned char *data; int count; @@ -348,7 +350,8 @@ static int aircable_attach (struct usb_serial *serial) } priv->rx_flags &= ~(THROTTLED | ACTUALLY_THROTTLED); - INIT_WORK(&priv->rx_work, aircable_read, port); + priv->port = port; + INIT_WORK(&priv->rx_work, aircable_read); usb_set_serial_port_data(serial->port[0], priv); @@ -515,7 +518,7 @@ static void aircable_read_bulk_callback(struct urb *urb) package_length - shift); } } - aircable_read(port); + aircable_read(&priv->rx_work); } /* Schedule the next read _if_ we are still open */ diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index bdb58100fc1d..fd159b040bfb 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -430,13 +430,14 @@ struct digi_port { int dp_in_close; /* close in progress */ wait_queue_head_t dp_close_wait; /* wait queue for close */ struct work_struct dp_wakeup_work; + struct usb_serial_port *dp_port; }; /* Local Function Declarations */ static void digi_wakeup_write( struct usb_serial_port *port ); -static void digi_wakeup_write_lock(void *); +static void digi_wakeup_write_lock(struct work_struct *work); static int digi_write_oob_command( struct usb_serial_port *port, unsigned char *buf, int count, int interruptible ); static int digi_write_inb_command( struct usb_serial_port *port, @@ -598,11 +599,12 @@ static inline long cond_wait_interruptible_timeout_irqrestore( * on writes. */ -static void digi_wakeup_write_lock(void *arg) +static void digi_wakeup_write_lock(struct work_struct *work) { - struct usb_serial_port *port = arg; + struct digi_port *priv = + container_of(work, struct digi_port, dp_wakeup_work); + struct usb_serial_port *port = priv->dp_port; unsigned long flags; - struct digi_port *priv = usb_get_serial_port_data(port); spin_lock_irqsave( &priv->dp_port_lock, flags ); @@ -1702,8 +1704,8 @@ dbg( "digi_startup: TOP" ); init_waitqueue_head( &priv->dp_flush_wait ); priv->dp_in_close = 0; init_waitqueue_head( &priv->dp_close_wait ); - INIT_WORK(&priv->dp_wakeup_work, - digi_wakeup_write_lock, serial->port[i]); + INIT_WORK(&priv->dp_wakeup_work, digi_wakeup_write_lock); + priv->dp_port = serial->port[i]; /* initialize write wait queue for this port */ init_waitqueue_head( &serial->port[i]->write_wait ); diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index c186b4e73c72..88ed5c1d236c 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -559,7 +559,8 @@ struct ftdi_private { char prev_status, diff_status; /* Used for TIOCMIWAIT */ __u8 rx_flags; /* receive state flags (throttling) */ spinlock_t rx_lock; /* spinlock for receive state */ - struct work_struct rx_work; + struct delayed_work rx_work; + struct usb_serial_port *port; int rx_processed; unsigned long rx_bytes; @@ -593,7 +594,7 @@ static int ftdi_write_room (struct usb_serial_port *port); static int ftdi_chars_in_buffer (struct usb_serial_port *port); static void ftdi_write_bulk_callback (struct urb *urb); static void ftdi_read_bulk_callback (struct urb *urb); -static void ftdi_process_read (void *param); +static void ftdi_process_read (struct work_struct *work); static void ftdi_set_termios (struct usb_serial_port *port, struct termios * old); static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file); static int ftdi_tiocmset (struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear); @@ -1201,7 +1202,8 @@ static int ftdi_sio_attach (struct usb_serial *serial) port->read_urb->transfer_buffer_length = BUFSZ; } - INIT_WORK(&priv->rx_work, ftdi_process_read, port); + INIT_DELAYED_WORK(&priv->rx_work, ftdi_process_read); + priv->port = port; /* Free port's existing write urb and transfer buffer. */ if (port->write_urb) { @@ -1641,17 +1643,18 @@ static void ftdi_read_bulk_callback (struct urb *urb) priv->rx_bytes += countread; spin_unlock_irqrestore(&priv->rx_lock, flags); - ftdi_process_read(port); + ftdi_process_read(&priv->rx_work.work); } /* ftdi_read_bulk_callback */ -static void ftdi_process_read (void *param) +static void ftdi_process_read (struct work_struct *work) { /* ftdi_process_read */ - struct usb_serial_port *port = (struct usb_serial_port*)param; + struct ftdi_private *priv = + container_of(work, struct ftdi_private, rx_work.work); + struct usb_serial_port *port = priv->port; struct urb *urb; struct tty_struct *tty; - struct ftdi_private *priv; char error_flag; unsigned char *data; @@ -2180,7 +2183,7 @@ static void ftdi_unthrottle (struct usb_serial_port *port) spin_unlock_irqrestore(&priv->rx_lock, flags); if (actually_throttled) - schedule_work(&priv->rx_work); + schedule_delayed_work(&priv->rx_work, 0); } static int __init ftdi_init (void) diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c index 909005107ea2..e09a0bfe6231 100644 --- a/drivers/usb/serial/keyspan_pda.c +++ b/drivers/usb/serial/keyspan_pda.c @@ -120,6 +120,8 @@ struct keyspan_pda_private { int tx_throttled; struct work_struct wakeup_work; struct work_struct unthrottle_work; + struct usb_serial *serial; + struct usb_serial_port *port; }; @@ -175,9 +177,11 @@ static struct usb_device_id id_table_fake_xircom [] = { }; #endif -static void keyspan_pda_wakeup_write( struct usb_serial_port *port ) +static void keyspan_pda_wakeup_write(struct work_struct *work) { - + struct keyspan_pda_private *priv = + container_of(work, struct keyspan_pda_private, wakeup_work); + struct usb_serial_port *port = priv->port; struct tty_struct *tty = port->tty; /* wake up port processes */ @@ -187,8 +191,11 @@ static void keyspan_pda_wakeup_write( struct usb_serial_port *port ) tty_wakeup(tty); } -static void keyspan_pda_request_unthrottle( struct usb_serial *serial ) +static void keyspan_pda_request_unthrottle(struct work_struct *work) { + struct keyspan_pda_private *priv = + container_of(work, struct keyspan_pda_private, unthrottle_work); + struct usb_serial *serial = priv->serial; int result; dbg(" request_unthrottle"); @@ -765,11 +772,10 @@ static int keyspan_pda_startup (struct usb_serial *serial) return (1); /* error */ usb_set_serial_port_data(serial->port[0], priv); init_waitqueue_head(&serial->port[0]->write_wait); - INIT_WORK(&priv->wakeup_work, (void *)keyspan_pda_wakeup_write, - (void *)(serial->port[0])); - INIT_WORK(&priv->unthrottle_work, - (void *)keyspan_pda_request_unthrottle, - (void *)(serial)); + INIT_WORK(&priv->wakeup_work, keyspan_pda_wakeup_write); + INIT_WORK(&priv->unthrottle_work, keyspan_pda_request_unthrottle); + priv->serial = serial; + priv->port = serial->port[0]; return (0); } diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index 8006e51c34bb..2cfba8488a93 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -533,9 +533,10 @@ void usb_serial_port_softint(struct usb_serial_port *port) schedule_work(&port->work); } -static void usb_serial_port_work(void *private) +static void usb_serial_port_work(struct work_struct *work) { - struct usb_serial_port *port = private; + struct usb_serial_port *port = + container_of(work, struct usb_serial_port, work); struct tty_struct *tty; dbg("%s - port %d", __FUNCTION__, port->number); @@ -799,7 +800,7 @@ int usb_serial_probe(struct usb_interface *interface, port->serial = serial; spin_lock_init(&port->lock); mutex_init(&port->mutex); - INIT_WORK(&port->work, usb_serial_port_work, port); + INIT_WORK(&port->work, usb_serial_port_work); serial->port[i] = port; } diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c index 4d1cd7aeccd3..154c7d290597 100644 --- a/drivers/usb/serial/whiteheat.c +++ b/drivers/usb/serial/whiteheat.c @@ -227,6 +227,7 @@ struct whiteheat_private { struct list_head rx_urbs_submitted; struct list_head rx_urb_q; struct work_struct rx_work; + struct usb_serial_port *port; struct list_head tx_urbs_free; struct list_head tx_urbs_submitted; }; @@ -241,7 +242,7 @@ static void command_port_read_callback(struct urb *urb); static int start_port_read(struct usb_serial_port *port); static struct whiteheat_urb_wrap *urb_to_wrap(struct urb *urb, struct list_head *head); static struct list_head *list_first(struct list_head *head); -static void rx_data_softint(void *private); +static void rx_data_softint(struct work_struct *work); static int firm_send_command(struct usb_serial_port *port, __u8 command, __u8 *data, __u8 datasize); static int firm_open(struct usb_serial_port *port); @@ -424,7 +425,8 @@ static int whiteheat_attach (struct usb_serial *serial) spin_lock_init(&info->lock); info->flags = 0; info->mcr = 0; - INIT_WORK(&info->rx_work, rx_data_softint, port); + INIT_WORK(&info->rx_work, rx_data_softint); + info->port = port; INIT_LIST_HEAD(&info->rx_urbs_free); INIT_LIST_HEAD(&info->rx_urbs_submitted); @@ -949,7 +951,7 @@ static void whiteheat_unthrottle (struct usb_serial_port *port) spin_unlock_irqrestore(&info->lock, flags); if (actually_throttled) - rx_data_softint(port); + rx_data_softint(&info->rx_work); return; } @@ -1400,10 +1402,11 @@ static struct list_head *list_first(struct list_head *head) } -static void rx_data_softint(void *private) +static void rx_data_softint(struct work_struct *work) { - struct usb_serial_port *port = (struct usb_serial_port *)private; - struct whiteheat_private *info = usb_get_serial_port_data(port); + struct whiteheat_private *info = + container_of(work, struct whiteheat_private, rx_work); + struct usb_serial_port *port = info->port; struct tty_struct *tty = port->tty; struct whiteheat_urb_wrap *wrap; struct urb *urb; diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c index 302174b8e477..31f476a64790 100644 --- a/drivers/video/console/fbcon.c +++ b/drivers/video/console/fbcon.c @@ -383,9 +383,9 @@ static void fbcon_update_softback(struct vc_data *vc) softback_top = 0; } -static void fb_flashcursor(void *private) +static void fb_flashcursor(struct work_struct *work) { - struct fb_info *info = private; + struct fb_info *info = container_of(work, struct fb_info, queue); struct fbcon_ops *ops = info->fbcon_par; struct display *p; struct vc_data *vc = NULL; @@ -442,7 +442,7 @@ static void fbcon_add_cursor_timer(struct fb_info *info) if ((!info->queue.func || info->queue.func == fb_flashcursor) && !(ops->flags & FBCON_FLAGS_CURSOR_TIMER)) { if (!info->queue.func) - INIT_WORK(&info->queue, fb_flashcursor, info); + INIT_WORK(&info->queue, fb_flashcursor); init_timer(&ops->cursor_timer); ops->cursor_timer.function = cursor_timer_handler; diff --git a/fs/9p/mux.c b/fs/9p/mux.c index 90a79c784549..944273c3dbff 100644 --- a/fs/9p/mux.c +++ b/fs/9p/mux.c @@ -110,8 +110,8 @@ struct v9fs_mux_rpc { }; static int v9fs_poll_proc(void *); -static void v9fs_read_work(void *); -static void v9fs_write_work(void *); +static void v9fs_read_work(struct work_struct *work); +static void v9fs_write_work(struct work_struct *work); static void v9fs_pollwait(struct file *filp, wait_queue_head_t * wait_address, poll_table * p); static u16 v9fs_mux_get_tag(struct v9fs_mux_data *); @@ -297,8 +297,8 @@ struct v9fs_mux_data *v9fs_mux_init(struct v9fs_transport *trans, int msize, m->rbuf = NULL; m->wpos = m->wsize = 0; m->wbuf = NULL; - INIT_WORK(&m->rq, v9fs_read_work, m); - INIT_WORK(&m->wq, v9fs_write_work, m); + INIT_WORK(&m->rq, v9fs_read_work); + INIT_WORK(&m->wq, v9fs_write_work); m->wsched = 0; memset(&m->poll_waddr, 0, sizeof(m->poll_waddr)); m->poll_task = NULL; @@ -458,13 +458,13 @@ static int v9fs_poll_proc(void *a) /** * v9fs_write_work - called when a transport can send some data */ -static void v9fs_write_work(void *a) +static void v9fs_write_work(struct work_struct *work) { int n, err; struct v9fs_mux_data *m; struct v9fs_req *req; - m = a; + m = container_of(work, struct v9fs_mux_data, wq); if (m->err < 0) { clear_bit(Wworksched, &m->wsched); @@ -564,7 +564,7 @@ static void process_request(struct v9fs_mux_data *m, struct v9fs_req *req) /** * v9fs_read_work - called when there is some data to be read from a transport */ -static void v9fs_read_work(void *a) +static void v9fs_read_work(struct work_struct *work) { int n, err; struct v9fs_mux_data *m; @@ -572,7 +572,7 @@ static void v9fs_read_work(void *a) struct v9fs_fcall *rcall; char *rbuf; - m = a; + m = container_of(work, struct v9fs_mux_data, rq); if (m->err < 0) return; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 78fe0fae23ff..55f5333dae99 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -35,7 +35,7 @@ struct greedy { struct gfs2_holder gr_gh; - struct work_struct gr_work; + struct delayed_work gr_work; }; struct gfs2_gl_hash_bucket { @@ -1368,9 +1368,9 @@ static void gfs2_glock_prefetch(struct gfs2_glock *gl, unsigned int state, glops->go_xmote_th(gl, state, flags); } -static void greedy_work(void *data) +static void greedy_work(struct work_struct *work) { - struct greedy *gr = data; + struct greedy *gr = container_of(work, struct greedy, gr_work.work); struct gfs2_holder *gh = &gr->gr_gh; struct gfs2_glock *gl = gh->gh_gl; const struct gfs2_glock_operations *glops = gl->gl_ops; @@ -1422,7 +1422,7 @@ int gfs2_glock_be_greedy(struct gfs2_glock *gl, unsigned int time) gfs2_holder_init(gl, 0, 0, gh); set_bit(HIF_GREEDY, &gh->gh_iflags); - INIT_WORK(&gr->gr_work, greedy_work, gr); + INIT_DELAYED_WORK(&gr->gr_work, greedy_work); set_bit(GLF_SKIP_WAITERS2, &gl->gl_flags); schedule_delayed_work(&gr->gr_work, time); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 42e3bef270c9..72dad552aa00 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -577,12 +577,12 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) server->rcv.ptr = (unsigned char*)&server->rcv.buf; server->rcv.len = 10; server->rcv.state = 0; - INIT_WORK(&server->rcv.tq, ncp_tcp_rcv_proc, server); - INIT_WORK(&server->tx.tq, ncp_tcp_tx_proc, server); + INIT_WORK(&server->rcv.tq, ncp_tcp_rcv_proc); + INIT_WORK(&server->tx.tq, ncp_tcp_tx_proc); sock->sk->sk_write_space = ncp_tcp_write_space; } else { - INIT_WORK(&server->rcv.tq, ncpdgram_rcv_proc, server); - INIT_WORK(&server->timeout_tq, ncpdgram_timeout_proc, server); + INIT_WORK(&server->rcv.tq, ncpdgram_rcv_proc); + INIT_WORK(&server->timeout_tq, ncpdgram_timeout_proc); server->timeout_tm.data = (unsigned long)server; server->timeout_tm.function = ncpdgram_timeout_call; } diff --git a/fs/ncpfs/sock.c b/fs/ncpfs/sock.c index 11c2b252ebed..e496d8b65e92 100644 --- a/fs/ncpfs/sock.c +++ b/fs/ncpfs/sock.c @@ -350,9 +350,10 @@ static void info_server(struct ncp_server *server, unsigned int id, const void * } } -void ncpdgram_rcv_proc(void *s) +void ncpdgram_rcv_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, rcv.tq); struct socket* sock; sock = server->ncp_sock; @@ -468,9 +469,10 @@ static void __ncpdgram_timeout_proc(struct ncp_server *server) } } -void ncpdgram_timeout_proc(void *s) +void ncpdgram_timeout_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, timeout_tq); mutex_lock(&server->rcv.creq_mutex); __ncpdgram_timeout_proc(server); mutex_unlock(&server->rcv.creq_mutex); @@ -652,18 +654,20 @@ skipdata:; } } -void ncp_tcp_rcv_proc(void *s) +void ncp_tcp_rcv_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, rcv.tq); mutex_lock(&server->rcv.creq_mutex); __ncptcp_rcv_proc(server); mutex_unlock(&server->rcv.creq_mutex); } -void ncp_tcp_tx_proc(void *s) +void ncp_tcp_tx_proc(struct work_struct *work) { - struct ncp_server *server = s; + struct ncp_server *server = + container_of(work, struct ncp_server, tx.tq); mutex_lock(&server->rcv.creq_mutex); __ncptcp_try_send(server); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 293b6495829f..e431e93ab503 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1829,9 +1829,8 @@ out: } static struct workqueue_struct *laundry_wq; -static struct work_struct laundromat_work; -static void laundromat_main(void *); -static DECLARE_WORK(laundromat_work, laundromat_main, NULL); +static void laundromat_main(struct work_struct *); +static DECLARE_DELAYED_WORK(laundromat_work, laundromat_main); __be32 nfsd4_renew(clientid_t *clid) @@ -1940,7 +1939,7 @@ nfs4_laundromat(void) } void -laundromat_main(void *not_used) +laundromat_main(struct work_struct *not_used) { time_t t; diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index f43bc5f18a35..0b2ad163005e 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -1205,10 +1205,12 @@ int ocfs2_flush_truncate_log(struct ocfs2_super *osb) return status; } -static void ocfs2_truncate_log_worker(void *data) +static void ocfs2_truncate_log_worker(struct work_struct *work) { int status; - struct ocfs2_super *osb = data; + struct ocfs2_super *osb = + container_of(work, struct ocfs2_super, + osb_truncate_log_wq.work); mlog_entry_void(); @@ -1441,7 +1443,8 @@ int ocfs2_truncate_log_init(struct ocfs2_super *osb) /* ocfs2_truncate_log_shutdown keys on the existence of * osb->osb_tl_inode so we don't set any of the osb variables * until we're sure all is well. */ - INIT_WORK(&osb->osb_truncate_log_wq, ocfs2_truncate_log_worker, osb); + INIT_DELAYED_WORK(&osb->osb_truncate_log_wq, + ocfs2_truncate_log_worker); osb->osb_tl_bh = tl_bh; osb->osb_tl_inode = tl_inode; diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 305cba3681fe..4cd9a9580456 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -141,7 +141,7 @@ struct o2hb_region { * recognizes a node going up and down in one iteration */ u64 hr_generation; - struct work_struct hr_write_timeout_work; + struct delayed_work hr_write_timeout_work; unsigned long hr_last_timeout_start; /* Used during o2hb_check_slot to hold a copy of the block @@ -156,9 +156,11 @@ struct o2hb_bio_wait_ctxt { int wc_error; }; -static void o2hb_write_timeout(void *arg) +static void o2hb_write_timeout(struct work_struct *work) { - struct o2hb_region *reg = arg; + struct o2hb_region *reg = + container_of(work, struct o2hb_region, + hr_write_timeout_work.work); mlog(ML_ERROR, "Heartbeat write timeout to device %s after %u " "milliseconds\n", reg->hr_dev_name, @@ -1404,7 +1406,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, goto out; } - INIT_WORK(®->hr_write_timeout_work, o2hb_write_timeout, reg); + INIT_DELAYED_WORK(®->hr_write_timeout_work, o2hb_write_timeout); /* * A node is considered live after it has beat LIVE_THRESHOLD diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index 7bba98fbfc15..4705d659fe57 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -88,7 +88,7 @@ void o2quo_disk_timeout(void) o2quo_fence_self(); } -static void o2quo_make_decision(void *arg) +static void o2quo_make_decision(struct work_struct *work) { int quorum; int lowest_hb, lowest_reachable = 0, fence = 0; @@ -306,7 +306,7 @@ void o2quo_init(void) struct o2quo_state *qs = &o2quo_state; spin_lock_init(&qs->qs_lock); - INIT_WORK(&qs->qs_work, o2quo_make_decision, NULL); + INIT_WORK(&qs->qs_work, o2quo_make_decision); } void o2quo_exit(void) diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index b650efa8c8be..9b3209dc0b16 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -140,11 +140,11 @@ static int o2net_sys_err_translations[O2NET_ERR_MAX] = [O2NET_ERR_DIED] = -EHOSTDOWN,}; /* can't quite avoid *all* internal declarations :/ */ -static void o2net_sc_connect_completed(void *arg); -static void o2net_rx_until_empty(void *arg); -static void o2net_shutdown_sc(void *arg); +static void o2net_sc_connect_completed(struct work_struct *work); +static void o2net_rx_until_empty(struct work_struct *work); +static void o2net_shutdown_sc(struct work_struct *work); static void o2net_listen_data_ready(struct sock *sk, int bytes); -static void o2net_sc_send_keep_req(void *arg); +static void o2net_sc_send_keep_req(struct work_struct *work); static void o2net_idle_timer(unsigned long data); static void o2net_sc_postpone_idle(struct o2net_sock_container *sc); @@ -308,10 +308,10 @@ static struct o2net_sock_container *sc_alloc(struct o2nm_node *node) o2nm_node_get(node); sc->sc_node = node; - INIT_WORK(&sc->sc_connect_work, o2net_sc_connect_completed, sc); - INIT_WORK(&sc->sc_rx_work, o2net_rx_until_empty, sc); - INIT_WORK(&sc->sc_shutdown_work, o2net_shutdown_sc, sc); - INIT_WORK(&sc->sc_keepalive_work, o2net_sc_send_keep_req, sc); + INIT_WORK(&sc->sc_connect_work, o2net_sc_connect_completed); + INIT_WORK(&sc->sc_rx_work, o2net_rx_until_empty); + INIT_WORK(&sc->sc_shutdown_work, o2net_shutdown_sc); + INIT_DELAYED_WORK(&sc->sc_keepalive_work, o2net_sc_send_keep_req); init_timer(&sc->sc_idle_timeout); sc->sc_idle_timeout.function = o2net_idle_timer; @@ -342,7 +342,7 @@ static void o2net_sc_queue_work(struct o2net_sock_container *sc, sc_put(sc); } static void o2net_sc_queue_delayed_work(struct o2net_sock_container *sc, - struct work_struct *work, + struct delayed_work *work, int delay) { sc_get(sc); @@ -350,7 +350,7 @@ static void o2net_sc_queue_delayed_work(struct o2net_sock_container *sc, sc_put(sc); } static void o2net_sc_cancel_delayed_work(struct o2net_sock_container *sc, - struct work_struct *work) + struct delayed_work *work) { if (cancel_delayed_work(work)) sc_put(sc); @@ -564,9 +564,11 @@ static void o2net_ensure_shutdown(struct o2net_node *nn, * ourselves as state_change couldn't get the nn_lock and call set_nn_state * itself. */ -static void o2net_shutdown_sc(void *arg) +static void o2net_shutdown_sc(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_shutdown_work); struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); sclog(sc, "shutting down\n"); @@ -1201,9 +1203,10 @@ out: /* this work func is triggerd by data ready. it reads until it can read no * more. it interprets 0, eof, as fatal. if data_ready hits while we're doing * our work the work struct will be marked and we'll be called again. */ -static void o2net_rx_until_empty(void *arg) +static void o2net_rx_until_empty(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, sc_rx_work); int ret; do { @@ -1249,9 +1252,11 @@ static int o2net_set_nodelay(struct socket *sock) /* called when a connect completes and after a sock is accepted. the * rx path will see the response and mark the sc valid */ -static void o2net_sc_connect_completed(void *arg) +static void o2net_sc_connect_completed(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_connect_work); mlog(ML_MSG, "sc sending handshake with ver %llu id %llx\n", (unsigned long long)O2NET_PROTOCOL_VERSION, @@ -1262,9 +1267,11 @@ static void o2net_sc_connect_completed(void *arg) } /* this is called as a work_struct func. */ -static void o2net_sc_send_keep_req(void *arg) +static void o2net_sc_send_keep_req(struct work_struct *work) { - struct o2net_sock_container *sc = arg; + struct o2net_sock_container *sc = + container_of(work, struct o2net_sock_container, + sc_keepalive_work.work); o2net_sendpage(sc, o2net_keep_req, sizeof(*o2net_keep_req)); sc_put(sc); @@ -1314,14 +1321,15 @@ static void o2net_sc_postpone_idle(struct o2net_sock_container *sc) * having a connect attempt fail, etc. This centralizes the logic which decides * if a connect attempt should be made or if we should give up and all future * transmit attempts should fail */ -static void o2net_start_connect(void *arg) +static void o2net_start_connect(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_connect_work.work); struct o2net_sock_container *sc = NULL; struct o2nm_node *node = NULL, *mynode = NULL; struct socket *sock = NULL; struct sockaddr_in myaddr = {0, }, remoteaddr = {0, }; - int ret = 0; + int ret = 0, stop; /* if we're greater we initiate tx, otherwise we accept */ if (o2nm_this_node() <= o2net_num_from_nn(nn)) @@ -1342,10 +1350,9 @@ static void o2net_start_connect(void *arg) spin_lock(&nn->nn_lock); /* see if we already have one pending or have given up */ - if (nn->nn_sc || nn->nn_persistent_error) - arg = NULL; + stop = (nn->nn_sc || nn->nn_persistent_error); spin_unlock(&nn->nn_lock); - if (arg == NULL) /* *shrug*, needed some indicator */ + if (stop) goto out; nn->nn_last_connect_attempt = jiffies; @@ -1421,9 +1428,10 @@ out: return; } -static void o2net_connect_expired(void *arg) +static void o2net_connect_expired(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_connect_expired.work); spin_lock(&nn->nn_lock); if (!nn->nn_sc_valid) { @@ -1436,9 +1444,10 @@ static void o2net_connect_expired(void *arg) spin_unlock(&nn->nn_lock); } -static void o2net_still_up(void *arg) +static void o2net_still_up(struct work_struct *work) { - struct o2net_node *nn = arg; + struct o2net_node *nn = + container_of(work, struct o2net_node, nn_still_up.work); o2quo_hb_still_up(o2net_num_from_nn(nn)); } @@ -1644,9 +1653,9 @@ out: return ret; } -static void o2net_accept_many(void *arg) +static void o2net_accept_many(struct work_struct *work) { - struct socket *sock = arg; + struct socket *sock = o2net_listen_sock; while (o2net_accept_one(sock) == 0) cond_resched(); } @@ -1700,7 +1709,7 @@ static int o2net_open_listening_sock(__be16 port) write_unlock_bh(&sock->sk->sk_callback_lock); o2net_listen_sock = sock; - INIT_WORK(&o2net_listen_work, o2net_accept_many, sock); + INIT_WORK(&o2net_listen_work, o2net_accept_many); sock->sk->sk_reuse = 1; ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin)); @@ -1819,9 +1828,10 @@ int o2net_init(void) struct o2net_node *nn = o2net_nn_from_num(i); spin_lock_init(&nn->nn_lock); - INIT_WORK(&nn->nn_connect_work, o2net_start_connect, nn); - INIT_WORK(&nn->nn_connect_expired, o2net_connect_expired, nn); - INIT_WORK(&nn->nn_still_up, o2net_still_up, nn); + INIT_DELAYED_WORK(&nn->nn_connect_work, o2net_start_connect); + INIT_DELAYED_WORK(&nn->nn_connect_expired, + o2net_connect_expired); + INIT_DELAYED_WORK(&nn->nn_still_up, o2net_still_up); /* until we see hb from a node we'll return einval */ nn->nn_persistent_error = -ENOTCONN; init_waitqueue_head(&nn->nn_sc_wq); diff --git a/fs/ocfs2/cluster/tcp_internal.h b/fs/ocfs2/cluster/tcp_internal.h index 4b46aac7d243..daebbd3a2c8c 100644 --- a/fs/ocfs2/cluster/tcp_internal.h +++ b/fs/ocfs2/cluster/tcp_internal.h @@ -86,18 +86,18 @@ struct o2net_node { * connect attempt fails and so can be self-arming. shutdown is * careful to first mark the nn such that no connects will be attempted * before canceling delayed connect work and flushing the queue. */ - struct work_struct nn_connect_work; + struct delayed_work nn_connect_work; unsigned long nn_last_connect_attempt; /* this is queued as nodes come up and is canceled when a connection is * established. this expiring gives up on the node and errors out * transmits */ - struct work_struct nn_connect_expired; + struct delayed_work nn_connect_expired; /* after we give up on a socket we wait a while before deciding * that it is still heartbeating and that we should do some * quorum work */ - struct work_struct nn_still_up; + struct delayed_work nn_still_up; }; struct o2net_sock_container { @@ -129,7 +129,7 @@ struct o2net_sock_container { struct work_struct sc_shutdown_work; struct timer_list sc_idle_timeout; - struct work_struct sc_keepalive_work; + struct delayed_work sc_keepalive_work; unsigned sc_handshake_ok:1; diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index fa968180b072..6b6ff76538c5 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -153,7 +153,7 @@ static inline struct hlist_head *dlm_lockres_hash(struct dlm_ctxt *dlm, unsigned * called functions that cannot be directly called from the * net message handlers for some reason, usually because * they need to send net messages of their own. */ -void dlm_dispatch_work(void *data); +void dlm_dispatch_work(struct work_struct *work); struct dlm_lock_resource; struct dlm_work_item; diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 8d1065f8b3bd..637646e6922e 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1296,7 +1296,7 @@ static struct dlm_ctxt *dlm_alloc_ctxt(const char *domain, spin_lock_init(&dlm->work_lock); INIT_LIST_HEAD(&dlm->work_list); - INIT_WORK(&dlm->dispatched_work, dlm_dispatch_work, dlm); + INIT_WORK(&dlm->dispatched_work, dlm_dispatch_work); kref_init(&dlm->dlm_refs); dlm->dlm_state = DLM_CTXT_NEW; diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 9d950d7cea38..fb3e2b0817f1 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -153,9 +153,10 @@ static inline void dlm_reset_recovery(struct dlm_ctxt *dlm) } /* Worker function used during recovery. */ -void dlm_dispatch_work(void *data) +void dlm_dispatch_work(struct work_struct *work) { - struct dlm_ctxt *dlm = (struct dlm_ctxt *)data; + struct dlm_ctxt *dlm = + container_of(work, struct dlm_ctxt, dispatched_work); LIST_HEAD(tmp_list); struct list_head *iter, *iter2; struct dlm_work_item *item; diff --git a/fs/ocfs2/dlm/userdlm.c b/fs/ocfs2/dlm/userdlm.c index eead48bbfac6..7d2f578b267d 100644 --- a/fs/ocfs2/dlm/userdlm.c +++ b/fs/ocfs2/dlm/userdlm.c @@ -171,15 +171,14 @@ static inline void user_dlm_grab_inode_ref(struct user_lock_res *lockres) BUG(); } -static void user_dlm_unblock_lock(void *opaque); +static void user_dlm_unblock_lock(struct work_struct *work); static void __user_dlm_queue_lockres(struct user_lock_res *lockres) { if (!(lockres->l_flags & USER_LOCK_QUEUED)) { user_dlm_grab_inode_ref(lockres); - INIT_WORK(&lockres->l_work, user_dlm_unblock_lock, - lockres); + INIT_WORK(&lockres->l_work, user_dlm_unblock_lock); queue_work(user_dlm_worker, &lockres->l_work); lockres->l_flags |= USER_LOCK_QUEUED; @@ -279,10 +278,11 @@ static inline void user_dlm_drop_inode_ref(struct user_lock_res *lockres) iput(inode); } -static void user_dlm_unblock_lock(void *opaque) +static void user_dlm_unblock_lock(struct work_struct *work) { int new_level, status; - struct user_lock_res *lockres = (struct user_lock_res *) opaque; + struct user_lock_res *lockres = + container_of(work, struct user_lock_res, l_work); struct dlm_ctxt *dlm = dlm_ctxt_from_user_lockres(lockres); mlog(0, "processing lockres %.*s\n", lockres->l_namelen, diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index fd9734def551..d95ee2720e6e 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -911,11 +911,12 @@ struct ocfs2_la_recovery_item { * NOTE: This function can and will sleep on recovery of other nodes * during cluster locking, just like any other ocfs2 process. */ -void ocfs2_complete_recovery(void *data) +void ocfs2_complete_recovery(struct work_struct *work) { int ret; - struct ocfs2_super *osb = data; - struct ocfs2_journal *journal = osb->journal; + struct ocfs2_journal *journal = + container_of(work, struct ocfs2_journal, j_recovery_work); + struct ocfs2_super *osb = journal->j_osb; struct ocfs2_dinode *la_dinode, *tl_dinode; struct ocfs2_la_recovery_item *item; struct list_head *p, *n; diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 2f3a6acdac45..5be161a4ad9f 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -172,7 +172,7 @@ static inline void ocfs2_handle_set_sync(struct ocfs2_journal_handle *handle, in } /* Exported only for the journal struct init code in super.c. Do not call. */ -void ocfs2_complete_recovery(void *data); +void ocfs2_complete_recovery(struct work_struct *work); /* * Journal Control: diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 0462a7f4e21b..9b1bad1d48ec 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -283,7 +283,7 @@ struct ocfs2_super /* Truncate log info */ struct inode *osb_tl_inode; struct buffer_head *osb_tl_bh; - struct work_struct osb_truncate_log_wq; + struct delayed_work osb_truncate_log_wq; struct ocfs2_node_map osb_recovering_orphan_dirs; unsigned int *osb_orphan_wipes; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 76b46ebbb10c..9a8089030f55 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1365,7 +1365,7 @@ static int ocfs2_initialize_super(struct super_block *sb, spin_lock_init(&journal->j_lock); journal->j_trans_id = (unsigned long) 1; INIT_LIST_HEAD(&journal->j_la_cleanups); - INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery, osb); + INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery); journal->j_state = OCFS2_JOURNAL_FREE; /* get some pseudo constants for clustersize bits */ diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 85ce23268302..cd1bb75ceb24 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -104,7 +104,7 @@ static int release_journal_dev(struct super_block *super, struct reiserfs_journal *journal); static int dirty_one_transaction(struct super_block *s, struct reiserfs_journal_list *jl); -static void flush_async_commits(void *p); +static void flush_async_commits(struct work_struct *work); static void queue_log_writer(struct super_block *s); /* values for join in do_journal_begin_r */ @@ -2836,7 +2836,8 @@ int journal_init(struct super_block *p_s_sb, const char *j_dev_name, if (reiserfs_mounted_fs_count <= 1) commit_wq = create_workqueue("reiserfs"); - INIT_WORK(&journal->j_work, flush_async_commits, p_s_sb); + INIT_DELAYED_WORK(&journal->j_work, flush_async_commits); + journal->j_work_sb = p_s_sb; return 0; free_and_return: free_journal_ram(p_s_sb); @@ -3447,10 +3448,11 @@ int journal_end_sync(struct reiserfs_transaction_handle *th, /* ** writeback the pending async commits to disk */ -static void flush_async_commits(void *p) +static void flush_async_commits(struct work_struct *work) { - struct super_block *p_s_sb = p; - struct reiserfs_journal *journal = SB_JOURNAL(p_s_sb); + struct reiserfs_journal *journal = + container_of(work, struct reiserfs_journal, j_work.work); + struct super_block *p_s_sb = journal->j_work_sb; struct reiserfs_journal_list *jl; struct list_head *entry; diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 09360cf1e1f2..8e6b56fc1cad 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -149,9 +149,10 @@ xfs_destroy_ioend( */ STATIC void xfs_end_bio_delalloc( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); xfs_destroy_ioend(ioend); } @@ -161,9 +162,10 @@ xfs_end_bio_delalloc( */ STATIC void xfs_end_bio_written( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); xfs_destroy_ioend(ioend); } @@ -176,9 +178,10 @@ xfs_end_bio_written( */ STATIC void xfs_end_bio_unwritten( - void *data) + struct work_struct *work) { - xfs_ioend_t *ioend = data; + xfs_ioend_t *ioend = + container_of(work, xfs_ioend_t, io_work); bhv_vnode_t *vp = ioend->io_vnode; xfs_off_t offset = ioend->io_offset; size_t size = ioend->io_size; @@ -220,11 +223,11 @@ xfs_alloc_ioend( ioend->io_size = 0; if (type == IOMAP_UNWRITTEN) - INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten); else if (type == IOMAP_DELAY) - INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc); else - INIT_WORK(&ioend->io_work, xfs_end_bio_written, ioend); + INIT_WORK(&ioend->io_work, xfs_end_bio_written); return ioend; } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index d3382843698e..eef4a0ba11e9 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -994,9 +994,10 @@ xfs_buf_wait_unpin( STATIC void xfs_buf_iodone_work( - void *v) + struct work_struct *work) { - xfs_buf_t *bp = (xfs_buf_t *)v; + xfs_buf_t *bp = + container_of(work, xfs_buf_t, b_iodone_work); if (bp->b_iodone) (*(bp->b_iodone))(bp); @@ -1017,10 +1018,10 @@ xfs_buf_ioend( if ((bp->b_iodone) || (bp->b_flags & XBF_ASYNC)) { if (schedule) { - INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work, bp); + INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work); queue_work(xfslogd_workqueue, &bp->b_iodone_work); } else { - xfs_buf_iodone_work(bp); + xfs_buf_iodone_work(&bp->b_iodone_work); } } else { up(&bp->b_iodonesema); diff --git a/include/linux/connector.h b/include/linux/connector.h index 4c02119c6ab9..3ea1cd58de97 100644 --- a/include/linux/connector.h +++ b/include/linux/connector.h @@ -133,7 +133,7 @@ struct cn_callback_data { struct cn_callback_entry { struct list_head callback_entry; struct cn_callback *cb; - struct work_struct work; + struct delayed_work work; struct cn_queue_dev *pdev; struct cn_callback_id id; @@ -170,7 +170,7 @@ void cn_queue_free_dev(struct cn_queue_dev *dev); int cn_cb_equal(struct cb_id *, struct cb_id *); -void cn_queue_wrapper(void *data); +void cn_queue_wrapper(struct work_struct *work); extern int cn_already_initialized; diff --git a/include/linux/i2o.h b/include/linux/i2o.h index c115e9e840b4..1fb02e17f6f6 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -461,7 +461,7 @@ struct i2o_driver { int (*reply) (struct i2o_controller *, u32, struct i2o_message *); /* Event handler */ - void (*event) (struct i2o_event *); + work_func_t event; struct workqueue_struct *event_queue; /* Event queue */ diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index 587264a58d56..8b08ef3820f2 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -110,7 +110,7 @@ struct mmc_host { struct mmc_card *card_busy; /* the MMC card claiming host */ struct mmc_card *card_selected; /* the selected MMC card */ - struct work_struct detect; + struct delayed_work detect; unsigned long private[0] ____cacheline_aligned; }; diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h index b089d9506283..a503052138bd 100644 --- a/include/linux/ncp_fs_sb.h +++ b/include/linux/ncp_fs_sb.h @@ -127,10 +127,10 @@ struct ncp_server { } unexpected_packet; }; -extern void ncp_tcp_rcv_proc(void *server); -extern void ncp_tcp_tx_proc(void *server); -extern void ncpdgram_rcv_proc(void *server); -extern void ncpdgram_timeout_proc(void *server); +extern void ncp_tcp_rcv_proc(struct work_struct *work); +extern void ncp_tcp_tx_proc(struct work_struct *work); +extern void ncpdgram_rcv_proc(struct work_struct *work); +extern void ncpdgram_timeout_proc(struct work_struct *work); extern void ncpdgram_timeout_call(unsigned long server); extern void ncp_tcp_data_ready(struct sock* sk, int len); extern void ncp_tcp_write_space(struct sock* sk); diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 73e0becec086..6610103f23e1 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -249,7 +249,8 @@ struct reiserfs_journal { int j_errno; /* when flushing ordered buffers, throttle new ordered writers */ - struct work_struct j_work; + struct delayed_work j_work; + struct super_block *j_work_sb; atomic_t j_async_throttle; }; diff --git a/include/linux/relay.h b/include/linux/relay.h index 24accb483849..0e3d91b76996 100644 --- a/include/linux/relay.h +++ b/include/linux/relay.h @@ -38,7 +38,7 @@ struct rchan_buf size_t subbufs_consumed; /* count of sub-buffers consumed */ struct rchan *chan; /* associated channel */ wait_queue_head_t read_wait; /* reader wait queue */ - struct work_struct wake_readers; /* reader wake-up work struct */ + struct delayed_work wake_readers; /* reader wake-up work struct */ struct dentry *dentry; /* channel file dentry */ struct kref kref; /* channel buffer refcount */ struct page **page_array; /* array of current buffer pages */ diff --git a/include/linux/usb.h b/include/linux/usb.h index 5482bfb3303d..06ce7a626040 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -382,7 +382,7 @@ struct usb_device { int pm_usage_cnt; /* usage counter for autosuspend */ #ifdef CONFIG_PM - struct work_struct autosuspend; /* for delayed autosuspends */ + struct delayed_work autosuspend; /* for delayed autosuspends */ struct mutex pm_mutex; /* protects PM operations */ unsigned auto_pm:1; /* autosuspend/resume in progress */ diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h index 617b672b1132..89119277553d 100644 --- a/include/net/ieee80211softmac.h +++ b/include/net/ieee80211softmac.h @@ -108,8 +108,8 @@ struct ieee80211softmac_assoc_info { /* Scan retries remaining */ int scan_retry; - struct work_struct work; - struct work_struct timeout; + struct delayed_work work; + struct delayed_work timeout; }; struct ieee80211softmac_bss_info { diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index c6d93bb0dcd2..60b684470db8 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1030,7 +1030,7 @@ void sctp_inq_init(struct sctp_inq *); void sctp_inq_free(struct sctp_inq *); void sctp_inq_push(struct sctp_inq *, struct sctp_chunk *packet); struct sctp_chunk *sctp_inq_pop(struct sctp_inq *); -void sctp_inq_set_th_handler(struct sctp_inq *, void (*)(void *), void *); +void sctp_inq_set_th_handler(struct sctp_inq *, work_func_t); /* This is the structure we use to hold outbound chunks. You push * chunks in and they automatically pop out the other end as bundled diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 9582e8401669..9ccc0365aa89 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -200,9 +200,14 @@ struct domain_device { void *lldd_dev; }; +struct sas_discovery_event { + struct work_struct work; + struct asd_sas_port *port; +}; + struct sas_discovery { spinlock_t disc_event_lock; - struct work_struct disc_work[DISC_NUM_EVENTS]; + struct sas_discovery_event disc_work[DISC_NUM_EVENTS]; unsigned long pending; u8 fanout_sas_addr[8]; u8 eeds_a[8]; @@ -248,14 +253,19 @@ struct asd_sas_port { void *lldd_port; /* not touched by the sas class code */ }; +struct asd_sas_event { + struct work_struct work; + struct asd_sas_phy *phy; +}; + /* The phy pretty much is controlled by the LLDD. * The class only reads those fields. */ struct asd_sas_phy { /* private: */ /* protected by ha->event_lock */ - struct work_struct port_events[PORT_NUM_EVENTS]; - struct work_struct phy_events[PHY_NUM_EVENTS]; + struct asd_sas_event port_events[PORT_NUM_EVENTS]; + struct asd_sas_event phy_events[PHY_NUM_EVENTS]; unsigned long port_events_pending; unsigned long phy_events_pending; @@ -307,10 +317,15 @@ struct scsi_core { int queue_thread_kill; }; +struct sas_ha_event { + struct work_struct work; + struct sas_ha_struct *ha; +}; + struct sas_ha_struct { /* private: */ spinlock_t event_lock; - struct work_struct ha_events[HA_NUM_EVENTS]; + struct sas_ha_event ha_events[HA_NUM_EVENTS]; unsigned long pending; struct scsi_core core; diff --git a/include/scsi/scsi_transport_fc.h b/include/scsi/scsi_transport_fc.h index fd352323378b..798f7c7ee426 100644 --- a/include/scsi/scsi_transport_fc.h +++ b/include/scsi/scsi_transport_fc.h @@ -206,9 +206,9 @@ struct fc_rport { /* aka fc_starget_attrs */ u8 flags; struct list_head peers; struct device dev; - struct work_struct dev_loss_work; + struct delayed_work dev_loss_work; struct work_struct scan_work; - struct work_struct fail_io_work; + struct delayed_work fail_io_work; struct work_struct stgt_delete_work; struct work_struct rport_delete_work; } __attribute__((aligned(sizeof(unsigned long)))); diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 4b95c89c95c9..d5c218ddc527 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -176,7 +176,7 @@ struct iscsi_cls_session { /* recovery fields */ int recovery_tmo; - struct work_struct recovery_work; + struct delayed_work recovery_work; int target_id; diff --git a/include/sound/ac97_codec.h b/include/sound/ac97_codec.h index 4c43521cc493..33720397a904 100644 --- a/include/sound/ac97_codec.h +++ b/include/sound/ac97_codec.h @@ -511,7 +511,7 @@ struct snd_ac97 { #ifdef CONFIG_SND_AC97_POWER_SAVE unsigned int power_up; /* power states */ struct workqueue_struct *power_workq; - struct work_struct power_work; + struct delayed_work power_work; #endif struct device dev; }; diff --git a/include/sound/ak4114.h b/include/sound/ak4114.h index 11702aa0bea9..2ee061625fd0 100644 --- a/include/sound/ak4114.h +++ b/include/sound/ak4114.h @@ -182,7 +182,7 @@ struct ak4114 { unsigned char rcs0; unsigned char rcs1; struct workqueue_struct *workqueue; - struct work_struct work; + struct delayed_work work; void *change_callback_private; void (*change_callback)(struct ak4114 *ak4114, unsigned char c0, unsigned char c1); }; diff --git a/kernel/relay.c b/kernel/relay.c index f04bbdb56ac2..2b92e8ece85b 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -308,9 +308,10 @@ static struct rchan_callbacks default_channel_callbacks = { * reason waking is deferred is that calling directly from write * causes problems if you're writing from say the scheduler. */ -static void wakeup_readers(void *private) +static void wakeup_readers(struct work_struct *work) { - struct rchan_buf *buf = private; + struct rchan_buf *buf = + container_of(work, struct rchan_buf, wake_readers.work); wake_up_interruptible(&buf->read_wait); } @@ -328,7 +329,7 @@ static inline void __relay_reset(struct rchan_buf *buf, unsigned int init) if (init) { init_waitqueue_head(&buf->read_wait); kref_init(&buf->kref); - INIT_WORK(&buf->wake_readers, NULL, NULL); + INIT_DELAYED_WORK(&buf->wake_readers, NULL); } else { cancel_delayed_work(&buf->wake_readers); flush_scheduled_work(); @@ -549,7 +550,8 @@ size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length) buf->padding[old_subbuf]; smp_mb(); if (waitqueue_active(&buf->read_wait)) { - PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf); + PREPARE_DELAYED_WORK(&buf->wake_readers, + wakeup_readers); schedule_delayed_work(&buf->wake_readers, 1); } } diff --git a/mm/swap.c b/mm/swap.c index 2e0e871f542f..d9a3770d8f3c 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -216,7 +216,7 @@ void lru_add_drain(void) } #ifdef CONFIG_NUMA -static void lru_add_drain_per_cpu(void *dummy) +static void lru_add_drain_per_cpu(struct work_struct *dummy) { lru_add_drain(); } @@ -226,7 +226,7 @@ static void lru_add_drain_per_cpu(void *dummy) */ int lru_add_drain_all(void) { - return schedule_on_each_cpu(lru_add_drain_per_cpu, NULL); + return schedule_on_each_cpu(lru_add_drain_per_cpu); } #else diff --git a/net/atm/lec.c b/net/atm/lec.c index 66c57c1091a8..e801fff69dc0 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -1458,7 +1458,7 @@ static void lane2_associate_ind(struct net_device *dev, u8 *mac_addr, #define LEC_ARP_REFRESH_INTERVAL (3*HZ) -static void lec_arp_check_expire(void *data); +static void lec_arp_check_expire(struct work_struct *work); static void lec_arp_expire_arp(unsigned long data); /* @@ -1481,7 +1481,7 @@ static void lec_arp_init(struct lec_priv *priv) INIT_HLIST_HEAD(&priv->lec_no_forward); INIT_HLIST_HEAD(&priv->mcast_fwds); spin_lock_init(&priv->lec_arp_lock); - INIT_WORK(&priv->lec_arp_work, lec_arp_check_expire, priv); + INIT_DELAYED_WORK(&priv->lec_arp_work, lec_arp_check_expire); schedule_delayed_work(&priv->lec_arp_work, LEC_ARP_REFRESH_INTERVAL); } @@ -1879,10 +1879,11 @@ static void lec_arp_expire_vcc(unsigned long data) * to ESI_FORWARD_DIRECT. This causes the flush period to end * regardless of the progress of the flush protocol. */ -static void lec_arp_check_expire(void *data) +static void lec_arp_check_expire(struct work_struct *work) { unsigned long flags; - struct lec_priv *priv = data; + struct lec_priv *priv = + container_of(work, struct lec_priv, lec_arp_work.work); struct hlist_node *node, *next; struct lec_arp_table *entry; unsigned long now; diff --git a/net/atm/lec.h b/net/atm/lec.h index 877f50939696..984e8e6e083a 100644 --- a/net/atm/lec.h +++ b/net/atm/lec.h @@ -92,7 +92,7 @@ struct lec_priv { spinlock_t lec_arp_lock; struct atm_vcc *mcast_vcc; /* Default Multicast Send VCC */ struct atm_vcc *lecd; - struct work_struct lec_arp_work; /* C10 */ + struct delayed_work lec_arp_work; /* C10 */ unsigned int maximum_unknown_frame_count; /* * Within the period of time defined by this variable, the client will send diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 3eeeb7a86e75..d4c935692ccf 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -237,9 +237,9 @@ static void bt_release(struct device *dev) kfree(data); } -static void add_conn(void *data) +static void add_conn(struct work_struct *work) { - struct hci_conn *conn = data; + struct hci_conn *conn = container_of(work, struct hci_conn, work); int i; if (device_register(&conn->dev) < 0) { @@ -272,14 +272,14 @@ void hci_conn_add_sysfs(struct hci_conn *conn) dev_set_drvdata(&conn->dev, conn); - INIT_WORK(&conn->work, add_conn, (void *) conn); + INIT_WORK(&conn->work, add_conn); schedule_work(&conn->work); } -static void del_conn(void *data) +static void del_conn(struct work_struct *work) { - struct hci_conn *conn = data; + struct hci_conn *conn = container_of(work, struct hci_conn, work); device_del(&conn->dev); } @@ -287,7 +287,7 @@ void hci_conn_del_sysfs(struct hci_conn *conn) { BT_DBG("conn %p", conn); - INIT_WORK(&conn->work, del_conn, (void *) conn); + INIT_WORK(&conn->work, del_conn); schedule_work(&conn->work); } diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index f753c40c11d2..55bb2634c088 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -77,12 +77,16 @@ static int port_cost(struct net_device *dev) * Called from work queue to allow for calling functions that * might sleep (such as speed check), and to debounce. */ -static void port_carrier_check(void *arg) +static void port_carrier_check(struct work_struct *work) { - struct net_device *dev = arg; struct net_bridge_port *p; + struct net_device *dev; struct net_bridge *br; + dev = container_of(work, struct net_bridge_port, + carrier_check.work)->dev; + work_release(work); + rtnl_lock(); p = dev->br_port; if (!p) @@ -276,7 +280,7 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br, p->port_no = index; br_init_port(p); p->state = BR_STATE_DISABLED; - INIT_WORK(&p->carrier_check, port_carrier_check, dev); + INIT_DELAYED_WORK_NAR(&p->carrier_check, port_carrier_check); br_stp_port_timer_init(p); kobject_init(&p->kobj); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 74258d86f256..3a534e94c7f3 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -82,7 +82,7 @@ struct net_bridge_port struct timer_list hold_timer; struct timer_list message_age_timer; struct kobject kobj; - struct work_struct carrier_check; + struct delayed_work carrier_check; struct rcu_head rcu; }; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6589adb14cbf..63f24c914ddb 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -56,7 +56,7 @@ static atomic_t trapped; static void zap_completion_queue(void); static void arp_reply(struct sk_buff *skb); -static void queue_process(void *p) +static void queue_process(struct work_struct *work) { unsigned long flags; struct sk_buff *skb; @@ -77,7 +77,7 @@ static void queue_process(void *p) } } -static DECLARE_WORK(send_queue, queue_process, NULL); +static DECLARE_WORK(send_queue, queue_process); void netpoll_queue(struct sk_buff *skb) { diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 9045438d6b36..36db5be2a9e9 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -31,8 +31,7 @@ struct inet_timewait_death_row dccp_death_row = { .tw_timer = TIMER_INITIALIZER(inet_twdr_hangman, 0, (unsigned long)&dccp_death_row), .twkill_work = __WORK_INITIALIZER(dccp_death_row.twkill_work, - inet_twdr_twkill_work, - &dccp_death_row), + inet_twdr_twkill_work), /* Short-time timewait calendar */ .twcal_hand = -1, diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c index cf51c87a971d..08386c102954 100644 --- a/net/ieee80211/softmac/ieee80211softmac_assoc.c +++ b/net/ieee80211/softmac/ieee80211softmac_assoc.c @@ -58,9 +58,11 @@ ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211soft } void -ieee80211softmac_assoc_timeout(void *d) +ieee80211softmac_assoc_timeout(struct work_struct *work) { - struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d; + struct ieee80211softmac_device *mac = + container_of(work, struct ieee80211softmac_device, + associnfo.timeout.work); struct ieee80211softmac_network *n; mutex_lock(&mac->associnfo.mutex); @@ -186,9 +188,11 @@ ieee80211softmac_assoc_notify_auth(struct net_device *dev, int event_type, void /* This function is called to handle userspace requests (asynchronously) */ void -ieee80211softmac_assoc_work(void *d) +ieee80211softmac_assoc_work(struct work_struct *work) { - struct ieee80211softmac_device *mac = (struct ieee80211softmac_device *)d; + struct ieee80211softmac_device *mac = + container_of(work, struct ieee80211softmac_device, + associnfo.work.work); struct ieee80211softmac_network *found = NULL; struct ieee80211_network *net = NULL, *best = NULL; int bssvalid; @@ -412,7 +416,7 @@ ieee80211softmac_handle_assoc_response(struct net_device * dev, network->authenticated = 0; /* we don't want to do this more than once ... */ network->auth_desynced_once = 1; - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); break; } default: @@ -446,7 +450,7 @@ ieee80211softmac_handle_disassoc(struct net_device * dev, ieee80211softmac_disassoc(mac); /* try to reassociate */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } @@ -466,7 +470,7 @@ ieee80211softmac_handle_reassoc_req(struct net_device * dev, dprintkl(KERN_INFO PFX "reassoc request from unknown network\n"); return 0; } - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c index 4cef39e171d0..2ae1833b657a 100644 --- a/net/ieee80211/softmac/ieee80211softmac_auth.c +++ b/net/ieee80211/softmac/ieee80211softmac_auth.c @@ -26,7 +26,7 @@ #include "ieee80211softmac_priv.h" -static void ieee80211softmac_auth_queue(void *data); +static void ieee80211softmac_auth_queue(struct work_struct *work); /* Queues an auth request to the desired AP */ int @@ -54,14 +54,14 @@ ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, auth->mac = mac; auth->retry = IEEE80211SOFTMAC_AUTH_RETRY_LIMIT; auth->state = IEEE80211SOFTMAC_AUTH_OPEN_REQUEST; - INIT_WORK(&auth->work, &ieee80211softmac_auth_queue, (void *)auth); + INIT_DELAYED_WORK(&auth->work, ieee80211softmac_auth_queue); /* Lock (for list) */ spin_lock_irqsave(&mac->lock, flags); /* add to list */ list_add_tail(&auth->list, &mac->auth_queue); - schedule_work(&auth->work); + schedule_delayed_work(&auth->work, 0); spin_unlock_irqrestore(&mac->lock, flags); return 0; @@ -70,14 +70,15 @@ ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, /* Sends an auth request to the desired AP and handles timeouts */ static void -ieee80211softmac_auth_queue(void *data) +ieee80211softmac_auth_queue(struct work_struct *work) { struct ieee80211softmac_device *mac; struct ieee80211softmac_auth_queue_item *auth; struct ieee80211softmac_network *net; unsigned long flags; - auth = (struct ieee80211softmac_auth_queue_item *)data; + auth = container_of(work, struct ieee80211softmac_auth_queue_item, + work.work); net = auth->net; mac = auth->mac; @@ -118,9 +119,11 @@ ieee80211softmac_auth_queue(void *data) /* Sends a response to an auth challenge (for shared key auth). */ static void -ieee80211softmac_auth_challenge_response(void *_aq) +ieee80211softmac_auth_challenge_response(struct work_struct *work) { - struct ieee80211softmac_auth_queue_item *aq = _aq; + struct ieee80211softmac_auth_queue_item *aq = + container_of(work, struct ieee80211softmac_auth_queue_item, + work.work); /* Send our response */ ieee80211softmac_send_mgt_frame(aq->mac, aq->net, IEEE80211_STYPE_AUTH, aq->state); @@ -228,8 +231,8 @@ ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth) * we have obviously already sent the initial auth * request. */ cancel_delayed_work(&aq->work); - INIT_WORK(&aq->work, &ieee80211softmac_auth_challenge_response, (void *)aq); - schedule_work(&aq->work); + INIT_DELAYED_WORK(&aq->work, &ieee80211softmac_auth_challenge_response); + schedule_delayed_work(&aq->work, 0); spin_unlock_irqrestore(&mac->lock, flags); return 0; case IEEE80211SOFTMAC_AUTH_SHARED_PASS: @@ -392,6 +395,6 @@ ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *de ieee80211softmac_deauth_from_net(mac, net); /* let's try to re-associate */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_event.c b/net/ieee80211/softmac/ieee80211softmac_event.c index f34fa2ef666b..b9015656cfb3 100644 --- a/net/ieee80211/softmac/ieee80211softmac_event.c +++ b/net/ieee80211/softmac/ieee80211softmac_event.c @@ -73,10 +73,12 @@ static char *event_descriptions[IEEE80211SOFTMAC_EVENT_LAST+1] = { static void -ieee80211softmac_notify_callback(void *d) +ieee80211softmac_notify_callback(struct work_struct *work) { - struct ieee80211softmac_event event = *(struct ieee80211softmac_event*) d; - kfree(d); + struct ieee80211softmac_event *pevent = + container_of(work, struct ieee80211softmac_event, work.work); + struct ieee80211softmac_event event = *pevent; + kfree(pevent); event.fun(event.mac->dev, event.event_type, event.context); } @@ -99,7 +101,7 @@ ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac, return -ENOMEM; eventptr->event_type = event; - INIT_WORK(&eventptr->work, ieee80211softmac_notify_callback, eventptr); + INIT_DELAYED_WORK(&eventptr->work, ieee80211softmac_notify_callback); eventptr->fun = fun; eventptr->context = context; eventptr->mac = mac; @@ -170,7 +172,7 @@ ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int eve /* User may have subscribed to ANY event, so * we tell them which event triggered it. */ eventptr->event_type = event; - schedule_work(&eventptr->work); + schedule_delayed_work(&eventptr->work, 0); } } } diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c index 33aff4f4a471..256207b71dc9 100644 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ b/net/ieee80211/softmac/ieee80211softmac_module.c @@ -58,8 +58,8 @@ struct net_device *alloc_ieee80211softmac(int sizeof_priv) INIT_LIST_HEAD(&softmac->events); mutex_init(&softmac->associnfo.mutex); - INIT_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work, softmac); - INIT_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout, softmac); + INIT_DELAYED_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work); + INIT_DELAYED_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout); softmac->start_scan = ieee80211softmac_start_scan_implementation; softmac->wait_for_scan = ieee80211softmac_wait_for_scan_implementation; softmac->stop_scan = ieee80211softmac_stop_scan_implementation; diff --git a/net/ieee80211/softmac/ieee80211softmac_priv.h b/net/ieee80211/softmac/ieee80211softmac_priv.h index 0642e090b8a7..c0dbe070e548 100644 --- a/net/ieee80211/softmac/ieee80211softmac_priv.h +++ b/net/ieee80211/softmac/ieee80211softmac_priv.h @@ -78,7 +78,7 @@ /* private definitions and prototypes */ /*** prototypes from _scan.c */ -void ieee80211softmac_scan(void *sm); +void ieee80211softmac_scan(struct work_struct *work); /* for internal use if scanning is needed */ int ieee80211softmac_start_scan(struct ieee80211softmac_device *mac); void ieee80211softmac_stop_scan(struct ieee80211softmac_device *mac); @@ -149,7 +149,7 @@ int ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *au int ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth); /*** prototypes from _assoc.c */ -void ieee80211softmac_assoc_work(void *d); +void ieee80211softmac_assoc_work(struct work_struct *work); int ieee80211softmac_handle_assoc_response(struct net_device * dev, struct ieee80211_assoc_response * resp, struct ieee80211_network * network); @@ -157,7 +157,7 @@ int ieee80211softmac_handle_disassoc(struct net_device * dev, struct ieee80211_disassoc * disassoc); int ieee80211softmac_handle_reassoc_req(struct net_device * dev, struct ieee80211_reassoc_request * reassoc); -void ieee80211softmac_assoc_timeout(void *d); +void ieee80211softmac_assoc_timeout(struct work_struct *work); void ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason); void ieee80211softmac_disassoc(struct ieee80211softmac_device *mac); @@ -207,7 +207,7 @@ struct ieee80211softmac_auth_queue_item { struct ieee80211softmac_device *mac; /* SoftMAC device */ u8 retry; /* Retry limit */ u8 state; /* Auth State */ - struct work_struct work; /* Work queue */ + struct delayed_work work; /* Work queue */ }; /* scanning information */ @@ -219,7 +219,8 @@ struct ieee80211softmac_scaninfo { stop:1; u8 skip_flags; struct completion finished; - struct work_struct softmac_scan; + struct delayed_work softmac_scan; + struct ieee80211softmac_device *mac; }; /* private event struct */ @@ -227,7 +228,7 @@ struct ieee80211softmac_event { struct list_head list; int event_type; void *event_context; - struct work_struct work; + struct delayed_work work; notify_function_ptr fun; void *context; struct ieee80211softmac_device *mac; diff --git a/net/ieee80211/softmac/ieee80211softmac_scan.c b/net/ieee80211/softmac/ieee80211softmac_scan.c index d31cf77498c4..a8326076581a 100644 --- a/net/ieee80211/softmac/ieee80211softmac_scan.c +++ b/net/ieee80211/softmac/ieee80211softmac_scan.c @@ -91,12 +91,14 @@ ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *sm) /* internal scanning implementation follows */ -void ieee80211softmac_scan(void *d) +void ieee80211softmac_scan(struct work_struct *work) { int invalid_channel; u8 current_channel_idx; - struct ieee80211softmac_device *sm = (struct ieee80211softmac_device *)d; - struct ieee80211softmac_scaninfo *si = sm->scaninfo; + struct ieee80211softmac_scaninfo *si = + container_of(work, struct ieee80211softmac_scaninfo, + softmac_scan.work); + struct ieee80211softmac_device *sm = si->mac; unsigned long flags; while (!(si->stop) && (si->current_channel_idx < si->number_channels)) { @@ -146,7 +148,8 @@ static inline struct ieee80211softmac_scaninfo *allocate_scaninfo(struct ieee802 struct ieee80211softmac_scaninfo *info = kmalloc(sizeof(struct ieee80211softmac_scaninfo), GFP_ATOMIC); if (unlikely(!info)) return NULL; - INIT_WORK(&info->softmac_scan, ieee80211softmac_scan, mac); + INIT_DELAYED_WORK(&info->softmac_scan, ieee80211softmac_scan); + info->mac = mac; init_completion(&info->finished); return info; } @@ -189,7 +192,7 @@ int ieee80211softmac_start_scan_implementation(struct net_device *dev) sm->scaninfo->started = 1; sm->scaninfo->stop = 0; INIT_COMPLETION(sm->scaninfo->finished); - schedule_work(&sm->scaninfo->softmac_scan); + schedule_delayed_work(&sm->scaninfo->softmac_scan, 0); spin_unlock_irqrestore(&sm->lock, flags); return 0; } diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c index 23068a830f7d..2ffaebd21c53 100644 --- a/net/ieee80211/softmac/ieee80211softmac_wx.c +++ b/net/ieee80211/softmac/ieee80211softmac_wx.c @@ -122,7 +122,7 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev, sm->associnfo.associating = 1; /* queue lower level code to do work (if necessary) */ - schedule_work(&sm->associnfo.work); + schedule_delayed_work(&sm->associnfo.work, 0); out: mutex_unlock(&sm->associnfo.mutex); @@ -356,7 +356,7 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, /* force reassociation */ mac->associnfo.bssvalid = 0; if (mac->associnfo.associated) - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); } else if (is_zero_ether_addr(data->ap_addr.sa_data)) { /* the bssid we have is no longer fixed */ mac->associnfo.bssfixed = 0; @@ -373,7 +373,7 @@ ieee80211softmac_wx_set_wap(struct net_device *net_dev, /* tell the other code that this bssid should be used no matter what */ mac->associnfo.bssfixed = 1; /* queue associate if new bssid or (old one again and not associated) */ - schedule_work(&mac->associnfo.work); + schedule_delayed_work(&mac->associnfo.work, 0); } out: diff --git a/net/ipv4/ipvs/ip_vs_ctl.c b/net/ipv4/ipvs/ip_vs_ctl.c index f261616e4602..9b933381ebbe 100644 --- a/net/ipv4/ipvs/ip_vs_ctl.c +++ b/net/ipv4/ipvs/ip_vs_ctl.c @@ -221,10 +221,10 @@ static void update_defense_level(void) * Timer for checking the defense */ #define DEFENSE_TIMER_PERIOD 1*HZ -static void defense_work_handler(void *data); -static DECLARE_WORK(defense_work, defense_work_handler, NULL); +static void defense_work_handler(struct work_struct *work); +static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); -static void defense_work_handler(void *data) +static void defense_work_handler(struct work_struct *work) { update_defense_level(); if (atomic_read(&ip_vs_dropentry)) diff --git a/net/irda/ircomm/ircomm_tty.c b/net/irda/ircomm/ircomm_tty.c index d50a02030ad7..262bda808d96 100644 --- a/net/irda/ircomm/ircomm_tty.c +++ b/net/irda/ircomm/ircomm_tty.c @@ -61,7 +61,7 @@ static void ircomm_tty_flush_buffer(struct tty_struct *tty); static void ircomm_tty_send_xchar(struct tty_struct *tty, char ch); static void ircomm_tty_wait_until_sent(struct tty_struct *tty, int timeout); static void ircomm_tty_hangup(struct tty_struct *tty); -static void ircomm_tty_do_softint(void *private_); +static void ircomm_tty_do_softint(struct work_struct *work); static void ircomm_tty_shutdown(struct ircomm_tty_cb *self); static void ircomm_tty_stop(struct tty_struct *tty); @@ -389,7 +389,7 @@ static int ircomm_tty_open(struct tty_struct *tty, struct file *filp) self->flow = FLOW_STOP; self->line = line; - INIT_WORK(&self->tqueue, ircomm_tty_do_softint, self); + INIT_WORK(&self->tqueue, ircomm_tty_do_softint); self->max_header_size = IRCOMM_TTY_HDR_UNINITIALISED; self->max_data_size = IRCOMM_TTY_DATA_UNINITIALISED; self->close_delay = 5*HZ/10; @@ -594,15 +594,16 @@ static void ircomm_tty_flush_buffer(struct tty_struct *tty) } /* - * Function ircomm_tty_do_softint (private_) + * Function ircomm_tty_do_softint (work) * * We use this routine to give the write wakeup to the user at at a * safe time (as fast as possible after write have completed). This * can be compared to the Tx interrupt. */ -static void ircomm_tty_do_softint(void *private_) +static void ircomm_tty_do_softint(struct work_struct *work) { - struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) private_; + struct ircomm_tty_cb *self = + container_of(work, struct ircomm_tty_cb, tqueue); struct tty_struct *tty; unsigned long flags; struct sk_buff *skb, *ctrl_skb; diff --git a/net/sctp/associola.c b/net/sctp/associola.c index ed0445fe85e7..88124696ba60 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -61,7 +61,7 @@ #include /* Forward declarations for internal functions. */ -static void sctp_assoc_bh_rcv(struct sctp_association *asoc); +static void sctp_assoc_bh_rcv(struct work_struct *work); /* 1st Level Abstractions. */ @@ -269,9 +269,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a /* Create an input queue. */ sctp_inq_init(&asoc->base.inqueue); - sctp_inq_set_th_handler(&asoc->base.inqueue, - (void (*)(void *))sctp_assoc_bh_rcv, - asoc); + sctp_inq_set_th_handler(&asoc->base.inqueue, sctp_assoc_bh_rcv); /* Create an output queue. */ sctp_outq_init(asoc, &asoc->outqueue); @@ -944,8 +942,11 @@ out: } /* Do delayed input processing. This is scheduled by sctp_rcv(). */ -static void sctp_assoc_bh_rcv(struct sctp_association *asoc) +static void sctp_assoc_bh_rcv(struct work_struct *work) { + struct sctp_association *asoc = + container_of(work, struct sctp_association, + base.inqueue.immediate); struct sctp_endpoint *ep; struct sctp_chunk *chunk; struct sock *sk; diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 9b6b394b66f6..a2b553721514 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -61,7 +61,7 @@ #include /* Forward declarations for internal helpers. */ -static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep); +static void sctp_endpoint_bh_rcv(struct work_struct *work); /* * Initialize the base fields of the endpoint structure. @@ -85,8 +85,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, sctp_inq_init(&ep->base.inqueue); /* Set its top-half handler */ - sctp_inq_set_th_handler(&ep->base.inqueue, - (void (*)(void *))sctp_endpoint_bh_rcv, ep); + sctp_inq_set_th_handler(&ep->base.inqueue, sctp_endpoint_bh_rcv); /* Initialize the bind addr area */ sctp_bind_addr_init(&ep->base.bind_addr, 0); @@ -311,8 +310,11 @@ int sctp_endpoint_is_peeled_off(struct sctp_endpoint *ep, /* Do delayed input processing. This is scheduled by sctp_rcv(). * This may be called on BH or task time. */ -static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep) +static void sctp_endpoint_bh_rcv(struct work_struct *work) { + struct sctp_endpoint *ep = + container_of(work, struct sctp_endpoint, + base.inqueue.immediate); struct sctp_association *asoc; struct sock *sk; struct sctp_transport *transport; diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index cf6deed7e849..71b07466e880 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -54,7 +54,7 @@ void sctp_inq_init(struct sctp_inq *queue) queue->in_progress = NULL; /* Create a task for delivering data. */ - INIT_WORK(&queue->immediate, NULL, NULL); + INIT_WORK(&queue->immediate, NULL); queue->malloced = 0; } @@ -97,7 +97,7 @@ void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *chunk) * on the BH related data structures. */ list_add_tail(&chunk->list, &q->in_chunk_list); - q->immediate.func(q->immediate.data); + q->immediate.func(&q->immediate); } /* Extract a chunk from an SCTP inqueue. @@ -205,9 +205,8 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) * The intent is that this routine will pull stuff out of the * inqueue and process it. */ -void sctp_inq_set_th_handler(struct sctp_inq *q, - void (*callback)(void *), void *arg) +void sctp_inq_set_th_handler(struct sctp_inq *q, work_func_t callback) { - INIT_WORK(&q->immediate, callback, arg); + INIT_WORK(&q->immediate, callback); } diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 7736b23c3f03..ba924d40df7d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -358,7 +358,7 @@ static void xfrm_policy_gc_kill(struct xfrm_policy *policy) xfrm_pol_put(policy); } -static void xfrm_policy_gc_task(void *data) +static void xfrm_policy_gc_task(struct work_struct *work) { struct xfrm_policy *policy; struct hlist_node *entry, *tmp; @@ -546,7 +546,7 @@ static inline int xfrm_byidx_should_resize(int total) static DEFINE_MUTEX(hash_resize_mutex); -static void xfrm_hash_resize(void *__unused) +static void xfrm_hash_resize(struct work_struct *__unused) { int dir, total; @@ -563,7 +563,7 @@ static void xfrm_hash_resize(void *__unused) mutex_unlock(&hash_resize_mutex); } -static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL); +static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize); /* Generate new index... KAME seems to generate them ordered by cost * of an absolute inpredictability of ordering of rules. This will not pass. */ @@ -2080,7 +2080,7 @@ static void __init xfrm_policy_init(void) panic("XFRM: failed to allocate bydst hash\n"); } - INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task, NULL); + INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task); register_netdevice_notifier(&xfrm_dev_notifier); } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 899de9ed22a6..40c527179843 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -115,7 +115,7 @@ static unsigned long xfrm_hash_new_size(void) static DEFINE_MUTEX(hash_resize_mutex); -static void xfrm_hash_resize(void *__unused) +static void xfrm_hash_resize(struct work_struct *__unused) { struct hlist_head *ndst, *nsrc, *nspi, *odst, *osrc, *ospi; unsigned long nsize, osize; @@ -168,7 +168,7 @@ out_unlock: mutex_unlock(&hash_resize_mutex); } -static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize, NULL); +static DECLARE_WORK(xfrm_hash_work, xfrm_hash_resize); DECLARE_WAIT_QUEUE_HEAD(km_waitq); EXPORT_SYMBOL(km_waitq); @@ -207,7 +207,7 @@ static void xfrm_state_gc_destroy(struct xfrm_state *x) kfree(x); } -static void xfrm_state_gc_task(void *data) +static void xfrm_state_gc_task(struct work_struct *data) { struct xfrm_state *x; struct hlist_node *entry, *tmp; @@ -1568,6 +1568,6 @@ void __init xfrm_state_init(void) panic("XFRM: Cannot allocate bydst/bysrc/byspi hashes."); xfrm_state_hmask = ((sz / sizeof(struct hlist_head)) - 1); - INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task, NULL); + INIT_WORK(&xfrm_state_gc_work, xfrm_state_gc_task); } diff --git a/sound/aoa/aoa-gpio.h b/sound/aoa/aoa-gpio.h index 3a61f3115573..ee64f5de8966 100644 --- a/sound/aoa/aoa-gpio.h +++ b/sound/aoa/aoa-gpio.h @@ -59,10 +59,10 @@ struct gpio_methods { }; struct gpio_notification { + struct delayed_work work; notify_func_t notify; void *data; void *gpio_private; - struct work_struct work; struct mutex mutex; }; diff --git a/sound/aoa/core/snd-aoa-gpio-feature.c b/sound/aoa/core/snd-aoa-gpio-feature.c index 40eb47eccf9a..2b03bc798bcb 100644 --- a/sound/aoa/core/snd-aoa-gpio-feature.c +++ b/sound/aoa/core/snd-aoa-gpio-feature.c @@ -195,9 +195,10 @@ static void ftr_gpio_all_amps_restore(struct gpio_runtime *rt) ftr_gpio_set_lineout(rt, (s>>2)&1); } -static void ftr_handle_notify(void *data) +static void ftr_handle_notify(struct work_struct *work) { - struct gpio_notification *notif = data; + struct gpio_notification *notif = + container_of(work, struct gpio_notification, work.work); mutex_lock(¬if->mutex); if (notif->notify) @@ -253,12 +254,9 @@ static void ftr_gpio_init(struct gpio_runtime *rt) ftr_gpio_all_amps_off(rt); rt->implementation_private = 0; - INIT_WORK(&rt->headphone_notify.work, ftr_handle_notify, - &rt->headphone_notify); - INIT_WORK(&rt->line_in_notify.work, ftr_handle_notify, - &rt->line_in_notify); - INIT_WORK(&rt->line_out_notify.work, ftr_handle_notify, - &rt->line_out_notify); + INIT_DELAYED_WORK(&rt->headphone_notify.work, ftr_handle_notify); + INIT_DELAYED_WORK(&rt->line_in_notify.work, ftr_handle_notify); + INIT_DELAYED_WORK(&rt->line_out_notify.work, ftr_handle_notify); mutex_init(&rt->headphone_notify.mutex); mutex_init(&rt->line_in_notify.mutex); mutex_init(&rt->line_out_notify.mutex); @@ -287,7 +285,7 @@ static irqreturn_t ftr_handle_notify_irq(int xx, void *data) { struct gpio_notification *notif = data; - schedule_work(¬if->work); + schedule_delayed_work(¬if->work, 0); return IRQ_HANDLED; } diff --git a/sound/aoa/core/snd-aoa-gpio-pmf.c b/sound/aoa/core/snd-aoa-gpio-pmf.c index 2836c3218391..5ca2220eac7d 100644 --- a/sound/aoa/core/snd-aoa-gpio-pmf.c +++ b/sound/aoa/core/snd-aoa-gpio-pmf.c @@ -69,9 +69,10 @@ static void pmf_gpio_all_amps_restore(struct gpio_runtime *rt) pmf_gpio_set_lineout(rt, (s>>2)&1); } -static void pmf_handle_notify(void *data) +static void pmf_handle_notify(struct work_struct *work) { - struct gpio_notification *notif = data; + struct gpio_notification *notif = + container_of(work, struct gpio_notification, work.work); mutex_lock(¬if->mutex); if (notif->notify) @@ -83,12 +84,9 @@ static void pmf_gpio_init(struct gpio_runtime *rt) { pmf_gpio_all_amps_off(rt); rt->implementation_private = 0; - INIT_WORK(&rt->headphone_notify.work, pmf_handle_notify, - &rt->headphone_notify); - INIT_WORK(&rt->line_in_notify.work, pmf_handle_notify, - &rt->line_in_notify); - INIT_WORK(&rt->line_out_notify.work, pmf_handle_notify, - &rt->line_out_notify); + INIT_DELAYED_WORK(&rt->headphone_notify.work, pmf_handle_notify); + INIT_DELAYED_WORK(&rt->line_in_notify.work, pmf_handle_notify); + INIT_DELAYED_WORK(&rt->line_out_notify.work, pmf_handle_notify); mutex_init(&rt->headphone_notify.mutex); mutex_init(&rt->line_in_notify.mutex); mutex_init(&rt->line_out_notify.mutex); @@ -129,7 +127,7 @@ static void pmf_handle_notify_irq(void *data) { struct gpio_notification *notif = data; - schedule_work(¬if->work); + schedule_delayed_work(¬if->work, 0); } static int pmf_set_notify(struct gpio_runtime *rt, diff --git a/sound/i2c/other/ak4114.c b/sound/i2c/other/ak4114.c index 12ffffc9e814..d2f2c5078e65 100644 --- a/sound/i2c/other/ak4114.c +++ b/sound/i2c/other/ak4114.c @@ -35,7 +35,7 @@ MODULE_LICENSE("GPL"); #define AK4114_ADDR 0x00 /* fixed address */ -static void ak4114_stats(void *); +static void ak4114_stats(struct work_struct *work); static void reg_write(struct ak4114 *ak4114, unsigned char reg, unsigned char val) { @@ -158,7 +158,7 @@ void snd_ak4114_reinit(struct ak4114 *chip) reg_write(chip, AK4114_REG_PWRDN, old | AK4114_RST | AK4114_PWN); /* bring up statistics / event queing */ chip->init = 0; - INIT_WORK(&chip->work, ak4114_stats, chip); + INIT_DELAYED_WORK(&chip->work, ak4114_stats); queue_delayed_work(chip->workqueue, &chip->work, HZ / 10); } @@ -561,9 +561,9 @@ int snd_ak4114_check_rate_and_errors(struct ak4114 *ak4114, unsigned int flags) return res; } -static void ak4114_stats(void *data) +static void ak4114_stats(struct work_struct *work) { - struct ak4114 *chip = (struct ak4114 *)data; + struct ak4114 *chip = container_of(work, struct ak4114, work.work); if (chip->init) return; diff --git a/sound/pci/ac97/ac97_codec.c b/sound/pci/ac97/ac97_codec.c index 6577b2325357..7abcb10b2754 100644 --- a/sound/pci/ac97/ac97_codec.c +++ b/sound/pci/ac97/ac97_codec.c @@ -1927,9 +1927,10 @@ static int snd_ac97_dev_disconnect(struct snd_device *device) static struct snd_ac97_build_ops null_build_ops; #ifdef CONFIG_SND_AC97_POWER_SAVE -static void do_update_power(void *data) +static void do_update_power(struct work_struct *work) { - update_power_regs(data); + update_power_regs( + container_of(work, struct snd_ac97, power_work.work)); } #endif @@ -1989,7 +1990,7 @@ int snd_ac97_mixer(struct snd_ac97_bus *bus, struct snd_ac97_template *template, mutex_init(&ac97->page_mutex); #ifdef CONFIG_SND_AC97_POWER_SAVE ac97->power_workq = create_workqueue("ac97"); - INIT_WORK(&ac97->power_work, do_update_power, ac97); + INIT_DELAYED_WORK(&ac97->power_work, do_update_power); #endif #ifdef CONFIG_PCI diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index 9c3d7ac08068..71482c15a852 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -272,10 +272,11 @@ EXPORT_SYMBOL(snd_hda_queue_unsol_event); /* * process queueud unsolicited events */ -static void process_unsol_events(void *data) +static void process_unsol_events(struct work_struct *work) { - struct hda_bus *bus = data; - struct hda_bus_unsolicited *unsol = bus->unsol; + struct hda_bus_unsolicited *unsol = + container_of(work, struct hda_bus_unsolicited, work); + struct hda_bus *bus = unsol->bus; struct hda_codec *codec; unsigned int rp, caddr, res; @@ -314,7 +315,8 @@ static int init_unsol_queue(struct hda_bus *bus) kfree(unsol); return -ENOMEM; } - INIT_WORK(&unsol->work, process_unsol_events, bus); + INIT_WORK(&unsol->work, process_unsol_events); + unsol->bus = bus; bus->unsol = unsol; return 0; } diff --git a/sound/pci/hda/hda_local.h b/sound/pci/hda/hda_local.h index f9416c36396e..9ca1baf860bd 100644 --- a/sound/pci/hda/hda_local.h +++ b/sound/pci/hda/hda_local.h @@ -206,6 +206,7 @@ struct hda_bus_unsolicited { /* workqueue */ struct workqueue_struct *workq; struct work_struct work; + struct hda_bus *bus; }; /* diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c index 2fbe1d183fce..8f074c7936e6 100644 --- a/sound/ppc/tumbler.c +++ b/sound/ppc/tumbler.c @@ -942,10 +942,11 @@ static void check_mute(struct snd_pmac *chip, struct pmac_gpio *gp, int val, int } static struct work_struct device_change; +static struct snd_pmac *device_change_chip; -static void device_change_handler(void *self) +static void device_change_handler(struct work_struct *work) { - struct snd_pmac *chip = self; + struct snd_pmac *chip = device_change_chip; struct pmac_tumbler *mix; int headphone, lineout; @@ -1417,7 +1418,8 @@ int __init snd_pmac_tumbler_init(struct snd_pmac *chip) chip->resume = tumbler_resume; #endif - INIT_WORK(&device_change, device_change_handler, (void *)chip); + INIT_WORK(&device_change, device_change_handler); + device_change_chip = chip; #ifdef PMAC_SUPPORT_AUTOMUTE if ((mix->headphone_irq >=0 || mix->lineout_irq >= 0) -- cgit v1.2.3 From ed7b1f6d6ea1054ea4fe293a7fd8015fc3803d93 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 23 Nov 2006 23:34:49 -0500 Subject: Input: serio - make serio_register_driver() return errors Perform actual driver registration right in serio_register_driver() instead of offloading it to kseriod and return proper error code to callers if driver registration fails. Note that driver <-> port matching is still done by kseriod to speed up boot process since probing for PS/2 mice and keyboards is pretty slow. Signed-off-by: Dmitry Torokhov --- drivers/input/serio/serio.c | 107 ++++++++++++++++++++++++++++++-------------- include/linux/serio.h | 7 +-- 2 files changed, 75 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index cd55ddb7df4f..8c717042f611 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -44,7 +44,7 @@ EXPORT_SYMBOL(serio_interrupt); EXPORT_SYMBOL(__serio_register_port); EXPORT_SYMBOL(serio_unregister_port); EXPORT_SYMBOL(serio_unregister_child_port); -EXPORT_SYMBOL(__serio_register_driver); +EXPORT_SYMBOL(serio_register_driver); EXPORT_SYMBOL(serio_unregister_driver); EXPORT_SYMBOL(serio_open); EXPORT_SYMBOL(serio_close); @@ -61,10 +61,10 @@ static LIST_HEAD(serio_list); static struct bus_type serio_bus; -static void serio_add_driver(struct serio_driver *drv); static void serio_add_port(struct serio *serio); static void serio_reconnect_port(struct serio *serio); static void serio_disconnect_port(struct serio *serio); +static void serio_attach_driver(struct serio_driver *drv); static int serio_connect_driver(struct serio *serio, struct serio_driver *drv) { @@ -168,10 +168,10 @@ static void serio_find_driver(struct serio *serio) */ enum serio_event_type { - SERIO_RESCAN, - SERIO_RECONNECT, + SERIO_RESCAN_PORT, + SERIO_RECONNECT_PORT, SERIO_REGISTER_PORT, - SERIO_REGISTER_DRIVER, + SERIO_ATTACH_DRIVER, }; struct serio_event { @@ -186,11 +186,12 @@ static LIST_HEAD(serio_event_list); static DECLARE_WAIT_QUEUE_HEAD(serio_wait); static struct task_struct *serio_task; -static void serio_queue_event(void *object, struct module *owner, - enum serio_event_type event_type) +static int serio_queue_event(void *object, struct module *owner, + enum serio_event_type event_type) { unsigned long flags; struct serio_event *event; + int retval = 0; spin_lock_irqsave(&serio_event_lock, flags); @@ -209,24 +210,34 @@ static void serio_queue_event(void *object, struct module *owner, } } - if ((event = kmalloc(sizeof(struct serio_event), GFP_ATOMIC))) { - if (!try_module_get(owner)) { - printk(KERN_WARNING "serio: Can't get module reference, dropping event %d\n", event_type); - kfree(event); - goto out; - } - - event->type = event_type; - event->object = object; - event->owner = owner; + event = kmalloc(sizeof(struct serio_event), GFP_ATOMIC); + if (!event) { + printk(KERN_ERR + "serio: Not enough memory to queue event %d\n", + event_type); + retval = -ENOMEM; + goto out; + } - list_add_tail(&event->node, &serio_event_list); - wake_up(&serio_wait); - } else { - printk(KERN_ERR "serio: Not enough memory to queue event %d\n", event_type); + if (!try_module_get(owner)) { + printk(KERN_WARNING + "serio: Can't get module reference, dropping event %d\n", + event_type); + kfree(event); + retval = -EINVAL; + goto out; } + + event->type = event_type; + event->object = object; + event->owner = owner; + + list_add_tail(&event->node, &serio_event_list); + wake_up(&serio_wait); + out: spin_unlock_irqrestore(&serio_event_lock, flags); + return retval; } static void serio_free_event(struct serio_event *event) @@ -304,17 +315,17 @@ static void serio_handle_event(void) serio_add_port(event->object); break; - case SERIO_RECONNECT: + case SERIO_RECONNECT_PORT: serio_reconnect_port(event->object); break; - case SERIO_RESCAN: + case SERIO_RESCAN_PORT: serio_disconnect_port(event->object); serio_find_driver(event->object); break; - case SERIO_REGISTER_DRIVER: - serio_add_driver(event->object); + case SERIO_ATTACH_DRIVER: + serio_attach_driver(event->object); break; default: @@ -666,12 +677,12 @@ static void serio_disconnect_port(struct serio *serio) void serio_rescan(struct serio *serio) { - serio_queue_event(serio, NULL, SERIO_RESCAN); + serio_queue_event(serio, NULL, SERIO_RESCAN_PORT); } void serio_reconnect(struct serio *serio) { - serio_queue_event(serio, NULL, SERIO_RECONNECT); + serio_queue_event(serio, NULL, SERIO_RECONNECT_PORT); } /* @@ -766,22 +777,52 @@ static int serio_driver_remove(struct device *dev) return 0; } -static void serio_add_driver(struct serio_driver *drv) +static void serio_attach_driver(struct serio_driver *drv) { int error; - error = driver_register(&drv->driver); + error = driver_attach(&drv->driver); if (error) - printk(KERN_ERR - "serio: driver_register() failed for %s, error: %d\n", + printk(KERN_WARNING + "serio: driver_attach() failed for %s with error %d\n", drv->driver.name, error); } -void __serio_register_driver(struct serio_driver *drv, struct module *owner) +int serio_register_driver(struct serio_driver *drv) { + int manual_bind = drv->manual_bind; + int error; + drv->driver.bus = &serio_bus; - serio_queue_event(drv, owner, SERIO_REGISTER_DRIVER); + /* + * Temporarily disable automatic binding because probing + * takes long time and we are better off doing it in kseriod + */ + drv->manual_bind = 1; + + error = driver_register(&drv->driver); + if (error) { + printk(KERN_ERR + "serio: driver_register() failed for %s, error: %d\n", + drv->driver.name, error); + return error; + } + + /* + * Restore original bind mode and let kseriod bind the + * driver to free ports + */ + if (!manual_bind) { + drv->manual_bind = 0; + error = serio_queue_event(drv, NULL, SERIO_ATTACH_DRIVER); + if (error) { + driver_unregister(&drv->driver); + return error; + } + } + + return 0; } void serio_unregister_driver(struct serio_driver *drv) diff --git a/include/linux/serio.h b/include/linux/serio.h index 8f52228390ab..0f478a8791a2 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -86,12 +86,7 @@ static inline void serio_register_port(struct serio *serio) void serio_unregister_port(struct serio *serio); void serio_unregister_child_port(struct serio *serio); -void __serio_register_driver(struct serio_driver *drv, struct module *owner); -static inline void serio_register_driver(struct serio_driver *drv) -{ - __serio_register_driver(drv, THIS_MODULE); -} - +int serio_register_driver(struct serio_driver *drv); void serio_unregister_driver(struct serio_driver *drv); static inline int serio_write(struct serio *serio, unsigned char data) -- cgit v1.2.3 From e4477d2d1bc3e6c76e83926f7fa8c8f94ba42615 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 24 Nov 2006 00:43:09 -0500 Subject: Input: add to kernel-api docbook Add input subsystem to kernel-api docbook. Enhance some function and parameter comments. Signed-off-by: Randy Dunlap Signed-off-by: Dmitry Torokhov --- Documentation/DocBook/kernel-api.tmpl | 8 ++++++++ drivers/input/ff-core.c | 4 ++-- drivers/input/ff-memless.c | 2 +- include/linux/input.h | 18 +++++++++--------- 4 files changed, 20 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index a166675c4303..a1af278f9901 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -533,4 +533,12 @@ X!Idrivers/video/console/fonts.c --> + + + Input Subsystem +!Iinclude/linux/input.h +!Edrivers/input/input.c +!Edrivers/input/ff-core.c +!Edrivers/input/ff-memless.c + diff --git a/drivers/input/ff-core.c b/drivers/input/ff-core.c index 35656cadc914..783b3412cead 100644 --- a/drivers/input/ff-core.c +++ b/drivers/input/ff-core.c @@ -203,7 +203,7 @@ static int erase_effect(struct input_dev *dev, int effect_id, } /** - * input_ff_erase - erase an effect from device + * input_ff_erase - erase a force-feedback effect from device * @dev: input device to erase effect from * @effect_id: id of the ffect to be erased * @file: purported owner of the request @@ -347,7 +347,7 @@ EXPORT_SYMBOL_GPL(input_ff_create); /** * input_ff_free() - frees force feedback portion of input device - * @dev: input device supporintg force feedback + * @dev: input device supporting force feedback * * This function is only needed in error path as input core will * automatically free force feedback structures when device is diff --git a/drivers/input/ff-memless.c b/drivers/input/ff-memless.c index cd8b7297e6df..eba18b6ac5e4 100644 --- a/drivers/input/ff-memless.c +++ b/drivers/input/ff-memless.c @@ -460,7 +460,7 @@ static void ml_ff_destroy(struct ff_device *ff) } /** - * input_ff_create_memless() - create memoryless FF device + * input_ff_create_memless() - create memoryless force-feedback device * @dev: input device supporting force-feedback * @data: driver-specific data to be passed into @play_effect * @play_effect: driver-specific method for playing FF effect diff --git a/include/linux/input.h b/include/linux/input.h index c38507ba38b5..4e61158b06a0 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -663,7 +663,7 @@ struct input_absinfo { #define BUS_GSC 0x1A /* - * Values describing the status of an effect + * Values describing the status of a force-feedback effect */ #define FF_STATUS_STOPPED 0x00 #define FF_STATUS_PLAYING 0x01 @@ -680,7 +680,7 @@ struct input_absinfo { */ /** - * struct ff_replay - defines scheduling of the effect + * struct ff_replay - defines scheduling of the force-feedback effect * @length: duration of the effect * @delay: delay before effect should start playing */ @@ -690,7 +690,7 @@ struct ff_replay { }; /** - * struct ff_trigger - defines what triggers the effect + * struct ff_trigger - defines what triggers the force-feedback effect * @button: number of the button triggering the effect * @interval: controls how soon the effect can be re-triggered */ @@ -700,7 +700,7 @@ struct ff_trigger { }; /** - * struct ff_envelope - generic effect envelope + * struct ff_envelope - generic force-feedback effect envelope * @attack_length: duration of the attack (ms) * @attack_level: level at the beginning of the attack * @fade_length: duration of fade (ms) @@ -719,7 +719,7 @@ struct ff_envelope { }; /** - * struct ff_constant_effect - defines parameters of a constant effect + * struct ff_constant_effect - defines parameters of a constant force-feedback effect * @level: strength of the effect; may be negative * @envelope: envelope data */ @@ -729,7 +729,7 @@ struct ff_constant_effect { }; /** - * struct ff_ramp_effect - defines parameters of a ramp effect + * struct ff_ramp_effect - defines parameters of a ramp force-feedback effect * @start_level: beginning strength of the effect; may be negative * @end_level: final strength of the effect; may be negative * @envelope: envelope data @@ -741,7 +741,7 @@ struct ff_ramp_effect { }; /** - * struct ff_condition_effect - defines a spring or friction effect + * struct ff_condition_effect - defines a spring or friction force-feedback effect * @right_saturation: maximum level when joystick moved all way to the right * @left_saturation: same for the left side * @right_coeff: controls how fast the force grows when the joystick moves @@ -762,7 +762,7 @@ struct ff_condition_effect { }; /** - * struct ff_periodic_effect - defines parameters of a periodic effect + * struct ff_periodic_effect - defines parameters of a periodic force-feedback effect * @waveform: kind of the effect (wave) * @period: period of the wave (ms) * @magnitude: peak value @@ -793,7 +793,7 @@ struct ff_periodic_effect { }; /** - * struct ff_rumble_effect - defines parameters of a periodic effect + * struct ff_rumble_effect - defines parameters of a periodic force-feedback effect * @strong_magnitude: magnitude of the heavy motor * @weak_magnitude: magnitude of the light one * -- cgit v1.2.3 From 701e054e0c2db82359f0454c7ed4fd24346d52eb Mon Sep 17 00:00:00 2001 From: Vasily Tarasov Date: Sat, 25 Nov 2006 11:09:22 -0800 Subject: [PATCH] mounstats NULL pointer dereference OpenVZ developers team has encountered the following problem in 2.6.19-rc6 kernel. After some seconds of running script while [[ 1 ]] do find /proc -name mountstats | xargs cat done this Oops appears: BUG: unable to handle kernel NULL pointer dereference at virtual address 00000010 printing eip: c01a6b70 *pde = 00000000 Oops: 0000 [#1] SMP Modules linked in: xt_length ipt_ttl xt_tcpmss ipt_TCPMSS iptable_mangle iptable_filter xt_multiport xt_limit ipt_tos ipt_REJECT ip_tables x_tables parport_pc lp parport sunrpc af_packet thermal processor fan button battery asus_acpi ac ohci_hcd ehci_hcd usbcore i2c_nforce2 i2c_core tg3 floppy pata_amd ide_cd cdrom sata_nv libata CPU: 1 EIP: 0060:[] Not tainted VLI EFLAGS: 00010246 (2.6.19-rc6 #2) EIP is at mountstats_open+0x70/0xf0 eax: 00000000 ebx: e6247030 ecx: e62470f8 edx: 00000000 esi: 00000000 edi: c01a6b00 ebp: c33b83c0 esp: f4105eb4 ds: 007b es: 007b ss: 0068 Process cat (pid: 6044, ti=f4105000 task=f4104a70 task.ti=f4105000) Stack: c33b83c0 c04ee940 f46a4a80 c33b83c0 e4df31b4 c01a6b00 f4105000 c0169231 e4df31b4 c33b83c0 c33b83c0 f4105f20 00000003 f4105000 c0169445 f2503cf0 f7f8c4c0 00008000 c33b83c0 00000000 00008000 c0169350 f4105f20 00008000 Call Trace: [] mountstats_open+0x0/0xf0 [] __dentry_open+0x181/0x250 [] nameidata_to_filp+0x35/0x50 [] do_filp_open+0x50/0x60 [] seq_read+0xc6/0x300 [] get_unused_fd+0x31/0xc0 [] do_sys_open+0x63/0x110 [] sys_open+0x27/0x30 [] sysenter_past_esp+0x56/0x79 ======================= Code: 45 74 8b 54 24 20 89 44 24 08 8b 42 f0 31 d2 e8 47 cb f8 ff 85 c0 89 c3 74 51 8d 80 a0 04 00 00 e8 46 06 2c 00 8b 83 48 04 00 00 <8b> 78 10 85 ff 74 03 f0 ff 07 b0 01 86 83 a0 04 00 00 f0 ff 4b EIP: [] mountstats_open+0x70/0xf0 SS:ESP 0068:f4105eb4 The problem is that task->nsproxy can be equal NULL for some time during task exit. This patch fixes the BUG. Signed-off-by: Vasily Tarasov Cc: Herbert Poetzl Cc: "Serge E. Hallyn" Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 3 ++- include/linux/nsproxy.h | 4 +++- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/base.c b/fs/proc/base.c index 8df27401d292..795319c54f72 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -442,7 +442,8 @@ static int mountstats_open(struct inode *inode, struct file *file) if (task) { task_lock(task); - namespace = task->nsproxy->namespace; + if (task->nsproxy) + namespace = task->nsproxy->namespace; if (namespace) get_namespace(namespace); task_unlock(task); diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index f6baecdeecd6..971d1c6dfc4b 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -45,8 +45,10 @@ static inline void exit_task_namespaces(struct task_struct *p) { struct nsproxy *ns = p->nsproxy; if (ns) { - put_nsproxy(ns); + task_lock(p); p->nsproxy = NULL; + task_unlock(p); + put_nsproxy(ns); } } #endif -- cgit v1.2.3 From ee3ce191e8eaa4cc15c51a28b34143b36404c4f5 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 25 Nov 2006 11:09:36 -0800 Subject: [PATCH] Enforce "unsigned long flags;" when spinlocking Make it break or warn if you pass to spin_lock_irqsave() and friends something different from "unsigned long flags;". Suprisingly large amount of these was caught by recent commit c53421b18f205c5f97c604ae55c6a921f034b0f6 and others. Idea is largely from FRV typechecking. Suggestions from Andrew Morton. All stupid typos in first version fixed. Passes allmodconfig on i386, x86_64, alpha, arm as well as my usual config. Note #1: checking with sparse is still needed, because a driver can save and pass around flags or something. So far patch is very intrusive. Note #2: techically, we should break only if sizeof(flags) < sizeof(unsigned long), however, the more pain for getting suspicious code into kernel, the better. Signed-off-by: Alexey Dobriyan Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irqflags.h | 37 ++++++++++++++++++++++++++++----- include/linux/spinlock.h | 53 ++++++++++++++++++++++++++++++++++++++++-------- 2 files changed, 76 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 412e025bc5c7..4fe740bf4eae 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -11,6 +11,12 @@ #ifndef _LINUX_TRACE_IRQFLAGS_H #define _LINUX_TRACE_IRQFLAGS_H +#define BUILD_CHECK_IRQ_FLAGS(flags) \ + do { \ + BUILD_BUG_ON(sizeof(flags) != sizeof(unsigned long)); \ + typecheck(unsigned long, flags); \ + } while (0) + #ifdef CONFIG_TRACE_IRQFLAGS extern void trace_hardirqs_on(void); extern void trace_hardirqs_off(void); @@ -50,10 +56,15 @@ #define local_irq_disable() \ do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0) #define local_irq_save(flags) \ - do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_local_irq_save(flags); \ + trace_hardirqs_off(); \ + } while (0) #define local_irq_restore(flags) \ do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ @@ -69,8 +80,16 @@ */ # define raw_local_irq_disable() local_irq_disable() # define raw_local_irq_enable() local_irq_enable() -# define raw_local_irq_save(flags) local_irq_save(flags) -# define raw_local_irq_restore(flags) local_irq_restore(flags) +# define raw_local_irq_save(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + local_irq_save(flags); \ + } while (0) +# define raw_local_irq_restore(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + local_irq_restore(flags); \ + } while (0) #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT @@ -80,7 +99,11 @@ raw_safe_halt(); \ } while (0) -#define local_save_flags(flags) raw_local_save_flags(flags) +#define local_save_flags(flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_local_save_flags(flags); \ + } while (0) #define irqs_disabled() \ ({ \ @@ -90,7 +113,11 @@ raw_irqs_disabled_flags(flags); \ }) -#define irqs_disabled_flags(flags) raw_irqs_disabled_flags(flags) +#define irqs_disabled_flags(flags) \ +({ \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + raw_irqs_disabled_flags(flags); \ +}) #endif /* CONFIG_X86 */ #endif diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index b800d2d68b32..54ad37089c49 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -52,6 +52,7 @@ #include #include #include +#include #include @@ -183,13 +184,37 @@ do { \ #define read_lock(lock) _read_lock(lock) #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -#define spin_lock_irqsave(lock, flags) flags = _spin_lock_irqsave(lock) -#define read_lock_irqsave(lock, flags) flags = _read_lock_irqsave(lock) -#define write_lock_irqsave(lock, flags) flags = _write_lock_irqsave(lock) +#define spin_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave(lock); \ + } while (0) +#define read_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _read_lock_irqsave(lock); \ + } while (0) +#define write_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _write_lock_irqsave(lock); \ + } while (0) #else -#define spin_lock_irqsave(lock, flags) _spin_lock_irqsave(lock, flags) -#define read_lock_irqsave(lock, flags) _read_lock_irqsave(lock, flags) -#define write_lock_irqsave(lock, flags) _write_lock_irqsave(lock, flags) +#define spin_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _spin_lock_irqsave(lock, flags); \ + } while (0) +#define read_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _read_lock_irqsave(lock, flags); \ + } while (0) +#define write_lock_irqsave(lock, flags) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _write_lock_irqsave(lock, flags); \ + } while (0) #endif #define spin_lock_irq(lock) _spin_lock_irq(lock) @@ -225,15 +250,24 @@ do { \ #endif #define spin_unlock_irqrestore(lock, flags) \ - _spin_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _spin_unlock_irqrestore(lock, flags); \ + } while (0) #define spin_unlock_bh(lock) _spin_unlock_bh(lock) #define read_unlock_irqrestore(lock, flags) \ - _read_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _read_unlock_irqrestore(lock, flags); \ + } while (0) #define read_unlock_bh(lock) _read_unlock_bh(lock) #define write_unlock_irqrestore(lock, flags) \ - _write_unlock_irqrestore(lock, flags) + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + _write_unlock_irqrestore(lock, flags); \ + } while (0) #define write_unlock_bh(lock) _write_unlock_bh(lock) #define spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) @@ -247,6 +281,7 @@ do { \ #define spin_trylock_irqsave(lock, flags) \ ({ \ + BUILD_CHECK_IRQ_FLAGS(flags); \ local_irq_save(flags); \ spin_trylock(lock) ? \ 1 : ({ local_irq_restore(flags); 0; }); \ -- cgit v1.2.3 From cfd3ef2346f924d6c0e82236c20fdb3a8840136a Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sat, 25 Nov 2006 11:09:37 -0800 Subject: [PATCH] lockdep: spin_lock_irqsave_nested() Introduce spin_lock_irqsave_nested(); implementation from: http://lkml.org/lkml/2006/6/1/122 Patch from: http://lkml.org/lkml/2006/9/13/258 [akpm@osdl.org: two compile fixes] Signed-off-by: Arjan van de Ven Signed-off-by: Jiri Kosina Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/spinlock.h | 18 ++++++++++++++++++ include/linux/spinlock_api_smp.h | 2 ++ kernel/spinlock.c | 21 +++++++++++++++++++++ 3 files changed, 41 insertions(+) (limited to 'include/linux') diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 54ad37089c49..57f670d78f7c 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -199,6 +199,21 @@ do { \ BUILD_CHECK_IRQ_FLAGS(flags); \ flags = _write_lock_irqsave(lock); \ } while (0) + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave_nested(lock, subclass); \ + } while (0) +#else +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + BUILD_CHECK_IRQ_FLAGS(flags); \ + flags = _spin_lock_irqsave(lock); \ + } while (0) +#endif + #else #define spin_lock_irqsave(lock, flags) \ do { \ @@ -215,6 +230,9 @@ do { \ BUILD_CHECK_IRQ_FLAGS(flags); \ _write_lock_irqsave(lock, flags); \ } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + spin_lock_irqsave(lock, flags) + #endif #define spin_lock_irq(lock) _spin_lock_irq(lock) diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index 8828b8155e9c..8a2307ce7296 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -32,6 +32,8 @@ void __lockfunc _read_lock_irq(rwlock_t *lock) __acquires(lock); void __lockfunc _write_lock_irq(rwlock_t *lock) __acquires(lock); unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock) __acquires(lock); +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) + __acquires(lock); unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock) __acquires(lock); unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock) diff --git a/kernel/spinlock.c b/kernel/spinlock.c index 476c3741511b..2c6c2bf85514 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -293,6 +293,27 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_spin_lock_nested); +unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) +{ + unsigned long flags; + + local_irq_save(flags); + preempt_disable(); + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + /* + * On lockdep we dont want the hand-coded irq-enable of + * _raw_spin_lock_flags() code, because lockdep assumes + * that interrupts are not re-enabled during lock-acquire: + */ +#ifdef CONFIG_PROVE_SPIN_LOCKING + _raw_spin_lock(lock); +#else + _raw_spin_lock_flags(lock, &flags); +#endif + return flags; +} + +EXPORT_SYMBOL(_spin_lock_irqsave_nested); #endif -- cgit v1.2.3 From b8e6ec865fd1d8838b6ce9516977b65e9f08f876 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 26 Nov 2006 16:27:17 -0800 Subject: Revert "[PATCH] Enforce "unsigned long flags;" when spinlocking" This reverts commit ee3ce191e8eaa4cc15c51a28b34143b36404c4f5, since it broke on at least ARM, MIPS and PA-RISC due to complicated header file dependencies. Conflicts in include/linux/spinlock.h (due to the "nested" variety fixes) fixed up by hand. Cc: Alexey Dobriyan Cc: Ralf Baechle Cc: Kyle McMartin Cc: Russell King Signed-off-by: Linus Torvalds --- include/linux/irqflags.h | 37 ++++---------------------- include/linux/spinlock.h | 69 +++++++++++------------------------------------- 2 files changed, 20 insertions(+), 86 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 4fe740bf4eae..412e025bc5c7 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -11,12 +11,6 @@ #ifndef _LINUX_TRACE_IRQFLAGS_H #define _LINUX_TRACE_IRQFLAGS_H -#define BUILD_CHECK_IRQ_FLAGS(flags) \ - do { \ - BUILD_BUG_ON(sizeof(flags) != sizeof(unsigned long)); \ - typecheck(unsigned long, flags); \ - } while (0) - #ifdef CONFIG_TRACE_IRQFLAGS extern void trace_hardirqs_on(void); extern void trace_hardirqs_off(void); @@ -56,15 +50,10 @@ #define local_irq_disable() \ do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0) #define local_irq_save(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_local_irq_save(flags); \ - trace_hardirqs_off(); \ - } while (0) + do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0) #define local_irq_restore(flags) \ do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ @@ -80,16 +69,8 @@ */ # define raw_local_irq_disable() local_irq_disable() # define raw_local_irq_enable() local_irq_enable() -# define raw_local_irq_save(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - local_irq_save(flags); \ - } while (0) -# define raw_local_irq_restore(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - local_irq_restore(flags); \ - } while (0) +# define raw_local_irq_save(flags) local_irq_save(flags) +# define raw_local_irq_restore(flags) local_irq_restore(flags) #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */ #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT @@ -99,11 +80,7 @@ raw_safe_halt(); \ } while (0) -#define local_save_flags(flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_local_save_flags(flags); \ - } while (0) +#define local_save_flags(flags) raw_local_save_flags(flags) #define irqs_disabled() \ ({ \ @@ -113,11 +90,7 @@ raw_irqs_disabled_flags(flags); \ }) -#define irqs_disabled_flags(flags) \ -({ \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - raw_irqs_disabled_flags(flags); \ -}) +#define irqs_disabled_flags(flags) raw_irqs_disabled_flags(flags) #endif /* CONFIG_X86 */ #endif diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 57f670d78f7c..8451052ca66f 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -52,7 +52,6 @@ #include #include #include -#include #include @@ -184,52 +183,24 @@ do { \ #define read_lock(lock) _read_lock(lock) #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) -#define spin_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave(lock); \ - } while (0) -#define read_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _read_lock_irqsave(lock); \ - } while (0) -#define write_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _write_lock_irqsave(lock); \ - } while (0) + +#define spin_lock_irqsave(lock, flags) flags = _spin_lock_irqsave(lock) +#define read_lock_irqsave(lock, flags) flags = _read_lock_irqsave(lock) +#define write_lock_irqsave(lock, flags) flags = _write_lock_irqsave(lock) #ifdef CONFIG_DEBUG_LOCK_ALLOC -#define spin_lock_irqsave_nested(lock, flags, subclass) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave_nested(lock, subclass); \ - } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + flags = _spin_lock_irqsave_nested(lock, subclass) #else -#define spin_lock_irqsave_nested(lock, flags, subclass) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - flags = _spin_lock_irqsave(lock); \ - } while (0) +#define spin_lock_irqsave_nested(lock, flags, subclass) \ + flags = _spin_lock_irqsave(lock) #endif #else -#define spin_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _spin_lock_irqsave(lock, flags); \ - } while (0) -#define read_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _read_lock_irqsave(lock, flags); \ - } while (0) -#define write_lock_irqsave(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _write_lock_irqsave(lock, flags); \ - } while (0) + +#define spin_lock_irqsave(lock, flags) _spin_lock_irqsave(lock, flags) +#define read_lock_irqsave(lock, flags) _read_lock_irqsave(lock, flags) +#define write_lock_irqsave(lock, flags) _write_lock_irqsave(lock, flags) #define spin_lock_irqsave_nested(lock, flags, subclass) \ spin_lock_irqsave(lock, flags) @@ -268,24 +239,15 @@ do { \ #endif #define spin_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _spin_unlock_irqrestore(lock, flags); \ - } while (0) + _spin_unlock_irqrestore(lock, flags) #define spin_unlock_bh(lock) _spin_unlock_bh(lock) #define read_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _read_unlock_irqrestore(lock, flags); \ - } while (0) + _read_unlock_irqrestore(lock, flags) #define read_unlock_bh(lock) _read_unlock_bh(lock) #define write_unlock_irqrestore(lock, flags) \ - do { \ - BUILD_CHECK_IRQ_FLAGS(flags); \ - _write_unlock_irqrestore(lock, flags); \ - } while (0) + _write_unlock_irqrestore(lock, flags) #define write_unlock_bh(lock) _write_unlock_bh(lock) #define spin_trylock_bh(lock) __cond_lock(lock, _spin_trylock_bh(lock)) @@ -299,7 +261,6 @@ do { \ #define spin_trylock_irqsave(lock, flags) \ ({ \ - BUILD_CHECK_IRQ_FLAGS(flags); \ local_irq_save(flags); \ spin_trylock(lock) ? \ 1 : ({ local_irq_restore(flags); 0; }); \ -- cgit v1.2.3 From 2ea5814472c3c910aed5c5b60f1f3b1000e353f1 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 26 Nov 2006 19:05:22 -0800 Subject: Fix 'ALIGN()' macro, take 2 You wouldn't think that doing an ALIGN() macro that aligns something up to a power-of-two boundary would be likely to have bugs, would you? But hey, in the wonderful world of mixing integer types, you have to be careful. This just makes sure that the alignment is interpreted in the same type as the thing to be aligned. Thanks to Roland Dreier, who noticed that the amso1100 driver got broken by the previous fix (that just extended the mask to "unsigned long", but was still broken in "unsigned long long" - it just happened to be the same on 64-bit architectures). See commit 4c8bd7eeee4c8f157fb61fb64b57500990b42e0e for the history of bugs here... Acked-by: Roland Dreier Cc: Andrew Morton Cc: David Miller Cc: Al Viro Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 24b611147adb..b9b5e4ba166a 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -30,8 +30,10 @@ extern const char linux_banner[]; #define STACK_MAGIC 0xdeadbeef +#define ALIGN(x,a) __ALIGN_MASK(x,(typeof(x))(a)-1) +#define __ALIGN_MASK(x,mask) (((x)+(mask))&~(mask)) + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) -#define ALIGN(x,a) (((x)+(a)-1UL)&~((a)-1UL)) #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f)) #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d)) #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y)) -- cgit v1.2.3 From e81c73596704793e73e6dbb478f41686f15a4b34 Mon Sep 17 00:00:00 2001 From: David S. Miller Date: Tue, 28 Nov 2006 20:53:39 -0800 Subject: [NET]: Fix MAX_HEADER setting. MAX_HEADER is either set to LL_MAX_HEADER or LL_MAX_HEADER + 48, and this is controlled by a set of CONFIG_* ifdef tests. It is trying to use LL_MAX_HEADER + 48 when any of the tunnels are enabled which set hard_header_len like this: dev->hard_header_len = LL_MAX_HEADER + sizeof(struct xxx); The correct set of tunnel drivers which do this are: ipip ip_gre ip6_tunnel sit so make the ifdef test match. Noticed by Patrick McHardy and with help from Herbert Xu. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 9264139bd8df..83b8c4f1d69d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -93,8 +93,10 @@ struct netpoll_info; #endif #endif -#if !defined(CONFIG_NET_IPIP) && \ - !defined(CONFIG_IPV6) && !defined(CONFIG_IPV6_MODULE) +#if !defined(CONFIG_NET_IPIP) && !defined(CONFIG_NET_IPIP_MODULE) && \ + !defined(CONFIG_NET_IPGRE) && !defined(CONFIG_NET_IPGRE_MODULE) && \ + !defined(CONFIG_IPV6_SIT) && !defined(CONFIG_IPV6_SIT_MODULE) && \ + !defined(CONFIG_IPV6_TUNNEL) && !defined(CONFIG_IPV6_TUNNEL_MODULE) #define MAX_HEADER LL_MAX_HEADER #else #define MAX_HEADER (LL_MAX_HEADER + 48) -- cgit v1.2.3 From 583bb86fbb9e85287f020fe4eb5352a0ec3c66a3 Mon Sep 17 00:00:00 2001 From: Nicolas Schichan Date: Wed, 18 Oct 2006 15:14:55 +0200 Subject: [MIPS] Add support for kexec A tiny userland application loading the kernel and invoking kexec_load for mips is available here: http://chac.le-poulpe.net/~nico/kexec/kexec-2006-10-18.tar.gz Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 17 ++++++++ arch/mips/kernel/Makefile | 2 + arch/mips/kernel/machine_kexec.c | 85 ++++++++++++++++++++++++++++++++++++++ arch/mips/kernel/relocate_kernel.S | 80 +++++++++++++++++++++++++++++++++++ arch/mips/kernel/scall32-o32.S | 2 +- arch/mips/kernel/scall64-64.S | 2 +- arch/mips/kernel/scall64-n32.S | 2 +- arch/mips/kernel/scall64-o32.S | 2 +- include/asm-mips/kexec.h | 32 ++++++++++++++ include/linux/kexec.h | 2 + 10 files changed, 222 insertions(+), 4 deletions(-) create mode 100644 arch/mips/kernel/machine_kexec.c create mode 100644 arch/mips/kernel/relocate_kernel.S create mode 100644 include/asm-mips/kexec.h (limited to 'include/linux') diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 1443024b1c7c..c0da0ffe8d57 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -766,6 +766,23 @@ config TOSHIBA_RBTX4938 endchoice +config KEXEC + bool "Kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but it is indepedent of the system firmware. And like a reboot + you can start any kernel with it, not just Linux. + + The name comes from the similiarity to the exec system call. + + It is an ongoing process to be certain the hardware in a machine + is properly shutdown, so do not be surprised if this code does not + initially work for you. It may help to enable device hotplugging + support. As of this writing the exact hardware interface is + strongly in flux, so no good recommendation can be made. + source "arch/mips/ddb5xxx/Kconfig" source "arch/mips/gt64120/ev64120/Kconfig" source "arch/mips/jazz/Kconfig" diff --git a/arch/mips/kernel/Makefile b/arch/mips/kernel/Makefile index 6bfbbed0897e..f35b739d0a12 100644 --- a/arch/mips/kernel/Makefile +++ b/arch/mips/kernel/Makefile @@ -67,6 +67,8 @@ obj-$(CONFIG_64BIT) += cpu-bugs64.o obj-$(CONFIG_I8253) += i8253.o +obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o + CFLAGS_cpu-bugs64.o = $(shell if $(CC) $(CFLAGS) -Wa,-mdaddi -c -o /dev/null -xc /dev/null >/dev/null 2>&1; then echo "-DHAVE_AS_SET_DADDI"; fi) EXTRA_AFLAGS := $(CFLAGS) diff --git a/arch/mips/kernel/machine_kexec.c b/arch/mips/kernel/machine_kexec.c new file mode 100644 index 000000000000..e0ad754c7edd --- /dev/null +++ b/arch/mips/kernel/machine_kexec.c @@ -0,0 +1,85 @@ +/* + * machine_kexec.c for kexec + * Created by on Thu Oct 12 15:15:06 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include + +#include +#include + +const extern unsigned char relocate_new_kernel[]; +const extern unsigned int relocate_new_kernel_size; + +extern unsigned long kexec_start_address; +extern unsigned long kexec_indirection_page; + +int +machine_kexec_prepare(struct kimage *kimage) +{ + return 0; +} + +void +machine_kexec_cleanup(struct kimage *kimage) +{ +} + +void +machine_shutdown(void) +{ +} + +void +machine_crash_shutdown(struct pt_regs *regs) +{ +} + +void +machine_kexec(struct kimage *image) +{ + unsigned long reboot_code_buffer; + unsigned long entry; + unsigned long *ptr; + + reboot_code_buffer = + (unsigned long)page_address(image->control_code_page); + + kexec_start_address = image->start; + kexec_indirection_page = phys_to_virt(image->head & PAGE_MASK); + + memcpy((void*)reboot_code_buffer, relocate_new_kernel, + relocate_new_kernel_size); + + /* + * The generic kexec code builds a page list with physical + * addresses. they are directly accessible through KSEG0 (or + * CKSEG0 or XPHYS if on 64bit system), hence the + * pys_to_virt() call. + */ + for (ptr = &image->head; (entry = *ptr) && !(entry &IND_DONE); + ptr = (entry & IND_INDIRECTION) ? + phys_to_virt(entry & PAGE_MASK) : ptr + 1) { + if (*ptr & IND_SOURCE || *ptr & IND_INDIRECTION || + *ptr & IND_DESTINATION) + *ptr = phys_to_virt(*ptr); + } + + /* + * we do not want to be bothered. + */ + local_irq_disable(); + + flush_icache_range(reboot_code_buffer, + reboot_code_buffer + KEXEC_CONTROL_CODE_SIZE); + + printk("Will call new kernel at %08x\n", image->start); + printk("Bye ...\n"); + flush_cache_all(); + ((void (*)(void))reboot_code_buffer)(); +} diff --git a/arch/mips/kernel/relocate_kernel.S b/arch/mips/kernel/relocate_kernel.S new file mode 100644 index 000000000000..a3f0d00c1334 --- /dev/null +++ b/arch/mips/kernel/relocate_kernel.S @@ -0,0 +1,80 @@ +/* + * relocate_kernel.S for kexec + * Created by on Thu Oct 12 17:49:57 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include +#include +#include +#include +#include + + .globl relocate_new_kernel +relocate_new_kernel: + + PTR_L s0, kexec_indirection_page + PTR_L s1, kexec_start_address + +process_entry: + PTR_L s2, (s0) + PTR_ADD s0, s0, SZREG + + /* destination page */ + and s3, s2, 0x1 + beq s3, zero, 1f + and s4, s2, ~0x1 /* store destination addr in s4 */ + move a0, s4 + b process_entry + +1: + /* indirection page, update s0 */ + and s3, s2, 0x2 + beq s3, zero, 1f + and s0, s2, ~0x2 + b process_entry + +1: + /* done page */ + and s3, s2, 0x4 + beq s3, zero, 1f + b done +1: + /* source page */ + and s3, s2, 0x8 + beq s3, zero, process_entry + and s2, s2, ~0x8 + li s6, (1 << PAGE_SHIFT) / SZREG + +copy_word: + /* copy page word by word */ + REG_L s5, (s2) + REG_S s5, (s4) + INT_ADD s4, s4, SZREG + INT_ADD s2, s2, SZREG + INT_SUB s6, s6, 1 + beq s6, zero, process_entry + b copy_word + b process_entry + +done: + /* jump to kexec_start_address */ + j s1 + + .globl kexec_start_address +kexec_start_address: + .long 0x0 + + .globl kexec_indirection_page +kexec_indirection_page: + .long 0x0 + +relocate_new_kernel_end: + + .globl relocate_new_kernel_size +relocate_new_kernel_size: + .long relocate_new_kernel_end - relocate_new_kernel diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index a95f37de080e..7c0b3936ba44 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -653,7 +653,7 @@ einval: li v0, -EINVAL sys sys_move_pages 6 sys sys_set_robust_list 2 sys sys_get_robust_list 3 /* 4310 */ - sys sys_ni_syscall 0 + sys sys_kexec_load 4 sys sys_getcpu 3 sys sys_epoll_pwait 6 .endm diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 8fb0f60f657b..e569b846e9a3 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -468,6 +468,6 @@ sys_call_table: PTR sys_move_pages PTR sys_set_robust_list PTR sys_get_robust_list - PTR sys_ni_syscall /* 5270 */ + PTR sys_kexec_load /* 5270 */ PTR sys_getcpu PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 0da5ca2040ff..5b18f265d75b 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -394,6 +394,6 @@ EXPORT(sysn32_call_table) PTR sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list - PTR sys_ni_syscall + PTR compat_sys_kexec_load PTR sys_getcpu PTR sys_epoll_pwait diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index b9d00cae8b5f..e91379c1be1d 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -516,7 +516,7 @@ sys_call_table: PTR compat_sys_move_pages PTR compat_sys_set_robust_list PTR compat_sys_get_robust_list /* 4310 */ - PTR sys_ni_syscall + PTR compat_sys_kexec_load PTR sys_getcpu PTR sys_epoll_pwait .size sys_call_table,.-sys_call_table diff --git a/include/asm-mips/kexec.h b/include/asm-mips/kexec.h new file mode 100644 index 000000000000..b25267ebcb09 --- /dev/null +++ b/include/asm-mips/kexec.h @@ -0,0 +1,32 @@ +/* + * kexec.h for kexec + * Created by on Thu Oct 12 14:59:34 2006 + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#ifndef _MIPS_KEXEC +# define _MIPS_KEXEC + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (0x20000000) +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (0x20000000) + /* Maximum address we can use for the control code buffer */ +#define KEXEC_CONTROL_MEMORY_LIMIT (0x20000000) + +#define KEXEC_CONTROL_CODE_SIZE 4096 + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_MIPS + +#define MAX_NOTE_BYTES 1024 + +static inline void crash_setup_regs(struct pt_regs *newregs, + struct pt_regs *oldregs) +{ + /* Dummy implementation for now */ +} + +#endif /* !_MIPS_KEXEC */ diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 6427949ddf99..a4ede62b339d 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -122,6 +122,8 @@ extern struct kimage *kexec_crash_image; #define KEXEC_ARCH_IA_64 (50 << 16) #define KEXEC_ARCH_S390 (22 << 16) #define KEXEC_ARCH_SH (42 << 16) +#define KEXEC_ARCH_MIPS_LE (10 << 16) +#define KEXEC_ARCH_MIPS ( 8 << 16) #define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ -- cgit v1.2.3 From 139a7bdc2b9391a4d0362190d9e5625dcf580105 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 30 Nov 2006 04:40:22 +0100 Subject: mqueue.h: don't include linux/types.h This #include is not required. Signed-off-by: Alexey Dobriyan Signed-off-by: Adrian Bunk --- include/linux/mqueue.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mqueue.h b/include/linux/mqueue.h index 8db9d75541a6..8b5a79615fbf 100644 --- a/include/linux/mqueue.h +++ b/include/linux/mqueue.h @@ -18,8 +18,6 @@ #ifndef _LINUX_MQUEUE_H #define _LINUX_MQUEUE_H -#include - #define MQ_PRIO_MAX 32768 /* per-uid limit of kernel memory used by mqueue, in bytes */ #define MQ_BYTES_MAX 819200 -- cgit v1.2.3 From e20ec9911bfef897459b9f8aeaf6eadb0920299a Mon Sep 17 00:00:00 2001 From: Jim Cromie Date: Thu, 30 Nov 2006 04:46:13 +0100 Subject: fix spelling error in include/linux/kernel.h Signed-off-by: Adrian Bunk --- include/linux/kernel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index b9b5e4ba166a..6738283ac385 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -65,7 +65,7 @@ struct user; * context (spinlock, irq-handler, ...). * * This is a useful debugging help to be able to catch problems early and not - * be biten later when the calling function happens to sleep when it is not + * be bitten later when the calling function happens to sleep when it is not * supposed to. */ #ifdef CONFIG_PREEMPT_VOLUNTARY -- cgit v1.2.3 From 98c4f0c336afe4318c12397bc74910d86ee036a2 Mon Sep 17 00:00:00 2001 From: Chase Venters Date: Thu, 30 Nov 2006 04:53:49 +0100 Subject: Fix jiffies.h comment jiffies.h includes a comment informing that jiffies_64 must be read with the assistance of the xtime_lock seqlock. The comment text, however, calls jiffies_64 "not volatile", which should probably read "not atomic". Signed-off-by: Chase Venters Signed-off-by: Adrian Bunk --- include/linux/jiffies.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index c8d5f207c3d4..0ec6e28bccd2 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -74,7 +74,7 @@ #define __jiffy_data __attribute__((section(".data"))) /* - * The 64-bit value is not volatile - you MUST NOT read it + * The 64-bit value is not atomic - you MUST NOT read it * without sampling the sequence number in xtime_lock. * get_jiffies_64() will do this for you as appropriate. */ -- cgit v1.2.3 From 0779bf2d2ecc4d9b1e9437ae659f50e6776a7666 Mon Sep 17 00:00:00 2001 From: Matt LaPlante Date: Thu, 30 Nov 2006 05:24:39 +0100 Subject: Fix misc .c/.h comment typos Fix various .c/.h typos in comments (no code changes). Signed-off-by: Matt LaPlante Signed-off-by: Adrian Bunk --- arch/cris/arch-v10/drivers/eeprom.c | 6 +++--- arch/cris/arch-v10/drivers/i2c.c | 2 +- arch/cris/arch-v10/kernel/kgdb.c | 2 +- arch/ia64/hp/common/sba_iommu.c | 8 ++++---- arch/sh64/lib/dbg.c | 2 +- drivers/atm/iphase.c | 2 +- drivers/char/rio/riocmd.c | 2 +- drivers/char/rio/rioinit.c | 2 +- drivers/char/rio/rioparam.c | 6 +++--- drivers/ide/ide-floppy.c | 2 +- drivers/isdn/hardware/eicon/os_4bri.c | 2 +- drivers/isdn/hisax/hfc4s8s_l1.h | 2 +- drivers/media/dvb/ttpci/budget-patch.c | 8 ++++---- drivers/net/e100.c | 2 +- drivers/net/e1000/e1000_hw.c | 2 +- drivers/net/sk98lin/h/skdrv2nd.h | 2 +- drivers/net/sk98lin/skdim.c | 4 ++-- drivers/net/wireless/ipw2200.c | 4 ++-- drivers/parisc/ccio-dma.c | 2 +- drivers/parisc/iosapic.c | 6 +++--- drivers/pci/hotplug/ibmphp_hpc.c | 2 +- drivers/s390/net/claw.h | 2 +- drivers/scsi/aic94xx/aic94xx_reg_def.h | 2 +- drivers/scsi/aic94xx/aic94xx_sds.c | 4 ++-- drivers/scsi/ncr53c8xx.c | 14 +++++++------- drivers/scsi/ncr53c8xx.h | 6 +++--- drivers/usb/host/u132-hcd.c | 6 +++--- drivers/usb/misc/usb_u132.h | 2 +- drivers/usb/serial/digi_acceleport.c | 2 +- fs/reiserfs/journal.c | 6 +++--- include/asm-m68knommu/mcfmbus.h | 2 +- include/asm-parisc/dma.h | 6 +++--- include/asm-parisc/pci.h | 2 +- include/asm-parisc/ropes.h | 2 +- include/linux/ixjuser.h | 2 +- include/linux/reiserfs_fs_sb.h | 2 +- net/wanrouter/af_wanpipe.c | 4 ++-- net/wanrouter/wanmain.c | 2 +- sound/oss/cs46xx.c | 6 +++--- 39 files changed, 71 insertions(+), 71 deletions(-) (limited to 'include/linux') diff --git a/arch/cris/arch-v10/drivers/eeprom.c b/arch/cris/arch-v10/drivers/eeprom.c index 6e1f191a71e3..284ebfda03f0 100644 --- a/arch/cris/arch-v10/drivers/eeprom.c +++ b/arch/cris/arch-v10/drivers/eeprom.c @@ -1,7 +1,7 @@ /*!***************************************************************************** *! -*! Implements an interface for i2c compatible eeproms to run under linux. -*! Supports 2k, 8k(?) and 16k. Uses adaptive timing adjustents by +*! Implements an interface for i2c compatible eeproms to run under Linux. +*! Supports 2k, 8k(?) and 16k. Uses adaptive timing adjustments by *! Johan.Adolfsson@axis.com *! *! Probing results: @@ -51,7 +51,7 @@ *! Revision 1.8 2001/06/15 13:24:29 jonashg *! * Added verification of pointers from userspace in read and write. *! * Made busy counter volatile. -*! * Added define for inital write delay. +*! * Added define for initial write delay. *! * Removed warnings by using loff_t instead of unsigned long. *! *! Revision 1.7 2001/06/14 15:26:54 jonashg diff --git a/arch/cris/arch-v10/drivers/i2c.c b/arch/cris/arch-v10/drivers/i2c.c index 6114596c3b33..092c724a645f 100644 --- a/arch/cris/arch-v10/drivers/i2c.c +++ b/arch/cris/arch-v10/drivers/i2c.c @@ -47,7 +47,7 @@ *! Update Port B register and shadow even when running with hardware support *! to avoid glitches when reading bits *! Never set direction to out in i2c_inbyte -*! Removed incorrect clock togling at end of i2c_inbyte +*! Removed incorrect clock toggling at end of i2c_inbyte *! *! Revision 1.8 2002/08/13 06:31:53 starvik *! Made SDA and SCL line configurable diff --git a/arch/cris/arch-v10/kernel/kgdb.c b/arch/cris/arch-v10/kernel/kgdb.c index 34528da98817..07628a13c6c4 100644 --- a/arch/cris/arch-v10/kernel/kgdb.c +++ b/arch/cris/arch-v10/kernel/kgdb.c @@ -33,7 +33,7 @@ *! *! Revision 1.2 2002/11/19 14:35:24 starvik *! Changes from linux 2.4 -*! Changed struct initializer syntax to the currently prefered notation +*! Changed struct initializer syntax to the currently preferred notation *! *! Revision 1.1 2001/12/17 13:59:27 bjornw *! Initial revision diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c index db8e1fcfa047..14691cda05c3 100644 --- a/arch/ia64/hp/common/sba_iommu.c +++ b/arch/ia64/hp/common/sba_iommu.c @@ -75,7 +75,7 @@ ** If a device prefetches beyond the end of a valid pdir entry, it will cause ** a hard failure, ie. MCA. Version 3.0 and later of the zx1 LBA should ** disconnect on 4k boundaries and prevent such issues. If the device is -** particularly agressive, this option will keep the entire pdir valid such +** particularly aggressive, this option will keep the entire pdir valid such ** that prefetching will hit a valid address. This could severely impact ** error containment, and is therefore off by default. The page that is ** used for spill-over is poisoned, so that should help debugging somewhat. @@ -258,10 +258,10 @@ static u64 prefetch_spill_page; /* ** DMA_CHUNK_SIZE is used by the SCSI mid-layer to break up -** (or rather not merge) DMA's into managable chunks. +** (or rather not merge) DMAs into manageable chunks. ** On parisc, this is more of the software/tuning constraint -** rather than the HW. I/O MMU allocation alogorithms can be -** faster with smaller size is (to some degree). +** rather than the HW. I/O MMU allocation algorithms can be +** faster with smaller sizes (to some degree). */ #define DMA_CHUNK_SIZE (BITS_PER_LONG*iovp_size) diff --git a/arch/sh64/lib/dbg.c b/arch/sh64/lib/dbg.c index 1326f45f31eb..4310fc87444e 100644 --- a/arch/sh64/lib/dbg.c +++ b/arch/sh64/lib/dbg.c @@ -383,7 +383,7 @@ void show_excp_regs(char *from, int trapnr, int signr, struct pt_regs *regs) /* ======================================================================= */ /* -** Depending on scan the MMU, Data or Instrction side +** Depending on scan the MMU, Data or Instruction side ** looking for a valid mapping matching Eaddr & asid. ** Return -1 if not found or the TLB id entry otherwise. ** Note: it works only for 4k pages! diff --git a/drivers/atm/iphase.c b/drivers/atm/iphase.c index 9ed1c60048f0..bb7ef570514c 100644 --- a/drivers/atm/iphase.c +++ b/drivers/atm/iphase.c @@ -305,7 +305,7 @@ static void clear_lockup (struct atm_vcc *vcc, IADEV *dev) { ** | R | NZ | 5-bit exponent | 9-bit mantissa | ** +----+----+------------------+-------------------------------+ ** -** R = reserverd (written as 0) +** R = reserved (written as 0) ** NZ = 0 if 0 cells/sec; 1 otherwise ** ** if NZ = 1, rate = 1.mmmmmmmmm x 2^(eeeee) cells/sec diff --git a/drivers/char/rio/riocmd.c b/drivers/char/rio/riocmd.c index 4df6ab2206a1..167ebc84e8d7 100644 --- a/drivers/char/rio/riocmd.c +++ b/drivers/char/rio/riocmd.c @@ -922,7 +922,7 @@ int RIOUnUse(unsigned long iPortP, struct CmdBlk *CmdBlkP) ** ** Packet is an actual packet structure to be filled in with the packet ** information associated with the command. You need to fill in everything, -** as the command processore doesn't process the command packet in any way. +** as the command processor doesn't process the command packet in any way. ** ** The PreFuncP is called before the packet is enqueued on the host rup. ** PreFuncP is called as (*PreFuncP)(PreArg, CmdBlkP);. PreFuncP must diff --git a/drivers/char/rio/rioinit.c b/drivers/char/rio/rioinit.c index 99f3df02b61c..0794844369d6 100644 --- a/drivers/char/rio/rioinit.c +++ b/drivers/char/rio/rioinit.c @@ -222,7 +222,7 @@ int RIOBoardTest(unsigned long paddr, void __iomem *caddr, unsigned char type, i ** which value will be written into memory. ** Call with op set to zero means that the RAM will not be read and checked ** before it is written. -** Call with op not zero, and the RAM will be read and compated with val[op-1] +** Call with op not zero and the RAM will be read and compared with val[op-1] ** to check that the data from the previous phase was retained. */ diff --git a/drivers/char/rio/rioparam.c b/drivers/char/rio/rioparam.c index 1066d9760704..bb498d24adcc 100644 --- a/drivers/char/rio/rioparam.c +++ b/drivers/char/rio/rioparam.c @@ -87,8 +87,8 @@ static char *_rioparam_c_sccs_ = "@(#)rioparam.c 1.3"; ** command bit set onto the port. The command bit is in the len field, ** and gets ORed in with the actual byte count. ** -** When you send a packet with the command bit set, then the first -** data byte ( data[0] ) is interpretted as the command to execute. +** When you send a packet with the command bit set the first +** data byte (data[0]) is interpreted as the command to execute. ** It also governs what data structure overlay should accompany the packet. ** Commands are defined in cirrus/cirrus.h ** @@ -103,7 +103,7 @@ static char *_rioparam_c_sccs_ = "@(#)rioparam.c 1.3"; ** ** Most commands do not use the remaining bytes in the data array. The ** exceptions are OPEN MOPEN and CONFIG. (NB. As with the SI CONFIG and -** OPEN are currently analagous). With these three commands the following +** OPEN are currently analogous). With these three commands the following ** 11 data bytes are all used to pass config information such as baud rate etc. ** The fields are also defined in cirrus.h. Some contain straightforward ** information such as the transmit XON character. Two contain the transmit and diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c index 8ccee9c769f8..e3a267622bb6 100644 --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -1635,7 +1635,7 @@ static int idefloppy_begin_format(ide_drive_t *drive, int __user *arg) /* ** Get ATAPI_FORMAT_UNIT progress indication. ** -** Userland gives a pointer to an int. The int is set to a progresss +** Userland gives a pointer to an int. The int is set to a progress ** indicator 0-65536, with 65536=100%. ** ** If the drive does not support format progress indication, we just check diff --git a/drivers/isdn/hardware/eicon/os_4bri.c b/drivers/isdn/hardware/eicon/os_4bri.c index 11e6f937c1e4..7b4ec3f60dbf 100644 --- a/drivers/isdn/hardware/eicon/os_4bri.c +++ b/drivers/isdn/hardware/eicon/os_4bri.c @@ -464,7 +464,7 @@ int diva_4bri_init_card(diva_os_xdi_adapter_t * a) /* ** Cleanup function will be called for master adapter only -** this is garanteed by design: cleanup callback is set +** this is guaranteed by design: cleanup callback is set ** by master adapter only */ static int diva_4bri_cleanup_adapter(diva_os_xdi_adapter_t * a) diff --git a/drivers/isdn/hisax/hfc4s8s_l1.h b/drivers/isdn/hisax/hfc4s8s_l1.h index e8f9c077fa85..9d5d2a56b4e9 100644 --- a/drivers/isdn/hisax/hfc4s8s_l1.h +++ b/drivers/isdn/hisax/hfc4s8s_l1.h @@ -16,7 +16,7 @@ /* * include Genero generated HFC-4S/8S header file hfc48scu.h -* for comlete register description. This will define _HFC48SCU_H_ +* for complete register description. This will define _HFC48SCU_H_ * to prevent redefinitions */ diff --git a/drivers/media/dvb/ttpci/budget-patch.c b/drivers/media/dvb/ttpci/budget-patch.c index fc1267b8c892..9a155396d6ac 100644 --- a/drivers/media/dvb/ttpci/budget-patch.c +++ b/drivers/media/dvb/ttpci/budget-patch.c @@ -500,14 +500,14 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte /* New design (By Emard) ** this rps1 code will copy internal HS event to GPIO3 pin. -** GPIO3 is in budget-patch hardware connectd to port B VSYNC +** GPIO3 is in budget-patch hardware connected to port B VSYNC ** HS is an internal event of 7146, accessible with RPS ** and temporarily raised high every n lines ** (n in defined in the RPS_THRESH1 counter threshold) ** I think HS is raised high on the beginning of the n-th line ** and remains high until this n-th line that triggered -** it is completely received. When the receiption of n-th line +** it is completely received. When the reception of n-th line ** ends, HS is lowered. ** To transmit data over DMA, 7146 needs changing state at @@ -541,7 +541,7 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte ** hardware debug note: a working budget card (including budget patch) ** with vpeirq() interrupt setup in mode "0x90" (every 64K) will ** generate 3 interrupts per 25-Hz DMA frame of 2*188*512 bytes -** and that means 3*25=75 Hz of interrupt freqency, as seen by +** and that means 3*25=75 Hz of interrupt frequency, as seen by ** watch cat /proc/interrupts ** ** If this frequency is 3x lower (and data received in the DMA @@ -550,7 +550,7 @@ static int budget_patch_attach (struct saa7146_dev* dev, struct saa7146_pci_exte ** this means VSYNC line is not connected in the hardware. ** (check soldering pcb and pins) ** The same behaviour of missing VSYNC can be duplicated on budget -** cards, by seting DD1_INIT trigger mode 7 in 3rd nibble. +** cards, by setting DD1_INIT trigger mode 7 in 3rd nibble. */ // Setup RPS1 "program" (p35) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 19ab3441269c..3a8df479cbda 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -1215,7 +1215,7 @@ static void e100_setup_ucode(struct nic *nic, struct cb *cb, struct sk_buff *skb * the literal in the instruction before the code is loaded, the * driver can change the algorithm. * -* INTDELAY - This loads the dead-man timer with its inital value. +* INTDELAY - This loads the dead-man timer with its initial value. * When this timer expires the interrupt is asserted, and the * timer is reset each time a new packet is received. (see * BUNDLEMAX below to set the limit on number of chained packets) diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index 65077f39da69..796c4f7d4260 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -3868,7 +3868,7 @@ e1000_phy_hw_reset(struct e1000_hw *hw) * * hw - Struct containing variables accessed by shared code * -* Sets bit 15 of the MII Control regiser +* Sets bit 15 of the MII Control register ******************************************************************************/ int32_t e1000_phy_reset(struct e1000_hw *hw) diff --git a/drivers/net/sk98lin/h/skdrv2nd.h b/drivers/net/sk98lin/h/skdrv2nd.h index 778d9e618ebd..3fa67171e832 100644 --- a/drivers/net/sk98lin/h/skdrv2nd.h +++ b/drivers/net/sk98lin/h/skdrv2nd.h @@ -160,7 +160,7 @@ struct s_IOCTL { /* ** Interim definition of SK_DRV_TIMER placed in this file until -** common modules have boon finallized +** common modules have been finalized */ #define SK_DRV_TIMER 11 #define SK_DRV_MODERATION_TIMER 1 diff --git a/drivers/net/sk98lin/skdim.c b/drivers/net/sk98lin/skdim.c index 07c1b4c8699d..37ce03fb8de3 100644 --- a/drivers/net/sk98lin/skdim.c +++ b/drivers/net/sk98lin/skdim.c @@ -252,7 +252,7 @@ SkDimEnableModerationIfNeeded(SK_AC *pAC) { /******************************************************************************* ** Function : SkDimDisplayModerationSettings -** Description : Displays the current settings regaring interrupt moderation +** Description : Displays the current settings regarding interrupt moderation ** Programmer : Ralph Roesler ** Last Modified: 22-mar-03 ** Returns : void (!) @@ -510,7 +510,7 @@ EnableIntMod(SK_AC *pAC) { /******************************************************************************* ** Function : DisableIntMod() -** Description : Disbles the interrupt moderation independent of what inter- +** Description : Disables the interrupt moderation independent of what inter- ** rupts are running or not ** Programmer : Ralph Roesler ** Last Modified: 23-mar-03 diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c index 1f742814a01c..72120d5c2f7b 100644 --- a/drivers/net/wireless/ipw2200.c +++ b/drivers/net/wireless/ipw2200.c @@ -6920,8 +6920,8 @@ static int ipw_qos_association(struct ipw_priv *priv, } /* -* handling the beaconing responces. if we get different QoS setting -* of the network from the the associated setting adjust the QoS +* handling the beaconing responses. if we get different QoS setting +* off the network from the associated setting, adjust the QoS * setting */ static int ipw_qos_association_resp(struct ipw_priv *priv, diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c index 68cb3a080050..fe3f5f5365c5 100644 --- a/drivers/parisc/ccio-dma.c +++ b/drivers/parisc/ccio-dma.c @@ -486,7 +486,7 @@ typedef unsigned long space_t; ** This bit tells U2 to do R/M/W for partial cachelines. "Streaming" ** data can avoid this if the mapping covers full cache lines. ** o STOP_MOST is needed for atomicity across cachelines. -** Apperently only "some EISA devices" need this. +** Apparently only "some EISA devices" need this. ** Using CONFIG_ISA is hack. Only the IOA with EISA under it needs ** to use this hint iff the EISA devices needs this feature. ** According to the U2 ERS, STOP_MOST enabled pages hurt performance. diff --git a/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c index c2949b4367e5..12bab64a62a1 100644 --- a/drivers/parisc/iosapic.c +++ b/drivers/parisc/iosapic.c @@ -50,12 +50,12 @@ ** ** PA Firmware ** ----------- -** PA-RISC platforms have two fundementally different types of firmware. +** PA-RISC platforms have two fundamentally different types of firmware. ** For PCI devices, "Legacy" PDC initializes the "INTERRUPT_LINE" register ** and BARs similar to a traditional PC BIOS. ** The newer "PAT" firmware supports PDC calls which return tables. -** PAT firmware only initializes PCI Console and Boot interface. -** With these tables, the OS can progam all other PCI devices. +** PAT firmware only initializes the PCI Console and Boot interface. +** With these tables, the OS can program all other PCI devices. ** ** One such PAT PDC call returns the "Interrupt Routing Table" (IRT). ** The IRT maps each PCI slot's INTA-D "output" line to an I/O SAPIC diff --git a/drivers/pci/hotplug/ibmphp_hpc.c b/drivers/pci/hotplug/ibmphp_hpc.c index c3ac98a0a6a6..f55ac3885cb3 100644 --- a/drivers/pci/hotplug/ibmphp_hpc.c +++ b/drivers/pci/hotplug/ibmphp_hpc.c @@ -531,7 +531,7 @@ static u8 hpc_readcmdtoindex (u8 cmd, u8 index) * * Action: issue a READ command to HPC * -* Input: pslot - can not be NULL for READ_ALLSTAT +* Input: pslot - cannot be NULL for READ_ALLSTAT * pstatus - can be NULL for READ_ALLSTAT * * Return 0 or error codes diff --git a/drivers/s390/net/claw.h b/drivers/s390/net/claw.h index 969be465309c..1ee9a6f06541 100644 --- a/drivers/s390/net/claw.h +++ b/drivers/s390/net/claw.h @@ -29,7 +29,7 @@ #define CLAW_COMPLETE 0xff /* flag to indicate i/o completed */ /*-----------------------------------------------------* -* CLAW control comand code * +* CLAW control command code * *------------------------------------------------------*/ #define SYSTEM_VALIDATE_REQUEST 0x01 /* System Validate request */ diff --git a/drivers/scsi/aic94xx/aic94xx_reg_def.h b/drivers/scsi/aic94xx/aic94xx_reg_def.h index b79f45f3ad47..a11f4e6d8bd9 100644 --- a/drivers/scsi/aic94xx/aic94xx_reg_def.h +++ b/drivers/scsi/aic94xx/aic94xx_reg_def.h @@ -2000,7 +2000,7 @@ * The host accesses this scratch in a different manner from the * central sequencer. The sequencer has to use CSEQ registers CSCRPAGE * and CMnSCRPAGE to access the scratch memory. A flat mapping of the - * scratch memory is avaliable for software convenience and to prevent + * scratch memory is available for software convenience and to prevent * corruption while the sequencer is running. This memory is mapped * onto addresses 800h - BFFh, total of 400h bytes. * diff --git a/drivers/scsi/aic94xx/aic94xx_sds.c b/drivers/scsi/aic94xx/aic94xx_sds.c index de7c04d4254d..e5a0ec37e954 100644 --- a/drivers/scsi/aic94xx/aic94xx_sds.c +++ b/drivers/scsi/aic94xx/aic94xx_sds.c @@ -64,7 +64,7 @@ struct asd_ocm_dir { #define OCM_INIT_DIR_ENTRIES 5 /*************************************************************************** -* OCM dircetory default +* OCM directory default ***************************************************************************/ static struct asd_ocm_dir OCMDirInit = { @@ -73,7 +73,7 @@ static struct asd_ocm_dir OCMDirInit = }; /*************************************************************************** -* OCM dircetory Entries default +* OCM directory Entries default ***************************************************************************/ static struct asd_ocm_dir_ent OCMDirEntriesInit[OCM_INIT_DIR_ENTRIES] = { diff --git a/drivers/scsi/ncr53c8xx.c b/drivers/scsi/ncr53c8xx.c index 6cc2bc2f62be..adb8eb4f5fd1 100644 --- a/drivers/scsi/ncr53c8xx.c +++ b/drivers/scsi/ncr53c8xx.c @@ -185,7 +185,7 @@ static inline struct list_head *ncr_list_pop(struct list_head *head) ** power of 2 cache line size. ** Enhanced in linux-2.3.44 to provide a memory pool ** per pcidev to support dynamic dma mapping. (I would -** have preferred a real bus astraction, btw). +** have preferred a real bus abstraction, btw). ** **========================================================== */ @@ -1438,7 +1438,7 @@ struct head { ** The first four bytes (scr_st[4]) are used inside the script by ** "COPY" commands. ** Because source and destination must have the same alignment -** in a DWORD, the fields HAVE to be at the choosen offsets. +** in a DWORD, the fields HAVE to be at the chosen offsets. ** xerr_st 0 (0x34) scratcha ** sync_st 1 (0x05) sxfer ** wide_st 3 (0x03) scntl3 @@ -1498,7 +1498,7 @@ struct head { ** the DSA (data structure address) register points ** to this substructure of the ccb. ** This substructure contains the header with -** the script-processor-changable data and +** the script-processor-changeable data and ** data blocks for the indirect move commands. ** **---------------------------------------------------------- @@ -5107,7 +5107,7 @@ void ncr_complete (struct ncb *np, struct ccb *cp) /* ** This CCB has been skipped by the NCR. -** Queue it in the correponding unit queue. +** Queue it in the corresponding unit queue. */ static void ncr_ccb_skipped(struct ncb *np, struct ccb *cp) { @@ -5896,8 +5896,8 @@ static void ncr_log_hard_error(struct ncb *np, u16 sist, u_char dstat) ** ** In normal cases, interrupt conditions occur one at a ** time. The ncr is able to stack in some extra registers -** other interrupts that will occurs after the first one. -** But severall interrupts may occur at the same time. +** other interrupts that will occur after the first one. +** But, several interrupts may occur at the same time. ** ** We probably should only try to deal with the normal ** case, but it seems that multiple interrupts occur in @@ -6796,7 +6796,7 @@ void ncr_int_sir (struct ncb *np) ** The host status field is set to HS_NEGOTIATE to mark this ** situation. ** -** If the target doesn't answer this message immidiately +** If the target doesn't answer this message immediately ** (as required by the standard), the SIR_NEGO_FAIL interrupt ** will be raised eventually. ** The handler removes the HS_NEGOTIATE status, and sets the diff --git a/drivers/scsi/ncr53c8xx.h b/drivers/scsi/ncr53c8xx.h index cb8b7701431e..b39357d9af8d 100644 --- a/drivers/scsi/ncr53c8xx.h +++ b/drivers/scsi/ncr53c8xx.h @@ -218,7 +218,7 @@ ** Same as option 1, but also deal with ** misconfigured interrupts. ** -** - Edge triggerred instead of level sensitive. +** - Edge triggered instead of level sensitive. ** - No interrupt line connected. ** - IRQ number misconfigured. ** @@ -549,7 +549,7 @@ struct ncr_driver_setup { /* ** Initial setup. -** Can be overriden at startup by a command line. +** Can be overridden at startup by a command line. */ #define SCSI_NCR_DRIVER_SETUP \ { \ @@ -1093,7 +1093,7 @@ struct scr_tblsel { **----------------------------------------------------------- ** On 810A, 860, 825A, 875, 895 and 896 chips the content ** of SFBR register can be used as data (SCR_SFBR_DATA). -** The 896 has additionnal IO registers starting at +** The 896 has additional IO registers starting at ** offset 0x80. Bit 7 of register offset is stored in ** bit 7 of the SCRIPTS instruction first DWORD. **----------------------------------------------------------- diff --git a/drivers/usb/host/u132-hcd.c b/drivers/usb/host/u132-hcd.c index 32c635ecbf31..a00d1595656c 100644 --- a/drivers/usb/host/u132-hcd.c +++ b/drivers/usb/host/u132-hcd.c @@ -211,7 +211,7 @@ int usb_ftdi_elan_read_pcimem(struct platform_device *pdev, u8 addressofs, int usb_ftdi_elan_write_pcimem(struct platform_device *pdev, u8 addressofs, u8 width, u32 data); /* -* these can not be inlines because we need the structure offset!! +* these cannot be inlines because we need the structure offset!! * Does anyone have a better way????? */ #define u132_read_pcimem(u132, member, data) \ @@ -3045,7 +3045,7 @@ static struct hc_driver u132_hc_driver = { * This function may be called by the USB core whilst the "usb_all_devices_rwsem" * is held for writing, thus this module must not call usb_remove_hcd() * synchronously - but instead should immediately stop activity to the -* device and ansynchronously call usb_remove_hcd() +* device and asynchronously call usb_remove_hcd() */ static int __devexit u132_remove(struct platform_device *pdev) { @@ -3241,7 +3241,7 @@ static int u132_resume(struct platform_device *pdev) #define u132_resume NULL #endif /* -* this driver is loaded explicitely by ftdi_u132 +* this driver is loaded explicitly by ftdi_u132 * * the platform_driver struct is static because it is per type of module */ diff --git a/drivers/usb/misc/usb_u132.h b/drivers/usb/misc/usb_u132.h index 551ba8906d62..5b5a3e621daa 100644 --- a/drivers/usb/misc/usb_u132.h +++ b/drivers/usb/misc/usb_u132.h @@ -52,7 +52,7 @@ * the kernel to load the "u132-hcd" module. * * The "ftdi-u132" module provides the interface to the inserted -* PC card and the "u132-hcd" module uses the API to send and recieve +* PC card and the "u132-hcd" module uses the API to send and receive * data. The API features call-backs, so that part of the "u132-hcd" * module code will run in the context of one of the kernel threads * of the "ftdi-u132" module. diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index bdb58100fc1d..5e3ac281a2f8 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -157,7 +157,7 @@ * to TASK_RUNNING will be lost and write_chan's subsequent call to * schedule() will never return (unless it catches a signal). * This race condition occurs because write_bulk_callback() (and thus -* the wakeup) are called asynchonously from an interrupt, rather than +* the wakeup) are called asynchronously from an interrupt, rather than * from the scheduler. We can avoid the race by calling the wakeup * from the scheduler queue and that's our fix: Now, at the end of * write_bulk_callback() we queue up a wakeup call on the scheduler diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 85ce23268302..ac93174c9639 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1464,7 +1464,7 @@ static int flush_journal_list(struct super_block *s, } /* if someone has this block in a newer transaction, just make - ** sure they are commited, and don't try writing it to disk + ** sure they are committed, and don't try writing it to disk */ if (pjl) { if (atomic_read(&pjl->j_commit_left)) @@ -3384,7 +3384,7 @@ static int remove_from_transaction(struct super_block *p_s_sb, /* ** for any cnode in a journal list, it can only be dirtied of all the -** transactions that include it are commited to disk. +** transactions that include it are committed to disk. ** this checks through each transaction, and returns 1 if you are allowed to dirty, ** and 0 if you aren't ** @@ -3426,7 +3426,7 @@ static int can_dirty(struct reiserfs_journal_cnode *cn) } /* syncs the commit blocks, but does not force the real buffers to disk -** will wait until the current transaction is done/commited before returning +** will wait until the current transaction is done/committed before returning */ int journal_end_sync(struct reiserfs_transaction_handle *th, struct super_block *p_s_sb, unsigned long nblocks) diff --git a/include/asm-m68knommu/mcfmbus.h b/include/asm-m68knommu/mcfmbus.h index 13df9d41bd1a..319899c47a2c 100644 --- a/include/asm-m68knommu/mcfmbus.h +++ b/include/asm-m68knommu/mcfmbus.h @@ -37,7 +37,7 @@ #define MCFMBUS_MFDR_MBC(a) ((a)&0x3F) /*M-Bus Clock*/ /* -* Define bit flags in Controll Register +* Define bit flags in Control Register */ #define MCFMBUS_MBCR_MEN (0x80) /* M-Bus Enable */ diff --git a/include/asm-parisc/dma.h b/include/asm-parisc/dma.h index da2cf373e31c..31ad0f05af3d 100644 --- a/include/asm-parisc/dma.h +++ b/include/asm-parisc/dma.h @@ -17,10 +17,10 @@ /* ** DMA_CHUNK_SIZE is used by the SCSI mid-layer to break up -** (or rather not merge) DMA's into managable chunks. +** (or rather not merge) DMAs into manageable chunks. ** On parisc, this is more of the software/tuning constraint -** rather than the HW. I/O MMU allocation alogorithms can be -** faster with smaller size is (to some degree). +** rather than the HW. I/O MMU allocation algorithms can be +** faster with smaller sizes (to some degree). */ #define DMA_CHUNK_SIZE (BITS_PER_LONG*PAGE_SIZE) diff --git a/include/asm-parisc/pci.h b/include/asm-parisc/pci.h index 7b8ad118d2fe..7b3be9ac0dda 100644 --- a/include/asm-parisc/pci.h +++ b/include/asm-parisc/pci.h @@ -149,7 +149,7 @@ extern int parisc_bus_is_phys; /* in arch/parisc/kernel/setup.c */ /* ** Most PCI devices (eg Tulip, NCR720) also export the same registers ** to both MMIO and I/O port space. Due to poor performance of I/O Port -** access under HP PCI bus adapters, strongly reccomend use of MMIO +** access under HP PCI bus adapters, strongly recommend the use of MMIO ** address space. ** ** While I'm at it more PA programming notes: diff --git a/include/asm-parisc/ropes.h b/include/asm-parisc/ropes.h index 5542dd00472b..007a880615eb 100644 --- a/include/asm-parisc/ropes.h +++ b/include/asm-parisc/ropes.h @@ -14,7 +14,7 @@ #endif /* -** The number of pdir entries to "free" before issueing +** The number of pdir entries to "free" before issuing ** a read to PCOM register to flush out PCOM writes. ** Interacts with allocation granularity (ie 4 or 8 entries ** allocated and free'd/purged at a time might make this diff --git a/include/linux/ixjuser.h b/include/linux/ixjuser.h index fd1756d3a47e..88b45895746d 100644 --- a/include/linux/ixjuser.h +++ b/include/linux/ixjuser.h @@ -315,7 +315,7 @@ typedef struct { * structures. If the freq0 variable is non-zero, the tone table contents * for the tone_index are updated to the frequencies and gains defined. It * should be noted that DTMF tones cannot be reassigned, so if DTMF tone -* table indexs are used in a cadence the frequency and gain variables will +* table indexes are used in a cadence the frequency and gain variables will * be ignored. * * If the array elements contain frequency parameters the driver will diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 73e0becec086..62a7169aed8b 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -429,7 +429,7 @@ enum reiserfs_mount_options { /* -o hash={tea, rupasov, r5, detect} is meant for properly mounting ** reiserfs disks from 3.5.19 or earlier. 99% of the time, this option ** is not required. If the normal autodection code can't determine which -** hash to use (because both hases had the same value for a file) +** hash to use (because both hashes had the same value for a file) ** use this option to force a specific hash. It won't allow you to override ** the existing hash on the FS, so if you have a tea hash disk, and mount ** with -o hash=rupasov, the mount will fail. diff --git a/net/wanrouter/af_wanpipe.c b/net/wanrouter/af_wanpipe.c index 6f39faa15832..c2059733e15a 100644 --- a/net/wanrouter/af_wanpipe.c +++ b/net/wanrouter/af_wanpipe.c @@ -13,7 +13,7 @@ * Due Credit: * Wanpipe socket layer is based on Packet and * the X25 socket layers. The above sockets were -* used for the specific use of Sangoma Technoloiges +* used for the specific use of Sangoma Technologies * API programs. * Packet socket Authors: Ross Biro, Fred N. van Kempen and * Alan Cox. @@ -23,7 +23,7 @@ * Apr 25, 2000 Nenad Corbic o Added the ability to send zero length packets. * Mar 13, 2000 Nenad Corbic o Added a tx buffer check via ioctl call. * Mar 06, 2000 Nenad Corbic o Fixed the corrupt sock lcn problem. -* Server and client applicaton can run +* Server and client application can run * simultaneously without conflicts. * Feb 29, 2000 Nenad Corbic o Added support for PVC protocols, such as * CHDLC, Frame Relay and HDLC API. diff --git a/net/wanrouter/wanmain.c b/net/wanrouter/wanmain.c index 9479659277ae..316211d9f17d 100644 --- a/net/wanrouter/wanmain.c +++ b/net/wanrouter/wanmain.c @@ -3,7 +3,7 @@ * * This module is completely hardware-independent and provides * the following common services for the WAN Link Drivers: -* o WAN device managenment (registering, unregistering) +* o WAN device management (registering, unregistering) * o Network interface management * o Physical connection management (dial-up, incoming calls) * o Logical connection management (switched virtual circuits) diff --git a/sound/oss/cs46xx.c b/sound/oss/cs46xx.c index 6e3c41f530e6..b1c5d8286e40 100644 --- a/sound/oss/cs46xx.c +++ b/sound/oss/cs46xx.c @@ -779,7 +779,7 @@ static unsigned int cs_set_adc_rate(struct cs_state *state, unsigned int rate) rate = 48000 / 9; /* - * We can not capture at at rate greater than the Input Rate (48000). + * We cannot capture at at rate greater than the Input Rate (48000). * Return an error if an attempt is made to stray outside that limit. */ if (rate > 48000) @@ -4754,8 +4754,8 @@ static int cs_hardware_init(struct cs_card *card) mdelay(5 * cs_laptop_wait); /* Shouldnt be needed ?? */ /* -* If we are resuming under 2.2.x then we can not schedule a timeout. -* so, just spin the CPU. +* If we are resuming under 2.2.x then we cannot schedule a timeout, +* so just spin the CPU. */ if (card->pm.flags & CS46XX_PM_IDLE) { /* -- cgit v1.2.3 From 03a67a46af8647b2c7825107045ecae641e103d3 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Thu, 30 Nov 2006 05:32:19 +0100 Subject: Fix typos in doc and comments Changes persistant -> persistent. www.dictionary.com does not know persistant (with an A), but should it be one of those things you can spell in more than one correct way, let me know. Signed-off-by: Jan Engelhardt Signed-off-by: Adrian Bunk --- Documentation/Changes | 2 +- Documentation/power/states.txt | 2 +- arch/um/drivers/chan_user.c | 2 +- drivers/message/fusion/mptbase.c | 2 +- drivers/mtd/maps/cfi_flagadm.c | 2 +- drivers/pci/Kconfig | 2 +- fs/Kconfig | 2 +- fs/jfs/jfs_filsys.h | 2 +- include/linux/textsearch.h | 4 ++-- lib/textsearch.c | 2 +- 10 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/Documentation/Changes b/Documentation/Changes index abee7f58c1ed..73a8617f1861 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -201,7 +201,7 @@ udev ---- udev is a userspace application for populating /dev dynamically with only entries for devices actually present. udev replaces the basic -functionality of devfs, while allowing persistant device naming for +functionality of devfs, while allowing persistent device naming for devices. FUSE diff --git a/Documentation/power/states.txt b/Documentation/power/states.txt index 3e5e5d3ff419..0931a330d362 100644 --- a/Documentation/power/states.txt +++ b/Documentation/power/states.txt @@ -62,7 +62,7 @@ setup via another operating system for it to use. Despite the inconvenience, this method requires minimal work by the kernel, since the firmware will also handle restoring memory contents on resume. -If the kernel is responsible for persistantly saving state, a mechanism +If the kernel is responsible for persistently saving state, a mechanism called 'swsusp' (Swap Suspend) is used to write memory contents to free swap space. swsusp has some restrictive requirements, but should work in most cases. Some, albeit outdated, documentation can be found diff --git a/arch/um/drivers/chan_user.c b/arch/um/drivers/chan_user.c index 2f880cb167a5..0cad3546cb89 100644 --- a/arch/um/drivers/chan_user.c +++ b/arch/um/drivers/chan_user.c @@ -120,7 +120,7 @@ static int winch_thread(void *arg) /* These are synchronization calls between various UML threads on the * host - since they are not different kernel threads, we cannot use * kernel semaphores. We don't use SysV semaphores because they are - * persistant. */ + * persistent. */ count = os_read_file(pipe_fd, &c, sizeof(c)); if(count != sizeof(c)) printk("winch_thread : failed to read synchronization byte, " diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index e5c72719debc..051b7c5b8f03 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -6185,7 +6185,7 @@ mpt_spi_log_info(MPT_ADAPTER *ioc, u32 log_info) "Abort", /* 12h */ "IO Not Yet Executed", /* 13h */ "IO Executed", /* 14h */ - "Persistant Reservation Out Not Affiliation Owner", /* 15h */ + "Persistent Reservation Out Not Affiliation Owner", /* 15h */ "Open Transmit DMA Abort", /* 16h */ "IO Device Missing Delay Retry", /* 17h */ NULL, /* 18h */ diff --git a/drivers/mtd/maps/cfi_flagadm.c b/drivers/mtd/maps/cfi_flagadm.c index 92b5d883d7b0..65e5ee552010 100644 --- a/drivers/mtd/maps/cfi_flagadm.c +++ b/drivers/mtd/maps/cfi_flagadm.c @@ -80,7 +80,7 @@ struct mtd_partition flagadm_parts[] = { .size = FLASH_PARTITION2_SIZE }, { - .name = "Persistant storage", + .name = "Persistent storage", .offset = FLASH_PARTITION3_ADDR, .size = FLASH_PARTITION3_SIZE } diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index d0ba112355cc..3cfb0a3575e6 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -34,7 +34,7 @@ config PCI_MULTITHREAD_PROBE It is very unwise to use this option if you are not using a boot process that can handle devices being created in any - order. A program that can create persistant block and network + order. A program that can create persistent block and network device names (like udev) is a good idea if you wish to use this option. diff --git a/fs/Kconfig b/fs/Kconfig index 8bec76bbc0c3..b3b5aa0edff9 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -972,7 +972,7 @@ config SYSFS Some system agents rely on the information in sysfs to operate. /sbin/hotplug uses device and object attributes in sysfs to assist in - delegating policy decisions, like persistantly naming devices. + delegating policy decisions, like persistently naming devices. sysfs is currently used by the block subsystem to mount the root partition. If sysfs is disabled you must specify the boot device on diff --git a/fs/jfs/jfs_filsys.h b/fs/jfs/jfs_filsys.h index 9901928668cf..eb550b339bb8 100644 --- a/fs/jfs/jfs_filsys.h +++ b/fs/jfs/jfs_filsys.h @@ -81,7 +81,7 @@ #define JFS_SWAP_BYTES 0x00100000 /* running on big endian computer */ /* Directory index */ -#define JFS_DIR_INDEX 0x00200000 /* Persistant index for */ +#define JFS_DIR_INDEX 0x00200000 /* Persistent index for */ /* directory entries */ diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h index 7dac8f04d28e..004808a6df1d 100644 --- a/include/linux/textsearch.h +++ b/include/linux/textsearch.h @@ -20,7 +20,7 @@ struct ts_config; /** * struct ts_state - search state * @offset: offset for next match - * @cb: control buffer, for persistant variables of get_next_block() + * @cb: control buffer, for persistent variables of get_next_block() */ struct ts_state { @@ -71,7 +71,7 @@ struct ts_config * Called repeatedly until 0 is returned. Must assign the * head of the next block of data to &*dst and return the length * of the block or 0 if at the end. consumed == 0 indicates - * a new search. May store/read persistant values in state->cb. + * a new search. May store/read persistent values in state->cb. */ unsigned int (*get_next_block)(unsigned int consumed, const u8 **dst, diff --git a/lib/textsearch.c b/lib/textsearch.c index 2cb4a437942e..98bcadc01185 100644 --- a/lib/textsearch.c +++ b/lib/textsearch.c @@ -40,7 +40,7 @@ * configuration according to the specified parameters. * (3) User starts the search(es) by calling _find() or _next() to * fetch subsequent occurrences. A state variable is provided - * to the algorihtm to store persistant variables. + * to the algorihtm to store persistent variables. * (4) Core eventually resets the search offset and forwards the find() * request to the algorithm. * (5) Algorithm calls get_next_block() provided by the user continously -- cgit v1.2.3 From beea494d5e09f9e2172894ec84324b94244826a9 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 7 Nov 2006 21:03:20 +0000 Subject: [ARM] Remove EEPROM slave emulation from i2c-pxa driver. The i2c-pxa driver should not contain EEPROM slave-mode emulation; this is something the platform should provide where required. Remove it. Signed-off-by: Russell King --- drivers/i2c/busses/i2c-pxa.c | 131 +------------------------------------------ include/linux/i2c-pxa.h | 31 ---------- 2 files changed, 1 insertion(+), 161 deletions(-) (limited to 'include/linux') diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index c95a6c154165..c3b1567c852a 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -357,133 +357,6 @@ static void i2c_pxa_reset(struct pxa_i2c *i2c) #ifdef CONFIG_I2C_PXA_SLAVE -/* - * I2C EEPROM emulation. - */ -static struct i2c_eeprom_emu eeprom = { - .size = I2C_EEPROM_EMU_SIZE, - .watch = LIST_HEAD_INIT(eeprom.watch), -}; - -struct i2c_eeprom_emu *i2c_pxa_get_eeprom(void) -{ - return &eeprom; -} - -int i2c_eeprom_emu_addwatcher(struct i2c_eeprom_emu *emu, void *data, - unsigned int addr, unsigned int size, - struct i2c_eeprom_emu_watcher *watcher) -{ - struct i2c_eeprom_emu_watch *watch; - unsigned long flags; - - if (addr + size > emu->size) - return -EINVAL; - - watch = kmalloc(sizeof(struct i2c_eeprom_emu_watch), GFP_KERNEL); - if (watch) { - watch->start = addr; - watch->end = addr + size - 1; - watch->ops = watcher; - watch->data = data; - - local_irq_save(flags); - list_add(&watch->node, &emu->watch); - local_irq_restore(flags); - } - - return watch ? 0 : -ENOMEM; -} - -void i2c_eeprom_emu_delwatcher(struct i2c_eeprom_emu *emu, void *data, - struct i2c_eeprom_emu_watcher *watcher) -{ - struct i2c_eeprom_emu_watch *watch, *n; - unsigned long flags; - - list_for_each_entry_safe(watch, n, &emu->watch, node) { - if (watch->ops == watcher && watch->data == data) { - local_irq_save(flags); - list_del(&watch->node); - local_irq_restore(flags); - kfree(watch); - } - } -} - -static void i2c_eeprom_emu_event(void *ptr, i2c_slave_event_t event) -{ - struct i2c_eeprom_emu *emu = ptr; - - eedbg(3, "i2c_eeprom_emu_event: %d\n", event); - - switch (event) { - case I2C_SLAVE_EVENT_START_WRITE: - emu->seen_start = 1; - eedbg(2, "i2c_eeprom: write initiated\n"); - break; - - case I2C_SLAVE_EVENT_START_READ: - emu->seen_start = 0; - eedbg(2, "i2c_eeprom: read initiated\n"); - break; - - case I2C_SLAVE_EVENT_STOP: - emu->seen_start = 0; - eedbg(2, "i2c_eeprom: received stop\n"); - break; - - default: - eedbg(0, "i2c_eeprom: unhandled event\n"); - break; - } -} - -static int i2c_eeprom_emu_read(void *ptr) -{ - struct i2c_eeprom_emu *emu = ptr; - int ret; - - ret = emu->bytes[emu->ptr]; - emu->ptr = (emu->ptr + 1) % emu->size; - - return ret; -} - -static void i2c_eeprom_emu_write(void *ptr, unsigned int val) -{ - struct i2c_eeprom_emu *emu = ptr; - struct i2c_eeprom_emu_watch *watch; - - if (emu->seen_start != 0) { - eedbg(2, "i2c_eeprom_emu_write: setting ptr %02x\n", val); - emu->ptr = val; - emu->seen_start = 0; - return; - } - - emu->bytes[emu->ptr] = val; - - eedbg(1, "i2c_eeprom_emu_write: ptr=0x%02x, val=0x%02x\n", - emu->ptr, val); - - list_for_each_entry(watch, &emu->watch, node) { - if (!watch->ops || !watch->ops->write) - continue; - if (watch->start <= emu->ptr && watch->end >= emu->ptr) - watch->ops->write(watch->data, emu->ptr, val); - } - - emu->ptr = (emu->ptr + 1) % emu->size; -} - -struct i2c_slave_client eeprom_client = { - .data = &eeprom, - .event = i2c_eeprom_emu_event, - .read = i2c_eeprom_emu_read, - .write = i2c_eeprom_emu_write -}; - /* * PXA I2C Slave mode */ @@ -963,11 +836,9 @@ static int i2c_pxa_probe(struct platform_device *dev) i2c->slave_addr = I2C_PXA_SLAVE_ADDR; #ifdef CONFIG_I2C_PXA_SLAVE - i2c->slave = &eeprom_client; if (plat) { i2c->slave_addr = plat->slave_addr; - if (plat->slave) - i2c->slave = plat->slave; + i2c->slave = plat->slave; } #endif diff --git a/include/linux/i2c-pxa.h b/include/linux/i2c-pxa.h index 5f3eaf802223..41dcdfe7f625 100644 --- a/include/linux/i2c-pxa.h +++ b/include/linux/i2c-pxa.h @@ -1,29 +1,6 @@ #ifndef _LINUX_I2C_ALGO_PXA_H #define _LINUX_I2C_ALGO_PXA_H -struct i2c_eeprom_emu_watcher { - void (*write)(void *, unsigned int addr, unsigned char newval); -}; - -struct i2c_eeprom_emu_watch { - struct list_head node; - unsigned int start; - unsigned int end; - struct i2c_eeprom_emu_watcher *ops; - void *data; -}; - -#define I2C_EEPROM_EMU_SIZE (256) - -struct i2c_eeprom_emu { - unsigned int size; - unsigned int ptr; - unsigned int seen_start; - struct list_head watch; - - unsigned char bytes[I2C_EEPROM_EMU_SIZE]; -}; - typedef enum i2c_slave_event_e { I2C_SLAVE_EVENT_START_READ, I2C_SLAVE_EVENT_START_WRITE, @@ -37,12 +14,4 @@ struct i2c_slave_client { void (*write)(void *ptr, unsigned int val); }; -extern int i2c_eeprom_emu_addwatcher(struct i2c_eeprom_emu *, void *data, - unsigned int addr, unsigned int size, - struct i2c_eeprom_emu_watcher *); - -extern void i2c_eeprom_emu_delwatcher(struct i2c_eeprom_emu *, void *data, struct i2c_eeprom_emu_watcher *watcher); - -extern struct i2c_eeprom_emu *i2c_pxa_get_eeprom(void); - #endif /* _LINUX_I2C_ALGO_PXA_H */ -- cgit v1.2.3 From 3ca68df6ee61e1a2034f3307b9edb9b3d87e5ca1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 20:11:25 -0400 Subject: [GFS2] split gfs2_dinode into on-disk and host variants The latter is used as part of gfs2-private part of struct inode. It actually stores a lot of fields differently; for now the declaration is just cloned, inode field is swtiched and changes propagated. Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 4 ++-- fs/gfs2/ondisk.c | 6 +++--- include/linux/gfs2_ondisk.h | 48 ++++++++++++++++++++++++++++++++++++++++++--- 3 files changed, 50 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index d470e5286ecd..191a3df93d67 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -48,7 +48,7 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) { struct inode *inode = &ip->i_inode; - struct gfs2_dinode *di = &ip->i_di; + struct gfs2_dinode_host *di = &ip->i_di; inode->i_ino = ip->i_num.no_addr; @@ -98,7 +98,7 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) void gfs2_inode_attr_out(struct gfs2_inode *ip) { struct inode *inode = &ip->i_inode; - struct gfs2_dinode *di = &ip->i_di; + struct gfs2_dinode_host *di = &ip->i_di; gfs2_assert_withdraw(GFS2_SB(inode), (di->di_mode & S_IFMT) == (inode->i_mode & S_IFMT)); di->di_mode = inode->i_mode; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 1025960b0e6e..52cb9a2dc058 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -153,7 +153,7 @@ void gfs2_quota_in(struct gfs2_quota *qu, const void *buf) qu->qu_value = be64_to_cpu(str->qu_value); } -void gfs2_dinode_in(struct gfs2_dinode *di, const void *buf) +void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf) { const struct gfs2_dinode *str = buf; @@ -187,7 +187,7 @@ void gfs2_dinode_in(struct gfs2_dinode *di, const void *buf) } -void gfs2_dinode_out(const struct gfs2_dinode *di, void *buf) +void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf) { struct gfs2_dinode *str = buf; @@ -221,7 +221,7 @@ void gfs2_dinode_out(const struct gfs2_dinode *di, void *buf) } -void gfs2_dinode_print(const struct gfs2_dinode *di) +void gfs2_dinode_print(const struct gfs2_dinode_host *di) { gfs2_meta_header_print(&di->di_header); gfs2_inum_print(&di->di_num); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index a7ae7c177cac..f334b4bd2915 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -270,6 +270,48 @@ struct gfs2_dinode { __u8 di_reserved[56]; }; +struct gfs2_dinode_host { + struct gfs2_meta_header di_header; + + struct gfs2_inum di_num; + + __be32 di_mode; /* mode of file */ + __be32 di_uid; /* owner's user id */ + __be32 di_gid; /* owner's group id */ + __be32 di_nlink; /* number of links to this file */ + __be64 di_size; /* number of bytes in file */ + __be64 di_blocks; /* number of blocks in file */ + __be64 di_atime; /* time last accessed */ + __be64 di_mtime; /* time last modified */ + __be64 di_ctime; /* time last changed */ + __be32 di_major; /* device major number */ + __be32 di_minor; /* device minor number */ + + /* This section varies from gfs1. Padding added to align with + * remainder of dinode + */ + __be64 di_goal_meta; /* rgrp to alloc from next */ + __be64 di_goal_data; /* data block goal */ + __be64 di_generation; /* generation number for NFS */ + + __be32 di_flags; /* GFS2_DIF_... */ + __be32 di_payload_format; /* GFS2_FORMAT_... */ + __u16 __pad1; /* Was ditype in gfs1 */ + __be16 di_height; /* height of metadata */ + __u32 __pad2; /* Unused incarnation number from gfs1 */ + + /* These only apply to directories */ + __u16 __pad3; /* Padding */ + __be16 di_depth; /* Number of bits in the table */ + __be32 di_entries; /* The number of entries in the directory */ + + struct gfs2_inum __pad4; /* Unused even in current gfs1 */ + + __be64 di_eattr; /* extended attribute block number */ + + __u8 di_reserved[56]; +}; + /* * directory structure - many of these per directory file */ @@ -422,8 +464,8 @@ extern void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota *qu, const void *buf); extern void gfs2_quota_out(const struct gfs2_quota *qu, void *buf); -extern void gfs2_dinode_in(struct gfs2_dinode *di, const void *buf); -extern void gfs2_dinode_out(const struct gfs2_dinode *di, void *buf); +extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); +extern void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); extern void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf); @@ -436,7 +478,7 @@ extern void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf); /* Printing functions */ extern void gfs2_rindex_print(const struct gfs2_rindex *ri); -extern void gfs2_dinode_print(const struct gfs2_dinode *di); +extern void gfs2_dinode_print(const struct gfs2_dinode_host *di); #endif /* __KERNEL__ */ -- cgit v1.2.3 From 5c6edb576f3800723bb65dbfaff82517089e32d0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 20:33:01 -0400 Subject: [GFS2] gfs2_dinode_host fields are host-endian Annotated scalar fields, dropped unused ones. Note that it's not at all obvious that we want to convert all of them to host-endian... Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/incore.h | 2 +- include/linux/gfs2_ondisk.h | 47 +++++++++++++++++++-------------------------- 2 files changed, 21 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 118dc693d111..1c876e0fb44a 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -229,7 +229,7 @@ struct gfs2_inode { unsigned long i_flags; /* GIF_... */ u64 i_vn; - struct gfs2_dinode i_di; /* To be replaced by ref to block */ + struct gfs2_dinode_host i_di; /* To be replaced by ref to block */ struct gfs2_glock *i_gl; /* Move into i_gh? */ struct gfs2_holder i_iopen_gh; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index f334b4bd2915..0e67a89a9699 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -275,41 +275,34 @@ struct gfs2_dinode_host { struct gfs2_inum di_num; - __be32 di_mode; /* mode of file */ - __be32 di_uid; /* owner's user id */ - __be32 di_gid; /* owner's group id */ - __be32 di_nlink; /* number of links to this file */ - __be64 di_size; /* number of bytes in file */ - __be64 di_blocks; /* number of blocks in file */ - __be64 di_atime; /* time last accessed */ - __be64 di_mtime; /* time last modified */ - __be64 di_ctime; /* time last changed */ - __be32 di_major; /* device major number */ - __be32 di_minor; /* device minor number */ + __u32 di_mode; /* mode of file */ + __u32 di_uid; /* owner's user id */ + __u32 di_gid; /* owner's group id */ + __u32 di_nlink; /* number of links to this file */ + __u64 di_size; /* number of bytes in file */ + __u64 di_blocks; /* number of blocks in file */ + __u64 di_atime; /* time last accessed */ + __u64 di_mtime; /* time last modified */ + __u64 di_ctime; /* time last changed */ + __u32 di_major; /* device major number */ + __u32 di_minor; /* device minor number */ /* This section varies from gfs1. Padding added to align with * remainder of dinode */ - __be64 di_goal_meta; /* rgrp to alloc from next */ - __be64 di_goal_data; /* data block goal */ - __be64 di_generation; /* generation number for NFS */ + __u64 di_goal_meta; /* rgrp to alloc from next */ + __u64 di_goal_data; /* data block goal */ + __u64 di_generation; /* generation number for NFS */ - __be32 di_flags; /* GFS2_DIF_... */ - __be32 di_payload_format; /* GFS2_FORMAT_... */ - __u16 __pad1; /* Was ditype in gfs1 */ - __be16 di_height; /* height of metadata */ - __u32 __pad2; /* Unused incarnation number from gfs1 */ + __u32 di_flags; /* GFS2_DIF_... */ + __u32 di_payload_format; /* GFS2_FORMAT_... */ + __u16 di_height; /* height of metadata */ /* These only apply to directories */ - __u16 __pad3; /* Padding */ - __be16 di_depth; /* Number of bits in the table */ - __be32 di_entries; /* The number of entries in the directory */ - - struct gfs2_inum __pad4; /* Unused even in current gfs1 */ + __u16 di_depth; /* Number of bits in the table */ + __u32 di_entries; /* The number of entries in the directory */ - __be64 di_eattr; /* extended attribute block number */ - - __u8 di_reserved[56]; + __u64 di_eattr; /* extended attribute block number */ }; /* -- cgit v1.2.3 From f50dfaf78c01df3cc2d8819f07d6661915567bae Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 20:45:02 -0400 Subject: [GFS2] split gfs2_sb Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/incore.h | 2 +- fs/gfs2/ondisk.c | 2 +- fs/gfs2/super.c | 2 +- fs/gfs2/super.h | 2 +- include/linux/gfs2_ondisk.h | 22 +++++++++++++++++++++- 5 files changed, 25 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 1c876e0fb44a..bd596ba74e56 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -450,7 +450,7 @@ struct gfs2_sbd { struct super_block *sd_vfs_meta; struct kobject sd_kobj; unsigned long sd_flags; /* SDF_... */ - struct gfs2_sb sd_sb; + struct gfs2_sb_host sd_sb; /* Constants computed on mount */ diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 52cb9a2dc058..5b32f1a35794 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -79,7 +79,7 @@ static void gfs2_meta_header_print(const struct gfs2_meta_header *mh) pv(mh, mh_format, "%u"); } -void gfs2_sb_in(struct gfs2_sb *sb, const void *buf) +void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) { const struct gfs2_sb *str = buf; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 6a78b1b32e25..52aa3221fc99 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -97,7 +97,7 @@ void gfs2_tune_init(struct gfs2_tune *gt) * changed. */ -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb *sb, int silent) +int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) { unsigned int x; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index 5bb443ae0f59..ac95064c1e5a 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -14,7 +14,7 @@ void gfs2_tune_init(struct gfs2_tune *gt); -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb *sb, int silent); +int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent); int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent); struct page *gfs2_read_super(struct super_block *sb, sector_t sector); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 0e67a89a9699..b7bdfef82e4e 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -128,6 +128,26 @@ struct gfs2_sb { /* In gfs1, quota and license dinodes followed */ }; +struct gfs2_sb_host { + struct gfs2_meta_header sb_header; + + __be32 sb_fs_format; + __be32 sb_multihost_format; + __u32 __pad0; /* Was superblock flags in gfs1 */ + + __be32 sb_bsize; + __be32 sb_bsize_shift; + __u32 __pad1; /* Was journal segment size in gfs1 */ + + struct gfs2_inum sb_master_dir; /* Was jindex dinode in gfs1 */ + struct gfs2_inum __pad2; /* Was rindex dinode in gfs1 */ + struct gfs2_inum sb_root_dir; + + char sb_lockproto[GFS2_LOCKNAME_LEN]; + char sb_locktable[GFS2_LOCKNAME_LEN]; + /* In gfs1, quota and license dinodes followed */ +}; + /* * resource index structure */ @@ -450,7 +470,7 @@ struct gfs2_quota_change { extern void gfs2_inum_in(struct gfs2_inum *no, const void *buf); extern void gfs2_inum_out(const struct gfs2_inum *no, void *buf); -extern void gfs2_sb_in(struct gfs2_sb *sb, const void *buf); +extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); extern void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex *ri, void *buf); extern void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf); -- cgit v1.2.3 From bc558c87bb7e50c4f728d32684a9f4f4c73ebde3 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 21:02:41 -0400 Subject: [GFS2] fields of gfs2_sb_host are host-endian ... and several could be killed, but that's another story. Annotate scalar ones, kill completely unused. Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- include/linux/gfs2_ondisk.h | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index b7bdfef82e4e..a5d36cdc46e9 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -131,16 +131,13 @@ struct gfs2_sb { struct gfs2_sb_host { struct gfs2_meta_header sb_header; - __be32 sb_fs_format; - __be32 sb_multihost_format; - __u32 __pad0; /* Was superblock flags in gfs1 */ + __u32 sb_fs_format; + __u32 sb_multihost_format; - __be32 sb_bsize; - __be32 sb_bsize_shift; - __u32 __pad1; /* Was journal segment size in gfs1 */ + __u32 sb_bsize; + __u32 sb_bsize_shift; struct gfs2_inum sb_master_dir; /* Was jindex dinode in gfs1 */ - struct gfs2_inum __pad2; /* Was rindex dinode in gfs1 */ struct gfs2_inum sb_root_dir; char sb_lockproto[GFS2_LOCKNAME_LEN]; -- cgit v1.2.3 From 68826664d12827d7a732192e2f00ba46fb899414 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 21:07:22 -0400 Subject: [GFS2] split and annotate gfs2_rgrp Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/incore.h | 2 +- fs/gfs2/ondisk.c | 4 ++-- include/linux/gfs2_ondisk.h | 13 +++++++++++-- 3 files changed, 14 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index bd596ba74e56..8ca7a7f35062 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -68,7 +68,7 @@ struct gfs2_rgrpd { struct list_head rd_recent; /* Recently used rgrps */ struct gfs2_glock *rd_gl; /* Glock for this rgrp */ struct gfs2_rindex rd_ri; - struct gfs2_rgrp rd_rg; + struct gfs2_rgrp_host rd_rg; u64 rd_rg_vn; struct gfs2_bitmap *rd_bits; unsigned int rd_bh_count; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 5b32f1a35794..3b156a15cd8d 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -120,7 +120,7 @@ void gfs2_rindex_print(const struct gfs2_rindex *ri) pv(ri, ri_bitbytes, "%u"); } -void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf) +void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf) { const struct gfs2_rgrp *str = buf; @@ -131,7 +131,7 @@ void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf) rg->rg_igeneration = be64_to_cpu(str->rg_igeneration); } -void gfs2_rgrp_out(const struct gfs2_rgrp *rg, void *buf) +void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf) { struct gfs2_rgrp *str = buf; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index a5d36cdc46e9..e4ca6e4176bb 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -193,6 +193,15 @@ struct gfs2_rgrp { __u8 rg_reserved[80]; /* Several fields from gfs1 now reserved */ }; +struct gfs2_rgrp_host { + struct gfs2_meta_header rg_header; + + __u32 rg_flags; + __u32 rg_free; + __u32 rg_dinodes; + __u64 rg_igeneration; +}; + /* * quota structure */ @@ -470,8 +479,8 @@ extern void gfs2_inum_out(const struct gfs2_inum *no, void *buf); extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); extern void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex *ri, void *buf); -extern void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf); -extern void gfs2_rgrp_out(const struct gfs2_rgrp *rg, void *buf); +extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); +extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota *qu, const void *buf); extern void gfs2_quota_out(const struct gfs2_quota *qu, void *buf); extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); -- cgit v1.2.3 From e697264709c86040271cdd7abee781d7adbb7f91 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 21:29:46 -0400 Subject: [GFS2] split and annotate gfs2_inum_range Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 4 ++-- fs/gfs2/ondisk.c | 4 ++-- include/linux/gfs2_ondisk.h | 9 +++++++-- 3 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 191a3df93d67..7eb6b440da6d 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -436,7 +436,7 @@ static int pick_formal_ino_1(struct gfs2_sbd *sdp, u64 *formal_ino) { struct gfs2_inode *ip = GFS2_I(sdp->sd_ir_inode); struct buffer_head *bh; - struct gfs2_inum_range ir; + struct gfs2_inum_range_host ir; int error; error = gfs2_trans_begin(sdp, RES_DINODE, 0); @@ -479,7 +479,7 @@ static int pick_formal_ino_2(struct gfs2_sbd *sdp, u64 *formal_ino) struct gfs2_inode *m_ip = GFS2_I(sdp->sd_inum_inode); struct gfs2_holder gh; struct buffer_head *bh; - struct gfs2_inum_range ir; + struct gfs2_inum_range_host ir; int error; error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 3b156a15cd8d..64f5f0c604aa 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -263,7 +263,7 @@ void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf) lh->lh_hash = be32_to_cpu(str->lh_hash); } -void gfs2_inum_range_in(struct gfs2_inum_range *ir, const void *buf) +void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf) { const struct gfs2_inum_range *str = buf; @@ -271,7 +271,7 @@ void gfs2_inum_range_in(struct gfs2_inum_range *ir, const void *buf) ir->ir_length = be64_to_cpu(str->ir_length); } -void gfs2_inum_range_out(const struct gfs2_inum_range *ir, void *buf) +void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf) { struct gfs2_inum_range *str = buf; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index e4ca6e4176bb..c035587d066e 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -445,6 +445,11 @@ struct gfs2_inum_range { __be64 ir_length; }; +struct gfs2_inum_range_host { + __u64 ir_start; + __u64 ir_length; +}; + /* * Statfs change * Describes an change to the pool of free and allocated @@ -488,8 +493,8 @@ extern void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); extern void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf); -extern void gfs2_inum_range_in(struct gfs2_inum_range *ir, const void *buf); -extern void gfs2_inum_range_out(const struct gfs2_inum_range *ir, void *buf); +extern void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf); +extern void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf); extern void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf); extern void gfs2_statfs_change_out(const struct gfs2_statfs_change *sc, void *buf); extern void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf); -- cgit v1.2.3 From 551676226163379c217e8ec54bd287eab9b8521e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 21:47:13 -0400 Subject: [GFS2] split and annotate gfs2_log_head Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/glops.c | 2 +- fs/gfs2/incore.h | 2 +- fs/gfs2/lops.c | 4 ++-- fs/gfs2/lops.h | 2 +- fs/gfs2/ondisk.c | 2 +- fs/gfs2/recovery.c | 22 +++++++++++----------- fs/gfs2/recovery.h | 2 +- fs/gfs2/super.c | 4 ++-- include/linux/gfs2_ondisk.h | 12 +++++++++++- 9 files changed, 31 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 41a6b6818a50..5406b193227c 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -491,7 +491,7 @@ static void trans_go_xmote_bh(struct gfs2_glock *gl) struct gfs2_sbd *sdp = gl->gl_sbd; struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode); struct gfs2_glock *j_gl = ip->i_gl; - struct gfs2_log_header head; + struct gfs2_log_header_host head; int error; if (gl->gl_state != LM_ST_UNLOCKED && diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 8ca7a7f35062..e69f3394a2ce 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -41,7 +41,7 @@ struct gfs2_log_operations { void (*lo_before_commit) (struct gfs2_sbd *sdp); void (*lo_after_commit) (struct gfs2_sbd *sdp, struct gfs2_ail *ai); void (*lo_before_scan) (struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass); + struct gfs2_log_header_host *head, int pass); int (*lo_scan_elements) (struct gfs2_jdesc *jd, unsigned int start, struct gfs2_log_descriptor *ld, __be64 *ptr, int pass); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index ab6d1115f95d..8a654cd253dd 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -182,7 +182,7 @@ static void buf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) } static void buf_lo_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass) + struct gfs2_log_header_host *head, int pass) { struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); @@ -328,7 +328,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) } static void revoke_lo_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass) + struct gfs2_log_header_host *head, int pass) { struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); diff --git a/fs/gfs2/lops.h b/fs/gfs2/lops.h index 5839c05ae6be..965bc65c7c64 100644 --- a/fs/gfs2/lops.h +++ b/fs/gfs2/lops.h @@ -60,7 +60,7 @@ static inline void lops_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) } static inline void lops_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, + struct gfs2_log_header_host *head, unsigned int pass) { int x; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 64f5f0c604aa..84b1ebc7569e 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -251,7 +251,7 @@ void gfs2_dinode_print(const struct gfs2_dinode_host *di) printk(KERN_INFO " di_eattr = %llu\n", (unsigned long long)di->di_eattr); } -void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf) +void gfs2_log_header_in(struct gfs2_log_header_host *lh, const void *buf) { const struct gfs2_log_header *str = buf; diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index 62cd223819b7..447816241626 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -132,10 +132,10 @@ void gfs2_revoke_clean(struct gfs2_sbd *sdp) */ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, - struct gfs2_log_header *head) + struct gfs2_log_header_host *head) { struct buffer_head *bh; - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; u32 hash; int error; @@ -143,7 +143,7 @@ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, if (error) return error; - memcpy(&lh, bh->b_data, sizeof(struct gfs2_log_header)); + memcpy(&lh, bh->b_data, sizeof(struct gfs2_log_header)); /* XXX */ lh.lh_hash = 0; hash = gfs2_disk_hash((char *)&lh, sizeof(struct gfs2_log_header)); gfs2_log_header_in(&lh, bh->b_data); @@ -174,7 +174,7 @@ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, */ static int find_good_lh(struct gfs2_jdesc *jd, unsigned int *blk, - struct gfs2_log_header *head) + struct gfs2_log_header_host *head) { unsigned int orig_blk = *blk; int error; @@ -205,10 +205,10 @@ static int find_good_lh(struct gfs2_jdesc *jd, unsigned int *blk, * Returns: errno */ -static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { unsigned int blk = head->lh_blkno; - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; int error; for (;;) { @@ -245,9 +245,9 @@ static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header *head) * Returns: errno */ -int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { - struct gfs2_log_header lh_1, lh_m; + struct gfs2_log_header_host lh_1, lh_m; u32 blk_1, blk_2, blk_m; int error; @@ -320,7 +320,7 @@ static int foreach_descriptor(struct gfs2_jdesc *jd, unsigned int start, length = be32_to_cpu(ld->ld_length); if (be32_to_cpu(ld->ld_header.mh_type) == GFS2_METATYPE_LH) { - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; error = get_log_header(jd, start, &lh); if (!error) { gfs2_replay_incr_blk(sdp, &start); @@ -363,7 +363,7 @@ static int foreach_descriptor(struct gfs2_jdesc *jd, unsigned int start, * Returns: errno */ -static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { struct gfs2_inode *ip = GFS2_I(jd->jd_inode); struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); @@ -425,7 +425,7 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd) { struct gfs2_inode *ip = GFS2_I(jd->jd_inode); struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); - struct gfs2_log_header head; + struct gfs2_log_header_host head; struct gfs2_holder j_gh, ji_gh, t_gh; unsigned long t; int ro = 0; diff --git a/fs/gfs2/recovery.h b/fs/gfs2/recovery.h index 961feedf4d8b..f7235e61c723 100644 --- a/fs/gfs2/recovery.h +++ b/fs/gfs2/recovery.h @@ -26,7 +26,7 @@ int gfs2_revoke_check(struct gfs2_sbd *sdp, u64 blkno, unsigned int where); void gfs2_revoke_clean(struct gfs2_sbd *sdp); int gfs2_find_jhead(struct gfs2_jdesc *jd, - struct gfs2_log_header *head); + struct gfs2_log_header_host *head); int gfs2_recover_journal(struct gfs2_jdesc *gfs2_jd); void gfs2_check_journals(struct gfs2_sbd *sdp); diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 52aa3221fc99..0faf563c83a9 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -508,7 +508,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode); struct gfs2_glock *j_gl = ip->i_gl; struct gfs2_holder t_gh; - struct gfs2_log_header head; + struct gfs2_log_header_host head; int error; error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, @@ -873,7 +873,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd; struct lfcc *lfcc; LIST_HEAD(list); - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; int error; error = gfs2_jindex_hold(sdp, &ji_gh); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index c035587d066e..fb69a64c70cd 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -405,6 +405,16 @@ struct gfs2_log_header { __be32 lh_hash; }; +struct gfs2_log_header_host { + struct gfs2_meta_header lh_header; + + __u64 lh_sequence; /* Sequence number of this transaction */ + __u32 lh_flags; /* GFS2_LOG_HEAD_... */ + __u32 lh_tail; /* Block number of log tail */ + __u32 lh_blkno; + __u32 lh_hash; +}; + /* * Log type descriptor */ @@ -492,7 +502,7 @@ extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); extern void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); -extern void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf); +extern void gfs2_log_header_in(struct gfs2_log_header_host *lh, const void *buf); extern void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf); extern void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf); extern void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf); -- cgit v1.2.3 From e928a76f959e89884f6186bb6f846c533847d5df Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 21:57:23 -0400 Subject: [GFS2] split and annotate gfs2_meta_header Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/ondisk.c | 6 +++--- include/linux/gfs2_ondisk.h | 14 ++++++++++---- 2 files changed, 13 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 84b1ebc7569e..b5aa7ab97f28 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -54,7 +54,7 @@ static void gfs2_inum_print(const struct gfs2_inum *no) printk(KERN_INFO " no_addr = %llu\n", (unsigned long long)no->no_addr); } -static void gfs2_meta_header_in(struct gfs2_meta_header *mh, const void *buf) +static void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) { const struct gfs2_meta_header *str = buf; @@ -63,7 +63,7 @@ static void gfs2_meta_header_in(struct gfs2_meta_header *mh, const void *buf) mh->mh_format = be32_to_cpu(str->mh_format); } -static void gfs2_meta_header_out(const struct gfs2_meta_header *mh, void *buf) +static void gfs2_meta_header_out(const struct gfs2_meta_header_host *mh, void *buf) { struct gfs2_meta_header *str = buf; @@ -72,7 +72,7 @@ static void gfs2_meta_header_out(const struct gfs2_meta_header *mh, void *buf) str->mh_format = cpu_to_be32(mh->mh_format); } -static void gfs2_meta_header_print(const struct gfs2_meta_header *mh) +static void gfs2_meta_header_print(const struct gfs2_meta_header_host *mh) { pv(mh, mh_magic, "0x%.8X"); pv(mh, mh_type, "%u"); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index fb69a64c70cd..76eb9e1bb773 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -89,6 +89,12 @@ struct gfs2_meta_header { __be32 __pad1; /* Was incarnation number in gfs1 */ }; +struct gfs2_meta_header_host { + __u32 mh_magic; + __u32 mh_type; + __u32 mh_format; +}; + /* * super-block structure * @@ -129,7 +135,7 @@ struct gfs2_sb { }; struct gfs2_sb_host { - struct gfs2_meta_header sb_header; + struct gfs2_meta_header_host sb_header; __u32 sb_fs_format; __u32 sb_multihost_format; @@ -194,7 +200,7 @@ struct gfs2_rgrp { }; struct gfs2_rgrp_host { - struct gfs2_meta_header rg_header; + struct gfs2_meta_header_host rg_header; __u32 rg_flags; __u32 rg_free; @@ -297,7 +303,7 @@ struct gfs2_dinode { }; struct gfs2_dinode_host { - struct gfs2_meta_header di_header; + struct gfs2_meta_header_host di_header; struct gfs2_inum di_num; @@ -406,7 +412,7 @@ struct gfs2_log_header { }; struct gfs2_log_header_host { - struct gfs2_meta_header lh_header; + struct gfs2_meta_header_host lh_header; __u64 lh_sequence; /* Sequence number of this transaction */ __u32 lh_flags; /* GFS2_LOG_HEAD_... */ -- cgit v1.2.3 From 1e81c4c3e0f55c95b6278a827262b80debd0dc7e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 22:51:24 -0400 Subject: [GFS2] split and annotate gfs_rindex Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/incore.h | 2 +- fs/gfs2/ondisk.c | 4 ++-- fs/gfs2/rgrp.c | 2 +- include/linux/gfs2_ondisk.h | 14 +++++++++++--- 4 files changed, 15 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index e69f3394a2ce..e4afc2c4d609 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -67,7 +67,7 @@ struct gfs2_rgrpd { struct list_head rd_list_mru; struct list_head rd_recent; /* Recently used rgrps */ struct gfs2_glock *rd_gl; /* Glock for this rgrp */ - struct gfs2_rindex rd_ri; + struct gfs2_rindex_host rd_ri; struct gfs2_rgrp_host rd_rg; u64 rd_rg_vn; struct gfs2_bitmap *rd_bits; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index b5aa7ab97f28..c4e099d582f7 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -97,7 +97,7 @@ void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); } -void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf) +void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf) { const struct gfs2_rindex *str = buf; @@ -109,7 +109,7 @@ void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf) } -void gfs2_rindex_print(const struct gfs2_rindex *ri) +void gfs2_rindex_print(const struct gfs2_rindex_host *ri) { printk(KERN_INFO " ri_addr = %llu\n", (unsigned long long)ri->ri_addr); pv(ri, ri_length, "%u"); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index b261385c0065..07dfd6305058 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -253,7 +253,7 @@ void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd) } -static inline int rgrp_contains_block(struct gfs2_rindex *ri, u64 block) +static inline int rgrp_contains_block(struct gfs2_rindex_host *ri, u64 block) { u64 first = ri->ri_data0; u64 last = first + ri->ri_data; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 76eb9e1bb773..7dd5e4c18a66 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -168,6 +168,14 @@ struct gfs2_rindex { __u8 ri_reserved[64]; }; +struct gfs2_rindex_host { + __u64 ri_addr; /* grp block disk address */ + __u64 ri_data0; /* first data location */ + __u32 ri_length; /* length of rgrp header in fs blocks */ + __u32 ri_data; /* num of data blocks in rgrp */ + __u32 ri_bitbytes; /* number of bytes in data bitmaps */ +}; + /* * resource group header structure */ @@ -498,8 +506,8 @@ struct gfs2_quota_change { extern void gfs2_inum_in(struct gfs2_inum *no, const void *buf); extern void gfs2_inum_out(const struct gfs2_inum *no, void *buf); extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); -extern void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf); -extern void gfs2_rindex_out(const struct gfs2_rindex *ri, void *buf); +extern void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf); +extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota *qu, const void *buf); @@ -517,7 +525,7 @@ extern void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf); /* Printing functions */ -extern void gfs2_rindex_print(const struct gfs2_rindex *ri); +extern void gfs2_rindex_print(const struct gfs2_rindex_host *ri); extern void gfs2_dinode_print(const struct gfs2_dinode_host *di); #endif /* __KERNEL__ */ -- cgit v1.2.3 From 629a21e7ecedf779c68dcaa9a186069f57a7c652 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 22:51:24 -0400 Subject: [GFS2] split and annotate gfs2_inum Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/dir.c | 8 ++++---- fs/gfs2/dir.h | 8 ++++---- fs/gfs2/incore.h | 2 +- fs/gfs2/inode.c | 22 +++++++++++----------- fs/gfs2/inode.h | 4 ++-- fs/gfs2/ondisk.c | 6 +++--- fs/gfs2/ops_dentry.c | 2 +- fs/gfs2/ops_export.c | 8 ++++---- fs/gfs2/ops_export.h | 2 +- fs/gfs2/ops_file.c | 2 +- fs/gfs2/ops_fstype.c | 4 ++-- include/linux/gfs2_ondisk.h | 19 ++++++++++++------- 12 files changed, 46 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index e24af28b1a12..d67a3760ca30 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -1194,7 +1194,7 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, int *copied) { const struct gfs2_dirent *dent, *dent_next; - struct gfs2_inum inum; + struct gfs2_inum_host inum; u64 off, off_next; unsigned int x, y; int run = 0; @@ -1456,7 +1456,7 @@ out: */ int gfs2_dir_search(struct inode *dir, const struct qstr *name, - struct gfs2_inum *inum, unsigned int *type) + struct gfs2_inum_host *inum, unsigned int *type) { struct buffer_head *bh; struct gfs2_dirent *dent; @@ -1531,7 +1531,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) */ int gfs2_dir_add(struct inode *inode, const struct qstr *name, - const struct gfs2_inum *inum, unsigned type) + const struct gfs2_inum_host *inum, unsigned type) { struct gfs2_inode *ip = GFS2_I(inode); struct buffer_head *bh; @@ -1666,7 +1666,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) */ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, - struct gfs2_inum *inum, unsigned int new_type) + struct gfs2_inum_host *inum, unsigned int new_type) { struct buffer_head *bh; struct gfs2_dirent *dent; diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index 371233419b07..b21b33668a5b 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -31,17 +31,17 @@ struct gfs2_inum; typedef int (*gfs2_filldir_t) (void *opaque, const char *name, unsigned int length, u64 offset, - struct gfs2_inum *inum, unsigned int type); + struct gfs2_inum_host *inum, unsigned int type); int gfs2_dir_search(struct inode *dir, const struct qstr *filename, - struct gfs2_inum *inum, unsigned int *type); + struct gfs2_inum_host *inum, unsigned int *type); int gfs2_dir_add(struct inode *inode, const struct qstr *filename, - const struct gfs2_inum *inum, unsigned int type); + const struct gfs2_inum_host *inum, unsigned int type); int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename); int gfs2_dir_read(struct inode *inode, u64 * offset, void *opaque, gfs2_filldir_t filldir); int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, - struct gfs2_inum *new_inum, unsigned int new_type); + struct gfs2_inum_host *new_inum, unsigned int new_type); int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip); diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index e4afc2c4d609..20c9b4f0e52f 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -224,7 +224,7 @@ enum { struct gfs2_inode { struct inode i_inode; - struct gfs2_inum i_num; + struct gfs2_inum_host i_num; unsigned long i_flags; /* GIF_... */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 7eb6b440da6d..dadd1f35c864 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -112,7 +112,7 @@ void gfs2_inode_attr_out(struct gfs2_inode *ip) static int iget_test(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_inum *inum = opaque; + struct gfs2_inum_host *inum = opaque; if (ip && ip->i_num.no_addr == inum->no_addr) return 1; @@ -123,19 +123,19 @@ static int iget_test(struct inode *inode, void *opaque) static int iget_set(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_inum *inum = opaque; + struct gfs2_inum_host *inum = opaque; ip->i_num = *inum; return 0; } -struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum *inum) +struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum_host *inum) { return ilookup5(sb, (unsigned long)inum->no_formal_ino, iget_test, inum); } -static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum *inum) +static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum_host *inum) { return iget5_locked(sb, (unsigned long)inum->no_formal_ino, iget_test, iget_set, inum); @@ -150,7 +150,7 @@ static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum *inum) * Returns: A VFS inode, or an error */ -struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, unsigned int type) +struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum_host *inum, unsigned int type) { struct inode *inode = gfs2_iget(sb, inum); struct gfs2_inode *ip = GFS2_I(inode); @@ -394,7 +394,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, struct super_block *sb = dir->i_sb; struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_holder d_gh; - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error = 0; struct inode *inode = NULL; @@ -610,7 +610,7 @@ static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, *gid = current->fsgid; } -static int alloc_dinode(struct gfs2_inode *dip, struct gfs2_inum *inum, +static int alloc_dinode(struct gfs2_inode *dip, struct gfs2_inum_host *inum, u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); @@ -650,7 +650,7 @@ out: */ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - const struct gfs2_inum *inum, unsigned int mode, + const struct gfs2_inum_host *inum, unsigned int mode, unsigned int uid, unsigned int gid, const u64 *generation) { @@ -707,7 +707,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, } static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - unsigned int mode, const struct gfs2_inum *inum, + unsigned int mode, const struct gfs2_inum_host *inum, const u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); @@ -866,7 +866,7 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, struct gfs2_inode *dip = ghs->gh_gl->gl_object; struct inode *dir = &dip->i_inode; struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct gfs2_inum inum; + struct gfs2_inum_host inum; int error; u64 generation; @@ -1018,7 +1018,7 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip) { - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error; diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index f5d861760579..d699b9209750 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -27,8 +27,8 @@ static inline int gfs2_is_dir(struct gfs2_inode *ip) void gfs2_inode_attr_in(struct gfs2_inode *ip); void gfs2_inode_attr_out(struct gfs2_inode *ip); -struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, unsigned type); -struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum *inum); +struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum_host *inum, unsigned type); +struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum_host *inum); int gfs2_inode_refresh(struct gfs2_inode *ip); diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index c4e099d582f7..32f592f4a3c0 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -32,7 +32,7 @@ * first arg: the cpu-order structure */ -void gfs2_inum_in(struct gfs2_inum *no, const void *buf) +void gfs2_inum_in(struct gfs2_inum_host *no, const void *buf) { const struct gfs2_inum *str = buf; @@ -40,7 +40,7 @@ void gfs2_inum_in(struct gfs2_inum *no, const void *buf) no->no_addr = be64_to_cpu(str->no_addr); } -void gfs2_inum_out(const struct gfs2_inum *no, void *buf) +void gfs2_inum_out(const struct gfs2_inum_host *no, void *buf) { struct gfs2_inum *str = buf; @@ -48,7 +48,7 @@ void gfs2_inum_out(const struct gfs2_inum *no, void *buf) str->no_addr = cpu_to_be64(no->no_addr); } -static void gfs2_inum_print(const struct gfs2_inum *no) +static void gfs2_inum_print(const struct gfs2_inum_host *no) { printk(KERN_INFO " no_formal_ino = %llu\n", (unsigned long long)no->no_formal_ino); printk(KERN_INFO " no_addr = %llu\n", (unsigned long long)no->no_addr); diff --git a/fs/gfs2/ops_dentry.c b/fs/gfs2/ops_dentry.c index 00041b1b8025..c36f9e342e66 100644 --- a/fs/gfs2/ops_dentry.c +++ b/fs/gfs2/ops_dentry.c @@ -43,7 +43,7 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd) struct inode *inode = dentry->d_inode; struct gfs2_holder d_gh; struct gfs2_inode *ip; - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error; diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c index 86127d93bd35..33423a5c329b 100644 --- a/fs/gfs2/ops_export.c +++ b/fs/gfs2/ops_export.c @@ -35,7 +35,7 @@ static struct dentry *gfs2_decode_fh(struct super_block *sb, void *context) { struct gfs2_fh_obj fh_obj; - struct gfs2_inum *this, parent; + struct gfs2_inum_host *this, parent; if (fh_type != fh_len) return NULL; @@ -114,12 +114,12 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *fh, int *len, } struct get_name_filldir { - struct gfs2_inum inum; + struct gfs2_inum_host inum; char *name; }; static int get_name_filldir(void *opaque, const char *name, unsigned int length, - u64 offset, struct gfs2_inum *inum, + u64 offset, struct gfs2_inum_host *inum, unsigned int type) { struct get_name_filldir *gnfd = (struct get_name_filldir *)opaque; @@ -202,7 +202,7 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, void *inum_obj) { struct gfs2_sbd *sdp = sb->s_fs_info; struct gfs2_fh_obj *fh_obj = (struct gfs2_fh_obj *)inum_obj; - struct gfs2_inum *inum = &fh_obj->this; + struct gfs2_inum_host *inum = &fh_obj->this; struct gfs2_holder i_gh, ri_gh, rgd_gh; struct gfs2_rgrpd *rgd; struct inode *inode; diff --git a/fs/gfs2/ops_export.h b/fs/gfs2/ops_export.h index 09aca5046fb1..f925a955b3b8 100644 --- a/fs/gfs2/ops_export.h +++ b/fs/gfs2/ops_export.h @@ -15,7 +15,7 @@ extern struct export_operations gfs2_export_ops; struct gfs2_fh_obj { - struct gfs2_inum this; + struct gfs2_inum_host this; __u32 imode; }; diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 3064f133bf3c..359965c368bf 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -139,7 +139,7 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int origin) */ static int filldir_func(void *opaque, const char *name, unsigned int length, - u64 offset, struct gfs2_inum *inum, + u64 offset, struct gfs2_inum_host *inum, unsigned int type) { struct filldir_reg *fdr = (struct filldir_reg *)opaque; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 882873a6bd69..d14e139d2674 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -237,7 +237,7 @@ fail: } static struct inode *gfs2_lookup_root(struct super_block *sb, - struct gfs2_inum *inum) + struct gfs2_inum_host *inum) { return gfs2_inode_lookup(sb, inum, DT_DIR); } @@ -246,7 +246,7 @@ static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) { struct super_block *sb = sdp->sd_vfs; struct gfs2_holder sb_gh; - struct gfs2_inum *inum; + struct gfs2_inum_host *inum; struct inode *inode; int error = 0; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 7dd5e4c18a66..b16df6e8f55c 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -54,8 +54,13 @@ struct gfs2_inum { __be64 no_addr; }; -static inline int gfs2_inum_equal(const struct gfs2_inum *ino1, - const struct gfs2_inum *ino2) +struct gfs2_inum_host { + __u64 no_formal_ino; + __u64 no_addr; +}; + +static inline int gfs2_inum_equal(const struct gfs2_inum_host *ino1, + const struct gfs2_inum_host *ino2) { return ino1->no_formal_ino == ino2->no_formal_ino && ino1->no_addr == ino2->no_addr; @@ -143,8 +148,8 @@ struct gfs2_sb_host { __u32 sb_bsize; __u32 sb_bsize_shift; - struct gfs2_inum sb_master_dir; /* Was jindex dinode in gfs1 */ - struct gfs2_inum sb_root_dir; + struct gfs2_inum_host sb_master_dir; /* Was jindex dinode in gfs1 */ + struct gfs2_inum_host sb_root_dir; char sb_lockproto[GFS2_LOCKNAME_LEN]; char sb_locktable[GFS2_LOCKNAME_LEN]; @@ -313,7 +318,7 @@ struct gfs2_dinode { struct gfs2_dinode_host { struct gfs2_meta_header_host di_header; - struct gfs2_inum di_num; + struct gfs2_inum_host di_num; __u32 di_mode; /* mode of file */ __u32 di_uid; /* owner's user id */ @@ -503,8 +508,8 @@ struct gfs2_quota_change { #ifdef __KERNEL__ /* Translation functions */ -extern void gfs2_inum_in(struct gfs2_inum *no, const void *buf); -extern void gfs2_inum_out(const struct gfs2_inum *no, void *buf); +extern void gfs2_inum_in(struct gfs2_inum_host *no, const void *buf); +extern void gfs2_inum_out(const struct gfs2_inum_host *no, void *buf); extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); extern void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); -- cgit v1.2.3 From b5bc9e8b065dbcd4c675e8c158d6e524f221b8e1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 23:31:55 -0400 Subject: [GFS2] split and annotate gfs2_quota Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/ondisk.c | 2 +- fs/gfs2/quota.c | 2 +- include/linux/gfs2_ondisk.h | 9 +++++++-- 3 files changed, 9 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 32f592f4a3c0..5db0737fcdb5 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -144,7 +144,7 @@ void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf) memset(&str->rg_reserved, 0, sizeof(str->rg_reserved)); } -void gfs2_quota_in(struct gfs2_quota *qu, const void *buf) +void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf) { const struct gfs2_quota *str = buf; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index a3deae7416c9..e3f5b8da484c 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -743,7 +743,7 @@ static int do_glock(struct gfs2_quota_data *qd, int force_refresh, struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd; struct gfs2_inode *ip = GFS2_I(sdp->sd_quota_inode); struct gfs2_holder i_gh; - struct gfs2_quota q; + struct gfs2_quota_host q; char buf[sizeof(struct gfs2_quota)]; struct file_ra_state ra_state; int error; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index b16df6e8f55c..431e03b85458 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -232,6 +232,12 @@ struct gfs2_quota { __u8 qu_reserved[64]; }; +struct gfs2_quota_host { + __u64 qu_limit; + __u64 qu_warn; + __u64 qu_value; +}; + /* * dinode structure */ @@ -515,8 +521,7 @@ extern void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); -extern void gfs2_quota_in(struct gfs2_quota *qu, const void *buf); -extern void gfs2_quota_out(const struct gfs2_quota *qu, void *buf); +extern void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf); extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); extern void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); -- cgit v1.2.3 From bd209cc017f231e8536550bdab1bf5da93c32798 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 23:43:19 -0400 Subject: [GFS2] split and annotate gfs2_statfs_change Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/incore.h | 4 ++-- fs/gfs2/ondisk.c | 4 ++-- fs/gfs2/ops_super.c | 2 +- fs/gfs2/super.c | 22 +++++++++++----------- fs/gfs2/super.h | 4 ++-- include/linux/gfs2_ondisk.h | 10 ++++++++-- 6 files changed, 26 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 20c9b4f0e52f..c0a8c3b6a852 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -503,8 +503,8 @@ struct gfs2_sbd { spinlock_t sd_statfs_spin; struct mutex sd_statfs_mutex; - struct gfs2_statfs_change sd_statfs_master; - struct gfs2_statfs_change sd_statfs_local; + struct gfs2_statfs_change_host sd_statfs_master; + struct gfs2_statfs_change_host sd_statfs_local; unsigned long sd_statfs_sync_time; /* Resource group stuff */ diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 5db0737fcdb5..84c59b165b63 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -279,7 +279,7 @@ void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf) str->ir_length = cpu_to_be64(ir->ir_length); } -void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf) +void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf) { const struct gfs2_statfs_change *str = buf; @@ -288,7 +288,7 @@ void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf) sc->sc_dinodes = be64_to_cpu(str->sc_dinodes); } -void gfs2_statfs_change_out(const struct gfs2_statfs_change *sc, void *buf) +void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf) { struct gfs2_statfs_change *str = buf; diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index b47d9598c047..9c786d1e702d 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -215,7 +215,7 @@ static int gfs2_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_inode->i_sb; struct gfs2_sbd *sdp = sb->s_fs_info; - struct gfs2_statfs_change sc; + struct gfs2_statfs_change_host sc; int error; if (gfs2_tune_get(sdp, gt_statfs_slow)) diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 0faf563c83a9..0ef831727086 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -587,9 +587,9 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) int gfs2_statfs_init(struct gfs2_sbd *sdp) { struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct buffer_head *m_bh, *l_bh; struct gfs2_holder gh; int error; @@ -634,7 +634,7 @@ void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, s64 dinodes) { struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct buffer_head *l_bh; int error; @@ -660,8 +660,8 @@ int gfs2_statfs_sync(struct gfs2_sbd *sdp) { struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct gfs2_holder gh; struct buffer_head *m_bh, *l_bh; int error; @@ -727,10 +727,10 @@ out: * Returns: errno */ -int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) +int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc) { - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; spin_lock(&sdp->sd_statfs_spin); @@ -760,7 +760,7 @@ int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) */ static int statfs_slow_fill(struct gfs2_rgrpd *rgd, - struct gfs2_statfs_change *sc) + struct gfs2_statfs_change_host *sc) { gfs2_rgrp_verify(rgd); sc->sc_total += rgd->rd_ri.ri_data; @@ -782,7 +782,7 @@ static int statfs_slow_fill(struct gfs2_rgrpd *rgd, * Returns: errno */ -int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) +int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc) { struct gfs2_holder ri_gh; struct gfs2_rgrpd *rgd_next; @@ -792,7 +792,7 @@ int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) int done; int error = 0, err; - memset(sc, 0, sizeof(struct gfs2_statfs_change)); + memset(sc, 0, sizeof(struct gfs2_statfs_change_host)); gha = kcalloc(slots, sizeof(struct gfs2_holder), GFP_KERNEL); if (!gha) return -ENOMEM; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index ac95064c1e5a..e590b2df11dc 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -45,8 +45,8 @@ int gfs2_statfs_init(struct gfs2_sbd *sdp); void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, s64 dinodes); int gfs2_statfs_sync(struct gfs2_sbd *sdp); -int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc); -int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc); +int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc); +int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc); int gfs2_freeze_fs(struct gfs2_sbd *sdp); void gfs2_unfreeze_fs(struct gfs2_sbd *sdp); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 431e03b85458..3ce3a47b720a 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -497,6 +497,12 @@ struct gfs2_statfs_change { __be64 sc_dinodes; }; +struct gfs2_statfs_change_host { + __u64 sc_total; + __u64 sc_free; + __u64 sc_dinodes; +}; + /* * Quota change * Describes an allocation change for a particular @@ -529,8 +535,8 @@ extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); extern void gfs2_log_header_in(struct gfs2_log_header_host *lh, const void *buf); extern void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf); extern void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf); -extern void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf); -extern void gfs2_statfs_change_out(const struct gfs2_statfs_change *sc, void *buf); +extern void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf); +extern void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf); extern void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf); /* Printing functions */ -- cgit v1.2.3 From b62f963e1fdf838fed91faec21228d421a834f2d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 13 Oct 2006 23:46:46 -0400 Subject: [GFS2] split and annotate gfs2_quota_change Signed-off-by: Al Viro Signed-off-by: Steven Whitehouse --- fs/gfs2/ondisk.c | 2 +- fs/gfs2/quota.c | 2 +- include/linux/gfs2_ondisk.h | 8 +++++++- 3 files changed, 9 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 84c59b165b63..2d1682ded4b7 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -297,7 +297,7 @@ void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf) str->sc_dinodes = cpu_to_be64(sc->sc_dinodes); } -void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf) +void gfs2_quota_change_in(struct gfs2_quota_change_host *qc, const void *buf) { const struct gfs2_quota_change *str = buf; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index e3f5b8da484c..009d86c0008f 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -1103,7 +1103,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp) for (y = 0; y < sdp->sd_qc_per_block && slot < sdp->sd_quota_slots; y++, slot++) { - struct gfs2_quota_change qc; + struct gfs2_quota_change_host qc; struct gfs2_quota_data *qd; gfs2_quota_change_in(&qc, bh->b_data + diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 3ce3a47b720a..10a507dfd834 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -517,6 +517,12 @@ struct gfs2_quota_change { __be32 qc_id; }; +struct gfs2_quota_change_host { + __u64 qc_change; + __u32 qc_flags; /* GFS2_QCF_... */ + __u32 qc_id; +}; + #ifdef __KERNEL__ /* Translation functions */ @@ -537,7 +543,7 @@ extern void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf) extern void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf); extern void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf); extern void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf); -extern void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf); +extern void gfs2_quota_change_in(struct gfs2_quota_change_host *qc, const void *buf); /* Printing functions */ -- cgit v1.2.3 From 539e5d6b7ae8612c0393fe940d2da5b591318d3d Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 31 Oct 2006 15:07:05 -0500 Subject: [GFS2] Change argument of gfs2_dinode_out Everywhere this was called, a struct gfs2_inode was available, but despite that, it was always called with a struct gfs2_dinode as an argument. By making this change it paves the way to start eliminating fields duplicated between the kernel's struct inode and the struct gfs2_dinode. Signed-off-by: Steven Whitehouse --- fs/gfs2/acl.c | 2 +- fs/gfs2/bmap.c | 12 ++++++------ fs/gfs2/dir.c | 20 ++++++++++---------- fs/gfs2/eattr.c | 14 +++++++------- fs/gfs2/inode.c | 6 +++--- fs/gfs2/ondisk.c | 5 ++++- fs/gfs2/ops_file.c | 2 +- fs/gfs2/ops_inode.c | 10 +++++----- include/linux/gfs2_ondisk.h | 3 ++- 9 files changed, 39 insertions(+), 35 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 5f959b8ce406..906e403b054f 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -201,7 +201,7 @@ static int munge_mode(struct gfs2_inode *ip, mode_t mode) (ip->i_di.di_mode & S_IFMT) == (mode & S_IFMT)); ip->i_di.di_mode = mode; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 51f6356bdcb5..8c092ab2b4ba 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -498,7 +498,7 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } set_buffer_new(bh_map); @@ -780,7 +780,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); up_write(&ip->i_rw_mutex); @@ -860,7 +860,7 @@ static int do_grow(struct gfs2_inode *ip, u64 size) goto out_end_trans; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out_end_trans: @@ -970,7 +970,7 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) ip->i_di.di_size = size; ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + size); error = 1; @@ -983,7 +983,7 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); ip->i_di.di_flags |= GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); } } @@ -1057,7 +1057,7 @@ static int trunc_end(struct gfs2_inode *ip) ip->i_di.di_flags &= ~GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out: diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 59dc823c2833..0742761e1e02 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -132,7 +132,7 @@ static int gfs2_dir_write_stuffed(struct gfs2_inode *ip, const char *buf, if (ip->i_di.di_size < offset + size) ip->i_di.di_size = offset + size; ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -232,7 +232,7 @@ out: ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); return copied; @@ -907,7 +907,7 @@ static int dir_make_exhash(struct inode *inode) for (x = sdp->sd_hash_ptrs, y = -1; x; x >>= 1, y++) ; dip->i_di.di_depth = y; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); @@ -1039,7 +1039,7 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name) error = gfs2_meta_inode_buffer(dip, &dibh); if (!gfs2_assert_withdraw(GFS2_SB(&dip->i_inode), !error)) { dip->i_di.di_blocks++; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); } @@ -1119,7 +1119,7 @@ static int dir_double_exhash(struct gfs2_inode *dip) error = gfs2_meta_inode_buffer(dip, &dibh); if (!gfs2_assert_withdraw(sdp, !error)) { dip->i_di.di_depth++; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); } @@ -1517,7 +1517,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) return error; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_blocks++; - gfs2_dinode_out(&ip->i_di, bh->b_data); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); return 0; } @@ -1561,7 +1561,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_entries++; ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&ip->i_di, bh->b_data); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); error = 0; break; @@ -1647,7 +1647,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) gfs2_trans_add_bh(dip->i_gl, bh, 1); dip->i_di.di_entries--; dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&dip->i_di, bh->b_data); + gfs2_dinode_out(dip, bh->b_data); brelse(bh); mark_inode_dirty(&dip->i_inode); @@ -1695,7 +1695,7 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, } dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&dip->i_di, bh->b_data); + gfs2_dinode_out(dip, bh->b_data); brelse(bh); return 0; } @@ -1875,7 +1875,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_end_trans; gfs2_trans_add_bh(dip->i_gl, dibh, 1); - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); out_end_trans: diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index 518f0c0786c8..9b7bb565b59d 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -302,7 +302,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, if (!error) { ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -717,7 +717,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, } ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -852,7 +852,7 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh, } ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out: gfs2_trans_end(GFS2_SB(&ip->i_inode)); @@ -1132,7 +1132,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el) if (!error) { ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1287,7 +1287,7 @@ int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el, gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error); gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1397,7 +1397,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1446,7 +1446,7 @@ static int ea_dealloc_block(struct gfs2_inode *ip) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index fb969302f181..b861ddba8688 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -338,7 +338,7 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) ip->i_inode.i_nlink = nlink; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); mark_inode_dirty(&ip->i_inode); @@ -792,7 +792,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, goto fail_end_trans; ip->i_di.di_nlink = 1; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); return 0; @@ -1349,7 +1349,7 @@ __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } return error; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 2d1682ded4b7..2c50fa0261f5 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -15,6 +15,8 @@ #include "gfs2.h" #include +#include +#include "incore.h" #define pv(struct, member, fmt) printk(KERN_INFO " "#member" = "fmt"\n", \ struct->member); @@ -187,8 +189,9 @@ void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf) } -void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf) +void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) { + const struct gfs2_dinode_host *di = &ip->i_di; struct gfs2_dinode *str = buf; gfs2_meta_header_out(&di->di_header, buf); diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 2fc8868e065c..7ea417573903 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -336,7 +336,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) goto out_trans_end; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_flags = new_flags; - gfs2_dinode_out(&ip->i_di, bh->b_data); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); out_trans_end: gfs2_trans_end(sdp); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index ef6e5ed70e94..bd268852c674 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -339,7 +339,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, error = gfs2_meta_inode_buffer(ip, &dibh); if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, size); brelse(dibh); @@ -414,7 +414,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) gfs2_inum_out(&dip->i_num, &dent->de_inum); dent->de_type = cpu_to_be16(DT_DIR); - gfs2_dinode_out(&ip->i_di, di); + gfs2_dinode_out(ip, di); brelse(dibh); } @@ -541,7 +541,7 @@ static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, error = gfs2_meta_inode_buffer(ip, &dibh); if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -762,7 +762,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, goto out_end_trans; ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -949,7 +949,7 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 10a507dfd834..550effaf0c2b 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -535,7 +535,8 @@ extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf); extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); -extern void gfs2_dinode_out(const struct gfs2_dinode_host *di, void *buf); +struct gfs2_inode; +extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); extern void gfs2_log_header_in(struct gfs2_log_header_host *lh, const void *buf); -- cgit v1.2.3 From 891ea14712da68e282de8583e5fa14f0d3f3731e Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 31 Oct 2006 15:22:10 -0500 Subject: [GFS2] Change argument to gfs2_dinode_in This is a preliminary patch to enable the removal of fields in gfs2_dinode_host which are duplicated in struct inode. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 2 +- fs/gfs2/ondisk.c | 4 ++-- include/linux/gfs2_ondisk.h | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index b861ddba8688..9875e9356cd3 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -229,7 +229,7 @@ int gfs2_inode_refresh(struct gfs2_inode *ip) return -EIO; } - gfs2_dinode_in(&ip->i_di, dibh->b_data); + gfs2_dinode_in(ip, dibh->b_data); brelse(dibh); diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 2c50fa0261f5..edf87567bb2b 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -155,8 +155,9 @@ void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf) qu->qu_value = be64_to_cpu(str->qu_value); } -void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf) +void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) { + struct gfs2_dinode_host *di = &ip->i_di; const struct gfs2_dinode *str = buf; gfs2_meta_header_in(&di->di_header, buf); @@ -186,7 +187,6 @@ void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf) di->di_entries = be32_to_cpu(str->di_entries); di->di_eattr = be64_to_cpu(str->di_eattr); - } void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 550effaf0c2b..08d8240ba533 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -534,8 +534,8 @@ extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf); -extern void gfs2_dinode_in(struct gfs2_dinode_host *di, const void *buf); struct gfs2_inode; +extern void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); -- cgit v1.2.3 From ea744d01c6a5acf1f6171b4c6e1658a742063613 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 31 Oct 2006 15:28:00 -0500 Subject: [GFS2] Move gfs2_dinode_in to inode.c gfs2_dinode_in() is only ever called from one place, so move it to that place (in inode.c) and make it static. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 34 ++++++++++++++++++++++++++++++++++ fs/gfs2/ondisk.c | 36 +----------------------------------- include/linux/gfs2_ondisk.h | 2 +- 3 files changed, 36 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 9875e9356cd3..b39cfcfe9276 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -208,6 +208,40 @@ fail: return ERR_PTR(error); } +static void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) +{ + struct gfs2_dinode_host *di = &ip->i_di; + const struct gfs2_dinode *str = buf; + + gfs2_meta_header_in(&di->di_header, buf); + gfs2_inum_in(&di->di_num, &str->di_num); + + di->di_mode = be32_to_cpu(str->di_mode); + di->di_uid = be32_to_cpu(str->di_uid); + di->di_gid = be32_to_cpu(str->di_gid); + di->di_nlink = be32_to_cpu(str->di_nlink); + di->di_size = be64_to_cpu(str->di_size); + di->di_blocks = be64_to_cpu(str->di_blocks); + di->di_atime = be64_to_cpu(str->di_atime); + di->di_mtime = be64_to_cpu(str->di_mtime); + di->di_ctime = be64_to_cpu(str->di_ctime); + di->di_major = be32_to_cpu(str->di_major); + di->di_minor = be32_to_cpu(str->di_minor); + + di->di_goal_meta = be64_to_cpu(str->di_goal_meta); + di->di_goal_data = be64_to_cpu(str->di_goal_data); + di->di_generation = be64_to_cpu(str->di_generation); + + di->di_flags = be32_to_cpu(str->di_flags); + di->di_payload_format = be32_to_cpu(str->di_payload_format); + di->di_height = be16_to_cpu(str->di_height); + + di->di_depth = be16_to_cpu(str->di_depth); + di->di_entries = be32_to_cpu(str->di_entries); + + di->di_eattr = be64_to_cpu(str->di_eattr); +} + /** * gfs2_inode_refresh - Refresh the incore copy of the dinode * @ip: The GFS2 inode diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index edf87567bb2b..062a44f87fec 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -56,7 +56,7 @@ static void gfs2_inum_print(const struct gfs2_inum_host *no) printk(KERN_INFO " no_addr = %llu\n", (unsigned long long)no->no_addr); } -static void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) +void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) { const struct gfs2_meta_header *str = buf; @@ -155,40 +155,6 @@ void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf) qu->qu_value = be64_to_cpu(str->qu_value); } -void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) -{ - struct gfs2_dinode_host *di = &ip->i_di; - const struct gfs2_dinode *str = buf; - - gfs2_meta_header_in(&di->di_header, buf); - gfs2_inum_in(&di->di_num, &str->di_num); - - di->di_mode = be32_to_cpu(str->di_mode); - di->di_uid = be32_to_cpu(str->di_uid); - di->di_gid = be32_to_cpu(str->di_gid); - di->di_nlink = be32_to_cpu(str->di_nlink); - di->di_size = be64_to_cpu(str->di_size); - di->di_blocks = be64_to_cpu(str->di_blocks); - di->di_atime = be64_to_cpu(str->di_atime); - di->di_mtime = be64_to_cpu(str->di_mtime); - di->di_ctime = be64_to_cpu(str->di_ctime); - di->di_major = be32_to_cpu(str->di_major); - di->di_minor = be32_to_cpu(str->di_minor); - - di->di_goal_meta = be64_to_cpu(str->di_goal_meta); - di->di_goal_data = be64_to_cpu(str->di_goal_data); - di->di_generation = be64_to_cpu(str->di_generation); - - di->di_flags = be32_to_cpu(str->di_flags); - di->di_payload_format = be32_to_cpu(str->di_payload_format); - di->di_height = be16_to_cpu(str->di_height); - - di->di_depth = be16_to_cpu(str->di_depth); - di->di_entries = be32_to_cpu(str->di_entries); - - di->di_eattr = be64_to_cpu(str->di_eattr); -} - void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) { const struct gfs2_dinode_host *di = &ip->i_di; diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 08d8240ba533..4fc297a1ef95 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -528,6 +528,7 @@ struct gfs2_quota_change_host { extern void gfs2_inum_in(struct gfs2_inum_host *no, const void *buf); extern void gfs2_inum_out(const struct gfs2_inum_host *no, void *buf); +extern void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf); extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); extern void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); @@ -535,7 +536,6 @@ extern void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf); extern void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf); extern void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf); struct gfs2_inode; -extern void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); extern void gfs2_ea_header_in(struct gfs2_ea_header *ea, const void *buf); extern void gfs2_ea_header_out(const struct gfs2_ea_header *ea, void *buf); -- cgit v1.2.3 From 4cc14f0b88bf3e0b508143e091eb5a8dff3e3b9c Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 31 Oct 2006 19:00:24 -0500 Subject: [GFS2] Change argument to gfs2_dinode_print Change argument for gfs2_dinode_print in order to prepare for removal of duplicate fields between struct inode and struct gfs2_dinode_host. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 8 ++++---- fs/gfs2/ondisk.c | 4 +++- fs/gfs2/ops_inode.c | 4 ++-- include/linux/gfs2_ondisk.h | 2 +- 4 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index b39cfcfe9276..4c5d286fefdb 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -269,7 +269,7 @@ int gfs2_inode_refresh(struct gfs2_inode *ip) if (ip->i_num.no_addr != ip->i_di.di_num.no_addr) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } if (ip->i_num.no_formal_ino != ip->i_di.di_num.no_formal_ino) @@ -289,7 +289,7 @@ int gfs2_dinode_dealloc(struct gfs2_inode *ip) if (ip->i_di.di_blocks != 1) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } @@ -359,7 +359,7 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) bigger than the old one, we must have underflowed. */ if (diff < 0 && nlink > ip->i_di.di_nlink) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } @@ -1010,7 +1010,7 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, if (ip->i_di.di_entries != 2) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 062a44f87fec..77bed440833d 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -190,8 +190,10 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) } -void gfs2_dinode_print(const struct gfs2_dinode_host *di) +void gfs2_dinode_print(const struct gfs2_inode *ip) { + const struct gfs2_dinode_host *di = &ip->i_di; + gfs2_meta_header_print(&di->di_header); gfs2_inum_print(&di->di_num); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index bd268852c674..b2c2fe613d70 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -467,7 +467,7 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) if (ip->i_di.di_entries < 2) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); error = -EIO; goto out_gunlock; } @@ -640,7 +640,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (S_ISDIR(nip->i_di.di_mode)) { if (nip->i_di.di_entries < 2) { if (gfs2_consist_inode(nip)) - gfs2_dinode_print(&nip->i_di); + gfs2_dinode_print(nip); error = -EIO; goto out_gunlock; } diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 4fc297a1ef95..cf4c655d0d5f 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -549,7 +549,7 @@ extern void gfs2_quota_change_in(struct gfs2_quota_change_host *qc, const void * /* Printing functions */ extern void gfs2_rindex_print(const struct gfs2_rindex_host *ri); -extern void gfs2_dinode_print(const struct gfs2_dinode_host *di); +extern void gfs2_dinode_print(const struct gfs2_inode *ip); #endif /* __KERNEL__ */ -- cgit v1.2.3 From af339c0241d0dd3b35f9097b4f4999bb22ffe502 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 10:34:15 -0500 Subject: [GFS2] Shrink gfs2_inode (1) - di_header/di_num The metadata header doesn't need to be stored in the incore struct gfs2_inode since its constant, and this patch removes it. Also, there is already a field for the inode's number in the struct gfs2_inode, so we don't need one in struct gfs2_dinode_host as well. This saves 28 bytes of space in the struct gfs2_inode. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 25 +++++++++++-------------- fs/gfs2/ondisk.c | 22 +++++++++------------- include/linux/gfs2_ondisk.h | 5 ----- 3 files changed, 20 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 4c5d286fefdb..7ba05fc553aa 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -208,13 +208,18 @@ fail: return ERR_PTR(error); } -static void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) +static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) { struct gfs2_dinode_host *di = &ip->i_di; const struct gfs2_dinode *str = buf; - gfs2_meta_header_in(&di->di_header, buf); - gfs2_inum_in(&di->di_num, &str->di_num); + if (ip->i_num.no_addr != be64_to_cpu(str->di_num.no_addr)) { + if (gfs2_consist_inode(ip)) + gfs2_dinode_print(ip); + return -EIO; + } + if (ip->i_num.no_formal_ino != be64_to_cpu(str->di_num.no_formal_ino)) + return -ESTALE; di->di_mode = be32_to_cpu(str->di_mode); di->di_uid = be32_to_cpu(str->di_uid); @@ -240,6 +245,7 @@ static void gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_entries = be32_to_cpu(str->di_entries); di->di_eattr = be64_to_cpu(str->di_eattr); + return 0; } /** @@ -263,21 +269,12 @@ int gfs2_inode_refresh(struct gfs2_inode *ip) return -EIO; } - gfs2_dinode_in(ip, dibh->b_data); + error = gfs2_dinode_in(ip, dibh->b_data); brelse(dibh); - - if (ip->i_num.no_addr != ip->i_di.di_num.no_addr) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - if (ip->i_num.no_formal_ino != ip->i_di.di_num.no_formal_ino) - return -ESTALE; - ip->i_vn = ip->i_gl->gl_vn; - return 0; + return error; } int gfs2_dinode_dealloc(struct gfs2_inode *ip) diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 77bed440833d..4bc590edb60e 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -56,7 +56,7 @@ static void gfs2_inum_print(const struct gfs2_inum_host *no) printk(KERN_INFO " no_addr = %llu\n", (unsigned long long)no->no_addr); } -void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) +static void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) { const struct gfs2_meta_header *str = buf; @@ -74,13 +74,6 @@ static void gfs2_meta_header_out(const struct gfs2_meta_header_host *mh, void *b str->mh_format = cpu_to_be32(mh->mh_format); } -static void gfs2_meta_header_print(const struct gfs2_meta_header_host *mh) -{ - pv(mh, mh_magic, "0x%.8X"); - pv(mh, mh_type, "%u"); - pv(mh, mh_format, "%u"); -} - void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) { const struct gfs2_sb *str = buf; @@ -160,8 +153,13 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) const struct gfs2_dinode_host *di = &ip->i_di; struct gfs2_dinode *str = buf; - gfs2_meta_header_out(&di->di_header, buf); - gfs2_inum_out(&di->di_num, (char *)&str->di_num); + str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); + str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); + str->di_header.__pad0 = 0; + str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); + str->di_header.__pad1 = 0; + + gfs2_inum_out(&ip->i_num, &str->di_num); str->di_mode = cpu_to_be32(di->di_mode); str->di_uid = cpu_to_be32(di->di_uid); @@ -187,15 +185,13 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) str->di_entries = cpu_to_be32(di->di_entries); str->di_eattr = cpu_to_be64(di->di_eattr); - } void gfs2_dinode_print(const struct gfs2_inode *ip) { const struct gfs2_dinode_host *di = &ip->i_di; - gfs2_meta_header_print(&di->di_header); - gfs2_inum_print(&di->di_num); + gfs2_inum_print(&ip->i_num); pv(di, di_mode, "0%o"); pv(di, di_uid, "%u"); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index cf4c655d0d5f..c0e76fc718c7 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -322,10 +322,6 @@ struct gfs2_dinode { }; struct gfs2_dinode_host { - struct gfs2_meta_header_host di_header; - - struct gfs2_inum_host di_num; - __u32 di_mode; /* mode of file */ __u32 di_uid; /* owner's user id */ __u32 di_gid; /* owner's group id */ @@ -528,7 +524,6 @@ struct gfs2_quota_change_host { extern void gfs2_inum_in(struct gfs2_inum_host *no, const void *buf); extern void gfs2_inum_out(const struct gfs2_inum_host *no, void *buf); -extern void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf); extern void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf); extern void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf); extern void gfs2_rindex_out(const struct gfs2_rindex_host *ri, void *buf); -- cgit v1.2.3 From e7f14f4d094ea1a9ce1953375f5bc1500c760c79 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 31 Oct 2006 21:45:08 -0500 Subject: [GFS2] Shrink gfs2_inode (2) - di_major/di_minor This removes the device numbers from this structure by using inode->i_rdev instead. It also cleans up the code in gfs2_mknod. It results in shrinking the gfs2_inode by 8 bytes. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 36 ++++++++++++++++-------------------- fs/gfs2/inode.h | 2 +- fs/gfs2/ondisk.c | 4 ---- fs/gfs2/ops_inode.c | 38 +++++--------------------------------- include/linux/gfs2_ondisk.h | 2 -- 5 files changed, 22 insertions(+), 60 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 7ba05fc553aa..a99591956544 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -51,17 +51,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) struct gfs2_dinode_host *di = &ip->i_di; inode->i_ino = ip->i_num.no_addr; - - switch (di->di_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - inode->i_rdev = MKDEV(di->di_major, di->di_minor); - break; - default: - inode->i_rdev = 0; - break; - }; - inode->i_mode = di->di_mode; inode->i_nlink = di->di_nlink; inode->i_uid = di->di_uid; @@ -222,6 +211,15 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) return -ESTALE; di->di_mode = be32_to_cpu(str->di_mode); + ip->i_inode.i_rdev = 0; + switch (di->di_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), + be32_to_cpu(str->di_minor)); + break; + }; + di->di_uid = be32_to_cpu(str->di_uid); di->di_gid = be32_to_cpu(str->di_gid); di->di_nlink = be32_to_cpu(str->di_nlink); @@ -230,8 +228,6 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_atime = be64_to_cpu(str->di_atime); di->di_mtime = be64_to_cpu(str->di_mtime); di->di_ctime = be64_to_cpu(str->di_ctime); - di->di_major = be32_to_cpu(str->di_major); - di->di_minor = be32_to_cpu(str->di_minor); di->di_goal_meta = be64_to_cpu(str->di_goal_meta); di->di_goal_data = be64_to_cpu(str->di_goal_data); @@ -270,7 +266,6 @@ int gfs2_inode_refresh(struct gfs2_inode *ip) } error = gfs2_dinode_in(ip, dibh->b_data); - brelse(dibh); ip->i_vn = ip->i_gl->gl_vn; @@ -684,7 +679,7 @@ out: static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, const struct gfs2_inum_host *inum, unsigned int mode, unsigned int uid, unsigned int gid, - const u64 *generation) + const u64 *generation, dev_t dev) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_dinode *di; @@ -705,7 +700,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_size = cpu_to_be64(0); di->di_blocks = cpu_to_be64(1); di->di_atime = di->di_mtime = di->di_ctime = cpu_to_be64(get_seconds()); - di->di_major = di->di_minor = cpu_to_be32(0); + di->di_major = cpu_to_be32(MAJOR(dev)); + di->di_minor = cpu_to_be32(MINOR(dev)); di->di_goal_meta = di->di_goal_data = cpu_to_be64(inum->no_addr); di->di_generation = cpu_to_be64(*generation); di->di_flags = cpu_to_be32(0); @@ -740,7 +736,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, unsigned int mode, const struct gfs2_inum_host *inum, - const u64 *generation) + const u64 *generation, dev_t dev) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); unsigned int uid, gid; @@ -761,7 +757,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, if (error) goto out_quota; - init_dinode(dip, gl, inum, mode, uid, gid, generation); + init_dinode(dip, gl, inum, mode, uid, gid, generation, dev); gfs2_quota_change(dip, +1, uid, gid); gfs2_trans_end(sdp); @@ -892,7 +888,7 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip) */ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode) + unsigned int mode, dev_t dev) { struct inode *inode; struct gfs2_inode *dip = ghs->gh_gl->gl_object; @@ -950,7 +946,7 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, goto fail_gunlock; } - error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation); + error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev); if (error) goto fail_gunlock2; diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index d699b9209750..33c9ea68f7e7 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -37,7 +37,7 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff); struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, int is_root, struct nameidata *nd); struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode); + unsigned int mode, dev_t dev); int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip); int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 4bc590edb60e..60dd943761ab 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -170,8 +170,6 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) str->di_atime = cpu_to_be64(di->di_atime); str->di_mtime = cpu_to_be64(di->di_mtime); str->di_ctime = cpu_to_be64(di->di_ctime); - str->di_major = cpu_to_be32(di->di_major); - str->di_minor = cpu_to_be32(di->di_minor); str->di_goal_meta = cpu_to_be64(di->di_goal_meta); str->di_goal_data = cpu_to_be64(di->di_goal_data); @@ -202,8 +200,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) printk(KERN_INFO " di_atime = %lld\n", (long long)di->di_atime); printk(KERN_INFO " di_mtime = %lld\n", (long long)di->di_mtime); printk(KERN_INFO " di_ctime = %lld\n", (long long)di->di_ctime); - pv(di, di_major, "%u"); - pv(di, di_minor, "%u"); printk(KERN_INFO " di_goal_meta = %llu\n", (unsigned long long)di->di_goal_meta); printk(KERN_INFO " di_goal_data = %llu\n", (unsigned long long)di->di_goal_data); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index b2c2fe613d70..c10b914bf8cc 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -59,7 +59,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, gfs2_holder_init(dip->i_gl, 0, 0, ghs); for (;;) { - inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0); if (!IS_ERR(inode)) { gfs2_trans_end(sdp); if (dip->i_alloc.al_rgd) @@ -326,7 +326,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO, 0); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); @@ -379,7 +379,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode, 0); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); @@ -504,47 +504,19 @@ out: static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) { - struct gfs2_inode *dip = GFS2_I(dir), *ip; + struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_sbd *sdp = GFS2_SB(dir); struct gfs2_holder ghs[2]; struct inode *inode; - struct buffer_head *dibh; - u32 major = 0, minor = 0; - int error; - - switch (mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - major = MAJOR(dev); - minor = MINOR(dev); - break; - case S_IFIFO: - case S_IFSOCK: - break; - default: - return -EOPNOTSUPP; - }; gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, mode); + inode = gfs2_createi(ghs, &dentry->d_name, mode, dev); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); } - ip = ghs[1].gh_gl->gl_object; - - ip->i_di.di_major = major; - ip->i_di.di_minor = minor; - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - } - gfs2_trans_end(sdp); if (dip->i_alloc.al_rgd) gfs2_inplace_release(dip); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index c0e76fc718c7..5bcf895b2731 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -331,8 +331,6 @@ struct gfs2_dinode_host { __u64 di_atime; /* time last accessed */ __u64 di_mtime; /* time last modified */ __u64 di_ctime; /* time last changed */ - __u32 di_major; /* device major number */ - __u32 di_minor; /* device minor number */ /* This section varies from gfs1. Padding added to align with * remainder of dinode -- cgit v1.2.3 From b60623c238b6a819bd04090139704e2cb57a751f Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 12:22:46 -0500 Subject: [GFS2] Shrink gfs2_inode (3) - di_mode This removes the duplicate di_mode field in favour of using the inode->i_mode field. This saves 4 bytes. Signed-off-by: Steven Whitehouse --- fs/gfs2/acl.c | 16 ++++++++-------- fs/gfs2/bmap.c | 2 +- fs/gfs2/eaops.c | 2 +- fs/gfs2/eattr.c | 8 ++++---- fs/gfs2/glock.c | 2 +- fs/gfs2/glops.c | 8 ++++---- fs/gfs2/inode.c | 18 +++++++----------- fs/gfs2/inode.h | 2 +- fs/gfs2/ondisk.c | 3 +-- fs/gfs2/ops_address.c | 1 - fs/gfs2/ops_dentry.c | 2 +- fs/gfs2/ops_file.c | 6 +++--- fs/gfs2/ops_inode.c | 17 ++++++++--------- fs/gfs2/ops_super.c | 2 +- include/linux/gfs2_ondisk.h | 1 - 15 files changed, 41 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 906e403b054f..87f6304f2f54 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -76,9 +76,9 @@ int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access) return -EOPNOTSUPP; if (current->fsuid != ip->i_di.di_uid && !capable(CAP_FOWNER)) return -EPERM; - if (S_ISLNK(ip->i_di.di_mode)) + if (S_ISLNK(ip->i_inode.i_mode)) return -EOPNOTSUPP; - if (!access && !S_ISDIR(ip->i_di.di_mode)) + if (!access && !S_ISDIR(ip->i_inode.i_mode)) return -EACCES; return 0; @@ -198,8 +198,8 @@ static int munge_mode(struct gfs2_inode *ip, mode_t mode) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_assert_withdraw(sdp, - (ip->i_di.di_mode & S_IFMT) == (mode & S_IFMT)); - ip->i_di.di_mode = mode; + (ip->i_inode.i_mode & S_IFMT) == (mode & S_IFMT)); + ip->i_inode.i_mode = mode; gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -215,12 +215,12 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct posix_acl *acl = NULL, *clone; struct gfs2_ea_request er; - mode_t mode = ip->i_di.di_mode; + mode_t mode = ip->i_inode.i_mode; int error; if (!sdp->sd_args.ar_posix_acl) return 0; - if (S_ISLNK(ip->i_di.di_mode)) + if (S_ISLNK(ip->i_inode.i_mode)) return 0; memset(&er, 0, sizeof(struct gfs2_ea_request)); @@ -232,7 +232,7 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) return error; if (!acl) { mode &= ~current->fs->umask; - if (mode != ip->i_di.di_mode) + if (mode != ip->i_inode.i_mode) error = munge_mode(ip, mode); return error; } @@ -244,7 +244,7 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) posix_acl_release(acl); acl = clone; - if (S_ISDIR(ip->i_di.di_mode)) { + if (S_ISDIR(ip->i_inode.i_mode)) { er.er_name = GFS2_POSIX_ACL_DEFAULT; er.er_name_len = GFS2_POSIX_ACL_DEFAULT_LEN; error = gfs2_system_eaops.eo_set(ip, &er); diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 8c092ab2b4ba..481a06882544 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1109,7 +1109,7 @@ int gfs2_truncatei(struct gfs2_inode *ip, u64 size) { int error; - if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_di.di_mode))) + if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_inode.i_mode))) return -EINVAL; if (size > ip->i_di.di_size) diff --git a/fs/gfs2/eaops.c b/fs/gfs2/eaops.c index 92c54e9b0dc3..cd747c00f670 100644 --- a/fs/gfs2/eaops.c +++ b/fs/gfs2/eaops.c @@ -120,7 +120,7 @@ static int system_eo_set(struct gfs2_inode *ip, struct gfs2_ea_request *er) if (GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len)) { if (!(er->er_flags & GFS2_ERF_MODE)) { - er->er_mode = ip->i_di.di_mode; + er->er_mode = ip->i_inode.i_mode; er->er_flags |= GFS2_ERF_MODE; } error = gfs2_acl_validate_set(ip, 1, er, diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index 9b7bb565b59d..5208fa94aad2 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -711,9 +711,9 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, if (!error) { if (er->er_flags & GFS2_ERF_MODE) { gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), - (ip->i_di.di_mode & S_IFMT) == + (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT)); - ip->i_di.di_mode = er->er_mode; + ip->i_inode.i_mode = er->er_mode; } ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); @@ -847,8 +847,8 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh, if (er->er_flags & GFS2_ERF_MODE) { gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), - (ip->i_di.di_mode & S_IFMT) == (er->er_mode & S_IFMT)); - ip->i_di.di_mode = er->er_mode; + (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT)); + ip->i_inode.i_mode = er->er_mode; } ip->i_di.di_ctime = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 78fe0fae23ff..44633c46717a 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -2078,7 +2078,7 @@ static int dump_inode(struct gfs2_inode *ip) printk(KERN_INFO " num = %llu %llu\n", (unsigned long long)ip->i_num.no_formal_ino, (unsigned long long)ip->i_num.no_addr); - printk(KERN_INFO " type = %u\n", IF2DT(ip->i_di.di_mode)); + printk(KERN_INFO " type = %u\n", IF2DT(ip->i_inode.i_mode)); printk(KERN_INFO " i_flags ="); for (x = 0; x < 32; x++) if (test_bit(x, &ip->i_flags)) diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 5406b193227c..aad45b7a927d 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -92,7 +92,7 @@ static void gfs2_pte_inval(struct gfs2_glock *gl) ip = gl->gl_object; inode = &ip->i_inode; - if (!ip || !S_ISREG(ip->i_di.di_mode)) + if (!ip || !S_ISREG(inode->i_mode)) return; if (!test_bit(GIF_PAGED, &ip->i_flags)) @@ -119,7 +119,7 @@ static void gfs2_page_inval(struct gfs2_glock *gl) ip = gl->gl_object; inode = &ip->i_inode; - if (!ip || !S_ISREG(ip->i_di.di_mode)) + if (!ip || !S_ISREG(inode->i_mode)) return; truncate_inode_pages(inode->i_mapping, 0); @@ -142,7 +142,7 @@ static void gfs2_page_wait(struct gfs2_glock *gl) struct address_space *mapping = inode->i_mapping; int error; - if (!S_ISREG(ip->i_di.di_mode)) + if (!S_ISREG(inode->i_mode)) return; error = filemap_fdatawait(mapping); @@ -164,7 +164,7 @@ static void gfs2_page_writeback(struct gfs2_glock *gl) struct inode *inode = &ip->i_inode; struct address_space *mapping = inode->i_mapping; - if (!S_ISREG(ip->i_di.di_mode)) + if (!S_ISREG(inode->i_mode)) return; filemap_fdatawrite(mapping); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index a99591956544..de466043c91a 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -51,7 +51,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) struct gfs2_dinode_host *di = &ip->i_di; inode->i_ino = ip->i_num.no_addr; - inode->i_mode = di->di_mode; inode->i_nlink = di->di_nlink; inode->i_uid = di->di_uid; inode->i_gid = di->di_gid; @@ -88,9 +87,6 @@ void gfs2_inode_attr_out(struct gfs2_inode *ip) { struct inode *inode = &ip->i_inode; struct gfs2_dinode_host *di = &ip->i_di; - gfs2_assert_withdraw(GFS2_SB(inode), - (di->di_mode & S_IFMT) == (inode->i_mode & S_IFMT)); - di->di_mode = inode->i_mode; di->di_uid = inode->i_uid; di->di_gid = inode->i_gid; di->di_atime = inode->i_atime.tv_sec; @@ -210,9 +206,9 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) if (ip->i_num.no_formal_ino != be64_to_cpu(str->di_num.no_formal_ino)) return -ESTALE; - di->di_mode = be32_to_cpu(str->di_mode); + ip->i_inode.i_mode = be32_to_cpu(str->di_mode); ip->i_inode.i_rdev = 0; - switch (di->di_mode & S_IFMT) { + switch (ip->i_inode.i_mode & S_IFMT) { case S_IFBLK: case S_IFCHR: ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), @@ -620,7 +616,7 @@ static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, unsigned int *uid, unsigned int *gid) { if (GFS2_SB(&dip->i_inode)->sd_args.ar_suiddir && - (dip->i_di.di_mode & S_ISUID) && dip->i_di.di_uid) { + (dip->i_inode.i_mode & S_ISUID) && dip->i_di.di_uid) { if (S_ISDIR(*mode)) *mode |= S_ISUID; else if (dip->i_di.di_uid != current->fsuid) @@ -629,7 +625,7 @@ static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, } else *uid = current->fsuid; - if (dip->i_di.di_mode & S_ISGID) { + if (dip->i_inode.i_mode & S_ISGID) { if (S_ISDIR(*mode)) *mode |= S_ISGID; *gid = dip->i_di.di_gid; @@ -810,7 +806,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, goto fail_quota_locks; } - error = gfs2_dir_add(&dip->i_inode, name, &ip->i_num, IF2DT(ip->i_di.di_mode)); + error = gfs2_dir_add(&dip->i_inode, name, &ip->i_num, IF2DT(ip->i_inode.i_mode)); if (error) goto fail_end_trans; @@ -1053,7 +1049,7 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) return -EPERM; - if ((dip->i_di.di_mode & S_ISVTX) && + if ((dip->i_inode.i_mode & S_ISVTX) && dip->i_di.di_uid != current->fsuid && ip->i_di.di_uid != current->fsuid && !capable(CAP_FOWNER)) return -EPERM; @@ -1072,7 +1068,7 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, if (!gfs2_inum_equal(&inum, &ip->i_num)) return -ENOENT; - if (IF2DT(ip->i_di.di_mode) != type) { + if (IF2DT(ip->i_inode.i_mode) != type) { gfs2_consist_inode(dip); return -EIO; } diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 33c9ea68f7e7..69cbf98509a9 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -22,7 +22,7 @@ static inline int gfs2_is_jdata(struct gfs2_inode *ip) static inline int gfs2_is_dir(struct gfs2_inode *ip) { - return S_ISDIR(ip->i_di.di_mode); + return S_ISDIR(ip->i_inode.i_mode); } void gfs2_inode_attr_in(struct gfs2_inode *ip); diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 60dd943761ab..6b50a5731a51 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -161,7 +161,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) gfs2_inum_out(&ip->i_num, &str->di_num); - str->di_mode = cpu_to_be32(di->di_mode); + str->di_mode = cpu_to_be32(ip->i_inode.i_mode); str->di_uid = cpu_to_be32(di->di_uid); str->di_gid = cpu_to_be32(di->di_gid); str->di_nlink = cpu_to_be32(di->di_nlink); @@ -191,7 +191,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) gfs2_inum_print(&ip->i_num); - pv(di, di_mode, "0%o"); pv(di, di_uid, "%u"); pv(di, di_gid, "%u"); pv(di, di_nlink, "%u"); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 015640b3f123..45a3d85b1d68 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -498,7 +498,6 @@ static int gfs2_commit_write(struct file *file, struct page *page, di->di_size = cpu_to_be64(inode->i_size); } - di->di_mode = cpu_to_be32(inode->i_mode); di->di_atime = cpu_to_be64(inode->i_atime.tv_sec); di->di_mtime = cpu_to_be64(inode->i_mtime.tv_sec); di->di_ctime = cpu_to_be64(inode->i_ctime.tv_sec); diff --git a/fs/gfs2/ops_dentry.c b/fs/gfs2/ops_dentry.c index c36f9e342e66..d355899585d8 100644 --- a/fs/gfs2/ops_dentry.c +++ b/fs/gfs2/ops_dentry.c @@ -76,7 +76,7 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd) if (!gfs2_inum_equal(&ip->i_num, &inum)) goto invalid_gunlock; - if (IF2DT(ip->i_di.di_mode) != type) { + if (IF2DT(ip->i_inode.i_mode) != type) { gfs2_consist_inode(dip); goto fail_gunlock; } diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 7ea417573903..b52b9db1a2bd 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -425,7 +425,7 @@ static int gfs2_open(struct inode *inode, struct file *file) gfs2_assert_warn(GFS2_SB(inode), !file->private_data); file->private_data = fp; - if (S_ISREG(ip->i_di.di_mode)) { + if (S_ISREG(ip->i_inode.i_mode)) { error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); if (error) @@ -515,7 +515,7 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_POSIX)) return -ENOLCK; - if ((ip->i_di.di_mode & (S_ISGID | S_IXGRP)) == S_ISGID) + if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID) return -ENOLCK; if (sdp->sd_args.ar_localflocks) { @@ -617,7 +617,7 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_FLOCK)) return -ENOLCK; - if ((ip->i_di.di_mode & (S_ISGID | S_IXGRP)) == S_ISGID) + if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID) return -ENOLCK; if (sdp->sd_args.ar_localflocks) diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index c10b914bf8cc..cf7a5bae3957 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -144,7 +144,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, int alloc_required; int error; - if (S_ISDIR(ip->i_di.di_mode)) + if (S_ISDIR(inode->i_mode)) return -EPERM; gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); @@ -220,7 +220,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, } error = gfs2_dir_add(dir, &dentry->d_name, &ip->i_num, - IF2DT(ip->i_di.di_mode)); + IF2DT(inode->i_mode)); if (error) goto out_end_trans; @@ -564,11 +564,10 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Make sure we aren't trying to move a dirctory into it's subdir */ - if (S_ISDIR(ip->i_di.di_mode) && odip != ndip) { + if (S_ISDIR(ip->i_inode.i_mode) && odip != ndip) { dir_rename = 1; - error = gfs2_glock_nq_init(sdp->sd_rename_gl, - LM_ST_EXCLUSIVE, 0, + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, 0, &r_gh); if (error) goto out; @@ -609,7 +608,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) goto out_gunlock; - if (S_ISDIR(nip->i_di.di_mode)) { + if (S_ISDIR(nip->i_inode.i_mode)) { if (nip->i_di.di_entries < 2) { if (gfs2_consist_inode(nip)) gfs2_dinode_print(nip); @@ -646,7 +645,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = -EFBIG; goto out_gunlock; } - if (S_ISDIR(ip->i_di.di_mode) && + if (S_ISDIR(ip->i_inode.i_mode) && ndip->i_di.di_nlink == (u32)-1) { error = -EMLINK; goto out_gunlock; @@ -701,7 +700,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Remove the target file, if it exists */ if (nip) { - if (S_ISDIR(nip->i_di.di_mode)) + if (S_ISDIR(nip->i_inode.i_mode)) error = gfs2_rmdiri(ndip, &ndentry->d_name, nip); else { error = gfs2_dir_del(ndip, &ndentry->d_name); @@ -743,7 +742,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, goto out_end_trans; error = gfs2_dir_add(ndir, &ndentry->d_name, &ip->i_num, - IF2DT(ip->i_di.di_mode)); + IF2DT(ip->i_inode.i_mode)); if (error) goto out_end_trans; diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index 9c786d1e702d..863517569223 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -407,7 +407,7 @@ static void gfs2_delete_inode(struct inode *inode) if (error) goto out_uninit; - if (S_ISDIR(ip->i_di.di_mode) && + if (S_ISDIR(inode->i_mode) && (ip->i_di.di_flags & GFS2_DIF_EXHASH)) { error = gfs2_dir_exhash_dealloc(ip); if (error) diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 5bcf895b2731..f1ea0b48060b 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -322,7 +322,6 @@ struct gfs2_dinode { }; struct gfs2_dinode_host { - __u32 di_mode; /* mode of file */ __u32 di_uid; /* owner's user id */ __u32 di_gid; /* owner's group id */ __u32 di_nlink; /* number of links to this file */ -- cgit v1.2.3 From 2933f9254a6af33db25270778c998a42029da668 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 13:23:29 -0500 Subject: [GFS2] Shrink gfs2_inode (4) - di_uid/di_gid Remove duplicate di_uid/di_gid fields in favour of using inode->i_uid/inode->i_gid instead. This saves 8 bytes. Signed-off-by: Steven Whitehouse --- fs/gfs2/acl.c | 2 +- fs/gfs2/bmap.c | 2 +- fs/gfs2/eattr.c | 2 +- fs/gfs2/inode.c | 23 +++++++++-------------- fs/gfs2/ondisk.c | 6 ++---- fs/gfs2/ops_address.c | 2 +- fs/gfs2/ops_inode.c | 10 ++++------ fs/gfs2/ops_vm.c | 2 +- fs/gfs2/quota.c | 8 ++++---- fs/gfs2/rgrp.c | 11 +++++------ include/linux/gfs2_ondisk.h | 2 -- 11 files changed, 29 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 87f6304f2f54..3908992b27a2 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -74,7 +74,7 @@ int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access) { if (!GFS2_SB(&ip->i_inode)->sd_args.ar_posix_acl) return -EOPNOTSUPP; - if (current->fsuid != ip->i_di.di_uid && !capable(CAP_FOWNER)) + if (current->fsuid != ip->i_inode.i_uid && !capable(CAP_FOWNER)) return -EPERM; if (S_ISLNK(ip->i_inode.i_mode)) return -EOPNOTSUPP; diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 481a06882544..0c913eecf884 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -819,7 +819,7 @@ static int do_grow(struct gfs2_inode *ip, u64 size) if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index 5208fa94aad2..935cc9a57168 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -687,7 +687,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index de466043c91a..0de9b22f454b 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -52,8 +52,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) inode->i_ino = ip->i_num.no_addr; inode->i_nlink = di->di_nlink; - inode->i_uid = di->di_uid; - inode->i_gid = di->di_gid; i_size_write(inode, di->di_size); inode->i_atime.tv_sec = di->di_atime; inode->i_mtime.tv_sec = di->di_mtime; @@ -87,8 +85,6 @@ void gfs2_inode_attr_out(struct gfs2_inode *ip) { struct inode *inode = &ip->i_inode; struct gfs2_dinode_host *di = &ip->i_di; - di->di_uid = inode->i_uid; - di->di_gid = inode->i_gid; di->di_atime = inode->i_atime.tv_sec; di->di_mtime = inode->i_mtime.tv_sec; di->di_ctime = inode->i_ctime.tv_sec; @@ -216,8 +212,8 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) break; }; - di->di_uid = be32_to_cpu(str->di_uid); - di->di_gid = be32_to_cpu(str->di_gid); + ip->i_inode.i_uid = be32_to_cpu(str->di_uid); + ip->i_inode.i_gid = be32_to_cpu(str->di_gid); di->di_nlink = be32_to_cpu(str->di_nlink); di->di_size = be64_to_cpu(str->di_size); di->di_blocks = be64_to_cpu(str->di_blocks); @@ -616,19 +612,19 @@ static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, unsigned int *uid, unsigned int *gid) { if (GFS2_SB(&dip->i_inode)->sd_args.ar_suiddir && - (dip->i_inode.i_mode & S_ISUID) && dip->i_di.di_uid) { + (dip->i_inode.i_mode & S_ISUID) && dip->i_inode.i_uid) { if (S_ISDIR(*mode)) *mode |= S_ISUID; - else if (dip->i_di.di_uid != current->fsuid) + else if (dip->i_inode.i_uid != current->fsuid) *mode &= ~07111; - *uid = dip->i_di.di_uid; + *uid = dip->i_inode.i_uid; } else *uid = current->fsuid; if (dip->i_inode.i_mode & S_ISGID) { if (S_ISDIR(*mode)) *mode |= S_ISGID; - *gid = dip->i_di.di_gid; + *gid = dip->i_inode.i_gid; } else *gid = current->fsgid; } @@ -783,8 +779,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, if (alloc_required < 0) goto fail; if (alloc_required) { - error = gfs2_quota_check(dip, dip->i_di.di_uid, - dip->i_di.di_gid); + error = gfs2_quota_check(dip, dip->i_inode.i_uid, dip->i_inode.i_gid); if (error) goto fail_quota_locks; @@ -1050,8 +1045,8 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, return -EPERM; if ((dip->i_inode.i_mode & S_ISVTX) && - dip->i_di.di_uid != current->fsuid && - ip->i_di.di_uid != current->fsuid && !capable(CAP_FOWNER)) + dip->i_inode.i_uid != current->fsuid && + ip->i_inode.i_uid != current->fsuid && !capable(CAP_FOWNER)) return -EPERM; if (IS_APPEND(&dip->i_inode)) diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 6b50a5731a51..e224f6a22641 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -162,8 +162,8 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) gfs2_inum_out(&ip->i_num, &str->di_num); str->di_mode = cpu_to_be32(ip->i_inode.i_mode); - str->di_uid = cpu_to_be32(di->di_uid); - str->di_gid = cpu_to_be32(di->di_gid); + str->di_uid = cpu_to_be32(ip->i_inode.i_uid); + str->di_gid = cpu_to_be32(ip->i_inode.i_gid); str->di_nlink = cpu_to_be32(di->di_nlink); str->di_size = cpu_to_be64(di->di_size); str->di_blocks = cpu_to_be64(di->di_blocks); @@ -191,8 +191,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) gfs2_inum_print(&ip->i_num); - pv(di, di_uid, "%u"); - pv(di, di_gid, "%u"); pv(di, di_nlink, "%u"); printk(KERN_INFO " di_size = %llu\n", (unsigned long long)di->di_size); printk(KERN_INFO " di_blocks = %llu\n", (unsigned long long)di->di_blocks); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 45a3d85b1d68..38b702a18244 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -386,7 +386,7 @@ static int gfs2_prepare_write(struct file *file, struct page *page, if (error) goto out_alloc_put; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_qunlock; diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index cf7a5bae3957..efbcec3311bb 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -196,8 +196,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (error) goto out_alloc; - error = gfs2_quota_check(dip, dip->i_di.di_uid, - dip->i_di.di_gid); + error = gfs2_quota_check(dip, dip->i_inode.i_uid, dip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -673,8 +672,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) goto out_alloc; - error = gfs2_quota_check(ndip, ndip->i_di.di_uid, - ndip->i_di.di_gid); + error = gfs2_quota_check(ndip, ndip->i_inode.i_uid, ndip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -885,8 +883,8 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) u32 ouid, ogid, nuid, ngid; int error; - ouid = ip->i_di.di_uid; - ogid = ip->i_di.di_gid; + ouid = inode->i_uid; + ogid = inode->i_gid; nuid = attr->ia_uid; ngid = attr->ia_gid; diff --git a/fs/gfs2/ops_vm.c b/fs/gfs2/ops_vm.c index 5453d2947ab3..45a5f11fc39a 100644 --- a/fs/gfs2/ops_vm.c +++ b/fs/gfs2/ops_vm.c @@ -76,7 +76,7 @@ static int alloc_page_backing(struct gfs2_inode *ip, struct page *page) if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 5d00e9b20973..d0db881b55d2 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -452,19 +452,19 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF) return 0; - error = qdsb_get(sdp, QUOTA_USER, ip->i_di.di_uid, CREATE, qd); + error = qdsb_get(sdp, QUOTA_USER, ip->i_inode.i_uid, CREATE, qd); if (error) goto out; al->al_qd_num++; qd++; - error = qdsb_get(sdp, QUOTA_GROUP, ip->i_di.di_gid, CREATE, qd); + error = qdsb_get(sdp, QUOTA_GROUP, ip->i_inode.i_gid, CREATE, qd); if (error) goto out; al->al_qd_num++; qd++; - if (uid != NO_QUOTA_CHANGE && uid != ip->i_di.di_uid) { + if (uid != NO_QUOTA_CHANGE && uid != ip->i_inode.i_uid) { error = qdsb_get(sdp, QUOTA_USER, uid, CREATE, qd); if (error) goto out; @@ -472,7 +472,7 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) qd++; } - if (gid != NO_QUOTA_CHANGE && gid != ip->i_di.di_gid) { + if (gid != NO_QUOTA_CHANGE && gid != ip->i_inode.i_gid) { error = qdsb_get(sdp, QUOTA_GROUP, gid, CREATE, qd); if (error) goto out; diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 07dfd6305058..ff0846528d54 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -1217,7 +1217,7 @@ u64 gfs2_alloc_data(struct gfs2_inode *ip) al->al_alloced++; gfs2_statfs_change(sdp, 0, -1, 0); - gfs2_quota_change(ip, +1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid); spin_lock(&sdp->sd_rindex_spin); rgd->rd_free_clone--; @@ -1261,7 +1261,7 @@ u64 gfs2_alloc_meta(struct gfs2_inode *ip) al->al_alloced++; gfs2_statfs_change(sdp, 0, -1, 0); - gfs2_quota_change(ip, +1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_trans_add_unrevoke(sdp, block); spin_lock(&sdp->sd_rindex_spin); @@ -1337,8 +1337,7 @@ void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen) gfs2_trans_add_rg(rgd); gfs2_statfs_change(sdp, 0, +blen, 0); - gfs2_quota_change(ip, -(s64)blen, - ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid); } /** @@ -1366,7 +1365,7 @@ void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen) gfs2_trans_add_rg(rgd); gfs2_statfs_change(sdp, 0, +blen, 0); - gfs2_quota_change(ip, -(s64)blen, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_meta_wipe(ip, bstart, blen); } @@ -1411,7 +1410,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) { gfs2_free_uninit_di(rgd, ip->i_num.no_addr); - gfs2_quota_change(ip, -1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_meta_wipe(ip, ip->i_num.no_addr, 1); } diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index f1ea0b48060b..896c7f81a637 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -322,8 +322,6 @@ struct gfs2_dinode { }; struct gfs2_dinode_host { - __u32 di_uid; /* owner's user id */ - __u32 di_gid; /* owner's group id */ __u32 di_nlink; /* number of links to this file */ __u64 di_size; /* number of bytes in file */ __u64 di_blocks; /* number of blocks in file */ -- cgit v1.2.3 From 4f56110a00af5fb2e22fbccfcaf944d62cae8fcf Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 14:04:17 -0500 Subject: [GFS2] Shrink gfs2_inode (5) - di_nlink Remove the di_nlink field in favour of inode->i_nlink and update the nlink handling to use the proper macros. This saves 4 bytes. Signed-off-by: Steven Whitehouse --- fs/gfs2/inode.c | 37 ++++++++++++++++++++++++------------- fs/gfs2/ondisk.c | 3 +-- fs/gfs2/ops_inode.c | 12 ++++++------ include/linux/gfs2_ondisk.h | 1 - 4 files changed, 31 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 0de9b22f454b..711203984823 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -51,7 +51,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) struct gfs2_dinode_host *di = &ip->i_di; inode->i_ino = ip->i_num.no_addr; - inode->i_nlink = di->di_nlink; i_size_write(inode, di->di_size); inode->i_atime.tv_sec = di->di_atime; inode->i_mtime.tv_sec = di->di_mtime; @@ -214,7 +213,12 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) ip->i_inode.i_uid = be32_to_cpu(str->di_uid); ip->i_inode.i_gid = be32_to_cpu(str->di_gid); - di->di_nlink = be32_to_cpu(str->di_nlink); + /* + * We will need to review setting the nlink count here in the + * light of the forthcoming ro bind mount work. This is a reminder + * to do that. + */ + ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink); di->di_size = be64_to_cpu(str->di_size); di->di_blocks = be64_to_cpu(str->di_blocks); di->di_atime = be64_to_cpu(str->di_atime); @@ -336,12 +340,12 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) u32 nlink; int error; - BUG_ON(ip->i_di.di_nlink != ip->i_inode.i_nlink); - nlink = ip->i_di.di_nlink + diff; + BUG_ON(diff != 1 && diff != -1); + nlink = ip->i_inode.i_nlink + diff; /* If we are reducing the nlink count, but the new value ends up being bigger than the old one, we must have underflowed. */ - if (diff < 0 && nlink > ip->i_di.di_nlink) { + if (diff < 0 && nlink > ip->i_inode.i_nlink) { if (gfs2_consist_inode(ip)) gfs2_dinode_print(ip); return -EIO; @@ -351,16 +355,19 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) if (error) return error; - ip->i_di.di_nlink = nlink; + if (diff > 0) + inc_nlink(&ip->i_inode); + else + drop_nlink(&ip->i_inode); + ip->i_di.di_ctime = get_seconds(); - ip->i_inode.i_nlink = nlink; gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); mark_inode_dirty(&ip->i_inode); - if (ip->i_di.di_nlink == 0) { + if (ip->i_inode.i_nlink == 0) { struct gfs2_rgrpd *rgd; struct gfs2_holder ri_gh, rg_gh; @@ -375,7 +382,6 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) if (error) goto out_norgrp; - clear_nlink(&ip->i_inode); gfs2_unlink_di(&ip->i_inode); /* mark inode unlinked */ gfs2_glock_dq_uninit(&rg_gh); out_norgrp: @@ -586,7 +592,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, return error; /* Don't create entries in an unlinked directory */ - if (!dip->i_di.di_nlink) + if (!dip->i_inode.i_nlink) return -EPERM; error = gfs2_dir_search(&dip->i_inode, name, NULL, NULL); @@ -602,7 +608,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, if (dip->i_di.di_entries == (u32)-1) return -EFBIG; - if (S_ISDIR(mode) && dip->i_di.di_nlink == (u32)-1) + if (S_ISDIR(mode) && dip->i_inode.i_nlink == (u32)-1) return -EMLINK; return 0; @@ -808,7 +814,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto fail_end_trans; - ip->i_di.di_nlink = 1; + ip->i_inode.i_nlink = 1; gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -1016,7 +1022,12 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, if (error) return error; - error = gfs2_change_nlink(ip, -2); + /* It looks odd, but it really should be done twice */ + error = gfs2_change_nlink(ip, -1); + if (error) + return error; + + error = gfs2_change_nlink(ip, -1); if (error) return error; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index e224f6a22641..b4e354b18815 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -164,7 +164,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) str->di_mode = cpu_to_be32(ip->i_inode.i_mode); str->di_uid = cpu_to_be32(ip->i_inode.i_uid); str->di_gid = cpu_to_be32(ip->i_inode.i_gid); - str->di_nlink = cpu_to_be32(di->di_nlink); + str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); str->di_size = cpu_to_be64(di->di_size); str->di_blocks = cpu_to_be64(di->di_blocks); str->di_atime = cpu_to_be64(di->di_atime); @@ -191,7 +191,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) gfs2_inum_print(&ip->i_num); - pv(di, di_nlink, "%u"); printk(KERN_INFO " di_size = %llu\n", (unsigned long long)di->di_size); printk(KERN_INFO " di_blocks = %llu\n", (unsigned long long)di->di_blocks); printk(KERN_INFO " di_atime = %lld\n", (long long)di->di_atime); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index efbcec3311bb..06176dee1550 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -169,7 +169,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, } error = -EINVAL; - if (!dip->i_di.di_nlink) + if (!dip->i_inode.i_nlink) goto out_gunlock; error = -EFBIG; if (dip->i_di.di_entries == (u32)-1) @@ -178,10 +178,10 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) goto out_gunlock; error = -EINVAL; - if (!ip->i_di.di_nlink) + if (!ip->i_inode.i_nlink) goto out_gunlock; error = -EMLINK; - if (ip->i_di.di_nlink == (u32)-1) + if (ip->i_inode.i_nlink == (u32)-1) goto out_gunlock; alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); @@ -386,7 +386,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) ip = ghs[1].gh_gl->gl_object; - ip->i_di.di_nlink = 2; + ip->i_inode.i_nlink = 2; ip->i_di.di_size = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode); ip->i_di.di_flags |= GFS2_DIF_JDATA; ip->i_di.di_payload_format = GFS2_FORMAT_DE; @@ -636,7 +636,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, }; if (odip != ndip) { - if (!ndip->i_di.di_nlink) { + if (!ndip->i_inode.i_nlink) { error = -EINVAL; goto out_gunlock; } @@ -645,7 +645,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, goto out_gunlock; } if (S_ISDIR(ip->i_inode.i_mode) && - ndip->i_di.di_nlink == (u32)-1) { + ndip->i_inode.i_nlink == (u32)-1) { error = -EMLINK; goto out_gunlock; } diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 896c7f81a637..c61517b35b2e 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -322,7 +322,6 @@ struct gfs2_dinode { }; struct gfs2_dinode_host { - __u32 di_nlink; /* number of links to this file */ __u64 di_size; /* number of bytes in file */ __u64 di_blocks; /* number of blocks in file */ __u64 di_atime; /* time last accessed */ -- cgit v1.2.3 From 1a7b1eed5802502fd649e04784becd58557fdcf1 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 14:35:17 -0500 Subject: [GFS2] Shrink gfs2_inode (6) - di_atime/di_mtime/di_ctime Remove the di_[amc]time fields and use inode->i_[amc]time fields instead. This saves 24 bytes from the gfs2_inode. Signed-off-by: Steven Whitehouse --- fs/gfs2/bmap.c | 10 +++++----- fs/gfs2/dir.c | 10 +++++----- fs/gfs2/eattr.c | 9 ++++----- fs/gfs2/inode.c | 44 +++++++++++--------------------------------- fs/gfs2/inode.h | 1 - fs/gfs2/ondisk.c | 10 +++------- fs/gfs2/ops_address.c | 4 ---- fs/gfs2/ops_inode.c | 3 +-- include/linux/gfs2_ondisk.h | 3 --- 9 files changed, 29 insertions(+), 65 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 0c913eecf884..692d4a3da1bc 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -778,7 +778,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, gfs2_free_data(ip, bstart, blen); } - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_dinode_out(ip, dibh->b_data); @@ -853,7 +853,7 @@ static int do_grow(struct gfs2_inode *ip, u64 size) } ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); error = gfs2_meta_inode_buffer(ip, &dibh); if (error) @@ -968,7 +968,7 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) if (gfs2_is_stuffed(ip)) { ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + size); @@ -980,7 +980,7 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) if (!error) { ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); ip->i_di.di_flags |= GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); @@ -1053,7 +1053,7 @@ static int trunc_end(struct gfs2_inode *ip) ip->i_num.no_addr; gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode)); } - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); ip->i_di.di_flags &= ~GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 0742761e1e02..ca23c8beb3f0 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -131,7 +131,7 @@ static int gfs2_dir_write_stuffed(struct gfs2_inode *ip, const char *buf, memcpy(dibh->b_data + offset + sizeof(struct gfs2_dinode), buf, size); if (ip->i_di.di_size < offset + size) ip->i_di.di_size = offset + size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -229,7 +229,7 @@ out: if (ip->i_di.di_size < offset + copied) ip->i_di.di_size = offset + copied; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); @@ -1560,7 +1560,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, break; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_entries++; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_dinode_out(ip, bh->b_data); brelse(bh); error = 0; @@ -1646,7 +1646,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) gfs2_consist_inode(dip); gfs2_trans_add_bh(dip->i_gl, bh, 1); dip->i_di.di_entries--; - dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); + dip->i_inode.i_mtime.tv_sec = dip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_dinode_out(dip, bh->b_data); brelse(bh); mark_inode_dirty(&dip->i_inode); @@ -1694,7 +1694,7 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, gfs2_trans_add_bh(dip->i_gl, bh, 1); } - dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); + dip->i_inode.i_mtime.tv_sec = dip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_dinode_out(dip, bh->b_data); brelse(bh); return 0; diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index 935cc9a57168..7dde84775ba7 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -300,7 +300,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -715,7 +715,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, (er->er_mode & S_IFMT)); ip->i_inode.i_mode = er->er_mode; } - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -850,7 +850,7 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh, (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT)); ip->i_inode.i_mode = er->er_mode; } - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -1130,7 +1130,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -1285,7 +1285,6 @@ int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el, if (!error) { error = inode_setattr(&ip->i_inode, attr); gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error); - gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 711203984823..c22ae3c3a444 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -52,12 +52,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) inode->i_ino = ip->i_num.no_addr; i_size_write(inode, di->di_size); - inode->i_atime.tv_sec = di->di_atime; - inode->i_mtime.tv_sec = di->di_mtime; - inode->i_ctime.tv_sec = di->di_ctime; - inode->i_atime.tv_nsec = 0; - inode->i_mtime.tv_nsec = 0; - inode->i_ctime.tv_nsec = 0; inode->i_blocks = di->di_blocks << (GFS2_SB(inode)->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT); @@ -72,23 +66,6 @@ void gfs2_inode_attr_in(struct gfs2_inode *ip) inode->i_flags &= ~S_APPEND; } -/** - * gfs2_inode_attr_out - Copy attributes from VFS inode into the dinode - * @ip: The GFS2 inode - * - * Only copy out the attributes that we want the VFS layer - * to be able to modify. - */ - -void gfs2_inode_attr_out(struct gfs2_inode *ip) -{ - struct inode *inode = &ip->i_inode; - struct gfs2_dinode_host *di = &ip->i_di; - di->di_atime = inode->i_atime.tv_sec; - di->di_mtime = inode->i_mtime.tv_sec; - di->di_ctime = inode->i_ctime.tv_sec; -} - static int iget_test(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); @@ -221,9 +198,12 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink); di->di_size = be64_to_cpu(str->di_size); di->di_blocks = be64_to_cpu(str->di_blocks); - di->di_atime = be64_to_cpu(str->di_atime); - di->di_mtime = be64_to_cpu(str->di_mtime); - di->di_ctime = be64_to_cpu(str->di_ctime); + ip->i_inode.i_atime.tv_sec = be64_to_cpu(str->di_atime); + ip->i_inode.i_atime.tv_nsec = 0; + ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); + ip->i_inode.i_mtime.tv_nsec = 0; + ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); + ip->i_inode.i_ctime.tv_nsec = 0; di->di_goal_meta = be64_to_cpu(str->di_goal_meta); di->di_goal_data = be64_to_cpu(str->di_goal_data); @@ -360,7 +340,7 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) else drop_nlink(&ip->i_inode); - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); @@ -1224,7 +1204,7 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) return 0; curtime = get_seconds(); - if (curtime - ip->i_di.di_atime >= quantum) { + if (curtime - ip->i_inode.i_atime.tv_sec >= quantum) { gfs2_glock_dq(gh); gfs2_holder_reinit(LM_ST_EXCLUSIVE, gh->gh_flags & ~LM_FLAG_ANY, gh); @@ -1236,7 +1216,7 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) trying to get exclusive lock. */ curtime = get_seconds(); - if (curtime - ip->i_di.di_atime >= quantum) { + if (curtime - ip->i_inode.i_atime.tv_sec >= quantum) { struct buffer_head *dibh; struct gfs2_dinode *di; @@ -1250,11 +1230,11 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) if (error) goto fail_end_trans; - ip->i_di.di_atime = curtime; + ip->i_inode.i_atime.tv_sec = curtime; gfs2_trans_add_bh(ip->i_gl, dibh, 1); di = (struct gfs2_dinode *)dibh->b_data; - di->di_atime = cpu_to_be64(ip->i_di.di_atime); + di->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); brelse(dibh); gfs2_trans_end(sdp); @@ -1375,8 +1355,6 @@ __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) if (!error) { error = inode_setattr(&ip->i_inode, attr); gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error); - gfs2_inode_attr_out(ip); - gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 69cbf98509a9..54d584eddf29 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -26,7 +26,6 @@ static inline int gfs2_is_dir(struct gfs2_inode *ip) } void gfs2_inode_attr_in(struct gfs2_inode *ip); -void gfs2_inode_attr_out(struct gfs2_inode *ip); struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum_host *inum, unsigned type); struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum_host *inum); diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index b4e354b18815..82003e872a37 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -167,9 +167,9 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); str->di_size = cpu_to_be64(di->di_size); str->di_blocks = cpu_to_be64(di->di_blocks); - str->di_atime = cpu_to_be64(di->di_atime); - str->di_mtime = cpu_to_be64(di->di_mtime); - str->di_ctime = cpu_to_be64(di->di_ctime); + str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); + str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); + str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); str->di_goal_meta = cpu_to_be64(di->di_goal_meta); str->di_goal_data = cpu_to_be64(di->di_goal_data); @@ -193,10 +193,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) printk(KERN_INFO " di_size = %llu\n", (unsigned long long)di->di_size); printk(KERN_INFO " di_blocks = %llu\n", (unsigned long long)di->di_blocks); - printk(KERN_INFO " di_atime = %lld\n", (long long)di->di_atime); - printk(KERN_INFO " di_mtime = %lld\n", (long long)di->di_mtime); - printk(KERN_INFO " di_ctime = %lld\n", (long long)di->di_ctime); - printk(KERN_INFO " di_goal_meta = %llu\n", (unsigned long long)di->di_goal_meta); printk(KERN_INFO " di_goal_data = %llu\n", (unsigned long long)di->di_goal_data); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 38b702a18244..5c3962c80e88 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -498,10 +498,6 @@ static int gfs2_commit_write(struct file *file, struct page *page, di->di_size = cpu_to_be64(inode->i_size); } - di->di_atime = cpu_to_be64(inode->i_atime.tv_sec); - di->di_mtime = cpu_to_be64(inode->i_mtime.tv_sec); - di->di_ctime = cpu_to_be64(inode->i_ctime.tv_sec); - brelse(dibh); gfs2_trans_end(sdp); if (al->al_requested) { diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 06176dee1550..585b43a94ace 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -729,7 +729,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto out_end_trans; - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -915,7 +915,6 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) error = inode_setattr(inode, attr); gfs2_assert_warn(sdp, !error); - gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index c61517b35b2e..7f5a4a16224b 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -324,9 +324,6 @@ struct gfs2_dinode { struct gfs2_dinode_host { __u64 di_size; /* number of bytes in file */ __u64 di_blocks; /* number of blocks in file */ - __u64 di_atime; /* time last accessed */ - __u64 di_mtime; /* time last modified */ - __u64 di_ctime; /* time last changed */ /* This section varies from gfs1. Padding added to align with * remainder of dinode -- cgit v1.2.3 From a9583c7983cbba9726bfe64ee46613d654fc9e26 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Wed, 1 Nov 2006 20:09:14 -0500 Subject: [GFS2] Shrink gfs2_inode (7) - di_payload_format This is almost never used. Its there for backward compatibility with GFS1. It doesn't need its own field since it can always be calculated from the inode mode & flags. This saves a bit more space in the gfs2_inode. Signed-off-by: Steven Whitehouse --- fs/gfs2/dir.c | 1 - fs/gfs2/inode.c | 3 +-- fs/gfs2/ondisk.c | 6 +++--- fs/gfs2/ops_inode.c | 1 - include/linux/gfs2_ondisk.h | 1 - 5 files changed, 4 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index ca23c8beb3f0..c82d7cb4a654 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -902,7 +902,6 @@ static int dir_make_exhash(struct inode *inode) dip->i_di.di_size = sdp->sd_sb.sb_bsize / 2; dip->i_di.di_blocks++; dip->i_di.di_flags |= GFS2_DIF_EXHASH; - dip->i_di.di_payload_format = 0; for (x = sdp->sd_hash_ptrs, y = -1; x; x >>= 1, y++) ; dip->i_di.di_depth = y; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index c22ae3c3a444..f6177fc68320 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -210,7 +210,6 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_generation = be64_to_cpu(str->di_generation); di->di_flags = be32_to_cpu(str->di_flags); - di->di_payload_format = be32_to_cpu(str->di_payload_format); di->di_height = be16_to_cpu(str->di_height); di->di_depth = be16_to_cpu(str->di_depth); @@ -699,7 +698,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, } di->__pad1 = 0; - di->di_payload_format = cpu_to_be32(0); + di->di_payload_format = cpu_to_be32(S_ISDIR(mode) ? GFS2_FORMAT_DE : 0); di->di_height = cpu_to_be32(0); di->__pad2 = 0; di->__pad3 = 0; diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 82003e872a37..b2baba5c50b7 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -176,9 +176,10 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) str->di_generation = cpu_to_be64(di->di_generation); str->di_flags = cpu_to_be32(di->di_flags); - str->di_payload_format = cpu_to_be32(di->di_payload_format); str->di_height = cpu_to_be16(di->di_height); - + str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && + !(ip->i_di.di_flags & GFS2_DIF_EXHASH) ? + GFS2_FORMAT_DE : 0); str->di_depth = cpu_to_be16(di->di_depth); str->di_entries = cpu_to_be32(di->di_entries); @@ -197,7 +198,6 @@ void gfs2_dinode_print(const struct gfs2_inode *ip) printk(KERN_INFO " di_goal_data = %llu\n", (unsigned long long)di->di_goal_data); pv(di, di_flags, "0x%.8X"); - pv(di, di_payload_format, "%u"); pv(di, di_height, "%u"); pv(di, di_depth, "%u"); diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 585b43a94ace..0e4eade47bf2 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -389,7 +389,6 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) ip->i_inode.i_nlink = 2; ip->i_di.di_size = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode); ip->i_di.di_flags |= GFS2_DIF_JDATA; - ip->i_di.di_payload_format = GFS2_FORMAT_DE; ip->i_di.di_entries = 2; error = gfs2_meta_inode_buffer(ip, &dibh); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 7f5a4a16224b..536575efb62a 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -333,7 +333,6 @@ struct gfs2_dinode_host { __u64 di_generation; /* generation number for NFS */ __u32 di_flags; /* GFS2_DIF_... */ - __u32 di_payload_format; /* GFS2_FORMAT_... */ __u16 di_height; /* height of metadata */ /* These only apply to directories */ -- cgit v1.2.3 From f6e58f01e8dc869803b9f73b2aa9d5bc3f32ca05 Mon Sep 17 00:00:00 2001 From: Steven Whitehouse Date: Tue, 7 Nov 2006 15:14:58 -0500 Subject: [GFS2] Don't copy meta_header for rgrp in and out The meta_header for an ondisk rgrp never changes, so there is no point copying it in and back out to disk. Also there is no reason to keep a copy for each rgrp in memory. The code already checks to ensure that the header is correct before it calls the routine to copy the data in, so that we don't even need to check whether its correct on disk in the functions in ondisk.c Signed-off-by: Steven Whitehouse --- fs/gfs2/ondisk.c | 11 ----------- include/linux/gfs2_ondisk.h | 2 -- 2 files changed, 13 deletions(-) (limited to 'include/linux') diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index b2baba5c50b7..f2495f1e21ad 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -65,15 +65,6 @@ static void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *bu mh->mh_format = be32_to_cpu(str->mh_format); } -static void gfs2_meta_header_out(const struct gfs2_meta_header_host *mh, void *buf) -{ - struct gfs2_meta_header *str = buf; - - str->mh_magic = cpu_to_be32(mh->mh_magic); - str->mh_type = cpu_to_be32(mh->mh_type); - str->mh_format = cpu_to_be32(mh->mh_format); -} - void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) { const struct gfs2_sb *str = buf; @@ -119,7 +110,6 @@ void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf) { const struct gfs2_rgrp *str = buf; - gfs2_meta_header_in(&rg->rg_header, buf); rg->rg_flags = be32_to_cpu(str->rg_flags); rg->rg_free = be32_to_cpu(str->rg_free); rg->rg_dinodes = be32_to_cpu(str->rg_dinodes); @@ -130,7 +120,6 @@ void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf) { struct gfs2_rgrp *str = buf; - gfs2_meta_header_out(&rg->rg_header, buf); str->rg_flags = cpu_to_be32(rg->rg_flags); str->rg_free = cpu_to_be32(rg->rg_free); str->rg_dinodes = cpu_to_be32(rg->rg_dinodes); diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h index 536575efb62a..8b7e4c1e32ae 100644 --- a/include/linux/gfs2_ondisk.h +++ b/include/linux/gfs2_ondisk.h @@ -213,8 +213,6 @@ struct gfs2_rgrp { }; struct gfs2_rgrp_host { - struct gfs2_meta_header_host rg_header; - __u32 rg_flags; __u32 rg_free; __u32 rg_dinodes; -- cgit v1.2.3 From be1c63411addba3ad750eb4fdfc50b97bc82825e Mon Sep 17 00:00:00 2001 From: Olaf Kirch Date: Fri, 1 Dec 2006 10:39:12 +0100 Subject: [PATCH] blktrace: add timestamp message This adds a new timestamp message to blktrace, giving the timeofday when we starting tracing. This helps user space correlate block trace events with eg an application strace. This requires a (compatible) update to blkparse. The changed blkparse is still able to process traces generated by older kernels, and older versions of blkparse should silently ignore the new records (because they have a pid of 0). Signed-off-by: Olaf Kirch Signed-off-by: Jens Axboe --- block/blktrace.c | 57 ++++++++++++++++++++++++++++++++++---------- include/linux/blktrace_api.h | 12 ++++++++++ 2 files changed, 57 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/block/blktrace.c b/block/blktrace.c index 135593c8e45b..562ca7cbf858 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -22,30 +22,61 @@ #include #include #include +#include #include static DEFINE_PER_CPU(unsigned long long, blk_trace_cpu_offset) = { 0, }; static unsigned int blktrace_seq __read_mostly = 1; +/* + * Send out a notify message. + */ +static inline unsigned int trace_note(struct blk_trace *bt, + pid_t pid, int action, + const void *data, size_t len) +{ + struct blk_io_trace *t; + int cpu = smp_processor_id(); + + t = relay_reserve(bt->rchan, sizeof(*t) + len); + if (t == NULL) + return 0; + + t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; + t->time = sched_clock() - per_cpu(blk_trace_cpu_offset, cpu); + t->device = bt->dev; + t->action = action; + t->pid = pid; + t->cpu = cpu; + t->pdu_len = len; + memcpy((void *) t + sizeof(*t), data, len); + return blktrace_seq; +} + /* * Send out a notify for this process, if we haven't done so since a trace * started */ static void trace_note_tsk(struct blk_trace *bt, struct task_struct *tsk) { - struct blk_io_trace *t; + tsk->btrace_seq = trace_note(bt, tsk->pid, + BLK_TN_PROCESS, + tsk->comm, sizeof(tsk->comm)); +} - t = relay_reserve(bt->rchan, sizeof(*t) + sizeof(tsk->comm)); - if (t) { - t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION; - t->device = bt->dev; - t->action = BLK_TC_ACT(BLK_TC_NOTIFY); - t->pid = tsk->pid; - t->cpu = smp_processor_id(); - t->pdu_len = sizeof(tsk->comm); - memcpy((void *) t + sizeof(*t), tsk->comm, t->pdu_len); - tsk->btrace_seq = blktrace_seq; - } +static void trace_note_time(struct blk_trace *bt) +{ + struct timespec now; + unsigned long flags; + u32 words[2]; + + getnstimeofday(&now); + words[0] = now.tv_sec; + words[1] = now.tv_nsec; + + local_irq_save(flags); + trace_note(bt, 0, BLK_TN_TIMESTAMP, words, sizeof(words)); + local_irq_restore(flags); } static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector, @@ -394,6 +425,8 @@ static int blk_trace_startstop(request_queue_t *q, int start) blktrace_seq++; smp_mb(); bt->trace_state = Blktrace_running; + + trace_note_time(bt); ret = 0; } } else { diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index b99a714fcac6..3680ff9a30ed 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -49,6 +49,15 @@ enum blktrace_act { __BLK_TA_REMAP, /* bio was remapped */ }; +/* + * Notify events. + */ +enum blktrace_notify { + __BLK_TN_PROCESS = 0, /* establish pid/name mapping */ + __BLK_TN_TIMESTAMP, /* include system clock */ +}; + + /* * Trace actions in full. Additionally, read or write is masked */ @@ -68,6 +77,9 @@ enum blktrace_act { #define BLK_TA_BOUNCE (__BLK_TA_BOUNCE) #define BLK_TA_REMAP (__BLK_TA_REMAP | BLK_TC_ACT(BLK_TC_QUEUE)) +#define BLK_TN_PROCESS (__BLK_TN_PROCESS | BLK_TC_ACT(BLK_TC_NOTIFY)) +#define BLK_TN_TIMESTAMP (__BLK_TN_TIMESTAMP | BLK_TC_ACT(BLK_TC_NOTIFY)) + #define BLK_IO_TRACE_MAGIC 0x65617400 #define BLK_IO_TRACE_VERSION 0x07 -- cgit v1.2.3 From 0e75f9063f5c55fb0b0b546a7c356f8ec186825e Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Fri, 1 Dec 2006 10:40:55 +0100 Subject: [PATCH] block: support larger block pc requests This patch modifies blk_rq_map/unmap_user() and the cdrom and scsi_ioctl.c users so that it supports requests larger than bio by chaining them together. Signed-off-by: Mike Christie Signed-off-by: Jens Axboe --- block/ll_rw_blk.c | 166 +++++++++++++++++++++++++++++++++++++------------ block/scsi_ioctl.c | 53 ++++++++-------- drivers/cdrom/cdrom.c | 6 +- fs/bio.c | 18 +----- include/linux/blkdev.h | 7 ++- 5 files changed, 160 insertions(+), 90 deletions(-) (limited to 'include/linux') diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 9eaee6640535..0f82e12f7b67 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -2322,6 +2322,84 @@ void blk_insert_request(request_queue_t *q, struct request *rq, EXPORT_SYMBOL(blk_insert_request); +static int __blk_rq_unmap_user(struct bio *bio) +{ + int ret = 0; + + if (bio) { + if (bio_flagged(bio, BIO_USER_MAPPED)) + bio_unmap_user(bio); + else + ret = bio_uncopy_user(bio); + } + + return ret; +} + +static int __blk_rq_map_user(request_queue_t *q, struct request *rq, + void __user *ubuf, unsigned int len) +{ + unsigned long uaddr; + struct bio *bio, *orig_bio; + int reading, ret; + + reading = rq_data_dir(rq) == READ; + + /* + * if alignment requirement is satisfied, map in user pages for + * direct dma. else, set up kernel bounce buffers + */ + uaddr = (unsigned long) ubuf; + if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) + bio = bio_map_user(q, NULL, uaddr, len, reading); + else + bio = bio_copy_user(q, uaddr, len, reading); + + if (IS_ERR(bio)) { + return PTR_ERR(bio); + } + + orig_bio = bio; + blk_queue_bounce(q, &bio); + /* + * We link the bounce buffer in and could have to traverse it + * later so we have to get a ref to prevent it from being freed + */ + bio_get(bio); + + /* + * for most (all? don't know of any) queues we could + * skip grabbing the queue lock here. only drivers with + * funky private ->back_merge_fn() function could be + * problematic. + */ + spin_lock_irq(q->queue_lock); + if (!rq->bio) + blk_rq_bio_prep(q, rq, bio); + else if (!q->back_merge_fn(q, rq, bio)) { + ret = -EINVAL; + spin_unlock_irq(q->queue_lock); + goto unmap_bio; + } else { + rq->biotail->bi_next = bio; + rq->biotail = bio; + + rq->nr_sectors += bio_sectors(bio); + rq->hard_nr_sectors = rq->nr_sectors; + rq->data_len += bio->bi_size; + } + spin_unlock_irq(q->queue_lock); + + return bio->bi_size; + +unmap_bio: + /* if it was boucned we must call the end io function */ + bio_endio(bio, bio->bi_size, 0); + __blk_rq_unmap_user(orig_bio); + bio_put(bio); + return ret; +} + /** * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage * @q: request queue where request should be inserted @@ -2343,42 +2421,44 @@ EXPORT_SYMBOL(blk_insert_request); * unmapping. */ int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf, - unsigned int len) + unsigned long len) { - unsigned long uaddr; - struct bio *bio; - int reading; + unsigned long bytes_read = 0; + int ret; if (len > (q->max_hw_sectors << 9)) return -EINVAL; if (!len || !ubuf) return -EINVAL; - reading = rq_data_dir(rq) == READ; + while (bytes_read != len) { + unsigned long map_len, end, start; - /* - * if alignment requirement is satisfied, map in user pages for - * direct dma. else, set up kernel bounce buffers - */ - uaddr = (unsigned long) ubuf; - if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q))) - bio = bio_map_user(q, NULL, uaddr, len, reading); - else - bio = bio_copy_user(q, uaddr, len, reading); + map_len = min_t(unsigned long, len - bytes_read, BIO_MAX_SIZE); + end = ((unsigned long)ubuf + map_len + PAGE_SIZE - 1) + >> PAGE_SHIFT; + start = (unsigned long)ubuf >> PAGE_SHIFT; - if (!IS_ERR(bio)) { - rq->bio = rq->biotail = bio; - blk_rq_bio_prep(q, rq, bio); + /* + * A bad offset could cause us to require BIO_MAX_PAGES + 1 + * pages. If this happens we just lower the requested + * mapping len by a page so that we can fit + */ + if (end - start > BIO_MAX_PAGES) + map_len -= PAGE_SIZE; - rq->buffer = rq->data = NULL; - rq->data_len = len; - return 0; + ret = __blk_rq_map_user(q, rq, ubuf, map_len); + if (ret < 0) + goto unmap_rq; + bytes_read += ret; + ubuf += ret; } - /* - * bio is the err-ptr - */ - return PTR_ERR(bio); + rq->buffer = rq->data = NULL; + return 0; +unmap_rq: + blk_rq_unmap_user(rq); + return ret; } EXPORT_SYMBOL(blk_rq_map_user); @@ -2404,7 +2484,7 @@ EXPORT_SYMBOL(blk_rq_map_user); * unmapping. */ int blk_rq_map_user_iov(request_queue_t *q, struct request *rq, - struct sg_iovec *iov, int iov_count) + struct sg_iovec *iov, int iov_count, unsigned int len) { struct bio *bio; @@ -2418,10 +2498,15 @@ int blk_rq_map_user_iov(request_queue_t *q, struct request *rq, if (IS_ERR(bio)) return PTR_ERR(bio); - rq->bio = rq->biotail = bio; + if (bio->bi_size != len) { + bio_endio(bio, bio->bi_size, 0); + bio_unmap_user(bio); + return -EINVAL; + } + + bio_get(bio); blk_rq_bio_prep(q, rq, bio); rq->buffer = rq->data = NULL; - rq->data_len = bio->bi_size; return 0; } @@ -2429,23 +2514,26 @@ EXPORT_SYMBOL(blk_rq_map_user_iov); /** * blk_rq_unmap_user - unmap a request with user data - * @bio: bio to be unmapped - * @ulen: length of user buffer + * @rq: rq to be unmapped * * Description: - * Unmap a bio previously mapped by blk_rq_map_user(). + * Unmap a rq previously mapped by blk_rq_map_user(). + * rq->bio must be set to the original head of the request. */ -int blk_rq_unmap_user(struct bio *bio, unsigned int ulen) +int blk_rq_unmap_user(struct request *rq) { - int ret = 0; + struct bio *bio, *mapped_bio; - if (bio) { - if (bio_flagged(bio, BIO_USER_MAPPED)) - bio_unmap_user(bio); + while ((bio = rq->bio)) { + if (bio_flagged(bio, BIO_BOUNCED)) + mapped_bio = bio->bi_private; else - ret = bio_uncopy_user(bio); - } + mapped_bio = bio; + __blk_rq_unmap_user(mapped_bio); + rq->bio = bio->bi_next; + bio_put(bio); + } return 0; } @@ -2476,11 +2564,8 @@ int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf, if (rq_data_dir(rq) == WRITE) bio->bi_rw |= (1 << BIO_RW); - rq->bio = rq->biotail = bio; blk_rq_bio_prep(q, rq, bio); - rq->buffer = rq->data = NULL; - rq->data_len = len; return 0; } @@ -3495,6 +3580,7 @@ void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *bio) rq->hard_cur_sectors = rq->current_nr_sectors; rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio); rq->buffer = bio_data(bio); + rq->data_len = bio->bi_size; rq->bio = rq->biotail = bio; } diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index e55a75621437..5493c2fbbab1 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -226,7 +226,6 @@ static int sg_io(struct file *file, request_queue_t *q, unsigned long start_time; int writing = 0, ret = 0; struct request *rq; - struct bio *bio; char sense[SCSI_SENSE_BUFFERSIZE]; unsigned char cmd[BLK_MAX_CDB]; @@ -258,30 +257,6 @@ static int sg_io(struct file *file, request_queue_t *q, if (!rq) return -ENOMEM; - if (hdr->iovec_count) { - const int size = sizeof(struct sg_iovec) * hdr->iovec_count; - struct sg_iovec *iov; - - iov = kmalloc(size, GFP_KERNEL); - if (!iov) { - ret = -ENOMEM; - goto out; - } - - if (copy_from_user(iov, hdr->dxferp, size)) { - kfree(iov); - ret = -EFAULT; - goto out; - } - - ret = blk_rq_map_user_iov(q, rq, iov, hdr->iovec_count); - kfree(iov); - } else if (hdr->dxfer_len) - ret = blk_rq_map_user(q, rq, hdr->dxferp, hdr->dxfer_len); - - if (ret) - goto out; - /* * fill in request structure */ @@ -294,7 +269,6 @@ static int sg_io(struct file *file, request_queue_t *q, rq->sense_len = 0; rq->cmd_type = REQ_TYPE_BLOCK_PC; - bio = rq->bio; /* * bounce this after holding a reference to the original bio, it's @@ -309,6 +283,31 @@ static int sg_io(struct file *file, request_queue_t *q, if (!rq->timeout) rq->timeout = BLK_DEFAULT_TIMEOUT; + if (hdr->iovec_count) { + const int size = sizeof(struct sg_iovec) * hdr->iovec_count; + struct sg_iovec *iov; + + iov = kmalloc(size, GFP_KERNEL); + if (!iov) { + ret = -ENOMEM; + goto out; + } + + if (copy_from_user(iov, hdr->dxferp, size)) { + kfree(iov); + ret = -EFAULT; + goto out; + } + + ret = blk_rq_map_user_iov(q, rq, iov, hdr->iovec_count, + hdr->dxfer_len); + kfree(iov); + } else if (hdr->dxfer_len) + ret = blk_rq_map_user(q, rq, hdr->dxferp, hdr->dxfer_len); + + if (ret) + goto out; + rq->retries = 0; start_time = jiffies; @@ -339,7 +338,7 @@ static int sg_io(struct file *file, request_queue_t *q, hdr->sb_len_wr = len; } - if (blk_rq_unmap_user(bio, hdr->dxfer_len)) + if (blk_rq_unmap_user(rq)) ret = -EFAULT; /* may not have succeeded, but output values written to control diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 7ea0f48f8fa6..2df5cf4ec743 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -2133,16 +2133,14 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf, rq->timeout = 60 * HZ; bio = rq->bio; - if (rq->bio) - blk_queue_bounce(q, &rq->bio); - if (blk_execute_rq(q, cdi->disk, rq, 0)) { struct request_sense *s = rq->sense; ret = -EIO; cdi->last_sense = s->sense_key; } - if (blk_rq_unmap_user(bio, len)) + rq->bio = bio; + if (blk_rq_unmap_user(rq)) ret = -EFAULT; if (ret) diff --git a/fs/bio.c b/fs/bio.c index d91cfbf7ebc4..aa4d09bd4e71 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -560,10 +560,8 @@ struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr, break; } - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { - ret = -EINVAL; + if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) break; - } len -= bytes; } @@ -750,7 +748,6 @@ struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev, int write_to_vm) { struct bio *bio; - int len = 0, i; bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); @@ -765,18 +762,7 @@ struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev, */ bio_get(bio); - for (i = 0; i < iov_count; i++) - len += iov[i].iov_len; - - if (bio->bi_size == len) - return bio; - - /* - * don't support partial mappings - */ - bio_endio(bio, bio->bi_size, 0); - bio_unmap_user(bio); - return ERR_PTR(-EINVAL); + return bio; } static void __bio_unmap_user(struct bio *bio) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 7bfcde2d5578..e1c7286165ff 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -678,10 +678,11 @@ extern void __blk_stop_queue(request_queue_t *q); extern void blk_run_queue(request_queue_t *); extern void blk_start_queueing(request_queue_t *); extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *); -extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int); -extern int blk_rq_unmap_user(struct bio *, unsigned int); +extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned long); +extern int blk_rq_unmap_user(struct request *); extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, gfp_t); -extern int blk_rq_map_user_iov(request_queue_t *, struct request *, struct sg_iovec *, int); +extern int blk_rq_map_user_iov(request_queue_t *, struct request *, + struct sg_iovec *, int, unsigned int); extern int blk_execute_rq(request_queue_t *, struct gendisk *, struct request *, int); extern void blk_execute_rq_nowait(request_queue_t *, struct gendisk *, -- cgit v1.2.3 From bb37b94c68e7b37eecea8576039ae9396ca07839 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Fri, 1 Dec 2006 10:42:33 +0100 Subject: [BLOCK] Cleanup unused variable passing - ->init_queue() does not need the elevator passed in - ->put_request() is a hot path and need not have the queue passed in - cfq_update_io_seektime() does not need cfqd passed in Signed-off-by: Jens Axboe --- block/as-iosched.c | 2 +- block/cfq-iosched.c | 9 ++++----- block/deadline-iosched.c | 2 +- block/elevator.c | 4 ++-- block/noop-iosched.c | 2 +- include/linux/elevator.h | 4 ++-- 6 files changed, 11 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/block/as-iosched.c b/block/as-iosched.c index 50b95e4c1425..00242111a457 100644 --- a/block/as-iosched.c +++ b/block/as-iosched.c @@ -1317,7 +1317,7 @@ static void as_exit_queue(elevator_t *e) /* * initialize elevator private data (as_data). */ -static void *as_init_queue(request_queue_t *q, elevator_t *e) +static void *as_init_queue(request_queue_t *q) { struct as_data *ad; diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 1d9c3c70a9a0..e9019ed39b73 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1464,8 +1464,7 @@ cfq_update_io_thinktime(struct cfq_data *cfqd, struct cfq_io_context *cic) } static void -cfq_update_io_seektime(struct cfq_data *cfqd, struct cfq_io_context *cic, - struct request *rq) +cfq_update_io_seektime(struct cfq_io_context *cic, struct request *rq) { sector_t sdist; u64 total; @@ -1617,7 +1616,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq, } cfq_update_io_thinktime(cfqd, cic); - cfq_update_io_seektime(cfqd, cic, rq); + cfq_update_io_seektime(cic, rq); cfq_update_idle_window(cfqd, cfqq, cic); cic->last_queue = jiffies; @@ -1770,7 +1769,7 @@ static int cfq_may_queue(request_queue_t *q, int rw) /* * queue lock held here */ -static void cfq_put_request(request_queue_t *q, struct request *rq) +static void cfq_put_request(struct request *rq) { struct cfq_queue *cfqq = RQ_CFQQ(rq); @@ -1951,7 +1950,7 @@ static void cfq_exit_queue(elevator_t *e) kfree(cfqd); } -static void *cfq_init_queue(request_queue_t *q, elevator_t *e) +static void *cfq_init_queue(request_queue_t *q) { struct cfq_data *cfqd; int i; diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c index b7c5b34cb7b4..6d673e938d3e 100644 --- a/block/deadline-iosched.c +++ b/block/deadline-iosched.c @@ -356,7 +356,7 @@ static void deadline_exit_queue(elevator_t *e) /* * initialize elevator private data (deadline_data). */ -static void *deadline_init_queue(request_queue_t *q, elevator_t *e) +static void *deadline_init_queue(request_queue_t *q) { struct deadline_data *dd; diff --git a/block/elevator.c b/block/elevator.c index 8ccd163254b8..c0063f345c5d 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -129,7 +129,7 @@ static struct elevator_type *elevator_get(const char *name) static void *elevator_init_queue(request_queue_t *q, struct elevator_queue *eq) { - return eq->ops->elevator_init_fn(q, eq); + return eq->ops->elevator_init_fn(q); } static void elevator_attach(request_queue_t *q, struct elevator_queue *eq, @@ -810,7 +810,7 @@ void elv_put_request(request_queue_t *q, struct request *rq) elevator_t *e = q->elevator; if (e->ops->elevator_put_req_fn) - e->ops->elevator_put_req_fn(q, rq); + e->ops->elevator_put_req_fn(rq); } int elv_may_queue(request_queue_t *q, int rw) diff --git a/block/noop-iosched.c b/block/noop-iosched.c index 79af43179421..1c3de2b9a6b5 100644 --- a/block/noop-iosched.c +++ b/block/noop-iosched.c @@ -65,7 +65,7 @@ noop_latter_request(request_queue_t *q, struct request *rq) return list_entry(rq->queuelist.next, struct request, queuelist); } -static void *noop_init_queue(request_queue_t *q, elevator_t *e) +static void *noop_init_queue(request_queue_t *q) { struct noop_data *nd; diff --git a/include/linux/elevator.h b/include/linux/elevator.h index 2fa9f1144228..a24931d24404 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -21,11 +21,11 @@ typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *); typedef int (elevator_may_queue_fn) (request_queue_t *, int); typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, gfp_t); -typedef void (elevator_put_req_fn) (request_queue_t *, struct request *); +typedef void (elevator_put_req_fn) (struct request *); typedef void (elevator_activate_req_fn) (request_queue_t *, struct request *); typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *); -typedef void *(elevator_init_fn) (request_queue_t *, elevator_t *); +typedef void *(elevator_init_fn) (request_queue_t *); typedef void (elevator_exit_fn) (elevator_t *); struct elevator_ops -- cgit v1.2.3 From bce40a36de574376f41f1ff3c4d212a7da2a3c90 Mon Sep 17 00:00:00 2001 From: Philip Langdale Date: Sat, 21 Oct 2006 12:35:02 +0200 Subject: [PATCH] mmc: Add support for mmc v4 high speed mode This adds support for the high-speed modes defined by mmc v4 (assuming the host controller is up to it). On a TI sdhci controller, it improves read speed from 1.3MBps to 2.3MBps. The TI controller can only go up to 24MHz, but everything helps. Another person has taken this basic patch and used it on a Nokia 770 to get a bigger boost because that controller can run at 48MHZ. Signed-off-by: Philip Langdale Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 121 ++++++++++++++++++++++++++++++++++++++++++- include/linux/mmc/card.h | 8 +++ include/linux/mmc/protocol.h | 47 +++++++++++++++-- 3 files changed, 171 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 766bc54406e5..2d5b93000dee 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -4,6 +4,7 @@ * Copyright (C) 2003-2004 Russell King, All Rights Reserved. * SD support Copyright (C) 2004 Ian Molton, All Rights Reserved. * SD support Copyright (C) 2005 Pierre Ossman, All Rights Reserved. + * MMCv4 support Copyright (C) 2006 Philip Langdale, All Rights Reserved. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -953,6 +954,114 @@ static void mmc_read_csds(struct mmc_host *host) } } +static void mmc_process_ext_csds(struct mmc_host *host) +{ + int err; + struct mmc_card *card; + + struct mmc_request mrq; + struct mmc_command cmd; + struct mmc_data data; + + struct scatterlist sg; + + /* + * As the ext_csd is so large and mostly unused, we don't store the + * raw block in mmc_card. + */ + u8 *ext_csd; + ext_csd = kmalloc(512, GFP_KERNEL); + if (!ext_csd) { + printk("%s: could not allocate a buffer to receive the ext_csd." + "mmc v4 cards will be treated as v3.\n", + mmc_hostname(host)); + return; + } + + list_for_each_entry(card, &host->cards, node) { + if (card->state & (MMC_STATE_DEAD|MMC_STATE_PRESENT)) + continue; + if (mmc_card_sd(card)) + continue; + if (card->csd.mmca_vsn < CSD_SPEC_VER_4) + continue; + + err = mmc_select_card(host, card); + if (err != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = MMC_SEND_EXT_CSD; + cmd.arg = 0; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 512; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, ext_csd, 512); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + switch (ext_csd[EXT_CSD_CARD_TYPE]) { + case EXT_CSD_CARD_TYPE_52 | EXT_CSD_CARD_TYPE_26: + card->ext_csd.hs_max_dtr = 52000000; + break; + case EXT_CSD_CARD_TYPE_26: + card->ext_csd.hs_max_dtr = 26000000; + break; + default: + /* MMC v4 spec says this cannot happen */ + printk("%s: card is mmc v4 but doesn't support " + "any high-speed modes.\n", + mmc_hostname(card->host)); + mmc_card_set_bad(card); + continue; + } + + /* Activate highspeed support. */ + cmd.opcode = MMC_SWITCH; + cmd.arg = (MMC_SWITCH_MODE_WRITE_BYTE << 24) | + (EXT_CSD_HS_TIMING << 16) | + (1 << 8) | + EXT_CSD_CMD_SET_NORMAL; + cmd.flags = MMC_RSP_R1B | MMC_CMD_AC; + + err = mmc_wait_for_cmd(host, &cmd, CMD_RETRIES); + if (err != MMC_ERR_NONE) { + printk("%s: failed to switch card to mmc v4 " + "high-speed mode.\n", + mmc_hostname(card->host)); + continue; + } + + mmc_card_set_highspeed(card); + } + + kfree(ext_csd); + + mmc_deselect_cards(host); +} + static void mmc_read_scrs(struct mmc_host *host) { int err; @@ -1031,8 +1140,14 @@ static unsigned int mmc_calculate_clock(struct mmc_host *host) unsigned int max_dtr = host->f_max; list_for_each_entry(card, &host->cards, node) - if (!mmc_card_dead(card) && max_dtr > card->csd.max_dtr) - max_dtr = card->csd.max_dtr; + if (!mmc_card_dead(card)) { + if (mmc_card_highspeed(card)) { + if (max_dtr > card->ext_csd.hs_max_dtr) + max_dtr = card->ext_csd.hs_max_dtr; + } else if (max_dtr > card->csd.max_dtr) { + max_dtr = card->csd.max_dtr; + } + } pr_debug("%s: selected %d.%03dMHz transfer rate\n", mmc_hostname(host), @@ -1152,6 +1267,8 @@ static void mmc_setup(struct mmc_host *host) if (host->mode == MMC_MODE_SD) mmc_read_scrs(host); + else + mmc_process_ext_csds(host); } diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index 991a37382a22..ce25256f80d5 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -39,6 +39,10 @@ struct mmc_csd { write_misalign:1; }; +struct mmc_ext_csd { + unsigned int hs_max_dtr; +}; + struct sd_scr { unsigned char sda_vsn; unsigned char bus_widths; @@ -62,11 +66,13 @@ struct mmc_card { #define MMC_STATE_BAD (1<<2) /* unrecognised device */ #define MMC_STATE_SDCARD (1<<3) /* is an SD card */ #define MMC_STATE_READONLY (1<<4) /* card is read-only */ +#define MMC_STATE_HIGHSPEED (1<<5) /* card is in mmc4 highspeed mode */ u32 raw_cid[4]; /* raw card CID */ u32 raw_csd[4]; /* raw card CSD */ u32 raw_scr[2]; /* raw card SCR */ struct mmc_cid cid; /* card identification */ struct mmc_csd csd; /* card specific */ + struct mmc_ext_csd ext_csd; /* mmc v4 extended card specific */ struct sd_scr scr; /* extra SD information */ }; @@ -75,12 +81,14 @@ struct mmc_card { #define mmc_card_bad(c) ((c)->state & MMC_STATE_BAD) #define mmc_card_sd(c) ((c)->state & MMC_STATE_SDCARD) #define mmc_card_readonly(c) ((c)->state & MMC_STATE_READONLY) +#define mmc_card_highspeed(c) ((c)->state & MMC_STATE_HIGHSPEED) #define mmc_card_set_present(c) ((c)->state |= MMC_STATE_PRESENT) #define mmc_card_set_dead(c) ((c)->state |= MMC_STATE_DEAD) #define mmc_card_set_bad(c) ((c)->state |= MMC_STATE_BAD) #define mmc_card_set_sd(c) ((c)->state |= MMC_STATE_SDCARD) #define mmc_card_set_readonly(c) ((c)->state |= MMC_STATE_READONLY) +#define mmc_card_set_highspeed(c) ((c)->state |= MMC_STATE_HIGHSPEED) #define mmc_card_name(c) ((c)->cid.prod_name) #define mmc_card_id(c) ((c)->dev.bus_id) diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 08dec8d9e703..311b6547f561 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -25,14 +25,16 @@ #ifndef MMC_MMC_PROTOCOL_H #define MMC_MMC_PROTOCOL_H -/* Standard MMC commands (3.1) type argument response */ +/* Standard MMC commands (4.1) type argument response */ /* class 1 */ #define MMC_GO_IDLE_STATE 0 /* bc */ #define MMC_SEND_OP_COND 1 /* bcr [31:0] OCR R3 */ #define MMC_ALL_SEND_CID 2 /* bcr R2 */ #define MMC_SET_RELATIVE_ADDR 3 /* ac [31:16] RCA R1 */ #define MMC_SET_DSR 4 /* bc [31:16] RCA */ +#define MMC_SWITCH 6 /* ac [31:0] See below R1b */ #define MMC_SELECT_CARD 7 /* ac [31:16] RCA R1 */ +#define MMC_SEND_EXT_CSD 8 /* adtc R1 */ #define MMC_SEND_CSD 9 /* ac [31:16] RCA R2 */ #define MMC_SEND_CID 10 /* ac [31:16] RCA R2 */ #define MMC_READ_DAT_UNTIL_STOP 11 /* adtc [31:0] dadr R1 */ @@ -87,6 +89,17 @@ #define SD_APP_OP_COND 41 /* bcr [31:0] OCR R3 */ #define SD_APP_SEND_SCR 51 /* adtc R1 */ +/* + * MMC_SWITCH argument format: + * + * [31:26] Always 0 + * [25:24] Access Mode + * [23:16] Location of target Byte in EXT_CSD + * [15:08] Value Byte + * [07:03] Always 0 + * [02:00] Command Set + */ + /* MMC status in R1 Type @@ -230,13 +243,41 @@ struct _mmc_csd { #define CSD_STRUCT_VER_1_0 0 /* Valid for system specification 1.0 - 1.2 */ #define CSD_STRUCT_VER_1_1 1 /* Valid for system specification 1.4 - 2.2 */ -#define CSD_STRUCT_VER_1_2 2 /* Valid for system specification 3.1 */ +#define CSD_STRUCT_VER_1_2 2 /* Valid for system specification 3.1 - 3.2 - 3.31 - 4.0 - 4.1 */ +#define CSD_STRUCT_EXT_CSD 3 /* Version is coded in CSD_STRUCTURE in EXT_CSD */ #define CSD_SPEC_VER_0 0 /* Implements system specification 1.0 - 1.2 */ #define CSD_SPEC_VER_1 1 /* Implements system specification 1.4 */ #define CSD_SPEC_VER_2 2 /* Implements system specification 2.0 - 2.2 */ -#define CSD_SPEC_VER_3 3 /* Implements system specification 3.1 */ +#define CSD_SPEC_VER_3 3 /* Implements system specification 3.1 - 3.2 - 3.31 */ +#define CSD_SPEC_VER_4 4 /* Implements system specification 4.0 - 4.1 */ + +/* + * EXT_CSD fields + */ + +#define EXT_CSD_HS_TIMING 185 /* R/W */ +#define EXT_CSD_CARD_TYPE 196 /* RO */ + +/* + * EXT_CSD field definitions + */ + +#define EXT_CSD_CMD_SET_NORMAL (1<<0) +#define EXT_CSD_CMD_SET_SECURE (1<<1) +#define EXT_CSD_CMD_SET_CPSECURE (1<<2) + +#define EXT_CSD_CARD_TYPE_26 (1<<0) /* Card can run at 26MHz */ +#define EXT_CSD_CARD_TYPE_52 (1<<1) /* Card can run at 52MHz */ + +/* + * MMC_SWITCH access modes + */ +#define MMC_SWITCH_MODE_CMD_SET 0x00 /* Change the command set */ +#define MMC_SWITCH_MODE_SET_BITS 0x01 /* Set bits which are 1 in value */ +#define MMC_SWITCH_MODE_CLEAR_BITS 0x02 /* Clear bits which are 1 in value */ +#define MMC_SWITCH_MODE_WRITE_BYTE 0x03 /* Set target to value */ /* * SD bus widths -- cgit v1.2.3 From e45a1bd20fa5b920901879e85cdf5eda21f78d7c Mon Sep 17 00:00:00 2001 From: Philip Langdale Date: Sun, 29 Oct 2006 10:14:19 +0100 Subject: mmc: Add support for mmc v4 wide-bus modes This change adds support for the mmc4 4-bit wide-bus mode. The mmc4 spec defines 8-bit and 4-bit transfer modes. As we do not support any 8-bit hardware, this patch only adds support for the 4-bit mode, but it can easily be built upon when the time comes. The 4-bit mode is electrically compatible with SD's 4-bit mode but the procedure for turning it on is different. This patch implements only the essential parts of the procedure as defined by the spec. Two additional steps are recommended but not compulsory. I am documenting them here so that there's a record. 1) A bus-test mechanism is implemented using dedicated mmc commands which allow for testing the functionality of the data bus at the electrical level. This is pretty paranoid and they way the commands work is not compatible with the mmc subsystem (they don't set valid CRC values). 2) MMC v4 cards can indicate they would like to draw more than the default amount of current in wide-bus modes. We currently will never switch the card into a higher draw mode. Supposedly, allowing the card to draw more current will let it perform better, but the specs seem to indicate that the card will function correctly without the mode change. Empirical testing supports this interpretation. Signed-off-by: Philip Langdale Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 51 ++++++++++++++++++++++++++++++++------------ include/linux/mmc/protocol.h | 5 +++++ 2 files changed, 42 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 2d5b93000dee..1593a6a632cf 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -397,23 +397,23 @@ static int mmc_select_card(struct mmc_host *host, struct mmc_card *card) return err; /* - * Default bus width is 1 bit. - */ - host->ios.bus_width = MMC_BUS_WIDTH_1; - - /* - * We can only change the bus width of the selected - * card so therefore we have to put the handling + * We can only change the bus width of SD cards when + * they are selected so we have to put the handling * here. + * + * The card is in 1 bit mode by default so + * we only need to change if it supports the + * wider version. */ - if (host->caps & MMC_CAP_4_BIT_DATA) { + if (mmc_card_sd(card) && + (card->scr.bus_widths & SD_SCR_BUS_WIDTH_4)) { + /* - * The card is in 1 bit mode by default so - * we only need to change if it supports the - * wider version. - */ - if (mmc_card_sd(card) && - (card->scr.bus_widths & SD_SCR_BUS_WIDTH_4)) { + * Default bus width is 1 bit. + */ + host->ios.bus_width = MMC_BUS_WIDTH_1; + + if (host->caps & MMC_CAP_4_BIT_DATA) { struct mmc_command cmd; cmd.opcode = SD_APP_SET_BUS_WIDTH; cmd.arg = SD_BUS_WIDTH_4; @@ -1055,6 +1055,29 @@ static void mmc_process_ext_csds(struct mmc_host *host) } mmc_card_set_highspeed(card); + + /* Check for host support for wide-bus modes. */ + if (!(host->caps & MMC_CAP_4_BIT_DATA)) { + continue; + } + + /* Activate 4-bit support. */ + cmd.opcode = MMC_SWITCH; + cmd.arg = (MMC_SWITCH_MODE_WRITE_BYTE << 24) | + (EXT_CSD_BUS_WIDTH << 16) | + (EXT_CSD_BUS_WIDTH_4 << 8) | + EXT_CSD_CMD_SET_NORMAL; + cmd.flags = MMC_RSP_R1B | MMC_CMD_AC; + + err = mmc_wait_for_cmd(host, &cmd, CMD_RETRIES); + if (err != MMC_ERR_NONE) { + printk("%s: failed to switch card to " + "mmc v4 4-bit bus mode.\n", + mmc_hostname(card->host)); + continue; + } + + host->ios.bus_width = MMC_BUS_WIDTH_4; } kfree(ext_csd); diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 311b6547f561..45c51fd85786 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -256,6 +256,7 @@ struct _mmc_csd { * EXT_CSD fields */ +#define EXT_CSD_BUS_WIDTH 183 /* R/W */ #define EXT_CSD_HS_TIMING 185 /* R/W */ #define EXT_CSD_CARD_TYPE 196 /* RO */ @@ -270,6 +271,10 @@ struct _mmc_csd { #define EXT_CSD_CARD_TYPE_26 (1<<0) /* Card can run at 26MHz */ #define EXT_CSD_CARD_TYPE_52 (1<<1) /* Card can run at 52MHz */ +#define EXT_CSD_BUS_WIDTH_1 0 /* Card is in 1 bit mode */ +#define EXT_CSD_BUS_WIDTH_4 1 /* Card is in 4 bit mode */ +#define EXT_CSD_BUS_WIDTH_8 2 /* Card is in 8 bit mode */ + /* * MMC_SWITCH access modes */ -- cgit v1.2.3 From 7ccd266e676a3f0c6f8f897f58b684cac3dd1650 Mon Sep 17 00:00:00 2001 From: Pierre Ossman Date: Wed, 8 Nov 2006 23:03:10 +0100 Subject: mmc: Support for high speed SD cards Modern SD cards support a clock speed of 50 MHz. Make sure we test for this capability and do the song and dance required to activate it. Activating high speed support actually modifies the TRAN_SPEED field of the CSD. But as the spec says that the cards must report 50 MHz, we might as well skip re-reading the CSD. Signed-off-by: Pierre Ossman --- drivers/mmc/mmc.c | 122 +++++++++++++++++++++++++++++++++++++++++-- include/linux/mmc/card.h | 7 ++- include/linux/mmc/protocol.h | 22 ++++++++ 3 files changed, 146 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/mmc/mmc.c b/drivers/mmc/mmc.c index 82b7643c1654..9d190022a490 100644 --- a/drivers/mmc/mmc.c +++ b/drivers/mmc/mmc.c @@ -1157,6 +1157,116 @@ static void mmc_read_scrs(struct mmc_host *host) mmc_deselect_cards(host); } +static void mmc_read_switch_caps(struct mmc_host *host) +{ + int err; + struct mmc_card *card; + struct mmc_request mrq; + struct mmc_command cmd; + struct mmc_data data; + unsigned char *status; + struct scatterlist sg; + + status = kmalloc(64, GFP_KERNEL); + if (!status) { + printk(KERN_WARNING "%s: Unable to allocate buffer for " + "reading switch capabilities.\n", + mmc_hostname(host)); + return; + } + + list_for_each_entry(card, &host->cards, node) { + if (card->state & (MMC_STATE_DEAD|MMC_STATE_PRESENT)) + continue; + if (!mmc_card_sd(card)) + continue; + if (card->scr.sda_vsn < SCR_SPEC_VER_1) + continue; + + err = mmc_select_card(host, card); + if (err != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = SD_SWITCH; + cmd.arg = 0x00FFFFF1; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 64; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, status, 64); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + if (status[13] & 0x02) + card->sw_caps.hs_max_dtr = 50000000; + + memset(&cmd, 0, sizeof(struct mmc_command)); + + cmd.opcode = SD_SWITCH; + cmd.arg = 0x80FFFFF1; + cmd.flags = MMC_RSP_R1 | MMC_CMD_ADTC; + + memset(&data, 0, sizeof(struct mmc_data)); + + mmc_set_data_timeout(&data, card, 0); + + data.blksz = 64; + data.blocks = 1; + data.flags = MMC_DATA_READ; + data.sg = &sg; + data.sg_len = 1; + + memset(&mrq, 0, sizeof(struct mmc_request)); + + mrq.cmd = &cmd; + mrq.data = &data; + + sg_init_one(&sg, status, 64); + + mmc_wait_for_req(host, &mrq); + + if (cmd.error != MMC_ERR_NONE || data.error != MMC_ERR_NONE) { + mmc_card_set_dead(card); + continue; + } + + if ((status[16] & 0xF) != 1) { + printk(KERN_WARNING "%s: Problem switching card " + "into high-speed mode!\n", + mmc_hostname(host)); + continue; + } + + mmc_card_set_highspeed(card); + } + + kfree(status); + + mmc_deselect_cards(host); +} + static unsigned int mmc_calculate_clock(struct mmc_host *host) { struct mmc_card *card; @@ -1164,9 +1274,12 @@ static unsigned int mmc_calculate_clock(struct mmc_host *host) list_for_each_entry(card, &host->cards, node) if (!mmc_card_dead(card)) { - if (mmc_card_highspeed(card)) { + if (mmc_card_highspeed(card) && mmc_card_sd(card)) { + if (max_dtr > card->sw_caps.hs_max_dtr) + max_dtr = card->sw_caps.hs_max_dtr; + } else if (mmc_card_highspeed(card) && !mmc_card_sd(card)) { if (max_dtr > card->ext_csd.hs_max_dtr) - max_dtr = card->ext_csd.hs_max_dtr; + max_dtr = card->ext_csd.hs_max_dtr; } else if (max_dtr > card->csd.max_dtr) { max_dtr = card->csd.max_dtr; } @@ -1288,9 +1401,10 @@ static void mmc_setup(struct mmc_host *host) mmc_read_csds(host); - if (host->mode == MMC_MODE_SD) + if (host->mode == MMC_MODE_SD) { mmc_read_scrs(host); - else + mmc_read_switch_caps(host); + } else mmc_process_ext_csds(host); } diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index ce25256f80d5..d0e6a5497614 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -50,6 +50,10 @@ struct sd_scr { #define SD_SCR_BUS_WIDTH_4 (1<<2) }; +struct sd_switch_caps { + unsigned int hs_max_dtr; +}; + struct mmc_host; /* @@ -66,7 +70,7 @@ struct mmc_card { #define MMC_STATE_BAD (1<<2) /* unrecognised device */ #define MMC_STATE_SDCARD (1<<3) /* is an SD card */ #define MMC_STATE_READONLY (1<<4) /* card is read-only */ -#define MMC_STATE_HIGHSPEED (1<<5) /* card is in mmc4 highspeed mode */ +#define MMC_STATE_HIGHSPEED (1<<5) /* card is in high speed mode */ u32 raw_cid[4]; /* raw card CID */ u32 raw_csd[4]; /* raw card CSD */ u32 raw_scr[2]; /* raw card SCR */ @@ -74,6 +78,7 @@ struct mmc_card { struct mmc_csd csd; /* card specific */ struct mmc_ext_csd ext_csd; /* mmc v4 extended card specific */ struct sd_scr scr; /* extra SD information */ + struct sd_switch_caps sw_caps; /* switch (CMD6) caps */ }; #define mmc_card_present(c) ((c)->state & MMC_STATE_PRESENT) diff --git a/include/linux/mmc/protocol.h b/include/linux/mmc/protocol.h index 45c51fd85786..2dce60c43f4b 100644 --- a/include/linux/mmc/protocol.h +++ b/include/linux/mmc/protocol.h @@ -82,6 +82,7 @@ /* class 8 */ /* This is basically the same command as for MMC with some quirks. */ #define SD_SEND_RELATIVE_ADDR 3 /* bcr R6 */ +#define SD_SWITCH 6 /* adtc [31:0] See below R1 */ /* Application commands */ #define SD_APP_SET_BUS_WIDTH 6 /* ac [1:0] bus width R1 */ @@ -100,6 +101,19 @@ * [02:00] Command Set */ +/* + * SD_SWITCH argument format: + * + * [31] Check (0) or switch (1) + * [30:24] Reserved (0) + * [23:20] Function group 6 + * [19:16] Function group 5 + * [15:12] Function group 4 + * [11:8] Function group 3 + * [7:4] Function group 2 + * [3:0] Function group 1 + */ + /* MMC status in R1 Type @@ -284,6 +298,14 @@ struct _mmc_csd { #define MMC_SWITCH_MODE_CLEAR_BITS 0x02 /* Clear bits which are 1 in value */ #define MMC_SWITCH_MODE_WRITE_BYTE 0x03 /* Set target to value */ +/* + * SCR field definitions + */ + +#define SCR_SPEC_VER_0 0 /* Implements system specification 1.0 - 1.01 */ +#define SCR_SPEC_VER_1 1 /* Implements system specification 1.10 */ +#define SCR_SPEC_VER_2 2 /* Implements system specification 2.00 */ + /* * SD bus widths */ -- cgit v1.2.3 From 437052516779fea608261a50682b124315f48f01 Mon Sep 17 00:00:00 2001 From: inaky@linux.intel.com Date: Wed, 11 Oct 2006 20:05:58 -0700 Subject: usb/hub: allow hubs up to 31 children Current Wireless USB host hardware (Intel i1480 for example) allows up to 22 devices to connect, thus bringing up the max number of children in the WUSB Host Controller to 22 'fake' ports. Upcoming hardware might raise that limit. Makes almost no difference to go to 31, as the bit arrays are byte-aligned (plus an extra bit in general), so 22 bits fit in 4 bytes as 31 do. As well, the only other array that depends on USB_MAXCHILDREN is 'struct usb_hub->indicator'. By declaring it 'u8' instead of 'enum hub_led_mode', we reduce the size of each entry from 4 bytes (in i386) to 1, which will add as we when are doubling USB_MAXCHILDREN (with 16 the size of that array is 64 bytes, with 31 would be 128; by using u8 that goes down to 31 bytes). Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hub.h | 2 +- include/linux/usb.h | 7 ++++++- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h index 0f8e82a4d480..1b05994de4dc 100644 --- a/drivers/usb/core/hub.h +++ b/drivers/usb/core/hub.h @@ -229,7 +229,7 @@ struct usb_hub { unsigned resume_root_hub:1; unsigned has_indicators:1; - enum hub_led_mode indicator[USB_MAXCHILDREN]; + u8 indicator[USB_MAXCHILDREN]; struct work_struct leds; }; diff --git a/include/linux/usb.h b/include/linux/usb.h index 5482bfb3303d..e5cb1690975a 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -313,8 +313,13 @@ struct usb_bus { /* This is arbitrary. * From USB 2.0 spec Table 11-13, offset 7, a hub can * have up to 255 ports. The most yet reported is 10. + * + * Current Wireless USB host hardware (Intel i1480 for example) allows + * up to 22 devices to connect. Upcoming hardware might raise that + * limit. Because the arrays need to add a bit for hub status data, we + * do 31, so plus one evens out to four bytes. */ -#define USB_MAXCHILDREN (16) +#define USB_MAXCHILDREN (31) struct usb_tt; -- cgit v1.2.3 From 0c1ac4f25f894f9df0ffe9b912c165fb6a185a3c Mon Sep 17 00:00:00 2001 From: Luiz Fernando N. Capitulino Date: Mon, 30 Oct 2006 14:53:03 -0300 Subject: USB: makes usb_endpoint_* functions inline. We have no benefits of having the usb_endpoint_* functions as functions, but making them inline saves text and data segment sizes: text data bss dec hex filename 14893634 3108770 1108840 19111244 1239d4c vmlinux.func 14893185 3108566 1108840 19110591 1239abf vmlinux.inline This is the result of a 2.6.19-rc3 kernel compiled with GCC 4.1.1 without CONFIG_MODULES, CONFIG_CC_OPTIMIZE_FOR_SIZE, CONFIG_REGPARM options set. USB support is fully enabled (while most of the other drivers are not), and that kernel has most of the USB code ported to use the endpoint functions. That happens because a call to those functions are expensive (in terms of bytes), while the function's size is smaller or have the same 'size' of the call. Signed-off-by: Luiz Fernando N. Capitulino Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/usb.c | 144 ------------------------------------------------- include/linux/usb.h | 142 ++++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 131 insertions(+), 155 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c index 467cb02832f3..a83c2d5065c1 100644 --- a/drivers/usb/core/usb.c +++ b/drivers/usb/core/usb.c @@ -537,138 +537,6 @@ int usb_get_current_frame_number(struct usb_device *dev) return usb_hcd_get_frame_number (dev); } -/** - * usb_endpoint_dir_in - check if the endpoint has IN direction - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type IN, otherwise it returns false. - */ -int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN); -} - -/** - * usb_endpoint_dir_out - check if the endpoint has OUT direction - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type OUT, otherwise it returns false. - */ -int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT); -} - -/** - * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type bulk, otherwise it returns false. - */ -int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_BULK); -} - -/** - * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type interrupt, otherwise it returns - * false. - */ -int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_INT); -} - -/** - * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type - * @epd: endpoint to be checked - * - * Returns true if the endpoint is of type isochronous, otherwise it returns - * false. - */ -int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd) -{ - return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == - USB_ENDPOINT_XFER_ISOC); -} - -/** - * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has bulk transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has bulk transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd)); -} - -/** - * usb_endpoint_is_int_in - check if the endpoint is interrupt IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has interrupt transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has interrupt transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd)); -} - -/** - * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN - * @epd: endpoint to be checked - * - * Returns true if the endpoint has isochronous transfer type and IN direction, - * otherwise it returns false. - */ -int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd)); -} - -/** - * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT - * @epd: endpoint to be checked - * - * Returns true if the endpoint has isochronous transfer type and OUT direction, - * otherwise it returns false. - */ -int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd) -{ - return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd)); -} - /*-------------------------------------------------------------------*/ /* * __usb_get_extra_descriptor() finds a descriptor of specific type in the @@ -1102,18 +970,6 @@ EXPORT_SYMBOL(__usb_get_extra_descriptor); EXPORT_SYMBOL(usb_find_device); EXPORT_SYMBOL(usb_get_current_frame_number); -EXPORT_SYMBOL_GPL(usb_endpoint_dir_in); -EXPORT_SYMBOL_GPL(usb_endpoint_dir_out); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_bulk); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_int); -EXPORT_SYMBOL_GPL(usb_endpoint_xfer_isoc); -EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_out); -EXPORT_SYMBOL_GPL(usb_endpoint_is_int_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_int_out); -EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_in); -EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_out); - EXPORT_SYMBOL (usb_buffer_alloc); EXPORT_SYMBOL (usb_buffer_free); diff --git a/include/linux/usb.h b/include/linux/usb.h index e5cb1690975a..e732e024a141 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -495,17 +495,137 @@ static inline int usb_make_path (struct usb_device *dev, char *buf, /*-------------------------------------------------------------------------*/ -extern int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd); -extern int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd); +/** + * usb_endpoint_dir_in - check if the endpoint has IN direction + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type IN, otherwise it returns false. + */ +static inline int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN); +} + +/** + * usb_endpoint_dir_out - check if the endpoint has OUT direction + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type OUT, otherwise it returns false. + */ +static inline int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT); +} + +/** + * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type bulk, otherwise it returns false. + */ +static inline int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_BULK); +} + +/** + * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type interrupt, otherwise it returns + * false. + */ +static inline int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_INT); +} + +/** + * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type + * @epd: endpoint to be checked + * + * Returns true if the endpoint is of type isochronous, otherwise it returns + * false. + */ +static inline int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd) +{ + return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == + USB_ENDPOINT_XFER_ISOC); +} + +/** + * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has bulk transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has bulk transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd)); +} + +/** + * usb_endpoint_is_int_in - check if the endpoint is interrupt IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has interrupt transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has interrupt transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd)); +} + +/** + * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN + * @epd: endpoint to be checked + * + * Returns true if the endpoint has isochronous transfer type and IN direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd)); +} + +/** + * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT + * @epd: endpoint to be checked + * + * Returns true if the endpoint has isochronous transfer type and OUT direction, + * otherwise it returns false. + */ +static inline int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd) +{ + return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd)); +} /*-------------------------------------------------------------------------*/ -- cgit v1.2.3 From 692a186c9d5f12d43cef28d40c25247dc4f302f0 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 30 Oct 2006 17:07:51 -0500 Subject: USB: expand autosuspend/autoresume API This patch (as814) adds usb_autopm_set_interface() to the autosuspend API. It also provides convenient wrapper routines, usb_autopm_enable() and usb_autopm_disable(), for drivers that want to specify directly whether autosuspend should be allowed. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 22 ++++++++++++++++++++++ include/linux/usb.h | 29 ++++++++++++++++++++++++++--- 2 files changed, 48 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index ca0e40ed2b72..204495fa6b3d 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -1322,6 +1322,28 @@ int usb_autopm_get_interface(struct usb_interface *intf) } EXPORT_SYMBOL_GPL(usb_autopm_get_interface); +/** + * usb_autopm_set_interface - set a USB interface's autosuspend state + * @intf: the usb_interface whose state should be set + * + * This routine sets the autosuspend state of @intf's device according + * to @intf's usage counter, which the caller must have set previously. + * If the counter is <= 0, the device is autosuspended (if it isn't + * already suspended and if nothing else prevents the autosuspend). If + * the counter is > 0, the device is autoresumed (if it isn't already + * awake). + */ +int usb_autopm_set_interface(struct usb_interface *intf) +{ + int status; + + status = usb_autopm_do_interface(intf, 0); + // dev_dbg(&intf->dev, "%s: status %d cnt %d\n", + // __FUNCTION__, status, intf->pm_usage_cnt); + return status; +} +EXPORT_SYMBOL_GPL(usb_autopm_set_interface); + #endif /* CONFIG_USB_SUSPEND */ static int usb_suspend(struct device *dev, pm_message_t message) diff --git a/include/linux/usb.h b/include/linux/usb.h index e732e024a141..864c6c21c21e 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -415,14 +415,37 @@ extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id); /* USB autosuspend and autoresume */ #ifdef CONFIG_USB_SUSPEND +extern int usb_autopm_set_interface(struct usb_interface *intf); extern int usb_autopm_get_interface(struct usb_interface *intf); extern void usb_autopm_put_interface(struct usb_interface *intf); +static inline void usb_autopm_enable(struct usb_interface *intf) +{ + intf->pm_usage_cnt = 0; + usb_autopm_set_interface(intf); +} + +static inline void usb_autopm_disable(struct usb_interface *intf) +{ + intf->pm_usage_cnt = 1; + usb_autopm_set_interface(intf); +} + #else -#define usb_autopm_get_interface(intf) 0 -#define usb_autopm_put_interface(intf) do {} while (0) -#endif +static inline int usb_autopm_set_interface(struct usb_interface *intf) +{ return 0; } + +static inline int usb_autopm_get_interface(struct usb_interface *intf) +{ return 0; } + +static inline void usb_autopm_put_interface(struct usb_interface *intf) +{ } +static inline void usb_autopm_enable(struct usb_interface *intf) +{ } +static inline void usb_autopm_disable(struct usb_interface *intf) +{ } +#endif /*-------------------------------------------------------------------------*/ -- cgit v1.2.3 From ce3615879ae85373c03744b45b7c2d7ae5e29b2a Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 20 Nov 2006 11:12:22 -0500 Subject: USB: struct usb_device: change flag to bitflag This patch (as816) changes an existing flag in the usb_device structure to a bitflag, preparing the way for more bitflags to come in the future. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/message.c | 2 +- include/linux/usb.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 7729c0744886..5684d8722922 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -764,7 +764,7 @@ int usb_string(struct usb_device *dev, int index, char *buf, size_t size) err = -EINVAL; goto errout; } else { - dev->have_langid = -1; + dev->have_langid = 1; dev->string_langid = tbuf[2] | (tbuf[3]<< 8); /* always use the first langid listed */ dev_dbg (&dev->dev, "default language 0x%04x\n", diff --git a/include/linux/usb.h b/include/linux/usb.h index 864c6c21c21e..5634a2d91ec0 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -362,7 +362,7 @@ struct usb_device { u8 portnum; /* Parent port number (origin 1) */ u8 level; /* Number of USB hub ancestors */ - int have_langid; /* whether string_langid is valid */ + unsigned have_langid:1; /* whether string_langid is valid */ int string_langid; /* language ID for strings */ /* static strings from the device */ -- cgit v1.2.3 From ee49fb5dc89d34f1794ac9362fa97c1a640f7ddd Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 22 Nov 2006 16:55:54 -0500 Subject: USB: keep count of unsuspended children This patch (as818b) simplifies autosuspend processing by keeping track of the number of unsuspended children of each USB hub. This will permit us to avoid a good deal of unnecessary work all the time; we will no longer have to create a bunch of workqueue entries to carry out autosuspend requests, only to have them fail because one of the hub's children isn't suspended. The basic idea is simple. There already is a usage counter in the usb_device structure for preventing autosuspends. The patch just increments that counter for every unsuspended child. There's only one tricky part: When a device disconnects we need to remember whether it was suspended at the time (leave the counter alone) or not (decrement the counter). Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 34 +++++++++++++++++++++++----------- drivers/usb/core/hub.c | 14 ++++++++++++++ include/linux/usb.h | 1 + 3 files changed, 38 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index 40c1bf09b2b7..0fa15bd62c48 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -1048,7 +1048,7 @@ int usb_suspend_both(struct usb_device *udev, pm_message_t msg) /* If the suspend succeeded, propagate it up the tree */ } else if (parent) - usb_autosuspend_device(parent, 0); + usb_autosuspend_device(parent, 1); // dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status); return status; @@ -1096,9 +1096,25 @@ int usb_resume_both(struct usb_device *udev) /* Propagate the resume up the tree, if necessary */ if (udev->state == USB_STATE_SUSPENDED) { if (parent) { - usb_pm_lock(parent); - parent->auto_pm = 1; - status = usb_resume_both(parent); + status = usb_autoresume_device(parent, 1); + if (status == 0) { + status = usb_resume_device(udev); + if (status) { + usb_autosuspend_device(parent, 1); + + /* It's possible usb_resume_device() + * failed after the port was + * unsuspended, causing udev to be + * logically disconnected. We don't + * want usb_disconnect() to autosuspend + * the parent again, so tell it that + * udev disconnected while still + * suspended. */ + if (udev->state == + USB_STATE_NOTATTACHED) + udev->discon_suspended = 1; + } + } } else { /* We can't progagate beyond the USB subsystem, @@ -1107,11 +1123,9 @@ int usb_resume_both(struct usb_device *udev) if (udev->dev.parent->power.power_state.event != PM_EVENT_ON) status = -EHOSTUNREACH; - } - if (status == 0) - status = usb_resume_device(udev); - if (parent) - usb_pm_unlock(parent); + else + status = usb_resume_device(udev); + } } else { /* Needed only for setting udev->dev.power.power_state.event @@ -1119,8 +1133,6 @@ int usb_resume_both(struct usb_device *udev) status = usb_resume_device(udev); } - /* Now the parent won't suspend until we are finished */ - if (status == 0 && udev->actconfig) { for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) { intf = udev->actconfig->interface[i]; diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 46df5e60764b..e46d38b18249 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1039,6 +1039,8 @@ static void recursively_mark_NOTATTACHED(struct usb_device *udev) if (udev->children[i]) recursively_mark_NOTATTACHED(udev->children[i]); } + if (udev->state == USB_STATE_SUSPENDED) + udev->discon_suspended = 1; udev->state = USB_STATE_NOTATTACHED; } @@ -1228,6 +1230,14 @@ void usb_disconnect(struct usb_device **pdev) *pdev = NULL; spin_unlock_irq(&device_state_lock); + /* Decrement the parent's count of unsuspended children */ + if (udev->parent) { + usb_pm_lock(udev); + if (!udev->discon_suspended) + usb_autosuspend_device(udev->parent, 1); + usb_pm_unlock(udev); + } + put_device(&udev->dev); } @@ -1356,6 +1366,10 @@ static int __usb_new_device(void *void_data) goto fail; } + /* Increment the parent's count of unsuspended children */ + if (udev->parent) + usb_autoresume_device(udev->parent, 1); + exit: module_put(THIS_MODULE); return err; diff --git a/include/linux/usb.h b/include/linux/usb.h index 5634a2d91ec0..0cd73edeef13 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -362,6 +362,7 @@ struct usb_device { u8 portnum; /* Parent port number (origin 1) */ u8 level; /* Number of USB hub ancestors */ + unsigned discon_suspended:1; /* Disconnected while suspended */ unsigned have_langid:1; /* whether string_langid is valid */ int string_langid; /* language ID for strings */ -- cgit v1.2.3 From e65e5fb5ceb02aaea7b65bf8b3b0d0c9057718b6 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 7 Nov 2006 18:21:21 +1100 Subject: PCI: Make some MSI-X #defines generic Move some MSI-X #defines into pci_regs.h so they can be used outside of drivers/pci. Signed-off-by: Michael Ellerman Signed-off-by: Greg Kroah-Hartman --- drivers/pci/msi.h | 8 -------- include/linux/pci_regs.h | 6 ++++++ 2 files changed, 6 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h index f0cca1772f9c..3898f5237144 100644 --- a/drivers/pci/msi.h +++ b/drivers/pci/msi.h @@ -6,14 +6,6 @@ #ifndef MSI_H #define MSI_H -/* - * MSI-X Address Register - */ -#define PCI_MSIX_FLAGS_QSIZE 0x7FF -#define PCI_MSIX_FLAGS_ENABLE (1 << 15) -#define PCI_MSIX_FLAGS_BIRMASK (7 << 0) -#define PCI_MSIX_FLAGS_BITMASK (1 << 0) - #define PCI_MSIX_ENTRY_SIZE 16 #define PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET 0 #define PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET 4 diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index c321316f1bc7..064b1dc71c22 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -292,6 +292,12 @@ #define PCI_MSI_DATA_64 12 /* 16 bits of data for 64-bit devices */ #define PCI_MSI_MASK_BIT 16 /* Mask bits register */ +/* MSI-X registers (these are at offset PCI_MSI_FLAGS) */ +#define PCI_MSIX_FLAGS_QSIZE 0x7FF +#define PCI_MSIX_FLAGS_ENABLE (1 << 15) +#define PCI_MSIX_FLAGS_BIRMASK (7 << 0) +#define PCI_MSIX_FLAGS_BITMASK (1 << 0) + /* CompactPCI Hotswap Register */ #define PCI_CHSWP_CSR 2 /* Control and Status Register */ -- cgit v1.2.3 From a2302c68d923537436b1114aa207787c1a31bd50 Mon Sep 17 00:00:00 2001 From: John Keller Date: Wed, 4 Oct 2006 16:49:52 -0500 Subject: Altix: Initial ACPI support - ROM shadowing. Support a shadowed ROM when running with an ACPI capable PROM. Define a new dev.resource flag IORESOURCE_ROM_BIOS_COPY to describe the case of a BIOS shadowed ROM, which can then be used to avoid pci_map_rom() making an unneeded call to pci_enable_rom(). Signed-off-by: John Keller Signed-off-by: Greg Kroah-Hartman --- arch/ia64/sn/kernel/io_acpi_init.c | 33 +++++++++++++++++++++++++++++++++ arch/ia64/sn/kernel/io_common.c | 5 +++-- arch/ia64/sn/kernel/io_init.c | 3 +++ drivers/pci/rom.c | 9 ++++++--- include/linux/ioport.h | 1 + 5 files changed, 46 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c index a9dc36901b19..99d7f278612a 100644 --- a/arch/ia64/sn/kernel/io_acpi_init.c +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -169,6 +169,39 @@ sn_acpi_bus_fixup(struct pci_bus *bus) } } +/* + * sn_acpi_slot_fixup - Perform any SN specific slot fixup. + * At present there does not appear to be + * any generic way to handle a ROM image + * that has been shadowed by the PROM, so + * we pass a pointer to it within the + * pcidev_info structure. + */ + +void +sn_acpi_slot_fixup(struct pci_dev *dev, struct pcidev_info *pcidev_info) +{ + void __iomem *addr; + size_t size; + + if (pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE]) { + /* + * A valid ROM image exists and has been shadowed by the + * PROM. Setup the pci_dev ROM resource to point to + * the shadowed copy. + */ + size = dev->resource[PCI_ROM_RESOURCE].end - + dev->resource[PCI_ROM_RESOURCE].start; + addr = + ioremap(pcidev_info->pdi_pio_mapped_addr[PCI_ROM_RESOURCE], + size); + dev->resource[PCI_ROM_RESOURCE].start = (unsigned long) addr; + dev->resource[PCI_ROM_RESOURCE].end = + (unsigned long) addr + size; + dev->resource[PCI_ROM_RESOURCE].flags |= IORESOURCE_ROM_BIOS_COPY; + } +} + static struct acpi_driver acpi_sn_hubdev_driver = { .name = "SGI HUBDEV Driver", .ids = "SGIHUB,SGITIO", diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c index 12531de6754c..d4dd8f4b6b8d 100644 --- a/arch/ia64/sn/kernel/io_common.c +++ b/arch/ia64/sn/kernel/io_common.c @@ -286,9 +286,10 @@ void sn_pci_fixup_slot(struct pci_dev *dev) list_add_tail(&pcidev_info->pdi_list, &(SN_PLATFORM_DATA(dev->bus)->pcidev_info)); - if (!SN_ACPI_BASE_SUPPORT()) + if (SN_ACPI_BASE_SUPPORT()) + sn_acpi_slot_fixup(dev, pcidev_info); + else sn_more_slot_fixup(dev, pcidev_info); - /* * Using the PROMs values for the PCI host bus, get the Linux * PCI host_pci_dev struct and set up host bus linkages diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index 990224a44121..9ad843e0383b 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -210,6 +210,9 @@ sn_more_slot_fixup(struct pci_dev *dev, struct pcidev_info *pcidev_info) dev->resource[idx].parent = &ioport_resource; else dev->resource[idx].parent = &iomem_resource; + /* If ROM, mark as shadowed in PROM */ + if (idx == PCI_ROM_RESOURCE) + dev->resource[idx].flags |= IORESOURCE_ROM_BIOS_COPY; } /* Create a pci_window in the pci_controller struct for * each device resource. diff --git a/drivers/pci/rom.c b/drivers/pci/rom.c index e1dcefc69bb4..d087e0817715 100644 --- a/drivers/pci/rom.c +++ b/drivers/pci/rom.c @@ -81,7 +81,8 @@ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size) start = (loff_t)0xC0000; *size = 0x20000; /* cover C000:0 through E000:0 */ } else { - if (res->flags & IORESOURCE_ROM_COPY) { + if (res->flags & + (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY)) { *size = pci_resource_len(pdev, PCI_ROM_RESOURCE); return (void __iomem *)(unsigned long) pci_resource_start(pdev, PCI_ROM_RESOURCE); @@ -165,7 +166,8 @@ void __iomem *pci_map_rom_copy(struct pci_dev *pdev, size_t *size) if (!rom) return NULL; - if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_SHADOW)) + if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_SHADOW | + IORESOURCE_ROM_BIOS_COPY)) return rom; res->start = (unsigned long)kmalloc(*size, GFP_KERNEL); @@ -191,7 +193,7 @@ void pci_unmap_rom(struct pci_dev *pdev, void __iomem *rom) { struct resource *res = &pdev->resource[PCI_ROM_RESOURCE]; - if (res->flags & IORESOURCE_ROM_COPY) + if (res->flags & (IORESOURCE_ROM_COPY | IORESOURCE_ROM_BIOS_COPY)) return; iounmap(rom); @@ -215,6 +217,7 @@ void pci_remove_rom(struct pci_dev *pdev) sysfs_remove_bin_file(&pdev->dev.kobj, pdev->rom_attr); if (!(res->flags & (IORESOURCE_ROM_ENABLE | IORESOURCE_ROM_SHADOW | + IORESOURCE_ROM_BIOS_COPY | IORESOURCE_ROM_COPY))) pci_disable_rom(pdev); } diff --git a/include/linux/ioport.h b/include/linux/ioport.h index d42c83399071..cf8696d4a138 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -89,6 +89,7 @@ struct resource_list { #define IORESOURCE_ROM_ENABLE (1<<0) /* ROM is enabled, same as PCI_ROM_ADDRESS_ENABLE */ #define IORESOURCE_ROM_SHADOW (1<<1) /* ROM is copy at C000:0 */ #define IORESOURCE_ROM_COPY (1<<2) /* ROM is alloc'd copy, resource field overlaid */ +#define IORESOURCE_ROM_BIOS_COPY (1<<3) /* ROM is BIOS copy, resource field overlaid */ /* PC/ISA/whatever - the normal PC address spaces: IO and memory */ extern struct resource ioport_resource; -- cgit v1.2.3 From bae94d02371c402408a4edfb95e71e88dbd3e973 Mon Sep 17 00:00:00 2001 From: Inaky Perez-Gonzalez Date: Wed, 22 Nov 2006 12:40:31 -0800 Subject: PCI: switch pci_{enable,disable}_device() to be nestable Changes the pci_{enable,disable}_device() functions to work in a nested basis, so that eg, three calls to enable_device() require three calls to disable_device(). The reason for this is to simplify PCI drivers for multi-interface/capability devices. These are devices that cram more than one interface in a single function. A relevant example of that is the Wireless [USB] Host Controller Interface (similar to EHCI) [see http://www.intel.com/technology/comms/wusb/whci.htm]. In these kind of devices, multiple interfaces are accessed through a single bar and IRQ line. For that, the drivers map only the smallest area of the bar to access their register banks and use shared IRQ handlers. However, because the order at which those drivers load cannot be known ahead of time, the sequence in which the calls to pci_enable_device() and pci_disable_device() cannot be predicted. Thus: 1. driverA starts pci_enable_device() 2. driverB starts pci_enable_device() 3. driverA shutdown pci_disable_device() 4. driverB shutdown pci_disable_device() between steps 3 and 4, driver B would loose access to it's device, even if it didn't intend to. By using this modification, the device won't be disabled until all the callers to enable() have called disable(). This is implemented by replacing 'struct pci_dev->is_enabled' from a bitfield to an atomic use count. Each caller to enable increments it, each caller to disable decrements it. When the count increments from 0 to 1, __pci_enable_device() is called to actually enable the device. When it drops to zero, pci_disable_device() actually does the disabling. We keep the backend __pci_enable_device() for pci_default_resume() to use and also change the sysfs method implementation, so that userspace enabling/disabling the device doesn't disable it one time too much. Signed-off-by: Inaky Perez-Gonzalez Signed-off-by: Greg Kroah-Hartman --- drivers/pci/pci-driver.c | 4 ++-- drivers/pci/pci-sysfs.c | 33 +++++++++++++++++++++------------ drivers/pci/pci.c | 40 +++++++++++++++++++++++++++++++++------- drivers/pci/pci.h | 1 + include/linux/pci.h | 3 ++- 5 files changed, 59 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 84ec9c8f6703..e5ae3a0c13bb 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -329,8 +329,8 @@ static int pci_default_resume(struct pci_dev *pci_dev) /* restore the PCI config space */ pci_restore_state(pci_dev); /* if the device was enabled before suspend, reenable */ - if (pci_dev->is_enabled) - retval = pci_enable_device(pci_dev); + if (atomic_read(&pci_dev->enable_cnt)) + retval = __pci_enable_device(pci_dev); /* if the device was busmaster before the suspend, make it busmaster again */ if (pci_dev->is_busmaster) pci_set_master(pci_dev); diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index f952bfea48a6..7a94076752d0 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -42,7 +42,6 @@ pci_config_attr(subsystem_vendor, "0x%04x\n"); pci_config_attr(subsystem_device, "0x%04x\n"); pci_config_attr(class, "0x%06x\n"); pci_config_attr(irq, "%u\n"); -pci_config_attr(is_enabled, "%u\n"); static ssize_t broken_parity_status_show(struct device *dev, struct device_attribute *attr, @@ -112,26 +111,36 @@ static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, (u8)(pci_dev->class >> 16), (u8)(pci_dev->class >> 8), (u8)(pci_dev->class)); } -static ssize_t -is_enabled_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) + +static ssize_t is_enabled_store(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) { + ssize_t result = -EINVAL; struct pci_dev *pdev = to_pci_dev(dev); - int retval = 0; /* this can crash the machine when done on the "wrong" device */ if (!capable(CAP_SYS_ADMIN)) return count; - if (*buf == '0') - pci_disable_device(pdev); + if (*buf == '0') { + if (atomic_read(&pdev->enable_cnt) != 0) + pci_disable_device(pdev); + else + result = -EIO; + } else if (*buf == '1') + result = pci_enable_device(pdev); + + return result < 0 ? result : count; +} - if (*buf == '1') - retval = pci_enable_device(pdev); +static ssize_t is_enabled_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct pci_dev *pdev; - if (retval) - return retval; - return count; + pdev = to_pci_dev (dev); + return sprintf (buf, "%u\n", atomic_read(&pdev->enable_cnt)); } static ssize_t diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 427991741cf3..5a14b73cf3a1 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -612,29 +612,50 @@ pci_enable_device_bars(struct pci_dev *dev, int bars) } /** - * pci_enable_device - Initialize device before it's used by a driver. + * __pci_enable_device - Initialize device before it's used by a driver. * @dev: PCI device to be initialized * * Initialize device before it's used by a driver. Ask low-level code * to enable I/O and memory. Wake up the device if it was suspended. * Beware, this function can fail. + * + * Note this function is a backend and is not supposed to be called by + * normal code, use pci_enable_device() instead. */ int -pci_enable_device(struct pci_dev *dev) +__pci_enable_device(struct pci_dev *dev) { int err; - if (dev->is_enabled) - return 0; - err = pci_enable_device_bars(dev, (1 << PCI_NUM_RESOURCES) - 1); if (err) return err; pci_fixup_device(pci_fixup_enable, dev); - dev->is_enabled = 1; return 0; } +/** + * pci_enable_device - Initialize device before it's used by a driver. + * @dev: PCI device to be initialized + * + * Initialize device before it's used by a driver. Ask low-level code + * to enable I/O and memory. Wake up the device if it was suspended. + * Beware, this function can fail. + * + * Note we don't actually enable the device many times if we call + * this function repeatedly (we just increment the count). + */ +int pci_enable_device(struct pci_dev *dev) +{ + int result; + if (atomic_add_return(1, &dev->enable_cnt) > 1) + return 0; /* already enabled */ + result = __pci_enable_device(dev); + if (result < 0) + atomic_dec(&dev->enable_cnt); + return result; +} + /** * pcibios_disable_device - disable arch specific PCI resources for device dev * @dev: the PCI device to disable @@ -651,12 +672,18 @@ void __attribute__ ((weak)) pcibios_disable_device (struct pci_dev *dev) {} * * Signal to the system that the PCI device is not in use by the system * anymore. This only involves disabling PCI bus-mastering, if active. + * + * Note we don't actually disable the device until all callers of + * pci_device_enable() have called pci_device_disable(). */ void pci_disable_device(struct pci_dev *dev) { u16 pci_command; + if (atomic_sub_return(1, &dev->enable_cnt) != 0) + return; + if (dev->msi_enabled) disable_msi_mode(dev, pci_find_capability(dev, PCI_CAP_ID_MSI), PCI_CAP_ID_MSI); @@ -672,7 +699,6 @@ pci_disable_device(struct pci_dev *dev) dev->is_busmaster = 0; pcibios_disable_device(dev); - dev->is_enabled = 0; } /** diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 6bf327db5c5e..398852f526a6 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -1,5 +1,6 @@ /* Functions internal to the PCI core code */ +extern int __must_check __pci_enable_device(struct pci_dev *); extern int pci_uevent(struct device *dev, char **envp, int num_envp, char *buffer, int buffer_size); extern int pci_create_sysfs_dev_files(struct pci_dev *pdev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 09be0f81b27b..01c707261f9c 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -51,6 +51,7 @@ #include #include #include +#include #include /* File state for mmap()s on /proc/bus/pci/X/Y */ @@ -159,7 +160,6 @@ struct pci_dev { unsigned int transparent:1; /* Transparent PCI bridge */ unsigned int multifunction:1;/* Part of multi-function device */ /* keep track of device state */ - unsigned int is_enabled:1; /* pci_enable_device has been called */ unsigned int is_busmaster:1; /* device is busmaster */ unsigned int no_msi:1; /* device may not use msi */ unsigned int no_d1d2:1; /* only allow d0 or d3 */ @@ -167,6 +167,7 @@ struct pci_dev { unsigned int broken_parity_status:1; /* Device generates false positive parity */ unsigned int msi_enabled:1; unsigned int msix_enabled:1; + atomic_t enable_cnt; /* pci_enable_device has been called */ u32 saved_config_space[16]; /* config space saved at suspend time */ struct hlist_head saved_cap_space; -- cgit v1.2.3 From 3b59d52d8c7925e7a9a396f2e31a66eb060c6c37 Mon Sep 17 00:00:00 2001 From: Jason Gaston Date: Wed, 22 Nov 2006 15:15:08 -0800 Subject: PCI: irq: irq and pci_ids patch for Intel ICH9 This updated patch adds the Intel ICH9 LPC and SMBus Controller DID's. Signed-off-by: Jason Gaston Signed-off-by: Greg Kroah-Hartman --- arch/i386/pci/irq.c | 6 ++++++ include/linux/pci_ids.h | 7 +++++++ 2 files changed, 13 insertions(+) (limited to 'include/linux') diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index 69163998adeb..e65551cd8216 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -543,6 +543,12 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH8_2: case PCI_DEVICE_ID_INTEL_ICH8_3: case PCI_DEVICE_ID_INTEL_ICH8_4: + case PCI_DEVICE_ID_INTEL_ICH9_0: + case PCI_DEVICE_ID_INTEL_ICH9_1: + case PCI_DEVICE_ID_INTEL_ICH9_2: + case PCI_DEVICE_ID_INTEL_ICH9_3: + case PCI_DEVICE_ID_INTEL_ICH9_4: + case PCI_DEVICE_ID_INTEL_ICH9_5: r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index fa4e1d799782..e060a7637947 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2211,6 +2211,13 @@ #define PCI_DEVICE_ID_INTEL_ICH8_4 0x2815 #define PCI_DEVICE_ID_INTEL_ICH8_5 0x283e #define PCI_DEVICE_ID_INTEL_ICH8_6 0x2850 +#define PCI_DEVICE_ID_INTEL_ICH9_0 0x2910 +#define PCI_DEVICE_ID_INTEL_ICH9_1 0x2911 +#define PCI_DEVICE_ID_INTEL_ICH9_2 0x2912 +#define PCI_DEVICE_ID_INTEL_ICH9_3 0x2913 +#define PCI_DEVICE_ID_INTEL_ICH9_4 0x2914 +#define PCI_DEVICE_ID_INTEL_ICH9_5 0x2915 +#define PCI_DEVICE_ID_INTEL_ICH9_6 0x2930 #define PCI_DEVICE_ID_INTEL_82855PM_HB 0x3340 #define PCI_DEVICE_ID_INTEL_82830_HB 0x3575 #define PCI_DEVICE_ID_INTEL_82830_CGC 0x3577 -- cgit v1.2.3 From 116af378201ef793424cd10508ccf18b06d8a021 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 25 Oct 2006 13:44:59 +1000 Subject: Driver core: add notification of bus events I finally did as you suggested and added the notifier to the struct bus_type itself. There are still problems to be expected is something attaches to a bus type where the code can hook in different struct device sub-classes (which is imho a big bogosity but I won't even try to argue that case now) but it will solve nicely a number of issues I've had so far. That also means that clients interested in registering for such notifications have to do it before devices are added and after bus types are registered. Fortunately, most bus types that matter for the various usage scenarios I have in mind are registerd at postcore_initcall time, which means I have a really nice spot at arch_initcall time to add my notifiers. There are 4 notifications provided. Device being added (before hooked to the bus) and removed (failure of previous case or after being unhooked from the bus), along with driver being bound to a device and about to be unbound. The usage I have for these are: - The 2 first ones are used to maintain a struct device_ext that is hooked to struct device.firmware_data. This structure contains for now a pointer to the Open Firmware node related to the device (if any), the NUMA node ID (for quick access to it) and the DMA operations pointers & iommu table instance for DMA to/from this device. For bus types I own (like IBM VIO or EBUS), I just maintain that structure directly from the bus code when creating the devices. But for bus types managed by generic code like PCI or platform (actually, of_platform which is a variation of platform linked to Open Firmware device-tree), I need this notifier. - The other two ones have a completely different usage scenario. I have cases where multiple devices and their drivers depend on each other. For example, the IBM EMAC network driver needs to attach to a MAL DMA engine which is a separate device, and a PHY interface which is also a separate device. They are all of_platform_device's (well, about to be with my upcoming patches) but there is no say in what precise order the core will "probe" them and instanciate the various modules. The solution I found for that is to have the drivers for emac to use multithread_probe, and wait for a driver to be bound to the target MAL and PHY control devices (the device-tree contains reference to the MAL and PHY interface nodes, which I can then match to of_platform_devices). Right now, I've been polling, but with that notifier, I can more cleanly wait (with a timeout of course). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 14 ++++++++++++++ drivers/base/core.c | 12 ++++++++++++ drivers/base/dd.c | 10 ++++++++++ include/linux/device.h | 25 +++++++++++++++++++++++++ 4 files changed, 61 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 7d8a7ce73fb3..ed3e8a2be64a 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -724,6 +724,8 @@ int bus_register(struct bus_type * bus) { int retval; + BLOCKING_INIT_NOTIFIER_HEAD(&bus->bus_notifier); + retval = kobject_set_name(&bus->subsys.kset.kobj, "%s", bus->name); if (retval) goto out; @@ -782,6 +784,18 @@ void bus_unregister(struct bus_type * bus) subsystem_unregister(&bus->subsys); } +int bus_register_notifier(struct bus_type *bus, struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&bus->bus_notifier, nb); +} +EXPORT_SYMBOL_GPL(bus_register_notifier); + +int bus_unregister_notifier(struct bus_type *bus, struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&bus->bus_notifier, nb); +} +EXPORT_SYMBOL_GPL(bus_unregister_notifier); + int __init buses_init(void) { return subsystem_register(&bus_subsys); diff --git a/drivers/base/core.c b/drivers/base/core.c index 002fde46d38d..d4f35d8902a2 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -17,6 +17,7 @@ #include #include #include +#include #include @@ -428,6 +429,11 @@ int device_add(struct device *dev) if (platform_notify) platform_notify(dev); + /* notify clients of device entry (new way) */ + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_ADD_DEVICE, dev); + dev->uevent_attr.attr.name = "uevent"; dev->uevent_attr.attr.mode = S_IWUSR; if (dev->driver) @@ -504,6 +510,9 @@ int device_add(struct device *dev) BusError: device_pm_remove(dev); PMError: + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_DEL_DEVICE, dev); device_remove_groups(dev); GroupError: device_remove_attrs(dev); @@ -622,6 +631,9 @@ void device_del(struct device * dev) */ if (platform_notify_remove) platform_notify_remove(dev); + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_DEL_DEVICE, dev); bus_remove_device(dev); device_pm_remove(dev); kobject_uevent(&dev->kobj, KOBJ_REMOVE); diff --git a/drivers/base/dd.c b/drivers/base/dd.c index c5d6bb4290ad..9c88b1e34bc3 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -52,6 +52,11 @@ int device_bind_driver(struct device *dev) pr_debug("bound device '%s' to driver '%s'\n", dev->bus_id, dev->driver->name); + + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_BOUND_DRIVER, dev); + klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices); ret = sysfs_create_link(&dev->driver->kobj, &dev->kobj, kobject_name(&dev->kobj)); @@ -288,6 +293,11 @@ static void __device_release_driver(struct device * dev) sysfs_remove_link(&dev->kobj, "driver"); klist_remove(&dev->knode_driver); + if (dev->bus) + blocking_notifier_call_chain(&dev->bus->bus_notifier, + BUS_NOTIFY_UNBIND_DRIVER, + dev); + if (dev->bus && dev->bus->remove) dev->bus->remove(dev); else if (drv->remove) diff --git a/include/linux/device.h b/include/linux/device.h index 9d4f6a963936..b00e02711393 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -42,6 +42,8 @@ struct bus_type { struct klist klist_devices; struct klist klist_drivers; + struct blocking_notifier_head bus_notifier; + struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; struct driver_attribute * drv_attrs; @@ -75,6 +77,29 @@ int __must_check bus_for_each_drv(struct bus_type *bus, struct device_driver *start, void *data, int (*fn)(struct device_driver *, void *)); +/* + * Bus notifiers: Get notified of addition/removal of devices + * and binding/unbinding of drivers to devices. + * In the long run, it should be a replacement for the platform + * notify hooks. + */ +struct notifier_block; + +extern int bus_register_notifier(struct bus_type *bus, + struct notifier_block *nb); +extern int bus_unregister_notifier(struct bus_type *bus, + struct notifier_block *nb); + +/* All 4 notifers below get called with the target struct device * + * as an argument. Note that those functions are likely to be called + * with the device semaphore held in the core, so be careful. + */ +#define BUS_NOTIFY_ADD_DEVICE 0x00000001 /* device added */ +#define BUS_NOTIFY_DEL_DEVICE 0x00000002 /* device removed */ +#define BUS_NOTIFY_BOUND_DRIVER 0x00000003 /* driver bound to device */ +#define BUS_NOTIFY_UNBIND_DRIVER 0x00000004 /* driver about to be + unbound */ + /* driverfs interface for exporting bus attributes */ struct bus_attribute { -- cgit v1.2.3 From f0ee61a6cecd100301a60d99feb187776533b2a2 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 23 Oct 2006 10:40:54 -0700 Subject: Driver Core: Move virtual_device_parent() to core.c It doesn't need to be global or in device.h Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 17 ----------------- drivers/base/core.c | 17 +++++++++++++++++ include/linux/device.h | 2 -- 3 files changed, 17 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/class.c b/drivers/base/class.c index 0ff267a248db..2e705f6abb4c 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -893,23 +893,6 @@ void class_interface_unregister(struct class_interface *class_intf) class_put(parent); } -int virtual_device_parent(struct device *dev) -{ - if (!dev->class) - return -ENODEV; - - if (!dev->class->virtual_dir) { - static struct kobject *virtual_dir = NULL; - - if (!virtual_dir) - virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual"); - dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name); - } - - dev->kobj.parent = dev->class->virtual_dir; - return 0; -} - int __init classes_init(void) { int retval; diff --git a/drivers/base/core.c b/drivers/base/core.c index d4f35d8902a2..dbcd40b987d2 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -384,6 +384,23 @@ void device_initialize(struct device *dev) device_init_wakeup(dev, 0); } +static int virtual_device_parent(struct device *dev) +{ + if (!dev->class) + return -ENODEV; + + if (!dev->class->virtual_dir) { + static struct kobject *virtual_dir = NULL; + + if (!virtual_dir) + virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual"); + dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name); + } + + dev->kobj.parent = dev->class->virtual_dir; + return 0; +} + /** * device_add - add device to device hierarchy. * @dev: device. diff --git a/include/linux/device.h b/include/linux/device.h index b00e02711393..00b29e0c5ce0 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -440,8 +440,6 @@ extern struct device *device_create(struct class *cls, struct device *parent, __attribute__((format(printf,4,5))); extern void device_destroy(struct class *cls, dev_t devt); -extern int virtual_device_parent(struct device *dev); - /* * Platform "fixup" functions - allow the platform to have their say * about devices and actions that the general device layer doesn't -- cgit v1.2.3 From 94fbcded4ea0dc14cbfb222a5c68372f150d1476 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 27 Jul 2006 16:16:04 -0700 Subject: Driver core: change misc class_devices to be real devices This also ment that some of the misc drivers had to also be fixed up as they were assuming the device was a class_device. Signed-off-by: Greg Kroah-Hartman --- drivers/char/hw_random/core.c | 38 +++++++++++++++++++------------------- drivers/char/misc.c | 13 ++++--------- drivers/char/tpm/tpm.c | 2 +- drivers/input/serio/serio_raw.c | 2 +- include/linux/miscdevice.h | 5 ++--- 5 files changed, 27 insertions(+), 33 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index 154a81d328c1..ebace201bec6 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -162,7 +162,8 @@ static struct miscdevice rng_miscdev = { }; -static ssize_t hwrng_attr_current_store(struct class_device *class, +static ssize_t hwrng_attr_current_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) { int err; @@ -192,7 +193,8 @@ static ssize_t hwrng_attr_current_store(struct class_device *class, return err ? : len; } -static ssize_t hwrng_attr_current_show(struct class_device *class, +static ssize_t hwrng_attr_current_show(struct device *dev, + struct device_attribute *attr, char *buf) { int err; @@ -210,7 +212,8 @@ static ssize_t hwrng_attr_current_show(struct class_device *class, return ret; } -static ssize_t hwrng_attr_available_show(struct class_device *class, +static ssize_t hwrng_attr_available_show(struct device *dev, + struct device_attribute *attr, char *buf) { int err; @@ -234,20 +237,18 @@ static ssize_t hwrng_attr_available_show(struct class_device *class, return ret; } -static CLASS_DEVICE_ATTR(rng_current, S_IRUGO | S_IWUSR, - hwrng_attr_current_show, - hwrng_attr_current_store); -static CLASS_DEVICE_ATTR(rng_available, S_IRUGO, - hwrng_attr_available_show, - NULL); +static DEVICE_ATTR(rng_current, S_IRUGO | S_IWUSR, + hwrng_attr_current_show, + hwrng_attr_current_store); +static DEVICE_ATTR(rng_available, S_IRUGO, + hwrng_attr_available_show, + NULL); static void unregister_miscdev(void) { - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_available); - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_current); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_available); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_current); misc_deregister(&rng_miscdev); } @@ -258,20 +259,19 @@ static int register_miscdev(void) err = misc_register(&rng_miscdev); if (err) goto out; - err = class_device_create_file(rng_miscdev.class, - &class_device_attr_rng_current); + err = device_create_file(rng_miscdev.this_device, + &dev_attr_rng_current); if (err) goto err_misc_dereg; - err = class_device_create_file(rng_miscdev.class, - &class_device_attr_rng_available); + err = device_create_file(rng_miscdev.this_device, + &dev_attr_rng_available); if (err) goto err_remove_current; out: return err; err_remove_current: - class_device_remove_file(rng_miscdev.class, - &class_device_attr_rng_current); + device_remove_file(rng_miscdev.this_device, &dev_attr_rng_current); err_misc_dereg: misc_deregister(&rng_miscdev); goto out; diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 62ebe09656e3..7a484fc7cb9e 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -169,11 +169,6 @@ fail: return err; } -/* - * TODO for 2.7: - * - add a struct kref to struct miscdevice and make all usages of - * them dynamic. - */ static struct class *misc_class; static const struct file_operations misc_fops = { @@ -228,10 +223,10 @@ int misc_register(struct miscdevice * misc) misc_minors[misc->minor >> 3] |= 1 << (misc->minor & 7); dev = MKDEV(MISC_MAJOR, misc->minor); - misc->class = class_device_create(misc_class, NULL, dev, misc->dev, + misc->this_device = device_create(misc_class, misc->parent, dev, "%s", misc->name); - if (IS_ERR(misc->class)) { - err = PTR_ERR(misc->class); + if (IS_ERR(misc->this_device)) { + err = PTR_ERR(misc->this_device); goto out; } @@ -264,7 +259,7 @@ int misc_deregister(struct miscdevice * misc) down(&misc_sem); list_del(&misc->list); - class_device_destroy(misc_class, MKDEV(MISC_MAJOR, misc->minor)); + device_destroy(misc_class, MKDEV(MISC_MAJOR, misc->minor)); if (i < DYNAMIC_MINORS && i>0) { misc_minors[i>>3] &= ~(1 << (misc->minor & 7)); } diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 6ad2d3bb945c..6e1329d404d2 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -1130,7 +1130,7 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend scnprintf(devname, DEVNAME_SIZE, "%s%d", "tpm", chip->dev_num); chip->vendor.miscdev.name = devname; - chip->vendor.miscdev.dev = dev; + chip->vendor.miscdev.parent = dev; chip->dev = get_device(dev); if (misc_register(&chip->vendor.miscdev)) { diff --git a/drivers/input/serio/serio_raw.c b/drivers/input/serio/serio_raw.c index ba2a2035d648..7c8d0399ae82 100644 --- a/drivers/input/serio/serio_raw.c +++ b/drivers/input/serio/serio_raw.c @@ -297,7 +297,7 @@ static int serio_raw_connect(struct serio *serio, struct serio_driver *drv) serio_raw->dev.minor = PSMOUSE_MINOR; serio_raw->dev.name = serio_raw->name; - serio_raw->dev.dev = &serio->dev; + serio_raw->dev.parent = &serio->dev; serio_raw->dev.fops = &serio_raw_fops; err = misc_register(&serio_raw->dev); diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index b03cfb91e228..326da7d500c7 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -31,15 +31,14 @@ #define HPET_MINOR 228 struct device; -struct class_device; struct miscdevice { int minor; const char *name; const struct file_operations *fops; struct list_head list; - struct device *dev; - struct class_device *class; + struct device *parent; + struct device *this_device; }; extern int misc_register(struct miscdevice * misc); -- cgit v1.2.3 From 01107d343076c34b9e1ce5d073292bd7f3097fda Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 7 Aug 2006 22:19:37 -0700 Subject: Driver core: convert tty core to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Also fixes up the isdn drivers that were putting something in the class device's directory. Signed-off-by: Greg Kroah-Hartman --- drivers/char/tty_io.c | 19 ++++++++++--------- drivers/isdn/gigaset/common.c | 2 +- drivers/isdn/gigaset/gigaset.h | 2 +- drivers/isdn/gigaset/interface.c | 10 +++++----- drivers/isdn/gigaset/proc.c | 19 ++++++++++--------- include/linux/tty.h | 5 ++--- 6 files changed, 29 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index e90ea39c7c4b..50dc49205a23 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -3612,7 +3612,8 @@ static struct class *tty_class; * This field is optional, if there is no known struct device * for this tty device it can be set to NULL safely. * - * Returns a pointer to the class device (or ERR_PTR(-EFOO) on error). + * Returns a pointer to the struct device for this tty device + * (or ERR_PTR(-EFOO) on error). * * This call is required to be made to register an individual tty device * if the tty driver's flags have the TTY_DRIVER_DYNAMIC_DEV bit set. If @@ -3622,8 +3623,8 @@ static struct class *tty_class; * Locking: ?? */ -struct class_device *tty_register_device(struct tty_driver *driver, - unsigned index, struct device *device) +struct device *tty_register_device(struct tty_driver *driver, unsigned index, + struct device *device) { char name[64]; dev_t dev = MKDEV(driver->major, driver->minor_start) + index; @@ -3639,7 +3640,7 @@ struct class_device *tty_register_device(struct tty_driver *driver, else tty_line_name(driver, index, name); - return class_device_create(tty_class, NULL, dev, device, "%s", name); + return device_create(tty_class, device, dev, name); } /** @@ -3655,7 +3656,7 @@ struct class_device *tty_register_device(struct tty_driver *driver, void tty_unregister_device(struct tty_driver *driver, unsigned index) { - class_device_destroy(tty_class, MKDEV(driver->major, driver->minor_start) + index); + device_destroy(tty_class, MKDEV(driver->major, driver->minor_start) + index); } EXPORT_SYMBOL(tty_register_device); @@ -3895,20 +3896,20 @@ static int __init tty_init(void) if (cdev_add(&tty_cdev, MKDEV(TTYAUX_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 0), 1, "/dev/tty") < 0) panic("Couldn't register /dev/tty driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), NULL, "tty"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), "tty"); cdev_init(&console_cdev, &console_fops); if (cdev_add(&console_cdev, MKDEV(TTYAUX_MAJOR, 1), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 1), 1, "/dev/console") < 0) panic("Couldn't register /dev/console driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), NULL, "console"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), "console"); #ifdef CONFIG_UNIX98_PTYS cdev_init(&ptmx_cdev, &ptmx_fops); if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0) panic("Couldn't register /dev/ptmx driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx"); + device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), "ptmx"); #endif #ifdef CONFIG_VT @@ -3916,7 +3917,7 @@ static int __init tty_init(void) if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0) panic("Couldn't register /dev/tty0 driver\n"); - class_device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0"); + device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), "tty0"); vty_init(); #endif diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index 5800beeebb85..defd5743dba6 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -702,7 +702,7 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, cs->open_count = 0; cs->dev = NULL; cs->tty = NULL; - cs->class = NULL; + cs->tty_dev = NULL; cs->cidmode = cidmode != 0; //if(onechannel) { //FIXME diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h index 884bd72c1bf4..06298cc52bf5 100644 --- a/drivers/isdn/gigaset/gigaset.h +++ b/drivers/isdn/gigaset/gigaset.h @@ -444,7 +444,7 @@ struct cardstate { struct gigaset_driver *driver; unsigned minor_index; struct device *dev; - struct class_device *class; + struct device *tty_dev; const struct gigaset_ops *ops; diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index 596f3aebe2f7..7edea015867e 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -625,13 +625,13 @@ void gigaset_if_init(struct cardstate *cs) return; tasklet_init(&cs->if_wake_tasklet, &if_wake, (unsigned long) cs); - cs->class = tty_register_device(drv->tty, cs->minor_index, NULL); + cs->tty_dev = tty_register_device(drv->tty, cs->minor_index, NULL); - if (!IS_ERR(cs->class)) - class_set_devdata(cs->class, cs); + if (!IS_ERR(cs->tty_dev)) + dev_set_drvdata(cs->tty_dev, cs); else { warn("could not register device to the tty subsystem"); - cs->class = NULL; + cs->tty_dev = NULL; } } @@ -645,7 +645,7 @@ void gigaset_if_free(struct cardstate *cs) tasklet_disable(&cs->if_wake_tasklet); tasklet_kill(&cs->if_wake_tasklet); - cs->class = NULL; + cs->tty_dev = NULL; tty_unregister_device(drv->tty, cs->minor_index); } diff --git a/drivers/isdn/gigaset/proc.c b/drivers/isdn/gigaset/proc.c index 9ad840e95dbe..e767afa55abf 100644 --- a/drivers/isdn/gigaset/proc.c +++ b/drivers/isdn/gigaset/proc.c @@ -16,11 +16,12 @@ #include "gigaset.h" #include -static ssize_t show_cidmode(struct class_device *class, char *buf) +static ssize_t show_cidmode(struct device *dev, + struct device_attribute *attr, char *buf) { int ret; unsigned long flags; - struct cardstate *cs = class_get_devdata(class); + struct cardstate *cs = dev_get_drvdata(dev); spin_lock_irqsave(&cs->lock, flags); ret = sprintf(buf, "%u\n", cs->cidmode); @@ -29,10 +30,10 @@ static ssize_t show_cidmode(struct class_device *class, char *buf) return ret; } -static ssize_t set_cidmode(struct class_device *class, +static ssize_t set_cidmode(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - struct cardstate *cs = class_get_devdata(class); + struct cardstate *cs = dev_get_drvdata(dev); long int value; char *end; @@ -64,25 +65,25 @@ static ssize_t set_cidmode(struct class_device *class, return count; } -static CLASS_DEVICE_ATTR(cidmode, S_IRUGO|S_IWUSR, show_cidmode, set_cidmode); +static DEVICE_ATTR(cidmode, S_IRUGO|S_IWUSR, show_cidmode, set_cidmode); /* free sysfs for device */ void gigaset_free_dev_sysfs(struct cardstate *cs) { - if (!cs->class) + if (!cs->tty_dev) return; gig_dbg(DEBUG_INIT, "removing sysfs entries"); - class_device_remove_file(cs->class, &class_device_attr_cidmode); + device_remove_file(cs->tty_dev, &dev_attr_cidmode); } /* initialize sysfs for device */ void gigaset_init_dev_sysfs(struct cardstate *cs) { - if (!cs->class) + if (!cs->tty_dev) return; gig_dbg(DEBUG_INIT, "setting up sysfs"); - if (class_device_create_file(cs->class, &class_device_attr_cidmode)) + if (device_create_file(cs->tty_dev, &dev_attr_cidmode)) dev_err(cs->dev, "could not create sysfs attribute\n"); } diff --git a/include/linux/tty.h b/include/linux/tty.h index 44091c0db0b4..65321f911c1e 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -276,9 +276,8 @@ extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc); extern int tty_unregister_ldisc(int disc); extern int tty_register_driver(struct tty_driver *driver); extern int tty_unregister_driver(struct tty_driver *driver); -extern struct class_device *tty_register_device(struct tty_driver *driver, - unsigned index, - struct device *dev); +extern struct device *tty_register_device(struct tty_driver *driver, + unsigned index, struct device *dev); extern void tty_unregister_device(struct tty_driver *driver, unsigned index); extern int tty_read_raw_data(struct tty_struct *tty, unsigned char *bufp, int buflen); -- cgit v1.2.3 From fcaf71fd51f9cfc504455d3e19ec242e4b2073ed Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 12 Sep 2006 17:00:10 +0200 Subject: Driver core: convert mmc code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- drivers/mmc/mmc_queue.c | 4 ++-- drivers/mmc/mmc_sysfs.c | 20 ++++++++++---------- drivers/mmc/wbsd.c | 6 +++--- include/linux/mmc/host.h | 8 ++++---- 4 files changed, 19 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/drivers/mmc/mmc_queue.c b/drivers/mmc/mmc_queue.c index 4ccdd82b680f..61a1de85cb23 100644 --- a/drivers/mmc/mmc_queue.c +++ b/drivers/mmc/mmc_queue.c @@ -130,8 +130,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock u64 limit = BLK_BOUNCE_HIGH; int ret; - if (host->dev->dma_mask && *host->dev->dma_mask) - limit = *host->dev->dma_mask; + if (mmc_dev(host)->dma_mask && *mmc_dev(host)->dma_mask) + limit = *mmc_dev(host)->dma_mask; mq->card = card; mq->queue = blk_init_queue(mmc_request, lock); diff --git a/drivers/mmc/mmc_sysfs.c b/drivers/mmc/mmc_sysfs.c index 10cc9734eaa0..ac5329636045 100644 --- a/drivers/mmc/mmc_sysfs.c +++ b/drivers/mmc/mmc_sysfs.c @@ -199,7 +199,7 @@ void mmc_init_card(struct mmc_card *card, struct mmc_host *host) memset(card, 0, sizeof(struct mmc_card)); card->host = host; device_initialize(&card->dev); - card->dev.parent = card->host->dev; + card->dev.parent = mmc_dev(host); card->dev.bus = &mmc_bus_type; card->dev.release = mmc_release_card; } @@ -242,7 +242,7 @@ void mmc_remove_card(struct mmc_card *card) } -static void mmc_host_classdev_release(struct class_device *dev) +static void mmc_host_classdev_release(struct device *dev) { struct mmc_host *host = cls_dev_to_mmc_host(dev); kfree(host); @@ -250,7 +250,7 @@ static void mmc_host_classdev_release(struct class_device *dev) static struct class mmc_host_class = { .name = "mmc_host", - .release = mmc_host_classdev_release, + .dev_release = mmc_host_classdev_release, }; static DEFINE_IDR(mmc_host_idr); @@ -267,10 +267,10 @@ struct mmc_host *mmc_alloc_host_sysfs(int extra, struct device *dev) if (host) { memset(host, 0, sizeof(struct mmc_host) + extra); - host->dev = dev; - host->class_dev.dev = host->dev; + host->parent = dev; + host->class_dev.parent = dev; host->class_dev.class = &mmc_host_class; - class_device_initialize(&host->class_dev); + device_initialize(&host->class_dev); } return host; @@ -292,10 +292,10 @@ int mmc_add_host_sysfs(struct mmc_host *host) if (err) return err; - snprintf(host->class_dev.class_id, BUS_ID_SIZE, + snprintf(host->class_dev.bus_id, BUS_ID_SIZE, "mmc%d", host->index); - return class_device_add(&host->class_dev); + return device_add(&host->class_dev); } /* @@ -303,7 +303,7 @@ int mmc_add_host_sysfs(struct mmc_host *host) */ void mmc_remove_host_sysfs(struct mmc_host *host) { - class_device_del(&host->class_dev); + device_del(&host->class_dev); spin_lock(&mmc_host_lock); idr_remove(&mmc_host_idr, host->index); @@ -315,7 +315,7 @@ void mmc_remove_host_sysfs(struct mmc_host *host) */ void mmc_free_host_sysfs(struct mmc_host *host) { - class_device_put(&host->class_dev); + put_device(&host->class_dev); } static struct workqueue_struct *workqueue; diff --git a/drivers/mmc/wbsd.c b/drivers/mmc/wbsd.c index ced309b37a8f..682e62b0b09d 100644 --- a/drivers/mmc/wbsd.c +++ b/drivers/mmc/wbsd.c @@ -1488,7 +1488,7 @@ static void __devinit wbsd_request_dma(struct wbsd_host *host, int dma) /* * Translate the address to a physical address. */ - host->dma_addr = dma_map_single(host->mmc->dev, host->dma_buffer, + host->dma_addr = dma_map_single(mmc_dev(host->mmc), host->dma_buffer, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); /* @@ -1512,7 +1512,7 @@ kfree: */ BUG_ON(1); - dma_unmap_single(host->mmc->dev, host->dma_addr, + dma_unmap_single(mmc_dev(host->mmc), host->dma_addr, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); host->dma_addr = (dma_addr_t)NULL; @@ -1530,7 +1530,7 @@ err: static void __devexit wbsd_release_dma(struct wbsd_host *host) { if (host->dma_addr) { - dma_unmap_single(host->mmc->dev, host->dma_addr, + dma_unmap_single(mmc_dev(host->mmc), host->dma_addr, WBSD_DMA_SIZE, DMA_BIDIRECTIONAL); } kfree(host->dma_buffer); diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index 587264a58d56..528e7d3fecb1 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -74,8 +74,8 @@ struct mmc_card; struct device; struct mmc_host { - struct device *dev; - struct class_device class_dev; + struct device *parent; + struct device class_dev; int index; const struct mmc_host_ops *ops; unsigned int f_min; @@ -125,8 +125,8 @@ static inline void *mmc_priv(struct mmc_host *host) return (void *)host->private; } -#define mmc_dev(x) ((x)->dev) -#define mmc_hostname(x) ((x)->class_dev.class_id) +#define mmc_dev(x) ((x)->parent) +#define mmc_hostname(x) ((x)->class_dev.bus_id) extern int mmc_suspend_host(struct mmc_host *, pm_message_t); extern int mmc_resume_host(struct mmc_host *); -- cgit v1.2.3 From 78cde0887930f5d11a56fc51b013f2672fba0e6f Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 14 Sep 2006 07:30:59 -0700 Subject: Driver core: convert fb code to use struct device Converts from using struct "class_device" to "struct device" making everything show up properly in /sys/devices/ with symlinks from the /sys/class directory. Signed-off-by: Greg Kroah-Hartman --- drivers/video/fbmem.c | 16 ++--- drivers/video/fbsysfs.c | 163 +++++++++++++++++++++++++++--------------------- include/linux/fb.h | 8 +-- 3 files changed, 103 insertions(+), 84 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 93ffcdd95f50..e973a87fbb01 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -1296,14 +1296,14 @@ register_framebuffer(struct fb_info *fb_info) break; fb_info->node = i; - fb_info->class_device = class_device_create(fb_class, NULL, MKDEV(FB_MAJOR, i), - fb_info->device, "fb%d", i); - if (IS_ERR(fb_info->class_device)) { + fb_info->dev = device_create(fb_class, fb_info->device, + MKDEV(FB_MAJOR, i), "fb%d", i); + if (IS_ERR(fb_info->dev)) { /* Not fatal */ - printk(KERN_WARNING "Unable to create class_device for framebuffer %d; errno = %ld\n", i, PTR_ERR(fb_info->class_device)); - fb_info->class_device = NULL; + printk(KERN_WARNING "Unable to create device for framebuffer %d; errno = %ld\n", i, PTR_ERR(fb_info->dev)); + fb_info->dev = NULL; } else - fb_init_class_device(fb_info); + fb_init_device(fb_info); if (fb_info->pixmap.addr == NULL) { fb_info->pixmap.addr = kmalloc(FBPIXMAPSIZE, GFP_KERNEL); @@ -1356,8 +1356,8 @@ unregister_framebuffer(struct fb_info *fb_info) fb_destroy_modelist(&fb_info->modelist); registered_fb[i]=NULL; num_registered_fb--; - fb_cleanup_class_device(fb_info); - class_device_destroy(fb_class, MKDEV(FB_MAJOR, i)); + fb_cleanup_device(fb_info); + device_destroy(fb_class, MKDEV(FB_MAJOR, i)); event.info = fb_info; fb_notifier_call_chain(FB_EVENT_FB_UNREGISTERED, &event); return 0; diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c index d3a50417ed9a..323bdf6fc7d5 100644 --- a/drivers/video/fbsysfs.c +++ b/drivers/video/fbsysfs.c @@ -73,7 +73,7 @@ EXPORT_SYMBOL(framebuffer_alloc); * * @info: frame buffer info structure * - * Drop the reference count of the class_device embedded in the + * Drop the reference count of the device embedded in the * framebuffer info structure. * */ @@ -120,10 +120,10 @@ static int mode_string(char *buf, unsigned int offset, m, mode->xres, mode->yres, v, mode->refresh); } -static ssize_t store_mode(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_mode(struct device *device, struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); char mstr[100]; struct fb_var_screeninfo var; struct fb_modelist *modelist; @@ -151,9 +151,10 @@ static ssize_t store_mode(struct class_device *class_device, const char * buf, return -EINVAL; } -static ssize_t show_mode(struct class_device *class_device, char *buf) +static ssize_t show_mode(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); if (!fb_info->mode) return 0; @@ -161,10 +162,11 @@ static ssize_t show_mode(struct class_device *class_device, char *buf) return mode_string(buf, 0, fb_info->mode); } -static ssize_t store_modes(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_modes(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); LIST_HEAD(old_list); int i = count / sizeof(struct fb_videomode); @@ -186,9 +188,10 @@ static ssize_t store_modes(struct class_device *class_device, const char * buf, return 0; } -static ssize_t show_modes(struct class_device *class_device, char *buf) +static ssize_t show_modes(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); unsigned int i; struct list_head *pos; struct fb_modelist *modelist; @@ -203,10 +206,10 @@ static ssize_t show_modes(struct class_device *class_device, char *buf) return i; } -static ssize_t store_bpp(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_bpp(struct device *device, struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char ** last = NULL; int err; @@ -218,16 +221,18 @@ static ssize_t store_bpp(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_bpp(struct class_device *class_device, char *buf) +static ssize_t show_bpp(struct device *device, struct device_attribute *attr, + char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->var.bits_per_pixel); } -static ssize_t store_rotate(struct class_device *class_device, const char *buf, - size_t count) +static ssize_t store_rotate(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char **last = NULL; int err; @@ -242,17 +247,19 @@ static ssize_t store_rotate(struct class_device *class_device, const char *buf, } -static ssize_t show_rotate(struct class_device *class_device, char *buf) +static ssize_t show_rotate(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->var.rotate); } -static ssize_t store_virtual(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_virtual(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char *last = NULL; int err; @@ -269,23 +276,26 @@ static ssize_t store_virtual(struct class_device *class_device, return count; } -static ssize_t show_virtual(struct class_device *class_device, char *buf) +static ssize_t show_virtual(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d,%d\n", fb_info->var.xres_virtual, fb_info->var.yres_virtual); } -static ssize_t show_stride(struct class_device *class_device, char *buf) +static ssize_t show_stride(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->fix.line_length); } -static ssize_t store_blank(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_blank(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); char *last = NULL; int err; @@ -299,42 +309,48 @@ static ssize_t store_blank(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_blank(struct class_device *class_device, char *buf) +static ssize_t show_blank(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_console(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_console(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t show_console(struct class_device *class_device, char *buf) +static ssize_t show_console(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_cursor(struct class_device *class_device, - const char * buf, size_t count) +static ssize_t store_cursor(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t show_cursor(struct class_device *class_device, char *buf) +static ssize_t show_cursor(struct device *device, + struct device_attribute *attr, char *buf) { -// struct fb_info *fb_info = class_get_devdata(class_device); +// struct fb_info *fb_info = dev_get_drvdata(device); return 0; } -static ssize_t store_pan(struct class_device *class_device, const char * buf, - size_t count) +static ssize_t store_pan(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); struct fb_var_screeninfo var; char *last = NULL; int err; @@ -355,24 +371,27 @@ static ssize_t store_pan(struct class_device *class_device, const char * buf, return count; } -static ssize_t show_pan(struct class_device *class_device, char *buf) +static ssize_t show_pan(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d,%d\n", fb_info->var.xoffset, fb_info->var.xoffset); } -static ssize_t show_name(struct class_device *class_device, char *buf) +static ssize_t show_name(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%s\n", fb_info->fix.id); } -static ssize_t store_fbstate(struct class_device *class_device, - const char *buf, size_t count) +static ssize_t store_fbstate(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); u32 state; char *last = NULL; @@ -385,17 +404,19 @@ static ssize_t store_fbstate(struct class_device *class_device, return count; } -static ssize_t show_fbstate(struct class_device *class_device, char *buf) +static ssize_t show_fbstate(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); return snprintf(buf, PAGE_SIZE, "%d\n", fb_info->state); } #ifdef CONFIG_FB_BACKLIGHT -static ssize_t store_bl_curve(struct class_device *class_device, - const char *buf, size_t count) +static ssize_t store_bl_curve(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); u8 tmp_curve[FB_BACKLIGHT_LEVELS]; unsigned int i; @@ -432,9 +453,10 @@ static ssize_t store_bl_curve(struct class_device *class_device, return count; } -static ssize_t show_bl_curve(struct class_device *class_device, char *buf) +static ssize_t show_bl_curve(struct device *device, + struct device_attribute *attr, char *buf) { - struct fb_info *fb_info = class_get_devdata(class_device); + struct fb_info *fb_info = dev_get_drvdata(device); ssize_t len = 0; unsigned int i; @@ -465,7 +487,7 @@ static ssize_t show_bl_curve(struct class_device *class_device, char *buf) /* When cmap is added back in it should be a binary attribute * not a text one. Consideration should also be given to converting * fbdev to use configfs instead of sysfs */ -static struct class_device_attribute class_device_attrs[] = { +static struct device_attribute device_attrs[] = { __ATTR(bits_per_pixel, S_IRUGO|S_IWUSR, show_bpp, store_bpp), __ATTR(blank, S_IRUGO|S_IWUSR, show_blank, store_blank), __ATTR(console, S_IRUGO|S_IWUSR, show_console, store_console), @@ -483,17 +505,16 @@ static struct class_device_attribute class_device_attrs[] = { #endif }; -int fb_init_class_device(struct fb_info *fb_info) +int fb_init_device(struct fb_info *fb_info) { int i, error = 0; - class_set_devdata(fb_info->class_device, fb_info); + dev_set_drvdata(fb_info->dev, fb_info); fb_info->class_flag |= FB_SYSFS_FLAG_ATTR; - for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) { - error = class_device_create_file(fb_info->class_device, - &class_device_attrs[i]); + for (i = 0; i < ARRAY_SIZE(device_attrs); i++) { + error = device_create_file(fb_info->dev, &device_attrs[i]); if (error) break; @@ -501,22 +522,20 @@ int fb_init_class_device(struct fb_info *fb_info) if (error) { while (--i >= 0) - class_device_remove_file(fb_info->class_device, - &class_device_attrs[i]); + device_remove_file(fb_info->dev, &device_attrs[i]); fb_info->class_flag &= ~FB_SYSFS_FLAG_ATTR; } return 0; } -void fb_cleanup_class_device(struct fb_info *fb_info) +void fb_cleanup_device(struct fb_info *fb_info) { unsigned int i; if (fb_info->class_flag & FB_SYSFS_FLAG_ATTR) { - for (i = 0; i < ARRAY_SIZE(class_device_attrs); i++) - class_device_remove_file(fb_info->class_device, - &class_device_attrs[i]); + for (i = 0; i < ARRAY_SIZE(device_attrs); i++) + device_remove_file(fb_info->dev, &device_attrs[i]); fb_info->class_flag &= ~FB_SYSFS_FLAG_ATTR; } diff --git a/include/linux/fb.h b/include/linux/fb.h index 3e69241e6a81..fa23e0671bb3 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -774,8 +774,8 @@ struct fb_info { #endif struct fb_ops *fbops; - struct device *device; - struct class_device *class_device; /* sysfs per device attrs */ + struct device *device; /* This is the parent */ + struct device *dev; /* This is this fb device */ int class_flag; /* private sysfs flags */ #ifdef CONFIG_FB_TILEBLITTING struct fb_tile_ops *tileops; /* Tile Blitting */ @@ -910,8 +910,8 @@ static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch, /* drivers/video/fbsysfs.c */ extern struct fb_info *framebuffer_alloc(size_t size, struct device *dev); extern void framebuffer_release(struct fb_info *info); -extern int fb_init_class_device(struct fb_info *fb_info); -extern void fb_cleanup_class_device(struct fb_info *head); +extern int fb_init_device(struct fb_info *fb_info); +extern void fb_cleanup_device(struct fb_info *head); extern void fb_bl_default_curve(struct fb_info *fb_info, u8 off, u8 min, u8 max); /* drivers/video/fbmon.c */ -- cgit v1.2.3 From c6dbaef22a2f78700e242915a13218dd780c89ff Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sat, 11 Nov 2006 17:18:39 +1100 Subject: Driver core: add dev_archdata to struct device Add arch specific dev_archdata to struct device Adds an arch specific struct dev_arch to struct device. This enables architecture to add specific fields to every device in the system, like DMA operation pointers, NUMA node ID, firmware specific data, etc... Signed-off-by: Benjamin Herrenschmidt Acked-by: Andi Kleen Acked-By: David Howells Signed-off-by: Greg Kroah-Hartman --- include/asm-alpha/device.h | 7 +++++++ include/asm-arm/device.h | 7 +++++++ include/asm-arm26/device.h | 7 +++++++ include/asm-avr32/device.h | 7 +++++++ include/asm-cris/device.h | 7 +++++++ include/asm-frv/device.h | 7 +++++++ include/asm-generic/device.h | 12 ++++++++++++ include/asm-h8300/device.h | 7 +++++++ include/asm-i386/device.h | 7 +++++++ include/asm-ia64/device.h | 7 +++++++ include/asm-m32r/device.h | 7 +++++++ include/asm-m68k/device.h | 7 +++++++ include/asm-m68knommu/device.h | 7 +++++++ include/asm-mips/device.h | 7 +++++++ include/asm-parisc/device.h | 7 +++++++ include/asm-powerpc/device.h | 7 +++++++ include/asm-ppc/device.h | 7 +++++++ include/asm-s390/device.h | 7 +++++++ include/asm-sh/device.h | 7 +++++++ include/asm-sh64/device.h | 7 +++++++ include/asm-sparc/device.h | 7 +++++++ include/asm-sparc64/device.h | 7 +++++++ include/asm-um/device.h | 7 +++++++ include/asm-v850/device.h | 7 +++++++ include/asm-x86_64/device.h | 7 +++++++ include/asm-xtensa/device.h | 7 +++++++ include/linux/device.h | 3 +++ 27 files changed, 190 insertions(+) create mode 100644 include/asm-alpha/device.h create mode 100644 include/asm-arm/device.h create mode 100644 include/asm-arm26/device.h create mode 100644 include/asm-avr32/device.h create mode 100644 include/asm-cris/device.h create mode 100644 include/asm-frv/device.h create mode 100644 include/asm-generic/device.h create mode 100644 include/asm-h8300/device.h create mode 100644 include/asm-i386/device.h create mode 100644 include/asm-ia64/device.h create mode 100644 include/asm-m32r/device.h create mode 100644 include/asm-m68k/device.h create mode 100644 include/asm-m68knommu/device.h create mode 100644 include/asm-mips/device.h create mode 100644 include/asm-parisc/device.h create mode 100644 include/asm-powerpc/device.h create mode 100644 include/asm-ppc/device.h create mode 100644 include/asm-s390/device.h create mode 100644 include/asm-sh/device.h create mode 100644 include/asm-sh64/device.h create mode 100644 include/asm-sparc/device.h create mode 100644 include/asm-sparc64/device.h create mode 100644 include/asm-um/device.h create mode 100644 include/asm-v850/device.h create mode 100644 include/asm-x86_64/device.h create mode 100644 include/asm-xtensa/device.h (limited to 'include/linux') diff --git a/include/asm-alpha/device.h b/include/asm-alpha/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-alpha/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-arm/device.h b/include/asm-arm/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-arm/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-arm26/device.h b/include/asm-arm26/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-arm26/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-avr32/device.h b/include/asm-avr32/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-avr32/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-cris/device.h b/include/asm-cris/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-cris/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-frv/device.h b/include/asm-frv/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-frv/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-generic/device.h b/include/asm-generic/device.h new file mode 100644 index 000000000000..c17c9600f220 --- /dev/null +++ b/include/asm-generic/device.h @@ -0,0 +1,12 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#ifndef _ASM_GENERIC_DEVICE_H +#define _ASM_GENERIC_DEVICE_H + +struct dev_archdata { +}; + +#endif /* _ASM_GENERIC_DEVICE_H */ diff --git a/include/asm-h8300/device.h b/include/asm-h8300/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-h8300/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-i386/device.h b/include/asm-i386/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-i386/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-ia64/device.h b/include/asm-ia64/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-ia64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m32r/device.h b/include/asm-m32r/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-m32r/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m68k/device.h b/include/asm-m68k/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-m68k/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-m68knommu/device.h b/include/asm-m68knommu/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-m68knommu/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-mips/device.h b/include/asm-mips/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-mips/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-parisc/device.h b/include/asm-parisc/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-parisc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-powerpc/device.h b/include/asm-powerpc/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-powerpc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-ppc/device.h b/include/asm-ppc/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-ppc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-s390/device.h b/include/asm-s390/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-s390/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sh/device.h b/include/asm-sh/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-sh/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sh64/device.h b/include/asm-sh64/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-sh64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sparc/device.h b/include/asm-sparc/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-sparc/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-sparc64/device.h b/include/asm-sparc64/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-sparc64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-um/device.h b/include/asm-um/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-um/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-v850/device.h b/include/asm-v850/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-v850/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-x86_64/device.h b/include/asm-x86_64/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-x86_64/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/asm-xtensa/device.h b/include/asm-xtensa/device.h new file mode 100644 index 000000000000..d8f9872b0e2d --- /dev/null +++ b/include/asm-xtensa/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include + diff --git a/include/linux/device.h b/include/linux/device.h index 00b29e0c5ce0..5b54d756cd54 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -21,6 +21,7 @@ #include #include #include +#include #define DEVICE_NAME_SIZE 50 #define DEVICE_NAME_HALF __stringify(20) /* Less than half to accommodate slop */ @@ -383,6 +384,8 @@ struct device { struct dma_coherent_mem *dma_mem; /* internal for coherent mem override */ + /* arch specific additions */ + struct dev_archdata archdata; /* class_device migration path */ struct list_head node; -- cgit v1.2.3 From 465ae641e4a3e5028aa9c85d3843259aa28a22ce Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sat, 11 Nov 2006 17:18:42 +1100 Subject: ACPI: Change ACPI to use dev_archdata instead of firmware_data Change ACPI to use dev_archdata instead of firmware_data This patch changes ACPI to use the new dev_archdata on i386, x86_64 and ia64 (is there any other arch using ACPI ?) to store it's acpi_handle. It also removes the firmware_data field from struct device as this was the only user. Only build-tested on x86 Signed-off-by: Benjamin Herrenschmidt Cc: Len Brown Signed-off-by: Greg Kroah-Hartman --- drivers/acpi/glue.c | 20 +++++++++++--------- include/acpi/acpi_bus.h | 2 +- include/asm-i386/device.h | 10 +++++++++- include/asm-ia64/device.h | 10 +++++++++- include/asm-x86_64/device.h | 10 +++++++++- include/linux/device.h | 2 -- 6 files changed, 39 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 10f160dc75b1..a2f46d587d55 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -267,9 +267,9 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) { acpi_status status; - if (dev->firmware_data) { + if (dev->archdata.acpi_handle) { printk(KERN_WARNING PREFIX - "Drivers changed 'firmware_data' for %s\n", dev->bus_id); + "Drivers changed 'acpi_handle' for %s\n", dev->bus_id); return -EINVAL; } get_device(dev); @@ -278,25 +278,26 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) put_device(dev); return -EINVAL; } - dev->firmware_data = handle; + dev->archdata.acpi_handle = handle; return 0; } static int acpi_unbind_one(struct device *dev) { - if (!dev->firmware_data) + if (!dev->archdata.acpi_handle) return 0; - if (dev == acpi_get_physical_device(dev->firmware_data)) { + if (dev == acpi_get_physical_device(dev->archdata.acpi_handle)) { /* acpi_get_physical_device increase refcnt by one */ put_device(dev); - acpi_detach_data(dev->firmware_data, acpi_glue_data_handler); - dev->firmware_data = NULL; + acpi_detach_data(dev->archdata.acpi_handle, + acpi_glue_data_handler); + dev->archdata.acpi_handle = NULL; /* acpi_bind_one increase refcnt by one */ put_device(dev); } else { printk(KERN_ERR PREFIX - "Oops, 'firmware_data' corrupt for %s\n", dev->bus_id); + "Oops, 'acpi_handle' corrupt for %s\n", dev->bus_id); } return 0; } @@ -328,7 +329,8 @@ static int acpi_platform_notify(struct device *dev) if (!ret) { struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; - acpi_get_name(dev->firmware_data, ACPI_FULL_PATHNAME, &buffer); + acpi_get_name(dev->archdata.acpi_handle, + ACPI_FULL_PATHNAME, &buffer); DBG("Device %s -> %s\n", dev->bus_id, (char *)buffer.pointer); kfree(buffer.pointer); } else diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index f338e40bd544..fdd10953b2b6 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -357,7 +357,7 @@ struct device *acpi_get_physical_device(acpi_handle); /* helper */ acpi_handle acpi_get_child(acpi_handle, acpi_integer); acpi_handle acpi_get_pci_rootbridge_handle(unsigned int, unsigned int); -#define DEVICE_ACPI_HANDLE(dev) ((acpi_handle)((dev)->firmware_data)) +#define DEVICE_ACPI_HANDLE(dev) ((acpi_handle)((dev)->archdata.acpi_handle)) #endif /* CONFIG_ACPI */ diff --git a/include/asm-i386/device.h b/include/asm-i386/device.h index d8f9872b0e2d..849604c70e6b 100644 --- a/include/asm-i386/device.h +++ b/include/asm-i386/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_I386_DEVICE_H +#define _ASM_I386_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_I386_DEVICE_H */ diff --git a/include/asm-ia64/device.h b/include/asm-ia64/device.h index d8f9872b0e2d..3db6daf7f251 100644 --- a/include/asm-ia64/device.h +++ b/include/asm-ia64/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_IA64_DEVICE_H +#define _ASM_IA64_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_IA64_DEVICE_H */ diff --git a/include/asm-x86_64/device.h b/include/asm-x86_64/device.h index d8f9872b0e2d..3afa03f33a36 100644 --- a/include/asm-x86_64/device.h +++ b/include/asm-x86_64/device.h @@ -3,5 +3,13 @@ * * This file is released under the GPLv2 */ -#include +#ifndef _ASM_X86_64_DEVICE_H +#define _ASM_X86_64_DEVICE_H +struct dev_archdata { +#ifdef CONFIG_ACPI + void *acpi_handle; +#endif +}; + +#endif /* _ASM_X86_64_DEVICE_H */ diff --git a/include/linux/device.h b/include/linux/device.h index 5b54d756cd54..2d9dc358c027 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -369,8 +369,6 @@ struct device { void *driver_data; /* data private to the driver */ void *platform_data; /* Platform specific data, device core doesn't touch it */ - void *firmware_data; /* Firmware specific data (e.g. ACPI, - BIOS data),reserved for device core*/ struct dev_pm_info power; u64 *dma_mask; /* dma mask (if dma'able device) */ -- cgit v1.2.3 From 5ab699810d46011ad2195c5916f3cbc684bfe3ee Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 16 Nov 2006 15:42:07 +0100 Subject: driver core: Introduce device_find_child(). Introduce device_find_child() to match device_for_each_child(). Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 33 +++++++++++++++++++++++++++++++++ include/linux/device.h | 2 ++ 2 files changed, 35 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index 5d11bbdfbd2f..a29e68545462 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -750,12 +750,45 @@ int device_for_each_child(struct device * parent, void * data, return error; } +/** + * device_find_child - device iterator for locating a particular device. + * @parent: parent struct device + * @data: Data to pass to match function + * @match: Callback function to check device + * + * This is similar to the device_for_each_child() function above, but it + * returns a reference to a device that is 'found' for later use, as + * determined by the @match callback. + * + * The callback should return 0 if the device doesn't match and non-zero + * if it does. If the callback returns non-zero and a reference to the + * current device can be obtained, this function will return to the caller + * and not iterate over any more devices. + */ +struct device * device_find_child(struct device *parent, void *data, + int (*match)(struct device *, void *)) +{ + struct klist_iter i; + struct device *child; + + if (!parent) + return NULL; + + klist_iter_init(&parent->klist_children, &i); + while ((child = next_device(&i))) + if (match(child, data) && get_device(child)) + break; + klist_iter_exit(&i); + return child; +} + int __init devices_init(void) { return subsystem_register(&devices_subsys); } EXPORT_SYMBOL_GPL(device_for_each_child); +EXPORT_SYMBOL_GPL(device_find_child); EXPORT_SYMBOL_GPL(device_initialize); EXPORT_SYMBOL_GPL(device_add); diff --git a/include/linux/device.h b/include/linux/device.h index 2d9dc358c027..0a0370c74181 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -421,6 +421,8 @@ extern int __must_check device_add(struct device * dev); extern void device_del(struct device * dev); extern int device_for_each_child(struct device *, void *, int (*fn)(struct device *, void *)); +extern struct device *device_find_child(struct device *, void *data, + int (*match)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); /* -- cgit v1.2.3 From 8a82472f86bf693b8e91ed56c9ca4f62fbbdcfa3 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Mon, 20 Nov 2006 17:07:51 +0100 Subject: driver core: Introduce device_move(): move a device to a new parent. Provide a function device_move() to move a device to a new parent device. Add auxilliary functions kobject_move() and sysfs_move_dir(). kobject_move() generates a new uevent of type KOBJ_MOVE, containing the previous path (DEVPATH_OLD) in addition to the usual values. For this, a new interface kobject_uevent_env() is created that allows to add further environmental data to the uevent at the kobject layer. Signed-off-by: Cornelia Huck Acked-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 92 +++++++++++++++++++++++++++++++++++++++++++++++++ fs/sysfs/dir.c | 45 ++++++++++++++++++++++++ include/linux/device.h | 1 + include/linux/kobject.h | 8 +++++ include/linux/sysfs.h | 8 +++++ lib/kobject.c | 50 +++++++++++++++++++++++++++ lib/kobject_uevent.c | 28 ++++++++++++--- 7 files changed, 228 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index 75b45a10935a..e4eaf46c4d93 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -955,3 +955,95 @@ int device_rename(struct device *dev, char *new_name) return error; } + + +static int device_move_class_links(struct device *dev, + struct device *old_parent, + struct device *new_parent) +{ +#ifdef CONFIG_SYSFS_DEPRECATED + int error; + char *class_name; + + class_name = make_class_name(dev->class->name, &dev->kobj); + if (!class_name) { + error = PTR_ERR(class_name); + class_name = NULL; + goto out; + } + if (old_parent) { + sysfs_remove_link(&dev->kobj, "device"); + sysfs_remove_link(&old_parent->kobj, class_name); + } + error = sysfs_create_link(&dev->kobj, &new_parent->kobj, "device"); + if (error) + goto out; + error = sysfs_create_link(&new_parent->kobj, &dev->kobj, class_name); + if (error) + sysfs_remove_link(&dev->kobj, "device"); +out: + kfree(class_name); + return error; +#else + return 0; +#endif +} + +/** + * device_move - moves a device to a new parent + * @dev: the pointer to the struct device to be moved + * @new_parent: the new parent of the device + */ +int device_move(struct device *dev, struct device *new_parent) +{ + int error; + struct device *old_parent; + + dev = get_device(dev); + if (!dev) + return -EINVAL; + + if (!device_is_registered(dev)) { + error = -EINVAL; + goto out; + } + new_parent = get_device(new_parent); + if (!new_parent) { + error = -EINVAL; + goto out; + } + pr_debug("DEVICE: moving '%s' to '%s'\n", dev->bus_id, + new_parent->bus_id); + error = kobject_move(&dev->kobj, &new_parent->kobj); + if (error) { + put_device(new_parent); + goto out; + } + old_parent = dev->parent; + dev->parent = new_parent; + if (old_parent) + klist_del(&dev->knode_parent); + klist_add_tail(&dev->knode_parent, &new_parent->klist_children); + if (!dev->class) + goto out_put; + error = device_move_class_links(dev, old_parent, new_parent); + if (error) { + /* We ignore errors on cleanup since we're hosed anyway... */ + device_move_class_links(dev, new_parent, old_parent); + if (!kobject_move(&dev->kobj, &old_parent->kobj)) { + klist_del(&dev->knode_parent); + if (old_parent) + klist_add_tail(&dev->knode_parent, + &old_parent->klist_children); + } + put_device(new_parent); + goto out; + } +out_put: + put_device(old_parent); +out: + put_device(dev); + return error; +} + +EXPORT_SYMBOL_GPL(device_move); diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 3aa3434621ca..a5782e8c7f07 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -372,6 +372,51 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name) return error; } +int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent) +{ + struct dentry *old_parent_dentry, *new_parent_dentry, *new_dentry; + struct sysfs_dirent *new_parent_sd, *sd; + int error; + + if (!new_parent) + return -EINVAL; + + old_parent_dentry = kobj->parent ? + kobj->parent->dentry : sysfs_mount->mnt_sb->s_root; + new_parent_dentry = new_parent->dentry; + +again: + mutex_lock(&old_parent_dentry->d_inode->i_mutex); + if (!mutex_trylock(&new_parent_dentry->d_inode->i_mutex)) { + mutex_unlock(&old_parent_dentry->d_inode->i_mutex); + goto again; + } + + new_parent_sd = new_parent_dentry->d_fsdata; + sd = kobj->dentry->d_fsdata; + + new_dentry = lookup_one_len(kobj->name, new_parent_dentry, + strlen(kobj->name)); + if (IS_ERR(new_dentry)) { + error = PTR_ERR(new_dentry); + goto out; + } else + error = 0; + d_add(new_dentry, NULL); + d_move(kobj->dentry, new_dentry); + dput(new_dentry); + + /* Remove from old parent's list and insert into new parent's list. */ + list_del_init(&sd->s_sibling); + list_add(&sd->s_sibling, &new_parent_sd->s_children); + +out: + mutex_unlock(&new_parent_dentry->d_inode->i_mutex); + mutex_unlock(&old_parent_dentry->d_inode->i_mutex); + + return error; +} + static int sysfs_dir_open(struct inode *inode, struct file *file) { struct dentry * dentry = file->f_dentry; diff --git a/include/linux/device.h b/include/linux/device.h index 0a0370c74181..583a341e016c 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -424,6 +424,7 @@ extern int device_for_each_child(struct device *, void *, extern struct device *device_find_child(struct device *, void *data, int (*match)(struct device *, void *)); extern int device_rename(struct device *dev, char *new_name); +extern int device_move(struct device *dev, struct device *new_parent); /* * Manual binding of a device to driver. See drivers/base/bus.c diff --git a/include/linux/kobject.h b/include/linux/kobject.h index bcd9cd173c2c..d1c8d28fa92e 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -47,6 +47,7 @@ enum kobject_action { KOBJ_UMOUNT = (__force kobject_action_t) 0x05, /* umount event for block devices (broken) */ KOBJ_OFFLINE = (__force kobject_action_t) 0x06, /* device offline */ KOBJ_ONLINE = (__force kobject_action_t) 0x07, /* device online */ + KOBJ_MOVE = (__force kobject_action_t) 0x08, /* device move */ }; struct kobject { @@ -76,6 +77,7 @@ extern int __must_check kobject_add(struct kobject *); extern void kobject_del(struct kobject *); extern int __must_check kobject_rename(struct kobject *, const char *new_name); +extern int __must_check kobject_move(struct kobject *, struct kobject *); extern int __must_check kobject_register(struct kobject *); extern void kobject_unregister(struct kobject *); @@ -264,6 +266,8 @@ extern int __must_check subsys_create_file(struct subsystem * , #if defined(CONFIG_HOTPLUG) void kobject_uevent(struct kobject *kobj, enum kobject_action action); +void kobject_uevent_env(struct kobject *kobj, enum kobject_action action, + char *envp[]); int add_uevent_var(char **envp, int num_envp, int *cur_index, char *buffer, int buffer_size, int *cur_len, @@ -271,6 +275,10 @@ int add_uevent_var(char **envp, int num_envp, int *cur_index, __attribute__((format (printf, 7, 8))); #else static inline void kobject_uevent(struct kobject *kobj, enum kobject_action action) { } +static inline void kobject_uevent_env(struct kobject *kobj, + enum kobject_action action, + char *envp[]) +{ } static inline int add_uevent_var(char **envp, int num_envp, int *cur_index, char *buffer, int buffer_size, int *cur_len, diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 6d5c43d31dec..2129d1b6c874 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -96,6 +96,9 @@ sysfs_remove_dir(struct kobject *); extern int __must_check sysfs_rename_dir(struct kobject *, const char *new_name); +extern int __must_check +sysfs_move_dir(struct kobject *, struct kobject *); + extern int __must_check sysfs_create_file(struct kobject *, const struct attribute *); @@ -142,6 +145,11 @@ static inline int sysfs_rename_dir(struct kobject * k, const char *new_name) return 0; } +static inline int sysfs_move_dir(struct kobject * k, struct kobject * new_parent) +{ + return 0; +} + static inline int sysfs_create_file(struct kobject * k, const struct attribute * a) { return 0; diff --git a/lib/kobject.c b/lib/kobject.c index 7dd5c0e9d996..744a4b102c7f 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -310,6 +310,56 @@ int kobject_rename(struct kobject * kobj, const char *new_name) return error; } +/** + * kobject_move - move object to another parent + * @kobj: object in question. + * @new_parent: object's new parent + */ + +int kobject_move(struct kobject *kobj, struct kobject *new_parent) +{ + int error; + struct kobject *old_parent; + const char *devpath = NULL; + char *devpath_string = NULL; + char *envp[2]; + + kobj = kobject_get(kobj); + if (!kobj) + return -EINVAL; + new_parent = kobject_get(new_parent); + if (!new_parent) { + error = -EINVAL; + goto out; + } + /* old object path */ + devpath = kobject_get_path(kobj, GFP_KERNEL); + if (!devpath) { + error = -ENOMEM; + goto out; + } + devpath_string = kmalloc(strlen(devpath) + 15, GFP_KERNEL); + if (!devpath_string) { + error = -ENOMEM; + goto out; + } + sprintf(devpath_string, "DEVPATH_OLD=%s", devpath); + envp[0] = devpath_string; + envp[1] = NULL; + error = sysfs_move_dir(kobj, new_parent); + if (error) + goto out; + old_parent = kobj->parent; + kobj->parent = new_parent; + kobject_put(old_parent); + kobject_uevent_env(kobj, KOBJ_MOVE, envp); +out: + kobject_put(kobj); + kfree(devpath_string); + kfree(devpath); + return error; +} + /** * kobject_del - unlink kobject from hierarchy. * @kobj: object. diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 7f20e7b857cb..a1922765ff31 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -50,18 +50,22 @@ static char *action_to_string(enum kobject_action action) return "offline"; case KOBJ_ONLINE: return "online"; + case KOBJ_MOVE: + return "move"; default: return NULL; } } /** - * kobject_uevent - notify userspace by ending an uevent + * kobject_uevent_env - send an uevent with environmental data * - * @action: action that is happening (usually KOBJ_ADD and KOBJ_REMOVE) + * @action: action that is happening (usually KOBJ_MOVE) * @kobj: struct kobject that the action is happening to + * @envp_ext: pointer to environmental data */ -void kobject_uevent(struct kobject *kobj, enum kobject_action action) +void kobject_uevent_env(struct kobject *kobj, enum kobject_action action, + char *envp_ext[]) { char **envp; char *buffer; @@ -76,6 +80,7 @@ void kobject_uevent(struct kobject *kobj, enum kobject_action action) char *seq_buff; int i = 0; int retval; + int j; pr_debug("%s\n", __FUNCTION__); @@ -134,7 +139,8 @@ void kobject_uevent(struct kobject *kobj, enum kobject_action action) scratch += sprintf (scratch, "DEVPATH=%s", devpath) + 1; envp [i++] = scratch; scratch += sprintf(scratch, "SUBSYSTEM=%s", subsystem) + 1; - + for (j = 0; envp_ext && envp_ext[j]; j++) + envp[i++] = envp_ext[j]; /* just reserve the space, overwrite it after kset call has returned */ envp[i++] = seq_buff = scratch; scratch += strlen("SEQNUM=18446744073709551616") + 1; @@ -200,6 +206,20 @@ exit: kfree(envp); return; } + +EXPORT_SYMBOL_GPL(kobject_uevent_env); + +/** + * kobject_uevent - notify userspace by ending an uevent + * + * @action: action that is happening (usually KOBJ_ADD and KOBJ_REMOVE) + * @kobj: struct kobject that the action is happening to + */ +void kobject_uevent(struct kobject *kobj, enum kobject_action action) +{ + kobject_uevent_env(kobj, action, NULL); +} + EXPORT_SYMBOL_GPL(kobject_uevent); /** -- cgit v1.2.3 From c67334fbdfbba533af767610cf3fde8a49710e62 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Thu, 16 Nov 2006 23:28:47 -0800 Subject: Driver core: platform_driver_probe(), can save codespace This defines a new platform_driver_probe() method allowing the driver's probe() method, and its support code+data, to safely live in __init sections for typical system configurations. Many system-on-chip processors could benefit from this API, to the tune of recovering hundreds to thousands of bytes per driver. That's memory which is currently wasted holding code which can never be called after system startup, yet can not be removed. It can't be removed because of the linkage requirement that pointers to init section code (like, ideally, probe support) must not live in other sections (like driver method tables) after those pointers would be invalid. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- drivers/base/platform.c | 48 +++++++++++++++++++++++++++++++++++++++++ include/linux/platform_device.h | 6 ++++++ 2 files changed, 54 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 940ce41f1887..d1df4a087924 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -388,6 +388,11 @@ static int platform_drv_probe(struct device *_dev) return drv->probe(dev); } +static int platform_drv_probe_fail(struct device *_dev) +{ + return -ENXIO; +} + static int platform_drv_remove(struct device *_dev) { struct platform_driver *drv = to_platform_driver(_dev->driver); @@ -451,6 +456,49 @@ void platform_driver_unregister(struct platform_driver *drv) } EXPORT_SYMBOL_GPL(platform_driver_unregister); +/** + * platform_driver_probe - register driver for non-hotpluggable device + * @drv: platform driver structure + * @probe: the driver probe routine, probably from an __init section + * + * Use this instead of platform_driver_register() when you know the device + * is not hotpluggable and has already been registered, and you want to + * remove its run-once probe() infrastructure from memory after the driver + * has bound to the device. + * + * One typical use for this would be with drivers for controllers integrated + * into system-on-chip processors, where the controller devices have been + * configured as part of board setup. + * + * Returns zero if the driver registered and bound to a device, else returns + * a negative error code and with the driver not registered. + */ +int platform_driver_probe(struct platform_driver *drv, + int (*probe)(struct platform_device *)) +{ + int retval, code; + + /* temporary section violation during probe() */ + drv->probe = probe; + retval = code = platform_driver_register(drv); + + /* Fixup that section violation, being paranoid about code scanning + * the list of drivers in order to probe new devices. Check to see + * if the probe was successful, and make sure any forced probes of + * new devices fail. + */ + spin_lock(&platform_bus_type.klist_drivers.k_lock); + drv->probe = NULL; + if (code == 0 && list_empty(&drv->driver.klist_devices.k_list)) + retval = -ENODEV; + drv->driver.probe = platform_drv_probe_fail; + spin_unlock(&platform_bus_type.klist_drivers.k_lock); + + if (code != retval) + platform_driver_unregister(drv); + return retval; +} +EXPORT_SYMBOL_GPL(platform_driver_probe); /* modalias support enables more hands-off userspace setup: * (a) environment variable lets new-style hotplug events work once system is diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h index 29cd6dee13db..20f47b81d3fa 100644 --- a/include/linux/platform_device.h +++ b/include/linux/platform_device.h @@ -58,6 +58,12 @@ struct platform_driver { extern int platform_driver_register(struct platform_driver *); extern void platform_driver_unregister(struct platform_driver *); +/* non-hotpluggable platform devices may use this so that probe() and + * its support may live in __init sections, conserving runtime memory. + */ +extern int platform_driver_probe(struct platform_driver *driver, + int (*probe)(struct platform_device *)); + #define platform_get_drvdata(_dev) dev_get_drvdata(&(_dev)->dev) #define platform_set_drvdata(_dev,data) dev_set_drvdata(&(_dev)->dev, (data)) -- cgit v1.2.3 From e17e0f51aeea4e59c7e450a1c0f26605b91c1260 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Fri, 24 Nov 2006 12:15:25 +0100 Subject: Driver core: show drivers in /sys/module/ Show the drivers, which belong to the module: $ ls -l /sys/module/usbcore/drivers/ hub -> ../../../bus/usb/drivers/hub usb -> ../../../bus/usb/drivers/usb usbfs -> ../../../bus/usb/drivers/usbfs Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- include/linux/module.h | 1 + kernel/module.c | 31 +++++++++++++++++++++++++------ 2 files changed, 26 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/module.h b/include/linux/module.h index d1d00ce8f4ed..9258ffd8a7f0 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -264,6 +264,7 @@ struct module struct module_attribute *modinfo_attrs; const char *version; const char *srcversion; + struct kobject *drivers_dir; /* Exported symbols */ const struct kernel_symbol *syms; diff --git a/kernel/module.c b/kernel/module.c index f0166563c602..45e01cb60101 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1086,22 +1086,35 @@ static int mod_sysfs_setup(struct module *mod, goto out; kobj_set_kset_s(&mod->mkobj, module_subsys); mod->mkobj.mod = mod; - err = kobject_register(&mod->mkobj.kobj); + + /* delay uevent until full sysfs population */ + kobject_init(&mod->mkobj.kobj); + err = kobject_add(&mod->mkobj.kobj); if (err) goto out; + mod->drivers_dir = kobject_add_dir(&mod->mkobj.kobj, "drivers"); + if (!mod->drivers_dir) + goto out_unreg; + err = module_param_sysfs_setup(mod, kparam, num_params); if (err) - goto out_unreg; + goto out_unreg_drivers; err = module_add_modinfo_attrs(mod); if (err) - goto out_unreg; + goto out_unreg_param; + kobject_uevent(&mod->mkobj.kobj, KOBJ_ADD); return 0; +out_unreg_drivers: + kobject_unregister(mod->drivers_dir); +out_unreg_param: + module_param_sysfs_remove(mod); out_unreg: - kobject_unregister(&mod->mkobj.kobj); + kobject_del(&mod->mkobj.kobj); + kobject_put(&mod->mkobj.kobj); out: return err; } @@ -1110,6 +1123,7 @@ static void mod_kobject_remove(struct module *mod) { module_remove_modinfo_attrs(mod); module_param_sysfs_remove(mod); + kobject_unregister(mod->drivers_dir); kobject_unregister(&mod->mkobj.kobj); } @@ -2275,11 +2289,14 @@ void print_modules(void) void module_add_driver(struct module *mod, struct device_driver *drv) { + int no_warn; + if (!mod || !drv) return; - /* Don't check return code; this call is idempotent */ - sysfs_create_link(&drv->kobj, &mod->mkobj.kobj, "module"); + /* Don't check return codes; these calls are idempotent */ + no_warn = sysfs_create_link(&drv->kobj, &mod->mkobj.kobj, "module"); + no_warn = sysfs_create_link(mod->drivers_dir, &drv->kobj, drv->name); } EXPORT_SYMBOL(module_add_driver); @@ -2288,6 +2305,8 @@ void module_remove_driver(struct device_driver *drv) if (!drv) return; sysfs_remove_link(&drv->kobj, "module"); + if (drv->owner && drv->owner->drivers_dir) + sysfs_remove_link(drv->owner->drivers_dir, drv->name); } EXPORT_SYMBOL(module_remove_driver); -- cgit v1.2.3 From fc085150b491bfc186efbca90a14cf907a3060a9 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Tue, 10 Oct 2006 14:28:11 -0700 Subject: [PATCH] libata: add 40pin "short" cable support, honour drive side speed detection [deweerdt@free.fr: build fix] Signed-off-by: Alan Cox Signed-off-by: Frederik Deweerdt Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/ata_piix.c | 31 ++++++++++++++++++++++++++++++- drivers/ata/libata-core.c | 7 +++++++ drivers/ata/pata_ali.c | 2 +- include/linux/ata.h | 14 ++++++++++++-- 4 files changed, 50 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 720174d628fa..14b726b9f985 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -93,7 +93,7 @@ #include #define DRV_NAME "ata_piix" -#define DRV_VERSION "2.00ac6" +#define DRV_VERSION "2.00ac7" enum { PIIX_IOCFG = 0x54, /* IDE I/O configuration register */ @@ -571,6 +571,23 @@ module_param(force_pcs, int, 0444); MODULE_PARM_DESC(force_pcs, "force honoring or ignoring PCS to work around " "device mis-detection (0=default, 1=ignore PCS, 2=honor PCS)"); +struct ich_laptop { + u16 device; + u16 subvendor; + u16 subdevice; +}; + +/* + * List of laptops that use short cables rather than 80 wire + */ + +static const struct ich_laptop ich_laptop[] = { + /* devid, subvendor, subdev */ + { 0x27DF, 0x0005, 0x0280 }, /* ICH7 on Acer 5602WLMi */ + /* end marker */ + { 0, } +}; + /** * piix_pata_cbl_detect - Probe host controller cable detect info * @ap: Port for which cable detect info is desired @@ -585,12 +602,24 @@ MODULE_PARM_DESC(force_pcs, "force honoring or ignoring PCS to work around " static void ich_pata_cbl_detect(struct ata_port *ap) { struct pci_dev *pdev = to_pci_dev(ap->host->dev); + const struct ich_laptop *lap = &ich_laptop[0]; u8 tmp, mask; /* no 80c support in host controller? */ if ((ap->udma_mask & ~ATA_UDMA_MASK_40C) == 0) goto cbl40; + /* Check for specials - Acer Aspire 5602WLMi */ + while (lap->device) { + if (lap->device == pdev->device && + lap->subvendor == pdev->subsystem_vendor && + lap->subdevice == pdev->subsystem_device) { + ap->cbl = ATA_CBL_PATA40_SHORT; + return; + } + lap++; + } + /* check BIOS cable detect results */ mask = ap->port_no == 0 ? PIIX_80C_PRI : PIIX_80C_SEC; pci_read_config_byte(pdev, PIIX_IOCFG, &tmp); diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 915a55a6cc14..b896119cccd3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -3119,6 +3119,13 @@ static void ata_dev_xfermask(struct ata_device *dev) */ if (ap->cbl == ATA_CBL_PATA40) xfer_mask &= ~(0xF8 << ATA_SHIFT_UDMA); + /* Apply drive side cable rule. Unknown or 80 pin cables reported + * host side are checked drive side as well. Cases where we know a + * 40wire cable is used safely for 80 are not checked here. + */ + if (ata_drive_40wire(dev->id) && (ap->cbl == ATA_CBL_PATA_UNK || ap->cbl == ATA_CBL_PATA80)) + xfer_mask &= ~(0xF8 << ATA_SHIFT_UDMA); + xfer_mask &= ata_pack_xfermask(dev->pio_mask, dev->mwdma_mask, dev->udma_mask); diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c index 64eed99f6814..e6d2b840e870 100644 --- a/drivers/ata/pata_ali.c +++ b/drivers/ata/pata_ali.c @@ -78,7 +78,7 @@ static int ali_c2_cable_detect(struct ata_port *ap) implement the detect logic */ if (ali_cable_override(pdev)) - return ATA_CBL_PATA80; + return ATA_CBL_PATA40_SHORT; /* Host view cable detect 0x4A bit 0 primary bit 1 secondary Bit set for 40 pin */ diff --git a/include/linux/ata.h b/include/linux/ata.h index d89441907024..1df941648a57 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -200,8 +200,9 @@ enum { ATA_CBL_NONE = 0, ATA_CBL_PATA40 = 1, ATA_CBL_PATA80 = 2, - ATA_CBL_PATA_UNK = 3, - ATA_CBL_SATA = 4, + ATA_CBL_PATA40_SHORT = 3, /* 40 wire cable to high UDMA spec */ + ATA_CBL_PATA_UNK = 4, + ATA_CBL_SATA = 5, /* SATA Status and Control Registers */ SCR_STATUS = 0, @@ -342,6 +343,15 @@ static inline int ata_id_is_cfa(const u16 *id) return 0; } +static inline int ata_drive_40wire(const u16 *dev_id) +{ + if (ata_id_major_version(dev_id) >= 5 && ata_id_is_sata(dev_id)) + return 0; /* SATA */ + if (dev_id[93] & 0x4000) + return 0; /* 80 wire */ + return 1; +} + static inline int atapi_cdb_len(const u16 *dev_id) { u16 tmp = dev_id[0] & 0x3; -- cgit v1.2.3 From 6919a0a6cfdad9e83d02cef5973826acd416560c Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 27 Oct 2006 19:08:46 -0700 Subject: [PATCH] libata: Revamp blacklist support to allow multiple kinds of blacklisting flaws Signed-off-by: Alan Cox Cc: Jeff Garzik Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 123 ++++++++++++++++++++++++++++------------------ include/linux/libata.h | 3 ++ 2 files changed, 78 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index b896119cccd3..b4fbfebafd26 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1345,7 +1345,10 @@ static void ata_dev_config_ncq(struct ata_device *dev, desc[0] = '\0'; return; } - + if (ata_device_blacklisted(dev) & ATA_HORKAGE_NONCQ) { + snprintf(desc, desc_sz, "NCQ (not used)"); + return; + } if (ap->flags & ATA_FLAG_NCQ) { hdepth = min(ap->scsi_host->can_queue, ATA_MAX_QUEUE - 1); dev->flags |= ATA_DFLAG_NCQ; @@ -3014,37 +3017,55 @@ int ata_dev_revalidate(struct ata_device *dev, int post_reset) return rc; } -static const char * const ata_dma_blacklist [] = { - "WDC AC11000H", NULL, - "WDC AC22100H", NULL, - "WDC AC32500H", NULL, - "WDC AC33100H", NULL, - "WDC AC31600H", NULL, - "WDC AC32100H", "24.09P07", - "WDC AC23200L", "21.10N21", - "Compaq CRD-8241B", NULL, - "CRD-8400B", NULL, - "CRD-8480B", NULL, - "CRD-8482B", NULL, - "CRD-84", NULL, - "SanDisk SDP3B", NULL, - "SanDisk SDP3B-64", NULL, - "SANYO CD-ROM CRD", NULL, - "HITACHI CDR-8", NULL, - "HITACHI CDR-8335", NULL, - "HITACHI CDR-8435", NULL, - "Toshiba CD-ROM XM-6202B", NULL, - "TOSHIBA CD-ROM XM-1702BC", NULL, - "CD-532E-A", NULL, - "E-IDE CD-ROM CR-840", NULL, - "CD-ROM Drive/F5A", NULL, - "WPI CDD-820", NULL, - "SAMSUNG CD-ROM SC-148C", NULL, - "SAMSUNG CD-ROM SC", NULL, - "SanDisk SDP3B-64", NULL, - "ATAPI CD-ROM DRIVE 40X MAXIMUM",NULL, - "_NEC DV5800A", NULL, - "SAMSUNG CD-ROM SN-124", "N001" +struct ata_blacklist_entry { + const char *model_num; + const char *model_rev; + unsigned long horkage; +}; + +static const struct ata_blacklist_entry ata_device_blacklist [] = { + /* Devices with DMA related problems under Linux */ + { "WDC AC11000H", NULL, ATA_HORKAGE_NODMA }, + { "WDC AC22100H", NULL, ATA_HORKAGE_NODMA }, + { "WDC AC32500H", NULL, ATA_HORKAGE_NODMA }, + { "WDC AC33100H", NULL, ATA_HORKAGE_NODMA }, + { "WDC AC31600H", NULL, ATA_HORKAGE_NODMA }, + { "WDC AC32100H", "24.09P07", ATA_HORKAGE_NODMA }, + { "WDC AC23200L", "21.10N21", ATA_HORKAGE_NODMA }, + { "Compaq CRD-8241B", NULL, ATA_HORKAGE_NODMA }, + { "CRD-8400B", NULL, ATA_HORKAGE_NODMA }, + { "CRD-8480B", NULL, ATA_HORKAGE_NODMA }, + { "CRD-8482B", NULL, ATA_HORKAGE_NODMA }, + { "CRD-84", NULL, ATA_HORKAGE_NODMA }, + { "SanDisk SDP3B", NULL, ATA_HORKAGE_NODMA }, + { "SanDisk SDP3B-64", NULL, ATA_HORKAGE_NODMA }, + { "SANYO CD-ROM CRD", NULL, ATA_HORKAGE_NODMA }, + { "HITACHI CDR-8", NULL, ATA_HORKAGE_NODMA }, + { "HITACHI CDR-8335", NULL, ATA_HORKAGE_NODMA }, + { "HITACHI CDR-8435", NULL, ATA_HORKAGE_NODMA }, + { "Toshiba CD-ROM XM-6202B", NULL, ATA_HORKAGE_NODMA }, + { "TOSHIBA CD-ROM XM-1702BC", NULL, ATA_HORKAGE_NODMA }, + { "CD-532E-A", NULL, ATA_HORKAGE_NODMA }, + { "E-IDE CD-ROM CR-840",NULL, ATA_HORKAGE_NODMA }, + { "CD-ROM Drive/F5A", NULL, ATA_HORKAGE_NODMA }, + { "WPI CDD-820", NULL, ATA_HORKAGE_NODMA }, + { "SAMSUNG CD-ROM SC-148C", NULL, ATA_HORKAGE_NODMA }, + { "SAMSUNG CD-ROM SC", NULL, ATA_HORKAGE_NODMA }, + { "SanDisk SDP3B-64", NULL, ATA_HORKAGE_NODMA }, + { "ATAPI CD-ROM DRIVE 40X MAXIMUM",NULL,ATA_HORKAGE_NODMA }, + { "_NEC DV5800A", NULL, ATA_HORKAGE_NODMA }, + { "SAMSUNG CD-ROM SN-124","N001", ATA_HORKAGE_NODMA }, + + /* Devices we expect to fail diagnostics */ + + /* Devices where NCQ should be avoided */ + /* NCQ is slow */ + { "WDC WD740ADFD-00", NULL, ATA_HORKAGE_NONCQ }, + + /* Devices with NCQ limits */ + + /* End Marker */ + { } }; static int ata_strim(char *s, size_t len) @@ -3059,20 +3080,12 @@ static int ata_strim(char *s, size_t len) return len; } -static int ata_dma_blacklisted(const struct ata_device *dev) +unsigned long ata_device_blacklisted(const struct ata_device *dev) { unsigned char model_num[40]; unsigned char model_rev[16]; unsigned int nlen, rlen; - int i; - - /* We don't support polling DMA. - * DMA blacklist those ATAPI devices with CDB-intr (and use PIO) - * if the LLDD handles only interrupts in the HSM_ST_LAST state. - */ - if ((dev->ap->flags & ATA_FLAG_PIO_POLLING) && - (dev->flags & ATA_DFLAG_CDB_INTR)) - return 1; + const struct ata_blacklist_entry *ad = ata_device_blacklist; ata_id_string(dev->id, model_num, ATA_ID_PROD_OFS, sizeof(model_num)); @@ -3081,17 +3094,30 @@ static int ata_dma_blacklisted(const struct ata_device *dev) nlen = ata_strim(model_num, sizeof(model_num)); rlen = ata_strim(model_rev, sizeof(model_rev)); - for (i = 0; i < ARRAY_SIZE(ata_dma_blacklist); i += 2) { - if (!strncmp(ata_dma_blacklist[i], model_num, nlen)) { - if (ata_dma_blacklist[i+1] == NULL) - return 1; - if (!strncmp(ata_dma_blacklist[i], model_rev, rlen)) - return 1; + while (ad->model_num) { + if (!strncmp(ad->model_num, model_num, nlen)) { + if (ad->model_rev == NULL) + return ad->horkage; + if (!strncmp(ad->model_rev, model_rev, rlen)) + return ad->horkage; } + ad++; } return 0; } +static int ata_dma_blacklisted(const struct ata_device *dev) +{ + /* We don't support polling DMA. + * DMA blacklist those ATAPI devices with CDB-intr (and use PIO) + * if the LLDD handles only interrupts in the HSM_ST_LAST state. + */ + if ((dev->ap->flags & ATA_FLAG_PIO_POLLING) && + (dev->flags & ATA_DFLAG_CDB_INTR)) + return 1; + return (ata_device_blacklisted(dev) & ATA_HORKAGE_NODMA) ? 1 : 0; +} + /** * ata_dev_xfermask - Compute supported xfermask of the given device * @dev: Device to compute xfermask for @@ -6153,6 +6179,7 @@ EXPORT_SYMBOL_GPL(ata_host_suspend); EXPORT_SYMBOL_GPL(ata_host_resume); EXPORT_SYMBOL_GPL(ata_id_string); EXPORT_SYMBOL_GPL(ata_id_c_string); +EXPORT_SYMBOL_GPL(ata_device_blacklisted); EXPORT_SYMBOL_GPL(ata_scsi_simulate); EXPORT_SYMBOL_GPL(ata_pio_need_iordy); diff --git a/include/linux/libata.h b/include/linux/libata.h index abd2debebca2..2300fcc37f80 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -307,6 +307,8 @@ enum { (some horkage may be drive/controller pair dependant */ ATA_HORKAGE_DIAGNOSTIC = (1 << 0), /* Failed boot diag */ + ATA_HORKAGE_NODMA = (1 << 1), /* DMA problems */ + ATA_HORKAGE_NONCQ = (1 << 2), /* Don't use NCQ */ }; enum hsm_task_states { @@ -787,6 +789,7 @@ extern void ata_id_string(const u16 *id, unsigned char *s, unsigned int ofs, unsigned int len); extern void ata_id_c_string(const u16 *id, unsigned char *s, unsigned int ofs, unsigned int len); +extern unsigned long ata_device_blacklisted(const struct ata_device *dev); extern void ata_bmdma_setup (struct ata_queued_cmd *qc); extern void ata_bmdma_start (struct ata_queued_cmd *qc); extern void ata_bmdma_stop(struct ata_queued_cmd *qc); -- cgit v1.2.3 From a20c9e820864e18b59d2a4f2f04e8b6053986c95 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 27 Oct 2006 19:08:48 -0700 Subject: [PATCH] ata: Generic platform_device libata driver needs a changelog Signed-off-by: Paul Mundt Acked-by: Russell King Cc: Jeff Garzik Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/Kconfig | 9 ++ drivers/ata/Makefile | 1 + drivers/ata/pata_platform.c | 295 ++++++++++++++++++++++++++++++++++++++++++ include/linux/pata_platform.h | 13 ++ 4 files changed, 318 insertions(+) create mode 100644 drivers/ata/pata_platform.c create mode 100644 include/linux/pata_platform.h (limited to 'include/linux') diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig index 9b07e70bee7d..98d0f01c90eb 100644 --- a/drivers/ata/Kconfig +++ b/drivers/ata/Kconfig @@ -499,6 +499,15 @@ config PATA_WINBOND_VLB Support for the Winbond W83759A controller on Vesa Local Bus systems. +config PATA_PLATFORM + tristate "Generic platform device PATA support" + depends on EMBEDDED + help + This option enables support for generic directly connected ATA + devices commonly found on embedded systems. + + If unsure, say N. + endif endmenu diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile index ceba7d824a62..f2b3ea2d15d0 100644 --- a/drivers/ata/Makefile +++ b/drivers/ata/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_PATA_WINBOND) += pata_sl82c105.o obj-$(CONFIG_PATA_WINBOND_VLB) += pata_winbond.o obj-$(CONFIG_PATA_SIS) += pata_sis.o obj-$(CONFIG_PATA_TRIFLEX) += pata_triflex.o +obj-$(CONFIG_PATA_PLATFORM) += pata_platform.o # Should be last but one libata driver obj-$(CONFIG_ATA_GENERIC) += ata_generic.o # Should be last libata driver diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c new file mode 100644 index 000000000000..63d6687f0ba9 --- /dev/null +++ b/drivers/ata/pata_platform.c @@ -0,0 +1,295 @@ +/* + * Generic platform device PATA driver + * + * Copyright (C) 2006 Paul Mundt + * + * Based on pata_pcmcia: + * + * Copyright 2005-2006 Red Hat Inc , all rights reserved. + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DRV_NAME "pata_platform" +#define DRV_VERSION "0.1.2" + +static int pio_mask = 1; + +/* + * Provide our own set_mode() as we don't want to change anything that has + * already been configured.. + */ +static void pata_platform_set_mode(struct ata_port *ap) +{ + int i; + + for (i = 0; i < ATA_MAX_DEVICES; i++) { + struct ata_device *dev = &ap->device[i]; + + if (ata_dev_enabled(dev)) { + /* We don't really care */ + dev->pio_mode = dev->xfer_mode = XFER_PIO_0; + dev->xfer_shift = ATA_SHIFT_PIO; + dev->flags |= ATA_DFLAG_PIO; + } + } +} + +static void pata_platform_host_stop(struct ata_host *host) +{ + int i; + + /* + * Unmap the bases for MMIO + */ + for (i = 0; i < host->n_ports; i++) { + struct ata_port *ap = host->ports[i]; + + if (ap->flags & ATA_FLAG_MMIO) { + iounmap((void __iomem *)ap->ioaddr.ctl_addr); + iounmap((void __iomem *)ap->ioaddr.cmd_addr); + } + } +} + +static struct scsi_host_template pata_platform_sht = { + .module = THIS_MODULE, + .name = DRV_NAME, + .ioctl = ata_scsi_ioctl, + .queuecommand = ata_scsi_queuecmd, + .can_queue = ATA_DEF_QUEUE, + .this_id = ATA_SHT_THIS_ID, + .sg_tablesize = LIBATA_MAX_PRD, + .max_sectors = ATA_MAX_SECTORS, + .cmd_per_lun = ATA_SHT_CMD_PER_LUN, + .emulated = ATA_SHT_EMULATED, + .use_clustering = ATA_SHT_USE_CLUSTERING, + .proc_name = DRV_NAME, + .dma_boundary = ATA_DMA_BOUNDARY, + .slave_configure = ata_scsi_slave_config, + .bios_param = ata_std_bios_param, +}; + +static struct ata_port_operations pata_platform_port_ops = { + .set_mode = pata_platform_set_mode, + + .port_disable = ata_port_disable, + .tf_load = ata_tf_load, + .tf_read = ata_tf_read, + .check_status = ata_check_status, + .exec_command = ata_exec_command, + .dev_select = ata_std_dev_select, + + .freeze = ata_bmdma_freeze, + .thaw = ata_bmdma_thaw, + .error_handler = ata_bmdma_error_handler, + .post_internal_cmd = ata_bmdma_post_internal_cmd, + + .qc_prep = ata_qc_prep, + .qc_issue = ata_qc_issue_prot, + + .data_xfer = ata_pio_data_xfer_noirq, + + .irq_handler = ata_interrupt, + .irq_clear = ata_bmdma_irq_clear, + + .port_start = ata_port_start, + .port_stop = ata_port_stop, + .host_stop = pata_platform_host_stop +}; + +static void pata_platform_setup_port(struct ata_ioports *ioaddr, + struct pata_platform_info *info) +{ + unsigned int shift = 0; + + /* Fixup the port shift for platforms that need it */ + if (info && info->ioport_shift) + shift = info->ioport_shift; + + ioaddr->data_addr = ioaddr->cmd_addr + (ATA_REG_DATA << shift); + ioaddr->error_addr = ioaddr->cmd_addr + (ATA_REG_ERR << shift); + ioaddr->feature_addr = ioaddr->cmd_addr + (ATA_REG_FEATURE << shift); + ioaddr->nsect_addr = ioaddr->cmd_addr + (ATA_REG_NSECT << shift); + ioaddr->lbal_addr = ioaddr->cmd_addr + (ATA_REG_LBAL << shift); + ioaddr->lbam_addr = ioaddr->cmd_addr + (ATA_REG_LBAM << shift); + ioaddr->lbah_addr = ioaddr->cmd_addr + (ATA_REG_LBAH << shift); + ioaddr->device_addr = ioaddr->cmd_addr + (ATA_REG_DEVICE << shift); + ioaddr->status_addr = ioaddr->cmd_addr + (ATA_REG_STATUS << shift); + ioaddr->command_addr = ioaddr->cmd_addr + (ATA_REG_CMD << shift); +} + +/** + * pata_platform_probe - attach a platform interface + * @pdev: platform device + * + * Register a platform bus IDE interface. Such interfaces are PIO and we + * assume do not support IRQ sharing. + * + * Platform devices are expected to contain 3 resources per port: + * + * - I/O Base (IORESOURCE_IO or IORESOURCE_MEM) + * - CTL Base (IORESOURCE_IO or IORESOURCE_MEM) + * - IRQ (IORESOURCE_IRQ) + * + * If the base resources are both mem types, the ioremap() is handled + * here. For IORESOURCE_IO, it's assumed that there's no remapping + * necessary. + */ +static int __devinit pata_platform_probe(struct platform_device *pdev) +{ + struct resource *io_res, *ctl_res; + struct ata_probe_ent ae; + unsigned int mmio; + int ret; + + /* + * Simple resource validation .. + */ + if (unlikely(pdev->num_resources != 3)) { + dev_err(&pdev->dev, "invalid number of resources\n"); + return -EINVAL; + } + + /* + * Get the I/O base first + */ + io_res = platform_get_resource(pdev, IORESOURCE_IO, 0); + if (io_res == NULL) { + io_res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (unlikely(io_res == NULL)) + return -EINVAL; + } + + /* + * Then the CTL base + */ + ctl_res = platform_get_resource(pdev, IORESOURCE_IO, 1); + if (ctl_res == NULL) { + ctl_res = platform_get_resource(pdev, IORESOURCE_MEM, 1); + if (unlikely(ctl_res == NULL)) + return -EINVAL; + } + + /* + * Check for MMIO + */ + mmio = (( io_res->flags == IORESOURCE_MEM) && + (ctl_res->flags == IORESOURCE_MEM)); + + /* + * Now that that's out of the way, wire up the port.. + */ + memset(&ae, 0, sizeof(struct ata_probe_ent)); + INIT_LIST_HEAD(&ae.node); + ae.dev = &pdev->dev; + ae.port_ops = &pata_platform_port_ops; + ae.sht = &pata_platform_sht; + ae.n_ports = 1; + ae.pio_mask = pio_mask; + ae.irq = platform_get_irq(pdev, 0); + ae.irq_flags = 0; + ae.port_flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST; + + /* + * Handle the MMIO case + */ + if (mmio) { + ae.port_flags |= ATA_FLAG_MMIO; + + ae.port[0].cmd_addr = (unsigned long)ioremap(io_res->start, + io_res->end - io_res->start + 1); + if (unlikely(!ae.port[0].cmd_addr)) { + dev_err(&pdev->dev, "failed to remap IO base\n"); + return -ENXIO; + } + + ae.port[0].ctl_addr = (unsigned long)ioremap(ctl_res->start, + ctl_res->end - ctl_res->start + 1); + if (unlikely(!ae.port[0].ctl_addr)) { + dev_err(&pdev->dev, "failed to remap CTL base\n"); + ret = -ENXIO; + goto bad_remap; + } + } else { + ae.port[0].cmd_addr = io_res->start; + ae.port[0].ctl_addr = ctl_res->start; + } + + ae.port[0].altstatus_addr = ae.port[0].ctl_addr; + + pata_platform_setup_port(&ae.port[0], pdev->dev.platform_data); + + if (unlikely(ata_device_add(&ae) == 0)) { + ret = -ENODEV; + goto add_failed; + } + + return 0; + +add_failed: + if (ae.port[0].ctl_addr && mmio) + iounmap((void __iomem *)ae.port[0].ctl_addr); +bad_remap: + if (ae.port[0].cmd_addr && mmio) + iounmap((void __iomem *)ae.port[0].cmd_addr); + + return ret; +} + +/** + * pata_platform_remove - unplug a platform interface + * @pdev: platform device + * + * A platform bus ATA device has been unplugged. Perform the needed + * cleanup. Also called on module unload for any active devices. + */ +static int __devexit pata_platform_remove(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct ata_host *host = dev_get_drvdata(dev); + + ata_host_remove(host); + dev_set_drvdata(dev, NULL); + + return 0; +} + +static struct platform_driver pata_platform_driver = { + .probe = pata_platform_probe, + .remove = __devexit_p(pata_platform_remove), + .driver = { + .name = DRV_NAME, + .owner = THIS_MODULE, + }, +}; + +static int __init pata_platform_init(void) +{ + return platform_driver_register(&pata_platform_driver); +} + +static void __exit pata_platform_exit(void) +{ + platform_driver_unregister(&pata_platform_driver); +} +module_init(pata_platform_init); +module_exit(pata_platform_exit); + +module_param(pio_mask, int, 0); + +MODULE_AUTHOR("Paul Mundt"); +MODULE_DESCRIPTION("low-level driver for platform device ATA"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_VERSION); diff --git a/include/linux/pata_platform.h b/include/linux/pata_platform.h new file mode 100644 index 000000000000..2d5fd647e0e9 --- /dev/null +++ b/include/linux/pata_platform.h @@ -0,0 +1,13 @@ +#ifndef __LINUX_PATA_PLATFORM_H +#define __LINUX_PATA_PLATFORM_H + +struct pata_platform_info { + /* + * I/O port shift, for platforms with ports that are + * constantly spaced and need larger than the 1-byte + * spacing used by ata_std_ports(). + */ + unsigned int ioport_shift; +}; + +#endif /* __LINUX_PATA_PLATFORM_H */ -- cgit v1.2.3 From d1adc1bbd6dde3e05a91e2d3e6ab42d202ea61d5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Oct 2006 18:32:15 +0900 Subject: [PATCH] libata: handle 0xff status properly libata waits for !BSY even when the status register reports 0xff. This causes long boot delays when D8 isn't pulled down properly. This patch does the followings. * don't wait if status register is 0xff in all wait functions * make ata_busy_sleep() return 0 on success and -errno on failure. -ENODEV is returned on 0xff status and -EBUSY on other failures. * make ata_bus_softreset() succeed on 0xff status. 0xff status is not reset failure. It indicates no device. This removes unnecessary retries on such ports. Note that the code change assumes unoccupied port reporting 0xff status does not produce valid device signature. Signed-off-by: Tejun Heo Cc: Joe Jin Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 30 ++++++++++++++++++------------ include/linux/libata.h | 9 ++++----- 2 files changed, 22 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index b4fbfebafd26..d2336673601c 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2328,11 +2328,14 @@ static inline void ata_tf_to_host(struct ata_port *ap, * Sleep until ATA Status register bit BSY clears, * or a timeout occurs. * - * LOCKING: None. + * LOCKING: + * Kernel thread context (may sleep). + * + * RETURNS: + * 0 on success, -errno otherwise. */ - -unsigned int ata_busy_sleep (struct ata_port *ap, - unsigned long tmout_pat, unsigned long tmout) +int ata_busy_sleep(struct ata_port *ap, + unsigned long tmout_pat, unsigned long tmout) { unsigned long timer_start, timeout; u8 status; @@ -2340,27 +2343,32 @@ unsigned int ata_busy_sleep (struct ata_port *ap, status = ata_busy_wait(ap, ATA_BUSY, 300); timer_start = jiffies; timeout = timer_start + tmout_pat; - while ((status & ATA_BUSY) && (time_before(jiffies, timeout))) { + while (status != 0xff && (status & ATA_BUSY) && + time_before(jiffies, timeout)) { msleep(50); status = ata_busy_wait(ap, ATA_BUSY, 3); } - if (status & ATA_BUSY) + if (status != 0xff && (status & ATA_BUSY)) ata_port_printk(ap, KERN_WARNING, "port is slow to respond, please be patient " "(Status 0x%x)\n", status); timeout = timer_start + tmout; - while ((status & ATA_BUSY) && (time_before(jiffies, timeout))) { + while (status != 0xff && (status & ATA_BUSY) && + time_before(jiffies, timeout)) { msleep(50); status = ata_chk_status(ap); } + if (status == 0xff) + return -ENODEV; + if (status & ATA_BUSY) { ata_port_printk(ap, KERN_ERR, "port failed to respond " "(%lu secs, Status 0x%x)\n", tmout / HZ, status); - return 1; + return -EBUSY; } return 0; @@ -2451,10 +2459,8 @@ static unsigned int ata_bus_softreset(struct ata_port *ap, * the bus shows 0xFF because the odd clown forgets the D7 * pulldown resistor. */ - if (ata_check_status(ap) == 0xFF) { - ata_port_printk(ap, KERN_ERR, "SRST failed (status 0xFF)\n"); - return AC_ERR_OTHER; - } + if (ata_check_status(ap) == 0xFF) + return 0; ata_bus_post_reset(ap, devmask); diff --git a/include/linux/libata.h b/include/linux/libata.h index 2300fcc37f80..6c003d852a88 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -746,9 +746,8 @@ extern int ata_scsi_device_suspend(struct scsi_device *, pm_message_t mesg); extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg); extern void ata_host_resume(struct ata_host *host); extern int ata_ratelimit(void); -extern unsigned int ata_busy_sleep(struct ata_port *ap, - unsigned long timeout_pat, - unsigned long timeout); +extern int ata_busy_sleep(struct ata_port *ap, + unsigned long timeout_pat, unsigned long timeout); extern void ata_port_queue_task(struct ata_port *ap, void (*fn)(void *), void *data, unsigned long delay); extern u32 ata_wait_register(void __iomem *reg, u32 mask, u32 val, @@ -1064,7 +1063,7 @@ static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits, udelay(10); status = ata_chk_status(ap); max--; - } while ((status & bits) && (max > 0)); + } while (status != 0xff && (status & bits) && (max > 0)); return status; } @@ -1085,7 +1084,7 @@ static inline u8 ata_wait_idle(struct ata_port *ap) { u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000); - if (status & (ATA_BUSY | ATA_DRQ)) { + if (status != 0xff && (status & (ATA_BUSY | ATA_DRQ))) { unsigned long l = ap->ioaddr.status_addr; if (ata_msg_warn(ap)) printk(KERN_WARNING "ATA: abnormal status 0x%X on port 0x%lX\n", -- cgit v1.2.3 From 90088bb41200b4da962282dfd45db82544adac3b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Oct 2006 11:10:26 +0900 Subject: [PATCH] libata: move ata_irq_on() into libata-sff.c ata_irq_on() isn't used outside of libata core layer. The function is TF/SFF interface specific but currently used by core path with some hack too. Move it from include/linux/libata.h to drivers/ata/libata-sff.c. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-sff.c | 29 +++++++++++++++++++++++++++++ drivers/ata/libata.h | 3 +++ include/linux/libata.h | 31 ------------------------------- 3 files changed, 32 insertions(+), 31 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index e178d6ae8b80..37471d3c1c28 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -38,6 +38,35 @@ #include "libata.h" +/** + * ata_irq_on - Enable interrupts on a port. + * @ap: Port on which interrupts are enabled. + * + * Enable interrupts on a legacy IDE device using MMIO or PIO, + * wait for idle, clear any pending interrupts. + * + * LOCKING: + * Inherited from caller. + */ +u8 ata_irq_on(struct ata_port *ap) +{ + struct ata_ioports *ioaddr = &ap->ioaddr; + u8 tmp; + + ap->ctl &= ~ATA_NIEN; + ap->last_ctl = ap->ctl; + + if (ap->flags & ATA_FLAG_MMIO) + writeb(ap->ctl, (void __iomem *) ioaddr->ctl_addr); + else + outb(ap->ctl, ioaddr->ctl_addr); + tmp = ata_wait_idle(ap); + + ap->ops->irq_clear(ap); + + return tmp; +} + /** * ata_tf_load_pio - send taskfile registers to host controller * @ap: Port to which output is sent diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index 0ed263be652a..c83300055ec5 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -120,4 +120,7 @@ extern void ata_scsi_error(struct Scsi_Host *host); extern void ata_port_wait_eh(struct ata_port *ap); extern void ata_qc_schedule_eh(struct ata_queued_cmd *qc); +/* libata-sff.c */ +extern u8 ata_irq_on(struct ata_port *ap); + #endif /* __LIBATA_H__ */ diff --git a/include/linux/libata.h b/include/linux/libata.h index 6c003d852a88..d3bf7b936c3f 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1149,37 +1149,6 @@ static inline void ata_qc_reinit(struct ata_queued_cmd *qc) qc->result_tf.feature = 0; } -/** - * ata_irq_on - Enable interrupts on a port. - * @ap: Port on which interrupts are enabled. - * - * Enable interrupts on a legacy IDE device using MMIO or PIO, - * wait for idle, clear any pending interrupts. - * - * LOCKING: - * Inherited from caller. - */ - -static inline u8 ata_irq_on(struct ata_port *ap) -{ - struct ata_ioports *ioaddr = &ap->ioaddr; - u8 tmp; - - ap->ctl &= ~ATA_NIEN; - ap->last_ctl = ap->ctl; - - if (ap->flags & ATA_FLAG_MMIO) - writeb(ap->ctl, (void __iomem *) ioaddr->ctl_addr); - else - outb(ap->ctl, ioaddr->ctl_addr); - tmp = ata_wait_idle(ap); - - ap->ops->irq_clear(ap); - - return tmp; -} - - /** * ata_irq_ack - Acknowledge a device interrupt. * @ap: Port on which interrupts are enabled. -- cgit v1.2.3 From b6103f6d1659e2024776bc759d28613fb36344a8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 1 Nov 2006 17:59:53 +0900 Subject: [PATCH] libata: separate out and export sata_port_hardreset() Separate out sata_port_hardreset() from sata_std_hardreset(). This will be used by LLD hardreset implementation and later by PMP. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 52 ++++++++++++++++++++++++++++++++++++++--------- include/linux/libata.h | 2 ++ 2 files changed, 44 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index d2336673601c..315f46841f68 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2786,9 +2786,9 @@ int ata_std_softreset(struct ata_port *ap, unsigned int *classes) } /** - * sata_std_hardreset - reset host port via SATA phy reset + * sata_port_hardreset - reset port via SATA phy reset * @ap: port to reset - * @class: resulting class of attached device + * @timing: timing parameters { interval, duratinon, timeout } in msec * * SATA phy-reset host port using DET bits of SControl register. * @@ -2798,10 +2798,8 @@ int ata_std_softreset(struct ata_port *ap, unsigned int *classes) * RETURNS: * 0 on success, -errno otherwise. */ -int sata_std_hardreset(struct ata_port *ap, unsigned int *class) +int sata_port_hardreset(struct ata_port *ap, const unsigned long *timing) { - struct ata_eh_context *ehc = &ap->eh_context; - const unsigned long *timing = sata_ehc_deb_timing(ehc); u32 scontrol; int rc; @@ -2814,24 +2812,24 @@ int sata_std_hardreset(struct ata_port *ap, unsigned int *class) * and Sil3124. */ if ((rc = sata_scr_read(ap, SCR_CONTROL, &scontrol))) - return rc; + goto out; scontrol = (scontrol & 0x0f0) | 0x304; if ((rc = sata_scr_write(ap, SCR_CONTROL, scontrol))) - return rc; + goto out; sata_set_spd(ap); } /* issue phy wake/reset */ if ((rc = sata_scr_read(ap, SCR_CONTROL, &scontrol))) - return rc; + goto out; scontrol = (scontrol & 0x0f0) | 0x301; if ((rc = sata_scr_write_flush(ap, SCR_CONTROL, scontrol))) - return rc; + goto out; /* Couldn't find anything in SATA I/II specs, but AHCI-1.1 * 10.4.2 says at least 1 ms. @@ -2839,7 +2837,40 @@ int sata_std_hardreset(struct ata_port *ap, unsigned int *class) msleep(1); /* bring phy back */ - sata_phy_resume(ap, timing); + rc = sata_phy_resume(ap, timing); + out: + DPRINTK("EXIT, rc=%d\n", rc); + return rc; +} + +/** + * sata_std_hardreset - reset host port via SATA phy reset + * @ap: port to reset + * @class: resulting class of attached device + * + * SATA phy-reset host port using DET bits of SControl register, + * wait for !BSY and classify the attached device. + * + * LOCKING: + * Kernel thread context (may sleep) + * + * RETURNS: + * 0 on success, -errno otherwise. + */ +int sata_std_hardreset(struct ata_port *ap, unsigned int *class) +{ + const unsigned long *timing = sata_ehc_deb_timing(&ap->eh_context); + int rc; + + DPRINTK("ENTER\n"); + + /* do hardreset */ + rc = sata_port_hardreset(ap, timing); + if (rc) { + ata_port_printk(ap, KERN_ERR, + "COMRESET failed (errno=%d)\n", rc); + return rc; + } /* TODO: phy layer with polling, timeouts, etc. */ if (ata_port_offline(ap)) { @@ -6159,6 +6190,7 @@ EXPORT_SYMBOL_GPL(__sata_phy_reset); EXPORT_SYMBOL_GPL(ata_bus_reset); EXPORT_SYMBOL_GPL(ata_std_prereset); EXPORT_SYMBOL_GPL(ata_std_softreset); +EXPORT_SYMBOL_GPL(sata_port_hardreset); EXPORT_SYMBOL_GPL(sata_std_hardreset); EXPORT_SYMBOL_GPL(ata_std_postreset); EXPORT_SYMBOL_GPL(ata_dev_classify); diff --git a/include/linux/libata.h b/include/linux/libata.h index d3bf7b936c3f..41fa0890dbfb 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -702,6 +702,8 @@ extern int sata_phy_debounce(struct ata_port *ap, const unsigned long *param); extern int sata_phy_resume(struct ata_port *ap, const unsigned long *param); extern int ata_std_prereset(struct ata_port *ap); extern int ata_std_softreset(struct ata_port *ap, unsigned int *classes); +extern int sata_port_hardreset(struct ata_port *ap, + const unsigned long *timing); extern int sata_std_hardreset(struct ata_port *ap, unsigned int *class); extern void ata_std_postreset(struct ata_port *ap, unsigned int *classes); extern void ata_port_disable(struct ata_port *); -- cgit v1.2.3 From efdaedc443e935eda82e9e78a6e65d1f993d242f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 1 Nov 2006 18:38:52 +0900 Subject: [PATCH] libata: implement ATA_EHI_PRINTINFO Implement ehi flag ATA_EHI_PRINTINFO. This flag is set when device configuration needs to print out device info. This used to be handled by @print_info argument to ata_dev_configure() but LLDs also need to know about it in ->dev_config() callback. This patch replaces @print_info w/ ATA_EHI_PRINTINFO and make sata_sil print workaround messages only on the initial configuration. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 10 ++++++---- drivers/ata/libata-eh.c | 7 +++++-- drivers/ata/libata.h | 2 +- drivers/ata/sata_sil.c | 11 +++++++---- include/linux/libata.h | 1 + 5 files changed, 20 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 315f46841f68..e294731a7edd 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1377,7 +1377,6 @@ static void ata_set_port_max_cmd_len(struct ata_port *ap) /** * ata_dev_configure - Configure the specified ATA/ATAPI device * @dev: Target device to configure - * @print_info: Enable device info printout * * Configure @dev according to @dev->id. Generic and low-level * driver specific fixups are also applied. @@ -1388,9 +1387,10 @@ static void ata_set_port_max_cmd_len(struct ata_port *ap) * RETURNS: * 0 on success, -errno otherwise */ -int ata_dev_configure(struct ata_device *dev, int print_info) +int ata_dev_configure(struct ata_device *dev) { struct ata_port *ap = dev->ap; + int print_info = ap->eh_context.i.flags & ATA_EHI_PRINTINFO; const u16 *id = dev->id; unsigned int xfer_mask; char revbuf[7]; /* XYZ-99\0 */ @@ -1638,7 +1638,9 @@ int ata_bus_probe(struct ata_port *ap) if (rc) goto fail; - rc = ata_dev_configure(dev, 1); + ap->eh_context.i.flags |= ATA_EHI_PRINTINFO; + rc = ata_dev_configure(dev); + ap->eh_context.i.flags &= ~ATA_EHI_PRINTINFO; if (rc) goto fail; } @@ -3045,7 +3047,7 @@ int ata_dev_revalidate(struct ata_device *dev, int post_reset) memcpy(dev->id, id, sizeof(id[0]) * ATA_ID_WORDS); /* configure device according to the new ID */ - rc = ata_dev_configure(dev, 0); + rc = ata_dev_configure(dev); if (rc == 0) return 0; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 02b2b2787d9b..7c446442616c 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1661,8 +1661,11 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, dev->class = ehc->classes[dev->devno]; rc = ata_dev_read_id(dev, &dev->class, 1, dev->id); - if (rc == 0) - rc = ata_dev_configure(dev, 1); + if (rc == 0) { + ehc->i.flags |= ATA_EHI_PRINTINFO; + rc = ata_dev_configure(dev); + ehc->i.flags &= ~ATA_EHI_PRINTINFO; + } if (rc) { dev->class = ATA_DEV_UNKNOWN; diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index c83300055ec5..e4ffb2e38992 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -54,7 +54,7 @@ extern unsigned int ata_do_simple_cmd(struct ata_device *dev, u8 cmd); extern int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class, int post_reset, u16 *id); extern int ata_dev_revalidate(struct ata_device *dev, int post_reset); -extern int ata_dev_configure(struct ata_device *dev, int print_info); +extern int ata_dev_configure(struct ata_device *dev); extern int sata_down_spd_limit(struct ata_port *ap); extern int sata_set_spd_needed(struct ata_port *ap); extern int ata_down_xfermask_limit(struct ata_device *dev, int force_pio0); diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c index ca8d99312472..f844a1faba18 100644 --- a/drivers/ata/sata_sil.c +++ b/drivers/ata/sata_sil.c @@ -534,6 +534,7 @@ static void sil_thaw(struct ata_port *ap) */ static void sil_dev_config(struct ata_port *ap, struct ata_device *dev) { + int print_info = ap->eh_context.i.flags & ATA_EHI_PRINTINFO; unsigned int n, quirks = 0; unsigned char model_num[41]; @@ -549,16 +550,18 @@ static void sil_dev_config(struct ata_port *ap, struct ata_device *dev) if (slow_down || ((ap->flags & SIL_FLAG_MOD15WRITE) && (quirks & SIL_QUIRK_MOD15WRITE))) { - ata_dev_printk(dev, KERN_INFO, "applying Seagate errata fix " - "(mod15write workaround)\n"); + if (print_info) + ata_dev_printk(dev, KERN_INFO, "applying Seagate " + "errata fix (mod15write workaround)\n"); dev->max_sectors = 15; return; } /* limit to udma5 */ if (quirks & SIL_QUIRK_UDMA5MAX) { - ata_dev_printk(dev, KERN_INFO, - "applying Maxtor errata fix %s\n", model_num); + if (print_info) + ata_dev_printk(dev, KERN_INFO, "applying Maxtor " + "errata fix %s\n", model_num); dev->udma_mask &= ATA_UDMA5; return; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 41fa0890dbfb..949484627e67 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -283,6 +283,7 @@ enum { ATA_EHI_QUIET = (1 << 3), /* be quiet */ ATA_EHI_DID_RESET = (1 << 16), /* already reset this port */ + ATA_EHI_PRINTINFO = (1 << 17), /* print configuration info */ ATA_EHI_RESET_MODIFIER_MASK = ATA_EHI_RESUME_LINK, -- cgit v1.2.3 From baa1e78a834c917984a4659fd282f712c17ee3bf Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 1 Nov 2006 18:39:27 +0900 Subject: [PATCH] libata: implement ATA_EHI_SETMODE and ATA_EHI_POST_SETMODE libata EH used to perform ata_set_mode() iff the EH session performed reset as indicated by ATA_EHI_DID_RESET. This is incorrect because ->dev_config() called by revalidation is allowed to modify transfer mode which ata_set_mode() should take care of. This patch implements the following two flags. * ATA_EHI_SETMODE: set during EH to schedule ata_set_mode(). Both new device attachment and revalidation set this flag. * ATA_EHI_POST_SETMODE: set while the device is revalidated after ata_set_mode(). Post-setmode revalidation is different from initial configuaration and EH revalidation in that ->dev_config() is not allowed tune transfer mode. LLD can use this flag to determine whether it's allowed to tune transfer mode. Note that POST_SETMODE ->dev_config() is guaranteed to be preceded by non-POST_SETMODE ->dev_config(). Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 3 +++ drivers/ata/libata-eh.c | 13 +++++++++++-- include/linux/libata.h | 2 ++ 3 files changed, 16 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index e294731a7edd..5028396029b3 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2158,6 +2158,7 @@ int ata_down_xfermask_limit(struct ata_device *dev, int force_pio0) static int ata_dev_set_mode(struct ata_device *dev) { + struct ata_eh_context *ehc = &dev->ap->eh_context; unsigned int err_mask; int rc; @@ -2172,7 +2173,9 @@ static int ata_dev_set_mode(struct ata_device *dev) return -EIO; } + ehc->i.flags |= ATA_EHI_POST_SETMODE; rc = ata_dev_revalidate(dev, 0); + ehc->i.flags &= ~ATA_EHI_POST_SETMODE; if (rc) return rc; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 7c446442616c..477648801a65 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1653,6 +1653,11 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, ata_eh_done(ap, dev, ATA_EH_REVALIDATE); + /* Configuration may have changed, reconfigure + * transfer mode. + */ + ehc->i.flags |= ATA_EHI_SETMODE; + /* schedule the scsi_rescan_device() here */ queue_work(ata_aux_wq, &(ap->scsi_rescan_task)); } else if (dev->class == ATA_DEV_UNKNOWN && @@ -1675,6 +1680,9 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, spin_lock_irqsave(ap->lock, flags); ap->pflags |= ATA_PFLAG_SCSI_HOTPLUG; spin_unlock_irqrestore(ap->lock, flags); + + /* new device discovered, configure transfer mode */ + ehc->i.flags |= ATA_EHI_SETMODE; } } @@ -1990,13 +1998,14 @@ static int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset, if (rc) goto dev_fail; - /* configure transfer mode if the port has been reset */ - if (ehc->i.flags & ATA_EHI_DID_RESET) { + /* configure transfer mode if necessary */ + if (ehc->i.flags & ATA_EHI_SETMODE) { rc = ata_set_mode(ap, &dev); if (rc) { down_xfermask = 1; goto dev_fail; } + ehc->i.flags &= ~ATA_EHI_SETMODE; } /* suspend devices */ diff --git a/include/linux/libata.h b/include/linux/libata.h index 949484627e67..0d0ddeaee73f 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -284,6 +284,8 @@ enum { ATA_EHI_DID_RESET = (1 << 16), /* already reset this port */ ATA_EHI_PRINTINFO = (1 << 17), /* print configuration info */ + ATA_EHI_SETMODE = (1 << 18), /* configure transfer mode */ + ATA_EHI_POST_SETMODE = (1 << 19), /* revaildating after setmode */ ATA_EHI_RESET_MODIFIER_MASK = ATA_EHI_RESUME_LINK, -- cgit v1.2.3 From 6fc49adb9417b9c793e8f88d485387bb89ceb733 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 11 Nov 2006 20:10:45 +0900 Subject: [PATCH] libata: use FLUSH_EXT only when driver is larger than LBA28 limit Many drives support LBA48 even when its capacity is smaller than 1<<28, as LBA48 is required for many functionalities. FLUSH_EXT is mandatory for drives w/ LBA48 support. Interestingly, at least one of such drives (ST960812A) has problems dealing with FLUSH_EXT. It eventually completes the command but takes around 7 seconds to finish in many cases thus drastically slowing down IO transactions. This seems to be a firmware bug which sneaked into production probably because no other ATA driver including linux IDE issues FLUSH_EXT to drives which report support for LBA48 & FLUSH_EXT but is smaller than 1<<28 blocks. This patch adds ATA_DFLAG_FLUSH_EXT which is set iff the drive supports LBA48 & FLUSH_EXT and is larger than LBA28 limit. Both cache flush paths are updated to issue FLUSH_EXT only when the flag is set. Note that the changed behavior is more inline with the rest of libata. libata prefers shorter commands whenever possible. Signed-off-by: Tejun Heo Cc: Danny Kukawka Cc: Stefan Seyfried Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 6 +++++- drivers/ata/libata-scsi.c | 3 +-- include/linux/libata.h | 1 + 3 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index f531a89c26c9..d94b8a02c340 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1457,6 +1457,10 @@ int ata_dev_configure(struct ata_device *dev) if (ata_id_has_lba48(id)) { dev->flags |= ATA_DFLAG_LBA48; lba_desc = "LBA48"; + + if (dev->n_sectors >= (1UL << 28) && + ata_id_has_flush_ext(id)) + dev->flags |= ATA_DFLAG_FLUSH_EXT; } /* config NCQ */ @@ -5128,7 +5132,7 @@ int ata_flush_cache(struct ata_device *dev) if (!ata_try_flush_cache(dev)) return 0; - if (ata_id_has_flush_ext(dev->id)) + if (dev->flags & ATA_DFLAG_FLUSH_EXT) cmd = ATA_CMD_FLUSH_EXT; else cmd = ATA_CMD_FLUSH; diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index f13dd07a5331..7a55c2e4ea6e 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1027,8 +1027,7 @@ static unsigned int ata_scsi_flush_xlat(struct ata_queued_cmd *qc, const u8 *scs tf->flags |= ATA_TFLAG_DEVICE; tf->protocol = ATA_PROT_NODATA; - if ((qc->dev->flags & ATA_DFLAG_LBA48) && - (ata_id_has_flush_ext(qc->dev->id))) + if (qc->dev->flags & ATA_DFLAG_FLUSH_EXT) tf->command = ATA_CMD_FLUSH_EXT; else tf->command = ATA_CMD_FLUSH; diff --git a/include/linux/libata.h b/include/linux/libata.h index 0d0ddeaee73f..36e233cc3886 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -140,6 +140,7 @@ enum { ATA_DFLAG_LBA48 = (1 << 1), /* device supports LBA48 */ ATA_DFLAG_CDB_INTR = (1 << 2), /* device asserts INTRQ when ready for CDB */ ATA_DFLAG_NCQ = (1 << 3), /* device supports NCQ */ + ATA_DFLAG_FLUSH_EXT = (1 << 4), /* do FLUSH_EXT instead of FLUSH */ ATA_DFLAG_CFG_MASK = (1 << 8) - 1, ATA_DFLAG_PIO = (1 << 8), /* device limited to PIO mode */ -- cgit v1.2.3 From 55a8e2c83ce50548dfef74bb19dfe2b809cb3099 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 10 Nov 2006 18:08:10 +0900 Subject: [PATCH] libata: implement presence detection via polling IDENTIFY On some controllers (ICHs in piix mode), there is *NO* reliable way to determine device presence other than issuing IDENTIFY and see how the transaction proceeds by watching the TF status register. libata acted this way before irq-pio and phantom devices caused very little problem but now that IDENTIFY is performed using IRQ drive PIO, such phantom devices now result in multiple 30sec timeouts during boot. This patch implements ATA_FLAG_DETECT_POLLING. If a LLD sets this flag, libata core issues the initial IDENTIFY in polling mode and if the initial data transfer fails w/ HSM violation, the port is considered to be empty thus replicating the old libata and IDE behavior. Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 19 +++++++++++++++++-- drivers/ata/libata-eh.c | 23 ++++++++++++++++++----- drivers/ata/libata.h | 2 ++ include/linux/libata.h | 3 +++ 4 files changed, 40 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 090abe443820..21f8d61e5879 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1272,9 +1272,20 @@ int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class, tf.protocol = ATA_PROT_PIO; + /* presence detection using polling IDENTIFY? */ + if (flags & ATA_READID_DETECT) + tf.flags |= ATA_TFLAG_POLLING; + err_mask = ata_exec_internal(dev, &tf, NULL, DMA_FROM_DEVICE, id, sizeof(id[0]) * ATA_ID_WORDS); if (err_mask) { + if ((flags & ATA_READID_DETECT) && + (err_mask & AC_ERR_NODEV_HINT)) { + DPRINTK("ata%u.%d: NODEV after polling detection\n", + ap->id, dev->devno); + return -ENOENT; + } + rc = -EIO; reason = "I/O error"; goto err_out; @@ -4285,8 +4296,12 @@ fsm_start: /* device stops HSM for abort/error */ qc->err_mask |= AC_ERR_DEV; else - /* HSM violation. Let EH handle this */ - qc->err_mask |= AC_ERR_HSM; + /* HSM violation. Let EH handle this. + * Phantom devices also trigger this + * condition. Mark hint. + */ + qc->err_mask |= AC_ERR_HSM | + AC_ERR_NODEV_HINT; ap->hsm_task_state = HSM_ST_ERR; goto fsm_start; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 755fc68b5374..e69f3df2ea39 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1667,12 +1667,23 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, ata_class_enabled(ehc->classes[dev->devno])) { dev->class = ehc->classes[dev->devno]; + if (ap->flags & ATA_FLAG_DETECT_POLLING) + readid_flags |= ATA_READID_DETECT; + rc = ata_dev_read_id(dev, &dev->class, readid_flags, dev->id); if (rc == 0) { ehc->i.flags |= ATA_EHI_PRINTINFO; rc = ata_dev_configure(dev); ehc->i.flags &= ~ATA_EHI_PRINTINFO; + } else if (rc == -ENOENT) { + /* IDENTIFY was issued to non-existent + * device. No need to reset. Just + * thaw and kill the device. + */ + ata_eh_thaw_port(ap); + dev->class = ATA_DEV_UNKNOWN; + rc = 0; } if (rc) { @@ -1680,12 +1691,14 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, break; } - spin_lock_irqsave(ap->lock, flags); - ap->pflags |= ATA_PFLAG_SCSI_HOTPLUG; - spin_unlock_irqrestore(ap->lock, flags); + if (ata_dev_enabled(dev)) { + spin_lock_irqsave(ap->lock, flags); + ap->pflags |= ATA_PFLAG_SCSI_HOTPLUG; + spin_unlock_irqrestore(ap->lock, flags); - /* new device discovered, configure transfer mode */ - ehc->i.flags |= ATA_EHI_SETMODE; + /* new device discovered, configure xfermode */ + ehc->i.flags |= ATA_EHI_SETMODE; + } } } diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index bb98390aa01a..be2ac39f013b 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -42,6 +42,8 @@ struct ata_scsi_args { enum { /* flags for ata_dev_read_id() */ ATA_READID_POSTRESET = (1 << 0), /* reading ID after reset */ + ATA_READID_DETECT = (1 << 1), /* perform presence detection + * using polling IDENTIFY */ }; extern struct workqueue_struct *ata_aux_wq; diff --git a/include/linux/libata.h b/include/linux/libata.h index 36e233cc3886..9080789913f7 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -176,6 +176,8 @@ enum { ATA_FLAG_SKIP_D2H_BSY = (1 << 12), /* can't wait for the first D2H * Register FIS clearing BSY */ ATA_FLAG_DEBUGMSG = (1 << 13), + ATA_FLAG_DETECT_POLLING = (1 << 14), /* detect device presence by + * polling IDENTIFY */ /* The following flag belongs to ap->pflags but is kept in * ap->flags because it's referenced in many LLDs and will be @@ -335,6 +337,7 @@ enum ata_completion_errors { AC_ERR_SYSTEM = (1 << 6), /* system error */ AC_ERR_INVALID = (1 << 7), /* invalid argument */ AC_ERR_OTHER = (1 << 8), /* unknown */ + AC_ERR_NODEV_HINT = (1 << 9), /* polling device detection hint */ }; /* forward declarations */ -- cgit v1.2.3 From 13df29f69749a61b5209d52b71fcbf7300e5d6fb Mon Sep 17 00:00:00 2001 From: Maciej W. Rozycki Date: Tue, 3 Oct 2006 16:18:28 +0100 Subject: [PATCH] 2.6.18: sb1250-mac: Missing inclusions from The uses some types and macros defined in , , and , but fails to include these headers. Signed-off-by: Maciej W. Rozycki patch-mips-2.6.18-20060920-include-phy-16 Signed-off-by: Jeff Garzik --- include/linux/phy.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/phy.h b/include/linux/phy.h index 9447a57ee8a9..ce8bc80b3c86 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -20,6 +20,10 @@ #include #include +#include +#include +#include +#include #define PHY_BASIC_FEATURES (SUPPORTED_10baseT_Half | \ SUPPORTED_10baseT_Full | \ -- cgit v1.2.3 From fe75f7471ba5604fe65435f717e3612a482c28cb Mon Sep 17 00:00:00 2001 From: Christian Lamparter Date: Mon, 2 Oct 2006 19:55:22 +0200 Subject: [PATCH] wext: extend MLME support This patch adds two new defines for the SIOCSIWMLME to cover all kinds MLMEs (well, except REASSOC) through a ioctl. Signed-off-by: Christian Lamparter Signed-off-by: John W. Linville --- include/linux/wireless.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/wireless.h b/include/linux/wireless.h index a50a0130fd9e..7c269f4992eb 100644 --- a/include/linux/wireless.h +++ b/include/linux/wireless.h @@ -546,6 +546,8 @@ /* MLME requests (SIOCSIWMLME / struct iw_mlme) */ #define IW_MLME_DEAUTH 0 #define IW_MLME_DISASSOC 1 +#define IW_MLME_AUTH 2 +#define IW_MLME_ASSOC 3 /* SIOCSIWAUTH/SIOCGIWAUTH struct iw_param flags */ #define IW_AUTH_INDEX 0x0FFF -- cgit v1.2.3 From 52d78d63310d9818458fd9800d24a4d5425aeac7 Mon Sep 17 00:00:00 2001 From: Ayaz Abdulla Date: Mon, 6 Nov 2006 00:43:39 -0800 Subject: [PATCH] forcedeth: add new NVIDIA pci ids Add pci device ids for the NVIDIA MCP67 chip. Signed-off-by: Ayaz Abdulla Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- include/linux/pci_ids.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index e060a7637947..fd5033b8a927 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1213,6 +1213,10 @@ #define PCI_DEVICE_ID_NVIDIA_NVENET_21 0x0451 #define PCI_DEVICE_ID_NVIDIA_NVENET_22 0x0452 #define PCI_DEVICE_ID_NVIDIA_NVENET_23 0x0453 +#define PCI_DEVICE_ID_NVIDIA_NVENET_24 0x054C +#define PCI_DEVICE_ID_NVIDIA_NVENET_25 0x054D +#define PCI_DEVICE_ID_NVIDIA_NVENET_26 0x054E +#define PCI_DEVICE_ID_NVIDIA_NVENET_27 0x054F #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE 0x0560 #define PCI_VENDOR_ID_IMS 0x10e0 -- cgit v1.2.3 From f789dfdc44d5bbc04fb7f06e1e4eb682169acaaf Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Thu, 30 Nov 2006 04:27:00 -0700 Subject: [PATCH] mv643xx_eth: fix unbalanced parentheses in macros Signed-off-by: Mariusz Kozlowski Signed-off-by: Dale Farnsworth Signed-off-by: Jeff Garzik --- include/linux/mv643xx.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mv643xx.h b/include/linux/mv643xx.h index edfa012fad3a..aff25c000abf 100644 --- a/include/linux/mv643xx.h +++ b/include/linux/mv643xx.h @@ -724,7 +724,7 @@ #define MV643XX_ETH_RX_FIFO_URGENT_THRESHOLD_REG(port) (0x2470 + (port<<10)) #define MV643XX_ETH_TX_FIFO_URGENT_THRESHOLD_REG(port) (0x2474 + (port<<10)) #define MV643XX_ETH_RX_MINIMAL_FRAME_SIZE_REG(port) (0x247c + (port<<10)) -#define MV643XX_ETH_RX_DISCARDED_FRAMES_COUNTER(port) (0x2484 + (port<<10) +#define MV643XX_ETH_RX_DISCARDED_FRAMES_COUNTER(port) (0x2484 + (port<<10)) #define MV643XX_ETH_PORT_DEBUG_0_REG(port) (0x248c + (port<<10)) #define MV643XX_ETH_PORT_DEBUG_1_REG(port) (0x2490 + (port<<10)) #define MV643XX_ETH_PORT_INTERNAL_ADDR_ERROR_REG(port) (0x2494 + (port<<10)) @@ -1135,7 +1135,7 @@ struct mv64xxx_i2c_pdata { #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_1 (1<<19) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_2 (1<<20) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_3 ((1<<20) | (1<<19)) -#define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_4 ((1<<21) +#define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_4 (1<<21) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_5 ((1<<21) | (1<<19)) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_6 ((1<<21) | (1<<20)) #define MV643XX_ETH_DEFAULT_RX_UDP_QUEUE_7 ((1<<21) | (1<<20) | (1<<19)) -- cgit v1.2.3 From e8a2b6a4207332a2d59628a12cece9e8c1d769e4 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Fri, 1 Dec 2006 12:01:06 -0600 Subject: [PATCH] PHY: Add support for configuring the PHY connection interface Most PHYs connect to an ethernet controller over a GMII or MII interface. However, a growing number are connected over different interfaces, such as RGMII or SGMII. The ethernet driver will tell the PHY what type of connection it is by setting it manually, or passing it in through phy_connect (or phy_attach). Changes include: * Updates to documentation * Updates to PHY Lib consumers * Changes to PHY Lib to add interface support * Some minor changes to whitespace in phy.h * gianfar driver now detects interface and passes appropriate value to PHY Lib Signed-off-by: Andrew Fleming Signed-off-by: Jeff Garzik --- Documentation/networking/phy.txt | 13 +++++++++---- drivers/net/au1000_eth.c | 3 ++- drivers/net/fs_enet/fs_enet-main.c | 3 ++- drivers/net/gianfar.c | 39 ++++++++++++++++++++++++++++++++++++-- drivers/net/gianfar.h | 3 +++ drivers/net/phy/phy_device.c | 29 +++++++++++++++++++--------- include/linux/phy.h | 26 +++++++++++++++++++------ 7 files changed, 93 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/Documentation/networking/phy.txt b/Documentation/networking/phy.txt index 29ccae409031..0bc95eab1512 100644 --- a/Documentation/networking/phy.txt +++ b/Documentation/networking/phy.txt @@ -1,7 +1,7 @@ ------- PHY Abstraction Layer -(Updated 2005-07-21) +(Updated 2006-11-30) Purpose @@ -97,11 +97,12 @@ Letting the PHY Abstraction Layer do Everything Next, you need to know the device name of the PHY connected to this device. The name will look something like, "phy0:0", where the first number is the - bus id, and the second is the PHY's address on that bus. + bus id, and the second is the PHY's address on that bus. Typically, + the bus is responsible for making its ID unique. Now, to connect, just call this function: - phydev = phy_connect(dev, phy_name, &adjust_link, flags); + phydev = phy_connect(dev, phy_name, &adjust_link, flags, interface); phydev is a pointer to the phy_device structure which represents the PHY. If phy_connect is successful, it will return the pointer. dev, here, is the @@ -115,6 +116,10 @@ Letting the PHY Abstraction Layer do Everything This is useful if the system has put hardware restrictions on the PHY/controller, of which the PHY needs to be aware. + interface is a u32 which specifies the connection type used + between the controller and the PHY. Examples are GMII, MII, + RGMII, and SGMII. For a full list, see include/linux/phy.h + Now just make sure that phydev->supported and phydev->advertising have any values pruned from them which don't make sense for your controller (a 10/100 controller may be connected to a gigabit capable PHY, so you would need to @@ -191,7 +196,7 @@ Doing it all yourself start, or disables then frees them for stop. struct phy_device * phy_attach(struct net_device *dev, const char *phy_id, - u32 flags); + u32 flags, phy_interface_t interface); Attaches a network device to a particular PHY, binding the PHY to a generic driver if none was found during bus initialization. Passes in diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 7db3c8af0894..f0b6879a1c7d 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -360,7 +360,8 @@ static int mii_probe (struct net_device *dev) BUG_ON(!phydev); BUG_ON(phydev->attached_dev); - phydev = phy_connect(dev, phydev->dev.bus_id, &au1000_adjust_link, 0); + phydev = phy_connect(dev, phydev->dev.bus_id, &au1000_adjust_link, 0, + PHY_INTERFACE_MODE_MII); if (IS_ERR(phydev)) { printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name); diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c index cb3958704a87..889d3a13e95e 100644 --- a/drivers/net/fs_enet/fs_enet-main.c +++ b/drivers/net/fs_enet/fs_enet-main.c @@ -779,7 +779,8 @@ static int fs_init_phy(struct net_device *dev) fep->oldspeed = 0; fep->oldduplex = -1; if(fep->fpi->bus_id) - phydev = phy_connect(dev, fep->fpi->bus_id, &fs_adjust_link, 0); + phydev = phy_connect(dev, fep->fpi->bus_id, &fs_adjust_link, 0, + PHY_INTERFACE_MODE_MII); else { printk("No phy bus ID specified in BSP code\n"); return -EINVAL; diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 6bf18c82083d..baa35144134c 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -9,7 +9,7 @@ * Author: Andy Fleming * Maintainer: Kumar Gala * - * Copyright (c) 2002-2004 Freescale Semiconductor, Inc. + * Copyright (c) 2002-2006 Freescale Semiconductor, Inc. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -398,6 +398,38 @@ static int gfar_remove(struct platform_device *pdev) } +/* Reads the controller's registers to determine what interface + * connects it to the PHY. + */ +static phy_interface_t gfar_get_interface(struct net_device *dev) +{ + struct gfar_private *priv = netdev_priv(dev); + u32 ecntrl = gfar_read(&priv->regs->ecntrl); + + if (ecntrl & ECNTRL_SGMII_MODE) + return PHY_INTERFACE_MODE_SGMII; + + if (ecntrl & ECNTRL_TBI_MODE) { + if (ecntrl & ECNTRL_REDUCED_MODE) + return PHY_INTERFACE_MODE_RTBI; + else + return PHY_INTERFACE_MODE_TBI; + } + + if (ecntrl & ECNTRL_REDUCED_MODE) { + if (ecntrl & ECNTRL_REDUCED_MII_MODE) + return PHY_INTERFACE_MODE_RMII; + else + return PHY_INTERFACE_MODE_RGMII; + } + + if (priv->einfo->device_flags & FSL_GIANFAR_DEV_HAS_GIGABIT) + return PHY_INTERFACE_MODE_GMII; + + return PHY_INTERFACE_MODE_MII; +} + + /* Initializes driver's PHY state, and attaches to the PHY. * Returns 0 on success. */ @@ -409,6 +441,7 @@ static int init_phy(struct net_device *dev) SUPPORTED_1000baseT_Full : 0; struct phy_device *phydev; char phy_id[BUS_ID_SIZE]; + phy_interface_t interface; priv->oldlink = 0; priv->oldspeed = 0; @@ -416,7 +449,9 @@ static int init_phy(struct net_device *dev) snprintf(phy_id, BUS_ID_SIZE, PHY_ID_FMT, priv->einfo->bus_id, priv->einfo->phy_id); - phydev = phy_connect(dev, phy_id, &adjust_link, 0); + interface = gfar_get_interface(dev); + + phydev = phy_connect(dev, phy_id, &adjust_link, 0, interface); if (IS_ERR(phydev)) { printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name); diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h index 9e81a50cf2be..39e9e321fcbc 100644 --- a/drivers/net/gianfar.h +++ b/drivers/net/gianfar.h @@ -160,7 +160,10 @@ extern const char gfar_driver_version[]; #define ECNTRL_INIT_SETTINGS 0x00001000 #define ECNTRL_TBI_MODE 0x00000020 +#define ECNTRL_REDUCED_MODE 0x00000010 #define ECNTRL_R100 0x00000008 +#define ECNTRL_REDUCED_MII_MODE 0x00000004 +#define ECNTRL_SGMII_MODE 0x00000002 #define MRBLR_INIT_SETTINGS DEFAULT_RX_BUFFER_SIZE diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 2a08b2b62c4c..b01fc70a57db 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -59,6 +59,7 @@ struct phy_device* phy_device_create(struct mii_bus *bus, int addr, int phy_id) dev->duplex = -1; dev->pause = dev->asym_pause = 0; dev->link = 1; + dev->interface = PHY_INTERFACE_MODE_GMII; dev->autoneg = AUTONEG_ENABLE; @@ -137,11 +138,12 @@ void phy_prepare_link(struct phy_device *phydev, * the desired functionality. */ struct phy_device * phy_connect(struct net_device *dev, const char *phy_id, - void (*handler)(struct net_device *), u32 flags) + void (*handler)(struct net_device *), u32 flags, + u32 interface) { struct phy_device *phydev; - phydev = phy_attach(dev, phy_id, flags); + phydev = phy_attach(dev, phy_id, flags, interface); if (IS_ERR(phydev)) return phydev; @@ -186,7 +188,7 @@ static int phy_compare_id(struct device *dev, void *data) } struct phy_device *phy_attach(struct net_device *dev, - const char *phy_id, u32 flags) + const char *phy_id, u32 flags, u32 interface) { struct bus_type *bus = &mdio_bus_type; struct phy_device *phydev; @@ -231,6 +233,20 @@ struct phy_device *phy_attach(struct net_device *dev, phydev->dev_flags = flags; + phydev->interface = interface; + + /* Do initial configuration here, now that + * we have certain key parameters + * (dev_flags and interface) */ + if (phydev->drv->config_init) { + int err; + + err = phydev->drv->config_init(phydev); + + if (err < 0) + return ERR_PTR(err); + } + return phydev; } EXPORT_SYMBOL(phy_attach); @@ -612,13 +628,8 @@ static int phy_probe(struct device *dev) spin_unlock(&phydev->lock); - if (err < 0) - return err; - - if (phydev->drv->config_init) - err = phydev->drv->config_init(phydev); - return err; + } static int phy_remove(struct device *dev) diff --git a/include/linux/phy.h b/include/linux/phy.h index ce8bc80b3c86..edd4c88ca7d8 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -47,15 +47,26 @@ #define PHY_HAS_INTERRUPT 0x00000001 #define PHY_HAS_MAGICANEG 0x00000002 +/* Interface Mode definitions */ +typedef enum { + PHY_INTERFACE_MODE_MII, + PHY_INTERFACE_MODE_GMII, + PHY_INTERFACE_MODE_SGMII, + PHY_INTERFACE_MODE_TBI, + PHY_INTERFACE_MODE_RMII, + PHY_INTERFACE_MODE_RGMII, + PHY_INTERFACE_MODE_RTBI +} phy_interface_t; + #define MII_BUS_MAX 4 -#define PHY_INIT_TIMEOUT 100000 +#define PHY_INIT_TIMEOUT 100000 #define PHY_STATE_TIME 1 #define PHY_FORCE_TIMEOUT 10 #define PHY_AN_TIMEOUT 10 -#define PHY_MAX_ADDR 32 +#define PHY_MAX_ADDR 32 /* Used when trying to connect to a specific phy (mii bus id:phy device id) */ #define PHY_ID_FMT "%x:%02x" @@ -87,8 +98,8 @@ struct mii_bus { int *irq; }; -#define PHY_INTERRUPT_DISABLED 0x0 -#define PHY_INTERRUPT_ENABLED 0x80000000 +#define PHY_INTERRUPT_DISABLED 0x0 +#define PHY_INTERRUPT_ENABLED 0x80000000 /* PHY state machine states: * @@ -230,6 +241,8 @@ struct phy_device { u32 dev_flags; + phy_interface_t interface; + /* Bus address of the PHY (0-32) */ int addr; @@ -345,9 +358,10 @@ struct phy_device* get_phy_device(struct mii_bus *bus, int addr); int phy_clear_interrupt(struct phy_device *phydev); int phy_config_interrupt(struct phy_device *phydev, u32 interrupts); struct phy_device * phy_attach(struct net_device *dev, - const char *phy_id, u32 flags); + const char *phy_id, u32 flags, phy_interface_t interface); struct phy_device * phy_connect(struct net_device *dev, const char *phy_id, - void (*handler)(struct net_device *), u32 flags); + void (*handler)(struct net_device *), u32 flags, + phy_interface_t interface); void phy_disconnect(struct phy_device *phydev); void phy_detach(struct phy_device *phydev); void phy_start(struct phy_device *phydev); -- cgit v1.2.3 From 3e6c8cd5669c1202fe806ce3e13d701f20a71c7e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:19:09 -0800 Subject: [TIPC]: endianness annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/tipc_config.h | 32 ++++++++++++++++---------------- include/net/tipc/tipc_bearer.h | 2 +- include/net/tipc/tipc_msg.h | 2 +- net/tipc/config.c | 32 +++++++++++--------------------- net/tipc/dbg.c | 3 +-- net/tipc/name_distr.c | 10 +++++----- net/tipc/node.c | 9 +++------ 7 files changed, 38 insertions(+), 52 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tipc_config.h b/include/linux/tipc_config.h index 33a653913d94..b0c916d1f375 100644 --- a/include/linux/tipc_config.h +++ b/include/linux/tipc_config.h @@ -194,34 +194,34 @@ struct tipc_node_info { - __u32 addr; /* network address of node */ - __u32 up; /* 0=down, 1= up */ + __be32 addr; /* network address of node */ + __be32 up; /* 0=down, 1= up */ }; struct tipc_link_info { - __u32 dest; /* network address of peer node */ - __u32 up; /* 0=down, 1=up */ + __be32 dest; /* network address of peer node */ + __be32 up; /* 0=down, 1=up */ char str[TIPC_MAX_LINK_NAME]; /* link name */ }; struct tipc_bearer_config { - __u32 priority; /* Range [1,31]. Override per link */ - __u32 detect_scope; + __be32 priority; /* Range [1,31]. Override per link */ + __be32 detect_scope; char name[TIPC_MAX_BEARER_NAME]; }; struct tipc_link_config { - __u32 value; + __be32 value; char name[TIPC_MAX_LINK_NAME]; }; #define TIPC_NTQ_ALLTYPES 0x80000000 struct tipc_name_table_query { - __u32 depth; /* 1:type, 2:+name info, 3:+port info, 4+:+debug info */ - __u32 type; /* {t,l,u} info ignored if high bit of "depth" is set */ - __u32 lowbound; /* (i.e. displays all entries of name table) */ - __u32 upbound; + __be32 depth; /* 1:type, 2:+name info, 3:+port info, 4+:+debug info */ + __be32 type; /* {t,l,u} info ignored if high bit of "depth" is set */ + __be32 lowbound; /* (i.e. displays all entries of name table) */ + __be32 upbound; }; /* @@ -262,8 +262,8 @@ struct tipc_route_info { */ struct tlv_desc { - __u16 tlv_len; /* TLV length (descriptor + value) */ - __u16 tlv_type; /* TLV identifier */ + __be16 tlv_len; /* TLV length (descriptor + value) */ + __be16 tlv_type; /* TLV identifier */ }; #define TLV_ALIGNTO 4 @@ -377,9 +377,9 @@ struct tipc_genlmsghdr { struct tipc_cfg_msg_hdr { - __u32 tcm_len; /* Message length (including header) */ - __u16 tcm_type; /* Command type */ - __u16 tcm_flags; /* Additional flags */ + __be32 tcm_len; /* Message length (including header) */ + __be16 tcm_type; /* Command type */ + __be16 tcm_flags; /* Additional flags */ char tcm_reserved[8]; /* Unused */ }; diff --git a/include/net/tipc/tipc_bearer.h b/include/net/tipc/tipc_bearer.h index e07136d74c2f..2151a80cdf30 100644 --- a/include/net/tipc/tipc_bearer.h +++ b/include/net/tipc/tipc_bearer.h @@ -58,7 +58,7 @@ */ struct tipc_media_addr { - __u32 type; /* bearer type (network byte order) */ + __be32 type; /* bearer type (network byte order) */ union { __u8 eth_addr[6]; /* 48 bit Ethernet addr (byte array) */ #if 0 diff --git a/include/net/tipc/tipc_msg.h b/include/net/tipc/tipc_msg.h index 4d096eebc93f..fb42eb7a86a5 100644 --- a/include/net/tipc/tipc_msg.h +++ b/include/net/tipc/tipc_msg.h @@ -40,7 +40,7 @@ #ifdef __KERNEL__ struct tipc_msg { - u32 hdr[15]; + __be32 hdr[15]; }; diff --git a/net/tipc/config.c b/net/tipc/config.c index ed1351ed05e1..458a2c46cef3 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -107,7 +107,7 @@ int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) { struct sk_buff *buf; - u32 value_net; + __be32 value_net; buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(value))); if (buf) { @@ -284,8 +284,7 @@ static struct sk_buff *cfg_set_own_addr(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - addr = *(u32 *)TLV_DATA(req_tlv_area); - addr = ntohl(addr); + addr = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (addr == tipc_own_addr) return tipc_cfg_reply_none(); if (!tipc_addr_node_valid(addr)) @@ -319,8 +318,7 @@ static struct sk_buff *cfg_set_remote_mng(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); tipc_remote_management = (value != 0); return tipc_cfg_reply_none(); } @@ -332,8 +330,7 @@ static struct sk_buff *cfg_set_max_publications(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 65535)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max publications must be 1-65535)"); @@ -348,8 +345,7 @@ static struct sk_buff *cfg_set_max_subscriptions(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 65535)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max subscriptions must be 1-65535"); @@ -363,8 +359,7 @@ static struct sk_buff *cfg_set_max_ports(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_ports) return tipc_cfg_reply_none(); if (value != delimit(value, 127, 65535)) @@ -383,8 +378,7 @@ static struct sk_buff *cfg_set_max_zones(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_zones) return tipc_cfg_reply_none(); if (value != delimit(value, 1, 255)) @@ -403,8 +397,7 @@ static struct sk_buff *cfg_set_max_clusters(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 1, 1)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (max clusters fixed at 1)"); @@ -417,8 +410,7 @@ static struct sk_buff *cfg_set_max_nodes(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_max_nodes) return tipc_cfg_reply_none(); if (value != delimit(value, 8, 2047)) @@ -437,8 +429,7 @@ static struct sk_buff *cfg_set_max_slaves(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != 0) return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (max secondary nodes fixed at 0)"); @@ -451,8 +442,7 @@ static struct sk_buff *cfg_set_netid(void) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value == tipc_net_id) return tipc_cfg_reply_none(); if (value != delimit(value, 1, 9999)) diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c index d8af4c28695d..627f99b7afdf 100644 --- a/net/tipc/dbg.c +++ b/net/tipc/dbg.c @@ -393,8 +393,7 @@ struct sk_buff *tipc_log_resize(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = *(u32 *)TLV_DATA(req_tlv_area); - value = ntohl(value); + value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (value != delimit(value, 0, 32768)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (log size must be 0-32768)"); diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 03bd659c43ca..7bf87cb26ef3 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -66,11 +66,11 @@ */ struct distr_item { - u32 type; - u32 lower; - u32 upper; - u32 ref; - u32 key; + __be32 type; + __be32 lower; + __be32 upper; + __be32 ref; + __be32 key; }; /** diff --git a/net/tipc/node.c b/net/tipc/node.c index 886bda5e88db..106cd0dfac78 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -597,8 +597,7 @@ struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - domain = *(u32 *)TLV_DATA(req_tlv_area); - domain = ntohl(domain); + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (!tipc_addr_domain_valid(domain)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); @@ -642,8 +641,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - domain = *(u32 *)TLV_DATA(req_tlv_area); - domain = ntohl(domain); + domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); if (!tipc_addr_domain_valid(domain)) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); @@ -664,8 +662,7 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) /* Add TLV for broadcast link */ - link_info.dest = tipc_own_addr & 0xfffff00; - link_info.dest = htonl(link_info.dest); + link_info.dest = htonl(tipc_own_addr & 0xfffff00); link_info.up = htonl(1); sprintf(link_info.str, tipc_bclink_name); tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); -- cgit v1.2.3 From 5a874db4d9bfd8a4c6324d844a4d1c7cfa5cf2c4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:19:38 -0800 Subject: [NET]: ipconfig and nfsroot annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- fs/nfs/nfsroot.c | 13 +++--- include/linux/nfs_fs.h | 2 +- include/net/ipconfig.h | 8 ++-- net/ipv4/ipconfig.c | 105 +++++++++++++++++++++++++------------------------ 4 files changed, 64 insertions(+), 64 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index 8dfefe41a8da..75f819dc0255 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -98,7 +98,7 @@ static char nfs_root_name[256] __initdata = ""; /* Address of NFS server */ -static __u32 servaddr __initdata = 0; +static __be32 servaddr __initdata = 0; /* Name of directory to mount */ static char nfs_path[NFS_MAXPATHLEN] __initdata = { 0, }; @@ -327,7 +327,7 @@ static int __init root_nfs_name(char *name) */ static int __init root_nfs_addr(void) { - if ((servaddr = root_server_addr) == INADDR_NONE) { + if ((servaddr = root_server_addr) == htonl(INADDR_NONE)) { printk(KERN_ERR "Root-NFS: No NFS server available, giving up.\n"); return -1; } @@ -411,7 +411,7 @@ __setup("nfsroot=", nfs_root_setup); * Construct sockaddr_in from address and port number. */ static inline void -set_sockaddr(struct sockaddr_in *sin, __u32 addr, __u16 port) +set_sockaddr(struct sockaddr_in *sin, __be32 addr, __be16 port) { sin->sin_family = AF_INET; sin->sin_addr.s_addr = addr; @@ -468,14 +468,13 @@ static int __init root_nfs_ports(void) dprintk("Root-NFS: Portmapper on server returned %d " "as nfsd port\n", port); } - nfs_port = htons(nfs_port); if ((port = root_nfs_getport(NFS_MNT_PROGRAM, mountd_ver, proto)) < 0) { printk(KERN_ERR "Root-NFS: Unable to get mountd port " "number from server, using default\n"); port = mountd_port; } - mount_port = htons(port); + mount_port = port; dprintk("Root-NFS: mountd port is %d\n", port); return 0; @@ -496,7 +495,7 @@ static int __init root_nfs_get_handle(void) int version = (nfs_data.flags & NFS_MOUNT_VER3) ? NFS_MNT3_VERSION : NFS_MNT_VERSION; - set_sockaddr(&sin, servaddr, mount_port); + set_sockaddr(&sin, servaddr, htons(mount_port)); status = nfsroot_mount(&sin, nfs_path, &fh, version, protocol); if (status < 0) printk(KERN_ERR "Root-NFS: Server returned error %d " @@ -519,6 +518,6 @@ void * __init nfs_root_data(void) || root_nfs_ports() < 0 || root_nfs_get_handle() < 0) return NULL; - set_sockaddr((struct sockaddr_in *) &nfs_data.addr, servaddr, nfs_port); + set_sockaddr((struct sockaddr_in *) &nfs_data.addr, servaddr, htons(nfs_port)); return (void*)&nfs_data; } diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 45228c1a1195..625ffea98561 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -318,7 +318,7 @@ extern void put_nfs_open_context(struct nfs_open_context *ctx); extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode); /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */ -extern u32 root_nfs_parse_addr(char *name); /*__init*/ +extern __be32 root_nfs_parse_addr(char *name); /*__init*/ static inline void nfs_fattr_init(struct nfs_fattr *fattr) { diff --git a/include/net/ipconfig.h b/include/net/ipconfig.h index 2a1fe996fbc6..3924d7d2cb11 100644 --- a/include/net/ipconfig.h +++ b/include/net/ipconfig.h @@ -11,12 +11,12 @@ extern int ic_proto_enabled; /* Protocols enabled (see IC_xxx) */ extern int ic_set_manually; /* IPconfig parameters set manually */ -extern u32 ic_myaddr; /* My IP address */ -extern u32 ic_gateway; /* Gateway IP address */ +extern __be32 ic_myaddr; /* My IP address */ +extern __be32 ic_gateway; /* Gateway IP address */ -extern u32 ic_servaddr; /* Boot server IP address */ +extern __be32 ic_servaddr; /* Boot server IP address */ -extern u32 root_server_addr; /* Address of NFS server */ +extern __be32 root_server_addr; /* Address of NFS server */ extern u8 root_server_path[]; /* Path to mount as root */ diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 955a07abb91d..afa60b9a003f 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -101,6 +101,7 @@ #define CONF_NAMESERVERS_MAX 3 /* Maximum number of nameservers - '3' from resolv.h */ +#define NONE __constant_htonl(INADDR_NONE) /* * Public IP configuration @@ -129,19 +130,19 @@ int ic_proto_enabled __initdata = 0 static int ic_host_name_set __initdata = 0; /* Host name set by us? */ -u32 ic_myaddr = INADDR_NONE; /* My IP address */ -static u32 ic_netmask = INADDR_NONE; /* Netmask for local subnet */ -u32 ic_gateway = INADDR_NONE; /* Gateway IP address */ +__be32 ic_myaddr = NONE; /* My IP address */ +static __be32 ic_netmask = NONE; /* Netmask for local subnet */ +__be32 ic_gateway = NONE; /* Gateway IP address */ -u32 ic_servaddr = INADDR_NONE; /* Boot server IP address */ +__be32 ic_servaddr = NONE; /* Boot server IP address */ -u32 root_server_addr = INADDR_NONE; /* Address of NFS server */ +__be32 root_server_addr = NONE; /* Address of NFS server */ u8 root_server_path[256] = { 0, }; /* Path to mount as root */ /* Persistent data: */ static int ic_proto_used; /* Protocol used, if any */ -static u32 ic_nameservers[CONF_NAMESERVERS_MAX]; /* DNS Server IP addresses */ +static __be32 ic_nameservers[CONF_NAMESERVERS_MAX]; /* DNS Server IP addresses */ static u8 ic_domain[64]; /* DNS (not NIS) domain name */ /* @@ -172,7 +173,7 @@ struct ic_device { struct net_device *dev; unsigned short flags; short able; - u32 xid; + __be32 xid; }; static struct ic_device *ic_first_dev __initdata = NULL;/* List of open device */ @@ -223,7 +224,7 @@ static int __init ic_open_devs(void) d->flags = oflags; d->able = able; if (able & IC_BOOTP) - get_random_bytes(&d->xid, sizeof(u32)); + get_random_bytes(&d->xid, sizeof(__be32)); else d->xid = 0; ic_proto_have_if |= able; @@ -269,7 +270,7 @@ static void __init ic_close_devs(void) */ static inline void -set_sockaddr(struct sockaddr_in *sin, u32 addr, u16 port) +set_sockaddr(struct sockaddr_in *sin, __be32 addr, __be16 port) { sin->sin_family = AF_INET; sin->sin_addr.s_addr = addr; @@ -332,7 +333,7 @@ static int __init ic_setup_routes(void) { /* No need to setup device routes, only the default route... */ - if (ic_gateway != INADDR_NONE) { + if (ic_gateway != NONE) { struct rtentry rm; int err; @@ -368,10 +369,10 @@ static int __init ic_defaults(void) if (!ic_host_name_set) sprintf(init_utsname()->nodename, "%u.%u.%u.%u", NIPQUAD(ic_myaddr)); - if (root_server_addr == INADDR_NONE) + if (root_server_addr == NONE) root_server_addr = ic_servaddr; - if (ic_netmask == INADDR_NONE) { + if (ic_netmask == NONE) { if (IN_CLASSA(ntohl(ic_myaddr))) ic_netmask = htonl(IN_CLASSA_NET); else if (IN_CLASSB(ntohl(ic_myaddr))) @@ -420,7 +421,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt { struct arphdr *rarp; unsigned char *rarp_ptr; - u32 sip, tip; + __be32 sip, tip; unsigned char *sha, *tha; /* s for "source", t for "target" */ struct ic_device *d; @@ -485,12 +486,12 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt goto drop_unlock; /* Discard packets which are not from specified server. */ - if (ic_servaddr != INADDR_NONE && ic_servaddr != sip) + if (ic_servaddr != NONE && ic_servaddr != sip) goto drop_unlock; /* We have a winner! */ ic_dev = dev; - if (ic_myaddr == INADDR_NONE) + if (ic_myaddr == NONE) ic_myaddr = tip; ic_servaddr = sip; ic_got_reply = IC_RARP; @@ -530,13 +531,13 @@ struct bootp_pkt { /* BOOTP packet format */ u8 htype; /* HW address type */ u8 hlen; /* HW address length */ u8 hops; /* Used only by gateways */ - u32 xid; /* Transaction ID */ - u16 secs; /* Seconds since we started */ - u16 flags; /* Just what it says */ - u32 client_ip; /* Client's IP address if known */ - u32 your_ip; /* Assigned IP address */ - u32 server_ip; /* (Next, e.g. NFS) Server's IP address */ - u32 relay_ip; /* IP address of BOOTP relay */ + __be32 xid; /* Transaction ID */ + __be16 secs; /* Seconds since we started */ + __be16 flags; /* Just what it says */ + __be32 client_ip; /* Client's IP address if known */ + __be32 your_ip; /* Assigned IP address */ + __be32 server_ip; /* (Next, e.g. NFS) Server's IP address */ + __be32 relay_ip; /* IP address of BOOTP relay */ u8 hw_addr[16]; /* Client's HW address */ u8 serv_name[64]; /* Server host name */ u8 boot_file[128]; /* Name of boot file */ @@ -576,7 +577,7 @@ static const u8 ic_bootp_cookie[4] = { 99, 130, 83, 99 }; static void __init ic_dhcp_init_options(u8 *options) { - u8 mt = ((ic_servaddr == INADDR_NONE) + u8 mt = ((ic_servaddr == NONE) ? DHCPDISCOVER : DHCPREQUEST); u8 *e = options; @@ -666,7 +667,7 @@ static inline void ic_bootp_init(void) int i; for (i = 0; i < CONF_NAMESERVERS_MAX; i++) - ic_nameservers[i] = INADDR_NONE; + ic_nameservers[i] = NONE; dev_add_pack(&bootp_packet_type); } @@ -708,7 +709,7 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d h->frag_off = htons(IP_DF); h->ttl = 64; h->protocol = IPPROTO_UDP; - h->daddr = INADDR_BROADCAST; + h->daddr = htonl(INADDR_BROADCAST); h->check = ip_fast_csum((unsigned char *) h, h->ihl); /* Construct UDP header */ @@ -730,8 +731,8 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d b->htype = dev->type; /* can cause undefined behavior */ } b->hlen = dev->addr_len; - b->your_ip = INADDR_NONE; - b->server_ip = INADDR_NONE; + b->your_ip = NONE; + b->server_ip = NONE; memcpy(b->hw_addr, dev->dev_addr, dev->addr_len); b->secs = htons(jiffies_diff / HZ); b->xid = d->xid; @@ -788,11 +789,11 @@ static void __init ic_do_bootp_ext(u8 *ext) switch (*ext++) { case 1: /* Subnet mask */ - if (ic_netmask == INADDR_NONE) + if (ic_netmask == NONE) memcpy(&ic_netmask, ext+1, 4); break; case 3: /* Default gateway */ - if (ic_gateway == INADDR_NONE) + if (ic_gateway == NONE) memcpy(&ic_gateway, ext+1, 4); break; case 6: /* DNS server */ @@ -800,7 +801,7 @@ static void __init ic_do_bootp_ext(u8 *ext) if (servers > CONF_NAMESERVERS_MAX) servers = CONF_NAMESERVERS_MAX; for (i = 0; i < servers; i++) { - if (ic_nameservers[i] == INADDR_NONE) + if (ic_nameservers[i] == NONE) memcpy(&ic_nameservers[i], ext+1+4*i, 4); } break; @@ -917,7 +918,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str #ifdef IPCONFIG_DHCP if (ic_proto_enabled & IC_USE_DHCP) { - u32 server_id = INADDR_NONE; + __be32 server_id = NONE; int mt = 0; ext = &b->exten[4]; @@ -949,7 +950,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str /* While in the process of accepting one offer, * ignore all others. */ - if (ic_myaddr != INADDR_NONE) + if (ic_myaddr != NONE) goto drop_unlock; /* Let's accept that offer. */ @@ -965,7 +966,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str * precedence over the bootp header one if * they are different. */ - if ((server_id != INADDR_NONE) && + if ((server_id != NONE) && (b->server_ip != server_id)) b->server_ip = ic_servaddr; break; @@ -979,8 +980,8 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str default: /* Urque. Forget it*/ - ic_myaddr = INADDR_NONE; - ic_servaddr = INADDR_NONE; + ic_myaddr = NONE; + ic_servaddr = NONE; goto drop_unlock; }; @@ -1004,9 +1005,9 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str ic_dev = dev; ic_myaddr = b->your_ip; ic_servaddr = b->server_ip; - if (ic_gateway == INADDR_NONE && b->relay_ip) + if (ic_gateway == NONE && b->relay_ip) ic_gateway = b->relay_ip; - if (ic_nameservers[0] == INADDR_NONE) + if (ic_nameservers[0] == NONE) ic_nameservers[0] = ic_servaddr; ic_got_reply = IC_BOOTP; @@ -1150,7 +1151,7 @@ static int __init ic_dynamic(void) #endif if (!ic_got_reply) { - ic_myaddr = INADDR_NONE; + ic_myaddr = NONE; return -1; } @@ -1182,12 +1183,12 @@ static int pnp_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "domain %s\n", ic_domain); for (i = 0; i < CONF_NAMESERVERS_MAX; i++) { - if (ic_nameservers[i] != INADDR_NONE) + if (ic_nameservers[i] != NONE) seq_printf(seq, "nameserver %u.%u.%u.%u\n", NIPQUAD(ic_nameservers[i])); } - if (ic_servaddr != INADDR_NONE) + if (ic_servaddr != NONE) seq_printf(seq, "bootserver %u.%u.%u.%u\n", NIPQUAD(ic_servaddr)); @@ -1213,9 +1214,9 @@ static struct file_operations pnp_seq_fops = { * need to have root_server_addr set _before_ IPConfig gets called as it * can override it. */ -u32 __init root_nfs_parse_addr(char *name) +__be32 __init root_nfs_parse_addr(char *name) { - u32 addr; + __be32 addr; int octets = 0; char *cp, *cq; @@ -1237,7 +1238,7 @@ u32 __init root_nfs_parse_addr(char *name) addr = in_aton(name); memmove(name, cp, strlen(cp) + 1); } else - addr = INADDR_NONE; + addr = NONE; return addr; } @@ -1248,7 +1249,7 @@ u32 __init root_nfs_parse_addr(char *name) static int __init ip_auto_config(void) { - u32 addr; + __be32 addr; #ifdef CONFIG_PROC_FS proc_net_fops_create("pnp", S_IRUGO, &pnp_seq_fops); @@ -1277,11 +1278,11 @@ static int __init ip_auto_config(void) * interfaces and no default was set), use BOOTP or RARP to get the * missing values. */ - if (ic_myaddr == INADDR_NONE || + if (ic_myaddr == NONE || #ifdef CONFIG_ROOT_NFS (MAJOR(ROOT_DEV) == UNNAMED_MAJOR - && root_server_addr == INADDR_NONE - && ic_servaddr == INADDR_NONE) || + && root_server_addr == NONE + && ic_servaddr == NONE) || #endif ic_first_dev->next) { #ifdef IPCONFIG_DYNAMIC @@ -1334,7 +1335,7 @@ static int __init ip_auto_config(void) } addr = root_nfs_parse_addr(root_server_path); - if (root_server_addr == INADDR_NONE) + if (root_server_addr == NONE) root_server_addr = addr; /* @@ -1461,19 +1462,19 @@ static int __init ip_auto_config_setup(char *addrs) switch (num) { case 0: if ((ic_myaddr = in_aton(ip)) == INADDR_ANY) - ic_myaddr = INADDR_NONE; + ic_myaddr = NONE; break; case 1: if ((ic_servaddr = in_aton(ip)) == INADDR_ANY) - ic_servaddr = INADDR_NONE; + ic_servaddr = NONE; break; case 2: if ((ic_gateway = in_aton(ip)) == INADDR_ANY) - ic_gateway = INADDR_NONE; + ic_gateway = NONE; break; case 3: if ((ic_netmask = in_aton(ip)) == INADDR_ANY) - ic_netmask = INADDR_NONE; + ic_netmask = NONE; break; case 4: if ((dp = strchr(ip, '.'))) { -- cgit v1.2.3 From a27ee7a4dd30feda1954950b5840455a51ae1507 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:21:21 -0800 Subject: [IPV6]: annotate icmpv6 headers Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index c771a7db9871..dc79396aac25 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -7,17 +7,17 @@ struct icmp6hdr { __u8 icmp6_type; __u8 icmp6_code; - __u16 icmp6_cksum; + __be16 icmp6_cksum; union { - __u32 un_data32[1]; - __u16 un_data16[2]; + __be32 un_data32[1]; + __be16 un_data16[2]; __u8 un_data8[4]; struct icmpv6_echo { - __u16 identifier; - __u16 sequence; + __be16 identifier; + __be16 sequence; } u_echo; struct icmpv6_nd_advt { @@ -53,7 +53,7 @@ struct icmp6hdr { #else #error "Please fix " #endif - __u16 rt_lifetime; + __be16 rt_lifetime; } u_nd_ra; } icmp6_dataun; -- cgit v1.2.3 From d5a0a1e3109339090769e40fdaa62482fcf2a717 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:23:14 -0800 Subject: [IPV4]: encapsulation annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/if_tunnel.h | 8 ++++---- include/linux/xfrm.h | 4 ++-- net/ipv4/ah4.c | 4 ++-- net/ipv4/esp4.c | 4 ++-- net/ipv4/ip_gre.c | 52 +++++++++++++++++++++++------------------------ net/ipv4/ipip.c | 16 +++++++-------- 6 files changed, 44 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index bef9f8fd93b3..8de079ba1107 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -19,10 +19,10 @@ struct ip_tunnel_parm { char name[IFNAMSIZ]; int link; - __u16 i_flags; - __u16 o_flags; - __u32 i_key; - __u32 o_key; + __be16 i_flags; + __be16 o_flags; + __be32 i_key; + __be32 o_key; struct iphdr iph; }; diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 8ae7f744917b..7907c42bd4e4 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -211,8 +211,8 @@ struct xfrm_user_tmpl { struct xfrm_encap_tmpl { __u16 encap_type; - __u16 encap_sport; - __u16 encap_dport; + __be16 encap_sport; + __be16 encap_dport; xfrm_address_t encap_oa; }; diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 99542977e47e..67a5509e26fc 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -14,7 +14,7 @@ * into IP header for icv calculation. Options are already checked * for validity, so paranoia is not required. */ -static int ip_clear_mutable_options(struct iphdr *iph, u32 *daddr) +static int ip_clear_mutable_options(struct iphdr *iph, __be32 *daddr) { unsigned char * optptr = (unsigned char*)(iph+1); int l = iph->ihl*4 - sizeof(struct iphdr); @@ -162,7 +162,7 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb) iph->frag_off = 0; iph->check = 0; if (ihl > sizeof(*iph)) { - u32 dummy; + __be32 dummy; if (ip_clear_mutable_options(iph, &dummy)) goto out; } diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index b5c205b57669..f2c6776ea0e6 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -67,7 +67,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) if (x->encap) { struct xfrm_encap_tmpl *encap = x->encap; struct udphdr *uh; - u32 *udpdata32; + __be32 *udpdata32; uh = (struct udphdr *)esph; uh->source = encap->encap_sport; @@ -81,7 +81,7 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) esph = (struct ip_esp_hdr *)(uh + 1); break; case UDP_ENCAP_ESPINUDP_NON_IKE: - udpdata32 = (u32 *)(uh + 1); + udpdata32 = (__be32 *)(uh + 1); udpdata32[0] = udpdata32[1] = 0; esph = (struct ip_esp_hdr *)(udpdata32 + 2); break; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index d5b5dec075b8..25221146debb 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -144,7 +144,7 @@ static struct net_device *ipgre_fb_tunnel_dev; */ #define HASH_SIZE 16 -#define HASH(addr) ((addr^(addr>>4))&0xF) +#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static struct ip_tunnel *tunnels[4][HASH_SIZE]; @@ -157,7 +157,7 @@ static DEFINE_RWLOCK(ipgre_lock); /* Given src, dst and key, find appropriate for input tunnel. */ -static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key) +static struct ip_tunnel * ipgre_tunnel_lookup(__be32 remote, __be32 local, __be32 key) { unsigned h0 = HASH(remote); unsigned h1 = HASH(key); @@ -194,9 +194,9 @@ static struct ip_tunnel * ipgre_tunnel_lookup(u32 remote, u32 local, u32 key) static struct ip_tunnel **ipgre_bucket(struct ip_tunnel *t) { - u32 remote = t->parms.iph.daddr; - u32 local = t->parms.iph.saddr; - u32 key = t->parms.i_key; + __be32 remote = t->parms.iph.daddr; + __be32 local = t->parms.iph.saddr; + __be32 key = t->parms.i_key; unsigned h = HASH(key); int prio = 0; @@ -236,9 +236,9 @@ static void ipgre_tunnel_unlink(struct ip_tunnel *t) static struct ip_tunnel * ipgre_tunnel_locate(struct ip_tunnel_parm *parms, int create) { - u32 remote = parms->iph.daddr; - u32 local = parms->iph.saddr; - u32 key = parms->i_key; + __be32 remote = parms->iph.daddr; + __be32 local = parms->iph.saddr; + __be32 key = parms->i_key; struct ip_tunnel *t, **tp, *nt; struct net_device *dev; unsigned h = HASH(key); @@ -319,12 +319,12 @@ static void ipgre_err(struct sk_buff *skb, u32 info) */ struct iphdr *iph = (struct iphdr*)skb->data; - u16 *p = (u16*)(skb->data+(iph->ihl<<2)); + __be16 *p = (__be16*)(skb->data+(iph->ihl<<2)); int grehlen = (iph->ihl<<2) + 4; int type = skb->h.icmph->type; int code = skb->h.icmph->code; struct ip_tunnel *t; - u16 flags; + __be16 flags; flags = p[0]; if (flags&(GRE_CSUM|GRE_KEY|GRE_SEQ|GRE_ROUTING|GRE_VERSION)) { @@ -370,7 +370,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info) } read_lock(&ipgre_lock); - t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((u32*)p) + (grehlen>>2) - 1) : 0); + t = ipgre_tunnel_lookup(iph->daddr, iph->saddr, (flags&GRE_KEY) ? *(((__be32*)p) + (grehlen>>2) - 1) : 0); if (t == NULL || t->parms.iph.daddr == 0 || MULTICAST(t->parms.iph.daddr)) goto out; @@ -388,14 +388,14 @@ out: #else struct iphdr *iph = (struct iphdr*)dp; struct iphdr *eiph; - u16 *p = (u16*)(dp+(iph->ihl<<2)); + __be16 *p = (__be16*)(dp+(iph->ihl<<2)); int type = skb->h.icmph->type; int code = skb->h.icmph->code; int rel_type = 0; int rel_code = 0; __be32 rel_info = 0; __u32 n = 0; - u16 flags; + __be16 flags; int grehlen = (iph->ihl<<2) + 4; struct sk_buff *skb2; struct flowi fl; @@ -556,9 +556,9 @@ static int ipgre_rcv(struct sk_buff *skb) { struct iphdr *iph; u8 *h; - u16 flags; + __be16 flags; u16 csum = 0; - u32 key = 0; + __be32 key = 0; u32 seqno = 0; struct ip_tunnel *tunnel; int offset = 4; @@ -568,7 +568,7 @@ static int ipgre_rcv(struct sk_buff *skb) iph = skb->nh.iph; h = skb->data; - flags = *(u16*)h; + flags = *(__be16*)h; if (flags&(GRE_CSUM|GRE_KEY|GRE_ROUTING|GRE_SEQ|GRE_VERSION)) { /* - Version must be 0. @@ -592,11 +592,11 @@ static int ipgre_rcv(struct sk_buff *skb) offset += 4; } if (flags&GRE_KEY) { - key = *(u32*)(h + offset); + key = *(__be32*)(h + offset); offset += 4; } if (flags&GRE_SEQ) { - seqno = ntohl(*(u32*)(h + offset)); + seqno = ntohl(*(__be32*)(h + offset)); offset += 4; } } @@ -605,7 +605,7 @@ static int ipgre_rcv(struct sk_buff *skb) if ((tunnel = ipgre_tunnel_lookup(iph->saddr, iph->daddr, key)) != NULL) { secpath_reset(skb); - skb->protocol = *(u16*)(h + 2); + skb->protocol = *(__be16*)(h + 2); /* WCCP version 1 and 2 protocol decoding. * - Change protocol to IP * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header @@ -673,13 +673,13 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct iphdr *old_iph = skb->nh.iph; struct iphdr *tiph; u8 tos; - u16 df; + __be16 df; struct rtable *rt; /* Route to the other host */ struct net_device *tdev; /* Device to other host */ struct iphdr *iph; /* Our new IP header */ int max_headroom; /* The extra header space needed */ int gre_hlen; - u32 dst; + __be32 dst; int mtu; if (tunnel->recursion++) { @@ -860,11 +860,11 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) iph->ttl = dst_metric(&rt->u.dst, RTAX_HOPLIMIT); } - ((u16*)(iph+1))[0] = tunnel->parms.o_flags; - ((u16*)(iph+1))[1] = skb->protocol; + ((__be16*)(iph+1))[0] = tunnel->parms.o_flags; + ((__be16*)(iph+1))[1] = skb->protocol; if (tunnel->parms.o_flags&(GRE_KEY|GRE_CSUM|GRE_SEQ)) { - u32 *ptr = (u32*)(((u8*)iph) + tunnel->hlen - 4); + __be32 *ptr = (__be32*)(((u8*)iph) + tunnel->hlen - 4); if (tunnel->parms.o_flags&GRE_SEQ) { ++tunnel->o_seqno; @@ -877,7 +877,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (tunnel->parms.o_flags&GRE_CSUM) { *ptr = 0; - *(__u16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr)); + *(__be16*)ptr = ip_compute_csum((void*)(iph+1), skb->len - sizeof(struct iphdr)); } } @@ -1068,7 +1068,7 @@ static int ipgre_header(struct sk_buff *skb, struct net_device *dev, unsigned sh { struct ip_tunnel *t = netdev_priv(dev); struct iphdr *iph = (struct iphdr *)skb_push(skb, t->hlen); - u16 *p = (u16*)(iph+1); + __be16 *p = (__be16*)(iph+1); memcpy(iph, &t->parms.iph, sizeof(struct iphdr)); p[0] = t->parms.o_flags; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 0c4556529228..9d719d664e5b 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -118,7 +118,7 @@ #include #define HASH_SIZE 16 -#define HASH(addr) ((addr^(addr>>4))&0xF) +#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF) static int ipip_fb_tunnel_init(struct net_device *dev); static int ipip_tunnel_init(struct net_device *dev); @@ -134,7 +134,7 @@ static struct ip_tunnel **tunnels[4] = { tunnels_wc, tunnels_l, tunnels_r, tunne static DEFINE_RWLOCK(ipip_lock); -static struct ip_tunnel * ipip_tunnel_lookup(u32 remote, u32 local) +static struct ip_tunnel * ipip_tunnel_lookup(__be32 remote, __be32 local) { unsigned h0 = HASH(remote); unsigned h1 = HASH(local); @@ -160,8 +160,8 @@ static struct ip_tunnel * ipip_tunnel_lookup(u32 remote, u32 local) static struct ip_tunnel **ipip_bucket(struct ip_tunnel *t) { - u32 remote = t->parms.iph.daddr; - u32 local = t->parms.iph.saddr; + __be32 remote = t->parms.iph.daddr; + __be32 local = t->parms.iph.saddr; unsigned h = 0; int prio = 0; @@ -203,8 +203,8 @@ static void ipip_tunnel_link(struct ip_tunnel *t) static struct ip_tunnel * ipip_tunnel_locate(struct ip_tunnel_parm *parms, int create) { - u32 remote = parms->iph.daddr; - u32 local = parms->iph.saddr; + __be32 remote = parms->iph.daddr; + __be32 local = parms->iph.saddr; struct ip_tunnel *t, **tp, *nt; struct net_device *dev; unsigned h = 0; @@ -519,13 +519,13 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) struct net_device_stats *stats = &tunnel->stat; struct iphdr *tiph = &tunnel->parms.iph; u8 tos = tunnel->parms.iph.tos; - u16 df = tiph->frag_off; + __be16 df = tiph->frag_off; struct rtable *rt; /* Route to the other host */ struct net_device *tdev; /* Device to other host */ struct iphdr *old_iph = skb->nh.iph; struct iphdr *iph; /* Our new IP header */ int max_headroom; /* The extra header space needed */ - u32 dst = tiph->daddr; + __be32 dst = tiph->daddr; int mtu; if (tunnel->recursion++) { -- cgit v1.2.3 From d29ef86b0a497dd2c4f9601644a15392f7e21f73 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:23:42 -0800 Subject: [AF_KEY]: annotate Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/pfkeyv2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h index d5dd471da225..0f0b880c4280 100644 --- a/include/linux/pfkeyv2.h +++ b/include/linux/pfkeyv2.h @@ -32,7 +32,7 @@ struct sadb_ext { struct sadb_sa { uint16_t sadb_sa_len; uint16_t sadb_sa_exttype; - uint32_t sadb_sa_spi; + __be32 sadb_sa_spi; uint8_t sadb_sa_replay; uint8_t sadb_sa_state; uint8_t sadb_sa_auth; @@ -211,7 +211,7 @@ struct sadb_x_nat_t_type { struct sadb_x_nat_t_port { uint16_t sadb_x_nat_t_port_len; uint16_t sadb_x_nat_t_port_exttype; - uint16_t sadb_x_nat_t_port_port; + __be16 sadb_x_nat_t_port_port; uint16_t sadb_x_nat_t_port_reserved; } __attribute__((packed)); /* sizeof(struct sadb_x_nat_t_port) == 8 */ -- cgit v1.2.3 From 90bcaf7b4a33bb9b100cc06869f0c033a870d4a0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:25:17 -0800 Subject: [IPV6]: flowlabels are net-endian Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/in6.h | 4 ++-- include/linux/ip6_tunnel.h | 2 +- include/linux/ipv6.h | 2 +- include/net/flow.h | 2 +- include/net/ipv6.h | 4 ++-- net/ipv6/af_inet6.c | 2 +- net/ipv6/datagram.c | 10 +++++----- net/ipv6/ip6_flowlabel.c | 8 ++++---- net/ipv6/ip6_output.c | 4 ++-- net/ipv6/ip6_tunnel.c | 6 +++--- net/ipv6/route.c | 2 +- 11 files changed, 23 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/include/linux/in6.h b/include/linux/in6.h index f28621f638e0..4e8350ae8869 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -54,7 +54,7 @@ extern const struct in6_addr in6addr_loopback; struct sockaddr_in6 { unsigned short int sin6_family; /* AF_INET6 */ __be16 sin6_port; /* Transport layer port # */ - __u32 sin6_flowinfo; /* IPv6 flow information */ + __be32 sin6_flowinfo; /* IPv6 flow information */ struct in6_addr sin6_addr; /* IPv6 address */ __u32 sin6_scope_id; /* scope id (new in RFC2553) */ }; @@ -72,7 +72,7 @@ struct ipv6_mreq { struct in6_flowlabel_req { struct in6_addr flr_dst; - __u32 flr_label; + __be32 flr_label; __u8 flr_action; __u8 flr_share; __u16 flr_flags; diff --git a/include/linux/ip6_tunnel.h b/include/linux/ip6_tunnel.h index 5c23aeb104ca..af3f4a70f3df 100644 --- a/include/linux/ip6_tunnel.h +++ b/include/linux/ip6_tunnel.h @@ -25,7 +25,7 @@ struct ip6_tnl_parm { __u8 proto; /* tunnel protocol */ __u8 encap_limit; /* encapsulation limit for tunnel */ __u8 hop_limit; /* hop limit for tunnel */ - __u32 flowinfo; /* traffic class and flowlabel for tunnel */ + __be32 flowinfo; /* traffic class and flowlabel for tunnel */ __u32 flags; /* tunnel flags */ struct in6_addr laddr; /* local tunnel end-point address */ struct in6_addr raddr; /* remote tunnel end-point address */ diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 4f435c59de06..f8241130f5ea 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -274,7 +274,7 @@ struct ipv6_pinfo { struct in6_addr *saddr_cache; #endif - __u32 flow_label; + __be32 flow_label; __u32 frag_size; __s16 hop_limit; __s16 mcast_hops; diff --git a/include/net/flow.h b/include/net/flow.h index 5cda27cd9deb..270d4c1761af 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -27,7 +27,7 @@ struct flowi { struct in6_addr daddr; struct in6_addr saddr; __u32 fwmark; - __u32 flowlabel; + __be32 flowlabel; } ip6_u; struct { diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 4953fac1d070..0b8c9b990ac4 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -191,7 +191,7 @@ struct ipv6_txoptions struct ip6_flowlabel { struct ip6_flowlabel *next; - u32 label; + __be32 label; struct in6_addr dst; struct ipv6_txoptions *opt; atomic_t users; @@ -211,7 +211,7 @@ struct ipv6_fl_socklist struct ip6_flowlabel *fl; }; -extern struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, u32 label); +extern struct ip6_flowlabel *fl6_sock_lookup(struct sock *sk, __be32 label); extern struct ipv6_txoptions *fl6_merge_options(struct ipv6_txoptions * opt_space, struct ip6_flowlabel * fl, struct ipv6_txoptions * fopt); diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 858cae29581c..92bfccf62cb7 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -678,7 +678,7 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb) if (np->rxopt.all) { if ((opt->hop && (np->rxopt.bits.hopopts || np->rxopt.bits.ohopopts)) || - ((IPV6_FLOWINFO_MASK & *(u32*)skb->nh.raw) && + ((IPV6_FLOWINFO_MASK & *(__be32*)skb->nh.raw) && np->rxopt.bits.rxflow) || (opt->srcrt && (np->rxopt.bits.srcrt || np->rxopt.bits.osrcrt)) || diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 7206747022fc..a67434af691f 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -318,7 +318,7 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) ipv6_addr_copy(&sin->sin6_addr, (struct in6_addr *)(skb->nh.raw + serr->addr_offset)); if (np->sndflow) - sin->sin6_flowinfo = *(u32*)(skb->nh.raw + serr->addr_offset - 24) & IPV6_FLOWINFO_MASK; + sin->sin6_flowinfo = *(__be32*)(skb->nh.raw + serr->addr_offset - 24) & IPV6_FLOWINFO_MASK; if (ipv6_addr_type(&sin->sin6_addr) & IPV6_ADDR_LINKLOCAL) sin->sin6_scope_id = IP6CB(skb)->iif; } else { @@ -401,8 +401,8 @@ int datagram_recv_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) put_cmsg(msg, SOL_IPV6, IPV6_TCLASS, sizeof(tclass), &tclass); } - if (np->rxopt.bits.rxflow && (*(u32*)skb->nh.raw & IPV6_FLOWINFO_MASK)) { - u32 flowinfo = *(u32*)skb->nh.raw & IPV6_FLOWINFO_MASK; + if (np->rxopt.bits.rxflow && (*(__be32*)skb->nh.raw & IPV6_FLOWINFO_MASK)) { + __be32 flowinfo = *(__be32*)skb->nh.raw & IPV6_FLOWINFO_MASK; put_cmsg(msg, SOL_IPV6, IPV6_FLOWINFO, sizeof(flowinfo), &flowinfo); } @@ -560,12 +560,12 @@ int datagram_send_ctl(struct msghdr *msg, struct flowi *fl, } if (fl->fl6_flowlabel&IPV6_FLOWINFO_MASK) { - if ((fl->fl6_flowlabel^*(u32 *)CMSG_DATA(cmsg))&~IPV6_FLOWINFO_MASK) { + if ((fl->fl6_flowlabel^*(__be32 *)CMSG_DATA(cmsg))&~IPV6_FLOWINFO_MASK) { err = -EINVAL; goto exit_f; } } - fl->fl6_flowlabel = IPV6_FLOWINFO_MASK & *(u32 *)CMSG_DATA(cmsg); + fl->fl6_flowlabel = IPV6_FLOWINFO_MASK & *(__be32 *)CMSG_DATA(cmsg); break; case IPV6_2292HOPOPTS: diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 6d4533b58dca..624fae251f4e 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -61,7 +61,7 @@ static DEFINE_RWLOCK(ip6_fl_lock); static DEFINE_RWLOCK(ip6_sk_fl_lock); -static __inline__ struct ip6_flowlabel * __fl_lookup(u32 label) +static __inline__ struct ip6_flowlabel * __fl_lookup(__be32 label) { struct ip6_flowlabel *fl; @@ -72,7 +72,7 @@ static __inline__ struct ip6_flowlabel * __fl_lookup(u32 label) return NULL; } -static struct ip6_flowlabel * fl_lookup(u32 label) +static struct ip6_flowlabel * fl_lookup(__be32 label) { struct ip6_flowlabel *fl; @@ -153,7 +153,7 @@ static void ip6_fl_gc(unsigned long dummy) write_unlock(&ip6_fl_lock); } -static int fl_intern(struct ip6_flowlabel *fl, __u32 label) +static int fl_intern(struct ip6_flowlabel *fl, __be32 label) { fl->label = label & IPV6_FLOWLABEL_MASK; @@ -182,7 +182,7 @@ static int fl_intern(struct ip6_flowlabel *fl, __u32 label) /* Socket flowlabel lists */ -struct ip6_flowlabel * fl6_sock_lookup(struct sock *sk, u32 label) +struct ip6_flowlabel * fl6_sock_lookup(struct sock *sk, __be32 label) { struct ipv6_fl_socklist *sfl; struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 66716911962e..9ff1be1bc2e7 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -217,7 +217,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (tclass < 0) tclass = 0; - *(u32 *)hdr = htonl(0x60000000 | (tclass << 20)) | fl->fl6_flowlabel; + *(__be32 *)hdr = htonl(0x60000000 | (tclass << 20)) | fl->fl6_flowlabel; hdr->payload_len = htons(seg_len); hdr->nexthdr = proto; @@ -1311,7 +1311,7 @@ int ip6_push_pending_frames(struct sock *sk) skb->nh.ipv6h = hdr = (struct ipv6hdr*) skb_push(skb, sizeof(struct ipv6hdr)); - *(u32*)hdr = fl->fl6_flowlabel | + *(__be32*)hdr = fl->fl6_flowlabel | htonl(0x60000000 | ((int)np->cork.tclass << 20)); if (skb->len <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 6a3026dd865a..4919f9294e2a 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -665,9 +665,9 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) dsfield = ipv6_get_dsfield(ipv6h); if ((t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_TCLASS_MASK); + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_TCLASS_MASK); if ((t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)) - fl.fl6_flowlabel |= (*(__u32 *) ipv6h & IPV6_FLOWLABEL_MASK); + fl.fl6_flowlabel |= (*(__be32 *) ipv6h & IPV6_FLOWLABEL_MASK); if (encap_limit >= 0 && (opt = create_tel(encap_limit)) == NULL) goto tx_err; @@ -735,7 +735,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) skb->nh.raw = skb_push(skb, sizeof(struct ipv6hdr)); ipv6h = skb->nh.ipv6h; - *(u32*)ipv6h = fl.fl6_flowlabel | htonl(0x60000000); + *(__be32*)ipv6h = fl.fl6_flowlabel | htonl(0x60000000); dsfield = INET_ECN_encapsulate(0, dsfield); ipv6_change_dsfield(ipv6h, ~INET_ECN_MASK, dsfield); ipv6h->payload_len = htons(skb->len - sizeof(struct ipv6hdr)); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b39ae99122d5..e9c1fc5f21b1 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -714,7 +714,7 @@ void ip6_route_input(struct sk_buff *skb) #ifdef CONFIG_IPV6_ROUTE_FWMARK .fwmark = skb->nfmark, #endif - .flowlabel = (* (u32 *) iph)&IPV6_FLOWINFO_MASK, + .flowlabel = (* (__be32 *) iph)&IPV6_FLOWINFO_MASK, }, }, .proto = iph->nexthdr, -- cgit v1.2.3 From 0e11c91e1e912bc4db5b71607d149e7e9a77e756 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:26:29 -0800 Subject: [AF_PACKET]: annotate Weirdness: the third argument of socket() is net-endian here. Oh, well - it's documented in packet(7). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/if_packet.h | 6 ++++-- net/packet/af_packet.c | 18 ++++++++++-------- 2 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h index b92558549d27..99393ef3af39 100644 --- a/include/linux/if_packet.h +++ b/include/linux/if_packet.h @@ -1,17 +1,19 @@ #ifndef __LINUX_IF_PACKET_H #define __LINUX_IF_PACKET_H +#include + struct sockaddr_pkt { unsigned short spkt_family; unsigned char spkt_device[14]; - unsigned short spkt_protocol; + __be16 spkt_protocol; }; struct sockaddr_ll { unsigned short sll_family; - unsigned short sll_protocol; + __be16 sll_protocol; int sll_ifindex; unsigned short sll_hatype; unsigned char sll_pkttype; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index f4ccb90e6739..271d2eed0699 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -201,7 +201,7 @@ struct packet_sock { spinlock_t bind_lock; char running; /* prot_hook is attached*/ int ifindex; /* bound device */ - unsigned short num; + __be16 num; #ifdef CONFIG_PACKET_MULTICAST struct packet_mclist *mclist; #endif @@ -331,7 +331,7 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, struct sockaddr_pkt *saddr=(struct sockaddr_pkt *)msg->msg_name; struct sk_buff *skb; struct net_device *dev; - unsigned short proto=0; + __be16 proto=0; int err; /* @@ -704,7 +704,7 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, struct sockaddr_ll *saddr=(struct sockaddr_ll *)msg->msg_name; struct sk_buff *skb; struct net_device *dev; - unsigned short proto; + __be16 proto; unsigned char *addr; int ifindex, err, reserve = 0; @@ -858,7 +858,7 @@ static int packet_release(struct socket *sock) * Attach a packet hook. */ -static int packet_do_bind(struct sock *sk, struct net_device *dev, int protocol) +static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 protocol) { struct packet_sock *po = pkt_sk(sk); /* @@ -983,6 +983,7 @@ static int packet_create(struct socket *sock, int protocol) { struct sock *sk; struct packet_sock *po; + __be16 proto = (__force __be16)protocol; /* weird, but documented */ int err; if (!capable(CAP_NET_RAW)) @@ -1010,7 +1011,7 @@ static int packet_create(struct socket *sock, int protocol) po = pkt_sk(sk); sk->sk_family = PF_PACKET; - po->num = protocol; + po->num = proto; sk->sk_destruct = packet_sock_destruct; atomic_inc(&packet_socks_nr); @@ -1027,8 +1028,8 @@ static int packet_create(struct socket *sock, int protocol) #endif po->prot_hook.af_packet_priv = sk; - if (protocol) { - po->prot_hook.type = protocol; + if (proto) { + po->prot_hook.type = proto; dev_add_pack(&po->prot_hook); sock_hold(sk); po->running = 1; @@ -1624,7 +1625,8 @@ static int packet_set_ring(struct sock *sk, struct tpacket_req *req, int closing { char **pg_vec = NULL; struct packet_sock *po = pkt_sk(sk); - int was_running, num, order = 0; + int was_running, order = 0; + __be16 num; int err = 0; if (req->tp_block_nr) { -- cgit v1.2.3 From 98a4a86128d7179b22365e16bf880e849e20bc7d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:26:51 -0800 Subject: [NETFILTER]: trivial annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter/nfnetlink.h | 2 +- include/linux/netfilter/nfnetlink_log.h | 10 +++++----- include/linux/netfilter/nfnetlink_queue.h | 18 +++++++++--------- include/linux/netfilter/xt_conntrack.h | 4 ++-- include/linux/netfilter/xt_policy.h | 2 +- include/linux/types.h | 2 ++ net/netfilter/nfnetlink_log.c | 22 ++++++++++------------ net/netfilter/nfnetlink_queue.c | 9 ++++----- net/netfilter/xt_multiport.c | 4 ++-- 9 files changed, 36 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 6d8e3e5a80e9..1e9c821f152d 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -78,7 +78,7 @@ struct nfattr struct nfgenmsg { u_int8_t nfgen_family; /* AF_xxx */ u_int8_t version; /* nfnetlink version */ - u_int16_t res_id; /* resource id */ + __be16 res_id; /* resource id */ }; #define NFNETLINK_V0 0 diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h index 87b92f8b988f..55a2a2b814ed 100644 --- a/include/linux/netfilter/nfnetlink_log.h +++ b/include/linux/netfilter/nfnetlink_log.h @@ -16,20 +16,20 @@ enum nfulnl_msg_types { }; struct nfulnl_msg_packet_hdr { - u_int16_t hw_protocol; /* hw protocol (network order) */ + __be16 hw_protocol; /* hw protocol (network order) */ u_int8_t hook; /* netfilter hook */ u_int8_t _pad; }; struct nfulnl_msg_packet_hw { - u_int16_t hw_addrlen; + __be16 hw_addrlen; u_int16_t _pad; u_int8_t hw_addr[8]; }; struct nfulnl_msg_packet_timestamp { - aligned_u64 sec; - aligned_u64 usec; + aligned_be64 sec; + aligned_be64 usec; }; #define NFULNL_PREFIXLEN 30 /* just like old log target */ @@ -67,7 +67,7 @@ struct nfulnl_msg_config_cmd { } __attribute__ ((packed)); struct nfulnl_msg_config_mode { - u_int32_t copy_range; + __be32 copy_range; u_int8_t copy_mode; u_int8_t _pad; } __attribute__ ((packed)); diff --git a/include/linux/netfilter/nfnetlink_queue.h b/include/linux/netfilter/nfnetlink_queue.h index 36af0360b56d..4beea3d6fda8 100644 --- a/include/linux/netfilter/nfnetlink_queue.h +++ b/include/linux/netfilter/nfnetlink_queue.h @@ -13,20 +13,20 @@ enum nfqnl_msg_types { }; struct nfqnl_msg_packet_hdr { - u_int32_t packet_id; /* unique ID of packet in queue */ - u_int16_t hw_protocol; /* hw protocol (network order) */ + __be32 packet_id; /* unique ID of packet in queue */ + __be16 hw_protocol; /* hw protocol (network order) */ u_int8_t hook; /* netfilter hook */ } __attribute__ ((packed)); struct nfqnl_msg_packet_hw { - u_int16_t hw_addrlen; + __be16 hw_addrlen; u_int16_t _pad; u_int8_t hw_addr[8]; }; struct nfqnl_msg_packet_timestamp { - aligned_u64 sec; - aligned_u64 usec; + aligned_be64 sec; + aligned_be64 usec; }; enum nfqnl_attr_type { @@ -47,8 +47,8 @@ enum nfqnl_attr_type { #define NFQA_MAX (__NFQA_MAX - 1) struct nfqnl_msg_verdict_hdr { - u_int32_t verdict; - u_int32_t id; + __be32 verdict; + __be32 id; }; @@ -63,7 +63,7 @@ enum nfqnl_msg_config_cmds { struct nfqnl_msg_config_cmd { u_int8_t command; /* nfqnl_msg_config_cmds */ u_int8_t _pad; - u_int16_t pf; /* AF_xxx for PF_[UN]BIND */ + __be16 pf; /* AF_xxx for PF_[UN]BIND */ }; enum nfqnl_config_mode { @@ -73,7 +73,7 @@ enum nfqnl_config_mode { }; struct nfqnl_msg_config_params { - u_int32_t copy_range; + __be32 copy_range; u_int8_t copy_mode; /* enum nfqnl_config_mode */ } __attribute__ ((packed)); diff --git a/include/linux/netfilter/xt_conntrack.h b/include/linux/netfilter/xt_conntrack.h index 4c2d9945ca54..70b6f718cf4c 100644 --- a/include/linux/netfilter/xt_conntrack.h +++ b/include/linux/netfilter/xt_conntrack.h @@ -29,14 +29,14 @@ struct ip_conntrack_old_tuple { struct { - __u32 ip; + __be32 ip; union { __u16 all; } u; } src; struct { - __u32 ip; + __be32 ip; union { __u16 all; } u; diff --git a/include/linux/netfilter/xt_policy.h b/include/linux/netfilter/xt_policy.h index a8132ec076fb..45654d359a68 100644 --- a/include/linux/netfilter/xt_policy.h +++ b/include/linux/netfilter/xt_policy.h @@ -39,7 +39,7 @@ struct xt_policy_elem union xt_policy_addr smask; union xt_policy_addr daddr; union xt_policy_addr dmask; - u_int32_t spi; + __be32 spi; u_int32_t reqid; u_int8_t proto; u_int8_t mode; diff --git a/include/linux/types.h b/include/linux/types.h index 750f085fa564..9f11fdd2bd72 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -128,6 +128,8 @@ typedef __s64 int64_t; /* this is a special 64bit data type that is 8-byte aligned */ #define aligned_u64 unsigned long long __attribute__((aligned(8))) +#define aligned_be64 __be64 __attribute__((aligned(8))) +#define aligned_le64 __le64 __attribute__((aligned(8))) /** * The type used for indexing onto a disc or disc partition. diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 1e5207b80fe5..856ed0d19974 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -414,7 +414,7 @@ __build_packet_message(struct nfulnl_instance *inst, struct nfulnl_msg_packet_hdr pmsg; struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; - u_int32_t tmp_uint; + __be32 tmp_uint; UDEBUG("entered\n"); @@ -508,11 +508,9 @@ __build_packet_message(struct nfulnl_instance *inst, if (indev && skb->dev && skb->dev->hard_header_parse) { struct nfulnl_msg_packet_hw phw; - - phw.hw_addrlen = - skb->dev->hard_header_parse((struct sk_buff *)skb, + int len = skb->dev->hard_header_parse((struct sk_buff *)skb, phw.hw_addr); - phw.hw_addrlen = htons(phw.hw_addrlen); + phw.hw_addrlen = htons(len); NFA_PUT(inst->skb, NFULA_HWADDR, sizeof(phw), &phw); } @@ -529,7 +527,7 @@ __build_packet_message(struct nfulnl_instance *inst, if (skb->sk) { read_lock_bh(&skb->sk->sk_callback_lock); if (skb->sk->sk_socket && skb->sk->sk_socket->file) { - u_int32_t uid = htonl(skb->sk->sk_socket->file->f_uid); + __be32 uid = htonl(skb->sk->sk_socket->file->f_uid); /* need to unlock here since NFA_PUT may goto */ read_unlock_bh(&skb->sk->sk_callback_lock); NFA_PUT(inst->skb, NFULA_UID, sizeof(uid), &uid); @@ -882,15 +880,15 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, } if (nfula[NFULA_CFG_TIMEOUT-1]) { - u_int32_t timeout = - *(u_int32_t *)NFA_DATA(nfula[NFULA_CFG_TIMEOUT-1]); + __be32 timeout = + *(__be32 *)NFA_DATA(nfula[NFULA_CFG_TIMEOUT-1]); nfulnl_set_timeout(inst, ntohl(timeout)); } if (nfula[NFULA_CFG_NLBUFSIZ-1]) { - u_int32_t nlbufsiz = - *(u_int32_t *)NFA_DATA(nfula[NFULA_CFG_NLBUFSIZ-1]); + __be32 nlbufsiz = + *(__be32 *)NFA_DATA(nfula[NFULA_CFG_NLBUFSIZ-1]); nfulnl_set_nlbufsiz(inst, ntohl(nlbufsiz)); } @@ -903,8 +901,8 @@ nfulnl_recv_config(struct sock *ctnl, struct sk_buff *skb, } if (nfula[NFULA_CFG_FLAGS-1]) { - u_int16_t flags = - *(u_int16_t *)NFA_DATA(nfula[NFULA_CFG_FLAGS-1]); + __be16 flags = + *(__be16 *)NFA_DATA(nfula[NFULA_CFG_FLAGS-1]); nfulnl_set_flags(inst, ntohs(flags)); } diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index e815a9aa6e95..4ab7b1416bb5 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -349,7 +349,7 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, struct sk_buff *entskb = entry->skb; struct net_device *indev; struct net_device *outdev; - unsigned int tmp_uint; + __be32 tmp_uint; QDEBUG("entered\n"); @@ -489,10 +489,9 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, && entskb->dev->hard_header_parse) { struct nfqnl_msg_packet_hw phw; - phw.hw_addrlen = - entskb->dev->hard_header_parse(entskb, + int len = entskb->dev->hard_header_parse(entskb, phw.hw_addr); - phw.hw_addrlen = htons(phw.hw_addrlen); + phw.hw_addrlen = htons(len); NFA_PUT(skb, NFQA_HWADDR, sizeof(phw), &phw); } @@ -835,7 +834,7 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, } if (nfqa[NFQA_MARK-1]) - entry->skb->nfmark = ntohl(*(u_int32_t *) + entry->skb->nfmark = ntohl(*(__be32 *) NFA_DATA(nfqa[NFQA_MARK-1])); issue_verdict(entry, verdict); diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c index d3aefd380930..b4293058c6ff 100644 --- a/net/netfilter/xt_multiport.c +++ b/net/netfilter/xt_multiport.c @@ -104,7 +104,7 @@ match(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - u16 _ports[2], *pptr; + __be16 _ports[2], *pptr; const struct xt_multiport *multiinfo = matchinfo; if (offset) @@ -135,7 +135,7 @@ match_v1(const struct sk_buff *skb, unsigned int protoff, int *hotdrop) { - u16 _ports[2], *pptr; + __be16 _ports[2], *pptr; const struct xt_multiport_v1 *multiinfo = matchinfo; if (offset) -- cgit v1.2.3 From ae08e1f092210619fe49551aa3ed0dc0003d5880 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Nov 2006 00:27:11 -0800 Subject: [IPV6]: ip6_output annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- net/ipv6/ip6_output.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 85577a4ffa61..7fc9a3aaa1c9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -139,7 +139,7 @@ struct skb_shared_info { /* Warning: this field is not always filled in (UFO)! */ unsigned short gso_segs; unsigned short gso_type; - unsigned int ip6_frag_id; + __be32 ip6_frag_id; struct sk_buff *frag_list; skb_frag_t frags[MAX_SKB_FRAGS]; }; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 9ff1be1bc2e7..93330685adfc 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -267,7 +267,7 @@ int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev, hdr = (struct ipv6hdr *) skb_put(skb, sizeof(struct ipv6hdr)); skb->nh.ipv6h = hdr; - *(u32*)hdr = htonl(0x60000000); + *(__be32*)hdr = htonl(0x60000000); hdr->payload_len = htons(len); hdr->nexthdr = proto; @@ -571,7 +571,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) struct ipv6hdr *tmp_hdr; struct frag_hdr *fh; unsigned int mtu, hlen, left, len; - u32 frag_id = 0; + __be32 frag_id = 0; int ptr, offset = 0, err=0; u8 *prevhdr, nexthdr = 0; -- cgit v1.2.3 From c1a856c9640c9ff3d70bbd8214b6a0974609eef8 Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:03:44 -0600 Subject: SELinux: Various xfrm labeling fixes Since the upstreaming of the mlsxfrm modification a few months back, testing has resulted in the identification of the following issues/bugs that are resolved in this patch set. 1. Fix the security context used in the IKE negotiation to be the context of the socket as opposed to the context of the SPD rule. 2. Fix SO_PEERSEC for tcp sockets to return the security context of the peer as opposed to the source. 3. Fix the selection of an SA for an outgoing packet to be at the same context as the originating socket/flow. The following would be the result of applying this patchset: - SO_PEERSEC will now correctly return the peer's context. - IKE deamons will receive the context of the source socket/flow as opposed to the SPD rule's context so that the negotiated SA will be at the same context as the source socket/flow. - The SELinux policy will require one or more of the following for a socket to be able to communicate with/without SAs: 1. To enable a socket to communicate without using labeled-IPSec SAs: allow socket_t unlabeled_t:association { sendto recvfrom } 2. To enable a socket to communicate with labeled-IPSec SAs: allow socket_t self:association { sendto }; allow socket_t peer_sa_t:association { recvfrom }; This Patch: Pass correct security context to IKE for use in negotiation Fix the security context passed to IKE for use in negotiation to be the context of the socket as opposed to the context of the SPD rule so that the SA carries the label of the originating socket/flow. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 21 ++++++++++----------- security/dummy.c | 4 ++-- security/selinux/include/xfrm.h | 4 ++-- security/selinux/xfrm.c | 35 +++++++++-------------------------- 4 files changed, 23 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/include/linux/security.h b/include/linux/security.h index b200b9856f32..a509329a669b 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -836,10 +836,8 @@ struct request_sock; * used by the XFRM system. * @sec_ctx contains the security context information being provided by * the user-level policy update program (e.g., setkey). - * @sk refers to the sock from which to derive the security context. * Allocate a security structure to the xp->security field; the security - * field is initialized to NULL when the xfrm_policy is allocated. Only - * one of sec_ctx or sock can be specified. + * field is initialized to NULL when the xfrm_policy is allocated. * Return 0 if operation was successful (memory to allocate, legal context) * @xfrm_policy_clone_security: * @old contains an existing xfrm_policy in the SPD. @@ -858,9 +856,6 @@ struct request_sock; * Database by the XFRM system. * @sec_ctx contains the security context information being provided by * the user-level SA generation program (e.g., setkey or racoon). - * @polsec contains the security context information associated with a xfrm - * policy rule from which to take the base context. polsec must be NULL - * when sec_ctx is specified. * @secid contains the secid from which to take the mls portion of the context. * Allocate a security structure to the x->security field; the security * field is initialized to NULL when the xfrm_state is allocated. Set the @@ -1378,12 +1373,12 @@ struct security_operations { #ifdef CONFIG_SECURITY_NETWORK_XFRM int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk); + struct xfrm_user_sec_ctx *sec_ctx); int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new); void (*xfrm_policy_free_security) (struct xfrm_policy *xp); int (*xfrm_policy_delete_security) (struct xfrm_policy *xp); int (*xfrm_state_alloc_security) (struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *polsec, + struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void (*xfrm_state_free_security) (struct xfrm_state *x); int (*xfrm_state_delete_security) (struct xfrm_state *x); @@ -3120,7 +3115,7 @@ static inline void security_inet_csk_clone(struct sock *newsk, #ifdef CONFIG_SECURITY_NETWORK_XFRM static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx) { - return security_ops->xfrm_policy_alloc_security(xp, sec_ctx, NULL); + return security_ops->xfrm_policy_alloc_security(xp, sec_ctx); } static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) @@ -3141,7 +3136,7 @@ static inline int security_xfrm_policy_delete(struct xfrm_policy *xp) static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx) { - return security_ops->xfrm_state_alloc_security(x, sec_ctx, NULL, 0); + return security_ops->xfrm_state_alloc_security(x, sec_ctx, 0); } static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x, @@ -3149,7 +3144,11 @@ static inline int security_xfrm_state_alloc_acquire(struct xfrm_state *x, { if (!polsec) return 0; - return security_ops->xfrm_state_alloc_security(x, NULL, polsec, secid); + /* + * We want the context to be taken from secid which is usually + * from the sock. + */ + return security_ops->xfrm_state_alloc_security(x, NULL, secid); } static inline int security_xfrm_state_delete(struct xfrm_state *x) diff --git a/security/dummy.c b/security/dummy.c index 43874c1e6e23..838d8442cf3c 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -836,7 +836,7 @@ static inline void dummy_req_classify_flow(const struct request_sock *req, #ifdef CONFIG_SECURITY_NETWORK_XFRM static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk) + struct xfrm_user_sec_ctx *sec_ctx) { return 0; } @@ -856,7 +856,7 @@ static int dummy_xfrm_policy_delete_security(struct xfrm_policy *xp) } static int dummy_xfrm_state_alloc_security(struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid) + struct xfrm_user_sec_ctx *sec_ctx, u32 secid) { return 0; } diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 526b28019aca..8e329ddb5e37 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -8,12 +8,12 @@ #define _SELINUX_XFRM_H_ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx, struct sock *sk); + struct xfrm_user_sec_ctx *sec_ctx); int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new); void selinux_xfrm_policy_free(struct xfrm_policy *xp); int selinux_xfrm_policy_delete(struct xfrm_policy *xp); int selinux_xfrm_state_alloc(struct xfrm_state *x, - struct xfrm_user_sec_ctx *sec_ctx, struct xfrm_sec_ctx *pol, u32 secid); + struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void selinux_xfrm_state_free(struct xfrm_state *x); int selinux_xfrm_state_delete(struct xfrm_state *x); int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 675b995a67c3..4d5a043cdfa1 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -226,16 +226,15 @@ int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall) * CTX does not have a meaningful value on input */ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, - struct xfrm_user_sec_ctx *uctx, struct xfrm_sec_ctx *pol, u32 sid) + struct xfrm_user_sec_ctx *uctx, u32 sid) { int rc = 0; struct task_security_struct *tsec = current->security; struct xfrm_sec_ctx *ctx = NULL; char *ctx_str = NULL; u32 str_len; - u32 ctx_sid; - BUG_ON(uctx && pol); + BUG_ON(uctx && sid); if (!uctx) goto not_from_user; @@ -279,15 +278,7 @@ static int selinux_xfrm_sec_ctx_alloc(struct xfrm_sec_ctx **ctxp, return rc; not_from_user: - if (pol) { - rc = security_sid_mls_copy(pol->ctx_sid, sid, &ctx_sid); - if (rc) - goto out; - } - else - ctx_sid = sid; - - rc = security_sid_to_context(ctx_sid, &ctx_str, &str_len); + rc = security_sid_to_context(sid, &ctx_str, &str_len); if (rc) goto out; @@ -302,7 +293,7 @@ not_from_user: ctx->ctx_doi = XFRM_SC_DOI_LSM; ctx->ctx_alg = XFRM_SC_ALG_SELINUX; - ctx->ctx_sid = ctx_sid; + ctx->ctx_sid = sid; ctx->ctx_len = str_len; memcpy(ctx->ctx_str, ctx_str, @@ -323,22 +314,14 @@ out2: * xfrm_policy. */ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *uctx, struct sock *sk) + struct xfrm_user_sec_ctx *uctx) { int err; - u32 sid; BUG_ON(!xp); - BUG_ON(uctx && sk); - - if (sk) { - struct sk_security_struct *ssec = sk->sk_security; - sid = ssec->sid; - } - else - sid = SECSID_NULL; + BUG_ON(!uctx); - err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, NULL, sid); + err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, 0); return err; } @@ -399,13 +382,13 @@ int selinux_xfrm_policy_delete(struct xfrm_policy *xp) * xfrm_state. */ int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *uctx, - struct xfrm_sec_ctx *pol, u32 secid) + u32 secid) { int err; BUG_ON(!x); - err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, pol, secid); + err = selinux_xfrm_sec_ctx_alloc(&x->security, uctx, secid); return err; } -- cgit v1.2.3 From 6b877699c6f1efede4545bcecc367786a472eedb Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:04:09 -0600 Subject: SELinux: Return correct context for SO_PEERSEC Fix SO_PEERSEC for tcp sockets to return the security context of the peer (as represented by the SA from the peer) as opposed to the SA used by the local/source socket. Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 16 +++++++++++++++- include/net/request_sock.h | 1 + net/ipv4/tcp_input.c | 2 ++ security/dummy.c | 6 ++++++ security/selinux/hooks.c | 21 +++++++++++++++++---- security/selinux/include/xfrm.h | 12 ++++++------ security/selinux/xfrm.c | 40 +++------------------------------------- 7 files changed, 50 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/include/linux/security.h b/include/linux/security.h index a509329a669b..84cebcdb3f83 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -826,6 +826,8 @@ struct request_sock; * Sets the openreq's sid to socket's sid with MLS portion taken from peer sid. * @inet_csk_clone: * Sets the new child socket's sid to the openreq sid. + * @inet_conn_established: + * Sets the connection's peersid to the secmark on skb. * @req_classify_flow: * Sets the flow's sid to the openreq sid. * @@ -1368,6 +1370,7 @@ struct security_operations { int (*inet_conn_request)(struct sock *sk, struct sk_buff *skb, struct request_sock *req); void (*inet_csk_clone)(struct sock *newsk, const struct request_sock *req); + void (*inet_conn_established)(struct sock *sk, struct sk_buff *skb); void (*req_classify_flow)(const struct request_sock *req, struct flowi *fl); #endif /* CONFIG_SECURITY_NETWORK */ @@ -2961,9 +2964,15 @@ static inline void security_inet_csk_clone(struct sock *newsk, { security_ops->inet_csk_clone(newsk, req); } + +static inline void security_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ + security_ops->inet_conn_established(sk, skb); +} #else /* CONFIG_SECURITY_NETWORK */ static inline int security_unix_stream_connect(struct socket * sock, - struct socket * other, + struct socket * other, struct sock * newsk) { return 0; @@ -3110,6 +3119,11 @@ static inline void security_inet_csk_clone(struct sock *newsk, const struct request_sock *req) { } + +static inline void security_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ +} #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 8e165ca16bd8..f743a941a4f2 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -54,6 +54,7 @@ struct request_sock { struct request_sock_ops *rsk_ops; struct sock *sk; u32 secid; + u32 peer_secid; }; static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index cf06accbe687..4a8c96cdec7d 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4230,6 +4230,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, mb(); tcp_set_state(sk, TCP_ESTABLISHED); + security_inet_conn_established(sk, skb); + /* Make sure socket is routed, for correct metrics. */ icsk->icsk_af_ops->rebuild_header(sk); diff --git a/security/dummy.c b/security/dummy.c index 838d8442cf3c..0148d1518dd1 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -828,6 +828,11 @@ static inline void dummy_inet_csk_clone(struct sock *newsk, { } +static inline void dummy_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ +} + static inline void dummy_req_classify_flow(const struct request_sock *req, struct flowi *fl) { @@ -1108,6 +1113,7 @@ void security_fixup_ops (struct security_operations *ops) set_to_dummy_if_null(ops, sock_graft); set_to_dummy_if_null(ops, inet_conn_request); set_to_dummy_if_null(ops, inet_csk_clone); + set_to_dummy_if_null(ops, inet_conn_established); set_to_dummy_if_null(ops, req_classify_flow); #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 28ee187ed224..5bbd599a4471 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3535,8 +3535,10 @@ static int selinux_socket_getpeersec_stream(struct socket *sock, char __user *op } else if (isec->sclass == SECCLASS_TCP_SOCKET) { peer_sid = selinux_netlbl_socket_getpeersec_stream(sock); - if (peer_sid == SECSID_NULL) - peer_sid = selinux_socket_getpeer_stream(sock->sk); + if (peer_sid == SECSID_NULL) { + ssec = sock->sk->sk_security; + peer_sid = ssec->peer_sid; + } if (peer_sid == SECSID_NULL) { err = -ENOPROTOOPT; goto out; @@ -3647,11 +3649,11 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb, return 0; } - err = selinux_xfrm_decode_session(skb, &peersid, 0); - BUG_ON(err); + selinux_skb_xfrm_sid(skb, &peersid); if (peersid == SECSID_NULL) { req->secid = sksec->sid; + req->peer_secid = 0; return 0; } @@ -3660,6 +3662,7 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb, return err; req->secid = newsid; + req->peer_secid = peersid; return 0; } @@ -3669,6 +3672,7 @@ static void selinux_inet_csk_clone(struct sock *newsk, struct sk_security_struct *newsksec = newsk->sk_security; newsksec->sid = req->secid; + newsksec->peer_sid = req->peer_secid; /* NOTE: Ideally, we should also get the isec->sid for the new socket in sync, but we don't have the isec available yet. So we will wait until sock_graft to do it, by which @@ -3677,6 +3681,14 @@ static void selinux_inet_csk_clone(struct sock *newsk, selinux_netlbl_sk_security_init(newsksec, req->rsk_ops->family); } +static void selinux_inet_conn_established(struct sock *sk, + struct sk_buff *skb) +{ + struct sk_security_struct *sksec = sk->sk_security; + + selinux_skb_xfrm_sid(skb, &sksec->peer_sid); +} + static void selinux_req_classify_flow(const struct request_sock *req, struct flowi *fl) { @@ -4739,6 +4751,7 @@ static struct security_operations selinux_ops = { .sock_graft = selinux_sock_graft, .inet_conn_request = selinux_inet_conn_request, .inet_csk_clone = selinux_inet_csk_clone, + .inet_conn_established = selinux_inet_conn_established, .req_classify_flow = selinux_req_classify_flow, #ifdef CONFIG_SECURITY_NETWORK_XFRM diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 8e329ddb5e37..27502365d706 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -39,7 +39,6 @@ int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb, struct avc_audit_data *ad); int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, struct avc_audit_data *ad); -u32 selinux_socket_getpeer_stream(struct sock *sk); u32 selinux_socket_getpeer_dgram(struct sk_buff *skb); int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall); #else @@ -55,11 +54,6 @@ static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, return 0; } -static inline int selinux_socket_getpeer_stream(struct sock *sk) -{ - return SECSID_NULL; -} - static inline int selinux_socket_getpeer_dgram(struct sk_buff *skb) { return SECSID_NULL; @@ -71,4 +65,10 @@ static inline int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int } #endif +static inline void selinux_skb_xfrm_sid(struct sk_buff *skb, u32 *sid) +{ + int err = selinux_xfrm_decode_session(skb, sid, 0); + BUG_ON(err); +} + #endif /* _SELINUX_XFRM_H_ */ diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 4d5a043cdfa1..8fef74271f22 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -184,7 +184,8 @@ int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, } /* - * LSM hook implementation that determines the sid for the session. + * LSM hook implementation that checks and/or returns the xfrm sid for the + * incoming packet. */ int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall) @@ -402,44 +403,9 @@ void selinux_xfrm_state_free(struct xfrm_state *x) kfree(ctx); } -/* - * SELinux internal function to retrieve the context of a connected - * (sk->sk_state == TCP_ESTABLISHED) TCP socket based on its security - * association used to connect to the remote socket. - * - * Retrieve via getsockopt SO_PEERSEC. - */ -u32 selinux_socket_getpeer_stream(struct sock *sk) -{ - struct dst_entry *dst, *dst_test; - u32 peer_sid = SECSID_NULL; - - if (sk->sk_state != TCP_ESTABLISHED) - goto out; - - dst = sk_dst_get(sk); - if (!dst) - goto out; - - for (dst_test = dst; dst_test != 0; - dst_test = dst_test->child) { - struct xfrm_state *x = dst_test->xfrm; - - if (x && selinux_authorizable_xfrm(x)) { - struct xfrm_sec_ctx *ctx = x->security; - peer_sid = ctx->ctx_sid; - break; - } - } - dst_release(dst); - -out: - return peer_sid; -} - /* * SELinux internal function to retrieve the context of a UDP packet - * based on its security association used to connect to the remote socket. + * based on its security association. * * Retrieve via setsockopt IP_PASSSEC and recvmsg with control message * type SCM_SECURITY. -- cgit v1.2.3 From 67f83cbf081a70426ff667e8d14f94e13ed3bdca Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Wed, 8 Nov 2006 17:04:26 -0600 Subject: SELinux: Fix SA selection semantics Fix the selection of an SA for an outgoing packet to be at the same context as the originating socket/flow. This eliminates the SELinux policy's ability to use/sendto SAs with contexts other than the socket's. With this patch applied, the SELinux policy will require one or more of the following for a socket to be able to communicate with/without SAs: 1. To enable a socket to communicate without using labeled-IPSec SAs: allow socket_t unlabeled_t:association { sendto recvfrom } 2. To enable a socket to communicate with labeled-IPSec SAs: allow socket_t self:association { sendto }; allow socket_t peer_sa_t:association { recvfrom }; Signed-off-by: Venkat Yekkirala Signed-off-by: James Morris --- include/linux/security.h | 19 -------- net/xfrm/xfrm_policy.c | 3 +- security/dummy.c | 7 --- security/selinux/hooks.c | 26 +++++++---- security/selinux/include/xfrm.h | 7 +-- security/selinux/xfrm.c | 101 +++++++++++++++++++--------------------- 6 files changed, 70 insertions(+), 93 deletions(-) (limited to 'include/linux') diff --git a/include/linux/security.h b/include/linux/security.h index 84cebcdb3f83..83cdefae9931 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -886,11 +886,6 @@ struct request_sock; * @xp contains the policy to check for a match. * @fl contains the flow to check for a match. * Return 1 if there is a match. - * @xfrm_flow_state_match: - * @fl contains the flow key to match. - * @xfrm points to the xfrm_state to match. - * @xp points to the xfrm_policy to match. - * Return 1 if there is a match. * @xfrm_decode_session: * @skb points to skb to decode. * @secid points to the flow key secid to set. @@ -1388,8 +1383,6 @@ struct security_operations { int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int (*xfrm_state_pol_flow_match)(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); - int (*xfrm_flow_state_match)(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp); int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall); #endif /* CONFIG_SECURITY_NETWORK_XFRM */ @@ -3186,12 +3179,6 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, return security_ops->xfrm_state_pol_flow_match(x, xp, fl); } -static inline int security_xfrm_flow_state_match(struct flowi *fl, - struct xfrm_state *xfrm, struct xfrm_policy *xp) -{ - return security_ops->xfrm_flow_state_match(fl, xfrm, xp); -} - static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) { return security_ops->xfrm_decode_session(skb, secid, 1); @@ -3255,12 +3242,6 @@ static inline int security_xfrm_state_pol_flow_match(struct xfrm_state *x, return 1; } -static inline int security_xfrm_flow_state_match(struct flowi *fl, - struct xfrm_state *xfrm, struct xfrm_policy *xp) -{ - return 1; -} - static inline int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid) { return 0; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 7736b23c3f03..b88b038530c9 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -1894,7 +1894,8 @@ int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, if (fl && !xfrm_selector_match(&dst->xfrm->sel, fl, family)) return 0; - if (fl && !security_xfrm_flow_state_match(fl, dst->xfrm, pol)) + if (fl && pol && + !security_xfrm_state_pol_flow_match(dst->xfrm, pol, fl)) return 0; if (dst->xfrm->km.state != XFRM_STATE_VALID) return 0; diff --git a/security/dummy.c b/security/dummy.c index 0148d1518dd1..558795b237d6 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -886,12 +886,6 @@ static int dummy_xfrm_state_pol_flow_match(struct xfrm_state *x, return 1; } -static int dummy_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp) -{ - return 1; -} - static int dummy_xfrm_decode_session(struct sk_buff *skb, u32 *fl, int ckall) { return 0; @@ -1126,7 +1120,6 @@ void security_fixup_ops (struct security_operations *ops) set_to_dummy_if_null(ops, xfrm_state_delete_security); set_to_dummy_if_null(ops, xfrm_policy_lookup); set_to_dummy_if_null(ops, xfrm_state_pol_flow_match); - set_to_dummy_if_null(ops, xfrm_flow_state_match); set_to_dummy_if_null(ops, xfrm_decode_session); #endif /* CONFIG_SECURITY_NETWORK_XFRM */ #ifdef CONFIG_KEYS diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 5bbd599a4471..956137baf3e7 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2889,7 +2889,8 @@ static void selinux_task_to_inode(struct task_struct *p, } /* Returns error only if unable to parse addresses */ -static int selinux_parse_skb_ipv4(struct sk_buff *skb, struct avc_audit_data *ad) +static int selinux_parse_skb_ipv4(struct sk_buff *skb, + struct avc_audit_data *ad, u8 *proto) { int offset, ihlen, ret = -EINVAL; struct iphdr _iph, *ih; @@ -2907,6 +2908,9 @@ static int selinux_parse_skb_ipv4(struct sk_buff *skb, struct avc_audit_data *ad ad->u.net.v4info.daddr = ih->daddr; ret = 0; + if (proto) + *proto = ih->protocol; + switch (ih->protocol) { case IPPROTO_TCP: { struct tcphdr _tcph, *th; @@ -2950,7 +2954,8 @@ out: #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) /* Returns error only if unable to parse addresses */ -static int selinux_parse_skb_ipv6(struct sk_buff *skb, struct avc_audit_data *ad) +static int selinux_parse_skb_ipv6(struct sk_buff *skb, + struct avc_audit_data *ad, u8 *proto) { u8 nexthdr; int ret = -EINVAL, offset; @@ -2971,6 +2976,9 @@ static int selinux_parse_skb_ipv6(struct sk_buff *skb, struct avc_audit_data *ad if (offset < 0) goto out; + if (proto) + *proto = nexthdr; + switch (nexthdr) { case IPPROTO_TCP: { struct tcphdr _tcph, *th; @@ -3007,13 +3015,13 @@ out: #endif /* IPV6 */ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad, - char **addrp, int *len, int src) + char **addrp, int *len, int src, u8 *proto) { int ret = 0; switch (ad->u.net.family) { case PF_INET: - ret = selinux_parse_skb_ipv4(skb, ad); + ret = selinux_parse_skb_ipv4(skb, ad, proto); if (ret || !addrp) break; *len = 4; @@ -3023,7 +3031,7 @@ static int selinux_parse_skb(struct sk_buff *skb, struct avc_audit_data *ad, #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: - ret = selinux_parse_skb_ipv6(skb, ad); + ret = selinux_parse_skb_ipv6(skb, ad, proto); if (ret || !addrp) break; *len = 16; @@ -3494,7 +3502,7 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb) ad.u.net.netif = skb->dev ? skb->dev->name : "[unknown]"; ad.u.net.family = family; - err = selinux_parse_skb(skb, &ad, &addrp, &len, 1); + err = selinux_parse_skb(skb, &ad, &addrp, &len, 1, NULL); if (err) goto out; @@ -3820,6 +3828,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, struct avc_audit_data ad; struct net_device *dev = (struct net_device *)out; struct sk_security_struct *sksec; + u8 proto; sk = skb->sk; if (!sk) @@ -3831,7 +3840,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, ad.u.net.netif = dev->name; ad.u.net.family = family; - err = selinux_parse_skb(skb, &ad, &addrp, &len, 0); + err = selinux_parse_skb(skb, &ad, &addrp, &len, 0, &proto); if (err) goto out; @@ -3845,7 +3854,7 @@ static unsigned int selinux_ip_postroute_last(unsigned int hooknum, if (err) goto out; - err = selinux_xfrm_postroute_last(sksec->sid, skb, &ad); + err = selinux_xfrm_postroute_last(sksec->sid, skb, &ad, proto); out: return err ? NF_DROP : NF_ACCEPT; } @@ -4764,7 +4773,6 @@ static struct security_operations selinux_ops = { .xfrm_state_delete_security = selinux_xfrm_state_delete, .xfrm_policy_lookup = selinux_xfrm_policy_lookup, .xfrm_state_pol_flow_match = selinux_xfrm_state_pol_flow_match, - .xfrm_flow_state_match = selinux_xfrm_flow_state_match, .xfrm_decode_session = selinux_xfrm_decode_session, #endif diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 27502365d706..ebd7246a4be5 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -19,9 +19,6 @@ int selinux_xfrm_state_delete(struct xfrm_state *x); int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp); - /* * Extract the security blob from the sock (it's actually on the socket) @@ -38,7 +35,7 @@ static inline struct inode_security_struct *get_sock_isec(struct sock *sk) int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb, struct avc_audit_data *ad); int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad); + struct avc_audit_data *ad, u8 proto); u32 selinux_socket_getpeer_dgram(struct sk_buff *skb); int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall); #else @@ -49,7 +46,7 @@ static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, } static inline int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad) + struct avc_audit_data *ad, u8 proto) { return 0; } diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 8fef74271f22..9b777140068f 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -115,71 +115,40 @@ int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy * struct flowi *fl) { u32 state_sid; - u32 pol_sid; - int err; + int rc; - if (xp->security) { - if (!x->security) - /* unlabeled SA and labeled policy can't match */ - return 0; - else - state_sid = x->security->ctx_sid; - pol_sid = xp->security->ctx_sid; - } else + if (!xp->security) if (x->security) /* unlabeled policy and labeled SA can't match */ return 0; else /* unlabeled policy and unlabeled SA match all flows */ return 1; - - err = avc_has_perm(state_sid, pol_sid, SECCLASS_ASSOCIATION, - ASSOCIATION__POLMATCH, - NULL); - - if (err) - return 0; - - err = avc_has_perm(fl->secid, state_sid, SECCLASS_ASSOCIATION, - ASSOCIATION__SENDTO, - NULL)? 0:1; - - return err; -} - -/* - * LSM hook implementation that authorizes that a particular outgoing flow - * can use a given security association. - */ - -int selinux_xfrm_flow_state_match(struct flowi *fl, struct xfrm_state *xfrm, - struct xfrm_policy *xp) -{ - int rc = 0; - u32 sel_sid = SECINITSID_UNLABELED; - struct xfrm_sec_ctx *ctx; - - if (!xp->security) - if (!xfrm->security) - return 1; - else - return 0; else - if (!xfrm->security) + if (!x->security) + /* unlabeled SA and labeled policy can't match */ return 0; + else + if (!selinux_authorizable_xfrm(x)) + /* Not a SELinux-labeled SA */ + return 0; - /* Context sid is either set to label or ANY_ASSOC */ - if ((ctx = xfrm->security)) { - if (!selinux_authorizable_ctx(ctx)) - return 0; + state_sid = x->security->ctx_sid; - sel_sid = ctx->ctx_sid; - } + if (fl->secid != state_sid) + return 0; - rc = avc_has_perm(fl->secid, sel_sid, SECCLASS_ASSOCIATION, + rc = avc_has_perm(fl->secid, state_sid, SECCLASS_ASSOCIATION, ASSOCIATION__SENDTO, NULL)? 0:1; + /* + * We don't need a separate SA Vs. policy polmatch check + * since the SA is now of the same label as the flow and + * a flow Vs. policy polmatch check had already happened + * in selinux_xfrm_policy_lookup() above. + */ + return rc; } @@ -481,6 +450,13 @@ int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, } } + /* + * This check even when there's no association involved is + * intended, according to Trent Jaeger, to make sure a + * process can't engage in non-ipsec communication unless + * explicitly allowed by policy. + */ + rc = avc_has_perm(isec_sid, sel_sid, SECCLASS_ASSOCIATION, ASSOCIATION__RECVFROM, ad); @@ -492,10 +468,10 @@ int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, * If we have no security association, then we need to determine * whether the socket is allowed to send to an unlabelled destination. * If we do have a authorizable security association, then it has already been - * checked in xfrm_policy_lookup hook. + * checked in the selinux_xfrm_state_pol_flow_match hook above. */ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, - struct avc_audit_data *ad) + struct avc_audit_data *ad, u8 proto) { struct dst_entry *dst; int rc = 0; @@ -514,6 +490,27 @@ int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, } } + switch (proto) { + case IPPROTO_AH: + case IPPROTO_ESP: + case IPPROTO_COMP: + /* + * We should have already seen this packet once before + * it underwent xfrm(s). No need to subject it to the + * unlabeled check. + */ + goto out; + default: + break; + } + + /* + * This check even when there's no association involved is + * intended, according to Trent Jaeger, to make sure a + * process can't engage in non-ipsec communication unless + * explicitly allowed by policy. + */ + rc = avc_has_perm(isec_sid, SECINITSID_UNLABELED, SECCLASS_ASSOCIATION, ASSOCIATION__SENDTO, ad); out: -- cgit v1.2.3 From 82e91ffef60e6eba9848fe149ce1eecd2b5aef12 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:19:14 -0800 Subject: [NET]: Turn nfmark into generic mark nfmark is being used in various subsystems and has become the defacto mark field for all kinds of packets. Therefore it makes sense to rename it to `mark' and remove the dependency on CONFIG_NETFILTER. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/bridge/netfilter/ebt_mark.c | 8 ++++---- net/bridge/netfilter/ebt_mark_m.c | 4 ++-- net/bridge/netfilter/ebt_ulog.c | 2 +- net/core/skbuff.c | 4 ++-- net/decnet/dn_route.c | 4 ++-- net/ipv4/ip_output.c | 2 +- net/ipv4/ipvs/ip_vs_proto_tcp.c | 2 +- net/ipv4/ipvs/ip_vs_proto_udp.c | 2 +- net/ipv4/netfilter.c | 2 +- net/ipv4/netfilter/ip_queue.c | 2 +- net/ipv4/netfilter/ipt_REJECT.c | 2 +- net/ipv4/netfilter/ipt_ULOG.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 6 +++--- net/ipv4/route.c | 10 +++++----- net/ipv6/ip6_output.c | 2 +- net/ipv6/netfilter/ip6_queue.c | 2 +- net/ipv6/netfilter/ip6table_mangle.c | 9 ++++----- net/ipv6/route.c | 2 +- net/netfilter/nfnetlink_log.c | 4 ++-- net/netfilter/nfnetlink_queue.c | 8 ++++---- net/netfilter/xt_CONNMARK.c | 10 +++++----- net/netfilter/xt_MARK.c | 12 ++++++------ net/netfilter/xt_mark.c | 2 +- net/sched/Kconfig | 2 +- net/sched/cls_fw.c | 6 +----- net/sched/cls_u32.c | 2 +- net/sched/em_meta.c | 10 +++------- 28 files changed, 59 insertions(+), 68 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7fc9a3aaa1c9..e3ae544b3956 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -216,7 +216,7 @@ enum { * @tail: Tail pointer * @end: End pointer * @destructor: Destruct function - * @nfmark: Can be used for communication between hooks + * @mark: Generic packet mark * @nfct: Associated connection, if any * @ipvs_property: skbuff is owned by ipvs * @nfctinfo: Relationship of this skb to the connection @@ -295,7 +295,6 @@ struct sk_buff { #ifdef CONFIG_BRIDGE_NETFILTER struct nf_bridge_info *nf_bridge; #endif - __u32 nfmark; #endif /* CONFIG_NETFILTER */ #ifdef CONFIG_NET_SCHED __u16 tc_index; /* traffic control index */ @@ -310,6 +309,7 @@ struct sk_buff { __u32 secmark; #endif + __u32 mark; /* These elements must be at the end, see alloc_skb() for details. */ unsigned int truesize; diff --git a/net/bridge/netfilter/ebt_mark.c b/net/bridge/netfilter/ebt_mark.c index b54306a934e5..2458638561cb 100644 --- a/net/bridge/netfilter/ebt_mark.c +++ b/net/bridge/netfilter/ebt_mark.c @@ -25,13 +25,13 @@ static int ebt_target_mark(struct sk_buff **pskb, unsigned int hooknr, int action = info->target & -16; if (action == MARK_SET_VALUE) - (*pskb)->nfmark = info->mark; + (*pskb)->mark = info->mark; else if (action == MARK_OR_VALUE) - (*pskb)->nfmark |= info->mark; + (*pskb)->mark |= info->mark; else if (action == MARK_AND_VALUE) - (*pskb)->nfmark &= info->mark; + (*pskb)->mark &= info->mark; else - (*pskb)->nfmark ^= info->mark; + (*pskb)->mark ^= info->mark; return info->target | -16; } diff --git a/net/bridge/netfilter/ebt_mark_m.c b/net/bridge/netfilter/ebt_mark_m.c index a6413e4b4982..025869ee0b68 100644 --- a/net/bridge/netfilter/ebt_mark_m.c +++ b/net/bridge/netfilter/ebt_mark_m.c @@ -19,8 +19,8 @@ static int ebt_filter_mark(const struct sk_buff *skb, struct ebt_mark_m_info *info = (struct ebt_mark_m_info *) data; if (info->bitmask & EBT_MARK_OR) - return !(!!(skb->nfmark & info->mask) ^ info->invert); - return !(((skb->nfmark & info->mask) == info->mark) ^ info->invert); + return !(!!(skb->mark & info->mask) ^ info->invert); + return !(((skb->mark & info->mask) == info->mark) ^ info->invert); } static int ebt_mark_check(const char *tablename, unsigned int hookmask, diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c index 9f950db3b76f..c1af68b5a29c 100644 --- a/net/bridge/netfilter/ebt_ulog.c +++ b/net/bridge/netfilter/ebt_ulog.c @@ -168,7 +168,7 @@ static void ebt_ulog_packet(unsigned int hooknr, const struct sk_buff *skb, if (ub->qlen == 1) skb_set_timestamp(ub->skb, &pm->stamp); pm->data_len = copy_len; - pm->mark = skb->nfmark; + pm->mark = skb->mark; pm->hook = hooknr; if (uloginfo->prefix != NULL) strcpy(pm->prefix, uloginfo->prefix); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b8b106358040..b3dea1ef9535 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -473,8 +473,8 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) #endif C(protocol); n->destructor = NULL; + C(mark); #ifdef CONFIG_NETFILTER - C(nfmark); C(nfct); nf_conntrack_get(skb->nfct); C(nfctinfo); @@ -534,8 +534,8 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->pkt_type = old->pkt_type; new->tstamp = old->tstamp; new->destructor = NULL; + new->mark = old->mark; #ifdef CONFIG_NETFILTER - new->nfmark = old->nfmark; new->nfct = old->nfct; nf_conntrack_get(old->nfct); new->nfctinfo = old->nfctinfo; diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 23489f7232d2..3482839af280 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1236,7 +1236,7 @@ static int dn_route_input_slow(struct sk_buff *skb) .saddr = cb->src, .scope = RT_SCOPE_UNIVERSE, #ifdef CONFIG_DECNET_ROUTE_FWMARK - .fwmark = skb->nfmark + .fwmark = skb->mark #endif } }, .iif = skb->dev->ifindex }; @@ -1458,7 +1458,7 @@ int dn_route_input(struct sk_buff *skb) (rt->fl.fld_dst == cb->dst) && (rt->fl.oif == 0) && #ifdef CONFIG_DECNET_ROUTE_FWMARK - (rt->fl.fld_fwmark == skb->nfmark) && + (rt->fl.fld_fwmark == skb->mark) && #endif (rt->fl.iif == cb->iif)) { rt->u.dst.lastuse = jiffies; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index fc195a44fc2e..23633bf042ba 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -386,6 +386,7 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) dst_release(to->dst); to->dst = dst_clone(from->dst); to->dev = from->dev; + to->mark = from->mark; /* Copy the flags to each fragment. */ IPCB(to)->flags = IPCB(from)->flags; @@ -394,7 +395,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) to->tc_index = from->tc_index; #endif #ifdef CONFIG_NETFILTER - to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ nf_conntrack_put(to->nfct); to->nfct = from->nfct; diff --git a/net/ipv4/ipvs/ip_vs_proto_tcp.c b/net/ipv4/ipvs/ip_vs_proto_tcp.c index 6ff05c3a32e6..7de385267b33 100644 --- a/net/ipv4/ipvs/ip_vs_proto_tcp.c +++ b/net/ipv4/ipvs/ip_vs_proto_tcp.c @@ -84,7 +84,7 @@ tcp_conn_schedule(struct sk_buff *skb, } if (th->syn && - (svc = ip_vs_service_get(skb->nfmark, skb->nh.iph->protocol, + (svc = ip_vs_service_get(skb->mark, skb->nh.iph->protocol, skb->nh.iph->daddr, th->dest))) { if (ip_vs_todrop()) { /* diff --git a/net/ipv4/ipvs/ip_vs_proto_udp.c b/net/ipv4/ipvs/ip_vs_proto_udp.c index 691c8b637b29..452cb9c384b3 100644 --- a/net/ipv4/ipvs/ip_vs_proto_udp.c +++ b/net/ipv4/ipvs/ip_vs_proto_udp.c @@ -89,7 +89,7 @@ udp_conn_schedule(struct sk_buff *skb, struct ip_vs_protocol *pp, return 0; } - if ((svc = ip_vs_service_get(skb->nfmark, skb->nh.iph->protocol, + if ((svc = ip_vs_service_get(skb->mark, skb->nh.iph->protocol, skb->nh.iph->daddr, uh->dest))) { if (ip_vs_todrop()) { /* diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index e2005c6810a4..bfc8d753a23a 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -28,7 +28,7 @@ int ip_route_me_harder(struct sk_buff **pskb, unsigned addr_type) fl.nl_u.ip4_u.tos = RT_TOS(iph->tos); fl.oif = (*pskb)->sk ? (*pskb)->sk->sk_bound_dev_if : 0; #ifdef CONFIG_IP_ROUTE_FWMARK - fl.nl_u.ip4_u.fwmark = (*pskb)->nfmark; + fl.nl_u.ip4_u.fwmark = (*pskb)->mark; #endif if (ip_route_output_key(&rt, &fl) != 0) return -1; diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index 97556cc2e4e0..cd520df4dcf4 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -243,7 +243,7 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->data_len = data_len; pmsg->timestamp_sec = entry->skb->tstamp.off_sec; pmsg->timestamp_usec = entry->skb->tstamp.off_usec; - pmsg->mark = entry->skb->nfmark; + pmsg->mark = entry->skb->mark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index 264763adc39b..f0319e5ee437 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -76,7 +76,7 @@ static void send_reset(struct sk_buff *oldskb, int hook) /* This packet will not be the same as the other: clear nf fields */ nf_reset(nskb); - nskb->nfmark = 0; + nskb->mark = 0; skb_init_secmark(nskb); tcph = (struct tcphdr *)((u_int32_t*)nskb->nh.iph + nskb->nh.iph->ihl); diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 2b104ea54f48..dbd34783a64d 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -239,7 +239,7 @@ static void ipt_ulog_packet(unsigned int hooknum, pm->data_len = copy_len; pm->timestamp_sec = skb->tstamp.off_sec; pm->timestamp_usec = skb->tstamp.off_usec; - pm->mark = skb->nfmark; + pm->mark = skb->mark; pm->hook = hooknum; if (prefix != NULL) strncpy(pm->prefix, prefix, sizeof(pm->prefix)); diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index b91f3582359b..62d4ccc259ca 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -132,7 +132,7 @@ ipt_local_hook(unsigned int hook, unsigned int ret; u_int8_t tos; __be32 saddr, daddr; - unsigned long nfmark; + u_int32_t mark; /* root is playing with raw sockets. */ if ((*pskb)->len < sizeof(struct iphdr) @@ -143,7 +143,7 @@ ipt_local_hook(unsigned int hook, } /* Save things which could affect route */ - nfmark = (*pskb)->nfmark; + mark = (*pskb)->mark; saddr = (*pskb)->nh.iph->saddr; daddr = (*pskb)->nh.iph->daddr; tos = (*pskb)->nh.iph->tos; @@ -154,7 +154,7 @@ ipt_local_hook(unsigned int hook, && ((*pskb)->nh.iph->saddr != saddr || (*pskb)->nh.iph->daddr != daddr #ifdef CONFIG_IP_ROUTE_FWMARK - || (*pskb)->nfmark != nfmark + || (*pskb)->mark != mark #endif || (*pskb)->nh.iph->tos != tos)) if (ip_route_me_harder(pskb, RTN_UNSPEC)) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 925ee4dfc32c..4de3e38fa1a8 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1644,7 +1644,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -1790,7 +1790,7 @@ static inline int __mkroute_input(struct sk_buff *skb, rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -1921,7 +1921,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, .tos = tos, .scope = RT_SCOPE_UNIVERSE, #ifdef CONFIG_IP_ROUTE_FWMARK - .fwmark = skb->nfmark + .fwmark = skb->mark #endif } }, .iif = dev->ifindex }; @@ -2035,7 +2035,7 @@ local_input: rth->rt_dst = daddr; rth->fl.fl4_tos = tos; #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark= skb->nfmark; + rth->fl.fl4_fwmark= skb->mark; #endif rth->fl.fl4_src = saddr; rth->rt_src = saddr; @@ -2114,7 +2114,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.iif == iif && rth->fl.oif == 0 && #ifdef CONFIG_IP_ROUTE_FWMARK - rth->fl.fl4_fwmark == skb->nfmark && + rth->fl.fl4_fwmark == skb->mark && #endif rth->fl.fl4_tos == tos) { rth->u.dst.lastuse = jiffies; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 93330685adfc..1bde3aca3466 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -499,12 +499,12 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) dst_release(to->dst); to->dst = dst_clone(from->dst); to->dev = from->dev; + to->mark = from->mark; #ifdef CONFIG_NET_SCHED to->tc_index = from->tc_index; #endif #ifdef CONFIG_NETFILTER - to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ nf_conntrack_put(to->nfct); to->nfct = from->nfct; diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index 9fec832ee08b..21908c9a10da 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -241,7 +241,7 @@ ipq_build_packet_message(struct ipq_queue_entry *entry, int *errp) pmsg->data_len = data_len; pmsg->timestamp_sec = entry->skb->tstamp.off_sec; pmsg->timestamp_usec = entry->skb->tstamp.off_usec; - pmsg->mark = entry->skb->nfmark; + pmsg->mark = entry->skb->mark; pmsg->hook = entry->info->hook; pmsg->hw_protocol = entry->skb->protocol; diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 386ea260e767..6250e86a6ddc 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -149,11 +149,10 @@ ip6t_local_hook(unsigned int hook, int (*okfn)(struct sk_buff *)) { - unsigned long nfmark; unsigned int ret; struct in6_addr saddr, daddr; u_int8_t hop_limit; - u_int32_t flowlabel; + u_int32_t flowlabel, mark; #if 0 /* root is playing with raw sockets. */ @@ -165,10 +164,10 @@ ip6t_local_hook(unsigned int hook, } #endif - /* save source/dest address, nfmark, hoplimit, flowlabel, priority, */ + /* save source/dest address, mark, hoplimit, flowlabel, priority, */ memcpy(&saddr, &(*pskb)->nh.ipv6h->saddr, sizeof(saddr)); memcpy(&daddr, &(*pskb)->nh.ipv6h->daddr, sizeof(daddr)); - nfmark = (*pskb)->nfmark; + mark = (*pskb)->mark; hop_limit = (*pskb)->nh.ipv6h->hop_limit; /* flowlabel and prio (includes version, which shouldn't change either */ @@ -179,7 +178,7 @@ ip6t_local_hook(unsigned int hook, if (ret != NF_DROP && ret != NF_STOLEN && (memcmp(&(*pskb)->nh.ipv6h->saddr, &saddr, sizeof(saddr)) || memcmp(&(*pskb)->nh.ipv6h->daddr, &daddr, sizeof(daddr)) - || (*pskb)->nfmark != nfmark + || (*pskb)->mark != mark || (*pskb)->nh.ipv6h->hop_limit != hop_limit)) return ip6_route_me_harder(*pskb) == 0 ? ret : NF_DROP; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e9c1fc5f21b1..aaabb1fad1cf 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -712,7 +712,7 @@ void ip6_route_input(struct sk_buff *skb) .daddr = iph->daddr, .saddr = iph->saddr, #ifdef CONFIG_IPV6_ROUTE_FWMARK - .fwmark = skb->nfmark, + .fwmark = skb->mark, #endif .flowlabel = (* (__be32 *) iph)&IPV6_FLOWINFO_MASK, }, diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 856ed0d19974..bd3ffa6f1a6d 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -501,8 +501,8 @@ __build_packet_message(struct nfulnl_instance *inst, #endif } - if (skb->nfmark) { - tmp_uint = htonl(skb->nfmark); + if (skb->mark) { + tmp_uint = htonl(skb->mark); NFA_PUT(inst->skb, NFULA_MARK, sizeof(tmp_uint), &tmp_uint); } diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 4ab7b1416bb5..82e4454659bf 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -480,8 +480,8 @@ nfqnl_build_packet_message(struct nfqnl_instance *queue, #endif } - if (entskb->nfmark) { - tmp_uint = htonl(entskb->nfmark); + if (entskb->mark) { + tmp_uint = htonl(entskb->mark); NFA_PUT(skb, NFQA_MARK, sizeof(u_int32_t), &tmp_uint); } @@ -834,8 +834,8 @@ nfqnl_recv_verdict(struct sock *ctnl, struct sk_buff *skb, } if (nfqa[NFQA_MARK-1]) - entry->skb->nfmark = ntohl(*(__be32 *) - NFA_DATA(nfqa[NFQA_MARK-1])); + entry->skb->mark = ntohl(*(__be32 *) + NFA_DATA(nfqa[NFQA_MARK-1])); issue_verdict(entry, verdict); instance_put(queue); diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c index c01524f817f0..67ed53152999 100644 --- a/net/netfilter/xt_CONNMARK.c +++ b/net/netfilter/xt_CONNMARK.c @@ -42,7 +42,7 @@ target(struct sk_buff **pskb, { const struct xt_connmark_target_info *markinfo = targinfo; u_int32_t diff; - u_int32_t nfmark; + u_int32_t mark; u_int32_t newmark; u_int32_t ctinfo; u_int32_t *ctmark = nf_ct_get_mark(*pskb, &ctinfo); @@ -62,7 +62,7 @@ target(struct sk_buff **pskb, break; case XT_CONNMARK_SAVE: newmark = (*ctmark & ~markinfo->mask) | - ((*pskb)->nfmark & markinfo->mask); + ((*pskb)->mark & markinfo->mask); if (*ctmark != newmark) { *ctmark = newmark; #if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) @@ -73,10 +73,10 @@ target(struct sk_buff **pskb, } break; case XT_CONNMARK_RESTORE: - nfmark = (*pskb)->nfmark; - diff = (*ctmark ^ nfmark) & markinfo->mask; + mark = (*pskb)->mark; + diff = (*ctmark ^ mark) & markinfo->mask; if (diff != 0) - (*pskb)->nfmark = nfmark ^ diff; + (*pskb)->mark = mark ^ diff; break; } } diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c index c6e860a7114f..0b48547e8d64 100644 --- a/net/netfilter/xt_MARK.c +++ b/net/netfilter/xt_MARK.c @@ -31,8 +31,8 @@ target_v0(struct sk_buff **pskb, { const struct xt_mark_target_info *markinfo = targinfo; - if((*pskb)->nfmark != markinfo->mark) - (*pskb)->nfmark = markinfo->mark; + if((*pskb)->mark != markinfo->mark) + (*pskb)->mark = markinfo->mark; return XT_CONTINUE; } @@ -54,16 +54,16 @@ target_v1(struct sk_buff **pskb, break; case XT_MARK_AND: - mark = (*pskb)->nfmark & markinfo->mark; + mark = (*pskb)->mark & markinfo->mark; break; case XT_MARK_OR: - mark = (*pskb)->nfmark | markinfo->mark; + mark = (*pskb)->mark | markinfo->mark; break; } - if((*pskb)->nfmark != mark) - (*pskb)->nfmark = mark; + if((*pskb)->mark != mark) + (*pskb)->mark = mark; return XT_CONTINUE; } diff --git a/net/netfilter/xt_mark.c b/net/netfilter/xt_mark.c index 934dddfbcd23..dfa1ee6914c0 100644 --- a/net/netfilter/xt_mark.c +++ b/net/netfilter/xt_mark.c @@ -31,7 +31,7 @@ match(const struct sk_buff *skb, { const struct xt_mark_info *info = matchinfo; - return ((skb->nfmark & info->mask) == info->mark) ^ info->invert; + return ((skb->mark & info->mask) == info->mark) ^ info->invert; } static int diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 8298ea9ffe19..b2437092978c 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -320,7 +320,7 @@ config CLS_U32_PERF config CLS_U32_MARK bool "Netfilter marks support" - depends on NET_CLS_U32 && NETFILTER + depends on NET_CLS_U32 ---help--- Say Y here to be able to use netfilter marks as u32 key. diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index e54acc6bcccd..f59a2c4aa039 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -101,11 +101,7 @@ static int fw_classify(struct sk_buff *skb, struct tcf_proto *tp, struct fw_head *head = (struct fw_head*)tp->root; struct fw_filter *f; int r; -#ifdef CONFIG_NETFILTER - u32 id = skb->nfmark & head->mask; -#else - u32 id = 0; -#endif + u32 id = skb->mark & head->mask; if (head != NULL) { for (f=head->ht[fw_hash(id)]; f; f=f->next) { diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 0a6cfa0005be..8b5194801995 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -143,7 +143,7 @@ next_knode: #endif #ifdef CONFIG_CLS_U32_MARK - if ((skb->nfmark & n->mark.mask) != n->mark.val) { + if ((skb->mark & n->mark.mask) != n->mark.val) { n = n->next; goto next_knode; } else { diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 61e3b740ab1a..d3ff3503326a 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -208,13 +208,9 @@ META_COLLECTOR(int_maclen) * Netfilter **************************************************************************/ -META_COLLECTOR(int_nfmark) +META_COLLECTOR(int_mark) { -#ifdef CONFIG_NETFILTER - dst->value = skb->nfmark; -#else - dst->value = 0; -#endif + dst->value = skb->mark; } /************************************************************************** @@ -490,7 +486,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [META_ID(PKTLEN)] = META_FUNC(int_pktlen), [META_ID(DATALEN)] = META_FUNC(int_datalen), [META_ID(MACLEN)] = META_FUNC(int_maclen), - [META_ID(NFMARK)] = META_FUNC(int_nfmark), + [META_ID(NFMARK)] = META_FUNC(int_mark), [META_ID(TCINDEX)] = META_FUNC(int_tcindex), [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), [META_ID(RTIIF)] = META_FUNC(int_rtiif), -- cgit v1.2.3 From b8964ed9fa727109c9084abc807652ebfb681c18 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:22:18 -0800 Subject: [NET] rules: Protocol independant mark selector Move mark selector currently implemented per protocol into the protocol independant part. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/fib_rules.h | 2 +- include/net/fib_rules.h | 2 ++ net/core/fib_rules.c | 29 +++++++++++++++++++++++++++++ net/decnet/dn_rules.c | 27 --------------------------- net/ipv4/fib_rules.c | 29 ----------------------------- net/ipv6/fib6_rules.c | 32 -------------------------------- 6 files changed, 32 insertions(+), 89 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h index 4418c8d9d479..adcdfbdd14d5 100644 --- a/include/linux/fib_rules.h +++ b/include/linux/fib_rules.h @@ -34,7 +34,7 @@ enum FRA_UNUSED3, FRA_UNUSED4, FRA_UNUSED5, - FRA_FWMARK, /* netfilter mark */ + FRA_FWMARK, /* mark */ FRA_FLOW, /* flow/class id */ FRA_UNUSED6, FRA_UNUSED7, diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 8e2f473d3e82..68542b565cf7 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -13,6 +13,8 @@ struct fib_rule atomic_t refcnt; int ifindex; char ifname[IFNAMSIZ]; + u32 mark; + u32 mark_mask; u32 pref; u32 flags; u32 table; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 6b0e63cacd93..da91bf2e6151 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -119,6 +119,9 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, if (rule->ifindex && (rule->ifindex != fl->iif)) continue; + if ((rule->mark ^ fl->mark) & rule->mark_mask) + continue; + if (!ops->match(rule, fl, flags)) continue; @@ -179,6 +182,18 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) rule->ifindex = dev->ifindex; } + if (tb[FRA_FWMARK]) { + rule->mark = nla_get_u32(tb[FRA_FWMARK]); + if (rule->mark) + /* compatibility: if the mark value is non-zero all bits + * are compared unless a mask is explicitly specified. + */ + rule->mark_mask = 0xFFFFFFFF; + } + + if (tb[FRA_FWMASK]) + rule->mark_mask = nla_get_u32(tb[FRA_FWMASK]); + rule->action = frh->action; rule->flags = frh->flags; rule->table = frh_get_table(frh, tb); @@ -250,6 +265,14 @@ int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) nla_strcmp(tb[FRA_IFNAME], rule->ifname)) continue; + if (tb[FRA_FWMARK] && + (rule->mark != nla_get_u32(tb[FRA_FWMARK]))) + continue; + + if (tb[FRA_FWMASK] && + (rule->mark_mask != nla_get_u32(tb[FRA_FWMASK]))) + continue; + if (!ops->compare(rule, frh, tb)) continue; @@ -298,6 +321,12 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, if (rule->pref) NLA_PUT_U32(skb, FRA_PRIORITY, rule->pref); + if (rule->mark) + NLA_PUT_U32(skb, FRA_FWMARK, rule->mark); + + if (rule->mark_mask || rule->mark) + NLA_PUT_U32(skb, FRA_FWMASK, rule->mark_mask); + if (ops->fill(rule, skb, nlh, frh) < 0) goto nla_put_failure; diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index a09c09f5572d..1f5d23c96681 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -45,8 +45,6 @@ struct dn_fib_rule __le16 dstmask; __le16 srcmap; u8 flags; - u32 fwmark; - u32 fwmask; }; static struct dn_fib_rule default_rule = { @@ -129,9 +127,6 @@ static int dn_fib_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) ((daddr ^ r->dst) & r->dstmask)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -165,18 +160,6 @@ static int dn_fib_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) r->dst = nla_get_u16(tb[FRA_DST]); - if (tb[FRA_FWMARK]) { - r->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (r->fwmark) - /* compatibility: if the mark value is non-zero all bits - * are compared unless a mask is explicitly specified. - */ - r->fwmask = 0xFFFFFFFF; - } - - if (tb[FRA_FWMASK]) - r->fwmask = nla_get_u32(tb[FRA_FWMASK]); - r->src_len = frh->src_len; r->srcmask = dnet_make_mask(r->src_len); r->dst_len = frh->dst_len; @@ -197,12 +180,6 @@ static int dn_fib_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->dst_len && (r->dst_len != frh->dst_len)) return 0; - if (tb[FRA_FWMARK] && (r->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (r->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - if (tb[FRA_SRC] && (r->src != nla_get_u16(tb[FRA_SRC]))) return 0; @@ -240,10 +217,6 @@ static int dn_fib_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = r->src_len; frh->tos = 0; - if (r->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, r->fwmark); - if (r->fwmask || r->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, r->fwmask); if (r->dst_len) NLA_PUT_U16(skb, FRA_DST, r->dst); if (r->src_len) diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index de8d5dd7099b..d2a190a35d65 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -44,8 +44,6 @@ struct fib4_rule __be32 srcmask; __be32 dst; __be32 dstmask; - u32 fwmark; - u32 fwmask; #ifdef CONFIG_NET_CLS_ROUTE u32 tclassid; #endif @@ -158,9 +156,6 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tos && (r->tos != fl->fl4_tos)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -216,18 +211,6 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (tb[FRA_DST]) rule4->dst = nla_get_be32(tb[FRA_DST]); - if (tb[FRA_FWMARK]) { - rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (rule4->fwmark) - /* compatibility: if the mark value is non-zero all bits - * are compared unless a mask is explicitly specified. - */ - rule4->fwmask = 0xFFFFFFFF; - } - - if (tb[FRA_FWMASK]) - rule4->fwmask = nla_get_u32(tb[FRA_FWMASK]); - #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW]) rule4->tclassid = nla_get_u32(tb[FRA_FLOW]); @@ -258,12 +241,6 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, if (frh->tos && (rule4->tos != frh->tos)) return 0; - if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (rule4->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - #ifdef CONFIG_NET_CLS_ROUTE if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW]))) return 0; @@ -288,12 +265,6 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb, frh->src_len = rule4->src_len; frh->tos = rule4->tos; - if (rule4->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark); - - if (rule4->fwmask || rule4->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, rule4->fwmask); - if (rule4->dst_len) NLA_PUT_BE32(skb, FRA_DST, rule4->dst); diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 89bea64eee1c..26374cbe2bc2 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -25,8 +25,6 @@ struct fib6_rule struct fib_rule common; struct rt6key src; struct rt6key dst; - u32 fwmark; - u32 fwmask; u8 tclass; }; @@ -128,9 +126,6 @@ static int fib6_rule_match(struct fib_rule *rule, struct flowi *fl, int flags) if (r->tclass && r->tclass != ((ntohl(fl->fl6_flowlabel) >> 20) & 0xff)) return 0; - if ((r->fwmark ^ fl->mark) & r->fwmask) - return 0; - return 1; } @@ -173,21 +168,6 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, nla_memcpy(&rule6->dst.addr, tb[FRA_DST], sizeof(struct in6_addr)); - if (tb[FRA_FWMARK]) { - rule6->fwmark = nla_get_u32(tb[FRA_FWMARK]); - if (rule6->fwmark) { - /* - * if the mark value is non-zero, - * all bits are compared by default - * unless a mask is explicitly specified. - */ - rule6->fwmask = 0xFFFFFFFF; - } - } - - if (tb[FRA_FWMASK]) - rule6->fwmask = nla_get_u32(tb[FRA_FWMASK]); - rule6->src.plen = frh->src_len; rule6->dst.plen = frh->dst_len; rule6->tclass = frh->tos; @@ -219,12 +199,6 @@ static int fib6_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh, nla_memcmp(tb[FRA_DST], &rule6->dst.addr, sizeof(struct in6_addr))) return 0; - if (tb[FRA_FWMARK] && (rule6->fwmark != nla_get_u32(tb[FRA_FWMARK]))) - return 0; - - if (tb[FRA_FWMASK] && (rule6->fwmask != nla_get_u32(tb[FRA_FWMASK]))) - return 0; - return 1; } @@ -246,12 +220,6 @@ static int fib6_rule_fill(struct fib_rule *rule, struct sk_buff *skb, NLA_PUT(skb, FRA_SRC, sizeof(struct in6_addr), &rule6->src.addr); - if (rule6->fwmark) - NLA_PUT_U32(skb, FRA_FWMARK, rule6->fwmark); - - if (rule6->fwmask || rule6->fwmark) - NLA_PUT_U32(skb, FRA_FWMASK, rule6->fwmask); - return 0; nla_put_failure: -- cgit v1.2.3 From 3dfbcc411e461db51a1ac1aa1c6ebe2c5a0275a0 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 9 Nov 2006 15:23:20 -0800 Subject: [NET] rules: Add support to invert selectors Introduces a new flag FIB_RULE_INVERT causing rules to apply if the specified selector doesn't match. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/fib_rules.h | 1 + net/core/fib_rules.c | 24 +++++++++++++++++------- 2 files changed, 18 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fib_rules.h b/include/linux/fib_rules.h index adcdfbdd14d5..8270aac2aa5d 100644 --- a/include/linux/fib_rules.h +++ b/include/linux/fib_rules.h @@ -6,6 +6,7 @@ /* rule is permanent, and cannot be deleted */ #define FIB_RULE_PERMANENT 1 +#define FIB_RULE_INVERT 2 struct fib_rule_hdr { diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index da91bf2e6151..4148e274a204 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -107,6 +107,22 @@ out: EXPORT_SYMBOL_GPL(fib_rules_unregister); +static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, + struct flowi *fl, int flags) +{ + int ret = 0; + + if (rule->ifindex && (rule->ifindex != fl->iif)) + goto out; + + if ((rule->mark ^ fl->mark) & rule->mark_mask) + goto out; + + ret = ops->match(rule, fl, flags); +out: + return (rule->flags & FIB_RULE_INVERT) ? !ret : ret; +} + int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, int flags, struct fib_lookup_arg *arg) { @@ -116,13 +132,7 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, rcu_read_lock(); list_for_each_entry_rcu(rule, ops->rules_list, list) { - if (rule->ifindex && (rule->ifindex != fl->iif)) - continue; - - if ((rule->mark ^ fl->mark) & rule->mark_mask) - continue; - - if (!ops->match(rule, fl, flags)) + if (!fib_rule_match(rule, ops, fl, flags)) continue; err = ops->action(rule, fl, flags, arg); -- cgit v1.2.3 From 3ff825b28d3345ef381eceae22bf9d92231f23dc Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:32:06 -0800 Subject: [TCP]: Add tcp_available_congestion_control sysctl. Create /proc/sys/net/ipv4/tcp_available_congestion_control that reflects currently available TCP choices. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 4 ++++ net/ipv4/sysctl_net_ipv4.c | 24 ++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 16 ++++++++++++++++ 5 files changed, 51 insertions(+) (limited to 'include/linux') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index fd3c0c012351..db4280856588 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,10 +351,16 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_available_congestion_control - STRING + Shows the available congestion control choices that are registered. + More congestion control algorithms may be available as modules, + but not loaded. + tcp_congestion_control - STRING Set the congestion control algorithm to be used for new connections. The algorithm "reno" is always available, but additional choices may be available based on kernel configuration. + Default is set as part of kernel configuration. somaxconn - INTEGER Limit of socket listen() backlog, known in userspace as SOMAXCONN. diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index d98562f1df76..28a48279654d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -426,6 +426,7 @@ enum NET_CIPSOV4_CACHE_BUCKET_SIZE=119, NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, + NET_TCP_AVAIL_CONG_CONTROL=122, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 246916c2321e..6af4baf5b769 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -620,6 +620,9 @@ enum tcp_ca_event { * Interface for adding new TCP congestion control handlers */ #define TCP_CA_NAME_MAX 16 +#define TCP_CA_MAX 128 +#define TCP_CA_BUF_MAX (TCP_CA_NAME_MAX*TCP_CA_MAX) + struct tcp_congestion_ops { struct list_head list; @@ -659,6 +662,7 @@ extern void tcp_init_congestion_control(struct sock *sk); extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); +extern void tcp_get_available_congestion_control(char *buf, size_t len); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 15061b314411..2e770f45d829 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -129,6 +129,23 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, return ret; } +static int proc_tcp_available_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + tcp_get_available_congestion_control(tbl.data, TCP_CA_BUF_MAX); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + kfree(tbl.data); + return ret; +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -731,6 +748,13 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_dointvec, }, #endif /* CONFIG_NETLABEL */ + { + .ctl_name = NET_TCP_AVAIL_CONG_CONTROL, + .procname = "tcp_available_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0444, + .proc_handler = &proc_tcp_available_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 1e2982f4acd4..d846d7b95e1f 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -139,6 +139,22 @@ static int __init tcp_congestion_default(void) late_initcall(tcp_congestion_default); +/* Build string with list of available congestion control values */ +void tcp_get_available_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + /* Get current default congestion control */ void tcp_get_default_congestion_control(char *name) { -- cgit v1.2.3 From ce7bc3bf15cbf5dc5a5587ccb6b04c5b4dde4336 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 9 Nov 2006 16:35:15 -0800 Subject: [TCP]: Restrict congestion control choices. Allow normal users to only choose among a restricted set of congestion control choices. The default is reno and what ever has been configured as default. But the policy can be changed by administrator at any time. For example, to allow any choice: cp /proc/sys/net/ipv4/tcp_available_congestion_control \ /proc/sys/net/ipv4/tcp_allowed_congestion_control Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 6 ++++ include/linux/sysctl.h | 1 + include/net/tcp.h | 3 ++ net/ipv4/sysctl_net_ipv4.c | 52 ++++++++++++++++++++++++++++ net/ipv4/tcp_cong.c | 63 ++++++++++++++++++++++++++++++++++ 5 files changed, 125 insertions(+) (limited to 'include/linux') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index db4280856588..bbcc8deda172 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -351,6 +351,12 @@ tcp_frto - BOOLEAN where packet loss is typically due to random radio interference rather than intermediate router congestion. +tcp_allowed_congestion_control - STRING + Show/set the congestion control choices available to non-privileged + processes. The list is a subset of those listed in + tcp_available_congestion_control. + Default is "reno" and the default setting (tcp_congestion_control). + tcp_available_congestion_control - STRING Shows the available congestion control choices that are registered. More congestion control algorithms may be available as modules, diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 28a48279654d..0725441621d0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -427,6 +427,7 @@ enum NET_CIPSOV4_RBM_OPTFMT=120, NET_CIPSOV4_RBM_STRICTVALID=121, NET_TCP_AVAIL_CONG_CONTROL=122, + NET_TCP_ALLOWED_CONG_CONTROL=123, }; enum { diff --git a/include/net/tcp.h b/include/net/tcp.h index 6af4baf5b769..e1a5d29d0a1f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -625,6 +625,7 @@ enum tcp_ca_event { struct tcp_congestion_ops { struct list_head list; + int non_restricted; /* initialize private data (optional) */ void (*init)(struct sock *sk); @@ -663,6 +664,8 @@ extern void tcp_cleanup_congestion_control(struct sock *sk); extern int tcp_set_default_congestion_control(const char *name); extern void tcp_get_default_congestion_control(char *name); extern void tcp_get_available_congestion_control(char *buf, size_t len); +extern void tcp_get_allowed_congestion_control(char *buf, size_t len); +extern int tcp_set_allowed_congestion_control(char *allowed); extern int tcp_set_congestion_control(struct sock *sk, const char *name); extern void tcp_slow_start(struct tcp_sock *tp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 2e770f45d829..dfcf47f10f88 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -146,6 +146,50 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl, return ret; } +static int proc_allowed_congestion_control(ctl_table *ctl, + int write, struct file * filp, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen); + ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos); + if (write && ret == 0) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + return ret; +} + +static int strategy_allowed_congestion_control(ctl_table *table, int __user *name, + int nlen, void __user *oldval, + size_t __user *oldlenp, + void __user *newval, size_t newlen, + void **context) +{ + ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + tcp_get_available_congestion_control(tbl.data, tbl.maxlen); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, + context); + if (ret == 0 && newval && newlen) + ret = tcp_set_allowed_congestion_control(tbl.data); + kfree(tbl.data); + + return ret; + +} + ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_TCP_TIMESTAMPS, @@ -755,6 +799,14 @@ ctl_table ipv4_table[] = { .mode = 0444, .proc_handler = &proc_tcp_available_congestion_control, }, + { + .ctl_name = NET_TCP_ALLOWED_CONG_CONTROL, + .procname = "tcp_allowed_congestion_control", + .maxlen = TCP_CA_BUF_MAX, + .mode = 0644, + .proc_handler = &proc_allowed_congestion_control, + .strategy = &strategy_allowed_congestion_control, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index d846d7b95e1f..343d6197c92e 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -123,6 +123,7 @@ int tcp_set_default_congestion_control(const char *name) #endif if (ca) { + ca->non_restricted = 1; /* default is always allowed */ list_move(&ca->list, &tcp_cong_list); ret = 0; } @@ -168,6 +169,64 @@ void tcp_get_default_congestion_control(char *name) rcu_read_unlock(); } +/* Built list of non-restricted congestion control values */ +void tcp_get_allowed_congestion_control(char *buf, size_t maxlen) +{ + struct tcp_congestion_ops *ca; + size_t offs = 0; + + *buf = '\0'; + rcu_read_lock(); + list_for_each_entry_rcu(ca, &tcp_cong_list, list) { + if (!ca->non_restricted) + continue; + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", ca->name); + + } + rcu_read_unlock(); +} + +/* Change list of non-restricted congestion control */ +int tcp_set_allowed_congestion_control(char *val) +{ + struct tcp_congestion_ops *ca; + char *clone, *name; + int ret = 0; + + clone = kstrdup(val, GFP_USER); + if (!clone) + return -ENOMEM; + + spin_lock(&tcp_cong_list_lock); + /* pass 1 check for bad entries */ + while ((name = strsep(&clone, " ")) && *name) { + ca = tcp_ca_find(name); + if (!ca) { + ret = -ENOENT; + goto out; + } + } + + /* pass 2 clear */ + list_for_each_entry_rcu(ca, &tcp_cong_list, list) + ca->non_restricted = 0; + + /* pass 3 mark as allowed */ + while ((name = strsep(&val, " ")) && *name) { + ca = tcp_ca_find(name); + WARN_ON(!ca); + if (ca) + ca->non_restricted = 1; + } +out: + spin_unlock(&tcp_cong_list_lock); + + return ret; +} + + /* Change congestion control for socket */ int tcp_set_congestion_control(struct sock *sk, const char *name) { @@ -183,6 +242,9 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) if (!ca) err = -ENOENT; + else if (!(ca->non_restricted || capable(CAP_NET_ADMIN))) + err = -EPERM; + else if (!try_module_get(ca->owner)) err = -EBUSY; @@ -284,6 +346,7 @@ EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd); struct tcp_congestion_ops tcp_reno = { .name = "reno", + .non_restricted = 1, .owner = THIS_MODULE, .ssthresh = tcp_reno_ssthresh, .cong_avoid = tcp_reno_cong_avoid, -- cgit v1.2.3 From 9b42078ed6edfe04e9dc9a59b946ad912aeef717 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Fri, 10 Nov 2006 11:22:32 -0200 Subject: [DCCP]: Combine allocating & zeroing header space on skb This is a code simplification: it combines three often recurring operations into one inline function, * allocate `len' bytes header space in skb * fill these `len' bytes with zeroes * cast the start of this header space as dccp_hdr Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 7 +++++++ net/dccp/ipv4.c | 9 ++------- net/dccp/ipv6.c | 8 ++------ net/dccp/output.c | 14 +++----------- 4 files changed, 14 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 53553c99cad6..90d04ffddae8 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -256,6 +256,13 @@ static inline struct dccp_hdr *dccp_hdr(const struct sk_buff *skb) return (struct dccp_hdr *)skb->h.raw; } +static inline struct dccp_hdr *dccp_zeroed_hdr(struct sk_buff *skb, int headlen) +{ + skb->h.raw = skb_push(skb, headlen); + memset(skb->h.raw, 0, headlen); + return dccp_hdr(skb); +} + static inline struct dccp_hdr_ext *dccp_hdrx(const struct sk_buff *skb) { return (struct dccp_hdr_ext *)(skb->h.raw + sizeof(struct dccp_hdr)); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index de64e6c7f93d..ce8eed32dbeb 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -212,12 +212,9 @@ static void dccp_v4_reqsk_send_ack(struct sk_buff *rxskb, /* Reserve space for headers. */ skb_reserve(skb, dccp_v4_ctl_socket->sk->sk_prot->max_header); - skb->dst = dst_clone(rxskb->dst); - skb->h.raw = skb_push(skb, dccp_hdr_ack_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_ack_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_ack_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_ACK; @@ -720,9 +717,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) skb_reserve(skb, dccp_v4_ctl_socket->sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->h.raw = skb_push(skb, dccp_hdr_reset_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_reset_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_reset_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_RESET; diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 6f1c2ad88608..116bddb64b80 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -537,9 +537,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); - skb->h.raw = skb_push(skb, dccp_hdr_reset_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_reset_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_reset_len); /* Swap the send and the receive. */ dh->dccph_type = DCCP_PKT_RESET; @@ -601,9 +599,7 @@ static void dccp_v6_reqsk_send_ack(struct sk_buff *rxskb, skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); - skb->h.raw = skb_push(skb, dccp_hdr_ack_len); - dh = dccp_hdr(skb); - memset(dh, 0, dccp_hdr_ack_len); + dh = dccp_zeroed_hdr(skb, dccp_hdr_ack_len); /* Build DCCP header and checksum it. */ dh->dccph_type = DCCP_PKT_ACK; diff --git a/net/dccp/output.c b/net/dccp/output.c index 2cc4f4b2a9dd..1ae2248557c6 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -88,11 +88,9 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) return -EPROTO; } - skb->h.raw = skb_push(skb, dccp_header_size); - dh = dccp_hdr(skb); /* Build DCCP header and checksum it. */ - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_type = dcb->dccpd_type; dh->dccph_sport = inet->sport; dh->dccph_dport = inet->dport; @@ -340,10 +338,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, return NULL; } - skb->h.raw = skb_push(skb, dccp_header_size); - - dh = dccp_hdr(skb); - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; dh->dccph_dport = inet_rsk(req)->rmt_port; @@ -392,10 +387,7 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, return NULL; } - skb->h.raw = skb_push(skb, dccp_header_size); - - dh = dccp_hdr(skb); - memset(dh, 0, dccp_header_size); + dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; dh->dccph_dport = inet_sk(sk)->dport; -- cgit v1.2.3 From 6f4e5fff1e4d46714ea554fd83e44eab534e8b11 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Fri, 10 Nov 2006 17:43:06 -0200 Subject: [DCCP]: Support for partial checksums (RFC 4340, sec. 9.2) This patch does the following: a) introduces variable-length checksums as specified in [RFC 4340, sec. 9.2] b) provides necessary socket options and documentation as to how to use them c) basic support and infrastructure for the Minimum Checksum Coverage feature [RFC 4340, sec. 9.2.1]: acceptability tests, user notification and user interface In addition, it (1) fixes two bugs in the DCCPv4 checksum computation: * pseudo-header used checksum_len instead of skb->len * incorrect checksum coverage calculation based on dccph_x (2) removes dccp_v4_verify_checksum() since it reduplicates code of the checksum computation; code calling this function is updated accordingly. (3) now uses skb_checksum(), which is safer than checksum_partial() if the sk_buff has is a non-linear buffer (has pages attached to it). (4) fixes an outstanding TODO item: * If P.CsCov is too large for the packet size, drop packet and return. The code has been tested with applications, the latest version of tcpdump now comes with support for partial DCCP checksums. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 16 ++++++ include/linux/dccp.h | 7 +++ net/dccp/dccp.h | 29 +++++++++-- net/dccp/ipv4.c | 105 +++++++++++++++++++------------------- net/dccp/ipv6.c | 74 +++++++++++++++++---------- net/dccp/output.c | 13 ++--- net/dccp/proto.c | 26 ++++++++-- 7 files changed, 173 insertions(+), 97 deletions(-) (limited to 'include/linux') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index 74563b38ffd9..a8142a81038a 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -47,6 +47,22 @@ the socket will fall back to 0 (which means that no meaningful service code is present). Connecting sockets set at most one service option; for listening sockets, multiple service codes can be specified. +DCCP_SOCKOPT_SEND_CSCOV and DCCP_SOCKOPT_RECV_CSCOV are used for setting the +partial checksum coverage (RFC 4340, sec. 9.2). The default is that checksums +always cover the entire packet and that only fully covered application data is +accepted by the receiver. Hence, when using this feature on the sender, it must +be enabled at the receiver, too with suitable choice of CsCov. + +DCCP_SOCKOPT_SEND_CSCOV sets the sender checksum coverage. Values in the + range 0..15 are acceptable. The default setting is 0 (full coverage), + values between 1..15 indicate partial coverage. +DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it + sets a threshold, where again values 0..15 are acceptable. The default + of 0 means that all packets with a partial coverage will be discarded. + Values in the range 1..15 indicate that packets with minimally such a + coverage value are also acceptable. The higher the number, the more + restrictive this setting (see [RFC 4340, sec. 9.2.1]). + Notes ===== diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 90d04ffddae8..0502dfa7f32c 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -183,6 +183,7 @@ enum { DCCPF_ACK_RATIO = 5, DCCPF_SEND_ACK_VECTOR = 6, DCCPF_SEND_NDP_COUNT = 7, + DCCPF_MIN_CSUM_COVER = 8, /* 10-127 reserved */ DCCPF_MIN_CCID_SPECIFIC = 128, DCCPF_MAX_CCID_SPECIFIC = 255, @@ -200,6 +201,8 @@ struct dccp_so_feat { #define DCCP_SOCKOPT_SERVICE 2 #define DCCP_SOCKOPT_CHANGE_L 3 #define DCCP_SOCKOPT_CHANGE_R 4 +#define DCCP_SOCKOPT_SEND_CSCOV 10 +#define DCCP_SOCKOPT_RECV_CSCOV 11 #define DCCP_SOCKOPT_CCID_RX_INFO 128 #define DCCP_SOCKOPT_CCID_TX_INFO 192 @@ -450,6 +453,8 @@ struct dccp_ackvec; * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option * @dccps_packet_size - Set thru setsockopt * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_pcslen - sender partial checksum coverage (via sockopt) + * @dccps_pcrlen - receiver partial checksum coverage (via sockopt) * @dccps_ndp_count - number of Non Data Packets since last data packet * @dccps_hc_rx_ackvec - rx half connection ack vector * @dccps_xmit_timer - timer for when CCID is not ready to send @@ -474,6 +479,8 @@ struct dccp_sock { __u32 dccps_packet_size; __u16 dccps_l_ack_ratio; __u16 dccps_r_ack_ratio; + __u16 dccps_pcslen; + __u16 dccps_pcrlen; unsigned long dccps_ndp_count; __u32 dccps_mss_cache; struct dccp_minisock dccps_minisock; diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index 2990bfb12587..d5c414bf7819 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -129,6 +129,30 @@ DECLARE_SNMP_STAT(struct dccp_mib, dccp_statistics); #define DCCP_ADD_STATS_USER(field, val) \ SNMP_ADD_STATS_USER(dccp_statistics, field, val) +/* + * Checksumming routines + */ +static inline int dccp_csum_coverage(const struct sk_buff *skb) +{ + const struct dccp_hdr* dh = dccp_hdr(skb); + + if (dh->dccph_cscov == 0) + return skb->len; + return (dh->dccph_doff + dh->dccph_cscov - 1) * sizeof(u32); +} + +static inline void dccp_csum_outgoing(struct sk_buff *skb) +{ + int cov = dccp_csum_coverage(skb); + + if (cov >= skb->len) + dccp_hdr(skb)->dccph_cscov = 0; + + skb->csum = skb_checksum(skb, 0, (cov > skb->len)? skb->len : cov, 0); +} + +extern void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb); + extern int dccp_retransmit_skb(struct sock *sk, struct sk_buff *skb); extern void dccp_send_ack(struct sock *sk); @@ -214,14 +238,9 @@ extern void dccp_shutdown(struct sock *sk, int how); extern int inet_dccp_listen(struct socket *sock, int backlog); extern unsigned int dccp_poll(struct file *file, struct socket *sock, poll_table *wait); -extern void dccp_v4_send_check(struct sock *sk, int len, - struct sk_buff *skb); extern int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len); -extern int dccp_v4_checksum(const struct sk_buff *skb, - const __be32 saddr, const __be32 daddr); - extern int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code); extern void dccp_send_close(struct sock *sk, const int active); extern int dccp_invalid_packet(struct sk_buff *skb); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 91bffaa761a6..496112080f3d 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -349,13 +349,19 @@ out: sock_put(sk); } -/* This routine computes an IPv4 DCCP checksum. */ -void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) +static inline u16 dccp_v4_csum_finish(struct sk_buff *skb, + __be32 src, __be32 dst) +{ + return csum_tcpudp_magic(src, dst, skb->len, IPPROTO_DCCP, skb->csum); +} + +void dccp_v4_send_check(struct sock *sk, int unused, struct sk_buff *skb) { const struct inet_sock *inet = inet_sk(sk); struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v4_checksum(skb, inet->saddr, inet->daddr); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v4_csum_finish(skb, inet->saddr, inet->daddr); } EXPORT_SYMBOL_GPL(dccp_v4_send_check); @@ -454,47 +460,6 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -int dccp_v4_checksum(const struct sk_buff *skb, const __be32 saddr, - const __be32 daddr) -{ - const struct dccp_hdr* dh = dccp_hdr(skb); - int checksum_len; - u32 tmp; - - if (dh->dccph_cscov == 0) - checksum_len = skb->len; - else { - checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32); - checksum_len = checksum_len < skb->len ? checksum_len : - skb->len; - } - - tmp = csum_partial((unsigned char *)dh, checksum_len, 0); - return csum_tcpudp_magic(saddr, daddr, checksum_len, - IPPROTO_DCCP, tmp); -} - -EXPORT_SYMBOL_GPL(dccp_v4_checksum); - -static int dccp_v4_verify_checksum(struct sk_buff *skb, - const __be32 saddr, const __be32 daddr) -{ - struct dccp_hdr *dh = dccp_hdr(skb); - int checksum_len; - u32 tmp; - - if (dh->dccph_cscov == 0) - checksum_len = skb->len; - else { - checksum_len = (dh->dccph_cscov + dh->dccph_x) * sizeof(u32); - checksum_len = checksum_len < skb->len ? checksum_len : - skb->len; - } - tmp = csum_partial((unsigned char *)dh, checksum_len, 0); - return csum_tcpudp_magic(saddr, daddr, checksum_len, - IPPROTO_DCCP, tmp) == 0 ? 0 : -1; -} - static struct dst_entry* dccp_v4_route_skb(struct sock *sk, struct sk_buff *skb) { @@ -536,8 +501,8 @@ static int dccp_v4_send_response(struct sock *sk, struct request_sock *req, const struct inet_request_sock *ireq = inet_rsk(req); struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v4_checksum(skb, ireq->loc_addr, - ireq->rmt_addr); + dh->dccph_checksum = dccp_v4_csum_finish(skb, ireq->loc_addr, + ireq->rmt_addr); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, @@ -602,8 +567,9 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); - dh->dccph_checksum = dccp_v4_checksum(skb, rxskb->nh.iph->saddr, - rxskb->nh.iph->daddr); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v4_csum_finish(skb, rxskb->nh.iph->saddr, + rxskb->nh.iph->daddr); bh_lock_sock(dccp_v4_ctl_socket->sk); err = ip_build_and_send_pkt(skb, dccp_v4_ctl_socket->sk, @@ -779,6 +745,7 @@ EXPORT_SYMBOL_GPL(dccp_v4_do_rcv); int dccp_invalid_packet(struct sk_buff *skb) { const struct dccp_hdr *dh; + unsigned int cscov; if (skb->pkt_type != PACKET_HOST) return 1; @@ -830,6 +797,22 @@ int dccp_invalid_packet(struct sk_buff *skb) return 1; } + /* + * If P.CsCov is too large for the packet size, drop packet and return. + * This must come _before_ checksumming (not as RFC 4340 suggests). + */ + cscov = dccp_csum_coverage(skb); + if (cscov > skb->len) { + LIMIT_NETDEBUG(KERN_WARNING + "DCCP: P.CsCov %u exceeds packet length %d\n", + dh->dccph_cscov, skb->len); + return 1; + } + + /* If header checksum is incorrect, drop packet and return. + * (This step is completed in the AF-dependent functions.) */ + skb->csum = skb_checksum(skb, 0, cscov, 0); + return 0; } @@ -840,16 +823,17 @@ static int dccp_v4_rcv(struct sk_buff *skb) { const struct dccp_hdr *dh; struct sock *sk; + int min_cov; - /* Step 1: Check header basics: */ + /* Step 1: Check header basics */ if (dccp_invalid_packet(skb)) goto discard_it; - /* If the header checksum is incorrect, drop packet and return */ - if (dccp_v4_verify_checksum(skb, skb->nh.iph->saddr, - skb->nh.iph->daddr) < 0) { - LIMIT_NETDEBUG(KERN_WARNING "%s: incorrect header checksum\n", + /* Step 1: If header checksum is incorrect, drop packet and return */ + if (dccp_v4_csum_finish(skb, skb->nh.iph->saddr, skb->nh.iph->daddr)) { + LIMIT_NETDEBUG(KERN_WARNING + "%s: dropped packet with invalid checksum\n", __FUNCTION__); goto discard_it; } @@ -905,6 +889,21 @@ static int dccp_v4_rcv(struct sk_buff *skb) goto no_dccp_socket; } + /* + * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + */ + min_cov = dccp_sk(sk)->dccps_pcrlen; + if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { + dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n", + dh->dccph_cscov, min_cov); + /* FIXME: "Such packets SHOULD be reported using Data Dropped + * options (Section 11.7) with Drop Code 0, Protocol + * Constraints." */ + goto discard_and_relse; + } + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; nf_reset(skb); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 201801e1532d..193b946fd039 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -58,12 +58,22 @@ static void dccp_v6_hash(struct sock *sk) } } -static inline u16 dccp_v6_check(struct dccp_hdr *dh, int len, - struct in6_addr *saddr, - struct in6_addr *daddr, - unsigned long base) +/* add pseudo-header to DCCP checksum stored in skb->csum */ +static inline u16 dccp_v6_csum_finish(struct sk_buff *skb, + struct in6_addr *saddr, + struct in6_addr *daddr) { - return csum_ipv6_magic(saddr, daddr, len, IPPROTO_DCCP, base); + return csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_DCCP, skb->csum); +} + +static inline void dccp_v6_send_check(struct sock *sk, int unused_value, + struct sk_buff *skb) +{ + struct ipv6_pinfo *np = inet6_sk(sk); + struct dccp_hdr *dh = dccp_hdr(skb); + + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v6_csum_finish(skb, &np->saddr, &np->daddr); } static __u32 dccp_v6_init_sequence(struct sock *sk, struct sk_buff *skb) @@ -280,12 +290,9 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, if (skb != NULL) { struct dccp_hdr *dh = dccp_hdr(skb); - dh->dccph_checksum = dccp_v6_check(dh, skb->len, - &ireq6->loc_addr, - &ireq6->rmt_addr, - csum_partial((char *)dh, - skb->len, - skb->csum)); + dh->dccph_checksum = dccp_v6_csum_finish(skb, + &ireq6->loc_addr, + &ireq6->rmt_addr); ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); if (err == NET_XMIT_CN) @@ -305,18 +312,6 @@ static void dccp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } -static void dccp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) -{ - struct ipv6_pinfo *np = inet6_sk(sk); - struct dccp_hdr *dh = dccp_hdr(skb); - - dh->dccph_checksum = csum_ipv6_magic(&np->saddr, &np->daddr, - len, IPPROTO_DCCP, - csum_partial((char *)dh, - dh->dccph_doff << 2, - skb->csum)); -} - static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) { struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; @@ -360,12 +355,14 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_csum_outgoing(skb); + dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, + &rxskb->nh.ipv6h->daddr); + memset(&fl, 0, sizeof(fl)); ipv6_addr_copy(&fl.fl6_dst, &rxskb->nh.ipv6h->saddr); ipv6_addr_copy(&fl.fl6_src, &rxskb->nh.ipv6h->daddr); - dh->dccph_checksum = csum_ipv6_magic(&fl.fl6_src, &fl.fl6_dst, - sizeof(*dh), IPPROTO_DCCP, - skb->csum); + fl.proto = IPPROTO_DCCP; fl.oif = inet6_iif(rxskb); fl.fl_ip_dport = dh->dccph_dport; @@ -825,12 +822,22 @@ static int dccp_v6_rcv(struct sk_buff **pskb) const struct dccp_hdr *dh; struct sk_buff *skb = *pskb; struct sock *sk; + int min_cov; - /* Step 1: Check header basics: */ + /* Step 1: Check header basics */ if (dccp_invalid_packet(skb)) goto discard_it; + /* Step 1: If header checksum is incorrect, drop packet and return. */ + if (dccp_v6_csum_finish(skb, &skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr)) { + LIMIT_NETDEBUG(KERN_WARNING + "%s: dropped packet with invalid checksum\n", + __FUNCTION__); + goto discard_it; + } + dh = dccp_hdr(skb); DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(skb); @@ -869,6 +876,19 @@ static int dccp_v6_rcv(struct sk_buff **pskb) goto no_dccp_socket; } + /* + * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + */ + min_cov = dccp_sk(sk)->dccps_pcrlen; + if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { + dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n", + dh->dccph_cscov, min_cov); + /* FIXME: send Data Dropped option (see also dccp_v4_rcv) */ + goto discard_and_relse; + } + if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; diff --git a/net/dccp/output.c b/net/dccp/output.c index 51654975e8ea..992caedd7725 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -96,6 +96,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) dh->dccph_dport = inet->dport; dh->dccph_doff = (dccp_header_size + dcb->dccpd_opt_len) / 4; dh->dccph_ccval = dcb->dccpd_ccval; + dh->dccph_cscov = dp->dccps_pcslen; /* XXX For now we're using only 48 bits sequence numbers */ dh->dccph_x = 1; @@ -115,7 +116,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) break; } - icsk->icsk_af_ops->send_check(sk, skb->len, skb); + icsk->icsk_af_ops->send_check(sk, 0, skb); if (set_ack) dccp_event_ack_sent(sk); @@ -329,7 +330,6 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, skb_reserve(skb, sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->csum = 0; dreq = dccp_rsk(req); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_RESPONSE; @@ -352,6 +352,8 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dreq->dreq_isr); dccp_hdr_response(skb)->dccph_resp_service = dreq->dreq_service; + dccp_csum_outgoing(skb); + DCCP_INC_STATS(DCCP_MIB_OUTSEGS); return skb; } @@ -376,7 +378,6 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, skb_reserve(skb, sk->sk_prot->max_header); skb->dst = dst_clone(dst); - skb->csum = 0; dccp_inc_seqno(&dp->dccps_gss); @@ -401,7 +402,7 @@ static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), dp->dccps_gsr); dccp_hdr_reset(skb)->dccph_reset_code = code; - inet_csk(sk)->icsk_af_ops->send_check(sk, skb->len, skb); + inet_csk(sk)->icsk_af_ops->send_check(sk, 0, skb); DCCP_INC_STATS(DCCP_MIB_OUTSEGS); return skb; @@ -475,7 +476,6 @@ int dccp_connect(struct sock *sk) skb_reserve(skb, sk->sk_prot->max_header); DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_REQUEST; - skb->csum = 0; dccp_skb_entail(sk, skb); dccp_transmit_skb(sk, skb_clone(skb, GFP_KERNEL)); @@ -507,7 +507,6 @@ void dccp_send_ack(struct sock *sk) /* Reserve space for headers */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = DCCP_PKT_ACK; dccp_transmit_skb(sk, skb); } @@ -561,7 +560,6 @@ void dccp_send_sync(struct sock *sk, const u64 seq, /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = pkt_type; DCCP_SKB_CB(skb)->dccpd_seq = seq; @@ -587,7 +585,6 @@ void dccp_send_close(struct sock *sk, const int active) /* Reserve space for headers and prepare control bits. */ skb_reserve(skb, sk->sk_prot->max_header); - skb->csum = 0; DCCP_SKB_CB(skb)->dccpd_type = dp->dccps_role == DCCP_ROLE_CLIENT ? DCCP_PKT_CLOSE : DCCP_PKT_CLOSEREQ; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 047d170a363a..db54e557eff1 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -472,7 +472,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, case DCCP_SOCKOPT_PACKET_SIZE: dp->dccps_packet_size = val; break; - case DCCP_SOCKOPT_CHANGE_L: if (optlen != sizeof(struct dccp_so_feat)) err = -EINVAL; @@ -481,7 +480,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, (struct dccp_so_feat __user *) optval); break; - case DCCP_SOCKOPT_CHANGE_R: if (optlen != sizeof(struct dccp_so_feat)) err = -EINVAL; @@ -490,12 +488,26 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, (struct dccp_so_feat __user *) optval); break; - + case DCCP_SOCKOPT_SEND_CSCOV: /* sender side, RFC 4340, sec. 9.2 */ + if (val < 0 || val > 15) + err = -EINVAL; + else + dp->dccps_pcslen = val; + break; + case DCCP_SOCKOPT_RECV_CSCOV: /* receiver side, RFC 4340 sec. 9.2.1 */ + if (val < 0 || val > 15) + err = -EINVAL; + else { + dp->dccps_pcrlen = val; + /* FIXME: add feature negotiation, + * ChangeL(MinimumChecksumCoverage, val) */ + } + break; default: err = -ENOPROTOOPT; break; } - + release_sock(sk); return err; } @@ -575,6 +587,12 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname, case DCCP_SOCKOPT_SERVICE: return dccp_getsockopt_service(sk, len, (__be32 __user *)optval, optlen); + case DCCP_SOCKOPT_SEND_CSCOV: + val = dp->dccps_pcslen; + break; + case DCCP_SOCKOPT_RECV_CSCOV: + val = dp->dccps_pcrlen; + break; case 128 ... 191: return ccid_hc_rx_getsockopt(dp->dccps_hc_rx_ccid, sk, optname, len, (u32 __user *)optval, optlen); -- cgit v1.2.3 From 339bf98ffc6a8d8eb16fc532ac57ffbced2f8a68 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Fri, 10 Nov 2006 14:10:15 -0800 Subject: [NETLINK]: Do precise netlink message allocations where possible Account for the netlink message header size directly in nlmsg_new() instead of relying on the caller calculate it correctly. Replaces error handling of message construction functions when constructing notifications with bug traps since a failure implies a bug in calculating the size of the skb. Signed-off-by: Thomas Graf Acked-by: Paul Moore Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + include/net/fib_rules.h | 1 + include/net/netlink.h | 9 ++--- kernel/taskstats.c | 3 +- net/bridge/br_netlink.c | 21 ++++++++---- net/core/fib_rules.c | 24 +++++++++++--- net/core/neighbour.c | 17 +++++++--- net/core/rtnetlink.c | 39 ++++++++++++++-------- net/decnet/dn_rules.c | 6 ++++ net/decnet/dn_table.c | 34 ++++++++++++++++--- net/ipv4/devinet.c | 18 +++++++--- net/ipv4/fib_rules.c | 8 +++++ net/ipv4/fib_semantics.c | 36 ++++++++++++++++---- net/ipv6/addrconf.c | 70 ++++++++++++++++++--------------------- net/ipv6/fib6_rules.c | 7 ++++ net/ipv6/route.c | 23 +++++++++---- net/netlabel/netlabel_cipso_v4.c | 2 +- net/netlabel/netlabel_mgmt.c | 4 +-- net/netlabel/netlabel_unlabeled.c | 2 +- net/netlink/af_netlink.c | 13 ++++---- net/netlink/genetlink.c | 2 +- 21 files changed, 233 insertions(+), 107 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 66411622e06e..e61e1e138421 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -174,6 +174,7 @@ int netlink_sendskb(struct sock *sk, struct sk_buff *skb, int protocol); */ #define NLMSG_GOODORDER 0 #define NLMSG_GOODSIZE (SKB_MAX_ORDER(0, NLMSG_GOODORDER)) +#define NLMSG_DEFAULT_SIZE (NLMSG_GOODSIZE - NLMSG_HDRLEN) struct netlink_callback diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index e4ba781d289f..bc3c26494c3d 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -52,6 +52,7 @@ struct fib_rules_ops struct nlmsghdr *, struct fib_rule_hdr *); u32 (*default_pref)(void); + size_t (*nlmsg_payload)(struct fib_rule *); int nlgroup; struct nla_policy *policy; diff --git a/include/net/netlink.h b/include/net/netlink.h index ce5cba19c393..30021339157c 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -500,14 +500,15 @@ static inline struct nlmsghdr *nlmsg_put_answer(struct sk_buff *skb, /** * nlmsg_new - Allocate a new netlink message - * @size: maximum size of message + * @payload: size of the message payload * @flags: the type of memory to allocate. * - * Use NLMSG_GOODSIZE if size isn't know and you need a good default size. + * Use NLMSG_DEFAULT_SIZE if the size of the payload isn't known + * and a good default is needed. */ -static inline struct sk_buff *nlmsg_new(int size, gfp_t flags) +static inline struct sk_buff *nlmsg_new(size_t payload, gfp_t flags) { - return alloc_skb(size, flags); + return alloc_skb(nlmsg_total_size(payload), flags); } /** diff --git a/kernel/taskstats.c b/kernel/taskstats.c index f45c5e70773c..4f3f0e48c845 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -77,8 +77,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * If new attributes are added, please revisit this allocation */ - size = nlmsg_total_size(genlmsg_total_size(size)); - skb = nlmsg_new(size, GFP_KERNEL); + skb = nlmsg_new(genlmsg_total_size(size), GFP_KERNEL); if (!skb) return -ENOMEM; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 8f661195d09d..15d6efbe7519 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -15,6 +15,18 @@ #include #include "br_private.h" +static inline size_t br_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(4) /* IFLA_MASTER */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size(1) /* IFLA_OPERSTATE */ + + nla_total_size(1); /* IFLA_PROTINFO */ +} + /* * Create one netlink message for one interface * Contains port and master info as well as carrier and bridge state. @@ -77,19 +89,16 @@ rtattr_failure: void br_ifinfo_notify(int event, struct net_bridge_port *port) { struct sk_buff *skb; - int payload = sizeof(struct ifinfomsg) + 128; int err = -ENOBUFS; pr_debug("bridge notify event=%d\n", event); - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(br_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = br_fill_ifinfo(skb, port, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in br_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_ATOMIC); errout: diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 4148e274a204..1df6cd4568d3 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -306,6 +306,22 @@ errout: return err; } +static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops, + struct fib_rule *rule) +{ + size_t payload = NLMSG_ALIGN(sizeof(struct fib_rule_hdr)) + + nla_total_size(IFNAMSIZ) /* FRA_IFNAME */ + + nla_total_size(4) /* FRA_PRIORITY */ + + nla_total_size(4) /* FRA_TABLE */ + + nla_total_size(4) /* FRA_FWMARK */ + + nla_total_size(4); /* FRA_FWMASK */ + + if (ops->nlmsg_payload) + payload += ops->nlmsg_payload(rule); + + return payload; +} + static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, u32 pid, u32 seq, int type, int flags, struct fib_rules_ops *ops) @@ -384,15 +400,13 @@ static void notify_rule_change(int event, struct fib_rule *rule, struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(fib_rule_nlmsg_size(ops, rule), GFP_KERNEL); if (skb == NULL) goto errout; err = fib_nl_fill_rule(skb, rule, pid, nlh->nlmsg_seq, event, 0, ops); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in fib_rule_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, ops->nlgroup, nlh, GFP_KERNEL); errout: diff --git a/net/core/neighbour.c b/net/core/neighbour.c index b4b478353b27..0e097ba14d73 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2410,20 +2410,27 @@ static struct file_operations neigh_stat_seq_fops = { #endif /* CONFIG_PROC_FS */ #ifdef CONFIG_ARPD +static inline size_t neigh_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ndmsg)) + + nla_total_size(MAX_ADDR_LEN) /* NDA_DST */ + + nla_total_size(MAX_ADDR_LEN) /* NDA_LLADDR */ + + nla_total_size(sizeof(struct nda_cacheinfo)) + + nla_total_size(4); /* NDA_PROBES */ +} + static void __neigh_notify(struct neighbour *n, int type, int flags) { struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_ATOMIC); + skb = nlmsg_new(neigh_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = neigh_fill_info(skb, n, 0, 0, type, flags); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in neigh_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_NEIGH, NULL, GFP_ATOMIC); errout: diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 02f3c7947898..50d6cb40c6e3 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -273,6 +273,25 @@ static void copy_rtnl_link_stats(struct rtnl_link_stats *a, a->tx_compressed = b->tx_compressed; }; +static inline size_t if_nlmsg_size(int iwbuflen) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(IFNAMSIZ) /* IFLA_QDISC */ + + nla_total_size(sizeof(struct rtnl_link_ifmap)) + + nla_total_size(sizeof(struct rtnl_link_stats)) + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_BROADCAST */ + + nla_total_size(4) /* IFLA_TXQLEN */ + + nla_total_size(4) /* IFLA_WEIGHT */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size(4) /* IFLA_MASTER */ + + nla_total_size(1) /* IFLA_OPERSTATE */ + + nla_total_size(1) /* IFLA_LINKMODE */ + + nla_total_size(iwbuflen); +} + static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, void *iwbuf, int iwbuflen, int type, u32 pid, u32 seq, u32 change, unsigned int flags) @@ -558,7 +577,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) struct sk_buff *nskb; char *iw_buf = NULL, *iw = NULL; int iw_buf_len = 0; - int err, payload; + int err; err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy); if (err < 0) @@ -587,9 +606,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) } #endif /* CONFIG_NET_WIRELESS_RTNETLINK */ - payload = NLMSG_ALIGN(sizeof(struct ifinfomsg) + - nla_total_size(iw_buf_len)); - nskb = nlmsg_new(nlmsg_total_size(payload), GFP_KERNEL); + nskb = nlmsg_new(if_nlmsg_size(iw_buf_len), GFP_KERNEL); if (nskb == NULL) { err = -ENOBUFS; goto errout; @@ -597,10 +614,8 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) err = rtnl_fill_ifinfo(nskb, dev, iw, iw_buf_len, RTM_NEWLINK, NETLINK_CB(skb).pid, nlh->nlmsg_seq, 0, 0); - if (err <= 0) { - kfree_skb(nskb); - goto errout; - } + /* failure impilies BUG in if_nlmsg_size or wireless_rtnetlink_get */ + BUG_ON(err < 0); err = rtnl_unicast(nskb, NETLINK_CB(skb).pid); errout: @@ -639,15 +654,13 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(if_nlmsg_size(0), GFP_KERNEL); if (skb == NULL) goto errout; err = rtnl_fill_ifinfo(skb, dev, NULL, 0, type, 0, 0, change, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in if_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_LINK, NULL, GFP_KERNEL); errout: diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index e32d0c3d5a96..b7dfd04a9638 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -241,6 +241,12 @@ static u32 dn_fib_rule_default_pref(void) return 0; } +static size_t dn_fib_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(2) /* dst */ + + nla_total_size(2); /* src */ +} + int dn_fib_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) { return fib_rules_dump(skb, cb, AF_DECnet); diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index 317904bb5896..e74b744254ab 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -263,6 +263,32 @@ static int dn_fib_nh_match(struct rtmsg *r, struct nlmsghdr *nlh, struct dn_kern return 0; } +static inline size_t dn_fib_nlmsg_size(struct dn_fib_info *fi) +{ + size_t payload = NLMSG_ALIGN(struct rtmsg) + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(2) /* RTA_DST */ + + nla_total_size(4); /* RTA_PRIORITY */ + + /* space for nested metrics */ + payload += nla_total_size((RTAX_MAX * nla_total_size(4))); + + if (fi->fib_nhs) { + /* Also handles the special case fib_nhs == 1 */ + + /* each nexthop is packed in an attribute */ + size_t nhsize = nla_total_size(sizeof(struct rtnexthop)); + + /* may contain a gateway attribute */ + nhsize += nla_total_size(4); + + /* all nexthops are packed in a nested attribute */ + payload += nla_total_size(fi->fib_nhs * nhsize); + } + + return payload; +} + static int dn_fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event, u32 tb_id, u8 type, u8 scope, void *dst, int dst_len, struct dn_fib_info *fi, unsigned int flags) @@ -335,17 +361,15 @@ static void dn_rtmsg_fib(int event, struct dn_fib_node *f, int z, u32 tb_id, u32 pid = req ? req->pid : 0; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(dn_fib_nlmsg_size(DN_FIB_INFO(f), GFP_KERNEL)); if (skb == NULL) goto errout; err = dn_fib_dump_info(skb, pid, nlh->nlmsg_seq, event, tb_id, f->fn_type, f->fn_scope, &f->fn_key, z, DN_FIB_INFO(f), 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in dn_fib_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_DECnet_ROUTE, nlh, GFP_KERNEL); errout: diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 7602c79a389b..f38cbbae0ae3 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1120,6 +1120,16 @@ static struct notifier_block ip_netdev_notifier = { .notifier_call =inetdev_event, }; +static inline size_t inet_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifaddrmsg)) + + nla_total_size(4) /* IFA_ADDRESS */ + + nla_total_size(4) /* IFA_LOCAL */ + + nla_total_size(4) /* IFA_BROADCAST */ + + nla_total_size(4) /* IFA_ANYCAST */ + + nla_total_size(IFNAMSIZ); /* IFA_LABEL */ +} + static int inet_fill_ifaddr(struct sk_buff *skb, struct in_ifaddr *ifa, u32 pid, u32 seq, int event, unsigned int flags) { @@ -1208,15 +1218,13 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa, struct nlmsghdr *nlh, u32 seq = nlh ? nlh->nlmsg_seq : 0; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(inet_nlmsg_size(), GFP_KERNEL); if (skb == NULL) goto errout; err = inet_fill_ifaddr(skb, ifa, pid, seq, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_IPV4_IFADDR, nlh, GFP_KERNEL); errout: diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index fd4a8cd4c06e..b837c33e0404 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -299,6 +299,13 @@ static u32 fib4_rule_default_pref(void) return 0; } +static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(4) /* dst */ + + nla_total_size(4) /* src */ + + nla_total_size(4); /* flow */ +} + static struct fib_rules_ops fib4_rules_ops = { .family = AF_INET, .rule_size = sizeof(struct fib4_rule), @@ -308,6 +315,7 @@ static struct fib_rules_ops fib4_rules_ops = { .compare = fib4_rule_compare, .fill = fib4_rule_fill, .default_pref = fib4_rule_default_pref, + .nlmsg_payload = fib4_rule_nlmsg_payload, .nlgroup = RTNLGRP_IPV4_RULE, .policy = fib4_rule_policy, .rules_list = &fib4_rules, diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 884d176e0082..e63b8a98fb4d 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -273,25 +273,49 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev) return -1; } +static inline size_t fib_nlmsg_size(struct fib_info *fi) +{ + size_t payload = NLMSG_ALIGN(sizeof(struct rtmsg)) + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(4) /* RTA_DST */ + + nla_total_size(4) /* RTA_PRIORITY */ + + nla_total_size(4); /* RTA_PREFSRC */ + + /* space for nested metrics */ + payload += nla_total_size((RTAX_MAX * nla_total_size(4))); + + if (fi->fib_nhs) { + /* Also handles the special case fib_nhs == 1 */ + + /* each nexthop is packed in an attribute */ + size_t nhsize = nla_total_size(sizeof(struct rtnexthop)); + + /* may contain flow and gateway attribute */ + nhsize += 2 * nla_total_size(4); + + /* all nexthops are packed in a nested attribute */ + payload += nla_total_size(fi->fib_nhs * nhsize); + } + + return payload; +} + void rtmsg_fib(int event, __be32 key, struct fib_alias *fa, int dst_len, u32 tb_id, struct nl_info *info) { struct sk_buff *skb; - int payload = sizeof(struct rtmsg) + 256; u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_KERNEL); + skb = nlmsg_new(fib_nlmsg_size(fa->fa_info), GFP_KERNEL); if (skb == NULL) goto errout; err = fib_dump_info(skb, info->pid, seq, event, tb_id, fa->fa_type, fa->fa_scope, key, dst_len, fa->fa_tos, fa->fa_info, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in fib_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, info->pid, RTNLGRP_IPV4_ROUTE, info->nlh, GFP_KERNEL); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6a98f68348cb..967ea320a9ca 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3098,10 +3098,9 @@ static inline int rt_scope(int ifa_scope) static inline int inet6_ifaddr_msgsize(void) { - return nlmsg_total_size(sizeof(struct ifaddrmsg) + - nla_total_size(16) + - nla_total_size(sizeof(struct ifa_cacheinfo)) + - 128); + return NLMSG_ALIGN(sizeof(struct ifaddrmsg)) + + nla_total_size(16) /* IFA_ADDRESS */ + + nla_total_size(sizeof(struct ifa_cacheinfo)); } static int inet6_fill_ifaddr(struct sk_buff *skb, struct inet6_ifaddr *ifa, @@ -3329,10 +3328,8 @@ static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh, err = inet6_fill_ifaddr(skb, ifa, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWADDR, 0); - if (err < 0) { - kfree_skb(skb); - goto errout_ifa; - } + /* failure implies BUG in inet6_ifaddr_msgsize() */ + BUG_ON(err < 0); err = rtnl_unicast(skb, NETLINK_CB(in_skb).pid); errout_ifa: @@ -3351,10 +3348,8 @@ static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa) goto errout; err = inet6_fill_ifaddr(skb, ifa, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_ifaddr_msgsize() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC); errout: @@ -3397,16 +3392,19 @@ static void inline ipv6_store_devconf(struct ipv6_devconf *cnf, array[DEVCONF_PROXY_NDP] = cnf->proxy_ndp; } -/* Maximum length of ifinfomsg attributes */ -#define INET6_IFINFO_RTA_SPACE \ - RTA_SPACE(IFNAMSIZ) /* IFNAME */ + \ - RTA_SPACE(MAX_ADDR_LEN) /* ADDRESS */ + \ - RTA_SPACE(sizeof(u32)) /* MTU */ + \ - RTA_SPACE(sizeof(int)) /* LINK */ + \ - RTA_SPACE(0) /* PROTINFO */ + \ - RTA_SPACE(sizeof(u32)) /* FLAGS */ + \ - RTA_SPACE(sizeof(struct ifla_cacheinfo)) /* CACHEINFO */ + \ - RTA_SPACE(sizeof(__s32[DEVCONF_MAX])) /* CONF */ +static inline size_t inet6_if_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct ifinfomsg)) + + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + + nla_total_size(MAX_ADDR_LEN) /* IFLA_ADDRESS */ + + nla_total_size(4) /* IFLA_MTU */ + + nla_total_size(4) /* IFLA_LINK */ + + nla_total_size( /* IFLA_PROTINFO */ + nla_total_size(4) /* IFLA_INET6_FLAGS */ + + nla_total_size(sizeof(struct ifla_cacheinfo)) + + nla_total_size(DEVCONF_MAX * 4) /* IFLA_INET6_CONF */ + ); +} static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, u32 pid, u32 seq, int event, unsigned int flags) @@ -3501,18 +3499,15 @@ static int inet6_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) void inet6_ifinfo_notify(int event, struct inet6_dev *idev) { struct sk_buff *skb; - int payload = sizeof(struct ifinfomsg) + INET6_IFINFO_RTA_SPACE; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(inet6_if_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = inet6_fill_ifinfo(skb, idev, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_if_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_IFADDR, NULL, GFP_ATOMIC); errout: @@ -3520,10 +3515,12 @@ errout: rtnl_set_sk_err(RTNLGRP_IPV6_IFADDR, err); } -/* Maximum length of prefix_cacheinfo attributes */ -#define INET6_PREFIX_RTA_SPACE \ - RTA_SPACE(sizeof(((struct prefix_info *)NULL)->prefix)) /* ADDRESS */ + \ - RTA_SPACE(sizeof(struct prefix_cacheinfo)) /* CACHEINFO */ +static inline size_t inet6_prefix_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct prefixmsg)) + + nla_total_size(sizeof(struct in6_addr)) + + nla_total_size(sizeof(struct prefix_cacheinfo)); +} static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, struct prefix_info *pinfo, u32 pid, u32 seq, @@ -3569,18 +3566,15 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, struct prefix_info *pinfo) { struct sk_buff *skb; - int payload = sizeof(struct prefixmsg) + INET6_PREFIX_RTA_SPACE; int err = -ENOBUFS; - skb = nlmsg_new(nlmsg_total_size(payload), GFP_ATOMIC); + skb = nlmsg_new(inet6_prefix_nlmsg_size(), GFP_ATOMIC); if (skb == NULL) goto errout; err = inet6_fill_prefix(skb, idev, pinfo, 0, 0, event, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in inet6_prefix_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, 0, RTNLGRP_IPV6_PREFIX, NULL, GFP_ATOMIC); errout: diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 25804cb69cf0..d587dde5897e 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -232,6 +232,12 @@ static u32 fib6_rule_default_pref(void) return 0x3FFF; } +static size_t fib6_rule_nlmsg_payload(struct fib_rule *rule) +{ + return nla_total_size(16) /* dst */ + + nla_total_size(16); /* src */ +} + static struct fib_rules_ops fib6_rules_ops = { .family = AF_INET6, .rule_size = sizeof(struct fib6_rule), @@ -241,6 +247,7 @@ static struct fib_rules_ops fib6_rules_ops = { .compare = fib6_rule_compare, .fill = fib6_rule_fill, .default_pref = fib6_rule_default_pref, + .nlmsg_payload = fib6_rule_nlmsg_payload, .nlgroup = RTNLGRP_IPV6_RULE, .policy = fib6_rule_policy, .rules_list = &fib6_rules, diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0ad07c9087a7..a6472cb9054c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2006,6 +2006,20 @@ int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) return ip6_route_add(&cfg); } +static inline size_t rt6_nlmsg_size(void) +{ + return NLMSG_ALIGN(sizeof(struct rtmsg)) + + nla_total_size(16) /* RTA_SRC */ + + nla_total_size(16) /* RTA_DST */ + + nla_total_size(16) /* RTA_GATEWAY */ + + nla_total_size(16) /* RTA_PREFSRC */ + + nla_total_size(4) /* RTA_TABLE */ + + nla_total_size(4) /* RTA_IIF */ + + nla_total_size(4) /* RTA_OIF */ + + nla_total_size(4) /* RTA_PRIORITY */ + + nla_total_size(sizeof(struct rta_cacheinfo)); +} + static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 pid, u32 seq, @@ -2200,7 +2214,6 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) struct sk_buff *skb; u32 pid = 0, seq = 0; struct nlmsghdr *nlh = NULL; - int payload = sizeof(struct rtmsg) + 256; int err = -ENOBUFS; if (info) { @@ -2210,15 +2223,13 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) seq = nlh->nlmsg_seq; } - skb = nlmsg_new(nlmsg_total_size(payload), gfp_any()); + skb = nlmsg_new(rt6_nlmsg_size(), gfp_any()); if (skb == NULL) goto errout; err = rt6_fill_node(skb, rt, NULL, NULL, 0, event, pid, seq, 0, 0); - if (err < 0) { - kfree_skb(skb); - goto errout; - } + /* failure implies BUG in rt6_nlmsg_size() */ + BUG_ON(err < 0); err = rtnl_notify(skb, pid, RTNLGRP_IPV6_ROUTE, nlh, gfp_any()); errout: diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index a6ce1d6d5c59..f1788bd290f8 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -452,7 +452,7 @@ static int netlbl_cipsov4_list(struct sk_buff *skb, struct genl_info *info) } list_start: - ans_skb = nlmsg_new(NLMSG_GOODSIZE * nlsze_mult, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE * nlsze_mult, GFP_KERNEL); if (ans_skb == NULL) { ret_val = -ENOMEM; goto list_failure; diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index 53c9079ad2c3..c529622ff0b7 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -356,7 +356,7 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info) void *data; struct netlbl_dom_map *entry; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; data = netlbl_netlink_hdr_put(ans_skb, @@ -492,7 +492,7 @@ static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info) struct sk_buff *ans_skb = NULL; void *data; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) return -ENOMEM; data = netlbl_netlink_hdr_put(ans_skb, diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 1833ad233b39..219dccade4e1 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -138,7 +138,7 @@ static int netlbl_unlabel_list(struct sk_buff *skb, struct genl_info *info) struct sk_buff *ans_skb; void *data; - ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + ans_skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (ans_skb == NULL) goto list_failure; data = netlbl_netlink_hdr_put(ans_skb, diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index d527c8977b1f..f61d81b3c61c 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1148,7 +1148,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, if (len > sk->sk_sndbuf - 32) goto out; err = -ENOBUFS; - skb = nlmsg_new(len, GFP_KERNEL); + skb = alloc_skb(len, GFP_KERNEL); if (skb==NULL) goto out; @@ -1435,14 +1435,13 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) struct sk_buff *skb; struct nlmsghdr *rep; struct nlmsgerr *errmsg; - int size; + size_t payload = sizeof(*errmsg); - if (err == 0) - size = nlmsg_total_size(sizeof(*errmsg)); - else - size = nlmsg_total_size(sizeof(*errmsg) + nlmsg_len(nlh)); + /* error messages get the original request appened */ + if (err) + payload += nlmsg_len(nlh); - skb = nlmsg_new(size, GFP_KERNEL); + skb = nlmsg_new(payload, GFP_KERNEL); if (!skb) { struct sock *sk; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 49bc2db7982b..70d60c818897 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -480,7 +480,7 @@ static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, struct sk_buff *skb; int err; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (skb == NULL) return ERR_PTR(-ENOBUFS); -- cgit v1.2.3 From 2e2e9e92bd723244ea20fa488b1780111f2b05e1 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 13 Nov 2006 13:23:52 -0200 Subject: [DCCP]: Add sysctls to control retransmission behaviour This adds 3 sysctls which govern the retransmission behaviour of DCCP control packets (3way handshake, feature negotiation). It removes 4 FIXMEs from the code. The close resemblance of sysctl variables to their TCP analogues is emphasised not only by their name, but also by giving them the same initial values. This is useful since there is not much practical experience with DCCP yet. Furthermore, with regard to the previous patch, it is now possible to limit the number of keepalive-Responses by setting net.dccp.default.request_retries (also a bit like in TCP). Lastly, added documentation of all existing DCCP sysctls. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 41 +++++++++++++++++++++++++++++++++++++++ include/linux/sysctl.h | 3 +++ net/dccp/dccp.h | 11 +++++++++++ net/dccp/feat.h | 7 ------- net/dccp/proto.c | 1 + net/dccp/sysctl.c | 25 ++++++++++++++++++++++++ net/dccp/timer.c | 16 +++++++++------ 7 files changed, 91 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index a8142a81038a..c2328c862e98 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -63,6 +63,47 @@ DCCP_SOCKOPT_SEND_CSCOV is for the receiver and has a different meaning: it coverage value are also acceptable. The higher the number, the more restrictive this setting (see [RFC 4340, sec. 9.2.1]). +Sysctl variables +================ +Several DCCP default parameters can be managed by the following sysctls +(sysctl net.dccp.default or /proc/sys/net/dccp/default): + +request_retries + The number of active connection initiation retries (the number of + Requests minus one) before timing out. In addition, it also governs + the behaviour of the other, passive side: this variable also sets + the number of times DCCP repeats sending a Response when the initial + handshake does not progress from RESPOND to OPEN (i.e. when no Ack + is received after the initial Request). This value should be greater + than 0, suggested is less than 10. Analogue of tcp_syn_retries. + +retries1 + How often a DCCP Response is retransmitted until the listening DCCP + side considers its connecting peer dead. Analogue of tcp_retries1. + +retries2 + The number of times a general DCCP packet is retransmitted. This has + importance for retransmitted acknowledgments and feature negotiation, + data packets are never retransmitted. Analogue of tcp_retries2. + +send_ndp = 1 + Whether or not to send NDP count options (sec. 7.7.2). + +send_ackvec = 1 + Whether or not to send Ack Vector options (sec. 11.5). + +ack_ratio = 2 + The default Ack Ratio (sec. 11.3) to use. + +tx_ccid = 2 + Default CCID for the sender-receiver half-connection. + +rx_ccid = 2 + Default CCID for the receiver-sender half-connection. + +seq_window = 100 + The initial sequence window (sec. 7.5.2). + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 0725441621d0..2e8c5ad82793 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -614,6 +614,9 @@ enum { NET_DCCP_DEFAULT_ACK_RATIO = 4, NET_DCCP_DEFAULT_SEND_ACKVEC = 5, NET_DCCP_DEFAULT_SEND_NDP = 6, + NET_DCCP_DEFAULT_REQ_RETRIES = 7, + NET_DCCP_DEFAULT_RETRIES1 = 8, + NET_DCCP_DEFAULT_RETRIES2 = 9, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h index d5c414bf7819..e7b96074a1b1 100644 --- a/net/dccp/dccp.h +++ b/net/dccp/dccp.h @@ -64,6 +64,17 @@ extern void dccp_time_wait(struct sock *sk, int state, int timeo); #define DCCP_XMIT_TIMEO 30000 /* Time/msecs for blocking transmit per packet */ +/* sysctl variables for DCCP */ +extern int sysctl_dccp_request_retries; +extern int sysctl_dccp_retries1; +extern int sysctl_dccp_retries2; +extern int dccp_feat_default_sequence_window; +extern int dccp_feat_default_rx_ccid; +extern int dccp_feat_default_tx_ccid; +extern int dccp_feat_default_ack_ratio; +extern int dccp_feat_default_send_ack_vector; +extern int dccp_feat_default_send_ndp_count; + /* is seq1 < seq2 ? */ static inline int before48(const u64 seq1, const u64 seq2) { diff --git a/net/dccp/feat.h b/net/dccp/feat.h index cee553d416ca..6048373c7186 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -26,11 +26,4 @@ extern void dccp_feat_clean(struct dccp_minisock *dmsk); extern int dccp_feat_clone(struct sock *oldsk, struct sock *newsk); extern int dccp_feat_init(struct dccp_minisock *dmsk); -extern int dccp_feat_default_sequence_window; -extern int dccp_feat_default_rx_ccid; -extern int dccp_feat_default_tx_ccid; -extern int dccp_feat_default_ack_ratio; -extern int dccp_feat_default_send_ack_vector; -extern int dccp_feat_default_send_ndp_count; - #endif /* _DCCP_FEAT_H */ diff --git a/net/dccp/proto.c b/net/dccp/proto.c index db54e557eff1..9c9c08cffdaf 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -212,6 +212,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) dccp_init_xmit_timers(sk); icsk->icsk_rto = DCCP_TIMEOUT_INIT; + icsk->icsk_syn_retries = sysctl_dccp_request_retries; sk->sk_state = DCCP_CLOSED; sk->sk_write_space = dccp_write_space; icsk->icsk_sync_mss = dccp_sync_mss; diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 38bc157876f3..7b09f2179985 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -11,6 +11,7 @@ #include #include +#include "dccp.h" #include "feat.h" #ifndef CONFIG_SYSCTL @@ -66,6 +67,30 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, + .procname = "request_retries", + .data = &sysctl_dccp_request_retries, + .maxlen = sizeof(sysctl_dccp_request_retries), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES1, + .procname = "retries1", + .data = &sysctl_dccp_retries1, + .maxlen = sizeof(sysctl_dccp_retries1), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = NET_DCCP_DEFAULT_RETRIES2, + .procname = "retries2", + .data = &sysctl_dccp_retries2, + .maxlen = sizeof(sysctl_dccp_retries2), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { .ctl_name = 0, } }; diff --git a/net/dccp/timer.c b/net/dccp/timer.c index bda0af639ae4..7b3f16e29a97 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -15,6 +15,11 @@ #include "dccp.h" +/* sysctl variables governing numbers of retransmission attempts */ +int sysctl_dccp_request_retries __read_mostly = TCP_SYN_RETRIES; +int sysctl_dccp_retries1 __read_mostly = TCP_RETR1; +int sysctl_dccp_retries2 __read_mostly = TCP_RETR2; + static void dccp_write_timer(unsigned long data); static void dccp_keepalive_timer(unsigned long data); static void dccp_delack_timer(unsigned long data); @@ -44,11 +49,10 @@ static int dccp_write_timeout(struct sock *sk) if (sk->sk_state == DCCP_REQUESTING || sk->sk_state == DCCP_PARTOPEN) { if (icsk->icsk_retransmits != 0) dst_negative_advice(&sk->sk_dst_cache); - retry_until = icsk->icsk_syn_retries ? : - /* FIXME! */ 3 /* FIXME! sysctl_tcp_syn_retries */; + retry_until = icsk->icsk_syn_retries ? + : sysctl_dccp_request_retries; } else { - if (icsk->icsk_retransmits >= - /* FIXME! sysctl_tcp_retries1 */ 5 /* FIXME! */) { + if (icsk->icsk_retransmits >= sysctl_dccp_retries1) { /* NOTE. draft-ietf-tcpimpl-pmtud-01.txt requires pmtu black hole detection. :-( @@ -72,7 +76,7 @@ static int dccp_write_timeout(struct sock *sk) dst_negative_advice(&sk->sk_dst_cache); } - retry_until = /* FIXME! */ 15 /* FIXME! sysctl_tcp_retries2 */; + retry_until = sysctl_dccp_retries2; /* * FIXME: see tcp_write_timout and tcp_out_of_resources */ @@ -196,7 +200,7 @@ backoff: icsk->icsk_rto = min(icsk->icsk_rto << 1, DCCP_RTO_MAX); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, DCCP_RTO_MAX); - if (icsk->icsk_retransmits > 3 /* FIXME: sysctl_dccp_retries1 */) + if (icsk->icsk_retransmits > sysctl_dccp_retries1) __sk_dst_reset(sk); out:; } -- cgit v1.2.3 From 90833aa4f496d69ca374af6acef7d1614c8693ff Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 13 Nov 2006 16:02:22 -0800 Subject: [NET]: The scheduled removal of the frame diverter. This patch contains the scheduled removal of the frame diverter. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- Documentation/feature-removal-schedule.txt | 15 - drivers/net/Space.c | 1 - include/linux/Kbuild | 1 - include/linux/divert.h | 132 ------- include/linux/netdevice.h | 6 - include/linux/sockios.h | 4 +- net/Kconfig | 27 -- net/core/Makefile | 1 - net/core/dev.c | 20 +- net/core/dv.c | 546 ----------------------------- net/core/sysctl_net_core.c | 14 - net/socket.c | 6 - 12 files changed, 6 insertions(+), 767 deletions(-) delete mode 100644 include/linux/divert.h delete mode 100644 net/core/dv.c (limited to 'include/linux') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index d52c4aaaf17f..b3949cd3d013 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -227,21 +227,6 @@ Who: Patrick McHardy --------------------------- -What: frame diverter -When: November 2006 -Why: The frame diverter is included in most distribution kernels, but is - broken. It does not correctly handle many things: - - IPV6 - - non-linear skb's - - network device RCU on removal - - input frames not correctly checked for protocol errors - It also adds allocation overhead even if not enabled. - It is not clear if anyone is still using it. -Who: Stephen Hemminger - ---------------------------- - - What: PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment When: October 2008 Why: The stacking of class devices makes these values misleading and diff --git a/drivers/net/Space.c b/drivers/net/Space.c index a67f5efc983f..602ed31a5dd9 100644 --- a/drivers/net/Space.c +++ b/drivers/net/Space.c @@ -33,7 +33,6 @@ #include #include #include -#include /* A unified ethernet device probe. This is the easiest way to have every ethernet adaptor have the name "eth[0123...]". diff --git a/include/linux/Kbuild b/include/linux/Kbuild index a1155a2beb32..d7e04689304c 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -193,7 +193,6 @@ unifdef-y += cuda.h unifdef-y += cyclades.h unifdef-y += dccp.h unifdef-y += dirent.h -unifdef-y += divert.h unifdef-y += dlm.h unifdef-y += elfcore.h unifdef-y += errno.h diff --git a/include/linux/divert.h b/include/linux/divert.h deleted file mode 100644 index 8fb4e9de6843..000000000000 --- a/include/linux/divert.h +++ /dev/null @@ -1,132 +0,0 @@ -/* - * Frame Diversion, Benoit Locher - * - * Changes: - * 06/09/2000 BL: initial version - * - */ - -#ifndef _LINUX_DIVERT_H -#define _LINUX_DIVERT_H - -#include - -#define MAX_DIVERT_PORTS 8 /* Max number of ports to divert (tcp, udp) */ - -/* Divertable protocols */ -#define DIVERT_PROTO_NONE 0x0000 -#define DIVERT_PROTO_IP 0x0001 -#define DIVERT_PROTO_ICMP 0x0002 -#define DIVERT_PROTO_TCP 0x0004 -#define DIVERT_PROTO_UDP 0x0008 - -/* - * This is an Ethernet Frame Diverter option block - */ -struct divert_blk -{ - int divert; /* are we active */ - unsigned int protos; /* protocols */ - __u16 tcp_dst[MAX_DIVERT_PORTS]; /* specific tcp dst ports to divert */ - __u16 tcp_src[MAX_DIVERT_PORTS]; /* specific tcp src ports to divert */ - __u16 udp_dst[MAX_DIVERT_PORTS]; /* specific udp dst ports to divert */ - __u16 udp_src[MAX_DIVERT_PORTS]; /* specific udp src ports to divert */ -}; - -/* - * Diversion control block, for configuration with the userspace tool - * divert - */ - -typedef union _divert_cf_arg -{ - __s16 int16; - __u16 uint16; - __s32 int32; - __u32 uint32; - __s64 int64; - __u64 uint64; - void __user *ptr; -} divert_cf_arg; - - -struct divert_cf -{ - int cmd; /* Command */ - divert_cf_arg arg1, - arg2, - arg3; - int dev_index; /* device index (eth0=0, etc...) */ -}; - - -/* Diversion commands */ -#define DIVCMD_DIVERT 1 /* ENABLE/DISABLE diversion */ -#define DIVCMD_IP 2 /* ENABLE/DISABLE whold IP diversion */ -#define DIVCMD_TCP 3 /* ENABLE/DISABLE whold TCP diversion */ -#define DIVCMD_TCPDST 4 /* ADD/REMOVE TCP DST port for diversion */ -#define DIVCMD_TCPSRC 5 /* ADD/REMOVE TCP SRC port for diversion */ -#define DIVCMD_UDP 6 /* ENABLE/DISABLE whole UDP diversion */ -#define DIVCMD_UDPDST 7 /* ADD/REMOVE UDP DST port for diversion */ -#define DIVCMD_UDPSRC 8 /* ADD/REMOVE UDP SRC port for diversion */ -#define DIVCMD_ICMP 9 /* ENABLE/DISABLE whole ICMP diversion */ -#define DIVCMD_GETSTATUS 10 /* GET the status of the diverter */ -#define DIVCMD_RESET 11 /* Reset the diverter on the specified dev */ -#define DIVCMD_GETVERSION 12 /* Retrieve the diverter code version (char[32]) */ - -/* General syntax of the commands: - * - * DIVCMD_xxxxxx(arg1, arg2, arg3, dev_index) - * - * SIOCSIFDIVERT: - * DIVCMD_DIVERT(DIVARG1_ENABLE|DIVARG1_DISABLE, , ,ifindex) - * DIVCMD_IP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_TCP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_TCPDST(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_TCPSRC(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_UDP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_UDPDST(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_UDPSRC(DIVARG1_ADD|DIVARG1_REMOVE, port, , ifindex) - * DIVCMD_ICMP(DIVARG1_ENABLE|DIVARG1_DISABLE, , , ifindex) - * DIVCMD_RESET(, , , ifindex) - * - * SIOGIFDIVERT: - * DIVCMD_GETSTATUS(divert_blk, , , ifindex) - * DIVCMD_GETVERSION(string[3]) - */ - - -/* Possible values for arg1 */ -#define DIVARG1_ENABLE 0 /* ENABLE something */ -#define DIVARG1_DISABLE 1 /* DISABLE something */ -#define DIVARG1_ADD 2 /* ADD something */ -#define DIVARG1_REMOVE 3 /* REMOVE something */ - - -#ifdef __KERNEL__ - -/* diverter functions */ -#include - -#ifdef CONFIG_NET_DIVERT -#include - -int alloc_divert_blk(struct net_device *); -void free_divert_blk(struct net_device *); -int divert_ioctl(unsigned int cmd, struct divert_cf __user *arg); -void divert_frame(struct sk_buff *skb); -static inline void handle_diverter(struct sk_buff *skb) -{ - /* if diversion is supported on device, then divert */ - if (skb->dev->divert && skb->dev->divert->divert) - divert_frame(skb); -} - -#else -# define alloc_divert_blk(dev) (0) -# define free_divert_blk(dev) do {} while (0) -# define divert_ioctl(cmd, arg) (-ENOPKG) -# define handle_diverter(skb) do {} while (0) -#endif -#endif -#endif /* _LINUX_DIVERT_H */ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 83b8c4f1d69d..4e967b2e22cc 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -38,7 +38,6 @@ #include #include -struct divert_blk; struct vlan_group; struct ethtool_ops; struct netpoll_info; @@ -517,11 +516,6 @@ struct net_device /* bridge stuff */ struct net_bridge_port *br_port; -#ifdef CONFIG_NET_DIVERT - /* this will get initialized at each interface type init routine */ - struct divert_blk *divert; -#endif /* CONFIG_NET_DIVERT */ - /* class/net/name entry */ struct class_device class_dev; /* space for optional statistics and wireless sysfs groups */ diff --git a/include/linux/sockios.h b/include/linux/sockios.h index e6b9d1d36ea2..abef7596655a 100644 --- a/include/linux/sockios.h +++ b/include/linux/sockios.h @@ -72,8 +72,8 @@ #define SIOCGIFTXQLEN 0x8942 /* Get the tx queue length */ #define SIOCSIFTXQLEN 0x8943 /* Set the tx queue length */ -#define SIOCGIFDIVERT 0x8944 /* Frame diversion support */ -#define SIOCSIFDIVERT 0x8945 /* Set frame diversion options */ +/* SIOCGIFDIVERT was: 0x8944 Frame diversion support */ +/* SIOCSIFDIVERT was: 0x8945 Set frame diversion options */ #define SIOCETHTOOL 0x8946 /* Ethtool interface */ diff --git a/net/Kconfig b/net/Kconfig index 67e39ad8b8b6..867f95032513 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -175,33 +175,6 @@ source "net/ipx/Kconfig" source "drivers/net/appletalk/Kconfig" source "net/x25/Kconfig" source "net/lapb/Kconfig" - -config NET_DIVERT - bool "Frame Diverter (EXPERIMENTAL)" - depends on EXPERIMENTAL && BROKEN - ---help--- - The Frame Diverter allows you to divert packets from the - network, that are not aimed at the interface receiving it (in - promisc. mode). Typically, a Linux box setup as an Ethernet bridge - with the Frames Diverter on, can do some *really* transparent www - caching using a Squid proxy for example. - - This is very useful when you don't want to change your router's - config (or if you simply don't have access to it). - - The other possible usages of diverting Ethernet Frames are - numberous: - - reroute smtp traffic to another interface - - traffic-shape certain network streams - - transparently proxy smtp connections - - etc... - - For more informations, please refer to: - - - - If unsure, say N. - source "net/econet/Kconfig" source "net/wanrouter/Kconfig" source "net/sched/Kconfig" diff --git a/net/core/Makefile b/net/core/Makefile index 119568077dab..73272d506e93 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -12,7 +12,6 @@ obj-y += dev.o ethtool.o dev_mcast.o dst.o netevent.o \ obj-$(CONFIG_XFRM) += flow.o obj-$(CONFIG_SYSFS) += net-sysfs.o -obj-$(CONFIG_NET_DIVERT) += dv.o obj-$(CONFIG_NET_PKTGEN) += pktgen.o obj-$(CONFIG_WIRELESS_EXT) += wireless.o obj-$(CONFIG_NETPOLL) += netpoll.o diff --git a/net/core/dev.c b/net/core/dev.c index 411c2428d268..5bf13b132dd7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -98,7 +98,6 @@ #include #include #include -#include #include #include #include @@ -1827,8 +1826,6 @@ int netif_receive_skb(struct sk_buff *skb) ncls: #endif - handle_diverter(skb); - if (handle_bridge(&skb, &pt_prev, &ret, orig_dev)) goto out; @@ -2898,10 +2895,6 @@ int register_netdevice(struct net_device *dev) spin_lock_init(&dev->ingress_lock); #endif - ret = alloc_divert_blk(dev); - if (ret) - goto out; - dev->iflink = -1; /* Init, if this function is available */ @@ -2910,13 +2903,13 @@ int register_netdevice(struct net_device *dev) if (ret) { if (ret > 0) ret = -EIO; - goto out_err; + goto out; } } if (!dev_valid_name(dev->name)) { ret = -EINVAL; - goto out_err; + goto out; } dev->ifindex = dev_new_index(); @@ -2930,7 +2923,7 @@ int register_netdevice(struct net_device *dev) = hlist_entry(p, struct net_device, name_hlist); if (!strncmp(d->name, dev->name, IFNAMSIZ)) { ret = -EEXIST; - goto out_err; + goto out; } } @@ -2974,7 +2967,7 @@ int register_netdevice(struct net_device *dev) ret = netdev_register_sysfs(dev); if (ret) - goto out_err; + goto out; dev->reg_state = NETREG_REGISTERED; /* @@ -3001,9 +2994,6 @@ int register_netdevice(struct net_device *dev) out: return ret; -out_err: - free_divert_blk(dev); - goto out; } /** @@ -3320,8 +3310,6 @@ int unregister_netdevice(struct net_device *dev) /* Notifier chain MUST detach us from master device. */ BUG_TRAP(!dev->master); - free_divert_blk(dev); - /* Finish processing unregister after unlock */ net_set_todo(dev); diff --git a/net/core/dv.c b/net/core/dv.c deleted file mode 100644 index 29ee77f15932..000000000000 --- a/net/core/dv.c +++ /dev/null @@ -1,546 +0,0 @@ -/* - * INET An implementation of the TCP/IP protocol suite for the LINUX - * operating system. INET is implemented using the BSD Socket - * interface as the means of communication with the user level. - * - * Generic frame diversion - * - * Authors: - * Benoit LOCHER: initial integration within the kernel with support for ethernet - * Dave Miller: improvement on the code (correctness, performance and source files) - * - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -const char sysctl_divert_version[32]="0.46"; /* Current version */ - -static int __init dv_init(void) -{ - return 0; -} -module_init(dv_init); - -/* - * Allocate a divert_blk for a device. This must be an ethernet nic. - */ -int alloc_divert_blk(struct net_device *dev) -{ - int alloc_size = (sizeof(struct divert_blk) + 3) & ~3; - - dev->divert = NULL; - if (dev->type == ARPHRD_ETHER) { - dev->divert = kzalloc(alloc_size, GFP_KERNEL); - if (dev->divert == NULL) { - printk(KERN_INFO "divert: unable to allocate divert_blk for %s\n", - dev->name); - return -ENOMEM; - } - dev_hold(dev); - } - - return 0; -} - -/* - * Free a divert_blk allocated by the above function, if it was - * allocated on that device. - */ -void free_divert_blk(struct net_device *dev) -{ - if (dev->divert) { - kfree(dev->divert); - dev->divert=NULL; - dev_put(dev); - } -} - -/* - * Adds a tcp/udp (source or dest) port to an array - */ -static int add_port(u16 ports[], u16 port) -{ - int i; - - if (port == 0) - return -EINVAL; - - /* Storing directly in network format for performance, - * thanks Dave :) - */ - port = htons(port); - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == port) - return -EALREADY; - } - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == 0) { - ports[i] = port; - return 0; - } - } - - return -ENOBUFS; -} - -/* - * Removes a port from an array tcp/udp (source or dest) - */ -static int remove_port(u16 ports[], u16 port) -{ - int i; - - if (port == 0) - return -EINVAL; - - /* Storing directly in network format for performance, - * thanks Dave ! - */ - port = htons(port); - - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - if (ports[i] == port) { - ports[i] = 0; - return 0; - } - } - - return -EINVAL; -} - -/* Some basic sanity checks on the arguments passed to divert_ioctl() */ -static int check_args(struct divert_cf *div_cf, struct net_device **dev) -{ - char devname[32]; - int ret; - - if (dev == NULL) - return -EFAULT; - - /* GETVERSION: all other args are unused */ - if (div_cf->cmd == DIVCMD_GETVERSION) - return 0; - - /* Network device index should reasonably be between 0 and 1000 :) */ - if (div_cf->dev_index < 0 || div_cf->dev_index > 1000) - return -EINVAL; - - /* Let's try to find the ifname */ - sprintf(devname, "eth%d", div_cf->dev_index); - *dev = dev_get_by_name(devname); - - /* dev should NOT be null */ - if (*dev == NULL) - return -EINVAL; - - ret = 0; - - /* user issuing the ioctl must be a super one :) */ - if (!capable(CAP_SYS_ADMIN)) { - ret = -EPERM; - goto out; - } - - /* Device must have a divert_blk member NOT null */ - if ((*dev)->divert == NULL) - ret = -EINVAL; -out: - dev_put(*dev); - return ret; -} - -/* - * control function of the diverter - */ -#if 0 -#define DVDBG(a) \ - printk(KERN_DEBUG "divert_ioctl() line %d %s\n", __LINE__, (a)) -#else -#define DVDBG(a) -#endif - -int divert_ioctl(unsigned int cmd, struct divert_cf __user *arg) -{ - struct divert_cf div_cf; - struct divert_blk *div_blk; - struct net_device *dev; - int ret; - - switch (cmd) { - case SIOCGIFDIVERT: - DVDBG("SIOCGIFDIVERT, copy_from_user"); - if (copy_from_user(&div_cf, arg, sizeof(struct divert_cf))) - return -EFAULT; - DVDBG("before check_args"); - ret = check_args(&div_cf, &dev); - if (ret) - return ret; - DVDBG("after checkargs"); - div_blk = dev->divert; - - DVDBG("befre switch()"); - switch (div_cf.cmd) { - case DIVCMD_GETSTATUS: - /* Now, just give the user the raw divert block - * for him to play with :) - */ - if (copy_to_user(div_cf.arg1.ptr, dev->divert, - sizeof(struct divert_blk))) - return -EFAULT; - break; - - case DIVCMD_GETVERSION: - DVDBG("GETVERSION: checking ptr"); - if (div_cf.arg1.ptr == NULL) - return -EINVAL; - DVDBG("GETVERSION: copying data to userland"); - if (copy_to_user(div_cf.arg1.ptr, - sysctl_divert_version, 32)) - return -EFAULT; - DVDBG("GETVERSION: data copied"); - break; - - default: - return -EINVAL; - } - - break; - - case SIOCSIFDIVERT: - if (copy_from_user(&div_cf, arg, sizeof(struct divert_cf))) - return -EFAULT; - - ret = check_args(&div_cf, &dev); - if (ret) - return ret; - - div_blk = dev->divert; - - switch(div_cf.cmd) { - case DIVCMD_RESET: - div_blk->divert = 0; - div_blk->protos = DIVERT_PROTO_NONE; - memset(div_blk->tcp_dst, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->tcp_src, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->udp_dst, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - memset(div_blk->udp_src, 0, - MAX_DIVERT_PORTS * sizeof(u16)); - return 0; - - case DIVCMD_DIVERT: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->divert) - return -EALREADY; - div_blk->divert = 1; - break; - - case DIVARG1_DISABLE: - if (!div_blk->divert) - return -EALREADY; - div_blk->divert = 0; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_IP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_IP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_IP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_IP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_IP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_TCP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_TCP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_TCP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_TCP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCPDST: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->tcp_dst, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->tcp_dst, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_TCPSRC: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->tcp_src, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->tcp_src, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_UDP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_UDP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_UDP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_UDP; - break; - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDPDST: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->udp_dst, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->udp_dst, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_UDPSRC: - switch(div_cf.arg1.int32) { - case DIVARG1_ADD: - return add_port(div_blk->udp_src, - div_cf.arg2.uint16); - - case DIVARG1_REMOVE: - return remove_port(div_blk->udp_src, - div_cf.arg2.uint16); - - default: - return -EINVAL; - } - - break; - - case DIVCMD_ICMP: - switch(div_cf.arg1.int32) { - case DIVARG1_ENABLE: - if (div_blk->protos & DIVERT_PROTO_ICMP) - return -EALREADY; - div_blk->protos |= DIVERT_PROTO_ICMP; - break; - - case DIVARG1_DISABLE: - if (!(div_blk->protos & DIVERT_PROTO_ICMP)) - return -EALREADY; - div_blk->protos &= ~DIVERT_PROTO_ICMP; - break; - - default: - return -EINVAL; - } - - break; - - default: - return -EINVAL; - } - - break; - - default: - return -EINVAL; - } - - return 0; -} - - -/* - * Check if packet should have its dest mac address set to the box itself - * for diversion - */ - -#define ETH_DIVERT_FRAME(skb) \ - memcpy(eth_hdr(skb), skb->dev->dev_addr, ETH_ALEN); \ - skb->pkt_type=PACKET_HOST - -void divert_frame(struct sk_buff *skb) -{ - struct ethhdr *eth = eth_hdr(skb); - struct iphdr *iph; - struct tcphdr *tcph; - struct udphdr *udph; - struct divert_blk *divert = skb->dev->divert; - int i, src, dst; - unsigned char *skb_data_end = skb->data + skb->len; - - /* Packet is already aimed at us, return */ - if (!compare_ether_addr(eth->h_dest, skb->dev->dev_addr)) - return; - - /* proto is not IP, do nothing */ - if (eth->h_proto != htons(ETH_P_IP)) - return; - - /* Divert all IP frames ? */ - if (divert->protos & DIVERT_PROTO_IP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP frame (thanks Dave) */ - iph = (struct iphdr *) skb->data; - if (((iph->ihl<<2)+(unsigned char*)(iph)) >= skb_data_end) { - printk(KERN_INFO "divert: malformed IP packet !\n"); - return; - } - - switch (iph->protocol) { - /* Divert all ICMP frames ? */ - case IPPROTO_ICMP: - if (divert->protos & DIVERT_PROTO_ICMP) { - ETH_DIVERT_FRAME(skb); - return; - } - break; - - /* Divert all TCP frames ? */ - case IPPROTO_TCP: - if (divert->protos & DIVERT_PROTO_TCP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP - * frame (thanx Dave) - */ - tcph = (struct tcphdr *) - (((unsigned char *)iph) + (iph->ihl<<2)); - if (((unsigned char *)(tcph+1)) >= skb_data_end) { - printk(KERN_INFO "divert: malformed TCP packet !\n"); - return; - } - - /* Divert some tcp dst/src ports only ?*/ - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - dst = divert->tcp_dst[i]; - src = divert->tcp_src[i]; - if ((dst && dst == tcph->dest) || - (src && src == tcph->source)) { - ETH_DIVERT_FRAME(skb); - return; - } - } - break; - - /* Divert all UDP frames ? */ - case IPPROTO_UDP: - if (divert->protos & DIVERT_PROTO_UDP) { - ETH_DIVERT_FRAME(skb); - return; - } - - /* Check for possible (maliciously) malformed IP - * packet (thanks Dave) - */ - udph = (struct udphdr *) - (((unsigned char *)iph) + (iph->ihl<<2)); - if (((unsigned char *)(udph+1)) >= skb_data_end) { - printk(KERN_INFO - "divert: malformed UDP packet !\n"); - return; - } - - /* Divert some udp dst/src ports only ? */ - for (i = 0; i < MAX_DIVERT_PORTS; i++) { - dst = divert->udp_dst[i]; - src = divert->udp_src[i]; - if ((dst && dst == udph->dest) || - (src && src == udph->source)) { - ETH_DIVERT_FRAME(skb); - return; - } - } - break; - } -} diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 02534131d88e..1e75b1585460 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -21,10 +21,6 @@ extern __u32 sysctl_rmem_max; extern int sysctl_core_destroy_delay; -#ifdef CONFIG_NET_DIVERT -extern char sysctl_divert_version[]; -#endif /* CONFIG_NET_DIVERT */ - #ifdef CONFIG_XFRM extern u32 sysctl_xfrm_aevent_etime; extern u32 sysctl_xfrm_aevent_rseqth; @@ -105,16 +101,6 @@ ctl_table core_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, -#ifdef CONFIG_NET_DIVERT - { - .ctl_name = NET_CORE_DIVERT_VERSION, - .procname = "divert_version", - .data = (void *)sysctl_divert_version, - .maxlen = 32, - .mode = 0444, - .proc_handler = &proc_dostring - }, -#endif /* CONFIG_NET_DIVERT */ #ifdef CONFIG_XFRM { .ctl_name = NET_CORE_AEVENT_ETIME, diff --git a/net/socket.c b/net/socket.c index 6c9b9b326d76..e8db54702a69 100644 --- a/net/socket.c +++ b/net/socket.c @@ -77,7 +77,6 @@ #include #include #include -#include #include #include #include @@ -852,11 +851,6 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) err = vlan_ioctl_hook(argp); mutex_unlock(&vlan_ioctl_mutex); break; - case SIOCGIFDIVERT: - case SIOCSIFDIVERT: - /* Convert this to call through a hook */ - err = divert_ioctl(cmd, argp); - break; case SIOCADDDLCI: case SIOCDELDLCI: err = -ENOPKG; -- cgit v1.2.3 From b9df3cb8cf9a96e63dfdcd3056a9cbc71f2459e7 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 11:21:36 -0200 Subject: [TCP/DCCP]: Introduce net_xmit_eval Throughout the TCP/DCCP (and tunnelling) code, it often happens that the return code of a transmit function needs to be tested against NET_XMIT_CN which is a value that does not indicate a strict error condition. This patch uses a macro for these recurring situations which is consistent with the already existing macro net_xmit_errno, saving on duplicated code. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/netdevice.h | 4 ++++ include/net/ipip.h | 2 +- net/dccp/ipv4.c | 5 ++--- net/dccp/ipv6.c | 3 +-- net/dccp/output.c | 14 ++------------ net/ipv4/tcp_ipv4.c | 3 +-- net/ipv4/tcp_output.c | 8 +------- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/tcp_ipv6.c | 3 +-- 9 files changed, 14 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4e967b2e22cc..caa3c2593719 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -66,6 +66,10 @@ struct netpoll_info; #define NET_RX_CN_HIGH 4 /* The storm is here */ #define NET_RX_BAD 5 /* packet dropped due to kernel error */ +/* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It + * indicates that the device will soon be dropping packets, or already drops + * some packets of the same priority; prompting us to send less aggressively. */ +#define net_xmit_eval(e) ((e) == NET_XMIT_CN? 0 : (e)) #define net_xmit_errno(e) ((e) != NET_XMIT_CN ? -ENOBUFS : 0) #endif diff --git a/include/net/ipip.h b/include/net/ipip.h index f490c3cbe377..84058858eea7 100644 --- a/include/net/ipip.h +++ b/include/net/ipip.h @@ -35,7 +35,7 @@ struct ip_tunnel ip_send_check(iph); \ \ err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev, dst_output);\ - if (err == NET_XMIT_SUCCESS || err == NET_XMIT_CN) { \ + if (net_xmit_eval(err) == 0) { \ stats->tx_bytes += pkt_len; \ stats->tx_packets++; \ } else { \ diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index bc400b2ba25e..61c09014dade 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -501,8 +501,7 @@ static int dccp_v4_send_response(struct sock *sk, struct request_sock *req, err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, ireq->opt); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } out: @@ -571,7 +570,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) rxskb->nh.iph->saddr, NULL); bh_unlock_sock(dccp_v4_ctl_socket->sk); - if (err == NET_XMIT_CN || err == 0) { + if (net_xmit_eval(err) == 0) { DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS); DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS); } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 8d6ddb6389a7..2165b1740c7c 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -294,8 +294,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, &ireq6->rmt_addr); ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } done: diff --git a/net/dccp/output.c b/net/dccp/output.c index 0994b13f0f15..ef22f3cc791a 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -125,16 +125,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); - if (err <= 0) - return err; - - /* NET_XMIT_CN is special. It does not guarantee, - * that this packet is lost. It tells that device - * is about to start to drop packets or already - * drops some packets of the same priority and - * invokes us to send less aggressively. - */ - return err == NET_XMIT_CN ? 0 : err; + return net_xmit_eval(err); } return -ENOBUFS; } @@ -426,8 +417,7 @@ int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code) if (skb != NULL) { memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, sk, 0); - if (err == NET_XMIT_CN) - err = 0; + return net_xmit_eval(err); } } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2eb58844403c..0ad0904bf56c 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -662,8 +662,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, ireq->rmt_addr, ireq->opt); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } out: diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f63e99aac2d5..6a8581ab9a23 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -484,13 +484,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, tcp_enter_cwr(sk); - /* NET_XMIT_CN is special. It does not guarantee, - * that this packet is lost. It tells that device - * is about to start to drop packets or already - * drops some packets of the same priority and - * invokes us to send less aggressively. - */ - return err == NET_XMIT_CN ? 0 : err; + return net_xmit_eval(err); #undef SYSCTL_FLAG_TSTAMPS #undef SYSCTL_FLAG_WSCALE diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 4919f9294e2a..80a11909159d 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -748,7 +748,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output); - if (err == NET_XMIT_SUCCESS || err == NET_XMIT_CN) { + if (net_xmit_eval(err) == 0) { stats->tx_bytes += pkt_len; stats->tx_packets++; } else { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9a8e690fdf7c..9a88395a7629 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -502,8 +502,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); - if (err == NET_XMIT_CN) - err = 0; + err = net_xmit_eval(err); } done: -- cgit v1.2.3 From c02fdc0e81e9c735d8d895af1e201b235df326d8 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 12:48:10 -0200 Subject: [DCCP]: Make feature negotiation more readable This patch replaces cryptic feature negotiation messages of type Oct 31 15:42:20 kernel: dccp_feat_change: feat change type=32 feat=1 Oct 31 15:42:21 kernel: dccp_feat_change: feat change type=34 feat=1 Oct 31 15:42:21 kernel: dccp_feat_change: feat change type=32 feat=5 into ones of type: Nov 2 13:54:45 kernel: dccp_feat_change: ChangeL(CCID (1), 3) Nov 2 13:54:45 kernel: dccp_feat_change: ChangeR(CCID (1), 3) Nov 2 13:54:45 kernel: dccp_feat_change: ChangeL(Ack Ratio (5), 2) Also, * completed the feature number list wrt RFC 4340 sec. 6.4 * annotating which ones have been implemented so far * implemented rudimentary sanity checking in feat.c (FIXMEs) * some minor fixes Commiter note: uninlined dccp_feat_name and dccp_feat_typename, for consistency with dccp_{state,packet}_name, that, BTW, should be compiled only if CONFIG_IP_DCCP_DEBUG is selected, leaving this to another cset tho. Also shortened dccp_feat_negotiation_debug to dccp_feat_debug. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 25 ++++++----- net/dccp/feat.c | 116 +++++++++++++++++++++++++++++++++++++++------------ net/dccp/feat.h | 41 +++++++++++++++++- net/dccp/options.c | 4 +- 4 files changed, 145 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 0502dfa7f32c..696319ee2d5b 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -175,18 +175,21 @@ enum { DCCPC_CCID3 = 3, }; -/* DCCP features */ -enum { - DCCPF_RESERVED = 0, - DCCPF_CCID = 1, - DCCPF_SEQUENCE_WINDOW = 3, - DCCPF_ACK_RATIO = 5, - DCCPF_SEND_ACK_VECTOR = 6, - DCCPF_SEND_NDP_COUNT = 7, +/* DCCP features (RFC 4340 section 6.4) */ + enum { + DCCPF_RESERVED = 0, + DCCPF_CCID = 1, + DCCPF_SHORT_SEQNOS = 2, /* XXX: not yet implemented */ + DCCPF_SEQUENCE_WINDOW = 3, + DCCPF_ECN_INCAPABLE = 4, /* XXX: not yet implemented */ + DCCPF_ACK_RATIO = 5, + DCCPF_SEND_ACK_VECTOR = 6, + DCCPF_SEND_NDP_COUNT = 7, DCCPF_MIN_CSUM_COVER = 8, - /* 10-127 reserved */ - DCCPF_MIN_CCID_SPECIFIC = 128, - DCCPF_MAX_CCID_SPECIFIC = 255, + DCCPF_DATA_CHECKSUM = 9, /* XXX: not yet implemented */ + /* 10-127 reserved */ + DCCPF_MIN_CCID_SPECIFIC = 128, + DCCPF_MAX_CCID_SPECIFIC = 255, }; /* this structure is argument to DCCP_SOCKOPT_CHANGE_X */ diff --git a/net/dccp/feat.c b/net/dccp/feat.c index a1b0682ee77c..12cde2f2f13b 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -12,7 +12,6 @@ #include -#include "dccp.h" #include "ccid.h" #include "feat.h" @@ -23,9 +22,17 @@ int dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature, { struct dccp_opt_pend *opt; - dccp_pr_debug("feat change type=%d feat=%d\n", type, feature); + dccp_feat_debug(type, feature, *val); - /* XXX sanity check feat change request */ + if (!dccp_feat_is_valid_type(type)) { + pr_info("option type %d invalid in negotiation\n", type); + return 1; + } + if (!dccp_feat_is_valid_length(type, feature, len)) { + pr_info("invalid length %d\n", len); + return 1; + } + /* XXX add further sanity checks */ /* check if that feature is already being negotiated */ list_for_each_entry(opt, &dmsk->dccpms_pending, dccpop_node) { @@ -95,14 +102,14 @@ static int dccp_feat_update_ccid(struct sock *sk, u8 type, u8 new_ccid_nr) /* XXX taking only u8 vals */ static int dccp_feat_update(struct sock *sk, u8 type, u8 feat, u8 val) { - dccp_pr_debug("changing [%d] feat %d to %d\n", type, feat, val); + dccp_feat_debug(type, feat, val); switch (feat) { case DCCPF_CCID: return dccp_feat_update_ccid(sk, type, val); default: - dccp_pr_debug("IMPLEMENT changing [%d] feat %d to %d\n", - type, feat, val); + dccp_pr_debug("UNIMPLEMENTED: %s(%d, ...)\n", + dccp_feat_typename(type), feat); break; } return 0; @@ -265,10 +272,10 @@ static int dccp_feat_nn(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) u8 *copy; int rc; - /* NN features must be change L */ - if (type == DCCPO_CHANGE_R) { - dccp_pr_debug("received CHANGE_R %d for NN feat %d\n", - type, feature); + /* NN features must be Change L (sec. 6.3.2) */ + if (type != DCCPO_CHANGE_L) { + dccp_pr_debug("received %s for NN feature %d\n", + dccp_feat_typename(type), feature); return -EFAULT; } @@ -299,7 +306,8 @@ static int dccp_feat_nn(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) return rc; } - dccp_pr_debug("Confirming NN feature %d (val=%d)\n", feature, *copy); + dccp_feat_debug(type, feature, *copy); + list_add_tail(&opt->dccpop_node, &dmsk->dccpms_conf); return 0; @@ -318,14 +326,19 @@ static void dccp_feat_empty_confirm(struct dccp_minisock *dmsk, return; } - opt->dccpop_type = type == DCCPO_CHANGE_L ? DCCPO_CONFIRM_R : - DCCPO_CONFIRM_L; + switch (type) { + case DCCPO_CHANGE_L: opt->dccpop_type = DCCPO_CONFIRM_R; break; + case DCCPO_CHANGE_R: opt->dccpop_type = DCCPO_CONFIRM_L; break; + default: pr_info("invalid type %d\n", type); return; + + } opt->dccpop_feat = feature; opt->dccpop_val = NULL; opt->dccpop_len = 0; /* change feature */ - dccp_pr_debug("Empty confirm feature %d type %d\n", feature, type); + dccp_pr_debug("Empty %s(%d)\n", dccp_feat_typename(type), feature); + list_add_tail(&opt->dccpop_node, &dmsk->dccpms_conf); } @@ -359,7 +372,7 @@ int dccp_feat_change_recv(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) { int rc; - dccp_pr_debug("got feat change type=%d feat=%d\n", type, feature); + dccp_feat_debug(type, feature, *val); /* figure out if it's SP or NN feature */ switch (feature) { @@ -375,6 +388,8 @@ int dccp_feat_change_recv(struct sock *sk, u8 type, u8 feature, u8 *val, u8 len) /* XXX implement other features */ default: + dccp_pr_debug("UNIMPLEMENTED: not handling %s(%d, ...)\n", + dccp_feat_typename(type), feature); rc = -EFAULT; break; } @@ -403,20 +418,27 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, u8 t; struct dccp_opt_pend *opt; struct dccp_minisock *dmsk = dccp_msk(sk); - int rc = 1; + int found = 0; int all_confirmed = 1; - dccp_pr_debug("got feat confirm type=%d feat=%d\n", type, feature); - - /* XXX sanity check type & feat */ + dccp_feat_debug(type, feature, *val); /* locate our change request */ - t = type == DCCPO_CONFIRM_L ? DCCPO_CHANGE_R : DCCPO_CHANGE_L; + switch (type) { + case DCCPO_CONFIRM_L: t = DCCPO_CHANGE_R; break; + case DCCPO_CONFIRM_R: t = DCCPO_CHANGE_L; break; + default: pr_info("invalid type %d\n", type); + return 1; + + } + /* XXX sanity check feature value */ list_for_each_entry(opt, &dmsk->dccpms_pending, dccpop_node) { if (!opt->dccpop_conf && opt->dccpop_type == t && opt->dccpop_feat == feature) { - /* we found it */ + found = 1; + dccp_pr_debug("feature %d found\n", opt->dccpop_feat); + /* XXX do sanity check */ opt->dccpop_conf = 1; @@ -425,9 +447,7 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, dccp_feat_update(sk, opt->dccpop_type, opt->dccpop_feat, *val); - dccp_pr_debug("feat %d type %d confirmed %d\n", - feature, type, *val); - rc = 0; + /* XXX check the return value of dccp_feat_update */ break; } @@ -446,9 +466,9 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS); } - if (rc) - dccp_pr_debug("feat %d type %d never requested\n", - feature, type); + if (!found) + dccp_pr_debug("%s(%d, ...) never requested\n", + dccp_feat_typename(type), feature); return 0; } @@ -583,3 +603,45 @@ out: } EXPORT_SYMBOL_GPL(dccp_feat_init); + +#ifdef CONFIG_IP_DCCP_DEBUG +const char *dccp_feat_typename(const u8 type) +{ + switch(type) { + case DCCPO_CHANGE_L: return("ChangeL"); + case DCCPO_CONFIRM_L: return("ConfirmL"); + case DCCPO_CHANGE_R: return("ChangeR"); + case DCCPO_CONFIRM_R: return("ConfirmR"); + /* the following case must not appear in feature negotation */ + default: dccp_pr_debug("unknown type %d [BUG!]\n", type); + } + return NULL; +} + +EXPORT_SYMBOL_GPL(dccp_feat_typename); + +const char *dccp_feat_name(const u8 feat) +{ + static const char *feature_names[] = { + [DCCPF_RESERVED] = "Reserved", + [DCCPF_CCID] = "CCID", + [DCCPF_SHORT_SEQNOS] = "Allow Short Seqnos", + [DCCPF_SEQUENCE_WINDOW] = "Sequence Window", + [DCCPF_ECN_INCAPABLE] = "ECN Incapable", + [DCCPF_ACK_RATIO] = "Ack Ratio", + [DCCPF_SEND_ACK_VECTOR] = "Send ACK Vector", + [DCCPF_SEND_NDP_COUNT] = "Send NDP Count", + [DCCPF_MIN_CSUM_COVER] = "Min. Csum Coverage", + [DCCPF_DATA_CHECKSUM] = "Send Data Checksum", + }; + if (feat >= DCCPF_MIN_CCID_SPECIFIC) + return "CCID-specific"; + + if (dccp_feat_is_reserved(feat)) + return feature_names[DCCPF_RESERVED]; + + return feature_names[feat]; +} + +EXPORT_SYMBOL_GPL(dccp_feat_name); +#endif /* CONFIG_IP_DCCP_DEBUG */ diff --git a/net/dccp/feat.h b/net/dccp/feat.h index 6048373c7186..2c373ad7edcf 100644 --- a/net/dccp/feat.h +++ b/net/dccp/feat.h @@ -12,9 +12,46 @@ */ #include +#include "dccp.h" -struct sock; -struct dccp_minisock; +static inline int dccp_feat_is_valid_length(u8 type, u8 feature, u8 len) +{ + /* sec. 6.1: Confirm has at least length 3, + * sec. 6.2: Change has at least length 4 */ + if (len < 3) + return 1; + if (len < 4 && (type == DCCPO_CHANGE_L || type == DCCPO_CHANGE_R)) + return 1; + /* XXX: add per-feature length validation (sec. 6.6.8) */ + return 0; +} + +static inline int dccp_feat_is_reserved(const u8 feat) +{ + return (feat > DCCPF_DATA_CHECKSUM && + feat < DCCPF_MIN_CCID_SPECIFIC) || + feat == DCCPF_RESERVED; +} + +/* feature negotiation knows only these four option types (RFC 4340, sec. 6) */ +static inline int dccp_feat_is_valid_type(const u8 optnum) +{ + return optnum >= DCCPO_CHANGE_L && optnum <= DCCPO_CONFIRM_R; + +} + +#ifdef CONFIG_IP_DCCP_DEBUG +extern const char *dccp_feat_typename(const u8 type); +extern const char *dccp_feat_name(const u8 feat); + +static inline void dccp_feat_debug(const u8 type, const u8 feat, const u8 val) +{ + dccp_pr_debug("%s(%s (%d), %d)\n", dccp_feat_typename(type), + dccp_feat_name(feat), feat, val); +} +#else +#define dccp_feat_debug(type, feat, val) +#endif /* CONFIG_IP_DCCP_DEBUG */ extern int dccp_feat_change(struct dccp_minisock *dmsk, u8 type, u8 feature, u8 *val, u8 len, gfp_t gfp); diff --git a/net/dccp/options.c b/net/dccp/options.c index 121e794fe454..2d0ef27f4ab9 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -465,8 +465,10 @@ static int dccp_insert_feat_opt(struct sk_buff *skb, u8 type, u8 feat, if (len) memcpy(to, val, len); - dccp_pr_debug("option %d feat %d len %d\n", type, feat, len); + dccp_pr_debug("%s(%s (%d), ...), length %d\n", + dccp_feat_typename(type), + dccp_feat_name(feat), feat, len); return 0; } -- cgit v1.2.3 From 09dbc3895e3242346bd434dae743c456fd28fc6a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 14 Nov 2006 12:57:34 -0200 Subject: [DCCP]: Miscellaneous code tidy-ups This patch does not change code; it performs some trivial clean/tidy-ups: * removal of a `debug_prefix' string in favour of the already existing dccp_role(sk) * add documentation of structures and constants * separated out the cases for invalid packets (step 1 of the packet validation) * removing duplicate statements * combining declaration & initialisation Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 18 +++++++++++++++++- net/dccp/ackvec.c | 27 ++++++--------------------- net/dccp/ipv4.c | 39 +++++++++++++++++++-------------------- net/dccp/ipv6.c | 6 ++---- net/dccp/options.c | 18 +++++++----------- net/dccp/output.c | 1 + net/dccp/proto.c | 8 ++------ 7 files changed, 54 insertions(+), 63 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 696319ee2d5b..d308f1228b61 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -209,6 +209,7 @@ struct dccp_so_feat { #define DCCP_SOCKOPT_CCID_RX_INFO 128 #define DCCP_SOCKOPT_CCID_TX_INFO 192 +/* maximum number of services provided on the same listening port */ #define DCCP_SERVICE_LIST_MAX_LEN 32 #ifdef __KERNEL__ @@ -355,6 +356,9 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb) * @dccpms_ccid - Congestion Control Id (CCID) (section 10) * @dccpms_send_ack_vector - Send Ack Vector Feature (section 11.5) * @dccpms_send_ndp_count - Send NDP Count Feature (7.7.2) + * @dccpms_ack_ratio - Ack Ratio Feature (section 11.3) + * @dccpms_pending - List of features being negotiated + * @dccpms_conf - */ struct dccp_minisock { __u64 dccpms_sequence_window; @@ -452,14 +456,26 @@ struct dccp_ackvec; * @dccps_gss - greatest sequence number sent * @dccps_gsr - greatest valid sequence number received * @dccps_gar - greatest valid ack number received on a non-Sync; initialized to %dccps_iss + * @dccps_service - first (passive sock) or unique (active sock) service code + * @dccps_service_list - second .. last service code on passive socket * @dccps_timestamp_time - time of latest TIMESTAMP option * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option * @dccps_packet_size - Set thru setsockopt - * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_l_ack_ratio - + * @dccps_r_ack_ratio - * @dccps_pcslen - sender partial checksum coverage (via sockopt) * @dccps_pcrlen - receiver partial checksum coverage (via sockopt) * @dccps_ndp_count - number of Non Data Packets since last data packet + * @dccps_mss_cache - + * @dccps_minisock - * @dccps_hc_rx_ackvec - rx half connection ack vector + * @dccps_hc_rx_ccid - + * @dccps_hc_tx_ccid - + * @dccps_options_received - + * @dccps_epoch - + * @dccps_role - Role of this sock, one of %dccp_role + * @dccps_hc_rx_insert_options - + * @dccps_hc_tx_insert_options - * @dccps_xmit_timer - timer for when CCID is not ready to send */ struct dccp_sock { diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index f8208874ac7d..0c54b89a4e9b 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -67,10 +67,6 @@ static void dccp_ackvec_insert_avr(struct dccp_ackvec *av, int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); -#ifdef CONFIG_IP_DCCP_DEBUG - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT tx: " : "server tx: "; -#endif struct dccp_ackvec *av = dp->dccps_hc_rx_ackvec; int len = av->dccpav_vec_len + 2; struct timeval now; @@ -129,9 +125,9 @@ int dccp_insert_option_ackvec(struct sock *sk, struct sk_buff *skb) dccp_ackvec_insert_avr(av, avr); - dccp_pr_debug("%sACK Vector 0, len=%d, ack_seqno=%llu, " + dccp_pr_debug("%s ACK Vector 0, len=%d, ack_seqno=%llu, " "ack_ackno=%llu\n", - debug_prefix, avr->dccpavr_sent_len, + dccp_role(sk), avr->dccpavr_sent_len, (unsigned long long)avr->dccpavr_ack_seqno, (unsigned long long)avr->dccpavr_ack_ackno); return 0; @@ -380,14 +376,9 @@ void dccp_ackvec_check_rcv_ackno(struct dccp_ackvec *av, struct sock *sk, */ list_for_each_entry_reverse(avr, &av->dccpav_records, dccpavr_node) { if (ackno == avr->dccpavr_ack_seqno) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK packet 0, len=%d, ack_seqno=%llu, " + dccp_pr_debug("%s ACK packet 0, len=%d, ack_seqno=%llu, " "ack_ackno=%llu, ACKED!\n", - debug_prefix, 1, + dccp_role(sk), 1, (unsigned long long)avr->dccpavr_ack_seqno, (unsigned long long)avr->dccpavr_ack_ackno); dccp_ackvec_throw_record(av, avr); @@ -437,16 +428,10 @@ found: if (between48(avr->dccpavr_ack_seqno, ackno_end_rl, ackno)) { const u8 state = *vector & DCCP_ACKVEC_STATE_MASK; if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED) { -#ifdef CONFIG_IP_DCCP_DEBUG - struct dccp_sock *dp = dccp_sk(sk); - const char *debug_prefix = - dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx ack: " : "server rx ack: "; -#endif - dccp_pr_debug("%sACK vector 0, len=%d, " + dccp_pr_debug("%s ACK vector 0, len=%d, " "ack_seqno=%llu, ack_ackno=%llu, " "ACKED!\n", - debug_prefix, len, + dccp_role(sk), len, (unsigned long long) avr->dccpavr_ack_seqno, (unsigned long long) diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 61c09014dade..34d6d197c3b2 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -518,7 +518,7 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) sizeof(struct dccp_hdr_reset); struct sk_buff *skb; struct dst_entry *dst; - u64 seqno; + u64 seqno = 0; /* Never send a reset in response to a reset. */ if (rxdh->dccph_type == DCCP_PKT_RESET) @@ -552,13 +552,11 @@ static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) DCCP_SKB_CB(rxskb)->dccpd_reset_code; /* See "8.3.1. Abnormal Termination" in RFC 4340 */ - seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); dccp_hdr_set_seq(dh, seqno); - dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), - DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v4_csum_finish(skb, rxskb->nh.iph->saddr, @@ -734,6 +732,11 @@ discard: EXPORT_SYMBOL_GPL(dccp_v4_do_rcv); +/** + * dccp_invalid_packet - check for malformed packets + * Implements RFC 4340, 8.5: Step 1: Check header basics + * Packets that fail these checks are ignored and do not receive Resets. + */ int dccp_invalid_packet(struct sk_buff *skb) { const struct dccp_hdr *dh; @@ -742,6 +745,7 @@ int dccp_invalid_packet(struct sk_buff *skb) if (skb->pkt_type != PACKET_HOST) return 1; + /* If the packet is shorter than 12 bytes, drop packet and return */ if (!pskb_may_pull(skb, sizeof(struct dccp_hdr))) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: pskb_may_pull failed\n"); return 1; @@ -749,42 +753,37 @@ int dccp_invalid_packet(struct sk_buff *skb) dh = dccp_hdr(skb); - /* If the packet type is not understood, drop packet and return */ + /* If P.type is not understood, drop packet and return */ if (dh->dccph_type >= DCCP_PKT_INVALID) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: invalid packet type\n"); return 1; } /* - * If P.Data Offset is too small for packet type, or too large for - * packet, drop packet and return + * If P.Data Offset is too small for packet type, drop packet and return */ if (dh->dccph_doff < dccp_hdr_len(skb) / sizeof(u32)) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.Data Offset(%u) " - "too small 1\n", - dh->dccph_doff); + "too small\n", dh->dccph_doff); return 1; } - + /* + * If P.Data Offset is too too large for packet, drop packet and return + */ if (!pskb_may_pull(skb, dh->dccph_doff * sizeof(u32))) { LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.Data Offset(%u) " - "too small 2\n", - dh->dccph_doff); + "too large\n", dh->dccph_doff); return 1; } - dh = dccp_hdr(skb); - /* * If P.type is not Data, Ack, or DataAck and P.X == 0 (the packet * has short sequence numbers), drop packet and return */ - if (dh->dccph_x == 0 && - dh->dccph_type != DCCP_PKT_DATA && - dh->dccph_type != DCCP_PKT_ACK && - dh->dccph_type != DCCP_PKT_DATAACK) { - LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.type (%s) not Data, Ack " - "nor DataAck and P.X == 0\n", + if (dh->dccph_type >= DCCP_PKT_DATA && + dh->dccph_type <= DCCP_PKT_DATAACK && dh->dccph_x == 0) { + LIMIT_NETDEBUG(KERN_WARNING "DCCP: P.type (%s) not Data||Ack||" + "DataAck, while P.X == 0\n", dccp_packet_name(dh->dccph_type)); return 1; } diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 2165b1740c7c..fc326173c215 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -318,7 +318,7 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) sizeof(struct dccp_hdr_reset); struct sk_buff *skb; struct flowi fl; - u64 seqno; + u64 seqno = 0; if (rxdh->dccph_type == DCCP_PKT_RESET) return; @@ -345,13 +345,11 @@ static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) DCCP_SKB_CB(rxskb)->dccpd_reset_code; /* See "8.3.1. Abnormal Termination" in RFC 4340 */ - seqno = 0; if (DCCP_SKB_CB(rxskb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_set_seqno(&seqno, DCCP_SKB_CB(rxskb)->dccpd_ack_seq + 1); dccp_hdr_set_seq(dh, seqno); - dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), - DCCP_SKB_CB(rxskb)->dccpd_seq); + dccp_hdr_set_ack(dccp_hdr_ack_bits(skb), DCCP_SKB_CB(rxskb)->dccpd_seq); dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, diff --git a/net/dccp/options.c b/net/dccp/options.c index 2d0ef27f4ab9..7e50678e2471 100644 --- a/net/dccp/options.c +++ b/net/dccp/options.c @@ -60,10 +60,6 @@ static u32 dccp_decode_value_var(const unsigned char *bf, const u8 len) int dccp_parse_options(struct sock *sk, struct sk_buff *skb) { struct dccp_sock *dp = dccp_sk(sk); -#ifdef CONFIG_IP_DCCP_DEBUG - const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ? - "CLIENT rx opt: " : "server rx opt: "; -#endif const struct dccp_hdr *dh = dccp_hdr(skb); const u8 pkt_type = DCCP_SKB_CB(skb)->dccpd_type; unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb); @@ -119,7 +115,7 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) goto out_invalid_option; opt_recv->dccpor_ndp = dccp_decode_value_var(value, len); - dccp_pr_debug("%sNDP count=%d\n", debug_prefix, + dccp_pr_debug("%s rx opt: NDP count=%d\n", dccp_role(sk), opt_recv->dccpor_ndp); break; case DCCPO_CHANGE_L: @@ -165,8 +161,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) dp->dccps_timestamp_echo = opt_recv->dccpor_timestamp; dccp_timestamp(sk, &dp->dccps_timestamp_time); - dccp_pr_debug("%sTIMESTAMP=%u, ackno=%llu\n", - debug_prefix, opt_recv->dccpor_timestamp, + dccp_pr_debug("%s rx opt: TIMESTAMP=%u, ackno=%llu\n", + dccp_role(sk), opt_recv->dccpor_timestamp, (unsigned long long) DCCP_SKB_CB(skb)->dccpd_ack_seq); break; @@ -176,8 +172,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) opt_recv->dccpor_timestamp_echo = ntohl(*(__be32 *)value); - dccp_pr_debug("%sTIMESTAMP_ECHO=%u, len=%d, ackno=%llu, ", - debug_prefix, + dccp_pr_debug("%s rx opt: TIMESTAMP_ECHO=%u, len=%d, " + "ackno=%llu, ", dccp_role(sk), opt_recv->dccpor_timestamp_echo, len + 2, (unsigned long long) @@ -211,8 +207,8 @@ int dccp_parse_options(struct sock *sk, struct sk_buff *skb) if (elapsed_time > opt_recv->dccpor_elapsed_time) opt_recv->dccpor_elapsed_time = elapsed_time; - dccp_pr_debug("%sELAPSED_TIME=%d\n", debug_prefix, - elapsed_time); + dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n", + dccp_role(sk), elapsed_time); break; /* * From RFC 4340, sec. 10.3: diff --git a/net/dccp/output.c b/net/dccp/output.c index ef22f3cc791a..c34eada7f025 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -333,6 +333,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, return NULL; } + /* Build and checksum header */ dh = dccp_zeroed_hdr(skb, dccp_header_size); dh->dccph_sport = inet_sk(sk)->sport; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 9c9c08cffdaf..0225bdacd3b1 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -452,9 +452,8 @@ out_free_val: static int do_dccp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - struct dccp_sock *dp; - int err; - int val; + struct dccp_sock *dp = dccp_sk(sk); + int val, err = 0; if (optlen < sizeof(int)) return -EINVAL; @@ -466,9 +465,6 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, return dccp_setsockopt_service(sk, val, optval, optlen); lock_sock(sk); - dp = dccp_sk(sk); - err = 0; - switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: dp->dccps_packet_size = val; -- cgit v1.2.3 From 93ec2c723e3f8a216dde2899aeb85c648672bc6b Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:50 -0700 Subject: netpoll info leak After looking harder, Steve noticed that the netpoll beast leaked a little every time it shutdown for a nap. Not a big leak, but a nuisance kind of thing. He took out his refcount duct tape and patched the leak. It was overkill since there was already other locking in that area, but it looked clean and wouldn't attract fleas. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 1 + net/core/netpoll.c | 25 +++++++++++++++++++------ 2 files changed, 20 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 1efe60c5c00c..39845fc975f9 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -25,6 +25,7 @@ struct netpoll { }; struct netpoll_info { + atomic_t refcnt; spinlock_t poll_lock; int poll_owner; int tries; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 4de62f1f4134..c66df2f45d26 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -658,8 +658,11 @@ int netpoll_setup(struct netpoll *np) npinfo->tries = MAX_RETRIES; spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); - } else + atomic_set(&npinfo->refcnt, 1); + } else { npinfo = ndev->npinfo; + atomic_inc(&npinfo->refcnt); + } if (!ndev->poll_controller) { printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n", @@ -766,12 +769,22 @@ void netpoll_cleanup(struct netpoll *np) if (np->dev) { npinfo = np->dev->npinfo; - if (npinfo && npinfo->rx_np == np) { - spin_lock_irqsave(&npinfo->rx_lock, flags); - npinfo->rx_np = NULL; - npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; - spin_unlock_irqrestore(&npinfo->rx_lock, flags); + if (npinfo) { + if (npinfo->rx_np == np) { + spin_lock_irqsave(&npinfo->rx_lock, flags); + npinfo->rx_np = NULL; + npinfo->rx_flags &= ~NETPOLL_RX_ENABLED; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + } + + np->dev->npinfo = NULL; + if (atomic_dec_and_test(&npinfo->refcnt)) { + skb_queue_purge(&npinfo->arp_tx); + + kfree(npinfo); + } } + dev_put(np->dev); } -- cgit v1.2.3 From b6cd27ed33886a5ffaf0925a6d98e13e18e8a1af Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:51 -0700 Subject: netpoll per device txq When the netpoll beast got really busy, it tended to clog things, so it stored them for later. But the beast was putting all it's skb's in one basket. This was bad because maybe some pipes were clogged and others were not. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 2 ++ net/core/netpoll.c | 50 +++++++++++++++---------------------------------- 2 files changed, 17 insertions(+), 35 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 39845fc975f9..93a8b7664423 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -33,6 +33,8 @@ struct netpoll_info { spinlock_t rx_lock; struct netpoll *rx_np; /* netpoll that registered an rx_hook */ struct sk_buff_head arp_tx; /* list of arp requests to reply to */ + struct sk_buff_head txq; + struct work_struct tx_work; }; void netpoll_poll(struct netpoll *np); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index c66df2f45d26..ac4e8b8f57d1 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -38,10 +38,6 @@ static struct sk_buff_head skb_pool; -static DEFINE_SPINLOCK(queue_lock); -static int queue_depth; -static struct sk_buff *queue_head, *queue_tail; - static atomic_t trapped; #define NETPOLL_RX_ENABLED 1 @@ -56,46 +52,25 @@ static void arp_reply(struct sk_buff *skb); static void queue_process(void *p) { - unsigned long flags; + struct netpoll_info *npinfo = p; struct sk_buff *skb; - while (queue_head) { - spin_lock_irqsave(&queue_lock, flags); - - skb = queue_head; - queue_head = skb->next; - if (skb == queue_tail) - queue_head = NULL; - - queue_depth--; - - spin_unlock_irqrestore(&queue_lock, flags); - + while ((skb = skb_dequeue(&npinfo->txq))) dev_queue_xmit(skb); - } -} -static DECLARE_WORK(send_queue, queue_process, NULL); +} void netpoll_queue(struct sk_buff *skb) { - unsigned long flags; + struct net_device *dev = skb->dev; + struct netpoll_info *npinfo = dev->npinfo; - if (queue_depth == MAX_QUEUE_DEPTH) { - __kfree_skb(skb); - return; + if (!npinfo) + kfree_skb(skb); + else { + skb_queue_tail(&npinfo->txq, skb); + schedule_work(&npinfo->tx_work); } - - spin_lock_irqsave(&queue_lock, flags); - if (!queue_head) - queue_head = skb; - else - queue_tail->next = skb; - queue_tail = skb; - queue_depth++; - spin_unlock_irqrestore(&queue_lock, flags); - - schedule_work(&send_queue); } static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, @@ -658,6 +633,9 @@ int netpoll_setup(struct netpoll *np) npinfo->tries = MAX_RETRIES; spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); + skb_queue_head_init(&npinfo->txq); + INIT_WORK(&npinfo->tx_work, queue_process, npinfo); + atomic_set(&npinfo->refcnt, 1); } else { npinfo = ndev->npinfo; @@ -780,6 +758,8 @@ void netpoll_cleanup(struct netpoll *np) np->dev->npinfo = NULL; if (atomic_dec_and_test(&npinfo->refcnt)) { skb_queue_purge(&npinfo->arp_tx); + skb_queue_purge(&npinfo->txq); + flush_scheduled_work(); kfree(npinfo); } -- cgit v1.2.3 From 2bdfe0baeca0e2750037b8fba71905c00ac3c515 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:54 -0700 Subject: netpoll retry cleanup The netpoll beast was still not happy. If the beast got clogged pipes, it tended to stare blankly off in space for a long time. The problem couldn't be completely fixed because the beast talked with irq's disabled. But it could be made less painful and shorter. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 1 - net/core/netpoll.c | 71 +++++++++++++++++++++++-------------------------- 2 files changed, 33 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 93a8b7664423..c65d12ec7bb0 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -28,7 +28,6 @@ struct netpoll_info { atomic_t refcnt; spinlock_t poll_lock; int poll_owner; - int tries; int rx_flags; spinlock_t rx_lock; struct netpoll *rx_np; /* netpoll that registered an rx_hook */ diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 93cb828f3aaf..6b34c394672f 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -34,12 +34,12 @@ #define MAX_UDP_CHUNK 1460 #define MAX_SKBS 32 #define MAX_QUEUE_DEPTH (MAX_SKBS / 2) -#define MAX_RETRIES 20000 static struct sk_buff_head skb_pool; static atomic_t trapped; +#define USEC_PER_POLL 50 #define NETPOLL_RX_ENABLED 1 #define NETPOLL_RX_DROP 2 @@ -72,6 +72,7 @@ static void queue_process(void *p) schedule_delayed_work(&npinfo->tx_work, HZ/10); return; } + netif_tx_unlock_bh(dev); } } @@ -244,50 +245,44 @@ repeat: static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) { - int status; - struct netpoll_info *npinfo; + int status = NETDEV_TX_BUSY; + unsigned long tries; + struct net_device *dev = np->dev; + struct netpoll_info *npinfo = np->dev->npinfo; + + if (!npinfo || !netif_running(dev) || !netif_device_present(dev)) { + __kfree_skb(skb); + return; + } + + /* don't get messages out of order, and no recursion */ + if ( !(np->drop == netpoll_queue && skb_queue_len(&npinfo->txq)) + && npinfo->poll_owner != smp_processor_id() + && netif_tx_trylock(dev)) { + + /* try until next clock tick */ + for(tries = jiffies_to_usecs(1)/USEC_PER_POLL; tries > 0; --tries) { + if (!netif_queue_stopped(dev)) + status = dev->hard_start_xmit(skb, dev); + + if (status == NETDEV_TX_OK) + break; - if (!np || !np->dev || !netif_running(np->dev)) { - __kfree_skb(skb); - return; - } + /* tickle device maybe there is some cleanup */ + netpoll_poll(np); - npinfo = np->dev->npinfo; + udelay(USEC_PER_POLL); + } + netif_tx_unlock(dev); + } - /* avoid recursion */ - if (npinfo->poll_owner == smp_processor_id() || - np->dev->xmit_lock_owner == smp_processor_id()) { + if (status != NETDEV_TX_OK) { + /* requeue for later */ if (np->drop) np->drop(skb); else __kfree_skb(skb); - return; } - - do { - npinfo->tries--; - netif_tx_lock(np->dev); - - /* - * network drivers do not expect to be called if the queue is - * stopped. - */ - status = NETDEV_TX_BUSY; - if (!netif_queue_stopped(np->dev)) - status = np->dev->hard_start_xmit(skb, np->dev); - - netif_tx_unlock(np->dev); - - /* success */ - if(!status) { - npinfo->tries = MAX_RETRIES; /* reset */ - return; - } - - /* transmit busy */ - netpoll_poll(np); - udelay(50); - } while (npinfo->tries > 0); } void netpoll_send_udp(struct netpoll *np, const char *msg, int len) @@ -649,7 +644,7 @@ int netpoll_setup(struct netpoll *np) npinfo->rx_np = NULL; spin_lock_init(&npinfo->poll_lock); npinfo->poll_owner = -1; - npinfo->tries = MAX_RETRIES; + spin_lock_init(&npinfo->rx_lock); skb_queue_head_init(&npinfo->arp_tx); skb_queue_head_init(&npinfo->txq); -- cgit v1.2.3 From 5de4a473bda49554e4e9bd93b78f43c49a7ea69c Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:55 -0700 Subject: netpoll queue cleanup The beast had a long and not very happy history. At one point, a friend (netdump) had asked that he open up a little. Well, the friend was long gone now, and the beast had this dangling piece hanging (netpoll_queue). It wasn't hard to stitch the netpoll_queue back in where it belonged and make everything tidy. Signed-off-by: Stephen Hemminger --- drivers/net/netconsole.c | 1 - include/linux/netpoll.h | 4 ++-- net/core/netpoll.c | 23 +++-------------------- 3 files changed, 5 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index e6e8a9797b6d..69233f6aa05c 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -60,7 +60,6 @@ static struct netpoll np = { .local_port = 6665, .remote_port = 6666, .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}, - .drop = netpoll_queue, }; static int configured = 0; diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index c65d12ec7bb0..b7eb008c43de 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -18,7 +18,7 @@ struct netpoll { struct net_device *dev; char dev_name[16], *name; void (*rx_hook)(struct netpoll *, int, char *, int); - void (*drop)(struct sk_buff *skb); + u32 local_ip, remote_ip; u16 local_port, remote_port; unsigned char local_mac[6], remote_mac[6]; @@ -44,7 +44,7 @@ int netpoll_trap(void); void netpoll_set_trap(int trap); void netpoll_cleanup(struct netpoll *np); int __netpoll_rx(struct sk_buff *skb); -void netpoll_queue(struct sk_buff *skb); + #ifdef CONFIG_NETPOLL static inline int netpoll_rx(struct sk_buff *skb) diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6b34c394672f..0d1de3c47a01 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -77,19 +77,6 @@ static void queue_process(void *p) } } -void netpoll_queue(struct sk_buff *skb) -{ - struct net_device *dev = skb->dev; - struct netpoll_info *npinfo = dev->npinfo; - - if (!npinfo) - kfree_skb(skb); - else { - skb_queue_tail(&npinfo->txq, skb); - schedule_work(&npinfo->tx_work); - } -} - static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, unsigned short ulen, u32 saddr, u32 daddr) { @@ -256,7 +243,7 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } /* don't get messages out of order, and no recursion */ - if ( !(np->drop == netpoll_queue && skb_queue_len(&npinfo->txq)) + if ( skb_queue_len(&npinfo->txq) == 0 && npinfo->poll_owner != smp_processor_id() && netif_tx_trylock(dev)) { @@ -277,11 +264,8 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } if (status != NETDEV_TX_OK) { - /* requeue for later */ - if (np->drop) - np->drop(skb); - else - __kfree_skb(skb); + skb_queue_tail(&npinfo->txq, skb); + schedule_work(&npinfo->tx_work); } } @@ -809,4 +793,3 @@ EXPORT_SYMBOL(netpoll_setup); EXPORT_SYMBOL(netpoll_cleanup); EXPORT_SYMBOL(netpoll_send_udp); EXPORT_SYMBOL(netpoll_poll); -EXPORT_SYMBOL(netpoll_queue); -- cgit v1.2.3 From bf6bce71eae386dbc37f93af7e5ad173450d9945 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 26 Oct 2006 15:46:56 -0700 Subject: netpoll header cleanup As Steve left netpoll beast, hopefully not to return soon. He noticed that the header was messy. He straightened it up and polished it a little, then waved goodbye. Signed-off-by: Stephen Hemminger --- include/linux/netpoll.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index b7eb008c43de..2cc9867b1626 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -12,16 +12,15 @@ #include #include -struct netpoll; - struct netpoll { struct net_device *dev; - char dev_name[16], *name; + char dev_name[IFNAMSIZ]; + const char *name; void (*rx_hook)(struct netpoll *, int, char *, int); u32 local_ip, remote_ip; u16 local_port, remote_port; - unsigned char local_mac[6], remote_mac[6]; + u8 local_mac[ETH_ALEN], remote_mac[ETH_ALEN]; }; struct netpoll_info { -- cgit v1.2.3 From cfb6eeb4c860592edd123fdea908d23c6ad1c7dc Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 14 Nov 2006 19:07:45 -0800 Subject: [TCP]: MD5 Signature Option (RFC2385) support. Based on implementation by Rick Payne. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- CREDITS | 3 + include/linux/tcp.h | 35 ++- include/net/request_sock.h | 3 +- include/net/tcp.h | 143 ++++++++++ include/net/timewait_sock.h | 3 + net/dccp/ipv4.c | 6 +- net/dccp/ipv6.c | 6 +- net/dccp/minisocks.c | 2 +- net/ipv4/Kconfig | 16 ++ net/ipv4/tcp.c | 137 +++++++++ net/ipv4/tcp_input.c | 8 + net/ipv4/tcp_ipv4.c | 673 ++++++++++++++++++++++++++++++++++++++++++-- net/ipv4/tcp_minisocks.c | 64 ++++- net/ipv4/tcp_output.c | 111 +++++++- net/ipv6/tcp_ipv6.c | 568 +++++++++++++++++++++++++++++++++++-- 15 files changed, 1714 insertions(+), 64 deletions(-) (limited to 'include/linux') diff --git a/CREDITS b/CREDITS index ccd4f9f4dd71..d0880082c19b 100644 --- a/CREDITS +++ b/CREDITS @@ -2598,6 +2598,9 @@ S: Ucitelska 1576 S: Prague 8 S: 182 00 Czech Republic +N: Rick Payne +D: RFC2385 Support for TCP + N: Barak A. Pearlmutter E: bap@cs.unm.edu W: http://www.cs.unm.edu/~bap/ diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 2d36f6db3706..0aecfc955591 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -19,6 +19,7 @@ #include #include +#include struct tcphdr { __be16 source; @@ -94,6 +95,7 @@ enum { #define TCP_INFO 11 /* Information about this connection. */ #define TCP_QUICKACK 12 /* Block/reenable quick acks */ #define TCP_CONGESTION 13 /* Congestion control algorithm */ +#define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ #define TCPI_OPT_TIMESTAMPS 1 #define TCPI_OPT_SACK 2 @@ -157,6 +159,17 @@ struct tcp_info __u32 tcpi_total_retrans; }; +/* for TCP_MD5SIG socket option */ +#define TCP_MD5SIG_MAXKEYLEN 80 + +struct tcp_md5sig { + struct __kernel_sockaddr_storage tcpm_addr; /* address associated */ + __u16 __tcpm_pad1; /* zero */ + __u16 tcpm_keylen; /* key length */ + __u32 __tcpm_pad2; /* zero */ + __u8 tcpm_key[TCP_MD5SIG_MAXKEYLEN]; /* key (binary) */ +}; + #ifdef __KERNEL__ #include @@ -197,9 +210,13 @@ struct tcp_options_received { }; struct tcp_request_sock { - struct inet_request_sock req; - __u32 rcv_isn; - __u32 snt_isn; + struct inet_request_sock req; +#ifdef CONFIG_TCP_MD5SIG + /* Only used by TCP MD5 Signature so far. */ + struct tcp_request_sock_ops *af_specific; +#endif + __u32 rcv_isn; + __u32 snt_isn; }; static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) @@ -363,6 +380,14 @@ struct tcp_sock { __u32 probe_seq_start; __u32 probe_seq_end; } mtu_probe; + +#ifdef CONFIG_TCP_MD5SIG +/* TCP AF-Specific parts; only used by MD5 Signature support so far */ + struct tcp_sock_af_ops *af_specific; + +/* TCP MD5 Signagure Option information */ + struct tcp_md5sig_info *md5sig_info; +#endif }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) @@ -377,6 +402,10 @@ struct tcp_timewait_sock { __u32 tw_rcv_wnd; __u32 tw_ts_recent; long tw_ts_recent_stamp; +#ifdef CONFIG_TCP_MD5SIG + __u16 tw_md5_keylen; + __u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; +#endif }; static inline struct tcp_timewait_sock *tcp_twsk(const struct sock *sk) diff --git a/include/net/request_sock.h b/include/net/request_sock.h index b5b023e79e5f..e37baaf2080b 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -35,7 +35,8 @@ struct request_sock_ops { struct dst_entry *dst); void (*send_ack)(struct sk_buff *skb, struct request_sock *req); - void (*send_reset)(struct sk_buff *skb); + void (*send_reset)(struct sock *sk, + struct sk_buff *skb); void (*destructor)(struct request_sock *req); }; diff --git a/include/net/tcp.h b/include/net/tcp.h index e1a5d29d0a1f..363960872de0 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -161,6 +162,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOPT_SACK_PERM 4 /* SACK Permitted */ #define TCPOPT_SACK 5 /* SACK Block */ #define TCPOPT_TIMESTAMP 8 /* Better RTT estimations/PAWS */ +#define TCPOPT_MD5SIG 19 /* MD5 Signature (RFC2385) */ /* * TCP option lengths @@ -170,6 +172,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_WINDOW 3 #define TCPOLEN_SACK_PERM 2 #define TCPOLEN_TIMESTAMP 10 +#define TCPOLEN_MD5SIG 18 /* But this is what stacks really send out. */ #define TCPOLEN_TSTAMP_ALIGNED 12 @@ -178,6 +181,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_SACK_BASE 2 #define TCPOLEN_SACK_BASE_ALIGNED 4 #define TCPOLEN_SACK_PERBLOCK 8 +#define TCPOLEN_MD5SIG_ALIGNED 20 /* Flags in tp->nonagle */ #define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */ @@ -299,6 +303,8 @@ extern void tcp_cleanup_rbuf(struct sock *sk, int copied); extern int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp); +extern void tcp_twsk_destructor(struct sock *sk); + static inline void tcp_dec_quickack_mode(struct sock *sk, const unsigned int pkts) { @@ -1064,6 +1070,114 @@ static inline void clear_all_retrans_hints(struct tcp_sock *tp){ tp->fastpath_skb_hint = NULL; } +/* MD5 Signature */ +struct crypto_hash; + +/* - key database */ +struct tcp_md5sig_key { + u8 *key; + u8 keylen; +}; + +struct tcp4_md5sig_key { + u8 *key; + u16 keylen; + __be32 addr; +}; + +struct tcp6_md5sig_key { + u8 *key; + u16 keylen; +#if 0 + u32 scope_id; /* XXX */ +#endif + struct in6_addr addr; +}; + +/* - sock block */ +struct tcp_md5sig_info { + struct tcp4_md5sig_key *keys4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct tcp6_md5sig_key *keys6; + u32 entries6; + u32 alloced6; +#endif + u32 entries4; + u32 alloced4; +}; + +/* - pseudo header */ +struct tcp4_pseudohdr { + __be32 saddr; + __be32 daddr; + __u8 pad; + __u8 protocol; + __be16 len; +}; + +struct tcp6_pseudohdr { + struct in6_addr saddr; + struct in6_addr daddr; + __be32 len; + __be32 protocol; /* including padding */ +}; + +union tcp_md5sum_block { + struct tcp4_pseudohdr ip4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct tcp6_pseudohdr ip6; +#endif +}; + +/* - pool: digest algorithm, hash description and scratch buffer */ +struct tcp_md5sig_pool { + struct hash_desc md5_desc; + union tcp_md5sum_block md5_blk; +}; + +#define TCP_MD5SIG_MAXKEYS (~(u32)0) /* really?! */ + +/* - functions */ +extern int tcp_v4_calc_md5_hash(char *md5_hash, + struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, + int protocol, int tcplen); +extern struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk, + struct sock *addr_sk); + +extern int tcp_v4_md5_do_add(struct sock *sk, + __be32 addr, + u8 *newkey, + u8 newkeylen); + +extern int tcp_v4_md5_do_del(struct sock *sk, + u32 addr); + +extern struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void); +extern void tcp_free_md5sig_pool(void); + +extern struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu); +extern void __tcp_put_md5sig_pool(void); + +static inline +struct tcp_md5sig_pool *tcp_get_md5sig_pool(void) +{ + int cpu = get_cpu(); + struct tcp_md5sig_pool *ret = __tcp_get_md5sig_pool(cpu); + if (!ret) + put_cpu(); + return ret; +} + +static inline void tcp_put_md5sig_pool(void) +{ + __tcp_put_md5sig_pool(); + put_cpu(); +} + /* /proc */ enum tcp_seq_states { TCP_SEQ_STATE_LISTENING, @@ -1103,6 +1217,35 @@ extern int tcp4_proc_init(void); extern void tcp4_proc_exit(void); #endif +/* TCP af-specific functions */ +struct tcp_sock_af_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct sock *addr_sk); + int (*calc_md5_hash) (char *location, + struct tcp_md5sig_key *md5, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, + int protocol, int len); + int (*md5_add) (struct sock *sk, + struct sock *addr_sk, + u8 *newkey, + u8 len); + int (*md5_parse) (struct sock *sk, + char __user *optval, + int optlen); +#endif +}; + +struct tcp_request_sock_ops { +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *(*md5_lookup) (struct sock *sk, + struct request_sock *req); +#endif +}; + extern void tcp_v4_init(struct net_proto_family *ops); extern void tcp_init(void); diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index be293d795e38..d7a306ea560d 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -31,6 +31,9 @@ static inline int twsk_unique(struct sock *sk, struct sock *sktw, void *twp) static inline void twsk_destructor(struct sock *sk) { + BUG_ON(sk == NULL); + BUG_ON(sk->sk_prot == NULL); + BUG_ON(sk->sk_prot->twsk_prot == NULL); if (sk->sk_prot->twsk_prot->twsk_destructor != NULL) sk->sk_prot->twsk_prot->twsk_destructor(sk); } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 34d6d197c3b2..35985334daee 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -509,7 +509,7 @@ out: return err; } -static void dccp_v4_ctl_send_reset(struct sk_buff *rxskb) +static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) { int err; struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; @@ -724,7 +724,7 @@ int dccp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v4_ctl_send_reset(skb); + dccp_v4_ctl_send_reset(sk, skb); discard: kfree_skb(skb); return 0; @@ -913,7 +913,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v4_ctl_send_reset(skb); + dccp_v4_ctl_send_reset(sk, skb); } discard_it: diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index fc326173c215..e0a0607862ef 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -310,7 +310,7 @@ static void dccp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } -static void dccp_v6_ctl_send_reset(struct sk_buff *rxskb) +static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) { struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh; const u32 dccp_hdr_reset_len = sizeof(struct dccp_hdr) + @@ -805,7 +805,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v6_ctl_send_reset(skb); + dccp_v6_ctl_send_reset(sk, skb); discard: if (opt_skb != NULL) __kfree_skb(opt_skb); @@ -902,7 +902,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v6_ctl_send_reset(skb); + dccp_v6_ctl_send_reset(sk, skb); } discard_it: diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 0c49733f5be1..3975048d8094 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -246,7 +246,7 @@ listen_overflow: DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY; drop: if (dccp_hdr(skb)->dccph_type != DCCP_PKT_RESET) - req->rsk_ops->send_reset(skb); + req->rsk_ops->send_reset(sk, skb); inet_csk_reqsk_queue_drop(sk, req, prev); goto out; diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index bc298bcc344e..39e0cb763588 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -618,5 +618,21 @@ config DEFAULT_TCP_CONG default "reno" if DEFAULT_RENO default "cubic" +config TCP_MD5SIG + bool "TCP: MD5 Signature Option support (RFC2385) (EXPERIMENTAL)" + depends on EXPERIMENTAL + select CRYPTO + select CRYPTO_MD5 + ---help--- + RFC2385 specifices a method of giving MD5 protection to TCP sessions. + Its main (only?) use is to protect BGP sessions between core routers + on the Internet. + + If unsure, say N. + +config TCP_MD5SIG_DEBUG + bool "TCP: MD5 Signature Option debugging" + depends on TCP_MD5SIG + source "net/ipv4/ipvs/Kconfig" diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c05e8edaf544..dadef867a3bb 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -258,6 +258,7 @@ #include #include #include +#include #include #include @@ -1942,6 +1943,13 @@ static int do_tcp_setsockopt(struct sock *sk, int level, } break; +#ifdef CONFIG_TCP_MD5SIG + case TCP_MD5SIG: + /* Read the IP->Key mappings from userspace */ + err = tp->af_specific->md5_parse(sk, optval, optlen); + break; +#endif + default: err = -ENOPROTOOPT; break; @@ -2231,6 +2239,135 @@ out: } EXPORT_SYMBOL(tcp_tso_segment); +#ifdef CONFIG_TCP_MD5SIG +static unsigned long tcp_md5sig_users; +static struct tcp_md5sig_pool **tcp_md5sig_pool; +static DEFINE_SPINLOCK(tcp_md5sig_pool_lock); + +static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool) +{ + int cpu; + for_each_possible_cpu(cpu) { + struct tcp_md5sig_pool *p = *per_cpu_ptr(pool, cpu); + if (p) { + if (p->md5_desc.tfm) + crypto_free_hash(p->md5_desc.tfm); + kfree(p); + p = NULL; + } + } + free_percpu(pool); +} + +void tcp_free_md5sig_pool(void) +{ + struct tcp_md5sig_pool **pool = NULL; + + spin_lock(&tcp_md5sig_pool_lock); + if (--tcp_md5sig_users == 0) { + pool = tcp_md5sig_pool; + tcp_md5sig_pool = NULL; + } + spin_unlock(&tcp_md5sig_pool_lock); + if (pool) + __tcp_free_md5sig_pool(pool); +} + +EXPORT_SYMBOL(tcp_free_md5sig_pool); + +struct tcp_md5sig_pool **__tcp_alloc_md5sig_pool(void) +{ + int cpu; + struct tcp_md5sig_pool **pool; + + pool = alloc_percpu(struct tcp_md5sig_pool *); + if (!pool) + return NULL; + + for_each_possible_cpu(cpu) { + struct tcp_md5sig_pool *p; + struct crypto_hash *hash; + + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) + goto out_free; + *per_cpu_ptr(pool, cpu) = p; + + hash = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC); + if (!hash || IS_ERR(hash)) + goto out_free; + + p->md5_desc.tfm = hash; + } + return pool; +out_free: + __tcp_free_md5sig_pool(pool); + return NULL; +} + +struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(void) +{ + struct tcp_md5sig_pool **pool; + int alloc = 0; + +retry: + spin_lock(&tcp_md5sig_pool_lock); + pool = tcp_md5sig_pool; + if (tcp_md5sig_users++ == 0) { + alloc = 1; + spin_unlock(&tcp_md5sig_pool_lock); + } else if (!pool) { + tcp_md5sig_users--; + spin_unlock(&tcp_md5sig_pool_lock); + cpu_relax(); + goto retry; + } else + spin_unlock(&tcp_md5sig_pool_lock); + + if (alloc) { + /* we cannot hold spinlock here because this may sleep. */ + struct tcp_md5sig_pool **p = __tcp_alloc_md5sig_pool(); + spin_lock(&tcp_md5sig_pool_lock); + if (!p) { + tcp_md5sig_users--; + spin_unlock(&tcp_md5sig_pool_lock); + return NULL; + } + pool = tcp_md5sig_pool; + if (pool) { + /* oops, it has already been assigned. */ + spin_unlock(&tcp_md5sig_pool_lock); + __tcp_free_md5sig_pool(p); + } else { + tcp_md5sig_pool = pool = p; + spin_unlock(&tcp_md5sig_pool_lock); + } + } + return pool; +} + +EXPORT_SYMBOL(tcp_alloc_md5sig_pool); + +struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu) +{ + struct tcp_md5sig_pool **p; + spin_lock(&tcp_md5sig_pool_lock); + p = tcp_md5sig_pool; + if (p) + tcp_md5sig_users++; + spin_unlock(&tcp_md5sig_pool_lock); + return (p ? *per_cpu_ptr(p, cpu) : NULL); +} + +EXPORT_SYMBOL(__tcp_get_md5sig_pool); + +void __tcp_put_md5sig_pool(void) { + __tcp_free_md5sig_pool(tcp_md5sig_pool); +} + +EXPORT_SYMBOL(__tcp_put_md5sig_pool); +#endif + extern void __skb_cb_too_small_for_tcp(int, int); extern struct tcp_congestion_ops tcp_reno; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 4a8c96cdec7d..6ab3423674bb 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2677,6 +2677,14 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, opt_rx->sack_ok) { TCP_SKB_CB(skb)->sacked = (ptr - 2) - (unsigned char *)th; } +#ifdef CONFIG_TCP_MD5SIG + case TCPOPT_MD5SIG: + /* + * The MD5 Hash has already been + * checked (see tcp_v{4,6}_do_rcv()). + */ + break; +#endif }; ptr+=opsize-2; length-=opsize; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0ad0904bf56c..8c8e8112f98d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -78,6 +78,9 @@ #include #include +#include +#include + int sysctl_tcp_tw_reuse __read_mostly; int sysctl_tcp_low_latency __read_mostly; @@ -89,6 +92,13 @@ static struct socket *tcp_socket; void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb); +#ifdef CONFIG_TCP_MD5SIG +static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr); +static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + __be32 saddr, __be32 daddr, struct tcphdr *th, + int protocol, int tcplen); +#endif + struct inet_hashinfo __cacheline_aligned tcp_hashinfo = { .lhash_lock = __RW_LOCK_UNLOCKED(tcp_hashinfo.lhash_lock), .lhash_users = ATOMIC_INIT(0), @@ -526,11 +536,19 @@ int tcp_v4_gso_send_check(struct sk_buff *skb) * Exception: precedence violation. We do not implement it in any case. */ -static void tcp_v4_send_reset(struct sk_buff *skb) +static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) { struct tcphdr *th = skb->h.th; - struct tcphdr rth; + struct { + struct tcphdr th; +#ifdef CONFIG_TCP_MD5SIG + u32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; +#endif + } rep; struct ip_reply_arg arg; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif /* Never send a reset in response to a reset. */ if (th->rst) @@ -540,29 +558,50 @@ static void tcp_v4_send_reset(struct sk_buff *skb) return; /* Swap the send and the receive. */ - memset(&rth, 0, sizeof(struct tcphdr)); - rth.dest = th->source; - rth.source = th->dest; - rth.doff = sizeof(struct tcphdr) / 4; - rth.rst = 1; + memset(&rep, 0, sizeof(rep)); + rep.th.dest = th->source; + rep.th.source = th->dest; + rep.th.doff = sizeof(struct tcphdr) / 4; + rep.th.rst = 1; if (th->ack) { - rth.seq = th->ack_seq; + rep.th.seq = th->ack_seq; } else { - rth.ack = 1; - rth.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin + - skb->len - (th->doff << 2)); + rep.th.ack = 1; + rep.th.ack_seq = htonl(ntohl(th->seq) + th->syn + th->fin + + skb->len - (th->doff << 2)); } memset(&arg, 0, sizeof arg); - arg.iov[0].iov_base = (unsigned char *)&rth; - arg.iov[0].iov_len = sizeof rth; + arg.iov[0].iov_base = (unsigned char *)&rep; + arg.iov[0].iov_len = sizeof(rep.th); + +#ifdef CONFIG_TCP_MD5SIG + key = sk ? tcp_v4_md5_do_lookup(sk, skb->nh.iph->daddr) : NULL; + if (key) { + rep.opt[0] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + /* Update length and the length the header thinks exists */ + arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED; + rep.th.doff = arg.iov[0].iov_len / 4; + + tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[1], + key, + skb->nh.iph->daddr, + skb->nh.iph->saddr, + &rep.th, IPPROTO_TCP, + arg.iov[0].iov_len); + } +#endif + arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr, skb->nh.iph->saddr, /*XXX*/ sizeof(struct tcphdr), IPPROTO_TCP, 0); arg.csumoffset = offsetof(struct tcphdr, check) / 2; - ip_send_reply(tcp_socket->sk, skb, &arg, sizeof rth); + ip_send_reply(tcp_socket->sk, skb, &arg, arg.iov[0].iov_len); TCP_INC_STATS_BH(TCP_MIB_OUTSEGS); TCP_INC_STATS_BH(TCP_MIB_OUTRSTS); @@ -572,15 +611,24 @@ static void tcp_v4_send_reset(struct sk_buff *skb) outside socket context is ugly, certainly. What can I do? */ -static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, +static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk, + struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) { struct tcphdr *th = skb->h.th; struct { struct tcphdr th; - u32 tsopt[TCPOLEN_TSTAMP_ALIGNED >> 2]; + u32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) +#ifdef CONFIG_TCP_MD5SIG + + (TCPOLEN_MD5SIG_ALIGNED >> 2) +#endif + ]; } rep; struct ip_reply_arg arg; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; + struct tcp_md5sig_key tw_key; +#endif memset(&rep.th, 0, sizeof(struct tcphdr)); memset(&arg, 0, sizeof arg); @@ -588,12 +636,12 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, arg.iov[0].iov_base = (unsigned char *)&rep; arg.iov[0].iov_len = sizeof(rep.th); if (ts) { - rep.tsopt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | - TCPOLEN_TIMESTAMP); - rep.tsopt[1] = htonl(tcp_time_stamp); - rep.tsopt[2] = htonl(ts); - arg.iov[0].iov_len = sizeof(rep); + rep.opt[0] = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | + TCPOLEN_TIMESTAMP); + rep.opt[1] = htonl(tcp_time_stamp); + rep.opt[2] = htonl(ts); + arg.iov[0].iov_len = TCPOLEN_TSTAMP_ALIGNED; } /* Swap the send and the receive. */ @@ -605,6 +653,44 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, rep.th.ack = 1; rep.th.window = htons(win); +#ifdef CONFIG_TCP_MD5SIG + /* + * The SKB holds an imcoming packet, but may not have a valid ->sk + * pointer. This is especially the case when we're dealing with a + * TIME_WAIT ack, because the sk structure is long gone, and only + * the tcp_timewait_sock remains. So the md5 key is stashed in that + * structure, and we use it in preference. I believe that (twsk || + * skb->sk) holds true, but we program defensively. + */ + if (!twsk && skb->sk) { + key = tcp_v4_md5_do_lookup(skb->sk, skb->nh.iph->daddr); + } else if (twsk && twsk->tw_md5_keylen) { + tw_key.key = twsk->tw_md5_key; + tw_key.keylen = twsk->tw_md5_keylen; + key = &tw_key; + } else { + key = NULL; + } + + if (key) { + int offset = (ts) ? 3 : 0; + + rep.opt[offset++] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + arg.iov[0].iov_len += TCPOLEN_MD5SIG_ALIGNED; + rep.th.doff = arg.iov[0].iov_len/4; + + tcp_v4_do_calc_md5_hash((__u8 *)&rep.opt[offset], + key, + skb->nh.iph->daddr, + skb->nh.iph->saddr, + &rep.th, IPPROTO_TCP, + arg.iov[0].iov_len); + } +#endif + arg.csum = csum_tcpudp_nofold(skb->nh.iph->daddr, skb->nh.iph->saddr, /*XXX*/ arg.iov[0].iov_len, IPPROTO_TCP, 0); @@ -618,9 +704,9 @@ static void tcp_v4_send_ack(struct sk_buff *skb, u32 seq, u32 ack, static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) { struct inet_timewait_sock *tw = inet_twsk(sk); - const struct tcp_timewait_sock *tcptw = tcp_twsk(sk); + struct tcp_timewait_sock *tcptw = tcp_twsk(sk); - tcp_v4_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, + tcp_v4_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale, tcptw->tw_ts_recent); inet_twsk_put(tw); @@ -628,7 +714,8 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, + tcp_v4_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1, + tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -714,6 +801,461 @@ static struct ip_options *tcp_v4_save_options(struct sock *sk, return dopt; } +#ifdef CONFIG_TCP_MD5SIG +/* + * RFC2385 MD5 checksumming requires a mapping of + * IP address->MD5 Key. + * We need to maintain these in the sk structure. + */ + +/* Find the Key structure for an address. */ +static struct tcp_md5sig_key *tcp_v4_md5_do_lookup(struct sock *sk, __be32 addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + if (!tp->md5sig_info || !tp->md5sig_info->entries4) + return NULL; + for (i = 0; i < tp->md5sig_info->entries4; i++) { + if (tp->md5sig_info->keys4[i].addr == addr) + return (struct tcp_md5sig_key *)&tp->md5sig_info->keys4[i]; + } + return NULL; +} + +struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk, + struct sock *addr_sk) +{ + return tcp_v4_md5_do_lookup(sk, inet_sk(addr_sk)->daddr); +} + +EXPORT_SYMBOL(tcp_v4_md5_lookup); + +struct tcp_md5sig_key *tcp_v4_reqsk_md5_lookup(struct sock *sk, + struct request_sock *req) +{ + return tcp_v4_md5_do_lookup(sk, inet_rsk(req)->rmt_addr); +} + +/* This can be called on a newly created socket, from other files */ +int tcp_v4_md5_do_add(struct sock *sk, __be32 addr, + u8 *newkey, u8 newkeylen) +{ + /* Add Key to the list */ + struct tcp4_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + struct tcp4_md5sig_key *keys; + + key = (struct tcp4_md5sig_key *) tcp_v4_md5_do_lookup(sk, addr); + if (key) { + /* Pre-existing entry - just update that one. */ + kfree (key->key); + key->key = newkey; + key->keylen = newkeylen; + } else { + if (!tp->md5sig_info) { + tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC); + if (!tp->md5sig_info) { + kfree(newkey); + return -ENOMEM; + } + } + if (tcp_alloc_md5sig_pool() == NULL) { + kfree(newkey); + return -ENOMEM; + } + if (tp->md5sig_info->alloced4 == tp->md5sig_info->entries4) { + keys = kmalloc((sizeof(struct tcp4_md5sig_key) * + (tp->md5sig_info->entries4 + 1)), GFP_ATOMIC); + if (!keys) { + kfree(newkey); + tcp_free_md5sig_pool(); + return -ENOMEM; + } + + if (tp->md5sig_info->entries4) + memcpy(keys, tp->md5sig_info->keys4, + (sizeof (struct tcp4_md5sig_key) * + tp->md5sig_info->entries4)); + + /* Free old key list, and reference new one */ + if (tp->md5sig_info->keys4) + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = keys; + tp->md5sig_info->alloced4++; + } + tp->md5sig_info->entries4++; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].addr = addr; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].key = newkey; + tp->md5sig_info->keys4[tp->md5sig_info->entries4 - 1].keylen = newkeylen; + } + return 0; +} + +EXPORT_SYMBOL(tcp_v4_md5_do_add); + +static int tcp_v4_md5_add_func(struct sock *sk, struct sock *addr_sk, + u8 *newkey, u8 newkeylen) +{ + return tcp_v4_md5_do_add(sk, inet_sk(addr_sk)->daddr, + newkey, newkeylen); +} + +int tcp_v4_md5_do_del(struct sock *sk, __be32 addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + for (i = 0; i < tp->md5sig_info->entries4; i++) { + if (tp->md5sig_info->keys4[i].addr == addr) { + /* Free the key */ + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4--; + + if (tp->md5sig_info->entries4 == 0) { + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + } else { + /* Need to do some manipulation */ + if (tp->md5sig_info->entries4 != i) + memcpy(&tp->md5sig_info->keys4[i], + &tp->md5sig_info->keys4[i+1], + (tp->md5sig_info->entries4 - i) + * sizeof (struct tcp4_md5sig_key)); + } + tcp_free_md5sig_pool(); + return 0; + } + } + return -ENOENT; +} + +EXPORT_SYMBOL(tcp_v4_md5_do_del); + +static void tcp_v4_clear_md5_list (struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + /* Free each key, then the set of key keys, + * the crypto element, and then decrement our + * hold on the last resort crypto. + */ + if (tp->md5sig_info->entries4) { + int i; + for (i = 0; i < tp->md5sig_info->entries4; i++) + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4 = 0; + tcp_free_md5sig_pool(); + } + if (tp->md5sig_info->keys4) { + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + tp->md5sig_info->alloced4 = 0; + } +} + +static int tcp_v4_parse_md5_keys (struct sock *sk, char __user *optval, + int optlen) +{ + struct tcp_md5sig cmd; + struct sockaddr_in *sin = (struct sockaddr_in *)&cmd.tcpm_addr; + u8 *newkey; + + if (optlen < sizeof(cmd)) + return -EINVAL; + + if (copy_from_user (&cmd, optval, sizeof(cmd))) + return -EFAULT; + + if (sin->sin_family != AF_INET) + return -EINVAL; + + if (!cmd.tcpm_key || !cmd.tcpm_keylen) { + if (!tcp_sk(sk)->md5sig_info) + return -ENOENT; + return tcp_v4_md5_do_del(sk, sin->sin_addr.s_addr); + } + + if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN) + return -EINVAL; + + if (!tcp_sk(sk)->md5sig_info) { + struct tcp_sock *tp = tcp_sk(sk); + struct tcp_md5sig_info *p; + + p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL); + if (!p) + return -EINVAL; + + tp->md5sig_info = p; + + } + + newkey = kmalloc(cmd.tcpm_keylen, GFP_KERNEL); + if (!newkey) + return -ENOMEM; + memcpy(newkey, cmd.tcpm_key, cmd.tcpm_keylen); + return tcp_v4_md5_do_add(sk, sin->sin_addr.s_addr, + newkey, cmd.tcpm_keylen); +} + +static int tcp_v4_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + __be32 saddr, __be32 daddr, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct scatterlist sg[4]; + __u16 data_len; + int block = 0; +#ifdef CONFIG_TCP_MD5SIG_DEBUG + int i; +#endif + __u16 old_checksum; + struct tcp_md5sig_pool *hp; + struct tcp4_pseudohdr *bp; + struct hash_desc *desc; + int err; + unsigned int nbytes = 0; + + /* + * Okay, so RFC2385 is turned on for this connection, + * so we need to generate the MD5 hash for the packet now. + */ + + hp = tcp_get_md5sig_pool(); + if (!hp) + goto clear_hash_noput; + + bp = &hp->md5_blk.ip4; + desc = &hp->md5_desc; + + /* + * 1. the TCP pseudo-header (in the order: source IP address, + * destination IP address, zero-padded protocol number, and + * segment length) + */ + bp->saddr = saddr; + bp->daddr = daddr; + bp->pad = 0; + bp->protocol = protocol; + bp->len = htons(tcplen); + sg_set_buf(&sg[block++], bp, sizeof(*bp)); + nbytes += sizeof(*bp); + +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk("Calcuating hash for: "); + for (i = 0; i < sizeof (*bp); i++) + printk ("%02x ", (unsigned int)((unsigned char *)bp)[i]); + printk(" "); +#endif + + /* 2. the TCP header, excluding options, and assuming a + * checksum of zero/ + */ + old_checksum = th->check; + th->check = 0; + sg_set_buf(&sg[block++], th, sizeof(struct tcphdr)); + nbytes += sizeof(struct tcphdr); +#ifdef CONFIG_TCP_MD5SIG_DEBUG + for (i = 0; i < sizeof (struct tcphdr); i++) + printk (" %02x", (unsigned int)((unsigned char *)th)[i]); +#endif + /* 3. the TCP segment data (if any) */ + data_len = tcplen - (th->doff << 2); + if (data_len > 0) { + unsigned char *data = (unsigned char *)th + (th->doff << 2); + sg_set_buf(&sg[block++], data, data_len); + nbytes += data_len; + } + + /* 4. an independently-specified key or password, known to both + * TCPs and presumably connection-specific + */ + sg_set_buf(&sg[block++], key->key, key->keylen); + nbytes += key->keylen; + +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk (" and password: "); + for (i = 0; i < key->keylen; i++) + printk ("%02x ", (unsigned int)key->key[i]); +#endif + + /* Now store the Hash into the packet */ + err = crypto_hash_init(desc); + if (err) + goto clear_hash; + err = crypto_hash_update(desc, sg, nbytes); + if (err) + goto clear_hash; + err = crypto_hash_final(desc, md5_hash); + if (err) + goto clear_hash; + + /* Reset header, and free up the crypto */ + tcp_put_md5sig_pool(); + th->check = old_checksum; + +out: +#ifdef CONFIG_TCP_MD5SIG_DEBUG + printk(" result:"); + for (i = 0; i < 16; i++) + printk (" %02x", (unsigned int)(((u8*)md5_hash)[i])); + printk("\n"); +#endif + return 0; +clear_hash: + tcp_put_md5sig_pool(); +clear_hash_noput: + memset(md5_hash, 0, 16); + goto out; +} + +int tcp_v4_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, int protocol, + int tcplen) +{ + __be32 saddr, daddr; + + if (sk) { + saddr = inet_sk(sk)->saddr; + daddr = inet_sk(sk)->daddr; + } else { + struct rtable *rt = (struct rtable *)dst; + BUG_ON(!rt); + saddr = rt->rt_src; + daddr = rt->rt_dst; + } + return tcp_v4_do_calc_md5_hash(md5_hash, key, + saddr, daddr, + th, protocol, tcplen); +} + +EXPORT_SYMBOL(tcp_v4_calc_md5_hash); + +static int tcp_v4_inbound_md5_hash (struct sock *sk, struct sk_buff *skb) +{ + /* + * This gets called for each TCP segment that arrives + * so we want to be efficient. + * We have 3 drop cases: + * o No MD5 hash and one expected. + * o MD5 hash and we're not expecting one. + * o MD5 hash and its wrong. + */ + __u8 *hash_location = NULL; + struct tcp_md5sig_key *hash_expected; + struct iphdr *iph = skb->nh.iph; + struct tcphdr *th = skb->h.th; + int length = (th->doff << 2) - sizeof (struct tcphdr); + int genhash; + unsigned char *ptr; + unsigned char newhash[16]; + + hash_expected = tcp_v4_md5_do_lookup(sk, iph->saddr); + + /* + * If the TCP option length is less than the TCP_MD5SIG + * option length, then we can shortcut + */ + if (length < TCPOLEN_MD5SIG) { + if (hash_expected) + return 1; + else + return 0; + } + + /* Okay, we can't shortcut - we have to grub through the options */ + ptr = (unsigned char *)(th + 1); + while (length > 0) { + int opcode = *ptr++; + int opsize; + + switch (opcode) { + case TCPOPT_EOL: + goto done_opts; + case TCPOPT_NOP: + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2) + goto done_opts; + if (opsize > length) + goto done_opts; + + if (opcode == TCPOPT_MD5SIG) { + hash_location = ptr; + goto done_opts; + } + } + ptr += opsize-2; + length -= opsize; + } +done_opts: + /* We've parsed the options - do we have a hash? */ + if (!hash_expected && !hash_location) + return 0; + + if (hash_expected && !hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest)); + } + return 1; + } + + if (!hash_expected && hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest)); + } + return 1; + } + + /* Okay, so this is hash_expected and hash_location - + * so we need to calculate the checksum. + */ + genhash = tcp_v4_do_calc_md5_hash(newhash, + hash_expected, + iph->saddr, iph->daddr, + th, sk->sk_protocol, + skb->len); + + if (genhash || memcmp(hash_location, newhash, 16) != 0) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash failed for " + "(" NIPQUAD_FMT ", %d)->(" NIPQUAD_FMT ", %d)%s\n", + NIPQUAD (iph->saddr), ntohs(th->source), + NIPQUAD (iph->daddr), ntohs(th->dest), + genhash ? " tcp_v4_calc_md5_hash failed" : ""); +#ifdef CONFIG_TCP_MD5SIG_DEBUG + do { + int i; + printk("Received: "); + for (i = 0; i < 16; i++) + printk("%02x ", 0xff & (int)hash_location[i]); + printk("\n"); + printk("Calculated: "); + for (i = 0; i < 16; i++) + printk("%02x ", 0xff & (int)newhash[i]); + printk("\n"); + } while(0); +#endif + } + return 1; + } + return 0; +} + +#endif + struct request_sock_ops tcp_request_sock_ops __read_mostly = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), @@ -723,9 +1265,16 @@ struct request_sock_ops tcp_request_sock_ops __read_mostly = { .send_reset = tcp_v4_send_reset, }; +struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_reqsk_md5_lookup, +#endif +}; + static struct timewait_sock_ops tcp_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp_timewait_sock), .twsk_unique = tcp_twsk_unique, + .twsk_destructor= tcp_twsk_destructor, }; int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) @@ -773,6 +1322,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (!req) goto drop; +#ifdef CONFIG_TCP_MD5SIG + tcp_rsk(req)->af_specific = &tcp_request_sock_ipv4_ops; +#endif + tcp_clear_options(&tmp_opt); tmp_opt.mss_clamp = 536; tmp_opt.user_mss = tcp_sk(sk)->rx_opt.user_mss; @@ -891,6 +1444,9 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (sk_acceptq_is_full(sk)) goto exit_overflow; @@ -925,6 +1481,24 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newtp->advmss = dst_metric(dst, RTAX_ADVMSS); tcp_initialize_rcv_mss(newsk); +#ifdef CONFIG_TCP_MD5SIG + /* Copy over the MD5 key from the original socket */ + if ((key = tcp_v4_md5_do_lookup(sk, newinet->daddr)) != NULL) { + /* + * We're using one, so create a matching key + * on the newsk structure. If we fail to get + * memory, then we end up not copying the key + * across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + memcpy(newkey, key->key, key->keylen); + tcp_v4_md5_do_add(newsk, inet_sk(sk)->daddr, + newkey, key->keylen); + } + } +#endif + __inet_hash(&tcp_hashinfo, newsk, 0); __inet_inherit_port(&tcp_hashinfo, sk, newsk); @@ -1000,10 +1574,24 @@ static int tcp_v4_checksum_init(struct sk_buff *skb) */ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) { + struct sock *rsk; +#ifdef CONFIG_TCP_MD5SIG + /* + * We really want to reject the packet as early as possible + * if: + * o We're expecting an MD5'd packet and this is no MD5 tcp option + * o There is an MD5 option and we're not expecting one + */ + if (tcp_v4_inbound_md5_hash (sk, skb)) + goto discard; +#endif + if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ TCP_CHECK_TIMER(sk); - if (tcp_rcv_established(sk, skb, skb->h.th, skb->len)) + if (tcp_rcv_established(sk, skb, skb->h.th, skb->len)) { + rsk = sk; goto reset; + } TCP_CHECK_TIMER(sk); return 0; } @@ -1017,20 +1605,24 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) goto discard; if (nsk != sk) { - if (tcp_child_process(sk, nsk, skb)) + if (tcp_child_process(sk, nsk, skb)) { + rsk = nsk; goto reset; + } return 0; } } TCP_CHECK_TIMER(sk); - if (tcp_rcv_state_process(sk, skb, skb->h.th, skb->len)) + if (tcp_rcv_state_process(sk, skb, skb->h.th, skb->len)) { + rsk = sk; goto reset; + } TCP_CHECK_TIMER(sk); return 0; reset: - tcp_v4_send_reset(skb); + tcp_v4_send_reset(rsk, skb); discard: kfree_skb(skb); /* Be careful here. If this function gets more complicated and @@ -1139,7 +1731,7 @@ no_tcp_socket: bad_packet: TCP_INC_STATS_BH(TCP_MIB_INERRS); } else { - tcp_v4_send_reset(skb); + tcp_v4_send_reset(NULL, skb); } discard_it: @@ -1262,6 +1854,15 @@ struct inet_connection_sock_af_ops ipv4_specific = { #endif }; +struct tcp_sock_af_ops tcp_sock_ipv4_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_md5_lookup, + .calc_md5_hash = tcp_v4_calc_md5_hash, + .md5_add = tcp_v4_md5_add_func, + .md5_parse = tcp_v4_parse_md5_keys, +#endif +}; + /* NOTE: A lot of things set to zero explicitly by call to * sk_alloc() so need not be done here. */ @@ -1301,6 +1902,9 @@ static int tcp_v4_init_sock(struct sock *sk) icsk->icsk_af_ops = &ipv4_specific; icsk->icsk_sync_mss = tcp_sync_mss; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv4_specific; +#endif sk->sk_sndbuf = sysctl_tcp_wmem[1]; sk->sk_rcvbuf = sysctl_tcp_rmem[1]; @@ -1324,6 +1928,15 @@ int tcp_v4_destroy_sock(struct sock *sk) /* Cleans up our, hopefully empty, out_of_order_queue. */ __skb_queue_purge(&tp->out_of_order_queue); +#ifdef CONFIG_TCP_MD5SIG + /* Clean up the MD5 key list, if any */ + if (tp->md5sig_info) { + tcp_v4_clear_md5_list(sk); + kfree(tp->md5sig_info); + tp->md5sig_info = NULL; + } +#endif + #ifdef CONFIG_NET_DMA /* Cleans up our sk_async_wait_queue */ __skb_queue_purge(&sk->sk_async_wait_queue); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0163d9826907..ac55d8892cf1 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -306,6 +306,28 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) tw->tw_ipv6only = np->ipv6only; } #endif + +#ifdef CONFIG_TCP_MD5SIG + /* + * The timewait bucket does not have the key DB from the + * sock structure. We just make a quick copy of the + * md5 key being used (if indeed we are using one) + * so the timewait ack generating code has the key. + */ + do { + struct tcp_md5sig_key *key; + memset(tcptw->tw_md5_key, 0, sizeof(tcptw->tw_md5_key)); + tcptw->tw_md5_keylen = 0; + key = tp->af_specific->md5_lookup(sk, sk); + if (key != NULL) { + memcpy(&tcptw->tw_md5_key, key->key, key->keylen); + tcptw->tw_md5_keylen = key->keylen; + if (tcp_alloc_md5sig_pool() == NULL) + BUG(); + } + } while(0); +#endif + /* Linkage updates. */ __inet_twsk_hashdance(tw, sk, &tcp_hashinfo); @@ -337,6 +359,17 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) tcp_done(sk); } +void tcp_twsk_destructor(struct sock *sk) +{ + struct tcp_timewait_sock *twsk = tcp_twsk(sk); +#ifdef CONFIG_TCP_MD5SIG + if (twsk->tw_md5_keylen) + tcp_put_md5sig_pool(); +#endif +} + +EXPORT_SYMBOL_GPL(tcp_twsk_destructor); + /* This is not only more efficient than what we used to do, it eliminates * a lot of code duplication between IPv4/IPv6 SYN recv processing. -DaveM * @@ -435,6 +468,11 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->rx_opt.ts_recent_stamp = 0; newtp->tcp_header_len = sizeof(struct tcphdr); } +#ifdef CONFIG_TCP_MD5SIG + newtp->md5sig_info = NULL; /*XXX*/ + if (newtp->af_specific->md5_lookup(sk, newsk)) + newtp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED; +#endif if (skb->len >= TCP_MIN_RCVMSS+newtp->tcp_header_len) newicsk->icsk_ack.last_seg_size = skb->len - newtp->tcp_header_len; newtp->rx_opt.mss_clamp = req->mss; @@ -617,6 +655,30 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, req, NULL); if (child == NULL) goto listen_overflow; +#ifdef CONFIG_TCP_MD5SIG + else { + /* Copy over the MD5 key from the original socket */ + struct tcp_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + key = tp->af_specific->md5_lookup(sk, child); + if (key != NULL) { + /* + * We're using one, so create a matching key on the + * newsk structure. If we fail to get memory then we + * end up not copying the key across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + if (!tcp_alloc_md5sig_pool()) + BUG(); + memcpy(newkey, key->key, key->keylen); + tp->af_specific->md5_add(child, child, + newkey, + key->keylen); + } + } + } +#endif inet_csk_reqsk_queue_unlink(sk, req, prev); inet_csk_reqsk_queue_removed(sk, req); @@ -633,7 +695,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, embryonic_reset: NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS); if (!(flg & TCP_FLAG_RST)) - req->rsk_ops->send_reset(skb); + req->rsk_ops->send_reset(sk, skb); inet_csk_reqsk_queue_drop(sk, req, prev); return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 6a8581ab9a23..32c1a972fa31 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -270,7 +270,7 @@ static u16 tcp_select_window(struct sock *sk) } static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp, - __u32 tstamp) + __u32 tstamp, __u8 **md5_hash) { if (tp->rx_opt.tstamp_ok) { *ptr++ = htonl((TCPOPT_NOP << 24) | @@ -298,16 +298,29 @@ static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp, tp->rx_opt.eff_sacks--; } } +#ifdef CONFIG_TCP_MD5SIG + if (md5_hash) { + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + *md5_hash = (__u8 *)ptr; + } +#endif } /* Construct a tcp options header for a SYN or SYN_ACK packet. * If this is every changed make sure to change the definition of * MAX_SYN_SIZE to match the new maximum number of options that you * can generate. + * + * Note - that with the RFC2385 TCP option, we make room for the + * 16 byte MD5 hash. This will be filled in later, so the pointer for the + * location to be filled is passed back up. */ static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack, int offer_wscale, int wscale, __u32 tstamp, - __u32 ts_recent) + __u32 ts_recent, __u8 **md5_hash) { /* We always get an MSS option. * The option bytes which will be seen in normal data @@ -346,6 +359,20 @@ static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack, (TCPOPT_WINDOW << 16) | (TCPOLEN_WINDOW << 8) | (wscale)); +#ifdef CONFIG_TCP_MD5SIG + /* + * If MD5 is enabled, then we set the option, and include the size + * (always 18). The actual MD5 hash is added just before the + * packet is sent. + */ + if (md5_hash) { + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + *md5_hash = (__u8 *) ptr; + } +#endif } /* This routine actually transmits TCP packets queued in by @@ -366,6 +393,10 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, struct tcp_sock *tp; struct tcp_skb_cb *tcb; int tcp_header_size; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *md5; + __u8 *md5_hash_location; +#endif struct tcphdr *th; int sysctl_flags; int err; @@ -424,6 +455,16 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, if (tcp_packets_in_flight(tp) == 0) tcp_ca_event(sk, CA_EVENT_TX_START); +#ifdef CONFIG_TCP_MD5SIG + /* + * Are we doing MD5 on this segment? If so - make + * room for it. + */ + md5 = tp->af_specific->md5_lookup(sk, sk); + if (md5) + tcp_header_size += TCPOLEN_MD5SIG_ALIGNED; +#endif + th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; @@ -460,13 +501,34 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, (sysctl_flags & SYSCTL_FLAG_WSCALE), tp->rx_opt.rcv_wscale, tcb->when, - tp->rx_opt.ts_recent); + tp->rx_opt.ts_recent, + +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL); } else { tcp_build_and_update_options((__be32 *)(th + 1), - tp, tcb->when); + tp, tcb->when, +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL); TCP_ECN_send(sk, tp, skb, tcp_header_size); } +#ifdef CONFIG_TCP_MD5SIG + /* Calculate the MD5 hash, as we have all we need now */ + if (md5) { + tp->af_specific->calc_md5_hash(md5_hash_location, + md5, + sk, NULL, NULL, + skb->h.th, + sk->sk_protocol, + skb->len); + } +#endif + icsk->icsk_af_ops->send_check(sk, skb->len, skb); if (likely(tcb->flags & TCPCB_FLAG_ACK)) @@ -840,6 +902,11 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed) mss_now -= (TCPOLEN_SACK_BASE_ALIGNED + (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK)); +#ifdef CONFIG_TCP_MD5SIG + if (tp->af_specific->md5_lookup(sk, sk)) + mss_now -= TCPOLEN_MD5SIG_ALIGNED; +#endif + xmit_size_goal = mss_now; if (doing_tso) { @@ -2033,6 +2100,10 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, struct tcphdr *th; int tcp_header_size; struct sk_buff *skb; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *md5; + __u8 *md5_hash_location; +#endif skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15, 1, GFP_ATOMIC); if (skb == NULL) @@ -2048,6 +2119,13 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); + +#ifdef CONFIG_TCP_MD5SIG + /* Are we doing MD5 on this segment? If so - make room for it */ + md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req); + if (md5) + tcp_header_size += TCPOLEN_MD5SIG_ALIGNED; +#endif skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); @@ -2085,11 +2163,29 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, - req->ts_recent); + req->ts_recent, + ( +#ifdef CONFIG_TCP_MD5SIG + md5 ? &md5_hash_location : +#endif + NULL) + ); skb->csum = 0; th->doff = (tcp_header_size >> 2); TCP_INC_STATS(TCP_MIB_OUTSEGS); + +#ifdef CONFIG_TCP_MD5SIG + /* Okay, we have all we need - do the md5 hash if needed */ + if (md5) { + tp->af_specific->calc_md5_hash(md5_hash_location, + md5, + NULL, dst, req, + skb->h.th, sk->sk_protocol, + skb->len); + } +#endif + return skb; } @@ -2108,6 +2204,11 @@ static void tcp_connect_init(struct sock *sk) tp->tcp_header_len = sizeof(struct tcphdr) + (sysctl_tcp_timestamps ? TCPOLEN_TSTAMP_ALIGNED : 0); +#ifdef CONFIG_TCP_MD5SIG + if (tp->af_specific->md5_lookup(sk, sk) != NULL) + tp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED; +#endif + /* If user gave his TCP_MAXSEG, record it to clamp */ if (tp->rx_opt.user_mss) tp->rx_opt.mss_clamp = tp->rx_opt.user_mss; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9a88395a7629..663d1d238014 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -66,10 +66,13 @@ #include #include +#include +#include + /* Socket used for sending RSTs and ACKs */ static struct socket *tcp6_socket; -static void tcp_v6_send_reset(struct sk_buff *skb); +static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb); static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb); @@ -78,6 +81,8 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb); static struct inet_connection_sock_af_ops ipv6_mapped; static struct inet_connection_sock_af_ops ipv6_specific; +static struct tcp_sock_af_ops tcp_sock_ipv6_specific; +static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific; static int tcp_v6_get_port(struct sock *sk, unsigned short snum) { @@ -208,6 +213,9 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, icsk->icsk_af_ops = &ipv6_mapped; sk->sk_backlog_rcv = tcp_v4_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_mapped_specific; +#endif err = tcp_v4_connect(sk, (struct sockaddr *)&sin, sizeof(sin)); @@ -215,6 +223,9 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, icsk->icsk_ext_hdr_len = exthdrlen; icsk->icsk_af_ops = &ipv6_specific; sk->sk_backlog_rcv = tcp_v6_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_specific; +#endif goto failure; } else { ipv6_addr_set(&np->saddr, 0, 0, htonl(0x0000FFFF), @@ -518,6 +529,396 @@ static void tcp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet6_rsk(req)->pktopts); } +#ifdef CONFIG_TCP_MD5SIG +static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(struct sock *sk, + struct in6_addr *addr) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + BUG_ON(tp == NULL); + + if (!tp->md5sig_info || !tp->md5sig_info->entries6) + return NULL; + + for (i = 0; i < tp->md5sig_info->entries6; i++) { + if (ipv6_addr_cmp(&tp->md5sig_info->keys6[i].addr, addr) == 0) + return (struct tcp_md5sig_key *)&tp->md5sig_info->keys6[i]; + } + return NULL; +} + +static struct tcp_md5sig_key *tcp_v6_md5_lookup(struct sock *sk, + struct sock *addr_sk) +{ + return tcp_v6_md5_do_lookup(sk, &inet6_sk(addr_sk)->daddr); +} + +static struct tcp_md5sig_key *tcp_v6_reqsk_md5_lookup(struct sock *sk, + struct request_sock *req) +{ + return tcp_v6_md5_do_lookup(sk, &inet6_rsk(req)->rmt_addr); +} + +static int tcp_v6_md5_do_add(struct sock *sk, struct in6_addr *peer, + char *newkey, u8 newkeylen) +{ + /* Add key to the list */ + struct tcp6_md5sig_key *key; + struct tcp_sock *tp = tcp_sk(sk); + struct tcp6_md5sig_key *keys; + + key = (struct tcp6_md5sig_key*) tcp_v6_md5_do_lookup(sk, peer); + if (key) { + /* modify existing entry - just update that one */ + kfree(key->key); + key->key = newkey; + key->keylen = newkeylen; + } else { + /* reallocate new list if current one is full. */ + if (!tp->md5sig_info) { + tp->md5sig_info = kzalloc(sizeof(*tp->md5sig_info), GFP_ATOMIC); + if (!tp->md5sig_info) { + kfree(newkey); + return -ENOMEM; + } + } + tcp_alloc_md5sig_pool(); + if (tp->md5sig_info->alloced6 == tp->md5sig_info->entries6) { + keys = kmalloc((sizeof (tp->md5sig_info->keys6[0]) * + (tp->md5sig_info->entries6 + 1)), GFP_ATOMIC); + + if (!keys) { + tcp_free_md5sig_pool(); + kfree(newkey); + return -ENOMEM; + } + + if (tp->md5sig_info->entries6) + memmove(keys, tp->md5sig_info->keys6, + (sizeof (tp->md5sig_info->keys6[0]) * + tp->md5sig_info->entries6)); + + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = keys; + tp->md5sig_info->alloced6++; + } + + ipv6_addr_copy(&tp->md5sig_info->keys6[tp->md5sig_info->entries6].addr, + peer); + tp->md5sig_info->keys6[tp->md5sig_info->entries6].key = newkey; + tp->md5sig_info->keys6[tp->md5sig_info->entries6].keylen = newkeylen; + + tp->md5sig_info->entries6++; + } + return 0; +} + +static int tcp_v6_md5_add_func(struct sock *sk, struct sock *addr_sk, + u8 *newkey, __u8 newkeylen) +{ + return tcp_v6_md5_do_add(sk, &inet6_sk(addr_sk)->daddr, + newkey, newkeylen); +} + +static int tcp_v6_md5_do_del(struct sock *sk, struct in6_addr *peer) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + for (i = 0; i < tp->md5sig_info->entries6; i++) { + if (ipv6_addr_cmp(&tp->md5sig_info->keys6[i].addr, peer) == 0) { + /* Free the key */ + kfree(tp->md5sig_info->keys6[i].key); + tp->md5sig_info->entries6--; + + if (tp->md5sig_info->entries6 == 0) { + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = NULL; + + tcp_free_md5sig_pool(); + + return 0; + } else { + /* shrink the database */ + if (tp->md5sig_info->entries6 != i) + memmove(&tp->md5sig_info->keys6[i], + &tp->md5sig_info->keys6[i+1], + (tp->md5sig_info->entries6 - i) + * sizeof (tp->md5sig_info->keys6[0])); + } + } + } + return -ENOENT; +} + +static void tcp_v6_clear_md5_list (struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + int i; + + if (tp->md5sig_info->entries6) { + for (i = 0; i < tp->md5sig_info->entries6; i++) + kfree(tp->md5sig_info->keys6[i].key); + tp->md5sig_info->entries6 = 0; + tcp_free_md5sig_pool(); + } + + kfree(tp->md5sig_info->keys6); + tp->md5sig_info->keys6 = NULL; + tp->md5sig_info->alloced6 = 0; + + if (tp->md5sig_info->entries4) { + for (i = 0; i < tp->md5sig_info->entries4; i++) + kfree(tp->md5sig_info->keys4[i].key); + tp->md5sig_info->entries4 = 0; + tcp_free_md5sig_pool(); + } + + kfree(tp->md5sig_info->keys4); + tp->md5sig_info->keys4 = NULL; + tp->md5sig_info->alloced4 = 0; +} + +static int tcp_v6_parse_md5_keys (struct sock *sk, char __user *optval, + int optlen) +{ + struct tcp_md5sig cmd; + struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&cmd.tcpm_addr; + u8 *newkey; + + if (optlen < sizeof(cmd)) + return -EINVAL; + + if (copy_from_user(&cmd, optval, sizeof(cmd))) + return -EFAULT; + + if (sin6->sin6_family != AF_INET6) + return -EINVAL; + + if (!cmd.tcpm_keylen) { + if (!tcp_sk(sk)->md5sig_info) + return -ENOENT; + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_MAPPED) + return tcp_v4_md5_do_del(sk, sin6->sin6_addr.s6_addr32[3]); + return tcp_v6_md5_do_del(sk, &sin6->sin6_addr); + } + + if (cmd.tcpm_keylen > TCP_MD5SIG_MAXKEYLEN) + return -EINVAL; + + if (!tcp_sk(sk)->md5sig_info) { + struct tcp_sock *tp = tcp_sk(sk); + struct tcp_md5sig_info *p; + + p = kzalloc(sizeof(struct tcp_md5sig_info), GFP_KERNEL); + if (!p) + return -ENOMEM; + + tp->md5sig_info = p; + } + + newkey = kmalloc(cmd.tcpm_keylen, GFP_KERNEL); + if (!newkey) + return -ENOMEM; + memcpy(newkey, cmd.tcpm_key, cmd.tcpm_keylen); + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_MAPPED) { + return tcp_v4_md5_do_add(sk, sin6->sin6_addr.s6_addr32[3], + newkey, cmd.tcpm_keylen); + } + return tcp_v6_md5_do_add(sk, &sin6->sin6_addr, newkey, cmd.tcpm_keylen); +} + +static int tcp_v6_do_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct in6_addr *saddr, + struct in6_addr *daddr, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct scatterlist sg[4]; + __u16 data_len; + int block = 0; + __u16 cksum; + struct tcp_md5sig_pool *hp; + struct tcp6_pseudohdr *bp; + struct hash_desc *desc; + int err; + unsigned int nbytes = 0; + + hp = tcp_get_md5sig_pool(); + if (!hp) { + printk(KERN_WARNING "%s(): hash pool not found...\n", __FUNCTION__); + goto clear_hash_noput; + } + bp = &hp->md5_blk.ip6; + desc = &hp->md5_desc; + + /* 1. TCP pseudo-header (RFC2460) */ + ipv6_addr_copy(&bp->saddr, saddr); + ipv6_addr_copy(&bp->daddr, daddr); + bp->len = htonl(tcplen); + bp->protocol = htonl(protocol); + + sg_set_buf(&sg[block++], bp, sizeof(*bp)); + nbytes += sizeof(*bp); + + /* 2. TCP header, excluding options */ + cksum = th->check; + th->check = 0; + sg_set_buf(&sg[block++], th, sizeof(*th)); + nbytes += sizeof(*th); + + /* 3. TCP segment data (if any) */ + data_len = tcplen - (th->doff << 2); + if (data_len > 0) { + u8 *data = (u8 *)th + (th->doff << 2); + sg_set_buf(&sg[block++], data, data_len); + nbytes += data_len; + } + + /* 4. shared key */ + sg_set_buf(&sg[block++], key->key, key->keylen); + nbytes += key->keylen; + + /* Now store the hash into the packet */ + err = crypto_hash_init(desc); + if (err) { + printk(KERN_WARNING "%s(): hash_init failed\n", __FUNCTION__); + goto clear_hash; + } + err = crypto_hash_update(desc, sg, nbytes); + if (err) { + printk(KERN_WARNING "%s(): hash_update failed\n", __FUNCTION__); + goto clear_hash; + } + err = crypto_hash_final(desc, md5_hash); + if (err) { + printk(KERN_WARNING "%s(): hash_final failed\n", __FUNCTION__); + goto clear_hash; + } + + /* Reset header, and free up the crypto */ + tcp_put_md5sig_pool(); + th->check = cksum; +out: + return 0; +clear_hash: + tcp_put_md5sig_pool(); +clear_hash_noput: + memset(md5_hash, 0, 16); + goto out; +} + +static int tcp_v6_calc_md5_hash(char *md5_hash, struct tcp_md5sig_key *key, + struct sock *sk, + struct dst_entry *dst, + struct request_sock *req, + struct tcphdr *th, int protocol, + int tcplen) +{ + struct in6_addr *saddr, *daddr; + + if (sk) { + saddr = &inet6_sk(sk)->saddr; + daddr = &inet6_sk(sk)->daddr; + } else { + saddr = &inet6_rsk(req)->loc_addr; + daddr = &inet6_rsk(req)->rmt_addr; + } + return tcp_v6_do_calc_md5_hash(md5_hash, key, + saddr, daddr, + th, protocol, tcplen); +} + +static int tcp_v6_inbound_md5_hash (struct sock *sk, struct sk_buff *skb) +{ + __u8 *hash_location = NULL; + struct tcp_md5sig_key *hash_expected; + struct ipv6hdr *ip6h = skb->nh.ipv6h; + struct tcphdr *th = skb->h.th; + int length = (th->doff << 2) - sizeof (*th); + int genhash; + u8 *ptr; + u8 newhash[16]; + + hash_expected = tcp_v6_md5_do_lookup(sk, &ip6h->saddr); + + /* If the TCP option is too short, we can short cut */ + if (length < TCPOLEN_MD5SIG) + return hash_expected ? 1 : 0; + + /* parse options */ + ptr = (u8*)(th + 1); + while (length > 0) { + int opcode = *ptr++; + int opsize; + + switch(opcode) { + case TCPOPT_EOL: + goto done_opts; + case TCPOPT_NOP: + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2 || opsize > length) + goto done_opts; + if (opcode == TCPOPT_MD5SIG) { + hash_location = ptr; + goto done_opts; + } + } + ptr += opsize - 2; + length -= opsize; + } + +done_opts: + /* do we have a hash as expected? */ + if (!hash_expected) { + if (!hash_location) + return 0; + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash NOT expected but found " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + + if (!hash_location) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash expected but NOT found " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + + /* check the signature */ + genhash = tcp_v6_do_calc_md5_hash(newhash, + hash_expected, + &ip6h->saddr, &ip6h->daddr, + th, sk->sk_protocol, + skb->len); + if (genhash || memcmp(hash_location, newhash, 16) != 0) { + if (net_ratelimit()) { + printk(KERN_INFO "MD5 Hash %s for " + "(" NIP6_FMT ", %u)->" + "(" NIP6_FMT ", %u)\n", + genhash ? "failed" : "mismatch", + NIP6(ip6h->saddr), ntohs(th->source), + NIP6(ip6h->daddr), ntohs(th->dest)); + } + return 1; + } + return 0; +} +#endif + static struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), @@ -527,9 +928,16 @@ static struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .send_reset = tcp_v6_send_reset }; +struct tcp_request_sock_ops tcp_request_sock_ipv6_ops = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v6_reqsk_md5_lookup, +#endif +}; + static struct timewait_sock_ops tcp6_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp6_timewait_sock), .twsk_unique = tcp_twsk_unique, + .twsk_destructor= tcp_twsk_destructor, }; static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) @@ -566,11 +974,15 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb) return 0; } -static void tcp_v6_send_reset(struct sk_buff *skb) +static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb) { struct tcphdr *th = skb->h.th, *t1; struct sk_buff *buff; struct flowi fl; + int tot_len = sizeof(*th); +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (th->rst) return; @@ -578,25 +990,35 @@ static void tcp_v6_send_reset(struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) return; +#ifdef CONFIG_TCP_MD5SIG + if (sk) + key = tcp_v6_md5_do_lookup(sk, &skb->nh.ipv6h->daddr); + else + key = NULL; + + if (key) + tot_len += TCPOLEN_MD5SIG_ALIGNED; +#endif + /* * We need to grab some memory, and put together an RST, * and then put it into the queue to be sent. */ - buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + sizeof(struct tcphdr), + buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len, GFP_ATOMIC); if (buff == NULL) return; - skb_reserve(buff, MAX_HEADER + sizeof(struct ipv6hdr) + sizeof(struct tcphdr)); + skb_reserve(buff, MAX_HEADER + sizeof(struct ipv6hdr) + tot_len); - t1 = (struct tcphdr *) skb_push(buff,sizeof(struct tcphdr)); + t1 = (struct tcphdr *) skb_push(buff, tot_len); /* Swap the send and the receive. */ memset(t1, 0, sizeof(*t1)); t1->dest = th->source; t1->source = th->dest; - t1->doff = sizeof(*t1)/4; + t1->doff = tot_len / 4; t1->rst = 1; if(th->ack) { @@ -607,6 +1029,22 @@ static void tcp_v6_send_reset(struct sk_buff *skb) + skb->len - (th->doff<<2)); } +#ifdef CONFIG_TCP_MD5SIG + if (key) { + u32 *opt = (u32*)(t1 + 1); + opt[0] = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | + TCPOLEN_MD5SIG); + tcp_v6_do_calc_md5_hash((__u8*)&opt[1], + key, + &skb->nh.ipv6h->daddr, + &skb->nh.ipv6h->saddr, + t1, IPPROTO_TCP, + tot_len); + } +#endif + buff->csum = csum_partial((char *)t1, sizeof(*t1), 0); memset(&fl, 0, sizeof(fl)); @@ -637,15 +1075,37 @@ static void tcp_v6_send_reset(struct sk_buff *skb) kfree_skb(buff); } -static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) +static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, + struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 ts) { struct tcphdr *th = skb->h.th, *t1; struct sk_buff *buff; struct flowi fl; int tot_len = sizeof(struct tcphdr); + u32 *topt; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; + struct tcp_md5sig_key tw_key; +#endif + +#ifdef CONFIG_TCP_MD5SIG + if (!tw && skb->sk) { + key = tcp_v6_md5_do_lookup(skb->sk, &skb->nh.ipv6h->daddr); + } else if (tw && tw->tw_md5_keylen) { + tw_key.key = tw->tw_md5_key; + tw_key.keylen = tw->tw_md5_keylen; + key = &tw_key; + } else { + key = NULL; + } +#endif if (ts) tot_len += TCPOLEN_TSTAMP_ALIGNED; +#ifdef CONFIG_TCP_MD5SIG + if (key) + tot_len += TCPOLEN_MD5SIG_ALIGNED; +#endif buff = alloc_skb(MAX_HEADER + sizeof(struct ipv6hdr) + tot_len, GFP_ATOMIC); @@ -665,15 +1125,29 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 t1->ack_seq = htonl(ack); t1->ack = 1; t1->window = htons(win); + + topt = (u32*)(t1 + 1); if (ts) { - u32 *ptr = (u32*)(t1 + 1); - *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | - (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); - *ptr++ = htonl(tcp_time_stamp); - *ptr = htonl(ts); + *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP); + *topt++ = htonl(tcp_time_stamp); + *topt = htonl(ts); } +#ifdef CONFIG_TCP_MD5SIG + if (key) { + *topt++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | + (TCPOPT_MD5SIG << 8) | TCPOLEN_MD5SIG); + tcp_v6_do_calc_md5_hash((__u8 *)topt, + key, + &skb->nh.ipv6h->daddr, + &skb->nh.ipv6h->saddr, + t1, IPPROTO_TCP, + tot_len); + } +#endif + buff->csum = csum_partial((char *)t1, tot_len, 0); memset(&fl, 0, sizeof(fl)); @@ -704,9 +1178,9 @@ static void tcp_v6_send_ack(struct sk_buff *skb, u32 seq, u32 ack, u32 win, u32 static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) { struct inet_timewait_sock *tw = inet_twsk(sk); - const struct tcp_timewait_sock *tcptw = tcp_twsk(sk); + struct tcp_timewait_sock *tcptw = tcp_twsk(sk); - tcp_v6_send_ack(skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, + tcp_v6_send_ack(tcptw, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt, tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale, tcptw->tw_ts_recent); @@ -715,7 +1189,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { - tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); + tcp_v6_send_ack(NULL, skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -786,6 +1260,10 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (req == NULL) goto drop; +#ifdef CONFIG_TCP_MD5SIG + tcp_rsk(req)->af_specific = &tcp_request_sock_ipv6_ops; +#endif + tcp_clear_options(&tmp_opt); tmp_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr); tmp_opt.user_mss = tp->rx_opt.user_mss; @@ -844,6 +1322,9 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct tcp_sock *newtp; struct sock *newsk; struct ipv6_txoptions *opt; +#ifdef CONFIG_TCP_MD5SIG + struct tcp_md5sig_key *key; +#endif if (skb->protocol == htons(ETH_P_IP)) { /* @@ -874,6 +1355,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, inet_csk(newsk)->icsk_af_ops = &ipv6_mapped; newsk->sk_backlog_rcv = tcp_v4_do_rcv; +#ifdef CONFIG_TCP_MD5SIG + newtp->af_specific = &tcp_sock_ipv6_mapped_specific; +#endif + newnp->pktoptions = NULL; newnp->opt = NULL; newnp->mcast_oif = inet6_iif(skb); @@ -1008,6 +1493,23 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newinet->daddr = newinet->saddr = newinet->rcv_saddr = LOOPBACK4_IPV6; +#ifdef CONFIG_TCP_MD5SIG + /* Copy over the MD5 key from the original socket */ + if ((key = tcp_v6_md5_do_lookup(sk, &newnp->daddr)) != NULL) { + /* We're using one, so create a matching key + * on the newsk structure. If we fail to get + * memory, then we end up not copying the key + * across. Shucks. + */ + char *newkey = kmalloc(key->keylen, GFP_ATOMIC); + if (newkey) { + memcpy(newkey, key->key, key->keylen); + tcp_v6_md5_do_add(newsk, &inet6_sk(sk)->daddr, + newkey, key->keylen); + } + } +#endif + __inet6_hash(&tcp_hashinfo, newsk); inet_inherit_port(&tcp_hashinfo, sk, newsk); @@ -1067,6 +1569,11 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) if (skb->protocol == htons(ETH_P_IP)) return tcp_v4_do_rcv(sk, skb); +#ifdef CONFIG_TCP_MD5SIG + if (tcp_v6_inbound_md5_hash (sk, skb)) + goto discard; +#endif + if (sk_filter(sk, skb)) goto discard; @@ -1132,7 +1639,7 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v6_send_reset(skb); + tcp_v6_send_reset(sk, skb); discard: if (opt_skb) __kfree_skb(opt_skb); @@ -1257,7 +1764,7 @@ no_tcp_socket: bad_packet: TCP_INC_STATS_BH(TCP_MIB_INERRS); } else { - tcp_v6_send_reset(skb); + tcp_v6_send_reset(NULL, skb); } discard_it: @@ -1336,6 +1843,15 @@ static struct inet_connection_sock_af_ops ipv6_specific = { #endif }; +static struct tcp_sock_af_ops tcp_sock_ipv6_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v6_md5_lookup, + .calc_md5_hash = tcp_v6_calc_md5_hash, + .md5_add = tcp_v6_md5_add_func, + .md5_parse = tcp_v6_parse_md5_keys, +#endif +}; + /* * TCP over IPv4 via INET6 API */ @@ -1358,6 +1874,15 @@ static struct inet_connection_sock_af_ops ipv6_mapped = { #endif }; +static struct tcp_sock_af_ops tcp_sock_ipv6_mapped_specific = { +#ifdef CONFIG_TCP_MD5SIG + .md5_lookup = tcp_v4_md5_lookup, + .calc_md5_hash = tcp_v4_calc_md5_hash, + .md5_add = tcp_v6_md5_add_func, + .md5_parse = tcp_v6_parse_md5_keys, +#endif +}; + /* NOTE: A lot of things set to zero explicitly by call to * sk_alloc() so need not be done here. */ @@ -1397,6 +1922,10 @@ static int tcp_v6_init_sock(struct sock *sk) sk->sk_write_space = sk_stream_write_space; sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); +#ifdef CONFIG_TCP_MD5SIG + tp->af_specific = &tcp_sock_ipv6_specific; +#endif + sk->sk_sndbuf = sysctl_tcp_wmem[1]; sk->sk_rcvbuf = sysctl_tcp_rmem[1]; @@ -1407,6 +1936,11 @@ static int tcp_v6_init_sock(struct sock *sk) static int tcp_v6_destroy_sock(struct sock *sk) { +#ifdef CONFIG_TCP_MD5SIG + /* Clean up the MD5 key list */ + if (tcp_sk(sk)->md5sig_info) + tcp_v6_clear_md5_list(sk); +#endif tcp_v4_destroy_sock(sk); return inet6_destroy_sock(sk); } -- cgit v1.2.3 From 6051e2f4fb68fc8e5343db58fa680ece376f405c Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 14 Nov 2006 19:54:19 -0800 Subject: [IPv6] prefix: Convert RTM_NEWPREFIX notifications to use the new netlink api RTM_GETPREFIX is completely unused and is thus removed. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 2 -- net/core/rtnetlink.c | 1 - net/ipv6/addrconf.c | 30 ++++++++++++++---------------- security/selinux/nlmsgtab.c | 1 - 4 files changed, 14 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 3a18addaed4c..33b3d0ab3a91 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -81,8 +81,6 @@ enum { RTM_NEWPREFIX = 52, #define RTM_NEWPREFIX RTM_NEWPREFIX - RTM_GETPREFIX = 54, -#define RTM_GETPREFIX RTM_GETPREFIX RTM_GETMULTICAST = 58, #define RTM_GETMULTICAST RTM_GETMULTICAST diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 50d6cb40c6e3..0cb4d9e53a07 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -108,7 +108,6 @@ static const int rtm_min[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWTCLASS)] = NLMSG_LENGTH(sizeof(struct tcmsg)), [RTM_FAM(RTM_NEWTFILTER)] = NLMSG_LENGTH(sizeof(struct tcmsg)), [RTM_FAM(RTM_NEWACTION)] = NLMSG_LENGTH(sizeof(struct tcamsg)), - [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), }; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 46cd941d296f..052f99eaf2ac 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3518,16 +3518,18 @@ static inline size_t inet6_prefix_nlmsg_size(void) } static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, - struct prefix_info *pinfo, u32 pid, u32 seq, - int event, unsigned int flags) + struct prefix_info *pinfo, u32 pid, u32 seq, + int event, unsigned int flags) { - struct prefixmsg *pmsg; - struct nlmsghdr *nlh; - unsigned char *b = skb->tail; + struct prefixmsg *pmsg; + struct nlmsghdr *nlh; struct prefix_cacheinfo ci; - nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*pmsg), flags); - pmsg = NLMSG_DATA(nlh); + nlh = nlmsg_put(skb, pid, seq, event, sizeof(*pmsg), flags); + if (nlh == NULL) + return -ENOBUFS; + + pmsg = nlmsg_data(nlh); pmsg->prefix_family = AF_INET6; pmsg->prefix_pad1 = 0; pmsg->prefix_pad2 = 0; @@ -3535,26 +3537,22 @@ static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev, pmsg->prefix_len = pinfo->prefix_len; pmsg->prefix_type = pinfo->type; pmsg->prefix_pad3 = 0; - pmsg->prefix_flags = 0; if (pinfo->onlink) pmsg->prefix_flags |= IF_PREFIX_ONLINK; if (pinfo->autoconf) pmsg->prefix_flags |= IF_PREFIX_AUTOCONF; - RTA_PUT(skb, PREFIX_ADDRESS, sizeof(pinfo->prefix), &pinfo->prefix); + NLA_PUT(skb, PREFIX_ADDRESS, sizeof(pinfo->prefix), &pinfo->prefix); ci.preferred_time = ntohl(pinfo->prefered); ci.valid_time = ntohl(pinfo->valid); - RTA_PUT(skb, PREFIX_CACHEINFO, sizeof(ci), &ci); + NLA_PUT(skb, PREFIX_CACHEINFO, sizeof(ci), &ci); - nlh->nlmsg_len = skb->tail - b; - return skb->len; + return nlmsg_end(skb, nlh); -nlmsg_failure: -rtattr_failure: - skb_trim(skb, b - skb->data); - return -1; +nla_put_failure: + return nlmsg_cancel(skb, nlh); } static void inet6_prefix_notify(int event, struct inet6_dev *idev, diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index b8f4d25cf335..ccfe8755735e 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -60,7 +60,6 @@ static struct nlmsg_perm nlmsg_route_perms[] = { RTM_DELACTION, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, { RTM_GETACTION, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_NEWPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, - { RTM_GETPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETMULTICAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETANYCAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_READ }, -- cgit v1.2.3 From ba4e58eca8aa9473b44fdfd312f26c4a2e7798b3 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 27 Nov 2006 11:10:57 -0800 Subject: [NET]: Supporting UDP-Lite (RFC 3828) in Linux This is a revision of the previously submitted patch, which alters the way files are organized and compiled in the following manner: * UDP and UDP-Lite now use separate object files * source file dependencies resolved via header files net/ipv{4,6}/udp_impl.h * order of inclusion files in udp.c/udplite.c adapted accordingly [NET/IPv4]: Support for the UDP-Lite protocol (RFC 3828) This patch adds support for UDP-Lite to the IPv4 stack, provided as an extension to the existing UDPv4 code: * generic routines are all located in net/ipv4/udp.c * UDP-Lite specific routines are in net/ipv4/udplite.c * MIB/statistics support in /proc/net/snmp and /proc/net/udplite * shared API with extensions for partial checksum coverage [NET/IPv6]: Extension for UDP-Lite over IPv6 It extends the existing UDPv6 code base with support for UDP-Lite in the same manner as per UDPv4. In particular, * UDPv6 generic and shared code is in net/ipv6/udp.c * UDP-Litev6 specific extensions are in net/ipv6/udplite.c * MIB/statistics support in /proc/net/snmp6 and /proc/net/udplite6 * support for IPV6_ADDRFORM * aligned the coding style of protocol initialisation with af_inet6.c * made the error handling in udpv6_queue_rcv_skb consistent; to return `-1' on error on all error cases * consolidation of shared code [NET]: UDP-Lite Documentation and basic XFRM/Netfilter support The UDP-Lite patch further provides * API documentation for UDP-Lite * basic xfrm support * basic netfilter support for IPv4 and IPv6 (LOG target) Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- Documentation/networking/udplite.txt | 281 +++++++++++++++++++ include/linux/in.h | 1 + include/linux/socket.h | 1 + include/linux/udp.h | 12 + include/net/ipv6.h | 12 +- include/net/transp_v6.h | 2 + include/net/udp.h | 91 +++++- include/net/udplite.h | 149 ++++++++++ include/net/xfrm.h | 2 + net/ipv4/Makefile | 3 +- net/ipv4/af_inet.c | 8 +- net/ipv4/netfilter/ipt_LOG.c | 11 +- net/ipv4/proc.c | 13 + net/ipv4/udp.c | 518 ++++++++++++++++++++--------------- net/ipv4/udp_impl.h | 38 +++ net/ipv4/udplite.c | 119 ++++++++ net/ipv4/xfrm4_policy.c | 1 + net/ipv6/Makefile | 4 +- net/ipv6/af_inet6.c | 21 +- net/ipv6/ipv6_sockglue.c | 11 +- net/ipv6/netfilter/ip6t_LOG.c | 10 +- net/ipv6/proc.c | 11 + net/ipv6/udp.c | 361 +++++++++++++----------- net/ipv6/udp_impl.h | 34 +++ net/ipv6/udplite.c | 105 +++++++ net/ipv6/xfrm6_policy.c | 1 + net/netfilter/xt_multiport.c | 5 +- net/netfilter/xt_tcpudp.c | 20 +- 28 files changed, 1442 insertions(+), 403 deletions(-) create mode 100644 Documentation/networking/udplite.txt create mode 100644 include/net/udplite.h create mode 100644 net/ipv4/udp_impl.h create mode 100644 net/ipv4/udplite.c create mode 100644 net/ipv6/udp_impl.h create mode 100644 net/ipv6/udplite.c (limited to 'include/linux') diff --git a/Documentation/networking/udplite.txt b/Documentation/networking/udplite.txt new file mode 100644 index 000000000000..dd6f46b83dab --- /dev/null +++ b/Documentation/networking/udplite.txt @@ -0,0 +1,281 @@ + =========================================================================== + The UDP-Lite protocol (RFC 3828) + =========================================================================== + + + UDP-Lite is a Standards-Track IETF transport protocol whose characteristic + is a variable-length checksum. This has advantages for transport of multimedia + (video, VoIP) over wireless networks, as partly damaged packets can still be + fed into the codec instead of being discarded due to a failed checksum test. + + This file briefly describes the existing kernel support and the socket API. + For in-depth information, you can consult: + + o The UDP-Lite Homepage: http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/ + Fom here you can also download some example application source code. + + o The UDP-Lite HOWTO on + http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/files/UDP-Lite-HOWTO.txt + + o The Wireshark UDP-Lite WiKi (with capture files): + http://wiki.wireshark.org/Lightweight_User_Datagram_Protocol + + o The Protocol Spec, RFC 3828, http://www.ietf.org/rfc/rfc3828.txt + + + I) APPLICATIONS + + Several applications have been ported successfully to UDP-Lite. Ethereal + (now called wireshark) has UDP-Litev4/v6 support by default. The tarball on + + http://www.erg.abdn.ac.uk/users/gerrit/udp-lite/files/udplite_linux.tar.gz + + has source code for several v4/v6 client-server and network testing examples. + + Porting applications to UDP-Lite is straightforward: only socket level and + IPPROTO need to be changed; senders additionally set the checksum coverage + length (default = header length = 8). Details are in the next section. + + + II) PROGRAMMING API + + UDP-Lite provides a connectionless, unreliable datagram service and hence + uses the same socket type as UDP. In fact, porting from UDP to UDP-Lite is + very easy: simply add `IPPROTO_UDPLITE' as the last argument of the socket(2) + call so that the statement looks like: + + s = socket(PF_INET, SOCK_DGRAM, IPPROTO_UDPLITE); + + or, respectively, + + s = socket(PF_INET6, SOCK_DGRAM, IPPROTO_UDPLITE); + + With just the above change you are able to run UDP-Lite services or connect + to UDP-Lite servers. The kernel will assume that you are not interested in + using partial checksum coverage and so emulate UDP mode (full coverage). + + To make use of the partial checksum coverage facilities requires setting a + single socket option, which takes an integer specifying the coverage length: + + * Sender checksum coverage: UDPLITE_SEND_CSCOV + + For example, + + int val = 20; + setsockopt(s, SOL_UDPLITE, UDPLITE_SEND_CSCOV, &val, sizeof(int)); + + sets the checksum coverage length to 20 bytes (12b data + 8b header). + Of each packet only the first 20 bytes (plus the pseudo-header) will be + checksummed. This is useful for RTP applications which have a 12-byte + base header. + + + * Receiver checksum coverage: UDPLITE_RECV_CSCOV + + This option is the receiver-side analogue. It is truly optional, i.e. not + required to enable traffic with partial checksum coverage. Its function is + that of a traffic filter: when enabled, it instructs the kernel to drop + all packets which have a coverage _less_ than this value. For example, if + RTP and UDP headers are to be protected, a receiver can enforce that only + packets with a minimum coverage of 20 are admitted: + + int min = 20; + setsockopt(s, SOL_UDPLITE, UDPLITE_RECV_CSCOV, &min, sizeof(int)); + + The calls to getsockopt(2) are analogous. Being an extension and not a stand- + alone protocol, all socket options known from UDP can be used in exactly the + same manner as before, e.g. UDP_CORK or UDP_ENCAP. + + A detailed discussion of UDP-Lite checksum coverage options is in section IV. + + + III) HEADER FILES + + The socket API requires support through header files in /usr/include: + + * /usr/include/netinet/in.h + to define IPPROTO_UDPLITE + + * /usr/include/netinet/udplite.h + for UDP-Lite header fields and protocol constants + + For testing purposes, the following can serve as a `mini' header file: + + #define IPPROTO_UDPLITE 136 + #define SOL_UDPLITE 136 + #define UDPLITE_SEND_CSCOV 10 + #define UDPLITE_RECV_CSCOV 11 + + Ready-made header files for various distros are in the UDP-Lite tarball. + + + IV) KERNEL BEHAVIOUR WITH REGARD TO THE VARIOUS SOCKET OPTIONS + + To enable debugging messages, the log level need to be set to 8, as most + messages use the KERN_DEBUG level (7). + + 1) Sender Socket Options + + If the sender specifies a value of 0 as coverage length, the module + assumes full coverage, transmits a packet with coverage length of 0 + and according checksum. If the sender specifies a coverage < 8 and + different from 0, the kernel assumes 8 as default value. Finally, + if the specified coverage length exceeds the packet length, the packet + length is used instead as coverage length. + + 2) Receiver Socket Options + + The receiver specifies the minimum value of the coverage length it + is willing to accept. A value of 0 here indicates that the receiver + always wants the whole of the packet covered. In this case, all + partially covered packets are dropped and an error is logged. + + It is not possible to specify illegal values (<0 and <8); in these + cases the default of 8 is assumed. + + All packets arriving with a coverage value less than the specified + threshold are discarded, these events are also logged. + + 3) Disabling the Checksum Computation + + On both sender and receiver, checksumming will always be performed + and can not be disabled using SO_NO_CHECK. Thus + + setsockopt(sockfd, SOL_SOCKET, SO_NO_CHECK, ... ); + + will always will be ignored, while the value of + + getsockopt(sockfd, SOL_SOCKET, SO_NO_CHECK, &value, ...); + + is meaningless (as in TCP). Packets with a zero checksum field are + illegal (cf. RFC 3828, sec. 3.1) will be silently discarded. + + 4) Fragmentation + + The checksum computation respects both buffersize and MTU. The size + of UDP-Lite packets is determined by the size of the send buffer. The + minimum size of the send buffer is 2048 (defined as SOCK_MIN_SNDBUF + in include/net/sock.h), the default value is configurable as + net.core.wmem_default or via setting the SO_SNDBUF socket(7) + option. The maximum upper bound for the send buffer is determined + by net.core.wmem_max. + + Given a payload size larger than the send buffer size, UDP-Lite will + split the payload into several individual packets, filling up the + send buffer size in each case. + + The precise value also depends on the interface MTU. The interface MTU, + in turn, may trigger IP fragmentation. In this case, the generated + UDP-Lite packet is split into several IP packets, of which only the + first one contains the L4 header. + + The send buffer size has implications on the checksum coverage length. + Consider the following example: + + Payload: 1536 bytes Send Buffer: 1024 bytes + MTU: 1500 bytes Coverage Length: 856 bytes + + UDP-Lite will ship the 1536 bytes in two separate packets: + + Packet 1: 1024 payload + 8 byte header + 20 byte IP header = 1052 bytes + Packet 2: 512 payload + 8 byte header + 20 byte IP header = 540 bytes + + The coverage packet covers the UDP-Lite header and 848 bytes of the + payload in the first packet, the second packet is fully covered. Note + that for the second packet, the coverage length exceeds the packet + length. The kernel always re-adjusts the coverage length to the packet + length in such cases. + + As an example of what happens when one UDP-Lite packet is split into + several tiny fragments, consider the following example. + + Payload: 1024 bytes Send buffer size: 1024 bytes + MTU: 300 bytes Coverage length: 575 bytes + + +-+-----------+--------------+--------------+--------------+ + |8| 272 | 280 | 280 | 280 | + +-+-----------+--------------+--------------+--------------+ + 280 560 840 1032 + ^ + *****checksum coverage************* + + The UDP-Lite module generates one 1032 byte packet (1024 + 8 byte + header). According to the interface MTU, these are split into 4 IP + packets (280 byte IP payload + 20 byte IP header). The kernel module + sums the contents of the entire first two packets, plus 15 bytes of + the last packet before releasing the fragments to the IP module. + + To see the analogous case for IPv6 fragmentation, consider a link + MTU of 1280 bytes and a write buffer of 3356 bytes. If the checksum + coverage is less than 1232 bytes (MTU minus IPv6/fragment header + lengths), only the first fragment needs to be considered. When using + larger checksum coverage lengths, each eligible fragment needs to be + checksummed. Suppose we have a checksum coverage of 3062. The buffer + of 3356 bytes will be split into the following fragments: + + Fragment 1: 1280 bytes carrying 1232 bytes of UDP-Lite data + Fragment 2: 1280 bytes carrying 1232 bytes of UDP-Lite data + Fragment 3: 948 bytes carrying 900 bytes of UDP-Lite data + + The first two fragments have to be checksummed in full, of the last + fragment only 598 (= 3062 - 2*1232) bytes are checksummed. + + While it is important that such cases are dealt with correctly, they + are (annoyingly) rare: UDP-Lite is designed for optimising multimedia + performance over wireless (or generally noisy) links and thus smaller + coverage lenghts are likely to be expected. + + + V) UDP-LITE RUNTIME STATISTICS AND THEIR MEANING + + Exceptional and error conditions are logged to syslog at the KERN_DEBUG + level. Live statistics about UDP-Lite are available in /proc/net/snmp + and can (with newer versions of netstat) be viewed using + + netstat -svu + + This displays UDP-Lite statistics variables, whose meaning is as follows. + + InDatagrams: Total number of received datagrams. + + NoPorts: Number of packets received to an unknown port. + These cases are counted separately (not as InErrors). + + InErrors: Number of erroneous UDP-Lite packets. Errors include: + * internal socket queue receive errors + * packet too short (less than 8 bytes or stated + coverage length exceeds received length) + * xfrm4_policy_check() returned with error + * application has specified larger min. coverage + length than that of incoming packet + * checksum coverage violated + * bad checksum + + OutDatagrams: Total number of sent datagrams. + + These statistics derive from the UDP MIB (RFC 2013). + + + VI) IPTABLES + + There is packet match support for UDP-Lite as well as support for the LOG target. + If you copy and paste the following line into /etc/protcols, + + udplite 136 UDP-Lite # UDP-Lite [RFC 3828] + + then + iptables -A INPUT -p udplite -j LOG + + will produce logging output to syslog. Dropping and rejecting packets also works. + + + VII) MAINTAINER ADDRESS + + The UDP-Lite patch was developed at + University of Aberdeen + Electronics Research Group + Department of Engineering + Fraser Noble Building + Aberdeen AB24 3UE; UK + The current maintainer is Gerrit Renker, . Initial + code was developed by William Stanislaus, . diff --git a/include/linux/in.h b/include/linux/in.h index 2619859f6e1b..1912e7c0bc26 100644 --- a/include/linux/in.h +++ b/include/linux/in.h @@ -45,6 +45,7 @@ enum { IPPROTO_COMP = 108, /* Compression Header protocol */ IPPROTO_SCTP = 132, /* Stream Control Transport Protocol */ + IPPROTO_UDPLITE = 136, /* UDP-Lite (RFC 3828) */ IPPROTO_RAW = 255, /* Raw IP packets */ IPPROTO_MAX diff --git a/include/linux/socket.h b/include/linux/socket.h index 361409094649..592b66679823 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -264,6 +264,7 @@ struct ucred { #define SOL_IPV6 41 #define SOL_ICMPV6 58 #define SOL_SCTP 132 +#define SOL_UDPLITE 136 /* UDP-Lite (RFC 3828) */ #define SOL_RAW 255 #define SOL_IPX 256 #define SOL_AX25 257 diff --git a/include/linux/udp.h b/include/linux/udp.h index 014b41d1e308..564f3b050105 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -38,6 +38,7 @@ struct udphdr { #include #include +#define UDP_HTABLE_SIZE 128 struct udp_sock { /* inet_sock has to be the first member */ @@ -50,12 +51,23 @@ struct udp_sock { * when the socket is uncorked. */ __u16 len; /* total length of pending frames */ + /* + * Fields specific to UDP-Lite. + */ + __u16 pcslen; + __u16 pcrlen; +/* indicator bits used by pcflag: */ +#define UDPLITE_BIT 0x1 /* set by udplite proto init function */ +#define UDPLITE_SEND_CC 0x2 /* set via udplite setsockopt */ +#define UDPLITE_RECV_CC 0x4 /* set via udplite setsocktopt */ + __u8 pcflag; /* marks socket as UDP-Lite if > 0 */ }; static inline struct udp_sock *udp_sk(const struct sock *sk) { return (struct udp_sock *)sk; } +#define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag) #endif diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 3c266ad99a02..9390649bbfec 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -158,9 +158,13 @@ DECLARE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); SNMP_INC_STATS_OFFSET_BH(icmpv6_statistics, field, _offset); \ }) DECLARE_SNMP_STAT(struct udp_mib, udp_stats_in6); -#define UDP6_INC_STATS(field) SNMP_INC_STATS(udp_stats_in6, field) -#define UDP6_INC_STATS_BH(field) SNMP_INC_STATS_BH(udp_stats_in6, field) -#define UDP6_INC_STATS_USER(field) SNMP_INC_STATS_USER(udp_stats_in6, field) +DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6); +#define UDP6_INC_STATS_BH(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_BH(udplite_stats_in6, field); \ + else SNMP_INC_STATS_BH(udp_stats_in6, field); } while(0) +#define UDP6_INC_STATS_USER(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_USER(udplite_stats_in6, field); \ + else SNMP_INC_STATS_USER(udp_stats_in6, field); } while(0) int snmp6_register_dev(struct inet6_dev *idev); int snmp6_unregister_dev(struct inet6_dev *idev); @@ -604,6 +608,8 @@ extern int tcp6_proc_init(void); extern void tcp6_proc_exit(void); extern int udp6_proc_init(void); extern void udp6_proc_exit(void); +extern int udplite6_proc_init(void); +extern void udplite6_proc_exit(void); extern int ipv6_misc_proc_init(void); extern void ipv6_misc_proc_exit(void); diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 61f724c1036f..409da3a9a455 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -11,6 +11,7 @@ extern struct proto rawv6_prot; extern struct proto udpv6_prot; +extern struct proto udplitev6_prot; extern struct proto tcpv6_prot; struct flowi; @@ -24,6 +25,7 @@ extern void ipv6_destopt_init(void); /* transport protocols */ extern void rawv6_init(void); extern void udpv6_init(void); +extern void udplitev6_init(void); extern void tcpv6_init(void); extern int udpv6_connect(struct sock *sk, diff --git a/include/net/udp.h b/include/net/udp.h index db0c05f67546..4f0626735ed3 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -26,9 +26,28 @@ #include #include #include +#include +#include #include -#define UDP_HTABLE_SIZE 128 +/** + * struct udp_skb_cb - UDP(-Lite) private variables + * + * @header: private variables used by IPv4/IPv6 + * @cscov: checksum coverage length (UDP-Lite only) + * @partial_cov: if set indicates partial csum coverage + */ +struct udp_skb_cb { + union { + struct inet_skb_parm h4; +#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) + struct inet6_skb_parm h6; +#endif + } header; + __u16 cscov; + __u8 partial_cov; +}; +#define UDP_SKB_CB(__skb) ((struct udp_skb_cb *)((__skb)->cb)) extern struct hlist_head udp_hash[UDP_HTABLE_SIZE]; extern rwlock_t udp_hash_lock; @@ -47,6 +66,62 @@ extern struct proto udp_prot; struct sk_buff; +/* + * Generic checksumming routines for UDP(-Lite) v4 and v6 + */ +static inline u16 __udp_lib_checksum_complete(struct sk_buff *skb) +{ + if (! UDP_SKB_CB(skb)->partial_cov) + return __skb_checksum_complete(skb); + return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, + skb->csum)); +} + +static __inline__ int udp_lib_checksum_complete(struct sk_buff *skb) +{ + return skb->ip_summed != CHECKSUM_UNNECESSARY && + __udp_lib_checksum_complete(skb); +} + +/** + * udp_csum_outgoing - compute UDPv4/v6 checksum over fragments + * @sk: socket we are writing to + * @skb: sk_buff containing the filled-in UDP header + * (checksum field must be zeroed out) + */ +static inline u32 udp_csum_outgoing(struct sock *sk, struct sk_buff *skb) +{ + u32 csum = csum_partial(skb->h.raw, sizeof(struct udphdr), 0); + + skb_queue_walk(&sk->sk_write_queue, skb) { + csum = csum_add(csum, skb->csum); + } + return csum; +} + +/* hash routines shared between UDPv4/6 and UDP-Litev4/6 */ +static inline void udp_lib_hash(struct sock *sk) +{ + BUG(); +} + +static inline void udp_lib_unhash(struct sock *sk) +{ + write_lock_bh(&udp_hash_lock); + if (sk_del_node_init(sk)) { + inet_sk(sk)->num = 0; + sock_prot_dec_use(sk->sk_prot); + } + write_unlock_bh(&udp_hash_lock); +} + +static inline void udp_lib_close(struct sock *sk, long timeout) +{ + sk_common_release(sk); +} + + +/* net/ipv4/udp.c */ extern int udp_get_port(struct sock *sk, unsigned short snum, int (*saddr_cmp)(const struct sock *, const struct sock *)); extern void udp_err(struct sk_buff *, u32); @@ -61,21 +136,29 @@ extern unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait); DECLARE_SNMP_STAT(struct udp_mib, udp_statistics); -#define UDP_INC_STATS(field) SNMP_INC_STATS(udp_statistics, field) -#define UDP_INC_STATS_BH(field) SNMP_INC_STATS_BH(udp_statistics, field) -#define UDP_INC_STATS_USER(field) SNMP_INC_STATS_USER(udp_statistics, field) +/* + * SNMP statistics for UDP and UDP-Lite + */ +#define UDP_INC_STATS_USER(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_USER(udplite_statistics, field); \ + else SNMP_INC_STATS_USER(udp_statistics, field); } while(0) +#define UDP_INC_STATS_BH(field, is_udplite) do { \ + if (is_udplite) SNMP_INC_STATS_BH(udplite_statistics, field); \ + else SNMP_INC_STATS_BH(udp_statistics, field); } while(0) /* /proc */ struct udp_seq_afinfo { struct module *owner; char *name; sa_family_t family; + struct hlist_head *hashtable; int (*seq_show) (struct seq_file *m, void *v); struct file_operations *seq_fops; }; struct udp_iter_state { sa_family_t family; + struct hlist_head *hashtable; int bucket; struct seq_operations seq_ops; }; diff --git a/include/net/udplite.h b/include/net/udplite.h new file mode 100644 index 000000000000..1473b3e49044 --- /dev/null +++ b/include/net/udplite.h @@ -0,0 +1,149 @@ +/* + * Definitions for the UDP-Lite (RFC 3828) code. + */ +#ifndef _UDPLITE_H +#define _UDPLITE_H + +/* UDP-Lite socket options */ +#define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */ +#define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */ + +extern struct proto udplite_prot; +extern struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; + +/* UDP-Lite does not have a standardized MIB yet, so we inherit from UDP */ +DECLARE_SNMP_STAT(struct udp_mib, udplite_statistics); + +/* + * Checksum computation is all in software, hence simpler getfrag. + */ +static __inline__ int udplite_getfrag(void *from, char *to, int offset, + int len, int odd, struct sk_buff *skb) +{ + return memcpy_fromiovecend(to, (struct iovec *) from, offset, len); +} + +/* Designate sk as UDP-Lite socket */ +static inline int udplite_sk_init(struct sock *sk) +{ + udp_sk(sk)->pcflag = UDPLITE_BIT; + return 0; +} + +/* + * Checksumming routines + */ +static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh) +{ + u16 cscov; + + /* In UDPv4 a zero checksum means that the transmitter generated no + * checksum. UDP-Lite (like IPv6) mandates checksums, hence packets + * with a zero checksum field are illegal. */ + if (uh->check == 0) { + LIMIT_NETDEBUG(KERN_DEBUG "UDPLITE: zeroed checksum field\n"); + return 1; + } + + UDP_SKB_CB(skb)->partial_cov = 0; + cscov = ntohs(uh->len); + + if (cscov == 0) /* Indicates that full coverage is required. */ + cscov = skb->len; + else if (cscov < 8 || cscov > skb->len) { + /* + * Coverage length violates RFC 3828: log and discard silently. + */ + LIMIT_NETDEBUG(KERN_DEBUG "UDPLITE: bad csum coverage %d/%d\n", + cscov, skb->len); + return 1; + + } else if (cscov < skb->len) + UDP_SKB_CB(skb)->partial_cov = 1; + + UDP_SKB_CB(skb)->cscov = cscov; + + /* + * There is no known NIC manufacturer supporting UDP-Lite yet, + * hence ip_summed is always (re-)set to CHECKSUM_NONE. + */ + skb->ip_summed = CHECKSUM_NONE; + + return 0; +} + +static __inline__ int udplite4_csum_init(struct sk_buff *skb, struct udphdr *uh) +{ + int rc = udplite_checksum_init(skb, uh); + + if (!rc) + skb->csum = csum_tcpudp_nofold(skb->nh.iph->saddr, + skb->nh.iph->daddr, + skb->len, IPPROTO_UDPLITE, 0); + return rc; +} + +static __inline__ int udplite6_csum_init(struct sk_buff *skb, struct udphdr *uh) +{ + int rc = udplite_checksum_init(skb, uh); + + if (!rc) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDPLITE, 0); + return rc; +} + +static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) +{ + int cscov = up->len; + + /* + * Sender has set `partial coverage' option on UDP-Lite socket + */ + if (up->pcflag & UDPLITE_SEND_CC) { + if (up->pcslen < up->len) { + /* up->pcslen == 0 means that full coverage is required, + * partial coverage only if 0 < up->pcslen < up->len */ + if (0 < up->pcslen) { + cscov = up->pcslen; + } + uh->len = htons(up->pcslen); + } + /* + * NOTE: Causes for the error case `up->pcslen > up->len': + * (i) Application error (will not be penalized). + * (ii) Payload too big for send buffer: data is split + * into several packets, each with its own header. + * In this case (e.g. last segment), coverage may + * exceed packet length. + * Since packets with coverage length > packet length are + * illegal, we fall back to the defaults here. + */ + } + return cscov; +} + +static inline u32 udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) +{ + u32 csum = 0; + int off, len, cscov = udplite_sender_cscov(udp_sk(sk), skb->h.uh); + + skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */ + + skb_queue_walk(&sk->sk_write_queue, skb) { + off = skb->h.raw - skb->data; + len = skb->len - off; + + csum = skb_checksum(skb, off, (cscov > len)? len : cscov, csum); + + if ((cscov -= len) <= 0) + break; + } + return csum; +} + +extern void udplite4_register(void); +extern int udplite_get_port(struct sock *sk, unsigned short snum, + int (*scmp)(const struct sock *, const struct sock *)); +#endif /* _UDPLITE_H */ diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 81c91e8a328f..3878a88ff618 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -468,6 +468,7 @@ __be16 xfrm_flowi_sport(struct flowi *fl) switch(fl->proto) { case IPPROTO_TCP: case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_SCTP: port = fl->fl_ip_sport; break; @@ -493,6 +494,7 @@ __be16 xfrm_flowi_dport(struct flowi *fl) switch(fl->proto) { case IPPROTO_TCP: case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_SCTP: port = fl->fl_ip_dport; break; diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 15645c51520c..7a068626feea 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -8,7 +8,8 @@ obj-y := route.o inetpeer.o protocol.o \ inet_timewait_sock.o inet_connection_sock.o \ tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ tcp_minisocks.o tcp_cong.o \ - datagram.o raw.o udp.o arp.o icmp.o devinet.o af_inet.o igmp.o \ + datagram.o raw.o udp.o udplite.o \ + arp.o icmp.o devinet.o af_inet.o igmp.o \ sysctl_net_ipv4.o fib_frontend.o fib_semantics.o obj-$(CONFIG_IP_FIB_HASH) += fib_hash.o diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 4a81d54a7569..8db39f7e3bf0 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -104,6 +104,7 @@ #include #include #include +#include #include #include #include @@ -1223,10 +1224,13 @@ static int __init init_ipv4_mibs(void) tcp_statistics[1] = alloc_percpu(struct tcp_mib); udp_statistics[0] = alloc_percpu(struct udp_mib); udp_statistics[1] = alloc_percpu(struct udp_mib); + udplite_statistics[0] = alloc_percpu(struct udp_mib); + udplite_statistics[1] = alloc_percpu(struct udp_mib); if (! (net_statistics[0] && net_statistics[1] && ip_statistics[0] && ip_statistics[1] && tcp_statistics[0] && tcp_statistics[1] - && udp_statistics[0] && udp_statistics[1])) + && udp_statistics[0] && udp_statistics[1] + && udplite_statistics[0] && udplite_statistics[1] ) ) return -ENOMEM; (void) tcp_mib_init(); @@ -1313,6 +1317,8 @@ static int __init inet_init(void) /* Setup TCP slab cache for open requests. */ tcp_init(); + /* Add UDP-Lite (RFC 3828) */ + udplite4_register(); /* * Set the ICMP layer up diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 7dc820df8bc5..46eee64a11f6 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -171,11 +171,15 @@ static void dump_packet(const struct nf_loginfo *info, } break; } - case IPPROTO_UDP: { + case IPPROTO_UDP: + case IPPROTO_UDPLITE: { struct udphdr _udph, *uh; - /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP "); + if (ih->protocol == IPPROTO_UDP) + /* Max length: 10 "PROTO=UDP " */ + printk("PROTO=UDP " ); + else /* Max length: 14 "PROTO=UDPLITE " */ + printk("PROTO=UDPLITE "); if (ntohs(ih->frag_off) & IP_OFFSET) break; @@ -341,6 +345,7 @@ static void dump_packet(const struct nf_loginfo *info, /* IP: 40+46+6+11+127 = 230 */ /* TCP: 10+max(25,20+30+13+9+32+11+127) = 252 */ /* UDP: 10+max(25,20) = 35 */ + /* UDPLITE: 14+max(25,20) = 39 */ /* ICMP: 11+max(25, 18+25+max(19,14,24+3+n+10,3+n+10)) = 91+n */ /* ESP: 10+max(25)+15 = 50 */ /* AH: 9+max(25)+15 = 49 */ diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 9c6cbe3d9fb8..cd873da54cbe 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -66,6 +67,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated), atomic_read(&tcp_memory_allocated)); seq_printf(seq, "UDP: inuse %d\n", fold_prot_inuse(&udp_prot)); + seq_printf(seq, "UDPLITE: inuse %d\n", fold_prot_inuse(&udplite_prot)); seq_printf(seq, "RAW: inuse %d\n", fold_prot_inuse(&raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", ip_frag_nqueues, atomic_read(&ip_frag_mem)); @@ -304,6 +306,17 @@ static int snmp_seq_show(struct seq_file *seq, void *v) fold_field((void **) udp_statistics, snmp4_udp_list[i].entry)); + /* the UDP and UDP-Lite MIBs are the same */ + seq_puts(seq, "\nUdpLite:"); + for (i = 0; snmp4_udp_list[i].name != NULL; i++) + seq_printf(seq, " %s", snmp4_udp_list[i].name); + + seq_puts(seq, "\nUdpLite:"); + for (i = 0; snmp4_udp_list[i].name != NULL; i++) + seq_printf(seq, " %lu", + fold_field((void **) udplite_statistics, + snmp4_udp_list[i].entry) ); + seq_putc(seq, '\n'); return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 9e1bd374875e..98ba75096175 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -92,22 +92,16 @@ #include #include #include -#include #include -#include -#include #include -#include #include #include #include -#include -#include #include #include -#include #include #include +#include "udp_impl.h" /* * Snmp MIB for the UDP layer @@ -120,26 +114,30 @@ DEFINE_RWLOCK(udp_hash_lock); static int udp_port_rover; -static inline int udp_lport_inuse(u16 num) +static inline int __udp_lib_lport_inuse(__be16 num, struct hlist_head udptable[]) { struct sock *sk; struct hlist_node *node; - sk_for_each(sk, node, &udp_hash[num & (UDP_HTABLE_SIZE - 1)]) + sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)]) if (inet_sk(sk)->num == num) return 1; return 0; } /** - * udp_get_port - common port lookup for IPv4 and IPv6 + * __udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6 * * @sk: socket struct in question * @snum: port number to look up + * @udptable: hash list table, must be of UDP_HTABLE_SIZE + * @port_rover: pointer to record of last unallocated port * @saddr_comp: AF-dependent comparison of bound local IP addresses */ -int udp_get_port(struct sock *sk, unsigned short snum, - int (*saddr_cmp)(const struct sock *sk1, const struct sock *sk2)) +int __udp_lib_get_port(struct sock *sk, unsigned short snum, + struct hlist_head udptable[], int *port_rover, + int (*saddr_comp)(const struct sock *sk1, + const struct sock *sk2 ) ) { struct hlist_node *node; struct hlist_head *head; @@ -150,15 +148,15 @@ int udp_get_port(struct sock *sk, unsigned short snum, if (snum == 0) { int best_size_so_far, best, result, i; - if (udp_port_rover > sysctl_local_port_range[1] || - udp_port_rover < sysctl_local_port_range[0]) - udp_port_rover = sysctl_local_port_range[0]; + if (*port_rover > sysctl_local_port_range[1] || + *port_rover < sysctl_local_port_range[0]) + *port_rover = sysctl_local_port_range[0]; best_size_so_far = 32767; - best = result = udp_port_rover; + best = result = *port_rover; for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) { int size; - head = &udp_hash[result & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[result & (UDP_HTABLE_SIZE - 1)]; if (hlist_empty(head)) { if (result > sysctl_local_port_range[1]) result = sysctl_local_port_range[0] + @@ -179,15 +177,15 @@ int udp_get_port(struct sock *sk, unsigned short snum, result = sysctl_local_port_range[0] + ((result - sysctl_local_port_range[0]) & (UDP_HTABLE_SIZE - 1)); - if (!udp_lport_inuse(result)) + if (! __udp_lib_lport_inuse(result, udptable)) break; } if (i >= (1 << 16) / UDP_HTABLE_SIZE) goto fail; gotit: - udp_port_rover = snum = result; + *port_rover = snum = result; } else { - head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[snum & (UDP_HTABLE_SIZE - 1)]; sk_for_each(sk2, node, head) if (inet_sk(sk2)->num == snum && @@ -195,12 +193,12 @@ gotit: (!sk2->sk_reuse || !sk->sk_reuse) && (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && - (*saddr_cmp)(sk, sk2) ) + (*saddr_comp)(sk, sk2) ) goto fail; } inet_sk(sk)->num = snum; if (sk_unhashed(sk)) { - head = &udp_hash[snum & (UDP_HTABLE_SIZE - 1)]; + head = &udptable[snum & (UDP_HTABLE_SIZE - 1)]; sk_add_node(sk, head); sock_prot_inc_use(sk->sk_prot); } @@ -210,7 +208,13 @@ fail: return error; } -static inline int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) +__inline__ int udp_get_port(struct sock *sk, unsigned short snum, + int (*scmp)(const struct sock *, const struct sock *)) +{ + return __udp_lib_get_port(sk, snum, udp_hash, &udp_port_rover, scmp); +} + +inline int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) { struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); @@ -224,34 +228,20 @@ static inline int udp_v4_get_port(struct sock *sk, unsigned short snum) return udp_get_port(sk, snum, ipv4_rcv_saddr_equal); } - -static void udp_v4_hash(struct sock *sk) -{ - BUG(); -} - -static void udp_v4_unhash(struct sock *sk) -{ - write_lock_bh(&udp_hash_lock); - if (sk_del_node_init(sk)) { - inet_sk(sk)->num = 0; - sock_prot_dec_use(sk->sk_prot); - } - write_unlock_bh(&udp_hash_lock); -} - /* UDP is nearly always wildcards out the wazoo, it makes no sense to try * harder than this. -DaveM */ -static struct sock *udp_v4_lookup_longway(__be32 saddr, __be16 sport, - __be32 daddr, __be16 dport, int dif) +static struct sock *__udp4_lib_lookup(__be32 saddr, __be16 sport, + __be32 daddr, __be16 dport, + int dif, struct hlist_head udptable[]) { struct sock *sk, *result = NULL; struct hlist_node *node; unsigned short hnum = ntohs(dport); int badness = -1; - sk_for_each(sk, node, &udp_hash[hnum & (UDP_HTABLE_SIZE - 1)]) { + read_lock(&udp_hash_lock); + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); if (inet->num == hnum && !ipv6_only_sock(sk)) { @@ -285,20 +275,10 @@ static struct sock *udp_v4_lookup_longway(__be32 saddr, __be16 sport, } } } - return result; -} - -static __inline__ struct sock *udp_v4_lookup(__be32 saddr, __be16 sport, - __be32 daddr, __be16 dport, int dif) -{ - struct sock *sk; - - read_lock(&udp_hash_lock); - sk = udp_v4_lookup_longway(saddr, sport, daddr, dport, dif); - if (sk) - sock_hold(sk); + if (result) + sock_hold(result); read_unlock(&udp_hash_lock); - return sk; + return result; } static inline struct sock *udp_v4_mcast_next(struct sock *sk, @@ -340,7 +320,7 @@ found: * to find the appropriate port. */ -void udp_err(struct sk_buff *skb, u32 info) +void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[]) { struct inet_sock *inet; struct iphdr *iph = (struct iphdr*)skb->data; @@ -351,7 +331,8 @@ void udp_err(struct sk_buff *skb, u32 info) int harderr; int err; - sk = udp_v4_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex); + sk = __udp4_lib_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, + skb->dev->ifindex, udptable ); if (sk == NULL) { ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); return; /* No socket for error */ @@ -405,6 +386,11 @@ out: sock_put(sk); } +__inline__ void udp_err(struct sk_buff *skb, u32 info) +{ + return __udp4_lib_err(skb, info, udp_hash); +} + /* * Throw away all pending data and cancel the corking. Socket is locked. */ @@ -419,16 +405,56 @@ static void udp_flush_pending_frames(struct sock *sk) } } +/** + * udp4_hwcsum_outgoing - handle outgoing HW checksumming + * @sk: socket we are sending on + * @skb: sk_buff containing the filled-in UDP header + * (checksum field must be zeroed out) + */ +static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, + __be32 src, __be32 dst, int len ) +{ + unsigned int csum = 0, offset; + struct udphdr *uh = skb->h.uh; + + if (skb_queue_len(&sk->sk_write_queue) == 1) { + /* + * Only one fragment on the socket. + */ + skb->csum = offsetof(struct udphdr, check); + uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); + } else { + /* + * HW-checksum won't work as there are two or more + * fragments on the socket so that all csums of sk_buffs + * should be together + */ + offset = skb->h.raw - skb->data; + skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); + + skb->ip_summed = CHECKSUM_NONE; + + skb_queue_walk(&sk->sk_write_queue, skb) { + csum = csum_add(csum, skb->csum); + } + + uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum); + if (uh->check == 0) + uh->check = -1; + } +} + /* * Push out all pending data as one UDP datagram. Socket is locked. */ -static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) +int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) { struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; struct sk_buff *skb; struct udphdr *uh; int err = 0; + u32 csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) @@ -443,52 +469,28 @@ static int udp_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->len = htons(up->len); uh->check = 0; - if (sk->sk_no_check == UDP_CSUM_NOXMIT) { + if (up->pcflag) /* UDP-Lite */ + csum = udplite_csum_outgoing(sk, skb); + + else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + skb->ip_summed = CHECKSUM_NONE; goto send; - } - if (skb_queue_len(&sk->sk_write_queue) == 1) { - /* - * Only one fragment on the socket. - */ - if (skb->ip_summed == CHECKSUM_PARTIAL) { - skb->csum = offsetof(struct udphdr, check); - uh->check = ~csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, 0); - } else { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, skb->csum); - if (uh->check == 0) - uh->check = -1; - } - } else { - unsigned int csum = 0; - /* - * HW-checksum won't work as there are two or more - * fragments on the socket so that all csums of sk_buffs - * should be together. - */ - if (skb->ip_summed == CHECKSUM_PARTIAL) { - int offset = (unsigned char *)uh - skb->data; - skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); + } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ - skb->ip_summed = CHECKSUM_NONE; - } else { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - } + udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len); + goto send; + + } else /* `normal' UDP */ + csum = udp_csum_outgoing(sk, skb); + + /* add protocol-dependent pseudo-header */ + uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len, + sk->sk_protocol, csum ); + if (uh->check == 0) + uh->check = -1; - skb_queue_walk(&sk->sk_write_queue, skb) { - csum = csum_add(csum, skb->csum); - } - uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, - up->len, IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = -1; - } send: err = ip_push_pending_frames(sk); out: @@ -497,12 +499,6 @@ out: return err; } - -static unsigned short udp_check(struct udphdr *uh, int len, __be32 saddr, __be32 daddr, unsigned long base) -{ - return(csum_tcpudp_magic(saddr, daddr, len, IPPROTO_UDP, base)); -} - int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len) { @@ -516,8 +512,9 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, __be32 daddr, faddr, saddr; __be16 dport; u8 tos; - int err; + int err, is_udplite = up->pcflag; int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); if (len > 0xFFFF) return -EMSGSIZE; @@ -622,7 +619,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, { .daddr = faddr, .saddr = saddr, .tos = tos } }, - .proto = IPPROTO_UDP, + .proto = sk->sk_protocol, .uli_u = { .ports = { .sport = inet->sport, .dport = dport } } }; @@ -668,8 +665,9 @@ back_from_confirm: do_append_data: up->len += ulen; - err = ip_append_data(sk, ip_generic_getfrag, msg->msg_iov, ulen, - sizeof(struct udphdr), &ipc, rt, + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip_append_data(sk, getfrag, msg->msg_iov, ulen, + sizeof(struct udphdr), &ipc, rt, corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); if (err) udp_flush_pending_frames(sk); @@ -684,7 +682,7 @@ out: if (free) kfree(ipc.opt); if (!err) { - UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS); + UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); return len; } /* @@ -695,7 +693,7 @@ out: * seems like overkill. */ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS); + UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); } return err; @@ -707,8 +705,8 @@ do_confirm: goto out; } -static int udp_sendpage(struct sock *sk, struct page *page, int offset, - size_t size, int flags) +int udp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags) { struct udp_sock *up = udp_sk(sk); int ret; @@ -795,29 +793,18 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) return(0); } -static __inline__ int __udp_checksum_complete(struct sk_buff *skb) -{ - return __skb_checksum_complete(skb); -} - -static __inline__ int udp_checksum_complete(struct sk_buff *skb) -{ - return skb->ip_summed != CHECKSUM_UNNECESSARY && - __udp_checksum_complete(skb); -} - /* * This should be easy, if there is something there we * return it, otherwise we block. */ -static int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len, int noblock, int flags, int *addr_len) +int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len) { struct inet_sock *inet = inet_sk(sk); struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name; struct sk_buff *skb; - int copied, err; + int copied, err, copy_only, is_udplite = IS_UDPLITE(sk); /* * Check any passed addresses @@ -839,15 +826,25 @@ try_again: msg->msg_flags |= MSG_TRUNC; } - if (skb->ip_summed==CHECKSUM_UNNECESSARY) { - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else if (msg->msg_flags&MSG_TRUNC) { - if (__udp_checksum_complete(skb)) + /* + * Decide whether to checksum and/or copy data. + * + * UDP: checksum may have been computed in HW, + * (re-)compute it if message is truncated. + * UDP-Lite: always needs to checksum, no HW support. + */ + copy_only = (skb->ip_summed==CHECKSUM_UNNECESSARY); + + if (is_udplite || (!copy_only && msg->msg_flags&MSG_TRUNC)) { + if (__udp_lib_checksum_complete(skb)) goto csum_copy_err; - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else { + copy_only = 1; + } + + if (copy_only) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); if (err == -EINVAL) @@ -880,7 +877,7 @@ out: return err; csum_copy_err: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); skb_kill_datagram(sk, skb, flags); @@ -912,11 +909,6 @@ int udp_disconnect(struct sock *sk, int flags) return 0; } -static void udp_close(struct sock *sk, long timeout) -{ - sk_common_release(sk); -} - /* return: * 1 if the the UDP system should process it * 0 if we should drop this packet @@ -1022,7 +1014,7 @@ static int udp_encap_rcv(struct sock * sk, struct sk_buff *skb) * Note that in the success and error cases, the skb is assumed to * have either been requeued or freed. */ -static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) { struct udp_sock *up = udp_sk(sk); int rc; @@ -1030,10 +1022,8 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) /* * Charge it to the socket, dropping if the queue is full. */ - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) { - kfree_skb(skb); - return -1; - } + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; nf_reset(skb); if (up->encap_type) { @@ -1057,31 +1047,68 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) if (ret < 0) { /* process the ESP packet */ ret = xfrm4_rcv_encap(skb, up->encap_type); - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return -ret; } /* FALLTHROUGH -- it's a UDP Packet */ } - if (sk->sk_filter && skb->ip_summed != CHECKSUM_UNNECESSARY) { - if (__udp_checksum_complete(skb)) { - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return -1; + /* + * UDP-Lite specific tests, ignored on UDP sockets + */ + if ((up->pcflag & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + /* + * MIB statistics other than incrementing the error count are + * disabled for the following two types of errors: these depend + * on the application settings, not on the functioning of the + * protocol stack as such. + * + * RFC 3828 here recommends (sec 3.3): "There should also be a + * way ... to ... at least let the receiving application block + * delivery of packets with coverage values less than a value + * provided by the application." + */ + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage " + "%d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; } + /* The next case involves violating the min. coverage requested + * by the receiver. This is subtle: if receiver wants x and x is + * greater than the buffersize/MTU then receiver will complain + * that it wants x while sender emits packets of smaller size y. + * Therefore the above ...()->partial_cov statement is essential. + */ + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING + "UDPLITE: coverage %d too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } + } + + if (sk->sk_filter && skb->ip_summed != CHECKSUM_UNNECESSARY) { + if (__udp_lib_checksum_complete(skb)) + goto drop; skb->ip_summed = CHECKSUM_UNNECESSARY; } if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { /* Note that an ENOMEM error is charged twice */ if (rc == -ENOMEM) - UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS); - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return -1; + UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag); + goto drop; } - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + + UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return 0; + +drop: + UDP_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); + kfree_skb(skb); + return -1; } /* @@ -1090,14 +1117,16 @@ static int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) * Note: called only from the BH handler context, * so we don't need to lock the hashes. */ -static int udp_v4_mcast_deliver(struct sk_buff *skb, struct udphdr *uh, - __be32 saddr, __be32 daddr) +static int __udp4_lib_mcast_deliver(struct sk_buff *skb, + struct udphdr *uh, + __be32 saddr, __be32 daddr, + struct hlist_head udptable[]) { struct sock *sk; int dif; read_lock(&udp_hash_lock); - sk = sk_head(&udp_hash[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); dif = skb->dev->ifindex; sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); if (sk) { @@ -1131,65 +1160,75 @@ static int udp_v4_mcast_deliver(struct sk_buff *skb, struct udphdr *uh, * Otherwise, csum completion requires chacksumming packet body, * including udp header and folding it to skb->csum. */ -static void udp_checksum_init(struct sk_buff *skb, struct udphdr *uh, - unsigned short ulen, __be32 saddr, __be32 daddr) +static inline void udp4_csum_init(struct sk_buff *skb, struct udphdr *uh) { if (uh->check == 0) { skb->ip_summed = CHECKSUM_UNNECESSARY; } else if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!udp_check(uh, ulen, saddr, daddr, skb->csum)) + if (!csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len, IPPROTO_UDP, skb->csum )) skb->ip_summed = CHECKSUM_UNNECESSARY; } if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0); + skb->csum = csum_tcpudp_nofold(skb->nh.iph->saddr, + skb->nh.iph->daddr, + skb->len, IPPROTO_UDP, 0); /* Probably, we should checksum udp header (it should be in cache * in any case) and data in tiny packets (< rx copybreak). */ + + /* UDP = UDP-Lite with a non-partial checksum coverage */ + UDP_SKB_CB(skb)->partial_cov = 0; } /* * All we need to do is get the socket, and then do a checksum. */ -int udp_rcv(struct sk_buff *skb) +int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], + int is_udplite) { struct sock *sk; - struct udphdr *uh; + struct udphdr *uh = skb->h.uh; unsigned short ulen; struct rtable *rt = (struct rtable*)skb->dst; __be32 saddr = skb->nh.iph->saddr; __be32 daddr = skb->nh.iph->daddr; - int len = skb->len; /* - * Validate the packet and the UDP length. + * Validate the packet. */ if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto no_header; - - uh = skb->h.uh; + goto drop; /* No space for header. */ ulen = ntohs(uh->len); - - if (ulen > len || ulen < sizeof(*uh)) + if (ulen > skb->len) goto short_packet; - if (pskb_trim_rcsum(skb, ulen)) - goto short_packet; + if(! is_udplite ) { /* UDP validates ulen. */ + + if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen)) + goto short_packet; - udp_checksum_init(skb, uh, ulen, saddr, daddr); + udp4_csum_init(skb, uh); + + } else { /* UDP-Lite validates cscov. */ + if (udplite4_csum_init(skb, uh)) + goto csum_error; + } if(rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST)) - return udp_v4_mcast_deliver(skb, uh, saddr, daddr); + return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable); - sk = udp_v4_lookup(saddr, uh->source, daddr, uh->dest, skb->dev->ifindex); + sk = __udp4_lib_lookup(saddr, uh->source, daddr, uh->dest, + skb->dev->ifindex, udptable ); if (sk != NULL) { int ret = udp_queue_rcv_skb(sk, skb); sock_put(sk); /* a return value > 0 means to resubmit the input, but - * it it wants the return to be -protocol, or 0 + * it wants the return to be -protocol, or 0 */ if (ret > 0) return -ret; @@ -1201,10 +1240,10 @@ int udp_rcv(struct sk_buff *skb) nf_reset(skb); /* No socket. Drop packet silently, if checksum is wrong */ - if (udp_checksum_complete(skb)) + if (udp_lib_checksum_complete(skb)) goto csum_error; - UDP_INC_STATS_BH(UDP_MIB_NOPORTS); + UDP_INC_STATS_BH(UDP_MIB_NOPORTS, is_udplite); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); /* @@ -1215,36 +1254,40 @@ int udp_rcv(struct sk_buff *skb) return(0); short_packet: - LIMIT_NETDEBUG(KERN_DEBUG "UDP: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", + is_udplite? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), ulen, - len, + skb->len, NIPQUAD(daddr), ntohs(uh->dest)); -no_header: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return(0); + goto drop; csum_error: /* * RFC1122: OK. Discards the bad packet silently (as far as * the network is concerned, anyway) as per 4.1.3.4 (MUST). */ - LIMIT_NETDEBUG(KERN_DEBUG "UDP: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", + is_udplite? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), NIPQUAD(daddr), ntohs(uh->dest), ulen); drop: - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); kfree_skb(skb); return(0); } -static int udp_destroy_sock(struct sock *sk) +__inline__ int udp_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udp_hash, 0); +} + +int udp_destroy_sock(struct sock *sk) { lock_sock(sk); udp_flush_pending_frames(sk); @@ -1293,6 +1336,32 @@ static int do_udp_setsockopt(struct sock *sk, int level, int optname, } break; + /* + * UDP-Lite's partial checksum coverage (RFC 3828). + */ + /* The sender sets actual checksum coverage length via this option. + * The case coverage > packet length is handled by send module. */ + case UDPLITE_SEND_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Illegal coverage: use default (8) */ + val = 8; + up->pcslen = val; + up->pcflag |= UDPLITE_SEND_CC; + break; + + /* The receiver specifies a minimum checksum coverage value. To make + * sense, this should be set to at least 8 (as done below). If zero is + * used, this again means full checksum coverage. */ + case UDPLITE_RECV_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Avoid silly minimal values. */ + val = 8; + up->pcrlen = val; + up->pcflag |= UDPLITE_RECV_CC; + break; + default: err = -ENOPROTOOPT; break; @@ -1301,21 +1370,21 @@ static int do_udp_setsockopt(struct sock *sk, int level, int optname, return err; } -static int udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return ip_setsockopt(sk, level, optname, optval, optlen); - return do_udp_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_setsockopt(sk, level, optname, optval, optlen); + return ip_setsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int compat_udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return compat_ip_setsockopt(sk, level, optname, optval, optlen); - return do_udp_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_setsockopt(sk, level, optname, optval, optlen); + return compat_ip_setsockopt(sk, level, optname, optval, optlen); } #endif @@ -1342,6 +1411,16 @@ static int do_udp_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + /* The following two cannot be changed on UDP sockets, the return is + * always 0 (which corresponds to the full checksum coverage of UDP). */ + case UDPLITE_SEND_CSCOV: + val = up->pcslen; + break; + + case UDPLITE_RECV_CSCOV: + val = up->pcrlen; + break; + default: return -ENOPROTOOPT; }; @@ -1353,21 +1432,21 @@ static int do_udp_getsockopt(struct sock *sk, int level, int optname, return 0; } -static int udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return ip_getsockopt(sk, level, optname, optval, optlen); - return do_udp_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_getsockopt(sk, level, optname, optval, optlen); + return ip_getsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udp_getsockopt(struct sock *sk, int level, int optname, +int compat_udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return compat_ip_getsockopt(sk, level, optname, optval, optlen); - return do_udp_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udp_getsockopt(sk, level, optname, optval, optlen); + return compat_ip_getsockopt(sk, level, optname, optval, optlen); } #endif /** @@ -1387,7 +1466,8 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) { unsigned int mask = datagram_poll(file, sock, wait); struct sock *sk = sock->sk; - + int is_lite = IS_UDPLITE(sk); + /* Check for false positives due to checksum errors */ if ( (mask & POLLRDNORM) && !(file->f_flags & O_NONBLOCK) && @@ -1397,8 +1477,8 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) spin_lock_bh(&rcvq->lock); while ((skb = skb_peek(rcvq)) != NULL) { - if (udp_checksum_complete(skb)) { - UDP_INC_STATS_BH(UDP_MIB_INERRORS); + if (udp_lib_checksum_complete(skb)) { + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_lite); __skb_unlink(skb, rcvq); kfree_skb(skb); } else { @@ -1420,7 +1500,7 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) struct proto udp_prot = { .name = "UDP", .owner = THIS_MODULE, - .close = udp_close, + .close = udp_lib_close, .connect = ip4_datagram_connect, .disconnect = udp_disconnect, .ioctl = udp_ioctl, @@ -1431,8 +1511,8 @@ struct proto udp_prot = { .recvmsg = udp_recvmsg, .sendpage = udp_sendpage, .backlog_rcv = udp_queue_rcv_skb, - .hash = udp_v4_hash, - .unhash = udp_v4_unhash, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, .get_port = udp_v4_get_port, .obj_size = sizeof(struct udp_sock), #ifdef CONFIG_COMPAT @@ -1451,7 +1531,7 @@ static struct sock *udp_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; - sk_for_each(sk, node, &udp_hash[state->bucket]) { + sk_for_each(sk, node, state->hashtable + state->bucket) { if (sk->sk_family == state->family) goto found; } @@ -1472,7 +1552,7 @@ try_again: } while (sk && sk->sk_family != state->family); if (!sk && ++state->bucket < UDP_HTABLE_SIZE) { - sk = sk_head(&udp_hash[state->bucket]); + sk = sk_head(state->hashtable + state->bucket); goto try_again; } return sk; @@ -1522,6 +1602,7 @@ static int udp_seq_open(struct inode *inode, struct file *file) if (!s) goto out; s->family = afinfo->family; + s->hashtable = afinfo->hashtable; s->seq_ops.start = udp_seq_start; s->seq_ops.next = udp_seq_next; s->seq_ops.show = afinfo->seq_show; @@ -1588,7 +1669,7 @@ static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket) atomic_read(&sp->sk_refcnt), sp); } -static int udp4_seq_show(struct seq_file *seq, void *v) +int udp4_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) seq_printf(seq, "%-127s\n", @@ -1611,6 +1692,7 @@ static struct udp_seq_afinfo udp4_seq_afinfo = { .owner = THIS_MODULE, .name = "udp", .family = AF_INET, + .hashtable = udp_hash, .seq_show = udp4_seq_show, .seq_fops = &udp4_seq_fops, }; diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h new file mode 100644 index 000000000000..f6f4277ba6dc --- /dev/null +++ b/net/ipv4/udp_impl.h @@ -0,0 +1,38 @@ +#ifndef _UDP4_IMPL_H +#define _UDP4_IMPL_H +#include +#include +#include +#include + +extern int __udp4_lib_rcv(struct sk_buff *, struct hlist_head [], int ); +extern void __udp4_lib_err(struct sk_buff *, u32, struct hlist_head []); + +extern int __udp_lib_get_port(struct sock *sk, unsigned short snum, + struct hlist_head udptable[], int *port_rover, + int (*)(const struct sock*,const struct sock*)); +extern int ipv4_rcv_saddr_equal(const struct sock *, const struct sock *); + + +extern int udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); + +#ifdef CONFIG_COMPAT +extern int compat_udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int compat_udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +#endif +extern int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len); +extern int udp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags); +extern int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb); +extern int udp_destroy_sock(struct sock *sk); + +#ifdef CONFIG_PROC_FS +extern int udp4_seq_show(struct seq_file *seq, void *v); +#endif +#endif /* _UDP4_IMPL_H */ diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c new file mode 100644 index 000000000000..561de6d8c734 --- /dev/null +++ b/net/ipv4/udplite.c @@ -0,0 +1,119 @@ +/* + * UDPLITE An implementation of the UDP-Lite protocol (RFC 3828). + * + * Version: $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" +DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics) __read_mostly; + +struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; +static int udplite_port_rover; + +__inline__ int udplite_get_port(struct sock *sk, unsigned short p, + int (*c)(const struct sock *, const struct sock *)) +{ + return __udp_lib_get_port(sk, p, udplite_hash, &udplite_port_rover, c); +} + +static __inline__ int udplite_v4_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal); +} + +__inline__ int udplite_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udplite_hash, 1); +} + +__inline__ void udplite_err(struct sk_buff *skb, u32 info) +{ + return __udp4_lib_err(skb, info, udplite_hash); +} + +static struct net_protocol udplite_protocol = { + .handler = udplite_rcv, + .err_handler = udplite_err, + .no_policy = 1, +}; + +struct proto udplite_prot = { + .name = "UDP-Lite", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip4_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udp_destroy_sock, + .setsockopt = udp_setsockopt, + .getsockopt = udp_getsockopt, + .sendmsg = udp_sendmsg, + .recvmsg = udp_recvmsg, + .sendpage = udp_sendpage, + .backlog_rcv = udp_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v4_get_port, + .obj_size = sizeof(struct udp_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udp_setsockopt, + .compat_getsockopt = compat_udp_getsockopt, +#endif +}; + +static struct inet_protosw udplite4_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplite_prot, + .ops = &inet_dgram_ops, + .capability = -1, + .no_check = 0, /* must checksum (RFC 3828) */ + .flags = INET_PROTOSW_PERMANENT, +}; + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite4_seq_fops; +static struct udp_seq_afinfo udplite4_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite", + .family = AF_INET, + .hashtable = udplite_hash, + .seq_show = udp4_seq_show, + .seq_fops = &udplite4_seq_fops, +}; +#endif + +void __init udplite4_register(void) +{ + if (proto_register(&udplite_prot, 1)) + goto out_register_err; + + if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0) + goto out_unregister_proto; + + inet_register_protosw(&udplite4_protosw); + +#ifdef CONFIG_PROC_FS + if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */ + printk(KERN_ERR "%s: Cannot register /proc!\n", __FUNCTION__); +#endif + return; + +out_unregister_proto: + proto_unregister(&udplite_prot); +out_register_err: + printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __FUNCTION__); +} + +EXPORT_SYMBOL(udplite_hash); +EXPORT_SYMBOL(udplite_prot); +EXPORT_SYMBOL(udplite_get_port); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index da766234607b..d4107bb701b5 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -199,6 +199,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl) if (!(iph->frag_off & htons(IP_MF | IP_OFFSET))) { switch (iph->protocol) { case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_TCP: case IPPROTO_SCTP: case IPPROTO_DCCP: diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index addcc011bc01..8bacda109b7f 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -5,8 +5,8 @@ obj-$(CONFIG_IPV6) += ipv6.o ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ - route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o raw.o \ - protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ + route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ + raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ exthdrs.o sysctl_net_ipv6.o datagram.o proc.o \ ip6_flowlabel.o ipv6_syms.o inet6_connection_sock.o diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 92bfccf62cb7..1eb1c7f261d4 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #include #include @@ -737,8 +738,13 @@ static int __init init_ipv6_mibs(void) if (snmp6_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib), __alignof__(struct udp_mib)) < 0) goto err_udp_mib; + if (snmp6_mib_init((void **)udplite_stats_in6, sizeof (struct udp_mib), + __alignof__(struct udp_mib)) < 0) + goto err_udplite_mib; return 0; +err_udplite_mib: + snmp6_mib_free((void **)udp_stats_in6); err_udp_mib: snmp6_mib_free((void **)icmpv6_statistics); err_icmp_mib: @@ -753,6 +759,7 @@ static void cleanup_ipv6_mibs(void) snmp6_mib_free((void **)ipv6_statistics); snmp6_mib_free((void **)icmpv6_statistics); snmp6_mib_free((void **)udp_stats_in6); + snmp6_mib_free((void **)udplite_stats_in6); } static int __init inet6_init(void) @@ -780,10 +787,14 @@ static int __init inet6_init(void) if (err) goto out_unregister_tcp_proto; - err = proto_register(&rawv6_prot, 1); + err = proto_register(&udplitev6_prot, 1); if (err) goto out_unregister_udp_proto; + err = proto_register(&rawv6_prot, 1); + if (err) + goto out_unregister_udplite_proto; + /* Register the socket-side information for inet6_create. */ for(r = &inetsw6[0]; r < &inetsw6[SOCK_MAX]; ++r) @@ -837,6 +848,8 @@ static int __init inet6_init(void) goto proc_tcp6_fail; if (udp6_proc_init()) goto proc_udp6_fail; + if (udplite6_proc_init()) + goto proc_udplite6_fail; if (ipv6_misc_proc_init()) goto proc_misc6_fail; @@ -862,6 +875,7 @@ static int __init inet6_init(void) /* Init v6 transport protocols. */ udpv6_init(); + udplitev6_init(); tcpv6_init(); ipv6_packet_init(); @@ -879,6 +893,8 @@ proc_if6_fail: proc_anycast6_fail: ipv6_misc_proc_exit(); proc_misc6_fail: + udplite6_proc_exit(); +proc_udplite6_fail: udp6_proc_exit(); proc_udp6_fail: tcp6_proc_exit(); @@ -902,6 +918,8 @@ out_unregister_sock: sock_unregister(PF_INET6); out_unregister_raw_proto: proto_unregister(&rawv6_prot); +out_unregister_udplite_proto: + proto_unregister(&udplitev6_prot); out_unregister_udp_proto: proto_unregister(&udpv6_prot); out_unregister_tcp_proto: @@ -919,6 +937,7 @@ static void __exit inet6_exit(void) ac6_proc_exit(); ipv6_misc_proc_exit(); udp6_proc_exit(); + udplite6_proc_exit(); tcp6_proc_exit(); raw6_proc_exit(); #endif diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index de6b91981b30..1eafcfc95e81 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include @@ -239,6 +240,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, struct sk_buff *pktopt; if (sk->sk_protocol != IPPROTO_UDP && + sk->sk_protocol != IPPROTO_UDPLITE && sk->sk_protocol != IPPROTO_TCP) break; @@ -276,11 +278,15 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, sk->sk_family = PF_INET; tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); } else { + struct proto *prot = &udp_prot; + + if (sk->sk_protocol == IPPROTO_UDPLITE) + prot = &udplite_prot; local_bh_disable(); sock_prot_dec_use(sk->sk_prot); - sock_prot_inc_use(&udp_prot); + sock_prot_inc_use(prot); local_bh_enable(); - sk->sk_prot = &udp_prot; + sk->sk_prot = prot; sk->sk_socket->ops = &inet_dgram_ops; sk->sk_family = PF_INET; } @@ -813,6 +819,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case IPV6_ADDRFORM: if (sk->sk_protocol != IPPROTO_UDP && + sk->sk_protocol != IPPROTO_UDPLITE && sk->sk_protocol != IPPROTO_TCP) return -EINVAL; if (sk->sk_state != TCP_ESTABLISHED) diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index 0cf537d30185..3cb6bb79cc05 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -270,11 +270,15 @@ static void dump_packet(const struct nf_loginfo *info, } break; } - case IPPROTO_UDP: { + case IPPROTO_UDP: + case IPPROTO_UDPLITE: { struct udphdr _udph, *uh; - /* Max length: 10 "PROTO=UDP " */ - printk("PROTO=UDP "); + if (currenthdr == IPPROTO_UDP) + /* Max length: 10 "PROTO=UDP " */ + printk("PROTO=UDP " ); + else /* Max length: 14 "PROTO=UDPLITE " */ + printk("PROTO=UDPLITE "); if (fragment) break; diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 4158d386b0aa..35249d8487bb 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -49,6 +49,8 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v) fold_prot_inuse(&tcpv6_prot)); seq_printf(seq, "UDP6: inuse %d\n", fold_prot_inuse(&udpv6_prot)); + seq_printf(seq, "UDPLITE6: inuse %d\n", + fold_prot_inuse(&udplitev6_prot)); seq_printf(seq, "RAW6: inuse %d\n", fold_prot_inuse(&rawv6_prot)); seq_printf(seq, "FRAG6: inuse %d memory %d\n", @@ -133,6 +135,14 @@ static struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_SENTINEL }; +static struct snmp_mib snmp6_udplite6_list[] = { + SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS), + SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS), + SNMP_MIB_ITEM("UdpLite6InErrors", UDP_MIB_INERRORS), + SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS), + SNMP_MIB_SENTINEL +}; + static unsigned long fold_field(void *mib[], int offt) { @@ -167,6 +177,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) snmp6_seq_show_item(seq, (void **)ipv6_statistics, snmp6_ipstats_list); snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list); snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list); + snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list); } return 0; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 886300d13a59..5a64027bf2fc 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -38,26 +38,18 @@ #include #include -#include -#include - -#include #include #include #include #include -#include -#include -#include #include -#include #include - #include #include #include #include +#include "udp_impl.h" DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly; @@ -66,23 +58,9 @@ static inline int udp_v6_get_port(struct sock *sk, unsigned short snum) return udp_get_port(sk, snum, ipv6_rcv_saddr_equal); } -static void udp_v6_hash(struct sock *sk) -{ - BUG(); -} - -static void udp_v6_unhash(struct sock *sk) -{ - write_lock_bh(&udp_hash_lock); - if (sk_del_node_init(sk)) { - inet_sk(sk)->num = 0; - sock_prot_dec_use(sk->sk_prot); - } - write_unlock_bh(&udp_hash_lock); -} - -static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, - struct in6_addr *daddr, u16 dport, int dif) +static struct sock *__udp6_lib_lookup(struct in6_addr *saddr, __be16 sport, + struct in6_addr *daddr, __be16 dport, + int dif, struct hlist_head udptable[]) { struct sock *sk, *result = NULL; struct hlist_node *node; @@ -90,7 +68,7 @@ static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, int badness = -1; read_lock(&udp_hash_lock); - sk_for_each(sk, node, &udp_hash[hnum & (UDP_HTABLE_SIZE - 1)]) { + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); if (inet->num == hnum && sk->sk_family == PF_INET6) { @@ -131,21 +109,12 @@ static struct sock *udp_v6_lookup(struct in6_addr *saddr, u16 sport, return result; } -/* - * - */ - -static void udpv6_close(struct sock *sk, long timeout) -{ - sk_common_release(sk); -} - /* * This should be easy, if there is something there we * return it, otherwise we block. */ -static int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, +int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len, int noblock, int flags, int *addr_len) { @@ -153,7 +122,7 @@ static int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, struct inet_sock *inet = inet_sk(sk); struct sk_buff *skb; size_t copied; - int err; + int err, copy_only, is_udplite = IS_UDPLITE(sk); if (addr_len) *addr_len=sizeof(struct sockaddr_in6); @@ -172,15 +141,21 @@ try_again: msg->msg_flags |= MSG_TRUNC; } - if (skb->ip_summed==CHECKSUM_UNNECESSARY) { - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else if (msg->msg_flags&MSG_TRUNC) { - if (__skb_checksum_complete(skb)) + /* + * Decide whether to checksum and/or copy data. + */ + copy_only = (skb->ip_summed==CHECKSUM_UNNECESSARY); + + if (is_udplite || (!copy_only && msg->msg_flags&MSG_TRUNC)) { + if (__udp_lib_checksum_complete(skb)) goto csum_copy_err; - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, - copied); - } else { + copy_only = 1; + } + + if (copy_only) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); if (err == -EINVAL) goto csum_copy_err; @@ -231,14 +206,15 @@ csum_copy_err: skb_kill_datagram(sk, skb, flags); if (flags & MSG_DONTWAIT) { - UDP6_INC_STATS_USER(UDP_MIB_INERRORS); + UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); return -EAGAIN; } goto try_again; } -static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __be32 info) +void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + int type, int code, int offset, __be32 info, + struct hlist_head udptable[] ) { struct ipv6_pinfo *np; struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; @@ -248,8 +224,8 @@ static void udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct sock *sk; int err; - sk = udp_v6_lookup(daddr, uh->dest, saddr, uh->source, inet6_iif(skb)); - + sk = __udp6_lib_lookup(daddr, uh->dest, + saddr, uh->source, inet6_iif(skb), udptable); if (sk == NULL) return; @@ -270,31 +246,55 @@ out: sock_put(sk); } -static inline int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +static __inline__ void udpv6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, int type, + int code, int offset, __u32 info ) +{ + return __udp6_lib_err(skb, opt, type, code, offset, info, udp_hash); +} + +int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) { + struct udp_sock *up = udp_sk(sk); int rc; - if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) { - kfree_skb(skb); - return -1; - } + if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; - if (skb_checksum_complete(skb)) { - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; + /* + * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c). + */ + if ((up->pcflag & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage" + " %d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; + } + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d " + "too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } } + if (udp_lib_checksum_complete(skb)) + goto drop; + if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { /* Note that an ENOMEM error is charged twice */ if (rc == -ENOMEM) - UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS); - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; + UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag); + goto drop; } - UDP6_INC_STATS_BH(UDP_MIB_INDATAGRAMS); + UDP6_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag); return 0; +drop: + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag); + kfree_skb(skb); + return -1; } static struct sock *udp_v6_mcast_next(struct sock *sk, @@ -338,15 +338,15 @@ static struct sock *udp_v6_mcast_next(struct sock *sk, * Note: called only from the BH handler context, * so we don't need to lock the hashes. */ -static void udpv6_mcast_deliver(struct udphdr *uh, - struct in6_addr *saddr, struct in6_addr *daddr, - struct sk_buff *skb) +static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr, + struct in6_addr *daddr, struct hlist_head udptable[]) { struct sock *sk, *sk2; + const struct udphdr *uh = skb->h.uh; int dif; read_lock(&udp_hash_lock); - sk = sk_head(&udp_hash[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); dif = inet6_iif(skb); sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); if (!sk) { @@ -364,9 +364,34 @@ static void udpv6_mcast_deliver(struct udphdr *uh, udpv6_queue_rcv_skb(sk, skb); out: read_unlock(&udp_hash_lock); + return 0; } -static int udpv6_rcv(struct sk_buff **pskb) +static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh) + +{ + if (uh->check == 0) { + /* RFC 2460 section 8.1 says that we SHOULD log + this error. Well, it is reasonable. + */ + LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); + return 1; + } + if (skb->ip_summed == CHECKSUM_COMPLETE && + !csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDP, skb->csum )) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, IPPROTO_UDP, 0); + + return (UDP_SKB_CB(skb)->partial_cov = 0); +} + +int __udp6_lib_rcv(struct sk_buff **pskb, struct hlist_head udptable[], + int is_udplite) { struct sk_buff *skb = *pskb; struct sock *sk; @@ -383,44 +408,39 @@ static int udpv6_rcv(struct sk_buff **pskb) uh = skb->h.uh; ulen = ntohs(uh->len); + if (ulen > skb->len) + goto short_packet; - /* Check for jumbo payload */ - if (ulen == 0) - ulen = skb->len; + if(! is_udplite ) { /* UDP validates ulen. */ - if (ulen > skb->len || ulen < sizeof(*uh)) - goto short_packet; + /* Check for jumbo payload */ + if (ulen == 0) + ulen = skb->len; - if (uh->check == 0) { - /* RFC 2460 section 8.1 says that we SHOULD log - this error. Well, it is reasonable. - */ - LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); - goto discard; - } + if (ulen < sizeof(*uh)) + goto short_packet; - if (ulen < skb->len) { - if (pskb_trim_rcsum(skb, ulen)) - goto discard; - saddr = &skb->nh.ipv6h->saddr; - daddr = &skb->nh.ipv6h->daddr; - uh = skb->h.uh; - } + if (ulen < skb->len) { + if (pskb_trim_rcsum(skb, ulen)) + goto short_packet; + saddr = &skb->nh.ipv6h->saddr; + daddr = &skb->nh.ipv6h->daddr; + uh = skb->h.uh; + } - if (skb->ip_summed == CHECKSUM_COMPLETE && - !csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (udp6_csum_init(skb, uh)) + goto discard; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->csum = ~csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, 0); + } else { /* UDP-Lite validates cscov. */ + if (udplite6_csum_init(skb, uh)) + goto discard; + } /* * Multicast receive code */ - if (ipv6_addr_is_multicast(daddr)) { - udpv6_mcast_deliver(uh, saddr, daddr, skb); - return 0; - } + if (ipv6_addr_is_multicast(daddr)) + return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable); /* Unicast */ @@ -428,15 +448,16 @@ static int udpv6_rcv(struct sk_buff **pskb) * check socket cache ... must talk to Alan about his plans * for sock caches... i'll skip this for now. */ - sk = udp_v6_lookup(saddr, uh->source, daddr, uh->dest, inet6_iif(skb)); + sk = __udp6_lib_lookup(saddr, uh->source, + daddr, uh->dest, inet6_iif(skb), udptable); if (sk == NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; - if (skb_checksum_complete(skb)) + if (udp_lib_checksum_complete(skb)) goto discard; - UDP6_INC_STATS_BH(UDP_MIB_NOPORTS); + UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, is_udplite); icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); @@ -451,14 +472,20 @@ static int udpv6_rcv(struct sk_buff **pskb) return(0); short_packet: - if (net_ratelimit()) - printk(KERN_DEBUG "UDP: short packet: %d/%u\n", ulen, skb->len); + LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n", + is_udplite? "-Lite" : "", ulen, skb->len); discard: - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); kfree_skb(skb); return(0); } + +static __inline__ int udpv6_rcv(struct sk_buff **pskb) +{ + return __udp6_lib_rcv(pskb, udp_hash, 0); +} + /* * Throw away all pending data and cancel the corking. Socket is locked. */ @@ -484,6 +511,7 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) struct inet_sock *inet = inet_sk(sk); struct flowi *fl = &inet->cork.fl; int err = 0; + u32 csum = 0; /* Grab the skbuff where UDP header space exists. */ if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) @@ -498,35 +526,17 @@ static int udp_v6_push_pending_frames(struct sock *sk, struct udp_sock *up) uh->len = htons(up->len); uh->check = 0; - if (sk->sk_no_check == UDP_CSUM_NOXMIT) { - skb->ip_summed = CHECKSUM_NONE; - goto send; - } - - if (skb_queue_len(&sk->sk_write_queue) == 1) { - skb->csum = csum_partial((char *)uh, - sizeof(struct udphdr), skb->csum); - uh->check = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - up->len, fl->proto, skb->csum); - } else { - u32 tmp_csum = 0; - - skb_queue_walk(&sk->sk_write_queue, skb) { - tmp_csum = csum_add(tmp_csum, skb->csum); - } - tmp_csum = csum_partial((char *)uh, - sizeof(struct udphdr), tmp_csum); - tmp_csum = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - up->len, fl->proto, tmp_csum); - uh->check = tmp_csum; + if (up->pcflag) + csum = udplite_csum_outgoing(sk, skb); + else + csum = udp_csum_outgoing(sk, skb); - } + /* add protocol-dependent pseudo-header */ + uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, + up->len, fl->proto, csum ); if (uh->check == 0) uh->check = -1; -send: err = ip6_push_pending_frames(sk); out: up->len = 0; @@ -534,7 +544,7 @@ out: return err; } -static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, +int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len) { struct ipv6_txoptions opt_space; @@ -554,6 +564,8 @@ static int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; int err; int connected = 0; + int is_udplite = up->pcflag; + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); /* destination address check */ if (sin6) { @@ -694,7 +706,7 @@ do_udp_sendmsg: opt = fl6_merge_options(&opt_space, flowlabel, opt); opt = ipv6_fixup_options(&opt_space, opt); - fl.proto = IPPROTO_UDP; + fl.proto = sk->sk_protocol; ipv6_addr_copy(&fl.fl6_dst, daddr); if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr)) ipv6_addr_copy(&fl.fl6_src, &np->saddr); @@ -761,7 +773,8 @@ back_from_confirm: do_append_data: up->len += ulen; - err = ip6_append_data(sk, ip_generic_getfrag, msg->msg_iov, ulen, + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen, sizeof(struct udphdr), hlimit, tclass, opt, &fl, (struct rt6_info*)dst, corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); @@ -793,7 +806,7 @@ do_append_data: out: fl6_sock_release(flowlabel); if (!err) { - UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS); + UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); return len; } /* @@ -804,7 +817,7 @@ out: * seems like overkill. */ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS); + UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); } return err; @@ -816,7 +829,7 @@ do_confirm: goto out; } -static int udpv6_destroy_sock(struct sock *sk) +int udpv6_destroy_sock(struct sock *sk) { lock_sock(sk); udp_v6_flush_pending_frames(sk); @@ -854,7 +867,6 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, release_sock(sk); } break; - case UDP_ENCAP: switch (val) { case 0: @@ -866,6 +878,24 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, } break; + case UDPLITE_SEND_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Illegal coverage: use default (8) */ + val = 8; + up->pcslen = val; + up->pcflag |= UDPLITE_SEND_CC; + break; + + case UDPLITE_RECV_CSCOV: + if (!up->pcflag) /* Disable the option on UDP sockets */ + return -ENOPROTOOPT; + if (val != 0 && val < 8) /* Avoid silly minimal values. */ + val = 8; + up->pcrlen = val; + up->pcflag |= UDPLITE_RECV_CC; + break; + default: err = -ENOPROTOOPT; break; @@ -874,22 +904,21 @@ static int do_udpv6_setsockopt(struct sock *sk, int level, int optname, return err; } -static int udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return ipv6_setsockopt(sk, level, optname, optval, optlen); - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return ipv6_setsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) +int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) { - if (level != SOL_UDP) - return compat_ipv6_setsockopt(sk, level, optname, - optval, optlen); - return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_setsockopt(sk, level, optname, optval, optlen); + return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); } #endif @@ -916,6 +945,14 @@ static int do_udpv6_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + case UDPLITE_SEND_CSCOV: + val = up->pcslen; + break; + + case UDPLITE_RECV_CSCOV: + val = up->pcrlen; + break; + default: return -ENOPROTOOPT; }; @@ -927,22 +964,21 @@ static int do_udpv6_getsockopt(struct sock *sk, int level, int optname, return 0; } -static int udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return ipv6_getsockopt(sk, level, optname, optval, optlen); - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return ipv6_getsockopt(sk, level, optname, optval, optlen); } #ifdef CONFIG_COMPAT -static int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) +int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) { - if (level != SOL_UDP) - return compat_ipv6_getsockopt(sk, level, optname, - optval, optlen); - return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + if (level == SOL_UDP || level == SOL_UDPLITE) + return do_udpv6_getsockopt(sk, level, optname, optval, optlen); + return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); } #endif @@ -983,7 +1019,7 @@ static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket atomic_read(&sp->sk_refcnt), sp); } -static int udp6_seq_show(struct seq_file *seq, void *v) +int udp6_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) seq_printf(seq, @@ -1002,6 +1038,7 @@ static struct udp_seq_afinfo udp6_seq_afinfo = { .owner = THIS_MODULE, .name = "udp6", .family = AF_INET6, + .hashtable = udp_hash, .seq_show = udp6_seq_show, .seq_fops = &udp6_seq_fops, }; @@ -1021,7 +1058,7 @@ void udp6_proc_exit(void) { struct proto udpv6_prot = { .name = "UDPv6", .owner = THIS_MODULE, - .close = udpv6_close, + .close = udp_lib_close, .connect = ip6_datagram_connect, .disconnect = udp_disconnect, .ioctl = udp_ioctl, @@ -1031,8 +1068,8 @@ struct proto udpv6_prot = { .sendmsg = udpv6_sendmsg, .recvmsg = udpv6_recvmsg, .backlog_rcv = udpv6_queue_rcv_skb, - .hash = udp_v6_hash, - .unhash = udp_v6_unhash, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, .get_port = udp_v6_get_port, .obj_size = sizeof(struct udp6_sock), #ifdef CONFIG_COMPAT diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h new file mode 100644 index 000000000000..ec9878899128 --- /dev/null +++ b/net/ipv6/udp_impl.h @@ -0,0 +1,34 @@ +#ifndef _UDP6_IMPL_H +#define _UDP6_IMPL_H +#include +#include +#include +#include +#include + +extern int __udp6_lib_rcv(struct sk_buff **, struct hlist_head [], int ); +extern void __udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *, + int , int , int , __be32 , struct hlist_head []); + +extern int udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +extern int udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +#ifdef CONFIG_COMPAT +extern int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen); +extern int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen); +#endif +extern int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len); +extern int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len, + int noblock, int flags, int *addr_len); +extern int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb); +extern int udpv6_destroy_sock(struct sock *sk); + +#ifdef CONFIG_PROC_FS +extern int udp6_seq_show(struct seq_file *seq, void *v); +#endif +#endif /* _UDP6_IMPL_H */ diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c new file mode 100644 index 000000000000..e0ec5e63004a --- /dev/null +++ b/net/ipv6/udplite.c @@ -0,0 +1,105 @@ +/* + * UDPLITEv6 An implementation of the UDP-Lite protocol over IPv6. + * See also net/ipv4/udplite.c + * + * Version: $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" + +DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly; + +static __inline__ int udplitev6_rcv(struct sk_buff **pskb) +{ + return __udp6_lib_rcv(pskb, udplite_hash, 1); +} + +static __inline__ void udplitev6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, + int type, int code, int offset, __u32 info) +{ + return __udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash); +} + +static struct inet6_protocol udplitev6_protocol = { + .handler = udplitev6_rcv, + .err_handler = udplitev6_err, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, +}; + +static __inline__ int udplite_v6_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal); +} + +struct proto udplitev6_prot = { + .name = "UDPLITEv6", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip6_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udpv6_destroy_sock, + .setsockopt = udpv6_setsockopt, + .getsockopt = udpv6_getsockopt, + .sendmsg = udpv6_sendmsg, + .recvmsg = udpv6_recvmsg, + .backlog_rcv = udpv6_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v6_get_port, + .obj_size = sizeof(struct udp6_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udpv6_setsockopt, + .compat_getsockopt = compat_udpv6_getsockopt, +#endif +}; + +static struct inet_protosw udplite6_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplitev6_prot, + .ops = &inet6_dgram_ops, + .capability = -1, + .no_check = 0, + .flags = INET_PROTOSW_PERMANENT, +}; + +void __init udplitev6_init(void) +{ + if (inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE) < 0) + printk(KERN_ERR "%s: Could not register.\n", __FUNCTION__); + + inet6_register_protosw(&udplite6_protosw); +} + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite6_seq_fops; +static struct udp_seq_afinfo udplite6_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite6", + .family = AF_INET6, + .hashtable = udplite_hash, + .seq_show = udp6_seq_show, + .seq_fops = &udplite6_seq_fops, +}; + +int __init udplite6_proc_init(void) +{ + return udp_proc_register(&udplite6_seq_afinfo); +} + +void udplite6_proc_exit(void) +{ + udp_proc_unregister(&udplite6_seq_afinfo); +} +#endif diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 2fba1f0739aa..8dffd4daae9c 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -274,6 +274,7 @@ _decode_session6(struct sk_buff *skb, struct flowi *fl) break; case IPPROTO_UDP: + case IPPROTO_UDPLITE: case IPPROTO_TCP: case IPPROTO_SCTP: case IPPROTO_DCCP: diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c index b4293058c6ff..1602086c7fd6 100644 --- a/net/netfilter/xt_multiport.c +++ b/net/netfilter/xt_multiport.c @@ -1,5 +1,5 @@ -/* Kernel module to match one of a list of TCP/UDP/SCTP/DCCP ports: ports are in - the same place so we can treat them as equal. */ +/* Kernel module to match one of a list of TCP/UDP(-Lite)/SCTP/DCCP ports: + ports are in the same place so we can treat them as equal. */ /* (C) 1999-2001 Paul `Rusty' Russell * (C) 2002-2004 Netfilter Core Team @@ -162,6 +162,7 @@ check(u_int16_t proto, { /* Must specify supported protocol, no unknown flags or bad count */ return (proto == IPPROTO_TCP || proto == IPPROTO_UDP + || proto == IPPROTO_UDPLITE || proto == IPPROTO_SCTP || proto == IPPROTO_DCCP) && !(ip_invflags & XT_INV_PROTO) && (match_flags == XT_MULTIPORT_SOURCE diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index e76a68e0bc66..46414b562a19 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -10,7 +10,7 @@ #include #include -MODULE_DESCRIPTION("x_tables match for TCP and UDP, supports IPv4 and IPv6"); +MODULE_DESCRIPTION("x_tables match for TCP and UDP(-Lite), supports IPv4 and IPv6"); MODULE_LICENSE("GPL"); MODULE_ALIAS("xt_tcp"); MODULE_ALIAS("xt_udp"); @@ -234,6 +234,24 @@ static struct xt_match xt_tcpudp_match[] = { .proto = IPPROTO_UDP, .me = THIS_MODULE, }, + { + .name = "udplite", + .family = AF_INET, + .checkentry = udp_checkentry, + .match = udp_match, + .matchsize = sizeof(struct xt_udp), + .proto = IPPROTO_UDPLITE, + .me = THIS_MODULE, + }, + { + .name = "udplite", + .family = AF_INET6, + .checkentry = udp_checkentry, + .match = udp_match, + .matchsize = sizeof(struct xt_udp), + .proto = IPPROTO_UDPLITE, + .me = THIS_MODULE, + }, }; static int __init xt_tcpudp_init(void) -- cgit v1.2.3 From 714e85be3557222bc25f69c252326207c900a7db Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:51:49 -0800 Subject: [IPV6]: Assorted trivial endianness annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 14 ++++++++------ include/net/arp.h | 2 +- include/net/ip.h | 11 ++++++----- net/ipv4/af_inet.c | 6 +++--- net/ipv4/arp.c | 2 +- net/ipv4/cipso_ipv4.c | 8 ++++---- net/ipv4/devinet.c | 10 +++++----- net/ipv4/ip_output.c | 2 +- net/ipv4/ip_sockglue.c | 2 +- net/ipv4/raw.c | 4 ++-- net/ipv4/route.c | 4 ++-- net/ipv4/syncookies.c | 18 +++++++++--------- net/ipv4/tcp_ipv4.c | 10 +++++----- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_timer.c | 2 +- 15 files changed, 50 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 5a0ab04627bc..c0f7aec331c2 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -124,12 +124,13 @@ static __inline__ int inet_ifa_match(__be32 addr, struct in_ifaddr *ifa) * Check if a mask is acceptable. */ -static __inline__ int bad_mask(u32 mask, u32 addr) +static __inline__ int bad_mask(__be32 mask, __be32 addr) { + __u32 hmask; if (addr & (mask = ~mask)) return 1; - mask = ntohl(mask); - if (mask & (mask+1)) + hmask = ntohl(mask); + if (hmask & (hmask+1)) return 1; return 0; } @@ -190,11 +191,12 @@ static __inline__ __be32 inet_make_mask(int logmask) return 0; } -static __inline__ int inet_mask_len(__u32 mask) +static __inline__ int inet_mask_len(__be32 mask) { - if (!(mask = ntohl(mask))) + __u32 hmask = ntohl(mask); + if (!hmask) return 0; - return 32 - ffz(~mask); + return 32 - ffz(~hmask); } diff --git a/include/net/arp.h b/include/net/arp.h index 6a3d9a7d302b..f02664568600 100644 --- a/include/net/arp.h +++ b/include/net/arp.h @@ -16,7 +16,7 @@ extern void arp_send(int type, int ptype, __be32 dest_ip, struct net_device *dev, __be32 src_ip, unsigned char *dest_hw, unsigned char *src_hw, unsigned char *th); extern int arp_bind_neighbour(struct dst_entry *dst); -extern int arp_mc_map(u32 addr, u8 *haddr, struct net_device *dev, int dir); +extern int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir); extern void arp_ifdown(struct net_device *dev); extern struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip, diff --git a/include/net/ip.h b/include/net/ip.h index 949fa8683626..412e8114667d 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -123,7 +123,7 @@ extern int ip4_datagram_connect(struct sock *sk, * multicast packets. */ -static inline void ip_tr_mc_map(u32 addr, char *buf) +static inline void ip_tr_mc_map(__be32 addr, char *buf) { buf[0]=0xC0; buf[1]=0x00; @@ -238,9 +238,9 @@ static inline void ip_select_ident_more(struct iphdr *iph, struct dst_entry *dst * Map a multicast IP onto multicast MAC for type ethernet. */ -static inline void ip_eth_mc_map(u32 addr, char *buf) +static inline void ip_eth_mc_map(__be32 naddr, char *buf) { - addr=ntohl(addr); + __u32 addr=ntohl(naddr); buf[0]=0x01; buf[1]=0x00; buf[2]=0x5e; @@ -256,13 +256,14 @@ static inline void ip_eth_mc_map(u32 addr, char *buf) * Leave P_Key as 0 to be filled in by driver. */ -static inline void ip_ib_mc_map(u32 addr, char *buf) +static inline void ip_ib_mc_map(__be32 naddr, char *buf) { + __u32 addr; buf[0] = 0; /* Reserved */ buf[1] = 0xff; /* Multicast QPN */ buf[2] = 0xff; buf[3] = 0xff; - addr = ntohl(addr); + addr = ntohl(naddr); buf[4] = 0xff; buf[5] = 0x12; /* link local scope */ buf[6] = 0x40; /* IPv4 signature */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 8db39f7e3bf0..1144900d37f6 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -644,7 +644,7 @@ int inet_getname(struct socket *sock, struct sockaddr *uaddr, sin->sin_port = inet->dport; sin->sin_addr.s_addr = inet->daddr; } else { - __u32 addr = inet->rcv_saddr; + __be32 addr = inet->rcv_saddr; if (!addr) addr = inet->saddr; sin->sin_port = inet->sport; @@ -995,8 +995,8 @@ static int inet_sk_reselect_saddr(struct sock *sk) struct inet_sock *inet = inet_sk(sk); int err; struct rtable *rt; - __u32 old_saddr = inet->saddr; - __u32 new_saddr; + __be32 old_saddr = inet->saddr; + __be32 new_saddr; __be32 daddr = inet->daddr; if (inet->opt && inet->opt->srr) diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index cfb5d3de9c84..3981e8be9ab8 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -203,7 +203,7 @@ struct neigh_table arp_tbl = { .gc_thresh3 = 1024, }; -int arp_mc_map(u32 addr, u8 *haddr, struct net_device *dev, int dir) +int arp_mc_map(__be32 addr, u8 *haddr, struct net_device *dev, int dir) { switch (dev->type) { case ARPHRD_ETHER: diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 6460233407c7..c3a61ebbadef 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -966,7 +966,7 @@ static int cipso_v4_gentag_hdr(const struct cipso_v4_doi *doi_def, buf[0] = IPOPT_CIPSO; buf[1] = CIPSO_V4_HDR_LEN + len; - *(u32 *)&buf[2] = htonl(doi_def->doi); + *(__be32 *)&buf[2] = htonl(doi_def->doi); return 0; } @@ -1140,7 +1140,7 @@ int cipso_v4_validate(unsigned char **option) } rcu_read_lock(); - doi_def = cipso_v4_doi_getdef(ntohl(*((u32 *)&opt[2]))); + doi_def = cipso_v4_doi_getdef(ntohl(*((__be32 *)&opt[2]))); if (doi_def == NULL) { err_offset = 2; goto validate_return_locked; @@ -1370,7 +1370,7 @@ int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) if (ret_val == 0) return ret_val; - doi = ntohl(*(u32 *)&cipso_ptr[2]); + doi = ntohl(*(__be32 *)&cipso_ptr[2]); rcu_read_lock(); doi_def = cipso_v4_doi_getdef(doi); if (doi_def == NULL) { @@ -1436,7 +1436,7 @@ int cipso_v4_skbuff_getattr(const struct sk_buff *skb, if (cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr) == 0) return 0; - doi = ntohl(*(u32 *)&cipso_ptr[2]); + doi = ntohl(*(__be32 *)&cipso_ptr[2]); rcu_read_lock(); doi_def = cipso_v4_doi_getdef(doi); if (doi_def == NULL) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index f38cbbae0ae3..a1b356c8aa59 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -577,20 +577,20 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg * Determine a default network mask, based on the IP address. */ -static __inline__ int inet_abc_len(u32 addr) +static __inline__ int inet_abc_len(__be32 addr) { int rc = -1; /* Something else, probably a multicast. */ if (ZERONET(addr)) rc = 0; else { - addr = ntohl(addr); + __u32 haddr = ntohl(addr); - if (IN_CLASSA(addr)) + if (IN_CLASSA(haddr)) rc = 8; - else if (IN_CLASSB(addr)) + else if (IN_CLASSB(haddr)) rc = 16; - else if (IN_CLASSC(addr)) + else if (IN_CLASSC(haddr)) rc = 24; } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index def32d8d3b06..90942a384a45 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -341,7 +341,7 @@ packet_routed: /* OK, we know where to send it, allocate and build IP header. */ iph = (struct iphdr *) skb_push(skb, sizeof(struct iphdr) + (opt ? opt->optlen : 0)); - *((__u16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff)); + *((__be16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff)); iph->tot_len = htons(skb->len); if (ip_dont_fragment(sk, &rt->u.dst) && !ipfragok) iph->frag_off = htons(IP_DF); diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 4b132953bcc2..57d4bae6f080 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -355,7 +355,7 @@ int ip_recv_error(struct sock *sk, struct msghdr *msg, int len) sin = (struct sockaddr_in *)msg->msg_name; if (sin) { sin->sin_family = AF_INET; - sin->sin_addr.s_addr = *(u32*)(skb->nh.raw + serr->addr_offset); + sin->sin_addr.s_addr = *(__be32*)(skb->nh.raw + serr->addr_offset); sin->sin_port = serr->port; memset(&sin->sin_zero, 0, sizeof(sin->sin_zero)); } diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 5c31dead2bdc..a6c63bbd9ddb 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -854,8 +854,8 @@ static void raw_seq_stop(struct seq_file *seq, void *v) static __inline__ char *get_raw_sock(struct sock *sp, char *tmpbuf, int i) { struct inet_sock *inet = inet_sk(sp); - unsigned int dest = inet->daddr, - src = inet->rcv_saddr; + __be32 dest = inet->daddr, + src = inet->rcv_saddr; __u16 destp = 0, srcp = inet->num; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index d7152b2b2c64..ee00b6506ab4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -566,8 +566,8 @@ static inline u32 rt_score(struct rtable *rt) static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) { - return ((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | - (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr) | + return ((__force u32)((fl1->nl_u.ip4_u.daddr ^ fl2->nl_u.ip4_u.daddr) | + (fl1->nl_u.ip4_u.saddr ^ fl2->nl_u.ip4_u.saddr)) | (fl1->mark ^ fl2->mark) | (*(u16 *)&fl1->nl_u.ip4_u.tos ^ *(u16 *)&fl2->nl_u.ip4_u.tos) | diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 661e0a4bca72..6b19530905af 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -35,23 +35,23 @@ module_init(init_syncookies); #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) -static u32 cookie_hash(u32 saddr, u32 daddr, u32 sport, u32 dport, +static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, u32 count, int c) { __u32 tmp[16 + 5 + SHA_WORKSPACE_WORDS]; memcpy(tmp + 3, syncookie_secret[c], sizeof(syncookie_secret[c])); - tmp[0] = saddr; - tmp[1] = daddr; - tmp[2] = (sport << 16) + dport; + tmp[0] = (__force u32)saddr; + tmp[1] = (__force u32)daddr; + tmp[2] = ((__force u32)sport << 16) + (__force u32)dport; tmp[3] = count; sha_transform(tmp + 16, (__u8 *)tmp, tmp + 16 + 5); return tmp[17]; } -static __u32 secure_tcp_syn_cookie(__u32 saddr, __u32 daddr, __u16 sport, - __u16 dport, __u32 sseq, __u32 count, +static __u32 secure_tcp_syn_cookie(__be32 saddr, __be32 daddr, __be16 sport, + __be16 dport, __u32 sseq, __u32 count, __u32 data) { /* @@ -80,8 +80,8 @@ static __u32 secure_tcp_syn_cookie(__u32 saddr, __u32 daddr, __u16 sport, * "maxdiff" if the current (passed-in) "count". The return value * is (__u32)-1 if this test fails. */ -static __u32 check_tcp_syn_cookie(__u32 cookie, __u32 saddr, __u32 daddr, - __u16 sport, __u16 dport, __u32 sseq, +static __u32 check_tcp_syn_cookie(__u32 cookie, __be32 saddr, __be32 daddr, + __be16 sport, __be16 dport, __u32 sseq, __u32 count, __u32 maxdiff) { __u32 diff; @@ -220,7 +220,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, } ireq = inet_rsk(req); treq = tcp_rsk(req); - treq->rcv_isn = htonl(skb->h.th->seq) - 1; + treq->rcv_isn = ntohl(skb->h.th->seq) - 1; treq->snt_isn = cookie; req->mss = mss; ireq->rmt_port = skb->h.th->source; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 8c8e8112f98d..0ca8dead03b0 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -542,7 +542,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) struct { struct tcphdr th; #ifdef CONFIG_TCP_MD5SIG - u32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; + __be32 opt[(TCPOLEN_MD5SIG_ALIGNED >> 2)]; #endif } rep; struct ip_reply_arg arg; @@ -618,9 +618,9 @@ static void tcp_v4_send_ack(struct tcp_timewait_sock *twsk, struct tcphdr *th = skb->h.th; struct { struct tcphdr th; - u32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) + __be32 opt[(TCPOLEN_TSTAMP_ALIGNED >> 2) #ifdef CONFIG_TCP_MD5SIG - + (TCPOLEN_MD5SIG_ALIGNED >> 2) + + (TCPOLEN_MD5SIG_ALIGNED >> 2) #endif ]; } rep; @@ -2333,8 +2333,8 @@ static void get_tcp4_sock(struct sock *sp, char *tmpbuf, int i) struct tcp_sock *tp = tcp_sk(sp); const struct inet_connection_sock *icsk = inet_csk(sp); struct inet_sock *inet = inet_sk(sp); - unsigned int dest = inet->daddr; - unsigned int src = inet->rcv_saddr; + __be32 dest = inet->daddr; + __be32 src = inet->rcv_saddr; __u16 destp = ntohs(inet->dport); __u16 srcp = ntohs(inet->sport); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 4a0ee901a888..383cb38461c5 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -493,7 +493,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, struct request_sock **prev) { struct tcphdr *th = skb->h.th; - u32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK); + __be32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK); int paws_reject = 0; struct tcp_options_received tmp_opt; struct sock *child; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index fb09ade5897b..3355c276b611 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -297,7 +297,7 @@ static void tcp_retransmit_timer(struct sock *sk) if (net_ratelimit()) { struct inet_sock *inet = inet_sk(sk); printk(KERN_DEBUG "TCP: Treason uncloaked! Peer %u.%u.%u.%u:%u/%u shrinks window %u:%u. Repaired.\n", - NIPQUAD(inet->daddr), htons(inet->dport), + NIPQUAD(inet->daddr), ntohs(inet->dport), inet->num, tp->snd_una, tp->snd_nxt); } #endif -- cgit v1.2.3 From b09b845ca6724c3bbdc00c0cb2313258c7189ca9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 20:52:19 -0800 Subject: [RANDOM]: Annotate random.h IP helpers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- drivers/char/random.c | 48 ++++++++++++++++++++++++------------------------ include/linux/random.h | 20 ++++++++++---------- 2 files changed, 34 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/random.c b/drivers/char/random.c index eb6b13f4211a..d40df30c2b10 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1466,8 +1466,8 @@ static __init int seqgen_init(void) late_initcall(seqgen_init); #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -__u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, - __u16 sport, __u16 dport) +__u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport) { struct timeval tv; __u32 seq; @@ -1479,10 +1479,10 @@ __u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, */ memcpy(hash, saddr, 16); - hash[4]=(sport << 16) + dport; + hash[4]=((__force u16)sport << 16) + (__force u16)dport; memcpy(&hash[5],keyptr->secret,sizeof(__u32) * 7); - seq = twothirdsMD4Transform(daddr, hash) & HASH_MASK; + seq = twothirdsMD4Transform((const __u32 *)daddr, hash) & HASH_MASK; seq += keyptr->count; do_gettimeofday(&tv); @@ -1496,7 +1496,7 @@ EXPORT_SYMBOL(secure_tcpv6_sequence_number); /* The code below is shamelessly stolen from secure_tcp_sequence_number(). * All blames to Andrey V. Savochkin . */ -__u32 secure_ip_id(__u32 daddr) +__u32 secure_ip_id(__be32 daddr) { struct keydata *keyptr; __u32 hash[4]; @@ -1508,7 +1508,7 @@ __u32 secure_ip_id(__u32 daddr) * The dest ip address is placed in the starting vector, * which is then hashed with random data. */ - hash[0] = daddr; + hash[0] = (__force __u32)daddr; hash[1] = keyptr->secret[9]; hash[2] = keyptr->secret[10]; hash[3] = keyptr->secret[11]; @@ -1518,8 +1518,8 @@ __u32 secure_ip_id(__u32 daddr) #ifdef CONFIG_INET -__u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport) +__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) { struct timeval tv; __u32 seq; @@ -1532,9 +1532,9 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, * Note that the words are placed into the starting vector, which is * then mixed with a partial MD4 over random data. */ - hash[0]=saddr; - hash[1]=daddr; - hash[2]=(sport << 16) + dport; + hash[0]=(__force u32)saddr; + hash[1]=(__force u32)daddr; + hash[2]=((__force u16)sport << 16) + (__force u16)dport; hash[3]=keyptr->secret[11]; seq = half_md4_transform(hash, keyptr->secret) & HASH_MASK; @@ -1559,7 +1559,7 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, EXPORT_SYMBOL(secure_tcp_sequence_number); /* Generate secure starting point for ephemeral IPV4 transport port search */ -u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport) +u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) { struct keydata *keyptr = get_keyptr(); u32 hash[4]; @@ -1568,25 +1568,25 @@ u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport) * Pick a unique starting offset for each ephemeral port search * (saddr, daddr, dport) and 48bits of random data. */ - hash[0] = saddr; - hash[1] = daddr; - hash[2] = dport ^ keyptr->secret[10]; + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = (__force u32)dport ^ keyptr->secret[10]; hash[3] = keyptr->secret[11]; return half_md4_transform(hash, keyptr->secret); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, __u16 dport) +u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, __be16 dport) { struct keydata *keyptr = get_keyptr(); u32 hash[12]; memcpy(hash, saddr, 16); - hash[4] = dport; + hash[4] = (__force u32)dport; memcpy(&hash[5],keyptr->secret,sizeof(__u32) * 7); - return twothirdsMD4Transform(daddr, hash); + return twothirdsMD4Transform((const __u32 *)daddr, hash); } #endif @@ -1595,17 +1595,17 @@ u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, __u16 dpo * bit's 32-47 increase every key exchange * 0-31 hash(source, dest) */ -u64 secure_dccp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport) +u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) { struct timeval tv; u64 seq; __u32 hash[4]; struct keydata *keyptr = get_keyptr(); - hash[0] = saddr; - hash[1] = daddr; - hash[2] = (sport << 16) + dport; + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = ((__force u16)sport << 16) + (__force u16)dport; hash[3] = keyptr->secret[11]; seq = half_md4_transform(hash, keyptr->secret); @@ -1641,7 +1641,7 @@ unsigned int get_random_int(void) * drain on it), and uses halfMD4Transform within the second. We * also mix it with jiffies and the PID: */ - return secure_ip_id(current->pid + jiffies); + return secure_ip_id((__force __be32)(current->pid + jiffies)); } /* diff --git a/include/linux/random.h b/include/linux/random.h index 0248b30e306d..01ad71033d65 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -51,16 +51,16 @@ extern void add_interrupt_randomness(int irq); extern void get_random_bytes(void *buf, int nbytes); void generate_random_uuid(unsigned char uuid_out[16]); -extern __u32 secure_ip_id(__u32 daddr); -extern u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport); -extern u32 secure_ipv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, - __u16 dport); -extern __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport); -extern __u32 secure_tcpv6_sequence_number(__u32 *saddr, __u32 *daddr, - __u16 sport, __u16 dport); -extern u64 secure_dccp_sequence_number(__u32 saddr, __u32 daddr, - __u16 sport, __u16 dport); +extern __u32 secure_ip_id(__be32 daddr); +extern u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport); +extern u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, + __be16 dport); +extern __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); +extern __u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport); +extern u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); #ifndef MODULE extern struct file_operations random_fops, urandom_fops; -- cgit v1.2.3 From 42d224aa170a4f7446cea6c972d9302d524545e5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:08 -0800 Subject: [NETFILTER]: More trivial annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack_tftp.h | 2 +- net/ipv4/netfilter/ip_conntrack_amanda.c | 2 +- net/ipv4/netfilter/ip_conntrack_helper_h323.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tftp.h b/include/linux/netfilter_ipv4/ip_conntrack_tftp.h index cde9729aa173..a404fc0abf0e 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_tftp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_tftp.h @@ -4,7 +4,7 @@ #define TFTP_PORT 69 struct tftphdr { - u_int16_t opcode; + __be16 opcode; }; #define TFTP_OPCODE_READ 1 diff --git a/net/ipv4/netfilter/ip_conntrack_amanda.c b/net/ipv4/netfilter/ip_conntrack_amanda.c index 6c7383a8e42b..7fca246615d5 100644 --- a/net/ipv4/netfilter/ip_conntrack_amanda.c +++ b/net/ipv4/netfilter/ip_conntrack_amanda.c @@ -180,7 +180,7 @@ static struct ip_conntrack_helper amanda_helper = { .help = help, .name = "amanda", - .tuple = { .src = { .u = { __constant_htons(10080) } }, + .tuple = { .src = { .u = { .udp = {.port = __constant_htons(10080) } } }, .dst = { .protonum = IPPROTO_UDP }, }, .mask = { .src = { .u = { 0xFFFF } }, diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323.c b/net/ipv4/netfilter/ip_conntrack_helper_h323.c index 6cb9070cd0bc..a06b340ea414 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323.c @@ -1153,7 +1153,7 @@ static struct ip_conntrack_helper ip_conntrack_helper_q931 = { .me = THIS_MODULE, .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4 /* T.120 and H.245 */ , .timeout = 240, - .tuple = {.src = {.u = {__constant_htons(Q931_PORT)}}, + .tuple = {.src = {.u = {.tcp = {.port = __constant_htons(Q931_PORT)}}}, .dst = {.protonum = IPPROTO_TCP}}, .mask = {.src = {.u = {0xFFFF}}, .dst = {.protonum = 0xFF}}, @@ -1746,7 +1746,7 @@ static struct ip_conntrack_helper ip_conntrack_helper_ras = { .me = THIS_MODULE, .max_expected = 32, .timeout = 240, - .tuple = {.src = {.u = {__constant_htons(RAS_PORT)}}, + .tuple = {.src = {.u = {.tcp = {.port = __constant_htons(RAS_PORT)}}}, .dst = {.protonum = IPPROTO_UDP}}, .mask = {.src = {.u = {0xFFFE}}, .dst = {.protonum = 0xFF}}, -- cgit v1.2.3 From 30d492da738a8d5f4ec884b3e1a13eef97714994 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:29 -0800 Subject: [ATM]: Annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/atmarp.h | 2 +- include/linux/atmbr2684.h | 4 ++-- include/linux/atmmpc.h | 16 ++++++++-------- include/net/atmclip.h | 2 +- net/atm/br2684.c | 2 +- net/atm/clip.c | 10 +++++----- net/atm/lec.c | 6 +++--- net/atm/lec.h | 6 +++--- net/atm/mpc.c | 26 +++++++++++++------------- net/atm/mpc.h | 6 +++--- net/atm/mpoa_caches.c | 14 +++++++------- net/atm/mpoa_caches.h | 16 ++++++++-------- net/atm/mpoa_proc.c | 6 +++--- 13 files changed, 58 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/include/linux/atmarp.h b/include/linux/atmarp.h index 24f82338f59a..ee108f9e9cb7 100644 --- a/include/linux/atmarp.h +++ b/include/linux/atmarp.h @@ -37,7 +37,7 @@ enum atmarp_ctrl_type { struct atmarp_ctrl { enum atmarp_ctrl_type type; /* message type */ int itf_num;/* interface number (if present) */ - uint32_t ip; /* IP address (act_need only) */ + __be32 ip; /* IP address (act_need only) */ }; #endif diff --git a/include/linux/atmbr2684.h b/include/linux/atmbr2684.h index 7981b733f1ef..969fb6c9e1cc 100644 --- a/include/linux/atmbr2684.h +++ b/include/linux/atmbr2684.h @@ -86,8 +86,8 @@ struct atm_backend_br2684 { * efficient per-if in/out filters, this support will be removed */ struct br2684_filter { - __u32 prefix; /* network byte order */ - __u32 netmask; /* 0 = disable filter */ + __be32 prefix; /* network byte order */ + __be32 netmask; /* 0 = disable filter */ }; struct br2684_filter_set { diff --git a/include/linux/atmmpc.h b/include/linux/atmmpc.h index 5fbfa68136d3..ea1650425a12 100644 --- a/include/linux/atmmpc.h +++ b/include/linux/atmmpc.h @@ -13,7 +13,7 @@ struct atmmpc_ioc { int dev_num; - uint32_t ipaddr; /* the IP address of the shortcut */ + __be32 ipaddr; /* the IP address of the shortcut */ int type; /* ingress or egress */ }; @@ -21,8 +21,8 @@ typedef struct in_ctrl_info { uint8_t Last_NHRP_CIE_code; uint8_t Last_Q2931_cause_value; uint8_t eg_MPC_ATM_addr[ATM_ESA_LEN]; - uint32_t tag; - uint32_t in_dst_ip; /* IP address this ingress MPC sends packets to */ + __be32 tag; + __be32 in_dst_ip; /* IP address this ingress MPC sends packets to */ uint16_t holding_time; uint32_t request_id; } in_ctrl_info; @@ -30,10 +30,10 @@ typedef struct in_ctrl_info { typedef struct eg_ctrl_info { uint8_t DLL_header[256]; uint8_t DH_length; - uint32_t cache_id; - uint32_t tag; - uint32_t mps_ip; - uint32_t eg_dst_ip; /* IP address to which ingress MPC sends packets */ + __be32 cache_id; + __be32 tag; + __be32 mps_ip; + __be32 eg_dst_ip; /* IP address to which ingress MPC sends packets */ uint8_t in_MPC_data_ATM_addr[ATM_ESA_LEN]; uint16_t holding_time; } eg_ctrl_info; @@ -49,7 +49,7 @@ struct mpc_parameters { struct k_message { uint16_t type; - uint32_t ip_mask; + __be32 ip_mask; uint8_t MPS_ctrl[ATM_ESA_LEN]; union { in_ctrl_info in_info; diff --git a/include/net/atmclip.h b/include/net/atmclip.h index 90fcc98e676f..b5a51a7bb364 100644 --- a/include/net/atmclip.h +++ b/include/net/atmclip.h @@ -36,7 +36,7 @@ struct clip_vcc { struct atmarp_entry { - u32 ip; /* IP address */ + __be32 ip; /* IP address */ struct clip_vcc *vccs; /* active VCCs; NULL if resolution is pending */ unsigned long expires; /* entry expiration time */ diff --git a/net/atm/br2684.c b/net/atm/br2684.c index d00cca97eb33..b04162f10d85 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -372,7 +372,7 @@ static int br2684_setfilt(struct atm_vcc *atmvcc, void __user *arg) /* Returns 1 if packet should be dropped */ static inline int -packet_fails_filter(u16 type, struct br2684_vcc *brvcc, struct sk_buff *skb) +packet_fails_filter(__be16 type, struct br2684_vcc *brvcc, struct sk_buff *skb) { if (brvcc->filter.netmask == 0) return 0; /* no filter in place */ diff --git a/net/atm/clip.c b/net/atm/clip.c index 7af2c411da82..1c416934b7c1 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -54,7 +54,7 @@ static struct atm_vcc *atmarpd; static struct neigh_table clip_tbl; static struct timer_list idle_timer; -static int to_atmarpd(enum atmarp_ctrl_type type, int itf, unsigned long ip) +static int to_atmarpd(enum atmarp_ctrl_type type, int itf, __be32 ip) { struct sock *sk; struct atmarp_ctrl *ctrl; @@ -220,7 +220,7 @@ static void clip_push(struct atm_vcc *vcc, struct sk_buff *skb) || memcmp(skb->data, llc_oui, sizeof (llc_oui))) skb->protocol = htons(ETH_P_IP); else { - skb->protocol = ((u16 *) skb->data)[3]; + skb->protocol = ((__be16 *) skb->data)[3]; skb_pull(skb, RFC1483LLC_LEN); if (skb->protocol == htons(ETH_P_ARP)) { PRIV(skb->dev)->stats.rx_packets++; @@ -430,7 +430,7 @@ static int clip_start_xmit(struct sk_buff *skb, struct net_device *dev) here = skb_push(skb, RFC1483LLC_LEN); memcpy(here, llc_oui, sizeof(llc_oui)); - ((u16 *) here)[3] = skb->protocol; + ((__be16 *) here)[3] = skb->protocol; } atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc); ATM_SKB(skb)->atm_options = vcc->atm_options; @@ -509,7 +509,7 @@ static int clip_mkip(struct atm_vcc *vcc, int timeout) return 0; } -static int clip_setentry(struct atm_vcc *vcc, u32 ip) +static int clip_setentry(struct atm_vcc *vcc, __be32 ip) { struct neighbour *neigh; struct atmarp_entry *entry; @@ -752,7 +752,7 @@ static int clip_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) err = clip_mkip(vcc, arg); break; case ATMARP_SETENTRY: - err = clip_setentry(vcc, arg); + err = clip_setentry(vcc, (__force __be32)arg); break; case ATMARP_ENCAP: err = clip_encap(vcc, arg); diff --git a/net/atm/lec.c b/net/atm/lec.c index 66c57c1091a8..9f5f931743bd 100644 --- a/net/atm/lec.c +++ b/net/atm/lec.c @@ -204,9 +204,9 @@ static unsigned char *get_tr_dst(unsigned char *packet, unsigned char *rdesc) memset(rdesc, 0, ETH_ALEN); /* offset 4 comes from LAN destination field in LE control frames */ if (trh->rcf & htons((uint16_t) TR_RCF_DIR_BIT)) - memcpy(&rdesc[4], &trh->rseg[num_rdsc - 2], sizeof(uint16_t)); + memcpy(&rdesc[4], &trh->rseg[num_rdsc - 2], sizeof(__be16)); else { - memcpy(&rdesc[4], &trh->rseg[1], sizeof(uint16_t)); + memcpy(&rdesc[4], &trh->rseg[1], sizeof(__be16)); rdesc[5] = ((ntohs(trh->rseg[0]) & 0x000f) | (rdesc[5] & 0xf0)); } @@ -775,7 +775,7 @@ static void lec_push(struct atm_vcc *vcc, struct sk_buff *skb) unsigned char *src, *dst; atm_return(vcc, skb->truesize); - if (*(uint16_t *) skb->data == htons(priv->lecid) || + if (*(__be16 *) skb->data == htons(priv->lecid) || !priv->lecd || !(dev->flags & IFF_UP)) { /* * Probably looping back, or if lecd is missing, diff --git a/net/atm/lec.h b/net/atm/lec.h index 877f50939696..24cc95f86741 100644 --- a/net/atm/lec.h +++ b/net/atm/lec.h @@ -14,14 +14,14 @@ #define LEC_HEADER_LEN 16 struct lecdatahdr_8023 { - unsigned short le_header; + __be16 le_header; unsigned char h_dest[ETH_ALEN]; unsigned char h_source[ETH_ALEN]; - unsigned short h_type; + __be16 h_type; }; struct lecdatahdr_8025 { - unsigned short le_header; + __be16 le_header; unsigned char ac_pad; unsigned char fc; unsigned char h_dest[ETH_ALEN]; diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 0d2b994af511..f15f5d847860 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -152,7 +152,7 @@ static struct mpoa_client *find_mpc_by_lec(struct net_device *dev) /* * Overwrites the old entry or makes a new one. */ -struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos) +struct atm_mpoa_qos *atm_mpoa_add_qos(__be32 dst_ip, struct atm_qos *qos) { struct atm_mpoa_qos *entry; @@ -177,7 +177,7 @@ struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos) return entry; } -struct atm_mpoa_qos *atm_mpoa_search_qos(uint32_t dst_ip) +struct atm_mpoa_qos *atm_mpoa_search_qos(__be32 dst_ip) { struct atm_mpoa_qos *qos; @@ -460,11 +460,11 @@ static int send_via_shortcut(struct sk_buff *skb, struct mpoa_client *mpc) in_cache_entry *entry; struct iphdr *iph; char *buff; - uint32_t ipaddr = 0; + __be32 ipaddr = 0; static struct { struct llc_snap_hdr hdr; - uint32_t tag; + __be32 tag; } tagged_llc_snap_hdr = { {0xaa, 0xaa, 0x03, {0x00, 0x00, 0x00}, {0x88, 0x4c}}, 0 @@ -559,7 +559,7 @@ static int atm_mpoa_vcc_attach(struct atm_vcc *vcc, void __user *arg) struct mpoa_client *mpc; struct atmmpc_ioc ioc_data; in_cache_entry *in_entry; - uint32_t ipaddr; + __be32 ipaddr; bytes_left = copy_from_user(&ioc_data, arg, sizeof(struct atmmpc_ioc)); if (bytes_left != 0) { @@ -638,7 +638,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) struct sk_buff *new_skb; eg_cache_entry *eg; struct mpoa_client *mpc; - uint32_t tag; + __be32 tag; char *tmp; ddprintk("mpoa: (%s) mpc_push:\n", dev->name); @@ -683,7 +683,7 @@ static void mpc_push(struct atm_vcc *vcc, struct sk_buff *skb) } tmp = skb->data + sizeof(struct llc_snap_hdr); - tag = *(uint32_t *)tmp; + tag = *(__be32 *)tmp; eg = mpc->eg_ops->get_by_tag(tag, mpc); if (eg == NULL) { @@ -1029,7 +1029,7 @@ static int mpoa_event_listener(struct notifier_block *mpoa_notifier, unsigned lo static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; in_cache_entry *entry; entry = mpc->in_ops->get(dst_ip, mpc); @@ -1066,7 +1066,7 @@ static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) */ static void check_qos_and_open_shortcut(struct k_message *msg, struct mpoa_client *client, in_cache_entry *entry) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; struct atm_mpoa_qos *qos = atm_mpoa_search_qos(dst_ip); eg_cache_entry *eg_entry = client->eg_ops->get_by_src_ip(dst_ip, client); @@ -1102,7 +1102,7 @@ static void check_qos_and_open_shortcut(struct k_message *msg, struct mpoa_clien static void MPOA_res_reply_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; + __be32 dst_ip = msg->content.in_info.in_dst_ip; in_cache_entry *entry = mpc->in_ops->get(dst_ip, mpc); dprintk("mpoa: (%s) MPOA_res_reply_rcvd: ip %u.%u.%u.%u\n", mpc->dev->name, NIPQUAD(dst_ip)); @@ -1148,8 +1148,8 @@ static void MPOA_res_reply_rcvd(struct k_message *msg, struct mpoa_client *mpc) static void ingress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t dst_ip = msg->content.in_info.in_dst_ip; - uint32_t mask = msg->ip_mask; + __be32 dst_ip = msg->content.in_info.in_dst_ip; + __be32 mask = msg->ip_mask; in_cache_entry *entry = mpc->in_ops->get_with_mask(dst_ip, mpc, mask); if(entry == NULL){ @@ -1173,7 +1173,7 @@ static void ingress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) static void egress_purge_rcvd(struct k_message *msg, struct mpoa_client *mpc) { - uint32_t cache_id = msg->content.eg_info.cache_id; + __be32 cache_id = msg->content.eg_info.cache_id; eg_cache_entry *entry = mpc->eg_ops->get_by_cache_id(cache_id, mpc); if (entry == NULL) { diff --git a/net/atm/mpc.h b/net/atm/mpc.h index 3c7981a229e8..51f460d005c3 100644 --- a/net/atm/mpc.h +++ b/net/atm/mpc.h @@ -36,14 +36,14 @@ struct mpoa_client { struct atm_mpoa_qos { struct atm_mpoa_qos *next; - uint32_t ipaddr; + __be32 ipaddr; struct atm_qos qos; }; /* MPOA QoS operations */ -struct atm_mpoa_qos *atm_mpoa_add_qos(uint32_t dst_ip, struct atm_qos *qos); -struct atm_mpoa_qos *atm_mpoa_search_qos(uint32_t dst_ip); +struct atm_mpoa_qos *atm_mpoa_add_qos(__be32 dst_ip, struct atm_qos *qos); +struct atm_mpoa_qos *atm_mpoa_search_qos(__be32 dst_ip); int atm_mpoa_delete_qos(struct atm_mpoa_qos *qos); /* Display QoS entries. This is for the procfs */ diff --git a/net/atm/mpoa_caches.c b/net/atm/mpoa_caches.c index fbf13cdcf46e..f3b99b38c91b 100644 --- a/net/atm/mpoa_caches.c +++ b/net/atm/mpoa_caches.c @@ -22,7 +22,7 @@ #define ddprintk(format,args...) #endif -static in_cache_entry *in_cache_get(uint32_t dst_ip, +static in_cache_entry *in_cache_get(__be32 dst_ip, struct mpoa_client *client) { in_cache_entry *entry; @@ -42,9 +42,9 @@ static in_cache_entry *in_cache_get(uint32_t dst_ip, return NULL; } -static in_cache_entry *in_cache_get_with_mask(uint32_t dst_ip, +static in_cache_entry *in_cache_get_with_mask(__be32 dst_ip, struct mpoa_client *client, - uint32_t mask) + __be32 mask) { in_cache_entry *entry; @@ -84,7 +84,7 @@ static in_cache_entry *in_cache_get_by_vcc(struct atm_vcc *vcc, return NULL; } -static in_cache_entry *in_cache_add_entry(uint32_t dst_ip, +static in_cache_entry *in_cache_add_entry(__be32 dst_ip, struct mpoa_client *client) { in_cache_entry* entry = kmalloc(sizeof(in_cache_entry), GFP_KERNEL); @@ -319,7 +319,7 @@ static void in_destroy_cache(struct mpoa_client *mpc) return; } -static eg_cache_entry *eg_cache_get_by_cache_id(uint32_t cache_id, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_cache_id(__be32 cache_id, struct mpoa_client *mpc) { eg_cache_entry *entry; @@ -339,7 +339,7 @@ static eg_cache_entry *eg_cache_get_by_cache_id(uint32_t cache_id, struct mpoa_c } /* This can be called from any context since it saves CPU flags */ -static eg_cache_entry *eg_cache_get_by_tag(uint32_t tag, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_tag(__be32 tag, struct mpoa_client *mpc) { unsigned long flags; eg_cache_entry *entry; @@ -380,7 +380,7 @@ static eg_cache_entry *eg_cache_get_by_vcc(struct atm_vcc *vcc, struct mpoa_clie return NULL; } -static eg_cache_entry *eg_cache_get_by_src_ip(uint32_t ipaddr, struct mpoa_client *mpc) +static eg_cache_entry *eg_cache_get_by_src_ip(__be32 ipaddr, struct mpoa_client *mpc) { eg_cache_entry *entry; diff --git a/net/atm/mpoa_caches.h b/net/atm/mpoa_caches.h index 6c9886a03d0b..84de977def2e 100644 --- a/net/atm/mpoa_caches.h +++ b/net/atm/mpoa_caches.h @@ -29,12 +29,12 @@ typedef struct in_cache_entry { } in_cache_entry; struct in_cache_ops{ - in_cache_entry *(*add_entry)(uint32_t dst_ip, + in_cache_entry *(*add_entry)(__be32 dst_ip, struct mpoa_client *client); - in_cache_entry *(*get)(uint32_t dst_ip, struct mpoa_client *client); - in_cache_entry *(*get_with_mask)(uint32_t dst_ip, + in_cache_entry *(*get)(__be32 dst_ip, struct mpoa_client *client); + in_cache_entry *(*get_with_mask)(__be32 dst_ip, struct mpoa_client *client, - uint32_t mask); + __be32 mask); in_cache_entry *(*get_by_vcc)(struct atm_vcc *vcc, struct mpoa_client *client); void (*put)(in_cache_entry *entry); @@ -56,17 +56,17 @@ typedef struct eg_cache_entry{ struct atm_vcc *shortcut; uint32_t packets_rcvd; uint16_t entry_state; - uint32_t latest_ip_addr; /* The src IP address of the last packet */ + __be32 latest_ip_addr; /* The src IP address of the last packet */ struct eg_ctrl_info ctrl_info; atomic_t use; } eg_cache_entry; struct eg_cache_ops{ eg_cache_entry *(*add_entry)(struct k_message *msg, struct mpoa_client *client); - eg_cache_entry *(*get_by_cache_id)(uint32_t cache_id, struct mpoa_client *client); - eg_cache_entry *(*get_by_tag)(uint32_t cache_id, struct mpoa_client *client); + eg_cache_entry *(*get_by_cache_id)(__be32 cache_id, struct mpoa_client *client); + eg_cache_entry *(*get_by_tag)(__be32 cache_id, struct mpoa_client *client); eg_cache_entry *(*get_by_vcc)(struct atm_vcc *vcc, struct mpoa_client *client); - eg_cache_entry *(*get_by_src_ip)(uint32_t ipaddr, struct mpoa_client *client); + eg_cache_entry *(*get_by_src_ip)(__be32 ipaddr, struct mpoa_client *client); void (*put)(eg_cache_entry *entry); void (*remove_entry)(eg_cache_entry *entry, struct mpoa_client *client); void (*update)(eg_cache_entry *entry, uint16_t holding_time); diff --git a/net/atm/mpoa_proc.c b/net/atm/mpoa_proc.c index d37b8911b3ab..3844c85d602f 100644 --- a/net/atm/mpoa_proc.c +++ b/net/atm/mpoa_proc.c @@ -231,14 +231,14 @@ static int parse_qos(const char *buff) */ unsigned char ip[4]; int tx_pcr, tx_sdu, rx_pcr, rx_sdu; - uint32_t ipaddr; + __be32 ipaddr; struct atm_qos qos; memset(&qos, 0, sizeof(struct atm_qos)); if (sscanf(buff, "del %hhu.%hhu.%hhu.%hhu", ip, ip+1, ip+2, ip+3) == 4) { - ipaddr = *(uint32_t *)ip; + ipaddr = *(__be32 *)ip; return atm_mpoa_delete_qos(atm_mpoa_search_qos(ipaddr)); } @@ -250,7 +250,7 @@ static int parse_qos(const char *buff) ip, ip+1, ip+2, ip+3, &tx_pcr, &tx_sdu, &rx_pcr, &rx_sdu) != 8) return 0; - ipaddr = *(uint32_t *)ip; + ipaddr = *(__be32 *)ip; qos.txtp.traffic_class = ATM_CBR; qos.txtp.max_pcr = tx_pcr; qos.txtp.max_sdu = tx_sdu; -- cgit v1.2.3 From 47c183fa5ea7feebc356da8ccbd9105a41f8e534 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:11:51 -0800 Subject: [BRIDGE]: Annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_bridge.h | 2 +- include/linux/netfilter_bridge/ebt_802_3.h | 10 +++++----- include/linux/netfilter_bridge/ebt_among.h | 2 +- include/linux/netfilter_bridge/ebt_arp.h | 14 +++++++------- include/linux/netfilter_bridge/ebt_ip.h | 8 ++++---- include/linux/netfilter_bridge/ebt_vlan.h | 2 +- include/linux/netfilter_bridge/ebtables.h | 2 +- net/bridge/br_netfilter.c | 2 +- net/bridge/netfilter/ebt_802_3.c | 2 +- net/bridge/netfilter/ebt_among.c | 22 +++++++++++----------- net/bridge/netfilter/ebt_arp.c | 6 +++--- net/bridge/netfilter/ebt_ip.c | 4 ++-- net/bridge/netfilter/ebt_log.c | 6 +++--- net/bridge/netfilter/ebt_vlan.c | 2 +- 14 files changed, 42 insertions(+), 42 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index 9a4dd11af86e..6c4613f8ad75 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -64,7 +64,7 @@ static inline int nf_bridge_pad(const struct sk_buff *skb) struct bridge_skb_cb { union { - __u32 ipv4; + __be32 ipv4; } daddr; }; diff --git a/include/linux/netfilter_bridge/ebt_802_3.h b/include/linux/netfilter_bridge/ebt_802_3.h index b9f712c14a0a..07f044ff1a6b 100644 --- a/include/linux/netfilter_bridge/ebt_802_3.h +++ b/include/linux/netfilter_bridge/ebt_802_3.h @@ -28,21 +28,21 @@ struct hdr_ui { uint8_t ssap; uint8_t ctrl; uint8_t orig[3]; - uint16_t type; + __be16 type; }; struct hdr_ni { uint8_t dsap; uint8_t ssap; - uint16_t ctrl; + __be16 ctrl; uint8_t orig[3]; - uint16_t type; + __be16 type; }; struct ebt_802_3_hdr { uint8_t daddr[6]; uint8_t saddr[6]; - uint16_t len; + __be16 len; union { struct hdr_ui ui; struct hdr_ni ni; @@ -61,7 +61,7 @@ static inline struct ebt_802_3_hdr *ebt_802_3_hdr(const struct sk_buff *skb) struct ebt_802_3_info { uint8_t sap; - uint16_t type; + __be16 type; uint8_t bitmask; uint8_t invflags; }; diff --git a/include/linux/netfilter_bridge/ebt_among.h b/include/linux/netfilter_bridge/ebt_among.h index 307c1fed8511..7654069233ca 100644 --- a/include/linux/netfilter_bridge/ebt_among.h +++ b/include/linux/netfilter_bridge/ebt_among.h @@ -32,7 +32,7 @@ struct ebt_mac_wormhash_tuple { uint32_t cmp[2]; - uint32_t ip; + __be32 ip; }; struct ebt_mac_wormhash diff --git a/include/linux/netfilter_bridge/ebt_arp.h b/include/linux/netfilter_bridge/ebt_arp.h index 537ec6b487a2..97e4dbde1f89 100644 --- a/include/linux/netfilter_bridge/ebt_arp.h +++ b/include/linux/netfilter_bridge/ebt_arp.h @@ -14,13 +14,13 @@ struct ebt_arp_info { - uint16_t htype; - uint16_t ptype; - uint16_t opcode; - uint32_t saddr; - uint32_t smsk; - uint32_t daddr; - uint32_t dmsk; + __be16 htype; + __be16 ptype; + __be16 opcode; + __be32 saddr; + __be32 smsk; + __be32 daddr; + __be32 dmsk; unsigned char smaddr[ETH_ALEN]; unsigned char smmsk[ETH_ALEN]; unsigned char dmaddr[ETH_ALEN]; diff --git a/include/linux/netfilter_bridge/ebt_ip.h b/include/linux/netfilter_bridge/ebt_ip.h index 7247385cdcb1..d6847475bf2e 100644 --- a/include/linux/netfilter_bridge/ebt_ip.h +++ b/include/linux/netfilter_bridge/ebt_ip.h @@ -28,10 +28,10 @@ /* the same values are used for the invflags */ struct ebt_ip_info { - uint32_t saddr; - uint32_t daddr; - uint32_t smsk; - uint32_t dmsk; + __be32 saddr; + __be32 daddr; + __be32 smsk; + __be32 dmsk; uint8_t tos; uint8_t protocol; uint8_t bitmask; diff --git a/include/linux/netfilter_bridge/ebt_vlan.h b/include/linux/netfilter_bridge/ebt_vlan.h index cb1fcc41565f..1d98be4031e7 100644 --- a/include/linux/netfilter_bridge/ebt_vlan.h +++ b/include/linux/netfilter_bridge/ebt_vlan.h @@ -10,7 +10,7 @@ struct ebt_vlan_info { uint16_t id; /* VLAN ID {1-4095} */ uint8_t prio; /* VLAN User Priority {0-7} */ - uint16_t encap; /* VLAN Encapsulated frame code {0-65535} */ + __be16 encap; /* VLAN Encapsulated frame code {0-65535} */ uint8_t bitmask; /* Args bitmask bit 1=1 - ID arg, bit 2=1 User-Priority arg, bit 3=1 encap*/ uint8_t invflags; /* Inverse bitmask bit 1=1 - inversed ID arg, diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index b1a7cc90877b..e6ea70de24d5 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -141,7 +141,7 @@ struct ebt_entry { /* this needs to be the first field */ unsigned int bitmask; unsigned int invflags; - uint16_t ethproto; + __be16 ethproto; /* the physical in-dev */ char in[IFNAMSIZ]; /* the logical in-dev */ diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index ac181be13d83..2a5d31b1a196 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -381,7 +381,7 @@ static int check_hbh_len(struct sk_buff *skb) case IPV6_TLV_JUMBO: if (skb->nh.raw[off + 1] != 4 || (off & 3) != 2) goto bad; - pkt_len = ntohl(*(u32 *) (skb->nh.raw + off + 2)); + pkt_len = ntohl(*(__be32 *) (skb->nh.raw + off + 2)); if (pkt_len <= IPV6_MAXPLEN || skb->nh.ipv6h->payload_len) goto bad; diff --git a/net/bridge/netfilter/ebt_802_3.c b/net/bridge/netfilter/ebt_802_3.c index d42f63f5e9f8..9abbc09ccdc3 100644 --- a/net/bridge/netfilter/ebt_802_3.c +++ b/net/bridge/netfilter/ebt_802_3.c @@ -17,7 +17,7 @@ static int ebt_filter_802_3(const struct sk_buff *skb, const struct net_device * { struct ebt_802_3_info *info = (struct ebt_802_3_info *)data; struct ebt_802_3_hdr *hdr = ebt_802_3_hdr(skb); - uint16_t type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; + __be16 type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; if (info->bitmask & EBT_802_3_SAP) { if (FWINV(info->sap != hdr->llc.ui.ssap, EBT_802_3_SAP)) diff --git a/net/bridge/netfilter/ebt_among.c b/net/bridge/netfilter/ebt_among.c index a614485828af..ce97c4285f9a 100644 --- a/net/bridge/netfilter/ebt_among.c +++ b/net/bridge/netfilter/ebt_among.c @@ -15,7 +15,7 @@ #include static int ebt_mac_wormhash_contains(const struct ebt_mac_wormhash *wh, - const char *mac, uint32_t ip) + const char *mac, __be32 ip) { /* You may be puzzled as to how this code works. * Some tricks were used, refer to @@ -70,7 +70,7 @@ static int ebt_mac_wormhash_check_integrity(const struct ebt_mac_wormhash return 0; } -static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) +static int get_ip_dst(const struct sk_buff *skb, __be32 *addr) { if (eth_hdr(skb)->h_proto == htons(ETH_P_IP)) { struct iphdr _iph, *ih; @@ -81,16 +81,16 @@ static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) *addr = ih->daddr; } else if (eth_hdr(skb)->h_proto == htons(ETH_P_ARP)) { struct arphdr _arph, *ah; - uint32_t buf, *bp; + __be32 buf, *bp; ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL || - ah->ar_pln != sizeof(uint32_t) || + ah->ar_pln != sizeof(__be32) || ah->ar_hln != ETH_ALEN) return -1; bp = skb_header_pointer(skb, sizeof(struct arphdr) + - 2 * ETH_ALEN + sizeof(uint32_t), - sizeof(uint32_t), &buf); + 2 * ETH_ALEN + sizeof(__be32), + sizeof(__be32), &buf); if (bp == NULL) return -1; *addr = *bp; @@ -98,7 +98,7 @@ static int get_ip_dst(const struct sk_buff *skb, uint32_t *addr) return 0; } -static int get_ip_src(const struct sk_buff *skb, uint32_t *addr) +static int get_ip_src(const struct sk_buff *skb, __be32 *addr) { if (eth_hdr(skb)->h_proto == htons(ETH_P_IP)) { struct iphdr _iph, *ih; @@ -109,15 +109,15 @@ static int get_ip_src(const struct sk_buff *skb, uint32_t *addr) *addr = ih->saddr; } else if (eth_hdr(skb)->h_proto == htons(ETH_P_ARP)) { struct arphdr _arph, *ah; - uint32_t buf, *bp; + __be32 buf, *bp; ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL || - ah->ar_pln != sizeof(uint32_t) || + ah->ar_pln != sizeof(__be32) || ah->ar_hln != ETH_ALEN) return -1; bp = skb_header_pointer(skb, sizeof(struct arphdr) + - ETH_ALEN, sizeof(uint32_t), &buf); + ETH_ALEN, sizeof(__be32), &buf); if (bp == NULL) return -1; *addr = *bp; @@ -133,7 +133,7 @@ static int ebt_filter_among(const struct sk_buff *skb, struct ebt_among_info *info = (struct ebt_among_info *) data; const char *dmac, *smac; const struct ebt_mac_wormhash *wh_dst, *wh_src; - uint32_t dip = 0, sip = 0; + __be32 dip = 0, sip = 0; wh_dst = ebt_among_wh_dst(info); wh_src = ebt_among_wh_src(info); diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index a6c81d9f73b8..9c599800a900 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -35,10 +35,10 @@ static int ebt_filter_arp(const struct sk_buff *skb, const struct net_device *in return EBT_NOMATCH; if (info->bitmask & (EBT_ARP_SRC_IP | EBT_ARP_DST_IP)) { - uint32_t _addr, *ap; + __be32 _addr, *ap; /* IPv4 addresses are always 4 bytes */ - if (ah->ar_pln != sizeof(uint32_t)) + if (ah->ar_pln != sizeof(__be32)) return EBT_NOMATCH; if (info->bitmask & EBT_ARP_SRC_IP) { ap = skb_header_pointer(skb, sizeof(struct arphdr) + @@ -53,7 +53,7 @@ static int ebt_filter_arp(const struct sk_buff *skb, const struct net_device *in if (info->bitmask & EBT_ARP_DST_IP) { ap = skb_header_pointer(skb, sizeof(struct arphdr) + - 2*ah->ar_hln+sizeof(uint32_t), + 2*ah->ar_hln+sizeof(__be32), sizeof(_addr), &_addr); if (ap == NULL) return EBT_NOMATCH; diff --git a/net/bridge/netfilter/ebt_ip.c b/net/bridge/netfilter/ebt_ip.c index 65b665ce57b5..e4c642448e1b 100644 --- a/net/bridge/netfilter/ebt_ip.c +++ b/net/bridge/netfilter/ebt_ip.c @@ -20,8 +20,8 @@ #include struct tcpudphdr { - uint16_t src; - uint16_t dst; + __be16 src; + __be16 dst; }; static int ebt_filter_ip(const struct sk_buff *skb, const struct net_device *in, diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c index 466ed3440b74..a184f879f253 100644 --- a/net/bridge/netfilter/ebt_log.c +++ b/net/bridge/netfilter/ebt_log.c @@ -38,8 +38,8 @@ static int ebt_log_check(const char *tablename, unsigned int hookmask, struct tcpudphdr { - uint16_t src; - uint16_t dst; + __be16 src; + __be16 dst; }; struct arppayload @@ -130,7 +130,7 @@ ebt_log_packet(unsigned int pf, unsigned int hooknum, * then log the ARP payload */ if (ah->ar_hrd == htons(1) && ah->ar_hln == ETH_ALEN && - ah->ar_pln == sizeof(uint32_t)) { + ah->ar_pln == sizeof(__be32)) { struct arppayload _arpp, *ap; ap = skb_header_pointer(skb, sizeof(_arph), diff --git a/net/bridge/netfilter/ebt_vlan.c b/net/bridge/netfilter/ebt_vlan.c index a2b452862b73..7ee377622964 100644 --- a/net/bridge/netfilter/ebt_vlan.c +++ b/net/bridge/netfilter/ebt_vlan.c @@ -55,7 +55,7 @@ ebt_filter_vlan(const struct sk_buff *skb, unsigned short id; /* VLAN ID, given from frame TCI */ unsigned char prio; /* user_priority, given from frame TCI */ /* VLAN encapsulated Type/Length field, given from orig frame */ - unsigned short encap; + __be16 encap; fp = skb_header_pointer(skb, 0, sizeof(_frame), &_frame); if (fp == NULL) -- cgit v1.2.3 From a64b78a077a71c9b9c0c1b0be699083379783c3d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:12:29 -0800 Subject: [NET]: Annotate net_srandom(). Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/net.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/net.h b/include/linux/net.h index 15c733b816f0..6f0dfeba509a 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -196,7 +196,7 @@ extern struct socket *sockfd_lookup(int fd, int *err); extern int net_ratelimit(void); #define net_random() random32() -#define net_srandom(seed) srandom32(seed) +#define net_srandom(seed) srandom32((__force u32)seed) extern int kernel_sendmsg(struct socket *sock, struct msghdr *msg, struct kvec *vec, size_t num, size_t len); -- cgit v1.2.3 From 2bc357987a6510e61d33f3b20fa989fb2b6a10b8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:14:18 -0800 Subject: [NET]: Introduce types for checksums. New types - for 16bit checksums and "unfolded" 32bit variant. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/types.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/types.h b/include/linux/types.h index 9f11fdd2bd72..745c409ebbb5 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -182,6 +182,8 @@ typedef __u32 __bitwise __be32; typedef __u64 __bitwise __le64; typedef __u64 __bitwise __be64; #endif +typedef __u16 __bitwise __sum16; +typedef __u32 __bitwise __wsum; #ifdef __KERNEL__ typedef unsigned __bitwise__ gfp_t; -- cgit v1.2.3 From 9981a0e36a572e9fcf84bfab915fdc93bed0e3c9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:24:30 -0800 Subject: [NET]: Annotate checksums in on-the-wire packets. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/dccp.h | 2 +- include/linux/icmp.h | 2 +- include/linux/icmpv6.h | 2 +- include/linux/igmp.h | 2 +- include/linux/ip.h | 2 +- include/linux/tcp.h | 2 +- include/linux/udp.h | 2 +- 7 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index d308f1228b61..72cc355d7a03 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -30,7 +30,7 @@ struct dccp_hdr { #else #error "Adjust your defines" #endif - __u16 dccph_checksum; + __sum16 dccph_checksum; #if defined(__LITTLE_ENDIAN_BITFIELD) __u8 dccph_x:1, dccph_type:4, diff --git a/include/linux/icmp.h b/include/linux/icmp.h index 878cfe4e587f..24da4fbc1a2f 100644 --- a/include/linux/icmp.h +++ b/include/linux/icmp.h @@ -68,7 +68,7 @@ struct icmphdr { __u8 type; __u8 code; - __be16 checksum; + __sum16 checksum; union { struct { __be16 id; diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index dc79396aac25..68d3526c3a05 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -7,7 +7,7 @@ struct icmp6hdr { __u8 icmp6_type; __u8 icmp6_code; - __be16 icmp6_cksum; + __sum16 icmp6_cksum; union { diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 21dd56905271..6e7ea2f0a57c 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -30,7 +30,7 @@ struct igmphdr { __u8 type; __u8 code; /* For newer IGMP */ - __be16 csum; + __sum16 csum; __be32 group; }; diff --git a/include/linux/ip.h b/include/linux/ip.h index ecee9bb27d0e..1d36b971a8b5 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -98,7 +98,7 @@ struct iphdr { __be16 frag_off; __u8 ttl; __u8 protocol; - __be16 check; + __sum16 check; __be32 saddr; __be32 daddr; /*The options start here. */ diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 0aecfc955591..dd61b172ac68 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -52,7 +52,7 @@ struct tcphdr { #error "Adjust your defines" #endif __be16 window; - __be16 check; + __sum16 check; __be16 urg_ptr; }; diff --git a/include/linux/udp.h b/include/linux/udp.h index 564f3b050105..7e08c07efe0f 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -23,7 +23,7 @@ struct udphdr { __be16 source; __be16 dest; __be16 len; - __be16 check; + __sum16 check; }; /* UDP socket options */ -- cgit v1.2.3 From 44bb93633f57a55979f3c2589b10fd6a2bfc7c08 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:36:14 -0800 Subject: [NET]: Annotate csum_partial() callers in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/socket.h | 2 +- include/linux/sunrpc/xdr.h | 2 +- net/core/iovec.c | 4 ++-- net/core/skbuff.c | 2 +- net/ipv4/ip_output.c | 8 ++++---- net/unix/af_unix.c | 3 ++- 6 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/socket.h b/include/linux/socket.h index 592b66679823..92cd38efad7f 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -293,7 +293,7 @@ extern int memcpy_fromiovecend(unsigned char *kdata, struct iovec *iov, extern int csum_partial_copy_fromiovecend(unsigned char *kdata, struct iovec *iov, int offset, - unsigned int len, int *csump); + unsigned int len, __wsum *csump); extern int verify_iovec(struct msghdr *m, struct iovec *iov, char *address, int mode); extern int memcpy_toiovec(struct iovec *v, unsigned char *kdata, int len); diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index ac69e5511606..9a527c364394 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -151,7 +151,7 @@ typedef struct { struct sk_buff *skb; unsigned int offset; size_t count; - unsigned int csum; + __wsum csum; } skb_reader_t; typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); diff --git a/net/core/iovec.c b/net/core/iovec.c index 65e4b56fbc77..04b249c40b5b 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -158,9 +158,9 @@ int memcpy_fromiovecend(unsigned char *kdata, struct iovec *iov, int offset, * call to this function will be unaligned also. */ int csum_partial_copy_fromiovecend(unsigned char *kdata, struct iovec *iov, - int offset, unsigned int len, int *csump) + int offset, unsigned int len, __wsum *csump) { - int csum = *csump; + __wsum csum = *csump; int partial_cnt = 0, err = 0; /* Skip over the finished iovecs */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index dfa02cc8d687..c0e3427057fc 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1265,7 +1265,7 @@ unsigned int skb_checksum(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 5f3e35c03637..f9194f7e39d3 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -682,7 +682,7 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk if (memcpy_fromiovecend(to, iov, offset, len) < 0) return -EFAULT; } else { - unsigned int csum = 0; + __wsum csum = 0; if (csum_partial_copy_fromiovecend(to, iov, offset, len, &csum) < 0) return -EFAULT; skb->csum = csum_block_add(skb->csum, csum, odd); @@ -690,11 +690,11 @@ ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk return 0; } -static inline unsigned int +static inline __wsum csum_page(struct page *page, int offset, int copy) { char *kaddr; - unsigned int csum; + __wsum csum; kaddr = kmap(page); csum = csum_partial(kaddr + offset, copy, 0); kunmap(page); @@ -1166,7 +1166,7 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, } if (skb->ip_summed == CHECKSUM_NONE) { - unsigned int csum; + __wsum csum; csum = csum_page(page, offset, len); skb->csum = csum_block_add(skb->csum, csum, skb->len); } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index b43a27828df5..2f208c7f4d43 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -151,8 +151,9 @@ static inline void unix_set_secdata(struct scm_cookie *scm, struct sk_buff *skb) * each socket state is protected by separate rwlock. */ -static inline unsigned unix_hash_fold(unsigned hash) +static inline unsigned unix_hash_fold(__wsum n) { + unsigned hash = (__force unsigned)n; hash ^= hash>>16; hash ^= hash>>8; return hash&(UNIX_HASH_SIZE-1); -- cgit v1.2.3 From 5084205faf45384fff25c4cf77dd5c96279283ad Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:36:34 -0800 Subject: [NET]: Annotate callers of csum_partial_copy_...() and csum_and_copy...() in net/* Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 +-- include/net/sock.h | 2 +- net/core/datagram.c | 6 +++--- net/core/skbuff.c | 2 +- net/ipv4/ip_output.c | 2 +- 5 files changed, 7 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e3ae544b3956..64fa7f4c702d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1199,8 +1199,7 @@ static inline int skb_add_data(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_NONE) { int err = 0; - unsigned int csum = csum_and_copy_from_user(from, - skb_put(skb, copy), + __wsum csum = csum_and_copy_from_user(from, skb_put(skb, copy), copy, 0, &err); if (!err) { skb->csum = csum_block_add(skb->csum, csum, off); diff --git a/include/net/sock.h b/include/net/sock.h index 35ffbdd35d3e..dc4b92b8abea 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1088,7 +1088,7 @@ static inline int skb_copy_to_page(struct sock *sk, char __user *from, { if (skb->ip_summed == CHECKSUM_NONE) { int err = 0; - unsigned int csum = csum_and_copy_from_user(from, + __wsum csum = csum_and_copy_from_user(from, page_address(page) + off, copy, 0, &err); if (err) diff --git a/net/core/datagram.c b/net/core/datagram.c index e5a05a046fef..0d9c9bac4006 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -321,7 +321,7 @@ fault: static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, u8 __user *to, int len, - unsigned int *csump) + __wsum *csump) { int start = skb_headlen(skb); int pos = 0; @@ -350,7 +350,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; int err = 0; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; @@ -386,7 +386,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset, end = start + list->len; if ((copy = end - offset) > 0) { - unsigned int csum2 = 0; + __wsum csum2 = 0; if (copy > len) copy = len; if (skb_copy_and_csum_datagram(list, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index c0e3427057fc..da6683f4b31d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1342,7 +1342,7 @@ unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, end = start + skb_shinfo(skb)->frags[i].size; if ((copy = end - offset) > 0) { - unsigned int csum2; + __wsum csum2; u8 *vaddr; skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index f9194f7e39d3..1da3d32f8289 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1314,7 +1314,7 @@ void ip_flush_pending_frames(struct sock *sk) static int ip_reply_glue_bits(void *dptr, char *to, int offset, int len, int odd, struct sk_buff *skb) { - unsigned int csum; + __wsum csum; csum = csum_partial_copy_nocheck(dptr+offset, to, len, 0); skb->csum = csum_block_add(skb->csum, csum, odd); -- cgit v1.2.3 From 2bbbc86890ac4c911c5057f69af93853e52a42a8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:37:14 -0800 Subject: [NET]: Annotate skb_checksum() and callers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/core/skbuff.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 64fa7f4c702d..784129fb61d4 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1334,8 +1334,8 @@ extern int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, extern void skb_free_datagram(struct sock *sk, struct sk_buff *skb); extern void skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); -extern unsigned int skb_checksum(const struct sk_buff *skb, int offset, - int len, unsigned int csum); +extern __wsum skb_checksum(const struct sk_buff *skb, int offset, + int len, __wsum csum); extern int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len); extern int skb_store_bits(const struct sk_buff *skb, int offset, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 6e8c15b39eb8..be9b541f536f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1240,8 +1240,8 @@ EXPORT_SYMBOL(skb_store_bits); /* Checksum skb data. */ -unsigned int skb_checksum(const struct sk_buff *skb, int offset, - int len, unsigned int csum) +__wsum skb_checksum(const struct sk_buff *skb, int offset, + int len, __wsum csum) { int start = skb_headlen(skb); int i, copy = start - offset; -- cgit v1.2.3 From 81d77662763ae527ba3a9b9275467901aaab7dfd Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:37:33 -0800 Subject: [NET]: Annotate skb_copy_and_csum_bits() and callers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- net/core/skbuff.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 784129fb61d4..874ca029fbb9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1340,9 +1340,9 @@ extern int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len); extern int skb_store_bits(const struct sk_buff *skb, int offset, void *from, int len); -extern unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, +extern __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, u8 *to, int len, - unsigned int csum); + __wsum csum); extern void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to); extern void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index be9b541f536f..07c25d601922 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1315,8 +1315,8 @@ __wsum skb_checksum(const struct sk_buff *skb, int offset, /* Both of above in one bottle. */ -unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, - u8 *to, int len, unsigned int csum) +__wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, + u8 *to, int len, __wsum csum) { int start = skb_headlen(skb); int i, copy = start - offset; @@ -1368,7 +1368,7 @@ unsigned int skb_copy_and_csum_bits(const struct sk_buff *skb, int offset, struct sk_buff *list = skb_shinfo(skb)->frag_list; for (; list; list = list->next) { - unsigned int csum2; + __wsum csum2; int end; BUG_TRAP(start <= offset + len); -- cgit v1.2.3 From b51655b958dfb1176bfcf99466231fdbef8751ff Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:40:42 -0800 Subject: [NET]: Annotate __skb_checksum_complete() and friends. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter.h | 6 +++--- include/linux/netfilter_ipv4.h | 2 +- include/linux/netfilter_ipv6.h | 2 +- include/linux/skbuff.h | 2 +- include/net/tcp.h | 2 +- include/net/udp.h | 8 ++++---- net/core/datagram.c | 2 +- net/core/netpoll.c | 4 ++-- net/ipv4/netfilter.c | 4 ++-- net/ipv4/tcp_input.c | 4 ++-- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/netfilter.c | 4 ++-- net/ipv6/tcp_ipv6.c | 2 +- 13 files changed, 22 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index b7e67d1d4382..707bb2e53c4e 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -290,7 +290,7 @@ extern u_int16_t nf_proto_csum_update(struct sk_buff *skb, struct nf_afinfo { unsigned short family; - unsigned int (*checksum)(struct sk_buff *skb, unsigned int hook, + __sum16 (*checksum)(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); void (*saveroute)(const struct sk_buff *skb, struct nf_info *info); @@ -305,12 +305,12 @@ static inline struct nf_afinfo *nf_get_afinfo(unsigned short family) return rcu_dereference(nf_afinfo[family]); } -static inline unsigned int +static inline __sum16 nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol, unsigned short family) { struct nf_afinfo *afinfo; - unsigned int csum = 0; + __sum16 csum = 0; rcu_read_lock(); afinfo = nf_get_afinfo(family); diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index 5b63a231a76b..5821eb5a0a3e 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -79,7 +79,7 @@ enum nf_ip_hook_priorities { #ifdef __KERNEL__ extern int ip_route_me_harder(struct sk_buff **pskb, unsigned addr_type); extern int ip_xfrm_me_harder(struct sk_buff **pskb); -extern unsigned int nf_ip_checksum(struct sk_buff *skb, unsigned int hook, +extern __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); #endif /*__KERNEL__*/ diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h index d97e268cdfe5..ab81a6dc94ea 100644 --- a/include/linux/netfilter_ipv6.h +++ b/include/linux/netfilter_ipv6.h @@ -74,7 +74,7 @@ enum nf_ip6_hook_priorities { #ifdef CONFIG_NETFILTER extern int ip6_route_me_harder(struct sk_buff *skb); -extern unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, +extern __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); extern int ipv6_netfilter_init(void); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 874ca029fbb9..41753667541d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1398,7 +1398,7 @@ static inline void skb_set_timestamp(struct sk_buff *skb, const struct timeval * extern void __net_timestamp(struct sk_buff *skb); -extern unsigned int __skb_checksum_complete(struct sk_buff *skb); +extern __sum16 __skb_checksum_complete(struct sk_buff *skb); /** * skb_checksum_complete - Calculate checksum of an entire packet diff --git a/include/net/tcp.h b/include/net/tcp.h index 826aaecdb994..aa7989c53791 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -814,7 +814,7 @@ static inline __sum16 tcp_v4_check(struct tcphdr *th, int len, return csum_tcpudp_magic(saddr,daddr,len,IPPROTO_TCP,base); } -static inline int __tcp_checksum_complete(struct sk_buff *skb) +static inline __sum16 __tcp_checksum_complete(struct sk_buff *skb) { return __skb_checksum_complete(skb); } diff --git a/include/net/udp.h b/include/net/udp.h index 39e825a6909a..c5ccd9a3387b 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -69,15 +69,15 @@ struct sk_buff; /* * Generic checksumming routines for UDP(-Lite) v4 and v6 */ -static inline u16 __udp_lib_checksum_complete(struct sk_buff *skb) +static inline __sum16 __udp_lib_checksum_complete(struct sk_buff *skb) { if (! UDP_SKB_CB(skb)->partial_cov) return __skb_checksum_complete(skb); - return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, - skb->csum)); + return csum_fold(skb_checksum(skb, 0, UDP_SKB_CB(skb)->cscov, + skb->csum)); } -static __inline__ int udp_lib_checksum_complete(struct sk_buff *skb) +static inline __sum16 udp_lib_checksum_complete(struct sk_buff *skb) { return skb->ip_summed != CHECKSUM_UNNECESSARY && __udp_lib_checksum_complete(skb); diff --git a/net/core/datagram.c b/net/core/datagram.c index 0d9c9bac4006..797fdd4352ce 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -411,7 +411,7 @@ fault: return -EFAULT; } -unsigned int __skb_checksum_complete(struct sk_buff *skb) +__sum16 __skb_checksum_complete(struct sk_buff *skb) { __sum16 sum; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 26ee1791aa02..8be3681d3d80 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -77,8 +77,8 @@ static void queue_process(void *p) } } -static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, - unsigned short ulen, __be32 saddr, __be32 daddr) +static __sum16 checksum_udp(struct sk_buff *skb, struct udphdr *uh, + unsigned short ulen, __be32 saddr, __be32 daddr) { __wsum psum; diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index b797a37c01ce..a68966059b50 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -162,11 +162,11 @@ static int nf_ip_reroute(struct sk_buff **pskb, const struct nf_info *info) return 0; } -unsigned int nf_ip_checksum(struct sk_buff *skb, unsigned int hook, +__sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol) { struct iphdr *iph = skb->nh.iph; - unsigned int csum = 0; + __sum16 csum = 0; switch (skb->ip_summed) { case CHECKSUM_COMPLETE: diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6ab3423674bb..9304034c0c47 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3790,9 +3790,9 @@ static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen) return err; } -static int __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) +static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) { - int result; + __sum16 result; if (sock_owned_by_user(sk)) { local_bh_enable(); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0ca8dead03b0..dadf80272413 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1544,7 +1544,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -static int tcp_v4_checksum_init(struct sk_buff *skb) +static __sum16 tcp_v4_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_COMPLETE) { if (!tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr, diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 8d1b542806c1..f6294e5bcb31 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -80,11 +80,11 @@ static int nf_ip6_reroute(struct sk_buff **pskb, const struct nf_info *info) return 0; } -unsigned int nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, +__sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol) { struct ipv6hdr *ip6h = skb->nh.ipv6h; - unsigned int csum = 0; + __sum16 csum = 0; switch (skb->ip_summed) { case CHECKSUM_COMPLETE: diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 394bc54c5c21..147ce499f509 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1527,7 +1527,7 @@ out: return NULL; } -static int tcp_v6_checksum_init(struct sk_buff *skb) +static __sum16 tcp_v6_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_COMPLETE) { if (!tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, -- cgit v1.2.3 From 43bc0ca7eadc024e9e5b935fa5e0892df4fec9eb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:43:23 -0800 Subject: [NET]: netfilter checksum annotations Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter.h | 26 +++++++++++++++---- .../linux/netfilter_ipv4/ip_conntrack_proto_gre.h | 4 +-- net/ipv4/netfilter/ip_nat_core.c | 6 ++--- net/ipv4/netfilter/ip_nat_helper.c | 30 ++++++++-------------- net/ipv4/netfilter/ip_nat_proto_gre.c | 8 +++--- net/ipv4/netfilter/ip_nat_proto_icmp.c | 6 ++--- net/ipv4/netfilter/ip_nat_proto_tcp.c | 5 ++-- net/ipv4/netfilter/ip_nat_proto_udp.c | 7 ++--- net/ipv4/netfilter/ipt_ECN.c | 11 +++----- net/ipv4/netfilter/ipt_TCPMSS.c | 24 ++++++----------- net/ipv4/netfilter/ipt_TOS.c | 5 ++-- net/ipv4/netfilter/ipt_TTL.c | 5 ++-- net/netfilter/core.c | 27 ++++++++----------- 13 files changed, 70 insertions(+), 94 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 707bb2e53c4e..6ab5e2d6133e 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -282,11 +282,27 @@ extern void nf_invalidate_cache(int pf); Returns true or false. */ extern int skb_make_writable(struct sk_buff **pskb, unsigned int writable_len); -extern u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval, - u_int32_t csum); -extern u_int16_t nf_proto_csum_update(struct sk_buff *skb, - u_int32_t oldval, u_int32_t newval, - u_int16_t csum, int pseudohdr); +static inline void nf_csum_replace4(__sum16 *sum, __be32 from, __be32 to) +{ + __be32 diff[] = { ~from, to }; + + *sum = csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum_unfold(*sum))); +} + +static inline void nf_csum_replace2(__sum16 *sum, __be16 from, __be16 to) +{ + nf_csum_replace4(sum, (__force __be32)from, (__force __be32)to); +} + +extern void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb, + __be32 from, __be32 to, int pseudohdr); + +static inline void nf_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb, + __be16 from, __be16 to, int pseudohdr) +{ + nf_proto_csum_replace4(sum, skb, (__force __be32)from, + (__force __be32)to, pseudohdr); +} struct nf_afinfo { unsigned short family; diff --git a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h index 1d853aa873eb..e371e0fc1672 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_proto_gre.h @@ -102,11 +102,11 @@ static inline __be32 *gre_key(struct gre_hdr *greh) } /* get pointer ot gre csum, if present */ -static inline u_int16_t *gre_csum(struct gre_hdr *greh) +static inline __sum16 *gre_csum(struct gre_hdr *greh) { if (!greh->csum) return NULL; - return (u_int16_t *) (greh+sizeof(*greh)); + return (__sum16 *) (greh+sizeof(*greh)); } #endif /* __KERNEL__ */ diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c index 4b6260a97408..9d1a5175dcd4 100644 --- a/net/ipv4/netfilter/ip_nat_core.c +++ b/net/ipv4/netfilter/ip_nat_core.c @@ -362,12 +362,10 @@ manip_pkt(u_int16_t proto, iph = (void *)(*pskb)->data + iphdroff; if (maniptype == IP_NAT_MANIP_SRC) { - iph->check = nf_csum_update(~iph->saddr, target->src.ip, - iph->check); + nf_csum_replace4(&iph->check, iph->saddr, target->src.ip); iph->saddr = target->src.ip; } else { - iph->check = nf_csum_update(~iph->daddr, target->dst.ip, - iph->check); + nf_csum_replace4(&iph->check, iph->daddr, target->dst.ip); iph->daddr = target->dst.ip; } return 1; diff --git a/net/ipv4/netfilter/ip_nat_helper.c b/net/ipv4/netfilter/ip_nat_helper.c index 3e7fd64c2162..ee80feb4b2a9 100644 --- a/net/ipv4/netfilter/ip_nat_helper.c +++ b/net/ipv4/netfilter/ip_nat_helper.c @@ -188,10 +188,8 @@ ip_nat_mangle_tcp_packet(struct sk_buff **pskb, csum_partial((char *)tcph, datalen, 0)); } else - tcph->check = nf_proto_csum_update(*pskb, - htons(oldlen) ^ htons(0xFFFF), - htons(datalen), - tcph->check, 1); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldlen), htons(datalen), 1); if (rep_len != match_len) { set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); @@ -266,10 +264,8 @@ ip_nat_mangle_udp_packet(struct sk_buff **pskb, if (!udph->check) udph->check = CSUM_MANGLED_0; } else - udph->check = nf_proto_csum_update(*pskb, - htons(oldlen) ^ htons(0xFFFF), - htons(datalen), - udph->check, 1); + nf_proto_csum_replace2(&udph->check, *pskb, + htons(oldlen), htons(datalen), 1); return 1; } EXPORT_SYMBOL(ip_nat_mangle_udp_packet); @@ -307,14 +303,10 @@ sack_adjust(struct sk_buff *skb, ntohl(sack->start_seq), new_start_seq, ntohl(sack->end_seq), new_end_seq); - tcph->check = nf_proto_csum_update(skb, - ~sack->start_seq, - new_start_seq, - tcph->check, 0); - tcph->check = nf_proto_csum_update(skb, - ~sack->end_seq, - new_end_seq, - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->start_seq, new_start_seq, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->end_seq, new_end_seq, 0); sack->start_seq = new_start_seq; sack->end_seq = new_end_seq; sackoff += sizeof(*sack); @@ -397,10 +389,8 @@ ip_nat_seq_adjust(struct sk_buff **pskb, else newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before); - tcph->check = nf_proto_csum_update(*pskb, ~tcph->seq, newseq, - tcph->check, 0); - tcph->check = nf_proto_csum_update(*pskb, ~tcph->ack_seq, newack, - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->seq, newseq, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->ack_seq, newack, 0); DEBUGP("Adjusting sequence number from %u->%u, ack from %u->%u\n", ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq), diff --git a/net/ipv4/netfilter/ip_nat_proto_gre.c b/net/ipv4/netfilter/ip_nat_proto_gre.c index bf91f9312b3c..95810202d849 100644 --- a/net/ipv4/netfilter/ip_nat_proto_gre.c +++ b/net/ipv4/netfilter/ip_nat_proto_gre.c @@ -129,11 +129,9 @@ gre_manip_pkt(struct sk_buff **pskb, } if (greh->csum) { /* FIXME: Never tested this code... */ - *(gre_csum(greh)) = - nf_proto_csum_update(*pskb, - ~*(gre_key(greh)), - tuple->dst.u.gre.key, - *(gre_csum(greh)), 0); + nf_proto_csum_replace4(gre_csum(greh), *pskb, + *(gre_key(greh)), + tuple->dst.u.gre.key, 0); } *(gre_key(greh)) = tuple->dst.u.gre.key; break; diff --git a/net/ipv4/netfilter/ip_nat_proto_icmp.c b/net/ipv4/netfilter/ip_nat_proto_icmp.c index 3f6efc13ac74..75266fe3e0fa 100644 --- a/net/ipv4/netfilter/ip_nat_proto_icmp.c +++ b/net/ipv4/netfilter/ip_nat_proto_icmp.c @@ -66,10 +66,8 @@ icmp_manip_pkt(struct sk_buff **pskb, return 0; hdr = (struct icmphdr *)((*pskb)->data + hdroff); - hdr->checksum = nf_proto_csum_update(*pskb, - hdr->un.echo.id ^ htons(0xFFFF), - tuple->src.u.icmp.id, - hdr->checksum, 0); + nf_proto_csum_replace2(&hdr->checksum, *pskb, + hdr->un.echo.id, tuple->src.u.icmp.id, 0); hdr->un.echo.id = tuple->src.u.icmp.id; return 1; } diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c index 12deb13b93b1..b586d18b3fb3 100644 --- a/net/ipv4/netfilter/ip_nat_proto_tcp.c +++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c @@ -129,9 +129,8 @@ tcp_manip_pkt(struct sk_buff **pskb, if (hdrsize < sizeof(*hdr)) return 1; - hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, hdr->check, 1); - hdr->check = nf_proto_csum_update(*pskb, oldport ^ htons(0xFFFF), newport, - hdr->check, 0); + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, oldport, newport, 0); return 1; } diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c index 82f8a6ab07ec..5ced0877b32f 100644 --- a/net/ipv4/netfilter/ip_nat_proto_udp.c +++ b/net/ipv4/netfilter/ip_nat_proto_udp.c @@ -115,11 +115,8 @@ udp_manip_pkt(struct sk_buff **pskb, } if (hdr->check || (*pskb)->ip_summed == CHECKSUM_PARTIAL) { - hdr->check = nf_proto_csum_update(*pskb, ~oldip, newip, - hdr->check, 1); - hdr->check = nf_proto_csum_update(*pskb, - *portptr ^ htons(0xFFFF), newport, - hdr->check, 0); + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, *portptr, newport, 0); if (!hdr->check) hdr->check = CSUM_MANGLED_0; } diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c index 1aa4517fbcdb..b55d670a24df 100644 --- a/net/ipv4/netfilter/ipt_ECN.c +++ b/net/ipv4/netfilter/ipt_ECN.c @@ -28,17 +28,16 @@ static inline int set_ect_ip(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) { struct iphdr *iph = (*pskb)->nh.iph; - u_int16_t oldtos; if ((iph->tos & IPT_ECN_IP_MASK) != (einfo->ip_ect & IPT_ECN_IP_MASK)) { + __u8 oldtos; if (!skb_make_writable(pskb, sizeof(struct iphdr))) return 0; iph = (*pskb)->nh.iph; oldtos = iph->tos; iph->tos &= ~IPT_ECN_IP_MASK; iph->tos |= (einfo->ip_ect & IPT_ECN_IP_MASK); - iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), - htons(iph->tos), iph->check); + nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos)); } return 1; } @@ -72,10 +71,8 @@ set_ect_tcp(struct sk_buff **pskb, const struct ipt_ECN_info *einfo) if (einfo->operation & IPT_ECN_OP_SET_CWR) tcph->cwr = einfo->proto.tcp.cwr; - tcph->check = nf_proto_csum_update((*pskb), - oldval ^ htons(0xFFFF), - ((__be16 *)tcph)[6], - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + oldval, ((__be16 *)tcph)[6], 0); return 1; } diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c index 108b6b76311f..93eb5c3c1884 100644 --- a/net/ipv4/netfilter/ipt_TCPMSS.c +++ b/net/ipv4/netfilter/ipt_TCPMSS.c @@ -97,10 +97,8 @@ ipt_tcpmss_target(struct sk_buff **pskb, opt[i+2] = (newmss & 0xff00) >> 8; opt[i+3] = (newmss & 0x00ff); - tcph->check = nf_proto_csum_update(*pskb, - htons(oldmss)^htons(0xFFFF), - htons(newmss), - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldmss), htons(newmss), 0); return IPT_CONTINUE; } } @@ -126,28 +124,22 @@ ipt_tcpmss_target(struct sk_buff **pskb, opt = (u_int8_t *)tcph + sizeof(struct tcphdr); memmove(opt + TCPOLEN_MSS, opt, tcplen - sizeof(struct tcphdr)); - tcph->check = nf_proto_csum_update(*pskb, - htons(tcplen) ^ htons(0xFFFF), - htons(tcplen + TCPOLEN_MSS), - tcph->check, 1); + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(tcplen), htons(tcplen + TCPOLEN_MSS), 1); opt[0] = TCPOPT_MSS; opt[1] = TCPOLEN_MSS; opt[2] = (newmss & 0xff00) >> 8; opt[3] = (newmss & 0x00ff); - tcph->check = nf_proto_csum_update(*pskb, htonl(~0), *((__be32 *)opt), - tcph->check, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, 0, *((__be32 *)opt), 0); oldval = ((__be16 *)tcph)[6]; tcph->doff += TCPOLEN_MSS/4; - tcph->check = nf_proto_csum_update(*pskb, - oldval ^ htons(0xFFFF), - ((__be16 *)tcph)[6], - tcph->check, 0); + nf_proto_csum_replace2(&tcph->check, *pskb, + oldval, ((__be16 *)tcph)[6], 0); newtotlen = htons(ntohs(iph->tot_len) + TCPOLEN_MSS); - iph->check = nf_csum_update(iph->tot_len ^ htons(0xFFFF), - newtotlen, iph->check); + nf_csum_replace2(&iph->check, iph->tot_len, newtotlen); iph->tot_len = newtotlen; return IPT_CONTINUE; } diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c index 83b80b3a5d2f..18e74ac4d425 100644 --- a/net/ipv4/netfilter/ipt_TOS.c +++ b/net/ipv4/netfilter/ipt_TOS.c @@ -30,16 +30,15 @@ target(struct sk_buff **pskb, { const struct ipt_tos_target_info *tosinfo = targinfo; struct iphdr *iph = (*pskb)->nh.iph; - u_int16_t oldtos; if ((iph->tos & IPTOS_TOS_MASK) != tosinfo->tos) { + __u8 oldtos; if (!skb_make_writable(pskb, sizeof(struct iphdr))) return NF_DROP; iph = (*pskb)->nh.iph; oldtos = iph->tos; iph->tos = (iph->tos & IPTOS_PREC_MASK) | tosinfo->tos; - iph->check = nf_csum_update(htons(oldtos) ^ htons(0xFFFF), - htons(iph->tos), iph->check); + nf_csum_replace2(&iph->check, htons(oldtos), htons(iph->tos)); } return IPT_CONTINUE; } diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c index ac9517d62af0..fffe5ca82e91 100644 --- a/net/ipv4/netfilter/ipt_TTL.c +++ b/net/ipv4/netfilter/ipt_TTL.c @@ -54,9 +54,8 @@ ipt_ttl_target(struct sk_buff **pskb, } if (new_ttl != iph->ttl) { - iph->check = nf_csum_update(htons((iph->ttl << 8)) ^ htons(0xFFFF), - htons(new_ttl << 8), - iph->check); + nf_csum_replace2(&iph->check, htons(iph->ttl << 8), + htons(new_ttl << 8)); iph->ttl = new_ttl; } diff --git a/net/netfilter/core.c b/net/netfilter/core.c index d80b935b3a92..17f9e1cbc73b 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -222,28 +222,21 @@ copy_skb: } EXPORT_SYMBOL(skb_make_writable); -u_int16_t nf_csum_update(u_int32_t oldval, u_int32_t newval, u_int32_t csum) -{ - u_int32_t diff[] = { oldval, newval }; - - return csum_fold(csum_partial((char *)diff, sizeof(diff), ~csum)); -} -EXPORT_SYMBOL(nf_csum_update); - -u_int16_t nf_proto_csum_update(struct sk_buff *skb, - u_int32_t oldval, u_int32_t newval, - u_int16_t csum, int pseudohdr) +void nf_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb, + __be32 from, __be32 to, int pseudohdr) { + __be32 diff[] = { ~from, to }; if (skb->ip_summed != CHECKSUM_PARTIAL) { - csum = nf_csum_update(oldval, newval, csum); + *sum = csum_fold(csum_partial((char *)diff, sizeof(diff), + ~csum_unfold(*sum))); if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr) - skb->csum = nf_csum_update(oldval, newval, skb->csum); + skb->csum = ~csum_partial((char *)diff, sizeof(diff), + ~skb->csum); } else if (pseudohdr) - csum = ~nf_csum_update(oldval, newval, ~csum); - - return csum; + *sum = ~csum_fold(csum_partial((char *)diff, sizeof(diff), + csum_unfold(*sum))); } -EXPORT_SYMBOL(nf_proto_csum_update); +EXPORT_SYMBOL(nf_proto_csum_replace4); /* This does not belong here, but locally generated errors need it if connection tracking in use: without this, connection may not be in hash table, and hence -- cgit v1.2.3 From 1f61ab5ca5cca939a6509892d84b34849e155036 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 14 Nov 2006 21:44:08 -0800 Subject: [NET]: Preliminaty annotation of skb->csum. It's still not completely right; we need to split it into anon unions of __wsum and unsigned - for cases when we use it for partial checksum and for offset of checksum in skb Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 41753667541d..fcab543d79ac 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -273,8 +273,8 @@ struct sk_buff { unsigned int len, data_len, - mac_len, - csum; + mac_len; + __wsum csum; __u32 priority; __u8 local_df:1, cloned:1, -- cgit v1.2.3 From bac0dff6cd194f7a28f7e840c9b6a7aa71c6ef97 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Sun, 19 Nov 2006 14:15:05 -0800 Subject: [BNX2]: Add 5709 PCI ID. Add PCI ID and detection for 5709 copper and SerDes chips. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 13 +++++++++++-- include/linux/pci_ids.h | 1 + 2 files changed, 12 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 92897efbc263..157463b4fa9f 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -86,6 +86,7 @@ typedef enum { NC370F, BCM5708, BCM5708S, + BCM5709, } board_t; /* indexed by board_t, above */ @@ -99,6 +100,7 @@ static const struct { { "HP NC370F Multifunction Gigabit Server Adapter" }, { "Broadcom NetXtreme II BCM5708 1000Base-T" }, { "Broadcom NetXtreme II BCM5708 1000Base-SX" }, + { "Broadcom NetXtreme II BCM5709 1000Base-T" }, }; static struct pci_device_id bnx2_pci_tbl[] = { @@ -116,6 +118,8 @@ static struct pci_device_id bnx2_pci_tbl[] = { PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706S }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5708S, PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5708S }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5709, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5709 }, { 0, } }; @@ -5854,10 +5858,15 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->phy_addr = 1; /* Disable WOL support if we are running on a SERDES chip. */ - if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) { + if (CHIP_NUM(bp) == CHIP_NUM_5709) { + if (CHIP_BOND_ID(bp) != BNX2_MISC_DUAL_MEDIA_CTRL_BOND_ID_C) + bp->phy_flags |= PHY_SERDES_FLAG; + } else if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) bp->phy_flags |= PHY_SERDES_FLAG; + + if (bp->phy_flags & PHY_SERDES_FLAG) { bp->flags |= NO_WOL_FLAG; - if (CHIP_NUM(bp) == CHIP_NUM_5708) { + if (CHIP_NUM(bp) != CHIP_NUM_5706) { bp->phy_addr = 2; reg = REG_RD_IND(bp, bp->shmem_base + BNX2_SHARED_HW_CFG_CONFIG); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index fd5033b8a927..c09da1e30c54 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1897,6 +1897,7 @@ #define PCI_VENDOR_ID_BROADCOM 0x14e4 #define PCI_DEVICE_ID_TIGON3_5752 0x1600 #define PCI_DEVICE_ID_TIGON3_5752M 0x1601 +#define PCI_DEVICE_ID_NX2_5709 0x1639 #define PCI_DEVICE_ID_TIGON3_5700 0x1644 #define PCI_DEVICE_ID_TIGON3_5701 0x1645 #define PCI_DEVICE_ID_TIGON3_5702 0x1646 -- cgit v1.2.3 From 82e3ab9dbeebd5c8d5402ad1607d22086271a56d Mon Sep 17 00:00:00 2001 From: Ian McDonald Date: Mon, 20 Nov 2006 19:19:32 -0200 Subject: [DCCP]: Adds the tx buffer sysctls This one got lost on the way from Ian to Gerrit to me, fix it. Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- Documentation/networking/dccp.txt | 4 ++++ include/linux/sysctl.h | 1 + net/dccp/sysctl.c | 9 +++++++++ 3 files changed, 14 insertions(+) (limited to 'include/linux') diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt index 1910d097a0ad..dda15886bcb5 100644 --- a/Documentation/networking/dccp.txt +++ b/Documentation/networking/dccp.txt @@ -100,6 +100,10 @@ rx_ccid = 2 seq_window = 100 The initial sequence window (sec. 7.5.2). +tx_qlen = 5 + The size of the transmit buffer in packets. A value of 0 corresponds + to an unbounded transmit buffer. + Notes ===== diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 2e8c5ad82793..61dd99c6c2a5 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -617,6 +617,7 @@ enum { NET_DCCP_DEFAULT_REQ_RETRIES = 7, NET_DCCP_DEFAULT_RETRIES1 = 8, NET_DCCP_DEFAULT_RETRIES2 = 9, + NET_DCCP_DEFAULT_TX_QLEN = 10, }; /* /proc/sys/net/ipx */ diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 8b62061e5701..4775ba3faa04 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -91,6 +91,15 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, + .procname = "tx_qlen", + .data = &sysctl_dccp_tx_qlen, + .maxlen = sizeof(sysctl_dccp_tx_qlen), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { .ctl_name = 0, } }; -- cgit v1.2.3 From f3ffaf14681e3cad61006873be8656ab41b793e0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 16:59:12 -0800 Subject: [SCTP]: Annotate SCTP headers. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/sctp.h | 72 ++++++++++++++++++++++++++-------------------------- 1 file changed, 36 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sctp.h b/include/linux/sctp.h index 6ec66dec29f7..35108fe7a686 100644 --- a/include/linux/sctp.h +++ b/include/linux/sctp.h @@ -57,17 +57,17 @@ /* Section 3.1. SCTP Common Header Format */ typedef struct sctphdr { - __u16 source; - __u16 dest; - __u32 vtag; - __u32 checksum; + __be16 source; + __be16 dest; + __be32 vtag; + __be32 checksum; } __attribute__((packed)) sctp_sctphdr_t; /* Section 3.2. Chunk Field Descriptions. */ typedef struct sctp_chunkhdr { __u8 type; __u8 flags; - __u16 length; + __be16 length; } __attribute__((packed)) sctp_chunkhdr_t; @@ -153,8 +153,8 @@ enum { SCTP_CHUNK_FLAG_T = 0x01 }; */ typedef struct sctp_paramhdr { - __u16 type; - __u16 length; + __be16 type; + __be16 length; } __attribute__((packed)) sctp_paramhdr_t; typedef enum { @@ -203,10 +203,10 @@ enum { SCTP_PARAM_ACTION_MASK = __constant_htons(0xc000), }; /* RFC 2960 Section 3.3.1 Payload Data (DATA) (0) */ typedef struct sctp_datahdr { - __u32 tsn; - __u16 stream; - __u16 ssn; - __u32 ppid; + __be32 tsn; + __be16 stream; + __be16 ssn; + __be32 ppid; __u8 payload[0]; } __attribute__((packed)) sctp_datahdr_t; @@ -232,11 +232,11 @@ enum { SCTP_DATA_FRAG_MASK = 0x03, }; * endpoints. */ typedef struct sctp_inithdr { - __u32 init_tag; - __u32 a_rwnd; - __u16 num_outbound_streams; - __u16 num_inbound_streams; - __u32 initial_tsn; + __be32 init_tag; + __be32 a_rwnd; + __be16 num_outbound_streams; + __be16 num_inbound_streams; + __be32 initial_tsn; __u8 params[0]; } __attribute__((packed)) sctp_inithdr_t; @@ -261,7 +261,7 @@ typedef struct sctp_ipv6addr_param { /* Section 3.3.2.1 Cookie Preservative (9) */ typedef struct sctp_cookie_preserve_param { sctp_paramhdr_t param_hdr; - uint32_t lifespan_increment; + __be32 lifespan_increment; } __attribute__((packed)) sctp_cookie_preserve_param_t; /* Section 3.3.2.1 Host Name Address (11) */ @@ -273,7 +273,7 @@ typedef struct sctp_hostname_param { /* Section 3.3.2.1 Supported Address Types (12) */ typedef struct sctp_supported_addrs_param { sctp_paramhdr_t param_hdr; - uint16_t types[0]; + __be16 types[0]; } __attribute__((packed)) sctp_supported_addrs_param_t; /* Appendix A. ECN Capable (32768) */ @@ -284,7 +284,7 @@ typedef struct sctp_ecn_capable_param { /* ADDIP Section 3.2.6 Adaption Layer Indication */ typedef struct sctp_adaption_ind_param { struct sctp_paramhdr param_hdr; - __u32 adaption_ind; + __be32 adaption_ind; } __attribute__((packed)) sctp_adaption_ind_param_t; /* RFC 2960. Section 3.3.3 Initiation Acknowledgement (INIT ACK) (2): @@ -316,11 +316,11 @@ typedef struct sctp_unrecognized_param { */ typedef struct sctp_gap_ack_block { - __u16 start; - __u16 end; + __be16 start; + __be16 end; } __attribute__((packed)) sctp_gap_ack_block_t; -typedef uint32_t sctp_dup_tsn_t; +typedef __be32 sctp_dup_tsn_t; typedef union { sctp_gap_ack_block_t gab; @@ -328,10 +328,10 @@ typedef union { } sctp_sack_variable_t; typedef struct sctp_sackhdr { - __u32 cum_tsn_ack; - __u32 a_rwnd; - __u16 num_gap_ack_blocks; - __u16 num_dup_tsns; + __be32 cum_tsn_ack; + __be32 a_rwnd; + __be16 num_gap_ack_blocks; + __be16 num_dup_tsns; sctp_sack_variable_t variable[0]; } __attribute__((packed)) sctp_sackhdr_t; @@ -371,7 +371,7 @@ typedef struct sctp_abort_chunk { * and the highest consecutive acking value. */ typedef struct sctp_shutdownhdr { - __u32 cum_tsn_ack; + __be32 cum_tsn_ack; } __attribute__((packed)) sctp_shutdownhdr_t; struct sctp_shutdown_chunk_t { @@ -382,8 +382,8 @@ struct sctp_shutdown_chunk_t { /* RFC 2960. Section 3.3.10 Operation Error (ERROR) (9) */ typedef struct sctp_errhdr { - __u16 cause; - __u16 length; + __be16 cause; + __be16 length; __u8 variable[0]; } __attribute__((packed)) sctp_errhdr_t; @@ -462,7 +462,7 @@ typedef enum { * Explicit Congestion Notification Echo (ECNE) (12) */ typedef struct sctp_ecnehdr { - __u32 lowest_tsn; + __be32 lowest_tsn; } sctp_ecnehdr_t; typedef struct sctp_ecne_chunk { @@ -474,7 +474,7 @@ typedef struct sctp_ecne_chunk { * Congestion Window Reduced (CWR) (13) */ typedef struct sctp_cwrhdr { - __u32 lowest_tsn; + __be32 lowest_tsn; } sctp_cwrhdr_t; typedef struct sctp_cwr_chunk { @@ -529,12 +529,12 @@ typedef struct sctp_cwr_chunk { * chunks this field MUST be filled in. */ struct sctp_fwdtsn_skip { - __u16 stream; - __u16 ssn; + __be16 stream; + __be16 ssn; } __attribute__((packed)); struct sctp_fwdtsn_hdr { - __u32 new_cum_tsn; + __be32 new_cum_tsn; struct sctp_fwdtsn_skip skip[0]; } __attribute((packed)); @@ -578,11 +578,11 @@ struct sctp_fwdtsn_chunk { */ typedef struct sctp_addip_param { sctp_paramhdr_t param_hdr; - __u32 crr_id; + __be32 crr_id; } __attribute__((packed)) sctp_addip_param_t; typedef struct sctp_addiphdr { - __u32 serial; + __be32 serial; __u8 params[0]; } __attribute__((packed)) sctp_addiphdr_t; -- cgit v1.2.3 From 962c837275e8a8c1df41f1882e971636093cdee4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 17:26:08 -0800 Subject: [SCTP]: Netfilter sctp annotations. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sctp.h | 2 +- net/netfilter/xt_sctp.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h index b8994d9fd1a9..5cf2c115cce4 100644 --- a/include/linux/netfilter/nf_conntrack_sctp.h +++ b/include/linux/netfilter/nf_conntrack_sctp.h @@ -20,7 +20,7 @@ struct ip_ct_sctp { enum sctp_conntrack state; - u_int32_t vtag[IP_CT_DIR_MAX]; + __be32 vtag[IP_CT_DIR_MAX]; u_int32_t ttag[IP_CT_DIR_MAX]; }; diff --git a/net/netfilter/xt_sctp.c b/net/netfilter/xt_sctp.c index 7956acaaa24b..71bf036f833c 100644 --- a/net/netfilter/xt_sctp.c +++ b/net/netfilter/xt_sctp.c @@ -71,7 +71,7 @@ match_packet(const struct sk_buff *skb, duprintf("Chunk num: %d\toffset: %d\ttype: %d\tlength: %d\tflags: %x\n", ++i, offset, sch->type, htons(sch->length), sch->flags); - offset += (htons(sch->length) + 3) & ~3; + offset += (ntohs(sch->length) + 3) & ~3; duprintf("skb->len: %d\toffset: %d\n", skb->len, offset); -- cgit v1.2.3 From ff1dcadb1b55dbf471c5ed109dbbdf06bd19ef3b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2006 18:07:29 -0800 Subject: [NET]: Split skb->csum ... into anonymous union of __wsum and __u32 (csum and csum_offset resp.) Signed-off-by: Al Viro Signed-off-by: David S. Miller --- drivers/net/cassini.c | 2 +- drivers/net/e1000/e1000_main.c | 2 +- drivers/net/ixgb/ixgb_main.c | 2 +- drivers/net/myri10ge/myri10ge.c | 2 +- drivers/net/sk98lin/skge.c | 4 ++-- drivers/net/skge.c | 2 +- drivers/net/sky2.c | 2 +- drivers/net/sungem.c | 2 +- drivers/net/sunhme.c | 2 +- include/linux/skbuff.h | 5 ++++- net/core/dev.c | 4 ++-- net/core/skbuff.c | 2 +- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/udp.c | 2 +- net/ipv6/tcp_ipv6.c | 4 ++-- 15 files changed, 22 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c index 521c5b71023c..fd2cc13f7d97 100644 --- a/drivers/net/cassini.c +++ b/drivers/net/cassini.c @@ -2825,7 +2825,7 @@ static inline int cas_xmit_tx_ringN(struct cas *cp, int ring, u64 csum_start_off, csum_stuff_off; csum_start_off = (u64) (skb->h.raw - skb->data); - csum_stuff_off = (u64) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; ctrl = TX_DESC_CSUM_EN | CAS_BASE(TX_DESC_CSUM_START, csum_start_off) | diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 7a0828869ecf..32dde0adb683 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2826,7 +2826,7 @@ e1000_tx_csum(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, context_desc = E1000_CONTEXT_DESC(*tx_ring, i); context_desc->upper_setup.tcp_fields.tucss = css; - context_desc->upper_setup.tcp_fields.tucso = css + skb->csum; + context_desc->upper_setup.tcp_fields.tucso = css + skb->csum_offset; context_desc->upper_setup.tcp_fields.tucse = 0; context_desc->tcp_seg_setup.data = 0; context_desc->cmd_and_length = cpu_to_le32(E1000_TXD_CMD_DEXT); diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index e09f575a3a38..7b127212e62b 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -1249,7 +1249,7 @@ ixgb_tx_csum(struct ixgb_adapter *adapter, struct sk_buff *skb) if(likely(skb->ip_summed == CHECKSUM_PARTIAL)) { struct ixgb_buffer *buffer_info; css = skb->h.raw - skb->data; - cso = (skb->h.raw + skb->csum) - skb->data; + cso = css + skb->csum_offset; i = adapter->tx_ring.next_to_use; context_desc = IXGB_CONTEXT_DESC(adapter->tx_ring, i); diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 806081b59733..36350e6db1c1 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -1955,7 +1955,7 @@ again: flags = (MXGEFW_FLAGS_NO_TSO | MXGEFW_FLAGS_FIRST); if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { cksum_offset = (skb->h.raw - skb->data); - pseudo_hdr_offset = (skb->h.raw + skb->csum) - skb->data; + pseudo_hdr_offset = cksum_offset + skb->csum_offset; /* If the headers are excessively large, then we must * fall back to a software checksum */ if (unlikely(cksum_offset > 255 || pseudo_hdr_offset > 127)) { diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index a5d41ebc9fb4..12cbfd190dd7 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -1562,7 +1562,7 @@ struct sk_buff *pMessage) /* pointer to send-message */ if (pMessage->ip_summed == CHECKSUM_PARTIAL) { u16 hdrlen = pMessage->h.raw - pMessage->data; - u16 offset = hdrlen + pMessage->csum; + u16 offset = hdrlen + pMessage->csum_offset; if ((pMessage->h.ipiph->protocol == IPPROTO_UDP ) && (pAC->GIni.GIChipRev == 0) && @@ -1681,7 +1681,7 @@ struct sk_buff *pMessage) /* pointer to send-message */ */ if (pMessage->ip_summed == CHECKSUM_PARTIAL) { u16 hdrlen = pMessage->h.raw - pMessage->data; - u16 offset = hdrlen + pMessage->csum; + u16 offset = hdrlen + pMessage->csum_offset; Control = BMU_STFWD; diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 27b537c8d5e3..5513907e8393 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -2565,7 +2565,7 @@ static int skge_xmit_frame(struct sk_buff *skb, struct net_device *dev) td->csum_offs = 0; td->csum_start = offset; - td->csum_write = offset + skb->csum; + td->csum_write = offset + skb->csum_offset; } else control = BMU_CHECK; diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 0ef1848b9761..842abd9396c6 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1350,7 +1350,7 @@ static int sky2_xmit_frame(struct sk_buff *skb, struct net_device *dev) u32 tcpsum; tcpsum = offset << 16; /* sum start */ - tcpsum |= offset + skb->csum; /* sum write */ + tcpsum |= offset + skb->csum_offset; /* sum write */ ctrl = CALSUM | WR_SUM | INIT_SUM | LOCK_SUM; if (skb->nh.iph->protocol == IPPROTO_UDP) diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 253e96e7ad20..334c6cfd6595 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -1030,7 +1030,7 @@ static int gem_start_xmit(struct sk_buff *skb, struct net_device *dev) u64 csum_start_off, csum_stuff_off; csum_start_off = (u64) (skb->h.raw - skb->data); - csum_stuff_off = (u64) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; ctrl = (TXDCTRL_CENAB | (csum_start_off << 15) | diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index 9d7cd130c19d..ec432ea879fb 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -2272,7 +2272,7 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev) u32 csum_start_off, csum_stuff_off; csum_start_off = (u32) (skb->h.raw - skb->data); - csum_stuff_off = (u32) ((skb->h.raw + skb->csum) - skb->data); + csum_stuff_off = csum_start_off + skb->csum_offset; tx_flags = (TXFLAG_OWN | TXFLAG_CSENABLE | ((csum_start_off << 14) & TXFLAG_CSBUFBEGIN) | diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fcab543d79ac..14ec16d2d9ba 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -274,7 +274,10 @@ struct sk_buff { unsigned int len, data_len, mac_len; - __wsum csum; + union { + __wsum csum; + __u32 csum_offset; + }; __u32 priority; __u8 local_df:1, cloned:1, diff --git a/net/core/dev.c b/net/core/dev.c index 1a36b17f4b51..59d058a3b504 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1191,9 +1191,9 @@ int skb_checksum_help(struct sk_buff *skb) offset = skb->tail - skb->h.raw; BUG_ON(offset <= 0); - BUG_ON(skb->csum + 2 > offset); + BUG_ON(skb->csum_offset + 2 > offset); - *(__sum16*)(skb->h.raw + skb->csum) = csum_fold(csum); + *(__sum16*)(skb->h.raw + skb->csum_offset) = csum_fold(csum); out_set_summed: skb->ip_summed = CHECKSUM_NONE; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 07c25d601922..a90bc439488e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1414,7 +1414,7 @@ void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to) skb->len - csstart, 0); if (skb->ip_summed == CHECKSUM_PARTIAL) { - long csstuff = csstart + skb->csum; + long csstuff = csstart + skb->csum_offset; *((__sum16 *)(to + csstuff)) = csum_fold(csum); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e9d467124c4d..4913f25e5ad5 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -504,7 +504,7 @@ void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) if (skb->ip_summed == CHECKSUM_PARTIAL) { th->check = ~tcp_v4_check(th, len, inet->saddr, inet->daddr, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); } else { th->check = tcp_v4_check(th, len, inet->saddr, inet->daddr, csum_partial((char *)th, @@ -526,7 +526,7 @@ int tcp_v4_gso_send_check(struct sk_buff *skb) th->check = 0; th->check = ~tcp_v4_check(th, skb->len, iph->saddr, iph->daddr, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); skb->ip_summed = CHECKSUM_PARTIAL; return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 28e4cf662ce0..1807a30694d9 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -422,7 +422,7 @@ static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, /* * Only one fragment on the socket. */ - skb->csum = offsetof(struct udphdr, check); + skb->csum_offset = offsetof(struct udphdr, check); uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); } else { /* diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0adb337c4b7e..517c50024bfc 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -948,7 +948,7 @@ static void tcp_v6_send_check(struct sock *sk, int len, struct sk_buff *skb) if (skb->ip_summed == CHECKSUM_PARTIAL) { th->check = ~csum_ipv6_magic(&np->saddr, &np->daddr, len, IPPROTO_TCP, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); } else { th->check = csum_ipv6_magic(&np->saddr, &np->daddr, len, IPPROTO_TCP, csum_partial((char *)th, th->doff<<2, @@ -970,7 +970,7 @@ static int tcp_v6_gso_send_check(struct sk_buff *skb) th->check = 0; th->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, skb->len, IPPROTO_TCP, 0); - skb->csum = offsetof(struct tcphdr, check); + skb->csum_offset = offsetof(struct tcphdr, check); skb->ip_summed = CHECKSUM_PARTIAL; return 0; } -- cgit v1.2.3 From d61c167dd0797a16584f7a922dd5d50efad1d28a Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 27 Nov 2006 12:31:45 -0200 Subject: [NET]: Add documentation for TFRC structures This adds documentation for the TFRC structure fields. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tfrc.h | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tfrc.h b/include/linux/tfrc.h index 7dab7831c3cb..31a9b25276fe 100644 --- a/include/linux/tfrc.h +++ b/include/linux/tfrc.h @@ -1,7 +1,8 @@ #ifndef _LINUX_TFRC_H_ #define _LINUX_TFRC_H_ /* - * include/linux/tfrc.h + * TFRC - Data Structures for the TCP-Friendly Rate Control congestion + * control mechanism as specified in RFC 3448. * * Copyright (c) 2005 The University of Waikato, Hamilton, New Zealand. * Copyright (c) 2005 Ian McDonald @@ -13,15 +14,30 @@ * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. */ - #include +/** tfrc_rx_info - TFRC Receiver Data Structure + * + * @tfrcrx_x_recv: receiver estimate of sending rate (3.2.2) + * @tfrcrx_rtt: round-trip-time (communicated by sender) + * @tfrcrx_p: current estimate of loss event rate (3.2.2) + */ struct tfrc_rx_info { __u32 tfrcrx_x_recv; __u32 tfrcrx_rtt; __u32 tfrcrx_p; }; +/** tfrc_tx_info - TFRC Sender Data Structure + * + * @tfrctx_x: computed transmit rate (4.3 (4)) + * @tfrctx_x_recv: receiver estimate of send rate (4.3) + * @tfrctx_x_calc: return value of throughput equation (3.1) + * @tfrctx_rtt: (moving average) estimate of RTT (4.3) + * @tfrctx_p: current loss event rate (5.4) + * @tfrctx_rto: estimate of RTO, equals 4*RTT (4.3) + * @tfrctx_ipi: inter-packet interval (4.6) + */ struct tfrc_tx_info { __u32 tfrctx_x; __u32 tfrctx_x_recv; -- cgit v1.2.3 From 4e9b82693542003b028c8494e9e3c49615b91ce7 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 27 Nov 2006 09:25:58 -0800 Subject: [NETLINK]: Remove unused dst_pid field in netlink_skb_parms The destination PID is passed directly to netlink_unicast() respectively netlink_multicast(). Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 - net/decnet/dn_route.c | 2 -- net/ipv4/fib_frontend.c | 1 - net/ipv4/netfilter/ip_conntrack_netlink.c | 4 +--- net/netfilter/nf_conntrack_netlink.c | 4 +--- net/netlink/af_netlink.c | 1 - net/xfrm/xfrm_user.c | 2 -- 7 files changed, 2 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index e61e1e138421..b3b9b609ee89 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -141,7 +141,6 @@ struct netlink_skb_parms { struct ucred creds; /* Skb credentials */ __u32 pid; - __u32 dst_pid; __u32 dst_group; kernel_cap_t eff_cap; __u32 loginuid; /* Login (audit) uid */ diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index f759d6f422ea..4eb985236aee 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1592,8 +1592,6 @@ int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg) if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; - err = dn_rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWROUTE, 0, 0); if (err == 0) diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 14025345cc56..d47b72af89ed 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -811,7 +811,6 @@ static void nl_fib_input(struct sock *sk, int len) pid = nlh->nlmsg_pid; /*pid of sending process */ NETLINK_CB(skb).pid = 0; /* from kernel */ - NETLINK_CB(skb).dst_pid = pid; NETLINK_CB(skb).dst_group = 0; /* unicast */ netlink_unicast(sk, skb, pid, MSG_DONTWAIT); } diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 55f0ae641081..1bb8ed33c5bc 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -743,7 +743,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, ip_conntrack_put(ct); return -ENOMEM; } - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; err = ctnetlink_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, 1, ct); @@ -1273,8 +1272,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) goto out; - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; - + err = ctnetlink_exp_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, 1, exp); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index ab67c2be2b5d..cfb35fd8d9b9 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -752,7 +752,6 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, nf_ct_put(ct); return -ENOMEM; } - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; err = ctnetlink_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, 1, ct); @@ -1300,8 +1299,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) goto out; - NETLINK_CB(skb2).dst_pid = NETLINK_CB(skb).pid; - + err = ctnetlink_exp_fill_info(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, 1, exp); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index f61d81b3c61c..3baafb10f8f3 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1153,7 +1153,6 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, goto out; NETLINK_CB(skb).pid = nlk->pid; - NETLINK_CB(skb).dst_pid = dst_pid; NETLINK_CB(skb).dst_group = dst_group; NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context); selinux_get_task_sid(current, &(NETLINK_CB(skb).sid)); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 6c4d2f49d9e5..8dbb38b91de2 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -650,7 +650,6 @@ static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb, if (!skb) return ERR_PTR(-ENOMEM); - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; info.in_skb = in_skb; info.out_skb = skb; info.nlmsg_seq = seq; @@ -1168,7 +1167,6 @@ static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb, if (!skb) return ERR_PTR(-ENOMEM); - NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid; info.in_skb = in_skb; info.out_skb = skb; info.nlmsg_seq = seq; -- cgit v1.2.3 From e3703b3de1f049b38733ba520e5038f23063068e Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Mon, 27 Nov 2006 09:27:07 -0800 Subject: [RTNETLINK]: Add rtnl_put_cacheinfo() to unify some code IPv4, IPv6, and DECNet all use struct rta_cacheinfo in a similiar way, therefore rtnl_put_cacheinfo() is added to reuse code. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 3 +++ net/core/rtnetlink.c | 20 ++++++++++++++++++++ net/decnet/dn_route.c | 16 +++++----------- net/ipv4/route.c | 26 +++++++++++--------------- net/ipv6/route.c | 19 ++++++------------- 5 files changed, 45 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 33b3d0ab3a91..493297acdae8 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -585,6 +585,9 @@ extern int rtnl_notify(struct sk_buff *skb, u32 pid, u32 group, struct nlmsghdr *nlh, gfp_t flags); extern void rtnl_set_sk_err(u32 group, int error); extern int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics); +extern int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, + u32 id, u32 ts, u32 tsage, long expires, + u32 error); extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 0cb4d9e53a07..e76539a5eb5e 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -212,6 +212,26 @@ nla_put_failure: return nla_nest_cancel(skb, mx); } +int rtnl_put_cacheinfo(struct sk_buff *skb, struct dst_entry *dst, u32 id, + u32 ts, u32 tsage, long expires, u32 error) +{ + struct rta_cacheinfo ci = { + .rta_lastuse = jiffies_to_clock_t(jiffies - dst->lastuse), + .rta_used = dst->__use, + .rta_clntref = atomic_read(&(dst->__refcnt)), + .rta_error = error, + .rta_id = id, + .rta_ts = ts, + .rta_tsage = tsage, + }; + + if (expires) + ci.rta_expires = jiffies_to_clock_t(expires); + + return nla_put(skb, RTA_CACHEINFO, sizeof(ci), &ci); +} + +EXPORT_SYMBOL_GPL(rtnl_put_cacheinfo); static void set_operstate(struct net_device *dev, unsigned char transition) { diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 4eb985236aee..9881933167bd 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1469,7 +1469,7 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, struct rtmsg *r; struct nlmsghdr *nlh; unsigned char *b = skb->tail; - struct rta_cacheinfo ci; + long expires; nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags); r = NLMSG_DATA(nlh); @@ -1502,16 +1502,10 @@ static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, RTA_PUT(skb, RTA_GATEWAY, 2, &rt->rt_gateway); if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0) goto rtattr_failure; - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - if (rt->u.dst.expires) - ci.rta_expires = jiffies_to_clock_t(rt->u.dst.expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_error = rt->u.dst.error; - ci.rta_id = ci.rta_ts = ci.rta_tsage = 0; - RTA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + expires = rt->u.dst.expires ? rt->u.dst.expires - jiffies : 0; + if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0, expires, + rt->u.dst.error) < 0) + goto rtattr_failure; if (rt->fl.iif) RTA_PUT(skb, RTA_IIF, sizeof(int), &rt->fl.iif); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ee00b6506ab4..9f3924c4905e 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2629,7 +2629,8 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, struct rtable *rt = (struct rtable*)skb->dst; struct rtmsg *r; struct nlmsghdr *nlh; - struct rta_cacheinfo ci; + long expires; + u32 id = 0, ts = 0, tsage = 0, error; nlh = nlmsg_put(skb, pid, seq, event, sizeof(*r), flags); if (nlh == NULL) @@ -2676,20 +2677,13 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0) goto nla_put_failure; - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - if (rt->u.dst.expires) - ci.rta_expires = jiffies_to_clock_t(rt->u.dst.expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_error = rt->u.dst.error; - ci.rta_id = ci.rta_ts = ci.rta_tsage = 0; + error = rt->u.dst.error; + expires = rt->u.dst.expires ? rt->u.dst.expires - jiffies : 0; if (rt->peer) { - ci.rta_id = rt->peer->ip_id_count; + id = rt->peer->ip_id_count; if (rt->peer->tcp_ts_stamp) { - ci.rta_ts = rt->peer->tcp_ts; - ci.rta_tsage = xtime.tv_sec - rt->peer->tcp_ts_stamp; + ts = rt->peer->tcp_ts; + tsage = xtime.tv_sec - rt->peer->tcp_ts_stamp; } } @@ -2708,7 +2702,7 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, } else { if (err == -EMSGSIZE) goto nla_put_failure; - ci.rta_error = err; + error = err; } } } else @@ -2716,7 +2710,9 @@ static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, NLA_PUT_U32(skb, RTA_IIF, rt->fl.iif); } - NLA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + if (rtnl_put_cacheinfo(skb, &rt->u.dst, id, ts, tsage, + expires, error) < 0) + goto nla_put_failure; return nlmsg_end(skb, nlh); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0bf17a3cf085..9f80518aacbd 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2027,7 +2027,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, { struct rtmsg *rtm; struct nlmsghdr *nlh; - struct rta_cacheinfo ci; + long expires; u32 table; if (prefix) { /* user wants prefix routes only */ @@ -2101,18 +2101,11 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, NLA_PUT_U32(skb, RTA_OIF, rt->rt6i_dev->ifindex); NLA_PUT_U32(skb, RTA_PRIORITY, rt->rt6i_metric); - ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse); - if (rt->rt6i_expires) - ci.rta_expires = jiffies_to_clock_t(rt->rt6i_expires - jiffies); - else - ci.rta_expires = 0; - ci.rta_used = rt->u.dst.__use; - ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt); - ci.rta_error = rt->u.dst.error; - ci.rta_id = 0; - ci.rta_ts = 0; - ci.rta_tsage = 0; - NLA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci); + + expires = rt->rt6i_expires ? rt->rt6i_expires - jiffies : 0; + if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0, + expires, rt->u.dst.error) < 0) + goto nla_put_failure; return nlmsg_end(skb, nlh); -- cgit v1.2.3 From d5c42c0ec4f7fd5a4e19e33a2d561758b67c55c8 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Mon, 27 Nov 2006 17:58:02 -0200 Subject: [NET]: Pack struct hh_cache [acme@newtoy net-2.6.20]$ pahole net/ipv4/tcp.o hh_cache /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/netdevice.h:190 */ struct hh_cache { struct hh_cache * hh_next; /* 0 4 */ atomic_t hh_refcnt; /* 4 4 */ __be16 hh_type; /* 8 2 */ /* XXX 2 bytes hole, try to pack */ int hh_len; /* 12 4 */ int (*hh_output)(); /* 16 4 */ rwlock_t hh_lock; /* 20 36 */ long unsigned int hh_data[24]; /* 56 96 */ }; /* size: 152, sum members: 150, holes: 1, sum holes: 2 */ [acme@newtoy net-2.6.20]$ find net -name "*.[ch]" | xargs grep 'hh_len.\+=' | sort -u net/atm/br2684.c: hh->hh_len = PADLEN + ETH_HLEN; net/ethernet/eth.c: hh->hh_len = ETH_HLEN; net/ipv4/ipconfig.c: int hh_len = LL_RESERVED_SPACE(dev); net/ipv4/ip_output.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv4/ip_output.c: int hh_len = LL_RESERVED_SPACE(dev); net/ipv4/netfilter.c: hh_len = (*pskb)->dst->dev->hard_header_len; net/ipv4/raw.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv6/ip6_output.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); net/ipv6/netfilter/ip6t_REJECT.c: hh_len = (dst->dev->hard_header_len + 15)&~15; net/ipv6/raw.c: hh_len = LL_RESERVED_SPACE(rt->u.dst.dev); [acme@newtoy net-2.6.20]$ [acme@newtoy net-2.6.20]$ find include -name "*.h" | xargs grep 'define ETH_HLEN' include/linux/if_ether.h:#define ETH_HLEN 14 /* Total octets in header. */ (((dev)->hard_header_len&~(HH_DATA_MOD - 1)) + HH_DATA_MOD) [acme@newtoy net-2.6.20]$ pahole net/ipv4/tcp.o net_device | grep hard_header_len short unsigned int hard_header_len; /* 106 2 */ [acme@newtoy net-2.6.20]$ So I think we're safe in turning hh_len an u16, end result: [acme@newtoy net-2.6.20]$ codiff -sV /tmp/tcp.o.before net/ipv4/tcp.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv4/tcp.c: struct hh_cache | -4 hh_len; from: int /* 12(0) 4(0) */ to: u16 /* 10(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index caa3c2593719..949eada46ce1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -196,7 +196,7 @@ struct hh_cache * NOTE: For VLANs, this will be the * encapuslated type. --BLG */ - int hh_len; /* length of header */ + u16 hh_len; /* length of header */ int (*hh_output)(struct sk_buff *skb); rwlock_t hh_lock; -- cgit v1.2.3 From 2ff52f282cf287d60e9eda1f3b5ec83e00a86130 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 28 Nov 2006 00:48:32 -0200 Subject: [TCP]: Change tcp_header_len member in tcp_sock to u16 With this we eliminate the last hole in struct tcp_sock. End result: [acme@newtoy net-2.6.20]$ codiff -sV /tmp/tcp.o.before net/ipv4/tcp.o /pub/scm/linux/kernel/git/acme/net-2.6.20/net/ipv4/tcp.c: struct tcp_sock | -4 tcp_header_len; from: int /* 1000(0) 4(0) */ to: u16 /* 1000(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ Now sizeof(tcp_sock) is just... [acme@newtoy net-2.6.20]$ pahole --sizes ../OUTPUT/qemu/net-2.6.20/net/ipv4/tcp.o | grep -w tcp_sock struct tcp_sock: 1500 0 1500 bytes ;-) Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tcp.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index dd61b172ac68..b42ff0efc2df 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -227,7 +227,8 @@ static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) struct tcp_sock { /* inet_connection_sock has to be the first member of tcp_sock */ struct inet_connection_sock inet_conn; - int tcp_header_len; /* Bytes of tcp header to send */ + u16 tcp_header_len; /* Bytes of tcp header to send */ + u16 xmit_size_goal; /* Goal for segmenting output packets */ /* * Header prediction flags @@ -268,8 +269,6 @@ struct tcp_sock { __u32 snd_wnd; /* The window we expect to receive */ __u32 max_window; /* Maximal window ever seen from peer */ __u32 mss_cache; /* Cached effective mss, not including SACKS */ - __u16 xmit_size_goal; /* Goal for segmenting output packets */ - /* XXX Two bytes hole, try to pack */ __u32 window_clamp; /* Maximal window to advertise */ __u32 rcv_ssthresh; /* Current window clamp */ -- cgit v1.2.3 From 3a137d2065571864be0301b9ebd72ddb01060997 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 28 Nov 2006 01:12:38 -0200 Subject: [TCP]: Renove the __ prefix on the struct tcp_sock members As this struct is not userland visible at all. Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tcp.h | 156 ++++++++++++++++++++++++++-------------------------- 1 file changed, 78 insertions(+), 78 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index b42ff0efc2df..3cc70d1a3504 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -185,17 +185,17 @@ struct tcp_sack_block_wire { }; struct tcp_sack_block { - __u32 start_seq; - __u32 end_seq; + u32 start_seq; + u32 end_seq; }; struct tcp_options_received { /* PAWS/RTTM data */ long ts_recent_stamp;/* Time we stored ts_recent (for aging) */ - __u32 ts_recent; /* Time stamp to echo next */ - __u32 rcv_tsval; /* Time stamp value */ - __u32 rcv_tsecr; /* Time stamp echo reply */ - __u16 saw_tstamp : 1, /* Saw TIMESTAMP on last packet */ + u32 ts_recent; /* Time stamp to echo next */ + u32 rcv_tsval; /* Time stamp value */ + u32 rcv_tsecr; /* Time stamp echo reply */ + u16 saw_tstamp : 1, /* Saw TIMESTAMP on last packet */ tstamp_ok : 1, /* TIMESTAMP seen on SYN packet */ dsack : 1, /* D-SACK is scheduled */ wscale_ok : 1, /* Wscale seen on SYN packet */ @@ -203,10 +203,10 @@ struct tcp_options_received { snd_wscale : 4, /* Window scaling received from sender */ rcv_wscale : 4; /* Window scaling to send to receiver */ /* SACKs data */ - __u8 eff_sacks; /* Size of SACK array to send with next packet */ - __u8 num_sacks; /* Number of SACK blocks */ - __u16 user_mss; /* mss requested by user in ioctl */ - __u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ + u8 eff_sacks; /* Size of SACK array to send with next packet */ + u8 num_sacks; /* Number of SACK blocks */ + u16 user_mss; /* mss requested by user in ioctl */ + u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ }; struct tcp_request_sock { @@ -215,8 +215,8 @@ struct tcp_request_sock { /* Only used by TCP MD5 Signature so far. */ struct tcp_request_sock_ops *af_specific; #endif - __u32 rcv_isn; - __u32 snt_isn; + u32 rcv_isn; + u32 snt_isn; }; static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) @@ -241,13 +241,13 @@ struct tcp_sock { * read the code and the spec side by side (and laugh ...) * See RFC793 and RFC1122. The RFC writes these in capitals. */ - __u32 rcv_nxt; /* What we want to receive next */ - __u32 snd_nxt; /* Next sequence we send */ + u32 rcv_nxt; /* What we want to receive next */ + u32 snd_nxt; /* Next sequence we send */ - __u32 snd_una; /* First byte we want an ack for */ - __u32 snd_sml; /* Last byte of the most recently transmitted small packet */ - __u32 rcv_tstamp; /* timestamp of last received ACK (for keepalives) */ - __u32 lsndtime; /* timestamp of last sent data packet (for restart window) */ + u32 snd_una; /* First byte we want an ack for */ + u32 snd_sml; /* Last byte of the most recently transmitted small packet */ + u32 rcv_tstamp; /* timestamp of last received ACK (for keepalives) */ + u32 lsndtime; /* timestamp of last sent data packet (for restart window) */ /* Data for direct copy to user */ struct { @@ -265,30 +265,30 @@ struct tcp_sock { #endif } ucopy; - __u32 snd_wl1; /* Sequence for window update */ - __u32 snd_wnd; /* The window we expect to receive */ - __u32 max_window; /* Maximal window ever seen from peer */ - __u32 mss_cache; /* Cached effective mss, not including SACKS */ + u32 snd_wl1; /* Sequence for window update */ + u32 snd_wnd; /* The window we expect to receive */ + u32 max_window; /* Maximal window ever seen from peer */ + u32 mss_cache; /* Cached effective mss, not including SACKS */ - __u32 window_clamp; /* Maximal window to advertise */ - __u32 rcv_ssthresh; /* Current window clamp */ + u32 window_clamp; /* Maximal window to advertise */ + u32 rcv_ssthresh; /* Current window clamp */ - __u32 frto_highmark; /* snd_nxt when RTO occurred */ - __u8 reordering; /* Packet reordering metric. */ - __u8 frto_counter; /* Number of new acks after RTO */ - __u8 nonagle; /* Disable Nagle algorithm? */ - __u8 keepalive_probes; /* num of allowed keep alive probes */ + u32 frto_highmark; /* snd_nxt when RTO occurred */ + u8 reordering; /* Packet reordering metric. */ + u8 frto_counter; /* Number of new acks after RTO */ + u8 nonagle; /* Disable Nagle algorithm? */ + u8 keepalive_probes; /* num of allowed keep alive probes */ /* RTT measurement */ - __u32 srtt; /* smoothed round trip time << 3 */ - __u32 mdev; /* medium deviation */ - __u32 mdev_max; /* maximal mdev for the last rtt period */ - __u32 rttvar; /* smoothed mdev_max */ - __u32 rtt_seq; /* sequence number to update rttvar */ - - __u32 packets_out; /* Packets which are "in flight" */ - __u32 left_out; /* Packets which leaved network */ - __u32 retrans_out; /* Retransmitted packets out */ + u32 srtt; /* smoothed round trip time << 3 */ + u32 mdev; /* medium deviation */ + u32 mdev_max; /* maximal mdev for the last rtt period */ + u32 rttvar; /* smoothed mdev_max */ + u32 rtt_seq; /* sequence number to update rttvar */ + + u32 packets_out; /* Packets which are "in flight" */ + u32 left_out; /* Packets which leaved network */ + u32 retrans_out; /* Retransmitted packets out */ /* * Options received (usually on last packet, some only on SYN packets). */ @@ -297,20 +297,20 @@ struct tcp_sock { /* * Slow start and congestion control (see also Nagle, and Karn & Partridge) */ - __u32 snd_ssthresh; /* Slow start size threshold */ - __u32 snd_cwnd; /* Sending congestion window */ - __u16 snd_cwnd_cnt; /* Linear increase counter */ - __u16 snd_cwnd_clamp; /* Do not allow snd_cwnd to grow above this */ - __u32 snd_cwnd_used; - __u32 snd_cwnd_stamp; + u32 snd_ssthresh; /* Slow start size threshold */ + u32 snd_cwnd; /* Sending congestion window */ + u16 snd_cwnd_cnt; /* Linear increase counter */ + u16 snd_cwnd_clamp; /* Do not allow snd_cwnd to grow above this */ + u32 snd_cwnd_used; + u32 snd_cwnd_stamp; struct sk_buff_head out_of_order_queue; /* Out of order segments go here */ - __u32 rcv_wnd; /* Current receiver window */ - __u32 rcv_wup; /* rcv_nxt on last window update sent */ - __u32 write_seq; /* Tail(+1) of data held in tcp send buffer */ - __u32 pushed_seq; /* Last pushed seq, required to talk to windows */ - __u32 copied_seq; /* Head of yet unread data */ + u32 rcv_wnd; /* Current receiver window */ + u32 rcv_wup; /* rcv_nxt on last window update sent */ + u32 write_seq; /* Tail(+1) of data held in tcp send buffer */ + u32 pushed_seq; /* Last pushed seq, required to talk to windows */ + u32 copied_seq; /* Head of yet unread data */ /* SACKs data */ struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */ @@ -331,26 +331,26 @@ struct tcp_sock { int retransmit_cnt_hint; int forward_cnt_hint; - __u16 advmss; /* Advertised MSS */ - __u16 prior_ssthresh; /* ssthresh saved at recovery start */ - __u32 lost_out; /* Lost packets */ - __u32 sacked_out; /* SACK'd packets */ - __u32 fackets_out; /* FACK'd packets */ - __u32 high_seq; /* snd_nxt at onset of congestion */ + u16 advmss; /* Advertised MSS */ + u16 prior_ssthresh; /* ssthresh saved at recovery start */ + u32 lost_out; /* Lost packets */ + u32 sacked_out; /* SACK'd packets */ + u32 fackets_out; /* FACK'd packets */ + u32 high_seq; /* snd_nxt at onset of congestion */ - __u32 retrans_stamp; /* Timestamp of the last retransmit, + u32 retrans_stamp; /* Timestamp of the last retransmit, * also used in SYN-SENT to remember stamp of * the first SYN. */ - __u32 undo_marker; /* tracking retrans started here. */ + u32 undo_marker; /* tracking retrans started here. */ int undo_retrans; /* number of undoable retransmissions. */ - __u32 urg_seq; /* Seq of received urgent pointer */ - __u16 urg_data; /* Saved octet of OOB data and control flags */ - __u8 urg_mode; /* In urgent mode */ - __u8 ecn_flags; /* ECN status bits. */ - __u32 snd_up; /* Urgent pointer */ + u32 urg_seq; /* Seq of received urgent pointer */ + u16 urg_data; /* Saved octet of OOB data and control flags */ + u8 urg_mode; /* In urgent mode */ + u8 ecn_flags; /* ECN status bits. */ + u32 snd_up; /* Urgent pointer */ - __u32 total_retrans; /* Total retransmits for entire connection */ - __u32 bytes_acked; /* Appropriate Byte Counting - RFC3465 */ + u32 total_retrans; /* Total retransmits for entire connection */ + u32 bytes_acked; /* Appropriate Byte Counting - RFC3465 */ unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ @@ -358,26 +358,26 @@ struct tcp_sock { unsigned long last_synq_overflow; - __u32 tso_deferred; + u32 tso_deferred; /* Receiver side RTT estimation */ struct { - __u32 rtt; - __u32 seq; - __u32 time; + u32 rtt; + u32 seq; + u32 time; } rcv_rtt_est; /* Receiver queue space */ struct { int space; - __u32 seq; - __u32 time; + u32 seq; + u32 time; } rcvq_space; /* TCP-specific MTU probe information. */ struct { - __u32 probe_seq_start; - __u32 probe_seq_end; + u32 probe_seq_start; + u32 probe_seq_end; } mtu_probe; #ifdef CONFIG_TCP_MD5SIG @@ -396,14 +396,14 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk) struct tcp_timewait_sock { struct inet_timewait_sock tw_sk; - __u32 tw_rcv_nxt; - __u32 tw_snd_nxt; - __u32 tw_rcv_wnd; - __u32 tw_ts_recent; + u32 tw_rcv_nxt; + u32 tw_snd_nxt; + u32 tw_rcv_wnd; + u32 tw_ts_recent; long tw_ts_recent_stamp; #ifdef CONFIG_TCP_MD5SIG - __u16 tw_md5_keylen; - __u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; + u16 tw_md5_keylen; + u8 tw_md5_key[TCP_MD5SIG_MAXKEYLEN]; #endif }; -- cgit v1.2.3 From 4384260443efe90a2ec0d907568dbc58ae792cd0 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 28 Nov 2006 18:14:10 -0200 Subject: [DCCP]: Remove allocation of sysctl numbers This is in response to a request sent earlier by Eric W. Biederman and replaces all sysctl numbers for net.dccp.default with CTL_UNNUMBERED. It has been tested to compile and to work. Commiter note: I've removed the use of CTL_UNNUMBERED, not setting .ctl_name sets it to 0, that is the what CTL_UNNUMBERED is, reason is to avoid unneeded source code cluttering. Signed-off-by: Gerrit Renker Signed-off-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/sysctl.h | 14 -------------- net/dccp/sysctl.c | 10 ---------- 2 files changed, 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 61dd99c6c2a5..94316a98e0d0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -606,20 +606,6 @@ enum { NET_DCCP_DEFAULT=1, }; -/* /proc/sys/net/dccp/default */ -enum { - NET_DCCP_DEFAULT_SEQ_WINDOW = 1, - NET_DCCP_DEFAULT_RX_CCID = 2, - NET_DCCP_DEFAULT_TX_CCID = 3, - NET_DCCP_DEFAULT_ACK_RATIO = 4, - NET_DCCP_DEFAULT_SEND_ACKVEC = 5, - NET_DCCP_DEFAULT_SEND_NDP = 6, - NET_DCCP_DEFAULT_REQ_RETRIES = 7, - NET_DCCP_DEFAULT_RETRIES1 = 8, - NET_DCCP_DEFAULT_RETRIES2 = 9, - NET_DCCP_DEFAULT_TX_QLEN = 10, -}; - /* /proc/sys/net/ipx */ enum { NET_IPX_PPROP_BROADCASTING=1, diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 4775ba3faa04..fdcfca3e9208 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -20,7 +20,6 @@ static struct ctl_table dccp_default_table[] = { { - .ctl_name = NET_DCCP_DEFAULT_SEQ_WINDOW, .procname = "seq_window", .data = &sysctl_dccp_feat_sequence_window, .maxlen = sizeof(sysctl_dccp_feat_sequence_window), @@ -28,7 +27,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RX_CCID, .procname = "rx_ccid", .data = &sysctl_dccp_feat_rx_ccid, .maxlen = sizeof(sysctl_dccp_feat_rx_ccid), @@ -36,7 +34,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_CCID, .procname = "tx_ccid", .data = &sysctl_dccp_feat_tx_ccid, .maxlen = sizeof(sysctl_dccp_feat_tx_ccid), @@ -44,7 +41,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_ACK_RATIO, .procname = "ack_ratio", .data = &sysctl_dccp_feat_ack_ratio, .maxlen = sizeof(sysctl_dccp_feat_ack_ratio), @@ -52,7 +48,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_ACKVEC, .procname = "send_ackvec", .data = &sysctl_dccp_feat_send_ack_vector, .maxlen = sizeof(sysctl_dccp_feat_send_ack_vector), @@ -60,7 +55,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_SEND_NDP, .procname = "send_ndp", .data = &sysctl_dccp_feat_send_ndp_count, .maxlen = sizeof(sysctl_dccp_feat_send_ndp_count), @@ -68,7 +62,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_REQ_RETRIES, .procname = "request_retries", .data = &sysctl_dccp_request_retries, .maxlen = sizeof(sysctl_dccp_request_retries), @@ -76,7 +69,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES1, .procname = "retries1", .data = &sysctl_dccp_retries1, .maxlen = sizeof(sysctl_dccp_retries1), @@ -84,7 +76,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_RETRIES2, .procname = "retries2", .data = &sysctl_dccp_retries2, .maxlen = sizeof(sysctl_dccp_retries2), @@ -92,7 +83,6 @@ static struct ctl_table dccp_default_table[] = { .proc_handler = proc_dointvec, }, { - .ctl_name = NET_DCCP_DEFAULT_TX_QLEN, .procname = "tx_qlen", .data = &sysctl_dccp_tx_qlen, .maxlen = sizeof(sysctl_dccp_tx_qlen), -- cgit v1.2.3 From 5aed324369c94a2c38469c8288e42eb1a9fac400 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Tue, 28 Nov 2006 19:33:36 -0200 Subject: [DCCP]: Tidy up unused structures This removes and cleans up unused variables and structures which have become unnecessary following the introduction of the EWMA patch to automatically track the CCID 3 receiver/sender packet sizes `s'. It deprecates the PACKET_SIZE socket option by returning an error code and printing a deprecation warning if an application tries to read or write this socket option. Signed-off-by: Gerrit Renker Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 4 +--- net/dccp/ccids/ccid3.h | 4 ---- net/dccp/proto.c | 8 ++++---- 3 files changed, 5 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index 72cc355d7a03..ed6cc8962d87 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -200,7 +200,7 @@ struct dccp_so_feat { }; /* DCCP socket options */ -#define DCCP_SOCKOPT_PACKET_SIZE 1 +#define DCCP_SOCKOPT_PACKET_SIZE 1 /* XXX deprecated, without effect */ #define DCCP_SOCKOPT_SERVICE 2 #define DCCP_SOCKOPT_CHANGE_L 3 #define DCCP_SOCKOPT_CHANGE_R 4 @@ -460,7 +460,6 @@ struct dccp_ackvec; * @dccps_service_list - second .. last service code on passive socket * @dccps_timestamp_time - time of latest TIMESTAMP option * @dccps_timestamp_echo - latest timestamp received on a TIMESTAMP option - * @dccps_packet_size - Set thru setsockopt * @dccps_l_ack_ratio - * @dccps_r_ack_ratio - * @dccps_pcslen - sender partial checksum coverage (via sockopt) @@ -495,7 +494,6 @@ struct dccp_sock { struct dccp_service_list *dccps_service_list; struct timeval dccps_timestamp_time; __u32 dccps_timestamp_echo; - __u32 dccps_packet_size; __u16 dccps_l_ack_ratio; __u16 dccps_r_ack_ratio; __u16 dccps_pcslen; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 970921700ce3..dbb884426dfa 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -42,10 +42,6 @@ #include #include "../ccid.h" -#define TFRC_MIN_PACKET_SIZE 16 -#define TFRC_STD_PACKET_SIZE 256 -#define TFRC_MAX_PACKET_SIZE 65535 - /* Two seconds as per RFC 3448 4.2 */ #define TFRC_INITIAL_TIMEOUT (2 * USEC_PER_SEC) diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 3c44d502e5c1..2604e34d8f38 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -470,7 +470,8 @@ static int do_dccp_setsockopt(struct sock *sk, int level, int optname, lock_sock(sk); switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: - dp->dccps_packet_size = val; + DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n"); + err = -EINVAL; break; case DCCP_SOCKOPT_CHANGE_L: if (optlen != sizeof(struct dccp_so_feat)) @@ -581,9 +582,8 @@ static int do_dccp_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case DCCP_SOCKOPT_PACKET_SIZE: - val = dp->dccps_packet_size; - len = sizeof(dp->dccps_packet_size); - break; + DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n"); + return -EINVAL; case DCCP_SOCKOPT_SERVICE: return dccp_getsockopt_service(sk, len, (__be32 __user *)optval, optlen); -- cgit v1.2.3 From d62f9ed4a490309bd9e5df0b42ba5d096e7b5902 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:17 +0100 Subject: [NETFILTER]: nf_conntrack: automatic sysctl registation for conntrack protocols Add helper functions for sysctl registration with optional instantiating of common path elements (like net/netfilter) and use it for support for automatic registation of conntrack protocol sysctls. Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 10 ++ include/net/netfilter/nf_conntrack_l3proto.h | 6 ++ include/net/netfilter/nf_conntrack_l4proto.h | 6 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_proto.c | 102 ++++++++++++++++++++ net/netfilter/nf_sysctl.c | 134 +++++++++++++++++++++++++++ 6 files changed, 259 insertions(+) create mode 100644 net/netfilter/nf_sysctl.c (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 6ab5e2d6133e..f6f3fcbd70ed 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -117,6 +117,16 @@ void nf_unregister_hooks(struct nf_hook_ops *reg, unsigned int n); int nf_register_sockopt(struct nf_sockopt_ops *reg); void nf_unregister_sockopt(struct nf_sockopt_ops *reg); +#ifdef CONFIG_SYSCTL +/* Sysctl registration */ +struct ctl_table_header *nf_register_sysctl_table(struct ctl_table *path, + struct ctl_table *table); +void nf_unregister_sysctl_table(struct ctl_table_header *header, + struct ctl_table *table); +extern struct ctl_table nf_net_netfilter_sysctl_path[]; +extern struct ctl_table nf_net_ipv4_netfilter_sysctl_path[]; +#endif /* CONFIG_SYSCTL */ + extern struct list_head nf_hooks[NPROTO][NF_MAX_HOOKS]; /* those NF_LOG_* defines and struct nf_loginfo are legacy definitios that will diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index 6364df059422..664ddcffe00d 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -75,6 +75,12 @@ struct nf_conntrack_l3proto int (*nfattr_to_tuple)(struct nfattr *tb[], struct nf_conntrack_tuple *t); +#ifdef CONFIG_SYSCTL + struct ctl_table_header *ctl_table_header; + struct ctl_table *ctl_table_path; + struct ctl_table *ctl_table; +#endif /* CONFIG_SYSCTL */ + /* Module (if any) which this is connected to. */ struct module *me; }; diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index c22804aa227e..fe1e8fa30d2f 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -76,6 +76,12 @@ struct nf_conntrack_l4proto int (*nfattr_to_tuple)(struct nfattr *tb[], struct nf_conntrack_tuple *t); +#ifdef CONFIG_SYSCTL + struct ctl_table_header **ctl_table_header; + struct ctl_table *ctl_table; + unsigned int *ctl_table_users; +#endif /* CONFIG_SYSCTL */ + /* Module (if any) which this is connected to. */ struct module *me; }; diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 627105df1040..84d529ded952 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -4,6 +4,7 @@ nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_exp nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o obj-$(CONFIG_NETFILTER) = netfilter.o +obj-$(CONFIG_SYSCTL) += nf_sysctl.o obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_QUEUE) += nfnetlink_queue.o diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index a6a3b1ddd00d..941b5c3754af 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -30,6 +31,34 @@ struct nf_conntrack_l4proto **nf_ct_protos[PF_MAX] __read_mostly; struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX] __read_mostly; +#ifdef CONFIG_SYSCTL +static DEFINE_MUTEX(nf_ct_proto_sysctl_mutex); + +static int +nf_ct_register_sysctl(struct ctl_table_header **header, struct ctl_table *path, + struct ctl_table *table, unsigned int *users) +{ + if (*header == NULL) { + *header = nf_register_sysctl_table(path, table); + if (*header == NULL) + return -ENOMEM; + } + if (users != NULL) + (*users)++; + return 0; +} + +static void +nf_ct_unregister_sysctl(struct ctl_table_header **header, + struct ctl_table *table, unsigned int *users) +{ + if (users != NULL && --*users > 0) + return; + nf_unregister_sysctl_table(*header, table); + *header = NULL; +} +#endif + struct nf_conntrack_l4proto * __nf_ct_l4proto_find(u_int16_t l3proto, u_int8_t l4proto) { @@ -124,6 +153,33 @@ static int kill_l4proto(struct nf_conn *i, void *data) l4proto->l3proto); } +static int nf_ct_l3proto_register_sysctl(struct nf_conntrack_l3proto *l3proto) +{ + int err = 0; + +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l3proto->ctl_table != NULL) { + err = nf_ct_register_sysctl(&l3proto->ctl_table_header, + l3proto->ctl_table_path, + l3proto->ctl_table, NULL); + } + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif + return err; +} + +static void nf_ct_l3proto_unregister_sysctl(struct nf_conntrack_l3proto *l3proto) +{ +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l3proto->ctl_table_header != NULL) + nf_ct_unregister_sysctl(&l3proto->ctl_table_header, + l3proto->ctl_table, NULL); + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif +} + int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) { int ret = 0; @@ -139,6 +195,12 @@ int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) goto out_unlock; } nf_ct_l3protos[proto->l3proto] = proto; + write_unlock_bh(&nf_conntrack_lock); + + ret = nf_ct_l3proto_register_sysctl(proto); + if (ret < 0) + nf_conntrack_l3proto_unregister(proto); + return ret; out_unlock: write_unlock_bh(&nf_conntrack_lock); @@ -165,6 +227,8 @@ int nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) nf_ct_l3protos[proto->l3proto] = &nf_conntrack_l3proto_generic; write_unlock_bh(&nf_conntrack_lock); + nf_ct_l3proto_unregister_sysctl(proto); + /* Somebody could be still looking at the proto in bh. */ synchronize_net(); @@ -175,6 +239,36 @@ out: return ret; } +static int nf_ct_l4proto_register_sysctl(struct nf_conntrack_l4proto *l4proto) +{ + int err = 0; + +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l4proto->ctl_table != NULL) { + err = nf_ct_register_sysctl(l4proto->ctl_table_header, + nf_net_netfilter_sysctl_path, + l4proto->ctl_table, + l4proto->ctl_table_users); + } + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif + return err; +} + +static void nf_ct_l4proto_unregister_sysctl(struct nf_conntrack_l4proto *l4proto) +{ +#ifdef CONFIG_SYSCTL + mutex_lock(&nf_ct_proto_sysctl_mutex); + if (l4proto->ctl_table_header != NULL && + *l4proto->ctl_table_header != NULL) + nf_ct_unregister_sysctl(l4proto->ctl_table_header, + l4proto->ctl_table, + l4proto->ctl_table_users); + mutex_unlock(&nf_ct_proto_sysctl_mutex); +#endif +} + /* FIXME: Allow NULL functions and sub in pointers to generic for them. --RR */ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) @@ -230,6 +324,12 @@ retry: } nf_ct_protos[l4proto->l3proto][l4proto->l4proto] = l4proto; + write_unlock_bh(&nf_conntrack_lock); + + ret = nf_ct_l4proto_register_sysctl(l4proto); + if (ret < 0) + nf_conntrack_l4proto_unregister(l4proto); + return ret; out_unlock: write_unlock_bh(&nf_conntrack_lock); @@ -257,6 +357,8 @@ int nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) = &nf_conntrack_l4proto_generic; write_unlock_bh(&nf_conntrack_lock); + nf_ct_l4proto_unregister_sysctl(l4proto); + /* Somebody could be still looking at the proto in bh. */ synchronize_net(); diff --git a/net/netfilter/nf_sysctl.c b/net/netfilter/nf_sysctl.c new file mode 100644 index 000000000000..06ddddb2911f --- /dev/null +++ b/net/netfilter/nf_sysctl.c @@ -0,0 +1,134 @@ +/* nf_sysctl.c netfilter sysctl registration/unregistation + * + * Copyright (c) 2006 Patrick McHardy + */ +#include +#include +#include +#include + +static void +path_free(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table *t, *next; + + for (t = path; t != NULL && t != table; t = next) { + next = t->child; + kfree(t); + } +} + +static struct ctl_table * +path_dup(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table *t, *last = NULL, *tmp; + + for (t = path; t != NULL; t = t->child) { + /* twice the size since path elements are terminated by an + * empty element */ + tmp = kmemdup(t, 2 * sizeof(*t), GFP_KERNEL); + if (tmp == NULL) { + if (last != NULL) + path_free(path, table); + return NULL; + } + + if (last != NULL) + last->child = tmp; + else + path = tmp; + last = tmp; + } + + if (last != NULL) + last->child = table; + else + path = table; + + return path; +} + +struct ctl_table_header * +nf_register_sysctl_table(struct ctl_table *path, struct ctl_table *table) +{ + struct ctl_table_header *header; + + path = path_dup(path, table); + if (path == NULL) + return NULL; + header = register_sysctl_table(path, 0); + if (header == NULL) + path_free(path, table); + return header; +} +EXPORT_SYMBOL_GPL(nf_register_sysctl_table); + +void +nf_unregister_sysctl_table(struct ctl_table_header *header, + struct ctl_table *table) +{ + struct ctl_table *path = header->ctl_table; + + unregister_sysctl_table(header); + path_free(path, table); +} +EXPORT_SYMBOL_GPL(nf_unregister_sysctl_table); + +/* net/netfilter */ +static struct ctl_table nf_net_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + }, + { + .ctl_name = 0 + } +}; +struct ctl_table nf_net_netfilter_sysctl_path[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_net_netfilter_table, + }, + { + .ctl_name = 0 + } +}; +EXPORT_SYMBOL_GPL(nf_net_netfilter_sysctl_path); + +/* net/ipv4/netfilter */ +static struct ctl_table nf_net_ipv4_netfilter_table[] = { + { + .ctl_name = NET_IPV4_NETFILTER, + .procname = "netfilter", + .mode = 0555, + }, + { + .ctl_name = 0 + } +}; +static struct ctl_table nf_net_ipv4_table[] = { + { + .ctl_name = NET_IPV4, + .procname = "ipv4", + .mode = 0555, + .child = nf_net_ipv4_netfilter_table, + }, + { + .ctl_name = 0 + } +}; +struct ctl_table nf_net_ipv4_netfilter_sysctl_path[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_net_ipv4_table, + }, + { + .ctl_name = 0 + } +}; +EXPORT_SYMBOL_GPL(nf_net_ipv4_netfilter_sysctl_path); -- cgit v1.2.3 From 468ec44bd5a863736d955f78b8c38896f26864a1 Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Wed, 29 Nov 2006 02:35:23 +0100 Subject: [NETFILTER]: conntrack: add '_get' to {ip, nf}_conntrack_expect_find We usually uses 'xxx_find_get' for function which increments reference count. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack.h | 2 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- net/ipv4/netfilter/ip_conntrack_core.c | 2 +- net/ipv4/netfilter/ip_conntrack_helper_pptp.c | 4 ++-- net/ipv4/netfilter/ip_conntrack_netlink.c | 4 ++-- net/ipv4/netfilter/ip_conntrack_standalone.c | 2 +- net/ipv4/netfilter/ip_nat_helper_pptp.c | 2 +- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 4 ++-- net/netfilter/nf_conntrack_standalone.c | 2 +- 10 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 64e868034c4a..61da56941dce 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -277,7 +277,7 @@ extern struct ip_conntrack_expect * __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_expect * -ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple); +ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple); extern struct ip_conntrack_tuple_hash * __ip_conntrack_find(const struct ip_conntrack_tuple *tuple, diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 5aa483e03455..2d335f024c85 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -53,7 +53,7 @@ struct nf_conntrack_expect * __nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); +nf_conntrack_expect_find_get(const struct nf_conntrack_tuple *tuple); struct nf_conntrack_expect * find_expectation(const struct nf_conntrack_tuple *tuple); diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 8b848aa77bfc..4d1f954d459b 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -233,7 +233,7 @@ __ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) /* Just find a expectation corresponding to a tuple. */ struct ip_conntrack_expect * -ip_conntrack_expect_find(const struct ip_conntrack_tuple *tuple) +ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple) { struct ip_conntrack_expect *i; diff --git a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c index a2af5e0c7f99..a5c057bcecf4 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_pptp.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_pptp.c @@ -142,7 +142,7 @@ static void pptp_expectfn(struct ip_conntrack *ct, DEBUGP("trying to unexpect other dir: "); DUMP_TUPLE(&inv_t); - exp_other = ip_conntrack_expect_find(&inv_t); + exp_other = ip_conntrack_expect_find_get(&inv_t); if (exp_other) { /* delete other expectation. */ DEBUGP("found\n"); @@ -176,7 +176,7 @@ static int destroy_sibling_or_exp(const struct ip_conntrack_tuple *t) ip_conntrack_put(sibling); return 1; } else { - exp = ip_conntrack_expect_find(t); + exp = ip_conntrack_expect_find_get(t); if (exp) { DEBUGP("unexpect_related of expect %p\n", exp); ip_conntrack_unexpect_related(exp); diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 1bb8ed33c5bc..3d277aa869dd 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -1256,7 +1256,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = ip_conntrack_expect_find(&tuple); + exp = ip_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; @@ -1309,7 +1309,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = ip_conntrack_expect_find(&tuple); + exp = ip_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index 02135756562e..2df67538ffb0 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -926,7 +926,7 @@ EXPORT_SYMBOL(__ip_ct_refresh_acct); EXPORT_SYMBOL(ip_conntrack_expect_alloc); EXPORT_SYMBOL(ip_conntrack_expect_put); EXPORT_SYMBOL_GPL(__ip_conntrack_expect_find); -EXPORT_SYMBOL_GPL(ip_conntrack_expect_find); +EXPORT_SYMBOL_GPL(ip_conntrack_expect_find_get); EXPORT_SYMBOL(ip_conntrack_expect_related); EXPORT_SYMBOL(ip_conntrack_unexpect_related); EXPORT_SYMBOL_GPL(ip_conntrack_expect_list); diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c index 329fdcd7d702..acf55d863100 100644 --- a/net/ipv4/netfilter/ip_nat_helper_pptp.c +++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c @@ -101,7 +101,7 @@ static void pptp_nat_expected(struct ip_conntrack *ct, DEBUGP("trying to unexpect other dir: "); DUMP_TUPLE(&t); - other_exp = ip_conntrack_expect_find(&t); + other_exp = ip_conntrack_expect_find_get(&t); if (other_exp) { ip_conntrack_unexpect_related(other_exp); ip_conntrack_expect_put(other_exp); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 79cfd79a42f0..aa5903e4da11 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -68,7 +68,7 @@ __nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) /* Just find a expectation corresponding to a tuple. */ struct nf_conntrack_expect * -nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +nf_conntrack_expect_find_get(const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index dc0830220130..7357b8f47acd 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1284,7 +1284,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = nf_conntrack_expect_find(&tuple); + exp = nf_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; @@ -1339,7 +1339,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = nf_conntrack_expect_find(&tuple); + exp = nf_conntrack_expect_find_get(&tuple); if (!exp) return -ENOENT; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 3db24f4f0f5b..be94b6359725 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -549,7 +549,7 @@ EXPORT_SYMBOL(__nf_conntrack_find); EXPORT_SYMBOL(nf_ct_unlink_expect); EXPORT_SYMBOL(nf_conntrack_hash_insert); EXPORT_SYMBOL(__nf_conntrack_expect_find); -EXPORT_SYMBOL(nf_conntrack_expect_find); +EXPORT_SYMBOL(nf_conntrack_expect_find_get); EXPORT_SYMBOL(nf_conntrack_expect_list); #if defined(CONFIG_NF_CT_NETLINK) || \ defined(CONFIG_NF_CT_NETLINK_MODULE) -- cgit v1.2.3 From 9d5b8baa4e9ace9be113c6151aaeeb3c07a26fc8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:26 +0100 Subject: [NETFILTER]: sip conntrack: minor cleanup - Use enum for header field enumeration - Use numerical value instead of pointer to header info structure to identify headers, unexport ct_sip_hdrs - group SIP and SDP entries in header info structure - remove double forward declaration of ct_sip_get_info Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 28 +++++------- net/ipv4/netfilter/ip_conntrack_sip.c | 60 +++++++++++++------------ net/ipv4/netfilter/ip_nat_sip.c | 30 ++++++------- 3 files changed, 55 insertions(+), 63 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 913dad66c0fb..2a15eb51fd6b 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -5,23 +5,15 @@ #define SIP_PORT 5060 #define SIP_TIMEOUT 3600 -#define POS_VIA 0 -#define POS_CONTACT 1 -#define POS_CONTENT 2 -#define POS_MEDIA 3 -#define POS_OWNER 4 -#define POS_CONNECTION 5 -#define POS_REQ_HEADER 6 -#define POS_SDP_HEADER 7 - -struct sip_header_nfo { - const char *lname; - const char *sname; - const char *ln_str; - size_t lnlen; - size_t snlen; - size_t ln_strlen; - int (*match_len)(const char *, const char *, int *); +enum sip_header_pos { + POS_REQ_HEADER, + POS_VIA, + POS_CONTACT, + POS_CONTENT, + POS_MEDIA, + POS_OWNER, + POS_CONNECTION, + POS_SDP_HEADER, }; extern unsigned int (*ip_nat_sip_hook)(struct sk_buff **pskb, @@ -36,7 +28,7 @@ extern unsigned int (*ip_nat_sdp_hook)(struct sk_buff **pskb, extern int ct_sip_get_info(const char *dptr, size_t dlen, unsigned int *matchoff, unsigned int *matchlen, - struct sip_header_nfo *hnfo); + enum sip_header_pos pos); extern int ct_sip_lnlen(const char *line, const char *limit); extern const char *ct_sip_search(const char *needle, const char *haystack, size_t needle_len, size_t haystack_len); diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index bf423db4ed46..cc3176581667 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -52,20 +52,32 @@ unsigned int (*ip_nat_sdp_hook)(struct sk_buff **pskb, const char *dptr); EXPORT_SYMBOL_GPL(ip_nat_sdp_hook); -int ct_sip_get_info(const char *dptr, size_t dlen, - unsigned int *matchoff, - unsigned int *matchlen, - struct sip_header_nfo *hnfo); -EXPORT_SYMBOL_GPL(ct_sip_get_info); - - static int digits_len(const char *dptr, const char *limit, int *shift); static int epaddr_len(const char *dptr, const char *limit, int *shift); static int skp_digits_len(const char *dptr, const char *limit, int *shift); static int skp_epaddr_len(const char *dptr, const char *limit, int *shift); -struct sip_header_nfo ct_sip_hdrs[] = { - { /* Via header */ +struct sip_header_nfo { + const char *lname; + const char *sname; + const char *ln_str; + size_t lnlen; + size_t snlen; + size_t ln_strlen; + int (*match_len)(const char *, const char *, int *); +}; + +static struct sip_header_nfo ct_sip_hdrs[] = { + [POS_REQ_HEADER] = { /* SIP Requests headers */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .sname = "sip:", + .snlen = sizeof("sip:") - 1, /* yes, i know.. ;) */ + .ln_str = "@", + .ln_strlen = sizeof("@") - 1, + .match_len = epaddr_len + }, + [POS_VIA] = { /* SIP Via header */ .lname = "Via:", .lnlen = sizeof("Via:") - 1, .sname = "\r\nv:", @@ -74,7 +86,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("UDP ") - 1, .match_len = epaddr_len, }, - { /* Contact header */ + [POS_CONTACT] = { /* SIP Contact header */ .lname = "Contact:", .lnlen = sizeof("Contact:") - 1, .sname = "\r\nm:", @@ -83,7 +95,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("sip:") - 1, .match_len = skp_epaddr_len }, - { /* Content length header */ + [POS_CONTENT] = { /* SIP Content length header */ .lname = "Content-Length:", .lnlen = sizeof("Content-Length:") - 1, .sname = "\r\nl:", @@ -92,7 +104,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof(":") - 1, .match_len = skp_digits_len }, - { /* SDP media info */ + [POS_MEDIA] = { /* SDP media info */ .lname = "\nm=", .lnlen = sizeof("\nm=") - 1, .sname = "\rm=", @@ -101,7 +113,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("audio ") - 1, .match_len = digits_len }, - { /* SDP owner address*/ + [POS_OWNER] = { /* SDP owner address*/ .lname = "\no=", .lnlen = sizeof("\no=") - 1, .sname = "\ro=", @@ -110,7 +122,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("IN IP4 ") - 1, .match_len = epaddr_len }, - { /* SDP connection info */ + [POS_CONNECTION] = { /* SDP connection info */ .lname = "\nc=", .lnlen = sizeof("\nc=") - 1, .sname = "\rc=", @@ -119,16 +131,7 @@ struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof("IN IP4 ") - 1, .match_len = epaddr_len }, - { /* Requests headers */ - .lname = "sip:", - .lnlen = sizeof("sip:") - 1, - .sname = "sip:", - .snlen = sizeof("sip:") - 1, /* yes, i know.. ;) */ - .ln_str = "@", - .ln_strlen = sizeof("@") - 1, - .match_len = epaddr_len - }, - { /* SDP version header */ + [POS_SDP_HEADER] = { /* SDP version header */ .lname = "\nv=", .lnlen = sizeof("\nv=") - 1, .sname = "\rv=", @@ -138,7 +141,6 @@ struct sip_header_nfo ct_sip_hdrs[] = { .match_len = digits_len } }; -EXPORT_SYMBOL_GPL(ct_sip_hdrs); /* get line lenght until first CR or LF seen. */ int ct_sip_lnlen(const char *line, const char *limit) @@ -263,8 +265,9 @@ static int skp_epaddr_len(const char *dptr, const char *limit, int *shift) int ct_sip_get_info(const char *dptr, size_t dlen, unsigned int *matchoff, unsigned int *matchlen, - struct sip_header_nfo *hnfo) + enum sip_header_pos pos) { + struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos]; const char *limit, *aux, *k = dptr; int shift = 0; @@ -298,6 +301,7 @@ int ct_sip_get_info(const char *dptr, size_t dlen, DEBUGP("%s header not found.\n", hnfo->lname); return 0; } +EXPORT_SYMBOL_GPL(ct_sip_get_info); static int set_expected_rtp(struct sk_buff **pskb, struct ip_conntrack *ct, @@ -393,7 +397,7 @@ static int sip_help(struct sk_buff **pskb, } /* Get ip and port address from SDP packet. */ if (ct_sip_get_info(dptr, datalen, &matchoff, &matchlen, - &ct_sip_hdrs[POS_CONNECTION]) > 0) { + POS_CONNECTION) > 0) { /* We'll drop only if there are parse problems. */ if (parse_ipaddr(dptr + matchoff, NULL, &ipaddr, @@ -402,7 +406,7 @@ static int sip_help(struct sk_buff **pskb, goto out; } if (ct_sip_get_info(dptr, datalen, &matchoff, &matchlen, - &ct_sip_hdrs[POS_MEDIA]) > 0) { + POS_MEDIA) > 0) { port = simple_strtoul(dptr + matchoff, NULL, 10); if (port < 1024) { diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index 9fad98099532..47097aac63d8 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -29,18 +29,16 @@ MODULE_DESCRIPTION("SIP NAT helper"); #define DEBUGP(format, args...) #endif -extern struct sip_header_nfo ct_sip_hdrs[]; - static unsigned int mangle_sip_packet(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, struct ip_conntrack *ct, const char **dptr, size_t dlen, char *buffer, int bufflen, - struct sip_header_nfo *hnfo) + enum sip_header_pos pos) { unsigned int matchlen, matchoff; - if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, hnfo) <= 0) + if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) return 0; if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, @@ -80,14 +78,13 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, if ((ctinfo) < IP_CT_IS_REPLY) { mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, - &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); return 1; } if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_VIA])) + buffer, bufflen, POS_VIA)) return 0; /* This search should ignore case, but later.. */ @@ -102,25 +99,24 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, return mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, - &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); } if ((ctinfo) < IP_CT_IS_REPLY) { if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_VIA])) + buffer, bufflen, POS_VIA)) return 0; /* Mangle Contact if exists only. - watch udp_nat_mangle()! */ mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_CONTACT]); + buffer, bufflen, POS_CONTACT); return 1; } /* This mangle requests headers. */ return mangle_sip_packet(pskb, ctinfo, ct, dptr, ct_sip_lnlen(*dptr, *dptr + (*pskb)->len - dataoff), - buffer, bufflen, &ct_sip_hdrs[POS_REQ_HEADER]); + buffer, bufflen, POS_REQ_HEADER); } static int mangle_content_len(struct sk_buff **pskb, @@ -136,7 +132,7 @@ static int mangle_content_len(struct sk_buff **pskb, /* Get actual SDP lenght */ if (ct_sip_get_info(dptr, (*pskb)->len - dataoff, &matchoff, - &matchlen, &ct_sip_hdrs[POS_SDP_HEADER]) > 0) { + &matchlen, POS_SDP_HEADER) > 0) { /* since ct_sip_get_info() give us a pointer passing 'v=' we need to add 2 bytes in this count. */ @@ -144,7 +140,7 @@ static int mangle_content_len(struct sk_buff **pskb, /* Now, update SDP lenght */ if (ct_sip_get_info(dptr, (*pskb)->len - dataoff, &matchoff, - &matchlen, &ct_sip_hdrs[POS_CONTENT]) > 0) { + &matchlen, POS_CONTENT) > 0) { bufflen = sprintf(buffer, "%u", c_len); @@ -170,17 +166,17 @@ static unsigned int mangle_sdp(struct sk_buff **pskb, /* Mangle owner and contact info. */ bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_OWNER])) + buffer, bufflen, POS_OWNER)) return 0; if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_CONNECTION])) + buffer, bufflen, POS_CONNECTION)) return 0; /* Mangle media port. */ bufflen = sprintf(buffer, "%u", port); if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, - buffer, bufflen, &ct_sip_hdrs[POS_MEDIA])) + buffer, bufflen, POS_MEDIA)) return 0; return mangle_content_len(pskb, ctinfo, ct, dptr); -- cgit v1.2.3 From 40883e8184947879f135605a05c0764c60656cc5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:27 +0100 Subject: [NETFILTER]: sip conntrack: do case insensitive SIP header search SIP headers are generally case-insensitive, only SDP headers are case sensitive. Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 3 ++- net/ipv4/netfilter/ip_conntrack_sip.c | 20 ++++++++++++++++---- net/ipv4/netfilter/ip_nat_sip.c | 7 ++++--- 3 files changed, 22 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 2a15eb51fd6b..51c65ac18c57 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -31,6 +31,7 @@ extern int ct_sip_get_info(const char *dptr, size_t dlen, enum sip_header_pos pos); extern int ct_sip_lnlen(const char *line, const char *limit); extern const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len); + size_t needle_len, size_t haystack_len, + int case_sensitive); #endif /* __KERNEL__ */ #endif /* __IP_CONNTRACK_SIP_H__ */ diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index cc3176581667..a9c0d1d3e56f 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -64,6 +64,7 @@ struct sip_header_nfo { size_t lnlen; size_t snlen; size_t ln_strlen; + int case_sensitive; int (*match_len)(const char *, const char *, int *); }; @@ -105,6 +106,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = skp_digits_len }, [POS_MEDIA] = { /* SDP media info */ + .case_sensitive = 1, .lname = "\nm=", .lnlen = sizeof("\nm=") - 1, .sname = "\rm=", @@ -114,6 +116,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = digits_len }, [POS_OWNER] = { /* SDP owner address*/ + .case_sensitive = 1, .lname = "\no=", .lnlen = sizeof("\no=") - 1, .sname = "\ro=", @@ -123,6 +126,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = epaddr_len }, [POS_CONNECTION] = { /* SDP connection info */ + .case_sensitive = 1, .lname = "\nc=", .lnlen = sizeof("\nc=") - 1, .sname = "\rc=", @@ -132,6 +136,7 @@ static struct sip_header_nfo ct_sip_hdrs[] = { .match_len = epaddr_len }, [POS_SDP_HEADER] = { /* SDP version header */ + .case_sensitive = 1, .lname = "\nv=", .lnlen = sizeof("\nv=") - 1, .sname = "\rv=", @@ -161,13 +166,19 @@ EXPORT_SYMBOL_GPL(ct_sip_lnlen); /* Linear string search, case sensitive. */ const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len) + size_t needle_len, size_t haystack_len, + int case_sensitive) { const char *limit = haystack + (haystack_len - needle_len); while (haystack <= limit) { - if (memcmp(haystack, needle, needle_len) == 0) - return haystack; + if (case_sensitive) { + if (strncmp(haystack, needle, needle_len) == 0) + return haystack; + } else { + if (strnicmp(haystack, needle, needle_len) == 0) + return haystack; + } haystack++; } return NULL; @@ -280,7 +291,8 @@ int ct_sip_get_info(const char *dptr, size_t dlen, continue; } aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen, - ct_sip_lnlen(dptr, limit)); + ct_sip_lnlen(dptr, limit), + hnfo->case_sensitive); if (!aux) { DEBUGP("'%s' not found in '%s'.\n", hnfo->ln_str, hnfo->lname); diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index 47097aac63d8..e16604c4339d 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -87,14 +87,15 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, buffer, bufflen, POS_VIA)) return 0; - /* This search should ignore case, but later.. */ aux = ct_sip_search("CSeq:", *dptr, sizeof("CSeq:") - 1, - (*pskb)->len - dataoff); + (*pskb)->len - dataoff, 0); if (!aux) return 0; if (!ct_sip_search("REGISTER", aux, sizeof("REGISTER"), - ct_sip_lnlen(aux, *dptr + (*pskb)->len - dataoff))) + ct_sip_lnlen(aux, + *dptr + (*pskb)->len - dataoff), + 1)) return 1; return mangle_sip_packet(pskb, ctinfo, ct, dptr, -- cgit v1.2.3 From 1b683b551209ca46ae59b29572018001db5af078 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:30 +0100 Subject: [NETFILTER]: sip conntrack: better NAT handling The NAT handling of the SIP helper has a few problems: - Request headers are only mangled in the reply direction, From/To headers not at all, which can lead to authentication failures with DNAT in case the authentication domain is the IP address - Contact headers in responses are only mangled for REGISTER responses - Headers may be mangled even though they contain addresses not participating in the connection, like alternative addresses - Packets are droppen when domain names are used where the helper expects IP addresses This patch takes a different approach, instead of fixed rules what field to mangle to what content, it adds symetric mapping of From/To/Via/Contact headers, which allows to deal properly with echoed addresses in responses and foreign addresses not belonging to the connection. Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4/ip_conntrack_sip.h | 5 +- net/ipv4/netfilter/ip_conntrack_sip.c | 29 +++- net/ipv4/netfilter/ip_nat_sip.c | 171 ++++++++++++++---------- 3 files changed, 135 insertions(+), 70 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sip.h b/include/linux/netfilter_ipv4/ip_conntrack_sip.h index 51c65ac18c57..bef6c646defa 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sip.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sip.h @@ -6,7 +6,10 @@ #define SIP_TIMEOUT 3600 enum sip_header_pos { - POS_REQ_HEADER, + POS_REG_REQ_URI, + POS_REQ_URI, + POS_FROM, + POS_TO, POS_VIA, POS_CONTACT, POS_CONTENT, diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index 0a6a13c45b07..3a26d63eed88 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -69,13 +69,38 @@ struct sip_header_nfo { }; static struct sip_header_nfo ct_sip_hdrs[] = { - [POS_REQ_HEADER] = { /* SIP Requests headers */ + [POS_REG_REQ_URI] = { /* SIP REGISTER request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = epaddr_len + }, + [POS_REQ_URI] = { /* SIP request URI */ .lname = "sip:", .lnlen = sizeof("sip:") - 1, .ln_str = "@", .ln_strlen = sizeof("@") - 1, .match_len = epaddr_len }, + [POS_FROM] = { /* SIP From header */ + .lname = "From:", + .lnlen = sizeof("From:") - 1, + .sname = "\r\nf:", + .snlen = sizeof("\r\nf:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, + [POS_TO] = { /* SIP To header */ + .lname = "To:", + .lnlen = sizeof("To:") - 1, + .sname = "\r\nt:", + .snlen = sizeof("\r\nt:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, [POS_VIA] = { /* SIP Via header */ .lname = "Via:", .lnlen = sizeof("Via:") - 1, @@ -284,7 +309,7 @@ int ct_sip_get_info(const char *dptr, size_t dlen, while (dptr <= limit) { if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) && - (hinfo->sname == NULL || + (hnfo->sname == NULL || strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) { dptr++; continue; diff --git a/net/ipv4/netfilter/ip_nat_sip.c b/net/ipv4/netfilter/ip_nat_sip.c index e16604c4339d..6223abc924ff 100644 --- a/net/ipv4/netfilter/ip_nat_sip.c +++ b/net/ipv4/netfilter/ip_nat_sip.c @@ -29,25 +29,70 @@ MODULE_DESCRIPTION("SIP NAT helper"); #define DEBUGP(format, args...) #endif -static unsigned int mangle_sip_packet(struct sk_buff **pskb, - enum ip_conntrack_info ctinfo, - struct ip_conntrack *ct, - const char **dptr, size_t dlen, - char *buffer, int bufflen, - enum sip_header_pos pos) +struct addr_map { + struct { + char src[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + char dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned int srclen, srciplen; + unsigned int dstlen, dstiplen; + } addr[IP_CT_DIR_MAX]; +}; + +static void addr_map_init(struct ip_conntrack *ct, struct addr_map *map) { - unsigned int matchlen, matchoff; + struct ip_conntrack_tuple *t; + enum ip_conntrack_dir dir; + unsigned int n; + + for (dir = 0; dir < IP_CT_DIR_MAX; dir++) { + t = &ct->tuplehash[dir].tuple; + + n = sprintf(map->addr[dir].src, "%u.%u.%u.%u", + NIPQUAD(t->src.ip)); + map->addr[dir].srciplen = n; + n += sprintf(map->addr[dir].src + n, ":%u", + ntohs(t->src.u.udp.port)); + map->addr[dir].srclen = n; + + n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u", + NIPQUAD(t->dst.ip)); + map->addr[dir].dstiplen = n; + n += sprintf(map->addr[dir].dst + n, ":%u", + ntohs(t->dst.u.udp.port)); + map->addr[dir].dstlen = n; + } +} + +static int map_sip_addr(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, + struct ip_conntrack *ct, const char **dptr, size_t dlen, + enum sip_header_pos pos, struct addr_map *map) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff, addrlen; + char *addr; if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) - return 0; + return 1; + + if ((matchlen == map->addr[dir].srciplen || + matchlen == map->addr[dir].srclen) && + memcmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { + addr = map->addr[!dir].dst; + addrlen = map->addr[!dir].dstlen; + } else if ((matchlen == map->addr[dir].dstiplen || + matchlen == map->addr[dir].dstlen) && + memcmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) { + addr = map->addr[!dir].src; + addrlen = map->addr[!dir].srclen; + } else + return 1; if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, - matchoff, matchlen, buffer, bufflen)) + matchoff, matchlen, addr, addrlen)) return 0; - - /* We need to reload this. Thanks Patrick. */ *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); return 1; + } static unsigned int ip_nat_sip(struct sk_buff **pskb, @@ -55,69 +100,61 @@ static unsigned int ip_nat_sip(struct sk_buff **pskb, struct ip_conntrack *ct, const char **dptr) { - enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; - unsigned int bufflen, dataoff; - __be32 ip; - __be16 port; + enum sip_header_pos pos; + struct addr_map map; + int dataoff, datalen; dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0") - 1) + return NF_DROP; + + addr_map_init(ct, &map); + + /* Basic rules: requests and responses. */ + if (strncmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) != 0) { + /* 10.2: Constructing the REGISTER Request: + * + * The "userinfo" and "@" components of the SIP URI MUST NOT + * be present. + */ + if (datalen >= sizeof("REGISTER") - 1 && + strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0) + pos = POS_REG_REQ_URI; + else + pos = POS_REQ_URI; + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, pos, &map)) + return NF_DROP; + } - ip = ct->tuplehash[!dir].tuple.dst.ip; - port = ct->tuplehash[!dir].tuple.dst.u.udp.port; - bufflen = sprintf(buffer, "%u.%u.%u.%u:%u", NIPQUAD(ip), ntohs(port)); + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_FROM, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_TO, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_VIA, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map)) + return NF_DROP; + return NF_ACCEPT; +} + +static unsigned int mangle_sip_packet(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct ip_conntrack *ct, + const char **dptr, size_t dlen, + char *buffer, int bufflen, + enum sip_header_pos pos) +{ + unsigned int matchlen, matchoff; - /* short packet ? */ - if (((*pskb)->len - dataoff) < (sizeof("SIP/2.0") - 1)) + if (ct_sip_get_info(*dptr, dlen, &matchoff, &matchlen, pos) <= 0) return 0; - /* Basic rules: requests and responses. */ - if (memcmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) == 0) { - const char *aux; - - if ((ctinfo) < IP_CT_IS_REPLY) { - mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - return 1; - } + if (!ip_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, buffer, bufflen)) + return 0; - if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_VIA)) - return 0; - - aux = ct_sip_search("CSeq:", *dptr, sizeof("CSeq:") - 1, - (*pskb)->len - dataoff, 0); - if (!aux) - return 0; - - if (!ct_sip_search("REGISTER", aux, sizeof("REGISTER"), - ct_sip_lnlen(aux, - *dptr + (*pskb)->len - dataoff), - 1)) - return 1; - - return mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - } - if ((ctinfo) < IP_CT_IS_REPLY) { - if (!mangle_sip_packet(pskb, ctinfo, ct, dptr, - (*pskb)->len - dataoff, - buffer, bufflen, POS_VIA)) - return 0; - - /* Mangle Contact if exists only. - watch udp_nat_mangle()! */ - mangle_sip_packet(pskb, ctinfo, ct, dptr, (*pskb)->len - dataoff, - buffer, bufflen, POS_CONTACT); - return 1; - } - /* This mangle requests headers. */ - return mangle_sip_packet(pskb, ctinfo, ct, dptr, - ct_sip_lnlen(*dptr, - *dptr + (*pskb)->len - dataoff), - buffer, bufflen, POS_REQ_HEADER); + /* We need to reload this. Thanks Patrick. */ + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; } static int mangle_content_len(struct sk_buff **pskb, -- cgit v1.2.3 From 829e17a1a602572ffa3beefe582dc103ee9fb9c7 Mon Sep 17 00:00:00 2001 From: Eric Leblond Date: Wed, 29 Nov 2006 02:35:33 +0100 Subject: [NETFILTER]: nfnetlink_queue: allow changing queue length through netlink Signed-off-by: Eric Leblond Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink_queue.h | 1 + net/netfilter/nfnetlink_queue.c | 8 ++++++++ 2 files changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/netfilter/nfnetlink_queue.h b/include/linux/netfilter/nfnetlink_queue.h index 4beea3d6fda8..83e789633e35 100644 --- a/include/linux/netfilter/nfnetlink_queue.h +++ b/include/linux/netfilter/nfnetlink_queue.h @@ -82,6 +82,7 @@ enum nfqnl_attr_config { NFQA_CFG_UNSPEC, NFQA_CFG_CMD, /* nfqnl_msg_config_cmd */ NFQA_CFG_PARAMS, /* nfqnl_msg_config_params */ + NFQA_CFG_QUEUE_MAXLEN, /* u_int32_t */ __NFQA_CFG_MAX }; #define NFQA_CFG_MAX (__NFQA_CFG_MAX-1) diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 82e4454659bf..a88a017da22c 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -947,6 +947,14 @@ nfqnl_recv_config(struct sock *ctnl, struct sk_buff *skb, ntohl(params->copy_range)); } + if (nfqa[NFQA_CFG_QUEUE_MAXLEN-1]) { + __be32 *queue_maxlen; + queue_maxlen = NFA_DATA(nfqa[NFQA_CFG_QUEUE_MAXLEN-1]); + spin_lock_bh(&queue->lock); + queue->queue_maxlen = ntohl(*queue_maxlen); + spin_unlock_bh(&queue->lock); + } + out_put: instance_put(queue); return ret; -- cgit v1.2.3 From d7a5c32442ed3d528b9ddfd3d5b837bad0ffa9da Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:34 +0100 Subject: [NETFILTER]: nfnetlink_log: remove useless prefix length limitation There is no reason for limiting netlink attributes in size. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink_log.h | 2 -- net/netfilter/nfnetlink_log.c | 19 ++++++++++--------- 2 files changed, 10 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nfnetlink_log.h b/include/linux/netfilter/nfnetlink_log.h index 55a2a2b814ed..5966afa026e9 100644 --- a/include/linux/netfilter/nfnetlink_log.h +++ b/include/linux/netfilter/nfnetlink_log.h @@ -32,8 +32,6 @@ struct nfulnl_msg_packet_timestamp { aligned_be64 usec; }; -#define NFULNL_PREFIXLEN 30 /* just like old log target */ - enum nfulnl_attr_type { NFULA_UNSPEC, NFULA_PACKET_HDR, diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index bd3ffa6f1a6d..d1505dd25c66 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -408,7 +408,7 @@ __build_packet_message(struct nfulnl_instance *inst, const struct net_device *indev, const struct net_device *outdev, const struct nf_loginfo *li, - const char *prefix) + const char *prefix, unsigned int plen) { unsigned char *old_tail; struct nfulnl_msg_packet_hdr pmsg; @@ -432,12 +432,8 @@ __build_packet_message(struct nfulnl_instance *inst, NFA_PUT(inst->skb, NFULA_PACKET_HDR, sizeof(pmsg), &pmsg); - if (prefix) { - int slen = strlen(prefix); - if (slen > NFULNL_PREFIXLEN) - slen = NFULNL_PREFIXLEN; - NFA_PUT(inst->skb, NFULA_PREFIX, slen, prefix); - } + if (prefix) + NFA_PUT(inst->skb, NFULA_PREFIX, plen, prefix); if (indev) { tmp_uint = htonl(indev->ifindex); @@ -601,6 +597,7 @@ nfulnl_log_packet(unsigned int pf, const struct nf_loginfo *li; unsigned int qthreshold; unsigned int nlbufsiz; + unsigned int plen; if (li_user && li_user->type == NF_LOG_TYPE_ULOG) li = li_user; @@ -616,6 +613,10 @@ nfulnl_log_packet(unsigned int pf, return; } + plen = 0; + if (prefix) + plen = strlen(prefix); + /* all macros expand to constant values at compile time */ /* FIXME: do we want to make the size calculation conditional based on * what is actually present? way more branches and checks, but more @@ -630,7 +631,7 @@ nfulnl_log_packet(unsigned int pf, #endif + NFA_SPACE(sizeof(u_int32_t)) /* mark */ + NFA_SPACE(sizeof(u_int32_t)) /* uid */ - + NFA_SPACE(NFULNL_PREFIXLEN) /* prefix */ + + NFA_SPACE(plen) /* prefix */ + NFA_SPACE(sizeof(struct nfulnl_msg_packet_hw)) + NFA_SPACE(sizeof(struct nfulnl_msg_packet_timestamp)); @@ -701,7 +702,7 @@ nfulnl_log_packet(unsigned int pf, inst->qlen++; __build_packet_message(inst, skb, data_len, pf, - hooknum, in, out, li, prefix); + hooknum, in, out, li, prefix, plen); /* timer_pending always called within inst->lock, so there * is no chance of a race here */ -- cgit v1.2.3 From 39b46fc6f0d1161a5585cd8af7b3a05e8118ab7e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:36 +0100 Subject: [NETFILTER]: x_tables: add port of hashlimit match for IPv4 and IPv6 Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_hashlimit.h | 40 ++ include/linux/netfilter_ipv4/ipt_hashlimit.h | 42 +- net/ipv4/netfilter/Kconfig | 14 - net/ipv4/netfilter/Makefile | 1 - net/ipv4/netfilter/ipt_hashlimit.c | 733 ------------------------- net/netfilter/Kconfig | 14 + net/netfilter/Makefile | 1 + net/netfilter/xt_hashlimit.c | 772 +++++++++++++++++++++++++++ 9 files changed, 836 insertions(+), 782 deletions(-) create mode 100644 include/linux/netfilter/xt_hashlimit.h delete mode 100644 net/ipv4/netfilter/ipt_hashlimit.c create mode 100644 net/netfilter/xt_hashlimit.c (limited to 'include/linux') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index 312bd2ffee33..e379a2d89ea0 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -14,6 +14,7 @@ header-y += xt_dscp.h header-y += xt_DSCP.h header-y += xt_esp.h header-y += xt_helper.h +header-y += xt_hashlimit.h header-y += xt_length.h header-y += xt_limit.h header-y += xt_mac.h diff --git a/include/linux/netfilter/xt_hashlimit.h b/include/linux/netfilter/xt_hashlimit.h new file mode 100644 index 000000000000..b4556b8edbfd --- /dev/null +++ b/include/linux/netfilter/xt_hashlimit.h @@ -0,0 +1,40 @@ +#ifndef _XT_HASHLIMIT_H +#define _XT_HASHLIMIT_H + +/* timings are in milliseconds. */ +#define XT_HASHLIMIT_SCALE 10000 +/* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 + seconds, or one every 59 hours. */ + +/* details of this structure hidden by the implementation */ +struct xt_hashlimit_htable; + +#define XT_HASHLIMIT_HASH_DIP 0x0001 +#define XT_HASHLIMIT_HASH_DPT 0x0002 +#define XT_HASHLIMIT_HASH_SIP 0x0004 +#define XT_HASHLIMIT_HASH_SPT 0x0008 + +struct hashlimit_cfg { + u_int32_t mode; /* bitmask of IPT_HASHLIMIT_HASH_* */ + u_int32_t avg; /* Average secs between packets * scale */ + u_int32_t burst; /* Period multiplier for upper limit. */ + + /* user specified */ + u_int32_t size; /* how many buckets */ + u_int32_t max; /* max number of entries */ + u_int32_t gc_interval; /* gc interval */ + u_int32_t expire; /* when do entries expire? */ +}; + +struct xt_hashlimit_info { + char name [IFNAMSIZ]; /* name */ + struct hashlimit_cfg cfg; + struct xt_hashlimit_htable *hinfo; + + /* Used internally by the kernel */ + union { + void *ptr; + struct xt_hashlimit_info *master; + } u; +}; +#endif /*_XT_HASHLIMIT_H*/ diff --git a/include/linux/netfilter_ipv4/ipt_hashlimit.h b/include/linux/netfilter_ipv4/ipt_hashlimit.h index ac2cb64ecd76..5662120a3d7b 100644 --- a/include/linux/netfilter_ipv4/ipt_hashlimit.h +++ b/include/linux/netfilter_ipv4/ipt_hashlimit.h @@ -1,40 +1,14 @@ #ifndef _IPT_HASHLIMIT_H #define _IPT_HASHLIMIT_H -/* timings are in milliseconds. */ -#define IPT_HASHLIMIT_SCALE 10000 -/* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 - seconds, or one every 59 hours. */ +#include -/* details of this structure hidden by the implementation */ -struct ipt_hashlimit_htable; +#define IPT_HASHLIMIT_SCALE XT_HASHLIMIT_SCALE +#define IPT_HASHLIMIT_HASH_DIP XT_HASHLIMIT_HASH_DIP +#define IPT_HASHLIMIT_HASH_DPT XT_HASHLIMIT_HASH_DPT +#define IPT_HASHLIMIT_HASH_SIP XT_HASHLIMIT_HASH_SIP +#define IPT_HASHLIMIT_HASH_SPT XT_HASHLIMIT_HASH_SPT -#define IPT_HASHLIMIT_HASH_DIP 0x0001 -#define IPT_HASHLIMIT_HASH_DPT 0x0002 -#define IPT_HASHLIMIT_HASH_SIP 0x0004 -#define IPT_HASHLIMIT_HASH_SPT 0x0008 +#define ipt_hashlimit_info xt_hashlimit_info -struct hashlimit_cfg { - u_int32_t mode; /* bitmask of IPT_HASHLIMIT_HASH_* */ - u_int32_t avg; /* Average secs between packets * scale */ - u_int32_t burst; /* Period multiplier for upper limit. */ - - /* user specified */ - u_int32_t size; /* how many buckets */ - u_int32_t max; /* max number of entries */ - u_int32_t gc_interval; /* gc interval */ - u_int32_t expire; /* when do entries expire? */ -}; - -struct ipt_hashlimit_info { - char name [IFNAMSIZ]; /* name */ - struct hashlimit_cfg cfg; - struct ipt_hashlimit_htable *hinfo; - - /* Used internally by the kernel */ - union { - void *ptr; - struct ipt_hashlimit_info *master; - } u; -}; -#endif /*_IPT_HASHLIMIT_H*/ +#endif /* _IPT_HASHLIMIT_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 4ac5b5c4678d..bc298a3f236f 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -326,20 +326,6 @@ config IP_NF_MATCH_ADDRTYPE If you want to compile it as a module, say M here and read . If unsure, say `N'. -config IP_NF_MATCH_HASHLIMIT - tristate 'hashlimit match support' - depends on IP_NF_IPTABLES - help - This option adds a new iptables `hashlimit' match. - - As opposed to `limit', this match dynamically creates a hash table - of limit buckets, based on your selection of source/destination - ip addresses and/or ports. - - It enables you to express policies like `10kpps for any given - destination IP' or `500pps from any given source IP' with a single - IPtables rule. - # `filter', generic and specific targets config IP_NF_FILTER tristate "Packet filtering" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 4ce20ebc4d6c..21359d83f0c7 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -53,7 +53,6 @@ obj-$(CONFIG_IP_NF_NAT) += iptable_nat.o obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o # matches -obj-$(CONFIG_IP_NF_MATCH_HASHLIMIT) += ipt_hashlimit.o obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c deleted file mode 100644 index 33ccdbf8e794..000000000000 --- a/net/ipv4/netfilter/ipt_hashlimit.c +++ /dev/null @@ -1,733 +0,0 @@ -/* iptables match extension to limit the number of packets per second - * seperately for each hashbucket (sourceip/sourceport/dstip/dstport) - * - * (C) 2003-2004 by Harald Welte - * - * $Id: ipt_hashlimit.c 3244 2004-10-20 16:24:29Z laforge@netfilter.org $ - * - * Development of this code was funded by Astaro AG, http://www.astaro.com/ - * - * based on ipt_limit.c by: - * Jérôme de Vivie - * Hervé Eychenne - * Rusty Russell - * - * The general idea is to create a hash table for every dstip and have a - * seperate limit counter per tuple. This way you can do something like 'limit - * the number of syn packets for each of my internal addresses. - * - * Ideally this would just be implemented as a general 'hash' match, which would - * allow us to attach any iptables target to it's hash buckets. But this is - * not possible in the current iptables architecture. As always, pkttables for - * 2.7.x will help ;) - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -/* FIXME: this is just for IP_NF_ASSERRT */ -#include -#include - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Harald Welte "); -MODULE_DESCRIPTION("iptables match for limiting per hash-bucket"); - -/* need to declare this at the top */ -static struct proc_dir_entry *hashlimit_procdir; -static struct file_operations dl_file_ops; - -/* hash table crap */ - -struct dsthash_dst { - __be32 src_ip; - __be32 dst_ip; - /* ports have to be consecutive !!! */ - __be16 src_port; - __be16 dst_port; -}; - -struct dsthash_ent { - /* static / read-only parts in the beginning */ - struct hlist_node node; - struct dsthash_dst dst; - - /* modified structure members in the end */ - unsigned long expires; /* precalculated expiry time */ - struct { - unsigned long prev; /* last modification */ - u_int32_t credit; - u_int32_t credit_cap, cost; - } rateinfo; -}; - -struct ipt_hashlimit_htable { - struct hlist_node node; /* global list of all htables */ - atomic_t use; - - struct hashlimit_cfg cfg; /* config */ - - /* used internally */ - spinlock_t lock; /* lock for list_head */ - u_int32_t rnd; /* random seed for hash */ - int rnd_initialized; - struct timer_list timer; /* timer for gc */ - atomic_t count; /* number entries in table */ - - /* seq_file stuff */ - struct proc_dir_entry *pde; - - struct hlist_head hash[0]; /* hashtable itself */ -}; - -static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ -static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ -static HLIST_HEAD(hashlimit_htables); -static kmem_cache_t *hashlimit_cachep __read_mostly; - -static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) -{ - return (ent->dst.dst_ip == b->dst_ip - && ent->dst.dst_port == b->dst_port - && ent->dst.src_port == b->src_port - && ent->dst.src_ip == b->src_ip); -} - -static inline u_int32_t -hash_dst(const struct ipt_hashlimit_htable *ht, const struct dsthash_dst *dst) -{ - return (jhash_3words((__force u32)dst->dst_ip, - ((__force u32)dst->dst_port<<16 | - (__force u32)dst->src_port), - (__force u32)dst->src_ip, ht->rnd) % ht->cfg.size); -} - -static inline struct dsthash_ent * -__dsthash_find(const struct ipt_hashlimit_htable *ht, struct dsthash_dst *dst) -{ - struct dsthash_ent *ent; - struct hlist_node *pos; - u_int32_t hash = hash_dst(ht, dst); - - if (!hlist_empty(&ht->hash[hash])) - hlist_for_each_entry(ent, pos, &ht->hash[hash], node) { - if (dst_cmp(ent, dst)) { - return ent; - } - } - - return NULL; -} - -/* allocate dsthash_ent, initialize dst, put in htable and lock it */ -static struct dsthash_ent * -__dsthash_alloc_init(struct ipt_hashlimit_htable *ht, struct dsthash_dst *dst) -{ - struct dsthash_ent *ent; - - /* initialize hash with random val at the time we allocate - * the first hashtable entry */ - if (!ht->rnd_initialized) { - get_random_bytes(&ht->rnd, 4); - ht->rnd_initialized = 1; - } - - if (ht->cfg.max && - atomic_read(&ht->count) >= ht->cfg.max) { - /* FIXME: do something. question is what.. */ - if (net_ratelimit()) - printk(KERN_WARNING - "ipt_hashlimit: max count of %u reached\n", - ht->cfg.max); - return NULL; - } - - ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC); - if (!ent) { - if (net_ratelimit()) - printk(KERN_ERR - "ipt_hashlimit: can't allocate dsthash_ent\n"); - return NULL; - } - - atomic_inc(&ht->count); - - ent->dst.dst_ip = dst->dst_ip; - ent->dst.dst_port = dst->dst_port; - ent->dst.src_ip = dst->src_ip; - ent->dst.src_port = dst->src_port; - - hlist_add_head(&ent->node, &ht->hash[hash_dst(ht, dst)]); - - return ent; -} - -static inline void -__dsthash_free(struct ipt_hashlimit_htable *ht, struct dsthash_ent *ent) -{ - hlist_del(&ent->node); - kmem_cache_free(hashlimit_cachep, ent); - atomic_dec(&ht->count); -} -static void htable_gc(unsigned long htlong); - -static int htable_create(struct ipt_hashlimit_info *minfo) -{ - int i; - unsigned int size; - struct ipt_hashlimit_htable *hinfo; - - if (minfo->cfg.size) - size = minfo->cfg.size; - else { - size = (((num_physpages << PAGE_SHIFT) / 16384) - / sizeof(struct list_head)); - if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) - size = 8192; - if (size < 16) - size = 16; - } - /* FIXME: don't use vmalloc() here or anywhere else -HW */ - hinfo = vmalloc(sizeof(struct ipt_hashlimit_htable) - + (sizeof(struct list_head) * size)); - if (!hinfo) { - printk(KERN_ERR "ipt_hashlimit: Unable to create hashtable\n"); - return -1; - } - minfo->hinfo = hinfo; - - /* copy match config into hashtable config */ - memcpy(&hinfo->cfg, &minfo->cfg, sizeof(hinfo->cfg)); - hinfo->cfg.size = size; - if (!hinfo->cfg.max) - hinfo->cfg.max = 8 * hinfo->cfg.size; - else if (hinfo->cfg.max < hinfo->cfg.size) - hinfo->cfg.max = hinfo->cfg.size; - - for (i = 0; i < hinfo->cfg.size; i++) - INIT_HLIST_HEAD(&hinfo->hash[i]); - - atomic_set(&hinfo->count, 0); - atomic_set(&hinfo->use, 1); - hinfo->rnd_initialized = 0; - spin_lock_init(&hinfo->lock); - hinfo->pde = create_proc_entry(minfo->name, 0, hashlimit_procdir); - if (!hinfo->pde) { - vfree(hinfo); - return -1; - } - hinfo->pde->proc_fops = &dl_file_ops; - hinfo->pde->data = hinfo; - - init_timer(&hinfo->timer); - hinfo->timer.expires = jiffies + msecs_to_jiffies(hinfo->cfg.gc_interval); - hinfo->timer.data = (unsigned long )hinfo; - hinfo->timer.function = htable_gc; - add_timer(&hinfo->timer); - - spin_lock_bh(&hashlimit_lock); - hlist_add_head(&hinfo->node, &hashlimit_htables); - spin_unlock_bh(&hashlimit_lock); - - return 0; -} - -static int select_all(struct ipt_hashlimit_htable *ht, struct dsthash_ent *he) -{ - return 1; -} - -static int select_gc(struct ipt_hashlimit_htable *ht, struct dsthash_ent *he) -{ - return (jiffies >= he->expires); -} - -static void htable_selective_cleanup(struct ipt_hashlimit_htable *ht, - int (*select)(struct ipt_hashlimit_htable *ht, - struct dsthash_ent *he)) -{ - int i; - - IP_NF_ASSERT(ht->cfg.size && ht->cfg.max); - - /* lock hash table and iterate over it */ - spin_lock_bh(&ht->lock); - for (i = 0; i < ht->cfg.size; i++) { - struct dsthash_ent *dh; - struct hlist_node *pos, *n; - hlist_for_each_entry_safe(dh, pos, n, &ht->hash[i], node) { - if ((*select)(ht, dh)) - __dsthash_free(ht, dh); - } - } - spin_unlock_bh(&ht->lock); -} - -/* hash table garbage collector, run by timer */ -static void htable_gc(unsigned long htlong) -{ - struct ipt_hashlimit_htable *ht = (struct ipt_hashlimit_htable *)htlong; - - htable_selective_cleanup(ht, select_gc); - - /* re-add the timer accordingly */ - ht->timer.expires = jiffies + msecs_to_jiffies(ht->cfg.gc_interval); - add_timer(&ht->timer); -} - -static void htable_destroy(struct ipt_hashlimit_htable *hinfo) -{ - /* remove timer, if it is pending */ - if (timer_pending(&hinfo->timer)) - del_timer(&hinfo->timer); - - /* remove proc entry */ - remove_proc_entry(hinfo->pde->name, hashlimit_procdir); - - htable_selective_cleanup(hinfo, select_all); - vfree(hinfo); -} - -static struct ipt_hashlimit_htable *htable_find_get(char *name) -{ - struct ipt_hashlimit_htable *hinfo; - struct hlist_node *pos; - - spin_lock_bh(&hashlimit_lock); - hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) { - if (!strcmp(name, hinfo->pde->name)) { - atomic_inc(&hinfo->use); - spin_unlock_bh(&hashlimit_lock); - return hinfo; - } - } - spin_unlock_bh(&hashlimit_lock); - - return NULL; -} - -static void htable_put(struct ipt_hashlimit_htable *hinfo) -{ - if (atomic_dec_and_test(&hinfo->use)) { - spin_lock_bh(&hashlimit_lock); - hlist_del(&hinfo->node); - spin_unlock_bh(&hashlimit_lock); - htable_destroy(hinfo); - } -} - - -/* The algorithm used is the Simple Token Bucket Filter (TBF) - * see net/sched/sch_tbf.c in the linux source tree - */ - -/* Rusty: This is my (non-mathematically-inclined) understanding of - this algorithm. The `average rate' in jiffies becomes your initial - amount of credit `credit' and the most credit you can ever have - `credit_cap'. The `peak rate' becomes the cost of passing the - test, `cost'. - - `prev' tracks the last packet hit: you gain one credit per jiffy. - If you get credit balance more than this, the extra credit is - discarded. Every time the match passes, you lose `cost' credits; - if you don't have that many, the test fails. - - See Alexey's formal explanation in net/sched/sch_tbf.c. - - To get the maximum range, we multiply by this factor (ie. you get N - credits per jiffy). We want to allow a rate as low as 1 per day - (slowest userspace tool allows), which means - CREDITS_PER_JIFFY*HZ*60*60*24 < 2^32 ie. -*/ -#define MAX_CPJ (0xFFFFFFFF / (HZ*60*60*24)) - -/* Repeated shift and or gives us all 1s, final shift and add 1 gives - * us the power of 2 below the theoretical max, so GCC simply does a - * shift. */ -#define _POW2_BELOW2(x) ((x)|((x)>>1)) -#define _POW2_BELOW4(x) (_POW2_BELOW2(x)|_POW2_BELOW2((x)>>2)) -#define _POW2_BELOW8(x) (_POW2_BELOW4(x)|_POW2_BELOW4((x)>>4)) -#define _POW2_BELOW16(x) (_POW2_BELOW8(x)|_POW2_BELOW8((x)>>8)) -#define _POW2_BELOW32(x) (_POW2_BELOW16(x)|_POW2_BELOW16((x)>>16)) -#define POW2_BELOW32(x) ((_POW2_BELOW32(x)>>1) + 1) - -#define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) - -/* Precision saver. */ -static inline u_int32_t -user2credits(u_int32_t user) -{ - /* If multiplying would overflow... */ - if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) - /* Divide first. */ - return (user / IPT_HASHLIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; - - return (user * HZ * CREDITS_PER_JIFFY) / IPT_HASHLIMIT_SCALE; -} - -static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now) -{ - dh->rateinfo.credit += (now - xchg(&dh->rateinfo.prev, now)) - * CREDITS_PER_JIFFY; - if (dh->rateinfo.credit > dh->rateinfo.credit_cap) - dh->rateinfo.credit = dh->rateinfo.credit_cap; -} - -static int -hashlimit_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const struct xt_match *match, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - struct ipt_hashlimit_info *r = - ((struct ipt_hashlimit_info *)matchinfo)->u.master; - struct ipt_hashlimit_htable *hinfo = r->hinfo; - unsigned long now = jiffies; - struct dsthash_ent *dh; - struct dsthash_dst dst; - - /* build 'dst' according to hinfo->cfg and current packet */ - memset(&dst, 0, sizeof(dst)); - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DIP) - dst.dst_ip = skb->nh.iph->daddr; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SIP) - dst.src_ip = skb->nh.iph->saddr; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DPT - ||hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SPT) { - __be16 _ports[2], *ports; - - switch (skb->nh.iph->protocol) { - case IPPROTO_TCP: - case IPPROTO_UDP: - case IPPROTO_SCTP: - case IPPROTO_DCCP: - ports = skb_header_pointer(skb, skb->nh.iph->ihl*4, - sizeof(_ports), &_ports); - break; - default: - _ports[0] = _ports[1] = 0; - ports = _ports; - break; - } - if (!ports) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - *hotdrop = 1; - return 0; - } - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_SPT) - dst.src_port = ports[0]; - if (hinfo->cfg.mode & IPT_HASHLIMIT_HASH_DPT) - dst.dst_port = ports[1]; - } - - spin_lock_bh(&hinfo->lock); - dh = __dsthash_find(hinfo, &dst); - if (!dh) { - dh = __dsthash_alloc_init(hinfo, &dst); - - if (!dh) { - /* enomem... don't match == DROP */ - if (net_ratelimit()) - printk(KERN_ERR "%s: ENOMEM\n", __FUNCTION__); - spin_unlock_bh(&hinfo->lock); - return 0; - } - - dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire); - - dh->rateinfo.prev = jiffies; - dh->rateinfo.credit = user2credits(hinfo->cfg.avg * - hinfo->cfg.burst); - dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg * - hinfo->cfg.burst); - dh->rateinfo.cost = user2credits(hinfo->cfg.avg); - } else { - /* update expiration timeout */ - dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire); - rateinfo_recalc(dh, now); - } - - if (dh->rateinfo.credit >= dh->rateinfo.cost) { - /* We're underlimit. */ - dh->rateinfo.credit -= dh->rateinfo.cost; - spin_unlock_bh(&hinfo->lock); - return 1; - } - - spin_unlock_bh(&hinfo->lock); - - /* default case: we're overlimit, thus don't match */ - return 0; -} - -static int -hashlimit_checkentry(const char *tablename, - const void *inf, - const struct xt_match *match, - void *matchinfo, - unsigned int hook_mask) -{ - struct ipt_hashlimit_info *r = matchinfo; - - /* Check for overflow. */ - if (r->cfg.burst == 0 - || user2credits(r->cfg.avg * r->cfg.burst) < - user2credits(r->cfg.avg)) { - printk(KERN_ERR "ipt_hashlimit: Overflow, try lower: %u/%u\n", - r->cfg.avg, r->cfg.burst); - return 0; - } - - if (r->cfg.mode == 0 - || r->cfg.mode > (IPT_HASHLIMIT_HASH_DPT - |IPT_HASHLIMIT_HASH_DIP - |IPT_HASHLIMIT_HASH_SIP - |IPT_HASHLIMIT_HASH_SPT)) - return 0; - - if (!r->cfg.gc_interval) - return 0; - - if (!r->cfg.expire) - return 0; - - if (r->name[sizeof(r->name) - 1] != '\0') - return 0; - - /* This is the best we've got: We cannot release and re-grab lock, - * since checkentry() is called before ip_tables.c grabs ipt_mutex. - * We also cannot grab the hashtable spinlock, since htable_create will - * call vmalloc, and that can sleep. And we cannot just re-search - * the list of htable's in htable_create(), since then we would - * create duplicate proc files. -HW */ - mutex_lock(&hlimit_mutex); - r->hinfo = htable_find_get(r->name); - if (!r->hinfo && (htable_create(r) != 0)) { - mutex_unlock(&hlimit_mutex); - return 0; - } - mutex_unlock(&hlimit_mutex); - - /* Ugly hack: For SMP, we only want to use one set */ - r->u.master = r; - - return 1; -} - -static void -hashlimit_destroy(const struct xt_match *match, void *matchinfo) -{ - struct ipt_hashlimit_info *r = matchinfo; - - htable_put(r->hinfo); -} - -#ifdef CONFIG_COMPAT -struct compat_ipt_hashlimit_info { - char name[IFNAMSIZ]; - struct hashlimit_cfg cfg; - compat_uptr_t hinfo; - compat_uptr_t master; -}; - -static void compat_from_user(void *dst, void *src) -{ - int off = offsetof(struct compat_ipt_hashlimit_info, hinfo); - - memcpy(dst, src, off); - memset(dst + off, 0, sizeof(struct compat_ipt_hashlimit_info) - off); -} - -static int compat_to_user(void __user *dst, void *src) -{ - int off = offsetof(struct compat_ipt_hashlimit_info, hinfo); - - return copy_to_user(dst, src, off) ? -EFAULT : 0; -} -#endif - -static struct ipt_match ipt_hashlimit = { - .name = "hashlimit", - .match = hashlimit_match, - .matchsize = sizeof(struct ipt_hashlimit_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_ipt_hashlimit_info), - .compat_from_user = compat_from_user, - .compat_to_user = compat_to_user, -#endif - .checkentry = hashlimit_checkentry, - .destroy = hashlimit_destroy, - .me = THIS_MODULE -}; - -/* PROC stuff */ - -static void *dl_seq_start(struct seq_file *s, loff_t *pos) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket; - - spin_lock_bh(&htable->lock); - if (*pos >= htable->cfg.size) - return NULL; - - bucket = kmalloc(sizeof(unsigned int), GFP_ATOMIC); - if (!bucket) - return ERR_PTR(-ENOMEM); - - *bucket = *pos; - return bucket; -} - -static void *dl_seq_next(struct seq_file *s, void *v, loff_t *pos) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - - *pos = ++(*bucket); - if (*pos >= htable->cfg.size) { - kfree(v); - return NULL; - } - return bucket; -} - -static void dl_seq_stop(struct seq_file *s, void *v) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - - kfree(bucket); - - spin_unlock_bh(&htable->lock); -} - -static inline int dl_seq_real_show(struct dsthash_ent *ent, struct seq_file *s) -{ - /* recalculate to show accurate numbers */ - rateinfo_recalc(ent, jiffies); - - return seq_printf(s, "%ld %u.%u.%u.%u:%u->%u.%u.%u.%u:%u %u %u %u\n", - (long)(ent->expires - jiffies)/HZ, - NIPQUAD(ent->dst.src_ip), ntohs(ent->dst.src_port), - NIPQUAD(ent->dst.dst_ip), ntohs(ent->dst.dst_port), - ent->rateinfo.credit, ent->rateinfo.credit_cap, - ent->rateinfo.cost); -} - -static int dl_seq_show(struct seq_file *s, void *v) -{ - struct proc_dir_entry *pde = s->private; - struct ipt_hashlimit_htable *htable = pde->data; - unsigned int *bucket = (unsigned int *)v; - struct dsthash_ent *ent; - struct hlist_node *pos; - - if (!hlist_empty(&htable->hash[*bucket])) - hlist_for_each_entry(ent, pos, &htable->hash[*bucket], node) { - if (dl_seq_real_show(ent, s)) { - /* buffer was filled and unable to print that tuple */ - return 1; - } - } - - return 0; -} - -static struct seq_operations dl_seq_ops = { - .start = dl_seq_start, - .next = dl_seq_next, - .stop = dl_seq_stop, - .show = dl_seq_show -}; - -static int dl_proc_open(struct inode *inode, struct file *file) -{ - int ret = seq_open(file, &dl_seq_ops); - - if (!ret) { - struct seq_file *sf = file->private_data; - sf->private = PDE(inode); - } - return ret; -} - -static struct file_operations dl_file_ops = { - .owner = THIS_MODULE, - .open = dl_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release -}; - -static int init_or_fini(int fini) -{ - int ret = 0; - - if (fini) - goto cleanup; - - if (ipt_register_match(&ipt_hashlimit)) { - ret = -EINVAL; - goto cleanup_nothing; - } - - hashlimit_cachep = kmem_cache_create("ipt_hashlimit", - sizeof(struct dsthash_ent), 0, - 0, NULL, NULL); - if (!hashlimit_cachep) { - printk(KERN_ERR "Unable to create ipt_hashlimit slab cache\n"); - ret = -ENOMEM; - goto cleanup_unreg_match; - } - - hashlimit_procdir = proc_mkdir("ipt_hashlimit", proc_net); - if (!hashlimit_procdir) { - printk(KERN_ERR "Unable to create proc dir entry\n"); - ret = -ENOMEM; - goto cleanup_free_slab; - } - - return ret; - -cleanup: - remove_proc_entry("ipt_hashlimit", proc_net); -cleanup_free_slab: - kmem_cache_destroy(hashlimit_cachep); -cleanup_unreg_match: - ipt_unregister_match(&ipt_hashlimit); -cleanup_nothing: - return ret; - -} - -static int __init ipt_hashlimit_init(void) -{ - return init_or_fini(0); -} - -static void __exit ipt_hashlimit_fini(void) -{ - init_or_fini(1); -} - -module_init(ipt_hashlimit_init); -module_exit(ipt_hashlimit_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index f619c6527266..7e6125467c12 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -464,5 +464,19 @@ config NETFILTER_XT_MATCH_TCPMSS To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_MATCH_HASHLIMIT + tristate '"hashlimit" match support' + depends on NETFILTER_XTABLES + help + This option adds a `hashlimit' match. + + As opposed to `limit', this match dynamically creates a hash table + of limit buckets, based on your selection of source/destination + addresses and/or ports. + + It enables you to express policies like `10kpps for any given + destination address' or `500pps from any given source address' + with a single rule. + endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 84d529ded952..f85811bfcfe5 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -59,3 +59,4 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_STATISTIC) += xt_statistic.o obj-$(CONFIG_NETFILTER_XT_MATCH_STRING) += xt_string.o obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o +obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c new file mode 100644 index 000000000000..46de566d3721 --- /dev/null +++ b/net/netfilter/xt_hashlimit.c @@ -0,0 +1,772 @@ +/* iptables match extension to limit the number of packets per second + * seperately for each hashbucket (sourceip/sourceport/dstip/dstport) + * + * (C) 2003-2004 by Harald Welte + * + * $Id: ipt_hashlimit.c 3244 2004-10-20 16:24:29Z laforge@netfilter.org $ + * + * Development of this code was funded by Astaro AG, http://www.astaro.com/ + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("iptables match for limiting per hash-bucket"); +MODULE_ALIAS("ipt_hashlimit"); +MODULE_ALIAS("ip6t_hashlimit"); + +/* need to declare this at the top */ +static struct proc_dir_entry *hashlimit_procdir4; +static struct proc_dir_entry *hashlimit_procdir6; +static struct file_operations dl_file_ops; + +/* hash table crap */ +struct dsthash_dst { + union { + struct { + __be32 src; + __be32 dst; + } ip; + struct { + __be32 src[4]; + __be32 dst[4]; + } ip6; + } addr; + __be16 src_port; + __be16 dst_port; +}; + +struct dsthash_ent { + /* static / read-only parts in the beginning */ + struct hlist_node node; + struct dsthash_dst dst; + + /* modified structure members in the end */ + unsigned long expires; /* precalculated expiry time */ + struct { + unsigned long prev; /* last modification */ + u_int32_t credit; + u_int32_t credit_cap, cost; + } rateinfo; +}; + +struct xt_hashlimit_htable { + struct hlist_node node; /* global list of all htables */ + atomic_t use; + int family; + + struct hashlimit_cfg cfg; /* config */ + + /* used internally */ + spinlock_t lock; /* lock for list_head */ + u_int32_t rnd; /* random seed for hash */ + int rnd_initialized; + unsigned int count; /* number entries in table */ + struct timer_list timer; /* timer for gc */ + + /* seq_file stuff */ + struct proc_dir_entry *pde; + + struct hlist_head hash[0]; /* hashtable itself */ +}; + +static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ +static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ +static HLIST_HEAD(hashlimit_htables); +static kmem_cache_t *hashlimit_cachep __read_mostly; + +static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) +{ + return !memcmp(&ent->dst, b, sizeof(ent->dst)); +} + +static u_int32_t +hash_dst(const struct xt_hashlimit_htable *ht, const struct dsthash_dst *dst) +{ + return jhash(dst, sizeof(*dst), ht->rnd) % ht->cfg.size; +} + +static struct dsthash_ent * +dsthash_find(const struct xt_hashlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + struct hlist_node *pos; + u_int32_t hash = hash_dst(ht, dst); + + if (!hlist_empty(&ht->hash[hash])) { + hlist_for_each_entry(ent, pos, &ht->hash[hash], node) + if (dst_cmp(ent, dst)) + return ent; + } + return NULL; +} + +/* allocate dsthash_ent, initialize dst, put in htable and lock it */ +static struct dsthash_ent * +dsthash_alloc_init(struct xt_hashlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + + /* initialize hash with random val at the time we allocate + * the first hashtable entry */ + if (!ht->rnd_initialized) { + get_random_bytes(&ht->rnd, 4); + ht->rnd_initialized = 1; + } + + if (ht->cfg.max && ht->count >= ht->cfg.max) { + /* FIXME: do something. question is what.. */ + if (net_ratelimit()) + printk(KERN_WARNING + "xt_hashlimit: max count of %u reached\n", + ht->cfg.max); + return NULL; + } + + ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC); + if (!ent) { + if (net_ratelimit()) + printk(KERN_ERR + "xt_hashlimit: can't allocate dsthash_ent\n"); + return NULL; + } + memcpy(&ent->dst, dst, sizeof(ent->dst)); + + hlist_add_head(&ent->node, &ht->hash[hash_dst(ht, dst)]); + ht->count++; + return ent; +} + +static inline void +dsthash_free(struct xt_hashlimit_htable *ht, struct dsthash_ent *ent) +{ + hlist_del(&ent->node); + kmem_cache_free(hashlimit_cachep, ent); + ht->count--; +} +static void htable_gc(unsigned long htlong); + +static int htable_create(struct xt_hashlimit_info *minfo, int family) +{ + struct xt_hashlimit_htable *hinfo; + unsigned int size; + unsigned int i; + + if (minfo->cfg.size) + size = minfo->cfg.size; + else { + size = ((num_physpages << PAGE_SHIFT) / 16384) / + sizeof(struct list_head); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + size = 8192; + if (size < 16) + size = 16; + } + /* FIXME: don't use vmalloc() here or anywhere else -HW */ + hinfo = vmalloc(sizeof(struct xt_hashlimit_htable) + + sizeof(struct list_head) * size); + if (!hinfo) { + printk(KERN_ERR "xt_hashlimit: unable to create hashtable\n"); + return -1; + } + minfo->hinfo = hinfo; + + /* copy match config into hashtable config */ + memcpy(&hinfo->cfg, &minfo->cfg, sizeof(hinfo->cfg)); + hinfo->cfg.size = size; + if (!hinfo->cfg.max) + hinfo->cfg.max = 8 * hinfo->cfg.size; + else if (hinfo->cfg.max < hinfo->cfg.size) + hinfo->cfg.max = hinfo->cfg.size; + + for (i = 0; i < hinfo->cfg.size; i++) + INIT_HLIST_HEAD(&hinfo->hash[i]); + + atomic_set(&hinfo->use, 1); + hinfo->count = 0; + hinfo->family = family; + hinfo->rnd_initialized = 0; + spin_lock_init(&hinfo->lock); + hinfo->pde = create_proc_entry(minfo->name, 0, + family == AF_INET ? hashlimit_procdir4 : + hashlimit_procdir6); + if (!hinfo->pde) { + vfree(hinfo); + return -1; + } + hinfo->pde->proc_fops = &dl_file_ops; + hinfo->pde->data = hinfo; + + init_timer(&hinfo->timer); + hinfo->timer.expires = jiffies + msecs_to_jiffies(hinfo->cfg.gc_interval); + hinfo->timer.data = (unsigned long )hinfo; + hinfo->timer.function = htable_gc; + add_timer(&hinfo->timer); + + spin_lock_bh(&hashlimit_lock); + hlist_add_head(&hinfo->node, &hashlimit_htables); + spin_unlock_bh(&hashlimit_lock); + + return 0; +} + +static int select_all(struct xt_hashlimit_htable *ht, struct dsthash_ent *he) +{ + return 1; +} + +static int select_gc(struct xt_hashlimit_htable *ht, struct dsthash_ent *he) +{ + return (jiffies >= he->expires); +} + +static void htable_selective_cleanup(struct xt_hashlimit_htable *ht, + int (*select)(struct xt_hashlimit_htable *ht, + struct dsthash_ent *he)) +{ + unsigned int i; + + /* lock hash table and iterate over it */ + spin_lock_bh(&ht->lock); + for (i = 0; i < ht->cfg.size; i++) { + struct dsthash_ent *dh; + struct hlist_node *pos, *n; + hlist_for_each_entry_safe(dh, pos, n, &ht->hash[i], node) { + if ((*select)(ht, dh)) + dsthash_free(ht, dh); + } + } + spin_unlock_bh(&ht->lock); +} + +/* hash table garbage collector, run by timer */ +static void htable_gc(unsigned long htlong) +{ + struct xt_hashlimit_htable *ht = (struct xt_hashlimit_htable *)htlong; + + htable_selective_cleanup(ht, select_gc); + + /* re-add the timer accordingly */ + ht->timer.expires = jiffies + msecs_to_jiffies(ht->cfg.gc_interval); + add_timer(&ht->timer); +} + +static void htable_destroy(struct xt_hashlimit_htable *hinfo) +{ + /* remove timer, if it is pending */ + if (timer_pending(&hinfo->timer)) + del_timer(&hinfo->timer); + + /* remove proc entry */ + remove_proc_entry(hinfo->pde->name, + hinfo->family == AF_INET ? hashlimit_procdir4 : + hashlimit_procdir6); + htable_selective_cleanup(hinfo, select_all); + vfree(hinfo); +} + +static struct xt_hashlimit_htable *htable_find_get(char *name, int family) +{ + struct xt_hashlimit_htable *hinfo; + struct hlist_node *pos; + + spin_lock_bh(&hashlimit_lock); + hlist_for_each_entry(hinfo, pos, &hashlimit_htables, node) { + if (!strcmp(name, hinfo->pde->name) && + hinfo->family == family) { + atomic_inc(&hinfo->use); + spin_unlock_bh(&hashlimit_lock); + return hinfo; + } + } + spin_unlock_bh(&hashlimit_lock); + return NULL; +} + +static void htable_put(struct xt_hashlimit_htable *hinfo) +{ + if (atomic_dec_and_test(&hinfo->use)) { + spin_lock_bh(&hashlimit_lock); + hlist_del(&hinfo->node); + spin_unlock_bh(&hashlimit_lock); + htable_destroy(hinfo); + } +} + +/* The algorithm used is the Simple Token Bucket Filter (TBF) + * see net/sched/sch_tbf.c in the linux source tree + */ + +/* Rusty: This is my (non-mathematically-inclined) understanding of + this algorithm. The `average rate' in jiffies becomes your initial + amount of credit `credit' and the most credit you can ever have + `credit_cap'. The `peak rate' becomes the cost of passing the + test, `cost'. + + `prev' tracks the last packet hit: you gain one credit per jiffy. + If you get credit balance more than this, the extra credit is + discarded. Every time the match passes, you lose `cost' credits; + if you don't have that many, the test fails. + + See Alexey's formal explanation in net/sched/sch_tbf.c. + + To get the maximum range, we multiply by this factor (ie. you get N + credits per jiffy). We want to allow a rate as low as 1 per day + (slowest userspace tool allows), which means + CREDITS_PER_JIFFY*HZ*60*60*24 < 2^32 ie. +*/ +#define MAX_CPJ (0xFFFFFFFF / (HZ*60*60*24)) + +/* Repeated shift and or gives us all 1s, final shift and add 1 gives + * us the power of 2 below the theoretical max, so GCC simply does a + * shift. */ +#define _POW2_BELOW2(x) ((x)|((x)>>1)) +#define _POW2_BELOW4(x) (_POW2_BELOW2(x)|_POW2_BELOW2((x)>>2)) +#define _POW2_BELOW8(x) (_POW2_BELOW4(x)|_POW2_BELOW4((x)>>4)) +#define _POW2_BELOW16(x) (_POW2_BELOW8(x)|_POW2_BELOW8((x)>>8)) +#define _POW2_BELOW32(x) (_POW2_BELOW16(x)|_POW2_BELOW16((x)>>16)) +#define POW2_BELOW32(x) ((_POW2_BELOW32(x)>>1) + 1) + +#define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) + +/* Precision saver. */ +static inline u_int32_t +user2credits(u_int32_t user) +{ + /* If multiplying would overflow... */ + if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) + /* Divide first. */ + return (user / XT_HASHLIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; + + return (user * HZ * CREDITS_PER_JIFFY) / XT_HASHLIMIT_SCALE; +} + +static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now) +{ + dh->rateinfo.credit += (now - dh->rateinfo.prev) * CREDITS_PER_JIFFY; + if (dh->rateinfo.credit > dh->rateinfo.credit_cap) + dh->rateinfo.credit = dh->rateinfo.credit_cap; + dh->rateinfo.prev = now; +} + +static int +hashlimit_init_dst(struct xt_hashlimit_htable *hinfo, struct dsthash_dst *dst, + const struct sk_buff *skb, unsigned int protoff) +{ + __be16 _ports[2], *ports; + int nexthdr; + + memset(dst, 0, sizeof(*dst)); + + switch (hinfo->family) { + case AF_INET: + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) + dst->addr.ip.dst = skb->nh.iph->daddr; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SIP) + dst->addr.ip.src = skb->nh.iph->saddr; + + if (!(hinfo->cfg.mode & + (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT))) + return 0; + nexthdr = skb->nh.iph->protocol; + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) + memcpy(&dst->addr.ip6.dst, &skb->nh.ipv6h->daddr, + sizeof(dst->addr.ip6.dst)); + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SIP) + memcpy(&dst->addr.ip6.src, &skb->nh.ipv6h->saddr, + sizeof(dst->addr.ip6.src)); + + if (!(hinfo->cfg.mode & + (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT))) + return 0; + nexthdr = ipv6_find_hdr(skb, &protoff, -1, NULL); + if (nexthdr < 0) + return -1; + break; +#endif + default: + BUG(); + return 0; + } + + switch (nexthdr) { + case IPPROTO_TCP: + case IPPROTO_UDP: + case IPPROTO_SCTP: + case IPPROTO_DCCP: + ports = skb_header_pointer(skb, protoff, sizeof(_ports), + &_ports); + break; + default: + _ports[0] = _ports[1] = 0; + ports = _ports; + break; + } + if (!ports) + return -1; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_SPT) + dst->src_port = ports[0]; + if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DPT) + dst->dst_port = ports[1]; + return 0; +} + +static int +hashlimit_match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const struct xt_match *match, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + struct xt_hashlimit_info *r = + ((struct xt_hashlimit_info *)matchinfo)->u.master; + struct xt_hashlimit_htable *hinfo = r->hinfo; + unsigned long now = jiffies; + struct dsthash_ent *dh; + struct dsthash_dst dst; + + if (hashlimit_init_dst(hinfo, &dst, skb, protoff) < 0) + goto hotdrop; + + spin_lock_bh(&hinfo->lock); + dh = dsthash_find(hinfo, &dst); + if (!dh) { + dh = dsthash_alloc_init(hinfo, &dst); + if (!dh) { + spin_unlock_bh(&hinfo->lock); + goto hotdrop; + } + + dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire); + dh->rateinfo.prev = jiffies; + dh->rateinfo.credit = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.cost = user2credits(hinfo->cfg.avg); + } else { + /* update expiration timeout */ + dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire); + rateinfo_recalc(dh, now); + } + + if (dh->rateinfo.credit >= dh->rateinfo.cost) { + /* We're underlimit. */ + dh->rateinfo.credit -= dh->rateinfo.cost; + spin_unlock_bh(&hinfo->lock); + return 1; + } + + spin_unlock_bh(&hinfo->lock); + + /* default case: we're overlimit, thus don't match */ + return 0; + +hotdrop: + *hotdrop = 1; + return 0; +} + +static int +hashlimit_checkentry(const char *tablename, + const void *inf, + const struct xt_match *match, + void *matchinfo, + unsigned int hook_mask) +{ + struct xt_hashlimit_info *r = matchinfo; + + /* Check for overflow. */ + if (r->cfg.burst == 0 || + user2credits(r->cfg.avg * r->cfg.burst) < user2credits(r->cfg.avg)) { + printk(KERN_ERR "xt_hashlimit: overflow, try lower: %u/%u\n", + r->cfg.avg, r->cfg.burst); + return 0; + } + if (r->cfg.mode == 0 || + r->cfg.mode > (XT_HASHLIMIT_HASH_DPT | + XT_HASHLIMIT_HASH_DIP | + XT_HASHLIMIT_HASH_SIP | + XT_HASHLIMIT_HASH_SPT)) + return 0; + if (!r->cfg.gc_interval) + return 0; + if (!r->cfg.expire) + return 0; + if (r->name[sizeof(r->name) - 1] != '\0') + return 0; + + /* This is the best we've got: We cannot release and re-grab lock, + * since checkentry() is called before x_tables.c grabs xt_mutex. + * We also cannot grab the hashtable spinlock, since htable_create will + * call vmalloc, and that can sleep. And we cannot just re-search + * the list of htable's in htable_create(), since then we would + * create duplicate proc files. -HW */ + mutex_lock(&hlimit_mutex); + r->hinfo = htable_find_get(r->name, match->family); + if (!r->hinfo && htable_create(r, match->family) != 0) { + mutex_unlock(&hlimit_mutex); + return 0; + } + mutex_unlock(&hlimit_mutex); + + /* Ugly hack: For SMP, we only want to use one set */ + r->u.master = r; + return 1; +} + +static void +hashlimit_destroy(const struct xt_match *match, void *matchinfo) +{ + struct xt_hashlimit_info *r = matchinfo; + + htable_put(r->hinfo); +} + +#ifdef CONFIG_COMPAT +struct compat_xt_hashlimit_info { + char name[IFNAMSIZ]; + struct hashlimit_cfg cfg; + compat_uptr_t hinfo; + compat_uptr_t master; +}; + +static void compat_from_user(void *dst, void *src) +{ + int off = offsetof(struct compat_xt_hashlimit_info, hinfo); + + memcpy(dst, src, off); + memset(dst + off, 0, sizeof(struct compat_xt_hashlimit_info) - off); +} + +static int compat_to_user(void __user *dst, void *src) +{ + int off = offsetof(struct compat_xt_hashlimit_info, hinfo); + + return copy_to_user(dst, src, off) ? -EFAULT : 0; +} +#endif + +static struct xt_match xt_hashlimit[] = { + { + .name = "hashlimit", + .family = AF_INET, + .match = hashlimit_match, + .matchsize = sizeof(struct xt_hashlimit_info), +#ifdef CONFIG_COMPAT + .compatsize = sizeof(struct compat_xt_hashlimit_info), + .compat_from_user = compat_from_user, + .compat_to_user = compat_to_user, +#endif + .checkentry = hashlimit_checkentry, + .destroy = hashlimit_destroy, + .me = THIS_MODULE + }, + { + .name = "hashlimit", + .family = AF_INET6, + .match = hashlimit_match, + .matchsize = sizeof(struct xt_hashlimit_info), +#ifdef CONFIG_COMPAT + .compatsize = sizeof(struct compat_xt_hashlimit_info), + .compat_from_user = compat_from_user, + .compat_to_user = compat_to_user, +#endif + .checkentry = hashlimit_checkentry, + .destroy = hashlimit_destroy, + .me = THIS_MODULE + }, +}; + +/* PROC stuff */ +static void *dl_seq_start(struct seq_file *s, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket; + + spin_lock_bh(&htable->lock); + if (*pos >= htable->cfg.size) + return NULL; + + bucket = kmalloc(sizeof(unsigned int), GFP_ATOMIC); + if (!bucket) + return ERR_PTR(-ENOMEM); + + *bucket = *pos; + return bucket; +} + +static void *dl_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + *pos = ++(*bucket); + if (*pos >= htable->cfg.size) { + kfree(v); + return NULL; + } + return bucket; +} + +static void dl_seq_stop(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + kfree(bucket); + spin_unlock_bh(&htable->lock); +} + +static int dl_seq_real_show(struct dsthash_ent *ent, int family, + struct seq_file *s) +{ + /* recalculate to show accurate numbers */ + rateinfo_recalc(ent, jiffies); + + switch (family) { + case AF_INET: + return seq_printf(s, "%ld %u.%u.%u.%u:%u->" + "%u.%u.%u.%u:%u %u %u %u\n", + (long)(ent->expires - jiffies)/HZ, + NIPQUAD(ent->dst.addr.ip.src), + ntohs(ent->dst.src_port), + NIPQUAD(ent->dst.addr.ip.dst), + ntohs(ent->dst.dst_port), + ent->rateinfo.credit, ent->rateinfo.credit_cap, + ent->rateinfo.cost); + case AF_INET6: + return seq_printf(s, "%ld " NIP6_FMT ":%u->" + NIP6_FMT ":%u %u %u %u\n", + (long)(ent->expires - jiffies)/HZ, + NIP6(*(struct in6_addr *)&ent->dst.addr.ip6.src), + ntohs(ent->dst.src_port), + NIP6(*(struct in6_addr *)&ent->dst.addr.ip6.dst), + ntohs(ent->dst.dst_port), + ent->rateinfo.credit, ent->rateinfo.credit_cap, + ent->rateinfo.cost); + default: + BUG(); + return 0; + } +} + +static int dl_seq_show(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct xt_hashlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + struct dsthash_ent *ent; + struct hlist_node *pos; + + if (!hlist_empty(&htable->hash[*bucket])) { + hlist_for_each_entry(ent, pos, &htable->hash[*bucket], node) + if (dl_seq_real_show(ent, htable->family, s)) + return 1; + } + return 0; +} + +static struct seq_operations dl_seq_ops = { + .start = dl_seq_start, + .next = dl_seq_next, + .stop = dl_seq_stop, + .show = dl_seq_show +}; + +static int dl_proc_open(struct inode *inode, struct file *file) +{ + int ret = seq_open(file, &dl_seq_ops); + + if (!ret) { + struct seq_file *sf = file->private_data; + sf->private = PDE(inode); + } + return ret; +} + +static struct file_operations dl_file_ops = { + .owner = THIS_MODULE, + .open = dl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static int __init xt_hashlimit_init(void) +{ + int err; + + err = xt_register_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); + if (err < 0) + goto err1; + + err = -ENOMEM; + hashlimit_cachep = kmem_cache_create("xt_hashlimit", + sizeof(struct dsthash_ent), 0, 0, + NULL, NULL); + if (!hashlimit_cachep) { + printk(KERN_ERR "xt_hashlimit: unable to create slab cache\n"); + goto err2; + } + hashlimit_procdir4 = proc_mkdir("ipt_hashlimit", proc_net); + if (!hashlimit_procdir4) { + printk(KERN_ERR "xt_hashlimit: unable to create proc dir " + "entry\n"); + goto err3; + } + hashlimit_procdir6 = proc_mkdir("ip6t_hashlimit", proc_net); + if (!hashlimit_procdir6) { + printk(KERN_ERR "xt_hashlimit: tnable to create proc dir " + "entry\n"); + goto err4; + } + return 0; +err4: + remove_proc_entry("ipt_hashlimit", proc_net); +err3: + kmem_cache_destroy(hashlimit_cachep); +err2: + xt_unregister_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); +err1: + return err; + +} + +static void __exit xt_hashlimit_fini(void) +{ + remove_proc_entry("ipt_hashlimit", proc_net); + remove_proc_entry("ip6t_hashlimit", proc_net); + kmem_cache_destroy(hashlimit_cachep); + xt_unregister_matches(xt_hashlimit, ARRAY_SIZE(xt_hashlimit)); +} + +module_init(xt_hashlimit_init); +module_exit(xt_hashlimit_fini); -- cgit v1.2.3 From baf7b1e11282127e068d149825cccec002091d61 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 29 Nov 2006 02:35:38 +0100 Subject: [NETFILTER]: x_tables: add NFLOG target Add new NFLOG target to allow use of nfnetlink_log for both IPv4 and IPv6. Currently we have two (unsupported by userspace) hacks in the LOG and ULOG targets to optionally call to the nflog API. They lack a few features, namely the IPv4 and IPv6 LOG targets can not specify a number of arguments related to nfnetlink_log, while the ULOG target is only available for IPv4. Remove those hacks and add a clean way to use nfnetlink_log. Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_NFLOG.h | 18 +++++++ include/linux/netfilter_ipv4/ipt_LOG.h | 2 +- include/linux/netfilter_ipv6/ip6t_LOG.h | 2 +- net/ipv4/netfilter/ipt_LOG.c | 9 +--- net/ipv6/netfilter/ip6t_LOG.c | 9 +--- net/netfilter/Kconfig | 11 +++++ net/netfilter/Makefile | 1 + net/netfilter/xt_NFLOG.c | 86 +++++++++++++++++++++++++++++++++ 9 files changed, 123 insertions(+), 16 deletions(-) create mode 100644 include/linux/netfilter/xt_NFLOG.h create mode 100644 net/netfilter/xt_NFLOG.c (limited to 'include/linux') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index e379a2d89ea0..6328175a1c3a 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -22,6 +22,7 @@ header-y += xt_mark.h header-y += xt_MARK.h header-y += xt_multiport.h header-y += xt_NFQUEUE.h +header-y += xt_NFLOG.h header-y += xt_pkttype.h header-y += xt_policy.h header-y += xt_realm.h diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h new file mode 100644 index 000000000000..cdcd0ed58f7a --- /dev/null +++ b/include/linux/netfilter/xt_NFLOG.h @@ -0,0 +1,18 @@ +#ifndef _XT_NFLOG_TARGET +#define _XT_NFLOG_TARGET + +#define XT_NFLOG_DEFAULT_GROUP 0x1 +#define XT_NFLOG_DEFAULT_THRESHOLD 1 + +#define XT_NFLOG_MASK 0x0 + +struct xt_nflog_info { + u_int32_t len; + u_int16_t group; + u_int16_t threshold; + u_int16_t flags; + u_int16_t pad; + char prefix[64]; +}; + +#endif /* _XT_NFLOG_TARGET */ diff --git a/include/linux/netfilter_ipv4/ipt_LOG.h b/include/linux/netfilter_ipv4/ipt_LOG.h index 892f9a33fea8..90fa6525ef9c 100644 --- a/include/linux/netfilter_ipv4/ipt_LOG.h +++ b/include/linux/netfilter_ipv4/ipt_LOG.h @@ -6,7 +6,7 @@ #define IPT_LOG_TCPOPT 0x02 /* Log TCP options */ #define IPT_LOG_IPOPT 0x04 /* Log IP options */ #define IPT_LOG_UID 0x08 /* Log UID owning local socket */ -#define IPT_LOG_NFLOG 0x10 /* Log using nf_log backend */ +#define IPT_LOG_NFLOG 0x10 /* Unsupported, don't reuse */ #define IPT_LOG_MASK 0x1f struct ipt_log_info { diff --git a/include/linux/netfilter_ipv6/ip6t_LOG.h b/include/linux/netfilter_ipv6/ip6t_LOG.h index 060c1a1c6c60..0d0119b0458c 100644 --- a/include/linux/netfilter_ipv6/ip6t_LOG.h +++ b/include/linux/netfilter_ipv6/ip6t_LOG.h @@ -6,7 +6,7 @@ #define IP6T_LOG_TCPOPT 0x02 /* Log TCP options */ #define IP6T_LOG_IPOPT 0x04 /* Log IP options */ #define IP6T_LOG_UID 0x08 /* Log UID owning local socket */ -#define IP6T_LOG_NFLOG 0x10 /* Log using nf_log backend */ +#define IP6T_LOG_NFLOG 0x10 /* Unsupported, don't use */ #define IP6T_LOG_MASK 0x1f struct ip6t_log_info { diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 46eee64a11f6..c96de16fefae 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -430,13 +430,8 @@ ipt_log_target(struct sk_buff **pskb, li.u.log.level = loginfo->level; li.u.log.logflags = loginfo->logflags; - if (loginfo->logflags & IPT_LOG_NFLOG) - nf_log_packet(PF_INET, hooknum, *pskb, in, out, &li, - "%s", loginfo->prefix); - else - ipt_log_packet(PF_INET, hooknum, *pskb, in, out, &li, - loginfo->prefix); - + ipt_log_packet(PF_INET, hooknum, *pskb, in, out, &li, + loginfo->prefix); return IPT_CONTINUE; } diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index f4857cf97f05..33b1faa90d74 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -440,13 +440,8 @@ ip6t_log_target(struct sk_buff **pskb, li.u.log.level = loginfo->level; li.u.log.logflags = loginfo->logflags; - if (loginfo->logflags & IP6T_LOG_NFLOG) - nf_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, - "%s", loginfo->prefix); - else - ip6t_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, - loginfo->prefix); - + ip6t_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, + loginfo->prefix); return IP6T_CONTINUE; } diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 7e6125467c12..d191dacead5e 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -184,6 +184,17 @@ config NETFILTER_XT_TARGET_NFQUEUE To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_TARGET_NFLOG + tristate '"NFLOG" target support' + depends on NETFILTER_XTABLES + help + This option enables the NFLOG target, which allows to LOG + messages through the netfilter logging API, which can use + either the old LOG target, the old ULOG target or nfnetlink_log + as backend. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_NOTRACK tristate '"NOTRACK" target support' depends on NETFILTER_XTABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index f85811bfcfe5..7f0089c584bf 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -31,6 +31,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_DSCP) += xt_DSCP.o obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o +obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c new file mode 100644 index 000000000000..901ed7abaa1b --- /dev/null +++ b/net/netfilter/xt_NFLOG.c @@ -0,0 +1,86 @@ +/* + * Copyright (c) 2006 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Patrick McHardy "); +MODULE_DESCRIPTION("x_tables NFLOG target"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_NFLOG"); +MODULE_ALIAS("ip6t_NFLOG"); + +static unsigned int +nflog_target(struct sk_buff **pskb, + const struct net_device *in, const struct net_device *out, + unsigned int hooknum, const struct xt_target *target, + const void *targinfo) +{ + const struct xt_nflog_info *info = targinfo; + struct nf_loginfo li; + + li.type = NF_LOG_TYPE_ULOG; + li.u.ulog.copy_len = info->len; + li.u.ulog.group = info->group; + li.u.ulog.qthreshold = info->threshold; + + nf_log_packet(target->family, hooknum, *pskb, in, out, &li, + "%s", info->prefix); + return XT_CONTINUE; +} + +static int +nflog_checkentry(const char *tablename, const void *entry, + const struct xt_target *target, void *targetinfo, + unsigned int hookmask) +{ + struct xt_nflog_info *info = targetinfo; + + if (info->flags & ~XT_NFLOG_MASK) + return 0; + if (info->prefix[sizeof(info->prefix) - 1] != '\0') + return 0; + return 1; +} + +static struct xt_target xt_nflog_target[] = { + { + .name = "NFLOG", + .family = AF_INET, + .checkentry = nflog_checkentry, + .target = nflog_target, + .targetsize = sizeof(struct xt_nflog_info), + .me = THIS_MODULE, + }, + { + .name = "NFLOG", + .family = AF_INET6, + .checkentry = nflog_checkentry, + .target = nflog_target, + .targetsize = sizeof(struct xt_nflog_info), + .me = THIS_MODULE, + }, +}; + +static int __init xt_nflog_init(void) +{ + return xt_register_targets(xt_nflog_target, + ARRAY_SIZE(xt_nflog_target)); +} + +static void __exit xt_nflog_fini(void) +{ + xt_unregister_targets(xt_nflog_target, ARRAY_SIZE(xt_nflog_target)); +} + +module_init(xt_nflog_init); +module_exit(xt_nflog_fini); -- cgit v1.2.3 From d12cdc3ccf140bd2febef1c1be92284571da983f Mon Sep 17 00:00:00 2001 From: Bart De Schuymer Date: Wed, 29 Nov 2006 02:35:40 +0100 Subject: [NETFILTER]: ebtables: add --snap-arp option The attached patch adds --snat-arp support, which makes it possible to change the source mac address in both the mac header and the arp header with one rule. Signed-off-by: Bart De Schuymer Signed-off-by: Patrick McHardy --- include/linux/netfilter_bridge/ebt_nat.h | 1 + include/linux/netfilter_bridge/ebtables.h | 4 ++++ net/bridge/netfilter/ebt_mark.c | 6 +++--- net/bridge/netfilter/ebt_snat.c | 27 ++++++++++++++++++++++++--- 4 files changed, 32 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_bridge/ebt_nat.h b/include/linux/netfilter_bridge/ebt_nat.h index 26fd90da4cd6..435b886a51aa 100644 --- a/include/linux/netfilter_bridge/ebt_nat.h +++ b/include/linux/netfilter_bridge/ebt_nat.h @@ -1,6 +1,7 @@ #ifndef __LINUX_BRIDGE_EBT_NAT_H #define __LINUX_BRIDGE_EBT_NAT_H +#define NAT_ARP_BIT (0x00000010) struct ebt_nat_info { unsigned char mac[ETH_ALEN]; diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index e6ea70de24d5..87775264ff0b 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -26,6 +26,10 @@ #define EBT_CONTINUE -3 #define EBT_RETURN -4 #define NUM_STANDARD_TARGETS 4 +/* ebtables target modules store the verdict inside an int. We can + * reclaim a part of this int for backwards compatible extensions. + * The 4 lsb are more than enough to store the verdict. */ +#define EBT_VERDICT_BITS 0x0000000F struct ebt_counter { diff --git a/net/bridge/netfilter/ebt_mark.c b/net/bridge/netfilter/ebt_mark.c index 2458638561cb..62d23c7b25e6 100644 --- a/net/bridge/netfilter/ebt_mark.c +++ b/net/bridge/netfilter/ebt_mark.c @@ -33,7 +33,7 @@ static int ebt_target_mark(struct sk_buff **pskb, unsigned int hooknr, else (*pskb)->mark ^= info->mark; - return info->target | -16; + return info->target | ~EBT_VERDICT_BITS; } static int ebt_target_mark_check(const char *tablename, unsigned int hookmask, @@ -44,13 +44,13 @@ static int ebt_target_mark_check(const char *tablename, unsigned int hookmask, if (datalen != EBT_ALIGN(sizeof(struct ebt_mark_t_info))) return -EINVAL; - tmp = info->target | -16; + tmp = info->target | ~EBT_VERDICT_BITS; if (BASE_CHAIN && tmp == EBT_RETURN) return -EINVAL; CLEAR_BASE_CHAIN_BIT; if (tmp < -NUM_STANDARD_TARGETS || tmp >= 0) return -EINVAL; - tmp = info->target & -16; + tmp = info->target & ~EBT_VERDICT_BITS; if (tmp != MARK_SET_VALUE && tmp != MARK_OR_VALUE && tmp != MARK_AND_VALUE && tmp != MARK_XOR_VALUE) return -EINVAL; diff --git a/net/bridge/netfilter/ebt_snat.c b/net/bridge/netfilter/ebt_snat.c index cbb33e24ca8a..a50722182bfe 100644 --- a/net/bridge/netfilter/ebt_snat.c +++ b/net/bridge/netfilter/ebt_snat.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include static int ebt_target_snat(struct sk_buff **pskb, unsigned int hooknr, const struct net_device *in, const struct net_device *out, @@ -31,24 +33,43 @@ static int ebt_target_snat(struct sk_buff **pskb, unsigned int hooknr, *pskb = nskb; } memcpy(eth_hdr(*pskb)->h_source, info->mac, ETH_ALEN); - return info->target; + if (!(info->target & NAT_ARP_BIT) && + eth_hdr(*pskb)->h_proto == htons(ETH_P_ARP)) { + struct arphdr _ah, *ap; + + ap = skb_header_pointer(*pskb, 0, sizeof(_ah), &_ah); + if (ap == NULL) + return EBT_DROP; + if (ap->ar_hln != ETH_ALEN) + goto out; + if (skb_store_bits(*pskb, sizeof(_ah), info->mac,ETH_ALEN)) + return EBT_DROP; + } +out: + return info->target | ~EBT_VERDICT_BITS; } static int ebt_target_snat_check(const char *tablename, unsigned int hookmask, const struct ebt_entry *e, void *data, unsigned int datalen) { struct ebt_nat_info *info = (struct ebt_nat_info *) data; + int tmp; if (datalen != EBT_ALIGN(sizeof(struct ebt_nat_info))) return -EINVAL; - if (BASE_CHAIN && info->target == EBT_RETURN) + tmp = info->target | ~EBT_VERDICT_BITS; + if (BASE_CHAIN && tmp == EBT_RETURN) return -EINVAL; CLEAR_BASE_CHAIN_BIT; if (strcmp(tablename, "nat")) return -EINVAL; if (hookmask & ~(1 << NF_BR_POST_ROUTING)) return -EINVAL; - if (INVALID_TARGET) + + if (tmp < -NUM_STANDARD_TARGETS || tmp >= 0) + return -EINVAL; + tmp = info->target | EBT_VERDICT_BITS; + if ((tmp & ~NAT_ARP_BIT) != ~NAT_ARP_BIT) return -EINVAL; return 0; } -- cgit v1.2.3 From 1e419cd9953f59d06d7b88d0e2911a68a0044f33 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 30 Nov 2006 19:28:48 -0800 Subject: [EBTABLES]: Split ebt_replace into user and kernel variants, annotate. Signed-off-by: Al Viro Signed-off-by: David S. Miller --- include/linux/netfilter_bridge/ebtables.h | 19 ++++++++++++++++++- net/bridge/netfilter/ebtable_broute.c | 2 +- net/bridge/netfilter/ebtable_filter.c | 2 +- net/bridge/netfilter/ebtable_nat.c | 2 +- net/bridge/netfilter/ebtables.c | 19 ++++++++++--------- 5 files changed, 31 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index 87775264ff0b..94e0a7dc0cb2 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -38,6 +38,23 @@ struct ebt_counter }; struct ebt_replace +{ + char name[EBT_TABLE_MAXNAMELEN]; + unsigned int valid_hooks; + /* nr of rules in the table */ + unsigned int nentries; + /* total size of the entries */ + unsigned int entries_size; + /* start of the chains */ + struct ebt_entries __user *hook_entry[NF_BR_NUMHOOKS]; + /* nr of counters userspace expects back */ + unsigned int num_counters; + /* where the kernel will put the old counters */ + struct ebt_counter __user *counters; + char __user *entries; +}; + +struct ebt_replace_kernel { char name[EBT_TABLE_MAXNAMELEN]; unsigned int valid_hooks; @@ -255,7 +272,7 @@ struct ebt_table { struct list_head list; char name[EBT_TABLE_MAXNAMELEN]; - struct ebt_replace *table; + struct ebt_replace_kernel *table; unsigned int valid_hooks; rwlock_t lock; /* e.g. could be the table explicitly only allows certain diff --git a/net/bridge/netfilter/ebtable_broute.c b/net/bridge/netfilter/ebtable_broute.c index 9a6e548e148b..d37ce0478938 100644 --- a/net/bridge/netfilter/ebtable_broute.c +++ b/net/bridge/netfilter/ebtable_broute.c @@ -23,7 +23,7 @@ static struct ebt_entries initial_chain = { .policy = EBT_ACCEPT, }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "broute", .valid_hooks = 1 << NF_BR_BROUTING, diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c index 3d5bd44f2395..127135ead2d5 100644 --- a/net/bridge/netfilter/ebtable_filter.c +++ b/net/bridge/netfilter/ebtable_filter.c @@ -30,7 +30,7 @@ static struct ebt_entries initial_chains[] = }, }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c index 04dd42efda1d..9c50488b62eb 100644 --- a/net/bridge/netfilter/ebtable_nat.c +++ b/net/bridge/netfilter/ebtable_nat.c @@ -30,7 +30,7 @@ static struct ebt_entries initial_chains[] = } }; -static struct ebt_replace initial_table = +static struct ebt_replace_kernel initial_table = { .name = "nat", .valid_hooks = NAT_VALID_HOOKS, diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 00a89705c1c4..bee558a41800 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -417,7 +417,8 @@ static int ebt_verify_pointers(struct ebt_replace *repl, for (i = 0; i < NF_BR_NUMHOOKS; i++) { if ((valid_hooks & (1 << i)) == 0) continue; - if ((char *)repl->hook_entry[i] == repl->entries + offset) + if ((char __user *)repl->hook_entry[i] == + repl->entries + offset) break; } @@ -1156,7 +1157,7 @@ int ebt_register_table(struct ebt_table *table) { struct ebt_table_info *newinfo; struct ebt_table *t; - struct ebt_replace *repl; + struct ebt_replace_kernel *repl; int ret, i, countersize; void *p; @@ -1320,33 +1321,33 @@ free_tmp: } static inline int ebt_make_matchname(struct ebt_entry_match *m, - char *base, char *ubase) + char *base, char __user *ubase) { - char *hlp = ubase - base + (char *)m; + char __user *hlp = ubase + ((char *)m - base); if (copy_to_user(hlp, m->u.match->name, EBT_FUNCTION_MAXNAMELEN)) return -EFAULT; return 0; } static inline int ebt_make_watchername(struct ebt_entry_watcher *w, - char *base, char *ubase) + char *base, char __user *ubase) { - char *hlp = ubase - base + (char *)w; + char __user *hlp = ubase + ((char *)w - base); if (copy_to_user(hlp , w->u.watcher->name, EBT_FUNCTION_MAXNAMELEN)) return -EFAULT; return 0; } -static inline int ebt_make_names(struct ebt_entry *e, char *base, char *ubase) +static inline int ebt_make_names(struct ebt_entry *e, char *base, char __user *ubase) { int ret; - char *hlp; + char __user *hlp; struct ebt_entry_target *t; if (e->bitmask == 0) return 0; - hlp = ubase - base + (char *)e + e->target_offset; + hlp = ubase + (((char *)e + e->target_offset) - base); t = (struct ebt_entry_target *)(((char *)e) + e->target_offset); ret = EBT_MATCH_ITERATE(e, ebt_make_matchname, base, ubase); -- cgit v1.2.3 From 9a7c9337a09b1ef07e55f95a4309957a2328a01f Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:04:04 -0800 Subject: [NET]: Accept wildcard delimiters in in[46]_pton Accept -1 as delimiter to abort parsing without an error at the first unknown character. This is needed by the upcoming nf_conntrack SIP helper, where addresses are delimited by either '\r' or '\n' characters. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/inet.h | 4 ++-- net/core/utils.c | 10 ++++++---- 2 files changed, 8 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/inet.h b/include/linux/inet.h index b7c6da7d6d32..675a7dbe86f8 100644 --- a/include/linux/inet.h +++ b/include/linux/inet.h @@ -46,7 +46,7 @@ #include extern __be32 in_aton(const char *str); -extern int in4_pton(const char *src, int srclen, u8 *dst, char delim, const char **end); -extern int in6_pton(const char *src, int srclen, u8 *dst, char delim, const char **end); +extern int in4_pton(const char *src, int srclen, u8 *dst, int delim, const char **end); +extern int in6_pton(const char *src, int srclen, u8 *dst, int delim, const char **end); #endif #endif /* _LINUX_INET_H */ diff --git a/net/core/utils.c b/net/core/utils.c index d93fe64f6693..61556065f07e 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -88,7 +88,7 @@ EXPORT_SYMBOL(in_aton); #define IN6PTON_NULL 0x20000000 /* first/tail */ #define IN6PTON_UNKNOWN 0x40000000 -static inline int digit2bin(char c, char delim) +static inline int digit2bin(char c, int delim) { if (c == delim || c == '\0') return IN6PTON_DELIM; @@ -99,7 +99,7 @@ static inline int digit2bin(char c, char delim) return IN6PTON_UNKNOWN; } -static inline int xdigit2bin(char c, char delim) +static inline int xdigit2bin(char c, int delim) { if (c == delim || c == '\0') return IN6PTON_DELIM; @@ -113,12 +113,14 @@ static inline int xdigit2bin(char c, char delim) return (IN6PTON_XDIGIT | (c - 'a' + 10)); if (c >= 'A' && c <= 'F') return (IN6PTON_XDIGIT | (c - 'A' + 10)); + if (delim == -1) + return IN6PTON_DELIM; return IN6PTON_UNKNOWN; } int in4_pton(const char *src, int srclen, u8 *dst, - char delim, const char **end) + int delim, const char **end) { const char *s; u8 *d; @@ -173,7 +175,7 @@ EXPORT_SYMBOL(in4_pton); int in6_pton(const char *src, int srclen, u8 *dst, - char delim, const char **end) + int delim, const char **end) { const char *s, *tok = NULL; u8 *d, *dc = NULL; -- cgit v1.2.3 From 5b1158e909ecbe1a052203e0d8df15633f829930 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Sat, 2 Dec 2006 22:07:13 -0800 Subject: [NETFILTER]: Add NAT support for nf_conntrack Add NAT support for nf_conntrack. Joint work of Jozsef Kadlecsik, Yasuyuki Kozakai, Martin Josefsson and myself. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter.h | 2 +- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 20 +- include/net/netfilter/nf_conntrack.h | 29 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- include/net/netfilter/nf_nat.h | 77 +++ include/net/netfilter/nf_nat_core.h | 27 ++ include/net/netfilter/nf_nat_helper.h | 32 ++ include/net/netfilter/nf_nat_protocol.h | 70 +++ include/net/netfilter/nf_nat_rule.h | 35 ++ net/ipv4/netfilter/Kconfig | 30 +- net/ipv4/netfilter/Makefile | 7 + net/ipv4/netfilter/ip_nat_standalone.c | 6 - net/ipv4/netfilter/ipt_MASQUERADE.c | 29 +- net/ipv4/netfilter/ipt_NETMAP.c | 4 + net/ipv4/netfilter/ipt_REDIRECT.c | 6 +- net/ipv4/netfilter/ipt_SAME.c | 12 +- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 7 +- net/ipv4/netfilter/nf_nat_core.c | 647 +++++++++++++++++++++++++ net/ipv4/netfilter/nf_nat_helper.c | 433 +++++++++++++++++ net/ipv4/netfilter/nf_nat_proto_icmp.c | 86 ++++ net/ipv4/netfilter/nf_nat_proto_tcp.c | 148 ++++++ net/ipv4/netfilter/nf_nat_proto_udp.c | 138 ++++++ net/ipv4/netfilter/nf_nat_proto_unknown.c | 54 +++ net/ipv4/netfilter/nf_nat_rule.c | 343 +++++++++++++ net/ipv4/netfilter/nf_nat_standalone.c | 406 ++++++++++++++++ net/netfilter/nf_conntrack_core.c | 23 +- net/netfilter/nf_conntrack_netlink.c | 46 +- net/netfilter/nf_conntrack_proto_tcp.c | 2 +- net/netfilter/nf_conntrack_standalone.c | 5 +- 29 files changed, 2667 insertions(+), 59 deletions(-) create mode 100644 include/net/netfilter/nf_nat.h create mode 100644 include/net/netfilter/nf_nat_core.h create mode 100644 include/net/netfilter/nf_nat_helper.h create mode 100644 include/net/netfilter/nf_nat_protocol.h create mode 100644 include/net/netfilter/nf_nat_rule.h create mode 100644 net/ipv4/netfilter/nf_nat_core.c create mode 100644 net/ipv4/netfilter/nf_nat_helper.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_icmp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_tcp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_udp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_unknown.c create mode 100644 net/ipv4/netfilter/nf_nat_rule.c create mode 100644 net/ipv4/netfilter/nf_nat_standalone.c (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index f6f3fcbd70ed..d4c4c5120bc0 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -357,7 +357,7 @@ extern void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *); static inline void nf_nat_decode_session(struct sk_buff *skb, struct flowi *fl, int family) { -#ifdef CONFIG_IP_NF_NAT_NEEDED +#if defined(CONFIG_IP_NF_NAT_NEEDED) || defined(CONFIG_NF_NAT_NEEDED) void (*decodefn)(struct sk_buff *, struct flowi *); if (family == AF_INET && (decodefn = ip_nat_decode_session) != NULL) diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 34b453a81a63..a1c57ee0a4fa 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -9,29 +9,23 @@ #ifndef _NF_CONNTRACK_IPV4_H #define _NF_CONNTRACK_IPV4_H -#ifdef CONFIG_IP_NF_NAT_NEEDED -#include +#ifdef CONFIG_NF_NAT_NEEDED +#include /* per conntrack: nat application helper private data */ -union ip_conntrack_nat_help { +union nf_conntrack_nat_help { /* insert nat helper private data here */ }; -struct nf_conntrack_ipv4_nat { - struct ip_nat_info info; - union ip_conntrack_nat_help help; +struct nf_conn_nat { + struct nf_nat_info info; + union nf_conntrack_nat_help help; #if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \ defined(CONFIG_IP_NF_TARGET_MASQUERADE_MODULE) int masq_index; #endif }; -#endif /* CONFIG_IP_NF_NAT_NEEDED */ - -struct nf_conntrack_ipv4 { -#ifdef CONFIG_IP_NF_NAT_NEEDED - struct nf_conntrack_ipv4_nat *nat; -#endif -}; +#endif /* CONFIG_NF_NAT_NEEDED */ /* Returns new sk_buff, or NULL */ struct sk_buff * diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index b4beb8c799e5..9948af068688 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -264,18 +264,45 @@ nf_conntrack_unregister_cache(u_int32_t features); /* valid combinations: * basic: nf_conn, nf_conn .. nf_conn_help - * nat: nf_conn .. nf_conn_nat, nf_conn .. nf_conn_nat, nf_conn help + * nat: nf_conn .. nf_conn_nat, nf_conn .. nf_conn_nat .. nf_conn help */ +#ifdef CONFIG_NF_NAT_NEEDED +static inline struct nf_conn_nat *nfct_nat(const struct nf_conn *ct) +{ + unsigned int offset = sizeof(struct nf_conn); + + if (!(ct->features & NF_CT_F_NAT)) + return NULL; + + offset = ALIGN(offset, __alignof__(struct nf_conn_nat)); + return (struct nf_conn_nat *) ((void *)ct + offset); +} + static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct) { unsigned int offset = sizeof(struct nf_conn); if (!(ct->features & NF_CT_F_HELP)) return NULL; + if (ct->features & NF_CT_F_NAT) { + offset = ALIGN(offset, __alignof__(struct nf_conn_nat)); + offset += sizeof(struct nf_conn_nat); + } offset = ALIGN(offset, __alignof__(struct nf_conn_help)); return (struct nf_conn_help *) ((void *)ct + offset); } +#else /* No NAT */ +static inline struct nf_conn_help *nfct_help(const struct nf_conn *ct) +{ + unsigned int offset = sizeof(struct nf_conn); + + if (!(ct->features & NF_CT_F_HELP)) + return NULL; + offset = ALIGN(offset, __alignof__(struct nf_conn_help)); + return (struct nf_conn_help *) ((void *)ct + offset); +} +#endif /* CONFIG_NF_NAT_NEEDED */ #endif /* __KERNEL__ */ #endif /* _NF_CONNTRACK_H */ diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 5d853e826d1d..b969c430b36a 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -43,7 +43,7 @@ struct nf_conntrack_expect #ifdef CONFIG_NF_NAT_NEEDED /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ - union nf_conntrack_manip_proto saved_proto; + union nf_conntrack_man_proto saved_proto; /* Direction relative to the master connection. */ enum ip_conntrack_dir dir; #endif diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h new file mode 100644 index 000000000000..61c62068ca6b --- /dev/null +++ b/include/net/netfilter/nf_nat.h @@ -0,0 +1,77 @@ +#ifndef _NF_NAT_H +#define _NF_NAT_H +#include +#include + +#define NF_NAT_MAPPING_TYPE_MAX_NAMELEN 16 + +enum nf_nat_manip_type +{ + IP_NAT_MANIP_SRC, + IP_NAT_MANIP_DST +}; + +/* SRC manip occurs POST_ROUTING or LOCAL_IN */ +#define HOOK2MANIP(hooknum) ((hooknum) != NF_IP_POST_ROUTING && (hooknum) != NF_IP_LOCAL_IN) + +#define IP_NAT_RANGE_MAP_IPS 1 +#define IP_NAT_RANGE_PROTO_SPECIFIED 2 + +/* NAT sequence number modifications */ +struct nf_nat_seq { + /* position of the last TCP sequence number modification (if any) */ + u_int32_t correction_pos; + + /* sequence number offset before and after last modification */ + int16_t offset_before, offset_after; +}; + +/* Single range specification. */ +struct nf_nat_range +{ + /* Set to OR of flags above. */ + unsigned int flags; + + /* Inclusive: network order. */ + __be32 min_ip, max_ip; + + /* Inclusive: network order */ + union nf_conntrack_man_proto min, max; +}; + +/* For backwards compat: don't use in modern code. */ +struct nf_nat_multi_range_compat +{ + unsigned int rangesize; /* Must be 1. */ + + /* hangs off end. */ + struct nf_nat_range range[1]; +}; + +#ifdef __KERNEL__ +#include + +/* The structure embedded in the conntrack structure. */ +struct nf_nat_info +{ + struct list_head bysource; + struct nf_nat_seq seq[IP_CT_DIR_MAX]; +}; + +struct nf_conn; + +/* Set up the info structure to map into this range. */ +extern unsigned int nf_nat_setup_info(struct nf_conn *ct, + const struct nf_nat_range *range, + unsigned int hooknum); + +/* Is this tuple already taken? (not by us)*/ +extern int nf_nat_used_tuple(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack); + +extern int nf_nat_module_is_loaded; + +#else /* !__KERNEL__: iptables wants this to compile. */ +#define nf_nat_multi_range nf_nat_multi_range_compat +#endif /*__KERNEL__*/ +#endif diff --git a/include/net/netfilter/nf_nat_core.h b/include/net/netfilter/nf_nat_core.h new file mode 100644 index 000000000000..9778ffa93440 --- /dev/null +++ b/include/net/netfilter/nf_nat_core.h @@ -0,0 +1,27 @@ +#ifndef _NF_NAT_CORE_H +#define _NF_NAT_CORE_H +#include +#include + +/* This header used to share core functionality between the standalone + NAT module, and the compatibility layer's use of NAT for masquerading. */ + +extern unsigned int nf_nat_packet(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb); + +extern int nf_nat_icmp_reply_translation(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb); + +static inline int nf_nat_initialized(struct nf_conn *ct, + enum nf_nat_manip_type manip) +{ + if (manip == IP_NAT_MANIP_SRC) + return test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status); + else + return test_bit(IPS_DST_NAT_DONE_BIT, &ct->status); +} +#endif /* _NF_NAT_CORE_H */ diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h new file mode 100644 index 000000000000..ec98ecf95fc8 --- /dev/null +++ b/include/net/netfilter/nf_nat_helper.h @@ -0,0 +1,32 @@ +#ifndef _NF_NAT_HELPER_H +#define _NF_NAT_HELPER_H +/* NAT protocol helper routines. */ + +#include + +struct sk_buff; + +/* These return true or false. */ +extern int nf_nat_mangle_tcp_packet(struct sk_buff **skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len); +extern int nf_nat_mangle_udp_packet(struct sk_buff **skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len); +extern int nf_nat_seq_adjust(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo); + +/* Setup NAT on this expected conntrack so it follows master, but goes + * to port ct->master->saved_proto. */ +extern void nf_nat_follow_master(struct nf_conn *ct, + struct nf_conntrack_expect *this); +#endif diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h new file mode 100644 index 000000000000..a9ec5ef61468 --- /dev/null +++ b/include/net/netfilter/nf_nat_protocol.h @@ -0,0 +1,70 @@ +/* Header for use in defining a given protocol. */ +#ifndef _NF_NAT_PROTOCOL_H +#define _NF_NAT_PROTOCOL_H +#include +#include + +struct nf_nat_range; + +struct nf_nat_protocol +{ + /* Protocol name */ + const char *name; + + /* Protocol number. */ + unsigned int protonum; + + struct module *me; + + /* Translate a packet to the target according to manip type. + Return true if succeeded. */ + int (*manip_pkt)(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype); + + /* Is the manipable part of the tuple between min and max incl? */ + int (*in_range)(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max); + + /* Alter the per-proto part of the tuple (depending on + maniptype), to give a unique tuple in the given range if + possible; return false if not. Per-protocol part of tuple + is initialized to the incoming packet. */ + int (*unique_tuple)(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct); + + int (*range_to_nfattr)(struct sk_buff *skb, + const struct nf_nat_range *range); + + int (*nfattr_to_range)(struct nfattr *tb[], + struct nf_nat_range *range); +}; + +/* Protocol registration. */ +extern int nf_nat_protocol_register(struct nf_nat_protocol *proto); +extern void nf_nat_protocol_unregister(struct nf_nat_protocol *proto); + +extern struct nf_nat_protocol *nf_nat_proto_find_get(u_int8_t protocol); +extern void nf_nat_proto_put(struct nf_nat_protocol *proto); + +/* Built-in protocols. */ +extern struct nf_nat_protocol nf_nat_protocol_tcp; +extern struct nf_nat_protocol nf_nat_protocol_udp; +extern struct nf_nat_protocol nf_nat_protocol_icmp; +extern struct nf_nat_protocol nf_nat_unknown_protocol; + +extern int init_protocols(void) __init; +extern void cleanup_protocols(void); +extern struct nf_nat_protocol *find_nat_proto(u_int16_t protonum); + +extern int nf_nat_port_range_to_nfattr(struct sk_buff *skb, + const struct nf_nat_range *range); +extern int nf_nat_port_nfattr_to_range(struct nfattr *tb[], + struct nf_nat_range *range); + +#endif /*_NF_NAT_PROTO_H*/ diff --git a/include/net/netfilter/nf_nat_rule.h b/include/net/netfilter/nf_nat_rule.h new file mode 100644 index 000000000000..f191c672bcc6 --- /dev/null +++ b/include/net/netfilter/nf_nat_rule.h @@ -0,0 +1,35 @@ +#ifndef _NF_NAT_RULE_H +#define _NF_NAT_RULE_H +#include +#include +#include + +/* Compatibility definitions for ipt_FOO modules */ +#define ip_nat_range nf_nat_range +#define ip_conntrack_tuple nf_conntrack_tuple +#define ip_conntrack_get nf_ct_get +#define ip_conntrack nf_conn +#define ip_nat_setup_info nf_nat_setup_info +#define ip_nat_multi_range_compat nf_nat_multi_range_compat +#define ip_ct_iterate_cleanup nf_ct_iterate_cleanup +#define IP_NF_ASSERT NF_CT_ASSERT + +extern int nf_nat_rule_init(void) __init; +extern void nf_nat_rule_cleanup(void); +extern int nf_nat_rule_find(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + struct nf_conn *ct, + struct nf_nat_info *info); + +extern unsigned int +alloc_null_binding(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum); + +extern unsigned int +alloc_null_binding_confirmed(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum); +#endif /* _NF_NAT_RULE_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 71485276b819..01789aeaeb5f 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -6,7 +6,7 @@ menu "IP: Netfilter Configuration" depends on INET && NETFILTER config NF_CONNTRACK_IPV4 - tristate "IPv4 support for new connection tracking (EXPERIMENTAL)" + tristate "IPv4 support for new connection tracking (required for NAT) (EXPERIMENTAL)" depends on EXPERIMENTAL && NF_CONNTRACK ---help--- Connection tracking keeps a record of what packets have passed @@ -387,7 +387,7 @@ config IP_NF_TARGET_TCPMSS To compile it as a module, choose M here. If unsure, say N. -# NAT + specific targets +# NAT + specific targets: ip_conntrack config IP_NF_NAT tristate "Full NAT" depends on IP_NF_IPTABLES && IP_NF_CONNTRACK @@ -398,14 +398,30 @@ config IP_NF_NAT To compile it as a module, choose M here. If unsure, say N. +# NAT + specific targets: nf_conntrack +config NF_NAT + tristate "Full NAT" + depends on IP_NF_IPTABLES && NF_CONNTRACK + help + The Full NAT option allows masquerading, port forwarding and other + forms of full Network Address Port Translation. It is controlled by + the `nat' table in iptables: see the man page for iptables(8). + + To compile it as a module, choose M here. If unsure, say N. + config IP_NF_NAT_NEEDED bool - depends on IP_NF_NAT != n + depends on IP_NF_NAT + default y + +config NF_NAT_NEEDED + bool + depends on NF_NAT default y config IP_NF_TARGET_MASQUERADE tristate "MASQUERADE target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help Masquerading is a special case of NAT: all outgoing connections are changed to seem to come from a particular interface's address, and @@ -417,7 +433,7 @@ config IP_NF_TARGET_MASQUERADE config IP_NF_TARGET_REDIRECT tristate "REDIRECT target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help REDIRECT is a special case of NAT: all incoming connections are mapped onto the incoming interface's address, causing the packets to @@ -428,7 +444,7 @@ config IP_NF_TARGET_REDIRECT config IP_NF_TARGET_NETMAP tristate "NETMAP target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help NETMAP is an implementation of static 1:1 NAT mapping of network addresses. It maps the network address part, while keeping the host @@ -439,7 +455,7 @@ config IP_NF_TARGET_NETMAP config IP_NF_TARGET_SAME tristate "SAME target support" - depends on IP_NF_NAT + depends on (NF_NAT || IP_NF_NAT) help This option adds a `SAME' target, which works like the standard SNAT target, but attempts to give clients the same IP for all connections. diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 21359d83f0c7..ec31690764ac 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -5,7 +5,12 @@ # objects for the standalone - connection tracking / NAT ip_conntrack-objs := ip_conntrack_standalone.o ip_conntrack_core.o ip_conntrack_proto_generic.o ip_conntrack_proto_tcp.o ip_conntrack_proto_udp.o ip_conntrack_proto_icmp.o ip_nat-objs := ip_nat_core.o ip_nat_helper.o ip_nat_proto_unknown.o ip_nat_proto_tcp.o ip_nat_proto_udp.o ip_nat_proto_icmp.o +nf_nat-objs := nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o +ifneq ($(CONFIG_NF_NAT),) +iptable_nat-objs := nf_nat_rule.o nf_nat_standalone.o +else iptable_nat-objs := ip_nat_rule.o ip_nat_standalone.o +endif ip_conntrack_pptp-objs := ip_conntrack_helper_pptp.o ip_conntrack_proto_gre.o ip_nat_pptp-objs := ip_nat_helper_pptp.o ip_nat_proto_gre.o @@ -16,6 +21,7 @@ ip_nat_h323-objs := ip_nat_helper_h323.o # connection tracking obj-$(CONFIG_IP_NF_CONNTRACK) += ip_conntrack.o obj-$(CONFIG_IP_NF_NAT) += ip_nat.o +obj-$(CONFIG_NF_NAT) += nf_nat.o # conntrack netlink interface obj-$(CONFIG_IP_NF_CONNTRACK_NETLINK) += ip_conntrack_netlink.o @@ -50,6 +56,7 @@ obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o obj-$(CONFIG_IP_NF_FILTER) += iptable_filter.o obj-$(CONFIG_IP_NF_MANGLE) += iptable_mangle.o obj-$(CONFIG_IP_NF_NAT) += iptable_nat.o +obj-$(CONFIG_NF_NAT) += iptable_nat.o obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o # matches diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index d85d2de50449..ad66328baa5d 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -44,12 +44,6 @@ #define DEBUGP(format, args...) #endif -#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \ - : ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \ - : ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \ - : ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \ - : "*ERROR*"))) - #ifdef CONFIG_XFRM static void nat_decode_session(struct sk_buff *skb, struct flowi *fl) { diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index 3dbfcfac8a84..28b9233956b5 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -2,7 +2,7 @@ (depending on route). */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team + * (C) 2002-2006 Netfilter Core Team * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -20,7 +20,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #include MODULE_LICENSE("GPL"); @@ -65,23 +69,33 @@ masquerade_target(struct sk_buff **pskb, const struct xt_target *target, const void *targinfo) { +#ifdef CONFIG_NF_NAT_NEEDED + struct nf_conn_nat *nat; +#endif struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; - const struct ip_nat_multi_range_compat *mr; struct ip_nat_range newrange; + const struct ip_nat_multi_range_compat *mr; struct rtable *rt; __be32 newsrc; IP_NF_ASSERT(hooknum == NF_IP_POST_ROUTING); ct = ip_conntrack_get(*pskb, &ctinfo); +#ifdef CONFIG_NF_NAT_NEEDED + nat = nfct_nat(ct); +#endif IP_NF_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED || ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)); /* Source address is 0.0.0.0 - locally generated packet that is * probably not supposed to be masqueraded. */ +#ifdef CONFIG_NF_NAT_NEEDED + if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip == 0) +#else if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip == 0) +#endif return NF_ACCEPT; mr = targinfo; @@ -93,7 +107,11 @@ masquerade_target(struct sk_buff **pskb, } write_lock_bh(&masq_lock); +#ifdef CONFIG_NF_NAT_NEEDED + nat->masq_index = out->ifindex; +#else ct->nat.masq_index = out->ifindex; +#endif write_unlock_bh(&masq_lock); /* Transfer from original range. */ @@ -109,10 +127,17 @@ masquerade_target(struct sk_buff **pskb, static inline int device_cmp(struct ip_conntrack *i, void *ifindex) { +#ifdef CONFIG_NF_NAT_NEEDED + struct nf_conn_nat *nat = nfct_nat(i); +#endif int ret; read_lock_bh(&masq_lock); +#ifdef CONFIG_NF_NAT_NEEDED + ret = (nat->masq_index == (int)(long)ifindex); +#else ret = (i->nat.masq_index == (int)(long)ifindex); +#endif read_unlock_bh(&masq_lock); return ret; diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c index 58a88f227108..9390e90f2b25 100644 --- a/net/ipv4/netfilter/ipt_NETMAP.c +++ b/net/ipv4/netfilter/ipt_NETMAP.c @@ -15,7 +15,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #define MODULENAME "NETMAP" MODULE_LICENSE("GPL"); diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c index c0dcfe9d610c..462eceb3a1b1 100644 --- a/net/ipv4/netfilter/ipt_REDIRECT.c +++ b/net/ipv4/netfilter/ipt_REDIRECT.c @@ -1,6 +1,6 @@ /* Redirect. Simple mapping which alters dst to a local IP address. */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team + * (C) 2002-2006 Netfilter Core Team * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -18,7 +18,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c index b38b13328d73..3dcf29411337 100644 --- a/net/ipv4/netfilter/ipt_SAME.c +++ b/net/ipv4/netfilter/ipt_SAME.c @@ -34,7 +34,11 @@ #include #include #include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#else #include +#endif #include MODULE_LICENSE("GPL"); @@ -152,11 +156,17 @@ same_target(struct sk_buff **pskb, Here we calculate the index in same->iparray which holds the ipaddress we should use */ +#ifdef CONFIG_NF_NAT_NEEDED + tmpip = ntohl(t->src.u3.ip); + + if (!(same->info & IPT_SAME_NODST)) + tmpip += ntohl(t->dst.u3.ip); +#else tmpip = ntohl(t->src.ip); if (!(same->info & IPT_SAME_NODST)) tmpip += ntohl(t->dst.ip); - +#endif aindex = tmpip % same->ipnum; new_ip = htonl(same->iparray[aindex]); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 653f57b8a107..5655109dcaff 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -111,10 +111,10 @@ ipv4_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, return NF_ACCEPT; } -int nat_module_is_loaded = 0; +int nf_nat_module_is_loaded = 0; static u_int32_t ipv4_get_features(const struct nf_conntrack_tuple *tuple) { - if (nat_module_is_loaded) + if (nf_nat_module_is_loaded) return NF_CT_F_NAT; return NF_CT_F_BASIC; @@ -532,3 +532,6 @@ module_init(nf_conntrack_l3proto_ipv4_init); module_exit(nf_conntrack_l3proto_ipv4_fini); EXPORT_SYMBOL(nf_ct_ipv4_gather_frags); +#ifdef CONFIG_NF_NAT_NEEDED +EXPORT_SYMBOL(nf_nat_module_is_loaded); +#endif diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c new file mode 100644 index 000000000000..86a92272b053 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -0,0 +1,647 @@ +/* NAT for netfilter; shared with compatibility layer. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include /* For tcp_prot in getorigdst */ +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static DEFINE_RWLOCK(nf_nat_lock); + +static struct nf_conntrack_l3proto *l3proto = NULL; + +/* Calculated at init based on memory size */ +static unsigned int nf_nat_htable_size; + +static struct list_head *bysource; + +#define MAX_IP_NAT_PROTO 256 +static struct nf_nat_protocol *nf_nat_protos[MAX_IP_NAT_PROTO]; + +static inline struct nf_nat_protocol * +__nf_nat_proto_find(u_int8_t protonum) +{ + return nf_nat_protos[protonum]; +} + +struct nf_nat_protocol * +nf_nat_proto_find_get(u_int8_t protonum) +{ + struct nf_nat_protocol *p; + + /* we need to disable preemption to make sure 'p' doesn't get + * removed until we've grabbed the reference */ + preempt_disable(); + p = __nf_nat_proto_find(protonum); + if (!try_module_get(p->me)) + p = &nf_nat_unknown_protocol; + preempt_enable(); + + return p; +} +EXPORT_SYMBOL_GPL(nf_nat_proto_find_get); + +void +nf_nat_proto_put(struct nf_nat_protocol *p) +{ + module_put(p->me); +} +EXPORT_SYMBOL_GPL(nf_nat_proto_put); + +/* We keep an extra hash for each conntrack, for fast searching. */ +static inline unsigned int +hash_by_src(const struct nf_conntrack_tuple *tuple) +{ + /* Original src, to ensure we map it consistently if poss. */ + return jhash_3words((__force u32)tuple->src.u3.ip, tuple->src.u.all, + tuple->dst.protonum, 0) % nf_nat_htable_size; +} + +/* Noone using conntrack by the time this called. */ +static void nf_nat_cleanup_conntrack(struct nf_conn *conn) +{ + struct nf_conn_nat *nat; + if (!(conn->status & IPS_NAT_DONE_MASK)) + return; + + nat = nfct_nat(conn); + write_lock_bh(&nf_nat_lock); + list_del(&nat->info.bysource); + write_unlock_bh(&nf_nat_lock); +} + +/* Is this tuple already taken? (not by us) */ +int +nf_nat_used_tuple(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + /* Conntrack tracking doesn't keep track of outgoing tuples; only + incoming ones. NAT means they don't have a fixed mapping, + so we invert the tuple and look for the incoming reply. + + We could keep a separate hash if this proves too slow. */ + struct nf_conntrack_tuple reply; + + nf_ct_invert_tuplepr(&reply, tuple); + return nf_conntrack_tuple_taken(&reply, ignored_conntrack); +} +EXPORT_SYMBOL(nf_nat_used_tuple); + +/* If we source map this tuple so reply looks like reply_tuple, will + * that meet the constraints of range. */ +static int +in_range(const struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range) +{ + struct nf_nat_protocol *proto; + + proto = __nf_nat_proto_find(tuple->dst.protonum); + /* If we are supposed to map IPs, then we must be in the + range specified, otherwise let this drag us onto a new src IP. */ + if (range->flags & IP_NAT_RANGE_MAP_IPS) { + if (ntohl(tuple->src.u3.ip) < ntohl(range->min_ip) || + ntohl(tuple->src.u3.ip) > ntohl(range->max_ip)) + return 0; + } + + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) || + proto->in_range(tuple, IP_NAT_MANIP_SRC, + &range->min, &range->max)) + return 1; + + return 0; +} + +static inline int +same_src(const struct nf_conn *ct, + const struct nf_conntrack_tuple *tuple) +{ + const struct nf_conntrack_tuple *t; + + t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + return (t->dst.protonum == tuple->dst.protonum && + t->src.u3.ip == tuple->src.u3.ip && + t->src.u.all == tuple->src.u.all); +} + +/* Only called for SRC manip */ +static int +find_appropriate_src(const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_tuple *result, + const struct nf_nat_range *range) +{ + unsigned int h = hash_by_src(tuple); + struct nf_conn_nat *nat; + struct nf_conn *ct; + + read_lock_bh(&nf_nat_lock); + list_for_each_entry(nat, &bysource[h], info.bysource) { + ct = (struct nf_conn *)((char *)nat - offsetof(struct nf_conn, data)); + if (same_src(ct, tuple)) { + /* Copy source part from reply tuple. */ + nf_ct_invert_tuplepr(result, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + result->dst = tuple->dst; + + if (in_range(result, range)) { + read_unlock_bh(&nf_nat_lock); + return 1; + } + } + } + read_unlock_bh(&nf_nat_lock); + return 0; +} + +/* For [FUTURE] fragmentation handling, we want the least-used + src-ip/dst-ip/proto triple. Fairness doesn't come into it. Thus + if the range specifies 1.2.3.4 ports 10000-10005 and 1.2.3.5 ports + 1-65535, we don't do pro-rata allocation based on ports; we choose + the ip with the lowest src-ip/dst-ip/proto usage. +*/ +static void +find_best_ips_proto(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + const struct nf_conn *ct, + enum nf_nat_manip_type maniptype) +{ + __be32 *var_ipp; + /* Host order */ + u_int32_t minip, maxip, j; + + /* No IP mapping? Do nothing. */ + if (!(range->flags & IP_NAT_RANGE_MAP_IPS)) + return; + + if (maniptype == IP_NAT_MANIP_SRC) + var_ipp = &tuple->src.u3.ip; + else + var_ipp = &tuple->dst.u3.ip; + + /* Fast path: only one choice. */ + if (range->min_ip == range->max_ip) { + *var_ipp = range->min_ip; + return; + } + + /* Hashing source and destination IPs gives a fairly even + * spread in practice (if there are a small number of IPs + * involved, there usually aren't that many connections + * anyway). The consistency means that servers see the same + * client coming from the same IP (some Internet Banking sites + * like this), even across reboots. */ + minip = ntohl(range->min_ip); + maxip = ntohl(range->max_ip); + j = jhash_2words((__force u32)tuple->src.u3.ip, + (__force u32)tuple->dst.u3.ip, 0); + *var_ipp = htonl(minip + j % (maxip - minip + 1)); +} + +/* Manipulate the tuple into the range given. For NF_IP_POST_ROUTING, + * we change the source to map into the range. For NF_IP_PRE_ROUTING + * and NF_IP_LOCAL_OUT, we change the destination to map into the + * range. It might not be possible to get a unique tuple, but we try. + * At worst (or if we race), we will end up with a final duplicate in + * __ip_conntrack_confirm and drop the packet. */ +static void +get_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig_tuple, + const struct nf_nat_range *range, + struct nf_conn *ct, + enum nf_nat_manip_type maniptype) +{ + struct nf_nat_protocol *proto; + + /* 1) If this srcip/proto/src-proto-part is currently mapped, + and that same mapping gives a unique tuple within the given + range, use that. + + This is only required for source (ie. NAT/masq) mappings. + So far, we don't do local source mappings, so multiple + manips not an issue. */ + if (maniptype == IP_NAT_MANIP_SRC) { + if (find_appropriate_src(orig_tuple, tuple, range)) { + DEBUGP("get_unique_tuple: Found current src map\n"); + if (!nf_nat_used_tuple(tuple, ct)) + return; + } + } + + /* 2) Select the least-used IP/proto combination in the given + range. */ + *tuple = *orig_tuple; + find_best_ips_proto(tuple, range, ct, maniptype); + + /* 3) The per-protocol part of the manip is made to map into + the range to make a unique tuple. */ + + proto = nf_nat_proto_find_get(orig_tuple->dst.protonum); + + /* Only bother mapping if it's not already in range and unique */ + if ((!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED) || + proto->in_range(tuple, maniptype, &range->min, &range->max)) && + !nf_nat_used_tuple(tuple, ct)) { + nf_nat_proto_put(proto); + return; + } + + /* Last change: get protocol to try to obtain unique tuple. */ + proto->unique_tuple(tuple, range, maniptype, ct); + + nf_nat_proto_put(proto); +} + +unsigned int +nf_nat_setup_info(struct nf_conn *ct, + const struct nf_nat_range *range, + unsigned int hooknum) +{ + struct nf_conntrack_tuple curr_tuple, new_tuple; + struct nf_conn_nat *nat = nfct_nat(ct); + struct nf_nat_info *info = &nat->info; + int have_to_hash = !(ct->status & IPS_NAT_DONE_MASK); + enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum); + + NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING || + hooknum == NF_IP_POST_ROUTING || + hooknum == NF_IP_LOCAL_IN || + hooknum == NF_IP_LOCAL_OUT); + BUG_ON(nf_nat_initialized(ct, maniptype)); + + /* What we've got will look like inverse of reply. Normally + this is what is in the conntrack, except for prior + manipulations (future optimization: if num_manips == 0, + orig_tp = + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple) */ + nf_ct_invert_tuplepr(&curr_tuple, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + get_unique_tuple(&new_tuple, &curr_tuple, range, ct, maniptype); + + if (!nf_ct_tuple_equal(&new_tuple, &curr_tuple)) { + struct nf_conntrack_tuple reply; + + /* Alter conntrack table so will recognize replies. */ + nf_ct_invert_tuplepr(&reply, &new_tuple); + nf_conntrack_alter_reply(ct, &reply); + + /* Non-atomic: we own this at the moment. */ + if (maniptype == IP_NAT_MANIP_SRC) + ct->status |= IPS_SRC_NAT; + else + ct->status |= IPS_DST_NAT; + } + + /* Place in source hash if this is the first time. */ + if (have_to_hash) { + unsigned int srchash; + + srchash = hash_by_src(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + write_lock_bh(&nf_nat_lock); + list_add(&info->bysource, &bysource[srchash]); + write_unlock_bh(&nf_nat_lock); + } + + /* It's done. */ + if (maniptype == IP_NAT_MANIP_DST) + set_bit(IPS_DST_NAT_DONE_BIT, &ct->status); + else + set_bit(IPS_SRC_NAT_DONE_BIT, &ct->status); + + return NF_ACCEPT; +} +EXPORT_SYMBOL(nf_nat_setup_info); + +/* Returns true if succeeded. */ +static int +manip_pkt(u_int16_t proto, + struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *target, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph; + struct nf_nat_protocol *p; + + if (!skb_make_writable(pskb, iphdroff + sizeof(*iph))) + return 0; + + iph = (void *)(*pskb)->data + iphdroff; + + /* Manipulate protcol part. */ + p = nf_nat_proto_find_get(proto); + if (!p->manip_pkt(pskb, iphdroff, target, maniptype)) { + nf_nat_proto_put(p); + return 0; + } + nf_nat_proto_put(p); + + iph = (void *)(*pskb)->data + iphdroff; + + if (maniptype == IP_NAT_MANIP_SRC) { + nf_csum_replace4(&iph->check, iph->saddr, target->src.u3.ip); + iph->saddr = target->src.u3.ip; + } else { + nf_csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip); + iph->daddr = target->dst.u3.ip; + } + return 1; +} + +/* Do packet manipulations according to nf_nat_setup_info. */ +unsigned int nf_nat_packet(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned long statusbit; + enum nf_nat_manip_type mtype = HOOK2MANIP(hooknum); + + if (mtype == IP_NAT_MANIP_SRC) + statusbit = IPS_SRC_NAT; + else + statusbit = IPS_DST_NAT; + + /* Invert if this is reply dir. */ + if (dir == IP_CT_DIR_REPLY) + statusbit ^= IPS_NAT_MASK; + + /* Non-atomic: these bits don't change. */ + if (ct->status & statusbit) { + struct nf_conntrack_tuple target; + + /* We are aiming to look like inverse of other direction. */ + nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); + + if (!manip_pkt(target.dst.protonum, pskb, 0, &target, mtype)) + return NF_DROP; + } + return NF_ACCEPT; +} +EXPORT_SYMBOL_GPL(nf_nat_packet); + +/* Dir is direction ICMP is coming from (opposite to packet it contains) */ +int nf_nat_icmp_reply_translation(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int hooknum, + struct sk_buff **pskb) +{ + struct { + struct icmphdr icmp; + struct iphdr ip; + } *inside; + struct nf_conntrack_tuple inner, target; + int hdrlen = (*pskb)->nh.iph->ihl * 4; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned long statusbit; + enum nf_nat_manip_type manip = HOOK2MANIP(hooknum); + + if (!skb_make_writable(pskb, hdrlen + sizeof(*inside))) + return 0; + + inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + + /* We're actually going to mangle it beyond trivial checksum + adjustment, so make sure the current checksum is correct. */ + if (nf_ip_checksum(*pskb, hooknum, hdrlen, 0)) + return 0; + + /* Must be RELATED */ + NF_CT_ASSERT((*pskb)->nfctinfo == IP_CT_RELATED || + (*pskb)->nfctinfo == IP_CT_RELATED+IP_CT_IS_REPLY); + + /* Redirects on non-null nats must be dropped, else they'll + start talking to each other without our translation, and be + confused... --RR */ + if (inside->icmp.type == ICMP_REDIRECT) { + /* If NAT isn't finished, assume it and drop. */ + if ((ct->status & IPS_NAT_DONE_MASK) != IPS_NAT_DONE_MASK) + return 0; + + if (ct->status & IPS_NAT_MASK) + return 0; + } + + DEBUGP("icmp_reply_translation: translating error %p manp %u dir %s\n", + *pskb, manip, dir == IP_CT_DIR_ORIGINAL ? "ORIG" : "REPLY"); + + if (!nf_ct_get_tuple(*pskb, + (*pskb)->nh.iph->ihl*4 + sizeof(struct icmphdr), + (*pskb)->nh.iph->ihl*4 + + sizeof(struct icmphdr) + inside->ip.ihl*4, + (u_int16_t)AF_INET, + inside->ip.protocol, + &inner, + l3proto, + __nf_ct_l4proto_find((u_int16_t)PF_INET, + inside->ip.protocol))) + return 0; + + /* Change inner back to look like incoming packet. We do the + opposite manip on this hook to normal, because it might not + pass all hooks (locally-generated ICMP). Consider incoming + packet: PREROUTING (DST manip), routing produces ICMP, goes + through POSTROUTING (which must correct the DST manip). */ + if (!manip_pkt(inside->ip.protocol, pskb, + (*pskb)->nh.iph->ihl*4 + sizeof(inside->icmp), + &ct->tuplehash[!dir].tuple, + !manip)) + return 0; + + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + /* Reloading "inside" here since manip_pkt inner. */ + inside = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + inside->icmp.checksum = 0; + inside->icmp.checksum = + csum_fold(skb_checksum(*pskb, hdrlen, + (*pskb)->len - hdrlen, 0)); + } + + /* Change outer to look the reply to an incoming packet + * (proto 0 means don't invert per-proto part). */ + if (manip == IP_NAT_MANIP_SRC) + statusbit = IPS_SRC_NAT; + else + statusbit = IPS_DST_NAT; + + /* Invert if this is reply dir. */ + if (dir == IP_CT_DIR_REPLY) + statusbit ^= IPS_NAT_MASK; + + if (ct->status & statusbit) { + nf_ct_invert_tuplepr(&target, &ct->tuplehash[!dir].tuple); + if (!manip_pkt(0, pskb, 0, &target, manip)) + return 0; + } + + return 1; +} +EXPORT_SYMBOL_GPL(nf_nat_icmp_reply_translation); + +/* Protocol registration. */ +int nf_nat_protocol_register(struct nf_nat_protocol *proto) +{ + int ret = 0; + + write_lock_bh(&nf_nat_lock); + if (nf_nat_protos[proto->protonum] != &nf_nat_unknown_protocol) { + ret = -EBUSY; + goto out; + } + nf_nat_protos[proto->protonum] = proto; + out: + write_unlock_bh(&nf_nat_lock); + return ret; +} +EXPORT_SYMBOL(nf_nat_protocol_register); + +/* Noone stores the protocol anywhere; simply delete it. */ +void nf_nat_protocol_unregister(struct nf_nat_protocol *proto) +{ + write_lock_bh(&nf_nat_lock); + nf_nat_protos[proto->protonum] = &nf_nat_unknown_protocol; + write_unlock_bh(&nf_nat_lock); + + /* Someone could be still looking at the proto in a bh. */ + synchronize_net(); +} +EXPORT_SYMBOL(nf_nat_protocol_unregister); + +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) +int +nf_nat_port_range_to_nfattr(struct sk_buff *skb, + const struct nf_nat_range *range) +{ + NFA_PUT(skb, CTA_PROTONAT_PORT_MIN, sizeof(__be16), + &range->min.tcp.port); + NFA_PUT(skb, CTA_PROTONAT_PORT_MAX, sizeof(__be16), + &range->max.tcp.port); + + return 0; + +nfattr_failure: + return -1; +} +EXPORT_SYMBOL_GPL(nf_nat_port_nfattr_to_range); + +int +nf_nat_port_nfattr_to_range(struct nfattr *tb[], struct nf_nat_range *range) +{ + int ret = 0; + + /* we have to return whether we actually parsed something or not */ + + if (tb[CTA_PROTONAT_PORT_MIN-1]) { + ret = 1; + range->min.tcp.port = + *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MIN-1]); + } + + if (!tb[CTA_PROTONAT_PORT_MAX-1]) { + if (ret) + range->max.tcp.port = range->min.tcp.port; + } else { + ret = 1; + range->max.tcp.port = + *(__be16 *)NFA_DATA(tb[CTA_PROTONAT_PORT_MAX-1]); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nf_nat_port_range_to_nfattr); +#endif + +static int __init nf_nat_init(void) +{ + size_t i; + + /* Leave them the same for the moment. */ + nf_nat_htable_size = nf_conntrack_htable_size; + + /* One vmalloc for both hash tables */ + bysource = vmalloc(sizeof(struct list_head) * nf_nat_htable_size); + if (!bysource) + return -ENOMEM; + + /* Sew in builtin protocols. */ + write_lock_bh(&nf_nat_lock); + for (i = 0; i < MAX_IP_NAT_PROTO; i++) + nf_nat_protos[i] = &nf_nat_unknown_protocol; + nf_nat_protos[IPPROTO_TCP] = &nf_nat_protocol_tcp; + nf_nat_protos[IPPROTO_UDP] = &nf_nat_protocol_udp; + nf_nat_protos[IPPROTO_ICMP] = &nf_nat_protocol_icmp; + write_unlock_bh(&nf_nat_lock); + + for (i = 0; i < nf_nat_htable_size; i++) { + INIT_LIST_HEAD(&bysource[i]); + } + + /* FIXME: Man, this is a hack. */ + NF_CT_ASSERT(nf_conntrack_destroyed == NULL); + nf_conntrack_destroyed = &nf_nat_cleanup_conntrack; + + /* Initialize fake conntrack so that NAT will skip it */ + nf_conntrack_untracked.status |= IPS_NAT_DONE_MASK; + + l3proto = nf_ct_l3proto_find_get((u_int16_t)AF_INET); + return 0; +} + +/* Clear NAT section of all conntracks, in case we're loaded again. */ +static int clean_nat(struct nf_conn *i, void *data) +{ + struct nf_conn_nat *nat = nfct_nat(i); + + if (!nat) + return 0; + memset(nat, 0, sizeof(nat)); + i->status &= ~(IPS_NAT_MASK | IPS_NAT_DONE_MASK | IPS_SEQ_ADJUST); + return 0; +} + +static void __exit nf_nat_cleanup(void) +{ + nf_ct_iterate_cleanup(&clean_nat, NULL); + nf_conntrack_destroyed = NULL; + vfree(bysource); + nf_ct_l3proto_put(l3proto); +} + +MODULE_LICENSE("GPL"); + +module_init(nf_nat_init); +module_exit(nf_nat_cleanup); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c new file mode 100644 index 000000000000..98fbfc84d183 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -0,0 +1,433 @@ +/* ip_nat_helper.c - generic support functions for NAT helpers + * + * (C) 2000-2002 Harald Welte + * (C) 2003-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#define DUMP_OFFSET(x) printk("offset_before=%d, offset_after=%d, correction_pos=%u\n", x->offset_before, x->offset_after, x->correction_pos); +#else +#define DEBUGP(format, args...) +#define DUMP_OFFSET(x) +#endif + +static DEFINE_SPINLOCK(nf_nat_seqofs_lock); + +/* Setup TCP sequence correction given this change at this sequence */ +static inline void +adjust_tcp_sequence(u32 seq, + int sizediff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + int dir; + struct nf_nat_seq *this_way, *other_way; + struct nf_conn_nat *nat = nfct_nat(ct); + + DEBUGP("nf_nat_resize_packet: old_size = %u, new_size = %u\n", + (*skb)->len, new_size); + + dir = CTINFO2DIR(ctinfo); + + this_way = &nat->info.seq[dir]; + other_way = &nat->info.seq[!dir]; + + DEBUGP("nf_nat_resize_packet: Seq_offset before: "); + DUMP_OFFSET(this_way); + + spin_lock_bh(&nf_nat_seqofs_lock); + + /* SYN adjust. If it's uninitialized, or this is after last + * correction, record it: we don't handle more than one + * adjustment in the window, but do deal with common case of a + * retransmit */ + if (this_way->offset_before == this_way->offset_after || + before(this_way->correction_pos, seq)) { + this_way->correction_pos = seq; + this_way->offset_before = this_way->offset_after; + this_way->offset_after += sizediff; + } + spin_unlock_bh(&nf_nat_seqofs_lock); + + DEBUGP("nf_nat_resize_packet: Seq_offset after: "); + DUMP_OFFSET(this_way); +} + +/* Frobs data inside this packet, which is linear. */ +static void mangle_contents(struct sk_buff *skb, + unsigned int dataoff, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + unsigned char *data; + + BUG_ON(skb_is_nonlinear(skb)); + data = (unsigned char *)skb->nh.iph + dataoff; + + /* move post-replacement */ + memmove(data + match_offset + rep_len, + data + match_offset + match_len, + skb->tail - (data + match_offset + match_len)); + + /* insert data from buffer */ + memcpy(data + match_offset, rep_buffer, rep_len); + + /* update skb info */ + if (rep_len > match_len) { + DEBUGP("nf_nat_mangle_packet: Extending packet by " + "%u from %u bytes\n", rep_len - match_len, + skb->len); + skb_put(skb, rep_len - match_len); + } else { + DEBUGP("nf_nat_mangle_packet: Shrinking packet from " + "%u from %u bytes\n", match_len - rep_len, + skb->len); + __skb_trim(skb, skb->len + rep_len - match_len); + } + + /* fix IP hdr checksum information */ + skb->nh.iph->tot_len = htons(skb->len); + ip_send_check(skb->nh.iph); +} + +/* Unusual, but possible case. */ +static int enlarge_skb(struct sk_buff **pskb, unsigned int extra) +{ + struct sk_buff *nskb; + + if ((*pskb)->len + extra > 65535) + return 0; + + nskb = skb_copy_expand(*pskb, skb_headroom(*pskb), extra, GFP_ATOMIC); + if (!nskb) + return 0; + + /* Transfer socket to new skb. */ + if ((*pskb)->sk) + skb_set_owner_w(nskb, (*pskb)->sk); + kfree_skb(*pskb); + *pskb = nskb; + return 1; +} + +/* Generic function for mangling variable-length address changes inside + * NATed TCP connections (like the PORT XXX,XXX,XXX,XXX,XXX,XXX + * command in FTP). + * + * Takes care about all the nasty sequence number changes, checksumming, + * skb enlargement, ... + * + * */ +int +nf_nat_mangle_tcp_packet(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + struct iphdr *iph; + struct tcphdr *tcph; + int oldlen, datalen; + + if (!skb_make_writable(pskb, (*pskb)->len)) + return 0; + + if (rep_len > match_len && + rep_len - match_len > skb_tailroom(*pskb) && + !enlarge_skb(pskb, rep_len - match_len)) + return 0; + + SKB_LINEAR_ASSERT(*pskb); + + iph = (*pskb)->nh.iph; + tcph = (void *)iph + iph->ihl*4; + + oldlen = (*pskb)->len - iph->ihl*4; + mangle_contents(*pskb, iph->ihl*4 + tcph->doff*4, + match_offset, match_len, rep_buffer, rep_len); + + datalen = (*pskb)->len - iph->ihl*4; + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + tcph->check = 0; + tcph->check = tcp_v4_check(tcph, datalen, + iph->saddr, iph->daddr, + csum_partial((char *)tcph, + datalen, 0)); + } else + nf_proto_csum_replace2(&tcph->check, *pskb, + htons(oldlen), htons(datalen), 1); + + if (rep_len != match_len) { + set_bit(IPS_SEQ_ADJUST_BIT, &ct->status); + adjust_tcp_sequence(ntohl(tcph->seq), + (int)rep_len - (int)match_len, + ct, ctinfo); + /* Tell TCP window tracking about seq change */ + nf_conntrack_tcp_update(*pskb, (*pskb)->nh.iph->ihl*4, + ct, CTINFO2DIR(ctinfo)); + } + return 1; +} +EXPORT_SYMBOL(nf_nat_mangle_tcp_packet); + +/* Generic function for mangling variable-length address changes inside + * NATed UDP connections (like the CONNECT DATA XXXXX MESG XXXXX INDEX XXXXX + * command in the Amanda protocol) + * + * Takes care about all the nasty sequence number changes, checksumming, + * skb enlargement, ... + * + * XXX - This function could be merged with nf_nat_mangle_tcp_packet which + * should be fairly easy to do. + */ +int +nf_nat_mangle_udp_packet(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int match_offset, + unsigned int match_len, + const char *rep_buffer, + unsigned int rep_len) +{ + struct iphdr *iph; + struct udphdr *udph; + int datalen, oldlen; + + /* UDP helpers might accidentally mangle the wrong packet */ + iph = (*pskb)->nh.iph; + if ((*pskb)->len < iph->ihl*4 + sizeof(*udph) + + match_offset + match_len) + return 0; + + if (!skb_make_writable(pskb, (*pskb)->len)) + return 0; + + if (rep_len > match_len && + rep_len - match_len > skb_tailroom(*pskb) && + !enlarge_skb(pskb, rep_len - match_len)) + return 0; + + iph = (*pskb)->nh.iph; + udph = (void *)iph + iph->ihl*4; + + oldlen = (*pskb)->len - iph->ihl*4; + mangle_contents(*pskb, iph->ihl*4 + sizeof(*udph), + match_offset, match_len, rep_buffer, rep_len); + + /* update the length of the UDP packet */ + datalen = (*pskb)->len - iph->ihl*4; + udph->len = htons(datalen); + + /* fix udp checksum if udp checksum was previously calculated */ + if (!udph->check && (*pskb)->ip_summed != CHECKSUM_PARTIAL) + return 1; + + if ((*pskb)->ip_summed != CHECKSUM_PARTIAL) { + udph->check = 0; + udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr, + datalen, IPPROTO_UDP, + csum_partial((char *)udph, + datalen, 0)); + if (!udph->check) + udph->check = CSUM_MANGLED_0; + } else + nf_proto_csum_replace2(&udph->check, *pskb, + htons(oldlen), htons(datalen), 1); + + return 1; +} +EXPORT_SYMBOL(nf_nat_mangle_udp_packet); + +/* Adjust one found SACK option including checksum correction */ +static void +sack_adjust(struct sk_buff *skb, + struct tcphdr *tcph, + unsigned int sackoff, + unsigned int sackend, + struct nf_nat_seq *natseq) +{ + while (sackoff < sackend) { + struct tcp_sack_block_wire *sack; + __be32 new_start_seq, new_end_seq; + + sack = (void *)skb->data + sackoff; + if (after(ntohl(sack->start_seq) - natseq->offset_before, + natseq->correction_pos)) + new_start_seq = htonl(ntohl(sack->start_seq) + - natseq->offset_after); + else + new_start_seq = htonl(ntohl(sack->start_seq) + - natseq->offset_before); + + if (after(ntohl(sack->end_seq) - natseq->offset_before, + natseq->correction_pos)) + new_end_seq = htonl(ntohl(sack->end_seq) + - natseq->offset_after); + else + new_end_seq = htonl(ntohl(sack->end_seq) + - natseq->offset_before); + + DEBUGP("sack_adjust: start_seq: %d->%d, end_seq: %d->%d\n", + ntohl(sack->start_seq), new_start_seq, + ntohl(sack->end_seq), new_end_seq); + + nf_proto_csum_replace4(&tcph->check, skb, + sack->start_seq, new_start_seq, 0); + nf_proto_csum_replace4(&tcph->check, skb, + sack->end_seq, new_end_seq, 0); + sack->start_seq = new_start_seq; + sack->end_seq = new_end_seq; + sackoff += sizeof(*sack); + } +} + +/* TCP SACK sequence number adjustment */ +static inline unsigned int +nf_nat_sack_adjust(struct sk_buff **pskb, + struct tcphdr *tcph, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + unsigned int dir, optoff, optend; + struct nf_conn_nat *nat = nfct_nat(ct); + + optoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct tcphdr); + optend = (*pskb)->nh.iph->ihl*4 + tcph->doff*4; + + if (!skb_make_writable(pskb, optend)) + return 0; + + dir = CTINFO2DIR(ctinfo); + + while (optoff < optend) { + /* Usually: option, length. */ + unsigned char *op = (*pskb)->data + optoff; + + switch (op[0]) { + case TCPOPT_EOL: + return 1; + case TCPOPT_NOP: + optoff++; + continue; + default: + /* no partial options */ + if (optoff + 1 == optend || + optoff + op[1] > optend || + op[1] < 2) + return 0; + if (op[0] == TCPOPT_SACK && + op[1] >= 2+TCPOLEN_SACK_PERBLOCK && + ((op[1] - 2) % TCPOLEN_SACK_PERBLOCK) == 0) + sack_adjust(*pskb, tcph, optoff+2, + optoff+op[1], + &nat->info.seq[!dir]); + optoff += op[1]; + } + } + return 1; +} + +/* TCP sequence number adjustment. Returns 1 on success, 0 on failure */ +int +nf_nat_seq_adjust(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct tcphdr *tcph; + int dir; + __be32 newseq, newack; + struct nf_conn_nat *nat = nfct_nat(ct); + struct nf_nat_seq *this_way, *other_way; + + dir = CTINFO2DIR(ctinfo); + + this_way = &nat->info.seq[dir]; + other_way = &nat->info.seq[!dir]; + + if (!skb_make_writable(pskb, (*pskb)->nh.iph->ihl*4+sizeof(*tcph))) + return 0; + + tcph = (void *)(*pskb)->data + (*pskb)->nh.iph->ihl*4; + if (after(ntohl(tcph->seq), this_way->correction_pos)) + newseq = htonl(ntohl(tcph->seq) + this_way->offset_after); + else + newseq = htonl(ntohl(tcph->seq) + this_way->offset_before); + + if (after(ntohl(tcph->ack_seq) - other_way->offset_before, + other_way->correction_pos)) + newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_after); + else + newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before); + + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->seq, newseq, 0); + nf_proto_csum_replace4(&tcph->check, *pskb, tcph->ack_seq, newack, 0); + + DEBUGP("Adjusting sequence number from %u->%u, ack from %u->%u\n", + ntohl(tcph->seq), ntohl(newseq), ntohl(tcph->ack_seq), + ntohl(newack)); + + tcph->seq = newseq; + tcph->ack_seq = newack; + + if (!nf_nat_sack_adjust(pskb, tcph, ct, ctinfo)) + return 0; + + nf_conntrack_tcp_update(*pskb, (*pskb)->nh.iph->ihl*4, ct, dir); + + return 1; +} +EXPORT_SYMBOL(nf_nat_seq_adjust); + +/* Setup NAT on this expected conntrack so it follows master. */ +/* If we fail to get a free NAT slot, we'll get dropped on confirm */ +void nf_nat_follow_master(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_nat_range range; + + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = exp->saved_proto; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.src.u3.ip; + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); +} +EXPORT_SYMBOL(nf_nat_follow_master); diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c new file mode 100644 index 000000000000..dcfd772972d7 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c @@ -0,0 +1,86 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static int +icmp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + return ntohs(tuple->src.u.icmp.id) >= ntohs(min->icmp.id) && + ntohs(tuple->src.u.icmp.id) <= ntohs(max->icmp.id); +} + +static int +icmp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t id; + unsigned int range_size; + unsigned int i; + + range_size = ntohs(range->max.icmp.id) - ntohs(range->min.icmp.id) + 1; + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) + range_size = 0xFFFF; + + for (i = 0; i < range_size; i++, id++) { + tuple->src.u.icmp.id = htons(ntohs(range->min.icmp.id) + + (id % range_size)); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +icmp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct icmphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + + if (!skb_make_writable(pskb, hdroff + sizeof(*hdr))) + return 0; + + hdr = (struct icmphdr *)((*pskb)->data + hdroff); + nf_proto_csum_replace2(&hdr->checksum, *pskb, + hdr->un.echo.id, tuple->src.u.icmp.id, 0); + hdr->un.echo.id = tuple->src.u.icmp.id; + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_icmp = { + .name = "ICMP", + .protonum = IPPROTO_ICMP, + .me = THIS_MODULE, + .manip_pkt = icmp_manip_pkt, + .in_range = icmp_in_range, + .unique_tuple = icmp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c new file mode 100644 index 000000000000..7e26a7e9bee1 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -0,0 +1,148 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +static int +tcp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 port; + + if (maniptype == IP_NAT_MANIP_SRC) + port = tuple->src.u.tcp.port; + else + port = tuple->dst.u.tcp.port; + + return ntohs(port) >= ntohs(min->tcp.port) && + ntohs(port) <= ntohs(max->tcp.port); +} + +static int +tcp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t port; + __be16 *portptr; + unsigned int range_size, min, i; + + if (maniptype == IP_NAT_MANIP_SRC) + portptr = &tuple->src.u.tcp.port; + else + portptr = &tuple->dst.u.tcp.port; + + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + /* If it's dst rewrite, can't change port */ + if (maniptype == IP_NAT_MANIP_DST) + return 0; + + /* Map privileged onto privileged. */ + if (ntohs(*portptr) < 1024) { + /* Loose convention: >> 512 is credential passing */ + if (ntohs(*portptr)<512) { + min = 1; + range_size = 511 - min + 1; + } else { + min = 600; + range_size = 1023 - min + 1; + } + } else { + min = 1024; + range_size = 65535 - 1024 + 1; + } + } else { + min = ntohs(range->min.tcp.port); + range_size = ntohs(range->max.tcp.port) - min + 1; + } + + for (i = 0; i < range_size; i++, port++) { + *portptr = htons(min + port % range_size); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +tcp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct tcphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + __be32 oldip, newip; + __be16 *portptr, newport, oldport; + int hdrsize = 8; /* TCP connection tracking guarantees this much */ + + /* this could be a inner header returned in icmp packet; in such + cases we cannot update the checksum field since it is outside of + the 8 bytes of transport layer headers we are guaranteed */ + if ((*pskb)->len >= hdroff + sizeof(struct tcphdr)) + hdrsize = sizeof(struct tcphdr); + + if (!skb_make_writable(pskb, hdroff + hdrsize)) + return 0; + + iph = (struct iphdr *)((*pskb)->data + iphdroff); + hdr = (struct tcphdr *)((*pskb)->data + hdroff); + + if (maniptype == IP_NAT_MANIP_SRC) { + /* Get rid of src ip and src pt */ + oldip = iph->saddr; + newip = tuple->src.u3.ip; + newport = tuple->src.u.tcp.port; + portptr = &hdr->source; + } else { + /* Get rid of dst ip and dst pt */ + oldip = iph->daddr; + newip = tuple->dst.u3.ip; + newport = tuple->dst.u.tcp.port; + portptr = &hdr->dest; + } + + oldport = *portptr; + *portptr = newport; + + if (hdrsize < sizeof(*hdr)) + return 1; + + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, oldport, newport, 0); + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_tcp = { + .name = "TCP", + .protonum = IPPROTO_TCP, + .me = THIS_MODULE, + .manip_pkt = tcp_manip_pkt, + .in_range = tcp_in_range, + .unique_tuple = tcp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c new file mode 100644 index 000000000000..ab0ce4c8699f --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -0,0 +1,138 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static int +udp_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 port; + + if (maniptype == IP_NAT_MANIP_SRC) + port = tuple->src.u.udp.port; + else + port = tuple->dst.u.udp.port; + + return ntohs(port) >= ntohs(min->udp.port) && + ntohs(port) <= ntohs(max->udp.port); +} + +static int +udp_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + static u_int16_t port; + __be16 *portptr; + unsigned int range_size, min, i; + + if (maniptype == IP_NAT_MANIP_SRC) + portptr = &tuple->src.u.udp.port; + else + portptr = &tuple->dst.u.udp.port; + + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + /* If it's dst rewrite, can't change port */ + if (maniptype == IP_NAT_MANIP_DST) + return 0; + + if (ntohs(*portptr) < 1024) { + /* Loose convention: >> 512 is credential passing */ + if (ntohs(*portptr)<512) { + min = 1; + range_size = 511 - min + 1; + } else { + min = 600; + range_size = 1023 - min + 1; + } + } else { + min = 1024; + range_size = 65535 - 1024 + 1; + } + } else { + min = ntohs(range->min.udp.port); + range_size = ntohs(range->max.udp.port) - min + 1; + } + + for (i = 0; i < range_size; i++, port++) { + *portptr = htons(min + port % range_size); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} + +static int +udp_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + struct udphdr *hdr; + unsigned int hdroff = iphdroff + iph->ihl*4; + __be32 oldip, newip; + __be16 *portptr, newport; + + if (!skb_make_writable(pskb, hdroff + sizeof(*hdr))) + return 0; + + iph = (struct iphdr *)((*pskb)->data + iphdroff); + hdr = (struct udphdr *)((*pskb)->data + hdroff); + + if (maniptype == IP_NAT_MANIP_SRC) { + /* Get rid of src ip and src pt */ + oldip = iph->saddr; + newip = tuple->src.u3.ip; + newport = tuple->src.u.udp.port; + portptr = &hdr->source; + } else { + /* Get rid of dst ip and dst pt */ + oldip = iph->daddr; + newip = tuple->dst.u3.ip; + newport = tuple->dst.u.udp.port; + portptr = &hdr->dest; + } + if (hdr->check || (*pskb)->ip_summed == CHECKSUM_PARTIAL) { + nf_proto_csum_replace4(&hdr->check, *pskb, oldip, newip, 1); + nf_proto_csum_replace2(&hdr->check, *pskb, *portptr, newport, + 0); + if (!hdr->check) + hdr->check = CSUM_MANGLED_0; + } + *portptr = newport; + return 1; +} + +struct nf_nat_protocol nf_nat_protocol_udp = { + .name = "UDP", + .protonum = IPPROTO_UDP, + .me = THIS_MODULE, + .manip_pkt = udp_manip_pkt, + .in_range = udp_in_range, + .unique_tuple = udp_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c new file mode 100644 index 000000000000..f50d0203f9c0 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c @@ -0,0 +1,54 @@ +/* The "unknown" protocol. This is what is used for protocols we + * don't understand. It's returned by ip_ct_find_proto(). + */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include + +#include +#include +#include +#include + +static int unknown_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type manip_type, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + return 1; +} + +static int unknown_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) +{ + /* Sorry: we can't help you; if it's not unique, we can't frob + anything. */ + return 0; +} + +static int +unknown_manip_pkt(struct sk_buff **pskb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + return 1; +} + +struct nf_nat_protocol nf_nat_unknown_protocol = { + .name = "unknown", + /* .me isn't set: getting a ref to this cannot fail. */ + .manip_pkt = unknown_manip_pkt, + .in_range = unknown_in_range, + .unique_tuple = unknown_unique_tuple, +}; diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c new file mode 100644 index 000000000000..b868ee0195d4 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -0,0 +1,343 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* Everything about the rules for NAT. */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define NAT_VALID_HOOKS ((1<range[0], hooknum); +} + +/* Before 2.6.11 we did implicit source NAT if required. Warn about change. */ +static void warn_if_extra_mangle(__be32 dstip, __be32 srcip) +{ + static int warned = 0; + struct flowi fl = { .nl_u = { .ip4_u = { .daddr = dstip } } }; + struct rtable *rt; + + if (ip_route_output_key(&rt, &fl) != 0) + return; + + if (rt->rt_src != srcip && !warned) { + printk("NAT: no longer support implicit source local NAT\n"); + printk("NAT: packet src %u.%u.%u.%u -> dst %u.%u.%u.%u\n", + NIPQUAD(srcip), NIPQUAD(dstip)); + warned = 1; + } + ip_rt_put(rt); +} + +static unsigned int ipt_dnat_target(struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + unsigned int hooknum, + const struct xt_target *target, + const void *targinfo) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + const struct nf_nat_multi_range_compat *mr = targinfo; + + NF_CT_ASSERT(hooknum == NF_IP_PRE_ROUTING || + hooknum == NF_IP_LOCAL_OUT); + + ct = nf_ct_get(*pskb, &ctinfo); + + /* Connection must be valid and new. */ + NF_CT_ASSERT(ct && (ctinfo == IP_CT_NEW || ctinfo == IP_CT_RELATED)); + + if (hooknum == NF_IP_LOCAL_OUT && + mr->range[0].flags & IP_NAT_RANGE_MAP_IPS) + warn_if_extra_mangle((*pskb)->nh.iph->daddr, + mr->range[0].min_ip); + + return nf_nat_setup_info(ct, &mr->range[0], hooknum); +} + +static int ipt_snat_checkentry(const char *tablename, + const void *entry, + const struct xt_target *target, + void *targinfo, + unsigned int hook_mask) +{ + struct nf_nat_multi_range_compat *mr = targinfo; + + /* Must be a valid range */ + if (mr->rangesize != 1) { + printk("SNAT: multiple ranges no longer supported\n"); + return 0; + } + return 1; +} + +static int ipt_dnat_checkentry(const char *tablename, + const void *entry, + const struct xt_target *target, + void *targinfo, + unsigned int hook_mask) +{ + struct nf_nat_multi_range_compat *mr = targinfo; + + /* Must be a valid range */ + if (mr->rangesize != 1) { + printk("DNAT: multiple ranges no longer supported\n"); + return 0; + } + return 1; +} + +inline unsigned int +alloc_null_binding(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum) +{ + /* Force range to this IP; let proto decide mapping for + per-proto parts (hence not IP_NAT_RANGE_PROTO_SPECIFIED). + Use reply in case it's already been mangled (eg local packet). + */ + __be32 ip + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); + struct nf_nat_range range + = { IP_NAT_RANGE_MAP_IPS, ip, ip, { 0 }, { 0 } }; + + DEBUGP("Allocating NULL binding for %p (%u.%u.%u.%u)\n", + ct, NIPQUAD(ip)); + return nf_nat_setup_info(ct, &range, hooknum); +} + +unsigned int +alloc_null_binding_confirmed(struct nf_conn *ct, + struct nf_nat_info *info, + unsigned int hooknum) +{ + __be32 ip + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); + u_int16_t all + = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC + ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.all + : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u.all); + struct nf_nat_range range + = { IP_NAT_RANGE_MAP_IPS, ip, ip, { all }, { all } }; + + DEBUGP("Allocating NULL binding for confirmed %p (%u.%u.%u.%u)\n", + ct, NIPQUAD(ip)); + return nf_nat_setup_info(ct, &range, hooknum); +} + +int nf_nat_rule_find(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + struct nf_conn *ct, + struct nf_nat_info *info) +{ + int ret; + + ret = ipt_do_table(pskb, hooknum, in, out, &nat_table); + + if (ret == NF_ACCEPT) { + if (!nf_nat_initialized(ct, HOOK2MANIP(hooknum))) + /* NUL mapping */ + ret = alloc_null_binding(ct, info, hooknum); + } + return ret; +} + +static struct ipt_target ipt_snat_reg = { + .name = "SNAT", + .target = ipt_snat_target, + .targetsize = sizeof(struct nf_nat_multi_range_compat), + .table = "nat", + .hooks = 1 << NF_IP_POST_ROUTING, + .checkentry = ipt_snat_checkentry, + .family = AF_INET, +}; + +static struct xt_target ipt_dnat_reg = { + .name = "DNAT", + .target = ipt_dnat_target, + .targetsize = sizeof(struct nf_nat_multi_range_compat), + .table = "nat", + .hooks = (1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_OUT), + .checkentry = ipt_dnat_checkentry, + .family = AF_INET, +}; + +int __init nf_nat_rule_init(void) +{ + int ret; + + ret = ipt_register_table(&nat_table, &nat_initial_table.repl); + if (ret != 0) + return ret; + ret = xt_register_target(&ipt_snat_reg); + if (ret != 0) + goto unregister_table; + + ret = xt_register_target(&ipt_dnat_reg); + if (ret != 0) + goto unregister_snat; + + return ret; + + unregister_snat: + xt_unregister_target(&ipt_snat_reg); + unregister_table: + ipt_unregister_table(&nat_table); + + return ret; +} + +void nf_nat_rule_cleanup(void) +{ + xt_unregister_target(&ipt_dnat_reg); + xt_unregister_target(&ipt_snat_reg); + ipt_unregister_table(&nat_table); +} diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c new file mode 100644 index 000000000000..730a7a44c883 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_standalone.c @@ -0,0 +1,406 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \ + : ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \ + : ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \ + : ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \ + : "*ERROR*"))) + +#ifdef CONFIG_XFRM +static void nat_decode_session(struct sk_buff *skb, struct flowi *fl) +{ + struct nf_conn *ct; + struct nf_conntrack_tuple *t; + enum ip_conntrack_info ctinfo; + enum ip_conntrack_dir dir; + unsigned long statusbit; + + ct = nf_ct_get(skb, &ctinfo); + if (ct == NULL) + return; + dir = CTINFO2DIR(ctinfo); + t = &ct->tuplehash[dir].tuple; + + if (dir == IP_CT_DIR_ORIGINAL) + statusbit = IPS_DST_NAT; + else + statusbit = IPS_SRC_NAT; + + if (ct->status & statusbit) { + fl->fl4_dst = t->dst.u3.ip; + if (t->dst.protonum == IPPROTO_TCP || + t->dst.protonum == IPPROTO_UDP) + fl->fl_ip_dport = t->dst.u.tcp.port; + } + + statusbit ^= IPS_NAT_MASK; + + if (ct->status & statusbit) { + fl->fl4_src = t->src.u3.ip; + if (t->dst.protonum == IPPROTO_TCP || + t->dst.protonum == IPPROTO_UDP) + fl->fl_ip_sport = t->src.u.tcp.port; + } +} +#endif + +static unsigned int +nf_nat_fn(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + struct nf_conn_nat *nat; + struct nf_nat_info *info; + /* maniptype == SRC for postrouting. */ + enum nf_nat_manip_type maniptype = HOOK2MANIP(hooknum); + + /* We never see fragments: conntrack defrags on pre-routing + and local-out, and nf_nat_out protects post-routing. */ + NF_CT_ASSERT(!((*pskb)->nh.iph->frag_off + & htons(IP_MF|IP_OFFSET))); + + ct = nf_ct_get(*pskb, &ctinfo); + /* Can't track? It's not due to stress, or conntrack would + have dropped it. Hence it's the user's responsibilty to + packet filter it out, or implement conntrack/NAT for that + protocol. 8) --RR */ + if (!ct) { + /* Exception: ICMP redirect to new connection (not in + hash table yet). We must not let this through, in + case we're doing NAT to the same network. */ + if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) { + struct icmphdr _hdr, *hp; + + hp = skb_header_pointer(*pskb, + (*pskb)->nh.iph->ihl*4, + sizeof(_hdr), &_hdr); + if (hp != NULL && + hp->type == ICMP_REDIRECT) + return NF_DROP; + } + return NF_ACCEPT; + } + + /* Don't try to NAT if this packet is not conntracked */ + if (ct == &nf_conntrack_untracked) + return NF_ACCEPT; + + nat = nfct_nat(ct); + if (!nat) + return NF_DROP; + + switch (ctinfo) { + case IP_CT_RELATED: + case IP_CT_RELATED+IP_CT_IS_REPLY: + if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) { + if (!nf_nat_icmp_reply_translation(ct, ctinfo, + hooknum, pskb)) + return NF_DROP; + else + return NF_ACCEPT; + } + /* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */ + case IP_CT_NEW: + info = &nat->info; + + /* Seen it before? This can happen for loopback, retrans, + or local packets.. */ + if (!nf_nat_initialized(ct, maniptype)) { + unsigned int ret; + + if (unlikely(nf_ct_is_confirmed(ct))) + /* NAT module was loaded late */ + ret = alloc_null_binding_confirmed(ct, info, + hooknum); + else if (hooknum == NF_IP_LOCAL_IN) + /* LOCAL_IN hook doesn't have a chain! */ + ret = alloc_null_binding(ct, info, hooknum); + else + ret = nf_nat_rule_find(pskb, hooknum, in, out, + ct, info); + + if (ret != NF_ACCEPT) { + return ret; + } + } else + DEBUGP("Already setup manip %s for ct %p\n", + maniptype == IP_NAT_MANIP_SRC ? "SRC" : "DST", + ct); + break; + + default: + /* ESTABLISHED */ + NF_CT_ASSERT(ctinfo == IP_CT_ESTABLISHED || + ctinfo == (IP_CT_ESTABLISHED+IP_CT_IS_REPLY)); + info = &nat->info; + } + + NF_CT_ASSERT(info); + return nf_nat_packet(ct, ctinfo, hooknum, pskb); +} + +static unsigned int +nf_nat_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + unsigned int ret; + __be32 daddr = (*pskb)->nh.iph->daddr; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); + if (ret != NF_DROP && ret != NF_STOLEN && + daddr != (*pskb)->nh.iph->daddr) { + dst_release((*pskb)->dst); + (*pskb)->dst = NULL; + } + return ret; +} + +static unsigned int +nf_nat_out(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ +#ifdef CONFIG_XFRM + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; +#endif + unsigned int ret; + + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) || + (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) + return NF_ACCEPT; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); +#ifdef CONFIG_XFRM + if (ret != NF_DROP && ret != NF_STOLEN && + (ct = nf_ct_get(*pskb, &ctinfo)) != NULL) { + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + + if (ct->tuplehash[dir].tuple.src.u3.ip != + ct->tuplehash[!dir].tuple.dst.u3.ip + || ct->tuplehash[dir].tuple.src.u.all != + ct->tuplehash[!dir].tuple.dst.u.all + ) + return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP; + } +#endif + return ret; +} + +static unsigned int +nf_nat_local_fn(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + unsigned int ret; + + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) || + (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) + return NF_ACCEPT; + + ret = nf_nat_fn(hooknum, pskb, in, out, okfn); + if (ret != NF_DROP && ret != NF_STOLEN && + (ct = nf_ct_get(*pskb, &ctinfo)) != NULL) { + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + + if (ct->tuplehash[dir].tuple.dst.u3.ip != + ct->tuplehash[!dir].tuple.src.u3.ip +#ifdef CONFIG_XFRM + || ct->tuplehash[dir].tuple.dst.u.all != + ct->tuplehash[!dir].tuple.src.u.all +#endif + ) + if (ip_route_me_harder(pskb, RTN_UNSPEC)) + ret = NF_DROP; + } + return ret; +} + +static unsigned int +nf_nat_adjust(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { + DEBUGP("nf_nat_standalone: adjusting sequence number\n"); + if (!nf_nat_seq_adjust(pskb, ct, ctinfo)) + return NF_DROP; + } + return NF_ACCEPT; +} + +/* We must be after connection tracking and before packet filtering. */ + +static struct nf_hook_ops nf_nat_ops[] = { + /* Before packet filtering, change destination */ + { + .hook = nf_nat_in, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_NAT_DST, + }, + /* After packet filtering, change source */ + { + .hook = nf_nat_out, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SRC, + }, + /* After conntrack, adjust sequence number */ + { + .hook = nf_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, + }, + /* Before packet filtering, change destination */ + { + .hook = nf_nat_local_fn, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_NAT_DST, + }, + /* After packet filtering, change source */ + { + .hook = nf_nat_fn, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_NAT_SRC, + }, + /* After conntrack, adjust sequence number */ + { + .hook = nf_nat_adjust, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_NAT_SEQ_ADJUST, + }, +}; + +static int __init nf_nat_standalone_init(void) +{ + int size, ret = 0; + + need_conntrack(); + + size = ALIGN(sizeof(struct nf_conn), __alignof__(struct nf_conn_nat)) + + sizeof(struct nf_conn_nat); + ret = nf_conntrack_register_cache(NF_CT_F_NAT, "nf_nat:base", size); + if (ret < 0) { + printk(KERN_ERR "nf_nat_init: Unable to create slab cache\n"); + return ret; + } + + size = ALIGN(size, __alignof__(struct nf_conn_help)) + + sizeof(struct nf_conn_help); + ret = nf_conntrack_register_cache(NF_CT_F_NAT|NF_CT_F_HELP, + "nf_nat:help", size); + if (ret < 0) { + printk(KERN_ERR "nf_nat_init: Unable to create slab cache\n"); + goto cleanup_register_cache; + } +#ifdef CONFIG_XFRM + BUG_ON(ip_nat_decode_session != NULL); + ip_nat_decode_session = nat_decode_session; +#endif + ret = nf_nat_rule_init(); + if (ret < 0) { + printk("nf_nat_init: can't setup rules.\n"); + goto cleanup_decode_session; + } + ret = nf_register_hooks(nf_nat_ops, ARRAY_SIZE(nf_nat_ops)); + if (ret < 0) { + printk("nf_nat_init: can't register hooks.\n"); + goto cleanup_rule_init; + } + nf_nat_module_is_loaded = 1; + return ret; + + cleanup_rule_init: + nf_nat_rule_cleanup(); + cleanup_decode_session: +#ifdef CONFIG_XFRM + ip_nat_decode_session = NULL; + synchronize_net(); +#endif + nf_conntrack_unregister_cache(NF_CT_F_NAT|NF_CT_F_HELP); + cleanup_register_cache: + nf_conntrack_unregister_cache(NF_CT_F_NAT); + return ret; +} + +static void __exit nf_nat_standalone_fini(void) +{ + nf_unregister_hooks(nf_nat_ops, ARRAY_SIZE(nf_nat_ops)); + nf_nat_rule_cleanup(); + nf_nat_module_is_loaded = 0; +#ifdef CONFIG_XFRM + ip_nat_decode_session = NULL; + synchronize_net(); +#endif + /* Conntrack caches are unregistered in nf_conntrack_cleanup */ +} + +module_init(nf_nat_standalone_init); +module_exit(nf_nat_standalone_fini); + +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat"); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index f952a7fb6ae3..aa8beabfeebb 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -579,7 +579,8 @@ __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, /* FIXME: protect helper list per RCU */ read_lock_bh(&nf_conntrack_lock); helper = __nf_ct_helper_find(repl); - if (helper) + /* NAT might want to assign a helper later */ + if (helper || features & NF_CT_F_NAT) features |= NF_CT_F_HELP; read_unlock_bh(&nf_conntrack_lock); @@ -850,6 +851,26 @@ int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, orig->dst.protonum)); } +/* Alter reply tuple (maybe alter helper). This is for NAT, and is + implicitly racy: see __nf_conntrack_confirm */ +void nf_conntrack_alter_reply(struct nf_conn *ct, + const struct nf_conntrack_tuple *newreply) +{ + struct nf_conn_help *help = nfct_help(ct); + + write_lock_bh(&nf_conntrack_lock); + /* Should be unconfirmed, so not in hash table yet */ + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + + DEBUGP("Altering reply tuple of %p to ", ct); + NF_CT_DUMP_TUPLE(newreply); + + ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!ct->master && help && help->expecting == 0) + help->helper = __nf_ct_helper_find(newreply); + write_unlock_bh(&nf_conntrack_lock); +} + /* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ void __nf_ct_refresh_acct(struct nf_conn *ct, enum ip_conntrack_info ctinfo, diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index f9c8ddd5973c..bd1d2de75e45 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -39,7 +39,11 @@ #include #include #include -#include +#include +#ifdef CONFIG_NF_NAT_NEEDED +#include +#include +#endif #include #include @@ -430,7 +434,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) restart: list_for_each_prev(i, &nf_conntrack_hash[cb->args[0]]) { h = (struct nf_conntrack_tuple_hash *) i; - if (DIRECTION(h) != IP_CT_DIR_ORIGINAL) + if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL) continue; ct = nf_ct_tuplehash_to_ctrack(h); /* Dump entries of a given L3 protocol number. @@ -556,28 +560,28 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct nf_conntrack_tuple *tuple, return 0; } -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED static const size_t cta_min_protonat[CTA_PROTONAT_MAX] = { [CTA_PROTONAT_PORT_MIN-1] = sizeof(u_int16_t), [CTA_PROTONAT_PORT_MAX-1] = sizeof(u_int16_t), }; -static int ctnetlink_parse_nat_proto(struct nfattr *attr, +static int nfnetlink_parse_nat_proto(struct nfattr *attr, const struct nf_conn *ct, - struct ip_nat_range *range) + struct nf_nat_range *range) { struct nfattr *tb[CTA_PROTONAT_MAX]; - struct ip_nat_protocol *npt; + struct nf_nat_protocol *npt; nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTONAT_MAX, cta_min_protonat)) return -EINVAL; - npt = ip_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + npt = nf_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); if (!npt->nfattr_to_range) { - ip_nat_proto_put(npt); + nf_nat_proto_put(npt); return 0; } @@ -585,7 +589,7 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, if (npt->nfattr_to_range(tb, range) > 0) range->flags |= IP_NAT_RANGE_PROTO_SPECIFIED; - ip_nat_proto_put(npt); + nf_nat_proto_put(npt); return 0; } @@ -596,8 +600,8 @@ static const size_t cta_min_nat[CTA_NAT_MAX] = { }; static inline int -ctnetlink_parse_nat(struct nfattr *nat, - const struct nf_conn *ct, struct ip_nat_range *range) +nfnetlink_parse_nat(struct nfattr *nat, + const struct nf_conn *ct, struct nf_nat_range *range) { struct nfattr *tb[CTA_NAT_MAX]; int err; @@ -623,7 +627,7 @@ ctnetlink_parse_nat(struct nfattr *nat, if (!tb[CTA_NAT_PROTO-1]) return 0; - err = ctnetlink_parse_nat_proto(tb[CTA_NAT_PROTO-1], ct, range); + err = nfnetlink_parse_nat_proto(tb[CTA_NAT_PROTO-1], ct, range); if (err < 0) return err; @@ -798,35 +802,35 @@ ctnetlink_change_status(struct nf_conn *ct, struct nfattr *cda[]) return -EINVAL; if (cda[CTA_NAT_SRC-1] || cda[CTA_NAT_DST-1]) { -#ifndef CONFIG_IP_NF_NAT_NEEDED +#ifndef CONFIG_NF_NAT_NEEDED return -EINVAL; #else - struct ip_nat_range range; + struct nf_nat_range range; if (cda[CTA_NAT_DST-1]) { - if (ctnetlink_parse_nat(cda[CTA_NAT_DST-1], ct, + if (nfnetlink_parse_nat(cda[CTA_NAT_DST-1], ct, &range) < 0) return -EINVAL; - if (ip_nat_initialized(ct, + if (nf_nat_initialized(ct, HOOK2MANIP(NF_IP_PRE_ROUTING))) return -EEXIST; - ip_nat_setup_info(ct, &range, hooknum); + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); } if (cda[CTA_NAT_SRC-1]) { - if (ctnetlink_parse_nat(cda[CTA_NAT_SRC-1], ct, + if (nfnetlink_parse_nat(cda[CTA_NAT_SRC-1], ct, &range) < 0) return -EINVAL; - if (ip_nat_initialized(ct, + if (nf_nat_initialized(ct, HOOK2MANIP(NF_IP_POST_ROUTING))) return -EEXIST; - ip_nat_setup_info(ct, &range, hooknum); + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); } #endif } /* Be careful here, modifying NAT bits can screw up things, * so don't let users modify them directly if they don't pass - * ip_nat_range. */ + * nf_nat_range. */ ct->status |= status & ~(IPS_NAT_DONE_MASK | IPS_NAT_MASK); return 0; } diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 71f492fc6413..8156e429b885 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -730,7 +730,7 @@ static int tcp_in_window(struct ip_ct_tcp *state, return res; } -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED /* Update sender->td_end after NAT successfully mangled the packet */ /* Caller must linearize skb at tcp header. */ void nf_conntrack_tcp_update(struct sk_buff *skb, diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index be94b6359725..3f56a3a6c399 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -530,8 +530,11 @@ EXPORT_SYMBOL(nf_conntrack_lock); EXPORT_SYMBOL(nf_conntrack_hash); EXPORT_SYMBOL(nf_conntrack_untracked); EXPORT_SYMBOL_GPL(nf_conntrack_find_get); -#ifdef CONFIG_IP_NF_NAT_NEEDED +#ifdef CONFIG_NF_NAT_NEEDED EXPORT_SYMBOL(nf_conntrack_tcp_update); +EXPORT_SYMBOL(nf_conntrack_register_cache); +EXPORT_SYMBOL(nf_conntrack_unregister_cache); +EXPORT_SYMBOL(nf_conntrack_alter_reply); #endif EXPORT_SYMBOL(__nf_conntrack_confirm); EXPORT_SYMBOL(nf_ct_get_tuple); -- cgit v1.2.3 From 55a733247d6d2883d9bb77825fafac3dfca13fc2 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Sat, 2 Dec 2006 22:07:44 -0800 Subject: [NETFILTER]: nf_nat: add FTP NAT helper port Add FTP NAT helper. Split out from Jozsef's big nf_nat patch with a few small fixes by myself. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_ftp.h | 20 +-- include/linux/netfilter_ipv4/ip_conntrack_ftp.h | 40 +++++- include/net/netfilter/nf_conntrack.h | 2 +- net/ipv4/netfilter/Kconfig | 25 ++-- net/ipv4/netfilter/Makefile | 5 +- net/ipv4/netfilter/nf_nat_ftp.c | 179 ++++++++++++++++++++++++ net/netfilter/nf_conntrack_ftp.c | 20 +-- 7 files changed, 260 insertions(+), 31 deletions(-) create mode 100644 net/ipv4/netfilter/nf_nat_ftp.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_ftp.h b/include/linux/netfilter/nf_conntrack_ftp.h index ad4a41c9ce93..81453ea7e4c2 100644 --- a/include/linux/netfilter/nf_conntrack_ftp.h +++ b/include/linux/netfilter/nf_conntrack_ftp.h @@ -3,16 +3,16 @@ /* FTP tracking. */ /* This enum is exposed to userspace */ -enum ip_ct_ftp_type +enum nf_ct_ftp_type { /* PORT command from client */ - IP_CT_FTP_PORT, + NF_CT_FTP_PORT, /* PASV response from server */ - IP_CT_FTP_PASV, + NF_CT_FTP_PASV, /* EPRT command from client */ - IP_CT_FTP_EPRT, + NF_CT_FTP_EPRT, /* EPSV response from server */ - IP_CT_FTP_EPSV, + NF_CT_FTP_EPSV, }; #ifdef __KERNEL__ @@ -21,23 +21,23 @@ enum ip_ct_ftp_type #define NUM_SEQ_TO_REMEMBER 2 /* This structure exists only once per master */ -struct ip_ct_ftp_master { +struct nf_ct_ftp_master { /* Valid seq positions for cmd matching after newline */ u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; /* 0 means seq_match_aft_nl not set */ int seq_aft_nl_num[IP_CT_DIR_MAX]; }; -struct ip_conntrack_expect; +struct nf_conntrack_expect; /* For NAT to hook in when we find a packet which describes what other * connection we should expect. */ -extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, +extern unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, - enum ip_ct_ftp_type type, + enum nf_ct_ftp_type type, unsigned int matchoff, unsigned int matchlen, - struct ip_conntrack_expect *exp, + struct nf_conntrack_expect *exp, u32 *seq); #endif /* __KERNEL__ */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h index 63811934de4d..2129fc3972ac 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h @@ -1,6 +1,44 @@ #ifndef _IP_CONNTRACK_FTP_H #define _IP_CONNTRACK_FTP_H +/* FTP tracking. */ -#include +/* This enum is exposed to userspace */ +enum ip_ct_ftp_type +{ + /* PORT command from client */ + IP_CT_FTP_PORT, + /* PASV response from server */ + IP_CT_FTP_PASV, + /* EPRT command from client */ + IP_CT_FTP_EPRT, + /* EPSV response from server */ + IP_CT_FTP_EPSV, +}; + +#ifdef __KERNEL__ + +#define FTP_PORT 21 + +#define NUM_SEQ_TO_REMEMBER 2 +/* This structure exists only once per master */ +struct ip_ct_ftp_master { + /* Valid seq positions for cmd matching after newline */ + u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; + /* 0 means seq_match_aft_nl not set */ + int seq_aft_nl_num[IP_CT_DIR_MAX]; +}; + +struct ip_conntrack_expect; + +/* For NAT to hook in when we find a packet which describes what other + * connection we should expect. */ +extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum ip_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct ip_conntrack_expect *exp, + u32 *seq); +#endif /* __KERNEL__ */ #endif /* _IP_CONNTRACK_FTP_H */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 9948af068688..83694cfdfa8f 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -45,7 +45,7 @@ union nf_conntrack_expect_proto { /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ - struct ip_ct_ftp_master ct_ftp_info; + struct nf_ct_ftp_master ct_ftp_info; }; #include diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 01789aeaeb5f..52f876db68f4 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -477,20 +477,29 @@ config IP_NF_NAT_SNMP_BASIC To compile it as a module, choose M here. If unsure, say N. +# If they want FTP, set to $CONFIG_IP_NF_NAT (m or y), +# or $CONFIG_IP_NF_FTP (m or y), whichever is weaker. +# From kconfig-language.txt: +# +# '&&' (6) +# +# (6) Returns the result of min(/expr/, /expr/). +config IP_NF_NAT_FTP + tristate + depends on IP_NF_IPTABLES && IP_NF_CONNTRACK && IP_NF_NAT + default IP_NF_NAT && IP_NF_FTP + +config NF_NAT_FTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_FTP + config IP_NF_NAT_IRC tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n default IP_NF_NAT if IP_NF_IRC=y default m if IP_NF_IRC=m -# If they want FTP, set to $CONFIG_IP_NF_NAT (m or y), -# or $CONFIG_IP_NF_FTP (m or y), whichever is weaker. Argh. -config IP_NF_NAT_FTP - tristate - depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n - default IP_NF_NAT if IP_NF_FTP=y - default m if IP_NF_FTP=m - config IP_NF_NAT_TFTP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index ec31690764ac..c0c6194bb275 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -40,7 +40,7 @@ obj-$(CONFIG_IP_NF_IRC) += ip_conntrack_irc.o obj-$(CONFIG_IP_NF_SIP) += ip_conntrack_sip.o obj-$(CONFIG_IP_NF_NETBIOS_NS) += ip_conntrack_netbios_ns.o -# NAT helpers +# NAT helpers (ip_conntrack) obj-$(CONFIG_IP_NF_NAT_H323) += ip_nat_h323.o obj-$(CONFIG_IP_NF_NAT_PPTP) += ip_nat_pptp.o obj-$(CONFIG_IP_NF_NAT_AMANDA) += ip_nat_amanda.o @@ -49,6 +49,9 @@ obj-$(CONFIG_IP_NF_NAT_FTP) += ip_nat_ftp.o obj-$(CONFIG_IP_NF_NAT_IRC) += ip_nat_irc.o obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o +# NAT helpers (nf_conntrack) +obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o + # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_ftp.c b/net/ipv4/netfilter/nf_nat_ftp.c new file mode 100644 index 000000000000..751b59801755 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_ftp.c @@ -0,0 +1,179 @@ +/* FTP extension for TCP NAT alteration. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Rusty Russell "); +MODULE_DESCRIPTION("ftp NAT helper"); +MODULE_ALIAS("ip_nat_ftp"); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/* FIXME: Time out? --RR */ + +static int +mangle_rfc959_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("nnn,nnn,nnn,nnn,nnn,nnn")]; + + sprintf(buffer, "%u,%u,%u,%u,%u,%u", + NIPQUAD(newip), port>>8, port&0xFF); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +/* |1|132.235.1.2|6275| */ +static int +mangle_eprt_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("|1|255.255.255.255|65535|")]; + + sprintf(buffer, "|1|%u.%u.%u.%u|%u|", NIPQUAD(newip), port); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +/* |1|132.235.1.2|6275| */ +static int +mangle_epsv_packet(struct sk_buff **pskb, + __be32 newip, + u_int16_t port, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + u32 *seq) +{ + char buffer[sizeof("|||65535|")]; + + sprintf(buffer, "|||%u|", port); + + DEBUGP("calling nf_nat_mangle_tcp_packet\n"); + + *seq += strlen(buffer) - matchlen; + return nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, matchoff, + matchlen, buffer, strlen(buffer)); +} + +static int (*mangle[])(struct sk_buff **, __be32, u_int16_t, + unsigned int, unsigned int, struct nf_conn *, + enum ip_conntrack_info, u32 *seq) += { + [NF_CT_FTP_PORT] = mangle_rfc959_packet, + [NF_CT_FTP_PASV] = mangle_rfc959_packet, + [NF_CT_FTP_EPRT] = mangle_eprt_packet, + [NF_CT_FTP_EPSV] = mangle_epsv_packet +}; + +/* So, this packet has hit the connection tracking matching code. + Mangle it, and change the expectation to match the new version. */ +static unsigned int nf_nat_ftp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum nf_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp, + u32 *seq) +{ + __be32 newip; + u_int16_t port; + int dir = CTINFO2DIR(ctinfo); + struct nf_conn *ct = exp->master; + + DEBUGP("FTP_NAT: type %i, off %u len %u\n", type, matchoff, matchlen); + + /* Connection will come from wherever this packet goes, hence !dir */ + newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = !dir; + + /* When you see the packet, we need to NAT it the same as the + * this one. */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + if (!mangle[type](pskb, newip, port, matchoff, matchlen, ct, ctinfo, + seq)) { + nf_conntrack_unexpect_related(exp); + return NF_DROP; + } + return NF_ACCEPT; +} + +static void __exit nf_nat_ftp_fini(void) +{ + rcu_assign_pointer(nf_nat_ftp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_ftp_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_ftp_hook)); + rcu_assign_pointer(nf_nat_ftp_hook, nf_nat_ftp); + return 0; +} + +/* Prior to 2.6.11, we had a ports param. No longer, but don't break users. */ +static int warn_set(const char *val, struct kernel_param *kp) +{ + printk(KERN_INFO KBUILD_MODNAME + ": kernel >= 2.6.10 only uses 'ports' for conntrack modules\n"); + return 0; +} +module_param_call(ports, warn_set, NULL, NULL, 0); + +module_init(nf_nat_ftp_init); +module_exit(nf_nat_ftp_fini); diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index e299d657e4fc..92a947168761 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -51,7 +51,7 @@ module_param(loose, bool, 0600); unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, - enum ip_ct_ftp_type type, + enum nf_ct_ftp_type type, unsigned int matchoff, unsigned int matchlen, struct nf_conntrack_expect *exp, @@ -74,7 +74,7 @@ static struct ftp_search { size_t plen; char skip; char term; - enum ip_ct_ftp_type ftptype; + enum nf_ct_ftp_type ftptype; int (*getnum)(const char *, size_t, struct nf_conntrack_man *, char); } search[IP_CT_DIR_MAX][2] = { [IP_CT_DIR_ORIGINAL] = { @@ -83,7 +83,7 @@ static struct ftp_search { .plen = sizeof("PORT") - 1, .skip = ' ', .term = '\r', - .ftptype = IP_CT_FTP_PORT, + .ftptype = NF_CT_FTP_PORT, .getnum = try_rfc959, }, { @@ -91,7 +91,7 @@ static struct ftp_search { .plen = sizeof("EPRT") - 1, .skip = ' ', .term = '\r', - .ftptype = IP_CT_FTP_EPRT, + .ftptype = NF_CT_FTP_EPRT, .getnum = try_eprt, }, }, @@ -101,7 +101,7 @@ static struct ftp_search { .plen = sizeof("227 ") - 1, .skip = '(', .term = ')', - .ftptype = IP_CT_FTP_PASV, + .ftptype = NF_CT_FTP_PASV, .getnum = try_rfc959, }, { @@ -109,7 +109,7 @@ static struct ftp_search { .plen = sizeof("229 ") - 1, .skip = '(', .term = ')', - .ftptype = IP_CT_FTP_EPSV, + .ftptype = NF_CT_FTP_EPSV, .getnum = try_epsv_response, }, }, @@ -320,7 +320,7 @@ static int find_pattern(const char *data, size_t dlen, } /* Look up to see if we're just after a \n. */ -static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) +static int find_nl_seq(u32 seq, const struct nf_ct_ftp_master *info, int dir) { unsigned int i; @@ -331,7 +331,7 @@ static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) } /* We don't update if it's older than what we have. */ -static void update_nl_seq(u32 nl_seq, struct ip_ct_ftp_master *info, int dir, +static void update_nl_seq(u32 nl_seq, struct nf_ct_ftp_master *info, int dir, struct sk_buff *skb) { unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; @@ -367,7 +367,7 @@ static int help(struct sk_buff **pskb, u32 seq; int dir = CTINFO2DIR(ctinfo); unsigned int matchlen, matchoff; - struct ip_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info; + struct nf_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info; struct nf_conntrack_expect *exp; struct nf_conntrack_man cmd = {}; unsigned int i; @@ -523,7 +523,7 @@ static int help(struct sk_buff **pskb, /* Now, NAT might want to mangle the packet, and register the * (possibly changed) expectation itself. */ nf_nat_ftp = rcu_dereference(nf_nat_ftp_hook); - if (nf_nat_ftp) + if (nf_nat_ftp && ct->status & IPS_NAT_MASK) ret = nf_nat_ftp(pskb, ctinfo, search[dir][i].ftptype, matchoff, matchlen, exp, &seq); else { -- cgit v1.2.3 From 16958900578b94585c2ab9a2d20d837b4d5e3ba6 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:08:26 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add amanda helper port Add IPv4 and IPv6 capable nf_conntrack port of the Amanda conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_amanda.h | 10 ++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_amanda.c | 78 +++++++++ net/netfilter/Kconfig | 15 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_amanda.c | 237 ++++++++++++++++++++++++++ 7 files changed, 347 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_amanda.h create mode 100644 net/ipv4/netfilter/nf_nat_amanda.c create mode 100644 net/netfilter/nf_conntrack_amanda.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_amanda.h b/include/linux/netfilter/nf_conntrack_amanda.h new file mode 100644 index 000000000000..26c223544ae8 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_amanda.h @@ -0,0 +1,10 @@ +#ifndef _NF_CONNTRACK_AMANDA_H +#define _NF_CONNTRACK_AMANDA_H +/* AMANDA tracking. */ + +extern unsigned int (*nf_nat_amanda_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp); +#endif /* _NF_CONNTRACK_AMANDA_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 52f876db68f4..6993ec53dc06 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -512,6 +512,11 @@ config IP_NF_NAT_AMANDA default IP_NF_NAT if IP_NF_AMANDA=y default m if IP_NF_AMANDA=m +config NF_NAT_AMANDA + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_AMANDA + config IP_NF_NAT_PPTP tristate depends on IP_NF_NAT!=n && IP_NF_PPTP!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index c0c6194bb275..8893249bbe98 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -50,6 +50,7 @@ obj-$(CONFIG_IP_NF_NAT_IRC) += ip_nat_irc.o obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o # NAT helpers (nf_conntrack) +obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o # generic IP tables diff --git a/net/ipv4/netfilter/nf_nat_amanda.c b/net/ipv4/netfilter/nf_nat_amanda.c new file mode 100644 index 000000000000..0f17098917bc --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_amanda.c @@ -0,0 +1,78 @@ +/* Amanda extension for TCP NAT alteration. + * (C) 2002 by Brian J. Murrell + * based on a copy of HW's ip_nat_irc.c as well as other modules + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Brian J. Murrell "); +MODULE_DESCRIPTION("Amanda NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_amanda"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) +{ + char buffer[sizeof("65535")]; + u_int16_t port; + unsigned int ret; + + /* Connection comes from client. */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = IP_CT_DIR_ORIGINAL; + + /* When you see the packet, we need to NAT it the same as the + * this one (ie. same IP: it will be TCP and master is UDP). */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + sprintf(buffer, "%u", port); + ret = nf_nat_mangle_udp_packet(pskb, exp->master, ctinfo, + matchoff, matchlen, + buffer, strlen(buffer)); + if (ret != NF_ACCEPT) + nf_conntrack_unexpect_related(exp); + return ret; +} + +static void __exit nf_nat_amanda_fini(void) +{ + rcu_assign_pointer(nf_nat_amanda_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_amanda_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_amanda_hook)); + rcu_assign_pointer(nf_nat_amanda_hook, help); + return 0; +} + +module_init(nf_nat_amanda_init); +module_exit(nf_nat_amanda_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index efe56f768f34..f85fd43b344b 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -131,6 +131,21 @@ config NF_CT_PROTO_SCTP If you want to compile it as a module, say M here and read Documentation/modules.txt. If unsure, say `N'. +config NF_CONNTRACK_AMANDA + tristate "Amanda backup protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + select TEXTSEARCH + select TEXTSEARCH_KMP + help + If you are running the Amanda backup package + on this machine or machines that will be MASQUERADED through this + machine, then you may want to enable this feature. This allows the + connection tracking and natting code to allow the sub-channels that + Amanda requires for communication of the backup data, messages and + index. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CONNTRACK_FTP tristate "FTP support on new connection tracking (EXPERIMENTAL)" depends on EXPERIMENTAL && NF_CONNTRACK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 7f0089c584bf..a5ee93817427 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o # connection tracking helpers +obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o # generic X tables diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c new file mode 100644 index 000000000000..5c495dc9d942 --- /dev/null +++ b/net/netfilter/nf_conntrack_amanda.c @@ -0,0 +1,237 @@ +/* Amanda extension for IP connection tracking + * + * (C) 2002 by Brian J. Murrell + * based on HW's ip_conntrack_irc.c as well as other modules + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +static unsigned int master_timeout __read_mostly = 300; +static char *ts_algo = "kmp"; + +MODULE_AUTHOR("Brian J. Murrell "); +MODULE_DESCRIPTION("Amanda connection tracking module"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_amanda"); + +module_param(master_timeout, uint, 0600); +MODULE_PARM_DESC(master_timeout, "timeout for the master connection"); +module_param(ts_algo, charp, 0400); +MODULE_PARM_DESC(ts_algo, "textsearch algorithm to use (default kmp)"); + +unsigned int (*nf_nat_amanda_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_amanda_hook); + +enum amanda_strings { + SEARCH_CONNECT, + SEARCH_NEWLINE, + SEARCH_DATA, + SEARCH_MESG, + SEARCH_INDEX, +}; + +static struct { + char *string; + size_t len; + struct ts_config *ts; +} search[] __read_mostly = { + [SEARCH_CONNECT] = { + .string = "CONNECT ", + .len = 8, + }, + [SEARCH_NEWLINE] = { + .string = "\n", + .len = 1, + }, + [SEARCH_DATA] = { + .string = "DATA ", + .len = 5, + }, + [SEARCH_MESG] = { + .string = "MESG ", + .len = 5, + }, + [SEARCH_INDEX] = { + .string = "INDEX ", + .len = 6, + }, +}; + +static int amanda_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct ts_state ts; + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + unsigned int dataoff, start, stop, off, i; + char pbuf[sizeof("65535")], *tmp; + u_int16_t len; + __be16 port; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int ret = NF_ACCEPT; + typeof(nf_nat_amanda_hook) nf_nat_amanda; + + /* Only look at packets from the Amanda server */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) + return NF_ACCEPT; + + /* increase the UDP timeout of the master connection as replies from + * Amanda clients to the server can be quite delayed */ + nf_ct_refresh(ct, *pskb, master_timeout * HZ); + + /* No data? */ + dataoff = protoff + sizeof(struct udphdr); + if (dataoff >= (*pskb)->len) { + if (net_ratelimit()) + printk("amanda_help: skblen = %u\n", (*pskb)->len); + return NF_ACCEPT; + } + + memset(&ts, 0, sizeof(ts)); + start = skb_find_text(*pskb, dataoff, (*pskb)->len, + search[SEARCH_CONNECT].ts, &ts); + if (start == UINT_MAX) + goto out; + start += dataoff + search[SEARCH_CONNECT].len; + + memset(&ts, 0, sizeof(ts)); + stop = skb_find_text(*pskb, start, (*pskb)->len, + search[SEARCH_NEWLINE].ts, &ts); + if (stop == UINT_MAX) + goto out; + stop += start; + + for (i = SEARCH_DATA; i <= SEARCH_INDEX; i++) { + memset(&ts, 0, sizeof(ts)); + off = skb_find_text(*pskb, start, stop, search[i].ts, &ts); + if (off == UINT_MAX) + continue; + off += start + search[i].len; + + len = min_t(unsigned int, sizeof(pbuf) - 1, stop - off); + if (skb_copy_bits(*pskb, off, pbuf, len)) + break; + pbuf[len] = '\0'; + + port = htons(simple_strtoul(pbuf, &tmp, 10)); + len = tmp - pbuf; + if (port == 0 || len > 5) + break; + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + nf_conntrack_expect_init(exp, family, + &tuple->src.u3, &tuple->dst.u3, + IPPROTO_TCP, NULL, &port); + + nf_nat_amanda = rcu_dereference(nf_nat_amanda_hook); + if (nf_nat_amanda && ct->status & IPS_NAT_MASK) + ret = nf_nat_amanda(pskb, ctinfo, off - dataoff, + len, exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + } + +out: + return ret; +} + +static struct nf_conntrack_helper amanda_helper[2] __read_mostly = { + { + .name = "amanda", + .max_expected = 3, + .timeout = 180, + .me = THIS_MODULE, + .help = amanda_help, + .tuple.src.l3num = AF_INET, + .tuple.src.u.udp.port = __constant_htons(10080), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + }, + { + .name = "amanda", + .max_expected = 3, + .timeout = 180, + .me = THIS_MODULE, + .help = amanda_help, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.udp.port = __constant_htons(10080), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + }, +}; + +static void __exit nf_conntrack_amanda_fini(void) +{ + int i; + + nf_conntrack_helper_unregister(&amanda_helper[0]); + nf_conntrack_helper_unregister(&amanda_helper[1]); + for (i = 0; i < ARRAY_SIZE(search); i++) + textsearch_destroy(search[i].ts); +} + +static int __init nf_conntrack_amanda_init(void) +{ + int ret, i; + + ret = -ENOMEM; + for (i = 0; i < ARRAY_SIZE(search); i++) { + search[i].ts = textsearch_prepare(ts_algo, search[i].string, + search[i].len, + GFP_KERNEL, TS_AUTOLOAD); + if (search[i].ts == NULL) + goto err1; + } + ret = nf_conntrack_helper_register(&amanda_helper[0]); + if (ret < 0) + goto err1; + ret = nf_conntrack_helper_register(&amanda_helper[1]); + if (ret < 0) + goto err2; + return 0; + +err2: + nf_conntrack_helper_unregister(&amanda_helper[0]); +err1: + for (; i >= 0; i--) { + if (search[i].ts) + textsearch_destroy(search[i].ts); + } + return ret; +} + +module_init(nf_conntrack_amanda_init); +module_exit(nf_conntrack_amanda_fini); -- cgit v1.2.3 From f587de0e2feb9eb9b94f98d0a7b7437e4d6617b4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:08:46 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add H.323 helper port Add IPv4 and IPv6 capable nf_conntrack port of the H.323 conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_h323.h | 92 + include/linux/netfilter/nf_conntrack_h323_asn1.h | 98 + include/linux/netfilter/nf_conntrack_h323_types.h | 951 ++++++++++ include/linux/netfilter_ipv4/Kbuild | 2 - include/linux/netfilter_ipv4/ip_conntrack_h323.h | 2 +- .../netfilter_ipv4/ip_conntrack_helper_h323_asn1.h | 98 - .../ip_conntrack_helper_h323_types.h | 939 ---------- include/net/netfilter/nf_conntrack.h | 2 + include/net/netfilter/nf_conntrack_expect.h | 1 + net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 3 +- net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c | 874 --------- .../netfilter/ip_conntrack_helper_h323_types.c | 1926 ------------------- net/ipv4/netfilter/nf_nat_h323.c | 596 ++++++ net/netfilter/Kconfig | 19 + net/netfilter/Makefile | 3 + net/netfilter/nf_conntrack_h323_asn1.c | 874 +++++++++ net/netfilter/nf_conntrack_h323_main.c | 1856 +++++++++++++++++++ net/netfilter/nf_conntrack_h323_types.c | 1927 ++++++++++++++++++++ 19 files changed, 6427 insertions(+), 3841 deletions(-) create mode 100644 include/linux/netfilter/nf_conntrack_h323.h create mode 100644 include/linux/netfilter/nf_conntrack_h323_asn1.h create mode 100644 include/linux/netfilter/nf_conntrack_h323_types.h delete mode 100644 include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h delete mode 100644 include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h delete mode 100644 net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c delete mode 100644 net/ipv4/netfilter/ip_conntrack_helper_h323_types.c create mode 100644 net/ipv4/netfilter/nf_nat_h323.c create mode 100644 net/netfilter/nf_conntrack_h323_asn1.c create mode 100644 net/netfilter/nf_conntrack_h323_main.c create mode 100644 net/netfilter/nf_conntrack_h323_types.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_h323.h b/include/linux/netfilter/nf_conntrack_h323.h new file mode 100644 index 000000000000..08e2f4977c2e --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323.h @@ -0,0 +1,92 @@ +#ifndef _NF_CONNTRACK_H323_H +#define _NF_CONNTRACK_H323_H + +#ifdef __KERNEL__ + +#include + +#define RAS_PORT 1719 +#define Q931_PORT 1720 +#define H323_RTP_CHANNEL_MAX 4 /* Audio, video, FAX and other */ + +/* This structure exists only once per master */ +struct nf_ct_h323_master { + + /* Original and NATed Q.931 or H.245 signal ports */ + __be16 sig_port[IP_CT_DIR_MAX]; + + /* Original and NATed RTP ports */ + __be16 rtp_port[H323_RTP_CHANNEL_MAX][IP_CT_DIR_MAX]; + + union { + /* RAS connection timeout */ + u_int32_t timeout; + + /* Next TPKT length (for separate TPKT header and data) */ + u_int16_t tpkt_len[IP_CT_DIR_MAX]; + }; +}; + +struct nf_conn; + +extern int get_h225_addr(struct nf_conn *ct, unsigned char *data, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port); +extern void nf_conntrack_h245_expect(struct nf_conn *new, + struct nf_conntrack_expect *this); +extern void nf_conntrack_q931_expect(struct nf_conn *new, + struct nf_conntrack_expect *this); +extern int (*set_h245_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, + __be16 port); +extern int (*set_h225_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, + __be16 port); +extern int (*set_sig_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count); +extern int (*set_ras_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count); +extern int (*nat_rtp_rtcp_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp); +extern int (*nat_t120_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_h245_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_callforwarding_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, + __be16 port, + struct nf_conntrack_expect *exp); +extern int (*nat_q931_hook) (struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, + int idx, __be16 port, + struct nf_conntrack_expect *exp); + +#endif + +#endif diff --git a/include/linux/netfilter/nf_conntrack_h323_asn1.h b/include/linux/netfilter/nf_conntrack_h323_asn1.h new file mode 100644 index 000000000000..8dab5968fc7e --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323_asn1.h @@ -0,0 +1,98 @@ +/**************************************************************************** + * ip_conntrack_h323_asn1.h - BER and PER decoding library for H.323 + * conntrack/NAT module. + * + * Copyright (c) 2006 by Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * + * This library is based on H.225 version 4, H.235 version 2 and H.245 + * version 7. It is extremely optimized to decode only the absolutely + * necessary objects in a signal for Linux kernel NAT module use, so don't + * expect it to be a full ASN.1 library. + * + * Features: + * + * 1. Small. The total size of code plus data is less than 20 KB (IA32). + * 2. Fast. Decoding Netmeeting's Setup signal 1 million times on a PIII 866 + * takes only 3.9 seconds. + * 3. No memory allocation. It uses a static object. No need to initialize or + * cleanup. + * 4. Thread safe. + * 5. Support embedded architectures that has no misaligned memory access + * support. + * + * Limitations: + * + * 1. At most 30 faststart entries. Actually this is limited by ethernet's MTU. + * If a Setup signal contains more than 30 faststart, the packet size will + * very likely exceed the MTU size, then the TPKT will be fragmented. I + * don't know how to handle this in a Netfilter module. Anybody can help? + * Although I think 30 is enough for most of the cases. + * 2. IPv4 addresses only. + * + ****************************************************************************/ + +#ifndef _NF_CONNTRACK_HELPER_H323_ASN1_H_ +#define _NF_CONNTRACK_HELPER_H323_ASN1_H_ + +/***************************************************************************** + * H.323 Types + ****************************************************************************/ +#include "nf_conntrack_h323_types.h" + +typedef struct { + enum { + Q931_NationalEscape = 0x00, + Q931_Alerting = 0x01, + Q931_CallProceeding = 0x02, + Q931_Connect = 0x07, + Q931_ConnectAck = 0x0F, + Q931_Progress = 0x03, + Q931_Setup = 0x05, + Q931_SetupAck = 0x0D, + Q931_Resume = 0x26, + Q931_ResumeAck = 0x2E, + Q931_ResumeReject = 0x22, + Q931_Suspend = 0x25, + Q931_SuspendAck = 0x2D, + Q931_SuspendReject = 0x21, + Q931_UserInformation = 0x20, + Q931_Disconnect = 0x45, + Q931_Release = 0x4D, + Q931_ReleaseComplete = 0x5A, + Q931_Restart = 0x46, + Q931_RestartAck = 0x4E, + Q931_Segment = 0x60, + Q931_CongestionCtrl = 0x79, + Q931_Information = 0x7B, + Q931_Notify = 0x6E, + Q931_Status = 0x7D, + Q931_StatusEnquiry = 0x75, + Q931_Facility = 0x62 + } MessageType; + H323_UserInformation UUIE; +} Q931; + +/***************************************************************************** + * Decode Functions Return Codes + ****************************************************************************/ + +#define H323_ERROR_NONE 0 /* Decoded successfully */ +#define H323_ERROR_STOP 1 /* Decoding stopped, not really an error */ +#define H323_ERROR_BOUND -1 +#define H323_ERROR_RANGE -2 + + +/***************************************************************************** + * Decode Functions + ****************************************************************************/ + +int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras); +int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931); +int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, + MultimediaSystemControlMessage * + mscm); + +#endif diff --git a/include/linux/netfilter/nf_conntrack_h323_types.h b/include/linux/netfilter/nf_conntrack_h323_types.h new file mode 100644 index 000000000000..38d74d5c9700 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_h323_types.h @@ -0,0 +1,951 @@ +/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + */ + +typedef struct TransportAddress_ipAddress { /* SEQUENCE */ + int options; /* No use */ + unsigned ip; +} TransportAddress_ipAddress; + +typedef struct TransportAddress_ip6Address { /* SEQUENCE */ + int options; /* No use */ + unsigned ip6; +} TransportAddress_ip6Address; + +typedef struct TransportAddress { /* CHOICE */ + enum { + eTransportAddress_ipAddress, + eTransportAddress_ipSourceRoute, + eTransportAddress_ipxAddress, + eTransportAddress_ip6Address, + eTransportAddress_netBios, + eTransportAddress_nsap, + eTransportAddress_nonStandardAddress, + } choice; + union { + TransportAddress_ipAddress ipAddress; + TransportAddress_ip6Address ip6Address; + }; +} TransportAddress; + +typedef struct DataProtocolCapability { /* CHOICE */ + enum { + eDataProtocolCapability_nonStandard, + eDataProtocolCapability_v14buffered, + eDataProtocolCapability_v42lapm, + eDataProtocolCapability_hdlcFrameTunnelling, + eDataProtocolCapability_h310SeparateVCStack, + eDataProtocolCapability_h310SingleVCStack, + eDataProtocolCapability_transparent, + eDataProtocolCapability_segmentationAndReassembly, + eDataProtocolCapability_hdlcFrameTunnelingwSAR, + eDataProtocolCapability_v120, + eDataProtocolCapability_separateLANStack, + eDataProtocolCapability_v76wCompression, + eDataProtocolCapability_tcp, + eDataProtocolCapability_udp, + } choice; +} DataProtocolCapability; + +typedef struct DataApplicationCapability_application { /* CHOICE */ + enum { + eDataApplicationCapability_application_nonStandard, + eDataApplicationCapability_application_t120, + eDataApplicationCapability_application_dsm_cc, + eDataApplicationCapability_application_userData, + eDataApplicationCapability_application_t84, + eDataApplicationCapability_application_t434, + eDataApplicationCapability_application_h224, + eDataApplicationCapability_application_nlpid, + eDataApplicationCapability_application_dsvdControl, + eDataApplicationCapability_application_h222DataPartitioning, + eDataApplicationCapability_application_t30fax, + eDataApplicationCapability_application_t140, + eDataApplicationCapability_application_t38fax, + eDataApplicationCapability_application_genericDataCapability, + } choice; + union { + DataProtocolCapability t120; + }; +} DataApplicationCapability_application; + +typedef struct DataApplicationCapability { /* SEQUENCE */ + int options; /* No use */ + DataApplicationCapability_application application; +} DataApplicationCapability; + +typedef struct DataType { /* CHOICE */ + enum { + eDataType_nonStandard, + eDataType_nullData, + eDataType_videoData, + eDataType_audioData, + eDataType_data, + eDataType_encryptionData, + eDataType_h235Control, + eDataType_h235Media, + eDataType_multiplexedStream, + } choice; + union { + DataApplicationCapability data; + }; +} DataType; + +typedef struct UnicastAddress_iPAddress { /* SEQUENCE */ + int options; /* No use */ + unsigned network; +} UnicastAddress_iPAddress; + +typedef struct UnicastAddress_iP6Address { /* SEQUENCE */ + int options; /* No use */ + unsigned network; +} UnicastAddress_iP6Address; + +typedef struct UnicastAddress { /* CHOICE */ + enum { + eUnicastAddress_iPAddress, + eUnicastAddress_iPXAddress, + eUnicastAddress_iP6Address, + eUnicastAddress_netBios, + eUnicastAddress_iPSourceRouteAddress, + eUnicastAddress_nsap, + eUnicastAddress_nonStandardAddress, + } choice; + union { + UnicastAddress_iPAddress iPAddress; + UnicastAddress_iP6Address iP6Address; + }; +} UnicastAddress; + +typedef struct H245_TransportAddress { /* CHOICE */ + enum { + eH245_TransportAddress_unicastAddress, + eH245_TransportAddress_multicastAddress, + } choice; + union { + UnicastAddress unicastAddress; + }; +} H245_TransportAddress; + +typedef struct H2250LogicalChannelParameters { /* SEQUENCE */ + enum { + eH2250LogicalChannelParameters_nonStandard = (1 << 31), + eH2250LogicalChannelParameters_associatedSessionID = + (1 << 30), + eH2250LogicalChannelParameters_mediaChannel = (1 << 29), + eH2250LogicalChannelParameters_mediaGuaranteedDelivery = + (1 << 28), + eH2250LogicalChannelParameters_mediaControlChannel = + (1 << 27), + eH2250LogicalChannelParameters_mediaControlGuaranteedDelivery + = (1 << 26), + eH2250LogicalChannelParameters_silenceSuppression = (1 << 25), + eH2250LogicalChannelParameters_destination = (1 << 24), + eH2250LogicalChannelParameters_dynamicRTPPayloadType = + (1 << 23), + eH2250LogicalChannelParameters_mediaPacketization = (1 << 22), + eH2250LogicalChannelParameters_transportCapability = + (1 << 21), + eH2250LogicalChannelParameters_redundancyEncoding = (1 << 20), + eH2250LogicalChannelParameters_source = (1 << 19), + } options; + H245_TransportAddress mediaChannel; + H245_TransportAddress mediaControlChannel; +} H2250LogicalChannelParameters; + +typedef struct OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_none, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannel_forwardLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannel_forwardLogicalChannelParameters_portNumber + = (1 << 31), + eOpenLogicalChannel_forwardLogicalChannelParameters_forwardLogicalChannelDependency + = (1 << 30), + eOpenLogicalChannel_forwardLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + DataType dataType; + OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannel_forwardLogicalChannelParameters; + +typedef struct OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannel_reverseLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters + = (1 << 31), + eOpenLogicalChannel_reverseLogicalChannelParameters_reverseLogicalChannelDependency + = (1 << 30), + eOpenLogicalChannel_reverseLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannel_reverseLogicalChannelParameters; + +typedef struct NetworkAccessParameters_networkAddress { /* CHOICE */ + enum { + eNetworkAccessParameters_networkAddress_q2931Address, + eNetworkAccessParameters_networkAddress_e164Address, + eNetworkAccessParameters_networkAddress_localAreaAddress, + } choice; + union { + H245_TransportAddress localAreaAddress; + }; +} NetworkAccessParameters_networkAddress; + +typedef struct NetworkAccessParameters { /* SEQUENCE */ + enum { + eNetworkAccessParameters_distribution = (1 << 31), + eNetworkAccessParameters_externalReference = (1 << 30), + eNetworkAccessParameters_t120SetupProcedure = (1 << 29), + } options; + NetworkAccessParameters_networkAddress networkAddress; +} NetworkAccessParameters; + +typedef struct OpenLogicalChannel { /* SEQUENCE */ + enum { + eOpenLogicalChannel_reverseLogicalChannelParameters = + (1 << 31), + eOpenLogicalChannel_separateStack = (1 << 30), + eOpenLogicalChannel_encryptionSync = (1 << 29), + } options; + OpenLogicalChannel_forwardLogicalChannelParameters + forwardLogicalChannelParameters; + OpenLogicalChannel_reverseLogicalChannelParameters + reverseLogicalChannelParameters; + NetworkAccessParameters separateStack; +} OpenLogicalChannel; + +typedef struct Setup_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Setup_UUIE_fastStart; + +typedef struct Setup_UUIE { /* SEQUENCE */ + enum { + eSetup_UUIE_h245Address = (1 << 31), + eSetup_UUIE_sourceAddress = (1 << 30), + eSetup_UUIE_destinationAddress = (1 << 29), + eSetup_UUIE_destCallSignalAddress = (1 << 28), + eSetup_UUIE_destExtraCallInfo = (1 << 27), + eSetup_UUIE_destExtraCRV = (1 << 26), + eSetup_UUIE_callServices = (1 << 25), + eSetup_UUIE_sourceCallSignalAddress = (1 << 24), + eSetup_UUIE_remoteExtensionAddress = (1 << 23), + eSetup_UUIE_callIdentifier = (1 << 22), + eSetup_UUIE_h245SecurityCapability = (1 << 21), + eSetup_UUIE_tokens = (1 << 20), + eSetup_UUIE_cryptoTokens = (1 << 19), + eSetup_UUIE_fastStart = (1 << 18), + eSetup_UUIE_mediaWaitForConnect = (1 << 17), + eSetup_UUIE_canOverlapSend = (1 << 16), + eSetup_UUIE_endpointIdentifier = (1 << 15), + eSetup_UUIE_multipleCalls = (1 << 14), + eSetup_UUIE_maintainConnection = (1 << 13), + eSetup_UUIE_connectionParameters = (1 << 12), + eSetup_UUIE_language = (1 << 11), + eSetup_UUIE_presentationIndicator = (1 << 10), + eSetup_UUIE_screeningIndicator = (1 << 9), + eSetup_UUIE_serviceControl = (1 << 8), + eSetup_UUIE_symmetricOperationRequired = (1 << 7), + eSetup_UUIE_capacity = (1 << 6), + eSetup_UUIE_circuitInfo = (1 << 5), + eSetup_UUIE_desiredProtocols = (1 << 4), + eSetup_UUIE_neededFeatures = (1 << 3), + eSetup_UUIE_desiredFeatures = (1 << 2), + eSetup_UUIE_supportedFeatures = (1 << 1), + eSetup_UUIE_parallelH245Control = (1 << 0), + } options; + TransportAddress h245Address; + TransportAddress destCallSignalAddress; + TransportAddress sourceCallSignalAddress; + Setup_UUIE_fastStart fastStart; +} Setup_UUIE; + +typedef struct CallProceeding_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} CallProceeding_UUIE_fastStart; + +typedef struct CallProceeding_UUIE { /* SEQUENCE */ + enum { + eCallProceeding_UUIE_h245Address = (1 << 31), + eCallProceeding_UUIE_callIdentifier = (1 << 30), + eCallProceeding_UUIE_h245SecurityMode = (1 << 29), + eCallProceeding_UUIE_tokens = (1 << 28), + eCallProceeding_UUIE_cryptoTokens = (1 << 27), + eCallProceeding_UUIE_fastStart = (1 << 26), + eCallProceeding_UUIE_multipleCalls = (1 << 25), + eCallProceeding_UUIE_maintainConnection = (1 << 24), + eCallProceeding_UUIE_fastConnectRefused = (1 << 23), + eCallProceeding_UUIE_featureSet = (1 << 22), + } options; + TransportAddress h245Address; + CallProceeding_UUIE_fastStart fastStart; +} CallProceeding_UUIE; + +typedef struct Connect_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Connect_UUIE_fastStart; + +typedef struct Connect_UUIE { /* SEQUENCE */ + enum { + eConnect_UUIE_h245Address = (1 << 31), + eConnect_UUIE_callIdentifier = (1 << 30), + eConnect_UUIE_h245SecurityMode = (1 << 29), + eConnect_UUIE_tokens = (1 << 28), + eConnect_UUIE_cryptoTokens = (1 << 27), + eConnect_UUIE_fastStart = (1 << 26), + eConnect_UUIE_multipleCalls = (1 << 25), + eConnect_UUIE_maintainConnection = (1 << 24), + eConnect_UUIE_language = (1 << 23), + eConnect_UUIE_connectedAddress = (1 << 22), + eConnect_UUIE_presentationIndicator = (1 << 21), + eConnect_UUIE_screeningIndicator = (1 << 20), + eConnect_UUIE_fastConnectRefused = (1 << 19), + eConnect_UUIE_serviceControl = (1 << 18), + eConnect_UUIE_capacity = (1 << 17), + eConnect_UUIE_featureSet = (1 << 16), + } options; + TransportAddress h245Address; + Connect_UUIE_fastStart fastStart; +} Connect_UUIE; + +typedef struct Alerting_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Alerting_UUIE_fastStart; + +typedef struct Alerting_UUIE { /* SEQUENCE */ + enum { + eAlerting_UUIE_h245Address = (1 << 31), + eAlerting_UUIE_callIdentifier = (1 << 30), + eAlerting_UUIE_h245SecurityMode = (1 << 29), + eAlerting_UUIE_tokens = (1 << 28), + eAlerting_UUIE_cryptoTokens = (1 << 27), + eAlerting_UUIE_fastStart = (1 << 26), + eAlerting_UUIE_multipleCalls = (1 << 25), + eAlerting_UUIE_maintainConnection = (1 << 24), + eAlerting_UUIE_alertingAddress = (1 << 23), + eAlerting_UUIE_presentationIndicator = (1 << 22), + eAlerting_UUIE_screeningIndicator = (1 << 21), + eAlerting_UUIE_fastConnectRefused = (1 << 20), + eAlerting_UUIE_serviceControl = (1 << 19), + eAlerting_UUIE_capacity = (1 << 18), + eAlerting_UUIE_featureSet = (1 << 17), + } options; + TransportAddress h245Address; + Alerting_UUIE_fastStart fastStart; +} Alerting_UUIE; + +typedef struct Information_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Information_UUIE_fastStart; + +typedef struct Information_UUIE { /* SEQUENCE */ + enum { + eInformation_UUIE_callIdentifier = (1 << 31), + eInformation_UUIE_tokens = (1 << 30), + eInformation_UUIE_cryptoTokens = (1 << 29), + eInformation_UUIE_fastStart = (1 << 28), + eInformation_UUIE_fastConnectRefused = (1 << 27), + eInformation_UUIE_circuitInfo = (1 << 26), + } options; + Information_UUIE_fastStart fastStart; +} Information_UUIE; + +typedef struct FacilityReason { /* CHOICE */ + enum { + eFacilityReason_routeCallToGatekeeper, + eFacilityReason_callForwarded, + eFacilityReason_routeCallToMC, + eFacilityReason_undefinedReason, + eFacilityReason_conferenceListChoice, + eFacilityReason_startH245, + eFacilityReason_noH245, + eFacilityReason_newTokens, + eFacilityReason_featureSetUpdate, + eFacilityReason_forwardedElements, + eFacilityReason_transportedInformation, + } choice; +} FacilityReason; + +typedef struct Facility_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Facility_UUIE_fastStart; + +typedef struct Facility_UUIE { /* SEQUENCE */ + enum { + eFacility_UUIE_alternativeAddress = (1 << 31), + eFacility_UUIE_alternativeAliasAddress = (1 << 30), + eFacility_UUIE_conferenceID = (1 << 29), + eFacility_UUIE_callIdentifier = (1 << 28), + eFacility_UUIE_destExtraCallInfo = (1 << 27), + eFacility_UUIE_remoteExtensionAddress = (1 << 26), + eFacility_UUIE_tokens = (1 << 25), + eFacility_UUIE_cryptoTokens = (1 << 24), + eFacility_UUIE_conferences = (1 << 23), + eFacility_UUIE_h245Address = (1 << 22), + eFacility_UUIE_fastStart = (1 << 21), + eFacility_UUIE_multipleCalls = (1 << 20), + eFacility_UUIE_maintainConnection = (1 << 19), + eFacility_UUIE_fastConnectRefused = (1 << 18), + eFacility_UUIE_serviceControl = (1 << 17), + eFacility_UUIE_circuitInfo = (1 << 16), + eFacility_UUIE_featureSet = (1 << 15), + eFacility_UUIE_destinationInfo = (1 << 14), + eFacility_UUIE_h245SecurityMode = (1 << 13), + } options; + TransportAddress alternativeAddress; + FacilityReason reason; + TransportAddress h245Address; + Facility_UUIE_fastStart fastStart; +} Facility_UUIE; + +typedef struct Progress_UUIE_fastStart { /* SEQUENCE OF */ + int count; + OpenLogicalChannel item[30]; +} Progress_UUIE_fastStart; + +typedef struct Progress_UUIE { /* SEQUENCE */ + enum { + eProgress_UUIE_h245Address = (1 << 31), + eProgress_UUIE_h245SecurityMode = (1 << 30), + eProgress_UUIE_tokens = (1 << 29), + eProgress_UUIE_cryptoTokens = (1 << 28), + eProgress_UUIE_fastStart = (1 << 27), + eProgress_UUIE_multipleCalls = (1 << 26), + eProgress_UUIE_maintainConnection = (1 << 25), + eProgress_UUIE_fastConnectRefused = (1 << 24), + } options; + TransportAddress h245Address; + Progress_UUIE_fastStart fastStart; +} Progress_UUIE; + +typedef struct H323_UU_PDU_h323_message_body { /* CHOICE */ + enum { + eH323_UU_PDU_h323_message_body_setup, + eH323_UU_PDU_h323_message_body_callProceeding, + eH323_UU_PDU_h323_message_body_connect, + eH323_UU_PDU_h323_message_body_alerting, + eH323_UU_PDU_h323_message_body_information, + eH323_UU_PDU_h323_message_body_releaseComplete, + eH323_UU_PDU_h323_message_body_facility, + eH323_UU_PDU_h323_message_body_progress, + eH323_UU_PDU_h323_message_body_empty, + eH323_UU_PDU_h323_message_body_status, + eH323_UU_PDU_h323_message_body_statusInquiry, + eH323_UU_PDU_h323_message_body_setupAcknowledge, + eH323_UU_PDU_h323_message_body_notify, + } choice; + union { + Setup_UUIE setup; + CallProceeding_UUIE callProceeding; + Connect_UUIE connect; + Alerting_UUIE alerting; + Information_UUIE information; + Facility_UUIE facility; + Progress_UUIE progress; + }; +} H323_UU_PDU_h323_message_body; + +typedef struct RequestMessage { /* CHOICE */ + enum { + eRequestMessage_nonStandard, + eRequestMessage_masterSlaveDetermination, + eRequestMessage_terminalCapabilitySet, + eRequestMessage_openLogicalChannel, + eRequestMessage_closeLogicalChannel, + eRequestMessage_requestChannelClose, + eRequestMessage_multiplexEntrySend, + eRequestMessage_requestMultiplexEntry, + eRequestMessage_requestMode, + eRequestMessage_roundTripDelayRequest, + eRequestMessage_maintenanceLoopRequest, + eRequestMessage_communicationModeRequest, + eRequestMessage_conferenceRequest, + eRequestMessage_multilinkRequest, + eRequestMessage_logicalChannelRateRequest, + } choice; + union { + OpenLogicalChannel openLogicalChannel; + }; +} RequestMessage; + +typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, + } choice; + union { + H2250LogicalChannelParameters h2250LogicalChannelParameters; + }; +} OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters; + +typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters { /* SEQUENCE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters_portNumber + = (1 << 31), + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters + = (1 << 30), + eOpenLogicalChannelAck_reverseLogicalChannelParameters_replacementFor + = (1 << 29), + } options; + OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters + multiplexParameters; +} OpenLogicalChannelAck_reverseLogicalChannelParameters; + +typedef struct H2250LogicalChannelAckParameters { /* SEQUENCE */ + enum { + eH2250LogicalChannelAckParameters_nonStandard = (1 << 31), + eH2250LogicalChannelAckParameters_sessionID = (1 << 30), + eH2250LogicalChannelAckParameters_mediaChannel = (1 << 29), + eH2250LogicalChannelAckParameters_mediaControlChannel = + (1 << 28), + eH2250LogicalChannelAckParameters_dynamicRTPPayloadType = + (1 << 27), + eH2250LogicalChannelAckParameters_flowControlToZero = + (1 << 26), + eH2250LogicalChannelAckParameters_portNumber = (1 << 25), + } options; + H245_TransportAddress mediaChannel; + H245_TransportAddress mediaControlChannel; +} H2250LogicalChannelAckParameters; + +typedef struct OpenLogicalChannelAck_forwardMultiplexAckParameters { /* CHOICE */ + enum { + eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters, + } choice; + union { + H2250LogicalChannelAckParameters + h2250LogicalChannelAckParameters; + }; +} OpenLogicalChannelAck_forwardMultiplexAckParameters; + +typedef struct OpenLogicalChannelAck { /* SEQUENCE */ + enum { + eOpenLogicalChannelAck_reverseLogicalChannelParameters = + (1 << 31), + eOpenLogicalChannelAck_separateStack = (1 << 30), + eOpenLogicalChannelAck_forwardMultiplexAckParameters = + (1 << 29), + eOpenLogicalChannelAck_encryptionSync = (1 << 28), + } options; + OpenLogicalChannelAck_reverseLogicalChannelParameters + reverseLogicalChannelParameters; + OpenLogicalChannelAck_forwardMultiplexAckParameters + forwardMultiplexAckParameters; +} OpenLogicalChannelAck; + +typedef struct ResponseMessage { /* CHOICE */ + enum { + eResponseMessage_nonStandard, + eResponseMessage_masterSlaveDeterminationAck, + eResponseMessage_masterSlaveDeterminationReject, + eResponseMessage_terminalCapabilitySetAck, + eResponseMessage_terminalCapabilitySetReject, + eResponseMessage_openLogicalChannelAck, + eResponseMessage_openLogicalChannelReject, + eResponseMessage_closeLogicalChannelAck, + eResponseMessage_requestChannelCloseAck, + eResponseMessage_requestChannelCloseReject, + eResponseMessage_multiplexEntrySendAck, + eResponseMessage_multiplexEntrySendReject, + eResponseMessage_requestMultiplexEntryAck, + eResponseMessage_requestMultiplexEntryReject, + eResponseMessage_requestModeAck, + eResponseMessage_requestModeReject, + eResponseMessage_roundTripDelayResponse, + eResponseMessage_maintenanceLoopAck, + eResponseMessage_maintenanceLoopReject, + eResponseMessage_communicationModeResponse, + eResponseMessage_conferenceResponse, + eResponseMessage_multilinkResponse, + eResponseMessage_logicalChannelRateAcknowledge, + eResponseMessage_logicalChannelRateReject, + } choice; + union { + OpenLogicalChannelAck openLogicalChannelAck; + }; +} ResponseMessage; + +typedef struct MultimediaSystemControlMessage { /* CHOICE */ + enum { + eMultimediaSystemControlMessage_request, + eMultimediaSystemControlMessage_response, + eMultimediaSystemControlMessage_command, + eMultimediaSystemControlMessage_indication, + } choice; + union { + RequestMessage request; + ResponseMessage response; + }; +} MultimediaSystemControlMessage; + +typedef struct H323_UU_PDU_h245Control { /* SEQUENCE OF */ + int count; + MultimediaSystemControlMessage item[4]; +} H323_UU_PDU_h245Control; + +typedef struct H323_UU_PDU { /* SEQUENCE */ + enum { + eH323_UU_PDU_nonStandardData = (1 << 31), + eH323_UU_PDU_h4501SupplementaryService = (1 << 30), + eH323_UU_PDU_h245Tunneling = (1 << 29), + eH323_UU_PDU_h245Control = (1 << 28), + eH323_UU_PDU_nonStandardControl = (1 << 27), + eH323_UU_PDU_callLinkage = (1 << 26), + eH323_UU_PDU_tunnelledSignallingMessage = (1 << 25), + eH323_UU_PDU_provisionalRespToH245Tunneling = (1 << 24), + eH323_UU_PDU_stimulusControl = (1 << 23), + eH323_UU_PDU_genericData = (1 << 22), + } options; + H323_UU_PDU_h323_message_body h323_message_body; + H323_UU_PDU_h245Control h245Control; +} H323_UU_PDU; + +typedef struct H323_UserInformation { /* SEQUENCE */ + enum { + eH323_UserInformation_user_data = (1 << 31), + } options; + H323_UU_PDU h323_uu_pdu; +} H323_UserInformation; + +typedef struct GatekeeperRequest { /* SEQUENCE */ + enum { + eGatekeeperRequest_nonStandardData = (1 << 31), + eGatekeeperRequest_gatekeeperIdentifier = (1 << 30), + eGatekeeperRequest_callServices = (1 << 29), + eGatekeeperRequest_endpointAlias = (1 << 28), + eGatekeeperRequest_alternateEndpoints = (1 << 27), + eGatekeeperRequest_tokens = (1 << 26), + eGatekeeperRequest_cryptoTokens = (1 << 25), + eGatekeeperRequest_authenticationCapability = (1 << 24), + eGatekeeperRequest_algorithmOIDs = (1 << 23), + eGatekeeperRequest_integrity = (1 << 22), + eGatekeeperRequest_integrityCheckValue = (1 << 21), + eGatekeeperRequest_supportsAltGK = (1 << 20), + eGatekeeperRequest_featureSet = (1 << 19), + eGatekeeperRequest_genericData = (1 << 18), + } options; + TransportAddress rasAddress; +} GatekeeperRequest; + +typedef struct GatekeeperConfirm { /* SEQUENCE */ + enum { + eGatekeeperConfirm_nonStandardData = (1 << 31), + eGatekeeperConfirm_gatekeeperIdentifier = (1 << 30), + eGatekeeperConfirm_alternateGatekeeper = (1 << 29), + eGatekeeperConfirm_authenticationMode = (1 << 28), + eGatekeeperConfirm_tokens = (1 << 27), + eGatekeeperConfirm_cryptoTokens = (1 << 26), + eGatekeeperConfirm_algorithmOID = (1 << 25), + eGatekeeperConfirm_integrity = (1 << 24), + eGatekeeperConfirm_integrityCheckValue = (1 << 23), + eGatekeeperConfirm_featureSet = (1 << 22), + eGatekeeperConfirm_genericData = (1 << 21), + } options; + TransportAddress rasAddress; +} GatekeeperConfirm; + +typedef struct RegistrationRequest_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationRequest_callSignalAddress; + +typedef struct RegistrationRequest_rasAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationRequest_rasAddress; + +typedef struct RegistrationRequest { /* SEQUENCE */ + enum { + eRegistrationRequest_nonStandardData = (1 << 31), + eRegistrationRequest_terminalAlias = (1 << 30), + eRegistrationRequest_gatekeeperIdentifier = (1 << 29), + eRegistrationRequest_alternateEndpoints = (1 << 28), + eRegistrationRequest_timeToLive = (1 << 27), + eRegistrationRequest_tokens = (1 << 26), + eRegistrationRequest_cryptoTokens = (1 << 25), + eRegistrationRequest_integrityCheckValue = (1 << 24), + eRegistrationRequest_keepAlive = (1 << 23), + eRegistrationRequest_endpointIdentifier = (1 << 22), + eRegistrationRequest_willSupplyUUIEs = (1 << 21), + eRegistrationRequest_maintainConnection = (1 << 20), + eRegistrationRequest_alternateTransportAddresses = (1 << 19), + eRegistrationRequest_additiveRegistration = (1 << 18), + eRegistrationRequest_terminalAliasPattern = (1 << 17), + eRegistrationRequest_supportsAltGK = (1 << 16), + eRegistrationRequest_usageReportingCapability = (1 << 15), + eRegistrationRequest_multipleCalls = (1 << 14), + eRegistrationRequest_supportedH248Packages = (1 << 13), + eRegistrationRequest_callCreditCapability = (1 << 12), + eRegistrationRequest_capacityReportingCapability = (1 << 11), + eRegistrationRequest_capacity = (1 << 10), + eRegistrationRequest_featureSet = (1 << 9), + eRegistrationRequest_genericData = (1 << 8), + } options; + RegistrationRequest_callSignalAddress callSignalAddress; + RegistrationRequest_rasAddress rasAddress; + unsigned timeToLive; +} RegistrationRequest; + +typedef struct RegistrationConfirm_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} RegistrationConfirm_callSignalAddress; + +typedef struct RegistrationConfirm { /* SEQUENCE */ + enum { + eRegistrationConfirm_nonStandardData = (1 << 31), + eRegistrationConfirm_terminalAlias = (1 << 30), + eRegistrationConfirm_gatekeeperIdentifier = (1 << 29), + eRegistrationConfirm_alternateGatekeeper = (1 << 28), + eRegistrationConfirm_timeToLive = (1 << 27), + eRegistrationConfirm_tokens = (1 << 26), + eRegistrationConfirm_cryptoTokens = (1 << 25), + eRegistrationConfirm_integrityCheckValue = (1 << 24), + eRegistrationConfirm_willRespondToIRR = (1 << 23), + eRegistrationConfirm_preGrantedARQ = (1 << 22), + eRegistrationConfirm_maintainConnection = (1 << 21), + eRegistrationConfirm_serviceControl = (1 << 20), + eRegistrationConfirm_supportsAdditiveRegistration = (1 << 19), + eRegistrationConfirm_terminalAliasPattern = (1 << 18), + eRegistrationConfirm_supportedPrefixes = (1 << 17), + eRegistrationConfirm_usageSpec = (1 << 16), + eRegistrationConfirm_featureServerAlias = (1 << 15), + eRegistrationConfirm_capacityReportingSpec = (1 << 14), + eRegistrationConfirm_featureSet = (1 << 13), + eRegistrationConfirm_genericData = (1 << 12), + } options; + RegistrationConfirm_callSignalAddress callSignalAddress; + unsigned timeToLive; +} RegistrationConfirm; + +typedef struct UnregistrationRequest_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} UnregistrationRequest_callSignalAddress; + +typedef struct UnregistrationRequest { /* SEQUENCE */ + enum { + eUnregistrationRequest_endpointAlias = (1 << 31), + eUnregistrationRequest_nonStandardData = (1 << 30), + eUnregistrationRequest_endpointIdentifier = (1 << 29), + eUnregistrationRequest_alternateEndpoints = (1 << 28), + eUnregistrationRequest_gatekeeperIdentifier = (1 << 27), + eUnregistrationRequest_tokens = (1 << 26), + eUnregistrationRequest_cryptoTokens = (1 << 25), + eUnregistrationRequest_integrityCheckValue = (1 << 24), + eUnregistrationRequest_reason = (1 << 23), + eUnregistrationRequest_endpointAliasPattern = (1 << 22), + eUnregistrationRequest_supportedPrefixes = (1 << 21), + eUnregistrationRequest_alternateGatekeeper = (1 << 20), + eUnregistrationRequest_genericData = (1 << 19), + } options; + UnregistrationRequest_callSignalAddress callSignalAddress; +} UnregistrationRequest; + +typedef struct AdmissionRequest { /* SEQUENCE */ + enum { + eAdmissionRequest_callModel = (1 << 31), + eAdmissionRequest_destinationInfo = (1 << 30), + eAdmissionRequest_destCallSignalAddress = (1 << 29), + eAdmissionRequest_destExtraCallInfo = (1 << 28), + eAdmissionRequest_srcCallSignalAddress = (1 << 27), + eAdmissionRequest_nonStandardData = (1 << 26), + eAdmissionRequest_callServices = (1 << 25), + eAdmissionRequest_canMapAlias = (1 << 24), + eAdmissionRequest_callIdentifier = (1 << 23), + eAdmissionRequest_srcAlternatives = (1 << 22), + eAdmissionRequest_destAlternatives = (1 << 21), + eAdmissionRequest_gatekeeperIdentifier = (1 << 20), + eAdmissionRequest_tokens = (1 << 19), + eAdmissionRequest_cryptoTokens = (1 << 18), + eAdmissionRequest_integrityCheckValue = (1 << 17), + eAdmissionRequest_transportQOS = (1 << 16), + eAdmissionRequest_willSupplyUUIEs = (1 << 15), + eAdmissionRequest_callLinkage = (1 << 14), + eAdmissionRequest_gatewayDataRate = (1 << 13), + eAdmissionRequest_capacity = (1 << 12), + eAdmissionRequest_circuitInfo = (1 << 11), + eAdmissionRequest_desiredProtocols = (1 << 10), + eAdmissionRequest_desiredTunnelledProtocol = (1 << 9), + eAdmissionRequest_featureSet = (1 << 8), + eAdmissionRequest_genericData = (1 << 7), + } options; + TransportAddress destCallSignalAddress; + TransportAddress srcCallSignalAddress; +} AdmissionRequest; + +typedef struct AdmissionConfirm { /* SEQUENCE */ + enum { + eAdmissionConfirm_irrFrequency = (1 << 31), + eAdmissionConfirm_nonStandardData = (1 << 30), + eAdmissionConfirm_destinationInfo = (1 << 29), + eAdmissionConfirm_destExtraCallInfo = (1 << 28), + eAdmissionConfirm_destinationType = (1 << 27), + eAdmissionConfirm_remoteExtensionAddress = (1 << 26), + eAdmissionConfirm_alternateEndpoints = (1 << 25), + eAdmissionConfirm_tokens = (1 << 24), + eAdmissionConfirm_cryptoTokens = (1 << 23), + eAdmissionConfirm_integrityCheckValue = (1 << 22), + eAdmissionConfirm_transportQOS = (1 << 21), + eAdmissionConfirm_willRespondToIRR = (1 << 20), + eAdmissionConfirm_uuiesRequested = (1 << 19), + eAdmissionConfirm_language = (1 << 18), + eAdmissionConfirm_alternateTransportAddresses = (1 << 17), + eAdmissionConfirm_useSpecifiedTransport = (1 << 16), + eAdmissionConfirm_circuitInfo = (1 << 15), + eAdmissionConfirm_usageSpec = (1 << 14), + eAdmissionConfirm_supportedProtocols = (1 << 13), + eAdmissionConfirm_serviceControl = (1 << 12), + eAdmissionConfirm_multipleCalls = (1 << 11), + eAdmissionConfirm_featureSet = (1 << 10), + eAdmissionConfirm_genericData = (1 << 9), + } options; + TransportAddress destCallSignalAddress; +} AdmissionConfirm; + +typedef struct LocationRequest { /* SEQUENCE */ + enum { + eLocationRequest_endpointIdentifier = (1 << 31), + eLocationRequest_nonStandardData = (1 << 30), + eLocationRequest_sourceInfo = (1 << 29), + eLocationRequest_canMapAlias = (1 << 28), + eLocationRequest_gatekeeperIdentifier = (1 << 27), + eLocationRequest_tokens = (1 << 26), + eLocationRequest_cryptoTokens = (1 << 25), + eLocationRequest_integrityCheckValue = (1 << 24), + eLocationRequest_desiredProtocols = (1 << 23), + eLocationRequest_desiredTunnelledProtocol = (1 << 22), + eLocationRequest_featureSet = (1 << 21), + eLocationRequest_genericData = (1 << 20), + eLocationRequest_hopCount = (1 << 19), + eLocationRequest_circuitInfo = (1 << 18), + } options; + TransportAddress replyAddress; +} LocationRequest; + +typedef struct LocationConfirm { /* SEQUENCE */ + enum { + eLocationConfirm_nonStandardData = (1 << 31), + eLocationConfirm_destinationInfo = (1 << 30), + eLocationConfirm_destExtraCallInfo = (1 << 29), + eLocationConfirm_destinationType = (1 << 28), + eLocationConfirm_remoteExtensionAddress = (1 << 27), + eLocationConfirm_alternateEndpoints = (1 << 26), + eLocationConfirm_tokens = (1 << 25), + eLocationConfirm_cryptoTokens = (1 << 24), + eLocationConfirm_integrityCheckValue = (1 << 23), + eLocationConfirm_alternateTransportAddresses = (1 << 22), + eLocationConfirm_supportedProtocols = (1 << 21), + eLocationConfirm_multipleCalls = (1 << 20), + eLocationConfirm_featureSet = (1 << 19), + eLocationConfirm_genericData = (1 << 18), + eLocationConfirm_circuitInfo = (1 << 17), + eLocationConfirm_serviceControl = (1 << 16), + } options; + TransportAddress callSignalAddress; + TransportAddress rasAddress; +} LocationConfirm; + +typedef struct InfoRequestResponse_callSignalAddress { /* SEQUENCE OF */ + int count; + TransportAddress item[10]; +} InfoRequestResponse_callSignalAddress; + +typedef struct InfoRequestResponse { /* SEQUENCE */ + enum { + eInfoRequestResponse_nonStandardData = (1 << 31), + eInfoRequestResponse_endpointAlias = (1 << 30), + eInfoRequestResponse_perCallInfo = (1 << 29), + eInfoRequestResponse_tokens = (1 << 28), + eInfoRequestResponse_cryptoTokens = (1 << 27), + eInfoRequestResponse_integrityCheckValue = (1 << 26), + eInfoRequestResponse_needResponse = (1 << 25), + eInfoRequestResponse_capacity = (1 << 24), + eInfoRequestResponse_irrStatus = (1 << 23), + eInfoRequestResponse_unsolicited = (1 << 22), + eInfoRequestResponse_genericData = (1 << 21), + } options; + TransportAddress rasAddress; + InfoRequestResponse_callSignalAddress callSignalAddress; +} InfoRequestResponse; + +typedef struct RasMessage { /* CHOICE */ + enum { + eRasMessage_gatekeeperRequest, + eRasMessage_gatekeeperConfirm, + eRasMessage_gatekeeperReject, + eRasMessage_registrationRequest, + eRasMessage_registrationConfirm, + eRasMessage_registrationReject, + eRasMessage_unregistrationRequest, + eRasMessage_unregistrationConfirm, + eRasMessage_unregistrationReject, + eRasMessage_admissionRequest, + eRasMessage_admissionConfirm, + eRasMessage_admissionReject, + eRasMessage_bandwidthRequest, + eRasMessage_bandwidthConfirm, + eRasMessage_bandwidthReject, + eRasMessage_disengageRequest, + eRasMessage_disengageConfirm, + eRasMessage_disengageReject, + eRasMessage_locationRequest, + eRasMessage_locationConfirm, + eRasMessage_locationReject, + eRasMessage_infoRequest, + eRasMessage_infoRequestResponse, + eRasMessage_nonStandardMessage, + eRasMessage_unknownMessageResponse, + eRasMessage_requestInProgress, + eRasMessage_resourcesAvailableIndicate, + eRasMessage_resourcesAvailableConfirm, + eRasMessage_infoRequestAck, + eRasMessage_infoRequestNak, + eRasMessage_serviceControlIndication, + eRasMessage_serviceControlResponse, + } choice; + union { + GatekeeperRequest gatekeeperRequest; + GatekeeperConfirm gatekeeperConfirm; + RegistrationRequest registrationRequest; + RegistrationConfirm registrationConfirm; + UnregistrationRequest unregistrationRequest; + AdmissionRequest admissionRequest; + AdmissionConfirm admissionConfirm; + LocationRequest locationRequest; + LocationConfirm locationConfirm; + InfoRequestResponse infoRequestResponse; + }; +} RasMessage; diff --git a/include/linux/netfilter_ipv4/Kbuild b/include/linux/netfilter_ipv4/Kbuild index 591c1a809c00..180337801a86 100644 --- a/include/linux/netfilter_ipv4/Kbuild +++ b/include/linux/netfilter_ipv4/Kbuild @@ -1,6 +1,4 @@ header-y += ip_conntrack_helper.h -header-y += ip_conntrack_helper_h323_asn1.h -header-y += ip_conntrack_helper_h323_types.h header-y += ip_conntrack_protocol.h header-y += ip_conntrack_sctp.h header-y += ip_conntrack_tcp.h diff --git a/include/linux/netfilter_ipv4/ip_conntrack_h323.h b/include/linux/netfilter_ipv4/ip_conntrack_h323.h index 943cc6a4871d..18f769818f4e 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_h323.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_h323.h @@ -3,7 +3,7 @@ #ifdef __KERNEL__ -#include +#include #define RAS_PORT 1719 #define Q931_PORT 1720 diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h b/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h deleted file mode 100644 index c6e9a0b6d30b..000000000000 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_asn1.h +++ /dev/null @@ -1,98 +0,0 @@ -/**************************************************************************** - * ip_conntrack_helper_h323_asn1.h - BER and PER decoding library for H.323 - * conntrack/NAT module. - * - * Copyright (c) 2006 by Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - * - * - * This library is based on H.225 version 4, H.235 version 2 and H.245 - * version 7. It is extremely optimized to decode only the absolutely - * necessary objects in a signal for Linux kernel NAT module use, so don't - * expect it to be a full ASN.1 library. - * - * Features: - * - * 1. Small. The total size of code plus data is less than 20 KB (IA32). - * 2. Fast. Decoding Netmeeting's Setup signal 1 million times on a PIII 866 - * takes only 3.9 seconds. - * 3. No memory allocation. It uses a static object. No need to initialize or - * cleanup. - * 4. Thread safe. - * 5. Support embedded architectures that has no misaligned memory access - * support. - * - * Limitations: - * - * 1. At most 30 faststart entries. Actually this is limited by ethernet's MTU. - * If a Setup signal contains more than 30 faststart, the packet size will - * very likely exceed the MTU size, then the TPKT will be fragmented. I - * don't know how to handle this in a Netfilter module. Anybody can help? - * Although I think 30 is enough for most of the cases. - * 2. IPv4 addresses only. - * - ****************************************************************************/ - -#ifndef _IP_CONNTRACK_HELPER_H323_ASN1_H_ -#define _IP_CONNTRACK_HELPER_H323_ASN1_H_ - -/***************************************************************************** - * H.323 Types - ****************************************************************************/ -#include "ip_conntrack_helper_h323_types.h" - -typedef struct { - enum { - Q931_NationalEscape = 0x00, - Q931_Alerting = 0x01, - Q931_CallProceeding = 0x02, - Q931_Connect = 0x07, - Q931_ConnectAck = 0x0F, - Q931_Progress = 0x03, - Q931_Setup = 0x05, - Q931_SetupAck = 0x0D, - Q931_Resume = 0x26, - Q931_ResumeAck = 0x2E, - Q931_ResumeReject = 0x22, - Q931_Suspend = 0x25, - Q931_SuspendAck = 0x2D, - Q931_SuspendReject = 0x21, - Q931_UserInformation = 0x20, - Q931_Disconnect = 0x45, - Q931_Release = 0x4D, - Q931_ReleaseComplete = 0x5A, - Q931_Restart = 0x46, - Q931_RestartAck = 0x4E, - Q931_Segment = 0x60, - Q931_CongestionCtrl = 0x79, - Q931_Information = 0x7B, - Q931_Notify = 0x6E, - Q931_Status = 0x7D, - Q931_StatusEnquiry = 0x75, - Q931_Facility = 0x62 - } MessageType; - H323_UserInformation UUIE; -} Q931; - -/***************************************************************************** - * Decode Functions Return Codes - ****************************************************************************/ - -#define H323_ERROR_NONE 0 /* Decoded successfully */ -#define H323_ERROR_STOP 1 /* Decoding stopped, not really an error */ -#define H323_ERROR_BOUND -1 -#define H323_ERROR_RANGE -2 - - -/***************************************************************************** - * Decode Functions - ****************************************************************************/ - -int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras); -int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931); -int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, - MultimediaSystemControlMessage * - mscm); - -#endif diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h b/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h deleted file mode 100644 index 3d4a773799fc..000000000000 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h +++ /dev/null @@ -1,939 +0,0 @@ -/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 - * - * Copyright (c) 2006 Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - */ - -typedef struct TransportAddress_ipAddress { /* SEQUENCE */ - int options; /* No use */ - unsigned ip; -} TransportAddress_ipAddress; - -typedef struct TransportAddress { /* CHOICE */ - enum { - eTransportAddress_ipAddress, - eTransportAddress_ipSourceRoute, - eTransportAddress_ipxAddress, - eTransportAddress_ip6Address, - eTransportAddress_netBios, - eTransportAddress_nsap, - eTransportAddress_nonStandardAddress, - } choice; - union { - TransportAddress_ipAddress ipAddress; - }; -} TransportAddress; - -typedef struct DataProtocolCapability { /* CHOICE */ - enum { - eDataProtocolCapability_nonStandard, - eDataProtocolCapability_v14buffered, - eDataProtocolCapability_v42lapm, - eDataProtocolCapability_hdlcFrameTunnelling, - eDataProtocolCapability_h310SeparateVCStack, - eDataProtocolCapability_h310SingleVCStack, - eDataProtocolCapability_transparent, - eDataProtocolCapability_segmentationAndReassembly, - eDataProtocolCapability_hdlcFrameTunnelingwSAR, - eDataProtocolCapability_v120, - eDataProtocolCapability_separateLANStack, - eDataProtocolCapability_v76wCompression, - eDataProtocolCapability_tcp, - eDataProtocolCapability_udp, - } choice; -} DataProtocolCapability; - -typedef struct DataApplicationCapability_application { /* CHOICE */ - enum { - eDataApplicationCapability_application_nonStandard, - eDataApplicationCapability_application_t120, - eDataApplicationCapability_application_dsm_cc, - eDataApplicationCapability_application_userData, - eDataApplicationCapability_application_t84, - eDataApplicationCapability_application_t434, - eDataApplicationCapability_application_h224, - eDataApplicationCapability_application_nlpid, - eDataApplicationCapability_application_dsvdControl, - eDataApplicationCapability_application_h222DataPartitioning, - eDataApplicationCapability_application_t30fax, - eDataApplicationCapability_application_t140, - eDataApplicationCapability_application_t38fax, - eDataApplicationCapability_application_genericDataCapability, - } choice; - union { - DataProtocolCapability t120; - }; -} DataApplicationCapability_application; - -typedef struct DataApplicationCapability { /* SEQUENCE */ - int options; /* No use */ - DataApplicationCapability_application application; -} DataApplicationCapability; - -typedef struct DataType { /* CHOICE */ - enum { - eDataType_nonStandard, - eDataType_nullData, - eDataType_videoData, - eDataType_audioData, - eDataType_data, - eDataType_encryptionData, - eDataType_h235Control, - eDataType_h235Media, - eDataType_multiplexedStream, - } choice; - union { - DataApplicationCapability data; - }; -} DataType; - -typedef struct UnicastAddress_iPAddress { /* SEQUENCE */ - int options; /* No use */ - unsigned network; -} UnicastAddress_iPAddress; - -typedef struct UnicastAddress { /* CHOICE */ - enum { - eUnicastAddress_iPAddress, - eUnicastAddress_iPXAddress, - eUnicastAddress_iP6Address, - eUnicastAddress_netBios, - eUnicastAddress_iPSourceRouteAddress, - eUnicastAddress_nsap, - eUnicastAddress_nonStandardAddress, - } choice; - union { - UnicastAddress_iPAddress iPAddress; - }; -} UnicastAddress; - -typedef struct H245_TransportAddress { /* CHOICE */ - enum { - eH245_TransportAddress_unicastAddress, - eH245_TransportAddress_multicastAddress, - } choice; - union { - UnicastAddress unicastAddress; - }; -} H245_TransportAddress; - -typedef struct H2250LogicalChannelParameters { /* SEQUENCE */ - enum { - eH2250LogicalChannelParameters_nonStandard = (1 << 31), - eH2250LogicalChannelParameters_associatedSessionID = - (1 << 30), - eH2250LogicalChannelParameters_mediaChannel = (1 << 29), - eH2250LogicalChannelParameters_mediaGuaranteedDelivery = - (1 << 28), - eH2250LogicalChannelParameters_mediaControlChannel = - (1 << 27), - eH2250LogicalChannelParameters_mediaControlGuaranteedDelivery - = (1 << 26), - eH2250LogicalChannelParameters_silenceSuppression = (1 << 25), - eH2250LogicalChannelParameters_destination = (1 << 24), - eH2250LogicalChannelParameters_dynamicRTPPayloadType = - (1 << 23), - eH2250LogicalChannelParameters_mediaPacketization = (1 << 22), - eH2250LogicalChannelParameters_transportCapability = - (1 << 21), - eH2250LogicalChannelParameters_redundancyEncoding = (1 << 20), - eH2250LogicalChannelParameters_source = (1 << 19), - } options; - H245_TransportAddress mediaChannel; - H245_TransportAddress mediaControlChannel; -} H2250LogicalChannelParameters; - -typedef struct OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_none, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannel_forwardLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannel_forwardLogicalChannelParameters_portNumber - = (1 << 31), - eOpenLogicalChannel_forwardLogicalChannelParameters_forwardLogicalChannelDependency - = (1 << 30), - eOpenLogicalChannel_forwardLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - DataType dataType; - OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannel_forwardLogicalChannelParameters; - -typedef struct OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h223LogicalChannelParameters, - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_v76LogicalChannelParameters, - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannel_reverseLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters - = (1 << 31), - eOpenLogicalChannel_reverseLogicalChannelParameters_reverseLogicalChannelDependency - = (1 << 30), - eOpenLogicalChannel_reverseLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannel_reverseLogicalChannelParameters; - -typedef struct NetworkAccessParameters_networkAddress { /* CHOICE */ - enum { - eNetworkAccessParameters_networkAddress_q2931Address, - eNetworkAccessParameters_networkAddress_e164Address, - eNetworkAccessParameters_networkAddress_localAreaAddress, - } choice; - union { - H245_TransportAddress localAreaAddress; - }; -} NetworkAccessParameters_networkAddress; - -typedef struct NetworkAccessParameters { /* SEQUENCE */ - enum { - eNetworkAccessParameters_distribution = (1 << 31), - eNetworkAccessParameters_externalReference = (1 << 30), - eNetworkAccessParameters_t120SetupProcedure = (1 << 29), - } options; - NetworkAccessParameters_networkAddress networkAddress; -} NetworkAccessParameters; - -typedef struct OpenLogicalChannel { /* SEQUENCE */ - enum { - eOpenLogicalChannel_reverseLogicalChannelParameters = - (1 << 31), - eOpenLogicalChannel_separateStack = (1 << 30), - eOpenLogicalChannel_encryptionSync = (1 << 29), - } options; - OpenLogicalChannel_forwardLogicalChannelParameters - forwardLogicalChannelParameters; - OpenLogicalChannel_reverseLogicalChannelParameters - reverseLogicalChannelParameters; - NetworkAccessParameters separateStack; -} OpenLogicalChannel; - -typedef struct Setup_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Setup_UUIE_fastStart; - -typedef struct Setup_UUIE { /* SEQUENCE */ - enum { - eSetup_UUIE_h245Address = (1 << 31), - eSetup_UUIE_sourceAddress = (1 << 30), - eSetup_UUIE_destinationAddress = (1 << 29), - eSetup_UUIE_destCallSignalAddress = (1 << 28), - eSetup_UUIE_destExtraCallInfo = (1 << 27), - eSetup_UUIE_destExtraCRV = (1 << 26), - eSetup_UUIE_callServices = (1 << 25), - eSetup_UUIE_sourceCallSignalAddress = (1 << 24), - eSetup_UUIE_remoteExtensionAddress = (1 << 23), - eSetup_UUIE_callIdentifier = (1 << 22), - eSetup_UUIE_h245SecurityCapability = (1 << 21), - eSetup_UUIE_tokens = (1 << 20), - eSetup_UUIE_cryptoTokens = (1 << 19), - eSetup_UUIE_fastStart = (1 << 18), - eSetup_UUIE_mediaWaitForConnect = (1 << 17), - eSetup_UUIE_canOverlapSend = (1 << 16), - eSetup_UUIE_endpointIdentifier = (1 << 15), - eSetup_UUIE_multipleCalls = (1 << 14), - eSetup_UUIE_maintainConnection = (1 << 13), - eSetup_UUIE_connectionParameters = (1 << 12), - eSetup_UUIE_language = (1 << 11), - eSetup_UUIE_presentationIndicator = (1 << 10), - eSetup_UUIE_screeningIndicator = (1 << 9), - eSetup_UUIE_serviceControl = (1 << 8), - eSetup_UUIE_symmetricOperationRequired = (1 << 7), - eSetup_UUIE_capacity = (1 << 6), - eSetup_UUIE_circuitInfo = (1 << 5), - eSetup_UUIE_desiredProtocols = (1 << 4), - eSetup_UUIE_neededFeatures = (1 << 3), - eSetup_UUIE_desiredFeatures = (1 << 2), - eSetup_UUIE_supportedFeatures = (1 << 1), - eSetup_UUIE_parallelH245Control = (1 << 0), - } options; - TransportAddress h245Address; - TransportAddress destCallSignalAddress; - TransportAddress sourceCallSignalAddress; - Setup_UUIE_fastStart fastStart; -} Setup_UUIE; - -typedef struct CallProceeding_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} CallProceeding_UUIE_fastStart; - -typedef struct CallProceeding_UUIE { /* SEQUENCE */ - enum { - eCallProceeding_UUIE_h245Address = (1 << 31), - eCallProceeding_UUIE_callIdentifier = (1 << 30), - eCallProceeding_UUIE_h245SecurityMode = (1 << 29), - eCallProceeding_UUIE_tokens = (1 << 28), - eCallProceeding_UUIE_cryptoTokens = (1 << 27), - eCallProceeding_UUIE_fastStart = (1 << 26), - eCallProceeding_UUIE_multipleCalls = (1 << 25), - eCallProceeding_UUIE_maintainConnection = (1 << 24), - eCallProceeding_UUIE_fastConnectRefused = (1 << 23), - eCallProceeding_UUIE_featureSet = (1 << 22), - } options; - TransportAddress h245Address; - CallProceeding_UUIE_fastStart fastStart; -} CallProceeding_UUIE; - -typedef struct Connect_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Connect_UUIE_fastStart; - -typedef struct Connect_UUIE { /* SEQUENCE */ - enum { - eConnect_UUIE_h245Address = (1 << 31), - eConnect_UUIE_callIdentifier = (1 << 30), - eConnect_UUIE_h245SecurityMode = (1 << 29), - eConnect_UUIE_tokens = (1 << 28), - eConnect_UUIE_cryptoTokens = (1 << 27), - eConnect_UUIE_fastStart = (1 << 26), - eConnect_UUIE_multipleCalls = (1 << 25), - eConnect_UUIE_maintainConnection = (1 << 24), - eConnect_UUIE_language = (1 << 23), - eConnect_UUIE_connectedAddress = (1 << 22), - eConnect_UUIE_presentationIndicator = (1 << 21), - eConnect_UUIE_screeningIndicator = (1 << 20), - eConnect_UUIE_fastConnectRefused = (1 << 19), - eConnect_UUIE_serviceControl = (1 << 18), - eConnect_UUIE_capacity = (1 << 17), - eConnect_UUIE_featureSet = (1 << 16), - } options; - TransportAddress h245Address; - Connect_UUIE_fastStart fastStart; -} Connect_UUIE; - -typedef struct Alerting_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Alerting_UUIE_fastStart; - -typedef struct Alerting_UUIE { /* SEQUENCE */ - enum { - eAlerting_UUIE_h245Address = (1 << 31), - eAlerting_UUIE_callIdentifier = (1 << 30), - eAlerting_UUIE_h245SecurityMode = (1 << 29), - eAlerting_UUIE_tokens = (1 << 28), - eAlerting_UUIE_cryptoTokens = (1 << 27), - eAlerting_UUIE_fastStart = (1 << 26), - eAlerting_UUIE_multipleCalls = (1 << 25), - eAlerting_UUIE_maintainConnection = (1 << 24), - eAlerting_UUIE_alertingAddress = (1 << 23), - eAlerting_UUIE_presentationIndicator = (1 << 22), - eAlerting_UUIE_screeningIndicator = (1 << 21), - eAlerting_UUIE_fastConnectRefused = (1 << 20), - eAlerting_UUIE_serviceControl = (1 << 19), - eAlerting_UUIE_capacity = (1 << 18), - eAlerting_UUIE_featureSet = (1 << 17), - } options; - TransportAddress h245Address; - Alerting_UUIE_fastStart fastStart; -} Alerting_UUIE; - -typedef struct Information_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Information_UUIE_fastStart; - -typedef struct Information_UUIE { /* SEQUENCE */ - enum { - eInformation_UUIE_callIdentifier = (1 << 31), - eInformation_UUIE_tokens = (1 << 30), - eInformation_UUIE_cryptoTokens = (1 << 29), - eInformation_UUIE_fastStart = (1 << 28), - eInformation_UUIE_fastConnectRefused = (1 << 27), - eInformation_UUIE_circuitInfo = (1 << 26), - } options; - Information_UUIE_fastStart fastStart; -} Information_UUIE; - -typedef struct FacilityReason { /* CHOICE */ - enum { - eFacilityReason_routeCallToGatekeeper, - eFacilityReason_callForwarded, - eFacilityReason_routeCallToMC, - eFacilityReason_undefinedReason, - eFacilityReason_conferenceListChoice, - eFacilityReason_startH245, - eFacilityReason_noH245, - eFacilityReason_newTokens, - eFacilityReason_featureSetUpdate, - eFacilityReason_forwardedElements, - eFacilityReason_transportedInformation, - } choice; -} FacilityReason; - -typedef struct Facility_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Facility_UUIE_fastStart; - -typedef struct Facility_UUIE { /* SEQUENCE */ - enum { - eFacility_UUIE_alternativeAddress = (1 << 31), - eFacility_UUIE_alternativeAliasAddress = (1 << 30), - eFacility_UUIE_conferenceID = (1 << 29), - eFacility_UUIE_callIdentifier = (1 << 28), - eFacility_UUIE_destExtraCallInfo = (1 << 27), - eFacility_UUIE_remoteExtensionAddress = (1 << 26), - eFacility_UUIE_tokens = (1 << 25), - eFacility_UUIE_cryptoTokens = (1 << 24), - eFacility_UUIE_conferences = (1 << 23), - eFacility_UUIE_h245Address = (1 << 22), - eFacility_UUIE_fastStart = (1 << 21), - eFacility_UUIE_multipleCalls = (1 << 20), - eFacility_UUIE_maintainConnection = (1 << 19), - eFacility_UUIE_fastConnectRefused = (1 << 18), - eFacility_UUIE_serviceControl = (1 << 17), - eFacility_UUIE_circuitInfo = (1 << 16), - eFacility_UUIE_featureSet = (1 << 15), - eFacility_UUIE_destinationInfo = (1 << 14), - eFacility_UUIE_h245SecurityMode = (1 << 13), - } options; - TransportAddress alternativeAddress; - FacilityReason reason; - TransportAddress h245Address; - Facility_UUIE_fastStart fastStart; -} Facility_UUIE; - -typedef struct Progress_UUIE_fastStart { /* SEQUENCE OF */ - int count; - OpenLogicalChannel item[30]; -} Progress_UUIE_fastStart; - -typedef struct Progress_UUIE { /* SEQUENCE */ - enum { - eProgress_UUIE_h245Address = (1 << 31), - eProgress_UUIE_h245SecurityMode = (1 << 30), - eProgress_UUIE_tokens = (1 << 29), - eProgress_UUIE_cryptoTokens = (1 << 28), - eProgress_UUIE_fastStart = (1 << 27), - eProgress_UUIE_multipleCalls = (1 << 26), - eProgress_UUIE_maintainConnection = (1 << 25), - eProgress_UUIE_fastConnectRefused = (1 << 24), - } options; - TransportAddress h245Address; - Progress_UUIE_fastStart fastStart; -} Progress_UUIE; - -typedef struct H323_UU_PDU_h323_message_body { /* CHOICE */ - enum { - eH323_UU_PDU_h323_message_body_setup, - eH323_UU_PDU_h323_message_body_callProceeding, - eH323_UU_PDU_h323_message_body_connect, - eH323_UU_PDU_h323_message_body_alerting, - eH323_UU_PDU_h323_message_body_information, - eH323_UU_PDU_h323_message_body_releaseComplete, - eH323_UU_PDU_h323_message_body_facility, - eH323_UU_PDU_h323_message_body_progress, - eH323_UU_PDU_h323_message_body_empty, - eH323_UU_PDU_h323_message_body_status, - eH323_UU_PDU_h323_message_body_statusInquiry, - eH323_UU_PDU_h323_message_body_setupAcknowledge, - eH323_UU_PDU_h323_message_body_notify, - } choice; - union { - Setup_UUIE setup; - CallProceeding_UUIE callProceeding; - Connect_UUIE connect; - Alerting_UUIE alerting; - Information_UUIE information; - Facility_UUIE facility; - Progress_UUIE progress; - }; -} H323_UU_PDU_h323_message_body; - -typedef struct RequestMessage { /* CHOICE */ - enum { - eRequestMessage_nonStandard, - eRequestMessage_masterSlaveDetermination, - eRequestMessage_terminalCapabilitySet, - eRequestMessage_openLogicalChannel, - eRequestMessage_closeLogicalChannel, - eRequestMessage_requestChannelClose, - eRequestMessage_multiplexEntrySend, - eRequestMessage_requestMultiplexEntry, - eRequestMessage_requestMode, - eRequestMessage_roundTripDelayRequest, - eRequestMessage_maintenanceLoopRequest, - eRequestMessage_communicationModeRequest, - eRequestMessage_conferenceRequest, - eRequestMessage_multilinkRequest, - eRequestMessage_logicalChannelRateRequest, - } choice; - union { - OpenLogicalChannel openLogicalChannel; - }; -} RequestMessage; - -typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters { /* CHOICE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h222LogicalChannelParameters, - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters, - } choice; - union { - H2250LogicalChannelParameters h2250LogicalChannelParameters; - }; -} OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters; - -typedef struct OpenLogicalChannelAck_reverseLogicalChannelParameters { /* SEQUENCE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters_portNumber - = (1 << 31), - eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters - = (1 << 30), - eOpenLogicalChannelAck_reverseLogicalChannelParameters_replacementFor - = (1 << 29), - } options; - OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters - multiplexParameters; -} OpenLogicalChannelAck_reverseLogicalChannelParameters; - -typedef struct H2250LogicalChannelAckParameters { /* SEQUENCE */ - enum { - eH2250LogicalChannelAckParameters_nonStandard = (1 << 31), - eH2250LogicalChannelAckParameters_sessionID = (1 << 30), - eH2250LogicalChannelAckParameters_mediaChannel = (1 << 29), - eH2250LogicalChannelAckParameters_mediaControlChannel = - (1 << 28), - eH2250LogicalChannelAckParameters_dynamicRTPPayloadType = - (1 << 27), - eH2250LogicalChannelAckParameters_flowControlToZero = - (1 << 26), - eH2250LogicalChannelAckParameters_portNumber = (1 << 25), - } options; - H245_TransportAddress mediaChannel; - H245_TransportAddress mediaControlChannel; -} H2250LogicalChannelAckParameters; - -typedef struct OpenLogicalChannelAck_forwardMultiplexAckParameters { /* CHOICE */ - enum { - eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters, - } choice; - union { - H2250LogicalChannelAckParameters - h2250LogicalChannelAckParameters; - }; -} OpenLogicalChannelAck_forwardMultiplexAckParameters; - -typedef struct OpenLogicalChannelAck { /* SEQUENCE */ - enum { - eOpenLogicalChannelAck_reverseLogicalChannelParameters = - (1 << 31), - eOpenLogicalChannelAck_separateStack = (1 << 30), - eOpenLogicalChannelAck_forwardMultiplexAckParameters = - (1 << 29), - eOpenLogicalChannelAck_encryptionSync = (1 << 28), - } options; - OpenLogicalChannelAck_reverseLogicalChannelParameters - reverseLogicalChannelParameters; - OpenLogicalChannelAck_forwardMultiplexAckParameters - forwardMultiplexAckParameters; -} OpenLogicalChannelAck; - -typedef struct ResponseMessage { /* CHOICE */ - enum { - eResponseMessage_nonStandard, - eResponseMessage_masterSlaveDeterminationAck, - eResponseMessage_masterSlaveDeterminationReject, - eResponseMessage_terminalCapabilitySetAck, - eResponseMessage_terminalCapabilitySetReject, - eResponseMessage_openLogicalChannelAck, - eResponseMessage_openLogicalChannelReject, - eResponseMessage_closeLogicalChannelAck, - eResponseMessage_requestChannelCloseAck, - eResponseMessage_requestChannelCloseReject, - eResponseMessage_multiplexEntrySendAck, - eResponseMessage_multiplexEntrySendReject, - eResponseMessage_requestMultiplexEntryAck, - eResponseMessage_requestMultiplexEntryReject, - eResponseMessage_requestModeAck, - eResponseMessage_requestModeReject, - eResponseMessage_roundTripDelayResponse, - eResponseMessage_maintenanceLoopAck, - eResponseMessage_maintenanceLoopReject, - eResponseMessage_communicationModeResponse, - eResponseMessage_conferenceResponse, - eResponseMessage_multilinkResponse, - eResponseMessage_logicalChannelRateAcknowledge, - eResponseMessage_logicalChannelRateReject, - } choice; - union { - OpenLogicalChannelAck openLogicalChannelAck; - }; -} ResponseMessage; - -typedef struct MultimediaSystemControlMessage { /* CHOICE */ - enum { - eMultimediaSystemControlMessage_request, - eMultimediaSystemControlMessage_response, - eMultimediaSystemControlMessage_command, - eMultimediaSystemControlMessage_indication, - } choice; - union { - RequestMessage request; - ResponseMessage response; - }; -} MultimediaSystemControlMessage; - -typedef struct H323_UU_PDU_h245Control { /* SEQUENCE OF */ - int count; - MultimediaSystemControlMessage item[4]; -} H323_UU_PDU_h245Control; - -typedef struct H323_UU_PDU { /* SEQUENCE */ - enum { - eH323_UU_PDU_nonStandardData = (1 << 31), - eH323_UU_PDU_h4501SupplementaryService = (1 << 30), - eH323_UU_PDU_h245Tunneling = (1 << 29), - eH323_UU_PDU_h245Control = (1 << 28), - eH323_UU_PDU_nonStandardControl = (1 << 27), - eH323_UU_PDU_callLinkage = (1 << 26), - eH323_UU_PDU_tunnelledSignallingMessage = (1 << 25), - eH323_UU_PDU_provisionalRespToH245Tunneling = (1 << 24), - eH323_UU_PDU_stimulusControl = (1 << 23), - eH323_UU_PDU_genericData = (1 << 22), - } options; - H323_UU_PDU_h323_message_body h323_message_body; - H323_UU_PDU_h245Control h245Control; -} H323_UU_PDU; - -typedef struct H323_UserInformation { /* SEQUENCE */ - enum { - eH323_UserInformation_user_data = (1 << 31), - } options; - H323_UU_PDU h323_uu_pdu; -} H323_UserInformation; - -typedef struct GatekeeperRequest { /* SEQUENCE */ - enum { - eGatekeeperRequest_nonStandardData = (1 << 31), - eGatekeeperRequest_gatekeeperIdentifier = (1 << 30), - eGatekeeperRequest_callServices = (1 << 29), - eGatekeeperRequest_endpointAlias = (1 << 28), - eGatekeeperRequest_alternateEndpoints = (1 << 27), - eGatekeeperRequest_tokens = (1 << 26), - eGatekeeperRequest_cryptoTokens = (1 << 25), - eGatekeeperRequest_authenticationCapability = (1 << 24), - eGatekeeperRequest_algorithmOIDs = (1 << 23), - eGatekeeperRequest_integrity = (1 << 22), - eGatekeeperRequest_integrityCheckValue = (1 << 21), - eGatekeeperRequest_supportsAltGK = (1 << 20), - eGatekeeperRequest_featureSet = (1 << 19), - eGatekeeperRequest_genericData = (1 << 18), - } options; - TransportAddress rasAddress; -} GatekeeperRequest; - -typedef struct GatekeeperConfirm { /* SEQUENCE */ - enum { - eGatekeeperConfirm_nonStandardData = (1 << 31), - eGatekeeperConfirm_gatekeeperIdentifier = (1 << 30), - eGatekeeperConfirm_alternateGatekeeper = (1 << 29), - eGatekeeperConfirm_authenticationMode = (1 << 28), - eGatekeeperConfirm_tokens = (1 << 27), - eGatekeeperConfirm_cryptoTokens = (1 << 26), - eGatekeeperConfirm_algorithmOID = (1 << 25), - eGatekeeperConfirm_integrity = (1 << 24), - eGatekeeperConfirm_integrityCheckValue = (1 << 23), - eGatekeeperConfirm_featureSet = (1 << 22), - eGatekeeperConfirm_genericData = (1 << 21), - } options; - TransportAddress rasAddress; -} GatekeeperConfirm; - -typedef struct RegistrationRequest_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationRequest_callSignalAddress; - -typedef struct RegistrationRequest_rasAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationRequest_rasAddress; - -typedef struct RegistrationRequest { /* SEQUENCE */ - enum { - eRegistrationRequest_nonStandardData = (1 << 31), - eRegistrationRequest_terminalAlias = (1 << 30), - eRegistrationRequest_gatekeeperIdentifier = (1 << 29), - eRegistrationRequest_alternateEndpoints = (1 << 28), - eRegistrationRequest_timeToLive = (1 << 27), - eRegistrationRequest_tokens = (1 << 26), - eRegistrationRequest_cryptoTokens = (1 << 25), - eRegistrationRequest_integrityCheckValue = (1 << 24), - eRegistrationRequest_keepAlive = (1 << 23), - eRegistrationRequest_endpointIdentifier = (1 << 22), - eRegistrationRequest_willSupplyUUIEs = (1 << 21), - eRegistrationRequest_maintainConnection = (1 << 20), - eRegistrationRequest_alternateTransportAddresses = (1 << 19), - eRegistrationRequest_additiveRegistration = (1 << 18), - eRegistrationRequest_terminalAliasPattern = (1 << 17), - eRegistrationRequest_supportsAltGK = (1 << 16), - eRegistrationRequest_usageReportingCapability = (1 << 15), - eRegistrationRequest_multipleCalls = (1 << 14), - eRegistrationRequest_supportedH248Packages = (1 << 13), - eRegistrationRequest_callCreditCapability = (1 << 12), - eRegistrationRequest_capacityReportingCapability = (1 << 11), - eRegistrationRequest_capacity = (1 << 10), - eRegistrationRequest_featureSet = (1 << 9), - eRegistrationRequest_genericData = (1 << 8), - } options; - RegistrationRequest_callSignalAddress callSignalAddress; - RegistrationRequest_rasAddress rasAddress; - unsigned timeToLive; -} RegistrationRequest; - -typedef struct RegistrationConfirm_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} RegistrationConfirm_callSignalAddress; - -typedef struct RegistrationConfirm { /* SEQUENCE */ - enum { - eRegistrationConfirm_nonStandardData = (1 << 31), - eRegistrationConfirm_terminalAlias = (1 << 30), - eRegistrationConfirm_gatekeeperIdentifier = (1 << 29), - eRegistrationConfirm_alternateGatekeeper = (1 << 28), - eRegistrationConfirm_timeToLive = (1 << 27), - eRegistrationConfirm_tokens = (1 << 26), - eRegistrationConfirm_cryptoTokens = (1 << 25), - eRegistrationConfirm_integrityCheckValue = (1 << 24), - eRegistrationConfirm_willRespondToIRR = (1 << 23), - eRegistrationConfirm_preGrantedARQ = (1 << 22), - eRegistrationConfirm_maintainConnection = (1 << 21), - eRegistrationConfirm_serviceControl = (1 << 20), - eRegistrationConfirm_supportsAdditiveRegistration = (1 << 19), - eRegistrationConfirm_terminalAliasPattern = (1 << 18), - eRegistrationConfirm_supportedPrefixes = (1 << 17), - eRegistrationConfirm_usageSpec = (1 << 16), - eRegistrationConfirm_featureServerAlias = (1 << 15), - eRegistrationConfirm_capacityReportingSpec = (1 << 14), - eRegistrationConfirm_featureSet = (1 << 13), - eRegistrationConfirm_genericData = (1 << 12), - } options; - RegistrationConfirm_callSignalAddress callSignalAddress; - unsigned timeToLive; -} RegistrationConfirm; - -typedef struct UnregistrationRequest_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} UnregistrationRequest_callSignalAddress; - -typedef struct UnregistrationRequest { /* SEQUENCE */ - enum { - eUnregistrationRequest_endpointAlias = (1 << 31), - eUnregistrationRequest_nonStandardData = (1 << 30), - eUnregistrationRequest_endpointIdentifier = (1 << 29), - eUnregistrationRequest_alternateEndpoints = (1 << 28), - eUnregistrationRequest_gatekeeperIdentifier = (1 << 27), - eUnregistrationRequest_tokens = (1 << 26), - eUnregistrationRequest_cryptoTokens = (1 << 25), - eUnregistrationRequest_integrityCheckValue = (1 << 24), - eUnregistrationRequest_reason = (1 << 23), - eUnregistrationRequest_endpointAliasPattern = (1 << 22), - eUnregistrationRequest_supportedPrefixes = (1 << 21), - eUnregistrationRequest_alternateGatekeeper = (1 << 20), - eUnregistrationRequest_genericData = (1 << 19), - } options; - UnregistrationRequest_callSignalAddress callSignalAddress; -} UnregistrationRequest; - -typedef struct AdmissionRequest { /* SEQUENCE */ - enum { - eAdmissionRequest_callModel = (1 << 31), - eAdmissionRequest_destinationInfo = (1 << 30), - eAdmissionRequest_destCallSignalAddress = (1 << 29), - eAdmissionRequest_destExtraCallInfo = (1 << 28), - eAdmissionRequest_srcCallSignalAddress = (1 << 27), - eAdmissionRequest_nonStandardData = (1 << 26), - eAdmissionRequest_callServices = (1 << 25), - eAdmissionRequest_canMapAlias = (1 << 24), - eAdmissionRequest_callIdentifier = (1 << 23), - eAdmissionRequest_srcAlternatives = (1 << 22), - eAdmissionRequest_destAlternatives = (1 << 21), - eAdmissionRequest_gatekeeperIdentifier = (1 << 20), - eAdmissionRequest_tokens = (1 << 19), - eAdmissionRequest_cryptoTokens = (1 << 18), - eAdmissionRequest_integrityCheckValue = (1 << 17), - eAdmissionRequest_transportQOS = (1 << 16), - eAdmissionRequest_willSupplyUUIEs = (1 << 15), - eAdmissionRequest_callLinkage = (1 << 14), - eAdmissionRequest_gatewayDataRate = (1 << 13), - eAdmissionRequest_capacity = (1 << 12), - eAdmissionRequest_circuitInfo = (1 << 11), - eAdmissionRequest_desiredProtocols = (1 << 10), - eAdmissionRequest_desiredTunnelledProtocol = (1 << 9), - eAdmissionRequest_featureSet = (1 << 8), - eAdmissionRequest_genericData = (1 << 7), - } options; - TransportAddress destCallSignalAddress; - TransportAddress srcCallSignalAddress; -} AdmissionRequest; - -typedef struct AdmissionConfirm { /* SEQUENCE */ - enum { - eAdmissionConfirm_irrFrequency = (1 << 31), - eAdmissionConfirm_nonStandardData = (1 << 30), - eAdmissionConfirm_destinationInfo = (1 << 29), - eAdmissionConfirm_destExtraCallInfo = (1 << 28), - eAdmissionConfirm_destinationType = (1 << 27), - eAdmissionConfirm_remoteExtensionAddress = (1 << 26), - eAdmissionConfirm_alternateEndpoints = (1 << 25), - eAdmissionConfirm_tokens = (1 << 24), - eAdmissionConfirm_cryptoTokens = (1 << 23), - eAdmissionConfirm_integrityCheckValue = (1 << 22), - eAdmissionConfirm_transportQOS = (1 << 21), - eAdmissionConfirm_willRespondToIRR = (1 << 20), - eAdmissionConfirm_uuiesRequested = (1 << 19), - eAdmissionConfirm_language = (1 << 18), - eAdmissionConfirm_alternateTransportAddresses = (1 << 17), - eAdmissionConfirm_useSpecifiedTransport = (1 << 16), - eAdmissionConfirm_circuitInfo = (1 << 15), - eAdmissionConfirm_usageSpec = (1 << 14), - eAdmissionConfirm_supportedProtocols = (1 << 13), - eAdmissionConfirm_serviceControl = (1 << 12), - eAdmissionConfirm_multipleCalls = (1 << 11), - eAdmissionConfirm_featureSet = (1 << 10), - eAdmissionConfirm_genericData = (1 << 9), - } options; - TransportAddress destCallSignalAddress; -} AdmissionConfirm; - -typedef struct LocationRequest { /* SEQUENCE */ - enum { - eLocationRequest_endpointIdentifier = (1 << 31), - eLocationRequest_nonStandardData = (1 << 30), - eLocationRequest_sourceInfo = (1 << 29), - eLocationRequest_canMapAlias = (1 << 28), - eLocationRequest_gatekeeperIdentifier = (1 << 27), - eLocationRequest_tokens = (1 << 26), - eLocationRequest_cryptoTokens = (1 << 25), - eLocationRequest_integrityCheckValue = (1 << 24), - eLocationRequest_desiredProtocols = (1 << 23), - eLocationRequest_desiredTunnelledProtocol = (1 << 22), - eLocationRequest_featureSet = (1 << 21), - eLocationRequest_genericData = (1 << 20), - eLocationRequest_hopCount = (1 << 19), - eLocationRequest_circuitInfo = (1 << 18), - } options; - TransportAddress replyAddress; -} LocationRequest; - -typedef struct LocationConfirm { /* SEQUENCE */ - enum { - eLocationConfirm_nonStandardData = (1 << 31), - eLocationConfirm_destinationInfo = (1 << 30), - eLocationConfirm_destExtraCallInfo = (1 << 29), - eLocationConfirm_destinationType = (1 << 28), - eLocationConfirm_remoteExtensionAddress = (1 << 27), - eLocationConfirm_alternateEndpoints = (1 << 26), - eLocationConfirm_tokens = (1 << 25), - eLocationConfirm_cryptoTokens = (1 << 24), - eLocationConfirm_integrityCheckValue = (1 << 23), - eLocationConfirm_alternateTransportAddresses = (1 << 22), - eLocationConfirm_supportedProtocols = (1 << 21), - eLocationConfirm_multipleCalls = (1 << 20), - eLocationConfirm_featureSet = (1 << 19), - eLocationConfirm_genericData = (1 << 18), - eLocationConfirm_circuitInfo = (1 << 17), - eLocationConfirm_serviceControl = (1 << 16), - } options; - TransportAddress callSignalAddress; - TransportAddress rasAddress; -} LocationConfirm; - -typedef struct InfoRequestResponse_callSignalAddress { /* SEQUENCE OF */ - int count; - TransportAddress item[10]; -} InfoRequestResponse_callSignalAddress; - -typedef struct InfoRequestResponse { /* SEQUENCE */ - enum { - eInfoRequestResponse_nonStandardData = (1 << 31), - eInfoRequestResponse_endpointAlias = (1 << 30), - eInfoRequestResponse_perCallInfo = (1 << 29), - eInfoRequestResponse_tokens = (1 << 28), - eInfoRequestResponse_cryptoTokens = (1 << 27), - eInfoRequestResponse_integrityCheckValue = (1 << 26), - eInfoRequestResponse_needResponse = (1 << 25), - eInfoRequestResponse_capacity = (1 << 24), - eInfoRequestResponse_irrStatus = (1 << 23), - eInfoRequestResponse_unsolicited = (1 << 22), - eInfoRequestResponse_genericData = (1 << 21), - } options; - TransportAddress rasAddress; - InfoRequestResponse_callSignalAddress callSignalAddress; -} InfoRequestResponse; - -typedef struct RasMessage { /* CHOICE */ - enum { - eRasMessage_gatekeeperRequest, - eRasMessage_gatekeeperConfirm, - eRasMessage_gatekeeperReject, - eRasMessage_registrationRequest, - eRasMessage_registrationConfirm, - eRasMessage_registrationReject, - eRasMessage_unregistrationRequest, - eRasMessage_unregistrationConfirm, - eRasMessage_unregistrationReject, - eRasMessage_admissionRequest, - eRasMessage_admissionConfirm, - eRasMessage_admissionReject, - eRasMessage_bandwidthRequest, - eRasMessage_bandwidthConfirm, - eRasMessage_bandwidthReject, - eRasMessage_disengageRequest, - eRasMessage_disengageConfirm, - eRasMessage_disengageReject, - eRasMessage_locationRequest, - eRasMessage_locationConfirm, - eRasMessage_locationReject, - eRasMessage_infoRequest, - eRasMessage_infoRequestResponse, - eRasMessage_nonStandardMessage, - eRasMessage_unknownMessageResponse, - eRasMessage_requestInProgress, - eRasMessage_resourcesAvailableIndicate, - eRasMessage_resourcesAvailableConfirm, - eRasMessage_infoRequestAck, - eRasMessage_infoRequestNak, - eRasMessage_serviceControlIndication, - eRasMessage_serviceControlResponse, - } choice; - union { - GatekeeperRequest gatekeeperRequest; - GatekeeperConfirm gatekeeperConfirm; - RegistrationRequest registrationRequest; - RegistrationConfirm registrationConfirm; - UnregistrationRequest unregistrationRequest; - AdmissionRequest admissionRequest; - AdmissionConfirm admissionConfirm; - LocationRequest locationRequest; - LocationConfirm locationConfirm; - InfoRequestResponse infoRequestResponse; - }; -} RasMessage; diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 83694cfdfa8f..1646076933b1 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -41,11 +41,13 @@ union nf_conntrack_expect_proto { /* Add protocol helper include file here */ #include +#include /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ struct nf_ct_ftp_master ct_ftp_info; + struct nf_ct_h323_master ct_h323_info; }; #include diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 54a3d038beaa..cef3136e22a3 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -41,6 +41,7 @@ struct nf_conntrack_expect unsigned int flags; #ifdef CONFIG_NF_NAT_NEEDED + __be32 saved_ip; /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ union nf_conntrack_man_proto saved_proto; diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 6993ec53dc06..e14156d1122e 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -529,6 +529,11 @@ config IP_NF_NAT_H323 default IP_NF_NAT if IP_NF_H323=y default m if IP_NF_H323=m +config NF_NAT_H323 + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_H323 + config IP_NF_NAT_SIP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 8893249bbe98..bdaba4700e3b 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -15,7 +15,7 @@ endif ip_conntrack_pptp-objs := ip_conntrack_helper_pptp.o ip_conntrack_proto_gre.o ip_nat_pptp-objs := ip_nat_helper_pptp.o ip_nat_proto_gre.o -ip_conntrack_h323-objs := ip_conntrack_helper_h323.o ip_conntrack_helper_h323_asn1.o +ip_conntrack_h323-objs := ip_conntrack_helper_h323.o ../../netfilter/nf_conntrack_h323_asn1.o ip_nat_h323-objs := ip_nat_helper_h323.o # connection tracking @@ -52,6 +52,7 @@ obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o # NAT helpers (nf_conntrack) obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o +obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c deleted file mode 100644 index 26dfecadb335..000000000000 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_asn1.c +++ /dev/null @@ -1,874 +0,0 @@ -/**************************************************************************** - * ip_conntrack_helper_h323_asn1.c - BER and PER decoding library for H.323 - * conntrack/NAT module. - * - * Copyright (c) 2006 by Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - * - * See ip_conntrack_helper_h323_asn1.h for details. - * - ****************************************************************************/ - -#ifdef __KERNEL__ -#include -#else -#include -#endif -#include - -/* Trace Flag */ -#ifndef H323_TRACE -#define H323_TRACE 0 -#endif - -#if H323_TRACE -#define TAB_SIZE 4 -#define IFTHEN(cond, act) if(cond){act;} -#ifdef __KERNEL__ -#define PRINT printk -#else -#define PRINT printf -#endif -#define FNAME(name) name, -#else -#define IFTHEN(cond, act) -#define PRINT(fmt, args...) -#define FNAME(name) -#endif - -/* ASN.1 Types */ -#define NUL 0 -#define BOOL 1 -#define OID 2 -#define INT 3 -#define ENUM 4 -#define BITSTR 5 -#define NUMSTR 6 -#define NUMDGT 6 -#define TBCDSTR 6 -#define OCTSTR 7 -#define PRTSTR 7 -#define IA5STR 7 -#define GENSTR 7 -#define BMPSTR 8 -#define SEQ 9 -#define SET 9 -#define SEQOF 10 -#define SETOF 10 -#define CHOICE 11 - -/* Constraint Types */ -#define FIXD 0 -/* #define BITS 1-8 */ -#define BYTE 9 -#define WORD 10 -#define CONS 11 -#define SEMI 12 -#define UNCO 13 - -/* ASN.1 Type Attributes */ -#define SKIP 0 -#define STOP 1 -#define DECODE 2 -#define EXT 4 -#define OPEN 8 -#define OPT 16 - - -/* ASN.1 Field Structure */ -typedef struct field_t { -#if H323_TRACE - char *name; -#endif - unsigned char type; - unsigned char sz; - unsigned char lb; - unsigned char ub; - unsigned short attr; - unsigned short offset; - struct field_t *fields; -} field_t; - -/* Bit Stream */ -typedef struct { - unsigned char *buf; - unsigned char *beg; - unsigned char *end; - unsigned char *cur; - unsigned bit; -} bitstr_t; - -/* Tool Functions */ -#define INC_BIT(bs) if((++bs->bit)>7){bs->cur++;bs->bit=0;} -#define INC_BITS(bs,b) if((bs->bit+=b)>7){bs->cur+=bs->bit>>3;bs->bit&=7;} -#define BYTE_ALIGN(bs) if(bs->bit){bs->cur++;bs->bit=0;} -#define CHECK_BOUND(bs,n) if(bs->cur+(n)>bs->end)return(H323_ERROR_BOUND) -static unsigned get_len(bitstr_t * bs); -static unsigned get_bit(bitstr_t * bs); -static unsigned get_bits(bitstr_t * bs, unsigned b); -static unsigned get_bitmap(bitstr_t * bs, unsigned b); -static unsigned get_uint(bitstr_t * bs, int b); - -/* Decoder Functions */ -static int decode_nul(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bool(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_oid(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_int(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_enum(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_seq(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level); -static int decode_choice(bitstr_t * bs, field_t * f, char *base, int level); - -/* Decoder Functions Vector */ -typedef int (*decoder_t) (bitstr_t *, field_t *, char *, int); -static decoder_t Decoders[] = { - decode_nul, - decode_bool, - decode_oid, - decode_int, - decode_enum, - decode_bitstr, - decode_numstr, - decode_octstr, - decode_bmpstr, - decode_seq, - decode_seqof, - decode_choice, -}; - -/**************************************************************************** - * H.323 Types - ****************************************************************************/ -#include "ip_conntrack_helper_h323_types.c" - -/**************************************************************************** - * Functions - ****************************************************************************/ -/* Assume bs is aligned && v < 16384 */ -unsigned get_len(bitstr_t * bs) -{ - unsigned v; - - v = *bs->cur++; - - if (v & 0x80) { - v &= 0x3f; - v <<= 8; - v += *bs->cur++; - } - - return v; -} - -/****************************************************************************/ -unsigned get_bit(bitstr_t * bs) -{ - unsigned b = (*bs->cur) & (0x80 >> bs->bit); - - INC_BIT(bs); - - return b; -} - -/****************************************************************************/ -/* Assume b <= 8 */ -unsigned get_bits(bitstr_t * bs, unsigned b) -{ - unsigned v, l; - - v = (*bs->cur) & (0xffU >> bs->bit); - l = b + bs->bit; - - if (l < 8) { - v >>= 8 - l; - bs->bit = l; - } else if (l == 8) { - bs->cur++; - bs->bit = 0; - } else { /* l > 8 */ - - v <<= 8; - v += *(++bs->cur); - v >>= 16 - l; - bs->bit = l - 8; - } - - return v; -} - -/****************************************************************************/ -/* Assume b <= 32 */ -unsigned get_bitmap(bitstr_t * bs, unsigned b) -{ - unsigned v, l, shift, bytes; - - if (!b) - return 0; - - l = bs->bit + b; - - if (l < 8) { - v = (unsigned) (*bs->cur) << (bs->bit + 24); - bs->bit = l; - } else if (l == 8) { - v = (unsigned) (*bs->cur++) << (bs->bit + 24); - bs->bit = 0; - } else { - for (bytes = l >> 3, shift = 24, v = 0; bytes; - bytes--, shift -= 8) - v |= (unsigned) (*bs->cur++) << shift; - - if (l < 32) { - v |= (unsigned) (*bs->cur) << shift; - v <<= bs->bit; - } else if (l > 32) { - v <<= bs->bit; - v |= (*bs->cur) >> (8 - bs->bit); - } - - bs->bit = l & 0x7; - } - - v &= 0xffffffff << (32 - b); - - return v; -} - -/**************************************************************************** - * Assume bs is aligned and sizeof(unsigned int) == 4 - ****************************************************************************/ -unsigned get_uint(bitstr_t * bs, int b) -{ - unsigned v = 0; - - switch (b) { - case 4: - v |= *bs->cur++; - v <<= 8; - case 3: - v |= *bs->cur++; - v <<= 8; - case 2: - v |= *bs->cur++; - v <<= 8; - case 1: - v |= *bs->cur++; - break; - } - return v; -} - -/****************************************************************************/ -int decode_nul(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bool(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - INC_BIT(bs); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_oid(bitstr_t * bs, field_t * f, char *base, int level) -{ - int len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = *bs->cur++; - bs->cur += len; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_int(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - bs->cur++; - break; - case WORD: /* 257 <= Range <= 64K */ - BYTE_ALIGN(bs); - bs->cur += 2; - break; - case CONS: /* 64K < Range < 4G */ - len = get_bits(bs, 2) + 1; - BYTE_ALIGN(bs); - if (base && (f->attr & DECODE)) { /* timeToLive */ - unsigned v = get_uint(bs, len) + f->lb; - PRINT(" = %u", v); - *((unsigned *) (base + f->offset)) = v; - } - bs->cur += len; - break; - case UNCO: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - len = get_len(bs); - bs->cur += len; - break; - default: /* 2 <= Range <= 255 */ - INC_BITS(bs, f->sz); - break; - } - - PRINT("\n"); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_enum(bitstr_t * bs, field_t * f, char *base, int level) -{ - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - if ((f->attr & EXT) && get_bit(bs)) { - INC_BITS(bs, 7); - } else { - INC_BITS(bs, f->sz); - } - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - BYTE_ALIGN(bs); - switch (f->sz) { - case FIXD: /* fixed length > 16 */ - len = f->lb; - break; - case WORD: /* 2-byte length */ - CHECK_BOUND(bs, 2); - len = (*bs->cur++) << 8; - len += (*bs->cur++) + f->lb; - break; - case SEMI: - CHECK_BOUND(bs, 2); - len = get_len(bs); - break; - default: - len = 0; - break; - } - - bs->cur += len >> 3; - bs->bit = len & 7; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - - BYTE_ALIGN(bs); - INC_BITS(bs, (len << 2)); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case FIXD: /* Range == 1 */ - if (f->lb > 2) { - BYTE_ALIGN(bs); - if (base && (f->attr & DECODE)) { - /* The IP Address */ - IFTHEN(f->lb == 4, - PRINT(" = %d.%d.%d.%d:%d", - bs->cur[0], bs->cur[1], - bs->cur[2], bs->cur[3], - bs->cur[4] * 256 + bs->cur[5])); - *((unsigned *) (base + f->offset)) = - bs->cur - bs->buf; - } - } - len = f->lb; - break; - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = (*bs->cur++) + f->lb; - break; - case SEMI: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - len = get_len(bs) + f->lb; - break; - default: /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - BYTE_ALIGN(bs); - break; - } - - bs->cur += len; - - PRINT("\n"); - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned len; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - switch (f->sz) { - case BYTE: /* Range == 256 */ - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - len = (*bs->cur++) + f->lb; - break; - default: /* 2 <= Range <= 255 */ - len = get_bits(bs, f->sz) + f->lb; - BYTE_ALIGN(bs); - break; - } - - bs->cur += len << 1; - - CHECK_BOUND(bs, 0); - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_seq(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned ext, bmp, i, opt, len = 0, bmp2, bmp2_len; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Extensible? */ - ext = (f->attr & EXT) ? get_bit(bs) : 0; - - /* Get fields bitmap */ - bmp = get_bitmap(bs, f->sz); - if (base) - *(unsigned *) base = bmp; - - /* Decode the root components */ - for (i = opt = 0, son = f->fields; i < f->lb; i++, son++) { - if (son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - return H323_ERROR_STOP; - } - - if (son->attr & OPT) { /* Optional component */ - if (!((0x80000000U >> (opt++)) & bmp)) /* Not exist */ - continue; - } - - /* Decode */ - if (son->attr & OPEN) { /* Open field */ - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, - " ", son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - /* Decode */ - if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - } - - /* No extension? */ - if (!ext) - return H323_ERROR_NONE; - - /* Get the extension bitmap */ - bmp2_len = get_bits(bs, 7) + 1; - CHECK_BOUND(bs, (bmp2_len + 7) >> 3); - bmp2 = get_bitmap(bs, bmp2_len); - bmp |= bmp2 >> f->sz; - if (base) - *(unsigned *) base = bmp; - BYTE_ALIGN(bs); - - /* Decode the extension components */ - for (opt = 0; opt < bmp2_len; opt++, i++, son++) { - if (i < f->ub && son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - return H323_ERROR_STOP; - } - - if (!((0x80000000 >> opt) & bmp2)) /* Not present */ - continue; - - /* Check Range */ - if (i >= f->ub) { /* Newer Version? */ - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - bs->cur += len; - continue; - } - - CHECK_BOUND(bs, 2); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, base, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned count, effective_count = 0, i, len = 0; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Decode item count */ - switch (f->sz) { - case BYTE: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 1); - count = *bs->cur++; - break; - case WORD: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - count = *bs->cur++; - count <<= 8; - count = *bs->cur++; - break; - case SEMI: - BYTE_ALIGN(bs); - CHECK_BOUND(bs, 2); - count = get_len(bs); - break; - default: - count = get_bits(bs, f->sz); - break; - } - count += f->lb; - - /* Write Count */ - if (base) { - effective_count = count > f->ub ? f->ub : count; - *(unsigned *) base = effective_count; - base += sizeof(unsigned); - } - - /* Decode nested field */ - son = f->fields; - if (base) - base -= son->offset; - for (i = 0; i < count; i++) { - if (son->attr & OPEN) { - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, - " ", son->name); - bs->cur += len; - continue; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, - i < - effective_count ? - base : NULL, - level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else - if ((err = (Decoders[son->type]) (bs, son, - i < - effective_count ? - base : NULL, - level + 1)) < - H323_ERROR_NONE) - return err; - - if (base) - base += son->offset; - } - - return H323_ERROR_NONE; -} - - -/****************************************************************************/ -int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) -{ - unsigned type, ext, len = 0; - int err; - field_t *son; - unsigned char *beg = NULL; - - PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); - - /* Decode? */ - base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; - - /* Decode the choice index number */ - if ((f->attr & EXT) && get_bit(bs)) { - ext = 1; - type = get_bits(bs, 7) + f->lb; - } else { - ext = 0; - type = get_bits(bs, f->sz); - } - - /* Write Type */ - if (base) - *(unsigned *) base = type; - - /* Check Range */ - if (type >= f->ub) { /* Newer version? */ - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - bs->cur += len; - return H323_ERROR_NONE; - } - - /* Transfer to son level */ - son = &f->fields[type]; - if (son->attr & STOP) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", son->name); - return H323_ERROR_STOP; - } - - if (ext || (son->attr & OPEN)) { - BYTE_ALIGN(bs); - len = get_len(bs); - CHECK_BOUND(bs, len); - if (!base || !(son->attr & DECODE)) { - PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", - son->name); - bs->cur += len; - return H323_ERROR_NONE; - } - beg = bs->cur; - - if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < - H323_ERROR_NONE) - return err; - - bs->cur = beg + len; - bs->bit = 0; - } else if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < - H323_ERROR_NONE) - return err; - - return H323_ERROR_NONE; -} - -/****************************************************************************/ -int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras) -{ - static field_t ras_message = { - FNAME("RasMessage") CHOICE, 5, 24, 32, DECODE | EXT, - 0, _RasMessage - }; - bitstr_t bs; - - bs.buf = bs.beg = bs.cur = buf; - bs.end = buf + sz; - bs.bit = 0; - - return decode_choice(&bs, &ras_message, (char *) ras, 0); -} - -/****************************************************************************/ -static int DecodeH323_UserInformation(unsigned char *buf, unsigned char *beg, - size_t sz, H323_UserInformation * uuie) -{ - static field_t h323_userinformation = { - FNAME("H323-UserInformation") SEQ, 1, 2, 2, DECODE | EXT, - 0, _H323_UserInformation - }; - bitstr_t bs; - - bs.buf = buf; - bs.beg = bs.cur = beg; - bs.end = beg + sz; - bs.bit = 0; - - return decode_seq(&bs, &h323_userinformation, (char *) uuie, 0); -} - -/****************************************************************************/ -int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, - MultimediaSystemControlMessage * - mscm) -{ - static field_t multimediasystemcontrolmessage = { - FNAME("MultimediaSystemControlMessage") CHOICE, 2, 4, 4, - DECODE | EXT, 0, _MultimediaSystemControlMessage - }; - bitstr_t bs; - - bs.buf = bs.beg = bs.cur = buf; - bs.end = buf + sz; - bs.bit = 0; - - return decode_choice(&bs, &multimediasystemcontrolmessage, - (char *) mscm, 0); -} - -/****************************************************************************/ -int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931) -{ - unsigned char *p = buf; - int len; - - if (!p || sz < 1) - return H323_ERROR_BOUND; - - /* Protocol Discriminator */ - if (*p != 0x08) { - PRINT("Unknown Protocol Discriminator\n"); - return H323_ERROR_RANGE; - } - p++; - sz--; - - /* CallReferenceValue */ - if (sz < 1) - return H323_ERROR_BOUND; - len = *p++; - sz--; - if (sz < len) - return H323_ERROR_BOUND; - p += len; - sz -= len; - - /* Message Type */ - if (sz < 1) - return H323_ERROR_BOUND; - q931->MessageType = *p++; - PRINT("MessageType = %02X\n", q931->MessageType); - if (*p & 0x80) { - p++; - sz--; - } - - /* Decode Information Elements */ - while (sz > 0) { - if (*p == 0x7e) { /* UserUserIE */ - if (sz < 3) - break; - p++; - len = *p++ << 8; - len |= *p++; - sz -= 3; - if (sz < len) - break; - p++; - len--; - return DecodeH323_UserInformation(buf, p, len, - &q931->UUIE); - } - p++; - sz--; - if (sz < 1) - break; - len = *p++; - if (sz < len) - break; - p += len; - sz -= len; - } - - PRINT("Q.931 UUIE not found\n"); - - return H323_ERROR_BOUND; -} diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c b/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c deleted file mode 100644 index 4b359618bedd..000000000000 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323_types.c +++ /dev/null @@ -1,1926 +0,0 @@ -/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 - * - * Copyright (c) 2006 Jing Min Zhao - * - * This source code is licensed under General Public License version 2. - */ - -static field_t _TransportAddress_ipAddress[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 4, 0, DECODE, - offsetof(TransportAddress_ipAddress, ip), NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute_route[] = { /* SEQUENCE OF */ - {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute_routing[] = { /* CHOICE */ - {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ipSourceRoute[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, - _TransportAddress_ipSourceRoute_route}, - {FNAME("routing") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _TransportAddress_ipSourceRoute_routing}, -}; - -static field_t _TransportAddress_ipxAddress[] = { /* SEQUENCE */ - {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, - {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("port") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress_ip6Address[] = { /* SEQUENCE */ - {FNAME("ip") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H221NonStandard[] = { /* SEQUENCE */ - {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _NonStandardIdentifier[] = { /* CHOICE */ - {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP | EXT, 0, - _H221NonStandard}, -}; - -static field_t _NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _NonStandardIdentifier}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TransportAddress[] = { /* CHOICE */ - {FNAME("ipAddress") SEQ, 0, 2, 2, DECODE, - offsetof(TransportAddress, ipAddress), _TransportAddress_ipAddress}, - {FNAME("ipSourceRoute") SEQ, 0, 4, 4, SKIP | EXT, 0, - _TransportAddress_ipSourceRoute}, - {FNAME("ipxAddress") SEQ, 0, 3, 3, SKIP, 0, - _TransportAddress_ipxAddress}, - {FNAME("ip6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _TransportAddress_ip6Address}, - {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, - _NonStandardParameter}, -}; - -static field_t _AliasAddress[] = { /* CHOICE */ - {FNAME("dialedDigits") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("h323-ID") BMPSTR, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("url-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("transportID") CHOICE, 3, 7, 7, SKIP | EXT, 0, NULL}, - {FNAME("email-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("partyNumber") CHOICE, 3, 5, 5, SKIP | EXT, 0, NULL}, - {FNAME("mobileUIM") CHOICE, 1, 2, 2, SKIP | EXT, 0, NULL}, -}; - -static field_t _Setup_UUIE_sourceAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _VendorIdentifier[] = { /* SEQUENCE */ - {FNAME("vendor") SEQ, 0, 3, 3, SKIP | EXT, 0, _H221NonStandard}, - {FNAME("productId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("versionId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _GatekeeperInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _H310Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H320Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H321Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H322Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H323Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H324Caps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _VoiceCaps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T120OnlyCaps[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _SupportedProtocols[] = { /* CHOICE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP, 0, - _NonStandardParameter}, - {FNAME("h310") SEQ, 1, 1, 3, SKIP | EXT, 0, _H310Caps}, - {FNAME("h320") SEQ, 1, 1, 3, SKIP | EXT, 0, _H320Caps}, - {FNAME("h321") SEQ, 1, 1, 3, SKIP | EXT, 0, _H321Caps}, - {FNAME("h322") SEQ, 1, 1, 3, SKIP | EXT, 0, _H322Caps}, - {FNAME("h323") SEQ, 1, 1, 3, SKIP | EXT, 0, _H323Caps}, - {FNAME("h324") SEQ, 1, 1, 3, SKIP | EXT, 0, _H324Caps}, - {FNAME("voice") SEQ, 1, 1, 3, SKIP | EXT, 0, _VoiceCaps}, - {FNAME("t120-only") SEQ, 1, 1, 3, SKIP | EXT, 0, _T120OnlyCaps}, - {FNAME("nonStandardProtocol") SEQ, 2, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("t38FaxAnnexbOnly") SEQ, 2, 5, 5, SKIP | EXT, 0, NULL}, -}; - -static field_t _GatewayInfo_protocol[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 4, 9, 11, SKIP | EXT, 0, _SupportedProtocols}, -}; - -static field_t _GatewayInfo[] = { /* SEQUENCE */ - {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _GatewayInfo_protocol}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _McuInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _TerminalInfo[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, -}; - -static field_t _EndpointType[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("vendor") SEQ, 2, 3, 3, SKIP | EXT | OPT, 0, - _VendorIdentifier}, - {FNAME("gatekeeper") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, - _GatekeeperInfo}, - {FNAME("gateway") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, _GatewayInfo}, - {FNAME("mcu") SEQ, 1, 1, 2, SKIP | EXT | OPT, 0, _McuInfo}, - {FNAME("terminal") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, _TerminalInfo}, - {FNAME("mc") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedNode") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("set") BITSTR, FIXD, 32, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedTunnelledProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, - 0, NULL}, -}; - -static field_t _Setup_UUIE_destinationAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _Setup_UUIE_destExtraCallInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _Setup_UUIE_destExtraCRV[] = { /* SEQUENCE OF */ - {FNAME("item") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Setup_UUIE_conferenceGoal[] = { /* CHOICE */ - {FNAME("create") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("join") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("invite") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("capability-negotiation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callIndependentSupplementaryService") NUL, FIXD, 0, 0, SKIP, - 0, NULL}, -}; - -static field_t _Q954Details[] = { /* SEQUENCE */ - {FNAME("conferenceCalling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threePartyService") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _QseriesOptions[] = { /* SEQUENCE */ - {FNAME("q932Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q951Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q952Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q953Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q955Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q956Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q957Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("q954Info") SEQ, 0, 2, 2, SKIP | EXT, 0, _Q954Details}, -}; - -static field_t _CallType[] = { /* CHOICE */ - {FNAME("pointToPoint") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("oneToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nToOne") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H245_NonStandardIdentifier_h221NonStandard[] = { /* SEQUENCE */ - {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H245_NonStandardIdentifier[] = { /* CHOICE */ - {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP, 0, - _H245_NonStandardIdentifier_h221NonStandard}, -}; - -static field_t _H245_NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP, 0, - _H245_NonStandardIdentifier}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H261VideoCapability[] = { /* SEQUENCE */ - {FNAME("qcifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("maxBitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("stillImageTransmission") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H262VideoCapability[] = { /* SEQUENCE */ - {FNAME("profileAndLevel-SPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-MPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SNRatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SNRatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-SpatialatH-14") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("profileAndLevel-HPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-HPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("profileAndLevel-HPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("framesPerSecond") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H263VideoCapability[] = { /* SEQUENCE */ - {FNAME("sqcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("qcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cif4MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("cif16MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("maxBitRate") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("unrestrictedVector") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("arithmeticCoding") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("advancedPrediction") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("pbFrames") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, - NULL}, - {FNAME("hrd-B") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("bppMaxKb") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowSqcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowQcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCif4MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("slowCif16MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("errorCompensation") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("enhancementLayerInfo") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("h263Options") SEQ, 5, 29, 31, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _IS11172VideoCapability[] = { /* SEQUENCE */ - {FNAME("constrainedBitstream") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("pictureRate") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _VideoCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("h261VideoCapability") SEQ, 2, 5, 6, SKIP | EXT, 0, - _H261VideoCapability}, - {FNAME("h262VideoCapability") SEQ, 6, 17, 18, SKIP | EXT, 0, - _H262VideoCapability}, - {FNAME("h263VideoCapability") SEQ, 7, 13, 21, SKIP | EXT, 0, - _H263VideoCapability}, - {FNAME("is11172VideoCapability") SEQ, 6, 7, 8, SKIP | EXT, 0, - _IS11172VideoCapability}, - {FNAME("genericVideoCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _AudioCapability_g7231[] = { /* SEQUENCE */ - {FNAME("maxAl-sduAudioFrames") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _IS11172AudioCapability[] = { /* SEQUENCE */ - {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _IS13818AudioCapability[] = { /* SEQUENCE */ - {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling16k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling22k05") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling24k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threeChannels2-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("threeChannels3-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels2-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels2-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fourChannels3-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fiveChannels3-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fiveChannels3-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("lowFrequencyEnhancement") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("multilingual") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _AudioCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("g711Alaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Alaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Ulaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g711Ulaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g722-48k") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g7231") SEQ, 0, 2, 2, SKIP, 0, _AudioCapability_g7231}, - {FNAME("g728") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729AnnexA") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("is11172AudioCapability") SEQ, 0, 9, 9, SKIP | EXT, 0, - _IS11172AudioCapability}, - {FNAME("is13818AudioCapability") SEQ, 0, 21, 21, SKIP | EXT, 0, - _IS13818AudioCapability}, - {FNAME("g729wAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g729AnnexAwAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, - {FNAME("g7231AnnexCCapability") SEQ, 1, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmHalfRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("gsmEnhancedFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("genericAudioCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, - {FNAME("g729Extensions") SEQ, 1, 8, 8, SKIP | EXT, 0, NULL}, -}; - -static field_t _DataProtocolCapability[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("v14buffered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v42lapm") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdlcFrameTunnelling") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h310SeparateVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h310SingleVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("transparent") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("segmentationAndReassembly") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdlcFrameTunnelingwSAR") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v120") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("separateLANStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("v76wCompression") CHOICE, 2, 3, 3, SKIP | EXT, 0, NULL}, - {FNAME("tcp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("udp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T84Profile_t84Restricted[] = { /* SEQUENCE */ - {FNAME("qcif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("cif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("ccir601Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("ccir601Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdtvSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("hdtvProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g3FacsMH200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g3FacsMH200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g4FacsMMR200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("g4FacsMMR200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig200x200Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig200x200Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig300x300Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("jbig300x300Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoLow") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoMedSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoMedProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoHighSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("digPhotoHighProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _T84Profile[] = { /* CHOICE */ - {FNAME("t84Unrestricted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("t84Restricted") SEQ, 0, 19, 19, SKIP | EXT, 0, - _T84Profile_t84Restricted}, -}; - -static field_t _DataApplicationCapability_application_t84[] = { /* SEQUENCE */ - {FNAME("t84Protocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t84Profile") CHOICE, 1, 2, 2, SKIP, 0, _T84Profile}, -}; - -static field_t _DataApplicationCapability_application_nlpid[] = { /* SEQUENCE */ - {FNAME("nlpidProtocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("nlpidData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _DataApplicationCapability_application[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("t120") CHOICE, 3, 7, 14, DECODE | EXT, - offsetof(DataApplicationCapability_application, t120), - _DataProtocolCapability}, - {FNAME("dsm-cc") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("userData") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t84") SEQ, 0, 2, 2, SKIP, 0, - _DataApplicationCapability_application_t84}, - {FNAME("t434") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("h224") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("nlpid") SEQ, 0, 2, 2, SKIP, 0, - _DataApplicationCapability_application_nlpid}, - {FNAME("dsvdControl") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h222DataPartitioning") CHOICE, 3, 7, 14, SKIP | EXT, 0, - _DataProtocolCapability}, - {FNAME("t30fax") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, - {FNAME("t140") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, - {FNAME("t38fax") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("genericDataCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _DataApplicationCapability[] = { /* SEQUENCE */ - {FNAME("application") CHOICE, 4, 10, 14, DECODE | EXT, - offsetof(DataApplicationCapability, application), - _DataApplicationCapability_application}, - {FNAME("maxBitRate") INT, CONS, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _EncryptionMode[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("h233Encryption") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _DataType[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("nullData") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("videoData") CHOICE, 3, 5, 6, SKIP | EXT, 0, _VideoCapability}, - {FNAME("audioData") CHOICE, 4, 14, 22, SKIP | EXT, 0, - _AudioCapability}, - {FNAME("data") SEQ, 0, 2, 2, DECODE | EXT, offsetof(DataType, data), - _DataApplicationCapability}, - {FNAME("encryptionData") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _EncryptionMode}, - {FNAME("h235Control") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("h235Media") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, - {FNAME("multiplexedStream") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, -}; - -static field_t _H222LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("resourceID") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("subChannelID") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("pcr-pid") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("programDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("streamDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters_adaptationLayerType_al3[] = { /* SEQUENCE */ - {FNAME("controlFieldOctets") INT, 2, 0, 0, SKIP, 0, NULL}, - {FNAME("sendBufferSize") INT, CONS, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters_adaptationLayerType[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, - _H245_NonStandardParameter}, - {FNAME("al1Framed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al1NotFramed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al2WithoutSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al2WithSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("al3") SEQ, 0, 2, 2, SKIP, 0, - _H223LogicalChannelParameters_adaptationLayerType_al3}, - {FNAME("al1M") SEQ, 0, 7, 8, SKIP | EXT, 0, NULL}, - {FNAME("al2M") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, - {FNAME("al3M") SEQ, 0, 5, 6, SKIP | EXT, 0, NULL}, -}; - -static field_t _H223LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("adaptationLayerType") CHOICE, 3, 6, 9, SKIP | EXT, 0, - _H223LogicalChannelParameters_adaptationLayerType}, - {FNAME("segmentableFlag") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CRCLength[] = { /* CHOICE */ - {FNAME("crc8bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("crc16bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("crc32bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76HDLCParameters[] = { /* SEQUENCE */ - {FNAME("crcLength") CHOICE, 2, 3, 3, SKIP | EXT, 0, _CRCLength}, - {FNAME("n401") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("loopbackTestProcedure") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_suspendResume[] = { /* CHOICE */ - {FNAME("noSuspendResume") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("suspendResumewAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("suspendResumewoAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_mode_eRM_recovery[] = { /* CHOICE */ - {FNAME("rej") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("sREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("mSREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters_mode_eRM[] = { /* SEQUENCE */ - {FNAME("windowSize") INT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("recovery") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode_eRM_recovery}, -}; - -static field_t _V76LogicalChannelParameters_mode[] = { /* CHOICE */ - {FNAME("eRM") SEQ, 0, 2, 2, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode_eRM}, - {FNAME("uNERM") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V75Parameters[] = { /* SEQUENCE */ - {FNAME("audioHeaderPresent") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _V76LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("hdlcParameters") SEQ, 0, 3, 3, SKIP | EXT, 0, - _V76HDLCParameters}, - {FNAME("suspendResume") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _V76LogicalChannelParameters_suspendResume}, - {FNAME("uIH") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("mode") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _V76LogicalChannelParameters_mode}, - {FNAME("v75Parameters") SEQ, 0, 1, 1, SKIP | EXT, 0, _V75Parameters}, -}; - -static field_t _H2250LogicalChannelParameters_nonStandard[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, -}; - -static field_t _UnicastAddress_iPAddress[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 4, 0, DECODE, - offsetof(UnicastAddress_iPAddress, network), NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPXAddress[] = { /* SEQUENCE */ - {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, - {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iP6Address[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress_routing[] = { /* CHOICE */ - {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress_route[] = { /* SEQUENCE OF */ - {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, -}; - -static field_t _UnicastAddress_iPSourceRouteAddress[] = { /* SEQUENCE */ - {FNAME("routing") CHOICE, 1, 2, 2, SKIP, 0, - _UnicastAddress_iPSourceRouteAddress_routing}, - {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, - _UnicastAddress_iPSourceRouteAddress_route}, -}; - -static field_t _UnicastAddress[] = { /* CHOICE */ - {FNAME("iPAddress") SEQ, 0, 2, 2, DECODE | EXT, - offsetof(UnicastAddress, iPAddress), _UnicastAddress_iPAddress}, - {FNAME("iPXAddress") SEQ, 0, 3, 3, SKIP | EXT, 0, - _UnicastAddress_iPXAddress}, - {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _UnicastAddress_iP6Address}, - {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("iPSourceRouteAddress") SEQ, 0, 4, 4, SKIP | EXT, 0, - _UnicastAddress_iPSourceRouteAddress}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress_iPAddress[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress_iP6Address[] = { /* SEQUENCE */ - {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _MulticastAddress[] = { /* CHOICE */ - {FNAME("iPAddress") SEQ, 0, 2, 2, SKIP | EXT, 0, - _MulticastAddress_iPAddress}, - {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, - _MulticastAddress_iP6Address}, - {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, - {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, -}; - -static field_t _H245_TransportAddress[] = { /* CHOICE */ - {FNAME("unicastAddress") CHOICE, 3, 5, 7, DECODE | EXT, - offsetof(H245_TransportAddress, unicastAddress), _UnicastAddress}, - {FNAME("multicastAddress") CHOICE, 1, 2, 4, SKIP | EXT, 0, - _MulticastAddress}, -}; - -static field_t _H2250LogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _H2250LogicalChannelParameters_nonStandard}, - {FNAME("sessionID") INT, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("associatedSessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelParameters, mediaChannel), - _H245_TransportAddress}, - {FNAME("mediaGuaranteedDelivery") BOOL, FIXD, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelParameters, mediaControlChannel), - _H245_TransportAddress}, - {FNAME("mediaControlGuaranteedDelivery") BOOL, FIXD, 0, 0, STOP | OPT, - 0, NULL}, - {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destination") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, STOP | OPT, 0, NULL}, - {FNAME("mediaPacketization") CHOICE, 0, 1, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("transportCapability") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, - NULL}, - {FNAME("redundancyEncoding") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("source") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, - _H222LogicalChannelParameters}, - {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, - _H223LogicalChannelParameters}, - {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, - _V76LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, - {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_forwardLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("dataType") CHOICE, 3, 6, 9, DECODE | EXT, - offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, - dataType), _DataType}, - {FNAME("multiplexParameters") CHOICE, 2, 3, 5, DECODE | EXT, - offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters}, - {FNAME("forwardLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, - 0, NULL}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, - _H223LogicalChannelParameters}, - {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, - _V76LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, -}; - -static field_t _OpenLogicalChannel_reverseLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("dataType") CHOICE, 3, 6, 9, SKIP | EXT, 0, _DataType}, - {FNAME("multiplexParameters") CHOICE, 1, 2, 3, DECODE | EXT | OPT, - offsetof(OpenLogicalChannel_reverseLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters}, - {FNAME("reverseLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, - 0, NULL}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _NetworkAccessParameters_distribution[] = { /* CHOICE */ - {FNAME("unicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("multicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Q2931Address_address[] = { /* CHOICE */ - {FNAME("internationalNumber") NUMSTR, 4, 1, 0, SKIP, 0, NULL}, - {FNAME("nsapAddress") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, -}; - -static field_t _Q2931Address[] = { /* SEQUENCE */ - {FNAME("address") CHOICE, 1, 2, 2, SKIP | EXT, 0, - _Q2931Address_address}, - {FNAME("subaddress") OCTSTR, 5, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _NetworkAccessParameters_networkAddress[] = { /* CHOICE */ - {FNAME("q2931Address") SEQ, 1, 2, 2, SKIP | EXT, 0, _Q2931Address}, - {FNAME("e164Address") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("localAreaAddress") CHOICE, 1, 2, 2, DECODE | EXT, - offsetof(NetworkAccessParameters_networkAddress, localAreaAddress), - _H245_TransportAddress}, -}; - -static field_t _NetworkAccessParameters[] = { /* SEQUENCE */ - {FNAME("distribution") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, - _NetworkAccessParameters_distribution}, - {FNAME("networkAddress") CHOICE, 2, 3, 3, DECODE | EXT, - offsetof(NetworkAccessParameters, networkAddress), - _NetworkAccessParameters_networkAddress}, - {FNAME("associateConference") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("externalReference") OCTSTR, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("t120SetupProcedure") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, -}; - -static field_t _OpenLogicalChannel[] = { /* SEQUENCE */ - {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("forwardLogicalChannelParameters") SEQ, 1, 3, 5, DECODE | EXT, - offsetof(OpenLogicalChannel, forwardLogicalChannelParameters), - _OpenLogicalChannel_forwardLogicalChannelParameters}, - {FNAME("reverseLogicalChannelParameters") SEQ, 1, 2, 4, - DECODE | EXT | OPT, offsetof(OpenLogicalChannel, - reverseLogicalChannelParameters), - _OpenLogicalChannel_reverseLogicalChannelParameters}, - {FNAME("separateStack") SEQ, 2, 4, 5, DECODE | EXT | OPT, - offsetof(OpenLogicalChannel, separateStack), - _NetworkAccessParameters}, - {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _Setup_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Setup_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, h245Address), _TransportAddress}, - {FNAME("sourceAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_sourceAddress}, - {FNAME("sourceInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("destinationAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destinationAddress}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, destCallSignalAddress), _TransportAddress}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destExtraCallInfo}, - {FNAME("destExtraCRV") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Setup_UUIE_destExtraCRV}, - {FNAME("activeMC") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("conferenceGoal") CHOICE, 2, 3, 5, SKIP | EXT, 0, - _Setup_UUIE_conferenceGoal}, - {FNAME("callServices") SEQ, 0, 8, 8, SKIP | EXT | OPT, 0, - _QseriesOptions}, - {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, - {FNAME("sourceCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Setup_UUIE, sourceCallSignalAddress), _TransportAddress}, - {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityCapability") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Setup_UUIE, fastStart), _Setup_UUIE_fastStart}, - {FNAME("mediaWaitForConnect") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("canOverlapSend") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("connectionParameters") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("symmetricOperationRequired") NUL, FIXD, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("neededFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("desiredFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("supportedFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("parallelH245Control") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("additionalSourceAddresses") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, -}; - -static field_t _CallProceeding_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _CallProceeding_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(CallProceeding_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(CallProceeding_UUIE, fastStart), - _CallProceeding_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Connect_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Connect_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Connect_UUIE, h245Address), _TransportAddress}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Connect_UUIE, fastStart), _Connect_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("connectedAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Alerting_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Alerting_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Alerting_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Alerting_UUIE, fastStart), _Alerting_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("alertingAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Information_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Information_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Information_UUIE, fastStart), _Information_UUIE_fastStart}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _ReleaseCompleteReason[] = { /* CHOICE */ - {FNAME("noBandwidth") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatekeeperResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("unreachableDestination") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationRejection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("invalidRevision") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("noPermission") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("unreachableGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatewayResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("badFormatAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("adaptiveBusy") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("inConf") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("facilityCallDeflection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("securityDenied") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("calledPartyNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callerNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("newConnectionNeeded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardReason") SEQ, 0, 2, 2, SKIP, 0, NULL}, - {FNAME("replaceWithConferenceInvite") OCTSTR, FIXD, 16, 0, SKIP, 0, - NULL}, - {FNAME("genericDataReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("neededFeatureNotSupported") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("tunnelledSignallingRejected") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _ReleaseComplete_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("reason") CHOICE, 4, 12, 22, SKIP | EXT | OPT, 0, - _ReleaseCompleteReason}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("busyAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, -}; - -static field_t _Facility_UUIE_alternativeAliasAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _FacilityReason[] = { /* CHOICE */ - {FNAME("routeCallToGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callForwarded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("routeCallToMC") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("conferenceListChoice") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("startH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("noH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("newTokens") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("featureSetUpdate") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("forwardedElements") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("transportedInformation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _Facility_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Facility_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("alternativeAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Facility_UUIE, alternativeAddress), _TransportAddress}, - {FNAME("alternativeAliasAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Facility_UUIE_alternativeAliasAddress}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, - {FNAME("reason") CHOICE, 2, 4, 11, DECODE | EXT, - offsetof(Facility_UUIE, reason), _FacilityReason}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("conferences") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Facility_UUIE, h245Address), _TransportAddress}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Facility_UUIE, fastStart), _Facility_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT | OPT, 0, NULL}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - NULL}, -}; - -static field_t _CallIdentifier[] = { /* SEQUENCE */ - {FNAME("guid") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, -}; - -static field_t _SecurityServiceMode[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, - {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("default") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _SecurityCapabilities[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("encryption") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, - {FNAME("authenticaton") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, - {FNAME("integrity") CHOICE, 2, 3, 3, SKIP | EXT, 0, - _SecurityServiceMode}, -}; - -static field_t _H245Security[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, - {FNAME("noSecurity") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("tls") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, - {FNAME("ipsec") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, -}; - -static field_t _DHset[] = { /* SEQUENCE */ - {FNAME("halfkey") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("modSize") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, - {FNAME("generator") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _TypedCertificate[] = { /* SEQUENCE */ - {FNAME("type") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("certificate") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _H235_NonStandardParameter[] = { /* SEQUENCE */ - {FNAME("nonStandardIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _ClearToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("password") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("dhkey") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, _DHset}, - {FNAME("challenge") OCTSTR, 7, 8, 0, SKIP | OPT, 0, NULL}, - {FNAME("random") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("certificate") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, - _TypedCertificate}, - {FNAME("generalID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, - _H235_NonStandardParameter}, - {FNAME("eckasdhkey") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, NULL}, - {FNAME("sendersID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _Progress_UUIE_tokens[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, -}; - -static field_t _Params[] = { /* SEQUENCE */ - {FNAME("ranInt") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("iv8") OCTSTR, FIXD, 8, 0, SKIP | OPT, 0, NULL}, - {FNAME("iv16") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdHash_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdHash[] = { /* SEQUENCE */ - {FNAME("alias") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdHash_token}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdHash_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdHash[] = { /* SEQUENCE */ - {FNAME("gatekeeperId") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdHash_token}, -}; - -static field_t _CryptoH323Token_cryptoEPPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoEPCert[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoGKCert[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoH323Token_cryptoFastStart[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoEncryptedToken_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoEncryptedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoEncryptedToken_token}, -}; - -static field_t _CryptoToken_cryptoSignedToken_token[] = { /* SEQUENCE */ - {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoSignedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("token") SEQ, 0, 4, 4, SKIP, 0, - _CryptoToken_cryptoSignedToken_token}, -}; - -static field_t _CryptoToken_cryptoHashedToken_token[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken_cryptoHashedToken[] = { /* SEQUENCE */ - {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("hashedVals") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, - {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoHashedToken_token}, -}; - -static field_t _CryptoToken_cryptoPwdEncr[] = { /* SEQUENCE */ - {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, - {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _CryptoToken[] = { /* CHOICE */ - {FNAME("cryptoEncryptedToken") SEQ, 0, 2, 2, SKIP, 0, - _CryptoToken_cryptoEncryptedToken}, - {FNAME("cryptoSignedToken") SEQ, 0, 2, 2, SKIP, 0, - _CryptoToken_cryptoSignedToken}, - {FNAME("cryptoHashedToken") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoHashedToken}, - {FNAME("cryptoPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoToken_cryptoPwdEncr}, -}; - -static field_t _CryptoH323Token[] = { /* CHOICE */ - {FNAME("cryptoEPPwdHash") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdHash}, - {FNAME("cryptoGKPwdHash") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdHash}, - {FNAME("cryptoEPPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoEPPwdEncr}, - {FNAME("cryptoGKPwdEncr") SEQ, 0, 3, 3, SKIP, 0, - _CryptoH323Token_cryptoGKPwdEncr}, - {FNAME("cryptoEPCert") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoEPCert}, - {FNAME("cryptoGKCert") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoGKCert}, - {FNAME("cryptoFastStart") SEQ, 0, 4, 4, SKIP, 0, - _CryptoH323Token_cryptoFastStart}, - {FNAME("nestedcryptoToken") CHOICE, 2, 4, 4, SKIP | EXT, 0, - _CryptoToken}, -}; - -static field_t _Progress_UUIE_cryptoTokens[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 8, 8, SKIP | EXT, 0, _CryptoH323Token}, -}; - -static field_t _Progress_UUIE_fastStart[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, - sizeof(OpenLogicalChannel), _OpenLogicalChannel} - , -}; - -static field_t _Progress_UUIE[] = { /* SEQUENCE */ - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, - _EndpointType}, - {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(Progress_UUIE, h245Address), _TransportAddress}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, - _CallIdentifier}, - {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, - _H245Security}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Progress_UUIE_tokens}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _Progress_UUIE_cryptoTokens}, - {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, - offsetof(Progress_UUIE, fastStart), _Progress_UUIE_fastStart}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H323_UU_PDU_h323_message_body[] = { /* CHOICE */ - {FNAME("setup") SEQ, 7, 13, 39, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, setup), _Setup_UUIE}, - {FNAME("callProceeding") SEQ, 1, 3, 12, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, callProceeding), - _CallProceeding_UUIE}, - {FNAME("connect") SEQ, 1, 4, 19, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, connect), _Connect_UUIE}, - {FNAME("alerting") SEQ, 1, 3, 17, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, alerting), _Alerting_UUIE}, - {FNAME("information") SEQ, 0, 1, 7, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, information), - _Information_UUIE}, - {FNAME("releaseComplete") SEQ, 1, 2, 11, SKIP | EXT, 0, - _ReleaseComplete_UUIE}, - {FNAME("facility") SEQ, 3, 5, 21, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, facility), _Facility_UUIE}, - {FNAME("progress") SEQ, 5, 8, 11, DECODE | EXT, - offsetof(H323_UU_PDU_h323_message_body, progress), _Progress_UUIE}, - {FNAME("empty") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("status") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("statusInquiry") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("setupAcknowledge") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, - {FNAME("notify") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, -}; - -static field_t _RequestMessage[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("masterSlaveDetermination") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("terminalCapabilitySet") SEQ, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("openLogicalChannel") SEQ, 1, 3, 5, DECODE | EXT, - offsetof(RequestMessage, openLogicalChannel), _OpenLogicalChannel}, - {FNAME("closeLogicalChannel") SEQ, 0, 2, 3, STOP | EXT, 0, NULL}, - {FNAME("requestChannelClose") SEQ, 0, 1, 3, STOP | EXT, 0, NULL}, - {FNAME("multiplexEntrySend") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntry") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestMode") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("roundTripDelayRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("communicationModeRequest") SEQ, 0, 0, 0, STOP | EXT, 0, NULL}, - {FNAME("conferenceRequest") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, - {FNAME("multilinkRequest") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("logicalChannelRateRequest") SEQ, 0, 3, 3, STOP | EXT, 0, - NULL}, -}; - -static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ - {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, - _H222LogicalChannelParameters}, - {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, - offsetof - (OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters, - h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, -}; - -static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters[] = { /* SEQUENCE */ - {FNAME("reverseLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("multiplexParameters") CHOICE, 0, 1, 2, DECODE | EXT | OPT, - offsetof(OpenLogicalChannelAck_reverseLogicalChannelParameters, - multiplexParameters), - _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters}, - {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _H2250LogicalChannelAckParameters_nonStandard[] = { /* SEQUENCE OF */ - {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, -}; - -static field_t _H2250LogicalChannelAckParameters[] = { /* SEQUENCE */ - {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _H2250LogicalChannelAckParameters_nonStandard}, - {FNAME("sessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelAckParameters, mediaChannel), - _H245_TransportAddress}, - {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, - offsetof(H2250LogicalChannelAckParameters, mediaControlChannel), - _H245_TransportAddress}, - {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, SKIP | OPT, 0, NULL}, - {FNAME("flowControlToZero") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, -}; - -static field_t _OpenLogicalChannelAck_forwardMultiplexAckParameters[] = { /* CHOICE */ - {FNAME("h2250LogicalChannelAckParameters") SEQ, 5, 5, 7, DECODE | EXT, - offsetof(OpenLogicalChannelAck_forwardMultiplexAckParameters, - h2250LogicalChannelAckParameters), - _H2250LogicalChannelAckParameters}, -}; - -static field_t _OpenLogicalChannelAck[] = { /* SEQUENCE */ - {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("reverseLogicalChannelParameters") SEQ, 2, 3, 4, - DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, - reverseLogicalChannelParameters), - _OpenLogicalChannelAck_reverseLogicalChannelParameters}, - {FNAME("separateStack") SEQ, 2, 4, 5, SKIP | EXT | OPT, 0, NULL}, - {FNAME("forwardMultiplexAckParameters") CHOICE, 0, 1, 1, - DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, - forwardMultiplexAckParameters), - _OpenLogicalChannelAck_forwardMultiplexAckParameters}, - {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _ResponseMessage[] = { /* CHOICE */ - {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("masterSlaveDeterminationAck") SEQ, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("masterSlaveDeterminationReject") SEQ, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("terminalCapabilitySetAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("terminalCapabilitySetReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("openLogicalChannelAck") SEQ, 1, 2, 5, DECODE | EXT, - offsetof(ResponseMessage, openLogicalChannelAck), - _OpenLogicalChannelAck}, - {FNAME("openLogicalChannelReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("closeLogicalChannelAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestChannelCloseAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestChannelCloseReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("multiplexEntrySendAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("multiplexEntrySendReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntryAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("requestMultiplexEntryReject") SEQ, 0, 2, 2, STOP | EXT, 0, - NULL}, - {FNAME("requestModeAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("requestModeReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("roundTripDelayResponse") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("maintenanceLoopReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, - {FNAME("communicationModeResponse") CHOICE, 0, 1, 1, STOP | EXT, 0, - NULL}, - {FNAME("conferenceResponse") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, - {FNAME("multilinkResponse") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("logicalChannelRateAcknowledge") SEQ, 0, 3, 3, STOP | EXT, 0, - NULL}, - {FNAME("logicalChannelRateReject") SEQ, 1, 4, 4, STOP | EXT, 0, NULL}, -}; - -static field_t _MultimediaSystemControlMessage[] = { /* CHOICE */ - {FNAME("request") CHOICE, 4, 11, 15, DECODE | EXT, - offsetof(MultimediaSystemControlMessage, request), _RequestMessage}, - {FNAME("response") CHOICE, 5, 19, 24, DECODE | EXT, - offsetof(MultimediaSystemControlMessage, response), - _ResponseMessage}, - {FNAME("command") CHOICE, 3, 7, 12, STOP | EXT, 0, NULL}, - {FNAME("indication") CHOICE, 4, 14, 23, STOP | EXT, 0, NULL}, -}; - -static field_t _H323_UU_PDU_h245Control[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 2, 4, 4, DECODE | OPEN | EXT, - sizeof(MultimediaSystemControlMessage), - _MultimediaSystemControlMessage} - , -}; - -static field_t _H323_UU_PDU[] = { /* SEQUENCE */ - {FNAME("h323-message-body") CHOICE, 3, 7, 13, DECODE | EXT, - offsetof(H323_UU_PDU, h323_message_body), - _H323_UU_PDU_h323_message_body}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("h4501SupplementaryService") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - NULL}, - {FNAME("h245Tunneling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("h245Control") SEQOF, SEMI, 0, 4, DECODE | OPT, - offsetof(H323_UU_PDU, h245Control), _H323_UU_PDU_h245Control}, - {FNAME("nonStandardControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("tunnelledSignallingMessage") SEQ, 2, 4, 4, STOP | EXT | OPT, - 0, NULL}, - {FNAME("provisionalRespToH245Tunneling") NUL, FIXD, 0, 0, STOP | OPT, - 0, NULL}, - {FNAME("stimulusControl") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _H323_UserInformation[] = { /* SEQUENCE */ - {FNAME("h323-uu-pdu") SEQ, 1, 2, 11, DECODE | EXT, - offsetof(H323_UserInformation, h323_uu_pdu), _H323_UU_PDU}, - {FNAME("user-data") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _GatekeeperRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(GatekeeperRequest, rasAddress), _TransportAddress}, - {FNAME("endpointType") SEQ, 6, 8, 10, STOP | EXT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("authenticationCapability") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("algorithmOIDs") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _GatekeeperConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(GatekeeperConfirm, rasAddress), _TransportAddress}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("authenticationMode") CHOICE, 3, 7, 8, STOP | EXT | OPT, 0, - NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("algorithmOID") OID, BYTE, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RegistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationRequest_rasAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationRequest_terminalAlias[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _RegistrationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("discoveryComplete") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationRequest, callSignalAddress), - _RegistrationRequest_callSignalAddress}, - {FNAME("rasAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationRequest, rasAddress), - _RegistrationRequest_rasAddress}, - {FNAME("terminalType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _RegistrationRequest_terminalAlias}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("endpointVendor") SEQ, 2, 3, 3, SKIP | EXT, 0, - _VendorIdentifier}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, - offsetof(RegistrationRequest, timeToLive), NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("keepAlive") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("additiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("usageReportingCapability") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, - NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportedH248Packages") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("callCreditCapability") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("capacityReportingCapability") SEQ, 0, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RegistrationConfirm_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _RegistrationConfirm_terminalAlias[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _RegistrationConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(RegistrationConfirm, callSignalAddress), - _RegistrationConfirm_callSignalAddress}, - {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _RegistrationConfirm_terminalAlias}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, - {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, - offsetof(RegistrationConfirm, timeToLive), NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("preGrantedARQ") SEQ, 0, 4, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportsAdditiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureServerAlias") CHOICE, 1, 2, 7, STOP | EXT | OPT, 0, - NULL}, - {FNAME("capacityReportingSpec") SEQ, 0, 1, 1, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _UnregistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _UnregistrationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(UnregistrationRequest, callSignalAddress), - _UnregistrationRequest_callSignalAddress}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("reason") CHOICE, 2, 4, 5, STOP | EXT | OPT, 0, NULL}, - {FNAME("endpointAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _CallModel[] = { /* CHOICE */ - {FNAME("direct") NUL, FIXD, 0, 0, SKIP, 0, NULL}, - {FNAME("gatekeeperRouted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, -}; - -static field_t _AdmissionRequest_destinationInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest_destExtraCallInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest_srcInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _AdmissionRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, - {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, _CallModel}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _AdmissionRequest_destinationInfo}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(AdmissionRequest, destCallSignalAddress), - _TransportAddress}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, - _AdmissionRequest_destExtraCallInfo}, - {FNAME("srcInfo") SEQOF, SEMI, 0, 0, SKIP, 0, - _AdmissionRequest_srcInfo}, - {FNAME("srcCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, - offsetof(AdmissionRequest, srcCallSignalAddress), _TransportAddress}, - {FNAME("bandWidth") INT, CONS, 0, 0, STOP, 0, NULL}, - {FNAME("callReferenceValue") INT, WORD, 0, 0, STOP, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, - {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, STOP, 0, NULL}, - {FNAME("activeMC") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("answerCall") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("callIdentifier") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, - {FNAME("srcAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("gatewayDataRate") SEQ, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _AdmissionConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("bandWidth") INT, CONS, 0, 0, SKIP, 0, NULL}, - {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT, 0, _CallModel}, - {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(AdmissionConfirm, destCallSignalAddress), - _TransportAddress}, - {FNAME("irrFrequency") INT, WORD, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("uuiesRequested") SEQ, 0, 9, 13, STOP | EXT, 0, NULL}, - {FNAME("language") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("useSpecifiedTransport") CHOICE, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _LocationRequest_destinationInfo[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, -}; - -static field_t _LocationRequest[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP, 0, - _LocationRequest_destinationInfo}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("replyAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationRequest, replyAddress), _TransportAddress}, - {FNAME("sourceInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, - NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("hopCount") INT, 8, 1, 0, STOP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, -}; - -static field_t _LocationConfirm[] = { /* SEQUENCE */ - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("callSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationConfirm, callSignalAddress), _TransportAddress}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(LocationConfirm, rasAddress), _TransportAddress}, - {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, - {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, - NULL}, - {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, - 0, NULL}, - {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, - {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _InfoRequestResponse_callSignalAddress[] = { /* SEQUENCE OF */ - {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, - sizeof(TransportAddress), _TransportAddress} - , -}; - -static field_t _InfoRequestResponse[] = { /* SEQUENCE */ - {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, - _NonStandardParameter}, - {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, - {FNAME("endpointType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, - {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, - {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, - offsetof(InfoRequestResponse, rasAddress), _TransportAddress}, - {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, - offsetof(InfoRequestResponse, callSignalAddress), - _InfoRequestResponse_callSignalAddress}, - {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("perCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, - {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, - {FNAME("needResponse") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, - {FNAME("irrStatus") CHOICE, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, - {FNAME("unsolicited") BOOL, FIXD, 0, 0, STOP, 0, NULL}, - {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, -}; - -static field_t _RasMessage[] = { /* CHOICE */ - {FNAME("gatekeeperRequest") SEQ, 4, 8, 18, DECODE | EXT, - offsetof(RasMessage, gatekeeperRequest), _GatekeeperRequest}, - {FNAME("gatekeeperConfirm") SEQ, 2, 5, 14, DECODE | EXT, - offsetof(RasMessage, gatekeeperConfirm), _GatekeeperConfirm}, - {FNAME("gatekeeperReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, - {FNAME("registrationRequest") SEQ, 3, 10, 31, DECODE | EXT, - offsetof(RasMessage, registrationRequest), _RegistrationRequest}, - {FNAME("registrationConfirm") SEQ, 3, 7, 24, DECODE | EXT, - offsetof(RasMessage, registrationConfirm), _RegistrationConfirm}, - {FNAME("registrationReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, - {FNAME("unregistrationRequest") SEQ, 3, 5, 15, DECODE | EXT, - offsetof(RasMessage, unregistrationRequest), _UnregistrationRequest}, - {FNAME("unregistrationConfirm") SEQ, 1, 2, 6, STOP | EXT, 0, NULL}, - {FNAME("unregistrationReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("admissionRequest") SEQ, 7, 16, 34, DECODE | EXT, - offsetof(RasMessage, admissionRequest), _AdmissionRequest}, - {FNAME("admissionConfirm") SEQ, 2, 6, 27, DECODE | EXT, - offsetof(RasMessage, admissionConfirm), _AdmissionConfirm}, - {FNAME("admissionReject") SEQ, 1, 3, 11, STOP | EXT, 0, NULL}, - {FNAME("bandwidthRequest") SEQ, 2, 7, 18, STOP | EXT, 0, NULL}, - {FNAME("bandwidthConfirm") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("bandwidthReject") SEQ, 1, 4, 9, STOP | EXT, 0, NULL}, - {FNAME("disengageRequest") SEQ, 1, 6, 19, STOP | EXT, 0, NULL}, - {FNAME("disengageConfirm") SEQ, 1, 2, 9, STOP | EXT, 0, NULL}, - {FNAME("disengageReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, - {FNAME("locationRequest") SEQ, 2, 5, 17, DECODE | EXT, - offsetof(RasMessage, locationRequest), _LocationRequest}, - {FNAME("locationConfirm") SEQ, 1, 4, 19, DECODE | EXT, - offsetof(RasMessage, locationConfirm), _LocationConfirm}, - {FNAME("locationReject") SEQ, 1, 3, 10, STOP | EXT, 0, NULL}, - {FNAME("infoRequest") SEQ, 2, 4, 15, STOP | EXT, 0, NULL}, - {FNAME("infoRequestResponse") SEQ, 3, 8, 16, DECODE | EXT, - offsetof(RasMessage, infoRequestResponse), _InfoRequestResponse}, - {FNAME("nonStandardMessage") SEQ, 0, 2, 7, STOP | EXT, 0, NULL}, - {FNAME("unknownMessageResponse") SEQ, 0, 1, 5, STOP | EXT, 0, NULL}, - {FNAME("requestInProgress") SEQ, 4, 6, 6, STOP | EXT, 0, NULL}, - {FNAME("resourcesAvailableIndicate") SEQ, 4, 9, 11, STOP | EXT, 0, - NULL}, - {FNAME("resourcesAvailableConfirm") SEQ, 4, 6, 7, STOP | EXT, 0, - NULL}, - {FNAME("infoRequestAck") SEQ, 4, 5, 5, STOP | EXT, 0, NULL}, - {FNAME("infoRequestNak") SEQ, 5, 7, 7, STOP | EXT, 0, NULL}, - {FNAME("serviceControlIndication") SEQ, 8, 10, 10, STOP | EXT, 0, - NULL}, - {FNAME("serviceControlResponse") SEQ, 7, 8, 8, STOP | EXT, 0, NULL}, -}; diff --git a/net/ipv4/netfilter/nf_nat_h323.c b/net/ipv4/netfilter/nf_nat_h323.c new file mode 100644 index 000000000000..fb9ab0114c23 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_h323.c @@ -0,0 +1,596 @@ +/* + * H.323 extension for NAT alteration. + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * Based on the 'brute force' H.323 NAT module by + * Jozsef Kadlecsik + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/****************************************************************************/ +static int set_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + unsigned int addroff, __be32 ip, __be16 port) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = ip_conntrack_get(*pskb, &ctinfo); + struct { + __be32 ip; + __be16 port; + } __attribute__ ((__packed__)) buf; + struct tcphdr _tcph, *th; + + buf.ip = ip; + buf.port = port; + addroff += dataoff; + + if ((*pskb)->nh.iph->protocol == IPPROTO_TCP) { + if (!nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + addroff, sizeof(buf), + (char *) &buf, sizeof(buf))) { + if (net_ratelimit()) + printk("nf_nat_h323: nf_nat_mangle_tcp_packet" + " error\n"); + return -1; + } + + /* Relocate data pointer */ + th = skb_header_pointer(*pskb, (*pskb)->nh.iph->ihl * 4, + sizeof(_tcph), &_tcph); + if (th == NULL) + return -1; + *data = (*pskb)->data + (*pskb)->nh.iph->ihl * 4 + + th->doff * 4 + dataoff; + } else { + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + addroff, sizeof(buf), + (char *) &buf, sizeof(buf))) { + if (net_ratelimit()) + printk("nf_nat_h323: nf_nat_mangle_udp_packet" + " error\n"); + return -1; + } + /* nf_nat_mangle_udp_packet uses skb_make_writable() to copy + * or pull everything in a linear buffer, so we can safely + * use the skb pointers now */ + *data = (*pskb)->data + (*pskb)->nh.iph->ihl * 4 + + sizeof(struct udphdr); + } + + return 0; +} + +/****************************************************************************/ +static int set_h225_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) +{ + return set_addr(pskb, data, dataoff, taddr->ipAddress.ip, + addr->ip, port); +} + +/****************************************************************************/ +static int set_h245_addr(struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) +{ + return set_addr(pskb, data, dataoff, + taddr->unicastAddress.iPAddress.network, + addr->ip, port); +} + +/****************************************************************************/ +static int set_sig_addr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int i; + __be16 port; + union nf_conntrack_address addr; + + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port)) { + if (addr.ip == ct->tuplehash[dir].tuple.src.u3.ip && + port == info->sig_port[dir]) { + /* GW->GK */ + + /* Fix for Gnomemeeting */ + if (i > 0 && + get_h225_addr(ct, *data, &taddr[0], + &addr, &port) && + (ntohl(addr.ip) & 0xff000000) == 0x7f000000) + i = 0; + + DEBUGP + ("nf_nat_ras: set signal address " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), port, + NIPQUAD(ct->tuplehash[!dir].tuple.dst. + ip), info->sig_port[!dir]); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir]. + tuple.dst.u3, + info->sig_port[!dir]); + } else if (addr.ip == ct->tuplehash[dir].tuple.dst.u3.ip && + port == info->sig_port[dir]) { + /* GK->GW */ + DEBUGP + ("nf_nat_ras: set signal address " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), port, + NIPQUAD(ct->tuplehash[!dir].tuple.src. + ip), info->sig_port[!dir]); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir]. + tuple.src.u3, + info->sig_port[!dir]); + } + } + } + + return 0; +} + +/****************************************************************************/ +static int set_ras_addr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + int dir = CTINFO2DIR(ctinfo); + int i; + __be16 port; + union nf_conntrack_address addr; + + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port) && + addr.ip == ct->tuplehash[dir].tuple.src.u3.ip && + port == ct->tuplehash[dir].tuple.src.u.udp.port) { + DEBUGP("nf_nat_ras: set rasAddress " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(ip), ntohs(port), + NIPQUAD(ct->tuplehash[!dir].tuple.dst.u3.ip), + ntohs(ct->tuplehash[!dir].tuple.dst.u.udp. + port)); + return set_h225_addr(pskb, data, 0, &taddr[i], + &ct->tuplehash[!dir].tuple.dst.u3, + ct->tuplehash[!dir].tuple. + dst.u.udp.port); + } + } + + return 0; +} + +/****************************************************************************/ +static int nat_rtp_rtcp(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int i; + u_int16_t nated_port; + + /* Set expectations for NAT */ + rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port; + rtp_exp->expectfn = nf_nat_follow_master; + rtp_exp->dir = !dir; + rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port; + rtcp_exp->expectfn = nf_nat_follow_master; + rtcp_exp->dir = !dir; + + /* Lookup existing expects */ + for (i = 0; i < H323_RTP_CHANNEL_MAX; i++) { + if (info->rtp_port[i][dir] == rtp_port) { + /* Expected */ + + /* Use allocated ports first. This will refresh + * the expects */ + rtp_exp->tuple.dst.u.udp.port = info->rtp_port[i][dir]; + rtcp_exp->tuple.dst.u.udp.port = + htons(ntohs(info->rtp_port[i][dir]) + 1); + break; + } else if (info->rtp_port[i][dir] == 0) { + /* Not expected */ + break; + } + } + + /* Run out of expectations */ + if (i >= H323_RTP_CHANNEL_MAX) { + if (net_ratelimit()) + printk("nf_nat_h323: out of expectations\n"); + return 0; + } + + /* Try to get a pair of ports. */ + for (nated_port = ntohs(rtp_exp->tuple.dst.u.udp.port); + nated_port != 0; nated_port += 2) { + rtp_exp->tuple.dst.u.udp.port = htons(nated_port); + if (nf_conntrack_expect_related(rtp_exp) == 0) { + rtcp_exp->tuple.dst.u.udp.port = + htons(nated_port + 1); + if (nf_conntrack_expect_related(rtcp_exp) == 0) + break; + nf_conntrack_unexpect_related(rtp_exp); + } + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_h323: out of RTP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h245_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons((port & htons(1)) ? nated_port + 1 : + nated_port)) == 0) { + /* Save ports */ + info->rtp_port[i][dir] = rtp_port; + info->rtp_port[i][!dir] = htons(nated_port); + } else { + nf_conntrack_unexpect_related(rtp_exp); + nf_conntrack_unexpect_related(rtcp_exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_h323: expect RTP %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(rtp_exp->tuple.src.ip), + ntohs(rtp_exp->tuple.src.u.udp.port), + NIPQUAD(rtp_exp->tuple.dst.ip), + ntohs(rtp_exp->tuple.dst.u.udp.port)); + DEBUGP("nf_nat_h323: expect RTCP %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(rtcp_exp->tuple.src.ip), + ntohs(rtcp_exp->tuple.src.u.udp.port), + NIPQUAD(rtcp_exp->tuple.dst.ip), + ntohs(rtcp_exp->tuple.dst.u.udp.port)); + + return 0; +} + +/****************************************************************************/ +static int nat_t120(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = nf_nat_follow_master; + exp->dir = !dir; + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_h323: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h245_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) < 0) { + nf_conntrack_unexpect_related(exp); + return -1; + } + + DEBUGP("nf_nat_h323: expect T.120 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static int nat_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = nf_nat_follow_master; + exp->dir = !dir; + + /* Check existing expects */ + if (info->sig_port[dir] == port) + nated_port = ntohs(info->sig_port[!dir]); + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_q931: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h225_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + } else { + nf_conntrack_unexpect_related(exp); + return -1; + } + + DEBUGP("nf_nat_q931: expect H.245 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/**************************************************************************** + * This conntrack expect function replaces nf_conntrack_q931_expect() + * which was set by nf_conntrack_h323.c. + ****************************************************************************/ +static void ip_nat_q931_expect(struct nf_conn *new, + struct nf_conntrack_expect *this) +{ + struct ip_nat_range range; + + if (this->tuple.src.u3.ip != 0) { /* Only accept calls from GK */ + nf_nat_follow_master(new, this); + return; + } + + /* This must be a fresh one. */ + BUG_ON(new->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = this->saved_proto; + range.min_ip = range.max_ip = + new->master->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING); +} + +/****************************************************************************/ +static int nat_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, int idx, + __be16 port, struct nf_conntrack_expect *exp) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port = ntohs(port); + union nf_conntrack_address addr; + + /* Set expectations for NAT */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = ip_nat_q931_expect; + exp->dir = !dir; + + /* Check existing expects */ + if (info->sig_port[dir] == port) + nated_port = ntohs(info->sig_port[!dir]); + + /* Try to get same port: if not, try to change it. */ + for (; nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_ras: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (set_h225_addr(pskb, data, 0, &taddr[idx], + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + /* Save ports */ + info->sig_port[dir] = port; + info->sig_port[!dir] = htons(nated_port); + + /* Fix for Gnomemeeting */ + if (idx > 0 && + get_h225_addr(ct, *data, &taddr[0], &addr, &port) && + (ntohl(addr.ip) & 0xff000000) == 0x7f000000) { + set_h225_addr_hook(pskb, data, 0, &taddr[0], + &ct->tuplehash[!dir].tuple.dst.u3, + info->sig_port[!dir]); + } + } else { + nf_conntrack_unexpect_related(exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_ras: expect Q.931 %u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static void ip_nat_callforwarding_expect(struct nf_conn *new, + struct nf_conntrack_expect *this) +{ + struct nf_nat_range range; + + /* This must be a fresh one. */ + BUG_ON(new->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip = new->tuplehash[!this->dir].tuple.src.u3.ip; + + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(new, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = this->saved_proto; + range.min_ip = range.max_ip = this->saved_ip; + + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(new, &range, NF_IP_PRE_ROUTING); +} + +/****************************************************************************/ +static int nat_callforwarding(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) +{ + int dir = CTINFO2DIR(ctinfo); + u_int16_t nated_port; + + /* Set expectations for NAT */ + exp->saved_ip = exp->tuple.dst.u3.ip; + exp->tuple.dst.u3.ip = ct->tuplehash[!dir].tuple.dst.u3.ip; + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->expectfn = ip_nat_callforwarding_expect; + exp->dir = !dir; + + /* Try to get same port: if not, try to change it. */ + for (nated_port = ntohs(port); nated_port != 0; nated_port++) { + exp->tuple.dst.u.tcp.port = htons(nated_port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (nated_port == 0) { /* No port available */ + if (net_ratelimit()) + printk("nf_nat_q931: out of TCP ports\n"); + return 0; + } + + /* Modify signal */ + if (!set_h225_addr(pskb, data, dataoff, taddr, + &ct->tuplehash[!dir].tuple.dst.u3, + htons(nated_port)) == 0) { + nf_conntrack_unexpect_related(exp); + return -1; + } + + /* Success */ + DEBUGP("nf_nat_q931: expect Call Forwarding " + "%u.%u.%u.%u:%hu->%u.%u.%u.%u:%hu\n", + NIPQUAD(exp->tuple.src.ip), ntohs(exp->tuple.src.u.tcp.port), + NIPQUAD(exp->tuple.dst.ip), ntohs(exp->tuple.dst.u.tcp.port)); + + return 0; +} + +/****************************************************************************/ +static int __init init(void) +{ + BUG_ON(rcu_dereference(set_h245_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_h225_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_sig_addr_hook) != NULL); + BUG_ON(rcu_dereference(set_ras_addr_hook) != NULL); + BUG_ON(rcu_dereference(nat_rtp_rtcp_hook) != NULL); + BUG_ON(rcu_dereference(nat_t120_hook) != NULL); + BUG_ON(rcu_dereference(nat_h245_hook) != NULL); + BUG_ON(rcu_dereference(nat_callforwarding_hook) != NULL); + BUG_ON(rcu_dereference(nat_q931_hook) != NULL); + + rcu_assign_pointer(set_h245_addr_hook, set_h245_addr); + rcu_assign_pointer(set_h225_addr_hook, set_h225_addr); + rcu_assign_pointer(set_sig_addr_hook, set_sig_addr); + rcu_assign_pointer(set_ras_addr_hook, set_ras_addr); + rcu_assign_pointer(nat_rtp_rtcp_hook, nat_rtp_rtcp); + rcu_assign_pointer(nat_t120_hook, nat_t120); + rcu_assign_pointer(nat_h245_hook, nat_h245); + rcu_assign_pointer(nat_callforwarding_hook, nat_callforwarding); + rcu_assign_pointer(nat_q931_hook, nat_q931); + + DEBUGP("nf_nat_h323: init success\n"); + return 0; +} + +/****************************************************************************/ +static void __exit fini(void) +{ + rcu_assign_pointer(set_h245_addr_hook, NULL); + rcu_assign_pointer(set_h225_addr_hook, NULL); + rcu_assign_pointer(set_sig_addr_hook, NULL); + rcu_assign_pointer(set_ras_addr_hook, NULL); + rcu_assign_pointer(nat_rtp_rtcp_hook, NULL); + rcu_assign_pointer(nat_t120_hook, NULL); + rcu_assign_pointer(nat_h245_hook, NULL); + rcu_assign_pointer(nat_callforwarding_hook, NULL); + rcu_assign_pointer(nat_q931_hook, NULL); + synchronize_rcu(); +} + +/****************************************************************************/ +module_init(init); +module_exit(fini); + +MODULE_AUTHOR("Jing Min Zhao "); +MODULE_DESCRIPTION("H.323 NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_h323"); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index f85fd43b344b..d8f3451c95b6 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -160,6 +160,25 @@ config NF_CONNTRACK_FTP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_H323 + tristate "H.323 protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + H.323 is a VoIP signalling protocol from ITU-T. As one of the most + important VoIP protocols, it is widely used by voice hardware and + software including voice gateways, IP phones, Netmeeting, OpenPhone, + Gnomemeeting, etc. + + With this module you can support H.323 on a connection tracking/NAT + firewall. + + This module supports RAS, Fast Start, H.245 Tunnelling, Call + Forwarding, RTP/RTCP and T.120 based audio, video, fax, chat, + whiteboard, file transfer, etc. For more information, please + visit http://nath323.sourceforge.net/. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index a5ee93817427..69b554576b6b 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -20,8 +20,11 @@ obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o # connection tracking helpers +nf_conntrack_h323-objs := nf_conntrack_h323_main.o nf_conntrack_h323_asn1.o + obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o +obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_h323_asn1.c b/net/netfilter/nf_conntrack_h323_asn1.c new file mode 100644 index 000000000000..f6fad713d484 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_asn1.c @@ -0,0 +1,874 @@ +/**************************************************************************** + * ip_conntrack_helper_h323_asn1.c - BER and PER decoding library for H.323 + * conntrack/NAT module. + * + * Copyright (c) 2006 by Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * See ip_conntrack_helper_h323_asn1.h for details. + * + ****************************************************************************/ + +#ifdef __KERNEL__ +#include +#else +#include +#endif +#include + +/* Trace Flag */ +#ifndef H323_TRACE +#define H323_TRACE 0 +#endif + +#if H323_TRACE +#define TAB_SIZE 4 +#define IFTHEN(cond, act) if(cond){act;} +#ifdef __KERNEL__ +#define PRINT printk +#else +#define PRINT printf +#endif +#define FNAME(name) name, +#else +#define IFTHEN(cond, act) +#define PRINT(fmt, args...) +#define FNAME(name) +#endif + +/* ASN.1 Types */ +#define NUL 0 +#define BOOL 1 +#define OID 2 +#define INT 3 +#define ENUM 4 +#define BITSTR 5 +#define NUMSTR 6 +#define NUMDGT 6 +#define TBCDSTR 6 +#define OCTSTR 7 +#define PRTSTR 7 +#define IA5STR 7 +#define GENSTR 7 +#define BMPSTR 8 +#define SEQ 9 +#define SET 9 +#define SEQOF 10 +#define SETOF 10 +#define CHOICE 11 + +/* Constraint Types */ +#define FIXD 0 +/* #define BITS 1-8 */ +#define BYTE 9 +#define WORD 10 +#define CONS 11 +#define SEMI 12 +#define UNCO 13 + +/* ASN.1 Type Attributes */ +#define SKIP 0 +#define STOP 1 +#define DECODE 2 +#define EXT 4 +#define OPEN 8 +#define OPT 16 + + +/* ASN.1 Field Structure */ +typedef struct field_t { +#if H323_TRACE + char *name; +#endif + unsigned char type; + unsigned char sz; + unsigned char lb; + unsigned char ub; + unsigned short attr; + unsigned short offset; + struct field_t *fields; +} field_t; + +/* Bit Stream */ +typedef struct { + unsigned char *buf; + unsigned char *beg; + unsigned char *end; + unsigned char *cur; + unsigned bit; +} bitstr_t; + +/* Tool Functions */ +#define INC_BIT(bs) if((++bs->bit)>7){bs->cur++;bs->bit=0;} +#define INC_BITS(bs,b) if((bs->bit+=b)>7){bs->cur+=bs->bit>>3;bs->bit&=7;} +#define BYTE_ALIGN(bs) if(bs->bit){bs->cur++;bs->bit=0;} +#define CHECK_BOUND(bs,n) if(bs->cur+(n)>bs->end)return(H323_ERROR_BOUND) +static unsigned get_len(bitstr_t * bs); +static unsigned get_bit(bitstr_t * bs); +static unsigned get_bits(bitstr_t * bs, unsigned b); +static unsigned get_bitmap(bitstr_t * bs, unsigned b); +static unsigned get_uint(bitstr_t * bs, int b); + +/* Decoder Functions */ +static int decode_nul(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bool(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_oid(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_int(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_enum(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_seq(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level); +static int decode_choice(bitstr_t * bs, field_t * f, char *base, int level); + +/* Decoder Functions Vector */ +typedef int (*decoder_t) (bitstr_t *, field_t *, char *, int); +static decoder_t Decoders[] = { + decode_nul, + decode_bool, + decode_oid, + decode_int, + decode_enum, + decode_bitstr, + decode_numstr, + decode_octstr, + decode_bmpstr, + decode_seq, + decode_seqof, + decode_choice, +}; + +/**************************************************************************** + * H.323 Types + ****************************************************************************/ +#include "nf_conntrack_h323_types.c" + +/**************************************************************************** + * Functions + ****************************************************************************/ +/* Assume bs is aligned && v < 16384 */ +unsigned get_len(bitstr_t * bs) +{ + unsigned v; + + v = *bs->cur++; + + if (v & 0x80) { + v &= 0x3f; + v <<= 8; + v += *bs->cur++; + } + + return v; +} + +/****************************************************************************/ +unsigned get_bit(bitstr_t * bs) +{ + unsigned b = (*bs->cur) & (0x80 >> bs->bit); + + INC_BIT(bs); + + return b; +} + +/****************************************************************************/ +/* Assume b <= 8 */ +unsigned get_bits(bitstr_t * bs, unsigned b) +{ + unsigned v, l; + + v = (*bs->cur) & (0xffU >> bs->bit); + l = b + bs->bit; + + if (l < 8) { + v >>= 8 - l; + bs->bit = l; + } else if (l == 8) { + bs->cur++; + bs->bit = 0; + } else { /* l > 8 */ + + v <<= 8; + v += *(++bs->cur); + v >>= 16 - l; + bs->bit = l - 8; + } + + return v; +} + +/****************************************************************************/ +/* Assume b <= 32 */ +unsigned get_bitmap(bitstr_t * bs, unsigned b) +{ + unsigned v, l, shift, bytes; + + if (!b) + return 0; + + l = bs->bit + b; + + if (l < 8) { + v = (unsigned) (*bs->cur) << (bs->bit + 24); + bs->bit = l; + } else if (l == 8) { + v = (unsigned) (*bs->cur++) << (bs->bit + 24); + bs->bit = 0; + } else { + for (bytes = l >> 3, shift = 24, v = 0; bytes; + bytes--, shift -= 8) + v |= (unsigned) (*bs->cur++) << shift; + + if (l < 32) { + v |= (unsigned) (*bs->cur) << shift; + v <<= bs->bit; + } else if (l > 32) { + v <<= bs->bit; + v |= (*bs->cur) >> (8 - bs->bit); + } + + bs->bit = l & 0x7; + } + + v &= 0xffffffff << (32 - b); + + return v; +} + +/**************************************************************************** + * Assume bs is aligned and sizeof(unsigned int) == 4 + ****************************************************************************/ +unsigned get_uint(bitstr_t * bs, int b) +{ + unsigned v = 0; + + switch (b) { + case 4: + v |= *bs->cur++; + v <<= 8; + case 3: + v |= *bs->cur++; + v <<= 8; + case 2: + v |= *bs->cur++; + v <<= 8; + case 1: + v |= *bs->cur++; + break; + } + return v; +} + +/****************************************************************************/ +int decode_nul(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bool(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + INC_BIT(bs); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_oid(bitstr_t * bs, field_t * f, char *base, int level) +{ + int len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = *bs->cur++; + bs->cur += len; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_int(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + bs->cur++; + break; + case WORD: /* 257 <= Range <= 64K */ + BYTE_ALIGN(bs); + bs->cur += 2; + break; + case CONS: /* 64K < Range < 4G */ + len = get_bits(bs, 2) + 1; + BYTE_ALIGN(bs); + if (base && (f->attr & DECODE)) { /* timeToLive */ + unsigned v = get_uint(bs, len) + f->lb; + PRINT(" = %u", v); + *((unsigned *) (base + f->offset)) = v; + } + bs->cur += len; + break; + case UNCO: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + len = get_len(bs); + bs->cur += len; + break; + default: /* 2 <= Range <= 255 */ + INC_BITS(bs, f->sz); + break; + } + + PRINT("\n"); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_enum(bitstr_t * bs, field_t * f, char *base, int level) +{ + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + if ((f->attr & EXT) && get_bit(bs)) { + INC_BITS(bs, 7); + } else { + INC_BITS(bs, f->sz); + } + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bitstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + BYTE_ALIGN(bs); + switch (f->sz) { + case FIXD: /* fixed length > 16 */ + len = f->lb; + break; + case WORD: /* 2-byte length */ + CHECK_BOUND(bs, 2); + len = (*bs->cur++) << 8; + len += (*bs->cur++) + f->lb; + break; + case SEMI: + CHECK_BOUND(bs, 2); + len = get_len(bs); + break; + default: + len = 0; + break; + } + + bs->cur += len >> 3; + bs->bit = len & 7; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_numstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + + BYTE_ALIGN(bs); + INC_BITS(bs, (len << 2)); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_octstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case FIXD: /* Range == 1 */ + if (f->lb > 2) { + BYTE_ALIGN(bs); + if (base && (f->attr & DECODE)) { + /* The IP Address */ + IFTHEN(f->lb == 4, + PRINT(" = %d.%d.%d.%d:%d", + bs->cur[0], bs->cur[1], + bs->cur[2], bs->cur[3], + bs->cur[4] * 256 + bs->cur[5])); + *((unsigned *) (base + f->offset)) = + bs->cur - bs->buf; + } + } + len = f->lb; + break; + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = (*bs->cur++) + f->lb; + break; + case SEMI: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + len = get_len(bs) + f->lb; + break; + default: /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + BYTE_ALIGN(bs); + break; + } + + bs->cur += len; + + PRINT("\n"); + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_bmpstr(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned len; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + switch (f->sz) { + case BYTE: /* Range == 256 */ + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + len = (*bs->cur++) + f->lb; + break; + default: /* 2 <= Range <= 255 */ + len = get_bits(bs, f->sz) + f->lb; + BYTE_ALIGN(bs); + break; + } + + bs->cur += len << 1; + + CHECK_BOUND(bs, 0); + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_seq(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned ext, bmp, i, opt, len = 0, bmp2, bmp2_len; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Extensible? */ + ext = (f->attr & EXT) ? get_bit(bs) : 0; + + /* Get fields bitmap */ + bmp = get_bitmap(bs, f->sz); + if (base) + *(unsigned *) base = bmp; + + /* Decode the root components */ + for (i = opt = 0, son = f->fields; i < f->lb; i++, son++) { + if (son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + return H323_ERROR_STOP; + } + + if (son->attr & OPT) { /* Optional component */ + if (!((0x80000000U >> (opt++)) & bmp)) /* Not exist */ + continue; + } + + /* Decode */ + if (son->attr & OPEN) { /* Open field */ + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, + " ", son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + /* Decode */ + if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + } + + /* No extension? */ + if (!ext) + return H323_ERROR_NONE; + + /* Get the extension bitmap */ + bmp2_len = get_bits(bs, 7) + 1; + CHECK_BOUND(bs, (bmp2_len + 7) >> 3); + bmp2 = get_bitmap(bs, bmp2_len); + bmp |= bmp2 >> f->sz; + if (base) + *(unsigned *) base = bmp; + BYTE_ALIGN(bs); + + /* Decode the extension components */ + for (opt = 0; opt < bmp2_len; opt++, i++, son++) { + if (i < f->ub && son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + return H323_ERROR_STOP; + } + + if (!((0x80000000 >> opt) & bmp2)) /* Not present */ + continue; + + /* Check Range */ + if (i >= f->ub) { /* Newer Version? */ + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + bs->cur += len; + continue; + } + + CHECK_BOUND(bs, 2); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, base, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int decode_seqof(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned count, effective_count = 0, i, len = 0; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Decode item count */ + switch (f->sz) { + case BYTE: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 1); + count = *bs->cur++; + break; + case WORD: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + count = *bs->cur++; + count <<= 8; + count = *bs->cur++; + break; + case SEMI: + BYTE_ALIGN(bs); + CHECK_BOUND(bs, 2); + count = get_len(bs); + break; + default: + count = get_bits(bs, f->sz); + break; + } + count += f->lb; + + /* Write Count */ + if (base) { + effective_count = count > f->ub ? f->ub : count; + *(unsigned *) base = effective_count; + base += sizeof(unsigned); + } + + /* Decode nested field */ + son = f->fields; + if (base) + base -= son->offset; + for (i = 0; i < count; i++) { + if (son->attr & OPEN) { + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, + " ", son->name); + bs->cur += len; + continue; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, + i < + effective_count ? + base : NULL, + level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else + if ((err = (Decoders[son->type]) (bs, son, + i < + effective_count ? + base : NULL, + level + 1)) < + H323_ERROR_NONE) + return err; + + if (base) + base += son->offset; + } + + return H323_ERROR_NONE; +} + + +/****************************************************************************/ +int decode_choice(bitstr_t * bs, field_t * f, char *base, int level) +{ + unsigned type, ext, len = 0; + int err; + field_t *son; + unsigned char *beg = NULL; + + PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name); + + /* Decode? */ + base = (base && (f->attr & DECODE)) ? base + f->offset : NULL; + + /* Decode the choice index number */ + if ((f->attr & EXT) && get_bit(bs)) { + ext = 1; + type = get_bits(bs, 7) + f->lb; + } else { + ext = 0; + type = get_bits(bs, f->sz); + } + + /* Write Type */ + if (base) + *(unsigned *) base = type; + + /* Check Range */ + if (type >= f->ub) { /* Newer version? */ + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + bs->cur += len; + return H323_ERROR_NONE; + } + + /* Transfer to son level */ + son = &f->fields[type]; + if (son->attr & STOP) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", son->name); + return H323_ERROR_STOP; + } + + if (ext || (son->attr & OPEN)) { + BYTE_ALIGN(bs); + len = get_len(bs); + CHECK_BOUND(bs, len); + if (!base || !(son->attr & DECODE)) { + PRINT("%*.s%s\n", (level + 1) * TAB_SIZE, " ", + son->name); + bs->cur += len; + return H323_ERROR_NONE; + } + beg = bs->cur; + + if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < + H323_ERROR_NONE) + return err; + + bs->cur = beg + len; + bs->bit = 0; + } else if ((err = (Decoders[son->type]) (bs, son, base, level + 1)) < + H323_ERROR_NONE) + return err; + + return H323_ERROR_NONE; +} + +/****************************************************************************/ +int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage * ras) +{ + static field_t ras_message = { + FNAME("RasMessage") CHOICE, 5, 24, 32, DECODE | EXT, + 0, _RasMessage + }; + bitstr_t bs; + + bs.buf = bs.beg = bs.cur = buf; + bs.end = buf + sz; + bs.bit = 0; + + return decode_choice(&bs, &ras_message, (char *) ras, 0); +} + +/****************************************************************************/ +static int DecodeH323_UserInformation(unsigned char *buf, unsigned char *beg, + size_t sz, H323_UserInformation * uuie) +{ + static field_t h323_userinformation = { + FNAME("H323-UserInformation") SEQ, 1, 2, 2, DECODE | EXT, + 0, _H323_UserInformation + }; + bitstr_t bs; + + bs.buf = buf; + bs.beg = bs.cur = beg; + bs.end = beg + sz; + bs.bit = 0; + + return decode_seq(&bs, &h323_userinformation, (char *) uuie, 0); +} + +/****************************************************************************/ +int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz, + MultimediaSystemControlMessage * + mscm) +{ + static field_t multimediasystemcontrolmessage = { + FNAME("MultimediaSystemControlMessage") CHOICE, 2, 4, 4, + DECODE | EXT, 0, _MultimediaSystemControlMessage + }; + bitstr_t bs; + + bs.buf = bs.beg = bs.cur = buf; + bs.end = buf + sz; + bs.bit = 0; + + return decode_choice(&bs, &multimediasystemcontrolmessage, + (char *) mscm, 0); +} + +/****************************************************************************/ +int DecodeQ931(unsigned char *buf, size_t sz, Q931 * q931) +{ + unsigned char *p = buf; + int len; + + if (!p || sz < 1) + return H323_ERROR_BOUND; + + /* Protocol Discriminator */ + if (*p != 0x08) { + PRINT("Unknown Protocol Discriminator\n"); + return H323_ERROR_RANGE; + } + p++; + sz--; + + /* CallReferenceValue */ + if (sz < 1) + return H323_ERROR_BOUND; + len = *p++; + sz--; + if (sz < len) + return H323_ERROR_BOUND; + p += len; + sz -= len; + + /* Message Type */ + if (sz < 1) + return H323_ERROR_BOUND; + q931->MessageType = *p++; + PRINT("MessageType = %02X\n", q931->MessageType); + if (*p & 0x80) { + p++; + sz--; + } + + /* Decode Information Elements */ + while (sz > 0) { + if (*p == 0x7e) { /* UserUserIE */ + if (sz < 3) + break; + p++; + len = *p++ << 8; + len |= *p++; + sz -= 3; + if (sz < len) + break; + p++; + len--; + return DecodeH323_UserInformation(buf, p, len, + &q931->UUIE); + } + p++; + sz--; + if (sz < 1) + break; + len = *p++; + if (sz < len) + break; + p += len; + sz -= len; + } + + PRINT("Q.931 UUIE not found\n"); + + return H323_ERROR_BOUND; +} diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c new file mode 100644 index 000000000000..6d8568959f82 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -0,0 +1,1856 @@ +/* + * H.323 connection tracking helper + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + * + * Based on the 'brute force' H.323 connection tracking module by + * Jozsef Kadlecsik + * + * For more information, please see http://nath323.sourceforge.net/ + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/* Parameters */ +static unsigned int default_rrq_ttl __read_mostly = 300; +module_param(default_rrq_ttl, uint, 0600); +MODULE_PARM_DESC(default_rrq_ttl, "use this TTL if it's missing in RRQ"); + +static int gkrouted_only __read_mostly = 1; +module_param(gkrouted_only, int, 0600); +MODULE_PARM_DESC(gkrouted_only, "only accept calls from gatekeeper"); + +static int callforward_filter __read_mostly = 1; +module_param(callforward_filter, bool, 0600); +MODULE_PARM_DESC(callforward_filter, "only create call forwarding expectations " + "if both endpoints are on different sides " + "(determined by routing information)"); + +/* Hooks for NAT */ +int (*set_h245_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) + __read_mostly; +int (*set_h225_addr_hook) (struct sk_buff **pskb, + unsigned char **data, int dataoff, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 port) + __read_mostly; +int (*set_sig_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) __read_mostly; +int (*set_ras_addr_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) __read_mostly; +int (*nat_rtp_rtcp_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, + __be16 port, __be16 rtp_port, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) __read_mostly; +int (*nat_t120_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_h245_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_callforwarding_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr, __be16 port, + struct nf_conntrack_expect *exp) __read_mostly; +int (*nat_q931_hook) (struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, TransportAddress *taddr, int idx, + __be16 port, struct nf_conntrack_expect *exp) + __read_mostly; + +static DEFINE_SPINLOCK(nf_h323_lock); +static char *h323_buffer; + +static struct nf_conntrack_helper nf_conntrack_helper_h245; +static struct nf_conntrack_helper nf_conntrack_helper_q931[]; +static struct nf_conntrack_helper nf_conntrack_helper_ras[]; + +/****************************************************************************/ +static int get_tpkt_data(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + unsigned char **data, int *datalen, int *dataoff) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + struct tcphdr _tcph, *th; + int tcpdatalen; + int tcpdataoff; + unsigned char *tpkt; + int tpktlen; + int tpktoff; + + /* Get TCP header */ + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return 0; + + /* Get TCP data offset */ + tcpdataoff = protoff + th->doff * 4; + + /* Get TCP data length */ + tcpdatalen = (*pskb)->len - tcpdataoff; + if (tcpdatalen <= 0) /* No TCP data */ + goto clear_out; + + if (*data == NULL) { /* first TPKT */ + /* Get first TPKT pointer */ + tpkt = skb_header_pointer(*pskb, tcpdataoff, tcpdatalen, + h323_buffer); + BUG_ON(tpkt == NULL); + + /* Validate TPKT identifier */ + if (tcpdatalen < 4 || tpkt[0] != 0x03 || tpkt[1] != 0) { + /* Netmeeting sends TPKT header and data separately */ + if (info->tpkt_len[dir] > 0) { + DEBUGP("nf_ct_h323: previous packet " + "indicated separate TPKT data of %hu " + "bytes\n", info->tpkt_len[dir]); + if (info->tpkt_len[dir] <= tcpdatalen) { + /* Yes, there was a TPKT header + * received */ + *data = tpkt; + *datalen = info->tpkt_len[dir]; + *dataoff = 0; + goto out; + } + + /* Fragmented TPKT */ + if (net_ratelimit()) + printk("nf_ct_h323: " + "fragmented TPKT\n"); + goto clear_out; + } + + /* It is not even a TPKT */ + return 0; + } + tpktoff = 0; + } else { /* Next TPKT */ + tpktoff = *dataoff + *datalen; + tcpdatalen -= tpktoff; + if (tcpdatalen <= 4) /* No more TPKT */ + goto clear_out; + tpkt = *data + *datalen; + + /* Validate TPKT identifier */ + if (tpkt[0] != 0x03 || tpkt[1] != 0) + goto clear_out; + } + + /* Validate TPKT length */ + tpktlen = tpkt[2] * 256 + tpkt[3]; + if (tpktlen < 4) + goto clear_out; + if (tpktlen > tcpdatalen) { + if (tcpdatalen == 4) { /* Separate TPKT header */ + /* Netmeeting sends TPKT header and data separately */ + DEBUGP("nf_ct_h323: separate TPKT header indicates " + "there will be TPKT data of %hu bytes\n", + tpktlen - 4); + info->tpkt_len[dir] = tpktlen - 4; + return 0; + } + + if (net_ratelimit()) + printk("nf_ct_h323: incomplete TPKT (fragmented?)\n"); + goto clear_out; + } + + /* This is the encapsulated data */ + *data = tpkt + 4; + *datalen = tpktlen - 4; + *dataoff = tpktoff + 4; + + out: + /* Clear TPKT length */ + info->tpkt_len[dir] = 0; + return 1; + + clear_out: + info->tpkt_len[dir] = 0; + return 0; +} + +/****************************************************************************/ +static int get_h245_addr(struct nf_conn *ct, unsigned char *data, + H245_TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port) +{ + unsigned char *p; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int len; + + if (taddr->choice != eH245_TransportAddress_unicastAddress) + return 0; + + switch (taddr->unicastAddress.choice) { + case eUnicastAddress_iPAddress: + if (family != AF_INET) + return 0; + p = data + taddr->unicastAddress.iPAddress.network; + len = 4; + break; + case eUnicastAddress_iP6Address: + if (family != AF_INET6) + return 0; + p = data + taddr->unicastAddress.iP6Address.network; + len = 16; + break; + default: + return 0; + } + + memcpy(addr, p, len); + memset((void *)addr + len, 0, sizeof(*addr) - len); + memcpy(port, p + len, sizeof(__be16)); + + return 1; +} + +/****************************************************************************/ +static int expect_rtp_rtcp(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + __be16 rtp_port, rtcp_port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *rtp_exp; + struct nf_conntrack_expect *rtcp_exp; + typeof(nat_rtp_rtcp_hook) nat_rtp_rtcp; + + /* Read RTP or RTCP address */ + if (!get_h245_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* RTP port is even */ + port &= htons(~1); + rtp_port = port; + rtcp_port = htons(ntohs(port) + 1); + + /* Create expect for RTP */ + if ((rtp_exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(rtp_exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_UDP, NULL, &rtp_port); + + /* Create expect for RTCP */ + if ((rtcp_exp = nf_conntrack_expect_alloc(ct)) == NULL) { + nf_conntrack_expect_put(rtp_exp); + return -1; + } + nf_conntrack_expect_init(rtcp_exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_UDP, NULL, &rtcp_port); + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_rtp_rtcp = rcu_dereference(nat_rtp_rtcp_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + taddr, port, rtp_port, rtp_exp, rtcp_exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(rtp_exp) == 0) { + if (nf_conntrack_expect_related(rtcp_exp) == 0) { + DEBUGP("nf_ct_h323: expect RTP "); + NF_CT_DUMP_TUPLE(&rtp_exp->tuple); + DEBUGP("nf_ct_h323: expect RTCP "); + NF_CT_DUMP_TUPLE(&rtcp_exp->tuple); + } else { + nf_conntrack_unexpect_related(rtp_exp); + ret = -1; + } + } else + ret = -1; + } + + nf_conntrack_expect_put(rtp_exp); + nf_conntrack_expect_put(rtcp_exp); + + return ret; +} + +/****************************************************************************/ +static int expect_t120(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H245_TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_t120_hook) nat_t120; + + /* Read T.120 address */ + if (!get_h245_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* Create expect for T.120 connections */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple channels */ + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_t120 = rcu_dereference(nat_t120_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_t120(pskb, ct, ctinfo, data, dataoff, taddr, + port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_h323: expect T.120 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_h245_channel(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + H2250LogicalChannelParameters *channel) +{ + int ret; + + if (channel->options & eH2250LogicalChannelParameters_mediaChannel) { + /* RTP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &channel->mediaChannel); + if (ret < 0) + return -1; + } + + if (channel-> + options & eH2250LogicalChannelParameters_mediaControlChannel) { + /* RTCP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &channel->mediaControlChannel); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_olc(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + OpenLogicalChannel *olc) +{ + int ret; + + DEBUGP("nf_ct_h323: OpenLogicalChannel\n"); + + if (olc->forwardLogicalChannelParameters.multiplexParameters.choice == + eOpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters) + { + ret = process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olc-> + forwardLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olc->options & + eOpenLogicalChannel_reverseLogicalChannelParameters) && + (olc->reverseLogicalChannelParameters.options & + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters) + && (olc->reverseLogicalChannelParameters.multiplexParameters. + choice == + eOpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters)) + { + ret = + process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olc-> + reverseLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olc->options & eOpenLogicalChannel_separateStack) && + olc->forwardLogicalChannelParameters.dataType.choice == + eDataType_data && + olc->forwardLogicalChannelParameters.dataType.data.application. + choice == eDataApplicationCapability_application_t120 && + olc->forwardLogicalChannelParameters.dataType.data.application. + t120.choice == eDataProtocolCapability_separateLANStack && + olc->separateStack.networkAddress.choice == + eNetworkAccessParameters_networkAddress_localAreaAddress) { + ret = expect_t120(pskb, ct, ctinfo, data, dataoff, + &olc->separateStack.networkAddress. + localAreaAddress); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_olca(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + OpenLogicalChannelAck *olca) +{ + H2250LogicalChannelAckParameters *ack; + int ret; + + DEBUGP("nf_ct_h323: OpenLogicalChannelAck\n"); + + if ((olca->options & + eOpenLogicalChannelAck_reverseLogicalChannelParameters) && + (olca->reverseLogicalChannelParameters.options & + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters) + && (olca->reverseLogicalChannelParameters.multiplexParameters. + choice == + eOpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters_h2250LogicalChannelParameters)) + { + ret = process_h245_channel(pskb, ct, ctinfo, data, dataoff, + &olca-> + reverseLogicalChannelParameters. + multiplexParameters. + h2250LogicalChannelParameters); + if (ret < 0) + return -1; + } + + if ((olca->options & + eOpenLogicalChannelAck_forwardMultiplexAckParameters) && + (olca->forwardMultiplexAckParameters.choice == + eOpenLogicalChannelAck_forwardMultiplexAckParameters_h2250LogicalChannelAckParameters)) + { + ack = &olca->forwardMultiplexAckParameters. + h2250LogicalChannelAckParameters; + if (ack->options & + eH2250LogicalChannelAckParameters_mediaChannel) { + /* RTP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &ack->mediaChannel); + if (ret < 0) + return -1; + } + + if (ack->options & + eH2250LogicalChannelAckParameters_mediaControlChannel) { + /* RTCP */ + ret = expect_rtp_rtcp(pskb, ct, ctinfo, data, dataoff, + &ack->mediaControlChannel); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + MultimediaSystemControlMessage *mscm) +{ + switch (mscm->choice) { + case eMultimediaSystemControlMessage_request: + if (mscm->request.choice == + eRequestMessage_openLogicalChannel) { + return process_olc(pskb, ct, ctinfo, data, dataoff, + &mscm->request.openLogicalChannel); + } + DEBUGP("nf_ct_h323: H.245 Request %d\n", + mscm->request.choice); + break; + case eMultimediaSystemControlMessage_response: + if (mscm->response.choice == + eResponseMessage_openLogicalChannelAck) { + return process_olca(pskb, ct, ctinfo, data, dataoff, + &mscm->response. + openLogicalChannelAck); + } + DEBUGP("nf_ct_h323: H.245 Response %d\n", + mscm->response.choice); + break; + default: + DEBUGP("nf_ct_h323: H.245 signal %d\n", mscm->choice); + break; + } + + return 0; +} + +/****************************************************************************/ +static int h245_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static MultimediaSystemControlMessage mscm; + unsigned char *data = NULL; + int datalen; + int dataoff; + int ret; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) { + return NF_ACCEPT; + } + DEBUGP("nf_ct_h245: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Process each TPKT */ + while (get_tpkt_data(pskb, protoff, ct, ctinfo, + &data, &datalen, &dataoff)) { + DEBUGP("nf_ct_h245: TPKT len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode H.245 signal */ + ret = DecodeMultimediaSystemControlMessage(data, datalen, + &mscm); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_h245: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + /* We don't drop when decoding error */ + break; + } + + /* Process H.245 signal */ + if (process_h245(pskb, ct, ctinfo, &data, dataoff, &mscm) < 0) + goto drop; + } + + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_h245: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_h245 __read_mostly = { + .name = "H.245", + .me = THIS_MODULE, + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */, + .timeout = 240, + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = h245_help +}; + +/****************************************************************************/ +int get_h225_addr(struct nf_conn *ct, unsigned char *data, + TransportAddress *taddr, + union nf_conntrack_address *addr, __be16 *port) +{ + unsigned char *p; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int len; + + switch (taddr->choice) { + case eTransportAddress_ipAddress: + if (family != AF_INET) + return 0; + p = data + taddr->ipAddress.ip; + len = 4; + break; + case eTransportAddress_ip6Address: + if (family != AF_INET6) + return 0; + p = data + taddr->ip6Address.ip6; + len = 16; + break; + default: + return 0; + } + + memcpy(addr, p, len); + memset((void *)addr + len, 0, sizeof(*addr) - len); + memcpy(port, p + len, sizeof(__be16)); + + return 1; +} + +/****************************************************************************/ +static int expect_h245(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_h245_hook) nat_h245; + + /* Read h245Address */ + if (!get_h225_addr(ct, *data, taddr, &addr, &port) || + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) || + port == 0) + return 0; + + /* Create expect for h245 connection */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->helper = &nf_conntrack_helper_h245; + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_h245 = rcu_dereference(nat_h245_hook)) && + ct->status & IPS_NAT_MASK) { + /* NAT needed */ + ret = nat_h245(pskb, ct, ctinfo, data, dataoff, taddr, + port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_q931: expect H.245 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/* If the calling party is on the same side of the forward-to party, + * we don't need to track the second call */ +static int callforward_do_filter(union nf_conntrack_address *src, + union nf_conntrack_address *dst, + int family) +{ + struct flowi fl1, fl2; + int ret = 0; + + memset(&fl1, 0, sizeof(fl1)); + memset(&fl2, 0, sizeof(fl2)); + + switch (family) { + case AF_INET: { + struct rtable *rt1, *rt2; + + fl1.fl4_dst = src->ip; + fl2.fl4_dst = dst->ip; + if (ip_route_output_key(&rt1, &fl1) == 0) { + if (ip_route_output_key(&rt2, &fl2) == 0) { + if (rt1->rt_gateway == rt2->rt_gateway && + rt1->u.dst.dev == rt2->u.dst.dev) + ret = 1; + dst_release(&rt2->u.dst); + } + dst_release(&rt1->u.dst); + } + break; + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: { + struct rt6_info *rt1, *rt2; + + memcpy(&fl1.fl6_dst, src, sizeof(fl1.fl6_dst)); + memcpy(&fl2.fl6_dst, dst, sizeof(fl2.fl6_dst)); + rt1 = (struct rt6_info *)ip6_route_output(NULL, &fl1); + if (rt1) { + rt2 = (struct rt6_info *)ip6_route_output(NULL, &fl2); + if (rt2) { + if (!memcmp(&rt1->rt6i_gateway, &rt2->rt6i_gateway, + sizeof(rt1->rt6i_gateway)) && + rt1->u.dst.dev == rt2->u.dst.dev) + ret = 1; + dst_release(&rt2->u.dst); + } + dst_release(&rt1->u.dst); + } + break; + } +#endif + } + return ret; + +} + +/****************************************************************************/ +static int expect_callforwarding(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + TransportAddress *taddr) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_callforwarding_hook) nat_callforwarding; + + /* Read alternativeAddress */ + if (!get_h225_addr(ct, *data, taddr, &addr, &port) || port == 0) + return 0; + + /* If the calling party is on the same side of the forward-to party, + * we don't need to track the second call */ + if (callforward_filter && + callforward_do_filter(&addr, &ct->tuplehash[!dir].tuple.src.u3, + ct->tuplehash[!dir].tuple.src.l3num)) { + DEBUGP("nf_ct_q931: Call Forwarding not tracked\n"); + return 0; + } + + /* Create expect for the second call leg */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->helper = nf_conntrack_helper_q931; + + if (memcmp(&ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[!dir].tuple.dst.u3, + sizeof(ct->tuplehash[dir].tuple.src.u3)) && + (nat_callforwarding = rcu_dereference(nat_callforwarding_hook)) && + ct->status & IPS_NAT_MASK) { + /* Need NAT */ + ret = nat_callforwarding(pskb, ct, ctinfo, data, dataoff, + taddr, port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_q931: expect Call Forwarding "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_setup(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Setup_UUIE *setup) +{ + int dir = CTINFO2DIR(ctinfo); + int ret; + int i; + __be16 port; + union nf_conntrack_address addr; + typeof(set_h225_addr_hook) set_h225_addr; + + DEBUGP("nf_ct_q931: Setup\n"); + + if (setup->options & eSetup_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &setup->h245Address); + if (ret < 0) + return -1; + } + + set_h225_addr = rcu_dereference(set_h225_addr_hook); + if ((setup->options & eSetup_UUIE_destCallSignalAddress) && + (set_h225_addr) && ct->status && IPS_NAT_MASK && + get_h225_addr(ct, *data, &setup->destCallSignalAddress, + &addr, &port) && + memcmp(&addr, &ct->tuplehash[!dir].tuple.src.u3, sizeof(addr))) { + DEBUGP("nf_ct_q931: set destCallSignalAddress " + NIP6_FMT ":%hu->" NIP6_FMT ":%hu\n", + NIP6(*(struct in6_addr *)&addr), ntohs(port), + NIP6(*(struct in6_addr *)&ct->tuplehash[!dir].tuple.src.u3), + ntohs(ct->tuplehash[!dir].tuple.src.u.tcp.port)); + ret = set_h225_addr(pskb, data, dataoff, + &setup->destCallSignalAddress, + &ct->tuplehash[!dir].tuple.src.u3, + ct->tuplehash[!dir].tuple.src.u.tcp.port); + if (ret < 0) + return -1; + } + + if ((setup->options & eSetup_UUIE_sourceCallSignalAddress) && + (set_h225_addr) && ct->status & IPS_NAT_MASK && + get_h225_addr(ct, *data, &setup->sourceCallSignalAddress, + &addr, &port) && + memcmp(&addr, &ct->tuplehash[!dir].tuple.dst.u3, sizeof(addr))) { + DEBUGP("nf_ct_q931: set sourceCallSignalAddress " + NIP6_FMT ":%hu->" NIP6_FMT ":%hu\n", + NIP6(*(struct in6_addr *)&addr), ntohs(port), + NIP6(*(struct in6_addr *)&ct->tuplehash[!dir].tuple.dst.u3), + ntohs(ct->tuplehash[!dir].tuple.dst.u.tcp.port)); + ret = set_h225_addr(pskb, data, dataoff, + &setup->sourceCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + ct->tuplehash[!dir].tuple.dst.u.tcp.port); + if (ret < 0) + return -1; + } + + if (setup->options & eSetup_UUIE_fastStart) { + for (i = 0; i < setup->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &setup->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_callproceeding(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + CallProceeding_UUIE *callproc) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: CallProceeding\n"); + + if (callproc->options & eCallProceeding_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &callproc->h245Address); + if (ret < 0) + return -1; + } + + if (callproc->options & eCallProceeding_UUIE_fastStart) { + for (i = 0; i < callproc->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &callproc->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_connect(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Connect_UUIE *connect) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Connect\n"); + + if (connect->options & eConnect_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &connect->h245Address); + if (ret < 0) + return -1; + } + + if (connect->options & eConnect_UUIE_fastStart) { + for (i = 0; i < connect->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &connect->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_alerting(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Alerting_UUIE *alert) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Alerting\n"); + + if (alert->options & eAlerting_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &alert->h245Address); + if (ret < 0) + return -1; + } + + if (alert->options & eAlerting_UUIE_fastStart) { + for (i = 0; i < alert->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &alert->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_information(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Information_UUIE *info) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Information\n"); + + if (info->options & eInformation_UUIE_fastStart) { + for (i = 0; i < info->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &info->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_facility(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Facility_UUIE *facility) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Facility\n"); + + if (facility->reason.choice == eFacilityReason_callForwarded) { + if (facility->options & eFacility_UUIE_alternativeAddress) + return expect_callforwarding(pskb, ct, ctinfo, data, + dataoff, + &facility-> + alternativeAddress); + return 0; + } + + if (facility->options & eFacility_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &facility->h245Address); + if (ret < 0) + return -1; + } + + if (facility->options & eFacility_UUIE_fastStart) { + for (i = 0; i < facility->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &facility->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_progress(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, + Progress_UUIE *progress) +{ + int ret; + int i; + + DEBUGP("nf_ct_q931: Progress\n"); + + if (progress->options & eProgress_UUIE_h245Address) { + ret = expect_h245(pskb, ct, ctinfo, data, dataoff, + &progress->h245Address); + if (ret < 0) + return -1; + } + + if (progress->options & eProgress_UUIE_fastStart) { + for (i = 0; i < progress->fastStart.count; i++) { + ret = process_olc(pskb, ct, ctinfo, data, dataoff, + &progress->fastStart.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int process_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, int dataoff, Q931 *q931) +{ + H323_UU_PDU *pdu = &q931->UUIE.h323_uu_pdu; + int i; + int ret = 0; + + switch (pdu->h323_message_body.choice) { + case eH323_UU_PDU_h323_message_body_setup: + ret = process_setup(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.setup); + break; + case eH323_UU_PDU_h323_message_body_callProceeding: + ret = process_callproceeding(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body. + callProceeding); + break; + case eH323_UU_PDU_h323_message_body_connect: + ret = process_connect(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.connect); + break; + case eH323_UU_PDU_h323_message_body_alerting: + ret = process_alerting(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.alerting); + break; + case eH323_UU_PDU_h323_message_body_information: + ret = process_information(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body. + information); + break; + case eH323_UU_PDU_h323_message_body_facility: + ret = process_facility(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.facility); + break; + case eH323_UU_PDU_h323_message_body_progress: + ret = process_progress(pskb, ct, ctinfo, data, dataoff, + &pdu->h323_message_body.progress); + break; + default: + DEBUGP("nf_ct_q931: Q.931 signal %d\n", + pdu->h323_message_body.choice); + break; + } + + if (ret < 0) + return -1; + + if (pdu->options & eH323_UU_PDU_h245Control) { + for (i = 0; i < pdu->h245Control.count; i++) { + ret = process_h245(pskb, ct, ctinfo, data, dataoff, + &pdu->h245Control.item[i]); + if (ret < 0) + return -1; + } + } + + return 0; +} + +/****************************************************************************/ +static int q931_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static Q931 q931; + unsigned char *data = NULL; + int datalen; + int dataoff; + int ret; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) { + return NF_ACCEPT; + } + DEBUGP("nf_ct_q931: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Process each TPKT */ + while (get_tpkt_data(pskb, protoff, ct, ctinfo, + &data, &datalen, &dataoff)) { + DEBUGP("nf_ct_q931: TPKT len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode Q.931 signal */ + ret = DecodeQ931(data, datalen, &q931); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_q931: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + /* We don't drop when decoding error */ + break; + } + + /* Process Q.931 signal */ + if (process_q931(pskb, ct, ctinfo, &data, dataoff, &q931) < 0) + goto drop; + } + + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_q931: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_q931[] __read_mostly = { + { + .name = "Q.931", + .me = THIS_MODULE, + /* T.120 and H.245 */ + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, + .timeout = 240, + .tuple.src.l3num = AF_INET, + .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.tcp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = q931_help + }, + { + .name = "Q.931", + .me = THIS_MODULE, + /* T.120 and H.245 */ + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, + .timeout = 240, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.tcp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = q931_help + }, +}; + +/****************************************************************************/ +static unsigned char *get_udp_data(struct sk_buff **pskb, unsigned int protoff, + int *datalen) +{ + struct udphdr _uh, *uh; + int dataoff; + + uh = skb_header_pointer(*pskb, protoff, sizeof(_uh), &_uh); + if (uh == NULL) + return NULL; + dataoff = protoff + sizeof(_uh); + if (dataoff >= (*pskb)->len) + return NULL; + *datalen = (*pskb)->len - dataoff; + return skb_header_pointer(*pskb, dataoff, *datalen, h323_buffer); +} + +/****************************************************************************/ +static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, + union nf_conntrack_address *addr, + __be16 port) +{ + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple tuple; + + memset(&tuple.src.u3, 0, sizeof(tuple.src.u3)); + tuple.src.u.tcp.port = 0; + memcpy(&tuple.dst.u3, addr, sizeof(tuple.dst.u3)); + tuple.dst.u.tcp.port = port; + tuple.dst.protonum = IPPROTO_TCP; + + exp = __nf_conntrack_expect_find(&tuple); + if (exp && exp->master == ct) + return exp; + return NULL; +} + +/****************************************************************************/ +static int set_expect_timeout(struct nf_conntrack_expect *exp, + unsigned timeout) +{ + if (!exp || !del_timer(&exp->timeout)) + return 0; + + exp->timeout.expires = jiffies + timeout * HZ; + add_timer(&exp->timeout); + + return 1; +} + +/****************************************************************************/ +static int expect_q931(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, + TransportAddress *taddr, int count) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + int i; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(nat_q931_hook) nat_q931; + + /* Look for the first related address */ + for (i = 0; i < count; i++) { + if (get_h225_addr(ct, *data, &taddr[i], &addr, &port) && + memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, + sizeof(addr)) == 0 && port != 0) + break; + } + + if (i >= count) /* Not found */ + return 0; + + /* Create expect for Q.931 */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + gkrouted_only ? /* only accept calls from GK? */ + &ct->tuplehash[!dir].tuple.src.u3 : + NULL, + &ct->tuplehash[!dir].tuple.dst.u3, + IPPROTO_TCP, NULL, &port); + exp->helper = nf_conntrack_helper_q931; + exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple calls */ + + nat_q931 = rcu_dereference(nat_q931_hook); + if (nat_q931 && ct->status & IPS_NAT_MASK) { /* Need NAT */ + ret = nat_q931(pskb, ct, ctinfo, data, taddr, i, port, exp); + } else { /* Conntrack only */ + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + + /* Save port for looking up expect in processing RCF */ + info->sig_port[dir] = port; + } else + ret = -1; + } + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_grq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, GatekeeperRequest *grq) +{ + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: GRQ\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) /* NATed */ + return set_ras_addr(pskb, ct, ctinfo, data, + &grq->rasAddress, 1); + return 0; +} + +/****************************************************************************/ +static int process_gcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, GatekeeperConfirm *gcf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + + DEBUGP("nf_ct_ras: GCF\n"); + + if (!get_h225_addr(ct, *data, &gcf->rasAddress, &addr, &port)) + return 0; + + /* Registration port is the same as discovery port */ + if (!memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + port == ct->tuplehash[dir].tuple.src.u.udp.port) + return 0; + + /* Avoid RAS expectation loops. A GCF is never expected. */ + if (test_bit(IPS_EXPECTED_BIT, &ct->status)) + return 0; + + /* Need new expect */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_UDP, NULL, &port); + exp->helper = nf_conntrack_helper_ras; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect RAS "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_rrq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RegistrationRequest *rrq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int ret; + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: RRQ\n"); + + ret = expect_q931(pskb, ct, ctinfo, data, + rrq->callSignalAddress.item, + rrq->callSignalAddress.count); + if (ret < 0) + return -1; + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) { + ret = set_ras_addr(pskb, ct, ctinfo, data, + rrq->rasAddress.item, + rrq->rasAddress.count); + if (ret < 0) + return -1; + } + + if (rrq->options & eRegistrationRequest_timeToLive) { + DEBUGP("nf_ct_ras: RRQ TTL = %u seconds\n", rrq->timeToLive); + info->timeout = rrq->timeToLive; + } else + info->timeout = default_rrq_ttl; + + return 0; +} + +/****************************************************************************/ +static int process_rcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RegistrationConfirm *rcf) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret; + struct nf_conntrack_expect *exp; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: RCF\n"); + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + rcf->callSignalAddress.item, + rcf->callSignalAddress.count); + if (ret < 0) + return -1; + } + + if (rcf->options & eRegistrationConfirm_timeToLive) { + DEBUGP("nf_ct_ras: RCF TTL = %u seconds\n", rcf->timeToLive); + info->timeout = rcf->timeToLive; + } + + if (info->timeout > 0) { + DEBUGP + ("nf_ct_ras: set RAS connection timeout to %u seconds\n", + info->timeout); + nf_ct_refresh(ct, *pskb, info->timeout * HZ); + + /* Set expect timeout */ + read_lock_bh(&nf_conntrack_lock); + exp = find_expect(ct, &ct->tuplehash[dir].tuple.dst.u3, + info->sig_port[!dir]); + if (exp) { + DEBUGP("nf_ct_ras: set Q.931 expect " + "timeout to %u seconds for", + info->timeout); + NF_CT_DUMP_TUPLE(&exp->tuple); + set_expect_timeout(exp, info->timeout); + } + read_unlock_bh(&nf_conntrack_lock); + } + + return 0; +} + +/****************************************************************************/ +static int process_urq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, UnregistrationRequest *urq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + int ret; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: URQ\n"); + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + urq->callSignalAddress.item, + urq->callSignalAddress.count); + if (ret < 0) + return -1; + } + + /* Clear old expect */ + nf_ct_remove_expectations(ct); + info->sig_port[dir] = 0; + info->sig_port[!dir] = 0; + + /* Give it 30 seconds for UCF or URJ */ + nf_ct_refresh(ct, *pskb, 30 * HZ); + + return 0; +} + +/****************************************************************************/ +static int process_arq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, AdmissionRequest *arq) +{ + struct nf_ct_h323_master *info = &nfct_help(ct)->help.ct_h323_info; + int dir = CTINFO2DIR(ctinfo); + __be16 port; + union nf_conntrack_address addr; + typeof(set_h225_addr_hook) set_h225_addr; + + DEBUGP("nf_ct_ras: ARQ\n"); + + set_h225_addr = rcu_dereference(set_h225_addr_hook); + if ((arq->options & eAdmissionRequest_destCallSignalAddress) && + get_h225_addr(ct, *data, &arq->destCallSignalAddress, + &addr, &port) && + !memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + port == info->sig_port[dir] && + set_h225_addr && ct->status & IPS_NAT_MASK) { + /* Answering ARQ */ + return set_h225_addr(pskb, data, 0, + &arq->destCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + info->sig_port[!dir]); + } + + if ((arq->options & eAdmissionRequest_srcCallSignalAddress) && + get_h225_addr(ct, *data, &arq->srcCallSignalAddress, + &addr, &port) && + !memcmp(&addr, &ct->tuplehash[dir].tuple.src.u3, sizeof(addr)) && + set_h225_addr && ct->status & IPS_NAT_MASK) { + /* Calling ARQ */ + return set_h225_addr(pskb, data, 0, + &arq->srcCallSignalAddress, + &ct->tuplehash[!dir].tuple.dst.u3, + port); + } + + return 0; +} + +/****************************************************************************/ +static int process_acf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, AdmissionConfirm *acf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: ACF\n"); + + if (!get_h225_addr(ct, *data, &acf->destCallSignalAddress, + &addr, &port)) + return 0; + + if (!memcmp(&addr, &ct->tuplehash[dir].tuple.dst.u3, sizeof(addr))) { + /* Answering ACF */ + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) + return set_sig_addr(pskb, ct, ctinfo, data, + &acf->destCallSignalAddress, 1); + return 0; + } + + /* Need new expect */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; + exp->helper = nf_conntrack_helper_q931; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + return ret; +} + +/****************************************************************************/ +static int process_lrq(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, LocationRequest *lrq) +{ + typeof(set_ras_addr_hook) set_ras_addr; + + DEBUGP("nf_ct_ras: LRQ\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) + return set_ras_addr(pskb, ct, ctinfo, data, + &lrq->replyAddress, 1); + return 0; +} + +/****************************************************************************/ +static int process_lcf(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, LocationConfirm *lcf) +{ + int dir = CTINFO2DIR(ctinfo); + int ret = 0; + __be16 port; + union nf_conntrack_address addr; + struct nf_conntrack_expect *exp; + + DEBUGP("nf_ct_ras: LCF\n"); + + if (!get_h225_addr(ct, *data, &lcf->callSignalAddress, + &addr, &port)) + return 0; + + /* Need new expect for call signal */ + if ((exp = nf_conntrack_expect_alloc(ct)) == NULL) + return -1; + nf_conntrack_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + &ct->tuplehash[!dir].tuple.src.u3, &addr, + IPPROTO_TCP, NULL, &port); + exp->flags = NF_CT_EXPECT_PERMANENT; + exp->helper = nf_conntrack_helper_q931; + + if (nf_conntrack_expect_related(exp) == 0) { + DEBUGP("nf_ct_ras: expect Q.931 "); + NF_CT_DUMP_TUPLE(&exp->tuple); + } else + ret = -1; + + nf_conntrack_expect_put(exp); + + /* Ignore rasAddress */ + + return ret; +} + +/****************************************************************************/ +static int process_irr(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, InfoRequestResponse *irr) +{ + int ret; + typeof(set_ras_addr_hook) set_ras_addr; + typeof(set_sig_addr_hook) set_sig_addr; + + DEBUGP("nf_ct_ras: IRR\n"); + + set_ras_addr = rcu_dereference(set_ras_addr_hook); + if (set_ras_addr && ct->status & IPS_NAT_MASK) { + ret = set_ras_addr(pskb, ct, ctinfo, data, + &irr->rasAddress, 1); + if (ret < 0) + return -1; + } + + set_sig_addr = rcu_dereference(set_sig_addr_hook); + if (set_sig_addr && ct->status & IPS_NAT_MASK) { + ret = set_sig_addr(pskb, ct, ctinfo, data, + irr->callSignalAddress.item, + irr->callSignalAddress.count); + if (ret < 0) + return -1; + } + + return 0; +} + +/****************************************************************************/ +static int process_ras(struct sk_buff **pskb, struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned char **data, RasMessage *ras) +{ + switch (ras->choice) { + case eRasMessage_gatekeeperRequest: + return process_grq(pskb, ct, ctinfo, data, + &ras->gatekeeperRequest); + case eRasMessage_gatekeeperConfirm: + return process_gcf(pskb, ct, ctinfo, data, + &ras->gatekeeperConfirm); + case eRasMessage_registrationRequest: + return process_rrq(pskb, ct, ctinfo, data, + &ras->registrationRequest); + case eRasMessage_registrationConfirm: + return process_rcf(pskb, ct, ctinfo, data, + &ras->registrationConfirm); + case eRasMessage_unregistrationRequest: + return process_urq(pskb, ct, ctinfo, data, + &ras->unregistrationRequest); + case eRasMessage_admissionRequest: + return process_arq(pskb, ct, ctinfo, data, + &ras->admissionRequest); + case eRasMessage_admissionConfirm: + return process_acf(pskb, ct, ctinfo, data, + &ras->admissionConfirm); + case eRasMessage_locationRequest: + return process_lrq(pskb, ct, ctinfo, data, + &ras->locationRequest); + case eRasMessage_locationConfirm: + return process_lcf(pskb, ct, ctinfo, data, + &ras->locationConfirm); + case eRasMessage_infoRequestResponse: + return process_irr(pskb, ct, ctinfo, data, + &ras->infoRequestResponse); + default: + DEBUGP("nf_ct_ras: RAS message %d\n", ras->choice); + break; + } + + return 0; +} + +/****************************************************************************/ +static int ras_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + static RasMessage ras; + unsigned char *data; + int datalen = 0; + int ret; + + DEBUGP("nf_ct_ras: skblen = %u\n", (*pskb)->len); + + spin_lock_bh(&nf_h323_lock); + + /* Get UDP data */ + data = get_udp_data(pskb, protoff, &datalen); + if (data == NULL) + goto accept; + DEBUGP("nf_ct_ras: RAS message len=%d ", datalen); + NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + + /* Decode RAS message */ + ret = DecodeRasMessage(data, datalen, &ras); + if (ret < 0) { + if (net_ratelimit()) + printk("nf_ct_ras: decoding error: %s\n", + ret == H323_ERROR_BOUND ? + "out of bound" : "out of range"); + goto accept; + } + + /* Process RAS message */ + if (process_ras(pskb, ct, ctinfo, &data, &ras) < 0) + goto drop; + + accept: + spin_unlock_bh(&nf_h323_lock); + return NF_ACCEPT; + + drop: + spin_unlock_bh(&nf_h323_lock); + if (net_ratelimit()) + printk("nf_ct_ras: packet dropped\n"); + return NF_DROP; +} + +/****************************************************************************/ +static struct nf_conntrack_helper nf_conntrack_helper_ras[] __read_mostly = { + { + .name = "RAS", + .me = THIS_MODULE, + .max_expected = 32, + .timeout = 240, + .tuple.src.l3num = AF_INET, + .tuple.src.u.udp.port = __constant_htons(RAS_PORT), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = ras_help, + }, + { + .name = "RAS", + .me = THIS_MODULE, + .max_expected = 32, + .timeout = 240, + .tuple.src.l3num = AF_INET6, + .tuple.src.u.udp.port = __constant_htons(RAS_PORT), + .tuple.dst.protonum = IPPROTO_UDP, + .mask.src.l3num = 0xFFFF, + .mask.src.u.udp.port = __constant_htons(0xFFFF), + .mask.dst.protonum = 0xFF, + .help = ras_help, + }, +}; + +/****************************************************************************/ +static void __exit nf_conntrack_h323_fini(void) +{ + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[1]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[0]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[1]); + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[0]); + kfree(h323_buffer); + DEBUGP("nf_ct_h323: fini\n"); +} + +/****************************************************************************/ +static int __init nf_conntrack_h323_init(void) +{ + int ret; + + h323_buffer = kmalloc(65536, GFP_KERNEL); + if (!h323_buffer) + return -ENOMEM; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_q931[0]); + if (ret < 0) + goto err1; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_q931[1]); + if (ret < 0) + goto err2; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_ras[0]); + if (ret < 0) + goto err3; + ret = nf_conntrack_helper_register(&nf_conntrack_helper_ras[1]); + if (ret < 0) + goto err4; + DEBUGP("nf_ct_h323: init success\n"); + return 0; + +err4: + nf_conntrack_helper_unregister(&nf_conntrack_helper_ras[0]); +err3: + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[1]); +err2: + nf_conntrack_helper_unregister(&nf_conntrack_helper_q931[0]); +err1: + return ret; +} + +/****************************************************************************/ +module_init(nf_conntrack_h323_init); +module_exit(nf_conntrack_h323_fini); + +EXPORT_SYMBOL_GPL(get_h225_addr); +EXPORT_SYMBOL_GPL(set_h245_addr_hook); +EXPORT_SYMBOL_GPL(set_h225_addr_hook); +EXPORT_SYMBOL_GPL(set_sig_addr_hook); +EXPORT_SYMBOL_GPL(set_ras_addr_hook); +EXPORT_SYMBOL_GPL(nat_rtp_rtcp_hook); +EXPORT_SYMBOL_GPL(nat_t120_hook); +EXPORT_SYMBOL_GPL(nat_h245_hook); +EXPORT_SYMBOL_GPL(nat_callforwarding_hook); +EXPORT_SYMBOL_GPL(nat_q931_hook); + +MODULE_AUTHOR("Jing Min Zhao "); +MODULE_DESCRIPTION("H.323 connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_h323"); diff --git a/net/netfilter/nf_conntrack_h323_types.c b/net/netfilter/nf_conntrack_h323_types.c new file mode 100644 index 000000000000..4c6f8b3b1208 --- /dev/null +++ b/net/netfilter/nf_conntrack_h323_types.c @@ -0,0 +1,1927 @@ +/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006 + * + * Copyright (c) 2006 Jing Min Zhao + * + * This source code is licensed under General Public License version 2. + */ + +static field_t _TransportAddress_ipAddress[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 4, 0, DECODE, + offsetof(TransportAddress_ipAddress, ip), NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute_route[] = { /* SEQUENCE OF */ + {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute_routing[] = { /* CHOICE */ + {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ipSourceRoute[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, + _TransportAddress_ipSourceRoute_route}, + {FNAME("routing") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _TransportAddress_ipSourceRoute_routing}, +}; + +static field_t _TransportAddress_ipxAddress[] = { /* SEQUENCE */ + {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, + {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("port") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress_ip6Address[] = { /* SEQUENCE */ + {FNAME("ip") OCTSTR, FIXD, 16, 0, DECODE, + offsetof(TransportAddress_ip6Address, ip6), NULL}, + {FNAME("port") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H221NonStandard[] = { /* SEQUENCE */ + {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _NonStandardIdentifier[] = { /* CHOICE */ + {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP | EXT, 0, + _H221NonStandard}, +}; + +static field_t _NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _NonStandardIdentifier}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TransportAddress[] = { /* CHOICE */ + {FNAME("ipAddress") SEQ, 0, 2, 2, DECODE, + offsetof(TransportAddress, ipAddress), _TransportAddress_ipAddress}, + {FNAME("ipSourceRoute") SEQ, 0, 4, 4, SKIP | EXT, 0, + _TransportAddress_ipSourceRoute}, + {FNAME("ipxAddress") SEQ, 0, 3, 3, SKIP, 0, + _TransportAddress_ipxAddress}, + {FNAME("ip6Address") SEQ, 0, 2, 2, DECODE | EXT, + offsetof(TransportAddress, ip6Address), _TransportAddress_ip6Address}, + {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, + _NonStandardParameter}, +}; + +static field_t _AliasAddress[] = { /* CHOICE */ + {FNAME("dialedDigits") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("h323-ID") BMPSTR, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("url-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("transportID") CHOICE, 3, 7, 7, SKIP | EXT, 0, NULL}, + {FNAME("email-ID") IA5STR, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("partyNumber") CHOICE, 3, 5, 5, SKIP | EXT, 0, NULL}, + {FNAME("mobileUIM") CHOICE, 1, 2, 2, SKIP | EXT, 0, NULL}, +}; + +static field_t _Setup_UUIE_sourceAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _VendorIdentifier[] = { /* SEQUENCE */ + {FNAME("vendor") SEQ, 0, 3, 3, SKIP | EXT, 0, _H221NonStandard}, + {FNAME("productId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("versionId") OCTSTR, BYTE, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _GatekeeperInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _H310Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H320Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H321Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H322Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H323Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H324Caps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _VoiceCaps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T120OnlyCaps[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("dataRatesSupported") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _SupportedProtocols[] = { /* CHOICE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP, 0, + _NonStandardParameter}, + {FNAME("h310") SEQ, 1, 1, 3, SKIP | EXT, 0, _H310Caps}, + {FNAME("h320") SEQ, 1, 1, 3, SKIP | EXT, 0, _H320Caps}, + {FNAME("h321") SEQ, 1, 1, 3, SKIP | EXT, 0, _H321Caps}, + {FNAME("h322") SEQ, 1, 1, 3, SKIP | EXT, 0, _H322Caps}, + {FNAME("h323") SEQ, 1, 1, 3, SKIP | EXT, 0, _H323Caps}, + {FNAME("h324") SEQ, 1, 1, 3, SKIP | EXT, 0, _H324Caps}, + {FNAME("voice") SEQ, 1, 1, 3, SKIP | EXT, 0, _VoiceCaps}, + {FNAME("t120-only") SEQ, 1, 1, 3, SKIP | EXT, 0, _T120OnlyCaps}, + {FNAME("nonStandardProtocol") SEQ, 2, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("t38FaxAnnexbOnly") SEQ, 2, 5, 5, SKIP | EXT, 0, NULL}, +}; + +static field_t _GatewayInfo_protocol[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 4, 9, 11, SKIP | EXT, 0, _SupportedProtocols}, +}; + +static field_t _GatewayInfo[] = { /* SEQUENCE */ + {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _GatewayInfo_protocol}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _McuInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("protocol") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _TerminalInfo[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, +}; + +static field_t _EndpointType[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("vendor") SEQ, 2, 3, 3, SKIP | EXT | OPT, 0, + _VendorIdentifier}, + {FNAME("gatekeeper") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, + _GatekeeperInfo}, + {FNAME("gateway") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, _GatewayInfo}, + {FNAME("mcu") SEQ, 1, 1, 2, SKIP | EXT | OPT, 0, _McuInfo}, + {FNAME("terminal") SEQ, 1, 1, 1, SKIP | EXT | OPT, 0, _TerminalInfo}, + {FNAME("mc") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedNode") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("set") BITSTR, FIXD, 32, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedTunnelledProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, + 0, NULL}, +}; + +static field_t _Setup_UUIE_destinationAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _Setup_UUIE_destExtraCallInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _Setup_UUIE_destExtraCRV[] = { /* SEQUENCE OF */ + {FNAME("item") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Setup_UUIE_conferenceGoal[] = { /* CHOICE */ + {FNAME("create") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("join") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("invite") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("capability-negotiation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callIndependentSupplementaryService") NUL, FIXD, 0, 0, SKIP, + 0, NULL}, +}; + +static field_t _Q954Details[] = { /* SEQUENCE */ + {FNAME("conferenceCalling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threePartyService") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _QseriesOptions[] = { /* SEQUENCE */ + {FNAME("q932Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q951Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q952Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q953Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q955Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q956Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q957Full") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("q954Info") SEQ, 0, 2, 2, SKIP | EXT, 0, _Q954Details}, +}; + +static field_t _CallType[] = { /* CHOICE */ + {FNAME("pointToPoint") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("oneToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nToOne") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nToN") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H245_NonStandardIdentifier_h221NonStandard[] = { /* SEQUENCE */ + {FNAME("t35CountryCode") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("t35Extension") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("manufacturerCode") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H245_NonStandardIdentifier[] = { /* CHOICE */ + {FNAME("object") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h221NonStandard") SEQ, 0, 3, 3, SKIP, 0, + _H245_NonStandardIdentifier_h221NonStandard}, +}; + +static field_t _H245_NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") CHOICE, 1, 2, 2, SKIP, 0, + _H245_NonStandardIdentifier}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H261VideoCapability[] = { /* SEQUENCE */ + {FNAME("qcifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cifMPI") INT, 2, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("maxBitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("stillImageTransmission") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H262VideoCapability[] = { /* SEQUENCE */ + {FNAME("profileAndLevel-SPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-MPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SNRatLL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SNRatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-SpatialatH-14") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("profileAndLevel-HPatML") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-HPatH-14") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("profileAndLevel-HPatHL") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("framesPerSecond") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H263VideoCapability[] = { /* SEQUENCE */ + {FNAME("sqcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("qcifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cifMPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cif4MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("cif16MPI") INT, 5, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("maxBitRate") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("unrestrictedVector") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("arithmeticCoding") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("advancedPrediction") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("pbFrames") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("temporalSpatialTradeOffCapability") BOOL, FIXD, 0, 0, SKIP, 0, + NULL}, + {FNAME("hrd-B") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("bppMaxKb") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowSqcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowQcifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCifMPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCif4MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("slowCif16MPI") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("errorCompensation") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("enhancementLayerInfo") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("h263Options") SEQ, 5, 29, 31, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _IS11172VideoCapability[] = { /* SEQUENCE */ + {FNAME("constrainedBitstream") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoBitRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("vbvBufferSize") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("samplesPerLine") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("linesPerFrame") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("pictureRate") INT, 4, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("luminanceSampleRate") INT, CONS, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("videoBadMBsCap") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _VideoCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("h261VideoCapability") SEQ, 2, 5, 6, SKIP | EXT, 0, + _H261VideoCapability}, + {FNAME("h262VideoCapability") SEQ, 6, 17, 18, SKIP | EXT, 0, + _H262VideoCapability}, + {FNAME("h263VideoCapability") SEQ, 7, 13, 21, SKIP | EXT, 0, + _H263VideoCapability}, + {FNAME("is11172VideoCapability") SEQ, 6, 7, 8, SKIP | EXT, 0, + _IS11172VideoCapability}, + {FNAME("genericVideoCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _AudioCapability_g7231[] = { /* SEQUENCE */ + {FNAME("maxAl-sduAudioFrames") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _IS11172AudioCapability[] = { /* SEQUENCE */ + {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _IS13818AudioCapability[] = { /* SEQUENCE */ + {FNAME("audioLayer1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioLayer3") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling16k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling22k05") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling24k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling32k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling44k1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("audioSampling48k") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("singleChannel") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("twoChannels") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threeChannels2-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("threeChannels3-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels2-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels2-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fourChannels3-1") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fiveChannels3-0-2-0") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fiveChannels3-2") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("lowFrequencyEnhancement") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("multilingual") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("bitRate") INT, WORD, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _AudioCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("g711Alaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Alaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Ulaw64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g711Ulaw56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-64k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-56k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g722-48k") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g7231") SEQ, 0, 2, 2, SKIP, 0, _AudioCapability_g7231}, + {FNAME("g728") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729AnnexA") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("is11172AudioCapability") SEQ, 0, 9, 9, SKIP | EXT, 0, + _IS11172AudioCapability}, + {FNAME("is13818AudioCapability") SEQ, 0, 21, 21, SKIP | EXT, 0, + _IS13818AudioCapability}, + {FNAME("g729wAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g729AnnexAwAnnexB") INT, BYTE, 1, 0, SKIP, 0, NULL}, + {FNAME("g7231AnnexCCapability") SEQ, 1, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmHalfRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("gsmEnhancedFullRate") SEQ, 0, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("genericAudioCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, + {FNAME("g729Extensions") SEQ, 1, 8, 8, SKIP | EXT, 0, NULL}, +}; + +static field_t _DataProtocolCapability[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("v14buffered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v42lapm") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdlcFrameTunnelling") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h310SeparateVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h310SingleVCStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("transparent") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("segmentationAndReassembly") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdlcFrameTunnelingwSAR") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v120") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("separateLANStack") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("v76wCompression") CHOICE, 2, 3, 3, SKIP | EXT, 0, NULL}, + {FNAME("tcp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("udp") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T84Profile_t84Restricted[] = { /* SEQUENCE */ + {FNAME("qcif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("cif") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("ccir601Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("ccir601Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdtvSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("hdtvProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g3FacsMH200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g3FacsMH200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g4FacsMMR200x100") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("g4FacsMMR200x200") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig200x200Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig200x200Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig300x300Seq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("jbig300x300Prog") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoLow") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoMedSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoMedProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoHighSeq") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("digPhotoHighProg") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _T84Profile[] = { /* CHOICE */ + {FNAME("t84Unrestricted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("t84Restricted") SEQ, 0, 19, 19, SKIP | EXT, 0, + _T84Profile_t84Restricted}, +}; + +static field_t _DataApplicationCapability_application_t84[] = { /* SEQUENCE */ + {FNAME("t84Protocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t84Profile") CHOICE, 1, 2, 2, SKIP, 0, _T84Profile}, +}; + +static field_t _DataApplicationCapability_application_nlpid[] = { /* SEQUENCE */ + {FNAME("nlpidProtocol") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("nlpidData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _DataApplicationCapability_application[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("t120") CHOICE, 3, 7, 14, DECODE | EXT, + offsetof(DataApplicationCapability_application, t120), + _DataProtocolCapability}, + {FNAME("dsm-cc") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("userData") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t84") SEQ, 0, 2, 2, SKIP, 0, + _DataApplicationCapability_application_t84}, + {FNAME("t434") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("h224") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("nlpid") SEQ, 0, 2, 2, SKIP, 0, + _DataApplicationCapability_application_nlpid}, + {FNAME("dsvdControl") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h222DataPartitioning") CHOICE, 3, 7, 14, SKIP | EXT, 0, + _DataProtocolCapability}, + {FNAME("t30fax") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, + {FNAME("t140") CHOICE, 3, 7, 14, SKIP | EXT, 0, NULL}, + {FNAME("t38fax") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("genericDataCapability") SEQ, 5, 6, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _DataApplicationCapability[] = { /* SEQUENCE */ + {FNAME("application") CHOICE, 4, 10, 14, DECODE | EXT, + offsetof(DataApplicationCapability, application), + _DataApplicationCapability_application}, + {FNAME("maxBitRate") INT, CONS, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _EncryptionMode[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("h233Encryption") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _DataType[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("nullData") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("videoData") CHOICE, 3, 5, 6, SKIP | EXT, 0, _VideoCapability}, + {FNAME("audioData") CHOICE, 4, 14, 22, SKIP | EXT, 0, + _AudioCapability}, + {FNAME("data") SEQ, 0, 2, 2, DECODE | EXT, offsetof(DataType, data), + _DataApplicationCapability}, + {FNAME("encryptionData") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _EncryptionMode}, + {FNAME("h235Control") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("h235Media") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, + {FNAME("multiplexedStream") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, +}; + +static field_t _H222LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("resourceID") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("subChannelID") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("pcr-pid") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("programDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("streamDescriptors") OCTSTR, SEMI, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters_adaptationLayerType_al3[] = { /* SEQUENCE */ + {FNAME("controlFieldOctets") INT, 2, 0, 0, SKIP, 0, NULL}, + {FNAME("sendBufferSize") INT, CONS, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters_adaptationLayerType[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, + _H245_NonStandardParameter}, + {FNAME("al1Framed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al1NotFramed") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al2WithoutSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al2WithSequenceNumbers") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("al3") SEQ, 0, 2, 2, SKIP, 0, + _H223LogicalChannelParameters_adaptationLayerType_al3}, + {FNAME("al1M") SEQ, 0, 7, 8, SKIP | EXT, 0, NULL}, + {FNAME("al2M") SEQ, 0, 2, 2, SKIP | EXT, 0, NULL}, + {FNAME("al3M") SEQ, 0, 5, 6, SKIP | EXT, 0, NULL}, +}; + +static field_t _H223LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("adaptationLayerType") CHOICE, 3, 6, 9, SKIP | EXT, 0, + _H223LogicalChannelParameters_adaptationLayerType}, + {FNAME("segmentableFlag") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CRCLength[] = { /* CHOICE */ + {FNAME("crc8bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("crc16bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("crc32bit") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76HDLCParameters[] = { /* SEQUENCE */ + {FNAME("crcLength") CHOICE, 2, 3, 3, SKIP | EXT, 0, _CRCLength}, + {FNAME("n401") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("loopbackTestProcedure") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_suspendResume[] = { /* CHOICE */ + {FNAME("noSuspendResume") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("suspendResumewAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("suspendResumewoAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_mode_eRM_recovery[] = { /* CHOICE */ + {FNAME("rej") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("sREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("mSREJ") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters_mode_eRM[] = { /* SEQUENCE */ + {FNAME("windowSize") INT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("recovery") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode_eRM_recovery}, +}; + +static field_t _V76LogicalChannelParameters_mode[] = { /* CHOICE */ + {FNAME("eRM") SEQ, 0, 2, 2, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode_eRM}, + {FNAME("uNERM") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V75Parameters[] = { /* SEQUENCE */ + {FNAME("audioHeaderPresent") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _V76LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("hdlcParameters") SEQ, 0, 3, 3, SKIP | EXT, 0, + _V76HDLCParameters}, + {FNAME("suspendResume") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _V76LogicalChannelParameters_suspendResume}, + {FNAME("uIH") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("mode") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _V76LogicalChannelParameters_mode}, + {FNAME("v75Parameters") SEQ, 0, 1, 1, SKIP | EXT, 0, _V75Parameters}, +}; + +static field_t _H2250LogicalChannelParameters_nonStandard[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, +}; + +static field_t _UnicastAddress_iPAddress[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 4, 0, DECODE, + offsetof(UnicastAddress_iPAddress, network), NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPXAddress[] = { /* SEQUENCE */ + {FNAME("node") OCTSTR, FIXD, 6, 0, SKIP, 0, NULL}, + {FNAME("netnum") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") OCTSTR, FIXD, 2, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iP6Address[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress_routing[] = { /* CHOICE */ + {FNAME("strict") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("loose") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress_route[] = { /* SEQUENCE OF */ + {FNAME("item") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, +}; + +static field_t _UnicastAddress_iPSourceRouteAddress[] = { /* SEQUENCE */ + {FNAME("routing") CHOICE, 1, 2, 2, SKIP, 0, + _UnicastAddress_iPSourceRouteAddress_routing}, + {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("route") SEQOF, SEMI, 0, 0, SKIP, 0, + _UnicastAddress_iPSourceRouteAddress_route}, +}; + +static field_t _UnicastAddress[] = { /* CHOICE */ + {FNAME("iPAddress") SEQ, 0, 2, 2, DECODE | EXT, + offsetof(UnicastAddress, iPAddress), _UnicastAddress_iPAddress}, + {FNAME("iPXAddress") SEQ, 0, 3, 3, SKIP | EXT, 0, + _UnicastAddress_iPXAddress}, + {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, + _UnicastAddress_iP6Address}, + {FNAME("netBios") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("iPSourceRouteAddress") SEQ, 0, 4, 4, SKIP | EXT, 0, + _UnicastAddress_iPSourceRouteAddress}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress_iPAddress[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 4, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress_iP6Address[] = { /* SEQUENCE */ + {FNAME("network") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("tsapIdentifier") INT, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _MulticastAddress[] = { /* CHOICE */ + {FNAME("iPAddress") SEQ, 0, 2, 2, SKIP | EXT, 0, + _MulticastAddress_iPAddress}, + {FNAME("iP6Address") SEQ, 0, 2, 2, SKIP | EXT, 0, + _MulticastAddress_iP6Address}, + {FNAME("nsap") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, + {FNAME("nonStandardAddress") SEQ, 0, 2, 2, SKIP, 0, NULL}, +}; + +static field_t _H245_TransportAddress[] = { /* CHOICE */ + {FNAME("unicastAddress") CHOICE, 3, 5, 7, DECODE | EXT, + offsetof(H245_TransportAddress, unicastAddress), _UnicastAddress}, + {FNAME("multicastAddress") CHOICE, 1, 2, 4, SKIP | EXT, 0, + _MulticastAddress}, +}; + +static field_t _H2250LogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _H2250LogicalChannelParameters_nonStandard}, + {FNAME("sessionID") INT, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("associatedSessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelParameters, mediaChannel), + _H245_TransportAddress}, + {FNAME("mediaGuaranteedDelivery") BOOL, FIXD, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelParameters, mediaControlChannel), + _H245_TransportAddress}, + {FNAME("mediaControlGuaranteedDelivery") BOOL, FIXD, 0, 0, STOP | OPT, + 0, NULL}, + {FNAME("silenceSuppression") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destination") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, STOP | OPT, 0, NULL}, + {FNAME("mediaPacketization") CHOICE, 0, 1, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("transportCapability") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, + NULL}, + {FNAME("redundancyEncoding") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("source") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, + _H222LogicalChannelParameters}, + {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, + _H223LogicalChannelParameters}, + {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, + _V76LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, + {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_forwardLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("dataType") CHOICE, 3, 6, 9, DECODE | EXT, + offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, + dataType), _DataType}, + {FNAME("multiplexParameters") CHOICE, 2, 3, 5, DECODE | EXT, + offsetof(OpenLogicalChannel_forwardLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannel_forwardLogicalChannelParameters_multiplexParameters}, + {FNAME("forwardLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, + 0, NULL}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h223LogicalChannelParameters") SEQ, 0, 2, 2, SKIP | EXT, 0, + _H223LogicalChannelParameters}, + {FNAME("v76LogicalChannelParameters") SEQ, 0, 5, 5, SKIP | EXT, 0, + _V76LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, +}; + +static field_t _OpenLogicalChannel_reverseLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("dataType") CHOICE, 3, 6, 9, SKIP | EXT, 0, _DataType}, + {FNAME("multiplexParameters") CHOICE, 1, 2, 3, DECODE | EXT | OPT, + offsetof(OpenLogicalChannel_reverseLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannel_reverseLogicalChannelParameters_multiplexParameters}, + {FNAME("reverseLogicalChannelDependency") INT, WORD, 1, 0, SKIP | OPT, + 0, NULL}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _NetworkAccessParameters_distribution[] = { /* CHOICE */ + {FNAME("unicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("multicast") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Q2931Address_address[] = { /* CHOICE */ + {FNAME("internationalNumber") NUMSTR, 4, 1, 0, SKIP, 0, NULL}, + {FNAME("nsapAddress") OCTSTR, 5, 1, 0, SKIP, 0, NULL}, +}; + +static field_t _Q2931Address[] = { /* SEQUENCE */ + {FNAME("address") CHOICE, 1, 2, 2, SKIP | EXT, 0, + _Q2931Address_address}, + {FNAME("subaddress") OCTSTR, 5, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _NetworkAccessParameters_networkAddress[] = { /* CHOICE */ + {FNAME("q2931Address") SEQ, 1, 2, 2, SKIP | EXT, 0, _Q2931Address}, + {FNAME("e164Address") NUMDGT, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("localAreaAddress") CHOICE, 1, 2, 2, DECODE | EXT, + offsetof(NetworkAccessParameters_networkAddress, localAreaAddress), + _H245_TransportAddress}, +}; + +static field_t _NetworkAccessParameters[] = { /* SEQUENCE */ + {FNAME("distribution") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, + _NetworkAccessParameters_distribution}, + {FNAME("networkAddress") CHOICE, 2, 3, 3, DECODE | EXT, + offsetof(NetworkAccessParameters, networkAddress), + _NetworkAccessParameters_networkAddress}, + {FNAME("associateConference") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("externalReference") OCTSTR, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("t120SetupProcedure") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, +}; + +static field_t _OpenLogicalChannel[] = { /* SEQUENCE */ + {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("forwardLogicalChannelParameters") SEQ, 1, 3, 5, DECODE | EXT, + offsetof(OpenLogicalChannel, forwardLogicalChannelParameters), + _OpenLogicalChannel_forwardLogicalChannelParameters}, + {FNAME("reverseLogicalChannelParameters") SEQ, 1, 2, 4, + DECODE | EXT | OPT, offsetof(OpenLogicalChannel, + reverseLogicalChannelParameters), + _OpenLogicalChannel_reverseLogicalChannelParameters}, + {FNAME("separateStack") SEQ, 2, 4, 5, DECODE | EXT | OPT, + offsetof(OpenLogicalChannel, separateStack), + _NetworkAccessParameters}, + {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _Setup_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Setup_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, h245Address), _TransportAddress}, + {FNAME("sourceAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_sourceAddress}, + {FNAME("sourceInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("destinationAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destinationAddress}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, destCallSignalAddress), _TransportAddress}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destExtraCallInfo}, + {FNAME("destExtraCRV") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Setup_UUIE_destExtraCRV}, + {FNAME("activeMC") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("conferenceGoal") CHOICE, 2, 3, 5, SKIP | EXT, 0, + _Setup_UUIE_conferenceGoal}, + {FNAME("callServices") SEQ, 0, 8, 8, SKIP | EXT | OPT, 0, + _QseriesOptions}, + {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, + {FNAME("sourceCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Setup_UUIE, sourceCallSignalAddress), _TransportAddress}, + {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityCapability") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Setup_UUIE, fastStart), _Setup_UUIE_fastStart}, + {FNAME("mediaWaitForConnect") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("canOverlapSend") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("connectionParameters") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("symmetricOperationRequired") NUL, FIXD, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("neededFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("desiredFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("supportedFeatures") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("parallelH245Control") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("additionalSourceAddresses") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, +}; + +static field_t _CallProceeding_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _CallProceeding_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(CallProceeding_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(CallProceeding_UUIE, fastStart), + _CallProceeding_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Connect_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Connect_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Connect_UUIE, h245Address), _TransportAddress}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Connect_UUIE, fastStart), _Connect_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("connectedAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Alerting_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Alerting_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Alerting_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Alerting_UUIE, fastStart), _Alerting_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("alertingAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Information_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Information_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Information_UUIE, fastStart), _Information_UUIE_fastStart}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _ReleaseCompleteReason[] = { /* CHOICE */ + {FNAME("noBandwidth") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatekeeperResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("unreachableDestination") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationRejection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("invalidRevision") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("noPermission") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("unreachableGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatewayResources") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("badFormatAddress") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("adaptiveBusy") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("inConf") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("facilityCallDeflection") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("securityDenied") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("calledPartyNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callerNotRegistered") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("newConnectionNeeded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardReason") SEQ, 0, 2, 2, SKIP, 0, NULL}, + {FNAME("replaceWithConferenceInvite") OCTSTR, FIXD, 16, 0, SKIP, 0, + NULL}, + {FNAME("genericDataReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("neededFeatureNotSupported") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("tunnelledSignallingRejected") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _ReleaseComplete_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("reason") CHOICE, 4, 12, 22, SKIP | EXT | OPT, 0, + _ReleaseCompleteReason}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("busyAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("presentationIndicator") CHOICE, 2, 3, 3, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("screeningIndicator") ENUM, 2, 0, 0, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, +}; + +static field_t _Facility_UUIE_alternativeAliasAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _FacilityReason[] = { /* CHOICE */ + {FNAME("routeCallToGatekeeper") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callForwarded") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("routeCallToMC") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("undefinedReason") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("conferenceListChoice") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("startH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("noH245") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("newTokens") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("featureSetUpdate") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("forwardedElements") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("transportedInformation") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _Facility_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Facility_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("alternativeAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Facility_UUIE, alternativeAddress), _TransportAddress}, + {FNAME("alternativeAliasAddress") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Facility_UUIE_alternativeAliasAddress}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, + {FNAME("reason") CHOICE, 2, 4, 11, DECODE | EXT, + offsetof(Facility_UUIE, reason), _FacilityReason}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") CHOICE, 1, 2, 7, SKIP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("conferences") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Facility_UUIE, h245Address), _TransportAddress}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Facility_UUIE, fastStart), _Facility_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, SKIP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, SKIP | EXT | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT | OPT, 0, NULL}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + NULL}, +}; + +static field_t _CallIdentifier[] = { /* SEQUENCE */ + {FNAME("guid") OCTSTR, FIXD, 16, 0, SKIP, 0, NULL}, +}; + +static field_t _SecurityServiceMode[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, + {FNAME("none") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("default") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _SecurityCapabilities[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("encryption") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, + {FNAME("authenticaton") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, + {FNAME("integrity") CHOICE, 2, 3, 3, SKIP | EXT, 0, + _SecurityServiceMode}, +}; + +static field_t _H245Security[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP, 0, _NonStandardParameter}, + {FNAME("noSecurity") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("tls") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, + {FNAME("ipsec") SEQ, 1, 4, 4, SKIP | EXT, 0, _SecurityCapabilities}, +}; + +static field_t _DHset[] = { /* SEQUENCE */ + {FNAME("halfkey") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("modSize") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, + {FNAME("generator") BITSTR, WORD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _TypedCertificate[] = { /* SEQUENCE */ + {FNAME("type") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("certificate") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _H235_NonStandardParameter[] = { /* SEQUENCE */ + {FNAME("nonStandardIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("data") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _ClearToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("password") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("dhkey") SEQ, 0, 3, 3, SKIP | EXT | OPT, 0, _DHset}, + {FNAME("challenge") OCTSTR, 7, 8, 0, SKIP | OPT, 0, NULL}, + {FNAME("random") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("certificate") SEQ, 0, 2, 2, SKIP | EXT | OPT, 0, + _TypedCertificate}, + {FNAME("generalID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("nonStandard") SEQ, 0, 2, 2, SKIP | OPT, 0, + _H235_NonStandardParameter}, + {FNAME("eckasdhkey") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, NULL}, + {FNAME("sendersID") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _Progress_UUIE_tokens[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, +}; + +static field_t _Params[] = { /* SEQUENCE */ + {FNAME("ranInt") INT, UNCO, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("iv8") OCTSTR, FIXD, 8, 0, SKIP | OPT, 0, NULL}, + {FNAME("iv16") OCTSTR, FIXD, 16, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdHash_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdHash[] = { /* SEQUENCE */ + {FNAME("alias") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdHash_token}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdHash_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdHash[] = { /* SEQUENCE */ + {FNAME("gatekeeperId") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("timeStamp") INT, CONS, 1, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdHash_token}, +}; + +static field_t _CryptoH323Token_cryptoEPPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoEPCert[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoGKCert[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoH323Token_cryptoFastStart[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoEncryptedToken_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoEncryptedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoEncryptedToken_token}, +}; + +static field_t _CryptoToken_cryptoSignedToken_token[] = { /* SEQUENCE */ + {FNAME("toBeSigned") SEQ, 8, 9, 11, SKIP | OPEN | EXT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("signature") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoSignedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("token") SEQ, 0, 4, 4, SKIP, 0, + _CryptoToken_cryptoSignedToken_token}, +}; + +static field_t _CryptoToken_cryptoHashedToken_token[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("hash") BITSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken_cryptoHashedToken[] = { /* SEQUENCE */ + {FNAME("tokenOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("hashedVals") SEQ, 8, 9, 11, SKIP | EXT, 0, _ClearToken}, + {FNAME("token") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoHashedToken_token}, +}; + +static field_t _CryptoToken_cryptoPwdEncr[] = { /* SEQUENCE */ + {FNAME("algorithmOID") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("paramS") SEQ, 2, 2, 3, SKIP | EXT, 0, _Params}, + {FNAME("encryptedData") OCTSTR, SEMI, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _CryptoToken[] = { /* CHOICE */ + {FNAME("cryptoEncryptedToken") SEQ, 0, 2, 2, SKIP, 0, + _CryptoToken_cryptoEncryptedToken}, + {FNAME("cryptoSignedToken") SEQ, 0, 2, 2, SKIP, 0, + _CryptoToken_cryptoSignedToken}, + {FNAME("cryptoHashedToken") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoHashedToken}, + {FNAME("cryptoPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoToken_cryptoPwdEncr}, +}; + +static field_t _CryptoH323Token[] = { /* CHOICE */ + {FNAME("cryptoEPPwdHash") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdHash}, + {FNAME("cryptoGKPwdHash") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdHash}, + {FNAME("cryptoEPPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoEPPwdEncr}, + {FNAME("cryptoGKPwdEncr") SEQ, 0, 3, 3, SKIP, 0, + _CryptoH323Token_cryptoGKPwdEncr}, + {FNAME("cryptoEPCert") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoEPCert}, + {FNAME("cryptoGKCert") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoGKCert}, + {FNAME("cryptoFastStart") SEQ, 0, 4, 4, SKIP, 0, + _CryptoH323Token_cryptoFastStart}, + {FNAME("nestedcryptoToken") CHOICE, 2, 4, 4, SKIP | EXT, 0, + _CryptoToken}, +}; + +static field_t _Progress_UUIE_cryptoTokens[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 8, 8, SKIP | EXT, 0, _CryptoH323Token}, +}; + +static field_t _Progress_UUIE_fastStart[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 1, 3, 5, DECODE | OPEN | EXT, + sizeof(OpenLogicalChannel), _OpenLogicalChannel} + , +}; + +static field_t _Progress_UUIE[] = { /* SEQUENCE */ + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQ, 6, 8, 10, SKIP | EXT, 0, + _EndpointType}, + {FNAME("h245Address") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(Progress_UUIE, h245Address), _TransportAddress}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, SKIP | EXT, 0, + _CallIdentifier}, + {FNAME("h245SecurityMode") CHOICE, 2, 4, 4, SKIP | EXT | OPT, 0, + _H245Security}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Progress_UUIE_tokens}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _Progress_UUIE_cryptoTokens}, + {FNAME("fastStart") SEQOF, SEMI, 0, 30, DECODE | OPT, + offsetof(Progress_UUIE, fastStart), _Progress_UUIE_fastStart}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("fastConnectRefused") NUL, FIXD, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H323_UU_PDU_h323_message_body[] = { /* CHOICE */ + {FNAME("setup") SEQ, 7, 13, 39, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, setup), _Setup_UUIE}, + {FNAME("callProceeding") SEQ, 1, 3, 12, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, callProceeding), + _CallProceeding_UUIE}, + {FNAME("connect") SEQ, 1, 4, 19, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, connect), _Connect_UUIE}, + {FNAME("alerting") SEQ, 1, 3, 17, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, alerting), _Alerting_UUIE}, + {FNAME("information") SEQ, 0, 1, 7, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, information), + _Information_UUIE}, + {FNAME("releaseComplete") SEQ, 1, 2, 11, SKIP | EXT, 0, + _ReleaseComplete_UUIE}, + {FNAME("facility") SEQ, 3, 5, 21, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, facility), _Facility_UUIE}, + {FNAME("progress") SEQ, 5, 8, 11, DECODE | EXT, + offsetof(H323_UU_PDU_h323_message_body, progress), _Progress_UUIE}, + {FNAME("empty") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("status") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("statusInquiry") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("setupAcknowledge") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, + {FNAME("notify") SEQ, 2, 4, 4, SKIP | EXT, 0, NULL}, +}; + +static field_t _RequestMessage[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("masterSlaveDetermination") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("terminalCapabilitySet") SEQ, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("openLogicalChannel") SEQ, 1, 3, 5, DECODE | EXT, + offsetof(RequestMessage, openLogicalChannel), _OpenLogicalChannel}, + {FNAME("closeLogicalChannel") SEQ, 0, 2, 3, STOP | EXT, 0, NULL}, + {FNAME("requestChannelClose") SEQ, 0, 1, 3, STOP | EXT, 0, NULL}, + {FNAME("multiplexEntrySend") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntry") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestMode") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("roundTripDelayRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopRequest") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("communicationModeRequest") SEQ, 0, 0, 0, STOP | EXT, 0, NULL}, + {FNAME("conferenceRequest") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, + {FNAME("multilinkRequest") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("logicalChannelRateRequest") SEQ, 0, 3, 3, STOP | EXT, 0, + NULL}, +}; + +static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters[] = { /* CHOICE */ + {FNAME("h222LogicalChannelParameters") SEQ, 3, 5, 5, SKIP | EXT, 0, + _H222LogicalChannelParameters}, + {FNAME("h2250LogicalChannelParameters") SEQ, 10, 11, 14, DECODE | EXT, + offsetof + (OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters, + h2250LogicalChannelParameters), _H2250LogicalChannelParameters}, +}; + +static field_t _OpenLogicalChannelAck_reverseLogicalChannelParameters[] = { /* SEQUENCE */ + {FNAME("reverseLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("multiplexParameters") CHOICE, 0, 1, 2, DECODE | EXT | OPT, + offsetof(OpenLogicalChannelAck_reverseLogicalChannelParameters, + multiplexParameters), + _OpenLogicalChannelAck_reverseLogicalChannelParameters_multiplexParameters}, + {FNAME("replacementFor") INT, WORD, 1, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _H2250LogicalChannelAckParameters_nonStandard[] = { /* SEQUENCE OF */ + {FNAME("item") SEQ, 0, 2, 2, SKIP, 0, _H245_NonStandardParameter}, +}; + +static field_t _H2250LogicalChannelAckParameters[] = { /* SEQUENCE */ + {FNAME("nonStandard") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _H2250LogicalChannelAckParameters_nonStandard}, + {FNAME("sessionID") INT, 8, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("mediaChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelAckParameters, mediaChannel), + _H245_TransportAddress}, + {FNAME("mediaControlChannel") CHOICE, 1, 2, 2, DECODE | EXT | OPT, + offsetof(H2250LogicalChannelAckParameters, mediaControlChannel), + _H245_TransportAddress}, + {FNAME("dynamicRTPPayloadType") INT, 5, 96, 0, SKIP | OPT, 0, NULL}, + {FNAME("flowControlToZero") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("portNumber") INT, WORD, 0, 0, SKIP | OPT, 0, NULL}, +}; + +static field_t _OpenLogicalChannelAck_forwardMultiplexAckParameters[] = { /* CHOICE */ + {FNAME("h2250LogicalChannelAckParameters") SEQ, 5, 5, 7, DECODE | EXT, + offsetof(OpenLogicalChannelAck_forwardMultiplexAckParameters, + h2250LogicalChannelAckParameters), + _H2250LogicalChannelAckParameters}, +}; + +static field_t _OpenLogicalChannelAck[] = { /* SEQUENCE */ + {FNAME("forwardLogicalChannelNumber") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("reverseLogicalChannelParameters") SEQ, 2, 3, 4, + DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, + reverseLogicalChannelParameters), + _OpenLogicalChannelAck_reverseLogicalChannelParameters}, + {FNAME("separateStack") SEQ, 2, 4, 5, SKIP | EXT | OPT, 0, NULL}, + {FNAME("forwardMultiplexAckParameters") CHOICE, 0, 1, 1, + DECODE | EXT | OPT, offsetof(OpenLogicalChannelAck, + forwardMultiplexAckParameters), + _OpenLogicalChannelAck_forwardMultiplexAckParameters}, + {FNAME("encryptionSync") SEQ, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _ResponseMessage[] = { /* CHOICE */ + {FNAME("nonStandard") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("masterSlaveDeterminationAck") SEQ, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("masterSlaveDeterminationReject") SEQ, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("terminalCapabilitySetAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("terminalCapabilitySetReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("openLogicalChannelAck") SEQ, 1, 2, 5, DECODE | EXT, + offsetof(ResponseMessage, openLogicalChannelAck), + _OpenLogicalChannelAck}, + {FNAME("openLogicalChannelReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("closeLogicalChannelAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestChannelCloseAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestChannelCloseReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("multiplexEntrySendAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("multiplexEntrySendReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntryAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("requestMultiplexEntryReject") SEQ, 0, 2, 2, STOP | EXT, 0, + NULL}, + {FNAME("requestModeAck") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("requestModeReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("roundTripDelayResponse") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopAck") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("maintenanceLoopReject") SEQ, 0, 2, 2, STOP | EXT, 0, NULL}, + {FNAME("communicationModeResponse") CHOICE, 0, 1, 1, STOP | EXT, 0, + NULL}, + {FNAME("conferenceResponse") CHOICE, 3, 8, 16, STOP | EXT, 0, NULL}, + {FNAME("multilinkResponse") CHOICE, 3, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("logicalChannelRateAcknowledge") SEQ, 0, 3, 3, STOP | EXT, 0, + NULL}, + {FNAME("logicalChannelRateReject") SEQ, 1, 4, 4, STOP | EXT, 0, NULL}, +}; + +static field_t _MultimediaSystemControlMessage[] = { /* CHOICE */ + {FNAME("request") CHOICE, 4, 11, 15, DECODE | EXT, + offsetof(MultimediaSystemControlMessage, request), _RequestMessage}, + {FNAME("response") CHOICE, 5, 19, 24, DECODE | EXT, + offsetof(MultimediaSystemControlMessage, response), + _ResponseMessage}, + {FNAME("command") CHOICE, 3, 7, 12, STOP | EXT, 0, NULL}, + {FNAME("indication") CHOICE, 4, 14, 23, STOP | EXT, 0, NULL}, +}; + +static field_t _H323_UU_PDU_h245Control[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 2, 4, 4, DECODE | OPEN | EXT, + sizeof(MultimediaSystemControlMessage), + _MultimediaSystemControlMessage} + , +}; + +static field_t _H323_UU_PDU[] = { /* SEQUENCE */ + {FNAME("h323-message-body") CHOICE, 3, 7, 13, DECODE | EXT, + offsetof(H323_UU_PDU, h323_message_body), + _H323_UU_PDU_h323_message_body}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("h4501SupplementaryService") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + NULL}, + {FNAME("h245Tunneling") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("h245Control") SEQOF, SEMI, 0, 4, DECODE | OPT, + offsetof(H323_UU_PDU, h245Control), _H323_UU_PDU_h245Control}, + {FNAME("nonStandardControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("tunnelledSignallingMessage") SEQ, 2, 4, 4, STOP | EXT | OPT, + 0, NULL}, + {FNAME("provisionalRespToH245Tunneling") NUL, FIXD, 0, 0, STOP | OPT, + 0, NULL}, + {FNAME("stimulusControl") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _H323_UserInformation[] = { /* SEQUENCE */ + {FNAME("h323-uu-pdu") SEQ, 1, 2, 11, DECODE | EXT, + offsetof(H323_UserInformation, h323_uu_pdu), _H323_UU_PDU}, + {FNAME("user-data") SEQ, 0, 2, 2, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _GatekeeperRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(GatekeeperRequest, rasAddress), _TransportAddress}, + {FNAME("endpointType") SEQ, 6, 8, 10, STOP | EXT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("authenticationCapability") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("algorithmOIDs") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _GatekeeperConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(GatekeeperConfirm, rasAddress), _TransportAddress}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("authenticationMode") CHOICE, 3, 7, 8, STOP | EXT | OPT, 0, + NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("algorithmOID") OID, BYTE, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrity") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RegistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationRequest_rasAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationRequest_terminalAlias[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _RegistrationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("discoveryComplete") BOOL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationRequest, callSignalAddress), + _RegistrationRequest_callSignalAddress}, + {FNAME("rasAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationRequest, rasAddress), + _RegistrationRequest_rasAddress}, + {FNAME("terminalType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _RegistrationRequest_terminalAlias}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("endpointVendor") SEQ, 2, 3, 3, SKIP | EXT, 0, + _VendorIdentifier}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, + offsetof(RegistrationRequest, timeToLive), NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("keepAlive") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("additiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportsAltGK") NUL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("usageReportingCapability") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, + NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportedH248Packages") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("callCreditCapability") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("capacityReportingCapability") SEQ, 0, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RegistrationConfirm_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _RegistrationConfirm_terminalAlias[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _RegistrationConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("protocolIdentifier") OID, BYTE, 0, 0, SKIP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(RegistrationConfirm, callSignalAddress), + _RegistrationConfirm_callSignalAddress}, + {FNAME("terminalAlias") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _RegistrationConfirm_terminalAlias}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, NULL}, + {FNAME("timeToLive") INT, CONS, 1, 0, DECODE | OPT, + offsetof(RegistrationConfirm, timeToLive), NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("preGrantedARQ") SEQ, 0, 4, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("maintainConnection") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportsAdditiveRegistration") NUL, FIXD, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("terminalAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureServerAlias") CHOICE, 1, 2, 7, STOP | EXT | OPT, 0, + NULL}, + {FNAME("capacityReportingSpec") SEQ, 0, 1, 1, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _UnregistrationRequest_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _UnregistrationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(UnregistrationRequest, callSignalAddress), + _UnregistrationRequest_callSignalAddress}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("reason") CHOICE, 2, 4, 5, STOP | EXT | OPT, 0, NULL}, + {FNAME("endpointAliasPattern") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("supportedPrefixes") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateGatekeeper") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _CallModel[] = { /* CHOICE */ + {FNAME("direct") NUL, FIXD, 0, 0, SKIP, 0, NULL}, + {FNAME("gatekeeperRouted") NUL, FIXD, 0, 0, SKIP, 0, NULL}, +}; + +static field_t _AdmissionRequest_destinationInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest_destExtraCallInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest_srcInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _AdmissionRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callType") CHOICE, 2, 4, 4, SKIP | EXT, 0, _CallType}, + {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT | OPT, 0, _CallModel}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _AdmissionRequest_destinationInfo}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(AdmissionRequest, destCallSignalAddress), + _TransportAddress}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, SKIP | OPT, 0, + _AdmissionRequest_destExtraCallInfo}, + {FNAME("srcInfo") SEQOF, SEMI, 0, 0, SKIP, 0, + _AdmissionRequest_srcInfo}, + {FNAME("srcCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT | OPT, + offsetof(AdmissionRequest, srcCallSignalAddress), _TransportAddress}, + {FNAME("bandWidth") INT, CONS, 0, 0, STOP, 0, NULL}, + {FNAME("callReferenceValue") INT, WORD, 0, 0, STOP, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("callServices") SEQ, 0, 8, 8, STOP | EXT | OPT, 0, NULL}, + {FNAME("conferenceID") OCTSTR, FIXD, 16, 0, STOP, 0, NULL}, + {FNAME("activeMC") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("answerCall") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("callIdentifier") SEQ, 0, 1, 1, STOP | EXT, 0, NULL}, + {FNAME("srcAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destAlternatives") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("willSupplyUUIEs") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("callLinkage") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("gatewayDataRate") SEQ, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _AdmissionConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("bandWidth") INT, CONS, 0, 0, SKIP, 0, NULL}, + {FNAME("callModel") CHOICE, 1, 2, 2, SKIP | EXT, 0, _CallModel}, + {FNAME("destCallSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(AdmissionConfirm, destCallSignalAddress), + _TransportAddress}, + {FNAME("irrFrequency") INT, WORD, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("transportQOS") CHOICE, 2, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("willRespondToIRR") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("uuiesRequested") SEQ, 0, 9, 13, STOP | EXT, 0, NULL}, + {FNAME("language") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("useSpecifiedTransport") CHOICE, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("usageSpec") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _LocationRequest_destinationInfo[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 1, 2, 7, SKIP | EXT, 0, _AliasAddress}, +}; + +static field_t _LocationRequest[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, SKIP, 0, + _LocationRequest_destinationInfo}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("replyAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationRequest, replyAddress), _TransportAddress}, + {FNAME("sourceInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("canMapAlias") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("gatekeeperIdentifier") BMPSTR, 7, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("desiredProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("desiredTunnelledProtocol") SEQ, 1, 2, 2, STOP | EXT | OPT, 0, + NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("hopCount") INT, 8, 1, 0, STOP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, +}; + +static field_t _LocationConfirm[] = { /* SEQUENCE */ + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("callSignalAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationConfirm, callSignalAddress), _TransportAddress}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(LocationConfirm, rasAddress), _TransportAddress}, + {FNAME("nonStandardData") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("destinationInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destExtraCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("destinationType") SEQ, 6, 8, 10, STOP | EXT | OPT, 0, NULL}, + {FNAME("remoteExtensionAddress") SEQOF, SEMI, 0, 0, STOP | OPT, 0, + NULL}, + {FNAME("alternateEndpoints") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("alternateTransportAddresses") SEQ, 1, 1, 1, STOP | EXT | OPT, + 0, NULL}, + {FNAME("supportedProtocols") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("multipleCalls") BOOL, FIXD, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("featureSet") SEQ, 3, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("circuitInfo") SEQ, 3, 3, 3, STOP | EXT | OPT, 0, NULL}, + {FNAME("serviceControl") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _InfoRequestResponse_callSignalAddress[] = { /* SEQUENCE OF */ + {FNAME("item") CHOICE, 3, 7, 7, DECODE | EXT, + sizeof(TransportAddress), _TransportAddress} + , +}; + +static field_t _InfoRequestResponse[] = { /* SEQUENCE */ + {FNAME("nonStandardData") SEQ, 0, 2, 2, SKIP | OPT, 0, + _NonStandardParameter}, + {FNAME("requestSeqNum") INT, WORD, 1, 0, SKIP, 0, NULL}, + {FNAME("endpointType") SEQ, 6, 8, 10, SKIP | EXT, 0, _EndpointType}, + {FNAME("endpointIdentifier") BMPSTR, 7, 1, 0, SKIP, 0, NULL}, + {FNAME("rasAddress") CHOICE, 3, 7, 7, DECODE | EXT, + offsetof(InfoRequestResponse, rasAddress), _TransportAddress}, + {FNAME("callSignalAddress") SEQOF, SEMI, 0, 10, DECODE, + offsetof(InfoRequestResponse, callSignalAddress), + _InfoRequestResponse_callSignalAddress}, + {FNAME("endpointAlias") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("perCallInfo") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("tokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("cryptoTokens") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, + {FNAME("integrityCheckValue") SEQ, 0, 2, 2, STOP | OPT, 0, NULL}, + {FNAME("needResponse") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("capacity") SEQ, 2, 2, 2, STOP | EXT | OPT, 0, NULL}, + {FNAME("irrStatus") CHOICE, 2, 4, 4, STOP | EXT | OPT, 0, NULL}, + {FNAME("unsolicited") BOOL, FIXD, 0, 0, STOP, 0, NULL}, + {FNAME("genericData") SEQOF, SEMI, 0, 0, STOP | OPT, 0, NULL}, +}; + +static field_t _RasMessage[] = { /* CHOICE */ + {FNAME("gatekeeperRequest") SEQ, 4, 8, 18, DECODE | EXT, + offsetof(RasMessage, gatekeeperRequest), _GatekeeperRequest}, + {FNAME("gatekeeperConfirm") SEQ, 2, 5, 14, DECODE | EXT, + offsetof(RasMessage, gatekeeperConfirm), _GatekeeperConfirm}, + {FNAME("gatekeeperReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, + {FNAME("registrationRequest") SEQ, 3, 10, 31, DECODE | EXT, + offsetof(RasMessage, registrationRequest), _RegistrationRequest}, + {FNAME("registrationConfirm") SEQ, 3, 7, 24, DECODE | EXT, + offsetof(RasMessage, registrationConfirm), _RegistrationConfirm}, + {FNAME("registrationReject") SEQ, 2, 5, 11, STOP | EXT, 0, NULL}, + {FNAME("unregistrationRequest") SEQ, 3, 5, 15, DECODE | EXT, + offsetof(RasMessage, unregistrationRequest), _UnregistrationRequest}, + {FNAME("unregistrationConfirm") SEQ, 1, 2, 6, STOP | EXT, 0, NULL}, + {FNAME("unregistrationReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("admissionRequest") SEQ, 7, 16, 34, DECODE | EXT, + offsetof(RasMessage, admissionRequest), _AdmissionRequest}, + {FNAME("admissionConfirm") SEQ, 2, 6, 27, DECODE | EXT, + offsetof(RasMessage, admissionConfirm), _AdmissionConfirm}, + {FNAME("admissionReject") SEQ, 1, 3, 11, STOP | EXT, 0, NULL}, + {FNAME("bandwidthRequest") SEQ, 2, 7, 18, STOP | EXT, 0, NULL}, + {FNAME("bandwidthConfirm") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("bandwidthReject") SEQ, 1, 4, 9, STOP | EXT, 0, NULL}, + {FNAME("disengageRequest") SEQ, 1, 6, 19, STOP | EXT, 0, NULL}, + {FNAME("disengageConfirm") SEQ, 1, 2, 9, STOP | EXT, 0, NULL}, + {FNAME("disengageReject") SEQ, 1, 3, 8, STOP | EXT, 0, NULL}, + {FNAME("locationRequest") SEQ, 2, 5, 17, DECODE | EXT, + offsetof(RasMessage, locationRequest), _LocationRequest}, + {FNAME("locationConfirm") SEQ, 1, 4, 19, DECODE | EXT, + offsetof(RasMessage, locationConfirm), _LocationConfirm}, + {FNAME("locationReject") SEQ, 1, 3, 10, STOP | EXT, 0, NULL}, + {FNAME("infoRequest") SEQ, 2, 4, 15, STOP | EXT, 0, NULL}, + {FNAME("infoRequestResponse") SEQ, 3, 8, 16, DECODE | EXT, + offsetof(RasMessage, infoRequestResponse), _InfoRequestResponse}, + {FNAME("nonStandardMessage") SEQ, 0, 2, 7, STOP | EXT, 0, NULL}, + {FNAME("unknownMessageResponse") SEQ, 0, 1, 5, STOP | EXT, 0, NULL}, + {FNAME("requestInProgress") SEQ, 4, 6, 6, STOP | EXT, 0, NULL}, + {FNAME("resourcesAvailableIndicate") SEQ, 4, 9, 11, STOP | EXT, 0, + NULL}, + {FNAME("resourcesAvailableConfirm") SEQ, 4, 6, 7, STOP | EXT, 0, + NULL}, + {FNAME("infoRequestAck") SEQ, 4, 5, 5, STOP | EXT, 0, NULL}, + {FNAME("infoRequestNak") SEQ, 5, 7, 7, STOP | EXT, 0, NULL}, + {FNAME("serviceControlIndication") SEQ, 8, 10, 10, STOP | EXT, 0, + NULL}, + {FNAME("serviceControlResponse") SEQ, 7, 8, 8, STOP | EXT, 0, NULL}, +}; -- cgit v1.2.3 From 869f37d8e48f3911eb70f38a994feaa8f8380008 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:06 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add IRC helper port Add nf_conntrack port of the IRC conntrack/NAT helper. Since DCC doesn't support IPv6 yet, the helper is still IPv4 only. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_irc.h | 15 ++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_irc.c | 101 +++++++++++ net/netfilter/Kconfig | 15 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_irc.c | 281 +++++++++++++++++++++++++++++ 7 files changed, 419 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_irc.h create mode 100644 net/ipv4/netfilter/nf_nat_irc.c create mode 100644 net/netfilter/nf_conntrack_irc.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_irc.h b/include/linux/netfilter/nf_conntrack_irc.h new file mode 100644 index 000000000000..2ab6b8255911 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_irc.h @@ -0,0 +1,15 @@ +#ifndef _NF_CONNTRACK_IRC_H +#define _NF_CONNTRACK_IRC_H + +#ifdef __KERNEL__ + +#define IRC_PORT 6667 + +extern unsigned int (*nf_nat_irc_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp); + +#endif /* __KERNEL__ */ +#endif /* _NF_CONNTRACK_IRC_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index e14156d1122e..4555f721dfc1 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -500,6 +500,11 @@ config IP_NF_NAT_IRC default IP_NF_NAT if IP_NF_IRC=y default m if IP_NF_IRC=m +config NF_NAT_IRC + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_IRC + config IP_NF_NAT_TFTP tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index bdaba4700e3b..56733c370327 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -53,6 +53,7 @@ obj-$(CONFIG_IP_NF_NAT_SIP) += ip_nat_sip.o obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o +obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_irc.c b/net/ipv4/netfilter/nf_nat_irc.c new file mode 100644 index 000000000000..9b8c0daea744 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_irc.c @@ -0,0 +1,101 @@ +/* IRC extension for TCP NAT alteration. + * + * (C) 2000-2001 by Harald Welte + * (C) 2004 Rusty Russell IBM Corporation + * based on a copy of RR's ip_nat_ftp.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("IRC (DCC) NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_irc"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) +{ + char buffer[sizeof("4294967296 65635")]; + u_int32_t ip; + u_int16_t port; + unsigned int ret; + + DEBUGP("IRC_NAT: info (seq %u + %u) in %u\n", + expect->seq, exp_irc_info->len, ntohl(tcph->seq)); + + /* Reply comes from server. */ + exp->saved_proto.tcp.port = exp->tuple.dst.u.tcp.port; + exp->dir = IP_CT_DIR_REPLY; + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.tcp.port); port != 0; port++) { + exp->tuple.dst.u.tcp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + ip = ntohl(exp->master->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip); + sprintf(buffer, "%u %u", ip, port); + DEBUGP("nf_nat_irc: inserting '%s' == %u.%u.%u.%u, port %u\n", + buffer, NIPQUAD(ip), port); + + ret = nf_nat_mangle_tcp_packet(pskb, exp->master, ctinfo, + matchoff, matchlen, buffer, + strlen(buffer)); + if (ret != NF_ACCEPT) + nf_conntrack_unexpect_related(exp); + return ret; +} + +static void __exit nf_nat_irc_fini(void) +{ + rcu_assign_pointer(nf_nat_irc_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_irc_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_irc_hook)); + rcu_assign_pointer(nf_nat_irc_hook, help); + return 0; +} + +/* Prior to 2.6.11, we had a ports param. No longer, but don't break users. */ +static int warn_set(const char *val, struct kernel_param *kp) +{ + printk(KERN_INFO KBUILD_MODNAME + ": kernel >= 2.6.10 only uses 'ports' for conntrack modules\n"); + return 0; +} +module_param_call(ports, warn_set, NULL, NULL, 0); + +module_init(nf_nat_irc_init); +module_exit(nf_nat_irc_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index d8f3451c95b6..bd50897d8fbb 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -179,6 +179,21 @@ config NF_CONNTRACK_H323 To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_IRC + tristate "IRC protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + There is a commonly-used extension to IRC called + Direct Client-to-Client Protocol (DCC). This enables users to send + files to each other, and also chat to each other without the need + of a server. DCC Sending is used anywhere you send files over IRC, + and DCC Chat is most commonly used by Eggdrop bots. If you are + using NAT, this extension will enable you to send files and initiate + chats. Note that you do NOT need this extension to get files or + have others initiate chats, or everything else in IRC. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 69b554576b6b..343fd4896406 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -25,6 +25,7 @@ nf_conntrack_h323-objs := nf_conntrack_h323_main.o nf_conntrack_h323_asn1.o obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o +obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c new file mode 100644 index 000000000000..ed01db634399 --- /dev/null +++ b/net/netfilter/nf_conntrack_irc.c @@ -0,0 +1,281 @@ +/* IRC extension for IP connection tracking, Version 1.21 + * (C) 2000-2002 by Harald Welte + * based on RR's ip_conntrack_ftp.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +static unsigned int max_dcc_channels = 8; +static unsigned int dcc_timeout __read_mostly = 300; +/* This is slow, but it's simple. --RR */ +static char *irc_buffer; +static DEFINE_SPINLOCK(irc_buffer_lock); + +unsigned int (*nf_nat_irc_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_irc_hook); + +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("IRC (DCC) connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_irc"); + +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "port numbers of IRC servers"); +module_param(max_dcc_channels, uint, 0400); +MODULE_PARM_DESC(max_dcc_channels, "max number of expected DCC channels per " + "IRC session"); +module_param(dcc_timeout, uint, 0400); +MODULE_PARM_DESC(dcc_timeout, "timeout on for unestablished DCC channels"); + +static const char *dccprotos[] = { + "SEND ", "CHAT ", "MOVE ", "TSEND ", "SCHAT " +}; + +#define MINMATCHLEN 5 + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s:" format, \ + __FILE__, __FUNCTION__ , ## args) +#else +#define DEBUGP(format, args...) +#endif + +/* tries to get the ip_addr and port out of a dcc command + * return value: -1 on failure, 0 on success + * data pointer to first byte of DCC command data + * data_end pointer to last byte of dcc command data + * ip returns parsed ip of dcc command + * port returns parsed port of dcc command + * ad_beg_p returns pointer to first byte of addr data + * ad_end_p returns pointer to last byte of addr data + */ +static int parse_dcc(char *data, char *data_end, u_int32_t *ip, + u_int16_t *port, char **ad_beg_p, char **ad_end_p) +{ + /* at least 12: "AAAAAAAA P\1\n" */ + while (*data++ != ' ') + if (data > data_end - 12) + return -1; + + *ad_beg_p = data; + *ip = simple_strtoul(data, &data, 10); + + /* skip blanks between ip and port */ + while (*data == ' ') { + if (data >= data_end) + return -1; + data++; + } + + *port = simple_strtoul(data, &data, 10); + *ad_end_p = data; + + return 0; +} + +static int help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + unsigned int dataoff; + struct tcphdr _tcph, *th; + char *data, *data_limit, *ib_ptr; + int dir = CTINFO2DIR(ctinfo); + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + u_int32_t dcc_ip; + u_int16_t dcc_port; + __be16 port; + int i, ret = NF_ACCEPT; + char *addr_beg_p, *addr_end_p; + typeof(nf_nat_irc_hook) nf_nat_irc; + + /* If packet is coming from IRC server */ + if (dir == IP_CT_DIR_REPLY) + return NF_ACCEPT; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) + return NF_ACCEPT; + + /* Not a full tcp header? */ + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return NF_ACCEPT; + + /* No data? */ + dataoff = protoff + th->doff*4; + if (dataoff >= (*pskb)->len) + return NF_ACCEPT; + + spin_lock_bh(&irc_buffer_lock); + ib_ptr = skb_header_pointer(*pskb, dataoff, (*pskb)->len - dataoff, + irc_buffer); + BUG_ON(ib_ptr == NULL); + + data = ib_ptr; + data_limit = ib_ptr + (*pskb)->len - dataoff; + + /* strlen("\1DCC SENT t AAAAAAAA P\1\n")=24 + * 5+MINMATCHLEN+strlen("t AAAAAAAA P\1\n")=14 */ + while (data < data_limit - (19 + MINMATCHLEN)) { + if (memcmp(data, "\1DCC ", 5)) { + data++; + continue; + } + data += 5; + /* we have at least (19+MINMATCHLEN)-5 bytes valid data left */ + + DEBUGP("DCC found in master %u.%u.%u.%u:%u %u.%u.%u.%u:%u...\n", + NIPQUAD(iph->saddr), ntohs(th->source), + NIPQUAD(iph->daddr), ntohs(th->dest)); + + for (i = 0; i < ARRAY_SIZE(dccprotos); i++) { + if (memcmp(data, dccprotos[i], strlen(dccprotos[i]))) { + /* no match */ + continue; + } + data += strlen(dccprotos[i]); + DEBUGP("DCC %s detected\n", dccprotos[i]); + + /* we have at least + * (19+MINMATCHLEN)-5-dccprotos[i].matchlen bytes valid + * data left (== 14/13 bytes) */ + if (parse_dcc((char *)data, data_limit, &dcc_ip, + &dcc_port, &addr_beg_p, &addr_end_p)) { + DEBUGP("unable to parse dcc command\n"); + continue; + } + DEBUGP("DCC bound ip/port: %u.%u.%u.%u:%u\n", + HIPQUAD(dcc_ip), dcc_port); + + /* dcc_ip can be the internal OR external (NAT'ed) IP */ + tuple = &ct->tuplehash[dir].tuple; + if (tuple->src.u3.ip != htonl(dcc_ip) && + tuple->dst.u3.ip != htonl(dcc_ip)) { + if (net_ratelimit()) + printk(KERN_WARNING + "Forged DCC command from " + "%u.%u.%u.%u: %u.%u.%u.%u:%u\n", + NIPQUAD(tuple->src.u3.ip), + HIPQUAD(dcc_ip), dcc_port); + continue; + } + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + tuple = &ct->tuplehash[!dir].tuple; + port = htons(dcc_port); + nf_conntrack_expect_init(exp, tuple->src.l3num, + NULL, &tuple->dst.u3, + IPPROTO_TCP, NULL, &port); + + nf_nat_irc = rcu_dereference(nf_nat_irc_hook); + if (nf_nat_irc && ct->status & IPS_NAT_MASK) + ret = nf_nat_irc(pskb, ctinfo, + addr_beg_p - ib_ptr, + addr_end_p - addr_beg_p, + exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + goto out; + } + } + out: + spin_unlock_bh(&irc_buffer_lock); + return ret; +} + +static struct nf_conntrack_helper irc[MAX_PORTS] __read_mostly; +static char irc_names[MAX_PORTS][sizeof("irc-65535")] __read_mostly; + +static void nf_conntrack_irc_fini(void); + +static int __init nf_conntrack_irc_init(void) +{ + int i, ret; + char *tmpname; + + if (max_dcc_channels < 1) { + printk("nf_ct_irc: max_dcc_channels must not be zero\n"); + return -EINVAL; + } + + irc_buffer = kmalloc(65536, GFP_KERNEL); + if (!irc_buffer) + return -ENOMEM; + + /* If no port given, default to standard irc port */ + if (ports_c == 0) + ports[ports_c++] = IRC_PORT; + + for (i = 0; i < ports_c; i++) { + irc[i].tuple.src.l3num = AF_INET; + irc[i].tuple.src.u.tcp.port = htons(ports[i]); + irc[i].tuple.dst.protonum = IPPROTO_TCP; + irc[i].mask.src.l3num = 0xFFFF; + irc[i].mask.src.u.tcp.port = htons(0xFFFF); + irc[i].mask.dst.protonum = 0xFF; + irc[i].max_expected = max_dcc_channels; + irc[i].timeout = dcc_timeout; + irc[i].me = THIS_MODULE; + irc[i].help = help; + + tmpname = &irc_names[i][0]; + if (ports[i] == IRC_PORT) + sprintf(tmpname, "irc"); + else + sprintf(tmpname, "irc-%u", i); + irc[i].name = tmpname; + + ret = nf_conntrack_helper_register(&irc[i]); + if (ret) { + printk("nf_ct_irc: failed to register helper " + "for pf: %u port: %u\n", + irc[i].tuple.src.l3num, ports[i]); + nf_conntrack_irc_fini(); + return ret; + } + } + return 0; +} + +/* This function is intentionally _NOT_ defined as __exit, because + * it is needed by the init function */ +static void nf_conntrack_irc_fini(void) +{ + int i; + + for (i = 0; i < ports_c; i++) + nf_conntrack_helper_unregister(&irc[i]); + kfree(irc_buffer); +} + +module_init(nf_conntrack_irc_init); +module_exit(nf_conntrack_irc_fini); -- cgit v1.2.3 From f09943fefe6b702e40893d35b4f10fd1064037fe Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:41 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add PPTP helper port Add nf_conntrack port of the PPtP conntrack/NAT helper. Since there seems to be no IPv6-capable PPtP implementation the helper only support IPv4. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_pptp.h | 321 ++++++++++++ include/linux/netfilter/nf_conntrack_proto_gre.h | 112 +++++ include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 2 + include/net/netfilter/nf_conntrack.h | 4 + include/net/netfilter/nf_conntrack_helper.h | 2 + include/net/netfilter/nf_conntrack_tuple.h | 6 + net/ipv4/netfilter/Kconfig | 10 + net/ipv4/netfilter/Makefile | 4 + net/ipv4/netfilter/nf_nat_pptp.c | 315 ++++++++++++ net/ipv4/netfilter/nf_nat_proto_gre.c | 179 +++++++ net/netfilter/Kconfig | 23 + net/netfilter/Makefile | 2 + net/netfilter/nf_conntrack_core.c | 4 + net/netfilter/nf_conntrack_pptp.c | 607 +++++++++++++++++++++++ net/netfilter/nf_conntrack_proto_gre.c | 305 ++++++++++++ 15 files changed, 1896 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_pptp.h create mode 100644 include/linux/netfilter/nf_conntrack_proto_gre.h create mode 100644 net/ipv4/netfilter/nf_nat_pptp.c create mode 100644 net/ipv4/netfilter/nf_nat_proto_gre.c create mode 100644 net/netfilter/nf_conntrack_pptp.c create mode 100644 net/netfilter/nf_conntrack_proto_gre.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_pptp.h b/include/linux/netfilter/nf_conntrack_pptp.h new file mode 100644 index 000000000000..fb049ec11ff2 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_pptp.h @@ -0,0 +1,321 @@ +/* PPTP constants and structs */ +#ifndef _NF_CONNTRACK_PPTP_H +#define _NF_CONNTRACK_PPTP_H + +/* state of the control session */ +enum pptp_ctrlsess_state { + PPTP_SESSION_NONE, /* no session present */ + PPTP_SESSION_ERROR, /* some session error */ + PPTP_SESSION_STOPREQ, /* stop_sess request seen */ + PPTP_SESSION_REQUESTED, /* start_sess request seen */ + PPTP_SESSION_CONFIRMED, /* session established */ +}; + +/* state of the call inside the control session */ +enum pptp_ctrlcall_state { + PPTP_CALL_NONE, + PPTP_CALL_ERROR, + PPTP_CALL_OUT_REQ, + PPTP_CALL_OUT_CONF, + PPTP_CALL_IN_REQ, + PPTP_CALL_IN_REP, + PPTP_CALL_IN_CONF, + PPTP_CALL_CLEAR_REQ, +}; + +/* conntrack private data */ +struct nf_ct_pptp_master { + enum pptp_ctrlsess_state sstate; /* session state */ + enum pptp_ctrlcall_state cstate; /* call state */ + __be16 pac_call_id; /* call id of PAC */ + __be16 pns_call_id; /* call id of PNS */ + + /* in pre-2.6.11 this used to be per-expect. Now it is per-conntrack + * and therefore imposes a fixed limit on the number of maps */ + struct nf_ct_gre_keymap *keymap[IP_CT_DIR_MAX]; +}; + +struct nf_nat_pptp { + __be16 pns_call_id; /* NAT'ed PNS call id */ + __be16 pac_call_id; /* NAT'ed PAC call id */ +}; + +#ifdef __KERNEL__ + +#define PPTP_CONTROL_PORT 1723 + +#define PPTP_PACKET_CONTROL 1 +#define PPTP_PACKET_MGMT 2 + +#define PPTP_MAGIC_COOKIE 0x1a2b3c4d + +struct pptp_pkt_hdr { + __u16 packetLength; + __be16 packetType; + __be32 magicCookie; +}; + +/* PptpControlMessageType values */ +#define PPTP_START_SESSION_REQUEST 1 +#define PPTP_START_SESSION_REPLY 2 +#define PPTP_STOP_SESSION_REQUEST 3 +#define PPTP_STOP_SESSION_REPLY 4 +#define PPTP_ECHO_REQUEST 5 +#define PPTP_ECHO_REPLY 6 +#define PPTP_OUT_CALL_REQUEST 7 +#define PPTP_OUT_CALL_REPLY 8 +#define PPTP_IN_CALL_REQUEST 9 +#define PPTP_IN_CALL_REPLY 10 +#define PPTP_IN_CALL_CONNECT 11 +#define PPTP_CALL_CLEAR_REQUEST 12 +#define PPTP_CALL_DISCONNECT_NOTIFY 13 +#define PPTP_WAN_ERROR_NOTIFY 14 +#define PPTP_SET_LINK_INFO 15 + +#define PPTP_MSG_MAX 15 + +/* PptpGeneralError values */ +#define PPTP_ERROR_CODE_NONE 0 +#define PPTP_NOT_CONNECTED 1 +#define PPTP_BAD_FORMAT 2 +#define PPTP_BAD_VALUE 3 +#define PPTP_NO_RESOURCE 4 +#define PPTP_BAD_CALLID 5 +#define PPTP_REMOVE_DEVICE_ERROR 6 + +struct PptpControlHeader { + __be16 messageType; + __u16 reserved; +}; + +/* FramingCapability Bitmap Values */ +#define PPTP_FRAME_CAP_ASYNC 0x1 +#define PPTP_FRAME_CAP_SYNC 0x2 + +/* BearerCapability Bitmap Values */ +#define PPTP_BEARER_CAP_ANALOG 0x1 +#define PPTP_BEARER_CAP_DIGITAL 0x2 + +struct PptpStartSessionRequest { + __be16 protocolVersion; + __u16 reserved1; + __be32 framingCapability; + __be32 bearerCapability; + __be16 maxChannels; + __be16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStartSessionResultCode Values */ +#define PPTP_START_OK 1 +#define PPTP_START_GENERAL_ERROR 2 +#define PPTP_START_ALREADY_CONNECTED 3 +#define PPTP_START_NOT_AUTHORIZED 4 +#define PPTP_START_UNKNOWN_PROTOCOL 5 + +struct PptpStartSessionReply { + __be16 protocolVersion; + __u8 resultCode; + __u8 generalErrorCode; + __be32 framingCapability; + __be32 bearerCapability; + __be16 maxChannels; + __be16 firmwareRevision; + __u8 hostName[64]; + __u8 vendorString[64]; +}; + +/* PptpStopReasons */ +#define PPTP_STOP_NONE 1 +#define PPTP_STOP_PROTOCOL 2 +#define PPTP_STOP_LOCAL_SHUTDOWN 3 + +struct PptpStopSessionRequest { + __u8 reason; + __u8 reserved1; + __u16 reserved2; +}; + +/* PptpStopSessionResultCode */ +#define PPTP_STOP_OK 1 +#define PPTP_STOP_GENERAL_ERROR 2 + +struct PptpStopSessionReply { + __u8 resultCode; + __u8 generalErrorCode; + __u16 reserved1; +}; + +struct PptpEchoRequest { + __be32 identNumber; +}; + +/* PptpEchoReplyResultCode */ +#define PPTP_ECHO_OK 1 +#define PPTP_ECHO_GENERAL_ERROR 2 + +struct PptpEchoReply { + __be32 identNumber; + __u8 resultCode; + __u8 generalErrorCode; + __u16 reserved; +}; + +/* PptpFramingType */ +#define PPTP_ASYNC_FRAMING 1 +#define PPTP_SYNC_FRAMING 2 +#define PPTP_DONT_CARE_FRAMING 3 + +/* PptpCallBearerType */ +#define PPTP_ANALOG_TYPE 1 +#define PPTP_DIGITAL_TYPE 2 +#define PPTP_DONT_CARE_BEARER_TYPE 3 + +struct PptpOutCallRequest { + __be16 callID; + __be16 callSerialNumber; + __be32 minBPS; + __be32 maxBPS; + __be32 bearerType; + __be32 framingType; + __be16 packetWindow; + __be16 packetProcDelay; + __be16 phoneNumberLength; + __u16 reserved1; + __u8 phoneNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpCallResultCode */ +#define PPTP_OUTCALL_CONNECT 1 +#define PPTP_OUTCALL_GENERAL_ERROR 2 +#define PPTP_OUTCALL_NO_CARRIER 3 +#define PPTP_OUTCALL_BUSY 4 +#define PPTP_OUTCALL_NO_DIAL_TONE 5 +#define PPTP_OUTCALL_TIMEOUT 6 +#define PPTP_OUTCALL_DONT_ACCEPT 7 + +struct PptpOutCallReply { + __be16 callID; + __be16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 causeCode; + __be32 connectSpeed; + __be16 packetWindow; + __be16 packetProcDelay; + __be32 physChannelID; +}; + +struct PptpInCallRequest { + __be16 callID; + __be16 callSerialNumber; + __be32 callBearerType; + __be32 physChannelID; + __be16 dialedNumberLength; + __be16 dialingNumberLength; + __u8 dialedNumber[64]; + __u8 dialingNumber[64]; + __u8 subAddress[64]; +}; + +/* PptpInCallResultCode */ +#define PPTP_INCALL_ACCEPT 1 +#define PPTP_INCALL_GENERAL_ERROR 2 +#define PPTP_INCALL_DONT_ACCEPT 3 + +struct PptpInCallReply { + __be16 callID; + __be16 peersCallID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 packetWindow; + __be16 packetProcDelay; + __u16 reserved; +}; + +struct PptpInCallConnected { + __be16 peersCallID; + __u16 reserved; + __be32 connectSpeed; + __be16 packetWindow; + __be16 packetProcDelay; + __be32 callFramingType; +}; + +struct PptpClearCallRequest { + __be16 callID; + __u16 reserved; +}; + +struct PptpCallDisconnectNotify { + __be16 callID; + __u8 resultCode; + __u8 generalErrorCode; + __be16 causeCode; + __u16 reserved; + __u8 callStatistics[128]; +}; + +struct PptpWanErrorNotify { + __be16 peersCallID; + __u16 reserved; + __be32 crcErrors; + __be32 framingErrors; + __be32 hardwareOverRuns; + __be32 bufferOverRuns; + __be32 timeoutErrors; + __be32 alignmentErrors; +}; + +struct PptpSetLinkInfo { + __be16 peersCallID; + __u16 reserved; + __be32 sendAccm; + __be32 recvAccm; +}; + +union pptp_ctrl_union { + struct PptpStartSessionRequest sreq; + struct PptpStartSessionReply srep; + struct PptpStopSessionRequest streq; + struct PptpStopSessionReply strep; + struct PptpOutCallRequest ocreq; + struct PptpOutCallReply ocack; + struct PptpInCallRequest icreq; + struct PptpInCallReply icack; + struct PptpInCallConnected iccon; + struct PptpClearCallRequest clrreq; + struct PptpCallDisconnectNotify disc; + struct PptpWanErrorNotify wanerr; + struct PptpSetLinkInfo setlink; +}; + +/* crap needed for nf_conntrack_compat.h */ +struct nf_conn; +struct nf_conntrack_expect; +enum ip_conntrack_info; + +extern int +(*nf_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern int +(*nf_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq); + +extern void +(*nf_nat_pptp_hook_exp_gre)(struct nf_conntrack_expect *exp_orig, + struct nf_conntrack_expect *exp_reply); + +extern void +(*nf_nat_pptp_hook_expectfn)(struct nf_conn *ct, + struct nf_conntrack_expect *exp); + +#endif /* __KERNEL__ */ +#endif /* _NF_CONNTRACK_PPTP_H */ diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h new file mode 100644 index 000000000000..4e6bbce04ff8 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -0,0 +1,112 @@ +#ifndef _CONNTRACK_PROTO_GRE_H +#define _CONNTRACK_PROTO_GRE_H +#include + +/* GRE PROTOCOL HEADER */ + +/* GRE Version field */ +#define GRE_VERSION_1701 0x0 +#define GRE_VERSION_PPTP 0x1 + +/* GRE Protocol field */ +#define GRE_PROTOCOL_PPTP 0x880B + +/* GRE Flags */ +#define GRE_FLAG_C 0x80 +#define GRE_FLAG_R 0x40 +#define GRE_FLAG_K 0x20 +#define GRE_FLAG_S 0x10 +#define GRE_FLAG_A 0x80 + +#define GRE_IS_C(f) ((f)&GRE_FLAG_C) +#define GRE_IS_R(f) ((f)&GRE_FLAG_R) +#define GRE_IS_K(f) ((f)&GRE_FLAG_K) +#define GRE_IS_S(f) ((f)&GRE_FLAG_S) +#define GRE_IS_A(f) ((f)&GRE_FLAG_A) + +/* GRE is a mess: Four different standards */ +struct gre_hdr { +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u16 rec:3, + srr:1, + seq:1, + key:1, + routing:1, + csum:1, + version:3, + reserved:4, + ack:1; +#elif defined(__BIG_ENDIAN_BITFIELD) + __u16 csum:1, + routing:1, + key:1, + seq:1, + srr:1, + rec:3, + ack:1, + reserved:4, + version:3; +#else +#error "Adjust your defines" +#endif + __be16 protocol; +}; + +/* modified GRE header for PPTP */ +struct gre_hdr_pptp { + __u8 flags; /* bitfield */ + __u8 version; /* should be GRE_VERSION_PPTP */ + __be16 protocol; /* should be GRE_PROTOCOL_PPTP */ + __be16 payload_len; /* size of ppp payload, not inc. gre header */ + __be16 call_id; /* peer's call_id for this session */ + __be32 seq; /* sequence number. Present if S==1 */ + __be32 ack; /* seq number of highest packet recieved by */ + /* sender in this session */ +}; + +struct nf_ct_gre { + unsigned int stream_timeout; + unsigned int timeout; +}; + +#ifdef __KERNEL__ +#include + +struct nf_conn; + +/* structure for original <-> reply keymap */ +struct nf_ct_gre_keymap { + struct list_head list; + struct nf_conntrack_tuple tuple; +}; + +/* add new tuple->key_reply pair to keymap */ +int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, + struct nf_conntrack_tuple *t); + +/* delete keymap entries */ +void nf_ct_gre_keymap_destroy(struct nf_conn *ct); + +/* get pointer to gre key, if present */ +static inline __be32 *gre_key(struct gre_hdr *greh) +{ + if (!greh->key) + return NULL; + if (greh->csum || greh->routing) + return (__be32 *)(greh+sizeof(*greh)+4); + return (__be32 *)(greh+sizeof(*greh)); +} + +/* get pointer ot gre csum, if present */ +static inline __sum16 *gre_csum(struct gre_hdr *greh) +{ + if (!greh->csum) + return NULL; + return (__sum16 *)(greh+sizeof(*greh)); +} + +extern void nf_ct_gre_keymap_flush(void); +extern void nf_nat_need_gre(void); + +#endif /* __KERNEL__ */ +#endif /* _CONNTRACK_PROTO_GRE_H */ diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index a1c57ee0a4fa..1401ccc051c4 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -11,10 +11,12 @@ #ifdef CONFIG_NF_NAT_NEEDED #include +#include /* per conntrack: nat application helper private data */ union nf_conntrack_nat_help { /* insert nat helper private data here */ + struct nf_nat_pptp nat_pptp_info; }; struct nf_conn_nat { diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 1646076933b1..032b36a0e378 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -21,6 +21,7 @@ #include #include +#include #include #include @@ -33,6 +34,7 @@ union nf_conntrack_proto { struct ip_ct_tcp tcp; struct ip_ct_icmp icmp; struct nf_ct_icmpv6 icmpv6; + struct nf_ct_gre gre; }; union nf_conntrack_expect_proto { @@ -41,12 +43,14 @@ union nf_conntrack_expect_proto { /* Add protocol helper include file here */ #include +#include #include /* per conntrack: application helper private data */ union nf_conntrack_help { /* insert conntrack helper private data (master) here */ struct nf_ct_ftp_master ct_ftp_info; + struct nf_ct_pptp_master ct_pptp_info; struct nf_ct_h323_master ct_h323_info; }; diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index fbba9e8b95fc..8c72ac9f0ab8 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -34,6 +34,8 @@ struct nf_conntrack_helper struct nf_conn *ct, enum ip_conntrack_info conntrackinfo); + void (*destroy)(struct nf_conn *ct); + int (*to_nfattr)(struct sk_buff *skb, const struct nf_conn *ct); }; diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index c96a9c576736..5d72b16e876f 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -49,6 +49,9 @@ union nf_conntrack_man_proto struct { __be16 port; } sctp; + struct { + __be16 key; /* GRE key is 32bit, PPtP only uses 16bit */ + } gre; }; /* The manipulable part of the tuple. */ @@ -84,6 +87,9 @@ struct nf_conntrack_tuple struct { __be16 port; } sctp; + struct { + __be16 key; + } gre; } u; /* The protocol. */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 4555f721dfc1..c3327ac024de 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -484,6 +484,10 @@ config IP_NF_NAT_SNMP_BASIC # '&&' (6) # # (6) Returns the result of min(/expr/, /expr/). +config NF_NAT_PROTO_GRE + tristate + depends on NF_NAT && NF_CT_PROTO_GRE + config IP_NF_NAT_FTP tristate depends on IP_NF_IPTABLES && IP_NF_CONNTRACK && IP_NF_NAT @@ -528,6 +532,12 @@ config IP_NF_NAT_PPTP default IP_NF_NAT if IP_NF_PPTP=y default m if IP_NF_PPTP=m +config NF_NAT_PPTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_PPTP + select NF_NAT_PROTO_GRE + config IP_NF_NAT_H323 tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 56733c370327..ef33ff2cdda9 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -54,6 +54,10 @@ obj-$(CONFIG_NF_NAT_AMANDA) += nf_nat_amanda.o obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o +obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o + +# NAT protocols (nf_nat) +obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o # generic IP tables obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c new file mode 100644 index 000000000000..0ae45b79a4eb --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -0,0 +1,315 @@ +/* + * nf_nat_pptp.c + * + * NAT support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * TODO: - NAT to a unique tuple, not to TCP source port + * (needs netfilter tuple reservation) + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#define NF_NAT_PPTP_VERSION "3.0" + +#define REQ_CID(req, off) (*(__be16 *)((char *)(req) + (off))) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT helper module for PPTP"); +MODULE_ALIAS("ip_nat_pptp"); + +#if 0 +extern const char *pptp_msg_name[]; +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +static void pptp_nat_expected(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_conn *master = ct->master; + struct nf_conntrack_expect *other_exp; + struct nf_conntrack_tuple t; + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + struct ip_nat_range range; + + ct_pptp_info = &nfct_help(master)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(master)->help.nat_pptp_info; + + /* And here goes the grand finale of corrosion... */ + if (exp->dir == IP_CT_DIR_ORIGINAL) { + DEBUGP("we are PNS->PAC\n"); + /* therefore, build tuple for PAC->PNS */ + t.src.l3num = AF_INET; + t.src.u3.ip = master->tuplehash[!exp->dir].tuple.src.u3.ip; + t.src.u.gre.key = ct_pptp_info->pac_call_id; + t.dst.u3.ip = master->tuplehash[!exp->dir].tuple.dst.u3.ip; + t.dst.u.gre.key = ct_pptp_info->pns_call_id; + t.dst.protonum = IPPROTO_GRE; + } else { + DEBUGP("we are PAC->PNS\n"); + /* build tuple for PNS->PAC */ + t.src.l3num = AF_INET; + t.src.u3.ip = master->tuplehash[exp->dir].tuple.src.u3.ip; + t.src.u.gre.key = nat_pptp_info->pns_call_id; + t.dst.u3.ip = master->tuplehash[exp->dir].tuple.dst.u3.ip; + t.dst.u.gre.key = nat_pptp_info->pac_call_id; + t.dst.protonum = IPPROTO_GRE; + } + + DEBUGP("trying to unexpect other dir: "); + NF_CT_DUMP_TUPLE(&t); + other_exp = nf_conntrack_expect_find_get(&t); + if (other_exp) { + nf_conntrack_unexpect_related(other_exp); + nf_conntrack_expect_put(other_exp); + DEBUGP("success\n"); + } else { + DEBUGP("not found!\n"); + } + + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; + if (exp->dir == IP_CT_DIR_ORIGINAL) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do source manip */ + nf_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.src.u3.ip; + if (exp->dir == IP_CT_DIR_REPLY) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do destination manip */ + nf_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); +} + +/* outbound packets == from PNS to PAC */ +static int +pptp_outbound_pkt(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) + +{ + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + u_int16_t msg; + __be16 new_callid; + unsigned int cid_off; + + ct_pptp_info = &nfct_help(ct)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + + new_callid = ct_pptp_info->pns_call_id; + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REQUEST: + cid_off = offsetof(union pptp_ctrl_union, ocreq.callID); + /* FIXME: ideally we would want to reserve a call ID + * here. current netfilter NAT core is not able to do + * this :( For now we use TCP source port. This breaks + * multiple calls within one control session */ + + /* save original call ID in nat_info */ + nat_pptp_info->pns_call_id = ct_pptp_info->pns_call_id; + + /* don't use tcph->source since we are at a DSTmanip + * hook (e.g. PREROUTING) and pkt is not mangled yet */ + new_callid = ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.tcp.port; + + /* save new call ID in ct info */ + ct_pptp_info->pns_call_id = new_callid; + break; + case PPTP_IN_CALL_REPLY: + cid_off = offsetof(union pptp_ctrl_union, icack.callID); + break; + case PPTP_CALL_CLEAR_REQUEST: + cid_off = offsetof(union pptp_ctrl_union, clrreq.callID); + break; + default: + DEBUGP("unknown outbound packet 0x%04x:%s\n", msg, + (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + case PPTP_SET_LINK_INFO: + /* only need to NAT in case PAC is behind NAT box */ + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REQUEST, IN_CALL_REPLY, CALL_CLEAR_REQUEST pass + * down to here */ + DEBUGP("altering call id from 0x%04x to 0x%04x\n", + ntohs(REQ_CID(pptpReq, cid_off)), ntohs(new_callid)); + + /* mangle packet */ + if (nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + cid_off + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader), + sizeof(new_callid), (char *)&new_callid, + sizeof(new_callid)) == 0) + return NF_DROP; + return NF_ACCEPT; +} + +static void +pptp_exp_gre(struct nf_conntrack_expect *expect_orig, + struct nf_conntrack_expect *expect_reply) +{ + struct nf_conn *ct = expect_orig->master; + struct nf_ct_pptp_master *ct_pptp_info; + struct nf_nat_pptp *nat_pptp_info; + + ct_pptp_info = &nfct_help(ct)->help.ct_pptp_info; + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + + /* save original PAC call ID in nat_info */ + nat_pptp_info->pac_call_id = ct_pptp_info->pac_call_id; + + /* alter expectation for PNS->PAC direction */ + expect_orig->saved_proto.gre.key = ct_pptp_info->pns_call_id; + expect_orig->tuple.src.u.gre.key = nat_pptp_info->pns_call_id; + expect_orig->tuple.dst.u.gre.key = ct_pptp_info->pac_call_id; + expect_orig->dir = IP_CT_DIR_ORIGINAL; + + /* alter expectation for PAC->PNS direction */ + expect_reply->saved_proto.gre.key = nat_pptp_info->pns_call_id; + expect_reply->tuple.src.u.gre.key = nat_pptp_info->pac_call_id; + expect_reply->tuple.dst.u.gre.key = ct_pptp_info->pns_call_id; + expect_reply->dir = IP_CT_DIR_REPLY; +} + +/* inbound packets == from PAC to PNS */ +static int +pptp_inbound_pkt(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) +{ + struct nf_nat_pptp *nat_pptp_info; + u_int16_t msg; + __be16 new_pcid; + unsigned int pcid_off; + + nat_pptp_info = &nfct_nat(ct)->help.nat_pptp_info; + new_pcid = nat_pptp_info->pns_call_id; + + switch (msg = ntohs(ctlh->messageType)) { + case PPTP_OUT_CALL_REPLY: + pcid_off = offsetof(union pptp_ctrl_union, ocack.peersCallID); + break; + case PPTP_IN_CALL_CONNECT: + pcid_off = offsetof(union pptp_ctrl_union, iccon.peersCallID); + break; + case PPTP_IN_CALL_REQUEST: + /* only need to nat in case PAC is behind NAT box */ + return NF_ACCEPT; + case PPTP_WAN_ERROR_NOTIFY: + pcid_off = offsetof(union pptp_ctrl_union, wanerr.peersCallID); + break; + case PPTP_CALL_DISCONNECT_NOTIFY: + pcid_off = offsetof(union pptp_ctrl_union, disc.callID); + break; + case PPTP_SET_LINK_INFO: + pcid_off = offsetof(union pptp_ctrl_union, setlink.peersCallID); + break; + default: + DEBUGP("unknown inbound packet %s\n", (msg <= PPTP_MSG_MAX)? + pptp_msg_name[msg]:pptp_msg_name[0]); + /* fall through */ + case PPTP_START_SESSION_REQUEST: + case PPTP_START_SESSION_REPLY: + case PPTP_STOP_SESSION_REQUEST: + case PPTP_STOP_SESSION_REPLY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* no need to alter packet */ + return NF_ACCEPT; + } + + /* only OUT_CALL_REPLY, IN_CALL_CONNECT, IN_CALL_REQUEST, + * WAN_ERROR_NOTIFY, CALL_DISCONNECT_NOTIFY pass down here */ + + /* mangle packet */ + DEBUGP("altering peer call id from 0x%04x to 0x%04x\n", + ntohs(REQ_CID(pptpReq, pcid_off)), ntohs(new_pcid)); + + if (nf_nat_mangle_tcp_packet(pskb, ct, ctinfo, + pcid_off + sizeof(struct pptp_pkt_hdr) + + sizeof(struct PptpControlHeader), + sizeof(new_pcid), (char *)&new_pcid, + sizeof(new_pcid)) == 0) + return NF_DROP; + return NF_ACCEPT; +} + +static int __init nf_nat_helper_pptp_init(void) +{ + nf_nat_need_gre(); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_outbound)); + rcu_assign_pointer(nf_nat_pptp_hook_outbound, pptp_outbound_pkt); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_inbound)); + rcu_assign_pointer(nf_nat_pptp_hook_inbound, pptp_inbound_pkt); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_exp_gre)); + rcu_assign_pointer(nf_nat_pptp_hook_exp_gre, pptp_exp_gre); + + BUG_ON(rcu_dereference(nf_nat_pptp_hook_expectfn)); + rcu_assign_pointer(nf_nat_pptp_hook_expectfn, pptp_nat_expected); + return 0; +} + +static void __exit nf_nat_helper_pptp_fini(void) +{ + rcu_assign_pointer(nf_nat_pptp_hook_expectfn, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_exp_gre, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_inbound, NULL); + rcu_assign_pointer(nf_nat_pptp_hook_outbound, NULL); + synchronize_rcu(); +} + +module_init(nf_nat_helper_pptp_init); +module_exit(nf_nat_helper_pptp_fini); diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c new file mode 100644 index 000000000000..d3de579e09d2 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -0,0 +1,179 @@ +/* + * nf_nat_proto_gre.c + * + * NAT protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include + +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE"); + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, \ + __FUNCTION__, ## args) +#else +#define DEBUGP(x, args...) +#endif + +/* is key in given range between min and max */ +static int +gre_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 key; + + if (maniptype == IP_NAT_MANIP_SRC) + key = tuple->src.u.gre.key; + else + key = tuple->dst.u.gre.key; + + return ntohs(key) >= ntohs(min->gre.key) && + ntohs(key) <= ntohs(max->gre.key); +} + +/* generate unique tuple ... */ +static int +gre_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *conntrack) +{ + static u_int16_t key; + __be16 *keyptr; + unsigned int min, i, range_size; + + if (maniptype == IP_NAT_MANIP_SRC) + keyptr = &tuple->src.u.gre.key; + else + keyptr = &tuple->dst.u.gre.key; + + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + DEBUGP("%p: NATing GRE PPTP\n", conntrack); + min = 1; + range_size = 0xffff; + } else { + min = ntohs(range->min.gre.key); + range_size = ntohs(range->max.gre.key) - min + 1; + } + + DEBUGP("min = %u, range_size = %u\n", min, range_size); + + for (i = 0; i < range_size; i++, key++) { + *keyptr = htons(min + key % range_size); + if (!nf_nat_used_tuple(tuple, conntrack)) + return 1; + } + + DEBUGP("%p: no NAT mapping\n", conntrack); + return 0; +} + +/* manipulate a GRE packet according to maniptype */ +static int +gre_manip_pkt(struct sk_buff **pskb, unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype) +{ + struct gre_hdr *greh; + struct gre_hdr_pptp *pgreh; + struct iphdr *iph = (struct iphdr *)((*pskb)->data + iphdroff); + unsigned int hdroff = iphdroff + iph->ihl * 4; + + /* pgreh includes two optional 32bit fields which are not required + * to be there. That's where the magic '8' comes from */ + if (!skb_make_writable(pskb, hdroff + sizeof(*pgreh) - 8)) + return 0; + + greh = (void *)(*pskb)->data + hdroff; + pgreh = (struct gre_hdr_pptp *)greh; + + /* we only have destination manip of a packet, since 'source key' + * is not present in the packet itself */ + if (maniptype != IP_NAT_MANIP_DST) + return 1; + switch (greh->version) { + case 0: + if (!greh->key) { + DEBUGP("can't nat GRE w/o key\n"); + break; + } + if (greh->csum) { + /* FIXME: Never tested this code... */ + nf_proto_csum_replace4(gre_csum(greh), *pskb, + *(gre_key(greh)), + tuple->dst.u.gre.key, 0); + } + *(gre_key(greh)) = tuple->dst.u.gre.key; + break; + case GRE_VERSION_PPTP: + DEBUGP("call_id -> 0x%04x\n", ntohs(tuple->dst.u.gre.key)); + pgreh->call_id = tuple->dst.u.gre.key; + break; + default: + DEBUGP("can't nat unknown GRE version\n"); + return 0; + } + return 1; +} + +static struct nf_nat_protocol gre __read_mostly = { + .name = "GRE", + .protonum = IPPROTO_GRE, + .manip_pkt = gre_manip_pkt, + .in_range = gre_in_range, + .unique_tuple = gre_unique_tuple, +#if defined(CONFIG_IP_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_IP_NF_CONNTRACK_NETLINK_MODULE) + .range_to_nfattr = nf_nat_port_range_to_nfattr, + .nfattr_to_range = nf_nat_port_nfattr_to_range, +#endif +}; + +int __init nf_nat_proto_gre_init(void) +{ + return nf_nat_protocol_register(&gre); +} + +void __exit nf_nat_proto_gre_fini(void) +{ + nf_nat_protocol_unregister(&gre); +} + +module_init(nf_nat_proto_gre_init); +module_exit(nf_nat_proto_gre_fini); + +void nf_nat_need_gre(void) +{ + return; +} +EXPORT_SYMBOL_GPL(nf_nat_need_gre); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index d1a365d83c53..6b2eb26ae03f 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -120,6 +120,10 @@ config NF_CONNTRACK_EVENTS If unsure, say `N'. +config NF_CT_PROTO_GRE + tristate + depends on EXPERIMENTAL && NF_CONNTRACK + config NF_CT_PROTO_SCTP tristate 'SCTP protocol on new connection tracking support (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK @@ -213,6 +217,25 @@ config NF_CONNTRACK_NETBIOS_NS To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_PPTP + tristate "PPtP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + select NF_CT_PROTO_GRE + help + This module adds support for PPTP (Point to Point Tunnelling + Protocol, RFC2637) connection tracking and NAT. + + If you are running PPTP sessions over a stateful firewall or NAT + box, you may want to enable this feature. + + Please note that not all PPTP modes of operation are supported yet. + Specifically these limitations exist: + - Blindy assumes that control connections are always established + in PNS->PAC direction. This is a violation of RFC2637. + - Only supports a single call within each session + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 67144b2af647..897bed4cbd79 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o # SCTP protocol connection tracking +obj-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o # netlink interface for nf_conntrack @@ -27,6 +28,7 @@ obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o +obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index aa8beabfeebb..ed756c928bc4 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -300,6 +300,7 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; + struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -310,6 +311,9 @@ destroy_conntrack(struct nf_conntrack *nfct) nf_conntrack_event(IPCT_DESTROY, ct); set_bit(IPS_DYING_BIT, &ct->status); + if (help && help->helper && help->helper->destroy) + help->helper->destroy(ct); + /* To make sure we don't get any weird locking issues here: * destroy_conntrack() MUST NOT be called with a write lock * to nf_conntrack_lock!!! -HW */ diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c new file mode 100644 index 000000000000..f0ff00e0d052 --- /dev/null +++ b/net/netfilter/nf_conntrack_pptp.c @@ -0,0 +1,607 @@ +/* + * Connection tracking support for PPTP (Point to Point Tunneling Protocol). + * PPTP is a a protocol for creating virtual private networks. + * It is a specification defined by Microsoft and some vendors + * working with Microsoft. PPTP is built on top of a modified + * version of the Internet Generic Routing Encapsulation Protocol. + * GRE is defined in RFC 1701 and RFC 1702. Documentation of + * PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + * Limitations: + * - We blindly assume that control connections are always + * established in PNS->PAC direction. This is a violation + * of RFFC2673 + * - We can only support one single call within each session + * TODO: + * - testing of incoming PPTP calls + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#define NF_CT_PPTP_VERSION "3.1" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("Netfilter connection tracking helper module for PPTP"); +MODULE_ALIAS("ip_conntrack_pptp"); + +static DEFINE_SPINLOCK(nf_pptp_lock); + +int +(*nf_nat_pptp_hook_outbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_outbound); + +int +(*nf_nat_pptp_hook_inbound)(struct sk_buff **pskb, + struct nf_conn *ct, enum ip_conntrack_info ctinfo, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_inbound); + +void +(*nf_nat_pptp_hook_exp_gre)(struct nf_conntrack_expect *expect_orig, + struct nf_conntrack_expect *expect_reply) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_exp_gre); + +void +(*nf_nat_pptp_hook_expectfn)(struct nf_conn *ct, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_expectfn); + +#if 0 +/* PptpControlMessageType names */ +const char *pptp_msg_name[] = { + "UNKNOWN_MESSAGE", + "START_SESSION_REQUEST", + "START_SESSION_REPLY", + "STOP_SESSION_REQUEST", + "STOP_SESSION_REPLY", + "ECHO_REQUEST", + "ECHO_REPLY", + "OUT_CALL_REQUEST", + "OUT_CALL_REPLY", + "IN_CALL_REQUEST", + "IN_CALL_REPLY", + "IN_CALL_CONNECT", + "CALL_CLEAR_REQUEST", + "CALL_DISCONNECT_NOTIFY", + "WAN_ERROR_NOTIFY", + "SET_LINK_INFO" +}; +EXPORT_SYMBOL(pptp_msg_name); +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#else +#define DEBUGP(format, args...) +#endif + +#define SECS *HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS + +#define PPTP_GRE_TIMEOUT (10 MINS) +#define PPTP_GRE_STREAM_TIMEOUT (5 HOURS) + +static void pptp_expectfn(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + typeof(nf_nat_pptp_hook_expectfn) nf_nat_pptp_expectfn; + DEBUGP("increasing timeouts\n"); + + /* increase timeout of GRE data channel conntrack entry */ + ct->proto.gre.timeout = PPTP_GRE_TIMEOUT; + ct->proto.gre.stream_timeout = PPTP_GRE_STREAM_TIMEOUT; + + /* Can you see how rusty this code is, compared with the pre-2.6.11 + * one? That's what happened to my shiny newnat of 2002 ;( -HW */ + + rcu_read_lock(); + nf_nat_pptp_expectfn = rcu_dereference(nf_nat_pptp_hook_expectfn); + if (nf_nat_pptp_expectfn && ct->status & IPS_NAT_MASK) + nf_nat_pptp_expectfn(ct, exp); + else { + struct nf_conntrack_tuple inv_t; + struct nf_conntrack_expect *exp_other; + + /* obviously this tuple inversion only works until you do NAT */ + nf_ct_invert_tuplepr(&inv_t, &exp->tuple); + DEBUGP("trying to unexpect other dir: "); + NF_CT_DUMP_TUPLE(&inv_t); + + exp_other = nf_conntrack_expect_find_get(&inv_t); + if (exp_other) { + /* delete other expectation. */ + DEBUGP("found\n"); + nf_conntrack_unexpect_related(exp_other); + nf_conntrack_expect_put(exp_other); + } else { + DEBUGP("not found\n"); + } + } + rcu_read_unlock(); +} + +static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) +{ + struct nf_conntrack_tuple_hash *h; + struct nf_conntrack_expect *exp; + struct nf_conn *sibling; + + DEBUGP("trying to timeout ct or exp for tuple "); + NF_CT_DUMP_TUPLE(t); + + h = nf_conntrack_find_get(t, NULL); + if (h) { + sibling = nf_ct_tuplehash_to_ctrack(h); + DEBUGP("setting timeout of conntrack %p to 0\n", sibling); + sibling->proto.gre.timeout = 0; + sibling->proto.gre.stream_timeout = 0; + if (del_timer(&sibling->timeout)) + sibling->timeout.function((unsigned long)sibling); + nf_ct_put(sibling); + return 1; + } else { + exp = nf_conntrack_expect_find_get(t); + if (exp) { + DEBUGP("unexpect_related of expect %p\n", exp); + nf_conntrack_unexpect_related(exp); + nf_conntrack_expect_put(exp); + return 1; + } + } + return 0; +} + +/* timeout GRE data connections */ +static void pptp_destroy_siblings(struct nf_conn *ct) +{ + struct nf_conn_help *help = nfct_help(ct); + struct nf_conntrack_tuple t; + + nf_ct_gre_keymap_destroy(ct); + + /* try original (pns->pac) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = help->help.ct_pptp_info.pns_call_id; + t.dst.u.gre.key = help->help.ct_pptp_info.pac_call_id; + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout original pns->pac ct/exp\n"); + + /* try reply (pac->pns) tuple */ + memcpy(&t, &ct->tuplehash[IP_CT_DIR_REPLY].tuple, sizeof(t)); + t.dst.protonum = IPPROTO_GRE; + t.src.u.gre.key = help->help.ct_pptp_info.pac_call_id; + t.dst.u.gre.key = help->help.ct_pptp_info.pns_call_id; + if (!destroy_sibling_or_exp(&t)) + DEBUGP("failed to timeout reply pac->pns ct/exp\n"); +} + +/* expect GRE connections (PNS->PAC and PAC->PNS direction) */ +static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) +{ + struct nf_conntrack_expect *exp_orig, *exp_reply; + enum ip_conntrack_dir dir; + int ret = 1; + typeof(nf_nat_pptp_hook_exp_gre) nf_nat_pptp_exp_gre; + + exp_orig = nf_conntrack_expect_alloc(ct); + if (exp_orig == NULL) + goto out; + + exp_reply = nf_conntrack_expect_alloc(ct); + if (exp_reply == NULL) + goto out_put_orig; + + /* original direction, PNS->PAC */ + dir = IP_CT_DIR_ORIGINAL; + nf_conntrack_expect_init(exp_orig, ct->tuplehash[dir].tuple.src.l3num, + &ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[dir].tuple.dst.u3, + IPPROTO_GRE, &peer_callid, &callid); + exp_orig->expectfn = pptp_expectfn; + + /* reply direction, PAC->PNS */ + dir = IP_CT_DIR_REPLY; + nf_conntrack_expect_init(exp_reply, ct->tuplehash[dir].tuple.src.l3num, + &ct->tuplehash[dir].tuple.src.u3, + &ct->tuplehash[dir].tuple.dst.u3, + IPPROTO_GRE, &callid, &peer_callid); + exp_reply->expectfn = pptp_expectfn; + + nf_nat_pptp_exp_gre = rcu_dereference(nf_nat_pptp_hook_exp_gre); + if (nf_nat_pptp_exp_gre && ct->status & IPS_NAT_MASK) + nf_nat_pptp_exp_gre(exp_orig, exp_reply); + if (nf_conntrack_expect_related(exp_orig) != 0) + goto out_put_both; + if (nf_conntrack_expect_related(exp_reply) != 0) + goto out_unexpect_orig; + + /* Add GRE keymap entries */ + if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_ORIGINAL, &exp_orig->tuple) != 0) + goto out_unexpect_both; + if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_REPLY, &exp_reply->tuple) != 0) { + nf_ct_gre_keymap_destroy(ct); + goto out_unexpect_both; + } + ret = 0; + +out_put_both: + nf_conntrack_expect_put(exp_reply); +out_put_orig: + nf_conntrack_expect_put(exp_orig); +out: + return ret; + +out_unexpect_both: + nf_conntrack_unexpect_related(exp_reply); +out_unexpect_orig: + nf_conntrack_unexpect_related(exp_orig); + goto out_put_both; +} + +static inline int +pptp_inbound_pkt(struct sk_buff **pskb, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq, + unsigned int reqlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + u_int16_t msg; + __be16 cid = 0, pcid = 0; + typeof(nf_nat_pptp_hook_inbound) nf_nat_pptp_inbound; + + msg = ntohs(ctlh->messageType); + DEBUGP("inbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REPLY: + /* server confirms new control session */ + if (info->sstate < PPTP_SESSION_REQUESTED) + goto invalid; + if (pptpReq->srep.resultCode == PPTP_START_OK) + info->sstate = PPTP_SESSION_CONFIRMED; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_STOP_SESSION_REPLY: + /* server confirms end of control session */ + if (info->sstate > PPTP_SESSION_STOPREQ) + goto invalid; + if (pptpReq->strep.resultCode == PPTP_STOP_OK) + info->sstate = PPTP_SESSION_NONE; + else + info->sstate = PPTP_SESSION_ERROR; + break; + + case PPTP_OUT_CALL_REPLY: + /* server accepted call, we now expect GRE frames */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + if (info->cstate != PPTP_CALL_OUT_REQ && + info->cstate != PPTP_CALL_OUT_CONF) + goto invalid; + + cid = pptpReq->ocack.callID; + pcid = pptpReq->ocack.peersCallID; + if (info->pns_call_id != pcid) + goto invalid; + DEBUGP("%s, CID=%X, PCID=%X\n", pptp_msg_name[msg], + ntohs(cid), ntohs(pcid)); + + if (pptpReq->ocack.resultCode == PPTP_OUTCALL_CONNECT) { + info->cstate = PPTP_CALL_OUT_CONF; + info->pac_call_id = cid; + exp_gre(ct, cid, pcid); + } else + info->cstate = PPTP_CALL_NONE; + break; + + case PPTP_IN_CALL_REQUEST: + /* server tells us about incoming call request */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + + cid = pptpReq->icreq.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->cstate = PPTP_CALL_IN_REQ; + info->pac_call_id = cid; + break; + + case PPTP_IN_CALL_CONNECT: + /* server tells us about incoming call established */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + if (info->cstate != PPTP_CALL_IN_REP && + info->cstate != PPTP_CALL_IN_CONF) + goto invalid; + + pcid = pptpReq->iccon.peersCallID; + cid = info->pac_call_id; + + if (info->pns_call_id != pcid) + goto invalid; + + DEBUGP("%s, PCID=%X\n", pptp_msg_name[msg], ntohs(pcid)); + info->cstate = PPTP_CALL_IN_CONF; + + /* we expect a GRE connection from PAC to PNS */ + exp_gre(ct, cid, pcid); + break; + + case PPTP_CALL_DISCONNECT_NOTIFY: + /* server confirms disconnect */ + cid = pptpReq->disc.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->cstate = PPTP_CALL_NONE; + + /* untrack this call id, unexpect GRE packets */ + pptp_destroy_siblings(ct); + break; + + case PPTP_WAN_ERROR_NOTIFY: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + + default: + goto invalid; + } + + nf_nat_pptp_inbound = rcu_dereference(nf_nat_pptp_hook_inbound); + if (nf_nat_pptp_inbound && ct->status & IPS_NAT_MASK) + return nf_nat_pptp_inbound(pskb, ct, ctinfo, ctlh, pptpReq); + return NF_ACCEPT; + +invalid: + DEBUGP("invalid %s: type=%d cid=%u pcid=%u " + "cstate=%d sstate=%d pns_cid=%u pac_cid=%u\n", + msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : pptp_msg_name[0], + msg, ntohs(cid), ntohs(pcid), info->cstate, info->sstate, + ntohs(info->pns_call_id), ntohs(info->pac_call_id)); + return NF_ACCEPT; +} + +static inline int +pptp_outbound_pkt(struct sk_buff **pskb, + struct PptpControlHeader *ctlh, + union pptp_ctrl_union *pptpReq, + unsigned int reqlen, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + u_int16_t msg; + __be16 cid = 0, pcid = 0; + typeof(nf_nat_pptp_hook_outbound) nf_nat_pptp_outbound; + + msg = ntohs(ctlh->messageType); + DEBUGP("outbound control message %s\n", pptp_msg_name[msg]); + + switch (msg) { + case PPTP_START_SESSION_REQUEST: + /* client requests for new control session */ + if (info->sstate != PPTP_SESSION_NONE) + goto invalid; + info->sstate = PPTP_SESSION_REQUESTED; + break; + + case PPTP_STOP_SESSION_REQUEST: + /* client requests end of control session */ + info->sstate = PPTP_SESSION_STOPREQ; + break; + + case PPTP_OUT_CALL_REQUEST: + /* client initiating connection to server */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + info->cstate = PPTP_CALL_OUT_REQ; + /* track PNS call id */ + cid = pptpReq->ocreq.callID; + DEBUGP("%s, CID=%X\n", pptp_msg_name[msg], ntohs(cid)); + info->pns_call_id = cid; + break; + + case PPTP_IN_CALL_REPLY: + /* client answers incoming call */ + if (info->cstate != PPTP_CALL_IN_REQ && + info->cstate != PPTP_CALL_IN_REP) + goto invalid; + + cid = pptpReq->icack.callID; + pcid = pptpReq->icack.peersCallID; + if (info->pac_call_id != pcid) + goto invalid; + DEBUGP("%s, CID=%X PCID=%X\n", pptp_msg_name[msg], + ntohs(cid), ntohs(pcid)); + + if (pptpReq->icack.resultCode == PPTP_INCALL_ACCEPT) { + /* part two of the three-way handshake */ + info->cstate = PPTP_CALL_IN_REP; + info->pns_call_id = cid; + } else + info->cstate = PPTP_CALL_NONE; + break; + + case PPTP_CALL_CLEAR_REQUEST: + /* client requests hangup of call */ + if (info->sstate != PPTP_SESSION_CONFIRMED) + goto invalid; + /* FUTURE: iterate over all calls and check if + * call ID is valid. We don't do this without newnat, + * because we only know about last call */ + info->cstate = PPTP_CALL_CLEAR_REQ; + break; + + case PPTP_SET_LINK_INFO: + case PPTP_ECHO_REQUEST: + case PPTP_ECHO_REPLY: + /* I don't have to explain these ;) */ + break; + + default: + goto invalid; + } + + nf_nat_pptp_outbound = rcu_dereference(nf_nat_pptp_hook_outbound); + if (nf_nat_pptp_outbound && ct->status & IPS_NAT_MASK) + return nf_nat_pptp_outbound(pskb, ct, ctinfo, ctlh, pptpReq); + return NF_ACCEPT; + +invalid: + DEBUGP("invalid %s: type=%d cid=%u pcid=%u " + "cstate=%d sstate=%d pns_cid=%u pac_cid=%u\n", + msg <= PPTP_MSG_MAX ? pptp_msg_name[msg] : pptp_msg_name[0], + msg, ntohs(cid), ntohs(pcid), info->cstate, info->sstate, + ntohs(info->pns_call_id), ntohs(info->pac_call_id)); + return NF_ACCEPT; +} + +static const unsigned int pptp_msg_size[] = { + [PPTP_START_SESSION_REQUEST] = sizeof(struct PptpStartSessionRequest), + [PPTP_START_SESSION_REPLY] = sizeof(struct PptpStartSessionReply), + [PPTP_STOP_SESSION_REQUEST] = sizeof(struct PptpStopSessionRequest), + [PPTP_STOP_SESSION_REPLY] = sizeof(struct PptpStopSessionReply), + [PPTP_OUT_CALL_REQUEST] = sizeof(struct PptpOutCallRequest), + [PPTP_OUT_CALL_REPLY] = sizeof(struct PptpOutCallReply), + [PPTP_IN_CALL_REQUEST] = sizeof(struct PptpInCallRequest), + [PPTP_IN_CALL_REPLY] = sizeof(struct PptpInCallReply), + [PPTP_IN_CALL_CONNECT] = sizeof(struct PptpInCallConnected), + [PPTP_CALL_CLEAR_REQUEST] = sizeof(struct PptpClearCallRequest), + [PPTP_CALL_DISCONNECT_NOTIFY] = sizeof(struct PptpCallDisconnectNotify), + [PPTP_WAN_ERROR_NOTIFY] = sizeof(struct PptpWanErrorNotify), + [PPTP_SET_LINK_INFO] = sizeof(struct PptpSetLinkInfo), +}; + +/* track caller id inside control connection, call expect_related */ +static int +conntrack_pptp_help(struct sk_buff **pskb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) + +{ + int dir = CTINFO2DIR(ctinfo); + struct nf_ct_pptp_master *info = &nfct_help(ct)->help.ct_pptp_info; + struct tcphdr _tcph, *tcph; + struct pptp_pkt_hdr _pptph, *pptph; + struct PptpControlHeader _ctlh, *ctlh; + union pptp_ctrl_union _pptpReq, *pptpReq; + unsigned int tcplen = (*pskb)->len - protoff; + unsigned int datalen, reqlen, nexthdr_off; + int oldsstate, oldcstate; + int ret; + u_int16_t msg; + + /* don't do any tracking before tcp handshake complete */ + if (ctinfo != IP_CT_ESTABLISHED && + ctinfo != IP_CT_ESTABLISHED + IP_CT_IS_REPLY) + return NF_ACCEPT; + + nexthdr_off = protoff; + tcph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_tcph), &_tcph); + BUG_ON(!tcph); + nexthdr_off += tcph->doff * 4; + datalen = tcplen - tcph->doff * 4; + + pptph = skb_header_pointer(*pskb, nexthdr_off, sizeof(_pptph), &_pptph); + if (!pptph) { + DEBUGP("no full PPTP header, can't track\n"); + return NF_ACCEPT; + } + nexthdr_off += sizeof(_pptph); + datalen -= sizeof(_pptph); + + /* if it's not a control message we can't do anything with it */ + if (ntohs(pptph->packetType) != PPTP_PACKET_CONTROL || + ntohl(pptph->magicCookie) != PPTP_MAGIC_COOKIE) { + DEBUGP("not a control packet\n"); + return NF_ACCEPT; + } + + ctlh = skb_header_pointer(*pskb, nexthdr_off, sizeof(_ctlh), &_ctlh); + if (!ctlh) + return NF_ACCEPT; + nexthdr_off += sizeof(_ctlh); + datalen -= sizeof(_ctlh); + + reqlen = datalen; + msg = ntohs(ctlh->messageType); + if (msg > 0 && msg <= PPTP_MSG_MAX && reqlen < pptp_msg_size[msg]) + return NF_ACCEPT; + if (reqlen > sizeof(*pptpReq)) + reqlen = sizeof(*pptpReq); + + pptpReq = skb_header_pointer(*pskb, nexthdr_off, reqlen, &_pptpReq); + if (!pptpReq) + return NF_ACCEPT; + + oldsstate = info->sstate; + oldcstate = info->cstate; + + spin_lock_bh(&nf_pptp_lock); + + /* FIXME: We just blindly assume that the control connection is always + * established from PNS->PAC. However, RFC makes no guarantee */ + if (dir == IP_CT_DIR_ORIGINAL) + /* client -> server (PNS -> PAC) */ + ret = pptp_outbound_pkt(pskb, ctlh, pptpReq, reqlen, ct, + ctinfo); + else + /* server -> client (PAC -> PNS) */ + ret = pptp_inbound_pkt(pskb, ctlh, pptpReq, reqlen, ct, + ctinfo); + DEBUGP("sstate: %d->%d, cstate: %d->%d\n", + oldsstate, info->sstate, oldcstate, info->cstate); + spin_unlock_bh(&nf_pptp_lock); + + return ret; +} + +/* control protocol helper */ +static struct nf_conntrack_helper pptp __read_mostly = { + .name = "pptp", + .me = THIS_MODULE, + .max_expected = 2, + .timeout = 5 * 60, + .tuple.src.l3num = AF_INET, + .tuple.src.u.tcp.port = __constant_htons(PPTP_CONTROL_PORT), + .tuple.dst.protonum = IPPROTO_TCP, + .mask.src.l3num = 0xffff, + .mask.src.u.tcp.port = __constant_htons(0xffff), + .mask.dst.protonum = 0xff, + .help = conntrack_pptp_help, + .destroy = pptp_destroy_siblings, +}; + +static int __init nf_conntrack_pptp_init(void) +{ + return nf_conntrack_helper_register(&pptp); +} + +static void __exit nf_conntrack_pptp_fini(void) +{ + nf_conntrack_helper_unregister(&pptp); + nf_ct_gre_keymap_flush(); +} + +module_init(nf_conntrack_pptp_init); +module_exit(nf_conntrack_pptp_fini); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c new file mode 100644 index 000000000000..ac193ce70249 --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -0,0 +1,305 @@ +/* + * ip_conntrack_proto_gre.c - Version 3.0 + * + * Connection tracking protocol helper module for GRE. + * + * GRE is a generic encapsulation protocol, which is generally not very + * suited for NAT, as it has no protocol-specific part as port numbers. + * + * It has an optional key field, which may help us distinguishing two + * connections between the same two hosts. + * + * GRE is defined in RFC 1701 and RFC 1702, as well as RFC 2784 + * + * PPTP is built on top of a modified version of GRE, and has a mandatory + * field called "CallID", which serves us for the same purpose as the key + * field in plain GRE. + * + * Documentation about PPTP can be found in RFC 2637 + * + * (C) 2000-2005 by Harald Welte + * + * Development of this code funded by Astaro AG (http://www.astaro.com/) + * + */ + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#define GRE_TIMEOUT (30 * HZ) +#define GRE_STREAM_TIMEOUT (180 * HZ) + +#if 0 +#define DEBUGP(format, args...) printk(KERN_DEBUG "%s:%s: " format, __FILE__, __FUNCTION__, ## args) +#else +#define DEBUGP(x, args...) +#endif + +static DEFINE_RWLOCK(nf_ct_gre_lock); +static LIST_HEAD(gre_keymap_list); + +void nf_ct_gre_keymap_flush(void) +{ + struct list_head *pos, *n; + + write_lock_bh(&nf_ct_gre_lock); + list_for_each_safe(pos, n, &gre_keymap_list) { + list_del(pos); + kfree(pos); + } + write_unlock_bh(&nf_ct_gre_lock); +} +EXPORT_SYMBOL(nf_ct_gre_keymap_flush); + +static inline int gre_key_cmpfn(const struct nf_ct_gre_keymap *km, + const struct nf_conntrack_tuple *t) +{ + return km->tuple.src.l3num == t->src.l3num && + !memcmp(&km->tuple.src.u3, &t->src.u3, sizeof(t->src.u3)) && + !memcmp(&km->tuple.dst.u3, &t->dst.u3, sizeof(t->dst.u3)) && + km->tuple.dst.protonum == t->dst.protonum && + km->tuple.dst.u.all == t->dst.u.all; +} + +/* look up the source key for a given tuple */ +static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t) +{ + struct nf_ct_gre_keymap *km; + __be16 key = 0; + + read_lock_bh(&nf_ct_gre_lock); + list_for_each_entry(km, &gre_keymap_list, list) { + if (gre_key_cmpfn(km, t)) { + key = km->tuple.src.u.gre.key; + break; + } + } + read_unlock_bh(&nf_ct_gre_lock); + + DEBUGP("lookup src key 0x%x for ", key); + NF_CT_DUMP_TUPLE(t); + + return key; +} + +/* add a single keymap entry, associate with specified master ct */ +int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, + struct nf_conntrack_tuple *t) +{ + struct nf_conn_help *help = nfct_help(ct); + struct nf_ct_gre_keymap **kmp, *km; + + BUG_ON(strcmp(help->helper->name, "pptp")); + kmp = &help->help.ct_pptp_info.keymap[dir]; + if (*kmp) { + /* check whether it's a retransmission */ + list_for_each_entry(km, &gre_keymap_list, list) { + if (gre_key_cmpfn(km, t) && km == *kmp) + return 0; + } + DEBUGP("trying to override keymap_%s for ct %p\n", + dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); + return -EEXIST; + } + + km = kmalloc(sizeof(*km), GFP_ATOMIC); + if (!km) + return -ENOMEM; + memcpy(&km->tuple, t, sizeof(*t)); + *kmp = km; + + DEBUGP("adding new entry %p: ", km); + NF_CT_DUMP_TUPLE(&km->tuple); + + write_lock_bh(&nf_ct_gre_lock); + list_add_tail(&km->list, &gre_keymap_list); + write_unlock_bh(&nf_ct_gre_lock); + + return 0; +} +EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); + +/* destroy the keymap entries associated with specified master ct */ +void nf_ct_gre_keymap_destroy(struct nf_conn *ct) +{ + struct nf_conn_help *help = nfct_help(ct); + enum ip_conntrack_dir dir; + + DEBUGP("entering for ct %p\n", ct); + BUG_ON(strcmp(help->helper->name, "pptp")); + + write_lock_bh(&nf_ct_gre_lock); + for (dir = IP_CT_DIR_ORIGINAL; dir < IP_CT_DIR_MAX; dir++) { + if (help->help.ct_pptp_info.keymap[dir]) { + DEBUGP("removing %p from list\n", + help->help.ct_pptp_info.keymap[dir]); + list_del(&help->help.ct_pptp_info.keymap[dir]->list); + kfree(help->help.ct_pptp_info.keymap[dir]); + help->help.ct_pptp_info.keymap[dir] = NULL; + } + } + write_unlock_bh(&nf_ct_gre_lock); +} +EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); + +/* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */ + +/* invert gre part of tuple */ +static int gre_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->dst.u.gre.key = orig->src.u.gre.key; + tuple->src.u.gre.key = orig->dst.u.gre.key; + return 1; +} + +/* gre hdr info to tuple */ +static int gre_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct gre_hdr_pptp _pgrehdr, *pgrehdr; + __be16 srckey; + struct gre_hdr _grehdr, *grehdr; + + /* first only delinearize old RFC1701 GRE header */ + grehdr = skb_header_pointer(skb, dataoff, sizeof(_grehdr), &_grehdr); + if (!grehdr || grehdr->version != GRE_VERSION_PPTP) { + /* try to behave like "nf_conntrack_proto_generic" */ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + return 1; + } + + /* PPTP header is variable length, only need up to the call_id field */ + pgrehdr = skb_header_pointer(skb, dataoff, 8, &_pgrehdr); + if (!pgrehdr) + return 1; + + if (ntohs(grehdr->protocol) != GRE_PROTOCOL_PPTP) { + DEBUGP("GRE_VERSION_PPTP but unknown proto\n"); + return 0; + } + + tuple->dst.u.gre.key = pgrehdr->call_id; + srckey = gre_keymap_lookup(tuple); + tuple->src.u.gre.key = srckey; + + return 1; +} + +/* print gre part of tuple */ +static int gre_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "srckey=0x%x dstkey=0x%x ", + ntohs(tuple->src.u.gre.key), + ntohs(tuple->dst.u.gre.key)); +} + +/* print private data for conntrack */ +static int gre_print_conntrack(struct seq_file *s, + const struct nf_conn *ct) +{ + return seq_printf(s, "timeout=%u, stream_timeout=%u ", + (ct->proto.gre.timeout / HZ), + (ct->proto.gre.stream_timeout / HZ)); +} + +/* Returns verdict for packet, and may modify conntrack */ +static int gre_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* If we've seen traffic both ways, this is a GRE connection. + * Extend timeout. */ + if (ct->status & IPS_SEEN_REPLY) { + nf_ct_refresh_acct(ct, ctinfo, skb, + ct->proto.gre.stream_timeout); + /* Also, more likely to be important, and not a probe. */ + set_bit(IPS_ASSURED_BIT, &ct->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } else + nf_ct_refresh_acct(ct, ctinfo, skb, + ct->proto.gre.timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int gre_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) +{ + DEBUGP(": "); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + + /* initialize to sane value. Ideally a conntrack helper + * (e.g. in case of pptp) is increasing them */ + ct->proto.gre.stream_timeout = GRE_STREAM_TIMEOUT; + ct->proto.gre.timeout = GRE_TIMEOUT; + + return 1; +} + +/* Called when a conntrack entry has already been removed from the hashes + * and is about to be deleted from memory */ +static void gre_destroy(struct nf_conn *ct) +{ + struct nf_conn *master = ct->master; + DEBUGP(" entering\n"); + + if (!master) + DEBUGP("no master !?!\n"); + else + nf_ct_gre_keymap_destroy(master); +} + +/* protocol helper struct */ +static struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 = { + .l3proto = AF_INET, + .l4proto = IPPROTO_GRE, + .name = "gre", + .pkt_to_tuple = gre_pkt_to_tuple, + .invert_tuple = gre_invert_tuple, + .print_tuple = gre_print_tuple, + .print_conntrack = gre_print_conntrack, + .packet = gre_packet, + .new = gre_new, + .destroy = gre_destroy, + .me = THIS_MODULE, +#if defined(CONFIG_NF_CONNTRACK_NETLINK) || \ + defined(CONFIG_NF_CONNTRACK_NETLINK_MODULE) + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif +}; + +static int __init nf_ct_proto_gre_init(void) +{ + return nf_conntrack_l4proto_register(&nf_conntrack_l4proto_gre4); +} + +static void nf_ct_proto_gre_fini(void) +{ + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_gre4); + nf_ct_gre_keymap_flush(); +} + +module_init(nf_ct_proto_gre_init); +module_exit(nf_ct_proto_gre_fini); + +MODULE_LICENSE("GPL"); -- cgit v1.2.3 From 9fafcd7b203229c3f3893a475741afc27e276306 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:09:57 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add SIP helper port Add IPv4 and IPv6 capable nf_conntrack port of the SIP conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 41 +++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_sip.c | 283 +++++++++++++++ net/netfilter/Kconfig | 12 + net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_sip.c | 530 +++++++++++++++++++++++++++++ 7 files changed, 873 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_sip.h create mode 100644 net/ipv4/netfilter/nf_nat_sip.c create mode 100644 net/netfilter/nf_conntrack_sip.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h new file mode 100644 index 000000000000..bb7f2041db74 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -0,0 +1,41 @@ +#ifndef __NF_CONNTRACK_SIP_H__ +#define __NF_CONNTRACK_SIP_H__ +#ifdef __KERNEL__ + +#define SIP_PORT 5060 +#define SIP_TIMEOUT 3600 + +enum sip_header_pos { + POS_REG_REQ_URI, + POS_REQ_URI, + POS_FROM, + POS_TO, + POS_VIA, + POS_CONTACT, + POS_CONTENT, + POS_MEDIA, + POS_OWNER_IP4, + POS_CONNECTION_IP4, + POS_OWNER_IP6, + POS_CONNECTION_IP6, + POS_SDP_HEADER, +}; + +extern unsigned int (*nf_nat_sip_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr); +extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr); + +extern int ct_sip_get_info(struct nf_conn *ct, const char *dptr, size_t dlen, + unsigned int *matchoff, unsigned int *matchlen, + enum sip_header_pos pos); +extern int ct_sip_lnlen(const char *line, const char *limit); +extern const char *ct_sip_search(const char *needle, const char *haystack, + size_t needle_len, size_t haystack_len, + int case_sensitive); +#endif /* __KERNEL__ */ +#endif /* __NF_CONNTRACK_SIP_H__ */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index c3327ac024de..83e83f553ccb 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -555,6 +555,11 @@ config IP_NF_NAT_SIP default IP_NF_NAT if IP_NF_SIP=y default m if IP_NF_SIP=m +config NF_NAT_SIP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_SIP + # mangle + specific targets config IP_NF_MANGLE tristate "Packet mangling" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index ef33ff2cdda9..167b65e89aca 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_NF_NAT_FTP) += nf_nat_ftp.o obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o +obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o # NAT protocols (nf_nat) obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c new file mode 100644 index 000000000000..3d524b957310 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -0,0 +1,283 @@ +/* SIP extension for UDP NAT alteration. + * + * (C) 2005 by Christian Hentschel + * based on RR's ip_nat_ftp.c and other modules. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christian Hentschel "); +MODULE_DESCRIPTION("SIP NAT helper"); +MODULE_ALIAS("ip_nat_sip"); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +struct addr_map { + struct { + char src[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + char dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned int srclen, srciplen; + unsigned int dstlen, dstiplen; + } addr[IP_CT_DIR_MAX]; +}; + +static void addr_map_init(struct nf_conn *ct, struct addr_map *map) +{ + struct nf_conntrack_tuple *t; + enum ip_conntrack_dir dir; + unsigned int n; + + for (dir = 0; dir < IP_CT_DIR_MAX; dir++) { + t = &ct->tuplehash[dir].tuple; + + n = sprintf(map->addr[dir].src, "%u.%u.%u.%u", + NIPQUAD(t->src.u3.ip)); + map->addr[dir].srciplen = n; + n += sprintf(map->addr[dir].src + n, ":%u", + ntohs(t->src.u.udp.port)); + map->addr[dir].srclen = n; + + n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u", + NIPQUAD(t->dst.u3.ip)); + map->addr[dir].dstiplen = n; + n += sprintf(map->addr[dir].dst + n, ":%u", + ntohs(t->dst.u.udp.port)); + map->addr[dir].dstlen = n; + } +} + +static int map_sip_addr(struct sk_buff **pskb, enum ip_conntrack_info ctinfo, + struct nf_conn *ct, const char **dptr, size_t dlen, + enum sip_header_pos pos, struct addr_map *map) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff, addrlen; + char *addr; + + if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) + return 1; + + if ((matchlen == map->addr[dir].srciplen || + matchlen == map->addr[dir].srclen) && + memcmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { + addr = map->addr[!dir].dst; + addrlen = map->addr[!dir].dstlen; + } else if ((matchlen == map->addr[dir].dstiplen || + matchlen == map->addr[dir].dstlen) && + memcmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) { + addr = map->addr[!dir].src; + addrlen = map->addr[!dir].srclen; + } else + return 1; + + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, addr, addrlen)) + return 0; + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; + +} + +static unsigned int ip_nat_sip(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr) +{ + enum sip_header_pos pos; + struct addr_map map; + int dataoff, datalen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0") - 1) + return NF_DROP; + + addr_map_init(ct, &map); + + /* Basic rules: requests and responses. */ + if (strncmp(*dptr, "SIP/2.0", sizeof("SIP/2.0") - 1) != 0) { + /* 10.2: Constructing the REGISTER Request: + * + * The "userinfo" and "@" components of the SIP URI MUST NOT + * be present. + */ + if (datalen >= sizeof("REGISTER") - 1 && + strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0) + pos = POS_REG_REQ_URI; + else + pos = POS_REQ_URI; + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, pos, &map)) + return NF_DROP; + } + + if (!map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_FROM, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_TO, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_VIA, &map) || + !map_sip_addr(pskb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map)) + return NF_DROP; + return NF_ACCEPT; +} + +static unsigned int mangle_sip_packet(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr, size_t dlen, + char *buffer, int bufflen, + enum sip_header_pos pos) +{ + unsigned int matchlen, matchoff; + + if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) + return 0; + + if (!nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, buffer, bufflen)) + return 0; + + /* We need to reload this. Thanks Patrick. */ + *dptr = (*pskb)->data + (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + return 1; +} + +static int mangle_content_len(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char *dptr) +{ + unsigned int dataoff, matchoff, matchlen; + char buffer[sizeof("65536")]; + int bufflen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + + /* Get actual SDP lenght */ + if (ct_sip_get_info(ct, dptr, (*pskb)->len - dataoff, &matchoff, + &matchlen, POS_SDP_HEADER) > 0) { + + /* since ct_sip_get_info() give us a pointer passing 'v=' + we need to add 2 bytes in this count. */ + int c_len = (*pskb)->len - dataoff - matchoff + 2; + + /* Now, update SDP length */ + if (ct_sip_get_info(ct, dptr, (*pskb)->len - dataoff, &matchoff, + &matchlen, POS_CONTENT) > 0) { + + bufflen = sprintf(buffer, "%u", c_len); + return nf_nat_mangle_udp_packet(pskb, ct, ctinfo, + matchoff, matchlen, + buffer, bufflen); + } + } + return 0; +} + +static unsigned int mangle_sdp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + __be32 newip, u_int16_t port, + const char *dptr) +{ + char buffer[sizeof("nnn.nnn.nnn.nnn")]; + unsigned int dataoff, bufflen; + + dataoff = (*pskb)->nh.iph->ihl*4 + sizeof(struct udphdr); + + /* Mangle owner and contact info. */ + bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_OWNER_IP4)) + return 0; + + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_CONNECTION_IP4)) + return 0; + + /* Mangle media port. */ + bufflen = sprintf(buffer, "%u", port); + if (!mangle_sip_packet(pskb, ctinfo, ct, &dptr, (*pskb)->len - dataoff, + buffer, bufflen, POS_MEDIA)) + return 0; + + return mangle_content_len(pskb, ctinfo, ct, dptr); +} + +/* So, this packet has hit the connection tracking matching code. + Mangle it, and change the expectation to match the new version. */ +static unsigned int ip_nat_sdp(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr) +{ + struct nf_conn *ct = exp->master; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + __be32 newip; + u_int16_t port; + + DEBUGP("ip_nat_sdp():\n"); + + /* Connection will come from reply */ + newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + + exp->tuple.dst.u3.ip = newip; + exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port; + exp->dir = !dir; + + /* When you see the packet, we need to NAT it the same as the + this one. */ + exp->expectfn = nf_nat_follow_master; + + /* Try to get same port: if not, try to change it. */ + for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) { + exp->tuple.dst.u.udp.port = htons(port); + if (nf_conntrack_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + if (!mangle_sdp(pskb, ctinfo, ct, newip, port, dptr)) { + nf_conntrack_unexpect_related(exp); + return NF_DROP; + } + return NF_ACCEPT; +} + +static void __exit nf_nat_sip_fini(void) +{ + rcu_assign_pointer(nf_nat_sip_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_sip_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_sip_hook)); + BUG_ON(rcu_dereference(nf_nat_sdp_hook)); + rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); + rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp); + return 0; +} + +module_init(nf_nat_sip_init); +module_exit(nf_nat_sip_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 6b2eb26ae03f..e73891194b31 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -236,6 +236,18 @@ config NF_CONNTRACK_PPTP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_SIP + tristate "SIP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + SIP is an application-layer control protocol that can establish, + modify, and terminate multimedia sessions (conferences) such as + Internet telephony calls. With the ip_conntrack_sip and + the nf_nat_sip modules you can support the protocol on a connection + tracking/NATing firewall. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 897bed4cbd79..86da21700aa3 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -29,6 +29,7 @@ obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o +obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c new file mode 100644 index 000000000000..e0c06795a943 --- /dev/null +++ b/net/netfilter/nf_conntrack_sip.c @@ -0,0 +1,530 @@ +/* SIP extension for IP connection tracking. + * + * (C) 2005 by Christian Hentschel + * based on RR's ip_conntrack_ftp.c and other modules. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Christian Hentschel "); +MODULE_DESCRIPTION("SIP connection tracking helper"); +MODULE_ALIAS("ip_conntrack_sip"); + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "port numbers of SIP servers"); + +static unsigned int sip_timeout __read_mostly = SIP_TIMEOUT; +module_param(sip_timeout, uint, 0600); +MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session"); + +unsigned int (*nf_nat_sip_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conn *ct, + const char **dptr) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sip_hook); + +unsigned int (*nf_nat_sdp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp, + const char *dptr) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); + +static int digits_len(struct nf_conn *, const char *, const char *, int *); +static int epaddr_len(struct nf_conn *, const char *, const char *, int *); +static int skp_digits_len(struct nf_conn *, const char *, const char *, int *); +static int skp_epaddr_len(struct nf_conn *, const char *, const char *, int *); + +struct sip_header_nfo { + const char *lname; + const char *sname; + const char *ln_str; + size_t lnlen; + size_t snlen; + size_t ln_strlen; + int case_sensitive; + int (*match_len)(struct nf_conn *, const char *, + const char *, int *); +}; + +static const struct sip_header_nfo ct_sip_hdrs[] = { + [POS_REG_REQ_URI] = { /* SIP REGISTER request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = epaddr_len, + }, + [POS_REQ_URI] = { /* SIP request URI */ + .lname = "sip:", + .lnlen = sizeof("sip:") - 1, + .ln_str = "@", + .ln_strlen = sizeof("@") - 1, + .match_len = epaddr_len, + }, + [POS_FROM] = { /* SIP From header */ + .lname = "From:", + .lnlen = sizeof("From:") - 1, + .sname = "\r\nf:", + .snlen = sizeof("\r\nf:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len, + }, + [POS_TO] = { /* SIP To header */ + .lname = "To:", + .lnlen = sizeof("To:") - 1, + .sname = "\r\nt:", + .snlen = sizeof("\r\nt:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len + }, + [POS_VIA] = { /* SIP Via header */ + .lname = "Via:", + .lnlen = sizeof("Via:") - 1, + .sname = "\r\nv:", + .snlen = sizeof("\r\nv:") - 1, /* rfc3261 "\r\n" */ + .ln_str = "UDP ", + .ln_strlen = sizeof("UDP ") - 1, + .match_len = epaddr_len, + }, + [POS_CONTACT] = { /* SIP Contact header */ + .lname = "Contact:", + .lnlen = sizeof("Contact:") - 1, + .sname = "\r\nm:", + .snlen = sizeof("\r\nm:") - 1, + .ln_str = "sip:", + .ln_strlen = sizeof("sip:") - 1, + .match_len = skp_epaddr_len + }, + [POS_CONTENT] = { /* SIP Content length header */ + .lname = "Content-Length:", + .lnlen = sizeof("Content-Length:") - 1, + .sname = "\r\nl:", + .snlen = sizeof("\r\nl:") - 1, + .ln_str = ":", + .ln_strlen = sizeof(":") - 1, + .match_len = skp_digits_len + }, + [POS_MEDIA] = { /* SDP media info */ + .case_sensitive = 1, + .lname = "\nm=", + .lnlen = sizeof("\nm=") - 1, + .sname = "\rm=", + .snlen = sizeof("\rm=") - 1, + .ln_str = "audio ", + .ln_strlen = sizeof("audio ") - 1, + .match_len = digits_len + }, + [POS_OWNER_IP4] = { /* SDP owner address*/ + .case_sensitive = 1, + .lname = "\no=", + .lnlen = sizeof("\no=") - 1, + .sname = "\ro=", + .snlen = sizeof("\ro=") - 1, + .ln_str = "IN IP4 ", + .ln_strlen = sizeof("IN IP4 ") - 1, + .match_len = epaddr_len + }, + [POS_CONNECTION_IP4] = {/* SDP connection info */ + .case_sensitive = 1, + .lname = "\nc=", + .lnlen = sizeof("\nc=") - 1, + .sname = "\rc=", + .snlen = sizeof("\rc=") - 1, + .ln_str = "IN IP4 ", + .ln_strlen = sizeof("IN IP4 ") - 1, + .match_len = epaddr_len + }, + [POS_OWNER_IP6] = { /* SDP owner address*/ + .case_sensitive = 1, + .lname = "\no=", + .lnlen = sizeof("\no=") - 1, + .sname = "\ro=", + .snlen = sizeof("\ro=") - 1, + .ln_str = "IN IP6 ", + .ln_strlen = sizeof("IN IP6 ") - 1, + .match_len = epaddr_len + }, + [POS_CONNECTION_IP6] = {/* SDP connection info */ + .case_sensitive = 1, + .lname = "\nc=", + .lnlen = sizeof("\nc=") - 1, + .sname = "\rc=", + .snlen = sizeof("\rc=") - 1, + .ln_str = "IN IP6 ", + .ln_strlen = sizeof("IN IP6 ") - 1, + .match_len = epaddr_len + }, + [POS_SDP_HEADER] = { /* SDP version header */ + .case_sensitive = 1, + .lname = "\nv=", + .lnlen = sizeof("\nv=") - 1, + .sname = "\rv=", + .snlen = sizeof("\rv=") - 1, + .ln_str = "=", + .ln_strlen = sizeof("=") - 1, + .match_len = digits_len + } +}; + +/* get line lenght until first CR or LF seen. */ +int ct_sip_lnlen(const char *line, const char *limit) +{ + const char *k = line; + + while ((line <= limit) && (*line == '\r' || *line == '\n')) + line++; + + while (line <= limit) { + if (*line == '\r' || *line == '\n') + break; + line++; + } + return line - k; +} +EXPORT_SYMBOL_GPL(ct_sip_lnlen); + +/* Linear string search, case sensitive. */ +const char *ct_sip_search(const char *needle, const char *haystack, + size_t needle_len, size_t haystack_len, + int case_sensitive) +{ + const char *limit = haystack + (haystack_len - needle_len); + + while (haystack <= limit) { + if (case_sensitive) { + if (strncmp(haystack, needle, needle_len) == 0) + return haystack; + } else { + if (strnicmp(haystack, needle, needle_len) == 0) + return haystack; + } + haystack++; + } + return NULL; +} +EXPORT_SYMBOL_GPL(ct_sip_search); + +static int digits_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int len = 0; + while (dptr <= limit && isdigit(*dptr)) { + dptr++; + len++; + } + return len; +} + +/* get digits lenght, skiping blank spaces. */ +static int skp_digits_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + for (; dptr <= limit && *dptr == ' '; dptr++) + (*shift)++; + + return digits_len(ct, dptr, limit, shift); +} + +static int parse_addr(struct nf_conn *ct, const char *cp, const char **endp, + union nf_conntrack_address *addr, const char *limit) +{ + const char *end; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + int ret = 0; + + switch (family) { + case AF_INET: + ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end); + break; + case AF_INET6: + ret = in6_pton(cp, limit - cp, (u8 *)&addr->ip6, -1, &end); + break; + default: + BUG(); + } + + if (ret == 0 || end == cp) + return 0; + if (endp) + *endp = end; + return 1; +} + +/* skip ip address. returns its length. */ +static int epaddr_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + union nf_conntrack_address addr; + const char *aux = dptr; + + if (!parse_addr(ct, dptr, &dptr, &addr, limit)) { + DEBUGP("ip: %s parse failed.!\n", dptr); + return 0; + } + + /* Port number */ + if (*dptr == ':') { + dptr++; + dptr += digits_len(ct, dptr, limit, shift); + } + return dptr - aux; +} + +/* get address length, skiping user info. */ +static int skp_epaddr_len(struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int s = *shift; + + for (; dptr <= limit && *dptr != '@'; dptr++) + (*shift)++; + + if (*dptr == '@') { + dptr++; + (*shift)++; + } else + *shift = s; + + return epaddr_len(ct, dptr, limit, shift); +} + +/* Returns 0 if not found, -1 error parsing. */ +int ct_sip_get_info(struct nf_conn *ct, + const char *dptr, size_t dlen, + unsigned int *matchoff, + unsigned int *matchlen, + enum sip_header_pos pos) +{ + const struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos]; + const char *limit, *aux, *k = dptr; + int shift = 0; + + limit = dptr + (dlen - hnfo->lnlen); + + while (dptr <= limit) { + if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) && + (strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) { + dptr++; + continue; + } + aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen, + ct_sip_lnlen(dptr, limit), + hnfo->case_sensitive); + if (!aux) { + DEBUGP("'%s' not found in '%s'.\n", hnfo->ln_str, + hnfo->lname); + return -1; + } + aux += hnfo->ln_strlen; + + *matchlen = hnfo->match_len(ct, aux, limit, &shift); + if (!*matchlen) + return -1; + + *matchoff = (aux - k) + shift; + + DEBUGP("%s match succeeded! - len: %u\n", hnfo->lname, + *matchlen); + return 1; + } + DEBUGP("%s header not found.\n", hnfo->lname); + return 0; +} +EXPORT_SYMBOL_GPL(ct_sip_get_info); + +static int set_expected_rtp(struct sk_buff **pskb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + union nf_conntrack_address *addr, + __be16 port, + const char *dptr) +{ + struct nf_conntrack_expect *exp; + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + int family = ct->tuplehash[!dir].tuple.src.l3num; + int ret; + typeof(nf_nat_sdp_hook) nf_nat_sdp; + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) + return NF_DROP; + nf_conntrack_expect_init(exp, family, + &ct->tuplehash[!dir].tuple.src.u3, addr, + IPPROTO_UDP, NULL, &port); + + nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); + if (nf_nat_sdp && ct->status & IPS_NAT_MASK) + ret = nf_nat_sdp(pskb, ctinfo, exp, dptr); + else { + if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + else + ret = NF_ACCEPT; + } + nf_conntrack_expect_put(exp); + + return ret; +} + +static int sip_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + union nf_conntrack_address addr; + unsigned int dataoff, datalen; + const char *dptr; + int ret = NF_ACCEPT; + int matchoff, matchlen; + u_int16_t port; + enum sip_header_pos pos; + typeof(nf_nat_sip_hook) nf_nat_sip; + + /* No Data ? */ + dataoff = protoff + sizeof(struct udphdr); + if (dataoff >= (*pskb)->len) + return NF_ACCEPT; + + nf_ct_refresh(ct, *pskb, sip_timeout * HZ); + + if (!skb_is_nonlinear(*pskb)) + dptr = (*pskb)->data + dataoff; + else { + DEBUGP("Copy of skbuff not supported yet.\n"); + goto out; + } + + nf_nat_sip = rcu_dereference(nf_nat_sip_hook); + if (nf_nat_sip && ct->status & IPS_NAT_MASK) { + if (!nf_nat_sip(pskb, ctinfo, ct, &dptr)) { + ret = NF_DROP; + goto out; + } + } + + datalen = (*pskb)->len - dataoff; + if (datalen < sizeof("SIP/2.0 200") - 1) + goto out; + + /* RTP info only in some SDP pkts */ + if (memcmp(dptr, "INVITE", sizeof("INVITE") - 1) != 0 && + memcmp(dptr, "SIP/2.0 200", sizeof("SIP/2.0 200") - 1) != 0) { + goto out; + } + /* Get address and port from SDP packet. */ + pos = family == AF_INET ? POS_CONNECTION_IP4 : POS_CONNECTION_IP6; + if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, pos) > 0) { + + /* We'll drop only if there are parse problems. */ + if (!parse_addr(ct, dptr + matchoff, NULL, &addr, + dptr + datalen)) { + ret = NF_DROP; + goto out; + } + if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, + POS_MEDIA) > 0) { + + port = simple_strtoul(dptr + matchoff, NULL, 10); + if (port < 1024) { + ret = NF_DROP; + goto out; + } + ret = set_expected_rtp(pskb, ct, ctinfo, &addr, + htons(port), dptr); + } + } +out: + return ret; +} + +static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; +static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; + +static void nf_conntrack_sip_fini(void) +{ + int i, j; + + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) { + if (sip[i][j].me == NULL) + continue; + nf_conntrack_helper_unregister(&sip[i][j]); + } + } +} + +static int __init nf_conntrack_sip_init(void) +{ + int i, j, ret; + char *tmpname; + + if (ports_c == 0) + ports[ports_c++] = SIP_PORT; + + for (i = 0; i < ports_c; i++) { + memset(&sip[i], 0, sizeof(sip[i])); + + sip[i][0].tuple.src.l3num = AF_INET; + sip[i][1].tuple.src.l3num = AF_INET6; + for (j = 0; j < 2; j++) { + sip[i][j].tuple.dst.protonum = IPPROTO_UDP; + sip[i][j].tuple.src.u.udp.port = htons(ports[i]); + sip[i][j].mask.src.l3num = 0xFFFF; + sip[i][j].mask.src.u.udp.port = htons(0xFFFF); + sip[i][j].mask.dst.protonum = 0xFF; + sip[i][j].max_expected = 2; + sip[i][j].timeout = 3 * 60; /* 3 minutes */ + sip[i][j].me = THIS_MODULE; + sip[i][j].help = sip_help; + + tmpname = &sip_names[i][j][0]; + if (ports[i] == SIP_PORT) + sprintf(tmpname, "sip"); + else + sprintf(tmpname, "sip-%u", i); + sip[i][j].name = tmpname; + + DEBUGP("port #%u: %u\n", i, ports[i]); + + ret = nf_conntrack_helper_register(&sip[i][j]); + if (ret) { + printk("nf_ct_sip: failed to register helper " + "for pf: %u port: %u\n", + sip[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_sip_fini(); + return ret; + } + } + } + return 0; +} + +module_init(nf_conntrack_sip_init); +module_exit(nf_conntrack_sip_fini); -- cgit v1.2.3 From a536df35b3a58caa9015bf7887a374b20f658368 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Sat, 2 Dec 2006 22:10:18 -0800 Subject: [NETFILTER]: nf_conntrack/nf_nat: add TFTP helper port Add IPv4 and IPv6 capable nf_conntrack port of the TFTP conntrack/NAT helper. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_tftp.h | 20 ++++ net/ipv4/netfilter/Kconfig | 5 + net/ipv4/netfilter/Makefile | 1 + net/ipv4/netfilter/nf_nat_tftp.c | 52 +++++++++ net/netfilter/Kconfig | 11 ++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_tftp.c | 159 ++++++++++++++++++++++++++++ 7 files changed, 249 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_tftp.h create mode 100644 net/ipv4/netfilter/nf_nat_tftp.c create mode 100644 net/netfilter/nf_conntrack_tftp.c (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_tftp.h b/include/linux/netfilter/nf_conntrack_tftp.h new file mode 100644 index 000000000000..0d79b7ae051f --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_tftp.h @@ -0,0 +1,20 @@ +#ifndef _NF_CONNTRACK_TFTP_H +#define _NF_CONNTRACK_TFTP_H + +#define TFTP_PORT 69 + +struct tftphdr { + __be16 opcode; +}; + +#define TFTP_OPCODE_READ 1 +#define TFTP_OPCODE_WRITE 2 +#define TFTP_OPCODE_DATA 3 +#define TFTP_OPCODE_ACK 4 +#define TFTP_OPCODE_ERROR 5 + +extern unsigned int (*nf_nat_tftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp); + +#endif /* _NF_CONNTRACK_TFTP_H */ diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 83e83f553ccb..1c0018b6bea4 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -515,6 +515,11 @@ config IP_NF_NAT_TFTP default IP_NF_NAT if IP_NF_TFTP=y default m if IP_NF_TFTP=m +config NF_NAT_TFTP + tristate + depends on IP_NF_IPTABLES && NF_CONNTRACK && NF_NAT + default NF_NAT && NF_CONNTRACK_TFTP + config IP_NF_NAT_AMANDA tristate depends on IP_NF_IPTABLES!=n && IP_NF_CONNTRACK!=n && IP_NF_NAT!=n diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 167b65e89aca..3ca0af0ed5f0 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -56,6 +56,7 @@ obj-$(CONFIG_NF_NAT_H323) += nf_nat_h323.o obj-$(CONFIG_NF_NAT_IRC) += nf_nat_irc.o obj-$(CONFIG_NF_NAT_PPTP) += nf_nat_pptp.o obj-$(CONFIG_NF_NAT_SIP) += nf_nat_sip.o +obj-$(CONFIG_NF_NAT_TFTP) += nf_nat_tftp.o # NAT protocols (nf_nat) obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o diff --git a/net/ipv4/netfilter/nf_nat_tftp.c b/net/ipv4/netfilter/nf_nat_tftp.c new file mode 100644 index 000000000000..2566b79de224 --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_tftp.c @@ -0,0 +1,52 @@ +/* (C) 2001-2002 Magnus Boden + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Magnus Boden "); +MODULE_DESCRIPTION("TFTP NAT helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_nat_tftp"); + +static unsigned int help(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp) +{ + struct nf_conn *ct = exp->master; + + exp->saved_proto.udp.port + = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port; + exp->dir = IP_CT_DIR_REPLY; + exp->expectfn = nf_nat_follow_master; + if (nf_conntrack_expect_related(exp) != 0) + return NF_DROP; + return NF_ACCEPT; +} + +static void __exit nf_nat_tftp_fini(void) +{ + rcu_assign_pointer(nf_nat_tftp_hook, NULL); + synchronize_rcu(); +} + +static int __init nf_nat_tftp_init(void) +{ + BUG_ON(rcu_dereference(nf_nat_tftp_hook)); + rcu_assign_pointer(nf_nat_tftp_hook, help); + return 0; +} + +module_init(nf_nat_tftp_init); +module_exit(nf_nat_tftp_fini); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index e73891194b31..e0deac17df12 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -248,6 +248,17 @@ config NF_CONNTRACK_SIP To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_TFTP + tristate "TFTP protocol support (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + TFTP connection tracking helper, this is required depending + on how restrictive your ruleset is. + If you are using a tftp client behind -j SNAT or -j MASQUERADING + you will need this. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CT_NETLINK tristate 'Connection tracking netlink interface (EXPERIMENTAL)' depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 86da21700aa3..5dc5574f7e99 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -30,6 +30,7 @@ obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o +obj-$(CONFIG_NF_CONNTRACK_TFTP) += nf_conntrack_tftp.o # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c new file mode 100644 index 000000000000..d46a1263fcb5 --- /dev/null +++ b/net/netfilter/nf_conntrack_tftp.c @@ -0,0 +1,159 @@ +/* (C) 2001-2002 Magnus Boden + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Magnus Boden "); +MODULE_DESCRIPTION("TFTP connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ip_conntrack_tftp"); + +#define MAX_PORTS 8 +static unsigned short ports[MAX_PORTS]; +static int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); +MODULE_PARM_DESC(ports, "Port numbers of TFTP servers"); + +#if 0 +#define DEBUGP(format, args...) printk("%s:%s:" format, \ + __FILE__, __FUNCTION__ , ## args) +#else +#define DEBUGP(format, args...) +#endif + +unsigned int (*nf_nat_tftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + struct nf_conntrack_expect *exp) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_tftp_hook); + +static int tftp_help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + struct tftphdr _tftph, *tfh; + struct nf_conntrack_expect *exp; + struct nf_conntrack_tuple *tuple; + unsigned int ret = NF_ACCEPT; + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + typeof(nf_nat_tftp_hook) nf_nat_tftp; + + tfh = skb_header_pointer(*pskb, protoff + sizeof(struct udphdr), + sizeof(_tftph), &_tftph); + if (tfh == NULL) + return NF_ACCEPT; + + switch (ntohs(tfh->opcode)) { + case TFTP_OPCODE_READ: + case TFTP_OPCODE_WRITE: + /* RRQ and WRQ works the same way */ + DEBUGP(""); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) + return NF_DROP; + tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; + nf_conntrack_expect_init(exp, family, + &tuple->src.u3, &tuple->dst.u3, + IPPROTO_UDP, + NULL, &tuple->dst.u.udp.port); + + DEBUGP("expect: "); + NF_CT_DUMP_TUPLE(&exp->tuple); + NF_CT_DUMP_TUPLE(&exp->mask); + + nf_nat_tftp = rcu_dereference(nf_nat_tftp_hook); + if (nf_nat_tftp && ct->status & IPS_NAT_MASK) + ret = nf_nat_tftp(pskb, ctinfo, exp); + else if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + nf_conntrack_expect_put(exp); + break; + case TFTP_OPCODE_DATA: + case TFTP_OPCODE_ACK: + DEBUGP("Data/ACK opcode\n"); + break; + case TFTP_OPCODE_ERROR: + DEBUGP("Error opcode\n"); + break; + default: + DEBUGP("Unknown opcode\n"); + } + return ret; +} + +static struct nf_conntrack_helper tftp[MAX_PORTS][2] __read_mostly; +static char tftp_names[MAX_PORTS][2][sizeof("tftp-65535")] __read_mostly; + +static void nf_conntrack_tftp_fini(void) +{ + int i, j; + + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) + nf_conntrack_helper_unregister(&tftp[i][j]); + } +} + +static int __init nf_conntrack_tftp_init(void) +{ + int i, j, ret; + char *tmpname; + + if (ports_c == 0) + ports[ports_c++] = TFTP_PORT; + + for (i = 0; i < ports_c; i++) { + memset(&tftp[i], 0, sizeof(tftp[i])); + + tftp[i][0].tuple.src.l3num = AF_INET; + tftp[i][1].tuple.src.l3num = AF_INET6; + for (j = 0; j < 2; j++) { + tftp[i][j].tuple.dst.protonum = IPPROTO_UDP; + tftp[i][j].tuple.src.u.udp.port = htons(ports[i]); + tftp[i][j].mask.src.l3num = 0xFFFF; + tftp[i][j].mask.dst.protonum = 0xFF; + tftp[i][j].mask.src.u.udp.port = htons(0xFFFF); + tftp[i][j].max_expected = 1; + tftp[i][j].timeout = 5 * 60; /* 5 minutes */ + tftp[i][j].me = THIS_MODULE; + tftp[i][j].help = tftp_help; + + tmpname = &tftp_names[i][j][0]; + if (ports[i] == TFTP_PORT) + sprintf(tmpname, "tftp"); + else + sprintf(tmpname, "tftp-%u", i); + tftp[i][j].name = tmpname; + + ret = nf_conntrack_helper_register(&tftp[i][j]); + if (ret) { + printk("nf_ct_tftp: failed to register helper " + "for pf: %u port: %u\n", + tftp[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_tftp_fini(); + return ret; + } + } + } + return 0; +} + +module_init(nf_conntrack_tftp_init); +module_exit(nf_conntrack_tftp_fini); -- cgit v1.2.3 From 2b5f6dcce5bf94b9b119e9ed8d537098ec61c3d2 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Dec 2006 22:22:25 -0800 Subject: [XFRM]: Fix aevent structuring to be more complete. aevents can not uniquely identify an SA. We break the ABI with this patch, but consensus is that since it is not yet utilized by any (known) application then it is fine (better do it now than later). Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- Documentation/networking/xfrm_sync.txt | 5 ++++- include/linux/xfrm.h | 2 ++ net/xfrm/xfrm_user.c | 4 +++- 3 files changed, 9 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/Documentation/networking/xfrm_sync.txt b/Documentation/networking/xfrm_sync.txt index 8be626f7c0b8..d7aac9dedeb4 100644 --- a/Documentation/networking/xfrm_sync.txt +++ b/Documentation/networking/xfrm_sync.txt @@ -47,10 +47,13 @@ aevent_id structure looks like: struct xfrm_aevent_id { struct xfrm_usersa_id sa_id; + xfrm_address_t saddr; __u32 flags; + __u32 reqid; }; -xfrm_usersa_id in this message layout identifies the SA. +The unique SA is identified by the combination of xfrm_usersa_id, +reqid and saddr. flags are used to indicate different things. The possible flags are: diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 7907c42bd4e4..088ba8113f7e 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -289,7 +289,9 @@ struct xfrm_usersa_id { struct xfrm_aevent_id { struct xfrm_usersa_id sa_id; + xfrm_address_t saddr; __u32 flags; + __u32 reqid; }; struct xfrm_userspi_info { diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 055b72fb37bc..6f97665983d2 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1281,10 +1281,12 @@ static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, struct km_eve id = NLMSG_DATA(nlh); nlh->nlmsg_flags = 0; - id->sa_id.daddr = x->id.daddr; + memcpy(&id->sa_id.daddr, &x->id.daddr,sizeof(x->id.daddr)); id->sa_id.spi = x->id.spi; id->sa_id.family = x->props.family; id->sa_id.proto = x->id.proto; + memcpy(&id->saddr, &x->props.saddr,sizeof(x->props.saddr)); + id->reqid = x->props.reqid; id->flags = c->data.aevent; RTA_PUT(skb, XFRMA_REPLAY_VAL, sizeof(x->replay), &x->replay); -- cgit v1.2.3 From 582982e6991d6718ddadf8751072b50a850dde48 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 17 Nov 2006 12:05:11 +0900 Subject: [PATCH] libata: remove unused HSM_ST_UNKNOWN HSM_ST_UNKNOWN is not used anywhere. Its value is zero and supposed to serve sanity check purpose but HSM_ST_IDLE is used for that purpose. This unused state causes confusion. After a port is initialized but before the first command is executed, the idle hsm state is UNKNOWN. However, once a command has completed, the idle hsm state is IDLE. This defeats sanity check in ata_pio_task() for the first command. This patch removes HSM_ST_UNKNOWN and consequently make HSM_ST_IDLE the default state. Signed-off-by: Tejun Heo --- include/linux/libata.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/libata.h b/include/linux/libata.h index 9080789913f7..6013211ac7de 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -318,7 +318,6 @@ enum { }; enum hsm_task_states { - HSM_ST_UNKNOWN, /* state unknown */ HSM_ST_IDLE, /* no command on going */ HSM_ST, /* (waiting the device to) transfer data */ HSM_ST_LAST, /* (waiting the device to) complete command */ -- cgit v1.2.3 From 3d3cca37559e3ab2b574eda11ed5207ccdb8980a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 16 Nov 2006 10:50:50 +0900 Subject: [PATCH] libata: implement ATA_FLAG_SETXFER_POLLING and use it in pata_via, take #2 This patch implements ATA_FLAG_SETXFER_POLLING and use in pata_via. If this flag is set, transfer mode setting performed by polling not by interrupt. This should help those controllers which raise interrupt before the command is actually complete on SETXFER. Rationale for this approach. * uses existing facility and relatively simple * no busy sleep in the interrupt handler * updating drivers is easy While at it, kill now unused flag ATA_FLAG_SRST in pata_via. Signed-off-by: Tejun Heo --- drivers/ata/libata-core.c | 8 ++++++++ drivers/ata/pata_via.c | 12 ++++++------ include/linux/libata.h | 1 + 3 files changed, 15 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 3fd7c7932707..b35fdcb104ec 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4821,6 +4821,14 @@ unsigned int ata_qc_issue_prot(struct ata_queued_cmd *qc) } } + /* Some controllers show flaky interrupt behavior after + * setting xfer mode. Use polling instead. + */ + if (unlikely(qc->tf.command == ATA_CMD_SET_FEATURES && + qc->tf.feature == SETFEATURES_XFER) && + (ap->flags & ATA_FLAG_SETXFER_POLLING)) + qc->tf.flags |= ATA_TFLAG_POLLING; + /* select the device */ ata_dev_select(ap, qc->dev->devno, 1, 0); diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index ee93dd0169be..a9077b617b8d 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -418,7 +418,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* Early VIA without UDMA support */ static struct ata_port_info via_mwdma_info = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .port_ops = &via_port_ops @@ -426,7 +426,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* Ditto with IRQ masking required */ static struct ata_port_info via_mwdma_info_borked = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .port_ops = &via_port_ops_noirq, @@ -434,7 +434,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* VIA UDMA 33 devices (and borked 66) */ static struct ata_port_info via_udma33_info = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .udma_mask = 0x7, @@ -443,7 +443,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* VIA UDMA 66 devices */ static struct ata_port_info via_udma66_info = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .udma_mask = 0x1f, @@ -452,7 +452,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* VIA UDMA 100 devices */ static struct ata_port_info via_udma100_info = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .udma_mask = 0x3f, @@ -461,7 +461,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id) /* UDMA133 with bad AST (All current 133) */ static struct ata_port_info via_udma133_info = { .sht = &via_sht, - .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SRST, + .flags = ATA_FLAG_SLAVE_POSS | ATA_FLAG_SETXFER_POLLING, .pio_mask = 0x1f, .mwdma_mask = 0x07, .udma_mask = 0x7f, /* FIXME: should check north bridge */ diff --git a/include/linux/libata.h b/include/linux/libata.h index 6013211ac7de..8b57b6a806cc 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -178,6 +178,7 @@ enum { ATA_FLAG_DEBUGMSG = (1 << 13), ATA_FLAG_DETECT_POLLING = (1 << 14), /* detect device presence by * polling IDENTIFY */ + ATA_FLAG_SETXFER_POLLING= (1 << 15), /* use polling for SETXFER */ /* The following flag belongs to ap->pflags but is kept in * ap->flags because it's referenced in many LLDs and will be -- cgit v1.2.3 From 800b399669ad495ad4361d134df87401ae36f44f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 3 Dec 2006 21:34:13 +0900 Subject: [PATCH] libata: always use polling IDENTIFY libata switched to IRQ-driven IDENTIFY when IRQ-driven PIO was introduced. This has caused a lot of problems including device misdetection and phantom device. ATA_FLAG_DETECT_POLLING was added recently to selectively use polling IDENTIFY on problemetic drivers but many controllers and devices are affected by this problem and trying to adding ATA_FLAG_DETECT_POLLING for each such case is diffcult and not very rewarding. This patch makes libata always use polling IDENTIFY. This is consistent with libata's original behavior and drivers/ide's behavior. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ata_piix.c | 5 ++--- drivers/ata/libata-core.c | 8 ++------ drivers/ata/libata-eh.c | 3 --- drivers/ata/libata.h | 2 -- include/linux/libata.h | 4 +--- 5 files changed, 5 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 788a269206e6..a2d84f7cf2bc 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -105,9 +105,8 @@ enum { PIIX_FLAG_AHCI = (1 << 27), /* AHCI possible */ PIIX_FLAG_CHECKINTR = (1 << 28), /* make sure PCI INTx enabled */ - PIIX_PATA_FLAGS = ATA_FLAG_SLAVE_POSS | ATA_FLAG_DETECT_POLLING, - PIIX_SATA_FLAGS = ATA_FLAG_SATA | PIIX_FLAG_CHECKINTR | - ATA_FLAG_DETECT_POLLING, + PIIX_PATA_FLAGS = ATA_FLAG_SLAVE_POSS, + PIIX_SATA_FLAGS = ATA_FLAG_SATA | PIIX_FLAG_CHECKINTR, /* combined mode. if set, PATA is channel 0. * if clear, PATA is channel 1. diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 883276388207..f8ec3896b793 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1473,16 +1473,12 @@ int ata_dev_read_id(struct ata_device *dev, unsigned int *p_class, } tf.protocol = ATA_PROT_PIO; - - /* presence detection using polling IDENTIFY? */ - if (flags & ATA_READID_DETECT) - tf.flags |= ATA_TFLAG_POLLING; + tf.flags |= ATA_TFLAG_POLLING; /* for polling presence detection */ err_mask = ata_exec_internal(dev, &tf, NULL, DMA_FROM_DEVICE, id, sizeof(id[0]) * ATA_ID_WORDS); if (err_mask) { - if ((flags & ATA_READID_DETECT) && - (err_mask & AC_ERR_NODEV_HINT)) { + if (err_mask & AC_ERR_NODEV_HINT) { DPRINTK("ata%u.%d: NODEV after polling detection\n", ap->id, dev->devno); return -ENOENT; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 2aad7b79d6dd..76a85dfb7307 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1692,9 +1692,6 @@ static int ata_eh_revalidate_and_attach(struct ata_port *ap, ata_class_enabled(ehc->classes[dev->devno])) { dev->class = ehc->classes[dev->devno]; - if (ap->flags & ATA_FLAG_DETECT_POLLING) - readid_flags |= ATA_READID_DETECT; - rc = ata_dev_read_id(dev, &dev->class, readid_flags, dev->id); if (rc == 0) { diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index 1ff3f59504c9..107b2b565229 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -42,8 +42,6 @@ struct ata_scsi_args { enum { /* flags for ata_dev_read_id() */ ATA_READID_POSTRESET = (1 << 0), /* reading ID after reset */ - ATA_READID_DETECT = (1 << 1), /* perform presence detection - * using polling IDENTIFY */ }; extern struct workqueue_struct *ata_aux_wq; diff --git a/include/linux/libata.h b/include/linux/libata.h index 8b57b6a806cc..202283b5df96 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -176,9 +176,7 @@ enum { ATA_FLAG_SKIP_D2H_BSY = (1 << 12), /* can't wait for the first D2H * Register FIS clearing BSY */ ATA_FLAG_DEBUGMSG = (1 << 13), - ATA_FLAG_DETECT_POLLING = (1 << 14), /* detect device presence by - * polling IDENTIFY */ - ATA_FLAG_SETXFER_POLLING= (1 << 15), /* use polling for SETXFER */ + ATA_FLAG_SETXFER_POLLING= (1 << 14), /* use polling for SETXFER */ /* The following flag belongs to ap->pflags but is kept in * ap->flags because it's referenced in many LLDs and will be -- cgit v1.2.3 From d916faace3efc0bf19fe9a615a1ab8fa1a24cd93 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Sun, 3 Dec 2006 22:22:41 -0500 Subject: Remove long-unmaintained ftape driver subsystem. It's bitrotten, long unmaintained, long hidden under BROKEN_ON_SMP, etc. As scheduled in feature-removal-schedule.txt, and ack'd several times on lkml. Signed-off-by: Jeff Garzik --- Documentation/00-INDEX | 2 - Documentation/feature-removal-schedule.txt | 8 - Documentation/ftape.txt | 307 ----- Documentation/kernel-parameters.txt | 3 - MAINTAINERS | 5 - drivers/char/Kconfig | 33 - drivers/char/Makefile | 1 - drivers/char/ftape/Kconfig | 330 ------ drivers/char/ftape/Makefile | 28 - drivers/char/ftape/README.PCI | 81 -- drivers/char/ftape/RELEASE-NOTES | 966 --------------- drivers/char/ftape/compressor/Makefile | 31 - drivers/char/ftape/compressor/lzrw3.c | 743 ------------ drivers/char/ftape/compressor/lzrw3.h | 253 ---- drivers/char/ftape/compressor/zftape-compress.c | 1203 ------------------- drivers/char/ftape/compressor/zftape-compress.h | 83 -- drivers/char/ftape/lowlevel/Makefile | 43 - drivers/char/ftape/lowlevel/fc-10.c | 175 --- drivers/char/ftape/lowlevel/fc-10.h | 39 - drivers/char/ftape/lowlevel/fdc-io.c | 1349 --------------------- drivers/char/ftape/lowlevel/fdc-io.h | 252 ---- drivers/char/ftape/lowlevel/fdc-isr.c | 1170 ------------------- drivers/char/ftape/lowlevel/fdc-isr.h | 55 - drivers/char/ftape/lowlevel/ftape-bsm.c | 491 -------- drivers/char/ftape/lowlevel/ftape-bsm.h | 66 -- drivers/char/ftape/lowlevel/ftape-buffer.c | 130 --- drivers/char/ftape/lowlevel/ftape-buffer.h | 32 - drivers/char/ftape/lowlevel/ftape-calibr.c | 275 ----- drivers/char/ftape/lowlevel/ftape-calibr.h | 37 - drivers/char/ftape/lowlevel/ftape-ctl.c | 896 -------------- drivers/char/ftape/lowlevel/ftape-ctl.h | 162 --- drivers/char/ftape/lowlevel/ftape-ecc.c | 853 -------------- drivers/char/ftape/lowlevel/ftape-ecc.h | 84 -- drivers/char/ftape/lowlevel/ftape-format.c | 344 ------ drivers/char/ftape/lowlevel/ftape-format.h | 37 - drivers/char/ftape/lowlevel/ftape-init.c | 160 --- drivers/char/ftape/lowlevel/ftape-init.h | 43 - drivers/char/ftape/lowlevel/ftape-io.c | 992 ---------------- drivers/char/ftape/lowlevel/ftape-io.h | 90 -- drivers/char/ftape/lowlevel/ftape-proc.c | 214 ---- drivers/char/ftape/lowlevel/ftape-proc.h | 35 - drivers/char/ftape/lowlevel/ftape-read.c | 621 ---------- drivers/char/ftape/lowlevel/ftape-read.h | 51 - drivers/char/ftape/lowlevel/ftape-rw.c | 1092 ----------------- drivers/char/ftape/lowlevel/ftape-rw.h | 111 -- drivers/char/ftape/lowlevel/ftape-setup.c | 104 -- drivers/char/ftape/lowlevel/ftape-tracing.c | 118 -- drivers/char/ftape/lowlevel/ftape-tracing.h | 179 --- drivers/char/ftape/lowlevel/ftape-write.c | 336 ------ drivers/char/ftape/lowlevel/ftape-write.h | 53 - drivers/char/ftape/lowlevel/ftape_syms.c | 87 -- drivers/char/ftape/zftape/Makefile | 36 - drivers/char/ftape/zftape/zftape-buffers.c | 149 --- drivers/char/ftape/zftape/zftape-buffers.h | 55 - drivers/char/ftape/zftape/zftape-ctl.c | 1417 ----------------------- drivers/char/ftape/zftape/zftape-ctl.h | 58 - drivers/char/ftape/zftape/zftape-eof.c | 199 ---- drivers/char/ftape/zftape/zftape-eof.h | 52 - drivers/char/ftape/zftape/zftape-init.c | 377 ------ drivers/char/ftape/zftape/zftape-init.h | 77 -- drivers/char/ftape/zftape/zftape-read.c | 377 ------ drivers/char/ftape/zftape/zftape-read.h | 53 - drivers/char/ftape/zftape/zftape-rw.c | 375 ------ drivers/char/ftape/zftape/zftape-rw.h | 101 -- drivers/char/ftape/zftape/zftape-vtbl.c | 757 ------------ drivers/char/ftape/zftape/zftape-vtbl.h | 227 ---- drivers/char/ftape/zftape/zftape-write.c | 483 -------- drivers/char/ftape/zftape/zftape-write.h | 38 - drivers/char/ftape/zftape/zftape_syms.c | 43 - include/linux/Kbuild | 4 - include/linux/ftape-header-segment.h | 122 -- include/linux/ftape-vendors.h | 137 --- include/linux/ftape.h | 201 ---- include/linux/zftape.h | 87 -- 74 files changed, 20278 deletions(-) delete mode 100644 Documentation/ftape.txt delete mode 100644 drivers/char/ftape/Kconfig delete mode 100644 drivers/char/ftape/Makefile delete mode 100644 drivers/char/ftape/README.PCI delete mode 100644 drivers/char/ftape/RELEASE-NOTES delete mode 100644 drivers/char/ftape/compressor/Makefile delete mode 100644 drivers/char/ftape/compressor/lzrw3.c delete mode 100644 drivers/char/ftape/compressor/lzrw3.h delete mode 100644 drivers/char/ftape/compressor/zftape-compress.c delete mode 100644 drivers/char/ftape/compressor/zftape-compress.h delete mode 100644 drivers/char/ftape/lowlevel/Makefile delete mode 100644 drivers/char/ftape/lowlevel/fc-10.c delete mode 100644 drivers/char/ftape/lowlevel/fc-10.h delete mode 100644 drivers/char/ftape/lowlevel/fdc-io.c delete mode 100644 drivers/char/ftape/lowlevel/fdc-io.h delete mode 100644 drivers/char/ftape/lowlevel/fdc-isr.c delete mode 100644 drivers/char/ftape/lowlevel/fdc-isr.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-bsm.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-bsm.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-buffer.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-buffer.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-calibr.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-calibr.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-ctl.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-ctl.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-ecc.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-ecc.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-format.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-format.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-init.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-init.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-io.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-io.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-proc.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-proc.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-read.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-read.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-rw.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-rw.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-setup.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-tracing.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-tracing.h delete mode 100644 drivers/char/ftape/lowlevel/ftape-write.c delete mode 100644 drivers/char/ftape/lowlevel/ftape-write.h delete mode 100644 drivers/char/ftape/lowlevel/ftape_syms.c delete mode 100644 drivers/char/ftape/zftape/Makefile delete mode 100644 drivers/char/ftape/zftape/zftape-buffers.c delete mode 100644 drivers/char/ftape/zftape/zftape-buffers.h delete mode 100644 drivers/char/ftape/zftape/zftape-ctl.c delete mode 100644 drivers/char/ftape/zftape/zftape-ctl.h delete mode 100644 drivers/char/ftape/zftape/zftape-eof.c delete mode 100644 drivers/char/ftape/zftape/zftape-eof.h delete mode 100644 drivers/char/ftape/zftape/zftape-init.c delete mode 100644 drivers/char/ftape/zftape/zftape-init.h delete mode 100644 drivers/char/ftape/zftape/zftape-read.c delete mode 100644 drivers/char/ftape/zftape/zftape-read.h delete mode 100644 drivers/char/ftape/zftape/zftape-rw.c delete mode 100644 drivers/char/ftape/zftape/zftape-rw.h delete mode 100644 drivers/char/ftape/zftape/zftape-vtbl.c delete mode 100644 drivers/char/ftape/zftape/zftape-vtbl.h delete mode 100644 drivers/char/ftape/zftape/zftape-write.c delete mode 100644 drivers/char/ftape/zftape/zftape-write.h delete mode 100644 drivers/char/ftape/zftape/zftape_syms.c delete mode 100644 include/linux/ftape-header-segment.h delete mode 100644 include/linux/ftape-vendors.h delete mode 100644 include/linux/ftape.h delete mode 100644 include/linux/zftape.h (limited to 'include/linux') diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX index 02457ec9c94f..f08ca9535733 100644 --- a/Documentation/00-INDEX +++ b/Documentation/00-INDEX @@ -104,8 +104,6 @@ firmware_class/ - request_firmware() hotplug interface info. floppy.txt - notes and driver options for the floppy disk driver. -ftape.txt - - notes about the floppy tape device driver. hayes-esp.txt - info on using the Hayes ESP serial driver. highuid.txt diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index f81819364b7a..226ecf2ffd56 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -234,14 +234,6 @@ Who: Jean Delvare --------------------------- -What: ftape -When: 2.6.20 -Why: Orphaned for ages. SMP bugs long unfixed. Few users left - in the world. -Who: Jeff Garzik - ---------------------------- - What: IPv4 only connection tracking/NAT/helpers When: 2.6.22 Why: The new layer 3 independant connection tracking replaces the old diff --git a/Documentation/ftape.txt b/Documentation/ftape.txt deleted file mode 100644 index 7d8bb3384031..000000000000 --- a/Documentation/ftape.txt +++ /dev/null @@ -1,307 +0,0 @@ -Intro -===== - -This file describes some issues involved when using the "ftape" -floppy tape device driver that comes with the Linux kernel. - -ftape has a home page at - -http://ftape.dot-heine.de/ - -which contains further information about ftape. Please cross check -this WWW address against the address given (if any) in the MAINTAINERS -file located in the top level directory of the Linux kernel source -tree. - -NOTE: This is an unmaintained set of drivers, and it is not guaranteed to work. -If you are interested in taking over maintenance, contact Claus-Justus Heine -, the former maintainer. - -Contents -======== - -A minus 1: Ftape documentation - -A. Changes - 1. Goal - 2. I/O Block Size - 3. Write Access when not at EOD (End Of Data) or BOT (Begin Of Tape) - 4. Formatting - 5. Interchanging cartridges with other operating systems - -B. Debugging Output - 1. Introduction - 2. Tuning the debugging output - -C. Boot and load time configuration - 1. Setting boot time parameters - 2. Module load time parameters - 3. Ftape boot- and load time options - 4. Example kernel parameter setting - 5. Example module parameter setting - -D. Support and contacts - -******************************************************************************* - -A minus 1. Ftape documentation -============================== - -Unluckily, the ftape-HOWTO is out of date. This really needs to be -changed. Up to date documentation as well as recent development -versions of ftape and useful links to related topics can be found at -the ftape home page at - -http://ftape.dot-heine.de/ - -******************************************************************************* - -A. Changes -========== - -1. Goal - ~~~~ - The goal of all that incompatibilities was to give ftape an interface - that resembles the interface provided by SCSI tape drives as close - as possible. Thus any Unix backup program that is known to work - with SCSI tape drives should also work. - - The concept of a fixed block size for read/write transfers is - rather unrelated to this SCSI tape compatibility at the file system - interface level. It developed out of a feature of zftape, a - block wise user transparent on-the-fly compression. That compression - support will not be dropped in future releases for compatibility - reasons with previous releases of zftape. - -2. I/O Block Size - ~~~~~~~~~~~~~~ - The block size defaults to 10k which is the default block size of - GNU tar. - - The block size can be tuned either during kernel configuration or - at runtime with the MTIOCTOP ioctl using the MTSETBLK operation - (i.e. do "mt -f /dev/qft0" setblk #BLKSZ). A block size of 0 - switches to variable block size mode i.e. "mt setblk 0" switches - off the block size restriction. However, this disables zftape's - built in on-the-fly compression which doesn't work with variable - block size mode. - - The BLKSZ parameter must be given as a byte count and must be a - multiple of 32k or 0, i.e. use "mt setblk 32768" to switch to a - block size of 32k. - - The typical symptom of a block size mismatch is an "invalid - argument" error message. - -3. Write Access when not at EOD (End Of Data) or BOT (Begin Of Tape) - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - zftape (the file system interface of ftape-3.x) denies write access - to the tape cartridge when it isn't positioned either at BOT or - EOD. - -4. Formatting - ~~~~~~~~~~ - ftape DOES support formatting of floppy tape cartridges. You need the - `ftformat' program that is shipped with the modules version of ftape. - Please get the latest version of ftape from - - ftp://sunsite.unc.edu/pub/Linux/kernel/tapes - - or from the ftape home page at - - http://ftape.dot-heine.de/ - - `ftformat' is contained in the `./contrib/' subdirectory of that - separate ftape package. - -5. Interchanging cartridges with other operating systems - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - - The internal emulation of Unix tape device file marks has changed - completely. ftape now uses the volume table segment as specified - by the QIC-40/80/3010/3020/113 standards to emulate file marks. As - a consequence there is limited support to interchange cartridges - with other operating systems. - - To be more precise: ftape will detect volumes written by other OS's - programs and other OS's programs will detect volumes written by - ftape. - - However, it isn't possible to extract the data dumped to the tape - by some MSDOS program with ftape. This exceeds the scope of a - kernel device driver. If you need such functionality, then go ahead - and write a user space utility that is able to do that. ftape already - provides all kernel level support necessary to do that. - -******************************************************************************* - -B. Debugging Output - ================ - -1. Introduction - ~~~~~~~~~~~~ - The ftape driver can be very noisy in that is can print lots of - debugging messages to the kernel log files and the system console. - While this is useful for debugging it might be annoying during - normal use and enlarges the size of the driver by several kilobytes. - - To reduce the size of the driver you can trim the maximal amount of - debugging information available during kernel configuration. Please - refer to the kernel configuration script and its on-line help - functionality. - - The amount of debugging output maps to the "tracing" boot time - option and the "ft_tracing" modules option as follows: - - 0 bugs - 1 + errors (with call-stack dump) - 2 + warnings - 3 + information - 4 + more information - 5 + program flow - 6 + fdc/dma info - 7 + data flow - 8 + everything else - -2. Tuning the debugging output - ~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To reduce the amount of debugging output printed to the system - console you can - - i) trim the debugging output at run-time with - - mt -f /dev/nqft0 setdensity #DBGLVL - - where "#DBGLVL" is a number between 0 and 9 - - ii) trim the debugging output at module load time with - - modprobe ftape ft_tracing=#DBGLVL - - Of course, this applies only if you have configured ftape to be - compiled as a module. - - iii) trim the debugging output during system boot time. Add the - following to the kernel command line: - - ftape=#DBGLVL,tracing - - Please refer also to the next section if you don't know how to - set boot time parameters. - -******************************************************************************* - -C. Boot and load time configuration - ================================ - -1. Setting boot time parameters - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Assuming that you use lilo, the LI)nux LO)ader, boot time kernel - parameters can be set by adding a line - - append some_kernel_boot_time_parameter - - to `/etc/lilo.conf' or at real boot time by typing in the options - at the prompt provided by LILO. I can't give you advice on how to - specify those parameters with other loaders as I don't use them. - - For ftape, each "some_kernel_boot_time_parameter" looks like - "ftape=value,option". As an example, the debugging output can be - increased with - - ftape=4,tracing - - NOTE: the value precedes the option name. - -2. Module load time parameters - ~~~~~~~~~~~~~~~~~~~~~~~~~~~ - Module parameters can be specified either directly when invoking - the program 'modprobe' at the shell prompt: - - modprobe ftape ft_tracing=4 - - or by editing the file `/etc/modprobe.conf' in which case they take - effect each time when the module is loaded with `modprobe' (please - refer to the respective manual pages). Thus, you should add a line - - options ftape ft_tracing=4 - - to `/etc/modprobe.conf` if you intend to increase the debugging - output of the driver. - - -3. Ftape boot- and load time options - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - - i. Controlling the amount of debugging output - DBGLVL has to be replaced by a number between 0 and 8. - - module | kernel command line - -----------------------|---------------------- - ft_tracing=DBGLVL | ftape=DBGLVL,tracing - - ii. Hardware setup - BASE is the base address of your floppy disk controller, - IRQ and DMA give its interrupt and DMA channel, respectively. - BOOL is an integer, "0" means "no"; any other value means - "yes". You don't need to specify anything if connecting your tape - drive to the standard floppy disk controller. All of these - values have reasonable defaults. The defaults can be modified - during kernel configuration, i.e. while running "make config", - "make menuconfig" or "make xconfig" in the top level directory - of the Linux kernel source tree. Please refer also to the on - line documentation provided during that kernel configuration - process. - - ft_probe_fc10 is set to a non-zero value if you wish for ftape to - probe for a Colorado FC-10 or FC-20 controller. - - ft_mach2 is set to a non-zero value if you wish for ftape to probe - for a Mountain MACH-2 controller. - - module | kernel command line - -----------------------|---------------------- - ft_fdc_base=BASE | ftape=BASE,ioport - ft_fdc_irq=IRQ | ftape=IRQ,irq - ft_fdc_dma=DMA | ftape=DMA,dma - ft_probe_fc10=BOOL | ftape=BOOL,fc10 - ft_mach2=BOOL | ftape=BOOL,mach2 - ft_fdc_threshold=THR | ftape=THR,threshold - ft_fdc_rate_limit=RATE | ftape=RATE,datarate - -4. Example kernel parameter setting - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To configure ftape to probe for a Colorado FC-10/FC-20 controller - and to increase the amount of debugging output a little bit, add - the following line to `/etc/lilo.conf': - - append ftape=1,fc10 ftape=4,tracing - -5. Example module parameter setting - ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - To do the same, but with ftape compiled as a loadable kernel - module, add the following line to `/etc/modprobe.conf': - - options ftape ft_probe_fc10=1 ft_tracing=4 - -******************************************************************************* - -D. Support and contacts - ==================== - - Ftape is distributed under the GNU General Public License. There is - absolutely no warranty for this software. However, you can reach - the current maintainer of the ftape package under the email address - given in the MAINTAINERS file which is located in the top level - directory of the Linux kernel source tree. There you'll find also - the relevant mailing list to use as a discussion forum and the web - page to query for the most recent documentation, related work and - development versions of ftape. - - Changelog: - ========== - -~1996: Original Document - -10-24-2004: General cleanup and updating, noting additional module options. - James Nelson diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 67473849f20e..15e4fed127f6 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -557,9 +557,6 @@ and is between 256 and 4096 characters. It is defined in the file floppy= [HW] See Documentation/floppy.txt. - ftape= [HW] Floppy Tape subsystem debugging options. - See Documentation/ftape.txt. - gamecon.map[2|3]= [HW,JOY] Multisystem joystick and NES/SNES/PSX pad support via parallel port (up to 5 devices per port) diff --git a/MAINTAINERS b/MAINTAINERS index 45df5d4e2ab3..8385a69138a8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1166,11 +1166,6 @@ P: David Howells M: dhowells@redhat.com S: Maintained -FTAPE/QIC-117 -L: linux-tape@vger.kernel.org -W: http://sourceforge.net/projects/ftape -S: Orphan - FUSE: FILESYSTEM IN USERSPACE P: Miklos Szeredi M: miklos@szeredi.hu diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index ad8b537ad47b..24f922f12783 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -855,39 +855,6 @@ config TANBAC_TB0219 depends TANBAC_TB022X select GPIO_VR41XX -menu "Ftape, the floppy tape device driver" - -config FTAPE - tristate "Ftape (QIC-80/Travan) support" - depends on BROKEN_ON_SMP && (ALPHA || X86) - ---help--- - If you have a tape drive that is connected to your floppy - controller, say Y here. - - Some tape drives (like the Seagate "Tape Store 3200" or the Iomega - "Ditto 3200" or the Exabyte "Eagle TR-3") come with a "high speed" - controller of their own. These drives (and their companion - controllers) are also supported if you say Y here. - - If you have a special controller (such as the CMS FC-10, FC-20, - Mountain Mach-II, or any controller that is based on the Intel 82078 - FDC like the high speed controllers by Seagate and Exabyte and - Iomega's "Ditto Dash") you must configure it by selecting the - appropriate entries from the "Floppy tape controllers" sub-menu - below and possibly modify the default values for the IRQ and DMA - channel and the IO base in ftape's configuration menu. - - If you want to use your floppy tape drive on a PCI-bus based system, - please read the file . - - The ftape kernel driver is also available as a runtime loadable - module. To compile this driver as a module, choose M here: the - module will be called ftape. - -source "drivers/char/ftape/Kconfig" - -endmenu - source "drivers/char/agp/Kconfig" source "drivers/char/drm/Kconfig" diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 777cad045094..b1fcdab90947 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -78,7 +78,6 @@ obj-$(CONFIG_TOSHIBA) += toshiba.o obj-$(CONFIG_I8K) += i8k.o obj-$(CONFIG_DS1620) += ds1620.o obj-$(CONFIG_HW_RANDOM) += hw_random/ -obj-$(CONFIG_FTAPE) += ftape/ obj-$(CONFIG_COBALT_LCD) += lcd.o obj-$(CONFIG_PPDEV) += ppdev.o obj-$(CONFIG_NWBUTTON) += nwbutton.o diff --git a/drivers/char/ftape/Kconfig b/drivers/char/ftape/Kconfig deleted file mode 100644 index 0d65189a7ae8..000000000000 --- a/drivers/char/ftape/Kconfig +++ /dev/null @@ -1,330 +0,0 @@ -# -# Ftape configuration -# -config ZFTAPE - tristate "Zftape, the VFS interface" - depends on FTAPE - ---help--- - Normally, you want to say Y or M. DON'T say N here or you - WON'T BE ABLE TO USE YOUR FLOPPY TAPE DRIVE. - - The ftape module itself no longer contains the routines necessary - to interface with the kernel VFS layer (i.e. to actually write data - to and read data from the tape drive). Instead the file system - interface (i.e. the hardware independent part of the driver) has - been moved to a separate module. - - To compile this driver as a module, choose M here: the - module will be called zftape. - - Regardless of whether you say Y or M here, an additional runtime - loadable module called `zft-compressor' which contains code to - support user transparent on-the-fly compression based on Ross - William's lzrw3 algorithm will be produced. If you have enabled the - kernel module loader (i.e. have said Y to "Kernel module loader - support", above) then `zft-compressor' will be loaded - automatically by zftape when needed. - - Despite its name, zftape does NOT use compression by default. - -config ZFT_DFLT_BLK_SZ - int "Default block size" - depends on ZFTAPE - default "10240" - ---help--- - If unsure leave this at its default value, i.e. 10240. Note that - you specify only the default block size here. The block size can be - changed at run time using the MTSETBLK tape operation with the - MTIOCTOP ioctl (i.e. with "mt -f /dev/qft0 setblk #BLKSZ" from the - shell command line). - - The probably most striking difference between zftape and previous - versions of ftape is the fact that all data must be written or read - in multiples of a fixed block size. The block size defaults to - 10240 which is what GNU tar uses. The values for the block size - should be either 1 or multiples of 1024 up to a maximum value of - 63488 (i.e. 62 K). If you specify `1' then zftape's builtin - compression will be disabled. - - Reasonable values are `10240' (GNU tar's default block size), - `5120' (afio's default block size), `32768' (default block size some - backup programs assume for SCSI tape drives) or `1' (no restriction - on block size, but disables builtin compression). - -comment "The compressor will be built as a module only!" - depends on FTAPE && ZFTAPE - -config ZFT_COMPRESSOR - tristate - depends on FTAPE!=n && ZFTAPE!=n - default m - -config FT_NR_BUFFERS - int "Number of ftape buffers (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "3" - help - Please leave this at `3' unless you REALLY know what you are doing. - It is not necessary to change this value. Values below 3 make the - proper use of ftape impossible, values greater than 3 are a waste of - memory. You can change the amount of DMA memory used by ftape at - runtime with "mt -f /dev/qft0 setdrvbuffer #NUMBUFFERS". Each buffer - wastes 32 KB of memory. Please note that this memory cannot be - swapped out. - -config FT_PROC_FS - bool "Enable procfs status report (+2kb)" - depends on FTAPE && PROC_FS - ---help--- - Optional. Saying Y will result in creation of a directory - `/proc/ftape' under the /proc file system. The files can be viewed - with your favorite pager (i.e. use "more /proc/ftape/history" or - "less /proc/ftape/history" or simply "cat /proc/ftape/history"). The - file will contain some status information about the inserted - cartridge, the kernel driver, your tape drive, the floppy disk - controller and the error history for the most recent use of the - kernel driver. Saying Y will enlarge the size of the ftape driver - by approximately 2 KB. - - WARNING: When compiling ftape as a module (i.e. saying M to "Floppy - tape drive") it is dangerous to use ftape's /proc file system - interface. Accessing `/proc/ftape' while the module is unloaded will - result in a kernel Oops. This cannot be fixed from inside ftape. - -choice - prompt "Debugging output" - depends on FTAPE - default FT_NORMAL_DEBUG - -config FT_NORMAL_DEBUG - bool "Normal" - ---help--- - This option controls the amount of debugging output the ftape driver - is ABLE to produce; it does not increase or diminish the debugging - level itself. If unsure, leave this at its default setting, - i.e. choose "Normal". - - Ftape can print lots of debugging messages to the system console - resp. kernel log files. Reducing the amount of possible debugging - output reduces the size of the kernel module by some KB, so it might - be a good idea to use "None" for emergency boot floppies. - - If you want to save memory then the following strategy is - recommended: leave this option at its default setting "Normal" until - you know that the driver works as expected, afterwards reconfigure - the kernel, this time specifying "Reduced" or "None" and recompile - and install the kernel as usual. Note that choosing "Excessive" - debugging output does not increase the amount of debugging output - printed to the console but only makes it possible to produce - "Excessive" debugging output. - - Please read for a short description - how to control the amount of debugging output. - -config FT_FULL_DEBUG - bool "Excessive" - help - Extremely verbose output for driver debugging purposes. - -config FT_NO_TRACE - bool "Reduced" - help - Reduced tape driver debugging output. - -config FT_NO_TRACE_AT_ALL - bool "None" - help - Suppress all debugging output from the tape drive. - -endchoice - -comment "Hardware configuration" - depends on FTAPE - -choice - prompt "Floppy tape controllers" - depends on FTAPE - default FT_STD_FDC - -config FT_STD_FDC - bool "Standard" - ---help--- - Only change this setting if you have a special controller. If you - didn't plug any add-on card into your computer system but just - plugged the floppy tape cable into the already existing floppy drive - controller then you don't want to change the default setting, - i.e. choose "Standard". - - Choose "MACH-2" if you have a Mountain Mach-2 controller. - Choose "FC-10/FC-20" if you have a Colorado FC-10 or FC-20 - controller. - Choose "Alt/82078" if you have another controller that is located at - an IO base address different from the standard floppy drive - controller's base address of `0x3f0', or uses an IRQ (interrupt) - channel different from `6', or a DMA channel different from - `2'. This is necessary for any controller card that is based on - Intel's 82078 FDC such as Seagate's, Exabyte's and Iomega's "high - speed" controllers. - - If you choose something other than "Standard" then please make - sure that the settings for the IO base address and the IRQ and DMA - channel in the configuration menus below are correct. Use the manual - of your tape drive to determine the correct settings! - - If you are already successfully using your tape drive with another - operating system then you definitely should use the same settings - for the IO base, the IRQ and DMA channel that have proven to work - with that other OS. - - Note that this menu lets you specify only the default setting for - the hardware setup. The hardware configuration can be changed at - boot time (when ftape is compiled into the kernel, i.e. if you - have said Y to "Floppy tape drive") or module load time (i.e. if you - have said M to "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. If you want to use your floppy tape drive on a - PCI-bus based system, please read the file - . - -config FT_MACH2 - bool "MACH-2" - -config FT_PROBE_FC10 - bool "FC-10/FC-20" - -config FT_ALT_FDC - bool "Alt/82078" - -endchoice - -comment "Consult the manuals of your tape drive for the correct settings!" - depends on FTAPE && !FT_STD_FDC - -config FT_FDC_BASE - hex "IO base of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the base IO address are correct: - <<< MACH-2 : 0x1E0 >>> - <<< FC-10/FC-20: 0x180 >>> - <<< Secondary : 0x370 >>> - Secondary refers to a secondary FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the IO base. The hardware configuration can be changed at boot time - (when ftape is compiled into the kernel, i.e. if you specified Y to - "Floppy tape drive") or module load time (i.e. if you have said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_IRQ - int "IRQ channel of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the interrupt channel are correct: - <<< MACH-2 : 6 >>> - <<< FC-10/FC-20: 9 >>> - <<< Secondary : 6 >>> - Secondary refers to secondary a FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the IRQ channel. The hardware configuration can be changed at boot - time (when ftape is compiled into the kernel, i.e. if you said Y to - "Floppy tape drive") or module load time (i.e. if you said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_DMA - int "DMA channel of the floppy disk controller" - depends on FTAPE && !FT_STD_FDC - default "0" - ---help--- - You don't need to specify a value if the following default - settings for the DMA channel are correct: - <<< MACH-2 : 2 >>> - <<< FC-10/FC-20: 3 >>> - <<< Secondary : 2 >>> - Secondary refers to a secondary FDC controller like the "high speed" - controllers delivered by Seagate or Exabyte or Iomega's Ditto Dash. - Please make sure that the setting for the IO base address - specified here is correct. USE THE MANUAL OF YOUR TAPE DRIVE OR - CONTROLLER CARD TO DETERMINE THE CORRECT SETTING. If you are already - successfully using the tape drive with another operating system then - you definitely should use the same settings for the IO base that has - proven to work with that other OS. - - Note that this menu lets you specify only the default setting for - the DMA channel. The hardware configuration can be changed at boot - time (when ftape is compiled into the kernel, i.e. if you said Y to - "Floppy tape drive") or module load time (i.e. if you said M to - "Floppy tape drive"). - - Please read also the file which - contains a short description of the parameters that can be set at - boot or load time. - -config FT_FDC_THR - int "Default FIFO threshold (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "8" - help - Set the FIFO threshold of the FDC. If this is higher the DMA - controller may serve the FDC after a higher latency time. If this is - lower, fewer DMA transfers occur leading to less bus contention. - You may try to tune this if ftape annoys you with "reduced data - rate because of excessive overrun errors" messages. However, this - doesn't seem to have too much effect. - - If unsure, don't touch the initial value, i.e. leave it at "8". - -config FT_FDC_MAX_RATE - int "Maximal data rate to use (EXPERIMENTAL)" - depends on FTAPE && EXPERIMENTAL - default "2000" - ---help--- - With some motherboard/FDC combinations ftape will not be able to - run your FDC/tape drive combination at the highest available - speed. If this is the case you'll encounter "reduced data rate - because of excessive overrun errors" messages and lots of retries - before ftape finally decides to reduce the data rate. - - In this case it might be desirable to tell ftape beforehand that - it need not try to run the tape drive at the highest available - speed. If unsure, leave this disabled, i.e. leave it at 2000 - bits/sec. - -config FT_ALPHA_CLOCK - int "CPU clock frequency of your DEC Alpha" if ALPHA - depends on FTAPE - default "0" - help - On some DEC Alpha machines the CPU clock frequency cannot be - determined automatically, so you need to specify it here ONLY if - running a DEC Alpha, otherwise this setting has no effect. - diff --git a/drivers/char/ftape/Makefile b/drivers/char/ftape/Makefile deleted file mode 100644 index 0e67d2f8b7ec..000000000000 --- a/drivers/char/ftape/Makefile +++ /dev/null @@ -1,28 +0,0 @@ -# -# Copyright (C) 1997 Claus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/05 19:17:56 $ -# -# Makefile for the QIC-40/80/3010/3020 floppy-tape driver for -# Linux. -# - -obj-$(CONFIG_FTAPE) += lowlevel/ -obj-$(CONFIG_ZFTAPE) += zftape/ -obj-$(CONFIG_ZFT_COMPRESSOR) += compressor/ diff --git a/drivers/char/ftape/README.PCI b/drivers/char/ftape/README.PCI deleted file mode 100644 index 18de159d36e0..000000000000 --- a/drivers/char/ftape/README.PCI +++ /dev/null @@ -1,81 +0,0 @@ -Some notes for ftape users with PCI motherboards: -================================================= - -The problem: ------------- - -There have been some problem reports from people using PCI-bus based -systems getting overrun errors. -I wasn't able to reproduce these until I ran ftape on a Intel Plato -(Premiere PCI II) motherboard with bios version 1.00.08AX1. -It turned out that if GAT (Guaranteed Access Timing) is enabled (?) -ftape gets a lot of overrun errors. -The problem disappears when disabling GAT in the bios. -Note that Intel removed this setting (permanently disabled) from the -1.00.10AX1 bios ! - -It looks like that if GAT is enabled there are often large periods -(greater than 120 us !??) on the ISA bus that the DMA controller cannot -service the floppy disk controller. -I cannot imagine this being acceptable in a decent PCI implementation. -Maybe this is a `feature' of the chipset. I can only speculate why -Intel choose to remove the option from the latest Bios... - -The lesson of this all is that there may be other motherboard -implementations having the same of similar problems. -If you experience a lot of overrun errors during a backup to tape, -see if there is some setting in the Bios that may influence the -bus timing. - -I judge this a hardware problem and not a limitation of ftape ;-) -My DOS backup software seems to be suffering from the same problems -and even refuses to run at 1 Mbps ! -Ftape will reduce the data-rate from 1 Mbps to 500 Kbps if the number -of overrun errors on a track exceeds a threshold. - - -Possible solutions: -------------------- - -Some of the problems were solved by upgrading the (flash) bios. -Other suggest that it has to do with the FDC being on the PCI -bus, but that is not the case with the Intel Premiere II boards. -[If upgrading the bios doesn't solve the problem you could try -a floppy disk controller on the isa-bus]. - -Here is a list of systems and recommended BIOS settings: - - - Intel Premiere PCI (Revenge): - -Bios version 1.00.09.AF2 is reported to work. - - - - Intel Premiere PCI II (Plato): - -Bios version 1.00.10.AX1 and version 11 beta are ok. -If using version 1.00.08.AX1, GAT must be disabled ! - - - - ASUS PCI/I-SP3G: - -Preferred settings: ISA-GAT-mode : disabled - DMA-linebuffer-mode : standard - ISA-masterbuffer-mode : standard - - - DELL Dimension XPS P90 - -Bios version A2 is reported to be broken, while bios version A5 works. -You can get a flash bios upgrade from http://www.dell.com - - -To see if you're having the GAT problem, try making a backup -under DOS. If it's very slow and often repositions you're -probably having this problem. - - --//-- - LocalWords: ftape PCI bios GAT ISA DMA chipset Mbps Kbps FDC isa AF ok ASUS - LocalWords: SP linebuffer masterbuffer XPS http www com diff --git a/drivers/char/ftape/RELEASE-NOTES b/drivers/char/ftape/RELEASE-NOTES deleted file mode 100644 index 03799dbc05a4..000000000000 --- a/drivers/char/ftape/RELEASE-NOTES +++ /dev/null @@ -1,966 +0,0 @@ -Hey, Emacs, we're -*-Text-*- mode! - -===== Release notes for ftape-3.04d 25/11/97 ===== -- The correct pre-processor statement for "else if" is "#elif" not - "elsif". -- Need to call zft_reset_position() when overwriting cartridges - previously written with ftape-2.x, sftape, or ancient - (pre-ftape-3.x) versions of zftape. - -===== Release notes for ftape-3.04c 16/11/97 ===== -- fdc_probe() was calling DUMPREGS with a result length of "1" which - was just fine. Undo previous change. - -===== Release notes for ftape-3.04b 14/11/97 ===== - -- patches/2.x.x/floppy.c.diff was somewhat broken, releasing i/o - regions it never had allocated. -- fdc_probe() was calling DUMPREGS with a result length of "1" instead - of "10" -- Writing deleted data marks if the first segents on track zero are - should work now. -- ftformat should now be able to handle those cases where the tape - drive sets the read only status bit (QIC-40/80 cartridges with - QIC-3010/3020 tape drives) because the header segment is damaged. -- the MTIOCFTCMD ioctl may now be issued by the superuser ONLY. - -===== Release notes for ftape-3.04a 12/11/97 ===== -- Fix an "infinite loop can't be killed by signal" bug in - ftape_get_drive_status(). Only relevant when trying to access - buggy/misconfigured hardware -- Try to compensate a bug in the HP Colorado T3000's firmware: it - doesn't set the write protect bit for QIC80/QIC40 cartridges. - -===== Release notes for ftape-3.04 06/11/97 ===== -- If positioning with fast seeking fails fall back to a slow seek - before giving up. -- (nearly) no retries on "no data errors" when verifying after - formatting. Improved tuning of the bad sector map after formatting. -- the directory layout has changed again to allow for easier kernel - integration -- Module parameter "ftape_tracing" now is called "ft_tracing" because - the "ftape_tracing" variable has the version checksum attached to it. -- `/proc/ftape' interface for 2.0.* kernels. `/proc/ftape' no longer - is a directory but a file that contains all the information formerly - provided in separate files under the `/proc/ftape/' directory. -- Most of the configuration options have been prefixed by "CONFIG_FT_" - in preparation of the kernel inclusion. The Makefiles under - "./ftape/" should be directly usable by the kernel. -- The MODVERSIONS stuff is now auto-detected. -- Broke backslashed multi line options in MCONFIG into separate lines - using GNU-make's "+=" feature. -- The html and dvi version of the manual is now installed under - '/usr/doc/ftape` with 'make install` -- New SMP define in MCONFIG. ftape works with SMP if this is defined. -- attempt to cope with "excessive overrun errors" by gradually - increasing FDC FIFO threshold. But this doesn't seem to have too - much an effect. -- New load time configuration parameter "ft_fdc_rate_limit". If you - encounter too many overrun errors with a 2Mb controller then you - might want to set this to 1000. -- overrun errors on the last sector in a segment sometimes result in - a zero DMA residue. Dunno why, but compensate for it. -- there were still fdc_read() timeout errors. I think I have fixed it - now, please FIXME. -- Sometimes ftape_write() failed to re-start the tape drive when a - segment without a good sector was reached ("wait for empty segment - failed"). This is fixed. Especially important for > QIC-3010. -- sftape (aka ftape-2.x) has vanished. I didn't work on it for - ages. It is probably still possible to use the old code with - ftape-3.04, if one really needs it (BUT RECOMPILE IT) -- zftape no longer alters the contents of already existing volume - table entries, which makes it possible to fill in missing fields, - like time stamps using some user space program. -- ./contrib/vtblc/ contains such a program. -- new perl script ./contrib/scripts/listtape that list the contents of a - floppy tape cartridge parsing the output of "mt volinfo" + "mt fsf" -- the MTWEOF implementation has changed a little bit (after I had a - look at amanda). Calling MTWEOF while the tape is still held open - after writing something to the tape now will terminate the current - volume, and start a new one at the current position. -- the volume table maintained by zftape now is a doubly linked list - that grows dynamically as needed. - - formatting floppy tape cartridges - --------------------------------- - * there is a new user space formatting program that does most of the - dirty work in user space (auto-detect, computing the sector - coordinates, adjusting time stamps and statistics). It has a - simple command line interface. - * ftape-format.o has vanished, it has been folded into the low level - ftape.o module, and the ioctl interface into zftape.o. Most of the - complicated stuff has been moved to user space, so there was no - need for a separate module anymore. - * there is a new ioctl MTIOCFTCMD that sends a bare QIC-117 command - to the tape drive. - * there is a new mmap() feature to map the dma buffers into user - space to be used by the user level formatting program. - * Formatting of yet unformatted or totally degaussed cartridges - should be possible now. FIXME. - -===== Release notes for ftape-3.03b, ==== - -ftape-3.03b was released as a beta release only. Its main new feature -was support of the DITTO-2GB drive. This was made possible by reverse -engineering done by after Iomega failed to support -ftape. Although they had promised to do so (this makes me feel a bit -sad and uncomfortable about Iomega). - -===== Release notes for ftape-3.03a, 22/05/97 ==== - -- Finally fixed auto-un-loading of modules for kernels > 2.1.18 -- Add an "uninstall" target to the Makefile -- removed the kdtime hack -- texi2www didn't properly set the back-reference from a footnote back - to the regular text. - - zftape specific - --------------- - * hide the old compression map volume. Taper doesn't accept the - presence of non-Taper volumes and Taper-written volume on the same - tape. - * EOD (End Of Data) handling was still broken: the expected behavior - is to return a zero byte count at the first attempt to read past - EOD, return a zero byte count at the second attempt to read past - EOD and THEN return -EIO. - - ftape-format specific - --------------------- - * Detection of QIC-40 cartridges in select_tape_format() was broken - and made it impossible to format QIC-3010/3020 cartridges. - * There are strange "TR-1 Extra" cartridges out there which weren't - detected properly because the don't strictly conform to the - QIC-80, Rev. N, spec. - -===== Release notes for ftape-3.03, 30/04/97 ===== - -- Removed kernel integration code from the package. I plan to provide - a package that can be integrated into the stock kernel separately - (hopefully soon). - As a result, a simple `make' command now will build everything. -- ALL compile time configuration options have been moved to the file - `MCONFIG'. -- Quite a few `low level' changes to allow formatting of cartridges. -- formatting is implemented as a separate module `ftape-format.o'. The - modified `mt' program contains sample code that shows how to use it. -- The VFS interface has been moved from the `ftape.o' module to the - high level modules `zftape.o' resp. `sftape.o'. `ftape.o' contains - the hardware support only. -- A bit of /proc support for kernels > 2.1.28 -- Moved documentation to Doc subdir. INSTALL now contains some real - installation notes. -- `install' target in Makefile. - -zftape specific: ----------------- - -- zftape works for large cartridges now ( > 2^31 bytes) -- MTIOCVOLINFO and MTIOCGETSIZE now return the size in KILOBYTES, - NO LONGER in bytes. - -- permissions for write access to a cartridge have changed: - * zftape now also takes the file access mode into account - * zftape no longer allows writing in the middle of the recorded - media. The tape has to be positioned at BOT or EOD for write - access. - -- MTBSF has changed. It used to position at the beginning of the - previous file when called with count 1. This was different from the - expected behavior for other Un*x tape drivers (i.e. SCSI). MTBSF - with count 1 should merely position at the beginning of the current - volume. Fixed. As a result, `tar --verify' now produces the desired - result: it verifies the last written volume, not the pre-last - written volume. - -- The compression map has vanished --> no need for `mt erase' any - more. Fast seeking in a compressed volume is still be possible, but - takes slightly longer. As a side effect, you may experience an - additional volume showing up in front of all others for old - cartridges. This is the tape volume that holds the compression map. - -- The compression support for zftape has been moved to a separate - module `zft-compressor'. DON'T forget to load it before trying to - read back compressed volumes. The stock `zftape.o' module probes for - the module `zft-compressor' using the kerneld message channel; you - have to install `zft-compressor.o' in a place where modprobe can - find it if you want to use this. - -- New experimental feature that tries to get the broken down GMT time - from user space via a kernel daemon message channel. You need to - compile and start the `kdtime' daemon contained in the contrib - directory to use it. Needed (?) for time stamps in the header - segments and the volume table. - -- variable block size mode via MTSETBLK 0 - -- keep modules locked in memory after the block size has been changed - -sftape specific: ----------------- - -- end of tape handling should be fixed, i.e. multi volume archives - written with `afio' can be read back now. - - -===== Release notes for ftape-3.02a, 09/01/97 ===== - -No big news: -- call zft_init() resp. sft_init() when compiling the entire stuff - into the kernel image. -- fix bug in ftape-setup.c when NO_TRACE_AT_ALL was defined. -- fix bug in sftape-eof.c/zftape-eof.c for old kernels (1.2.*) -- add support for new module interface for recent kernels - -===== Release notes for ftape-3.02, 16/12/96 ===== -- Fixed the `FDC unlock command failed' bug in fdc-io.c. When the FIFO - was already locked when ftape was loaded, ftape failed to unlock it. -- Fixed compilation of `contrib/gnumt'. It now finds `mtio.h' even if - ftape is NOT included into the kernel source tree. -- fc-10.c: include for inb() and outb(). -- ftape/sftape/zftape: all global variable now have either a `ftape_', - a `ft_', `sft_', `zft_' or `qic_' prefix to prevent name clashes - with other parts of the kernel when including ftape into the kernel - source tree. -- Kerneld support has changed. `ftape' now searches for a module - `ftape-frontend' when none of the frontend (`sftape' or `zftape') is - loaded. Please refer to the `Installation/Loading ftape' section of - the TeXinfo manual. -- Add load resp. boot-time configuration of ftape. There are now - variables ft_fdc_base, ft_fdc_dma and ft_fdc_irq corresponding to - the former FDC_BASE etc. compile time definitions. One can also use - the kernel command line parameters to configure the driver if it is - compiled into the kernel. Also, the FC-10/FC-20 support is load-time - configurable now as well as the MACH-II hack (ft_probe_fc10, - resp. ft_mach2). Please refer to the section `Installation/Configure - ftape' of the TeXinfo manual. -- I removed the MODVERSIONS option from `Makefile.module'. Let me alone - with ftape and MODVERSIONS unless you include the ftape sources into - the kernel source tree. -- new vendors in `vendors.h': - * HP Colorado T3000 - * ComByte DoublePlay (including a bug fix for their broken - formatting software, thanks to whraven@njackn.com) - * Iomega DITTO 2GIG. NOTE: this drive cannot work with ftape because - the logical data layout of the cartridges used by this drive does - NOT conform to the QIC standards, it is a special Iomega specific - format. I've sent mail to Iomega but didn't receive an answer - yet. If you want this drive to be supported by ftape, ask Iomega - to give me information about it. -- zftape: - * re-introduced the MTIOC_ZFTAPE_GETBLKSZ ioctl for compatibility - with zftape 1.06a and earlier. Please don't use it when writing - new software, use the MTIOCVOLINFO ioctl instead. - * Major overhaul of the code that updates the header segments. Never - change the tape label unless erasing the tape. Thus we almost - never need to write the header segments, unless we would modify - the bad sector map which isn't done yet. Updating of volume table - and compression map more secure now although it takes a bit - longer. - * Fixed bug when aborting a write operation with a signal: zftape - now finishes the current volume (i.e. writes an eof marker) at the - current position. It didn't before which led to somehow *strange* - behavior in this cases. - * Keep module locked in memory when using it with the non-rewinding - devices and the tape is not logical at BOT. Needed for kerneld - support. -- sftape: - * Keep module locked in memory when using it with the non-rewinding - devices and the tape is not logical at BOT. Needed for kerneld - support. - -===== Release notes for ftape-3.01, 14/11/96 ===== - -- Fixed silly bugs in ftape-3.00: - * MAKEDEV.ftape: major device number must be 27, not 23 - * sftape/sftape-read.c: sftape_read_header_segments() called - itself recursively instead of calling ftape_read_header_segment() - * zftape/qic-vtbl.h: conversion of ftape's file marks to zftape's - internal volume table was broken. - * patches/2.x.x/linux-2.0.21.dif: my RCS (resp. CVS) system replaced - the `$Revison:' etc. macros in the `ftape.h' concerning part of the - patch :-( Fixed. - * info/ftape.info: Fixed misspellings (`cp' <-> `cp -r' etc.) - * when ftape/sftape or ftape/zftape was compiled into the kernel the - variable ftape_status was declared twice. Fixed. - * removed reference to undeclared variable kernel_version when not - compiling as module - * fixed a bug introduced by the use of bit-fields for some flags - (i.e. write_protected, no_cartridge, formatted) - * flag `header_read' is now reset correctly to zero when tape is - removed. -- fixed a bug in sftape/sftape-eof.c that was already in the original - ftape code. MTFSF/BSF was not handled correctly when positioned - right before the file mark (think of tar) -- Changed TRACE macros (following a suggestion of Marcin Dalecki) to use - the predefined __FUNCTION__ macro of GCC. Spares about 4k of code. -- added new vendor id for Iomega DITTO 2GIG -- fixed a bug already present in zftape-1.06 when aborting a write - with a signal: we now finish the current volume at that - position. Header segments remain NOT up to date until an explicit call - to MTREW or MTOFFL is done. - -===== Release notes for ftape-3.00, 14/10/96 ===== - -- Merged ftape with zftape. There are three modules now: - ftape for the hardware support, sftape for the implementation of the - original ftape eof mark stuff and zftape that implements zftape's way - of handling things (compression, volume table, tape blocks of - constant length) -- Documentation in TeXinfo format in the `info' subdirectory. -- New ioctls for zftape. See zftape/zftape.h -- Dummy formatting ioctl for ftape. See ftape.h -- Kernel patch files for the 2.*.* series to include ftape-3.00 in the - kernel source tree. These includes a kernel compatible Config.in - script and fairly large online information for the kernel configure - script. -- Support for compiling with Linux-1.2.13. -- Modified GNU mt from their cpio package that can handle the new - ioctls. -- ftape/sftape/zftape is kerneld save now! - -Notes on sftape: -- sftape implements the eof handling code of the original ftape. If - you like to stick with the original ftape stuff, you have to use - this module, not zftape. -- sftape is kerneld save, unlike the original ftape. -- we keep the entire header segment now in memory, so no need to read - it before updating the header segments. Additional memory - consumption: 256 bytes. - -Notes for zftape: -- zftape has support for tapes with format code 6 now, which use a - slightly different volume table format compared with other floppy - tapes. -- new ioctls for zftape. Have a look at zftape/zftape.h -- The internal volume table representation has changed for zftape. Old - cartridges are converted automatically. -- zftape no longer uses compression map segments, which have vanished - from the QIC specs, but creates volume table entry that reserves - enough space for the compression map. -- zftape is kerneld save now. -- we keep the entire header segment now in memory, so no need to read - it before updating the header segments. Additional memory - consumption: 256 bytes. - -Notes for contrib/gnumt: -- modified mt from the GNU cpio package that supports all the new - ioctls of zftape. -Notes for contrib/swapout: -- This contains the swapout.c program that was written by Kai - Harrekilde-Pederson. I simply added a Makefile. - -===== Release notes for ftape-2.10, 14/10/96 ===== - -The ftape maintainer has changed. -Kai Harrekilde-Petersen -has resigned from maintaining ftape, and I, -Claus-Justus Heine , -have taken over. - -- Added support for tapes with `format code 6', i.e. QIC-3020 tapes - with more than 2^16 segments. -- merged changes made by Bas Laarhoven with ftape-2.09. Refer - to his release notes below. I've included them into this - file unchanged for your reference. -- disabled call stack back trace for now. This new feature - introduced by the interim release 2.0.x still seems to - be buggy. -- Tried to minimize differences between the ftape version - to be included into the kernel source tree and the standalone - module version. -- Reintroduced support for Linux-1.2.13. Please refer to the - Install-guide. - -===== Release notes for ftape-2.09, 16/06/96 ===== - -There aren't any really big news in this release, mostly just that I -(the maintainer) have changed my email address (due to a new job). My -new address is - -- The CLK_48MHZ and FDC_82078SL options has gone (all 2Mbps cards seem - to use a 48MHz oscillator anyway and I haven't heard of an 'SL - chip out there). -- The S82078B has been `downgraded' to i82077AA compability. -- TESTING option revived. Right now, it'll enable the (seriously broken) - 2Mbps code. If you enable it, you'll experience a tape drive that's - *really* out to lunch! -- Some (bold) changes in the init code. Please notify me if they - break things for you. - -===== Release notes for ftape-2.08, 14/03/96 ===== - -If you correct a problem with ftape, please send your patch to -khp@dolphinics.no too. - -- Updated to reflect that NR_MEM_LISTS is gone in 1.3.74 -- Teac 700 added to list of known drives. -- The registered device name is now "ft" rather than "ftape". - -===== Release notes for ftape-2.07a, 14/03/96 ===== - -Bugfixes by Marcin Dalecki : -- In the last release it just compiled against 1.3.70; - now the params to request_irq() and free_irq are() are fixed, so it also - works in 1.3.73 :-) -- Support for modules is now correct for newer kernels. - -===== Release notes for ftape-2.07, 04/03/96 ===== - - -- ftape updated to compile against 1.3.70. -- Iomega 700 and Wangtek 3200 recognised. - - -===== Release notes for ftape-2.06b, 13/02/96 ===== - -Another simple bugfix version. - -- Jumbo 700 recognised. -- Typo in vendors.h fixed. - - -===== Release notes for ftape-2.06a, 10/02/96 ===== - -This release is a simple bugfix version. - -- Linux/SMP: ftape *should* work. -- FC-10/20: Only accepts IRQs 3-7, or 9. If IRQ 9, properly tell the card - to use IRQ 2. Thanks to Greg Crider (gcrider@iclnet.org) for finding and - locating this bug and testing the patch. -- Insight drive recognised correctly again. -- Motor-on wakeup version of the Iomega 250 drive added - - -===== Release notes for ftape-2.06, 28/01/96 ===== - -Special thanks go to Neal Friedman and Steven Sorbom for their -help in producing and testing this release. - -I have continued to clean up the code, with an eye towards inclusion -of ftape in Linus' official kernel (In fact, as I type this, I am -running on a kernel with ftape support statically linked). I have -test-compiled ftape against my 1.2.13 tree without problems. -Hopefully, everything should be OK for the v1.2.x people. - -WARNING! Alan Cox has mailed me that ftape does *NOT* work with -Linux/SMP. If you try to run ftape under Linux/SMP, it will cause a -kernel deadlock (which is worse than a panic). - -- QIC-3020/TR-3: 1Mbps support works. Neal is capable of reading and - writing data to a tape. ftape will automatically detect the type of - tape (e.g. TR-3 vs QIC-80) and move the fdc in and out of - "perpendicular mode" as necessary. -- 2Mbps support is disabled by default, since it is not fully - debugged. If you are adventurous, remove -DFDC_82078SL in the - Makefile and see what happens :-) -- fdc detection: silly bugs removed (Only 2Mbps fdcs were affected) - and added detection of the National Semiconductors PC8744 fdc chip - (used in the PC873xx "super-IO" chips). -- Removed warning about incompatible types when compiling with Linux - 1.2.x. -- README.PCI updated with info about the DELL Dimension XPS P90. -- Connor TST3200R added to detected drives. -- `swapout' utility added to distribution. It will dirty 5Meg of - memory, trying to swap out other programs. Just say `make swapout' - to build it. ftape will do this automatically Real Soon Now (ie: - when I have found out which kernel memory alloc function to call). - - -===== Release notes for ftape-2.05, 08/01/96 ===== - -- For v1.2.x Kernels, you must apply the patch linux-1.2/ksyms.patch to - the kernel and rebuild it (it adds the __get_dma_pages symbol to - ksyms.c). -- Included new asm-i386/io.h file from v1.3.x kernel series, to enable - gcc v.2.7.[12] to compile v1.2.x kernels (linux-1.2/io.h). -- Module versions: If you wish to compile ftape as a versioned module, - you must first compile your kernel with CONFIG_MODVERSIONS=y. - Otherwise, you will get complaints that does not - exist (if that happens, a `touch modversions.h' will help you out). -- CLK_48MHZ: new define in the Makefile (default: non-zero). If you have - a tape controller card that uses the i82078(-1) chip, but cannot get - it to work with ftape, try set it to 0 (and please report this). -- QIC-3010/3020: Complete support is still missing, but will hopefully - come soon. Steven Sorbom has kindly provided me with hints about - this. Writing of QIC-3020 tapes definitely does NOT work (do not try - it! - the drive will not be in "perpendicular mode" and this will ruin - the formatting info on the tape). -- ftape_num_buffers is out of fashion: use NR_BUFFERS instead (and - recompile if you want to change it :-). - - -===== Release notes for ftape-2.04, 01/01/96 ===== - -This version by Kai Harrekilde-Petersen - -- ALERT! Support for Kernels earlier then v1.1.85 is about to go away. - I intend to clean up some of the code (getting rid of an annoyingly - large numbers of #ifdef mostly), which means that support for - pre-1.1.85 kernels must go as well. -- NR_FTAPE_BUFFERS is gone; You can instead select the number of dma - buffers by saying `insmod ftape.o ftape_num_buffer=' instead. -- Configure script gone. ftape will now automagically determine your - kernel version by /usr/include/linux/version.h instead. -- CONFIG_MODVERSIONS now work. All combinations of versioned / - unversioned kernel and ftape module works (at least with my 1.3.52 - kernel). -- If you have problems with inserting ftape into an old (1.2.x) - kernel (e.g. insmod says "1.2.8 does not match 1.2.8), recompile - your modules utilities with your new compiler. -- Reveal TB1400 drive added to vendors.h -- Support for the i82078-1 (2Mbps) chip is coming along. The - biggest problem is that I don't have such a card, which makes - testing / debugging somewhat problematic. The second biggest - problem is that I do not have the QIC-3010/3020 standards either. - Status right now is that the chip is detected, and it should be - possible to put it into 2Mbps mode. However, I do not know what - "extras" are needed to complete the support. Although putting the - i82078 into 1Mbps mode ought to work out of the box, it doesn't - (right now, ftape complains about id am errors). - - -===== Release notes for ftape-2.04beta5, 29/12/95 ===== - -Bas offline linux-tape ----------------------- -For reasons only known to the majordomo mail list processor, Bas was -kicked off the linux-tape list sometime during the summer. Being -overworked at his for-pay job, he didn't notice it much. Instead I -(Kai, khp@dolphinics.no) has worked on ftape to produce the 2.04(beta) -version. - -zftape ------- -Note that there exists a much improved version of ftape, written by -Claus-Justus Heine which is named -zftape, which conforms to the QIC-80 specs on how to mark backups, and -is capable of doing automatic compression. However, zftape makes -substantial changes to ftape, and I (Kai) have therefore declined to -integrate zftape into ftape. Hopefully, this will happen soon. - -CONFIG_QIC117 removed from the kernel -------------------------------------- -The biggest change of all is that ftape now will allocate its dma -buffers when it is inserted. The means that the CONFIG_QIC117 option -has disappeared from the Linux kernel as of v1.3.34. If you have an -earlier kernel, simply answer 'no' to the question will do the trick -(if you get complains about __get_free_pages() missing, contact the -linux-tape mailing list). - -Note that ftape-2.04beta will work equally well on kernels with and -without `ftape support'. The only catch is, that you will waste -around 96-128Kb of precious DMA'able memory on a box that has ftape -support compiled in. - -Now for the real changes: - -- FC-20 can now use DMA channels 1, 2, and 3. Thanks to Daniel - Cohen, catman@wpi.edu. -- ftape no longer requires a (gigantic) 96Kb buffer to be statically - allocated by the kernel. -- Added new Iomega drive (8882) to vendors.h -- -fno-strength-reduce added to Makefile, since GCC is broken. -- i82078-1 (2Mbps) FDC support started. - - -===== Release notes for ftape-2.03b, 27/05/95 ===== - -- Prevented verify_area to return error if called with zero length. -- Fixed a bug in flush_buffers that caused too much padding to be - written when a final segment had bad sectors. -- Increased maximum fast-seek overshoot value from 5 to 10 segments. -- Breaking loop after 5 retries when positioning fails. -- Fixed wrong calculation of tape length for QIC-3010 and QIC-3020 - tapes (densities were swapped). -- Fixed wrong calculation of overshoot on seek_forward: Wrong sign - of error. -- Suppress (false) error message due to new tape loaded. -- Added two new CMS drives (11c3 and 11c5) to vendors.h. - - -===== Release notes for ftape-2.03a, 09/05/95 ===== - -- Fixed display of old error (even if already cleared) in ftape_open. -- Improved tape length detection, ioctls would fail for 425 ft tapes. - Until the tape length is calculated with data from the header - segment, we'll use worst-case values. -- Clear eof_mark after rewinding ioctls. -- Fixed wrong version message (2.03 had 2.02g id). -- Fixed bug that caused the fdc to be reset very frequently. - This shouldn't affect normal operation but the timing of the - report routines has changed again and that may cause problems. - We'll just have to find out.... -- Implemented correct write precompensation setting for QIC-3010/3020. -- Cleaned up fdc_interrupt_wait routine. Hope it still works :-) -- Finally removed (already disabled) special eof mark handling for - gnu tar. -- Changed order of get_dma_residue and disable_dma in fdc-isr.c - because the current order would fail on at least one system. - We're back to the original order again, hope (and expect) this - doesn't break any other system. - - -===== Release notes for ftape-2.03, 07/05/95 ===== - -(Changes refer to the first ftape-2.02 release) - -Support for wide and extended length tapes ------------------------------------------- -The Conner TSM 420 and 850 drives are reported to be working. -I haven't received any reports about other brands; the TSM 420 -and 850 seem to be the most widely used wide drives. -Extended length tapes (425 ft) with normal QIC-80 drives -are operating too (At least I've had no reports stating otherwise). -_Not_ yet completely supported (although they may work) are -QIC-3020 drives and 2 Mbps floppy disk controllers won't work at -the highest speed. -If someone is kind enough to send me one of these, I'll include -support for it too ;-) - -Easier configuration --------------------- -Problems due to wrong settings in the Makefile are prevented -by using a configuration script that sets the necessary (kernel -version dependent) compile time options. -This kernel version is now determined from the sources found -at /usr/src/linux, or if not found, the old way using -/proc/version. -Versioned modules will be used automatically when supported -by- and configured in- the kernel. -Note that the current modules code (1.1.87) is still broken -and _needs_ the fix included in the insmod directory. -Please don't send me any more Oops reports caused by insmod :-( - -Reduced module size -------------------- -The standard module size is much reduced and some compile time -options can even reduce it further. (I don't recommend this -for normal use but it can be handy for rescue diskettes) - -Option: Approx. module size: - - 150 Kb -NO_TRACE 125 Kb -NO_TRACE_AT_ALL 67 Kb - - -Much improved driver interruption ---------------------------------- -Most possible loops have been broken and signal detection -has been improved. -In most cases the driver can be aborted by ^C (SIGINT) and -SIGKILL (kill -9) will generate be a sure kill. -(Note that aborting a tape operation may damage the last -data written to tape) - -Improved error recovery ------------------------ -Ftape now returns an error (ENODATA) to the application if -a segment proves to be unrecoverable and then skips the -bad segment. -This causes most applications to continue to work (tar -and afio) loosing only a small amount (up to 29 Kb) of data. -Retried read operations will now be done slightly off-track -to improve the chance of success. Serious head off-track -errors will be detected. - -FC-10 and FC-20 controllers ---------------------------- -Ftape now supports both the old CMS FC-10 and the newer FC-20 -controllers. -Because the operation of these cards is still undocumented, -thus far they will only work with the default settings (See -Makefile). Any feed-back on how to use them with other settings -will be welcome ! -Compilation will fail if one changes the settings to illegal -values. - -Kernels and compilers ---------------------- -Ftape is currently being developed using the 2.5.8 compiler. -The older 2.4.5 probably works too (Set option in Makefile!). -I have no experience with any later compilers nor Elf support. -Any information on this is welcome. -The latest kernel I have tested ftape with is 1.2.6. - -Compression ------------ -An impressive collection of changes for ftape including -on-the-fly compression is still lying on my desk. -If 2.03 proves to be reliable I might start integrating these -but as usual, I'm short in time :-( - -Formatting ----------- -There is still no way to format tapes under Linux. As far as -I know all attempts to write such a program have died now. -Since formatted tapes are rather common now, I think all we -need is a utility that writes a worst case pattern and verifies -that with the drive put in verify mode, reducing margins. -Any takers ? - -Furthermore ------------ -Cleaned up messages. -Prepared to support multiple tape drives on one fdc. -Thanks to all the people who sent bug reports and helped me -improve the driver. Without trying to be complete I'll mention -Gary Anderson (without his accurate reports and unreliable -hardware there wouldn't be a 2.03), Stefan Kneifel (FC-20), -Robert Broughton (FC-20, you were almost there ;-), Bjorn -Ekwall (for the versioned modules and buggy insmod ;-), Peter -Fox, Christopher Oliver, Ralph Whittaker and not the least -Linus Torvalds (for Linux and keeping me busy because of -changes to the kernel ;-) -Thanks to anyone I forgot, for the bug reports, the ftape -bashing and the mental support... - - -That's it for now. Have Fun, - -Bas. - - -===== Release notes for ftape-2.02g, 06/05/95 ===== - -- Added extra test to break read-id loop with signal. -- Changed rewind code to handle negative overshoot for drives - that take very long to start or stop. -- Let use of get/set i/o-regions depend on kernel version. -- Changed code to use a more general test for conditional - compilations depending on kernel version. -- Improved micro-step functionality to go off-track only - while reading (id & data). -- Added failure on tape-not-referenced bit in ftape_command. -- Added FOREVER option to read-wait routine. -- Changed read-id to use shorter timeout causing smaller - rewinds on timeout. -- Made kernel-interface functions static. - - -===== Release notes for ftape-2.02f, 03/05/95 ===== - -- Added support for dual tape drives on my system, extended Configure - script to detect host 'dodo'. -- Log media defect in history if ecc failed and no data was returned. -- Fixed Configure script that was failing for kernel versions with - double digit version or revision numbers. - - -===== Release notes for ftape-2.02e, 01/05/95 ===== - -- Fixed reposition loop at logical eot (failing read_id). -- Fixed 34 segment offset when rewinding. -- Added fast seek capability for more than 255 segments. -- Fixed wrong busy result from ftape_command causing reverse - seek to fail. -- Added breakout from infinite rewind loop (if something fails). - - -===== Release notes for ftape-2.02d, 30/04/95 ===== - -- Improved abortion on signals: Interrupt will make a graceful - exit, Kill will be less nice and should be used if everything - else fails. -- Included check for tape-head off track. -- Implemented exit from tape-start loop. -- Added kernel io-port registration. -- Implemented skip of failing segment (ENODATA) on ecc failure. - This allows afio and tar to continue when the tape is damaged. -- Made distinction between drive names with different codes. - - -===== Release notes for ftape-2.02c, 22/04/95 ===== - -- Fixed too tight command queueing after tape stop/pause command - issued from within interrupt service routine (Showed as timeout - on Acknowledge errors during retries on some systems) -- Tried to fix timeouts when using 425 ft tape because the extended - length doesn't seem to be detected by the hardware. - We now use the format code from the header segment so adjust the - timing after reading the header segment. -- Fixed some messages stating 'unexpected something...' being not - unexpected anymore. -- Started preparations for merge of dynamic buffer allocation and - compression code. -- Changed some debug messages to include relevant segment information - at level 4. -- Included early bail-out when drive offline, preventing a lot of - false messages. -- Moved ftape_parameter_xxx() offsets into function instead of in calls. -- Removed 'weird, drive busy but no data' error when caused by - an error during a read-id. -- Improved 'timeout on acknowledge' diagnostics. -- Moved MODULE option into Configure. -- Reduced code size when no tracing at all was set (Claus Heine). -- No longer log error code 0 (no error) as an error. - - -===== Release notes for ftape-2.02b, 09/04/95 ===== - -- Relaxed timing for status operation and displaying - abnormal results. Hopefully this shows what's going - wrong with the Conner TSM850R drives. -- Created script for configuration, using version number - of kernel source if available, otherwise /proc/version. -- Fixed conditionals in kernel-interface.c. -- Removed unavoidable TRACE output. - - -===== Release notes for ftape-2.02a, 01/04/95 ===== - -- Implemented `new-style' (versioned) modules support for new - kernels. -- Reduced size of module by moving static data to bss. -- Now using version number of kernel source instead of running - kernel for kernel versions >= 1.1.82 -- Added feedback on drive speeds to vendor information. -- Included fixed insmod sources to distribution (Let's hope - the modules distribution get fixed soon :-/). - -Note that I haven't yet implemented any of the code extension I -received. I hope to find some time to do this soon. - - -===== Release notes for ftape-2.02, 15/01/95 ===== - - -- Fixed failing repositioning when overshoot was incremented. -- Fixed rate selection: Because of a deficiency in the QIC-117 - specification one cannot distinguish between a not implemented - and a failing command. Therefor we now try to find out if the - drive does support this command before usage. -- Fixed error retry using wrong offset in fdc-isr. -- Improved retry code to retry only once on a single no-data - error in a segment. -- Validate sector number extracted from eof mark because an - invalid file mark (due to ???) could cause kernel panic. -- Split ftape-io.c into ftape-io.c and ftape-ctl.c files. -- Corrected too high media error count after writing to - a bad tape. -- Added #include again because old kernel versions - need it. -- Fixed fdc not being disabled when open failed because no tape - drive was found. -- Fixed problem with soft error in sector 32 (shift operator with - shiftcount 32 is not defined). - - -===== Release notes for ftape-2.01, 08/01/95 ===== - - -- Removed TESTING setting from distributed Makefile. -- Fixed `mt asf' failure: Rewind was deferred to close which - overruled the fsf ioctl. -- Prevented non-interruptible commands being interrupted. -- Added missing timeout.pause setting. -- Maximum tape speed read from drive type information table. - If the information is not in the table (0) the drive will - determine the speed itself and put a message in the logfile. - This information should then be added to the table in the - vendors.h file (and reported to me). -- Added call to ftape_init_drive after soft reset for those - (antique) drives that don't do an implicit seek_load_point - after a reset or power up. -- Don't try to set data rate if reset failed. -- Prevent update of seek variables when starting from the - beginning or the end of the tape. -- Fixed wrong adjustment of overshoot in seek_forward(). -- Added sync to Makefile (again). -- Added code to diagnose timer problems (calibr.c). -- Replaced time differences by timediff calls. -- Removed reference to do_floppy from object for recent kernels. -- Fixed wrong display of 'failing dma controller' message. -- Removed various no longer used #include statements. -- Added max. tape speed value to vendor-struct. -- Changed ftape-command to check pre-conditions and wait - if needed. -- Further updated qic117.h to rev G. -- Combined command name table and restrictions table to one. - Extended this table with some new fields. -- Increased timeout on Ack timer value and included code to - report out of spec behaviour. -- Increased rewind timeout margin to calculated + 20%. -- Improved data rate selection so it won't fail on some - older (pre standard) drives. -- Changed initialisation code so drive will be rewound if the - driver is reloaded and the tape is not at bot. -- Moved some of the flush operations from close to the ioctls. -- Added exit code value to failing verify area message. -- Loop until tape halted in smart-stop. -- Fast seek handled specially if located at bot or eot. -- Being more conservative on overshoot value. - - -===== Release notes for ftape-2.00, 31/12/94 ===== - - The Install-guide is completely rewritten and now also includes -some information on how to use the driver. If you're either new -to ftape or new to Unix tape devices make sure to read it ! - - If you own a pci system and experience problems with the -ftape driver make sure to read the README.PCI file. It contains -some hints on how to fix your hardware. - - For anybody who hasn't noticed: The version number of the -driver has been incremented (The latest released version has -been version 1.14d). - This has been done for two major reasons: - - o A new (better) error recovery scheme is implemented. - o Support for new drive types has been added. - - All these improvements/changes will probably include a couple -of new (and old?) bugs. If you encounter any problems that you think -I'm not yet aware of, feel free to send a report to . - I recommend keeping a version of ftape-1.14d available, just -in case ;-) - - This version should work with all kernel versions from 1.0.9 up -to 1.1.72 (and probably earlier and later versions too). - - -Major new features: - -- Better handling of tapes with defects: When a sector repeatedly - (SOFT_RETRIES in ftape.h) cannot be written to or read from it is - marked as an hard error and gets skipped. - The error correction code can handle up to three of these hard - errors provided there are no other errors in that segment (32 Kb). - -- Allows writing to tapes with defects (although the risk of loosing - data increases !) - Look for the media-defects entry printed with the statistics when - the tape is closed. A non-zero value here shows a bad tape. - [the actual count is wrong (too high), this is a known bug]. - -- Use of backup header segment if first one is failing. - -- Support for extended length tapes with QIC-80: both 425 and 1100 ft. - 0.25 inch tapes are now recognized and handled. - -- Support for new QIC-80 drives with 8 mm `wide' tapes (e.g. Conner - TSM 420). - -- Support for new QIC-3010 and QIC-3020 drives (experimental) with - both 0.25 inch and 8 mm tapes. - -Some minor features were added, a couple of small bugs were fixed and -probably some new ones introduced ;-). - -[lseek() didn't make it into this version] - -Have fun, - -Bas. ----- - LocalWords: ftape MCONFIG mt VFS zftape resp sftape proc subdir MTIOCVOLINFO - LocalWords: MTIOCGETSIZE BOT EOD MTBSF zft kerneld modprobe kdtime contrib TR - LocalWords: MTSETBLK afio uninstall texi www EIO QIC init sft eof aka dma GB - LocalWords: SIGKILL MTIOCFTCMD mmap Iomega FDC fdc io gnumt mtio fc asm inb - LocalWords: outb ft qic frontend TeXinfo irq mach MODVERSIONS CONFIG html dvi - LocalWords: usr doc SMP Mb Dunno FIXME vtblc perl listtape volinfo fsf MTWEOF - LocalWords: amanda degaussed ComByte DoublePlay whraven njackn com MTIOC vtbl - LocalWords: GETBLKSZ MAKEDEV zftape's linux dif CVS Revison cp MTREW MTOFFL - LocalWords: MTFSF BSF Marcin Dalecki GCC Config cpio swapout Kai Harrekilde - LocalWords: Pederson khp dolphinics Justus claus momo rwth aachen Laarhoven diff --git a/drivers/char/ftape/compressor/Makefile b/drivers/char/ftape/compressor/Makefile deleted file mode 100644 index 1fbd6c4019db..000000000000 --- a/drivers/char/ftape/compressor/Makefile +++ /dev/null @@ -1,31 +0,0 @@ -# -# Copyright (C) 1997 Claus-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/compressor/Makefile,v $ -# $Revision: 1.1 $ -# $Date: 1997/10/05 19:12:28 $ -# -# Makefile for the optional compressor for th zftape VFS -# interface to the QIC-40/80/3010/3020 floppy-tape driver for -# Linux. -# - -obj-$(CONFIG_ZFT_COMPRESSOR) += zft-compressor.o - -zft-compressor-objs := zftape-compress.o lzrw3.o - -CFLAGS_lzrw3.o := -O6 -funroll-all-loops diff --git a/drivers/char/ftape/compressor/lzrw3.c b/drivers/char/ftape/compressor/lzrw3.c deleted file mode 100644 index a032a0ee2a99..000000000000 --- a/drivers/char/ftape/compressor/lzrw3.c +++ /dev/null @@ -1,743 +0,0 @@ -/* - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/lzrw3.c,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:29 $ - * - * Implementation of Ross Williams lzrw3 algorithm. Adaption for zftape. - * - */ - -#include "../compressor/lzrw3.h" /* Defines single exported function "compress". */ - -/******************************************************************************/ -/* */ -/* LZRW3.C */ -/* */ -/******************************************************************************/ -/* */ -/* Author : Ross Williams. */ -/* Date : 30-Jun-1991. */ -/* Release : 1. */ -/* */ -/******************************************************************************/ -/* */ -/* This file contains an implementation of the LZRW3 data compression */ -/* algorithm in C. */ -/* */ -/* The algorithm is a general purpose compression algorithm that runs fast */ -/* and gives reasonable compression. The algorithm is a member of the Lempel */ -/* Ziv family of algorithms and bases its compression on the presence in the */ -/* data of repeated substrings. */ -/* */ -/* This algorithm is unpatented and the code is public domain. As the */ -/* algorithm is based on the LZ77 class of algorithms, it is unlikely to be */ -/* the subject of a patent challenge. */ -/* */ -/* Unlike the LZRW1 and LZRW1-A algorithms, the LZRW3 algorithm is */ -/* deterministic and is guaranteed to yield the same compressed */ -/* representation for a given file each time it is run. */ -/* */ -/* The LZRW3 algorithm was originally designed and implemented */ -/* by Ross Williams on 31-Dec-1990. */ -/* */ -/* Here are the results of applying this code, compiled under THINK C 4.0 */ -/* and running on a Mac-SE (8MHz 68000), to the standard calgary corpus. */ -/* */ -/* +----------------------------------------------------------------+ */ -/* | DATA COMPRESSION TEST | */ -/* | ===================== | */ -/* | Time of run : Sun 30-Jun-1991 09:31PM | */ -/* | Timing accuracy : One part in 100 | */ -/* | Context length : 262144 bytes (= 256.0000K) | */ -/* | Test suite : Calgary Corpus Suite | */ -/* | Files in suite : 14 | */ -/* | Algorithm : LZRW3 | */ -/* | Note: All averages are calculated from the un-rounded values. | */ -/* +----------------------------------------------------------------+ */ -/* | File Name Length CxB ComLen %Remn Bits Com K/s Dec K/s | */ -/* | ---------- ------ --- ------ ----- ---- ------- ------- | */ -/* | rpus:Bib.D 111261 1 55033 49.5 3.96 19.46 32.27 | */ -/* | us:Book1.D 768771 3 467962 60.9 4.87 17.03 31.07 | */ -/* | us:Book2.D 610856 3 317102 51.9 4.15 19.39 34.15 | */ -/* | rpus:Geo.D 102400 1 82424 80.5 6.44 11.65 18.18 | */ -/* | pus:News.D 377109 2 205670 54.5 4.36 17.14 27.47 | */ -/* | pus:Obj1.D 21504 1 13027 60.6 4.85 13.40 18.95 | */ -/* | pus:Obj2.D 246814 1 116286 47.1 3.77 19.31 30.10 | */ -/* | s:Paper1.D 53161 1 27522 51.8 4.14 18.60 31.15 | */ -/* | s:Paper2.D 82199 1 45160 54.9 4.40 18.45 32.84 | */ -/* | rpus:Pic.D 513216 2 122388 23.8 1.91 35.29 51.05 | */ -/* | us:Progc.D 39611 1 19669 49.7 3.97 18.87 30.64 | */ -/* | us:Progl.D 71646 1 28247 39.4 3.15 24.34 40.66 | */ -/* | us:Progp.D 49379 1 19377 39.2 3.14 23.91 39.23 | */ -/* | us:Trans.D 93695 1 33481 35.7 2.86 25.48 40.37 | */ -/* +----------------------------------------------------------------+ */ -/* | Average 224401 1 110953 50.0 4.00 20.17 32.72 | */ -/* +----------------------------------------------------------------+ */ -/* */ -/******************************************************************************/ - -/******************************************************************************/ - -/* The following structure is returned by the "compress" function below when */ -/* the user asks the function to return identifying information. */ -/* The most important field in the record is the working memory field which */ -/* tells the calling program how much working memory should be passed to */ -/* "compress" when it is called to perform a compression or decompression. */ -/* LZRW3 uses the same amount of memory during compression and decompression. */ -/* For more information on this structure see "compress.h". */ - -#define U(X) ((ULONG) X) -#define SIZE_P_BYTE (U(sizeof(UBYTE *))) -#define SIZE_WORD (U(sizeof(UWORD ))) -#define ALIGNMENT_FUDGE (U(16)) -#define MEM_REQ ( U(4096)*(SIZE_P_BYTE) + ALIGNMENT_FUDGE ) - -static struct compress_identity identity = -{ - U(0x032DDEA8), /* Algorithm identification number. */ - MEM_REQ, /* Working memory (bytes) required. */ - "LZRW3", /* Name of algorithm. */ - "1.0", /* Version number of algorithm. */ - "31-Dec-1990", /* Date of algorithm. */ - "Public Domain", /* Copyright notice. */ - "Ross N. Williams", /* Author of algorithm. */ - "Renaissance Software", /* Affiliation of author. */ - "Public Domain" /* Vendor of algorithm. */ -}; - -LOCAL void compress_compress (UBYTE *,UBYTE *,ULONG,UBYTE *, LONG *); -LOCAL void compress_decompress(UBYTE *,UBYTE *,LONG, UBYTE *, ULONG *); - -/******************************************************************************/ - -/* This function is the only function exported by this module. */ -/* Depending on its first parameter, the function can be requested to */ -/* compress a block of memory, decompress a block of memory, or to identify */ -/* itself. For more information, see the specification file "compress.h". */ - -EXPORT void lzrw3_compress( - UWORD action, /* Action to be performed. */ - UBYTE *wrk_mem, /* Address of working memory we can use.*/ - UBYTE *src_adr, /* Address of input data. */ - LONG src_len, /* Length of input data. */ - UBYTE *dst_adr, /* Address to put output data. */ - void *p_dst_len /* Address of longword for length of output data.*/ -) -{ - switch (action) - { - case COMPRESS_ACTION_IDENTITY: - *((struct compress_identity **)p_dst_len)= &identity; - break; - case COMPRESS_ACTION_COMPRESS: - compress_compress(wrk_mem,src_adr,src_len,dst_adr,(LONG *)p_dst_len); - break; - case COMPRESS_ACTION_DECOMPRESS: - compress_decompress(wrk_mem,src_adr,src_len,dst_adr,(LONG *)p_dst_len); - break; - } -} - -/******************************************************************************/ -/* */ -/* BRIEF DESCRIPTION OF THE LZRW3 ALGORITHM */ -/* ======================================== */ -/* The LZRW3 algorithm is identical to the LZRW1-A algorithm except that */ -/* instead of transmitting history offsets, it transmits hash table indexes. */ -/* In order to decode the indexes, the decompressor must maintain an */ -/* identical hash table. Copy items are straightforward:when the decompressor */ -/* receives a copy item, it simply looks up the hash table to translate the */ -/* index into a pointer into the data already decompressed. To update the */ -/* hash table, it replaces the same table entry with a pointer to the start */ -/* of the newly decoded phrase. The tricky part is with literal items, for at */ -/* the time that the decompressor receives a literal item the decompressor */ -/* does not have the three bytes in the Ziv (that the compressor has) to */ -/* perform the three-byte hash. To solve this problem, in LZRW3, both the */ -/* compressor and decompressor are wired up so that they "buffer" these */ -/* literals and update their hash tables only when three bytes are available. */ -/* This makes the maximum buffering 2 bytes. */ -/* */ -/* Replacement of offsets by hash table indexes yields a few percent extra */ -/* compression at the cost of some speed. LZRW3 is slower than LZRW1, LZRW1-A */ -/* and LZRW2, but yields better compression. */ -/* */ -/* Extra compression could be obtained by using a hash table of depth two. */ -/* However, increasing the depth above one incurs a significant decrease in */ -/* compression speed which was not considered worthwhile. Another reason for */ -/* keeping the depth down to one was to allow easy comparison with the */ -/* LZRW1-A and LZRW2 algorithms so as to demonstrate the exact effect of the */ -/* use of direct hash indexes. */ -/* */ -/* +---+ */ -/* |___|4095 */ -/* |___| */ -/* +---------------------*_|<---+ /----+---\ */ -/* | |___| +---|Hash | */ -/* | |___| |Function| */ -/* | |___| \--------/ */ -/* | |___|0 ^ */ -/* | +---+ | */ -/* | Hash +-----+ */ -/* | Table | */ -/* | --- */ -/* v ^^^ */ -/* +-------------------------------------|----------------+ */ -/* |||||||||||||||||||||||||||||||||||||||||||||||||||||||| */ -/* +-------------------------------------|----------------+ */ -/* | |1......18| | */ -/* |<------- Lempel=History ------------>|<--Ziv-->| | */ -/* | (=bytes already processed) |<-Still to go-->| */ -/* |<-------------------- INPUT BLOCK ------------------->| */ -/* */ -/* The diagram above for LZRW3 looks almost identical to the diagram for */ -/* LZRW1. The difference is that in LZRW3, the compressor transmits hash */ -/* table indices instead of Lempel offsets. For this to work, the */ -/* decompressor must maintain a hash table as well as the compressor and both */ -/* compressor and decompressor must "buffer" literals, as the decompressor */ -/* cannot hash phrases commencing with a literal until another two bytes have */ -/* arrived. */ -/* */ -/* LZRW3 Algorithm Execution Summary */ -/* --------------------------------- */ -/* 1. Hash the first three bytes of the Ziv to yield a hash table index h. */ -/* 2. Look up the hash table yielding history pointer p. */ -/* 3. Match where p points with the Ziv. If there is a match of three or */ -/* more bytes, code those bytes (in the Ziv) as a copy item, otherwise */ -/* code the next byte in the Ziv as a literal item. */ -/* 4. Update the hash table as possible subject to the constraint that only */ -/* phrases commencing three bytes back from the Ziv can be hashed and */ -/* entered into the hash table. (This enables the decompressor to keep */ -/* pace). See the description and code for more details. */ -/* */ -/******************************************************************************/ -/* */ -/* DEFINITION OF COMPRESSED FILE FORMAT */ -/* ==================================== */ -/* * A compressed file consists of a COPY FLAG followed by a REMAINDER. */ -/* * The copy flag CF uses up four bytes with the first byte being the */ -/* least significant. */ -/* * If CF=1, then the compressed file represents the remainder of the file */ -/* exactly. Otherwise CF=0 and the remainder of the file consists of zero */ -/* or more GROUPS, each of which represents one or more bytes. */ -/* * Each group consists of two bytes of CONTROL information followed by */ -/* sixteen ITEMs except for the last group which can contain from one */ -/* to sixteen items. */ -/* * An item can be either a LITERAL item or a COPY item. */ -/* * Each item corresponds to a bit in the control bytes. */ -/* * The first control byte corresponds to the first 8 items in the group */ -/* with bit 0 corresponding to the first item in the group and bit 7 to */ -/* the eighth item in the group. */ -/* * The second control byte corresponds to the second 8 items in the group */ -/* with bit 0 corresponding to the ninth item in the group and bit 7 to */ -/* the sixteenth item in the group. */ -/* * A zero bit in a control word means that the corresponding item is a */ -/* literal item. A one bit corresponds to a copy item. */ -/* * A literal item consists of a single byte which represents itself. */ -/* * A copy item consists of two bytes that represent from 3 to 18 bytes. */ -/* * The first byte in a copy item will be denoted C1. */ -/* * The second byte in a copy item will be denoted C2. */ -/* * Bits will be selected using square brackets. */ -/* For example: C1[0..3] is the low nibble of the first control byte. */ -/* of copy item C1. */ -/* * The LENGTH of a copy item is defined to be C1[0..3]+3 which is a number */ -/* in the range [3,18]. */ -/* * The INDEX of a copy item is defined to be C1[4..7]*256+C2[0..8] which */ -/* is a number in the range [0,4095]. */ -/* * A copy item represents the sequence of bytes */ -/* text[POS-OFFSET..POS-OFFSET+LENGTH-1] where */ -/* text is the entire text of the uncompressed string. */ -/* POS is the index in the text of the character following the */ -/* string represented by all the items preceeding the item */ -/* being defined. */ -/* OFFSET is obtained from INDEX by looking up the hash table. */ -/* */ -/******************************************************************************/ - -/* The following #define defines the length of the copy flag that appears at */ -/* the start of the compressed file. The value of four bytes was chosen */ -/* because the fast_copy routine on my Macintosh runs faster if the source */ -/* and destination blocks are relatively longword aligned. */ -/* The actual flag data appears in the first byte. The rest are zeroed so as */ -/* to normalize the compressed representation (i.e. not non-deterministic). */ -#define FLAG_BYTES 4 - -/* The following #defines define the meaning of the values of the copy */ -/* flag at the start of the compressed file. */ -#define FLAG_COMPRESS 0 /* Signals that output was result of compression. */ -#define FLAG_COPY 1 /* Signals that output was simply copied over. */ - -/* The 68000 microprocessor (on which this algorithm was originally developed */ -/* is fussy about non-aligned arrays of words. To avoid these problems the */ -/* following macro can be used to "waste" from 0 to 3 bytes so as to align */ -/* the argument pointer. */ -#define ULONG_ALIGN_UP(X) ((((ULONG)X)+sizeof(ULONG)-1)&~(sizeof(ULONG)-1)) - - -/* The following constant defines the maximum length of an uncompressed item. */ -/* This definition must not be changed; its value is hardwired into the code. */ -/* The longest number of bytes that can be spanned by a single item is 18 */ -/* for the longest copy item. */ -#define MAX_RAW_ITEM (18) - -/* The following constant defines the maximum length of an uncompressed group.*/ -/* This definition must not be changed; its value is hardwired into the code. */ -/* A group contains at most 16 items which explains this definition. */ -#define MAX_RAW_GROUP (16*MAX_RAW_ITEM) - -/* The following constant defines the maximum length of a compressed group. */ -/* This definition must not be changed; its value is hardwired into the code. */ -/* A compressed group consists of two control bytes followed by up to 16 */ -/* compressed items each of which can have a maximum length of two bytes. */ -#define MAX_CMP_GROUP (2+16*2) - -/* The following constant defines the number of entries in the hash table. */ -/* This definition must not be changed; its value is hardwired into the code. */ -#define HASH_TABLE_LENGTH (4096) - -/* LZRW3, unlike LZRW1(-A), must initialize its hash table so as to enable */ -/* the compressor and decompressor to stay in step maintaining identical hash */ -/* tables. In an early version of the algorithm, the tables were simply */ -/* initialized to zero and a check for zero was included just before the */ -/* matching code. However, this test costs time. A better solution is to */ -/* initialize all the entries in the hash table to point to a constant */ -/* string. The decompressor does the same. This solution requires no extra */ -/* test. The contents of the string do not matter so long as the string is */ -/* the same for the compressor and decompressor and contains at least */ -/* MAX_RAW_ITEM bytes. I chose consecutive decimal digits because they do not */ -/* have white space problems (e.g. there is no chance that the compiler will */ -/* replace more than one space by a TAB) and because they make the length of */ -/* the string obvious by inspection. */ -#define START_STRING_18 ((UBYTE *) "123456789012345678") - -/* In this algorithm, hash values have to be calculated at more than one */ -/* point. The following macro neatens the code up for this. */ -#define HASH(PTR) \ - (((40543*(((*(PTR))<<8)^((*((PTR)+1))<<4)^(*((PTR)+2))))>>4) & 0xFFF) - -/******************************************************************************/ - -/* Input : Hand over the required amount of working memory in p_wrk_mem. */ -/* Input : Specify input block using p_src_first and src_len. */ -/* Input : Point p_dst_first to the start of the output zone (OZ). */ -/* Input : Point p_dst_len to a ULONG to receive the output length. */ -/* Input : Input block and output zone must not overlap. */ -/* Output : Length of output block written to *p_dst_len. */ -/* Output : Output block in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. May */ -/* Output : write in OZ=Mem[p_dst_first..p_dst_first+src_len+MAX_CMP_GROUP-1].*/ -/* Output : Upon completion guaranteed *p_dst_len<=src_len+FLAG_BYTES. */ -LOCAL void compress_compress(UBYTE *p_wrk_mem, - UBYTE *p_src_first, ULONG src_len, - UBYTE *p_dst_first, LONG *p_dst_len) -{ - /* p_src and p_dst step through the source and destination blocks. */ - register UBYTE *p_src = p_src_first; - register UBYTE *p_dst = p_dst_first; - - /* The following variables are never modified and are used in the */ - /* calculations that determine when the main loop terminates. */ - UBYTE *p_src_post = p_src_first+src_len; - UBYTE *p_dst_post = p_dst_first+src_len; - UBYTE *p_src_max1 = p_src_first+src_len-MAX_RAW_ITEM; - UBYTE *p_src_max16 = p_src_first+src_len-MAX_RAW_ITEM*16; - - /* The variables 'p_control' and 'control' are used to buffer control bits. */ - /* Before each group is processed, the next two bytes of the output block */ - /* are set aside for the control word for the group about to be processed. */ - /* 'p_control' is set to point to the first byte of that word. Meanwhile, */ - /* 'control' buffers the control bits being generated during the processing */ - /* of the group. Instead of having a counter to keep track of how many items */ - /* have been processed (=the number of bits in the control word), at the */ - /* start of each group, the top word of 'control' is filled with 1 bits. */ - /* As 'control' is shifted for each item, the 1 bits in the top word are */ - /* absorbed or destroyed. When they all run out (i.e. when the top word is */ - /* all zero bits, we know that we are at the end of a group. */ -# define TOPWORD 0xFFFF0000 - UBYTE *p_control; - register ULONG control=TOPWORD; - - /* THe variable 'hash' always points to the first element of the hash table. */ - UBYTE **hash= (UBYTE **) ULONG_ALIGN_UP(p_wrk_mem); - - /* The following two variables represent the literal buffer. p_h1 points to */ - /* the hash table entry corresponding to the youngest literal. p_h2 points */ - /* to the hash table entry corresponding to the second youngest literal. */ - /* Note: p_h1=0=>p_h2=0 because zero values denote absence of a pending */ - /* literal. The variables are initialized to zero meaning an empty "buffer". */ - UBYTE **p_h1=NULL; - UBYTE **p_h2=NULL; - - /* To start, we write the flag bytes. Being optimistic, we set the flag to */ - /* FLAG_COMPRESS. The remaining flag bytes are zeroed so as to keep the */ - /* algorithm deterministic. */ - *p_dst++=FLAG_COMPRESS; - {UWORD i; for (i=2;i<=FLAG_BYTES;i++) *p_dst++=0;} - - /* Reserve the first word of output as the control word for the first group. */ - /* Note: This is undone at the end if the input block is empty. */ - p_control=p_dst; p_dst+=2; - - /* Initialize all elements of the hash table to point to a constant string. */ - /* Use of an unrolled loop speeds this up considerably. */ - {UWORD i; UBYTE **p_h=hash; -# define ZH *p_h++=START_STRING_18 - for (i=0;i<256;i++) /* 256=HASH_TABLE_LENGTH/16. */ - {ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH; - ZH;ZH;ZH;ZH;} - } - - /* The main loop processes either 1 or 16 items per iteration. As its */ - /* termination logic is complicated, I have opted for an infinite loop */ - /* structure containing 'break' and 'goto' statements. */ - while (TRUE) - {/* Begin main processing loop. */ - - /* Note: All the variables here except unroll should be defined within */ - /* the inner loop. Unfortunately the loop hasn't got a block. */ - register UBYTE *p; /* Scans through targ phrase during matching. */ - register UBYTE *p_ziv= NULL ; /* Points to first byte of current Ziv. */ - register UWORD unroll; /* Loop counter for unrolled inner loop. */ - register UWORD index; /* Index of current hash table entry. */ - register UBYTE **p_h0 = NULL ; /* Pointer to current hash table entry. */ - - /* Test for overrun and jump to overrun code if necessary. */ - if (p_dst>p_dst_post) - goto overrun; - - /* The following cascade of if statements efficiently catches and deals */ - /* with varying degrees of closeness to the end of the input block. */ - /* When we get very close to the end, we stop updating the table and */ - /* code the remaining bytes as literals. This makes the code simpler. */ - unroll=16; - if (p_src>p_src_max16) - { - unroll=1; - if (p_src>p_src_max1) - { - if (p_src==p_src_post) - break; - else - goto literal; - } - } - - /* This inner unrolled loop processes 'unroll' (whose value is either 1 */ - /* or 16) items. I have chosen to implement this loop with labels and */ - /* gotos to heighten the ease with which the loop may be implemented with */ - /* a single decrement and branch instruction in assembly language and */ - /* also because the labels act as highly readable place markers. */ - /* (Also because we jump into the loop for endgame literals (see above)). */ - - begin_unrolled_loop: - - /* To process the next phrase, we hash the next three bytes and use */ - /* the resultant hash table index to look up the hash table. A pointer */ - /* to the entry is stored in p_h0 so as to avoid an array lookup. The */ - /* hash table entry *p_h0 is looked up yielding a pointer p to a */ - /* potential match of the Ziv in the history. */ - index=HASH(p_src); - p_h0=&hash[index]; - p=*p_h0; - - /* Having looked up the candidate position, we are in a position to */ - /* attempt a match. The match loop has been unrolled using the PS */ - /* macro so that failure within the first three bytes automatically */ - /* results in the literal branch being taken. The coding is simple. */ - /* p_ziv saves p_src so we can let p_src wander. */ -# define PS *p++!=*p_src++ - p_ziv=p_src; - if (PS || PS || PS) - { - /* Literal. */ - - /* Code the literal byte as itself and a zero control bit. */ - p_src=p_ziv; literal: *p_dst++=*p_src++; control&=0xFFFEFFFF; - - /* We have just coded a literal. If we had two pending ones, that */ - /* makes three and we can update the hash table. */ - if (p_h2!=0) - {*p_h2=p_ziv-2;} - - /* In any case, rotate the hash table pointers for next time. */ - p_h2=p_h1; p_h1=p_h0; - - } - else - { - /* Copy */ - - /* Match up to 15 remaining bytes using an unrolled loop and code. */ -#if 0 - PS || PS || PS || PS || PS || PS || PS || PS || - PS || PS || PS || PS || PS || PS || PS || p_src++; -#else - if ( - !( PS || PS || PS || PS || PS || PS || PS || PS || - PS || PS || PS || PS || PS || PS || PS ) - ) p_src++; -#endif - *p_dst++=((index&0xF00)>>4)|(--p_src-p_ziv-3); - *p_dst++=index&0xFF; - - /* As we have just coded three bytes, we are now in a position to */ - /* update the hash table with the literal bytes that were pending */ - /* upon the arrival of extra context bytes. */ - if (p_h1!=0) - { - if (p_h2) - {*p_h2=p_ziv-2; p_h2=NULL;} - *p_h1=p_ziv-1; p_h1=NULL; - } - - /* In any case, we can update the hash table based on the current */ - /* position as we just coded at least three bytes in a copy items. */ - *p_h0=p_ziv; - - } - control>>=1; - - /* This loop is all set up for a decrement and jump instruction! */ -#ifndef linux -` end_unrolled_loop: if (--unroll) goto begin_unrolled_loop; -#else - /* end_unrolled_loop: */ if (--unroll) goto begin_unrolled_loop; -#endif - - /* At this point it will nearly always be the end of a group in which */ - /* case, we have to do some control-word processing. However, near the */ - /* end of the input block, the inner unrolled loop is only executed once. */ - /* This necessitates the 'if' test. */ - if ((control&TOPWORD)==0) - { - /* Write the control word to the place we saved for it in the output. */ - *p_control++= control &0xFF; - *p_control = (control>>8) &0xFF; - - /* Reserve the next word in the output block for the control word */ - /* for the group about to be processed. */ - p_control=p_dst; p_dst+=2; - - /* Reset the control bits buffer. */ - control=TOPWORD; - } - - } /* End main processing loop. */ - - /* After the main processing loop has executed, all the input bytes have */ - /* been processed. However, the control word has still to be written to the */ - /* word reserved for it in the output at the start of the most recent group. */ - /* Before writing, the control word has to be shifted so that all the bits */ - /* are in the right place. The "empty" bit positions are filled with 1s */ - /* which partially fill the top word. */ - while(control&TOPWORD) control>>=1; - *p_control++= control &0xFF; - *p_control++=(control>>8) &0xFF; - - /* If the last group contained no items, delete the control word too. */ - if (p_control==p_dst) p_dst-=2; - - /* Write the length of the output block to the dst_len parameter and return. */ - *p_dst_len=p_dst-p_dst_first; - return; - - /* Jump here as soon as an overrun is detected. An overrun is defined to */ - /* have occurred if p_dst>p_dst_first+src_len. That is, the moment the */ - /* length of the output written so far exceeds the length of the input block.*/ - /* The algorithm checks for overruns at least at the end of each group */ - /* which means that the maximum overrun is MAX_CMP_GROUP bytes. */ - /* Once an overrun occurs, the only thing to do is to set the copy flag and */ - /* copy the input over. */ - overrun: -#if 0 - *p_dst_first=FLAG_COPY; - fast_copy(p_src_first,p_dst_first+FLAG_BYTES,src_len); - *p_dst_len=src_len+FLAG_BYTES; -#else - fast_copy(p_src_first,p_dst_first,src_len); - *p_dst_len= -src_len; /* return a negative number to indicate uncompressed data */ -#endif -} - -/******************************************************************************/ - -/* Input : Hand over the required amount of working memory in p_wrk_mem. */ -/* Input : Specify input block using p_src_first and src_len. */ -/* Input : Point p_dst_first to the start of the output zone. */ -/* Input : Point p_dst_len to a ULONG to receive the output length. */ -/* Input : Input block and output zone must not overlap. User knows */ -/* Input : upperbound on output block length from earlier compression. */ -/* Input : In any case, maximum expansion possible is nine times. */ -/* Output : Length of output block written to *p_dst_len. */ -/* Output : Output block in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. */ -/* Output : Writes only in Mem[p_dst_first..p_dst_first+*p_dst_len-1]. */ -LOCAL void compress_decompress( UBYTE *p_wrk_mem, - UBYTE *p_src_first, LONG src_len, - UBYTE *p_dst_first, ULONG *p_dst_len) -{ - /* Byte pointers p_src and p_dst scan through the input and output blocks. */ - register UBYTE *p_src = p_src_first+FLAG_BYTES; - register UBYTE *p_dst = p_dst_first; - /* we need to avoid a SEGV when trying to uncompress corrupt data */ - register UBYTE *p_dst_post = p_dst_first + *p_dst_len; - - /* The following two variables are never modified and are used to control */ - /* the main loop. */ - UBYTE *p_src_post = p_src_first+src_len; - UBYTE *p_src_max16 = p_src_first+src_len-(MAX_CMP_GROUP-2); - - /* The hash table is the only resident of the working memory. The hash table */ - /* contains HASH_TABLE_LENGTH=4096 pointers to positions in the history. To */ - /* keep Macintoshes happy, it is longword aligned. */ - UBYTE **hash = (UBYTE **) ULONG_ALIGN_UP(p_wrk_mem); - - /* The variable 'control' is used to buffer the control bits which appear in */ - /* groups of 16 bits (control words) at the start of each compressed group. */ - /* When each group is read, bit 16 of the register is set to one. Whenever */ - /* a new bit is needed, the register is shifted right. When the value of the */ - /* register becomes 1, we know that we have reached the end of a group. */ - /* Initializing the register to 1 thus instructs the code to follow that it */ - /* should read a new control word immediately. */ - register ULONG control=1; - - /* The value of 'literals' is always in the range 0..3. It is the number of */ - /* consecutive literal items just seen. We have to record this number so as */ - /* to know when to update the hash table. When literals gets to 3, there */ - /* have been three consecutive literals and we can update at the position of */ - /* the oldest of the three. */ - register UWORD literals=0; - - /* Check the leading copy flag to see if the compressor chose to use a copy */ - /* operation instead of a compression operation. If a copy operation was */ - /* used, then all we need to do is copy the data over, set the output length */ - /* and return. */ -#if 0 - if (*p_src_first==FLAG_COPY) - { - fast_copy(p_src_first+FLAG_BYTES,p_dst_first,src_len-FLAG_BYTES); - *p_dst_len=src_len-FLAG_BYTES; - return; - } -#else - if ( src_len < 0 ) - { - fast_copy(p_src_first,p_dst_first,-src_len ); - *p_dst_len = (ULONG)-src_len; - return; - } -#endif - - /* Initialize all elements of the hash table to point to a constant string. */ - /* Use of an unrolled loop speeds this up considerably. */ - {UWORD i; UBYTE **p_h=hash; -# define ZJ *p_h++=START_STRING_18 - for (i=0;i<256;i++) /* 256=HASH_TABLE_LENGTH/16. */ - {ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ; - ZJ;ZJ;ZJ;ZJ;} - } - - /* The outer loop processes either 1 or 16 items per iteration depending on */ - /* how close p_src is to the end of the input block. */ - while (p_src!=p_src_post) - {/* Start of outer loop */ - - register UWORD unroll; /* Counts unrolled loop executions. */ - - /* When 'control' has the value 1, it means that the 16 buffered control */ - /* bits that were read in at the start of the current group have all been */ - /* shifted out and that all that is left is the 1 bit that was injected */ - /* into bit 16 at the start of the current group. When we reach the end */ - /* of a group, we have to load a new control word and inject a new 1 bit. */ - if (control==1) - { - control=0x10000|*p_src++; - control|=(*p_src++)<<8; - } - - /* If it is possible that we are within 16 groups from the end of the */ - /* input, execute the unrolled loop only once, else process a whole group */ - /* of 16 items by looping 16 times. */ - unroll= p_src<=p_src_max16 ? 16 : 1; - - /* This inner loop processes one phrase (item) per iteration. */ - while (unroll--) - { /* Begin unrolled inner loop. */ - - /* Process a literal or copy item depending on the next control bit. */ - if (control&1) - { - /* Copy item. */ - - register UBYTE *p; /* Points to place from which to copy. */ - register UWORD lenmt; /* Length of copy item minus three. */ - register UBYTE **p_hte; /* Pointer to current hash table entry.*/ - register UBYTE *p_ziv=p_dst; /* Pointer to start of current Ziv. */ - - /* Read and dismantle the copy word. Work out from where to copy. */ - lenmt=*p_src++; - p_hte=&hash[((lenmt&0xF0)<<4)|*p_src++]; - p=*p_hte; - lenmt&=0xF; - - /* Now perform the copy using a half unrolled loop. */ - *p_dst++=*p++; - *p_dst++=*p++; - *p_dst++=*p++; - while (lenmt--) - *p_dst++=*p++; - - /* Because we have just received 3 or more bytes in a copy item */ - /* (whose bytes we have just installed in the output), we are now */ - /* in a position to flush all the pending literal hashings that had */ - /* been postponed for lack of bytes. */ - if (literals>0) - { - register UBYTE *r=p_ziv-literals; - hash[HASH(r)]=r; - if (literals==2) - {r++; hash[HASH(r)]=r;} - literals=0; - } - - /* In any case, we can immediately update the hash table with the */ - /* current position. We don't need to do a HASH(...) to work out */ - /* where to put the pointer, as the compressor just told us!!! */ - *p_hte=p_ziv; - - } - else - { - /* Literal item. */ - - /* Copy over the literal byte. */ - *p_dst++=*p_src++; - - /* If we now have three literals waiting to be hashed into the hash */ - /* table, we can do one of them now (because there are three). */ - if (++literals == 3) - {register UBYTE *p=p_dst-3; hash[HASH(p)]=p; literals=2;} - } - - /* Shift the control buffer so the next control bit is in bit 0. */ - control>>=1; -#if 1 - if (p_dst > p_dst_post) - { - /* Shit: we tried to decompress corrupt data */ - *p_dst_len = 0; - return; - } -#endif - } /* End unrolled inner loop. */ - - } /* End of outer loop */ - - /* Write the length of the decompressed data before returning. */ - *p_dst_len=p_dst-p_dst_first; -} - -/******************************************************************************/ -/* End of LZRW3.C */ -/******************************************************************************/ diff --git a/drivers/char/ftape/compressor/lzrw3.h b/drivers/char/ftape/compressor/lzrw3.h deleted file mode 100644 index 533feba47526..000000000000 --- a/drivers/char/ftape/compressor/lzrw3.h +++ /dev/null @@ -1,253 +0,0 @@ -#ifndef _LZRW3_H -#define _LZRW3_H -/* - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/lzrw3.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:30 $ - * - * include files for lzrw3. Only slighty modified from the original - * version. Assembles the three include files compress.h, port.h and - * fastcopy.h from the original lzrw3 package. - * - */ - -#include -#include - -/******************************************************************************/ -/* */ -/* COMPRESS.H */ -/* */ -/******************************************************************************/ -/* */ -/* Author : Ross Williams. */ -/* Date : December 1989. */ -/* */ -/* This header file defines the interface to a set of functions called */ -/* 'compress', each member of which implements a particular data compression */ -/* algorithm. */ -/* */ -/* Normally in C programming, for each .H file, there is a corresponding .C */ -/* file that implements the functions promised in the .H file. */ -/* Here, there are many .C files corresponding to this header file. */ -/* Each comforming implementation file contains a single function */ -/* called 'compress' that implements a single data compression */ -/* algorithm that conforms with the interface specified in this header file. */ -/* Only one algorithm can be linked in at a time in this organization. */ -/* */ -/******************************************************************************/ -/* */ -/* DEFINITION OF FUNCTION COMPRESS */ -/* =============================== */ -/* */ -/* Summary of Function Compress */ -/* ---------------------------- */ -/* The action that 'compress' takes depends on its first argument called */ -/* 'action'. The function provides three actions: */ -/* */ -/* - Return information about the algorithm. */ -/* - Compress a block of memory. */ -/* - Decompress a block of memory. */ -/* */ -/* Parameters */ -/* ---------- */ -/* See the formal C definition later for a description of the parameters. */ -/* */ -/* Constants */ -/* --------- */ -/* COMPRESS_OVERRUN: The constant COMPRESS_OVERRUN defines by how many bytes */ -/* an algorithm is allowed to expand a block during a compression operation. */ -/* */ -/* Although compression algorithms usually compress data, there will always */ -/* be data that a given compressor will expand (this can be proven). */ -/* Fortunately, the degree of expansion can be limited to a single bit, by */ -/* copying over the input data if the data gets bigger during compression. */ -/* To allow for this possibility, the first bit of a compressed */ -/* representation can be used as a flag indicating whether the */ -/* input data was copied over, or truly compressed. In practice, the first */ -/* byte would be used to store this bit so as to maintain byte alignment. */ -/* */ -/* Unfortunately, in general, the only way to tell if an algorithm will */ -/* expand a particular block of data is to run the algorithm on the data. */ -/* If the algorithm does not continuously monitor how many output bytes it */ -/* has written, it might write an output block far larger than the input */ -/* block before realizing that it has done so. */ -/* On the other hand, continuous checks on output length are inefficient. */ -/* */ -/* To cater for all these problems, this interface definition: */ -/* > Allows a compression algorithm to return an output block that is up to */ -/* COMPRESS_OVERRUN bytes longer than the input block. */ -/* > Allows a compression algorithm to write up to COMPRESS_OVERRUN bytes */ -/* more than the length of the input block to the memory of the output */ -/* block regardless of the length of the output block eventually returned. */ -/* This allows an algorithm to overrun the length of the input block in the */ -/* output block by up to COMPRESS_OVERRUN bytes between expansion checks. */ -/* */ -/* The problem does not arise for decompression. */ -/* */ -/* Identity Action */ -/* --------------- */ -/* > action must be COMPRESS_ACTION_IDENTITY. */ -/* > p_dst_len must point to a longword to receive a longword address. */ -/* > The value of the other parameters does not matter. */ -/* > After execution, the longword that p_dst_len points to will be a pointer */ -/* to a structure of type compress_identity. */ -/* Thus, for example, after the call, (*p_dst_len)->memory will return the */ -/* number of bytes of working memory that the algorithm requires to run. */ -/* > The values of the identity structure returned are fixed constant */ -/* attributes of the algorithm and must not vary from call to call. */ -/* */ -/* Common Requirements for Compression and Decompression Actions */ -/* ------------------------------------------------------------- */ -/* > wrk_mem must point to an unused block of memory of a length specified in */ -/* the algorithm's identity block. The identity block can be obtained by */ -/* making a separate call to compress, specifying the identity action. */ -/* > The INPUT BLOCK is defined to be Memory[src_addr,src_addr+src_len-1]. */ -/* > dst_len will be used to denote *p_dst_len. */ -/* > dst_len is not read by compress, only written. */ -/* > The value of dst_len is defined only upon termination. */ -/* > The OUTPUT BLOCK is defined to be Memory[dst_addr,dst_addr+dst_len-1]. */ -/* */ -/* Compression Action */ -/* ------------------ */ -/* > action must be COMPRESS_ACTION_COMPRESS. */ -/* > src_len must be in the range [0,COMPRESS_MAX_ORG]. */ -/* > The OUTPUT ZONE is defined to be */ -/* Memory[dst_addr,dst_addr+src_len-1+COMPRESS_OVERRUN]. */ -/* > The function can modify any part of the output zone regardless of the */ -/* final length of the output block. */ -/* > The input block and the output zone must not overlap. */ -/* > dst_len will be in the range [0,src_len+COMPRESS_OVERRUN]. */ -/* > dst_len will be in the range [0,COMPRESS_MAX_COM] (from prev fact). */ -/* > The output block will consist of a representation of the input block. */ -/* */ -/* Decompression Action */ -/* -------------------- */ -/* > action must be COMPRESS_ACTION_DECOMPRESS. */ -/* > The input block must be the result of an earlier compression operation. */ -/* > If the previous fact is true, the following facts must also be true: */ -/* > src_len will be in the range [0,COMPRESS_MAX_COM]. */ -/* > dst_len will be in the range [0,COMPRESS_MAX_ORG]. */ -/* > The input and output blocks must not overlap. */ -/* > Only the output block is modified. */ -/* > Upon termination, the output block will consist of the bytes contained */ -/* in the input block passed to the earlier compression operation. */ -/* */ -/******************************************************************************/ - -/******************************************************************************/ -/* */ -/* PORT.H */ -/* */ -/******************************************************************************/ -/* */ -/* This module contains macro definitions and types that are likely to */ -/* change between computers. */ -/* */ -/******************************************************************************/ - -#ifndef DONE_PORT /* Only do this if not previously done. */ - - #ifdef THINK_C - #define UBYTE unsigned char /* Unsigned byte */ - #define UWORD unsigned int /* Unsigned word (2 bytes) */ - #define ULONG unsigned long /* Unsigned word (4 bytes) */ - #define BOOL unsigned char /* Boolean */ - #define FOPEN_BINARY_READ "rb" /* Mode string for binary reading. */ - #define FOPEN_BINARY_WRITE "wb" /* Mode string for binary writing. */ - #define FOPEN_TEXT_APPEND "a" /* Mode string for text appending. */ - #define REAL double /* USed for floating point stuff. */ - #endif - #if defined(LINUX) || defined(linux) - #define UBYTE __u8 /* Unsigned byte */ - #define UWORD __u16 /* Unsigned word (2 bytes) */ - #define ULONG __u32 /* Unsigned word (4 bytes) */ - #define LONG __s32 /* Signed word (4 bytes) */ - #define BOOL is not used here /* Boolean */ - #define FOPEN_BINARY_READ not used /* Mode string for binary reading. */ - #define FOPEN_BINARY_WRITE not used /* Mode string for binary writing. */ - #define FOPEN_TEXT_APPEND not used /* Mode string for text appending. */ - #define REAL not used /* USed for floating point stuff. */ - #ifndef TRUE - #define TRUE 1 - #endif - #endif - - #define DONE_PORT /* Don't do all this again. */ - #define MALLOC_FAIL NULL /* Failure status from malloc() */ - #define LOCAL static /* For non-exported routines. */ - #define EXPORT /* Signals exported function. */ - #define then /* Useful for aligning ifs. */ - -#endif - -/******************************************************************************/ -/* End of PORT.H */ -/******************************************************************************/ - -#define COMPRESS_ACTION_IDENTITY 0 -#define COMPRESS_ACTION_COMPRESS 1 -#define COMPRESS_ACTION_DECOMPRESS 2 - -#define COMPRESS_OVERRUN 1024 -#define COMPRESS_MAX_COM 0x70000000 -#define COMPRESS_MAX_ORG (COMPRESS_MAX_COM-COMPRESS_OVERRUN) - -#define COMPRESS_MAX_STRLEN 255 - -/* The following structure provides information about the algorithm. */ -/* > The top bit of id must be zero. The remaining bits must be chosen by */ -/* the author of the algorithm by tossing a coin 31 times. */ -/* > The amount of memory requested by the algorithm is specified in bytes */ -/* and must be in the range [0,0x70000000]. */ -/* > All strings s must be such that strlen(s)<=COMPRESS_MAX_STRLEN. */ -struct compress_identity - { - ULONG id; /* Identifying number of algorithm. */ - ULONG memory; /* Number of bytes of working memory required. */ - - char *name; /* Name of algorithm. */ - char *version; /* Version number. */ - char *date; /* Date of release of this version. */ - char *copyright; /* Copyright message. */ - - char *author; /* Author of algorithm. */ - char *affiliation; /* Affiliation of author. */ - char *vendor; /* Where the algorithm can be obtained. */ - }; - -void lzrw3_compress( /* Single function interface to compression algorithm. */ -UWORD action, /* Action to be performed. */ -UBYTE *wrk_mem, /* Working memory temporarily given to routine to use. */ -UBYTE *src_adr, /* Address of input data. */ -LONG src_len, /* Length of input data. */ -UBYTE *dst_adr, /* Address of output data. */ -void *p_dst_len /* Pointer to a longword where routine will write: */ - /* If action=..IDENTITY => Adr of id structure. */ - /* If action=..COMPRESS => Length of output data. */ - /* If action=..DECOMPRESS => Length of output data. */ -); - -/******************************************************************************/ -/* End of COMPRESS.H */ -/******************************************************************************/ - - -/******************************************************************************/ -/* fast_copy.h */ -/******************************************************************************/ - -/* This function copies a block of memory very quickly. */ -/* The exact speed depends on the relative alignment of the blocks of memory. */ -/* PRE : 0<=src_len<=(2^32)-1 . */ -/* PRE : Source and destination blocks must not overlap. */ -/* POST : MEM[dst_adr,dst_adr+src_len-1]=MEM[src_adr,src_adr+src_len-1]. */ -/* POST : MEM[dst_adr,dst_adr+src_len-1] is the only memory changed. */ - -#define fast_copy(src,dst,len) memcpy(dst,src,len) - -/******************************************************************************/ -/* End of fast_copy.h */ -/******************************************************************************/ - -#endif diff --git a/drivers/char/ftape/compressor/zftape-compress.c b/drivers/char/ftape/compressor/zftape-compress.c deleted file mode 100644 index 65ffc0be3df9..000000000000 --- a/drivers/char/ftape/compressor/zftape-compress.c +++ /dev/null @@ -1,1203 +0,0 @@ -/* - * Copyright (C) 1994-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * This file implements a "generic" interface between the * - * zftape-driver and a compression-algorithm. The * - * compression-algorithm currently used is a LZ77. I use the * - * implementation lzrw3 by Ross N. Williams (Renaissance * - * Software). The compression program itself is in the file - * lzrw3.c * and lzrw3.h. To adopt another compression algorithm - * the functions * zft_compress() and zft_uncompress() must be - * changed * appropriately. See below. - */ - -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../compressor/zftape-compress.h" -#include "../zftape/zftape-vtbl.h" -#include "../compressor/lzrw3.h" - -/* - * global variables - */ - -/* I handle the allocation of this buffer as a special case, because - * it's size varies depending on the tape length inserted. - */ - -/* local variables - */ -static void *zftc_wrk_mem = NULL; -static __u8 *zftc_buf = NULL; -static void *zftc_scratch_buf = NULL; - -/* compression statistics - */ -static unsigned int zftc_wr_uncompressed = 0; -static unsigned int zftc_wr_compressed = 0; -static unsigned int zftc_rd_uncompressed = 0; -static unsigned int zftc_rd_compressed = 0; - -/* forward */ -static int zftc_write(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume); -static int zftc_read(int *read_cnt, - __u8 __user *dst_buf, const int to_do, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume); -static int zftc_seek(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, - __u8 *buffer); -static void zftc_lock (void); -static void zftc_reset (void); -static void zftc_cleanup(void); -static void zftc_stats (void); - -/* compressed segment. This conforms to QIC-80-MC, Revision K. - * - * Rev. K applies to tapes with `fixed length format' which is - * indicated by format code 2,3 and 5. See below for format code 4 and 6 - * - * 2 bytes: offset of compression segment structure - * 29k > offset >= 29k-18: data from previous segment ens in this - * segment and no compressed block starts - * in this segment - * offset == 0: data from previous segment occupies entire - * segment and continues in next segment - * n bytes: remainder from previous segment - * - * Rev. K: - * 4 bytes: 4 bytes: files set byte offset - * Post Rev. K and QIC-3020/3020: - * 8 bytes: 8 bytes: files set byte offset - * 2 bytes: byte count N (amount of data following) - * bit 15 is set if data is compressed, bit 15 is not - * set if data is uncompressed - * N bytes: data (as much as specified in the byte count) - * 2 bytes: byte count N_1 of next cluster - * N_1 bytes: data of next cluset - * 2 bytes: byte count N_2 of next cluster - * N_2 bytes: ... - * - * Note that the `N' byte count accounts only for the bytes that in the - * current segment if the cluster spans to the next segment. - */ - -typedef struct -{ - int cmpr_pos; /* actual position in compression buffer */ - int cmpr_sz; /* what is left in the compression buffer - * when copying the compressed data to the - * deblock buffer - */ - unsigned int first_block; /* location of header information in - * this segment - */ - unsigned int count; /* amount of data of current block - * contained in current segment - */ - unsigned int offset; /* offset in current segment */ - unsigned int spans:1; /* might continue in next segment */ - unsigned int uncmpr; /* 0x8000 if this block contains - * uncompressed data - */ - __s64 foffs; /* file set byte offset, same as in - * compression map segment - */ -} cmpr_info; - -static cmpr_info cseg; /* static data. Must be kept uptodate and shared by - * read, write and seek functions - */ - -#define DUMP_CMPR_INFO(level, msg, info) \ - TRACE(level, msg "\n" \ - KERN_INFO "cmpr_pos : %d\n" \ - KERN_INFO "cmpr_sz : %d\n" \ - KERN_INFO "first_block: %d\n" \ - KERN_INFO "count : %d\n" \ - KERN_INFO "offset : %d\n" \ - KERN_INFO "spans : %d\n" \ - KERN_INFO "uncmpr : 0x%04x\n" \ - KERN_INFO "foffs : " LL_X, \ - (info)->cmpr_pos, (info)->cmpr_sz, (info)->first_block, \ - (info)->count, (info)->offset, (info)->spans == 1, \ - (info)->uncmpr, LL((info)->foffs)) - -/* dispatch compression segment info, return error code - * - * afterwards, cseg->offset points to start of data of the NEXT - * compressed block, and cseg->count contains the amount of data - * left in the actual compressed block. cseg->spans is set to 1 if - * the block is continued in the following segment. Otherwise it is - * set to 0. - */ -static int get_cseg (cmpr_info *cinfo, const __u8 *buff, - const unsigned int seg_sz, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - cinfo->first_block = GET2(buff, 0); - if (cinfo->first_block == 0) { /* data spans to next segment */ - cinfo->count = seg_sz - sizeof(__u16); - cinfo->offset = seg_sz; - cinfo->spans = 1; - } else { /* cluster definetely ends in this segment */ - if (cinfo->first_block > seg_sz) { - /* data corrupted */ - TRACE_ABORT(-EIO, ft_t_err, "corrupted data:\n" - KERN_INFO "segment size: %d\n" - KERN_INFO "first block : %d", - seg_sz, cinfo->first_block); - } - cinfo->count = cinfo->first_block - sizeof(__u16); - cinfo->offset = cinfo->first_block; - cinfo->spans = 0; - } - /* now get the offset the first block should have in the - * uncompressed data stream. - * - * For this magic `18' refer to CRF-3 standard or QIC-80MC, - * Rev. K. - */ - if ((seg_sz - cinfo->offset) > 18) { - if (volume->qic113) { /* > revision K */ - TRACE(ft_t_data_flow, "New QIC-113 compliance"); - cinfo->foffs = GET8(buff, cinfo->offset); - cinfo->offset += sizeof(__s64); - } else { - TRACE(/* ft_t_data_flow */ ft_t_noise, "pre QIC-113 version"); - cinfo->foffs = (__s64)GET4(buff, cinfo->offset); - cinfo->offset += sizeof(__u32); - } - } - if (cinfo->foffs > volume->size) { - TRACE_ABORT(-EIO, ft_t_err, "Inconsistency:\n" - KERN_INFO "offset in current volume: %d\n" - KERN_INFO "size of current volume : %d", - (int)(cinfo->foffs>>10), (int)(volume->size>>10)); - } - if (cinfo->cmpr_pos + cinfo->count > volume->blk_sz) { - TRACE_ABORT(-EIO, ft_t_err, "Inconsistency:\n" - KERN_INFO "block size : %d\n" - KERN_INFO "data record: %d", - volume->blk_sz, cinfo->cmpr_pos + cinfo->count); - } - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */, "", cinfo); - TRACE_EXIT 0; -} - -/* This one is called, when a new cluster starts in same segment. - * - * Note: if this is the first cluster in the current segment, we must - * not check whether there are more than 18 bytes available because - * this have already been done in get_cseg() and there may be less - * than 18 bytes available due to header information. - * - */ -static void get_next_cluster(cmpr_info *cluster, const __u8 *buff, - const int seg_sz, const int finish) -{ - TRACE_FUN(ft_t_flow); - - if (seg_sz - cluster->offset > 18 || cluster->foffs != 0) { - cluster->count = GET2(buff, cluster->offset); - cluster->uncmpr = cluster->count & 0x8000; - cluster->count -= cluster->uncmpr; - cluster->offset += sizeof(__u16); - cluster->foffs = 0; - if ((cluster->offset + cluster->count) < seg_sz) { - cluster->spans = 0; - } else if (cluster->offset + cluster->count == seg_sz) { - cluster->spans = !finish; - } else { - /* either an error or a volume written by an - * old version. If this is a data error, then we'll - * catch it later. - */ - TRACE(ft_t_data_flow, "Either error or old volume"); - cluster->spans = 1; - cluster->count = seg_sz - cluster->offset; - } - } else { - cluster->count = 0; - cluster->spans = 0; - cluster->foffs = 0; - } - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */ , "", cluster); - TRACE_EXIT; -} - -static void zftc_lock(void) -{ -} - -/* this function is needed for zftape_reset_position in zftape-io.c - */ -static void zftc_reset(void) -{ - TRACE_FUN(ft_t_flow); - - memset((void *)&cseg, '\0', sizeof(cseg)); - zftc_stats(); - TRACE_EXIT; -} - -static int cmpr_mem_initialized = 0; -static unsigned int alloc_blksz = 0; - -static int zft_allocate_cmpr_mem(unsigned int blksz) -{ - TRACE_FUN(ft_t_flow); - - if (cmpr_mem_initialized && blksz == alloc_blksz) { - TRACE_EXIT 0; - } - TRACE_CATCH(zft_vmalloc_once(&zftc_wrk_mem, CMPR_WRK_MEM_SIZE), - zftc_cleanup()); - TRACE_CATCH(zft_vmalloc_always(&zftc_buf, blksz + CMPR_OVERRUN), - zftc_cleanup()); - alloc_blksz = blksz; - TRACE_CATCH(zft_vmalloc_always(&zftc_scratch_buf, blksz+CMPR_OVERRUN), - zftc_cleanup()); - cmpr_mem_initialized = 1; - TRACE_EXIT 0; -} - -static void zftc_cleanup(void) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(&zftc_wrk_mem, CMPR_WRK_MEM_SIZE); - zft_vfree(&zftc_buf, alloc_blksz + CMPR_OVERRUN); - zft_vfree(&zftc_scratch_buf, alloc_blksz + CMPR_OVERRUN); - cmpr_mem_initialized = alloc_blksz = 0; - TRACE_EXIT; -} - -/***************************************************************************** - * * - * The following two functions "ftape_compress()" and * - * "ftape_uncompress()" are the interface to the actual compression * - * algorithm (i.e. they are calling the "compress()" function from * - * the lzrw3 package for now). These routines could quite easily be * - * changed to adopt another compression algorithm instead of lzrw3, * - * which currently is used. * - * * - *****************************************************************************/ - -/* called by zft_compress_write() to perform the compression. Must - * return the size of the compressed data. - * - * NOTE: The size of the compressed data should not exceed the size of - * the uncompressed data. Most compression algorithms have means - * to store data unchanged if the "compressed" data amount would - * exceed the original one. Mostly this is done by storing some - * flag-bytes in front of the compressed data to indicate if it - * is compressed or not. Thus the worst compression result - * length is the original length plus those flag-bytes. - * - * We don't want that, as the QIC-80 standard provides a means - * of marking uncompressed blocks by simply setting bit 15 of - * the compressed block's length. Thus a compessed block can - * have at most a length of 2^15-1 bytes. The QIC-80 standard - * restricts the block-length even further, allowing only 29k - - * 6 bytes. - * - * Currently, the maximum blocksize used by zftape is 28k. - * - * In short: don't exceed the length of the input-package, set - * bit 15 of the compressed size to 1 if you have copied data - * instead of compressing it. - */ -static int zft_compress(__u8 *in_buffer, unsigned int in_sz, __u8 *out_buffer) -{ - __s32 compressed_sz; - TRACE_FUN(ft_t_flow); - - - lzrw3_compress(COMPRESS_ACTION_COMPRESS, zftc_wrk_mem, - in_buffer, in_sz, out_buffer, &compressed_sz); - if (TRACE_LEVEL >= ft_t_info) { - /* the compiler will optimize this away when - * compiled with NO_TRACE_AT_ALL option - */ - TRACE(ft_t_data_flow, "\n" - KERN_INFO "before compression: %d bytes\n" - KERN_INFO "after compresison : %d bytes", - in_sz, - (int)(compressed_sz < 0 - ? -compressed_sz : compressed_sz)); - /* for statistical purposes - */ - zftc_wr_compressed += (compressed_sz < 0 - ? -compressed_sz : compressed_sz); - zftc_wr_uncompressed += in_sz; - } - TRACE_EXIT (int)compressed_sz; -} - -/* called by zft_compress_read() to decompress the data. Must - * return the size of the decompressed data for sanity checks - * (compared with zft_blk_sz) - * - * NOTE: Read the note for zft_compress() above! If bit 15 of the - * parameter in_sz is set, then the data in in_buffer isn't - * compressed, which must be handled by the un-compression - * algorithm. (I changed lzrw3 to handle this.) - * - * The parameter max_out_sz is needed to prevent buffer overruns when - * uncompressing corrupt data. - */ -static unsigned int zft_uncompress(__u8 *in_buffer, - int in_sz, - __u8 *out_buffer, - unsigned int max_out_sz) -{ - TRACE_FUN(ft_t_flow); - - lzrw3_compress(COMPRESS_ACTION_DECOMPRESS, zftc_wrk_mem, - in_buffer, (__s32)in_sz, - out_buffer, (__u32 *)&max_out_sz); - - if (TRACE_LEVEL >= ft_t_info) { - TRACE(ft_t_data_flow, "\n" - KERN_INFO "before decompression: %d bytes\n" - KERN_INFO "after decompression : %d bytes", - in_sz < 0 ? -in_sz : in_sz,(int)max_out_sz); - /* for statistical purposes - */ - zftc_rd_compressed += in_sz < 0 ? -in_sz : in_sz; - zftc_rd_uncompressed += max_out_sz; - } - TRACE_EXIT (unsigned int)max_out_sz; -} - -/* print some statistics about the efficiency of the compression to - * the kernel log - */ -static void zftc_stats(void) -{ - TRACE_FUN(ft_t_flow); - - if (TRACE_LEVEL < ft_t_info) { - TRACE_EXIT; - } - if (zftc_wr_uncompressed != 0) { - if (zftc_wr_compressed > (1<<14)) { - TRACE(ft_t_info, "compression statistics (writing):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - (((zftc_wr_compressed>>10) * 100) - / (zftc_wr_uncompressed>>10))); - } else { - TRACE(ft_t_info, "compression statistics (writing):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - ((zftc_wr_compressed * 100) - / zftc_wr_uncompressed)); - } - } - if (zftc_rd_uncompressed != 0) { - if (zftc_rd_compressed > (1<<14)) { - TRACE(ft_t_info, "compression statistics (reading):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - (((zftc_rd_compressed>>10) * 100) - / (zftc_rd_uncompressed>>10))); - } else { - TRACE(ft_t_info, "compression statistics (reading):\n" - KERN_INFO " compr./uncmpr. : %3d %%", - ((zftc_rd_compressed * 100) - / zftc_rd_uncompressed)); - } - } - /* only print it once: */ - zftc_wr_uncompressed = - zftc_wr_compressed = - zftc_rd_uncompressed = - zftc_rd_compressed = 0; - TRACE_EXIT; -} - -/* start new compressed block - */ -static int start_new_cseg(cmpr_info *cluster, - char *dst_buf, - const zft_position *pos, - const unsigned int blk_sz, - const char *src_buf, - const int this_segs_sz, - const int qic113) -{ - int size_left; - int cp_cnt; - int buf_pos; - TRACE_FUN(ft_t_flow); - - size_left = this_segs_sz - sizeof(__u16) - cluster->cmpr_sz; - TRACE(ft_t_data_flow,"\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "compressed_sz: %d\n" - KERN_INFO "size_left : %d", - this_segs_sz, cluster->cmpr_sz, size_left); - if (size_left > 18) { /* start a new cluseter */ - cp_cnt = cluster->cmpr_sz; - cluster->cmpr_sz = 0; - buf_pos = cp_cnt + sizeof(__u16); - PUT2(dst_buf, 0, buf_pos); - - if (qic113) { - __s64 foffs = pos->volume_pos; - if (cp_cnt) foffs += (__s64)blk_sz; - - TRACE(ft_t_data_flow, "new style QIC-113 header"); - PUT8(dst_buf, buf_pos, foffs); - buf_pos += sizeof(__s64); - } else { - __u32 foffs = (__u32)pos->volume_pos; - if (cp_cnt) foffs += (__u32)blk_sz; - - TRACE(ft_t_data_flow, "old style QIC-80MC header"); - PUT4(dst_buf, buf_pos, foffs); - buf_pos += sizeof(__u32); - } - } else if (size_left >= 0) { - cp_cnt = cluster->cmpr_sz; - cluster->cmpr_sz = 0; - buf_pos = cp_cnt + sizeof(__u16); - PUT2(dst_buf, 0, buf_pos); - /* zero unused part of segment. */ - memset(dst_buf + buf_pos, '\0', size_left); - buf_pos = this_segs_sz; - } else { /* need entire segment and more space */ - PUT2(dst_buf, 0, 0); - cp_cnt = this_segs_sz - sizeof(__u16); - cluster->cmpr_sz -= cp_cnt; - buf_pos = this_segs_sz; - } - memcpy(dst_buf + sizeof(__u16), src_buf + cluster->cmpr_pos, cp_cnt); - cluster->cmpr_pos += cp_cnt; - TRACE_EXIT buf_pos; -} - -/* return-value: the number of bytes removed from the user-buffer - * `src_buf' or error code - * - * int *write_cnt : how much actually has been moved to the - * dst_buf. Need not be initialized when - * function returns with an error code - * (negativ return value) - * __u8 *dst_buf : kernel space buffer where the has to be - * copied to. The contents of this buffers - * goes to a specific segment. - * const int seg_sz : the size of the segment dst_buf will be - * copied to. - * const zft_position *pos : struct containing the coordinates in - * the current volume (byte position, - * segment id of current segment etc) - * const zft_volinfo *volume: information about the current volume, - * size etc. - * const __u8 *src_buf : user space buffer that contains the - * data the user wants to be written to - * tape. - * const int req_len : the amount of data the user wants to be - * written to tape. - */ -static int zftc_write(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume) -{ - int req_len_left = req_len; - int result; - int len_left; - int buf_pos_write = pos->seg_byte_pos; - TRACE_FUN(ft_t_flow); - - /* Note: we do not unlock the module because - * there are some values cached in that `cseg' variable. We - * don't don't want to use this information when being - * unloaded by kerneld even when the tape is full or when we - * cannot allocate enough memory. - */ - if (pos->tape_pos > (volume->size-volume->blk_sz-ZFT_CMPR_OVERHEAD)) { - TRACE_EXIT -ENOSPC; - } - if (zft_allocate_cmpr_mem(volume->blk_sz) < 0) { - /* should we unlock the module? But it shouldn't - * be locked anyway ... - */ - TRACE_EXIT -ENOMEM; - } - if (buf_pos_write == 0) { /* fill a new segment */ - *write_cnt = buf_pos_write = start_new_cseg(&cseg, - dst_buf, - pos, - volume->blk_sz, - zftc_buf, - seg_sz, - volume->qic113); - if (cseg.cmpr_sz == 0 && cseg.cmpr_pos != 0) { - req_len_left -= result = volume->blk_sz; - cseg.cmpr_pos = 0; - } else { - result = 0; - } - } else { - *write_cnt = result = 0; - } - - len_left = seg_sz - buf_pos_write; - while ((req_len_left > 0) && (len_left > 18)) { - /* now we have some size left for a new compressed - * block. We know, that the compression buffer is - * empty (else there wouldn't be any space left). - */ - if (copy_from_user(zftc_scratch_buf, src_buf + result, - volume->blk_sz) != 0) { - TRACE_EXIT -EFAULT; - } - req_len_left -= volume->blk_sz; - cseg.cmpr_sz = zft_compress(zftc_scratch_buf, volume->blk_sz, - zftc_buf); - if (cseg.cmpr_sz < 0) { - cseg.uncmpr = 0x8000; - cseg.cmpr_sz = -cseg.cmpr_sz; - } else { - cseg.uncmpr = 0; - } - /* increment "result" iff we copied the entire - * compressed block to the zft_deblock_buf - */ - len_left -= sizeof(__u16); - if (len_left >= cseg.cmpr_sz) { - len_left -= cseg.count = cseg.cmpr_sz; - cseg.cmpr_pos = cseg.cmpr_sz = 0; - result += volume->blk_sz; - } else { - cseg.cmpr_sz -= - cseg.cmpr_pos = - cseg.count = len_left; - len_left = 0; - } - PUT2(dst_buf, buf_pos_write, cseg.uncmpr | cseg.count); - buf_pos_write += sizeof(__u16); - memcpy(dst_buf + buf_pos_write, zftc_buf, cseg.count); - buf_pos_write += cseg.count; - *write_cnt += cseg.count + sizeof(__u16); - FT_SIGNAL_EXIT(_DONT_BLOCK); - } - /* erase the remainder of the segment if less than 18 bytes - * left (18 bytes is due to the QIC-80 standard) - */ - if (len_left <= 18) { - memset(dst_buf + buf_pos_write, '\0', len_left); - (*write_cnt) += len_left; - } - TRACE(ft_t_data_flow, "returning %d", result); - TRACE_EXIT result; -} - -/* out: - * - * int *read_cnt: the number of bytes we removed from the zft_deblock_buf - * (result) - * int *to_do : the remaining size of the read-request. - * - * in: - * - * char *buff : buff is the address of the upper part of the user - * buffer, that hasn't been filled with data yet. - - * int buf_pos_read : copy of from _ftape_read() - * int buf_len_read : copy of buf_len_rd from _ftape_read() - * char *zft_deblock_buf: zft_deblock_buf - * unsigned short blk_sz: the block size valid for this volume, may differ - * from zft_blk_sz. - * int finish: if != 0 means that this is the last segment belonging - * to this volume - * returns the amount of data actually copied to the user-buffer - * - * to_do MUST NOT SHRINK except to indicate an EOF. In this case *to_do has to - * be set to 0 - */ -static int zftc_read (int *read_cnt, - __u8 __user *dst_buf, const int to_do, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume) -{ - int uncompressed_sz; - int result = 0; - int remaining = to_do; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(zft_allocate_cmpr_mem(volume->blk_sz),); - if (pos->seg_byte_pos == 0) { - /* new segment just read - */ - TRACE_CATCH(get_cseg(&cseg, src_buf, seg_sz, volume), - *read_cnt = 0); - memcpy(zftc_buf + cseg.cmpr_pos, src_buf + sizeof(__u16), - cseg.count); - cseg.cmpr_pos += cseg.count; - *read_cnt = cseg.offset; - DUMP_CMPR_INFO(ft_t_noise /* ft_t_any */, "", &cseg); - } else { - *read_cnt = 0; - } - /* loop and uncompress until user buffer full or - * deblock-buffer empty - */ - TRACE(ft_t_data_flow, "compressed_sz: %d, compos : %d, *read_cnt: %d", - cseg.cmpr_sz, cseg.cmpr_pos, *read_cnt); - while ((cseg.spans == 0) && (remaining > 0)) { - if (cseg.cmpr_pos != 0) { /* cmpr buf is not empty */ - uncompressed_sz = - zft_uncompress(zftc_buf, - cseg.uncmpr == 0x8000 ? - -cseg.cmpr_pos : cseg.cmpr_pos, - zftc_scratch_buf, - volume->blk_sz); - if (uncompressed_sz != volume->blk_sz) { - *read_cnt = 0; - TRACE_ABORT(-EIO, ft_t_warn, - "Uncompressed blk (%d) != blk size (%d)", - uncompressed_sz, volume->blk_sz); - } - if (copy_to_user(dst_buf + result, - zftc_scratch_buf, - uncompressed_sz) != 0 ) { - TRACE_EXIT -EFAULT; - } - remaining -= uncompressed_sz; - result += uncompressed_sz; - cseg.cmpr_pos = 0; - } - if (remaining > 0) { - get_next_cluster(&cseg, src_buf, seg_sz, - volume->end_seg == pos->seg_pos); - if (cseg.count != 0) { - memcpy(zftc_buf, src_buf + cseg.offset, - cseg.count); - cseg.cmpr_pos = cseg.count; - cseg.offset += cseg.count; - *read_cnt += cseg.count + sizeof(__u16); - } else { - remaining = 0; - } - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "compressed_sz: %d\n" - KERN_INFO "compos : %d\n" - KERN_INFO "*read_cnt : %d", - cseg.cmpr_sz, cseg.cmpr_pos, *read_cnt); - } - if (seg_sz - cseg.offset <= 18) { - *read_cnt += seg_sz - cseg.offset; - TRACE(ft_t_data_flow, "expanding read cnt to: %d", *read_cnt); - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "read count : %d\n" - KERN_INFO "buf_pos_read : %d\n" - KERN_INFO "remaining : %d", - seg_sz, *read_cnt, pos->seg_byte_pos, - seg_sz - *read_cnt - pos->seg_byte_pos); - TRACE(ft_t_data_flow, "returning: %d", result); - TRACE_EXIT result; -} - -/* seeks to the new data-position. Reads sometimes a segment. - * - * start_seg and end_seg give the boundaries of the current volume - * blk_sz is the blk_sz of the current volume as stored in the - * volume label - * - * We don't allow blocksizes less than 1024 bytes, therefore we don't need - * a 64 bit argument for new_block_pos. - */ - -static int seek_in_segment(const unsigned int to_do, cmpr_info *c_info, - const char *src_buf, const int seg_sz, - const int seg_pos, const zft_volinfo *volume); -static int slow_seek_forward_until_error(const unsigned int distance, - cmpr_info *c_info, zft_position *pos, - const zft_volinfo *volume, __u8 *buf); -static int search_valid_segment(unsigned int segment, - const unsigned int end_seg, - const unsigned int max_foffs, - zft_position *pos, cmpr_info *c_info, - const zft_volinfo *volume, __u8 *buf); -static int slow_seek_forward(unsigned int dest, cmpr_info *c_info, - zft_position *pos, const zft_volinfo *volume, - __u8 *buf); -static int compute_seg_pos(unsigned int dest, zft_position *pos, - const zft_volinfo *volume); - -#define ZFT_SLOW_SEEK_THRESHOLD 10 /* segments */ -#define ZFT_FAST_SEEK_MAX_TRIALS 10 /* times */ -#define ZFT_FAST_SEEK_BACKUP 10 /* segments */ - -static int zftc_seek(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, __u8 *buf) -{ - unsigned int dest; - int limit; - int distance; - int result = 0; - int seg_dist; - int new_seg; - int old_seg = 0; - int fast_seek_trials = 0; - TRACE_FUN(ft_t_flow); - - if (new_block_pos == 0) { - pos->seg_pos = volume->start_seg; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - zftc_reset(); - TRACE_EXIT 0; - } - dest = new_block_pos * (volume->blk_sz >> 10); - distance = dest - (pos->volume_pos >> 10); - while (distance != 0) { - seg_dist = compute_seg_pos(dest, pos, volume); - TRACE(ft_t_noise, "\n" - KERN_INFO "seg_dist: %d\n" - KERN_INFO "distance: %d\n" - KERN_INFO "dest : %d\n" - KERN_INFO "vpos : %d\n" - KERN_INFO "seg_pos : %d\n" - KERN_INFO "trials : %d", - seg_dist, distance, dest, - (unsigned int)(pos->volume_pos>>10), pos->seg_pos, - fast_seek_trials); - if (distance > 0) { - if (seg_dist < 0) { - TRACE(ft_t_bug, "BUG: distance %d > 0, " - "segment difference %d < 0", - distance, seg_dist); - result = -EIO; - break; - } - new_seg = pos->seg_pos + seg_dist; - if (new_seg > volume->end_seg) { - new_seg = volume->end_seg; - } - if (old_seg == new_seg || /* loop */ - seg_dist <= ZFT_SLOW_SEEK_THRESHOLD || - fast_seek_trials >= ZFT_FAST_SEEK_MAX_TRIALS) { - TRACE(ft_t_noise, "starting slow seek:\n" - KERN_INFO "fast seek failed too often: %s\n" - KERN_INFO "near target position : %s\n" - KERN_INFO "looping between two segs : %s", - (fast_seek_trials >= - ZFT_FAST_SEEK_MAX_TRIALS) - ? "yes" : "no", - (seg_dist <= ZFT_SLOW_SEEK_THRESHOLD) - ? "yes" : "no", - (old_seg == new_seg) - ? "yes" : "no"); - result = slow_seek_forward(dest, &cseg, - pos, volume, buf); - break; - } - old_seg = new_seg; - limit = volume->end_seg; - fast_seek_trials ++; - for (;;) { - result = search_valid_segment(new_seg, limit, - volume->size, - pos, &cseg, - volume, buf); - if (result == 0 || result == -EINTR) { - break; - } - if (new_seg == volume->start_seg) { - result = -EIO; /* set errror - * condition - */ - break; - } - limit = new_seg; - new_seg -= ZFT_FAST_SEEK_BACKUP; - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - } - if (result < 0) { - TRACE(ft_t_warn, - "Couldn't find a readable segment"); - break; - } - } else /* if (distance < 0) */ { - if (seg_dist > 0) { - TRACE(ft_t_bug, "BUG: distance %d < 0, " - "segment difference %d >0", - distance, seg_dist); - result = -EIO; - break; - } - new_seg = pos->seg_pos + seg_dist; - if (fast_seek_trials > 0 && seg_dist == 0) { - /* this avoids sticking to the same - * segment all the time. On the other hand: - * if we got here for the first time, and the - * deblock_buffer still contains a valid - * segment, then there is no need to skip to - * the previous segment if the desired position - * is inside this segment. - */ - new_seg --; - } - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - limit = pos->seg_pos; - fast_seek_trials ++; - for (;;) { - result = search_valid_segment(new_seg, limit, - pos->volume_pos, - pos, &cseg, - volume, buf); - if (result == 0 || result == -EINTR) { - break; - } - if (new_seg == volume->start_seg) { - result = -EIO; /* set errror - * condition - */ - break; - } - limit = new_seg; - new_seg -= ZFT_FAST_SEEK_BACKUP; - if (new_seg < volume->start_seg) { - new_seg = volume->start_seg; - } - } - if (result < 0) { - TRACE(ft_t_warn, - "Couldn't find a readable segment"); - break; - } - } - distance = dest - (pos->volume_pos >> 10); - } - TRACE_EXIT result; -} - - -/* advance inside the given segment at most to_do bytes. - * of kilobytes moved - */ - -static int seek_in_segment(const unsigned int to_do, - cmpr_info *c_info, - const char *src_buf, - const int seg_sz, - const int seg_pos, - const zft_volinfo *volume) -{ - int result = 0; - int blk_sz = volume->blk_sz >> 10; - int remaining = to_do; - TRACE_FUN(ft_t_flow); - - if (c_info->offset == 0) { - /* new segment just read - */ - TRACE_CATCH(get_cseg(c_info, src_buf, seg_sz, volume),); - c_info->cmpr_pos += c_info->count; - DUMP_CMPR_INFO(ft_t_noise, "", c_info); - } - /* loop and uncompress until user buffer full or - * deblock-buffer empty - */ - TRACE(ft_t_noise, "compressed_sz: %d, compos : %d", - c_info->cmpr_sz, c_info->cmpr_pos); - while (c_info->spans == 0 && remaining > 0) { - if (c_info->cmpr_pos != 0) { /* cmpr buf is not empty */ - result += blk_sz; - remaining -= blk_sz; - c_info->cmpr_pos = 0; - } - if (remaining > 0) { - get_next_cluster(c_info, src_buf, seg_sz, - volume->end_seg == seg_pos); - if (c_info->count != 0) { - c_info->cmpr_pos = c_info->count; - c_info->offset += c_info->count; - } else { - break; - } - } - /* Allow escape from this loop on signal! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - DUMP_CMPR_INFO(ft_t_noise, "", c_info); - TRACE(ft_t_noise, "to_do: %d", remaining); - } - if (seg_sz - c_info->offset <= 18) { - c_info->offset = seg_sz; - } - TRACE(ft_t_noise, "\n" - KERN_INFO "segment size : %d\n" - KERN_INFO "buf_pos_read : %d\n" - KERN_INFO "remaining : %d", - seg_sz, c_info->offset, - seg_sz - c_info->offset); - TRACE_EXIT result; -} - -static int slow_seek_forward_until_error(const unsigned int distance, - cmpr_info *c_info, - zft_position *pos, - const zft_volinfo *volume, - __u8 *buf) -{ - unsigned int remaining = distance; - int seg_sz; - int seg_pos; - int result; - TRACE_FUN(ft_t_flow); - - seg_pos = pos->seg_pos; - do { - TRACE_CATCH(seg_sz = zft_fetch_segment(seg_pos, buf, - FT_RD_AHEAD),); - /* now we have the contents of the actual segment in - * the deblock buffer - */ - TRACE_CATCH(result = seek_in_segment(remaining, c_info, buf, - seg_sz, seg_pos,volume),); - remaining -= result; - pos->volume_pos += result<<10; - pos->seg_pos = seg_pos; - pos->seg_byte_pos = c_info->offset; - seg_pos ++; - if (seg_pos <= volume->end_seg && c_info->offset == seg_sz) { - pos->seg_pos ++; - pos->seg_byte_pos = 0; - c_info->offset = 0; - } - /* Allow escape from this loop on signal! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_noise, "\n" - KERN_INFO "remaining: %d\n" - KERN_INFO "seg_pos: %d\n" - KERN_INFO "end_seg: %d\n" - KERN_INFO "result: %d", - remaining, seg_pos, volume->end_seg, result); - } while (remaining > 0 && seg_pos <= volume->end_seg); - TRACE_EXIT 0; -} - -/* return segment id of next segment containing valid data, -EIO otherwise - */ -static int search_valid_segment(unsigned int segment, - const unsigned int end_seg, - const unsigned int max_foffs, - zft_position *pos, - cmpr_info *c_info, - const zft_volinfo *volume, - __u8 *buf) -{ - cmpr_info tmp_info; - int seg_sz; - TRACE_FUN(ft_t_flow); - - memset(&tmp_info, 0, sizeof(cmpr_info)); - while (segment <= end_seg) { - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_noise, - "Searching readable segment between %d and %d", - segment, end_seg); - seg_sz = zft_fetch_segment(segment, buf, FT_RD_AHEAD); - if ((seg_sz > 0) && - (get_cseg (&tmp_info, buf, seg_sz, volume) >= 0) && - (tmp_info.foffs != 0 || segment == volume->start_seg)) { - if ((tmp_info.foffs>>10) > max_foffs) { - TRACE_ABORT(-EIO, ft_t_noise, "\n" - KERN_INFO "cseg.foff: %d\n" - KERN_INFO "dest : %d", - (int)(tmp_info.foffs >> 10), - max_foffs); - } - DUMP_CMPR_INFO(ft_t_noise, "", &tmp_info); - *c_info = tmp_info; - pos->seg_pos = segment; - pos->volume_pos = c_info->foffs; - pos->seg_byte_pos = c_info->offset; - TRACE(ft_t_noise, "found segment at %d", segment); - TRACE_EXIT 0; - } - segment++; - } - TRACE_EXIT -EIO; -} - -static int slow_seek_forward(unsigned int dest, - cmpr_info *c_info, - zft_position *pos, - const zft_volinfo *volume, - __u8 *buf) -{ - unsigned int distance; - int result = 0; - TRACE_FUN(ft_t_flow); - - distance = dest - (pos->volume_pos >> 10); - while ((distance > 0) && - (result = slow_seek_forward_until_error(distance, - c_info, - pos, - volume, - buf)) < 0) { - if (result == -EINTR) { - break; - } - TRACE(ft_t_noise, "seg_pos: %d", pos->seg_pos); - /* the failing segment is either pos->seg_pos or - * pos->seg_pos + 1. There is no need to further try - * that segment, because ftape_read_segment() already - * has tried very much to read it. So we start with - * following segment, which is pos->seg_pos + 1 - */ - if(search_valid_segment(pos->seg_pos+1, volume->end_seg, dest, - pos, c_info, - volume, buf) < 0) { - TRACE(ft_t_noise, "search_valid_segment() failed"); - result = -EIO; - break; - } - distance = dest - (pos->volume_pos >> 10); - result = 0; - TRACE(ft_t_noise, "segment: %d", pos->seg_pos); - /* found valid segment, retry the seek */ - } - TRACE_EXIT result; -} - -static int compute_seg_pos(const unsigned int dest, - zft_position *pos, - const zft_volinfo *volume) -{ - int segment; - int distance = dest - (pos->volume_pos >> 10); - unsigned int raw_size; - unsigned int virt_size; - unsigned int factor; - TRACE_FUN(ft_t_flow); - - if (distance >= 0) { - raw_size = volume->end_seg - pos->seg_pos + 1; - virt_size = ((unsigned int)(volume->size>>10) - - (unsigned int)(pos->volume_pos>>10) - + FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS - 1); - virt_size /= FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS; - if (virt_size == 0 || raw_size == 0) { - TRACE_EXIT 0; - } - if (raw_size >= (1<<25)) { - factor = raw_size/(virt_size>>7); - } else { - factor = (raw_size<<7)/virt_size; - } - segment = distance/(FT_SECTORS_PER_SEGMENT-FT_ECC_SECTORS); - segment = (segment * factor)>>7; - } else { - raw_size = pos->seg_pos - volume->start_seg + 1; - virt_size = ((unsigned int)(pos->volume_pos>>10) - + FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS - 1); - virt_size /= FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS; - if (virt_size == 0 || raw_size == 0) { - TRACE_EXIT 0; - } - if (raw_size >= (1<<25)) { - factor = raw_size/(virt_size>>7); - } else { - factor = (raw_size<<7)/virt_size; - } - segment = distance/(FT_SECTORS_PER_SEGMENT-FT_ECC_SECTORS); - } - TRACE(ft_t_noise, "factor: %d/%d", factor, 1<<7); - TRACE_EXIT segment; -} - -static struct zft_cmpr_ops cmpr_ops = { - zftc_write, - zftc_read, - zftc_seek, - zftc_lock, - zftc_reset, - zftc_cleanup -}; - -int zft_compressor_init(void) -{ - TRACE_FUN(ft_t_flow); - -#ifdef MODULE - printk(KERN_INFO "zftape compressor v1.00a 970514 for " FTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO "(c) 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO "Compressor for zftape (lzrw3 algorithm)\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO "zftape compressor v1.00a 970514\n"); - printk(KERN_INFO "For use with " FTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "zft_compressor_init @ 0x%p", zft_compressor_init); - TRACE(ft_t_info, "installing compressor for zftape ..."); - TRACE_CATCH(zft_cmpr_register(&cmpr_ops),); - TRACE_EXIT 0; -} - -#ifdef MODULE - -MODULE_AUTHOR( - "(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de"); -MODULE_DESCRIPTION( -"Compression routines for zftape. Uses the lzrw3 algorithm by Ross Williams"); -MODULE_LICENSE("GPL"); - -/* Called by modules package when installing the driver - */ -int init_module(void) -{ - return zft_compressor_init(); -} - -#endif /* MODULE */ diff --git a/drivers/char/ftape/compressor/zftape-compress.h b/drivers/char/ftape/compressor/zftape-compress.h deleted file mode 100644 index f200741e33bf..000000000000 --- a/drivers/char/ftape/compressor/zftape-compress.h +++ /dev/null @@ -1,83 +0,0 @@ -#ifndef _ZFTAPE_COMPRESS_H -#define _ZFTAPE_COMPRESS_H -/* - * Copyright (c) 1994-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/compressor/zftape-compress.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/10/05 19:12:32 $ - * - * This file contains macros and definitions for zftape's - * builtin compression code. - * - */ - -#include "../zftape/zftape-buffers.h" -#include "../zftape/zftape-vtbl.h" -#include "../compressor/lzrw3.h" - -/* CMPR_WRK_MEM_SIZE gives the size of the compression wrk_mem */ -/* I got these out of lzrw3.c */ -#define U(X) ((__u32) X) -#define SIZE_P_BYTE (U(sizeof(__u8 *))) -#define ALIGNMENT_FUDGE (U(16)) - -#define CMPR_WRK_MEM_SIZE (U(4096)*(SIZE_P_BYTE) + ALIGNMENT_FUDGE) - -/* the maximum number of bytes the size of the "compressed" data can - * exceed the uncompressed data. As it is quite useless to compress - * data twice it is sometimes the case that it is more efficient to - * copy a block of data but to feed it to the "compression" - * algorithm. In this case there are some flag bytes or the like - * proceding the "compressed" data. THAT MUST NOT BE THE CASE for the - * algorithm we use for this driver. Instead, the high bit 15 of - * compressed_size: - * - * compressed_size = ftape_compress() - * - * must be set in such a case. - * - * Nevertheless, it might also be as for lzrw3 that there is an - * "intermediate" overrun that exceeds the amount of the compressed - * data that is actually produced. During the algorithm we need in the - * worst case MAX_CMP_GROUP bytes more than the input-size. - */ -#define MAX_CMP_GROUP (2+16*2) /* from lzrw3.c */ - -#define CMPR_OVERRUN MAX_CMP_GROUP /* during compression */ - -/****************************************************/ - -#define CMPR_BUFFER_SIZE (MAX_BLOCK_SIZE + CMPR_OVERRUN) - -/* the compression map stores the byte offset compressed blocks within - * the current volume for catridges with format code 2,3 and 5 - * (and old versions of zftape) and the offset measured in kilobytes for - * format code 4 and 6. This gives us a possible max. size of a - * compressed volume of 1024*4GIG which should be enough. - */ -typedef __u32 CmprMap; - -/* globals - */ - -/* exported functions - */ - -#endif /* _ZFTAPE_COMPRESS_H */ diff --git a/drivers/char/ftape/lowlevel/Makefile b/drivers/char/ftape/lowlevel/Makefile deleted file mode 100644 index febab07ba427..000000000000 --- a/drivers/char/ftape/lowlevel/Makefile +++ /dev/null @@ -1,43 +0,0 @@ -# -# Copyright (C) 1996, 1997 Clau-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/07 09:26:02 $ -# -# Makefile for the lowlevel part QIC-40/80/3010/3020 floppy-tape -# driver for Linux. -# - -obj-$(CONFIG_FTAPE) += ftape.o - -ftape-objs := ftape-init.o fdc-io.o fdc-isr.o \ - ftape-bsm.o ftape-ctl.o ftape-read.o ftape-rw.o \ - ftape-write.o ftape-io.o ftape-calibr.o ftape-ecc.o fc-10.o \ - ftape-buffer.o ftape-format.o ftape_syms.o - -ifeq ($(CONFIG_FTAPE),y) -ftape-objs += ftape-setup.o -endif - -ifndef CONFIG_FT_NO_TRACE_AT_ALL -ftape-objs += ftape-tracing.o -endif - -ifeq ($(CONFIG_FT_PROC_FS),y) -ftape-objs += ftape-proc.o -endif diff --git a/drivers/char/ftape/lowlevel/fc-10.c b/drivers/char/ftape/lowlevel/fc-10.c deleted file mode 100644 index 9bc1cddade76..000000000000 --- a/drivers/char/ftape/lowlevel/fc-10.c +++ /dev/null @@ -1,175 +0,0 @@ -/* - * - - Copyright (C) 1993,1994 Jon Tombs. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - The entire guts of this program was written by dosemu, modified to - record reads and writes to the ports in the 0x180-0x188 address space, - while running the CMS program TAPE.EXE V2.0.5 supplied with the drive. - - Modified to use an array of addresses and generally cleaned up (made - much shorter) 4 June 94, dosemu isn't that good at writing short code it - would seem :-). Made independent of 0x180, but I doubt it will work - at any other address. - - Modified for distribution with ftape source. 21 June 94, SJL. - - Modifications on 20 October 95, by Daniel Cohen (catman@wpi.edu): - Modified to support different DMA, IRQ, and IO Ports. Borland's - Turbo Debugger in virtual 8086 mode (TD386.EXE with hardware breakpoints - provided by the TDH386.SYS Device Driver) was used on the CMS program - TAPE V4.0.5. I set breakpoints on I/O to ports 0x180-0x187. Note that - CMS's program will not successfully configure the tape drive if you set - breakpoints on IO Reads, but you can set them on IO Writes without problems. - Known problems: - - You can not use DMA Channels 5 or 7. - - Modification on 29 January 96, by Daniel Cohen (catman@wpi.edu): - Modified to only accept IRQs 3 - 7, or 9. Since we can only send a 3 bit - number representing the IRQ to the card, special handling is required when - IRQ 9 is selected. IRQ 2 and 9 are the same, and we should request IRQ 9 - from the kernel while telling the card to use IRQ 2. Thanks to Greg - Crider (gcrider@iclnet.org) for finding and locating this bug, as well as - testing the patch. - - Modification on 11 December 96, by Claus Heine (claus@momo.math.rwth-aachen.de): - Modified a little to use variahle ft_fdc_base, ft_fdc_irq, ft_fdc_dma - instead of preprocessor symbols. Thus we can compile this into the module - or kernel and let the user specify the options as command line arguments. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fc-10.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:04 $ - * - * This file contains code for the CMS FC-10/FC-20 card. - */ - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/fc-10.h" - -static __u16 inbs_magic[] = { - 0x3, 0x3, 0x0, 0x4, 0x7, 0x2, 0x5, 0x3, 0x1, 0x4, - 0x3, 0x5, 0x2, 0x0, 0x3, 0x7, 0x4, 0x2, - 0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7 -}; - -static __u16 fc10_ports[] = { - 0x180, 0x210, 0x2A0, 0x300, 0x330, 0x340, 0x370 -}; - -int fc10_enable(void) -{ - int i; - __u8 cardConfig = 0x00; - __u8 x; - TRACE_FUN(ft_t_flow); - -/* This code will only work if the FC-10 (or FC-20) is set to - * use DMA channels 1, 2, or 3. DMA channels 5 and 7 seem to be - * initialized by the same command as channels 1 and 3, respectively. - */ - if (ft_fdc_dma > 3) { - TRACE_ABORT(0, ft_t_err, -"Error: The FC-10/20 must be set to use DMA channels 1, 2, or 3!"); - } -/* Only allow the FC-10/20 to use IRQ 3-7, or 9. Note that CMS's program - * only accepts IRQ's 2-7, but in linux, IRQ 2 is the same as IRQ 9. - */ - if (ft_fdc_irq < 3 || ft_fdc_irq == 8 || ft_fdc_irq > 9) { - TRACE_ABORT(0, ft_t_err, -"Error: The FC-10/20 must be set to use IRQ levels 3 - 7, or 9!\n" -KERN_INFO "Note: IRQ 9 is the same as IRQ 2"); - } - /* Clear state machine ??? - */ - for (i = 0; i < NR_ITEMS(inbs_magic); i++) { - inb(ft_fdc_base + inbs_magic[i]); - } - outb(0x0, ft_fdc_base); - - x = inb(ft_fdc_base); - if (x == 0x13 || x == 0x93) { - for (i = 1; i < 8; i++) { - if (inb(ft_fdc_base + i) != x) { - TRACE_EXIT 0; - } - } - } else { - TRACE_EXIT 0; - } - - outb(0x8, ft_fdc_base); - - for (i = 0; i < 8; i++) { - if (inb(ft_fdc_base + i) != 0x0) { - TRACE_EXIT 0; - } - } - outb(0x10, ft_fdc_base); - - for (i = 0; i < 8; i++) { - if (inb(ft_fdc_base + i) != 0xff) { - TRACE_EXIT 0; - } - } - - /* Okay, we found a FC-10 card ! ??? - */ - outb(0x0, fdc.ccr); - - /* Clear state machine again ??? - */ - for (i = 0; i < NR_ITEMS(inbs_magic); i++) { - inb(ft_fdc_base + inbs_magic[i]); - } - /* Send io port */ - for (i = 0; i < NR_ITEMS(fc10_ports); i++) - if (ft_fdc_base == fc10_ports[i]) - cardConfig = i + 1; - if (cardConfig == 0) { - TRACE_EXIT 0; /* Invalid I/O Port */ - } - /* and IRQ - If using IRQ 9, tell the FC card it is actually IRQ 2 */ - if (ft_fdc_irq != 9) - cardConfig |= ft_fdc_irq << 3; - else - cardConfig |= 2 << 3; - - /* and finally DMA Channel */ - cardConfig |= ft_fdc_dma << 6; - outb(cardConfig, ft_fdc_base); /* DMA [2 bits]/IRQ [3 bits]/BASE [3 bits] */ - - /* Enable FC-10 ??? - */ - outb(0, fdc.ccr); - outb(0, fdc.dor2); - outb(FDC_DMA_MODE /* 8 */, fdc.dor); - outb(FDC_DMA_MODE /* 8 */, fdc.dor); - outb(1, fdc.dor2); - - /************************************* - * - * cH: why the hell should this be necessary? This is done - * by fdc_reset()!!! - * - *************************************/ - /* Initialize fdc, select drive B: - */ - outb(FDC_DMA_MODE, fdc.dor); /* assert reset, dma & irq enabled */ - /* 0x08 */ - outb(FDC_DMA_MODE|FDC_RESET_NOT, fdc.dor); /* release reset */ - /* 0x08 | 0x04 = 0x0c */ - outb(FDC_DMA_MODE|FDC_RESET_NOT|FDC_MOTOR_1|FTAPE_SEL_B, fdc.dor); - /* 0x08 | 0x04 | 0x20 | 0x01 = 0x2d */ - /* select drive 1 */ /* why not drive 0 ???? */ - TRACE_EXIT (x == 0x93) ? 2 : 1; -} diff --git a/drivers/char/ftape/lowlevel/fc-10.h b/drivers/char/ftape/lowlevel/fc-10.h deleted file mode 100644 index da7b88bca889..000000000000 --- a/drivers/char/ftape/lowlevel/fc-10.h +++ /dev/null @@ -1,39 +0,0 @@ -#ifndef _FC_10_H -#define _FC_10_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fc-10.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/09/19 09:05:22 $ - * - * This file contains definitions for the FC-10 code - * of the QIC-40/80 floppy-tape driver for Linux. - */ - -/* - * fc-10.c defined global vars. - */ - -/* - * fc-10.c defined global functions. - */ -extern int fc10_enable(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/fdc-io.c b/drivers/char/ftape/lowlevel/fdc-io.c deleted file mode 100644 index bbcf918f056f..000000000000 --- a/drivers/char/ftape/lowlevel/fdc-io.c +++ /dev/null @@ -1,1349 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-io.c,v $ - * $Revision: 1.7.4.2 $ - * $Date: 1997/11/16 14:48:17 $ - * - * This file contains the low-level floppy disk interface code - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/fdc-isr.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/fc-10.h" - -/* Global vars. - */ -static int ftape_motor; -volatile int ftape_current_cylinder = -1; -volatile fdc_mode_enum fdc_mode = fdc_idle; -fdc_config_info fdc; -DECLARE_WAIT_QUEUE_HEAD(ftape_wait_intr); - -unsigned int ft_fdc_base = CONFIG_FT_FDC_BASE; -unsigned int ft_fdc_irq = CONFIG_FT_FDC_IRQ; -unsigned int ft_fdc_dma = CONFIG_FT_FDC_DMA; -unsigned int ft_fdc_threshold = CONFIG_FT_FDC_THR; /* bytes */ -unsigned int ft_fdc_rate_limit = CONFIG_FT_FDC_MAX_RATE; /* bits/sec */ -int ft_probe_fc10 = CONFIG_FT_PROBE_FC10; -int ft_mach2 = CONFIG_FT_MACH2; - -/* Local vars. - */ -static spinlock_t fdc_io_lock; -static unsigned int fdc_calibr_count; -static unsigned int fdc_calibr_time; -static int fdc_status; -volatile __u8 fdc_head; /* FDC head from sector id */ -volatile __u8 fdc_cyl; /* FDC track from sector id */ -volatile __u8 fdc_sect; /* FDC sector from sector id */ -static int fdc_data_rate = 500; /* data rate (Kbps) */ -static int fdc_rate_code; /* data rate code (0 == 500 Kbps) */ -static int fdc_seek_rate = 2; /* step rate (msec) */ -static void (*do_ftape) (void); -static int fdc_fifo_state; /* original fifo setting - fifo enabled */ -static int fdc_fifo_thr; /* original fifo setting - threshold */ -static int fdc_lock_state; /* original lock setting - locked */ -static int fdc_fifo_locked; /* has fifo && lock set ? */ -static __u8 fdc_precomp; /* default precomp. value (nsec) */ -static __u8 fdc_prec_code; /* fdc precomp. select code */ - -static char ftape_id[] = "ftape"; /* used by request irq and free irq */ - -static int fdc_set_seek_rate(int seek_rate); - -void fdc_catch_stray_interrupts(int count) -{ - unsigned long flags; - - spin_lock_irqsave(&fdc_io_lock, flags); - if (count == 0) { - ft_expected_stray_interrupts = 0; - } else { - ft_expected_stray_interrupts += count; - } - spin_unlock_irqrestore(&fdc_io_lock, flags); -} - -/* Wait during a timeout period for a given FDC status. - * If usecs == 0 then just test status, else wait at least for usecs. - * Returns -ETIME on timeout. Function must be calibrated first ! - */ -static int fdc_wait(unsigned int usecs, __u8 mask, __u8 state) -{ - int count_1 = (fdc_calibr_count * usecs + - fdc_calibr_count - 1) / fdc_calibr_time; - - do { - fdc_status = inb_p(fdc.msr); - if ((fdc_status & mask) == state) { - return 0; - } - } while (count_1-- >= 0); - return -ETIME; -} - -int fdc_ready_wait(unsigned int usecs) -{ - return fdc_wait(usecs, FDC_DATA_READY | FDC_BUSY, FDC_DATA_READY); -} - -/* Why can't we just use udelay()? - */ -static void fdc_usec_wait(unsigned int usecs) -{ - fdc_wait(usecs, 0, 1); /* will always timeout ! */ -} - -static int fdc_ready_out_wait(unsigned int usecs) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - return fdc_wait(usecs, FDC_DATA_OUT_READY, FDC_DATA_OUT_READY); -} - -void fdc_wait_calibrate(void) -{ - ftape_calibrate("fdc_wait", - fdc_usec_wait, &fdc_calibr_count, &fdc_calibr_time); -} - -/* Wait for a (short) while for the FDC to become ready - * and transfer the next command byte. - * Return -ETIME on timeout on getting ready (depends on hardware!). - */ -static int fdc_write(const __u8 data) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - if (fdc_wait(150, FDC_DATA_READY_MASK, FDC_DATA_IN_READY) < 0) { - return -ETIME; - } else { - outb(data, fdc.fifo); - return 0; - } -} - -/* Wait for a (short) while for the FDC to become ready - * and transfer the next result byte. - * Return -ETIME if timeout on getting ready (depends on hardware!). - */ -static int fdc_read(__u8 * data) -{ - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - if (fdc_wait(150, FDC_DATA_READY_MASK, FDC_DATA_OUT_READY) < 0) { - return -ETIME; - } else { - *data = inb(fdc.fifo); - return 0; - } -} - -/* Output a cmd_len long command string to the FDC. - * The FDC should be ready to receive a new command or - * an error (EBUSY or ETIME) will occur. - */ -int fdc_command(const __u8 * cmd_data, int cmd_len) -{ - int result = 0; - unsigned long flags; - int count = cmd_len; - int retry = 0; -#ifdef TESTING - static unsigned int last_time; - unsigned int time; -#endif - TRACE_FUN(ft_t_any); - - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - spin_lock_irqsave(&fdc_io_lock, flags); - if (!in_interrupt()) - /* Yes, I know, too much comments inside this function - * ... - * - * Yet another bug in the original driver. All that - * havoc is caused by the fact that the isr() sends - * itself a command to the floppy tape driver (pause, - * micro step pause). Now, the problem is that - * commands are transmitted via the fdc_seek - * command. But: the fdc performs seeks in the - * background i.e. it doesn't signal busy while - * sending the step pulses to the drive. Therefore the - * non-interrupt level driver has no chance to tell - * whether the isr() just has issued a seek. Therefore - * we HAVE TO have a look at the ft_hide_interrupt - * flag: it signals the non-interrupt level part of - * the driver that it has to wait for the fdc until it - * has completet seeking. - * - * THIS WAS PRESUMABLY THE REASON FOR ALL THAT - * "fdc_read timeout" errors, I HOPE :-) - */ - if (ft_hide_interrupt) { - restore_flags(flags); - TRACE(ft_t_info, - "Waiting for the isr() completing fdc_seek()"); - if (fdc_interrupt_wait(2 * FT_SECOND) < 0) { - TRACE(ft_t_warn, - "Warning: timeout waiting for isr() seek to complete"); - } - if (ft_hide_interrupt || !ft_seek_completed) { - /* There cannot be another - * interrupt. The isr() only stops - * the tape and the next interrupt - * won't come until we have send our - * command to the drive. - */ - TRACE_ABORT(-EIO, ft_t_bug, - "BUG? isr() is still seeking?\n" - KERN_INFO "hide: %d\n" - KERN_INFO "seek: %d", - ft_hide_interrupt, - ft_seek_completed); - - } - fdc_usec_wait(FT_RQM_DELAY); /* wait for valid RQM status */ - spin_lock_irqsave(&fdc_io_lock, flags); - } - fdc_status = inb(fdc.msr); - if ((fdc_status & FDC_DATA_READY_MASK) != FDC_DATA_IN_READY) { - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_ABORT(-EBUSY, ft_t_err, "fdc not ready"); - } - fdc_mode = *cmd_data; /* used by isr */ -#ifdef TESTING - if (fdc_mode == FDC_SEEK) { - time = ftape_timediff(last_time, ftape_timestamp()); - if (time < 6000) { - TRACE(ft_t_bug,"Warning: short timeout between seek commands: %d", - time); - } - } -#endif - if (!in_interrupt()) { - /* shouldn't be cleared if called from isr - */ - ft_interrupt_seen = 0; - } - while (count) { - result = fdc_write(*cmd_data); - if (result < 0) { - TRACE(ft_t_fdc_dma, - "fdc_mode = %02x, status = %02x at index %d", - (int) fdc_mode, (int) fdc_status, - cmd_len - count); - if (++retry <= 3) { - TRACE(ft_t_warn, "fdc_write timeout, retry"); - } else { - TRACE(ft_t_err, "fdc_write timeout, fatal"); - /* recover ??? */ - break; - } - } else { - --count; - ++cmd_data; - } - } -#ifdef TESTING - if (fdc_mode == FDC_SEEK) { - last_time = ftape_timestamp(); - } -#endif - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_EXIT result; -} - -/* Input a res_len long result string from the FDC. - * The FDC should be ready to send the result or an error - * (EBUSY or ETIME) will occur. - */ -int fdc_result(__u8 * res_data, int res_len) -{ - int result = 0; - unsigned long flags; - int count = res_len; - int retry = 0; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&fdc_io_lock, flags); - fdc_status = inb(fdc.msr); - if ((fdc_status & FDC_DATA_READY_MASK) != FDC_DATA_OUT_READY) { - TRACE(ft_t_err, "fdc not ready"); - result = -EBUSY; - } else while (count) { - if (!(fdc_status & FDC_BUSY)) { - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_ABORT(-EIO, ft_t_err, "premature end of result phase"); - } - result = fdc_read(res_data); - if (result < 0) { - TRACE(ft_t_fdc_dma, - "fdc_mode = %02x, status = %02x at index %d", - (int) fdc_mode, - (int) fdc_status, - res_len - count); - if (++retry <= 3) { - TRACE(ft_t_warn, "fdc_read timeout, retry"); - } else { - TRACE(ft_t_err, "fdc_read timeout, fatal"); - /* recover ??? */ - break; - ++retry; - } - } else { - --count; - ++res_data; - } - } - spin_unlock_irqrestore(&fdc_io_lock, flags); - fdc_usec_wait(FT_RQM_DELAY); /* allow FDC to negate BSY */ - TRACE_EXIT result; -} - -/* Handle command and result phases for - * commands without data phase. - */ -static int fdc_issue_command(const __u8 * out_data, int out_count, - __u8 * in_data, int in_count) -{ - TRACE_FUN(ft_t_any); - - if (out_count > 0) { - TRACE_CATCH(fdc_command(out_data, out_count),); - } - /* will take 24 - 30 usec for fdc_sense_drive_status and - * fdc_sense_interrupt_status commands. - * 35 fails sometimes (5/9/93 SJL) - * On a loaded system it incidentally takes longer than - * this for the fdc to get ready ! ?????? WHY ?????? - * So until we know what's going on use a very long timeout. - */ - TRACE_CATCH(fdc_ready_out_wait(500 /* usec */),); - if (in_count > 0) { - TRACE_CATCH(fdc_result(in_data, in_count), - TRACE(ft_t_err, "result phase aborted")); - } - TRACE_EXIT 0; -} - -/* Wait for FDC interrupt with timeout (in milliseconds). - * Signals are blocked so the wait will not be aborted. - * Note: interrupts must be enabled ! (23/05/93 SJL) - */ -int fdc_interrupt_wait(unsigned int time) -{ - DECLARE_WAITQUEUE(wait,current); - sigset_t old_sigmask; - static int resetting; - long timeout; - - TRACE_FUN(ft_t_fdc_dma); - - if (waitqueue_active(&ftape_wait_intr)) { - TRACE_ABORT(-EIO, ft_t_err, "error: nested call"); - } - /* timeout time will be up to USPT microseconds too long ! */ - timeout = (1000 * time + FT_USPT - 1) / FT_USPT; - - spin_lock_irq(¤t->sighand->siglock); - old_sigmask = current->blocked; - sigfillset(¤t->blocked); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&ftape_wait_intr, &wait); - while (!ft_interrupt_seen && timeout) - timeout = schedule_timeout_interruptible(timeout); - - spin_lock_irq(¤t->sighand->siglock); - current->blocked = old_sigmask; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - remove_wait_queue(&ftape_wait_intr, &wait); - /* the following IS necessary. True: as well - * wake_up_interruptible() as the schedule() set TASK_RUNNING - * when they wakeup a task, BUT: it may very well be that - * ft_interrupt_seen is already set to 1 when we enter here - * in which case schedule() gets never called, and - * TASK_RUNNING never set. This has the funny effect that we - * execute all the code until we leave kernel space, but then - * the task is stopped (a task CANNOT be preempted while in - * kernel mode. Sending a pair of SIGSTOP/SIGCONT to the - * tasks wakes it up again. Funny! :-) - */ - current->state = TASK_RUNNING; - if (ft_interrupt_seen) { /* woken up by interrupt */ - ft_interrupt_seen = 0; - TRACE_EXIT 0; - } - /* Original comment: - * In first instance, next statement seems unnecessary since - * it will be cleared in fdc_command. However, a small part of - * the software seems to rely on this being cleared here - * (ftape_close might fail) so stick to it until things get fixed ! - */ - /* My deeply sought of knowledge: - * Behold NO! It is obvious. fdc_reset() doesn't call fdc_command() - * but nevertheless uses fdc_interrupt_wait(). OF COURSE this needs to - * be reset here. - */ - ft_interrupt_seen = 0; /* clear for next call */ - if (!resetting) { - resetting = 1; /* break infinite recursion if reset fails */ - TRACE(ft_t_any, "cleanup reset"); - fdc_reset(); - resetting = 0; - } - TRACE_EXIT (signal_pending(current)) ? -EINTR : -ETIME; -} - -/* Start/stop drive motor. Enable DMA mode. - */ -void fdc_motor(int motor) -{ - int unit = ft_drive_sel; - int data = unit | FDC_RESET_NOT | FDC_DMA_MODE; - TRACE_FUN(ft_t_any); - - ftape_motor = motor; - if (ftape_motor) { - data |= FDC_MOTOR_0 << unit; - TRACE(ft_t_noise, "turning motor %d on", unit); - } else { - TRACE(ft_t_noise, "turning motor %d off", unit); - } - if (ft_mach2) { - outb_p(data, fdc.dor2); - } else { - outb_p(data, fdc.dor); - } - ftape_sleep(10 * FT_MILLISECOND); - TRACE_EXIT; -} - -static void fdc_update_dsr(void) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "rate = %d Kbps, precomp = %d ns", - fdc_data_rate, fdc_precomp); - if (fdc.type >= i82077) { - outb_p((fdc_rate_code & 0x03) | fdc_prec_code, fdc.dsr); - } else { - outb_p(fdc_rate_code & 0x03, fdc.ccr); - } - TRACE_EXIT; -} - -void fdc_set_write_precomp(int precomp) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_noise, "New precomp: %d nsec", precomp); - fdc_precomp = precomp; - /* write precompensation can be set in multiples of 41.67 nsec. - * round the parameter to the nearest multiple and convert it - * into a fdc setting. Note that 0 means default to the fdc, - * 7 is used instead of that. - */ - fdc_prec_code = ((fdc_precomp + 21) / 42) << 2; - if (fdc_prec_code == 0 || fdc_prec_code > (6 << 2)) { - fdc_prec_code = 7 << 2; - } - fdc_update_dsr(); - TRACE_EXIT; -} - -/* Reprogram the 82078 registers to use Data Rate Table 1 on all drives. - */ -static void fdc_set_drive_specs(void) -{ - __u8 cmd[] = { FDC_DRIVE_SPEC, 0x00, 0x00, 0x00, 0x00, 0xc0}; - int result; - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "Setting of drive specs called"); - if (fdc.type >= i82078_1) { - cmd[1] = (0 << 5) | (2 << 2); - cmd[2] = (1 << 5) | (2 << 2); - cmd[3] = (2 << 5) | (2 << 2); - cmd[4] = (3 << 5) | (2 << 2); - result = fdc_command(cmd, NR_ITEMS(cmd)); - if (result < 0) { - TRACE(ft_t_err, "Setting of drive specs failed"); - } - } - TRACE_EXIT; -} - -/* Select clock for fdc, must correspond with tape drive setting ! - * This also influences the fdc timing so we must adjust some values. - */ -int fdc_set_data_rate(int rate) -{ - int bad_rate = 0; - TRACE_FUN(ft_t_any); - - /* Select clock for fdc, must correspond with tape drive setting ! - * This also influences the fdc timing so we must adjust some values. - */ - TRACE(ft_t_fdc_dma, "new rate = %d", rate); - switch (rate) { - case 250: - fdc_rate_code = fdc_data_rate_250; - break; - case 500: - fdc_rate_code = fdc_data_rate_500; - break; - case 1000: - if (fdc.type < i82077) { - bad_rate = 1; - } else { - fdc_rate_code = fdc_data_rate_1000; - } - break; - case 2000: - if (fdc.type < i82078_1) { - bad_rate = 1; - } else { - fdc_rate_code = fdc_data_rate_2000; - } - break; - default: - bad_rate = 1; - } - if (bad_rate) { - TRACE_ABORT(-EIO, - ft_t_fdc_dma, "%d is not a valid data rate", rate); - } - fdc_data_rate = rate; - fdc_update_dsr(); - fdc_set_seek_rate(fdc_seek_rate); /* clock changed! */ - ftape_udelay(1000); - TRACE_EXIT 0; -} - -/* keep the unit select if keep_select is != 0, - */ -static void fdc_dor_reset(int keep_select) -{ - __u8 fdc_ctl = ft_drive_sel; - - if (keep_select != 0) { - fdc_ctl |= FDC_DMA_MODE; - if (ftape_motor) { - fdc_ctl |= FDC_MOTOR_0 << ft_drive_sel; - } - } - ftape_udelay(10); /* ??? but seems to be necessary */ - if (ft_mach2) { - outb_p(fdc_ctl & 0x0f, fdc.dor); - outb_p(fdc_ctl, fdc.dor2); - } else { - outb_p(fdc_ctl, fdc.dor); - } - fdc_usec_wait(10); /* delay >= 14 fdc clocks */ - if (keep_select == 0) { - fdc_ctl = 0; - } - fdc_ctl |= FDC_RESET_NOT; - if (ft_mach2) { - outb_p(fdc_ctl & 0x0f, fdc.dor); - outb_p(fdc_ctl, fdc.dor2); - } else { - outb_p(fdc_ctl, fdc.dor); - } -} - -/* Reset the floppy disk controller. Leave the ftape_unit selected. - */ -void fdc_reset(void) -{ - int st0; - int i; - int dummy; - unsigned long flags; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&fdc_io_lock, flags); - - fdc_dor_reset(1); /* keep unit selected */ - - fdc_mode = fdc_idle; - - /* maybe the spin_lock_irq* pair is not necessary, BUT: - * the following line MUST be here. Otherwise fdc_interrupt_wait() - * won't wait. Note that fdc_reset() is called from - * ftape_dumb_stop() when the fdc is busy transferring data. In this - * case fdc_isr() MOST PROBABLY sets ft_interrupt_seen, and tries - * to get the result bytes from the fdc etc. CLASH. - */ - ft_interrupt_seen = 0; - - /* Program data rate - */ - fdc_update_dsr(); /* restore data rate and precomp */ - - spin_unlock_irqrestore(&fdc_io_lock, flags); - - /* - * Wait for first polling cycle to complete - */ - if (fdc_interrupt_wait(1 * FT_SECOND) < 0) { - TRACE(ft_t_err, "no drive polling interrupt!"); - } else { /* clear all disk-changed statuses */ - for (i = 0; i < 4; ++i) { - if(fdc_sense_interrupt_status(&st0, &dummy) != 0) { - TRACE(ft_t_err, "sense failed for %d", i); - } - if (i == ft_drive_sel) { - ftape_current_cylinder = dummy; - } - } - TRACE(ft_t_noise, "drive polling completed"); - } - /* - * SPECIFY COMMAND - */ - fdc_set_seek_rate(fdc_seek_rate); - /* - * DRIVE SPECIFICATION COMMAND (if fdc type known) - */ - if (fdc.type >= i82078_1) { - fdc_set_drive_specs(); - } - TRACE_EXIT; -} - -#if !defined(CLK_48MHZ) -# define CLK_48MHZ 1 -#endif - -/* When we're done, put the fdc into reset mode so that the regular - * floppy disk driver will figure out that something is wrong and - * initialize the controller the way it wants. - */ -void fdc_disable(void) -{ - __u8 cmd1[] = {FDC_CONFIGURE, 0x00, 0x00, 0x00}; - __u8 cmd2[] = {FDC_LOCK}; - __u8 cmd3[] = {FDC_UNLOCK}; - __u8 stat[1]; - TRACE_FUN(ft_t_flow); - - if (!fdc_fifo_locked) { - fdc_reset(); - TRACE_EXIT; - } - if (fdc_issue_command(cmd3, 1, stat, 1) < 0 || stat[0] != 0x00) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, - "couldn't unlock fifo, configuration remains changed"); - } - fdc_fifo_locked = 0; - if (CLK_48MHZ && fdc.type >= i82078) { - cmd1[0] |= FDC_CLK48_BIT; - } - cmd1[2] = ((fdc_fifo_state) ? 0 : 0x20) + (fdc_fifo_thr - 1); - if (fdc_command(cmd1, NR_ITEMS(cmd1)) < 0) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, - "couldn't reconfigure fifo to old state"); - } - if (fdc_lock_state && - fdc_issue_command(cmd2, 1, stat, 1) < 0) { - fdc_dor_reset(0); - TRACE_ABORT(/**/, ft_t_bug, "couldn't lock old state again"); - } - TRACE(ft_t_noise, "fifo restored: %sabled, thr. %d, %slocked", - fdc_fifo_state ? "en" : "dis", - fdc_fifo_thr, (fdc_lock_state) ? "" : "not "); - fdc_dor_reset(0); - TRACE_EXIT; -} - -/* Specify FDC seek-rate (milliseconds) - */ -static int fdc_set_seek_rate(int seek_rate) -{ - /* set step rate, dma mode, and minimal head load and unload times - */ - __u8 in[3] = { FDC_SPECIFY, 1, (1 << 1)}; - - fdc_seek_rate = seek_rate; - in[1] |= (16 - (fdc_data_rate * fdc_seek_rate) / 500) << 4; - - return fdc_command(in, 3); -} - -/* Sense drive status: get unit's drive status (ST3) - */ -int fdc_sense_drive_status(int *st3) -{ - __u8 out[2]; - __u8 in[1]; - TRACE_FUN(ft_t_any); - - out[0] = FDC_SENSED; - out[1] = ft_drive_sel; - TRACE_CATCH(fdc_issue_command(out, 2, in, 1),); - *st3 = in[0]; - TRACE_EXIT 0; -} - -/* Sense Interrupt Status command: - * should be issued at the end of each seek. - * get ST0 and current cylinder. - */ -int fdc_sense_interrupt_status(int *st0, int *current_cylinder) -{ - __u8 out[1]; - __u8 in[2]; - TRACE_FUN(ft_t_any); - - out[0] = FDC_SENSEI; - TRACE_CATCH(fdc_issue_command(out, 1, in, 2),); - *st0 = in[0]; - *current_cylinder = in[1]; - TRACE_EXIT 0; -} - -/* step to track - */ -int fdc_seek(int track) -{ - __u8 out[3]; - int st0, pcn; -#ifdef TESTING - unsigned int time; -#endif - TRACE_FUN(ft_t_any); - - out[0] = FDC_SEEK; - out[1] = ft_drive_sel; - out[2] = track; -#ifdef TESTING - time = ftape_timestamp(); -#endif - /* We really need this command to work ! - */ - ft_seek_completed = 0; - TRACE_CATCH(fdc_command(out, 3), - fdc_reset(); - TRACE(ft_t_noise, "destination was: %d, resetting FDC...", - track)); - /* Handle interrupts until ft_seek_completed or timeout. - */ - for (;;) { - TRACE_CATCH(fdc_interrupt_wait(2 * FT_SECOND),); - if (ft_seek_completed) { - TRACE_CATCH(fdc_sense_interrupt_status(&st0, &pcn),); - if ((st0 & ST0_SEEK_END) == 0) { - TRACE_ABORT(-EIO, ft_t_err, - "no seek-end after seek completion !??"); - } - break; - } - } -#ifdef TESTING - time = ftape_timediff(time, ftape_timestamp()) / abs(track - ftape_current_cylinder); - if ((time < 900 || time > 3100) && abs(track - ftape_current_cylinder) > 5) { - TRACE(ft_t_warn, "Wrong FDC STEP interval: %d usecs (%d)", - time, track - ftape_current_cylinder); - } -#endif - /* Verify whether we issued the right tape command. - */ - /* Verify that we seek to the proper track. */ - if (pcn != track) { - TRACE_ABORT(-EIO, ft_t_err, "bad seek.."); - } - ftape_current_cylinder = track; - TRACE_EXIT 0; -} - -static int perpend_mode; /* set if fdc is in perpendicular mode */ - -static int perpend_off(void) -{ - __u8 perpend[] = {FDC_PERPEND, 0x00}; - TRACE_FUN(ft_t_any); - - if (perpend_mode) { - /* Turn off perpendicular mode */ - perpend[1] = 0x80; - TRACE_CATCH(fdc_command(perpend, 2), - TRACE(ft_t_err,"Perpendicular mode exit failed!")); - perpend_mode = 0; - } - TRACE_EXIT 0; -} - -static int handle_perpend(int segment_id) -{ - __u8 perpend[] = {FDC_PERPEND, 0x00}; - TRACE_FUN(ft_t_any); - - /* When writing QIC-3020 tapes, turn on perpendicular mode - * if tape is moving in forward direction (even tracks). - */ - if (ft_qic_std == QIC_TAPE_QIC3020 && - ((segment_id / ft_segments_per_track) & 1) == 0) { -/* FIXME: some i82077 seem to support perpendicular mode as - * well. - */ -#if 0 - if (fdc.type < i82077AA) {} -#else - if (fdc.type < i82077 && ft_data_rate < 1000) { -#endif - /* fdc does not support perpendicular mode: complain - */ - TRACE_ABORT(-EIO, ft_t_err, - "Your FDC does not support QIC-3020."); - } - perpend[1] = 0x03 /* 0x83 + (0x4 << ft_drive_sel) */ ; - TRACE_CATCH(fdc_command(perpend, 2), - TRACE(ft_t_err,"Perpendicular mode entry failed!")); - TRACE(ft_t_flow, "Perpendicular mode set"); - perpend_mode = 1; - TRACE_EXIT 0; - } - TRACE_EXIT perpend_off(); -} - -static inline void fdc_setup_dma(char mode, - volatile void *addr, unsigned int count) -{ - /* Program the DMA controller. - */ - disable_dma(fdc.dma); - clear_dma_ff(fdc.dma); - set_dma_mode(fdc.dma, mode); - set_dma_addr(fdc.dma, virt_to_bus((void*)addr)); - set_dma_count(fdc.dma, count); - enable_dma(fdc.dma); -} - -/* Setup fdc and dma for formatting the next segment - */ -int fdc_setup_formatting(buffer_struct * buff) -{ - unsigned long flags; - __u8 out[6] = { - FDC_FORMAT, 0x00, 3, 4 * FT_SECTORS_PER_SEGMENT, 0x00, 0x6b - }; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(handle_perpend(buff->segment_id),); - /* Program the DMA controller. - */ - TRACE(ft_t_fdc_dma, - "phys. addr. = %lx", virt_to_bus((void*) buff->ptr)); - spin_lock_irqsave(&fdc_io_lock, flags); - fdc_setup_dma(DMA_MODE_WRITE, buff->ptr, FT_SECTORS_PER_SEGMENT * 4); - /* Issue FDC command to start reading/writing. - */ - out[1] = ft_drive_sel; - out[4] = buff->gap3; - TRACE_CATCH(fdc_setup_error = fdc_command(out, sizeof(out)), - restore_flags(flags); fdc_mode = fdc_idle); - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_EXIT 0; -} - - -/* Setup Floppy Disk Controller and DMA to read or write the next cluster - * of good sectors from or to the current segment. - */ -int fdc_setup_read_write(buffer_struct * buff, __u8 operation) -{ - unsigned long flags; - __u8 out[9]; - int dma_mode; - TRACE_FUN(ft_t_any); - - switch(operation) { - case FDC_VERIFY: - if (fdc.type < i82077) { - operation = FDC_READ; - } - case FDC_READ: - case FDC_READ_DELETED: - dma_mode = DMA_MODE_READ; - TRACE(ft_t_fdc_dma, "xfer %d sectors to 0x%p", - buff->sector_count, buff->ptr); - TRACE_CATCH(perpend_off(),); - break; - case FDC_WRITE_DELETED: - TRACE(ft_t_noise, "deleting segment %d", buff->segment_id); - case FDC_WRITE: - dma_mode = DMA_MODE_WRITE; - /* When writing QIC-3020 tapes, turn on perpendicular mode - * if tape is moving in forward direction (even tracks). - */ - TRACE_CATCH(handle_perpend(buff->segment_id),); - TRACE(ft_t_fdc_dma, "xfer %d sectors from 0x%p", - buff->sector_count, buff->ptr); - break; - default: - TRACE_ABORT(-EIO, - ft_t_bug, "bug: invalid operation parameter"); - } - TRACE(ft_t_fdc_dma, "phys. addr. = %lx",virt_to_bus((void*)buff->ptr)); - spin_lock_irqsave(&fdc_io_lock, flags); - if (operation != FDC_VERIFY) { - fdc_setup_dma(dma_mode, buff->ptr, - FT_SECTOR_SIZE * buff->sector_count); - } - /* Issue FDC command to start reading/writing. - */ - out[0] = operation; - out[1] = ft_drive_sel; - out[2] = buff->cyl; - out[3] = buff->head; - out[4] = buff->sect + buff->sector_offset; - out[5] = 3; /* Sector size of 1K. */ - out[6] = out[4] + buff->sector_count - 1; /* last sector */ - out[7] = 109; /* Gap length. */ - out[8] = 0xff; /* No limit to transfer size. */ - TRACE(ft_t_fdc_dma, "C: 0x%02x, H: 0x%02x, R: 0x%02x, cnt: 0x%02x", - out[2], out[3], out[4], out[6] - out[4] + 1); - spin_unlock_irqrestore(&fdc_io_lock, flags); - TRACE_CATCH(fdc_setup_error = fdc_command(out, 9),fdc_mode = fdc_idle); - TRACE_EXIT 0; -} - -int fdc_fifo_threshold(__u8 threshold, - int *fifo_state, int *lock_state, int *fifo_thr) -{ - const __u8 cmd0[] = {FDC_DUMPREGS}; - __u8 cmd1[] = {FDC_CONFIGURE, 0, (0x0f & (threshold - 1)), 0}; - const __u8 cmd2[] = {FDC_LOCK}; - const __u8 cmd3[] = {FDC_UNLOCK}; - __u8 reg[10]; - __u8 stat; - int i; - int result; - TRACE_FUN(ft_t_any); - - if (CLK_48MHZ && fdc.type >= i82078) { - cmd1[0] |= FDC_CLK48_BIT; - } - /* Dump fdc internal registers for examination - */ - TRACE_CATCH(fdc_command(cmd0, NR_ITEMS(cmd0)), - TRACE(ft_t_warn, "dumpreg cmd failed, fifo unchanged")); - /* Now read fdc internal registers from fifo - */ - for (i = 0; i < (int)NR_ITEMS(reg); ++i) { - fdc_read(®[i]); - TRACE(ft_t_fdc_dma, "Register %d = 0x%02x", i, reg[i]); - } - if (fifo_state && lock_state && fifo_thr) { - *fifo_state = (reg[8] & 0x20) == 0; - *lock_state = reg[7] & 0x80; - *fifo_thr = 1 + (reg[8] & 0x0f); - } - TRACE(ft_t_noise, - "original fifo state: %sabled, threshold %d, %slocked", - ((reg[8] & 0x20) == 0) ? "en" : "dis", - 1 + (reg[8] & 0x0f), (reg[7] & 0x80) ? "" : "not "); - /* If fdc is already locked, unlock it first ! */ - if (reg[7] & 0x80) { - fdc_ready_wait(100); - TRACE_CATCH(fdc_issue_command(cmd3, NR_ITEMS(cmd3), &stat, 1), - TRACE(ft_t_bug, "FDC unlock command failed, " - "configuration unchanged")); - } - fdc_fifo_locked = 0; - /* Enable fifo and set threshold at xx bytes to allow a - * reasonably large latency and reduce number of dma bursts. - */ - fdc_ready_wait(100); - if ((result = fdc_command(cmd1, NR_ITEMS(cmd1))) < 0) { - TRACE(ft_t_bug, "configure cmd failed, fifo unchanged"); - } - /* Now lock configuration so reset will not change it - */ - if(fdc_issue_command(cmd2, NR_ITEMS(cmd2), &stat, 1) < 0 || - stat != 0x10) { - TRACE_ABORT(-EIO, ft_t_bug, - "FDC lock command failed, stat = 0x%02x", stat); - } - fdc_fifo_locked = 1; - TRACE_EXIT result; -} - -static int fdc_fifo_enable(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc_fifo_locked) { - TRACE_ABORT(0, ft_t_warn, "Fifo not enabled because locked"); - } - TRACE_CATCH(fdc_fifo_threshold(ft_fdc_threshold /* bytes */, - &fdc_fifo_state, - &fdc_lock_state, - &fdc_fifo_thr),); - TRACE_CATCH(fdc_fifo_threshold(ft_fdc_threshold /* bytes */, - NULL, NULL, NULL),); - TRACE_EXIT 0; -} - -/* Determine fd controller type - */ -static __u8 fdc_save_state[2]; - -static int fdc_probe(void) -{ - __u8 cmd[1]; - __u8 stat[16]; /* must be able to hold dumpregs & save results */ - int i; - TRACE_FUN(ft_t_any); - - /* Try to find out what kind of fd controller we have to deal with - * Scheme borrowed from floppy driver: - * first try if FDC_DUMPREGS command works - * (this indicates that we have a 82072 or better) - * then try the FDC_VERSION command (82072 doesn't support this) - * then try the FDC_UNLOCK command (some older 82077's don't support this) - * then try the FDC_PARTID command (82078's support this) - */ - cmd[0] = FDC_DUMPREGS; - if (fdc_issue_command(cmd, 1, stat, 1) != 0) { - TRACE_ABORT(no_fdc, ft_t_bug, "No FDC found"); - } - if (stat[0] == 0x80) { - /* invalid command: must be pre 82072 */ - TRACE_ABORT(i8272, - ft_t_warn, "Type 8272A/765A compatible FDC found"); - } - fdc_result(&stat[1], 9); - fdc_save_state[0] = stat[7]; - fdc_save_state[1] = stat[8]; - cmd[0] = FDC_VERSION; - if (fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] == 0x80) { - TRACE_ABORT(i8272, ft_t_warn, "Type 82072 FDC found"); - } - if (*stat != 0x90) { - TRACE_ABORT(i8272, ft_t_warn, "Unknown FDC found"); - } - cmd[0] = FDC_UNLOCK; - if(fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] != 0x00) { - TRACE_ABORT(i8272, ft_t_warn, - "Type pre-1991 82077 FDC found, " - "treating it like a 82072"); - } - if (fdc_save_state[0] & 0x80) { /* was locked */ - cmd[0] = FDC_LOCK; /* restore lock */ - (void)fdc_issue_command(cmd, 1, stat, 1); - TRACE(ft_t_warn, "FDC is already locked"); - } - /* Test for a i82078 FDC */ - cmd[0] = FDC_PARTID; - if (fdc_issue_command(cmd, 1, stat, 1) < 0 || stat[0] == 0x80) { - /* invalid command: not a i82078xx type FDC */ - for (i = 0; i < 4; ++i) { - outb_p(i, fdc.tdr); - if ((inb_p(fdc.tdr) & 0x03) != i) { - TRACE_ABORT(i82077, - ft_t_warn, "Type 82077 FDC found"); - } - } - TRACE_ABORT(i82077AA, ft_t_warn, "Type 82077AA FDC found"); - } - /* FDC_PARTID cmd succeeded */ - switch (stat[0] >> 5) { - case 0x0: - /* i82078SL or i82078-1. The SL part cannot run at - * 2Mbps (the SL and -1 dies are identical; they are - * speed graded after production, according to Intel). - * Some SL's can be detected by doing a SAVE cmd and - * look at bit 7 of the first byte (the SEL3V# bit). - * If it is 0, the part runs off 3Volts, and hence it - * is a SL. - */ - cmd[0] = FDC_SAVE; - if(fdc_issue_command(cmd, 1, stat, 16) < 0) { - TRACE(ft_t_err, "FDC_SAVE failed. Dunno why"); - /* guess we better claim the fdc to be a i82078 */ - TRACE_ABORT(i82078, - ft_t_warn, - "Type i82078 FDC (i suppose) found"); - } - if ((stat[0] & FDC_SEL3V_BIT)) { - /* fdc running off 5Volts; Pray that it's a i82078-1 - */ - TRACE_ABORT(i82078_1, ft_t_warn, - "Type i82078-1 or 5Volt i82078SL FDC found"); - } - TRACE_ABORT(i82078, ft_t_warn, - "Type 3Volt i82078SL FDC (1Mbps) found"); - case 0x1: - case 0x2: /* S82078B */ - /* The '78B isn't '78 compatible. Detect it as a '77AA */ - TRACE_ABORT(i82077AA, ft_t_warn, "Type i82077AA FDC found"); - case 0x3: /* NSC PC8744 core; used in several super-IO chips */ - TRACE_ABORT(i82077AA, - ft_t_warn, "Type 82077AA compatible FDC found"); - default: - TRACE(ft_t_warn, "A previously undetected FDC found"); - TRACE_ABORT(i82077AA, ft_t_warn, - "Treating it as a 82077AA. Please report partid= %d", - stat[0]); - } /* switch(stat[ 0] >> 5) */ - TRACE_EXIT no_fdc; -} - -static int fdc_request_regions(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_mach2 || ft_probe_fc10) { - if (!request_region(fdc.sra, 8, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra); -#endif - } - } else { - if (!request_region(fdc.sra, 6, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra); -#endif - } - if (!request_region(fdc.sra + 7, 1, "fdc (ft)")) { -#ifndef BROKEN_FLOPPY_DRIVER - release_region(fdc.sra, 6); - TRACE_EXIT -EBUSY; -#else - TRACE(ft_t_warn, -"address 0x%03x occupied (by floppy driver?), using it anyway", fdc.sra + 7); -#endif - } - } - TRACE_EXIT 0; -} - -void fdc_release_regions(void) -{ - TRACE_FUN(ft_t_flow); - - if (fdc.sra != 0) { - if (fdc.dor2 != 0) { - release_region(fdc.sra, 8); - } else { - release_region(fdc.sra, 6); - release_region(fdc.dir, 1); - } - } - TRACE_EXIT; -} - -static int fdc_config_regs(unsigned int fdc_base, - unsigned int fdc_irq, - unsigned int fdc_dma) -{ - TRACE_FUN(ft_t_flow); - - fdc.irq = fdc_irq; - fdc.dma = fdc_dma; - fdc.sra = fdc_base; - fdc.srb = fdc_base + 1; - fdc.dor = fdc_base + 2; - fdc.tdr = fdc_base + 3; - fdc.msr = fdc.dsr = fdc_base + 4; - fdc.fifo = fdc_base + 5; - fdc.dir = fdc.ccr = fdc_base + 7; - fdc.dor2 = (ft_mach2 || ft_probe_fc10) ? fdc_base + 6 : 0; - TRACE_CATCH(fdc_request_regions(), fdc.sra = 0); - TRACE_EXIT 0; -} - -static int fdc_config(void) -{ - static int already_done; - TRACE_FUN(ft_t_any); - - if (already_done) { - TRACE_CATCH(fdc_request_regions(),); - *(fdc.hook) = fdc_isr; /* hook our handler in */ - TRACE_EXIT 0; - } - if (ft_probe_fc10) { - int fc_type; - - TRACE_CATCH(fdc_config_regs(ft_fdc_base, - ft_fdc_irq, ft_fdc_dma),); - fc_type = fc10_enable(); - if (fc_type != 0) { - TRACE(ft_t_warn, "FC-%c0 controller found", '0' + fc_type); - fdc.type = fc10; - fdc.hook = &do_ftape; - *(fdc.hook) = fdc_isr; /* hook our handler in */ - already_done = 1; - TRACE_EXIT 0; - } else { - TRACE(ft_t_warn, "FC-10/20 controller not found"); - fdc_release_regions(); - fdc.type = no_fdc; - ft_probe_fc10 = 0; - ft_fdc_base = 0x3f0; - ft_fdc_irq = 6; - ft_fdc_dma = 2; - } - } - TRACE(ft_t_warn, "fdc base: 0x%x, irq: %d, dma: %d", - ft_fdc_base, ft_fdc_irq, ft_fdc_dma); - TRACE_CATCH(fdc_config_regs(ft_fdc_base, ft_fdc_irq, ft_fdc_dma),); - fdc.hook = &do_ftape; - *(fdc.hook) = fdc_isr; /* hook our handler in */ - already_done = 1; - TRACE_EXIT 0; -} - -static irqreturn_t ftape_interrupt(int irq, void *dev_id) -{ - void (*handler) (void) = *fdc.hook; - int handled = 0; - TRACE_FUN(ft_t_any); - - *fdc.hook = NULL; - if (handler) { - handled = 1; - handler(); - } else { - TRACE(ft_t_bug, "Unexpected ftape interrupt"); - } - TRACE_EXIT IRQ_RETVAL(handled); -} - -static int fdc_grab_irq_and_dma(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc.hook == &do_ftape) { - /* Get fast interrupt handler. - */ - if (request_irq(fdc.irq, ftape_interrupt, - IRQF_DISABLED, "ft", ftape_id)) { - TRACE_ABORT(-EIO, ft_t_bug, - "Unable to grab IRQ%d for ftape driver", - fdc.irq); - } - if (request_dma(fdc.dma, ftape_id)) { - free_irq(fdc.irq, ftape_id); - TRACE_ABORT(-EIO, ft_t_bug, - "Unable to grab DMA%d for ftape driver", - fdc.dma); - } - } - if (ft_fdc_base != 0x3f0 && (ft_fdc_dma == 2 || ft_fdc_irq == 6)) { - /* Using same dma channel or irq as standard fdc, need - * to disable the dma-gate on the std fdc. This - * couldn't be done in the floppy driver as some - * laptops are using the dma-gate to enter a low power - * or even suspended state :-( - */ - outb_p(FDC_RESET_NOT, 0x3f2); - TRACE(ft_t_noise, "DMA-gate on standard fdc disabled"); - } - TRACE_EXIT 0; -} - -int fdc_release_irq_and_dma(void) -{ - TRACE_FUN(ft_t_any); - - if (fdc.hook == &do_ftape) { - disable_dma(fdc.dma); /* just in case... */ - free_dma(fdc.dma); - free_irq(fdc.irq, ftape_id); - } - if (ft_fdc_base != 0x3f0 && (ft_fdc_dma == 2 || ft_fdc_irq == 6)) { - /* Using same dma channel as standard fdc, need to - * disable the dma-gate on the std fdc. This couldn't - * be done in the floppy driver as some laptops are - * using the dma-gate to enter a low power or even - * suspended state :-( - */ - outb_p(FDC_RESET_NOT | FDC_DMA_MODE, 0x3f2); - TRACE(ft_t_noise, "DMA-gate on standard fdc enabled again"); - } - TRACE_EXIT 0; -} - -int fdc_init(void) -{ - TRACE_FUN(ft_t_any); - - /* find a FDC to use */ - TRACE_CATCH(fdc_config(),); - TRACE_CATCH(fdc_grab_irq_and_dma(), fdc_release_regions()); - ftape_motor = 0; - fdc_catch_stray_interrupts(0); /* clear number of awainted - * stray interrupte - */ - fdc_catch_stray_interrupts(1); /* one always comes (?) */ - TRACE(ft_t_flow, "resetting fdc"); - fdc_set_seek_rate(2); /* use nominal QIC step rate */ - fdc_reset(); /* init fdc & clear track counters */ - if (fdc.type == no_fdc) { /* no FC-10 or FC-20 found */ - fdc.type = fdc_probe(); - fdc_reset(); /* update with new knowledge */ - } - if (fdc.type == no_fdc) { - fdc_release_irq_and_dma(); - fdc_release_regions(); - TRACE_EXIT -ENXIO; - } - if (fdc.type >= i82077) { - if (fdc_fifo_enable() < 0) { - TRACE(ft_t_warn, "couldn't enable fdc fifo !"); - } else { - TRACE(ft_t_flow, "fdc fifo enabled and locked"); - } - } - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/fdc-io.h b/drivers/char/ftape/lowlevel/fdc-io.h deleted file mode 100644 index 7ec3c72178bb..000000000000 --- a/drivers/char/ftape/lowlevel/fdc-io.h +++ /dev/null @@ -1,252 +0,0 @@ -#ifndef _FDC_IO_H -#define _FDC_IO_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-io.h,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:18:06 $ - * - * This file contains the declarations for the low level - * functions that communicate with the floppy disk controller, - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include - -#include "../lowlevel/ftape-bsm.h" - -#define FDC_SK_BIT (0x20) -#define FDC_MT_BIT (0x80) - -#define FDC_READ (FD_READ & ~(FDC_SK_BIT | FDC_MT_BIT)) -#define FDC_WRITE (FD_WRITE & ~FDC_MT_BIT) -#define FDC_READ_DELETED (0x4c) -#define FDC_WRITE_DELETED (0x49) -#define FDC_VERIFY (0x56) -#define FDC_READID (0x4a) -#define FDC_SENSED (0x04) -#define FDC_SENSEI (FD_SENSEI) -#define FDC_FORMAT (FD_FORMAT) -#define FDC_RECAL (FD_RECALIBRATE) -#define FDC_SEEK (FD_SEEK) -#define FDC_SPECIFY (FD_SPECIFY) -#define FDC_RECALIBR (FD_RECALIBRATE) -#define FDC_VERSION (FD_VERSION) -#define FDC_PERPEND (FD_PERPENDICULAR) -#define FDC_DUMPREGS (FD_DUMPREGS) -#define FDC_LOCK (FD_LOCK) -#define FDC_UNLOCK (FD_UNLOCK) -#define FDC_CONFIGURE (FD_CONFIGURE) -#define FDC_DRIVE_SPEC (0x8e) /* i82078 has this (any others?) */ -#define FDC_PARTID (0x18) /* i82078 has this */ -#define FDC_SAVE (0x2e) /* i82078 has this (any others?) */ -#define FDC_RESTORE (0x4e) /* i82078 has this (any others?) */ - -#define FDC_STATUS_MASK (STATUS_BUSY | STATUS_DMA | STATUS_DIR | STATUS_READY) -#define FDC_DATA_READY (STATUS_READY) -#define FDC_DATA_OUTPUT (STATUS_DIR) -#define FDC_DATA_READY_MASK (STATUS_READY | STATUS_DIR) -#define FDC_DATA_OUT_READY (STATUS_READY | STATUS_DIR) -#define FDC_DATA_IN_READY (STATUS_READY) -#define FDC_BUSY (STATUS_BUSY) -#define FDC_CLK48_BIT (0x80) -#define FDC_SEL3V_BIT (0x40) - -#define ST0_INT_MASK (ST0_INTR) -#define FDC_INT_NORMAL (ST0_INTR & 0x00) -#define FDC_INT_ABNORMAL (ST0_INTR & 0x40) -#define FDC_INT_INVALID (ST0_INTR & 0x80) -#define FDC_INT_READYCH (ST0_INTR & 0xC0) -#define ST0_SEEK_END (ST0_SE) -#define ST3_TRACK_0 (ST3_TZ) - -#define FDC_RESET_NOT (0x04) -#define FDC_DMA_MODE (0x08) -#define FDC_MOTOR_0 (0x10) -#define FDC_MOTOR_1 (0x20) - -typedef struct { - void (**hook) (void); /* our wedge into the isr */ - enum { - no_fdc, i8272, i82077, i82077AA, fc10, - i82078, i82078_1 - } type; /* FDC type */ - unsigned int irq; /* FDC irq nr */ - unsigned int dma; /* FDC dma channel nr */ - __u16 sra; /* Status register A (PS/2 only) */ - __u16 srb; /* Status register B (PS/2 only) */ - __u16 dor; /* Digital output register */ - __u16 tdr; /* Tape Drive Register (82077SL-1 & - 82078 only) */ - __u16 msr; /* Main Status Register */ - __u16 dsr; /* Datarate Select Register (8207x only) */ - __u16 fifo; /* Data register / Fifo on 8207x */ - __u16 dir; /* Digital Input Register */ - __u16 ccr; /* Configuration Control Register */ - __u16 dor2; /* Alternate dor on MACH-2 controller, - also used with FC-10, meaning unknown */ -} fdc_config_info; - -typedef enum { - fdc_data_rate_250 = 2, - fdc_data_rate_300 = 1, /* any fdc in default configuration */ - fdc_data_rate_500 = 0, - fdc_data_rate_1000 = 3, - fdc_data_rate_2000 = 1, /* i82078-1: when using Data Rate Table #2 */ -} fdc_data_rate_type; - -typedef enum { - fdc_idle = 0, - fdc_reading_data = FDC_READ, - fdc_seeking = FDC_SEEK, - fdc_writing_data = FDC_WRITE, - fdc_deleting = FDC_WRITE_DELETED, - fdc_reading_id = FDC_READID, - fdc_recalibrating = FDC_RECAL, - fdc_formatting = FDC_FORMAT, - fdc_verifying = FDC_VERIFY -} fdc_mode_enum; - -typedef enum { - waiting = 0, - reading, - writing, - formatting, - verifying, - deleting, - done, - error, - mmapped, -} buffer_state_enum; - -typedef struct { - __u8 *address; - volatile buffer_state_enum status; - volatile __u8 *ptr; - volatile unsigned int bytes; - volatile unsigned int segment_id; - - /* bitmap for remainder of segment not yet handled. - * one bit set for each bad sector that must be skipped. - */ - volatile SectorMap bad_sector_map; - - /* bitmap with bad data blocks in data buffer. - * the errors in this map may be retried. - */ - volatile SectorMap soft_error_map; - - /* bitmap with bad data blocks in data buffer - * the errors in this map may not be retried. - */ - volatile SectorMap hard_error_map; - - /* retry counter for soft errors. - */ - volatile int retry; - - /* sectors to skip on retry ??? - */ - volatile unsigned int skip; - - /* nr of data blocks in data buffer - */ - volatile unsigned int data_offset; - - /* offset in segment for first sector to be handled. - */ - volatile unsigned int sector_offset; - - /* size of cluster of good sectors to be handled. - */ - volatile unsigned int sector_count; - - /* size of remaining part of segment to be handled. - */ - volatile unsigned int remaining; - - /* points to next segment (contiguous) to be handled, - * or is zero if no read-ahead is allowed. - */ - volatile unsigned int next_segment; - - /* flag being set if deleted data was read. - */ - volatile int deleted; - - /* floppy coordinates of first sector in segment */ - volatile __u8 head; - volatile __u8 cyl; - volatile __u8 sect; - - /* gap to use when formatting */ - __u8 gap3; - /* flag set when buffer is mmaped */ - int mmapped; -} buffer_struct; - -/* - * fdc-io.c defined public variables - */ -extern volatile fdc_mode_enum fdc_mode; -extern int fdc_setup_error; /* outdated ??? */ -extern wait_queue_head_t ftape_wait_intr; -extern volatile int ftape_current_cylinder; /* track nr FDC thinks we're on */ -extern volatile __u8 fdc_head; /* FDC head */ -extern volatile __u8 fdc_cyl; /* FDC track */ -extern volatile __u8 fdc_sect; /* FDC sector */ -extern fdc_config_info fdc; /* FDC hardware configuration */ - -extern unsigned int ft_fdc_base; -extern unsigned int ft_fdc_irq; -extern unsigned int ft_fdc_dma; -extern unsigned int ft_fdc_threshold; -extern unsigned int ft_fdc_rate_limit; -extern int ft_probe_fc10; -extern int ft_mach2; -/* - * fdc-io.c defined public functions - */ -extern void fdc_catch_stray_interrupts(int count); -extern int fdc_ready_wait(unsigned int timeout); -extern int fdc_command(const __u8 * cmd_data, int cmd_len); -extern int fdc_result(__u8 * res_data, int res_len); -extern int fdc_interrupt_wait(unsigned int time); -extern int fdc_seek(int track); -extern int fdc_sense_drive_status(int *st3); -extern void fdc_motor(int motor); -extern void fdc_reset(void); -extern void fdc_disable(void); -extern int fdc_fifo_threshold(__u8 threshold, - int *fifo_state, int *lock_state, int *fifo_thr); -extern void fdc_wait_calibrate(void); -extern int fdc_sense_interrupt_status(int *st0, int *current_cylinder); -extern void fdc_save_drive_specs(void); -extern void fdc_restore_drive_specs(void); -extern int fdc_set_data_rate(int rate); -extern void fdc_set_write_precomp(int precomp); -extern int fdc_release_irq_and_dma(void); -extern void fdc_release_regions(void); -extern int fdc_init(void); -extern int fdc_setup_read_write(buffer_struct * buff, __u8 operation); -extern int fdc_setup_formatting(buffer_struct * buff); -#endif diff --git a/drivers/char/ftape/lowlevel/fdc-isr.c b/drivers/char/ftape/lowlevel/fdc-isr.c deleted file mode 100644 index ad2bc733ae1b..000000000000 --- a/drivers/char/ftape/lowlevel/fdc-isr.c +++ /dev/null @@ -1,1170 +0,0 @@ -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-isr.c,v $ - * $Revision: 1.9 $ - * $Date: 1997/10/17 23:01:53 $ - * - * This file contains the interrupt service routine and - * associated code for the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -#include -#include - -#define volatile /* */ - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-isr.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -volatile int ft_expected_stray_interrupts; -volatile int ft_interrupt_seen; -volatile int ft_seek_completed; -volatile int ft_hide_interrupt; -/* Local vars. - */ -typedef enum { - no_error = 0, id_am_error = 0x01, id_crc_error = 0x02, - data_am_error = 0x04, data_crc_error = 0x08, - no_data_error = 0x10, overrun_error = 0x20, -} error_cause; -static int stop_read_ahead; - - -static void print_error_cause(int cause) -{ - TRACE_FUN(ft_t_any); - - switch (cause) { - case no_data_error: - TRACE(ft_t_noise, "no data error"); - break; - case id_am_error: - TRACE(ft_t_noise, "id am error"); - break; - case id_crc_error: - TRACE(ft_t_noise, "id crc error"); - break; - case data_am_error: - TRACE(ft_t_noise, "data am error"); - break; - case data_crc_error: - TRACE(ft_t_noise, "data crc error"); - break; - case overrun_error: - TRACE(ft_t_noise, "overrun error"); - break; - default:; - } - TRACE_EXIT; -} - -static char *fdc_mode_txt(fdc_mode_enum mode) -{ - switch (mode) { - case fdc_idle: - return "fdc_idle"; - case fdc_reading_data: - return "fdc_reading_data"; - case fdc_seeking: - return "fdc_seeking"; - case fdc_writing_data: - return "fdc_writing_data"; - case fdc_reading_id: - return "fdc_reading_id"; - case fdc_recalibrating: - return "fdc_recalibrating"; - case fdc_formatting: - return "fdc_formatting"; - case fdc_verifying: - return "fdc_verifying"; - default: - return "unknown"; - } -} - -static inline error_cause decode_irq_cause(fdc_mode_enum mode, __u8 st[]) -{ - error_cause cause = no_error; - TRACE_FUN(ft_t_any); - - /* Valid st[], decode cause of interrupt. - */ - switch (st[0] & ST0_INT_MASK) { - case FDC_INT_NORMAL: - TRACE(ft_t_fdc_dma,"normal completion: %s",fdc_mode_txt(mode)); - break; - case FDC_INT_ABNORMAL: - TRACE(ft_t_flow, "abnormal completion %s", fdc_mode_txt(mode)); - TRACE(ft_t_fdc_dma, "ST0: 0x%02x, ST1: 0x%02x, ST2: 0x%02x", - st[0], st[1], st[2]); - TRACE(ft_t_fdc_dma, - "C: 0x%02x, H: 0x%02x, R: 0x%02x, N: 0x%02x", - st[3], st[4], st[5], st[6]); - if (st[1] & 0x01) { - if (st[2] & 0x01) { - cause = data_am_error; - } else { - cause = id_am_error; - } - } else if (st[1] & 0x20) { - if (st[2] & 0x20) { - cause = data_crc_error; - } else { - cause = id_crc_error; - } - } else if (st[1] & 0x04) { - cause = no_data_error; - } else if (st[1] & 0x10) { - cause = overrun_error; - } - print_error_cause(cause); - break; - case FDC_INT_INVALID: - TRACE(ft_t_flow, "invalid completion %s", fdc_mode_txt(mode)); - break; - case FDC_INT_READYCH: - if (st[0] & ST0_SEEK_END) { - TRACE(ft_t_flow, "drive poll completed"); - } else { - TRACE(ft_t_flow, "ready change %s",fdc_mode_txt(mode)); - } - break; - default: - break; - } - TRACE_EXIT cause; -} - -static void update_history(error_cause cause) -{ - switch (cause) { - case id_am_error: - ft_history.id_am_errors++; - break; - case id_crc_error: - ft_history.id_crc_errors++; - break; - case data_am_error: - ft_history.data_am_errors++; - break; - case data_crc_error: - ft_history.data_crc_errors++; - break; - case overrun_error: - ft_history.overrun_errors++; - break; - case no_data_error: - ft_history.no_data_errors++; - break; - default:; - } -} - -static void skip_bad_sector(buffer_struct * buff) -{ - TRACE_FUN(ft_t_any); - - /* Mark sector as soft error and skip it - */ - if (buff->remaining > 0) { - ++buff->sector_offset; - ++buff->data_offset; - --buff->remaining; - buff->ptr += FT_SECTOR_SIZE; - buff->bad_sector_map >>= 1; - } else { - /* Hey, what is this????????????? C code: if we shift - * more than 31 bits, we get no shift. That's bad!!!!!! - */ - ++buff->sector_offset; /* hack for error maps */ - TRACE(ft_t_warn, "skipping last sector in segment"); - } - TRACE_EXIT; -} - -static void update_error_maps(buffer_struct * buff, unsigned int error_offset) -{ - int hard = 0; - TRACE_FUN(ft_t_any); - - if (buff->retry < FT_SOFT_RETRIES) { - buff->soft_error_map |= (1 << error_offset); - } else { - buff->hard_error_map |= (1 << error_offset); - buff->soft_error_map &= ~buff->hard_error_map; - buff->retry = -1; /* will be set to 0 in setup_segment */ - hard = 1; - } - TRACE(ft_t_noise, "sector %d : %s error\n" - KERN_INFO "hard map: 0x%08lx\n" - KERN_INFO "soft map: 0x%08lx", - FT_SECTOR(error_offset), hard ? "hard" : "soft", - (long) buff->hard_error_map, (long) buff->soft_error_map); - TRACE_EXIT; -} - -static void print_progress(buffer_struct *buff, error_cause cause) -{ - TRACE_FUN(ft_t_any); - - switch (cause) { - case no_error: - TRACE(ft_t_flow,"%d Sector(s) transferred", buff->sector_count); - break; - case no_data_error: - TRACE(ft_t_flow, "Sector %d not found", - FT_SECTOR(buff->sector_offset)); - break; - case overrun_error: - /* got an overrun error on the first byte, must be a - * hardware problem - */ - TRACE(ft_t_bug, - "Unexpected error: failing DMA or FDC controller ?"); - break; - case data_crc_error: - TRACE(ft_t_flow, "Error in sector %d", - FT_SECTOR(buff->sector_offset - 1)); - break; - case id_crc_error: - case id_am_error: - case data_am_error: - TRACE(ft_t_flow, "Error in sector %d", - FT_SECTOR(buff->sector_offset)); - break; - default: - TRACE(ft_t_flow, "Unexpected error at sector %d", - FT_SECTOR(buff->sector_offset)); - break; - } - TRACE_EXIT; -} - -/* - * Error cause: Amount xferred: Action: - * - * id_am_error 0 mark bad and skip - * id_crc_error 0 mark bad and skip - * data_am_error 0 mark bad and skip - * data_crc_error % 1024 mark bad and skip - * no_data_error 0 retry on write - * mark bad and skip on read - * overrun_error [ 0..all-1 ] mark bad and skip - * no_error all continue - */ - -/* the arg `sector' is returned by the fdc and tells us at which sector we - * are positioned at (relative to starting sector of segment) - */ -static void determine_verify_progress(buffer_struct *buff, - error_cause cause, - __u8 sector) -{ - TRACE_FUN(ft_t_any); - - if (cause == no_error && sector == 1) { - buff->sector_offset = FT_SECTORS_PER_SEGMENT; - buff->remaining = 0; - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - } else { - buff->sector_offset = sector - buff->sect; - buff->remaining = FT_SECTORS_PER_SEGMENT - buff->sector_offset; - TRACE(ft_t_noise, "%ssector offset: 0x%04x", - (cause == no_error) ? "unexpected " : "", - buff->sector_offset); - switch (cause) { - case overrun_error: - break; -#if 0 - case no_data_error: - buff->retry = FT_SOFT_RETRIES; - if (buff->hard_error_map && - buff->sector_offset > 1 && - (buff->hard_error_map & - (1 << (buff->sector_offset-2)))) { - buff->retry --; - } - break; -#endif - default: - buff->retry = FT_SOFT_RETRIES; - break; - } - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - /* Sector_offset points to the problem area Now adjust - * sector_offset so it always points one past he failing - * sector. I.e. skip the bad sector. - */ - ++buff->sector_offset; - --buff->remaining; - update_error_maps(buff, buff->sector_offset - 1); - } - TRACE_EXIT; -} - -static void determine_progress(buffer_struct *buff, - error_cause cause, - __u8 sector) -{ - unsigned int dma_residue; - TRACE_FUN(ft_t_any); - - /* Using less preferred order of disable_dma and - * get_dma_residue because this seems to fail on at least one - * system if reversed! - */ - dma_residue = get_dma_residue(fdc.dma); - disable_dma(fdc.dma); - if (cause != no_error || dma_residue != 0) { - TRACE(ft_t_noise, "%sDMA residue: 0x%04x", - (cause == no_error) ? "unexpected " : "", - dma_residue); - /* adjust to actual value: */ - if (dma_residue == 0) { - /* this happens sometimes with overrun errors. - * I don't know whether we could ignore the - * overrun error. Play save. - */ - buff->sector_count --; - } else { - buff->sector_count -= ((dma_residue + - (FT_SECTOR_SIZE - 1)) / - FT_SECTOR_SIZE); - } - } - /* Update var's influenced by the DMA operation. - */ - if (buff->sector_count > 0) { - buff->sector_offset += buff->sector_count; - buff->data_offset += buff->sector_count; - buff->ptr += (buff->sector_count * - FT_SECTOR_SIZE); - buff->remaining -= buff->sector_count; - buff->bad_sector_map >>= buff->sector_count; - } - if (TRACE_LEVEL >= ft_t_flow) { - print_progress(buff, cause); - } - if (cause != no_error) { - if (buff->remaining == 0) { - TRACE(ft_t_warn, "foo?\n" - KERN_INFO "count : %d\n" - KERN_INFO "offset: %d\n" - KERN_INFO "soft : %08x\n" - KERN_INFO "hard : %08x", - buff->sector_count, - buff->sector_offset, - buff->soft_error_map, - buff->hard_error_map); - } - /* Sector_offset points to the problem area, except if we got - * a data_crc_error. In that case it points one past the - * failing sector. - * - * Now adjust sector_offset so it always points one past he - * failing sector. I.e. skip the bad sector. - */ - if (cause != data_crc_error) { - skip_bad_sector(buff); - } - update_error_maps(buff, buff->sector_offset - 1); - } - TRACE_EXIT; -} - -static int calc_steps(int cmd) -{ - if (ftape_current_cylinder > cmd) { - return ftape_current_cylinder - cmd; - } else { - return ftape_current_cylinder + cmd; - } -} - -static void pause_tape(int retry, int mode) -{ - int result; - __u8 out[3] = {FDC_SEEK, ft_drive_sel, 0}; - TRACE_FUN(ft_t_any); - - /* We'll use a raw seek command to get the tape to rewind and - * stop for a retry. - */ - ++ft_history.rewinds; - if (qic117_cmds[ftape_current_command].non_intr) { - TRACE(ft_t_warn, "motion command may be issued too soon"); - } - if (retry && (mode == fdc_reading_data || - mode == fdc_reading_id || - mode == fdc_verifying)) { - ftape_current_command = QIC_MICRO_STEP_PAUSE; - ftape_might_be_off_track = 1; - } else { - ftape_current_command = QIC_PAUSE; - } - out[2] = calc_steps(ftape_current_command); - result = fdc_command(out, 3); /* issue QIC_117 command */ - ftape_current_cylinder = out[ 2]; - if (result < 0) { - TRACE(ft_t_noise, "qic-pause failed, status = %d", result); - } else { - ft_location.known = 0; - ft_runner_status = idle; - ft_hide_interrupt = 1; - ftape_tape_running = 0; - } - TRACE_EXIT; -} - -static void continue_xfer(buffer_struct *buff, - fdc_mode_enum mode, - unsigned int skip) -{ - int write = 0; - TRACE_FUN(ft_t_any); - - if (mode == fdc_writing_data || mode == fdc_deleting) { - write = 1; - } - /* This part can be removed if it never happens - */ - if (skip > 0 && - (ft_runner_status != running || - (write && (buff->status != writing)) || - (!write && (buff->status != reading && - buff->status != verifying)))) { - TRACE(ft_t_err, "unexpected runner/buffer state %d/%d", - ft_runner_status, buff->status); - buff->status = error; - /* finish this buffer: */ - (void)ftape_next_buffer(ft_queue_head); - ft_runner_status = aborting; - fdc_mode = fdc_idle; - } else if (buff->remaining > 0 && ftape_calc_next_cluster(buff) > 0) { - /* still sectors left in current segment, continue - * with this segment - */ - if (fdc_setup_read_write(buff, mode) < 0) { - /* failed, abort operation - */ - buff->bytes = buff->ptr - buff->address; - buff->status = error; - /* finish this buffer: */ - (void)ftape_next_buffer(ft_queue_head); - ft_runner_status = aborting; - fdc_mode = fdc_idle; - } - } else { - /* current segment completed - */ - unsigned int last_segment = buff->segment_id; - int eot = ((last_segment + 1) % ft_segments_per_track) == 0; - unsigned int next = buff->next_segment; /* 0 means stop ! */ - - buff->bytes = buff->ptr - buff->address; - buff->status = done; - buff = ftape_next_buffer(ft_queue_head); - if (eot) { - /* finished last segment on current track, - * can't continue - */ - ft_runner_status = logical_eot; - fdc_mode = fdc_idle; - TRACE_EXIT; - } - if (next <= 0) { - /* don't continue with next segment - */ - TRACE(ft_t_noise, "no %s allowed, stopping tape", - (write) ? "write next" : "read ahead"); - pause_tape(0, mode); - ft_runner_status = idle; /* not quite true until - * next irq - */ - TRACE_EXIT; - } - /* continue with next segment - */ - if (buff->status != waiting) { - TRACE(ft_t_noise, "all input buffers %s, pausing tape", - (write) ? "empty" : "full"); - pause_tape(0, mode); - ft_runner_status = idle; /* not quite true until - * next irq - */ - TRACE_EXIT; - } - if (write && next != buff->segment_id) { - TRACE(ft_t_noise, - "segments out of order, aborting write"); - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - TRACE_EXIT; - } - ftape_setup_new_segment(buff, next, 0); - if (stop_read_ahead) { - buff->next_segment = 0; - stop_read_ahead = 0; - } - if (ftape_calc_next_cluster(buff) == 0 || - fdc_setup_read_write(buff, mode) != 0) { - TRACE(ft_t_err, "couldn't start %s-ahead", - write ? "write" : "read"); - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - } else { - /* keep on going */ - switch (ft_driver_state) { - case reading: buff->status = reading; break; - case verifying: buff->status = verifying; break; - case writing: buff->status = writing; break; - case deleting: buff->status = deleting; break; - default: - TRACE(ft_t_err, - "BUG: ft_driver_state %d should be one out of " - "{reading, writing, verifying, deleting}", - ft_driver_state); - buff->status = write ? writing : reading; - break; - } - } - } - TRACE_EXIT; -} - -static void retry_sector(buffer_struct *buff, - int mode, - unsigned int skip) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_noise, "%s error, will retry", - (mode == fdc_writing_data || mode == fdc_deleting) ? "write" : "read"); - pause_tape(1, mode); - ft_runner_status = aborting; - buff->status = error; - buff->skip = skip; - TRACE_EXIT; -} - -static unsigned int find_resume_point(buffer_struct *buff) -{ - int i = 0; - SectorMap mask; - SectorMap map; - TRACE_FUN(ft_t_any); - - /* This function is to be called after all variables have been - * updated to point past the failing sector. - * If there are any soft errors before the failing sector, - * find the first soft error and return the sector offset. - * Otherwise find the last hard error. - * Note: there should always be at least one hard or soft error ! - */ - if (buff->sector_offset < 1 || buff->sector_offset > 32) { - TRACE(ft_t_bug, "BUG: sector_offset = %d", - buff->sector_offset); - TRACE_EXIT 0; - } - if (buff->sector_offset >= 32) { /* C-limitation on shift ! */ - mask = 0xffffffff; - } else { - mask = (1 << buff->sector_offset) - 1; - } - map = buff->soft_error_map & mask; - if (map) { - while ((map & (1 << i)) == 0) { - ++i; - } - TRACE(ft_t_noise, "at sector %d", FT_SECTOR(i)); - } else { - map = buff->hard_error_map & mask; - i = buff->sector_offset - 1; - if (map) { - while ((map & (1 << i)) == 0) { - --i; - } - TRACE(ft_t_noise, "after sector %d", FT_SECTOR(i)); - ++i; /* first sector after last hard error */ - } else { - TRACE(ft_t_bug, "BUG: no soft or hard errors"); - } - } - TRACE_EXIT i; -} - -/* check possible dma residue when formatting, update position record in - * buffer struct. This is, of course, modelled after determine_progress(), but - * we don't need to set up for retries because the format process cannot be - * interrupted (except at the end of the tape track). - */ -static int determine_fmt_progress(buffer_struct *buff, error_cause cause) -{ - unsigned int dma_residue; - TRACE_FUN(ft_t_any); - - /* Using less preferred order of disable_dma and - * get_dma_residue because this seems to fail on at least one - * system if reversed! - */ - dma_residue = get_dma_residue(fdc.dma); - disable_dma(fdc.dma); - if (cause != no_error || dma_residue != 0) { - TRACE(ft_t_info, "DMA residue = 0x%04x", dma_residue); - fdc_mode = fdc_idle; - switch(cause) { - case no_error: - ft_runner_status = aborting; - buff->status = idle; - break; - case overrun_error: - /* got an overrun error on the first byte, must be a - * hardware problem - */ - TRACE(ft_t_bug, - "Unexpected error: failing DMA controller ?"); - ft_runner_status = do_abort; - buff->status = error; - break; - default: - TRACE(ft_t_noise, "Unexpected error at segment %d", - buff->segment_id); - ft_runner_status = do_abort; - buff->status = error; - break; - } - TRACE_EXIT -EIO; /* can only retry entire track in format mode - */ - } - /* Update var's influenced by the DMA operation. - */ - buff->ptr += FT_SECTORS_PER_SEGMENT * 4; - buff->bytes -= FT_SECTORS_PER_SEGMENT * 4; - buff->remaining -= FT_SECTORS_PER_SEGMENT; - buff->segment_id ++; /* done with segment */ - TRACE_EXIT 0; -} - -/* - * Continue formatting, switch buffers if there is no data left in - * current buffer. This is, of course, modelled after - * continue_xfer(), but we don't need to set up for retries because - * the format process cannot be interrupted (except at the end of the - * tape track). - */ -static void continue_formatting(buffer_struct *buff) -{ - TRACE_FUN(ft_t_any); - - if (buff->remaining <= 0) { /* no space left in dma buffer */ - unsigned int next = buff->next_segment; - - if (next == 0) { /* end of tape track */ - buff->status = done; - ft_runner_status = logical_eot; - fdc_mode = fdc_idle; - TRACE(ft_t_noise, "Done formatting track %d", - ft_location.track); - TRACE_EXIT; - } - /* - * switch to next buffer! - */ - buff->status = done; - buff = ftape_next_buffer(ft_queue_head); - - if (buff->status != waiting || next != buff->segment_id) { - goto format_setup_error; - } - } - if (fdc_setup_formatting(buff) < 0) { - goto format_setup_error; - } - buff->status = formatting; - TRACE(ft_t_fdc_dma, "Formatting segment %d on track %d", - buff->segment_id, ft_location.track); - TRACE_EXIT; - format_setup_error: - ft_runner_status = do_abort; - fdc_mode = fdc_idle; - buff->status = error; - TRACE(ft_t_err, "Error setting up for segment %d on track %d", - buff->segment_id, ft_location.track); - TRACE_EXIT; - -} - -/* this handles writing, read id, reading and formatting - */ -static void handle_fdc_busy(buffer_struct *buff) -{ - static int no_data_error_count; - int retry = 0; - error_cause cause; - __u8 in[7]; - int skip; - fdc_mode_enum fmode = fdc_mode; - TRACE_FUN(ft_t_any); - - if (fdc_result(in, 7) < 0) { /* better get it fast ! */ - TRACE(ft_t_err, - "Probably fatal error during FDC Result Phase\n" - KERN_INFO - "drive may hang until (power on) reset :-("); - /* what to do next ???? - */ - TRACE_EXIT; - } - cause = decode_irq_cause(fdc_mode, in); -#ifdef TESTING - { int i; - for (i = 0; i < (int)ft_nr_buffers; ++i) - TRACE(ft_t_any, "buffer[%d] status: %d, segment_id: %d", - i, ft_buffer[i]->status, ft_buffer[i]->segment_id); - } -#endif - if (fmode == fdc_reading_data && ft_driver_state == verifying) { - fmode = fdc_verifying; - } - switch (fmode) { - case fdc_verifying: - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_noise,"aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - switch (cause) { - case no_error: - no_data_error_count = 0; - determine_verify_progress(buff, cause, in[5]); - if (in[2] & 0x40) { - /* This should not happen when verifying - */ - TRACE(ft_t_warn, - "deleted data in segment %d/%d", - buff->segment_id, - FT_SECTOR(buff->sector_offset - 1)); - buff->remaining = 0; /* abort transfer */ - buff->hard_error_map = EMPTY_SEGMENT; - skip = 1; - } else { - skip = 0; - } - continue_xfer(buff, fdc_mode, skip); - break; - case no_data_error: - no_data_error_count ++; - case overrun_error: - retry ++; - case id_am_error: - case id_crc_error: - case data_am_error: - case data_crc_error: - determine_verify_progress(buff, cause, in[5]); - if (cause == no_data_error) { - if (no_data_error_count >= 2) { - TRACE(ft_t_warn, - "retrying because of successive " - "no data errors"); - no_data_error_count = 0; - } else { - retry --; - } - } else { - no_data_error_count = 0; - } - if (retry) { - skip = find_resume_point(buff); - } else { - skip = buff->sector_offset; - } - if (retry && skip < 32) { - retry_sector(buff, fdc_mode, skip); - } else { - continue_xfer(buff, fdc_mode, skip); - } - update_history(cause); - break; - default: - /* Don't know why this could happen - * but find out. - */ - determine_verify_progress(buff, cause, in[5]); - retry_sector(buff, fdc_mode, 0); - TRACE(ft_t_err, "Error: unexpected error"); - break; - } - break; - case fdc_reading_data: -#ifdef TESTING - /* I'm sorry, but: NOBODY ever used this trace - * messages for ages. I guess that Bas was the last person - * that ever really used this (thank you, between the lines) - */ - if (cause == no_error) { - TRACE(ft_t_flow,"reading segment %d",buff->segment_id); - } else { - TRACE(ft_t_noise, "error reading segment %d", - buff->segment_id); - TRACE(ft_t_noise, "\n" - KERN_INFO - "IRQ:C: 0x%02x, H: 0x%02x, R: 0x%02x, N: 0x%02x\n" - KERN_INFO - "BUF:C: 0x%02x, H: 0x%02x, R: 0x%02x", - in[3], in[4], in[5], in[6], - buff->cyl, buff->head, buff->sect); - } -#endif - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_noise,"aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->bad_sector_map == FAKE_SEGMENT) { - /* This condition occurs when reading a `fake' - * sector that's not accessible. Doesn't - * really matter as we would have ignored it - * anyway ! - * - * Chance is that we're past the next segment - * now, so the next operation may fail and - * result in a retry. - */ - buff->remaining = 0; /* skip failing sector */ - /* buff->ptr = buff->address; */ - /* fake success: */ - continue_xfer(buff, fdc_mode, 1); - /* trace calls are expensive: place them AFTER - * the real stuff has been done. - * - */ - TRACE(ft_t_noise, "skipping empty segment %d (read), size? %d", - buff->segment_id, buff->ptr - buff->address); - TRACE_EXIT; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - switch (cause) { - case no_error: - determine_progress(buff, cause, in[5]); - if (in[2] & 0x40) { - /* Handle deleted data in header segments. - * Skip segment and force read-ahead. - */ - TRACE(ft_t_warn, - "deleted data in segment %d/%d", - buff->segment_id, - FT_SECTOR(buff->sector_offset - 1)); - buff->deleted = 1; - buff->remaining = 0;/*abort transfer */ - buff->soft_error_map |= - (-1L << buff->sector_offset); - if (buff->segment_id == 0) { - /* stop on next segment */ - stop_read_ahead = 1; - } - /* force read-ahead: */ - buff->next_segment = - buff->segment_id + 1; - skip = (FT_SECTORS_PER_SEGMENT - - buff->sector_offset); - } else { - skip = 0; - } - continue_xfer(buff, fdc_mode, skip); - break; - case no_data_error: - /* Tape started too far ahead of or behind the - * right sector. This may also happen in the - * middle of a segment ! - * - * Handle no-data as soft error. If next - * sector fails too, a retry (with needed - * reposition) will follow. - */ - retry ++; - case id_am_error: - case id_crc_error: - case data_am_error: - case data_crc_error: - case overrun_error: - retry += (buff->soft_error_map != 0 || - buff->hard_error_map != 0); - determine_progress(buff, cause, in[5]); -#if 1 || defined(TESTING) - if (cause == overrun_error) retry ++; -#endif - if (retry) { - skip = find_resume_point(buff); - } else { - skip = buff->sector_offset; - } - /* Try to resume with next sector on single - * errors (let ecc correct it), but retry on - * no_data (we'll be past the target when we - * get here so we cannot retry) or on - * multiple errors (reduce chance on ecc - * failure). - */ - /* cH: 23/02/97: if the last sector in the - * segment was a hard error, then there is - * no sense in a retry. This occasion seldom - * occurs but ... @:³²¸`@%&§$ - */ - if (retry && skip < 32) { - retry_sector(buff, fdc_mode, skip); - } else { - continue_xfer(buff, fdc_mode, skip); - } - update_history(cause); - break; - default: - /* Don't know why this could happen - * but find out. - */ - determine_progress(buff, cause, in[5]); - retry_sector(buff, fdc_mode, 0); - TRACE(ft_t_err, "Error: unexpected error"); - break; - } - break; - case fdc_reading_id: - if (cause == no_error) { - fdc_cyl = in[3]; - fdc_head = in[4]; - fdc_sect = in[5]; - TRACE(ft_t_fdc_dma, - "id read: C: 0x%02x, H: 0x%02x, R: 0x%02x", - fdc_cyl, fdc_head, fdc_sect); - } else { /* no valid information, use invalid sector */ - fdc_cyl = fdc_head = fdc_sect = 0; - TRACE(ft_t_flow, "Didn't find valid sector Id"); - } - fdc_mode = fdc_idle; - break; - case fdc_deleting: - case fdc_writing_data: -#ifdef TESTING - if (cause == no_error) { - TRACE(ft_t_flow, "writing segment %d", buff->segment_id); - } else { - TRACE(ft_t_noise, "error writing segment %d", - buff->segment_id); - } -#endif - if (ft_runner_status == aborting || - ft_runner_status == do_abort) { - TRACE(ft_t_flow, "aborting %s",fdc_mode_txt(fdc_mode)); - break; - } - if (buff->retry > 0) { - TRACE(ft_t_flow, "this is retry nr %d", buff->retry); - } - if (buff->bad_sector_map == FAKE_SEGMENT) { - /* This condition occurs when trying to write to a - * `fake' sector that's not accessible. Doesn't really - * matter as it isn't used anyway ! Might be located - * at wrong segment, then we'll fail on the next - * segment. - */ - TRACE(ft_t_noise, "skipping empty segment (write)"); - buff->remaining = 0; /* skip failing sector */ - /* fake success: */ - continue_xfer(buff, fdc_mode, 1); - break; - } - switch (cause) { - case no_error: - determine_progress(buff, cause, in[5]); - continue_xfer(buff, fdc_mode, 0); - break; - case no_data_error: - case id_am_error: - case id_crc_error: - case data_am_error: - case overrun_error: - update_history(cause); - determine_progress(buff, cause, in[5]); - skip = find_resume_point(buff); - retry_sector(buff, fdc_mode, skip); - break; - default: - if (in[1] & 0x02) { - TRACE(ft_t_err, "media not writable"); - } else { - TRACE(ft_t_bug, "unforeseen write error"); - } - fdc_mode = fdc_idle; - break; - } - break; /* fdc_deleting || fdc_writing_data */ - case fdc_formatting: - /* The interrupt comes after formatting a segment. We then - * have to set up QUICKLY for the next segment. But - * afterwards, there is plenty of time. - */ - switch (cause) { - case no_error: - /* would like to keep most of the formatting stuff - * outside the isr code, but timing is too critical - */ - if (determine_fmt_progress(buff, cause) >= 0) { - continue_formatting(buff); - } - break; - case no_data_error: - case id_am_error: - case id_crc_error: - case data_am_error: - case overrun_error: - default: - determine_fmt_progress(buff, cause); - update_history(cause); - if (in[1] & 0x02) { - TRACE(ft_t_err, "media not writable"); - } else { - TRACE(ft_t_bug, "unforeseen write error"); - } - break; - } /* cause */ - break; - default: - TRACE(ft_t_warn, "Warning: unexpected irq during: %s", - fdc_mode_txt(fdc_mode)); - fdc_mode = fdc_idle; - break; - } - TRACE_EXIT; -} - -/* FDC interrupt service routine. - */ -void fdc_isr(void) -{ - static int isr_active; -#ifdef TESTING - unsigned int t0 = ftape_timestamp(); -#endif - TRACE_FUN(ft_t_any); - - if (isr_active++) { - --isr_active; - TRACE(ft_t_bug, "BUG: nested interrupt, not good !"); - *fdc.hook = fdc_isr; /* hook our handler into the fdc - * code again - */ - TRACE_EXIT; - } - sti(); - if (inb_p(fdc.msr) & FDC_BUSY) { /* Entering Result Phase */ - ft_hide_interrupt = 0; - handle_fdc_busy(ftape_get_buffer(ft_queue_head)); - if (ft_runner_status == do_abort) { - /* cease operation, remember tape position - */ - TRACE(ft_t_flow, "runner aborting"); - ft_runner_status = aborting; - ++ft_expected_stray_interrupts; - } - } else { /* !FDC_BUSY */ - /* clear interrupt, cause should be gotten by issuing - * a Sense Interrupt Status command. - */ - if (fdc_mode == fdc_recalibrating || fdc_mode == fdc_seeking) { - if (ft_hide_interrupt) { - int st0; - int pcn; - - if (fdc_sense_interrupt_status(&st0, &pcn) < 0) - TRACE(ft_t_err, - "sense interrupt status failed"); - ftape_current_cylinder = pcn; - TRACE(ft_t_flow, "handled hidden interrupt"); - } - ft_seek_completed = 1; - fdc_mode = fdc_idle; - } else if (!waitqueue_active(&ftape_wait_intr)) { - if (ft_expected_stray_interrupts == 0) { - TRACE(ft_t_warn, "unexpected stray interrupt"); - } else { - TRACE(ft_t_flow, "expected stray interrupt"); - --ft_expected_stray_interrupts; - } - } else { - if (fdc_mode == fdc_reading_data || - fdc_mode == fdc_verifying || - fdc_mode == fdc_writing_data || - fdc_mode == fdc_deleting || - fdc_mode == fdc_formatting || - fdc_mode == fdc_reading_id) { - if (inb_p(fdc.msr) & FDC_BUSY) { - TRACE(ft_t_bug, - "***** FDC failure, busy too late"); - } else { - TRACE(ft_t_bug, - "***** FDC failure, no busy"); - } - } else { - TRACE(ft_t_fdc_dma, "awaited stray interrupt"); - } - } - ft_hide_interrupt = 0; - } - /* Handle sleep code. - */ - if (!ft_hide_interrupt) { - ft_interrupt_seen ++; - if (waitqueue_active(&ftape_wait_intr)) { - wake_up_interruptible(&ftape_wait_intr); - } - } else { - TRACE(ft_t_flow, "hiding interrupt while %s", - waitqueue_active(&ftape_wait_intr) ? "waiting":"active"); - } -#ifdef TESTING - t0 = ftape_timediff(t0, ftape_timestamp()); - if (t0 >= 1000) { - /* only tell us about long calls */ - TRACE(ft_t_noise, "isr() duration: %5d usec", t0); - } -#endif - *fdc.hook = fdc_isr; /* hook our handler into the fdc code again */ - --isr_active; - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/fdc-isr.h b/drivers/char/ftape/lowlevel/fdc-isr.h deleted file mode 100644 index 065aa978942d..000000000000 --- a/drivers/char/ftape/lowlevel/fdc-isr.h +++ /dev/null @@ -1,55 +0,0 @@ -#ifndef _FDC_ISR_H -#define _FDC_ISR_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/fdc-isr.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:07 $ - * - * This file declares the global variables necessary to - * synchronize the interrupt service routine (isr) with the - * remainder of the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -/* - * fdc-isr.c defined public variables - */ -extern volatile int ft_expected_stray_interrupts; /* masks stray interrupts */ -extern volatile int ft_seek_completed; /* flag set by isr */ -extern volatile int ft_interrupt_seen; /* flag set by isr */ -extern volatile int ft_hide_interrupt; /* flag set by isr */ - -/* - * fdc-io.c defined public functions - */ -extern void fdc_isr(void); - -/* - * A kernel hook that steals one interrupt from the floppy - * driver (Should be fixed when the new fdc driver gets ready) - * See the linux kernel source files: - * drivers/block/floppy.c & drivers/block/blk.h - * for the details. - */ -extern void (*do_floppy) (void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-bsm.c b/drivers/char/ftape/lowlevel/ftape-bsm.c deleted file mode 100644 index d1a301cc344f..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-bsm.c +++ /dev/null @@ -1,491 +0,0 @@ -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-bsm.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:15:15 $ - * - * This file contains the bad-sector map handling code for - * the QIC-117 floppy tape driver for Linux. - * QIC-40, QIC-80, QIC-3010 and QIC-3020 maps are implemented. - */ - -#include - -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" - -/* Global vars. - */ - -/* Local vars. - */ -static __u8 *bad_sector_map; -static SectorCount *bsm_hash_ptr; - -typedef enum { - forward, backward -} mode_type; - -#if 0 -static void ftape_put_bad_sector_entry(int segment_id, SectorMap new_map); -#endif - -#if 0 -/* fix_tape converts a normal QIC-80 tape into a 'wide' tape. - * For testing purposes only ! - */ -void fix_tape(__u8 * buffer, ft_format_type new_code) -{ - static __u8 list[BAD_SECTOR_MAP_SIZE]; - SectorMap *src_ptr = (SectorMap *) list; - __u8 *dst_ptr = bad_sector_map; - SectorMap map; - unsigned int sector = 1; - int i; - - if (format_code != fmt_var && format_code != fmt_big) { - memcpy(list, bad_sector_map, sizeof(list)); - memset(bad_sector_map, 0, sizeof(bad_sector_map)); - while ((__u8 *) src_ptr - list < sizeof(list)) { - map = *src_ptr++; - if (map == EMPTY_SEGMENT) { - *(SectorMap *) dst_ptr = 0x800000 + sector; - dst_ptr += 3; - sector += SECTORS_PER_SEGMENT; - } else { - for (i = 0; i < SECTORS_PER_SEGMENT; ++i) { - if (map & 1) { - *(SewctorMap *) dst_ptr = sector; - dst_ptr += 3; - } - map >>= 1; - ++sector; - } - } - } - } - bad_sector_map_changed = 1; - *(buffer + 4) = new_code; /* put new format code */ - if (format_code != fmt_var && new_code == fmt_big) { - PUT4(buffer, FT_6_HSEG_1, (__u32)GET2(buffer, 6)); - PUT4(buffer, FT_6_HSEG_2, (__u32)GET2(buffer, 8)); - PUT4(buffer, FT_6_FRST_SEG, (__u32)GET2(buffer, 10)); - PUT4(buffer, FT_6_LAST_SEG, (__u32)GET2(buffer, 12)); - memset(buffer+6, '\0', 8); - } - format_code = new_code; -} - -#endif - -/* given buffer that contains a header segment, find the end of - * of the bsm list - */ -__u8 * ftape_find_end_of_bsm_list(__u8 * address) -{ - __u8 *ptr = address + FT_HEADER_END; /* start of bsm list */ - __u8 *limit = address + FT_SEGMENT_SIZE; - while (ptr + 2 < limit) { - if (ptr[0] || ptr[1] || ptr[2]) { - ptr += 3; - } else { - return ptr; - } - } - return NULL; -} - -static inline void put_sector(SectorCount *ptr, unsigned int sector) -{ - ptr->bytes[0] = sector & 0xff; - sector >>= 8; - ptr->bytes[1] = sector & 0xff; - sector >>= 8; - ptr->bytes[2] = sector & 0xff; -} - -static inline unsigned int get_sector(SectorCount *ptr) -{ -#if 1 - unsigned int sector; - - sector = ptr->bytes[0]; - sector += ptr->bytes[1] << 8; - sector += ptr->bytes[2] << 16; - - return sector; -#else - /* GET4 gets the next four bytes in Intel little endian order - * and converts them to host byte order and handles unaligned - * access. - */ - return (GET4(ptr, 0) & 0x00ffffff); /* back to host byte order */ -#endif -} - -static void bsm_debug_fake(void) -{ - /* for testing of bad sector handling at end of tape - */ -#if 0 - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 3, - 0x000003e0; - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 2, - 0xff3fffff; - ftape_put_bad_sector_entry(segments_per_track * tracks_per_tape - 1, - 0xffffe000; -#endif - /* Enable to test bad sector handling - */ -#if 0 - ftape_put_bad_sector_entry(30, 0xfffffffe) - ftape_put_bad_sector_entry(32, 0x7fffffff); - ftape_put_bad_sector_entry(34, 0xfffeffff); - ftape_put_bad_sector_entry(36, 0x55555555); - ftape_put_bad_sector_entry(38, 0xffffffff); - ftape_put_bad_sector_entry(50, 0xffff0000); - ftape_put_bad_sector_entry(51, 0xffffffff); - ftape_put_bad_sector_entry(52, 0xffffffff); - ftape_put_bad_sector_entry(53, 0x0000ffff); -#endif - /* Enable when testing multiple volume tar dumps. - */ -#if 0 - { - int i; - - for (i = ft_first_data_segment; - i <= ft_last_data_segment - 7; ++i) { - ftape_put_bad_sector_entry(i, EMPTY_SEGMENT); - } - } -#endif - /* Enable when testing bit positions in *_error_map - */ -#if 0 - { - int i; - - for (i = first_data_segment; i <= last_data_segment; ++i) { - ftape_put_bad_sector_entry(i, - ftape_get_bad_sector_entry(i) - | 0x00ff00ff); - } - } -#endif -} - -static void print_bad_sector_map(void) -{ - unsigned int good_sectors; - unsigned int total_bad = 0; - int i; - TRACE_FUN(ft_t_flow); - - if (ft_format_code == fmt_big || - ft_format_code == fmt_var || - ft_format_code == fmt_1100ft) { - SectorCount *ptr = (SectorCount *)bad_sector_map; - unsigned int sector; - __u16 *ptr16; - - while((sector = get_sector(ptr++)) != 0) { - if ((ft_format_code == fmt_big || - ft_format_code == fmt_var) && - sector & 0x800000) { - total_bad += FT_SECTORS_PER_SEGMENT - 3; - TRACE(ft_t_noise, "bad segment at sector: %6d", - sector & 0x7fffff); - } else { - ++total_bad; - TRACE(ft_t_noise, "bad sector: %6d", sector); - } - } - /* Display old ftape's end-of-file marks - */ - ptr16 = (__u16*)ptr; - while ((sector = get_unaligned(ptr16++)) != 0) { - TRACE(ft_t_noise, "Old ftape eof mark: %4d/%2d", - sector, get_unaligned(ptr16++)); - } - } else { /* fixed size format */ - for (i = ft_first_data_segment; - i < (int)(ft_segments_per_track * ft_tracks_per_tape); ++i) { - SectorMap map = ((SectorMap *) bad_sector_map)[i]; - - if (map) { - TRACE(ft_t_noise, - "bsm for segment %4d: 0x%08x", i, (unsigned int)map); - total_bad += ((map == EMPTY_SEGMENT) - ? FT_SECTORS_PER_SEGMENT - 3 - : count_ones(map)); - } - } - } - good_sectors = - ((ft_segments_per_track * ft_tracks_per_tape - ft_first_data_segment) - * (FT_SECTORS_PER_SEGMENT - 3)) - total_bad; - TRACE(ft_t_info, "%d Kb usable on this tape", good_sectors); - if (total_bad == 0) { - TRACE(ft_t_info, - "WARNING: this tape has no bad blocks registered !"); - } else { - TRACE(ft_t_info, "%d bad sectors", total_bad); - } - TRACE_EXIT; -} - - -void ftape_extract_bad_sector_map(__u8 * buffer) -{ - TRACE_FUN(ft_t_any); - - /* Fill the bad sector map with the contents of buffer. - */ - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - /* QIC-3010/3020 and wide QIC-80 tapes no longer have a failed - * sector log but use this area to extend the bad sector map. - */ - bad_sector_map = &buffer[FT_HEADER_END]; - } else { - /* non-wide QIC-80 tapes have a failed sector log area that - * mustn't be included in the bad sector map. - */ - bad_sector_map = &buffer[FT_FSL + FT_FSL_SIZE]; - } - if (ft_format_code == fmt_1100ft || - ft_format_code == fmt_var || - ft_format_code == fmt_big) { - bsm_hash_ptr = (SectorCount *)bad_sector_map; - } else { - bsm_hash_ptr = NULL; - } - bsm_debug_fake(); - if (TRACE_LEVEL >= ft_t_info) { - print_bad_sector_map(); - } - TRACE_EXIT; -} - -static inline SectorMap cvt2map(unsigned int sector) -{ - return 1 << (((sector & 0x7fffff) - 1) % FT_SECTORS_PER_SEGMENT); -} - -static inline int cvt2segment(unsigned int sector) -{ - return ((sector & 0x7fffff) - 1) / FT_SECTORS_PER_SEGMENT; -} - -static int forward_seek_entry(int segment_id, - SectorCount **ptr, - SectorMap *map) -{ - unsigned int sector; - int segment; - - do { - sector = get_sector((*ptr)++); - segment = cvt2segment(sector); - } while (sector != 0 && segment < segment_id); - (*ptr) --; /* point to first sector >= segment_id */ - /* Get all sectors in segment_id - */ - if (sector == 0 || segment != segment_id) { - *map = 0; - return 0; - } else if ((sector & 0x800000) && - (ft_format_code == fmt_var || ft_format_code == fmt_big)) { - *map = EMPTY_SEGMENT; - return FT_SECTORS_PER_SEGMENT; - } else { - int count = 1; - SectorCount *tmp_ptr = (*ptr) + 1; - - *map = cvt2map(sector); - while ((sector = get_sector(tmp_ptr++)) != 0 && - (segment = cvt2segment(sector)) == segment_id) { - *map |= cvt2map(sector); - ++count; - } - return count; - } -} - -static int backwards_seek_entry(int segment_id, - SectorCount **ptr, - SectorMap *map) -{ - unsigned int sector; - int segment; /* max unsigned int */ - - if (*ptr <= (SectorCount *)bad_sector_map) { - *map = 0; - return 0; - } - do { - sector = get_sector(--(*ptr)); - segment = cvt2segment(sector); - } while (*ptr > (SectorCount *)bad_sector_map && segment > segment_id); - if (segment > segment_id) { /* at start of list, no entry found */ - *map = 0; - return 0; - } else if (segment < segment_id) { - /* before smaller entry, adjust for overshoot */ - (*ptr) ++; - *map = 0; - return 0; - } else if ((sector & 0x800000) && - (ft_format_code == fmt_big || ft_format_code == fmt_var)) { - *map = EMPTY_SEGMENT; - return FT_SECTORS_PER_SEGMENT; - } else { /* get all sectors in segment_id */ - int count = 1; - - *map = cvt2map(sector); - while(*ptr > (SectorCount *)bad_sector_map) { - sector = get_sector(--(*ptr)); - segment = cvt2segment(sector); - if (segment != segment_id) { - break; - } - *map |= cvt2map(sector); - ++count; - } - if (segment < segment_id) { - (*ptr) ++; - } - return count; - } -} - -#if 0 -static void ftape_put_bad_sector_entry(int segment_id, SectorMap new_map) -{ - SectorCount *ptr = (SectorCount *)bad_sector_map; - int count; - int new_count; - SectorMap map; - TRACE_FUN(ft_t_any); - - if (ft_format_code == fmt_1100ft || - ft_format_code == fmt_var || - ft_format_code == fmt_big) { - count = forward_seek_entry(segment_id, &ptr, &map); - new_count = count_ones(new_map); - /* If format code == 4 put empty segment instead of 32 - * bad sectors. - */ - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - if (new_count == FT_SECTORS_PER_SEGMENT) { - new_count = 1; - } - if (count == FT_SECTORS_PER_SEGMENT) { - count = 1; - } - } - if (count != new_count) { - /* insert (or delete if < 0) new_count - count - * entries. Move trailing part of list - * including terminating 0. - */ - SectorCount *hi_ptr = ptr; - - do { - } while (get_sector(hi_ptr++) != 0); - /* Note: ptr is of type byte *, and each bad sector - * consumes 3 bytes. - */ - memmove(ptr + new_count, ptr + count, - (size_t)(hi_ptr - (ptr + count))*sizeof(SectorCount)); - } - TRACE(ft_t_noise, "putting map 0x%08x at %p, segment %d", - (unsigned int)new_map, ptr, segment_id); - if (new_count == 1 && new_map == EMPTY_SEGMENT) { - put_sector(ptr++, (0x800001 + - segment_id * - FT_SECTORS_PER_SEGMENT)); - } else { - int i = 0; - - while (new_map) { - if (new_map & 1) { - put_sector(ptr++, - 1 + segment_id * - FT_SECTORS_PER_SEGMENT + i); - } - ++i; - new_map >>= 1; - } - } - } else { - ((SectorMap *) bad_sector_map)[segment_id] = new_map; - } - TRACE_EXIT; -} -#endif /* 0 */ - -SectorMap ftape_get_bad_sector_entry(int segment_id) -{ - if (ft_used_header_segment == -1) { - /* When reading header segment we'll need a blank map. - */ - return 0; - } else if (bsm_hash_ptr != NULL) { - /* Invariants: - * map - mask value returned on last call. - * bsm_hash_ptr - points to first sector greater or equal to - * first sector in last_referenced segment. - * last_referenced - segment id used in the last call, - * sector and map belong to this id. - * This code is designed for sequential access and retries. - * For true random access it may have to be redesigned. - */ - static int last_reference = -1; - static SectorMap map; - - if (segment_id > last_reference) { - /* Skip all sectors before segment_id - */ - forward_seek_entry(segment_id, &bsm_hash_ptr, &map); - } else if (segment_id < last_reference) { - /* Skip backwards until begin of buffer or - * first sector in segment_id - */ - backwards_seek_entry(segment_id, &bsm_hash_ptr, &map); - } /* segment_id == last_reference : keep map */ - last_reference = segment_id; - return map; - } else { - return ((SectorMap *) bad_sector_map)[segment_id]; - } -} - -/* This is simply here to prevent us from overwriting other kernel - * data. Writes will result in NULL Pointer dereference. - */ -void ftape_init_bsm(void) -{ - bad_sector_map = NULL; - bsm_hash_ptr = NULL; -} diff --git a/drivers/char/ftape/lowlevel/ftape-bsm.h b/drivers/char/ftape/lowlevel/ftape-bsm.h deleted file mode 100644 index ed45465af4d4..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-bsm.h +++ /dev/null @@ -1,66 +0,0 @@ -#ifndef _FTAPE_BSM_H -#define _FTAPE_BSM_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-bsm.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:07 $ - * - * This file contains definitions for the bad sector map handling - * routines for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#define EMPTY_SEGMENT (0xffffffff) -#define FAKE_SEGMENT (0xfffffffe) - -/* maximum (format code 4) bad sector map size (bytes). - */ -#define BAD_SECTOR_MAP_SIZE (29 * SECTOR_SIZE - 256) - -/* format code 4 bad sector entry, ftape uses this - * internally for all format codes - */ -typedef __u32 SectorMap; -/* variable and 1100 ft bad sector map entry. These three bytes represent - * a single sector address measured from BOT. - */ -typedef struct NewSectorMap { - __u8 bytes[3]; -} SectorCount; - - -/* - * ftape-bsm.c defined global vars. - */ - -/* - * ftape-bsm.c defined global functions. - */ -extern void update_bad_sector_map(__u8 * buffer); -extern void ftape_extract_bad_sector_map(__u8 * buffer); -extern SectorMap ftape_get_bad_sector_entry(int segment_id); -extern __u8 *ftape_find_end_of_bsm_list(__u8 * address); -extern void ftape_init_bsm(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-buffer.c b/drivers/char/ftape/lowlevel/ftape-buffer.c deleted file mode 100644 index c706ff162771..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-buffer.c +++ /dev/null @@ -1,130 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-buffer.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/16 23:33:11 $ - * - * This file contains the allocator/dealloctor for ftape's dynamic dma - * buffer. - */ - -#include -#include -#include -#include - -#include -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-buffer.h" - -/* DMA'able memory allocation stuff. - */ - -static inline void *dmaalloc(size_t size) -{ - unsigned long addr; - - if (size == 0) { - return NULL; - } - addr = __get_dma_pages(GFP_KERNEL, get_order(size)); - if (addr) { - struct page *page; - - for (page = virt_to_page(addr); page < virt_to_page(addr+size); page++) - SetPageReserved(page); - } - return (void *)addr; -} - -static inline void dmafree(void *addr, size_t size) -{ - if (size > 0) { - struct page *page; - - for (page = virt_to_page((unsigned long)addr); - page < virt_to_page((unsigned long)addr+size); page++) - ClearPageReserved(page); - free_pages((unsigned long) addr, get_order(size)); - } -} - -static int add_one_buffer(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_nr_buffers >= FT_MAX_NR_BUFFERS) { - TRACE_EXIT -ENOMEM; - } - ft_buffer[ft_nr_buffers] = kmalloc(sizeof(buffer_struct), GFP_KERNEL); - if (ft_buffer[ft_nr_buffers] == NULL) { - TRACE_EXIT -ENOMEM; - } - memset(ft_buffer[ft_nr_buffers], 0, sizeof(buffer_struct)); - ft_buffer[ft_nr_buffers]->address = dmaalloc(FT_BUFF_SIZE); - if (ft_buffer[ft_nr_buffers]->address == NULL) { - kfree(ft_buffer[ft_nr_buffers]); - ft_buffer[ft_nr_buffers] = NULL; - TRACE_EXIT -ENOMEM; - } - ft_nr_buffers ++; - TRACE(ft_t_info, "buffer nr #%d @ %p, dma area @ %p", - ft_nr_buffers, - ft_buffer[ft_nr_buffers-1], - ft_buffer[ft_nr_buffers-1]->address); - TRACE_EXIT 0; -} - -static void del_one_buffer(void) -{ - TRACE_FUN(ft_t_flow); - if (ft_nr_buffers > 0) { - TRACE(ft_t_info, "releasing buffer nr #%d @ %p, dma area @ %p", - ft_nr_buffers, - ft_buffer[ft_nr_buffers-1], - ft_buffer[ft_nr_buffers-1]->address); - ft_nr_buffers --; - dmafree(ft_buffer[ft_nr_buffers]->address, FT_BUFF_SIZE); - kfree(ft_buffer[ft_nr_buffers]); - ft_buffer[ft_nr_buffers] = NULL; - } - TRACE_EXIT; -} - -int ftape_set_nr_buffers(int cnt) -{ - int delta = cnt - ft_nr_buffers; - TRACE_FUN(ft_t_flow); - - if (delta > 0) { - while (delta--) { - if (add_one_buffer() < 0) { - TRACE_EXIT -ENOMEM; - } - } - } else if (delta < 0) { - while (delta++) { - del_one_buffer(); - } - } - ftape_zap_read_buffers(); - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/ftape-buffer.h b/drivers/char/ftape/lowlevel/ftape-buffer.h deleted file mode 100644 index eec99cee8f82..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-buffer.h +++ /dev/null @@ -1,32 +0,0 @@ -#ifndef _FTAPE_BUFFER_H -#define _FTAPE_BUFFER_H - -/* - * Copyright (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-buffer.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:08 $ - * - * This file contains the allocator/dealloctor for ftape's dynamic dma - * buffer. - */ - -extern int ftape_set_nr_buffers(int cnt); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-calibr.c b/drivers/char/ftape/lowlevel/ftape-calibr.c deleted file mode 100644 index 8e50bfd35a52..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-calibr.c +++ /dev/null @@ -1,275 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-calibr.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:08 $ - * - * GP calibration routine for processor speed dependent - * functions. - */ - -#include -#include -#include -#include -#if defined(__alpha__) -# include -#elif defined(__x86_64__) -# include -# include -#elif defined(__i386__) -# include -#endif -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-calibr.h" -#include "../lowlevel/fdc-io.h" - -#undef DEBUG - -#if !defined(__alpha__) && !defined(__i386__) && !defined(__x86_64__) -# error Ftape is not implemented for this architecture! -#endif - -#if defined(__alpha__) || defined(__x86_64__) -static unsigned long ps_per_cycle = 0; -#endif - -static spinlock_t calibr_lock; - -/* - * Note: On Intel PCs, the clock ticks at 100 Hz (HZ==100) which is - * too slow for certain timeouts (and that clock doesn't even tick - * when interrupts are disabled). For that reason, the 8254 timer is - * used directly to implement fine-grained timeouts. However, on - * Alpha PCs, the 8254 is *not* used to implement the clock tick - * (which is 1024 Hz, normally) and the 8254 timer runs at some - * "random" frequency (it seems to run at 18Hz, but it's not safe to - * rely on this value). Instead, we use the Alpha's "rpcc" - * instruction to read cycle counts. As this is a 32 bit counter, - * it will overflow only once per 30 seconds (on a 200MHz machine), - * which is plenty. - */ - -unsigned int ftape_timestamp(void) -{ -#if defined(__alpha__) - unsigned long r; - - asm volatile ("rpcc %0" : "=r" (r)); - return r; -#elif defined(__x86_64__) - unsigned long r; - rdtscl(r); - return r; -#elif defined(__i386__) - -/* - * Note that there is some time between counter underflowing and jiffies - * increasing, so the code below won't always give correct output. - * -Vojtech - */ - - unsigned long flags; - __u16 lo; - __u16 hi; - - spin_lock_irqsave(&calibr_lock, flags); - outb_p(0x00, 0x43); /* latch the count ASAP */ - lo = inb_p(0x40); /* read the latched count */ - lo |= inb(0x40) << 8; - hi = jiffies; - spin_unlock_irqrestore(&calibr_lock, flags); - return ((hi + 1) * (unsigned int) LATCH) - lo; /* downcounter ! */ -#endif -} - -static unsigned int short_ftape_timestamp(void) -{ -#if defined(__alpha__) || defined(__x86_64__) - return ftape_timestamp(); -#elif defined(__i386__) - unsigned int count; - unsigned long flags; - - spin_lock_irqsave(&calibr_lock, flags); - outb_p(0x00, 0x43); /* latch the count ASAP */ - count = inb_p(0x40); /* read the latched count */ - count |= inb(0x40) << 8; - spin_unlock_irqrestore(&calibr_lock, flags); - return (LATCH - count); /* normal: downcounter */ -#endif -} - -static unsigned int diff(unsigned int t0, unsigned int t1) -{ -#if defined(__alpha__) || defined(__x86_64__) - return (t1 - t0); -#elif defined(__i386__) - /* - * This is tricky: to work for both short and full ftape_timestamps - * we'll have to discriminate between these. - * If it _looks_ like short stamps with wrapping around we'll - * asume it are. This will generate a small error if it really - * was a (very large) delta from full ftape_timestamps. - */ - return (t1 <= t0 && t0 <= LATCH) ? t1 + LATCH - t0 : t1 - t0; -#endif -} - -static unsigned int usecs(unsigned int count) -{ -#if defined(__alpha__) || defined(__x86_64__) - return (ps_per_cycle * count) / 1000000UL; -#elif defined(__i386__) - return (10000 * count) / ((CLOCK_TICK_RATE + 50) / 100); -#endif -} - -unsigned int ftape_timediff(unsigned int t0, unsigned int t1) -{ - /* - * Calculate difference in usec for ftape_timestamp results t0 & t1. - * Note that on the i386 platform with short time-stamps, the - * maximum allowed timespan is 1/HZ or we'll lose ticks! - */ - return usecs(diff(t0, t1)); -} - -/* To get an indication of the I/O performance, - * measure the duration of the inb() function. - */ -static void time_inb(void) -{ - int i; - int t0, t1; - unsigned long flags; - int status; - TRACE_FUN(ft_t_any); - - spin_lock_irqsave(&calibr_lock, flags); - t0 = short_ftape_timestamp(); - for (i = 0; i < 1000; ++i) { - status = inb(fdc.msr); - } - t1 = short_ftape_timestamp(); - spin_unlock_irqrestore(&calibr_lock, flags); - TRACE(ft_t_info, "inb() duration: %d nsec", ftape_timediff(t0, t1)); - TRACE_EXIT; -} - -static void init_clock(void) -{ - TRACE_FUN(ft_t_any); - -#if defined(__x86_64__) - ps_per_cycle = 1000000000UL / cpu_khz; -#elif defined(__alpha__) - extern struct hwrpb_struct *hwrpb; - ps_per_cycle = (1000*1000*1000*1000UL) / hwrpb->cycle_freq; -#endif - TRACE_EXIT; -} - -/* - * Input: function taking int count as parameter. - * pointers to calculated calibration variables. - */ -void ftape_calibrate(char *name, - void (*fun) (unsigned int), - unsigned int *calibr_count, - unsigned int *calibr_time) -{ - static int first_time = 1; - int i; - unsigned int tc = 0; - unsigned int count; - unsigned int time; -#if defined(__i386__) - unsigned int old_tc = 0; - unsigned int old_count = 1; - unsigned int old_time = 1; -#endif - TRACE_FUN(ft_t_flow); - - if (first_time) { /* get idea of I/O performance */ - init_clock(); - time_inb(); - first_time = 0; - } - /* value of timeout must be set so that on very slow systems - * it will give a time less than one jiffy, and on - * very fast systems it'll give reasonable precision. - */ - - count = 40; - for (i = 0; i < 15; ++i) { - unsigned int t0; - unsigned int t1; - unsigned int once; - unsigned int multiple; - unsigned long flags; - - *calibr_count = - *calibr_time = count; /* set TC to 1 */ - spin_lock_irqsave(&calibr_lock, flags); - fun(0); /* dummy, get code into cache */ - t0 = short_ftape_timestamp(); - fun(0); /* overhead + one test */ - t1 = short_ftape_timestamp(); - once = diff(t0, t1); - t0 = short_ftape_timestamp(); - fun(count); /* overhead + count tests */ - t1 = short_ftape_timestamp(); - multiple = diff(t0, t1); - spin_unlock_irqrestore(&calibr_lock, flags); - time = ftape_timediff(0, multiple - once); - tc = (1000 * time) / (count - 1); - TRACE(ft_t_any, "once:%3d us,%6d times:%6d us, TC:%5d ns", - usecs(once), count - 1, usecs(multiple), tc); -#if defined(__alpha__) || defined(__x86_64__) - /* - * Increase the calibration count exponentially until the - * calibration time exceeds 100 ms. - */ - if (time >= 100*1000) { - break; - } -#elif defined(__i386__) - /* - * increase the count until the resulting time nears 2/HZ, - * then the tc will drop sharply because we lose LATCH counts. - */ - if (tc <= old_tc / 2) { - time = old_time; - count = old_count; - break; - } - old_tc = tc; - old_count = count; - old_time = time; -#endif - count *= 2; - } - *calibr_count = count - 1; - *calibr_time = time; - TRACE(ft_t_info, "TC for `%s()' = %d nsec (at %d counts)", - name, (1000 * *calibr_time) / *calibr_count, *calibr_count); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/ftape-calibr.h b/drivers/char/ftape/lowlevel/ftape-calibr.h deleted file mode 100644 index 0c7e75246c7d..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-calibr.h +++ /dev/null @@ -1,37 +0,0 @@ -#ifndef _FTAPE_CALIBR_H -#define _FTAPE_CALIBR_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-calibr.h,v $ - * $Revision: 1.1 $ - * $Date: 1997/09/19 09:05:26 $ - * - * This file contains a gp calibration routine for - * hardware dependent timeout functions. - */ - -extern void ftape_calibrate(char *name, - void (*fun) (unsigned int), - unsigned int *calibr_count, - unsigned int *calibr_time); -extern unsigned int ftape_timestamp(void); -extern unsigned int ftape_timediff(unsigned int t0, unsigned int t1); - -#endif /* _FTAPE_CALIBR_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-ctl.c b/drivers/char/ftape/lowlevel/ftape-ctl.c deleted file mode 100644 index 5d7c1ce92d59..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-ctl.c +++ /dev/null @@ -1,896 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ctl.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/11/11 14:37:44 $ - * - * This file contains the non-read/write ftape functions for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include -#include - -#include -#include -#include -#include - -/* ease porting between pre-2.4.x and later kernels */ -#define vma_get_pgoff(v) ((v)->vm_pgoff) - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -ftape_info ftape_status = { -/* vendor information */ - { 0, }, /* drive type */ -/* data rates */ - 500, /* used data rate */ - 500, /* drive max rate */ - 500, /* fdc max rate */ -/* drive selection, either FTAPE_SEL_A/B/C/D */ - -1, /* drive selection */ -/* flags set after decode the drive and tape status */ - 0, /* formatted */ - 1, /* no tape */ - 1, /* write protected */ - 1, /* new tape */ -/* values of last queried drive/tape status and error */ - {{0,}}, /* last error code */ - {{0,}}, /* drive status, configuration, tape status */ -/* cartridge geometry */ - 20, /* tracks_per_tape */ - 102, /* segments_per_track */ -/* location of header segments, etc. */ - -1, /* used_header_segment */ - -1, /* header_segment_1 */ - -1, /* header_segment_2 */ - -1, /* first_data_segment */ - -1, /* last_data_segment */ -/* the format code as stored in the header segment */ - fmt_normal, /* format code */ -/* the default for the qic std: unknown */ - -1, -/* is tape running? */ - idle, /* runner_state */ -/* is tape reading/writing/verifying/formatting/deleting */ - idle, /* driver state */ -/* flags fatal hardware error */ - 1, /* failure */ -/* history record */ - { 0, } /* history record */ -}; - -int ftape_segments_per_head = 1020; -int ftape_segments_per_cylinder = 4; -int ftape_init_drive_needed = 1; /* need to be global for ftape_reset_drive() - * in ftape-io.c - */ - -/* Local vars. - */ -static const vendor_struct vendors[] = QIC117_VENDORS; -static const wakeup_method methods[] = WAKEUP_METHODS; - -const ftape_info *ftape_get_status(void) -{ -#if defined(STATUS_PARANOYA) - static ftape_info get_status; - - get_status = ftape_status; - return &get_status; -#else - return &ftape_status; /* maybe return only a copy of it to assure - * read only access - */ -#endif -} - -static int ftape_not_operational(int status) -{ - /* return true if status indicates tape can not be used. - */ - return ((status ^ QIC_STATUS_CARTRIDGE_PRESENT) & - (QIC_STATUS_ERROR | - QIC_STATUS_CARTRIDGE_PRESENT | - QIC_STATUS_NEW_CARTRIDGE)); -} - -int ftape_seek_to_eot(void) -{ - int status; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - while ((status & QIC_STATUS_AT_EOT) == 0) { - if (ftape_not_operational(status)) { - TRACE_EXIT -EIO; - } - TRACE_CATCH(ftape_command_wait(QIC_PHYSICAL_FORWARD, - ftape_timeout.rewind,&status),); - } - TRACE_EXIT 0; -} - -int ftape_seek_to_bot(void) -{ - int status; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - while ((status & QIC_STATUS_AT_BOT) == 0) { - if (ftape_not_operational(status)) { - TRACE_EXIT -EIO; - } - TRACE_CATCH(ftape_command_wait(QIC_PHYSICAL_REVERSE, - ftape_timeout.rewind,&status),); - } - TRACE_EXIT 0; -} - -static int ftape_new_cartridge(void) -{ - ft_location.track = -1; /* force seek on first access */ - ftape_zap_read_buffers(); - ftape_zap_write_buffers(); - return 0; -} - -int ftape_abort_operation(void) -{ - int result = 0; - int status; - TRACE_FUN(ft_t_flow); - - if (ft_runner_status == running) { - TRACE(ft_t_noise, "aborting runner, waiting"); - - ft_runner_status = do_abort; - /* set timeout so that the tape will run to logical EOT - * if we missed the last sector and there are no queue pulses. - */ - result = ftape_dumb_stop(); - } - if (ft_runner_status != idle) { - if (ft_runner_status == do_abort) { - TRACE(ft_t_noise, "forcing runner abort"); - } - TRACE(ft_t_noise, "stopping tape"); - result = ftape_stop_tape(&status); - ft_location.known = 0; - ft_runner_status = idle; - } - ftape_reset_buffer(); - ftape_zap_read_buffers(); - ftape_set_state(idle); - TRACE_EXIT result; -} - -static int lookup_vendor_id(unsigned int vendor_id) -{ - int i = 0; - - while (vendors[i].vendor_id != vendor_id) { - if (++i >= NR_ITEMS(vendors)) { - return -1; - } - } - return i; -} - -static void ftape_detach_drive(void) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "disabling tape drive and fdc"); - ftape_put_drive_to_sleep(ft_drive_type.wake_up); - fdc_catch_stray_interrupts(1); /* one always comes */ - fdc_disable(); - fdc_release_irq_and_dma(); - fdc_release_regions(); - TRACE_EXIT; -} - -static void clear_history(void) -{ - ft_history.used = 0; - ft_history.id_am_errors = - ft_history.id_crc_errors = - ft_history.data_am_errors = - ft_history.data_crc_errors = - ft_history.overrun_errors = - ft_history.no_data_errors = - ft_history.retries = - ft_history.crc_errors = - ft_history.crc_failures = - ft_history.ecc_failures = - ft_history.corrected = - ft_history.defects = - ft_history.rewinds = 0; -} - -static int ftape_activate_drive(vendor_struct * drive_type) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - /* If we already know the drive type, wake it up. - * Else try to find out what kind of drive is attached. - */ - if (drive_type->wake_up != unknown_wake_up) { - TRACE(ft_t_flow, "enabling tape drive and fdc"); - result = ftape_wakeup_drive(drive_type->wake_up); - if (result < 0) { - TRACE(ft_t_err, "known wakeup method failed"); - } - } else { - wake_up_types method; - const ft_trace_t old_tracing = TRACE_LEVEL; - if (TRACE_LEVEL < ft_t_flow) { - SET_TRACE_LEVEL(ft_t_bug); - } - - /* Try to awaken the drive using all known methods. - * Lower tracing for a while. - */ - for (method=no_wake_up; method < NR_ITEMS(methods); ++method) { - drive_type->wake_up = method; -#ifdef CONFIG_FT_TWO_DRIVES - /* Test setup for dual drive configuration. - * /dev/rft2 uses mountain wakeup - * /dev/rft3 uses colorado wakeup - * Other systems will use the normal scheme. - */ - if ((ft_drive_sel < 2) || - (ft_drive_sel == 2 && method == FT_WAKE_UP_1) || - (ft_drive_sel == 3 && method == FT_WAKE_UP_2)) { - result=ftape_wakeup_drive(drive_type->wake_up); - } else { - result = -EIO; - } -#else - result = ftape_wakeup_drive(drive_type->wake_up); -#endif - if (result >= 0) { - TRACE(ft_t_warn, "drive wakeup method: %s", - methods[drive_type->wake_up].name); - break; - } - } - SET_TRACE_LEVEL(old_tracing); - - if (method >= NR_ITEMS(methods)) { - /* no response at all, cannot open this drive */ - drive_type->wake_up = unknown_wake_up; - TRACE(ft_t_err, "no tape drive found !"); - result = -ENODEV; - } - } - TRACE_EXIT result; -} - -static int ftape_get_drive_status(void) -{ - int result; - int status; - TRACE_FUN(ft_t_flow); - - ft_no_tape = ft_write_protected = 0; - /* Tape drive is activated now. - * First clear error status if present. - */ - do { - result = ftape_ready_wait(ftape_timeout.reset, &status); - if (result < 0) { - if (result == -ETIME) { - TRACE(ft_t_err, "ftape_ready_wait timeout"); - } else if (result == -EINTR) { - TRACE(ft_t_err, "ftape_ready_wait aborted"); - } else { - TRACE(ft_t_err, "ftape_ready_wait failed"); - } - TRACE_EXIT -EIO; - } - /* Clear error condition (drive is ready !) - */ - if (status & QIC_STATUS_ERROR) { - unsigned int error; - qic117_cmd_t command; - - TRACE(ft_t_err, "error status set"); - result = ftape_report_error(&error, &command, 1); - if (result < 0) { - TRACE(ft_t_err, - "report_error_code failed: %d", result); - /* hope it's working next time */ - ftape_reset_drive(); - TRACE_EXIT -EIO; - } else if (error != 0) { - TRACE(ft_t_noise, "error code : %d", error); - TRACE(ft_t_noise, "error command: %d", command); - } - } - if (status & QIC_STATUS_NEW_CARTRIDGE) { - unsigned int error; - qic117_cmd_t command; - const ft_trace_t old_tracing = TRACE_LEVEL; - SET_TRACE_LEVEL(ft_t_bug); - - /* Undocumented feature: Must clear (not present!) - * error here or we'll fail later. - */ - ftape_report_error(&error, &command, 1); - - SET_TRACE_LEVEL(old_tracing); - TRACE(ft_t_info, "status: new cartridge"); - ft_new_tape = 1; - } else { - ft_new_tape = 0; - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - } while (status & QIC_STATUS_ERROR); - - ft_no_tape = !(status & QIC_STATUS_CARTRIDGE_PRESENT); - ft_write_protected = (status & QIC_STATUS_WRITE_PROTECT) != 0; - if (ft_no_tape) { - TRACE(ft_t_warn, "no cartridge present"); - } else { - if (ft_write_protected) { - TRACE(ft_t_noise, "Write protected cartridge"); - } - } - TRACE_EXIT 0; -} - -static void ftape_log_vendor_id(void) -{ - int vendor_index; - TRACE_FUN(ft_t_flow); - - ftape_report_vendor_id(&ft_drive_type.vendor_id); - vendor_index = lookup_vendor_id(ft_drive_type.vendor_id); - if (ft_drive_type.vendor_id == UNKNOWN_VENDOR && - ft_drive_type.wake_up == wake_up_colorado) { - vendor_index = 0; - /* hack to get rid of all this mail */ - ft_drive_type.vendor_id = 0; - } - if (vendor_index < 0) { - /* Unknown vendor id, first time opening device. The - * drive_type remains set to type found at wakeup - * time, this will probably keep the driver operating - * for this new vendor. - */ - TRACE(ft_t_warn, "\n" - KERN_INFO "============ unknown vendor id ===========\n" - KERN_INFO "A new, yet unsupported tape drive is found\n" - KERN_INFO "Please report the following values:\n" - KERN_INFO " Vendor id : 0x%04x\n" - KERN_INFO " Wakeup method : %s\n" - KERN_INFO "And a description of your tape drive\n" - KERN_INFO "to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - ft_drive_type.vendor_id, - methods[ft_drive_type.wake_up].name); - ft_drive_type.speed = 0; /* unknown */ - } else { - ft_drive_type.name = vendors[vendor_index].name; - ft_drive_type.speed = vendors[vendor_index].speed; - TRACE(ft_t_info, "tape drive type: %s", ft_drive_type.name); - /* scan all methods for this vendor_id in table */ - while(ft_drive_type.wake_up != vendors[vendor_index].wake_up) { - if (vendor_index < NR_ITEMS(vendors) - 1 && - vendors[vendor_index + 1].vendor_id - == - ft_drive_type.vendor_id) { - ++vendor_index; - } else { - break; - } - } - if (ft_drive_type.wake_up != vendors[vendor_index].wake_up) { - TRACE(ft_t_warn, "\n" - KERN_INFO "==========================================\n" - KERN_INFO "wakeup type mismatch:\n" - KERN_INFO "found: %s, expected: %s\n" - KERN_INFO "please report this to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - methods[ft_drive_type.wake_up].name, - methods[vendors[vendor_index].wake_up].name); - } - } - TRACE_EXIT; -} - -void ftape_calc_timeouts(unsigned int qic_std, - unsigned int data_rate, - unsigned int tape_len) -{ - int speed; /* deci-ips ! */ - int ff_speed; - int length; - TRACE_FUN(ft_t_any); - - /* tape transport speed - * data rate: QIC-40 QIC-80 QIC-3010 QIC-3020 - * - * 250 Kbps 25 ips n/a n/a n/a - * 500 Kbps 50 ips 34 ips 22.6 ips n/a - * 1 Mbps n/a 68 ips 45.2 ips 22.6 ips - * 2 Mbps n/a n/a n/a 45.2 ips - * - * fast tape transport speed is at least 68 ips. - */ - switch (qic_std) { - case QIC_TAPE_QIC40: - speed = (data_rate == 250) ? 250 : 500; - break; - case QIC_TAPE_QIC80: - speed = (data_rate == 500) ? 340 : 680; - break; - case QIC_TAPE_QIC3010: - speed = (data_rate == 500) ? 226 : 452; - break; - case QIC_TAPE_QIC3020: - speed = (data_rate == 1000) ? 226 : 452; - break; - default: - TRACE(ft_t_bug, "Unknown qic_std (bug) ?"); - speed = 500; - break; - } - if (ft_drive_type.speed == 0) { - unsigned long t0; - static int dt = 0; /* keep gcc from complaining */ - static int first_time = 1; - - /* Measure the time it takes to wind to EOT and back to BOT. - * If the tape length is known, calculate the rewind speed. - * Else keep the time value for calculation of the rewind - * speed later on, when the length _is_ known. - * Ask for a report only when length and speed are both known. - */ - if (first_time) { - ftape_seek_to_bot(); - t0 = jiffies; - ftape_seek_to_eot(); - ftape_seek_to_bot(); - dt = (int) (((jiffies - t0) * FT_USPT) / 1000); - if (dt < 1) { - dt = 1; /* prevent div by zero on failures */ - } - first_time = 0; - TRACE(ft_t_info, - "trying to determine seek timeout, got %d msec", - dt); - } - if (tape_len != 0) { - ft_drive_type.speed = - (2 * 12 * tape_len * 1000) / dt; - TRACE(ft_t_warn, "\n" - KERN_INFO "==========================================\n" - KERN_INFO "drive type: %s\n" - KERN_INFO "delta time = %d ms, length = %d ft\n" - KERN_INFO "has a maximum tape speed of %d ips\n" - KERN_INFO "please report this to "THE_FTAPE_MAINTAINER"\n" - KERN_INFO "==========================================", - ft_drive_type.name, dt, tape_len, - ft_drive_type.speed); - } - } - /* Handle unknown length tapes as very long ones. We'll - * determine the actual length from a header segment later. - * This is normal for all modern (Wide,TR1/2/3) formats. - */ - if (tape_len <= 0) { - TRACE(ft_t_noise, - "Unknown tape length, using maximal timeouts"); - length = QIC_TOP_TAPE_LEN; /* use worst case values */ - } else { - length = tape_len; /* use actual values */ - } - if (ft_drive_type.speed == 0) { - ff_speed = speed; - } else { - ff_speed = ft_drive_type.speed; - } - /* time to go from bot to eot at normal speed (data rate): - * time = (1+delta) * length (ft) * 12 (inch/ft) / speed (ips) - * delta = 10 % for seek speed, 20 % for rewind speed. - */ - ftape_timeout.seek = (length * 132 * FT_SECOND) / speed; - ftape_timeout.rewind = (length * 144 * FT_SECOND) / (10 * ff_speed); - ftape_timeout.reset = 20 * FT_SECOND + ftape_timeout.rewind; - TRACE(ft_t_noise, "timeouts for speed = %d, length = %d\n" - KERN_INFO "seek timeout : %d sec\n" - KERN_INFO "rewind timeout: %d sec\n" - KERN_INFO "reset timeout : %d sec", - speed, length, - (ftape_timeout.seek + 500) / 1000, - (ftape_timeout.rewind + 500) / 1000, - (ftape_timeout.reset + 500) / 1000); - TRACE_EXIT; -} - -/* This function calibrates the datarate (i.e. determines the maximal - * usable data rate) and sets the global variable ft_qic_std to qic_std - * - */ -int ftape_calibrate_data_rate(unsigned int qic_std) -{ - int rate = ft_fdc_rate_limit; - int result; - TRACE_FUN(ft_t_flow); - - ft_qic_std = qic_std; - - if (ft_qic_std == -1) { - TRACE_ABORT(-EIO, ft_t_err, - "Unable to determine data rate if QIC standard is unknown"); - } - - /* Select highest rate supported by both fdc and drive. - * Start with highest rate supported by the fdc. - */ - while (fdc_set_data_rate(rate) < 0 && rate > 250) { - rate /= 2; - } - TRACE(ft_t_info, - "Highest FDC supported data rate: %d Kbps", rate); - ft_fdc_max_rate = rate; - do { - result = ftape_set_data_rate(rate, ft_qic_std); - } while (result == -EINVAL && (rate /= 2) > 250); - if (result < 0) { - TRACE_ABORT(-EIO, ft_t_err, "set datarate failed"); - } - ft_data_rate = rate; - TRACE_EXIT 0; -} - -static int ftape_init_drive(void) -{ - int status; - qic_model model; - unsigned int qic_std; - unsigned int data_rate; - TRACE_FUN(ft_t_flow); - - ftape_init_drive_needed = 0; /* don't retry if this fails ? */ - TRACE_CATCH(ftape_report_raw_drive_status(&status),); - if (status & QIC_STATUS_CARTRIDGE_PRESENT) { - if (!(status & QIC_STATUS_AT_BOT)) { - /* Antique drives will get here after a soft reset, - * modern ones only if the driver is loaded when the - * tape wasn't rewound properly. - */ - /* Tape should be at bot if new cartridge ! */ - ftape_seek_to_bot(); - } - if (!(status & QIC_STATUS_REFERENCED)) { - TRACE(ft_t_flow, "starting seek_load_point"); - TRACE_CATCH(ftape_command_wait(QIC_SEEK_LOAD_POINT, - ftape_timeout.reset, - &status),); - } - } - ft_formatted = (status & QIC_STATUS_REFERENCED) != 0; - if (!ft_formatted) { - TRACE(ft_t_warn, "Warning: tape is not formatted !"); - } - - /* report configuration aborts when ftape_tape_len == -1 - * unknown qic_std is okay if not formatted. - */ - TRACE_CATCH(ftape_report_configuration(&model, - &data_rate, - &qic_std, - &ftape_tape_len),); - - /* Maybe add the following to the /proc entry - */ - TRACE(ft_t_info, "%s drive @ %d Kbps", - (model == prehistoric) ? "prehistoric" : - ((model == pre_qic117c) ? "pre QIC-117C" : - ((model == post_qic117b) ? "post QIC-117B" : - "post QIC-117D")), data_rate); - - if (ft_formatted) { - /* initialize ft_used_data_rate to maximum value - * and set ft_qic_std - */ - TRACE_CATCH(ftape_calibrate_data_rate(qic_std),); - if (ftape_tape_len == 0) { - TRACE(ft_t_info, "unknown length QIC-%s tape", - (ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) - ? "3010" : "3020"))); - } else { - TRACE(ft_t_info, "%d ft. QIC-%s tape", ftape_tape_len, - (ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) - ? "3010" : "3020"))); - } - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - /* soft write-protect QIC-40/QIC-80 cartridges used with a - * Colorado T3000 drive. Buggy hardware! - */ - if ((ft_drive_type.vendor_id == 0x011c6) && - ((ft_qic_std == QIC_TAPE_QIC40 || - ft_qic_std == QIC_TAPE_QIC80) && - !ft_write_protected)) { - TRACE(ft_t_warn, "\n" - KERN_INFO "The famous Colorado T3000 bug:\n" - KERN_INFO "%s drives can't write QIC40 and QIC80\n" - KERN_INFO "cartridges but don't set the write-protect flag!", - ft_drive_type.name); - ft_write_protected = 1; - } - } else { - /* Doesn't make too much sense to set the data rate - * because we don't know what to use for the write - * precompensation. - * Need to do this again when formatting the cartridge. - */ - ft_data_rate = data_rate; - ftape_calc_timeouts(QIC_TAPE_QIC40, - data_rate, - ftape_tape_len); - } - ftape_new_cartridge(); - TRACE_EXIT 0; -} - -static void ftape_munmap(void) -{ - int i; - TRACE_FUN(ft_t_flow); - - for (i = 0; i < ft_nr_buffers; i++) { - ft_buffer[i]->mmapped = 0; - } - TRACE_EXIT; -} - -/* Map the dma buffers into the virtual address range given by vma. - * We only check the caller doesn't map non-existent buffers. We - * don't check for multiple mappings. - */ -int ftape_mmap(struct vm_area_struct *vma) -{ - int num_buffers; - int i; - TRACE_FUN(ft_t_flow); - - if (ft_failure) { - TRACE_EXIT -ENODEV; - } - if (!(vma->vm_flags & (VM_READ|VM_WRITE))) { - TRACE_ABORT(-EINVAL, ft_t_err, "Undefined mmap() access"); - } - if (vma_get_pgoff(vma) != 0) { - TRACE_ABORT(-EINVAL, ft_t_err, "page offset must be 0"); - } - if ((vma->vm_end - vma->vm_start) % FT_BUFF_SIZE != 0) { - TRACE_ABORT(-EINVAL, ft_t_err, - "size = %ld, should be a multiple of %d", - vma->vm_end - vma->vm_start, - FT_BUFF_SIZE); - } - num_buffers = (vma->vm_end - vma->vm_start) / FT_BUFF_SIZE; - if (num_buffers > ft_nr_buffers) { - TRACE_ABORT(-EINVAL, - ft_t_err, "size = %ld, should be less than %d", - vma->vm_end - vma->vm_start, - ft_nr_buffers * FT_BUFF_SIZE); - } - if (ft_driver_state != idle) { - /* this also clears the buffer states - */ - ftape_abort_operation(); - } else { - ftape_reset_buffer(); - } - for (i = 0; i < num_buffers; i++) { - unsigned long pfn; - - pfn = virt_to_phys(ft_buffer[i]->address) >> PAGE_SHIFT; - TRACE_CATCH(remap_pfn_range(vma, vma->vm_start + - i * FT_BUFF_SIZE, - pfn, - FT_BUFF_SIZE, - vma->vm_page_prot), - _res = -EAGAIN); - TRACE(ft_t_noise, "remapped dma buffer @ %p to location @ %p", - ft_buffer[i]->address, - (void *)(vma->vm_start + i * FT_BUFF_SIZE)); - } - for (i = 0; i < num_buffers; i++) { - memset(ft_buffer[i]->address, 0xAA, FT_BUFF_SIZE); - ft_buffer[i]->mmapped++; - } - TRACE_EXIT 0; -} - -static void ftape_init_driver(void); /* forward declaration */ - -/* OPEN routine called by kernel-interface code - */ -int ftape_enable(int drive_selection) -{ - TRACE_FUN(ft_t_any); - - if (ft_drive_sel == -1 || ft_drive_sel != drive_selection) { - /* Other selection than last time - */ - ftape_init_driver(); - } - ft_drive_sel = FTAPE_SEL(drive_selection); - ft_failure = 0; - TRACE_CATCH(fdc_init(),); /* init & detect fdc */ - TRACE_CATCH(ftape_activate_drive(&ft_drive_type), - fdc_disable(); - fdc_release_irq_and_dma(); - fdc_release_regions()); - TRACE_CATCH(ftape_get_drive_status(), ftape_detach_drive()); - if (ft_drive_type.vendor_id == UNKNOWN_VENDOR) { - ftape_log_vendor_id(); - } - if (ft_new_tape) { - ftape_init_drive_needed = 1; - } - if (!ft_no_tape && ftape_init_drive_needed) { - TRACE_CATCH(ftape_init_drive(), ftape_detach_drive()); - } - ftape_munmap(); /* clear the mmap flag */ - clear_history(); - TRACE_EXIT 0; -} - -/* release routine called by the high level interface modules - * zftape or sftape. - */ -void ftape_disable(void) -{ - int i; - TRACE_FUN(ft_t_any); - - for (i = 0; i < ft_nr_buffers; i++) { - if (ft_buffer[i]->mmapped) { - TRACE(ft_t_noise, "first byte of buffer %d: 0x%02x", - i, *ft_buffer[i]->address); - } - } - if (sigtestsetmask(¤t->pending.signal, _DONT_BLOCK) && - !(sigtestsetmask(¤t->pending.signal, _NEVER_BLOCK)) && - ftape_tape_running) { - TRACE(ft_t_warn, - "Interrupted by fatal signal and tape still running"); - ftape_dumb_stop(); - ftape_abort_operation(); /* it's annoying */ - } else { - ftape_set_state(idle); - } - ftape_detach_drive(); - if (ft_history.used) { - TRACE(ft_t_info, "== Non-fatal errors this run: =="); - TRACE(ft_t_info, "fdc isr statistics:\n" - KERN_INFO " id_am_errors : %3d\n" - KERN_INFO " id_crc_errors : %3d\n" - KERN_INFO " data_am_errors : %3d\n" - KERN_INFO " data_crc_errors : %3d\n" - KERN_INFO " overrun_errors : %3d\n" - KERN_INFO " no_data_errors : %3d\n" - KERN_INFO " retries : %3d", - ft_history.id_am_errors, ft_history.id_crc_errors, - ft_history.data_am_errors, ft_history.data_crc_errors, - ft_history.overrun_errors, ft_history.no_data_errors, - ft_history.retries); - if (ft_history.used & 1) { - TRACE(ft_t_info, "ecc statistics:\n" - KERN_INFO " crc_errors : %3d\n" - KERN_INFO " crc_failures : %3d\n" - KERN_INFO " ecc_failures : %3d\n" - KERN_INFO " sectors corrected: %3d", - ft_history.crc_errors, ft_history.crc_failures, - ft_history.ecc_failures, ft_history.corrected); - } - if (ft_history.defects > 0) { - TRACE(ft_t_warn, "Warning: %d media defects!", - ft_history.defects); - } - if (ft_history.rewinds > 0) { - TRACE(ft_t_info, "tape motion statistics:\n" - KERN_INFO "repositions : %3d", - ft_history.rewinds); - } - } - ft_failure = 1; - TRACE_EXIT; -} - -static void ftape_init_driver(void) -{ - TRACE_FUN(ft_t_flow); - - ft_drive_type.vendor_id = UNKNOWN_VENDOR; - ft_drive_type.speed = 0; - ft_drive_type.wake_up = unknown_wake_up; - ft_drive_type.name = "Unknown"; - - ftape_timeout.seek = 650 * FT_SECOND; - ftape_timeout.reset = 670 * FT_SECOND; - ftape_timeout.rewind = 650 * FT_SECOND; - ftape_timeout.head_seek = 15 * FT_SECOND; - ftape_timeout.stop = 5 * FT_SECOND; - ftape_timeout.pause = 16 * FT_SECOND; - - ft_qic_std = -1; - ftape_tape_len = 0; /* unknown */ - ftape_current_command = 0; - ftape_current_cylinder = -1; - - ft_segments_per_track = 102; - ftape_segments_per_head = 1020; - ftape_segments_per_cylinder = 4; - ft_tracks_per_tape = 20; - - ft_failure = 1; - - ft_formatted = 0; - ft_no_tape = 1; - ft_write_protected = 1; - ft_new_tape = 1; - - ft_driver_state = idle; - - ft_data_rate = - ft_fdc_max_rate = 500; - ft_drive_max_rate = 0; /* triggers set_rate_test() */ - - ftape_init_drive_needed = 1; - - ft_header_segment_1 = -1; - ft_header_segment_2 = -1; - ft_used_header_segment = -1; - ft_first_data_segment = -1; - ft_last_data_segment = -1; - - ft_location.track = -1; - ft_location.known = 0; - - ftape_tape_running = 0; - ftape_might_be_off_track = 1; - - ftape_new_cartridge(); /* init some tape related variables */ - ftape_init_bsm(); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/lowlevel/ftape-ctl.h b/drivers/char/ftape/lowlevel/ftape-ctl.h deleted file mode 100644 index 5f5e30bc3615..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-ctl.h +++ /dev/null @@ -1,162 +0,0 @@ -#ifndef _FTAPE_CTL_H -#define _FTAPE_CTL_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ctl.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:09 $ - * - * This file contains the non-standard IOCTL related definitions - * for the QIC-40/80/3010/3020 floppy-tape driver "ftape" for - * Linux. - */ - -#include -#include -#include - -#include "../lowlevel/ftape-rw.h" -#include - -typedef struct { - int used; /* any reading or writing done */ - /* isr statistics */ - unsigned int id_am_errors; /* id address mark not found */ - unsigned int id_crc_errors; /* crc error in id address mark */ - unsigned int data_am_errors; /* data address mark not found */ - unsigned int data_crc_errors; /* crc error in data field */ - unsigned int overrun_errors; /* fdc access timing problem */ - unsigned int no_data_errors; /* sector not found */ - unsigned int retries; /* number of tape retries */ - /* ecc statistics */ - unsigned int crc_errors; /* crc error in data */ - unsigned int crc_failures; /* bad data without crc error */ - unsigned int ecc_failures; /* failed to correct */ - unsigned int corrected; /* total sectors corrected */ - /* general statistics */ - unsigned int rewinds; /* number of tape rewinds */ - unsigned int defects; /* bad sectors due to media defects */ -} history_record; - -/* this structure contains * ALL * information that we want - * our child modules to know about, but don't want them to - * modify. - */ -typedef struct { - /* vendor information */ - vendor_struct fti_drive_type; - /* data rates */ - unsigned int fti_used_data_rate; - unsigned int fti_drive_max_rate; - unsigned int fti_fdc_max_rate; - /* drive selection, either FTAPE_SEL_A/B/C/D */ - int fti_drive_sel; - /* flags set after decode the drive and tape status */ - unsigned int fti_formatted :1; - unsigned int fti_no_tape :1; - unsigned int fti_write_protected:1; - unsigned int fti_new_tape :1; - /* values of last queried drive/tape status and error */ - ft_drive_error fti_last_error; - ft_drive_status fti_last_status; - /* cartridge geometry */ - unsigned int fti_tracks_per_tape; - unsigned int fti_segments_per_track; - /* location of header segments, etc. */ - int fti_used_header_segment; - int fti_header_segment_1; - int fti_header_segment_2; - int fti_first_data_segment; - int fti_last_data_segment; - /* the format code as stored in the header segment */ - ft_format_type fti_format_code; - /* the following is the sole reason for the ftape_set_status() call */ - unsigned int fti_qic_std; - /* is tape running? */ - volatile enum runner_status_enum fti_runner_status; - /* is tape reading/writing/verifying/formatting/deleting */ - buffer_state_enum fti_state; - /* flags fatal hardware error */ - unsigned int fti_failure:1; - /* history record */ - history_record fti_history; -} ftape_info; - -/* vendor information */ -#define ft_drive_type ftape_status.fti_drive_type -/* data rates */ -#define ft_data_rate ftape_status.fti_used_data_rate -#define ft_drive_max_rate ftape_status.fti_drive_max_rate -#define ft_fdc_max_rate ftape_status.fti_fdc_max_rate -/* drive selection, either FTAPE_SEL_A/B/C/D */ -#define ft_drive_sel ftape_status.fti_drive_sel -/* flags set after decode the drive and tape status */ -#define ft_formatted ftape_status.fti_formatted -#define ft_no_tape ftape_status.fti_no_tape -#define ft_write_protected ftape_status.fti_write_protected -#define ft_new_tape ftape_status.fti_new_tape -/* values of last queried drive/tape status and error */ -#define ft_last_error ftape_status.fti_last_error -#define ft_last_status ftape_status.fti_last_status -/* cartridge geometry */ -#define ft_tracks_per_tape ftape_status.fti_tracks_per_tape -#define ft_segments_per_track ftape_status.fti_segments_per_track -/* the format code as stored in the header segment */ -#define ft_format_code ftape_status.fti_format_code -/* the qic status as returned by report drive configuration */ -#define ft_qic_std ftape_status.fti_qic_std -#define ft_used_header_segment ftape_status.fti_used_header_segment -#define ft_header_segment_1 ftape_status.fti_header_segment_1 -#define ft_header_segment_2 ftape_status.fti_header_segment_2 -#define ft_first_data_segment ftape_status.fti_first_data_segment -#define ft_last_data_segment ftape_status.fti_last_data_segment -/* is tape running? */ -#define ft_runner_status ftape_status.fti_runner_status -/* is tape reading/writing/verifying/formatting/deleting */ -#define ft_driver_state ftape_status.fti_state -/* flags fatal hardware error */ -#define ft_failure ftape_status.fti_failure -/* history record */ -#define ft_history ftape_status.fti_history - -/* - * ftape-ctl.c defined global vars. - */ -extern ftape_info ftape_status; -extern int ftape_segments_per_head; -extern int ftape_segments_per_cylinder; -extern int ftape_init_drive_needed; - -/* - * ftape-ctl.c defined global functions. - */ -extern int ftape_mmap(struct vm_area_struct *vma); -extern int ftape_enable(int drive_selection); -extern void ftape_disable(void); -extern int ftape_seek_to_bot(void); -extern int ftape_seek_to_eot(void); -extern int ftape_abort_operation(void); -extern void ftape_calc_timeouts(unsigned int qic_std, - unsigned int data_rate, - unsigned int tape_len); -extern int ftape_calibrate_data_rate(unsigned int qic_std); -extern const ftape_info *ftape_get_status(void); -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-ecc.c b/drivers/char/ftape/lowlevel/ftape-ecc.c deleted file mode 100644 index e5632f674bc8..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-ecc.c +++ /dev/null @@ -1,853 +0,0 @@ -/* - * - * Copyright (c) 1993 Ning and David Mosberger. - - This is based on code originally written by Bas Laarhoven (bas@vimec.nl) - and David L. Brown, Jr., and incorporates improvements suggested by - Kai Harrekilde-Petersen. - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ecc.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:18:10 $ - * - * This file contains the Reed-Solomon error correction code - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-ecc.h" - -/* Machines that are big-endian should define macro BIG_ENDIAN. - * Unfortunately, there doesn't appear to be a standard include file - * that works for all OSs. - */ - -#if defined(__sparc__) || defined(__hppa) -#define BIG_ENDIAN -#endif /* __sparc__ || __hppa */ - -#if defined(__mips__) -#error Find a smart way to determine the Endianness of the MIPS CPU -#endif - -/* Notice: to minimize the potential for confusion, we use r to - * denote the independent variable of the polynomials in the - * Galois Field GF(2^8). We reserve x for polynomials that - * that have coefficients in GF(2^8). - * - * The Galois Field in which coefficient arithmetic is performed are - * the polynomials over Z_2 (i.e., 0 and 1) modulo the irreducible - * polynomial f(r), where f(r)=r^8 + r^7 + r^2 + r + 1. A polynomial - * is represented as a byte with the MSB as the coefficient of r^7 and - * the LSB as the coefficient of r^0. For example, the binary - * representation of f(x) is 0x187 (of course, this doesn't fit into 8 - * bits). In this field, the polynomial r is a primitive element. - * That is, r^i with i in 0,...,255 enumerates all elements in the - * field. - * - * The generator polynomial for the QIC-80 ECC is - * - * g(x) = x^3 + r^105*x^2 + r^105*x + 1 - * - * which can be factored into: - * - * g(x) = (x-r^-1)(x-r^0)(x-r^1) - * - * the byte representation of the coefficients are: - * - * r^105 = 0xc0 - * r^-1 = 0xc3 - * r^0 = 0x01 - * r^1 = 0x02 - * - * Notice that r^-1 = r^254 as exponent arithmetic is performed - * modulo 2^8-1 = 255. - * - * For more information on Galois Fields and Reed-Solomon codes, refer - * to any good book. I found _An Introduction to Error Correcting - * Codes with Applications_ by S. A. Vanstone and P. C. van Oorschot - * to be a good introduction into the former. _CODING THEORY: The - * Essentials_ I found very useful for its concise description of - * Reed-Solomon encoding/decoding. - * - */ - -typedef __u8 Matrix[3][3]; - -/* - * gfpow[] is defined such that gfpow[i] returns r^i if - * i is in the range [0..255]. - */ -static const __u8 gfpow[] = -{ - 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, - 0x87, 0x89, 0x95, 0xad, 0xdd, 0x3d, 0x7a, 0xf4, - 0x6f, 0xde, 0x3b, 0x76, 0xec, 0x5f, 0xbe, 0xfb, - 0x71, 0xe2, 0x43, 0x86, 0x8b, 0x91, 0xa5, 0xcd, - 0x1d, 0x3a, 0x74, 0xe8, 0x57, 0xae, 0xdb, 0x31, - 0x62, 0xc4, 0x0f, 0x1e, 0x3c, 0x78, 0xf0, 0x67, - 0xce, 0x1b, 0x36, 0x6c, 0xd8, 0x37, 0x6e, 0xdc, - 0x3f, 0x7e, 0xfc, 0x7f, 0xfe, 0x7b, 0xf6, 0x6b, - 0xd6, 0x2b, 0x56, 0xac, 0xdf, 0x39, 0x72, 0xe4, - 0x4f, 0x9e, 0xbb, 0xf1, 0x65, 0xca, 0x13, 0x26, - 0x4c, 0x98, 0xb7, 0xe9, 0x55, 0xaa, 0xd3, 0x21, - 0x42, 0x84, 0x8f, 0x99, 0xb5, 0xed, 0x5d, 0xba, - 0xf3, 0x61, 0xc2, 0x03, 0x06, 0x0c, 0x18, 0x30, - 0x60, 0xc0, 0x07, 0x0e, 0x1c, 0x38, 0x70, 0xe0, - 0x47, 0x8e, 0x9b, 0xb1, 0xe5, 0x4d, 0x9a, 0xb3, - 0xe1, 0x45, 0x8a, 0x93, 0xa1, 0xc5, 0x0d, 0x1a, - 0x34, 0x68, 0xd0, 0x27, 0x4e, 0x9c, 0xbf, 0xf9, - 0x75, 0xea, 0x53, 0xa6, 0xcb, 0x11, 0x22, 0x44, - 0x88, 0x97, 0xa9, 0xd5, 0x2d, 0x5a, 0xb4, 0xef, - 0x59, 0xb2, 0xe3, 0x41, 0x82, 0x83, 0x81, 0x85, - 0x8d, 0x9d, 0xbd, 0xfd, 0x7d, 0xfa, 0x73, 0xe6, - 0x4b, 0x96, 0xab, 0xd1, 0x25, 0x4a, 0x94, 0xaf, - 0xd9, 0x35, 0x6a, 0xd4, 0x2f, 0x5e, 0xbc, 0xff, - 0x79, 0xf2, 0x63, 0xc6, 0x0b, 0x16, 0x2c, 0x58, - 0xb0, 0xe7, 0x49, 0x92, 0xa3, 0xc1, 0x05, 0x0a, - 0x14, 0x28, 0x50, 0xa0, 0xc7, 0x09, 0x12, 0x24, - 0x48, 0x90, 0xa7, 0xc9, 0x15, 0x2a, 0x54, 0xa8, - 0xd7, 0x29, 0x52, 0xa4, 0xcf, 0x19, 0x32, 0x64, - 0xc8, 0x17, 0x2e, 0x5c, 0xb8, 0xf7, 0x69, 0xd2, - 0x23, 0x46, 0x8c, 0x9f, 0xb9, 0xf5, 0x6d, 0xda, - 0x33, 0x66, 0xcc, 0x1f, 0x3e, 0x7c, 0xf8, 0x77, - 0xee, 0x5b, 0xb6, 0xeb, 0x51, 0xa2, 0xc3, 0x01 -}; - -/* - * This is a log table. That is, gflog[r^i] returns i (modulo f(r)). - * gflog[0] is undefined and the first element is therefore not valid. - */ -static const __u8 gflog[256] = -{ - 0xff, 0x00, 0x01, 0x63, 0x02, 0xc6, 0x64, 0x6a, - 0x03, 0xcd, 0xc7, 0xbc, 0x65, 0x7e, 0x6b, 0x2a, - 0x04, 0x8d, 0xce, 0x4e, 0xc8, 0xd4, 0xbd, 0xe1, - 0x66, 0xdd, 0x7f, 0x31, 0x6c, 0x20, 0x2b, 0xf3, - 0x05, 0x57, 0x8e, 0xe8, 0xcf, 0xac, 0x4f, 0x83, - 0xc9, 0xd9, 0xd5, 0x41, 0xbe, 0x94, 0xe2, 0xb4, - 0x67, 0x27, 0xde, 0xf0, 0x80, 0xb1, 0x32, 0x35, - 0x6d, 0x45, 0x21, 0x12, 0x2c, 0x0d, 0xf4, 0x38, - 0x06, 0x9b, 0x58, 0x1a, 0x8f, 0x79, 0xe9, 0x70, - 0xd0, 0xc2, 0xad, 0xa8, 0x50, 0x75, 0x84, 0x48, - 0xca, 0xfc, 0xda, 0x8a, 0xd6, 0x54, 0x42, 0x24, - 0xbf, 0x98, 0x95, 0xf9, 0xe3, 0x5e, 0xb5, 0x15, - 0x68, 0x61, 0x28, 0xba, 0xdf, 0x4c, 0xf1, 0x2f, - 0x81, 0xe6, 0xb2, 0x3f, 0x33, 0xee, 0x36, 0x10, - 0x6e, 0x18, 0x46, 0xa6, 0x22, 0x88, 0x13, 0xf7, - 0x2d, 0xb8, 0x0e, 0x3d, 0xf5, 0xa4, 0x39, 0x3b, - 0x07, 0x9e, 0x9c, 0x9d, 0x59, 0x9f, 0x1b, 0x08, - 0x90, 0x09, 0x7a, 0x1c, 0xea, 0xa0, 0x71, 0x5a, - 0xd1, 0x1d, 0xc3, 0x7b, 0xae, 0x0a, 0xa9, 0x91, - 0x51, 0x5b, 0x76, 0x72, 0x85, 0xa1, 0x49, 0xeb, - 0xcb, 0x7c, 0xfd, 0xc4, 0xdb, 0x1e, 0x8b, 0xd2, - 0xd7, 0x92, 0x55, 0xaa, 0x43, 0x0b, 0x25, 0xaf, - 0xc0, 0x73, 0x99, 0x77, 0x96, 0x5c, 0xfa, 0x52, - 0xe4, 0xec, 0x5f, 0x4a, 0xb6, 0xa2, 0x16, 0x86, - 0x69, 0xc5, 0x62, 0xfe, 0x29, 0x7d, 0xbb, 0xcc, - 0xe0, 0xd3, 0x4d, 0x8c, 0xf2, 0x1f, 0x30, 0xdc, - 0x82, 0xab, 0xe7, 0x56, 0xb3, 0x93, 0x40, 0xd8, - 0x34, 0xb0, 0xef, 0x26, 0x37, 0x0c, 0x11, 0x44, - 0x6f, 0x78, 0x19, 0x9a, 0x47, 0x74, 0xa7, 0xc1, - 0x23, 0x53, 0x89, 0xfb, 0x14, 0x5d, 0xf8, 0x97, - 0x2e, 0x4b, 0xb9, 0x60, 0x0f, 0xed, 0x3e, 0xe5, - 0xf6, 0x87, 0xa5, 0x17, 0x3a, 0xa3, 0x3c, 0xb7 -}; - -/* This is a multiplication table for the factor 0xc0 (i.e., r^105 (mod f(r)). - * gfmul_c0[f] returns r^105 * f(r) (modulo f(r)). - */ -static const __u8 gfmul_c0[256] = -{ - 0x00, 0xc0, 0x07, 0xc7, 0x0e, 0xce, 0x09, 0xc9, - 0x1c, 0xdc, 0x1b, 0xdb, 0x12, 0xd2, 0x15, 0xd5, - 0x38, 0xf8, 0x3f, 0xff, 0x36, 0xf6, 0x31, 0xf1, - 0x24, 0xe4, 0x23, 0xe3, 0x2a, 0xea, 0x2d, 0xed, - 0x70, 0xb0, 0x77, 0xb7, 0x7e, 0xbe, 0x79, 0xb9, - 0x6c, 0xac, 0x6b, 0xab, 0x62, 0xa2, 0x65, 0xa5, - 0x48, 0x88, 0x4f, 0x8f, 0x46, 0x86, 0x41, 0x81, - 0x54, 0x94, 0x53, 0x93, 0x5a, 0x9a, 0x5d, 0x9d, - 0xe0, 0x20, 0xe7, 0x27, 0xee, 0x2e, 0xe9, 0x29, - 0xfc, 0x3c, 0xfb, 0x3b, 0xf2, 0x32, 0xf5, 0x35, - 0xd8, 0x18, 0xdf, 0x1f, 0xd6, 0x16, 0xd1, 0x11, - 0xc4, 0x04, 0xc3, 0x03, 0xca, 0x0a, 0xcd, 0x0d, - 0x90, 0x50, 0x97, 0x57, 0x9e, 0x5e, 0x99, 0x59, - 0x8c, 0x4c, 0x8b, 0x4b, 0x82, 0x42, 0x85, 0x45, - 0xa8, 0x68, 0xaf, 0x6f, 0xa6, 0x66, 0xa1, 0x61, - 0xb4, 0x74, 0xb3, 0x73, 0xba, 0x7a, 0xbd, 0x7d, - 0x47, 0x87, 0x40, 0x80, 0x49, 0x89, 0x4e, 0x8e, - 0x5b, 0x9b, 0x5c, 0x9c, 0x55, 0x95, 0x52, 0x92, - 0x7f, 0xbf, 0x78, 0xb8, 0x71, 0xb1, 0x76, 0xb6, - 0x63, 0xa3, 0x64, 0xa4, 0x6d, 0xad, 0x6a, 0xaa, - 0x37, 0xf7, 0x30, 0xf0, 0x39, 0xf9, 0x3e, 0xfe, - 0x2b, 0xeb, 0x2c, 0xec, 0x25, 0xe5, 0x22, 0xe2, - 0x0f, 0xcf, 0x08, 0xc8, 0x01, 0xc1, 0x06, 0xc6, - 0x13, 0xd3, 0x14, 0xd4, 0x1d, 0xdd, 0x1a, 0xda, - 0xa7, 0x67, 0xa0, 0x60, 0xa9, 0x69, 0xae, 0x6e, - 0xbb, 0x7b, 0xbc, 0x7c, 0xb5, 0x75, 0xb2, 0x72, - 0x9f, 0x5f, 0x98, 0x58, 0x91, 0x51, 0x96, 0x56, - 0x83, 0x43, 0x84, 0x44, 0x8d, 0x4d, 0x8a, 0x4a, - 0xd7, 0x17, 0xd0, 0x10, 0xd9, 0x19, 0xde, 0x1e, - 0xcb, 0x0b, 0xcc, 0x0c, 0xc5, 0x05, 0xc2, 0x02, - 0xef, 0x2f, 0xe8, 0x28, 0xe1, 0x21, 0xe6, 0x26, - 0xf3, 0x33, 0xf4, 0x34, 0xfd, 0x3d, 0xfa, 0x3a -}; - - -/* Returns V modulo 255 provided V is in the range -255,-254,...,509. - */ -static inline __u8 mod255(int v) -{ - if (v > 0) { - if (v < 255) { - return v; - } else { - return v - 255; - } - } else { - return v + 255; - } -} - - -/* Add two numbers in the field. Addition in this field is equivalent - * to a bit-wise exclusive OR operation---subtraction is therefore - * identical to addition. - */ -static inline __u8 gfadd(__u8 a, __u8 b) -{ - return a ^ b; -} - - -/* Add two vectors of numbers in the field. Each byte in A and B gets - * added individually. - */ -static inline unsigned long gfadd_long(unsigned long a, unsigned long b) -{ - return a ^ b; -} - - -/* Multiply two numbers in the field: - */ -static inline __u8 gfmul(__u8 a, __u8 b) -{ - if (a && b) { - return gfpow[mod255(gflog[a] + gflog[b])]; - } else { - return 0; - } -} - - -/* Just like gfmul, except we have already looked up the log of the - * second number. - */ -static inline __u8 gfmul_exp(__u8 a, int b) -{ - if (a) { - return gfpow[mod255(gflog[a] + b)]; - } else { - return 0; - } -} - - -/* Just like gfmul_exp, except that A is a vector of numbers. That - * is, each byte in A gets multiplied by gfpow[mod255(B)]. - */ -static inline unsigned long gfmul_exp_long(unsigned long a, int b) -{ - __u8 t; - - if (sizeof(long) == 4) { - return ( - ((t = (__u32)a >> 24 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 24) : 0) | - ((t = (__u32)a >> 16 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 16) : 0) | - ((t = (__u32)a >> 8 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 8) : 0) | - ((t = (__u32)a >> 0 & 0xff) ? - (((__u32) gfpow[mod255(gflog[t] + b)]) << 0) : 0)); - } else if (sizeof(long) == 8) { - return ( - ((t = (__u64)a >> 56 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 56) : 0) | - ((t = (__u64)a >> 48 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 48) : 0) | - ((t = (__u64)a >> 40 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 40) : 0) | - ((t = (__u64)a >> 32 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 32) : 0) | - ((t = (__u64)a >> 24 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 24) : 0) | - ((t = (__u64)a >> 16 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 16) : 0) | - ((t = (__u64)a >> 8 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 8) : 0) | - ((t = (__u64)a >> 0 & 0xff) ? - (((__u64) gfpow[mod255(gflog[t] + b)]) << 0) : 0)); - } else { - TRACE_FUN(ft_t_any); - TRACE_ABORT(-1, ft_t_err, "Error: size of long is %d bytes", - (int)sizeof(long)); - } -} - - -/* Divide two numbers in the field. Returns a/b (modulo f(x)). - */ -static inline __u8 gfdiv(__u8 a, __u8 b) -{ - if (!b) { - TRACE_FUN(ft_t_any); - TRACE_ABORT(0xff, ft_t_bug, "Error: division by zero"); - } else if (a == 0) { - return 0; - } else { - return gfpow[mod255(gflog[a] - gflog[b])]; - } -} - - -/* The following functions return the inverse of the matrix of the - * linear system that needs to be solved to determine the error - * magnitudes. The first deals with matrices of rank 3, while the - * second deals with matrices of rank 2. The error indices are passed - * in arguments L0,..,L2 (0=first sector, 31=last sector). The error - * indices must be sorted in ascending order, i.e., L00 CRC failures)"); - } - log_det = 255 - gflog[det]; - - /* Now, calculate all of the coefficients: - */ - Ainv[0][0]= gfmul_exp(gfadd(gfpow[l1], gfpow[l2]), log_det); - Ainv[0][1]= gfmul_exp(gfadd(t21, t12), log_det); - Ainv[0][2]= gfmul_exp(gfadd(gfpow[255 - l1], gfpow[255 - l2]),log_det); - - Ainv[1][0]= gfmul_exp(gfadd(gfpow[l0], gfpow[l2]), log_det); - Ainv[1][1]= gfmul_exp(gfadd(t20, t02), log_det); - Ainv[1][2]= gfmul_exp(gfadd(gfpow[255 - l0], gfpow[255 - l2]),log_det); - - Ainv[2][0]= gfmul_exp(gfadd(gfpow[l0], gfpow[l1]), log_det); - Ainv[2][1]= gfmul_exp(gfadd(t10, t01), log_det); - Ainv[2][2]= gfmul_exp(gfadd(gfpow[255 - l0], gfpow[255 - l1]),log_det); - - return 1; -} - - -static inline int gfinv2(__u8 l0, __u8 l1, Matrix Ainv) -{ - __u8 det; - __u8 t1, t2; - int log_det; - - t1 = gfpow[255 - l0]; - t2 = gfpow[255 - l1]; - det = gfadd(t1, t2); - if (!det) { - TRACE_FUN(ft_t_any); - TRACE_ABORT(0, ft_t_err, - "Inversion failed (2 CRC errors, >0 CRC failures)"); - } - log_det = 255 - gflog[det]; - - /* Now, calculate all of the coefficients: - */ - Ainv[0][0] = Ainv[1][0] = gfpow[log_det]; - - Ainv[0][1] = gfmul_exp(t2, log_det); - Ainv[1][1] = gfmul_exp(t1, log_det); - - return 1; -} - - -/* Multiply matrix A by vector S and return result in vector B. M is - * assumed to be of order NxN, S and B of order Nx1. - */ -static inline void gfmat_mul(int n, Matrix A, - __u8 *s, __u8 *b) -{ - int i, j; - __u8 dot_prod; - - for (i = 0; i < n; ++i) { - dot_prod = 0; - for (j = 0; j < n; ++j) { - dot_prod = gfadd(dot_prod, gfmul(A[i][j], s[j])); - } - b[i] = dot_prod; - } -} - - - -/* The Reed Solomon ECC codes are computed over the N-th byte of each - * block, where N=SECTOR_SIZE. There are up to 29 blocks of data, and - * 3 blocks of ECC. The blocks are stored contiguously in memory. A - * segment, consequently, is assumed to have at least 4 blocks: one or - * more data blocks plus three ECC blocks. - * - * Notice: In QIC-80 speak, a CRC error is a sector with an incorrect - * CRC. A CRC failure is a sector with incorrect data, but - * a valid CRC. In the error control literature, the former - * is usually called "erasure", the latter "error." - */ -/* Compute the parity bytes for C columns of data, where C is the - * number of bytes that fit into a long integer. We use a linear - * feed-back register to do this. The parity bytes P[0], P[STRIDE], - * P[2*STRIDE] are computed such that: - * - * x^k * p(x) + m(x) = 0 (modulo g(x)) - * - * where k = NBLOCKS, - * p(x) = P[0] + P[STRIDE]*x + P[2*STRIDE]*x^2, and - * m(x) = sum_{i=0}^k m_i*x^i. - * m_i = DATA[i*SECTOR_SIZE] - */ -static inline void set_parity(unsigned long *data, - int nblocks, - unsigned long *p, - int stride) -{ - unsigned long p0, p1, p2, t1, t2, *end; - - end = data + nblocks * (FT_SECTOR_SIZE / sizeof(long)); - p0 = p1 = p2 = 0; - while (data < end) { - /* The new parity bytes p0_i, p1_i, p2_i are computed - * from the old values p0_{i-1}, p1_{i-1}, p2_{i-1} - * recursively as: - * - * p0_i = p1_{i-1} + r^105 * (m_{i-1} - p0_{i-1}) - * p1_i = p2_{i-1} + r^105 * (m_{i-1} - p0_{i-1}) - * p2_i = (m_{i-1} - p0_{i-1}) - * - * With the initial condition: p0_0 = p1_0 = p2_0 = 0. - */ - t1 = gfadd_long(*data, p0); - /* - * Multiply each byte in t1 by 0xc0: - */ - if (sizeof(long) == 4) { - t2= (((__u32) gfmul_c0[(__u32)t1 >> 24 & 0xff]) << 24 | - ((__u32) gfmul_c0[(__u32)t1 >> 16 & 0xff]) << 16 | - ((__u32) gfmul_c0[(__u32)t1 >> 8 & 0xff]) << 8 | - ((__u32) gfmul_c0[(__u32)t1 >> 0 & 0xff]) << 0); - } else if (sizeof(long) == 8) { - t2= (((__u64) gfmul_c0[(__u64)t1 >> 56 & 0xff]) << 56 | - ((__u64) gfmul_c0[(__u64)t1 >> 48 & 0xff]) << 48 | - ((__u64) gfmul_c0[(__u64)t1 >> 40 & 0xff]) << 40 | - ((__u64) gfmul_c0[(__u64)t1 >> 32 & 0xff]) << 32 | - ((__u64) gfmul_c0[(__u64)t1 >> 24 & 0xff]) << 24 | - ((__u64) gfmul_c0[(__u64)t1 >> 16 & 0xff]) << 16 | - ((__u64) gfmul_c0[(__u64)t1 >> 8 & 0xff]) << 8 | - ((__u64) gfmul_c0[(__u64)t1 >> 0 & 0xff]) << 0); - } else { - TRACE_FUN(ft_t_any); - TRACE(ft_t_err, "Error: long is of size %d", - (int) sizeof(long)); - TRACE_EXIT; - } - p0 = gfadd_long(t2, p1); - p1 = gfadd_long(t2, p2); - p2 = t1; - data += FT_SECTOR_SIZE / sizeof(long); - } - *p = p0; - p += stride; - *p = p1; - p += stride; - *p = p2; - return; -} - - -/* Compute the 3 syndrome values. DATA should point to the first byte - * of the column for which the syndromes are desired. The syndromes - * are computed over the first NBLOCKS of rows. The three bytes will - * be placed in S[0], S[1], and S[2]. - * - * S[i] is the value of the "message" polynomial m(x) evaluated at the - * i-th root of the generator polynomial g(x). - * - * As g(x)=(x-r^-1)(x-1)(x-r^1) we evaluate the message polynomial at - * x=r^-1 to get S[0], at x=r^0=1 to get S[1], and at x=r to get S[2]. - * This could be done directly and efficiently via the Horner scheme. - * However, it would require multiplication tables for the factors - * r^-1 (0xc3) and r (0x02). The following scheme does not require - * any multiplication tables beyond what's needed for set_parity() - * anyway and is slightly faster if there are no errors and slightly - * slower if there are errors. The latter is hopefully the infrequent - * case. - * - * To understand the alternative algorithm, notice that set_parity(m, - * k, p) computes parity bytes such that: - * - * x^k * p(x) = m(x) (modulo g(x)). - * - * That is, to evaluate m(r^m), where r^m is a root of g(x), we can - * simply evaluate (r^m)^k*p(r^m). Also, notice that p is 0 if and - * only if s is zero. That is, if all parity bytes are 0, we know - * there is no error in the data and consequently there is no need to - * compute s(x) at all! In all other cases, we compute s(x) from p(x) - * by evaluating (r^m)^k*p(r^m) for m=-1, m=0, and m=1. The p(x) - * polynomial is evaluated via the Horner scheme. - */ -static int compute_syndromes(unsigned long *data, int nblocks, unsigned long *s) -{ - unsigned long p[3]; - - set_parity(data, nblocks, p, 1); - if (p[0] | p[1] | p[2]) { - /* Some of the checked columns do not have a zero - * syndrome. For simplicity, we compute the syndromes - * for all columns that we have computed the - * remainders for. - */ - s[0] = gfmul_exp_long( - gfadd_long(p[0], - gfmul_exp_long( - gfadd_long(p[1], - gfmul_exp_long(p[2], -1)), - -1)), - -nblocks); - s[1] = gfadd_long(gfadd_long(p[2], p[1]), p[0]); - s[2] = gfmul_exp_long( - gfadd_long(p[0], - gfmul_exp_long( - gfadd_long(p[1], - gfmul_exp_long(p[2], 1)), - 1)), - nblocks); - return 0; - } else { - return 1; - } -} - - -/* Correct the block in the column pointed to by DATA. There are NBAD - * CRC errors and their indices are in BAD_LOC[0], up to - * BAD_LOC[NBAD-1]. If NBAD>1, Ainv holds the inverse of the matrix - * of the linear system that needs to be solved to determine the error - * magnitudes. S[0], S[1], and S[2] are the syndrome values. If row - * j gets corrected, then bit j will be set in CORRECTION_MAP. - */ -static inline int correct_block(__u8 *data, int nblocks, - int nbad, int *bad_loc, Matrix Ainv, - __u8 *s, - SectorMap * correction_map) -{ - int ncorrected = 0; - int i; - __u8 t1, t2; - __u8 c0, c1, c2; /* check bytes */ - __u8 error_mag[3], log_error_mag; - __u8 *dp, l, e; - TRACE_FUN(ft_t_any); - - switch (nbad) { - case 0: - /* might have a CRC failure: */ - if (s[0] == 0) { - /* more than one error */ - TRACE_ABORT(-1, ft_t_err, - "ECC failed (0 CRC errors, >1 CRC failures)"); - } - t1 = gfdiv(s[1], s[0]); - if ((bad_loc[nbad++] = gflog[t1]) >= nblocks) { - TRACE(ft_t_err, - "ECC failed (0 CRC errors, >1 CRC failures)"); - TRACE_ABORT(-1, ft_t_err, - "attempt to correct data at %d", bad_loc[0]); - } - error_mag[0] = s[1]; - break; - case 1: - t1 = gfadd(gfmul_exp(s[1], bad_loc[0]), s[2]); - t2 = gfadd(gfmul_exp(s[0], bad_loc[0]), s[1]); - if (t1 == 0 && t2 == 0) { - /* one erasure, no error: */ - Ainv[0][0] = gfpow[bad_loc[0]]; - } else if (t1 == 0 || t2 == 0) { - /* one erasure and more than one error: */ - TRACE_ABORT(-1, ft_t_err, - "ECC failed (1 erasure, >1 error)"); - } else { - /* one erasure, one error: */ - if ((bad_loc[nbad++] = gflog[gfdiv(t1, t2)]) - >= nblocks) { - TRACE(ft_t_err, "ECC failed " - "(1 CRC errors, >1 CRC failures)"); - TRACE_ABORT(-1, ft_t_err, - "attempt to correct data at %d", - bad_loc[1]); - } - if (!gfinv2(bad_loc[0], bad_loc[1], Ainv)) { - /* inversion failed---must have more - * than one error - */ - TRACE_EXIT -1; - } - } - /* FALL THROUGH TO ERROR MAGNITUDE COMPUTATION: - */ - case 2: - case 3: - /* compute error magnitudes: */ - gfmat_mul(nbad, Ainv, s, error_mag); - break; - - default: - TRACE_ABORT(-1, ft_t_err, - "Internal Error: number of CRC errors > 3"); - } - - /* Perform correction by adding ERROR_MAG[i] to the byte at - * offset BAD_LOC[i]. Also add the value of the computed - * error polynomial to the syndrome values. If the correction - * was successful, the resulting check bytes should be zero - * (i.e., the corrected data is a valid code word). - */ - c0 = s[0]; - c1 = s[1]; - c2 = s[2]; - for (i = 0; i < nbad; ++i) { - e = error_mag[i]; - if (e) { - /* correct the byte at offset L by magnitude E: */ - l = bad_loc[i]; - dp = &data[l * FT_SECTOR_SIZE]; - *dp = gfadd(*dp, e); - *correction_map |= 1 << l; - ++ncorrected; - - log_error_mag = gflog[e]; - c0 = gfadd(c0, gfpow[mod255(log_error_mag - l)]); - c1 = gfadd(c1, e); - c2 = gfadd(c2, gfpow[mod255(log_error_mag + l)]); - } - } - if (c0 || c1 || c2) { - TRACE_ABORT(-1, ft_t_err, - "ECC self-check failed, too many errors"); - } - TRACE_EXIT ncorrected; -} - - -#if defined(ECC_SANITY_CHECK) || defined(ECC_PARANOID) - -/* Perform a sanity check on the computed parity bytes: - */ -static int sanity_check(unsigned long *data, int nblocks) -{ - TRACE_FUN(ft_t_any); - unsigned long s[3]; - - if (!compute_syndromes(data, nblocks, s)) { - TRACE_ABORT(0, ft_bug, - "Internal Error: syndrome self-check failed"); - } - TRACE_EXIT 1; -} - -#endif /* defined(ECC_SANITY_CHECK) || defined(ECC_PARANOID) */ - -/* Compute the parity for an entire segment of data. - */ -int ftape_ecc_set_segment_parity(struct memory_segment *mseg) -{ - int i; - __u8 *parity_bytes; - - parity_bytes = &mseg->data[(mseg->blocks - 3) * FT_SECTOR_SIZE]; - for (i = 0; i < FT_SECTOR_SIZE; i += sizeof(long)) { - set_parity((unsigned long *) &mseg->data[i], mseg->blocks - 3, - (unsigned long *) &parity_bytes[i], - FT_SECTOR_SIZE / sizeof(long)); -#ifdef ECC_PARANOID - if (!sanity_check((unsigned long *) &mseg->data[i], - mseg->blocks)) { - return -1; - } -#endif /* ECC_PARANOID */ - } - return 0; -} - - -/* Checks and corrects (if possible) the segment MSEG. Returns one of - * ECC_OK, ECC_CORRECTED, and ECC_FAILED. - */ -int ftape_ecc_correct_data(struct memory_segment *mseg) -{ - int col, i, result; - int ncorrected = 0; - int nerasures = 0; /* # of erasures (CRC errors) */ - int erasure_loc[3]; /* erasure locations */ - unsigned long ss[3]; - __u8 s[3]; - Matrix Ainv; - TRACE_FUN(ft_t_flow); - - mseg->corrected = 0; - - /* find first column that has non-zero syndromes: */ - for (col = 0; col < FT_SECTOR_SIZE; col += sizeof(long)) { - if (!compute_syndromes((unsigned long *) &mseg->data[col], - mseg->blocks, ss)) { - /* something is wrong---have to fix things */ - break; - } - } - if (col >= FT_SECTOR_SIZE) { - /* all syndromes are ok, therefore nothing to correct */ - TRACE_EXIT ECC_OK; - } - /* count the number of CRC errors if there were any: */ - if (mseg->read_bad) { - for (i = 0; i < mseg->blocks; i++) { - if (BAD_CHECK(mseg->read_bad, i)) { - if (nerasures >= 3) { - /* this is too much for ECC */ - TRACE_ABORT(ECC_FAILED, ft_t_err, - "ECC failed (>3 CRC errors)"); - } /* if */ - erasure_loc[nerasures++] = i; - } - } - } - /* - * If there are at least 2 CRC errors, determine inverse of matrix - * of linear system to be solved: - */ - switch (nerasures) { - case 2: - if (!gfinv2(erasure_loc[0], erasure_loc[1], Ainv)) { - TRACE_EXIT ECC_FAILED; - } - break; - case 3: - if (!gfinv3(erasure_loc[0], erasure_loc[1], - erasure_loc[2], Ainv)) { - TRACE_EXIT ECC_FAILED; - } - break; - default: - /* this is not an error condition... */ - break; - } - - do { - for (i = 0; i < sizeof(long); ++i) { - s[0] = ss[0]; - s[1] = ss[1]; - s[2] = ss[2]; - if (s[0] | s[1] | s[2]) { -#ifdef BIG_ENDIAN - result = correct_block( - &mseg->data[col + sizeof(long) - 1 - i], - mseg->blocks, - nerasures, - erasure_loc, - Ainv, - s, - &mseg->corrected); -#else - result = correct_block(&mseg->data[col + i], - mseg->blocks, - nerasures, - erasure_loc, - Ainv, - s, - &mseg->corrected); -#endif - if (result < 0) { - TRACE_EXIT ECC_FAILED; - } - ncorrected += result; - } - ss[0] >>= 8; - ss[1] >>= 8; - ss[2] >>= 8; - } - -#ifdef ECC_SANITY_CHECK - if (!sanity_check((unsigned long *) &mseg->data[col], - mseg->blocks)) { - TRACE_EXIT ECC_FAILED; - } -#endif /* ECC_SANITY_CHECK */ - - /* find next column with non-zero syndromes: */ - while ((col += sizeof(long)) < FT_SECTOR_SIZE) { - if (!compute_syndromes((unsigned long *) - &mseg->data[col], mseg->blocks, ss)) { - /* something is wrong---have to fix things */ - break; - } - } - } while (col < FT_SECTOR_SIZE); - if (ncorrected && nerasures == 0) { - TRACE(ft_t_warn, "block contained error not caught by CRC"); - } - TRACE((ncorrected > 0) ? ft_t_noise : ft_t_any, "number of corrections: %d", ncorrected); - TRACE_EXIT ncorrected ? ECC_CORRECTED : ECC_OK; -} diff --git a/drivers/char/ftape/lowlevel/ftape-ecc.h b/drivers/char/ftape/lowlevel/ftape-ecc.h deleted file mode 100644 index 4829146fe9a0..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-ecc.h +++ /dev/null @@ -1,84 +0,0 @@ -#ifndef _FTAPE_ECC_H_ -#define _FTAPE_ECC_H_ - -/* - * Copyright (C) 1993 Ning and David Mosberger. - * Original: - * Copyright (C) 1993 Bas Laarhoven. - * Copyright (C) 1992 David L. Brown, Jr. - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-ecc.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:11 $ - * - * This file contains the definitions for the - * Reed-Solomon error correction code - * for the QIC-40/80 tape streamer device driver. - */ - -#include "../lowlevel/ftape-bsm.h" - -#define BAD_CLEAR(entry) ((entry)=0) -#define BAD_SET(entry,sector) ((entry)|=(1<<(sector))) -#define BAD_CHECK(entry,sector) ((entry)&(1<<(sector))) - -/* - * Return values for ecc_correct_data: - */ -enum { - ECC_OK, /* Data was correct. */ - ECC_CORRECTED, /* Correctable error in data. */ - ECC_FAILED, /* Could not correct data. */ -}; - -/* - * Representation of an in memory segment. MARKED_BAD lists the - * sectors that were marked bad during formatting. If the N-th sector - * in a segment is marked bad, bit 1< -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-format.h" - -#if defined(TESTING) -#define FT_FMT_SEGS_PER_BUF 50 -#else -#define FT_FMT_SEGS_PER_BUF (FT_BUFF_SIZE/(4*FT_SECTORS_PER_SEGMENT)) -#endif - -static spinlock_t ftape_format_lock; - -/* - * first segment of the new buffer - */ -static int switch_segment; - -/* - * at most 256 segments fit into one 32 kb buffer. Even TR-1 cartridges have - * more than this many segments per track, so better be careful. - * - * buffer_struct *buff: buffer to store the formatting coordinates in - * int start: starting segment for this buffer. - * int spt: segments per track - * - * Note: segment ids are relative to the start of the track here. - */ -static void setup_format_buffer(buffer_struct *buff, int start, int spt, - __u8 gap3) -{ - int to_do = spt - start; - TRACE_FUN(ft_t_flow); - - if (to_do > FT_FMT_SEGS_PER_BUF) { - to_do = FT_FMT_SEGS_PER_BUF; - } - buff->ptr = buff->address; - buff->remaining = to_do * FT_SECTORS_PER_SEGMENT; /* # sectors */ - buff->bytes = buff->remaining * 4; /* need 4 bytes per sector */ - buff->gap3 = gap3; - buff->segment_id = start; - buff->next_segment = start + to_do; - if (buff->next_segment >= spt) { - buff->next_segment = 0; /* 0 means: stop runner */ - } - buff->status = waiting; /* tells the isr that it can use - * this buffer - */ - TRACE_EXIT; -} - - -/* - * start formatting a new track. - */ -int ftape_format_track(const unsigned int track, const __u8 gap3) -{ - unsigned long flags; - buffer_struct *tail, *head; - int status; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(ftape_ready_wait(ftape_timeout.pause, &status),); - if (track & 1) { - if (!(status & QIC_STATUS_AT_EOT)) { - TRACE_CATCH(ftape_seek_to_eot(),); - } - } else { - if (!(status & QIC_STATUS_AT_BOT)) { - TRACE_CATCH(ftape_seek_to_bot(),); - } - } - ftape_abort_operation(); /* this sets ft_head = ft_tail = 0 */ - ftape_set_state(formatting); - - TRACE(ft_t_noise, - "Formatting track %d, logical: from segment %d to %d", - track, track * ft_segments_per_track, - (track + 1) * ft_segments_per_track - 1); - - /* - * initialize the buffer switching protocol for this track - */ - head = ftape_get_buffer(ft_queue_head); /* tape isn't running yet */ - tail = ftape_get_buffer(ft_queue_tail); /* tape isn't running yet */ - switch_segment = 0; - do { - FT_SIGNAL_EXIT(_DONT_BLOCK); - setup_format_buffer(tail, switch_segment, - ft_segments_per_track, gap3); - switch_segment = tail->next_segment; - } while ((switch_segment != 0) && - ((tail = ftape_next_buffer(ft_queue_tail)) != head)); - /* go */ - head->status = formatting; - TRACE_CATCH(ftape_seek_head_to_track(track),); - TRACE_CATCH(ftape_command(QIC_LOGICAL_FORWARD),); - spin_lock_irqsave(&ftape_format_lock, flags); - TRACE_CATCH(fdc_setup_formatting(head), restore_flags(flags)); - spin_unlock_irqrestore(&ftape_format_lock, flags); - TRACE_EXIT 0; -} - -/* return segment id of segment currently being formatted and do the - * buffer switching stuff. - */ -int ftape_format_status(unsigned int *segment_id) -{ - buffer_struct *tail = ftape_get_buffer(ft_queue_tail); - int result; - TRACE_FUN(ft_t_flow); - - while (switch_segment != 0 && - ftape_get_buffer(ft_queue_head) != tail) { - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* need more buffers, first wait for empty buffer - */ - TRACE_CATCH(ftape_wait_segment(formatting),); - /* don't worry for gap3. If we ever hit this piece of code, - * then all buffer already have the correct gap3 set! - */ - setup_format_buffer(tail, switch_segment, - ft_segments_per_track, tail->gap3); - switch_segment = tail->next_segment; - if (switch_segment != 0) { - tail = ftape_next_buffer(ft_queue_tail); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting || ft_runner_status == do_abort) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - TRACE(ft_t_warn, "Error formatting segment %d", - ftape_get_buffer(ft_queue_head)->segment_id); - (void)ftape_abort_operation(); - TRACE_EXIT (head->status != error) ? -EAGAIN : -EIO; - } - /* - * don't care if the timer expires, this is just kind of a - * "select" operation that lets the calling process sleep - * until something has happened - */ - if (fdc_interrupt_wait(5 * FT_SECOND) < 0) { - TRACE(ft_t_noise, "End of track %d at segment %d", - ft_location.track, - ftape_get_buffer(ft_queue_head)->segment_id); - result = 1; /* end of track, unlock module */ - } else { - result = 0; - } - /* - * the calling process should use the seg id to determine - * which parts of the dma buffers can be safely overwritten - * with new data. - */ - *segment_id = ftape_get_buffer(ft_queue_head)->segment_id; - /* - * Internally we start counting segment ids from the start of - * each track when formatting, but externally we keep them - * relative to the start of the tape: - */ - *segment_id += ft_location.track * ft_segments_per_track; - TRACE_EXIT result; -} - -/* - * The segment id is relative to the start of the tape - */ -int ftape_verify_segment(const unsigned int segment_id, SectorMap *bsm) -{ - int result; - int verify_done = 0; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Verifying segment %d", segment_id); - - if (ft_driver_state != verifying) { - TRACE(ft_t_noise, "calling ftape_abort_operation"); - if (ftape_abort_operation() < 0) { - TRACE(ft_t_err, "ftape_abort_operation failed"); - TRACE_EXIT -EIO; - } - } - *bsm = 0x00000000; - ftape_set_state(verifying); - for (;;) { - buffer_struct *tail; - /* - * Allow escape from this loop on signal - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* - * Search all full buffers for the first matching the - * wanted segment. Clear other buffers on the fly. - */ - tail = ftape_get_buffer(ft_queue_tail); - while (!verify_done && tail->status == done) { - /* - * Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (tail->segment_id == segment_id) { - /* If out buffer is already full, - * return its contents. - */ - TRACE(ft_t_flow, "found segment in cache: %d", - segment_id); - if ((tail->soft_error_map | - tail->hard_error_map) != 0) { - TRACE(ft_t_info,"bsm[%d] = 0x%08lx", - segment_id, - (unsigned long) - (tail->soft_error_map | - tail->hard_error_map)); - *bsm = (tail->soft_error_map | - tail->hard_error_map); - } - verify_done = 1; - } else { - TRACE(ft_t_flow,"zapping segment in cache: %d", - tail->segment_id); - } - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - if (!verify_done && tail->status == verifying) { - if (tail->segment_id == segment_id) { - switch(ftape_wait_segment(verifying)) { - case 0: - break; - case -EINTR: - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by " - "non-blockable signal"); - break; - default: - ftape_abort_operation(); - ftape_set_state(verifying); - /* be picky */ - TRACE_ABORT(-EIO, ft_t_warn, - "wait_segment failed"); - } - } else { - /* We're reading the wrong segment, - * stop runner. - */ - TRACE(ft_t_noise, "verifying wrong segment"); - ftape_abort_operation(); - ftape_set_state(verifying); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - if (head->status == error || - head->status == verifying) { - /* no data or overrun error */ - head->status = waiting; - } - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait - * for BOT or EOT mark. Sets ft_runner_status to - * idle if at lEOT and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - if (verify_done) { - TRACE_EXIT 0; - } - /* Now at least one buffer is idle! - * Restart runner & tape if needed. - */ - /* We could optimize the following a little bit. We know that - * the bad sector map is empty. - */ - tail = ftape_get_buffer(ft_queue_tail); - if (tail->status == waiting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - - ftape_setup_new_segment(head, segment_id, -1); - ftape_calc_next_cluster(head); - if (ft_runner_status == idle) { - result = ftape_start_tape(segment_id, - head->sector_offset); - switch(result) { - case 0: - break; - case -ETIME: - case -EINTR: - TRACE_ABORT(result, ft_t_err, "Error: " - "segment %d unreachable", - segment_id); - break; - default: - *bsm = EMPTY_SEGMENT; - TRACE_EXIT 0; - break; - } - } - head->status = verifying; - fdc_setup_read_write(head, FDC_VERIFY); - } - } - /* not reached */ - TRACE_EXIT -EIO; -} diff --git a/drivers/char/ftape/lowlevel/ftape-format.h b/drivers/char/ftape/lowlevel/ftape-format.h deleted file mode 100644 index f15161566643..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-format.h +++ /dev/null @@ -1,37 +0,0 @@ -#ifndef _FTAPE_FORMAT_H -#define _FTAPE_FORMAT_H - -/* - * Copyright (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-format.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:13 $ - * - * This file contains the low level definitions for the - * formatting support for the QIC-40/80/3010/3020 floppy-tape - * driver "ftape" for Linux. - */ - -#ifdef __KERNEL__ -extern int ftape_format_track(const unsigned int track, const __u8 gap3); -extern int ftape_format_status(unsigned int *segment_id); -extern int ftape_verify_segment(const unsigned int segment_id, SectorMap *bsm); -#endif /* __KERNEL__ */ - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-init.c b/drivers/char/ftape/lowlevel/ftape-init.c deleted file mode 100644 index 4998132a81d1..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-init.c +++ /dev/null @@ -1,160 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * This file contains the code that interfaces the kernel - * for the QIC-40/80/3010/3020 floppy-tape driver for Linux. - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#ifdef CONFIG_ZFTAPE -#include -#endif - -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-proc.h" -#include "../lowlevel/ftape-tracing.h" - - -#if defined(MODULE) && !defined(CONFIG_FT_NO_TRACE_AT_ALL) -static int ft_tracing = -1; -#endif - - -/* Called by modules package when installing the driver - * or by kernel during the initialization phase - */ -static int __init ftape_init(void) -{ - TRACE_FUN(ft_t_flow); - -#ifdef MODULE -#ifndef CONFIG_FT_NO_TRACE_AT_ALL - if (ft_tracing != -1) { - ftape_tracing = ft_tracing; - } -#endif - printk(KERN_INFO FTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO "(c) 1993-1996 Bas Laarhoven (bas@vimec.nl)\n" -KERN_INFO "(c) 1995-1996 Kai Harrekilde-Petersen (khp@dolphinics.no)\n" -KERN_INFO "(c) 1996-1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO "QIC-117 driver for QIC-40/80/3010/3020 floppy tape drives\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO FTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "installing QIC-117 floppy tape hardware drive ... "); - TRACE(ft_t_info, "ftape_init @ 0x%p", ftape_init); - /* Allocate the DMA buffers. They are deallocated at cleanup() time. - */ -#ifdef TESTING -#ifdef MODULE - while (ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS) < 0) { - ftape_sleep(FT_SECOND/20); - if (signal_pending(current)) { - (void)ftape_set_nr_buffers(0); - TRACE(ft_t_bug, - "Killed by signal while allocating buffers."); - TRACE_ABORT(-EINTR, - ft_t_bug, "Free up memory and retry"); - } - } -#else - TRACE_CATCH(ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS), - (void)ftape_set_nr_buffers(0)); -#endif -#else - TRACE_CATCH(ftape_set_nr_buffers(CONFIG_FT_NR_BUFFERS), - (void)ftape_set_nr_buffers(0)); -#endif - ft_drive_sel = -1; - ft_failure = 1; /* inhibit any operation but open */ - ftape_udelay_calibrate(); /* must be before fdc_wait_calibrate ! */ - fdc_wait_calibrate(); -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - (void)ftape_proc_init(); -#endif -#ifdef CONFIG_ZFTAPE - (void)zft_init(); -#endif - TRACE_EXIT 0; -} - -module_param(ft_fdc_base, uint, 0); -MODULE_PARM_DESC(ft_fdc_base, "Base address of FDC controller."); -module_param(ft_fdc_irq, uint, 0); -MODULE_PARM_DESC(ft_fdc_irq, "IRQ (interrupt channel) to use."); -module_param(ft_fdc_dma, uint, 0); -MODULE_PARM_DESC(ft_fdc_dma, "DMA channel to use."); -module_param(ft_fdc_threshold, uint, 0); -MODULE_PARM_DESC(ft_fdc_threshold, "Threshold of the FDC Fifo."); -module_param(ft_fdc_rate_limit, uint, 0); -MODULE_PARM_DESC(ft_fdc_rate_limit, "Maximal data rate for FDC."); -module_param(ft_probe_fc10, bool, 0); -MODULE_PARM_DESC(ft_probe_fc10, - "If non-zero, probe for a Colorado FC-10/FC-20 controller."); -module_param(ft_mach2, bool, 0); -MODULE_PARM_DESC(ft_mach2, - "If non-zero, probe for a Mountain MACH-2 controller."); -#if defined(MODULE) && !defined(CONFIG_FT_NO_TRACE_AT_ALL) -module_param(ft_tracing, int, 0644); -MODULE_PARM_DESC(ft_tracing, - "Amount of debugging output, 0 <= tracing <= 8, default 3."); -#endif - -MODULE_AUTHOR( - "(c) 1993-1996 Bas Laarhoven (bas@vimec.nl), " - "(c) 1995-1996 Kai Harrekilde-Petersen (khp@dolphinics.no), " - "(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)"); -MODULE_DESCRIPTION( - "QIC-117 driver for QIC-40/80/3010/3020 floppy tape drives."); -MODULE_LICENSE("GPL"); - -static void __exit ftape_exit(void) -{ - TRACE_FUN(ft_t_flow); - -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - ftape_proc_destroy(); -#endif - (void)ftape_set_nr_buffers(0); - printk(KERN_INFO "ftape: unloaded.\n"); - TRACE_EXIT; -} - -module_init(ftape_init); -module_exit(ftape_exit); diff --git a/drivers/char/ftape/lowlevel/ftape-init.h b/drivers/char/ftape/lowlevel/ftape-init.h deleted file mode 100644 index 99a7b8ab086f..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-init.h +++ /dev/null @@ -1,43 +0,0 @@ -#ifndef _FTAPE_INIT_H -#define _FTAPE_INIT_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-init.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:16 $ - * - * This file contains the definitions for the interface to - * the Linux kernel for floppy tape driver ftape. - * - */ - -#include -#include - -#define _NEVER_BLOCK (sigmask(SIGKILL) | sigmask(SIGSTOP)) -#define _DONT_BLOCK (_NEVER_BLOCK | sigmask(SIGINT)) -#define _DO_BLOCK (sigmask(SIGPIPE)) - -#ifndef QIC117_TAPE_MAJOR -#define QIC117_TAPE_MAJOR 27 -#endif - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-io.c b/drivers/char/ftape/lowlevel/ftape-io.c deleted file mode 100644 index 259015aeff55..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-io.c +++ /dev/null @@ -1,992 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996 Kai Harrekilde-Petersen, - * (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-io.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/11/11 14:02:36 $ - * - * This file contains the general control functions for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-calibr.h" - -/* Global vars. - */ -/* NOTE: sectors start numbering at 1, all others at 0 ! */ -ft_timeout_table ftape_timeout; -unsigned int ftape_tape_len; -volatile qic117_cmd_t ftape_current_command; -const struct qic117_command_table qic117_cmds[] = QIC117_COMMANDS; -int ftape_might_be_off_track; - -/* Local vars. - */ -static int diagnostic_mode; -static unsigned int ftape_udelay_count; -static unsigned int ftape_udelay_time; - -void ftape_udelay(unsigned int usecs) -{ - volatile int count = (ftape_udelay_count * usecs + - ftape_udelay_count - 1) / ftape_udelay_time; - volatile int i; - - while (count-- > 0) { - for (i = 0; i < 20; ++i); - } -} - -void ftape_udelay_calibrate(void) -{ - ftape_calibrate("ftape_udelay", - ftape_udelay, &ftape_udelay_count, &ftape_udelay_time); -} - -/* Delay (msec) routine. - */ -void ftape_sleep(unsigned int time) -{ - TRACE_FUN(ft_t_any); - - time *= 1000; /* msecs -> usecs */ - if (time < FT_USPT) { - /* Time too small for scheduler, do a busy wait ! */ - ftape_udelay(time); - } else { - long timeout; - unsigned long flags; - unsigned int ticks = (time + FT_USPT - 1) / FT_USPT; - - TRACE(ft_t_any, "%d msec, %d ticks", time/1000, ticks); - timeout = ticks; - save_flags(flags); - sti(); - msleep_interruptible(jiffies_to_msecs(timeout)); - /* Mmm. Isn't current->blocked == 0xffffffff ? - */ - if (signal_pending(current)) { - TRACE(ft_t_err, "awoken by non-blocked signal :-("); - } - restore_flags(flags); - } - TRACE_EXIT; -} - -/* send a command or parameter to the drive - * Generates # of step pulses. - */ -static inline int ft_send_to_drive(int arg) -{ - /* Always wait for a command_timeout period to separate - * individuals commands and/or parameters. - */ - ftape_sleep(3 * FT_MILLISECOND); - /* Keep cylinder nr within range, step towards home if possible. - */ - if (ftape_current_cylinder >= arg) { - return fdc_seek(ftape_current_cylinder - arg); - } else { - return fdc_seek(ftape_current_cylinder + arg); - } -} - -/* forward */ int ftape_report_raw_drive_status(int *status); - -static int ft_check_cmd_restrictions(qic117_cmd_t command) -{ - int status = -1; - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "%s", qic117_cmds[command].name); - /* A new motion command during an uninterruptible (motion) - * command requires a ready status before the new command can - * be issued. Otherwise a new motion command needs to be - * checked against required status. - */ - if (qic117_cmds[command].cmd_type == motion && - qic117_cmds[ftape_current_command].non_intr) { - ftape_report_raw_drive_status(&status); - if ((status & QIC_STATUS_READY) == 0) { - TRACE(ft_t_noise, - "motion cmd (%d) during non-intr cmd (%d)", - command, ftape_current_command); - TRACE(ft_t_noise, "waiting until drive gets ready"); - ftape_ready_wait(ftape_timeout.seek, - &status); - } - } - if (qic117_cmds[command].mask != 0) { - __u8 difference; - /* Some commands do require a certain status: - */ - if (status == -1) { /* not yet set */ - ftape_report_raw_drive_status(&status); - } - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - /* Wait until the drive gets - * ready. This may last forever if - * the drive never gets ready... - */ - while ((difference & QIC_STATUS_READY) != 0) { - TRACE(ft_t_noise, "command %d issued while not ready", - command); - TRACE(ft_t_noise, "waiting until drive gets ready"); - if (ftape_ready_wait(ftape_timeout.seek, - &status) == -EINTR) { - /* Bail out on signal ! - */ - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by non-blockable signal"); - } - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - } - while ((difference & QIC_STATUS_ERROR) != 0) { - int err; - qic117_cmd_t cmd; - - TRACE(ft_t_noise, - "command %d issued while error pending", - command); - TRACE(ft_t_noise, "clearing error status"); - ftape_report_error(&err, &cmd, 1); - ftape_report_raw_drive_status(&status); - difference = ((status ^ qic117_cmds[command].state) & - qic117_cmds[command].mask); - if ((difference & QIC_STATUS_ERROR) != 0) { - /* Bail out on fatal signal ! - */ - FT_SIGNAL_EXIT(_NEVER_BLOCK); - } - } - if (difference) { - /* Any remaining difference can't be solved - * here. - */ - if (difference & (QIC_STATUS_CARTRIDGE_PRESENT | - QIC_STATUS_NEW_CARTRIDGE | - QIC_STATUS_REFERENCED)) { - TRACE(ft_t_warn, - "Fatal: tape removed or reinserted !"); - ft_failure = 1; - } else { - TRACE(ft_t_err, "wrong state: 0x%02x should be: 0x%02x", - status & qic117_cmds[command].mask, - qic117_cmds[command].state); - } - TRACE_EXIT -EIO; - } - if (~status & QIC_STATUS_READY & qic117_cmds[command].mask) { - TRACE_ABORT(-EBUSY, ft_t_err, "Bad: still busy!"); - } - } - TRACE_EXIT 0; -} - -/* Issue a tape command: - */ -int ftape_command(qic117_cmd_t command) -{ - int result = 0; - static int level; - TRACE_FUN(ft_t_any); - - if ((unsigned int)command > NR_ITEMS(qic117_cmds)) { - /* This is a bug we'll want to know about too. - */ - TRACE_ABORT(-EIO, ft_t_bug, "bug - bad command: %d", command); - } - if (++level > 5) { /* This is a bug we'll want to know about. */ - --level; - TRACE_ABORT(-EIO, ft_t_bug, "bug - recursion for command: %d", - command); - } - /* disable logging and restriction check for some commands, - * check all other commands that have a prescribed starting - * status. - */ - if (diagnostic_mode) { - TRACE(ft_t_flow, "diagnostic command %d", command); - } else if (command == QIC_REPORT_DRIVE_STATUS || - command == QIC_REPORT_NEXT_BIT) { - TRACE(ft_t_any, "%s", qic117_cmds[command].name); - } else { - TRACE_CATCH(ft_check_cmd_restrictions(command), --level); - } - /* Now all conditions are met or result was < 0. - */ - result = ft_send_to_drive((unsigned int)command); - if (qic117_cmds[command].cmd_type == motion && - command != QIC_LOGICAL_FORWARD && command != QIC_STOP_TAPE) { - ft_location.known = 0; - } - ftape_current_command = command; - --level; - TRACE_EXIT result; -} - -/* Send a tape command parameter: - * Generates command # of step pulses. - * Skips tape-status call ! - */ -int ftape_parameter(unsigned int parameter) -{ - TRACE_FUN(ft_t_any); - - TRACE(ft_t_flow, "called with parameter = %d", parameter); - TRACE_EXIT ft_send_to_drive(parameter + 2); -} - -/* Wait for the drive to get ready. - * timeout time in milli-seconds - * Returned status is valid if result != -EIO - * - * Should we allow to be killed by SIGINT? (^C) - * Would be nice at least for large timeouts. - */ -int ftape_ready_wait(unsigned int timeout, int *status) -{ - unsigned long t0; - unsigned int poll_delay; - int signal_retries; - TRACE_FUN(ft_t_any); - - /* the following ** REALLY ** reduces the system load when - * e.g. one simply rewinds or retensions. The tape is slow - * anyway. It is really not necessary to detect error - * conditions with 1/10 seconds granularity - * - * On my AMD 133MHZ 486: 100 ms: 23% system load - * 1 sec: 5% - * 5 sec: 0.6%, yeah - */ - if (timeout <= FT_SECOND) { - poll_delay = 100 * FT_MILLISECOND; - signal_retries = 20; /* two seconds */ - } else if (timeout < 20 * FT_SECOND) { - TRACE(ft_t_flow, "setting poll delay to 1 second"); - poll_delay = FT_SECOND; - signal_retries = 2; /* two seconds */ - } else { - TRACE(ft_t_flow, "setting poll delay to 5 seconds"); - poll_delay = 5 * FT_SECOND; - signal_retries = 1; /* five seconds */ - } - for (;;) { - t0 = jiffies; - TRACE_CATCH(ftape_report_raw_drive_status(status),); - if (*status & QIC_STATUS_READY) { - TRACE_EXIT 0; - } - if (!signal_retries--) { - FT_SIGNAL_EXIT(_NEVER_BLOCK); - } - if ((int)timeout >= 0) { - /* this will fail when jiffies wraps around about - * once every year :-) - */ - timeout -= ((jiffies - t0) * FT_SECOND) / HZ; - if (timeout <= 0) { - TRACE_ABORT(-ETIME, ft_t_err, "timeout"); - } - ftape_sleep(poll_delay); - timeout -= poll_delay; - } else { - ftape_sleep(poll_delay); - } - } - TRACE_EXIT -ETIME; -} - -/* Issue command and wait up to timeout milli seconds for drive ready - */ -int ftape_command_wait(qic117_cmd_t command, unsigned int timeout, int *status) -{ - int result; - - /* Drive should be ready, issue command - */ - result = ftape_command(command); - if (result >= 0) { - result = ftape_ready_wait(timeout, status); - } - return result; -} - -static int ftape_parameter_wait(unsigned int parm, unsigned int timeout, int *status) -{ - int result; - - /* Drive should be ready, issue command - */ - result = ftape_parameter(parm); - if (result >= 0) { - result = ftape_ready_wait(timeout, status); - } - return result; -} - -/*-------------------------------------------------------------------------- - * Report operations - */ - -/* Query the drive about its status. The command is sent and - result_length bits of status are returned (2 extra bits are read - for start and stop). */ - -int ftape_report_operation(int *status, - qic117_cmd_t command, - int result_length) -{ - int i, st3; - unsigned int t0; - unsigned int dt; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_command(command),); - t0 = ftape_timestamp(); - i = 0; - do { - ++i; - ftape_sleep(3 * FT_MILLISECOND); /* see remark below */ - TRACE_CATCH(fdc_sense_drive_status(&st3),); - dt = ftape_timediff(t0, ftape_timestamp()); - /* Ack should be asserted within Ttimout + Tack = 6 msec. - * Looks like some drives fail to do this so extend this - * period to 300 msec. - */ - } while (!(st3 & ST3_TRACK_0) && dt < 300000); - if (!(st3 & ST3_TRACK_0)) { - TRACE(ft_t_err, - "No acknowledge after %u msec. (%i iter)", dt / 1000, i); - TRACE_ABORT(-EIO, ft_t_err, "timeout on Acknowledge"); - } - /* dt may be larger than expected because of other tasks - * scheduled while we were sleeping. - */ - if (i > 1 && dt > 6000) { - TRACE(ft_t_err, "Acknowledge after %u msec. (%i iter)", - dt / 1000, i); - } - *status = 0; - for (i = 0; i < result_length + 1; i++) { - TRACE_CATCH(ftape_command(QIC_REPORT_NEXT_BIT),); - TRACE_CATCH(fdc_sense_drive_status(&st3),); - if (i < result_length) { - *status |= ((st3 & ST3_TRACK_0) ? 1 : 0) << i; - } else if ((st3 & ST3_TRACK_0) == 0) { - TRACE_ABORT(-EIO, ft_t_err, "missing status stop bit"); - } - } - /* this command will put track zero and index back into normal state */ - (void)ftape_command(QIC_REPORT_NEXT_BIT); - TRACE_EXIT 0; -} - -/* Report the current drive status. */ - -int ftape_report_raw_drive_status(int *status) -{ - int result; - int count = 0; - TRACE_FUN(ft_t_any); - - do { - result = ftape_report_operation(status, - QIC_REPORT_DRIVE_STATUS, 8); - } while (result < 0 && ++count <= 3); - if (result < 0) { - TRACE_ABORT(-EIO, ft_t_err, - "report_operation failed after %d trials", count); - } - if ((*status & 0xff) == 0xff) { - TRACE_ABORT(-EIO, ft_t_err, - "impossible drive status 0xff"); - } - if (*status & QIC_STATUS_READY) { - ftape_current_command = QIC_NO_COMMAND; /* completed */ - } - ft_last_status.status.drive_status = (__u8)(*status & 0xff); - TRACE_EXIT 0; -} - -int ftape_report_drive_status(int *status) -{ - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_report_raw_drive_status(status),); - if (*status & QIC_STATUS_NEW_CARTRIDGE || - !(*status & QIC_STATUS_CARTRIDGE_PRESENT)) { - ft_failure = 1; /* will inhibit further operations */ - TRACE_EXIT -EIO; - } - if (*status & QIC_STATUS_READY && *status & QIC_STATUS_ERROR) { - /* Let caller handle all errors */ - TRACE_ABORT(1, ft_t_warn, "warning: error status set!"); - } - TRACE_EXIT 0; -} - -int ftape_report_error(unsigned int *error, - qic117_cmd_t *command, int report) -{ - static const ftape_error ftape_errors[] = QIC117_ERRORS; - int code; - TRACE_FUN(ft_t_any); - - TRACE_CATCH(ftape_report_operation(&code, QIC_REPORT_ERROR_CODE, 16),); - *error = (unsigned int)(code & 0xff); - *command = (qic117_cmd_t)((code>>8)&0xff); - /* remember hardware status, maybe useful for status ioctls - */ - ft_last_error.error.command = (__u8)*command; - ft_last_error.error.error = (__u8)*error; - if (!report) { - TRACE_EXIT 0; - } - if (*error == 0) { - TRACE_ABORT(0, ft_t_info, "No error"); - } - TRACE(ft_t_info, "errorcode: %d", *error); - if (*error < NR_ITEMS(ftape_errors)) { - TRACE(ft_t_noise, "%sFatal ERROR:", - (ftape_errors[*error].fatal ? "" : "Non-")); - TRACE(ft_t_noise, "%s ...", ftape_errors[*error].message); - } else { - TRACE(ft_t_noise, "Unknown ERROR !"); - } - if ((unsigned int)*command < NR_ITEMS(qic117_cmds) && - qic117_cmds[*command].name != NULL) { - TRACE(ft_t_noise, "... caused by command \'%s\'", - qic117_cmds[*command].name); - } else { - TRACE(ft_t_noise, "... caused by unknown command %d", - *command); - } - TRACE_EXIT 0; -} - -int ftape_report_configuration(qic_model *model, - unsigned int *rate, - int *qic_std, - int *tape_len) -{ - int result; - int config; - int status; - static const unsigned int qic_rates[ 4] = { 250, 2000, 500, 1000 }; - TRACE_FUN(ft_t_any); - - result = ftape_report_operation(&config, - QIC_REPORT_DRIVE_CONFIGURATION, 8); - if (result < 0) { - ft_last_status.status.drive_config = (__u8)0x00; - *model = prehistoric; - *rate = 500; - *qic_std = QIC_TAPE_QIC40; - *tape_len = 205; - TRACE_EXIT 0; - } else { - ft_last_status.status.drive_config = (__u8)(config & 0xff); - } - *rate = qic_rates[(config & QIC_CONFIG_RATE_MASK) >> QIC_CONFIG_RATE_SHIFT]; - result = ftape_report_operation(&status, QIC_REPORT_TAPE_STATUS, 8); - if (result < 0) { - ft_last_status.status.tape_status = (__u8)0x00; - /* pre- QIC117 rev C spec. drive, QIC_CONFIG_80 bit is valid. - */ - *qic_std = (config & QIC_CONFIG_80) ? - QIC_TAPE_QIC80 : QIC_TAPE_QIC40; - /* ?? how's about 425ft tapes? */ - *tape_len = (config & QIC_CONFIG_LONG) ? 307 : 0; - *model = pre_qic117c; - result = 0; - } else { - ft_last_status.status.tape_status = (__u8)(status & 0xff); - *model = post_qic117b; - TRACE(ft_t_any, "report tape status result = %02x", status); - /* post- QIC117 rev C spec. drive, QIC_CONFIG_80 bit is - * invalid. - */ - switch (status & QIC_TAPE_STD_MASK) { - case QIC_TAPE_QIC40: - case QIC_TAPE_QIC80: - case QIC_TAPE_QIC3020: - case QIC_TAPE_QIC3010: - *qic_std = status & QIC_TAPE_STD_MASK; - break; - default: - *qic_std = -1; - break; - } - switch (status & QIC_TAPE_LEN_MASK) { - case QIC_TAPE_205FT: - /* 205 or 425+ ft 550 Oe tape */ - *tape_len = 0; - break; - case QIC_TAPE_307FT: - /* 307.5 ft 550 Oe Extended Length (XL) tape */ - *tape_len = 307; - break; - case QIC_TAPE_VARIABLE: - /* Variable length 550 Oe tape */ - *tape_len = 0; - break; - case QIC_TAPE_1100FT: - /* 1100 ft 550 Oe tape */ - *tape_len = 1100; - break; - case QIC_TAPE_FLEX: - /* Variable length 900 Oe tape */ - *tape_len = 0; - break; - default: - *tape_len = -1; - break; - } - if (*qic_std == -1 || *tape_len == -1) { - TRACE(ft_t_any, - "post qic-117b spec drive with unknown tape"); - } - result = *tape_len == -1 ? -EIO : 0; - if (status & QIC_TAPE_WIDE) { - switch (*qic_std) { - case QIC_TAPE_QIC80: - TRACE(ft_t_info, "TR-1 tape detected"); - break; - case QIC_TAPE_QIC3010: - TRACE(ft_t_info, "TR-2 tape detected"); - break; - case QIC_TAPE_QIC3020: - TRACE(ft_t_info, "TR-3 tape detected"); - break; - default: - TRACE(ft_t_warn, - "Unknown Travan tape type detected"); - break; - } - } - } - TRACE_EXIT (result < 0) ? -EIO : 0; -} - -static int ftape_report_rom_version(int *version) -{ - - if (ftape_report_operation(version, QIC_REPORT_ROM_VERSION, 8) < 0) { - return -EIO; - } else { - return 0; - } -} - -void ftape_report_vendor_id(unsigned int *id) -{ - int result; - TRACE_FUN(ft_t_any); - - /* We'll try to get a vendor id from the drive. First - * according to the QIC-117 spec, a 16-bit id is requested. - * If that fails we'll try an 8-bit version, otherwise we'll - * try an undocumented query. - */ - result = ftape_report_operation((int *) id, QIC_REPORT_VENDOR_ID, 16); - if (result < 0) { - result = ftape_report_operation((int *) id, - QIC_REPORT_VENDOR_ID, 8); - if (result < 0) { - /* The following is an undocumented call found - * in the CMS code. - */ - result = ftape_report_operation((int *) id, 24, 8); - if (result < 0) { - *id = UNKNOWN_VENDOR; - } else { - TRACE(ft_t_noise, "got old 8 bit id: %04x", - *id); - *id |= 0x20000; - } - } else { - TRACE(ft_t_noise, "got 8 bit id: %04x", *id); - *id |= 0x10000; - } - } else { - TRACE(ft_t_noise, "got 16 bit id: %04x", *id); - } - if (*id == 0x0047) { - int version; - int sign; - - if (ftape_report_rom_version(&version) < 0) { - TRACE(ft_t_bug, "report rom version failed"); - TRACE_EXIT; - } - TRACE(ft_t_noise, "CMS rom version: %d", version); - ftape_command(QIC_ENTER_DIAGNOSTIC_1); - ftape_command(QIC_ENTER_DIAGNOSTIC_1); - diagnostic_mode = 1; - if (ftape_report_operation(&sign, 9, 8) < 0) { - unsigned int error; - qic117_cmd_t command; - - ftape_report_error(&error, &command, 1); - ftape_command(QIC_ENTER_PRIMARY_MODE); - diagnostic_mode = 0; - TRACE_EXIT; /* failure ! */ - } else { - TRACE(ft_t_noise, "CMS signature: %02x", sign); - } - if (sign == 0xa5) { - result = ftape_report_operation(&sign, 37, 8); - if (result < 0) { - if (version >= 63) { - *id = 0x8880; - TRACE(ft_t_noise, - "This is an Iomega drive !"); - } else { - *id = 0x0047; - TRACE(ft_t_noise, - "This is a real CMS drive !"); - } - } else { - *id = 0x0047; - TRACE(ft_t_noise, "CMS status: %d", sign); - } - } else { - *id = UNKNOWN_VENDOR; - } - ftape_command(QIC_ENTER_PRIMARY_MODE); - diagnostic_mode = 0; - } - TRACE_EXIT; -} - -static int qic_rate_code(unsigned int rate) -{ - switch (rate) { - case 250: - return QIC_CONFIG_RATE_250; - case 500: - return QIC_CONFIG_RATE_500; - case 1000: - return QIC_CONFIG_RATE_1000; - case 2000: - return QIC_CONFIG_RATE_2000; - default: - return QIC_CONFIG_RATE_500; - } -} - -static int ftape_set_rate_test(unsigned int *max_rate) -{ - unsigned int error; - qic117_cmd_t command; - int status; - int supported = 0; - TRACE_FUN(ft_t_any); - - /* Check if the drive does support the select rate command - * by testing all different settings. If any one is accepted - * we assume the command is supported, else not. - */ - for (*max_rate = 2000; *max_rate >= 250; *max_rate /= 2) { - if (ftape_command(QIC_SELECT_RATE) < 0) { - continue; - } - if (ftape_parameter_wait(qic_rate_code(*max_rate), - 1 * FT_SECOND, &status) < 0) { - continue; - } - if (status & QIC_STATUS_ERROR) { - ftape_report_error(&error, &command, 0); - continue; - } - supported = 1; /* did accept a request */ - break; - } - TRACE(ft_t_noise, "Select Rate command is%s supported", - supported ? "" : " not"); - TRACE_EXIT supported; -} - -int ftape_set_data_rate(unsigned int new_rate /* Kbps */, unsigned int qic_std) -{ - int status; - int result = 0; - unsigned int data_rate = new_rate; - static int supported; - int rate_changed = 0; - qic_model dummy_model; - unsigned int dummy_qic_std, dummy_tape_len; - TRACE_FUN(ft_t_any); - - if (ft_drive_max_rate == 0) { /* first time */ - supported = ftape_set_rate_test(&ft_drive_max_rate); - } - if (supported) { - ftape_command(QIC_SELECT_RATE); - result = ftape_parameter_wait(qic_rate_code(new_rate), - 1 * FT_SECOND, &status); - if (result >= 0 && !(status & QIC_STATUS_ERROR)) { - rate_changed = 1; - } - } - TRACE_CATCH(result = ftape_report_configuration(&dummy_model, - &data_rate, - &dummy_qic_std, - &dummy_tape_len),); - if (data_rate != new_rate) { - if (!supported) { - TRACE(ft_t_warn, "Rate change not supported!"); - } else if (rate_changed) { - TRACE(ft_t_warn, "Requested: %d, got %d", - new_rate, data_rate); - } else { - TRACE(ft_t_warn, "Rate change failed!"); - } - result = -EINVAL; - } - /* - * Set data rate and write precompensation as specified: - * - * | QIC-40/80 | QIC-3010/3020 - * rate | precomp | precomp - * ----------+-------------+-------------- - * 250 Kbps. | 250 ns. | 0 ns. - * 500 Kbps. | 125 ns. | 0 ns. - * 1 Mbps. | 42 ns. | 0 ns. - * 2 Mbps | N/A | 0 ns. - */ - if ((qic_std == QIC_TAPE_QIC40 && data_rate > 500) || - (qic_std == QIC_TAPE_QIC80 && data_rate > 1000)) { - TRACE_ABORT(-EINVAL, - ft_t_warn, "Datarate too high for QIC-mode"); - } - TRACE_CATCH(fdc_set_data_rate(data_rate),_res = -EINVAL); - ft_data_rate = data_rate; - if (qic_std == QIC_TAPE_QIC40 || qic_std == QIC_TAPE_QIC80) { - switch (data_rate) { - case 250: - fdc_set_write_precomp(250); - break; - default: - case 500: - fdc_set_write_precomp(125); - break; - case 1000: - fdc_set_write_precomp(42); - break; - } - } else { - fdc_set_write_precomp(0); - } - TRACE_EXIT result; -} - -/* The next two functions are used to cope with excessive overrun errors - */ -int ftape_increase_threshold(void) -{ - TRACE_FUN(ft_t_flow); - - if (fdc.type < i82077 || ft_fdc_threshold >= 12) { - TRACE_ABORT(-EIO, ft_t_err, "cannot increase fifo threshold"); - } - if (fdc_fifo_threshold(++ft_fdc_threshold, NULL, NULL, NULL) < 0) { - TRACE(ft_t_err, "cannot increase fifo threshold"); - ft_fdc_threshold --; - fdc_reset(); - } - TRACE(ft_t_info, "New FIFO threshold: %d", ft_fdc_threshold); - TRACE_EXIT 0; -} - -int ftape_half_data_rate(void) -{ - if (ft_data_rate < 500) { - return -1; - } - if (ftape_set_data_rate(ft_data_rate / 2, ft_qic_std) < 0) { - return -EIO; - } - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - return 0; -} - -/* Seek the head to the specified track. - */ -int ftape_seek_head_to_track(unsigned int track) -{ - int status; - TRACE_FUN(ft_t_any); - - ft_location.track = -1; /* remains set in case of error */ - if (track >= ft_tracks_per_tape) { - TRACE_ABORT(-EINVAL, ft_t_bug, "track out of bounds"); - } - TRACE(ft_t_flow, "seeking track %d", track); - TRACE_CATCH(ftape_command(QIC_SEEK_HEAD_TO_TRACK),); - TRACE_CATCH(ftape_parameter_wait(track, ftape_timeout.head_seek, - &status),); - ft_location.track = track; - ftape_might_be_off_track = 0; - TRACE_EXIT 0; -} - -int ftape_wakeup_drive(wake_up_types method) -{ - int status; - int motor_on = 0; - TRACE_FUN(ft_t_any); - - switch (method) { - case wake_up_colorado: - TRACE_CATCH(ftape_command(QIC_PHANTOM_SELECT),); - TRACE_CATCH(ftape_parameter(0 /* ft_drive_sel ?? */),); - break; - case wake_up_mountain: - TRACE_CATCH(ftape_command(QIC_SOFT_SELECT),); - ftape_sleep(FT_MILLISECOND); /* NEEDED */ - TRACE_CATCH(ftape_parameter(18),); - break; - case wake_up_insight: - ftape_sleep(100 * FT_MILLISECOND); - motor_on = 1; - fdc_motor(motor_on); /* enable is done by motor-on */ - case no_wake_up: - break; - default: - TRACE_EXIT -ENODEV; /* unknown wakeup method */ - break; - } - /* If wakeup succeeded we shouldn't get an error here.. - */ - TRACE_CATCH(ftape_report_raw_drive_status(&status), - if (motor_on) { - fdc_motor(0); - }); - TRACE_EXIT 0; -} - -int ftape_put_drive_to_sleep(wake_up_types method) -{ - TRACE_FUN(ft_t_any); - - switch (method) { - case wake_up_colorado: - TRACE_CATCH(ftape_command(QIC_PHANTOM_DESELECT),); - break; - case wake_up_mountain: - TRACE_CATCH(ftape_command(QIC_SOFT_DESELECT),); - break; - case wake_up_insight: - fdc_motor(0); /* enable is done by motor-on */ - case no_wake_up: /* no wakeup / no sleep ! */ - break; - default: - TRACE_EXIT -ENODEV; /* unknown wakeup method */ - } - TRACE_EXIT 0; -} - -int ftape_reset_drive(void) -{ - int result = 0; - int status; - unsigned int err_code; - qic117_cmd_t err_command; - int i; - TRACE_FUN(ft_t_any); - - /* We want to re-establish contact with our drive. Fire a - * number of reset commands (single step pulses) and pray for - * success. - */ - for (i = 0; i < 2; ++i) { - TRACE(ft_t_flow, "Resetting fdc"); - fdc_reset(); - ftape_sleep(10 * FT_MILLISECOND); - TRACE(ft_t_flow, "Reset command to drive"); - result = ftape_command(QIC_RESET); - if (result == 0) { - ftape_sleep(1 * FT_SECOND); /* drive not - * accessible - * during 1 second - */ - TRACE(ft_t_flow, "Re-selecting drive"); - - /* Strange, the QIC-117 specs don't mention - * this but the drive gets deselected after a - * soft reset ! So we need to enable it - * again. - */ - if (ftape_wakeup_drive(ft_drive_type.wake_up) < 0) { - TRACE(ft_t_err, "Wakeup failed !"); - } - TRACE(ft_t_flow, "Waiting until drive gets ready"); - result= ftape_ready_wait(ftape_timeout.reset, &status); - if (result == 0 && (status & QIC_STATUS_ERROR)) { - result = ftape_report_error(&err_code, - &err_command, 1); - if (result == 0 && err_code == 27) { - /* Okay, drive saw reset - * command and responded as it - * should - */ - break; - } else { - result = -EIO; - } - } else { - result = -EIO; - } - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - } - if (result != 0) { - TRACE(ft_t_err, "General failure to reset tape drive"); - } else { - /* Restore correct settings: keep original rate - */ - ftape_set_data_rate(ft_data_rate, ft_qic_std); - } - ftape_init_drive_needed = 1; - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/lowlevel/ftape-io.h b/drivers/char/ftape/lowlevel/ftape-io.h deleted file mode 100644 index 26a7baad8717..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-io.h +++ /dev/null @@ -1,90 +0,0 @@ -#ifndef _FTAPE_IO_H -#define _FTAPE_IO_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-io.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:18 $ - * - * This file contains definitions for the glue part of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include -#include - -typedef struct { - unsigned int seek; - unsigned int reset; - unsigned int rewind; - unsigned int head_seek; - unsigned int stop; - unsigned int pause; -} ft_timeout_table; - -typedef enum { - prehistoric, pre_qic117c, post_qic117b, post_qic117d -} qic_model; - -/* - * ftape-io.c defined global vars. - */ -extern ft_timeout_table ftape_timeout; -extern unsigned int ftape_tape_len; -extern volatile qic117_cmd_t ftape_current_command; -extern const struct qic117_command_table qic117_cmds[]; -extern int ftape_might_be_off_track; - -/* - * ftape-io.c defined global functions. - */ -extern void ftape_udelay(unsigned int usecs); -extern void ftape_udelay_calibrate(void); -extern void ftape_sleep(unsigned int time); -extern void ftape_report_vendor_id(unsigned int *id); -extern int ftape_command(qic117_cmd_t command); -extern int ftape_command_wait(qic117_cmd_t command, - unsigned int timeout, - int *status); -extern int ftape_parameter(unsigned int parameter); -extern int ftape_report_operation(int *status, - qic117_cmd_t command, - int result_length); -extern int ftape_report_configuration(qic_model *model, - unsigned int *rate, - int *qic_std, - int *tape_len); -extern int ftape_report_drive_status(int *status); -extern int ftape_report_raw_drive_status(int *status); -extern int ftape_report_status(int *status); -extern int ftape_ready_wait(unsigned int timeout, int *status); -extern int ftape_seek_head_to_track(unsigned int track); -extern int ftape_set_data_rate(unsigned int new_rate, unsigned int qic_std); -extern int ftape_report_error(unsigned int *error, - qic117_cmd_t *command, - int report); -extern int ftape_reset_drive(void); -extern int ftape_put_drive_to_sleep(wake_up_types method); -extern int ftape_wakeup_drive(wake_up_types method); -extern int ftape_increase_threshold(void); -extern int ftape_half_data_rate(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-proc.c b/drivers/char/ftape/lowlevel/ftape-proc.c deleted file mode 100644 index e805b15e0a12..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-proc.c +++ /dev/null @@ -1,214 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-proc.c,v $ - * $Revision: 1.11 $ - * $Date: 1997/10/24 14:47:37 $ - * - * This file contains the procfs interface for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - - * Old code removed, switched to dynamic proc entry. - */ - - -#if defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) - -#include - -#include -#include -#include - -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-proc.h" -#include "../lowlevel/ftape-tracing.h" - -static size_t get_driver_info(char *buf) -{ - const char *debug_level[] = { "bugs" , - "errors", - "warnings", - "informational", - "noisy", - "program flow", - "fdc and dma", - "data flow", - "anything" }; - - return sprintf(buf, - "version : %s\n" - "used data rate: %d kbit/sec\n" - "dma memory : %d kb\n" - "debug messages: %s\n", - FTAPE_VERSION, - ft_data_rate, - FT_BUFF_SIZE * ft_nr_buffers >> 10, - debug_level[TRACE_LEVEL]); -} - -static size_t get_tapedrive_info(char *buf) -{ - return sprintf(buf, - "vendor id : 0x%04x\n" - "drive name: %s\n" - "wind speed: %d ips\n" - "wakeup : %s\n" - "max. rate : %d kbit/sec\n", - ft_drive_type.vendor_id, - ft_drive_type.name, - ft_drive_type.speed, - ((ft_drive_type.wake_up == no_wake_up) - ? "No wakeup needed" : - ((ft_drive_type.wake_up == wake_up_colorado) - ? "Colorado" : - ((ft_drive_type.wake_up == wake_up_mountain) - ? "Mountain" : - ((ft_drive_type.wake_up == wake_up_insight) - ? "Motor on" : - "Unknown")))), - ft_drive_max_rate); -} - -static size_t get_cartridge_info(char *buf) -{ - if (ftape_init_drive_needed) { - return sprintf(buf, "uninitialized\n"); - } - if (ft_no_tape) { - return sprintf(buf, "no cartridge inserted\n"); - } - return sprintf(buf, - "segments : %5d\n" - "tracks : %5d\n" - "length : %5dft\n" - "formatted : %3s\n" - "writable : %3s\n" - "QIC spec. : QIC-%s\n" - "fmt-code : %1d\n", - ft_segments_per_track, - ft_tracks_per_tape, - ftape_tape_len, - (ft_formatted == 1) ? "yes" : "no", - (ft_write_protected == 1) ? "no" : "yes", - ((ft_qic_std == QIC_TAPE_QIC40) ? "40" : - ((ft_qic_std == QIC_TAPE_QIC80) ? "80" : - ((ft_qic_std == QIC_TAPE_QIC3010) ? "3010" : - ((ft_qic_std == QIC_TAPE_QIC3020) ? "3020" : - "???")))), - ft_format_code); -} - -static size_t get_controller_info(char *buf) -{ - const char *fdc_name[] = { "no fdc", - "i8272", - "i82077", - "i82077AA", - "Colorado FC-10 or FC-20", - "i82078", - "i82078_1" }; - - return sprintf(buf, - "FDC type : %s\n" - "FDC base : 0x%03x\n" - "FDC irq : %d\n" - "FDC dma : %d\n" - "FDC thr. : %d\n" - "max. rate : %d kbit/sec\n", - ft_mach2 ? "Mountain MACH-2" : fdc_name[fdc.type], - fdc.sra, fdc.irq, fdc.dma, - ft_fdc_threshold, ft_fdc_max_rate); -} - -static size_t get_history_info(char *buf) -{ - size_t len; - - len = sprintf(buf, - "\nFDC isr statistics\n" - " id_am_errors : %3d\n" - " id_crc_errors : %3d\n" - " data_am_errors : %3d\n" - " data_crc_errors : %3d\n" - " overrun_errors : %3d\n" - " no_data_errors : %3d\n" - " retries : %3d\n", - ft_history.id_am_errors, ft_history.id_crc_errors, - ft_history.data_am_errors, ft_history.data_crc_errors, - ft_history.overrun_errors, ft_history.no_data_errors, - ft_history.retries); - len += sprintf(buf + len, - "\nECC statistics\n" - " crc_errors : %3d\n" - " crc_failures : %3d\n" - " ecc_failures : %3d\n" - " sectors corrected: %3d\n", - ft_history.crc_errors, ft_history.crc_failures, - ft_history.ecc_failures, ft_history.corrected); - len += sprintf(buf + len, - "\ntape quality statistics\n" - " media defects : %3d\n", - ft_history.defects); - len += sprintf(buf + len, - "\ntape motion statistics\n" - " repositions : %3d\n", - ft_history.rewinds); - return len; -} - -static int ftape_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - char *ptr = page; - size_t len; - - ptr += sprintf(ptr, "Kernel Driver\n\n"); - ptr += get_driver_info(ptr); - ptr += sprintf(ptr, "\nTape Drive\n\n"); - ptr += get_tapedrive_info(ptr); - ptr += sprintf(ptr, "\nFDC Controller\n\n"); - ptr += get_controller_info(ptr); - ptr += sprintf(ptr, "\nTape Cartridge\n\n"); - ptr += get_cartridge_info(ptr); - ptr += sprintf(ptr, "\nHistory Record\n\n"); - ptr += get_history_info(ptr); - - len = strlen(page); - *start = NULL; - if (off+count >= len) { - *eof = 1; - } else { - *eof = 0; - } - return len; -} - -int __init ftape_proc_init(void) -{ - return create_proc_read_entry("ftape", 0, &proc_root, - ftape_read_proc, NULL) != NULL; -} - -void ftape_proc_destroy(void) -{ - remove_proc_entry("ftape", &proc_root); -} - -#endif /* defined(CONFIG_PROC_FS) && defined(CONFIG_FT_PROC_FS) */ diff --git a/drivers/char/ftape/lowlevel/ftape-proc.h b/drivers/char/ftape/lowlevel/ftape-proc.h deleted file mode 100644 index 264dfcc1d22d..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-proc.h +++ /dev/null @@ -1,35 +0,0 @@ -#ifndef _FTAPE_PROC_H -#define _FTAPE_PROC_H - -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-proc.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:20 $ - * - * This file contains definitions for the procfs interface of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include - -extern int ftape_proc_init(void); -extern void ftape_proc_destroy(void); - -#endif diff --git a/drivers/char/ftape/lowlevel/ftape-read.c b/drivers/char/ftape/lowlevel/ftape-read.c deleted file mode 100644 index d967d8cd86dc..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-read.c +++ /dev/null @@ -1,621 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-read.c,v $ - * $Revision: 1.6 $ - * $Date: 1997/10/21 14:39:22 $ - * - * This file contains the reading code - * for the QIC-117 floppy-tape driver for Linux. - * - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ - -/* Local vars. - */ - -void ftape_zap_read_buffers(void) -{ - int i; - - for (i = 0; i < ft_nr_buffers; ++i) { -/* changed to "fit" with dynamic allocation of tape_buffer. --khp */ - ft_buffer[i]->status = waiting; - ft_buffer[i]->bytes = 0; - ft_buffer[i]->skip = 0; - ft_buffer[i]->retry = 0; - } -/* ftape_reset_buffer(); */ -} - -static SectorMap convert_sector_map(buffer_struct * buff) -{ - int i = 0; - SectorMap bad_map = ftape_get_bad_sector_entry(buff->segment_id); - SectorMap src_map = buff->soft_error_map | buff->hard_error_map; - SectorMap dst_map = 0; - TRACE_FUN(ft_t_any); - - if (bad_map || src_map) { - TRACE(ft_t_flow, "bad_map = 0x%08lx", (long) bad_map); - TRACE(ft_t_flow, "src_map = 0x%08lx", (long) src_map); - } - while (bad_map) { - while ((bad_map & 1) == 0) { - if (src_map & 1) { - dst_map |= (1 << i); - } - src_map >>= 1; - bad_map >>= 1; - ++i; - } - /* (bad_map & 1) == 1 */ - src_map >>= 1; - bad_map >>= 1; - } - if (src_map) { - dst_map |= (src_map << i); - } - if (dst_map) { - TRACE(ft_t_flow, "dst_map = 0x%08lx", (long) dst_map); - } - TRACE_EXIT dst_map; -} - -static int correct_and_copy_fraction(buffer_struct *buff, __u8 * destination, - int start, int size) -{ - struct memory_segment mseg; - int result; - SectorMap read_bad; - TRACE_FUN(ft_t_any); - - mseg.read_bad = convert_sector_map(buff); - mseg.marked_bad = 0; /* not used... */ - mseg.blocks = buff->bytes / FT_SECTOR_SIZE; - mseg.data = buff->address; - /* If there are no data sectors we can skip this segment. - */ - if (mseg.blocks <= 3) { - TRACE_ABORT(0, ft_t_noise, "empty segment"); - } - read_bad = mseg.read_bad; - ft_history.crc_errors += count_ones(read_bad); - result = ftape_ecc_correct_data(&mseg); - if (read_bad != 0 || mseg.corrected != 0) { - TRACE(ft_t_noise, "crc error map: 0x%08lx", (unsigned long)read_bad); - TRACE(ft_t_noise, "corrected map: 0x%08lx", (unsigned long)mseg.corrected); - ft_history.corrected += count_ones(mseg.corrected); - } - if (result == ECC_CORRECTED || result == ECC_OK) { - if (result == ECC_CORRECTED) { - TRACE(ft_t_info, "ecc corrected segment: %d", buff->segment_id); - } - if(start < 0) { - start= 0; - } - if((start+size) > ((mseg.blocks - 3) * FT_SECTOR_SIZE)) { - size = (mseg.blocks - 3) * FT_SECTOR_SIZE - start; - } - if (size < 0) { - size= 0; - } - if(size > 0) { - memcpy(destination + start, mseg.data + start, size); - } - if ((read_bad ^ mseg.corrected) & mseg.corrected) { - /* sectors corrected without crc errors set */ - ft_history.crc_failures++; - } - TRACE_EXIT size; /* (mseg.blocks - 3) * FT_SECTOR_SIZE; */ - } else { - ft_history.ecc_failures++; - TRACE_ABORT(-EAGAIN, - ft_t_err, "ecc failure on segment %d", - buff->segment_id); - } - TRACE_EXIT 0; -} - -/* Read given segment into buffer at address. - */ -int ftape_read_segment_fraction(const int segment_id, - void *address, - const ft_read_mode_t read_mode, - const int start, - const int size) -{ - int result = 0; - int retry = 0; - int bytes_read = 0; - int read_done = 0; - TRACE_FUN(ft_t_flow); - - ft_history.used |= 1; - TRACE(ft_t_data_flow, "segment_id = %d", segment_id); - if (ft_driver_state != reading) { - TRACE(ft_t_noise, "calling ftape_abort_operation"); - TRACE_CATCH(ftape_abort_operation(),); - ftape_set_state(reading); - } - for(;;) { - buffer_struct *tail; - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* Search all full buffers for the first matching the - * wanted segment. Clear other buffers on the fly. - */ - tail = ftape_get_buffer(ft_queue_tail); - while (!read_done && tail->status == done) { - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (tail->segment_id == segment_id) { - /* If out buffer is already full, - * return its contents. - */ - TRACE(ft_t_flow, "found segment in cache: %d", - segment_id); - if (tail->deleted) { - /* Return a value that - * read_header_segment - * understands. As this - * should only occur when - * searching for the header - * segments it shouldn't be - * misinterpreted elsewhere. - */ - TRACE_EXIT 0; - } - result = correct_and_copy_fraction( - tail, - address, - start, - size); - TRACE(ft_t_flow, "segment contains (bytes): %d", - result); - if (result < 0) { - if (result != -EAGAIN) { - TRACE_EXIT result; - } - /* keep read_done == 0, will - * trigger - * ftape_abort_operation - * because reading wrong - * segment. - */ - TRACE(ft_t_err, "ecc failed, retry"); - ++retry; - } else { - read_done = 1; - bytes_read = result; - } - } else { - TRACE(ft_t_flow,"zapping segment in cache: %d", - tail->segment_id); - } - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - if (!read_done && tail->status == reading) { - if (tail->segment_id == segment_id) { - switch(ftape_wait_segment(reading)) { - case 0: - break; - case -EINTR: - TRACE_ABORT(-EINTR, ft_t_warn, - "interrupted by " - "non-blockable signal"); - break; - default: - TRACE(ft_t_noise, - "wait_segment failed"); - ftape_abort_operation(); - ftape_set_state(reading); - break; - } - } else { - /* We're reading the wrong segment, - * stop runner. - */ - TRACE(ft_t_noise, "reading wrong segment"); - ftape_abort_operation(); - ftape_set_state(reading); - } - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - switch(head->status) { - case error: - ft_history.defects += - count_ones(head->hard_error_map); - case reading: - head->status = waiting; - break; - default: - break; - } - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait - * for BOT or EOT mark. Sets ft_runner_status to - * idle if at lEOT and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - /* If we got a segment: quit, or else retry up to limit. - * - * If segment to read is empty, do not start runner for it, - * but wait for next read call. - */ - if (read_done || - ftape_get_bad_sector_entry(segment_id) == EMPTY_SEGMENT ) { - /* bytes_read = 0; should still be zero */ - TRACE_EXIT bytes_read; - - } - if (retry > FT_RETRIES_ON_ECC_ERROR) { - ft_history.defects++; - TRACE_ABORT(-ENODATA, ft_t_err, - "too many retries on ecc failure"); - } - /* Now at least one buffer is empty ! - * Restart runner & tape if needed. - */ - TRACE(ft_t_any, "head: %d, tail: %d, ft_runner_status: %d", - ftape_buffer_id(ft_queue_head), - ftape_buffer_id(ft_queue_tail), - ft_runner_status); - TRACE(ft_t_any, "buffer[].status, [head]: %d, [tail]: %d", - ftape_get_buffer(ft_queue_head)->status, - ftape_get_buffer(ft_queue_tail)->status); - tail = ftape_get_buffer(ft_queue_tail); - if (tail->status == waiting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - - ftape_setup_new_segment(head, segment_id, -1); - if (read_mode == FT_RD_SINGLE) { - /* disable read-ahead */ - head->next_segment = 0; - } - ftape_calc_next_cluster(head); - if (ft_runner_status == idle) { - result = ftape_start_tape(segment_id, - head->sector_offset); - if (result < 0) { - TRACE_ABORT(result, ft_t_err, "Error: " - "segment %d unreachable", - segment_id); - } - } - head->status = reading; - fdc_setup_read_write(head, FDC_READ); - } - } - /* not reached */ - TRACE_EXIT -EIO; -} - -int ftape_read_header_segment(__u8 *address) -{ - int result; - int header_segment; - int first_failed = 0; - int status; - TRACE_FUN(ft_t_flow); - - ft_used_header_segment = -1; - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_flow, "reading..."); - /* We're looking for the first header segment. - * A header segment cannot contain bad sectors, therefor at the - * tape start, segments with bad sectors are (according to QIC-40/80) - * written with deleted data marks and must be skipped. - */ - memset(address, '\0', (FT_SECTORS_PER_SEGMENT - 3) * FT_SECTOR_SIZE); - result = 0; -#define HEADER_SEGMENT_BOUNDARY 68 /* why not 42? */ - for (header_segment = 0; - header_segment < HEADER_SEGMENT_BOUNDARY && result == 0; - ++header_segment) { - /* Set no read-ahead, the isr will force read-ahead whenever - * it encounters deleted data ! - */ - result = ftape_read_segment(header_segment, - address, - FT_RD_SINGLE); - if (result < 0 && !first_failed) { - TRACE(ft_t_err, "header segment damaged, trying backup"); - first_failed = 1; - result = 0; /* force read of next (backup) segment */ - } - } - if (result < 0 || header_segment >= HEADER_SEGMENT_BOUNDARY) { - TRACE_ABORT(-EIO, ft_t_err, - "no readable header segment found"); - } - TRACE_CATCH(ftape_abort_operation(),); - ft_used_header_segment = header_segment; - result = ftape_decode_header_segment(address); - TRACE_EXIT result; -} - -int ftape_decode_header_segment(__u8 *address) -{ - unsigned int max_floppy_side; - unsigned int max_floppy_track; - unsigned int max_floppy_sector; - unsigned int new_tape_len; - TRACE_FUN(ft_t_flow); - - if (GET4(address, FT_SIGNATURE) == FT_D2G_MAGIC) { - /* Ditto 2GB header segment. They encrypt the bad sector map. - * We decrypt it and store them in normal format. - * I hope this is correct. - */ - int i; - TRACE(ft_t_warn, - "Found Ditto 2GB tape, " - "trying to decrypt bad sector map"); - for (i=256; i < 29 * FT_SECTOR_SIZE; i++) { - address[i] = ~(address[i] - (i&0xff)); - } - PUT4(address, 0,FT_HSEG_MAGIC); - } else if (GET4(address, FT_SIGNATURE) != FT_HSEG_MAGIC) { - TRACE_ABORT(-EIO, ft_t_err, - "wrong signature in header segment"); - } - ft_format_code = (ft_format_type) address[FT_FMT_CODE]; - if (ft_format_code != fmt_big) { - ft_header_segment_1 = GET2(address, FT_HSEG_1); - ft_header_segment_2 = GET2(address, FT_HSEG_2); - ft_first_data_segment = GET2(address, FT_FRST_SEG); - ft_last_data_segment = GET2(address, FT_LAST_SEG); - } else { - ft_header_segment_1 = GET4(address, FT_6_HSEG_1); - ft_header_segment_2 = GET4(address, FT_6_HSEG_2); - ft_first_data_segment = GET4(address, FT_6_FRST_SEG); - ft_last_data_segment = GET4(address, FT_6_LAST_SEG); - } - TRACE(ft_t_noise, "first data segment: %d", ft_first_data_segment); - TRACE(ft_t_noise, "last data segment: %d", ft_last_data_segment); - TRACE(ft_t_noise, "header segments are %d and %d", - ft_header_segment_1, ft_header_segment_2); - - /* Verify tape parameters... - * QIC-40/80 spec: tape_parameters: - * - * segments-per-track segments_per_track - * tracks-per-cartridge tracks_per_tape - * max-floppy-side (segments_per_track * - * tracks_per_tape - 1) / - * ftape_segments_per_head - * max-floppy-track ftape_segments_per_head / - * ftape_segments_per_cylinder - 1 - * max-floppy-sector ftape_segments_per_cylinder * - * FT_SECTORS_PER_SEGMENT - */ - ft_segments_per_track = GET2(address, FT_SPT); - ft_tracks_per_tape = address[FT_TPC]; - max_floppy_side = address[FT_FHM]; - max_floppy_track = address[FT_FTM]; - max_floppy_sector = address[FT_FSM]; - TRACE(ft_t_noise, "(fmt/spt/tpc/fhm/ftm/fsm) = %d/%d/%d/%d/%d/%d", - ft_format_code, ft_segments_per_track, ft_tracks_per_tape, - max_floppy_side, max_floppy_track, max_floppy_sector); - new_tape_len = ftape_tape_len; - switch (ft_format_code) { - case fmt_425ft: - new_tape_len = 425; - break; - case fmt_normal: - if (ftape_tape_len == 0) { /* otherwise 307 ft */ - new_tape_len = 205; - } - break; - case fmt_1100ft: - new_tape_len = 1100; - break; - case fmt_var:{ - int segments_per_1000_inch = 1; /* non-zero default for switch */ - switch (ft_qic_std) { - case QIC_TAPE_QIC40: - segments_per_1000_inch = 332; - break; - case QIC_TAPE_QIC80: - segments_per_1000_inch = 488; - break; - case QIC_TAPE_QIC3010: - segments_per_1000_inch = 730; - break; - case QIC_TAPE_QIC3020: - segments_per_1000_inch = 1430; - break; - } - new_tape_len = (1000 * ft_segments_per_track + - (segments_per_1000_inch - 1)) / segments_per_1000_inch; - break; - } - case fmt_big:{ - int segments_per_1000_inch = 1; /* non-zero default for switch */ - switch (ft_qic_std) { - case QIC_TAPE_QIC40: - segments_per_1000_inch = 332; - break; - case QIC_TAPE_QIC80: - segments_per_1000_inch = 488; - break; - case QIC_TAPE_QIC3010: - segments_per_1000_inch = 730; - break; - case QIC_TAPE_QIC3020: - segments_per_1000_inch = 1430; - break; - default: - TRACE_ABORT(-EIO, ft_t_bug, - "%x QIC-standard with fmt-code %d, please report", - ft_qic_std, ft_format_code); - } - new_tape_len = ((1000 * ft_segments_per_track + - (segments_per_1000_inch - 1)) / - segments_per_1000_inch); - break; - } - default: - TRACE_ABORT(-EIO, ft_t_err, - "unknown tape format, please report !"); - } - if (new_tape_len != ftape_tape_len) { - ftape_tape_len = new_tape_len; - TRACE(ft_t_info, "calculated tape length is %d ft", - ftape_tape_len); - ftape_calc_timeouts(ft_qic_std, ft_data_rate, ftape_tape_len); - } - if (ft_segments_per_track == 0 && ft_tracks_per_tape == 0 && - max_floppy_side == 0 && max_floppy_track == 0 && - max_floppy_sector == 0) { - /* QIC-40 Rev E and earlier has no values in the header. - */ - ft_segments_per_track = 68; - ft_tracks_per_tape = 20; - max_floppy_side = 1; - max_floppy_track = 169; - max_floppy_sector = 128; - } - /* This test will compensate for the wrong parameter on tapes - * formatted by Conner software. - */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 7 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous CONNER bug: max_floppy_side off by one !"); - max_floppy_side = 6; - } - /* These tests will compensate for the wrong parameter on tapes - * formatted by ComByte Windows software. - * - * First, for 205 foot tapes - */ - if (ft_segments_per_track == 100 && - ft_tracks_per_tape == 28 && - max_floppy_side == 9 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the ComByte bug: max_floppy_side incorrect!"); - max_floppy_side = 4; - } - /* Next, for 307 foot tapes. */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 9 && - max_floppy_track == 149 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the ComByte bug: max_floppy_side incorrect!"); - max_floppy_side = 6; - } - /* This test will compensate for the wrong parameter on tapes - * formatted by Colorado Windows software. - */ - if (ft_segments_per_track == 150 && - ft_tracks_per_tape == 28 && - max_floppy_side == 6 && - max_floppy_track == 150 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous Colorado bug: max_floppy_track off by one !"); - max_floppy_track = 149; - } - ftape_segments_per_head = ((max_floppy_sector/FT_SECTORS_PER_SEGMENT) * - (max_floppy_track + 1)); - /* This test will compensate for some bug reported by Dima - * Brodsky. Seems to be a Colorado bug, either. (freebee - * Imation tape shipped together with Colorado T3000 - */ - if ((ft_format_code == fmt_var || ft_format_code == fmt_big) && - ft_tracks_per_tape == 50 && - max_floppy_side == 54 && - max_floppy_track == 255 && - max_floppy_sector == 128) { -TRACE(ft_t_info, "the famous ??? bug: max_floppy_track off by one !"); - max_floppy_track = 254; - } - /* - * Verify drive_configuration with tape parameters - */ - if (ftape_segments_per_head == 0 || ftape_segments_per_cylinder == 0 || - ((ft_segments_per_track * ft_tracks_per_tape - 1) / ftape_segments_per_head - != max_floppy_side) || - (ftape_segments_per_head / ftape_segments_per_cylinder - 1 != max_floppy_track) || - (ftape_segments_per_cylinder * FT_SECTORS_PER_SEGMENT != max_floppy_sector) -#ifdef TESTING - || ((ft_format_code == fmt_var || ft_format_code == fmt_big) && - (max_floppy_track != 254 || max_floppy_sector != 128)) -#endif - ) { - char segperheadz = ftape_segments_per_head ? ' ' : '?'; - char segpercylz = ftape_segments_per_cylinder ? ' ' : '?'; - TRACE(ft_t_err,"Tape parameters inconsistency, please report"); - TRACE(ft_t_err, "reported = %d/%d/%d/%d/%d/%d", - ft_format_code, - ft_segments_per_track, - ft_tracks_per_tape, - max_floppy_side, - max_floppy_track, - max_floppy_sector); - TRACE(ft_t_err, "required = %d/%d/%d/%d%c/%d%c/%d", - ft_format_code, - ft_segments_per_track, - ft_tracks_per_tape, - ftape_segments_per_head ? - ((ft_segments_per_track * ft_tracks_per_tape -1) / - ftape_segments_per_head ) : - (ft_segments_per_track * ft_tracks_per_tape -1), - segperheadz, - ftape_segments_per_cylinder ? - (ftape_segments_per_head / - ftape_segments_per_cylinder - 1 ) : - ftape_segments_per_head - 1, - segpercylz, - (ftape_segments_per_cylinder * FT_SECTORS_PER_SEGMENT)); - TRACE_EXIT -EIO; - } - ftape_extract_bad_sector_map(address); - TRACE_EXIT 0; -} diff --git a/drivers/char/ftape/lowlevel/ftape-read.h b/drivers/char/ftape/lowlevel/ftape-read.h deleted file mode 100644 index 069f99f2a984..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-read.h +++ /dev/null @@ -1,51 +0,0 @@ -#ifndef _FTAPE_READ_H -#define _FTAPE_READ_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-read.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:22 $ - * - * This file contains the definitions for the read functions - * for the QIC-117 floppy-tape driver for Linux. - * - */ - -/* ftape-read.c defined global functions. - */ -typedef enum { - FT_RD_SINGLE = 0, - FT_RD_AHEAD = 1, -} ft_read_mode_t; - -extern int ftape_read_header_segment(__u8 *address); -extern int ftape_decode_header_segment(__u8 *address); -extern int ftape_read_segment_fraction(const int segment, - void *address, - const ft_read_mode_t read_mode, - const int start, - const int size); -#define ftape_read_segment(segment, address, read_mode) \ - ftape_read_segment_fraction(segment, address, read_mode, \ - 0, FT_SEGMENT_SIZE) -extern void ftape_zap_read_buffers(void); - -#endif /* _FTAPE_READ_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-rw.c b/drivers/char/ftape/lowlevel/ftape-rw.c deleted file mode 100644 index c0d6dc2cbfd3..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-rw.c +++ /dev/null @@ -1,1092 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-rw.c,v $ - * $Revision: 1.7 $ - * $Date: 1997/10/28 14:26:49 $ - * - * This file contains some common code for the segment read and - * segment write routines for the QIC-117 floppy-tape driver for - * Linux. - */ - -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" - -/* Global vars. - */ -int ft_nr_buffers; -buffer_struct *ft_buffer[FT_MAX_NR_BUFFERS]; -static volatile int ft_head; -static volatile int ft_tail; /* not volatile but need same type as head */ -int fdc_setup_error; -location_record ft_location = {-1, 0}; -volatile int ftape_tape_running; - -/* Local vars. - */ -static int overrun_count_offset; -static int inhibit_correction; - -/* maxmimal allowed overshoot when fast seeking - */ -#define OVERSHOOT_LIMIT 10 - -/* Increment cyclic buffer nr. - */ -buffer_struct *ftape_next_buffer(ft_buffer_queue_t pos) -{ - switch (pos) { - case ft_queue_head: - if (++ft_head >= ft_nr_buffers) { - ft_head = 0; - } - return ft_buffer[ft_head]; - case ft_queue_tail: - if (++ft_tail >= ft_nr_buffers) { - ft_tail = 0; - } - return ft_buffer[ft_tail]; - default: - return NULL; - } -} -int ftape_buffer_id(ft_buffer_queue_t pos) -{ - switch(pos) { - case ft_queue_head: return ft_head; - case ft_queue_tail: return ft_tail; - default: return -1; - } -} -buffer_struct *ftape_get_buffer(ft_buffer_queue_t pos) -{ - switch(pos) { - case ft_queue_head: return ft_buffer[ft_head]; - case ft_queue_tail: return ft_buffer[ft_tail]; - default: return NULL; - } -} -void ftape_reset_buffer(void) -{ - ft_head = ft_tail = 0; -} - -buffer_state_enum ftape_set_state(buffer_state_enum new_state) -{ - buffer_state_enum old_state = ft_driver_state; - - ft_driver_state = new_state; - return old_state; -} -/* Calculate Floppy Disk Controller and DMA parameters for a segment. - * head: selects buffer struct in array. - * offset: number of physical sectors to skip (including bad ones). - * count: number of physical sectors to handle (including bad ones). - */ -static int setup_segment(buffer_struct * buff, - int segment_id, - unsigned int sector_offset, - unsigned int sector_count, - int retry) -{ - SectorMap offset_mask; - SectorMap mask; - TRACE_FUN(ft_t_any); - - buff->segment_id = segment_id; - buff->sector_offset = sector_offset; - buff->remaining = sector_count; - buff->head = segment_id / ftape_segments_per_head; - buff->cyl = (segment_id % ftape_segments_per_head) / ftape_segments_per_cylinder; - buff->sect = (segment_id % ftape_segments_per_cylinder) * FT_SECTORS_PER_SEGMENT + 1; - buff->deleted = 0; - offset_mask = (1 << buff->sector_offset) - 1; - mask = ftape_get_bad_sector_entry(segment_id) & offset_mask; - while (mask) { - if (mask & 1) { - offset_mask >>= 1; /* don't count bad sector */ - } - mask >>= 1; - } - buff->data_offset = count_ones(offset_mask); /* good sectors to skip */ - buff->ptr = buff->address + buff->data_offset * FT_SECTOR_SIZE; - TRACE(ft_t_flow, "data offset = %d sectors", buff->data_offset); - if (retry) { - buff->soft_error_map &= offset_mask; /* keep skipped part */ - } else { - buff->hard_error_map = buff->soft_error_map = 0; - } - buff->bad_sector_map = ftape_get_bad_sector_entry(buff->segment_id); - if (buff->bad_sector_map != 0) { - TRACE(ft_t_noise, "segment: %d, bad sector map: %08lx", - buff->segment_id, (long)buff->bad_sector_map); - } else { - TRACE(ft_t_flow, "segment: %d", buff->segment_id); - } - if (buff->sector_offset > 0) { - buff->bad_sector_map >>= buff->sector_offset; - } - if (buff->sector_offset != 0 || buff->remaining != FT_SECTORS_PER_SEGMENT) { - TRACE(ft_t_flow, "sector offset = %d, count = %d", - buff->sector_offset, buff->remaining); - } - /* Segments with 3 or less sectors are not written with valid - * data because there is no space left for the ecc. The - * data written is whatever happens to be in the buffer. - * Reading such a segment will return a zero byte-count. - * To allow us to read/write segments with all bad sectors - * we fake one readable sector in the segment. This - * prevents having to handle these segments in a very - * special way. It is not important if the reading of this - * bad sector fails or not (the data is ignored). It is - * only read to keep the driver running. - * - * The QIC-40/80 spec. has no information on how to handle - * this case, so this is my interpretation. - */ - if (buff->bad_sector_map == EMPTY_SEGMENT) { - TRACE(ft_t_flow, "empty segment %d, fake first sector good", - buff->segment_id); - if (buff->ptr != buff->address) { - TRACE(ft_t_bug, "This is a bug: %p/%p", - buff->ptr, buff->address); - } - buff->bad_sector_map = FAKE_SEGMENT; - } - fdc_setup_error = 0; - buff->next_segment = segment_id + 1; - TRACE_EXIT 0; -} - -/* Calculate Floppy Disk Controller and DMA parameters for a new segment. - */ -int ftape_setup_new_segment(buffer_struct * buff, int segment_id, int skip) -{ - int result = 0; - static int old_segment_id = -1; - static buffer_state_enum old_ft_driver_state = idle; - int retry = 0; - unsigned offset = 0; - int count = FT_SECTORS_PER_SEGMENT; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_flow, "%s segment %d (old = %d)", - (ft_driver_state == reading || ft_driver_state == verifying) - ? "reading" : "writing", - segment_id, old_segment_id); - if (ft_driver_state != old_ft_driver_state) { /* when verifying */ - old_segment_id = -1; - old_ft_driver_state = ft_driver_state; - } - if (segment_id == old_segment_id) { - ++buff->retry; - ++ft_history.retries; - TRACE(ft_t_flow, "setting up for retry nr %d", buff->retry); - retry = 1; - if (skip && buff->skip > 0) { /* allow skip on retry */ - offset = buff->skip; - count -= offset; - TRACE(ft_t_flow, "skipping %d sectors", offset); - } - } else { - buff->retry = 0; - buff->skip = 0; - old_segment_id = segment_id; - } - result = setup_segment(buff, segment_id, offset, count, retry); - TRACE_EXIT result; -} - -/* Determine size of next cluster of good sectors. - */ -int ftape_calc_next_cluster(buffer_struct * buff) -{ - /* Skip bad sectors. - */ - while (buff->remaining > 0 && (buff->bad_sector_map & 1) != 0) { - buff->bad_sector_map >>= 1; - ++buff->sector_offset; - --buff->remaining; - } - /* Find next cluster of good sectors - */ - if (buff->bad_sector_map == 0) { /* speed up */ - buff->sector_count = buff->remaining; - } else { - SectorMap map = buff->bad_sector_map; - - buff->sector_count = 0; - while (buff->sector_count < buff->remaining && (map & 1) == 0) { - ++buff->sector_count; - map >>= 1; - } - } - return buff->sector_count; -} - -/* if just passed the last segment on a track, wait for BOT - * or EOT mark. - */ -int ftape_handle_logical_eot(void) -{ - TRACE_FUN(ft_t_flow); - - if (ft_runner_status == logical_eot) { - int status; - - TRACE(ft_t_noise, "tape at logical EOT"); - TRACE_CATCH(ftape_ready_wait(ftape_timeout.seek, &status),); - if ((status & (QIC_STATUS_AT_BOT | QIC_STATUS_AT_EOT)) == 0) { - TRACE_ABORT(-EIO, ft_t_err, "eot/bot not reached"); - } - ft_runner_status = end_of_tape; - } - if (ft_runner_status == end_of_tape) { - TRACE(ft_t_noise, "runner stopped because of logical EOT"); - ft_runner_status = idle; - } - TRACE_EXIT 0; -} - -static int check_bot_eot(int status) -{ - TRACE_FUN(ft_t_flow); - - if (status & (QIC_STATUS_AT_BOT | QIC_STATUS_AT_EOT)) { - ft_location.bot = ((ft_location.track & 1) == 0 ? - (status & QIC_STATUS_AT_BOT) != 0: - (status & QIC_STATUS_AT_EOT) != 0); - ft_location.eot = !ft_location.bot; - ft_location.segment = (ft_location.track + - (ft_location.bot ? 0 : 1)) * ft_segments_per_track - 1; - ft_location.sector = -1; - ft_location.known = 1; - TRACE(ft_t_flow, "tape at logical %s", - ft_location.bot ? "bot" : "eot"); - TRACE(ft_t_flow, "segment = %d", ft_location.segment); - } else { - ft_location.known = 0; - } - TRACE_EXIT ft_location.known; -} - -/* Read Id of first sector passing tape head. - */ -static int ftape_read_id(void) -{ - int status; - __u8 out[2]; - TRACE_FUN(ft_t_any); - - /* Assume tape is running on entry, be able to handle - * situation where it stopped or is stopping. - */ - ft_location.known = 0; /* default is location not known */ - out[0] = FDC_READID; - out[1] = ft_drive_sel; - TRACE_CATCH(fdc_command(out, 2),); - switch (fdc_interrupt_wait(20 * FT_SECOND)) { - case 0: - if (fdc_sect == 0) { - if (ftape_report_drive_status(&status) >= 0 && - (status & QIC_STATUS_READY)) { - ftape_tape_running = 0; - TRACE(ft_t_flow, "tape has stopped"); - check_bot_eot(status); - } - } else { - ft_location.known = 1; - ft_location.segment = (ftape_segments_per_head - * fdc_head - + ftape_segments_per_cylinder - * fdc_cyl - + (fdc_sect - 1) - / FT_SECTORS_PER_SEGMENT); - ft_location.sector = ((fdc_sect - 1) - % FT_SECTORS_PER_SEGMENT); - ft_location.eot = ft_location.bot = 0; - } - break; - case -ETIME: - /* Didn't find id on tape, must be near end: Wait - * until stopped. - */ - if (ftape_ready_wait(FT_FOREVER, &status) >= 0) { - ftape_tape_running = 0; - TRACE(ft_t_flow, "tape has stopped"); - check_bot_eot(status); - } - break; - default: - /* Interrupted or otherwise failing - * fdc_interrupt_wait() - */ - TRACE(ft_t_err, "fdc_interrupt_wait failed"); - break; - } - if (!ft_location.known) { - TRACE_ABORT(-EIO, ft_t_flow, "no id found"); - } - if (ft_location.sector == 0) { - TRACE(ft_t_flow, "passing segment %d/%d", - ft_location.segment, ft_location.sector); - } else { - TRACE(ft_t_fdc_dma, "passing segment %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int logical_forward(void) -{ - ftape_tape_running = 1; - return ftape_command(QIC_LOGICAL_FORWARD); -} - -int ftape_stop_tape(int *pstatus) -{ - int retry = 0; - int result; - TRACE_FUN(ft_t_flow); - - do { - result = ftape_command_wait(QIC_STOP_TAPE, - ftape_timeout.stop, pstatus); - if (result == 0) { - if ((*pstatus & QIC_STATUS_READY) == 0) { - result = -EIO; - } else { - ftape_tape_running = 0; - } - } - } while (result < 0 && ++retry <= 3); - if (result < 0) { - TRACE(ft_t_err, "failed ! (fatal)"); - } - TRACE_EXIT result; -} - -int ftape_dumb_stop(void) -{ - int result; - int status; - TRACE_FUN(ft_t_flow); - - /* Abort current fdc operation if it's busy (probably read - * or write operation pending) with a reset. - */ - if (fdc_ready_wait(100 /* usec */) < 0) { - TRACE(ft_t_noise, "aborting fdc operation"); - fdc_reset(); - } - /* Reading id's after the last segment on a track may fail - * but eventually the drive will become ready (logical eot). - */ - result = ftape_report_drive_status(&status); - ft_location.known = 0; - do { - if (result == 0 && status & QIC_STATUS_READY) { - /* Tape is not running any more. - */ - TRACE(ft_t_noise, "tape already halted"); - check_bot_eot(status); - ftape_tape_running = 0; - } else if (ftape_tape_running) { - /* Tape is (was) still moving. - */ -#ifdef TESTING - ftape_read_id(); -#endif - result = ftape_stop_tape(&status); - } else { - /* Tape not yet ready but stopped. - */ - result = ftape_ready_wait(ftape_timeout.pause,&status); - } - } while (ftape_tape_running - && !(sigtestsetmask(¤t->pending.signal, _NEVER_BLOCK))); -#ifndef TESTING - ft_location.known = 0; -#endif - if (ft_runner_status == aborting || ft_runner_status == do_abort) { - ft_runner_status = idle; - } - TRACE_EXIT result; -} - -/* Wait until runner has finished tail buffer. - * - */ -int ftape_wait_segment(buffer_state_enum state) -{ - int status; - int result = 0; - TRACE_FUN(ft_t_flow); - - while (ft_buffer[ft_tail]->status == state) { - TRACE(ft_t_flow, "state: %d", ft_buffer[ft_tail]->status); - /* First buffer still being worked on, wait up to timeout. - * - * Note: we check two times for being killed. 50 - * seconds are quite long. Note that - * fdc_interrupt_wait() is not killable by any - * means. ftape_read_segment() wants us to return - * -EINTR in case of a signal. - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - result = fdc_interrupt_wait(50 * FT_SECOND); - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (result < 0) { - TRACE_ABORT(result, - ft_t_err, "fdc_interrupt_wait failed"); - } - if (fdc_setup_error) { - /* recover... FIXME */ - TRACE_ABORT(-EIO, ft_t_err, "setup error"); - } - } - if (ft_buffer[ft_tail]->status != error) { - TRACE_EXIT 0; - } - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_noise, "ftape_report_drive_status: 0x%02x", status); - if ((status & QIC_STATUS_READY) && - (status & QIC_STATUS_ERROR)) { - unsigned int error; - qic117_cmd_t command; - - /* Report and clear error state. - * In case the drive can't operate at the selected - * rate, select the next lower data rate. - */ - ftape_report_error(&error, &command, 1); - if (error == 31 && command == QIC_LOGICAL_FORWARD) { - /* drive does not accept this data rate */ - if (ft_data_rate > 250) { - TRACE(ft_t_info, - "Probable data rate conflict"); - TRACE(ft_t_info, - "Lowering data rate to %d Kbps", - ft_data_rate / 2); - ftape_half_data_rate(); - if (ft_buffer[ft_tail]->retry > 0) { - /* give it a chance */ - --ft_buffer[ft_tail]->retry; - } - } else { - /* no rate is accepted... */ - TRACE(ft_t_err, "We're dead :("); - } - } else { - TRACE(ft_t_err, "Unknown error"); - } - TRACE_EXIT -EIO; /* g.p. error */ - } - TRACE_EXIT 0; -} - -/* forward */ static int seek_forward(int segment_id, int fast); - -static int fast_seek(int count, int reverse) -{ - int result = 0; - int status; - TRACE_FUN(ft_t_flow); - - if (count > 0) { - /* If positioned at begin or end of tape, fast seeking needs - * special treatment. - * Starting from logical bot needs a (slow) seek to the first - * segment before the high speed seek. Most drives do this - * automatically but some older don't, so we treat them - * all the same. - * Starting from logical eot is even more difficult because - * we cannot (slow) reverse seek to the last segment. - * TO BE IMPLEMENTED. - */ - inhibit_correction = 0; - if (ft_location.known && - ((ft_location.bot && !reverse) || - (ft_location.eot && reverse))) { - if (!reverse) { - /* (slow) skip to first segment on a track - */ - seek_forward(ft_location.track * ft_segments_per_track, 0); - --count; - } else { - /* When seeking backwards from - * end-of-tape the number of erased - * gaps found seems to be higher than - * expected. Therefor the drive must - * skip some more segments than - * calculated, but we don't know how - * many. Thus we will prevent the - * re-calculation of offset and - * overshoot when seeking backwards. - */ - inhibit_correction = 1; - count += 3; /* best guess */ - } - } - } else { - TRACE(ft_t_flow, "warning: zero or negative count: %d", count); - } - if (count > 0) { - int i; - int nibbles = count > 255 ? 3 : 2; - - if (count > 4095) { - TRACE(ft_t_noise, "skipping clipped at 4095 segment"); - count = 4095; - } - /* Issue this tape command first. */ - if (!reverse) { - TRACE(ft_t_noise, "skipping %d segment(s)", count); - result = ftape_command(nibbles == 3 ? - QIC_SKIP_EXTENDED_FORWARD : QIC_SKIP_FORWARD); - } else { - TRACE(ft_t_noise, "backing up %d segment(s)", count); - result = ftape_command(nibbles == 3 ? - QIC_SKIP_EXTENDED_REVERSE : QIC_SKIP_REVERSE); - } - if (result < 0) { - TRACE(ft_t_noise, "Skip command failed"); - } else { - --count; /* 0 means one gap etc. */ - for (i = 0; i < nibbles; ++i) { - if (result >= 0) { - result = ftape_parameter(count & 15); - count /= 16; - } - } - result = ftape_ready_wait(ftape_timeout.rewind, &status); - if (result >= 0) { - ftape_tape_running = 0; - } - } - } - TRACE_EXIT result; -} - -static int validate(int id) -{ - /* Check to see if position found is off-track as reported - * once. Because all tracks in one direction lie next to - * each other, if off-track the error will be approximately - * 2 * ft_segments_per_track. - */ - if (ft_location.track == -1) { - return 1; /* unforseen situation, don't generate error */ - } else { - /* Use margin of ft_segments_per_track on both sides - * because ftape needs some margin and the error we're - * looking for is much larger ! - */ - int lo = (ft_location.track - 1) * ft_segments_per_track; - int hi = (ft_location.track + 2) * ft_segments_per_track; - - return (id >= lo && id < hi); - } -} - -static int seek_forward(int segment_id, int fast) -{ - int failures = 0; - int count; - static int margin = 1; /* fixed: stop this before target */ - static int overshoot = 1; - static int min_count = 8; - int expected = -1; - int target = segment_id - margin; - int fast_seeking; - int prev_segment = ft_location.segment; - TRACE_FUN(ft_t_flow); - - if (!ft_location.known) { - TRACE_ABORT(-EIO, ft_t_err, - "fatal: cannot seek from unknown location"); - } - if (!validate(segment_id)) { - ftape_sleep(1 * FT_SECOND); - ft_failure = 1; - TRACE_ABORT(-EIO, ft_t_err, - "fatal: head off track (bad hardware?)"); - } - TRACE(ft_t_noise, "from %d/%d to %d/0 - %d", - ft_location.segment, ft_location.sector,segment_id,margin); - count = target - ft_location.segment - overshoot; - fast_seeking = (fast && - count > (min_count + (ft_location.bot ? 1 : 0))); - if (fast_seeking) { - TRACE(ft_t_noise, "fast skipping %d segments", count); - expected = segment_id - margin; - fast_seek(count, 0); - } - if (!ftape_tape_running) { - logical_forward(); - } - while (ft_location.segment < segment_id) { - /* This requires at least one sector in a (bad) segment to - * have a valid and readable sector id ! - * It looks like this is not guaranteed, so we must try - * to find a way to skip an EMPTY_SEGMENT. !!! FIXME !!! - */ - if (ftape_read_id() < 0 || !ft_location.known || - sigtestsetmask(¤t->pending.signal, _DONT_BLOCK)) { - ft_location.known = 0; - if (!ftape_tape_running || - ++failures > FT_SECTORS_PER_SEGMENT) { - TRACE_ABORT(-EIO, ft_t_err, - "read_id failed completely"); - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE(ft_t_flow, "read_id failed, retry (%d)", - failures); - continue; - } - if (fast_seeking) { - TRACE(ft_t_noise, "ended at %d/%d (%d,%d)", - ft_location.segment, ft_location.sector, - overshoot, inhibit_correction); - if (!inhibit_correction && - (ft_location.segment < expected || - ft_location.segment > expected + margin)) { - int error = ft_location.segment - expected; - TRACE(ft_t_noise, - "adjusting overshoot from %d to %d", - overshoot, overshoot + error); - overshoot += error; - /* All overshoots have the same - * direction, so it should never - * become negative, but who knows. - */ - if (overshoot < -5 || - overshoot > OVERSHOOT_LIMIT) { - if (overshoot < 0) { - /* keep sane value */ - overshoot = -5; - } else { - /* keep sane value */ - overshoot = OVERSHOOT_LIMIT; - } - TRACE(ft_t_noise, - "clipped overshoot to %d", - overshoot); - } - } - fast_seeking = 0; - } - if (ft_location.known) { - if (ft_location.segment > prev_segment + 1) { - TRACE(ft_t_noise, - "missed segment %d while skipping", - prev_segment + 1); - } - prev_segment = ft_location.segment; - } - } - if (ft_location.segment > segment_id) { - TRACE_ABORT(-EIO, - ft_t_noise, "failed: skip ended at segment %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int skip_reverse(int segment_id, int *pstatus) -{ - int failures = 0; - static int overshoot = 1; - static int min_rewind = 2; /* 1 + overshoot */ - static const int margin = 1; /* stop this before target */ - int expected = 0; - int count = 1; - int short_seek; - int target = segment_id - margin; - TRACE_FUN(ft_t_flow); - - if (ft_location.known && !validate(segment_id)) { - ftape_sleep(1 * FT_SECOND); - ft_failure = 1; - TRACE_ABORT(-EIO, ft_t_err, - "fatal: head off track (bad hardware?)"); - } - do { - if (!ft_location.known) { - TRACE(ft_t_warn, "warning: location not known"); - } - TRACE(ft_t_noise, "from %d/%d to %d/0 - %d", - ft_location.segment, ft_location.sector, - segment_id, margin); - /* min_rewind == 1 + overshoot_when_doing_minimum_rewind - * overshoot == overshoot_when_doing_larger_rewind - * Initially min_rewind == 1 + overshoot, optimization - * of both values will be done separately. - * overshoot and min_rewind can be negative as both are - * sums of three components: - * any_overshoot == rewind_overshoot - - * stop_overshoot - - * start_overshoot - */ - if (ft_location.segment - target - (min_rewind - 1) < 1) { - short_seek = 1; - } else { - count = ft_location.segment - target - overshoot; - short_seek = (count < 1); - } - if (short_seek) { - count = 1; /* do shortest rewind */ - expected = ft_location.segment - min_rewind; - if (expected/ft_segments_per_track != ft_location.track) { - expected = (ft_location.track * - ft_segments_per_track); - } - } else { - expected = target; - } - fast_seek(count, 1); - logical_forward(); - if (ftape_read_id() < 0 || !ft_location.known || - (sigtestsetmask(¤t->pending.signal, _DONT_BLOCK))) { - if ((!ftape_tape_running && !ft_location.known) || - ++failures > FT_SECTORS_PER_SEGMENT) { - TRACE_ABORT(-EIO, ft_t_err, - "read_id failed completely"); - } - FT_SIGNAL_EXIT(_DONT_BLOCK); - TRACE_CATCH(ftape_report_drive_status(pstatus),); - TRACE(ft_t_noise, "ftape_read_id failed, retry (%d)", - failures); - continue; - } - TRACE(ft_t_noise, "ended at %d/%d (%d,%d,%d)", - ft_location.segment, ft_location.sector, - min_rewind, overshoot, inhibit_correction); - if (!inhibit_correction && - (ft_location.segment < expected || - ft_location.segment > expected + margin)) { - int error = expected - ft_location.segment; - if (short_seek) { - TRACE(ft_t_noise, - "adjusting min_rewind from %d to %d", - min_rewind, min_rewind + error); - min_rewind += error; - if (min_rewind < -5) { - /* is this right ? FIXME ! */ - /* keep sane value */ - min_rewind = -5; - TRACE(ft_t_noise, - "clipped min_rewind to %d", - min_rewind); - } - } else { - TRACE(ft_t_noise, - "adjusting overshoot from %d to %d", - overshoot, overshoot + error); - overshoot += error; - if (overshoot < -5 || - overshoot > OVERSHOOT_LIMIT) { - if (overshoot < 0) { - /* keep sane value */ - overshoot = -5; - } else { - /* keep sane value */ - overshoot = OVERSHOOT_LIMIT; - } - TRACE(ft_t_noise, - "clipped overshoot to %d", - overshoot); - } - } - } - } while (ft_location.segment > segment_id); - if (ft_location.known) { - TRACE(ft_t_noise, "current location: %d/%d", - ft_location.segment, ft_location.sector); - } - TRACE_EXIT 0; -} - -static int determine_position(void) -{ - int retry = 0; - int status; - int result; - TRACE_FUN(ft_t_flow); - - if (!ftape_tape_running) { - /* This should only happen if tape is stopped by isr. - */ - TRACE(ft_t_flow, "waiting for tape stop"); - if (ftape_ready_wait(ftape_timeout.pause, &status) < 0) { - TRACE(ft_t_flow, "drive still running (fatal)"); - ftape_tape_running = 1; /* ? */ - } - } else { - ftape_report_drive_status(&status); - } - if (status & QIC_STATUS_READY) { - /* Drive must be ready to check error state ! - */ - TRACE(ft_t_flow, "drive is ready"); - if (status & QIC_STATUS_ERROR) { - unsigned int error; - qic117_cmd_t command; - - /* Report and clear error state, try to continue. - */ - TRACE(ft_t_flow, "error status set"); - ftape_report_error(&error, &command, 1); - ftape_ready_wait(ftape_timeout.reset, &status); - ftape_tape_running = 0; /* ? */ - } - if (check_bot_eot(status)) { - if (ft_location.bot) { - if ((status & QIC_STATUS_READY) == 0) { - /* tape moving away from - * bot/eot, let's see if we - * can catch up with the first - * segment on this track. - */ - } else { - TRACE(ft_t_flow, - "start tape from logical bot"); - logical_forward(); /* start moving */ - } - } else { - if ((status & QIC_STATUS_READY) == 0) { - TRACE(ft_t_noise, "waiting for logical end of track"); - result = ftape_ready_wait(ftape_timeout.reset, &status); - /* error handling needed ? */ - } else { - TRACE(ft_t_noise, - "tape at logical end of track"); - } - } - } else { - TRACE(ft_t_flow, "start tape"); - logical_forward(); /* start moving */ - ft_location.known = 0; /* not cleared by logical forward ! */ - } - } - /* tape should be moving now, start reading id's - */ - while (!ft_location.known && - retry++ < FT_SECTORS_PER_SEGMENT && - (result = ftape_read_id()) < 0) { - - TRACE(ft_t_flow, "location unknown"); - - /* exit on signal - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - - /* read-id somehow failed, tape may - * have reached end or some other - * error happened. - */ - TRACE(ft_t_flow, "read-id failed"); - TRACE_CATCH(ftape_report_drive_status(&status),); - TRACE(ft_t_err, "ftape_report_drive_status: 0x%02x", status); - if (status & QIC_STATUS_READY) { - ftape_tape_running = 0; - TRACE(ft_t_noise, "tape stopped for unknown reason! " - "status = 0x%02x", status); - if (status & QIC_STATUS_ERROR || - !check_bot_eot(status)) { - /* oops, tape stopped but not at end! - */ - TRACE_EXIT -EIO; - } - } - } - TRACE(ft_t_flow, - "tape is positioned at segment %d", ft_location.segment); - TRACE_EXIT ft_location.known ? 0 : -EIO; -} - -/* Get the tape running and position it just before the - * requested segment. - * Seek tape-track and reposition as needed. - */ -int ftape_start_tape(int segment_id, int sector_offset) -{ - int track = segment_id / ft_segments_per_track; - int result = -EIO; - int status; - static int last_segment = -1; - static int bad_bus_timing = 0; - /* number of segments passing the head between starting the tape - * and being able to access the first sector. - */ - static int start_offset = 1; - int retry; - TRACE_FUN(ft_t_flow); - - /* If sector_offset > 0, seek into wanted segment instead of - * into previous. - * This allows error recovery if a part of the segment is bad - * (erased) causing the tape drive to generate an index pulse - * thus causing a no-data error before the requested sector - * is reached. - */ - ftape_tape_running = 0; - TRACE(ft_t_noise, "target segment: %d/%d%s", segment_id, sector_offset, - ft_buffer[ft_head]->retry > 0 ? " retry" : ""); - if (ft_buffer[ft_head]->retry > 0) { /* this is a retry */ - int dist = segment_id - last_segment; - - if ((int)ft_history.overrun_errors < overrun_count_offset) { - overrun_count_offset = ft_history.overrun_errors; - } else if (dist < 0 || dist > 50) { - overrun_count_offset = ft_history.overrun_errors; - } else if ((ft_history.overrun_errors - - overrun_count_offset) >= 8) { - if (ftape_increase_threshold() >= 0) { - --ft_buffer[ft_head]->retry; - overrun_count_offset = - ft_history.overrun_errors; - TRACE(ft_t_warn, "increased threshold because " - "of excessive overrun errors"); - } else if (!bad_bus_timing && ft_data_rate >= 1000) { - ftape_half_data_rate(); - --ft_buffer[ft_head]->retry; - bad_bus_timing = 1; - overrun_count_offset = - ft_history.overrun_errors; - TRACE(ft_t_warn, "reduced datarate because " - "of excessive overrun errors"); - } - } - } - last_segment = segment_id; - if (ft_location.track != track || - (ftape_might_be_off_track && ft_buffer[ft_head]->retry== 0)) { - /* current track unknown or not equal to destination - */ - ftape_ready_wait(ftape_timeout.seek, &status); - ftape_seek_head_to_track(track); - /* overrun_count_offset = ft_history.overrun_errors; */ - } - result = -EIO; - retry = 0; - while (result < 0 && - retry++ <= 5 && - !ft_failure && - !(sigtestsetmask(¤t->pending.signal, _DONT_BLOCK))) { - - if (retry && start_offset < 5) { - start_offset ++; - } - /* Check if we are able to catch the requested - * segment in time. - */ - if ((ft_location.known || (determine_position() == 0)) && - ft_location.segment >= - (segment_id - - ((ftape_tape_running || ft_location.bot) - ? 0 : start_offset))) { - /* Too far ahead (in or past target segment). - */ - if (ftape_tape_running) { - if ((result = ftape_stop_tape(&status)) < 0) { - TRACE(ft_t_err, - "stop tape failed with code %d", - result); - break; - } - TRACE(ft_t_noise, "tape stopped"); - ftape_tape_running = 0; - } - TRACE(ft_t_noise, "repositioning"); - ++ft_history.rewinds; - if (segment_id % ft_segments_per_track < start_offset){ - TRACE(ft_t_noise, "end of track condition\n" - KERN_INFO "segment_id : %d\n" - KERN_INFO "ft_segments_per_track: %d\n" - KERN_INFO "start_offset : %d", - segment_id, ft_segments_per_track, - start_offset); - - /* If seeking to first segments on - * track better do a complete rewind - * to logical begin of track to get a - * more steady tape motion. - */ - result = ftape_command_wait( - (ft_location.track & 1) - ? QIC_PHYSICAL_FORWARD - : QIC_PHYSICAL_REVERSE, - ftape_timeout.rewind, &status); - check_bot_eot(status); /* update location */ - } else { - result= skip_reverse(segment_id - start_offset, - &status); - } - } - if (!ft_location.known) { - TRACE(ft_t_bug, "panic: location not known"); - result = -EIO; - continue; /* while() will check for failure */ - } - TRACE(ft_t_noise, "current segment: %d/%d", - ft_location.segment, ft_location.sector); - /* We're on the right track somewhere before the - * wanted segment. Start tape movement if needed and - * skip to just before or inside the requested - * segment. Keep tape running. - */ - result = 0; - if (ft_location.segment < - (segment_id - ((ftape_tape_running || ft_location.bot) - ? 0 : start_offset))) { - if (sector_offset > 0) { - result = seek_forward(segment_id, - retry <= 3); - } else { - result = seek_forward(segment_id - 1, - retry <= 3); - } - } - if (result == 0 && - ft_location.segment != - (segment_id - (sector_offset > 0 ? 0 : 1))) { - result = -EIO; - } - } - if (result < 0) { - TRACE(ft_t_err, "failed to reposition"); - } else { - ft_runner_status = running; - } - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/lowlevel/ftape-rw.h b/drivers/char/ftape/lowlevel/ftape-rw.h deleted file mode 100644 index 32f4feeb887c..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-rw.h +++ /dev/null @@ -1,111 +0,0 @@ -#ifndef _FTAPE_RW_H -#define _FTAPE_RW_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-rw.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:25 $ - * - * This file contains the definitions for the read and write - * functions for the QIC-117 floppy-tape driver for Linux. - * - * Claus-Justus Heine (1996/09/20): Add definition of format code 6 - * Claus-Justus Heine (1996/10/04): Changed GET/PUT macros to cast to (__u8 *) - * - */ - -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/ftape-bsm.h" - -#include - -#define GET2(address, offset) get_unaligned((__u16*)((__u8 *)address + offset)) -#define GET4(address, offset) get_unaligned((__u32*)((__u8 *)address + offset)) -#define GET8(address, offset) get_unaligned((__u64*)((__u8 *)address + offset)) -#define PUT2(address, offset , value) put_unaligned((value), (__u16*)((__u8 *)address + offset)) -#define PUT4(address, offset , value) put_unaligned((value), (__u32*)((__u8 *)address + offset)) -#define PUT8(address, offset , value) put_unaligned((value), (__u64*)((__u8 *)address + offset)) - -enum runner_status_enum { - idle = 0, - running, - do_abort, - aborting, - logical_eot, - end_of_tape, -}; - -typedef enum ft_buffer_queue { - ft_queue_head = 0, - ft_queue_tail = 1 -} ft_buffer_queue_t; - - -typedef struct { - int track; /* tape head position */ - volatile int segment; /* current segment */ - volatile int sector; /* sector offset within current segment */ - volatile unsigned int bot; /* logical begin of track */ - volatile unsigned int eot; /* logical end of track */ - volatile unsigned int known; /* validates bot, segment, sector */ -} location_record; - -/* Count nr of 1's in pattern. - */ -static inline int count_ones(unsigned long mask) -{ - int bits; - - for (bits = 0; mask != 0; mask >>= 1) { - if (mask & 1) { - ++bits; - } - } - return bits; -} - -#define FT_MAX_NR_BUFFERS 16 /* arbitrary value */ -/* ftape-rw.c defined global vars. - */ -extern buffer_struct *ft_buffer[FT_MAX_NR_BUFFERS]; -extern int ft_nr_buffers; -extern location_record ft_location; -extern volatile int ftape_tape_running; - -/* ftape-rw.c defined global functions. - */ -extern int ftape_setup_new_segment(buffer_struct * buff, - int segment_id, - int offset); -extern int ftape_calc_next_cluster(buffer_struct * buff); -extern buffer_struct *ftape_next_buffer (ft_buffer_queue_t pos); -extern buffer_struct *ftape_get_buffer (ft_buffer_queue_t pos); -extern int ftape_buffer_id (ft_buffer_queue_t pos); -extern void ftape_reset_buffer(void); -extern void ftape_tape_parameters(__u8 drive_configuration); -extern int ftape_wait_segment(buffer_state_enum state); -extern int ftape_dumb_stop(void); -extern int ftape_start_tape(int segment_id, int offset); -extern int ftape_stop_tape(int *pstatus); -extern int ftape_handle_logical_eot(void); -extern buffer_state_enum ftape_set_state(buffer_state_enum new_state); -#endif /* _FTAPE_RW_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-setup.c b/drivers/char/ftape/lowlevel/ftape-setup.c deleted file mode 100644 index 678340acd0b7..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-setup.c +++ /dev/null @@ -1,104 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-setup.c,v $ - * $Revision: 1.7 $ - * $Date: 1997/10/10 09:57:06 $ - * - * This file contains the code for processing the kernel command - * line options for the QIC-40/80/3010/3020 floppy-tape driver - * "ftape" for Linux. - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/fdc-io.h" - -static struct param_table { - const char *name; - int *var; - int def_param; - int min; - int max; -} config_params[] __initdata = { -#ifndef CONFIG_FT_NO_TRACE_AT_ALL - { "tracing", &ftape_tracing, 3, ft_t_bug, ft_t_any}, -#endif - { "ioport", &ft_fdc_base, CONFIG_FT_FDC_BASE, 0x0, 0xfff}, - { "irq", &ft_fdc_irq, CONFIG_FT_FDC_IRQ, 2, 15}, - { "dma", &ft_fdc_dma, CONFIG_FT_FDC_DMA, 0, 3}, - { "threshold", &ft_fdc_threshold, CONFIG_FT_FDC_THR, 1, 16}, - { "datarate", &ft_fdc_rate_limit, CONFIG_FT_FDC_MAX_RATE, 500, 2000}, - { "fc10", &ft_probe_fc10, CONFIG_FT_PROBE_FC10, 0, 1}, - { "mach2", &ft_mach2, CONFIG_FT_MACH2, 0, 1} -}; - -static int __init ftape_setup(char *str) -{ - int i; - int param; - int ints[2]; - - TRACE_FUN(ft_t_flow); - - str = get_options(str, ARRAY_SIZE(ints), ints); - if (str) { - for (i=0; i < NR_ITEMS(config_params); i++) { - if (strcmp(str,config_params[i].name) == 0){ - if (ints[0]) { - param = ints[1]; - } else { - param = config_params[i].def_param; - } - if (param < config_params[i].min || - param > config_params[i].max) { - TRACE(ft_t_err, - "parameter %s out of range %d ... %d", - config_params[i].name, - config_params[i].min, - config_params[i].max); - goto out; - } - if(config_params[i].var) { - TRACE(ft_t_info, "%s=%d", str, param); - *config_params[i].var = param; - } - goto out; - } - } - } - if (str) { - TRACE(ft_t_err, "unknown ftape option [%s]", str); - - TRACE(ft_t_err, "allowed options are:"); - for (i=0; i < NR_ITEMS(config_params); i++) { - TRACE(ft_t_err, " %s",config_params[i].name); - } - } else { - TRACE(ft_t_err, "botched ftape option"); - } - out: - TRACE_EXIT 1; -} - -__setup("ftape=", ftape_setup); diff --git a/drivers/char/ftape/lowlevel/ftape-tracing.c b/drivers/char/ftape/lowlevel/ftape-tracing.c deleted file mode 100644 index 7fdc6567440b..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-tracing.c +++ /dev/null @@ -1,118 +0,0 @@ -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-tracing.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:27 $ - * - * This file contains the reading code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include "../lowlevel/ftape-tracing.h" - -/* Global vars. - */ -/* tracing - * set it to: to log : - * 0 bugs - * 1 + errors - * 2 + warnings - * 3 + information - * 4 + more information - * 5 + program flow - * 6 + fdc/dma info - * 7 + data flow - * 8 + everything else - */ -ft_trace_t ftape_tracing = ft_t_info; /* Default level: information and up */ -int ftape_function_nest_level; - -/* Local vars. - */ -static __u8 trace_id; -static char spacing[] = "* "; - -void ftape_trace_call(const char *file, const char *name) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", - ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s+%s (%s)\n", - (int) trace_id++, indent, file, name); -} - -void ftape_trace_exit(const char *file, const char *name) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s-%s (%s)\n", - (int) trace_id++, indent, file, name); -} - -void ftape_trace_log(const char *file, const char *function) -{ - char *indent; - - /* Since printk seems not to work with "%*s" format - * we'll use this work-around. - */ - if (ftape_function_nest_level < 0) { - printk(KERN_INFO "function nest level (%d) < 0\n", ftape_function_nest_level); - ftape_function_nest_level = 0; - } - if (ftape_function_nest_level < sizeof(spacing)) { - indent = (spacing + - sizeof(spacing) - 1 - - ftape_function_nest_level); - } else { - indent = spacing; - } - printk(KERN_INFO "[%03d]%s%s (%s) - ", - (int) trace_id++, indent, file, function); -} diff --git a/drivers/char/ftape/lowlevel/ftape-tracing.h b/drivers/char/ftape/lowlevel/ftape-tracing.h deleted file mode 100644 index 2950810c7085..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-tracing.h +++ /dev/null @@ -1,179 +0,0 @@ -#ifndef _FTAPE_TRACING_H -#define _FTAPE_TRACING_H - -/* - * Copyright (C) 1994-1996 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-tracing.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:28 $ - * - * This file contains definitions that eases the debugging of the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - */ - -#include - -/* - * Be very careful with TRACE_EXIT and TRACE_ABORT. - * - * if (something) TRACE_EXIT error; - * - * will NOT work. Use - * - * if (something) { - * TRACE_EXIT error; - * } - * - * instead. Maybe a bit dangerous, but save lots of lines of code. - */ - -#define LL_X "%d/%d KB" -#define LL(x) (unsigned int)((__u64)(x)>>10), (unsigned int)((x)&1023) - -typedef enum { - ft_t_nil = -1, - ft_t_bug, - ft_t_err, - ft_t_warn, - ft_t_info, - ft_t_noise, - ft_t_flow, - ft_t_fdc_dma, - ft_t_data_flow, - ft_t_any -} ft_trace_t; - -#ifdef CONFIG_FT_NO_TRACE_AT_ALL -/* the compiler will optimize away most TRACE() macros - */ -#define FT_TRACE_TOP_LEVEL ft_t_bug -#define TRACE_FUN(level) do {} while(0) -#define TRACE_EXIT return -#define TRACE(l, m, i...) \ -{ \ - if ((ft_trace_t)(l) == FT_TRACE_TOP_LEVEL) { \ - printk(KERN_INFO"ftape%s(%s):\n" \ - KERN_INFO m".\n" ,__FILE__, __FUNCTION__ , ##i); \ - } \ -} -#define SET_TRACE_LEVEL(l) if ((l) == (l)) do {} while(0) -#define TRACE_LEVEL FT_TRACE_TOP_LEVEL - -#else - -#ifdef CONFIG_FT_NO_TRACE -/* the compiler will optimize away many TRACE() macros - * the ftape_simple_trace_call() function simply increments - * the function nest level. - */ -#define FT_TRACE_TOP_LEVEL ft_t_warn -#define TRACE_FUN(level) ftape_function_nest_level++ -#define TRACE_EXIT ftape_function_nest_level--; return - -#else -#ifdef CONFIG_FT_FULL_DEBUG -#define FT_TRACE_TOP_LEVEL ft_t_any -#else -#define FT_TRACE_TOP_LEVEL ft_t_flow -#endif -#define TRACE_FUN(level) \ - const ft_trace_t _tracing = level; \ - if (ftape_tracing >= (ft_trace_t)(level) && \ - (ft_trace_t)(level) <= FT_TRACE_TOP_LEVEL) \ - ftape_trace_call(__FILE__, __FUNCTION__); \ - ftape_function_nest_level ++; - -#define TRACE_EXIT \ - --ftape_function_nest_level; \ - if (ftape_tracing >= (ft_trace_t)(_tracing) && \ - (ft_trace_t)(_tracing) <= FT_TRACE_TOP_LEVEL) \ - ftape_trace_exit(__FILE__, __FUNCTION__); \ - return - -#endif - -#define TRACE(l, m, i...) \ -{ \ - if (ftape_tracing >= (ft_trace_t)(l) && \ - (ft_trace_t)(l) <= FT_TRACE_TOP_LEVEL) { \ - ftape_trace_log(__FILE__, __FUNCTION__); \ - printk(m".\n" ,##i); \ - } \ -} - -#define SET_TRACE_LEVEL(l) \ -{ \ - if ((ft_trace_t)(l) <= FT_TRACE_TOP_LEVEL) { \ - ftape_tracing = (ft_trace_t)(l); \ - } else { \ - ftape_tracing = FT_TRACE_TOP_LEVEL; \ - } \ -} -#define TRACE_LEVEL \ -((ftape_tracing <= FT_TRACE_TOP_LEVEL) ? ftape_tracing : FT_TRACE_TOP_LEVEL) - - -/* Global variables declared in tracing.c - */ -extern ft_trace_t ftape_tracing; /* sets default level */ -extern int ftape_function_nest_level; - -/* Global functions declared in tracing.c - */ -extern void ftape_trace_call(const char *file, const char *name); -extern void ftape_trace_exit(const char *file, const char *name); -extern void ftape_trace_log (const char *file, const char *name); - -#endif /* !defined(CONFIG_FT_NO_TRACE_AT_ALL) */ - -/* - * Abort with a message. - */ -#define TRACE_ABORT(res, i...) \ -{ \ - TRACE(i); \ - TRACE_EXIT res; \ -} - -/* The following transforms the common "if(result < 0) ... " into a - * one-liner. - */ -#define _TRACE_CATCH(level, fun, action) \ -{ \ - int _res = (fun); \ - if (_res < 0) { \ - do { action /* */ ; } while(0); \ - TRACE_ABORT(_res, level, "%s failed: %d", #fun, _res); \ - } \ -} - -#define TRACE_CATCH(fun, fail) _TRACE_CATCH(ft_t_err, fun, fail) - -/* Abort the current function when signalled. This doesn't belong here, - * but rather into ftape-rw.h (maybe) - */ -#define FT_SIGNAL_EXIT(sig_mask) \ - if (sigtestsetmask(¤t->pending.signal, sig_mask)) { \ - TRACE_ABORT(-EINTR, \ - ft_t_warn, \ - "interrupted by non-blockable signal"); \ - } - -#endif /* _FTAPE_TRACING_H */ diff --git a/drivers/char/ftape/lowlevel/ftape-write.c b/drivers/char/ftape/lowlevel/ftape-write.c deleted file mode 100644 index 45601ec801ee..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-write.c +++ /dev/null @@ -1,336 +0,0 @@ -/* - * Copyright (C) 1993-1995 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-write.c,v $ - * $Revision: 1.3.4.1 $ - * $Date: 1997/11/14 18:07:04 $ - * - * This file contains the writing code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include -#include - -#include -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-ecc.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/fdc-isr.h" - -/* Global vars. - */ - -/* Local vars. - */ -static int last_write_failed; - -void ftape_zap_write_buffers(void) -{ - int i; - - for (i = 0; i < ft_nr_buffers; ++i) { - ft_buffer[i]->status = done; - } - ftape_reset_buffer(); -} - -static int copy_and_gen_ecc(void *destination, - const void *source, - const SectorMap bad_sector_map) -{ - int result; - struct memory_segment mseg; - int bads = count_ones(bad_sector_map); - TRACE_FUN(ft_t_any); - - if (bads > 0) { - TRACE(ft_t_noise, "bad sectors in map: %d", bads); - } - if (bads + 3 >= FT_SECTORS_PER_SEGMENT) { - TRACE(ft_t_noise, "empty segment"); - mseg.blocks = 0; /* skip entire segment */ - result = 0; /* nothing written */ - } else { - mseg.blocks = FT_SECTORS_PER_SEGMENT - bads; - mseg.data = destination; - memcpy(mseg.data, source, (mseg.blocks - 3) * FT_SECTOR_SIZE); - result = ftape_ecc_set_segment_parity(&mseg); - if (result < 0) { - TRACE(ft_t_err, "ecc_set_segment_parity failed"); - } else { - result = (mseg.blocks - 3) * FT_SECTOR_SIZE; - } - } - TRACE_EXIT result; -} - - -int ftape_start_writing(const ft_write_mode_t mode) -{ - buffer_struct *head = ftape_get_buffer(ft_queue_head); - int segment_id = head->segment_id; - int result; - buffer_state_enum wanted_state = (mode == FT_WR_DELETE - ? deleting - : writing); - TRACE_FUN(ft_t_flow); - - if ((ft_driver_state != wanted_state) || head->status != waiting) { - TRACE_EXIT 0; - } - ftape_setup_new_segment(head, segment_id, 1); - if (mode == FT_WR_SINGLE) { - /* stop tape instead of pause */ - head->next_segment = 0; - } - ftape_calc_next_cluster(head); /* prepare */ - head->status = ft_driver_state; /* either writing or deleting */ - if (ft_runner_status == idle) { - TRACE(ft_t_noise, - "starting runner for segment %d", segment_id); - TRACE_CATCH(ftape_start_tape(segment_id,head->sector_offset),); - } else { - TRACE(ft_t_noise, "runner not idle, not starting tape"); - } - /* go */ - result = fdc_setup_read_write(head, (mode == FT_WR_DELETE - ? FDC_WRITE_DELETED : FDC_WRITE)); - ftape_set_state(wanted_state); /* should not be necessary */ - TRACE_EXIT result; -} - -/* Wait until all data is actually written to tape. - * - * There is a problem: when the tape runs into logical EOT, then this - * failes. We need to restart the runner in this case. - */ -int ftape_loop_until_writes_done(void) -{ - buffer_struct *head; - TRACE_FUN(ft_t_flow); - - while ((ft_driver_state == writing || ft_driver_state == deleting) && - ftape_get_buffer(ft_queue_head)->status != done) { - /* set the runner status to idle if at lEOT */ - TRACE_CATCH(ftape_handle_logical_eot(), last_write_failed = 1); - /* restart the tape if necessary */ - if (ft_runner_status == idle) { - TRACE(ft_t_noise, "runner is idle, restarting"); - if (ft_driver_state == deleting) { - TRACE_CATCH(ftape_start_writing(FT_WR_DELETE), - last_write_failed = 1); - } else { - TRACE_CATCH(ftape_start_writing(FT_WR_MULTI), - last_write_failed = 1); - } - } - TRACE(ft_t_noise, "tail: %d, head: %d", - ftape_buffer_id(ft_queue_tail), - ftape_buffer_id(ft_queue_head)); - TRACE_CATCH(fdc_interrupt_wait(5 * FT_SECOND), - last_write_failed = 1); - head = ftape_get_buffer(ft_queue_head); - if (head->status == error) { - /* Allow escape from loop when signaled ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - if (head->hard_error_map != 0) { - /* Implement hard write error recovery here - */ - } - /* retry this one */ - head->status = waiting; - if (ft_runner_status == aborting) { - ftape_dumb_stop(); - } - if (ft_runner_status != idle) { - TRACE_ABORT(-EIO, ft_t_err, - "unexpected state: " - "ft_runner_status != idle"); - } - ftape_start_writing(ft_driver_state == deleting - ? FT_WR_MULTI : FT_WR_DELETE); - } - TRACE(ft_t_noise, "looping until writes done"); - } - ftape_set_state(idle); - TRACE_EXIT 0; -} - -/* Write given segment from buffer at address to tape. - */ -static int write_segment(const int segment_id, - const void *address, - const ft_write_mode_t write_mode) -{ - int bytes_written = 0; - buffer_struct *tail; - buffer_state_enum wanted_state = (write_mode == FT_WR_DELETE - ? deleting : writing); - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "segment_id = %d", segment_id); - if (ft_driver_state != wanted_state) { - if (ft_driver_state == deleting || - wanted_state == deleting) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - } - TRACE(ft_t_noise, "calling ftape_abort_operation"); - TRACE_CATCH(ftape_abort_operation(),); - ftape_zap_write_buffers(); - ftape_set_state(wanted_state); - } - /* if all buffers full we'll have to wait... - */ - ftape_wait_segment(wanted_state); - tail = ftape_get_buffer(ft_queue_tail); - switch(tail->status) { - case done: - ft_history.defects += count_ones(tail->hard_error_map); - break; - case waiting: - /* this could happen with multiple EMPTY_SEGMENTs, but - * shouldn't happen any more as we re-start the runner even - * with an empty segment. - */ - bytes_written = -EAGAIN; - break; - case error: - /* setup for a retry - */ - tail->status = waiting; - bytes_written = -EAGAIN; /* force retry */ - if (tail->hard_error_map != 0) { - TRACE(ft_t_warn, - "warning: %d hard error(s) in written segment", - count_ones(tail->hard_error_map)); - TRACE(ft_t_noise, "hard_error_map = 0x%08lx", - (long)tail->hard_error_map); - /* Implement hard write error recovery here - */ - } - break; - default: - TRACE_ABORT(-EIO, ft_t_err, - "wait for empty segment failed, tail status: %d", - tail->status); - } - /* should runner stop ? - */ - if (ft_runner_status == aborting) { - buffer_struct *head = ftape_get_buffer(ft_queue_head); - if (head->status == wanted_state) { - head->status = done; /* ???? */ - } - /* don't call abort_operation(), we don't want to zap - * the dma buffers - */ - TRACE_CATCH(ftape_dumb_stop(),); - } else { - /* If just passed last segment on tape: wait for BOT - * or EOT mark. Sets ft_runner_status to idle if at lEOT - * and successful - */ - TRACE_CATCH(ftape_handle_logical_eot(),); - } - if (tail->status == done) { - /* now at least one buffer is empty, fill it with our - * data. skip bad sectors and generate ecc. - * copy_and_gen_ecc return nr of bytes written, range - * 0..29 Kb inclusive! - * - * Empty segments are handled inside coyp_and_gen_ecc() - */ - if (write_mode != FT_WR_DELETE) { - TRACE_CATCH(bytes_written = copy_and_gen_ecc( - tail->address, address, - ftape_get_bad_sector_entry(segment_id)),); - } - tail->segment_id = segment_id; - tail->status = waiting; - tail = ftape_next_buffer(ft_queue_tail); - } - /* Start tape only if all buffers full or flush mode. - * This will give higher probability of streaming. - */ - if (ft_runner_status != running && - ((tail->status == waiting && - ftape_get_buffer(ft_queue_head) == tail) || - write_mode != FT_WR_ASYNC)) { - TRACE_CATCH(ftape_start_writing(write_mode),); - } - TRACE_EXIT bytes_written; -} - -/* Write as much as fits from buffer to the given segment on tape - * and handle retries. - * Return the number of bytes written (>= 0), or: - * -EIO write failed - * -EINTR interrupted by signal - * -ENOSPC device full - */ -int ftape_write_segment(const int segment_id, - const void *buffer, - const ft_write_mode_t flush) -{ - int retry = 0; - int result; - TRACE_FUN(ft_t_flow); - - ft_history.used |= 2; - if (segment_id >= ft_tracks_per_tape*ft_segments_per_track) { - /* tape full */ - TRACE_ABORT(-ENOSPC, ft_t_err, - "invalid segment id: %d (max %d)", - segment_id, - ft_tracks_per_tape * ft_segments_per_track -1); - } - for (;;) { - if ((result = write_segment(segment_id, buffer, flush)) >= 0) { - if (result == 0) { /* empty segment */ - TRACE(ft_t_noise, - "empty segment, nothing written"); - } - TRACE_EXIT result; - } - if (result == -EAGAIN) { - if (++retry > 100) { /* give up */ - TRACE_ABORT(-EIO, ft_t_err, - "write failed, >100 retries in segment"); - } - TRACE(ft_t_warn, "write error, retry %d (%d)", - retry, - ftape_get_buffer(ft_queue_tail)->segment_id); - } else { - TRACE_ABORT(result, ft_t_err, - "write_segment failed, error: %d", result); - } - /* Allow escape from loop when signaled ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - } -} diff --git a/drivers/char/ftape/lowlevel/ftape-write.h b/drivers/char/ftape/lowlevel/ftape-write.h deleted file mode 100644 index 0e7f898b7af9..000000000000 --- a/drivers/char/ftape/lowlevel/ftape-write.h +++ /dev/null @@ -1,53 +0,0 @@ -#ifndef _FTAPE_WRITE_H -#define _FTAPE_WRITE_H - -/* - * Copyright (C) 1994-1995 Bas Laarhoven, - * (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape-write.h,v $ - $Author: claus $ - * - $Revision: 1.2 $ - $Date: 1997/10/05 19:18:30 $ - $State: Exp $ - * - * This file contains the definitions for the write functions - * for the QIC-117 floppy-tape driver for Linux. - * - */ - - -/* ftape-write.c defined global functions. - */ -typedef enum { - FT_WR_ASYNC = 0, /* start tape only when all buffers are full */ - FT_WR_MULTI = 1, /* start tape, but don't necessarily stop */ - FT_WR_SINGLE = 2, /* write a single segment and stop afterwards */ - FT_WR_DELETE = 3 /* write deleted data marks */ -} ft_write_mode_t; - -extern int ftape_start_writing(const ft_write_mode_t mode); -extern int ftape_write_segment(const int segment, - const void *address, - const ft_write_mode_t flushing); -extern void ftape_zap_write_buffers(void); -extern int ftape_loop_until_writes_done(void); - -#endif /* _FTAPE_WRITE_H */ - diff --git a/drivers/char/ftape/lowlevel/ftape_syms.c b/drivers/char/ftape/lowlevel/ftape_syms.c deleted file mode 100644 index 8e0dc4a07ca6..000000000000 --- a/drivers/char/ftape/lowlevel/ftape_syms.c +++ /dev/null @@ -1,87 +0,0 @@ -/* - * Copyright (C) 1996-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/lowlevel/ftape_syms.c,v $ - * $Revision: 1.4 $ - * $Date: 1997/10/17 00:03:51 $ - * - * This file contains the symbols that the ftape low level - * part of the QIC-40/80/3010/3020 floppy-tape driver "ftape" - * exports to its high level clients - */ - -#include - -#include -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-init.h" -#include "../lowlevel/fdc-io.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-rw.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-format.h" - -/* bad sector handling from ftape-bsm.c */ -EXPORT_SYMBOL(ftape_get_bad_sector_entry); -EXPORT_SYMBOL(ftape_find_end_of_bsm_list); -/* from ftape-rw.c */ -EXPORT_SYMBOL(ftape_set_state); -/* from ftape-ctl.c */ -EXPORT_SYMBOL(ftape_seek_to_bot); -EXPORT_SYMBOL(ftape_seek_to_eot); -EXPORT_SYMBOL(ftape_abort_operation); -EXPORT_SYMBOL(ftape_get_status); -EXPORT_SYMBOL(ftape_enable); -EXPORT_SYMBOL(ftape_disable); -EXPORT_SYMBOL(ftape_mmap); -EXPORT_SYMBOL(ftape_calibrate_data_rate); -/* from ftape-io.c */ -EXPORT_SYMBOL(ftape_reset_drive); -EXPORT_SYMBOL(ftape_command); -EXPORT_SYMBOL(ftape_parameter); -EXPORT_SYMBOL(ftape_ready_wait); -EXPORT_SYMBOL(ftape_report_operation); -EXPORT_SYMBOL(ftape_report_error); -/* from ftape-read.c */ -EXPORT_SYMBOL(ftape_read_segment_fraction); -EXPORT_SYMBOL(ftape_zap_read_buffers); -EXPORT_SYMBOL(ftape_read_header_segment); -EXPORT_SYMBOL(ftape_decode_header_segment); -/* from ftape-write.c */ -EXPORT_SYMBOL(ftape_write_segment); -EXPORT_SYMBOL(ftape_start_writing); -EXPORT_SYMBOL(ftape_loop_until_writes_done); -/* from ftape-buffer.h */ -EXPORT_SYMBOL(ftape_set_nr_buffers); -/* from ftape-format.h */ -EXPORT_SYMBOL(ftape_format_track); -EXPORT_SYMBOL(ftape_format_status); -EXPORT_SYMBOL(ftape_verify_segment); -/* from tracing.c */ -#ifndef CONFIG_FT_NO_TRACE_AT_ALL -EXPORT_SYMBOL(ftape_tracing); -EXPORT_SYMBOL(ftape_function_nest_level); -EXPORT_SYMBOL(ftape_trace_call); -EXPORT_SYMBOL(ftape_trace_exit); -EXPORT_SYMBOL(ftape_trace_log); -#endif - diff --git a/drivers/char/ftape/zftape/Makefile b/drivers/char/ftape/zftape/Makefile deleted file mode 100644 index 6d91c1f77c05..000000000000 --- a/drivers/char/ftape/zftape/Makefile +++ /dev/null @@ -1,36 +0,0 @@ -# -# Copyright (C) 1996, 1997 Claus-Justus Heine. -# -# This program is free software; you can redistribute it and/or modify -# it under the terms of the GNU General Public License as published by -# the Free Software Foundation; either version 2, or (at your option) -# any later version. -# -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -# GNU General Public License for more details. -# -# You should have received a copy of the GNU General Public License -# along with this program; see the file COPYING. If not, write to -# the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. -# -# $Source: /homes/cvs/ftape-stacked/ftape/zftape/Makefile,v $ -# $Revision: 1.4 $ -# $Date: 1997/10/05 19:18:58 $ -# -# Makefile for the QIC-40/80/3010/3020 zftape interface VFS to -# ftape -# - - -# ZFT_OBSOLETE - enable the MTIOC_ZFTAPE_GETBLKSZ ioctl. You should -# leave this enabled for compatibility with taper. - -obj-$(CONFIG_ZFTAPE) += zftape.o - -zftape-objs := zftape-rw.o zftape-ctl.o zftape-read.o \ - zftape-write.o zftape-vtbl.o zftape-eof.o \ - zftape-init.o zftape-buffers.o zftape_syms.o - -EXTRA_CFLAGS := -DZFT_OBSOLETE diff --git a/drivers/char/ftape/zftape/zftape-buffers.c b/drivers/char/ftape/zftape/zftape-buffers.c deleted file mode 100644 index 7ebce2ec7897..000000000000 --- a/drivers/char/ftape/zftape/zftape-buffers.c +++ /dev/null @@ -1,149 +0,0 @@ -/* - * Copyright (C) 1995-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-buffers.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:59 $ - * - * This file contains the dynamic buffer allocation routines - * of zftape - */ - -#include -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* global variables - */ - -/* local varibales - */ -static unsigned int used_memory; -static unsigned int peak_memory; - -void zft_memory_stats(void) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Memory usage (vmalloc allocations):\n" - KERN_INFO "total allocated: %d\n" - KERN_INFO "peak allocation: %d", - used_memory, peak_memory); - peak_memory = used_memory; - TRACE_EXIT; -} - -int zft_vcalloc_once(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - if (zft_vmalloc_once(new, size) < 0) { - TRACE_EXIT -ENOMEM; - } - memset(*(void **)new, '\0', size); - TRACE_EXIT 0; -} -int zft_vmalloc_once(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - - if (*(void **)new != NULL || size == 0) { - TRACE_EXIT 0; - } - if ((*(void **)new = vmalloc(size)) == NULL) { - TRACE_EXIT -ENOMEM; - } - used_memory += size; - if (peak_memory < used_memory) { - peak_memory = used_memory; - } - TRACE_ABORT(0, ft_t_noise, - "allocated buffer @ %p, %zd bytes", *(void **)new, size); -} -int zft_vmalloc_always(void *new, size_t size) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(new, size); - TRACE_EXIT zft_vmalloc_once(new, size); -} -void zft_vfree(void *old, size_t size) -{ - TRACE_FUN(ft_t_flow); - - if (*(void **)old) { - vfree(*(void **)old); - used_memory -= size; - TRACE(ft_t_noise, "released buffer @ %p, %zd bytes", - *(void **)old, size); - *(void **)old = NULL; - } - TRACE_EXIT; -} - -void *zft_kmalloc(size_t size) -{ - void *new; - - while ((new = kmalloc(size, GFP_KERNEL)) == NULL) { - msleep_interruptible(100); - } - memset(new, 0, size); - used_memory += size; - if (peak_memory < used_memory) { - peak_memory = used_memory; - } - return new; -} - -void zft_kfree(void *old, size_t size) -{ - kfree(old); - used_memory -= size; -} - -/* there are some more buffers that are allocated on demand. - * cleanup_module() calles this function to be sure to have released - * them - */ -void zft_uninit_mem(void) -{ - TRACE_FUN(ft_t_flow); - - zft_vfree(&zft_hseg_buf, FT_SEGMENT_SIZE); - zft_vfree(&zft_deblock_buf, FT_SEGMENT_SIZE); zft_deblock_segment = -1; - zft_free_vtbl(); - if (zft_cmpr_lock(0 /* don't load */) == 0) { - (*zft_cmpr_ops->cleanup)(); - (*zft_cmpr_ops->reset)(); /* unlock it again */ - } - zft_memory_stats(); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-buffers.h b/drivers/char/ftape/zftape/zftape-buffers.h deleted file mode 100644 index 798e3128c682..000000000000 --- a/drivers/char/ftape/zftape/zftape-buffers.h +++ /dev/null @@ -1,55 +0,0 @@ -#ifndef _FTAPE_DYNMEM_H -#define _FTAPE_DYNMEM_H - -/* - * Copyright (C) 1995-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-buffers.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:18:59 $ - * - * memory allocation routines. - * - */ - -/* we do not allocate all of the really large buffer memory before - * someone tries to open the drive. ftape_open() may fail with - * -ENOMEM, but that's better having 200k of vmalloced memory which - * cannot be swapped out. - */ - -extern void zft_memory_stats(void); -extern int zft_vmalloc_once(void *new, size_t size); -extern int zft_vcalloc_once(void *new, size_t size); -extern int zft_vmalloc_always(void *new, size_t size); -extern void zft_vfree(void *old, size_t size); -extern void *zft_kmalloc(size_t size); -extern void zft_kfree(void *old, size_t size); - -/* called by cleanup_module() - */ -extern void zft_uninit_mem(void); - -#endif - - - - - - - diff --git a/drivers/char/ftape/zftape/zftape-ctl.c b/drivers/char/ftape/zftape/zftape-ctl.c deleted file mode 100644 index 22ba0f5d00cf..000000000000 --- a/drivers/char/ftape/zftape/zftape-ctl.c +++ /dev/null @@ -1,1417 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-ctl.c,v $ - * $Revision: 1.2.6.2 $ - * $Date: 1997/11/14 18:07:33 $ - * - * This file contains the non-read/write zftape functions - * for the QIC-40/80/3010/3020 floppy-tape driver for Linux. - */ - -#include -#include -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ -int zft_write_protected; /* this is when cartridge rdonly or O_RDONLY */ -int zft_header_read; -int zft_offline; -unsigned int zft_unit; -int zft_resid; -int zft_mt_compression; - -/* Local vars. - */ -static int going_offline; - -typedef int (mt_fun)(int *argptr); -typedef int (*mt_funp)(int *argptr); -typedef struct -{ - mt_funp function; - unsigned offline : 1; /* op permitted if offline or no_tape */ - unsigned write_protected : 1; /* op permitted if write-protected */ - unsigned not_formatted : 1; /* op permitted if tape not formatted */ - unsigned raw_mode : 1; /* op permitted if zft_mode == 0 */ - unsigned need_idle_state : 1; /* need to call def_idle_state */ - char *name; -} fun_entry; - -static mt_fun mt_dummy, mt_reset, mt_fsr, mt_bsr, mt_rew, mt_offl, mt_nop, - mt_weof, mt_erase, mt_ras2, mt_setblk, mt_setdensity, - mt_seek, mt_tell, mt_reten, mt_eom, mt_fsf, mt_bsf, - mt_fsfm, mt_bsfm, mt_setdrvbuffer, mt_compression; - -static fun_entry mt_funs[]= -{ - {mt_reset , 1, 1, 1, 1, 0, "MT_RESET" }, /* 0 */ - {mt_fsf , 0, 1, 0, 0, 1, "MT_FSF" }, - {mt_bsf , 0, 1, 0, 0, 1, "MT_BSF" }, - {mt_fsr , 0, 1, 0, 1, 1, "MT_FSR" }, - {mt_bsr , 0, 1, 0, 1, 1, "MT_BSR" }, - {mt_weof , 0, 0, 0, 0, 0, "MT_WEOF" }, /* 5 */ - {mt_rew , 0, 1, 1, 1, 0, "MT_REW" }, - {mt_offl , 0, 1, 1, 1, 0, "MT_OFFL" }, - {mt_nop , 1, 1, 1, 1, 0, "MT_NOP" }, - {mt_reten , 0, 1, 1, 1, 0, "MT_RETEN" }, - {mt_bsfm , 0, 1, 0, 0, 1, "MT_BSFM" }, /* 10 */ - {mt_fsfm , 0, 1, 0, 0, 1, "MT_FSFM" }, - {mt_eom , 0, 1, 0, 0, 1, "MT_EOM" }, - {mt_erase , 0, 0, 0, 1, 0, "MT_ERASE" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_RAS1" }, - {mt_ras2 , 0, 0, 0, 1, 0, "MT_RAS2" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_RAS3" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_dummy , 1, 1, 1, 1, 0, "UNKNOWN" }, - {mt_setblk , 1, 1, 1, 1, 1, "MT_SETBLK"}, /* 20 */ - {mt_setdensity , 1, 1, 1, 1, 0, "MT_SETDENSITY"}, - {mt_seek , 0, 1, 0, 1, 1, "MT_SEEK" }, - {mt_dummy , 0, 1, 0, 1, 1, "MT_TELL" }, /* wr-only ?! */ - {mt_setdrvbuffer, 1, 1, 1, 1, 0, "MT_SETDRVBUFFER" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_FSS" }, /* 25 */ - {mt_dummy , 1, 1, 1, 1, 0, "MT_BSS" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_WSM" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_LOCK" }, - {mt_dummy , 1, 1, 1, 1, 0, "MT_UNLOCK"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_LOAD" }, /* 30 */ - {mt_dummy , 1, 1, 1, 1, 0, "MT_UNLOAD"}, - {mt_compression , 1, 1, 1, 0, 1, "MT_COMPRESSION"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_SETPART"}, - {mt_dummy , 1, 1, 1, 1, 0, "MT_MKPART"} -}; - -#define NR_MT_CMDS NR_ITEMS(mt_funs) - -void zft_reset_position(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - pos->seg_byte_pos = - pos->volume_pos = 0; - if (zft_header_read) { - /* need to keep track of the volume table and - * compression map. We therefor simply - * position at the beginning of the first - * volume. This covers old ftape archives as - * well has various flavours of the - * compression map segments. The worst case is - * that the compression map shows up as a - * additional volume in front of all others. - */ - pos->seg_pos = zft_find_volume(0)->start_seg; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - } else { - pos->tape_pos = 0; - pos->seg_pos = -1; - } - zft_just_before_eof = 0; - zft_deblock_segment = -1; - zft_io_state = zft_idle; - zft_zap_read_buffers(); - zft_prevent_flush(); - /* unlock the compresison module if it is loaded. - * The zero arg means not to try to load the module. - */ - if (zft_cmpr_lock(0) == 0) { - (*zft_cmpr_ops->reset)(); /* unlock */ - } - TRACE_EXIT; -} - -static void zft_init_driver(void) -{ - TRACE_FUN(ft_t_flow); - - zft_resid = - zft_header_read = - zft_old_ftape = - zft_offline = - zft_write_protected = - going_offline = - zft_mt_compression = - zft_header_changed = - zft_volume_table_changed = - zft_written_segments = 0; - zft_blk_sz = CONFIG_ZFT_DFLT_BLK_SZ; - zft_reset_position(&zft_pos); /* does most of the stuff */ - ftape_zap_read_buffers(); - ftape_set_state(idle); - TRACE_EXIT; -} - -int zft_def_idle_state(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (!zft_header_read) { - result = zft_read_header_segments(); - } else if ((result = zft_flush_buffers()) >= 0 && zft_qic_mode) { - /* don't move past eof - */ - (void)zft_close_volume(&zft_pos); - } - if (ftape_abort_operation() < 0) { - TRACE(ft_t_warn, "ftape_abort_operation() failed"); - result = -EIO; - } - /* clear remaining read buffers */ - zft_zap_read_buffers(); - zft_io_state = zft_idle; - TRACE_EXIT result; -} - -/***************************************************************************** - * * - * functions for the MTIOCTOP commands * - * * - *****************************************************************************/ - -static int mt_dummy(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - TRACE_EXIT -ENOSYS; -} - -static int mt_reset(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - (void)ftape_seek_to_bot(); - TRACE_CATCH(ftape_reset_drive(), - zft_init_driver(); zft_uninit_mem(); zft_offline = 1); - /* fake a re-open of the device. This will set all flage and - * allocate buffers as appropriate. The new tape condition will - * force the open routine to do anything we need. - */ - TRACE_CATCH(_zft_open(-1 /* fake reopen */, 0 /* dummy */),); - TRACE_EXIT 0; -} - -static int mt_fsf(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = zft_skip_volumes(*arg, &zft_pos); - zft_just_before_eof = 0; - TRACE_EXIT result; -} - -static int mt_bsf(int *arg) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (*arg != 0) { - result = zft_skip_volumes(-*arg + 1, &zft_pos); - } - TRACE_EXIT result; -} - -static int seek_block(__s64 data_offset, - __s64 block_increment, - zft_position *pos) -{ - int result = 0; - __s64 new_block_pos; - __s64 vol_block_count; - const zft_volinfo *volume; - int exceed; - TRACE_FUN(ft_t_flow); - - volume = zft_find_volume(pos->seg_pos); - if (volume->start_seg == 0 || volume->end_seg == 0) { - TRACE_EXIT -EIO; - } - new_block_pos = (zft_div_blksz(data_offset, volume->blk_sz) - + block_increment); - vol_block_count = zft_div_blksz(volume->size, volume->blk_sz); - if (new_block_pos < 0) { - TRACE(ft_t_noise, - "new_block_pos " LL_X " < 0", LL(new_block_pos)); - zft_resid = (int)new_block_pos; - new_block_pos = 0; - exceed = 1; - } else if (new_block_pos > vol_block_count) { - TRACE(ft_t_noise, - "new_block_pos " LL_X " exceeds size of volume " LL_X, - LL(new_block_pos), LL(vol_block_count)); - zft_resid = (int)(vol_block_count - new_block_pos); - new_block_pos = vol_block_count; - exceed = 1; - } else { - exceed = 0; - } - if (zft_use_compression && volume->use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - result = (*zft_cmpr_ops->seek)(new_block_pos, pos, volume, - zft_deblock_buf); - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - pos->tape_pos += pos->seg_byte_pos; - } else { - pos->volume_pos = zft_mul_blksz(new_block_pos, volume->blk_sz); - pos->tape_pos = zft_calc_tape_pos(volume->start_seg); - pos->tape_pos += pos->volume_pos; - pos->seg_pos = zft_calc_seg_byte_coord(&pos->seg_byte_pos, - pos->tape_pos); - } - zft_just_before_eof = volume->size == pos->volume_pos; - if (zft_just_before_eof) { - /* why this? because zft_file_no checks agains start - * and end segment of a volume. We do not want to - * advance to the next volume with this function. - */ - TRACE(ft_t_noise, "set zft_just_before_eof"); - zft_position_before_eof(pos, volume); - } - TRACE(ft_t_noise, "\n" - KERN_INFO "new_seg_pos : %d\n" - KERN_INFO "new_tape_pos: " LL_X "\n" - KERN_INFO "vol_size : " LL_X "\n" - KERN_INFO "seg_byte_pos: %d\n" - KERN_INFO "blk_sz : %d", - pos->seg_pos, LL(pos->tape_pos), - LL(volume->size), pos->seg_byte_pos, - volume->blk_sz); - if (!exceed) { - zft_resid = new_block_pos - zft_div_blksz(pos->volume_pos, - volume->blk_sz); - } - if (zft_resid < 0) { - zft_resid = -zft_resid; - } - TRACE_EXIT ((exceed || zft_resid != 0) && result >= 0) ? -EINVAL : result; -} - -static int mt_fsr(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = seek_block(zft_pos.volume_pos, *arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_bsr(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = seek_block(zft_pos.volume_pos, -*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_weof(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(zft_flush_buffers(),); - result = zft_weof(*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_rew(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - if(zft_header_read) { - (void)zft_def_idle_state(); - } - result = ftape_seek_to_bot(); - zft_reset_position(&zft_pos); - TRACE_EXIT result; -} - -static int mt_offl(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - going_offline= 1; - result = mt_rew(NULL); - TRACE_EXIT result; -} - -static int mt_nop(int *dummy) -{ - TRACE_FUN(ft_t_flow); - /* should we set tape status? - */ - if (!zft_offline) { /* offline includes no_tape */ - (void)zft_def_idle_state(); - } - TRACE_EXIT 0; -} - -static int mt_reten(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - if(zft_header_read) { - (void)zft_def_idle_state(); - } - result = ftape_seek_to_eot(); - if (result >= 0) { - result = ftape_seek_to_bot(); - } - TRACE_EXIT(result); -} - -static int fsfbsfm(int arg, zft_position *pos) -{ - const zft_volinfo *vtbl; - __s64 block_pos; - TRACE_FUN(ft_t_flow); - - /* What to do? This should seek to the next file-mark and - * position BEFORE. That is, a next write would just extend - * the current file. Well. Let's just seek to the end of the - * current file, if count == 1. If count > 1, then do a - * "mt_fsf(count - 1)", and then seek to the end of that file. - * If count == 0, do nothing - */ - if (arg == 0) { - TRACE_EXIT 0; - } - zft_just_before_eof = 0; - TRACE_CATCH(zft_skip_volumes(arg < 0 ? arg : arg-1, pos), - if (arg > 0) { - zft_resid ++; - }); - vtbl = zft_find_volume(pos->seg_pos); - block_pos = zft_div_blksz(vtbl->size, vtbl->blk_sz); - (void)seek_block(0, block_pos, pos); - if (pos->volume_pos != vtbl->size) { - zft_just_before_eof = 0; - zft_resid = 1; - /* we didn't managed to go there */ - TRACE_ABORT(-EIO, ft_t_err, - "wanted file position " LL_X ", arrived at " LL_X, - LL(vtbl->size), LL(pos->volume_pos)); - } - zft_just_before_eof = 1; - TRACE_EXIT 0; -} - -static int mt_bsfm(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = fsfbsfm(-*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_fsfm(int *arg) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = fsfbsfm(*arg, &zft_pos); - TRACE_EXIT result; -} - -static int mt_eom(int *dummy) -{ - TRACE_FUN(ft_t_flow); - - zft_skip_to_eom(&zft_pos); - TRACE_EXIT 0; -} - -static int mt_erase(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = zft_erase(); - TRACE_EXIT result; -} - -static int mt_ras2(int *dummy) -{ - int result; - TRACE_FUN(ft_t_flow); - - result = -ENOSYS; - TRACE_EXIT result; -} - -/* Sets the new blocksize in BYTES - * - */ -static int mt_setblk(int *new_size) -{ - TRACE_FUN(ft_t_flow); - - if((unsigned int)(*new_size) > ZFT_MAX_BLK_SZ) { - TRACE_ABORT(-EINVAL, ft_t_info, - "desired blk_sz (%d) should be <= %d bytes", - *new_size, ZFT_MAX_BLK_SZ); - } - if ((*new_size & (FT_SECTOR_SIZE-1)) != 0) { - TRACE_ABORT(-EINVAL, ft_t_info, - "desired blk_sz (%d) must be a multiple of %d bytes", - *new_size, FT_SECTOR_SIZE); - } - if (*new_size == 0) { - if (zft_use_compression) { - TRACE_ABORT(-EINVAL, ft_t_info, - "Variable block size not yet " - "supported with compression"); - } - *new_size = 1; - } - zft_blk_sz = *new_size; - TRACE_EXIT 0; -} - -static int mt_setdensity(int *arg) -{ - TRACE_FUN(ft_t_flow); - - SET_TRACE_LEVEL(*arg); - TRACE(TRACE_LEVEL, "tracing set to %d", TRACE_LEVEL); - if ((int)TRACE_LEVEL != *arg) { - TRACE_EXIT -EINVAL; - } - TRACE_EXIT 0; -} - -static int mt_seek(int *new_block_pos) -{ - int result= 0; - TRACE_FUN(ft_t_any); - - result = seek_block(0, (__s64)*new_block_pos, &zft_pos); - TRACE_EXIT result; -} - -/* OK, this is totally different from SCSI, but the worst thing that can - * happen is that there is not enough defragmentated memory that can be - * allocated. Also, there is a hardwired limit of 16 dma buffers in the - * stock ftape module. This shouldn't bring the system down. - * - * NOTE: the argument specifies the total number of dma buffers to use. - * The driver needs at least 3 buffers to function at all. - * - */ -static int mt_setdrvbuffer(int *cnt) -{ - TRACE_FUN(ft_t_flow); - - if (*cnt < 3) { - TRACE_EXIT -EINVAL; - } - TRACE_CATCH(ftape_set_nr_buffers(*cnt),); - TRACE_EXIT 0; -} -/* return the block position from start of volume - */ -static int mt_tell(int *arg) -{ - TRACE_FUN(ft_t_flow); - - *arg = zft_div_blksz(zft_pos.volume_pos, - zft_find_volume(zft_pos.seg_pos)->blk_sz); - TRACE_EXIT 0; -} - -static int mt_compression(int *arg) -{ - TRACE_FUN(ft_t_flow); - - /* Ok. We could also check whether compression is available at - * all by trying to load the compression module. We could - * also check for a block size of 1 byte which is illegal - * with compression. Instead of doing it here we rely on - * zftape_write() to do the proper checks. - */ - if ((unsigned int)*arg > 1) { - TRACE_EXIT -EINVAL; - } - if (*arg != 0 && zft_blk_sz == 1) { /* variable block size */ - TRACE_ABORT(-EINVAL, ft_t_info, - "Compression not yet supported " - "with variable block size"); - } - zft_mt_compression = *arg; - if ((zft_unit & ZFT_ZIP_MODE) == 0) { - zft_use_compression = zft_mt_compression; - } - TRACE_EXIT 0; -} - -/* check whether write access is allowed. Write access is denied when - * + zft_write_protected == 1 -- this accounts for either hard write - * protection of the cartridge or for - * O_RDONLY access mode of the tape device - * + zft_offline == 1 -- this meany that there is either no tape - * or that the MTOFFLINE ioctl has been - * previously issued (`soft eject') - * + ft_formatted == 0 -- this means that the cartridge is not - * formatted - * Then we distinuguish two cases. When zft_qic_mode is TRUE, then we try - * to emulate a `traditional' (aka SCSI like) UN*X tape device. Therefore we - * deny writes when - * + zft_qic_mode ==1 && - * (!zft_tape_at_lbot() && -- tape no at logical BOT - * !(zft_tape_at_eom() || -- tape not at logical EOM (or EOD) - * (zft_tape_at_eom() && - * zft_old_ftape()))) -- we can't add new volume to tapes - * written by old ftape because ftape - * don't use the volume table - * - * when the drive is in true raw mode (aka /dev/rawft0) then we don't - * care about LBOT and EOM conditions. This device is intended for a - * user level program that wants to truly implement the QIC-80 compliance - * at the logical data layout level of the cartridge, i.e. implement all - * that volume table and volume directory stuff etc.< - */ -int zft_check_write_access(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - if (zft_offline) { /* offline includes no_tape */ - TRACE_ABORT(-ENXIO, - ft_t_info, "tape is offline or no cartridge"); - } - if (!ft_formatted) { - TRACE_ABORT(-EACCES, ft_t_info, "tape is not formatted"); - } - if (zft_write_protected) { - TRACE_ABORT(-EACCES, ft_t_info, "cartridge write protected"); - } - if (zft_qic_mode) { - /* check BOT condition */ - if (!zft_tape_at_lbot(pos)) { - /* protect cartridges written by old ftape if - * not at BOT because they use the vtbl - * segment for storing data - */ - if (zft_old_ftape) { - TRACE_ABORT(-EACCES, ft_t_warn, - "Cannot write to cartridges written by old ftape when not at BOT"); - } - /* not at BOT, but allow writes at EOD, of course - */ - if (!zft_tape_at_eod(pos)) { - TRACE_ABORT(-EACCES, ft_t_info, - "tape not at BOT and not at EOD"); - } - } - /* fine. Now the tape is either at BOT or at EOD. */ - } - /* or in raw mode in which case we don't care about BOT and EOD */ - TRACE_EXIT 0; -} - -/* OPEN routine called by kernel-interface code - * - * NOTE: this is also called by mt_reset() with dev_minor == -1 - * to fake a reopen after a reset. - */ -int _zft_open(unsigned int dev_minor, unsigned int access_mode) -{ - static unsigned int tape_unit; - static unsigned int file_access_mode; - int result; - TRACE_FUN(ft_t_flow); - - if ((int)dev_minor == -1) { - /* fake reopen */ - zft_unit = tape_unit; - access_mode = file_access_mode; - zft_init_driver(); /* reset all static data to defaults */ - } else { - tape_unit = dev_minor; - file_access_mode = access_mode; - if ((result = ftape_enable(FTAPE_SEL(dev_minor))) < 0) { - TRACE_ABORT(-ENXIO, ft_t_err, - "ftape_enable failed: %d", result); - } - if (ft_new_tape || ft_no_tape || !ft_formatted || - (FTAPE_SEL(zft_unit) != FTAPE_SEL(dev_minor)) || - (zft_unit & ZFT_RAW_MODE) != (dev_minor & ZFT_RAW_MODE)) { - /* reset all static data to defaults, - */ - zft_init_driver(); - } - zft_unit = dev_minor; - } - zft_set_flags(zft_unit); /* decode the minor bits */ - if (zft_blk_sz == 1 && zft_use_compression) { - ftape_disable(); /* resets ft_no_tape */ - TRACE_ABORT(-ENODEV, ft_t_warn, "Variable block size not yet " - "supported with compression"); - } - /* no need for most of the buffers when no tape or not - * formatted. for the read/write operations, it is the - * regardless whether there is no tape, a not-formatted tape - * or the whether the driver is soft offline. - * Nevertheless we allow some ioctls with non-formatted tapes, - * like rewind and reset. - */ - if (ft_no_tape || !ft_formatted) { - zft_uninit_mem(); - } - if (ft_no_tape) { - zft_offline = 1; /* so we need not test two variables */ - } - if ((access_mode == O_WRONLY || access_mode == O_RDWR) && - (ft_write_protected || ft_no_tape)) { - ftape_disable(); /* resets ft_no_tape */ - TRACE_ABORT(ft_no_tape ? -ENXIO : -EROFS, - ft_t_warn, "wrong access mode %s cartridge", - ft_no_tape ? "without a" : "with write protected"); - } - zft_write_protected = (access_mode == O_RDONLY || - ft_write_protected != 0); - if (zft_write_protected) { - TRACE(ft_t_noise, - "read only access mode: %d, " - "drive write protected: %d", - access_mode == O_RDONLY, - ft_write_protected != 0); - } - if (!zft_offline) { - TRACE_CATCH(zft_vmalloc_once(&zft_deblock_buf,FT_SEGMENT_SIZE), - ftape_disable()); - } - /* zft_seg_pos should be greater than the vtbl segpos but not - * if in compatibility mode and only after we read in the - * header segments - * - * might also be a problem if the user makes a backup with a - * *qft* device and rewinds it with a raw device. - */ - if (zft_qic_mode && - !zft_old_ftape && - zft_pos.seg_pos >= 0 && - zft_header_read && - zft_pos.seg_pos <= ft_first_data_segment) { - TRACE(ft_t_noise, "you probably mixed up the zftape devices!"); - zft_reset_position(&zft_pos); - } - TRACE_EXIT 0; -} - -/* RELEASE routine called by kernel-interface code - */ -int _zft_close(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (zft_offline) { - /* call the hardware release routine. Puts the drive offline */ - ftape_disable(); - TRACE_EXIT 0; - } - if (!(ft_write_protected || zft_old_ftape)) { - result = zft_flush_buffers(); - TRACE(ft_t_noise, "writing file mark at current position"); - if (zft_qic_mode && zft_close_volume(&zft_pos) == 0) { - zft_move_past_eof(&zft_pos); - } - if ((zft_tape_at_lbot(&zft_pos) || - !(zft_unit & FTAPE_NO_REWIND))) { - if (result >= 0) { - result = zft_update_header_segments(); - } else { - TRACE(ft_t_err, - "Error: unable to update header segments"); - } - } - } - ftape_abort_operation(); - if (!(zft_unit & FTAPE_NO_REWIND)) { - TRACE(ft_t_noise, "rewinding tape"); - if (ftape_seek_to_bot() < 0 && result >= 0) { - result = -EIO; /* keep old value */ - } - zft_reset_position(&zft_pos); - } - zft_zap_read_buffers(); - /* now free up memory as much as possible. We don't destroy - * the deblock buffer if it containes a valid segment. - */ - if (zft_deblock_segment == -1) { - zft_vfree(&zft_deblock_buf, FT_SEGMENT_SIZE); - } - /* high level driver status, forces creation of a new volume - * when calling ftape_write again and not zft_just_before_eof - */ - zft_io_state = zft_idle; - if (going_offline) { - zft_init_driver(); - zft_uninit_mem(); - going_offline = 0; - zft_offline = 1; - } else if (zft_cmpr_lock(0 /* don't load */) == 0) { - (*zft_cmpr_ops->reset)(); /* unlock it again */ - } - zft_memory_stats(); - /* call the hardware release routine. Puts the drive offline */ - ftape_disable(); - TRACE_EXIT result; -} - -/* - * the wrapper function around the wrapper MTIOCTOP ioctl - */ -static int mtioctop(struct mtop *mtop, int arg_size) -{ - int result = 0; - fun_entry *mt_fun_entry; - TRACE_FUN(ft_t_flow); - - if (arg_size != sizeof(struct mtop) || mtop->mt_op >= NR_MT_CMDS) { - TRACE_EXIT -EINVAL; - } - TRACE(ft_t_noise, "calling MTIOCTOP command: %s", - mt_funs[mtop->mt_op].name); - mt_fun_entry= &mt_funs[mtop->mt_op]; - zft_resid = mtop->mt_count; - if (!mt_fun_entry->offline && zft_offline) { - if (ft_no_tape) { - TRACE_ABORT(-ENXIO, ft_t_info, "no tape present"); - } else { - TRACE_ABORT(-ENXIO, ft_t_info, "drive is offline"); - } - } - if (!mt_fun_entry->not_formatted && !ft_formatted) { - TRACE_ABORT(-EACCES, ft_t_info, "tape is not formatted"); - } - if (!mt_fun_entry->write_protected) { - TRACE_CATCH(zft_check_write_access(&zft_pos),); - } - if (mt_fun_entry->need_idle_state && !(zft_offline || !ft_formatted)) { - TRACE_CATCH(zft_def_idle_state(),); - } - if (!zft_qic_mode && !mt_fun_entry->raw_mode) { - TRACE_ABORT(-EACCES, ft_t_info, -"Drive needs to be in QIC-80 compatibility mode for this command"); - } - result = (mt_fun_entry->function)(&mtop->mt_count); - if (zft_tape_at_lbot(&zft_pos)) { - TRACE_CATCH(zft_update_header_segments(),); - } - if (result >= 0) { - zft_resid = 0; - } - TRACE_EXIT result; -} - -/* - * standard MTIOCGET ioctl - */ -static int mtiocget(struct mtget *mtget, int arg_size) -{ - const zft_volinfo *volume; - __s64 max_tape_pos; - TRACE_FUN(ft_t_flow); - - if (arg_size != sizeof(struct mtget)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - mtget->mt_type = ft_drive_type.vendor_id + 0x800000; - mtget->mt_dsreg = ft_last_status.space; - mtget->mt_erreg = ft_last_error.space; /* error register */ - mtget->mt_resid = zft_resid; /* residuum of writes, reads and - * MTIOCTOP commands - */ - if (!zft_offline) { /* neither no_tape nor soft offline */ - mtget->mt_gstat = GMT_ONLINE(~0UL); - /* should rather return the status of the cartridge - * than the access mode of the file, therefor use - * ft_write_protected, not zft_write_protected - */ - if (ft_write_protected) { - mtget->mt_gstat |= GMT_WR_PROT(~0UL); - } - if(zft_header_read) { /* this catches non-formatted */ - volume = zft_find_volume(zft_pos.seg_pos); - mtget->mt_fileno = volume->count; - max_tape_pos = zft_capacity - zft_blk_sz; - if (zft_use_compression) { - max_tape_pos -= ZFT_CMPR_OVERHEAD; - } - if (zft_tape_at_eod(&zft_pos)) { - mtget->mt_gstat |= GMT_EOD(~0UL); - } - if (zft_pos.tape_pos > max_tape_pos) { - mtget->mt_gstat |= GMT_EOT(~0UL); - } - mtget->mt_blkno = zft_div_blksz(zft_pos.volume_pos, - volume->blk_sz); - if (zft_just_before_eof) { - mtget->mt_gstat |= GMT_EOF(~0UL); - } - if (zft_tape_at_lbot(&zft_pos)) { - mtget->mt_gstat |= GMT_BOT(~0UL); - } - } else { - mtget->mt_fileno = mtget->mt_blkno = -1; - if (mtget->mt_dsreg & QIC_STATUS_AT_BOT) { - mtget->mt_gstat |= GMT_BOT(~0UL); - } - } - } else { - if (ft_no_tape) { - mtget->mt_gstat = GMT_DR_OPEN(~0UL); - } else { - mtget->mt_gstat = 0UL; - } - mtget->mt_fileno = mtget->mt_blkno = -1; - } - TRACE_EXIT 0; -} - -#ifdef MTIOCRDFTSEG -/* - * Read a floppy tape segment. This is useful for manipulating the - * volume table, and read the old header segment before re-formatting - * the cartridge. - */ -static int mtiocrdftseg(struct mtftseg * mtftseg, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCRDFTSEG"); - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (arg_size != sizeof(struct mtftseg)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_RD_SINGLE && - mtftseg->mt_mode != FT_RD_AHEAD) { - TRACE_ABORT(-EINVAL, ft_t_info, "invalid read mode"); - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; /* -ENXIO ? */ - - } - if (!zft_header_read) { - TRACE_CATCH(zft_def_idle_state(),); - } - if (mtftseg->mt_segno > ft_last_data_segment) { - TRACE_ABORT(-EINVAL, ft_t_info, "segment number is too large"); - } - mtftseg->mt_result = ftape_read_segment(mtftseg->mt_segno, - zft_deblock_buf, - mtftseg->mt_mode); - if (mtftseg->mt_result < 0) { - /* a negativ result is not an ioctl error. if - * the user wants to read damaged tapes, - * it's up to her/him - */ - TRACE_EXIT 0; - } - if (copy_to_user(mtftseg->mt_data, - zft_deblock_buf, - mtftseg->mt_result) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCWRFTSEG -/* - * write a floppy tape segment. This version features writing of - * deleted address marks, and gracefully ignores the (software) - * ft_formatted flag to support writing of header segments after - * formatting. - */ -static int mtiocwrftseg(struct mtftseg * mtftseg, int arg_size) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCWRFTSEG"); - if (zft_write_protected || zft_qic_mode) { - TRACE_EXIT -EACCES; - } - if (arg_size != sizeof(struct mtftseg)) { - TRACE_ABORT(-EINVAL, ft_t_info, "bad argument size: %d", - arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_WR_ASYNC && - mtftseg->mt_mode != FT_WR_MULTI && - mtftseg->mt_mode != FT_WR_SINGLE && - mtftseg->mt_mode != FT_WR_DELETE) { - TRACE_ABORT(-EINVAL, ft_t_info, "invalid write mode"); - } - /* - * We don't check for ft_formatted, because this gives - * only the software status of the driver. - * - * We assume that the user knows what it is - * doing. And rely on the low level stuff to fail - * when the tape isn't formatted. We only make sure - * that The header segment buffer is allocated, - * because it holds the bad sector map. - */ - if (zft_hseg_buf == NULL) { - TRACE_EXIT -ENXIO; - } - if (mtftseg->mt_mode != FT_WR_DELETE) { - if (copy_from_user(zft_deblock_buf, - mtftseg->mt_data, - FT_SEGMENT_SIZE) != 0) { - TRACE_EXIT -EFAULT; - } - } - mtftseg->mt_result = ftape_write_segment(mtftseg->mt_segno, - zft_deblock_buf, - mtftseg->mt_mode); - if (mtftseg->mt_result >= 0 && mtftseg->mt_mode == FT_WR_SINGLE) { - /* - * a negativ result is not an ioctl error. if - * the user wants to write damaged tapes, - * it's up to her/him - */ - if ((result = ftape_loop_until_writes_done()) < 0) { - mtftseg->mt_result = result; - } - } - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCVOLINFO -/* - * get information about volume positioned at. - */ -static int mtiocvolinfo(struct mtvolinfo *volinfo, int arg_size) -{ - const zft_volinfo *volume; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCVOLINFO"); - if (arg_size != sizeof(struct mtvolinfo)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - volume = zft_find_volume(zft_pos.seg_pos); - volinfo->mt_volno = volume->count; - volinfo->mt_blksz = volume->blk_sz == 1 ? 0 : volume->blk_sz; - volinfo->mt_size = volume->size >> 10; - volinfo->mt_rawsize = ((zft_calc_tape_pos(volume->end_seg + 1) >> 10) - - (zft_calc_tape_pos(volume->start_seg) >> 10)); - volinfo->mt_cmpr = volume->use_compression; - TRACE_EXIT 0; -} -#endif - -#ifdef ZFT_OBSOLETE -static int mtioc_zftape_getblksz(struct mtblksz *blksz, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "\n" - KERN_INFO "Mag tape ioctl command: MTIOC_ZTAPE_GETBLKSZ\n" - KERN_INFO "This ioctl is here merely for compatibility.\n" - KERN_INFO "Please use MTIOCVOLINFO instead"); - if (arg_size != sizeof(struct mtblksz)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - blksz->mt_blksz = zft_find_volume(zft_pos.seg_pos)->blk_sz; - TRACE_EXIT 0; -} -#endif - -#ifdef MTIOCGETSIZE -/* - * get the capacity of the tape cartridge. - */ -static int mtiocgetsize(struct mttapesize *size, int arg_size) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOC_ZFTAPE_GETSIZE"); - if (arg_size != sizeof(struct mttapesize)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (zft_offline) { - TRACE_EXIT -ENXIO; - } - if (!ft_formatted) { - TRACE_EXIT -EACCES; - } - TRACE_CATCH(zft_def_idle_state(),); - size->mt_capacity = (unsigned int)(zft_capacity>>10); - size->mt_used = (unsigned int)(zft_get_eom_pos()>>10); - TRACE_EXIT 0; -} -#endif - -static int mtiocpos(struct mtpos *mtpos, int arg_size) -{ - int result; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCPOS"); - if (arg_size != sizeof(struct mtpos)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - result = mt_tell((int *)&mtpos->mt_blkno); - TRACE_EXIT result; -} - -#ifdef MTIOCFTFORMAT -/* - * formatting of floppy tape cartridges. This is intended to be used - * together with the MTIOCFTCMD ioctl and the new mmap feature - */ - -/* - * This function uses ftape_decode_header_segment() to inform the low - * level ftape module about the new parameters. - * - * It erases the hseg_buf. The calling process must specify all - * parameters to assure proper operation. - * - * return values: -EINVAL - wrong argument size - * -EINVAL - if ftape_decode_header_segment() failed. - */ -static int set_format_parms(struct ftfmtparms *p, __u8 *hseg_buf) -{ - ft_trace_t old_level = TRACE_LEVEL; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "MTIOCFTFORMAT operation FTFMT_SETPARMS"); - memset(hseg_buf, 0, FT_SEGMENT_SIZE); - PUT4(hseg_buf, FT_SIGNATURE, FT_HSEG_MAGIC); - - /* fill in user specified parameters - */ - hseg_buf[FT_FMT_CODE] = (__u8)p->ft_fmtcode; - PUT2(hseg_buf, FT_SPT, p->ft_spt); - hseg_buf[FT_TPC] = (__u8)p->ft_tpc; - hseg_buf[FT_FHM] = (__u8)p->ft_fhm; - hseg_buf[FT_FTM] = (__u8)p->ft_ftm; - - /* fill in sane defaults to make ftape happy. - */ - hseg_buf[FT_FSM] = (__u8)128; /* 128 is hard wired all over ftape */ - if (p->ft_fmtcode == fmt_big) { - PUT4(hseg_buf, FT_6_HSEG_1, 0); - PUT4(hseg_buf, FT_6_HSEG_2, 1); - PUT4(hseg_buf, FT_6_FRST_SEG, 2); - PUT4(hseg_buf, FT_6_LAST_SEG, p->ft_spt * p->ft_tpc - 1); - } else { - PUT2(hseg_buf, FT_HSEG_1, 0); - PUT2(hseg_buf, FT_HSEG_2, 1); - PUT2(hseg_buf, FT_FRST_SEG, 2); - PUT2(hseg_buf, FT_LAST_SEG, p->ft_spt * p->ft_tpc - 1); - } - - /* Synchronize with the low level module. This is particularly - * needed for unformatted cartridges as the QIC std was previously - * unknown BUT is needed to set data rate and to calculate timeouts. - */ - TRACE_CATCH(ftape_calibrate_data_rate(p->ft_qicstd&QIC_TAPE_STD_MASK), - _res = -EINVAL); - - /* The following will also recalcualte the timeouts for the tape - * length and QIC std we want to format to. - * abort with -EINVAL rather than -EIO - */ - SET_TRACE_LEVEL(ft_t_warn); - TRACE_CATCH(ftape_decode_header_segment(hseg_buf), - SET_TRACE_LEVEL(old_level); _res = -EINVAL); - SET_TRACE_LEVEL(old_level); - TRACE_EXIT 0; -} - -/* - * Return the internal SOFTWARE status of the kernel driver. This does - * NOT query the tape drive about its status. - */ -static int get_format_parms(struct ftfmtparms *p, __u8 *hseg_buffer) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "MTIOCFTFORMAT operation FTFMT_GETPARMS"); - p->ft_qicstd = ft_qic_std; - p->ft_fmtcode = ft_format_code; - p->ft_fhm = hseg_buffer[FT_FHM]; - p->ft_ftm = hseg_buffer[FT_FTM]; - p->ft_spt = ft_segments_per_track; - p->ft_tpc = ft_tracks_per_tape; - TRACE_EXIT 0; -} - -static int mtiocftformat(struct mtftformat *mtftformat, int arg_size) -{ - int result; - union fmt_arg *arg = &mtftformat->fmt_arg; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCFTFORMAT"); - if (zft_offline) { - if (ft_no_tape) { - TRACE_ABORT(-ENXIO, ft_t_info, "no tape present"); - } else { - TRACE_ABORT(-ENXIO, ft_t_info, "drive is offline"); - } - } - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (zft_hseg_buf == NULL) { - TRACE_CATCH(zft_vcalloc_once(&zft_hseg_buf, FT_SEGMENT_SIZE),); - } - zft_header_read = 0; - switch(mtftformat->fmt_op) { - case FTFMT_SET_PARMS: - TRACE_CATCH(set_format_parms(&arg->fmt_parms, zft_hseg_buf),); - TRACE_EXIT 0; - case FTFMT_GET_PARMS: - TRACE_CATCH(get_format_parms(&arg->fmt_parms, zft_hseg_buf),); - TRACE_EXIT 0; - case FTFMT_FORMAT_TRACK: - if ((ft_formatted && zft_check_write_access(&zft_pos) < 0) || - (!ft_formatted && zft_write_protected)) { - TRACE_ABORT(-EACCES, ft_t_info, "Write access denied"); - } - TRACE_CATCH(ftape_format_track(arg->fmt_track.ft_track, - arg->fmt_track.ft_gap3),); - TRACE_EXIT 0; - case FTFMT_STATUS: - TRACE_CATCH(ftape_format_status(&arg->fmt_status.ft_segment),); - TRACE_EXIT 0; - case FTFMT_VERIFY: - TRACE_CATCH(ftape_verify_segment(arg->fmt_verify.ft_segment, - (SectorMap *)&arg->fmt_verify.ft_bsm),); - TRACE_EXIT 0; - default: - TRACE_ABORT(-EINVAL, ft_t_err, "Invalid format operation"); - } - TRACE_EXIT result; -} -#endif - -#ifdef MTIOCFTCMD -/* - * send a QIC-117 command to the drive, with optional timeouts, - * parameter and result bits. This is intended to be used together - * with the formatting ioctl. - */ -static int mtiocftcmd(struct mtftcmd *ftcmd, int arg_size) -{ - int i; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "Mag tape ioctl command: MTIOCFTCMD"); - if (!capable(CAP_SYS_ADMIN)) { - TRACE_ABORT(-EPERM, ft_t_info, - "need CAP_SYS_ADMIN capability to send raw qic-117 commands"); - } - if (zft_qic_mode) { - TRACE_ABORT(-EACCES, ft_t_info, - "driver needs to be in raw mode for this ioctl"); - } - if (arg_size != sizeof(struct mtftcmd)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (ftcmd->ft_wait_before) { - TRACE_CATCH(ftape_ready_wait(ftcmd->ft_wait_before, - &ftcmd->ft_status),); - } - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - if (ftcmd->ft_result_bits != 0) { - TRACE_CATCH(ftape_report_operation(&ftcmd->ft_result, - ftcmd->ft_cmd, - ftcmd->ft_result_bits),); - } else { - TRACE_CATCH(ftape_command(ftcmd->ft_cmd),); - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - for (i = 0; i < ftcmd->ft_parm_cnt; i++) { - TRACE_CATCH(ftape_parameter(ftcmd->ft_parms[i]&0x0f),); - if (ftcmd->ft_status & QIC_STATUS_ERROR) - goto ftmtcmd_error; - } - } - if (ftcmd->ft_wait_after != 0) { - TRACE_CATCH(ftape_ready_wait(ftcmd->ft_wait_after, - &ftcmd->ft_status),); - } -ftmtcmd_error: - if (ftcmd->ft_status & QIC_STATUS_ERROR) { - TRACE(ft_t_noise, "error status set"); - TRACE_CATCH(ftape_report_error(&ftcmd->ft_error, - &ftcmd->ft_cmd, 1),); - } - TRACE_EXIT 0; /* this is not an i/o error */ -} -#endif - -/* IOCTL routine called by kernel-interface code - */ -int _zft_ioctl(unsigned int command, void __user * arg) -{ - int result; - union { struct mtop mtop; - struct mtget mtget; - struct mtpos mtpos; -#ifdef MTIOCRDFTSEG - struct mtftseg mtftseg; -#endif -#ifdef MTIOCVOLINFO - struct mtvolinfo mtvolinfo; -#endif -#ifdef MTIOCGETSIZE - struct mttapesize mttapesize; -#endif -#ifdef MTIOCFTFORMAT - struct mtftformat mtftformat; -#endif -#ifdef ZFT_OBSOLETE - struct mtblksz mtblksz; -#endif -#ifdef MTIOCFTCMD - struct mtftcmd mtftcmd; -#endif - } krnl_arg; - int arg_size = _IOC_SIZE(command); - int dir = _IOC_DIR(command); - TRACE_FUN(ft_t_flow); - - /* This check will only catch arguments that are too large ! - */ - if (dir & (_IOC_READ | _IOC_WRITE) && arg_size > sizeof(krnl_arg)) { - TRACE_ABORT(-EINVAL, - ft_t_info, "bad argument size: %d", arg_size); - } - if (dir & _IOC_WRITE) { - if (copy_from_user(&krnl_arg, arg, arg_size) != 0) { - TRACE_EXIT -EFAULT; - } - } - TRACE(ft_t_flow, "called with ioctl command: 0x%08x", command); - switch (command) { - case MTIOCTOP: - result = mtioctop(&krnl_arg.mtop, arg_size); - break; - case MTIOCGET: - result = mtiocget(&krnl_arg.mtget, arg_size); - break; - case MTIOCPOS: - result = mtiocpos(&krnl_arg.mtpos, arg_size); - break; -#ifdef MTIOCVOLINFO - case MTIOCVOLINFO: - result = mtiocvolinfo(&krnl_arg.mtvolinfo, arg_size); - break; -#endif -#ifdef ZFT_OBSOLETE - case MTIOC_ZFTAPE_GETBLKSZ: - result = mtioc_zftape_getblksz(&krnl_arg.mtblksz, arg_size); - break; -#endif -#ifdef MTIOCRDFTSEG - case MTIOCRDFTSEG: /* read a segment via ioctl */ - result = mtiocrdftseg(&krnl_arg.mtftseg, arg_size); - break; -#endif -#ifdef MTIOCWRFTSEG - case MTIOCWRFTSEG: /* write a segment via ioctl */ - result = mtiocwrftseg(&krnl_arg.mtftseg, arg_size); - break; -#endif -#ifdef MTIOCGETSIZE - case MTIOCGETSIZE: - result = mtiocgetsize(&krnl_arg.mttapesize, arg_size); - break; -#endif -#ifdef MTIOCFTFORMAT - case MTIOCFTFORMAT: - result = mtiocftformat(&krnl_arg.mtftformat, arg_size); - break; -#endif -#ifdef MTIOCFTCMD - case MTIOCFTCMD: - result = mtiocftcmd(&krnl_arg.mtftcmd, arg_size); - break; -#endif - default: - result = -EINVAL; - break; - } - if ((result >= 0) && (dir & _IOC_READ)) { - if (copy_to_user(arg, &krnl_arg, arg_size) != 0) { - TRACE_EXIT -EFAULT; - } - } - TRACE_EXIT result; -} diff --git a/drivers/char/ftape/zftape/zftape-ctl.h b/drivers/char/ftape/zftape/zftape-ctl.h deleted file mode 100644 index 8e6f2d7ac74e..000000000000 --- a/drivers/char/ftape/zftape/zftape-ctl.h +++ /dev/null @@ -1,58 +0,0 @@ -#ifndef _ZFTAPE_CTL_H -#define _ZFTAPE_CTL_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-ctl.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:02 $ - * - * This file contains the non-standard IOCTL related definitions - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include - -#include "../zftape/zftape-rw.h" - -#ifdef CONFIG_ZFTAPE_MODULE -#define ftape_status (*zft_status) -#endif - -extern int zft_offline; -extern int zft_mt_compression; -extern int zft_write_protected; -extern int zft_header_read; -extern unsigned int zft_unit; -extern int zft_resid; - -extern void zft_reset_position(zft_position *pos); -extern int zft_check_write_access(zft_position *pos); -extern int zft_def_idle_state(void); - -/* hooks for the VFS interface - */ -extern int _zft_open(unsigned int dev_minor, unsigned int access_mode); -extern int _zft_close(void); -extern int _zft_ioctl(unsigned int command, void __user *arg); -#endif - - - diff --git a/drivers/char/ftape/zftape/zftape-eof.c b/drivers/char/ftape/zftape/zftape-eof.c deleted file mode 100644 index dcadcaee9ac1..000000000000 --- a/drivers/char/ftape/zftape/zftape-eof.c +++ /dev/null @@ -1,199 +0,0 @@ -/* - * I use these routines just to decide when I have to fake a - * volume-table to preserve compatibility to original ftape. - */ -/* - * Copyright (C) 1994-1995 Bas Laarhoven. - * - * Modified for zftape 1996, 1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-eof.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:02 $ - * - * This file contains the eof mark handling code - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-eof.h" - -/* Global vars. - */ - -/* a copy of the failed sector log from the header segment. - */ -eof_mark_union *zft_eof_map; - -/* number of eof marks (entries in bad sector log) on tape. - */ -int zft_nr_eof_marks = -1; - - -/* Local vars. - */ - -static char linux_tape_label[] = "Linux raw format V"; -enum { - min_fmt_version = 1, max_fmt_version = 2 -}; -static unsigned ftape_fmt_version = 0; - - -/* Ftape (mis)uses the bad sector log to record end-of-file marks. - * Initially (when the tape is erased) all entries in the bad sector - * log are added to the tape's bad sector map. The bad sector log then - * is cleared. - * - * The bad sector log normally contains entries of the form: - * even 16-bit word: segment number of bad sector - * odd 16-bit word: encoded date - * There can be a total of 448 entries (1792 bytes). - * - * My guess is that no program is using this bad sector log (the * - * format seems useless as there is no indication of the bad sector - * itself, only the segment) However, if any program does use the bad - * sector log, the format used by ftape will let the program think - * there are some bad sectors and no harm is done. - * - * The eof mark entries that ftape stores in the bad sector log: even - * 16-bit word: segment number of eof mark odd 16-bit word: sector - * number of eof mark [1..32] - * - * The zft_eof_map as maintained is a sorted list of eof mark entries. - * - * - * The tape name field in the header segments is used to store a linux - * tape identification string and a version number. This way the tape - * can be recognized as a Linux raw format tape when using tools under - * other OS's. - * - * 'Wide' QIC tapes (format code 4) don't have a failed sector list - * anymore. That space is used for the (longer) bad sector map that - * now is a variable length list too. We now store our end-of-file - * marker list after the bad-sector-map on tape. The list is delimited - * by a (__u32) 0 entry. - */ - -int zft_ftape_validate_label(char *label) -{ - static char tmp_label[45]; - int result = 0; - TRACE_FUN(ft_t_any); - - memcpy(tmp_label, label, FT_LABEL_SZ); - tmp_label[FT_LABEL_SZ] = '\0'; - TRACE(ft_t_noise, "tape label = `%s'", tmp_label); - ftape_fmt_version = 0; - if (memcmp(label, linux_tape_label, strlen(linux_tape_label)) == 0) { - int pos = strlen(linux_tape_label); - while (label[pos] >= '0' && label[pos] <= '9') { - ftape_fmt_version *= 10; - ftape_fmt_version = label[ pos++] - '0'; - } - result = (ftape_fmt_version >= min_fmt_version && - ftape_fmt_version <= max_fmt_version); - } - TRACE(ft_t_noise, "format version = %d", ftape_fmt_version); - TRACE_EXIT result; -} - -static __u8 * find_end_of_eof_list(__u8 * ptr, __u8 * limit) -{ - while (ptr + 3 < limit) { - - if (get_unaligned((__u32*)ptr)) { - ptr += sizeof(__u32); - } else { - return ptr; - } - } - return NULL; -} - -void zft_ftape_extract_file_marks(__u8* address) -{ - int i; - TRACE_FUN(ft_t_any); - - zft_eof_map = NULL; - if (ft_format_code == fmt_var || ft_format_code == fmt_big) { - __u8* end; - __u8* start = ftape_find_end_of_bsm_list(address); - - zft_nr_eof_marks = 0; - if (start) { - start += 3; /* skip end of list mark */ - end = find_end_of_eof_list(start, - address + FT_SEGMENT_SIZE); - if (end && end - start <= FT_FSL_SIZE) { - zft_nr_eof_marks = ((end - start) / - sizeof(eof_mark_union)); - zft_eof_map = (eof_mark_union *)start; - } else { - TRACE(ft_t_err, - "EOF Mark List is too long or damaged!"); - } - } else { - TRACE(ft_t_err, - "Bad Sector List is too long or damaged !"); - } - } else { - zft_eof_map = (eof_mark_union *)&address[FT_FSL]; - zft_nr_eof_marks = GET2(address, FT_FSL_CNT); - } - TRACE(ft_t_noise, "number of file marks: %d", zft_nr_eof_marks); - if (ftape_fmt_version == 1) { - TRACE(ft_t_info, "swapping version 1 fields"); - /* version 1 format uses swapped sector and segment - * fields, correct that ! - */ - for (i = 0; i < zft_nr_eof_marks; ++i) { - __u16 tmp = GET2(&zft_eof_map[i].mark.segment,0); - PUT2(&zft_eof_map[i].mark.segment, 0, - GET2(&zft_eof_map[i].mark.date,0)); - PUT2(&zft_eof_map[i].mark.date, 0, tmp); - } - } - for (i = 0; i < zft_nr_eof_marks; ++i) { - TRACE(ft_t_noise, "eof mark: %5d/%2d", - GET2(&zft_eof_map[i].mark.segment, 0), - GET2(&zft_eof_map[i].mark.date,0)); - } - TRACE_EXIT; -} - -void zft_clear_ftape_file_marks(void) -{ - TRACE_FUN(ft_t_flow); - /* Clear failed sector log: remove all tape marks. We - * don't use old ftape-style EOF-marks. - */ - TRACE(ft_t_info, "Clearing old ftape's eof map"); - memset(zft_eof_map, 0, zft_nr_eof_marks * sizeof(__u32)); - zft_nr_eof_marks = 0; - PUT2(zft_hseg_buf, FT_FSL_CNT, 0); /* nr of eof-marks */ - zft_header_changed = 1; - zft_update_label(zft_hseg_buf); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-eof.h b/drivers/char/ftape/zftape/zftape-eof.h deleted file mode 100644 index 26568c26c518..000000000000 --- a/drivers/char/ftape/zftape/zftape-eof.h +++ /dev/null @@ -1,52 +0,0 @@ -#ifndef _ZFTAPE_EOF_H -#define _ZFTAPE_EOF_H - -/* - * Copyright (C) 1994-1995 Bas Laarhoven. - * adaptaed for zftape 1996, 1997 by Claus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-eof.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:03 $ - * - * Definitions and declarations for the end of file markers - * for the QIC-40/80 floppy-tape driver for Linux. - */ - -#include -#include "../zftape/zftape-buffers.h" -/* failed sector log size (only used if format code != 4). - */ - -typedef union { - ft_fsl_entry mark; - __u32 entry; -} eof_mark_union; - -/* ftape-eof.c defined global vars. - */ -extern int zft_nr_eof_marks; -extern eof_mark_union *zft_eof_map; - -/* ftape-eof.c defined global functions. - */ -extern void zft_ftape_extract_file_marks(__u8* address); -extern int zft_ftape_validate_label(char* label); -extern void zft_clear_ftape_file_marks(void); - -#endif diff --git a/drivers/char/ftape/zftape/zftape-init.c b/drivers/char/ftape/zftape/zftape-init.c deleted file mode 100644 index 164a1aa77a2f..000000000000 --- a/drivers/char/ftape/zftape/zftape-init.c +++ /dev/null @@ -1,377 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * This file contains the code that registers the zftape frontend - * to the ftape floppy tape driver for Linux - */ - -#include -#include -#include -#include -#include -#include -#include -#ifdef CONFIG_KMOD -#include -#endif -#include -#include - -#include -#include -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-buffers.h" - -MODULE_AUTHOR("(c) 1996, 1997 Claus-Justus Heine " - "(claus@momo.math.rwth-aachen.de)"); -MODULE_DESCRIPTION(ZFTAPE_VERSION " - " - "VFS interface for the Linux floppy tape driver. " - "Support for QIC-113 compatible volume table " - "and builtin compression (lzrw3 algorithm)"); -MODULE_SUPPORTED_DEVICE("char-major-27"); -MODULE_LICENSE("GPL"); - -/* Global vars. - */ -struct zft_cmpr_ops *zft_cmpr_ops = NULL; -const ftape_info *zft_status; - -/* Local vars. - */ -static unsigned long busy_flag; - -static sigset_t orig_sigmask; - -/* the interface to the kernel vfs layer - */ - -/* Note about llseek(): - * - * st.c and tpqic.c update fp->f_pos but don't implment llseek() and - * initialize the llseek component of the file_ops struct with NULL. - * This means that the user will get the default seek, but the tape - * device will not respect the new position, but happily read from the - * old position. Think a zftape specific llseek() function would be - * better, returning -ESPIPE. TODO. - */ - -static int zft_open (struct inode *ino, struct file *filep); -static int zft_close(struct inode *ino, struct file *filep); -static int zft_ioctl(struct inode *ino, struct file *filep, - unsigned int command, unsigned long arg); -static int zft_mmap(struct file *filep, struct vm_area_struct *vma); -static ssize_t zft_read (struct file *fp, char __user *buff, - size_t req_len, loff_t *ppos); -static ssize_t zft_write(struct file *fp, const char __user *buff, - size_t req_len, loff_t *ppos); - -static const struct file_operations zft_cdev = -{ - .owner = THIS_MODULE, - .read = zft_read, - .write = zft_write, - .ioctl = zft_ioctl, - .mmap = zft_mmap, - .open = zft_open, - .release = zft_close, -}; - -static struct class *zft_class; - -/* Open floppy tape device - */ -static int zft_open(struct inode *ino, struct file *filep) -{ - int result; - TRACE_FUN(ft_t_flow); - - nonseekable_open(ino, filep); - TRACE(ft_t_flow, "called for minor %d", iminor(ino)); - if ( test_and_set_bit(0,&busy_flag) ) { - TRACE_ABORT(-EBUSY, ft_t_warn, "failed: already busy"); - } - if ((iminor(ino) & ~(ZFT_MINOR_OP_MASK | FTAPE_NO_REWIND)) - > - FTAPE_SEL_D) { - clear_bit(0,&busy_flag); - TRACE_ABORT(-ENXIO, ft_t_err, "failed: invalid unit nr"); - } - orig_sigmask = current->blocked; - sigfillset(¤t->blocked); - result = _zft_open(iminor(ino), filep->f_flags & O_ACCMODE); - if (result < 0) { - current->blocked = orig_sigmask; /* restore mask */ - clear_bit(0,&busy_flag); - TRACE_ABORT(result, ft_t_err, "_ftape_open failed"); - } else { - /* Mask signals that will disturb proper operation of the - * program that is calling. - */ - current->blocked = orig_sigmask; - sigaddsetmask (¤t->blocked, _DO_BLOCK); - TRACE_EXIT 0; - } -} - -/* Close floppy tape device - */ -static int zft_close(struct inode *ino, struct file *filep) -{ - int result; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || iminor(ino) != zft_unit) { - TRACE(ft_t_err, "failed: not busy or wrong unit"); - TRACE_EXIT 0; - } - sigfillset(¤t->blocked); - result = _zft_close(); - if (result < 0) { - TRACE(ft_t_err, "_zft_close failed"); - } - current->blocked = orig_sigmask; /* restore before open state */ - clear_bit(0,&busy_flag); - TRACE_EXIT 0; -} - -/* Ioctl for floppy tape device - */ -static int zft_ioctl(struct inode *ino, struct file *filep, - unsigned int command, unsigned long arg) -{ - int result = -EIO; - sigset_t old_sigmask; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - /* This will work as long as sizeof(void *) == sizeof(long) */ - result = _zft_ioctl(command, (void __user *) arg); - current->blocked = old_sigmask; /* restore mask */ - TRACE_EXIT result; -} - -/* Ioctl for floppy tape device - */ -static int zft_mmap(struct file *filep, struct vm_area_struct *vma) -{ - int result = -EIO; - sigset_t old_sigmask; - TRACE_FUN(ft_t_flow); - - if ( !test_bit(0,&busy_flag) || - iminor(filep->f_dentry->d_inode) != zft_unit || - ft_failure) - { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - if ((result = ftape_mmap(vma)) >= 0) { -#ifndef MSYNC_BUG_WAS_FIXED - static struct vm_operations_struct dummy = { NULL, }; - vma->vm_ops = &dummy; -#endif - } - current->blocked = old_sigmask; /* restore mask */ - TRACE_EXIT result; -} - -/* Read from floppy tape device - */ -static ssize_t zft_read(struct file *fp, char __user *buff, - size_t req_len, loff_t *ppos) -{ - int result = -EIO; - sigset_t old_sigmask; - struct inode *ino = fp->f_dentry->d_inode; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, "called with count: %ld", (unsigned long)req_len); - if (!test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - result = _zft_read(buff, req_len); - current->blocked = old_sigmask; /* restore mask */ - TRACE(ft_t_data_flow, "return with count: %d", result); - TRACE_EXIT result; -} - -/* Write to tape device - */ -static ssize_t zft_write(struct file *fp, const char __user *buff, - size_t req_len, loff_t *ppos) -{ - int result = -EIO; - sigset_t old_sigmask; - struct inode *ino = fp->f_dentry->d_inode; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_flow, "called with count: %ld", (unsigned long)req_len); - if (!test_bit(0,&busy_flag) || iminor(ino) != zft_unit || ft_failure) { - TRACE_ABORT(-EIO, ft_t_err, - "failed: not busy, failure or wrong unit"); - } - old_sigmask = current->blocked; /* save mask */ - sigfillset(¤t->blocked); - result = _zft_write(buff, req_len); - current->blocked = old_sigmask; /* restore mask */ - TRACE(ft_t_data_flow, "return with count: %d", result); - TRACE_EXIT result; -} - -/* END OF VFS INTERFACE - * - *****************************************************************************/ - -/* driver/module initialization - */ - -/* the compression module has to call this function to hook into the zftape - * code - */ -int zft_cmpr_register(struct zft_cmpr_ops *new_ops) -{ - TRACE_FUN(ft_t_flow); - - if (zft_cmpr_ops != NULL) { - TRACE_EXIT -EBUSY; - } else { - zft_cmpr_ops = new_ops; - TRACE_EXIT 0; - } -} - -/* lock the zft-compressor() module. - */ -int zft_cmpr_lock(int try_to_load) -{ - if (zft_cmpr_ops == NULL) { -#ifdef CONFIG_KMOD - if (try_to_load) { - request_module("zft-compressor"); - if (zft_cmpr_ops == NULL) { - return -ENOSYS; - } - } else { - return -ENOSYS; - } -#else - return -ENOSYS; -#endif - } - (*zft_cmpr_ops->lock)(); - return 0; -} - -#ifdef CONFIG_ZFT_COMPRESSOR -extern int zft_compressor_init(void); -#endif - -/* Called by modules package when installing the driver or by kernel - * during the initialization phase - */ -int __init zft_init(void) -{ - int i; - TRACE_FUN(ft_t_flow); - -#ifdef MODULE - printk(KERN_INFO ZFTAPE_VERSION "\n"); - if (TRACE_LEVEL >= ft_t_info) { - printk( -KERN_INFO -"(c) 1996, 1997 Claus-Justus Heine (claus@momo.math.rwth-aachen.de)\n" -KERN_INFO -"vfs interface for ftape floppy tape driver.\n" -KERN_INFO -"Support for QIC-113 compatible volume table, dynamic memory allocation\n" -KERN_INFO -"and builtin compression (lzrw3 algorithm).\n"); - } -#else /* !MODULE */ - /* print a short no-nonsense boot message */ - printk(KERN_INFO ZFTAPE_VERSION "\n"); -#endif /* MODULE */ - TRACE(ft_t_info, "zft_init @ 0x%p", zft_init); - TRACE(ft_t_info, - "installing zftape VFS interface for ftape driver ..."); - TRACE_CATCH(register_chrdev(QIC117_TAPE_MAJOR, "zft", &zft_cdev),); - - zft_class = class_create(THIS_MODULE, "zft"); - for (i = 0; i < 4; i++) { - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i), NULL, "qft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 4), NULL, "nqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 16), NULL, "zqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 20), NULL, "nzqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 32), NULL, "rawqft%i", i); - class_device_create(zft_class, NULL, MKDEV(QIC117_TAPE_MAJOR, i + 36), NULL, "nrawrawqft%i", i); - } - -#ifdef CONFIG_ZFT_COMPRESSOR - (void)zft_compressor_init(); -#endif - zft_status = ftape_get_status(); /* fetch global data of ftape - * hardware driver - */ - TRACE_EXIT 0; -} - - -/* Called by modules package when removing the driver - */ -static void zft_exit(void) -{ - int i; - TRACE_FUN(ft_t_flow); - - if (unregister_chrdev(QIC117_TAPE_MAJOR, "zft") != 0) { - TRACE(ft_t_warn, "failed"); - } else { - TRACE(ft_t_info, "successful"); - } - for (i = 0; i < 4; i++) { - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 4)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 16)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 20)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 32)); - class_device_destroy(zft_class, MKDEV(QIC117_TAPE_MAJOR, i + 36)); - } - class_destroy(zft_class); - zft_uninit_mem(); /* release remaining memory, if any */ - printk(KERN_INFO "zftape successfully unloaded.\n"); - TRACE_EXIT; -} - -module_init(zft_init); -module_exit(zft_exit); diff --git a/drivers/char/ftape/zftape/zftape-init.h b/drivers/char/ftape/zftape/zftape-init.h deleted file mode 100644 index 937e5d48c20e..000000000000 --- a/drivers/char/ftape/zftape/zftape-init.h +++ /dev/null @@ -1,77 +0,0 @@ -#ifndef _ZFTAPE_INIT_H -#define _ZFTAPE_INIT_H - -/* - * Copyright (C) 1996, 1997 Claus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-init.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:05 $ - * - * This file contains definitions and macro for the vfs - * interface defined by zftape - * - */ - -#include - -#include "../lowlevel/ftape-tracing.h" -#include "../lowlevel/ftape-ctl.h" -#include "../lowlevel/ftape-read.h" -#include "../lowlevel/ftape-write.h" -#include "../lowlevel/ftape-bsm.h" -#include "../lowlevel/ftape-io.h" -#include "../lowlevel/ftape-buffer.h" -#include "../lowlevel/ftape-format.h" - -#include "../zftape/zftape-rw.h" - -#ifdef MODULE -#define ftape_status (*zft_status) -#endif - -extern const ftape_info *zft_status; /* needed for zftape-vtbl.h */ - -#include "../zftape/zftape-vtbl.h" - -struct zft_cmpr_ops { - int (*write)(int *write_cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos, const zft_volinfo *volume); - int (*read)(int *read_cnt, - __u8 __user *dst_buf, const int req_len, - const __u8 *src_buf, const int seg_sz, - const zft_position *pos, const zft_volinfo *volume); - int (*seek)(unsigned int new_block_pos, - zft_position *pos, const zft_volinfo *volume, - __u8 *buffer); - void (*lock) (void); - void (*reset) (void); - void (*cleanup)(void); -}; - -extern struct zft_cmpr_ops *zft_cmpr_ops; -/* zftape-init.c defined global functions. - */ -extern int zft_cmpr_register(struct zft_cmpr_ops *new_ops); -extern int zft_cmpr_lock(int try_to_load); - -#endif - - diff --git a/drivers/char/ftape/zftape/zftape-read.c b/drivers/char/ftape/zftape/zftape-read.c deleted file mode 100644 index 214bf03dce68..000000000000 --- a/drivers/char/ftape/zftape/zftape-read.c +++ /dev/null @@ -1,377 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-read.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:06 $ - * - * This file contains the high level reading code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ -int zft_just_before_eof; - -/* Local vars. - */ -static int buf_len_rd; - -void zft_zap_read_buffers(void) -{ - buf_len_rd = 0; -} - -int zft_read_header_segments(void) -{ - TRACE_FUN(ft_t_flow); - - zft_header_read = 0; - TRACE_CATCH(zft_vmalloc_once(&zft_hseg_buf, FT_SEGMENT_SIZE),); - TRACE_CATCH(ftape_read_header_segment(zft_hseg_buf),); - TRACE(ft_t_info, "Segments written since first format: %d", - (int)GET4(zft_hseg_buf, FT_SEG_CNT)); - zft_qic113 = (ft_format_code != fmt_normal && - ft_format_code != fmt_1100ft && - ft_format_code != fmt_425ft); - TRACE(ft_t_info, "ft_first_data_segment: %d, ft_last_data_segment: %d", - ft_first_data_segment, ft_last_data_segment); - zft_capacity = zft_get_capacity(); - zft_old_ftape = zft_ftape_validate_label(&zft_hseg_buf[FT_LABEL]); - if (zft_old_ftape) { - TRACE(ft_t_info, -"Found old ftaped tape, emulating eof marks, entering read-only mode"); - zft_ftape_extract_file_marks(zft_hseg_buf); - TRACE_CATCH(zft_fake_volume_headers(zft_eof_map, - zft_nr_eof_marks),); - } else { - /* the specs say that the volume table must be - * initialized with zeroes during formatting, so it - * MUST be readable, i.e. contain vaid ECC - * information. - */ - TRACE_CATCH(ftape_read_segment(ft_first_data_segment, - zft_deblock_buf, - FT_RD_SINGLE),); - TRACE_CATCH(zft_extract_volume_headers(zft_deblock_buf),); - } - zft_header_read = 1; - zft_set_flags(zft_unit); - zft_reset_position(&zft_pos); - TRACE_EXIT 0; -} - -int zft_fetch_segment_fraction(const unsigned int segment, void *buffer, - const ft_read_mode_t read_mode, - const unsigned int start, - const unsigned int size) -{ - int seg_sz; - TRACE_FUN(ft_t_flow); - - if (segment == zft_deblock_segment) { - TRACE(ft_t_data_flow, - "re-using segment %d already in deblock buffer", - segment); - seg_sz = zft_get_seg_sz(segment); - if (start > seg_sz) { - TRACE_ABORT(-EINVAL, ft_t_bug, - "trying to read beyond end of segment:\n" - KERN_INFO "seg_sz : %d\n" - KERN_INFO "start : %d\n" - KERN_INFO "segment: %d", - seg_sz, start, segment); - } - if ((start + size) > seg_sz) { - TRACE_EXIT seg_sz - start; - } - TRACE_EXIT size; - } - seg_sz = ftape_read_segment_fraction(segment, buffer, read_mode, - start, size); - TRACE(ft_t_data_flow, "segment %d, result %d", segment, seg_sz); - if ((int)seg_sz >= 0 && start == 0 && size == FT_SEGMENT_SIZE) { - /* this implicitly assumes that we are always called with - * buffer == zft_deblock_buf - */ - zft_deblock_segment = segment; - } else { - zft_deblock_segment = -1; - } - TRACE_EXIT seg_sz; -} - -/* - * out: - * - * int *read_cnt: the number of bytes we removed from the - * zft_deblock_buf (result) - * - * int *to_do : the remaining size of the read-request. Is changed. - * - * in: - * - * char *buff : buff is the address of the upper part of the user - * buffer, that hasn't been filled with data yet. - * int buf_pos_read: copy of buf_pos_rd - * int buf_len_read: copy of buf_len_rd - * char *zft_deblock_buf: ftape_zft_deblock_buf - * - * returns the amount of data actually copied to the user-buffer - * - * to_do MUST NOT SHRINK except to indicate an EOT. In this case to_do - * has to be set to 0. We cannot return -ENOSPC, because we return the - * amount of data actually * copied to the user-buffer - */ -static int zft_simple_read (int *read_cnt, - __u8 __user *dst_buf, - const int to_do, - const __u8 *src_buf, - const int seg_sz, - const zft_position *pos, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if (seg_sz - pos->seg_byte_pos < to_do) { - *read_cnt = seg_sz - pos->seg_byte_pos; - } else { - *read_cnt = to_do; - } - if (copy_to_user(dst_buf, - src_buf + pos->seg_byte_pos, *read_cnt) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE(ft_t_noise, "nr bytes just read: %d", *read_cnt); - TRACE_EXIT *read_cnt; -} - -/* req_len: gets clipped due to EOT of EOF. - * req_clipped: is a flag indicating whether req_len was clipped or not - * volume: contains information on current volume (blk_sz etc.) - */ -static int check_read_access(int *req_len, - const zft_volinfo **volume, - int *req_clipped, - const zft_position *pos) -{ - static __s64 remaining; - static int eod; - TRACE_FUN(ft_t_flow); - - if (zft_io_state != zft_reading) { - if (zft_offline) { /* offline includes no_tape */ - TRACE_ABORT(-ENXIO, ft_t_warn, - "tape is offline or no cartridge"); - } - if (!ft_formatted) { - TRACE_ABORT(-EACCES, - ft_t_warn, "tape is not formatted"); - } - /* now enter defined state, read header segment if not - * already done and flush write buffers - */ - TRACE_CATCH(zft_def_idle_state(),); - zft_io_state = zft_reading; - if (zft_tape_at_eod(pos)) { - eod = 1; - TRACE_EXIT 1; - } - eod = 0; - *volume = zft_find_volume(pos->seg_pos); - /* get the space left until EOF */ - remaining = zft_check_for_eof(*volume, pos); - buf_len_rd = 0; - TRACE(ft_t_noise, "remaining: " LL_X ", vol_no: %d", - LL(remaining), (*volume)->count); - } else if (zft_tape_at_eod(pos)) { - if (++eod > 2) { - TRACE_EXIT -EIO; /* st.c also returns -EIO */ - } else { - TRACE_EXIT 1; - } - } - if ((*req_len % (*volume)->blk_sz) != 0) { - /* this message is informational only. The user gets the - * proper return value - */ - TRACE_ABORT(-EINVAL, ft_t_info, - "req_len %d not a multiple of block size %d", - *req_len, (*volume)->blk_sz); - } - /* As GNU tar doesn't accept partial read counts when the - * multiple volume flag is set, we make sure to return the - * requested amount of data. Except, of course, at the end of - * the tape or file mark. - */ - remaining -= *req_len; - if (remaining <= 0) { - TRACE(ft_t_noise, - "clipped request from %d to %d.", - *req_len, (int)(*req_len + remaining)); - *req_len += remaining; - *req_clipped = 1; - } else { - *req_clipped = 0; - } - TRACE_EXIT 0; -} - -/* this_segs_size: the current segment's size. - * buff: the USER-SPACE buffer provided by the calling function. - * req_len: how much data should be read at most. - * volume: contains information on current volume (blk_sz etc.) - */ -static int empty_deblock_buf(__u8 __user *usr_buf, const int req_len, - const __u8 *src_buf, const int seg_sz, - zft_position *pos, - const zft_volinfo *volume) -{ - int cnt; - int result = 0; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, "this_segs_size: %d", seg_sz); - if (zft_use_compression && volume->use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - TRACE_CATCH(result= (*zft_cmpr_ops->read)(&cnt, - usr_buf, req_len, - src_buf, seg_sz, - pos, volume),); - } else { - TRACE_CATCH(result= zft_simple_read (&cnt, - usr_buf, req_len, - src_buf, seg_sz, - pos, volume),); - } - pos->volume_pos += result; - pos->tape_pos += cnt; - pos->seg_byte_pos += cnt; - buf_len_rd -= cnt; /* remaining bytes in buffer */ - TRACE(ft_t_data_flow, "buf_len_rd: %d, cnt: %d", buf_len_rd, cnt); - if(pos->seg_byte_pos >= seg_sz) { - pos->seg_pos++; - pos->seg_byte_pos = 0; - } - TRACE(ft_t_data_flow, "bytes moved out of deblock-buffer: %d", cnt); - TRACE_EXIT result; -} - - -/* note: we store the segment id of the segment that is inside the - * deblock buffer. This spares a lot of ftape_read_segment()s when we - * use small block-sizes. The block-size may be 1kb (SECTOR_SIZE). In - * this case a MTFSR 28 maybe still inside the same segment. - */ -int _zft_read(char __user *buff, int req_len) -{ - int req_clipped; - int result = 0; - int bytes_read = 0; - static unsigned int seg_sz = 0; - static const zft_volinfo *volume = NULL; - TRACE_FUN(ft_t_flow); - - zft_resid = req_len; - result = check_read_access(&req_len, &volume, - &req_clipped, &zft_pos); - switch(result) { - case 0: - break; /* nothing special */ - case 1: - TRACE(ft_t_noise, "EOD reached"); - TRACE_EXIT 0; /* EOD */ - default: - TRACE_ABORT(result, ft_t_noise, - "check_read_access() failed with result %d", - result); - TRACE_EXIT result; - } - while (req_len > 0) { - /* Allow escape from this loop on signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - /* buf_len_rd == 0 means that we need to read a new - * segment. - */ - if (buf_len_rd == 0) { - while((result = zft_fetch_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_RD_AHEAD)) == 0) { - zft_pos.seg_pos ++; - zft_pos.seg_byte_pos = 0; - } - if (result < 0) { - zft_resid -= bytes_read; - TRACE_ABORT(result, ft_t_noise, - "zft_fetch_segment(): %d", - result); - } - seg_sz = result; - buf_len_rd = seg_sz - zft_pos.seg_byte_pos; - } - TRACE_CATCH(result = empty_deblock_buf(buff, - req_len, - zft_deblock_buf, - seg_sz, - &zft_pos, - volume), - zft_resid -= bytes_read); - TRACE(ft_t_data_flow, "bytes just read: %d", result); - bytes_read += result; /* what we got so far */ - buff += result; /* index in user-buffer */ - req_len -= result; /* what's left from req_len */ - } /* while (req_len > 0) */ - if (req_clipped) { - TRACE(ft_t_data_flow, - "maybe partial count because of eof mark"); - if (zft_just_before_eof && bytes_read == 0) { - /* req_len was > 0, but user didn't get - * anything the user has read in the eof-mark - */ - zft_move_past_eof(&zft_pos); - ftape_abort_operation(); - } else { - /* don't skip to the next file before the user - * tried to read a second time past EOF Just - * mark that we are at EOF and maybe decrement - * zft_seg_pos to stay in the same volume; - */ - zft_just_before_eof = 1; - zft_position_before_eof(&zft_pos, volume); - TRACE(ft_t_noise, "just before eof"); - } - } - zft_resid -= result; /* for MTSTATUS */ - TRACE_EXIT bytes_read; -} diff --git a/drivers/char/ftape/zftape/zftape-read.h b/drivers/char/ftape/zftape/zftape-read.h deleted file mode 100644 index 42941de0c23a..000000000000 --- a/drivers/char/ftape/zftape/zftape-read.h +++ /dev/null @@ -1,53 +0,0 @@ -#ifndef _ZFTAPE_READ_H -#define _ZFTAPE_READ_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-read.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:07 $ - * - * This file contains the definitions for the read functions - * for the zftape driver for Linux. - * - */ - -#include "../lowlevel/ftape-read.h" - -/* ftape-read.c defined global vars. - */ -extern int zft_just_before_eof; - -/* ftape-read.c defined global functions. - */ -extern void zft_zap_read_buffers(void); -extern int zft_read_header_segments(void); -extern int zft_fetch_segment_fraction(const unsigned int segment, - void *buffer, - const ft_read_mode_t read_mode, - const unsigned int start, - const unsigned int size); -#define zft_fetch_segment(segment, address, read_mode) \ - zft_fetch_segment_fraction(segment, address, read_mode, \ - 0, FT_SEGMENT_SIZE) -/* hook for the VFS interface - */ -extern int _zft_read(char __user *buff, int req_len); - -#endif /* _ZFTAPE_READ_H */ diff --git a/drivers/char/ftape/zftape/zftape-rw.c b/drivers/char/ftape/zftape/zftape-rw.c deleted file mode 100644 index dab634686885..000000000000 --- a/drivers/char/ftape/zftape/zftape-rw.c +++ /dev/null @@ -1,375 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-rw.c,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:08 $ - * - * This file contains some common code for the r/w code for - * zftape. - */ - -#include -#include - -#include -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ - -__u8 *zft_deblock_buf; -__u8 *zft_hseg_buf; -int zft_deblock_segment = -1; -zft_status_enum zft_io_state = zft_idle; -int zft_header_changed; -int zft_qic113; /* conform to old specs. and old zftape */ -int zft_use_compression; -zft_position zft_pos = { - -1, /* seg_pos */ - 0, /* seg_byte_pos */ - 0, /* tape_pos */ - 0 /* volume_pos */ -}; -unsigned int zft_blk_sz = CONFIG_ZFT_DFLT_BLK_SZ; -__s64 zft_capacity; - -unsigned int zft_written_segments; -int zft_label_changed; - -/* Local vars. - */ - -unsigned int zft_get_seg_sz(unsigned int segment) -{ - int size; - TRACE_FUN(ft_t_any); - - size = FT_SEGMENT_SIZE - - count_ones(ftape_get_bad_sector_entry(segment))*FT_SECTOR_SIZE; - if (size > 0) { - TRACE_EXIT (unsigned)size; - } else { - TRACE_EXIT 0; - } -} - -/* ftape_set_flags(). Claus-Justus Heine, 1994/1995 - */ -void zft_set_flags(unsigned minor_unit) -{ - TRACE_FUN(ft_t_flow); - - zft_use_compression = zft_qic_mode = 0; - switch (minor_unit & ZFT_MINOR_OP_MASK) { - case (ZFT_Q80_MODE | ZFT_ZIP_MODE): - case ZFT_ZIP_MODE: - zft_use_compression = 1; - case 0: - case ZFT_Q80_MODE: - zft_qic_mode = 1; - if (zft_mt_compression) { /* override the default */ - zft_use_compression = 1; - } - break; - case ZFT_RAW_MODE: - TRACE(ft_t_noise, "switching to raw mode"); - break; - default: - TRACE(ft_t_warn, "Warning:\n" - KERN_INFO "Wrong combination of minor device bits.\n" - KERN_INFO "Switching to raw read-only mode."); - zft_write_protected = 1; - break; - } - TRACE_EXIT; -} - -/* computes the segment and byte offset inside the segment - * corresponding to tape_pos. - * - * tape_pos gives the offset in bytes from the beginning of the - * ft_first_data_segment *seg_byte_pos is the offset in the current - * segment in bytes - * - * Of, if this routine was called often one should cache the last data - * pos it was called with, but actually this is only needed in - * ftape_seek_block(), that is, almost never. - */ -int zft_calc_seg_byte_coord(int *seg_byte_pos, __s64 tape_pos) -{ - int segment; - int seg_sz; - TRACE_FUN(ft_t_flow); - - if (tape_pos == 0) { - *seg_byte_pos = 0; - segment = ft_first_data_segment; - } else { - seg_sz = 0; - - for (segment = ft_first_data_segment; - ((tape_pos > 0) && (segment <= ft_last_data_segment)); - segment++) { - seg_sz = zft_get_seg_sz(segment); - tape_pos -= seg_sz; - } - if(tape_pos >= 0) { - /* the case tape_pos > != 0 means that the - * argument tape_pos lies beyond the EOT. - */ - *seg_byte_pos= 0; - } else { /* tape_pos < 0 */ - segment--; - *seg_byte_pos= tape_pos + seg_sz; - } - } - TRACE_EXIT(segment); -} - -/* ftape_calc_tape_pos(). - * - * computes the offset in bytes from the beginning of the - * ft_first_data_segment inverse to ftape_calc_seg_byte_coord - * - * We should do some caching. But how: - * - * Each time the header segments are read in, this routine is called - * with ft_tracks_per_tape*segments_per_track argumnet. So this should be - * the time to reset the cache. - * - * Also, it might be in the future that the bad sector map gets - * changed. -> reset the cache - */ -static int seg_pos; -static __s64 tape_pos; - -__s64 zft_get_capacity(void) -{ - seg_pos = ft_first_data_segment; - tape_pos = 0; - - while (seg_pos <= ft_last_data_segment) { - tape_pos += zft_get_seg_sz(seg_pos ++); - } - return tape_pos; -} - -__s64 zft_calc_tape_pos(int segment) -{ - int d1, d2, d3; - TRACE_FUN(ft_t_any); - - if (segment > ft_last_data_segment) { - TRACE_EXIT zft_capacity; - } - if (segment < ft_first_data_segment) { - TRACE_EXIT 0; - } - d2 = segment - seg_pos; - if (-d2 > 10) { - d1 = segment - ft_first_data_segment; - if (-d2 > d1) { - tape_pos = 0; - seg_pos = ft_first_data_segment; - d2 = d1; - } - } - if (d2 > 10) { - d3 = ft_last_data_segment - segment; - if (d2 > d3) { - tape_pos = zft_capacity; - seg_pos = ft_last_data_segment + 1; - d2 = -d3; - } - } - if (d2 > 0) { - while (seg_pos < segment) { - tape_pos += zft_get_seg_sz(seg_pos++); - } - } else { - while (seg_pos > segment) { - tape_pos -= zft_get_seg_sz(--seg_pos); - } - } - TRACE(ft_t_noise, "new cached pos: %d", seg_pos); - - TRACE_EXIT tape_pos; -} - -/* copy Z-label string to buffer, keeps track of the correct offset in - * `buffer' - */ -void zft_update_label(__u8 *buffer) -{ - TRACE_FUN(ft_t_flow); - - if (strncmp(&buffer[FT_LABEL], ZFTAPE_LABEL, - sizeof(ZFTAPE_LABEL)-1) != 0) { - TRACE(ft_t_info, "updating label from \"%s\" to \"%s\"", - &buffer[FT_LABEL], ZFTAPE_LABEL); - strcpy(&buffer[FT_LABEL], ZFTAPE_LABEL); - memset(&buffer[FT_LABEL] + sizeof(ZFTAPE_LABEL) - 1, ' ', - FT_LABEL_SZ - sizeof(ZFTAPE_LABEL + 1)); - PUT4(buffer, FT_LABEL_DATE, 0); - zft_label_changed = zft_header_changed = 1; /* changed */ - } - TRACE_EXIT; -} - -int zft_verify_write_segments(unsigned int segment, - __u8 *data, size_t size, - __u8 *buffer) -{ - int result; - __u8 *write_buf; - __u8 *src_buf; - int single; - int seg_pos; - int seg_sz; - int remaining; - ft_write_mode_t write_mode; - TRACE_FUN(ft_t_flow); - - seg_pos = segment; - seg_sz = zft_get_seg_sz(seg_pos); - src_buf = data; - single = size <= seg_sz; - remaining = size; - do { - TRACE(ft_t_noise, "\n" - KERN_INFO "remaining: %d\n" - KERN_INFO "seg_sz : %d\n" - KERN_INFO "segment : %d", - remaining, seg_sz, seg_pos); - if (remaining == seg_sz) { - write_buf = src_buf; - write_mode = single ? FT_WR_SINGLE : FT_WR_MULTI; - remaining = 0; - } else if (remaining > seg_sz) { - write_buf = src_buf; - write_mode = FT_WR_ASYNC; /* don't start tape */ - remaining -= seg_sz; - } else { /* remaining < seg_sz */ - write_buf = buffer; - memcpy(write_buf, src_buf, remaining); - memset(&write_buf[remaining],'\0',seg_sz-remaining); - write_mode = single ? FT_WR_SINGLE : FT_WR_MULTI; - remaining = 0; - } - if ((result = ftape_write_segment(seg_pos, - write_buf, - write_mode)) != seg_sz) { - TRACE(ft_t_err, "Error: " - "Couldn't write segment %d", seg_pos); - TRACE_EXIT result < 0 ? result : -EIO; /* bail out */ - } - zft_written_segments ++; - seg_sz = zft_get_seg_sz(++seg_pos); - src_buf += result; - } while (remaining > 0); - if (ftape_get_status()->fti_state == writing) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - TRACE_CATCH(ftape_abort_operation(),); - zft_prevent_flush(); - } - seg_pos = segment; - src_buf = data; - remaining = size; - do { - TRACE_CATCH(result = ftape_read_segment(seg_pos, buffer, - single ? FT_RD_SINGLE - : FT_RD_AHEAD),); - if (memcmp(src_buf, buffer, - remaining > result ? result : remaining) != 0) { - TRACE_ABORT(-EIO, ft_t_err, - "Failed to verify written segment %d", - seg_pos); - } - remaining -= result; - TRACE(ft_t_noise, "verify successful:\n" - KERN_INFO "segment : %d\n" - KERN_INFO "segsize : %d\n" - KERN_INFO "remaining: %d", - seg_pos, result, remaining); - src_buf += seg_sz; - seg_pos++; - } while (remaining > 0); - TRACE_EXIT size; -} - - -/* zft_erase(). implemented compression-handling - * - * calculate the first data-segment when using/not using compression. - * - * update header-segment and compression-map-segment. - */ -int zft_erase(void) -{ - int result = 0; - TRACE_FUN(ft_t_flow); - - if (!zft_header_read) { - TRACE_CATCH(zft_vmalloc_once((void **)&zft_hseg_buf, - FT_SEGMENT_SIZE),); - /* no need to read the vtbl and compression map */ - TRACE_CATCH(ftape_read_header_segment(zft_hseg_buf),); - if ((zft_old_ftape = - zft_ftape_validate_label(&zft_hseg_buf[FT_LABEL]))) { - zft_ftape_extract_file_marks(zft_hseg_buf); - } - TRACE(ft_t_noise, - "ft_first_data_segment: %d, ft_last_data_segment: %d", - ft_first_data_segment, ft_last_data_segment); - zft_qic113 = (ft_format_code != fmt_normal && - ft_format_code != fmt_1100ft && - ft_format_code != fmt_425ft); - } - if (zft_old_ftape) { - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - PUT2(zft_hseg_buf, FT_CMAP_START, 0); - zft_volume_table_changed = 1; - zft_capacity = zft_get_capacity(); - zft_init_vtbl(); - /* the rest must be done in ftape_update_header_segments - */ - zft_header_read = 1; - zft_header_changed = 1; /* force update of timestamp */ - result = zft_update_header_segments(); - - ftape_abort_operation(); - - zft_reset_position(&zft_pos); - zft_set_flags (zft_unit); - TRACE_EXIT result; -} - -unsigned int zft_get_time(void) -{ - unsigned int date = FT_TIME_STAMP(2097, 11, 30, 23, 59, 59); /* fun */ - return date; -} diff --git a/drivers/char/ftape/zftape/zftape-rw.h b/drivers/char/ftape/zftape/zftape-rw.h deleted file mode 100644 index 1ceec22b60bd..000000000000 --- a/drivers/char/ftape/zftape/zftape-rw.h +++ /dev/null @@ -1,101 +0,0 @@ -#ifndef _ZFTAPE_RW_H -#define _ZFTAPE_RW_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-rw.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:09 $ - * - * This file contains the definitions for the read and write - * functions for the QIC-117 floppy-tape driver for Linux. - * - */ - -#include "../zftape/zftape-buffers.h" - -#define SEGMENTS_PER_TAPE (ft_segments_per_track * ft_tracks_per_tape) - -/* QIC-113 Rev. G says that `a maximum of 63488 raw bytes may be - * compressed into a single frame'. - * Maybe we should stick to 32kb to make it more `beautiful' - */ -#define ZFT_MAX_BLK_SZ (62*1024) /* bytes */ -#if !defined(CONFIG_ZFT_DFLT_BLK_SZ) -# define CONFIG_ZFT_DFLT_BLK_SZ (10*1024) /* bytes, default of gnu tar */ -#elif CONFIG_ZFT_DFLT_BLK_SZ == 0 -# undef CONFIG_ZFT_DFLT_BLK_SZ -# define CONFIG_ZFT_DFLT_BLK_SZ 1 -#elif (CONFIG_ZFT_DFLT_BLK_SZ % 1024) != 0 -# error CONFIG_ZFT_DFLT_BLK_SZ must be 1 or a multiple of 1024 -#endif -/* The *optional* compression routines need some overhead per tape - * block for their purposes. Instead of asking the actual compression - * implementation how much it needs, we restrict this overhead to be - * maximal of ZFT_CMPT_OVERHEAD size. We need this for EOT - * conditions. The tape is assumed to be logical at EOT when the - * distance from the physical EOT is less than - * one tape block + ZFT_CMPR_OVERHEAD - */ -#define ZFT_CMPR_OVERHEAD 16 /* bytes */ - -typedef enum -{ - zft_idle = 0, - zft_reading, - zft_writing, -} zft_status_enum; - -typedef struct /* all values measured in bytes */ -{ - int seg_pos; /* segment currently positioned at */ - int seg_byte_pos; /* offset in current segment */ - __s64 tape_pos; /* real offset from BOT */ - __s64 volume_pos; /* pos. in uncompressed data stream in - * current volume - */ -} zft_position; - -extern zft_position zft_pos; -extern __u8 *zft_deblock_buf; -extern __u8 *zft_hseg_buf; -extern int zft_deblock_segment; -extern zft_status_enum zft_io_state; -extern int zft_header_changed; -extern int zft_qic113; /* conform to old specs. and old zftape */ -extern int zft_use_compression; -extern unsigned int zft_blk_sz; -extern __s64 zft_capacity; -extern unsigned int zft_written_segments; -extern int zft_label_changed; - -/* zftape-rw.c exported functions - */ -extern unsigned int zft_get_seg_sz(unsigned int segment); -extern void zft_set_flags(unsigned int minor_unit); -extern int zft_calc_seg_byte_coord(int *seg_byte_pos, __s64 tape_pos); -extern __s64 zft_calc_tape_pos(int segment); -extern __s64 zft_get_capacity(void); -extern void zft_update_label(__u8 *buffer); -extern int zft_erase(void); -extern int zft_verify_write_segments(unsigned int segment, - __u8 *data, size_t size, __u8 *buffer); -extern unsigned int zft_get_time(void); -#endif /* _ZFTAPE_RW_H */ - diff --git a/drivers/char/ftape/zftape/zftape-vtbl.c b/drivers/char/ftape/zftape/zftape-vtbl.c deleted file mode 100644 index ad7f8be6340b..000000000000 --- a/drivers/char/ftape/zftape/zftape-vtbl.c +++ /dev/null @@ -1,757 +0,0 @@ -/* - * Copyright (c) 1995-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-vtbl.c,v $ - * $Revision: 1.7.6.1 $ - * $Date: 1997/11/24 13:48:31 $ - * - * This file defines a volume table as defined in various QIC - * standards. - * - * This is a minimal implementation, just allowing ordinary DOS - * :( prgrams to identify the cartridge as used. - */ - -#include -#include -#include - -#include -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -#define ZFT_CMAP_HACK /* leave this defined to hide the compression map */ - -/* - * global variables - */ -int zft_qic_mode = 1; /* use the vtbl */ -int zft_old_ftape; /* prevents old ftaped tapes to be overwritten */ -int zft_volume_table_changed; /* for write_header_segments() */ - -/* - * private variables (only exported for inline functions) - */ -LIST_HEAD(zft_vtbl); - -/* We could also allocate these dynamically when extracting the volume table - * sizeof(zft_volinfo) is about 32 or something close to that - */ -static zft_volinfo tape_vtbl; -static zft_volinfo eot_vtbl; -static zft_volinfo *cur_vtbl; - -static inline void zft_new_vtbl_entry(void) -{ - struct list_head *tmp = &zft_last_vtbl->node; - zft_volinfo *new = zft_kmalloc(sizeof(zft_volinfo)); - - list_add(&new->node, tmp); - new->count = zft_eom_vtbl->count ++; -} - -void zft_free_vtbl(void) -{ - for (;;) { - struct list_head *tmp = zft_vtbl.prev; - zft_volinfo *vtbl; - - if (tmp == &zft_vtbl) - break; - list_del(tmp); - vtbl = list_entry(tmp, zft_volinfo, node); - zft_kfree(vtbl, sizeof(zft_volinfo)); - } - INIT_LIST_HEAD(&zft_vtbl); - cur_vtbl = NULL; -} - -/* initialize vtbl, called by ftape_new_cartridge() - */ -void zft_init_vtbl(void) -{ - zft_volinfo *new; - - zft_free_vtbl(); - - /* Create the two dummy vtbl entries - */ - new = zft_kmalloc(sizeof(zft_volinfo)); - list_add(&new->node, &zft_vtbl); - new = zft_kmalloc(sizeof(zft_volinfo)); - list_add(&new->node, &zft_vtbl); - zft_head_vtbl->end_seg = ft_first_data_segment; - zft_head_vtbl->blk_sz = zft_blk_sz; - zft_head_vtbl->count = -1; - zft_eom_vtbl->start_seg = ft_first_data_segment + 1; - zft_eom_vtbl->end_seg = ft_last_data_segment + 1; - zft_eom_vtbl->blk_sz = zft_blk_sz; - zft_eom_vtbl->count = 0; - - /* Reset the pointer for zft_find_volume() - */ - cur_vtbl = zft_eom_vtbl; - - /* initialize the dummy vtbl entries for zft_qic_mode == 0 - */ - eot_vtbl.start_seg = ft_last_data_segment + 1; - eot_vtbl.end_seg = ft_last_data_segment + 1; - eot_vtbl.blk_sz = zft_blk_sz; - eot_vtbl.count = -1; - tape_vtbl.start_seg = ft_first_data_segment; - tape_vtbl.end_seg = ft_last_data_segment; - tape_vtbl.blk_sz = zft_blk_sz; - tape_vtbl.size = zft_capacity; - tape_vtbl.count = 0; -} - -/* check for a valid VTBL signature. - */ -static int vtbl_signature_valid(__u8 signature[4]) -{ - const char *vtbl_ids[] = VTBL_IDS; /* valid signatures */ - int j; - - for (j = 0; - (j < NR_ITEMS(vtbl_ids)) && (memcmp(signature, vtbl_ids[j], 4) != 0); - j++); - return j < NR_ITEMS(vtbl_ids); -} - -/* We used to store the block-size of the volume in the volume-label, - * using the keyword "blocksize". The blocksize written to the - * volume-label is in bytes. - * - * We use this now only for compatibility with old zftape version. We - * store the blocksize directly as binary number in the vendor - * extension part of the volume entry. - */ -static int check_volume_label(const char *label, int *blk_sz) -{ - int valid_format; - char *blocksize; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "called with \"%s\" / \"%s\"", label, ZFT_VOL_NAME); - if (strncmp(label, ZFT_VOL_NAME, strlen(ZFT_VOL_NAME)) != 0) { - *blk_sz = 1; /* smallest block size that we allow */ - valid_format = 0; - } else { - TRACE(ft_t_noise, "got old style zftape vtbl entry"); - /* get the default blocksize */ - /* use the kernel strstr() */ - blocksize= strstr(label, " blocksize "); - if (blocksize) { - blocksize += strlen(" blocksize "); - for(*blk_sz= 0; - *blocksize >= '0' && *blocksize <= '9'; - blocksize++) { - *blk_sz *= 10; - *blk_sz += *blocksize - '0'; - } - if (*blk_sz > ZFT_MAX_BLK_SZ) { - *blk_sz= 1; - valid_format= 0; - } else { - valid_format = 1; - } - } else { - *blk_sz= 1; - valid_format= 0; - } - } - TRACE_EXIT valid_format; -} - -/* check for a zftape volume - */ -static int check_volume(__u8 *entry, zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if(strncmp(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) { - TRACE(ft_t_noise, "got new style zftape vtbl entry"); - volume->blk_sz = GET2(entry, VTBL_EXT+EXT_ZFTAPE_BLKSZ); - volume->qic113 = entry[VTBL_EXT+EXT_ZFTAPE_QIC113]; - TRACE_EXIT 1; - } else { - TRACE_EXIT check_volume_label(&entry[VTBL_DESC], &volume->blk_sz); - } -} - - -/* create zftape specific vtbl entry, the volume bounds are inserted - * in the calling function, zft_create_volume_headers() - */ -static void create_zft_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - memset(entry, 0, VTBL_SIZE); - memcpy(&entry[VTBL_SIG], VTBL_ID, 4); - sprintf(&entry[VTBL_DESC], ZFT_VOL_NAME" %03d", vtbl->count); - entry[VTBL_FLAGS] = (VTBL_FL_NOT_VERIFIED | VTBL_FL_SEG_SPANNING); - entry[VTBL_M_NO] = 1; /* multi_cartridge_count */ - strcpy(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG); - PUT2(entry, VTBL_EXT+EXT_ZFTAPE_BLKSZ, vtbl->blk_sz); - if (zft_qic113) { - PUT8(entry, VTBL_DATA_SIZE, vtbl->size); - entry[VTBL_CMPR] = VTBL_CMPR_UNREG; - if (vtbl->use_compression) { /* use compression: */ - entry[VTBL_CMPR] |= VTBL_CMPR_USED; - } - entry[VTBL_EXT+EXT_ZFTAPE_QIC113] = 1; - } else { - PUT4(entry, VTBL_DATA_SIZE, vtbl->size); - entry[VTBL_K_CMPR] = VTBL_CMPR_UNREG; - if (vtbl->use_compression) { /* use compression: */ - entry[VTBL_K_CMPR] |= VTBL_CMPR_USED; - } - } - if (ft_format_code == fmt_big) { - /* SCSI like vtbl, store the number of used - * segments as 4 byte value - */ - PUT4(entry, VTBL_SCSI_SEGS, vtbl->end_seg-vtbl->start_seg + 1); - } else { - /* normal, QIC-80MC like vtbl - */ - PUT2(entry, VTBL_START, vtbl->start_seg); - PUT2(entry, VTBL_END, vtbl->end_seg); - } - TRACE_EXIT; -} - -/* this one creates the volume headers for each volume. It is assumed - * that buffer already contains the old volume-table, so that vtbl - * entries without the zft_volume flag set can savely be ignored. - */ -static void zft_create_volume_headers(__u8 *buffer) -{ - __u8 *entry; - struct list_head *tmp; - zft_volinfo *vtbl; - TRACE_FUN(ft_t_flow); - -#ifdef ZFT_CMAP_HACK - if((strncmp(&buffer[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) && - buffer[VTBL_EXT+EXT_ZFTAPE_CMAP] != 0) { - TRACE(ft_t_noise, "deleting cmap volume"); - memmove(buffer, buffer + VTBL_SIZE, - FT_SEGMENT_SIZE - VTBL_SIZE); - } -#endif - entry = buffer; - for (tmp = zft_head_vtbl->node.next; - tmp != &zft_eom_vtbl->node; - tmp = tmp->next) { - vtbl = list_entry(tmp, zft_volinfo, node); - /* we now fill in the values only for newly created volumes. - */ - if (vtbl->new_volume) { - create_zft_volume(entry, vtbl); - vtbl->new_volume = 0; /* clear the flag */ - } - - DUMP_VOLINFO(ft_t_noise, &entry[VTBL_DESC], vtbl); - entry += VTBL_SIZE; - } - memset(entry, 0, FT_SEGMENT_SIZE - zft_eom_vtbl->count * VTBL_SIZE); - TRACE_EXIT; -} - -/* write volume table to tape. Calls zft_create_volume_headers() - */ -int zft_update_volume_table(unsigned int segment) -{ - int result = 0; - __u8 *verify_buf = NULL; - TRACE_FUN(ft_t_flow); - - TRACE_CATCH(result = ftape_read_segment(ft_first_data_segment, - zft_deblock_buf, - FT_RD_SINGLE),); - zft_create_volume_headers(zft_deblock_buf); - TRACE(ft_t_noise, "writing volume table segment %d", segment); - if (zft_vmalloc_once(&verify_buf, FT_SEGMENT_SIZE) == 0) { - TRACE_CATCH(zft_verify_write_segments(segment, - zft_deblock_buf, result, - verify_buf), - zft_vfree(&verify_buf, FT_SEGMENT_SIZE)); - zft_vfree(&verify_buf, FT_SEGMENT_SIZE); - } else { - TRACE_CATCH(ftape_write_segment(segment, zft_deblock_buf, - FT_WR_SINGLE),); - } - TRACE_EXIT 0; -} - -/* non zftape volumes are handled in raw mode. Thus we need to - * calculate the raw amount of data contained in those segments. - */ -static void extract_alien_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - vtbl->size = (zft_calc_tape_pos(zft_last_vtbl->end_seg+1) - - zft_calc_tape_pos(zft_last_vtbl->start_seg)); - vtbl->use_compression = 0; - vtbl->qic113 = zft_qic113; - if (vtbl->qic113) { - TRACE(ft_t_noise, - "Fake alien volume's size from " LL_X " to " LL_X, - LL(GET8(entry, VTBL_DATA_SIZE)), LL(vtbl->size)); - } else { - TRACE(ft_t_noise, - "Fake alien volume's size from %d to " LL_X, - (int)GET4(entry, VTBL_DATA_SIZE), LL(vtbl->size)); - } - TRACE_EXIT; -} - - -/* extract an zftape specific volume - */ -static void extract_zft_volume(__u8 *entry, zft_volinfo *vtbl) -{ - TRACE_FUN(ft_t_flow); - - if (vtbl->qic113) { - vtbl->size = GET8(entry, VTBL_DATA_SIZE); - vtbl->use_compression = - (entry[VTBL_CMPR] & VTBL_CMPR_USED) != 0; - } else { - vtbl->size = GET4(entry, VTBL_DATA_SIZE); - if (entry[VTBL_K_CMPR] & VTBL_CMPR_UNREG) { - vtbl->use_compression = - (entry[VTBL_K_CMPR] & VTBL_CMPR_USED) != 0; - } else if (entry[VTBL_CMPR] & VTBL_CMPR_UNREG) { - vtbl->use_compression = - (entry[VTBL_CMPR] & VTBL_CMPR_USED) != 0; - } else { - TRACE(ft_t_warn, "Geeh! There is something wrong:\n" - KERN_INFO "QIC compression (Rev = K): %x\n" - KERN_INFO "QIC compression (Rev > K): %x", - entry[VTBL_K_CMPR], entry[VTBL_CMPR]); - } - } - TRACE_EXIT; -} - -/* extract the volume table from buffer. "buffer" must already contain - * the vtbl-segment - */ -int zft_extract_volume_headers(__u8 *buffer) -{ - __u8 *entry; - TRACE_FUN(ft_t_flow); - - zft_init_vtbl(); - entry = buffer; -#ifdef ZFT_CMAP_HACK - if ((strncmp(&entry[VTBL_EXT+EXT_ZFTAPE_SIG], ZFTAPE_SIG, - strlen(ZFTAPE_SIG)) == 0) && - entry[VTBL_EXT+EXT_ZFTAPE_CMAP] != 0) { - TRACE(ft_t_noise, "ignoring cmap volume"); - entry += VTBL_SIZE; - } -#endif - /* the end of the vtbl is indicated by an invalid signature - */ - while (vtbl_signature_valid(&entry[VTBL_SIG]) && - (entry - buffer) < FT_SEGMENT_SIZE) { - zft_new_vtbl_entry(); - if (ft_format_code == fmt_big) { - /* SCSI like vtbl, stores only the number of - * segments used - */ - unsigned int num_segments= GET4(entry, VTBL_SCSI_SEGS); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = - zft_last_vtbl->start_seg + num_segments - 1; - } else { - /* `normal', QIC-80 like vtbl - */ - zft_last_vtbl->start_seg = GET2(entry, VTBL_START); - zft_last_vtbl->end_seg = GET2(entry, VTBL_END); - } - zft_eom_vtbl->start_seg = zft_last_vtbl->end_seg + 1; - /* check if we created this volume and get the - * blk_sz - */ - zft_last_vtbl->zft_volume = check_volume(entry, zft_last_vtbl); - if (zft_last_vtbl->zft_volume == 0) { - extract_alien_volume(entry, zft_last_vtbl); - } else { - extract_zft_volume(entry, zft_last_vtbl); - } - DUMP_VOLINFO(ft_t_noise, &entry[VTBL_DESC], zft_last_vtbl); - entry +=VTBL_SIZE; - } -#if 0 -/* - * undefine to test end of tape handling - */ - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = ft_last_data_segment - 10; - zft_last_vtbl->blk_sz = zft_blk_sz; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->qic113 = zft_qic113; - zft_last_vtbl->size = (zft_calc_tape_pos(zft_last_vtbl->end_seg+1) - - zft_calc_tape_pos(zft_last_vtbl->start_seg)); -#endif - TRACE_EXIT 0; -} - -/* this functions translates the failed_sector_log, misused as - * EOF-marker list, into a virtual volume table. The table mustn't be - * written to tape, because this would occupy the first data segment, - * which should be the volume table, but is actually the first segment - * that is filled with data (when using standard ftape). We assume, - * that we get a non-empty failed_sector_log. - */ -int zft_fake_volume_headers (eof_mark_union *eof_map, int num_failed_sectors) -{ - unsigned int segment, sector; - int have_eom = 0; - int vol_no; - TRACE_FUN(ft_t_flow); - - if ((num_failed_sectors >= 2) && - (GET2(&eof_map[num_failed_sectors - 1].mark.segment, 0) - == - GET2(&eof_map[num_failed_sectors - 2].mark.segment, 0) + 1) && - (GET2(&eof_map[num_failed_sectors - 1].mark.date, 0) == 1)) { - /* this should be eom. We keep the remainder of the - * tape as another volume. - */ - have_eom = 1; - } - zft_init_vtbl(); - zft_eom_vtbl->start_seg = ft_first_data_segment; - for(vol_no = 0; vol_no < num_failed_sectors - have_eom; vol_no ++) { - zft_new_vtbl_entry(); - - segment = GET2(&eof_map[vol_no].mark.segment, 0); - sector = GET2(&eof_map[vol_no].mark.date, 0); - - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = segment; - zft_eom_vtbl->start_seg = segment + 1; - zft_last_vtbl->blk_sz = 1; - zft_last_vtbl->size = - (zft_calc_tape_pos(zft_last_vtbl->end_seg) - - zft_calc_tape_pos(zft_last_vtbl->start_seg) - + (sector-1) * FT_SECTOR_SIZE); - TRACE(ft_t_noise, - "failed sector log: segment: %d, sector: %d", - segment, sector); - DUMP_VOLINFO(ft_t_noise, "Faked volume", zft_last_vtbl); - } - if (!have_eom) { - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = zft_eom_vtbl->start_seg; - zft_last_vtbl->end_seg = ft_last_data_segment; - zft_eom_vtbl->start_seg = ft_last_data_segment + 1; - zft_last_vtbl->size = zft_capacity; - zft_last_vtbl->size -= zft_calc_tape_pos(zft_last_vtbl->start_seg); - zft_last_vtbl->blk_sz = 1; - DUMP_VOLINFO(ft_t_noise, "Faked volume",zft_last_vtbl); - } - TRACE_EXIT 0; -} - -/* update the internal volume table - * - * if before start of last volume: erase all following volumes if - * inside a volume: set end of volume to infinity - * - * this function is intended to be called every time _ftape_write() is - * called - * - * return: 0 if no new volume was created, 1 if a new volume was - * created - * - * NOTE: we don't need to check for zft_mode as ftape_write() does - * that already. This function gets never called without accessing - * zftape via the *qft* devices - */ - -int zft_open_volume(zft_position *pos, int blk_sz, int use_compression) -{ - TRACE_FUN(ft_t_flow); - - if (!zft_qic_mode) { - TRACE_EXIT 0; - } - if (zft_tape_at_lbot(pos)) { - zft_init_vtbl(); - if(zft_old_ftape) { - /* clear old ftape's eof marks */ - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - zft_reset_position(pos); - } - if (pos->seg_pos != zft_last_vtbl->end_seg + 1) { - TRACE_ABORT(-EIO, ft_t_bug, - "BUG: seg_pos: %d, zft_last_vtbl->end_seg: %d", - pos->seg_pos, zft_last_vtbl->end_seg); - } - TRACE(ft_t_noise, "create new volume"); - if (zft_eom_vtbl->count >= ZFT_MAX_VOLUMES) { - TRACE_ABORT(-ENOSPC, ft_t_err, - "Error: maxmimal number of volumes exhausted " - "(maxmimum is %d)", ZFT_MAX_VOLUMES); - } - zft_new_vtbl_entry(); - pos->volume_pos = pos->seg_byte_pos = 0; - zft_last_vtbl->start_seg = pos->seg_pos; - zft_last_vtbl->end_seg = ft_last_data_segment; /* infinity */ - zft_last_vtbl->blk_sz = blk_sz; - zft_last_vtbl->size = zft_capacity; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->use_compression = use_compression; - zft_last_vtbl->qic113 = zft_qic113; - zft_last_vtbl->new_volume = 1; - zft_last_vtbl->open = 1; - zft_volume_table_changed = 1; - zft_eom_vtbl->start_seg = ft_last_data_segment + 1; - TRACE_EXIT 0; -} - -/* perform mtfsf, mtbsf, not allowed without zft_qic_mode - */ -int zft_skip_volumes(int count, zft_position *pos) -{ - const zft_volinfo *vtbl; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "count: %d", count); - - vtbl= zft_find_volume(pos->seg_pos); - while (count > 0 && vtbl != zft_eom_vtbl) { - vtbl = list_entry(vtbl->node.next, zft_volinfo, node); - count --; - } - while (count < 0 && vtbl != zft_first_vtbl) { - vtbl = list_entry(vtbl->node.prev, zft_volinfo, node); - count ++; - } - pos->seg_pos = vtbl->start_seg; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - zft_just_before_eof = vtbl->size == 0; - if (zft_cmpr_ops) { - (*zft_cmpr_ops->reset)(); - } - zft_deblock_segment = -1; /* no need to keep cache */ - TRACE(ft_t_noise, "repositioning to:\n" - KERN_INFO "zft_seg_pos : %d\n" - KERN_INFO "zft_seg_byte_pos : %d\n" - KERN_INFO "zft_tape_pos : " LL_X "\n" - KERN_INFO "zft_volume_pos : " LL_X "\n" - KERN_INFO "file number : %d", - pos->seg_pos, pos->seg_byte_pos, - LL(pos->tape_pos), LL(pos->volume_pos), vtbl->count); - zft_resid = count < 0 ? -count : count; - TRACE_EXIT zft_resid ? -EINVAL : 0; -} - -/* the following simply returns the raw data position of the EOM - * marker, MTIOCSIZE ioctl - */ -__s64 zft_get_eom_pos(void) -{ - if (zft_qic_mode) { - return zft_calc_tape_pos(zft_eom_vtbl->start_seg); - } else { - /* there is only one volume in raw mode */ - return zft_capacity; - } -} - -/* skip to eom, used for MTEOM - */ -void zft_skip_to_eom(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - pos->seg_pos = zft_eom_vtbl->start_seg; - pos->seg_byte_pos = - pos->volume_pos = - zft_just_before_eof = 0; - pos->tape_pos = zft_calc_tape_pos(pos->seg_pos); - TRACE(ft_t_noise, "ftape positioned to segment %d, data pos " LL_X, - pos->seg_pos, LL(pos->tape_pos)); - TRACE_EXIT; -} - -/* write an EOF-marker by setting zft_last_vtbl->end_seg to seg_pos. - * NOTE: this function assumes that zft_last_vtbl points to a valid - * vtbl entry - * - * NOTE: this routine always positions before the EOF marker - */ -int zft_close_volume(zft_position *pos) -{ - TRACE_FUN(ft_t_any); - - if (zft_vtbl_empty || !zft_last_vtbl->open) { /* should not happen */ - TRACE(ft_t_noise, "There are no volumes to finish"); - TRACE_EXIT -EIO; - } - if (pos->seg_byte_pos == 0 && - pos->seg_pos != zft_last_vtbl->start_seg) { - pos->seg_pos --; - pos->seg_byte_pos = zft_get_seg_sz(pos->seg_pos); - } - zft_last_vtbl->end_seg = pos->seg_pos; - zft_last_vtbl->size = pos->volume_pos; - zft_volume_table_changed = 1; - zft_just_before_eof = 1; - zft_eom_vtbl->start_seg = zft_last_vtbl->end_seg + 1; - zft_last_vtbl->open = 0; /* closed */ - TRACE_EXIT 0; -} - -/* write count file-marks at current position. - * - * The tape is positioned after the eof-marker, that is at byte 0 of - * the segment following the eof-marker - * - * this function is only allowed in zft_qic_mode - * - * Only allowed when tape is at BOT or EOD. - */ -int zft_weof(unsigned int count, zft_position *pos) -{ - - TRACE_FUN(ft_t_flow); - - if (!count) { /* write zero EOF marks should be a real no-op */ - TRACE_EXIT 0; - } - zft_volume_table_changed = 1; - if (zft_tape_at_lbot(pos)) { - zft_init_vtbl(); - if(zft_old_ftape) { - /* clear old ftape's eof marks */ - zft_clear_ftape_file_marks(); - zft_old_ftape = 0; /* no longer old ftape */ - } - } - if (zft_last_vtbl->open) { - zft_close_volume(pos); - zft_move_past_eof(pos); - count --; - } - /* now it's easy, just append eof-marks, that is empty - * volumes, to the end of the already recorded media. - */ - while (count > 0 && - pos->seg_pos <= ft_last_data_segment && - zft_eom_vtbl->count < ZFT_MAX_VOLUMES) { - TRACE(ft_t_noise, - "Writing zero sized file at segment %d", pos->seg_pos); - zft_new_vtbl_entry(); - zft_last_vtbl->start_seg = pos->seg_pos; - zft_last_vtbl->end_seg = pos->seg_pos; - zft_last_vtbl->size = 0; - zft_last_vtbl->blk_sz = zft_blk_sz; - zft_last_vtbl->zft_volume = 1; - zft_last_vtbl->use_compression = 0; - pos->tape_pos += zft_get_seg_sz(pos->seg_pos); - zft_eom_vtbl->start_seg = ++ pos->seg_pos; - count --; - } - if (count > 0) { - /* there are two possibilities: end of tape, or the - * maximum number of files is exhausted. - */ - zft_resid = count; - TRACE(ft_t_noise,"Number of marks NOT written: %d", zft_resid); - if (zft_eom_vtbl->count == ZFT_MAX_VOLUMES) { - TRACE_ABORT(-EINVAL, ft_t_warn, - "maximum allowed number of files " - "exhausted: %d", ZFT_MAX_VOLUMES); - } else { - TRACE_ABORT(-ENOSPC, - ft_t_noise, "reached end of tape"); - } - } - TRACE_EXIT 0; -} - -const zft_volinfo *zft_find_volume(unsigned int seg_pos) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_any, "called with seg_pos %d",seg_pos); - if (!zft_qic_mode) { - if (seg_pos > ft_last_data_segment) { - TRACE_EXIT &eot_vtbl; - } - tape_vtbl.blk_sz = zft_blk_sz; - TRACE_EXIT &tape_vtbl; - } - if (seg_pos < zft_first_vtbl->start_seg) { - TRACE_EXIT (cur_vtbl = zft_first_vtbl); - } - while (seg_pos > cur_vtbl->end_seg) { - cur_vtbl = list_entry(cur_vtbl->node.next, zft_volinfo, node); - TRACE(ft_t_noise, "%d - %d", cur_vtbl->start_seg, cur_vtbl->end_seg); - } - while (seg_pos < cur_vtbl->start_seg) { - cur_vtbl = list_entry(cur_vtbl->node.prev, zft_volinfo, node); - TRACE(ft_t_noise, "%d - %d", cur_vtbl->start_seg, cur_vtbl->end_seg); - } - if (seg_pos > cur_vtbl->end_seg || seg_pos < cur_vtbl->start_seg) { - TRACE(ft_t_bug, "This cannot happen"); - } - DUMP_VOLINFO(ft_t_noise, "", cur_vtbl); - TRACE_EXIT cur_vtbl; -} - -/* this function really assumes that we are just before eof - */ -void zft_move_past_eof(zft_position *pos) -{ - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_noise, "old seg. pos: %d", pos->seg_pos); - pos->tape_pos += zft_get_seg_sz(pos->seg_pos++) - pos->seg_byte_pos; - pos->seg_byte_pos = 0; - pos->volume_pos = 0; - if (zft_cmpr_ops) { - (*zft_cmpr_ops->reset)(); - } - zft_just_before_eof = 0; - zft_deblock_segment = -1; /* no need to cache it anymore */ - TRACE(ft_t_noise, "new seg. pos: %d", pos->seg_pos); - TRACE_EXIT; -} diff --git a/drivers/char/ftape/zftape/zftape-vtbl.h b/drivers/char/ftape/zftape/zftape-vtbl.h deleted file mode 100644 index f31d196d1759..000000000000 --- a/drivers/char/ftape/zftape/zftape-vtbl.h +++ /dev/null @@ -1,227 +0,0 @@ -#ifndef _ZFTAPE_VTBL_H -#define _ZFTAPE_VTBL_H - -/* - * Copyright (c) 1995-1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or - modify it under the terms of the GNU General Public License as - published by the Free Software Foundation; either version 2, or (at - your option) any later version. - - This program is distributed in the hope that it will be useful, but - WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, - USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-vtbl.h,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/28 14:30:09 $ - * - * This file defines a volume table as defined in the QIC-80 - * development standards. - */ - -#include - -#include "../lowlevel/ftape-tracing.h" - -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-rw.h" - -#define VTBL_SIZE 128 /* bytes */ - -/* The following are offsets in the vtbl. */ -#define VTBL_SIG 0 -#define VTBL_START 4 -#define VTBL_END 6 -#define VTBL_DESC 8 -#define VTBL_DATE 52 -#define VTBL_FLAGS 56 -#define VTBL_FL_VENDOR_SPECIFIC (1<<0) -#define VTBL_FL_MUTLI_CARTRIDGE (1<<1) -#define VTBL_FL_NOT_VERIFIED (1<<2) -#define VTBL_FL_REDIR_INHIBIT (1<<3) -#define VTBL_FL_SEG_SPANNING (1<<4) -#define VTBL_FL_DIRECTORY_LAST (1<<5) -#define VTBL_FL_RESERVED_6 (1<<6) -#define VTBL_FL_RESERVED_7 (1<<7) -#define VTBL_M_NO 57 -#define VTBL_EXT 58 -#define EXT_ZFTAPE_SIG 0 -#define EXT_ZFTAPE_BLKSZ 10 -#define EXT_ZFTAPE_CMAP 12 -#define EXT_ZFTAPE_QIC113 13 -#define VTBL_PWD 84 -#define VTBL_DIR_SIZE 92 -#define VTBL_DATA_SIZE 96 -#define VTBL_OS_VERSION 104 -#define VTBL_SRC_DRIVE 106 -#define VTBL_DEV 122 -#define VTBL_RESERVED_1 123 -#define VTBL_CMPR 124 -#define VTBL_CMPR_UNREG 0x3f -#define VTBL_CMPR_USED 0x80 -#define VTBL_FMT 125 -#define VTBL_RESERVED_2 126 -#define VTBL_RESERVED_3 127 -/* compatibility with pre revision K */ -#define VTBL_K_CMPR 120 - -/* the next is used by QIC-3020 tapes with format code 6 (>2^16 - * segments) It is specified in QIC-113, Rev. G, Section 5 (SCSI - * volume table). The difference is simply, that we only store the - * number of segments used, not the starting segment. - */ -#define VTBL_SCSI_SEGS 4 /* is a 4 byte value */ - -/* one vtbl is 128 bytes, that results in a maximum number of - * 29*1024/128 = 232 volumes. - */ -#define ZFT_MAX_VOLUMES (FT_SEGMENT_SIZE/VTBL_SIZE) -#define VTBL_ID "VTBL" -#define VTBL_IDS { VTBL_ID, "XTBL", "UTID", "EXVT" } /* other valid ids */ -#define ZFT_VOL_NAME "zftape volume" /* volume label used by me */ -#define ZFTAPE_SIG "LINUX ZFT" - -/* global variables - */ -typedef struct zft_internal_vtbl -{ - struct list_head node; - int count; - unsigned int start_seg; /* 32 bits are enough for now */ - unsigned int end_seg; /* 32 bits are enough for now */ - __s64 size; /* uncompressed size */ - unsigned int blk_sz; /* block size for this volume */ - unsigned int zft_volume :1; /* zftape created this volume */ - unsigned int use_compression:1; /* compressed volume */ - unsigned int qic113 :1; /* layout of compressed block - * info and vtbl conforms to - * QIC-113, Rev. G - */ - unsigned int new_volume :1; /* it was created by us, this - * run. this allows the - * fields that aren't really - * used by zftape to be filled - * in by some user level - * program. - */ - unsigned int open :1; /* just in progress of being - * written - */ -} zft_volinfo; - -extern struct list_head zft_vtbl; -#define zft_head_vtbl list_entry(zft_vtbl.next, zft_volinfo, node) -#define zft_eom_vtbl list_entry(zft_vtbl.prev, zft_volinfo, node) -#define zft_last_vtbl list_entry(zft_eom_vtbl->node.prev, zft_volinfo, node) -#define zft_first_vtbl list_entry(zft_head_vtbl->node.next, zft_volinfo, node) -#define zft_vtbl_empty (zft_eom_vtbl->node.prev == &zft_head_vtbl->node) - -#define DUMP_VOLINFO(level, desc, info) \ -{ \ - char tmp[21]; \ - strlcpy(tmp, desc, sizeof(tmp)); \ - TRACE(level, "Volume %d:\n" \ - KERN_INFO "description : %s\n" \ - KERN_INFO "first segment: %d\n" \ - KERN_INFO "last segment: %d\n" \ - KERN_INFO "size : " LL_X "\n" \ - KERN_INFO "block size : %d\n" \ - KERN_INFO "compression : %d\n" \ - KERN_INFO "zftape volume: %d\n" \ - KERN_INFO "QIC-113 conf.: %d", \ - (info)->count, tmp, (info)->start_seg, (info)->end_seg, \ - LL((info)->size), (info)->blk_sz, \ - (info)->use_compression != 0, (info)->zft_volume != 0, \ - (info)->qic113 != 0); \ -} - -extern int zft_qic_mode; -extern int zft_old_ftape; -extern int zft_volume_table_changed; - -/* exported functions */ -extern void zft_init_vtbl (void); -extern void zft_free_vtbl (void); -extern int zft_extract_volume_headers(__u8 *buffer); -extern int zft_update_volume_table (unsigned int segment); -extern int zft_open_volume (zft_position *pos, - int blk_sz, int use_compression); -extern int zft_close_volume (zft_position *pos); -extern const zft_volinfo *zft_find_volume(unsigned int seg_pos); -extern int zft_skip_volumes (int count, zft_position *pos); -extern __s64 zft_get_eom_pos (void); -extern void zft_skip_to_eom (zft_position *pos); -extern int zft_fake_volume_headers (eof_mark_union *eof_map, - int num_failed_sectors); -extern int zft_weof (unsigned int count, zft_position *pos); -extern void zft_move_past_eof (zft_position *pos); - -static inline int zft_tape_at_eod (const zft_position *pos); -static inline int zft_tape_at_lbot (const zft_position *pos); -static inline void zft_position_before_eof (zft_position *pos, - const zft_volinfo *volume); -static inline __s64 zft_check_for_eof(const zft_volinfo *vtbl, - const zft_position *pos); - -/* this function decrements the zft_seg_pos counter if we are right - * at the beginning of a segment. This is to handle fsfm/bsfm -- we - * need to position before the eof mark. NOTE: zft_tape_pos is not - * changed - */ -static inline void zft_position_before_eof(zft_position *pos, - const zft_volinfo *volume) -{ - TRACE_FUN(ft_t_flow); - - if (pos->seg_pos == volume->end_seg + 1 && pos->seg_byte_pos == 0) { - pos->seg_pos --; - pos->seg_byte_pos = zft_get_seg_sz(pos->seg_pos); - } - TRACE_EXIT; -} - -/* Mmmh. Is the position at the end of the last volume, that is right - * before the last EOF mark also logical an EOD condition? - */ -static inline int zft_tape_at_eod(const zft_position *pos) -{ - TRACE_FUN(ft_t_any); - - if (zft_qic_mode) { - TRACE_EXIT (pos->seg_pos >= zft_eom_vtbl->start_seg || - zft_last_vtbl->open); - } else { - TRACE_EXIT pos->seg_pos > ft_last_data_segment; - } -} - -static inline int zft_tape_at_lbot(const zft_position *pos) -{ - if (zft_qic_mode) { - return (pos->seg_pos <= zft_first_vtbl->start_seg && - pos->volume_pos == 0); - } else { - return (pos->seg_pos <= ft_first_data_segment && - pos->volume_pos == 0); - } -} - -/* This one checks for EOF. return remaing space (may be negative) - */ -static inline __s64 zft_check_for_eof(const zft_volinfo *vtbl, - const zft_position *pos) -{ - return (__s64)(vtbl->size - pos->volume_pos); -} - -#endif /* _ZFTAPE_VTBL_H */ diff --git a/drivers/char/ftape/zftape/zftape-write.c b/drivers/char/ftape/zftape/zftape-write.c deleted file mode 100644 index 94327b8c97b9..000000000000 --- a/drivers/char/ftape/zftape/zftape-write.c +++ /dev/null @@ -1,483 +0,0 @@ -/* - * Copyright (C) 1996, 1997 Claus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-write.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/11/06 00:50:29 $ - * - * This file contains the writing code - * for the QIC-117 floppy-tape driver for Linux. - */ - -#include -#include - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-eof.h" -#include "../zftape/zftape-ctl.h" -#include "../zftape/zftape-write.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-rw.h" -#include "../zftape/zftape-vtbl.h" - -/* Global vars. - */ - -/* Local vars. - */ -static int last_write_failed; -static int need_flush; - -void zft_prevent_flush(void) -{ - need_flush = 0; -} - -static int zft_write_header_segments(__u8* buffer) -{ - int header_1_ok = 0; - int header_2_ok = 0; - unsigned int time_stamp; - TRACE_FUN(ft_t_noise); - - TRACE_CATCH(ftape_abort_operation(),); - ftape_seek_to_bot(); /* prevents extra rewind */ - if (GET4(buffer, 0) != FT_HSEG_MAGIC) { - TRACE_ABORT(-EIO, ft_t_err, - "wrong header signature found, aborting"); - } - /* Be optimistic: */ - PUT4(buffer, FT_SEG_CNT, - zft_written_segments + GET4(buffer, FT_SEG_CNT) + 2); - if ((time_stamp = zft_get_time()) != 0) { - PUT4(buffer, FT_WR_DATE, time_stamp); - if (zft_label_changed) { - PUT4(buffer, FT_LABEL_DATE, time_stamp); - } - } - TRACE(ft_t_noise, - "writing first header segment %d", ft_header_segment_1); - header_1_ok = zft_verify_write_segments(ft_header_segment_1, - buffer, FT_SEGMENT_SIZE, - zft_deblock_buf) >= 0; - TRACE(ft_t_noise, - "writing second header segment %d", ft_header_segment_2); - header_2_ok = zft_verify_write_segments(ft_header_segment_2, - buffer, FT_SEGMENT_SIZE, - zft_deblock_buf) >= 0; - if (!header_1_ok) { - TRACE(ft_t_warn, "Warning: " - "update of first header segment failed"); - } - if (!header_2_ok) { - TRACE(ft_t_warn, "Warning: " - "update of second header segment failed"); - } - if (!header_1_ok && !header_2_ok) { - TRACE_ABORT(-EIO, ft_t_err, "Error: " - "update of both header segments failed."); - } - TRACE_EXIT 0; -} - -int zft_update_header_segments(void) -{ - TRACE_FUN(ft_t_noise); - - /* must NOT use zft_write_protected, as it also includes the - * file access mode. But we also want to update when soft - * write protection is enabled (O_RDONLY) - */ - if (ft_write_protected || zft_old_ftape) { - TRACE_ABORT(0, ft_t_noise, "Tape set read-only: no update"); - } - if (!zft_header_read) { - TRACE_ABORT(0, ft_t_noise, "Nothing to update"); - } - if (!zft_header_changed) { - zft_header_changed = zft_written_segments > 0; - } - if (!zft_header_changed && !zft_volume_table_changed) { - TRACE_ABORT(0, ft_t_noise, "Nothing to update"); - } - TRACE(ft_t_noise, "Updating header segments"); - if (ftape_get_status()->fti_state == writing) { - TRACE_CATCH(ftape_loop_until_writes_done(),); - } - TRACE_CATCH(ftape_abort_operation(),); - - zft_deblock_segment = -1; /* invalidate the cache */ - if (zft_header_changed) { - TRACE_CATCH(zft_write_header_segments(zft_hseg_buf),); - } - if (zft_volume_table_changed) { - TRACE_CATCH(zft_update_volume_table(ft_first_data_segment),); - } - zft_header_changed = - zft_volume_table_changed = - zft_label_changed = - zft_written_segments = 0; - TRACE_CATCH(ftape_abort_operation(),); - ftape_seek_to_bot(); - TRACE_EXIT 0; -} - -static int read_merge_buffer(int seg_pos, __u8 *buffer, int offset, int seg_sz) -{ - int result = 0; - const ft_trace_t old_tracing = TRACE_LEVEL; - TRACE_FUN(ft_t_flow); - - if (zft_qic_mode) { - /* writing in the middle of a volume is NOT allowed - * - */ - TRACE(ft_t_noise, "No need to read a segment"); - memset(buffer + offset, 0, seg_sz - offset); - TRACE_EXIT 0; - } - TRACE(ft_t_any, "waiting"); - ftape_start_writing(FT_WR_MULTI); - TRACE_CATCH(ftape_loop_until_writes_done(),); - - TRACE(ft_t_noise, "trying to read segment %d from offset %d", - seg_pos, offset); - SET_TRACE_LEVEL(ft_t_bug); - result = zft_fetch_segment_fraction(seg_pos, buffer, - FT_RD_SINGLE, - offset, seg_sz - offset); - SET_TRACE_LEVEL(old_tracing); - if (result != (seg_sz - offset)) { - TRACE(ft_t_noise, "Ignore error: read_segment() result: %d", - result); - memset(buffer + offset, 0, seg_sz - offset); - } - TRACE_EXIT 0; -} - -/* flush the write buffer to tape and write an eof-marker at the - * current position if not in raw mode. This function always - * positions the tape before the eof-marker. _ftape_close() should - * then advance to the next segment. - * - * the parameter "finish_volume" describes whether to position before - * or after the possibly created file-mark. We always position after - * the file-mark when called from ftape_close() and a flush was needed - * (that is ftape_write() was the last tape operation before calling - * ftape_flush) But we always position before the file-mark when this - * function get's called from outside ftape_close() - */ -int zft_flush_buffers(void) -{ - int result; - int data_remaining; - int this_segs_size; - TRACE_FUN(ft_t_flow); - - TRACE(ft_t_data_flow, - "entered, ftape_state = %d", ftape_get_status()->fti_state); - if (ftape_get_status()->fti_state != writing && !need_flush) { - TRACE_ABORT(0, ft_t_noise, "no need for flush"); - } - zft_io_state = zft_idle; /* triggers some initializations for the - * read and write routines - */ - if (last_write_failed) { - ftape_abort_operation(); - TRACE_EXIT -EIO; - } - TRACE(ft_t_noise, "flushing write buffers"); - this_segs_size = zft_get_seg_sz(zft_pos.seg_pos); - if (this_segs_size == zft_pos.seg_byte_pos) { - zft_pos.seg_pos ++; - data_remaining = zft_pos.seg_byte_pos = 0; - } else { - data_remaining = zft_pos.seg_byte_pos; - } - /* If there is any data not written to tape yet, append zero's - * up to the end of the sector (if using compression) or merge - * it with the data existing on the tape Then write the - * segment(s) to tape. - */ - TRACE(ft_t_noise, "Position:\n" - KERN_INFO "seg_pos : %d\n" - KERN_INFO "byte pos : %d\n" - KERN_INFO "remaining: %d", - zft_pos.seg_pos, zft_pos.seg_byte_pos, data_remaining); - if (data_remaining > 0) { - do { - this_segs_size = zft_get_seg_sz(zft_pos.seg_pos); - if (this_segs_size > data_remaining) { - TRACE_CATCH(read_merge_buffer(zft_pos.seg_pos, - zft_deblock_buf, - data_remaining, - this_segs_size), - last_write_failed = 1); - } - result = ftape_write_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_WR_MULTI); - if (result != this_segs_size) { - TRACE(ft_t_err, "flush buffers failed"); - zft_pos.tape_pos -= zft_pos.seg_byte_pos; - zft_pos.seg_byte_pos = 0; - - last_write_failed = 1; - TRACE_EXIT result; - } - zft_written_segments ++; - TRACE(ft_t_data_flow, - "flush, moved out buffer: %d", result); - /* need next segment for more data (empty segments?) - */ - if (result < data_remaining) { - if (result > 0) { - /* move remainder to buffer beginning - */ - memmove(zft_deblock_buf, - zft_deblock_buf + result, - FT_SEGMENT_SIZE - result); - } - } - data_remaining -= result; - zft_pos.seg_pos ++; - } while (data_remaining > 0); - TRACE(ft_t_any, "result: %d", result); - zft_deblock_segment = --zft_pos.seg_pos; - if (data_remaining == 0) { /* first byte next segment */ - zft_pos.seg_byte_pos = this_segs_size; - } else { /* after data previous segment, data_remaining < 0 */ - zft_pos.seg_byte_pos = data_remaining + result; - } - } else { - TRACE(ft_t_noise, "zft_deblock_buf empty"); - zft_pos.seg_pos --; - zft_pos.seg_byte_pos = zft_get_seg_sz (zft_pos.seg_pos); - ftape_start_writing(FT_WR_MULTI); - } - TRACE(ft_t_any, "waiting"); - if ((result = ftape_loop_until_writes_done()) < 0) { - /* that's really bad. What to to with zft_tape_pos? - */ - TRACE(ft_t_err, "flush buffers failed"); - } - TRACE(ft_t_any, "zft_seg_pos: %d, zft_seg_byte_pos: %d", - zft_pos.seg_pos, zft_pos.seg_byte_pos); - last_write_failed = - need_flush = 0; - TRACE_EXIT result; -} - -/* return-value: the number of bytes removed from the user-buffer - * - * out: - * int *write_cnt: how much actually has been moved to the - * zft_deblock_buf - * int req_len : MUST NOT BE CHANGED, except at EOT, in - * which case it may be adjusted - * in : - * char *buff : the user buffer - * int buf_pos_write : copy of buf_len_wr int - * this_segs_size : the size in bytes of the actual segment - * char - * *zft_deblock_buf : zft_deblock_buf - */ -static int zft_simple_write(int *cnt, - __u8 *dst_buf, const int seg_sz, - const __u8 __user *src_buf, const int req_len, - const zft_position *pos,const zft_volinfo *volume) -{ - int space_left; - TRACE_FUN(ft_t_flow); - - /* volume->size holds the tape capacity while volume is open */ - if (pos->tape_pos + volume->blk_sz > volume->size) { - TRACE_EXIT -ENOSPC; - } - /* remaining space in this segment, NOT zft_deblock_buf - */ - space_left = seg_sz - pos->seg_byte_pos; - *cnt = req_len < space_left ? req_len : space_left; - if (copy_from_user(dst_buf + pos->seg_byte_pos, src_buf, *cnt) != 0) { - TRACE_EXIT -EFAULT; - } - TRACE_EXIT *cnt; -} - -static int check_write_access(int req_len, - const zft_volinfo **volume, - zft_position *pos, - const unsigned int blk_sz) -{ - int result; - TRACE_FUN(ft_t_flow); - - if ((req_len % zft_blk_sz) != 0) { - TRACE_ABORT(-EINVAL, ft_t_info, - "write-count %d must be multiple of block-size %d", - req_len, blk_sz); - } - if (zft_io_state == zft_writing) { - /* all other error conditions have been checked earlier - */ - TRACE_EXIT 0; - } - zft_io_state = zft_idle; - TRACE_CATCH(zft_check_write_access(pos),); - /* If we haven't read the header segment yet, do it now. - * This will verify the configuration, get the bad sector - * table and read the volume table segment - */ - if (!zft_header_read) { - TRACE_CATCH(zft_read_header_segments(),); - } - /* fine. Now the tape is either at BOT or at EOD, - * Write start of volume now - */ - TRACE_CATCH(zft_open_volume(pos, blk_sz, zft_use_compression),); - *volume = zft_find_volume(pos->seg_pos); - DUMP_VOLINFO(ft_t_noise, "", *volume); - zft_just_before_eof = 0; - /* now merge with old data if necessary */ - if (!zft_qic_mode && pos->seg_byte_pos != 0){ - result = zft_fetch_segment(pos->seg_pos, - zft_deblock_buf, - FT_RD_SINGLE); - if (result < 0) { - if (result == -EINTR || result == -ENOSPC) { - TRACE_EXIT result; - } - TRACE(ft_t_noise, - "ftape_read_segment() result: %d. " - "This might be normal when using " - "a newly\nformatted tape", result); - memset(zft_deblock_buf, '\0', pos->seg_byte_pos); - } - } - zft_io_state = zft_writing; - TRACE_EXIT 0; -} - -static int fill_deblock_buf(__u8 *dst_buf, const int seg_sz, - zft_position *pos, const zft_volinfo *volume, - const char __user *usr_buf, const int req_len) -{ - int cnt = 0; - int result = 0; - TRACE_FUN(ft_t_flow); - - if (seg_sz == 0) { - TRACE_ABORT(0, ft_t_data_flow, "empty segment"); - } - TRACE(ft_t_data_flow, "\n" - KERN_INFO "remaining req_len: %d\n" - KERN_INFO " buf_pos: %d", - req_len, pos->seg_byte_pos); - /* zft_deblock_buf will not contain a valid segment any longer */ - zft_deblock_segment = -1; - if (zft_use_compression) { - TRACE_CATCH(zft_cmpr_lock(1 /* try to load */),); - TRACE_CATCH(result= (*zft_cmpr_ops->write)(&cnt, - dst_buf, seg_sz, - usr_buf, req_len, - pos, volume),); - } else { - TRACE_CATCH(result= zft_simple_write(&cnt, - dst_buf, seg_sz, - usr_buf, req_len, - pos, volume),); - } - pos->volume_pos += result; - pos->seg_byte_pos += cnt; - pos->tape_pos += cnt; - TRACE(ft_t_data_flow, "\n" - KERN_INFO "removed from user-buffer : %d bytes.\n" - KERN_INFO "copied to zft_deblock_buf: %d bytes.\n" - KERN_INFO "zft_tape_pos : " LL_X " bytes.", - result, cnt, LL(pos->tape_pos)); - TRACE_EXIT result; -} - - -/* called by the kernel-interface routine "zft_write()" - */ -int _zft_write(const char __user *buff, int req_len) -{ - int result = 0; - int written = 0; - int write_cnt; - int seg_sz; - static const zft_volinfo *volume = NULL; - TRACE_FUN(ft_t_flow); - - zft_resid = req_len; - last_write_failed = 1; /* reset to 0 when successful */ - /* check if write is allowed - */ - TRACE_CATCH(check_write_access(req_len, &volume,&zft_pos,zft_blk_sz),); - while (req_len > 0) { - /* Allow us to escape from this loop with a signal ! - */ - FT_SIGNAL_EXIT(_DONT_BLOCK); - seg_sz = zft_get_seg_sz(zft_pos.seg_pos); - if ((write_cnt = fill_deblock_buf(zft_deblock_buf, - seg_sz, - &zft_pos, - volume, - buff, - req_len)) < 0) { - zft_resid -= written; - if (write_cnt == -ENOSPC) { - /* leave the remainder to flush_buffers() - */ - TRACE(ft_t_info, "No space left on device"); - last_write_failed = 0; - if (!need_flush) { - need_flush = written > 0; - } - TRACE_EXIT written > 0 ? written : -ENOSPC; - } else { - TRACE_EXIT result; - } - } - if (zft_pos.seg_byte_pos == seg_sz) { - TRACE_CATCH(ftape_write_segment(zft_pos.seg_pos, - zft_deblock_buf, - FT_WR_ASYNC), - zft_resid -= written); - zft_written_segments ++; - zft_pos.seg_byte_pos = 0; - zft_deblock_segment = zft_pos.seg_pos; - ++zft_pos.seg_pos; - } - written += write_cnt; - buff += write_cnt; - req_len -= write_cnt; - } /* while (req_len > 0) */ - TRACE(ft_t_data_flow, "remaining in blocking buffer: %d", - zft_pos.seg_byte_pos); - TRACE(ft_t_data_flow, "just written bytes: %d", written); - last_write_failed = 0; - zft_resid -= written; - need_flush = need_flush || written > 0; - TRACE_EXIT written; /* bytes written */ -} diff --git a/drivers/char/ftape/zftape/zftape-write.h b/drivers/char/ftape/zftape/zftape-write.h deleted file mode 100644 index ea887015b493..000000000000 --- a/drivers/char/ftape/zftape/zftape-write.h +++ /dev/null @@ -1,38 +0,0 @@ -#ifndef _ZFTAPE_WRITE_H -#define _ZFTAPE_WRITE_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape-write.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:13 $ - * - * This file contains the definitions for the write functions - * for the zftape driver for Linux. - * - */ - -extern int zft_flush_buffers(void); -extern int zft_update_header_segments(void); -extern void zft_prevent_flush(void); - -/* hook for the VFS interface - */ -extern int _zft_write(const char __user *buff, int req_len); -#endif /* _ZFTAPE_WRITE_H */ diff --git a/drivers/char/ftape/zftape/zftape_syms.c b/drivers/char/ftape/zftape/zftape_syms.c deleted file mode 100644 index 2db1401682df..000000000000 --- a/drivers/char/ftape/zftape/zftape_syms.c +++ /dev/null @@ -1,43 +0,0 @@ -/* - * Copyright (C) 1997 Claus-Justus Heine - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/ftape/zftape/zftape_syms.c,v $ - * $Revision: 1.3 $ - * $Date: 1997/10/05 19:19:14 $ - * - * This file contains the symbols that the zftape frontend to - * the ftape floppy tape driver exports - */ - -#include - -#include - -#include "../zftape/zftape-init.h" -#include "../zftape/zftape-read.h" -#include "../zftape/zftape-buffers.h" -#include "../zftape/zftape-ctl.h" - -/* zftape-init.c */ -EXPORT_SYMBOL(zft_cmpr_register); -/* zftape-read.c */ -EXPORT_SYMBOL(zft_fetch_segment_fraction); -/* zftape-buffers.c */ -EXPORT_SYMBOL(zft_vmalloc_once); -EXPORT_SYMBOL(zft_vmalloc_always); -EXPORT_SYMBOL(zft_vfree); diff --git a/include/linux/Kbuild b/include/linux/Kbuild index d7e04689304c..ff433126361f 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -60,8 +60,6 @@ header-y += fadvise.h header-y += fd.h header-y += fdreg.h header-y += fib_rules.h -header-y += ftape-header-segment.h -header-y += ftape-vendors.h header-y += fuse.h header-y += futex.h header-y += genetlink.h @@ -206,7 +204,6 @@ unifdef-y += fcntl.h unifdef-y += filter.h unifdef-y += flat.h unifdef-y += fs.h -unifdef-y += ftape.h unifdef-y += gameport.h unifdef-y += generic_serial.h unifdef-y += genhd.h @@ -341,6 +338,5 @@ unifdef-y += wait.h unifdef-y += wanrouter.h unifdef-y += watchdog.h unifdef-y += xfrm.h -unifdef-y += zftape.h objhdr-y += version.h diff --git a/include/linux/ftape-header-segment.h b/include/linux/ftape-header-segment.h deleted file mode 100644 index 4732218f0708..000000000000 --- a/include/linux/ftape-header-segment.h +++ /dev/null @@ -1,122 +0,0 @@ -#ifndef _FTAPE_HEADER_SEGMENT_H -#define _FTAPE_HEADER_SEGMENT_H - -/* - * Copyright (C) 1996-1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/include/linux/ftape-header-segment.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:28 $ - * - * This file defines some offsets into the header segment of a - * floppy tape cartridge. For use with the QIC-40/80/3010/3020 - * floppy-tape driver "ftape" for Linux. - */ - -#define FT_SIGNATURE 0 /* must be 0xaa55aa55 */ -#define FT_FMT_CODE 4 -#define FT_REV_LEVEL 5 /* only for QIC-80 since. Rev. L (== 0x0c) */ -#define FT_HSEG_1 6 /* first header segment, except for format code 6 */ -#define FT_HSEG_2 8 /* second header segment, except for format code 6 */ -#define FT_FRST_SEG 10 /* first data segment, except for format code 6 */ -#define FT_LAST_SEG 12 /* last data segment, except for format code 6 */ -#define FT_FMT_DATE 14 /* date and time of most recent format, see below */ -#define FT_WR_DATE 18 /* date and time of most recent write or format */ -#define FT_SPT 24 /* segments per track */ -#define FT_TPC 26 /* tracks per cartridge */ -#define FT_FHM 27 /* floppy drive head (maximum of it) */ -#define FT_FTM 28 /* floppy track max. */ -#define FT_FSM 29 /* floppy sector max. (128) */ -#define FT_LABEL 30 /* floppy tape label */ -#define FT_LABEL_DATE 74 /* date and time the tape label was written */ -#define FT_LABEL_SZ (FT_LABEL_DATE - FT_LABEL) -#define FT_CMAP_START 78 /* starting segment of compression map */ -#define FT_FMT_ERROR 128 /* must be set to 0xff if remainder gets lost during - * tape format - */ -#define FT_SEG_CNT 130 /* number of seg. written, formatted or verified - * through lifetime of tape (why not read?) - */ -#define FT_INIT_DATE 138 /* date and time of initial tape format */ -#define FT_FMT_CNT 142 /* number of times tape has been formatted */ -#define FT_FSL_CNT 144 /* number of segments in failed sector log */ -#define FT_MK_CODE 146 /* id string of tape manufacturer */ -#define FT_LOT_CODE 190 /* tape manufacturer lot code */ -#define FT_6_HSEG_1 234 /* first header segment for format code 6 */ -#define FT_6_HSEG_2 238 /* second header segment for format code 6 */ -#define FT_6_FRST_SEG 242 /* first data segment for format code 6 */ -#define FT_6_LAST_SEG 246 /* last data segment for format code 6 */ - -#define FT_FSL 256 -#define FT_HEADER_END 256 /* space beyond this point: - * format codes 2, 3 and 5: - * - failed sector log until byte 2047 - * - bad sector map in the reamining part of segment - * format codes 4 and 6: - * - bad sector map starts hear - */ - - -/* value to be stored at the FT_SIGNATURE offset - */ -#define FT_HSEG_MAGIC 0xaa55aa55 -#define FT_D2G_MAGIC 0x82288228 /* Ditto 2GB */ - -/* data and time encoding: */ -#define FT_YEAR_SHIFT 25 -#define FT_YEAR_MASK 0xfe000000 -#define FT_YEAR_0 1970 -#define FT_YEAR_MAX 127 -#define FT_YEAR(year) ((((year)-FT_YEAR_0)< -#include -#endif -#include -#include - -#define FT_SECTOR(x) (x+1) /* sector offset into real sector */ -#define FT_SECTOR_SIZE 1024 -#define FT_SECTORS_PER_SEGMENT 32 -#define FT_ECC_SECTORS 3 -#define FT_SEGMENT_SIZE ((FT_SECTORS_PER_SEGMENT - FT_ECC_SECTORS) * FT_SECTOR_SIZE) -#define FT_BUFF_SIZE (FT_SECTORS_PER_SEGMENT * FT_SECTOR_SIZE) - -/* - * bits of the minor device number that define drive selection - * methods. Could be used one day to access multiple tape - * drives on the same controller. - */ -#define FTAPE_SEL_A 0 -#define FTAPE_SEL_B 1 -#define FTAPE_SEL_C 2 -#define FTAPE_SEL_D 3 -#define FTAPE_SEL_MASK 3 -#define FTAPE_SEL(unit) ((unit) & FTAPE_SEL_MASK) -#define FTAPE_NO_REWIND 4 /* mask for minor nr */ - -/* the following two may be reported when MTIOCGET is requested ... */ -typedef union { - struct { - __u8 error; - __u8 command; - } error; - long space; -} ft_drive_error; -typedef union { - struct { - __u8 drive_status; - __u8 drive_config; - __u8 tape_status; - } status; - long space; -} ft_drive_status; - -#ifdef __KERNEL__ - -#define FT_RQM_DELAY 12 -#define FT_MILLISECOND 1 -#define FT_SECOND 1000 -#define FT_FOREVER -1 -#ifndef HZ -#error "HZ undefined." -#endif -#define FT_USPT (1000000/HZ) /* microseconds per tick */ - -/* This defines the number of retries that the driver will allow - * before giving up (and letting a higher level handle the error). - */ -#ifdef TESTING -#define FT_SOFT_RETRIES 1 /* number of low level retries */ -#define FT_RETRIES_ON_ECC_ERROR 3 /* ecc error when correcting segment */ -#else -#define FT_SOFT_RETRIES 6 /* number of low level retries (triple) */ -#define FT_RETRIES_ON_ECC_ERROR 3 /* ecc error when correcting segment */ -#endif - -#ifndef THE_FTAPE_MAINTAINER -#define THE_FTAPE_MAINTAINER "the ftape maintainer" -#endif - -/* Initialize missing configuration parameters. - */ -#ifndef CONFIG_FT_NR_BUFFERS -# define CONFIG_FT_NR_BUFFERS 3 -#endif -#ifndef CONFIG_FT_FDC_THR -# define CONFIG_FT_FDC_THR 8 -#endif -#ifndef CONFIG_FT_FDC_MAX_RATE -# define CONFIG_FT_FDC_MAX_RATE 2000 -#endif -#ifndef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0 -#endif -#ifndef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 0 -#endif -#ifndef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 0 -#endif - -/* Turn some booleans into numbers. - */ -#ifdef CONFIG_FT_PROBE_FC10 -# undef CONFIG_FT_PROBE_FC10 -# define CONFIG_FT_PROBE_FC10 1 -#else -# define CONFIG_FT_PROBE_FC10 0 -#endif -#ifdef CONFIG_FT_MACH2 -# undef CONFIG_FT_MACH2 -# define CONFIG_FT_MACH2 1 -#else -# define CONFIG_FT_MACH2 0 -#endif - -/* Insert default settings - */ -#if CONFIG_FT_PROBE_FC10 == 1 -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x180 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 9 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 3 -# endif -#elif CONFIG_FT_MACH2 == 1 /* CONFIG_FT_PROBE_FC10 == 1 */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x1E0 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#elif defined(CONFIG_FT_ALT_FDC) /* CONFIG_FT_MACH2 */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x370 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#else /* CONFIG_FT_ALT_FDC */ -# if CONFIG_FT_FDC_BASE == 0 -# undef CONFIG_FT_FDC_BASE -# define CONFIG_FT_FDC_BASE 0x3f0 -# endif -# if CONFIG_FT_FDC_IRQ == 0 -# undef CONFIG_FT_FDC_IRQ -# define CONFIG_FT_FDC_IRQ 6 -# endif -# if CONFIG_FT_FDC_DMA == 0 -# undef CONFIG_FT_FDC_DMA -# define CONFIG_FT_FDC_DMA 2 -# endif -#endif /* standard FDC */ - -/* some useful macro's - */ -#define NR_ITEMS(x) (int)(sizeof(x)/ sizeof(*x)) - -#endif /* __KERNEL__ */ - -#endif diff --git a/include/linux/zftape.h b/include/linux/zftape.h deleted file mode 100644 index b057c65366c6..000000000000 --- a/include/linux/zftape.h +++ /dev/null @@ -1,87 +0,0 @@ -#ifndef _ZFTAPE_H -#define _ZFTAPE_H - -/* - * Copyright (C) 1996, 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/include/linux/zftape.h,v $ - * $Revision: 1.12 $ - * $Date: 1997/10/21 11:02:37 $ - * - * Special ioctl and other global info for the zftape VFS - * interface for the QIC-40/80/3010/3020 floppy-tape driver for - * Linux. - */ - -#define ZFTAPE_VERSION "zftape for " FTAPE_VERSION - -#include - -#define ZFTAPE_LABEL "Ftape - The Linux Floppy Tape Project!" - -/* Bits of the minor device number that control the operation mode */ -#define ZFT_Q80_MODE (1 << 3) -#define ZFT_ZIP_MODE (1 << 4) -#define ZFT_RAW_MODE (1 << 5) -#define ZFT_MINOR_OP_MASK (ZFT_Q80_MODE | \ - ZFT_ZIP_MODE | \ - ZFT_RAW_MODE) -#define ZFT_MINOR_MASK (FTAPE_SEL_MASK | \ - ZFT_MINOR_OP_MASK | \ - FTAPE_NO_REWIND) - -#ifdef ZFT_OBSOLETE -struct mtblksz { - unsigned int mt_blksz; -}; -#define MTIOC_ZFTAPE_GETBLKSZ _IOR('m', 104, struct mtblksz) -#endif - -#ifdef __KERNEL__ - -extern int zft_init(void); - -static inline __s64 zft_div_blksz(__s64 value, __u32 blk_sz) -{ - if (blk_sz == 1) { - return value; - } else { - return (__s64)(((__u32)(value >> 10) + (blk_sz >> 10) - 1) - / (blk_sz >> 10)); - } -} - -static inline __s64 zft_mul_blksz(__s64 value, __u32 blk_sz) -{ - if (blk_sz == 1) { - return value; - } else { - /* if blk_sz != 1, then it is a multiple of 1024. In - * this case, `value' will also fit into 32 bits. - * - * Actually, this limits the capacity to 42 - * bits. This is (2^32)*1024, roughly a thousand - * times 2GB, or 3 Terabytes. Hopefully this is enough - */ - return(__s64)(((__u32)(value)*(blk_sz>>10))<<10); - } -} - -#endif - -#endif -- cgit v1.2.3 From f6a570333e554b48ad589e7137c77c57809eee81 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Oct 2006 01:47:25 -0400 Subject: [PATCH] severing module.h->sched.h Signed-off-by: Al Viro --- arch/i386/kernel/alternative.c | 1 + arch/i386/kernel/cpu/mcheck/therm_throt.c | 1 + drivers/base/cpu.c | 1 + drivers/hwmon/abituguru.c | 1 + drivers/leds/ledtrig-ide-disk.c | 1 + drivers/leds/ledtrig-timer.c | 1 + drivers/pci/access.c | 1 + drivers/scsi/scsi_transport_sas.c | 1 + drivers/w1/slaves/w1_therm.c | 1 + include/asm-x86_64/elf.h | 1 - include/linux/acct.h | 1 + include/linux/module.h | 13 +------------ include/scsi/libiscsi.h | 2 ++ kernel/latency.c | 1 + kernel/module.c | 15 ++++++++++++++- lib/random32.c | 1 + 16 files changed, 29 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c index 583c238e17fb..535f9794fba1 100644 --- a/arch/i386/kernel/alternative.c +++ b/arch/i386/kernel/alternative.c @@ -1,4 +1,5 @@ #include +#include #include #include #include diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c index 2d8703b7ce65..bad8b4420709 100644 --- a/arch/i386/kernel/cpu/mcheck/therm_throt.c +++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c @@ -20,6 +20,7 @@ #include #include #include +#include #include /* How long to wait between reporting thermal events */ diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 4bef76a2f3f2..1f745f12f94e 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/hwmon/abituguru.c b/drivers/hwmon/abituguru.c index e5cb0fdab9b1..b1dc63e4ac7b 100644 --- a/drivers/hwmon/abituguru.c +++ b/drivers/hwmon/abituguru.c @@ -21,6 +21,7 @@ etc voltage & frequency control is not supported! */ #include +#include #include #include #include diff --git a/drivers/leds/ledtrig-ide-disk.c b/drivers/leds/ledtrig-ide-disk.c index fa651886ab4f..54b155c7026f 100644 --- a/drivers/leds/ledtrig-ide-disk.c +++ b/drivers/leds/ledtrig-ide-disk.c @@ -12,6 +12,7 @@ */ #include +#include #include #include #include diff --git a/drivers/leds/ledtrig-timer.c b/drivers/leds/ledtrig-timer.c index 29a8818a32ec..d756bdb01c59 100644 --- a/drivers/leds/ledtrig-timer.c +++ b/drivers/leds/ledtrig-timer.c @@ -12,6 +12,7 @@ */ #include +#include #include #include #include diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 73a58c73d526..fc405f0165d9 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -1,5 +1,6 @@ #include #include +#include #include #include diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c index b5b0c2cba96b..5c0b75bbfa10 100644 --- a/drivers/scsi/scsi_transport_sas.c +++ b/drivers/scsi/scsi_transport_sas.c @@ -25,6 +25,7 @@ #include #include +#include #include #include #include diff --git a/drivers/w1/slaves/w1_therm.c b/drivers/w1/slaves/w1_therm.c index 5372cfcbd054..b022fffd8c51 100644 --- a/drivers/w1/slaves/w1_therm.c +++ b/drivers/w1/slaves/w1_therm.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include diff --git a/include/asm-x86_64/elf.h b/include/asm-x86_64/elf.h index a406fcb1e924..6d24ea7c4d9d 100644 --- a/include/asm-x86_64/elf.h +++ b/include/asm-x86_64/elf.h @@ -45,7 +45,6 @@ typedef struct user_i387_struct elf_fpregset_t; #ifdef __KERNEL__ #include -#include /* * This is used to ensure we don't load something for the wrong architecture. diff --git a/include/linux/acct.h b/include/linux/acct.h index 0496d1f09952..302eb727ecb8 100644 --- a/include/linux/acct.h +++ b/include/linux/acct.h @@ -119,6 +119,7 @@ struct acct_v3 #ifdef CONFIG_BSD_PROCESS_ACCT struct vfsmount; struct super_block; +struct pacct_struct; extern void acct_auto_close_mnt(struct vfsmount *m); extern void acct_auto_close(struct super_block *sb); extern void acct_init_pacct(struct pacct_struct *pacct); diff --git a/include/linux/module.h b/include/linux/module.h index 9258ffd8a7f0..d33df2408e05 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -6,7 +6,6 @@ * Rewritten by Richard Henderson Dec 1996 * Rewritten again by Rusty Russell, 2002 */ -#include #include #include #include @@ -411,17 +410,7 @@ static inline int try_module_get(struct module *module) return ret; } -static inline void module_put(struct module *module) -{ - if (module) { - unsigned int cpu = get_cpu(); - local_dec(&module->ref[cpu].count); - /* Maybe they're waiting for us to drop reference? */ - if (unlikely(!module_is_live(module))) - wake_up_process(module->waiter); - put_cpu(); - } -} +extern void module_put(struct module *module); #else /*!CONFIG_MODULE_UNLOAD*/ static inline int try_module_get(struct module *module) diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h index 61eebec00a7b..ea0816d4904d 100644 --- a/include/scsi/libiscsi.h +++ b/include/scsi/libiscsi.h @@ -25,6 +25,8 @@ #include #include +#include +#include #include #include diff --git a/kernel/latency.c b/kernel/latency.c index 258f2555abbc..e63fcacb61a7 100644 --- a/kernel/latency.c +++ b/kernel/latency.c @@ -36,6 +36,7 @@ #include #include #include +#include #include struct latency_info { diff --git a/kernel/module.c b/kernel/module.c index 45e01cb60101..e2d09d604ca0 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -34,10 +34,10 @@ #include #include #include +#include #include #include #include -#include #include #include #include @@ -790,6 +790,19 @@ static struct module_attribute refcnt = { .show = show_refcnt, }; +void module_put(struct module *module) +{ + if (module) { + unsigned int cpu = get_cpu(); + local_dec(&module->ref[cpu].count); + /* Maybe they're waiting for us to drop reference? */ + if (unlikely(!module_is_live(module))) + wake_up_process(module->waiter); + put_cpu(); + } +} +EXPORT_SYMBOL(module_put); + #else /* !CONFIG_MODULE_UNLOAD */ static void print_unload_info(struct seq_file *m, struct module *mod) { diff --git a/lib/random32.c b/lib/random32.c index 4a15ce51cea7..ec7f81d3fb18 100644 --- a/lib/random32.c +++ b/lib/random32.c @@ -36,6 +36,7 @@ #include #include #include +#include #include struct rnd_state { -- cgit v1.2.3 From 914e26379decf1fd984b22e51fd2e4209b7a7f1b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 18 Oct 2006 13:55:46 -0400 Subject: [PATCH] severing fs.h, radix-tree.h -> sched.h Signed-off-by: Al Viro --- arch/i386/kernel/acpi/cstate.c | 1 + drivers/acpi/dock.c | 1 + drivers/char/hw_random/core.c | 1 + drivers/char/tpm/tpm.h | 1 + drivers/hwmon/hdaps.c | 1 + drivers/spi/spi_butterfly.c | 1 + fs/9p/conv.c | 1 + fs/9p/fcall.c | 1 + fs/9p/fid.c | 1 + fs/9p/v9fs.c | 1 + fs/9p/vfs_dir.c | 1 + fs/9p/vfs_file.c | 1 + fs/inotify.c | 1 + fs/jffs2/acl.c | 1 + fs/jffs2/wbuf.c | 1 + fs/jfs/ioctl.c | 1 + fs/proc/root.c | 1 + fs/super.c | 18 ++++++++++++++++++ fs/sync.c | 1 + fs/utimes.c | 1 + include/linux/fs.h | 37 +++++++------------------------------ include/linux/radix-tree.h | 1 - 22 files changed, 44 insertions(+), 31 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/acpi/cstate.c b/arch/i386/kernel/acpi/cstate.c index 20563e52c622..4664b55f623e 100644 --- a/arch/i386/kernel/acpi/cstate.c +++ b/arch/i386/kernel/acpi/cstate.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 578b99b71d9c..bf5b79ed3613 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index ebace201bec6..26a860adcb38 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 050ced247f68..bb9a43c6cf3d 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/hwmon/hdaps.c b/drivers/hwmon/hdaps.c index 26be4ea8a38a..e8ef62b83d6b 100644 --- a/drivers/hwmon/hdaps.c +++ b/drivers/hwmon/hdaps.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #define HDAPS_LOW_PORT 0x1600 /* first port used by hdaps */ diff --git a/drivers/spi/spi_butterfly.c b/drivers/spi/spi_butterfly.c index 39d9b20f2038..c2f601f8e4f2 100644 --- a/drivers/spi/spi_butterfly.c +++ b/drivers/spi/spi_butterfly.c @@ -23,6 +23,7 @@ #include #include +#include #include #include #include diff --git a/fs/9p/conv.c b/fs/9p/conv.c index 56d88c1a09c5..a3ed571eee31 100644 --- a/fs/9p/conv.c +++ b/fs/9p/conv.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include "debug.h" diff --git a/fs/9p/fcall.c b/fs/9p/fcall.c index 8556097fcda8..dc336a67592f 100644 --- a/fs/9p/fcall.c +++ b/fs/9p/fcall.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include "debug.h" diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 70492ccb4385..27507201f9e7 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include "debug.h" diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 0f628041e3f7..0b96fae8b479 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index e32d5971039b..905c882f4e2f 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index c3c47eda7574..79e6f9cd7340 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/inotify.c b/fs/inotify.c index 723836a1f718..f5099d86fd91 100644 --- a/fs/inotify.c +++ b/fs/inotify.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 0ae3cd10702c..73f0d60f73a5 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index b9b700730dfe..70707309dfa1 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -19,6 +19,7 @@ #include #include #include +#include #include "nodelist.h" diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c index 37db52488262..ed814b1ff4d9 100644 --- a/fs/jfs/ioctl.c +++ b/fs/jfs/ioctl.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include diff --git a/fs/proc/root.c b/fs/proc/root.c index ffe66c38488b..64d242b6dcfa 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/super.c b/fs/super.c index 47e554c12e76..84c320f6ad7e 100644 --- a/fs/super.c +++ b/fs/super.c @@ -220,6 +220,24 @@ static int grab_super(struct super_block *s) __releases(sb_lock) return 0; } +/* + * Superblock locking. We really ought to get rid of these two. + */ +void lock_super(struct super_block * sb) +{ + get_fs_excl(); + mutex_lock(&sb->s_lock); +} + +void unlock_super(struct super_block * sb) +{ + put_fs_excl(); + mutex_unlock(&sb->s_lock); +} + +EXPORT_SYMBOL(lock_super); +EXPORT_SYMBOL(unlock_super); + /* * Write out and wait upon all dirty data associated with this * superblock. Filesystem data as well as the underlying block diff --git a/fs/sync.c b/fs/sync.c index 1de747b5ddb9..865f32be386e 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/utimes.c b/fs/utimes.c index 1bcd852fc4a9..99cf2cb11fec 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include diff --git a/include/linux/fs.h b/include/linux/fs.h index 2fe6e3f900ba..cac7b1ef9543 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -276,7 +276,7 @@ extern int dir_notify_enable; #include #include #include -#include +#include #include #include @@ -977,36 +977,13 @@ enum { #define vfs_check_frozen(sb, level) \ wait_event((sb)->s_wait_unfrozen, ((sb)->s_frozen < (level))) -static inline void get_fs_excl(void) -{ - atomic_inc(¤t->fs_excl); -} - -static inline void put_fs_excl(void) -{ - atomic_dec(¤t->fs_excl); -} - -static inline int has_fs_excl(void) -{ - return atomic_read(¤t->fs_excl); -} +#define get_fs_excl() atomic_inc(¤t->fs_excl) +#define put_fs_excl() atomic_dec(¤t->fs_excl) +#define has_fs_excl() atomic_read(¤t->fs_excl) - -/* - * Superblock locking. - */ -static inline void lock_super(struct super_block * sb) -{ - get_fs_excl(); - mutex_lock(&sb->s_lock); -} - -static inline void unlock_super(struct super_block * sb) -{ - put_fs_excl(); - mutex_unlock(&sb->s_lock); -} +/* not quite ready to be deprecated, but... */ +extern void lock_super(struct super_block *); +extern void unlock_super(struct super_block *); /* * VFS helper functions.. diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index 9158a68140c9..cbfa11537421 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -19,7 +19,6 @@ #ifndef _LINUX_RADIX_TREE_H #define _LINUX_RADIX_TREE_H -#include #include #include -- cgit v1.2.3 From a1f8e7f7fb9d7e2cbcb53170edca7c0ac4680697 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 16:08:53 -0400 Subject: [PATCH] severing skbuff.h -> highmem.h Signed-off-by: Al Viro --- drivers/infiniband/ulp/iser/iser_memory.c | 1 + drivers/net/sun3lance.c | 1 + fs/compat.c | 1 + include/linux/skbuff.h | 19 ------------------- kernel/auditsc.c | 1 + net/appletalk/ddp.c | 1 + net/core/kmap_skb.h | 19 +++++++++++++++++++ net/core/skbuff.c | 3 ++- net/core/sock.c | 1 + net/ipv4/ip_output.c | 1 + net/packet/af_packet.c | 1 + 11 files changed, 29 insertions(+), 20 deletions(-) create mode 100644 net/core/kmap_skb.h (limited to 'include/linux') diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index 0606744c3f84..5e122501fd80 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/sun3lance.c b/drivers/net/sun3lance.c index b865db363ba0..47a1c09d19ac 100644 --- a/drivers/net/sun3lance.c +++ b/drivers/net/sun3lance.c @@ -38,6 +38,7 @@ static char *version = "sun3lance.c: v1.2 1/12/2001 Sam Creasey (sammy@sammy.ne #include #include +#include #include #include #include diff --git a/fs/compat.c b/fs/compat.c index 8d0a0018a7d2..fde52d40c0b6 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -45,6 +45,7 @@ #include #include #include +#include #include #include /* siocdevprivate_ioctl */ diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 14ec16d2d9ba..24ce0add6c54 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -23,7 +23,6 @@ #include #include #include -#include #include #include #include @@ -1295,24 +1294,6 @@ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len) return __pskb_trim(skb, len); } -static inline void *kmap_skb_frag(const skb_frag_t *frag) -{ -#ifdef CONFIG_HIGHMEM - BUG_ON(in_irq()); - - local_bh_disable(); -#endif - return kmap_atomic(frag->page, KM_SKB_DATA_SOFTIRQ); -} - -static inline void kunmap_skb_frag(void *vaddr) -{ - kunmap_atomic(vaddr, KM_SKB_DATA_SOFTIRQ); -#ifdef CONFIG_HIGHMEM - local_bh_enable(); -#endif -} - #define skb_queue_walk(queue, skb) \ for (skb = (queue)->next; \ prefetch(skb->next), (skb != (struct sk_buff *)(queue)); \ diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 42f2f1179711..ab97e5101232 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -64,6 +64,7 @@ #include #include #include +#include #include #include "audit.h" diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 485e35c3b28b..3a7052207708 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -61,6 +61,7 @@ #include #include #include +#include "../core/kmap_skb.h" struct datalink_proto *ddp_dl, *aarp_dl; static const struct proto_ops atalk_dgram_ops; diff --git a/net/core/kmap_skb.h b/net/core/kmap_skb.h new file mode 100644 index 000000000000..283c2b993fb8 --- /dev/null +++ b/net/core/kmap_skb.h @@ -0,0 +1,19 @@ +#include + +static inline void *kmap_skb_frag(const skb_frag_t *frag) +{ +#ifdef CONFIG_HIGHMEM + BUG_ON(in_irq()); + + local_bh_disable(); +#endif + return kmap_atomic(frag->page, KM_SKB_DATA_SOFTIRQ); +} + +static inline void kunmap_skb_frag(void *vaddr) +{ + kunmap_atomic(vaddr, KM_SKB_DATA_SOFTIRQ); +#ifdef CONFIG_HIGHMEM + local_bh_enable(); +#endif +} diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a90bc439488e..8e1c385e5ba9 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -56,7 +56,6 @@ #include #include #include -#include #include #include @@ -67,6 +66,8 @@ #include #include +#include "kmap_skb.h" + static kmem_cache_t *skbuff_head_cache __read_mostly; static kmem_cache_t *skbuff_fclone_cache __read_mostly; diff --git a/net/core/sock.c b/net/core/sock.c index ab8fafadb4ba..419c7d3289c7 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -111,6 +111,7 @@ #include #include #include +#include #include #include diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 1da3d32f8289..a35209d517ad 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -53,6 +53,7 @@ #include #include #include +#include #include #include diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 271d2eed0699..08e68b67bbf6 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -71,6 +71,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From bd01f843c3368dcee735c19603251669f23f4477 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Oct 2006 17:23:57 -0400 Subject: [PATCH] severing skbuff.h -> poll.h Signed-off-by: Al Viro --- fs/compat.c | 1 + fs/gfs2/locking/dlm/plock.c | 1 + include/linux/skbuff.h | 1 - include/net/inet_connection_sock.h | 1 + include/net/udp.h | 1 + 5 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/compat.c b/fs/compat.c index fde52d40c0b6..06dad665b88f 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include /* siocdevprivate_ioctl */ diff --git a/fs/gfs2/locking/dlm/plock.c b/fs/gfs2/locking/dlm/plock.c index 7365aec9511b..3799f19b282f 100644 --- a/fs/gfs2/locking/dlm/plock.c +++ b/fs/gfs2/locking/dlm/plock.c @@ -8,6 +8,7 @@ #include #include +#include #include "lock_dlm.h" diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 24ce0add6c54..5c5a08576dcc 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -23,7 +23,6 @@ #include #include #include -#include #include #include #include diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index cccea051e922..bf16d98d372c 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -18,6 +18,7 @@ #include #include #include +#include #include #include diff --git a/include/net/udp.h b/include/net/udp.h index 1548d68d45da..1b921fa81474 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -29,6 +29,7 @@ #include #include #include +#include /** * struct udp_skb_cb - UDP(-Lite) private variables -- cgit v1.2.3 From d7fe0f241dceade9c8d4af75498765c5ff7f27e6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 3 Dec 2006 23:15:30 -0500 Subject: [PATCH] severing skbuff.h -> mm.h Signed-off-by: Al Viro --- drivers/isdn/divert/isdn_divert.c | 2 ++ drivers/net/ehea/ehea_qmr.c | 1 + drivers/net/lance.c | 1 + drivers/net/ne3210.c | 1 + drivers/net/sk98lin/skge.c | 1 + drivers/net/starfire.c | 1 + drivers/net/sungem.c | 1 + drivers/net/sunhme.c | 1 + drivers/net/typhoon.c | 1 + include/linux/igmp.h | 1 + include/linux/kernelcapi.h | 1 + include/linux/netdevice.h | 1 + include/linux/netfilter_ipv4/ip_conntrack.h | 1 + include/linux/skbuff.h | 1 - include/net/irda/timer.h | 1 + include/net/netfilter/nf_conntrack.h | 1 + include/net/netlink.h | 1 + include/net/sock.h | 1 + net/ieee80211/ieee80211_crypt_tkip.c | 1 + net/ieee80211/ieee80211_crypt_wep.c | 1 + net/ipv4/ipvs/ip_vs_lblc.c | 1 + net/ipv4/ipvs/ip_vs_lblcr.c | 1 + net/irda/discovery.c | 1 + net/irda/iriap.c | 1 + net/irda/irttp.c | 1 + net/netfilter/nf_conntrack_core.c | 1 + net/netfilter/x_tables.c | 1 + net/netfilter/xt_hashlimit.c | 1 + 28 files changed, 28 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/isdn/divert/isdn_divert.c b/drivers/isdn/divert/isdn_divert.c index 1f5ebe9ee72c..03319ea5aa0c 100644 --- a/drivers/isdn/divert/isdn_divert.c +++ b/drivers/isdn/divert/isdn_divert.c @@ -10,6 +10,8 @@ */ #include +#include +#include #include "isdn_divert.h" diff --git a/drivers/net/ehea/ehea_qmr.c b/drivers/net/ehea/ehea_qmr.c index 72ef7bde3346..f143e13b229d 100644 --- a/drivers/net/ehea/ehea_qmr.c +++ b/drivers/net/ehea/ehea_qmr.c @@ -26,6 +26,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ +#include #include "ehea.h" #include "ehea_phyp.h" #include "ehea_qmr.h" diff --git a/drivers/net/lance.c b/drivers/net/lance.c index 6efbd499d752..4256c13c73c2 100644 --- a/drivers/net/lance.c +++ b/drivers/net/lance.c @@ -57,6 +57,7 @@ static const char version[] = "lance.c:v1.16 2006/11/09 dplatt@3do.com, becker@c #include #include #include +#include #include #include diff --git a/drivers/net/ne3210.c b/drivers/net/ne3210.c index d66328975425..1a6fed76d4cc 100644 --- a/drivers/net/ne3210.c +++ b/drivers/net/ne3210.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index 12cbfd190dd7..92d11b961db8 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -114,6 +114,7 @@ #include #include #include +#include #include "h/skdrv1st.h" #include "h/skdrv2nd.h" diff --git a/drivers/net/starfire.c b/drivers/net/starfire.c index 7a0aee6c869d..bf873ea25797 100644 --- a/drivers/net/starfire.c +++ b/drivers/net/starfire.c @@ -41,6 +41,7 @@ #include #include #include +#include #include /* Processor type for cache alignment. */ #include #include diff --git a/drivers/net/sungem.c b/drivers/net/sungem.c index 334c6cfd6595..cf44e72399b9 100644 --- a/drivers/net/sungem.c +++ b/drivers/net/sungem.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index ec432ea879fb..4d59ece67292 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c index 8ddea1da7c05..9781b16bb8b6 100644 --- a/drivers/net/typhoon.c +++ b/drivers/net/typhoon.c @@ -117,6 +117,7 @@ static const int multicast_filter_limit = 32; #include #include #include +#include #include #include #include diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 6e7ea2f0a57c..9dbb525c5178 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -127,6 +127,7 @@ struct igmpv3_query { #ifdef __KERNEL__ #include +#include #include extern int sysctl_igmp_max_memberships; diff --git a/include/linux/kernelcapi.h b/include/linux/kernelcapi.h index 891bb2cf0aa8..f8a0ff86dacc 100644 --- a/include/linux/kernelcapi.h +++ b/include/linux/kernelcapi.h @@ -47,6 +47,7 @@ typedef struct kcapi_carddef { #include #include +#include #define KCI_CONTRUP 0 /* arg: struct capi_profile */ #define KCI_CONTRDOWN 1 /* arg: NULL */ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 949eada46ce1..c57088f575a3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -30,6 +30,7 @@ #include #ifdef __KERNEL__ +#include #include #include #include diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 61da56941dce..33581c13d947 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -9,6 +9,7 @@ #include #include +#include #include #include #include diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 5c5a08576dcc..6bdff9b148d0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -22,7 +22,6 @@ #include #include #include -#include #include #include #include diff --git a/include/net/irda/timer.h b/include/net/irda/timer.h index 2c5d8864ab77..cb61568547d1 100644 --- a/include/net/irda/timer.h +++ b/include/net/irda/timer.h @@ -28,6 +28,7 @@ #define TIMER_H #include +#include #include /* for HZ */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 032b36a0e378..bd01b4633ee2 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -56,6 +56,7 @@ union nf_conntrack_help { #include #include +#include #ifdef CONFIG_NETFILTER_DEBUG #define NF_CT_ASSERT(x) \ diff --git a/include/net/netlink.h b/include/net/netlink.h index fd75fd65d59e..bcaf67b7a19d 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -3,6 +3,7 @@ #include #include +#include /* ======================================================================== * Netlink Messages and Attributes Interface (As Seen On TV) diff --git a/include/net/sock.h b/include/net/sock.h index 26fc0b16bc0c..fe3a33fad03f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -47,6 +47,7 @@ #include #include #include /* struct sk_buff */ +#include #include #include diff --git a/net/ieee80211/ieee80211_crypt_tkip.c b/net/ieee80211/ieee80211_crypt_tkip.c index 4200ec509866..fc1f99a59732 100644 --- a/net/ieee80211/ieee80211_crypt_tkip.c +++ b/net/ieee80211/ieee80211_crypt_tkip.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include diff --git a/net/ieee80211/ieee80211_crypt_wep.c b/net/ieee80211/ieee80211_crypt_wep.c index 1b2efff11d39..7a95c3d81314 100644 --- a/net/ieee80211/ieee80211_crypt_wep.c +++ b/net/ieee80211/ieee80211_crypt_wep.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include diff --git a/net/ipv4/ipvs/ip_vs_lblc.c b/net/ipv4/ipvs/ip_vs_lblc.c index 524751e031de..a4385a2180ee 100644 --- a/net/ipv4/ipvs/ip_vs_lblc.c +++ b/net/ipv4/ipvs/ip_vs_lblc.c @@ -45,6 +45,7 @@ #include #include #include +#include /* for sysctl */ #include diff --git a/net/ipv4/ipvs/ip_vs_lblcr.c b/net/ipv4/ipvs/ip_vs_lblcr.c index 08990192b6ec..fe1af5d079af 100644 --- a/net/ipv4/ipvs/ip_vs_lblcr.c +++ b/net/ipv4/ipvs/ip_vs_lblcr.c @@ -43,6 +43,7 @@ #include #include #include +#include /* for sysctl */ #include diff --git a/net/irda/discovery.c b/net/irda/discovery.c index 3fefc822c1c0..89fd2a2cbca6 100644 --- a/net/irda/discovery.c +++ b/net/irda/discovery.c @@ -32,6 +32,7 @@ #include #include +#include #include #include diff --git a/net/irda/iriap.c b/net/irda/iriap.c index 8cfd076c4c12..8f1c6d65b247 100644 --- a/net/irda/iriap.c +++ b/net/irda/iriap.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include diff --git a/net/irda/irttp.c b/net/irda/irttp.c index 9c446a72ff1f..252f11012566 100644 --- a/net/irda/irttp.c +++ b/net/irda/irttp.c @@ -26,6 +26,7 @@ #include #include +#include #include #include diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 93d97d9f9da8..eaa0f8a1adb6 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 58522fc65d33..8996584b8499 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index 501c564e247f..a98de0b54d65 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From f23f6e08c47acbdd20e9c49a79da8c404ea168e1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 20 Oct 2006 15:17:02 -0400 Subject: [PATCH] severing poll.h -> mm.h Signed-off-by: Al Viro --- drivers/char/hpet.c | 1 + drivers/media/dvb/cinergyT2/cinergyT2.c | 1 + include/linux/poll.h | 3 ++- include/sound/pcm.h | 1 + sound/oss/cs46xx.c | 1 + sound/oss/dmabuf.c | 1 + sound/oss/emu10k1/audio.c | 1 + sound/oss/es1371.c | 1 + sound/oss/i810_audio.c | 1 + sound/oss/soundcard.c | 1 + sound/oss/sscape.c | 1 + sound/oss/trident.c | 1 + 12 files changed, 13 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 091a11cd878c..20dc3be5ecfc 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index 55bc891768c2..8a7dd507cf6e 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "dmxdev.h" #include "dvb_demux.h" diff --git a/include/linux/poll.h b/include/linux/poll.h index 51e1b56741fb..27690798623f 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -8,7 +8,8 @@ #include #include #include -#include +#include +#include #include /* ~832 bytes of stack space used max in sys_select/sys_poll before allocating diff --git a/include/sound/pcm.h b/include/sound/pcm.h index afaf3e88e086..2f645dfd7f70 100644 --- a/include/sound/pcm.h +++ b/include/sound/pcm.h @@ -26,6 +26,7 @@ #include #include #include +#include #include #define snd_pcm_substream_chip(substream) ((substream)->private_data) diff --git a/sound/oss/cs46xx.c b/sound/oss/cs46xx.c index b1c5d8286e40..147c8a951137 100644 --- a/sound/oss/cs46xx.c +++ b/sound/oss/cs46xx.c @@ -91,6 +91,7 @@ #include #include #include +#include #include #include diff --git a/sound/oss/dmabuf.c b/sound/oss/dmabuf.c index b256c0401161..eaf69971bf92 100644 --- a/sound/oss/dmabuf.c +++ b/sound/oss/dmabuf.c @@ -25,6 +25,7 @@ #define BE_CONSERVATIVE #define SAMPLE_ROUNDUP 0 +#include #include "sound_config.h" #define DMAP_FREE_ON_CLOSE 0 diff --git a/sound/oss/emu10k1/audio.c b/sound/oss/emu10k1/audio.c index cde4d59d5430..86dd23974e05 100644 --- a/sound/oss/emu10k1/audio.c +++ b/sound/oss/emu10k1/audio.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include "hwaccess.h" diff --git a/sound/oss/es1371.c b/sound/oss/es1371.c index ddf6b0a0bca5..cc282a0cd539 100644 --- a/sound/oss/es1371.c +++ b/sound/oss/es1371.c @@ -130,6 +130,7 @@ #include #include #include +#include #include #include diff --git a/sound/oss/i810_audio.c b/sound/oss/i810_audio.c index 240cc7939b69..c3c8a720d555 100644 --- a/sound/oss/i810_audio.c +++ b/sound/oss/i810_audio.c @@ -101,6 +101,7 @@ #include #include #include +#include #include diff --git a/sound/oss/soundcard.c b/sound/oss/soundcard.c index 75c5e745705f..8fb8e7f99556 100644 --- a/sound/oss/soundcard.c +++ b/sound/oss/soundcard.c @@ -42,6 +42,7 @@ #include #include #include +#include /* * This ought to be moved into include/asm/dma.h diff --git a/sound/oss/sscape.c b/sound/oss/sscape.c index 51f2fa615413..30c36d1f35d7 100644 --- a/sound/oss/sscape.c +++ b/sound/oss/sscape.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include "coproc.h" diff --git a/sound/oss/trident.c b/sound/oss/trident.c index 7a363a178afd..6b1f8c9cdcf8 100644 --- a/sound/oss/trident.c +++ b/sound/oss/trident.c @@ -216,6 +216,7 @@ #include #include #include +#include #include #include -- cgit v1.2.3 From bf1ab978be2318c5a564de9aa0f1a217b44170d4 Mon Sep 17 00:00:00 2001 From: Dwayne Grant McConnell Date: Thu, 23 Nov 2006 00:46:37 +0100 Subject: [POWERPC] coredump: Add SPU elf notes to coredump. This patch adds SPU elf notes to the coredump. It creates a separate note for each of /regs, /fpcr, /lslr, /decr, /decr_status, /mem, /signal1, /signal1_type, /signal2, /signal2_type, /event_mask, /event_status, /mbox_info, /ibox_info, /wbox_info, /dma_info, /proxydma_info, /object-id. A new macro, ARCH_HAVE_EXTRA_NOTES, was created for architectures to specify they have extra elf core notes. A new macro, ELF_CORE_EXTRA_NOTES_SIZE, was created so the size of the additional notes could be calculated and added to the notes phdr entry. A new macro, ELF_CORE_WRITE_EXTRA_NOTES, was created so the new notes would be written after the existing notes. The SPU coredump code resides in spufs. Stub functions are provided in the kernel which are hooked into the spufs code which does the actual work via register_arch_coredump_calls(). A new set of __spufs__read/get() functions was provided to allow the coredump code to read from the spufs files without having to lock the SPU context for each file read from. Cc: Signed-off-by: Dwayne Grant McConnell Signed-off-by: Arnd Bergmann --- arch/powerpc/platforms/cell/Makefile | 1 + arch/powerpc/platforms/cell/spu_coredump.c | 81 +++++++ arch/powerpc/platforms/cell/spufs/Makefile | 2 +- arch/powerpc/platforms/cell/spufs/coredump.c | 238 +++++++++++++++++++ arch/powerpc/platforms/cell/spufs/file.c | 327 ++++++++++++++++++++------- arch/powerpc/platforms/cell/spufs/inode.c | 7 + arch/powerpc/platforms/cell/spufs/spufs.h | 11 + fs/binfmt_elf.c | 8 + include/asm-powerpc/elf.h | 13 ++ include/asm-powerpc/spu.h | 10 + include/linux/elf.h | 7 + 11 files changed, 618 insertions(+), 87 deletions(-) create mode 100644 arch/powerpc/platforms/cell/spu_coredump.c create mode 100644 arch/powerpc/platforms/cell/spufs/coredump.c (limited to 'include/linux') diff --git a/arch/powerpc/platforms/cell/Makefile b/arch/powerpc/platforms/cell/Makefile index 62c011e80929..f90e8337796c 100644 --- a/arch/powerpc/platforms/cell/Makefile +++ b/arch/powerpc/platforms/cell/Makefile @@ -15,5 +15,6 @@ spufs-modular-$(CONFIG_SPU_FS) += spu_syscalls.o spu-priv1-$(CONFIG_PPC_CELL_NATIVE) += spu_priv1_mmio.o obj-$(CONFIG_SPU_BASE) += spu_callbacks.o spu_base.o \ + spu_coredump.o \ $(spufs-modular-m) \ $(spu-priv1-y) spufs/ diff --git a/arch/powerpc/platforms/cell/spu_coredump.c b/arch/powerpc/platforms/cell/spu_coredump.c new file mode 100644 index 000000000000..6915b418ee73 --- /dev/null +++ b/arch/powerpc/platforms/cell/spu_coredump.c @@ -0,0 +1,81 @@ +/* + * SPU core dump code + * + * (C) Copyright 2006 IBM Corp. + * + * Author: Dwayne Grant McConnell + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include +#include +#include + +#include + +static struct spu_coredump_calls spu_coredump_calls; +static DEFINE_MUTEX(spu_coredump_mutex); + +int arch_notes_size(void) +{ + long ret; + struct module *owner = spu_coredump_calls.owner; + + ret = -ENOSYS; + mutex_lock(&spu_coredump_mutex); + if (owner && try_module_get(owner)) { + ret = spu_coredump_calls.arch_notes_size(); + module_put(owner); + } + mutex_unlock(&spu_coredump_mutex); + return ret; +} + +void arch_write_notes(struct file *file) +{ + struct module *owner = spu_coredump_calls.owner; + + mutex_lock(&spu_coredump_mutex); + if (owner && try_module_get(owner)) { + spu_coredump_calls.arch_write_notes(file); + module_put(owner); + } + mutex_unlock(&spu_coredump_mutex); +} + +int register_arch_coredump_calls(struct spu_coredump_calls *calls) +{ + if (spu_coredump_calls.owner) + return -EBUSY; + + mutex_lock(&spu_coredump_mutex); + spu_coredump_calls.arch_notes_size = calls->arch_notes_size; + spu_coredump_calls.arch_write_notes = calls->arch_write_notes; + spu_coredump_calls.owner = calls->owner; + mutex_unlock(&spu_coredump_mutex); + return 0; +} +EXPORT_SYMBOL_GPL(register_arch_coredump_calls); + +void unregister_arch_coredump_calls(struct spu_coredump_calls *calls) +{ + BUG_ON(spu_coredump_calls.owner != calls->owner); + + mutex_lock(&spu_coredump_mutex); + spu_coredump_calls.owner = NULL; + mutex_unlock(&spu_coredump_mutex); +} +EXPORT_SYMBOL_GPL(unregister_arch_coredump_calls); diff --git a/arch/powerpc/platforms/cell/spufs/Makefile b/arch/powerpc/platforms/cell/spufs/Makefile index ecdfbb35f82e..472217d19faf 100644 --- a/arch/powerpc/platforms/cell/spufs/Makefile +++ b/arch/powerpc/platforms/cell/spufs/Makefile @@ -1,7 +1,7 @@ obj-y += switch.o obj-$(CONFIG_SPU_FS) += spufs.o -spufs-y += inode.o file.o context.o syscalls.o +spufs-y += inode.o file.o context.o syscalls.o coredump.o spufs-y += sched.o backing_ops.o hw_ops.o run.o gang.o # Rules to build switch.o with the help of SPU tool chain diff --git a/arch/powerpc/platforms/cell/spufs/coredump.c b/arch/powerpc/platforms/cell/spufs/coredump.c new file mode 100644 index 000000000000..26945c491f6b --- /dev/null +++ b/arch/powerpc/platforms/cell/spufs/coredump.c @@ -0,0 +1,238 @@ +/* + * SPU core dump code + * + * (C) Copyright 2006 IBM Corp. + * + * Author: Dwayne Grant McConnell + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include +#include +#include +#include +#include +#include + +#include + +#include "spufs.h" + +struct spufs_ctx_info { + struct list_head list; + int dfd; + int memsize; /* in bytes */ + struct spu_context *ctx; +}; + +static LIST_HEAD(ctx_info_list); + +static ssize_t do_coredump_read(int num, struct spu_context *ctx, void __user *buffer, + size_t size, loff_t *off) +{ + u64 data; + int ret; + + if (spufs_coredump_read[num].read) + return spufs_coredump_read[num].read(ctx, buffer, size, off); + + data = spufs_coredump_read[num].get(ctx); + ret = copy_to_user(buffer, &data, 8); + return ret ? -EFAULT : 8; +} + +/* + * These are the only things you should do on a core-file: use only these + * functions to write out all the necessary info. + */ +static int spufs_dump_write(struct file *file, const void *addr, int nr) +{ + return file->f_op->write(file, addr, nr, &file->f_pos) == nr; +} + +static int spufs_dump_seek(struct file *file, loff_t off) +{ + if (file->f_op->llseek) { + if (file->f_op->llseek(file, off, 0) != off) + return 0; + } else + file->f_pos = off; + return 1; +} + +static void spufs_fill_memsize(struct spufs_ctx_info *ctx_info) +{ + struct spu_context *ctx; + unsigned long long lslr; + + ctx = ctx_info->ctx; + lslr = ctx->csa.priv2.spu_lslr_RW; + ctx_info->memsize = lslr + 1; +} + +static int spufs_ctx_note_size(struct spufs_ctx_info *ctx_info) +{ + int dfd, memsize, i, sz, total = 0; + char *name; + char fullname[80]; + + dfd = ctx_info->dfd; + memsize = ctx_info->memsize; + + for (i = 0; spufs_coredump_read[i].name; i++) { + name = spufs_coredump_read[i].name; + sz = spufs_coredump_read[i].size; + + sprintf(fullname, "SPU/%d/%s", dfd, name); + + total += sizeof(struct elf_note); + total += roundup(strlen(fullname) + 1, 4); + if (!strcmp(name, "mem")) + total += roundup(memsize, 4); + else + total += roundup(sz, 4); + } + + return total; +} + +static int spufs_add_one_context(struct file *file, int dfd) +{ + struct spu_context *ctx; + struct spufs_ctx_info *ctx_info; + int size; + + ctx = SPUFS_I(file->f_dentry->d_inode)->i_ctx; + if (ctx->flags & SPU_CREATE_NOSCHED) + return 0; + + ctx_info = kzalloc(sizeof(*ctx_info), GFP_KERNEL); + if (unlikely(!ctx_info)) + return -ENOMEM; + + ctx_info->dfd = dfd; + ctx_info->ctx = ctx; + + spufs_fill_memsize(ctx_info); + + size = spufs_ctx_note_size(ctx_info); + list_add(&ctx_info->list, &ctx_info_list); + return size; +} + +/* + * The additional architecture-specific notes for Cell are various + * context files in the spu context. + * + * This function iterates over all open file descriptors and sees + * if they are a directory in spufs. In that case we use spufs + * internal functionality to dump them without needing to actually + * open the files. + */ +static int spufs_arch_notes_size(void) +{ + struct fdtable *fdt = files_fdtable(current->files); + int size = 0, fd; + + for (fd = 0; fd < fdt->max_fdset && fd < fdt->max_fds; fd++) { + if (FD_ISSET(fd, fdt->open_fds)) { + struct file *file = fcheck(fd); + + if (file && file->f_op == &spufs_context_fops) { + int rval = spufs_add_one_context(file, fd); + if (rval < 0) + break; + size += rval; + } + } + } + + return size; +} + +static void spufs_arch_write_note(struct spufs_ctx_info *ctx_info, int i, + struct file *file) +{ + struct spu_context *ctx; + loff_t pos = 0; + int sz, dfd, rc, total = 0; + const int bufsz = 4096; + char *name; + char fullname[80], *buf; + struct elf_note en; + + buf = kmalloc(bufsz, GFP_KERNEL); + if (!buf) + return; + + dfd = ctx_info->dfd; + name = spufs_coredump_read[i].name; + + if (!strcmp(name, "mem")) + sz = ctx_info->memsize; + else + sz = spufs_coredump_read[i].size; + + ctx = ctx_info->ctx; + if (!ctx) { + return; + } + + sprintf(fullname, "SPU/%d/%s", dfd, name); + en.n_namesz = strlen(fullname) + 1; + en.n_descsz = sz; + en.n_type = NT_SPU; + + if (!spufs_dump_write(file, &en, sizeof(en))) + return; + if (!spufs_dump_write(file, fullname, en.n_namesz)) + return; + if (!spufs_dump_seek(file, roundup((unsigned long)file->f_pos, 4))) + return; + + do { + rc = do_coredump_read(i, ctx, buf, bufsz, &pos); + if (rc > 0) { + if (!spufs_dump_write(file, buf, rc)) + return; + total += rc; + } + } while (rc == bufsz && total < sz); + + spufs_dump_seek(file, roundup((unsigned long)file->f_pos + - total + sz, 4)); +} + +static void spufs_arch_write_notes(struct file *file) +{ + int j; + struct spufs_ctx_info *ctx_info, *next; + + list_for_each_entry_safe(ctx_info, next, &ctx_info_list, list) { + spu_acquire_saved(ctx_info->ctx); + for (j = 0; j < spufs_coredump_num_notes; j++) + spufs_arch_write_note(ctx_info, j, file); + spu_release(ctx_info->ctx); + list_del(&ctx_info->list); + kfree(ctx_info); + } +} + +struct spu_coredump_calls spufs_coredump_calls = { + .arch_notes_size = spufs_arch_notes_size, + .arch_write_notes = spufs_arch_write_notes, + .owner = THIS_MODULE, +}; diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c index 50e0afc46ad2..347eff56fcbd 100644 --- a/arch/powerpc/platforms/cell/spufs/file.c +++ b/arch/powerpc/platforms/cell/spufs/file.c @@ -39,7 +39,6 @@ #define SPUFS_MMAP_4K (PAGE_SIZE == 0x1000) - static int spufs_mem_open(struct inode *inode, struct file *file) { @@ -51,19 +50,24 @@ spufs_mem_open(struct inode *inode, struct file *file) return 0; } +static ssize_t +__spufs_mem_read(struct spu_context *ctx, char __user *buffer, + size_t size, loff_t *pos) +{ + char *local_store = ctx->ops->get_ls(ctx); + return simple_read_from_buffer(buffer, size, pos, local_store, + LS_SIZE); +} + static ssize_t spufs_mem_read(struct file *file, char __user *buffer, size_t size, loff_t *pos) { - struct spu_context *ctx = file->private_data; - char *local_store; int ret; + struct spu_context *ctx = file->private_data; spu_acquire(ctx); - - local_store = ctx->ops->get_ls(ctx); - ret = simple_read_from_buffer(buffer, size, pos, local_store, LS_SIZE); - + ret = __spufs_mem_read(ctx, buffer, size, pos); spu_release(ctx); return ret; } @@ -261,19 +265,24 @@ spufs_regs_open(struct inode *inode, struct file *file) return 0; } +static ssize_t +__spufs_regs_read(struct spu_context *ctx, char __user *buffer, + size_t size, loff_t *pos) +{ + struct spu_lscsa *lscsa = ctx->csa.lscsa; + return simple_read_from_buffer(buffer, size, pos, + lscsa->gprs, sizeof lscsa->gprs); +} + static ssize_t spufs_regs_read(struct file *file, char __user *buffer, size_t size, loff_t *pos) { - struct spu_context *ctx = file->private_data; - struct spu_lscsa *lscsa = ctx->csa.lscsa; int ret; + struct spu_context *ctx = file->private_data; spu_acquire_saved(ctx); - - ret = simple_read_from_buffer(buffer, size, pos, - lscsa->gprs, sizeof lscsa->gprs); - + ret = __spufs_regs_read(ctx, buffer, size, pos); spu_release(ctx); return ret; } @@ -307,19 +316,24 @@ static struct file_operations spufs_regs_fops = { .llseek = generic_file_llseek, }; +static ssize_t +__spufs_fpcr_read(struct spu_context *ctx, char __user * buffer, + size_t size, loff_t * pos) +{ + struct spu_lscsa *lscsa = ctx->csa.lscsa; + return simple_read_from_buffer(buffer, size, pos, + &lscsa->fpcr, sizeof(lscsa->fpcr)); +} + static ssize_t spufs_fpcr_read(struct file *file, char __user * buffer, size_t size, loff_t * pos) { - struct spu_context *ctx = file->private_data; - struct spu_lscsa *lscsa = ctx->csa.lscsa; int ret; + struct spu_context *ctx = file->private_data; spu_acquire_saved(ctx); - - ret = simple_read_from_buffer(buffer, size, pos, - &lscsa->fpcr, sizeof(lscsa->fpcr)); - + ret = __spufs_fpcr_read(ctx, buffer, size, pos); spu_release(ctx); return ret; } @@ -719,22 +733,19 @@ static int spufs_signal1_open(struct inode *inode, struct file *file) return nonseekable_open(inode, file); } -static ssize_t spufs_signal1_read(struct file *file, char __user *buf, +static ssize_t __spufs_signal1_read(struct spu_context *ctx, char __user *buf, size_t len, loff_t *pos) { - struct spu_context *ctx = file->private_data; int ret = 0; u32 data; if (len < 4) return -EINVAL; - spu_acquire_saved(ctx); if (ctx->csa.spu_chnlcnt_RW[3]) { data = ctx->csa.spu_chnldata_RW[3]; ret = 4; } - spu_release(ctx); if (!ret) goto out; @@ -746,6 +757,19 @@ out: return ret; } +static ssize_t spufs_signal1_read(struct file *file, char __user *buf, + size_t len, loff_t *pos) +{ + int ret; + struct spu_context *ctx = file->private_data; + + spu_acquire_saved(ctx); + ret = __spufs_signal1_read(ctx, buf, len, pos); + spu_release(ctx); + + return ret; +} + static ssize_t spufs_signal1_write(struct file *file, const char __user *buf, size_t len, loff_t *pos) { @@ -816,22 +840,19 @@ static int spufs_signal2_open(struct inode *inode, struct file *file) return nonseekable_open(inode, file); } -static ssize_t spufs_signal2_read(struct file *file, char __user *buf, +static ssize_t __spufs_signal2_read(struct spu_context *ctx, char __user *buf, size_t len, loff_t *pos) { - struct spu_context *ctx = file->private_data; int ret = 0; u32 data; if (len < 4) return -EINVAL; - spu_acquire_saved(ctx); if (ctx->csa.spu_chnlcnt_RW[4]) { data = ctx->csa.spu_chnldata_RW[4]; ret = 4; } - spu_release(ctx); if (!ret) goto out; @@ -840,7 +861,20 @@ static ssize_t spufs_signal2_read(struct file *file, char __user *buf, return -EFAULT; out: - return 4; + return ret; +} + +static ssize_t spufs_signal2_read(struct file *file, char __user *buf, + size_t len, loff_t *pos) +{ + struct spu_context *ctx = file->private_data; + int ret; + + spu_acquire_saved(ctx); + ret = __spufs_signal2_read(ctx, buf, len, pos); + spu_release(ctx); + + return ret; } static ssize_t spufs_signal2_write(struct file *file, const char __user *buf, @@ -916,13 +950,19 @@ static void spufs_signal1_type_set(void *data, u64 val) spu_release(ctx); } +static u64 __spufs_signal1_type_get(void *data) +{ + struct spu_context *ctx = data; + return ctx->ops->signal1_type_get(ctx); +} + static u64 spufs_signal1_type_get(void *data) { struct spu_context *ctx = data; u64 ret; spu_acquire(ctx); - ret = ctx->ops->signal1_type_get(ctx); + ret = __spufs_signal1_type_get(data); spu_release(ctx); return ret; @@ -939,13 +979,19 @@ static void spufs_signal2_type_set(void *data, u64 val) spu_release(ctx); } +static u64 __spufs_signal2_type_get(void *data) +{ + struct spu_context *ctx = data; + return ctx->ops->signal2_type_get(ctx); +} + static u64 spufs_signal2_type_get(void *data) { struct spu_context *ctx = data; u64 ret; spu_acquire(ctx); - ret = ctx->ops->signal2_type_get(ctx); + ret = __spufs_signal2_type_get(data); spu_release(ctx); return ret; @@ -1387,13 +1433,19 @@ static void spufs_decr_set(void *data, u64 val) spu_release(ctx); } -static u64 spufs_decr_get(void *data) +static u64 __spufs_decr_get(void *data) { struct spu_context *ctx = data; struct spu_lscsa *lscsa = ctx->csa.lscsa; + return lscsa->decr.slot[0]; +} + +static u64 spufs_decr_get(void *data) +{ + struct spu_context *ctx = data; u64 ret; spu_acquire_saved(ctx); - ret = lscsa->decr.slot[0]; + ret = __spufs_decr_get(data); spu_release(ctx); return ret; } @@ -1409,13 +1461,19 @@ static void spufs_decr_status_set(void *data, u64 val) spu_release(ctx); } -static u64 spufs_decr_status_get(void *data) +static u64 __spufs_decr_status_get(void *data) { struct spu_context *ctx = data; struct spu_lscsa *lscsa = ctx->csa.lscsa; + return lscsa->decr_status.slot[0]; +} + +static u64 spufs_decr_status_get(void *data) +{ + struct spu_context *ctx = data; u64 ret; spu_acquire_saved(ctx); - ret = lscsa->decr_status.slot[0]; + ret = __spufs_decr_status_get(data); spu_release(ctx); return ret; } @@ -1431,30 +1489,43 @@ static void spufs_event_mask_set(void *data, u64 val) spu_release(ctx); } -static u64 spufs_event_mask_get(void *data) +static u64 __spufs_event_mask_get(void *data) { struct spu_context *ctx = data; struct spu_lscsa *lscsa = ctx->csa.lscsa; + return lscsa->event_mask.slot[0]; +} + +static u64 spufs_event_mask_get(void *data) +{ + struct spu_context *ctx = data; u64 ret; spu_acquire_saved(ctx); - ret = lscsa->event_mask.slot[0]; + ret = __spufs_event_mask_get(data); spu_release(ctx); return ret; } DEFINE_SIMPLE_ATTRIBUTE(spufs_event_mask_ops, spufs_event_mask_get, spufs_event_mask_set, "0x%llx\n") -static u64 spufs_event_status_get(void *data) +static u64 __spufs_event_status_get(void *data) { struct spu_context *ctx = data; struct spu_state *state = &ctx->csa; - u64 ret = 0; u64 stat; - - spu_acquire_saved(ctx); stat = state->spu_chnlcnt_RW[0]; if (stat) - ret = state->spu_chnldata_RW[0]; + return state->spu_chnldata_RW[0]; + return 0; +} + +static u64 spufs_event_status_get(void *data) +{ + struct spu_context *ctx = data; + u64 ret = 0; + + spu_acquire_saved(ctx); + ret = __spufs_event_status_get(data); spu_release(ctx); return ret; } @@ -1499,12 +1570,18 @@ static u64 spufs_id_get(void *data) } DEFINE_SIMPLE_ATTRIBUTE(spufs_id_ops, spufs_id_get, NULL, "0x%llx\n") -static u64 spufs_object_id_get(void *data) +static u64 __spufs_object_id_get(void *data) { struct spu_context *ctx = data; return ctx->object_id; } +static u64 spufs_object_id_get(void *data) +{ + /* FIXME: Should there really be no locking here? */ + return __spufs_object_id_get(data); +} + static void spufs_object_id_set(void *data, u64 id) { struct spu_context *ctx = data; @@ -1514,13 +1591,19 @@ static void spufs_object_id_set(void *data, u64 id) DEFINE_SIMPLE_ATTRIBUTE(spufs_object_id_ops, spufs_object_id_get, spufs_object_id_set, "0x%llx\n"); +static u64 __spufs_lslr_get(void *data) +{ + struct spu_context *ctx = data; + return ctx->csa.priv2.spu_lslr_RW; +} + static u64 spufs_lslr_get(void *data) { struct spu_context *ctx = data; u64 ret; spu_acquire_saved(ctx); - ret = ctx->csa.priv2.spu_lslr_RW; + ret = __spufs_lslr_get(data); spu_release(ctx); return ret; @@ -1535,26 +1618,36 @@ static int spufs_info_open(struct inode *inode, struct file *file) return 0; } +static ssize_t __spufs_mbox_info_read(struct spu_context *ctx, + char __user *buf, size_t len, loff_t *pos) +{ + u32 mbox_stat; + u32 data; + + mbox_stat = ctx->csa.prob.mb_stat_R; + if (mbox_stat & 0x0000ff) { + data = ctx->csa.prob.pu_mb_R; + } + + return simple_read_from_buffer(buf, len, pos, &data, sizeof data); +} + static ssize_t spufs_mbox_info_read(struct file *file, char __user *buf, size_t len, loff_t *pos) { + int ret; struct spu_context *ctx = file->private_data; - u32 mbox_stat; - u32 data; if (!access_ok(VERIFY_WRITE, buf, len)) return -EFAULT; spu_acquire_saved(ctx); spin_lock(&ctx->csa.register_lock); - mbox_stat = ctx->csa.prob.mb_stat_R; - if (mbox_stat & 0x0000ff) { - data = ctx->csa.prob.pu_mb_R; - } + ret = __spufs_mbox_info_read(ctx, buf, len, pos); spin_unlock(&ctx->csa.register_lock); spu_release(ctx); - return simple_read_from_buffer(buf, len, pos, &data, sizeof data); + return ret; } static struct file_operations spufs_mbox_info_fops = { @@ -1563,26 +1656,36 @@ static struct file_operations spufs_mbox_info_fops = { .llseek = generic_file_llseek, }; +static ssize_t __spufs_ibox_info_read(struct spu_context *ctx, + char __user *buf, size_t len, loff_t *pos) +{ + u32 ibox_stat; + u32 data; + + ibox_stat = ctx->csa.prob.mb_stat_R; + if (ibox_stat & 0xff0000) { + data = ctx->csa.priv2.puint_mb_R; + } + + return simple_read_from_buffer(buf, len, pos, &data, sizeof data); +} + static ssize_t spufs_ibox_info_read(struct file *file, char __user *buf, size_t len, loff_t *pos) { struct spu_context *ctx = file->private_data; - u32 ibox_stat; - u32 data; + int ret; if (!access_ok(VERIFY_WRITE, buf, len)) return -EFAULT; spu_acquire_saved(ctx); spin_lock(&ctx->csa.register_lock); - ibox_stat = ctx->csa.prob.mb_stat_R; - if (ibox_stat & 0xff0000) { - data = ctx->csa.priv2.puint_mb_R; - } + ret = __spufs_ibox_info_read(ctx, buf, len, pos); spin_unlock(&ctx->csa.register_lock); spu_release(ctx); - return simple_read_from_buffer(buf, len, pos, &data, sizeof data); + return ret; } static struct file_operations spufs_ibox_info_fops = { @@ -1591,29 +1694,39 @@ static struct file_operations spufs_ibox_info_fops = { .llseek = generic_file_llseek, }; -static ssize_t spufs_wbox_info_read(struct file *file, char __user *buf, - size_t len, loff_t *pos) +static ssize_t __spufs_wbox_info_read(struct spu_context *ctx, + char __user *buf, size_t len, loff_t *pos) { - struct spu_context *ctx = file->private_data; int i, cnt; u32 data[4]; u32 wbox_stat; + wbox_stat = ctx->csa.prob.mb_stat_R; + cnt = 4 - ((wbox_stat & 0x00ff00) >> 8); + for (i = 0; i < cnt; i++) { + data[i] = ctx->csa.spu_mailbox_data[i]; + } + + return simple_read_from_buffer(buf, len, pos, &data, + cnt * sizeof(u32)); +} + +static ssize_t spufs_wbox_info_read(struct file *file, char __user *buf, + size_t len, loff_t *pos) +{ + struct spu_context *ctx = file->private_data; + int ret; + if (!access_ok(VERIFY_WRITE, buf, len)) return -EFAULT; spu_acquire_saved(ctx); spin_lock(&ctx->csa.register_lock); - wbox_stat = ctx->csa.prob.mb_stat_R; - cnt = (wbox_stat & 0x00ff00) >> 8; - for (i = 0; i < cnt; i++) { - data[i] = ctx->csa.spu_mailbox_data[i]; - } + ret = __spufs_wbox_info_read(ctx, buf, len, pos); spin_unlock(&ctx->csa.register_lock); spu_release(ctx); - return simple_read_from_buffer(buf, len, pos, &data, - cnt * sizeof(u32)); + return ret; } static struct file_operations spufs_wbox_info_fops = { @@ -1622,19 +1735,13 @@ static struct file_operations spufs_wbox_info_fops = { .llseek = generic_file_llseek, }; -static ssize_t spufs_dma_info_read(struct file *file, char __user *buf, - size_t len, loff_t *pos) +static ssize_t __spufs_dma_info_read(struct spu_context *ctx, + char __user *buf, size_t len, loff_t *pos) { - struct spu_context *ctx = file->private_data; struct spu_dma_info info; struct mfc_cq_sr *qp, *spuqp; int i; - if (!access_ok(VERIFY_WRITE, buf, len)) - return -EFAULT; - - spu_acquire_saved(ctx); - spin_lock(&ctx->csa.register_lock); info.dma_info_type = ctx->csa.priv2.spu_tag_status_query_RW; info.dma_info_mask = ctx->csa.lscsa->tag_mask.slot[0]; info.dma_info_status = ctx->csa.spu_chnldata_RW[24]; @@ -1649,25 +1756,40 @@ static ssize_t spufs_dma_info_read(struct file *file, char __user *buf, qp->mfc_cq_data2_RW = spuqp->mfc_cq_data2_RW; qp->mfc_cq_data3_RW = spuqp->mfc_cq_data3_RW; } - spin_unlock(&ctx->csa.register_lock); - spu_release(ctx); return simple_read_from_buffer(buf, len, pos, &info, sizeof info); } +static ssize_t spufs_dma_info_read(struct file *file, char __user *buf, + size_t len, loff_t *pos) +{ + struct spu_context *ctx = file->private_data; + int ret; + + if (!access_ok(VERIFY_WRITE, buf, len)) + return -EFAULT; + + spu_acquire_saved(ctx); + spin_lock(&ctx->csa.register_lock); + ret = __spufs_dma_info_read(ctx, buf, len, pos); + spin_unlock(&ctx->csa.register_lock); + spu_release(ctx); + + return ret; +} + static struct file_operations spufs_dma_info_fops = { .open = spufs_info_open, .read = spufs_dma_info_read, }; -static ssize_t spufs_proxydma_info_read(struct file *file, char __user *buf, - size_t len, loff_t *pos) +static ssize_t __spufs_proxydma_info_read(struct spu_context *ctx, + char __user *buf, size_t len, loff_t *pos) { - struct spu_context *ctx = file->private_data; struct spu_proxydma_info info; - int ret = sizeof info; struct mfc_cq_sr *qp, *puqp; + int ret = sizeof info; int i; if (len < ret) @@ -1676,8 +1798,6 @@ static ssize_t spufs_proxydma_info_read(struct file *file, char __user *buf, if (!access_ok(VERIFY_WRITE, buf, len)) return -EFAULT; - spu_acquire_saved(ctx); - spin_lock(&ctx->csa.register_lock); info.proxydma_info_type = ctx->csa.prob.dma_querytype_RW; info.proxydma_info_mask = ctx->csa.prob.dma_querymask_RW; info.proxydma_info_status = ctx->csa.prob.dma_tagstatus_R; @@ -1690,12 +1810,23 @@ static ssize_t spufs_proxydma_info_read(struct file *file, char __user *buf, qp->mfc_cq_data2_RW = puqp->mfc_cq_data2_RW; qp->mfc_cq_data3_RW = puqp->mfc_cq_data3_RW; } + + return simple_read_from_buffer(buf, len, pos, &info, + sizeof info); +} + +static ssize_t spufs_proxydma_info_read(struct file *file, char __user *buf, + size_t len, loff_t *pos) +{ + struct spu_context *ctx = file->private_data; + int ret; + + spu_acquire_saved(ctx); + spin_lock(&ctx->csa.register_lock); + ret = __spufs_proxydma_info_read(ctx, buf, len, pos); spin_unlock(&ctx->csa.register_lock); spu_release(ctx); - if (copy_to_user(buf, &info, sizeof info)) - ret = -EFAULT; - return ret; } @@ -1760,3 +1891,27 @@ struct tree_descr spufs_dir_nosched_contents[] = { { "object-id", &spufs_object_id_ops, 0666, }, {}, }; + +struct spufs_coredump_reader spufs_coredump_read[] = { + { "regs", __spufs_regs_read, NULL, 128 * 16 }, + { "fpcr", __spufs_fpcr_read, NULL, 16 }, + { "lslr", NULL, __spufs_lslr_get, 11 }, + { "decr", NULL, __spufs_decr_get, 11 }, + { "decr_status", NULL, __spufs_decr_status_get, 11 }, + { "mem", __spufs_mem_read, NULL, 256 * 1024, }, + { "signal1", __spufs_signal1_read, NULL, 4 }, + { "signal1_type", NULL, __spufs_signal1_type_get, 2 }, + { "signal2", __spufs_signal2_read, NULL, 4 }, + { "signal2_type", NULL, __spufs_signal2_type_get, 2 }, + { "event_mask", NULL, __spufs_event_mask_get, 8 }, + { "event_status", NULL, __spufs_event_status_get, 8 }, + { "mbox_info", __spufs_mbox_info_read, NULL, 4 }, + { "ibox_info", __spufs_ibox_info_read, NULL, 4 }, + { "wbox_info", __spufs_wbox_info_read, NULL, 16 }, + { "dma_info", __spufs_dma_info_read, NULL, 69 * 8 }, + { "proxydma_info", __spufs_proxydma_info_read, NULL, 35 * 8 }, + { "object-id", NULL, __spufs_object_id_get, 19 }, + { }, +}; +int spufs_coredump_num_notes = ARRAY_SIZE(spufs_coredump_read) - 1; + diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index d5f0a21a19d8..a3ca06bd0ca1 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -232,6 +232,7 @@ struct file_operations spufs_context_fops = { .readdir = dcache_readdir, .fsync = simple_sync_file, }; +EXPORT_SYMBOL_GPL(spufs_context_fops); static int spufs_mkdir(struct inode *dir, struct dentry *dentry, unsigned int flags, @@ -647,6 +648,7 @@ static struct file_system_type spufs_type = { static int __init spufs_init(void) { int ret; + ret = -ENOMEM; spufs_inode_cache = kmem_cache_create("spufs_inode_cache", sizeof(struct spufs_inode_info), 0, @@ -662,10 +664,14 @@ static int __init spufs_init(void) if (ret) goto out_cache; ret = register_spu_syscalls(&spufs_calls); + if (ret) + goto out_fs; + ret = register_arch_coredump_calls(&spufs_coredump_calls); if (ret) goto out_fs; spufs_init_isolated_loader(); + return 0; out_fs: unregister_filesystem(&spufs_type); @@ -679,6 +685,7 @@ module_init(spufs_init); static void __exit spufs_exit(void) { spu_sched_exit(); + unregister_arch_coredump_calls(&spufs_coredump_calls); unregister_spu_syscalls(&spufs_calls); unregister_filesystem(&spufs_type); kmem_cache_destroy(spufs_inode_cache); diff --git a/arch/powerpc/platforms/cell/spufs/spufs.h b/arch/powerpc/platforms/cell/spufs/spufs.h index 23d20f380560..70fb13395c04 100644 --- a/arch/powerpc/platforms/cell/spufs/spufs.h +++ b/arch/powerpc/platforms/cell/spufs/spufs.h @@ -223,4 +223,15 @@ void spufs_stop_callback(struct spu *spu); void spufs_mfc_callback(struct spu *spu); void spufs_dma_callback(struct spu *spu, int type); +extern struct spu_coredump_calls spufs_coredump_calls; +struct spufs_coredump_reader { + char *name; + ssize_t (*read)(struct spu_context *ctx, + char __user *buffer, size_t size, loff_t *pos); + u64 (*get)(void *data); + size_t size; +}; +extern struct spufs_coredump_reader spufs_coredump_read[]; +extern int spufs_coredump_num_notes; + #endif diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 79b05a1a4365..cc72bb43061d 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1582,6 +1582,10 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) sz += thread_status_size; +#ifdef ELF_CORE_WRITE_EXTRA_NOTES + sz += ELF_CORE_EXTRA_NOTES_SIZE; +#endif + fill_elf_note_phdr(&phdr, sz, offset); offset += sz; DUMP_WRITE(&phdr, sizeof(phdr)); @@ -1622,6 +1626,10 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) if (!writenote(notes + i, file, &foffset)) goto end_coredump; +#ifdef ELF_CORE_WRITE_EXTRA_NOTES + ELF_CORE_WRITE_EXTRA_NOTES; +#endif + /* write out the thread status notes section */ list_for_each(t, &thread_list) { struct elf_thread_status *tmp = diff --git a/include/asm-powerpc/elf.h b/include/asm-powerpc/elf.h index 9a83a987d396..4545aa682509 100644 --- a/include/asm-powerpc/elf.h +++ b/include/asm-powerpc/elf.h @@ -411,4 +411,17 @@ do { \ /* Keep this the last entry. */ #define R_PPC64_NUM 107 +#ifdef CONFIG_PPC_CELL +/* Notes used in ET_CORE. Note name is "SPU//". */ +#define NT_SPU 1 + +extern int arch_notes_size(void); +extern void arch_write_notes(struct file *file); + +#define ELF_CORE_EXTRA_NOTES_SIZE arch_notes_size() +#define ELF_CORE_WRITE_EXTRA_NOTES arch_write_notes(file) + +#define ARCH_HAVE_EXTRA_ELF_NOTES +#endif /* CONFIG_PPC_CELL */ + #endif /* _ASM_POWERPC_ELF_H */ diff --git a/include/asm-powerpc/spu.h b/include/asm-powerpc/spu.h index ffa4df083609..f968f8697538 100644 --- a/include/asm-powerpc/spu.h +++ b/include/asm-powerpc/spu.h @@ -172,6 +172,13 @@ extern struct spufs_calls { struct module *owner; } spufs_calls; +/* coredump calls implemented in spufs */ +struct spu_coredump_calls { + asmlinkage int (*arch_notes_size)(void); + asmlinkage void (*arch_write_notes)(struct file *file); + struct module *owner; +}; + /* return status from spu_run, same as in libspe */ #define SPE_EVENT_DMA_ALIGNMENT 0x0008 /*A DMA alignment error */ #define SPE_EVENT_SPE_ERROR 0x0010 /*An illegal instruction error*/ @@ -203,6 +210,9 @@ static inline void unregister_spu_syscalls(struct spufs_calls *calls) } #endif /* MODULE */ +int register_arch_coredump_calls(struct spu_coredump_calls *calls); +void unregister_arch_coredump_calls(struct spu_coredump_calls *calls); + int spu_add_sysdev_attr(struct sysdev_attribute *attr); void spu_remove_sysdev_attr(struct sysdev_attribute *attr); diff --git a/include/linux/elf.h b/include/linux/elf.h index b70d1d2c8d28..743d5c8e6d36 100644 --- a/include/linux/elf.h +++ b/include/linux/elf.h @@ -368,5 +368,12 @@ extern Elf64_Dyn _DYNAMIC []; #endif +#ifndef ARCH_HAVE_EXTRA_ELF_NOTES +static inline int arch_notes_size(void) { return 0; } +static inline void arch_write_notes(struct file *file) { } + +#define ELF_CORE_EXTRA_NOTES_SIZE arch_notes_size() +#define ELF_CORE_WRITE_EXTRA_NOTES arch_write_notes(file) +#endif /* ARCH_HAVE_EXTRA_ELF_NOTES */ #endif /* _LINUX_ELF_H */ -- cgit v1.2.3 From 0c789ff64e874bbece03b607f200566ab24ea9ab Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 4 Dec 2006 10:52:29 -0800 Subject: netfilter.h needs rcuupdate.h for RCU locking functions This was exposed by Al's recent header file dependency reduction patches.. Cc: Al Viro Cc: David S. Miller Signed-off-by: Linus Torvalds --- include/linux/netfilter.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index d4c4c5120bc0..9389cf7d4062 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -11,6 +11,7 @@ #include #endif #include +#include /* Responses from hook functions. */ #define NF_DROP 0 -- cgit v1.2.3 From a80958f4849316a18c06f75b9e850ccecbf20df8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 4 Dec 2006 20:41:19 +0000 Subject: [PATCH] fix fallout from header dependency trimming OK, that seems to be enough to deal with the mess. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/kernelcapi.h | 1 + include/linux/netfilter.h | 1 - include/linux/skbuff.h | 1 + 3 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kernelcapi.h b/include/linux/kernelcapi.h index f8a0ff86dacc..aea34e74c496 100644 --- a/include/linux/kernelcapi.h +++ b/include/linux/kernelcapi.h @@ -48,6 +48,7 @@ typedef struct kcapi_carddef { #include #include #include +#include #define KCI_CONTRUP 0 /* arg: struct capi_profile */ #define KCI_CONTRDOWN 1 /* arg: NULL */ diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 9389cf7d4062..d4c4c5120bc0 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -11,7 +11,6 @@ #include #endif #include -#include /* Responses from hook functions. */ #define NF_DROP 0 diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 6bdff9b148d0..a05a5f7c0b73 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -25,6 +25,7 @@ #include #include #include +#include #include #define HAVE_ALLOC_SKB /* For the drivers to know */ -- cgit v1.2.3 From e62438630ca37539c8cc1553710bbfaa3cf960a7 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 4 Dec 2006 03:38:31 -0700 Subject: [PATCH] Centralise definitions of sector_t and blkcnt_t CONFIG_LBD and CONFIG_LSF are spread into asm/types.h for no particularly good reason. Centralising the definition in linux/types.h means that arch maintainers don't need to bother adding it, as well as fixing the problem with x86-64 users being asked to make a decision that has absolutely no effect. The H8/300 porters seem particularly confused since I'm not aware of any microcontrollers that need to support 2TB filesystems. Signed-off-by: Matthew Wilcox Signed-off-by: Linus Torvalds --- block/Kconfig | 6 ++---- include/asm-avr32/types.h | 5 ----- include/asm-h8300/types.h | 6 ------ include/asm-i386/types.h | 10 ---------- include/asm-mips/types.h | 10 ---------- include/asm-powerpc/types.h | 10 ---------- include/asm-s390/types.h | 10 ---------- include/asm-sh/types.h | 10 ---------- include/asm-x86_64/types.h | 3 --- include/linux/types.h | 14 +++++++++----- 10 files changed, 11 insertions(+), 73 deletions(-) (limited to 'include/linux') diff --git a/block/Kconfig b/block/Kconfig index 83766a6bdee2..a50f48111647 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -19,11 +19,9 @@ config BLOCK if BLOCK -#XXX - it makes sense to enable this only for 32-bit subarch's, not for x86_64 -#for instance. config LBD bool "Support for Large Block Devices" - depends on X86 || (MIPS && 32BIT) || PPC32 || (S390 && !64BIT) || SUPERH || UML + depends on !64BIT help Say Y here if you want to attach large (bigger than 2TB) discs to your machine, or if you want to have a raid or loopback device @@ -44,7 +42,7 @@ config BLK_DEV_IO_TRACE config LSF bool "Support for Large Single Files" - depends on X86 || (MIPS && 32BIT) || PPC32 || ARCH_S390_31 || SUPERH || UML + depends on !64BIT help Say Y here if you want to be able to handle very large files (bigger than 2TB), otherwise say N. diff --git a/include/asm-avr32/types.h b/include/asm-avr32/types.h index 3f47db9675af..2bff153a32ed 100644 --- a/include/asm-avr32/types.h +++ b/include/asm-avr32/types.h @@ -57,11 +57,6 @@ typedef unsigned long long u64; typedef u32 dma_addr_t; -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-h8300/types.h b/include/asm-h8300/types.h index da2402b86540..2a8b1b2be782 100644 --- a/include/asm-h8300/types.h +++ b/include/asm-h8300/types.h @@ -55,12 +55,6 @@ typedef unsigned long long u64; typedef u32 dma_addr_t; -#define HAVE_SECTOR_T -typedef u64 sector_t; - -#define HAVE_BLKCNT_T -typedef u64 blkcnt_t; - #endif /* __KERNEL__ */ #endif /* __ASSEMBLY__ */ diff --git a/include/asm-i386/types.h b/include/asm-i386/types.h index 4b4b295ccdb9..ad0a55bd782f 100644 --- a/include/asm-i386/types.h +++ b/include/asm-i386/types.h @@ -57,16 +57,6 @@ typedef u32 dma_addr_t; #endif typedef u64 dma64_addr_t; -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - -#ifdef CONFIG_LSF -typedef u64 blkcnt_t; -#define HAVE_BLKCNT_T -#endif - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-mips/types.h b/include/asm-mips/types.h index 2b52e180c6f2..63a13c5bd832 100644 --- a/include/asm-mips/types.h +++ b/include/asm-mips/types.h @@ -93,16 +93,6 @@ typedef unsigned long long phys_t; typedef unsigned long phys_t; #endif -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - -#ifdef CONFIG_LSF -typedef u64 blkcnt_t; -#define HAVE_BLKCNT_T -#endif - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-powerpc/types.h b/include/asm-powerpc/types.h index d6fb56b80453..3b363757a2bb 100644 --- a/include/asm-powerpc/types.h +++ b/include/asm-powerpc/types.h @@ -97,16 +97,6 @@ typedef struct { unsigned long env; } func_descr_t; -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - -#ifdef CONFIG_LSF -typedef u64 blkcnt_t; -#define HAVE_BLKCNT_T -#endif - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-s390/types.h b/include/asm-s390/types.h index ae2951cc83ac..fc5d7cf19324 100644 --- a/include/asm-s390/types.h +++ b/include/asm-s390/types.h @@ -87,16 +87,6 @@ typedef union { } subreg; } register_pair; -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - -#ifdef CONFIG_LSF -typedef u64 blkcnt_t; -#define HAVE_BLKCNT_T -#endif - #endif /* ! __s390x__ */ #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-sh/types.h b/include/asm-sh/types.h index 3c09dd4ca31c..fd00dbb82f84 100644 --- a/include/asm-sh/types.h +++ b/include/asm-sh/types.h @@ -52,16 +52,6 @@ typedef unsigned long long u64; typedef u32 dma_addr_t; -#ifdef CONFIG_LBD -typedef u64 sector_t; -#define HAVE_SECTOR_T -#endif - -#ifdef CONFIG_LSF -typedef u64 blkcnt_t; -#define HAVE_BLKCNT_T -#endif - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-x86_64/types.h b/include/asm-x86_64/types.h index c86c2e6793e2..2d4491aae281 100644 --- a/include/asm-x86_64/types.h +++ b/include/asm-x86_64/types.h @@ -48,9 +48,6 @@ typedef unsigned long long u64; typedef u64 dma64_addr_t; typedef u64 dma_addr_t; -typedef u64 sector_t; -#define HAVE_SECTOR_T - #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/include/linux/types.h b/include/linux/types.h index 745c409ebbb5..0351bf2fac85 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -136,15 +136,19 @@ typedef __s64 int64_t; * * Linux always considers sectors to be 512 bytes long independently * of the devices real block size. - * - * If required, asm/types.h can override it and define - * HAVE_SECTOR_T */ -#ifndef HAVE_SECTOR_T +#ifdef CONFIG_LBD +typedef u64 sector_t; +#else typedef unsigned long sector_t; #endif -#ifndef HAVE_BLKCNT_T +/* + * The type of the inode's block count. + */ +#ifdef CONFIG_LSF +typedef u64 blkcnt_t; +#else typedef unsigned long blkcnt_t; #endif -- cgit v1.2.3 From 6d5aefb8eaa38e44b5b8cf60c812aceafc02d924 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 5 Dec 2006 19:36:26 +0000 Subject: WorkQueue: Fix up arch-specific work items where possible Fix up arch-specific work items where possible to use the new work_struct and delayed_work structs. Three places that enqueue bits of their stack and then return have been marked with #error as this is not permitted. Signed-Off-By: David Howells --- arch/arm/common/sharpsl_pm.c | 22 +++++++++++----------- arch/arm/mach-omap1/board-h3.c | 3 ++- arch/arm/mach-omap1/board-nokia770.c | 6 +++--- arch/arm/mach-omap1/leds-osk.c | 4 ++-- arch/arm/mach-omap2/board-h4.c | 3 ++- arch/arm/mach-pxa/akita-ioexp.c | 6 +++--- arch/ia64/hp/sim/simserial.c | 4 ++-- arch/ia64/kernel/mca.c | 8 ++++---- arch/ia64/kernel/smpboot.c | 12 +++++++----- arch/mips/kernel/kspd.c | 4 ++-- arch/powerpc/platforms/embedded6xx/ls_uart.c | 4 ++-- arch/powerpc/platforms/powermac/backlight.c | 12 ++++++------ arch/ppc/8260_io/fcc_enet.c | 21 +++++++++++++-------- arch/ppc/8xx_io/fec.c | 21 +++++++++++++-------- arch/s390/appldata/appldata_base.c | 6 +++--- arch/um/drivers/chan_kern.c | 2 +- arch/um/drivers/mconsole_kern.c | 4 ++-- arch/um/drivers/net_kern.c | 1 + arch/um/drivers/port_kern.c | 4 ++-- drivers/macintosh/rack-meter.c | 16 ++++++++++------ drivers/net/chelsio/cphy.h | 2 +- drivers/net/chelsio/my3126.c | 8 +++++--- drivers/net/netxen/netxen_nic.h | 3 ++- drivers/net/netxen/netxen_nic_init.c | 5 +++-- drivers/net/netxen/netxen_nic_main.c | 19 ++++++++++--------- drivers/net/smc91x.c | 15 +++++++++------ drivers/net/wireless/zd1211rw/zd_mac.c | 23 +++++++++++++---------- drivers/net/wireless/zd1211rw/zd_mac.h | 4 ++-- drivers/spi/pxa2xx_spi.c | 9 +++++---- drivers/usb/core/hub.c | 2 +- drivers/usb/misc/appledisplay.c | 11 ++++++----- drivers/video/pxafb.c | 7 ++++--- include/asm-arm/arch-omap/irda.h | 2 +- include/linux/netpoll.h | 2 +- 34 files changed, 154 insertions(+), 121 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/common/sharpsl_pm.c b/arch/arm/common/sharpsl_pm.c index 605dedf96790..b3599743093b 100644 --- a/arch/arm/common/sharpsl_pm.c +++ b/arch/arm/common/sharpsl_pm.c @@ -60,16 +60,16 @@ static int sharpsl_ac_check(void); static int sharpsl_fatal_check(void); static int sharpsl_average_value(int ad); static void sharpsl_average_clear(void); -static void sharpsl_charge_toggle(void *private_); -static void sharpsl_battery_thread(void *private_); +static void sharpsl_charge_toggle(struct work_struct *private_); +static void sharpsl_battery_thread(struct work_struct *private_); /* * Variables */ struct sharpsl_pm_status sharpsl_pm; -DECLARE_WORK(toggle_charger, sharpsl_charge_toggle, NULL); -DECLARE_WORK(sharpsl_bat, sharpsl_battery_thread, NULL); +DECLARE_DELAYED_WORK(toggle_charger, sharpsl_charge_toggle); +DECLARE_DELAYED_WORK(sharpsl_bat, sharpsl_battery_thread); DEFINE_LED_TRIGGER(sharpsl_charge_led_trigger); @@ -116,7 +116,7 @@ void sharpsl_battery_kick(void) EXPORT_SYMBOL(sharpsl_battery_kick); -static void sharpsl_battery_thread(void *private_) +static void sharpsl_battery_thread(struct work_struct *private_) { int voltage, percent, apm_status, i = 0; @@ -128,7 +128,7 @@ static void sharpsl_battery_thread(void *private_) /* Corgi cannot confirm when battery fully charged so periodically kick! */ if (!sharpsl_pm.machinfo->batfull_irq && (sharpsl_pm.charge_mode == CHRG_ON) && time_after(jiffies, sharpsl_pm.charge_start_time + SHARPSL_CHARGE_ON_TIME_INTERVAL)) - schedule_work(&toggle_charger); + schedule_delayed_work(&toggle_charger, 0); while(1) { voltage = sharpsl_pm.machinfo->read_devdata(SHARPSL_BATT_VOLT); @@ -212,7 +212,7 @@ static void sharpsl_charge_off(void) sharpsl_pm_led(SHARPSL_LED_OFF); sharpsl_pm.charge_mode = CHRG_OFF; - schedule_work(&sharpsl_bat); + schedule_delayed_work(&sharpsl_bat, 0); } static void sharpsl_charge_error(void) @@ -222,7 +222,7 @@ static void sharpsl_charge_error(void) sharpsl_pm.charge_mode = CHRG_ERROR; } -static void sharpsl_charge_toggle(void *private_) +static void sharpsl_charge_toggle(struct work_struct *private_) { dev_dbg(sharpsl_pm.dev, "Toogling Charger at time: %lx\n", jiffies); @@ -254,7 +254,7 @@ static void sharpsl_ac_timer(unsigned long data) else if (sharpsl_pm.charge_mode == CHRG_ON) sharpsl_charge_off(); - schedule_work(&sharpsl_bat); + schedule_delayed_work(&sharpsl_bat, 0); } @@ -279,10 +279,10 @@ static void sharpsl_chrg_full_timer(unsigned long data) sharpsl_charge_off(); } else if (sharpsl_pm.full_count < 2) { dev_dbg(sharpsl_pm.dev, "Charge Full: Count too low\n"); - schedule_work(&toggle_charger); + schedule_delayed_work(&toggle_charger, 0); } else if (time_after(jiffies, sharpsl_pm.charge_start_time + SHARPSL_CHARGE_FINISH_TIME)) { dev_dbg(sharpsl_pm.dev, "Charge Full: Interrupt generated too slowly - retry.\n"); - schedule_work(&toggle_charger); + schedule_delayed_work(&toggle_charger, 0); } else { sharpsl_charge_off(); sharpsl_pm.charge_mode = CHRG_DONE; diff --git a/arch/arm/mach-omap1/board-h3.c b/arch/arm/mach-omap1/board-h3.c index f225a083dee1..9d2346fb68f4 100644 --- a/arch/arm/mach-omap1/board-h3.c +++ b/arch/arm/mach-omap1/board-h3.c @@ -323,7 +323,8 @@ static int h3_transceiver_mode(struct device *dev, int mode) cancel_delayed_work(&irda_config->gpio_expa); PREPARE_WORK(&irda_config->gpio_expa, set_trans_mode, &mode); - schedule_work(&irda_config->gpio_expa); +#error this is not permitted - mode is an argument variable + schedule_delayed_work(&irda_config->gpio_expa, 0); return 0; } diff --git a/arch/arm/mach-omap1/board-nokia770.c b/arch/arm/mach-omap1/board-nokia770.c index dbc555d209ff..cbe909bad79b 100644 --- a/arch/arm/mach-omap1/board-nokia770.c +++ b/arch/arm/mach-omap1/board-nokia770.c @@ -74,7 +74,7 @@ static struct omap_kp_platform_data nokia770_kp_data = { .rows = 8, .cols = 8, .keymap = nokia770_keymap, - .keymapsize = ARRAY_SIZE(nokia770_keymap) + .keymapsize = ARRAY_SIZE(nokia770_keymap), .delay = 4, }; @@ -191,7 +191,7 @@ static void nokia770_audio_pwr_up(void) printk("HP connected\n"); } -static void codec_delayed_power_down(void *arg) +static void codec_delayed_power_down(struct work_struct *work) { down(&audio_pwr_sem); if (audio_pwr_state == -1) @@ -200,7 +200,7 @@ static void codec_delayed_power_down(void *arg) up(&audio_pwr_sem); } -static DECLARE_WORK(codec_power_down_work, codec_delayed_power_down, NULL); +static DECLARE_DELAYED_WORK(codec_power_down_work, codec_delayed_power_down); static void nokia770_audio_pwr_down(void) { diff --git a/arch/arm/mach-omap1/leds-osk.c b/arch/arm/mach-omap1/leds-osk.c index 3b29e59b0e6f..0cbf1b0071f8 100644 --- a/arch/arm/mach-omap1/leds-osk.c +++ b/arch/arm/mach-omap1/leds-osk.c @@ -35,7 +35,7 @@ static u8 hw_led_state; static u8 tps_leds_change; -static void tps_work(void *unused) +static void tps_work(struct work_struct *unused) { for (;;) { u8 leds; @@ -61,7 +61,7 @@ static void tps_work(void *unused) } } -static DECLARE_WORK(work, tps_work, NULL); +static DECLARE_WORK(work, tps_work); #ifdef CONFIG_OMAP_OSK_MISTRAL diff --git a/arch/arm/mach-omap2/board-h4.c b/arch/arm/mach-omap2/board-h4.c index 26a95a642ad7..3b1ad1d981a3 100644 --- a/arch/arm/mach-omap2/board-h4.c +++ b/arch/arm/mach-omap2/board-h4.c @@ -206,7 +206,8 @@ static int h4_transceiver_mode(struct device *dev, int mode) cancel_delayed_work(&irda_config->gpio_expa); PREPARE_WORK(&irda_config->gpio_expa, set_trans_mode, &mode); - schedule_work(&irda_config->gpio_expa); +#error this is not permitted - mode is an argument variable + schedule_delayed_work(&irda_config->gpio_expa, 0); return 0; } diff --git a/arch/arm/mach-pxa/akita-ioexp.c b/arch/arm/mach-pxa/akita-ioexp.c index 1b398742ab56..12d2fe0ceff6 100644 --- a/arch/arm/mach-pxa/akita-ioexp.c +++ b/arch/arm/mach-pxa/akita-ioexp.c @@ -36,11 +36,11 @@ I2C_CLIENT_INSMOD; static int max7310_write(struct i2c_client *client, int address, int data); static struct i2c_client max7310_template; -static void akita_ioexp_work(void *private_); +static void akita_ioexp_work(struct work_struct *private_); static struct device *akita_ioexp_device; static unsigned char ioexp_output_value = AKITA_IOEXP_IO_OUT; -DECLARE_WORK(akita_ioexp, akita_ioexp_work, NULL); +DECLARE_WORK(akita_ioexp, akita_ioexp_work); /* @@ -158,7 +158,7 @@ void akita_reset_ioexp(struct device *dev, unsigned char bit) EXPORT_SYMBOL(akita_set_ioexp); EXPORT_SYMBOL(akita_reset_ioexp); -static void akita_ioexp_work(void *private_) +static void akita_ioexp_work(struct work_struct *private_) { if (akita_ioexp_device) max7310_set_ouputs(akita_ioexp_device, ioexp_output_value); diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c index caab986af70c..b62f0c4d2c7c 100644 --- a/arch/ia64/hp/sim/simserial.c +++ b/arch/ia64/hp/sim/simserial.c @@ -209,7 +209,7 @@ static void do_serial_bh(void) } #endif -static void do_softint(void *private_) +static void do_softint(struct work_struct *private_) { printk(KERN_ERR "simserial: do_softint called\n"); } @@ -698,7 +698,7 @@ static int get_async_struct(int line, struct async_struct **ret_info) info->flags = sstate->flags; info->xmit_fifo_size = sstate->xmit_fifo_size; info->line = line; - INIT_WORK(&info->work, do_softint, info); + INIT_WORK(&info->work, do_softint); info->state = sstate; if (sstate->info) { kfree(info); diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 7cfa63a98cb3..6bedd97570ca 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -678,7 +678,7 @@ ia64_mca_cmc_vector_enable (void *dummy) * disable the cmc interrupt vector. */ static void -ia64_mca_cmc_vector_disable_keventd(void *unused) +ia64_mca_cmc_vector_disable_keventd(struct work_struct *unused) { on_each_cpu(ia64_mca_cmc_vector_disable, NULL, 1, 0); } @@ -690,7 +690,7 @@ ia64_mca_cmc_vector_disable_keventd(void *unused) * enable the cmc interrupt vector. */ static void -ia64_mca_cmc_vector_enable_keventd(void *unused) +ia64_mca_cmc_vector_enable_keventd(struct work_struct *unused) { on_each_cpu(ia64_mca_cmc_vector_enable, NULL, 1, 0); } @@ -1247,8 +1247,8 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, monarch_cpu = -1; } -static DECLARE_WORK(cmc_disable_work, ia64_mca_cmc_vector_disable_keventd, NULL); -static DECLARE_WORK(cmc_enable_work, ia64_mca_cmc_vector_enable_keventd, NULL); +static DECLARE_WORK(cmc_disable_work, ia64_mca_cmc_vector_disable_keventd); +static DECLARE_WORK(cmc_enable_work, ia64_mca_cmc_vector_enable_keventd); /* * ia64_mca_cmc_int_handler diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index f7d7f5668144..b21ddecea943 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -463,15 +463,17 @@ struct pt_regs * __devinit idle_regs(struct pt_regs *regs) } struct create_idle { + struct work_struct work; struct task_struct *idle; struct completion done; int cpu; }; void -do_fork_idle(void *_c_idle) +do_fork_idle(struct work_struct *work) { - struct create_idle *c_idle = _c_idle; + struct create_idle *c_idle = + container_of(work, struct create_idle, work); c_idle->idle = fork_idle(c_idle->cpu); complete(&c_idle->done); @@ -482,10 +484,10 @@ do_boot_cpu (int sapicid, int cpu) { int timeout; struct create_idle c_idle = { + .work = __WORK_INITIALIZER(c_idle.work, do_fork_idle), .cpu = cpu, .done = COMPLETION_INITIALIZER(c_idle.done), }; - DECLARE_WORK(work, do_fork_idle, &c_idle); c_idle.idle = get_idle_for_cpu(cpu); if (c_idle.idle) { @@ -497,9 +499,9 @@ do_boot_cpu (int sapicid, int cpu) * We can't use kernel_thread since we must avoid to reschedule the child. */ if (!keventd_up() || current_is_keventd()) - work.func(work.data); + c_idle.work.func(&c_idle.work); else { - schedule_work(&work); + schedule_work(&c_idle.work); wait_for_completion(&c_idle.done); } diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index f06a144c7881..2c82412b9efe 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -319,7 +319,7 @@ static void sp_cleanup(void) static int channel_open = 0; /* the work handler */ -static void sp_work(void *data) +static void sp_work(struct work_struct *unused) { if (!channel_open) { if( rtlx_open(RTLX_CHANNEL_SYSIO, 1) != 0) { @@ -354,7 +354,7 @@ static void startwork(int vpe) return; } - INIT_WORK(&work, sp_work, NULL); + INIT_WORK(&work, sp_work); queue_work(workqueue, &work); } else queue_work(workqueue, &work); diff --git a/arch/powerpc/platforms/embedded6xx/ls_uart.c b/arch/powerpc/platforms/embedded6xx/ls_uart.c index 31bcdae84823..0e837762cc5b 100644 --- a/arch/powerpc/platforms/embedded6xx/ls_uart.c +++ b/arch/powerpc/platforms/embedded6xx/ls_uart.c @@ -14,7 +14,7 @@ static unsigned long avr_clock; static struct work_struct wd_work; -static void wd_stop(void *unused) +static void wd_stop(struct work_struct *unused) { const char string[] = "AAAAFFFFJJJJ>>>>VVVV>>>>ZZZZVVVVKKKK"; int i = 0, rescue = 8; @@ -122,7 +122,7 @@ static int __init ls_uarts_init(void) ls_uart_init(); - INIT_WORK(&wd_work, wd_stop, NULL); + INIT_WORK(&wd_work, wd_stop); schedule_work(&wd_work); return 0; diff --git a/arch/powerpc/platforms/powermac/backlight.c b/arch/powerpc/platforms/powermac/backlight.c index afa593a8544a..c3a89414ddc0 100644 --- a/arch/powerpc/platforms/powermac/backlight.c +++ b/arch/powerpc/platforms/powermac/backlight.c @@ -18,11 +18,11 @@ #define OLD_BACKLIGHT_MAX 15 -static void pmac_backlight_key_worker(void *data); -static void pmac_backlight_set_legacy_worker(void *data); +static void pmac_backlight_key_worker(struct work_struct *work); +static void pmac_backlight_set_legacy_worker(struct work_struct *work); -static DECLARE_WORK(pmac_backlight_key_work, pmac_backlight_key_worker, NULL); -static DECLARE_WORK(pmac_backlight_set_legacy_work, pmac_backlight_set_legacy_worker, NULL); +static DECLARE_WORK(pmac_backlight_key_work, pmac_backlight_key_worker); +static DECLARE_WORK(pmac_backlight_set_legacy_work, pmac_backlight_set_legacy_worker); /* Although these variables are used in interrupt context, it makes no sense to * protect them. No user is able to produce enough key events per second and @@ -94,7 +94,7 @@ int pmac_backlight_curve_lookup(struct fb_info *info, int value) return level; } -static void pmac_backlight_key_worker(void *data) +static void pmac_backlight_key_worker(struct work_struct *work) { if (atomic_read(&kernel_backlight_disabled)) return; @@ -166,7 +166,7 @@ static int __pmac_backlight_set_legacy_brightness(int brightness) return error; } -static void pmac_backlight_set_legacy_worker(void *data) +static void pmac_backlight_set_legacy_worker(struct work_struct *work) { if (atomic_read(&kernel_backlight_disabled)) return; diff --git a/arch/ppc/8260_io/fcc_enet.c b/arch/ppc/8260_io/fcc_enet.c index 2e1943e27819..709952c25f29 100644 --- a/arch/ppc/8260_io/fcc_enet.c +++ b/arch/ppc/8260_io/fcc_enet.c @@ -385,6 +385,7 @@ struct fcc_enet_private { phy_info_t *phy; struct work_struct phy_relink; struct work_struct phy_display_config; + struct net_device *dev; uint sequence_done; @@ -1391,10 +1392,11 @@ static phy_info_t *phy_info[] = { NULL }; -static void mii_display_status(void *data) +static void mii_display_status(struct work_struct *work) { - struct net_device *dev = data; - volatile struct fcc_enet_private *fep = dev->priv; + volatile struct fcc_enet_private *fep = + container_of(work, struct fcc_enet_private, phy_relink); + struct net_device *dev = fep->dev; uint s = fep->phy_status; if (!fep->link && !fep->old_link) { @@ -1428,10 +1430,12 @@ static void mii_display_status(void *data) printk(".\n"); } -static void mii_display_config(void *data) +static void mii_display_config(struct work_struct *work) { - struct net_device *dev = data; - volatile struct fcc_enet_private *fep = dev->priv; + volatile struct fcc_enet_private *fep = + container_of(work, struct fcc_enet_private, + phy_display_config); + struct net_device *dev = fep->dev; uint s = fep->phy_status; printk("%s: config: auto-negotiation ", dev->name); @@ -1758,8 +1762,9 @@ static int __init fec_enet_init(void) cep->phy_id_done = 0; cep->phy_addr = fip->fc_phyaddr; mii_queue(dev, mk_mii_read(MII_PHYSID1), mii_discover_phy); - INIT_WORK(&cep->phy_relink, mii_display_status, dev); - INIT_WORK(&cep->phy_display_config, mii_display_config, dev); + INIT_WORK(&cep->phy_relink, mii_display_status); + INIT_WORK(&cep->phy_display_config, mii_display_config); + cep->dev = dev; #endif /* CONFIG_USE_MDIO */ fip++; diff --git a/arch/ppc/8xx_io/fec.c b/arch/ppc/8xx_io/fec.c index 2f9fa9e3d331..e6c28fb423b2 100644 --- a/arch/ppc/8xx_io/fec.c +++ b/arch/ppc/8xx_io/fec.c @@ -173,6 +173,7 @@ struct fec_enet_private { uint phy_speed; phy_info_t *phy; struct work_struct phy_task; + struct net_device *dev; uint sequence_done; @@ -1263,10 +1264,11 @@ static void mii_display_status(struct net_device *dev) printk(".\n"); } -static void mii_display_config(void *priv) +static void mii_display_config(struct work_struct *work) { - struct net_device *dev = (struct net_device *)priv; - struct fec_enet_private *fep = dev->priv; + struct fec_enet_private *fep = + container_of(work, struct fec_enet_private, phy_task); + struct net_device *dev = fep->dev; volatile uint *s = &(fep->phy_status); printk("%s: config: auto-negotiation ", dev->name); @@ -1295,10 +1297,11 @@ static void mii_display_config(void *priv) fep->sequence_done = 1; } -static void mii_relink(void *priv) +static void mii_relink(struct work_struct *work) { - struct net_device *dev = (struct net_device *)priv; - struct fec_enet_private *fep = dev->priv; + struct fec_enet_private *fep = + container_of(work, struct fec_enet_private, phy_task); + struct net_device *dev = fep->dev; int duplex; fep->link = (fep->phy_status & PHY_STAT_LINK) ? 1 : 0; @@ -1325,7 +1328,8 @@ static void mii_queue_relink(uint mii_reg, struct net_device *dev) { struct fec_enet_private *fep = dev->priv; - INIT_WORK(&fep->phy_task, mii_relink, (void *)dev); + fep->dev = dev; + INIT_WORK(&fep->phy_task, mii_relink); schedule_work(&fep->phy_task); } @@ -1333,7 +1337,8 @@ static void mii_queue_config(uint mii_reg, struct net_device *dev) { struct fec_enet_private *fep = dev->priv; - INIT_WORK(&fep->phy_task, mii_display_config, (void *)dev); + fep->dev = dev; + INIT_WORK(&fep->phy_task, mii_display_config); schedule_work(&fep->phy_task); } diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index af1e8fc7d985..67d5cf9cba83 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -92,8 +92,8 @@ static int appldata_timer_active; * Work queue */ static struct workqueue_struct *appldata_wq; -static void appldata_work_fn(void *data); -static DECLARE_WORK(appldata_work, appldata_work_fn, NULL); +static void appldata_work_fn(struct work_struct *work); +static DECLARE_WORK(appldata_work, appldata_work_fn); /* @@ -125,7 +125,7 @@ static void appldata_timer_function(unsigned long data) * * call data gathering function for each (active) module */ -static void appldata_work_fn(void *data) +static void appldata_work_fn(struct work_struct *work) { struct list_head *lh; struct appldata_ops *ops; diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c index 3576b3cc505e..7d4190e55654 100644 --- a/arch/um/drivers/chan_kern.c +++ b/arch/um/drivers/chan_kern.c @@ -638,7 +638,7 @@ int chan_out_fd(struct list_head *chans) return -1; } -void chan_interrupt(struct list_head *chans, struct work_struct *task, +void chan_interrupt(struct list_head *chans, struct delayed_work *task, struct tty_struct *tty, int irq) { struct list_head *ele, *next; diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index 7b172160fe04..96f0189327af 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -56,7 +56,7 @@ static struct notifier_block reboot_notifier = { static LIST_HEAD(mc_requests); -static void mc_work_proc(void *unused) +static void mc_work_proc(struct work_struct *unused) { struct mconsole_entry *req; unsigned long flags; @@ -72,7 +72,7 @@ static void mc_work_proc(void *unused) } } -static DECLARE_WORK(mconsole_work, mc_work_proc, NULL); +static DECLARE_WORK(mconsole_work, mc_work_proc); static irqreturn_t mconsole_interrupt(int irq, void *dev_id) { diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c index ec9eb8bd9432..286bc0b3207f 100644 --- a/arch/um/drivers/net_kern.c +++ b/arch/um/drivers/net_kern.c @@ -99,6 +99,7 @@ irqreturn_t uml_net_interrupt(int irq, void *dev_id) * same device, since it tests for (dev->flags & IFF_UP). So * there's no harm in delaying the device shutdown. */ schedule_work(&close_work); +#error this is not permitted - close_work will go out of scope goto out; } reactivate_fd(lp->fd, UM_ETH_IRQ); diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c index ce9f3733f73e..6dfe632f1c14 100644 --- a/arch/um/drivers/port_kern.c +++ b/arch/um/drivers/port_kern.c @@ -132,7 +132,7 @@ static int port_accept(struct port_list *port) DECLARE_MUTEX(ports_sem); struct list_head ports = LIST_HEAD_INIT(ports); -void port_work_proc(void *unused) +void port_work_proc(struct work_struct *unused) { struct port_list *port; struct list_head *ele; @@ -150,7 +150,7 @@ void port_work_proc(void *unused) local_irq_restore(flags); } -DECLARE_WORK(port_work, port_work_proc, NULL); +DECLARE_WORK(port_work, port_work_proc); static irqreturn_t port_interrupt(int irq, void *data) { diff --git a/drivers/macintosh/rack-meter.c b/drivers/macintosh/rack-meter.c index f1b6f563673a..5ed41fe84e57 100644 --- a/drivers/macintosh/rack-meter.c +++ b/drivers/macintosh/rack-meter.c @@ -48,7 +48,8 @@ struct rackmeter_dma { } ____cacheline_aligned; struct rackmeter_cpu { - struct work_struct sniffer; + struct delayed_work sniffer; + struct rackmeter *rm; cputime64_t prev_wall; cputime64_t prev_idle; int zero; @@ -208,11 +209,12 @@ static void rackmeter_setup_dbdma(struct rackmeter *rm) rackmeter_do_pause(rm, 0); } -static void rackmeter_do_timer(void *data) +static void rackmeter_do_timer(struct work_struct *work) { - struct rackmeter *rm = data; + struct rackmeter_cpu *rcpu = + container_of(work, struct rackmeter_cpu, sniffer.work); + struct rackmeter *rm = rcpu->rm; unsigned int cpu = smp_processor_id(); - struct rackmeter_cpu *rcpu = &rm->cpu[cpu]; cputime64_t cur_jiffies, total_idle_ticks; unsigned int total_ticks, idle_ticks; int i, offset, load, cumm, pause; @@ -263,8 +265,10 @@ static void __devinit rackmeter_init_cpu_sniffer(struct rackmeter *rm) * on those machines yet */ - INIT_WORK(&rm->cpu[0].sniffer, rackmeter_do_timer, rm); - INIT_WORK(&rm->cpu[1].sniffer, rackmeter_do_timer, rm); + rm->cpu[0].rm = rm; + INIT_DELAYED_WORK(&rm->cpu[0].sniffer, rackmeter_do_timer); + rm->cpu[1].rm = rm; + INIT_DELAYED_WORK(&rm->cpu[1].sniffer, rackmeter_do_timer); for_each_online_cpu(cpu) { struct rackmeter_cpu *rcpu; diff --git a/drivers/net/chelsio/cphy.h b/drivers/net/chelsio/cphy.h index 60901f25014e..cf9143499882 100644 --- a/drivers/net/chelsio/cphy.h +++ b/drivers/net/chelsio/cphy.h @@ -91,7 +91,7 @@ struct cphy { int state; /* Link status state machine */ adapter_t *adapter; /* associated adapter */ - struct work_struct phy_update; + struct delayed_work phy_update; u16 bmsr; int count; diff --git a/drivers/net/chelsio/my3126.c b/drivers/net/chelsio/my3126.c index 0b90014d5b3e..c7731b6f9de3 100644 --- a/drivers/net/chelsio/my3126.c +++ b/drivers/net/chelsio/my3126.c @@ -93,9 +93,11 @@ static int my3126_interrupt_handler(struct cphy *cphy) return cphy_cause_link_change; } -static void my3216_poll(void *arg) +static void my3216_poll(struct work_struct *work) { - my3126_interrupt_handler(arg); + struct cphy *cphy = container_of(work, struct cphy, phy_update.work); + + my3126_interrupt_handler(cphy); } static int my3126_set_loopback(struct cphy *cphy, int on) @@ -171,7 +173,7 @@ static struct cphy *my3126_phy_create(adapter_t *adapter, if (cphy) cphy_init(cphy, adapter, phy_addr, &my3126_ops, mdio_ops); - INIT_WORK(&cphy->phy_update, my3216_poll, cphy); + INIT_DELAYED_WORK(&cphy->phy_update, my3216_poll); cphy->bmsr = 0; return (cphy); diff --git a/drivers/net/netxen/netxen_nic.h b/drivers/net/netxen/netxen_nic.h index d925053fe597..9c588af8ab74 100644 --- a/drivers/net/netxen/netxen_nic.h +++ b/drivers/net/netxen/netxen_nic.h @@ -714,6 +714,7 @@ struct netxen_adapter { spinlock_t lock; struct work_struct watchdog_task; struct work_struct tx_timeout_task; + struct net_device *netdev; struct timer_list watchdog_timer; u32 curr_window; @@ -921,7 +922,7 @@ netxen_nic_do_ioctl(struct netxen_adapter *adapter, void *u_data, struct netxen_port *port); int netxen_nic_rx_has_work(struct netxen_adapter *adapter); int netxen_nic_tx_has_work(struct netxen_adapter *adapter); -void netxen_watchdog_task(unsigned long v); +void netxen_watchdog_task(struct work_struct *work); void netxen_post_rx_buffers(struct netxen_adapter *adapter, u32 ctx, u32 ringid); void netxen_process_cmd_ring(unsigned long data); diff --git a/drivers/net/netxen/netxen_nic_init.c b/drivers/net/netxen/netxen_nic_init.c index 0dca029bc3e5..eae18236aefa 100644 --- a/drivers/net/netxen/netxen_nic_init.c +++ b/drivers/net/netxen/netxen_nic_init.c @@ -710,12 +710,13 @@ static inline int netxen_nic_check_temp(struct netxen_adapter *adapter) return rv; } -void netxen_watchdog_task(unsigned long v) +void netxen_watchdog_task(struct work_struct *work) { int port_num; struct netxen_port *port; struct net_device *netdev; - struct netxen_adapter *adapter = (struct netxen_adapter *)v; + struct netxen_adapter *adapter = + container_of(work, struct netxen_adapter, watchdog_task); if (netxen_nic_check_temp(adapter)) return; diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c index 1cb662d5bd76..df0bb36a1cfb 100644 --- a/drivers/net/netxen/netxen_nic_main.c +++ b/drivers/net/netxen/netxen_nic_main.c @@ -64,7 +64,7 @@ static int netxen_nic_open(struct net_device *netdev); static int netxen_nic_close(struct net_device *netdev); static int netxen_nic_xmit_frame(struct sk_buff *, struct net_device *); static void netxen_tx_timeout(struct net_device *netdev); -static void netxen_tx_timeout_task(struct net_device *netdev); +static void netxen_tx_timeout_task(struct work_struct *work); static void netxen_watchdog(unsigned long); static int netxen_handle_int(struct netxen_adapter *, struct net_device *); static int netxen_nic_ioctl(struct net_device *netdev, @@ -274,8 +274,7 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) adapter->ahw.xg_linkup = 0; adapter->watchdog_timer.function = &netxen_watchdog; adapter->watchdog_timer.data = (unsigned long)adapter; - INIT_WORK(&adapter->watchdog_task, - (void (*)(void *))netxen_watchdog_task, adapter); + INIT_WORK(&adapter->watchdog_task, netxen_watchdog_task); adapter->ahw.pdev = pdev; adapter->proc_cmd_buf_counter = 0; pci_read_config_byte(pdev, PCI_REVISION_ID, &adapter->ahw.revision_id); @@ -379,8 +378,8 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) dev_addr); } } - INIT_WORK(&adapter->tx_timeout_task, - (void (*)(void *))netxen_tx_timeout_task, netdev); + adapter->netdev = netdev; + INIT_WORK(&adapter->tx_timeout_task, netxen_tx_timeout_task); netif_carrier_off(netdev); netif_stop_queue(netdev); @@ -938,18 +937,20 @@ static void netxen_tx_timeout(struct net_device *netdev) schedule_work(&adapter->tx_timeout_task); } -static void netxen_tx_timeout_task(struct net_device *netdev) +static void netxen_tx_timeout_task(struct work_struct *work) { - struct netxen_port *port = (struct netxen_port *)netdev_priv(netdev); + struct netxen_adapter *adapter = + container_of(work, struct netxen_adapter, tx_timeout_task); + struct net_device *netdev = adapter->netdev; unsigned long flags; printk(KERN_ERR "%s %s: transmit timeout, resetting.\n", netxen_nic_driver_name, netdev->name); - spin_lock_irqsave(&port->adapter->lock, flags); + spin_lock_irqsave(&adapter->lock, flags); netxen_nic_close(netdev); netxen_nic_open(netdev); - spin_unlock_irqrestore(&port->adapter->lock, flags); + spin_unlock_irqrestore(&adapter->lock, flags); netdev->trans_start = jiffies; netif_wake_queue(netdev); } diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c index 95b6478f55c6..e62a9586fb95 100644 --- a/drivers/net/smc91x.c +++ b/drivers/net/smc91x.c @@ -210,6 +210,7 @@ struct smc_local { /* work queue */ struct work_struct phy_configure; + struct net_device *dev; int work_pending; spinlock_t lock; @@ -1114,10 +1115,11 @@ static void smc_phy_check_media(struct net_device *dev, int init) * of autonegotiation.) If the RPC ANEG bit is cleared, the selection * is controlled by the RPC SPEED and RPC DPLX bits. */ -static void smc_phy_configure(void *data) +static void smc_phy_configure(struct work_struct *work) { - struct net_device *dev = data; - struct smc_local *lp = netdev_priv(dev); + struct smc_local *lp = + container_of(work, struct smc_local, phy_configure); + struct net_device *dev = lp->dev; void __iomem *ioaddr = lp->base; int phyaddr = lp->mii.phy_id; int my_phy_caps; /* My PHY capabilities */ @@ -1592,7 +1594,7 @@ smc_open(struct net_device *dev) /* Configure the PHY, initialize the link state */ if (lp->phy_type != 0) - smc_phy_configure(dev); + smc_phy_configure(&lp->phy_configure); else { spin_lock_irq(&lp->lock); smc_10bt_check_media(dev, 1); @@ -1972,7 +1974,8 @@ static int __init smc_probe(struct net_device *dev, void __iomem *ioaddr) #endif tasklet_init(&lp->tx_task, smc_hardware_send_pkt, (unsigned long)dev); - INIT_WORK(&lp->phy_configure, smc_phy_configure, dev); + INIT_WORK(&lp->phy_configure, smc_phy_configure); + lp->dev = dev; lp->mii.phy_id_mask = 0x1f; lp->mii.reg_num_mask = 0x1f; lp->mii.force_media = 0; @@ -2322,7 +2325,7 @@ static int smc_drv_resume(struct platform_device *dev) smc_reset(ndev); smc_enable(ndev); if (lp->phy_type != 0) - smc_phy_configure(ndev); + smc_phy_configure(&lp->phy_configure); netif_device_attach(ndev); } } diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 44f3cfd4cc1d..f1573a9c2336 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -32,8 +32,8 @@ static void ieee_init(struct ieee80211_device *ieee); static void softmac_init(struct ieee80211softmac_device *sm); -static void set_rts_cts_work(void *d); -static void set_basic_rates_work(void *d); +static void set_rts_cts_work(struct work_struct *work); +static void set_basic_rates_work(struct work_struct *work); static void housekeeping_init(struct zd_mac *mac); static void housekeeping_enable(struct zd_mac *mac); @@ -48,8 +48,8 @@ int zd_mac_init(struct zd_mac *mac, memset(mac, 0, sizeof(*mac)); spin_lock_init(&mac->lock); mac->netdev = netdev; - INIT_WORK(&mac->set_rts_cts_work, set_rts_cts_work, mac); - INIT_WORK(&mac->set_basic_rates_work, set_basic_rates_work, mac); + INIT_DELAYED_WORK(&mac->set_rts_cts_work, set_rts_cts_work); + INIT_DELAYED_WORK(&mac->set_basic_rates_work, set_basic_rates_work); ieee_init(ieee); softmac_init(ieee80211_priv(netdev)); @@ -366,9 +366,10 @@ static void try_enable_tx(struct zd_mac *mac) spin_unlock_irqrestore(&mac->lock, flags); } -static void set_rts_cts_work(void *d) +static void set_rts_cts_work(struct work_struct *work) { - struct zd_mac *mac = d; + struct zd_mac *mac = + container_of(work, struct zd_mac, set_rts_cts_work.work); unsigned long flags; u8 rts_rate; unsigned int short_preamble; @@ -387,9 +388,10 @@ static void set_rts_cts_work(void *d) try_enable_tx(mac); } -static void set_basic_rates_work(void *d) +static void set_basic_rates_work(struct work_struct *work) { - struct zd_mac *mac = d; + struct zd_mac *mac = + container_of(work, struct zd_mac, set_basic_rates_work.work); unsigned long flags; u16 basic_rates; @@ -467,12 +469,13 @@ static void bssinfo_change(struct net_device *netdev, u32 changes) if (need_set_rts_cts && !mac->updating_rts_rate) { mac->updating_rts_rate = 1; netif_stop_queue(mac->netdev); - queue_work(zd_workqueue, &mac->set_rts_cts_work); + queue_delayed_work(zd_workqueue, &mac->set_rts_cts_work, 0); } if (need_set_rates && !mac->updating_basic_rates) { mac->updating_basic_rates = 1; netif_stop_queue(mac->netdev); - queue_work(zd_workqueue, &mac->set_basic_rates_work); + queue_delayed_work(zd_workqueue, &mac->set_basic_rates_work, + 0); } spin_unlock_irqrestore(&mac->lock, flags); } diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h index 08d6b8c08e75..d4e8b870409d 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.h +++ b/drivers/net/wireless/zd1211rw/zd_mac.h @@ -133,8 +133,8 @@ struct zd_mac { struct iw_statistics iw_stats; struct housekeeping housekeeping; - struct work_struct set_rts_cts_work; - struct work_struct set_basic_rates_work; + struct delayed_work set_rts_cts_work; + struct delayed_work set_basic_rates_work; unsigned int stats_count; u8 qual_buffer[ZD_MAC_STATS_BUFFER_SIZE]; diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c index 72025df5561d..494d9b856488 100644 --- a/drivers/spi/pxa2xx_spi.c +++ b/drivers/spi/pxa2xx_spi.c @@ -148,7 +148,7 @@ struct chip_data { void (*cs_control)(u32 command); }; -static void pump_messages(void *data); +static void pump_messages(struct work_struct *work); static int flush(struct driver_data *drv_data) { @@ -884,9 +884,10 @@ static void pump_transfers(unsigned long data) } } -static void pump_messages(void *data) +static void pump_messages(struct work_struct *work) { - struct driver_data *drv_data = data; + struct driver_data *drv_data = + container_of(work, struct driver_data, pump_messages); unsigned long flags; /* Lock queue and check for queue work */ @@ -1098,7 +1099,7 @@ static int init_queue(struct driver_data *drv_data) tasklet_init(&drv_data->pump_transfers, pump_transfers, (unsigned long)drv_data); - INIT_WORK(&drv_data->pump_messages, pump_messages, drv_data); + INIT_WORK(&drv_data->pump_messages, pump_messages); drv_data->workqueue = create_singlethread_workqueue( drv_data->master->cdev.dev->bus_id); if (drv_data->workqueue == NULL) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 39186db1015f..9be41ed1f9a6 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -68,7 +68,7 @@ struct usb_hub { unsigned has_indicators:1; u8 indicator[USB_MAXCHILDREN]; - struct work_struct leds; + struct delayed_work leds; }; diff --git a/drivers/usb/misc/appledisplay.c b/drivers/usb/misc/appledisplay.c index ba30ca6a14aa..02cbb7fff24f 100644 --- a/drivers/usb/misc/appledisplay.c +++ b/drivers/usb/misc/appledisplay.c @@ -76,7 +76,7 @@ struct appledisplay { char *urbdata; /* interrupt URB data buffer */ char *msgdata; /* control message data buffer */ - struct work_struct work; + struct delayed_work work; int button_pressed; spinlock_t lock; }; @@ -117,7 +117,7 @@ static void appledisplay_complete(struct urb *urb) case ACD_BTN_BRIGHT_UP: case ACD_BTN_BRIGHT_DOWN: pdata->button_pressed = 1; - queue_work(wq, &pdata->work); + queue_delayed_work(wq, &pdata->work, 0); break; case ACD_BTN_NONE: default: @@ -184,9 +184,10 @@ static struct backlight_properties appledisplay_bl_data = { .max_brightness = 0xFF }; -static void appledisplay_work(void *private) +static void appledisplay_work(struct work_struct *work) { - struct appledisplay *pdata = private; + struct appledisplay *pdata = + container_of(work, struct appledisplay, work.work); int retval; up(&pdata->bd->sem); @@ -238,7 +239,7 @@ static int appledisplay_probe(struct usb_interface *iface, pdata->udev = udev; spin_lock_init(&pdata->lock); - INIT_WORK(&pdata->work, appledisplay_work, pdata); + INIT_DELAYED_WORK(&pdata->work, appledisplay_work); /* Allocate buffer for control messages */ pdata->msgdata = kmalloc(ACD_MSG_BUFFER_LEN, GFP_KERNEL); diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c index 8a8ae55a7403..38eb0b69c2d7 100644 --- a/drivers/video/pxafb.c +++ b/drivers/video/pxafb.c @@ -964,9 +964,10 @@ static void set_ctrlr_state(struct pxafb_info *fbi, u_int state) * Our LCD controller task (which is called when we blank or unblank) * via keventd. */ -static void pxafb_task(void *dummy) +static void pxafb_task(struct work_struct *work) { - struct pxafb_info *fbi = dummy; + struct pxafb_info *fbi = + container_of(work, struct pxafb_info, task); u_int state = xchg(&fbi->task_state, -1); set_ctrlr_state(fbi, state); @@ -1159,7 +1160,7 @@ static struct pxafb_info * __init pxafb_init_fbinfo(struct device *dev) } init_waitqueue_head(&fbi->ctrlr_wait); - INIT_WORK(&fbi->task, pxafb_task, fbi); + INIT_WORK(&fbi->task, pxafb_task); init_MUTEX(&fbi->ctrlr_sem); return fbi; diff --git a/include/asm-arm/arch-omap/irda.h b/include/asm-arm/arch-omap/irda.h index 805ae3575e44..345a649ec838 100644 --- a/include/asm-arm/arch-omap/irda.h +++ b/include/asm-arm/arch-omap/irda.h @@ -24,7 +24,7 @@ struct omap_irda_config { /* Very specific to the needs of some platforms (h3,h4) * having calls which can sleep in irda_set_speed. */ - struct work_struct gpio_expa; + struct delayed_work gpio_expa; int rx_channel; int tx_channel; unsigned long dest_start; diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 2cc9867b1626..29930b71a9aa 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -32,7 +32,7 @@ struct netpoll_info { struct netpoll *rx_np; /* netpoll that registered an rx_hook */ struct sk_buff_head arp_tx; /* list of arp requests to reply to */ struct sk_buff_head txq; - struct work_struct tx_work; + struct delayed_work tx_work; }; void netpoll_poll(struct netpoll *np); -- cgit v1.2.3 From e6b3c4db6fbcd0d33720696f37790d6b8be12313 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 11 Nov 2006 22:18:03 -0500 Subject: Fix a second potential rpc_wakeup race... Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 12 +++---- include/linux/sunrpc/sched.h | 8 +---- net/sunrpc/clnt.c | 19 ++++++----- net/sunrpc/pmap_clnt.c | 2 +- net/sunrpc/sched.c | 80 ++++++++++++++++++++++++++------------------ net/sunrpc/sunrpc_syms.c | 1 - 6 files changed, 65 insertions(+), 57 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 8118036cc449..93ac05889cbc 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -636,7 +636,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) smp_wmb(); } else status = data->rpc_status; - rpc_release_task(task); + rpc_put_task(task); return status; } @@ -742,7 +742,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) smp_wmb(); } else status = data->rpc_status; - rpc_release_task(task); + rpc_put_task(task); if (status != 0) return status; @@ -3067,7 +3067,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co if (status == 0) nfs_post_op_update_inode(inode, &data->fattr); } - rpc_release_task(task); + rpc_put_task(task); return status; } @@ -3314,7 +3314,7 @@ static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock * if (IS_ERR(task)) goto out; status = nfs4_wait_for_completion_rpc_task(task); - rpc_release_task(task); + rpc_put_task(task); out: return status; } @@ -3430,7 +3430,7 @@ static void nfs4_lock_release(void *calldata) task = nfs4_do_unlck(&data->fl, data->ctx, data->lsp, data->arg.lock_seqid); if (!IS_ERR(task)) - rpc_release_task(task); + rpc_put_task(task); dprintk("%s: cancelling lock!\n", __FUNCTION__); } else nfs_free_seqid(data->arg.lock_seqid); @@ -3472,7 +3472,7 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f ret = -EAGAIN; } else data->cancelled = 1; - rpc_release_task(task); + rpc_put_task(task); dprintk("%s: done, ret = %d!\n", __FUNCTION__, ret); return ret; } diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index f399c138f79d..9fdb8c9d09f2 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -178,13 +178,6 @@ struct rpc_call_ops { } while (0) #define RPC_IS_ACTIVATED(t) (test_bit(RPC_TASK_ACTIVE, &(t)->tk_runstate)) -#define rpc_set_active(t) (set_bit(RPC_TASK_ACTIVE, &(t)->tk_runstate)) -#define rpc_clear_active(t) \ - do { \ - smp_mb__before_clear_bit(); \ - clear_bit(RPC_TASK_ACTIVE, &(t)->tk_runstate); \ - smp_mb__after_clear_bit(); \ - } while(0) /* * Task priorities. @@ -254,6 +247,7 @@ struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags, void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *ops, void *data); +void rpc_put_task(struct rpc_task *); void rpc_release_task(struct rpc_task *); void rpc_exit_task(struct rpc_task *); void rpc_killall_tasks(struct rpc_clnt *); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index dfeea4fea95a..a323abc7ea85 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -466,10 +466,9 @@ int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) BUG_ON(flags & RPC_TASK_ASYNC); - status = -ENOMEM; task = rpc_new_task(clnt, flags, &rpc_default_ops, NULL); if (task == NULL) - goto out; + return -ENOMEM; /* Mask signals on RPC calls _and_ GSS_AUTH upcalls */ rpc_task_sigmask(task, &oldset); @@ -478,15 +477,17 @@ int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) /* Set up the call info struct and execute the task */ status = task->tk_status; - if (status == 0) { - atomic_inc(&task->tk_count); - status = rpc_execute(task); - if (status == 0) - status = task->tk_status; + if (status != 0) { + rpc_release_task(task); + goto out; } - rpc_restore_sigmask(&oldset); - rpc_release_task(task); + atomic_inc(&task->tk_count); + status = rpc_execute(task); + if (status == 0) + status = task->tk_status; + rpc_put_task(task); out: + rpc_restore_sigmask(&oldset); return status; } diff --git a/net/sunrpc/pmap_clnt.c b/net/sunrpc/pmap_clnt.c index e52afab413de..8d2e10fc3df9 100644 --- a/net/sunrpc/pmap_clnt.c +++ b/net/sunrpc/pmap_clnt.c @@ -134,7 +134,7 @@ void rpc_getport(struct rpc_task *task) child = rpc_run_task(pmap_clnt, RPC_TASK_ASYNC, &pmap_getport_ops, map); if (IS_ERR(child)) goto bailout; - rpc_release_task(child); + rpc_put_task(child); task->tk_xprt->stat.bind_count++; return; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index b57d4062d429..66d01365f3a5 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -266,12 +266,28 @@ static int rpc_wait_bit_interruptible(void *word) return 0; } +static void rpc_set_active(struct rpc_task *task) +{ + if (test_and_set_bit(RPC_TASK_ACTIVE, &task->tk_runstate) != 0) + return; + spin_lock(&rpc_sched_lock); +#ifdef RPC_DEBUG + task->tk_magic = RPC_TASK_MAGIC_ID; + task->tk_pid = rpc_task_id++; +#endif + /* Add to global list of all tasks */ + list_add_tail(&task->tk_task, &all_tasks); + spin_unlock(&rpc_sched_lock); +} + /* * Mark an RPC call as having completed by clearing the 'active' bit */ -static inline void rpc_mark_complete_task(struct rpc_task *task) +static void rpc_mark_complete_task(struct rpc_task *task) { - rpc_clear_active(task); + smp_mb__before_clear_bit(); + clear_bit(RPC_TASK_ACTIVE, &task->tk_runstate); + smp_mb__after_clear_bit(); wake_up_bit(&task->tk_runstate, RPC_TASK_ACTIVE); } @@ -335,9 +351,6 @@ static void __rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task, return; } - /* Mark the task as being activated if so needed */ - rpc_set_active(task); - __rpc_add_wait_queue(q, task); BUG_ON(task->tk_callback != NULL); @@ -348,6 +361,9 @@ static void __rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task, void rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task, rpc_action action, rpc_action timer) { + /* Mark the task as being activated if so needed */ + rpc_set_active(task); + /* * Protect the queue operations. */ @@ -673,8 +689,6 @@ static int __rpc_execute(struct rpc_task *task) } dprintk("RPC: %4d, return %d, status %d\n", task->tk_pid, status, task->tk_status); - /* Wake up anyone who is waiting for task completion */ - rpc_mark_complete_task(task); /* Release all resources associated with the task */ rpc_release_task(task); return status; @@ -788,15 +802,6 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, cons task->tk_flags |= RPC_TASK_NOINTR; } -#ifdef RPC_DEBUG - task->tk_magic = RPC_TASK_MAGIC_ID; - task->tk_pid = rpc_task_id++; -#endif - /* Add to global list of all tasks */ - spin_lock(&rpc_sched_lock); - list_add_tail(&task->tk_task, &all_tasks); - spin_unlock(&rpc_sched_lock); - BUG_ON(task->tk_ops == NULL); /* starting timestamp */ @@ -849,16 +854,35 @@ cleanup: goto out; } -void rpc_release_task(struct rpc_task *task) + +void rpc_put_task(struct rpc_task *task) { const struct rpc_call_ops *tk_ops = task->tk_ops; void *calldata = task->tk_calldata; + if (!atomic_dec_and_test(&task->tk_count)) + return; + /* Release resources */ + if (task->tk_rqstp) + xprt_release(task); + if (task->tk_msg.rpc_cred) + rpcauth_unbindcred(task); + if (task->tk_client) { + rpc_release_client(task->tk_client); + task->tk_client = NULL; + } + if (task->tk_flags & RPC_TASK_DYNAMIC) + rpc_free_task(task); + if (tk_ops->rpc_release) + tk_ops->rpc_release(calldata); +} +EXPORT_SYMBOL(rpc_put_task); + +void rpc_release_task(struct rpc_task *task) +{ #ifdef RPC_DEBUG BUG_ON(task->tk_magic != RPC_TASK_MAGIC_ID); #endif - if (!atomic_dec_and_test(&task->tk_count)) - return; dprintk("RPC: %4d release task\n", task->tk_pid); /* Remove from global task list */ @@ -871,23 +895,13 @@ void rpc_release_task(struct rpc_task *task) /* Synchronously delete any running timer */ rpc_delete_timer(task); - /* Release resources */ - if (task->tk_rqstp) - xprt_release(task); - if (task->tk_msg.rpc_cred) - rpcauth_unbindcred(task); - if (task->tk_client) { - rpc_release_client(task->tk_client); - task->tk_client = NULL; - } - #ifdef RPC_DEBUG task->tk_magic = 0; #endif - if (task->tk_flags & RPC_TASK_DYNAMIC) - rpc_free_task(task); - if (tk_ops->rpc_release) - tk_ops->rpc_release(calldata); + /* Wake up anyone who is waiting for task completion */ + rpc_mark_complete_task(task); + + rpc_put_task(task); } /** diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c index 192dff5dabcb..faaf81e97720 100644 --- a/net/sunrpc/sunrpc_syms.c +++ b/net/sunrpc/sunrpc_syms.c @@ -33,7 +33,6 @@ EXPORT_SYMBOL(rpciod_down); EXPORT_SYMBOL(rpciod_up); EXPORT_SYMBOL(rpc_new_task); EXPORT_SYMBOL(rpc_wake_up_status); -EXPORT_SYMBOL(rpc_release_task); /* RPC client functions */ EXPORT_SYMBOL(rpc_clone_client); -- cgit v1.2.3 From 8aca67f0ae2d8811165c22326825a645cc8e1b48 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 13 Nov 2006 16:23:44 -0500 Subject: SUNRPC: Fix a potential race in rpc_wake_up_task() Use RCU to ensure that we can safely call rpc_finish_wakeup after we've called __rpc_do_wake_up_task. If not, there is a theoretical race, in which the rpc_task finishes executing, and gets freed first. Signed-off-by: Trond Myklebust --- fs/nfs/read.c | 8 +++++++- fs/nfs/write.c | 20 ++++++++++++++++---- include/linux/nfs_fs.h | 7 ++----- include/linux/sunrpc/sched.h | 2 ++ net/sunrpc/sched.c | 30 ++++++++++++++++++++---------- 5 files changed, 47 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/read.c b/fs/nfs/read.c index c2e49c397a27..8b58bbf6e39e 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -65,13 +65,19 @@ struct nfs_read_data *nfs_readdata_alloc(size_t len) return p; } -static void nfs_readdata_free(struct nfs_read_data *p) +static void nfs_readdata_rcu_free(struct rcu_head *head) { + struct nfs_read_data *p = container_of(head, struct nfs_read_data, task.u.tk_rcu); if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); mempool_free(p, nfs_rdata_mempool); } +static void nfs_readdata_free(struct nfs_read_data *rdata) +{ + call_rcu_bh(&rdata->task.u.tk_rcu, nfs_readdata_rcu_free); +} + void nfs_readdata_release(void *data) { nfs_readdata_free(data); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 883dd4a1c157..29d88209199d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -102,13 +102,19 @@ struct nfs_write_data *nfs_commit_alloc(void) return p; } -void nfs_commit_free(struct nfs_write_data *p) +void nfs_commit_rcu_free(struct rcu_head *head) { + struct nfs_write_data *p = container_of(head, struct nfs_write_data, task.u.tk_rcu); if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); mempool_free(p, nfs_commit_mempool); } +void nfs_commit_free(struct nfs_write_data *wdata) +{ + call_rcu_bh(&wdata->task.u.tk_rcu, nfs_commit_rcu_free); +} + struct nfs_write_data *nfs_writedata_alloc(size_t len) { unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -131,13 +137,19 @@ struct nfs_write_data *nfs_writedata_alloc(size_t len) return p; } -static void nfs_writedata_free(struct nfs_write_data *p) +static void nfs_writedata_rcu_free(struct rcu_head *head) { + struct nfs_write_data *p = container_of(head, struct nfs_write_data, task.u.tk_rcu); if (p && (p->pagevec != &p->page_array[0])) kfree(p->pagevec); mempool_free(p, nfs_wdata_mempool); } +static void nfs_writedata_free(struct nfs_write_data *wdata) +{ + call_rcu_bh(&wdata->task.u.tk_rcu, nfs_writedata_rcu_free); +} + void nfs_writedata_release(void *wdata) { nfs_writedata_free(wdata); @@ -258,7 +270,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, io_error: nfs_end_data_update(inode); end_page_writeback(page); - nfs_writedata_free(wdata); + nfs_writedata_release(wdata); return written ? written : result; } @@ -1043,7 +1055,7 @@ out_bad: while (!list_empty(&list)) { data = list_entry(list.next, struct nfs_write_data, pages); list_del(&data->pages); - nfs_writedata_free(data); + nfs_writedata_release(data); } nfs_mark_request_dirty(req); nfs_clear_page_writeback(req); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 625ffea98561..02f38189d180 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -428,11 +428,6 @@ extern int nfs_updatepage(struct file *, struct page *, unsigned int, unsigned extern int nfs_writeback_done(struct rpc_task *, struct nfs_write_data *); extern void nfs_writedata_release(void *); -#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) -struct nfs_write_data *nfs_commit_alloc(void); -void nfs_commit_free(struct nfs_write_data *p); -#endif - /* * Try to write back everything synchronously (but check the * return value!) @@ -440,6 +435,8 @@ void nfs_commit_free(struct nfs_write_data *p); extern int nfs_sync_inode_wait(struct inode *, unsigned long, unsigned int, int); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) extern int nfs_commit_inode(struct inode *, int); +extern struct nfs_write_data *nfs_commit_alloc(void); +extern void nfs_commit_free(struct nfs_write_data *wdata); extern void nfs_commit_release(void *wdata); #else static inline int diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 9fdb8c9d09f2..14fc813ddd0c 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -11,6 +11,7 @@ #include #include +#include #include #include #include @@ -85,6 +86,7 @@ struct rpc_task { union { struct work_struct tk_work; /* Async task work queue */ struct rpc_wait tk_wait; /* RPC wait */ + struct rcu_head tk_rcu; /* for task deletion */ } u; unsigned short tk_timeouts; /* maj timeouts */ diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 66d01365f3a5..6b808c03fb72 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -427,16 +427,19 @@ __rpc_default_timer(struct rpc_task *task) */ void rpc_wake_up_task(struct rpc_task *task) { + rcu_read_lock_bh(); if (rpc_start_wakeup(task)) { if (RPC_IS_QUEUED(task)) { struct rpc_wait_queue *queue = task->u.tk_wait.rpc_waitq; - spin_lock_bh(&queue->lock); + /* Note: we're already in a bh-safe context */ + spin_lock(&queue->lock); __rpc_do_wake_up_task(task); - spin_unlock_bh(&queue->lock); + spin_unlock(&queue->lock); } rpc_finish_wakeup(task); } + rcu_read_unlock_bh(); } /* @@ -499,14 +502,16 @@ struct rpc_task * rpc_wake_up_next(struct rpc_wait_queue *queue) struct rpc_task *task = NULL; dprintk("RPC: wake_up_next(%p \"%s\")\n", queue, rpc_qname(queue)); - spin_lock_bh(&queue->lock); + rcu_read_lock_bh(); + spin_lock(&queue->lock); if (RPC_IS_PRIORITY(queue)) task = __rpc_wake_up_next_priority(queue); else { task_for_first(task, &queue->tasks[0]) __rpc_wake_up_task(task); } - spin_unlock_bh(&queue->lock); + spin_unlock(&queue->lock); + rcu_read_unlock_bh(); return task; } @@ -522,7 +527,8 @@ void rpc_wake_up(struct rpc_wait_queue *queue) struct rpc_task *task, *next; struct list_head *head; - spin_lock_bh(&queue->lock); + rcu_read_lock_bh(); + spin_lock(&queue->lock); head = &queue->tasks[queue->maxpriority]; for (;;) { list_for_each_entry_safe(task, next, head, u.tk_wait.list) @@ -531,7 +537,8 @@ void rpc_wake_up(struct rpc_wait_queue *queue) break; head--; } - spin_unlock_bh(&queue->lock); + spin_unlock(&queue->lock); + rcu_read_unlock_bh(); } /** @@ -546,7 +553,8 @@ void rpc_wake_up_status(struct rpc_wait_queue *queue, int status) struct rpc_task *task, *next; struct list_head *head; - spin_lock_bh(&queue->lock); + rcu_read_lock_bh(); + spin_lock(&queue->lock); head = &queue->tasks[queue->maxpriority]; for (;;) { list_for_each_entry_safe(task, next, head, u.tk_wait.list) { @@ -557,7 +565,8 @@ void rpc_wake_up_status(struct rpc_wait_queue *queue, int status) break; head--; } - spin_unlock_bh(&queue->lock); + spin_unlock(&queue->lock); + rcu_read_unlock_bh(); } static void __rpc_atrun(struct rpc_task *task) @@ -817,8 +826,9 @@ rpc_alloc_task(void) return (struct rpc_task *)mempool_alloc(rpc_task_mempool, GFP_NOFS); } -static void rpc_free_task(struct rpc_task *task) +static void rpc_free_task(struct rcu_head *rcu) { + struct rpc_task *task = container_of(rcu, struct rpc_task, u.tk_rcu); dprintk("RPC: %4d freeing task\n", task->tk_pid); mempool_free(task, rpc_task_mempool); } @@ -872,7 +882,7 @@ void rpc_put_task(struct rpc_task *task) task->tk_client = NULL; } if (task->tk_flags & RPC_TASK_DYNAMIC) - rpc_free_task(task); + call_rcu_bh(&task->u.tk_rcu, rpc_free_task); if (tk_ops->rpc_release) tk_ops->rpc_release(calldata); } -- cgit v1.2.3 From 3e32a5d99a467b9d4d416323c8c292479b4915e5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 16 Nov 2006 11:37:27 -0500 Subject: SUNRPC: Give cloned RPC clients their own rpc_pipefs directory Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 1 + net/sunrpc/clnt.c | 23 +++++++++++++---------- 2 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index f6d1d646ce05..a1be89deb3af 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -53,6 +53,7 @@ struct rpc_clnt { struct dentry * cl_dentry; /* inode */ struct rpc_clnt * cl_parent; /* Points to parent of clones */ struct rpc_rtt cl_rtt_default; + struct rpc_program * cl_program; char cl_inline_name[32]; }; diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index de8bbfcd5030..8b78177e7575 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -144,6 +144,7 @@ static struct rpc_clnt * rpc_new_client(struct rpc_xprt *xprt, char *servname, s err = -ENOMEM; if (clnt->cl_metrics == NULL) goto out_no_stats; + clnt->cl_program = program; if (!xprt_bound(clnt->cl_xprt)) clnt->cl_autobind = 1; @@ -257,6 +258,7 @@ struct rpc_clnt * rpc_clone_client(struct rpc_clnt *clnt) { struct rpc_clnt *new; + int err = -ENOMEM; new = kmemdup(clnt, sizeof(*new), GFP_KERNEL); if (!new) @@ -266,6 +268,9 @@ rpc_clone_client(struct rpc_clnt *clnt) new->cl_metrics = rpc_alloc_iostats(clnt); if (new->cl_metrics == NULL) goto out_no_stats; + err = rpc_setup_pipedir(new, clnt->cl_program->pipe_dir_name); + if (err != 0) + goto out_no_path; new->cl_parent = clnt; atomic_inc(&clnt->cl_count); new->cl_xprt = xprt_get(clnt->cl_xprt); @@ -273,17 +278,17 @@ rpc_clone_client(struct rpc_clnt *clnt) new->cl_autobind = 0; new->cl_oneshot = 0; new->cl_dead = 0; - if (!IS_ERR(new->cl_dentry)) - dget(new->cl_dentry); rpc_init_rtt(&new->cl_rtt_default, clnt->cl_xprt->timeout.to_initval); if (new->cl_auth) atomic_inc(&new->cl_auth->au_count); return new; +out_no_path: + rpc_free_iostats(new->cl_metrics); out_no_stats: kfree(new); out_no_clnt: - printk(KERN_INFO "RPC: out of memory in %s\n", __FUNCTION__); - return ERR_PTR(-ENOMEM); + dprintk("RPC: %s returned error %d\n", __FUNCTION__, err); + return ERR_PTR(err); } /* @@ -336,16 +341,14 @@ rpc_destroy_client(struct rpc_clnt *clnt) rpcauth_destroy(clnt->cl_auth); clnt->cl_auth = NULL; } - if (clnt->cl_parent != clnt) { - if (!IS_ERR(clnt->cl_dentry)) - dput(clnt->cl_dentry); - rpc_destroy_client(clnt->cl_parent); - goto out_free; - } if (!IS_ERR(clnt->cl_dentry)) { rpc_rmdir(clnt->cl_dentry); rpc_put_mount(); } + if (clnt->cl_parent != clnt) { + rpc_destroy_client(clnt->cl_parent); + goto out_free; + } if (clnt->cl_server != clnt->cl_inline_name) kfree(clnt->cl_server); out_free: -- cgit v1.2.3 From bbd5a1f9fc9fad0f8725812d91c51b052e847de8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 18 Oct 2006 16:01:05 -0400 Subject: SUNRPC: Fix up missing BKL in asynchronous RPC callback functions Signed-off-by: Trond Myklebust --- include/linux/sunrpc/sched.h | 1 + net/sunrpc/clnt.c | 3 +-- net/sunrpc/sched.c | 15 +++++++++++---- 3 files changed, 13 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 14fc813ddd0c..b6b6ad6253b4 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -252,6 +252,7 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, void rpc_put_task(struct rpc_task *); void rpc_release_task(struct rpc_task *); void rpc_exit_task(struct rpc_task *); +void rpc_release_calldata(const struct rpc_call_ops *, void *); void rpc_killall_tasks(struct rpc_clnt *); int rpc_execute(struct rpc_task *); void rpc_init_priority_wait_queue(struct rpc_wait_queue *, const char *); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 8b78177e7575..37d345c04aa5 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -541,8 +541,7 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags, rpc_restore_sigmask(&oldset); return status; out_release: - if (tk_ops->rpc_release != NULL) - tk_ops->rpc_release(data); + rpc_release_calldata(tk_ops, data); return status; } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 6b808c03fb72..9c13050d23eb 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -608,6 +608,15 @@ void rpc_exit_task(struct rpc_task *task) } EXPORT_SYMBOL(rpc_exit_task); +void rpc_release_calldata(const struct rpc_call_ops *ops, void *calldata) +{ + if (ops->rpc_release != NULL) { + lock_kernel(); + ops->rpc_release(calldata); + unlock_kernel(); + } +} + /* * This is the RPC `scheduler' (or rather, the finite state machine). */ @@ -883,8 +892,7 @@ void rpc_put_task(struct rpc_task *task) } if (task->tk_flags & RPC_TASK_DYNAMIC) call_rcu_bh(&task->u.tk_rcu, rpc_free_task); - if (tk_ops->rpc_release) - tk_ops->rpc_release(calldata); + rpc_release_calldata(tk_ops, calldata); } EXPORT_SYMBOL(rpc_put_task); @@ -928,8 +936,7 @@ struct rpc_task *rpc_run_task(struct rpc_clnt *clnt, int flags, struct rpc_task *task; task = rpc_new_task(clnt, flags, ops, data); if (task == NULL) { - if (ops->rpc_release != NULL) - ops->rpc_release(data); + rpc_release_calldata(ops, data); return ERR_PTR(-ENOMEM); } atomic_inc(&task->tk_count); -- cgit v1.2.3 From 1e78957e0a8f882df6a3660b62f9aae441f54891 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 31 Aug 2006 15:09:19 -0400 Subject: SUNRPC: Clean up argument types in xdr.c Converts various integer buffer offsets and sizes to unsigned integer. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 12 +++---- net/sunrpc/xdr.c | 86 +++++++++++++++++++++++----------------------- 2 files changed, 49 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 9a527c364394..441b91dbafe5 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -139,10 +139,10 @@ xdr_adjust_iovec(struct kvec *iov, __be32 *p) */ extern void xdr_shift_buf(struct xdr_buf *, size_t); extern void xdr_buf_from_iov(struct kvec *, struct xdr_buf *); -extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, int, int); -extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, int); -extern int read_bytes_from_xdr_buf(struct xdr_buf *, int, void *, int); -extern int write_bytes_to_xdr_buf(struct xdr_buf *, int, void *, int); +extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, unsigned int, unsigned int); +extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, unsigned int); +extern int read_bytes_from_xdr_buf(struct xdr_buf *, unsigned int, void *, unsigned int); +extern int write_bytes_to_xdr_buf(struct xdr_buf *, unsigned int, void *, unsigned int); /* * Helper structure for copying from an sk_buff. @@ -160,8 +160,8 @@ extern int csum_partial_copy_to_xdr(struct xdr_buf *, struct sk_buff *); extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); -extern int xdr_encode_word(struct xdr_buf *, int, u32); -extern int xdr_decode_word(struct xdr_buf *, int, u32 *); +extern int xdr_encode_word(struct xdr_buf *, unsigned int, u32); +extern int xdr_decode_word(struct xdr_buf *, unsigned int, u32 *); struct xdr_array2_desc; typedef int (*xdr_xcode_elem_t)(struct xdr_array2_desc *desc, void *elem); diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 9022eb8b37ed..b474edbab3f1 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -640,41 +640,30 @@ xdr_buf_from_iov(struct kvec *iov, struct xdr_buf *buf) buf->buflen = buf->len = iov->iov_len; } -/* Sets subiov to the intersection of iov with the buffer of length len - * starting base bytes after iov. Indicates empty intersection by setting - * length of subiov to zero. Decrements len by length of subiov, sets base - * to zero (or decrements it by length of iov if subiov is empty). */ -static void -iov_subsegment(struct kvec *iov, struct kvec *subiov, int *base, int *len) -{ - if (*base > iov->iov_len) { - subiov->iov_base = NULL; - subiov->iov_len = 0; - *base -= iov->iov_len; - } else { - subiov->iov_base = iov->iov_base + *base; - subiov->iov_len = min(*len, (int)iov->iov_len - *base); - *base = 0; - } - *len -= subiov->iov_len; -} - /* Sets subbuf to the portion of buf of length len beginning base bytes * from the start of buf. Returns -1 if base of length are out of bounds. */ int xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf, - int base, int len) + unsigned int base, unsigned int len) { - int i; - subbuf->buflen = subbuf->len = len; - iov_subsegment(buf->head, subbuf->head, &base, &len); + if (base < buf->head[0].iov_len) { + subbuf->head[0].iov_base = buf->head[0].iov_base + base; + subbuf->head[0].iov_len = min_t(unsigned int, len, + buf->head[0].iov_len - base); + len -= subbuf->head[0].iov_len; + base = 0; + } else { + subbuf->head[0].iov_base = NULL; + subbuf->head[0].iov_len = 0; + base -= buf->head[0].iov_len; + } if (base < buf->page_len) { - i = (base + buf->page_base) >> PAGE_CACHE_SHIFT; - subbuf->pages = &buf->pages[i]; - subbuf->page_base = (base + buf->page_base) & ~PAGE_CACHE_MASK; - subbuf->page_len = min((int)buf->page_len - base, len); + subbuf->page_len = min(buf->page_len - base, len); + base += buf->page_base; + subbuf->page_base = base & ~PAGE_CACHE_MASK; + subbuf->pages = &buf->pages[base >> PAGE_CACHE_SHIFT]; len -= subbuf->page_len; base = 0; } else { @@ -682,7 +671,18 @@ xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf, subbuf->page_len = 0; } - iov_subsegment(buf->tail, subbuf->tail, &base, &len); + if (base < buf->tail[0].iov_len) { + subbuf->tail[0].iov_base = buf->tail[0].iov_base + base; + subbuf->tail[0].iov_len = min_t(unsigned int, len, + buf->tail[0].iov_len - base); + len -= subbuf->tail[0].iov_len; + base = 0; + } else { + subbuf->tail[0].iov_base = NULL; + subbuf->tail[0].iov_len = 0; + base -= buf->tail[0].iov_len; + } + if (base || len) return -1; return 0; @@ -690,25 +690,25 @@ xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf, /* obj is assumed to point to allocated memory of size at least len: */ int -read_bytes_from_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len) +read_bytes_from_xdr_buf(struct xdr_buf *buf, unsigned int base, void *obj, unsigned int len) { struct xdr_buf subbuf; - int this_len; + unsigned int this_len; int status; status = xdr_buf_subsegment(buf, &subbuf, base, len); if (status) goto out; - this_len = min(len, (int)subbuf.head[0].iov_len); + this_len = min_t(unsigned int, len, subbuf.head[0].iov_len); memcpy(obj, subbuf.head[0].iov_base, this_len); len -= this_len; obj += this_len; - this_len = min(len, (int)subbuf.page_len); + this_len = min_t(unsigned int, len, subbuf.page_len); if (this_len) _copy_from_pages(obj, subbuf.pages, subbuf.page_base, this_len); len -= this_len; obj += this_len; - this_len = min(len, (int)subbuf.tail[0].iov_len); + this_len = min_t(unsigned int, len, subbuf.tail[0].iov_len); memcpy(obj, subbuf.tail[0].iov_base, this_len); out: return status; @@ -716,32 +716,32 @@ out: /* obj is assumed to point to allocated memory of size at least len: */ int -write_bytes_to_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len) +write_bytes_to_xdr_buf(struct xdr_buf *buf, unsigned int base, void *obj, unsigned int len) { struct xdr_buf subbuf; - int this_len; + unsigned int this_len; int status; status = xdr_buf_subsegment(buf, &subbuf, base, len); if (status) goto out; - this_len = min(len, (int)subbuf.head[0].iov_len); + this_len = min_t(unsigned int, len, subbuf.head[0].iov_len); memcpy(subbuf.head[0].iov_base, obj, this_len); len -= this_len; obj += this_len; - this_len = min(len, (int)subbuf.page_len); + this_len = min_t(unsigned int, len, subbuf.page_len); if (this_len) _copy_to_pages(subbuf.pages, subbuf.page_base, obj, this_len); len -= this_len; obj += this_len; - this_len = min(len, (int)subbuf.tail[0].iov_len); + this_len = min_t(unsigned int, len, subbuf.tail[0].iov_len); memcpy(subbuf.tail[0].iov_base, obj, this_len); out: return status; } int -xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj) +xdr_decode_word(struct xdr_buf *buf, unsigned int base, u32 *obj) { __be32 raw; int status; @@ -754,7 +754,7 @@ xdr_decode_word(struct xdr_buf *buf, int base, u32 *obj) } int -xdr_encode_word(struct xdr_buf *buf, int base, u32 obj) +xdr_encode_word(struct xdr_buf *buf, unsigned int base, u32 obj) { __be32 raw = htonl(obj); @@ -766,10 +766,10 @@ xdr_encode_word(struct xdr_buf *buf, int base, u32 obj) * try to find space for it at the end of the tail, copy it there, and * set obj to point to it. */ int -xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, int offset) +xdr_buf_read_netobj(struct xdr_buf *buf, struct xdr_netobj *obj, unsigned int offset) { - u32 tail_offset = buf->head[0].iov_len + buf->page_len; - u32 obj_end_offset; + unsigned int tail_offset = buf->head[0].iov_len + buf->page_len; + unsigned int obj_end_offset; if (xdr_decode_word(buf, offset, &obj->len)) goto out; -- cgit v1.2.3 From e744cf2e3ab8535a8494a4cf0177de26f94586da Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 17 Oct 2006 14:44:24 -0400 Subject: SUNRPC: minor optimization of "xid" field in rpc_xprt Move the xid field in the rpc_xprt structure to be in the same cache line as the reserve_lock, since these are used at the same time. Test plan: None. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 60394fbc4c70..0321fb53bd02 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -157,11 +157,6 @@ struct rpc_xprt { unsigned char shutdown : 1, /* being shut down */ resvport : 1; /* use a reserved port */ - /* - * XID - */ - __u32 xid; /* Next XID value to use */ - /* * State of TCP reply receive stuff */ @@ -193,8 +188,8 @@ struct rpc_xprt { */ spinlock_t transport_lock; /* lock transport info */ spinlock_t reserve_lock; /* lock slot table */ + u32 xid; /* Next XID value to use */ struct rpc_task * snd_task; /* Task blocked in send */ - struct list_head recv; struct { -- cgit v1.2.3 From c8541ecdd5692bcfbcb5305cab9a873288d29175 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 17 Oct 2006 14:44:27 -0400 Subject: SUNRPC: Make the transport-specific setup routine allocate rpc_xprt Change the location where the rpc_xprt structure is allocated so each transport implementation can allocate a private area from the same chunk of memory. Note also that xprt->ops->destroy, rather than xprt_destroy, is now responsible for freeing rpc_xprt when the transport is destroyed. Test plan: Connectathon. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 4 +-- net/sunrpc/xprt.c | 31 +++++++------------ net/sunrpc/xprtsock.c | 73 ++++++++++++++++++++++++++++++--------------- 3 files changed, 61 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 0321fb53bd02..d7919010863d 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -265,8 +265,8 @@ void xprt_disconnect(struct rpc_xprt *xprt); /* * Socket transport setup operations */ -int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to); -int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to); +struct rpc_xprt * xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to); +struct rpc_xprt * xs_setup_tcp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to); /* * Reserved bit positions in xprt->state diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 80857470dc11..8cc2afa2942c 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -891,39 +891,25 @@ void xprt_set_timeout(struct rpc_timeout *to, unsigned int retr, unsigned long i */ struct rpc_xprt *xprt_create_transport(int proto, struct sockaddr *ap, size_t size, struct rpc_timeout *to) { - int result; struct rpc_xprt *xprt; struct rpc_rqst *req; - if ((xprt = kzalloc(sizeof(struct rpc_xprt), GFP_KERNEL)) == NULL) { - dprintk("RPC: xprt_create_transport: no memory\n"); - return ERR_PTR(-ENOMEM); - } - if (size <= sizeof(xprt->addr)) { - memcpy(&xprt->addr, ap, size); - xprt->addrlen = size; - } else { - kfree(xprt); - dprintk("RPC: xprt_create_transport: address too large\n"); - return ERR_PTR(-EBADF); - } - switch (proto) { case IPPROTO_UDP: - result = xs_setup_udp(xprt, to); + xprt = xs_setup_udp(ap, size, to); break; case IPPROTO_TCP: - result = xs_setup_tcp(xprt, to); + xprt = xs_setup_tcp(ap, size, to); break; default: printk(KERN_ERR "RPC: unrecognized transport protocol: %d\n", proto); return ERR_PTR(-EIO); } - if (result) { - kfree(xprt); - dprintk("RPC: xprt_create_transport: failed, %d\n", result); - return ERR_PTR(result); + if (IS_ERR(xprt)) { + dprintk("RPC: xprt_create_transport: failed, %ld\n", + -PTR_ERR(xprt)); + return xprt; } kref_init(&xprt->kref); @@ -969,8 +955,11 @@ static void xprt_destroy(struct kref *kref) dprintk("RPC: destroying transport %p\n", xprt); xprt->shutdown = 1; del_timer_sync(&xprt->timer); + + /* + * Tear down transport state and free the rpc_xprt + */ xprt->ops->destroy(xprt); - kfree(xprt); } /** diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 25620851d4bf..ec3462f141b4 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -518,6 +518,7 @@ static void xs_destroy(struct rpc_xprt *xprt) xs_close(xprt); xs_free_peer_addresses(xprt); kfree(xprt->slot); + kfree(xprt); } static inline struct rpc_xprt *xprt_from_sock(struct sock *sk) @@ -1339,26 +1340,53 @@ static struct rpc_xprt_ops xs_tcp_ops = { .print_stats = xs_tcp_print_stats, }; +static struct rpc_xprt *xs_setup_xprt(struct sockaddr *addr, size_t addrlen, unsigned int slot_table_size) +{ + struct rpc_xprt *xprt; + + if (addrlen > sizeof(xprt->addr)) { + dprintk("RPC: xs_setup_xprt: address too large\n"); + return ERR_PTR(-EBADF); + } + + xprt = kzalloc(sizeof(struct rpc_xprt), GFP_KERNEL); + if (xprt == NULL) { + dprintk("RPC: xs_setup_xprt: couldn't allocate rpc_xprt\n"); + return ERR_PTR(-ENOMEM); + } + + xprt->max_reqs = slot_table_size; + xprt->slot = kcalloc(xprt->max_reqs, sizeof(struct rpc_rqst), GFP_KERNEL); + if (xprt->slot == NULL) { + kfree(xprt); + dprintk("RPC: xs_setup_xprt: couldn't allocate slot table\n"); + return ERR_PTR(-ENOMEM); + } + + memcpy(&xprt->addr, addr, addrlen); + xprt->addrlen = addrlen; + xprt->port = xs_get_random_port(); + + return xprt; +} + /** * xs_setup_udp - Set up transport to use a UDP socket - * @xprt: transport to set up + * @addr: address of remote server + * @addrlen: length of address in bytes * @to: timeout parameters * */ -int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to) +struct rpc_xprt *xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to) { - size_t slot_table_size; - struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr; + struct rpc_xprt *xprt; - xprt->max_reqs = xprt_udp_slot_table_entries; - slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]); - xprt->slot = kzalloc(slot_table_size, GFP_KERNEL); - if (xprt->slot == NULL) - return -ENOMEM; + xprt = xs_setup_xprt(addr, addrlen, xprt_udp_slot_table_entries); + if (IS_ERR(xprt)) + return xprt; - if (ntohs(addr->sin_port) != 0) + if (ntohs(((struct sockaddr_in *)addr)->sin_port) != 0) xprt_set_bound(xprt); - xprt->port = xs_get_random_port(); xprt->prot = IPPROTO_UDP; xprt->tsh_size = 0; @@ -1382,29 +1410,26 @@ int xs_setup_udp(struct rpc_xprt *xprt, struct rpc_timeout *to) dprintk("RPC: set up transport to address %s\n", xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); - return 0; + return xprt; } /** * xs_setup_tcp - Set up transport to use a TCP socket - * @xprt: transport to set up + * @addr: address of remote server + * @addrlen: length of address in bytes * @to: timeout parameters * */ -int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to) +struct rpc_xprt *xs_setup_tcp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to) { - size_t slot_table_size; - struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr; + struct rpc_xprt *xprt; - xprt->max_reqs = xprt_tcp_slot_table_entries; - slot_table_size = xprt->max_reqs * sizeof(xprt->slot[0]); - xprt->slot = kzalloc(slot_table_size, GFP_KERNEL); - if (xprt->slot == NULL) - return -ENOMEM; + xprt = xs_setup_xprt(addr, addrlen, xprt_tcp_slot_table_entries); + if (IS_ERR(xprt)) + return xprt; - if (ntohs(addr->sin_port) != 0) + if (ntohs(((struct sockaddr_in *)addr)->sin_port) != 0) xprt_set_bound(xprt); - xprt->port = xs_get_random_port(); xprt->prot = IPPROTO_TCP; xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); @@ -1427,5 +1452,5 @@ int xs_setup_tcp(struct rpc_xprt *xprt, struct rpc_timeout *to) dprintk("RPC: set up transport to address %s\n", xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); - return 0; + return xprt; } -- cgit v1.2.3 From 3f442547b76bf9fb70d7aecc41cf1980459253c9 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sun, 17 Sep 2006 14:46:44 -0400 Subject: NFS: Clean up nfs_scan_dirty() Pass down struct writeback control. Signed-off-by: Trond Myklebust --- fs/nfs/pagelist.c | 76 ++++++++++++++++++++++++++++++++---------------- fs/nfs/write.c | 54 ++++++++++------------------------ include/linux/nfs_fs.h | 2 +- include/linux/nfs_page.h | 5 ++-- 4 files changed, 71 insertions(+), 66 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 829af323f288..e046c9b6a9dd 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -17,6 +17,7 @@ #include #include #include +#include #define NFS_PARANOIA 1 @@ -268,11 +269,10 @@ nfs_coalesce_requests(struct list_head *head, struct list_head *dst, #define NFS_SCAN_MAXENTRIES 16 /** - * nfs_scan_lock_dirty - Scan the radix tree for dirty requests - * @nfsi: NFS inode + * nfs_scan_dirty - Scan the radix tree for dirty requests + * @mapping: pointer to address space + * @wbc: writeback_control structure * @dst: Destination list - * @idx_start: lower bound of page->index to scan - * @npages: idx_start + npages sets the upper bound to scan. * * Moves elements from one of the inode request lists. * If the number of requests is set to 0, the entire address_space @@ -280,46 +280,72 @@ nfs_coalesce_requests(struct list_head *head, struct list_head *dst, * The requests are *not* checked to ensure that they form a contiguous set. * You must be holding the inode's req_lock when calling this function */ -int -nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, - unsigned long idx_start, unsigned int npages) +long nfs_scan_dirty(struct address_space *mapping, + struct writeback_control *wbc, + struct list_head *dst) { + struct nfs_inode *nfsi = NFS_I(mapping->host); struct nfs_page *pgvec[NFS_SCAN_MAXENTRIES]; struct nfs_page *req; - unsigned long idx_end; + pgoff_t idx_start, idx_end; + long count = wbc->nr_to_write; + long res = 0; int found, i; - int res; - res = 0; - if (npages == 0) - idx_end = ~0; - else - idx_end = idx_start + npages - 1; + if (nfsi->ndirty == 0 || count <= 0) + return 0; + if (wbc->range_cyclic) { + idx_start = 0; + idx_end = ULONG_MAX; + } else if (wbc->range_end == 0) { + idx_start = wbc->range_start >> PAGE_CACHE_SHIFT; + idx_end = ULONG_MAX; + } else { + idx_start = wbc->range_start >> PAGE_CACHE_SHIFT; + idx_end = wbc->range_end >> PAGE_CACHE_SHIFT; + } for (;;) { + unsigned int toscan = NFS_SCAN_MAXENTRIES; + + if (toscan > count) + toscan = count; found = radix_tree_gang_lookup_tag(&nfsi->nfs_page_tree, - (void **)&pgvec[0], idx_start, NFS_SCAN_MAXENTRIES, + (void **)&pgvec[0], idx_start, toscan, NFS_PAGE_TAG_DIRTY); + + /* Did we make progress? */ if (found <= 0) break; + for (i = 0; i < found; i++) { req = pgvec[i]; - if (req->wb_index > idx_end) + if (!wbc->range_cyclic && req->wb_index > idx_end) goto out; - idx_start = req->wb_index + 1; + /* Try to lock request and mark it for writeback */ + if (!nfs_set_page_writeback_locked(req)) + goto next; + radix_tree_tag_clear(&nfsi->nfs_page_tree, + req->wb_index, NFS_PAGE_TAG_DIRTY); + nfsi->ndirty--; + nfs_list_remove_request(req); + nfs_list_add_request(req, dst); + dec_zone_page_state(req->wb_page, NR_FILE_DIRTY); + res++; + if (res == LONG_MAX) + goto out; + count--; + if (count == 0) + goto out; - if (nfs_set_page_writeback_locked(req)) { - radix_tree_tag_clear(&nfsi->nfs_page_tree, - req->wb_index, NFS_PAGE_TAG_DIRTY); - nfs_list_remove_request(req); - nfs_list_add_request(req, dst); - dec_zone_page_state(req->wb_page, NR_FILE_DIRTY); - res++; - } +next: + idx_start = req->wb_index + 1; } } out: + wbc->nr_to_write = count; + WARN_ON ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty)); return res; } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 1d72096c4d22..dbc89fa7e9d5 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -79,7 +79,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context*, unsigned int, unsigned int); static int nfs_wait_on_write_congestion(struct address_space *, int); static int nfs_wait_on_requests(struct inode *, unsigned long, unsigned int); -static int nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how); +static long nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how); static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; @@ -400,10 +400,8 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) goto out; } err = nfs_commit_inode(inode, wb_priority(wbc)); - if (err > 0) { - wbc->nr_to_write -= err; + if (err > 0) err = 0; - } out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); @@ -607,31 +605,6 @@ static void nfs_cancel_commit_list(struct list_head *head) } } -/* - * nfs_scan_dirty - Scan an inode for dirty requests - * @inode: NFS inode to scan - * @dst: destination list - * @idx_start: lower bound of page->index to scan. - * @npages: idx_start + npages sets the upper bound to scan. - * - * Moves requests from the inode's dirty page list. - * The requests are *not* checked to ensure that they form a contiguous set. - */ -static int -nfs_scan_dirty(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages) -{ - struct nfs_inode *nfsi = NFS_I(inode); - int res = 0; - - if (nfsi->ndirty != 0) { - res = nfs_scan_lock_dirty(nfsi, dst, idx_start, npages); - nfsi->ndirty -= res; - if ((nfsi->ndirty == 0) != list_empty(&nfsi->dirty)) - printk(KERN_ERR "NFS: desynchronized value of nfs_i.ndirty.\n"); - } - return res; -} - #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) /* * nfs_scan_commit - Scan an inode for commit requests @@ -1467,22 +1440,19 @@ static inline int nfs_commit_list(struct inode *inode, struct list_head *head, i } #endif -static int nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how) +static long nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how) { struct nfs_inode *nfsi = NFS_I(mapping->host); LIST_HEAD(head); - pgoff_t index = wbc->range_start >> PAGE_CACHE_SHIFT; - unsigned long npages = 1 + (wbc->range_end >> PAGE_CACHE_SHIFT) - index; - int res; + long res; spin_lock(&nfsi->req_lock); - res = nfs_scan_dirty(mapping->host, &head, index, npages); + res = nfs_scan_dirty(mapping, wbc, &head); spin_unlock(&nfsi->req_lock); if (res) { int error = nfs_flush_list(mapping->host, &head, res, how); if (error < 0) return error; - wbc->nr_to_write -= res; } return res; } @@ -1506,13 +1476,21 @@ int nfs_commit_inode(struct inode *inode, int how) } #endif -int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, +long nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, unsigned int npages, int how) { struct nfs_inode *nfsi = NFS_I(inode); + struct address_space *mapping = inode->i_mapping; + struct writeback_control wbc = { + .bdi = mapping->backing_dev_info, + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_start = ((loff_t)idx_start) << PAGE_CACHE_SHIFT, + .range_end = ((loff_t)(idx_start + npages - 1)) << PAGE_CACHE_SHIFT, + }; LIST_HEAD(head); int nocommit = how & FLUSH_NOCOMMIT; - int pages, ret; + long pages, ret; how &= ~FLUSH_NOCOMMIT; spin_lock(&nfsi->req_lock); @@ -1520,7 +1498,7 @@ int nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, ret = nfs_wait_on_requests_locked(inode, idx_start, npages); if (ret != 0) continue; - pages = nfs_scan_dirty(inode, &head, idx_start, npages); + pages = nfs_scan_dirty(mapping, &wbc, &head); if (pages != 0) { spin_unlock(&nfsi->req_lock); if (how & FLUSH_INVALIDATE) { diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 02f38189d180..f8190ae9e3fb 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -432,7 +432,7 @@ extern void nfs_writedata_release(void *); * Try to write back everything synchronously (but check the * return value!) */ -extern int nfs_sync_inode_wait(struct inode *, unsigned long, unsigned int, int); +extern long nfs_sync_inode_wait(struct inode *, unsigned long, unsigned int, int); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) extern int nfs_commit_inode(struct inode *, int); extern struct nfs_write_data *nfs_commit_alloc(void); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 1f7bd287c230..38aa15fea638 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -60,8 +60,9 @@ extern void nfs_clear_request(struct nfs_page *req); extern void nfs_release_request(struct nfs_page *req); -extern int nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, - unsigned long idx_start, unsigned int npages); +extern long nfs_scan_dirty(struct address_space *mapping, + struct writeback_control *wbc, + struct list_head *dst); extern int nfs_scan_list(struct nfs_inode *nfsi, struct list_head *head, struct list_head *dst, unsigned long idx_start, unsigned int npages); extern int nfs_coalesce_requests(struct list_head *, struct list_head *, -- cgit v1.2.3 From 1c75950b9a2254ef08f986e00ad20266ae9ba7f1 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 9 Oct 2006 16:18:38 -0400 Subject: NFS: cleanup of nfs_sync_inode_wait() Allow callers to directly pass it a struct writeback_control. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 9 +++-- fs/nfs/inode.c | 2 +- fs/nfs/write.c | 93 +++++++++++++++++++++++++++++++++++++++++++------- include/linux/nfs_fs.h | 27 +++------------ 4 files changed, 92 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index cc93865cea93..d6ee60fc3ba6 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -307,11 +307,14 @@ static int nfs_commit_write(struct file *file, struct page *page, unsigned offse static void nfs_invalidate_page(struct page *page, unsigned long offset) { - struct inode *inode = page->mapping->host; + loff_t range_start, range_end; + if (offset != 0) + return; /* Cancel any unstarted writes on this page */ - if (offset == 0) - nfs_sync_inode_wait(inode, page->index, 1, FLUSH_INVALIDATE); + range_start = page_offset(page); + range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); + nfs_sync_mapping_range(page->mapping, range_start, range_end, FLUSH_INVALIDATE); } static int nfs_release_page(struct page *page, gfp_t gfp) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 08cc4c5919ab..7c32187f953e 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -422,7 +422,7 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) int err; /* Flush out writes to the server in order to update c/mtime */ - nfs_sync_inode_wait(inode, 0, 0, FLUSH_NOCOMMIT); + nfs_sync_mapping_range(inode->i_mapping, 0, 0, FLUSH_NOCOMMIT); /* * We may force a getattr if the user cares about atime. diff --git a/fs/nfs/write.c b/fs/nfs/write.c index dbc89fa7e9d5..310fdeca6250 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -80,6 +80,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context*, static int nfs_wait_on_write_congestion(struct address_space *, int); static int nfs_wait_on_requests(struct inode *, unsigned long, unsigned int); static long nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how); +static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how); static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; @@ -1476,29 +1477,38 @@ int nfs_commit_inode(struct inode *inode, int how) } #endif -long nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, - unsigned int npages, int how) +long nfs_sync_mapping_wait(struct address_space *mapping, struct writeback_control *wbc, int how) { + struct inode *inode = mapping->host; struct nfs_inode *nfsi = NFS_I(inode); - struct address_space *mapping = inode->i_mapping; - struct writeback_control wbc = { - .bdi = mapping->backing_dev_info, - .sync_mode = WB_SYNC_ALL, - .nr_to_write = LONG_MAX, - .range_start = ((loff_t)idx_start) << PAGE_CACHE_SHIFT, - .range_end = ((loff_t)(idx_start + npages - 1)) << PAGE_CACHE_SHIFT, - }; + unsigned long idx_start, idx_end; + unsigned int npages = 0; LIST_HEAD(head); int nocommit = how & FLUSH_NOCOMMIT; long pages, ret; + /* FIXME */ + if (wbc->range_cyclic) + idx_start = 0; + else { + idx_start = wbc->range_start >> PAGE_CACHE_SHIFT; + idx_end = wbc->range_end >> PAGE_CACHE_SHIFT; + if (idx_end > idx_start) { + unsigned long l_npages = 1 + idx_end - idx_start; + npages = l_npages; + if (sizeof(npages) != sizeof(l_npages) && + (unsigned long)npages != l_npages) + npages = 0; + } + } how &= ~FLUSH_NOCOMMIT; spin_lock(&nfsi->req_lock); do { + wbc->pages_skipped = 0; ret = nfs_wait_on_requests_locked(inode, idx_start, npages); if (ret != 0) continue; - pages = nfs_scan_dirty(mapping, &wbc, &head); + pages = nfs_scan_dirty(mapping, wbc, &head); if (pages != 0) { spin_unlock(&nfsi->req_lock); if (how & FLUSH_INVALIDATE) { @@ -1509,11 +1519,16 @@ long nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, spin_lock(&nfsi->req_lock); continue; } + if (wbc->pages_skipped != 0) + continue; if (nocommit) break; pages = nfs_scan_commit(inode, &head, idx_start, npages); - if (pages == 0) + if (pages == 0) { + if (wbc->pages_skipped != 0) + continue; break; + } if (how & FLUSH_INVALIDATE) { spin_unlock(&nfsi->req_lock); nfs_cancel_commit_list(&head); @@ -1530,6 +1545,60 @@ long nfs_sync_inode_wait(struct inode *inode, unsigned long idx_start, return ret; } +/* + * flush the inode to disk. + */ +int nfs_wb_all(struct inode *inode) +{ + struct address_space *mapping = inode->i_mapping; + struct writeback_control wbc = { + .bdi = mapping->backing_dev_info, + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_cyclic = 1, + }; + int ret; + + ret = nfs_sync_mapping_wait(mapping, &wbc, 0); + if (ret >= 0) + return 0; + return ret; +} + +int nfs_sync_mapping_range(struct address_space *mapping, loff_t range_start, loff_t range_end, int how) +{ + struct writeback_control wbc = { + .bdi = mapping->backing_dev_info, + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_start = range_start, + .range_end = range_end, + }; + int ret; + + ret = nfs_sync_mapping_wait(mapping, &wbc, how); + if (ret >= 0) + return 0; + return ret; +} + +static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how) +{ + loff_t range_start = page_offset(page); + loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); + + return nfs_sync_mapping_range(inode->i_mapping, range_start, range_end, how | FLUSH_STABLE); +} + +/* + * Write back all requests on one page - we do this before reading it. + */ +int nfs_wb_page(struct inode *inode, struct page* page) +{ + return nfs_wb_page_priority(inode, page, 0); +} + + int __init nfs_init_writepagecache(void) { nfs_wdata_cachep = kmem_cache_create("nfs_write_data", diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f8190ae9e3fb..f2ec9be1e22f 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -432,7 +432,10 @@ extern void nfs_writedata_release(void *); * Try to write back everything synchronously (but check the * return value!) */ -extern long nfs_sync_inode_wait(struct inode *, unsigned long, unsigned int, int); +extern long nfs_sync_mapping_wait(struct address_space *, struct writeback_control *, int); +extern int nfs_sync_mapping_range(struct address_space *, loff_t, loff_t, int); +extern int nfs_wb_all(struct inode *inode); +extern int nfs_wb_page(struct inode *inode, struct page* page); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) extern int nfs_commit_inode(struct inode *, int); extern struct nfs_write_data *nfs_commit_alloc(void); @@ -452,28 +455,6 @@ nfs_have_writebacks(struct inode *inode) return NFS_I(inode)->npages != 0; } -static inline int -nfs_wb_all(struct inode *inode) -{ - int error = nfs_sync_inode_wait(inode, 0, 0, 0); - return (error < 0) ? error : 0; -} - -/* - * Write back all requests on one page - we do this before reading it. - */ -static inline int nfs_wb_page_priority(struct inode *inode, struct page* page, int how) -{ - int error = nfs_sync_inode_wait(inode, page->index, 1, - how | FLUSH_STABLE); - return (error < 0) ? error : 0; -} - -static inline int nfs_wb_page(struct inode *inode, struct page* page) -{ - return nfs_wb_page_priority(inode, page, 0); -} - /* * Allocate nfs_write_data structures */ -- cgit v1.2.3 From 200baa2112012dd8a13db9da3ee6885403f9c013 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Dec 2006 00:35:40 -0500 Subject: NFS: Remove nfs_writepage_sync() Maintaining two parallel ways of doing synchronous writes is rather pointless. This patch gets rid of the legacy nfs_writepage_sync(), and replaces it with the faster asynchronous writes. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 6 +++ fs/nfs/nfs3proc.c | 47 ----------------------- fs/nfs/nfs4proc.c | 85 ---------------------------------------- fs/nfs/proc.c | 28 -------------- fs/nfs/write.c | 100 ++---------------------------------------------- include/linux/nfs_xdr.h | 2 - 6 files changed, 10 insertions(+), 258 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index d6ee60fc3ba6..143a19037ce8 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -378,6 +378,12 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov, nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count); result = generic_file_aio_write(iocb, iov, nr_segs, pos); + /* Return error values for O_SYNC and IS_SYNC() */ + if (result >= 0 && (IS_SYNC(inode) || (iocb->ki_filp->f_flags & O_SYNC))) { + int err = nfs_fsync(iocb->ki_filp, dentry, 1); + if (err < 0) + result = err; + } out: return result; diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 7fba06306b82..510ae524f3fd 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -276,51 +276,6 @@ static int nfs3_proc_read(struct nfs_read_data *rdata) return status; } -static int nfs3_proc_write(struct nfs_write_data *wdata) -{ - int rpcflags = wdata->flags; - struct inode * inode = wdata->inode; - struct nfs_fattr * fattr = wdata->res.fattr; - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_WRITE], - .rpc_argp = &wdata->args, - .rpc_resp = &wdata->res, - .rpc_cred = wdata->cred, - }; - int status; - - dprintk("NFS call write %d @ %Ld\n", wdata->args.count, - (long long) wdata->args.offset); - nfs_fattr_init(fattr); - status = rpc_call_sync(NFS_CLIENT(inode), &msg, rpcflags); - if (status >= 0) - nfs_post_op_update_inode(inode, fattr); - dprintk("NFS reply write: %d\n", status); - return status < 0? status : wdata->res.count; -} - -static int nfs3_proc_commit(struct nfs_write_data *cdata) -{ - struct inode * inode = cdata->inode; - struct nfs_fattr * fattr = cdata->res.fattr; - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT], - .rpc_argp = &cdata->args, - .rpc_resp = &cdata->res, - .rpc_cred = cdata->cred, - }; - int status; - - dprintk("NFS call commit %d @ %Ld\n", cdata->args.count, - (long long) cdata->args.offset); - nfs_fattr_init(fattr); - status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); - if (status >= 0) - nfs_post_op_update_inode(inode, fattr); - dprintk("NFS reply commit: %d\n", status); - return status; -} - /* * Create a regular file. * For now, we don't implement O_EXCL. @@ -901,8 +856,6 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .access = nfs3_proc_access, .readlink = nfs3_proc_readlink, .read = nfs3_proc_read, - .write = nfs3_proc_write, - .commit = nfs3_proc_commit, .create = nfs3_proc_create, .remove = nfs3_proc_remove, .unlink_setup = nfs3_proc_unlink_setup, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 4b5fc714d786..ee458aeab24a 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1775,89 +1775,6 @@ static int nfs4_proc_read(struct nfs_read_data *rdata) return err; } -static int _nfs4_proc_write(struct nfs_write_data *wdata) -{ - int rpcflags = wdata->flags; - struct inode *inode = wdata->inode; - struct nfs_fattr *fattr = wdata->res.fattr; - struct nfs_server *server = NFS_SERVER(inode); - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE], - .rpc_argp = &wdata->args, - .rpc_resp = &wdata->res, - .rpc_cred = wdata->cred, - }; - int status; - - dprintk("NFS call write %d @ %Ld\n", wdata->args.count, - (long long) wdata->args.offset); - - wdata->args.bitmask = server->attr_bitmask; - wdata->res.server = server; - wdata->timestamp = jiffies; - nfs_fattr_init(fattr); - status = rpc_call_sync(server->client, &msg, rpcflags); - dprintk("NFS reply write: %d\n", status); - if (status < 0) - return status; - renew_lease(server, wdata->timestamp); - nfs_post_op_update_inode(inode, fattr); - return wdata->res.count; -} - -static int nfs4_proc_write(struct nfs_write_data *wdata) -{ - struct nfs4_exception exception = { }; - int err; - do { - err = nfs4_handle_exception(NFS_SERVER(wdata->inode), - _nfs4_proc_write(wdata), - &exception); - } while (exception.retry); - return err; -} - -static int _nfs4_proc_commit(struct nfs_write_data *cdata) -{ - struct inode *inode = cdata->inode; - struct nfs_fattr *fattr = cdata->res.fattr; - struct nfs_server *server = NFS_SERVER(inode); - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT], - .rpc_argp = &cdata->args, - .rpc_resp = &cdata->res, - .rpc_cred = cdata->cred, - }; - int status; - - dprintk("NFS call commit %d @ %Ld\n", cdata->args.count, - (long long) cdata->args.offset); - - cdata->args.bitmask = server->attr_bitmask; - cdata->res.server = server; - cdata->timestamp = jiffies; - nfs_fattr_init(fattr); - status = rpc_call_sync(server->client, &msg, 0); - if (status >= 0) - renew_lease(server, cdata->timestamp); - dprintk("NFS reply commit: %d\n", status); - if (status >= 0) - nfs_post_op_update_inode(inode, fattr); - return status; -} - -static int nfs4_proc_commit(struct nfs_write_data *cdata) -{ - struct nfs4_exception exception = { }; - int err; - do { - err = nfs4_handle_exception(NFS_SERVER(cdata->inode), - _nfs4_proc_commit(cdata), - &exception); - } while (exception.retry); - return err; -} - /* * Got race? * We will need to arrange for the VFS layer to provide an atomic open. @@ -3730,8 +3647,6 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .access = nfs4_proc_access, .readlink = nfs4_proc_readlink, .read = nfs4_proc_read, - .write = nfs4_proc_write, - .commit = nfs4_proc_commit, .create = nfs4_proc_create, .remove = nfs4_proc_remove, .unlink_setup = nfs4_proc_unlink_setup, diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 7a9013dd5d0a..10f5e80ca157 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -215,32 +215,6 @@ static int nfs_proc_read(struct nfs_read_data *rdata) return status; } -static int nfs_proc_write(struct nfs_write_data *wdata) -{ - int flags = wdata->flags; - struct inode * inode = wdata->inode; - struct nfs_fattr * fattr = wdata->res.fattr; - struct rpc_message msg = { - .rpc_proc = &nfs_procedures[NFSPROC_WRITE], - .rpc_argp = &wdata->args, - .rpc_resp = &wdata->res, - .rpc_cred = wdata->cred, - }; - int status; - - dprintk("NFS call write %d @ %Ld\n", wdata->args.count, - (long long) wdata->args.offset); - nfs_fattr_init(fattr); - status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags); - if (status >= 0) { - nfs_post_op_update_inode(inode, fattr); - wdata->res.count = wdata->args.count; - wdata->verf.committed = NFS_FILE_SYNC; - } - dprintk("NFS reply write: %d\n", status); - return status < 0? status : wdata->res.count; -} - static int nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, int flags, struct nameidata *nd) @@ -693,8 +667,6 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .access = NULL, /* access */ .readlink = nfs_proc_readlink, .read = nfs_proc_read, - .write = nfs_proc_write, - .commit = NULL, /* commit */ .create = nfs_proc_create, .remove = nfs_proc_remove, .unlink_setup = nfs_proc_unlink_setup, diff --git a/fs/nfs/write.c b/fs/nfs/write.c index de9a16a8f7e4..f0720b544b12 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -210,78 +210,6 @@ static void nfs_mark_uptodate(struct page *page, unsigned int base, unsigned int SetPageUptodate(page); } -/* - * Write a page synchronously. - * Offset is the data offset within the page. - */ -static int nfs_writepage_sync(struct nfs_open_context *ctx, struct page *page, - unsigned int offset, unsigned int count, int how) -{ - struct inode *inode = page->mapping->host; - unsigned int wsize = NFS_SERVER(inode)->wsize; - int result, written = 0; - struct nfs_write_data *wdata; - - wdata = nfs_writedata_alloc(wsize); - if (!wdata) - return -ENOMEM; - - wdata->flags = how; - wdata->cred = ctx->cred; - wdata->inode = inode; - wdata->args.fh = NFS_FH(inode); - wdata->args.context = ctx; - wdata->args.pages = &page; - wdata->args.stable = NFS_FILE_SYNC; - wdata->args.pgbase = offset; - wdata->args.count = wsize; - wdata->res.fattr = &wdata->fattr; - wdata->res.verf = &wdata->verf; - - dprintk("NFS: nfs_writepage_sync(%s/%Ld %d@%Ld)\n", - inode->i_sb->s_id, - (long long)NFS_FILEID(inode), - count, (long long)(page_offset(page) + offset)); - - set_page_writeback(page); - nfs_begin_data_update(inode); - do { - if (count < wsize) - wdata->args.count = count; - wdata->args.offset = page_offset(page) + wdata->args.pgbase; - - result = NFS_PROTO(inode)->write(wdata); - - if (result < 0) { - /* Must mark the page invalid after I/O error */ - ClearPageUptodate(page); - goto io_error; - } - if (result < wdata->args.count) - printk(KERN_WARNING "NFS: short write, count=%u, result=%d\n", - wdata->args.count, result); - - wdata->args.offset += result; - wdata->args.pgbase += result; - written += result; - count -= result; - nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, result); - } while (count); - /* Update file length */ - nfs_grow_file(page, offset, written); - /* Set the PG_uptodate flag? */ - nfs_mark_uptodate(page, offset, written); - - if (PageError(page)) - ClearPageError(page); - -io_error: - nfs_end_data_update(inode); - end_page_writeback(page); - nfs_writedata_release(wdata); - return written ? written : result; -} - static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, unsigned int offset, unsigned int count) { @@ -342,22 +270,12 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc) err = -EBADF; goto out; } - lock_kernel(); - if (!IS_SYNC(inode)) { - err = nfs_writepage_setup(ctx, page, 0, offset); - if (!wbc->for_writepages) - nfs_flush_mapping(page->mapping, wbc, wb_priority(wbc)); - } else { - err = nfs_writepage_sync(ctx, page, 0, offset, wb_priority(wbc)); - if (err >= 0) { - if (err != offset) - redirty_page_for_writepage(wbc, page); - err = 0; - } - } - unlock_kernel(); + err = nfs_writepage_setup(ctx, page, 0, offset); put_nfs_open_context(ctx); + out: + if (!wbc->for_writepages) + nfs_flush_mapping(page->mapping, wbc, wb_priority(wbc)); unlock_page(page); return err; } @@ -777,16 +695,6 @@ int nfs_updatepage(struct file *file, struct page *page, file->f_dentry->d_name.name, count, (long long)(page_offset(page) +offset)); - if (IS_SYNC(inode)) { - status = nfs_writepage_sync(ctx, page, offset, count, 0); - if (status > 0) { - if (offset == 0 && status == PAGE_CACHE_SIZE) - SetPageUptodate(page); - return 0; - } - return status; - } - /* If we're not using byte range locks, and we know the page * is entirely in cache, it may be more efficient to avoid * fragmenting write requests. diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 768c1ad5ff6f..9ee9da5e1cc9 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -785,8 +785,6 @@ struct nfs_rpc_ops { int (*readlink)(struct inode *, struct page *, unsigned int, unsigned int); int (*read) (struct nfs_read_data *); - int (*write) (struct nfs_write_data *); - int (*commit) (struct nfs_write_data *); int (*create) (struct inode *, struct dentry *, struct iattr *, int, struct nameidata *); int (*remove) (struct inode *, struct qstr *); -- cgit v1.2.3 From 1a54533ec8d92a5edae97ec6ae10023ee71c4b46 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Dec 2006 00:35:40 -0500 Subject: NFS: Add nfs_set_page_dirty() We will want to allow nfs_writepage() to distinguish between pages that have been marked as dirty by the VM, and those that have been marked as dirty by nfs_updatepage(). In the former case, the entire page will want to be written out, and so any requests that were pending need to be flushed out first. Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 2 +- fs/nfs/write.c | 49 +++++++++++++++++++++++++++++++++++------------- include/linux/nfs_fs.h | 1 + include/linux/nfs_page.h | 1 + 4 files changed, 39 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 143a19037ce8..c2fe3bd83ab1 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -331,7 +331,7 @@ static int nfs_release_page(struct page *page, gfp_t gfp) const struct address_space_operations nfs_file_aops = { .readpage = nfs_readpage, .readpages = nfs_readpages, - .set_page_dirty = __set_page_dirty_nobuffers, + .set_page_dirty = nfs_set_page_dirty, .writepage = nfs_writepage, .writepages = nfs_writepages, .prepare_write = nfs_prepare_write, diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f0720b544b12..266fea71cfc0 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -251,16 +251,23 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc) { struct nfs_open_context *ctx; struct inode *inode = page->mapping->host; + struct nfs_page *req; unsigned offset; - int err; + int err = 0; nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); nfs_add_stats(inode, NFSIOS_WRITEPAGES, 1); - /* Ensure we've flushed out any previous writes */ - nfs_wb_page_priority(inode, page, wb_priority(wbc)); + req = nfs_page_find_request(page); + if (req != NULL) { + int flushme = test_bit(PG_NEED_FLUSH, &req->wb_flags); + nfs_release_request(req); + if (!flushme) + goto out; + /* Ensure we've flushed out the invalid write */ + nfs_wb_page_priority(inode, page, wb_priority(wbc)); + } - err = 0; offset = nfs_page_length(page); if (!offset) goto out; @@ -655,7 +662,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page) { struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data; struct nfs_page *req; - int status = 0; + int do_flush, status; /* * Look for a request corresponding to this page. If there * is one, and it belongs to another file, we flush it out @@ -664,15 +671,18 @@ int nfs_flush_incompatible(struct file *file, struct page *page) * Also do the same if we find a request from an existing * dropped page. */ - req = nfs_page_find_request(page); - if (req != NULL) { - int do_flush = req->wb_page != page || req->wb_context != ctx; - + do { + req = nfs_page_find_request(page); + if (req == NULL) + return 0; + do_flush = req->wb_page != page || req->wb_context != ctx + || test_bit(PG_NEED_FLUSH, &req->wb_flags); nfs_release_request(req); - if (do_flush) - status = nfs_wb_page(page->mapping->host, page); - } - return (status < 0) ? status : 0; + if (!do_flush) + return 0; + status = nfs_wb_page(page->mapping->host, page); + } while (status == 0); + return status; } /* @@ -1437,6 +1447,19 @@ int nfs_wb_page(struct inode *inode, struct page* page) return nfs_wb_page_priority(inode, page, 0); } +int nfs_set_page_dirty(struct page *page) +{ + struct nfs_page *req; + + req = nfs_page_find_request(page); + if (req != NULL) { + /* Mark any existing write requests for flushing */ + set_bit(PG_NEED_FLUSH, &req->wb_flags); + nfs_release_request(req); + } + return __set_page_dirty_nobuffers(page); +} + int __init nfs_init_writepagecache(void) { diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f2ec9be1e22f..1b38c2085a53 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -427,6 +427,7 @@ extern int nfs_flush_incompatible(struct file *file, struct page *page); extern int nfs_updatepage(struct file *, struct page *, unsigned int, unsigned int); extern int nfs_writeback_done(struct rpc_task *, struct nfs_write_data *); extern void nfs_writedata_release(void *); +extern int nfs_set_page_dirty(struct page *); /* * Try to write back everything synchronously (but check the diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 38aa15fea638..d111be639140 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -30,6 +30,7 @@ #define PG_BUSY 0 #define PG_NEED_COMMIT 1 #define PG_NEED_RESCHED 2 +#define PG_NEED_FLUSH 3 struct nfs_inode; struct nfs_page { -- cgit v1.2.3 From 4d770ccf4257b23a7ca2a85de1b1c22657b581d8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Dec 2006 00:35:41 -0500 Subject: NFS: Ensure that nfs_wb_page() calls writepage when necessary. Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 34 ++++++++++++++++++++++++++++++---- include/linux/nfs_fs.h | 1 + 2 files changed, 31 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 266fea71cfc0..0eca6a542106 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -247,7 +247,7 @@ static int wb_priority(struct writeback_control *wbc) /* * Write an mmapped page to the server. */ -int nfs_writepage(struct page *page, struct writeback_control *wbc) +static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc) { struct nfs_open_context *ctx; struct inode *inode = page->mapping->host; @@ -265,7 +265,7 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc) if (!flushme) goto out; /* Ensure we've flushed out the invalid write */ - nfs_wb_page_priority(inode, page, wb_priority(wbc)); + nfs_wb_page_priority(inode, page, wb_priority(wbc) | FLUSH_STABLE | FLUSH_NOWRITEPAGE); } offset = nfs_page_length(page); @@ -283,6 +283,14 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc) out: if (!wbc->for_writepages) nfs_flush_mapping(page->mapping, wbc, wb_priority(wbc)); + return err; +} + +int nfs_writepage(struct page *page, struct writeback_control *wbc) +{ + int err; + + err = nfs_writepage_locked(page, wbc); unlock_page(page); return err; } @@ -1435,8 +1443,26 @@ static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how) { loff_t range_start = page_offset(page); loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); + struct writeback_control wbc = { + .bdi = page->mapping->backing_dev_info, + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_start = range_start, + .range_end = range_end, + }; + int ret; - return nfs_sync_mapping_range(inode->i_mapping, range_start, range_end, how | FLUSH_STABLE); + BUG_ON(!PageLocked(page)); + if (!(how & FLUSH_NOWRITEPAGE) && clear_page_dirty_for_io(page)) { + ret = nfs_writepage_locked(page, &wbc); + if (ret < 0) + goto out; + } + ret = nfs_sync_mapping_wait(page->mapping, &wbc, how); + if (ret >= 0) + return 0; +out: + return ret; } /* @@ -1444,7 +1470,7 @@ static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how) */ int nfs_wb_page(struct inode *inode, struct page* page) { - return nfs_wb_page_priority(inode, page, 0); + return nfs_wb_page_priority(inode, page, FLUSH_STABLE); } int nfs_set_page_dirty(struct page *page) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 1b38c2085a53..3de334c72b1f 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -33,6 +33,7 @@ #define FLUSH_HIGHPRI 16 /* high priority memory reclaim flush */ #define FLUSH_NOCOMMIT 32 /* Don't send the NFSv3/v4 COMMIT */ #define FLUSH_INVALIDATE 64 /* Invalidate the page cache */ +#define FLUSH_NOWRITEPAGE 128 /* Don't call writepage() */ #ifdef __KERNEL__ -- cgit v1.2.3 From e261f51f25b98c213e0b3d7f2109b117d714f69d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Dec 2006 00:35:41 -0500 Subject: NFS: Make nfs_updatepage() mark the page as dirty. This will ensure that we can call set_page_writeback() from within nfs_writepage(), which is always called with the page lock set. Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 73 +++++++++++++++++++++++++++++++++++++----------- include/linux/nfs_page.h | 1 + 2 files changed, 57 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 0eca6a542106..130528d09a26 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -77,6 +77,7 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context*, struct page *, unsigned int, unsigned int); +static void nfs_mark_request_dirty(struct nfs_page *req); static int nfs_wait_on_write_congestion(struct address_space *, int); static int nfs_wait_on_requests(struct inode *, unsigned long, unsigned int); static long nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how); @@ -244,6 +245,48 @@ static int wb_priority(struct writeback_control *wbc) return 0; } +/* + * Find an associated nfs write request, and prepare to flush it out + * Returns 1 if there was no write request, or if the request was + * already tagged by nfs_set_page_dirty.Returns 0 if the request + * was not tagged. + * May also return an error if the user signalled nfs_wait_on_request(). + */ +static int nfs_page_mark_flush(struct page *page) +{ + struct nfs_page *req; + spinlock_t *req_lock = &NFS_I(page->mapping->host)->req_lock; + int ret; + + spin_lock(req_lock); + for(;;) { + req = nfs_page_find_request_locked(page); + if (req == NULL) { + spin_unlock(req_lock); + return 1; + } + if (nfs_lock_request_dontget(req)) + break; + /* Note: If we hold the page lock, as is the case in nfs_writepage, + * then the call to nfs_lock_request_dontget() will always + * succeed provided that someone hasn't already marked the + * request as dirty (in which case we don't care). + */ + spin_unlock(req_lock); + ret = nfs_wait_on_request(req); + nfs_release_request(req); + if (ret != 0) + return ret; + spin_lock(req_lock); + } + spin_unlock(req_lock); + if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) + nfs_mark_request_dirty(req); + ret = test_bit(PG_NEED_FLUSH, &req->wb_flags); + nfs_unlock_request(req); + return ret; +} + /* * Write an mmapped page to the server. */ @@ -251,23 +294,16 @@ static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc { struct nfs_open_context *ctx; struct inode *inode = page->mapping->host; - struct nfs_page *req; unsigned offset; - int err = 0; + int err; nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); nfs_add_stats(inode, NFSIOS_WRITEPAGES, 1); - req = nfs_page_find_request(page); - if (req != NULL) { - int flushme = test_bit(PG_NEED_FLUSH, &req->wb_flags); - nfs_release_request(req); - if (!flushme) - goto out; - /* Ensure we've flushed out the invalid write */ - nfs_wb_page_priority(inode, page, wb_priority(wbc) | FLUSH_STABLE | FLUSH_NOWRITEPAGE); - } - + err = nfs_page_mark_flush(page); + if (err <= 0) + goto out; + err = 0; offset = nfs_page_length(page); if (!offset) goto out; @@ -279,7 +315,11 @@ static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc } err = nfs_writepage_setup(ctx, page, 0, offset); put_nfs_open_context(ctx); - + if (err != 0) + goto out; + err = nfs_page_mark_flush(page); + if (err > 0) + err = 0; out: if (!wbc->for_writepages) nfs_flush_mapping(page->mapping, wbc, wb_priority(wbc)); @@ -409,8 +449,7 @@ nfs_mark_request_dirty(struct nfs_page *req) static inline int nfs_dirty_request(struct nfs_page *req) { - struct nfs_inode *nfsi = NFS_I(req->wb_context->dentry->d_inode); - return !list_empty(&req->wb_list) && req->wb_list_head == &nfsi->dirty; + return test_bit(PG_FLUSHING, &req->wb_flags) == 0; } #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) @@ -628,7 +667,6 @@ static struct nfs_page * nfs_update_request(struct nfs_open_context* ctx, return ERR_PTR(error); } spin_unlock(&nfsi->req_lock); - nfs_mark_request_dirty(new); return new; } spin_unlock(&nfsi->req_lock); @@ -684,7 +722,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page) if (req == NULL) return 0; do_flush = req->wb_page != page || req->wb_context != ctx - || test_bit(PG_NEED_FLUSH, &req->wb_flags); + || !nfs_dirty_request(req); nfs_release_request(req); if (!do_flush) return 0; @@ -723,6 +761,7 @@ int nfs_updatepage(struct file *file, struct page *page, } status = nfs_writepage_setup(ctx, page, offset, count); + __set_page_dirty_nobuffers(page); dprintk("NFS: nfs_updatepage returns %d (isize %Ld)\n", status, (long long)i_size_read(inode)); diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index d111be639140..2e555d49c9b7 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -31,6 +31,7 @@ #define PG_NEED_COMMIT 1 #define PG_NEED_RESCHED 2 #define PG_NEED_FLUSH 3 +#define PG_FLUSHING 4 struct nfs_inode; struct nfs_page { -- cgit v1.2.3 From 61822ab5e3ed09fcfc49e37227b655202adf6130 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 5 Dec 2006 00:35:42 -0500 Subject: NFS: Ensure we only call set_page_writeback() under the page lock Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 6 +----- fs/nfs/write.c | 38 ++++++++++++++++++++++++++++---------- include/linux/nfs_fs.h | 1 + 3 files changed, 30 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/file.c b/fs/nfs/file.c index c2fe3bd83ab1..238fb6641aae 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -307,14 +307,10 @@ static int nfs_commit_write(struct file *file, struct page *page, unsigned offse static void nfs_invalidate_page(struct page *page, unsigned long offset) { - loff_t range_start, range_end; - if (offset != 0) return; /* Cancel any unstarted writes on this page */ - range_start = page_offset(page); - range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); - nfs_sync_mapping_range(page->mapping, range_start, range_end, FLUSH_INVALIDATE); + nfs_wb_page_priority(page->mapping->host, page, FLUSH_INVALIDATE); } static int nfs_release_page(struct page *page, gfp_t gfp) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 130528d09a26..bd4dff9dbd69 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -81,7 +81,6 @@ static void nfs_mark_request_dirty(struct nfs_page *req); static int nfs_wait_on_write_congestion(struct address_space *, int); static int nfs_wait_on_requests(struct inode *, unsigned long, unsigned int); static long nfs_flush_mapping(struct address_space *mapping, struct writeback_control *wbc, int how); -static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how); static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; @@ -280,8 +279,10 @@ static int nfs_page_mark_flush(struct page *page) spin_lock(req_lock); } spin_unlock(req_lock); - if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) + if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) { nfs_mark_request_dirty(req); + set_page_writeback(page); + } ret = test_bit(PG_NEED_FLUSH, &req->wb_flags); nfs_unlock_request(req); return ret; @@ -443,6 +444,13 @@ nfs_mark_request_dirty(struct nfs_page *req) mark_inode_dirty(inode); } +static void +nfs_redirty_request(struct nfs_page *req) +{ + clear_bit(PG_FLUSHING, &req->wb_flags); + __set_page_dirty_nobuffers(req->wb_page); +} + /* * Check if a request is dirty */ @@ -777,7 +785,7 @@ static void nfs_writepage_release(struct nfs_page *req) #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) if (!PageError(req->wb_page)) { if (NFS_NEED_RESCHED(req)) { - nfs_mark_request_dirty(req); + nfs_redirty_request(req); goto out; } else if (NFS_NEED_COMMIT(req)) { nfs_mark_request_commit(req); @@ -893,7 +901,6 @@ static int nfs_flush_multi(struct inode *inode, struct list_head *head, int how) atomic_set(&req->wb_complete, requests); ClearPageError(page); - set_page_writeback(page); offset = 0; nbytes = req->wb_bytes; do { @@ -923,7 +930,7 @@ out_bad: list_del(&data->pages); nfs_writedata_release(data); } - nfs_mark_request_dirty(req); + nfs_redirty_request(req); nfs_clear_page_writeback(req); return -ENOMEM; } @@ -954,7 +961,6 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, int how) nfs_list_remove_request(req); nfs_list_add_request(req, &data->pages); ClearPageError(req->wb_page); - set_page_writeback(req->wb_page); *pages++ = req->wb_page; count += req->wb_bytes; } @@ -969,7 +975,7 @@ static int nfs_flush_one(struct inode *inode, struct list_head *head, int how) while (!list_empty(head)) { struct nfs_page *req = nfs_list_entry(head->next); nfs_list_remove_request(req); - nfs_mark_request_dirty(req); + nfs_redirty_request(req); nfs_clear_page_writeback(req); } return -ENOMEM; @@ -1004,7 +1010,7 @@ out_err: while (!list_empty(head)) { req = nfs_list_entry(head->next); nfs_list_remove_request(req); - nfs_mark_request_dirty(req); + nfs_redirty_request(req); nfs_clear_page_writeback(req); } return error; @@ -1320,7 +1326,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) } /* We have a mismatch. Write the page again */ dprintk(" mismatch\n"); - nfs_mark_request_dirty(req); + nfs_redirty_request(req); next: nfs_clear_page_writeback(req); } @@ -1451,13 +1457,18 @@ int nfs_wb_all(struct inode *inode) .bdi = mapping->backing_dev_info, .sync_mode = WB_SYNC_ALL, .nr_to_write = LONG_MAX, + .for_writepages = 1, .range_cyclic = 1, }; int ret; + ret = generic_writepages(mapping, &wbc); + if (ret < 0) + goto out; ret = nfs_sync_mapping_wait(mapping, &wbc, 0); if (ret >= 0) return 0; +out: return ret; } @@ -1469,16 +1480,23 @@ int nfs_sync_mapping_range(struct address_space *mapping, loff_t range_start, lo .nr_to_write = LONG_MAX, .range_start = range_start, .range_end = range_end, + .for_writepages = 1, }; int ret; + if (!(how & FLUSH_NOWRITEPAGE)) { + ret = generic_writepages(mapping, &wbc); + if (ret < 0) + goto out; + } ret = nfs_sync_mapping_wait(mapping, &wbc, how); if (ret >= 0) return 0; +out: return ret; } -static int nfs_wb_page_priority(struct inode *inode, struct page *page, int how) +int nfs_wb_page_priority(struct inode *inode, struct page *page, int how) { loff_t range_start = page_offset(page); loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 3de334c72b1f..04963063e620 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -438,6 +438,7 @@ extern long nfs_sync_mapping_wait(struct address_space *, struct writeback_contr extern int nfs_sync_mapping_range(struct address_space *, loff_t, loff_t, int); extern int nfs_wb_all(struct inode *inode); extern int nfs_wb_page(struct inode *inode, struct page* page); +extern int nfs_wb_page_priority(struct inode *inode, struct page* page, int how); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) extern int nfs_commit_inode(struct inode *, int); extern struct nfs_write_data *nfs_commit_alloc(void); -- cgit v1.2.3 From 8fc7500bb8ea3b5c909869d00628635e964ae882 Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Mon, 4 Dec 2006 20:22:31 -0500 Subject: rpc: gss: eliminate print_hexl()'s Dumping all this data to the logs is wasteful (even when debugging is turned off), and creates too much output to be useful when it's turned on. Fix a minor style bug or two while we're at it. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/auth_gss.h | 2 -- net/sunrpc/auth_gss/auth_gss.c | 40 ---------------------------------- net/sunrpc/auth_gss/gss_krb5_crypto.c | 16 ++------------ net/sunrpc/auth_gss/gss_krb5_seal.c | 3 --- net/sunrpc/auth_gss/gss_krb5_wrap.c | 3 --- net/sunrpc/auth_gss/gss_spkm3_unseal.c | 4 ---- 6 files changed, 2 insertions(+), 66 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/auth_gss.h b/include/linux/sunrpc/auth_gss.h index 97b62e97dd8d..2db2fbf34947 100644 --- a/include/linux/sunrpc/auth_gss.h +++ b/include/linux/sunrpc/auth_gss.h @@ -90,8 +90,6 @@ struct gss_cred { #define gc_flags gc_base.cr_flags #define gc_expire gc_base.cr_expire -void print_hexl(u32 *p, u_int length, u_int offset); - #endif /* __KERNEL__ */ #endif /* _LINUX_SUNRPC_AUTH_GSS_H */ diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index e5a84a482e57..d12ee5f54c0c 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -94,46 +94,6 @@ struct gss_auth { static void gss_destroy_ctx(struct gss_cl_ctx *); static struct rpc_pipe_ops gss_upcall_ops; -void -print_hexl(u32 *p, u_int length, u_int offset) -{ - u_int i, j, jm; - u8 c, *cp; - - dprintk("RPC: print_hexl: length %d\n",length); - dprintk("\n"); - cp = (u8 *) p; - - for (i = 0; i < length; i += 0x10) { - dprintk(" %04x: ", (u_int)(i + offset)); - jm = length - i; - jm = jm > 16 ? 16 : jm; - - for (j = 0; j < jm; j++) { - if ((j % 2) == 1) - dprintk("%02x ", (u_int)cp[i+j]); - else - dprintk("%02x", (u_int)cp[i+j]); - } - for (; j < 16; j++) { - if ((j % 2) == 1) - dprintk(" "); - else - dprintk(" "); - } - dprintk(" "); - - for (j = 0; j < jm; j++) { - c = cp[i+j]; - c = isprint(c) ? c : '.'; - dprintk("%c", c); - } - dprintk("\n"); - } -} - -EXPORT_SYMBOL(print_hexl); - static inline struct gss_cl_ctx * gss_get_ctx(struct gss_cl_ctx *ctx) { diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index e11a40b25cce..4c53896f1b08 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -61,9 +61,6 @@ krb5_encrypt( u8 local_iv[16] = {0}; struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv }; - dprintk("RPC: krb5_encrypt: input data:\n"); - print_hexl((u32 *)in, length, 0); - if (length % crypto_blkcipher_blocksize(tfm) != 0) goto out; @@ -80,12 +77,9 @@ krb5_encrypt( sg_set_buf(sg, out, length); ret = crypto_blkcipher_encrypt_iv(&desc, sg, sg, length); - - dprintk("RPC: krb5_encrypt: output data:\n"); - print_hexl((u32 *)out, length, 0); out: dprintk("RPC: krb5_encrypt returns %d\n",ret); - return(ret); + return ret; } EXPORT_SYMBOL(krb5_encrypt); @@ -103,9 +97,6 @@ krb5_decrypt( u8 local_iv[16] = {0}; struct blkcipher_desc desc = { .tfm = tfm, .info = local_iv }; - dprintk("RPC: krb5_decrypt: input data:\n"); - print_hexl((u32 *)in, length, 0); - if (length % crypto_blkcipher_blocksize(tfm) != 0) goto out; @@ -121,12 +112,9 @@ krb5_decrypt( sg_set_buf(sg, out, length); ret = crypto_blkcipher_decrypt_iv(&desc, sg, sg, length); - - dprintk("RPC: krb5_decrypt: output_data:\n"); - print_hexl((u32 *)out, length, 0); out: dprintk("RPC: gss_k5decrypt returns %d\n",ret); - return(ret); + return ret; } EXPORT_SYMBOL(krb5_decrypt); diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c index 08601ee4cd73..dc58af0b8b4c 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seal.c +++ b/net/sunrpc/auth_gss/gss_krb5_seal.c @@ -129,9 +129,6 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, memcpy(krb5_hdr + 16, md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, KRB5_CKSUM_LENGTH); - - dprintk("RPC: make_seal_token: cksum data: \n"); - print_hexl((u32 *) (krb5_hdr + 16), KRB5_CKSUM_LENGTH, 0); break; default: BUG(); diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c index cc45c1605f80..0f512e8e0d19 100644 --- a/net/sunrpc/auth_gss/gss_krb5_wrap.c +++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c @@ -199,9 +199,6 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, memcpy(krb5_hdr + 16, md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, KRB5_CKSUM_LENGTH); - - dprintk("RPC: make_seal_token: cksum data: \n"); - print_hexl((u32 *) (krb5_hdr + 16), KRB5_CKSUM_LENGTH, 0); break; default: BUG(); diff --git a/net/sunrpc/auth_gss/gss_spkm3_unseal.c b/net/sunrpc/auth_gss/gss_spkm3_unseal.c index 8537f581ef9b..544eb6fd9094 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_unseal.c +++ b/net/sunrpc/auth_gss/gss_spkm3_unseal.c @@ -103,10 +103,6 @@ spkm3_read_token(struct spkm3_ctx *ctx, dprintk("RPC: spkm3_read_token: digest wire_cksum.len %d:\n", wire_cksum.len); - dprintk(" md5cksum.data\n"); - print_hexl((u32 *) md5cksum.data, 16, 0); - dprintk(" cksum.data:\n"); - print_hexl((u32 *) wire_cksum.data, wire_cksum.len, 0); ret = GSS_S_BAD_SIG; code = memcmp(md5cksum.data, wire_cksum.data, wire_cksum.len); -- cgit v1.2.3 From 37a4e6cb0391f2293ba3d59e3a63ec0e56ed720d Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Mon, 4 Dec 2006 20:22:33 -0500 Subject: rpc: move process_xdr_buf Since process_xdr_buf() is useful outside of the kerberos-specific code, we move it to net/sunrpc/xdr.c, export it, and rename it in keeping with xdr_* naming convention of xdr.c. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 2 + net/sunrpc/auth_gss/gss_krb5_crypto.c | 73 ++--------------------------------- net/sunrpc/xdr.c | 68 ++++++++++++++++++++++++++++++++ 3 files changed, 74 insertions(+), 69 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 441b91dbafe5..4d17846cd78a 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -11,6 +11,7 @@ #include #include +#include /* * Buffer adjustment @@ -196,6 +197,7 @@ extern void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 extern __be32 *xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes); extern void xdr_read_pages(struct xdr_stream *xdr, unsigned int len); extern void xdr_enter_page(struct xdr_stream *xdr, unsigned int len); +extern int xdr_process_buf(struct xdr_buf *buf, unsigned int offset, unsigned int len, int (*actor)(struct scatterlist *, void *), void *data); #endif /* __KERNEL__ */ diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index 4c53896f1b08..10d05ea37213 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -43,6 +43,7 @@ #include #include #include +#include #ifdef RPC_DEBUG # define RPCDBG_FACILITY RPCDBG_AUTH @@ -119,72 +120,6 @@ out: EXPORT_SYMBOL(krb5_decrypt); -static int -process_xdr_buf(struct xdr_buf *buf, int offset, int len, - int (*actor)(struct scatterlist *, void *), void *data) -{ - int i, page_len, thislen, page_offset, ret = 0; - struct scatterlist sg[1]; - - if (offset >= buf->head[0].iov_len) { - offset -= buf->head[0].iov_len; - } else { - thislen = buf->head[0].iov_len - offset; - if (thislen > len) - thislen = len; - sg_set_buf(sg, buf->head[0].iov_base + offset, thislen); - ret = actor(sg, data); - if (ret) - goto out; - offset = 0; - len -= thislen; - } - if (len == 0) - goto out; - - if (offset >= buf->page_len) { - offset -= buf->page_len; - } else { - page_len = buf->page_len - offset; - if (page_len > len) - page_len = len; - len -= page_len; - page_offset = (offset + buf->page_base) & (PAGE_CACHE_SIZE - 1); - i = (offset + buf->page_base) >> PAGE_CACHE_SHIFT; - thislen = PAGE_CACHE_SIZE - page_offset; - do { - if (thislen > page_len) - thislen = page_len; - sg->page = buf->pages[i]; - sg->offset = page_offset; - sg->length = thislen; - ret = actor(sg, data); - if (ret) - goto out; - page_len -= thislen; - i++; - page_offset = 0; - thislen = PAGE_CACHE_SIZE; - } while (page_len != 0); - offset = 0; - } - if (len == 0) - goto out; - - if (offset < buf->tail[0].iov_len) { - thislen = buf->tail[0].iov_len - offset; - if (thislen > len) - thislen = len; - sg_set_buf(sg, buf->tail[0].iov_base + offset, thislen); - ret = actor(sg, data); - len -= thislen; - } - if (len != 0) - ret = -EINVAL; -out: - return ret; -} - static int checksummer(struct scatterlist *sg, void *data) { @@ -225,7 +160,7 @@ make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, err = crypto_hash_update(&desc, sg, hdrlen); if (err) goto out; - err = process_xdr_buf(body, body_offset, body->len - body_offset, + err = xdr_process_buf(body, body_offset, body->len - body_offset, checksummer, &desc); if (err) goto out; @@ -323,7 +258,7 @@ gss_encrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf, desc.fragno = 0; desc.fraglen = 0; - ret = process_xdr_buf(buf, offset, buf->len - offset, encryptor, &desc); + ret = xdr_process_buf(buf, offset, buf->len - offset, encryptor, &desc); return ret; } @@ -389,7 +324,7 @@ gss_decrypt_xdr_buf(struct crypto_blkcipher *tfm, struct xdr_buf *buf, desc.desc.flags = 0; desc.fragno = 0; desc.fraglen = 0; - return process_xdr_buf(buf, offset, buf->len - offset, decryptor, &desc); + return xdr_process_buf(buf, offset, buf->len - offset, decryptor, &desc); } EXPORT_SYMBOL(gss_decrypt_xdr_buf); diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 5a6485946f3c..a0af250ca319 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -1021,3 +1021,71 @@ xdr_encode_array2(struct xdr_buf *buf, unsigned int base, return xdr_xcode_array2(buf, base, desc, 1); } + +int +xdr_process_buf(struct xdr_buf *buf, unsigned int offset, unsigned int len, + int (*actor)(struct scatterlist *, void *), void *data) +{ + int i, ret = 0; + unsigned page_len, thislen, page_offset; + struct scatterlist sg[1]; + + if (offset >= buf->head[0].iov_len) { + offset -= buf->head[0].iov_len; + } else { + thislen = buf->head[0].iov_len - offset; + if (thislen > len) + thislen = len; + sg_set_buf(sg, buf->head[0].iov_base + offset, thislen); + ret = actor(sg, data); + if (ret) + goto out; + offset = 0; + len -= thislen; + } + if (len == 0) + goto out; + + if (offset >= buf->page_len) { + offset -= buf->page_len; + } else { + page_len = buf->page_len - offset; + if (page_len > len) + page_len = len; + len -= page_len; + page_offset = (offset + buf->page_base) & (PAGE_CACHE_SIZE - 1); + i = (offset + buf->page_base) >> PAGE_CACHE_SHIFT; + thislen = PAGE_CACHE_SIZE - page_offset; + do { + if (thislen > page_len) + thislen = page_len; + sg->page = buf->pages[i]; + sg->offset = page_offset; + sg->length = thislen; + ret = actor(sg, data); + if (ret) + goto out; + page_len -= thislen; + i++; + page_offset = 0; + thislen = PAGE_CACHE_SIZE; + } while (page_len != 0); + offset = 0; + } + if (len == 0) + goto out; + if (offset < buf->tail[0].iov_len) { + thislen = buf->tail[0].iov_len - offset; + if (thislen > len) + thislen = len; + sg_set_buf(sg, buf->tail[0].iov_base + offset, thislen); + ret = actor(sg, data); + len -= thislen; + } + if (len != 0) + ret = -EINVAL; +out: + return ret; +} +EXPORT_SYMBOL(xdr_process_buf); + -- cgit v1.2.3 From adeb8133dd57f380e70a389a89a2ea3ae227f9e2 Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Mon, 4 Dec 2006 20:22:34 -0500 Subject: rpc: spkm3 update This updates the spkm3 code to bring it up to date with our current understanding of the spkm3 spec. In doing so, we're changing the downcall format used by gssd in the spkm3 case, which will cause an incompatilibity with old userland spkm3 support. Since the old code a) didn't implement the protocol correctly, and b) was never distributed except in the form of some experimental patches from the citi web site, we're assuming this is OK. We do detect the old downcall format and print warning (and fail). We also include a version number in the new downcall format, to be used in the future in case any further change is required. In some more detail: - fix integrity support - removed dependency on NIDs. instead OIDs are used - known OID values for algorithms added. - fixed some context fields and types Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_spkm3.h | 34 ++++----- net/sunrpc/auth_gss/auth_gss.c | 2 +- net/sunrpc/auth_gss/gss_spkm3_mech.c | 131 +++++++++------------------------ net/sunrpc/auth_gss/gss_spkm3_seal.c | 101 +++++++++++++++++++------ net/sunrpc/auth_gss/gss_spkm3_token.c | 6 +- net/sunrpc/auth_gss/gss_spkm3_unseal.c | 88 +++++++++++----------- 6 files changed, 180 insertions(+), 182 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_spkm3.h b/include/linux/sunrpc/gss_spkm3.h index 2cf3fbb40b4f..e3e6a3437f8b 100644 --- a/include/linux/sunrpc/gss_spkm3.h +++ b/include/linux/sunrpc/gss_spkm3.h @@ -12,27 +12,19 @@ #include struct spkm3_ctx { - struct xdr_netobj ctx_id; /* per message context id */ - int qop; /* negotiated qop */ + struct xdr_netobj ctx_id; /* per message context id */ + int endtime; /* endtime of the context */ struct xdr_netobj mech_used; unsigned int ret_flags ; - unsigned int req_flags ; - struct xdr_netobj share_key; - int conf_alg; - struct crypto_blkcipher *derived_conf_key; - int intg_alg; - struct crypto_blkcipher *derived_integ_key; - int keyestb_alg; /* alg used to get share_key */ - int owf_alg; /* one way function */ + struct xdr_netobj conf_alg; + struct xdr_netobj derived_conf_key; + struct xdr_netobj intg_alg; + struct xdr_netobj derived_integ_key; }; -/* from openssl/objects.h */ -/* XXX need SEAL_ALG_NONE */ -#define NID_md5 4 -#define NID_dhKeyAgreement 28 -#define NID_des_cbc 31 -#define NID_sha1 64 -#define NID_cast5_cbc 108 +/* OIDs declarations for K-ALG, I-ALG, C-ALG, and OWF-ALG */ +extern const struct xdr_netobj hmac_md5_oid; +extern const struct xdr_netobj cast5_cbc_oid; /* SPKM InnerContext Token types */ @@ -46,11 +38,13 @@ u32 spkm3_make_token(struct spkm3_ctx *ctx, struct xdr_buf * text, struct xdr_ne u32 spkm3_read_token(struct spkm3_ctx *ctx, struct xdr_netobj *read_token, struct xdr_buf *message_buffer, int toktype); #define CKSUMTYPE_RSA_MD5 0x0007 +#define CKSUMTYPE_HMAC_MD5 0x0008 -s32 make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, - int body_offset, struct xdr_netobj *cksum); +s32 make_spkm3_checksum(s32 cksumtype, struct xdr_netobj *key, char *header, + unsigned int hdrlen, struct xdr_buf *body, + unsigned int body_offset, struct xdr_netobj *cksum); void asn1_bitstring_len(struct xdr_netobj *in, int *enclen, int *zerobits); -int decode_asn1_bitstring(struct xdr_netobj *out, char *in, int enclen, +int decode_asn1_bitstring(struct xdr_netobj *out, char *in, int enclen, int explen); void spkm3_mic_header(unsigned char **hdrbuf, unsigned int *hdrlen, unsigned char *ctxhdr, int elen, int zbit); diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index d12ee5f54c0c..a02ecc1f230d 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -68,7 +68,7 @@ static struct rpc_credops gss_credops; #define GSS_CRED_SLACK 1024 /* XXX: unused */ /* length of a krb5 verifier (48), plus data added before arguments when * using integrity (two 4-byte integers): */ -#define GSS_VERF_SLACK 56 +#define GSS_VERF_SLACK 100 /* XXX this define must match the gssd define * as it is passed to gssd to signal the use of diff --git a/net/sunrpc/auth_gss/gss_spkm3_mech.c b/net/sunrpc/auth_gss/gss_spkm3_mech.c index d57f60838895..41465072d0b5 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_mech.c +++ b/net/sunrpc/auth_gss/gss_spkm3_mech.c @@ -82,133 +82,73 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *res) return q; } -static inline const void * -get_key(const void *p, const void *end, struct crypto_blkcipher **res, - int *resalg) -{ - struct xdr_netobj key = { 0 }; - int setkey = 0; - char *alg_name; - - p = simple_get_bytes(p, end, resalg, sizeof(*resalg)); - if (IS_ERR(p)) - goto out_err; - p = simple_get_netobj(p, end, &key); - if (IS_ERR(p)) - goto out_err; - - switch (*resalg) { - case NID_des_cbc: - alg_name = "cbc(des)"; - setkey = 1; - break; - case NID_cast5_cbc: - /* XXXX here in name only, not used */ - alg_name = "cbc(cast5)"; - setkey = 0; /* XXX will need to set to 1 */ - break; - case NID_md5: - if (key.len == 0) { - dprintk("RPC: SPKM3 get_key: NID_md5 zero Key length\n"); - } - alg_name = "md5"; - setkey = 0; - break; - default: - dprintk("gss_spkm3_mech: unsupported algorithm %d\n", *resalg); - goto out_err_free_key; - } - *res = crypto_alloc_blkcipher(alg_name, 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(*res)) { - printk("gss_spkm3_mech: unable to initialize crypto algorthm %s\n", alg_name); - *res = NULL; - goto out_err_free_key; - } - if (setkey) { - if (crypto_blkcipher_setkey(*res, key.data, key.len)) { - printk("gss_spkm3_mech: error setting key for crypto algorthm %s\n", alg_name); - goto out_err_free_tfm; - } - } - - if(key.len > 0) - kfree(key.data); - return p; - -out_err_free_tfm: - crypto_free_blkcipher(*res); -out_err_free_key: - if(key.len > 0) - kfree(key.data); - p = ERR_PTR(-EINVAL); -out_err: - return p; -} - static int gss_import_sec_context_spkm3(const void *p, size_t len, struct gss_ctx *ctx_id) { const void *end = (const void *)((const char *)p + len); struct spkm3_ctx *ctx; + int version; if (!(ctx = kzalloc(sizeof(*ctx), GFP_KERNEL))) goto out_err; + p = simple_get_bytes(p, end, &version, sizeof(version)); + if (IS_ERR(p)) + goto out_err_free_ctx; + if (version != 1) { + dprintk("RPC: unknown spkm3 token format: obsolete nfs-utils?\n"); + goto out_err_free_ctx; + } + p = simple_get_netobj(p, end, &ctx->ctx_id); if (IS_ERR(p)) goto out_err_free_ctx; - p = simple_get_bytes(p, end, &ctx->qop, sizeof(ctx->qop)); + p = simple_get_bytes(p, end, &ctx->endtime, sizeof(ctx->endtime)); if (IS_ERR(p)) goto out_err_free_ctx_id; p = simple_get_netobj(p, end, &ctx->mech_used); if (IS_ERR(p)) - goto out_err_free_mech; + goto out_err_free_ctx_id; p = simple_get_bytes(p, end, &ctx->ret_flags, sizeof(ctx->ret_flags)); if (IS_ERR(p)) goto out_err_free_mech; - p = simple_get_bytes(p, end, &ctx->req_flags, sizeof(ctx->req_flags)); + p = simple_get_netobj(p, end, &ctx->conf_alg); if (IS_ERR(p)) goto out_err_free_mech; - p = simple_get_netobj(p, end, &ctx->share_key); - if (IS_ERR(p)) - goto out_err_free_s_key; - - p = get_key(p, end, &ctx->derived_conf_key, &ctx->conf_alg); + p = simple_get_netobj(p, end, &ctx->derived_conf_key); if (IS_ERR(p)) - goto out_err_free_s_key; + goto out_err_free_conf_alg; - p = get_key(p, end, &ctx->derived_integ_key, &ctx->intg_alg); + p = simple_get_netobj(p, end, &ctx->intg_alg); if (IS_ERR(p)) - goto out_err_free_key1; + goto out_err_free_conf_key; - p = simple_get_bytes(p, end, &ctx->keyestb_alg, sizeof(ctx->keyestb_alg)); + p = simple_get_netobj(p, end, &ctx->derived_integ_key); if (IS_ERR(p)) - goto out_err_free_key2; - - p = simple_get_bytes(p, end, &ctx->owf_alg, sizeof(ctx->owf_alg)); - if (IS_ERR(p)) - goto out_err_free_key2; + goto out_err_free_intg_alg; if (p != end) - goto out_err_free_key2; + goto out_err_free_intg_key; ctx_id->internal_ctx_id = ctx; dprintk("Successfully imported new spkm context.\n"); return 0; -out_err_free_key2: - crypto_free_blkcipher(ctx->derived_integ_key); -out_err_free_key1: - crypto_free_blkcipher(ctx->derived_conf_key); -out_err_free_s_key: - kfree(ctx->share_key.data); +out_err_free_intg_key: + kfree(ctx->derived_integ_key.data); +out_err_free_intg_alg: + kfree(ctx->intg_alg.data); +out_err_free_conf_key: + kfree(ctx->derived_conf_key.data); +out_err_free_conf_alg: + kfree(ctx->conf_alg.data); out_err_free_mech: kfree(ctx->mech_used.data); out_err_free_ctx_id: @@ -220,13 +160,16 @@ out_err: } static void -gss_delete_sec_context_spkm3(void *internal_ctx) { +gss_delete_sec_context_spkm3(void *internal_ctx) +{ struct spkm3_ctx *sctx = internal_ctx; - crypto_free_blkcipher(sctx->derived_integ_key); - crypto_free_blkcipher(sctx->derived_conf_key); - kfree(sctx->share_key.data); + kfree(sctx->derived_integ_key.data); + kfree(sctx->intg_alg.data); + kfree(sctx->derived_conf_key.data); + kfree(sctx->conf_alg.data); kfree(sctx->mech_used.data); + kfree(sctx->ctx_id.data); kfree(sctx); } @@ -238,7 +181,6 @@ gss_verify_mic_spkm3(struct gss_ctx *ctx, u32 maj_stat = 0; struct spkm3_ctx *sctx = ctx->internal_ctx_id; - dprintk("RPC: gss_verify_mic_spkm3 calling spkm3_read_token\n"); maj_stat = spkm3_read_token(sctx, checksum, signbuf, SPKM_MIC_TOK); dprintk("RPC: gss_verify_mic_spkm3 returning %d\n", maj_stat); @@ -253,10 +195,9 @@ gss_get_mic_spkm3(struct gss_ctx *ctx, u32 err = 0; struct spkm3_ctx *sctx = ctx->internal_ctx_id; - dprintk("RPC: gss_get_mic_spkm3\n"); - err = spkm3_make_token(sctx, message_buffer, - message_token, SPKM_MIC_TOK); + message_token, SPKM_MIC_TOK); + dprintk("RPC: gss_get_mic_spkm3 returning %d\n", err); return err; } diff --git a/net/sunrpc/auth_gss/gss_spkm3_seal.c b/net/sunrpc/auth_gss/gss_spkm3_seal.c index 18c7862bc234..b179d58c6249 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_seal.c +++ b/net/sunrpc/auth_gss/gss_spkm3_seal.c @@ -39,11 +39,17 @@ #include #include #include +#include +#include +#include #ifdef RPC_DEBUG # define RPCDBG_FACILITY RPCDBG_AUTH #endif +const struct xdr_netobj hmac_md5_oid = { 8, "\x2B\x06\x01\x05\x05\x08\x01\x01"}; +const struct xdr_netobj cast5_cbc_oid = {9, "\x2A\x86\x48\x86\xF6\x7D\x07\x42\x0A"}; + /* * spkm3_make_token() * @@ -66,29 +72,23 @@ spkm3_make_token(struct spkm3_ctx *ctx, int ctxelen = 0, ctxzbit = 0; int md5elen = 0, md5zbit = 0; - dprintk("RPC: spkm3_make_token\n"); - now = jiffies; if (ctx->ctx_id.len != 16) { dprintk("RPC: spkm3_make_token BAD ctx_id.len %d\n", - ctx->ctx_id.len); + ctx->ctx_id.len); goto out_err; } - - switch (ctx->intg_alg) { - case NID_md5: - checksum_type = CKSUMTYPE_RSA_MD5; - break; - default: - dprintk("RPC: gss_spkm3_seal: ctx->signalg %d not" - " supported\n", ctx->intg_alg); - goto out_err; - } - /* XXX since we don't support WRAP, perhaps we don't care... */ - if (ctx->conf_alg != NID_cast5_cbc) { - dprintk("RPC: gss_spkm3_seal: ctx->sealalg %d not supported\n", - ctx->conf_alg); + + if (!g_OID_equal(&ctx->intg_alg, &hmac_md5_oid)) { + dprintk("RPC: gss_spkm3_seal: unsupported I-ALG algorithm." + "only support hmac-md5 I-ALG.\n"); + goto out_err; + } else + checksum_type = CKSUMTYPE_HMAC_MD5; + + if (!g_OID_equal(&ctx->conf_alg, &cast5_cbc_oid)) { + dprintk("RPC: gss_spkm3_seal: unsupported C-ALG algorithm\n"); goto out_err; } @@ -96,10 +96,10 @@ spkm3_make_token(struct spkm3_ctx *ctx, /* Calculate checksum over the mic-header */ asn1_bitstring_len(&ctx->ctx_id, &ctxelen, &ctxzbit); spkm3_mic_header(&mic_hdr.data, &mic_hdr.len, ctx->ctx_id.data, - ctxelen, ctxzbit); - - if (make_checksum(checksum_type, mic_hdr.data, mic_hdr.len, - text, 0, &md5cksum)) + ctxelen, ctxzbit); + if (make_spkm3_checksum(checksum_type, &ctx->derived_integ_key, + (char *)mic_hdr.data, mic_hdr.len, + text, 0, &md5cksum)) goto out_err; asn1_bitstring_len(&md5cksum, &md5elen, &md5zbit); @@ -121,7 +121,66 @@ spkm3_make_token(struct spkm3_ctx *ctx, return GSS_S_COMPLETE; out_err: + if (md5cksum.data) + kfree(md5cksum.data); + token->data = NULL; token->len = 0; return GSS_S_FAILURE; } + +static int +spkm3_checksummer(struct scatterlist *sg, void *data) +{ + struct hash_desc *desc = data; + + return crypto_hash_update(desc, sg, sg->length); +} + +/* checksum the plaintext data and hdrlen bytes of the token header */ +s32 +make_spkm3_checksum(s32 cksumtype, struct xdr_netobj *key, char *header, + unsigned int hdrlen, struct xdr_buf *body, + unsigned int body_offset, struct xdr_netobj *cksum) +{ + char *cksumname; + struct hash_desc desc; /* XXX add to ctx? */ + struct scatterlist sg[1]; + int err; + + switch (cksumtype) { + case CKSUMTYPE_HMAC_MD5: + cksumname = "md5"; + break; + default: + dprintk("RPC: spkm3_make_checksum:" + " unsupported checksum %d", cksumtype); + return GSS_S_FAILURE; + } + + if (key->data == NULL || key->len <= 0) return GSS_S_FAILURE; + + desc.tfm = crypto_alloc_hash(cksumname, 0, CRYPTO_ALG_ASYNC); + if (IS_ERR(desc.tfm)) + return GSS_S_FAILURE; + cksum->len = crypto_hash_digestsize(desc.tfm); + desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP; + + err = crypto_hash_setkey(desc.tfm, key->data, key->len); + if (err) + goto out; + + sg_set_buf(sg, header, hdrlen); + crypto_hash_update(&desc, sg, 1); + + xdr_process_buf(body, body_offset, body->len - body_offset, + spkm3_checksummer, &desc); + crypto_hash_final(&desc, cksum->data); + +out: + crypto_free_hash(desc.tfm); + + return err ? GSS_S_FAILURE : 0; +} + +EXPORT_SYMBOL(make_spkm3_checksum); diff --git a/net/sunrpc/auth_gss/gss_spkm3_token.c b/net/sunrpc/auth_gss/gss_spkm3_token.c index 854a983ccf26..35188b6ea8f7 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_token.c +++ b/net/sunrpc/auth_gss/gss_spkm3_token.c @@ -172,10 +172,10 @@ spkm3_mic_header(unsigned char **hdrbuf, unsigned int *hdrlen, unsigned char *ct *(u8 *)hptr++ = zbit; memcpy(hptr, ctxdata, elen); hptr += elen; - *hdrlen = hptr - top; + *hdrlen = hptr - top; } - -/* + +/* * spkm3_mic_innercontext_token() * * *tokp points to the beginning of the SPKM_MIC token described diff --git a/net/sunrpc/auth_gss/gss_spkm3_unseal.c b/net/sunrpc/auth_gss/gss_spkm3_unseal.c index 544eb6fd9094..e54581ca7570 100644 --- a/net/sunrpc/auth_gss/gss_spkm3_unseal.c +++ b/net/sunrpc/auth_gss/gss_spkm3_unseal.c @@ -54,66 +54,70 @@ spkm3_read_token(struct spkm3_ctx *ctx, struct xdr_buf *message_buffer, /* signbuf */ int toktype) { + s32 checksum_type; s32 code; struct xdr_netobj wire_cksum = {.len =0, .data = NULL}; char cksumdata[16]; struct xdr_netobj md5cksum = {.len = 0, .data = cksumdata}; unsigned char *ptr = (unsigned char *)read_token->data; - unsigned char *cksum; + unsigned char *cksum; int bodysize, md5elen; int mic_hdrlen; u32 ret = GSS_S_DEFECTIVE_TOKEN; - dprintk("RPC: spkm3_read_token read_token->len %d\n", read_token->len); - if (g_verify_token_header((struct xdr_netobj *) &ctx->mech_used, &bodysize, &ptr, read_token->len)) goto out; /* decode the token */ - if (toktype == SPKM_MIC_TOK) { - - if ((ret = spkm3_verify_mic_token(&ptr, &mic_hdrlen, &cksum))) - goto out; - - if (*cksum++ != 0x03) { - dprintk("RPC: spkm3_read_token BAD checksum type\n"); - goto out; - } - md5elen = *cksum++; - cksum++; /* move past the zbit */ - - if(!decode_asn1_bitstring(&wire_cksum, cksum, md5elen - 1, 16)) - goto out; - - /* HARD CODED FOR MD5 */ - - /* compute the checksum of the message. - * ptr + 2 = start of header piece of checksum - * mic_hdrlen + 2 = length of header piece of checksum - */ - ret = GSS_S_DEFECTIVE_TOKEN; - code = make_checksum(CKSUMTYPE_RSA_MD5, ptr + 2, - mic_hdrlen + 2, - message_buffer, 0, &md5cksum); - - if (code) - goto out; - - dprintk("RPC: spkm3_read_token: digest wire_cksum.len %d:\n", - wire_cksum.len); - - ret = GSS_S_BAD_SIG; - code = memcmp(md5cksum.data, wire_cksum.data, wire_cksum.len); - if (code) - goto out; - - } else { - dprintk("RPC: BAD or UNSUPPORTED SPKM3 token type: %d\n",toktype); + if (toktype != SPKM_MIC_TOK) { + dprintk("RPC: BAD SPKM3 token type: %d\n", toktype); + goto out; + } + + if ((ret = spkm3_verify_mic_token(&ptr, &mic_hdrlen, &cksum))) + goto out; + + if (*cksum++ != 0x03) { + dprintk("RPC: spkm3_read_token BAD checksum type\n"); + goto out; + } + md5elen = *cksum++; + cksum++; /* move past the zbit */ + + if (!decode_asn1_bitstring(&wire_cksum, cksum, md5elen - 1, 16)) + goto out; + + /* HARD CODED FOR MD5 */ + + /* compute the checksum of the message. + * ptr + 2 = start of header piece of checksum + * mic_hdrlen + 2 = length of header piece of checksum + */ + ret = GSS_S_DEFECTIVE_TOKEN; + if (!g_OID_equal(&ctx->intg_alg, &hmac_md5_oid)) { + dprintk("RPC: gss_spkm3_seal: unsupported I-ALG algorithm\n"); + goto out; + } + + checksum_type = CKSUMTYPE_HMAC_MD5; + + code = make_spkm3_checksum(checksum_type, + &ctx->derived_integ_key, ptr + 2, mic_hdrlen + 2, + message_buffer, 0, &md5cksum); + + if (code) + goto out; + + ret = GSS_S_BAD_SIG; + code = memcmp(md5cksum.data, wire_cksum.data, wire_cksum.len); + if (code) { + dprintk("RPC: bad MIC checksum\n"); goto out; } + /* XXX: need to add expiration and sequencing */ ret = GSS_S_COMPLETE; out: -- cgit v1.2.3 From e678e06bf8fa25981a6fa1f08b979fd086d713f8 Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Mon, 4 Dec 2006 20:22:35 -0500 Subject: gss: krb5: remove signalg and sealalg We designed the krb5 context import without completely understanding the context. Now it's clear that there are a number of fields that we ignore, or that we depend on having one single value. In particular, we only support one value of signalg currently; so let's check the signalg field in the downcall (in case we decide there's something else we could support here eventually), but ignore it otherwise. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 1 - net/sunrpc/auth_gss/gss_krb5_mech.c | 5 ++++- net/sunrpc/auth_gss/gss_krb5_seal.c | 34 ++++++++++------------------------ net/sunrpc/auth_gss/gss_krb5_wrap.c | 30 ++++++++---------------------- 4 files changed, 22 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index e30ba201910a..f680ed3b1b5e 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -44,7 +44,6 @@ struct krb5_ctx { int initiate; /* 1 = initiating, 0 = accepting */ int seed_init; unsigned char seed[16]; - int signalg; int sealalg; struct crypto_blkcipher *enc; struct crypto_blkcipher *seq; diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index 754b8cd6439f..17587163fcae 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -129,6 +129,7 @@ gss_import_sec_context_kerberos(const void *p, { const void *end = (const void *)((const char *)p + len); struct krb5_ctx *ctx; + int tmp; if (!(ctx = kzalloc(sizeof(*ctx), GFP_KERNEL))) goto out_err; @@ -142,9 +143,11 @@ gss_import_sec_context_kerberos(const void *p, p = simple_get_bytes(p, end, ctx->seed, sizeof(ctx->seed)); if (IS_ERR(p)) goto out_err_free_ctx; - p = simple_get_bytes(p, end, &ctx->signalg, sizeof(ctx->signalg)); + p = simple_get_bytes(p, end, &tmp, sizeof(tmp)); if (IS_ERR(p)) goto out_err_free_ctx; + if (tmp != SGN_ALG_DES_MAC_MD5) + goto out_err_free_ctx; p = simple_get_bytes(p, end, &ctx->sealalg, sizeof(ctx->sealalg)); if (IS_ERR(p)) goto out_err_free_ctx; diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c index dc58af0b8b4c..a496af585a08 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seal.c +++ b/net/sunrpc/auth_gss/gss_krb5_seal.c @@ -88,15 +88,7 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, now = get_seconds(); - switch (ctx->signalg) { - case SGN_ALG_DES_MAC_MD5: - checksum_type = CKSUMTYPE_RSA_MD5; - break; - default: - dprintk("RPC: gss_krb5_seal: ctx->signalg %d not" - " supported\n", ctx->signalg); - goto out_err; - } + checksum_type = CKSUMTYPE_RSA_MD5; if (ctx->sealalg != SEAL_ALG_NONE && ctx->sealalg != SEAL_ALG_DES) { dprintk("RPC: gss_krb5_seal: ctx->sealalg %d not supported\n", ctx->sealalg); @@ -115,24 +107,18 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, krb5_hdr = ptr - 2; msg_start = krb5_hdr + 24; - *(__be16 *)(krb5_hdr + 2) = htons(ctx->signalg); + *(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5); memset(krb5_hdr + 4, 0xff, 4); if (make_checksum(checksum_type, krb5_hdr, 8, text, 0, &md5cksum)) - goto out_err; - - switch (ctx->signalg) { - case SGN_ALG_DES_MAC_MD5: - if (krb5_encrypt(ctx->seq, NULL, md5cksum.data, - md5cksum.data, md5cksum.len)) - goto out_err; - memcpy(krb5_hdr + 16, - md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, - KRB5_CKSUM_LENGTH); - break; - default: - BUG(); - } + goto out_err; + + if (krb5_encrypt(ctx->seq, NULL, md5cksum.data, + md5cksum.data, md5cksum.len)) + goto out_err; + memcpy(krb5_hdr + 16, + md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, + KRB5_CKSUM_LENGTH); spin_lock(&krb5_seq_lock); seq_send = ctx->seq_send++; diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c index ad243872f547..eee49f4c4c6a 100644 --- a/net/sunrpc/auth_gss/gss_krb5_wrap.c +++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c @@ -134,15 +134,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, now = get_seconds(); - switch (kctx->signalg) { - case SGN_ALG_DES_MAC_MD5: - checksum_type = CKSUMTYPE_RSA_MD5; - break; - default: - dprintk("RPC: gss_krb5_seal: kctx->signalg %d not" - " supported\n", kctx->signalg); - goto out_err; - } + checksum_type = CKSUMTYPE_RSA_MD5; if (kctx->sealalg != SEAL_ALG_NONE && kctx->sealalg != SEAL_ALG_DES) { dprintk("RPC: gss_krb5_seal: kctx->sealalg %d not supported\n", kctx->sealalg); @@ -177,7 +169,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, msg_start = krb5_hdr + 24; /* XXXJBF: */ BUG_ON(buf->head[0].iov_base + offset + headlen != msg_start + blocksize); - *(__be16 *)(krb5_hdr + 2) = htons(kctx->signalg); + *(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5); memset(krb5_hdr + 4, 0xff, 4); *(__be16 *)(krb5_hdr + 4) = htons(kctx->sealalg); @@ -191,18 +183,12 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, goto out_err; buf->pages = tmp_pages; - switch (kctx->signalg) { - case SGN_ALG_DES_MAC_MD5: - if (krb5_encrypt(kctx->seq, NULL, md5cksum.data, - md5cksum.data, md5cksum.len)) - goto out_err; - memcpy(krb5_hdr + 16, - md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, - KRB5_CKSUM_LENGTH); - break; - default: - BUG(); - } + if (krb5_encrypt(kctx->seq, NULL, md5cksum.data, + md5cksum.data, md5cksum.len)) + goto out_err; + memcpy(krb5_hdr + 16, + md5cksum.data + md5cksum.len - KRB5_CKSUM_LENGTH, + KRB5_CKSUM_LENGTH); spin_lock(&krb5_seq_lock); seq_send = kctx->seq_send++; -- cgit v1.2.3 From ca54f896454852f0bc8d50e6e4c55d9defedbd0a Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Mon, 4 Dec 2006 20:22:38 -0500 Subject: rpcgss: simplify make_checksum We're doing some pointless translation between krb5 constants and kernel crypto string names. Also clean up some related spkm3 code as necessary. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 2 +- net/sunrpc/auth_gss/gss_krb5_crypto.c | 12 +----------- net/sunrpc/auth_gss/gss_krb5_seal.c | 2 +- net/sunrpc/auth_gss/gss_krb5_unseal.c | 3 +-- net/sunrpc/auth_gss/gss_krb5_wrap.c | 4 ++-- 5 files changed, 6 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index f680ed3b1b5e..abfa1f32d914 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -116,7 +116,7 @@ enum seal_alg { #define ENCTYPE_UNKNOWN 0x01ff s32 -make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, +make_checksum(char *, char *header, int hdrlen, struct xdr_buf *body, int body_offset, struct xdr_netobj *cksum); u32 gss_get_mic_kerberos(struct gss_ctx *, struct xdr_buf *, diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c index 10d05ea37213..d926cda88623 100644 --- a/net/sunrpc/auth_gss/gss_krb5_crypto.c +++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c @@ -130,23 +130,13 @@ checksummer(struct scatterlist *sg, void *data) /* checksum the plaintext data and hdrlen bytes of the token header */ s32 -make_checksum(s32 cksumtype, char *header, int hdrlen, struct xdr_buf *body, +make_checksum(char *cksumname, char *header, int hdrlen, struct xdr_buf *body, int body_offset, struct xdr_netobj *cksum) { - char *cksumname; struct hash_desc desc; /* XXX add to ctx? */ struct scatterlist sg[1]; int err; - switch (cksumtype) { - case CKSUMTYPE_RSA_MD5: - cksumname = "md5"; - break; - default: - dprintk("RPC: krb5_make_checksum:" - " unsupported checksum %d", cksumtype); - return GSS_S_FAILURE; - } desc.tfm = crypto_alloc_hash(cksumname, 0, CRYPTO_ALG_ASYNC); if (IS_ERR(desc.tfm)) return GSS_S_FAILURE; diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c index 2bc22776be55..c187f7f1520f 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seal.c +++ b/net/sunrpc/auth_gss/gss_krb5_seal.c @@ -108,7 +108,7 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, *(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5); memset(krb5_hdr + 4, 0xff, 4); - if (make_checksum(CKSUMTYPE_RSA_MD5, krb5_hdr, 8, text, 0, &md5cksum)) + if (make_checksum("md5", krb5_hdr, 8, text, 0, &md5cksum)) goto out_err; if (krb5_encrypt(ctx->seq, NULL, md5cksum.data, diff --git a/net/sunrpc/auth_gss/gss_krb5_unseal.c b/net/sunrpc/auth_gss/gss_krb5_unseal.c index 60469d9ab226..62807ac1e2ca 100644 --- a/net/sunrpc/auth_gss/gss_krb5_unseal.c +++ b/net/sunrpc/auth_gss/gss_krb5_unseal.c @@ -114,8 +114,7 @@ gss_verify_mic_kerberos(struct gss_ctx *gss_ctx, if (signalg != SGN_ALG_DES_MAC_MD5) goto out; - ret = make_checksum(CKSUMTYPE_RSA_MD5, ptr - 2, 8, - message_buffer, 0, &md5cksum); + ret = make_checksum("md5", ptr - 2, 8, message_buffer, 0, &md5cksum); if (ret) goto out; diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c index 206937187274..6d508d77adf9 100644 --- a/net/sunrpc/auth_gss/gss_krb5_wrap.c +++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c @@ -176,7 +176,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, /* XXXJBF: UGH!: */ tmp_pages = buf->pages; buf->pages = pages; - if (make_checksum(CKSUMTYPE_RSA_MD5, krb5_hdr, 8, buf, + if (make_checksum("md5", krb5_hdr, 8, buf, offset + headlen - blocksize, &md5cksum)) goto out_err; buf->pages = tmp_pages; @@ -272,7 +272,7 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf) ptr + 22 - (unsigned char *)buf->head[0].iov_base)) goto out; - ret = make_checksum(CKSUMTYPE_RSA_MD5, ptr - 2, 8, buf, + ret = make_checksum("md5", ptr - 2, 8, buf, ptr + 22 - (unsigned char *)buf->head[0].iov_base, &md5cksum); if (ret) goto out; -- cgit v1.2.3 From d922a84a8bf1d627810906d033223d4fa629fdbf Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Mon, 4 Dec 2006 20:22:40 -0500 Subject: rpcgss: krb5: sanity check sealalg value in the downcall The sealalg is checked in several places, giving the impression it could be either SEAL_ALG_NONE or SEAL_ALG_DES. But in fact SEAL_ALG_NONE seems to be sufficient only for making mic's, and all the contexts we get must be capable of wrapping as well. So the sealalg must be SEAL_ALG_DES. As with signalg, just check for the right value on the downcall and ignore it otherwise. Similarly, tighten expectations for the sealalg on incoming tokens, in case we do support other values eventually. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 1 - net/sunrpc/auth_gss/gss_krb5_mech.c | 4 +++- net/sunrpc/auth_gss/gss_krb5_seal.c | 6 ------ net/sunrpc/auth_gss/gss_krb5_wrap.c | 25 ++----------------------- 4 files changed, 5 insertions(+), 31 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index abfa1f32d914..01c5e4314632 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -44,7 +44,6 @@ struct krb5_ctx { int initiate; /* 1 = initiating, 0 = accepting */ int seed_init; unsigned char seed[16]; - int sealalg; struct crypto_blkcipher *enc; struct crypto_blkcipher *seq; s32 endtime; diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index 17587163fcae..bf5435db8785 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -148,9 +148,11 @@ gss_import_sec_context_kerberos(const void *p, goto out_err_free_ctx; if (tmp != SGN_ALG_DES_MAC_MD5) goto out_err_free_ctx; - p = simple_get_bytes(p, end, &ctx->sealalg, sizeof(ctx->sealalg)); + p = simple_get_bytes(p, end, &tmp, sizeof(tmp)); if (IS_ERR(p)) goto out_err_free_ctx; + if (tmp != SEAL_ALG_DES) + goto out_err_free_ctx; p = simple_get_bytes(p, end, &ctx->endtime, sizeof(ctx->endtime)); if (IS_ERR(p)) goto out_err_free_ctx; diff --git a/net/sunrpc/auth_gss/gss_krb5_seal.c b/net/sunrpc/auth_gss/gss_krb5_seal.c index f3f42a4465cf..f42e453e63ea 100644 --- a/net/sunrpc/auth_gss/gss_krb5_seal.c +++ b/net/sunrpc/auth_gss/gss_krb5_seal.c @@ -87,12 +87,6 @@ gss_get_mic_kerberos(struct gss_ctx *gss_ctx, struct xdr_buf *text, now = get_seconds(); - if (ctx->sealalg != SEAL_ALG_NONE && ctx->sealalg != SEAL_ALG_DES) { - dprintk("RPC: gss_krb5_seal: ctx->sealalg %d not supported\n", - ctx->sealalg); - return GSS_S_FAILURE; - } - token->len = g_token_size(&ctx->mech_used, 22); ptr = token->data; diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c index 63b06ee2d542..bf25f4d9acd1 100644 --- a/net/sunrpc/auth_gss/gss_krb5_wrap.c +++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c @@ -133,12 +133,6 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, now = get_seconds(); - if (kctx->sealalg != SEAL_ALG_NONE && kctx->sealalg != SEAL_ALG_DES) { - dprintk("RPC: gss_krb5_seal: kctx->sealalg %d not supported\n", - kctx->sealalg); - return GSS_S_FAILURE; - } - blocksize = crypto_blkcipher_blocksize(kctx->enc); gss_krb5_add_padding(buf, offset, blocksize); BUG_ON((buf->len - offset) % blocksize); @@ -169,7 +163,7 @@ gss_wrap_kerberos(struct gss_ctx *ctx, int offset, *(__be16 *)(krb5_hdr + 2) = htons(SGN_ALG_DES_MAC_MD5); memset(krb5_hdr + 4, 0xff, 4); - *(__be16 *)(krb5_hdr + 4) = htons(kctx->sealalg); + *(__be16 *)(krb5_hdr + 4) = htons(SEAL_ALG_DES); make_confounder(msg_start, blocksize); @@ -245,26 +239,11 @@ gss_unwrap_kerberos(struct gss_ctx *ctx, int offset, struct xdr_buf *buf) if ((ptr[4] != 0xff) || (ptr[5] != 0xff)) return GSS_S_DEFECTIVE_TOKEN; - if (sealalg == 0xffff) + if (sealalg != SEAL_ALG_DES) return GSS_S_DEFECTIVE_TOKEN; if (signalg != SGN_ALG_DES_MAC_MD5) return GSS_S_DEFECTIVE_TOKEN; - /* in the current spec, there is only one valid seal algorithm per - key type, so a simple comparison is ok */ - - if (sealalg != kctx->sealalg) - return GSS_S_DEFECTIVE_TOKEN; - - /* there are several mappings of seal algorithms to sign algorithms, - but few enough that we can try them all. */ - - if ((kctx->sealalg == SEAL_ALG_NONE && signalg > 1) || - (kctx->sealalg == SEAL_ALG_1 && signalg != SGN_ALG_3) || - (kctx->sealalg == SEAL_ALG_DES3KD && - signalg != SGN_ALG_HMAC_SHA1_DES3_KD)) - return GSS_S_DEFECTIVE_TOKEN; - if (gss_decrypt_xdr_buf(kctx->enc, buf, ptr + 22 - (unsigned char *)buf->head[0].iov_base)) return GSS_S_DEFECTIVE_TOKEN; -- cgit v1.2.3 From 717757ad1038ab6aacb89bad579c89b006efd913 Mon Sep 17 00:00:00 2001 From: J. Bruce Fields Date: Mon, 4 Dec 2006 20:22:41 -0500 Subject: rpcgss: krb5: ignore seed We're currently not actually using seed or seed_init. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/sunrpc/gss_krb5.h | 2 -- net/sunrpc/auth_gss/gss_krb5_mech.c | 11 ++++++----- 2 files changed, 6 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/gss_krb5.h b/include/linux/sunrpc/gss_krb5.h index 01c5e4314632..5a4b1e0206e3 100644 --- a/include/linux/sunrpc/gss_krb5.h +++ b/include/linux/sunrpc/gss_krb5.h @@ -42,8 +42,6 @@ struct krb5_ctx { int initiate; /* 1 = initiating, 0 = accepting */ - int seed_init; - unsigned char seed[16]; struct crypto_blkcipher *enc; struct crypto_blkcipher *seq; s32 endtime; diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index bf5435db8785..05d4bee86fc0 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -137,12 +137,13 @@ gss_import_sec_context_kerberos(const void *p, p = simple_get_bytes(p, end, &ctx->initiate, sizeof(ctx->initiate)); if (IS_ERR(p)) goto out_err_free_ctx; - p = simple_get_bytes(p, end, &ctx->seed_init, sizeof(ctx->seed_init)); - if (IS_ERR(p)) - goto out_err_free_ctx; - p = simple_get_bytes(p, end, ctx->seed, sizeof(ctx->seed)); - if (IS_ERR(p)) + /* The downcall format was designed before we completely understood + * the uses of the context fields; so it includes some stuff we + * just give some minimal sanity-checking, and some we ignore + * completely (like the next twenty bytes): */ + if (unlikely(p + 20 > end || p + 20 < p)) goto out_err_free_ctx; + p += 20; p = simple_get_bytes(p, end, &tmp, sizeof(tmp)); if (IS_ERR(p)) goto out_err_free_ctx; -- cgit v1.2.3 From ee0ac0c227c2a2b6dd1b33c23831100ee895dacf Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:15 -0500 Subject: SUNRPC: Remove sock and inet fields from rpc_xprt The "sock" and "inet" fields are socket-specific. Move them to a private data structure maintained entirely within net/sunrpc/xprtsock.c Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 4 +-- net/sunrpc/xprtsock.c | 62 ++++++++++++++++++++++++++++----------------- 2 files changed, 41 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index d7919010863d..4c074a73670c 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -17,6 +17,8 @@ #include #include +#include + extern unsigned int xprt_udp_slot_table_entries; extern unsigned int xprt_tcp_slot_table_entries; @@ -126,8 +128,6 @@ struct rpc_xprt_ops { struct rpc_xprt { struct kref kref; /* Reference count */ struct rpc_xprt_ops * ops; /* transport methods */ - struct socket * sock; /* BSD socket layer */ - struct sock * inet; /* INET layer */ struct rpc_timeout timeout; /* timeout parms */ struct sockaddr_storage addr; /* server address */ diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index dc4a21f1a129..0f455fd1820c 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -127,6 +127,12 @@ static inline void xs_pktdump(char *msg, u32 *packet, unsigned int count) struct sock_xprt { struct rpc_xprt xprt; + + /* + * Network layer + */ + struct socket * sock; + struct sock * inet; }; static void xs_format_peer_addresses(struct rpc_xprt *xprt) @@ -285,19 +291,20 @@ static void xs_nospace(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); dprintk("RPC: %4d xmit incomplete (%u left of %u)\n", task->tk_pid, req->rq_slen - req->rq_bytes_sent, req->rq_slen); - if (test_bit(SOCK_ASYNC_NOSPACE, &xprt->sock->flags)) { + if (test_bit(SOCK_ASYNC_NOSPACE, &transport->sock->flags)) { /* Protect against races with write_space */ spin_lock_bh(&xprt->transport_lock); /* Don't race with disconnect */ if (!xprt_connected(xprt)) task->tk_status = -ENOTCONN; - else if (test_bit(SOCK_NOSPACE, &xprt->sock->flags)) + else if (test_bit(SOCK_NOSPACE, &transport->sock->flags)) xprt_wait_for_buffer_space(task); spin_unlock_bh(&xprt->transport_lock); @@ -321,6 +328,7 @@ static int xs_udp_send_request(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct xdr_buf *xdr = &req->rq_snd_buf; int status; @@ -329,8 +337,10 @@ static int xs_udp_send_request(struct rpc_task *task) req->rq_svec->iov_len); req->rq_xtime = jiffies; - status = xs_sendpages(xprt->sock, (struct sockaddr *) &xprt->addr, - xprt->addrlen, xdr, req->rq_bytes_sent); + status = xs_sendpages(transport->sock, + (struct sockaddr *) &xprt->addr, + xprt->addrlen, xdr, + req->rq_bytes_sent); dprintk("RPC: xs_udp_send_request(%u) = %d\n", xdr->len - req->rq_bytes_sent, status); @@ -386,6 +396,7 @@ static int xs_tcp_send_request(struct rpc_task *task) { struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct xdr_buf *xdr = &req->rq_snd_buf; int status, retry = 0; @@ -400,8 +411,8 @@ static int xs_tcp_send_request(struct rpc_task *task) * called sendmsg(). */ while (1) { req->rq_xtime = jiffies; - status = xs_sendpages(xprt->sock, NULL, 0, xdr, - req->rq_bytes_sent); + status = xs_sendpages(transport->sock, + NULL, 0, xdr, req->rq_bytes_sent); dprintk("RPC: xs_tcp_send_request(%u) = %d\n", xdr->len - req->rq_bytes_sent, status); @@ -479,8 +490,9 @@ out_release: */ static void xs_close(struct rpc_xprt *xprt) { - struct socket *sock = xprt->sock; - struct sock *sk = xprt->inet; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct socket *sock = transport->sock; + struct sock *sk = transport->inet; if (!sk) goto clear_close_wait; @@ -488,8 +500,8 @@ static void xs_close(struct rpc_xprt *xprt) dprintk("RPC: xs_close xprt %p\n", xprt); write_lock_bh(&sk->sk_callback_lock); - xprt->inet = NULL; - xprt->sock = NULL; + transport->inet = NULL; + transport->sock = NULL; sk->sk_user_data = NULL; sk->sk_data_ready = xprt->old_data_ready; @@ -946,7 +958,8 @@ static void xs_tcp_write_space(struct sock *sk) static void xs_udp_do_set_buffer_size(struct rpc_xprt *xprt) { - struct sock *sk = xprt->inet; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct sock *sk = transport->inet; if (xprt->rcvsize) { sk->sk_userlocks |= SOCK_RCVBUF_LOCK; @@ -1062,7 +1075,8 @@ static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock) static void xs_udp_connect_worker(void *args) { struct rpc_xprt *xprt = (struct rpc_xprt *) args; - struct socket *sock = xprt->sock; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct socket *sock = transport->sock; int err, status = -EIO; if (xprt->shutdown || !xprt_bound(xprt)) @@ -1084,7 +1098,7 @@ static void xs_udp_connect_worker(void *args) dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); - if (!xprt->inet) { + if (!transport->inet) { struct sock *sk = sock->sk; write_lock_bh(&sk->sk_callback_lock); @@ -1101,8 +1115,8 @@ static void xs_udp_connect_worker(void *args) xprt_set_connected(xprt); /* Reset to new socket */ - xprt->sock = sock; - xprt->inet = sk; + transport->sock = sock; + transport->inet = sk; write_unlock_bh(&sk->sk_callback_lock); } @@ -1120,7 +1134,7 @@ out: static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) { int result; - struct socket *sock = xprt->sock; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct sockaddr any; dprintk("RPC: disconnecting xprt %p to reuse port\n", xprt); @@ -1131,7 +1145,7 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) */ memset(&any, 0, sizeof(any)); any.sa_family = AF_UNSPEC; - result = kernel_connect(sock, &any, sizeof(any), 0); + result = kernel_connect(transport->sock, &any, sizeof(any), 0); if (result) dprintk("RPC: AF_UNSPEC connect return code %d\n", result); @@ -1146,13 +1160,14 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) static void xs_tcp_connect_worker(void *args) { struct rpc_xprt *xprt = (struct rpc_xprt *)args; - struct socket *sock = xprt->sock; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct socket *sock = transport->sock; int err, status = -EIO; if (xprt->shutdown || !xprt_bound(xprt)) goto out; - if (!xprt->sock) { + if (!sock) { /* start from scratch */ if ((err = sock_create_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock)) < 0) { dprintk("RPC: can't create TCP transport socket (%d).\n", -err); @@ -1170,7 +1185,7 @@ static void xs_tcp_connect_worker(void *args) dprintk("RPC: worker connecting xprt %p to address: %s\n", xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); - if (!xprt->inet) { + if (!transport->inet) { struct sock *sk = sock->sk; write_lock_bh(&sk->sk_callback_lock); @@ -1193,8 +1208,8 @@ static void xs_tcp_connect_worker(void *args) xprt_clear_connected(xprt); /* Reset to new socket */ - xprt->sock = sock; - xprt->inet = sk; + transport->sock = sock; + transport->inet = sk; write_unlock_bh(&sk->sk_callback_lock); } @@ -1243,11 +1258,12 @@ out_clear: static void xs_connect(struct rpc_task *task) { struct rpc_xprt *xprt = task->tk_xprt; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); if (xprt_test_and_set_connecting(xprt)) return; - if (xprt->sock != NULL) { + if (transport->sock != NULL) { dprintk("RPC: xs_connect delayed xprt %p for %lu seconds\n", xprt, xprt->reestablish_timeout / HZ); schedule_delayed_work(&xprt->connect_worker, -- cgit v1.2.3 From 51971139b2342fa1005e87bbfcb52305da3fe891 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:19 -0500 Subject: SUNRPC: Move TCP receive state variables into private data structure Move the TCP receive state variables from the generic rpc_xprt structure to a private structure maintained inside net/sunrpc/xprtsock.c. Also rename a function/variable pair to refer to RPC fragment headers instead of record markers, to be consistent with types defined in sunrpc/*.h. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 9 --- net/sunrpc/xprtsock.c | 154 +++++++++++++++++++++++++------------------- 2 files changed, 86 insertions(+), 77 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 4c074a73670c..0a0ad1ce70e4 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -157,15 +157,6 @@ struct rpc_xprt { unsigned char shutdown : 1, /* being shut down */ resvport : 1; /* use a reserved port */ - /* - * State of TCP reply receive stuff - */ - __be32 tcp_recm, /* Fragment header */ - tcp_xid; /* Current XID */ - u32 tcp_reclen, /* fragment length */ - tcp_offset; /* fragment offset */ - unsigned long tcp_copied, /* copied to request */ - tcp_flags; /* * Connection of transports */ diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 0f455fd1820c..417bd91ea53e 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -133,6 +133,18 @@ struct sock_xprt { */ struct socket * sock; struct sock * inet; + + /* + * State of TCP reply receive + */ + __be32 tcp_fraghdr, + tcp_xid; + + u32 tcp_offset, + tcp_reclen; + + unsigned long tcp_copied, + tcp_flags; }; static void xs_format_peer_addresses(struct rpc_xprt *xprt) @@ -628,73 +640,73 @@ static inline size_t xs_tcp_copy_data(skb_reader_t *desc, void *p, size_t len) static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc) { + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); size_t len, used; char *p; - p = ((char *) &xprt->tcp_recm) + xprt->tcp_offset; - len = sizeof(xprt->tcp_recm) - xprt->tcp_offset; + p = ((char *) &transport->tcp_fraghdr) + transport->tcp_offset; + len = sizeof(transport->tcp_fraghdr) - transport->tcp_offset; used = xs_tcp_copy_data(desc, p, len); - xprt->tcp_offset += used; + transport->tcp_offset += used; if (used != len) return; - xprt->tcp_reclen = ntohl(xprt->tcp_recm); - if (xprt->tcp_reclen & RPC_LAST_STREAM_FRAGMENT) - xprt->tcp_flags |= XPRT_LAST_FRAG; + transport->tcp_reclen = ntohl(transport->tcp_fraghdr); + if (transport->tcp_reclen & RPC_LAST_STREAM_FRAGMENT) + transport->tcp_flags |= XPRT_LAST_FRAG; else - xprt->tcp_flags &= ~XPRT_LAST_FRAG; - xprt->tcp_reclen &= RPC_FRAGMENT_SIZE_MASK; + transport->tcp_flags &= ~XPRT_LAST_FRAG; + transport->tcp_reclen &= RPC_FRAGMENT_SIZE_MASK; - xprt->tcp_flags &= ~XPRT_COPY_RECM; - xprt->tcp_offset = 0; + transport->tcp_flags &= ~XPRT_COPY_RECM; + transport->tcp_offset = 0; /* Sanity check of the record length */ - if (unlikely(xprt->tcp_reclen < 4)) { + if (unlikely(transport->tcp_reclen < 4)) { dprintk("RPC: invalid TCP record fragment length\n"); xprt_disconnect(xprt); return; } dprintk("RPC: reading TCP record fragment of length %d\n", - xprt->tcp_reclen); + transport->tcp_reclen); } -static void xs_tcp_check_recm(struct rpc_xprt *xprt) +static void xs_tcp_check_fraghdr(struct sock_xprt *transport) { - dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u, tcp_flags = %lx\n", - xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen, xprt->tcp_flags); - if (xprt->tcp_offset == xprt->tcp_reclen) { - xprt->tcp_flags |= XPRT_COPY_RECM; - xprt->tcp_offset = 0; - if (xprt->tcp_flags & XPRT_LAST_FRAG) { - xprt->tcp_flags &= ~XPRT_COPY_DATA; - xprt->tcp_flags |= XPRT_COPY_XID; - xprt->tcp_copied = 0; + if (transport->tcp_offset == transport->tcp_reclen) { + transport->tcp_flags |= XPRT_COPY_RECM; + transport->tcp_offset = 0; + if (transport->tcp_flags & XPRT_LAST_FRAG) { + transport->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags |= XPRT_COPY_XID; + transport->tcp_copied = 0; } } } -static inline void xs_tcp_read_xid(struct rpc_xprt *xprt, skb_reader_t *desc) +static inline void xs_tcp_read_xid(struct sock_xprt *transport, skb_reader_t *desc) { size_t len, used; char *p; - len = sizeof(xprt->tcp_xid) - xprt->tcp_offset; + len = sizeof(transport->tcp_xid) - transport->tcp_offset; dprintk("RPC: reading XID (%Zu bytes)\n", len); - p = ((char *) &xprt->tcp_xid) + xprt->tcp_offset; + p = ((char *) &transport->tcp_xid) + transport->tcp_offset; used = xs_tcp_copy_data(desc, p, len); - xprt->tcp_offset += used; + transport->tcp_offset += used; if (used != len) return; - xprt->tcp_flags &= ~XPRT_COPY_XID; - xprt->tcp_flags |= XPRT_COPY_DATA; - xprt->tcp_copied = 4; + transport->tcp_flags &= ~XPRT_COPY_XID; + transport->tcp_flags |= XPRT_COPY_DATA; + transport->tcp_copied = 4; dprintk("RPC: reading reply for XID %08x\n", - ntohl(xprt->tcp_xid)); - xs_tcp_check_recm(xprt); + ntohl(transport->tcp_xid)); + xs_tcp_check_fraghdr(transport); } static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) { + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct rpc_rqst *req; struct xdr_buf *rcvbuf; size_t len; @@ -702,34 +714,34 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc /* Find and lock the request corresponding to this xid */ spin_lock(&xprt->transport_lock); - req = xprt_lookup_rqst(xprt, xprt->tcp_xid); + req = xprt_lookup_rqst(xprt, transport->tcp_xid); if (!req) { - xprt->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags &= ~XPRT_COPY_DATA; dprintk("RPC: XID %08x request not found!\n", - ntohl(xprt->tcp_xid)); + ntohl(transport->tcp_xid)); spin_unlock(&xprt->transport_lock); return; } rcvbuf = &req->rq_private_buf; len = desc->count; - if (len > xprt->tcp_reclen - xprt->tcp_offset) { + if (len > transport->tcp_reclen - transport->tcp_offset) { skb_reader_t my_desc; - len = xprt->tcp_reclen - xprt->tcp_offset; + len = transport->tcp_reclen - transport->tcp_offset; memcpy(&my_desc, desc, sizeof(my_desc)); my_desc.count = len; - r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, + r = xdr_partial_copy_from_skb(rcvbuf, transport->tcp_copied, &my_desc, xs_tcp_copy_data); desc->count -= r; desc->offset += r; } else - r = xdr_partial_copy_from_skb(rcvbuf, xprt->tcp_copied, + r = xdr_partial_copy_from_skb(rcvbuf, transport->tcp_copied, desc, xs_tcp_copy_data); if (r > 0) { - xprt->tcp_copied += r; - xprt->tcp_offset += r; + transport->tcp_copied += r; + transport->tcp_offset += r; } if (r != len) { /* Error when copying to the receive buffer, @@ -741,77 +753,79 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc * Any remaining data from this record will * be discarded. */ - xprt->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags &= ~XPRT_COPY_DATA; dprintk("RPC: XID %08x truncated request\n", - ntohl(xprt->tcp_xid)); + ntohl(transport->tcp_xid)); dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n", - xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen); + xprt, transport->tcp_copied, transport->tcp_offset, + transport->tcp_reclen); goto out; } dprintk("RPC: XID %08x read %Zd bytes\n", - ntohl(xprt->tcp_xid), r); + ntohl(transport->tcp_xid), r); dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n", - xprt, xprt->tcp_copied, xprt->tcp_offset, xprt->tcp_reclen); - - if (xprt->tcp_copied == req->rq_private_buf.buflen) - xprt->tcp_flags &= ~XPRT_COPY_DATA; - else if (xprt->tcp_offset == xprt->tcp_reclen) { - if (xprt->tcp_flags & XPRT_LAST_FRAG) - xprt->tcp_flags &= ~XPRT_COPY_DATA; + xprt, transport->tcp_copied, transport->tcp_offset, + transport->tcp_reclen); + + if (transport->tcp_copied == req->rq_private_buf.buflen) + transport->tcp_flags &= ~XPRT_COPY_DATA; + else if (transport->tcp_offset == transport->tcp_reclen) { + if (transport->tcp_flags & XPRT_LAST_FRAG) + transport->tcp_flags &= ~XPRT_COPY_DATA; } out: - if (!(xprt->tcp_flags & XPRT_COPY_DATA)) - xprt_complete_rqst(req->rq_task, xprt->tcp_copied); + if (!(transport->tcp_flags & XPRT_COPY_DATA)) + xprt_complete_rqst(req->rq_task, transport->tcp_copied); spin_unlock(&xprt->transport_lock); - xs_tcp_check_recm(xprt); + xs_tcp_check_fraghdr(transport); } -static inline void xs_tcp_read_discard(struct rpc_xprt *xprt, skb_reader_t *desc) +static inline void xs_tcp_read_discard(struct sock_xprt *transport, skb_reader_t *desc) { size_t len; - len = xprt->tcp_reclen - xprt->tcp_offset; + len = transport->tcp_reclen - transport->tcp_offset; if (len > desc->count) len = desc->count; desc->count -= len; desc->offset += len; - xprt->tcp_offset += len; + transport->tcp_offset += len; dprintk("RPC: discarded %Zu bytes\n", len); - xs_tcp_check_recm(xprt); + xs_tcp_check_fraghdr(transport); } static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, unsigned int offset, size_t len) { struct rpc_xprt *xprt = rd_desc->arg.data; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); skb_reader_t desc = { .skb = skb, .offset = offset, .count = len, - .csum = 0 }; dprintk("RPC: xs_tcp_data_recv started\n"); do { /* Read in a new fragment marker if necessary */ /* Can we ever really expect to get completely empty fragments? */ - if (xprt->tcp_flags & XPRT_COPY_RECM) { + if (transport->tcp_flags & XPRT_COPY_RECM) { xs_tcp_read_fraghdr(xprt, &desc); continue; } /* Read in the xid if necessary */ - if (xprt->tcp_flags & XPRT_COPY_XID) { - xs_tcp_read_xid(xprt, &desc); + if (transport->tcp_flags & XPRT_COPY_XID) { + xs_tcp_read_xid(transport, &desc); continue; } /* Read in the request data */ - if (xprt->tcp_flags & XPRT_COPY_DATA) { + if (transport->tcp_flags & XPRT_COPY_DATA) { xs_tcp_read_request(xprt, &desc); continue; } /* Skip over any trailing bytes on short reads */ - xs_tcp_read_discard(xprt, &desc); + xs_tcp_read_discard(transport, &desc); } while (desc.count); dprintk("RPC: xs_tcp_data_recv done\n"); return len - desc.count; @@ -865,11 +879,15 @@ static void xs_tcp_state_change(struct sock *sk) case TCP_ESTABLISHED: spin_lock_bh(&xprt->transport_lock); if (!xprt_test_and_set_connected(xprt)) { + struct sock_xprt *transport = container_of(xprt, + struct sock_xprt, xprt); + /* Reset TCP record info */ - xprt->tcp_offset = 0; - xprt->tcp_reclen = 0; - xprt->tcp_copied = 0; - xprt->tcp_flags = XPRT_COPY_RECM | XPRT_COPY_XID; + transport->tcp_offset = 0; + transport->tcp_reclen = 0; + transport->tcp_copied = 0; + transport->tcp_flags = XPRT_COPY_RECM | XPRT_COPY_XID; + xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; xprt_wake_pending_tasks(xprt, 0); } -- cgit v1.2.3 From e136d0926ef6a048f6e65b35263c0a9faae3abbe Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:23 -0500 Subject: SUNRPC: Move TCP state flags into xprtsock.c Move "XPRT_LAST_FRAG" and friends from xprt.h into xprtsock.c, and rename them to use the naming scheme in use in xprtsock.c. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 5 ----- net/sunrpc/xprtsock.c | 49 +++++++++++++++++++++++++++------------------ 2 files changed, 29 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 0a0ad1ce70e4..3ff8230c7dd9 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -203,11 +203,6 @@ struct rpc_xprt { char * address_strings[RPC_DISPLAY_MAX]; }; -#define XPRT_LAST_FRAG (1 << 0) -#define XPRT_COPY_RECM (1 << 1) -#define XPRT_COPY_XID (1 << 2) -#define XPRT_COPY_DATA (1 << 3) - #ifdef __KERNEL__ /* diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 417bd91ea53e..c737acf61c75 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -147,6 +147,14 @@ struct sock_xprt { tcp_flags; }; +/* + * TCP receive state flags + */ +#define TCP_RCV_LAST_FRAG (1UL << 0) +#define TCP_RCV_COPY_FRAGHDR (1UL << 1) +#define TCP_RCV_COPY_XID (1UL << 2) +#define TCP_RCV_COPY_DATA (1UL << 3) + static void xs_format_peer_addresses(struct rpc_xprt *xprt) { struct sockaddr_in *addr = (struct sockaddr_in *) &xprt->addr; @@ -653,12 +661,12 @@ static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc transport->tcp_reclen = ntohl(transport->tcp_fraghdr); if (transport->tcp_reclen & RPC_LAST_STREAM_FRAGMENT) - transport->tcp_flags |= XPRT_LAST_FRAG; + transport->tcp_flags |= TCP_RCV_LAST_FRAG; else - transport->tcp_flags &= ~XPRT_LAST_FRAG; + transport->tcp_flags &= ~TCP_RCV_LAST_FRAG; transport->tcp_reclen &= RPC_FRAGMENT_SIZE_MASK; - transport->tcp_flags &= ~XPRT_COPY_RECM; + transport->tcp_flags &= ~TCP_RCV_COPY_FRAGHDR; transport->tcp_offset = 0; /* Sanity check of the record length */ @@ -674,11 +682,11 @@ static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc static void xs_tcp_check_fraghdr(struct sock_xprt *transport) { if (transport->tcp_offset == transport->tcp_reclen) { - transport->tcp_flags |= XPRT_COPY_RECM; + transport->tcp_flags |= TCP_RCV_COPY_FRAGHDR; transport->tcp_offset = 0; - if (transport->tcp_flags & XPRT_LAST_FRAG) { - transport->tcp_flags &= ~XPRT_COPY_DATA; - transport->tcp_flags |= XPRT_COPY_XID; + if (transport->tcp_flags & TCP_RCV_LAST_FRAG) { + transport->tcp_flags &= ~TCP_RCV_COPY_DATA; + transport->tcp_flags |= TCP_RCV_COPY_XID; transport->tcp_copied = 0; } } @@ -696,8 +704,8 @@ static inline void xs_tcp_read_xid(struct sock_xprt *transport, skb_reader_t *de transport->tcp_offset += used; if (used != len) return; - transport->tcp_flags &= ~XPRT_COPY_XID; - transport->tcp_flags |= XPRT_COPY_DATA; + transport->tcp_flags &= ~TCP_RCV_COPY_XID; + transport->tcp_flags |= TCP_RCV_COPY_DATA; transport->tcp_copied = 4; dprintk("RPC: reading reply for XID %08x\n", ntohl(transport->tcp_xid)); @@ -716,7 +724,7 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc spin_lock(&xprt->transport_lock); req = xprt_lookup_rqst(xprt, transport->tcp_xid); if (!req) { - transport->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags &= ~TCP_RCV_COPY_DATA; dprintk("RPC: XID %08x request not found!\n", ntohl(transport->tcp_xid)); spin_unlock(&xprt->transport_lock); @@ -747,13 +755,13 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc /* Error when copying to the receive buffer, * usually because we weren't able to allocate * additional buffer pages. All we can do now - * is turn off XPRT_COPY_DATA, so the request + * is turn off TCP_RCV_COPY_DATA, so the request * will not receive any additional updates, * and time out. * Any remaining data from this record will * be discarded. */ - transport->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags &= ~TCP_RCV_COPY_DATA; dprintk("RPC: XID %08x truncated request\n", ntohl(transport->tcp_xid)); dprintk("RPC: xprt = %p, tcp_copied = %lu, tcp_offset = %u, tcp_reclen = %u\n", @@ -769,14 +777,14 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc transport->tcp_reclen); if (transport->tcp_copied == req->rq_private_buf.buflen) - transport->tcp_flags &= ~XPRT_COPY_DATA; + transport->tcp_flags &= ~TCP_RCV_COPY_DATA; else if (transport->tcp_offset == transport->tcp_reclen) { - if (transport->tcp_flags & XPRT_LAST_FRAG) - transport->tcp_flags &= ~XPRT_COPY_DATA; + if (transport->tcp_flags & TCP_RCV_LAST_FRAG) + transport->tcp_flags &= ~TCP_RCV_COPY_DATA; } out: - if (!(transport->tcp_flags & XPRT_COPY_DATA)) + if (!(transport->tcp_flags & TCP_RCV_COPY_DATA)) xprt_complete_rqst(req->rq_task, transport->tcp_copied); spin_unlock(&xprt->transport_lock); xs_tcp_check_fraghdr(transport); @@ -810,17 +818,17 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns do { /* Read in a new fragment marker if necessary */ /* Can we ever really expect to get completely empty fragments? */ - if (transport->tcp_flags & XPRT_COPY_RECM) { + if (transport->tcp_flags & TCP_RCV_COPY_FRAGHDR) { xs_tcp_read_fraghdr(xprt, &desc); continue; } /* Read in the xid if necessary */ - if (transport->tcp_flags & XPRT_COPY_XID) { + if (transport->tcp_flags & TCP_RCV_COPY_XID) { xs_tcp_read_xid(transport, &desc); continue; } /* Read in the request data */ - if (transport->tcp_flags & XPRT_COPY_DATA) { + if (transport->tcp_flags & TCP_RCV_COPY_DATA) { xs_tcp_read_request(xprt, &desc); continue; } @@ -886,7 +894,8 @@ static void xs_tcp_state_change(struct sock *sk) transport->tcp_offset = 0; transport->tcp_reclen = 0; transport->tcp_copied = 0; - transport->tcp_flags = XPRT_COPY_RECM | XPRT_COPY_XID; + transport->tcp_flags = + TCP_RCV_COPY_FRAGHDR | TCP_RCV_COPY_XID; xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; xprt_wake_pending_tasks(xprt, 0); -- cgit v1.2.3 From c8475461829fd94f30208fbfa4eab7e5584c6495 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:26 -0500 Subject: SUNRPC: Move rpc_xprt socket connect fields into private data structure Move the socket-specific connection management fields out of the generic rpc_xprt structure into a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 2 -- net/sunrpc/xprtsock.c | 51 +++++++++++++++++++++++++++++---------------- 2 files changed, 33 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 3ff8230c7dd9..18bf72c06488 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -163,8 +163,6 @@ struct rpc_xprt { unsigned long connect_timeout, bind_timeout, reestablish_timeout; - struct work_struct connect_worker; - unsigned short port; /* * Disconnection of idle transports diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index c737acf61c75..4797a4608c07 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -145,6 +145,12 @@ struct sock_xprt { unsigned long tcp_copied, tcp_flags; + + /* + * Connection of transports + */ + struct work_struct connect_worker; + unsigned short port; }; /* @@ -545,9 +551,11 @@ clear_close_wait: */ static void xs_destroy(struct rpc_xprt *xprt) { + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + dprintk("RPC: xs_destroy xprt %p\n", xprt); - cancel_delayed_work(&xprt->connect_worker); + cancel_delayed_work(&transport->connect_worker); flush_scheduled_work(); xprt_disconnect(xprt); @@ -1065,20 +1073,20 @@ static void xs_set_port(struct rpc_xprt *xprt, unsigned short port) sap->sin_port = htons(port); } -static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock) +static int xs_bindresvport(struct sock_xprt *transport, struct socket *sock) { struct sockaddr_in myaddr = { .sin_family = AF_INET, }; int err; - unsigned short port = xprt->port; + unsigned short port = transport->port; do { myaddr.sin_port = htons(port); err = kernel_bind(sock, (struct sockaddr *) &myaddr, sizeof(myaddr)); if (err == 0) { - xprt->port = port; + transport->port = port; dprintk("RPC: xs_bindresvport bound to port %u\n", port); return 0; @@ -1087,7 +1095,7 @@ static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock) port = xprt_max_resvport; else port--; - } while (err == -EADDRINUSE && port != xprt->port); + } while (err == -EADDRINUSE && port != transport->port); dprintk("RPC: can't bind to reserved port (%d).\n", -err); return err; @@ -1101,8 +1109,8 @@ static int xs_bindresvport(struct rpc_xprt *xprt, struct socket *sock) */ static void xs_udp_connect_worker(void *args) { - struct rpc_xprt *xprt = (struct rpc_xprt *) args; - struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct sock_xprt *transport = (struct sock_xprt *)args; + struct rpc_xprt *xprt = &transport->xprt; struct socket *sock = transport->sock; int err, status = -EIO; @@ -1117,7 +1125,7 @@ static void xs_udp_connect_worker(void *args) goto out; } - if (xprt->resvport && xs_bindresvport(xprt, sock) < 0) { + if (xprt->resvport && xs_bindresvport(transport, sock) < 0) { sock_release(sock); goto out; } @@ -1186,8 +1194,8 @@ static void xs_tcp_reuse_connection(struct rpc_xprt *xprt) */ static void xs_tcp_connect_worker(void *args) { - struct rpc_xprt *xprt = (struct rpc_xprt *)args; - struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + struct sock_xprt *transport = (struct sock_xprt *)args; + struct rpc_xprt *xprt = &transport->xprt; struct socket *sock = transport->sock; int err, status = -EIO; @@ -1201,7 +1209,7 @@ static void xs_tcp_connect_worker(void *args) goto out; } - if (xprt->resvport && xs_bindresvport(xprt, sock) < 0) { + if (xprt->resvport && xs_bindresvport(transport, sock) < 0) { sock_release(sock); goto out; } @@ -1293,14 +1301,14 @@ static void xs_connect(struct rpc_task *task) if (transport->sock != NULL) { dprintk("RPC: xs_connect delayed xprt %p for %lu seconds\n", xprt, xprt->reestablish_timeout / HZ); - schedule_delayed_work(&xprt->connect_worker, + schedule_delayed_work(&transport->connect_worker, xprt->reestablish_timeout); xprt->reestablish_timeout <<= 1; if (xprt->reestablish_timeout > XS_TCP_MAX_REEST_TO) xprt->reestablish_timeout = XS_TCP_MAX_REEST_TO; } else { dprintk("RPC: xs_connect scheduled xprt %p\n", xprt); - schedule_work(&xprt->connect_worker); + schedule_work(&transport->connect_worker); /* flush_scheduled_work can sleep... */ if (!RPC_IS_ASYNC(task)) @@ -1316,8 +1324,10 @@ static void xs_connect(struct rpc_task *task) */ static void xs_udp_print_stats(struct rpc_xprt *xprt, struct seq_file *seq) { + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + seq_printf(seq, "\txprt:\tudp %u %lu %lu %lu %lu %Lu %Lu\n", - xprt->port, + transport->port, xprt->stat.bind_count, xprt->stat.sends, xprt->stat.recvs, @@ -1334,13 +1344,14 @@ static void xs_udp_print_stats(struct rpc_xprt *xprt, struct seq_file *seq) */ static void xs_tcp_print_stats(struct rpc_xprt *xprt, struct seq_file *seq) { + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); long idle_time = 0; if (xprt_connected(xprt)) idle_time = (long)(jiffies - xprt->last_used) / HZ; seq_printf(seq, "\txprt:\ttcp %u %lu %lu %lu %ld %lu %lu %lu %Lu %Lu\n", - xprt->port, + transport->port, xprt->stat.bind_count, xprt->stat.connect_count, xprt->stat.connect_time, @@ -1414,7 +1425,7 @@ static struct rpc_xprt *xs_setup_xprt(struct sockaddr *addr, size_t addrlen, uns memcpy(&xprt->addr, addr, addrlen); xprt->addrlen = addrlen; - xprt->port = xs_get_random_port(); + new->port = xs_get_random_port(); return xprt; } @@ -1429,10 +1440,12 @@ static struct rpc_xprt *xs_setup_xprt(struct sockaddr *addr, size_t addrlen, uns struct rpc_xprt *xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to) { struct rpc_xprt *xprt; + struct sock_xprt *transport; xprt = xs_setup_xprt(addr, addrlen, xprt_udp_slot_table_entries); if (IS_ERR(xprt)) return xprt; + transport = container_of(xprt, struct sock_xprt, xprt); if (ntohs(((struct sockaddr_in *)addr)->sin_port) != 0) xprt_set_bound(xprt); @@ -1442,7 +1455,7 @@ struct rpc_xprt *xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_ /* XXX: header size can vary due to auth type, IPv6, etc. */ xprt->max_payload = (1U << 16) - (MAX_HEADER << 3); - INIT_WORK(&xprt->connect_worker, xs_udp_connect_worker, xprt); + INIT_WORK(&transport->connect_worker, xs_udp_connect_worker, transport); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_UDP_CONN_TO; xprt->reestablish_timeout = XS_UDP_REEST_TO; @@ -1472,10 +1485,12 @@ struct rpc_xprt *xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_ struct rpc_xprt *xs_setup_tcp(struct sockaddr *addr, size_t addrlen, struct rpc_timeout *to) { struct rpc_xprt *xprt; + struct sock_xprt *transport; xprt = xs_setup_xprt(addr, addrlen, xprt_tcp_slot_table_entries); if (IS_ERR(xprt)) return xprt; + transport = container_of(xprt, struct sock_xprt, xprt); if (ntohs(((struct sockaddr_in *)addr)->sin_port) != 0) xprt_set_bound(xprt); @@ -1484,7 +1499,7 @@ struct rpc_xprt *xs_setup_tcp(struct sockaddr *addr, size_t addrlen, struct rpc_ xprt->tsh_size = sizeof(rpc_fraghdr) / sizeof(u32); xprt->max_payload = RPC_MAX_FRAGMENT_SIZE; - INIT_WORK(&xprt->connect_worker, xs_tcp_connect_worker, xprt); + INIT_WORK(&transport->connect_worker, xs_tcp_connect_worker, transport); xprt->bind_timeout = XS_BIND_TO; xprt->connect_timeout = XS_TCP_CONN_TO; xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO; -- cgit v1.2.3 From 7c6e066ec29290bf062f5bff2984bad9be5809c7 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:30 -0500 Subject: SUNRPC: Move the UDP socket bufsize parameters to a private data structure Move the socket-specific buffer size parameters for UDP sockets to a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 3 --- net/sunrpc/xprtsock.c | 24 ++++++++++++++++-------- 2 files changed, 16 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 18bf72c06488..21beb56fb84c 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -137,9 +137,6 @@ struct rpc_xprt { unsigned long cong; /* current congestion */ unsigned long cwnd; /* congestion window */ - size_t rcvsize, /* transport rcv buffer size */ - sndsize; /* transport send buffer size */ - size_t max_payload; /* largest RPC payload size, in bytes */ unsigned int tsh_size; /* size of transport specific diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 4797a4608c07..b804381d5bee 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -151,6 +151,12 @@ struct sock_xprt { */ struct work_struct connect_worker; unsigned short port; + + /* + * UDP socket buffer size parameters + */ + size_t rcvsize, + sndsize; }; /* @@ -996,13 +1002,13 @@ static void xs_udp_do_set_buffer_size(struct rpc_xprt *xprt) struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct sock *sk = transport->inet; - if (xprt->rcvsize) { + if (transport->rcvsize) { sk->sk_userlocks |= SOCK_RCVBUF_LOCK; - sk->sk_rcvbuf = xprt->rcvsize * xprt->max_reqs * 2; + sk->sk_rcvbuf = transport->rcvsize * xprt->max_reqs * 2; } - if (xprt->sndsize) { + if (transport->sndsize) { sk->sk_userlocks |= SOCK_SNDBUF_LOCK; - sk->sk_sndbuf = xprt->sndsize * xprt->max_reqs * 2; + sk->sk_sndbuf = transport->sndsize * xprt->max_reqs * 2; sk->sk_write_space(sk); } } @@ -1017,12 +1023,14 @@ static void xs_udp_do_set_buffer_size(struct rpc_xprt *xprt) */ static void xs_udp_set_buffer_size(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize) { - xprt->sndsize = 0; + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + + transport->sndsize = 0; if (sndsize) - xprt->sndsize = sndsize + 1024; - xprt->rcvsize = 0; + transport->sndsize = sndsize + 1024; + transport->rcvsize = 0; if (rcvsize) - xprt->rcvsize = rcvsize + 1024; + transport->rcvsize = rcvsize + 1024; xs_udp_do_set_buffer_size(xprt); } -- cgit v1.2.3 From 314dfd7987c71d7ba0c43ac3bf3d243c102ce025 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:34 -0500 Subject: SUNRPC: move saved socket callback functions to a private data structure Move the three fields for saving socket callback functions out of the rpc_xprt structure and into a private data structure maintained in net/sunrpc/xprtsock.c. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 6 ------ net/sunrpc/xprtsock.c | 25 ++++++++++++++++--------- 2 files changed, 16 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 21beb56fb84c..17f322794e91 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -17,8 +17,6 @@ #include #include -#include - extern unsigned int xprt_udp_slot_table_entries; extern unsigned int xprt_tcp_slot_table_entries; @@ -191,10 +189,6 @@ struct rpc_xprt { bklog_u; /* backlog queue utilization */ } stat; - void (*old_data_ready)(struct sock *, int); - void (*old_state_change)(struct sock *); - void (*old_write_space)(struct sock *); - char * address_strings[RPC_DISPLAY_MAX]; }; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index b804381d5bee..faa6bfebf258 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -157,6 +157,13 @@ struct sock_xprt { */ size_t rcvsize, sndsize; + + /* + * Saved socket callback addresses + */ + void (*old_data_ready)(struct sock *, int); + void (*old_state_change)(struct sock *); + void (*old_write_space)(struct sock *); }; /* @@ -536,9 +543,9 @@ static void xs_close(struct rpc_xprt *xprt) transport->sock = NULL; sk->sk_user_data = NULL; - sk->sk_data_ready = xprt->old_data_ready; - sk->sk_state_change = xprt->old_state_change; - sk->sk_write_space = xprt->old_write_space; + sk->sk_data_ready = transport->old_data_ready; + sk->sk_state_change = transport->old_state_change; + sk->sk_write_space = transport->old_write_space; write_unlock_bh(&sk->sk_callback_lock); sk->sk_no_check = 0; @@ -1147,9 +1154,9 @@ static void xs_udp_connect_worker(void *args) write_lock_bh(&sk->sk_callback_lock); sk->sk_user_data = xprt; - xprt->old_data_ready = sk->sk_data_ready; - xprt->old_state_change = sk->sk_state_change; - xprt->old_write_space = sk->sk_write_space; + transport->old_data_ready = sk->sk_data_ready; + transport->old_state_change = sk->sk_state_change; + transport->old_write_space = sk->sk_write_space; sk->sk_data_ready = xs_udp_data_ready; sk->sk_write_space = xs_udp_write_space; sk->sk_no_check = UDP_CSUM_NORCV; @@ -1234,9 +1241,9 @@ static void xs_tcp_connect_worker(void *args) write_lock_bh(&sk->sk_callback_lock); sk->sk_user_data = xprt; - xprt->old_data_ready = sk->sk_data_ready; - xprt->old_state_change = sk->sk_state_change; - xprt->old_write_space = sk->sk_write_space; + transport->old_data_ready = sk->sk_data_ready; + transport->old_state_change = sk->sk_state_change; + transport->old_write_space = sk->sk_write_space; sk->sk_data_ready = xs_tcp_data_ready; sk->sk_state_change = xs_tcp_state_change; sk->sk_write_space = xs_tcp_write_space; -- cgit v1.2.3 From 7559c7a28fbcaa0bca028eeebd5f251b09befe6b Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:37 -0500 Subject: SUNRPC: Make address format buffers more generic For now we will assume that all transports will use the address format buffers in the rpc_xprt struct to store their addresses. Change rpc_peer2str() to be a generic routine to handle this, and get rid of the print_address() op in the rpc_xprt_ops vector. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xprt.h | 1 - net/sunrpc/clnt.c | 6 +++++- net/sunrpc/xprtsock.c | 23 ++++------------------- 3 files changed, 9 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 17f322794e91..f780e72fc417 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -106,7 +106,6 @@ struct rpc_rqst { struct rpc_xprt_ops { void (*set_buffer_size)(struct rpc_xprt *xprt, size_t sndsize, size_t rcvsize); - char * (*print_addr)(struct rpc_xprt *xprt, enum rpc_display_format_t format); int (*reserve_xprt)(struct rpc_task *task); void (*release_xprt)(struct rpc_xprt *xprt, struct rpc_task *task); void (*rpcbind)(struct rpc_task *task); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 7d65e19550cc..aba528b9ae76 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -594,7 +594,11 @@ EXPORT_SYMBOL_GPL(rpc_peeraddr); char *rpc_peeraddr2str(struct rpc_clnt *clnt, enum rpc_display_format_t format) { struct rpc_xprt *xprt = clnt->cl_xprt; - return xprt->ops->print_addr(xprt, format); + + if (xprt->address_strings[format] != NULL) + return xprt->address_strings[format]; + else + return "unprintable"; } EXPORT_SYMBOL_GPL(rpc_peeraddr2str); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index faa6bfebf258..a94aff42d08c 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -1060,19 +1060,6 @@ static unsigned short xs_get_random_port(void) return rand + xprt_min_resvport; } -/** - * xs_print_peer_address - format an IPv4 address for printing - * @xprt: generic transport - * @format: flags field indicating which parts of the address to render - */ -static char *xs_print_peer_address(struct rpc_xprt *xprt, enum rpc_display_format_t format) -{ - if (xprt->address_strings[format] != NULL) - return xprt->address_strings[format]; - else - return "unprintable"; -} - /** * xs_set_port - reset the port number in the remote endpoint address * @xprt: generic transport @@ -1146,7 +1133,7 @@ static void xs_udp_connect_worker(void *args) } dprintk("RPC: worker connecting xprt %p to address: %s\n", - xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); + xprt, xprt->address_strings[RPC_DISPLAY_ALL]); if (!transport->inet) { struct sock *sk = sock->sk; @@ -1233,7 +1220,7 @@ static void xs_tcp_connect_worker(void *args) xs_tcp_reuse_connection(xprt); dprintk("RPC: worker connecting xprt %p to address: %s\n", - xprt, xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); + xprt, xprt->address_strings[RPC_DISPLAY_ALL]); if (!transport->inet) { struct sock *sk = sock->sk; @@ -1380,7 +1367,6 @@ static void xs_tcp_print_stats(struct rpc_xprt *xprt, struct seq_file *seq) static struct rpc_xprt_ops xs_udp_ops = { .set_buffer_size = xs_udp_set_buffer_size, - .print_addr = xs_print_peer_address, .reserve_xprt = xprt_reserve_xprt_cong, .release_xprt = xprt_release_xprt_cong, .rpcbind = rpc_getport, @@ -1398,7 +1384,6 @@ static struct rpc_xprt_ops xs_udp_ops = { }; static struct rpc_xprt_ops xs_tcp_ops = { - .print_addr = xs_print_peer_address, .reserve_xprt = xprt_reserve_xprt, .release_xprt = xs_tcp_release_xprt, .rpcbind = rpc_getport, @@ -1485,7 +1470,7 @@ struct rpc_xprt *xs_setup_udp(struct sockaddr *addr, size_t addrlen, struct rpc_ xs_format_peer_addresses(xprt); dprintk("RPC: set up transport to address %s\n", - xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); + xprt->address_strings[RPC_DISPLAY_ALL]); return xprt; } @@ -1529,7 +1514,7 @@ struct rpc_xprt *xs_setup_tcp(struct sockaddr *addr, size_t addrlen, struct rpc_ xs_format_peer_addresses(xprt); dprintk("RPC: set up transport to address %s\n", - xs_print_peer_address(xprt, RPC_DISPLAY_ALL)); + xprt->address_strings[RPC_DISPLAY_ALL]); return xprt; } -- cgit v1.2.3 From 9d29231690925915015c21c1fff73c7118099843 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:41 -0500 Subject: SUNRPC: skb_read_bits is the same as xs_tcp_copy_data Clean-up: eliminate xs_tcp_copy_data -- it's exactly the same logic as the common routine skb_read_bits. The UDP and TCP socket read code now share the same routine for copying data into an xdr_buf. Now that skb_read_bits() is exported, rename it to avoid confusing it with a generic skb_* function. As these functions are XDR-specific, they should not have names that suggest they are of generic use. Also rename skb_read_and_csum_bits() to be consistent. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 1 + net/sunrpc/socklib.c | 14 +++++++------- net/sunrpc/xprtsock.c | 24 ++++-------------------- 3 files changed, 12 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 4d17846cd78a..a06aab278fe0 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -157,6 +157,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); +size_t xdr_skb_read_bits(skb_reader_t *desc, void *to, size_t len); extern int csum_partial_copy_to_xdr(struct xdr_buf *, struct sk_buff *); extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); diff --git a/net/sunrpc/socklib.c b/net/sunrpc/socklib.c index 2635c543ba06..9c40d67c1ffb 100644 --- a/net/sunrpc/socklib.c +++ b/net/sunrpc/socklib.c @@ -16,7 +16,7 @@ /** - * skb_read_bits - copy some data bits from skb to internal buffer + * xdr_skb_read_bits - copy some data bits from skb to internal buffer * @desc: sk_buff copy helper * @to: copy destination * @len: number of bytes to copy @@ -24,11 +24,11 @@ * Possibly called several times to iterate over an sk_buff and copy * data out of it. */ -static size_t skb_read_bits(skb_reader_t *desc, void *to, size_t len) +size_t xdr_skb_read_bits(skb_reader_t *desc, void *to, size_t len) { if (len > desc->count) len = desc->count; - if (skb_copy_bits(desc->skb, desc->offset, to, len)) + if (unlikely(skb_copy_bits(desc->skb, desc->offset, to, len))) return 0; desc->count -= len; desc->offset += len; @@ -36,14 +36,14 @@ static size_t skb_read_bits(skb_reader_t *desc, void *to, size_t len) } /** - * skb_read_and_csum_bits - copy and checksum from skb to buffer + * xdr_skb_read_and_csum_bits - copy and checksum from skb to buffer * @desc: sk_buff copy helper * @to: copy destination * @len: number of bytes to copy * * Same as skb_read_bits, but calculate a checksum at the same time. */ -static size_t skb_read_and_csum_bits(skb_reader_t *desc, void *to, size_t len) +static size_t xdr_skb_read_and_csum_bits(skb_reader_t *desc, void *to, size_t len) { unsigned int pos; __wsum csum2; @@ -158,7 +158,7 @@ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) goto no_checksum; desc.csum = csum_partial(skb->data, desc.offset, skb->csum); - if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_and_csum_bits) < 0) + if (xdr_partial_copy_from_skb(xdr, 0, &desc, xdr_skb_read_and_csum_bits) < 0) return -1; if (desc.offset != skb->len) { __wsum csum2; @@ -173,7 +173,7 @@ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) netdev_rx_csum_fault(skb->dev); return 0; no_checksum: - if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits) < 0) + if (xdr_partial_copy_from_skb(xdr, 0, &desc, xdr_skb_read_bits) < 0) return -1; if (desc.count) return -1; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index a94aff42d08c..450c350a0c62 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -651,22 +651,6 @@ static void xs_udp_data_ready(struct sock *sk, int len) read_unlock(&sk->sk_callback_lock); } -static inline size_t xs_tcp_copy_data(skb_reader_t *desc, void *p, size_t len) -{ - if (len > desc->count) - len = desc->count; - if (skb_copy_bits(desc->skb, desc->offset, p, len)) { - dprintk("RPC: failed to copy %zu bytes from skb. %zu bytes remain\n", - len, desc->count); - return 0; - } - desc->offset += len; - desc->count -= len; - dprintk("RPC: copied %zu bytes from skb. %zu bytes remain\n", - len, desc->count); - return len; -} - static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); @@ -675,7 +659,7 @@ static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc p = ((char *) &transport->tcp_fraghdr) + transport->tcp_offset; len = sizeof(transport->tcp_fraghdr) - transport->tcp_offset; - used = xs_tcp_copy_data(desc, p, len); + used = xdr_skb_read_bits(desc, p, len); transport->tcp_offset += used; if (used != len) return; @@ -721,7 +705,7 @@ static inline void xs_tcp_read_xid(struct sock_xprt *transport, skb_reader_t *de len = sizeof(transport->tcp_xid) - transport->tcp_offset; dprintk("RPC: reading XID (%Zu bytes)\n", len); p = ((char *) &transport->tcp_xid) + transport->tcp_offset; - used = xs_tcp_copy_data(desc, p, len); + used = xdr_skb_read_bits(desc, p, len); transport->tcp_offset += used; if (used != len) return; @@ -761,12 +745,12 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc memcpy(&my_desc, desc, sizeof(my_desc)); my_desc.count = len; r = xdr_partial_copy_from_skb(rcvbuf, transport->tcp_copied, - &my_desc, xs_tcp_copy_data); + &my_desc, xdr_skb_read_bits); desc->count -= r; desc->offset += r; } else r = xdr_partial_copy_from_skb(rcvbuf, transport->tcp_copied, - desc, xs_tcp_copy_data); + desc, xdr_skb_read_bits); if (r > 0) { transport->tcp_copied += r; -- cgit v1.2.3 From dd4564715eae2c4136f278da9ae1c3bb5af3e509 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:35:44 -0500 Subject: SUNRPC: Rename skb_reader_t and friends Clean-up: hch suggested that the RPC client shouldn't pollute the name space used by the generic skb manipulation routines in net/core/skbuff.c. Rename a couple of types in xdr.h to adhere to this convention. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 10 +++++----- net/sunrpc/socklib.c | 8 ++++---- net/sunrpc/xprtsock.c | 12 ++++++------ 3 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index a06aab278fe0..9e340fa23c06 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -148,19 +148,19 @@ extern int write_bytes_to_xdr_buf(struct xdr_buf *, unsigned int, void *, unsign /* * Helper structure for copying from an sk_buff. */ -typedef struct { +struct xdr_skb_reader { struct sk_buff *skb; unsigned int offset; size_t count; __wsum csum; -} skb_reader_t; +}; -typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); +typedef size_t (*xdr_skb_read_actor)(struct xdr_skb_reader *desc, void *to, size_t len); -size_t xdr_skb_read_bits(skb_reader_t *desc, void *to, size_t len); +size_t xdr_skb_read_bits(struct xdr_skb_reader *desc, void *to, size_t len); extern int csum_partial_copy_to_xdr(struct xdr_buf *, struct sk_buff *); extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, - skb_reader_t *, skb_read_actor_t); + struct xdr_skb_reader *, xdr_skb_read_actor); extern int xdr_encode_word(struct xdr_buf *, unsigned int, u32); extern int xdr_decode_word(struct xdr_buf *, unsigned int, u32 *); diff --git a/net/sunrpc/socklib.c b/net/sunrpc/socklib.c index 9c40d67c1ffb..634885b0c04d 100644 --- a/net/sunrpc/socklib.c +++ b/net/sunrpc/socklib.c @@ -24,7 +24,7 @@ * Possibly called several times to iterate over an sk_buff and copy * data out of it. */ -size_t xdr_skb_read_bits(skb_reader_t *desc, void *to, size_t len) +size_t xdr_skb_read_bits(struct xdr_skb_reader *desc, void *to, size_t len) { if (len > desc->count) len = desc->count; @@ -43,7 +43,7 @@ size_t xdr_skb_read_bits(skb_reader_t *desc, void *to, size_t len) * * Same as skb_read_bits, but calculate a checksum at the same time. */ -static size_t xdr_skb_read_and_csum_bits(skb_reader_t *desc, void *to, size_t len) +static size_t xdr_skb_read_and_csum_bits(struct xdr_skb_reader *desc, void *to, size_t len) { unsigned int pos; __wsum csum2; @@ -66,7 +66,7 @@ static size_t xdr_skb_read_and_csum_bits(skb_reader_t *desc, void *to, size_t le * @copy_actor: virtual method for copying data * */ -ssize_t xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, skb_reader_t *desc, skb_read_actor_t copy_actor) +ssize_t xdr_partial_copy_from_skb(struct xdr_buf *xdr, unsigned int base, struct xdr_skb_reader *desc, xdr_skb_read_actor copy_actor) { struct page **ppage = xdr->pages; unsigned int len, pglen = xdr->page_len; @@ -148,7 +148,7 @@ out: */ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) { - skb_reader_t desc; + struct xdr_skb_reader desc; desc.skb = skb; desc.offset = sizeof(struct udphdr); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 450c350a0c62..ad61b3e9123b 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -651,7 +651,7 @@ static void xs_udp_data_ready(struct sock *sk, int len) read_unlock(&sk->sk_callback_lock); } -static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, skb_reader_t *desc) +static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, struct xdr_skb_reader *desc) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); size_t len, used; @@ -697,7 +697,7 @@ static void xs_tcp_check_fraghdr(struct sock_xprt *transport) } } -static inline void xs_tcp_read_xid(struct sock_xprt *transport, skb_reader_t *desc) +static inline void xs_tcp_read_xid(struct sock_xprt *transport, struct xdr_skb_reader *desc) { size_t len, used; char *p; @@ -717,7 +717,7 @@ static inline void xs_tcp_read_xid(struct sock_xprt *transport, skb_reader_t *de xs_tcp_check_fraghdr(transport); } -static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc) +static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_reader *desc) { struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); struct rpc_rqst *req; @@ -739,7 +739,7 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, skb_reader_t *desc rcvbuf = &req->rq_private_buf; len = desc->count; if (len > transport->tcp_reclen - transport->tcp_offset) { - skb_reader_t my_desc; + struct xdr_skb_reader my_desc; len = transport->tcp_reclen - transport->tcp_offset; memcpy(&my_desc, desc, sizeof(my_desc)); @@ -795,7 +795,7 @@ out: xs_tcp_check_fraghdr(transport); } -static inline void xs_tcp_read_discard(struct sock_xprt *transport, skb_reader_t *desc) +static inline void xs_tcp_read_discard(struct sock_xprt *transport, struct xdr_skb_reader *desc) { size_t len; @@ -813,7 +813,7 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns { struct rpc_xprt *xprt = rd_desc->arg.data; struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); - skb_reader_t desc = { + struct xdr_skb_reader desc = { .skb = skb, .offset = offset, .count = len, -- cgit v1.2.3 From 5847e1f4d058677c5e46dc6c3e3c70e8855ea3ba Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 5 Dec 2006 16:36:14 -0500 Subject: SUNRPC: Remove pprintk() from net/sunrpc/xprt.c These appear to be deprecated. Removing them also gets rid of some sparse noise. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- include/linux/sunrpc/debug.h | 6 ------ net/sunrpc/xprt.c | 2 -- 2 files changed, 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/debug.h b/include/linux/sunrpc/debug.h index e4729aa67654..60fce3c92857 100644 --- a/include/linux/sunrpc/debug.h +++ b/include/linux/sunrpc/debug.h @@ -62,12 +62,6 @@ extern unsigned int nlm_debug; # define RPC_IFDEBUG(x) #endif -#ifdef RPC_PROFILE -# define pprintk(args...) printk(## args) -#else -# define pprintk(args...) do ; while (0) -#endif - /* * Sysctl interface for RPC debugging */ diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 8cc2afa2942c..f8ca0a93454c 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -459,7 +459,6 @@ int xprt_adjust_timeout(struct rpc_rqst *req) if (to->to_maxval && req->rq_timeout >= to->to_maxval) req->rq_timeout = to->to_maxval; req->rq_retries++; - pprintk("RPC: %lu retrans\n", jiffies); } else { req->rq_timeout = to->to_initval; req->rq_retries = 0; @@ -468,7 +467,6 @@ int xprt_adjust_timeout(struct rpc_rqst *req) spin_lock_bh(&xprt->transport_lock); rpc_init_rtt(req->rq_task->tk_client->cl_rtt, to->to_initval); spin_unlock_bh(&xprt->transport_lock); - pprintk("RPC: %lu timeout\n", jiffies); status = -ETIMEDOUT; } -- cgit v1.2.3 From bb81a09e55eaf7e5f798468ab971469b6f66a259 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Thu, 7 Dec 2006 02:14:01 +0100 Subject: [PATCH] x86: all cpu backtrace When a spinlock lockup occurs, arrange for the NMI code to emit an all-cpu backtrace, so we get to see which CPU is holding the lock, and where. Cc: Andi Kleen Cc: Ingo Molnar Cc: Badari Pulavarty Signed-off-by: Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 26 ++++++++++++++++++++++++++ arch/x86_64/kernel/nmi.c | 29 ++++++++++++++++++++++++++++- include/asm-i386/nmi.h | 8 ++++++++ include/asm-x86_64/nmi.h | 3 +++ include/linux/nmi.h | 5 +++++ lib/spinlock_debug.c | 4 ++++ 6 files changed, 74 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index eaafe233a5da..171194ccb7bc 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -42,6 +43,8 @@ int nmi_watchdog_enabled; static DEFINE_PER_CPU(unsigned long, perfctr_nmi_owner); static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]); +static cpumask_t backtrace_mask = CPU_MASK_NONE; + /* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. It will be the max for all platforms (for now) */ @@ -907,6 +910,16 @@ __kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) touched = 1; } + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk("NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + sum = per_cpu(irq_stat, cpu).apic_timer_irqs; /* if the apic timer isn't firing, this cpu isn't doing much */ @@ -1033,6 +1046,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, #endif +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} + EXPORT_SYMBOL(nmi_active); EXPORT_SYMBOL(nmi_watchdog); EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi); diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 7af9cb3e2d99..27e95e7922c1 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -12,14 +12,15 @@ * Mikael Pettersson : PM converted to driver model. Disable/enable API. */ +#include #include #include #include #include #include -#include #include #include +#include #include #include @@ -41,6 +42,8 @@ int panic_on_unrecovered_nmi; static DEFINE_PER_CPU(unsigned, perfctr_nmi_owner); static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]); +static cpumask_t backtrace_mask = CPU_MASK_NONE; + /* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. It will be the max for all platforms (for now) */ @@ -782,6 +785,7 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) { int sum; int touched = 0; + int cpu = smp_processor_id(); struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); u64 dummy; int rc=0; @@ -799,6 +803,16 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason) touched = 1; } + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk("NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + #ifdef CONFIG_X86_MCE /* Could check oops_in_progress here too, but it's safer not too */ @@ -931,6 +945,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, #endif +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} + EXPORT_SYMBOL(nmi_active); EXPORT_SYMBOL(nmi_watchdog); EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi); diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h index 269d315719ca..b04333ea6f31 100644 --- a/include/asm-i386/nmi.h +++ b/include/asm-i386/nmi.h @@ -5,6 +5,9 @@ #define ASM_NMI_H #include +#include + +#ifdef ARCH_HAS_NMI_WATCHDOG /** * do_nmi_callback @@ -42,4 +45,9 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, void __user *, size_t *, loff_t *); extern int unknown_nmi_panic; +void __trigger_all_cpu_backtrace(void); +#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() + +#endif + #endif /* ASM_NMI_H */ diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h index f367d4014b42..72375e7d32a8 100644 --- a/include/asm-x86_64/nmi.h +++ b/include/asm-x86_64/nmi.h @@ -77,4 +77,7 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, extern int unknown_nmi_panic; +void __trigger_all_cpu_backtrace(void); +#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace() + #endif /* ASM_NMI_H */ diff --git a/include/linux/nmi.h b/include/linux/nmi.h index e16904e28c3a..acb4ed130247 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -15,9 +15,14 @@ * disables interrupts for a long time. This call is stateless. */ #ifdef ARCH_HAS_NMI_WATCHDOG +#include extern void touch_nmi_watchdog(void); #else # define touch_nmi_watchdog() touch_softlockup_watchdog() #endif +#ifndef trigger_all_cpu_backtrace +#define trigger_all_cpu_backtrace() do { } while (0) +#endif + #endif diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index b6c4f898197c..479fd462eaa9 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -7,6 +7,7 @@ */ #include +#include #include #include #include @@ -117,6 +118,9 @@ static void __spin_lock_debug(spinlock_t *lock) raw_smp_processor_id(), current->comm, current->pid, lock); dump_stack(); +#ifdef CONFIG_SMP + trigger_all_cpu_backtrace(); +#endif } } } -- cgit v1.2.3 From 968de4f02621db35b8ae5239c8cfc6664fb872d8 Mon Sep 17 00:00:00 2001 From: Eric W. Biederman Date: Thu, 7 Dec 2006 02:14:04 +0100 Subject: [PATCH] i386: Relocatable kernel support This patch modifies the i386 kernel so that if CONFIG_RELOCATABLE is selected it will be able to be loaded at any 4K aligned address below 1G. The technique used is to compile the decompressor with -fPIC and modify it so the decompressor is fully relocatable. For the main kernel relocations are generated. Resulting in a kernel that is relocatable with no runtime overhead and no need to modify the source code. A reserved 32bit word in the parameters has been assigned to serve as a stack so we figure out where are running. Signed-off-by: Eric W. Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andi Kleen --- arch/i386/Kconfig | 12 + arch/i386/Makefile | 4 +- arch/i386/boot/compressed/Makefile | 28 +- arch/i386/boot/compressed/head.S | 184 +++++++---- arch/i386/boot/compressed/misc.c | 261 ++++++++-------- arch/i386/boot/compressed/relocs.c | 563 ++++++++++++++++++++++++++++++++++ arch/i386/boot/compressed/vmlinux.lds | 43 +++ arch/i386/boot/compressed/vmlinux.scr | 3 +- arch/i386/boot/setup.S | 29 +- include/linux/screen_info.h | 3 +- 10 files changed, 920 insertions(+), 210 deletions(-) create mode 100644 arch/i386/boot/compressed/relocs.c create mode 100644 arch/i386/boot/compressed/vmlinux.lds (limited to 'include/linux') diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 8ff1c6fb5aa1..d588ca874bb4 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -773,6 +773,18 @@ config CRASH_DUMP PHYSICAL_START. For more details see Documentation/kdump/kdump.txt +config RELOCATABLE + bool "Build a relocatable kernel" + help + This build a kernel image that retains relocation information + so it can be loaded someplace besides the default 1MB. + The relocations tend to the kernel binary about 10% larger, + but are discarded at runtime. + + One use is for the kexec on panic case where the recovery kernel + must live at a different physical address than the primary + kernel. + config PHYSICAL_START hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 0677908dfa06..d1aca52bf690 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -26,7 +26,9 @@ endif LDFLAGS := -m elf_i386 OBJCOPYFLAGS := -O binary -R .note -R .comment -S -LDFLAGS_vmlinux := +ifdef CONFIG_RELOCATABLE +LDFLAGS_vmlinux := --emit-relocs +endif CHECKFLAGS += -D__i386__ CFLAGS += -pipe -msoft-float diff --git a/arch/i386/boot/compressed/Makefile b/arch/i386/boot/compressed/Makefile index 258ea95224f6..cc28da3a881e 100644 --- a/arch/i386/boot/compressed/Makefile +++ b/arch/i386/boot/compressed/Makefile @@ -4,22 +4,42 @@ # create a compressed vmlinux image from the original vmlinux # -targets := vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o +targets := vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o \ + vmlinux.bin.all vmlinux.relocs EXTRA_AFLAGS := -traditional -LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup_32 +LDFLAGS_vmlinux := -T +CFLAGS_misc.o += -fPIC +hostprogs-y := relocs -$(obj)/vmlinux: $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE +$(obj)/vmlinux: $(src)/vmlinux.lds $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o FORCE $(call if_changed,ld) @: $(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) +quiet_cmd_relocs = RELOCS $@ + cmd_relocs = $(obj)/relocs $< > $@ +$(obj)/vmlinux.relocs: vmlinux $(obj)/relocs FORCE + $(call if_changed,relocs) + +vmlinux.bin.all-y := $(obj)/vmlinux.bin +vmlinux.bin.all-$(CONFIG_RELOCATABLE) += $(obj)/vmlinux.relocs +quiet_cmd_relocbin = BUILD $@ + cmd_relocbin = cat $(filter-out FORCE,$^) > $@ +$(obj)/vmlinux.bin.all: $(vmlinux.bin.all-y) FORCE + $(call if_changed,relocbin) + +ifdef CONFIG_RELOCATABLE +$(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE + $(call if_changed,gzip) +else $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE $(call if_changed,gzip) +endif LDFLAGS_piggy.o := -r --format binary --oformat elf32-i386 -T -$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE +$(obj)/piggy.o: $(src)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE $(call if_changed,ld) diff --git a/arch/i386/boot/compressed/head.S b/arch/i386/boot/compressed/head.S index 40a8de8270a9..e4dd7a6b9b0f 100644 --- a/arch/i386/boot/compressed/head.S +++ b/arch/i386/boot/compressed/head.S @@ -25,9 +25,11 @@ #include #include +#include +.section ".text.head" .globl startup_32 - + startup_32: cld cli @@ -36,93 +38,141 @@ startup_32: movl %eax,%es movl %eax,%fs movl %eax,%gs + movl %eax,%ss - lss stack_start,%esp - xorl %eax,%eax -1: incl %eax # check that A20 really IS enabled - movl %eax,0x000000 # loop forever if it isn't - cmpl %eax,0x100000 - je 1b +/* Calculate the delta between where we were compiled to run + * at and where we were actually loaded at. This can only be done + * with a short local call on x86. Nothing else will tell us what + * address we are running at. The reserved chunk of the real-mode + * data at 0x34-0x3f are used as the stack for this calculation. + * Only 4 bytes are needed. + */ + leal 0x40(%esi), %esp + call 1f +1: popl %ebp + subl $1b, %ebp + +/* Compute the delta between where we were compiled to run at + * and where the code will actually run at. + */ + /* Start with the delta to where the kernel will run at. If we are + * a relocatable kernel this is the delta to our load address otherwise + * this is the delta to CONFIG_PHYSICAL start. + */ +#ifdef CONFIG_RELOCATABLE + movl %ebp, %ebx +#else + movl $(CONFIG_PHYSICAL_START - startup_32), %ebx +#endif + + /* Replace the compressed data size with the uncompressed size */ + subl input_len(%ebp), %ebx + movl output_len(%ebp), %eax + addl %eax, %ebx + /* Add 8 bytes for every 32K input block */ + shrl $12, %eax + addl %eax, %ebx + /* Add 32K + 18 bytes of extra slack */ + addl $(32768 + 18), %ebx + /* Align on a 4K boundary */ + addl $4095, %ebx + andl $~4095, %ebx + +/* Copy the compressed kernel to the end of our buffer + * where decompression in place becomes safe. + */ + pushl %esi + leal _end(%ebp), %esi + leal _end(%ebx), %edi + movl $(_end - startup_32), %ecx + std + rep + movsb + cld + popl %esi + +/* Compute the kernel start address. + */ +#ifdef CONFIG_RELOCATABLE + leal startup_32(%ebp), %ebp +#else + movl $CONFIG_PHYSICAL_START, %ebp +#endif /* - * Initialize eflags. Some BIOS's leave bits like NT set. This would - * confuse the debugger if this code is traced. - * XXX - best to initialize before switching to protected mode. + * Jump to the relocated address. */ - pushl $0 - popfl + leal relocated(%ebx), %eax + jmp *%eax +.section ".text" +relocated: + /* * Clear BSS */ xorl %eax,%eax - movl $_edata,%edi - movl $_end,%ecx + leal _edata(%ebx),%edi + leal _end(%ebx), %ecx subl %edi,%ecx cld rep stosb + +/* + * Setup the stack for the decompressor + */ + leal stack_end(%ebx), %esp + /* * Do the decompression, and jump to the new kernel.. */ - subl $16,%esp # place for structure on the stack - movl %esp,%eax + movl output_len(%ebx), %eax + pushl %eax + pushl %ebp # output address + movl input_len(%ebx), %eax + pushl %eax # input_len + leal input_data(%ebx), %eax + pushl %eax # input_data + leal _end(%ebx), %eax + pushl %eax # end of the image as third argument pushl %esi # real mode pointer as second arg - pushl %eax # address of structure as first arg call decompress_kernel - orl %eax,%eax - jnz 3f - popl %esi # discard address - popl %esi # real mode pointer - xorl %ebx,%ebx - ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START + addl $20, %esp + popl %ecx + +#if CONFIG_RELOCATABLE +/* Find the address of the relocations. + */ + movl %ebp, %edi + addl %ecx, %edi + +/* Calculate the delta between where vmlinux was compiled to run + * and where it was actually loaded. + */ + movl %ebp, %ebx + subl $CONFIG_PHYSICAL_START, %ebx /* - * We come here, if we were loaded high. - * We need to move the move-in-place routine down to 0x1000 - * and then start it with the buffer addresses in registers, - * which we got from the stack. + * Process relocations. */ -3: - movl $move_routine_start,%esi - movl $0x1000,%edi - movl $move_routine_end,%ecx - subl %esi,%ecx - addl $3,%ecx - shrl $2,%ecx - cld - rep - movsl - - popl %esi # discard the address - popl %ebx # real mode pointer - popl %esi # low_buffer_start - popl %ecx # lcount - popl %edx # high_buffer_start - popl %eax # hcount - movl $CONFIG_PHYSICAL_START,%edi - cli # make sure we don't get interrupted - ljmp $(__BOOT_CS), $0x1000 # and jump to the move routine + +1: subl $4, %edi + movl 0(%edi), %ecx + testl %ecx, %ecx + jz 2f + addl %ebx, -__PAGE_OFFSET(%ebx, %ecx) + jmp 1b +2: +#endif /* - * Routine (template) for moving the decompressed kernel in place, - * if we were high loaded. This _must_ PIC-code ! + * Jump to the decompressed kernel. */ -move_routine_start: - movl %ecx,%ebp - shrl $2,%ecx - rep - movsl - movl %ebp,%ecx - andl $3,%ecx - rep - movsb - movl %edx,%esi - movl %eax,%ecx # NOTE: rep movsb won't move if %ecx == 0 - addl $3,%ecx - shrl $2,%ecx - rep - movsl - movl %ebx,%esi # Restore setup pointer xorl %ebx,%ebx - ljmp $(__BOOT_CS), $CONFIG_PHYSICAL_START -move_routine_end: + jmp *%ebp + +.bss +.balign 4 +stack: + .fill 4096, 1, 0 +stack_end: diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index 20970ff44119..4eac24e95a10 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -13,6 +13,88 @@ #include #include #include +#include + +/* WARNING!! + * This code is compiled with -fPIC and it is relocated dynamically + * at run time, but no relocation processing is performed. + * This means that it is not safe to place pointers in static structures. + */ + +/* + * Getting to provable safe in place decompression is hard. + * Worst case behaviours need to be analized. + * Background information: + * + * The file layout is: + * magic[2] + * method[1] + * flags[1] + * timestamp[4] + * extraflags[1] + * os[1] + * compressed data blocks[N] + * crc[4] orig_len[4] + * + * resulting in 18 bytes of non compressed data overhead. + * + * Files divided into blocks + * 1 bit (last block flag) + * 2 bits (block type) + * + * 1 block occurs every 32K -1 bytes or when there 50% compression has been achieved. + * The smallest block type encoding is always used. + * + * stored: + * 32 bits length in bytes. + * + * fixed: + * magic fixed tree. + * symbols. + * + * dynamic: + * dynamic tree encoding. + * symbols. + * + * + * The buffer for decompression in place is the length of the + * uncompressed data, plus a small amount extra to keep the algorithm safe. + * The compressed data is placed at the end of the buffer. The output + * pointer is placed at the start of the buffer and the input pointer + * is placed where the compressed data starts. Problems will occur + * when the output pointer overruns the input pointer. + * + * The output pointer can only overrun the input pointer if the input + * pointer is moving faster than the output pointer. A condition only + * triggered by data whose compressed form is larger than the uncompressed + * form. + * + * The worst case at the block level is a growth of the compressed data + * of 5 bytes per 32767 bytes. + * + * The worst case internal to a compressed block is very hard to figure. + * The worst case can at least be boundined by having one bit that represents + * 32764 bytes and then all of the rest of the bytes representing the very + * very last byte. + * + * All of which is enough to compute an amount of extra data that is required + * to be safe. To avoid problems at the block level allocating 5 extra bytes + * per 32767 bytes of data is sufficient. To avoind problems internal to a block + * adding an extra 32767 bytes (the worst case uncompressed block size) is + * sufficient, to ensure that in the worst case the decompressed data for + * block will stop the byte before the compressed data for a block begins. + * To avoid problems with the compressed data's meta information an extra 18 + * bytes are needed. Leading to the formula: + * + * extra_bytes = (uncompressed_size >> 12) + 32768 + 18 + decompressor_size. + * + * Adding 8 bytes per 32K is a bit excessive but much easier to calculate. + * Adding 32768 instead of 32767 just makes for round numbers. + * Adding the decompressor_size is necessary as it musht live after all + * of the data as well. Last I measured the decompressor is about 14K. + * 10K of actuall data and 4K of bss. + * + */ /* * gzip declarations @@ -29,15 +111,20 @@ typedef unsigned char uch; typedef unsigned short ush; typedef unsigned long ulg; -#define WSIZE 0x8000 /* Window size must be at least 32k, */ - /* and a power of two */ +#define WSIZE 0x80000000 /* Window size must be at least 32k, + * and a power of two + * We don't actually have a window just + * a huge output buffer so I report + * a 2G windows size, as that should + * always be larger than our output buffer. + */ -static uch *inbuf; /* input buffer */ -static uch window[WSIZE]; /* Sliding window buffer */ +static uch *inbuf; /* input buffer */ +static uch *window; /* Sliding window buffer, (and final output buffer) */ -static unsigned insize = 0; /* valid bytes in inbuf */ -static unsigned inptr = 0; /* index of next byte to be processed in inbuf */ -static unsigned outcnt = 0; /* bytes in output buffer */ +static unsigned insize; /* valid bytes in inbuf */ +static unsigned inptr; /* index of next byte to be processed in inbuf */ +static unsigned outcnt; /* bytes in output buffer */ /* gzip flag byte */ #define ASCII_FLAG 0x01 /* bit 0 set: file probably ASCII text */ @@ -88,8 +175,6 @@ extern unsigned char input_data[]; extern int input_len; static long bytes_out = 0; -static uch *output_data; -static unsigned long output_ptr = 0; static void *malloc(int size); static void free(void *where); @@ -99,17 +184,10 @@ static void *memcpy(void *dest, const void *src, unsigned n); static void putstr(const char *); -extern int end; -static long free_mem_ptr = (long)&end; -static long free_mem_end_ptr; +static unsigned long free_mem_ptr; +static unsigned long free_mem_end_ptr; -#define INPLACE_MOVE_ROUTINE 0x1000 -#define LOW_BUFFER_START 0x2000 -#define LOW_BUFFER_MAX 0x90000 #define HEAP_SIZE 0x3000 -static unsigned int low_buffer_end, low_buffer_size; -static int high_loaded =0; -static uch *high_buffer_start /* = (uch *)(((ulg)&end) + HEAP_SIZE)*/; static char *vidmem = (char *)0xb8000; static int vidport; @@ -150,7 +228,7 @@ static void gzip_mark(void **ptr) static void gzip_release(void **ptr) { - free_mem_ptr = (long) *ptr; + free_mem_ptr = (unsigned long) *ptr; } static void scroll(void) @@ -178,7 +256,7 @@ static void putstr(const char *s) y--; } } else { - vidmem [ ( x + cols * y ) * 2 ] = c; + vidmem [ ( x + cols * y ) * 2 ] = c; if ( ++x >= cols ) { x = 0; if ( ++y >= lines ) { @@ -223,58 +301,31 @@ static void* memcpy(void* dest, const void* src, unsigned n) */ static int fill_inbuf(void) { - if (insize != 0) { - error("ran out of input data"); - } - - inbuf = input_data; - insize = input_len; - inptr = 1; - return inbuf[0]; + error("ran out of input data"); + return 0; } /* =========================================================================== * Write the output window window[0..outcnt-1] and update crc and bytes_out. * (Used for the decompressed data only.) */ -static void flush_window_low(void) -{ - ulg c = crc; /* temporary variable */ - unsigned n; - uch *in, *out, ch; - - in = window; - out = &output_data[output_ptr]; - for (n = 0; n < outcnt; n++) { - ch = *out++ = *in++; - c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); - } - crc = c; - bytes_out += (ulg)outcnt; - output_ptr += (ulg)outcnt; - outcnt = 0; -} - -static void flush_window_high(void) -{ - ulg c = crc; /* temporary variable */ - unsigned n; - uch *in, ch; - in = window; - for (n = 0; n < outcnt; n++) { - ch = *output_data++ = *in++; - if ((ulg)output_data == low_buffer_end) output_data=high_buffer_start; - c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); - } - crc = c; - bytes_out += (ulg)outcnt; - outcnt = 0; -} - static void flush_window(void) { - if (high_loaded) flush_window_high(); - else flush_window_low(); + /* With my window equal to my output buffer + * I only need to compute the crc here. + */ + ulg c = crc; /* temporary variable */ + unsigned n; + uch *in, ch; + + in = window; + for (n = 0; n < outcnt; n++) { + ch = *in++; + c = crc_32_tab[((int)c ^ ch) & 0xff] ^ (c >> 8); + } + crc = c; + bytes_out += (ulg)outcnt; + outcnt = 0; } static void error(char *x) @@ -286,66 +337,8 @@ static void error(char *x) while(1); /* Halt */ } -#define STACK_SIZE (4096) - -long user_stack [STACK_SIZE]; - -struct { - long * a; - short b; - } stack_start = { & user_stack [STACK_SIZE] , __BOOT_DS }; - -static void setup_normal_output_buffer(void) -{ -#ifdef STANDARD_MEMORY_BIOS_CALL - if (RM_EXT_MEM_K < 1024) error("Less than 2MB of memory"); -#else - if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < 1024) error("Less than 2MB of memory"); -#endif - output_data = (unsigned char *)CONFIG_PHYSICAL_START; /* Normally Points to 1M */ - free_mem_end_ptr = (long)real_mode; -} - -struct moveparams { - uch *low_buffer_start; int lcount; - uch *high_buffer_start; int hcount; -}; - -static void setup_output_buffer_if_we_run_high(struct moveparams *mv) -{ - high_buffer_start = (uch *)(((ulg)&end) + HEAP_SIZE); -#ifdef STANDARD_MEMORY_BIOS_CALL - if (RM_EXT_MEM_K < (3*1024)) error("Less than 4MB of memory"); -#else - if ((RM_ALT_MEM_K > RM_EXT_MEM_K ? RM_ALT_MEM_K : RM_EXT_MEM_K) < (3*1024)) error("Less than 4MB of memory"); -#endif - mv->low_buffer_start = output_data = (unsigned char *)LOW_BUFFER_START; - low_buffer_end = ((unsigned int)real_mode > LOW_BUFFER_MAX - ? LOW_BUFFER_MAX : (unsigned int)real_mode) & ~0xfff; - low_buffer_size = low_buffer_end - LOW_BUFFER_START; - high_loaded = 1; - free_mem_end_ptr = (long)high_buffer_start; - if ( (CONFIG_PHYSICAL_START + low_buffer_size) > ((ulg)high_buffer_start)) { - high_buffer_start = (uch *)(CONFIG_PHYSICAL_START + low_buffer_size); - mv->hcount = 0; /* say: we need not to move high_buffer */ - } - else mv->hcount = -1; - mv->high_buffer_start = high_buffer_start; -} - -static void close_output_buffer_if_we_run_high(struct moveparams *mv) -{ - if (bytes_out > low_buffer_size) { - mv->lcount = low_buffer_size; - if (mv->hcount) - mv->hcount = bytes_out - low_buffer_size; - } else { - mv->lcount = bytes_out; - mv->hcount = 0; - } -} - -asmlinkage int decompress_kernel(struct moveparams *mv, void *rmode) +asmlinkage void decompress_kernel(void *rmode, unsigned long end, + uch *input_data, unsigned long input_len, uch *output) { real_mode = rmode; @@ -360,13 +353,25 @@ asmlinkage int decompress_kernel(struct moveparams *mv, void *rmode) lines = RM_SCREEN_INFO.orig_video_lines; cols = RM_SCREEN_INFO.orig_video_cols; - if (free_mem_ptr < 0x100000) setup_normal_output_buffer(); - else setup_output_buffer_if_we_run_high(mv); + window = output; /* Output buffer (Normally at 1M) */ + free_mem_ptr = end; /* Heap */ + free_mem_end_ptr = end + HEAP_SIZE; + inbuf = input_data; /* Input buffer */ + insize = input_len; + inptr = 0; + + if (((u32)output - CONFIG_PHYSICAL_START) & 0x3fffff) + error("Destination address not 4M aligned"); + if (end > ((-__PAGE_OFFSET-(512 <<20)-1) & 0x7fffffff)) + error("Destination address too large"); +#ifndef CONFIG_RELOCATABLE + if ((u32)output != CONFIG_PHYSICAL_START) + error("Wrong destination address"); +#endif makecrc(); putstr("Uncompressing Linux... "); gunzip(); putstr("Ok, booting the kernel.\n"); - if (high_loaded) close_output_buffer_if_we_run_high(mv); - return high_loaded; + return; } diff --git a/arch/i386/boot/compressed/relocs.c b/arch/i386/boot/compressed/relocs.c new file mode 100644 index 000000000000..0551ceb21bed --- /dev/null +++ b/arch/i386/boot/compressed/relocs.c @@ -0,0 +1,563 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#define USE_BSD +#include + +#define MAX_SHDRS 100 +static Elf32_Ehdr ehdr; +static Elf32_Shdr shdr[MAX_SHDRS]; +static Elf32_Sym *symtab[MAX_SHDRS]; +static Elf32_Rel *reltab[MAX_SHDRS]; +static char *strtab[MAX_SHDRS]; +static unsigned long reloc_count, reloc_idx; +static unsigned long *relocs; + +static void die(char *fmt, ...) +{ + va_list ap; + va_start(ap, fmt); + vfprintf(stderr, fmt, ap); + va_end(ap); + exit(1); +} + +static const char *sym_type(unsigned type) +{ + static const char *type_name[] = { +#define SYM_TYPE(X) [X] = #X + SYM_TYPE(STT_NOTYPE), + SYM_TYPE(STT_OBJECT), + SYM_TYPE(STT_FUNC), + SYM_TYPE(STT_SECTION), + SYM_TYPE(STT_FILE), + SYM_TYPE(STT_COMMON), + SYM_TYPE(STT_TLS), +#undef SYM_TYPE + }; + const char *name = "unknown sym type name"; + if (type < sizeof(type_name)/sizeof(type_name[0])) { + name = type_name[type]; + } + return name; +} + +static const char *sym_bind(unsigned bind) +{ + static const char *bind_name[] = { +#define SYM_BIND(X) [X] = #X + SYM_BIND(STB_LOCAL), + SYM_BIND(STB_GLOBAL), + SYM_BIND(STB_WEAK), +#undef SYM_BIND + }; + const char *name = "unknown sym bind name"; + if (bind < sizeof(bind_name)/sizeof(bind_name[0])) { + name = bind_name[bind]; + } + return name; +} + +static const char *sym_visibility(unsigned visibility) +{ + static const char *visibility_name[] = { +#define SYM_VISIBILITY(X) [X] = #X + SYM_VISIBILITY(STV_DEFAULT), + SYM_VISIBILITY(STV_INTERNAL), + SYM_VISIBILITY(STV_HIDDEN), + SYM_VISIBILITY(STV_PROTECTED), +#undef SYM_VISIBILITY + }; + const char *name = "unknown sym visibility name"; + if (visibility < sizeof(visibility_name)/sizeof(visibility_name[0])) { + name = visibility_name[visibility]; + } + return name; +} + +static const char *rel_type(unsigned type) +{ + static const char *type_name[] = { +#define REL_TYPE(X) [X] = #X + REL_TYPE(R_386_NONE), + REL_TYPE(R_386_32), + REL_TYPE(R_386_PC32), + REL_TYPE(R_386_GOT32), + REL_TYPE(R_386_PLT32), + REL_TYPE(R_386_COPY), + REL_TYPE(R_386_GLOB_DAT), + REL_TYPE(R_386_JMP_SLOT), + REL_TYPE(R_386_RELATIVE), + REL_TYPE(R_386_GOTOFF), + REL_TYPE(R_386_GOTPC), +#undef REL_TYPE + }; + const char *name = "unknown type rel type name"; + if (type < sizeof(type_name)/sizeof(type_name[0])) { + name = type_name[type]; + } + return name; +} + +static const char *sec_name(unsigned shndx) +{ + const char *sec_strtab; + const char *name; + sec_strtab = strtab[ehdr.e_shstrndx]; + name = ""; + if (shndx < ehdr.e_shnum) { + name = sec_strtab + shdr[shndx].sh_name; + } + else if (shndx == SHN_ABS) { + name = "ABSOLUTE"; + } + else if (shndx == SHN_COMMON) { + name = "COMMON"; + } + return name; +} + +static const char *sym_name(const char *sym_strtab, Elf32_Sym *sym) +{ + const char *name; + name = ""; + if (sym->st_name) { + name = sym_strtab + sym->st_name; + } + else { + name = sec_name(shdr[sym->st_shndx].sh_name); + } + return name; +} + + + +#if BYTE_ORDER == LITTLE_ENDIAN +#define le16_to_cpu(val) (val) +#define le32_to_cpu(val) (val) +#endif +#if BYTE_ORDER == BIG_ENDIAN +#define le16_to_cpu(val) bswap_16(val) +#define le32_to_cpu(val) bswap_32(val) +#endif + +static uint16_t elf16_to_cpu(uint16_t val) +{ + return le16_to_cpu(val); +} + +static uint32_t elf32_to_cpu(uint32_t val) +{ + return le32_to_cpu(val); +} + +static void read_ehdr(FILE *fp) +{ + if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1) { + die("Cannot read ELF header: %s\n", + strerror(errno)); + } + if (memcmp(ehdr.e_ident, ELFMAG, 4) != 0) { + die("No ELF magic\n"); + } + if (ehdr.e_ident[EI_CLASS] != ELFCLASS32) { + die("Not a 32 bit executable\n"); + } + if (ehdr.e_ident[EI_DATA] != ELFDATA2LSB) { + die("Not a LSB ELF executable\n"); + } + if (ehdr.e_ident[EI_VERSION] != EV_CURRENT) { + die("Unknown ELF version\n"); + } + /* Convert the fields to native endian */ + ehdr.e_type = elf16_to_cpu(ehdr.e_type); + ehdr.e_machine = elf16_to_cpu(ehdr.e_machine); + ehdr.e_version = elf32_to_cpu(ehdr.e_version); + ehdr.e_entry = elf32_to_cpu(ehdr.e_entry); + ehdr.e_phoff = elf32_to_cpu(ehdr.e_phoff); + ehdr.e_shoff = elf32_to_cpu(ehdr.e_shoff); + ehdr.e_flags = elf32_to_cpu(ehdr.e_flags); + ehdr.e_ehsize = elf16_to_cpu(ehdr.e_ehsize); + ehdr.e_phentsize = elf16_to_cpu(ehdr.e_phentsize); + ehdr.e_phnum = elf16_to_cpu(ehdr.e_phnum); + ehdr.e_shentsize = elf16_to_cpu(ehdr.e_shentsize); + ehdr.e_shnum = elf16_to_cpu(ehdr.e_shnum); + ehdr.e_shstrndx = elf16_to_cpu(ehdr.e_shstrndx); + + if ((ehdr.e_type != ET_EXEC) && (ehdr.e_type != ET_DYN)) { + die("Unsupported ELF header type\n"); + } + if (ehdr.e_machine != EM_386) { + die("Not for x86\n"); + } + if (ehdr.e_version != EV_CURRENT) { + die("Unknown ELF version\n"); + } + if (ehdr.e_ehsize != sizeof(Elf32_Ehdr)) { + die("Bad Elf header size\n"); + } + if (ehdr.e_phentsize != sizeof(Elf32_Phdr)) { + die("Bad program header entry\n"); + } + if (ehdr.e_shentsize != sizeof(Elf32_Shdr)) { + die("Bad section header entry\n"); + } + if (ehdr.e_shstrndx >= ehdr.e_shnum) { + die("String table index out of bounds\n"); + } +} + +static void read_shdrs(FILE *fp) +{ + int i; + if (ehdr.e_shnum > MAX_SHDRS) { + die("%d section headers supported: %d\n", + ehdr.e_shnum, MAX_SHDRS); + } + if (fseek(fp, ehdr.e_shoff, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + ehdr.e_shoff, strerror(errno)); + } + if (fread(&shdr, sizeof(shdr[0]), ehdr.e_shnum, fp) != ehdr.e_shnum) { + die("Cannot read ELF section headers: %s\n", + strerror(errno)); + } + for(i = 0; i < ehdr.e_shnum; i++) { + shdr[i].sh_name = elf32_to_cpu(shdr[i].sh_name); + shdr[i].sh_type = elf32_to_cpu(shdr[i].sh_type); + shdr[i].sh_flags = elf32_to_cpu(shdr[i].sh_flags); + shdr[i].sh_addr = elf32_to_cpu(shdr[i].sh_addr); + shdr[i].sh_offset = elf32_to_cpu(shdr[i].sh_offset); + shdr[i].sh_size = elf32_to_cpu(shdr[i].sh_size); + shdr[i].sh_link = elf32_to_cpu(shdr[i].sh_link); + shdr[i].sh_info = elf32_to_cpu(shdr[i].sh_info); + shdr[i].sh_addralign = elf32_to_cpu(shdr[i].sh_addralign); + shdr[i].sh_entsize = elf32_to_cpu(shdr[i].sh_entsize); + } + +} + +static void read_strtabs(FILE *fp) +{ + int i; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_STRTAB) { + continue; + } + strtab[i] = malloc(shdr[i].sh_size); + if (!strtab[i]) { + die("malloc of %d bytes for strtab failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(strtab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + } +} + +static void read_symtabs(FILE *fp) +{ + int i,j; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_SYMTAB) { + continue; + } + symtab[i] = malloc(shdr[i].sh_size); + if (!symtab[i]) { + die("malloc of %d bytes for symtab failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(symtab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + for(j = 0; j < shdr[i].sh_size/sizeof(symtab[i][0]); j++) { + symtab[i][j].st_name = elf32_to_cpu(symtab[i][j].st_name); + symtab[i][j].st_value = elf32_to_cpu(symtab[i][j].st_value); + symtab[i][j].st_size = elf32_to_cpu(symtab[i][j].st_size); + symtab[i][j].st_shndx = elf16_to_cpu(symtab[i][j].st_shndx); + } + } +} + + +static void read_relocs(FILE *fp) +{ + int i,j; + for(i = 0; i < ehdr.e_shnum; i++) { + if (shdr[i].sh_type != SHT_REL) { + continue; + } + reltab[i] = malloc(shdr[i].sh_size); + if (!reltab[i]) { + die("malloc of %d bytes for relocs failed\n", + shdr[i].sh_size); + } + if (fseek(fp, shdr[i].sh_offset, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + shdr[i].sh_offset, strerror(errno)); + } + if (fread(reltab[i], 1, shdr[i].sh_size, fp) != shdr[i].sh_size) { + die("Cannot read symbol table: %s\n", + strerror(errno)); + } + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + reltab[i][j].r_offset = elf32_to_cpu(reltab[i][j].r_offset); + reltab[i][j].r_info = elf32_to_cpu(reltab[i][j].r_info); + } + } +} + + +static void print_absolute_symbols(void) +{ + int i; + printf("Absolute symbols\n"); + printf(" Num: Value Size Type Bind Visibility Name\n"); + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + int j; + if (shdr[i].sh_type != SHT_SYMTAB) { + continue; + } + sh_symtab = symtab[i]; + sym_strtab = strtab[shdr[i].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(symtab[0][0]); j++) { + Elf32_Sym *sym; + const char *name; + sym = &symtab[i][j]; + name = sym_name(sym_strtab, sym); + if (sym->st_shndx != SHN_ABS) { + continue; + } + printf("%5d %08x %5d %10s %10s %12s %s\n", + j, sym->st_value, sym->st_size, + sym_type(ELF32_ST_TYPE(sym->st_info)), + sym_bind(ELF32_ST_BIND(sym->st_info)), + sym_visibility(ELF32_ST_VISIBILITY(sym->st_other)), + name); + } + } + printf("\n"); +} + +static void print_absolute_relocs(void) +{ + int i; + printf("Absolute relocations\n"); + printf("Offset Info Type Sym.Value Sym.Name\n"); + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + unsigned sec_applies, sec_symtab; + int j; + if (shdr[i].sh_type != SHT_REL) { + continue; + } + sec_symtab = shdr[i].sh_link; + sec_applies = shdr[i].sh_info; + if (!(shdr[sec_applies].sh_flags & SHF_ALLOC)) { + continue; + } + sh_symtab = symtab[sec_symtab]; + sym_strtab = strtab[shdr[sec_symtab].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + Elf32_Rel *rel; + Elf32_Sym *sym; + const char *name; + rel = &reltab[i][j]; + sym = &sh_symtab[ELF32_R_SYM(rel->r_info)]; + name = sym_name(sym_strtab, sym); + if (sym->st_shndx != SHN_ABS) { + continue; + } + printf("%08x %08x %10s %08x %s\n", + rel->r_offset, + rel->r_info, + rel_type(ELF32_R_TYPE(rel->r_info)), + sym->st_value, + name); + } + } + printf("\n"); +} + +static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) +{ + int i; + /* Walk through the relocations */ + for(i = 0; i < ehdr.e_shnum; i++) { + char *sym_strtab; + Elf32_Sym *sh_symtab; + unsigned sec_applies, sec_symtab; + int j; + if (shdr[i].sh_type != SHT_REL) { + continue; + } + sec_symtab = shdr[i].sh_link; + sec_applies = shdr[i].sh_info; + if (!(shdr[sec_applies].sh_flags & SHF_ALLOC)) { + continue; + } + sh_symtab = symtab[sec_symtab]; + sym_strtab = strtab[shdr[sec_symtab].sh_link]; + for(j = 0; j < shdr[i].sh_size/sizeof(reltab[0][0]); j++) { + Elf32_Rel *rel; + Elf32_Sym *sym; + unsigned r_type; + rel = &reltab[i][j]; + sym = &sh_symtab[ELF32_R_SYM(rel->r_info)]; + r_type = ELF32_R_TYPE(rel->r_info); + /* Don't visit relocations to absolute symbols */ + if (sym->st_shndx == SHN_ABS) { + continue; + } + if (r_type == R_386_PC32) { + /* PC relative relocations don't need to be adjusted */ + } + else if (r_type == R_386_32) { + /* Visit relocations that need to be adjusted */ + visit(rel, sym); + } + else { + die("Unsupported relocation type: %d\n", r_type); + } + } + } +} + +static void count_reloc(Elf32_Rel *rel, Elf32_Sym *sym) +{ + reloc_count += 1; +} + +static void collect_reloc(Elf32_Rel *rel, Elf32_Sym *sym) +{ + /* Remember the address that needs to be adjusted. */ + relocs[reloc_idx++] = rel->r_offset; +} + +static int cmp_relocs(const void *va, const void *vb) +{ + const unsigned long *a, *b; + a = va; b = vb; + return (*a == *b)? 0 : (*a > *b)? 1 : -1; +} + +static void emit_relocs(int as_text) +{ + int i; + /* Count how many relocations I have and allocate space for them. */ + reloc_count = 0; + walk_relocs(count_reloc); + relocs = malloc(reloc_count * sizeof(relocs[0])); + if (!relocs) { + die("malloc of %d entries for relocs failed\n", + reloc_count); + } + /* Collect up the relocations */ + reloc_idx = 0; + walk_relocs(collect_reloc); + + /* Order the relocations for more efficient processing */ + qsort(relocs, reloc_count, sizeof(relocs[0]), cmp_relocs); + + /* Print the relocations */ + if (as_text) { + /* Print the relocations in a form suitable that + * gas will like. + */ + printf(".section \".data.reloc\",\"a\"\n"); + printf(".balign 4\n"); + for(i = 0; i < reloc_count; i++) { + printf("\t .long 0x%08lx\n", relocs[i]); + } + printf("\n"); + } + else { + unsigned char buf[4]; + buf[0] = buf[1] = buf[2] = buf[3] = 0; + /* Print a stop */ + printf("%c%c%c%c", buf[0], buf[1], buf[2], buf[3]); + /* Now print each relocation */ + for(i = 0; i < reloc_count; i++) { + buf[0] = (relocs[i] >> 0) & 0xff; + buf[1] = (relocs[i] >> 8) & 0xff; + buf[2] = (relocs[i] >> 16) & 0xff; + buf[3] = (relocs[i] >> 24) & 0xff; + printf("%c%c%c%c", buf[0], buf[1], buf[2], buf[3]); + } + } +} + +static void usage(void) +{ + die("i386_reloc [--abs | --text] vmlinux\n"); +} + +int main(int argc, char **argv) +{ + int show_absolute; + int as_text; + const char *fname; + FILE *fp; + int i; + + show_absolute = 0; + as_text = 0; + fname = NULL; + for(i = 1; i < argc; i++) { + char *arg = argv[i]; + if (*arg == '-') { + if (strcmp(argv[1], "--abs") == 0) { + show_absolute = 1; + continue; + } + else if (strcmp(argv[1], "--text") == 0) { + as_text = 1; + continue; + } + } + else if (!fname) { + fname = arg; + continue; + } + usage(); + } + if (!fname) { + usage(); + } + fp = fopen(fname, "r"); + if (!fp) { + die("Cannot open %s: %s\n", + fname, strerror(errno)); + } + read_ehdr(fp); + read_shdrs(fp); + read_strtabs(fp); + read_symtabs(fp); + read_relocs(fp); + if (show_absolute) { + print_absolute_symbols(); + print_absolute_relocs(); + return 0; + } + emit_relocs(as_text); + return 0; +} diff --git a/arch/i386/boot/compressed/vmlinux.lds b/arch/i386/boot/compressed/vmlinux.lds new file mode 100644 index 000000000000..cc4854f6c6c1 --- /dev/null +++ b/arch/i386/boot/compressed/vmlinux.lds @@ -0,0 +1,43 @@ +OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386") +OUTPUT_ARCH(i386) +ENTRY(startup_32) +SECTIONS +{ + /* Be careful parts of head.S assume startup_32 is at + * address 0. + */ + . = 0 ; + .text.head : { + _head = . ; + *(.text.head) + _ehead = . ; + } + .data.compressed : { + *(.data.compressed) + } + .text : { + _text = .; /* Text */ + *(.text) + *(.text.*) + _etext = . ; + } + .rodata : { + _rodata = . ; + *(.rodata) /* read-only data */ + *(.rodata.*) + _erodata = . ; + } + .data : { + _data = . ; + *(.data) + *(.data.*) + _edata = . ; + } + .bss : { + _bss = . ; + *(.bss) + *(.bss.*) + *(COMMON) + _end = . ; + } +} diff --git a/arch/i386/boot/compressed/vmlinux.scr b/arch/i386/boot/compressed/vmlinux.scr index 1ed9d791f863..707a88f7f29e 100644 --- a/arch/i386/boot/compressed/vmlinux.scr +++ b/arch/i386/boot/compressed/vmlinux.scr @@ -1,9 +1,10 @@ SECTIONS { - .data : { + .data.compressed : { input_len = .; LONG(input_data_end - input_data) input_data = .; *(.data) + output_len = . - 4; input_data_end = .; } } diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S index 3aec4538a113..9aa8b0518184 100644 --- a/arch/i386/boot/setup.S +++ b/arch/i386/boot/setup.S @@ -588,11 +588,6 @@ rmodeswtch_normal: call default_switch rmodeswtch_end: -# we get the code32 start address and modify the below 'jmpi' -# (loader may have changed it) - movl %cs:code32_start, %eax - movl %eax, %cs:code32 - # Now we move the system to its rightful place ... but we check if we have a # big-kernel. In that case we *must* not move it ... testb $LOADED_HIGH, %cs:loadflags @@ -788,11 +783,12 @@ a20_err_msg: a20_done: #endif /* CONFIG_X86_VOYAGER */ -# set up gdt and idt +# set up gdt and idt and 32bit start address lidt idt_48 # load idt with 0,0 xorl %eax, %eax # Compute gdt_base movw %ds, %ax # (Convert %ds:gdt to a linear ptr) shll $4, %eax + addl %eax, code32 addl $gdt, %eax movl %eax, (gdt_48+2) lgdt gdt_48 # load gdt with whatever is @@ -851,9 +847,26 @@ flush_instr: # Manual, Mixing 16-bit and 32-bit code, page 16-6) .byte 0x66, 0xea # prefix + jmpi-opcode -code32: .long 0x1000 # will be set to 0x100000 - # for big kernels +code32: .long startup_32 # will be set to %cs+startup_32 .word __BOOT_CS +.code32 +startup_32: + movl $(__BOOT_DS), %eax + movl %eax, %ds + movl %eax, %es + movl %eax, %fs + movl %eax, %gs + movl %eax, %ss + + xorl %eax, %eax +1: incl %eax # check that A20 really IS enabled + movl %eax, 0x00000000 # loop forever if it isn't + cmpl %eax, 0x00100000 + je 1b + + # Jump to the 32bit entry point + jmpl *(code32_start - start + (DELTA_INITSEG << 4))(%esi) +.code16 # Here's a bunch of information about your current kernel.. kernel_version: .ascii UTS_RELEASE diff --git a/include/linux/screen_info.h b/include/linux/screen_info.h index 2925e66a6732..b02308ee7667 100644 --- a/include/linux/screen_info.h +++ b/include/linux/screen_info.h @@ -42,7 +42,8 @@ struct screen_info { u16 pages; /* 0x32 */ u16 vesa_attributes; /* 0x34 */ u32 capabilities; /* 0x36 */ - /* 0x3a -- 0x3f reserved for future expansion */ + /* 0x3a -- 0x3b reserved for future expansion */ + /* 0x3c -- 0x3f micro stack for relocatable kernels */ }; extern struct screen_info screen_info; -- cgit v1.2.3 From 2fff0a48416af891dce38fd425246e337831e0bb Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Thu, 7 Dec 2006 02:14:05 +0100 Subject: [PATCH] Generic: Move __user cast into probe_kernel_address Caller of probe_kernel_address shouldn't need to know that pka is internally implemented with __get_user. So move the __user cast into pka. Signed-off-by: Andi Kleen --- include/linux/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index a48d7f11c7be..65a68da8bd5d 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -36,7 +36,7 @@ static inline unsigned long __copy_from_user_nocache(void *to, long ret; \ \ inc_preempt_count(); \ - ret = __get_user(retval, addr); \ + ret = __get_user(retval, (__force typeof(*addr) __user *)addr);\ dec_preempt_count(); \ ret; \ }) -- cgit v1.2.3 From d7cd56111f30259e1b532a12e06f59f8e0a20355 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 7 Dec 2006 02:14:08 +0100 Subject: [PATCH] i386: cpu_detect extraction Both lhype and Xen want to call the core of the x86 cpu detect code before calling start_kernel. (extracted from larger patch) AK: folded in start_kernel header patch Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Rusty Russell Signed-off-by: Andi Kleen Cc: Jeremy Fitzhardinge Cc: Andi Kleen Signed-off-by: Andrew Morton --- arch/i386/kernel/cpu/common.c | 37 +++++++++++++++++++++---------------- include/asm-i386/processor.h | 3 +++ include/linux/start_kernel.h | 12 ++++++++++++ init/main.c | 1 + 4 files changed, 37 insertions(+), 16 deletions(-) create mode 100644 include/linux/start_kernel.h (limited to 'include/linux') diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c index cda41aef79ad..68bcb687019a 100644 --- a/arch/i386/kernel/cpu/common.c +++ b/arch/i386/kernel/cpu/common.c @@ -236,29 +236,14 @@ static int __cpuinit have_cpuid_p(void) return flag_is_changeable_p(X86_EFLAGS_ID); } -/* Do minimum CPU detection early. - Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment. - The others are not touched to avoid unwanted side effects. - - WARNING: this function is only called on the BP. Don't add code here - that is supposed to run on all CPUs. */ -static void __init early_cpu_detect(void) +void __init cpu_detect(struct cpuinfo_x86 *c) { - struct cpuinfo_x86 *c = &boot_cpu_data; - - c->x86_cache_alignment = 32; - - if (!have_cpuid_p()) - return; - /* Get vendor name */ cpuid(0x00000000, &c->cpuid_level, (int *)&c->x86_vendor_id[0], (int *)&c->x86_vendor_id[8], (int *)&c->x86_vendor_id[4]); - get_cpu_vendor(c, 1); - c->x86 = 4; if (c->cpuid_level >= 0x00000001) { u32 junk, tfms, cap0, misc; @@ -275,6 +260,26 @@ static void __init early_cpu_detect(void) } } +/* Do minimum CPU detection early. + Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment. + The others are not touched to avoid unwanted side effects. + + WARNING: this function is only called on the BP. Don't add code here + that is supposed to run on all CPUs. */ +static void __init early_cpu_detect(void) +{ + struct cpuinfo_x86 *c = &boot_cpu_data; + + c->x86_cache_alignment = 32; + + if (!have_cpuid_p()) + return; + + cpu_detect(c); + + get_cpu_vendor(c, 1); +} + static void __cpuinit generic_identify(struct cpuinfo_x86 * c) { u32 tfms, xlvl; diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h index 5f0418d0078c..a52d65440429 100644 --- a/include/asm-i386/processor.h +++ b/include/asm-i386/processor.h @@ -20,6 +20,7 @@ #include #include #include +#include /* flag for disabling the tsc */ extern int tsc_disable; @@ -112,6 +113,8 @@ extern struct cpuinfo_x86 cpu_data[]; extern int cpu_llc_id[NR_CPUS]; extern char ignore_fpu_irq; +void __init cpu_detect(struct cpuinfo_x86 *c); + extern void identify_cpu(struct cpuinfo_x86 *); extern void print_cpu_info(struct cpuinfo_x86 *); extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c); diff --git a/include/linux/start_kernel.h b/include/linux/start_kernel.h new file mode 100644 index 000000000000..d3e5f2756545 --- /dev/null +++ b/include/linux/start_kernel.h @@ -0,0 +1,12 @@ +#ifndef _LINUX_START_KERNEL_H +#define _LINUX_START_KERNEL_H + +#include +#include + +/* Define the prototype for start_kernel here, rather than cluttering + up something else. */ + +extern asmlinkage void __init start_kernel(void); + +#endif /* _LINUX_START_KERNEL_H */ diff --git a/init/main.c b/init/main.c index 36f608a7cfba..985c9ed86050 100644 --- a/init/main.c +++ b/init/main.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From 72486f1f8f0a2bc828b9d30cf4690cf2dd6807fc Mon Sep 17 00:00:00 2001 From: Siddha, Suresh B Date: Thu, 7 Dec 2006 02:14:10 +0100 Subject: [PATCH] i386: change the 'no_control' field to 'hotpluggable' in the struct cpu Change the 'no_control' field in the cpu struct to a more positive and better term 'hotpluggable'. And change(/cleanup) the logic accordingly. Signed-off-by: Suresh Siddha Signed-off-by: Andi Kleen Cc: Andi Kleen Cc: "Li, Shaohua" Signed-off-by: Andrew Morton --- arch/i386/kernel/topology.c | 4 ++-- arch/ia64/kernel/topology.c | 8 ++++---- arch/powerpc/kernel/sysfs.c | 8 ++++---- drivers/base/cpu.c | 6 +++--- include/linux/cpu.h | 2 +- 5 files changed, 14 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c index 844c08fdb225..79cf608e14ca 100644 --- a/arch/i386/kernel/topology.c +++ b/arch/i386/kernel/topology.c @@ -44,8 +44,8 @@ int arch_register_cpu(int num) * Also certain PCI quirks require not to enable hotplug control * for all CPU's. */ - if (!num || !enable_cpu_hotplug) - cpu_devices[num].cpu.no_control = 1; + if (num && enable_cpu_hotplug) + cpu_devices[num].cpu.hotpluggable = 1; return register_cpu(&cpu_devices[num].cpu, num); } diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 5629b45e89c6..687500ddb4b8 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -31,11 +31,11 @@ int arch_register_cpu(int num) { #if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) /* - * If CPEI cannot be re-targetted, and this is - * CPEI target, then dont create the control file + * If CPEI can be re-targetted or if this is not + * CPEI target, then it is hotpluggable */ - if (!can_cpei_retarget() && is_cpu_cpei_target(num)) - sysfs_cpus[num].cpu.no_control = 1; + if (can_cpei_retarget() || !is_cpu_cpei_target(num)) + sysfs_cpus[num].cpu.hotpluggable = 1; map_cpu_to_node(num, node_cpuid[num].nid); #endif diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index 22123a0d5416..63ed265b7f09 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -239,7 +239,7 @@ static void unregister_cpu_online(unsigned int cpu) struct cpu *c = &per_cpu(cpu_devices, cpu); struct sys_device *s = &c->sysdev; - BUG_ON(c->no_control); + BUG_ON(!c->hotpluggable); if (!firmware_has_feature(FW_FEATURE_ISERIES) && cpu_has_feature(CPU_FTR_SMT)) @@ -424,10 +424,10 @@ static int __init topology_init(void) * CPU. For instance, the boot cpu might never be valid * for hotplugging. */ - if (!ppc_md.cpu_die) - c->no_control = 1; + if (ppc_md.cpu_die) + c->hotpluggable = 1; - if (cpu_online(cpu) || (c->no_control == 0)) { + if (cpu_online(cpu) || c->hotpluggable) { register_cpu(c, cpu); sysdev_create_file(&c->sysdev, &attr_physical_id); diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index 1f745f12f94e..7fd095efaebd 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -104,8 +104,8 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL); /* * register_cpu - Setup a driverfs device for a CPU. - * @cpu - Callers can set the cpu->no_control field to 1, to indicate not to - * generate a control file in sysfs for this CPU. + * @cpu - cpu->hotpluggable field set to 1 will generate a control file in + * sysfs for this CPU. * @num - CPU number to use when creating the device. * * Initialize and register the CPU device. @@ -119,7 +119,7 @@ int __devinit register_cpu(struct cpu *cpu, int num) error = sysdev_register(&cpu->sysdev); - if (!error && !cpu->no_control) + if (!error && cpu->hotpluggable) register_cpu_control(cpu); if (!error) cpu_sys_devices[num] = &cpu->sysdev; diff --git a/include/linux/cpu.h b/include/linux/cpu.h index f02d71bf6894..ad90340e7dba 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -27,7 +27,7 @@ struct cpu { int node_id; /* The node which contains the CPU */ - int no_control; /* Should the sysfs control file be created? */ + int hotpluggable; /* creates sysfs control file if hotpluggable */ struct sys_device sysdev; }; -- cgit v1.2.3 From e1cccf48b182dd743c3c83a4fdf8dc570a43b393 Mon Sep 17 00:00:00 2001 From: Artiom Myaskouvskey Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: call efi_get_time during suspend Function efi_get_time called not only during init kernel phase but also during suspend (from get_cmos_time). When it is called from get_cmos_time the corresponding runtime service should be called in virtual and not in physical mode. Signed-off-by: Artiom Myaskouvskey Signed-off-by: Andi Kleen Cc: "Narayanan, Chandramouli" Cc: "Jiossy, Rami" Cc: "Satt, Shai" Cc: Andi Kleen Cc: Matt Domsch Signed-off-by: Andrew Morton --- arch/i386/kernel/efi.c | 17 ++++++++++++----- include/linux/efi.h | 2 +- 2 files changed, 13 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index 8b40648d0ef0..b92c7f0a358a 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -194,17 +194,24 @@ inline int efi_set_rtc_mmss(unsigned long nowtime) return 0; } /* - * This should only be used during kernel init and before runtime - * services have been remapped, therefore, we'll need to call in physical - * mode. Note, this call isn't used later, so mark it __init. + * This is used during kernel init before runtime + * services have been remapped and also during suspend, therefore, + * we'll need to call both in physical and virtual modes. */ -inline unsigned long __init efi_get_time(void) +inline unsigned long efi_get_time(void) { efi_status_t status; efi_time_t eft; efi_time_cap_t cap; - status = phys_efi_get_time(&eft, &cap); + if (efi.get_time) { + /* if we are in virtual mode use remapped function */ + status = efi.get_time(&eft, &cap); + } else { + /* we are in physical mode */ + status = phys_efi_get_time(&eft, &cap); + } + if (status != EFI_SUCCESS) printk("Oops: efitime: can't read time status: 0x%lx\n",status); diff --git a/include/linux/efi.h b/include/linux/efi.h index 66d621dbcb6c..91ecf49fbf21 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -300,7 +300,7 @@ extern int efi_mem_attribute_range (unsigned long phys_addr, unsigned long size, extern int __init efi_uart_console_only (void); extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); -extern unsigned long __init efi_get_time(void); +extern unsigned long efi_get_time(void); extern int __init efi_set_rtc_mmss(unsigned long nowtime); extern struct efi_memory_map memmap; -- cgit v1.2.3 From bf7e6a196318316e921f357557fca9d11d15f486 Mon Sep 17 00:00:00 2001 From: Artiom Myaskouvskey Date: Thu, 7 Dec 2006 02:14:11 +0100 Subject: [PATCH] i386: Preserve EFI run time regions with memmap parameter When using memmap kernel parameter in EFI boot we should also add to memory map memory regions of runtime services to enable their mapping later. AK: merged and cleaned up the patch Signed-off-by: Artiom Myaskouvskey Signed-off-by: Andi Kleen --- arch/i386/kernel/e820.c | 60 +++++++++++++++++++++++++++++++++++-------------- arch/i386/mm/init.c | 2 -- include/linux/efi.h | 1 + 3 files changed, 44 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/e820.c b/arch/i386/kernel/e820.c index b704790f7969..2f7d0a92fd7c 100644 --- a/arch/i386/kernel/e820.c +++ b/arch/i386/kernel/e820.c @@ -742,29 +742,55 @@ void __init print_memory_map(char *who) } } -void __init limit_regions(unsigned long long size) +static __init __always_inline void efi_limit_regions(unsigned long long size) { unsigned long long current_addr = 0; + efi_memory_desc_t *md, *next_md; + void *p, *p1; + int i, j; + + j = 0; + p1 = memmap.map; + for (p = p1, i = 0; p < memmap.map_end; p += memmap.desc_size, i++) { + md = p; + next_md = p1; + current_addr = md->phys_addr + + PFN_PHYS(md->num_pages); + if (is_available_memory(md)) { + if (md->phys_addr >= size) continue; + memcpy(next_md, md, memmap.desc_size); + if (current_addr >= size) { + next_md->num_pages -= + PFN_UP(current_addr-size); + } + p1 += memmap.desc_size; + next_md = p1; + j++; + } else if ((md->attribute & EFI_MEMORY_RUNTIME) == + EFI_MEMORY_RUNTIME) { + /* In order to make runtime services + * available we have to include runtime + * memory regions in memory map */ + memcpy(next_md, md, memmap.desc_size); + p1 += memmap.desc_size; + next_md = p1; + j++; + } + } + memmap.nr_map = j; + memmap.map_end = memmap.map + + (memmap.nr_map * memmap.desc_size); +} + +void __init limit_regions(unsigned long long size) +{ + unsigned long long current_addr; int i; print_memory_map("limit_regions start"); if (efi_enabled) { - efi_memory_desc_t *md; - void *p; - - for (p = memmap.map, i = 0; p < memmap.map_end; - p += memmap.desc_size, i++) { - md = p; - current_addr = md->phys_addr + (md->num_pages << 12); - if (md->type == EFI_CONVENTIONAL_MEMORY) { - if (current_addr >= size) { - md->num_pages -= - (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT); - memmap.nr_map = i + 1; - return; - } - } - } + efi_limit_regions(size); + return; } for (i = 0; i < e820.nr_map; i++) { current_addr = e820.map[i].addr + e820.map[i].size; diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 167416155ee4..f4dd048187ff 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -192,8 +192,6 @@ static inline int page_kills_ppro(unsigned long pagenr) return 0; } -extern int is_available_memory(efi_memory_desc_t *); - int page_is_ram(unsigned long pagenr) { int i; diff --git a/include/linux/efi.h b/include/linux/efi.h index 91ecf49fbf21..df1c91855f0e 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -302,6 +302,7 @@ extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); extern unsigned long efi_get_time(void); extern int __init efi_set_rtc_mmss(unsigned long nowtime); +extern int is_available_memory(efi_memory_desc_t * md); extern struct efi_memory_map memmap; /** -- cgit v1.2.3 From 334c29a64507dda187565dd0db0403de3d70ec8b Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 4 Dec 2006 19:31:51 -0800 Subject: [GENETLINK]: Move command capabilities to flags. This patch moves command capabilities to command flags. Other than being cleaner, saves several bytes. We increment the nlctrl version so as to signal to user space that to not expect the attributes. We will try to be careful not to do this too often ;-> Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/linux/genetlink.h | 6 +++--- net/netlink/genetlink.c | 18 ++++++++---------- 2 files changed, 11 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h index 9049dc65ae51..f7a93770e1be 100644 --- a/include/linux/genetlink.h +++ b/include/linux/genetlink.h @@ -17,6 +17,9 @@ struct genlmsghdr { #define GENL_HDRLEN NLMSG_ALIGN(sizeof(struct genlmsghdr)) #define GENL_ADMIN_PERM 0x01 +#define GENL_CMD_CAP_DO 0x02 +#define GENL_CMD_CAP_DUMP 0x04 +#define GENL_CMD_CAP_HASPOL 0x08 /* * List of reserved static generic netlink identifiers: @@ -58,9 +61,6 @@ enum { CTRL_ATTR_OP_UNSPEC, CTRL_ATTR_OP_ID, CTRL_ATTR_OP_FLAGS, - CTRL_ATTR_OP_POLICY, - CTRL_ATTR_OP_DOIT, - CTRL_ATTR_OP_DUMPIT, __CTRL_ATTR_OP_MAX, }; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index b9b03747c1f3..b5df749cba8f 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -143,6 +143,13 @@ int genl_register_ops(struct genl_family *family, struct genl_ops *ops) goto errout; } + if (ops->dumpit) + ops->flags |= GENL_CMD_CAP_DO; + if (ops->doit) + ops->flags |= GENL_CMD_CAP_DUMP; + if (ops->policy) + ops->flags |= GENL_CMD_CAP_HASPOL; + genl_lock(); list_add_tail(&ops->ops_list, &family->ops_list); genl_unlock(); @@ -387,7 +394,7 @@ static void genl_rcv(struct sock *sk, int len) static struct genl_family genl_ctrl = { .id = GENL_ID_CTRL, .name = "nlctrl", - .version = 0x1, + .version = 0x2, .maxattr = CTRL_ATTR_MAX, }; @@ -425,15 +432,6 @@ static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, NLA_PUT_U32(skb, CTRL_ATTR_OP_ID, ops->cmd); NLA_PUT_U32(skb, CTRL_ATTR_OP_FLAGS, ops->flags); - if (ops->policy) - NLA_PUT_FLAG(skb, CTRL_ATTR_OP_POLICY); - - if (ops->doit) - NLA_PUT_FLAG(skb, CTRL_ATTR_OP_DOIT); - - if (ops->dumpit) - NLA_PUT_FLAG(skb, CTRL_ATTR_OP_DUMPIT); - nla_nest_end(skb, nest); } -- cgit v1.2.3 From 7cf4c1a5fd13820d7591179c0b925d739b2be9a7 Mon Sep 17 00:00:00 2001 From: Kazunori MIYAZAWA Date: Sat, 28 Oct 2006 13:21:22 +1000 Subject: [IPSEC]: Add support for AES-XCBC-MAC The glue of xfrm. Signed-off-by: Kazunori MIYAZAWA Signed-off-by: Herbert Xu --- include/linux/pfkeyv2.h | 1 + net/xfrm/xfrm_algo.c | 17 +++++++++++++++++ 2 files changed, 18 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h index 0f0b880c4280..265bafab6494 100644 --- a/include/linux/pfkeyv2.h +++ b/include/linux/pfkeyv2.h @@ -285,6 +285,7 @@ struct sadb_x_sec_ctx { #define SADB_X_AALG_SHA2_384HMAC 6 #define SADB_X_AALG_SHA2_512HMAC 7 #define SADB_X_AALG_RIPEMD160HMAC 8 +#define SADB_X_AALG_AES_XCBC_MAC 9 #define SADB_X_AALG_NULL 251 /* kame */ #define SADB_AALG_MAX 251 diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c index 5a0dbeb6bbe8..6b381fc0383d 100644 --- a/net/xfrm/xfrm_algo.c +++ b/net/xfrm/xfrm_algo.c @@ -119,6 +119,23 @@ static struct xfrm_algo_desc aalg_list[] = { .sadb_alg_maxbits = 160 } }, +{ + .name = "xcbc(aes)", + + .uinfo = { + .auth = { + .icv_truncbits = 96, + .icv_fullbits = 128, + } + }, + + .desc = { + .sadb_alg_id = SADB_X_AALG_AES_XCBC_MAC, + .sadb_alg_ivlen = 0, + .sadb_alg_minbits = 128, + .sadb_alg_maxbits = 128 + } +}, }; static struct xfrm_algo_desc ealg_list[] = { -- cgit v1.2.3 From cc44215eaaa5e4032946b962353526ae6c370c0e Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 22 Nov 2006 17:55:00 +1100 Subject: [CRYPTO] api: Remove unused functions This patch removes the following no longer used functions: - api.c: crypto_alg_available() - digest.c: crypto_digest_init() - digest.c: crypto_digest_update() - digest.c: crypto_digest_final() - digest.c: crypto_digest_digest() Signed-off-by: Adrian Bunk Signed-off-by: Herbert Xu --- crypto/api.c | 15 --------------- crypto/digest.c | 48 ------------------------------------------------ include/linux/crypto.h | 22 ---------------------- 3 files changed, 85 deletions(-) (limited to 'include/linux') diff --git a/crypto/api.c b/crypto/api.c index 4fb7fa45cb0d..8c446871cd5b 100644 --- a/crypto/api.c +++ b/crypto/api.c @@ -466,23 +466,8 @@ void crypto_free_tfm(struct crypto_tfm *tfm) kfree(tfm); } -int crypto_alg_available(const char *name, u32 flags) -{ - int ret = 0; - struct crypto_alg *alg = crypto_alg_mod_lookup(name, 0, - CRYPTO_ALG_ASYNC); - - if (!IS_ERR(alg)) { - crypto_mod_put(alg); - ret = 1; - } - - return ret; -} - EXPORT_SYMBOL_GPL(crypto_alloc_tfm); EXPORT_SYMBOL_GPL(crypto_free_tfm); -EXPORT_SYMBOL_GPL(crypto_alg_available); int crypto_has_alg(const char *name, u32 type, u32 mask) { diff --git a/crypto/digest.c b/crypto/digest.c index 0155a94e4b15..8f4593268ce0 100644 --- a/crypto/digest.c +++ b/crypto/digest.c @@ -21,54 +21,6 @@ #include "internal.h" #include "scatterwalk.h" -void crypto_digest_init(struct crypto_tfm *tfm) -{ - struct crypto_hash *hash = crypto_hash_cast(tfm); - struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags }; - - crypto_hash_init(&desc); -} -EXPORT_SYMBOL_GPL(crypto_digest_init); - -void crypto_digest_update(struct crypto_tfm *tfm, - struct scatterlist *sg, unsigned int nsg) -{ - struct crypto_hash *hash = crypto_hash_cast(tfm); - struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags }; - unsigned int nbytes = 0; - unsigned int i; - - for (i = 0; i < nsg; i++) - nbytes += sg[i].length; - - crypto_hash_update(&desc, sg, nbytes); -} -EXPORT_SYMBOL_GPL(crypto_digest_update); - -void crypto_digest_final(struct crypto_tfm *tfm, u8 *out) -{ - struct crypto_hash *hash = crypto_hash_cast(tfm); - struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags }; - - crypto_hash_final(&desc, out); -} -EXPORT_SYMBOL_GPL(crypto_digest_final); - -void crypto_digest_digest(struct crypto_tfm *tfm, - struct scatterlist *sg, unsigned int nsg, u8 *out) -{ - struct crypto_hash *hash = crypto_hash_cast(tfm); - struct hash_desc desc = { .tfm = hash, .flags = tfm->crt_flags }; - unsigned int nbytes = 0; - unsigned int i; - - for (i = 0; i < nsg; i++) - nbytes += sg[i].length; - - crypto_hash_digest(&desc, sg, nbytes, out); -} -EXPORT_SYMBOL_GPL(crypto_digest_digest); - static int init(struct hash_desc *desc) { struct crypto_tfm *tfm = crypto_hash_tfm(desc->tfm); diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 6485e9716b36..4aa9046601da 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -241,12 +241,8 @@ int crypto_unregister_alg(struct crypto_alg *alg); * Algorithm query interface. */ #ifdef CONFIG_CRYPTO -int crypto_alg_available(const char *name, u32 flags) - __deprecated_for_modules; int crypto_has_alg(const char *name, u32 type, u32 mask); #else -static int crypto_alg_available(const char *name, u32 flags) - __deprecated_for_modules; static inline int crypto_alg_available(const char *name, u32 flags) { return 0; @@ -707,16 +703,6 @@ static inline void crypto_cipher_decrypt_one(struct crypto_cipher *tfm, dst, src); } -void crypto_digest_init(struct crypto_tfm *tfm) __deprecated_for_modules; -void crypto_digest_update(struct crypto_tfm *tfm, - struct scatterlist *sg, unsigned int nsg) - __deprecated_for_modules; -void crypto_digest_final(struct crypto_tfm *tfm, u8 *out) - __deprecated_for_modules; -void crypto_digest_digest(struct crypto_tfm *tfm, - struct scatterlist *sg, unsigned int nsg, u8 *out) - __deprecated_for_modules; - static inline struct crypto_hash *__crypto_hash_cast(struct crypto_tfm *tfm) { return (struct crypto_hash *)tfm; @@ -729,14 +715,6 @@ static inline struct crypto_hash *crypto_hash_cast(struct crypto_tfm *tfm) return __crypto_hash_cast(tfm); } -static int crypto_digest_setkey(struct crypto_tfm *tfm, const u8 *key, - unsigned int keylen) __deprecated; -static inline int crypto_digest_setkey(struct crypto_tfm *tfm, - const u8 *key, unsigned int keylen) -{ - return tfm->crt_hash.setkey(crypto_hash_cast(tfm), key, keylen); -} - static inline struct crypto_hash *crypto_alloc_hash(const char *alg_name, u32 type, u32 mask) { -- cgit v1.2.3 From 9ee0779e994c6916863045297b831212e285da3b Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Tue, 5 Dec 2006 13:44:31 -0800 Subject: [NETFILTER]: nf_conntrack: fix warning in PPTP helper Signed-off-by: Yasuyuki Kozakai Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_pptp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_pptp.h b/include/linux/netfilter/nf_conntrack_pptp.h index fb049ec11ff2..9d8144a488cd 100644 --- a/include/linux/netfilter/nf_conntrack_pptp.h +++ b/include/linux/netfilter/nf_conntrack_pptp.h @@ -2,6 +2,8 @@ #ifndef _NF_CONNTRACK_PPTP_H #define _NF_CONNTRACK_PPTP_H +#include + /* state of the control session */ enum pptp_ctrlsess_state { PPTP_SESSION_NONE, /* no session present */ @@ -295,7 +297,6 @@ union pptp_ctrl_union { /* crap needed for nf_conntrack_compat.h */ struct nf_conn; struct nf_conntrack_expect; -enum ip_conntrack_info; extern int (*nf_nat_pptp_hook_outbound)(struct sk_buff **pskb, -- cgit v1.2.3 From 161a09e737f0761ca064ee6a907313402f7a54b6 Mon Sep 17 00:00:00 2001 From: Joy Latten Date: Mon, 27 Nov 2006 13:11:54 -0600 Subject: audit: Add auditing to ipsec An audit message occurs when an ipsec SA or ipsec policy is created/deleted. Signed-off-by: Joy Latten Signed-off-by: James Morris Signed-off-by: David S. Miller --- include/linux/audit.h | 6 +++ include/net/xfrm.h | 19 +++++--- kernel/auditsc.c | 6 ++- net/key/af_key.c | 27 +++++++++-- net/xfrm/xfrm_policy.c | 120 ++++++++++++++++++++++++++++++++++++++++++++++++- net/xfrm/xfrm_state.c | 17 ++++++- net/xfrm/xfrm_user.c | 33 ++++++++++++-- 7 files changed, 213 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index b2ca666d9997..0e07db6cc0d0 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -101,6 +101,10 @@ #define AUDIT_MAC_CIPSOV4_DEL 1408 /* NetLabel: del CIPSOv4 DOI entry */ #define AUDIT_MAC_MAP_ADD 1409 /* NetLabel: add LSM domain mapping */ #define AUDIT_MAC_MAP_DEL 1410 /* NetLabel: del LSM domain mapping */ +#define AUDIT_MAC_IPSEC_ADDSA 1411 /* Add a XFRM state */ +#define AUDIT_MAC_IPSEC_DELSA 1412 /* Delete a XFRM state */ +#define AUDIT_MAC_IPSEC_ADDSPD 1413 /* Add a XFRM policy */ +#define AUDIT_MAC_IPSEC_DELSPD 1414 /* Delete a XFRM policy */ #define AUDIT_FIRST_KERN_ANOM_MSG 1700 #define AUDIT_LAST_KERN_ANOM_MSG 1799 @@ -377,6 +381,7 @@ extern void auditsc_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial); extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); +extern void audit_log_task_context(struct audit_buffer *ab); extern int __audit_ipc_obj(struct kern_ipc_perm *ipcp); extern int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); extern int audit_bprm(struct linux_binprm *bprm); @@ -449,6 +454,7 @@ extern int audit_n_rules; #define audit_inode_update(i) do { ; } while (0) #define auditsc_get_stamp(c,t,s) do { BUG(); } while (0) #define audit_get_loginuid(c) ({ -1; }) +#define audit_log_task_context(b) do { ; } while (0) #define audit_ipc_obj(i) ({ 0; }) #define audit_ipc_set_perm(q,u,g,m) ({ 0; }) #define audit_bprm(p) ({ 0; }) diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 15ec19dcf9c8..f699cdcab406 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -392,6 +392,15 @@ extern int xfrm_unregister_km(struct xfrm_mgr *km); extern unsigned int xfrm_policy_count[XFRM_POLICY_MAX*2]; +/* Audit Information */ +struct xfrm_audit +{ + uid_t loginuid; + u32 secid; +}; +void xfrm_audit_log(uid_t auid, u32 secid, int type, int result, + struct xfrm_policy *xp, struct xfrm_state *x); + static inline void xfrm_pol_hold(struct xfrm_policy *policy) { if (likely(policy != NULL)) @@ -906,7 +915,7 @@ static inline int xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **s #endif extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq); extern int xfrm_state_delete(struct xfrm_state *x); -extern void xfrm_state_flush(u8 proto); +extern void xfrm_state_flush(u8 proto, struct xfrm_audit *audit_info); extern int xfrm_replay_check(struct xfrm_state *x, __be32 seq); extern void xfrm_replay_advance(struct xfrm_state *x, __be32 seq); extern void xfrm_replay_notify(struct xfrm_state *x, int event); @@ -959,13 +968,13 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir, struct xfrm_selector *sel, struct xfrm_sec_ctx *ctx, int delete); struct xfrm_policy *xfrm_policy_byid(u8, int dir, u32 id, int delete); -void xfrm_policy_flush(u8 type); +void xfrm_policy_flush(u8 type, struct xfrm_audit *audit_info); u32 xfrm_get_acqseq(void); void xfrm_alloc_spi(struct xfrm_state *x, __be32 minspi, __be32 maxspi); -struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto, - xfrm_address_t *daddr, xfrm_address_t *saddr, +struct xfrm_state * xfrm_find_acq(u8 mode, u32 reqid, u8 proto, + xfrm_address_t *daddr, xfrm_address_t *saddr, int create, unsigned short family); -extern void xfrm_policy_flush(u8 type); +extern void xfrm_policy_flush(u8 type, struct xfrm_audit *audit_info); extern int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol); extern int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *xdst, struct flowi *fl, int family, int strict); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index ab97e5101232..40722e26de98 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -731,7 +731,7 @@ static inline void audit_free_context(struct audit_context *context) printk(KERN_ERR "audit: freed %d contexts\n", count); } -static void audit_log_task_context(struct audit_buffer *ab) +void audit_log_task_context(struct audit_buffer *ab) { char *ctx = NULL; ssize_t len = 0; @@ -760,6 +760,8 @@ error_path: return; } +EXPORT_SYMBOL(audit_log_task_context); + static void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk) { char name[sizeof(tsk->comm)]; @@ -1488,6 +1490,8 @@ uid_t audit_get_loginuid(struct audit_context *ctx) return ctx ? ctx->loginuid : -1; } +EXPORT_SYMBOL(audit_get_loginuid); + /** * __audit_mq_open - record audit data for a POSIX MQ open * @oflag: open flag diff --git a/net/key/af_key.c b/net/key/af_key.c index 0e1dbfbb9b10..5dd5094659a1 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -27,6 +27,7 @@ #include #include #include +#include #include @@ -1420,6 +1421,9 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, else err = xfrm_state_update(x); + xfrm_audit_log(audit_get_loginuid(current->audit_context), 0, + AUDIT_MAC_IPSEC_ADDSA, err ? 0 : 1, NULL, x); + if (err < 0) { x->km.state = XFRM_STATE_DEAD; __xfrm_state_put(x); @@ -1460,8 +1464,12 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h err = -EPERM; goto out; } - + err = xfrm_state_delete(x); + + xfrm_audit_log(audit_get_loginuid(current->audit_context), 0, + AUDIT_MAC_IPSEC_DELSA, err ? 0 : 1, NULL, x); + if (err < 0) goto out; @@ -1637,12 +1645,15 @@ static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hd { unsigned proto; struct km_event c; + struct xfrm_audit audit_info; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - xfrm_state_flush(proto); + audit_info.loginuid = audit_get_loginuid(current->audit_context); + audit_info.secid = 0; + xfrm_state_flush(proto, &audit_info); c.data.proto = proto; c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; @@ -2205,6 +2216,9 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h err = xfrm_policy_insert(pol->sadb_x_policy_dir-1, xp, hdr->sadb_msg_type != SADB_X_SPDUPDATE); + xfrm_audit_log(audit_get_loginuid(current->audit_context), 0, + AUDIT_MAC_IPSEC_ADDSPD, err ? 0 : 1, xp, NULL); + if (err) goto out; @@ -2282,6 +2296,10 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN, pol->sadb_x_policy_dir-1, &sel, tmp.security, 1); security_xfrm_policy_free(&tmp); + + xfrm_audit_log(audit_get_loginuid(current->audit_context), 0, + AUDIT_MAC_IPSEC_DELSPD, (xp) ? 1 : 0, xp, NULL); + if (xp == NULL) return -ENOENT; @@ -2416,8 +2434,11 @@ static int key_notify_policy_flush(struct km_event *c) static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { struct km_event c; + struct xfrm_audit audit_info; - xfrm_policy_flush(XFRM_POLICY_TYPE_MAIN); + audit_info.loginuid = audit_get_loginuid(current->audit_context); + audit_info.secid = 0; + xfrm_policy_flush(XFRM_POLICY_TYPE_MAIN, &audit_info); c.data.type = XFRM_POLICY_TYPE_MAIN; c.event = XFRM_MSG_FLUSHPOLICY; c.pid = hdr->sadb_msg_pid; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 4f04222698d9..47c13649bac1 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "xfrm_hash.h" @@ -804,7 +805,7 @@ struct xfrm_policy *xfrm_policy_byid(u8 type, int dir, u32 id, int delete) } EXPORT_SYMBOL(xfrm_policy_byid); -void xfrm_policy_flush(u8 type) +void xfrm_policy_flush(u8 type, struct xfrm_audit *audit_info) { int dir; @@ -824,6 +825,9 @@ void xfrm_policy_flush(u8 type) hlist_del(&pol->byidx); write_unlock_bh(&xfrm_policy_lock); + xfrm_audit_log(audit_info->loginuid, audit_info->secid, + AUDIT_MAC_IPSEC_DELSPD, 1, pol, NULL); + xfrm_policy_kill(pol); killed++; @@ -842,6 +846,11 @@ void xfrm_policy_flush(u8 type) hlist_del(&pol->byidx); write_unlock_bh(&xfrm_policy_lock); + xfrm_audit_log(audit_info->loginuid, + audit_info->secid, + AUDIT_MAC_IPSEC_DELSPD, 1, + pol, NULL); + xfrm_policy_kill(pol); killed++; @@ -1977,6 +1986,115 @@ int xfrm_bundle_ok(struct xfrm_policy *pol, struct xfrm_dst *first, EXPORT_SYMBOL(xfrm_bundle_ok); +/* Audit addition and deletion of SAs and ipsec policy */ + +void xfrm_audit_log(uid_t auid, u32 sid, int type, int result, + struct xfrm_policy *xp, struct xfrm_state *x) +{ + + char *secctx; + u32 secctx_len; + struct xfrm_sec_ctx *sctx = NULL; + struct audit_buffer *audit_buf; + int family; + extern int audit_enabled; + + if (audit_enabled == 0) + return; + + audit_buf = audit_log_start(current->audit_context, GFP_ATOMIC, type); + if (audit_buf == NULL) + return; + + switch(type) { + case AUDIT_MAC_IPSEC_ADDSA: + audit_log_format(audit_buf, "SAD add: auid=%u", auid); + break; + case AUDIT_MAC_IPSEC_DELSA: + audit_log_format(audit_buf, "SAD delete: auid=%u", auid); + break; + case AUDIT_MAC_IPSEC_ADDSPD: + audit_log_format(audit_buf, "SPD add: auid=%u", auid); + break; + case AUDIT_MAC_IPSEC_DELSPD: + audit_log_format(audit_buf, "SPD delete: auid=%u", auid); + break; + default: + return; + } + + if (sid != 0 && + security_secid_to_secctx(sid, &secctx, &secctx_len) == 0) + audit_log_format(audit_buf, " subj=%s", secctx); + else + audit_log_task_context(audit_buf); + + if (xp) { + family = xp->selector.family; + if (xp->security) + sctx = xp->security; + } else { + family = x->props.family; + if (x->security) + sctx = x->security; + } + + if (sctx) + audit_log_format(audit_buf, + " sec_alg=%u sec_doi=%u sec_obj=%s", + sctx->ctx_alg, sctx->ctx_doi, sctx->ctx_str); + + switch(family) { + case AF_INET: + { + struct in_addr saddr, daddr; + if (xp) { + saddr.s_addr = xp->selector.saddr.a4; + daddr.s_addr = xp->selector.daddr.a4; + } else { + saddr.s_addr = x->props.saddr.a4; + daddr.s_addr = x->id.daddr.a4; + } + audit_log_format(audit_buf, + " src=%u.%u.%u.%u dst=%u.%u.%u.%u", + NIPQUAD(saddr), NIPQUAD(daddr)); + } + break; + case AF_INET6: + { + struct in6_addr saddr6, daddr6; + if (xp) { + memcpy(&saddr6, xp->selector.saddr.a6, + sizeof(struct in6_addr)); + memcpy(&daddr6, xp->selector.daddr.a6, + sizeof(struct in6_addr)); + } else { + memcpy(&saddr6, x->props.saddr.a6, + sizeof(struct in6_addr)); + memcpy(&daddr6, x->id.daddr.a6, + sizeof(struct in6_addr)); + } + audit_log_format(audit_buf, + " src=" NIP6_FMT "dst=" NIP6_FMT, + NIP6(saddr6), NIP6(daddr6)); + } + break; + } + + if (x) + audit_log_format(audit_buf, " spi=%lu(0x%lx) protocol=%s", + (unsigned long)ntohl(x->id.spi), + (unsigned long)ntohl(x->id.spi), + x->id.proto == IPPROTO_AH ? "AH" : + (x->id.proto == IPPROTO_ESP ? + "ESP" : "IPCOMP")); + + audit_log_format(audit_buf, " res=%u", result); + audit_log_end(audit_buf); +} + +EXPORT_SYMBOL(xfrm_audit_log); + int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo) { int err = 0; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index a14c88bf17f0..d5d3a6f1f609 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -20,6 +20,7 @@ #include #include #include +#include #include "xfrm_hash.h" @@ -238,6 +239,7 @@ static void xfrm_timer_handler(unsigned long data) unsigned long now = (unsigned long)xtime.tv_sec; long next = LONG_MAX; int warn = 0; + int err = 0; spin_lock(&x->lock); if (x->km.state == XFRM_STATE_DEAD) @@ -295,9 +297,14 @@ expired: next = 2; goto resched; } - if (!__xfrm_state_delete(x) && x->id.spi) + + err = __xfrm_state_delete(x); + if (!err && x->id.spi) km_state_expired(x, 1, 0); + xfrm_audit_log(audit_get_loginuid(current->audit_context), 0, + AUDIT_MAC_IPSEC_DELSA, err ? 0 : 1, NULL, x); + out: spin_unlock(&x->lock); } @@ -384,9 +391,10 @@ int xfrm_state_delete(struct xfrm_state *x) } EXPORT_SYMBOL(xfrm_state_delete); -void xfrm_state_flush(u8 proto) +void xfrm_state_flush(u8 proto, struct xfrm_audit *audit_info) { int i; + int err = 0; spin_lock_bh(&xfrm_state_lock); for (i = 0; i <= xfrm_state_hmask; i++) { @@ -400,6 +408,11 @@ restart: spin_unlock_bh(&xfrm_state_lock); xfrm_state_delete(x); + err = xfrm_state_delete(x); + xfrm_audit_log(audit_info->loginuid, + audit_info->secid, + AUDIT_MAC_IPSEC_DELSA, + err ? 0 : 1, NULL, x); xfrm_state_put(x); spin_lock_bh(&xfrm_state_lock); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 311205ffa775..e5372b11fc8f 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -31,6 +31,7 @@ #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) #include #endif +#include static int verify_one_alg(struct rtattr **xfrma, enum xfrm_attr_type_t type) { @@ -454,6 +455,9 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) else err = xfrm_state_update(x); + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_ADDSA, err ? 0 : 1, NULL, x); + if (err < 0) { x->km.state = XFRM_STATE_DEAD; __xfrm_state_put(x); @@ -523,6 +527,10 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) } err = xfrm_state_delete(x); + + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_DELSA, err ? 0 : 1, NULL, x); + if (err < 0) goto out; @@ -1030,6 +1038,9 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr * a type XFRM_MSG_UPDPOLICY - JHS */ excl = nlh->nlmsg_type == XFRM_MSG_NEWPOLICY; err = xfrm_policy_insert(p->dir, xp, excl); + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_DELSPD, err ? 0 : 1, xp, NULL); + if (err) { security_xfrm_policy_free(xp); kfree(xp); @@ -1257,6 +1268,10 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security, delete); security_xfrm_policy_free(&tmp); } + if (delete) + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_DELSPD, (xp) ? 1 : 0, xp, NULL); + if (xp == NULL) return -ENOENT; @@ -1291,8 +1306,11 @@ static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma { struct km_event c; struct xfrm_usersa_flush *p = NLMSG_DATA(nlh); + struct xfrm_audit audit_info; - xfrm_state_flush(p->proto); + audit_info.loginuid = NETLINK_CB(skb).loginuid; + audit_info.secid = NETLINK_CB(skb).sid; + xfrm_state_flush(p->proto, &audit_info); c.data.proto = p->proto; c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; @@ -1442,12 +1460,15 @@ static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **x struct km_event c; u8 type = XFRM_POLICY_TYPE_MAIN; int err; + struct xfrm_audit audit_info; err = copy_from_user_policy_type(&type, (struct rtattr **)xfrma); if (err) return err; - xfrm_policy_flush(type); + audit_info.loginuid = NETLINK_CB(skb).loginuid; + audit_info.secid = NETLINK_CB(skb).sid; + xfrm_policy_flush(type, &audit_info); c.data.type = type; c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; @@ -1502,6 +1523,9 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, void * err = 0; if (up->hard) { xfrm_policy_delete(xp, p->dir); + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_DELSPD, 1, xp, NULL); + } else { // reset the timers here? printk("Dont know what to do with soft policy expire\n"); @@ -1533,8 +1557,11 @@ static int xfrm_add_sa_expire(struct sk_buff *skb, struct nlmsghdr *nlh, void ** goto out; km_state_expired(x, ue->hard, current->pid); - if (ue->hard) + if (ue->hard) { __xfrm_state_delete(x); + xfrm_audit_log(NETLINK_CB(skb).loginuid, NETLINK_CB(skb).sid, + AUDIT_MAC_IPSEC_DELSA, 1, NULL, x); + } out: spin_unlock_bh(&x->lock); xfrm_state_put(x); -- cgit v1.2.3 From 676917d488212303ce4a7d033815ce8799201010 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 7 Dec 2006 00:20:22 -0800 Subject: [TG3]: Add 5787F device ID. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 4 +++- include/linux/pci_ids.h | 1 + 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index e6561c1d85d6..5514828b7929 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -192,6 +192,7 @@ static struct pci_device_id tg3_pci_tbl[] = { {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5786)}, {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787)}, {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787M)}, + {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787F)}, {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5714)}, {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5714S)}, {PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5715)}, @@ -10859,7 +10860,8 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->pdev->device == PCI_DEVICE_ID_TIGON3_5705F)) || (tp->pdev->vendor == PCI_VENDOR_ID_BROADCOM && (tp->pdev->device == PCI_DEVICE_ID_TIGON3_5751F || - tp->pdev->device == PCI_DEVICE_ID_TIGON3_5753F)) || + tp->pdev->device == PCI_DEVICE_ID_TIGON3_5753F || + tp->pdev->device == PCI_DEVICE_ID_TIGON3_5787F)) || GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) tp->tg3_flags |= TG3_FLAG_10_100_ONLY; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index c09da1e30c54..edddcce68b94 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1931,6 +1931,7 @@ #define PCI_DEVICE_ID_TIGON3_5750M 0x167c #define PCI_DEVICE_ID_TIGON3_5751M 0x167d #define PCI_DEVICE_ID_TIGON3_5751F 0x167e +#define PCI_DEVICE_ID_TIGON3_5787F 0x167f #define PCI_DEVICE_ID_TIGON3_5787M 0x1693 #define PCI_DEVICE_ID_TIGON3_5782 0x1696 #define PCI_DEVICE_ID_TIGON3_5786 0x169a -- cgit v1.2.3 From 9d26e213423923c9e033ccd373705118131827c9 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 7 Dec 2006 00:21:14 -0800 Subject: [TG3]: Add TG3_FLG2_IS_NIC flag. Add Tg3_FLG2_IS_NIC flag to unambiguously determine whether the device is NIC or onboard. Previously, the EEPROM_WRITE_PROT flag was overloaded to also mean onboard. With the separation, we can support some devices that are onboard but do not use eeprom write protect. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 37 ++++++++++++++++++++++++------------- drivers/net/tg3.h | 1 + include/linux/pci_ids.h | 2 ++ 3 files changed, 27 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 5514828b7929..16bc05fe531f 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -1062,7 +1062,7 @@ static void tg3_frob_aux_power(struct tg3 *tp) { struct tg3 *tp_peer = tp; - if ((tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) != 0) + if ((tp->tg3_flags2 & TG3_FLG2_IS_NIC) == 0) return; if ((GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5704) || @@ -1213,8 +1213,8 @@ static int tg3_set_power_state(struct tg3 *tp, pci_power_t state) power_control); udelay(100); /* Delay after power state change */ - /* Switch out of Vaux if it is not a LOM */ - if (!(tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT)) + /* Switch out of Vaux if it is a NIC */ + if (tp->tg3_flags2 & TG3_FLG2_IS_NIC) tw32_wait_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl, 100); return 0; @@ -6397,16 +6397,17 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy) udelay(40); /* tp->grc_local_ctrl is partially set up during tg3_get_invariants(). - * If TG3_FLAG_EEPROM_WRITE_PROT is set, we should read the + * If TG3_FLG2_IS_NIC is zero, we should read the * register to preserve the GPIO settings for LOMs. The GPIOs, * whether used as inputs or outputs, are set by boot code after * reset. */ - if (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) { + if (!(tp->tg3_flags2 & TG3_FLG2_IS_NIC)) { u32 gpio_mask; - gpio_mask = GRC_LCLCTRL_GPIO_OE0 | GRC_LCLCTRL_GPIO_OE2 | - GRC_LCLCTRL_GPIO_OUTPUT0 | GRC_LCLCTRL_GPIO_OUTPUT2; + gpio_mask = GRC_LCLCTRL_GPIO_OE0 | GRC_LCLCTRL_GPIO_OE1 | + GRC_LCLCTRL_GPIO_OE2 | GRC_LCLCTRL_GPIO_OUTPUT0 | + GRC_LCLCTRL_GPIO_OUTPUT1 | GRC_LCLCTRL_GPIO_OUTPUT2; if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5752) gpio_mask |= GRC_LCLCTRL_GPIO_OE3 | @@ -6418,8 +6419,9 @@ static int tg3_reset_hw(struct tg3 *tp, int reset_phy) tp->grc_local_ctrl |= tr32(GRC_LOCAL_CTRL) & gpio_mask; /* GPIO1 must be driven high for eeprom write protect */ - tp->grc_local_ctrl |= (GRC_LCLCTRL_GPIO_OE1 | - GRC_LCLCTRL_GPIO_OUTPUT1); + if (tp->tg3_flags & TG3_FLAG_EEPROM_WRITE_PROT) + tp->grc_local_ctrl |= (GRC_LCLCTRL_GPIO_OE1 | + GRC_LCLCTRL_GPIO_OUTPUT1); } tw32_f(GRC_LOCAL_CTRL, tp->grc_local_ctrl); udelay(100); @@ -9963,8 +9965,10 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp) tp->tg3_flags |= TG3_FLAG_EEPROM_WRITE_PROT; if (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5906) { - if (!(tr32(PCIE_TRANSACTION_CFG) & PCIE_TRANS_CFG_LOM)) + if (!(tr32(PCIE_TRANSACTION_CFG) & PCIE_TRANS_CFG_LOM)) { tp->tg3_flags &= ~TG3_FLAG_EEPROM_WRITE_PROT; + tp->tg3_flags2 |= TG3_FLG2_IS_NIC; + } return; } @@ -10064,10 +10068,17 @@ static void __devinit tg3_get_eeprom_hw_cfg(struct tg3 *tp) tp->pdev->subsystem_vendor == PCI_VENDOR_ID_DELL) tp->led_ctrl = LED_CTRL_MODE_PHY_2; - if (nic_cfg & NIC_SRAM_DATA_CFG_EEPROM_WP) + if (nic_cfg & NIC_SRAM_DATA_CFG_EEPROM_WP) { tp->tg3_flags |= TG3_FLAG_EEPROM_WRITE_PROT; - else + if ((tp->pdev->subsystem_vendor == + PCI_VENDOR_ID_ARIMA) && + (tp->pdev->subsystem_device == 0x205a || + tp->pdev->subsystem_device == 0x2063)) + tp->tg3_flags &= ~TG3_FLAG_EEPROM_WRITE_PROT; + } else { tp->tg3_flags &= ~TG3_FLAG_EEPROM_WRITE_PROT; + tp->tg3_flags2 |= TG3_FLG2_IS_NIC; + } if (nic_cfg & NIC_SRAM_DATA_CFG_ASF_ENABLE) { tp->tg3_flags |= TG3_FLAG_ENABLE_ASF; @@ -10693,7 +10704,7 @@ static int __devinit tg3_get_invariants(struct tg3 *tp) tp->tg3_flags |= TG3_FLAG_SRAM_USE_CONFIG; /* Get eeprom hw config before calling tg3_set_power_state(). - * In particular, the TG3_FLAG_EEPROM_WRITE_PROT flag must be + * In particular, the TG3_FLG2_IS_NIC flag must be * determined before calling tg3_set_power_state() so that * we know whether or not to switch out of Vaux power. * When the flag is set, it means that GPIO1 is used for eeprom diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h index 92f53000bce6..dfaf4ed127bd 100644 --- a/drivers/net/tg3.h +++ b/drivers/net/tg3.h @@ -2233,6 +2233,7 @@ struct tg3 { #define TG3_FLG2_PCI_EXPRESS 0x00000200 #define TG3_FLG2_ASF_NEW_HANDSHAKE 0x00000400 #define TG3_FLG2_HW_AUTONEG 0x00000800 +#define TG3_FLG2_IS_NIC 0x00001000 #define TG3_FLG2_PHY_SERDES 0x00002000 #define TG3_FLG2_CAPACITIVE_COUPLING 0x00004000 #define TG3_FLG2_FLASH 0x00008000 diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index edddcce68b94..ebc597d57242 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2003,6 +2003,8 @@ #define PCI_DEVICE_ID_FARSITE_TE1 0x1610 #define PCI_DEVICE_ID_FARSITE_TE1C 0x1612 +#define PCI_VENDOR_ID_ARIMA 0x161f + #define PCI_VENDOR_ID_SIBYTE 0x166d #define PCI_DEVICE_ID_BCM1250_PCI 0x0001 #define PCI_DEVICE_ID_BCM1250_HT 0x0002 -- cgit v1.2.3 From 676dcb8bc2ec78d80091037773598d6ec8c673d6 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:31:30 -0800 Subject: [PATCH] add bottom_half.h With CONFIG_SMP=n: drivers/input/ff-memless.c:384: warning: implicit declaration of function 'local_bh_disable' drivers/input/ff-memless.c:393: warning: implicit declaration of function 'local_bh_enable' Really linux/spinlock.h should include linux/interrupt.h. But interrupt.h includes sched.h which will need spinlock.h. So the patch breaks the _bh declarations out into a separate header and includes it in both interrupt.h and spinlock.h. Cc: "Randy.Dunlap" Cc: Andi Kleen Cc: Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bottom_half.h | 10 ++++++++++ include/linux/interrupt.h | 7 +------ include/linux/spinlock.h | 1 + 3 files changed, 12 insertions(+), 6 deletions(-) create mode 100644 include/linux/bottom_half.h (limited to 'include/linux') diff --git a/include/linux/bottom_half.h b/include/linux/bottom_half.h new file mode 100644 index 000000000000..777dbf695d44 --- /dev/null +++ b/include/linux/bottom_half.h @@ -0,0 +1,10 @@ +#ifndef _LINUX_BH_H +#define _LINUX_BH_H + +extern void local_bh_disable(void); +extern void __local_bh_enable(void); +extern void _local_bh_enable(void); +extern void local_bh_enable(void); +extern void local_bh_enable_ip(unsigned long ip); + +#endif /* _LINUX_BH_H */ diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 5b83e7b59621..de7593f4e895 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -217,12 +218,6 @@ static inline void __deprecated save_and_cli(unsigned long *x) #define save_and_cli(x) save_and_cli(&x) #endif /* CONFIG_SMP */ -extern void local_bh_disable(void); -extern void __local_bh_enable(void); -extern void _local_bh_enable(void); -extern void local_bh_enable(void); -extern void local_bh_enable_ip(unsigned long ip); - /* PLEASE, avoid to allocate new softirqs, if you need not _really_ high frequency threaded job scheduling. For almost all the purposes tasklets are more than enough. F.e. all serial device BHs et diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 8451052ca66f..94b767d64275 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -52,6 +52,7 @@ #include #include #include +#include #include -- cgit v1.2.3 From 89689ae7f95995723fbcd5c116c47933a3bb8b13 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:31:45 -0800 Subject: [PATCH] Get rid of zone_table[] The zone table is mostly not needed. If we have a node in the page flags then we can get to the zone via NODE_DATA() which is much more likely to be already in the cpu cache. In case of SMP and UP NODE_DATA() is a constant pointer which allows us to access an exact replica of zonetable in the node_zones field. In all of the above cases there will be no need at all for the zone table. The only remaining case is if in a NUMA system the node numbers do not fit into the page flags. In that case we make sparse generate a table that maps sections to nodes and use that table to to figure out the node number. This table is sized to fit in a single cache line for the known 32 bit NUMA platform which makes it very likely that the information can be obtained without a cache miss. For sparsemem the zone table seems to be have been fairly large based on the maximum possible number of sections and the number of zones per node. There is some memory saving by removing zone_table. The main benefit is to reduce the cache foootprint of the VM from the frequent lookups of zones. Plus it simplifies the page allocator. [akpm@osdl.org: build fix] Signed-off-by: Christoph Lameter Cc: Dave Hansen Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 57 +++++++++++++++++++++++++++++++++-------------------- mm/memory_hotplug.c | 1 - mm/page_alloc.c | 22 --------------------- mm/sparse.c | 23 +++++++++++++++++++++ 4 files changed, 59 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index d538de901965..ab6e4974f379 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -396,7 +396,9 @@ void split_page(struct page *page, unsigned int order); * We are going to use the flags for the page to node mapping if its in * there. This includes the case where there is no node, so it is implicit. */ -#define FLAGS_HAS_NODE (NODES_WIDTH > 0 || NODES_SHIFT == 0) +#if !(NODES_WIDTH > 0 || NODES_SHIFT == 0) +#define NODE_NOT_IN_PAGE_FLAGS +#endif #ifndef PFN_SECTION_SHIFT #define PFN_SECTION_SHIFT 0 @@ -411,13 +413,18 @@ void split_page(struct page *page, unsigned int order); #define NODES_PGSHIFT (NODES_PGOFF * (NODES_WIDTH != 0)) #define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_WIDTH != 0)) -/* NODE:ZONE or SECTION:ZONE is used to lookup the zone from a page. */ -#if FLAGS_HAS_NODE -#define ZONETABLE_SHIFT (NODES_SHIFT + ZONES_SHIFT) +/* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allcator */ +#ifdef NODE_NOT_IN_PAGEFLAGS +#define ZONEID_SHIFT (SECTIONS_SHIFT + ZONES_SHIFT) #else -#define ZONETABLE_SHIFT (SECTIONS_SHIFT + ZONES_SHIFT) +#define ZONEID_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#endif + +#if ZONES_WIDTH > 0 +#define ZONEID_PGSHIFT ZONES_PGSHIFT +#else +#define ZONEID_PGSHIFT NODES_PGOFF #endif -#define ZONETABLE_PGSHIFT ZONES_PGSHIFT #if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED #error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED @@ -426,23 +433,25 @@ void split_page(struct page *page, unsigned int order); #define ZONES_MASK ((1UL << ZONES_WIDTH) - 1) #define NODES_MASK ((1UL << NODES_WIDTH) - 1) #define SECTIONS_MASK ((1UL << SECTIONS_WIDTH) - 1) -#define ZONETABLE_MASK ((1UL << ZONETABLE_SHIFT) - 1) +#define ZONEID_MASK ((1UL << ZONEID_SHIFT) - 1) static inline enum zone_type page_zonenum(struct page *page) { return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; } -struct zone; -extern struct zone *zone_table[]; - +/* + * The identification function is only used by the buddy allocator for + * determining if two pages could be buddies. We are not really + * identifying a zone since we could be using a the section number + * id if we have not node id available in page flags. + * We guarantee only that it will return the same value for two + * combinable pages in a zone. + */ static inline int page_zone_id(struct page *page) { - return (page->flags >> ZONETABLE_PGSHIFT) & ZONETABLE_MASK; -} -static inline struct zone *page_zone(struct page *page) -{ - return zone_table[page_zone_id(page)]; + BUILD_BUG_ON(ZONEID_PGSHIFT == 0 && ZONEID_MASK); + return (page->flags >> ZONEID_PGSHIFT) & ZONEID_MASK; } static inline unsigned long zone_to_nid(struct zone *zone) @@ -454,13 +463,20 @@ static inline unsigned long zone_to_nid(struct zone *zone) #endif } +#ifdef NODE_NOT_IN_PAGE_FLAGS +extern unsigned long page_to_nid(struct page *page); +#else static inline unsigned long page_to_nid(struct page *page) { - if (FLAGS_HAS_NODE) - return (page->flags >> NODES_PGSHIFT) & NODES_MASK; - else - return zone_to_nid(page_zone(page)); + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; } +#endif + +static inline struct zone *page_zone(struct page *page) +{ + return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)]; +} + static inline unsigned long page_to_section(struct page *page) { return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK; @@ -477,6 +493,7 @@ static inline void set_page_node(struct page *page, unsigned long node) page->flags &= ~(NODES_MASK << NODES_PGSHIFT); page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; } + static inline void set_page_section(struct page *page, unsigned long section) { page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT); @@ -947,8 +964,6 @@ extern void mem_init(void); extern void show_mem(void); extern void si_meminfo(struct sysinfo * val); extern void si_meminfo_node(struct sysinfo *val, int nid); -extern void zonetable_add(struct zone *zone, int nid, enum zone_type zid, - unsigned long pfn, unsigned long size); #ifdef CONFIG_NUMA extern void setup_per_cpu_pageset(void); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index fd678a662eae..0c055a090f4d 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -72,7 +72,6 @@ static int __add_zone(struct zone *zone, unsigned long phys_start_pfn) return ret; } memmap_init_zone(nr_pages, nid, zone_type, phys_start_pfn); - zonetable_add(zone, nid, zone_type, phys_start_pfn, nr_pages); return 0; } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 08360aa111f9..23bc5bcbdcf9 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -83,13 +83,6 @@ int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { EXPORT_SYMBOL(totalram_pages); -/* - * Used by page_zone() to look up the address of the struct zone whose - * id is encoded in the upper bits of page->flags - */ -struct zone *zone_table[1 << ZONETABLE_SHIFT] __read_mostly; -EXPORT_SYMBOL(zone_table); - static char *zone_names[MAX_NR_ZONES] = { "DMA", #ifdef CONFIG_ZONE_DMA32 @@ -1715,20 +1708,6 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone, } } -#define ZONETABLE_INDEX(x, zone_nr) ((x << ZONES_SHIFT) | zone_nr) -void zonetable_add(struct zone *zone, int nid, enum zone_type zid, - unsigned long pfn, unsigned long size) -{ - unsigned long snum = pfn_to_section_nr(pfn); - unsigned long end = pfn_to_section_nr(pfn + size); - - if (FLAGS_HAS_NODE) - zone_table[ZONETABLE_INDEX(nid, zid)] = zone; - else - for (; snum <= end; snum++) - zone_table[ZONETABLE_INDEX(snum, zid)] = zone; -} - #ifndef __HAVE_ARCH_MEMMAP_INIT #define memmap_init(size, nid, zone, start_pfn) \ memmap_init_zone((size), (nid), (zone), (start_pfn)) @@ -2421,7 +2400,6 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat, if (!size) continue; - zonetable_add(zone, nid, j, zone_start_pfn, size); ret = init_currently_empty_zone(zone, zone_start_pfn, size); BUG_ON(ret); zone_start_pfn += size; diff --git a/mm/sparse.c b/mm/sparse.c index b3c82ba30012..158d6a2a5263 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -24,6 +24,25 @@ struct mem_section mem_section[NR_SECTION_ROOTS][SECTIONS_PER_ROOT] #endif EXPORT_SYMBOL(mem_section); +#ifdef NODE_NOT_IN_PAGE_FLAGS +/* + * If we did not store the node number in the page then we have to + * do a lookup in the section_to_node_table in order to find which + * node the page belongs to. + */ +#if MAX_NUMNODES <= 256 +static u8 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned; +#else +static u16 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned; +#endif + +unsigned long page_to_nid(struct page *page) +{ + return section_to_node_table[page_to_section(page)]; +} +EXPORT_SYMBOL(page_to_nid); +#endif + #ifdef CONFIG_SPARSEMEM_EXTREME static struct mem_section *sparse_index_alloc(int nid) { @@ -49,6 +68,10 @@ static int sparse_index_init(unsigned long section_nr, int nid) struct mem_section *section; int ret = 0; +#ifdef NODE_NOT_IN_PAGE_FLAGS + section_to_node_table[section_nr] = nid; +#endif + if (mem_section[root]) return -EEXIST; -- cgit v1.2.3 From 9276b1bc96a132f4068fdee00983c532f43d3a26 Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Wed, 6 Dec 2006 20:31:48 -0800 Subject: [PATCH] memory page_alloc zonelist caching speedup Optimize the critical zonelist scanning for free pages in the kernel memory allocator by caching the zones that were found to be full recently, and skipping them. Remembers the zones in a zonelist that were short of free memory in the last second. And it stashes a zone-to-node table in the zonelist struct, to optimize that conversion (minimize its cache footprint.) Recent changes: This differs in a significant way from a similar patch that I posted a week ago. Now, instead of having a nodemask_t of recently full nodes, I have a bitmask of recently full zones. This solves a problem that last weeks patch had, which on systems with multiple zones per node (such as DMA zone) would take seeing any of these zones full as meaning that all zones on that node were full. Also I changed names - from "zonelist faster" to "zonelist cache", as that seemed to better convey what we're doing here - caching some of the key zonelist state (for faster access.) See below for some performance benchmark results. After all that discussion with David on why I didn't need them, I went and got some ;). I wanted to verify that I had not hurt the normal case of memory allocation noticeably. At least for my one little microbenchmark, I found (1) the normal case wasn't affected, and (2) workloads that forced scanning across multiple nodes for memory improved up to 10% fewer System CPU cycles and lower elapsed clock time ('sys' and 'real'). Good. See details, below. I didn't have the logic in get_page_from_freelist() for various full nodes and zone reclaim failures correct. That should be fixed up now - notice the new goto labels zonelist_scan, this_zone_full, and try_next_zone, in get_page_from_freelist(). There are two reasons I persued this alternative, over some earlier proposals that would have focused on optimizing the fake numa emulation case by caching the last useful zone: 1) Contrary to what I said before, we (SGI, on large ia64 sn2 systems) have seen real customer loads where the cost to scan the zonelist was a problem, due to many nodes being full of memory before we got to a node we could use. Or at least, I think we have. This was related to me by another engineer, based on experiences from some time past. So this is not guaranteed. Most likely, though. The following approach should help such real numa systems just as much as it helps fake numa systems, or any combination thereof. 2) The effort to distinguish fake from real numa, using node_distance, so that we could cache a fake numa node and optimize choosing it over equivalent distance fake nodes, while continuing to properly scan all real nodes in distance order, was going to require a nasty blob of zonelist and node distance munging. The following approach has no new dependency on node distances or zone sorting. See comment in the patch below for a description of what it actually does. Technical details of note (or controversy): - See the use of "zlc_active" and "did_zlc_setup" below, to delay adding any work for this new mechanism until we've looked at the first zone in zonelist. I figured the odds of the first zone having the memory we needed were high enough that we should just look there, first, then get fancy only if we need to keep looking. - Some odd hackery was needed to add items to struct zonelist, while not tripping up the custom zonelists built by the mm/mempolicy.c code for MPOL_BIND. My usual wordy comments below explain this. Search for "MPOL_BIND". - Some per-node data in the struct zonelist is now modified frequently, with no locking. Multiple CPU cores on a node could hit and mangle this data. The theory is that this is just performance hint data, and the memory allocator will work just fine despite any such mangling. The fields at risk are the struct 'zonelist_cache' fields 'fullzones' (a bitmask) and 'last_full_zap' (unsigned long jiffies). It should all be self correcting after at most a one second delay. - This still does a linear scan of the same lengths as before. All I've optimized is making the scan faster, not algorithmically shorter. It is now able to scan a compact array of 'unsigned short' in the case of many full nodes, so one cache line should cover quite a few nodes, rather than each node hitting another one or two new and distinct cache lines. - If both Andi and Nick don't find this too complicated, I will be (pleasantly) flabbergasted. - I removed the comment claiming we only use one cachline's worth of zonelist. We seem, at least in the fake numa case, to have put the lie to that claim. - I pay no attention to the various watermarks and such in this performance hint. A node could be marked full for one watermark, and then skipped over when searching for a page using a different watermark. I think that's actually quite ok, as it will tend to slightly increase the spreading of memory over other nodes, away from a memory stressed node. =============== Performance - some benchmark results and analysis: This benchmark runs a memory hog program that uses multiple threads to touch alot of memory as quickly as it can. Multiple runs were made, touching 12, 38, 64 or 90 GBytes out of the total 96 GBytes on the system, and using 1, 19, 37, or 55 threads (on a 56 CPU system.) System, user and real (elapsed) timings were recorded for each run, shown in units of seconds, in the table below. Two kernels were tested - 2.6.18-mm3 and the same kernel with this zonelist caching patch added. The table also shows the percentage improvement the zonelist caching sys time is over (lower than) the stock *-mm kernel. number 2.6.18-mm3 zonelist-cache delta (< 0 good) percent GBs N ------------ -------------- ---------------- systime mem threads sys user real sys user real sys user real better 12 1 153 24 177 151 24 176 -2 0 -1 1% 12 19 99 22 8 99 22 8 0 0 0 0% 12 37 111 25 6 112 25 6 1 0 0 -0% 12 55 115 25 5 110 23 5 -5 -2 0 4% 38 1 502 74 576 497 73 570 -5 -1 -6 0% 38 19 426 78 48 373 76 39 -53 -2 -9 12% 38 37 544 83 36 547 82 36 3 -1 0 -0% 38 55 501 77 23 511 80 24 10 3 1 -1% 64 1 917 125 1042 890 124 1014 -27 -1 -28 2% 64 19 1118 138 119 965 141 103 -153 3 -16 13% 64 37 1202 151 94 1136 150 81 -66 -1 -13 5% 64 55 1118 141 61 1072 140 58 -46 -1 -3 4% 90 1 1342 177 1519 1275 174 1450 -67 -3 -69 4% 90 19 2392 199 192 2116 189 176 -276 -10 -16 11% 90 37 3313 238 175 2972 225 145 -341 -13 -30 10% 90 55 1948 210 104 1843 213 100 -105 3 -4 5% Notes: 1) This test ran a memory hog program that started a specified number N of threads, and had each thread allocate and touch 1/N'th of the total memory to be used in the test run in a single loop, writing a constant word to memory, one store every 4096 bytes. Watching this test during some earlier trial runs, I would see each of these threads sit down on one CPU and stay there, for the remainder of the pass, a different CPU for each thread. 2) The 'real' column is not comparable to the 'sys' or 'user' columns. The 'real' column is seconds wall clock time elapsed, from beginning to end of that test pass. The 'sys' and 'user' columns are total CPU seconds spent on that test pass. For a 19 thread test run, for example, the sum of 'sys' and 'user' could be up to 19 times the number of 'real' elapsed wall clock seconds. 3) Tests were run on a fresh, single-user boot, to minimize the amount of memory already in use at the start of the test, and to minimize the amount of background activity that might interfere. 4) Tests were done on a 56 CPU, 28 Node system with 96 GBytes of RAM. 5) Notice that the 'real' time gets large for the single thread runs, even though the measured 'sys' and 'user' times are modest. I'm not sure what that means - probably something to do with it being slow for one thread to be accessing memory along ways away. Perhaps the fake numa system, running ostensibly the same workload, would not show this substantial degradation of 'real' time for one thread on many nodes -- lets hope not. 6) The high thread count passes (one thread per CPU - on 55 of 56 CPUs) ran quite efficiently, as one might expect. Each pair of threads needed to allocate and touch the memory on the node the two threads shared, a pleasantly parallizable workload. 7) The intermediate thread count passes, when asking for alot of memory forcing them to go to a few neighboring nodes, improved the most with this zonelist caching patch. Conclusions: * This zonelist cache patch probably makes little difference one way or the other for most workloads on real numa hardware, if those workloads avoid heavy off node allocations. * For memory intensive workloads requiring substantial off-node allocations on real numa hardware, this patch improves both kernel and elapsed timings up to ten per-cent. * For fake numa systems, I'm optimistic, but will have to leave that up to Rohit Seth to actually test (once I get him a 2.6.18 backport.) Signed-off-by: Paul Jackson Cc: Rohit Seth Cc: Christoph Lameter Cc: David Rientjes Cc: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpuset.h | 2 + include/linux/mmzone.h | 85 ++++++++++++++++++++-- mm/mempolicy.c | 2 + mm/page_alloc.c | 188 +++++++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 265 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 4d8adf663681..748d2c996631 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -23,6 +23,7 @@ extern void cpuset_fork(struct task_struct *p); extern void cpuset_exit(struct task_struct *p); extern cpumask_t cpuset_cpus_allowed(struct task_struct *p); extern nodemask_t cpuset_mems_allowed(struct task_struct *p); +#define cpuset_current_mems_allowed (current->mems_allowed) void cpuset_init_current_mems_allowed(void); void cpuset_update_task_memory_state(void); #define cpuset_nodes_subset_current_mems_allowed(nodes) \ @@ -83,6 +84,7 @@ static inline nodemask_t cpuset_mems_allowed(struct task_struct *p) return node_possible_map; } +#define cpuset_current_mems_allowed (node_online_map) static inline void cpuset_init_current_mems_allowed(void) {} static inline void cpuset_update_task_memory_state(void) {} #define cpuset_nodes_subset_current_mems_allowed(nodes) (1) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index e06683e2bea3..09bf9d8d7b72 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -288,19 +288,94 @@ struct zone { */ #define DEF_PRIORITY 12 +/* Maximum number of zones on a zonelist */ +#define MAX_ZONES_PER_ZONELIST (MAX_NUMNODES * MAX_NR_ZONES) + +#ifdef CONFIG_NUMA +/* + * We cache key information from each zonelist for smaller cache + * footprint when scanning for free pages in get_page_from_freelist(). + * + * 1) The BITMAP fullzones tracks which zones in a zonelist have come + * up short of free memory since the last time (last_fullzone_zap) + * we zero'd fullzones. + * 2) The array z_to_n[] maps each zone in the zonelist to its node + * id, so that we can efficiently evaluate whether that node is + * set in the current tasks mems_allowed. + * + * Both fullzones and z_to_n[] are one-to-one with the zonelist, + * indexed by a zones offset in the zonelist zones[] array. + * + * The get_page_from_freelist() routine does two scans. During the + * first scan, we skip zones whose corresponding bit in 'fullzones' + * is set or whose corresponding node in current->mems_allowed (which + * comes from cpusets) is not set. During the second scan, we bypass + * this zonelist_cache, to ensure we look methodically at each zone. + * + * Once per second, we zero out (zap) fullzones, forcing us to + * reconsider nodes that might have regained more free memory. + * The field last_full_zap is the time we last zapped fullzones. + * + * This mechanism reduces the amount of time we waste repeatedly + * reexaming zones for free memory when they just came up low on + * memory momentarilly ago. + * + * The zonelist_cache struct members logically belong in struct + * zonelist. However, the mempolicy zonelists constructed for + * MPOL_BIND are intentionally variable length (and usually much + * shorter). A general purpose mechanism for handling structs with + * multiple variable length members is more mechanism than we want + * here. We resort to some special case hackery instead. + * + * The MPOL_BIND zonelists don't need this zonelist_cache (in good + * part because they are shorter), so we put the fixed length stuff + * at the front of the zonelist struct, ending in a variable length + * zones[], as is needed by MPOL_BIND. + * + * Then we put the optional zonelist cache on the end of the zonelist + * struct. This optional stuff is found by a 'zlcache_ptr' pointer in + * the fixed length portion at the front of the struct. This pointer + * both enables us to find the zonelist cache, and in the case of + * MPOL_BIND zonelists, (which will just set the zlcache_ptr to NULL) + * to know that the zonelist cache is not there. + * + * The end result is that struct zonelists come in two flavors: + * 1) The full, fixed length version, shown below, and + * 2) The custom zonelists for MPOL_BIND. + * The custom MPOL_BIND zonelists have a NULL zlcache_ptr and no zlcache. + * + * Even though there may be multiple CPU cores on a node modifying + * fullzones or last_full_zap in the same zonelist_cache at the same + * time, we don't lock it. This is just hint data - if it is wrong now + * and then, the allocator will still function, perhaps a bit slower. + */ + + +struct zonelist_cache { + DECLARE_BITMAP(fullzones, MAX_ZONES_PER_ZONELIST); /* zone full? */ + unsigned short z_to_n[MAX_ZONES_PER_ZONELIST]; /* zone->nid */ + unsigned long last_full_zap; /* when last zap'd (jiffies) */ +}; +#else +struct zonelist_cache; +#endif + /* * One allocation request operates on a zonelist. A zonelist * is a list of zones, the first one is the 'goal' of the * allocation, the other zones are fallback zones, in decreasing * priority. * - * Right now a zonelist takes up less than a cacheline. We never - * modify it apart from boot-up, and only a few indices are used, - * so despite the zonelist table being relatively big, the cache - * footprint of this construct is very small. + * If zlcache_ptr is not NULL, then it is just the address of zlcache, + * as explained above. If zlcache_ptr is NULL, there is no zlcache. */ + struct zonelist { - struct zone *zones[MAX_NUMNODES * MAX_NR_ZONES + 1]; // NULL delimited + struct zonelist_cache *zlcache_ptr; // NULL or &zlcache + struct zone *zones[MAX_ZONES_PER_ZONELIST + 1]; // NULL delimited +#ifdef CONFIG_NUMA + struct zonelist_cache zlcache; // optional ... +#endif }; #ifdef CONFIG_ARCH_POPULATES_NODE_MAP diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 617fb31086ee..fb907236bbd8 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -141,9 +141,11 @@ static struct zonelist *bind_zonelist(nodemask_t *nodes) enum zone_type k; max = 1 + MAX_NR_ZONES * nodes_weight(*nodes); + max++; /* space for zlcache_ptr (see mmzone.h) */ zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL); if (!zl) return NULL; + zl->zlcache_ptr = NULL; num = 0; /* First put in the highest zones from all nodes, then all the next lower zones etc. Avoid empty zones because the memory allocator diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 23bc5bcbdcf9..230771d3c6b6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -918,6 +918,126 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, return 1; } +#ifdef CONFIG_NUMA +/* + * zlc_setup - Setup for "zonelist cache". Uses cached zone data to + * skip over zones that are not allowed by the cpuset, or that have + * been recently (in last second) found to be nearly full. See further + * comments in mmzone.h. Reduces cache footprint of zonelist scans + * that have to skip over alot of full or unallowed zones. + * + * If the zonelist cache is present in the passed in zonelist, then + * returns a pointer to the allowed node mask (either the current + * tasks mems_allowed, or node_online_map.) + * + * If the zonelist cache is not available for this zonelist, does + * nothing and returns NULL. + * + * If the fullzones BITMAP in the zonelist cache is stale (more than + * a second since last zap'd) then we zap it out (clear its bits.) + * + * We hold off even calling zlc_setup, until after we've checked the + * first zone in the zonelist, on the theory that most allocations will + * be satisfied from that first zone, so best to examine that zone as + * quickly as we can. + */ +static nodemask_t *zlc_setup(struct zonelist *zonelist, int alloc_flags) +{ + struct zonelist_cache *zlc; /* cached zonelist speedup info */ + nodemask_t *allowednodes; /* zonelist_cache approximation */ + + zlc = zonelist->zlcache_ptr; + if (!zlc) + return NULL; + + if (jiffies - zlc->last_full_zap > 1 * HZ) { + bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); + zlc->last_full_zap = jiffies; + } + + allowednodes = !in_interrupt() && (alloc_flags & ALLOC_CPUSET) ? + &cpuset_current_mems_allowed : + &node_online_map; + return allowednodes; +} + +/* + * Given 'z' scanning a zonelist, run a couple of quick checks to see + * if it is worth looking at further for free memory: + * 1) Check that the zone isn't thought to be full (doesn't have its + * bit set in the zonelist_cache fullzones BITMAP). + * 2) Check that the zones node (obtained from the zonelist_cache + * z_to_n[] mapping) is allowed in the passed in allowednodes mask. + * Return true (non-zero) if zone is worth looking at further, or + * else return false (zero) if it is not. + * + * This check -ignores- the distinction between various watermarks, + * such as GFP_HIGH, GFP_ATOMIC, PF_MEMALLOC, ... If a zone is + * found to be full for any variation of these watermarks, it will + * be considered full for up to one second by all requests, unless + * we are so low on memory on all allowed nodes that we are forced + * into the second scan of the zonelist. + * + * In the second scan we ignore this zonelist cache and exactly + * apply the watermarks to all zones, even it is slower to do so. + * We are low on memory in the second scan, and should leave no stone + * unturned looking for a free page. + */ +static int zlc_zone_worth_trying(struct zonelist *zonelist, struct zone **z, + nodemask_t *allowednodes) +{ + struct zonelist_cache *zlc; /* cached zonelist speedup info */ + int i; /* index of *z in zonelist zones */ + int n; /* node that zone *z is on */ + + zlc = zonelist->zlcache_ptr; + if (!zlc) + return 1; + + i = z - zonelist->zones; + n = zlc->z_to_n[i]; + + /* This zone is worth trying if it is allowed but not full */ + return node_isset(n, *allowednodes) && !test_bit(i, zlc->fullzones); +} + +/* + * Given 'z' scanning a zonelist, set the corresponding bit in + * zlc->fullzones, so that subsequent attempts to allocate a page + * from that zone don't waste time re-examining it. + */ +static void zlc_mark_zone_full(struct zonelist *zonelist, struct zone **z) +{ + struct zonelist_cache *zlc; /* cached zonelist speedup info */ + int i; /* index of *z in zonelist zones */ + + zlc = zonelist->zlcache_ptr; + if (!zlc) + return; + + i = z - zonelist->zones; + + set_bit(i, zlc->fullzones); +} + +#else /* CONFIG_NUMA */ + +static nodemask_t *zlc_setup(struct zonelist *zonelist, int alloc_flags) +{ + return NULL; +} + +static int zlc_zone_worth_trying(struct zonelist *zonelist, struct zone **z, + nodemask_t *allowednodes) +{ + return 1; +} + +static void zlc_mark_zone_full(struct zonelist *zonelist, struct zone **z) +{ +} +#endif /* CONFIG_NUMA */ + /* * get_page_from_freelist goes through the zonelist trying to allocate * a page. @@ -926,23 +1046,32 @@ static struct page * get_page_from_freelist(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, int alloc_flags) { - struct zone **z = zonelist->zones; + struct zone **z; struct page *page = NULL; - int classzone_idx = zone_idx(*z); + int classzone_idx = zone_idx(zonelist->zones[0]); struct zone *zone; + nodemask_t *allowednodes = NULL;/* zonelist_cache approximation */ + int zlc_active = 0; /* set if using zonelist_cache */ + int did_zlc_setup = 0; /* just call zlc_setup() one time */ +zonelist_scan: /* - * Go through the zonelist once, looking for a zone with enough free. + * Scan zonelist, looking for a zone with enough free. * See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ + z = zonelist->zones; + do { + if (NUMA_BUILD && zlc_active && + !zlc_zone_worth_trying(zonelist, z, allowednodes)) + continue; zone = *z; if (unlikely(NUMA_BUILD && (gfp_mask & __GFP_THISNODE) && zone->zone_pgdat != zonelist->zones[0]->zone_pgdat)) break; if ((alloc_flags & ALLOC_CPUSET) && !cpuset_zone_allowed(zone, gfp_mask)) - continue; + goto try_next_zone; if (!(alloc_flags & ALLOC_NO_WATERMARKS)) { unsigned long mark; @@ -956,15 +1085,30 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, classzone_idx, alloc_flags)) { if (!zone_reclaim_mode || !zone_reclaim(zone, gfp_mask, order)) - continue; + goto this_zone_full; } } page = buffered_rmqueue(zonelist, zone, order, gfp_mask); if (page) break; - +this_zone_full: + if (NUMA_BUILD) + zlc_mark_zone_full(zonelist, z); +try_next_zone: + if (NUMA_BUILD && !did_zlc_setup) { + /* we do zlc_setup after the first zone is tried */ + allowednodes = zlc_setup(zonelist, alloc_flags); + zlc_active = 1; + did_zlc_setup = 1; + } } while (*(++z) != NULL); + + if (unlikely(NUMA_BUILD && page == NULL && zlc_active)) { + /* Disable zlc cache for second zonelist scan */ + zlc_active = 0; + goto zonelist_scan; + } return page; } @@ -1535,6 +1679,24 @@ static void __meminit build_zonelists(pg_data_t *pgdat) } } +/* Construct the zonelist performance cache - see further mmzone.h */ +static void __meminit build_zonelist_cache(pg_data_t *pgdat) +{ + int i; + + for (i = 0; i < MAX_NR_ZONES; i++) { + struct zonelist *zonelist; + struct zonelist_cache *zlc; + struct zone **z; + + zonelist = pgdat->node_zonelists + i; + zonelist->zlcache_ptr = zlc = &zonelist->zlcache; + bitmap_zero(zlc->fullzones, MAX_ZONES_PER_ZONELIST); + for (z = zonelist->zones; *z; z++) + zlc->z_to_n[z - zonelist->zones] = zone_to_nid(*z); + } +} + #else /* CONFIG_NUMA */ static void __meminit build_zonelists(pg_data_t *pgdat) @@ -1572,14 +1734,26 @@ static void __meminit build_zonelists(pg_data_t *pgdat) } } +/* non-NUMA variant of zonelist performance cache - just NULL zlcache_ptr */ +static void __meminit build_zonelist_cache(pg_data_t *pgdat) +{ + int i; + + for (i = 0; i < MAX_NR_ZONES; i++) + pgdat->node_zonelists[i].zlcache_ptr = NULL; +} + #endif /* CONFIG_NUMA */ /* return values int ....just for stop_machine_run() */ static int __meminit __build_all_zonelists(void *dummy) { int nid; - for_each_online_node(nid) + + for_each_online_node(nid) { build_zonelists(NODE_DATA(nid)); + build_zonelist_cache(NODE_DATA(nid)); + } return 0; } -- cgit v1.2.3 From 7253f4ef04b1cd138baf2b29a95473743ac0a307 Mon Sep 17 00:00:00 2001 From: Paul Jackson Date: Wed, 6 Dec 2006 20:31:49 -0800 Subject: [PATCH] memory page_alloc zonelist caching reorder structure Rearrange the struct members in the 'struct zonelist_cache' structure, so as to put the readonly (once initialized) z_to_n[] array first, where it will come right after the zones[] array in struct zonelist. This pretty much eliminates the chance that the two frequently written elements of 'struct zonelist_cache', the fullzones bitmap and last_full_zap times, will end up on the same cache line as the performance sensitive, frequently read, never (after init) written zones[] array. Keeping frequently written data off frequently read cache lines is good for performance. Thanks to Rohit Seth for the suggestion. Signed-off-by: Paul Jackson Cc: Rohit Seth Cc: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 09bf9d8d7b72..da6002dec205 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -352,8 +352,8 @@ struct zone { struct zonelist_cache { - DECLARE_BITMAP(fullzones, MAX_ZONES_PER_ZONELIST); /* zone full? */ unsigned short z_to_n[MAX_ZONES_PER_ZONELIST]; /* zone->nid */ + DECLARE_BITMAP(fullzones, MAX_ZONES_PER_ZONELIST); /* zone full? */ unsigned long last_full_zap; /* when last zap'd (jiffies) */ }; #else -- cgit v1.2.3 From 7602bdf2fd14a40dd9b104e516fdc05e1bd17952 Mon Sep 17 00:00:00 2001 From: Ashwin Chaugule Date: Wed, 6 Dec 2006 20:31:57 -0800 Subject: [PATCH] new scheme to preempt swap token The new swap token patches replace the current token traversal algo. The old algo had a crude timeout parameter that was used to handover the token from one task to another. This algo, transfers the token to the tasks that are in need of the token. The urgency for the token is based on the number of times a task is required to swap-in pages. Accordingly, the priority of a task is incremented if it has been badly affected due to swap-outs. To ensure that the token doesnt bounce around rapidly, the token holders are given a priority boost. The priority of tasks is also decremented, if their rate of swap-in's keeps reducing. This way, the condition to check whether to pre-empt the swap token, is a matter of comparing two task's priority fields. [akpm@osdl.org: cleanups] Signed-off-by: Ashwin Chaugule Cc: Rik van Riel Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 13 ++++-- include/linux/swap.h | 1 - kernel/fork.c | 8 ++++ kernel/sysctl.c | 11 ----- mm/thrash.c | 116 ++++++++++++++++++++------------------------------ 5 files changed, 63 insertions(+), 86 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index eafe4a7b8237..cad6a16260f7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -344,9 +344,16 @@ struct mm_struct { /* Architecture-specific MM context */ mm_context_t context; - /* Token based thrashing protection. */ - unsigned long swap_token_time; - char recent_pagein; + /* Swap token stuff */ + /* + * Last value of global fault stamp as seen by this process. + * In other words, this value gives an indication of how long + * it has been since this task got the token. + * Look at mm/thrash.c + */ + unsigned int faultstamp; + unsigned int token_priority; + unsigned int last_interval; /* coredumping support */ int core_waiters; diff --git a/include/linux/swap.h b/include/linux/swap.h index e7c36ba2a2db..89f8a39773bf 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -259,7 +259,6 @@ extern spinlock_t swap_lock; /* linux/mm/thrash.c */ extern struct mm_struct * swap_token_mm; -extern unsigned long swap_token_default_timeout; extern void grab_swap_token(void); extern void __put_swap_token(struct mm_struct *); diff --git a/kernel/fork.c b/kernel/fork.c index 8cdd3e72ba55..5678e6c61ef2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -479,6 +479,10 @@ static struct mm_struct *dup_mm(struct task_struct *tsk) memcpy(mm, oldmm, sizeof(*mm)); + /* Initializing for Swap token stuff */ + mm->token_priority = 0; + mm->last_interval = 0; + if (!mm_init(mm)) goto fail_nomem; @@ -542,6 +546,10 @@ static int copy_mm(unsigned long clone_flags, struct task_struct * tsk) goto fail_nomem; good_mm: + /* Initializing for Swap token stuff */ + mm->token_priority = 0; + mm->last_interval = 0; + tsk->mm = mm; tsk->active_mm = mm; return 0; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 09e569f4792b..7abe9704e75a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -977,17 +977,6 @@ static ctl_table vm_table[] = { .extra1 = &zero, }, #endif -#ifdef CONFIG_SWAP - { - .ctl_name = VM_SWAP_TOKEN_TIMEOUT, - .procname = "swap_token_timeout", - .data = &swap_token_default_timeout, - .maxlen = sizeof(swap_token_default_timeout), - .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, - }, -#endif #ifdef CONFIG_NUMA { .ctl_name = VM_ZONE_RECLAIM_MODE, diff --git a/mm/thrash.c b/mm/thrash.c index f4c560b4a2b7..19e428ca8b03 100644 --- a/mm/thrash.c +++ b/mm/thrash.c @@ -7,100 +7,74 @@ * * Simple token based thrashing protection, using the algorithm * described in: http://www.cs.wm.edu/~sjiang/token.pdf + * + * Sep 2006, Ashwin Chaugule + * Improved algorithm to pass token: + * Each task has a priority which is incremented if it contended + * for the token in an interval less than its previous attempt. + * If the token is acquired, that task's priority is boosted to prevent + * the token from bouncing around too often and to let the task make + * some progress in its execution. */ + #include #include #include #include static DEFINE_SPINLOCK(swap_token_lock); -static unsigned long swap_token_timeout; -static unsigned long swap_token_check; -struct mm_struct * swap_token_mm = &init_mm; - -#define SWAP_TOKEN_CHECK_INTERVAL (HZ * 2) -#define SWAP_TOKEN_TIMEOUT (300 * HZ) -/* - * Currently disabled; Needs further code to work at HZ * 300. - */ -unsigned long swap_token_default_timeout = SWAP_TOKEN_TIMEOUT; - -/* - * Take the token away if the process had no page faults - * in the last interval, or if it has held the token for - * too long. - */ -#define SWAP_TOKEN_ENOUGH_RSS 1 -#define SWAP_TOKEN_TIMED_OUT 2 -static int should_release_swap_token(struct mm_struct *mm) -{ - int ret = 0; - if (!mm->recent_pagein) - ret = SWAP_TOKEN_ENOUGH_RSS; - else if (time_after(jiffies, swap_token_timeout)) - ret = SWAP_TOKEN_TIMED_OUT; - mm->recent_pagein = 0; - return ret; -} +struct mm_struct *swap_token_mm; +unsigned int global_faults; -/* - * Try to grab the swapout protection token. We only try to - * grab it once every TOKEN_CHECK_INTERVAL, both to prevent - * SMP lock contention and to check that the process that held - * the token before is no longer thrashing. - */ void grab_swap_token(void) { - struct mm_struct *mm; - int reason; + int current_interval; - /* We have the token. Let others know we still need it. */ - if (has_swap_token(current->mm)) { - current->mm->recent_pagein = 1; - if (unlikely(!swap_token_default_timeout)) - disable_swap_token(); - return; - } - - if (time_after(jiffies, swap_token_check)) { + global_faults++; - if (!swap_token_default_timeout) { - swap_token_check = jiffies + SWAP_TOKEN_CHECK_INTERVAL; - return; - } - - /* ... or if we recently held the token. */ - if (time_before(jiffies, current->mm->swap_token_time)) - return; + current_interval = global_faults - current->mm->faultstamp; - if (!spin_trylock(&swap_token_lock)) - return; + if (!spin_trylock(&swap_token_lock)) + return; - swap_token_check = jiffies + SWAP_TOKEN_CHECK_INTERVAL; + /* First come first served */ + if (swap_token_mm == NULL) { + current->mm->token_priority = current->mm->token_priority + 2; + swap_token_mm = current->mm; + goto out; + } - mm = swap_token_mm; - if ((reason = should_release_swap_token(mm))) { - unsigned long eligible = jiffies; - if (reason == SWAP_TOKEN_TIMED_OUT) { - eligible += swap_token_default_timeout; - } - mm->swap_token_time = eligible; - swap_token_timeout = jiffies + swap_token_default_timeout; + if (current->mm != swap_token_mm) { + if (current_interval < current->mm->last_interval) + current->mm->token_priority++; + else { + current->mm->token_priority--; + if (unlikely(current->mm->token_priority < 0)) + current->mm->token_priority = 0; + } + /* Check if we deserve the token */ + if (current->mm->token_priority > + swap_token_mm->token_priority) { + current->mm->token_priority += 2; swap_token_mm = current->mm; } - spin_unlock(&swap_token_lock); + } else { + /* Token holder came in again! */ + current->mm->token_priority += 2; } - return; + +out: + current->mm->faultstamp = global_faults; + current->mm->last_interval = current_interval; + spin_unlock(&swap_token_lock); +return; } /* Called on process exit. */ void __put_swap_token(struct mm_struct *mm) { spin_lock(&swap_token_lock); - if (likely(mm == swap_token_mm)) { - mm->swap_token_time = jiffies + SWAP_TOKEN_CHECK_INTERVAL; - swap_token_mm = &init_mm; - swap_token_check = jiffies; - } + if (likely(mm == swap_token_mm)) + swap_token_mm = NULL; spin_unlock(&swap_token_lock); } -- cgit v1.2.3 From cc102509074bba0316f2b5deebd7ef4447da295e Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 6 Dec 2006 20:32:00 -0800 Subject: [PATCH] mm: add arch_alloc_page Add an arch_alloc_page to match arch_free_page. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 3 +++ mm/page_alloc.c | 2 ++ 2 files changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index bf2b6bc3f6fd..00c314aedab7 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -116,6 +116,9 @@ static inline enum zone_type gfp_zone(gfp_t flags) #ifndef HAVE_ARCH_FREE_PAGE static inline void arch_free_page(struct page *page, int order) { } #endif +#ifndef HAVE_ARCH_ALLOC_PAGE +static inline void arch_alloc_page(struct page *page, int order) { } +#endif extern struct page * FASTCALL(__alloc_pages(gfp_t, unsigned int, struct zonelist *)); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 230771d3c6b6..cd47e8f7bd5b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -598,6 +598,8 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags) 1 << PG_checked | 1 << PG_mappedtodisk); set_page_private(page, 0); set_page_refcounted(page); + + arch_alloc_page(page, order); kernel_map_pages(page, 1 << order, 1); if (gfp_flags & __GFP_ZERO) -- cgit v1.2.3 From 39dde65c9940c97fcd178a3d2b1c57ed8b7b68aa Mon Sep 17 00:00:00 2001 From: Chen, Kenneth W Date: Wed, 6 Dec 2006 20:32:03 -0800 Subject: [PATCH] shared page table for hugetlb page Following up with the work on shared page table done by Dave McCracken. This set of patch target shared page table for hugetlb memory only. The shared page table is particular useful in the situation of large number of independent processes sharing large shared memory segments. In the normal page case, the amount of memory saved from process' page table is quite significant. For hugetlb, the saving on page table memory is not the primary objective (as hugetlb itself already cuts down page table overhead significantly), instead, the purpose of using shared page table on hugetlb is to allow faster TLB refill and smaller cache pollution upon TLB miss. With PT sharing, pte entries are shared among hundreds of processes, the cache consumption used by all the page table is smaller and in return, application gets much higher cache hit ratio. One other effect is that cache hit ratio with hardware page walker hitting on pte in cache will be higher and this helps to reduce tlb miss latency. These two effects contribute to higher application performance. Signed-off-by: Ken Chen Acked-by: Hugh Dickins Cc: Dave McCracken Cc: William Lee Irwin III Cc: "Luck, Tony" Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: David Gibson Cc: Adam Litke Cc: Paul Mundt Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/hugetlbpage.c | 112 +++++++++++++++++++++++++++++++++++++++++- arch/ia64/mm/hugetlbpage.c | 5 ++ arch/powerpc/mm/hugetlbpage.c | 5 ++ arch/sh/mm/hugetlbpage.c | 5 ++ arch/sh64/mm/hugetlbpage.c | 5 ++ arch/sparc64/mm/hugetlbpage.c | 5 ++ include/linux/hugetlb.h | 1 + mm/hugetlb.c | 7 +++ 8 files changed, 144 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index 1719a8141f81..34728e4afe48 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -17,6 +17,113 @@ #include #include +static unsigned long page_table_shareable(struct vm_area_struct *svma, + struct vm_area_struct *vma, + unsigned long addr, pgoff_t idx) +{ + unsigned long saddr = ((idx - svma->vm_pgoff) << PAGE_SHIFT) + + svma->vm_start; + unsigned long sbase = saddr & PUD_MASK; + unsigned long s_end = sbase + PUD_SIZE; + + /* + * match the virtual addresses, permission and the alignment of the + * page table page. + */ + if (pmd_index(addr) != pmd_index(saddr) || + vma->vm_flags != svma->vm_flags || + sbase < svma->vm_start || svma->vm_end < s_end) + return 0; + + return saddr; +} + +static int vma_shareable(struct vm_area_struct *vma, unsigned long addr) +{ + unsigned long base = addr & PUD_MASK; + unsigned long end = base + PUD_SIZE; + + /* + * check on proper vm_flags and page table alignment + */ + if (vma->vm_flags & VM_MAYSHARE && + vma->vm_start <= base && end <= vma->vm_end) + return 1; + return 0; +} + +/* + * search for a shareable pmd page for hugetlb. + */ +static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) +{ + struct vm_area_struct *vma = find_vma(mm, addr); + struct address_space *mapping = vma->vm_file->f_mapping; + pgoff_t idx = ((addr - vma->vm_start) >> PAGE_SHIFT) + + vma->vm_pgoff; + struct prio_tree_iter iter; + struct vm_area_struct *svma; + unsigned long saddr; + pte_t *spte = NULL; + + if (!vma_shareable(vma, addr)) + return; + + spin_lock(&mapping->i_mmap_lock); + vma_prio_tree_foreach(svma, &iter, &mapping->i_mmap, idx, idx) { + if (svma == vma) + continue; + + saddr = page_table_shareable(svma, vma, addr, idx); + if (saddr) { + spte = huge_pte_offset(svma->vm_mm, saddr); + if (spte) { + get_page(virt_to_page(spte)); + break; + } + } + } + + if (!spte) + goto out; + + spin_lock(&mm->page_table_lock); + if (pud_none(*pud)) + pud_populate(mm, pud, (unsigned long) spte & PAGE_MASK); + else + put_page(virt_to_page(spte)); + spin_unlock(&mm->page_table_lock); +out: + spin_unlock(&mapping->i_mmap_lock); +} + +/* + * unmap huge page backed by shared pte. + * + * Hugetlb pte page is ref counted at the time of mapping. If pte is shared + * indicated by page_count > 1, unmap is achieved by clearing pud and + * decrementing the ref count. If count == 1, the pte page is not shared. + * + * called with vma->vm_mm->page_table_lock held. + * + * returns: 1 successfully unmapped a shared pte page + * 0 the underlying pte page is not shared, or it is the last user + */ +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + pgd_t *pgd = pgd_offset(mm, *addr); + pud_t *pud = pud_offset(pgd, *addr); + + BUG_ON(page_count(virt_to_page(ptep)) == 0); + if (page_count(virt_to_page(ptep)) == 1) + return 0; + + pud_clear(pud); + put_page(virt_to_page(ptep)); + *addr = ALIGN(*addr, HPAGE_SIZE * PTRS_PER_PTE) - HPAGE_SIZE; + return 1; +} + pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; @@ -25,8 +132,11 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) pgd = pgd_offset(mm, addr); pud = pud_alloc(mm, pgd, addr); - if (pud) + if (pud) { + if (pud_none(*pud)) + huge_pmd_share(mm, addr, pud); pte = (pte_t *) pmd_alloc(mm, pud, addr); + } BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte)); return pte; diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index f3a9585e98a8..0c7e94edc20e 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -64,6 +64,11 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 506d89768d45..424a8f57e155 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -146,6 +146,11 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) return hugepte_offset(hpdp, addr); } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp) { pte_t *hugepte = hugepd_page(*hpdp); diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 329059d6b54a..cf2c2ee35a37 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -63,6 +63,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c index 187cf01750b8..4b455f611146 100644 --- a/arch/sh64/mm/hugetlbpage.c +++ b/arch/sh64/mm/hugetlbpage.c @@ -53,6 +53,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index 53b9b1f528e5..33fd0b265e70 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -235,6 +235,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return pte; } +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t entry) { diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index ace64e57e17f..a60995afe334 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -35,6 +35,7 @@ extern int sysctl_hugetlb_shm_group; pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr); pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr); +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep); struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write); struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, diff --git a/mm/hugetlb.c b/mm/hugetlb.c index f7355bf2f285..9244971b6791 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -386,6 +386,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; + pte = huge_ptep_get_and_clear(mm, address, ptep); if (pte_none(pte)) continue; @@ -658,11 +661,14 @@ void hugetlb_change_protection(struct vm_area_struct *vma, BUG_ON(address >= end); flush_cache_range(vma, address, end); + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); spin_lock(&mm->page_table_lock); for (; address < end; address += HPAGE_SIZE) { ptep = huge_pte_offset(mm, address); if (!ptep) continue; + if (huge_pmd_unshare(mm, &address, ptep)) + continue; if (!pte_none(*ptep)) { pte = huge_ptep_get_and_clear(mm, address, ptep); pte = pte_mkhuge(pte_modify(pte, newprot)); @@ -671,6 +677,7 @@ void hugetlb_change_protection(struct vm_area_struct *vma, } } spin_unlock(&mm->page_table_lock); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); flush_tlb_range(vma, start, end); } -- cgit v1.2.3 From a866374aecc90c7d90619727ccd851ac096b2fc7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:32:20 -0800 Subject: [PATCH] mm: pagefault_{disable,enable}() Introduce pagefault_{disable,enable}() and use these where previously we did manual preempt increments/decrements to make the pagefault handler do the atomic thing. Currently they still rely on the increased preempt count, but do not rely on the disabled preemption, this might go away in the future. (NOTE: the extra barrier() in pagefault_disable might fix some holes on machines which have too many registers for their own good) [heiko.carstens@de.ibm.com: s390 fix] Signed-off-by: Peter Zijlstra Acked-by: Nick Piggin Cc: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/futex.c | 4 ++-- arch/i386/mm/highmem.c | 10 ++++------ arch/mips/mm/highmem.c | 10 ++++------ arch/s390/lib/uaccess_std.c | 6 +++--- arch/sparc/mm/highmem.c | 8 +++----- include/asm-frv/highmem.h | 5 ++--- include/asm-generic/futex.h | 4 ++-- include/asm-i386/futex.h | 4 ++-- include/asm-ia64/futex.h | 4 ++-- include/asm-mips/futex.h | 4 ++-- include/asm-parisc/futex.h | 4 ++-- include/asm-powerpc/futex.h | 4 ++-- include/asm-ppc/highmem.h | 8 +++----- include/asm-sparc64/futex.h | 4 ++-- include/asm-x86_64/futex.h | 4 ++-- include/linux/uaccess.h | 39 +++++++++++++++++++++++++++++++++++++-- kernel/futex.c | 28 ++++++++++++++-------------- 17 files changed, 88 insertions(+), 62 deletions(-) (limited to 'include/linux') diff --git a/arch/frv/kernel/futex.c b/arch/frv/kernel/futex.c index eae874a970c6..53dc5ed1ebda 100644 --- a/arch/frv/kernel/futex.c +++ b/arch/frv/kernel/futex.c @@ -200,7 +200,7 @@ int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -223,7 +223,7 @@ int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) break; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c index f9f647cdbc7b..178bbfe6cbac 100644 --- a/arch/i386/mm/highmem.c +++ b/arch/i386/mm/highmem.c @@ -32,7 +32,7 @@ void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -52,8 +52,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) #ifdef CONFIG_DEBUG_HIGHMEM if (vaddr >= PAGE_OFFSET && vaddr < (unsigned long)high_memory) { - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -68,8 +67,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) */ kpte_clear_flush(kmap_pte-idx, vaddr); - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } /* This is the same as kmap_atomic() but can map memory that doesn't @@ -80,7 +78,7 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) enum fixed_addresses idx; unsigned long vaddr; - inc_preempt_count(); + pagefault_disable(); idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); diff --git a/arch/mips/mm/highmem.c b/arch/mips/mm/highmem.c index 99ebf3ccc222..675502ada5a2 100644 --- a/arch/mips/mm/highmem.c +++ b/arch/mips/mm/highmem.c @@ -39,7 +39,7 @@ void *__kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -62,8 +62,7 @@ void __kunmap_atomic(void *kvaddr, enum km_type type) enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < FIXADDR_START) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -78,8 +77,7 @@ void __kunmap_atomic(void *kvaddr, enum km_type type) local_flush_tlb_one(vaddr); #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } #ifndef CONFIG_LIMITED_DMA @@ -92,7 +90,7 @@ void *kmap_atomic_pfn(unsigned long pfn, enum km_type type) enum fixed_addresses idx; unsigned long vaddr; - inc_preempt_count(); + pagefault_disable(); idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); diff --git a/arch/s390/lib/uaccess_std.c b/arch/s390/lib/uaccess_std.c index 2d549ed2e113..bbaca66fa293 100644 --- a/arch/s390/lib/uaccess_std.c +++ b/arch/s390/lib/uaccess_std.c @@ -11,7 +11,7 @@ #include #include -#include +#include #include #ifndef __s390x__ @@ -258,7 +258,7 @@ int futex_atomic_op(int op, int __user *uaddr, int oparg, int *old) { int oldval = 0, newval, ret; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -284,7 +284,7 @@ int futex_atomic_op(int op, int __user *uaddr, int oparg, int *old) default: ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); *old = oldval; return ret; } diff --git a/arch/sparc/mm/highmem.c b/arch/sparc/mm/highmem.c index 4d8ed9c65182..01fc6c254292 100644 --- a/arch/sparc/mm/highmem.c +++ b/arch/sparc/mm/highmem.c @@ -35,7 +35,7 @@ void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -70,8 +70,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) unsigned long idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < FIXADDR_START) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -97,8 +96,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type) #endif #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } /* We may be fed a pagetable here by ptep_to_xxx and others. */ diff --git a/include/asm-frv/highmem.h b/include/asm-frv/highmem.h index 0f390f41f816..ff4d6cdeb152 100644 --- a/include/asm-frv/highmem.h +++ b/include/asm-frv/highmem.h @@ -115,7 +115,7 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) { unsigned long paddr; - inc_preempt_count(); + pagefault_disable(); paddr = page_to_phys(page); switch (type) { @@ -170,8 +170,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) default: BUG(); } - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } #endif /* !__ASSEMBLY__ */ diff --git a/include/asm-generic/futex.h b/include/asm-generic/futex.h index df893c160318..f422df0956a2 100644 --- a/include/asm-generic/futex.h +++ b/include/asm-generic/futex.h @@ -21,7 +21,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -33,7 +33,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-i386/futex.h b/include/asm-i386/futex.h index 946d97cfea23..438ef0ec7101 100644 --- a/include/asm-i386/futex.h +++ b/include/asm-i386/futex.h @@ -56,7 +56,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); if (op == FUTEX_OP_SET) __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); @@ -88,7 +88,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) } } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-ia64/futex.h b/include/asm-ia64/futex.h index 07d77f3a8cbe..8a98a2654139 100644 --- a/include/asm-ia64/futex.h +++ b/include/asm-ia64/futex.h @@ -59,7 +59,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -83,7 +83,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-mips/futex.h b/include/asm-mips/futex.h index 927a216bd530..47e5679c2353 100644 --- a/include/asm-mips/futex.h +++ b/include/asm-mips/futex.h @@ -88,7 +88,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -115,7 +115,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-parisc/futex.h b/include/asm-parisc/futex.h index d84bbb283fd1..dbee6e60aa81 100644 --- a/include/asm-parisc/futex.h +++ b/include/asm-parisc/futex.h @@ -21,7 +21,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -33,7 +33,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-powerpc/futex.h b/include/asm-powerpc/futex.h index 936422e54891..3f3673fd3ff3 100644 --- a/include/asm-powerpc/futex.h +++ b/include/asm-powerpc/futex.h @@ -43,7 +43,7 @@ static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -65,7 +65,7 @@ static inline int futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-ppc/highmem.h b/include/asm-ppc/highmem.h index 1d2c4ef81c22..f7b21ee302b4 100644 --- a/include/asm-ppc/highmem.h +++ b/include/asm-ppc/highmem.h @@ -79,7 +79,7 @@ static inline void *kmap_atomic(struct page *page, enum km_type type) unsigned long vaddr; /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */ - inc_preempt_count(); + pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -101,8 +101,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) unsigned int idx = type + KM_TYPE_NR*smp_processor_id(); if (vaddr < KMAP_FIX_BEGIN) { // FIXME - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); return; } @@ -115,8 +114,7 @@ static inline void kunmap_atomic(void *kvaddr, enum km_type type) pte_clear(&init_mm, vaddr, kmap_pte+idx); flush_tlb_page(NULL, vaddr); #endif - dec_preempt_count(); - preempt_check_resched(); + pagefault_enable(); } static inline struct page *kmap_atomic_to_page(void *ptr) diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h index 7392fc4a954e..876312fe82cc 100644 --- a/include/asm-sparc64/futex.h +++ b/include/asm-sparc64/futex.h @@ -45,7 +45,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -67,7 +67,7 @@ static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/asm-x86_64/futex.h b/include/asm-x86_64/futex.h index 9804bf07b092..5cdfb08013c3 100644 --- a/include/asm-x86_64/futex.h +++ b/include/asm-x86_64/futex.h @@ -55,7 +55,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - inc_preempt_count(); + pagefault_disable(); switch (op) { case FUTEX_OP_SET: @@ -78,7 +78,7 @@ futex_atomic_op_inuser (int encoded_op, int __user *uaddr) ret = -ENOSYS; } - dec_preempt_count(); + pagefault_enable(); if (!ret) { switch (cmp) { diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index a48d7f11c7be..67918c22339c 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -1,8 +1,43 @@ #ifndef __LINUX_UACCESS_H__ #define __LINUX_UACCESS_H__ +#include #include +/* + * These routines enable/disable the pagefault handler in that + * it will not take any locks and go straight to the fixup table. + * + * They have great resemblance to the preempt_disable/enable calls + * and in fact they are identical; this is because currently there is + * no other way to make the pagefault handlers do this. So we do + * disable preemption but we don't necessarily care about that. + */ +static inline void pagefault_disable(void) +{ + inc_preempt_count(); + /* + * make sure to have issued the store before a pagefault + * can hit. + */ + barrier(); +} + +static inline void pagefault_enable(void) +{ + /* + * make sure to issue those last loads/stores before enabling + * the pagefault handler again. + */ + barrier(); + dec_preempt_count(); + /* + * make sure we do.. + */ + barrier(); + preempt_check_resched(); +} + #ifndef ARCH_HAS_NOCACHE_UACCESS static inline unsigned long __copy_from_user_inatomic_nocache(void *to, @@ -35,9 +70,9 @@ static inline unsigned long __copy_from_user_nocache(void *to, ({ \ long ret; \ \ - inc_preempt_count(); \ + pagefault_disable(); \ ret = __get_user(retval, addr); \ - dec_preempt_count(); \ + pagefault_enable(); \ ret; \ }) diff --git a/kernel/futex.c b/kernel/futex.c index 93ef30ba209f..af7b81cbde30 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -282,9 +282,9 @@ static inline int get_futex_value_locked(u32 *dest, u32 __user *from) { int ret; - inc_preempt_count(); + pagefault_disable(); ret = __copy_from_user_inatomic(dest, from, sizeof(u32)); - dec_preempt_count(); + pagefault_enable(); return ret ? -EFAULT : 0; } @@ -585,9 +585,9 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) if (!(uval & FUTEX_OWNER_DIED)) { newval = FUTEX_WAITERS | new_owner->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (curval == -EFAULT) return -EFAULT; if (curval != uval) @@ -618,9 +618,9 @@ static int unlock_futex_pi(u32 __user *uaddr, u32 uval) * There is no waiter, so we unlock the futex. The owner died * bit has not to be preserved here. We are the owner: */ - inc_preempt_count(); + pagefault_disable(); oldval = futex_atomic_cmpxchg_inatomic(uaddr, uval, 0); - dec_preempt_count(); + pagefault_enable(); if (oldval == -EFAULT) return oldval; @@ -1158,9 +1158,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, */ newval = current->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, 0, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1183,9 +1183,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, uval = curval; newval = uval | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1215,10 +1215,10 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, newval = current->pid | FUTEX_OWNER_DIED | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1390,9 +1390,9 @@ retry_locked: * anyone else up: */ if (!(uval & FUTEX_OWNER_DIED)) { - inc_preempt_count(); + pagefault_disable(); uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); - dec_preempt_count(); + pagefault_enable(); } if (unlikely(uval == -EFAULT)) -- cgit v1.2.3 From ad76fb6b5a5183255279e0ab5260715481770678 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:32:21 -0800 Subject: [PATCH] mm: k{,um}map_atomic() vs in_atomic() Make kmap_atomic/kunmap_atomic denote a pagefault disabled scope. All non trivial implementations already do this anyway. Signed-off-by: Peter Zijlstra Acked-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-mips/highmem.h | 10 ++++++++-- include/linux/highmem.h | 8 +++++--- 2 files changed, 13 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/asm-mips/highmem.h b/include/asm-mips/highmem.h index c976bfaaba83..f8c8182f7f2e 100644 --- a/include/asm-mips/highmem.h +++ b/include/asm-mips/highmem.h @@ -21,6 +21,7 @@ #include #include +#include #include /* undef for production */ @@ -70,11 +71,16 @@ static inline void *kmap(struct page *page) static inline void *kmap_atomic(struct page *page, enum km_type type) { + pagefault_disable(); return page_address(page); } -static inline void kunmap_atomic(void *kvaddr, enum km_type type) { } -#define kmap_atomic_pfn(pfn, idx) page_address(pfn_to_page(pfn)) +static inline void kunmap_atomic(void *kvaddr, enum km_type type) +{ + pagefault_enable(); +} + +#define kmap_atomic_pfn(pfn, idx) kmap_atomic(pfn_to_page(pfn), (idx)) #define kmap_atomic_to_page(ptr) virt_to_page(ptr) diff --git a/include/linux/highmem.h b/include/linux/highmem.h index fd7d12daa94f..3d8768b619e9 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -3,6 +3,7 @@ #include #include +#include #include @@ -41,9 +42,10 @@ static inline void *kmap(struct page *page) #define kunmap(page) do { (void) (page); } while (0) -#define kmap_atomic(page, idx) page_address(page) -#define kunmap_atomic(addr, idx) do { } while (0) -#define kmap_atomic_pfn(pfn, idx) page_address(pfn_to_page(pfn)) +#define kmap_atomic(page, idx) \ + ({ pagefault_disable(); page_address(page); }) +#define kunmap_atomic(addr, idx) do { pagefault_enable(); } while (0) +#define kmap_atomic_pfn(pfn, idx) kmap_atomic(pfn_to_page(pfn), (idx)) #define kmap_atomic_to_page(ptr) virt_to_page(ptr) #endif -- cgit v1.2.3 From 8b98c1699eba23cfd2e8b366625c50ff5fd1415b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 6 Dec 2006 20:32:30 -0800 Subject: [PATCH] leak tracking for kmalloc_node We have variants of kmalloc and kmem_cache_alloc that leave leak tracking to the caller. This is used for subsystem-specific allocators like skb_alloc. To make skb_alloc node-aware we need similar routines for the node-aware slab allocator, which this patch adds. Note that the code is rather ugly, but it mirrors the non-node-aware code 1:1: [akpm@osdl.org: add module export] Signed-off-by: Christoph Hellwig Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 23 ++++++++++++++++++++++ mm/slab.c | 55 +++++++++++++++++++++++++++++++++++++++------------- 2 files changed, 65 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index c4947b8a2c03..66c4640d3656 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -236,7 +236,25 @@ found: } return __kmalloc_node(size, flags, node); } + +/* + * kmalloc_node_track_caller is a special version of kmalloc_node that + * records the calling function of the routine calling it for slab leak + * tracking instead of just the calling function (confusing, eh?). + * It's useful when the call to kmalloc_node comes from a widely-used + * standard allocator where we care about the real place the memory + * allocation request comes from. + */ +#ifndef CONFIG_DEBUG_SLAB +#define kmalloc_node_track_caller(size, flags, node) \ + __kmalloc_node(size, flags, node) #else +extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, void *); +#define kmalloc_node_track_caller(size, flags, node) \ + __kmalloc_node_track_caller(size, flags, node, \ + __builtin_return_address(0)) +#endif +#else /* CONFIG_NUMA */ static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, gfp_t flags, int node) { return kmem_cache_alloc(cachep, flags); @@ -245,6 +263,9 @@ static inline void *kmalloc_node(size_t size, gfp_t flags, int node) { return kmalloc(size, flags); } + +#define kmalloc_node_track_caller(size, flags, node) \ + kmalloc_track_caller(size, flags) #endif extern int FASTCALL(kmem_cache_reap(int)); @@ -283,6 +304,8 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) #define kzalloc(s, f) __kzalloc(s, f) #define kmalloc_track_caller kmalloc +#define kmalloc_node_track_caller kmalloc_node + #endif /* CONFIG_SLOB */ /* System wide caches */ diff --git a/mm/slab.c b/mm/slab.c index bfd654c0ef41..8f3f61cacb5c 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1015,7 +1015,7 @@ static inline void *alternate_node_alloc(struct kmem_cache *cachep, return NULL; } -static inline void *__cache_alloc_node(struct kmem_cache *cachep, +static inline void *____cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid) { return NULL; @@ -1023,7 +1023,7 @@ static inline void *__cache_alloc_node(struct kmem_cache *cachep, #else /* CONFIG_NUMA */ -static void *__cache_alloc_node(struct kmem_cache *, gfp_t, int); +static void *____cache_alloc_node(struct kmem_cache *, gfp_t, int); static void *alternate_node_alloc(struct kmem_cache *, gfp_t); static struct array_cache **alloc_alien_cache(int node, int limit) @@ -3130,10 +3130,10 @@ static __always_inline void *__cache_alloc(struct kmem_cache *cachep, objp = ____cache_alloc(cachep, flags); /* * We may just have run out of memory on the local node. - * __cache_alloc_node() knows how to locate memory on other nodes + * ____cache_alloc_node() knows how to locate memory on other nodes */ if (NUMA_BUILD && !objp) - objp = __cache_alloc_node(cachep, flags, numa_node_id()); + objp = ____cache_alloc_node(cachep, flags, numa_node_id()); local_irq_restore(save_flags); objp = cache_alloc_debugcheck_after(cachep, flags, objp, caller); @@ -3160,7 +3160,7 @@ static void *alternate_node_alloc(struct kmem_cache *cachep, gfp_t flags) else if (current->mempolicy) nid_alloc = slab_node(current->mempolicy); if (nid_alloc != nid_here) - return __cache_alloc_node(cachep, flags, nid_alloc); + return ____cache_alloc_node(cachep, flags, nid_alloc); return NULL; } @@ -3183,7 +3183,7 @@ void *fallback_alloc(struct kmem_cache *cache, gfp_t flags) if (zone_idx(*z) <= ZONE_NORMAL && cpuset_zone_allowed(*z, flags) && cache->nodelists[nid]) - obj = __cache_alloc_node(cache, + obj = ____cache_alloc_node(cache, flags | __GFP_THISNODE, nid); } return obj; @@ -3192,7 +3192,7 @@ void *fallback_alloc(struct kmem_cache *cache, gfp_t flags) /* * A interface to enable slab creation on nodeid */ -static void *__cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, +static void *____cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid) { struct list_head *entry; @@ -3465,7 +3465,9 @@ out: * New and improved: it will now make sure that the object gets * put on the correct node list so that there is no false sharing. */ -void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid) +static __always_inline void * +__cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, + int nodeid, void *caller) { unsigned long save_flags; void *ptr; @@ -3477,17 +3479,23 @@ void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid) !cachep->nodelists[nodeid]) ptr = ____cache_alloc(cachep, flags); else - ptr = __cache_alloc_node(cachep, flags, nodeid); + ptr = ____cache_alloc_node(cachep, flags, nodeid); local_irq_restore(save_flags); - ptr = cache_alloc_debugcheck_after(cachep, flags, ptr, - __builtin_return_address(0)); + ptr = cache_alloc_debugcheck_after(cachep, flags, ptr, caller); return ptr; } + +void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid) +{ + return __cache_alloc_node(cachep, flags, nodeid, + __builtin_return_address(0)); +} EXPORT_SYMBOL(kmem_cache_alloc_node); -void *__kmalloc_node(size_t size, gfp_t flags, int node) +static __always_inline void * +__do_kmalloc_node(size_t size, gfp_t flags, int node, void *caller) { struct kmem_cache *cachep; @@ -3496,8 +3504,29 @@ void *__kmalloc_node(size_t size, gfp_t flags, int node) return NULL; return kmem_cache_alloc_node(cachep, flags, node); } + +#ifdef CONFIG_DEBUG_SLAB +void *__kmalloc_node(size_t size, gfp_t flags, int node) +{ + return __do_kmalloc_node(size, flags, node, + __builtin_return_address(0)); +} EXPORT_SYMBOL(__kmalloc_node); -#endif + +void *__kmalloc_node_track_caller(size_t size, gfp_t flags, + int node, void *caller) +{ + return __do_kmalloc_node(size, flags, node, caller); +} +EXPORT_SYMBOL(__kmalloc_node_track_caller); +#else +void *__kmalloc_node(size_t size, gfp_t flags, int node) +{ + return __do_kmalloc_node(size, flags, node, NULL); +} +EXPORT_SYMBOL(__kmalloc_node); +#endif /* CONFIG_DEBUG_SLAB */ +#endif /* CONFIG_NUMA */ /** * __do_kmalloc - allocate memory -- cgit v1.2.3 From 873481367edb18a7d0d7e5a285e6728c16bb44a9 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 6 Dec 2006 20:32:33 -0800 Subject: [PATCH] add numa node information to struct device For node-aware skb allocations we need information about the node in struct net_device or struct device. Davem suggested to put it into struct device which this patch does. In particular: - struct device gets a new int numa_node member if CONFIG_NUMA is set - there are two new helpers, dev_to_node and set_dev_node to transparently deal with the non-numa case - for pci devices the node-info is set to the value we get from pcibus_to_node. Note that for some architectures pcibus_to_node doesn't work yet at the time we call it currently. This is harmless and will just mean skb allocations aren't node-local on this architectures until the implementation of pcibus_to_node on these architectures have been updated (There are patches for x86 and x86_64 floating around) [akpm@osdl.org: cleanup] Signed-off-by: Christoph Hellwig Cc: Christoph Lameter Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/base/core.c | 1 + drivers/pci/probe.c | 1 + include/linux/device.h | 22 ++++++++++++++++++++++ 3 files changed, 24 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/core.c b/drivers/base/core.c index e4b530ef757d..67b79a7592a9 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -386,6 +386,7 @@ void device_initialize(struct device *dev) INIT_LIST_HEAD(&dev->node); init_MUTEX(&dev->sem); device_init_wakeup(dev, 0); + set_dev_node(dev, -1); } #ifdef CONFIG_SYSFS_DEPRECATED diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 0eeac60042b3..6a3c1e728900 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -873,6 +873,7 @@ void __devinit pci_device_add(struct pci_dev *dev, struct pci_bus *bus) dev->dev.release = pci_release_dev; pci_dev_get(dev); + set_dev_node(&dev->dev, pcibus_to_node(bus)); dev->dev.dma_mask = &dev->dma_mask; dev->dev.coherent_dma_mask = 0xffffffffull; diff --git a/include/linux/device.h b/include/linux/device.h index 583a341e016c..49ab53ce92dc 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -371,6 +371,9 @@ struct device { core doesn't touch it */ struct dev_pm_info power; +#ifdef CONFIG_NUMA + int numa_node; /* NUMA node this device is close to */ +#endif u64 *dma_mask; /* dma mask (if dma'able device) */ u64 coherent_dma_mask;/* Like dma_mask, but for alloc_coherent mappings as @@ -394,6 +397,25 @@ struct device { void (*release)(struct device * dev); }; +#ifdef CONFIG_NUMA +static inline int dev_to_node(struct device *dev) +{ + return dev->numa_node; +} +static inline void set_dev_node(struct device *dev, int node) +{ + dev->numa_node = node; +} +#else +static inline int dev_to_node(struct device *dev) +{ + return -1; +} +static inline void set_dev_node(struct device *dev, int node) +{ +} +#endif + static inline void * dev_get_drvdata (struct device *dev) { -- cgit v1.2.3 From b30973f877fea1a3fb84e05599890fcc082a88e5 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 6 Dec 2006 20:32:36 -0800 Subject: [PATCH] node-aware skb allocation Node-aware allocation of skbs for the receive path. Details: - __alloc_skb gets a new node argument and cals the node-aware slab functions with it. - netdev_alloc_skb passed the node number it gets from dev_to_node to it, everyone else passes -1 (any node) Signed-off-by: Christoph Hellwig Cc: Christoph Lameter Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/skbuff.h | 6 +++--- net/core/skbuff.c | 12 +++++++----- 2 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a05a5f7c0b73..1d649f3eb006 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -332,17 +332,17 @@ struct sk_buff { extern void kfree_skb(struct sk_buff *skb); extern void __kfree_skb(struct sk_buff *skb); extern struct sk_buff *__alloc_skb(unsigned int size, - gfp_t priority, int fclone); + gfp_t priority, int fclone, int node); static inline struct sk_buff *alloc_skb(unsigned int size, gfp_t priority) { - return __alloc_skb(size, priority, 0); + return __alloc_skb(size, priority, 0, -1); } static inline struct sk_buff *alloc_skb_fclone(unsigned int size, gfp_t priority) { - return __alloc_skb(size, priority, 1); + return __alloc_skb(size, priority, 1, -1); } extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8e1c385e5ba9..7217fb8928f2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -132,6 +132,7 @@ EXPORT_SYMBOL(skb_truesize_bug); * @gfp_mask: allocation mask * @fclone: allocate from fclone cache instead of head cache * and allocate a cloned (child) skb + * @node: numa node to allocate memory on * * Allocate a new &sk_buff. The returned buffer has no headroom and a * tail room of size bytes. The object has a reference count of one. @@ -141,7 +142,7 @@ EXPORT_SYMBOL(skb_truesize_bug); * %GFP_ATOMIC. */ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, - int fclone) + int fclone, int node) { kmem_cache_t *cache; struct skb_shared_info *shinfo; @@ -151,14 +152,14 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, cache = fclone ? skbuff_fclone_cache : skbuff_head_cache; /* Get the HEAD */ - skb = kmem_cache_alloc(cache, gfp_mask & ~__GFP_DMA); + skb = kmem_cache_alloc_node(cache, gfp_mask & ~__GFP_DMA, node); if (!skb) goto out; /* Get the DATA. Size must match skb_add_mtu(). */ size = SKB_DATA_ALIGN(size); - data = kmalloc_track_caller(size + sizeof(struct skb_shared_info), - gfp_mask); + data = kmalloc_node_track_caller(size + sizeof(struct skb_shared_info), + gfp_mask, node); if (!data) goto nodata; @@ -267,9 +268,10 @@ nodata: struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length, gfp_t gfp_mask) { + int node = dev->class_dev.dev ? dev_to_node(dev->class_dev.dev) : -1; struct sk_buff *skb; - skb = alloc_skb(length + NET_SKB_PAD, gfp_mask); + skb = __alloc_skb(length + NET_SKB_PAD, gfp_mask, 0, node); if (likely(skb)) { skb_reserve(skb, NET_SKB_PAD); skb->dev = dev; -- cgit v1.2.3 From 54cc211ce3fc73a9d21c6316886db0676beaca95 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:45 -0800 Subject: [PATCH] Remove bio_cachep from slab.h Remove bio_cachep from slab.h - it no longer exists. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index 66c4640d3656..6b7d096f056b 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -315,7 +315,6 @@ extern kmem_cache_t *files_cachep; extern kmem_cache_t *filp_cachep; extern kmem_cache_t *fs_cachep; extern kmem_cache_t *sighand_cachep; -extern kmem_cache_t *bio_cachep; #endif /* __KERNEL__ */ -- cgit v1.2.3 From 298ec1e2ac85cecce3eddd167286359358c44d5d Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:47 -0800 Subject: [PATCH] Move sighand_cachep to include/signal.h Move sighand_cachep definitioni to linux/signal.h The sighand cache is only used in fs/exec.c and kernel/fork.c. It is defined in kernel/fork.c but only used in fs/exec.c. The sighand_cachep is related to signal processing. So add the definition to signal.h. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/signal.h | 2 ++ include/linux/slab.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/signal.h b/include/linux/signal.h index 117135e33d67..14749056dd63 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -241,6 +241,8 @@ extern int sigprocmask(int, sigset_t *, sigset_t *); struct pt_regs; extern int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka, struct pt_regs *regs, void *cookie); +extern struct kmem_cache *sighand_cachep; + #endif /* __KERNEL__ */ #endif /* _LINUX_SIGNAL_H */ diff --git a/include/linux/slab.h b/include/linux/slab.h index 6b7d096f056b..467b297d2c6f 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -314,7 +314,6 @@ extern kmem_cache_t *names_cachep; extern kmem_cache_t *files_cachep; extern kmem_cache_t *filp_cachep; extern kmem_cache_t *fs_cachep; -extern kmem_cache_t *sighand_cachep; #endif /* __KERNEL__ */ -- cgit v1.2.3 From c43692e85f306667545b91194c748a6e46c1f8b4 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:48 -0800 Subject: [PATCH] Move vm_area_cachep to include/mm.h vm_area_cachep is used to store vm_area_structs. So move to mm.h. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 2 ++ include/linux/slab.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index ab6e4974f379..840303769c11 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -114,6 +114,8 @@ struct vm_area_struct { #endif }; +extern struct kmem_cache *vm_area_cachep; + /* * This struct defines the per-mm list of VMAs for uClinux. If CONFIG_MMU is * disabled, then there's a single shared list of VMAs maintained by the diff --git a/include/linux/slab.h b/include/linux/slab.h index 467b297d2c6f..a30a4028a92e 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -309,7 +309,6 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) #endif /* CONFIG_SLOB */ /* System wide caches */ -extern kmem_cache_t *vm_area_cachep; extern kmem_cache_t *names_cachep; extern kmem_cache_t *files_cachep; extern kmem_cache_t *filp_cachep; -- cgit v1.2.3 From 5d6538fcf231faccb2ac42f92851d259d00e62f9 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:50 -0800 Subject: [PATCH] Move files_cachep to include/file.h Proper place is in file.h since files_cachep uses are rated to file I/O. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/file.h | 2 ++ include/linux/slab.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/file.h b/include/linux/file.h index 74183e6f7f45..be37423e9920 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -114,4 +114,6 @@ struct files_struct *get_files_struct(struct task_struct *); void FASTCALL(put_files_struct(struct files_struct *fs)); void reset_files_struct(struct task_struct *, struct files_struct *); +extern struct kmem_cache *files_cachep; + #endif /* __LINUX_FILE_H */ diff --git a/include/linux/slab.h b/include/linux/slab.h index a30a4028a92e..dd9efecff410 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -310,7 +310,6 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) /* System wide caches */ extern kmem_cache_t *names_cachep; -extern kmem_cache_t *files_cachep; extern kmem_cache_t *filp_cachep; extern kmem_cache_t *fs_cachep; -- cgit v1.2.3 From 8b7d91eb7f6a3e8f0caaa613937bda5ab7dc7dc2 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:52 -0800 Subject: [PATCH] Move filep_cachep to include/file.h filp_cachep is only used in fs/file_table.c and in fs/dcache.c where it is defined. Move it to related definitions in linux/file.h. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/file.h | 2 ++ include/linux/slab.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/file.h b/include/linux/file.h index be37423e9920..6e77b9177f9e 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -64,6 +64,8 @@ struct files_struct { #define files_fdtable(files) (rcu_dereference((files)->fdt)) +extern struct kmem_cache *filp_cachep; + extern void FASTCALL(__fput(struct file *)); extern void FASTCALL(fput(struct file *)); diff --git a/include/linux/slab.h b/include/linux/slab.h index dd9efecff410..f9202d60d769 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -310,7 +310,6 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) /* System wide caches */ extern kmem_cache_t *names_cachep; -extern kmem_cache_t *filp_cachep; extern kmem_cache_t *fs_cachep; #endif /* __KERNEL__ */ -- cgit v1.2.3 From aa362a83e78d2e9320da588805cf2a0b53356bc3 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:54 -0800 Subject: [PATCH] Move fs_cachep to linux/fs_struct.h fs_cachep is only used in kernel/exit.c and in kernel/fork.c. It is used to store fs_struct items so it should be placed in linux/fs_struct.h Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs_struct.h | 2 ++ include/linux/slab.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h index c623d12a486e..11a36ceddf73 100644 --- a/include/linux/fs_struct.h +++ b/include/linux/fs_struct.h @@ -18,6 +18,8 @@ struct fs_struct { .umask = 0022, \ } +extern struct kmem_cache *fs_cachep; + extern void exit_fs(struct task_struct *); extern void set_fs_altroot(void); extern void set_fs_root(struct fs_struct *, struct vfsmount *, struct dentry *); diff --git a/include/linux/slab.h b/include/linux/slab.h index f9202d60d769..bfc063ee94e8 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -310,7 +310,6 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) /* System wide caches */ extern kmem_cache_t *names_cachep; -extern kmem_cache_t *fs_cachep; #endif /* __KERNEL__ */ -- cgit v1.2.3 From b86c089b83b8ae2bc814db865057768a9ba787b5 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:57 -0800 Subject: [PATCH] Move names_cachep to linux/fs.h The names_cachep is used for getname() and putname(). So lets put it into fs.h near those two definitions. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 2 ++ include/linux/slab.h | 3 --- 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index cac7b1ef9543..a8039c8d8cbb 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1481,6 +1481,8 @@ extern char * getname(const char __user *); extern void __init vfs_caches_init_early(void); extern void __init vfs_caches_init(unsigned long); +extern struct kmem_cache *names_cachep; + #define __getname() kmem_cache_alloc(names_cachep, SLAB_KERNEL) #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) #ifndef CONFIG_AUDITSYSCALL diff --git a/include/linux/slab.h b/include/linux/slab.h index bfc063ee94e8..e67314e4a0a0 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -308,9 +308,6 @@ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) #endif /* CONFIG_SLOB */ -/* System wide caches */ -extern kmem_cache_t *names_cachep; - #endif /* __KERNEL__ */ #endif /* _LINUX_SLAB_H */ -- cgit v1.2.3 From ebe29738f3934ad6a93c8bd76e30aa5d797a269d Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:32:59 -0800 Subject: [PATCH] Remove uses of kmem_cache_t from mm/* and include/linux/slab.h Remove all uses of kmem_cache_t (the most were left in slab.h). The typedef for kmem_cache_t is then only necessary for other kernel subsystems. Add a comment to that effect. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index e67314e4a0a0..b831776b2fc7 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -9,6 +9,7 @@ #if defined(__KERNEL__) +/* kmem_cache_t exists for legacy reasons and is not used by code in mm */ typedef struct kmem_cache kmem_cache_t; #include @@ -57,22 +58,23 @@ typedef struct kmem_cache kmem_cache_t; /* prototypes */ extern void __init kmem_cache_init(void); -extern kmem_cache_t *kmem_cache_create(const char *, size_t, size_t, unsigned long, - void (*)(void *, kmem_cache_t *, unsigned long), - void (*)(void *, kmem_cache_t *, unsigned long)); -extern void kmem_cache_destroy(kmem_cache_t *); -extern int kmem_cache_shrink(kmem_cache_t *); -extern void *kmem_cache_alloc(kmem_cache_t *, gfp_t); +extern struct kmem_cache *kmem_cache_create(const char *, size_t, size_t, + unsigned long, + void (*)(void *, struct kmem_cache *, unsigned long), + void (*)(void *, struct kmem_cache *, unsigned long)); +extern void kmem_cache_destroy(struct kmem_cache *); +extern int kmem_cache_shrink(struct kmem_cache *); +extern void *kmem_cache_alloc(struct kmem_cache *, gfp_t); extern void *kmem_cache_zalloc(struct kmem_cache *, gfp_t); -extern void kmem_cache_free(kmem_cache_t *, void *); -extern unsigned int kmem_cache_size(kmem_cache_t *); -extern const char *kmem_cache_name(kmem_cache_t *); +extern void kmem_cache_free(struct kmem_cache *, void *); +extern unsigned int kmem_cache_size(struct kmem_cache *); +extern const char *kmem_cache_name(struct kmem_cache *); /* Size description struct for general caches. */ struct cache_sizes { - size_t cs_size; - kmem_cache_t *cs_cachep; - kmem_cache_t *cs_dmacachep; + size_t cs_size; + struct kmem_cache *cs_cachep; + struct kmem_cache *cs_dmacachep; }; extern struct cache_sizes malloc_sizes[]; @@ -211,7 +213,7 @@ extern unsigned int ksize(const void *); extern int slab_is_available(void); #ifdef CONFIG_NUMA -extern void *kmem_cache_alloc_node(kmem_cache_t *, gfp_t flags, int node); +extern void *kmem_cache_alloc_node(struct kmem_cache *, gfp_t flags, int node); extern void *__kmalloc_node(size_t size, gfp_t flags, int node); static inline void *kmalloc_node(size_t size, gfp_t flags, int node) @@ -255,7 +257,8 @@ extern void *__kmalloc_node_track_caller(size_t, gfp_t, int, void *); __builtin_return_address(0)) #endif #else /* CONFIG_NUMA */ -static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, gfp_t flags, int node) +static inline void *kmem_cache_alloc_node(struct kmem_cache *cachep, + gfp_t flags, int node) { return kmem_cache_alloc(cachep, flags); } @@ -269,7 +272,7 @@ static inline void *kmalloc_node(size_t size, gfp_t flags, int node) #endif extern int FASTCALL(kmem_cache_reap(int)); -extern int FASTCALL(kmem_ptr_validate(kmem_cache_t *cachep, void *ptr)); +extern int FASTCALL(kmem_ptr_validate(struct kmem_cache *cachep, void *ptr)); #else /* CONFIG_SLOB */ -- cgit v1.2.3 From 25ba77c141dbcd2602dd0171824d0d72aa023a01 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 6 Dec 2006 20:33:03 -0800 Subject: [PATCH] numa node ids are int, page_to_nid and zone_to_nid should return int NUMA node ids are passed as either int or unsigned int almost exclusivly page_to_nid and zone_to_nid both return unsigned long. This is a throw back to when page_to_nid was a #define and was thus exposing the real type of the page flags field. In addition to fixing up the definitions of page_to_nid and zone_to_nid I audited the users of these functions identifying the following incorrect uses: 1) mm/page_alloc.c show_node() -- printk dumping the node id, 2) include/asm-ia64/pgalloc.h pgtable_quicklist_free() -- comparison against numa_node_id() which returns an int from cpu_to_node(), and 3) mm/mpolicy.c check_pte_range -- used as an index in node_isset which uses bit_set which in generic code takes an int. Signed-off-by: Andy Whitcroft Cc: Christoph Lameter Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-ia64/pgalloc.h | 2 +- include/linux/mm.h | 6 +++--- mm/mempolicy.c | 2 +- mm/page_alloc.c | 2 +- mm/sparse.c | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/asm-ia64/pgalloc.h b/include/asm-ia64/pgalloc.h index 9cb68e9b377e..393e04c42a2c 100644 --- a/include/asm-ia64/pgalloc.h +++ b/include/asm-ia64/pgalloc.h @@ -60,7 +60,7 @@ static inline void *pgtable_quicklist_alloc(void) static inline void pgtable_quicklist_free(void *pgtable_entry) { #ifdef CONFIG_NUMA - unsigned long nid = page_to_nid(virt_to_page(pgtable_entry)); + int nid = page_to_nid(virt_to_page(pgtable_entry)); if (unlikely(nid != numa_node_id())) { free_page((unsigned long)pgtable_entry); diff --git a/include/linux/mm.h b/include/linux/mm.h index 840303769c11..0e266fe1b4c4 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -456,7 +456,7 @@ static inline int page_zone_id(struct page *page) return (page->flags >> ZONEID_PGSHIFT) & ZONEID_MASK; } -static inline unsigned long zone_to_nid(struct zone *zone) +static inline int zone_to_nid(struct zone *zone) { #ifdef CONFIG_NUMA return zone->node; @@ -466,9 +466,9 @@ static inline unsigned long zone_to_nid(struct zone *zone) } #ifdef NODE_NOT_IN_PAGE_FLAGS -extern unsigned long page_to_nid(struct page *page); +extern int page_to_nid(struct page *page); #else -static inline unsigned long page_to_nid(struct page *page) +static inline int page_to_nid(struct page *page) { return (page->flags >> NODES_PGSHIFT) & NODES_MASK; } diff --git a/mm/mempolicy.c b/mm/mempolicy.c index fb907236bbd8..e7b69c90cfd6 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -221,7 +221,7 @@ static int check_pte_range(struct vm_area_struct *vma, pmd_t *pmd, orig_pte = pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); do { struct page *page; - unsigned int nid; + int nid; if (!pte_present(*pte)) continue; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 86f2984f8b79..614d427854a8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1407,7 +1407,7 @@ unsigned int nr_free_pagecache_pages(void) static inline void show_node(struct zone *zone) { if (NUMA_BUILD) - printk("Node %ld ", zone_to_nid(zone)); + printk("Node %d ", zone_to_nid(zone)); } void si_meminfo(struct sysinfo *val) diff --git a/mm/sparse.c b/mm/sparse.c index 158d6a2a5263..ac26eb0d73cd 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -36,7 +36,7 @@ static u8 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned; static u16 section_to_node_table[NR_MEM_SECTIONS] __cacheline_aligned; #endif -unsigned long page_to_nid(struct page *page) +int page_to_nid(struct page *page) { return section_to_node_table[page_to_section(page)]; } -- cgit v1.2.3 From 6e0eaa4b05cf53ca5caa702fd2760a5b3376be69 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:10 -0800 Subject: [PATCH] slab: remove SLAB_NO_GROW It is only used internally in the slab. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 2 -- mm/slab.c | 4 ++-- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index b831776b2fc7..9ffd1c1616bd 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -28,8 +28,6 @@ typedef struct kmem_cache kmem_cache_t; #define SLAB_LEVEL_MASK GFP_LEVEL_MASK -#define SLAB_NO_GROW __GFP_NO_GROW /* don't grow a cache */ - /* flags to pass to kmem_cache_create(). * The first 3 are only valid when the allocator as been build * SLAB_DEBUG_SUPPORT. diff --git a/mm/slab.c b/mm/slab.c index 8f3f61cacb5c..e853dfe8fd7b 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2721,8 +2721,8 @@ static int cache_grow(struct kmem_cache *cachep, gfp_t flags, int nodeid) * Be lazy and only check for valid flags here, keeping it out of the * critical path in kmem_cache_alloc(). */ - BUG_ON(flags & ~(SLAB_DMA | SLAB_LEVEL_MASK | SLAB_NO_GROW)); - if (flags & SLAB_NO_GROW) + BUG_ON(flags & ~(SLAB_DMA | SLAB_LEVEL_MASK | __GFP_NO_GROW)); + if (flags & __GFP_NO_GROW) return 0; ctor_flags = SLAB_CTOR_CONSTRUCTOR; -- cgit v1.2.3 From a06d72c1dcbff015250df6ad9f0b1d18c02113bf Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:12 -0800 Subject: [PATCH] slab: remove SLAB_LEVEL_MASK SLAB_LEVEL_MASK is only used internally to the slab and is and alias of GFP_LEVEL_MASK. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 2 -- mm/slab.c | 4 ++-- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index 9ffd1c1616bd..6f7b9bb43065 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -26,8 +26,6 @@ typedef struct kmem_cache kmem_cache_t; #define SLAB_KERNEL GFP_KERNEL #define SLAB_DMA GFP_DMA -#define SLAB_LEVEL_MASK GFP_LEVEL_MASK - /* flags to pass to kmem_cache_create(). * The first 3 are only valid when the allocator as been build * SLAB_DEBUG_SUPPORT. diff --git a/mm/slab.c b/mm/slab.c index e853dfe8fd7b..9f34b4946fba 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2721,12 +2721,12 @@ static int cache_grow(struct kmem_cache *cachep, gfp_t flags, int nodeid) * Be lazy and only check for valid flags here, keeping it out of the * critical path in kmem_cache_alloc(). */ - BUG_ON(flags & ~(SLAB_DMA | SLAB_LEVEL_MASK | __GFP_NO_GROW)); + BUG_ON(flags & ~(SLAB_DMA | GFP_LEVEL_MASK | __GFP_NO_GROW)); if (flags & __GFP_NO_GROW) return 0; ctor_flags = SLAB_CTOR_CONSTRUCTOR; - local_flags = (flags & SLAB_LEVEL_MASK); + local_flags = (flags & GFP_LEVEL_MASK); if (!(local_flags & __GFP_WAIT)) /* * Not allowed to sleep. Need to tell a constructor about -- cgit v1.2.3 From 55acbda0965ca0a29b0ca276e7d17a55edc11d1b Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:13 -0800 Subject: [PATCH] slab: remove SLAB_NOIO SLAB_NOIO is an alias of GFP_NOIO with a single instance of use. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/usb/storage/transport.c | 2 +- include/linux/slab.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c index 47644b5b6155..323293a3e61f 100644 --- a/drivers/usb/storage/transport.c +++ b/drivers/usb/storage/transport.c @@ -427,7 +427,7 @@ static int usb_stor_bulk_transfer_sglist(struct us_data *us, unsigned int pipe, US_DEBUGP("%s: xfer %u bytes, %d entries\n", __FUNCTION__, length, num_sg); result = usb_sg_init(&us->current_sg, us->pusb_dev, pipe, 0, - sg, num_sg, length, SLAB_NOIO); + sg, num_sg, length, GFP_NOIO); if (result) { US_DEBUGP("usb_sg_init returned %d\n", result); return USB_STOR_XFER_ERROR; diff --git a/include/linux/slab.h b/include/linux/slab.h index 6f7b9bb43065..43ced80c327b 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -20,7 +20,6 @@ typedef struct kmem_cache kmem_cache_t; /* flags for kmem_cache_alloc() */ #define SLAB_NOFS GFP_NOFS -#define SLAB_NOIO GFP_NOIO #define SLAB_ATOMIC GFP_ATOMIC #define SLAB_USER GFP_USER #define SLAB_KERNEL GFP_KERNEL -- cgit v1.2.3 From e6b4f8da3a88457148038bc952043e99a7fdba64 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:14 -0800 Subject: [PATCH] slab: remove SLAB_NOFS SLAB_NOFS is an alias of GFP_NOFS. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/cifs/misc.c | 4 ++-- fs/cifs/transport.c | 2 +- fs/dquot.c | 2 +- fs/ext3/super.c | 2 +- fs/ext4/super.c | 2 +- fs/hpfs/super.c | 2 +- fs/nfs/read.c | 2 +- fs/nfs/write.c | 4 ++-- fs/ntfs/attrib.c | 2 +- fs/ntfs/index.c | 2 +- fs/ntfs/inode.c | 4 ++-- fs/ntfs/unistr.c | 2 +- fs/ocfs2/dlm/dlmfs.c | 2 +- fs/ocfs2/super.c | 2 +- include/linux/slab.h | 1 - 15 files changed, 17 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index bbc9cd34b6ea..8355daff504c 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -153,7 +153,7 @@ cifs_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_req_poolp, SLAB_KERNEL | SLAB_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_req_poolp, SLAB_KERNEL | GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -192,7 +192,7 @@ cifs_small_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, SLAB_KERNEL | SLAB_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, SLAB_KERNEL | GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 48d47b46b1fb..7514237cf31a 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -51,7 +51,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) } temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - SLAB_KERNEL | SLAB_NOFS); + SLAB_KERNEL | GFP_NOFS); if (temp == NULL) return temp; else { diff --git a/fs/dquot.c b/fs/dquot.c index 9af789567e51..c6ae6c0e0cfc 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -600,7 +600,7 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type) { struct dquot *dquot; - dquot = kmem_cache_alloc(dquot_cachep, SLAB_NOFS); + dquot = kmem_cache_alloc(dquot_cachep, GFP_NOFS); if(!dquot) return NODQUOT; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index afc2d4f42d77..0cf633f0cfa3 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -445,7 +445,7 @@ static struct inode *ext3_alloc_inode(struct super_block *sb) { struct ext3_inode_info *ei; - ei = kmem_cache_alloc(ext3_inode_cachep, SLAB_NOFS); + ei = kmem_cache_alloc(ext3_inode_cachep, GFP_NOFS); if (!ei) return NULL; #ifdef CONFIG_EXT3_FS_POSIX_ACL diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b4b022aa2bc2..c730cbc84030 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -495,7 +495,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) { struct ext4_inode_info *ei; - ei = kmem_cache_alloc(ext4_inode_cachep, SLAB_NOFS); + ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS); if (!ei) return NULL; #ifdef CONFIG_EXT4DEV_FS_POSIX_ACL diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 450b5e0b4785..46ceadd6f16a 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -165,7 +165,7 @@ static kmem_cache_t * hpfs_inode_cachep; static struct inode *hpfs_alloc_inode(struct super_block *sb) { struct hpfs_inode_info *ei; - ei = (struct hpfs_inode_info *)kmem_cache_alloc(hpfs_inode_cachep, SLAB_NOFS); + ei = (struct hpfs_inode_info *)kmem_cache_alloc(hpfs_inode_cachep, GFP_NOFS); if (!ei) return NULL; ei->vfs_inode.i_version = 1; diff --git a/fs/nfs/read.c b/fs/nfs/read.c index c2e49c397a27..56f66f0ccb6a 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -46,7 +46,7 @@ static mempool_t *nfs_rdata_mempool; struct nfs_read_data *nfs_readdata_alloc(size_t len) { unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; - struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, SLAB_NOFS); + struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 883dd4a1c157..f7dd0d005957 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -93,7 +93,7 @@ static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion); struct nfs_write_data *nfs_commit_alloc(void) { - struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, SLAB_NOFS); + struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); @@ -112,7 +112,7 @@ void nfs_commit_free(struct nfs_write_data *p) struct nfs_write_data *nfs_writedata_alloc(size_t len) { unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; - struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, SLAB_NOFS); + struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index 9f08e851cfb6..c577d8e1bd95 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -1272,7 +1272,7 @@ ntfs_attr_search_ctx *ntfs_attr_get_search_ctx(ntfs_inode *ni, MFT_RECORD *mrec) { ntfs_attr_search_ctx *ctx; - ctx = kmem_cache_alloc(ntfs_attr_ctx_cache, SLAB_NOFS); + ctx = kmem_cache_alloc(ntfs_attr_ctx_cache, GFP_NOFS); if (ctx) ntfs_attr_init_search_ctx(ctx, ni, mrec); return ctx; diff --git a/fs/ntfs/index.c b/fs/ntfs/index.c index e32cde486362..2194eff49743 100644 --- a/fs/ntfs/index.c +++ b/fs/ntfs/index.c @@ -38,7 +38,7 @@ ntfs_index_context *ntfs_index_ctx_get(ntfs_inode *idx_ni) { ntfs_index_context *ictx; - ictx = kmem_cache_alloc(ntfs_index_ctx_cache, SLAB_NOFS); + ictx = kmem_cache_alloc(ntfs_index_ctx_cache, GFP_NOFS); if (ictx) *ictx = (ntfs_index_context){ .idx_ni = idx_ni }; return ictx; diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 2d3de9c89818..247989891b4b 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -324,7 +324,7 @@ struct inode *ntfs_alloc_big_inode(struct super_block *sb) ntfs_inode *ni; ntfs_debug("Entering."); - ni = kmem_cache_alloc(ntfs_big_inode_cache, SLAB_NOFS); + ni = kmem_cache_alloc(ntfs_big_inode_cache, GFP_NOFS); if (likely(ni != NULL)) { ni->state = 0; return VFS_I(ni); @@ -349,7 +349,7 @@ static inline ntfs_inode *ntfs_alloc_extent_inode(void) ntfs_inode *ni; ntfs_debug("Entering."); - ni = kmem_cache_alloc(ntfs_inode_cache, SLAB_NOFS); + ni = kmem_cache_alloc(ntfs_inode_cache, GFP_NOFS); if (likely(ni != NULL)) { ni->state = 0; return ni; diff --git a/fs/ntfs/unistr.c b/fs/ntfs/unistr.c index 6a495f7369f9..005ca4b0f132 100644 --- a/fs/ntfs/unistr.c +++ b/fs/ntfs/unistr.c @@ -266,7 +266,7 @@ int ntfs_nlstoucs(const ntfs_volume *vol, const char *ins, /* We do not trust outside sources. */ if (likely(ins)) { - ucs = kmem_cache_alloc(ntfs_name_cache, SLAB_NOFS); + ucs = kmem_cache_alloc(ntfs_name_cache, GFP_NOFS); if (likely(ucs)) { for (i = o = 0; i < ins_len; i += wc_len) { wc_len = nls->char2uni(ins + i, ins_len - i, diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 16b8d1ba7066..01f91501f70a 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -276,7 +276,7 @@ static struct inode *dlmfs_alloc_inode(struct super_block *sb) { struct dlmfs_inode_private *ip; - ip = kmem_cache_alloc(dlmfs_inode_cache, SLAB_NOFS); + ip = kmem_cache_alloc(dlmfs_inode_cache, GFP_NOFS); if (!ip) return NULL; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index d9b4214a12da..7574d26ee0ff 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -303,7 +303,7 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb) { struct ocfs2_inode_info *oi; - oi = kmem_cache_alloc(ocfs2_inode_cachep, SLAB_NOFS); + oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS); if (!oi) return NULL; diff --git a/include/linux/slab.h b/include/linux/slab.h index 43ced80c327b..5b70d52b4f86 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -19,7 +19,6 @@ typedef struct kmem_cache kmem_cache_t; #include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ /* flags for kmem_cache_alloc() */ -#define SLAB_NOFS GFP_NOFS #define SLAB_ATOMIC GFP_ATOMIC #define SLAB_USER GFP_USER #define SLAB_KERNEL GFP_KERNEL -- cgit v1.2.3 From f7267c0c0721fd02ad3dc37c3d6dd24ccd81d4d6 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:15 -0800 Subject: [PATCH] slab: remove SLAB_USER SLAB_USER is an alias of GFP_USER Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/crypto.c | 4 ++-- fs/ecryptfs/inode.c | 2 +- include/linux/slab.h | 1 - 3 files changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f63a7755fe86..776b2eed371e 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1334,7 +1334,7 @@ int ecryptfs_write_headers(struct dentry *ecryptfs_dentry, goto out; } /* Released in this function */ - page_virt = kmem_cache_alloc(ecryptfs_header_cache_0, SLAB_USER); + page_virt = kmem_cache_alloc(ecryptfs_header_cache_0, GFP_USER); if (!page_virt) { ecryptfs_printk(KERN_ERR, "Out of memory\n"); rc = -ENOMEM; @@ -1493,7 +1493,7 @@ int ecryptfs_read_headers(struct dentry *ecryptfs_dentry, &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; /* Read the first page from the underlying file */ - page_virt = kmem_cache_alloc(ecryptfs_header_cache_1, SLAB_USER); + page_virt = kmem_cache_alloc(ecryptfs_header_cache_1, GFP_USER); if (!page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Unable to allocate page_virt\n"); diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index dfcc68484f47..70911412044d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -404,7 +404,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, /* Released in this function */ page_virt = (char *)kmem_cache_alloc(ecryptfs_header_cache_2, - SLAB_USER); + GFP_USER); if (!page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, diff --git a/include/linux/slab.h b/include/linux/slab.h index 5b70d52b4f86..d7ee28e51330 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -20,7 +20,6 @@ typedef struct kmem_cache kmem_cache_t; /* flags for kmem_cache_alloc() */ #define SLAB_ATOMIC GFP_ATOMIC -#define SLAB_USER GFP_USER #define SLAB_KERNEL GFP_KERNEL #define SLAB_DMA GFP_DMA -- cgit v1.2.3 From 54e6ecb23951b195d02433a741c7f7cb0b796c78 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:16 -0800 Subject: [PATCH] slab: remove SLAB_ATOMIC SLAB_ATOMIC is an alias of GFP_ATOMIC Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/atm/he.c | 2 +- drivers/base/dmapool.c | 2 +- drivers/block/DAC960.c | 4 ++-- drivers/char/watchdog/pcwd_usb.c | 2 +- drivers/ieee1394/raw1394.c | 30 +++++++++++++++--------------- drivers/infiniband/hw/amso1100/c2_vq.c | 2 +- drivers/infiniband/hw/mthca/mthca_av.c | 2 +- drivers/isdn/gigaset/bas-gigaset.c | 20 ++++++++++---------- drivers/isdn/gigaset/usb-gigaset.c | 6 +++--- drivers/media/dvb/dvb-usb/usb-urb.c | 2 +- drivers/media/dvb/ttusb-dec/ttusb_dec.c | 2 +- drivers/s390/scsi/zfcp_fsf.c | 2 +- drivers/usb/core/hub.c | 2 +- drivers/usb/core/message.c | 2 +- drivers/usb/host/ehci-dbg.c | 2 +- drivers/usb/host/hc_crisv10.c | 10 +++++----- drivers/usb/host/ohci-dbg.c | 2 +- drivers/usb/host/uhci-q.c | 2 +- drivers/usb/input/aiptek.c | 2 +- drivers/usb/input/ati_remote.c | 6 +++--- drivers/usb/input/hid-core.c | 10 +++++----- drivers/usb/input/keyspan_remote.c | 2 +- drivers/usb/input/mtouchusb.c | 2 +- drivers/usb/input/powermate.c | 4 ++-- drivers/usb/input/touchkitusb.c | 2 +- drivers/usb/input/usbkbd.c | 8 ++++---- drivers/usb/input/usbmouse.c | 4 ++-- drivers/usb/input/xpad.c | 2 +- drivers/usb/input/yealink.c | 6 +++--- drivers/usb/misc/phidgetkit.c | 4 ++-- drivers/usb/misc/phidgetmotorcontrol.c | 4 ++-- drivers/usb/misc/usbtest.c | 8 ++++---- drivers/usb/mon/mon_text.c | 4 ++-- drivers/usb/net/catc.c | 2 +- drivers/usb/net/net1080.c | 2 +- drivers/usb/net/pegasus.c | 2 +- drivers/usb/net/rtl8150.c | 2 +- drivers/usb/serial/mos7720.c | 2 +- drivers/usb/serial/mos7840.c | 4 ++-- drivers/usb/storage/onetouch.c | 4 ++-- include/linux/slab.h | 1 - include/net/request_sock.h | 2 +- net/core/dst.c | 2 +- net/core/flow.c | 2 +- net/core/neighbour.c | 2 +- net/dccp/ccids/ccid3.c | 6 +++--- net/dccp/ccids/lib/loss_interval.c | 2 +- net/ipv4/inet_hashtables.c | 2 +- net/ipv4/inet_timewait_sock.c | 2 +- net/ipv6/ip6_fib.c | 2 +- net/ipv6/xfrm6_tunnel.c | 2 +- net/sctp/sm_make_chunk.c | 2 +- net/sctp/socket.c | 2 +- net/xfrm/xfrm_input.c | 2 +- security/selinux/avc.c | 2 +- 55 files changed, 107 insertions(+), 108 deletions(-) (limited to 'include/linux') diff --git a/drivers/atm/he.c b/drivers/atm/he.c index c7314a79da0f..2a2f0fc2288f 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -1724,7 +1724,7 @@ __alloc_tpd(struct he_dev *he_dev) struct he_tpd *tpd; dma_addr_t dma_handle; - tpd = pci_pool_alloc(he_dev->tpd_pool, SLAB_ATOMIC|SLAB_DMA, &dma_handle); + tpd = pci_pool_alloc(he_dev->tpd_pool, GFP_ATOMIC|SLAB_DMA, &dma_handle); if (tpd == NULL) return NULL; diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index b2efbd4cf710..fa4675254f67 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -297,7 +297,7 @@ restart: } } } - if (!(page = pool_alloc_page (pool, SLAB_ATOMIC))) { + if (!(page = pool_alloc_page (pool, GFP_ATOMIC))) { if (mem_flags & __GFP_WAIT) { DECLARE_WAITQUEUE (wait, current); diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 742d07403101..8d81a3a64c07 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -324,13 +324,13 @@ static boolean DAC960_CreateAuxiliaryStructures(DAC960_Controller_T *Controller) Command->Next = Controller->FreeCommands; Controller->FreeCommands = Command; Controller->Commands[CommandIdentifier-1] = Command; - ScatterGatherCPU = pci_pool_alloc(ScatterGatherPool, SLAB_ATOMIC, + ScatterGatherCPU = pci_pool_alloc(ScatterGatherPool, GFP_ATOMIC, &ScatterGatherDMA); if (ScatterGatherCPU == NULL) return DAC960_Failure(Controller, "AUXILIARY STRUCTURE CREATION"); if (RequestSensePool != NULL) { - RequestSenseCPU = pci_pool_alloc(RequestSensePool, SLAB_ATOMIC, + RequestSenseCPU = pci_pool_alloc(RequestSensePool, GFP_ATOMIC, &RequestSenseDMA); if (RequestSenseCPU == NULL) { pci_pool_free(ScatterGatherPool, ScatterGatherCPU, diff --git a/drivers/char/watchdog/pcwd_usb.c b/drivers/char/watchdog/pcwd_usb.c index e275dd4a705d..61138726b501 100644 --- a/drivers/char/watchdog/pcwd_usb.c +++ b/drivers/char/watchdog/pcwd_usb.c @@ -634,7 +634,7 @@ static int usb_pcwd_probe(struct usb_interface *interface, const struct usb_devi usb_pcwd->intr_size = (le16_to_cpu(endpoint->wMaxPacketSize) > 8 ? le16_to_cpu(endpoint->wMaxPacketSize) : 8); /* set up the memory buffer's */ - if (!(usb_pcwd->intr_buffer = usb_buffer_alloc(udev, usb_pcwd->intr_size, SLAB_ATOMIC, &usb_pcwd->intr_dma))) { + if (!(usb_pcwd->intr_buffer = usb_buffer_alloc(udev, usb_pcwd->intr_size, GFP_ATOMIC, &usb_pcwd->intr_dma))) { printk(KERN_ERR PFX "Out of memory\n"); goto error; } diff --git a/drivers/ieee1394/raw1394.c b/drivers/ieee1394/raw1394.c index 5ec4f5eb6b19..47f6a4e29b40 100644 --- a/drivers/ieee1394/raw1394.c +++ b/drivers/ieee1394/raw1394.c @@ -259,7 +259,7 @@ static void host_reset(struct hpsb_host *host) if (hi != NULL) { list_for_each_entry(fi, &hi->file_info_list, list) { if (fi->notification == RAW1394_NOTIFY_ON) { - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (req != NULL) { req->file_info = fi; @@ -306,13 +306,13 @@ static void iso_receive(struct hpsb_host *host, int channel, quadlet_t * data, if (!(fi->listen_channels & (1ULL << channel))) continue; - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) break; if (!ibs) { ibs = kmalloc(sizeof(*ibs) + length, - SLAB_ATOMIC); + GFP_ATOMIC); if (!ibs) { kfree(req); break; @@ -367,13 +367,13 @@ static void fcp_request(struct hpsb_host *host, int nodeid, int direction, if (!fi->fcp_buffer) continue; - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) break; if (!ibs) { ibs = kmalloc(sizeof(*ibs) + length, - SLAB_ATOMIC); + GFP_ATOMIC); if (!ibs) { kfree(req); break; @@ -593,7 +593,7 @@ static int state_initialized(struct file_info *fi, struct pending_request *req) switch (req->req.type) { case RAW1394_REQ_LIST_CARDS: spin_lock_irqsave(&host_info_lock, flags); - khl = kmalloc(sizeof(*khl) * host_count, SLAB_ATOMIC); + khl = kmalloc(sizeof(*khl) * host_count, GFP_ATOMIC); if (khl) { req->req.misc = host_count; @@ -1045,7 +1045,7 @@ static int arm_read(struct hpsb_host *host, int nodeid, quadlet_t * buffer, } if (arm_addr->notification_options & ARM_READ) { DBGMSG("arm_read -> entering notification-section"); - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) { DBGMSG("arm_read -> rcode_conflict_error"); spin_unlock_irqrestore(&host_info_lock, irqflags); @@ -1064,7 +1064,7 @@ static int arm_read(struct hpsb_host *host, int nodeid, quadlet_t * buffer, sizeof(struct arm_response) + sizeof(struct arm_request_response); } - req->data = kmalloc(size, SLAB_ATOMIC); + req->data = kmalloc(size, GFP_ATOMIC); if (!(req->data)) { free_pending_request(req); DBGMSG("arm_read -> rcode_conflict_error"); @@ -1198,7 +1198,7 @@ static int arm_write(struct hpsb_host *host, int nodeid, int destid, } if (arm_addr->notification_options & ARM_WRITE) { DBGMSG("arm_write -> entering notification-section"); - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) { DBGMSG("arm_write -> rcode_conflict_error"); spin_unlock_irqrestore(&host_info_lock, irqflags); @@ -1209,7 +1209,7 @@ static int arm_write(struct hpsb_host *host, int nodeid, int destid, sizeof(struct arm_request) + sizeof(struct arm_response) + (length) * sizeof(byte_t) + sizeof(struct arm_request_response); - req->data = kmalloc(size, SLAB_ATOMIC); + req->data = kmalloc(size, GFP_ATOMIC); if (!(req->data)) { free_pending_request(req); DBGMSG("arm_write -> rcode_conflict_error"); @@ -1400,7 +1400,7 @@ static int arm_lock(struct hpsb_host *host, int nodeid, quadlet_t * store, if (arm_addr->notification_options & ARM_LOCK) { byte_t *buf1, *buf2; DBGMSG("arm_lock -> entering notification-section"); - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) { DBGMSG("arm_lock -> rcode_conflict_error"); spin_unlock_irqrestore(&host_info_lock, irqflags); @@ -1408,7 +1408,7 @@ static int arm_lock(struct hpsb_host *host, int nodeid, quadlet_t * store, The request may be retried */ } size = sizeof(struct arm_request) + sizeof(struct arm_response) + 3 * sizeof(*store) + sizeof(struct arm_request_response); /* maximum */ - req->data = kmalloc(size, SLAB_ATOMIC); + req->data = kmalloc(size, GFP_ATOMIC); if (!(req->data)) { free_pending_request(req); DBGMSG("arm_lock -> rcode_conflict_error"); @@ -1628,7 +1628,7 @@ static int arm_lock64(struct hpsb_host *host, int nodeid, octlet_t * store, if (arm_addr->notification_options & ARM_LOCK) { byte_t *buf1, *buf2; DBGMSG("arm_lock64 -> entering notification-section"); - req = __alloc_pending_request(SLAB_ATOMIC); + req = __alloc_pending_request(GFP_ATOMIC); if (!req) { spin_unlock_irqrestore(&host_info_lock, irqflags); DBGMSG("arm_lock64 -> rcode_conflict_error"); @@ -1636,7 +1636,7 @@ static int arm_lock64(struct hpsb_host *host, int nodeid, octlet_t * store, The request may be retried */ } size = sizeof(struct arm_request) + sizeof(struct arm_response) + 3 * sizeof(*store) + sizeof(struct arm_request_response); /* maximum */ - req->data = kmalloc(size, SLAB_ATOMIC); + req->data = kmalloc(size, GFP_ATOMIC); if (!(req->data)) { free_pending_request(req); spin_unlock_irqrestore(&host_info_lock, irqflags); @@ -2443,7 +2443,7 @@ static void queue_rawiso_event(struct file_info *fi) /* only one ISO activity event may be in the queue */ if (!__rawiso_event_in_queue(fi)) { struct pending_request *req = - __alloc_pending_request(SLAB_ATOMIC); + __alloc_pending_request(GFP_ATOMIC); if (req) { req->file_info = fi; diff --git a/drivers/infiniband/hw/amso1100/c2_vq.c b/drivers/infiniband/hw/amso1100/c2_vq.c index 40caeb5f41b4..36620a22413c 100644 --- a/drivers/infiniband/hw/amso1100/c2_vq.c +++ b/drivers/infiniband/hw/amso1100/c2_vq.c @@ -164,7 +164,7 @@ void vq_req_put(struct c2_dev *c2dev, struct c2_vq_req *r) */ void *vq_repbuf_alloc(struct c2_dev *c2dev) { - return kmem_cache_alloc(c2dev->host_msg_cache, SLAB_ATOMIC); + return kmem_cache_alloc(c2dev->host_msg_cache, GFP_ATOMIC); } /* diff --git a/drivers/infiniband/hw/mthca/mthca_av.c b/drivers/infiniband/hw/mthca/mthca_av.c index 57cdc1bc5f50..27caf3b0648a 100644 --- a/drivers/infiniband/hw/mthca/mthca_av.c +++ b/drivers/infiniband/hw/mthca/mthca_av.c @@ -189,7 +189,7 @@ int mthca_create_ah(struct mthca_dev *dev, on_hca_fail: if (ah->type == MTHCA_AH_PCI_POOL) { ah->av = pci_pool_alloc(dev->av_table.pool, - SLAB_ATOMIC, &ah->avdma); + GFP_ATOMIC, &ah->avdma); if (!ah->av) return -ENOMEM; diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c index 0c937325a1b3..5857e7e23f84 100644 --- a/drivers/isdn/gigaset/bas-gigaset.c +++ b/drivers/isdn/gigaset/bas-gigaset.c @@ -572,7 +572,7 @@ static int atread_submit(struct cardstate *cs, int timeout) ucs->rcvbuf, ucs->rcvbuf_size, read_ctrl_callback, cs->inbuf); - if ((ret = usb_submit_urb(ucs->urb_cmd_in, SLAB_ATOMIC)) != 0) { + if ((ret = usb_submit_urb(ucs->urb_cmd_in, GFP_ATOMIC)) != 0) { update_basstate(ucs, 0, BS_ATRDPEND); dev_err(cs->dev, "could not submit HD_READ_ATMESSAGE: %s\n", get_usb_rcmsg(ret)); @@ -747,7 +747,7 @@ static void read_int_callback(struct urb *urb) check_pending(ucs); resubmit: - rc = usb_submit_urb(urb, SLAB_ATOMIC); + rc = usb_submit_urb(urb, GFP_ATOMIC); if (unlikely(rc != 0 && rc != -ENODEV)) { dev_err(cs->dev, "could not resubmit interrupt URB: %s\n", get_usb_rcmsg(rc)); @@ -807,7 +807,7 @@ static void read_iso_callback(struct urb *urb) urb->number_of_packets = BAS_NUMFRAMES; gig_dbg(DEBUG_ISO, "%s: isoc read overrun/resubmit", __func__); - rc = usb_submit_urb(urb, SLAB_ATOMIC); + rc = usb_submit_urb(urb, GFP_ATOMIC); if (unlikely(rc != 0 && rc != -ENODEV)) { dev_err(bcs->cs->dev, "could not resubmit isochronous read " @@ -900,7 +900,7 @@ static int starturbs(struct bc_state *bcs) } dump_urb(DEBUG_ISO, "Initial isoc read", urb); - if ((rc = usb_submit_urb(urb, SLAB_ATOMIC)) != 0) + if ((rc = usb_submit_urb(urb, GFP_ATOMIC)) != 0) goto error; } @@ -935,7 +935,7 @@ static int starturbs(struct bc_state *bcs) /* submit two URBs, keep third one */ for (k = 0; k < 2; ++k) { dump_urb(DEBUG_ISO, "Initial isoc write", urb); - rc = usb_submit_urb(ubc->isoouturbs[k].urb, SLAB_ATOMIC); + rc = usb_submit_urb(ubc->isoouturbs[k].urb, GFP_ATOMIC); if (rc != 0) goto error; } @@ -1042,7 +1042,7 @@ static int submit_iso_write_urb(struct isow_urbctx_t *ucx) return 0; /* no data to send */ urb->number_of_packets = nframe; - rc = usb_submit_urb(urb, SLAB_ATOMIC); + rc = usb_submit_urb(urb, GFP_ATOMIC); if (unlikely(rc)) { if (rc == -ENODEV) /* device removed - give up silently */ @@ -1341,7 +1341,7 @@ static void read_iso_tasklet(unsigned long data) urb->dev = bcs->cs->hw.bas->udev; urb->transfer_flags = URB_ISO_ASAP; urb->number_of_packets = BAS_NUMFRAMES; - rc = usb_submit_urb(urb, SLAB_ATOMIC); + rc = usb_submit_urb(urb, GFP_ATOMIC); if (unlikely(rc != 0 && rc != -ENODEV)) { dev_err(cs->dev, "could not resubmit isochronous read URB: %s\n", @@ -1458,7 +1458,7 @@ static void write_ctrl_callback(struct urb *urb) ucs->retry_ctrl); /* urb->dev is clobbered by USB subsystem */ urb->dev = ucs->udev; - rc = usb_submit_urb(urb, SLAB_ATOMIC); + rc = usb_submit_urb(urb, GFP_ATOMIC); if (unlikely(rc)) { dev_err(&ucs->interface->dev, "could not resubmit request 0x%02x: %s\n", @@ -1517,7 +1517,7 @@ static int req_submit(struct bc_state *bcs, int req, int val, int timeout) (unsigned char*) &ucs->dr_ctrl, NULL, 0, write_ctrl_callback, ucs); ucs->retry_ctrl = 0; - ret = usb_submit_urb(ucs->urb_ctrl, SLAB_ATOMIC); + ret = usb_submit_urb(ucs->urb_ctrl, GFP_ATOMIC); if (unlikely(ret)) { dev_err(bcs->cs->dev, "could not submit request 0x%02x: %s\n", req, get_usb_rcmsg(ret)); @@ -1763,7 +1763,7 @@ static int atwrite_submit(struct cardstate *cs, unsigned char *buf, int len) usb_sndctrlpipe(ucs->udev, 0), (unsigned char*) &ucs->dr_cmd_out, buf, len, write_command_callback, cs); - rc = usb_submit_urb(ucs->urb_cmd_out, SLAB_ATOMIC); + rc = usb_submit_urb(ucs->urb_cmd_out, GFP_ATOMIC); if (unlikely(rc)) { update_basstate(ucs, 0, BS_ATWRPEND); dev_err(cs->dev, "could not submit HD_WRITE_ATMESSAGE: %s\n", diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c index 5ebf49ac9b23..af89ce188f2a 100644 --- a/drivers/isdn/gigaset/usb-gigaset.c +++ b/drivers/isdn/gigaset/usb-gigaset.c @@ -410,7 +410,7 @@ static void gigaset_read_int_callback(struct urb *urb) if (resubmit) { spin_lock_irqsave(&cs->lock, flags); - r = cs->connected ? usb_submit_urb(urb, SLAB_ATOMIC) : -ENODEV; + r = cs->connected ? usb_submit_urb(urb, GFP_ATOMIC) : -ENODEV; spin_unlock_irqrestore(&cs->lock, flags); if (r) dev_err(cs->dev, "error %d when resubmitting urb.\n", @@ -486,7 +486,7 @@ static int send_cb(struct cardstate *cs, struct cmdbuf_t *cb) atomic_set(&ucs->busy, 1); spin_lock_irqsave(&cs->lock, flags); - status = cs->connected ? usb_submit_urb(ucs->bulk_out_urb, SLAB_ATOMIC) : -ENODEV; + status = cs->connected ? usb_submit_urb(ucs->bulk_out_urb, GFP_ATOMIC) : -ENODEV; spin_unlock_irqrestore(&cs->lock, flags); if (status) { @@ -664,7 +664,7 @@ static int write_modem(struct cardstate *cs) ucs->bulk_out_endpointAddr & 0x0f), ucs->bulk_out_buffer, count, gigaset_write_bulk_callback, cs); - ret = usb_submit_urb(ucs->bulk_out_urb, SLAB_ATOMIC); + ret = usb_submit_urb(ucs->bulk_out_urb, GFP_ATOMIC); } else { ret = -ENODEV; } diff --git a/drivers/media/dvb/dvb-usb/usb-urb.c b/drivers/media/dvb/dvb-usb/usb-urb.c index 78035ee824ca..397f51a7b2ad 100644 --- a/drivers/media/dvb/dvb-usb/usb-urb.c +++ b/drivers/media/dvb/dvb-usb/usb-urb.c @@ -116,7 +116,7 @@ static int usb_allocate_stream_buffers(struct usb_data_stream *stream, int num, for (stream->buf_num = 0; stream->buf_num < num; stream->buf_num++) { deb_mem("allocating buffer %d\n",stream->buf_num); if (( stream->buf_list[stream->buf_num] = - usb_buffer_alloc(stream->udev, size, SLAB_ATOMIC, + usb_buffer_alloc(stream->udev, size, GFP_ATOMIC, &stream->dma_addr[stream->buf_num]) ) == NULL) { deb_mem("not enough memory for urb-buffer allocation.\n"); usb_free_stream_buffers(stream); diff --git a/drivers/media/dvb/ttusb-dec/ttusb_dec.c b/drivers/media/dvb/ttusb-dec/ttusb_dec.c index 8135f3e76aeb..10b121ada833 100644 --- a/drivers/media/dvb/ttusb-dec/ttusb_dec.c +++ b/drivers/media/dvb/ttusb-dec/ttusb_dec.c @@ -1244,7 +1244,7 @@ static int ttusb_dec_init_usb(struct ttusb_dec *dec) return -ENOMEM; } dec->irq_buffer = usb_buffer_alloc(dec->udev,IRQ_PACKET_SIZE, - SLAB_ATOMIC, &dec->irq_dma_handle); + GFP_ATOMIC, &dec->irq_dma_handle); if(!dec->irq_buffer) { return -ENOMEM; } diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index 277826cdd0c8..067f1519eb04 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -109,7 +109,7 @@ zfcp_fsf_req_alloc(mempool_t *pool, int req_flags) ptr = kmalloc(size, GFP_ATOMIC); else ptr = kmem_cache_alloc(zfcp_data.fsf_req_qtcb_cache, - SLAB_ATOMIC); + GFP_ATOMIC); } if (unlikely(!ptr)) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 9be41ed1f9a6..0a46acf557ac 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -460,7 +460,7 @@ void usb_hub_tt_clear_buffer (struct usb_device *udev, int pipe) * since each TT has "at least two" buffers that can need it (and * there can be many TTs per hub). even if they're uncommon. */ - if ((clear = kmalloc (sizeof *clear, SLAB_ATOMIC)) == NULL) { + if ((clear = kmalloc (sizeof *clear, GFP_ATOMIC)) == NULL) { dev_err (&udev->dev, "can't save CLEAR_TT_BUFFER state\n"); /* FIXME recover somehow ... RESET_TT? */ return; diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 7390b67c609d..149aa8bfb1fe 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -488,7 +488,7 @@ void usb_sg_wait (struct usb_sg_request *io) int retval; io->urbs [i]->dev = io->dev; - retval = usb_submit_urb (io->urbs [i], SLAB_ATOMIC); + retval = usb_submit_urb (io->urbs [i], GFP_ATOMIC); /* after we submit, let completions or cancelations fire; * we handshake using io->status. diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c index 34b7a31cd85b..56349d21e6ea 100644 --- a/drivers/usb/host/ehci-dbg.c +++ b/drivers/usb/host/ehci-dbg.c @@ -492,7 +492,7 @@ show_periodic (struct class_device *class_dev, char *buf) unsigned i; __le32 tag; - if (!(seen = kmalloc (DBG_SCHED_LIMIT * sizeof *seen, SLAB_ATOMIC))) + if (!(seen = kmalloc (DBG_SCHED_LIMIT * sizeof *seen, GFP_ATOMIC))) return 0; seen_count = 0; diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 87eca6aeacf2..396dc69d4b4c 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -188,7 +188,7 @@ static DEFINE_TIMER(bulk_eot_timer, NULL, 0, 0); #define CHECK_ALIGN(x) if (((__u32)(x)) & 0x00000003) \ {panic("Alignment check (DWORD) failed at %s:%s:%d\n", __FILE__, __FUNCTION__, __LINE__);} -#define SLAB_FLAG (in_interrupt() ? SLAB_ATOMIC : SLAB_KERNEL) +#define SLAB_FLAG (in_interrupt() ? GFP_ATOMIC : SLAB_KERNEL) #define KMALLOC_FLAG (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) /* Most helpful debugging aid */ @@ -1743,7 +1743,7 @@ static irqreturn_t etrax_usb_tx_interrupt(int irq, void *vhc) *R_DMA_CH8_SUB3_CLR_INTR = IO_STATE(R_DMA_CH8_SUB3_CLR_INTR, clr_descr, do); - comp_data = (usb_isoc_complete_data_t*)kmem_cache_alloc(isoc_compl_cache, SLAB_ATOMIC); + comp_data = (usb_isoc_complete_data_t*)kmem_cache_alloc(isoc_compl_cache, GFP_ATOMIC); assert(comp_data != NULL); INIT_WORK(&comp_data->usb_bh, etrax_usb_isoc_descr_interrupt_bottom_half, comp_data); @@ -3010,7 +3010,7 @@ static void etrax_usb_add_to_isoc_sb_list(struct urb *urb, int epid) if (!urb->iso_frame_desc[i].length) continue; - next_sb_desc = (USB_SB_Desc_t*)kmem_cache_alloc(usb_desc_cache, SLAB_ATOMIC); + next_sb_desc = (USB_SB_Desc_t*)kmem_cache_alloc(usb_desc_cache, GFP_ATOMIC); assert(next_sb_desc != NULL); if (urb->iso_frame_desc[i].length > 0) { @@ -3063,7 +3063,7 @@ static void etrax_usb_add_to_isoc_sb_list(struct urb *urb, int epid) if (TxIsocEPList[epid].sub == 0) { dbg_isoc("Isoc traffic not already running, allocating SB"); - next_sb_desc = (USB_SB_Desc_t*)kmem_cache_alloc(usb_desc_cache, SLAB_ATOMIC); + next_sb_desc = (USB_SB_Desc_t*)kmem_cache_alloc(usb_desc_cache, GFP_ATOMIC); assert(next_sb_desc != NULL); next_sb_desc->command = (IO_STATE(USB_SB_command, tt, in) | @@ -3317,7 +3317,7 @@ static irqreturn_t etrax_usb_hc_interrupt_top_half(int irq, void *vhc) restore_flags(flags); - reg = (usb_interrupt_registers_t *)kmem_cache_alloc(top_half_reg_cache, SLAB_ATOMIC); + reg = (usb_interrupt_registers_t *)kmem_cache_alloc(top_half_reg_cache, GFP_ATOMIC); assert(reg != NULL); diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c index 8293c1d4be3f..0f47a57dac28 100644 --- a/drivers/usb/host/ohci-dbg.c +++ b/drivers/usb/host/ohci-dbg.c @@ -505,7 +505,7 @@ show_periodic (struct class_device *class_dev, char *buf) char *next; unsigned i; - if (!(seen = kmalloc (DBG_SCHED_LIMIT * sizeof *seen, SLAB_ATOMIC))) + if (!(seen = kmalloc (DBG_SCHED_LIMIT * sizeof *seen, GFP_ATOMIC))) return 0; seen_count = 0; diff --git a/drivers/usb/host/uhci-q.c b/drivers/usb/host/uhci-q.c index 06115f22a4fa..30b88459ac7d 100644 --- a/drivers/usb/host/uhci-q.c +++ b/drivers/usb/host/uhci-q.c @@ -498,7 +498,7 @@ static inline struct urb_priv *uhci_alloc_urb_priv(struct uhci_hcd *uhci, { struct urb_priv *urbp; - urbp = kmem_cache_alloc(uhci_up_cachep, SLAB_ATOMIC); + urbp = kmem_cache_alloc(uhci_up_cachep, GFP_ATOMIC); if (!urbp) return NULL; diff --git a/drivers/usb/input/aiptek.c b/drivers/usb/input/aiptek.c index bf428184608f..9f52429ce654 100644 --- a/drivers/usb/input/aiptek.c +++ b/drivers/usb/input/aiptek.c @@ -1988,7 +1988,7 @@ aiptek_probe(struct usb_interface *intf, const struct usb_device_id *id) goto fail1; aiptek->data = usb_buffer_alloc(usbdev, AIPTEK_PACKET_LENGTH, - SLAB_ATOMIC, &aiptek->data_dma); + GFP_ATOMIC, &aiptek->data_dma); if (!aiptek->data) goto fail1; diff --git a/drivers/usb/input/ati_remote.c b/drivers/usb/input/ati_remote.c index ff23318dc301..b724e36f7b92 100644 --- a/drivers/usb/input/ati_remote.c +++ b/drivers/usb/input/ati_remote.c @@ -592,7 +592,7 @@ static void ati_remote_irq_in(struct urb *urb) __FUNCTION__, urb->status); } - retval = usb_submit_urb(urb, SLAB_ATOMIC); + retval = usb_submit_urb(urb, GFP_ATOMIC); if (retval) dev_err(&ati_remote->interface->dev, "%s: usb_submit_urb()=%d\n", __FUNCTION__, retval); @@ -604,12 +604,12 @@ static void ati_remote_irq_in(struct urb *urb) static int ati_remote_alloc_buffers(struct usb_device *udev, struct ati_remote *ati_remote) { - ati_remote->inbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, SLAB_ATOMIC, + ati_remote->inbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, GFP_ATOMIC, &ati_remote->inbuf_dma); if (!ati_remote->inbuf) return -1; - ati_remote->outbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, SLAB_ATOMIC, + ati_remote->outbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, GFP_ATOMIC, &ati_remote->outbuf_dma); if (!ati_remote->outbuf) return -1; diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 4295bab4f1e2..f1d0e1d69828 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1079,7 +1079,7 @@ static void hid_irq_in(struct urb *urb) warn("input irq status %d received", urb->status); } - status = usb_submit_urb(urb, SLAB_ATOMIC); + status = usb_submit_urb(urb, GFP_ATOMIC); if (status) { clear_bit(HID_IN_RUNNING, &hid->iofl); if (status != -EPERM) { @@ -1864,13 +1864,13 @@ static void hid_find_max_report(struct hid_device *hid, unsigned int type, int * static int hid_alloc_buffers(struct usb_device *dev, struct hid_device *hid) { - if (!(hid->inbuf = usb_buffer_alloc(dev, hid->bufsize, SLAB_ATOMIC, &hid->inbuf_dma))) + if (!(hid->inbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->inbuf_dma))) return -1; - if (!(hid->outbuf = usb_buffer_alloc(dev, hid->bufsize, SLAB_ATOMIC, &hid->outbuf_dma))) + if (!(hid->outbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->outbuf_dma))) return -1; - if (!(hid->cr = usb_buffer_alloc(dev, sizeof(*(hid->cr)), SLAB_ATOMIC, &hid->cr_dma))) + if (!(hid->cr = usb_buffer_alloc(dev, sizeof(*(hid->cr)), GFP_ATOMIC, &hid->cr_dma))) return -1; - if (!(hid->ctrlbuf = usb_buffer_alloc(dev, hid->bufsize, SLAB_ATOMIC, &hid->ctrlbuf_dma))) + if (!(hid->ctrlbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->ctrlbuf_dma))) return -1; return 0; diff --git a/drivers/usb/input/keyspan_remote.c b/drivers/usb/input/keyspan_remote.c index 50aa8108a50b..98bd323369c7 100644 --- a/drivers/usb/input/keyspan_remote.c +++ b/drivers/usb/input/keyspan_remote.c @@ -456,7 +456,7 @@ static int keyspan_probe(struct usb_interface *interface, const struct usb_devic remote->in_endpoint = endpoint; remote->toggle = -1; /* Set to -1 so we will always not match the toggle from the first remote message. */ - remote->in_buffer = usb_buffer_alloc(udev, RECV_SIZE, SLAB_ATOMIC, &remote->in_dma); + remote->in_buffer = usb_buffer_alloc(udev, RECV_SIZE, GFP_ATOMIC, &remote->in_dma); if (!remote->in_buffer) { retval = -ENOMEM; goto fail1; diff --git a/drivers/usb/input/mtouchusb.c b/drivers/usb/input/mtouchusb.c index 79a85d46cb13..92c4e07da4c8 100644 --- a/drivers/usb/input/mtouchusb.c +++ b/drivers/usb/input/mtouchusb.c @@ -164,7 +164,7 @@ static int mtouchusb_alloc_buffers(struct usb_device *udev, struct mtouch_usb *m dbg("%s - called", __FUNCTION__); mtouch->data = usb_buffer_alloc(udev, MTOUCHUSB_REPORT_DATA_SIZE, - SLAB_ATOMIC, &mtouch->data_dma); + GFP_ATOMIC, &mtouch->data_dma); if (!mtouch->data) return -1; diff --git a/drivers/usb/input/powermate.c b/drivers/usb/input/powermate.c index 0bf91778c40d..fea97e5437f8 100644 --- a/drivers/usb/input/powermate.c +++ b/drivers/usb/input/powermate.c @@ -277,12 +277,12 @@ static int powermate_input_event(struct input_dev *dev, unsigned int type, unsig static int powermate_alloc_buffers(struct usb_device *udev, struct powermate_device *pm) { pm->data = usb_buffer_alloc(udev, POWERMATE_PAYLOAD_SIZE_MAX, - SLAB_ATOMIC, &pm->data_dma); + GFP_ATOMIC, &pm->data_dma); if (!pm->data) return -1; pm->configcr = usb_buffer_alloc(udev, sizeof(*(pm->configcr)), - SLAB_ATOMIC, &pm->configcr_dma); + GFP_ATOMIC, &pm->configcr_dma); if (!pm->configcr) return -1; diff --git a/drivers/usb/input/touchkitusb.c b/drivers/usb/input/touchkitusb.c index 05c0d1ca39ab..2a314b065922 100644 --- a/drivers/usb/input/touchkitusb.c +++ b/drivers/usb/input/touchkitusb.c @@ -248,7 +248,7 @@ static int touchkit_alloc_buffers(struct usb_device *udev, struct touchkit_usb *touchkit) { touchkit->data = usb_buffer_alloc(udev, TOUCHKIT_REPORT_DATA_SIZE, - SLAB_ATOMIC, &touchkit->data_dma); + GFP_ATOMIC, &touchkit->data_dma); if (!touchkit->data) return -1; diff --git a/drivers/usb/input/usbkbd.c b/drivers/usb/input/usbkbd.c index dac88640eab6..8505824848f6 100644 --- a/drivers/usb/input/usbkbd.c +++ b/drivers/usb/input/usbkbd.c @@ -122,7 +122,7 @@ static void usb_kbd_irq(struct urb *urb) memcpy(kbd->old, kbd->new, 8); resubmit: - i = usb_submit_urb (urb, SLAB_ATOMIC); + i = usb_submit_urb (urb, GFP_ATOMIC); if (i) err ("can't resubmit intr, %s-%s/input0, status %d", kbd->usbdev->bus->bus_name, @@ -196,11 +196,11 @@ static int usb_kbd_alloc_mem(struct usb_device *dev, struct usb_kbd *kbd) return -1; if (!(kbd->led = usb_alloc_urb(0, GFP_KERNEL))) return -1; - if (!(kbd->new = usb_buffer_alloc(dev, 8, SLAB_ATOMIC, &kbd->new_dma))) + if (!(kbd->new = usb_buffer_alloc(dev, 8, GFP_ATOMIC, &kbd->new_dma))) return -1; - if (!(kbd->cr = usb_buffer_alloc(dev, sizeof(struct usb_ctrlrequest), SLAB_ATOMIC, &kbd->cr_dma))) + if (!(kbd->cr = usb_buffer_alloc(dev, sizeof(struct usb_ctrlrequest), GFP_ATOMIC, &kbd->cr_dma))) return -1; - if (!(kbd->leds = usb_buffer_alloc(dev, 1, SLAB_ATOMIC, &kbd->leds_dma))) + if (!(kbd->leds = usb_buffer_alloc(dev, 1, GFP_ATOMIC, &kbd->leds_dma))) return -1; return 0; diff --git a/drivers/usb/input/usbmouse.c b/drivers/usb/input/usbmouse.c index 68a55642c082..64a33e420cfb 100644 --- a/drivers/usb/input/usbmouse.c +++ b/drivers/usb/input/usbmouse.c @@ -86,7 +86,7 @@ static void usb_mouse_irq(struct urb *urb) input_sync(dev); resubmit: - status = usb_submit_urb (urb, SLAB_ATOMIC); + status = usb_submit_urb (urb, GFP_ATOMIC); if (status) err ("can't resubmit intr, %s-%s/input0, status %d", mouse->usbdev->bus->bus_name, @@ -137,7 +137,7 @@ static int usb_mouse_probe(struct usb_interface *intf, const struct usb_device_i if (!mouse || !input_dev) goto fail1; - mouse->data = usb_buffer_alloc(dev, 8, SLAB_ATOMIC, &mouse->data_dma); + mouse->data = usb_buffer_alloc(dev, 8, GFP_ATOMIC, &mouse->data_dma); if (!mouse->data) goto fail1; diff --git a/drivers/usb/input/xpad.c b/drivers/usb/input/xpad.c index df97e5c803f9..e4bc76ebc835 100644 --- a/drivers/usb/input/xpad.c +++ b/drivers/usb/input/xpad.c @@ -325,7 +325,7 @@ static int xpad_probe(struct usb_interface *intf, const struct usb_device_id *id goto fail1; xpad->idata = usb_buffer_alloc(udev, XPAD_PKT_LEN, - SLAB_ATOMIC, &xpad->idata_dma); + GFP_ATOMIC, &xpad->idata_dma); if (!xpad->idata) goto fail1; diff --git a/drivers/usb/input/yealink.c b/drivers/usb/input/yealink.c index 2268ca311ade..caff8e6d7448 100644 --- a/drivers/usb/input/yealink.c +++ b/drivers/usb/input/yealink.c @@ -874,17 +874,17 @@ static int usb_probe(struct usb_interface *intf, const struct usb_device_id *id) /* allocate usb buffers */ yld->irq_data = usb_buffer_alloc(udev, USB_PKT_LEN, - SLAB_ATOMIC, &yld->irq_dma); + GFP_ATOMIC, &yld->irq_dma); if (yld->irq_data == NULL) return usb_cleanup(yld, -ENOMEM); yld->ctl_data = usb_buffer_alloc(udev, USB_PKT_LEN, - SLAB_ATOMIC, &yld->ctl_dma); + GFP_ATOMIC, &yld->ctl_dma); if (!yld->ctl_data) return usb_cleanup(yld, -ENOMEM); yld->ctl_req = usb_buffer_alloc(udev, sizeof(*(yld->ctl_req)), - SLAB_ATOMIC, &yld->ctl_req_dma); + GFP_ATOMIC, &yld->ctl_req_dma); if (yld->ctl_req == NULL) return usb_cleanup(yld, -ENOMEM); diff --git a/drivers/usb/misc/phidgetkit.c b/drivers/usb/misc/phidgetkit.c index 9659c79e187e..371bf2b1197d 100644 --- a/drivers/usb/misc/phidgetkit.c +++ b/drivers/usb/misc/phidgetkit.c @@ -377,7 +377,7 @@ static void interfacekit_irq(struct urb *urb) schedule_delayed_work(&kit->do_notify, 0); resubmit: - status = usb_submit_urb(urb, SLAB_ATOMIC); + status = usb_submit_urb(urb, GFP_ATOMIC); if (status) err("can't resubmit intr, %s-%s/interfacekit0, status %d", kit->udev->bus->bus_name, @@ -568,7 +568,7 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic kit->dev_no = -1; kit->ifkit = ifkit; - kit->data = usb_buffer_alloc(dev, URB_INT_SIZE, SLAB_ATOMIC, &kit->data_dma); + kit->data = usb_buffer_alloc(dev, URB_INT_SIZE, GFP_ATOMIC, &kit->data_dma); if (!kit->data) goto out; diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c index 2bb4fa572bb7..5727e1ea2f91 100644 --- a/drivers/usb/misc/phidgetmotorcontrol.c +++ b/drivers/usb/misc/phidgetmotorcontrol.c @@ -151,7 +151,7 @@ static void motorcontrol_irq(struct urb *urb) schedule_delayed_work(&mc->do_notify, 0); resubmit: - status = usb_submit_urb(urb, SLAB_ATOMIC); + status = usb_submit_urb(urb, GFP_ATOMIC); if (status) dev_err(&mc->intf->dev, "can't resubmit intr, %s-%s/motorcontrol0, status %d", @@ -338,7 +338,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic goto out; mc->dev_no = -1; - mc->data = usb_buffer_alloc(dev, URB_INT_SIZE, SLAB_ATOMIC, &mc->data_dma); + mc->data = usb_buffer_alloc(dev, URB_INT_SIZE, GFP_ATOMIC, &mc->data_dma); if (!mc->data) goto out; diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c index 194065dbb51f..ea04dccdc651 100644 --- a/drivers/usb/misc/usbtest.c +++ b/drivers/usb/misc/usbtest.c @@ -819,7 +819,7 @@ error: /* resubmit if we need to, else mark this as done */ if ((status == 0) && (ctx->pending < ctx->count)) { - if ((status = usb_submit_urb (urb, SLAB_ATOMIC)) != 0) { + if ((status = usb_submit_urb (urb, GFP_ATOMIC)) != 0) { dbg ("can't resubmit ctrl %02x.%02x, err %d", reqp->bRequestType, reqp->bRequest, status); urb->dev = NULL; @@ -999,7 +999,7 @@ test_ctrl_queue (struct usbtest_dev *dev, struct usbtest_param *param) context.urb = urb; spin_lock_irq (&context.lock); for (i = 0; i < param->sglen; i++) { - context.status = usb_submit_urb (urb [i], SLAB_ATOMIC); + context.status = usb_submit_urb (urb [i], GFP_ATOMIC); if (context.status != 0) { dbg ("can't submit urb[%d], status %d", i, context.status); @@ -1041,7 +1041,7 @@ static void unlink1_callback (struct urb *urb) // we "know" -EPIPE (stall) never happens if (!status) - status = usb_submit_urb (urb, SLAB_ATOMIC); + status = usb_submit_urb (urb, GFP_ATOMIC); if (status) { urb->status = status; complete ((struct completion *) urb->context); @@ -1481,7 +1481,7 @@ test_iso_queue (struct usbtest_dev *dev, struct usbtest_param *param, spin_lock_irq (&context.lock); for (i = 0; i < param->sglen; i++) { ++context.pending; - status = usb_submit_urb (urbs [i], SLAB_ATOMIC); + status = usb_submit_urb (urbs [i], GFP_ATOMIC); if (status < 0) { ERROR (dev, "submit iso[%d], error %d\n", i, status); if (i == 0) { diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c index 7a2346c53284..46aecc8754f1 100644 --- a/drivers/usb/mon/mon_text.c +++ b/drivers/usb/mon/mon_text.c @@ -147,7 +147,7 @@ static void mon_text_event(struct mon_reader_text *rp, struct urb *urb, stamp = mon_get_timestamp(); if (rp->nevents >= EVENT_MAX || - (ep = kmem_cache_alloc(rp->e_slab, SLAB_ATOMIC)) == NULL) { + (ep = kmem_cache_alloc(rp->e_slab, GFP_ATOMIC)) == NULL) { rp->r.m_bus->cnt_text_lost++; return; } @@ -188,7 +188,7 @@ static void mon_text_error(void *data, struct urb *urb, int error) struct mon_event_text *ep; if (rp->nevents >= EVENT_MAX || - (ep = kmem_cache_alloc(rp->e_slab, SLAB_ATOMIC)) == NULL) { + (ep = kmem_cache_alloc(rp->e_slab, GFP_ATOMIC)) == NULL) { rp->r.m_bus->cnt_text_lost++; return; } diff --git a/drivers/usb/net/catc.c b/drivers/usb/net/catc.c index 907b820a5faf..4852012735f6 100644 --- a/drivers/usb/net/catc.c +++ b/drivers/usb/net/catc.c @@ -345,7 +345,7 @@ static void catc_irq_done(struct urb *urb) } } resubmit: - status = usb_submit_urb (urb, SLAB_ATOMIC); + status = usb_submit_urb (urb, GFP_ATOMIC); if (status) err ("can't resubmit intr, %s-%s, status %d", catc->usbdev->bus->bus_name, diff --git a/drivers/usb/net/net1080.c b/drivers/usb/net/net1080.c index a77410562e12..493635954513 100644 --- a/drivers/usb/net/net1080.c +++ b/drivers/usb/net/net1080.c @@ -383,7 +383,7 @@ static void nc_ensure_sync(struct usbnet *dev) int status; /* Send a flush */ - urb = usb_alloc_urb(0, SLAB_ATOMIC); + urb = usb_alloc_urb(0, GFP_ATOMIC); if (!urb) return; diff --git a/drivers/usb/net/pegasus.c b/drivers/usb/net/pegasus.c index b5690b3834e3..d48c024cff59 100644 --- a/drivers/usb/net/pegasus.c +++ b/drivers/usb/net/pegasus.c @@ -856,7 +856,7 @@ static void intr_callback(struct urb *urb) pegasus->stats.rx_missed_errors += ((d[3] & 0x7f) << 8) | d[4]; } - status = usb_submit_urb(urb, SLAB_ATOMIC); + status = usb_submit_urb(urb, GFP_ATOMIC); if (status == -ENODEV) netif_device_detach(pegasus->net); if (status && netif_msg_timer(pegasus)) diff --git a/drivers/usb/net/rtl8150.c b/drivers/usb/net/rtl8150.c index 72171f94ded4..c54235f73cb6 100644 --- a/drivers/usb/net/rtl8150.c +++ b/drivers/usb/net/rtl8150.c @@ -587,7 +587,7 @@ static void intr_callback(struct urb *urb) } resubmit: - status = usb_submit_urb (urb, SLAB_ATOMIC); + status = usb_submit_urb (urb, GFP_ATOMIC); if (status == -ENODEV) netif_device_detach(dev->netdev); else if (status) diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c index 82cd15b894b0..70f93b18292f 100644 --- a/drivers/usb/serial/mos7720.c +++ b/drivers/usb/serial/mos7720.c @@ -363,7 +363,7 @@ static int mos7720_open(struct usb_serial_port *port, struct file * filp) /* Initialising the write urb pool */ for (j = 0; j < NUM_URBS; ++j) { - urb = usb_alloc_urb(0,SLAB_ATOMIC); + urb = usb_alloc_urb(0,GFP_ATOMIC); mos7720_port->write_urb_pool[j] = urb; if (urb == NULL) { diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c index 02c89e10b2cf..5432c6340086 100644 --- a/drivers/usb/serial/mos7840.c +++ b/drivers/usb/serial/mos7840.c @@ -826,7 +826,7 @@ static int mos7840_open(struct usb_serial_port *port, struct file *filp) /* Initialising the write urb pool */ for (j = 0; j < NUM_URBS; ++j) { - urb = usb_alloc_urb(0, SLAB_ATOMIC); + urb = usb_alloc_urb(0, GFP_ATOMIC); mos7840_port->write_urb_pool[j] = urb; if (urb == NULL) { @@ -2786,7 +2786,7 @@ static int mos7840_startup(struct usb_serial *serial) i + 1, status); } - mos7840_port->control_urb = usb_alloc_urb(0, SLAB_ATOMIC); + mos7840_port->control_urb = usb_alloc_urb(0, GFP_ATOMIC); mos7840_port->ctrl_buf = kmalloc(16, GFP_KERNEL); } diff --git a/drivers/usb/storage/onetouch.c b/drivers/usb/storage/onetouch.c index 3a158d58441f..e565d3d2ab29 100644 --- a/drivers/usb/storage/onetouch.c +++ b/drivers/usb/storage/onetouch.c @@ -76,7 +76,7 @@ static void usb_onetouch_irq(struct urb *urb) input_sync(dev); resubmit: - status = usb_submit_urb (urb, SLAB_ATOMIC); + status = usb_submit_urb (urb, GFP_ATOMIC); if (status) err ("can't resubmit intr, %s-%s/input0, status %d", onetouch->udev->bus->bus_name, @@ -154,7 +154,7 @@ int onetouch_connect_input(struct us_data *ss) goto fail1; onetouch->data = usb_buffer_alloc(udev, ONETOUCH_PKT_LEN, - SLAB_ATOMIC, &onetouch->data_dma); + GFP_ATOMIC, &onetouch->data_dma); if (!onetouch->data) goto fail1; diff --git a/include/linux/slab.h b/include/linux/slab.h index d7ee28e51330..34b046ea88f1 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -19,7 +19,6 @@ typedef struct kmem_cache kmem_cache_t; #include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ /* flags for kmem_cache_alloc() */ -#define SLAB_ATOMIC GFP_ATOMIC #define SLAB_KERNEL GFP_KERNEL #define SLAB_DMA GFP_DMA diff --git a/include/net/request_sock.h b/include/net/request_sock.h index e37baaf2080b..426f0fe774ef 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -60,7 +60,7 @@ struct request_sock { static inline struct request_sock *reqsk_alloc(const struct request_sock_ops *ops) { - struct request_sock *req = kmem_cache_alloc(ops->slab, SLAB_ATOMIC); + struct request_sock *req = kmem_cache_alloc(ops->slab, GFP_ATOMIC); if (req != NULL) req->rsk_ops = ops; diff --git a/net/core/dst.c b/net/core/dst.c index 1a5e49da0e77..836ec6606925 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -125,7 +125,7 @@ void * dst_alloc(struct dst_ops * ops) if (ops->gc()) return NULL; } - dst = kmem_cache_alloc(ops->kmem_cachep, SLAB_ATOMIC); + dst = kmem_cache_alloc(ops->kmem_cachep, GFP_ATOMIC); if (!dst) return NULL; memset(dst, 0, ops->entry_size); diff --git a/net/core/flow.c b/net/core/flow.c index b16d31ae5e54..5df3e297f817 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -211,7 +211,7 @@ void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir, if (flow_count(cpu) > flow_hwm) flow_cache_shrink(cpu); - fle = kmem_cache_alloc(flow_cachep, SLAB_ATOMIC); + fle = kmem_cache_alloc(flow_cachep, GFP_ATOMIC); if (fle) { fle->next = *head; *head = fle; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index ba509a4a8e92..0ab1987b9348 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -251,7 +251,7 @@ static struct neighbour *neigh_alloc(struct neigh_table *tbl) goto out_entries; } - n = kmem_cache_alloc(tbl->kmem_cachep, SLAB_ATOMIC); + n = kmem_cache_alloc(tbl->kmem_cachep, GFP_ATOMIC); if (!n) goto out_entries; diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index cf8c07b2704f..66a27b9688ca 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -295,7 +295,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) new_packet = dccp_tx_hist_head(&hctx->ccid3hctx_hist); if (new_packet == NULL || new_packet->dccphtx_sent) { new_packet = dccp_tx_hist_entry_new(ccid3_tx_hist, - SLAB_ATOMIC); + GFP_ATOMIC); if (unlikely(new_packet == NULL)) { DCCP_WARN("%s, sk=%p, not enough mem to add to history," @@ -889,7 +889,7 @@ static void ccid3_hc_rx_update_li(struct sock *sk, u64 seq_loss, u8 win_loss) /* new loss event detected */ /* calculate last interval length */ seq_temp = dccp_delta_seqno(head->dccplih_seqno, seq_loss); - entry = dccp_li_hist_entry_new(ccid3_li_hist, SLAB_ATOMIC); + entry = dccp_li_hist_entry_new(ccid3_li_hist, GFP_ATOMIC); if (entry == NULL) { DCCP_BUG("out of memory - can not allocate entry"); @@ -1011,7 +1011,7 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) } packet = dccp_rx_hist_entry_new(ccid3_rx_hist, sk, opt_recv->dccpor_ndp, - skb, SLAB_ATOMIC); + skb, GFP_ATOMIC); if (unlikely(packet == NULL)) { DCCP_WARN("%s, sk=%p, Not enough mem to add rx packet " "to history, consider it lost!\n", dccp_role(sk), sk); diff --git a/net/dccp/ccids/lib/loss_interval.c b/net/dccp/ccids/lib/loss_interval.c index 48b9b93f8acb..0a0baef16b3e 100644 --- a/net/dccp/ccids/lib/loss_interval.c +++ b/net/dccp/ccids/lib/loss_interval.c @@ -125,7 +125,7 @@ int dccp_li_hist_interval_new(struct dccp_li_hist *hist, int i; for (i = 0; i < DCCP_LI_HIST_IVAL_F_LENGTH; i++) { - entry = dccp_li_hist_entry_new(hist, SLAB_ATOMIC); + entry = dccp_li_hist_entry_new(hist, GFP_ATOMIC); if (entry == NULL) { dccp_li_hist_purge(hist, list); DCCP_BUG("loss interval list entry is NULL"); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 244c4f445c7d..bd6c9bc41893 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -31,7 +31,7 @@ struct inet_bind_bucket *inet_bind_bucket_create(kmem_cache_t *cachep, struct inet_bind_hashbucket *head, const unsigned short snum) { - struct inet_bind_bucket *tb = kmem_cache_alloc(cachep, SLAB_ATOMIC); + struct inet_bind_bucket *tb = kmem_cache_alloc(cachep, GFP_ATOMIC); if (tb != NULL) { tb->port = snum; diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index 8c74f9168b7d..e28330aa4139 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -91,7 +91,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, const int stat { struct inet_timewait_sock *tw = kmem_cache_alloc(sk->sk_prot_creator->twsk_prot->twsk_slab, - SLAB_ATOMIC); + GFP_ATOMIC); if (tw != NULL) { const struct inet_sock *inet = inet_sk(sk); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index bf526115e518..97a8cfbb61a1 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -150,7 +150,7 @@ static __inline__ struct fib6_node * node_alloc(void) { struct fib6_node *fn; - if ((fn = kmem_cache_alloc(fib6_node_kmem, SLAB_ATOMIC)) != NULL) + if ((fn = kmem_cache_alloc(fib6_node_kmem, GFP_ATOMIC)) != NULL) memset(fn, 0, sizeof(struct fib6_node)); return fn; diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index 01a5c52a2be3..d4f68b0f27d5 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -180,7 +180,7 @@ try_next_2:; spi = 0; goto out; alloc_spi: - x6spi = kmem_cache_alloc(xfrm6_tunnel_spi_kmem, SLAB_ATOMIC); + x6spi = kmem_cache_alloc(xfrm6_tunnel_spi_kmem, GFP_ATOMIC); if (!x6spi) goto out; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 04954e5f6846..8d55d10041f9 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -979,7 +979,7 @@ struct sctp_chunk *sctp_chunkify(struct sk_buff *skb, { struct sctp_chunk *retval; - retval = kmem_cache_alloc(sctp_chunk_cachep, SLAB_ATOMIC); + retval = kmem_cache_alloc(sctp_chunk_cachep, GFP_ATOMIC); if (!retval) goto nodata; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 02b27145b279..49607792cbd3 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4989,7 +4989,7 @@ static struct sctp_bind_bucket *sctp_bucket_create( { struct sctp_bind_bucket *pp; - pp = kmem_cache_alloc(sctp_bucket_cachep, SLAB_ATOMIC); + pp = kmem_cache_alloc(sctp_bucket_cachep, GFP_ATOMIC); SCTP_DBG_OBJCNT_INC(bind_bucket); if (pp) { pp->port = snum; diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index e8198a2c785d..a898a6a83a56 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -27,7 +27,7 @@ struct sec_path *secpath_dup(struct sec_path *src) { struct sec_path *sp; - sp = kmem_cache_alloc(secpath_cachep, SLAB_ATOMIC); + sp = kmem_cache_alloc(secpath_cachep, GFP_ATOMIC); if (!sp) return NULL; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index e73ac1ab7cfd..65b4ec9c699a 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -332,7 +332,7 @@ static struct avc_node *avc_alloc_node(void) { struct avc_node *node; - node = kmem_cache_alloc(avc_node_cachep, SLAB_ATOMIC); + node = kmem_cache_alloc(avc_node_cachep, GFP_ATOMIC); if (!node) goto out; -- cgit v1.2.3 From e94b1766097d53e6f3ccfb36c8baa562ffeda3fc Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:17 -0800 Subject: [PATCH] slab: remove SLAB_KERNEL SLAB_KERNEL is an alias of GFP_KERNEL. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/sysenter.c | 2 +- arch/ia64/ia32/binfmt_elf32.c | 8 ++++---- arch/ia64/kernel/perfmon.c | 2 +- arch/ia64/mm/init.c | 4 ++-- arch/powerpc/kernel/vdso.c | 2 +- arch/powerpc/platforms/cell/spufs/inode.c | 2 +- arch/sh/kernel/vsyscall/vsyscall.c | 2 +- arch/x86_64/ia32/ia32_binfmt.c | 2 +- arch/x86_64/ia32/syscall32.c | 2 +- drivers/atm/he.c | 4 ++-- drivers/base/dmapool.c | 2 +- drivers/dma/ioatdma.c | 4 ++-- drivers/ieee1394/hosts.c | 2 +- drivers/ieee1394/ohci1394.c | 8 ++++---- drivers/ieee1394/pcilynx.c | 2 +- drivers/ieee1394/raw1394.c | 10 +++++----- drivers/infiniband/hw/ehca/ehca_av.c | 2 +- drivers/infiniband/hw/ehca/ehca_cq.c | 2 +- drivers/infiniband/hw/ehca/ehca_main.c | 2 +- drivers/infiniband/hw/ehca/ehca_mrmw.c | 4 ++-- drivers/infiniband/hw/ehca/ehca_pd.c | 2 +- drivers/infiniband/hw/ehca/ehca_qp.c | 2 +- drivers/input/touchscreen/ads7846.c | 2 +- drivers/isdn/gigaset/bas-gigaset.c | 14 +++++++------- drivers/isdn/gigaset/usb-gigaset.c | 6 +++--- drivers/media/dvb/cinergyT2/cinergyT2.c | 2 +- drivers/mtd/devices/m25p80.c | 2 +- drivers/scsi/ipr.c | 2 +- drivers/spi/spi.c | 4 ++-- drivers/spi/spi_bitbang.c | 2 +- drivers/usb/core/hub.c | 4 ++-- drivers/usb/gadget/gmidi.c | 2 +- drivers/usb/gadget/goku_udc.c | 2 +- drivers/usb/gadget/inode.c | 6 +++--- drivers/usb/gadget/net2280.c | 2 +- drivers/usb/gadget/omap_udc.c | 2 +- drivers/usb/gadget/zero.c | 2 +- drivers/usb/host/hc_crisv10.c | 2 +- drivers/usb/host/ohci-pnx4008.c | 2 +- drivers/usb/input/acecad.c | 2 +- drivers/usb/input/usbtouchscreen.c | 2 +- drivers/usb/misc/usbtest.c | 28 ++++++++++++++-------------- drivers/usb/net/rndis_host.c | 2 +- drivers/usb/net/usbnet.c | 4 ++-- fs/adfs/super.c | 2 +- fs/affs/super.c | 2 +- fs/afs/super.c | 2 +- fs/befs/linuxvfs.c | 2 +- fs/bfs/inode.c | 2 +- fs/block_dev.c | 2 +- fs/cifs/cifsfs.c | 2 +- fs/cifs/misc.c | 4 ++-- fs/cifs/transport.c | 4 ++-- fs/coda/inode.c | 2 +- fs/dnotify.c | 2 +- fs/ecryptfs/crypto.c | 2 +- fs/ecryptfs/file.c | 2 +- fs/ecryptfs/inode.c | 4 ++-- fs/ecryptfs/keystore.c | 2 +- fs/ecryptfs/main.c | 4 ++-- fs/ecryptfs/super.c | 2 +- fs/efs/super.c | 2 +- fs/eventpoll.c | 4 ++-- fs/exec.c | 2 +- fs/ext2/super.c | 2 +- fs/fat/cache.c | 2 +- fs/fat/inode.c | 2 +- fs/fcntl.c | 2 +- fs/freevxfs/vxfs_inode.c | 4 ++-- fs/fuse/dev.c | 2 +- fs/fuse/inode.c | 2 +- fs/hfs/super.c | 2 +- fs/hfsplus/super.c | 2 +- fs/hugetlbfs/inode.c | 2 +- fs/inode.c | 2 +- fs/isofs/inode.c | 2 +- fs/jffs2/super.c | 2 +- fs/locks.c | 2 +- fs/minix/inode.c | 2 +- fs/ncpfs/inode.c | 2 +- fs/nfs/direct.c | 2 +- fs/nfs/inode.c | 2 +- fs/nfs/pagelist.c | 2 +- fs/openpromfs/inode.c | 2 +- fs/proc/inode.c | 2 +- fs/qnx4/inode.c | 2 +- fs/reiserfs/super.c | 2 +- fs/romfs/inode.c | 2 +- fs/smbfs/inode.c | 2 +- fs/smbfs/request.c | 2 +- fs/sysv/inode.c | 2 +- fs/udf/super.c | 2 +- fs/ufs/super.c | 2 +- include/linux/fs.h | 2 +- include/linux/rmap.h | 2 +- include/linux/slab.h | 1 - include/linux/taskstats_kern.h | 2 +- ipc/mqueue.c | 2 +- kernel/delayacct.c | 2 +- kernel/fork.c | 6 +++--- kernel/taskstats.c | 2 +- kernel/user.c | 2 +- mm/mempolicy.c | 2 +- mm/mmap.c | 4 ++-- mm/shmem.c | 2 +- mm/slab.c | 2 +- net/decnet/dn_table.c | 2 +- net/ipv4/fib_hash.c | 4 ++-- net/ipv4/fib_trie.c | 4 ++-- net/socket.c | 2 +- net/sunrpc/rpc_pipe.c | 2 +- security/keys/key.c | 2 +- security/selinux/hooks.c | 2 +- security/selinux/ss/avtab.c | 2 +- 114 files changed, 164 insertions(+), 165 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 713ba39d32c6..0bbacd0ec175 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -132,7 +132,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) goto up_fail; } - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) { ret = -ENOMEM; goto up_fail; diff --git a/arch/ia64/ia32/binfmt_elf32.c b/arch/ia64/ia32/binfmt_elf32.c index daa6b91bc921..578737ec7629 100644 --- a/arch/ia64/ia32/binfmt_elf32.c +++ b/arch/ia64/ia32/binfmt_elf32.c @@ -91,7 +91,7 @@ ia64_elf32_init (struct pt_regs *regs) * it with privilege level 3 because the IVE uses non-privileged accesses to these * tables. IA-32 segmentation is used to protect against IA-32 accesses to them. */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -117,7 +117,7 @@ ia64_elf32_init (struct pt_regs *regs) * code is locked in specific gate page, which is pointed by pretcode * when setup_frame_ia32 */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -142,7 +142,7 @@ ia64_elf32_init (struct pt_regs *regs) * Install LDT as anonymous memory. This gives us all-zero segment descriptors * until a task modifies them via modify_ldt(). */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -214,7 +214,7 @@ ia32_setup_arg_pages (struct linux_binprm *bprm, int executable_stack) bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 3aaede0d6981..e2321536ee4c 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2302,7 +2302,7 @@ pfm_smpl_buffer_alloc(struct task_struct *task, pfm_context_t *ctx, unsigned lon DPRINT(("smpl_buf @%p\n", smpl_buf)); /* allocate vma */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!vma) { DPRINT(("Cannot allocate vma\n")); goto error_kmem; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ff87a5cba399..56dc2024220e 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -156,7 +156,7 @@ ia64_init_addr_space (void) * the problem. When the process attempts to write to the register backing store * for the first time, it will get a SEGFAULT in this case. */ - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; @@ -175,7 +175,7 @@ ia64_init_addr_space (void) /* map NaT-page at address zero to speed up speculative dereferencing of NULL: */ if (!(current->personality & MMAP_PAGE_ZERO)) { - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (vma) { memset(vma, 0, sizeof(*vma)); vma->vm_mm = current->mm; diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index c913ad5cad29..a4b28c73bba0 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -264,7 +264,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, /* Allocate a VMA structure and fill it up */ - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (vma == NULL) { rc = -ENOMEM; goto fail_mmapsem; diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index c7d010749a18..7edfcc9d2853 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -48,7 +48,7 @@ spufs_alloc_inode(struct super_block *sb) { struct spufs_inode_info *ei; - ei = kmem_cache_alloc(spufs_inode_cache, SLAB_KERNEL); + ei = kmem_cache_alloc(spufs_inode_cache, GFP_KERNEL); if (!ei) return NULL; diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c index 075d6cc1a2d7..deb46941f315 100644 --- a/arch/sh/kernel/vsyscall/vsyscall.c +++ b/arch/sh/kernel/vsyscall/vsyscall.c @@ -97,7 +97,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, goto up_fail; } - vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) { ret = -ENOMEM; goto up_fail; diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 82ef182de6ae..932a62ad6c83 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -351,7 +351,7 @@ int ia32_setup_arg_pages(struct linux_binprm *bprm, unsigned long stack_top, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index 3a01329473ab..3e5ed20cba45 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -49,7 +49,7 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) struct mm_struct *mm = current->mm; int ret; - vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!vma) return -ENOMEM; diff --git a/drivers/atm/he.c b/drivers/atm/he.c index 2a2f0fc2288f..ec8a7a633e68 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -820,7 +820,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPS_POOL - cpuaddr = pci_pool_alloc(he_dev->rbps_pool, SLAB_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbps_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else @@ -884,7 +884,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPL_POOL - cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, SLAB_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index fa4675254f67..dbe0735f8c9e 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -126,7 +126,7 @@ dma_pool_create (const char *name, struct device *dev, } else if (allocation < size) return NULL; - if (!(retval = kmalloc (sizeof *retval, SLAB_KERNEL))) + if (!(retval = kmalloc (sizeof *retval, GFP_KERNEL))) return retval; strlcpy (retval->name, name, sizeof retval->name); diff --git a/drivers/dma/ioatdma.c b/drivers/dma/ioatdma.c index 0358419a0e48..8e8726104619 100644 --- a/drivers/dma/ioatdma.c +++ b/drivers/dma/ioatdma.c @@ -636,10 +636,10 @@ static int ioat_self_test(struct ioat_device *device) dma_cookie_t cookie; int err = 0; - src = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL); + src = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, GFP_KERNEL); if (!src) return -ENOMEM; - dest = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL); + dest = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, GFP_KERNEL); if (!dest) { kfree(src); return -ENOMEM; diff --git a/drivers/ieee1394/hosts.c b/drivers/ieee1394/hosts.c index 8f4378a1631c..b935e08695a9 100644 --- a/drivers/ieee1394/hosts.c +++ b/drivers/ieee1394/hosts.c @@ -123,7 +123,7 @@ struct hpsb_host *hpsb_alloc_host(struct hpsb_host_driver *drv, size_t extra, int i; int hostnum = 0; - h = kzalloc(sizeof(*h) + extra, SLAB_KERNEL); + h = kzalloc(sizeof(*h) + extra, GFP_KERNEL); if (!h) return NULL; diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c index 6e8ea9110c46..eae97d8dcf03 100644 --- a/drivers/ieee1394/ohci1394.c +++ b/drivers/ieee1394/ohci1394.c @@ -1225,7 +1225,7 @@ static int ohci_iso_recv_init(struct hpsb_iso *iso) int ctx; int ret = -ENOMEM; - recv = kmalloc(sizeof(*recv), SLAB_KERNEL); + recv = kmalloc(sizeof(*recv), GFP_KERNEL); if (!recv) return -ENOMEM; @@ -1918,7 +1918,7 @@ static int ohci_iso_xmit_init(struct hpsb_iso *iso) int ctx; int ret = -ENOMEM; - xmit = kmalloc(sizeof(*xmit), SLAB_KERNEL); + xmit = kmalloc(sizeof(*xmit), GFP_KERNEL); if (!xmit) return -ENOMEM; @@ -3021,7 +3021,7 @@ alloc_dma_rcv_ctx(struct ti_ohci *ohci, struct dma_rcv_ctx *d, return -ENOMEM; } - d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, SLAB_KERNEL, d->prg_bus+i); + d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, GFP_KERNEL, d->prg_bus+i); OHCI_DMA_ALLOC("pool dma_rcv prg[%d]", i); if (d->prg_cpu[i] != NULL) { @@ -3117,7 +3117,7 @@ alloc_dma_trm_ctx(struct ti_ohci *ohci, struct dma_trm_ctx *d, OHCI_DMA_ALLOC("dma_rcv prg pool"); for (i = 0; i < d->num_desc; i++) { - d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, SLAB_KERNEL, d->prg_bus+i); + d->prg_cpu[i] = pci_pool_alloc(d->prg_pool, GFP_KERNEL, d->prg_bus+i); OHCI_DMA_ALLOC("pool dma_trm prg[%d]", i); if (d->prg_cpu[i] != NULL) { diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c index 0a7412e27eb4..9cab1d661472 100644 --- a/drivers/ieee1394/pcilynx.c +++ b/drivers/ieee1394/pcilynx.c @@ -1428,7 +1428,7 @@ static int __devinit add_card(struct pci_dev *dev, struct i2c_algo_bit_data i2c_adapter_data; error = -ENOMEM; - i2c_ad = kmalloc(sizeof(*i2c_ad), SLAB_KERNEL); + i2c_ad = kmalloc(sizeof(*i2c_ad), GFP_KERNEL); if (!i2c_ad) FAIL("failed to allocate I2C adapter memory"); memcpy(i2c_ad, &bit_ops, sizeof(struct i2c_adapter)); diff --git a/drivers/ieee1394/raw1394.c b/drivers/ieee1394/raw1394.c index 47f6a4e29b40..bf71e069eaf5 100644 --- a/drivers/ieee1394/raw1394.c +++ b/drivers/ieee1394/raw1394.c @@ -112,7 +112,7 @@ static struct pending_request *__alloc_pending_request(gfp_t flags) static inline struct pending_request *alloc_pending_request(void) { - return __alloc_pending_request(SLAB_KERNEL); + return __alloc_pending_request(GFP_KERNEL); } static void free_pending_request(struct pending_request *req) @@ -1737,7 +1737,7 @@ static int arm_register(struct file_info *fi, struct pending_request *req) return (-EINVAL); } /* addr-list-entry for fileinfo */ - addr = kmalloc(sizeof(*addr), SLAB_KERNEL); + addr = kmalloc(sizeof(*addr), GFP_KERNEL); if (!addr) { req->req.length = 0; return (-ENOMEM); @@ -2103,7 +2103,7 @@ static int write_phypacket(struct file_info *fi, struct pending_request *req) static int get_config_rom(struct file_info *fi, struct pending_request *req) { int ret = sizeof(struct raw1394_request); - quadlet_t *data = kmalloc(req->req.length, SLAB_KERNEL); + quadlet_t *data = kmalloc(req->req.length, GFP_KERNEL); int status; if (!data) @@ -2133,7 +2133,7 @@ static int get_config_rom(struct file_info *fi, struct pending_request *req) static int update_config_rom(struct file_info *fi, struct pending_request *req) { int ret = sizeof(struct raw1394_request); - quadlet_t *data = kmalloc(req->req.length, SLAB_KERNEL); + quadlet_t *data = kmalloc(req->req.length, GFP_KERNEL); if (!data) return -ENOMEM; if (copy_from_user(data, int2ptr(req->req.sendb), req->req.length)) { @@ -2779,7 +2779,7 @@ static int raw1394_open(struct inode *inode, struct file *file) { struct file_info *fi; - fi = kzalloc(sizeof(*fi), SLAB_KERNEL); + fi = kzalloc(sizeof(*fi), GFP_KERNEL); if (!fi) return -ENOMEM; diff --git a/drivers/infiniband/hw/ehca/ehca_av.c b/drivers/infiniband/hw/ehca/ehca_av.c index 214e2fdddeef..0d6e2c4bb245 100644 --- a/drivers/infiniband/hw/ehca/ehca_av.c +++ b/drivers/infiniband/hw/ehca/ehca_av.c @@ -57,7 +57,7 @@ struct ib_ah *ehca_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr) struct ehca_shca *shca = container_of(pd->device, struct ehca_shca, ib_device); - av = kmem_cache_alloc(av_cache, SLAB_KERNEL); + av = kmem_cache_alloc(av_cache, GFP_KERNEL); if (!av) { ehca_err(pd->device, "Out of memory pd=%p ah_attr=%p", pd, ah_attr); diff --git a/drivers/infiniband/hw/ehca/ehca_cq.c b/drivers/infiniband/hw/ehca/ehca_cq.c index 458fe19648a1..93995b658d94 100644 --- a/drivers/infiniband/hw/ehca/ehca_cq.c +++ b/drivers/infiniband/hw/ehca/ehca_cq.c @@ -134,7 +134,7 @@ struct ib_cq *ehca_create_cq(struct ib_device *device, int cqe, if (cqe >= 0xFFFFFFFF - 64 - additional_cqe) return ERR_PTR(-EINVAL); - my_cq = kmem_cache_alloc(cq_cache, SLAB_KERNEL); + my_cq = kmem_cache_alloc(cq_cache, GFP_KERNEL); if (!my_cq) { ehca_err(device, "Out of memory for ehca_cq struct device=%p", device); diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c index 3d1c1c535038..cc47e4c13a18 100644 --- a/drivers/infiniband/hw/ehca/ehca_main.c +++ b/drivers/infiniband/hw/ehca/ehca_main.c @@ -108,7 +108,7 @@ static struct kmem_cache *ctblk_cache = NULL; void *ehca_alloc_fw_ctrlblock(void) { - void *ret = kmem_cache_zalloc(ctblk_cache, SLAB_KERNEL); + void *ret = kmem_cache_zalloc(ctblk_cache, GFP_KERNEL); if (!ret) ehca_gen_err("Out of memory for ctblk"); return ret; diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.c b/drivers/infiniband/hw/ehca/ehca_mrmw.c index abce676c0ae0..0a5e2214cc5f 100644 --- a/drivers/infiniband/hw/ehca/ehca_mrmw.c +++ b/drivers/infiniband/hw/ehca/ehca_mrmw.c @@ -53,7 +53,7 @@ static struct ehca_mr *ehca_mr_new(void) { struct ehca_mr *me; - me = kmem_cache_alloc(mr_cache, SLAB_KERNEL); + me = kmem_cache_alloc(mr_cache, GFP_KERNEL); if (me) { memset(me, 0, sizeof(struct ehca_mr)); spin_lock_init(&me->mrlock); @@ -72,7 +72,7 @@ static struct ehca_mw *ehca_mw_new(void) { struct ehca_mw *me; - me = kmem_cache_alloc(mw_cache, SLAB_KERNEL); + me = kmem_cache_alloc(mw_cache, GFP_KERNEL); if (me) { memset(me, 0, sizeof(struct ehca_mw)); spin_lock_init(&me->mwlock); diff --git a/drivers/infiniband/hw/ehca/ehca_pd.c b/drivers/infiniband/hw/ehca/ehca_pd.c index 2c3cdc6f7b39..d5345e5b3cd6 100644 --- a/drivers/infiniband/hw/ehca/ehca_pd.c +++ b/drivers/infiniband/hw/ehca/ehca_pd.c @@ -50,7 +50,7 @@ struct ib_pd *ehca_alloc_pd(struct ib_device *device, { struct ehca_pd *pd; - pd = kmem_cache_alloc(pd_cache, SLAB_KERNEL); + pd = kmem_cache_alloc(pd_cache, GFP_KERNEL); if (!pd) { ehca_err(device, "device=%p context=%p out of memory", device, context); diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c index 8682aa50c707..c6c9cef203e3 100644 --- a/drivers/infiniband/hw/ehca/ehca_qp.c +++ b/drivers/infiniband/hw/ehca/ehca_qp.c @@ -450,7 +450,7 @@ struct ib_qp *ehca_create_qp(struct ib_pd *pd, if (pd->uobject && udata) context = pd->uobject->context; - my_qp = kmem_cache_alloc(qp_cache, SLAB_KERNEL); + my_qp = kmem_cache_alloc(qp_cache, GFP_KERNEL); if (!my_qp) { ehca_err(pd->device, "pd=%p not enough memory to alloc qp", pd); return ERR_PTR(-ENOMEM); diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c index f56d6a0f0624..0517c7387d67 100644 --- a/drivers/input/touchscreen/ads7846.c +++ b/drivers/input/touchscreen/ads7846.c @@ -189,7 +189,7 @@ static int ads7846_read12_ser(struct device *dev, unsigned command) { struct spi_device *spi = to_spi_device(dev); struct ads7846 *ts = dev_get_drvdata(dev); - struct ser_req *req = kzalloc(sizeof *req, SLAB_KERNEL); + struct ser_req *req = kzalloc(sizeof *req, GFP_KERNEL); int status; int sample; int i; diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c index 5857e7e23f84..63b629b1cdb2 100644 --- a/drivers/isdn/gigaset/bas-gigaset.c +++ b/drivers/isdn/gigaset/bas-gigaset.c @@ -2218,21 +2218,21 @@ static int gigaset_probe(struct usb_interface *interface, * - three for the different uses of the default control pipe * - three for each isochronous pipe */ - if (!(ucs->urb_int_in = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_cmd_in = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_cmd_out = usb_alloc_urb(0, SLAB_KERNEL)) || - !(ucs->urb_ctrl = usb_alloc_urb(0, SLAB_KERNEL))) + if (!(ucs->urb_int_in = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_cmd_in = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_cmd_out = usb_alloc_urb(0, GFP_KERNEL)) || + !(ucs->urb_ctrl = usb_alloc_urb(0, GFP_KERNEL))) goto allocerr; for (j = 0; j < 2; ++j) { ubc = cs->bcs[j].hw.bas; for (i = 0; i < BAS_OUTURBS; ++i) if (!(ubc->isoouturbs[i].urb = - usb_alloc_urb(BAS_NUMFRAMES, SLAB_KERNEL))) + usb_alloc_urb(BAS_NUMFRAMES, GFP_KERNEL))) goto allocerr; for (i = 0; i < BAS_INURBS; ++i) if (!(ubc->isoinurbs[i] = - usb_alloc_urb(BAS_NUMFRAMES, SLAB_KERNEL))) + usb_alloc_urb(BAS_NUMFRAMES, GFP_KERNEL))) goto allocerr; } @@ -2246,7 +2246,7 @@ static int gigaset_probe(struct usb_interface *interface, (endpoint->bEndpointAddress) & 0x0f), ucs->int_in_buf, 3, read_int_callback, cs, endpoint->bInterval); - if ((rc = usb_submit_urb(ucs->urb_int_in, SLAB_KERNEL)) != 0) { + if ((rc = usb_submit_urb(ucs->urb_int_in, GFP_KERNEL)) != 0) { dev_err(cs->dev, "could not submit interrupt URB: %s\n", get_usb_rcmsg(rc)); goto error; diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c index af89ce188f2a..04f2ad7ba8b0 100644 --- a/drivers/isdn/gigaset/usb-gigaset.c +++ b/drivers/isdn/gigaset/usb-gigaset.c @@ -763,7 +763,7 @@ static int gigaset_probe(struct usb_interface *interface, goto error; } - ucs->bulk_out_urb = usb_alloc_urb(0, SLAB_KERNEL); + ucs->bulk_out_urb = usb_alloc_urb(0, GFP_KERNEL); if (!ucs->bulk_out_urb) { dev_err(cs->dev, "Couldn't allocate bulk_out_urb\n"); retval = -ENOMEM; @@ -774,7 +774,7 @@ static int gigaset_probe(struct usb_interface *interface, atomic_set(&ucs->busy, 0); - ucs->read_urb = usb_alloc_urb(0, SLAB_KERNEL); + ucs->read_urb = usb_alloc_urb(0, GFP_KERNEL); if (!ucs->read_urb) { dev_err(cs->dev, "No free urbs available\n"); retval = -ENOMEM; @@ -797,7 +797,7 @@ static int gigaset_probe(struct usb_interface *interface, gigaset_read_int_callback, cs->inbuf + 0, endpoint->bInterval); - retval = usb_submit_urb(ucs->read_urb, SLAB_KERNEL); + retval = usb_submit_urb(ucs->read_urb, GFP_KERNEL); if (retval) { dev_err(cs->dev, "Could not submit URB (error %d)\n", -retval); goto error; diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c index 206c13e47a06..9123147e376f 100644 --- a/drivers/media/dvb/cinergyT2/cinergyT2.c +++ b/drivers/media/dvb/cinergyT2/cinergyT2.c @@ -287,7 +287,7 @@ static int cinergyt2_alloc_stream_urbs (struct cinergyt2 *cinergyt2) int i; cinergyt2->streambuf = usb_buffer_alloc(cinergyt2->udev, STREAM_URB_COUNT*STREAM_BUF_SIZE, - SLAB_KERNEL, &cinergyt2->streambuf_dmahandle); + GFP_KERNEL, &cinergyt2->streambuf_dmahandle); if (!cinergyt2->streambuf) { dprintk(1, "failed to alloc consistent stream memory area, bailing out!\n"); return -ENOMEM; diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c index ef4a731ca5c2..334e078ffaff 100644 --- a/drivers/mtd/devices/m25p80.c +++ b/drivers/mtd/devices/m25p80.c @@ -451,7 +451,7 @@ static int __devinit m25p_probe(struct spi_device *spi) return -ENODEV; } - flash = kzalloc(sizeof *flash, SLAB_KERNEL); + flash = kzalloc(sizeof *flash, GFP_KERNEL); if (!flash) return -ENOMEM; diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index ccd4dafce8e2..b318500785e5 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -6940,7 +6940,7 @@ static int __devinit ipr_alloc_cmd_blks(struct ipr_ioa_cfg *ioa_cfg) return -ENOMEM; for (i = 0; i < IPR_NUM_CMD_BLKS; i++) { - ipr_cmd = pci_pool_alloc (ioa_cfg->ipr_cmd_pool, SLAB_KERNEL, &dma_addr); + ipr_cmd = pci_pool_alloc (ioa_cfg->ipr_cmd_pool, GFP_KERNEL, &dma_addr); if (!ipr_cmd) { ipr_free_cmd_blks(ioa_cfg); diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index c3c0626f550b..09f2c74a40c5 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -360,7 +360,7 @@ spi_alloc_master(struct device *dev, unsigned size) if (!dev) return NULL; - master = kzalloc(size + sizeof *master, SLAB_KERNEL); + master = kzalloc(size + sizeof *master, GFP_KERNEL); if (!master) return NULL; @@ -607,7 +607,7 @@ static int __init spi_init(void) { int status; - buf = kmalloc(SPI_BUFSIZ, SLAB_KERNEL); + buf = kmalloc(SPI_BUFSIZ, GFP_KERNEL); if (!buf) { status = -ENOMEM; goto err0; diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c index 08c1c57c6128..57289b61d0be 100644 --- a/drivers/spi/spi_bitbang.c +++ b/drivers/spi/spi_bitbang.c @@ -196,7 +196,7 @@ int spi_bitbang_setup(struct spi_device *spi) return -EINVAL; if (!cs) { - cs = kzalloc(sizeof *cs, SLAB_KERNEL); + cs = kzalloc(sizeof *cs, GFP_KERNEL); if (!cs) return -ENOMEM; spi->controller_state = cs; diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 0a46acf557ac..77c05be5241a 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -2371,7 +2371,7 @@ check_highspeed (struct usb_hub *hub, struct usb_device *udev, int port1) struct usb_qualifier_descriptor *qual; int status; - qual = kmalloc (sizeof *qual, SLAB_KERNEL); + qual = kmalloc (sizeof *qual, GFP_KERNEL); if (qual == NULL) return; @@ -2922,7 +2922,7 @@ static int config_descriptors_changed(struct usb_device *udev) if (len < le16_to_cpu(udev->config[index].desc.wTotalLength)) len = le16_to_cpu(udev->config[index].desc.wTotalLength); } - buf = kmalloc (len, SLAB_KERNEL); + buf = kmalloc (len, GFP_KERNEL); if (buf == NULL) { dev_err(&udev->dev, "no mem to re-read configs after reset\n"); /* assume the worst */ diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c index 64554acad63f..31351826f2ba 100644 --- a/drivers/usb/gadget/gmidi.c +++ b/drivers/usb/gadget/gmidi.c @@ -1236,7 +1236,7 @@ autoconf_fail: /* ok, we made sense of the hardware ... */ - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) { return -ENOMEM; } diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c index a3076da3f4eb..805a9826842d 100644 --- a/drivers/usb/gadget/goku_udc.c +++ b/drivers/usb/gadget/goku_udc.c @@ -1864,7 +1864,7 @@ static int goku_probe(struct pci_dev *pdev, const struct pci_device_id *id) } /* alloc, and start init */ - dev = kmalloc (sizeof *dev, SLAB_KERNEL); + dev = kmalloc (sizeof *dev, GFP_KERNEL); if (dev == NULL){ pr_debug("enomem %s\n", pci_name(pdev)); retval = -ENOMEM; diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index 86924f9cdd7e..3fb1044a4db0 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -412,7 +412,7 @@ ep_read (struct file *fd, char __user *buf, size_t len, loff_t *ptr) /* FIXME readahead for O_NONBLOCK and poll(); careful with ZLPs */ value = -ENOMEM; - kbuf = kmalloc (len, SLAB_KERNEL); + kbuf = kmalloc (len, GFP_KERNEL); if (unlikely (!kbuf)) goto free1; @@ -456,7 +456,7 @@ ep_write (struct file *fd, const char __user *buf, size_t len, loff_t *ptr) /* FIXME writebehind for O_NONBLOCK and poll(), qlen = 1 */ value = -ENOMEM; - kbuf = kmalloc (len, SLAB_KERNEL); + kbuf = kmalloc (len, GFP_KERNEL); if (!kbuf) goto free1; if (copy_from_user (kbuf, buf, len)) { @@ -1898,7 +1898,7 @@ dev_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr) buf += 4; length -= 4; - kbuf = kmalloc (length, SLAB_KERNEL); + kbuf = kmalloc (length, GFP_KERNEL); if (!kbuf) return -ENOMEM; if (copy_from_user (kbuf, buf, length)) { diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c index 0b590831582c..3024c679e38e 100644 --- a/drivers/usb/gadget/net2280.c +++ b/drivers/usb/gadget/net2280.c @@ -2861,7 +2861,7 @@ static int net2280_probe (struct pci_dev *pdev, const struct pci_device_id *id) } /* alloc, and start init */ - dev = kzalloc (sizeof *dev, SLAB_KERNEL); + dev = kzalloc (sizeof *dev, GFP_KERNEL); if (dev == NULL){ retval = -ENOMEM; goto done; diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c index 48a09fd89d18..030d87c28c2f 100644 --- a/drivers/usb/gadget/omap_udc.c +++ b/drivers/usb/gadget/omap_udc.c @@ -2581,7 +2581,7 @@ omap_udc_setup(struct platform_device *odev, struct otg_transceiver *xceiv) /* UDC_PULLUP_EN gates the chip clock */ // OTG_SYSCON_1_REG |= DEV_IDLE_EN; - udc = kzalloc(sizeof(*udc), SLAB_KERNEL); + udc = kzalloc(sizeof(*udc), GFP_KERNEL); if (!udc) return -ENOMEM; diff --git a/drivers/usb/gadget/zero.c b/drivers/usb/gadget/zero.c index 0f809dd68492..40710ea1b490 100644 --- a/drivers/usb/gadget/zero.c +++ b/drivers/usb/gadget/zero.c @@ -1190,7 +1190,7 @@ autoconf_fail: /* ok, we made sense of the hardware ... */ - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; spin_lock_init (&dev->lock); diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 396dc69d4b4c..7fd872aa654a 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -188,7 +188,7 @@ static DEFINE_TIMER(bulk_eot_timer, NULL, 0, 0); #define CHECK_ALIGN(x) if (((__u32)(x)) & 0x00000003) \ {panic("Alignment check (DWORD) failed at %s:%s:%d\n", __FILE__, __FUNCTION__, __LINE__);} -#define SLAB_FLAG (in_interrupt() ? GFP_ATOMIC : SLAB_KERNEL) +#define SLAB_FLAG (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) #define KMALLOC_FLAG (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) /* Most helpful debugging aid */ diff --git a/drivers/usb/host/ohci-pnx4008.c b/drivers/usb/host/ohci-pnx4008.c index 2dbb77414905..7f26f9bdbaf1 100644 --- a/drivers/usb/host/ohci-pnx4008.c +++ b/drivers/usb/host/ohci-pnx4008.c @@ -134,7 +134,7 @@ static int isp1301_attach(struct i2c_adapter *adap, int addr, int kind) { struct i2c_client *c; - c = (struct i2c_client *)kzalloc(sizeof(*c), SLAB_KERNEL); + c = (struct i2c_client *)kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return -ENOMEM; diff --git a/drivers/usb/input/acecad.c b/drivers/usb/input/acecad.c index 0096373b5f98..909138e5aa04 100644 --- a/drivers/usb/input/acecad.c +++ b/drivers/usb/input/acecad.c @@ -152,7 +152,7 @@ static int usb_acecad_probe(struct usb_interface *intf, const struct usb_device_ if (!acecad || !input_dev) goto fail1; - acecad->data = usb_buffer_alloc(dev, 8, SLAB_KERNEL, &acecad->data_dma); + acecad->data = usb_buffer_alloc(dev, 8, GFP_KERNEL, &acecad->data_dma); if (!acecad->data) goto fail1; diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c index 49704d4ed0e2..7f3c57da9bc0 100644 --- a/drivers/usb/input/usbtouchscreen.c +++ b/drivers/usb/input/usbtouchscreen.c @@ -680,7 +680,7 @@ static int usbtouch_probe(struct usb_interface *intf, type->process_pkt = usbtouch_process_pkt; usbtouch->data = usb_buffer_alloc(udev, type->rept_size, - SLAB_KERNEL, &usbtouch->data_dma); + GFP_KERNEL, &usbtouch->data_dma); if (!usbtouch->data) goto out_free; diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c index ea04dccdc651..fb321864a92d 100644 --- a/drivers/usb/misc/usbtest.c +++ b/drivers/usb/misc/usbtest.c @@ -213,7 +213,7 @@ static struct urb *simple_alloc_urb ( if (bytes < 0) return NULL; - urb = usb_alloc_urb (0, SLAB_KERNEL); + urb = usb_alloc_urb (0, GFP_KERNEL); if (!urb) return urb; usb_fill_bulk_urb (urb, udev, pipe, NULL, bytes, simple_callback, NULL); @@ -223,7 +223,7 @@ static struct urb *simple_alloc_urb ( urb->transfer_flags = URB_NO_TRANSFER_DMA_MAP; if (usb_pipein (pipe)) urb->transfer_flags |= URB_SHORT_NOT_OK; - urb->transfer_buffer = usb_buffer_alloc (udev, bytes, SLAB_KERNEL, + urb->transfer_buffer = usb_buffer_alloc (udev, bytes, GFP_KERNEL, &urb->transfer_dma); if (!urb->transfer_buffer) { usb_free_urb (urb); @@ -315,7 +315,7 @@ static int simple_io ( init_completion (&completion); if (usb_pipeout (urb->pipe)) simple_fill_buf (urb); - if ((retval = usb_submit_urb (urb, SLAB_KERNEL)) != 0) + if ((retval = usb_submit_urb (urb, GFP_KERNEL)) != 0) break; /* NOTE: no timeouts; can't be broken out of by interrupt */ @@ -374,7 +374,7 @@ alloc_sglist (int nents, int max, int vary) unsigned i; unsigned size = max; - sg = kmalloc (nents * sizeof *sg, SLAB_KERNEL); + sg = kmalloc (nents * sizeof *sg, GFP_KERNEL); if (!sg) return NULL; @@ -382,7 +382,7 @@ alloc_sglist (int nents, int max, int vary) char *buf; unsigned j; - buf = kzalloc (size, SLAB_KERNEL); + buf = kzalloc (size, GFP_KERNEL); if (!buf) { free_sglist (sg, i); return NULL; @@ -428,7 +428,7 @@ static int perform_sglist ( (udev->speed == USB_SPEED_HIGH) ? (INTERRUPT_RATE << 3) : INTERRUPT_RATE, - sg, nents, 0, SLAB_KERNEL); + sg, nents, 0, GFP_KERNEL); if (retval) break; @@ -855,7 +855,7 @@ test_ctrl_queue (struct usbtest_dev *dev, struct usbtest_param *param) * as with bulk/intr sglists, sglen is the queue depth; it also * controls which subtests run (more tests than sglen) or rerun. */ - urb = kcalloc(param->sglen, sizeof(struct urb *), SLAB_KERNEL); + urb = kcalloc(param->sglen, sizeof(struct urb *), GFP_KERNEL); if (!urb) return -ENOMEM; for (i = 0; i < param->sglen; i++) { @@ -981,7 +981,7 @@ test_ctrl_queue (struct usbtest_dev *dev, struct usbtest_param *param) if (!u) goto cleanup; - reqp = usb_buffer_alloc (udev, sizeof *reqp, SLAB_KERNEL, + reqp = usb_buffer_alloc (udev, sizeof *reqp, GFP_KERNEL, &u->setup_dma); if (!reqp) goto cleanup; @@ -1067,7 +1067,7 @@ static int unlink1 (struct usbtest_dev *dev, int pipe, int size, int async) * FIXME want additional tests for when endpoint is STALLing * due to errors, or is just NAKing requests. */ - if ((retval = usb_submit_urb (urb, SLAB_KERNEL)) != 0) { + if ((retval = usb_submit_urb (urb, GFP_KERNEL)) != 0) { dev_dbg (&dev->intf->dev, "submit fail %d\n", retval); return retval; } @@ -1251,7 +1251,7 @@ static int ctrl_out (struct usbtest_dev *dev, if (length < 1 || length > 0xffff || vary >= length) return -EINVAL; - buf = kmalloc(length, SLAB_KERNEL); + buf = kmalloc(length, GFP_KERNEL); if (!buf) return -ENOMEM; @@ -1403,7 +1403,7 @@ static struct urb *iso_alloc_urb ( maxp *= 1 + (0x3 & (le16_to_cpu(desc->wMaxPacketSize) >> 11)); packets = (bytes + maxp - 1) / maxp; - urb = usb_alloc_urb (packets, SLAB_KERNEL); + urb = usb_alloc_urb (packets, GFP_KERNEL); if (!urb) return urb; urb->dev = udev; @@ -1411,7 +1411,7 @@ static struct urb *iso_alloc_urb ( urb->number_of_packets = packets; urb->transfer_buffer_length = bytes; - urb->transfer_buffer = usb_buffer_alloc (udev, bytes, SLAB_KERNEL, + urb->transfer_buffer = usb_buffer_alloc (udev, bytes, GFP_KERNEL, &urb->transfer_dma); if (!urb->transfer_buffer) { usb_free_urb (urb); @@ -1900,7 +1900,7 @@ usbtest_probe (struct usb_interface *intf, const struct usb_device_id *id) } #endif - dev = kzalloc(sizeof(*dev), SLAB_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; info = (struct usbtest_info *) id->driver_info; @@ -1910,7 +1910,7 @@ usbtest_probe (struct usb_interface *intf, const struct usb_device_id *id) dev->intf = intf; /* cacheline-aligned scratch for i/o */ - if ((dev->buf = kmalloc (TBUF_SIZE, SLAB_KERNEL)) == NULL) { + if ((dev->buf = kmalloc (TBUF_SIZE, GFP_KERNEL)) == NULL) { kfree (dev); return -ENOMEM; } diff --git a/drivers/usb/net/rndis_host.c b/drivers/usb/net/rndis_host.c index c2a28d88ef3c..99f26b3e502f 100644 --- a/drivers/usb/net/rndis_host.c +++ b/drivers/usb/net/rndis_host.c @@ -469,7 +469,7 @@ static void rndis_unbind(struct usbnet *dev, struct usb_interface *intf) struct rndis_halt *halt; /* try to clear any rndis state/activity (no i/o from stack!) */ - halt = kcalloc(1, sizeof *halt, SLAB_KERNEL); + halt = kcalloc(1, sizeof *halt, GFP_KERNEL); if (halt) { halt->msg_type = RNDIS_MSG_HALT; halt->msg_len = ccpu2(sizeof *halt); diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c index 327f97555679..6e39e9988259 100644 --- a/drivers/usb/net/usbnet.c +++ b/drivers/usb/net/usbnet.c @@ -179,9 +179,9 @@ static int init_status (struct usbnet *dev, struct usb_interface *intf) period = max ((int) dev->status->desc.bInterval, (dev->udev->speed == USB_SPEED_HIGH) ? 7 : 3); - buf = kmalloc (maxp, SLAB_KERNEL); + buf = kmalloc (maxp, GFP_KERNEL); if (buf) { - dev->interrupt = usb_alloc_urb (0, SLAB_KERNEL); + dev->interrupt = usb_alloc_urb (0, GFP_KERNEL); if (!dev->interrupt) { kfree (buf); return -ENOMEM; diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 9ade139086fc..52eb10ca654e 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -217,7 +217,7 @@ static kmem_cache_t *adfs_inode_cachep; static struct inode *adfs_alloc_inode(struct super_block *sb) { struct adfs_inode_info *ei; - ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, SLAB_KERNEL); + ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/affs/super.c b/fs/affs/super.c index 5ea72c3a16c3..81c73ec09f66 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -71,7 +71,7 @@ static kmem_cache_t * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { struct affs_inode_info *ei; - ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, SLAB_KERNEL); + ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; diff --git a/fs/afs/super.c b/fs/afs/super.c index 67d1f5c819ec..c6ead009bf78 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -412,7 +412,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb) struct afs_vnode *vnode; vnode = (struct afs_vnode *) - kmem_cache_alloc(afs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(afs_inode_cachep, GFP_KERNEL); if (!vnode) return NULL; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 07f7144f0e2e..995348df94ad 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -277,7 +277,7 @@ befs_alloc_inode(struct super_block *sb) { struct befs_inode_info *bi; bi = (struct befs_inode_info *)kmem_cache_alloc(befs_inode_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index ed27ffb3459e..2e45123c8f7a 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -233,7 +233,7 @@ static kmem_cache_t * bfs_inode_cachep; static struct inode *bfs_alloc_inode(struct super_block *sb) { struct bfs_inode_info *bi; - bi = kmem_cache_alloc(bfs_inode_cachep, SLAB_KERNEL); + bi = kmem_cache_alloc(bfs_inode_cachep, GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; diff --git a/fs/block_dev.c b/fs/block_dev.c index 36c0e7af9d0f..063506705f27 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -239,7 +239,7 @@ static kmem_cache_t * bdev_cachep __read_mostly; static struct inode *bdev_alloc_inode(struct super_block *sb) { - struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, SLAB_KERNEL); + struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 84976cdbe713..84168629cea3 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -245,7 +245,7 @@ static struct inode * cifs_alloc_inode(struct super_block *sb) { struct cifsInodeInfo *cifs_inode; - cifs_inode = kmem_cache_alloc(cifs_inode_cachep, SLAB_KERNEL); + cifs_inode = kmem_cache_alloc(cifs_inode_cachep, GFP_KERNEL); if (!cifs_inode) return NULL; cifs_inode->cifsAttrs = 0x20; /* default */ diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 8355daff504c..aedf683f011f 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -153,7 +153,7 @@ cifs_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_req_poolp, SLAB_KERNEL | GFP_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_req_poolp, GFP_KERNEL | GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -192,7 +192,7 @@ cifs_small_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, SLAB_KERNEL | GFP_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, GFP_KERNEL | GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 7514237cf31a..1f727765a8ea 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -51,7 +51,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) } temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - SLAB_KERNEL | GFP_NOFS); + GFP_KERNEL | GFP_NOFS); if (temp == NULL) return temp; else { @@ -118,7 +118,7 @@ AllocOplockQEntry(struct inode * pinode, __u16 fid, struct cifsTconInfo * tcon) return NULL; } temp = (struct oplock_q_entry *) kmem_cache_alloc(cifs_oplock_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (temp == NULL) return temp; else { diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 88d123321164..50cedd2617d6 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -43,7 +43,7 @@ static kmem_cache_t * coda_inode_cachep; static struct inode *coda_alloc_inode(struct super_block *sb) { struct coda_inode_info *ei; - ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, SLAB_KERNEL); + ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, GFP_KERNEL); if (!ei) return NULL; memset(&ei->c_fid, 0, sizeof(struct CodaFid)); diff --git a/fs/dnotify.c b/fs/dnotify.c index 2b0442db67e0..e778b1737b79 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -77,7 +77,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) inode = filp->f_dentry->d_inode; if (!S_ISDIR(inode->i_mode)) return -ENOTDIR; - dn = kmem_cache_alloc(dn_cache, SLAB_KERNEL); + dn = kmem_cache_alloc(dn_cache, GFP_KERNEL); if (dn == NULL) return -ENOMEM; spin_lock(&inode->i_lock); diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 776b2eed371e..7196f50fe152 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -628,7 +628,7 @@ int ecryptfs_decrypt_page(struct file *file, struct page *page) num_extents_per_page = PAGE_CACHE_SIZE / crypt_stat->extent_size; base_extent = (page->index * num_extents_per_page); lower_page_virt = kmem_cache_alloc(ecryptfs_lower_page_cache, - SLAB_KERNEL); + GFP_KERNEL); if (!lower_page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Error getting page for encrypted " diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index a92ef05eff8f..42099e779a56 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -250,7 +250,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file) int lower_flags; /* Released in ecryptfs_release or end of function if failure */ - file_info = kmem_cache_alloc(ecryptfs_file_info_cache, SLAB_KERNEL); + file_info = kmem_cache_alloc(ecryptfs_file_info_cache, GFP_KERNEL); ecryptfs_set_file_private(file, file_info); if (!file_info) { ecryptfs_printk(KERN_ERR, diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 70911412044d..8a1945a84c36 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -369,7 +369,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, BUG_ON(!atomic_read(&lower_dentry->d_count)); ecryptfs_set_dentry_private(dentry, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(dentry)) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Out of memory whilst attempting " @@ -795,7 +795,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) /* Released at out_free: label */ ecryptfs_set_file_private(&fake_ecryptfs_file, kmem_cache_alloc(ecryptfs_file_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (unlikely(!ecryptfs_file_to_private(&fake_ecryptfs_file))) { rc = -ENOMEM; goto out; diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index c3746f56d162..745c0f1bfbbd 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -207,7 +207,7 @@ parse_tag_3_packet(struct ecryptfs_crypt_stat *crypt_stat, /* Released: wipe_auth_tok_list called in ecryptfs_parse_packet_set or * at end of function upon failure */ auth_tok_list_item = - kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, SLAB_KERNEL); + kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, GFP_KERNEL); if (!auth_tok_list_item) { ecryptfs_printk(KERN_ERR, "Unable to allocate memory\n"); rc = -ENOMEM; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a78d87d14baf..a2c6ccbce300 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -378,7 +378,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* Released in ecryptfs_put_super() */ ecryptfs_set_superblock_private(sb, kmem_cache_alloc(ecryptfs_sb_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_superblock_to_private(sb)) { ecryptfs_printk(KERN_WARNING, "Out of memory\n"); rc = -ENOMEM; @@ -402,7 +402,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* through deactivate_super(sb) from get_sb_nodev() */ ecryptfs_set_dentry_private(sb->s_root, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(sb->s_root)) { ecryptfs_printk(KERN_ERR, "dentry_info_cache alloc failed\n"); diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 825757ae4867..eaa5daaf106e 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -50,7 +50,7 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb) struct inode *inode = NULL; ecryptfs_inode = kmem_cache_alloc(ecryptfs_inode_info_cache, - SLAB_KERNEL); + GFP_KERNEL); if (unlikely(!ecryptfs_inode)) goto out; ecryptfs_init_crypt_stat(&ecryptfs_inode->crypt_stat); diff --git a/fs/efs/super.c b/fs/efs/super.c index b3f50651eb6b..69b15a996cfc 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -57,7 +57,7 @@ static kmem_cache_t * efs_inode_cachep; static struct inode *efs_alloc_inode(struct super_block *sb) { struct efs_inode_info *ei; - ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, SLAB_KERNEL); + ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index ae228ec54e94..f5c88435c6be 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -961,7 +961,7 @@ static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead, struct epitem *epi = ep_item_from_epqueue(pt); struct eppoll_entry *pwq; - if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, SLAB_KERNEL))) { + if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) { init_waitqueue_func_entry(&pwq->wait, ep_poll_callback); pwq->whead = whead; pwq->base = epi; @@ -1004,7 +1004,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, struct ep_pqueue epq; error = -ENOMEM; - if (!(epi = kmem_cache_alloc(epi_cache, SLAB_KERNEL))) + if (!(epi = kmem_cache_alloc(epi_cache, GFP_KERNEL))) goto eexit_1; /* Item initialization follow here ... */ diff --git a/fs/exec.c b/fs/exec.c index d993ea1a81ae..2092bd207463 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -404,7 +404,7 @@ int setup_arg_pages(struct linux_binprm *bprm, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index d8b9abd95d07..85c237e73853 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -140,7 +140,7 @@ static kmem_cache_t * ext2_inode_cachep; static struct inode *ext2_alloc_inode(struct super_block *sb) { struct ext2_inode_info *ei; - ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, SLAB_KERNEL); + ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; #ifdef CONFIG_EXT2_FS_POSIX_ACL diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 82cc4f59e3ba..8c272278455c 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -63,7 +63,7 @@ void fat_cache_destroy(void) static inline struct fat_cache *fat_cache_alloc(struct inode *inode) { - return kmem_cache_alloc(fat_cache_cachep, SLAB_KERNEL); + return kmem_cache_alloc(fat_cache_cachep, GFP_KERNEL); } static inline void fat_cache_free(struct fat_cache *cache) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 78945b53b0f8..b58fd0c9f3cd 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -482,7 +482,7 @@ static kmem_cache_t *fat_inode_cachep; static struct inode *fat_alloc_inode(struct super_block *sb) { struct msdos_inode_info *ei; - ei = kmem_cache_alloc(fat_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(fat_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/fcntl.c b/fs/fcntl.c index e4f26165f12a..c03dc9cb21cb 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -567,7 +567,7 @@ int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fap int result = 0; if (on) { - new = kmem_cache_alloc(fasync_cache, SLAB_KERNEL); + new = kmem_cache_alloc(fasync_cache, GFP_KERNEL); if (!new) return -ENOMEM; } diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 4786d51ad3bd..d2dd0d700077 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -103,7 +103,7 @@ vxfs_blkiget(struct super_block *sbp, u_long extent, ino_t ino) struct vxfs_inode_info *vip; struct vxfs_dinode *dip; - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(bp->b_data + offset); memcpy(vip, dip, sizeof(*vip)); @@ -145,7 +145,7 @@ __vxfs_iget(ino_t ino, struct inode *ilistp) struct vxfs_dinode *dip; caddr_t kaddr = (char *)page_address(pp); - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(kaddr + offset); memcpy(vip, dip, sizeof(*vip)); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 66571eafbb1e..8c15139f2756 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -41,7 +41,7 @@ static void fuse_request_init(struct fuse_req *req) struct fuse_req *fuse_request_alloc(void) { - struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, SLAB_KERNEL); + struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, GFP_KERNEL); if (req) fuse_request_init(req); return req; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index fc4203570370..e039e2047cce 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -46,7 +46,7 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) struct inode *inode; struct fuse_inode *fi; - inode = kmem_cache_alloc(fuse_inode_cachep, SLAB_KERNEL); + inode = kmem_cache_alloc(fuse_inode_cachep, GFP_KERNEL); if (!inode) return NULL; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 85b17b3fa4a0..ffc6409132c8 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -145,7 +145,7 @@ static struct inode *hfs_alloc_inode(struct super_block *sb) { struct hfs_inode_info *i; - i = kmem_cache_alloc(hfs_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfs_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 194eede52fa4..4a0c70c76c8a 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -440,7 +440,7 @@ static struct inode *hfsplus_alloc_inode(struct super_block *sb) { struct hfsplus_inode_info *i; - i = kmem_cache_alloc(hfsplus_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfsplus_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 7f4756963d05..36e52173a54a 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -522,7 +522,7 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo))) return NULL; - p = kmem_cache_alloc(hugetlbfs_inode_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(hugetlbfs_inode_cachep, GFP_KERNEL); if (unlikely(!p)) { hugetlbfs_inc_free_inodes(sbinfo); return NULL; diff --git a/fs/inode.c b/fs/inode.c index 26cdb115ce67..dd15984d51a8 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -109,7 +109,7 @@ static struct inode *alloc_inode(struct super_block *sb) if (sb->s_op->alloc_inode) inode = sb->s_op->alloc_inode(sb); else - inode = (struct inode *) kmem_cache_alloc(inode_cachep, SLAB_KERNEL); + inode = (struct inode *) kmem_cache_alloc(inode_cachep, GFP_KERNEL); if (inode) { struct address_space * const mapping = &inode->i_data; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index c34b862cdbf2..4b6381cd2cf4 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -62,7 +62,7 @@ static kmem_cache_t *isofs_inode_cachep; static struct inode *isofs_alloc_inode(struct super_block *sb) { struct iso_inode_info *ei; - ei = kmem_cache_alloc(isofs_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(isofs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index bc4b8106a490..77be534ce422 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -33,7 +33,7 @@ static kmem_cache_t *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { struct jffs2_inode_info *ei; - ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, SLAB_KERNEL); + ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/locks.c b/fs/locks.c index e0b6a80649a0..a7b97d50c1e0 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -147,7 +147,7 @@ static kmem_cache_t *filelock_cache __read_mostly; /* Allocate an empty lock structure. */ static struct file_lock *locks_alloc_lock(void) { - return kmem_cache_alloc(filelock_cache, SLAB_KERNEL); + return kmem_cache_alloc(filelock_cache, GFP_KERNEL); } static void locks_release_private(struct file_lock *fl) diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 1e36bae4d0eb..ce532c2deda8 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -56,7 +56,7 @@ static kmem_cache_t * minix_inode_cachep; static struct inode *minix_alloc_inode(struct super_block *sb) { struct minix_inode_info *ei; - ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, SLAB_KERNEL); + ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 72dad552aa00..ed84d899220c 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -45,7 +45,7 @@ static kmem_cache_t * ncp_inode_cachep; static struct inode *ncp_alloc_inode(struct super_block *sb) { struct ncp_inode_info *ei; - ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, SLAB_KERNEL); + ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index bdfabf854a51..769fd0a0c772 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -143,7 +143,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) { struct nfs_direct_req *dreq; - dreq = kmem_cache_alloc(nfs_direct_cachep, SLAB_KERNEL); + dreq = kmem_cache_alloc(nfs_direct_cachep, GFP_KERNEL); if (!dreq) return NULL; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 08cc4c5919ab..6b53aae4ed2c 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1080,7 +1080,7 @@ void nfs4_clear_inode(struct inode *inode) struct inode *nfs_alloc_inode(struct super_block *sb) { struct nfs_inode *nfsi; - nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, SLAB_KERNEL); + nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, GFP_KERNEL); if (!nfsi) return NULL; nfsi->flags = 0UL; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 829af323f288..a1561a820abe 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -26,7 +26,7 @@ static inline struct nfs_page * nfs_page_alloc(void) { struct nfs_page *p; - p = kmem_cache_alloc(nfs_page_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(nfs_page_cachep, GFP_KERNEL); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->wb_list); diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 592a6402e851..911d1bcfc567 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -336,7 +336,7 @@ static struct inode *openprom_alloc_inode(struct super_block *sb) { struct op_inode_info *oi; - oi = kmem_cache_alloc(op_inode_cachep, SLAB_KERNEL); + oi = kmem_cache_alloc(op_inode_cachep, GFP_KERNEL); if (!oi) return NULL; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 49dfb2ab783e..b24cdb2f17c1 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -88,7 +88,7 @@ static struct inode *proc_alloc_inode(struct super_block *sb) struct proc_inode *ei; struct inode *inode; - ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, SLAB_KERNEL); + ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->pid = NULL; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5a41db2a218d..5b943eb11d76 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -520,7 +520,7 @@ static kmem_cache_t *qnx4_inode_cachep; static struct inode *qnx4_alloc_inode(struct super_block *sb) { struct qnx4_inode_info *ei; - ei = kmem_cache_alloc(qnx4_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(qnx4_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 17249994110f..32332516d656 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -496,7 +496,7 @@ static struct inode *reiserfs_alloc_inode(struct super_block *sb) { struct reiserfs_inode_info *ei; ei = (struct reiserfs_inode_info *) - kmem_cache_alloc(reiserfs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index ddcd9e1ef282..d1b455f9b669 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -555,7 +555,7 @@ static kmem_cache_t * romfs_inode_cachep; static struct inode *romfs_alloc_inode(struct super_block *sb) { struct romfs_inode_info *ei; - ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, SLAB_KERNEL); + ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 2c122ee83adb..221617103484 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -55,7 +55,7 @@ static kmem_cache_t *smb_inode_cachep; static struct inode *smb_alloc_inode(struct super_block *sb) { struct smb_inode_info *ei; - ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, SLAB_KERNEL); + ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 0fb74697abc4..3eb1402191b9 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -61,7 +61,7 @@ static struct smb_request *smb_do_alloc_request(struct smb_sb_info *server, struct smb_request *req; unsigned char *buf = NULL; - req = kmem_cache_alloc(req_cachep, SLAB_KERNEL); + req = kmem_cache_alloc(req_cachep, GFP_KERNEL); VERBOSE("allocating request: %p\n", req); if (!req) goto out; diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index d63c5e48b050..a6ca12b747cf 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -307,7 +307,7 @@ static struct inode *sysv_alloc_inode(struct super_block *sb) { struct sysv_inode_info *si; - si = kmem_cache_alloc(sysv_inode_cachep, SLAB_KERNEL); + si = kmem_cache_alloc(sysv_inode_cachep, GFP_KERNEL); if (!si) return NULL; return &si->vfs_inode; diff --git a/fs/udf/super.c b/fs/udf/super.c index 1aea6a4f9a4a..e50f24221dea 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -112,7 +112,7 @@ static kmem_cache_t * udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { struct udf_inode_info *ei; - ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, SLAB_KERNEL); + ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); if (!ei) return NULL; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index ec79e3091d1b..85a88c0c5e68 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1209,7 +1209,7 @@ static kmem_cache_t * ufs_inode_cachep; static struct inode *ufs_alloc_inode(struct super_block *sb) { struct ufs_inode_info *ei; - ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, SLAB_KERNEL); + ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; diff --git a/include/linux/fs.h b/include/linux/fs.h index a8039c8d8cbb..94b831b8157c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1483,7 +1483,7 @@ extern void __init vfs_caches_init(unsigned long); extern struct kmem_cache *names_cachep; -#define __getname() kmem_cache_alloc(names_cachep, SLAB_KERNEL) +#define __getname() kmem_cache_alloc(names_cachep, GFP_KERNEL) #define __putname(name) kmem_cache_free(names_cachep, (void *)(name)) #ifndef CONFIG_AUDITSYSCALL #define putname(name) __putname(name) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index db2c1df4fef9..61c2ab634b00 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -34,7 +34,7 @@ extern kmem_cache_t *anon_vma_cachep; static inline struct anon_vma *anon_vma_alloc(void) { - return kmem_cache_alloc(anon_vma_cachep, SLAB_KERNEL); + return kmem_cache_alloc(anon_vma_cachep, GFP_KERNEL); } static inline void anon_vma_free(struct anon_vma *anon_vma) diff --git a/include/linux/slab.h b/include/linux/slab.h index 34b046ea88f1..639f65efa46e 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -19,7 +19,6 @@ typedef struct kmem_cache kmem_cache_t; #include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ /* flags for kmem_cache_alloc() */ -#define SLAB_KERNEL GFP_KERNEL #define SLAB_DMA GFP_DMA /* flags to pass to kmem_cache_create(). diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index 6562a2050a25..f81a5af8a4f8 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -35,7 +35,7 @@ static inline void taskstats_tgid_alloc(struct task_struct *tsk) return; /* No problem if kmem_cache_zalloc() fails */ - stats = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); + stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); spin_lock_irq(&tsk->sighand->siglock); if (!sig->stats) { diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 7c274002c9f5..813bb941342b 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -224,7 +224,7 @@ static struct inode *mqueue_alloc_inode(struct super_block *sb) { struct mqueue_inode_info *ei; - ei = kmem_cache_alloc(mqueue_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(mqueue_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 66a0ea48751d..70e9ec603082 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -41,7 +41,7 @@ void delayacct_init(void) void __delayacct_tsk_init(struct task_struct *tsk) { - tsk->delays = kmem_cache_zalloc(delayacct_cache, SLAB_KERNEL); + tsk->delays = kmem_cache_zalloc(delayacct_cache, GFP_KERNEL); if (tsk->delays) spin_lock_init(&tsk->delays->lock); } diff --git a/kernel/fork.c b/kernel/fork.c index 5678e6c61ef2..711aa5f10da7 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -237,7 +237,7 @@ static inline int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) goto fail_nomem; charge = len; } - tmp = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!tmp) goto fail_nomem; *tmp = *mpnt; @@ -319,7 +319,7 @@ static inline void mm_free_pgd(struct mm_struct * mm) __cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock); -#define allocate_mm() (kmem_cache_alloc(mm_cachep, SLAB_KERNEL)) +#define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL)) #define free_mm(mm) (kmem_cache_free(mm_cachep, (mm))) #include @@ -621,7 +621,7 @@ static struct files_struct *alloc_files(void) struct files_struct *newf; struct fdtable *fdt; - newf = kmem_cache_alloc(files_cachep, SLAB_KERNEL); + newf = kmem_cache_alloc(files_cachep, GFP_KERNEL); if (!newf) goto out; diff --git a/kernel/taskstats.c b/kernel/taskstats.c index d3d28919d4b4..1b2b326cf703 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -425,7 +425,7 @@ void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) *mycpu = raw_smp_processor_id(); *ptidstats = NULL; - tmp = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); + tmp = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); if (!tmp) return; diff --git a/kernel/user.c b/kernel/user.c index 220e586127a0..c1f93c164c93 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -132,7 +132,7 @@ struct user_struct * alloc_uid(uid_t uid) if (!up) { struct user_struct *new; - new = kmem_cache_alloc(uid_cachep, SLAB_KERNEL); + new = kmem_cache_alloc(uid_cachep, GFP_KERNEL); if (!new) return NULL; new->uid = uid; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index e7b69c90cfd6..ad864f8708b0 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1326,7 +1326,7 @@ struct mempolicy *__mpol_copy(struct mempolicy *old) atomic_set(&new->refcnt, 1); if (new->policy == MPOL_BIND) { int sz = ksize(old->v.zonelist); - new->v.zonelist = kmemdup(old->v.zonelist, sz, SLAB_KERNEL); + new->v.zonelist = kmemdup(old->v.zonelist, sz, GFP_KERNEL); if (!new->v.zonelist) { kmem_cache_free(policy_cache, new); return ERR_PTR(-ENOMEM); diff --git a/mm/mmap.c b/mm/mmap.c index 7b40abd7cba2..7be110e98d4c 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1736,7 +1736,7 @@ int split_vma(struct mm_struct * mm, struct vm_area_struct * vma, if (mm->map_count >= sysctl_max_map_count) return -ENOMEM; - new = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!new) return -ENOMEM; @@ -2057,7 +2057,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, vma_start < new_vma->vm_end) *vmap = new_vma; } else { - new_vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + new_vma = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (new_vma) { *new_vma = *vma; pol = mpol_copy(vma_policy(vma)); diff --git a/mm/shmem.c b/mm/shmem.c index 4959535fc14c..bdaecfdaabd4 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2263,7 +2263,7 @@ static struct kmem_cache *shmem_inode_cachep; static struct inode *shmem_alloc_inode(struct super_block *sb) { struct shmem_inode_info *p; - p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, SLAB_KERNEL); + p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL); if (!p) return NULL; return &p->vfs_inode; diff --git a/mm/slab.c b/mm/slab.c index 9f34b4946fba..1f374c1df018 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2237,7 +2237,7 @@ kmem_cache_create (const char *name, size_t size, size_t align, align = ralign; /* Get cache's description obj. */ - cachep = kmem_cache_zalloc(&cache_cache, SLAB_KERNEL); + cachep = kmem_cache_zalloc(&cache_cache, GFP_KERNEL); if (!cachep) goto oops; diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index bdbc3f431668..101e5ccaf096 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -590,7 +590,7 @@ create: replace: err = -ENOBUFS; - new_f = kmem_cache_alloc(dn_hash_kmem, SLAB_KERNEL); + new_f = kmem_cache_alloc(dn_hash_kmem, GFP_KERNEL); if (new_f == NULL) goto out; diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 107bb6cbb0b3..4463443e42cd 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -485,13 +485,13 @@ static int fn_hash_insert(struct fib_table *tb, struct fib_config *cfg) goto out; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; new_f = NULL; if (!f) { - new_f = kmem_cache_alloc(fn_hash_kmem, SLAB_KERNEL); + new_f = kmem_cache_alloc(fn_hash_kmem, GFP_KERNEL); if (new_f == NULL) goto out_free_new_fa; diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index d17990ec724f..6be6caf1af37 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1187,7 +1187,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg) u8 state; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; @@ -1232,7 +1232,7 @@ static int fn_trie_insert(struct fib_table *tb, struct fib_config *cfg) goto out; err = -ENOBUFS; - new_fa = kmem_cache_alloc(fn_alias_kmem, SLAB_KERNEL); + new_fa = kmem_cache_alloc(fn_alias_kmem, GFP_KERNEL); if (new_fa == NULL) goto out; diff --git a/net/socket.c b/net/socket.c index e8db54702a69..4f417c2ddc15 100644 --- a/net/socket.c +++ b/net/socket.c @@ -236,7 +236,7 @@ static struct inode *sock_alloc_inode(struct super_block *sb) { struct socket_alloc *ei; - ei = kmem_cache_alloc(sock_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(sock_inode_cachep, GFP_KERNEL); if (!ei) return NULL; init_waitqueue_head(&ei->socket.wait); diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index 49dba5febbbd..df753d0a884b 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -143,7 +143,7 @@ static struct inode * rpc_alloc_inode(struct super_block *sb) { struct rpc_inode *rpci; - rpci = (struct rpc_inode *)kmem_cache_alloc(rpc_inode_cachep, SLAB_KERNEL); + rpci = (struct rpc_inode *)kmem_cache_alloc(rpc_inode_cachep, GFP_KERNEL); if (!rpci) return NULL; return &rpci->vfs_inode; diff --git a/security/keys/key.c b/security/keys/key.c index 70eacbe5abde..157bac658bf9 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -285,7 +285,7 @@ struct key *key_alloc(struct key_type *type, const char *desc, } /* allocate and initialise the key and its description */ - key = kmem_cache_alloc(key_jar, SLAB_KERNEL); + key = kmem_cache_alloc(key_jar, GFP_KERNEL); if (!key) goto no_memory_2; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 78f98fe084eb..ac1aeed0b289 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -181,7 +181,7 @@ static int inode_alloc_security(struct inode *inode) struct task_security_struct *tsec = current->security; struct inode_security_struct *isec; - isec = kmem_cache_alloc(sel_inode_cache, SLAB_KERNEL); + isec = kmem_cache_alloc(sel_inode_cache, GFP_KERNEL); if (!isec) return -ENOMEM; diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c index d049c7acbc8b..2dfc6134c2c2 100644 --- a/security/selinux/ss/avtab.c +++ b/security/selinux/ss/avtab.c @@ -36,7 +36,7 @@ avtab_insert_node(struct avtab *h, int hvalue, struct avtab_key *key, struct avtab_datum *datum) { struct avtab_node * newnode; - newnode = kmem_cache_alloc(avtab_node_cachep, SLAB_KERNEL); + newnode = kmem_cache_alloc(avtab_node_cachep, GFP_KERNEL); if (newnode == NULL) return NULL; memset(newnode, 0, sizeof(struct avtab_node)); -- cgit v1.2.3 From 441e143e95f5aa1e04026cb0aa71c801ba53982f Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:19 -0800 Subject: [PATCH] slab: remove SLAB_DMA SLAB_DMA is an alias of GFP_DMA. This is the last one so we remove the leftover comment too. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/atm/he.c | 6 +++--- drivers/s390/block/dasd_eckd.c | 2 +- drivers/s390/block/dasd_fba.c | 2 +- drivers/usb/core/buffer.c | 2 +- include/linux/slab.h | 3 --- mm/slab.c | 4 ++-- 6 files changed, 8 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/atm/he.c b/drivers/atm/he.c index ec8a7a633e68..7d9b4e52f0bf 100644 --- a/drivers/atm/he.c +++ b/drivers/atm/he.c @@ -820,7 +820,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPS_POOL - cpuaddr = pci_pool_alloc(he_dev->rbps_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbps_pool, GFP_KERNEL|GFP_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else @@ -884,7 +884,7 @@ he_init_group(struct he_dev *he_dev, int group) void *cpuaddr; #ifdef USE_RBPL_POOL - cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, GFP_KERNEL|SLAB_DMA, &dma_handle); + cpuaddr = pci_pool_alloc(he_dev->rbpl_pool, GFP_KERNEL|GFP_DMA, &dma_handle); if (cpuaddr == NULL) return -ENOMEM; #else @@ -1724,7 +1724,7 @@ __alloc_tpd(struct he_dev *he_dev) struct he_tpd *tpd; dma_addr_t dma_handle; - tpd = pci_pool_alloc(he_dev->tpd_pool, GFP_ATOMIC|SLAB_DMA, &dma_handle); + tpd = pci_pool_alloc(he_dev->tpd_pool, GFP_ATOMIC|GFP_DMA, &dma_handle); if (tpd == NULL) return NULL; diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index 5ecea3e4fdef..fdaa471e845f 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -1215,7 +1215,7 @@ dasd_eckd_build_cp(struct dasd_device * device, struct request *req) dst = page_address(bv->bv_page) + bv->bv_offset; if (dasd_page_cache) { char *copy = kmem_cache_alloc(dasd_page_cache, - SLAB_DMA | __GFP_NOWARN); + GFP_DMA | __GFP_NOWARN); if (copy && rq_data_dir(req) == WRITE) memcpy(copy + bv->bv_offset, dst, bv->bv_len); if (copy) diff --git a/drivers/s390/block/dasd_fba.c b/drivers/s390/block/dasd_fba.c index 80926c548228..b857fd5893fd 100644 --- a/drivers/s390/block/dasd_fba.c +++ b/drivers/s390/block/dasd_fba.c @@ -308,7 +308,7 @@ dasd_fba_build_cp(struct dasd_device * device, struct request *req) dst = page_address(bv->bv_page) + bv->bv_offset; if (dasd_page_cache) { char *copy = kmem_cache_alloc(dasd_page_cache, - SLAB_DMA | __GFP_NOWARN); + GFP_DMA | __GFP_NOWARN); if (copy && rq_data_dir(req) == WRITE) memcpy(copy + bv->bv_offset, dst, bv->bv_len); if (copy) diff --git a/drivers/usb/core/buffer.c b/drivers/usb/core/buffer.c index 840442a25b61..c3915dc28608 100644 --- a/drivers/usb/core/buffer.c +++ b/drivers/usb/core/buffer.c @@ -93,7 +93,7 @@ void hcd_buffer_destroy (struct usb_hcd *hcd) } -/* sometimes alloc/free could use kmalloc with SLAB_DMA, for +/* sometimes alloc/free could use kmalloc with GFP_DMA, for * better sharing and to leverage mm/slab.c intelligence. */ diff --git a/include/linux/slab.h b/include/linux/slab.h index 639f65efa46e..fbcfc208f52b 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -18,9 +18,6 @@ typedef struct kmem_cache kmem_cache_t; #include /* kmalloc_sizes.h needs PAGE_SIZE */ #include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ -/* flags for kmem_cache_alloc() */ -#define SLAB_DMA GFP_DMA - /* flags to pass to kmem_cache_create(). * The first 3 are only valid when the allocator as been build * SLAB_DEBUG_SUPPORT. diff --git a/mm/slab.c b/mm/slab.c index 1f374c1df018..bb831ba63e1e 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2637,7 +2637,7 @@ static void cache_init_objs(struct kmem_cache *cachep, static void kmem_flagcheck(struct kmem_cache *cachep, gfp_t flags) { - if (flags & SLAB_DMA) + if (flags & GFP_DMA) BUG_ON(!(cachep->gfpflags & GFP_DMA)); else BUG_ON(cachep->gfpflags & GFP_DMA); @@ -2721,7 +2721,7 @@ static int cache_grow(struct kmem_cache *cachep, gfp_t flags, int nodeid) * Be lazy and only check for valid flags here, keeping it out of the * critical path in kmem_cache_alloc(). */ - BUG_ON(flags & ~(SLAB_DMA | GFP_LEVEL_MASK | __GFP_NO_GROW)); + BUG_ON(flags & ~(GFP_DMA | GFP_LEVEL_MASK | __GFP_NO_GROW)); if (flags & __GFP_NO_GROW) return 0; -- cgit v1.2.3 From e18b890bb0881bbab6f4f1a6cd20d9c60d66b003 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Wed, 6 Dec 2006 20:33:20 -0800 Subject: [PATCH] slab: remove kmem_cache_t Replace all uses of kmem_cache_t with struct kmem_cache. The patch was generated using the following script: #!/bin/sh # # Replace one string by another in all the kernel sources. # set -e for file in `find * -name "*.c" -o -name "*.h"|xargs grep -l $1`; do quilt add $file sed -e "1,\$s/$1/$2/g" $file >/tmp/$$ mv /tmp/$$ $file quilt refresh done The script was run like this sh replace kmem_cache_t "struct kmem_cache" Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/DMA-API.txt | 4 ++-- arch/arm/mach-s3c2410/dma.c | 4 ++-- arch/arm26/mm/memc.c | 6 +++--- arch/frv/mm/pgalloc.c | 6 +++--- arch/i386/mm/init.c | 4 ++-- arch/i386/mm/pgtable.c | 6 +++--- arch/ia64/ia32/ia32_support.c | 2 +- arch/ia64/ia32/sys_ia32.c | 2 +- arch/powerpc/kernel/rtas_flash.c | 4 ++-- arch/powerpc/mm/hugetlbpage.c | 2 +- arch/powerpc/mm/init_64.c | 6 +++--- arch/powerpc/platforms/cell/spufs/inode.c | 4 ++-- arch/sh/kernel/cpu/sh4/sq.c | 2 +- arch/sh/mm/pmb.c | 6 +++--- arch/sparc64/mm/init.c | 4 ++-- arch/sparc64/mm/tsb.c | 2 +- block/cfq-iosched.c | 4 ++-- block/ll_rw_blk.c | 6 +++--- drivers/block/aoe/aoeblk.c | 2 +- drivers/ieee1394/eth1394.c | 2 +- drivers/md/dm-crypt.c | 2 +- drivers/md/dm-mpath.c | 2 +- drivers/md/dm-snap.c | 6 +++--- drivers/md/dm.c | 4 ++-- drivers/md/kcopyd.c | 2 +- drivers/md/raid5.c | 4 ++-- drivers/message/i2o/i2o_block.h | 2 +- drivers/pci/msi.c | 2 +- drivers/s390/block/dasd_devmap.c | 2 +- drivers/s390/block/dasd_int.h | 2 +- drivers/s390/scsi/zfcp_def.h | 6 +++--- drivers/scsi/aic94xx/aic94xx.h | 4 ++-- drivers/scsi/aic94xx/aic94xx_hwi.c | 2 +- drivers/scsi/aic94xx/aic94xx_init.c | 4 ++-- drivers/scsi/libsas/sas_init.c | 2 +- drivers/scsi/qla2xxx/qla_os.c | 2 +- drivers/scsi/qla4xxx/ql4_os.c | 2 +- drivers/scsi/scsi.c | 2 +- drivers/scsi/scsi_lib.c | 4 ++-- drivers/scsi/scsi_tgt_lib.c | 2 +- drivers/usb/host/hc_crisv10.c | 6 +++--- drivers/usb/host/uhci-hcd.c | 2 +- drivers/usb/mon/mon_text.c | 6 +++--- fs/adfs/super.c | 4 ++-- fs/affs/super.c | 4 ++-- fs/afs/super.c | 6 +++--- fs/aio.c | 4 ++-- fs/befs/linuxvfs.c | 4 ++-- fs/bfs/inode.c | 4 ++-- fs/bio.c | 4 ++-- fs/block_dev.c | 4 ++-- fs/buffer.c | 4 ++-- fs/cifs/cifsfs.c | 14 +++++++------- fs/cifs/transport.c | 2 +- fs/coda/inode.c | 4 ++-- fs/configfs/configfs_internal.h | 2 +- fs/configfs/mount.c | 2 +- fs/dcache.c | 6 +++--- fs/dcookies.c | 2 +- fs/dlm/memory.c | 2 +- fs/dnotify.c | 2 +- fs/dquot.c | 2 +- fs/ecryptfs/main.c | 2 +- fs/efs/super.c | 4 ++-- fs/eventpoll.c | 4 ++-- fs/ext2/super.c | 4 ++-- fs/ext3/super.c | 4 ++-- fs/ext4/super.c | 4 ++-- fs/fat/cache.c | 4 ++-- fs/fat/inode.c | 4 ++-- fs/fcntl.c | 2 +- fs/freevxfs/vxfs_inode.c | 2 +- fs/fuse/dev.c | 2 +- fs/fuse/inode.c | 4 ++-- fs/gfs2/main.c | 4 ++-- fs/gfs2/util.c | 6 +++--- fs/gfs2/util.h | 6 +++--- fs/hfs/super.c | 4 ++-- fs/hfsplus/super.c | 4 ++-- fs/hpfs/super.c | 4 ++-- fs/hugetlbfs/inode.c | 4 ++-- fs/inode.c | 4 ++-- fs/inotify_user.c | 4 ++-- fs/isofs/inode.c | 4 ++-- fs/jbd/journal.c | 6 +++--- fs/jbd/revoke.c | 4 ++-- fs/jbd2/journal.c | 6 +++--- fs/jbd2/revoke.c | 4 ++-- fs/jffs/inode-v23.c | 4 ++-- fs/jffs/jffs_fm.c | 4 ++-- fs/jffs2/malloc.c | 18 +++++++++--------- fs/jffs2/super.c | 4 ++-- fs/jfs/jfs_metapage.c | 4 ++-- fs/jfs/super.c | 4 ++-- fs/locks.c | 4 ++-- fs/mbcache.c | 2 +- fs/minix/inode.c | 4 ++-- fs/namespace.c | 2 +- fs/ncpfs/inode.c | 4 ++-- fs/nfs/direct.c | 2 +- fs/nfs/inode.c | 4 ++-- fs/nfs/pagelist.c | 2 +- fs/nfs/read.c | 2 +- fs/nfs/write.c | 2 +- fs/nfsd/nfs4state.c | 10 +++++----- fs/ocfs2/dlm/dlmfs.c | 4 ++-- fs/ocfs2/dlm/dlmmaster.c | 2 +- fs/ocfs2/extent_map.c | 2 +- fs/ocfs2/inode.h | 2 +- fs/ocfs2/super.c | 4 ++-- fs/ocfs2/uptodate.c | 2 +- fs/openpromfs/inode.c | 4 ++-- fs/proc/inode.c | 4 ++-- fs/qnx4/inode.c | 4 ++-- fs/reiserfs/super.c | 4 ++-- fs/romfs/inode.c | 4 ++-- fs/smbfs/inode.c | 4 ++-- fs/smbfs/request.c | 2 +- fs/sysfs/mount.c | 2 +- fs/sysfs/sysfs.h | 2 +- fs/sysv/inode.c | 4 ++-- fs/udf/super.c | 4 ++-- fs/ufs/super.c | 4 ++-- include/acpi/platform/aclinux.h | 2 +- include/asm-arm26/pgalloc.h | 2 +- include/asm-i386/pgtable.h | 10 +++++----- include/asm-powerpc/pgalloc.h | 2 +- include/asm-sparc64/pgalloc.h | 2 +- include/linux/delayacct.h | 2 +- include/linux/i2o.h | 2 +- include/linux/jbd.h | 2 +- include/linux/jbd2.h | 2 +- include/linux/raid/raid5.h | 2 +- include/linux/rmap.h | 2 +- include/linux/skbuff.h | 2 +- include/linux/taskstats_kern.h | 2 +- include/net/dst.h | 2 +- include/net/inet_hashtables.h | 6 +++--- include/net/neighbour.h | 2 +- include/net/netfilter/nf_conntrack_expect.h | 2 +- include/net/request_sock.h | 2 +- include/net/sock.h | 2 +- include/net/timewait_sock.h | 2 +- include/scsi/libsas.h | 4 ++-- ipc/mqueue.c | 4 ++-- kernel/delayacct.c | 2 +- kernel/fork.c | 16 ++++++++-------- kernel/pid.c | 2 +- kernel/posix-timers.c | 2 +- kernel/signal.c | 2 +- kernel/taskstats.c | 2 +- kernel/user.c | 2 +- lib/idr.c | 4 ++-- lib/radix-tree.c | 4 ++-- net/bridge/br_fdb.c | 2 +- net/core/flow.c | 2 +- net/core/skbuff.c | 8 ++++---- net/core/sock.c | 2 +- net/dccp/ackvec.c | 4 ++-- net/dccp/ccid.c | 6 +++--- net/dccp/ccid.h | 4 ++-- net/dccp/ccids/lib/loss_interval.h | 2 +- net/dccp/ccids/lib/packet_history.h | 4 ++-- net/decnet/dn_table.c | 2 +- net/ipv4/fib_hash.c | 4 ++-- net/ipv4/fib_trie.c | 2 +- net/ipv4/inet_hashtables.c | 4 ++-- net/ipv4/inetpeer.c | 2 +- net/ipv4/ipmr.c | 2 +- net/ipv4/ipvs/ip_vs_conn.c | 2 +- net/ipv4/netfilter/ip_conntrack_core.c | 4 ++-- net/ipv6/ip6_fib.c | 2 +- net/ipv6/xfrm6_tunnel.c | 2 +- net/netfilter/nf_conntrack_core.c | 6 +++--- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/xt_hashlimit.c | 2 +- net/sctp/protocol.c | 4 ++-- net/sctp/sm_make_chunk.c | 2 +- net/sctp/socket.c | 2 +- net/socket.c | 4 ++-- net/sunrpc/rpc_pipe.c | 4 ++-- net/sunrpc/sched.c | 4 ++-- net/tipc/handler.c | 2 +- net/xfrm/xfrm_input.c | 2 +- net/xfrm/xfrm_policy.c | 2 +- security/keys/key.c | 2 +- security/selinux/avc.c | 2 +- security/selinux/hooks.c | 2 +- security/selinux/ss/avtab.c | 2 +- 189 files changed, 332 insertions(+), 332 deletions(-) (limited to 'include/linux') diff --git a/Documentation/DMA-API.txt b/Documentation/DMA-API.txt index 05431621c861..8621a064f7e1 100644 --- a/Documentation/DMA-API.txt +++ b/Documentation/DMA-API.txt @@ -77,7 +77,7 @@ To get this part of the dma_ API, you must #include Many drivers need lots of small dma-coherent memory regions for DMA descriptors or I/O buffers. Rather than allocating in units of a page or more using dma_alloc_coherent(), you can use DMA pools. These work -much like a kmem_cache_t, except that they use the dma-coherent allocator +much like a struct kmem_cache, except that they use the dma-coherent allocator not __get_free_pages(). Also, they understand common hardware constraints for alignment, like queue heads needing to be aligned on N byte boundaries. @@ -94,7 +94,7 @@ The pool create() routines initialize a pool of dma-coherent buffers for use with a given device. It must be called in a context which can sleep. -The "name" is for diagnostics (like a kmem_cache_t name); dev and size +The "name" is for diagnostics (like a struct kmem_cache name); dev and size are like what you'd pass to dma_alloc_coherent(). The device's hardware alignment requirement for this type of data is "align" (which is expressed in bytes, and must be a power of two). If your device has no boundary diff --git a/arch/arm/mach-s3c2410/dma.c b/arch/arm/mach-s3c2410/dma.c index 3d211dc2f2f9..01abb0ace234 100644 --- a/arch/arm/mach-s3c2410/dma.c +++ b/arch/arm/mach-s3c2410/dma.c @@ -40,7 +40,7 @@ /* io map for dma */ static void __iomem *dma_base; -static kmem_cache_t *dma_kmem; +static struct kmem_cache *dma_kmem; struct s3c24xx_dma_selection dma_sel; @@ -1271,7 +1271,7 @@ struct sysdev_class dma_sysclass = { /* kmem cache implementation */ -static void s3c2410_dma_cache_ctor(void *p, kmem_cache_t *c, unsigned long f) +static void s3c2410_dma_cache_ctor(void *p, struct kmem_cache *c, unsigned long f) { memset(p, 0, sizeof(struct s3c2410_dma_buf)); } diff --git a/arch/arm26/mm/memc.c b/arch/arm26/mm/memc.c index 34def6397c3c..f2901581d4da 100644 --- a/arch/arm26/mm/memc.c +++ b/arch/arm26/mm/memc.c @@ -24,7 +24,7 @@ #define MEMC_TABLE_SIZE (256*sizeof(unsigned long)) -kmem_cache_t *pte_cache, *pgd_cache; +struct kmem_cache *pte_cache, *pgd_cache; int page_nr; /* @@ -162,12 +162,12 @@ void __init create_memmap_holes(struct meminfo *mi) { } -static void pte_cache_ctor(void *pte, kmem_cache_t *cache, unsigned long flags) +static void pte_cache_ctor(void *pte, struct kmem_cache *cache, unsigned long flags) { memzero(pte, sizeof(pte_t) * PTRS_PER_PTE); } -static void pgd_cache_ctor(void *pgd, kmem_cache_t *cache, unsigned long flags) +static void pgd_cache_ctor(void *pgd, struct kmem_cache *cache, unsigned long flags) { memzero(pgd + MEMC_TABLE_SIZE, USER_PTRS_PER_PGD * sizeof(pgd_t)); } diff --git a/arch/frv/mm/pgalloc.c b/arch/frv/mm/pgalloc.c index f76dd03ddd99..19b13be114a2 100644 --- a/arch/frv/mm/pgalloc.c +++ b/arch/frv/mm/pgalloc.c @@ -18,7 +18,7 @@ #include pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((aligned(PAGE_SIZE))); -kmem_cache_t *pgd_cache; +struct kmem_cache *pgd_cache; pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { @@ -100,7 +100,7 @@ static inline void pgd_list_del(pgd_t *pgd) set_page_private(next, (unsigned long) pprev); } -void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_ctor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; @@ -120,7 +120,7 @@ void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) } /* never called when PTRS_PER_PMD > 1 */ -void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_dtor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; /* can be called from interrupt context */ diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 167416155ee4..a6a8e397dd88 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -699,8 +699,8 @@ int remove_memory(u64 start, u64 size) #endif #endif -kmem_cache_t *pgd_cache; -kmem_cache_t *pmd_cache; +struct kmem_cache *pgd_cache; +struct kmem_cache *pmd_cache; void __init pgtable_cache_init(void) { diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index 10126e3f8174..33be236fc6af 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -193,7 +193,7 @@ struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) return pte; } -void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags) +void pmd_ctor(void *pmd, struct kmem_cache *cache, unsigned long flags) { memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t)); } @@ -233,7 +233,7 @@ static inline void pgd_list_del(pgd_t *pgd) set_page_private(next, (unsigned long)pprev); } -void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_ctor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; @@ -253,7 +253,7 @@ void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused) } /* never called when PTRS_PER_PMD > 1 */ -void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused) +void pgd_dtor(void *pgd, struct kmem_cache *cache, unsigned long unused) { unsigned long flags; /* can be called from interrupt context */ diff --git a/arch/ia64/ia32/ia32_support.c b/arch/ia64/ia32/ia32_support.c index c187743965a0..6af400a12ca1 100644 --- a/arch/ia64/ia32/ia32_support.c +++ b/arch/ia64/ia32/ia32_support.c @@ -249,7 +249,7 @@ ia32_init (void) #if PAGE_SHIFT > IA32_PAGE_SHIFT { - extern kmem_cache_t *partial_page_cachep; + extern struct kmem_cache *partial_page_cachep; partial_page_cachep = kmem_cache_create("partial_page_cache", sizeof(struct partial_page), 0, 0, diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 9d6a3f210148..a4a6e1463af8 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -254,7 +254,7 @@ mmap_subpage (struct file *file, unsigned long start, unsigned long end, int pro } /* SLAB cache for partial_page structures */ -kmem_cache_t *partial_page_cachep; +struct kmem_cache *partial_page_cachep; /* * init partial_page_list. diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index b9561d300516..7d0f13fecc0e 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -101,7 +101,7 @@ struct flash_block_list_header { /* just the header of flash_block_list */ static struct flash_block_list_header rtas_firmware_flash_list = {0, NULL}; /* Use slab cache to guarantee 4k alignment */ -static kmem_cache_t *flash_block_cache = NULL; +static struct kmem_cache *flash_block_cache = NULL; #define FLASH_BLOCK_LIST_VERSION (1UL) @@ -286,7 +286,7 @@ static ssize_t rtas_flash_read(struct file *file, char __user *buf, } /* constructor for flash_block_cache */ -void rtas_block_ctor(void *ptr, kmem_cache_t *cache, unsigned long flags) +void rtas_block_ctor(void *ptr, struct kmem_cache *cache, unsigned long flags) { memset(ptr, 0, RTAS_BLK_SIZE); } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 424a8f57e155..89c836d54809 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -1047,7 +1047,7 @@ repeat: return err; } -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { memset(addr, 0, kmem_cache_size(cache)); } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 9a178549cbcf..d12a87ec5ae9 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -141,7 +141,7 @@ static int __init setup_kcore(void) } module_init(setup_kcore); -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { memset(addr, 0, kmem_cache_size(cache)); } @@ -166,9 +166,9 @@ static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { /* Hugepages need one extra cache, initialized in hugetlbpage.c. We * can't put into the tables above, because HPAGE_SHIFT is not compile * time constant. */ -kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1]; +struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1]; #else -kmem_cache_t *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; +struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; #endif void pgtable_cache_init(void) diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 7edfcc9d2853..e3af9112c026 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -40,7 +40,7 @@ #include "spufs.h" -static kmem_cache_t *spufs_inode_cache; +static struct kmem_cache *spufs_inode_cache; char *isolated_loader; static struct inode * @@ -65,7 +65,7 @@ spufs_destroy_inode(struct inode *inode) } static void -spufs_init_once(void *p, kmem_cache_t * cachep, unsigned long flags) +spufs_init_once(void *p, struct kmem_cache * cachep, unsigned long flags) { struct spufs_inode_info *ei = p; diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c index 55f43506995a..0c9ea38d2caa 100644 --- a/arch/sh/kernel/cpu/sh4/sq.c +++ b/arch/sh/kernel/cpu/sh4/sq.c @@ -38,7 +38,7 @@ struct sq_mapping { static struct sq_mapping *sq_mapping_list; static DEFINE_SPINLOCK(sq_mapping_lock); -static kmem_cache_t *sq_cache; +static struct kmem_cache *sq_cache; static unsigned long *sq_bitmap; #define store_queue_barrier() \ diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c index 92e745341e4d..b60ad83a7635 100644 --- a/arch/sh/mm/pmb.c +++ b/arch/sh/mm/pmb.c @@ -30,7 +30,7 @@ #define NR_PMB_ENTRIES 16 -static kmem_cache_t *pmb_cache; +static struct kmem_cache *pmb_cache; static unsigned long pmb_map; static struct pmb_entry pmb_init_map[] = { @@ -283,7 +283,7 @@ void pmb_unmap(unsigned long addr) } while (pmbe); } -static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags) +static void pmb_cache_ctor(void *pmb, struct kmem_cache *cachep, unsigned long flags) { struct pmb_entry *pmbe = pmb; @@ -297,7 +297,7 @@ static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags) spin_unlock_irq(&pmb_list_lock); } -static void pmb_cache_dtor(void *pmb, kmem_cache_t *cachep, unsigned long flags) +static void pmb_cache_dtor(void *pmb, struct kmem_cache *cachep, unsigned long flags) { spin_lock_irq(&pmb_list_lock); pmb_list_del(pmb); diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 09cb7fccc03a..a8e8802eed4d 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -176,9 +176,9 @@ unsigned long sparc64_kern_sec_context __read_mostly; int bigkernel = 0; -kmem_cache_t *pgtable_cache __read_mostly; +struct kmem_cache *pgtable_cache __read_mostly; -static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) +static void zero_ctor(void *addr, struct kmem_cache *cache, unsigned long flags) { clear_page(addr); } diff --git a/arch/sparc64/mm/tsb.c b/arch/sparc64/mm/tsb.c index beaa02810f0e..236d02f41a01 100644 --- a/arch/sparc64/mm/tsb.c +++ b/arch/sparc64/mm/tsb.c @@ -239,7 +239,7 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign } } -static kmem_cache_t *tsb_caches[8] __read_mostly; +static struct kmem_cache *tsb_caches[8] __read_mostly; static const char *tsb_cache_names[8] = { "tsb_8KB", diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 84e9be073180..78c6b312bd30 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -43,8 +43,8 @@ static int cfq_slice_idle = HZ / 125; #define RQ_CIC(rq) ((struct cfq_io_context*)(rq)->elevator_private) #define RQ_CFQQ(rq) ((rq)->elevator_private2) -static kmem_cache_t *cfq_pool; -static kmem_cache_t *cfq_ioc_pool; +static struct kmem_cache *cfq_pool; +static struct kmem_cache *cfq_ioc_pool; static DEFINE_PER_CPU(unsigned long, ioc_count); static struct completion *ioc_gone; diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index cc6e95f8e5d9..a4ff3271d4a8 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -44,17 +44,17 @@ static struct io_context *current_io_context(gfp_t gfp_flags, int node); /* * For the allocated request tables */ -static kmem_cache_t *request_cachep; +static struct kmem_cache *request_cachep; /* * For queue allocation */ -static kmem_cache_t *requestq_cachep; +static struct kmem_cache *requestq_cachep; /* * For io context allocations */ -static kmem_cache_t *iocontext_cachep; +static struct kmem_cache *iocontext_cachep; /* * Controlling structure to kblockd diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c index aa25f8b09fe3..478489c568a4 100644 --- a/drivers/block/aoe/aoeblk.c +++ b/drivers/block/aoe/aoeblk.c @@ -12,7 +12,7 @@ #include #include "aoe.h" -static kmem_cache_t *buf_pool_cache; +static struct kmem_cache *buf_pool_cache; static ssize_t aoedisk_show_state(struct gendisk * disk, char *page) { diff --git a/drivers/ieee1394/eth1394.c b/drivers/ieee1394/eth1394.c index 31e5cc49d61a..27d6c642415d 100644 --- a/drivers/ieee1394/eth1394.c +++ b/drivers/ieee1394/eth1394.c @@ -133,7 +133,7 @@ struct eth1394_node_info { #define ETH1394_DRIVER_NAME "eth1394" static const char driver_name[] = ETH1394_DRIVER_NAME; -static kmem_cache_t *packet_task_cache; +static struct kmem_cache *packet_task_cache; static struct hpsb_highlevel eth1394_highlevel; diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index ed2d4ef27fd8..c7bee4f2eedb 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -101,7 +101,7 @@ struct crypt_config { #define MIN_POOL_PAGES 32 #define MIN_BIO_PAGES 8 -static kmem_cache_t *_crypt_io_pool; +static struct kmem_cache *_crypt_io_pool; /* * Different IV generation algorithms: diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index e77ee6fd1044..cf8bf052138e 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -101,7 +101,7 @@ typedef int (*action_fn) (struct pgpath *pgpath); #define MIN_IOS 256 /* Mempool size */ -static kmem_cache_t *_mpio_cache; +static struct kmem_cache *_mpio_cache; struct workqueue_struct *kmultipathd; static void process_queued_ios(struct work_struct *work); diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 91c7aa1fed0e..b0ce2ce82278 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -88,8 +88,8 @@ struct pending_exception { * Hash table mapping origin volumes to lists of snapshots and * a lock to protect it */ -static kmem_cache_t *exception_cache; -static kmem_cache_t *pending_cache; +static struct kmem_cache *exception_cache; +static struct kmem_cache *pending_cache; static mempool_t *pending_pool; /* @@ -228,7 +228,7 @@ static int init_exception_table(struct exception_table *et, uint32_t size) return 0; } -static void exit_exception_table(struct exception_table *et, kmem_cache_t *mem) +static void exit_exception_table(struct exception_table *et, struct kmem_cache *mem) { struct list_head *slot; struct exception *ex, *next; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index fc4f743f3b53..7ec1b112a6d5 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -121,8 +121,8 @@ struct mapped_device { }; #define MIN_IOS 256 -static kmem_cache_t *_io_cache; -static kmem_cache_t *_tio_cache; +static struct kmem_cache *_io_cache; +static struct kmem_cache *_tio_cache; static int __init local_init(void) { diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c index b3c01496c737..b46f6c575f7e 100644 --- a/drivers/md/kcopyd.c +++ b/drivers/md/kcopyd.c @@ -203,7 +203,7 @@ struct kcopyd_job { /* FIXME: this should scale with the number of pages */ #define MIN_JOBS 512 -static kmem_cache_t *_job_cache; +static struct kmem_cache *_job_cache; static mempool_t *_job_pool; /* diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 69c3e201fa3b..52914d5cec76 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -348,7 +348,7 @@ static int grow_one_stripe(raid5_conf_t *conf) static int grow_stripes(raid5_conf_t *conf, int num) { - kmem_cache_t *sc; + struct kmem_cache *sc; int devs = conf->raid_disks; sprintf(conf->cache_name[0], "raid5/%s", mdname(conf->mddev)); @@ -397,7 +397,7 @@ static int resize_stripes(raid5_conf_t *conf, int newsize) LIST_HEAD(newstripes); struct disk_info *ndisks; int err = 0; - kmem_cache_t *sc; + struct kmem_cache *sc; int i; if (newsize <= conf->pool_size) diff --git a/drivers/message/i2o/i2o_block.h b/drivers/message/i2o/i2o_block.h index d9fdc95b440d..67f921b4419b 100644 --- a/drivers/message/i2o/i2o_block.h +++ b/drivers/message/i2o/i2o_block.h @@ -64,7 +64,7 @@ /* I2O Block OSM mempool struct */ struct i2o_block_mempool { - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *pool; }; diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 9fc9a34ef24a..9168401401bc 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -26,7 +26,7 @@ static DEFINE_SPINLOCK(msi_lock); static struct msi_desc* msi_desc[NR_IRQS] = { [0 ... NR_IRQS-1] = NULL }; -static kmem_cache_t* msi_cachep; +static struct kmem_cache* msi_cachep; static int pci_msi_enable = 1; diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c index 17fdd8c9f740..cf28ccc57948 100644 --- a/drivers/s390/block/dasd_devmap.c +++ b/drivers/s390/block/dasd_devmap.c @@ -25,7 +25,7 @@ #include "dasd_int.h" -kmem_cache_t *dasd_page_cache; +struct kmem_cache *dasd_page_cache; EXPORT_SYMBOL_GPL(dasd_page_cache); /* diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 9f52004f6fc2..dc5dd509434d 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -474,7 +474,7 @@ extern struct dasd_profile_info_t dasd_global_profile; extern unsigned int dasd_profile_level; extern struct block_device_operations dasd_device_operations; -extern kmem_cache_t *dasd_page_cache; +extern struct kmem_cache *dasd_page_cache; struct dasd_ccw_req * dasd_kmalloc_request(char *, int, int, struct dasd_device *); diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 74c0eac083e4..32933ed54b8a 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -1032,9 +1032,9 @@ struct zfcp_data { wwn_t init_wwpn; fcp_lun_t init_fcp_lun; char *driver_version; - kmem_cache_t *fsf_req_qtcb_cache; - kmem_cache_t *sr_buffer_cache; - kmem_cache_t *gid_pn_cache; + struct kmem_cache *fsf_req_qtcb_cache; + struct kmem_cache *sr_buffer_cache; + struct kmem_cache *gid_pn_cache; }; /** diff --git a/drivers/scsi/aic94xx/aic94xx.h b/drivers/scsi/aic94xx/aic94xx.h index 71a031df7a34..32f513b1b78a 100644 --- a/drivers/scsi/aic94xx/aic94xx.h +++ b/drivers/scsi/aic94xx/aic94xx.h @@ -56,8 +56,8 @@ /* 2*ITNL timeout + 1 second */ #define AIC94XX_SCB_TIMEOUT (5*HZ) -extern kmem_cache_t *asd_dma_token_cache; -extern kmem_cache_t *asd_ascb_cache; +extern struct kmem_cache *asd_dma_token_cache; +extern struct kmem_cache *asd_ascb_cache; extern char sas_addr_str[2*SAS_ADDR_SIZE + 1]; static inline void asd_stringify_sas_addr(char *p, const u8 *sas_addr) diff --git a/drivers/scsi/aic94xx/aic94xx_hwi.c b/drivers/scsi/aic94xx/aic94xx_hwi.c index af7e01134364..da94e126ca83 100644 --- a/drivers/scsi/aic94xx/aic94xx_hwi.c +++ b/drivers/scsi/aic94xx/aic94xx_hwi.c @@ -1047,7 +1047,7 @@ irqreturn_t asd_hw_isr(int irq, void *dev_id) static inline struct asd_ascb *asd_ascb_alloc(struct asd_ha_struct *asd_ha, gfp_t gfp_flags) { - extern kmem_cache_t *asd_ascb_cache; + extern struct kmem_cache *asd_ascb_cache; struct asd_seq_data *seq = &asd_ha->seq; struct asd_ascb *ascb; unsigned long flags; diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c index 42302ef05ee5..fbc82b00a418 100644 --- a/drivers/scsi/aic94xx/aic94xx_init.c +++ b/drivers/scsi/aic94xx/aic94xx_init.c @@ -450,8 +450,8 @@ static inline void asd_destroy_ha_caches(struct asd_ha_struct *asd_ha) asd_ha->scb_pool = NULL; } -kmem_cache_t *asd_dma_token_cache; -kmem_cache_t *asd_ascb_cache; +struct kmem_cache *asd_dma_token_cache; +struct kmem_cache *asd_ascb_cache; static int asd_create_global_caches(void) { diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c index d65bc4e0f214..2f0c07fc3f48 100644 --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -36,7 +36,7 @@ #include "../scsi_sas_internal.h" -kmem_cache_t *sas_task_cache; +struct kmem_cache *sas_task_cache; /*------------ SAS addr hash -----------*/ void sas_hash_addr(u8 *hashed, const u8 *sas_addr) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index cbe0cad83b68..d03523d3bf38 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -24,7 +24,7 @@ char qla2x00_version_str[40]; /* * SRB allocation cache */ -static kmem_cache_t *srb_cachep; +static struct kmem_cache *srb_cachep; /* * Ioctl related information. diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 969c9e431028..9ef693c8809a 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -19,7 +19,7 @@ char qla4xxx_version_str[40]; /* * SRB allocation cache */ -static kmem_cache_t *srb_cachep; +static struct kmem_cache *srb_cachep; /* * Module parameter information and variables diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index fafc00deaade..24cffd98ee63 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -136,7 +136,7 @@ const char * scsi_device_type(unsigned type) EXPORT_SYMBOL(scsi_device_type); struct scsi_host_cmd_pool { - kmem_cache_t *slab; + struct kmem_cache *slab; unsigned int users; char *name; unsigned int slab_flags; diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index fb616c69151f..1748e27501cd 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -36,7 +36,7 @@ struct scsi_host_sg_pool { size_t size; char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *pool; }; @@ -241,7 +241,7 @@ struct scsi_io_context { char sense[SCSI_SENSE_BUFFERSIZE]; }; -static kmem_cache_t *scsi_io_context_cache; +static struct kmem_cache *scsi_io_context_cache; static void scsi_end_async(struct request *req, int uptodate) { diff --git a/drivers/scsi/scsi_tgt_lib.c b/drivers/scsi/scsi_tgt_lib.c index 386dbae17b44..d402aff5f314 100644 --- a/drivers/scsi/scsi_tgt_lib.c +++ b/drivers/scsi/scsi_tgt_lib.c @@ -33,7 +33,7 @@ #include "scsi_tgt_priv.h" static struct workqueue_struct *scsi_tgtd; -static kmem_cache_t *scsi_tgt_cmd_cache; +static struct kmem_cache *scsi_tgt_cmd_cache; /* * TODO: this struct will be killed when the block layer supports large bios diff --git a/drivers/usb/host/hc_crisv10.c b/drivers/usb/host/hc_crisv10.c index 7fd872aa654a..9325e46a68c0 100644 --- a/drivers/usb/host/hc_crisv10.c +++ b/drivers/usb/host/hc_crisv10.c @@ -275,13 +275,13 @@ static volatile USB_SB_Desc_t TxIntrSB_zout __attribute__ ((aligned (4))); static int zout_buffer[4] __attribute__ ((aligned (4))); /* Cache for allocating new EP and SB descriptors. */ -static kmem_cache_t *usb_desc_cache; +static struct kmem_cache *usb_desc_cache; /* Cache for the registers allocated in the top half. */ -static kmem_cache_t *top_half_reg_cache; +static struct kmem_cache *top_half_reg_cache; /* Cache for the data allocated in the isoc descr top half. */ -static kmem_cache_t *isoc_compl_cache; +static struct kmem_cache *isoc_compl_cache; static struct usb_bus *etrax_usb_bus; diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c index 226bf3de8edd..e87692c31be4 100644 --- a/drivers/usb/host/uhci-hcd.c +++ b/drivers/usb/host/uhci-hcd.c @@ -81,7 +81,7 @@ MODULE_PARM_DESC(debug, "Debug level"); static char *errbuf; #define ERRBUF_LEN (32 * 1024) -static kmem_cache_t *uhci_up_cachep; /* urb_priv */ +static struct kmem_cache *uhci_up_cachep; /* urb_priv */ static void suspend_rh(struct uhci_hcd *uhci, enum uhci_rh_state new_state); static void wakeup_rh(struct uhci_hcd *uhci); diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c index 46aecc8754f1..05cf2c9a8f84 100644 --- a/drivers/usb/mon/mon_text.c +++ b/drivers/usb/mon/mon_text.c @@ -50,7 +50,7 @@ struct mon_event_text { #define SLAB_NAME_SZ 30 struct mon_reader_text { - kmem_cache_t *e_slab; + struct kmem_cache *e_slab; int nevents; struct list_head e_list; struct mon_reader r; /* In C, parent class can be placed anywhere */ @@ -63,7 +63,7 @@ struct mon_reader_text { char slab_name[SLAB_NAME_SZ]; }; -static void mon_text_ctor(void *, kmem_cache_t *, unsigned long); +static void mon_text_ctor(void *, struct kmem_cache *, unsigned long); /* * mon_text_submit @@ -450,7 +450,7 @@ const struct file_operations mon_fops_text = { /* * Slab interface: constructor. */ -static void mon_text_ctor(void *mem, kmem_cache_t *slab, unsigned long sflags) +static void mon_text_ctor(void *mem, struct kmem_cache *slab, unsigned long sflags) { /* * Nothing to initialize. No, really! diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 52eb10ca654e..e5a205c9f42f 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -212,7 +212,7 @@ static int adfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t *adfs_inode_cachep; +static struct kmem_cache *adfs_inode_cachep; static struct inode *adfs_alloc_inode(struct super_block *sb) { @@ -228,7 +228,7 @@ static void adfs_destroy_inode(struct inode *inode) kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; diff --git a/fs/affs/super.c b/fs/affs/super.c index 81c73ec09f66..3de93e799949 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -66,7 +66,7 @@ affs_write_super(struct super_block *sb) pr_debug("AFFS: write_super() at %lu, clean=%d\n", get_seconds(), clean); } -static kmem_cache_t * affs_inode_cachep; +static struct kmem_cache * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { @@ -83,7 +83,7 @@ static void affs_destroy_inode(struct inode *inode) kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; diff --git a/fs/afs/super.c b/fs/afs/super.c index c6ead009bf78..9a351c4c7564 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -35,7 +35,7 @@ struct afs_mount_params { struct afs_volume *volume; }; -static void afs_i_init_once(void *foo, kmem_cache_t *cachep, +static void afs_i_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags); static int afs_get_sb(struct file_system_type *fs_type, @@ -65,7 +65,7 @@ static struct super_operations afs_super_ops = { .put_super = afs_put_super, }; -static kmem_cache_t *afs_inode_cachep; +static struct kmem_cache *afs_inode_cachep; static atomic_t afs_count_active_inodes; /*****************************************************************************/ @@ -384,7 +384,7 @@ static void afs_put_super(struct super_block *sb) /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(void *_vnode, kmem_cache_t *cachep, +static void afs_i_init_once(void *_vnode, struct kmem_cache *cachep, unsigned long flags) { struct afs_vnode *vnode = (struct afs_vnode *) _vnode; diff --git a/fs/aio.c b/fs/aio.c index 287a1bc7a182..13aa9298abf4 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -47,8 +47,8 @@ unsigned long aio_nr; /* current system wide number of aio requests */ unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ /*----end sysctl variables---*/ -static kmem_cache_t *kiocb_cachep; -static kmem_cache_t *kioctx_cachep; +static struct kmem_cache *kiocb_cachep; +static struct kmem_cache *kioctx_cachep; static struct workqueue_struct *aio_wq; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 995348df94ad..bce402eee554 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -61,7 +61,7 @@ static const struct super_operations befs_sops = { }; /* slab cache for befs_inode_info objects */ -static kmem_cache_t *befs_inode_cachep; +static struct kmem_cache *befs_inode_cachep; static const struct file_operations befs_dir_operations = { .read = generic_read_dir, @@ -289,7 +289,7 @@ befs_destroy_inode(struct inode *inode) kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 2e45123c8f7a..eac175ed9f44 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -228,7 +228,7 @@ static void bfs_write_super(struct super_block *s) unlock_kernel(); } -static kmem_cache_t * bfs_inode_cachep; +static struct kmem_cache * bfs_inode_cachep; static struct inode *bfs_alloc_inode(struct super_block *sb) { @@ -244,7 +244,7 @@ static void bfs_destroy_inode(struct inode *inode) kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bfs_inode_info *bi = foo; diff --git a/fs/bio.c b/fs/bio.c index 50c40ce2cead..7ec737eda72b 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -30,7 +30,7 @@ #define BIO_POOL_SIZE 256 -static kmem_cache_t *bio_slab __read_mostly; +static struct kmem_cache *bio_slab __read_mostly; #define BIOVEC_NR_POOLS 6 @@ -44,7 +44,7 @@ mempool_t *bio_split_pool __read_mostly; struct biovec_slab { int nr_vecs; char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; }; /* diff --git a/fs/block_dev.c b/fs/block_dev.c index 063506705f27..13816b4d76f6 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -235,7 +235,7 @@ static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(bdev_lock); -static kmem_cache_t * bdev_cachep __read_mostly; +static struct kmem_cache * bdev_cachep __read_mostly; static struct inode *bdev_alloc_inode(struct super_block *sb) { @@ -253,7 +253,7 @@ static void bdev_destroy_inode(struct inode *inode) kmem_cache_free(bdev_cachep, bdi); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; diff --git a/fs/buffer.c b/fs/buffer.c index 35527dca1dbc..a8ca0ac21488 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2908,7 +2908,7 @@ asmlinkage long sys_bdflush(int func, long data) /* * Buffer-head allocation */ -static kmem_cache_t *bh_cachep; +static struct kmem_cache *bh_cachep; /* * Once the number of bh's in the machine exceeds this level, we start @@ -2961,7 +2961,7 @@ void free_buffer_head(struct buffer_head *bh) EXPORT_SYMBOL(free_buffer_head); static void -init_buffer_head(void *data, kmem_cache_t *cachep, unsigned long flags) +init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags) { if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 84168629cea3..e6b5866e5001 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -81,7 +81,7 @@ extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static int cifs_read_super(struct super_block *sb, void *data, @@ -232,11 +232,11 @@ static int cifs_permission(struct inode * inode, int mask, struct nameidata *nd) return generic_permission(inode, mask, NULL); } -static kmem_cache_t *cifs_inode_cachep; -static kmem_cache_t *cifs_req_cachep; -static kmem_cache_t *cifs_mid_cachep; -kmem_cache_t *cifs_oplock_cachep; -static kmem_cache_t *cifs_sm_req_cachep; +static struct kmem_cache *cifs_inode_cachep; +static struct kmem_cache *cifs_req_cachep; +static struct kmem_cache *cifs_mid_cachep; +struct kmem_cache *cifs_oplock_cachep; +static struct kmem_cache *cifs_sm_req_cachep; mempool_t *cifs_sm_req_poolp; mempool_t *cifs_req_poolp; mempool_t *cifs_mid_poolp; @@ -668,7 +668,7 @@ const struct file_operations cifs_dir_ops = { }; static void -cifs_init_once(void *inode, kmem_cache_t * cachep, unsigned long flags) +cifs_init_once(void *inode, struct kmem_cache * cachep, unsigned long flags) { struct cifsInodeInfo *cifsi = inode; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 1f727765a8ea..f80007eaebf4 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -34,7 +34,7 @@ #include "cifs_debug.h" extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static struct mid_q_entry * AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 50cedd2617d6..b64659fa82d0 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -38,7 +38,7 @@ static void coda_clear_inode(struct inode *); static void coda_put_super(struct super_block *); static int coda_statfs(struct dentry *dentry, struct kstatfs *buf); -static kmem_cache_t * coda_inode_cachep; +static struct kmem_cache * coda_inode_cachep; static struct inode *coda_alloc_inode(struct super_block *sb) { @@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode) kmem_cache_free(coda_inode_cachep, ITOC(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index 3f4ff7a242b9..f92cd303d2c9 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -49,7 +49,7 @@ struct configfs_dirent { #define CONFIGFS_NOT_PINNED (CONFIGFS_ITEM_ATTR) extern struct vfsmount * configfs_mount; -extern kmem_cache_t *configfs_dir_cachep; +extern struct kmem_cache *configfs_dir_cachep; extern int configfs_is_root(struct config_item *item); diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index 68bd5c93ca52..ed678529ebb2 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -38,7 +38,7 @@ struct vfsmount * configfs_mount = NULL; struct super_block * configfs_sb = NULL; -kmem_cache_t *configfs_dir_cachep; +struct kmem_cache *configfs_dir_cachep; static int configfs_mnt_count = 0; static struct super_operations configfs_ops = { diff --git a/fs/dcache.c b/fs/dcache.c index fd4a428998ef..a7c67cee5d83 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -43,7 +43,7 @@ static __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); EXPORT_SYMBOL(dcache_lock); -static kmem_cache_t *dentry_cache __read_mostly; +static struct kmem_cache *dentry_cache __read_mostly; #define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname)) @@ -2072,10 +2072,10 @@ static void __init dcache_init(unsigned long mempages) } /* SLAB cache for __getname() consumers */ -kmem_cache_t *names_cachep __read_mostly; +struct kmem_cache *names_cachep __read_mostly; /* SLAB cache for file structures */ -kmem_cache_t *filp_cachep __read_mostly; +struct kmem_cache *filp_cachep __read_mostly; EXPORT_SYMBOL(d_genocide); diff --git a/fs/dcookies.c b/fs/dcookies.c index 0c4b0674854b..21af1629f9bc 100644 --- a/fs/dcookies.c +++ b/fs/dcookies.c @@ -37,7 +37,7 @@ struct dcookie_struct { static LIST_HEAD(dcookie_users); static DEFINE_MUTEX(dcookie_mutex); -static kmem_cache_t *dcookie_cache __read_mostly; +static struct kmem_cache *dcookie_cache __read_mostly; static struct list_head *dcookie_hashtable __read_mostly; static size_t hash_size __read_mostly; diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index 989b608fd836..5352b03ff5aa 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -15,7 +15,7 @@ #include "config.h" #include "memory.h" -static kmem_cache_t *lkb_cache; +static struct kmem_cache *lkb_cache; int dlm_memory_init(void) diff --git a/fs/dnotify.c b/fs/dnotify.c index e778b1737b79..1f26a2b9eee1 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -23,7 +23,7 @@ int dir_notify_enable __read_mostly = 1; -static kmem_cache_t *dn_cache __read_mostly; +static struct kmem_cache *dn_cache __read_mostly; static void redo_inode_mask(struct inode *inode) { diff --git a/fs/dquot.c b/fs/dquot.c index c6ae6c0e0cfc..f9cd5e23ebdf 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -131,7 +131,7 @@ static struct quota_format_type *quota_formats; /* List of registered formats */ static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES; /* SLAB cache for dquot structures */ -static kmem_cache_t *dquot_cachep; +static struct kmem_cache *dquot_cachep; int register_quota_format(struct quota_format_type *fmt) { diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a2c6ccbce300..306f8fbd1a08 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -546,7 +546,7 @@ inode_info_init_once(void *vptr, struct kmem_cache *cachep, unsigned long flags) } static struct ecryptfs_cache_info { - kmem_cache_t **cache; + struct kmem_cache **cache; const char *name; size_t size; void (*ctor)(void*, struct kmem_cache *, unsigned long); diff --git a/fs/efs/super.c b/fs/efs/super.c index 69b15a996cfc..dfebf21289f4 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -52,7 +52,7 @@ static struct pt_types sgi_pt_types[] = { }; -static kmem_cache_t * efs_inode_cachep; +static struct kmem_cache * efs_inode_cachep; static struct inode *efs_alloc_inode(struct super_block *sb) { @@ -68,7 +68,7 @@ static void efs_destroy_inode(struct inode *inode) kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index f5c88435c6be..88a6f8d0b88e 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -283,10 +283,10 @@ static struct mutex epmutex; static struct poll_safewake psw; /* Slab cache used to allocate "struct epitem" */ -static kmem_cache_t *epi_cache __read_mostly; +static struct kmem_cache *epi_cache __read_mostly; /* Slab cache used to allocate "struct eppoll_entry" */ -static kmem_cache_t *pwq_cache __read_mostly; +static struct kmem_cache *pwq_cache __read_mostly; /* Virtual fs used to allocate inodes for eventpoll files */ static struct vfsmount *eventpoll_mnt __read_mostly; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 85c237e73853..3aafb1dd917a 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -135,7 +135,7 @@ static void ext2_put_super (struct super_block * sb) return; } -static kmem_cache_t * ext2_inode_cachep; +static struct kmem_cache * ext2_inode_cachep; static struct inode *ext2_alloc_inode(struct super_block *sb) { @@ -156,7 +156,7 @@ static void ext2_destroy_inode(struct inode *inode) kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 0cf633f0cfa3..9856565d4ddc 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -436,7 +436,7 @@ static void ext3_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext3_inode_cachep; +static struct kmem_cache *ext3_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -462,7 +462,7 @@ static void ext3_destroy_inode(struct inode *inode) kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index c730cbc84030..f2e8c4aa0fd2 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -486,7 +486,7 @@ static void ext4_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext4_inode_cachep; +static struct kmem_cache *ext4_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -513,7 +513,7 @@ static void ext4_destroy_inode(struct inode *inode) kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 8c272278455c..05c2941c74f2 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -34,9 +34,9 @@ static inline int fat_max_cache(struct inode *inode) return FAT_MAX_CACHE; } -static kmem_cache_t *fat_cache_cachep; +static struct kmem_cache *fat_cache_cachep; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct fat_cache *cache = (struct fat_cache *)foo; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index b58fd0c9f3cd..a9e4688582a2 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -477,7 +477,7 @@ static void fat_put_super(struct super_block *sb) kfree(sbi); } -static kmem_cache_t *fat_inode_cachep; +static struct kmem_cache *fat_inode_cachep; static struct inode *fat_alloc_inode(struct super_block *sb) { @@ -493,7 +493,7 @@ static void fat_destroy_inode(struct inode *inode) kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; diff --git a/fs/fcntl.c b/fs/fcntl.c index c03dc9cb21cb..4740d35e52cd 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -553,7 +553,7 @@ int send_sigurg(struct fown_struct *fown) } static DEFINE_RWLOCK(fasync_lock); -static kmem_cache_t *fasync_cache __read_mostly; +static struct kmem_cache *fasync_cache __read_mostly; /* * fasync_helper() is used by some character device drivers (mainly mice) diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index d2dd0d700077..0b7ae897cb78 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -46,7 +46,7 @@ extern const struct address_space_operations vxfs_immed_aops; extern struct inode_operations vxfs_immed_symlink_iops; -kmem_cache_t *vxfs_inode_cachep; +struct kmem_cache *vxfs_inode_cachep; #ifdef DIAGNOSTIC diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 8c15139f2756..357764d85ff1 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -19,7 +19,7 @@ MODULE_ALIAS_MISCDEV(FUSE_MINOR); -static kmem_cache_t *fuse_req_cachep; +static struct kmem_cache *fuse_req_cachep; static struct fuse_conn *fuse_get_conn(struct file *file) { diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index e039e2047cce..2bdc652b8b46 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -22,7 +22,7 @@ MODULE_AUTHOR("Miklos Szeredi "); MODULE_DESCRIPTION("Filesystem in Userspace"); MODULE_LICENSE("GPL"); -static kmem_cache_t *fuse_inode_cachep; +static struct kmem_cache *fuse_inode_cachep; struct list_head fuse_conn_list; DEFINE_MUTEX(fuse_mutex); @@ -601,7 +601,7 @@ static struct file_system_type fuse_fs_type = { static decl_subsys(fuse, NULL, NULL); static decl_subsys(connections, NULL, NULL); -static void fuse_inode_init_once(void *foo, kmem_cache_t *cachep, +static void fuse_inode_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct inode * inode = foo; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 9889c1eacec1..7c1a9e22a526 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -25,7 +25,7 @@ #include "util.h" #include "glock.h" -static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_inode_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_inode *ip = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == @@ -37,7 +37,7 @@ static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long } } -static void gfs2_init_glock_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_glock_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_glock *gl = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 196c604faadc..e5707a9f78c2 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -23,9 +23,9 @@ #include "lm.h" #include "util.h" -kmem_cache_t *gfs2_glock_cachep __read_mostly; -kmem_cache_t *gfs2_inode_cachep __read_mostly; -kmem_cache_t *gfs2_bufdata_cachep __read_mostly; +struct kmem_cache *gfs2_glock_cachep __read_mostly; +struct kmem_cache *gfs2_inode_cachep __read_mostly; +struct kmem_cache *gfs2_bufdata_cachep __read_mostly; void gfs2_assert_i(struct gfs2_sbd *sdp) { diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index 76a50899fe9e..7984dcf89ad0 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -146,9 +146,9 @@ int gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, gfs2_io_error_bh_i((sdp), (bh), __FUNCTION__, __FILE__, __LINE__); -extern kmem_cache_t *gfs2_glock_cachep; -extern kmem_cache_t *gfs2_inode_cachep; -extern kmem_cache_t *gfs2_bufdata_cachep; +extern struct kmem_cache *gfs2_glock_cachep; +extern struct kmem_cache *gfs2_inode_cachep; +extern struct kmem_cache *gfs2_bufdata_cachep; static inline unsigned int gfs2_tune_get_i(struct gfs2_tune *gt, unsigned int *p) diff --git a/fs/hfs/super.c b/fs/hfs/super.c index ffc6409132c8..a36987966004 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -24,7 +24,7 @@ #include "hfs_fs.h" #include "btree.h" -static kmem_cache_t *hfs_inode_cachep; +static struct kmem_cache *hfs_inode_cachep; MODULE_LICENSE("GPL"); @@ -430,7 +430,7 @@ static struct file_system_type hfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfs_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfs_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfs_inode_info *i = p; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 4a0c70c76c8a..0f513c6bf843 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -434,7 +434,7 @@ MODULE_AUTHOR("Brad Boyer"); MODULE_DESCRIPTION("Extended Macintosh Filesystem"); MODULE_LICENSE("GPL"); -static kmem_cache_t *hfsplus_inode_cachep; +static struct kmem_cache *hfsplus_inode_cachep; static struct inode *hfsplus_alloc_inode(struct super_block *sb) { @@ -467,7 +467,7 @@ static struct file_system_type hfsplus_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfsplus_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfsplus_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfsplus_inode_info *i = p; diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 46ceadd6f16a..34d68e211714 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -160,7 +160,7 @@ static int hpfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * hpfs_inode_cachep; +static struct kmem_cache * hpfs_inode_cachep; static struct inode *hpfs_alloc_inode(struct super_block *sb) { @@ -177,7 +177,7 @@ static void hpfs_destroy_inode(struct inode *inode) kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 36e52173a54a..0706f5aac6a2 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -513,7 +513,7 @@ static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo) } -static kmem_cache_t *hugetlbfs_inode_cachep; +static struct kmem_cache *hugetlbfs_inode_cachep; static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) { @@ -545,7 +545,7 @@ static const struct address_space_operations hugetlbfs_aops = { }; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; diff --git a/fs/inode.c b/fs/inode.c index dd15984d51a8..699aa4f7aa74 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -97,7 +97,7 @@ static DEFINE_MUTEX(iprune_mutex); */ struct inodes_stat_t inodes_stat; -static kmem_cache_t * inode_cachep __read_mostly; +static struct kmem_cache * inode_cachep __read_mostly; static struct inode *alloc_inode(struct super_block *sb) { @@ -209,7 +209,7 @@ void inode_init_once(struct inode *inode) EXPORT_SYMBOL(inode_init_once); -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct inode * inode = (struct inode *) foo; diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 017cb0f134d6..e1956e6f116c 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -34,8 +34,8 @@ #include -static kmem_cache_t *watch_cachep __read_mostly; -static kmem_cache_t *event_cachep __read_mostly; +static struct kmem_cache *watch_cachep __read_mostly; +static struct kmem_cache *event_cachep __read_mostly; static struct vfsmount *inotify_mnt __read_mostly; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 4b6381cd2cf4..ea55b6c469ec 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -57,7 +57,7 @@ static void isofs_put_super(struct super_block *sb) static void isofs_read_inode(struct inode *); static int isofs_statfs (struct dentry *, struct kstatfs *); -static kmem_cache_t *isofs_inode_cachep; +static struct kmem_cache *isofs_inode_cachep; static struct inode *isofs_alloc_inode(struct super_block *sb) { @@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode) kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct iso_inode_info *ei = foo; diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index b85c686b60db..a8774bed20b6 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -1630,7 +1630,7 @@ void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" }; @@ -1693,7 +1693,7 @@ void jbd_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *journal_head_cache; +static struct kmem_cache *journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -1996,7 +1996,7 @@ static void __exit remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd_handle_cache; +struct kmem_cache *jbd_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index c532429d8d9b..d204ab394f36 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -70,8 +70,8 @@ #include #endif -static kmem_cache_t *revoke_record_cache; -static kmem_cache_t *revoke_table_cache; +static struct kmem_cache *revoke_record_cache; +static struct kmem_cache *revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c60f378b0f76..50356019ae30 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1641,7 +1641,7 @@ void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd2_1k", "jbd2_2k", "jbd2_4k", NULL, "jbd2_8k" }; @@ -1704,7 +1704,7 @@ void jbd2_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *jbd2_journal_head_cache; +static struct kmem_cache *jbd2_journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -2007,7 +2007,7 @@ static void __exit jbd2_remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd2_handle_cache; +struct kmem_cache *jbd2_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 380d19917f37..f506646ad0ff 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -70,8 +70,8 @@ #include #endif -static kmem_cache_t *jbd2_revoke_record_cache; -static kmem_cache_t *jbd2_revoke_table_cache; +static struct kmem_cache *jbd2_revoke_record_cache; +static struct kmem_cache *jbd2_revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 3f7899ea4cba..9f15bce92022 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -61,8 +61,8 @@ static const struct file_operations jffs_dir_operations; static struct inode_operations jffs_dir_inode_operations; static const struct address_space_operations jffs_address_operations; -kmem_cache_t *node_cache = NULL; -kmem_cache_t *fm_cache = NULL; +struct kmem_cache *node_cache = NULL; +struct kmem_cache *fm_cache = NULL; /* Called by the VFS at mount time to initialize the whole file system. */ static int jffs_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c index 29b68d939bd9..077258b2103e 100644 --- a/fs/jffs/jffs_fm.c +++ b/fs/jffs/jffs_fm.c @@ -29,8 +29,8 @@ static int jffs_mark_obsolete(struct jffs_fmcontrol *fmc, __u32 fm_offset); static struct jffs_fm *jffs_alloc_fm(void); static void jffs_free_fm(struct jffs_fm *n); -extern kmem_cache_t *fm_cache; -extern kmem_cache_t *node_cache; +extern struct kmem_cache *fm_cache; +extern struct kmem_cache *node_cache; #if CONFIG_JFFS_FS_VERBOSE > 0 void diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 33f291005012..83f9881ec4cc 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -19,16 +19,16 @@ /* These are initialised to NULL in the kernel startup code. If you're porting to other operating systems, beware */ -static kmem_cache_t *full_dnode_slab; -static kmem_cache_t *raw_dirent_slab; -static kmem_cache_t *raw_inode_slab; -static kmem_cache_t *tmp_dnode_info_slab; -static kmem_cache_t *raw_node_ref_slab; -static kmem_cache_t *node_frag_slab; -static kmem_cache_t *inode_cache_slab; +static struct kmem_cache *full_dnode_slab; +static struct kmem_cache *raw_dirent_slab; +static struct kmem_cache *raw_inode_slab; +static struct kmem_cache *tmp_dnode_info_slab; +static struct kmem_cache *raw_node_ref_slab; +static struct kmem_cache *node_frag_slab; +static struct kmem_cache *inode_cache_slab; #ifdef CONFIG_JFFS2_FS_XATTR -static kmem_cache_t *xattr_datum_cache; -static kmem_cache_t *xattr_ref_cache; +static struct kmem_cache *xattr_datum_cache; +static struct kmem_cache *xattr_ref_cache; #endif int __init jffs2_create_slab_caches(void) diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 77be534ce422..7deb78254021 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -28,7 +28,7 @@ static void jffs2_put_super(struct super_block *); -static kmem_cache_t *jffs2_inode_cachep; +static struct kmem_cache *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { @@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode) kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } -static void jffs2_i_init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void jffs2_i_init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct jffs2_inode_info *ei = (struct jffs2_inode_info *) foo; diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 0cccd1c39d75..b1a1c7296014 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -74,7 +74,7 @@ static inline void lock_metapage(struct metapage *mp) } #define METAPOOL_MIN_PAGES 32 -static kmem_cache_t *metapage_cache; +static struct kmem_cache *metapage_cache; static mempool_t *metapage_mempool; #define MPS_PER_PAGE (PAGE_CACHE_SIZE >> L2PSIZE) @@ -180,7 +180,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) #endif -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct metapage *mp = (struct metapage *)foo; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 9c1c6e0e633d..ca3e19128532 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -44,7 +44,7 @@ MODULE_DESCRIPTION("The Journaled Filesystem (JFS)"); MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM"); MODULE_LICENSE("GPL"); -static kmem_cache_t * jfs_inode_cachep; +static struct kmem_cache * jfs_inode_cachep; static struct super_operations jfs_super_operations; static struct export_operations jfs_export_operations; @@ -748,7 +748,7 @@ static struct file_system_type jfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo; diff --git a/fs/locks.c b/fs/locks.c index a7b97d50c1e0..1cb0c57fedbd 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -142,7 +142,7 @@ int lease_break_time = 45; static LIST_HEAD(file_lock_list); static LIST_HEAD(blocked_list); -static kmem_cache_t *filelock_cache __read_mostly; +static struct kmem_cache *filelock_cache __read_mostly; /* Allocate an empty lock structure. */ static struct file_lock *locks_alloc_lock(void) @@ -199,7 +199,7 @@ EXPORT_SYMBOL(locks_init_lock); * Initialises the fields of the file lock which are invariant for * free file_locks. */ -static void init_once(void *foo, kmem_cache_t *cache, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cache, unsigned long flags) { struct file_lock *lock = (struct file_lock *) foo; diff --git a/fs/mbcache.c b/fs/mbcache.c index 0ff71256e65b..deeb9dc062d9 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -85,7 +85,7 @@ struct mb_cache { #ifndef MB_CACHE_INDEXES_COUNT int c_indexes_count; #endif - kmem_cache_t *c_entry_cache; + struct kmem_cache *c_entry_cache; struct list_head *c_block_hash; struct list_head *c_indexes_hash[0]; }; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index ce532c2deda8..629e09b38c5c 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -51,7 +51,7 @@ static void minix_put_super(struct super_block *sb) return; } -static kmem_cache_t * minix_inode_cachep; +static struct kmem_cache * minix_inode_cachep; static struct inode *minix_alloc_inode(struct super_block *sb) { @@ -67,7 +67,7 @@ static void minix_destroy_inode(struct inode *inode) kmem_cache_free(minix_inode_cachep, minix_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; diff --git a/fs/namespace.c b/fs/namespace.c index 55442a6cf221..b00ac84ebbdd 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -36,7 +36,7 @@ static int event; static struct list_head *mount_hashtable __read_mostly; static int hash_mask __read_mostly, hash_bits __read_mostly; -static kmem_cache_t *mnt_cache __read_mostly; +static struct kmem_cache *mnt_cache __read_mostly; static struct rw_semaphore namespace_sem; /* /sys/fs */ diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index ed84d899220c..fae53243bb92 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -40,7 +40,7 @@ static void ncp_delete_inode(struct inode *); static void ncp_put_super(struct super_block *); static int ncp_statfs(struct dentry *, struct kstatfs *); -static kmem_cache_t * ncp_inode_cachep; +static struct kmem_cache * ncp_inode_cachep; static struct inode *ncp_alloc_inode(struct super_block *sb) { @@ -56,7 +56,7 @@ static void ncp_destroy_inode(struct inode *inode) kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 769fd0a0c772..2f488e1d9b6c 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -58,7 +58,7 @@ #define NFSDBG_FACILITY NFSDBG_VFS -static kmem_cache_t *nfs_direct_cachep; +static struct kmem_cache *nfs_direct_cachep; /* * This represents a set of asynchronous requests that we're waiting on diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 6b53aae4ed2c..15afa460e629 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -55,7 +55,7 @@ static int nfs_update_inode(struct inode *, struct nfs_fattr *); static void nfs_zap_acl_cache(struct inode *); -static kmem_cache_t * nfs_inode_cachep; +static struct kmem_cache * nfs_inode_cachep; static inline unsigned long nfs_fattr_to_ino_t(struct nfs_fattr *fattr) @@ -1111,7 +1111,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi) #endif } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index a1561a820abe..3fbfc2f03307 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -20,7 +20,7 @@ #define NFS_PARANOIA 1 -static kmem_cache_t *nfs_page_cachep; +static struct kmem_cache *nfs_page_cachep; static inline struct nfs_page * nfs_page_alloc(void) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 56f66f0ccb6a..244a8c45b68e 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -38,7 +38,7 @@ static int nfs_pagein_one(struct list_head *, struct inode *); static const struct rpc_call_ops nfs_read_partial_ops; static const struct rpc_call_ops nfs_read_full_ops; -static kmem_cache_t *nfs_rdata_cachep; +static struct kmem_cache *nfs_rdata_cachep; static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f7dd0d005957..41b07288f99e 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -85,7 +85,7 @@ static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; -static kmem_cache_t *nfs_wdata_cachep; +static struct kmem_cache *nfs_wdata_cachep; static mempool_t *nfs_wdata_mempool; static mempool_t *nfs_commit_mempool; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e431e93ab503..640c92b2a9f7 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -84,10 +84,10 @@ static void nfs4_set_recdir(char *recdir); */ static DEFINE_MUTEX(client_mutex); -static kmem_cache_t *stateowner_slab = NULL; -static kmem_cache_t *file_slab = NULL; -static kmem_cache_t *stateid_slab = NULL; -static kmem_cache_t *deleg_slab = NULL; +static struct kmem_cache *stateowner_slab = NULL; +static struct kmem_cache *file_slab = NULL; +static struct kmem_cache *stateid_slab = NULL; +static struct kmem_cache *deleg_slab = NULL; void nfs4_lock_state(void) @@ -1003,7 +1003,7 @@ alloc_init_file(struct inode *ino) } static void -nfsd4_free_slab(kmem_cache_t **slab) +nfsd4_free_slab(struct kmem_cache **slab) { if (*slab == NULL) return; diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 01f91501f70a..941acf14e61f 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -66,7 +66,7 @@ static struct file_operations dlmfs_file_operations; static struct inode_operations dlmfs_dir_inode_operations; static struct inode_operations dlmfs_root_inode_operations; static struct inode_operations dlmfs_file_inode_operations; -static kmem_cache_t *dlmfs_inode_cache; +static struct kmem_cache *dlmfs_inode_cache; struct workqueue_struct *user_dlm_worker; @@ -257,7 +257,7 @@ static ssize_t dlmfs_file_write(struct file *filp, } static void dlmfs_init_once(void *foo, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct dlmfs_inode_private *ip = diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index f784177b6241..856012b4fa49 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -221,7 +221,7 @@ EXPORT_SYMBOL_GPL(dlm_dump_all_mles); #endif /* 0 */ -static kmem_cache_t *dlm_mle_cache = NULL; +static struct kmem_cache *dlm_mle_cache = NULL; static void dlm_mle_release(struct kref *kref); diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index fcd4475d1f89..80ac69f11d9f 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -61,7 +61,7 @@ struct ocfs2_em_insert_context { struct ocfs2_extent_map_entry *right_ent; }; -static kmem_cache_t *ocfs2_em_ent_cachep = NULL; +static struct kmem_cache *ocfs2_em_ent_cachep = NULL; static struct ocfs2_extent_map_entry * diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 46a378fbc40b..1a7dd2945b34 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -106,7 +106,7 @@ static inline struct ocfs2_inode_info *OCFS2_I(struct inode *inode) #define INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags & OCFS2_INODE_JOURNAL) #define SET_INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags |= OCFS2_INODE_JOURNAL) -extern kmem_cache_t *ocfs2_inode_cache; +extern struct kmem_cache *ocfs2_inode_cache; extern const struct address_space_operations ocfs2_aops; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 7574d26ee0ff..0524f8a04ad6 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -68,7 +68,7 @@ #include "buffer_head_io.h" -static kmem_cache_t *ocfs2_inode_cachep = NULL; +static struct kmem_cache *ocfs2_inode_cachep = NULL; /* OCFS2 needs to schedule several differnt types of work which * require cluster locking, disk I/O, recovery waits, etc. Since these @@ -914,7 +914,7 @@ bail: } static void ocfs2_inode_init_once(void *data, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct ocfs2_inode_info *oi = data; diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c index 9707ed7a3206..39814b900fc0 100644 --- a/fs/ocfs2/uptodate.c +++ b/fs/ocfs2/uptodate.c @@ -69,7 +69,7 @@ struct ocfs2_meta_cache_item { sector_t c_block; }; -static kmem_cache_t *ocfs2_uptodate_cachep = NULL; +static struct kmem_cache *ocfs2_uptodate_cachep = NULL; void ocfs2_metadata_cache_init(struct inode *inode) { diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 911d1bcfc567..26f44e0074ec 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -330,7 +330,7 @@ out: return 0; } -static kmem_cache_t *op_inode_cachep; +static struct kmem_cache *op_inode_cachep; static struct inode *openprom_alloc_inode(struct super_block *sb) { @@ -415,7 +415,7 @@ static struct file_system_type openprom_fs_type = { .kill_sb = kill_anon_super, }; -static void op_inode_init_once(void *data, kmem_cache_t * cachep, unsigned long flags) +static void op_inode_init_once(void *data, struct kmem_cache * cachep, unsigned long flags) { struct op_inode_info *oi = (struct op_inode_info *) data; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index b24cdb2f17c1..e26945ba685b 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -81,7 +81,7 @@ static void proc_read_inode(struct inode * inode) inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; } -static kmem_cache_t * proc_inode_cachep; +static struct kmem_cache * proc_inode_cachep; static struct inode *proc_alloc_inode(struct super_block *sb) { @@ -105,7 +105,7 @@ static void proc_destroy_inode(struct inode *inode) kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5b943eb11d76..c047dc654d5c 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -515,7 +515,7 @@ static void qnx4_read_inode(struct inode *inode) brelse(bh); } -static kmem_cache_t *qnx4_inode_cachep; +static struct kmem_cache *qnx4_inode_cachep; static struct inode *qnx4_alloc_inode(struct super_block *sb) { @@ -531,7 +531,7 @@ static void qnx4_destroy_inode(struct inode *inode) kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 32332516d656..745bc714ba91 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -490,7 +490,7 @@ static void reiserfs_put_super(struct super_block *s) return; } -static kmem_cache_t *reiserfs_inode_cachep; +static struct kmem_cache *reiserfs_inode_cachep; static struct inode *reiserfs_alloc_inode(struct super_block *sb) { @@ -507,7 +507,7 @@ static void reiserfs_destroy_inode(struct inode *inode) kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index d1b455f9b669..c5af088d4a4c 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -550,7 +550,7 @@ romfs_read_inode(struct inode *i) } } -static kmem_cache_t * romfs_inode_cachep; +static struct kmem_cache * romfs_inode_cachep; static struct inode *romfs_alloc_inode(struct super_block *sb) { @@ -566,7 +566,7 @@ static void romfs_destroy_inode(struct inode *inode) kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct romfs_inode_info *ei = (struct romfs_inode_info *) foo; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 221617103484..4af4cd729a5a 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -50,7 +50,7 @@ static void smb_put_super(struct super_block *); static int smb_statfs(struct dentry *, struct kstatfs *); static int smb_show_options(struct seq_file *, struct vfsmount *); -static kmem_cache_t *smb_inode_cachep; +static struct kmem_cache *smb_inode_cachep; static struct inode *smb_alloc_inode(struct super_block *sb) { @@ -66,7 +66,7 @@ static void smb_destroy_inode(struct inode *inode) kmem_cache_free(smb_inode_cachep, SMB_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct smb_inode_info *ei = (struct smb_inode_info *) foo; unsigned long flagmask = SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 3eb1402191b9..a4bcae8a9aff 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -25,7 +25,7 @@ #define ROUND_UP(x) (((x)+3) & ~3) /* cache for request structures */ -static kmem_cache_t *req_cachep; +static struct kmem_cache *req_cachep; static int smb_request_send_req(struct smb_request *req); diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 20551a1b8a56..e503f858fba8 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -16,7 +16,7 @@ struct vfsmount *sysfs_mount; struct super_block * sysfs_sb = NULL; -kmem_cache_t *sysfs_dir_cachep; +struct kmem_cache *sysfs_dir_cachep; static struct super_operations sysfs_ops = { .statfs = simple_statfs, diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 6f3d6bd52887..bd7cec295dab 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -1,6 +1,6 @@ extern struct vfsmount * sysfs_mount; -extern kmem_cache_t *sysfs_dir_cachep; +extern struct kmem_cache *sysfs_dir_cachep; extern struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent *); extern int sysfs_create(struct dentry *, int mode, int (*init)(struct inode *)); diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index a6ca12b747cf..ead9864567e3 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -301,7 +301,7 @@ static void sysv_delete_inode(struct inode *inode) unlock_kernel(); } -static kmem_cache_t *sysv_inode_cachep; +static struct kmem_cache *sysv_inode_cachep; static struct inode *sysv_alloc_inode(struct super_block *sb) { @@ -318,7 +318,7 @@ static void sysv_destroy_inode(struct inode *inode) kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } -static void init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/udf/super.c b/fs/udf/super.c index e50f24221dea..397f54aaf667 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -107,7 +107,7 @@ static struct file_system_type udf_fstype = { .fs_flags = FS_REQUIRES_DEV, }; -static kmem_cache_t * udf_inode_cachep; +static struct kmem_cache * udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { @@ -130,7 +130,7 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct udf_inode_info *ei = (struct udf_inode_info *) foo; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 85a88c0c5e68..0b18d2cc2efa 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1204,7 +1204,7 @@ static int ufs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * ufs_inode_cachep; +static struct kmem_cache * ufs_inode_cachep; static struct inode *ufs_alloc_inode(struct super_block *sb) { @@ -1221,7 +1221,7 @@ static void ufs_destroy_inode(struct inode *inode) kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/include/acpi/platform/aclinux.h b/include/acpi/platform/aclinux.h index 47faf27913a5..7f1e92930b62 100644 --- a/include/acpi/platform/aclinux.h +++ b/include/acpi/platform/aclinux.h @@ -64,7 +64,7 @@ /* Host-dependent types and defines */ #define ACPI_MACHINE_WIDTH BITS_PER_LONG -#define acpi_cache_t kmem_cache_t +#define acpi_cache_t struct kmem_cache #define acpi_spinlock spinlock_t * #define ACPI_EXPORT_SYMBOL(symbol) EXPORT_SYMBOL(symbol); #define strtoul simple_strtoul diff --git a/include/asm-arm26/pgalloc.h b/include/asm-arm26/pgalloc.h index 6437167b1ffe..7725af3ddb4d 100644 --- a/include/asm-arm26/pgalloc.h +++ b/include/asm-arm26/pgalloc.h @@ -15,7 +15,7 @@ #include #include -extern kmem_cache_t *pte_cache; +extern struct kmem_cache *pte_cache; static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long addr){ return kmem_cache_alloc(pte_cache, GFP_KERNEL); diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 7d398f493dde..bfee7ddfff53 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -34,14 +34,14 @@ struct vm_area_struct; #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page)) extern unsigned long empty_zero_page[1024]; extern pgd_t swapper_pg_dir[1024]; -extern kmem_cache_t *pgd_cache; -extern kmem_cache_t *pmd_cache; +extern struct kmem_cache *pgd_cache; +extern struct kmem_cache *pmd_cache; extern spinlock_t pgd_lock; extern struct page *pgd_list; -void pmd_ctor(void *, kmem_cache_t *, unsigned long); -void pgd_ctor(void *, kmem_cache_t *, unsigned long); -void pgd_dtor(void *, kmem_cache_t *, unsigned long); +void pmd_ctor(void *, struct kmem_cache *, unsigned long); +void pgd_ctor(void *, struct kmem_cache *, unsigned long); +void pgd_dtor(void *, struct kmem_cache *, unsigned long); void pgtable_cache_init(void); void paging_init(void); diff --git a/include/asm-powerpc/pgalloc.h b/include/asm-powerpc/pgalloc.h index ae63db7b3e7d..b0830db68f8a 100644 --- a/include/asm-powerpc/pgalloc.h +++ b/include/asm-powerpc/pgalloc.h @@ -11,7 +11,7 @@ #include #include -extern kmem_cache_t *pgtable_cache[]; +extern struct kmem_cache *pgtable_cache[]; #ifdef CONFIG_PPC_64K_PAGES #define PTE_CACHE_NUM 0 diff --git a/include/asm-sparc64/pgalloc.h b/include/asm-sparc64/pgalloc.h index 010f9cd0a672..5891ff7ba760 100644 --- a/include/asm-sparc64/pgalloc.h +++ b/include/asm-sparc64/pgalloc.h @@ -13,7 +13,7 @@ #include /* Page table allocation/freeing. */ -extern kmem_cache_t *pgtable_cache; +extern struct kmem_cache *pgtable_cache; static inline pgd_t *pgd_alloc(struct mm_struct *mm) { diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h index 561e2a77805c..55d1ca5e60f5 100644 --- a/include/linux/delayacct.h +++ b/include/linux/delayacct.h @@ -30,7 +30,7 @@ #ifdef CONFIG_TASK_DELAY_ACCT extern int delayacct_on; /* Delay accounting turned on/off */ -extern kmem_cache_t *delayacct_cache; +extern struct kmem_cache *delayacct_cache; extern void delayacct_init(void); extern void __delayacct_tsk_init(struct task_struct *); extern void __delayacct_tsk_exit(struct task_struct *); diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 1fb02e17f6f6..2514f4e286b7 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -490,7 +490,7 @@ struct i2o_dma { */ struct i2o_pool { char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; mempool_t *mempool; }; diff --git a/include/linux/jbd.h b/include/linux/jbd.h index fe89444b1c6f..dacd566c9f6c 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -949,7 +949,7 @@ void journal_put_journal_head(struct journal_head *jh); /* * handle management */ -extern kmem_cache_t *jbd_handle_cache; +extern struct kmem_cache *jbd_handle_cache; static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) { diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index ddb128795781..98a0ae52660f 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -958,7 +958,7 @@ void jbd2_journal_put_journal_head(struct journal_head *jh); /* * handle management */ -extern kmem_cache_t *jbd2_handle_cache; +extern struct kmem_cache *jbd2_handle_cache; static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags) { diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h index f13299a15591..03636d7918fe 100644 --- a/include/linux/raid/raid5.h +++ b/include/linux/raid/raid5.h @@ -235,7 +235,7 @@ struct raid5_private_data { */ int active_name; char cache_name[2][20]; - kmem_cache_t *slab_cache; /* for allocating stripes */ + struct kmem_cache *slab_cache; /* for allocating stripes */ int seq_flush, seq_write; int quiesce; diff --git a/include/linux/rmap.h b/include/linux/rmap.h index 61c2ab634b00..36f850373d2c 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -30,7 +30,7 @@ struct anon_vma { #ifdef CONFIG_MMU -extern kmem_cache_t *anon_vma_cachep; +extern struct kmem_cache *anon_vma_cachep; static inline struct anon_vma *anon_vma_alloc(void) { diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 1d649f3eb006..4ff3940210d8 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -345,7 +345,7 @@ static inline struct sk_buff *alloc_skb_fclone(unsigned int size, return __alloc_skb(size, priority, 1, -1); } -extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, +extern struct sk_buff *alloc_skb_from_cache(struct kmem_cache *cp, unsigned int size, gfp_t priority); extern void kfree_skbmem(struct sk_buff *skb); diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index f81a5af8a4f8..ce8a912e5426 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -12,7 +12,7 @@ #include #ifdef CONFIG_TASKSTATS -extern kmem_cache_t *taskstats_cache; +extern struct kmem_cache *taskstats_cache; extern struct mutex taskstats_exit_mutex; static inline void taskstats_exit_free(struct taskstats *tidstats) diff --git a/include/net/dst.h b/include/net/dst.h index e156e38e4ac3..62b7e7598e9a 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -98,7 +98,7 @@ struct dst_ops int entry_size; atomic_t entries; - kmem_cache_t *kmem_cachep; + struct kmem_cache *kmem_cachep; }; #ifdef __KERNEL__ diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index a9eb2eaf094e..34cc76e3ddb4 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -125,7 +125,7 @@ struct inet_hashinfo { rwlock_t lhash_lock ____cacheline_aligned; atomic_t lhash_users; wait_queue_head_t lhash_wait; - kmem_cache_t *bind_bucket_cachep; + struct kmem_cache *bind_bucket_cachep; }; static inline struct inet_ehash_bucket *inet_ehash_bucket( @@ -136,10 +136,10 @@ static inline struct inet_ehash_bucket *inet_ehash_bucket( } extern struct inet_bind_bucket * - inet_bind_bucket_create(kmem_cache_t *cachep, + inet_bind_bucket_create(struct kmem_cache *cachep, struct inet_bind_hashbucket *head, const unsigned short snum); -extern void inet_bind_bucket_destroy(kmem_cache_t *cachep, +extern void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket *tb); static inline int inet_bhashfn(const __u16 lport, const int bhash_size) diff --git a/include/net/neighbour.h b/include/net/neighbour.h index c8aacbd2e333..23967031ddb7 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -160,7 +160,7 @@ struct neigh_table atomic_t entries; rwlock_t lock; unsigned long last_rand; - kmem_cache_t *kmem_cachep; + struct kmem_cache *kmem_cachep; struct neigh_statistics *stats; struct neighbour **hash_buckets; unsigned int hash_mask; diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index cef3136e22a3..41bcc9eb4206 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -7,7 +7,7 @@ #include extern struct list_head nf_conntrack_expect_list; -extern kmem_cache_t *nf_conntrack_expect_cachep; +extern struct kmem_cache *nf_conntrack_expect_cachep; extern struct file_operations exp_file_ops; struct nf_conntrack_expect diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 426f0fe774ef..7aed02ce2b65 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -29,7 +29,7 @@ struct proto; struct request_sock_ops { int family; int obj_size; - kmem_cache_t *slab; + struct kmem_cache *slab; int (*rtx_syn_ack)(struct sock *sk, struct request_sock *req, struct dst_entry *dst); diff --git a/include/net/sock.h b/include/net/sock.h index fe3a33fad03f..730899ce5162 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -571,7 +571,7 @@ struct proto { int *sysctl_rmem; int max_header; - kmem_cache_t *slab; + struct kmem_cache *slab; unsigned int obj_size; atomic_t *orphan_count; diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index d7a306ea560d..1e1ee3253fd8 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -15,7 +15,7 @@ #include struct timewait_sock_ops { - kmem_cache_t *twsk_slab; + struct kmem_cache *twsk_slab; unsigned int twsk_obj_size; int (*twsk_unique)(struct sock *sk, struct sock *sktw, void *twp); diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 9233ed5de664..0c775fceb675 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -557,7 +557,7 @@ struct sas_task { static inline struct sas_task *sas_alloc_task(gfp_t flags) { - extern kmem_cache_t *sas_task_cache; + extern struct kmem_cache *sas_task_cache; struct sas_task *task = kmem_cache_alloc(sas_task_cache, flags); if (task) { @@ -575,7 +575,7 @@ static inline struct sas_task *sas_alloc_task(gfp_t flags) static inline void sas_free_task(struct sas_task *task) { if (task) { - extern kmem_cache_t *sas_task_cache; + extern struct kmem_cache *sas_task_cache; BUG_ON(!list_empty(&task->list)); kmem_cache_free(sas_task_cache, task); } diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 813bb941342b..3acc1661e517 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -90,7 +90,7 @@ static struct super_operations mqueue_super_ops; static void remove_notification(struct mqueue_inode_info *info); static spinlock_t mq_lock; -static kmem_cache_t *mqueue_inode_cachep; +static struct kmem_cache *mqueue_inode_cachep; static struct vfsmount *mqueue_mnt; static unsigned int queues_count; @@ -211,7 +211,7 @@ static int mqueue_get_sb(struct file_system_type *fs_type, return get_sb_single(fs_type, flags, data, mqueue_fill_super, mnt); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo; diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 70e9ec603082..766d5912b26a 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -20,7 +20,7 @@ #include int delayacct_on __read_mostly = 1; /* Delay accounting turned on/off */ -kmem_cache_t *delayacct_cache; +struct kmem_cache *delayacct_cache; static int __init delayacct_setup_disable(char *str) { diff --git a/kernel/fork.c b/kernel/fork.c index 711aa5f10da7..2cf74edd3295 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -82,26 +82,26 @@ int nr_processes(void) #ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR # define alloc_task_struct() kmem_cache_alloc(task_struct_cachep, GFP_KERNEL) # define free_task_struct(tsk) kmem_cache_free(task_struct_cachep, (tsk)) -static kmem_cache_t *task_struct_cachep; +static struct kmem_cache *task_struct_cachep; #endif /* SLAB cache for signal_struct structures (tsk->signal) */ -static kmem_cache_t *signal_cachep; +static struct kmem_cache *signal_cachep; /* SLAB cache for sighand_struct structures (tsk->sighand) */ -kmem_cache_t *sighand_cachep; +struct kmem_cache *sighand_cachep; /* SLAB cache for files_struct structures (tsk->files) */ -kmem_cache_t *files_cachep; +struct kmem_cache *files_cachep; /* SLAB cache for fs_struct structures (tsk->fs) */ -kmem_cache_t *fs_cachep; +struct kmem_cache *fs_cachep; /* SLAB cache for vm_area_struct structures */ -kmem_cache_t *vm_area_cachep; +struct kmem_cache *vm_area_cachep; /* SLAB cache for mm_struct structures (tsk->mm) */ -static kmem_cache_t *mm_cachep; +static struct kmem_cache *mm_cachep; void free_task(struct task_struct *tsk) { @@ -1421,7 +1421,7 @@ long do_fork(unsigned long clone_flags, #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif -static void sighand_ctor(void *data, kmem_cache_t *cachep, unsigned long flags) +static void sighand_ctor(void *data, struct kmem_cache *cachep, unsigned long flags) { struct sighand_struct *sighand = data; diff --git a/kernel/pid.c b/kernel/pid.c index b914392085f9..a48879b0b921 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -31,7 +31,7 @@ #define pid_hashfn(nr) hash_long((unsigned long)nr, pidhash_shift) static struct hlist_head *pid_hash; static int pidhash_shift; -static kmem_cache_t *pid_cachep; +static struct kmem_cache *pid_cachep; int pid_max = PID_MAX_DEFAULT; diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 9cbb5d1be06f..5fe87de10ff0 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -70,7 +70,7 @@ /* * Lets keep our timers in a slab cache :-) */ -static kmem_cache_t *posix_timers_cache; +static struct kmem_cache *posix_timers_cache; static struct idr posix_timers_id; static DEFINE_SPINLOCK(idr_lock); diff --git a/kernel/signal.c b/kernel/signal.c index df18c167a2a7..8e19d2785486 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -33,7 +33,7 @@ * SLAB caches for signal bits. */ -static kmem_cache_t *sigqueue_cachep; +static struct kmem_cache *sigqueue_cachep; /* * In POSIX a signal is sent either to a specific thread (Linux task) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 1b2b326cf703..f5f92014ae98 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -34,7 +34,7 @@ static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 }; static int family_registered; -kmem_cache_t *taskstats_cache; +struct kmem_cache *taskstats_cache; static struct genl_family family = { .id = GENL_ID_GENERATE, diff --git a/kernel/user.c b/kernel/user.c index c1f93c164c93..4869563080e9 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -26,7 +26,7 @@ #define __uidhashfn(uid) (((uid >> UIDHASH_BITS) + uid) & UIDHASH_MASK) #define uidhashentry(uid) (uidhash_table + __uidhashfn((uid))) -static kmem_cache_t *uid_cachep; +static struct kmem_cache *uid_cachep; static struct list_head uidhash_table[UIDHASH_SZ]; /* diff --git a/lib/idr.c b/lib/idr.c index 16d2143fea48..71853531d3b0 100644 --- a/lib/idr.c +++ b/lib/idr.c @@ -33,7 +33,7 @@ #include #include -static kmem_cache_t *idr_layer_cache; +static struct kmem_cache *idr_layer_cache; static struct idr_layer *alloc_layer(struct idr *idp) { @@ -445,7 +445,7 @@ void *idr_replace(struct idr *idp, void *ptr, int id) } EXPORT_SYMBOL(idr_replace); -static void idr_cache_ctor(void * idr_layer, kmem_cache_t *idr_layer_cache, +static void idr_cache_ctor(void * idr_layer, struct kmem_cache *idr_layer_cache, unsigned long flags) { memset(idr_layer, 0, sizeof(struct idr_layer)); diff --git a/lib/radix-tree.c b/lib/radix-tree.c index aa9bfd0bdbd1..9eb25955019f 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -63,7 +63,7 @@ static unsigned long height_to_maxindex[RADIX_TREE_MAX_PATH] __read_mostly; /* * Radix tree node cache. */ -static kmem_cache_t *radix_tree_node_cachep; +static struct kmem_cache *radix_tree_node_cachep; /* * Per-cpu pool of preloaded nodes @@ -846,7 +846,7 @@ int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag) EXPORT_SYMBOL(radix_tree_tagged); static void -radix_tree_node_ctor(void *node, kmem_cache_t *cachep, unsigned long flags) +radix_tree_node_ctor(void *node, struct kmem_cache *cachep, unsigned long flags) { memset(node, 0, sizeof(struct radix_tree_node)); } diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index d9f04864d15d..8ca448db7a0d 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -23,7 +23,7 @@ #include #include "br_private.h" -static kmem_cache_t *br_fdb_cache __read_mostly; +static struct kmem_cache *br_fdb_cache __read_mostly; static int fdb_insert(struct net_bridge *br, struct net_bridge_port *source, const unsigned char *addr); diff --git a/net/core/flow.c b/net/core/flow.c index 5df3e297f817..104c25d00a1d 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -44,7 +44,7 @@ static DEFINE_PER_CPU(struct flow_cache_entry **, flow_tables) = { NULL }; #define flow_table(cpu) (per_cpu(flow_tables, cpu)) -static kmem_cache_t *flow_cachep __read_mostly; +static struct kmem_cache *flow_cachep __read_mostly; static int flow_lwm, flow_hwm; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 7217fb8928f2..de7801d589e7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -68,8 +68,8 @@ #include "kmap_skb.h" -static kmem_cache_t *skbuff_head_cache __read_mostly; -static kmem_cache_t *skbuff_fclone_cache __read_mostly; +static struct kmem_cache *skbuff_head_cache __read_mostly; +static struct kmem_cache *skbuff_fclone_cache __read_mostly; /* * Keep out-of-line to prevent kernel bloat. @@ -144,7 +144,7 @@ EXPORT_SYMBOL(skb_truesize_bug); struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, int fclone, int node) { - kmem_cache_t *cache; + struct kmem_cache *cache; struct skb_shared_info *shinfo; struct sk_buff *skb; u8 *data; @@ -211,7 +211,7 @@ nodata: * Buffers may only be allocated from interrupts using a @gfp_mask of * %GFP_ATOMIC. */ -struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, +struct sk_buff *alloc_skb_from_cache(struct kmem_cache *cp, unsigned int size, gfp_t gfp_mask) { diff --git a/net/core/sock.c b/net/core/sock.c index 419c7d3289c7..4a432da441e9 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -841,7 +841,7 @@ struct sock *sk_alloc(int family, gfp_t priority, struct proto *prot, int zero_it) { struct sock *sk = NULL; - kmem_cache_t *slab = prot->slab; + struct kmem_cache *slab = prot->slab; if (slab != NULL) sk = kmem_cache_alloc(slab, priority); diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index bdf1bb7a82c0..1f4727ddbdbf 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -21,8 +21,8 @@ #include -static kmem_cache_t *dccp_ackvec_slab; -static kmem_cache_t *dccp_ackvec_record_slab; +static struct kmem_cache *dccp_ackvec_slab; +static struct kmem_cache *dccp_ackvec_record_slab; static struct dccp_ackvec_record *dccp_ackvec_record_new(void) { diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index ff05e59043cd..d8cf92f09e68 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -55,9 +55,9 @@ static inline void ccids_read_unlock(void) #define ccids_read_unlock() do { } while(0) #endif -static kmem_cache_t *ccid_kmem_cache_create(int obj_size, const char *fmt,...) +static struct kmem_cache *ccid_kmem_cache_create(int obj_size, const char *fmt,...) { - kmem_cache_t *slab; + struct kmem_cache *slab; char slab_name_fmt[32], *slab_name; va_list args; @@ -75,7 +75,7 @@ static kmem_cache_t *ccid_kmem_cache_create(int obj_size, const char *fmt,...) return slab; } -static void ccid_kmem_cache_destroy(kmem_cache_t *slab) +static void ccid_kmem_cache_destroy(struct kmem_cache *slab) { if (slab != NULL) { const char *name = kmem_cache_name(slab); diff --git a/net/dccp/ccid.h b/net/dccp/ccid.h index c7c29514dce8..bcc2d12ae81c 100644 --- a/net/dccp/ccid.h +++ b/net/dccp/ccid.h @@ -27,9 +27,9 @@ struct ccid_operations { unsigned char ccid_id; const char *ccid_name; struct module *ccid_owner; - kmem_cache_t *ccid_hc_rx_slab; + struct kmem_cache *ccid_hc_rx_slab; __u32 ccid_hc_rx_obj_size; - kmem_cache_t *ccid_hc_tx_slab; + struct kmem_cache *ccid_hc_tx_slab; __u32 ccid_hc_tx_obj_size; int (*ccid_hc_rx_init)(struct ccid *ccid, struct sock *sk); int (*ccid_hc_tx_init)(struct ccid *ccid, struct sock *sk); diff --git a/net/dccp/ccids/lib/loss_interval.h b/net/dccp/ccids/lib/loss_interval.h index 0ae85f0340b2..eb257014dd74 100644 --- a/net/dccp/ccids/lib/loss_interval.h +++ b/net/dccp/ccids/lib/loss_interval.h @@ -20,7 +20,7 @@ #define DCCP_LI_HIST_IVAL_F_LENGTH 8 struct dccp_li_hist { - kmem_cache_t *dccplih_slab; + struct kmem_cache *dccplih_slab; }; extern struct dccp_li_hist *dccp_li_hist_new(const char *name); diff --git a/net/dccp/ccids/lib/packet_history.h b/net/dccp/ccids/lib/packet_history.h index 067cf1c85a37..9a8bcf224aa7 100644 --- a/net/dccp/ccids/lib/packet_history.h +++ b/net/dccp/ccids/lib/packet_history.h @@ -68,14 +68,14 @@ struct dccp_rx_hist_entry { }; struct dccp_tx_hist { - kmem_cache_t *dccptxh_slab; + struct kmem_cache *dccptxh_slab; }; extern struct dccp_tx_hist *dccp_tx_hist_new(const char *name); extern void dccp_tx_hist_delete(struct dccp_tx_hist *hist); struct dccp_rx_hist { - kmem_cache_t *dccprxh_slab; + struct kmem_cache *dccprxh_slab; }; extern struct dccp_rx_hist *dccp_rx_hist_new(const char *name); diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index 101e5ccaf096..13b2421991ba 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -79,7 +79,7 @@ for( ; ((f) = *(fp)) != NULL && dn_key_eq((f)->fn_key, (key)); (fp) = &(f)->fn_n static struct hlist_head dn_fib_table_hash[DN_FIB_TABLE_HASHSZ]; static DEFINE_RWLOCK(dn_fib_tables_lock); -static kmem_cache_t *dn_hash_kmem __read_mostly; +static struct kmem_cache *dn_hash_kmem __read_mostly; static int dn_fib_hash_zombies; static inline dn_fib_idx_t dn_hash(dn_fib_key_t key, struct dn_zone *dz) diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 4463443e42cd..648f47c1c399 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -45,8 +45,8 @@ #include "fib_lookup.h" -static kmem_cache_t *fn_hash_kmem __read_mostly; -static kmem_cache_t *fn_alias_kmem __read_mostly; +static struct kmem_cache *fn_hash_kmem __read_mostly; +static struct kmem_cache *fn_alias_kmem __read_mostly; struct fib_node { struct hlist_node fn_hash; diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 6be6caf1af37..cfb249cc0a58 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -172,7 +172,7 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn); static struct tnode *halve(struct trie *t, struct tnode *tn); static void tnode_free(struct tnode *tn); -static kmem_cache_t *fn_alias_kmem __read_mostly; +static struct kmem_cache *fn_alias_kmem __read_mostly; static struct trie *trie_local = NULL, *trie_main = NULL; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index bd6c9bc41893..8c79c8a4ea5c 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -27,7 +27,7 @@ * Allocate and initialize a new local port bind bucket. * The bindhash mutex for snum's hash chain must be held here. */ -struct inet_bind_bucket *inet_bind_bucket_create(kmem_cache_t *cachep, +struct inet_bind_bucket *inet_bind_bucket_create(struct kmem_cache *cachep, struct inet_bind_hashbucket *head, const unsigned short snum) { @@ -45,7 +45,7 @@ struct inet_bind_bucket *inet_bind_bucket_create(kmem_cache_t *cachep, /* * Caller must hold hashbucket lock for this tb with local BH disabled */ -void inet_bind_bucket_destroy(kmem_cache_t *cachep, struct inet_bind_bucket *tb) +void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket *tb) { if (hlist_empty(&tb->owners)) { __hlist_del(&tb->node); diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index f072f3875af8..711eb6d0285a 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -73,7 +73,7 @@ /* Exported for inet_getid inline function. */ DEFINE_SPINLOCK(inet_peer_idlock); -static kmem_cache_t *peer_cachep __read_mostly; +static struct kmem_cache *peer_cachep __read_mostly; #define node_height(x) x->avl_height static struct inet_peer peer_fake_node = { diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index efcf45ecc818..ecb5422ea237 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -105,7 +105,7 @@ static DEFINE_SPINLOCK(mfc_unres_lock); In this case data path is free of exclusive locks at all. */ -static kmem_cache_t *mrt_cachep __read_mostly; +static struct kmem_cache *mrt_cachep __read_mostly; static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local); static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); diff --git a/net/ipv4/ipvs/ip_vs_conn.c b/net/ipv4/ipvs/ip_vs_conn.c index 8832eb517d52..8086787a2c51 100644 --- a/net/ipv4/ipvs/ip_vs_conn.c +++ b/net/ipv4/ipvs/ip_vs_conn.c @@ -44,7 +44,7 @@ static struct list_head *ip_vs_conn_tab; /* SLAB cache for IPVS connections */ -static kmem_cache_t *ip_vs_conn_cachep __read_mostly; +static struct kmem_cache *ip_vs_conn_cachep __read_mostly; /* counter for current IPVS connections */ static atomic_t ip_vs_conn_count = ATOMIC_INIT(0); diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index f4b0e68a16d2..8556a4f4f60a 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -65,8 +65,8 @@ static LIST_HEAD(helpers); unsigned int ip_conntrack_htable_size __read_mostly = 0; int ip_conntrack_max __read_mostly; struct list_head *ip_conntrack_hash __read_mostly; -static kmem_cache_t *ip_conntrack_cachep __read_mostly; -static kmem_cache_t *ip_conntrack_expect_cachep __read_mostly; +static struct kmem_cache *ip_conntrack_cachep __read_mostly; +static struct kmem_cache *ip_conntrack_expect_cachep __read_mostly; struct ip_conntrack ip_conntrack_untracked; unsigned int ip_ct_log_invalid __read_mostly; static LIST_HEAD(unconfirmed); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 97a8cfbb61a1..96d8310ae9c8 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -50,7 +50,7 @@ struct rt6_statistics rt6_stats; -static kmem_cache_t * fib6_node_kmem __read_mostly; +static struct kmem_cache * fib6_node_kmem __read_mostly; enum fib_walk_state_t { diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index d4f68b0f27d5..12e426b9aacd 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -50,7 +50,7 @@ static u32 xfrm6_tunnel_spi; #define XFRM6_TUNNEL_SPI_MIN 1 #define XFRM6_TUNNEL_SPI_MAX 0xffffffff -static kmem_cache_t *xfrm6_tunnel_spi_kmem __read_mostly; +static struct kmem_cache *xfrm6_tunnel_spi_kmem __read_mostly; #define XFRM6_TUNNEL_SPI_BYADDR_HSIZE 256 #define XFRM6_TUNNEL_SPI_BYSPI_HSIZE 256 diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index eaa0f8a1adb6..a9638ff52a72 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -108,7 +108,7 @@ static struct { size_t size; /* slab cache pointer */ - kmem_cache_t *cachep; + struct kmem_cache *cachep; /* allocated slab cache + modules which uses this slab cache */ int use; @@ -147,7 +147,7 @@ int nf_conntrack_register_cache(u_int32_t features, const char *name, { int ret = 0; char *cache_name; - kmem_cache_t *cachep; + struct kmem_cache *cachep; DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n", features, name, size); @@ -226,7 +226,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_register_cache); /* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */ void nf_conntrack_unregister_cache(u_int32_t features) { - kmem_cache_t *cachep; + struct kmem_cache *cachep; char *name; /* diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 588d37937046..c20f901fa177 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -29,7 +29,7 @@ LIST_HEAD(nf_conntrack_expect_list); EXPORT_SYMBOL_GPL(nf_conntrack_expect_list); -kmem_cache_t *nf_conntrack_expect_cachep __read_mostly; +struct kmem_cache *nf_conntrack_expect_cachep __read_mostly; static unsigned int nf_conntrack_expect_next_id; /* nf_conntrack_expect helper functions */ diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index a98de0b54d65..a5a6e192ac2d 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -92,7 +92,7 @@ struct xt_hashlimit_htable { static DEFINE_SPINLOCK(hashlimit_lock); /* protects htables list */ static DEFINE_MUTEX(hlimit_mutex); /* additional checkentry protection */ static HLIST_HEAD(hashlimit_htables); -static kmem_cache_t *hashlimit_cachep __read_mostly; +static struct kmem_cache *hashlimit_cachep __read_mostly; static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) { diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 11f3b549f4a4..f2ba8615895b 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -79,8 +79,8 @@ static struct sctp_pf *sctp_pf_inet_specific; static struct sctp_af *sctp_af_v4_specific; static struct sctp_af *sctp_af_v6_specific; -kmem_cache_t *sctp_chunk_cachep __read_mostly; -kmem_cache_t *sctp_bucket_cachep __read_mostly; +struct kmem_cache *sctp_chunk_cachep __read_mostly; +struct kmem_cache *sctp_bucket_cachep __read_mostly; /* Return the address of the control sock. */ struct sock *sctp_get_ctl_sock(void) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 8d55d10041f9..30927d3a597f 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -65,7 +65,7 @@ #include #include -extern kmem_cache_t *sctp_chunk_cachep; +extern struct kmem_cache *sctp_chunk_cachep; SCTP_STATIC struct sctp_chunk *sctp_make_chunk(const struct sctp_association *asoc, diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 49607792cbd3..1e8132b8c4d9 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -107,7 +107,7 @@ static void sctp_sock_migrate(struct sock *, struct sock *, struct sctp_association *, sctp_socket_type_t); static char *sctp_hmac_alg = SCTP_COOKIE_HMAC_ALG; -extern kmem_cache_t *sctp_bucket_cachep; +extern struct kmem_cache *sctp_bucket_cachep; /* Get the sndbuf space available at the time on the association. */ static inline int sctp_wspace(struct sctp_association *asoc) diff --git a/net/socket.c b/net/socket.c index 4f417c2ddc15..43eff489c878 100644 --- a/net/socket.c +++ b/net/socket.c @@ -230,7 +230,7 @@ int move_addr_to_user(void *kaddr, int klen, void __user *uaddr, #define SOCKFS_MAGIC 0x534F434B -static kmem_cache_t *sock_inode_cachep __read_mostly; +static struct kmem_cache *sock_inode_cachep __read_mostly; static struct inode *sock_alloc_inode(struct super_block *sb) { @@ -257,7 +257,7 @@ static void sock_destroy_inode(struct inode *inode) container_of(inode, struct socket_alloc, vfs_inode)); } -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct socket_alloc *ei = (struct socket_alloc *)foo; diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index df753d0a884b..19703aa9659e 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -33,7 +33,7 @@ static int rpc_mount_count; static struct file_system_type rpc_pipe_fs_type; -static kmem_cache_t *rpc_inode_cachep __read_mostly; +static struct kmem_cache *rpc_inode_cachep __read_mostly; #define RPC_UPCALL_TIMEOUT (30*HZ) @@ -824,7 +824,7 @@ static struct file_system_type rpc_pipe_fs_type = { }; static void -init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct rpc_inode *rpci = (struct rpc_inode *) foo; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index eff44bcdc95a..225e6510b523 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -34,8 +34,8 @@ static int rpc_task_id; #define RPC_BUFFER_MAXSIZE (2048) #define RPC_BUFFER_POOLSIZE (8) #define RPC_TASK_POOLSIZE (8) -static kmem_cache_t *rpc_task_slabp __read_mostly; -static kmem_cache_t *rpc_buffer_slabp __read_mostly; +static struct kmem_cache *rpc_task_slabp __read_mostly; +static struct kmem_cache *rpc_buffer_slabp __read_mostly; static mempool_t *rpc_task_mempool __read_mostly; static mempool_t *rpc_buffer_mempool __read_mostly; diff --git a/net/tipc/handler.c b/net/tipc/handler.c index ae6ddf00a1aa..eb80778d6d9c 100644 --- a/net/tipc/handler.c +++ b/net/tipc/handler.c @@ -42,7 +42,7 @@ struct queue_item { unsigned long data; }; -static kmem_cache_t *tipc_queue_item_cache; +static struct kmem_cache *tipc_queue_item_cache; static struct list_head signal_queue_head; static DEFINE_SPINLOCK(qitem_lock); static int handler_enabled = 0; diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index a898a6a83a56..414f89070380 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -12,7 +12,7 @@ #include #include -static kmem_cache_t *secpath_cachep __read_mostly; +static struct kmem_cache *secpath_cachep __read_mostly; void __secpath_destroy(struct sec_path *sp) { diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index f6c77bd36fdd..3f3f563eb4ab 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -39,7 +39,7 @@ EXPORT_SYMBOL(xfrm_policy_count); static DEFINE_RWLOCK(xfrm_policy_afinfo_lock); static struct xfrm_policy_afinfo *xfrm_policy_afinfo[NPROTO]; -static kmem_cache_t *xfrm_dst_cache __read_mostly; +static struct kmem_cache *xfrm_dst_cache __read_mostly; static struct work_struct xfrm_policy_gc_work; static HLIST_HEAD(xfrm_policy_gc_list); diff --git a/security/keys/key.c b/security/keys/key.c index 157bac658bf9..0db816f10f85 100644 --- a/security/keys/key.c +++ b/security/keys/key.c @@ -20,7 +20,7 @@ #include #include "internal.h" -static kmem_cache_t *key_jar; +static struct kmem_cache *key_jar; struct rb_root key_serial_tree; /* tree of keys indexed by serial */ DEFINE_SPINLOCK(key_serial_lock); diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 65b4ec9c699a..e7c0b5e2066b 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -124,7 +124,7 @@ DEFINE_PER_CPU(struct avc_cache_stats, avc_cache_stats) = { 0 }; static struct avc_cache avc_cache; static struct avc_callback_node *avc_callbacks; -static kmem_cache_t *avc_node_cachep; +static struct kmem_cache *avc_node_cachep; static inline int avc_hash(u32 ssid, u32 tsid, u16 tclass) { diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index ac1aeed0b289..44e9cd470543 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -124,7 +124,7 @@ static struct security_operations *secondary_ops = NULL; static LIST_HEAD(superblock_security_head); static DEFINE_SPINLOCK(sb_security_lock); -static kmem_cache_t *sel_inode_cache; +static struct kmem_cache *sel_inode_cache; /* Return security context for a given sid or just the context length if the buffer is null or length is 0 */ diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c index 2dfc6134c2c2..ebb993c5c244 100644 --- a/security/selinux/ss/avtab.c +++ b/security/selinux/ss/avtab.c @@ -28,7 +28,7 @@ (keyp->source_type << 9)) & \ AVTAB_HASH_MASK) -static kmem_cache_t *avtab_node_cachep; +static struct kmem_cache *avtab_node_cachep; static struct avtab_node* avtab_insert_node(struct avtab *h, int hvalue, -- cgit v1.2.3 From 1b1cec4bbc59feac89670d5d6d222a02545bac94 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:33:22 -0800 Subject: [PATCH] slab: deprecate kmem_cache_t Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index fbcfc208f52b..2271886744f8 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -7,16 +7,17 @@ #ifndef _LINUX_SLAB_H #define _LINUX_SLAB_H -#if defined(__KERNEL__) +#ifdef __KERNEL__ -/* kmem_cache_t exists for legacy reasons and is not used by code in mm */ -typedef struct kmem_cache kmem_cache_t; +#include +#include +#include +#include /* kmalloc_sizes.h needs PAGE_SIZE */ +#include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ +#include -#include -#include -#include -#include /* kmalloc_sizes.h needs PAGE_SIZE */ -#include /* kmalloc_sizes.h needs L1_CACHE_BYTES */ +/* kmem_cache_t exists for legacy reasons and is not used by code in mm */ +typedef struct kmem_cache kmem_cache_t __deprecated; /* flags to pass to kmem_cache_create(). * The first 3 are only valid when the allocator as been build -- cgit v1.2.3 From 33f2ef89f8e181486b63fdbdc97c6afa6ca9f34b Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 6 Dec 2006 20:33:32 -0800 Subject: [PATCH] mm: make compound page destructor handling explicit Currently we we use the lru head link of the second page of a compound page to hold its destructor. This was ok when it was purely an internal implmentation detail. However, hugetlbfs overrides this destructor violating the layering. Abstract this out as explicit calls, also introduce a type for the callback function allowing them to be type checked. For each callback we pre-declare the function, causing a type error on definition rather than on use elsewhere. [akpm@osdl.org: cleanups] Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 18 ++++++++++++++++++ mm/hugetlb.c | 2 +- mm/page_alloc.c | 2 +- mm/swap.c | 4 ++-- 4 files changed, 22 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 0e266fe1b4c4..a17b147c61e7 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -295,6 +295,24 @@ void put_pages_list(struct list_head *pages); void split_page(struct page *page, unsigned int order); +/* + * Compound pages have a destructor function. Provide a + * prototype for that function and accessor functions. + * These are _only_ valid on the head of a PG_compound page. + */ +typedef void compound_page_dtor(struct page *); + +static inline void set_compound_page_dtor(struct page *page, + compound_page_dtor *dtor) +{ + page[1].lru.next = (void *)dtor; +} + +static inline compound_page_dtor *get_compound_page_dtor(struct page *page) +{ + return (compound_page_dtor *)page[1].lru.next; +} + /* * Multiple processes may "see" the same page. E.g. for untouched * mappings of /dev/null, all processes see the same page full of diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 2911a364481e..0ccc7f230252 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -109,7 +109,7 @@ static int alloc_fresh_huge_page(void) if (nid == MAX_NUMNODES) nid = first_node(node_online_map); if (page) { - page[1].lru.next = (void *)free_huge_page; /* dtor */ + set_compound_page_dtor(page, free_huge_page); spin_lock(&hugetlb_lock); nr_huge_pages++; nr_huge_pages_node[page_to_nid(page)]++; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index dc8753bdd47e..d539f83c62b6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -230,7 +230,7 @@ static void prep_compound_page(struct page *page, unsigned long order) int i; int nr_pages = 1 << order; - page[1].lru.next = (void *)free_compound_page; /* set dtor */ + set_compound_page_dtor(page, free_compound_page); page[1].lru.prev = (void *)order; for (i = 0; i < nr_pages; i++) { struct page *p = page + i; diff --git a/mm/swap.c b/mm/swap.c index d9a3770d8f3c..017e72ca9bbb 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -57,9 +57,9 @@ static void put_compound_page(struct page *page) { page = (struct page *)page_private(page); if (put_page_testzero(page)) { - void (*dtor)(struct page *page); + compound_page_dtor *dtor; - dtor = (void (*)(struct page *))page[1].lru.next; + dtor = get_compound_page_dtor(page); (*dtor)(page); } } -- cgit v1.2.3 From 36de6437866bbb1d37e2312ff4f95ee4ed6d2b61 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 6 Dec 2006 20:33:42 -0800 Subject: [PATCH] Save some bytes in struct mm_struct Before: [acme@newtoy net-2.6.20]$ pahole --cacheline 32 kernel/sched.o mm_struct /* include2/asm/processor.h:542 */ struct mm_struct { struct vm_area_struct * mmap; /* 0 4 */ struct rb_root mm_rb; /* 4 4 */ struct vm_area_struct * mmap_cache; /* 8 4 */ long unsigned int (*get_unmapped_area)(); /* 12 4 */ void (*unmap_area)(); /* 16 4 */ long unsigned int mmap_base; /* 20 4 */ long unsigned int task_size; /* 24 4 */ long unsigned int cached_hole_size; /* 28 4 */ /* ---------- cacheline 1 boundary ---------- */ long unsigned int free_area_cache; /* 32 4 */ pgd_t * pgd; /* 36 4 */ atomic_t mm_users; /* 40 4 */ atomic_t mm_count; /* 44 4 */ int map_count; /* 48 4 */ struct rw_semaphore mmap_sem; /* 52 64 */ spinlock_t page_table_lock; /* 116 40 */ struct list_head mmlist; /* 156 8 */ mm_counter_t _file_rss; /* 164 4 */ mm_counter_t _anon_rss; /* 168 4 */ long unsigned int hiwater_rss; /* 172 4 */ long unsigned int hiwater_vm; /* 176 4 */ long unsigned int total_vm; /* 180 4 */ long unsigned int locked_vm; /* 184 4 */ long unsigned int shared_vm; /* 188 4 */ /* ---------- cacheline 6 boundary ---------- */ long unsigned int exec_vm; /* 192 4 */ long unsigned int stack_vm; /* 196 4 */ long unsigned int reserved_vm; /* 200 4 */ long unsigned int def_flags; /* 204 4 */ long unsigned int nr_ptes; /* 208 4 */ long unsigned int start_code; /* 212 4 */ long unsigned int end_code; /* 216 4 */ long unsigned int start_data; /* 220 4 */ /* ---------- cacheline 7 boundary ---------- */ long unsigned int end_data; /* 224 4 */ long unsigned int start_brk; /* 228 4 */ long unsigned int brk; /* 232 4 */ long unsigned int start_stack; /* 236 4 */ long unsigned int arg_start; /* 240 4 */ long unsigned int arg_end; /* 244 4 */ long unsigned int env_start; /* 248 4 */ long unsigned int env_end; /* 252 4 */ /* ---------- cacheline 8 boundary ---------- */ long unsigned int saved_auxv[44]; /* 256 176 */ unsigned int dumpable:2; /* 432 4 */ cpumask_t cpu_vm_mask; /* 436 4 */ mm_context_t context; /* 440 68 */ long unsigned int swap_token_time; /* 508 4 */ /* ---------- cacheline 16 boundary ---------- */ char recent_pagein; /* 512 1 */ /* XXX 3 bytes hole, try to pack */ int core_waiters; /* 516 4 */ struct completion * core_startup_done; /* 520 4 */ struct completion core_done; /* 524 52 */ rwlock_t ioctx_list_lock; /* 576 36 */ struct kioctx * ioctx_list; /* 612 4 */ }; /* size: 616, sum members: 613, holes: 1, sum holes: 3, cachelines: 20, last cacheline: 8 bytes */ After: [acme@newtoy net-2.6.20]$ pahole --cacheline 32 kernel/sched.o mm_struct /* include2/asm/processor.h:542 */ struct mm_struct { struct vm_area_struct * mmap; /* 0 4 */ struct rb_root mm_rb; /* 4 4 */ struct vm_area_struct * mmap_cache; /* 8 4 */ long unsigned int (*get_unmapped_area)(); /* 12 4 */ void (*unmap_area)(); /* 16 4 */ long unsigned int mmap_base; /* 20 4 */ long unsigned int task_size; /* 24 4 */ long unsigned int cached_hole_size; /* 28 4 */ /* ---------- cacheline 1 boundary ---------- */ long unsigned int free_area_cache; /* 32 4 */ pgd_t * pgd; /* 36 4 */ atomic_t mm_users; /* 40 4 */ atomic_t mm_count; /* 44 4 */ int map_count; /* 48 4 */ struct rw_semaphore mmap_sem; /* 52 64 */ spinlock_t page_table_lock; /* 116 40 */ struct list_head mmlist; /* 156 8 */ mm_counter_t _file_rss; /* 164 4 */ mm_counter_t _anon_rss; /* 168 4 */ long unsigned int hiwater_rss; /* 172 4 */ long unsigned int hiwater_vm; /* 176 4 */ long unsigned int total_vm; /* 180 4 */ long unsigned int locked_vm; /* 184 4 */ long unsigned int shared_vm; /* 188 4 */ /* ---------- cacheline 6 boundary ---------- */ long unsigned int exec_vm; /* 192 4 */ long unsigned int stack_vm; /* 196 4 */ long unsigned int reserved_vm; /* 200 4 */ long unsigned int def_flags; /* 204 4 */ long unsigned int nr_ptes; /* 208 4 */ long unsigned int start_code; /* 212 4 */ long unsigned int end_code; /* 216 4 */ long unsigned int start_data; /* 220 4 */ /* ---------- cacheline 7 boundary ---------- */ long unsigned int end_data; /* 224 4 */ long unsigned int start_brk; /* 228 4 */ long unsigned int brk; /* 232 4 */ long unsigned int start_stack; /* 236 4 */ long unsigned int arg_start; /* 240 4 */ long unsigned int arg_end; /* 244 4 */ long unsigned int env_start; /* 248 4 */ long unsigned int env_end; /* 252 4 */ /* ---------- cacheline 8 boundary ---------- */ long unsigned int saved_auxv[44]; /* 256 176 */ cpumask_t cpu_vm_mask; /* 432 4 */ mm_context_t context; /* 436 68 */ long unsigned int swap_token_time; /* 504 4 */ char recent_pagein; /* 508 1 */ unsigned char dumpable:2; /* 509 1 */ /* XXX 2 bytes hole, try to pack */ int core_waiters; /* 512 4 */ struct completion * core_startup_done; /* 516 4 */ struct completion core_done; /* 520 52 */ rwlock_t ioctx_list_lock; /* 572 36 */ struct kioctx * ioctx_list; /* 608 4 */ }; /* size: 612, sum members: 610, holes: 1, sum holes: 2, cachelines: 20, last cacheline: 4 bytes */ [acme@newtoy net-2.6.20]$ codiff -V /tmp/sched.o.before kernel/sched.o /pub/scm/linux/kernel/git/acme/net-2.6.20/kernel/sched.c: struct mm_struct | -4 dumpable:2; from: unsigned int /* 432(30) 4(2) */ to: unsigned char /* 509(6) 1(2) */ < SNIP other offset changes > 1 struct changed [acme@newtoy net-2.6.20]$ I'm not aware of any problem about using 2 byte wide bitfields where previously a 4 byte wide one was, holler if there is any, I wouldn't be surprised, bitfields are things from hell. For the curious, 432(30) means: at offset 432 from the struct start, at offset 30 in the bitfield (yeah, it comes backwards, hellish, huh?) ditto for 509(6), while 4(2) and 1(2) means "struct field size(bitfield size)". Now we have a 2 bytes hole and are using only 4 bytes of the last 32 bytes cacheline, any takers? :-) Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index cad6a16260f7..acfd2e15c5f2 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -338,7 +338,6 @@ struct mm_struct { unsigned long saved_auxv[AT_VECTOR_SIZE]; /* for /proc/PID/auxv */ - unsigned dumpable:2; cpumask_t cpu_vm_mask; /* Architecture-specific MM context */ @@ -355,6 +354,8 @@ struct mm_struct { unsigned int token_priority; unsigned int last_interval; + unsigned char dumpable:2; + /* coredumping support */ int core_waiters; struct completion *core_startup_done, core_done; -- cgit v1.2.3 From 7cf9c2c76c1a17b32f2da85b50cd4fe468ed44b5 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 6 Dec 2006 20:33:44 -0800 Subject: [PATCH] radix-tree: RCU lockless readside Make radix tree lookups safe to be performed without locks. Readers are protected against nodes being deleted by using RCU based freeing. Readers are protected against new node insertion by using memory barriers to ensure the node itself will be properly written before it is visible in the radix tree. Each radix tree node keeps a record of their height (above leaf nodes). This height does not change after insertion -- when the radix tree is extended, higher nodes are only inserted in the top. So a lookup can take the pointer to what is *now* the root node, and traverse down it even if the tree is concurrently extended and this node becomes a subtree of a new root. "Direct" pointers (tree height of 0, where root->rnode points directly to the data item) are handled by using the low bit of the pointer to signal whether rnode is a direct pointer or a pointer to a radix tree node. When a reader wants to traverse the next branch, they will take a copy of the pointer. This pointer will be either NULL (and the branch is empty) or non-NULL (and will point to a valid node). [akpm@osdl.org: cleanups] [Lee.Schermerhorn@hp.com: bugfixes, comments, simplifications] [clameter@sgi.com: build fix] Signed-off-by: Nick Piggin Cc: "Paul E. McKenney" Signed-off-by: Lee Schermerhorn Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 101 ++++++++++++++ lib/radix-tree.c | 327 +++++++++++++++++++++++++++++++-------------- mm/migrate.c | 19 ++- 3 files changed, 340 insertions(+), 107 deletions(-) (limited to 'include/linux') diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index cbfa11537421..0deb842541ac 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -1,6 +1,7 @@ /* * Copyright (C) 2001 Momchil Velikov * Portions Copyright (C) 2001 Christoph Hellwig + * Copyright (C) 2006 Nick Piggin * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as @@ -21,6 +22,35 @@ #include #include +#include +#include + +/* + * A direct pointer (root->rnode pointing directly to a data item, + * rather than another radix_tree_node) is signalled by the low bit + * set in the root->rnode pointer. + * + * In this case root->height is also NULL, but the direct pointer tests are + * needed for RCU lookups when root->height is unreliable. + */ +#define RADIX_TREE_DIRECT_PTR 1 + +static inline void *radix_tree_ptr_to_direct(void *ptr) +{ + return (void *)((unsigned long)ptr | RADIX_TREE_DIRECT_PTR); +} + +static inline void *radix_tree_direct_to_ptr(void *ptr) +{ + return (void *)((unsigned long)ptr & ~RADIX_TREE_DIRECT_PTR); +} + +static inline int radix_tree_is_direct_ptr(void *ptr) +{ + return (int)((unsigned long)ptr & RADIX_TREE_DIRECT_PTR); +} + +/*** radix-tree API starts here ***/ #define RADIX_TREE_MAX_TAGS 2 @@ -47,6 +77,77 @@ do { \ (root)->rnode = NULL; \ } while (0) +/** + * Radix-tree synchronization + * + * The radix-tree API requires that users provide all synchronisation (with + * specific exceptions, noted below). + * + * Synchronization of access to the data items being stored in the tree, and + * management of their lifetimes must be completely managed by API users. + * + * For API usage, in general, + * - any function _modifying_ the the tree or tags (inserting or deleting + * items, setting or clearing tags must exclude other modifications, and + * exclude any functions reading the tree. + * - any function _reading_ the the tree or tags (looking up items or tags, + * gang lookups) must exclude modifications to the tree, but may occur + * concurrently with other readers. + * + * The notable exceptions to this rule are the following functions: + * radix_tree_lookup + * radix_tree_tag_get + * radix_tree_gang_lookup + * radix_tree_gang_lookup_tag + * radix_tree_tagged + * + * The first 4 functions are able to be called locklessly, using RCU. The + * caller must ensure calls to these functions are made within rcu_read_lock() + * regions. Other readers (lock-free or otherwise) and modifications may be + * running concurrently. + * + * It is still required that the caller manage the synchronization and lifetimes + * of the items. So if RCU lock-free lookups are used, typically this would mean + * that the items have their own locks, or are amenable to lock-free access; and + * that the items are freed by RCU (or only freed after having been deleted from + * the radix tree *and* a synchronize_rcu() grace period). + * + * (Note, rcu_assign_pointer and rcu_dereference are not needed to control + * access to data items when inserting into or looking up from the radix tree) + * + * radix_tree_tagged is able to be called without locking or RCU. + */ + +/** + * radix_tree_deref_slot - dereference a slot + * @pslot: pointer to slot, returned by radix_tree_lookup_slot + * Returns: item that was stored in that slot with any direct pointer flag + * removed. + * + * For use with radix_tree_lookup_slot(). Caller must hold tree at least read + * locked across slot lookup and dereference. More likely, will be used with + * radix_tree_replace_slot(), as well, so caller will hold tree write locked. + */ +static inline void *radix_tree_deref_slot(void **pslot) +{ + return radix_tree_direct_to_ptr(*pslot); +} +/** + * radix_tree_replace_slot - replace item in a slot + * @pslot: pointer to slot, returned by radix_tree_lookup_slot + * @item: new item to store in the slot. + * + * For use with radix_tree_lookup_slot(). Caller must hold tree write locked + * across slot lookup and replacement. + */ +static inline void radix_tree_replace_slot(void **pslot, void *item) +{ + BUG_ON(radix_tree_is_direct_ptr(item)); + rcu_assign_pointer(*pslot, + (void *)((unsigned long)item | + ((unsigned long)*pslot & RADIX_TREE_DIRECT_PTR))); +} + int radix_tree_insert(struct radix_tree_root *, unsigned long, void *); void *radix_tree_lookup(struct radix_tree_root *, unsigned long); void **radix_tree_lookup_slot(struct radix_tree_root *, unsigned long); diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 9eb25955019f..e2cefabb5aa0 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -2,6 +2,7 @@ * Copyright (C) 2001 Momchil Velikov * Portions Copyright (C) 2001 Christoph Hellwig * Copyright (C) 2005 SGI, Christoph Lameter + * Copyright (C) 2006 Nick Piggin * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as @@ -30,6 +31,7 @@ #include #include #include +#include #ifdef __KERNEL__ @@ -45,7 +47,9 @@ ((RADIX_TREE_MAP_SIZE + BITS_PER_LONG - 1) / BITS_PER_LONG) struct radix_tree_node { + unsigned int height; /* Height from the bottom */ unsigned int count; + struct rcu_head rcu_head; void *slots[RADIX_TREE_MAP_SIZE]; unsigned long tags[RADIX_TREE_MAX_TAGS][RADIX_TREE_TAG_LONGS]; }; @@ -100,13 +104,21 @@ radix_tree_node_alloc(struct radix_tree_root *root) rtp->nr--; } } + BUG_ON(radix_tree_is_direct_ptr(ret)); return ret; } +static void radix_tree_node_rcu_free(struct rcu_head *head) +{ + struct radix_tree_node *node = + container_of(head, struct radix_tree_node, rcu_head); + kmem_cache_free(radix_tree_node_cachep, node); +} + static inline void radix_tree_node_free(struct radix_tree_node *node) { - kmem_cache_free(radix_tree_node_cachep, node); + call_rcu(&node->rcu_head, radix_tree_node_rcu_free); } /* @@ -222,11 +234,12 @@ static int radix_tree_extend(struct radix_tree_root *root, unsigned long index) } do { + unsigned int newheight; if (!(node = radix_tree_node_alloc(root))) return -ENOMEM; /* Increase the height. */ - node->slots[0] = root->rnode; + node->slots[0] = radix_tree_direct_to_ptr(root->rnode); /* Propagate the aggregated tag info into the new root */ for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) { @@ -234,9 +247,11 @@ static int radix_tree_extend(struct radix_tree_root *root, unsigned long index) tag_set(node, tag, 0); } + newheight = root->height+1; + node->height = newheight; node->count = 1; - root->rnode = node; - root->height++; + rcu_assign_pointer(root->rnode, node); + root->height = newheight; } while (height > root->height); out: return 0; @@ -258,6 +273,8 @@ int radix_tree_insert(struct radix_tree_root *root, int offset; int error; + BUG_ON(radix_tree_is_direct_ptr(item)); + /* Make sure the tree is high enough. */ if (index > radix_tree_maxindex(root->height)) { error = radix_tree_extend(root, index); @@ -275,11 +292,12 @@ int radix_tree_insert(struct radix_tree_root *root, /* Have to add a child node. */ if (!(slot = radix_tree_node_alloc(root))) return -ENOMEM; + slot->height = height; if (node) { - node->slots[offset] = slot; + rcu_assign_pointer(node->slots[offset], slot); node->count++; } else - root->rnode = slot; + rcu_assign_pointer(root->rnode, slot); } /* Go a level down */ @@ -295,11 +313,11 @@ int radix_tree_insert(struct radix_tree_root *root, if (node) { node->count++; - node->slots[offset] = item; + rcu_assign_pointer(node->slots[offset], item); BUG_ON(tag_get(node, 0, offset)); BUG_ON(tag_get(node, 1, offset)); } else { - root->rnode = item; + rcu_assign_pointer(root->rnode, radix_tree_ptr_to_direct(item)); BUG_ON(root_tag_get(root, 0)); BUG_ON(root_tag_get(root, 1)); } @@ -308,49 +326,54 @@ int radix_tree_insert(struct radix_tree_root *root, } EXPORT_SYMBOL(radix_tree_insert); -static inline void **__lookup_slot(struct radix_tree_root *root, - unsigned long index) +/** + * radix_tree_lookup_slot - lookup a slot in a radix tree + * @root: radix tree root + * @index: index key + * + * Returns: the slot corresponding to the position @index in the + * radix tree @root. This is useful for update-if-exists operations. + * + * This function cannot be called under rcu_read_lock, it must be + * excluded from writers, as must the returned slot for subsequent + * use by radix_tree_deref_slot() and radix_tree_replace slot. + * Caller must hold tree write locked across slot lookup and + * replace. + */ +void **radix_tree_lookup_slot(struct radix_tree_root *root, unsigned long index) { unsigned int height, shift; - struct radix_tree_node **slot; - - height = root->height; + struct radix_tree_node *node, **slot; - if (index > radix_tree_maxindex(height)) + node = root->rnode; + if (node == NULL) return NULL; - if (height == 0 && root->rnode) + if (radix_tree_is_direct_ptr(node)) { + if (index > 0) + return NULL; return (void **)&root->rnode; + } + + height = node->height; + if (index > radix_tree_maxindex(height)) + return NULL; shift = (height-1) * RADIX_TREE_MAP_SHIFT; - slot = &root->rnode; - while (height > 0) { - if (*slot == NULL) + do { + slot = (struct radix_tree_node **) + (node->slots + ((index>>shift) & RADIX_TREE_MAP_MASK)); + node = *slot; + if (node == NULL) return NULL; - slot = (struct radix_tree_node **) - ((*slot)->slots + - ((index >> shift) & RADIX_TREE_MAP_MASK)); shift -= RADIX_TREE_MAP_SHIFT; height--; - } + } while (height > 0); return (void **)slot; } - -/** - * radix_tree_lookup_slot - lookup a slot in a radix tree - * @root: radix tree root - * @index: index key - * - * Lookup the slot corresponding to the position @index in the radix tree - * @root. This is useful for update-if-exists operations. - */ -void **radix_tree_lookup_slot(struct radix_tree_root *root, unsigned long index) -{ - return __lookup_slot(root, index); -} EXPORT_SYMBOL(radix_tree_lookup_slot); /** @@ -359,13 +382,45 @@ EXPORT_SYMBOL(radix_tree_lookup_slot); * @index: index key * * Lookup the item at the position @index in the radix tree @root. + * + * This function can be called under rcu_read_lock, however the caller + * must manage lifetimes of leaf nodes (eg. RCU may also be used to free + * them safely). No RCU barriers are required to access or modify the + * returned item, however. */ void *radix_tree_lookup(struct radix_tree_root *root, unsigned long index) { - void **slot; + unsigned int height, shift; + struct radix_tree_node *node, **slot; + + node = rcu_dereference(root->rnode); + if (node == NULL) + return NULL; + + if (radix_tree_is_direct_ptr(node)) { + if (index > 0) + return NULL; + return radix_tree_direct_to_ptr(node); + } + + height = node->height; + if (index > radix_tree_maxindex(height)) + return NULL; + + shift = (height-1) * RADIX_TREE_MAP_SHIFT; - slot = __lookup_slot(root, index); - return slot != NULL ? *slot : NULL; + do { + slot = (struct radix_tree_node **) + (node->slots + ((index>>shift) & RADIX_TREE_MAP_MASK)); + node = rcu_dereference(*slot); + if (node == NULL) + return NULL; + + shift -= RADIX_TREE_MAP_SHIFT; + height--; + } while (height > 0); + + return node; } EXPORT_SYMBOL(radix_tree_lookup); @@ -495,27 +550,30 @@ int radix_tree_tag_get(struct radix_tree_root *root, unsigned long index, unsigned int tag) { unsigned int height, shift; - struct radix_tree_node *slot; + struct radix_tree_node *node; int saw_unset_tag = 0; - height = root->height; - if (index > radix_tree_maxindex(height)) - return 0; - /* check the root's tag bit */ if (!root_tag_get(root, tag)) return 0; - if (height == 0) - return 1; + node = rcu_dereference(root->rnode); + if (node == NULL) + return 0; + + if (radix_tree_is_direct_ptr(node)) + return (index == 0); + + height = node->height; + if (index > radix_tree_maxindex(height)) + return 0; shift = (height - 1) * RADIX_TREE_MAP_SHIFT; - slot = root->rnode; for ( ; ; ) { int offset; - if (slot == NULL) + if (node == NULL) return 0; offset = (index >> shift) & RADIX_TREE_MAP_MASK; @@ -524,15 +582,15 @@ int radix_tree_tag_get(struct radix_tree_root *root, * This is just a debug check. Later, we can bale as soon as * we see an unset tag. */ - if (!tag_get(slot, tag, offset)) + if (!tag_get(node, tag, offset)) saw_unset_tag = 1; if (height == 1) { - int ret = tag_get(slot, tag, offset); + int ret = tag_get(node, tag, offset); BUG_ON(ret && saw_unset_tag); return !!ret; } - slot = slot->slots[offset]; + node = rcu_dereference(node->slots[offset]); shift -= RADIX_TREE_MAP_SHIFT; height--; } @@ -541,47 +599,45 @@ EXPORT_SYMBOL(radix_tree_tag_get); #endif static unsigned int -__lookup(struct radix_tree_root *root, void **results, unsigned long index, +__lookup(struct radix_tree_node *slot, void **results, unsigned long index, unsigned int max_items, unsigned long *next_index) { unsigned int nr_found = 0; unsigned int shift, height; - struct radix_tree_node *slot; unsigned long i; - height = root->height; - if (height == 0) { - if (root->rnode && index == 0) - results[nr_found++] = root->rnode; + height = slot->height; + if (height == 0) goto out; - } - shift = (height-1) * RADIX_TREE_MAP_SHIFT; - slot = root->rnode; for ( ; height > 1; height--) { - - for (i = (index >> shift) & RADIX_TREE_MAP_MASK ; - i < RADIX_TREE_MAP_SIZE; i++) { + i = (index >> shift) & RADIX_TREE_MAP_MASK; + for (;;) { if (slot->slots[i] != NULL) break; index &= ~((1UL << shift) - 1); index += 1UL << shift; if (index == 0) goto out; /* 32-bit wraparound */ + i++; + if (i == RADIX_TREE_MAP_SIZE) + goto out; } - if (i == RADIX_TREE_MAP_SIZE) - goto out; shift -= RADIX_TREE_MAP_SHIFT; - slot = slot->slots[i]; + slot = rcu_dereference(slot->slots[i]); + if (slot == NULL) + goto out; } /* Bottom level: grab some items */ for (i = index & RADIX_TREE_MAP_MASK; i < RADIX_TREE_MAP_SIZE; i++) { + struct radix_tree_node *node; index++; - if (slot->slots[i]) { - results[nr_found++] = slot->slots[i]; + node = slot->slots[i]; + if (node) { + results[nr_found++] = rcu_dereference(node); if (nr_found == max_items) goto out; } @@ -603,28 +659,51 @@ out: * *@results. * * The implementation is naive. + * + * Like radix_tree_lookup, radix_tree_gang_lookup may be called under + * rcu_read_lock. In this case, rather than the returned results being + * an atomic snapshot of the tree at a single point in time, the semantics + * of an RCU protected gang lookup are as though multiple radix_tree_lookups + * have been issued in individual locks, and results stored in 'results'. */ unsigned int radix_tree_gang_lookup(struct radix_tree_root *root, void **results, unsigned long first_index, unsigned int max_items) { - const unsigned long max_index = radix_tree_maxindex(root->height); + unsigned long max_index; + struct radix_tree_node *node; unsigned long cur_index = first_index; - unsigned int ret = 0; + unsigned int ret; + + node = rcu_dereference(root->rnode); + if (!node) + return 0; + if (radix_tree_is_direct_ptr(node)) { + if (first_index > 0) + return 0; + node = radix_tree_direct_to_ptr(node); + results[0] = rcu_dereference(node); + return 1; + } + + max_index = radix_tree_maxindex(node->height); + + ret = 0; while (ret < max_items) { unsigned int nr_found; unsigned long next_index; /* Index of next search */ if (cur_index > max_index) break; - nr_found = __lookup(root, results + ret, cur_index, + nr_found = __lookup(node, results + ret, cur_index, max_items - ret, &next_index); ret += nr_found; if (next_index == 0) break; cur_index = next_index; } + return ret; } EXPORT_SYMBOL(radix_tree_gang_lookup); @@ -634,55 +713,64 @@ EXPORT_SYMBOL(radix_tree_gang_lookup); * open-coding the search. */ static unsigned int -__lookup_tag(struct radix_tree_root *root, void **results, unsigned long index, +__lookup_tag(struct radix_tree_node *slot, void **results, unsigned long index, unsigned int max_items, unsigned long *next_index, unsigned int tag) { unsigned int nr_found = 0; - unsigned int shift; - unsigned int height = root->height; - struct radix_tree_node *slot; + unsigned int shift, height; - if (height == 0) { - if (root->rnode && index == 0) - results[nr_found++] = root->rnode; + height = slot->height; + if (height == 0) goto out; - } - - shift = (height - 1) * RADIX_TREE_MAP_SHIFT; - slot = root->rnode; + shift = (height-1) * RADIX_TREE_MAP_SHIFT; - do { - unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK; + while (height > 0) { + unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK ; - for ( ; i < RADIX_TREE_MAP_SIZE; i++) { - if (tag_get(slot, tag, i)) { - BUG_ON(slot->slots[i] == NULL); + for (;;) { + if (tag_get(slot, tag, i)) break; - } index &= ~((1UL << shift) - 1); index += 1UL << shift; if (index == 0) goto out; /* 32-bit wraparound */ + i++; + if (i == RADIX_TREE_MAP_SIZE) + goto out; } - if (i == RADIX_TREE_MAP_SIZE) - goto out; height--; if (height == 0) { /* Bottom level: grab some items */ unsigned long j = index & RADIX_TREE_MAP_MASK; for ( ; j < RADIX_TREE_MAP_SIZE; j++) { + struct radix_tree_node *node; index++; - if (tag_get(slot, tag, j)) { - BUG_ON(slot->slots[j] == NULL); - results[nr_found++] = slot->slots[j]; + if (!tag_get(slot, tag, j)) + continue; + node = slot->slots[j]; + /* + * Even though the tag was found set, we need to + * recheck that we have a non-NULL node, because + * if this lookup is lockless, it may have been + * subsequently deleted. + * + * Similar care must be taken in any place that + * lookup ->slots[x] without a lock (ie. can't + * rely on its value remaining the same). + */ + if (node) { + node = rcu_dereference(node); + results[nr_found++] = node; if (nr_found == max_items) goto out; } } } shift -= RADIX_TREE_MAP_SHIFT; - slot = slot->slots[i]; - } while (height > 0); + slot = rcu_dereference(slot->slots[i]); + if (slot == NULL) + break; + } out: *next_index = index; return nr_found; @@ -706,27 +794,44 @@ radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results, unsigned long first_index, unsigned int max_items, unsigned int tag) { - const unsigned long max_index = radix_tree_maxindex(root->height); + struct radix_tree_node *node; + unsigned long max_index; unsigned long cur_index = first_index; - unsigned int ret = 0; + unsigned int ret; /* check the root's tag bit */ if (!root_tag_get(root, tag)) return 0; + node = rcu_dereference(root->rnode); + if (!node) + return 0; + + if (radix_tree_is_direct_ptr(node)) { + if (first_index > 0) + return 0; + node = radix_tree_direct_to_ptr(node); + results[0] = rcu_dereference(node); + return 1; + } + + max_index = radix_tree_maxindex(node->height); + + ret = 0; while (ret < max_items) { unsigned int nr_found; unsigned long next_index; /* Index of next search */ if (cur_index > max_index) break; - nr_found = __lookup_tag(root, results + ret, cur_index, + nr_found = __lookup_tag(node, results + ret, cur_index, max_items - ret, &next_index, tag); ret += nr_found; if (next_index == 0) break; cur_index = next_index; } + return ret; } EXPORT_SYMBOL(radix_tree_gang_lookup_tag); @@ -742,8 +847,19 @@ static inline void radix_tree_shrink(struct radix_tree_root *root) root->rnode->count == 1 && root->rnode->slots[0]) { struct radix_tree_node *to_free = root->rnode; + void *newptr; - root->rnode = to_free->slots[0]; + /* + * We don't need rcu_assign_pointer(), since we are simply + * moving the node from one part of the tree to another. If + * it was safe to dereference the old pointer to it + * (to_free->slots[0]), it will be safe to dereference the new + * one (root->rnode). + */ + newptr = to_free->slots[0]; + if (root->height == 1) + newptr = radix_tree_ptr_to_direct(newptr); + root->rnode = newptr; root->height--; /* must only free zeroed nodes into the slab */ tag_clear(to_free, 0, 0); @@ -767,6 +883,7 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index) { struct radix_tree_path path[RADIX_TREE_MAX_PATH], *pathp = path; struct radix_tree_node *slot = NULL; + struct radix_tree_node *to_free; unsigned int height, shift; int tag; int offset; @@ -777,6 +894,7 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index) slot = root->rnode; if (height == 0 && root->rnode) { + slot = radix_tree_direct_to_ptr(slot); root_tag_clear_all(root); root->rnode = NULL; goto out; @@ -809,10 +927,17 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index) radix_tree_tag_clear(root, index, tag); } + to_free = NULL; /* Now free the nodes we do not need anymore */ while (pathp->node) { pathp->node->slots[pathp->offset] = NULL; pathp->node->count--; + /* + * Queue the node for deferred freeing after the + * last reference to it disappears (set NULL, above). + */ + if (to_free) + radix_tree_node_free(to_free); if (pathp->node->count) { if (pathp->node == root->rnode) @@ -821,13 +946,15 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index) } /* Node with zero slots in use so free it */ - radix_tree_node_free(pathp->node); - + to_free = pathp->node; pathp--; + } root_tag_clear_all(root); root->height = 0; root->rnode = NULL; + if (to_free) + radix_tree_node_free(to_free); out: return slot; diff --git a/mm/migrate.c b/mm/migrate.c index b4979d423d2b..e9b161bde95b 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -294,7 +294,7 @@ out: static int migrate_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page) { - struct page **radix_pointer; + void **pslot; if (!mapping) { /* Anonymous page */ @@ -305,12 +305,11 @@ static int migrate_page_move_mapping(struct address_space *mapping, write_lock_irq(&mapping->tree_lock); - radix_pointer = (struct page **)radix_tree_lookup_slot( - &mapping->page_tree, - page_index(page)); + pslot = radix_tree_lookup_slot(&mapping->page_tree, + page_index(page)); if (page_count(page) != 2 + !!PagePrivate(page) || - *radix_pointer != page) { + (struct page *)radix_tree_deref_slot(pslot) != page) { write_unlock_irq(&mapping->tree_lock); return -EAGAIN; } @@ -318,7 +317,7 @@ static int migrate_page_move_mapping(struct address_space *mapping, /* * Now we know that no one else is looking at the page. */ - get_page(newpage); + get_page(newpage); /* add cache reference */ #ifdef CONFIG_SWAP if (PageSwapCache(page)) { SetPageSwapCache(newpage); @@ -326,8 +325,14 @@ static int migrate_page_move_mapping(struct address_space *mapping, } #endif - *radix_pointer = newpage; + radix_tree_replace_slot(pslot, newpage); + + /* + * Drop cache reference from old page. + * We know this isn't the last reference. + */ __put_page(page); + write_unlock_irq(&mapping->tree_lock); return 0; -- cgit v1.2.3 From 915bae9ebe41e52d71ad8b06d50e4ab26189f964 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Wed, 6 Dec 2006 20:34:07 -0800 Subject: [PATCH] swsusp: use partition device and offset to identify swap areas The Linux kernel handles swap files almost in the same way as it handles swap partitions and there are only two differences between these two types of swap areas: (1) swap files need not be contiguous, (2) the header of a swap file is not in the first block of the partition that holds it. From the swsusp's point of view (1) is not a problem, because it is already taken care of by the swap-handling code, but (2) has to be taken into consideration. In principle the location of a swap file's header may be determined with the help of appropriate filesystem driver. Unfortunately, however, it requires the filesystem holding the swap file to be mounted, and if this filesystem is journaled, it cannot be mounted during a resume from disk. For this reason we need some other means by which swap areas can be identified. For example, to identify a swap area we can use the partition that holds the area and the offset from the beginning of this partition at which the swap header is located. The following patch allows swsusp to identify swap areas this way. It changes swap_type_of() so that it takes an additional argument representing an offset of the swap header within the partition represented by its first argument. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 2 +- kernel/power/swap.c | 2 +- kernel/power/user.c | 5 +++-- mm/swapfile.c | 38 ++++++++++++++++++++++++++------------ 4 files changed, 31 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index 89f8a39773bf..d51e35e4e168 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -247,7 +247,7 @@ extern int swap_duplicate(swp_entry_t); extern int valid_swaphandles(swp_entry_t, unsigned long *); extern void swap_free(swp_entry_t); extern void free_swap_and_cache(swp_entry_t); -extern int swap_type_of(dev_t); +extern int swap_type_of(dev_t, sector_t); extern unsigned int count_swap_pages(int, int); extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t); extern struct swap_info_struct *get_swap_info_struct(unsigned); diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 1a3b0dd2c3fc..7b10da16389e 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -74,7 +74,7 @@ static int mark_swapfiles(swp_entry_t start) static int swsusp_swap_check(void) /* This is called before saving image */ { - int res = swap_type_of(swsusp_resume_device); + int res = swap_type_of(swsusp_resume_device, 0); if (res >= 0) { root_swap = res; diff --git a/kernel/power/user.c b/kernel/power/user.c index 4c24ca5d62e3..a327b18a5ffd 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -55,7 +55,8 @@ static int snapshot_open(struct inode *inode, struct file *filp) filp->private_data = data; memset(&data->handle, 0, sizeof(struct snapshot_handle)); if ((filp->f_flags & O_ACCMODE) == O_RDONLY) { - data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device) : -1; + data->swap = swsusp_resume_device ? + swap_type_of(swsusp_resume_device, 0) : -1; data->mode = O_RDONLY; } else { data->swap = -1; @@ -265,7 +266,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, * so we need to recode them */ if (old_decode_dev(arg)) { - data->swap = swap_type_of(old_decode_dev(arg)); + data->swap = swap_type_of(old_decode_dev(arg), 0); if (data->swap < 0) error = -ENODEV; } else { diff --git a/mm/swapfile.c b/mm/swapfile.c index f315131db006..2bfacbac0f4c 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -427,34 +427,48 @@ void free_swap_and_cache(swp_entry_t entry) #ifdef CONFIG_SOFTWARE_SUSPEND /* - * Find the swap type that corresponds to given device (if any) + * Find the swap type that corresponds to given device (if any). * - * This is needed for software suspend and is done in such a way that inode - * aliasing is allowed. + * @offset - number of the PAGE_SIZE-sized block of the device, starting + * from 0, in which the swap header is expected to be located. + * + * This is needed for the suspend to disk (aka swsusp). */ -int swap_type_of(dev_t device) +int swap_type_of(dev_t device, sector_t offset) { + struct block_device *bdev = NULL; int i; + if (device) + bdev = bdget(device); + spin_lock(&swap_lock); for (i = 0; i < nr_swapfiles; i++) { - struct inode *inode; + struct swap_info_struct *sis = swap_info + i; - if (!(swap_info[i].flags & SWP_WRITEOK)) + if (!(sis->flags & SWP_WRITEOK)) continue; - if (!device) { + if (!bdev) { spin_unlock(&swap_lock); return i; } - inode = swap_info[i].swap_file->f_dentry->d_inode; - if (S_ISBLK(inode->i_mode) && - device == MKDEV(imajor(inode), iminor(inode))) { - spin_unlock(&swap_lock); - return i; + if (bdev == sis->bdev) { + struct swap_extent *se; + + se = list_entry(sis->extent_list.next, + struct swap_extent, list); + if (se->start_block == offset) { + spin_unlock(&swap_lock); + bdput(bdev); + return i; + } } } spin_unlock(&swap_lock); + if (bdev) + bdput(bdev); + return -ENODEV; } -- cgit v1.2.3 From 3aef83e0ef1ffb8ea3bea97be46821a45c952173 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Wed, 6 Dec 2006 20:34:10 -0800 Subject: [PATCH] swsusp: use block device offsets to identify swap locations Make swsusp use block device offsets instead of swap offsets to identify swap locations and make it use the same code paths for writing as well as for reading data. This allows us to use the same code for handling swap files and swap partitions and to simplify the code, eg. by dropping rw_swap_page_sync(). Signed-off-by: Rafael J. Wysocki Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 3 +- kernel/power/power.h | 2 +- kernel/power/swap.c | 133 ++++++++++++++++++++++++++------------------------ kernel/power/swsusp.c | 10 ++-- kernel/power/user.c | 7 +-- mm/page_io.c | 45 ----------------- mm/swapfile.c | 17 +++++++ 7 files changed, 98 insertions(+), 119 deletions(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index d51e35e4e168..add51cebc8d9 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -218,8 +218,6 @@ extern void swap_unplug_io_fn(struct backing_dev_info *, struct page *); /* linux/mm/page_io.c */ extern int swap_readpage(struct file *, struct page *); extern int swap_writepage(struct page *page, struct writeback_control *wbc); -extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page, - struct bio **bio_chain); extern int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err); /* linux/mm/swap_state.c */ @@ -250,6 +248,7 @@ extern void free_swap_and_cache(swp_entry_t); extern int swap_type_of(dev_t, sector_t); extern unsigned int count_swap_pages(int, int); extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t); +extern sector_t swapdev_block(int, pgoff_t); extern struct swap_info_struct *get_swap_info_struct(unsigned); extern int can_share_swap_page(struct page *); extern int remove_exclusive_swap_page(struct page *); diff --git a/kernel/power/power.h b/kernel/power/power.h index 87ecb1856ee8..210ebba26020 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -146,7 +146,7 @@ struct bitmap_page { extern void free_bitmap(struct bitmap_page *bitmap); extern struct bitmap_page *alloc_bitmap(unsigned int nr_bits); -extern unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap); +extern sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap); extern void free_all_swap_pages(int swap, struct bitmap_page *bitmap); extern int swsusp_check(void); diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 7768c2ba7550..1b08f46bbb7e 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -34,8 +34,8 @@ extern char resume_file[]; #define SWSUSP_SIG "S1SUSPEND" static struct swsusp_header { - char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)]; - swp_entry_t image; + char reserved[PAGE_SIZE - 20 - sizeof(sector_t)]; + sector_t image; char orig_sig[10]; char sig[10]; } __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header; @@ -100,9 +100,9 @@ static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain) return submit(READ, page_off, virt_to_page(addr), bio_chain); } -static int bio_write_page(pgoff_t page_off, void *addr) +static int bio_write_page(pgoff_t page_off, void *addr, struct bio **bio_chain) { - return submit(WRITE, page_off, virt_to_page(addr), NULL); + return submit(WRITE, page_off, virt_to_page(addr), bio_chain); } static int wait_on_bio_chain(struct bio **bio_chain) @@ -157,22 +157,19 @@ static void show_speed(struct timeval *start, struct timeval *stop, * Saving part */ -static int mark_swapfiles(swp_entry_t start) +static int mark_swapfiles(sector_t start) { int error; - rw_swap_page_sync(READ, swp_entry(root_swap, 0), - virt_to_page((unsigned long)&swsusp_header), NULL); + bio_read_page(0, &swsusp_header, NULL); if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) || !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) { memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10); memcpy(swsusp_header.sig,SWSUSP_SIG, 10); swsusp_header.image = start; - error = rw_swap_page_sync(WRITE, swp_entry(root_swap, 0), - virt_to_page((unsigned long)&swsusp_header), - NULL); + error = bio_write_page(0, &swsusp_header, NULL); } else { - pr_debug("swsusp: Partition is not swap space.\n"); + printk(KERN_ERR "swsusp: Swap header not found!\n"); error = -ENODEV; } return error; @@ -185,12 +182,21 @@ static int mark_swapfiles(swp_entry_t start) static int swsusp_swap_check(void) /* This is called before saving image */ { - int res = swap_type_of(swsusp_resume_device, 0); + int res; + + res = swap_type_of(swsusp_resume_device, 0); + if (res < 0) + return res; + + root_swap = res; + resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_WRITE); + if (IS_ERR(resume_bdev)) + return PTR_ERR(resume_bdev); + + res = set_blocksize(resume_bdev, PAGE_SIZE); + if (res < 0) + blkdev_put(resume_bdev); - if (res >= 0) { - root_swap = res; - return 0; - } return res; } @@ -201,36 +207,26 @@ static int swsusp_swap_check(void) /* This is called before saving image */ * @bio_chain: Link the next write BIO here */ -static int write_page(void *buf, unsigned long offset, struct bio **bio_chain) +static int write_page(void *buf, sector_t offset, struct bio **bio_chain) { - swp_entry_t entry; - int error = -ENOSPC; - - if (offset) { - struct page *page = virt_to_page(buf); - - if (bio_chain) { - /* - * Whether or not we successfully allocated a copy page, - * we take a ref on the page here. It gets undone in - * wait_on_bio_chain(). - */ - struct page *page_copy; - page_copy = alloc_page(GFP_ATOMIC); - if (page_copy == NULL) { - WARN_ON_ONCE(1); - bio_chain = NULL; /* Go synchronous */ - get_page(page); - } else { - memcpy(page_address(page_copy), - page_address(page), PAGE_SIZE); - page = page_copy; - } + void *src; + + if (!offset) + return -ENOSPC; + + if (bio_chain) { + src = (void *)__get_free_page(GFP_ATOMIC); + if (src) { + memcpy(src, buf, PAGE_SIZE); + } else { + WARN_ON_ONCE(1); + bio_chain = NULL; /* Go synchronous */ + src = buf; } - entry = swp_entry(root_swap, offset); - error = rw_swap_page_sync(WRITE, entry, page, bio_chain); + } else { + src = buf; } - return error; + return bio_write_page(offset, src, bio_chain); } /* @@ -248,11 +244,11 @@ static int write_page(void *buf, unsigned long offset, struct bio **bio_chain) * at a time. */ -#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(long) - 1) +#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(sector_t) - 1) struct swap_map_page { - unsigned long entries[MAP_PAGE_ENTRIES]; - unsigned long next_swap; + sector_t entries[MAP_PAGE_ENTRIES]; + sector_t next_swap; }; /** @@ -262,7 +258,7 @@ struct swap_map_page { struct swap_map_handle { struct swap_map_page *cur; - unsigned long cur_swap; + sector_t cur_swap; struct bitmap_page *bitmap; unsigned int k; }; @@ -287,7 +283,7 @@ static int get_swap_writer(struct swap_map_handle *handle) release_swap_writer(handle); return -ENOMEM; } - handle->cur_swap = alloc_swap_page(root_swap, handle->bitmap); + handle->cur_swap = alloc_swapdev_block(root_swap, handle->bitmap); if (!handle->cur_swap) { release_swap_writer(handle); return -ENOSPC; @@ -300,11 +296,11 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, struct bio **bio_chain) { int error = 0; - unsigned long offset; + sector_t offset; if (!handle->cur) return -EINVAL; - offset = alloc_swap_page(root_swap, handle->bitmap); + offset = alloc_swapdev_block(root_swap, handle->bitmap); error = write_page(buf, offset, bio_chain); if (error) return error; @@ -313,7 +309,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, error = wait_on_bio_chain(bio_chain); if (error) goto out; - offset = alloc_swap_page(root_swap, handle->bitmap); + offset = alloc_swapdev_block(root_swap, handle->bitmap); if (!offset) return -ENOSPC; handle->cur->next_swap = offset; @@ -413,37 +409,47 @@ int swsusp_write(void) struct swsusp_info *header; int error; - if ((error = swsusp_swap_check())) { + error = swsusp_swap_check(); + if (error) { printk(KERN_ERR "swsusp: Cannot find swap device, try " "swapon -a.\n"); return error; } memset(&snapshot, 0, sizeof(struct snapshot_handle)); error = snapshot_read_next(&snapshot, PAGE_SIZE); - if (error < PAGE_SIZE) - return error < 0 ? error : -EFAULT; + if (error < PAGE_SIZE) { + if (error >= 0) + error = -EFAULT; + + goto out; + } header = (struct swsusp_info *)data_of(snapshot); if (!enough_swap(header->pages)) { printk(KERN_ERR "swsusp: Not enough free swap\n"); - return -ENOSPC; + error = -ENOSPC; + goto out; } error = get_swap_writer(&handle); if (!error) { - unsigned long start = handle.cur_swap; + sector_t start = handle.cur_swap; + error = swap_write_page(&handle, header, NULL); if (!error) error = save_image(&handle, &snapshot, header->pages - 1); + if (!error) { flush_swap_writer(&handle); printk("S"); - error = mark_swapfiles(swp_entry(root_swap, start)); + error = mark_swapfiles(start); printk("|\n"); } } if (error) free_all_swap_pages(root_swap, handle.bitmap); release_swap_writer(&handle); +out: + swsusp_close(); return error; } @@ -459,17 +465,18 @@ static void release_swap_reader(struct swap_map_handle *handle) handle->cur = NULL; } -static int get_swap_reader(struct swap_map_handle *handle, - swp_entry_t start) +static int get_swap_reader(struct swap_map_handle *handle, sector_t start) { int error; - if (!swp_offset(start)) + if (!start) return -EINVAL; + handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC); if (!handle->cur) return -ENOMEM; - error = bio_read_page(swp_offset(start), handle->cur, NULL); + + error = bio_read_page(start, handle->cur, NULL); if (error) { release_swap_reader(handle); return error; @@ -481,7 +488,7 @@ static int get_swap_reader(struct swap_map_handle *handle, static int swap_read_page(struct swap_map_handle *handle, void *buf, struct bio **bio_chain) { - unsigned long offset; + sector_t offset; int error; if (!handle->cur) @@ -608,7 +615,7 @@ int swsusp_check(void) if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) { memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10); /* Reset swap signature now */ - error = bio_write_page(0, &swsusp_header); + error = bio_write_page(0, &swsusp_header, NULL); } else { return -EINVAL; } diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 0b66659dc516..4147a756a8c7 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -134,18 +134,18 @@ static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit) return 0; } -unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap) +sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap) { unsigned long offset; offset = swp_offset(get_swap_page_of_type(swap)); if (offset) { - if (bitmap_set(bitmap, offset)) { + if (bitmap_set(bitmap, offset)) swap_free(swp_entry(swap, offset)); - offset = 0; - } + else + return swapdev_block(swap, offset); } - return offset; + return 0; } void free_all_swap_pages(int swap, struct bitmap_page *bitmap) diff --git a/kernel/power/user.c b/kernel/power/user.c index a327b18a5ffd..f0b7ef8bdd74 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -126,7 +126,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, { int error = 0; struct snapshot_data *data; - loff_t offset, avail; + loff_t avail; + sector_t offset; if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC) return -ENOTTY; @@ -240,10 +241,10 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; } } - offset = alloc_swap_page(data->swap, data->bitmap); + offset = alloc_swapdev_block(data->swap, data->bitmap); if (offset) { offset <<= PAGE_SHIFT; - error = put_user(offset, (loff_t __user *)arg); + error = put_user(offset, (sector_t __user *)arg); } else { error = -ENOSPC; } diff --git a/mm/page_io.c b/mm/page_io.c index d4840ecbf8f9..dbffec0d78c9 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -147,48 +147,3 @@ int swap_readpage(struct file *file, struct page *page) out: return ret; } - -#ifdef CONFIG_SOFTWARE_SUSPEND -/* - * A scruffy utility function to read or write an arbitrary swap page - * and wait on the I/O. The caller must have a ref on the page. - * - * We use end_swap_bio_read() even for writes, because it happens to do what - * we want. - */ -int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page, - struct bio **bio_chain) -{ - struct bio *bio; - int ret = 0; - int bio_rw; - - lock_page(page); - - bio = get_swap_bio(GFP_KERNEL, entry.val, page, end_swap_bio_read); - if (bio == NULL) { - unlock_page(page); - ret = -ENOMEM; - goto out; - } - - bio_rw = rw; - if (!bio_chain) - bio_rw |= (1 << BIO_RW_SYNC); - if (bio_chain) - bio_get(bio); - submit_bio(bio_rw, bio); - if (bio_chain == NULL) { - wait_on_page_locked(page); - - if (!PageUptodate(page) || PageError(page)) - ret = -EIO; - } - if (bio_chain) { - bio->bi_private = *bio_chain; - *bio_chain = bio; - } -out: - return ret; -} -#endif diff --git a/mm/swapfile.c b/mm/swapfile.c index 2bfacbac0f4c..55242363de64 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -945,6 +945,23 @@ sector_t map_swap_page(struct swap_info_struct *sis, pgoff_t offset) } } +#ifdef CONFIG_SOFTWARE_SUSPEND +/* + * Get the (PAGE_SIZE) block corresponding to given offset on the swapdev + * corresponding to given index in swap_info (swap type). + */ +sector_t swapdev_block(int swap_type, pgoff_t offset) +{ + struct swap_info_struct *sis; + + if (swap_type >= nr_swapfiles) + return 0; + + sis = swap_info + swap_type; + return (sis->flags & SWP_WRITEOK) ? map_swap_page(sis, offset) : 0; +} +#endif /* CONFIG_SOFTWARE_SUSPEND */ + /* * Free all of a swapdev's extent information */ -- cgit v1.2.3 From 8357376d3df21b7d6f857931a57ac50da9c66e26 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Wed, 6 Dec 2006 20:34:18 -0800 Subject: [PATCH] swsusp: Improve handling of highmem Currently swsusp saves the contents of highmem pages by copying them to the normal zone which is quite inefficient (eg. it requires two normal pages to be used for saving one highmem page). This may be improved by using highmem for saving the contents of saveable highmem pages. Namely, during the suspend phase of the suspend-resume cycle we try to allocate as many free highmem pages as there are saveable highmem pages. If there are not enough highmem image pages to store the contents of all of the saveable highmem pages, some of them will be stored in the "normal" memory. Next, we allocate as many free "normal" pages as needed to store the (remaining) image data. We use a memory bitmap to mark the allocated free pages (ie. highmem as well as "normal" image pages). Now, we use another memory bitmap to mark all of the saveable pages (highmem as well as "normal") and the contents of the saveable pages are copied into the image pages. Then, the second bitmap is used to save the pfns corresponding to the saveable pages and the first one is used to save their data. During the resume phase the pfns of the pages that were saveable during the suspend are loaded from the image and used to mark the "unsafe" page frames. Next, we try to allocate as many free highmem page frames as to load all of the image data that had been in the highmem before the suspend and we allocate so many free "normal" page frames that the total number of allocated free pages (highmem and "normal") is equal to the size of the image. While doing this we have to make sure that there will be some extra free "normal" and "safe" page frames for two lists of PBEs constructed later. Now, the image data are loaded, if possible, into their "original" page frames. The image data that cannot be written into their "original" page frames are loaded into "safe" page frames and their "original" kernel virtual addresses, as well as the addresses of the "safe" pages containing their copies, are stored in one of two lists of PBEs. One list of PBEs is for the copies of "normal" suspend pages (ie. "normal" pages that were saveable during the suspend) and it is used in the same way as previously (ie. by the architecture-dependent parts of swsusp). The other list of PBEs is for the copies of highmem suspend pages. The pages in this list are restored (in a reversible way) right before the arch-dependent code is called. Signed-off-by: Rafael J. Wysocki Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/suspend.h | 9 +- kernel/power/power.h | 2 +- kernel/power/snapshot.c | 851 ++++++++++++++++++++++++++++++++++++------------ kernel/power/swap.c | 2 +- kernel/power/swsusp.c | 53 +-- kernel/power/user.c | 2 +- mm/vmscan.c | 3 + 7 files changed, 680 insertions(+), 242 deletions(-) (limited to 'include/linux') diff --git a/include/linux/suspend.h b/include/linux/suspend.h index b1237f16ecde..bf99bd49f8ef 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -9,10 +9,13 @@ #include #include -/* page backup entry */ +/* struct pbe is used for creating lists of pages that should be restored + * atomically during the resume from disk, because the page frames they have + * occupied before the suspend are in use. + */ struct pbe { - unsigned long address; /* address of the copy */ - unsigned long orig_address; /* original address of page */ + void *address; /* address of the copy */ + void *orig_address; /* original address of a page */ struct pbe *next; }; diff --git a/kernel/power/power.h b/kernel/power/power.h index 7dbfd9f67e1c..3763343bde2f 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -103,8 +103,8 @@ struct snapshot_handle { extern unsigned int snapshot_additional_pages(struct zone *zone); extern int snapshot_read_next(struct snapshot_handle *handle, size_t count); extern int snapshot_write_next(struct snapshot_handle *handle, size_t count); +extern void snapshot_write_finalize(struct snapshot_handle *handle); extern int snapshot_image_loaded(struct snapshot_handle *handle); -extern void snapshot_free_unused_memory(struct snapshot_handle *handle); /* * This structure is used to pass the values needed for the identification diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 99f9b7d177d6..fd8251d40eb8 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1,15 +1,15 @@ /* * linux/kernel/power/snapshot.c * - * This file provide system snapshot/restore functionality. + * This file provides system snapshot/restore functionality for swsusp. * * Copyright (C) 1998-2005 Pavel Machek + * Copyright (C) 2006 Rafael J. Wysocki * - * This file is released under the GPLv2, and is based on swsusp.c. + * This file is released under the GPLv2. * */ - #include #include #include @@ -34,137 +34,24 @@ #include "power.h" -/* List of PBEs used for creating and restoring the suspend image */ +/* List of PBEs needed for restoring the pages that were allocated before + * the suspend and included in the suspend image, but have also been + * allocated by the "resume" kernel, so their contents cannot be written + * directly to their "original" page frames. + */ struct pbe *restore_pblist; -static unsigned int nr_copy_pages; -static unsigned int nr_meta_pages; +/* Pointer to an auxiliary buffer (1 page) */ static void *buffer; -#ifdef CONFIG_HIGHMEM -unsigned int count_highmem_pages(void) -{ - struct zone *zone; - unsigned long zone_pfn; - unsigned int n = 0; - - for_each_zone (zone) - if (is_highmem(zone)) { - mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; zone_pfn++) { - struct page *page; - unsigned long pfn = zone_pfn + zone->zone_start_pfn; - if (!pfn_valid(pfn)) - continue; - page = pfn_to_page(pfn); - if (PageReserved(page)) - continue; - if (PageNosaveFree(page)) - continue; - n++; - } - } - return n; -} - -struct highmem_page { - char *data; - struct page *page; - struct highmem_page *next; -}; - -static struct highmem_page *highmem_copy; - -static int save_highmem_zone(struct zone *zone) -{ - unsigned long zone_pfn; - mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) { - struct page *page; - struct highmem_page *save; - void *kaddr; - unsigned long pfn = zone_pfn + zone->zone_start_pfn; - - if (!(pfn%10000)) - printk("."); - if (!pfn_valid(pfn)) - continue; - page = pfn_to_page(pfn); - /* - * This condition results from rvmalloc() sans vmalloc_32() - * and architectural memory reservations. This should be - * corrected eventually when the cases giving rise to this - * are better understood. - */ - if (PageReserved(page)) - continue; - BUG_ON(PageNosave(page)); - if (PageNosaveFree(page)) - continue; - save = kmalloc(sizeof(struct highmem_page), GFP_ATOMIC); - if (!save) - return -ENOMEM; - save->next = highmem_copy; - save->page = page; - save->data = (void *) get_zeroed_page(GFP_ATOMIC); - if (!save->data) { - kfree(save); - return -ENOMEM; - } - kaddr = kmap_atomic(page, KM_USER0); - memcpy(save->data, kaddr, PAGE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - highmem_copy = save; - } - return 0; -} - -int save_highmem(void) -{ - struct zone *zone; - int res = 0; - - pr_debug("swsusp: Saving Highmem"); - drain_local_pages(); - for_each_zone (zone) { - if (is_highmem(zone)) - res = save_highmem_zone(zone); - if (res) - return res; - } - printk("\n"); - return 0; -} - -int restore_highmem(void) -{ - printk("swsusp: Restoring Highmem\n"); - while (highmem_copy) { - struct highmem_page *save = highmem_copy; - void *kaddr; - highmem_copy = save->next; - - kaddr = kmap_atomic(save->page, KM_USER0); - memcpy(kaddr, save->data, PAGE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - free_page((long) save->data); - kfree(save); - } - return 0; -} -#else -static inline unsigned int count_highmem_pages(void) {return 0;} -static inline int save_highmem(void) {return 0;} -static inline int restore_highmem(void) {return 0;} -#endif - /** * @safe_needed - on resume, for storing the PBE list and the image, * we can only use memory pages that do not conflict with the pages - * used before suspend. + * used before suspend. The unsafe pages have PageNosaveFree set + * and we count them using unsafe_pages. * - * The unsafe pages are marked with the PG_nosave_free flag - * and we count them using unsafe_pages + * Each allocated image page is marked as PageNosave and PageNosaveFree + * so that swsusp_free() can release it. */ #define PG_ANY 0 @@ -174,7 +61,7 @@ static inline int restore_highmem(void) {return 0;} static unsigned int allocated_unsafe_pages; -static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) +static void *get_image_page(gfp_t gfp_mask, int safe_needed) { void *res; @@ -195,20 +82,38 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) unsigned long get_safe_page(gfp_t gfp_mask) { - return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE); + return (unsigned long)get_image_page(gfp_mask, PG_SAFE); +} + +static struct page *alloc_image_page(gfp_t gfp_mask) { + struct page *page; + + page = alloc_page(gfp_mask); + if (page) { + SetPageNosave(page); + SetPageNosaveFree(page); + } + return page; } /** * free_image_page - free page represented by @addr, allocated with - * alloc_image_page (page flags set by it must be cleared) + * get_image_page (page flags set by it must be cleared) */ static inline void free_image_page(void *addr, int clear_nosave_free) { - ClearPageNosave(virt_to_page(addr)); + struct page *page; + + BUG_ON(!virt_addr_valid(addr)); + + page = virt_to_page(addr); + + ClearPageNosave(page); if (clear_nosave_free) - ClearPageNosaveFree(virt_to_page(addr)); - free_page((unsigned long)addr); + ClearPageNosaveFree(page); + + __free_page(page); } /* struct linked_page is used to build chains of pages */ @@ -269,7 +174,7 @@ static void *chain_alloc(struct chain_allocator *ca, unsigned int size) if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) { struct linked_page *lp; - lp = alloc_image_page(ca->gfp_mask, ca->safe_needed); + lp = get_image_page(ca->gfp_mask, ca->safe_needed); if (!lp) return NULL; @@ -446,8 +351,8 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) /* Compute the number of zones */ nr = 0; - for_each_zone (zone) - if (populated_zone(zone) && !is_highmem(zone)) + for_each_zone(zone) + if (populated_zone(zone)) nr++; /* Allocate the list of zones bitmap objects */ @@ -459,10 +364,10 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) } /* Initialize the zone bitmap objects */ - for_each_zone (zone) { + for_each_zone(zone) { unsigned long pfn; - if (!populated_zone(zone) || is_highmem(zone)) + if (!populated_zone(zone)) continue; zone_bm->start_pfn = zone->zone_start_pfn; @@ -481,7 +386,7 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) while (bb) { unsigned long *ptr; - ptr = alloc_image_page(gfp_mask, safe_needed); + ptr = get_image_page(gfp_mask, safe_needed); bb->data = ptr; if (!ptr) goto Free; @@ -669,9 +574,81 @@ unsigned int snapshot_additional_pages(struct zone *zone) res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK); res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE); - return res; + return 2 * res; } +#ifdef CONFIG_HIGHMEM +/** + * count_free_highmem_pages - compute the total number of free highmem + * pages, system-wide. + */ + +static unsigned int count_free_highmem_pages(void) +{ + struct zone *zone; + unsigned int cnt = 0; + + for_each_zone(zone) + if (populated_zone(zone) && is_highmem(zone)) + cnt += zone->free_pages; + + return cnt; +} + +/** + * saveable_highmem_page - Determine whether a highmem page should be + * included in the suspend image. + * + * We should save the page if it isn't Nosave or NosaveFree, or Reserved, + * and it isn't a part of a free chunk of pages. + */ + +static struct page *saveable_highmem_page(unsigned long pfn) +{ + struct page *page; + + if (!pfn_valid(pfn)) + return NULL; + + page = pfn_to_page(pfn); + + BUG_ON(!PageHighMem(page)); + + if (PageNosave(page) || PageReserved(page) || PageNosaveFree(page)) + return NULL; + + return page; +} + +/** + * count_highmem_pages - compute the total number of saveable highmem + * pages. + */ + +unsigned int count_highmem_pages(void) +{ + struct zone *zone; + unsigned int n = 0; + + for_each_zone(zone) { + unsigned long pfn, max_zone_pfn; + + if (!is_highmem(zone)) + continue; + + mark_free_pages(zone); + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + if (saveable_highmem_page(pfn)) + n++; + } + return n; +} +#else +static inline void *saveable_highmem_page(unsigned long pfn) { return NULL; } +static inline unsigned int count_highmem_pages(void) { return 0; } +#endif /* CONFIG_HIGHMEM */ + /** * pfn_is_nosave - check if given pfn is in the 'nosave' section */ @@ -684,12 +661,12 @@ static inline int pfn_is_nosave(unsigned long pfn) } /** - * saveable - Determine whether a page should be cloned or not. - * @pfn: The page + * saveable - Determine whether a non-highmem page should be included in + * the suspend image. * - * We save a page if it isn't Nosave, and is not in the range of pages - * statically defined as 'unsaveable', and it - * isn't a part of a free chunk of pages. + * We should save the page if it isn't Nosave, and is not in the range + * of pages statically defined as 'unsaveable', and it isn't a part of + * a free chunk of pages. */ static struct page *saveable_page(unsigned long pfn) @@ -701,76 +678,130 @@ static struct page *saveable_page(unsigned long pfn) page = pfn_to_page(pfn); - if (PageNosave(page)) + BUG_ON(PageHighMem(page)); + + if (PageNosave(page) || PageNosaveFree(page)) return NULL; + if (PageReserved(page) && pfn_is_nosave(pfn)) return NULL; - if (PageNosaveFree(page)) - return NULL; return page; } +/** + * count_data_pages - compute the total number of saveable non-highmem + * pages. + */ + unsigned int count_data_pages(void) { struct zone *zone; unsigned long pfn, max_zone_pfn; unsigned int n = 0; - for_each_zone (zone) { + for_each_zone(zone) { if (is_highmem(zone)) continue; + mark_free_pages(zone); max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) - n += !!saveable_page(pfn); + if(saveable_page(pfn)) + n++; } return n; } -static inline void copy_data_page(long *dst, long *src) +/* This is needed, because copy_page and memcpy are not usable for copying + * task structs. + */ +static inline void do_copy_page(long *dst, long *src) { int n; - /* copy_page and memcpy are not usable for copying task structs. */ for (n = PAGE_SIZE / sizeof(long); n; n--) *dst++ = *src++; } +#ifdef CONFIG_HIGHMEM +static inline struct page * +page_is_saveable(struct zone *zone, unsigned long pfn) +{ + return is_highmem(zone) ? + saveable_highmem_page(pfn) : saveable_page(pfn); +} + +static inline void +copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +{ + struct page *s_page, *d_page; + void *src, *dst; + + s_page = pfn_to_page(src_pfn); + d_page = pfn_to_page(dst_pfn); + if (PageHighMem(s_page)) { + src = kmap_atomic(s_page, KM_USER0); + dst = kmap_atomic(d_page, KM_USER1); + do_copy_page(dst, src); + kunmap_atomic(src, KM_USER0); + kunmap_atomic(dst, KM_USER1); + } else { + src = page_address(s_page); + if (PageHighMem(d_page)) { + /* Page pointed to by src may contain some kernel + * data modified by kmap_atomic() + */ + do_copy_page(buffer, src); + dst = kmap_atomic(pfn_to_page(dst_pfn), KM_USER0); + memcpy(dst, buffer, PAGE_SIZE); + kunmap_atomic(dst, KM_USER0); + } else { + dst = page_address(d_page); + do_copy_page(dst, src); + } + } +} +#else +#define page_is_saveable(zone, pfn) saveable_page(pfn) + +static inline void +copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +{ + do_copy_page(page_address(pfn_to_page(dst_pfn)), + page_address(pfn_to_page(src_pfn))); +} +#endif /* CONFIG_HIGHMEM */ + static void copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) { struct zone *zone; unsigned long pfn; - for_each_zone (zone) { + for_each_zone(zone) { unsigned long max_zone_pfn; - if (is_highmem(zone)) - continue; - mark_free_pages(zone); max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) - if (saveable_page(pfn)) + if (page_is_saveable(zone, pfn)) memory_bm_set_bit(orig_bm, pfn); } memory_bm_position_reset(orig_bm); memory_bm_position_reset(copy_bm); do { pfn = memory_bm_next_pfn(orig_bm); - if (likely(pfn != BM_END_OF_MAP)) { - struct page *page; - void *src; - - page = pfn_to_page(pfn); - src = page_address(page); - page = pfn_to_page(memory_bm_next_pfn(copy_bm)); - copy_data_page(page_address(page), src); - } + if (likely(pfn != BM_END_OF_MAP)) + copy_data_page(memory_bm_next_pfn(copy_bm), pfn); } while (pfn != BM_END_OF_MAP); } +/* Total number of image pages */ +static unsigned int nr_copy_pages; +/* Number of pages needed for saving the original pfns of the image pages */ +static unsigned int nr_meta_pages; + /** * swsusp_free - free pages allocated for the suspend. * @@ -792,7 +823,7 @@ void swsusp_free(void) if (PageNosave(page) && PageNosaveFree(page)) { ClearPageNosave(page); ClearPageNosaveFree(page); - free_page((long) page_address(page)); + __free_page(page); } } } @@ -802,34 +833,108 @@ void swsusp_free(void) buffer = NULL; } +#ifdef CONFIG_HIGHMEM +/** + * count_pages_for_highmem - compute the number of non-highmem pages + * that will be necessary for creating copies of highmem pages. + */ + +static unsigned int count_pages_for_highmem(unsigned int nr_highmem) +{ + unsigned int free_highmem = count_free_highmem_pages(); + + if (free_highmem >= nr_highmem) + nr_highmem = 0; + else + nr_highmem -= free_highmem; + + return nr_highmem; +} +#else +static unsigned int +count_pages_for_highmem(unsigned int nr_highmem) { return 0; } +#endif /* CONFIG_HIGHMEM */ /** - * enough_free_mem - Make sure we enough free memory to snapshot. - * - * Returns TRUE or FALSE after checking the number of available - * free pages. + * enough_free_mem - Make sure we have enough free memory for the + * snapshot image. */ -static int enough_free_mem(unsigned int nr_pages) +static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem) { struct zone *zone; unsigned int free = 0, meta = 0; - for_each_zone (zone) - if (!is_highmem(zone)) { + for_each_zone(zone) { + meta += snapshot_additional_pages(zone); + if (!is_highmem(zone)) free += zone->free_pages; - meta += snapshot_additional_pages(zone); - } + } - pr_debug("swsusp: pages needed: %u + %u + %u, available pages: %u\n", + nr_pages += count_pages_for_highmem(nr_highmem); + pr_debug("swsusp: Normal pages needed: %u + %u + %u, available pages: %u\n", nr_pages, PAGES_FOR_IO, meta, free); return free > nr_pages + PAGES_FOR_IO + meta; } +#ifdef CONFIG_HIGHMEM +/** + * get_highmem_buffer - if there are some highmem pages in the suspend + * image, we may need the buffer to copy them and/or load their data. + */ + +static inline int get_highmem_buffer(int safe_needed) +{ + buffer = get_image_page(GFP_ATOMIC | __GFP_COLD, safe_needed); + return buffer ? 0 : -ENOMEM; +} + +/** + * alloc_highmem_image_pages - allocate some highmem pages for the image. + * Try to allocate as many pages as needed, but if the number of free + * highmem pages is lesser than that, allocate them all. + */ + +static inline unsigned int +alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem) +{ + unsigned int to_alloc = count_free_highmem_pages(); + + if (to_alloc > nr_highmem) + to_alloc = nr_highmem; + + nr_highmem -= to_alloc; + while (to_alloc-- > 0) { + struct page *page; + + page = alloc_image_page(__GFP_HIGHMEM); + memory_bm_set_bit(bm, page_to_pfn(page)); + } + return nr_highmem; +} +#else +static inline int get_highmem_buffer(int safe_needed) { return 0; } + +static inline unsigned int +alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int n) { return 0; } +#endif /* CONFIG_HIGHMEM */ + +/** + * swsusp_alloc - allocate memory for the suspend image + * + * We first try to allocate as many highmem pages as there are + * saveable highmem pages in the system. If that fails, we allocate + * non-highmem pages for the copies of the remaining highmem ones. + * + * In this approach it is likely that the copies of highmem pages will + * also be located in the high memory, because of the way in which + * copy_data_pages() works. + */ + static int swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, - unsigned int nr_pages) + unsigned int nr_pages, unsigned int nr_highmem) { int error; @@ -841,13 +946,19 @@ swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, if (error) goto Free; + if (nr_highmem > 0) { + error = get_highmem_buffer(PG_ANY); + if (error) + goto Free; + + nr_pages += alloc_highmem_image_pages(copy_bm, nr_highmem); + } while (nr_pages-- > 0) { - struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD); + struct page *page = alloc_image_page(GFP_ATOMIC | __GFP_COLD); + if (!page) goto Free; - SetPageNosave(page); - SetPageNosaveFree(page); memory_bm_set_bit(copy_bm, page_to_pfn(page)); } return 0; @@ -857,30 +968,39 @@ Free: return -ENOMEM; } -/* Memory bitmap used for marking saveable pages */ +/* Memory bitmap used for marking saveable pages (during suspend) or the + * suspend image pages (during resume) + */ static struct memory_bitmap orig_bm; -/* Memory bitmap used for marking allocated pages that will contain the copies - * of saveable pages +/* Memory bitmap used on suspend for marking allocated pages that will contain + * the copies of saveable pages. During resume it is initially used for + * marking the suspend image pages, but then its set bits are duplicated in + * @orig_bm and it is released. Next, on systems with high memory, it may be + * used for marking "safe" highmem pages, but it has to be reinitialized for + * this purpose. */ static struct memory_bitmap copy_bm; asmlinkage int swsusp_save(void) { - unsigned int nr_pages; + unsigned int nr_pages, nr_highmem; - pr_debug("swsusp: critical section: \n"); + printk("swsusp: critical section: \n"); drain_local_pages(); nr_pages = count_data_pages(); - printk("swsusp: Need to copy %u pages\n", nr_pages); + nr_highmem = count_highmem_pages(); + printk("swsusp: Need to copy %u pages\n", nr_pages + nr_highmem); - if (!enough_free_mem(nr_pages)) { + if (!enough_free_mem(nr_pages, nr_highmem)) { printk(KERN_ERR "swsusp: Not enough free memory\n"); return -ENOMEM; } - if (swsusp_alloc(&orig_bm, ©_bm, nr_pages)) + if (swsusp_alloc(&orig_bm, ©_bm, nr_pages, nr_highmem)) { + printk(KERN_ERR "swsusp: Memory allocation failed\n"); return -ENOMEM; + } /* During allocating of suspend pagedir, new cold pages may appear. * Kill them. @@ -894,10 +1014,12 @@ asmlinkage int swsusp_save(void) * touch swap space! Except we must write out our image of course. */ + nr_pages += nr_highmem; nr_copy_pages = nr_pages; - nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT; + nr_meta_pages = DIV_ROUND_UP(nr_pages * sizeof(long), PAGE_SIZE); printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages); + return 0; } @@ -960,7 +1082,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ - buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); + buffer = get_image_page(GFP_ATOMIC, PG_ANY); if (!buffer) return -ENOMEM; } @@ -975,9 +1097,23 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) memset(buffer, 0, PAGE_SIZE); pack_pfns(buffer, &orig_bm); } else { - unsigned long pfn = memory_bm_next_pfn(©_bm); + struct page *page; - handle->buffer = page_address(pfn_to_page(pfn)); + page = pfn_to_page(memory_bm_next_pfn(©_bm)); + if (PageHighMem(page)) { + /* Highmem pages are copied to the buffer, + * because we can't return with a kmapped + * highmem page (we may not be called again). + */ + void *kaddr; + + kaddr = kmap_atomic(page, KM_USER0); + memcpy(buffer, kaddr, PAGE_SIZE); + kunmap_atomic(kaddr, KM_USER0); + handle->buffer = buffer; + } else { + handle->buffer = page_address(page); + } } handle->prev = handle->cur; } @@ -1005,7 +1141,7 @@ static int mark_unsafe_pages(struct memory_bitmap *bm) unsigned long pfn, max_zone_pfn; /* Clear page flags */ - for_each_zone (zone) { + for_each_zone(zone) { max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) if (pfn_valid(pfn)) @@ -1101,6 +1237,218 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) } } +/* List of "safe" pages that may be used to store data loaded from the suspend + * image + */ +static struct linked_page *safe_pages_list; + +#ifdef CONFIG_HIGHMEM +/* struct highmem_pbe is used for creating the list of highmem pages that + * should be restored atomically during the resume from disk, because the page + * frames they have occupied before the suspend are in use. + */ +struct highmem_pbe { + struct page *copy_page; /* data is here now */ + struct page *orig_page; /* data was here before the suspend */ + struct highmem_pbe *next; +}; + +/* List of highmem PBEs needed for restoring the highmem pages that were + * allocated before the suspend and included in the suspend image, but have + * also been allocated by the "resume" kernel, so their contents cannot be + * written directly to their "original" page frames. + */ +static struct highmem_pbe *highmem_pblist; + +/** + * count_highmem_image_pages - compute the number of highmem pages in the + * suspend image. The bits in the memory bitmap @bm that correspond to the + * image pages are assumed to be set. + */ + +static unsigned int count_highmem_image_pages(struct memory_bitmap *bm) +{ + unsigned long pfn; + unsigned int cnt = 0; + + memory_bm_position_reset(bm); + pfn = memory_bm_next_pfn(bm); + while (pfn != BM_END_OF_MAP) { + if (PageHighMem(pfn_to_page(pfn))) + cnt++; + + pfn = memory_bm_next_pfn(bm); + } + return cnt; +} + +/** + * prepare_highmem_image - try to allocate as many highmem pages as + * there are highmem image pages (@nr_highmem_p points to the variable + * containing the number of highmem image pages). The pages that are + * "safe" (ie. will not be overwritten when the suspend image is + * restored) have the corresponding bits set in @bm (it must be + * unitialized). + * + * NOTE: This function should not be called if there are no highmem + * image pages. + */ + +static unsigned int safe_highmem_pages; + +static struct memory_bitmap *safe_highmem_bm; + +static int +prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p) +{ + unsigned int to_alloc; + + if (memory_bm_create(bm, GFP_ATOMIC, PG_SAFE)) + return -ENOMEM; + + if (get_highmem_buffer(PG_SAFE)) + return -ENOMEM; + + to_alloc = count_free_highmem_pages(); + if (to_alloc > *nr_highmem_p) + to_alloc = *nr_highmem_p; + else + *nr_highmem_p = to_alloc; + + safe_highmem_pages = 0; + while (to_alloc-- > 0) { + struct page *page; + + page = alloc_page(__GFP_HIGHMEM); + if (!PageNosaveFree(page)) { + /* The page is "safe", set its bit the bitmap */ + memory_bm_set_bit(bm, page_to_pfn(page)); + safe_highmem_pages++; + } + /* Mark the page as allocated */ + SetPageNosave(page); + SetPageNosaveFree(page); + } + memory_bm_position_reset(bm); + safe_highmem_bm = bm; + return 0; +} + +/** + * get_highmem_page_buffer - for given highmem image page find the buffer + * that suspend_write_next() should set for its caller to write to. + * + * If the page is to be saved to its "original" page frame or a copy of + * the page is to be made in the highmem, @buffer is returned. Otherwise, + * the copy of the page is to be made in normal memory, so the address of + * the copy is returned. + * + * If @buffer is returned, the caller of suspend_write_next() will write + * the page's contents to @buffer, so they will have to be copied to the + * right location on the next call to suspend_write_next() and it is done + * with the help of copy_last_highmem_page(). For this purpose, if + * @buffer is returned, @last_highmem page is set to the page to which + * the data will have to be copied from @buffer. + */ + +static struct page *last_highmem_page; + +static void * +get_highmem_page_buffer(struct page *page, struct chain_allocator *ca) +{ + struct highmem_pbe *pbe; + void *kaddr; + + if (PageNosave(page) && PageNosaveFree(page)) { + /* We have allocated the "original" page frame and we can + * use it directly to store the loaded page. + */ + last_highmem_page = page; + return buffer; + } + /* The "original" page frame has not been allocated and we have to + * use a "safe" page frame to store the loaded page. + */ + pbe = chain_alloc(ca, sizeof(struct highmem_pbe)); + if (!pbe) { + swsusp_free(); + return NULL; + } + pbe->orig_page = page; + if (safe_highmem_pages > 0) { + struct page *tmp; + + /* Copy of the page will be stored in high memory */ + kaddr = buffer; + tmp = pfn_to_page(memory_bm_next_pfn(safe_highmem_bm)); + safe_highmem_pages--; + last_highmem_page = tmp; + pbe->copy_page = tmp; + } else { + /* Copy of the page will be stored in normal memory */ + kaddr = safe_pages_list; + safe_pages_list = safe_pages_list->next; + pbe->copy_page = virt_to_page(kaddr); + } + pbe->next = highmem_pblist; + highmem_pblist = pbe; + return kaddr; +} + +/** + * copy_last_highmem_page - copy the contents of a highmem image from + * @buffer, where the caller of snapshot_write_next() has place them, + * to the right location represented by @last_highmem_page . + */ + +static void copy_last_highmem_page(void) +{ + if (last_highmem_page) { + void *dst; + + dst = kmap_atomic(last_highmem_page, KM_USER0); + memcpy(dst, buffer, PAGE_SIZE); + kunmap_atomic(dst, KM_USER0); + last_highmem_page = NULL; + } +} + +static inline int last_highmem_page_copied(void) +{ + return !last_highmem_page; +} + +static inline void free_highmem_data(void) +{ + if (safe_highmem_bm) + memory_bm_free(safe_highmem_bm, PG_UNSAFE_CLEAR); + + if (buffer) + free_image_page(buffer, PG_UNSAFE_CLEAR); +} +#else +static inline int get_safe_write_buffer(void) { return 0; } + +static unsigned int +count_highmem_image_pages(struct memory_bitmap *bm) { return 0; } + +static inline int +prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p) +{ + return 0; +} + +static inline void * +get_highmem_page_buffer(struct page *page, struct chain_allocator *ca) +{ + return NULL; +} + +static inline void copy_last_highmem_page(void) {} +static inline int last_highmem_page_copied(void) { return 1; } +static inline void free_highmem_data(void) {} +#endif /* CONFIG_HIGHMEM */ + /** * prepare_image - use the memory bitmap @bm to mark the pages that will * be overwritten in the process of restoring the system memory state @@ -1110,20 +1458,25 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) * The idea is to allocate a new memory bitmap first and then allocate * as many pages as needed for the image data, but not to assign these * pages to specific tasks initially. Instead, we just mark them as - * allocated and create a list of "safe" pages that will be used later. + * allocated and create a lists of "safe" pages that will be used + * later. On systems with high memory a list of "safe" highmem pages is + * also created. */ #define PBES_PER_LINKED_PAGE (LINKED_PAGE_DATA_SIZE / sizeof(struct pbe)) -static struct linked_page *safe_pages_list; - static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) { - unsigned int nr_pages; + unsigned int nr_pages, nr_highmem; struct linked_page *sp_list, *lp; int error; + /* If there is no highmem, the buffer will not be necessary */ + free_image_page(buffer, PG_UNSAFE_CLEAR); + buffer = NULL; + + nr_highmem = count_highmem_image_pages(bm); error = mark_unsafe_pages(bm); if (error) goto Free; @@ -1134,6 +1487,11 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) duplicate_memory_bitmap(new_bm, bm); memory_bm_free(bm, PG_UNSAFE_KEEP); + if (nr_highmem > 0) { + error = prepare_highmem_image(bm, &nr_highmem); + if (error) + goto Free; + } /* Reserve some safe pages for potential later use. * * NOTE: This way we make sure there will be enough safe pages for the @@ -1142,10 +1500,10 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) */ sp_list = NULL; /* nr_copy_pages cannot be lesser than allocated_unsafe_pages */ - nr_pages = nr_copy_pages - allocated_unsafe_pages; + nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages; nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE); while (nr_pages > 0) { - lp = alloc_image_page(GFP_ATOMIC, PG_SAFE); + lp = get_image_page(GFP_ATOMIC, PG_SAFE); if (!lp) { error = -ENOMEM; goto Free; @@ -1156,7 +1514,7 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) } /* Preallocate memory for the image */ safe_pages_list = NULL; - nr_pages = nr_copy_pages - allocated_unsafe_pages; + nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages; while (nr_pages > 0) { lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC); if (!lp) { @@ -1196,6 +1554,9 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca) struct pbe *pbe; struct page *page = pfn_to_page(memory_bm_next_pfn(bm)); + if (PageHighMem(page)) + return get_highmem_page_buffer(page, ca); + if (PageNosave(page) && PageNosaveFree(page)) /* We have allocated the "original" page frame and we can * use it directly to store the loaded page. @@ -1210,12 +1571,12 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca) swsusp_free(); return NULL; } - pbe->orig_address = (unsigned long)page_address(page); - pbe->address = (unsigned long)safe_pages_list; + pbe->orig_address = page_address(page); + pbe->address = safe_pages_list; safe_pages_list = safe_pages_list->next; pbe->next = restore_pblist; restore_pblist = pbe; - return (void *)pbe->address; + return pbe->address; } /** @@ -1249,14 +1610,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) return 0; - if (!buffer) { - /* This makes the buffer be freed by swsusp_free() */ - buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); + if (handle->offset == 0) { + if (!buffer) + /* This makes the buffer be freed by swsusp_free() */ + buffer = get_image_page(GFP_ATOMIC, PG_ANY); + if (!buffer) return -ENOMEM; - } - if (!handle->offset) + handle->buffer = buffer; + } handle->sync_read = 1; if (handle->prev < handle->cur) { if (handle->prev == 0) { @@ -1284,8 +1647,10 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) return -ENOMEM; } } else { + copy_last_highmem_page(); handle->buffer = get_buffer(&orig_bm, &ca); - handle->sync_read = 0; + if (handle->buffer != buffer) + handle->sync_read = 0; } handle->prev = handle->cur; } @@ -1301,15 +1666,73 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) return count; } +/** + * snapshot_write_finalize - must be called after the last call to + * snapshot_write_next() in case the last page in the image happens + * to be a highmem page and its contents should be stored in the + * highmem. Additionally, it releases the memory that will not be + * used any more. + */ + +void snapshot_write_finalize(struct snapshot_handle *handle) +{ + copy_last_highmem_page(); + /* Free only if we have loaded the image entirely */ + if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) { + memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR); + free_highmem_data(); + } +} + int snapshot_image_loaded(struct snapshot_handle *handle) { - return !(!nr_copy_pages || + return !(!nr_copy_pages || !last_highmem_page_copied() || handle->cur <= nr_meta_pages + nr_copy_pages); } -void snapshot_free_unused_memory(struct snapshot_handle *handle) +#ifdef CONFIG_HIGHMEM +/* Assumes that @buf is ready and points to a "safe" page */ +static inline void +swap_two_pages_data(struct page *p1, struct page *p2, void *buf) { - /* Free only if we have loaded the image entirely */ - if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) - memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR); + void *kaddr1, *kaddr2; + + kaddr1 = kmap_atomic(p1, KM_USER0); + kaddr2 = kmap_atomic(p2, KM_USER1); + memcpy(buf, kaddr1, PAGE_SIZE); + memcpy(kaddr1, kaddr2, PAGE_SIZE); + memcpy(kaddr2, buf, PAGE_SIZE); + kunmap_atomic(kaddr1, KM_USER0); + kunmap_atomic(kaddr2, KM_USER1); +} + +/** + * restore_highmem - for each highmem page that was allocated before + * the suspend and included in the suspend image, and also has been + * allocated by the "resume" kernel swap its current (ie. "before + * resume") contents with the previous (ie. "before suspend") one. + * + * If the resume eventually fails, we can call this function once + * again and restore the "before resume" highmem state. + */ + +int restore_highmem(void) +{ + struct highmem_pbe *pbe = highmem_pblist; + void *buf; + + if (!pbe) + return 0; + + buf = get_image_page(GFP_ATOMIC, PG_SAFE); + if (!buf) + return -ENOMEM; + + while (pbe) { + swap_two_pages_data(pbe->copy_page, pbe->orig_page, buf); + pbe = pbe->next; + } + free_image_page(buf, PG_UNSAFE_CLEAR); + return 0; } +#endif /* CONFIG_HIGHMEM */ diff --git a/kernel/power/swap.c b/kernel/power/swap.c index aa5a9bff01f1..cbd187e90410 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -558,7 +558,7 @@ static int load_image(struct swap_map_handle *handle, error = err2; if (!error) { printk("\b\b\b\bdone\n"); - snapshot_free_unused_memory(snapshot); + snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; } diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 4147a756a8c7..68de5c1dbd78 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -64,10 +64,8 @@ int in_suspend __nosavedata = 0; #ifdef CONFIG_HIGHMEM unsigned int count_highmem_pages(void); -int save_highmem(void); int restore_highmem(void); #else -static inline int save_highmem(void) { return 0; } static inline int restore_highmem(void) { return 0; } static inline unsigned int count_highmem_pages(void) { return 0; } #endif @@ -184,7 +182,7 @@ static inline unsigned long __shrink_memory(long tmp) int swsusp_shrink_memory(void) { - long size, tmp; + long tmp; struct zone *zone; unsigned long pages = 0; unsigned int i = 0; @@ -192,15 +190,27 @@ int swsusp_shrink_memory(void) printk("Shrinking memory... "); do { - size = 2 * count_highmem_pages(); - size += size / 50 + count_data_pages() + PAGES_FOR_IO; + long size, highmem_size; + + highmem_size = count_highmem_pages(); + size = count_data_pages() + PAGES_FOR_IO; tmp = size; + size += highmem_size; for_each_zone (zone) - if (!is_highmem(zone) && populated_zone(zone)) { - tmp -= zone->free_pages; - tmp += zone->lowmem_reserve[ZONE_NORMAL]; - tmp += snapshot_additional_pages(zone); + if (populated_zone(zone)) { + if (is_highmem(zone)) { + highmem_size -= zone->free_pages; + } else { + tmp -= zone->free_pages; + tmp += zone->lowmem_reserve[ZONE_NORMAL]; + tmp += snapshot_additional_pages(zone); + } } + + if (highmem_size < 0) + highmem_size = 0; + + tmp += highmem_size; if (tmp > 0) { tmp = __shrink_memory(tmp); if (!tmp) @@ -223,6 +233,7 @@ int swsusp_suspend(void) if ((error = arch_prepare_suspend())) return error; + local_irq_disable(); /* At this point, device_suspend() has been called, but *not* * device_power_down(). We *must* device_power_down() now. @@ -235,18 +246,11 @@ int swsusp_suspend(void) goto Enable_irqs; } - if ((error = save_highmem())) { - printk(KERN_ERR "swsusp: Not enough free pages for highmem\n"); - goto Restore_highmem; - } - save_processor_state(); if ((error = swsusp_arch_suspend())) printk(KERN_ERR "Error %d suspending\n", error); /* Restore control flow magically appears here */ restore_processor_state(); -Restore_highmem: - restore_highmem(); /* NOTE: device_power_up() is just a resume() for devices * that suspended with irqs off ... no overall powerup. */ @@ -268,18 +272,23 @@ int swsusp_resume(void) printk(KERN_ERR "Some devices failed to power down, very bad\n"); /* We'll ignore saved state, but this gets preempt count (etc) right */ save_processor_state(); - error = swsusp_arch_resume(); - /* Code below is only ever reached in case of failure. Otherwise - * execution continues at place where swsusp_arch_suspend was called - */ - BUG_ON(!error); + error = restore_highmem(); + if (!error) { + error = swsusp_arch_resume(); + /* The code below is only ever reached in case of a failure. + * Otherwise execution continues at place where + * swsusp_arch_suspend() was called + */ + BUG_ON(!error); + /* This call to restore_highmem() undos the previous one */ + restore_highmem(); + } /* The only reason why swsusp_arch_resume() can fail is memory being * very tight, so we have to free it as soon as we can to avoid * subsequent failures */ swsusp_free(); restore_processor_state(); - restore_highmem(); touch_softlockup_watchdog(); device_power_up(); local_irq_enable(); diff --git a/kernel/power/user.c b/kernel/power/user.c index 05c58a2c0dd4..a63b25c63b49 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -194,12 +194,12 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; case SNAPSHOT_ATOMIC_RESTORE: + snapshot_write_finalize(&data->handle); if (data->mode != O_WRONLY || !data->frozen || !snapshot_image_loaded(&data->handle)) { error = -EPERM; break; } - snapshot_free_unused_memory(&data->handle); down(&pm_sem); pm_prepare_console(); suspend_console(); diff --git a/mm/vmscan.c b/mm/vmscan.c index 2e97baa3b2aa..2a6a79f68138 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1260,6 +1260,9 @@ out: } if (!all_zones_ok) { cond_resched(); + + try_to_freeze(); + goto loop_again; } -- cgit v1.2.3 From 7dfb71030f7636a0d65200158113c37764552f93 Mon Sep 17 00:00:00 2001 From: Nigel Cunningham Date: Wed, 6 Dec 2006 20:34:23 -0800 Subject: [PATCH] Add include/linux/freezer.h and move definitions from sched.h Move process freezing functions from include/linux/sched.h to freezer.h, so that modifications to the freezer or the kernel configuration don't require recompiling just about everything. [akpm@osdl.org: fix ueagle driver] Signed-off-by: Nigel Cunningham Cc: "Rafael J. Wysocki" Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/signal.c | 1 + arch/avr32/kernel/signal.c | 2 +- arch/frv/kernel/signal.c | 2 +- arch/h8300/kernel/signal.c | 2 +- arch/i386/kernel/io_apic.c | 1 + arch/m32r/kernel/signal.c | 2 +- arch/powerpc/kernel/signal_32.c | 2 +- arch/sh/kernel/signal.c | 1 + arch/sh64/kernel/signal.c | 2 +- drivers/block/pktcdvd.c | 2 +- drivers/char/hvc_console.c | 1 + drivers/edac/edac_mc.c | 1 + drivers/ieee1394/nodemgr.c | 1 + drivers/input/gameport/gameport.c | 1 + drivers/input/serio/serio.c | 1 + drivers/macintosh/therm_adt746x.c | 1 + drivers/macintosh/via-pmu.c | 2 +- drivers/macintosh/windfarm_core.c | 1 + drivers/md/md.c | 2 +- drivers/media/dvb/dvb-core/dvb_frontend.c | 2 +- drivers/media/video/msp3400-driver.c | 2 +- drivers/media/video/tvaudio.c | 1 + drivers/media/video/video-buf-dvb.c | 2 +- drivers/media/video/vivi.c | 1 + drivers/mfd/ucb1x00-ts.c | 2 +- drivers/net/irda/stir4200.c | 1 + drivers/net/wireless/airo.c | 1 + drivers/pcmcia/cs.c | 1 + drivers/pnp/pnpbios/core.c | 1 + drivers/usb/atm/ueagle-atm.c | 2 + drivers/usb/core/hub.c | 1 + drivers/usb/gadget/file_storage.c | 2 +- drivers/usb/storage/usb.c | 2 +- drivers/w1/w1.c | 1 + fs/afs/kafsasyncd.c | 1 + fs/afs/kafstimod.c | 1 + fs/cifs/cifsfs.c | 1 + fs/cifs/connect.c | 1 + fs/jbd/journal.c | 2 +- fs/jbd2/journal.c | 2 +- fs/jffs/intrep.c | 1 + fs/jffs2/background.c | 1 + fs/jfs/jfs_logmgr.c | 2 +- fs/jfs/jfs_txnmgr.c | 2 +- fs/lockd/clntproc.c | 1 + fs/xfs/linux-2.6/xfs_buf.c | 1 + fs/xfs/linux-2.6/xfs_super.c | 1 + include/linux/freezer.h | 84 +++++++++++++++++++++++++++++++ include/linux/sched.h | 81 ----------------------------- init/do_mounts_initrd.c | 1 + kernel/audit.c | 1 + kernel/power/disk.c | 1 + kernel/power/main.c | 1 + kernel/power/process.c | 1 + kernel/power/user.c | 1 + kernel/rtmutex-tester.c | 1 + kernel/sched.c | 2 +- kernel/signal.c | 1 + mm/pdflush.c | 1 + mm/vmscan.c | 1 + net/rxrpc/krxiod.c | 1 + net/rxrpc/krxsecd.c | 1 + net/rxrpc/krxtimod.c | 1 + net/sunrpc/svcsock.c | 1 + 64 files changed, 147 insertions(+), 101 deletions(-) create mode 100644 include/linux/freezer.h (limited to 'include/linux') diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index 48cf7fffddf2..f38a60a03b8c 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include diff --git a/arch/avr32/kernel/signal.c b/arch/avr32/kernel/signal.c index 33096651c24f..0ec14854a200 100644 --- a/arch/avr32/kernel/signal.c +++ b/arch/avr32/kernel/signal.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/frv/kernel/signal.c b/arch/frv/kernel/signal.c index b8a5882b8625..85baeae9666a 100644 --- a/arch/frv/kernel/signal.c +++ b/arch/frv/kernel/signal.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/h8300/kernel/signal.c b/arch/h8300/kernel/signal.c index 7787f70a05bb..02955604d760 100644 --- a/arch/h8300/kernel/signal.c +++ b/arch/h8300/kernel/signal.c @@ -38,7 +38,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 3b7a63e0ed1a..44c5a3206b2a 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c index b60cea4aebaa..092ea86bb079 100644 --- a/arch/m32r/kernel/signal.c +++ b/arch/m32r/kernel/signal.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 320353f0926f..e4ebe1a6228e 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -36,7 +36,7 @@ #include #include #include -#include +#include #endif #include diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c index 50d7c4993bef..bb1c480a59c7 100644 --- a/arch/sh/kernel/signal.c +++ b/arch/sh/kernel/signal.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include diff --git a/arch/sh64/kernel/signal.c b/arch/sh64/kernel/signal.c index 9e2ffc45c0e0..1666d3efb52e 100644 --- a/arch/sh64/kernel/signal.c +++ b/arch/sh64/kernel/signal.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index f2904f67af47..e45eaa264119 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -54,7 +54,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c index 9902ffad3b12..cc2cd46bedc6 100644 --- a/drivers/char/hvc_console.c +++ b/drivers/char/hvc_console.c @@ -38,6 +38,7 @@ #include #include #include +#include #include diff --git a/drivers/edac/edac_mc.c b/drivers/edac/edac_mc.c index 75e9e38330ff..1b4fc9221803 100644 --- a/drivers/edac/edac_mc.c +++ b/drivers/edac/edac_mc.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c index 8e7b83f84485..e829c9336b3c 100644 --- a/drivers/ieee1394/nodemgr.c +++ b/drivers/ieee1394/nodemgr.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include "csr.h" diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c index a0af97efe6ac..79dfb4b25c97 100644 --- a/drivers/input/gameport/gameport.c +++ b/drivers/input/gameport/gameport.c @@ -23,6 +23,7 @@ #include #include /* HZ */ #include +#include /*#include */ diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c index 211943f85cb6..5f1d4032fd57 100644 --- a/drivers/input/serio/serio.c +++ b/drivers/input/serio/serio.c @@ -35,6 +35,7 @@ #include #include #include +#include MODULE_AUTHOR("Vojtech Pavlik "); MODULE_DESCRIPTION("Serio abstraction core"); diff --git a/drivers/macintosh/therm_adt746x.c b/drivers/macintosh/therm_adt746x.c index 13b953ae8ebc..3d3bf1643e73 100644 --- a/drivers/macintosh/therm_adt746x.c +++ b/drivers/macintosh/therm_adt746x.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index e63ea1c1f3c1..c8558d4ed506 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -42,7 +42,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/macintosh/windfarm_core.c b/drivers/macintosh/windfarm_core.c index ab3faa702d58..e947af982f93 100644 --- a/drivers/macintosh/windfarm_core.c +++ b/drivers/macintosh/windfarm_core.c @@ -34,6 +34,7 @@ #include #include #include +#include #include diff --git a/drivers/md/md.c b/drivers/md/md.c index 8cbf9c9df1c3..6c4345bde07e 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -39,10 +39,10 @@ #include #include #include /* for invalidate_bdev */ -#include #include #include #include +#include #include diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c index a2ab2eebfc68..e85972222ab4 100644 --- a/drivers/media/dvb/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb/dvb-core/dvb_frontend.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/media/video/msp3400-driver.c b/drivers/media/video/msp3400-driver.c index cf43df3fe708..e1b56dc13c3f 100644 --- a/drivers/media/video/msp3400-driver.c +++ b/drivers/media/video/msp3400-driver.c @@ -56,7 +56,7 @@ #include #include #include -#include +#include #include "msp3400-driver.h" /* ---------------------------------------------------------------------- */ diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index fcaef4bf8289..d506dfaa45a9 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include diff --git a/drivers/media/video/video-buf-dvb.c b/drivers/media/video/video-buf-dvb.c index f53edf1923b7..fcc5467e7636 100644 --- a/drivers/media/video/video-buf-dvb.c +++ b/drivers/media/video/video-buf-dvb.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/media/video/vivi.c b/drivers/media/video/vivi.c index 3c8dc72dc8e9..9986de5cb3d6 100644 --- a/drivers/media/video/vivi.c +++ b/drivers/media/video/vivi.c @@ -36,6 +36,7 @@ #include #include #include +#include /* Wake up at about 30 fps */ #define WAKE_NUMERATOR 30 diff --git a/drivers/mfd/ucb1x00-ts.c b/drivers/mfd/ucb1x00-ts.c index 82938ad6ddbd..ce1a48108210 100644 --- a/drivers/mfd/ucb1x00-ts.c +++ b/drivers/mfd/ucb1x00-ts.c @@ -28,7 +28,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/net/irda/stir4200.c b/drivers/net/irda/stir4200.c index 3b4c47875935..c14a74634fd5 100644 --- a/drivers/net/irda/stir4200.c +++ b/drivers/net/irda/stir4200.c @@ -50,6 +50,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c index efcdaf1c5f73..44a22701da97 100644 --- a/drivers/net/wireless/airo.c +++ b/drivers/net/wireless/airo.c @@ -49,6 +49,7 @@ #include #include #include +#include #include "airo.h" diff --git a/drivers/pcmcia/cs.c b/drivers/pcmcia/cs.c index f9cd831a3f31..606a46740338 100644 --- a/drivers/pcmcia/cs.c +++ b/drivers/pcmcia/cs.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include diff --git a/drivers/pnp/pnpbios/core.c b/drivers/pnp/pnpbios/core.c index 81a6c83d89a6..81186f479a3f 100644 --- a/drivers/pnp/pnpbios/core.c +++ b/drivers/pnp/pnpbios/core.c @@ -61,6 +61,7 @@ #include #include #include +#include #include #include diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c index f2d196fa1e8b..dae4ef1e8fe5 100644 --- a/drivers/usb/atm/ueagle-atm.c +++ b/drivers/usb/atm/ueagle-atm.c @@ -64,6 +64,8 @@ #include #include #include +#include + #include #include "usbatm.h" diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 77c05be5241a..2651c2e2a89f 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 8b975d15538d..c98316ce8384 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -250,7 +250,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c index b401084b3d22..70644506651f 100644 --- a/drivers/usb/storage/usb.c +++ b/drivers/usb/storage/usb.c @@ -49,7 +49,7 @@ #include #include -#include +#include #include #include #include diff --git a/drivers/w1/w1.c b/drivers/w1/w1.c index de3e9791f80d..63c07243993c 100644 --- a/drivers/w1/w1.c +++ b/drivers/w1/w1.c @@ -31,6 +31,7 @@ #include #include #include +#include #include diff --git a/fs/afs/kafsasyncd.c b/fs/afs/kafsasyncd.c index f09a794f248e..615df2407cb2 100644 --- a/fs/afs/kafsasyncd.c +++ b/fs/afs/kafsasyncd.c @@ -20,6 +20,7 @@ #include #include #include +#include #include "cell.h" #include "server.h" #include "volume.h" diff --git a/fs/afs/kafstimod.c b/fs/afs/kafstimod.c index 65bc05ab8182..694344e4d3c7 100644 --- a/fs/afs/kafstimod.c +++ b/fs/afs/kafstimod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "cell.h" #include "volume.h" #include "kafstimod.h" diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index e6b5866e5001..71bc87a37fc1 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -34,6 +34,7 @@ #include #include #include +#include #include "cifsfs.h" #include "cifspdu.h" #define DECLARE_GLOBALS_HERE diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 71f77914ce93..2caca06b4bae 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include "cifspdu.h" diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index a8774bed20b6..10fff9443938 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 50356019ae30..44fc32bfd7f1 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c index 4a543e114970..478a74e2e9d5 100644 --- a/fs/jffs/intrep.c +++ b/fs/jffs/intrep.c @@ -66,6 +66,7 @@ #include #include #include +#include #include "intrep.h" #include "jffs_fm.h" diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index ff2a872e80e7..6eb3daebd563 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "nodelist.h" diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index b89c9aba0466..5065baa530b6 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -67,7 +67,7 @@ #include #include /* for sync_blockdev() */ #include -#include +#include #include #include #include "jfs_incore.h" diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 81f6f04af192..d558e51b0df8 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -46,7 +46,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 3d84f600b633..50643b6a5556 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index eef4a0ba11e9..b971237c5a9f 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -32,6 +32,7 @@ #include #include #include +#include STATIC kmem_zone_t *xfs_buf_zone; STATIC kmem_shaker_t xfs_buf_shake; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index de05abbbe7fd..b93265b7c79c 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -56,6 +56,7 @@ #include #include #include +#include STATIC struct quotactl_ops xfs_quotactl_operations; STATIC struct super_operations xfs_super_operations; diff --git a/include/linux/freezer.h b/include/linux/freezer.h new file mode 100644 index 000000000000..266373f74445 --- /dev/null +++ b/include/linux/freezer.h @@ -0,0 +1,84 @@ +/* Freezer declarations */ + +#ifdef CONFIG_PM +/* + * Check if a process has been frozen + */ +static inline int frozen(struct task_struct *p) +{ + return p->flags & PF_FROZEN; +} + +/* + * Check if there is a request to freeze a process + */ +static inline int freezing(struct task_struct *p) +{ + return p->flags & PF_FREEZE; +} + +/* + * Request that a process be frozen + * FIXME: SMP problem. We may not modify other process' flags! + */ +static inline void freeze(struct task_struct *p) +{ + p->flags |= PF_FREEZE; +} + +/* + * Sometimes we may need to cancel the previous 'freeze' request + */ +static inline void do_not_freeze(struct task_struct *p) +{ + p->flags &= ~PF_FREEZE; +} + +/* + * Wake up a frozen process + */ +static inline int thaw_process(struct task_struct *p) +{ + if (frozen(p)) { + p->flags &= ~PF_FROZEN; + wake_up_process(p); + return 1; + } + return 0; +} + +/* + * freezing is complete, mark process as frozen + */ +static inline void frozen_process(struct task_struct *p) +{ + p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN; +} + +extern void refrigerator(void); +extern int freeze_processes(void); +extern void thaw_processes(void); + +static inline int try_to_freeze(void) +{ + if (freezing(current)) { + refrigerator(); + return 1; + } else + return 0; +} +#else +static inline int frozen(struct task_struct *p) { return 0; } +static inline int freezing(struct task_struct *p) { return 0; } +static inline void freeze(struct task_struct *p) { BUG(); } +static inline int thaw_process(struct task_struct *p) { return 1; } +static inline void frozen_process(struct task_struct *p) { BUG(); } + +static inline void refrigerator(void) {} +static inline int freeze_processes(void) { BUG(); return 0; } +static inline void thaw_processes(void) {} + +static inline int try_to_freeze(void) { return 0; } + + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index acfd2e15c5f2..837a012f573c 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1618,87 +1618,6 @@ extern int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls); extern void normalize_rt_tasks(void); -#ifdef CONFIG_PM -/* - * Check if a process has been frozen - */ -static inline int frozen(struct task_struct *p) -{ - return p->flags & PF_FROZEN; -} - -/* - * Check if there is a request to freeze a process - */ -static inline int freezing(struct task_struct *p) -{ - return p->flags & PF_FREEZE; -} - -/* - * Request that a process be frozen - * FIXME: SMP problem. We may not modify other process' flags! - */ -static inline void freeze(struct task_struct *p) -{ - p->flags |= PF_FREEZE; -} - -/* - * Sometimes we may need to cancel the previous 'freeze' request - */ -static inline void do_not_freeze(struct task_struct *p) -{ - p->flags &= ~PF_FREEZE; -} - -/* - * Wake up a frozen process - */ -static inline int thaw_process(struct task_struct *p) -{ - if (frozen(p)) { - p->flags &= ~PF_FROZEN; - wake_up_process(p); - return 1; - } - return 0; -} - -/* - * freezing is complete, mark process as frozen - */ -static inline void frozen_process(struct task_struct *p) -{ - p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN; -} - -extern void refrigerator(void); -extern int freeze_processes(void); -extern void thaw_processes(void); - -static inline int try_to_freeze(void) -{ - if (freezing(current)) { - refrigerator(); - return 1; - } else - return 0; -} -#else -static inline int frozen(struct task_struct *p) { return 0; } -static inline int freezing(struct task_struct *p) { return 0; } -static inline void freeze(struct task_struct *p) { BUG(); } -static inline int thaw_process(struct task_struct *p) { return 1; } -static inline void frozen_process(struct task_struct *p) { BUG(); } - -static inline void refrigerator(void) {} -static inline int freeze_processes(void) { BUG(); return 0; } -static inline void thaw_processes(void) {} - -static inline int try_to_freeze(void) { return 0; } - -#endif /* CONFIG_PM */ #endif /* __KERNEL__ */ #endif diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c index 919a80cb322e..2cfd7cb36e79 100644 --- a/init/do_mounts_initrd.c +++ b/init/do_mounts_initrd.c @@ -6,6 +6,7 @@ #include #include #include +#include #include "do_mounts.h" diff --git a/kernel/audit.c b/kernel/audit.c index 98106f6078b0..d9b690ac684b 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -57,6 +57,7 @@ #include #include #include +#include #include "audit.h" diff --git a/kernel/power/disk.c b/kernel/power/disk.c index f5079231383e..53b3b57c0223 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -20,6 +20,7 @@ #include #include #include +#include #include "power.h" diff --git a/kernel/power/main.c b/kernel/power/main.c index 873228c71dab..6096c71b182b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "power.h" diff --git a/kernel/power/process.c b/kernel/power/process.c index 72e72d2c61e6..29be608e8349 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -13,6 +13,7 @@ #include #include #include +#include /* * Timeout for stopping processes diff --git a/kernel/power/user.c b/kernel/power/user.c index a63b25c63b49..26c66941c001 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -22,6 +22,7 @@ #include #include #include +#include #include diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index 6dcea9dd8c94..015fc633c96c 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "rtmutex.h" diff --git a/kernel/sched.c b/kernel/sched.c index 3399701c680e..12fdbef1d9bf 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -34,7 +34,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/signal.c b/kernel/signal.c index 8e19d2785486..bc972d7e6319 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include diff --git a/mm/pdflush.c b/mm/pdflush.c index b02102feeb4b..8ce0900dc95c 100644 --- a/mm/pdflush.c +++ b/mm/pdflush.c @@ -21,6 +21,7 @@ #include // Prototypes pdflush_operation() #include #include +#include /* diff --git a/mm/vmscan.c b/mm/vmscan.c index 2a6a79f68138..f6616e81fac7 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include diff --git a/net/rxrpc/krxiod.c b/net/rxrpc/krxiod.c index dada34a77b21..49effd92144e 100644 --- a/net/rxrpc/krxiod.c +++ b/net/rxrpc/krxiod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/net/rxrpc/krxsecd.c b/net/rxrpc/krxsecd.c index cea4eb5e2497..3ab0f77409f4 100644 --- a/net/rxrpc/krxsecd.c +++ b/net/rxrpc/krxsecd.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include "internal.h" diff --git a/net/rxrpc/krxtimod.c b/net/rxrpc/krxtimod.c index 3e7466900bd4..9a9b6132dba4 100644 --- a/net/rxrpc/krxtimod.c +++ b/net/rxrpc/krxtimod.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 64ca1f61dd94..1c68956824e3 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From ff39593ad0ff7a79a3717edac6634407aa8200c2 Mon Sep 17 00:00:00 2001 From: Nigel Cunningham Date: Wed, 6 Dec 2006 20:34:28 -0800 Subject: [PATCH] swsusp: thaw userspace and kernel space separately Modify process thawing so that we can thaw kernel space without thawing userspace, and thaw kernelspace first. This will be useful in later patches, where I intend to get swsusp thawing kernel threads only before seeking to free memory. Signed-off-by: Nigel Cunningham Cc: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/freezer.h | 9 ++++++++- kernel/power/process.c | 25 ++++++++++++++++++------- 2 files changed, 26 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 266373f74445..294ebea859c9 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -1,5 +1,8 @@ /* Freezer declarations */ +#define FREEZER_KERNEL_THREADS 0 +#define FREEZER_ALL_THREADS 1 + #ifdef CONFIG_PM /* * Check if a process has been frozen @@ -57,7 +60,8 @@ static inline void frozen_process(struct task_struct *p) extern void refrigerator(void); extern int freeze_processes(void); -extern void thaw_processes(void); +#define thaw_processes() do { thaw_some_processes(FREEZER_ALL_THREADS); } while(0) +#define thaw_kernel_threads() do { thaw_some_processes(FREEZER_KERNEL_THREADS); } while(0) static inline int try_to_freeze(void) { @@ -67,6 +71,9 @@ static inline int try_to_freeze(void) } else return 0; } + +extern void thaw_some_processes(int all); + #else static inline int frozen(struct task_struct *p) { return 0; } static inline int freezing(struct task_struct *p) { return 0; } diff --git a/kernel/power/process.c b/kernel/power/process.c index fedabad5a180..cba8a5890eda 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -153,18 +153,29 @@ int freeze_processes(void) return 0; } -void thaw_processes(void) +void thaw_some_processes(int all) { struct task_struct *g, *p; + int pass = 0; /* Pass 0 = Kernel space, 1 = Userspace */ printk("Restarting tasks... "); read_lock(&tasklist_lock); - do_each_thread(g, p) { - if (!freezeable(p)) - continue; - if (!thaw_process(p)) - printk(KERN_INFO "Strange, %s not stopped\n", p->comm); - } while_each_thread(g, p); + do { + do_each_thread(g, p) { + /* + * is_user = 0 if kernel thread or borrowed mm, + * 1 otherwise. + */ + int is_user = !!(p->mm && !(p->flags & PF_BORROWED_MM)); + if (!freezeable(p) || (is_user != pass)) + continue; + if (!thaw_process(p)) + printk(KERN_INFO + "Strange, %s not stopped\n", p->comm); + } while_each_thread(g, p); + + pass++; + } while (pass < 2 && all); read_unlock(&tasklist_lock); schedule(); -- cgit v1.2.3 From a9b6f562f14dc28fb4b2415f0f275cede0abe9b5 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Wed, 6 Dec 2006 20:34:37 -0800 Subject: [PATCH] swsusp: Untangle thaw_processes Move the loop from thaw_processes() to a separate function and call it independently for kernel threads and user space processes so that the order of thawing tasks is clearly visible. Drop thaw_kernel_threads() which is never used. Signed-off-by: Rafael J. Wysocki Cc: Pavel Machek Cc: Nigel Cunningham Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/freezer.h | 6 +----- kernel/power/process.c | 49 +++++++++++++++++++++++++++---------------------- 2 files changed, 28 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 294ebea859c9..6e05e3e7ce39 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -1,8 +1,5 @@ /* Freezer declarations */ -#define FREEZER_KERNEL_THREADS 0 -#define FREEZER_ALL_THREADS 1 - #ifdef CONFIG_PM /* * Check if a process has been frozen @@ -60,8 +57,7 @@ static inline void frozen_process(struct task_struct *p) extern void refrigerator(void); extern int freeze_processes(void); -#define thaw_processes() do { thaw_some_processes(FREEZER_ALL_THREADS); } while(0) -#define thaw_kernel_threads() do { thaw_some_processes(FREEZER_KERNEL_THREADS); } while(0) +extern void thaw_processes(void); static inline int try_to_freeze(void) { diff --git a/kernel/power/process.c b/kernel/power/process.c index 1badb9a89ade..fd0ebb942f50 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -20,6 +20,8 @@ */ #define TIMEOUT (20 * HZ) +#define FREEZER_KERNEL_THREADS 0 +#define FREEZER_USER_SPACE 1 static inline int freezeable(struct task_struct * p) { @@ -79,6 +81,11 @@ static void cancel_freezing(struct task_struct *p) } } +static inline int is_user_space(struct task_struct *p) +{ + return p->mm && !(p->flags & PF_BORROWED_MM); +} + /* 0 = success, else # of processes that we failed to stop */ int freeze_processes(void) { @@ -103,10 +110,9 @@ int freeze_processes(void) cancel_freezing(p); continue; } - if (p->mm && !(p->flags & PF_BORROWED_MM)) { - /* The task is a user-space one. - * Freeze it unless there's a vfork completion - * pending + if (is_user_space(p)) { + /* Freeze the task unless there is a vfork + * completion pending */ if (!p->vfork_done) freeze_process(p); @@ -155,31 +161,30 @@ int freeze_processes(void) return 0; } -void thaw_some_processes(int all) +static void thaw_tasks(int thaw_user_space) { struct task_struct *g, *p; - int pass = 0; /* Pass 0 = Kernel space, 1 = Userspace */ - printk("Restarting tasks... "); read_lock(&tasklist_lock); - do { - do_each_thread(g, p) { - /* - * is_user = 0 if kernel thread or borrowed mm, - * 1 otherwise. - */ - int is_user = !!(p->mm && !(p->flags & PF_BORROWED_MM)); - if (!freezeable(p) || (is_user != pass)) - continue; - if (!thaw_process(p)) - printk(KERN_INFO - "Strange, %s not stopped\n", p->comm); - } while_each_thread(g, p); + do_each_thread(g, p) { + if (!freezeable(p)) + continue; - pass++; - } while (pass < 2 && all); + if (is_user_space(p) == !thaw_user_space) + continue; + if (!thaw_process(p)) + printk(KERN_WARNING " Strange, %s not stopped\n", + p->comm ); + } while_each_thread(g, p); read_unlock(&tasklist_lock); +} + +void thaw_processes(void) +{ + printk("Restarting tasks ... "); + thaw_tasks(FREEZER_KERNEL_THREADS); + thaw_tasks(FREEZER_USER_SPACE); schedule(); printk("done.\n"); } -- cgit v1.2.3 From 341a595850dac1b0503df34260257d71b4fdf72c Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Wed, 6 Dec 2006 20:34:49 -0800 Subject: [PATCH] Support for freezeable workqueues Make it possible to create a workqueue the worker thread of which will be frozen during suspend, along with other kernel threads. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Cc: Nigel Cunningham Cc: David Chinner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/workqueue.h | 8 +++++--- kernel/workqueue.c | 20 ++++++++++++++------ 2 files changed, 19 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 4a3ea83c6d16..f0cb1df7b475 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -147,9 +147,11 @@ struct execute_work { extern struct workqueue_struct *__create_workqueue(const char *name, - int singlethread); -#define create_workqueue(name) __create_workqueue((name), 0) -#define create_singlethread_workqueue(name) __create_workqueue((name), 1) + int singlethread, + int freezeable); +#define create_workqueue(name) __create_workqueue((name), 0, 0) +#define create_freezeable_workqueue(name) __create_workqueue((name), 0, 1) +#define create_singlethread_workqueue(name) __create_workqueue((name), 1, 0) extern void destroy_workqueue(struct workqueue_struct *wq); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 8d1e7cb8a51a..2945b094d871 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -29,6 +29,7 @@ #include #include #include +#include /* * The per-CPU workqueue (if single thread, we always use the first @@ -55,6 +56,8 @@ struct cpu_workqueue_struct { struct task_struct *thread; int run_depth; /* Detect run_workqueue() recursion depth */ + + int freezeable; /* Freeze the thread during suspend */ } ____cacheline_aligned; /* @@ -265,7 +268,8 @@ static int worker_thread(void *__cwq) struct k_sigaction sa; sigset_t blocked; - current->flags |= PF_NOFREEZE; + if (!cwq->freezeable) + current->flags |= PF_NOFREEZE; set_user_nice(current, -5); @@ -288,6 +292,9 @@ static int worker_thread(void *__cwq) set_current_state(TASK_INTERRUPTIBLE); while (!kthread_should_stop()) { + if (cwq->freezeable) + try_to_freeze(); + add_wait_queue(&cwq->more_work, &wait); if (list_empty(&cwq->worklist)) schedule(); @@ -364,7 +371,7 @@ void fastcall flush_workqueue(struct workqueue_struct *wq) EXPORT_SYMBOL_GPL(flush_workqueue); static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, - int cpu) + int cpu, int freezeable) { struct cpu_workqueue_struct *cwq = per_cpu_ptr(wq->cpu_wq, cpu); struct task_struct *p; @@ -374,6 +381,7 @@ static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, cwq->thread = NULL; cwq->insert_sequence = 0; cwq->remove_sequence = 0; + cwq->freezeable = freezeable; INIT_LIST_HEAD(&cwq->worklist); init_waitqueue_head(&cwq->more_work); init_waitqueue_head(&cwq->work_done); @@ -389,7 +397,7 @@ static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, } struct workqueue_struct *__create_workqueue(const char *name, - int singlethread) + int singlethread, int freezeable) { int cpu, destroy = 0; struct workqueue_struct *wq; @@ -409,7 +417,7 @@ struct workqueue_struct *__create_workqueue(const char *name, mutex_lock(&workqueue_mutex); if (singlethread) { INIT_LIST_HEAD(&wq->list); - p = create_workqueue_thread(wq, singlethread_cpu); + p = create_workqueue_thread(wq, singlethread_cpu, freezeable); if (!p) destroy = 1; else @@ -417,7 +425,7 @@ struct workqueue_struct *__create_workqueue(const char *name, } else { list_add(&wq->list, &workqueues); for_each_online_cpu(cpu) { - p = create_workqueue_thread(wq, cpu); + p = create_workqueue_thread(wq, cpu, freezeable); if (p) { kthread_bind(p, cpu); wake_up_process(p); @@ -667,7 +675,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, mutex_lock(&workqueue_mutex); /* Create a new workqueue thread for it. */ list_for_each_entry(wq, &workqueues, list) { - if (!create_workqueue_thread(wq, hotcpu)) { + if (!create_workqueue_thread(wq, hotcpu, 0)) { printk("workqueue for %i failed\n", hotcpu); return NOTIFY_BAD; } -- cgit v1.2.3 From 799202cbd0ef6a201446d99fcbd78b9f0bda6ae5 Mon Sep 17 00:00:00 2001 From: Mike Miller Date: Wed, 6 Dec 2006 20:35:12 -0800 Subject: [PATCH] cciss: add support for 1024 logical volumes Add the support for a large number of logical volumes. We will soon have hardware that support up to 1024 logical volumes. Signed-off-by: Mike Miller Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/cciss.c | 114 ++++++++++++++++++++++++++++++-------------- drivers/block/cciss.h | 3 +- drivers/block/cciss_cmd.h | 2 +- include/linux/cciss_ioctl.h | 2 +- 4 files changed, 80 insertions(+), 41 deletions(-) (limited to 'include/linux') diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 70cf9321ef54..c99cb7ed57f7 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -1315,6 +1315,11 @@ static void cciss_update_drive_info(int ctlr, int drv_index) /* if it's the controller it's already added */ if (drv_index) { disk->queue = blk_init_queue(do_cciss_request, &h->lock); + sprintf(disk->disk_name, "cciss/c%dd%d", ctlr, drv_index); + disk->major = h->major; + disk->first_minor = drv_index << NWD_SHIFT; + disk->fops = &cciss_fops; + disk->private_data = &h->drv[drv_index]; /* Set up queue information */ disk->queue->backing_dev_info.ra_pages = READ_AHEAD; @@ -1393,11 +1398,6 @@ static int rebuild_lun_table(ctlr_info_t *h, struct gendisk *del_disk) /* Set busy_configuring flag for this operation */ spin_lock_irqsave(CCISS_LOCK(h->ctlr), flags); - if (h->num_luns >= CISS_MAX_LUN) { - spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); - return -EINVAL; - } - if (h->busy_configuring) { spin_unlock_irqrestore(CCISS_LOCK(h->ctlr), flags); return -EBUSY; @@ -1430,17 +1430,8 @@ static int rebuild_lun_table(ctlr_info_t *h, struct gendisk *del_disk) 0, 0, TYPE_CMD); if (return_code == IO_OK) { - listlength |= - (0xff & (unsigned int)(ld_buff->LUNListLength[0])) - << 24; - listlength |= - (0xff & (unsigned int)(ld_buff->LUNListLength[1])) - << 16; - listlength |= - (0xff & (unsigned int)(ld_buff->LUNListLength[2])) - << 8; - listlength |= - 0xff & (unsigned int)(ld_buff->LUNListLength[3]); + listlength = + be32_to_cpu(*(__u32 *) ld_buff->LUNListLength); } else { /* reading number of logical volumes failed */ printk(KERN_WARNING "cciss: report logical volume" " command failed\n"); @@ -1491,6 +1482,14 @@ static int rebuild_lun_table(ctlr_info_t *h, struct gendisk *del_disk) if (drv_index == -1) goto freeret; + /*Check if the gendisk needs to be allocated */ + if (!h->gendisk[drv_index]){ + h->gendisk[drv_index] = alloc_disk(1 << NWD_SHIFT); + if (!h->gendisk[drv_index]){ + printk(KERN_ERR "cciss: could not allocate new disk %d\n", drv_index); + goto mem_msg; + } + } } h->drv[drv_index].LunID = lunid; cciss_update_drive_info(ctlr, drv_index); @@ -1528,6 +1527,7 @@ static int rebuild_lun_table(ctlr_info_t *h, struct gendisk *del_disk) static int deregister_disk(struct gendisk *disk, drive_info_struct *drv, int clear_all) { + int i; ctlr_info_t *h = get_host(disk); if (!capable(CAP_SYS_RAWIO)) @@ -1551,9 +1551,35 @@ static int deregister_disk(struct gendisk *disk, drive_info_struct *drv, del_gendisk(disk); if (q) { blk_cleanup_queue(q); + /* Set drv->queue to NULL so that we do not try + * to call blk_start_queue on this queue in the + * interrupt handler + */ drv->queue = NULL; } + /* If clear_all is set then we are deleting the logical + * drive, not just refreshing its info. For drives + * other than disk 0 we will call put_disk. We do not + * do this for disk 0 as we need it to be able to + * configure the controller. + */ + if (clear_all){ + /* This isn't pretty, but we need to find the + * disk in our array and NULL our the pointer. + * This is so that we will call alloc_disk if + * this index is used again later. + */ + for (i=0; i < CISS_MAX_LUN; i++){ + if(h->gendisk[i] == disk){ + h->gendisk[i] = NULL; + break; + } + } + put_disk(disk); + } } + } else { + set_capacity(disk, 0); } --h->num_luns; @@ -3119,13 +3145,7 @@ geo_inq: /* Returns -1 if no free entries are left. */ static int alloc_cciss_hba(void) { - struct gendisk *disk[NWD]; - int i, n; - for (n = 0; n < NWD; n++) { - disk[n] = alloc_disk(1 << NWD_SHIFT); - if (!disk[n]) - goto out; - } + int i; for (i = 0; i < MAX_CTLR; i++) { if (!hba[i]) { @@ -3133,20 +3153,18 @@ static int alloc_cciss_hba(void) p = kzalloc(sizeof(ctlr_info_t), GFP_KERNEL); if (!p) goto Enomem; - for (n = 0; n < NWD; n++) - p->gendisk[n] = disk[n]; + p->gendisk[0] = alloc_disk(1 << NWD_SHIFT); + if (!p->gendisk[0]) + goto Enomem; hba[i] = p; return i; } } printk(KERN_WARNING "cciss: This driver supports a maximum" " of %d controllers.\n", MAX_CTLR); - goto out; - Enomem: + return -1; +Enomem: printk(KERN_ERR "cciss: out of memory.\n"); - out: - while (n--) - put_disk(disk[n]); return -1; } @@ -3156,7 +3174,7 @@ static void free_hba(int i) int n; hba[i] = NULL; - for (n = 0; n < NWD; n++) + for (n = 0; n < CISS_MAX_LUN; n++) put_disk(p->gendisk[n]); kfree(p); } @@ -3169,9 +3187,8 @@ static void free_hba(int i) static int __devinit cciss_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { - request_queue_t *q; int i; - int j; + int j = 0; int rc; int dac; @@ -3283,16 +3300,29 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, hba[i]->busy_initializing = 0; - for (j = 0; j < NWD; j++) { /* mfm */ + do { drive_info_struct *drv = &(hba[i]->drv[j]); struct gendisk *disk = hba[i]->gendisk[j]; + request_queue_t *q; + + /* Check if the disk was allocated already */ + if (!disk){ + hba[i]->gendisk[j] = alloc_disk(1 << NWD_SHIFT); + disk = hba[i]->gendisk[j]; + } + + /* Check that the disk was able to be allocated */ + if (!disk) { + printk(KERN_ERR "cciss: unable to allocate memory for disk %d\n", j); + goto clean4; + } q = blk_init_queue(do_cciss_request, &hba[i]->lock); if (!q) { printk(KERN_ERR "cciss: unable to allocate queue for disk %d\n", j); - break; + goto clean4; } drv->queue = q; @@ -3324,7 +3354,8 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, blk_queue_hardsect_size(q, drv->block_size); set_capacity(disk, drv->nr_blocks); add_disk(disk); - } + j++; + } while (j <= hba[i]->highest_lun); return 1; @@ -3347,6 +3378,15 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, unregister_blkdev(hba[i]->major, hba[i]->devname); clean1: hba[i]->busy_initializing = 0; + /* cleanup any queues that may have been initialized */ + for (j=0; j <= hba[i]->highest_lun; j++){ + drive_info_struct *drv = &(hba[i]->drv[j]); + if (drv->queue) + blk_cleanup_queue(drv->queue); + } + pci_release_regions(pdev); + pci_disable_device(pdev); + pci_set_drvdata(pdev, NULL); free_hba(i); return -1; } @@ -3394,7 +3434,7 @@ static void __devexit cciss_remove_one(struct pci_dev *pdev) remove_proc_entry(hba[i]->devname, proc_cciss); /* remove it from the disk list */ - for (j = 0; j < NWD; j++) { + for (j = 0; j < CISS_MAX_LUN; j++) { struct gendisk *disk = hba[i]->gendisk[j]; if (disk) { request_queue_t *q = disk->queue; diff --git a/drivers/block/cciss.h b/drivers/block/cciss.h index c3df673f860b..b70988dd33ec 100644 --- a/drivers/block/cciss.h +++ b/drivers/block/cciss.h @@ -6,7 +6,6 @@ #include "cciss_cmd.h" -#define NWD 16 #define NWD_SHIFT 4 #define MAX_PART (1 << NWD_SHIFT) @@ -112,7 +111,7 @@ struct ctlr_info int next_to_run; // Disk structures we need to pass back - struct gendisk *gendisk[NWD]; + struct gendisk *gendisk[CISS_MAX_LUN]; #ifdef CONFIG_CISS_SCSI_TAPE void *scsi_ctlr; /* ptr to structure containing scsi related stuff */ /* list of block side commands the scsi error handling sucked up */ diff --git a/drivers/block/cciss_cmd.h b/drivers/block/cciss_cmd.h index b2147cca5acd..43bf5593b59b 100644 --- a/drivers/block/cciss_cmd.h +++ b/drivers/block/cciss_cmd.h @@ -89,7 +89,7 @@ typedef union _u64bit //########################################################################### //STRUCTURES //########################################################################### -#define CISS_MAX_LUN 16 +#define CISS_MAX_LUN 1024 #define CISS_MAX_PHYS_LUN 1024 // SCSI-3 Cmmands diff --git a/include/linux/cciss_ioctl.h b/include/linux/cciss_ioctl.h index 6e27f42e3a57..cb57c30081a8 100644 --- a/include/linux/cciss_ioctl.h +++ b/include/linux/cciss_ioctl.h @@ -80,7 +80,7 @@ typedef __u32 DriverVer_type; #define HWORD __u16 #define DWORD __u32 -#define CISS_MAX_LUN 16 +#define CISS_MAX_LUN 1024 #define LEVEL2LUN 1 // index into Target(x) structure, due to byte swapping #define LEVEL3LUN 0 -- cgit v1.2.3 From 238b8721a554a33a451a3f13bdb5be8fe5cfc927 Mon Sep 17 00:00:00 2001 From: Peter Korsgaard Date: Wed, 6 Dec 2006 20:35:17 -0800 Subject: [PATCH] serial uartlite driver Add a driver for the Xilinx uartlite serial controller used in boards with the PPC405 core in the Xilinx V2P/V4 fpgas. The hardware is very simple (baudrate/start/stopbits fixed and no break support). See the datasheet for details: http://www.xilinx.com/bvdocs/ipcenter/data_sheet/opb_uartlite.pdf See http://thread.gmane.org/gmane.linux.serial/1237/ for the email thread. Signed-off-by: Peter Korsgaard Acked-by: Olof Johansson Cc: Russell King Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 6 + drivers/serial/Kconfig | 19 ++ drivers/serial/Makefile | 1 + drivers/serial/uartlite.c | 505 ++++++++++++++++++++++++++++++++++++++++++++ include/linux/serial_core.h | 2 + 5 files changed, 533 insertions(+) create mode 100644 drivers/serial/uartlite.c (limited to 'include/linux') diff --git a/MAINTAINERS b/MAINTAINERS index 5dff26814a06..fa1bba8de1f8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3454,6 +3454,12 @@ W: http://oss.sgi.com/projects/xfs T: git git://oss.sgi.com:8090/xfs/xfs-2.6 S: Supported +XILINX UARTLITE SERIAL DRIVER +P: Peter Korsgaard +M: jacmet@sunsite.dk +L: linux-serial@vger.kernel.org +S: Maintained + X86 3-LEVEL PAGING (PAE) SUPPORT P: Ingo Molnar M: mingo@redhat.com diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index 0b71e7d18903..e936c91640b4 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -511,6 +511,25 @@ config SERIAL_IMX_CONSOLE your boot loader (lilo or loadlin) about how to pass options to the kernel at boot time.) +config SERIAL_UARTLITE + tristate "Xilinx uartlite serial port support" + depends on PPC32 + select SERIAL_CORE + help + Say Y here if you want to use the Xilinx uartlite serial controller. + + To compile this driver as a module, choose M here: the + module will be called uartlite.ko. + +config SERIAL_UARTLITE_CONSOLE + bool "Support for console on Xilinx uartlite serial port" + depends on SERIAL_UARTLITE=y + select SERIAL_CORE_CONSOLE + help + Say Y here if you wish to use a Xilinx uartlite as the system + console (the system console is the device which receives all kernel + messages and warnings and which allows logins in single user mode). + config SERIAL_SUNCORE bool depends on SPARC diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile index b4d8a7c182e3..0dba0017a006 100644 --- a/drivers/serial/Makefile +++ b/drivers/serial/Makefile @@ -55,4 +55,5 @@ obj-$(CONFIG_SERIAL_VR41XX) += vr41xx_siu.o obj-$(CONFIG_SERIAL_SGI_IOC4) += ioc4_serial.o obj-$(CONFIG_SERIAL_SGI_IOC3) += ioc3_serial.o obj-$(CONFIG_SERIAL_ATMEL) += atmel_serial.o +obj-$(CONFIG_SERIAL_UARTLITE) += uartlite.o obj-$(CONFIG_SERIAL_NETX) += netx-serial.o diff --git a/drivers/serial/uartlite.c b/drivers/serial/uartlite.c new file mode 100644 index 000000000000..83690653b78b --- /dev/null +++ b/drivers/serial/uartlite.c @@ -0,0 +1,505 @@ +/* + * uartlite.c: Serial driver for Xilinx uartlite serial controller + * + * Peter Korsgaard + * + * This file is licensed under the terms of the GNU General Public License + * version 2. This program is licensed "as is" without any warranty of any + * kind, whether express or implied. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define ULITE_MAJOR 204 +#define ULITE_MINOR 187 +#define ULITE_NR_UARTS 4 + +/* For register details see datasheet: + http://www.xilinx.com/bvdocs/ipcenter/data_sheet/opb_uartlite.pdf +*/ +#define ULITE_RX 0x00 +#define ULITE_TX 0x04 +#define ULITE_STATUS 0x08 +#define ULITE_CONTROL 0x0c + +#define ULITE_REGION 16 + +#define ULITE_STATUS_RXVALID 0x01 +#define ULITE_STATUS_RXFULL 0x02 +#define ULITE_STATUS_TXEMPTY 0x04 +#define ULITE_STATUS_TXFULL 0x08 +#define ULITE_STATUS_IE 0x10 +#define ULITE_STATUS_OVERRUN 0x20 +#define ULITE_STATUS_FRAME 0x40 +#define ULITE_STATUS_PARITY 0x80 + +#define ULITE_CONTROL_RST_TX 0x01 +#define ULITE_CONTROL_RST_RX 0x02 +#define ULITE_CONTROL_IE 0x10 + + +static struct uart_port ports[ULITE_NR_UARTS]; + +static int ulite_receive(struct uart_port *port, int stat) +{ + struct tty_struct *tty = port->info->tty; + unsigned char ch = 0; + char flag = TTY_NORMAL; + + if ((stat & (ULITE_STATUS_RXVALID | ULITE_STATUS_OVERRUN + | ULITE_STATUS_FRAME)) == 0) + return 0; + + /* stats */ + if (stat & ULITE_STATUS_RXVALID) { + port->icount.rx++; + ch = readb(port->membase + ULITE_RX); + + if (stat & ULITE_STATUS_PARITY) + port->icount.parity++; + } + + if (stat & ULITE_STATUS_OVERRUN) + port->icount.overrun++; + + if (stat & ULITE_STATUS_FRAME) + port->icount.frame++; + + + /* drop byte with parity error if IGNPAR specificed */ + if (stat & port->ignore_status_mask & ULITE_STATUS_PARITY) + stat &= ~ULITE_STATUS_RXVALID; + + stat &= port->read_status_mask; + + if (stat & ULITE_STATUS_PARITY) + flag = TTY_PARITY; + + + stat &= ~port->ignore_status_mask; + + if (stat & ULITE_STATUS_RXVALID) + tty_insert_flip_char(tty, ch, flag); + + if (stat & ULITE_STATUS_FRAME) + tty_insert_flip_char(tty, 0, TTY_FRAME); + + if (stat & ULITE_STATUS_OVERRUN) + tty_insert_flip_char(tty, 0, TTY_OVERRUN); + + return 1; +} + +static int ulite_transmit(struct uart_port *port, int stat) +{ + struct circ_buf *xmit = &port->info->xmit; + + if (stat & ULITE_STATUS_TXFULL) + return 0; + + if (port->x_char) { + writeb(port->x_char, port->membase + ULITE_TX); + port->x_char = 0; + port->icount.tx++; + return 1; + } + + if (uart_circ_empty(xmit) || uart_tx_stopped(port)) + return 0; + + writeb(xmit->buf[xmit->tail], port->membase + ULITE_TX); + xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE-1); + port->icount.tx++; + + /* wake up */ + if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS) + uart_write_wakeup(port); + + return 1; +} + +static irqreturn_t ulite_isr(int irq, void *dev_id) +{ + struct uart_port *port = (struct uart_port *)dev_id; + int busy; + + do { + int stat = readb(port->membase + ULITE_STATUS); + busy = ulite_receive(port, stat); + busy |= ulite_transmit(port, stat); + } while (busy); + + tty_flip_buffer_push(port->info->tty); + + return IRQ_HANDLED; +} + +static unsigned int ulite_tx_empty(struct uart_port *port) +{ + unsigned long flags; + unsigned int ret; + + spin_lock_irqsave(&port->lock, flags); + ret = readb(port->membase + ULITE_STATUS); + spin_unlock_irqrestore(&port->lock, flags); + + return ret & ULITE_STATUS_TXEMPTY ? TIOCSER_TEMT : 0; +} + +static unsigned int ulite_get_mctrl(struct uart_port *port) +{ + return TIOCM_CTS | TIOCM_DSR | TIOCM_CAR; +} + +static void ulite_set_mctrl(struct uart_port *port, unsigned int mctrl) +{ + /* N/A */ +} + +static void ulite_stop_tx(struct uart_port *port) +{ + /* N/A */ +} + +static void ulite_start_tx(struct uart_port *port) +{ + ulite_transmit(port, readb(port->membase + ULITE_STATUS)); +} + +static void ulite_stop_rx(struct uart_port *port) +{ + /* don't forward any more data (like !CREAD) */ + port->ignore_status_mask = ULITE_STATUS_RXVALID | ULITE_STATUS_PARITY + | ULITE_STATUS_FRAME | ULITE_STATUS_OVERRUN; +} + +static void ulite_enable_ms(struct uart_port *port) +{ + /* N/A */ +} + +static void ulite_break_ctl(struct uart_port *port, int ctl) +{ + /* N/A */ +} + +static int ulite_startup(struct uart_port *port) +{ + int ret; + + ret = request_irq(port->irq, ulite_isr, + IRQF_DISABLED | IRQF_SAMPLE_RANDOM, "uartlite", port); + if (ret) + return ret; + + writeb(ULITE_CONTROL_RST_RX | ULITE_CONTROL_RST_TX, + port->membase + ULITE_CONTROL); + writeb(ULITE_CONTROL_IE, port->membase + ULITE_CONTROL); + + return 0; +} + +static void ulite_shutdown(struct uart_port *port) +{ + writeb(0, port->membase + ULITE_CONTROL); + readb(port->membase + ULITE_CONTROL); /* dummy */ + free_irq(port->irq, port); +} + +static void ulite_set_termios(struct uart_port *port, struct termios *termios, + struct termios *old) +{ + unsigned long flags; + unsigned int baud; + + spin_lock_irqsave(&port->lock, flags); + + port->read_status_mask = ULITE_STATUS_RXVALID | ULITE_STATUS_OVERRUN + | ULITE_STATUS_TXFULL; + + if (termios->c_iflag & INPCK) + port->read_status_mask |= + ULITE_STATUS_PARITY | ULITE_STATUS_FRAME; + + port->ignore_status_mask = 0; + if (termios->c_iflag & IGNPAR) + port->ignore_status_mask |= ULITE_STATUS_PARITY + | ULITE_STATUS_FRAME | ULITE_STATUS_OVERRUN; + + /* ignore all characters if CREAD is not set */ + if ((termios->c_cflag & CREAD) == 0) + port->ignore_status_mask |= + ULITE_STATUS_RXVALID | ULITE_STATUS_PARITY + | ULITE_STATUS_FRAME | ULITE_STATUS_OVERRUN; + + /* update timeout */ + baud = uart_get_baud_rate(port, termios, old, 0, 460800); + uart_update_timeout(port, termios->c_cflag, baud); + + spin_unlock_irqrestore(&port->lock, flags); +} + +static const char *ulite_type(struct uart_port *port) +{ + return port->type == PORT_UARTLITE ? "uartlite" : NULL; +} + +static void ulite_release_port(struct uart_port *port) +{ + release_mem_region(port->mapbase, ULITE_REGION); + iounmap(port->membase); + port->membase = 0; +} + +static int ulite_request_port(struct uart_port *port) +{ + if (!request_mem_region(port->mapbase, ULITE_REGION, "uartlite")) { + dev_err(port->dev, "Memory region busy\n"); + return -EBUSY; + } + + port->membase = ioremap(port->mapbase, ULITE_REGION); + if (!port->membase) { + dev_err(port->dev, "Unable to map registers\n"); + release_mem_region(port->mapbase, ULITE_REGION); + return -EBUSY; + } + + return 0; +} + +static void ulite_config_port(struct uart_port *port, int flags) +{ + ulite_request_port(port); + port->type = PORT_UARTLITE; +} + +static int ulite_verify_port(struct uart_port *port, struct serial_struct *ser) +{ + /* we don't want the core code to modify any port params */ + return -EINVAL; +} + +static struct uart_ops ulite_ops = { + .tx_empty = ulite_tx_empty, + .set_mctrl = ulite_set_mctrl, + .get_mctrl = ulite_get_mctrl, + .stop_tx = ulite_stop_tx, + .start_tx = ulite_start_tx, + .stop_rx = ulite_stop_rx, + .enable_ms = ulite_enable_ms, + .break_ctl = ulite_break_ctl, + .startup = ulite_startup, + .shutdown = ulite_shutdown, + .set_termios = ulite_set_termios, + .type = ulite_type, + .release_port = ulite_release_port, + .request_port = ulite_request_port, + .config_port = ulite_config_port, + .verify_port = ulite_verify_port +}; + +#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE +static void ulite_console_wait_tx(struct uart_port *port) +{ + int i; + + /* wait up to 10ms for the character(s) to be sent */ + for (i = 0; i < 10000; i++) { + if (readb(port->membase + ULITE_STATUS) & ULITE_STATUS_TXEMPTY) + break; + udelay(1); + } +} + +static void ulite_console_putchar(struct uart_port *port, int ch) +{ + ulite_console_wait_tx(port); + writeb(ch, port->membase + ULITE_TX); +} + +static void ulite_console_write(struct console *co, const char *s, + unsigned int count) +{ + struct uart_port *port = &ports[co->index]; + unsigned long flags; + unsigned int ier; + int locked = 1; + + if (oops_in_progress) { + locked = spin_trylock_irqsave(&port->lock, flags); + } else + spin_lock_irqsave(&port->lock, flags); + + /* save and disable interrupt */ + ier = readb(port->membase + ULITE_STATUS) & ULITE_STATUS_IE; + writeb(0, port->membase + ULITE_CONTROL); + + uart_console_write(port, s, count, ulite_console_putchar); + + ulite_console_wait_tx(port); + + /* restore interrupt state */ + if (ier) + writeb(ULITE_CONTROL_IE, port->membase + ULITE_CONTROL); + + if (locked) + spin_unlock_irqrestore(&port->lock, flags); +} + +static int __init ulite_console_setup(struct console *co, char *options) +{ + struct uart_port *port; + int baud = 9600; + int bits = 8; + int parity = 'n'; + int flow = 'n'; + + if (co->index < 0 || co->index >= ULITE_NR_UARTS) + return -EINVAL; + + port = &ports[co->index]; + + /* not initialized yet? */ + if (!port->membase) + return -ENODEV; + + if (options) + uart_parse_options(options, &baud, &parity, &bits, &flow); + + return uart_set_options(port, co, baud, parity, bits, flow); +} + +static struct uart_driver ulite_uart_driver; + +static struct console ulite_console = { + .name = "ttyUL", + .write = ulite_console_write, + .device = uart_console_device, + .setup = ulite_console_setup, + .flags = CON_PRINTBUFFER, + .index = -1, /* Specified on the cmdline (e.g. console=ttyUL0 ) */ + .data = &ulite_uart_driver, +}; + +static int __init ulite_console_init(void) +{ + register_console(&ulite_console); + return 0; +} + +console_initcall(ulite_console_init); + +#endif /* CONFIG_SERIAL_UARTLITE_CONSOLE */ + +static struct uart_driver ulite_uart_driver = { + .owner = THIS_MODULE, + .driver_name = "uartlite", + .dev_name = "ttyUL", + .major = ULITE_MAJOR, + .minor = ULITE_MINOR, + .nr = ULITE_NR_UARTS, +#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE + .cons = &ulite_console, +#endif +}; + +static int __devinit ulite_probe(struct platform_device *pdev) +{ + struct resource *res, *res2; + struct uart_port *port; + + if (pdev->id < 0 || pdev->id >= ULITE_NR_UARTS) + return -EINVAL; + + if (ports[pdev->id].membase) + return -EBUSY; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return -ENODEV; + + res2 = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + if (!res2) + return -ENODEV; + + port = &ports[pdev->id]; + + port->fifosize = 16; + port->regshift = 2; + port->iotype = UPIO_MEM; + port->iobase = 1; /* mark port in use */ + port->mapbase = res->start; + port->membase = 0; + port->ops = &ulite_ops; + port->irq = res2->start; + port->flags = UPF_BOOT_AUTOCONF; + port->dev = &pdev->dev; + port->type = PORT_UNKNOWN; + port->line = pdev->id; + + uart_add_one_port(&ulite_uart_driver, port); + platform_set_drvdata(pdev, port); + + return 0; +} + +static int ulite_remove(struct platform_device *pdev) +{ + struct uart_port *port = platform_get_drvdata(pdev); + + platform_set_drvdata(pdev, NULL); + + if (port) + uart_remove_one_port(&ulite_uart_driver, port); + + /* mark port as free */ + port->membase = 0; + + return 0; +} + +static struct platform_driver ulite_platform_driver = { + .probe = ulite_probe, + .remove = ulite_remove, + .driver = { + .owner = THIS_MODULE, + .name = "uartlite", + }, +}; + +int __init ulite_init(void) +{ + int ret; + + ret = uart_register_driver(&ulite_uart_driver); + if (ret) + return ret; + + ret = platform_driver_register(&ulite_platform_driver); + if (ret) + uart_unregister_driver(&ulite_uart_driver); + + return ret; +} + +void __exit ulite_exit(void) +{ + platform_driver_unregister(&ulite_platform_driver); + uart_unregister_driver(&ulite_uart_driver); +} + +module_init(ulite_init); +module_exit(ulite_exit); + +MODULE_AUTHOR("Peter Korsgaard "); +MODULE_DESCRIPTION("Xilinx uartlite serial driver"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 463ab953b092..827672136646 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -132,6 +132,8 @@ #define PORT_S3C2412 73 +/* Xilinx uartlite */ +#define PORT_UARTLITE 74 #ifdef __KERNEL__ -- cgit v1.2.3 From 48ed214d10ae3c3999af938970f7b5b58df77be3 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 6 Dec 2006 20:35:37 -0800 Subject: [PATCH] constify inode accessors Change the signature of i_size_read(), IMINOR() and IMAJOR() because they, or the functions they call, will never modify the argument. Signed-off-by: Jan Engelhardt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 94b831b8157c..d791bae9de9c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -636,7 +636,7 @@ extern void inode_double_unlock(struct inode *inode1, struct inode *inode2); * cmpxchg8b without the need of the lock prefix). For SMP compiles * and 64bit archs it makes no difference if preempt is enabled or not. */ -static inline loff_t i_size_read(struct inode *inode) +static inline loff_t i_size_read(const struct inode *inode) { #if BITS_PER_LONG==32 && defined(CONFIG_SMP) loff_t i_size; @@ -679,12 +679,12 @@ static inline void i_size_write(struct inode *inode, loff_t i_size) #endif } -static inline unsigned iminor(struct inode *inode) +static inline unsigned iminor(const struct inode *inode) { return MINOR(inode->i_rdev); } -static inline unsigned imajor(struct inode *inode) +static inline unsigned imajor(const struct inode *inode) { return MAJOR(inode->i_rdev); } -- cgit v1.2.3 From e9168c189fd54171124b5d25644024d99869e6a8 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 6 Dec 2006 20:35:38 -0800 Subject: [PATCH] fuse: update userspace interface to version 7.8 Add a flag to the RELEASE message which specifies that a FLUSH operation should be performed as well. This interface update is needed for the FreeBSD port, and doesn't actually touch the Linux implementation at all. Also rename the unused 'flush_flags' in the FLUSH message to 'unused'. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fuse.h | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 9fc48a674b82..76336327a941 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -15,7 +15,7 @@ #define FUSE_KERNEL_VERSION 7 /** Minor version number of this interface */ -#define FUSE_KERNEL_MINOR_VERSION 7 +#define FUSE_KERNEL_MINOR_VERSION 8 /** The node ID of the root inode */ #define FUSE_ROOT_ID 1 @@ -92,6 +92,11 @@ struct fuse_file_lock { #define FUSE_ASYNC_READ (1 << 0) #define FUSE_POSIX_LOCKS (1 << 1) +/** + * Release flags + */ +#define FUSE_RELEASE_FLUSH (1 << 0) + enum fuse_opcode { FUSE_LOOKUP = 1, FUSE_FORGET = 2, /* no reply */ @@ -205,12 +210,13 @@ struct fuse_open_out { struct fuse_release_in { __u64 fh; __u32 flags; - __u32 padding; + __u32 release_flags; + __u64 lock_owner; }; struct fuse_flush_in { __u64 fh; - __u32 flush_flags; + __u32 unused; __u32 padding; __u64 lock_owner; }; -- cgit v1.2.3 From b2d2272fae1e1df26ec8f93a6d5baea891dcce37 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 6 Dec 2006 20:35:51 -0800 Subject: [PATCH] fuse: add bmap support Add support for the BMAP operation for block device based filesystems. This is needed to support swap-files and lilo. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/dir.c | 2 ++ fs/fuse/file.c | 37 +++++++++++++++++++++++++++++++++++++ fs/fuse/fuse_i.h | 3 +++ include/linux/fuse.h | 11 +++++++++++ 4 files changed, 53 insertions(+) (limited to 'include/linux') diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 677f3ed7f987..1cabdb229adb 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1024,6 +1024,8 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) if (attr->ia_valid & ATTR_SIZE) { unsigned long limit; is_truncate = 1; + if (IS_SWAPFILE(inode)) + return -ETXTBSY; limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; if (limit != RLIM_INFINITY && attr->ia_size > (loff_t) limit) { send_sig(SIGXFSZ, current, 0); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 763a50daf1c0..128f79c40803 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -754,6 +754,42 @@ static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl) return err; } +static sector_t fuse_bmap(struct address_space *mapping, sector_t block) +{ + struct inode *inode = mapping->host; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_req *req; + struct fuse_bmap_in inarg; + struct fuse_bmap_out outarg; + int err; + + if (!inode->i_sb->s_bdev || fc->no_bmap) + return 0; + + req = fuse_get_req(fc); + if (IS_ERR(req)) + return 0; + + memset(&inarg, 0, sizeof(inarg)); + inarg.block = block; + inarg.blocksize = inode->i_sb->s_blocksize; + req->in.h.opcode = FUSE_BMAP; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + req->out.numargs = 1; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + request_send(fc, req); + err = req->out.h.error; + fuse_put_request(fc, req); + if (err == -ENOSYS) + fc->no_bmap = 1; + + return err ? 0 : outarg.block; +} + static const struct file_operations fuse_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -787,6 +823,7 @@ static const struct address_space_operations fuse_file_aops = { .commit_write = fuse_commit_write, .readpages = fuse_readpages, .set_page_dirty = fuse_set_page_dirty, + .bmap = fuse_bmap, }; void fuse_init_file_inode(struct inode *inode) diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 91edb8932d90..58d482d9f6bb 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -339,6 +339,9 @@ struct fuse_conn { /** Is interrupt not implemented by fs? */ unsigned no_interrupt : 1; + /** Is bmap not implemented by fs? */ + unsigned no_bmap : 1; + /** The number of requests waiting for completion */ atomic_t num_waiting; diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 76336327a941..162a754f4db7 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -132,6 +132,7 @@ enum fuse_opcode { FUSE_ACCESS = 34, FUSE_CREATE = 35, FUSE_INTERRUPT = 36, + FUSE_BMAP = 37, }; /* The read buffer is required to be at least 8k, but may be much larger */ @@ -302,6 +303,16 @@ struct fuse_interrupt_in { __u64 unique; }; +struct fuse_bmap_in { + __u64 block; + __u32 blocksize; + __u32 padding; +}; + +struct fuse_bmap_out { + __u64 block; +}; + struct fuse_in_header { __u32 len; __u32 opcode; -- cgit v1.2.3 From 0ec7ca41f6f0f74a394a7d686bc0ee8afef84887 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 6 Dec 2006 20:35:52 -0800 Subject: [PATCH] fuse: add DESTROY operation Add a DESTROY operation for block device based filesystems. With the help of this operation, such a filesystem can flush dirty data to the device synchronously before the umount returns. This is needed in situations where the filesystem is assumed to be clean immediately after unmount (e.g. ejecting removable media). Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/fuse_i.h | 6 ++++++ fs/fuse/inode.c | 22 ++++++++++++++++++++++ include/linux/fuse.h | 1 + 3 files changed, 29 insertions(+) (limited to 'include/linux') diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 58d482d9f6bb..b98b20de7405 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -298,6 +298,9 @@ struct fuse_conn { reply, before any other request, and never cleared */ unsigned conn_error : 1; + /** Connection successful. Only set in INIT */ + unsigned conn_init : 1; + /** Do readpages asynchronously? Only set in INIT */ unsigned async_read : 1; @@ -368,6 +371,9 @@ struct fuse_conn { /** Key for lock owner ID scrambling */ u32 scramble_key[4]; + + /** Reserved request for the DESTROY message */ + struct fuse_req *destroy_req; }; static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb) diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 1baaaeb2e850..437d61c65268 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -206,10 +206,23 @@ static void fuse_umount_begin(struct vfsmount *vfsmnt, int flags) fuse_abort_conn(get_fuse_conn_super(vfsmnt->mnt_sb)); } +static void fuse_send_destroy(struct fuse_conn *fc) +{ + struct fuse_req *req = fc->destroy_req; + if (req && fc->conn_init) { + fc->destroy_req = NULL; + req->in.h.opcode = FUSE_DESTROY; + req->force = 1; + request_send(fc, req); + fuse_put_request(fc, req); + } +} + static void fuse_put_super(struct super_block *sb) { struct fuse_conn *fc = get_fuse_conn_super(sb); + fuse_send_destroy(fc); spin_lock(&fc->lock); fc->connected = 0; fc->blocked = 0; @@ -410,6 +423,8 @@ static struct fuse_conn *new_conn(void) void fuse_conn_put(struct fuse_conn *fc) { if (atomic_dec_and_test(&fc->count)) { + if (fc->destroy_req) + fuse_request_free(fc->destroy_req); mutex_destroy(&fc->inst_mutex); kfree(fc); } @@ -466,6 +481,7 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->bdi.ra_pages = min(fc->bdi.ra_pages, ra_pages); fc->minor = arg->minor; fc->max_write = arg->minor < 5 ? 4096 : arg->max_write; + fc->conn_init = 1; } fuse_put_request(fc, req); fc->blocked = 0; @@ -563,6 +579,12 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (!init_req) goto err_put_root; + if (is_bdev) { + fc->destroy_req = fuse_request_alloc(); + if (!fc->destroy_req) + goto err_put_root; + } + mutex_lock(&fuse_mutex); err = -EINVAL; if (file->private_data) diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 162a754f4db7..534744efe30d 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -133,6 +133,7 @@ enum fuse_opcode { FUSE_CREATE = 35, FUSE_INTERRUPT = 36, FUSE_BMAP = 37, + FUSE_DESTROY = 38, }; /* The read buffer is required to be at least 8k, but may be much larger */ -- cgit v1.2.3 From e59e2ae2c29700117a54e85c106017c24837119f Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:35:59 -0800 Subject: [PATCH] SysRq-X: show blocked tasks Add SysRq-X support: show blocked (TASK_UNINTERRUPTIBLE) tasks only. Useful for debugging IO stalls. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/sysrq.c | 14 +++++++++++++- include/linux/sched.h | 11 ++++++++++- kernel/sched.c | 11 ++++++++--- 3 files changed, 31 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index c64f5bcff947..05810c8d20bc 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -182,6 +182,18 @@ static struct sysrq_key_op sysrq_showstate_op = { .enable_mask = SYSRQ_ENABLE_DUMP, }; +static void sysrq_handle_showstate_blocked(int key, struct tty_struct *tty) +{ + show_state_filter(TASK_UNINTERRUPTIBLE); +} +static struct sysrq_key_op sysrq_showstate_blocked_op = { + .handler = sysrq_handle_showstate_blocked, + .help_msg = "showBlockedTasks", + .action_msg = "Show Blocked State", + .enable_mask = SYSRQ_ENABLE_DUMP, +}; + + static void sysrq_handle_showmem(int key, struct tty_struct *tty) { show_mem(); @@ -304,7 +316,7 @@ static struct sysrq_key_op *sysrq_key_table[36] = { /* May be assigned at init time by SMP VOYAGER */ NULL, /* v */ NULL, /* w */ - NULL, /* x */ + &sysrq_showstate_blocked_op, /* x */ NULL, /* y */ NULL /* z */ }; diff --git a/include/linux/sched.h b/include/linux/sched.h index 837a012f573c..0a90cefb0b0d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -194,7 +194,16 @@ extern void init_idle(struct task_struct *idle, int cpu); extern cpumask_t nohz_cpu_mask; -extern void show_state(void); +/* + * Only dump TASK_* tasks. (-1 for all tasks) + */ +extern void show_state_filter(unsigned long state_filter); + +static inline void show_state(void) +{ + show_state_filter(-1); +} + extern void show_regs(struct pt_regs *); /* diff --git a/kernel/sched.c b/kernel/sched.c index 12fdbef1d9bf..1848e280504d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4804,7 +4804,7 @@ static void show_task(struct task_struct *p) show_stack(p, NULL); } -void show_state(void) +void show_state_filter(unsigned long state_filter) { struct task_struct *g, *p; @@ -4824,11 +4824,16 @@ void show_state(void) * console might take alot of time: */ touch_nmi_watchdog(); - show_task(p); + if (p->state & state_filter) + show_task(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); - debug_show_all_locks(); + /* + * Only show locks if all tasks are dumped: + */ + if (state_filter == -1) + debug_show_all_locks(); } /** -- cgit v1.2.3 From 5ec68b2e310437e99c297ba04e1afc5297aa6de1 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 6 Dec 2006 20:36:14 -0800 Subject: [PATCH] pull in necessary header files for cdev.h linux/cdev.h uses struct kobject and other structs and should therefore include them. Currently, a module either needs to add the missing includes itself, or, in case a module includes other headers already, needs to put last, which goes against a alphabetically-sorted include list. Signed-off-by: Jan Engelhardt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cdev.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/cdev.h b/include/linux/cdev.h index ee5f53f2ca15..f309b00e986e 100644 --- a/include/linux/cdev.h +++ b/include/linux/cdev.h @@ -2,6 +2,10 @@ #define _LINUX_CDEV_H #ifdef __KERNEL__ +#include +#include +#include + struct cdev { struct kobject kobj; struct module *owner; -- cgit v1.2.3 From c140e110019f25ffa1c6f3f365b0c9103d0b8475 Mon Sep 17 00:00:00 2001 From: Ryan Underwood Date: Wed, 6 Dec 2006 20:36:38 -0800 Subject: [PATCH] parport_pc: Add support for OX16PCI952 parallel port Add support for the parallel port (implemented as separate PCI function) on the Oxford Semiconductor OX16PCI952. Signed-off-by: Ryan Underwood Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/parport/parport_pc.c | 4 ++++ include/linux/pci_ids.h | 1 + 2 files changed, 5 insertions(+) (limited to 'include/linux') diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c index 39c96641bc72..5749500f45f0 100644 --- a/drivers/parport/parport_pc.c +++ b/drivers/parport/parport_pc.c @@ -2747,6 +2747,7 @@ enum parport_pc_pci_cards { titan_1284p2, avlab_1p, avlab_2p, + oxsemi_952, oxsemi_954, oxsemi_840, aks_0100, @@ -2822,6 +2823,7 @@ static struct parport_pc_pci { /* avlab_2p */ { 2, { { 0, 1}, { 2, 3 },} }, /* The Oxford Semi cards are unusual: 954 doesn't support ECP, * and 840 locks up if you write 1 to bit 2! */ + /* oxsemi_952 */ { 1, { { 0, 1 }, } }, /* oxsemi_954 */ { 1, { { 0, -1 }, } }, /* oxsemi_840 */ { 1, { { 0, -1 }, } }, /* aks_0100 */ { 1, { { 0, -1 }, } }, @@ -2895,6 +2897,8 @@ static const struct pci_device_id parport_pc_pci_tbl[] = { /* PCI_VENDOR_ID_AVLAB/Intek21 has another bunch of cards ...*/ { 0x14db, 0x2120, PCI_ANY_ID, PCI_ANY_ID, 0, 0, avlab_1p}, /* AFAVLAB_TK9902 */ { 0x14db, 0x2121, PCI_ANY_ID, PCI_ANY_ID, 0, 0, avlab_2p}, + { PCI_VENDOR_ID_OXSEMI, PCI_DEVICE_ID_OXSEMI_16PCI952PP, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, oxsemi_952 }, { PCI_VENDOR_ID_OXSEMI, PCI_DEVICE_ID_OXSEMI_16PCI954PP, PCI_ANY_ID, PCI_ANY_ID, 0, 0, oxsemi_954 }, { PCI_VENDOR_ID_OXSEMI, PCI_DEVICE_ID_OXSEMI_12PCI840, diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index c09da1e30c54..dcdb90f06d73 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1864,6 +1864,7 @@ #define PCI_DEVICE_ID_OXSEMI_16PCI95N 0x9511 #define PCI_DEVICE_ID_OXSEMI_16PCI954PP 0x9513 #define PCI_DEVICE_ID_OXSEMI_16PCI952 0x9521 +#define PCI_DEVICE_ID_OXSEMI_16PCI952PP 0x9523 #define PCI_VENDOR_ID_SAMSUNG 0x144d -- cgit v1.2.3 From 20aa7b21b1cbd1aa3fbf5fc14da5f7484a61a824 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:36:40 -0800 Subject: [PATCH] probe_kernel_address() needs to do set_fs() probe_kernel_address() purports to be generic, only it forgot to select KERNEL_DS, so it presently won't work right on all architectures. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/uaccess.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index 67918c22339c..76c3fe325101 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -65,14 +65,22 @@ static inline unsigned long __copy_from_user_nocache(void *to, * do_page_fault() doesn't attempt to take mmap_sem. This makes * probe_kernel_address() suitable for use within regions where the caller * already holds mmap_sem, or other locks which nest inside mmap_sem. + * This must be a macro because __get_user() needs to know the types of the + * args. + * + * We don't include enough header files to be able to do the set_fs(). We + * require that the probe_kernel_address() caller will do that. */ #define probe_kernel_address(addr, retval) \ ({ \ long ret; \ + mm_segment_t old_fs = get_fs(); \ \ + set_fs(KERNEL_DS); \ pagefault_disable(); \ ret = __get_user(retval, addr); \ pagefault_enable(); \ + set_fs(old_fs); \ ret; \ }) -- cgit v1.2.3 From 115085ea0794c0f339be8f9d25505c7f9861d824 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 6 Dec 2006 20:36:51 -0800 Subject: [PATCH] taskstats: cleanup do_exit() path do_exit: taskstats_exit_alloc() ... taskstats_exit_send() taskstats_exit_free() I think this is not good, let it be a single function exported to the core kernel, taskstats_exit(), which does alloc + send + free itself. Signed-off-by: Oleg Nesterov Cc: Balbir Singh Cc: Shailabh Nagar Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats_kern.h | 17 ++--------------- kernel/exit.c | 8 ++------ kernel/taskstats.c | 41 +++++++++++++++-------------------------- 3 files changed, 19 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index ce8a912e5426..f1261a532496 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -15,12 +15,6 @@ extern struct kmem_cache *taskstats_cache; extern struct mutex taskstats_exit_mutex; -static inline void taskstats_exit_free(struct taskstats *tidstats) -{ - if (tidstats) - kmem_cache_free(taskstats_cache, tidstats); -} - static inline void taskstats_tgid_init(struct signal_struct *sig) { sig->stats = NULL; @@ -54,17 +48,10 @@ static inline void taskstats_tgid_free(struct signal_struct *sig) kmem_cache_free(taskstats_cache, sig->stats); } -extern void taskstats_exit_alloc(struct taskstats **, unsigned int *); -extern void taskstats_exit_send(struct task_struct *, struct taskstats *, int, unsigned int); +extern void taskstats_exit(struct task_struct *, int group_dead); extern void taskstats_init_early(void); #else -static inline void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) -{} -static inline void taskstats_exit_free(struct taskstats *ptidstats) -{} -static inline void taskstats_exit_send(struct task_struct *tsk, - struct taskstats *tidstats, - int group_dead, unsigned int cpu) +static inline void taskstats_exit(struct task_struct *tsk, int group_dead) {} static inline void taskstats_tgid_init(struct signal_struct *sig) {} diff --git a/kernel/exit.c b/kernel/exit.c index 06de6c4e8ca3..4e3f919edc48 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -850,9 +850,7 @@ static void exit_notify(struct task_struct *tsk) fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; - struct taskstats *tidstats; int group_dead; - unsigned int mycpu; profile_task_exit(tsk); @@ -890,8 +888,6 @@ fastcall NORET_TYPE void do_exit(long code) current->comm, current->pid, preempt_count()); - taskstats_exit_alloc(&tidstats, &mycpu); - acct_update_integrals(tsk); if (tsk->mm) { update_hiwater_rss(tsk->mm); @@ -911,8 +907,8 @@ fastcall NORET_TYPE void do_exit(long code) #endif if (unlikely(tsk->audit_context)) audit_free(tsk); - taskstats_exit_send(tsk, tidstats, group_dead, mycpu); - taskstats_exit_free(tidstats); + + taskstats_exit(tsk, group_dead); exit_mm(tsk); diff --git a/kernel/taskstats.c b/kernel/taskstats.c index d9d7c3576238..2654886fe058 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -119,10 +119,10 @@ static int send_reply(struct sk_buff *skb, pid_t pid) /* * Send taskstats data in @skb to listeners registered for @cpu's exit data */ -static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) +static void send_cpu_listeners(struct sk_buff *skb, + struct listener_list *listeners) { struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data); - struct listener_list *listeners; struct listener *s, *tmp; struct sk_buff *skb_next, *skb_cur = skb; void *reply = genlmsg_data(genlhdr); @@ -135,7 +135,6 @@ static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) } rc = 0; - listeners = &per_cpu(listener_array, cpu); down_read(&listeners->sem); list_for_each_entry(s, &listeners->list, list) { skb_next = NULL; @@ -413,28 +412,12 @@ err: return rc; } -void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) -{ - struct listener_list *listeners; - /* - * This is the cpu on which the task is exiting currently and will - * be the one for which the exit event is sent, even if the cpu - * on which this function is running changes later. - */ - *mycpu = raw_smp_processor_id(); - - listeners = &per_cpu(listener_array, *mycpu); - - *ptidstats = NULL; - if (!list_empty(&listeners->list)) - *ptidstats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); -} - /* Send pid data out on exit */ -void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, - int group_dead, unsigned int mycpu) +void taskstats_exit(struct task_struct *tsk, int group_dead) { int rc; + struct listener_list *listeners; + struct taskstats *tidstats; struct sk_buff *rep_skb; void *reply; size_t size; @@ -458,12 +441,17 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, fill_tgid_exit(tsk); } + listeners = &__raw_get_cpu_var(listener_array); + if (list_empty(&listeners->list)) + return; + + tidstats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); if (!tidstats) return; rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); if (rc < 0) - goto ret; + goto free_stats; rc = fill_pid(tsk->pid, tsk, tidstats); if (rc < 0) @@ -492,15 +480,16 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, nla_nest_end(rep_skb, na); send: - send_cpu_listeners(rep_skb, mycpu); + send_cpu_listeners(rep_skb, listeners); +free_stats: + kmem_cache_free(taskstats_cache, tidstats); return; nla_put_failure: genlmsg_cancel(rep_skb, reply); err_skb: nlmsg_free(rep_skb); -ret: - return; + goto free_stats; } static struct genl_ops taskstats_ops = { -- cgit v1.2.3 From 34ec12349c8a9505adc59d72f92b4595bc2483ff Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 6 Dec 2006 20:36:52 -0800 Subject: [PATCH] taskstats: cleanup ->signal->stats allocation Allocate ->signal->stats on demand in taskstats_exit(), this allows us to remove taskstats_tgid_alloc() (the last non-trivial inline) from taskstat's public interface. Signed-off-by: Oleg Nesterov Cc: Balbir Singh Cc: Shailabh Nagar Cc: Jay Lan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats_kern.h | 24 ------------------------ kernel/fork.c | 1 - kernel/taskstats.c | 26 +++++++++++++++++++++++++- 3 files changed, 25 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/include/linux/taskstats_kern.h b/include/linux/taskstats_kern.h index f1261a532496..7e9680f4afdd 100644 --- a/include/linux/taskstats_kern.h +++ b/include/linux/taskstats_kern.h @@ -20,28 +20,6 @@ static inline void taskstats_tgid_init(struct signal_struct *sig) sig->stats = NULL; } -static inline void taskstats_tgid_alloc(struct task_struct *tsk) -{ - struct signal_struct *sig = tsk->signal; - struct taskstats *stats; - - if (sig->stats != NULL) - return; - - /* No problem if kmem_cache_zalloc() fails */ - stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); - - spin_lock_irq(&tsk->sighand->siglock); - if (!sig->stats) { - sig->stats = stats; - stats = NULL; - } - spin_unlock_irq(&tsk->sighand->siglock); - - if (stats) - kmem_cache_free(taskstats_cache, stats); -} - static inline void taskstats_tgid_free(struct signal_struct *sig) { if (sig->stats) @@ -55,8 +33,6 @@ static inline void taskstats_exit(struct task_struct *tsk, int group_dead) {} static inline void taskstats_tgid_init(struct signal_struct *sig) {} -static inline void taskstats_tgid_alloc(struct task_struct *tsk) -{} static inline void taskstats_tgid_free(struct signal_struct *sig) {} static inline void taskstats_init_early(void) diff --git a/kernel/fork.c b/kernel/fork.c index f37980df1d58..658838148647 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -847,7 +847,6 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts if (clone_flags & CLONE_THREAD) { atomic_inc(¤t->signal->count); atomic_inc(¤t->signal->live); - taskstats_tgid_alloc(current); return 0; } sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 2654886fe058..7d793d6b1e90 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -412,6 +412,30 @@ err: return rc; } +static struct taskstats *taskstats_tgid_alloc(struct task_struct *tsk) +{ + struct signal_struct *sig = tsk->signal; + struct taskstats *stats; + + if (sig->stats || thread_group_empty(tsk)) + goto ret; + + /* No problem if kmem_cache_zalloc() fails */ + stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); + + spin_lock_irq(&tsk->sighand->siglock); + if (!sig->stats) { + sig->stats = stats; + stats = NULL; + } + spin_unlock_irq(&tsk->sighand->siglock); + + if (stats) + kmem_cache_free(taskstats_cache, stats); +ret: + return sig->stats; +} + /* Send pid data out on exit */ void taskstats_exit(struct task_struct *tsk, int group_dead) { @@ -433,7 +457,7 @@ void taskstats_exit(struct task_struct *tsk, int group_dead) size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); - is_thread_group = (tsk->signal->stats != NULL); + is_thread_group = !!taskstats_tgid_alloc(tsk); if (is_thread_group) { /* PID + STATS + TGID + STATS */ size = 2 * size; -- cgit v1.2.3 From 9774a1f54f173ad18e816496c8979f1bf8ef666a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 6 Dec 2006 20:36:56 -0800 Subject: [PATCH] Compile-time check re world-writeable module params One of the mistakes a module_param() user can make is to supply default value of module parameter as the last argument. module_param() accepts permissions instead. If default value is, say, 3 (-------wx), parameter becomes world-writeable. So far, the only remedy was to apply grep(1) and read drivers submitted to -mm. BTDT. With this patch applied, compiler will finally do some job. *) bounds checking on permissions *) world-writeable bit checking on permissions *) compile breakage if checks trigger First version of this check (only "& 2" part) directly caught 4 out of 7 places during my last grep. Subject: Neverending module_param() bugs [X] drivers/acpi/sbs.c:101:module_param(capacity_mode, int, CAPACITY_UNIT); [X] drivers/acpi/sbs.c:102:module_param(update_mode, int, UPDATE_MODE); [ ] drivers/acpi/sbs.c:103:module_param(update_info_mode, int, UPDATE_INFO_MODE); [ ] drivers/acpi/sbs.c:104:module_param(update_time, int, UPDATE_TIME); [ ] drivers/acpi/sbs.c:105:module_param(update_time2, int, UPDATE_TIME2); [X] drivers/char/watchdog/sbc8360.c:203:module_param(timeout, int, 27); [X] drivers/media/video/tuner-simple.c:13:module_param(offset, int, 0666); Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/moduleparam.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/moduleparam.h b/include/linux/moduleparam.h index 7c0c2c198f1f..4a189dadb160 100644 --- a/include/linux/moduleparam.h +++ b/include/linux/moduleparam.h @@ -63,6 +63,9 @@ struct kparam_array not there, read bits mean it's readable, write bits mean it's writable. */ #define __module_param_call(prefix, name, set, get, arg, perm) \ + /* Default value instead of permissions? */ \ + static int __param_perm_check_##name __attribute__((unused)) = \ + BUILD_BUG_ON_ZERO((perm) < 0 || (perm) > 0777 || ((perm) & 2)); \ static char __param_str_##name[] = prefix #name; \ static struct kernel_param const __param_##name \ __attribute_used__ \ -- cgit v1.2.3 From e0980dafa329d33bb88edc8a3ef9fab4e070590c Mon Sep 17 00:00:00 2001 From: Paul B Schroeder Date: Wed, 6 Dec 2006 20:37:03 -0800 Subject: [PATCH] Exar quad port serial This is on our "Envoy" boxes which we have, according to the documentation, an "Exar ST16C554/554D Quad UART with 16-byte Fifo's". The box also has two other "on-board" serial ports and a modem chip. The two on-board serial UARTs were being detected along with the first two Exar UARTs. The last two Exar UARTs were not showing up and neither was the modem. This patch was the only way I could the kernel to see beyond the standard four serial ports and get all four of the Exar UARTs to show up. [akpm@osdl.org: build fix] Signed-off-by: Paul B Schroeder Cc: Lennart Sorensen Acked-by: Alan Cox Cc: Russell King Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250_exar_st16c554.c | 52 +++++++++++++++++++++++++++++++++++++ drivers/serial/Kconfig | 11 ++++++++ drivers/serial/Makefile | 1 + include/linux/serial_8250.h | 1 + 4 files changed, 65 insertions(+) create mode 100644 drivers/serial/8250_exar_st16c554.c (limited to 'include/linux') diff --git a/drivers/serial/8250_exar_st16c554.c b/drivers/serial/8250_exar_st16c554.c new file mode 100644 index 000000000000..567143ace159 --- /dev/null +++ b/drivers/serial/8250_exar_st16c554.c @@ -0,0 +1,52 @@ +/* + * linux/drivers/serial/8250_exar.c + * + * Written by Paul B Schroeder < pschroeder "at" uplogix "dot" com > + * Based on 8250_boca. + * + * Copyright (C) 2005 Russell King. + * Data taken from include/asm-i386/serial.h + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include + +#define PORT(_base,_irq) \ + { \ + .iobase = _base, \ + .irq = _irq, \ + .uartclk = 1843200, \ + .iotype = UPIO_PORT, \ + .flags = UPF_BOOT_AUTOCONF, \ + } + +static struct plat_serial8250_port exar_data[] = { + PORT(0x100, 5), + PORT(0x108, 5), + PORT(0x110, 5), + PORT(0x118, 5), + { }, +}; + +static struct platform_device exar_device = { + .name = "serial8250", + .id = PLAT8250_DEV_EXAR_ST16C554, + .dev = { + .platform_data = exar_data, + }, +}; + +static int __init exar_init(void) +{ + return platform_device_register(&exar_device); +} + +module_init(exar_init); + +MODULE_AUTHOR("Paul B Schroeder"); +MODULE_DESCRIPTION("8250 serial probe module for Exar cards"); +MODULE_LICENSE("GPL"); diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index e936c91640b4..fc12d5df10e2 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -210,6 +210,17 @@ config SERIAL_8250_BOCA To compile this driver as a module, choose M here: the module will be called 8250_boca. +config SERIAL_8250_EXAR_ST16C554 + tristate "Support Exar ST16C554/554D Quad UART" + depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS + help + The Uplogix Envoy TU301 uses this Exar Quad UART. If you are + tinkering with your Envoy TU301, or have a machine with this UART, + say Y here. + + To compile this driver as a module, choose M here: the module + will be called 8250_exar_st16c554. + config SERIAL_8250_HUB6 tristate "Support Hub6 cards" depends on SERIAL_8250 != n && ISA && SERIAL_8250_MANY_PORTS diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile index 0dba0017a006..df3632cd7df9 100644 --- a/drivers/serial/Makefile +++ b/drivers/serial/Makefile @@ -17,6 +17,7 @@ obj-$(CONFIG_SERIAL_8250_CONSOLE) += 8250_early.o obj-$(CONFIG_SERIAL_8250_FOURPORT) += 8250_fourport.o obj-$(CONFIG_SERIAL_8250_ACCENT) += 8250_accent.o obj-$(CONFIG_SERIAL_8250_BOCA) += 8250_boca.o +obj-$(CONFIG_SERIAL_8250_EXAR_ST16C554) += 8250_exar_st16c554.o obj-$(CONFIG_SERIAL_8250_HUB6) += 8250_hub6.o obj-$(CONFIG_SERIAL_8250_MCA) += 8250_mca.o obj-$(CONFIG_SERIAL_8250_AU1X00) += 8250_au1x00.o diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h index 8e9681413726..71310d80c09a 100644 --- a/include/linux/serial_8250.h +++ b/include/linux/serial_8250.h @@ -41,6 +41,7 @@ enum { PLAT8250_DEV_FOURPORT, PLAT8250_DEV_ACCENT, PLAT8250_DEV_BOCA, + PLAT8250_DEV_EXAR_ST16C554, PLAT8250_DEV_HUB6, PLAT8250_DEV_MCA, PLAT8250_DEV_AU1X00, -- cgit v1.2.3 From 3a229b39eb8497ae5f8077f81f7c8c3e1aacd624 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:37:14 -0800 Subject: [PATCH] ext3: uninline large functions Saves nearly 4kbytes on x86. Cc: Arnaldo Carvalho de Melo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext3/Makefile | 2 +- fs/ext3/ext3_jbd.c | 59 +++++++++++++++++++++++++++++++++++++ include/linux/ext3_jbd.h | 76 +++++++++++------------------------------------- 3 files changed, 77 insertions(+), 60 deletions(-) create mode 100644 fs/ext3/ext3_jbd.c (limited to 'include/linux') diff --git a/fs/ext3/Makefile b/fs/ext3/Makefile index 704cd44a40c2..e77766a8b3f0 100644 --- a/fs/ext3/Makefile +++ b/fs/ext3/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_EXT3_FS) += ext3.o ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ - ioctl.o namei.o super.o symlink.o hash.o resize.o + ioctl.o namei.o super.o symlink.o hash.o resize.o ext3_jbd.o ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o diff --git a/fs/ext3/ext3_jbd.c b/fs/ext3/ext3_jbd.c new file mode 100644 index 000000000000..e1f91fd26a93 --- /dev/null +++ b/fs/ext3/ext3_jbd.c @@ -0,0 +1,59 @@ +/* + * Interface between ext3 and JBD + */ + +#include + +int __ext3_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_undo_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_write_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_forget(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_revoke(const char *where, handle_t *handle, + unsigned long blocknr, struct buffer_head *bh) +{ + int err = journal_revoke(handle, blocknr, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_get_create_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_dirty_metadata(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h index ce0e6109aff0..8c43b13a02fe 100644 --- a/include/linux/ext3_jbd.h +++ b/include/linux/ext3_jbd.h @@ -109,74 +109,32 @@ int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode); * been done yet. */ -void ext3_journal_abort_handle(const char *caller, const char *err_fn, - struct buffer_head *bh, handle_t *handle, int err); - -static inline int -__ext3_journal_get_undo_access(const char *where, handle_t *handle, - struct buffer_head *bh) +static inline void ext3_journal_release_buffer(handle_t *handle, + struct buffer_head *bh) { - int err = journal_get_undo_access(handle, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; + journal_release_buffer(handle, bh); } -static inline int -__ext3_journal_get_write_access(const char *where, handle_t *handle, - struct buffer_head *bh) -{ - int err = journal_get_write_access(handle, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +void ext3_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err); -static inline void -ext3_journal_release_buffer(handle_t *handle, struct buffer_head *bh) -{ - journal_release_buffer(handle, bh); -} +int __ext3_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext3_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) -{ - int err = journal_forget(handle, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext3_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext3_journal_revoke(const char *where, handle_t *handle, - unsigned long blocknr, struct buffer_head *bh) -{ - int err = journal_revoke(handle, blocknr, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext3_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext3_journal_get_create_access(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = journal_get_create_access(handle, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext3_journal_revoke(const char *where, handle_t *handle, + unsigned long blocknr, struct buffer_head *bh); -static inline int -__ext3_journal_dirty_metadata(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = journal_dirty_metadata(handle, bh); - if (err) - ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext3_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh); +int __ext3_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh); #define ext3_journal_get_undo_access(handle, bh) \ __ext3_journal_get_undo_access(__FUNCTION__, (handle), (bh)) -- cgit v1.2.3 From 8984d137df669a6e94dbce7b87095e4ce80b9e67 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:37:15 -0800 Subject: [PATCH] ext4: uninline large functions Saves nearly 4kbytes on x86. Cc: Arnaldo Carvalho de Melo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ext4/Makefile | 3 +- fs/ext4/ext4_jbd2.c | 59 ++++++++++++++++++++++++++++++++++++ include/linux/ext4_jbd2.h | 76 +++++++++++------------------------------------ 3 files changed, 78 insertions(+), 60 deletions(-) create mode 100644 fs/ext4/ext4_jbd2.c (limited to 'include/linux') diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index a6acb96ebeb9..ae6e7e502ac9 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -5,7 +5,8 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ - ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o + ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ + ext4_jbd2.o ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c new file mode 100644 index 000000000000..d6afe4e27340 --- /dev/null +++ b/fs/ext4/ext4_jbd2.c @@ -0,0 +1,59 @@ +/* + * Interface between ext4 and JBD + */ + +#include + +int __ext4_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_get_undo_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_get_write_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_forget(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_revoke(const char *where, handle_t *handle, + ext4_fsblk_t blocknr, struct buffer_head *bh) +{ + int err = jbd2_journal_revoke(handle, blocknr, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = jbd2_journal_get_create_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = jbd2_journal_dirty_metadata(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} diff --git a/include/linux/ext4_jbd2.h b/include/linux/ext4_jbd2.h index 72dd631912e4..d716e6392cf6 100644 --- a/include/linux/ext4_jbd2.h +++ b/include/linux/ext4_jbd2.h @@ -114,74 +114,32 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode); * been done yet. */ -void ext4_journal_abort_handle(const char *caller, const char *err_fn, - struct buffer_head *bh, handle_t *handle, int err); - -static inline int -__ext4_journal_get_undo_access(const char *where, handle_t *handle, - struct buffer_head *bh) +static inline void ext4_journal_release_buffer(handle_t *handle, + struct buffer_head *bh) { - int err = jbd2_journal_get_undo_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; + jbd2_journal_release_buffer(handle, bh); } -static inline int -__ext4_journal_get_write_access(const char *where, handle_t *handle, - struct buffer_head *bh) -{ - int err = jbd2_journal_get_write_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +void ext4_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err); -static inline void -ext4_journal_release_buffer(handle_t *handle, struct buffer_head *bh) -{ - jbd2_journal_release_buffer(handle, bh); -} +int __ext4_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext4_journal_forget(const char *where, handle_t *handle, struct buffer_head *bh) -{ - int err = jbd2_journal_forget(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext4_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext4_journal_revoke(const char *where, handle_t *handle, - ext4_fsblk_t blocknr, struct buffer_head *bh) -{ - int err = jbd2_journal_revoke(handle, blocknr, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext4_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh); -static inline int -__ext4_journal_get_create_access(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = jbd2_journal_get_create_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext4_journal_revoke(const char *where, handle_t *handle, + ext4_fsblk_t blocknr, struct buffer_head *bh); -static inline int -__ext4_journal_dirty_metadata(const char *where, - handle_t *handle, struct buffer_head *bh) -{ - int err = jbd2_journal_dirty_metadata(handle, bh); - if (err) - ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); - return err; -} +int __ext4_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh); +int __ext4_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh); #define ext4_journal_get_undo_access(handle, bh) \ __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) -- cgit v1.2.3 From 6cfd76a26d9fe2ba54b9d496a48c1d9285e5c5ed Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:37:22 -0800 Subject: [PATCH] lockdep: name some old style locks Name some of the remaning 'old_style_spin_init' locks Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/traps.c | 2 +- include/asm-i386/rwsem.h | 4 ++-- include/linux/init_task.h | 2 +- include/linux/mutex.h | 2 +- include/linux/rtmutex.h | 2 +- include/linux/rwsem-spinlock.h | 3 ++- include/linux/sunrpc/sched.h | 4 ++-- kernel/acct.c | 3 ++- kernel/irq/handle.c | 2 +- net/sunrpc/svcauth.c | 3 ++- security/keys/process_keys.c | 2 +- 11 files changed, 16 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index fe9c5e8e7e6f..3124f1b04d67 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -452,7 +452,7 @@ void die(const char * str, struct pt_regs * regs, long err) u32 lock_owner; int lock_owner_depth; } die = { - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(die.lock), .lock_owner = -1, .lock_owner_depth = 0 }; diff --git a/include/asm-i386/rwsem.h b/include/asm-i386/rwsem.h index bc598d6388e3..041906f3c6df 100644 --- a/include/asm-i386/rwsem.h +++ b/include/asm-i386/rwsem.h @@ -75,8 +75,8 @@ struct rw_semaphore { #define __RWSEM_INITIALIZER(name) \ -{ RWSEM_UNLOCKED_VALUE, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) \ - __RWSEM_DEP_MAP_INIT(name) } +{ RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait_lock), \ + LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) } #define DECLARE_RWSEM(name) \ struct rw_semaphore name = __RWSEM_INITIALIZER(name) diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 33c5daacc743..733790d4f7db 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -73,7 +73,7 @@ extern struct nsproxy init_nsproxy; #define INIT_NSPROXY(nsproxy) { \ .count = ATOMIC_INIT(1), \ - .nslock = SPIN_LOCK_UNLOCKED, \ + .nslock = __SPIN_LOCK_UNLOCKED(nsproxy.nslock), \ .uts_ns = &init_uts_ns, \ .namespace = NULL, \ INIT_IPC_NS(ipc_ns) \ diff --git a/include/linux/mutex.h b/include/linux/mutex.h index 27c48daa3183..b2b91c477563 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -94,7 +94,7 @@ do { \ #define __MUTEX_INITIALIZER(lockname) \ { .count = ATOMIC_INIT(1) \ - , .wait_lock = SPIN_LOCK_UNLOCKED \ + , .wait_lock = __SPIN_LOCK_UNLOCKED(lockname.wait_lock) \ , .wait_list = LIST_HEAD_INIT(lockname.wait_list) \ __DEBUG_MUTEX_INITIALIZER(lockname) \ __DEP_MAP_MUTEX_INITIALIZER(lockname) } diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index 5d41dee82f80..b0090e9f7884 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -63,7 +63,7 @@ struct hrtimer_sleeper; #endif #define __RT_MUTEX_INITIALIZER(mutexname) \ - { .wait_lock = SPIN_LOCK_UNLOCKED \ + { .wait_lock = __SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ , .wait_list = PLIST_HEAD_INIT(mutexname.wait_list, mutexname.wait_lock) \ , .owner = NULL \ __DEBUG_RT_MUTEX_INITIALIZER(mutexname)} diff --git a/include/linux/rwsem-spinlock.h b/include/linux/rwsem-spinlock.h index ae1fcadd598e..813cee13da0d 100644 --- a/include/linux/rwsem-spinlock.h +++ b/include/linux/rwsem-spinlock.h @@ -44,7 +44,8 @@ struct rw_semaphore { #endif #define __RWSEM_INITIALIZER(name) \ -{ 0, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) } +{ 0, __SPIN_LOCK_UNLOCKED(name.wait_lock), LIST_HEAD_INIT((name).wait_list) \ + __RWSEM_DEP_MAP_INIT(name) } #define DECLARE_RWSEM(name) \ struct rw_semaphore name = __RWSEM_INITIALIZER(name) diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index f399c138f79d..0746c3b16f3a 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -222,7 +222,7 @@ struct rpc_wait_queue { #ifndef RPC_DEBUG # define RPC_WAITQ_INIT(var,qname) { \ - .lock = SPIN_LOCK_UNLOCKED, \ + .lock = __SPIN_LOCK_UNLOCKED(var.lock), \ .tasks = { \ [0] = LIST_HEAD_INIT(var.tasks[0]), \ [1] = LIST_HEAD_INIT(var.tasks[1]), \ @@ -231,7 +231,7 @@ struct rpc_wait_queue { } #else # define RPC_WAITQ_INIT(var,qname) { \ - .lock = SPIN_LOCK_UNLOCKED, \ + .lock = __SPIN_LOCK_UNLOCKED(var.lock), \ .tasks = { \ [0] = LIST_HEAD_INIT(var.tasks[0]), \ [1] = LIST_HEAD_INIT(var.tasks[1]), \ diff --git a/kernel/acct.c b/kernel/acct.c index 0aad5ca36a81..dc12db8600e7 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -89,7 +89,8 @@ struct acct_glbs { struct timer_list timer; }; -static struct acct_glbs acct_globals __cacheline_aligned = {SPIN_LOCK_UNLOCKED}; +static struct acct_glbs acct_globals __cacheline_aligned = + {__SPIN_LOCK_UNLOCKED(acct_globals.lock)}; /* * Called whenever the timer says to check the free space. diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index a681912bc89a..aff1f0fabb0d 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -54,7 +54,7 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned = { .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), #ifdef CONFIG_SMP .affinity = CPU_MASK_ALL #endif diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index ee9bb1522d5e..c7bb5f7f21a5 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -119,7 +119,8 @@ EXPORT_SYMBOL(svc_auth_unregister); #define DN_HASHMASK (DN_HASHMAX-1) static struct hlist_head auth_domain_table[DN_HASHMAX]; -static spinlock_t auth_domain_lock = SPIN_LOCK_UNLOCKED; +static spinlock_t auth_domain_lock = + __SPIN_LOCK_UNLOCKED(auth_domain_lock); void auth_domain_put(struct auth_domain *dom) { diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 32150cf7c37f..b6f86808475a 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -27,7 +27,7 @@ static DEFINE_MUTEX(key_session_mutex); struct key_user root_key_user = { .usage = ATOMIC_INIT(3), .consq = LIST_HEAD_INIT(root_key_user.consq), - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(root_key_user.lock), .nkeys = ATOMIC_INIT(2), .nikeys = ATOMIC_INIT(2), .uid = 0, -- cgit v1.2.3 From ece8a684c75df215320b4155944979e3f78c5c93 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:37:24 -0800 Subject: [PATCH] sleep profiling Implement prof=sleep profiling. TASK_UNINTERRUPTIBLE sleeps will be taken as a profile hit, and every millisecond spent sleeping causes a profile-hit for the call site that initiated the sleep. Sample readprofile output on i386: 306 ps2_sendbyte 1.3973 432 call_usermodehelper_keys 1.9548 484 ps2_command 0.6453 790 __driver_attach 4.7879 1593 msleep 44.2500 3976 sync_buffer 64.1290 4076 do_lookup 12.4648 8587 sync_page 122.6714 20820 total 0.0067 (NOTE: architectures need to check whether get_wchan() can be called from deep within the wakeup path.) akpm: we need to mark more functions __sched. lock_sock(), msleep(), others.. akpm: the contention in do_lookup() is a surprise. Presumably doing disk reads for directory contents while holding i_mutex. [akpm@osdl.org: various fixes] Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 1 + include/linux/profile.h | 24 ++++++++++++++++++++++- kernel/profile.c | 39 ++++++++++++++++++++++++++++--------- kernel/sched.c | 11 +++++++++++ 4 files changed, 65 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 8fe6b834ef27..2a40d9f6ffad 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1294,6 +1294,7 @@ and is between 256 and 4096 characters. It is defined in the file Param: "schedule" - profile schedule points. Param: - step/bucket size as a power of 2 for statistical time based profiling. + Param: "sleep" - profile D-state sleeping (millisecs) processor.max_cstate= [HW,ACPI] Limit processor to maximum C-state diff --git a/include/linux/profile.h b/include/linux/profile.h index acce53fd38b6..5670b340c4ef 100644 --- a/include/linux/profile.h +++ b/include/linux/profile.h @@ -6,10 +6,15 @@ #include #include #include +#include + #include +extern int prof_on __read_mostly; + #define CPU_PROFILING 1 #define SCHED_PROFILING 2 +#define SLEEP_PROFILING 3 struct proc_dir_entry; struct pt_regs; @@ -18,7 +23,24 @@ struct notifier_block; /* init basic kernel profiler */ void __init profile_init(void); void profile_tick(int); -void profile_hit(int, void *); + +/* + * Add multiple profiler hits to a given address: + */ +void profile_hits(int, void *ip, unsigned int nr_hits); + +/* + * Single profiler hit: + */ +static inline void profile_hit(int type, void *ip) +{ + /* + * Speedup for the common (no profiling enabled) case: + */ + if (unlikely(prof_on == type)) + profile_hits(type, ip, 1); +} + #ifdef CONFIG_PROC_FS void create_prof_cpu_mask(struct proc_dir_entry *); #else diff --git a/kernel/profile.c b/kernel/profile.c index 15b012df4ff1..04fd84e8cdbe 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -40,7 +40,7 @@ int (*timer_hook)(struct pt_regs *) __read_mostly; static atomic_t *prof_buffer; static unsigned long prof_len, prof_shift; -static int prof_on __read_mostly; +int prof_on __read_mostly; static cpumask_t prof_cpu_mask = CPU_MASK_ALL; #ifdef CONFIG_SMP static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits); @@ -51,9 +51,19 @@ static DEFINE_MUTEX(profile_flip_mutex); static int __init profile_setup(char * str) { static char __initdata schedstr[] = "schedule"; + static char __initdata sleepstr[] = "sleep"; int par; - if (!strncmp(str, schedstr, strlen(schedstr))) { + if (!strncmp(str, sleepstr, strlen(sleepstr))) { + prof_on = SLEEP_PROFILING; + if (str[strlen(sleepstr)] == ',') + str += strlen(sleepstr) + 1; + if (get_option(&str, &par)) + prof_shift = par; + printk(KERN_INFO + "kernel sleep profiling enabled (shift: %ld)\n", + prof_shift); + } else if (!strncmp(str, sleepstr, strlen(sleepstr))) { prof_on = SCHED_PROFILING; if (str[strlen(schedstr)] == ',') str += strlen(schedstr) + 1; @@ -204,7 +214,8 @@ EXPORT_SYMBOL_GPL(profile_event_unregister); * positions to which hits are accounted during short intervals (e.g. * several seconds) is usually very small. Exclusion from buffer * flipping is provided by interrupt disablement (note that for - * SCHED_PROFILING profile_hit() may be called from process context). + * SCHED_PROFILING or SLEEP_PROFILING profile_hit() may be called from + * process context). * The hash function is meant to be lightweight as opposed to strong, * and was vaguely inspired by ppc64 firmware-supported inverted * pagetable hash functions, but uses a full hashtable full of finite @@ -257,7 +268,7 @@ static void profile_discard_flip_buffers(void) mutex_unlock(&profile_flip_mutex); } -void profile_hit(int type, void *__pc) +void profile_hits(int type, void *__pc, unsigned int nr_hits) { unsigned long primary, secondary, flags, pc = (unsigned long)__pc; int i, j, cpu; @@ -274,21 +285,31 @@ void profile_hit(int type, void *__pc) put_cpu(); return; } + /* + * We buffer the global profiler buffer into a per-CPU + * queue and thus reduce the number of global (and possibly + * NUMA-alien) accesses. The write-queue is self-coalescing: + */ local_irq_save(flags); do { for (j = 0; j < PROFILE_GRPSZ; ++j) { if (hits[i + j].pc == pc) { - hits[i + j].hits++; + hits[i + j].hits += nr_hits; goto out; } else if (!hits[i + j].hits) { hits[i + j].pc = pc; - hits[i + j].hits = 1; + hits[i + j].hits = nr_hits; goto out; } } i = (i + secondary) & (NR_PROFILE_HIT - 1); } while (i != primary); - atomic_inc(&prof_buffer[pc]); + + /* + * Add the current hit(s) and flush the write-queue out + * to the global buffer: + */ + atomic_add(nr_hits, &prof_buffer[pc]); for (i = 0; i < NR_PROFILE_HIT; ++i) { atomic_add(hits[i].hits, &prof_buffer[hits[i].pc]); hits[i].pc = hits[i].hits = 0; @@ -356,14 +377,14 @@ static int __devinit profile_cpu_callback(struct notifier_block *info, #define profile_flip_buffers() do { } while (0) #define profile_discard_flip_buffers() do { } while (0) -void profile_hit(int type, void *__pc) +void profile_hits(int type, void *__pc, unsigned int nr_hits) { unsigned long pc; if (prof_on != type || !prof_buffer) return; pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift; - atomic_inc(&prof_buffer[min(pc, prof_len - 1)]); + atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]); } #endif /* !CONFIG_SMP */ diff --git a/kernel/sched.c b/kernel/sched.c index 343e1794233e..75a005ed4eda 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -948,6 +948,17 @@ static void activate_task(struct task_struct *p, struct rq *rq, int local) } #endif + /* + * Sleep time is in units of nanosecs, so shift by 20 to get a + * milliseconds-range estimation of the amount of time that the task + * spent sleeping: + */ + if (unlikely(prof_on == SLEEP_PROFILING)) { + if (p->state == TASK_UNINTERRUPTIBLE) + profile_hits(SLEEP_PROFILING, (void *)get_wchan(p), + (now - p->timestamp) >> 20); + } + if (!rt_task(p)) p->prio = recalc_task_prio(p, now); -- cgit v1.2.3 From d5abe669172f20a4129a711de0f250a4e07db298 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 6 Dec 2006 20:37:26 -0800 Subject: [PATCH] debug: workqueue locking sanity Workqueue functions should not leak locks, assert so, printing the last function ran. Use macros in lockdep.h to avoid include dependency pains. [akpm@osdl.org: build fix] Signed-off-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/lockdep.h | 5 +++++ kernel/workqueue.c | 13 +++++++++++++ 2 files changed, 18 insertions(+) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 819f08f1310d..da19aeb0dbe8 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -243,6 +243,8 @@ extern void lock_release(struct lockdep_map *lock, int nested, # define INIT_LOCKDEP .lockdep_recursion = 0, +#define lockdep_depth(tsk) ((tsk)->lockdep_depth) + #else /* !LOCKDEP */ static inline void lockdep_off(void) @@ -277,6 +279,9 @@ static inline int lockdep_internal(void) * The class key takes no space if lockdep is disabled: */ struct lock_class_key { }; + +#define lockdep_depth(tsk) (0) + #endif /* !LOCKDEP */ #if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_GENERIC_HARDIRQS) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 2945b094d871..5484d6e045c2 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -30,6 +30,8 @@ #include #include #include +#include +#include /* * The per-CPU workqueue (if single thread, we always use the first @@ -253,6 +255,17 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) work_release(work); f(work); + if (unlikely(in_atomic() || lockdep_depth(current) > 0)) { + printk(KERN_ERR "BUG: workqueue leaked lock or atomic: " + "%s/0x%08x/%d\n", + current->comm, preempt_count(), + current->pid); + printk(KERN_ERR " last function: "); + print_symbol("%s\n", (unsigned long)f); + debug_show_held_locks(current); + dump_stack(); + } + spin_lock_irqsave(&cwq->lock, flags); cwq->remove_sequence++; wake_up(&cwq->work_done); -- cgit v1.2.3 From 04903664325acb3f199dd8a4b8f1aa437e9fd6b2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:37:33 -0800 Subject: [PATCH] remove HASH_HIGHMEM It has no users and it's doubtful that we'll need it again. Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 3 +-- mm/page_alloc.c | 2 +- 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 31e9abb6d977..2275f2748708 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -119,8 +119,7 @@ extern void *alloc_large_system_hash(const char *tablename, unsigned int *_hash_mask, unsigned long limit); -#define HASH_HIGHMEM 0x00000001 /* Consider highmem? */ -#define HASH_EARLY 0x00000002 /* Allocating during early boot? */ +#define HASH_EARLY 0x00000001 /* Allocating during early boot? */ /* Only NUMA needs hash distribution. * IA64 is known to have sufficient vmalloc space. diff --git a/mm/page_alloc.c b/mm/page_alloc.c index d539f83c62b6..2273952300d4 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3226,7 +3226,7 @@ void *__init alloc_large_system_hash(const char *tablename, /* allow the kernel cmdline to have a say */ if (!numentries) { /* round applicable memory size up to nearest megabyte */ - numentries = (flags & HASH_HIGHMEM) ? nr_all_pages : nr_kernel_pages; + numentries = nr_kernel_pages; numentries += (1UL << (20 - PAGE_SHIFT)) - 1; numentries >>= 20 - PAGE_SHIFT; numentries <<= 20 - PAGE_SHIFT; -- cgit v1.2.3 From cfd1893477fa94bb0915e39afa2f044ac978b5c6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 6 Dec 2006 20:37:38 -0800 Subject: [PATCH] ktime: Fix signed / unsigned mismatch in ktime_to_ns The 32 bit implementation of ktime_to_ns returns unsigned value, while the 64 bit version correctly returns an signed value. There is no current user affected by this, but it has to be fixed, as ktime values can be negative. Pointed-out-by: Helmut Duregger Signed-off-by: Thomas Gleixner Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ktime.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ktime.h b/include/linux/ktime.h index 84eeecd60a02..611f17f79eef 100644 --- a/include/linux/ktime.h +++ b/include/linux/ktime.h @@ -248,9 +248,9 @@ static inline struct timeval ktime_to_timeval(const ktime_t kt) * * Returns the scalar nanoseconds representation of kt */ -static inline u64 ktime_to_ns(const ktime_t kt) +static inline s64 ktime_to_ns(const ktime_t kt) { - return (u64) kt.tv.sec * NSEC_PER_SEC + kt.tv.nsec; + return (s64) kt.tv.sec * NSEC_PER_SEC + kt.tv.nsec; } #endif -- cgit v1.2.3 From 651971cb7242e8f6d7ebd153e69bd271cb731223 Mon Sep 17 00:00:00 2001 From: suzuki Date: Wed, 6 Dec 2006 20:37:48 -0800 Subject: [PATCH] Fix the size limit of compat space msgsize Currently we allocate 64k space on the user stack and use it the msgbuf for sys_{msgrcv,msgsnd} for compat and the results are later copied in user [ by copy_in_user]. This patch introduces helper routines for sys_{msgrcv,msgsnd} as below: do_msgsnd() : Accepts the mtype and user space ptr to the buffer along with the msqid and msgflg. do_msgrcv() : Accepts a kernel space ptr to mtype and a userspace ptr to the buffer. The mtype has to be copied back the user space msgbuf by the caller. These changes avoid the need to allocate the msgsize on the userspace ( thus removing the size limt ) and the overhead of an extra copy_in_user(). Signed-off-by: Suzuki K P Cc: Arnd Bergmann Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/msg.h | 6 ++++++ ipc/compat.c | 23 +++++++---------------- ipc/msg.c | 44 +++++++++++++++++++++++++++++++++----------- 3 files changed, 46 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/msg.h b/include/linux/msg.h index acc7c174ff00..f1b60740d641 100644 --- a/include/linux/msg.h +++ b/include/linux/msg.h @@ -92,6 +92,12 @@ struct msg_queue { struct list_head q_senders; }; +/* Helper routines for sys_msgsnd and sys_msgrcv */ +extern long do_msgsnd(int msqid, long mtype, void __user *mtext, + size_t msgsz, int msgflg); +extern long do_msgrcv(int msqid, long *pmtype, void __user *mtext, + size_t msgsz, long msgtyp, int msgflg); + #endif /* __KERNEL__ */ #endif /* _LINUX_MSG_H */ diff --git a/ipc/compat.c b/ipc/compat.c index 4d20cfd38f0a..fa18141539fb 100644 --- a/ipc/compat.c +++ b/ipc/compat.c @@ -115,7 +115,6 @@ struct compat_shm_info { extern int sem_ctls[]; #define sc_semopm (sem_ctls[2]) -#define MAXBUF (64*1024) static inline int compat_ipc_parse_version(int *cmd) { @@ -307,35 +306,30 @@ long compat_sys_semctl(int first, int second, int third, void __user *uptr) long compat_sys_msgsnd(int first, int second, int third, void __user *uptr) { - struct msgbuf __user *p; struct compat_msgbuf __user *up = uptr; long type; if (first < 0) return -EINVAL; - if (second < 0 || (second >= MAXBUF - sizeof(struct msgbuf))) + if (second < 0) return -EINVAL; - p = compat_alloc_user_space(second + sizeof(struct msgbuf)); - if (get_user(type, &up->mtype) || - put_user(type, &p->mtype) || - copy_in_user(p->mtext, up->mtext, second)) + if (get_user(type, &up->mtype)) return -EFAULT; - return sys_msgsnd(first, p, second, third); + return do_msgsnd(first, type, up->mtext, second, third); } long compat_sys_msgrcv(int first, int second, int msgtyp, int third, int version, void __user *uptr) { - struct msgbuf __user *p; struct compat_msgbuf __user *up; long type; int err; if (first < 0) return -EINVAL; - if (second < 0 || (second >= MAXBUF - sizeof(struct msgbuf))) + if (second < 0) return -EINVAL; if (!version) { @@ -349,14 +343,11 @@ long compat_sys_msgrcv(int first, int second, int msgtyp, int third, uptr = compat_ptr(ipck.msgp); msgtyp = ipck.msgtyp; } - p = compat_alloc_user_space(second + sizeof(struct msgbuf)); - err = sys_msgrcv(first, p, second, msgtyp, third); + up = uptr; + err = do_msgrcv(first, &type, up->mtext, second, msgtyp, third); if (err < 0) goto out; - up = uptr; - if (get_user(type, &p->mtype) || - put_user(type, &up->mtype) || - copy_in_user(up->mtext, p->mtext, err)) + if (put_user(type, &up->mtype)) err = -EFAULT; out: return err; diff --git a/ipc/msg.c b/ipc/msg.c index 1266b1d0c8e3..a388824740e7 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -626,12 +626,11 @@ static inline int pipelined_send(struct msg_queue *msq, struct msg_msg *msg) return 0; } -asmlinkage long -sys_msgsnd(int msqid, struct msgbuf __user *msgp, size_t msgsz, int msgflg) +long do_msgsnd(int msqid, long mtype, void __user *mtext, + size_t msgsz, int msgflg) { struct msg_queue *msq; struct msg_msg *msg; - long mtype; int err; struct ipc_namespace *ns; @@ -639,12 +638,10 @@ sys_msgsnd(int msqid, struct msgbuf __user *msgp, size_t msgsz, int msgflg) if (msgsz > ns->msg_ctlmax || (long) msgsz < 0 || msqid < 0) return -EINVAL; - if (get_user(mtype, &msgp->mtype)) - return -EFAULT; if (mtype < 1) return -EINVAL; - msg = load_msg(msgp->mtext, msgsz); + msg = load_msg(mtext, msgsz); if (IS_ERR(msg)) return PTR_ERR(msg); @@ -723,6 +720,16 @@ out_free: return err; } +asmlinkage long +sys_msgsnd(int msqid, struct msgbuf __user *msgp, size_t msgsz, int msgflg) +{ + long mtype; + + if (get_user(mtype, &msgp->mtype)) + return -EFAULT; + return do_msgsnd(msqid, mtype, msgp->mtext, msgsz, msgflg); +} + static inline int convert_mode(long *msgtyp, int msgflg) { /* @@ -742,8 +749,8 @@ static inline int convert_mode(long *msgtyp, int msgflg) return SEARCH_EQUAL; } -asmlinkage long sys_msgrcv(int msqid, struct msgbuf __user *msgp, size_t msgsz, - long msgtyp, int msgflg) +long do_msgrcv(int msqid, long *pmtype, void __user *mtext, + size_t msgsz, long msgtyp, int msgflg) { struct msg_queue *msq; struct msg_msg *msg; @@ -889,15 +896,30 @@ out_unlock: return PTR_ERR(msg); msgsz = (msgsz > msg->m_ts) ? msg->m_ts : msgsz; - if (put_user (msg->m_type, &msgp->mtype) || - store_msg(msgp->mtext, msg, msgsz)) { + *pmtype = msg->m_type; + if (store_msg(mtext, msg, msgsz)) msgsz = -EFAULT; - } + free_msg(msg); return msgsz; } +asmlinkage long sys_msgrcv(int msqid, struct msgbuf __user *msgp, size_t msgsz, + long msgtyp, int msgflg) +{ + long err, mtype; + + err = do_msgrcv(msqid, &mtype, msgp->mtext, msgsz, msgtyp, msgflg); + if (err < 0) + goto out; + + if (put_user(mtype, &msgp->mtype)) + err = -EFAULT; +out: + return err; +} + #ifdef CONFIG_PROC_FS static int sysvipc_msg_proc_show(struct seq_file *s, void *it) { -- cgit v1.2.3 From 386d9a7edd9f3492c99124b0a659e9ed7abb30f9 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 6 Dec 2006 20:37:53 -0800 Subject: [PATCH] elf: Always define elf_addr_t in linux/elf.h Define elf_addr_t in linux/elf.h. The size of the type is determined using ELF_CLASS. This allows us to remove the defines that today are spread all over .c and .h files. Signed-off-by: Magnus Damm Cc: Daniel Jacobowitz Cc: Roland McGrath Cc: Jakub Jelinek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/ia32priv.h | 2 -- arch/mips/kernel/binfmt_elfn32.c | 1 - arch/mips/kernel/binfmt_elfo32.c | 1 - arch/mips/kernel/irixelf.c | 4 ---- arch/parisc/kernel/binfmt_elf32.c | 1 - arch/s390/kernel/binfmt_elf32.c | 1 - arch/sparc64/kernel/binfmt_elf32.c | 1 - arch/x86_64/ia32/ia32_binfmt.c | 2 -- fs/binfmt_elf.c | 4 ---- fs/binfmt_elf_fdpic.c | 3 --- include/asm-powerpc/elf.h | 2 -- include/linux/elf.h | 2 ++ 12 files changed, 2 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h index 703a67c934f8..cfa0bc0026b5 100644 --- a/arch/ia64/ia32/ia32priv.h +++ b/arch/ia64/ia32/ia32priv.h @@ -330,8 +330,6 @@ struct old_linux32_dirent { void ia64_elf32_init(struct pt_regs *regs); #define ELF_PLAT_INIT(_r, load_addr) ia64_elf32_init(_r) -#define elf_addr_t u32 - /* This macro yields a bitmask that programs can use to figure out what instruction set this CPU supports. */ #define ELF_HWCAP 0 diff --git a/arch/mips/kernel/binfmt_elfn32.c b/arch/mips/kernel/binfmt_elfn32.c index 4a9f1ecefaf2..9b34238d41c0 100644 --- a/arch/mips/kernel/binfmt_elfn32.c +++ b/arch/mips/kernel/binfmt_elfn32.c @@ -90,7 +90,6 @@ struct elf_prpsinfo32 char pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */ }; -#define elf_addr_t u32 #define elf_caddr_t u32 #define init_elf_binfmt init_elfn32_binfmt diff --git a/arch/mips/kernel/binfmt_elfo32.c b/arch/mips/kernel/binfmt_elfo32.c index e31813779895..993f7ec70f35 100644 --- a/arch/mips/kernel/binfmt_elfo32.c +++ b/arch/mips/kernel/binfmt_elfo32.c @@ -92,7 +92,6 @@ struct elf_prpsinfo32 char pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */ }; -#define elf_addr_t u32 #define elf_caddr_t u32 #define init_elf_binfmt init_elf32_binfmt diff --git a/arch/mips/kernel/irixelf.c b/arch/mips/kernel/irixelf.c index ab12c8f01518..a82fae221ce4 100644 --- a/arch/mips/kernel/irixelf.c +++ b/arch/mips/kernel/irixelf.c @@ -52,10 +52,6 @@ static struct linux_binfmt irix_format = { irix_core_dump, PAGE_SIZE }; -#ifndef elf_addr_t -#define elf_addr_t unsigned long -#endif - #ifdef DEBUG /* Debugging routines. */ static char *get_elf_p_type(Elf32_Word p_type) diff --git a/arch/parisc/kernel/binfmt_elf32.c b/arch/parisc/kernel/binfmt_elf32.c index 1e64e7b88110..ecb10a4f63c6 100644 --- a/arch/parisc/kernel/binfmt_elf32.c +++ b/arch/parisc/kernel/binfmt_elf32.c @@ -75,7 +75,6 @@ struct elf_prpsinfo32 char pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */ }; -#define elf_addr_t unsigned int #define init_elf_binfmt init_elf32_binfmt #define ELF_PLATFORM ("PARISC32\0") diff --git a/arch/s390/kernel/binfmt_elf32.c b/arch/s390/kernel/binfmt_elf32.c index 9565a2dcfadc..5c46054195cb 100644 --- a/arch/s390/kernel/binfmt_elf32.c +++ b/arch/s390/kernel/binfmt_elf32.c @@ -176,7 +176,6 @@ struct elf_prpsinfo32 #include -#define elf_addr_t u32 /* #define init_elf_binfmt init_elf32_binfmt */ diff --git a/arch/sparc64/kernel/binfmt_elf32.c b/arch/sparc64/kernel/binfmt_elf32.c index a98f3ae175a3..9ad84ff10a17 100644 --- a/arch/sparc64/kernel/binfmt_elf32.c +++ b/arch/sparc64/kernel/binfmt_elf32.c @@ -141,7 +141,6 @@ cputime_to_compat_timeval(const cputime_t cputime, struct compat_timeval *value) value->tv_sec = jiffies / HZ; } -#define elf_addr_t u32 #undef start_thread #define start_thread start_thread32 #define init_elf_binfmt init_elf32_binfmt diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 932a62ad6c83..543ef4f405e9 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -305,8 +305,6 @@ MODULE_AUTHOR("Eric Youngdale, Andi Kleen"); #undef MODULE_DESCRIPTION #undef MODULE_AUTHOR -#define elf_addr_t __u32 - static void elf32_init(struct pt_regs *); #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 68e20d5bfe1b..14ea630a857c 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -47,10 +47,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs); static int load_elf_library(struct file *); static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int); -#ifndef elf_addr_t -#define elf_addr_t unsigned long -#endif - /* * If we don't support core dumping, then supply a NULL so we * don't even try. diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index f86d5c9ce5eb..ed9a61c6beb3 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -40,9 +40,6 @@ #include typedef char *elf_caddr_t; -#ifndef elf_addr_t -#define elf_addr_t unsigned long -#endif #if 0 #define kdebug(fmt, ...) printk("FDPIC "fmt"\n" ,##__VA_ARGS__ ) diff --git a/include/asm-powerpc/elf.h b/include/asm-powerpc/elf.h index b5436642a109..d36426c01b6b 100644 --- a/include/asm-powerpc/elf.h +++ b/include/asm-powerpc/elf.h @@ -124,12 +124,10 @@ typedef elf_greg_t32 elf_gregset_t32[ELF_NGREG]; # define ELF_DATA ELFDATA2MSB typedef elf_greg_t64 elf_greg_t; typedef elf_gregset_t64 elf_gregset_t; -# define elf_addr_t unsigned long #else /* Assumption: ELF_ARCH == EM_PPC and ELF_CLASS == ELFCLASS32 */ typedef elf_greg_t32 elf_greg_t; typedef elf_gregset_t32 elf_gregset_t; -# define elf_addr_t __u32 #endif /* ELF_ARCH */ /* Floating point registers */ diff --git a/include/linux/elf.h b/include/linux/elf.h index 743d5c8e6d36..b403516d5c3d 100644 --- a/include/linux/elf.h +++ b/include/linux/elf.h @@ -358,6 +358,7 @@ extern Elf32_Dyn _DYNAMIC []; #define elfhdr elf32_hdr #define elf_phdr elf32_phdr #define elf_note elf32_note +#define elf_addr_t Elf32_Off #else @@ -365,6 +366,7 @@ extern Elf64_Dyn _DYNAMIC []; #define elfhdr elf64_hdr #define elf_phdr elf64_phdr #define elf_note elf64_note +#define elf_addr_t Elf64_Off #endif -- cgit v1.2.3 From b4c6c34a530b4d1c626f4ac0a884e0a9b849378c Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Wed, 6 Dec 2006 20:38:11 -0800 Subject: [PATCH] kprobes: enable booster on the preemptible kernel When we are unregistering a kprobe-booster, we can't release its instruction buffer immediately on the preemptive kernel, because some processes might be preempted on the buffer. The freeze_processes() and thaw_processes() functions can clean most of processes up from the buffer. There are still some non-frozen threads who have the PF_NOFREEZE flag. If those threads are sleeping (not preempted) at the known place outside the buffer, we can ensure safety of freeing. However, the processing of this check routine takes a long time. So, this patch introduces the garbage collection mechanism of insn_slot. It also introduces the "dirty" flag to free_insn_slot because of efficiency. The "clean" instruction slots (dirty flag is cleared) are released immediately. But the "dirty" slots which are used by boosted kprobes, are marked as garbages. collect_garbage_slots() will be invoked to release "dirty" slots if there are more than INSNS_PER_PAGE garbage slots or if there are no unused slots. Cc: "Keshavamurthy, Anil S" Cc: Ananth N Mavinakayanahalli Cc: "bibo,mao" Cc: Prasanna S Panchamukhi Cc: Yumiko Sugita Cc: Satoshi Oshima Cc: Hideo Aoki Signed-off-by: Masami Hiramatsu Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 4 +- arch/ia64/kernel/kprobes.c | 2 +- arch/powerpc/kernel/kprobes.c | 2 +- arch/s390/kernel/kprobes.c | 2 +- arch/x86_64/kernel/kprobes.c | 2 +- include/linux/kprobes.h | 2 +- kernel/kprobes.c | 117 ++++++++++++++++++++++++++++++++++-------- 7 files changed, 103 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index fc79e1e859c4..af1d53344993 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -184,7 +184,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1)); mutex_unlock(&kprobe_mutex); } @@ -333,7 +333,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) return 1; ss_probe: -#ifndef CONFIG_PREEMPT +#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM) if (p->ainsn.boostable == 1 && !p->post_handler){ /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 51217d63285e..4d592ee9300b 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -481,7 +481,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } /* diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index 7b8d12b9026c..4657563f8813 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -85,7 +85,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 67914fe7f317..576368c4f605 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -200,7 +200,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index ac241567e682..209c8c0bec71 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -224,7 +224,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn); + free_insn_slot(p->ainsn.insn, 0); mutex_unlock(&kprobe_mutex); } diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index ac4c0559f751..769be39b9681 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -165,7 +165,7 @@ extern void arch_disarm_kprobe(struct kprobe *p); extern int arch_init_kprobes(void); extern void show_registers(struct pt_regs *regs); extern kprobe_opcode_t *get_insn_slot(void); -extern void free_insn_slot(kprobe_opcode_t *slot); +extern void free_insn_slot(kprobe_opcode_t *slot, int dirty); extern void kprobes_inc_nmissed_count(struct kprobe *p); /* Get the kprobe at this addr (if any) - called with preemption disabled */ diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 610c837ad9e0..17ec4afb0994 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -83,9 +84,36 @@ struct kprobe_insn_page { kprobe_opcode_t *insns; /* Page of instruction slots */ char slot_used[INSNS_PER_PAGE]; int nused; + int ngarbage; }; static struct hlist_head kprobe_insn_pages; +static int kprobe_garbage_slots; +static int collect_garbage_slots(void); + +static int __kprobes check_safety(void) +{ + int ret = 0; +#if defined(CONFIG_PREEMPT) && defined(CONFIG_PM) + ret = freeze_processes(); + if (ret == 0) { + struct task_struct *p, *q; + do_each_thread(p, q) { + if (p != current && p->state == TASK_RUNNING && + p->pid != 0) { + printk("Check failed: %s is running\n",p->comm); + ret = -1; + goto loop_end; + } + } while_each_thread(p, q); + } +loop_end: + thaw_processes(); +#else + synchronize_sched(); +#endif + return ret; +} /** * get_insn_slot() - Find a slot on an executable page for an instruction. @@ -96,6 +124,7 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) struct kprobe_insn_page *kip; struct hlist_node *pos; + retry: hlist_for_each(pos, &kprobe_insn_pages) { kip = hlist_entry(pos, struct kprobe_insn_page, hlist); if (kip->nused < INSNS_PER_PAGE) { @@ -112,7 +141,11 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) } } - /* All out of space. Need to allocate a new page. Use slot 0.*/ + /* If there are any garbage slots, collect it and try again. */ + if (kprobe_garbage_slots && collect_garbage_slots() == 0) { + goto retry; + } + /* All out of space. Need to allocate a new page. Use slot 0. */ kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL); if (!kip) { return NULL; @@ -133,10 +166,62 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) memset(kip->slot_used, 0, INSNS_PER_PAGE); kip->slot_used[0] = 1; kip->nused = 1; + kip->ngarbage = 0; return kip->insns; } -void __kprobes free_insn_slot(kprobe_opcode_t *slot) +/* Return 1 if all garbages are collected, otherwise 0. */ +static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) +{ + kip->slot_used[idx] = 0; + kip->nused--; + if (kip->nused == 0) { + /* + * Page is no longer in use. Free it unless + * it's the last one. We keep the last one + * so as not to have to set it up again the + * next time somebody inserts a probe. + */ + hlist_del(&kip->hlist); + if (hlist_empty(&kprobe_insn_pages)) { + INIT_HLIST_NODE(&kip->hlist); + hlist_add_head(&kip->hlist, + &kprobe_insn_pages); + } else { + module_free(NULL, kip->insns); + kfree(kip); + } + return 1; + } + return 0; +} + +static int __kprobes collect_garbage_slots(void) +{ + struct kprobe_insn_page *kip; + struct hlist_node *pos, *next; + + /* Ensure no-one is preepmted on the garbages */ + if (check_safety() != 0) + return -EAGAIN; + + hlist_for_each_safe(pos, next, &kprobe_insn_pages) { + int i; + kip = hlist_entry(pos, struct kprobe_insn_page, hlist); + if (kip->ngarbage == 0) + continue; + kip->ngarbage = 0; /* we will collect all garbages */ + for (i = 0; i < INSNS_PER_PAGE; i++) { + if (kip->slot_used[i] == -1 && + collect_one_slot(kip, i)) + break; + } + } + kprobe_garbage_slots = 0; + return 0; +} + +void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) { struct kprobe_insn_page *kip; struct hlist_node *pos; @@ -146,28 +231,18 @@ void __kprobes free_insn_slot(kprobe_opcode_t *slot) if (kip->insns <= slot && slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; - kip->slot_used[i] = 0; - kip->nused--; - if (kip->nused == 0) { - /* - * Page is no longer in use. Free it unless - * it's the last one. We keep the last one - * so as not to have to set it up again the - * next time somebody inserts a probe. - */ - hlist_del(&kip->hlist); - if (hlist_empty(&kprobe_insn_pages)) { - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, - &kprobe_insn_pages); - } else { - module_free(NULL, kip->insns); - kfree(kip); - } + if (dirty) { + kip->slot_used[i] = -1; + kip->ngarbage++; + } else { + collect_one_slot(kip, i); } - return; + break; } } + if (dirty && (++kprobe_garbage_slots > INSNS_PER_PAGE)) { + collect_garbage_slots(); + } } #endif -- cgit v1.2.3 From 83df8db9e62129975fab6d800fb381faf0dfee74 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 6 Dec 2006 20:38:14 -0800 Subject: [PATCH] declare smp_call_function_single in generic code smp_call_function_single() needs to be visible in non-SMP builds, to fix: arch/x86_64/kernel/vsyscall.c:283: warning: implicit declaration of function 'smp_call_function_single' Signed-off-by: Randy Dunlap Cc: Ingo Molnar Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-x86_64/smp.h | 7 ------- include/linux/smp.h | 7 +++++++ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h index d6b7c057edba..f1bdd500d7ac 100644 --- a/include/asm-x86_64/smp.h +++ b/include/asm-x86_64/smp.h @@ -118,13 +118,6 @@ static __inline int logical_smp_processor_id(void) #define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu] #else #define cpu_physical_id(cpu) boot_cpu_id -static inline int smp_call_function_single(int cpuid, void (*func) (void *info), - void *info, int retry, int wait) -{ - /* Disable interrupts here? */ - func(info); - return 0; -} #endif /* !CONFIG_SMP */ #endif diff --git a/include/linux/smp.h b/include/linux/smp.h index 51649987f691..7ba23ec8211b 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -99,6 +99,13 @@ static inline int up_smp_call_function(void) static inline void smp_send_reschedule(int cpu) { } #define num_booting_cpus() 1 #define smp_prepare_boot_cpu() do {} while (0) +static inline int smp_call_function_single(int cpuid, void (*func) (void *info), + void *info, int retry, int wait) +{ + /* Disable interrupts here? */ + func(info); + return 0; +} #endif /* !SMP */ -- cgit v1.2.3 From 02316067852187b8bec781bec07410e91af79627 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:38:17 -0800 Subject: [PATCH] hotplug CPU: clean up hotcpu_notifier() use There was lots of #ifdef noise in the kernel due to hotcpu_notifier(fn, prio) not correctly marking 'fn' as used in the !HOTPLUG_CPU case, and thus generating compiler warnings of unused symbols, hence forcing people to add #ifdefs. the compiler can skip truly unused functions just fine: text data bss dec hex filename 1624412 728710 3674856 6027978 5bfaca vmlinux.before 1624412 728710 3674856 6027978 5bfaca vmlinux.after [akpm@osdl.org: topology.c fix] Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/mcheck/therm_throt.c | 2 -- arch/i386/kernel/cpuid.c | 2 -- arch/i386/kernel/microcode.c | 2 -- arch/i386/kernel/msr.c | 2 -- arch/ia64/kernel/palinfo.c | 2 -- arch/ia64/kernel/salinfo.c | 2 -- arch/s390/appldata/appldata_base.c | 2 -- arch/x86_64/kernel/mce.c | 2 -- arch/x86_64/kernel/mce_amd.c | 4 ++-- arch/x86_64/kernel/vsyscall.c | 2 -- block/ll_rw_blk.c | 4 ---- drivers/base/topology.c | 2 -- drivers/cpufreq/cpufreq.c | 2 -- fs/buffer.c | 2 -- include/linux/cpu.h | 6 +++--- kernel/cpuset.c | 4 ---- kernel/profile.c | 3 +-- kernel/sched.c | 3 --- kernel/workqueue.c | 2 -- lib/radix-tree.c | 2 -- mm/page_alloc.c | 4 ---- mm/swap.c | 2 ++ mm/vmscan.c | 2 -- net/core/dev.c | 2 -- net/core/flow.c | 2 -- 25 files changed, 8 insertions(+), 56 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c index bad8b4420709..065005c3f168 100644 --- a/arch/i386/kernel/cpu/mcheck/therm_throt.c +++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c @@ -116,7 +116,6 @@ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) return sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group); } -#ifdef CONFIG_HOTPLUG_CPU static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev) { return sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group); @@ -153,7 +152,6 @@ static struct notifier_block thermal_throttle_cpu_notifier = { .notifier_call = thermal_throttle_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ static __init int thermal_throttle_init_device(void) { diff --git a/arch/i386/kernel/cpuid.c b/arch/i386/kernel/cpuid.c index ab0c327e79dc..23b2cc748d4e 100644 --- a/arch/i386/kernel/cpuid.c +++ b/arch/i386/kernel/cpuid.c @@ -167,7 +167,6 @@ static int cpuid_device_create(int i) return err; } -#ifdef CONFIG_HOTPLUG_CPU static int cpuid_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -187,7 +186,6 @@ static struct notifier_block __cpuinitdata cpuid_class_cpu_notifier = { .notifier_call = cpuid_class_cpu_callback, }; -#endif /* !CONFIG_HOTPLUG_CPU */ static int __init cpuid_init(void) { diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c index 23f5984d0654..972346604f9d 100644 --- a/arch/i386/kernel/microcode.c +++ b/arch/i386/kernel/microcode.c @@ -703,7 +703,6 @@ static struct sysdev_driver mc_sysdev_driver = { .resume = mc_sysdev_resume, }; -#ifdef CONFIG_HOTPLUG_CPU static __cpuinit int mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { @@ -726,7 +725,6 @@ mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) static struct notifier_block mc_cpu_notifier = { .notifier_call = mc_cpu_callback, }; -#endif static int __init microcode_init (void) { diff --git a/arch/i386/kernel/msr.c b/arch/i386/kernel/msr.c index a773f776c9ea..7763c67ca282 100644 --- a/arch/i386/kernel/msr.c +++ b/arch/i386/kernel/msr.c @@ -250,7 +250,6 @@ static int msr_device_create(int i) return err; } -#ifdef CONFIG_HOTPLUG_CPU static int msr_class_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -271,7 +270,6 @@ static struct notifier_block __cpuinitdata msr_class_cpu_notifier = { .notifier_call = msr_class_cpu_callback, }; -#endif static int __init msr_init(void) { diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c index 0b546e2b36ac..c4c10a0b99d9 100644 --- a/arch/ia64/kernel/palinfo.c +++ b/arch/ia64/kernel/palinfo.c @@ -952,7 +952,6 @@ remove_palinfo_proc_entries(unsigned int hcpu) } } -#ifdef CONFIG_HOTPLUG_CPU static int palinfo_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -974,7 +973,6 @@ static struct notifier_block palinfo_cpu_notifier = .notifier_call = palinfo_cpu_callback, .priority = 0, }; -#endif static int __init palinfo_init(void) diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index e63b8ca5344a..fd607ca51a8d 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -575,7 +575,6 @@ static struct file_operations salinfo_data_fops = { .write = salinfo_log_write, }; -#ifdef CONFIG_HOTPLUG_CPU static int __devinit salinfo_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu) { @@ -620,7 +619,6 @@ static struct notifier_block salinfo_cpu_notifier = .notifier_call = salinfo_cpu_callback, .priority = 0, }; -#endif /* CONFIG_HOTPLUG_CPU */ static int __init salinfo_init(void) diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 67d5cf9cba83..b8c237290263 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -561,7 +561,6 @@ appldata_offline_cpu(int cpu) spin_unlock(&appldata_timer_lock); } -#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit appldata_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) @@ -582,7 +581,6 @@ appldata_cpu_notify(struct notifier_block *self, static struct notifier_block appldata_nb = { .notifier_call = appldata_cpu_notify, }; -#endif /* * appldata_init() diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c index c7587fc39015..bc863c464a1f 100644 --- a/arch/x86_64/kernel/mce.c +++ b/arch/x86_64/kernel/mce.c @@ -641,7 +641,6 @@ static __cpuinit int mce_create_device(unsigned int cpu) return err; } -#ifdef CONFIG_HOTPLUG_CPU static void mce_remove_device(unsigned int cpu) { int i; @@ -674,7 +673,6 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) static struct notifier_block mce_cpu_notifier = { .notifier_call = mce_cpu_callback, }; -#endif static __init int mce_init_device(void) { diff --git a/arch/x86_64/kernel/mce_amd.c b/arch/x86_64/kernel/mce_amd.c index 883fe747f64c..fa09debad4b7 100644 --- a/arch/x86_64/kernel/mce_amd.c +++ b/arch/x86_64/kernel/mce_amd.c @@ -551,7 +551,6 @@ out: return err; } -#ifdef CONFIG_HOTPLUG_CPU /* * let's be hotplug friendly. * in case of multiple core processors, the first core always takes ownership @@ -594,12 +593,14 @@ static void threshold_remove_bank(unsigned int cpu, int bank) sprintf(name, "threshold_bank%i", bank); +#ifdef CONFIG_SMP /* sibling symlink */ if (shared_bank[bank] && b->blocks->cpu != cpu) { sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name); per_cpu(threshold_banks, cpu)[bank] = NULL; return; } +#endif /* remove all sibling symlinks before unregistering */ for_each_cpu_mask(i, b->cpus) { @@ -656,7 +657,6 @@ static int threshold_cpu_callback(struct notifier_block *nfb, static struct notifier_block threshold_cpu_notifier = { .notifier_call = threshold_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ static __init int threshold_init_device(void) { diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 630036c06c75..3785e4954734 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -275,7 +275,6 @@ static void __cpuinit cpu_vsyscall_init(void *arg) vsyscall_set_cpu(raw_smp_processor_id()); } -#ifdef CONFIG_HOTPLUG_CPU static int __cpuinit cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) { @@ -284,7 +283,6 @@ cpu_vsyscall_notifier(struct notifier_block *n, unsigned long action, void *arg) smp_call_function_single(cpu, cpu_vsyscall_init, NULL, 0, 1); return NOTIFY_DONE; } -#endif static void __init map_vsyscall(void) { diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index a4ff3271d4a8..31512cd9f3ad 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -3459,8 +3459,6 @@ static void blk_done_softirq(struct softirq_action *h) } } -#ifdef CONFIG_HOTPLUG_CPU - static int blk_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -3486,8 +3484,6 @@ static struct notifier_block __devinitdata blk_cpu_notifier = { .notifier_call = blk_cpu_notify, }; -#endif /* CONFIG_HOTPLUG_CPU */ - /** * blk_complete_request - end I/O on a request * @req: the request being processed diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 3d12b85b0962..067a9e8bc377 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -108,7 +108,6 @@ static int __cpuinit topology_add_dev(unsigned int cpu) return rc; } -#ifdef CONFIG_HOTPLUG_CPU static void __cpuinit topology_remove_dev(unsigned int cpu) { struct sys_device *sys_dev = get_cpu_sysdev(cpu); @@ -136,7 +135,6 @@ static int __cpuinit topology_cpu_callback(struct notifier_block *nfb, } return rc ? NOTIFY_BAD : NOTIFY_OK; } -#endif static int __cpuinit topology_sysfs_init(void) { diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 7a7c6e6dfe4f..47ab42db122a 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1537,7 +1537,6 @@ int cpufreq_update_policy(unsigned int cpu) } EXPORT_SYMBOL(cpufreq_update_policy); -#ifdef CONFIG_HOTPLUG_CPU static int cpufreq_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { @@ -1577,7 +1576,6 @@ static struct notifier_block __cpuinitdata cpufreq_cpu_notifier = { .notifier_call = cpufreq_cpu_callback, }; -#endif /* CONFIG_HOTPLUG_CPU */ /********************************************************************* * REGISTER / UNREGISTER CPUFREQ DRIVER * diff --git a/fs/buffer.c b/fs/buffer.c index a8ca0ac21488..517860f2d75b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2972,7 +2972,6 @@ init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags) } } -#ifdef CONFIG_HOTPLUG_CPU static void buffer_exit_cpu(int cpu) { int i; @@ -2994,7 +2993,6 @@ static int buffer_cpu_notify(struct notifier_block *self, buffer_exit_cpu((unsigned long)hcpu); return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init buffer_init(void) { diff --git a/include/linux/cpu.h b/include/linux/cpu.h index f02d71bf6894..71dc6ba4f73f 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -89,9 +89,9 @@ int cpu_down(unsigned int cpu); #define lock_cpu_hotplug() do { } while (0) #define unlock_cpu_hotplug() do { } while (0) #define lock_cpu_hotplug_interruptible() 0 -#define hotcpu_notifier(fn, pri) do { } while (0) -#define register_hotcpu_notifier(nb) do { } while (0) -#define unregister_hotcpu_notifier(nb) do { } while (0) +#define hotcpu_notifier(fn, pri) do { (void)(fn); } while (0) +#define register_hotcpu_notifier(nb) do { (void)(nb); } while (0) +#define unregister_hotcpu_notifier(nb) do { (void)(nb); } while (0) /* CPUs don't go offline once they're online w/o CONFIG_HOTPLUG_CPU */ static inline int cpu_is_offline(int cpu) { return 0; } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index bd1e89c4c96a..9b62b4c03ad0 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2044,7 +2044,6 @@ out: return err; } -#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG) /* * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, @@ -2108,9 +2107,7 @@ static void common_cpu_mem_hotplug_unplug(void) mutex_unlock(&callback_mutex); mutex_unlock(&manage_mutex); } -#endif -#ifdef CONFIG_HOTPLUG_CPU /* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent @@ -2127,7 +2124,6 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb, common_cpu_mem_hotplug_unplug(); return 0; } -#endif #ifdef CONFIG_MEMORY_HOTPLUG /* diff --git a/kernel/profile.c b/kernel/profile.c index 04fd84e8cdbe..0961d93e1d91 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -319,7 +319,6 @@ out: put_cpu(); } -#ifdef CONFIG_HOTPLUG_CPU static int __devinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) { @@ -372,10 +371,10 @@ static int __devinit profile_cpu_callback(struct notifier_block *info, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ #else /* !CONFIG_SMP */ #define profile_flip_buffers() do { } while (0) #define profile_discard_flip_buffers() do { } while (0) +#define profile_cpu_callback NULL void profile_hits(int type, void *__pc, unsigned int nr_hits) { diff --git a/kernel/sched.c b/kernel/sched.c index 75a005ed4eda..c83f531c2886 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6740,8 +6740,6 @@ SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, sched_smt_power_savings_store); #endif - -#ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains * and groups cannot be updated in place without racing with the balancing @@ -6774,7 +6772,6 @@ static int update_sched_domains(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void __init sched_init_smp(void) { diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 5484d6e045c2..c5257316f4b9 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -655,7 +655,6 @@ int current_is_keventd(void) } -#ifdef CONFIG_HOTPLUG_CPU /* Take the work from this (downed) CPU. */ static void take_over_work(struct workqueue_struct *wq, unsigned int cpu) { @@ -738,7 +737,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void init_workqueues(void) { diff --git a/lib/radix-tree.c b/lib/radix-tree.c index e2cefabb5aa0..d69ddbe43865 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -996,7 +996,6 @@ static __init void radix_tree_init_maxindex(void) height_to_maxindex[i] = __maxindex(i); } -#ifdef CONFIG_HOTPLUG_CPU static int radix_tree_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -1016,7 +1015,6 @@ static int radix_tree_callback(struct notifier_block *nfb, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init radix_tree_init(void) { diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2273952300d4..27ec7a1b8022 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -701,7 +701,6 @@ void drain_node_pages(int nodeid) } #endif -#if defined(CONFIG_PM) || defined(CONFIG_HOTPLUG_CPU) static void __drain_pages(unsigned int cpu) { unsigned long flags; @@ -723,7 +722,6 @@ static void __drain_pages(unsigned int cpu) } } } -#endif /* CONFIG_PM || CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_PM @@ -2907,7 +2905,6 @@ void __init free_area_init(unsigned long *zones_size) __pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } -#ifdef CONFIG_HOTPLUG_CPU static int page_alloc_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -2922,7 +2919,6 @@ static int page_alloc_cpu_notify(struct notifier_block *self, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init page_alloc_init(void) { diff --git a/mm/swap.c b/mm/swap.c index 017e72ca9bbb..2ed7be39795e 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -514,5 +514,7 @@ void __init swap_setup(void) * Right now other parts of the system means that we * _really_ don't want to cluster much more */ +#ifdef CONFIG_HOTPLUG_CPU hotcpu_notifier(cpu_swap_callback, 0); +#endif } diff --git a/mm/vmscan.c b/mm/vmscan.c index f6616e81fac7..093f5fe6dd77 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1513,7 +1513,6 @@ out: } #endif -#ifdef CONFIG_HOTPLUG_CPU /* It's optimal to keep kswapds on the same CPUs as their memory, but not required for correctness. So if the last cpu in a node goes away, we get changed to run anywhere: as the first one comes back, @@ -1534,7 +1533,6 @@ static int __devinit cpu_callback(struct notifier_block *nfb, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ /* * This kswapd start function will be called by init and node-hot-add. diff --git a/net/core/dev.c b/net/core/dev.c index 59d058a3b504..e660cb57e42a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3340,7 +3340,6 @@ void unregister_netdev(struct net_device *dev) EXPORT_SYMBOL(unregister_netdev); -#ifdef CONFIG_HOTPLUG_CPU static int dev_cpu_callback(struct notifier_block *nfb, unsigned long action, void *ocpu) @@ -3384,7 +3383,6 @@ static int dev_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_NET_DMA /** diff --git a/net/core/flow.c b/net/core/flow.c index 104c25d00a1d..d137f971f97d 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -340,7 +340,6 @@ static void __devinit flow_cache_cpu_prepare(int cpu) tasklet_init(tasklet, flow_cache_flush_tasklet, 0); } -#ifdef CONFIG_HOTPLUG_CPU static int flow_cache_cpu(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -349,7 +348,6 @@ static int flow_cache_cpu(struct notifier_block *nfb, __flow_cache_shrink((unsigned long)hcpu, 0); return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ static int __init flow_cache_init(void) { -- cgit v1.2.3 From ebe7e5fe4b41deeb2731c5b52d8c8e6ac08b1f74 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:38:21 -0800 Subject: [PATCH] remove kernel/lockdep.c:lockdep_internal Remove the no longer used lockdep_internal(). Signed-off-by: Adrian Bunk Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/lockdep.h | 6 ------ kernel/lockdep.c | 7 ------- 2 files changed, 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index da19aeb0dbe8..498bfbd3b4e1 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -193,7 +193,6 @@ extern void lockdep_free_key_range(void *start, unsigned long size); extern void lockdep_off(void); extern void lockdep_on(void); -extern int lockdep_internal(void); /* * These methods are used by specific locking variants (spinlocks, @@ -255,11 +254,6 @@ static inline void lockdep_on(void) { } -static inline int lockdep_internal(void) -{ - return 0; -} - # define lock_acquire(l, s, t, r, c, i) do { } while (0) # define lock_release(l, n, i) do { } while (0) # define lockdep_init() do { } while (0) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 300d61bb314b..3926c3674354 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -140,13 +140,6 @@ void lockdep_on(void) EXPORT_SYMBOL(lockdep_on); -int lockdep_internal(void) -{ - return current->lockdep_recursion != 0; -} - -EXPORT_SYMBOL(lockdep_internal); - /* * Debugging switches: */ -- cgit v1.2.3 From d3228a887cae75ef2b8b1211c31c539bef5a5698 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:38:22 -0800 Subject: [PATCH] make kernel/signal.c:kill_proc_info() static Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 - kernel/signal.c | 3 +-- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 0a90cefb0b0d..3a767242e72f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1305,7 +1305,6 @@ extern int kill_pgrp(struct pid *pid, int sig, int priv); extern int kill_pid(struct pid *pid, int sig, int priv); extern int __kill_pg_info(int sig, struct siginfo *info, pid_t pgrp); extern int kill_pg_info(int, struct siginfo *, pid_t); -extern int kill_proc_info(int, struct siginfo *, pid_t); extern void do_notify_parent(struct task_struct *, int); extern void force_sig(int, struct task_struct *); extern void force_sig_specific(int, struct task_struct *); diff --git a/kernel/signal.c b/kernel/signal.c index bc972d7e6319..ec81defde339 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1134,8 +1134,7 @@ int kill_pid_info(int sig, struct siginfo *info, struct pid *pid) return error; } -int -kill_proc_info(int sig, struct siginfo *info, pid_t pid) +static int kill_proc_info(int sig, struct siginfo *info, pid_t pid) { int error; rcu_read_lock(); -- cgit v1.2.3 From d394e122bc1adba0f3eb1ebec1cedb8a8c524741 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:38:26 -0800 Subject: [PATCH] make fs/jbd/transaction.c:__journal_temp_unlink_buffer() static Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd/transaction.c | 4 +++- include/linux/jbd.h | 1 - 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 4f82bcd63e48..d38e0d575e48 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -27,6 +27,8 @@ #include #include +static void __journal_temp_unlink_buffer(struct journal_head *jh); + /* * get_transaction: obtain a new transaction_t object. * @@ -1499,7 +1501,7 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh) * * Called under j_list_lock. The journal may not be locked. */ -void __journal_temp_unlink_buffer(struct journal_head *jh) +static void __journal_temp_unlink_buffer(struct journal_head *jh) { struct journal_head **list = NULL; transaction_t *transaction; diff --git a/include/linux/jbd.h b/include/linux/jbd.h index dacd566c9f6c..452737551260 100644 --- a/include/linux/jbd.h +++ b/include/linux/jbd.h @@ -839,7 +839,6 @@ struct journal_s */ /* Filing buffers */ -extern void __journal_temp_unlink_buffer(struct journal_head *jh); extern void journal_unfile_buffer(journal_t *, struct journal_head *); extern void __journal_unfile_buffer(struct journal_head *); extern void __journal_refile_buffer(struct journal_head *); -- cgit v1.2.3 From 7ddae86095794cce4364740edd8463c77654a265 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:38:27 -0800 Subject: [PATCH] make fs/jbd2/transaction.c:__kbd2_journal_temp_unlink_buffer() static Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/jbd2/transaction.c | 2 ++ include/linux/jbd2.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index c051a94c8a97..3a8700153cb0 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -27,6 +27,8 @@ #include #include +static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); + /* * jbd2_get_transaction: obtain a new transaction_t object. * diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 98a0ae52660f..0e0fedd2039a 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -848,7 +848,6 @@ struct journal_s */ /* Filing buffers */ -extern void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); extern void jbd2_journal_unfile_buffer(journal_t *, struct journal_head *); extern void __jbd2_journal_unfile_buffer(struct journal_head *); extern void __jbd2_journal_refile_buffer(struct journal_head *); -- cgit v1.2.3 From c585646dd1d98caf0a5f2e85c794c1441df6fac1 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:38:29 -0800 Subject: [PATCH] fs/lockd/host.c: make 2 functions static Make the following needlessly global functions static: - nlm_lookup_host() - nsm_find() Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/host.c | 53 ++++++++++++++++++++++++--------------------- include/linux/lockd/lockd.h | 2 -- 2 files changed, 28 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/host.c b/fs/lockd/host.c index fb24a9730345..3d4610c2a266 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -36,34 +36,14 @@ static DEFINE_MUTEX(nlm_host_mutex); static void nlm_gc_hosts(void); static struct nsm_handle * __nsm_find(const struct sockaddr_in *, const char *, int, int); - -/* - * Find an NLM server handle in the cache. If there is none, create it. - */ -struct nlm_host * -nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, - const char *hostname, int hostname_len) -{ - return nlm_lookup_host(0, sin, proto, version, - hostname, hostname_len); -} - -/* - * Find an NLM client handle in the cache. If there is none, create it. - */ -struct nlm_host * -nlmsvc_lookup_host(struct svc_rqst *rqstp, - const char *hostname, int hostname_len) -{ - return nlm_lookup_host(1, &rqstp->rq_addr, - rqstp->rq_prot, rqstp->rq_vers, - hostname, hostname_len); -} +static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, + const char *hostname, + int hostname_len); /* * Common host lookup routine for server & client */ -struct nlm_host * +static struct nlm_host * nlm_lookup_host(int server, const struct sockaddr_in *sin, int proto, int version, const char *hostname, @@ -194,6 +174,29 @@ nlm_destroy_host(struct nlm_host *host) kfree(host); } +/* + * Find an NLM server handle in the cache. If there is none, create it. + */ +struct nlm_host * +nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, + const char *hostname, int hostname_len) +{ + return nlm_lookup_host(0, sin, proto, version, + hostname, hostname_len); +} + +/* + * Find an NLM client handle in the cache. If there is none, create it. + */ +struct nlm_host * +nlmsvc_lookup_host(struct svc_rqst *rqstp, + const char *hostname, int hostname_len) +{ + return nlm_lookup_host(1, &rqstp->rq_addr, + rqstp->rq_prot, rqstp->rq_vers, + hostname, hostname_len); +} + /* * Create the NLM RPC client for an NLM peer */ @@ -495,7 +498,7 @@ out: return nsm; } -struct nsm_handle * +static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, const char *hostname, int hostname_len) { return __nsm_find(sin, hostname, hostname_len, 1); diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 862d9730a60d..8c39654549d8 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -164,14 +164,12 @@ void nlmclnt_next_cookie(struct nlm_cookie *); */ struct nlm_host * nlmclnt_lookup_host(const struct sockaddr_in *, int, int, const char *, int); struct nlm_host * nlmsvc_lookup_host(struct svc_rqst *, const char *, int); -struct nlm_host * nlm_lookup_host(int server, const struct sockaddr_in *, int, int, const char *, int); struct rpc_clnt * nlm_bind_host(struct nlm_host *); void nlm_rebind_host(struct nlm_host *); struct nlm_host * nlm_get_host(struct nlm_host *); void nlm_release_host(struct nlm_host *); void nlm_shutdown_hosts(void); extern void nlm_host_rebooted(const struct sockaddr_in *, const char *, int, u32); -struct nsm_handle *nsm_find(const struct sockaddr_in *, const char *, int); void nsm_release(struct nsm_handle *); -- cgit v1.2.3 From d9489fb60614794cbca4b6b173c60ed9388974c6 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 6 Dec 2006 20:38:43 -0800 Subject: [PATCH] kernel-doc: fix fusion and i2o docs Correct lots of typos, kernel-doc warnings, & kernel-doc usage in fusion and i2o drivers. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/fusion/mptbase.c | 255 ++++++++++++++++++++------------------ drivers/message/fusion/mptfc.c | 5 +- drivers/message/fusion/mptscsih.c | 24 ++-- drivers/message/fusion/mptspi.c | 4 +- drivers/message/i2o/bus-osm.c | 3 + drivers/message/i2o/device.c | 8 +- drivers/message/i2o/driver.c | 20 ++- drivers/message/i2o/exec-osm.c | 10 +- drivers/message/i2o/i2o_block.c | 15 ++- drivers/message/i2o/i2o_proc.c | 2 +- drivers/message/i2o/i2o_scsi.c | 11 +- drivers/message/i2o/pci.c | 5 +- include/linux/i2o.h | 18 +-- 13 files changed, 203 insertions(+), 177 deletions(-) (limited to 'include/linux') diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index 051b7c5b8f03..6e068cf1049b 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -347,7 +347,7 @@ mpt_reply(MPT_ADAPTER *ioc, u32 pa) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_interrupt - MPT adapter (IOC) specific interrupt handler. * @irq: irq number (not used) * @bus_id: bus identifier cookie == pointer to MPT_ADAPTER structure @@ -387,14 +387,16 @@ mpt_interrupt(int irq, void *bus_id) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * mpt_base_reply - MPT base driver's callback routine; all base driver - * "internal" request/reply processing is routed here. - * Currently used for EventNotification and EventAck handling. +/** + * mpt_base_reply - MPT base driver's callback routine * @ioc: Pointer to MPT_ADAPTER structure * @mf: Pointer to original MPT request frame * @reply: Pointer to MPT reply frame (NULL if TurboReply) * + * MPT base driver's callback routine; all base driver + * "internal" request/reply processing is routed here. + * Currently used for EventNotification and EventAck handling. + * * Returns 1 indicating original alloc'd request frame ptr * should be freed, or 0 if it shouldn't. */ @@ -530,7 +532,7 @@ mpt_base_reply(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *reply) * @dclass: Protocol driver's class (%MPT_DRIVER_CLASS enum value) * * This routine is called by a protocol-specific driver (SCSI host, - * LAN, SCSI target) to register it's reply callback routine. Each + * LAN, SCSI target) to register its reply callback routine. Each * protocol-specific driver must do this before it will be able to * use any IOC resources, such as obtaining request frames. * @@ -572,7 +574,7 @@ mpt_register(MPT_CALLBACK cbfunc, MPT_DRIVER_CLASS dclass) * mpt_deregister - Deregister a protocol drivers resources. * @cb_idx: previously registered callback handle * - * Each protocol-specific driver should call this routine when it's + * Each protocol-specific driver should call this routine when its * module is unloaded. */ void @@ -617,7 +619,7 @@ mpt_event_register(int cb_idx, MPT_EVHANDLER ev_cbfunc) * * Each protocol-specific driver should call this routine * when it does not (or can no longer) handle events, - * or when it's module is unloaded. + * or when its module is unloaded. */ void mpt_event_deregister(int cb_idx) @@ -656,7 +658,7 @@ mpt_reset_register(int cb_idx, MPT_RESETHANDLER reset_func) * * Each protocol-specific driver should call this routine * when it does not (or can no longer) handle IOC reset handling, - * or when it's module is unloaded. + * or when its module is unloaded. */ void mpt_reset_deregister(int cb_idx) @@ -670,6 +672,8 @@ mpt_reset_deregister(int cb_idx) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mpt_device_driver_register - Register device driver hooks + * @dd_cbfunc: driver callbacks struct + * @cb_idx: MPT protocol driver index */ int mpt_device_driver_register(struct mpt_pci_driver * dd_cbfunc, int cb_idx) @@ -696,6 +700,7 @@ mpt_device_driver_register(struct mpt_pci_driver * dd_cbfunc, int cb_idx) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mpt_device_driver_deregister - DeRegister device driver hooks + * @cb_idx: MPT protocol driver index */ void mpt_device_driver_deregister(int cb_idx) @@ -887,8 +892,7 @@ mpt_add_sge(char *pAddr, u32 flagslength, dma_addr_t dma_addr) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mpt_send_handshake_request - Send MPT request via doorbell - * handshake method. + * mpt_send_handshake_request - Send MPT request via doorbell handshake method. * @handle: Handle of registered MPT protocol driver * @ioc: Pointer to MPT adapter structure * @reqBytes: Size of the request in bytes @@ -981,10 +985,13 @@ mpt_send_handshake_request(int handle, MPT_ADAPTER *ioc, int reqBytes, u32 *req, /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mpt_host_page_access_control - provides mechanism for the host - * driver to control the IOC's Host Page Buffer access. + * mpt_host_page_access_control - control the IOC's Host Page Buffer access * @ioc: Pointer to MPT adapter structure * @access_control_value: define bits below + * @sleepFlag: Specifies whether the process can sleep + * + * Provides mechanism for the host driver to control the IOC's + * Host Page Buffer access. * * Access Control Value - bits[15:12] * 0h Reserved @@ -1022,10 +1029,10 @@ mpt_host_page_access_control(MPT_ADAPTER *ioc, u8 access_control_value, int slee /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mpt_host_page_alloc - allocate system memory for the fw - * If we already allocated memory in past, then resend the same pointer. - * ioc@: Pointer to pointer to IOC adapter - * ioc_init@: Pointer to ioc init config page + * @ioc: Pointer to pointer to IOC adapter + * @ioc_init: Pointer to ioc init config page * + * If we already allocated memory in past, then resend the same pointer. * Returns 0 for success, non-zero for failure. */ static int @@ -1091,12 +1098,15 @@ return 0; /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mpt_verify_adapter - Given a unique IOC identifier, set pointer to - * the associated MPT adapter structure. + * mpt_verify_adapter - Given IOC identifier, set pointer to its adapter structure. * @iocid: IOC unique identifier (integer) * @iocpp: Pointer to pointer to IOC adapter * - * Returns iocid and sets iocpp. + * Given a unique IOC identifier, set pointer to the associated MPT + * adapter structure. + * + * Returns iocid and sets iocpp if iocid is found. + * Returns -1 if iocid is not found. */ int mpt_verify_adapter(int iocid, MPT_ADAPTER **iocpp) @@ -1115,9 +1125,10 @@ mpt_verify_adapter(int iocid, MPT_ADAPTER **iocpp) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_attach - Install a PCI intelligent MPT adapter. * @pdev: Pointer to pci_dev structure + * @id: PCI device ID information * * This routine performs all the steps necessary to bring the IOC of * a MPT adapter to a OPERATIONAL state. This includes registering @@ -1417,10 +1428,9 @@ mpt_attach(struct pci_dev *pdev, const struct pci_device_id *id) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_detach - Remove a PCI intelligent MPT adapter. * @pdev: Pointer to pci_dev structure - * */ void @@ -1466,10 +1476,10 @@ mpt_detach(struct pci_dev *pdev) */ #ifdef CONFIG_PM /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_suspend - Fusion MPT base driver suspend routine. - * - * + * @pdev: Pointer to pci_dev structure + * @state: new state to enter */ int mpt_suspend(struct pci_dev *pdev, pm_message_t state) @@ -1505,10 +1515,9 @@ mpt_suspend(struct pci_dev *pdev, pm_message_t state) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_resume - Fusion MPT base driver resume routine. - * - * + * @pdev: Pointer to pci_dev structure */ int mpt_resume(struct pci_dev *pdev) @@ -1566,7 +1575,7 @@ mpt_signal_reset(int index, MPT_ADAPTER *ioc, int reset_phase) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_do_ioc_recovery - Initialize or recover MPT adapter. * @ioc: Pointer to MPT adapter structure * @reason: Event word / reason @@ -1892,13 +1901,15 @@ mpt_do_ioc_recovery(MPT_ADAPTER *ioc, u32 reason, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * mpt_detect_bound_ports - Search for PCI bus/dev_function - * which matches PCI bus/dev_function (+/-1) for newly discovered 929, - * 929X, 1030 or 1035. +/** + * mpt_detect_bound_ports - Search for matching PCI bus/dev_function * @ioc: Pointer to MPT adapter structure * @pdev: Pointer to (struct pci_dev) structure * + * Search for PCI bus/dev_function which matches + * PCI bus/dev_function (+/-1) for newly discovered 929, + * 929X, 1030 or 1035. + * * If match on PCI dev_function +/-1 is found, bind the two MPT adapters * using alt_ioc pointer fields in their %MPT_ADAPTER structures. */ @@ -1945,9 +1956,9 @@ mpt_detect_bound_ports(MPT_ADAPTER *ioc, struct pci_dev *pdev) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_adapter_disable - Disable misbehaving MPT adapter. - * @this: Pointer to MPT adapter structure + * @ioc: Pointer to MPT adapter structure */ static void mpt_adapter_disable(MPT_ADAPTER *ioc) @@ -2046,9 +2057,8 @@ mpt_adapter_disable(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * mpt_adapter_dispose - Free all resources associated with a MPT - * adapter. +/** + * mpt_adapter_dispose - Free all resources associated with an MPT adapter * @ioc: Pointer to MPT adapter structure * * This routine unregisters h/w resources and frees all alloc'd memory @@ -2099,8 +2109,8 @@ mpt_adapter_dispose(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * MptDisplayIocCapabilities - Disply IOC's capacilities. +/** + * MptDisplayIocCapabilities - Disply IOC's capabilities. * @ioc: Pointer to MPT adapter structure */ static void @@ -2142,7 +2152,7 @@ MptDisplayIocCapabilities(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * MakeIocReady - Get IOC to a READY state, using KickStart if needed. * @ioc: Pointer to MPT_ADAPTER structure * @force: Force hard KickStart of IOC @@ -2279,7 +2289,7 @@ MakeIocReady(MPT_ADAPTER *ioc, int force, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_GetIocState - Get the current state of a MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @cooked: Request raw or cooked IOC state @@ -2304,7 +2314,7 @@ mpt_GetIocState(MPT_ADAPTER *ioc, int cooked) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * GetIocFacts - Send IOCFacts request to MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @sleepFlag: Specifies whether the process can sleep @@ -2478,7 +2488,7 @@ GetIocFacts(MPT_ADAPTER *ioc, int sleepFlag, int reason) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * GetPortFacts - Send PortFacts request to MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @portnum: Port number @@ -2545,7 +2555,7 @@ GetPortFacts(MPT_ADAPTER *ioc, int portnum, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * SendIocInit - Send IOCInit request to MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @sleepFlag: Specifies whether the process can sleep @@ -2630,7 +2640,7 @@ SendIocInit(MPT_ADAPTER *ioc, int sleepFlag) } /* No need to byte swap the multibyte fields in the reply - * since we don't even look at it's contents. + * since we don't even look at its contents. */ dhsprintk((MYIOC_s_INFO_FMT "Sending PortEnable (req @ %p)\n", @@ -2672,7 +2682,7 @@ SendIocInit(MPT_ADAPTER *ioc, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * SendPortEnable - Send PortEnable request to MPT adapter port. * @ioc: Pointer to MPT_ADAPTER structure * @portnum: Port number to enable @@ -2723,9 +2733,13 @@ SendPortEnable(MPT_ADAPTER *ioc, int portnum, int sleepFlag) return rc; } -/* - * ioc: Pointer to MPT_ADAPTER structure - * size - total FW bytes +/** + * mpt_alloc_fw_memory - allocate firmware memory + * @ioc: Pointer to MPT_ADAPTER structure + * @size: total FW bytes + * + * If memory has already been allocated, the same (cached) value + * is returned. */ void mpt_alloc_fw_memory(MPT_ADAPTER *ioc, int size) @@ -2742,9 +2756,12 @@ mpt_alloc_fw_memory(MPT_ADAPTER *ioc, int size) ioc->alloc_total += size; } } -/* - * If alt_img is NULL, delete from ioc structure. - * Else, delete a secondary image in same format. +/** + * mpt_free_fw_memory - free firmware memory + * @ioc: Pointer to MPT_ADAPTER structure + * + * If alt_img is NULL, delete from ioc structure. + * Else, delete a secondary image in same format. */ void mpt_free_fw_memory(MPT_ADAPTER *ioc) @@ -2763,7 +2780,7 @@ mpt_free_fw_memory(MPT_ADAPTER *ioc) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_do_upload - Construct and Send FWUpload request to MPT adapter port. * @ioc: Pointer to MPT_ADAPTER structure * @sleepFlag: Specifies whether the process can sleep @@ -2865,10 +2882,10 @@ mpt_do_upload(MPT_ADAPTER *ioc, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_downloadboot - DownloadBoot code * @ioc: Pointer to MPT_ADAPTER structure - * @flag: Specify which part of IOC memory is to be uploaded. + * @pFwHeader: Pointer to firmware header info * @sleepFlag: Specifies whether the process can sleep * * FwDownloadBoot requires Programmed IO access. @@ -3071,7 +3088,7 @@ mpt_downloadboot(MPT_ADAPTER *ioc, MpiFwHeader_t *pFwHeader, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * KickStart - Perform hard reset of MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @force: Force hard reset @@ -3145,12 +3162,12 @@ KickStart(MPT_ADAPTER *ioc, int force, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_diag_reset - Perform hard reset of the adapter. * @ioc: Pointer to MPT_ADAPTER structure * @ignore: Set if to honor and clear to ignore * the reset history bit - * @sleepflag: CAN_SLEEP if called in a non-interrupt thread, + * @sleepFlag: CAN_SLEEP if called in a non-interrupt thread, * else set to NO_SLEEP (use mdelay instead) * * This routine places the adapter in diagnostic mode via the @@ -3436,11 +3453,12 @@ mpt_diag_reset(MPT_ADAPTER *ioc, int ignore, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * SendIocReset - Send IOCReset request to MPT adapter. * @ioc: Pointer to MPT_ADAPTER structure * @reset_type: reset type, expected values are * %MPI_FUNCTION_IOC_MESSAGE_UNIT_RESET or %MPI_FUNCTION_IO_UNIT_RESET + * @sleepFlag: Specifies whether the process can sleep * * Send IOCReset request to the MPT adapter. * @@ -3494,11 +3512,12 @@ SendIocReset(MPT_ADAPTER *ioc, u8 reset_type, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * initChainBuffers - Allocate memory for and initialize - * chain buffers, chain buffer control arrays and spinlock. - * @hd: Pointer to MPT_SCSI_HOST structure - * @init: If set, initialize the spin lock. +/** + * initChainBuffers - Allocate memory for and initialize chain buffers + * @ioc: Pointer to MPT_ADAPTER structure + * + * Allocates memory for and initializes chain buffers, + * chain buffer control arrays and spinlock. */ static int initChainBuffers(MPT_ADAPTER *ioc) @@ -3594,7 +3613,7 @@ initChainBuffers(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * PrimeIocFifos - Initialize IOC request and reply FIFOs. * @ioc: Pointer to MPT_ADAPTER structure * @@ -3891,15 +3910,15 @@ mpt_handshake_req_reply_wait(MPT_ADAPTER *ioc, int reqBytes, u32 *req, } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * WaitForDoorbellAck - Wait for IOC to clear the IOP_DOORBELL_STATUS bit - * in it's IntStatus register. +/** + * WaitForDoorbellAck - Wait for IOC doorbell handshake acknowledge * @ioc: Pointer to MPT_ADAPTER structure * @howlong: How long to wait (in seconds) * @sleepFlag: Specifies whether the process can sleep * * This routine waits (up to ~2 seconds max) for IOC doorbell - * handshake ACKnowledge. + * handshake ACKnowledge, indicated by the IOP_DOORBELL_STATUS + * bit in its IntStatus register being clear. * * Returns a negative value on failure, else wait loop count. */ @@ -3942,14 +3961,14 @@ WaitForDoorbellAck(MPT_ADAPTER *ioc, int howlong, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * WaitForDoorbellInt - Wait for IOC to set the HIS_DOORBELL_INTERRUPT bit - * in it's IntStatus register. +/** + * WaitForDoorbellInt - Wait for IOC to set its doorbell interrupt bit * @ioc: Pointer to MPT_ADAPTER structure * @howlong: How long to wait (in seconds) * @sleepFlag: Specifies whether the process can sleep * - * This routine waits (up to ~2 seconds max) for IOC doorbell interrupt. + * This routine waits (up to ~2 seconds max) for IOC doorbell interrupt + * (MPI_HIS_DOORBELL_INTERRUPT) to be set in the IntStatus register. * * Returns a negative value on failure, else wait loop count. */ @@ -3991,8 +4010,8 @@ WaitForDoorbellInt(MPT_ADAPTER *ioc, int howlong, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * WaitForDoorbellReply - Wait for and capture a IOC handshake reply. +/** + * WaitForDoorbellReply - Wait for and capture an IOC handshake reply. * @ioc: Pointer to MPT_ADAPTER structure * @howlong: How long to wait (in seconds) * @sleepFlag: Specifies whether the process can sleep @@ -4077,7 +4096,7 @@ WaitForDoorbellReply(MPT_ADAPTER *ioc, int howlong, int sleepFlag) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * GetLanConfigPages - Fetch LANConfig pages. * @ioc: Pointer to MPT_ADAPTER structure * @@ -4188,12 +4207,9 @@ GetLanConfigPages(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * mptbase_sas_persist_operation - Perform operation on SAS Persitent Table +/** + * mptbase_sas_persist_operation - Perform operation on SAS Persistent Table * @ioc: Pointer to MPT_ADAPTER structure - * @sas_address: 64bit SAS Address for operation. - * @target_id: specified target for operation - * @bus: specified bus for operation * @persist_opcode: see below * * MPI_SAS_OP_CLEAR_NOT_PRESENT - Free all persist TargetID mappings for @@ -4202,7 +4218,7 @@ GetLanConfigPages(MPT_ADAPTER *ioc) * * NOTE: Don't use not this function during interrupt time. * - * Returns: 0 for success, non-zero error + * Returns 0 for success, non-zero error */ /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ @@ -4399,7 +4415,7 @@ mptbase_raid_process_event_data(MPT_ADAPTER *ioc, } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * GetIoUnitPage2 - Retrieve BIOS version and boot order information. * @ioc: Pointer to MPT_ADAPTER structure * @@ -4457,7 +4473,8 @@ GetIoUnitPage2(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* mpt_GetScsiPortSettings - read SCSI Port Page 0 and 2 +/** + * mpt_GetScsiPortSettings - read SCSI Port Page 0 and 2 * @ioc: Pointer to a Adapter Strucutre * @portnum: IOC port number * @@ -4644,7 +4661,8 @@ mpt_GetScsiPortSettings(MPT_ADAPTER *ioc, int portnum) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* mpt_readScsiDevicePageHeaders - save version and length of SDP1 +/** + * mpt_readScsiDevicePageHeaders - save version and length of SDP1 * @ioc: Pointer to a Adapter Strucutre * @portnum: IOC port number * @@ -4996,9 +5014,8 @@ mpt_read_ioc_pg_1(MPT_ADAPTER *ioc) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * SendEventNotification - Send EventNotification (on or off) request - * to MPT adapter. +/** + * SendEventNotification - Send EventNotification (on or off) request to adapter * @ioc: Pointer to MPT_ADAPTER structure * @EvSwitch: Event switch flags */ @@ -5062,8 +5079,8 @@ SendEventAck(MPT_ADAPTER *ioc, EventNotificationReply_t *evnp) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mpt_config - Generic function to issue config message - * @ioc - Pointer to an adapter structure - * @cfg - Pointer to a configuration structure. Struct contains + * @ioc: Pointer to an adapter structure + * @pCfg: Pointer to a configuration structure. Struct contains * action, page address, direction, physical address * and pointer to a configuration page header * Page header is updated. @@ -5188,8 +5205,8 @@ mpt_config(MPT_ADAPTER *ioc, CONFIGPARMS *pCfg) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * mpt_timer_expired - Call back for timer process. +/** + * mpt_timer_expired - Callback for timer process. * Used only internal config functionality. * @data: Pointer to MPT_SCSI_HOST recast as an unsigned long */ @@ -5214,12 +5231,12 @@ mpt_timer_expired(unsigned long data) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_ioc_reset - Base cleanup for hard reset * @ioc: Pointer to the adapter structure * @reset_phase: Indicates pre- or post-reset functionality * - * Remark: Free's resources with internally generated commands. + * Remark: Frees resources with internally generated commands. */ static int mpt_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) @@ -5271,7 +5288,7 @@ mpt_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) * procfs (%MPT_PROCFS_MPTBASEDIR/...) support stuff... */ /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * procmpt_create - Create %MPT_PROCFS_MPTBASEDIR entries. * * Returns 0 for success, non-zero for failure. @@ -5297,7 +5314,7 @@ procmpt_create(void) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * procmpt_destroy - Tear down %MPT_PROCFS_MPTBASEDIR entries. * * Returns 0 for success, non-zero for failure. @@ -5311,16 +5328,16 @@ procmpt_destroy(void) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * procmpt_summary_read - Handle read request from /proc/mpt/summary - * or from /proc/mpt/iocN/summary. +/** + * procmpt_summary_read - Handle read request of a summary file * @buf: Pointer to area to write information * @start: Pointer to start pointer * @offset: Offset to start writing - * @request: + * @request: Amount of read data requested * @eof: Pointer to EOF integer * @data: Pointer * + * Handles read request from /proc/mpt/summary or /proc/mpt/iocN/summary. * Returns number of characters written to process performing the read. */ static int @@ -5355,12 +5372,12 @@ procmpt_summary_read(char *buf, char **start, off_t offset, int request, int *eo } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * procmpt_version_read - Handle read request from /proc/mpt/version. * @buf: Pointer to area to write information * @start: Pointer to start pointer * @offset: Offset to start writing - * @request: + * @request: Amount of read data requested * @eof: Pointer to EOF integer * @data: Pointer * @@ -5411,12 +5428,12 @@ procmpt_version_read(char *buf, char **start, off_t offset, int request, int *eo } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * procmpt_iocinfo_read - Handle read request from /proc/mpt/iocN/info. * @buf: Pointer to area to write information * @start: Pointer to start pointer * @offset: Offset to start writing - * @request: + * @request: Amount of read data requested * @eof: Pointer to EOF integer * @data: Pointer * @@ -5577,16 +5594,17 @@ mpt_print_ioc_summary(MPT_ADAPTER *ioc, char *buffer, int *size, int len, int sh */ /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mpt_HardResetHandler - Generic reset handler, issue SCSI Task - * Management call based on input arg values. If TaskMgmt fails, - * return associated SCSI request. + * mpt_HardResetHandler - Generic reset handler * @ioc: Pointer to MPT_ADAPTER structure * @sleepFlag: Indicates if sleep or schedule must be called. * + * Issues SCSI Task Management call based on input arg values. + * If TaskMgmt fails, returns associated SCSI request. + * * Remark: _HardResetHandler can be invoked from an interrupt thread (timer) * or a non-interrupt thread. In the former, must not call schedule(). * - * Remark: A return of -1 is a FATAL error case, as it means a + * Note: A return of -1 is a FATAL error case, as it means a * FW reload/initialization failed. * * Returns 0 for SUCCESS or -1 if FAILED. @@ -5935,13 +5953,14 @@ EventDescriptionStr(u8 event, u32 evData0, char *evStr) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* - * ProcessEventNotification - Route a received EventNotificationReply to - * all currently regeistered event handlers. +/** + * ProcessEventNotification - Route EventNotificationReply to all event handlers * @ioc: Pointer to MPT_ADAPTER structure * @pEventReply: Pointer to EventNotification reply frame * @evHandlers: Pointer to integer, number of event handlers * + * Routes a received EventNotificationReply to all currently registered + * event handlers. * Returns sum of event handlers return values. */ static int @@ -6056,7 +6075,7 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_fc_log_info - Log information returned from Fibre Channel IOC. * @ioc: Pointer to MPT_ADAPTER structure * @log_info: U32 LogInfo reply word from the IOC @@ -6077,7 +6096,7 @@ mpt_fc_log_info(MPT_ADAPTER *ioc, u32 log_info) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_spi_log_info - Log information returned from SCSI Parallel IOC. * @ioc: Pointer to MPT_ADAPTER structure * @mr: Pointer to MPT reply frame @@ -6200,7 +6219,7 @@ mpt_spi_log_info(MPT_ADAPTER *ioc, u32 log_info) }; /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_sas_log_info - Log information returned from SAS IOC. * @ioc: Pointer to MPT_ADAPTER structure * @log_info: U32 LogInfo reply word from the IOC @@ -6255,7 +6274,7 @@ union loginfo_type { } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * mpt_sp_ioc_info - IOC information returned from SCSI Parallel IOC. * @ioc: Pointer to MPT_ADAPTER structure * @ioc_status: U32 IOCStatus word from IOC @@ -6416,7 +6435,7 @@ EXPORT_SYMBOL(mpt_free_fw_memory); EXPORT_SYMBOL(mptbase_sas_persist_operation); /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * fusion_init - Fusion MPT base driver initialization routine. * * Returns 0 for success, non-zero for failure. @@ -6456,7 +6475,7 @@ fusion_init(void) } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ -/* +/** * fusion_exit - Perform driver unload cleanup. * * This routine frees all resources associated with each MPT adapter diff --git a/drivers/message/fusion/mptfc.c b/drivers/message/fusion/mptfc.c index ef2b55e19910..ca2f9107f145 100644 --- a/drivers/message/fusion/mptfc.c +++ b/drivers/message/fusion/mptfc.c @@ -1395,8 +1395,7 @@ mptfc_ioc_reset(MPT_ADAPTER *ioc, int reset_phase) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mptfc_init - Register MPT adapter(s) as SCSI host(s) with - * linux scsi mid-layer. + * mptfc_init - Register MPT adapter(s) as SCSI host(s) with SCSI mid-layer. * * Returns 0 for success, non-zero for failure. */ @@ -1440,7 +1439,7 @@ mptfc_init(void) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mptfc_remove - Removed fc infrastructure for devices + * mptfc_remove - Remove fc infrastructure for devices * @pdev: Pointer to pci_dev structure * */ diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index 30524dc54b16..2c72c36b8171 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -1230,15 +1230,15 @@ mptscsih_host_info(MPT_ADAPTER *ioc, char *pbuf, off_t offset, int len) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mptscsih_proc_info - Return information about MPT adapter + * @host: scsi host struct + * @buffer: if write, user data; if read, buffer for user + * @start: returns the buffer address + * @offset: if write, 0; if read, the current offset into the buffer from + * the previous read. + * @length: if write, return length; + * @func: write = 1; read = 0 * * (linux scsi_host_template.info routine) - * - * buffer: if write, user data; if read, buffer for user - * length: if write, return length; - * offset: if write, 0; if read, the current offset into the buffer from - * the previous read. - * hostno: scsi host number - * func: if write = 1; if read = 0 */ int mptscsih_proc_info(struct Scsi_Host *host, char *buffer, char **start, off_t offset, @@ -1902,8 +1902,7 @@ mptscsih_bus_reset(struct scsi_cmnd * SCpnt) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mptscsih_host_reset - Perform a SCSI host adapter RESET! - * new_eh variant + * mptscsih_host_reset - Perform a SCSI host adapter RESET (new_eh variant) * @SCpnt: Pointer to scsi_cmnd structure, IO which reset is due to * * (linux scsi_host_template.eh_host_reset_handler routine) @@ -1949,8 +1948,7 @@ mptscsih_host_reset(struct scsi_cmnd *SCpnt) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mptscsih_tm_pending_wait - wait for pending task management request to - * complete. + * mptscsih_tm_pending_wait - wait for pending task management request to complete * @hd: Pointer to MPT host structure. * * Returns {SUCCESS,FAILED}. @@ -1982,6 +1980,7 @@ mptscsih_tm_pending_wait(MPT_SCSI_HOST * hd) /** * mptscsih_tm_wait_for_completion - wait for completion of TM task * @hd: Pointer to MPT host structure. + * @timeout: timeout in seconds * * Returns {SUCCESS,FAILED}. */ @@ -3429,8 +3428,7 @@ mptscsih_do_cmd(MPT_SCSI_HOST *hd, INTERNAL_CMD *io) /** * mptscsih_synchronize_cache - Send SYNCHRONIZE_CACHE to all disks. * @hd: Pointer to a SCSI HOST structure - * @vtarget: per device private data - * @lun: lun + * @vdevice: virtual target device * * Uses the ISR, but with special processing. * MUST be single-threaded. diff --git a/drivers/message/fusion/mptspi.c b/drivers/message/fusion/mptspi.c index f422c0d0621c..36641da59289 100644 --- a/drivers/message/fusion/mptspi.c +++ b/drivers/message/fusion/mptspi.c @@ -1100,8 +1100,7 @@ static struct pci_driver mptspi_driver = { /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** - * mptspi_init - Register MPT adapter(s) as SCSI host(s) with - * linux scsi mid-layer. + * mptspi_init - Register MPT adapter(s) as SCSI host(s) with SCSI mid-layer. * * Returns 0 for success, non-zero for failure. */ @@ -1135,7 +1134,6 @@ mptspi_init(void) /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ /** * mptspi_exit - Unregisters MPT adapter(s) - * */ static void __exit mptspi_exit(void) diff --git a/drivers/message/i2o/bus-osm.c b/drivers/message/i2o/bus-osm.c index d96c687aee93..c463dc2efc09 100644 --- a/drivers/message/i2o/bus-osm.c +++ b/drivers/message/i2o/bus-osm.c @@ -56,6 +56,9 @@ static int i2o_bus_scan(struct i2o_device *dev) /** * i2o_bus_store_scan - Scan the I2O Bus Adapter * @d: device which should be scanned + * @attr: device_attribute + * @buf: output buffer + * @count: buffer size * * Returns count. */ diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index 55757af4e3b2..b9df143e4ff1 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -54,8 +54,8 @@ static inline int i2o_device_issue_claim(struct i2o_device *dev, u32 cmd, * @dev: I2O device to claim * @drv: I2O driver which wants to claim the device * - * Do the leg work to assign a device to a given OSM. If the claim succeed - * the owner of the rimary. If the attempt fails a negative errno code + * Do the leg work to assign a device to a given OSM. If the claim succeeds, + * the owner is the primary. If the attempt fails a negative errno code * is returned. On success zero is returned. */ int i2o_device_claim(struct i2o_device *dev) @@ -208,7 +208,7 @@ static struct i2o_device *i2o_device_alloc(void) /** * i2o_device_add - allocate a new I2O device and add it to the IOP - * @iop: I2O controller where the device is on + * @c: I2O controller that the device is on * @entry: LCT entry of the I2O device * * Allocate a new I2O device and initialize it with the LCT entry. The @@ -280,7 +280,7 @@ err: /** * i2o_device_remove - remove an I2O device from the I2O core - * @dev: I2O device which should be released + * @i2o_dev: I2O device which should be released * * Is used on I2O controller removal or LCT modification, when the device * is removed from the system. Note that the device could still hang diff --git a/drivers/message/i2o/driver.c b/drivers/message/i2o/driver.c index 7fc7399bd2ec..9104b65ff70f 100644 --- a/drivers/message/i2o/driver.c +++ b/drivers/message/i2o/driver.c @@ -34,9 +34,7 @@ static spinlock_t i2o_drivers_lock; static struct i2o_driver **i2o_drivers; /** - * i2o_bus_match - Tell if a I2O device class id match the class ids of - * the I2O driver (OSM) - * + * i2o_bus_match - Tell if I2O device class id matches the class ids of the I2O driver (OSM) * @dev: device which should be verified * @drv: the driver to match against * @@ -248,7 +246,7 @@ int i2o_driver_dispatch(struct i2o_controller *c, u32 m) /** * i2o_driver_notify_controller_add_all - Send notify of added controller - * to all I2O drivers + * @c: newly added controller * * Send notifications to all registered drivers that a new controller was * added. @@ -267,8 +265,8 @@ void i2o_driver_notify_controller_add_all(struct i2o_controller *c) } /** - * i2o_driver_notify_controller_remove_all - Send notify of removed - * controller to all I2O drivers + * i2o_driver_notify_controller_remove_all - Send notify of removed controller + * @c: controller that is being removed * * Send notifications to all registered drivers that a controller was * removed. @@ -287,8 +285,8 @@ void i2o_driver_notify_controller_remove_all(struct i2o_controller *c) } /** - * i2o_driver_notify_device_add_all - Send notify of added device to all - * I2O drivers + * i2o_driver_notify_device_add_all - Send notify of added device + * @i2o_dev: newly added I2O device * * Send notifications to all registered drivers that a device was added. */ @@ -306,8 +304,8 @@ void i2o_driver_notify_device_add_all(struct i2o_device *i2o_dev) } /** - * i2o_driver_notify_device_remove_all - Send notify of removed device to - * all I2O drivers + * i2o_driver_notify_device_remove_all - Send notify of removed device + * @i2o_dev: device that is being removed * * Send notifications to all registered drivers that a device was removed. */ @@ -362,7 +360,7 @@ int __init i2o_driver_init(void) /** * i2o_driver_exit - clean up I2O drivers (OSMs) * - * Unregisters the I2O bus and free driver array. + * Unregisters the I2O bus and frees driver array. */ void __exit i2o_driver_exit(void) { diff --git a/drivers/message/i2o/exec-osm.c b/drivers/message/i2o/exec-osm.c index 9e529d8dd5cb..902753b2c661 100644 --- a/drivers/message/i2o/exec-osm.c +++ b/drivers/message/i2o/exec-osm.c @@ -94,8 +94,8 @@ static struct i2o_exec_wait *i2o_exec_wait_alloc(void) }; /** - * i2o_exec_wait_free - Free a i2o_exec_wait struct - * @i2o_exec_wait: I2O wait data which should be cleaned up + * i2o_exec_wait_free - Free an i2o_exec_wait struct + * @wait: I2O wait data which should be cleaned up */ static void i2o_exec_wait_free(struct i2o_exec_wait *wait) { @@ -105,7 +105,7 @@ static void i2o_exec_wait_free(struct i2o_exec_wait *wait) /** * i2o_msg_post_wait_mem - Post and wait a message with DMA buffers * @c: controller - * @m: message to post + * @msg: message to post * @timeout: time in seconds to wait * @dma: i2o_dma struct of the DMA buffer to free on failure * @@ -269,6 +269,7 @@ static int i2o_msg_post_wait_complete(struct i2o_controller *c, u32 m, /** * i2o_exec_show_vendor_id - Displays Vendor ID of controller * @d: device of which the Vendor ID should be displayed + * @attr: device_attribute to display * @buf: buffer into which the Vendor ID should be printed * * Returns number of bytes printed into buffer. @@ -290,6 +291,7 @@ static ssize_t i2o_exec_show_vendor_id(struct device *d, /** * i2o_exec_show_product_id - Displays Product ID of controller * @d: device of which the Product ID should be displayed + * @attr: device_attribute to display * @buf: buffer into which the Product ID should be printed * * Returns number of bytes printed into buffer. @@ -365,7 +367,7 @@ static int i2o_exec_remove(struct device *dev) /** * i2o_exec_lct_modified - Called on LCT NOTIFY reply - * @c: I2O controller on which the LCT has modified + * @work: work struct for a specific controller * * This function handles asynchronus LCT NOTIFY replies. It parses the * new LCT and if the buffer for the LCT was to small sends a LCT NOTIFY diff --git a/drivers/message/i2o/i2o_block.c b/drivers/message/i2o/i2o_block.c index 70ae00253321..da9859f2caf2 100644 --- a/drivers/message/i2o/i2o_block.c +++ b/drivers/message/i2o/i2o_block.c @@ -259,7 +259,7 @@ static int i2o_block_device_unlock(struct i2o_device *dev, u32 media_id) /** * i2o_block_device_power - Power management for device dev * @dev: I2O device which should receive the power management request - * @operation: Operation which should be send + * @op: Operation to send * * Send a power management request to the device dev. * @@ -315,7 +315,7 @@ static inline struct i2o_block_request *i2o_block_request_alloc(void) * i2o_block_request_free - Frees a I2O block request * @ireq: I2O block request which should be freed * - * Fres the allocated memory (give it back to the request mempool). + * Frees the allocated memory (give it back to the request mempool). */ static inline void i2o_block_request_free(struct i2o_block_request *ireq) { @@ -326,6 +326,7 @@ static inline void i2o_block_request_free(struct i2o_block_request *ireq) * i2o_block_sglist_alloc - Allocate the SG list and map it * @c: I2O controller to which the request belongs * @ireq: I2O block request + * @mptr: message body pointer * * Builds the SG list and map it to be accessable by the controller. * @@ -490,7 +491,7 @@ static void i2o_block_end_request(struct request *req, int uptodate, * i2o_block_reply - Block OSM reply handler. * @c: I2O controller from which the message arrives * @m: message id of reply - * qmsg: the actuall I2O message reply + * @msg: the actual I2O message reply * * This function gets all the message replies. * @@ -602,6 +603,8 @@ static void i2o_block_biosparam(unsigned long capacity, unsigned short *cyls, /** * i2o_block_open - Open the block device + * @inode: inode for block device being opened + * @file: file to open * * Power up the device, mount and lock the media. This function is called, * if the block device is opened for access. @@ -629,6 +632,8 @@ static int i2o_block_open(struct inode *inode, struct file *file) /** * i2o_block_release - Release the I2O block device + * @inode: inode for block device being released + * @file: file to close * * Unlock and unmount the media, and power down the device. Gets called if * the block device is closed. @@ -675,6 +680,8 @@ static int i2o_block_getgeo(struct block_device *bdev, struct hd_geometry *geo) /** * i2o_block_ioctl - Issue device specific ioctl calls. + * @inode: inode for block device ioctl + * @file: file for ioctl * @cmd: ioctl command * @arg: arg * @@ -902,7 +909,7 @@ static int i2o_block_transfer(struct request *req) /** * i2o_block_request_fn - request queue handling function - * q: request queue from which the request could be fetched + * @q: request queue from which the request could be fetched * * Takes the next request from the queue, transfers it and if no error * occurs dequeue it from the queue. On arrival of the reply the message diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c index 3d2e76eea93e..a61cb17c5c12 100644 --- a/drivers/message/i2o/i2o_proc.c +++ b/drivers/message/i2o/i2o_proc.c @@ -163,7 +163,7 @@ static int print_serial_number(struct seq_file *seq, u8 * serialno, int max_len) * i2o_get_class_name - do i2o class name lookup * @class: class number * - * Return a descriptive string for an i2o class + * Return a descriptive string for an i2o class. */ static const char *i2o_get_class_name(int class) { diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c index d5f93b12b6f1..1045c8a518bb 100644 --- a/drivers/message/i2o/i2o_scsi.c +++ b/drivers/message/i2o/i2o_scsi.c @@ -411,8 +411,7 @@ static void i2o_scsi_notify_device_add(struct i2o_device *i2o_dev) }; /** - * i2o_scsi_notify_device_remove - Retrieve notifications of removed - * devices + * i2o_scsi_notify_device_remove - Retrieve notifications of removed devices * @i2o_dev: the I2O device which was removed * * If a I2O device is removed, we catch the notification to remove the @@ -432,8 +431,7 @@ static void i2o_scsi_notify_device_remove(struct i2o_device *i2o_dev) }; /** - * i2o_scsi_notify_controller_add - Retrieve notifications of added - * controllers + * i2o_scsi_notify_controller_add - Retrieve notifications of added controllers * @c: the controller which was added * * If a I2O controller is added, we catch the notification to add a @@ -463,8 +461,7 @@ static void i2o_scsi_notify_controller_add(struct i2o_controller *c) }; /** - * i2o_scsi_notify_controller_remove - Retrieve notifications of removed - * controllers + * i2o_scsi_notify_controller_remove - Retrieve notifications of removed controllers * @c: the controller which was removed * * If a I2O controller is removed, we catch the notification to remove the @@ -751,7 +748,7 @@ static int i2o_scsi_abort(struct scsi_cmnd *SCpnt) * @capacity: size in sectors * @ip: geometry array * - * This is anyones guess quite frankly. We use the same rules everyone + * This is anyone's guess quite frankly. We use the same rules everyone * else appears to and hope. It seems to work. */ diff --git a/drivers/message/i2o/pci.c b/drivers/message/i2o/pci.c index 8287f95c8c42..3661e6e065d2 100644 --- a/drivers/message/i2o/pci.c +++ b/drivers/message/i2o/pci.c @@ -259,6 +259,7 @@ static irqreturn_t i2o_pci_interrupt(int irq, void *dev_id) /** * i2o_pci_irq_enable - Allocate interrupt for I2O controller + * @c: i2o_controller that the request is for * * Allocate an interrupt for the I2O controller, and activate interrupts * on the I2O controller. @@ -305,7 +306,7 @@ static void i2o_pci_irq_disable(struct i2o_controller *c) /** * i2o_pci_probe - Probe the PCI device for an I2O controller - * @dev: PCI device to test + * @pdev: PCI device to test * @id: id which matched with the PCI device id table * * Probe the PCI device for any device which is a memory of the @@ -447,7 +448,7 @@ static int __devinit i2o_pci_probe(struct pci_dev *pdev, /** * i2o_pci_remove - Removes a I2O controller from the system - * pdev: I2O controller which should be removed + * @pdev: I2O controller which should be removed * * Reset the I2O controller, disable interrupts and remove all allocated * resources. diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 2514f4e286b7..52f53e2e70c3 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -986,7 +986,8 @@ extern void i2o_driver_unregister(struct i2o_driver *); /** * i2o_driver_notify_controller_add - Send notification of added controller - * to a single I2O driver + * @drv: I2O driver + * @c: I2O controller * * Send notification of added controller to a single registered driver. */ @@ -998,8 +999,9 @@ static inline void i2o_driver_notify_controller_add(struct i2o_driver *drv, }; /** - * i2o_driver_notify_controller_remove - Send notification of removed - * controller to a single I2O driver + * i2o_driver_notify_controller_remove - Send notification of removed controller + * @drv: I2O driver + * @c: I2O controller * * Send notification of removed controller to a single registered driver. */ @@ -1011,8 +1013,9 @@ static inline void i2o_driver_notify_controller_remove(struct i2o_driver *drv, }; /** - * i2o_driver_notify_device_add - Send notification of added device to a - * single I2O driver + * i2o_driver_notify_device_add - Send notification of added device + * @drv: I2O driver + * @i2o_dev: the added i2o_device * * Send notification of added device to a single registered driver. */ @@ -1025,7 +1028,8 @@ static inline void i2o_driver_notify_device_add(struct i2o_driver *drv, /** * i2o_driver_notify_device_remove - Send notification of removed device - * to a single I2O driver + * @drv: I2O driver + * @i2o_dev: the added i2o_device * * Send notification of removed device to a single registered driver. */ @@ -1148,7 +1152,7 @@ static inline void i2o_msg_post(struct i2o_controller *c, /** * i2o_msg_post_wait - Post and wait a message and wait until return * @c: controller - * @m: message to post + * @msg: message to post * @timeout: time in seconds to wait * * This API allows an OSM to post a message and then be told whether or -- cgit v1.2.3 From 83b7b44e1c1e9e493ccd4146558481ab5af0116a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 6 Dec 2006 20:38:53 -0800 Subject: [PATCH] fs: reorder some 'struct inode' fields to speedup i_size manipulations On 32bits SMP platforms, 64bits i_size is protected by a seqcount (i_size_seqcount). When i_size is read or written, i_size_seqcount is read/written as well, so it make sense to group these two fields together in the same cache line. This patch moves i_size_seqcount next to i_size, and also moves i_version to let offsetof(struct inode, i_size) being 0x40 instead of 0x3c (for 32bits platforms). For 64 bits platforms, i_size_seqcount doesnt exist, and the move of a 'long i_version' should not introduce a new hole because of padding. Signed-off-by: Eric Dumazet Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index d791bae9de9c..3a1927e516d0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -548,12 +548,15 @@ struct inode { uid_t i_uid; gid_t i_gid; dev_t i_rdev; + unsigned long i_version; loff_t i_size; +#ifdef __NEED_I_SIZE_ORDERED + seqcount_t i_size_seqcount; +#endif struct timespec i_atime; struct timespec i_mtime; struct timespec i_ctime; unsigned int i_blkbits; - unsigned long i_version; blkcnt_t i_blocks; unsigned short i_bytes; spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ @@ -598,9 +601,6 @@ struct inode { void *i_security; #endif void *i_private; /* fs or device private pointer */ -#ifdef __NEED_I_SIZE_ORDERED - seqcount_t i_size_seqcount; -#endif }; /* -- cgit v1.2.3 From f7dff2b12654149c9cac8d8c79b6588759edd5a9 Mon Sep 17 00:00:00 2001 From: Gautham R Shenoy Date: Wed, 6 Dec 2006 20:38:58 -0800 Subject: [PATCH] Handle per-subsystem mutexes for CONFIG_HOTPLUG_CPU not set Provide a common interface for all the subsystems to lock and unlock their per-subsystem hotcpu mutexes. When CONFIG_HOTPLUG_CPU is not set, these operations would be no-ops. [akpm@osdl.org: macros -> inlines] Signed-off-by: Gautham R Shenoy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpu.h | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 71dc6ba4f73f..bf00ce6ecadf 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -24,6 +24,7 @@ #include #include #include +#include struct cpu { int node_id; /* The node which contains the CPU */ @@ -74,6 +75,17 @@ extern struct sysdev_class cpu_sysdev_class; #ifdef CONFIG_HOTPLUG_CPU /* Stop CPUs going up and down. */ + +static inline void cpuhotplug_mutex_lock(struct mutex *cpu_hp_mutex) +{ + mutex_lock(cpu_hp_mutex); +} + +static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex) +{ + mutex_unlock(cpu_hp_mutex); +} + extern void lock_cpu_hotplug(void); extern void unlock_cpu_hotplug(void); #define hotcpu_notifier(fn, pri) { \ @@ -85,7 +97,14 @@ extern void unlock_cpu_hotplug(void); #define unregister_hotcpu_notifier(nb) unregister_cpu_notifier(nb) int cpu_down(unsigned int cpu); #define cpu_is_offline(cpu) unlikely(!cpu_online(cpu)) -#else + +#else /* CONFIG_HOTPLUG_CPU */ + +static inline void cpuhotplug_mutex_lock(struct mutex *cpu_hp_mutex) +{ } +static inline void cpuhotplug_mutex_unlock(struct mutex *cpu_hp_mutex) +{ } + #define lock_cpu_hotplug() do { } while (0) #define unlock_cpu_hotplug() do { } while (0) #define lock_cpu_hotplug_interruptible() 0 @@ -95,7 +114,7 @@ int cpu_down(unsigned int cpu); /* CPUs don't go offline once they're online w/o CONFIG_HOTPLUG_CPU */ static inline int cpu_is_offline(int cpu) { return 0; } -#endif +#endif /* CONFIG_HOTPLUG_CPU */ #ifdef CONFIG_SUSPEND_SMP extern int disable_nonboot_cpus(void); -- cgit v1.2.3 From 0b71c8e76d20d0329bf7e54f172389f3c343dc41 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 6 Dec 2006 20:39:15 -0800 Subject: [PATCH] remove useless carta_random32.h Remove the carta_random32.h header file. The carta_random32() function was was put in and removed in favor of random32(). In the removal process, the header file was forgotten. Signed-off-by: Stephane Eranian Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/carta_random32.h | 29 ----------------------------- 1 file changed, 29 deletions(-) delete mode 100644 include/linux/carta_random32.h (limited to 'include/linux') diff --git a/include/linux/carta_random32.h b/include/linux/carta_random32.h deleted file mode 100644 index f6f3bd9f20b5..000000000000 --- a/include/linux/carta_random32.h +++ /dev/null @@ -1,29 +0,0 @@ -/* - * Fast, simple, yet decent quality random number generator based on - * a paper by David G. Carta ("Two Fast Implementations of the - * `Minimal Standard' Random Number Generator," Communications of the - * ACM, January, 1990). - * - * Copyright (c) 2002-2006 Hewlett-Packard Development Company, L.P. - * Contributed by Stephane Eranian - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of version 2 of the GNU General Public - * License as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA - * 02111-1307 USA - */ -#ifndef _LINUX_CARTA_RANDOM32_H_ -#define _LINUX_CARTA_RANDOM32_H_ - -u64 carta_random32(u64 seed); - -#endif /* _LINUX_CARTA_RANDOM32_H_ */ -- cgit v1.2.3 From 2ee91f197c0bc654b24eed5831fd12aa0d566a7d Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 6 Dec 2006 20:39:32 -0800 Subject: [PATCH] lockdep: show more details about self-test failures Make the locking self-test failures (of 'FAILURE' type) easier to debug by printing more information. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/debug_locks.h | 2 +- kernel/mutex-debug.c | 3 +++ lib/locking-selftest.c | 2 ++ 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h index 952bee79a8f3..a1c10b0c4cf0 100644 --- a/include/linux/debug_locks.h +++ b/include/linux/debug_locks.h @@ -24,7 +24,7 @@ extern int debug_locks_off(void); int __ret = 0; \ \ if (unlikely(c)) { \ - if (debug_locks_off()) \ + if (debug_locks_silent || debug_locks_off()) \ WARN_ON(1); \ __ret = 1; \ } \ diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 18651641a7b5..841539d72c55 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -77,6 +77,9 @@ void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, void debug_mutex_unlock(struct mutex *lock) { + if (unlikely(!debug_locks)) + return; + DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info()); DEBUG_LOCKS_WARN_ON(lock->magic != lock); DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next); diff --git a/lib/locking-selftest.c b/lib/locking-selftest.c index 7945787f439a..280332c1827c 100644 --- a/lib/locking-selftest.c +++ b/lib/locking-selftest.c @@ -963,7 +963,9 @@ static void dotest(void (*testcase_fn)(void), int expected, int lockclass_mask) printk("failed|"); } else { unexpected_testcase_failures++; + printk("FAILED|"); + dump_stack(); } } else { testcase_successes++; -- cgit v1.2.3 From 12d40e43d251de4fa1f982567fc8b4ee5e858367 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 6 Dec 2006 20:39:53 -0800 Subject: [PATCH] Save some bytes in struct inode [acme@newtoy net-2.6.20]$ pahole --cacheline 64 fs/inode.o inode /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/dcache.h:86 */ struct inode { struct hlist_node i_hash; /* 0 8 */ struct list_head i_list; /* 8 8 */ struct list_head i_sb_list; /* 16 8 */ struct list_head i_dentry; /* 24 8 */ long unsigned int i_ino; /* 32 4 */ atomic_t i_count; /* 36 4 */ umode_t i_mode; /* 40 2 */ /* XXX 2 bytes hole, try to pack */ unsigned int i_nlink; /* 44 4 */ uid_t i_uid; /* 48 4 */ gid_t i_gid; /* 52 4 */ dev_t i_rdev; /* 56 4 */ loff_t i_size; /* 60 8 */ struct timespec i_atime; /* 68 8 */ struct timespec i_mtime; /* 76 8 */ struct timespec i_ctime; /* 84 8 */ unsigned int i_blkbits; /* 92 4 */ long unsigned int i_version; /* 96 4 */ blkcnt_t i_blocks; /* 100 4 */ short unsigned int i_bytes; /* 104 2 */ /* XXX 2 bytes hole, try to pack */ spinlock_t i_lock; /* 108 40 */ struct mutex i_mutex; /* 148 76 */ struct rw_semaphore i_alloc_sem; /* 224 64 */ struct inode_operations * i_op; /* 288 4 */ const struct file_operations * i_fop; /* 292 4 */ struct super_block * i_sb; /* 296 4 */ struct file_lock * i_flock; /* 300 4 */ struct address_space * i_mapping; /* 304 4 */ struct address_space i_data; /* 308 188 */ struct list_head i_devices; /* 496 8 */ union ; /* 504 4 */ int i_cindex; /* 508 4 */ __u32 i_generation; /* 512 4 */ /* ---------- cacheline 8 boundary ---------- */ long unsigned int i_dnotify_mask; /* 516 4 */ struct dnotify_struct * i_dnotify; /* 520 4 */ struct list_head inotify_watches; /* 524 8 */ struct mutex inotify_mutex; /* 532 76 */ long unsigned int i_state; /* 608 4 */ long unsigned int dirtied_when; /* 612 4 */ unsigned int i_flags; /* 616 4 */ atomic_t i_writecount; /* 620 4 */ void * i_security; /* 624 4 */ void * i_private; /* 628 4 */ }; /* size: 632, sum members: 628, holes: 2, sum holes: 4 */ [acme@newtoy net-2.6.20]$ So just moving i_mode to after i_bytes we save 4 bytes by nuking both holes: [acme@newtoy net-2.6.20]$ codiff -V /tmp/inode.o.before fs/inode.o /pub/scm/linux/kernel/git/acme/net-2.6.20/fs/inode.c: struct inode | -4 i_mode; from: umode_t /* 40(0) 2(0) */ to: umode_t /* 102(0) 2(0) */ 1 struct changed [acme@newtoy net-2.6.20]$ I've prunned all the other offset changes, only this one is of interest here. So now we have: [acme@newtoy net-2.6.20]$ pahole --cacheline 64 ../OUTPUT/qemu/net-2.6.20/fs/inode.o inode /* /pub/scm/linux/kernel/git/acme/net-2.6.20/include/linux/dcache.h:86 */ struct inode { struct hlist_node i_hash; /* 0 8 */ struct list_head i_list; /* 8 8 */ struct list_head i_sb_list; /* 16 8 */ struct list_head i_dentry; /* 24 8 */ long unsigned int i_ino; /* 32 4 */ atomic_t i_count; /* 36 4 */ unsigned int i_nlink; /* 40 4 */ uid_t i_uid; /* 44 4 */ gid_t i_gid; /* 48 4 */ dev_t i_rdev; /* 52 4 */ loff_t i_size; /* 56 8 */ /* ---------- cacheline 1 boundary ---------- */ struct timespec i_atime; /* 64 8 */ struct timespec i_mtime; /* 72 8 */ struct timespec i_ctime; /* 80 8 */ unsigned int i_blkbits; /* 88 4 */ long unsigned int i_version; /* 92 4 */ blkcnt_t i_blocks; /* 96 4 */ short unsigned int i_bytes; /* 100 2 */ umode_t i_mode; /* 102 2 */ spinlock_t i_lock; /* 104 40 */ struct mutex i_mutex; /* 144 76 */ struct rw_semaphore i_alloc_sem; /* 220 64 */ struct inode_operations * i_op; /* 284 4 */ const struct file_operations * i_fop; /* 288 4 */ struct super_block * i_sb; /* 292 4 */ struct file_lock * i_flock; /* 296 4 */ struct address_space * i_mapping; /* 300 4 */ struct address_space i_data; /* 304 188 */ struct list_head i_devices; /* 492 8 */ union ; /* 500 4 */ int i_cindex; /* 504 4 */ __u32 i_generation; /* 508 4 */ /* ---------- cacheline 8 boundary ---------- */ long unsigned int i_dnotify_mask; /* 512 4 */ struct dnotify_struct * i_dnotify; /* 516 4 */ struct list_head inotify_watches; /* 520 8 */ struct mutex inotify_mutex; /* 528 76 */ long unsigned int i_state; /* 604 4 */ long unsigned int dirtied_when; /* 608 4 */ unsigned int i_flags; /* 612 4 */ atomic_t i_writecount; /* 616 4 */ void * i_security; /* 620 4 */ void * i_private; /* 624 4 */ }; /* size: 628 */ [acme@newtoy net-2.6.20]$ Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 3a1927e516d0..70b99fbb560b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -543,7 +543,6 @@ struct inode { struct list_head i_dentry; unsigned long i_ino; atomic_t i_count; - umode_t i_mode; unsigned int i_nlink; uid_t i_uid; gid_t i_gid; @@ -559,6 +558,7 @@ struct inode { unsigned int i_blkbits; blkcnt_t i_blocks; unsigned short i_bytes; + umode_t i_mode; spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ struct mutex i_mutex; struct rw_semaphore i_alloc_sem; -- cgit v1.2.3 From 0da1480ec33d4bac8c32051c1d33202be6dc439f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 6 Dec 2006 20:40:03 -0800 Subject: [PATCH] proper prototype for remove_inode_dquot_ref() Add a proper prototype for remove_inode_dquot_ref() in include/linux/quotaops.h Signed-off-by: Adrian Bunk Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 3 --- include/linux/quotaops.h | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/inode.c b/fs/inode.c index 699aa4f7aa74..9ecccab7326d 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1242,9 +1242,6 @@ EXPORT_SYMBOL(inode_needs_sync); */ #ifdef CONFIG_QUOTA -/* Function back in dquot.c */ -int remove_inode_dquot_ref(struct inode *, int, struct list_head *); - void remove_dquot_ref(struct super_block *sb, int type, struct list_head *tofree_head) { diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h index 5110201a4159..90c23f690c0d 100644 --- a/include/linux/quotaops.h +++ b/include/linux/quotaops.h @@ -37,6 +37,9 @@ extern int dquot_release(struct dquot *dquot); extern int dquot_commit_info(struct super_block *sb, int type); extern int dquot_mark_dquot_dirty(struct dquot *dquot); +int remove_inode_dquot_ref(struct inode *inode, int type, + struct list_head *tofree_head); + extern int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path); extern int vfs_quota_on_mount(struct super_block *sb, char *qf_name, int format_id, int type); -- cgit v1.2.3 From 65867beb0de4d055637476327b533e5ffbec2b97 Mon Sep 17 00:00:00 2001 From: Jordan Crouse Date: Wed, 6 Dec 2006 20:40:07 -0800 Subject: [PATCH] Trivial cleanup in the PCI IDs for the CS5535 Rename a poorly worded PCI ID for the Geode GX and CS5535 companion chips. The graphics processor and host bridge actually live in the northbridge on the integrated processor, not in the companion chip. Signed-off-by: Jordan Crouse Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/geode/gxfb_core.c | 2 +- include/linux/pci_ids.h | 5 ++--- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/geode/gxfb_core.c b/drivers/video/geode/gxfb_core.c index 0d3643fc6293..a454dcb8e215 100644 --- a/drivers/video/geode/gxfb_core.c +++ b/drivers/video/geode/gxfb_core.c @@ -380,7 +380,7 @@ static void gxfb_remove(struct pci_dev *pdev) } static struct pci_device_id gxfb_id_table[] = { - { PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_VIDEO, + { PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_GX_VIDEO, PCI_ANY_ID, PCI_ANY_ID, PCI_BASE_CLASS_DISPLAY << 16, 0xff0000, 0 }, { 0, } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index dcdb90f06d73..ff2dcb436cd0 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -390,7 +390,7 @@ #define PCI_DEVICE_ID_NS_CS5535_IDE 0x002d #define PCI_DEVICE_ID_NS_CS5535_AUDIO 0x002e #define PCI_DEVICE_ID_NS_CS5535_USB 0x002f -#define PCI_DEVICE_ID_NS_CS5535_VIDEO 0x0030 +#define PCI_DEVICE_ID_NS_GX_VIDEO 0x0030 #define PCI_DEVICE_ID_NS_SATURN 0x0035 #define PCI_DEVICE_ID_NS_SCx200_BRIDGE 0x0500 #define PCI_DEVICE_ID_NS_SCx200_SMI 0x0501 @@ -403,8 +403,7 @@ #define PCI_DEVICE_ID_NS_SC1100_XBUS 0x0515 #define PCI_DEVICE_ID_NS_87410 0xd001 -#define PCI_DEVICE_ID_NS_CS5535_HOST_BRIDGE 0x0028 -#define PCI_DEVICE_ID_NS_CS5535_ISA_BRIDGE 0x002b +#define PCI_DEVICE_ID_NS_GX_HOST_BRIDGE 0x0028 #define PCI_VENDOR_ID_TSENG 0x100c #define PCI_DEVICE_ID_TSENG_W32P_2 0x3202 -- cgit v1.2.3 From 5296c7bec8c85aa0a6964eca9ff4a1a8847fba8a Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Wed, 6 Dec 2006 20:40:10 -0800 Subject: [PATCH] fs: reiserfs add missing brackets Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/reiserfs_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 7bc6bfb86253..d0e4dce33ad5 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -739,7 +739,7 @@ struct block_head { #define PUT_B_FREE_SPACE(p_s_bh,val) do { set_blkh_free_space(B_BLK_HEAD(p_s_bh),val); } while (0) /* Get right delimiting key. -- little endian */ -#define B_PRIGHT_DELIM_KEY(p_s_bh) (&(blk_right_delim_key(B_BLK_HEAD(p_s_bh)) +#define B_PRIGHT_DELIM_KEY(p_s_bh) (&(blk_right_delim_key(B_BLK_HEAD(p_s_bh)))) /* Does the buffer contain a disk leaf. */ #define B_IS_ITEMS_LEVEL(p_s_bh) (B_LEVEL(p_s_bh) == DISK_LEAF_NODE_LEVEL) -- cgit v1.2.3 From a0e7688df1484fbf4d6d61c31f7d61a5d8cacf3c Mon Sep 17 00:00:00 2001 From: Robert P. J. Day Date: Wed, 6 Dec 2006 20:40:12 -0800 Subject: [PATCH] Kbuild: add 3 more header files to get properly "unifdef"ed Add 3 more files to get "unifdef"ed when creating sanitized headers with "make headers_install". Signed-off-by: Robert P. J. Day Acked-by: David Woodhouse Acked-by: "John W. Linville" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/Kbuild | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index ff433126361f..e618b25b5add 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -221,6 +221,7 @@ unifdef-y += if_bridge.h unifdef-y += if_ec.h unifdef-y += if_eql.h unifdef-y += if_ether.h +unifdef-y += if_fddi.h unifdef-y += if_frad.h unifdef-y += if_ltalk.h unifdef-y += if_pppox.h @@ -282,6 +283,7 @@ unifdef-y += nvram.h unifdef-y += parport.h unifdef-y += patchkey.h unifdef-y += pci.h +unifdef-y += personality.h unifdef-y += pktcdvd.h unifdef-y += pmu.h unifdef-y += poll.h @@ -337,6 +339,7 @@ unifdef-y += videodev.h unifdef-y += wait.h unifdef-y += wanrouter.h unifdef-y += watchdog.h +unifdef-y += wireless.h unifdef-y += xfrm.h objhdr-y += version.h -- cgit v1.2.3 From 15ad7cdcfd76450d4beebc789ec646664238184d Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Wed, 6 Dec 2006 20:40:36 -0800 Subject: [PATCH] struct seq_operations and struct file_operations constification - move some file_operations structs into the .rodata section - move static strings from policy_types[] array into the .rodata section - fix generic seq_operations usages, so that those structs may be defined as "const" as well [akpm@osdl.org: couple of fixes] Signed-off-by: Helge Deller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/seq_file.c | 4 ++-- include/linux/cpuset.h | 2 +- include/linux/mmzone.h | 2 +- include/linux/relay.h | 2 +- include/linux/sched.h | 2 +- include/linux/seq_file.h | 4 ++-- kernel/configs.c | 2 +- kernel/cpuset.c | 4 ++-- kernel/dma.c | 2 +- kernel/futex.c | 2 +- kernel/kallsyms.c | 4 ++-- kernel/lockdep_proc.c | 6 +++--- kernel/module.c | 2 +- kernel/power/user.c | 2 +- kernel/profile.c | 2 +- kernel/relay.c | 2 +- kernel/resource.c | 6 +++--- kernel/sched.c | 2 +- kernel/sysctl.c | 2 +- mm/mempolicy.c | 4 ++-- mm/page_alloc.c | 2 +- mm/shmem.c | 4 ++-- mm/slab.c | 4 ++-- mm/swapfile.c | 4 ++-- mm/vmstat.c | 8 ++++---- 25 files changed, 40 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/fs/seq_file.c b/fs/seq_file.c index 555b9ac04c25..10690aa401c7 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -26,7 +26,7 @@ * ERR_PTR(error). In the end of sequence they return %NULL. ->show() * returns 0 in case of success and negative number in case of error. */ -int seq_open(struct file *file, struct seq_operations *op) +int seq_open(struct file *file, const struct seq_operations *op) { struct seq_file *p = file->private_data; @@ -408,7 +408,7 @@ EXPORT_SYMBOL(single_open); int single_release(struct inode *inode, struct file *file) { - struct seq_operations *op = ((struct seq_file *)file->private_data)->op; + const struct seq_operations *op = ((struct seq_file *)file->private_data)->op; int res = seq_release(inode, file); kfree(op); return res; diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 748d2c996631..8821e1f75b44 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -46,7 +46,7 @@ extern int cpuset_excl_nodes_overlap(const struct task_struct *p); extern int cpuset_memory_pressure_enabled; extern void __cpuset_memory_pressure_bump(void); -extern struct file_operations proc_cpuset_operations; +extern const struct file_operations proc_cpuset_operations; extern char *cpuset_task_status_allowed(struct task_struct *task, char *buffer); extern void cpuset_lock(void); diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index da6002dec205..e339a7345f25 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -278,7 +278,7 @@ struct zone { /* * rarely used fields: */ - char *name; + const char *name; } ____cacheline_internodealigned_in_smp; /* diff --git a/include/linux/relay.h b/include/linux/relay.h index 0e3d91b76996..c6a48bfc8b14 100644 --- a/include/linux/relay.h +++ b/include/linux/relay.h @@ -274,7 +274,7 @@ static inline void subbuf_start_reserve(struct rchan_buf *buf, /* * exported relay file operations, kernel/relay.c */ -extern struct file_operations relay_file_operations; +extern const struct file_operations relay_file_operations; #endif /* _LINUX_RELAY_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 3a767242e72f..dede82c63445 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -573,7 +573,7 @@ struct sched_info { #endif /* defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) */ #ifdef CONFIG_SCHEDSTATS -extern struct file_operations proc_schedstat_operations; +extern const struct file_operations proc_schedstat_operations; #endif /* CONFIG_SCHEDSTATS */ #ifdef CONFIG_TASK_DELAY_ACCT diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index b95f6eb7254c..3e3cccbb1cac 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -20,7 +20,7 @@ struct seq_file { loff_t index; loff_t version; struct mutex lock; - struct seq_operations *op; + const struct seq_operations *op; void *private; }; @@ -31,7 +31,7 @@ struct seq_operations { int (*show) (struct seq_file *m, void *v); }; -int seq_open(struct file *, struct seq_operations *); +int seq_open(struct file *, const struct seq_operations *); ssize_t seq_read(struct file *, char __user *, size_t, loff_t *); loff_t seq_lseek(struct file *, loff_t, int); int seq_release(struct inode *, struct file *); diff --git a/kernel/configs.c b/kernel/configs.c index f9e31974f4ad..8fa1fb28f8a7 100644 --- a/kernel/configs.c +++ b/kernel/configs.c @@ -75,7 +75,7 @@ ikconfig_read_current(struct file *file, char __user *buf, return count; } -static struct file_operations ikconfig_file_ops = { +static const struct file_operations ikconfig_file_ops = { .owner = THIS_MODULE, .read = ikconfig_read_current, }; diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 9b62b4c03ad0..4ef1d29297ca 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -1531,7 +1531,7 @@ static int cpuset_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } -static struct file_operations cpuset_file_operations = { +static const struct file_operations cpuset_file_operations = { .read = cpuset_file_read, .write = cpuset_file_write, .llseek = generic_file_llseek, @@ -2605,7 +2605,7 @@ static int cpuset_open(struct inode *inode, struct file *file) return single_open(file, proc_cpuset_show, pid); } -struct file_operations proc_cpuset_operations = { +const struct file_operations proc_cpuset_operations = { .open = cpuset_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/dma.c b/kernel/dma.c index 2020644c938a..937b13ca33ba 100644 --- a/kernel/dma.c +++ b/kernel/dma.c @@ -140,7 +140,7 @@ static int proc_dma_open(struct inode *inode, struct file *file) return single_open(file, proc_dma_show, NULL); } -static struct file_operations proc_dma_operations = { +static const struct file_operations proc_dma_operations = { .open = proc_dma_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/futex.c b/kernel/futex.c index a8302a1620ea..95989a3b4168 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1492,7 +1492,7 @@ static unsigned int futex_poll(struct file *filp, return ret; } -static struct file_operations futex_fops = { +static const struct file_operations futex_fops = { .release = futex_close, .poll = futex_poll, }; diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 54befe36ee0b..ab63cfc42992 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -398,7 +398,7 @@ static int s_show(struct seq_file *m, void *p) return 0; } -static struct seq_operations kallsyms_op = { +static const struct seq_operations kallsyms_op = { .start = s_start, .next = s_next, .stop = s_stop, @@ -433,7 +433,7 @@ static int kallsyms_release(struct inode *inode, struct file *file) return seq_release(inode, file); } -static struct file_operations kallsyms_operations = { +static const struct file_operations kallsyms_operations = { .open = kallsyms_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index f6e72eaab3fa..b554b40a4aa6 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -113,7 +113,7 @@ static int l_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations lockdep_ops = { +static const struct seq_operations lockdep_ops = { .start = l_start, .next = l_next, .stop = l_stop, @@ -135,7 +135,7 @@ static int lockdep_open(struct inode *inode, struct file *file) return res; } -static struct file_operations proc_lockdep_operations = { +static const struct file_operations proc_lockdep_operations = { .open = lockdep_open, .read = seq_read, .llseek = seq_lseek, @@ -319,7 +319,7 @@ static int lockdep_stats_open(struct inode *inode, struct file *file) return single_open(file, lockdep_stats_show, NULL); } -static struct file_operations proc_lockdep_stats_operations = { +static const struct file_operations proc_lockdep_stats_operations = { .open = lockdep_stats_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/module.c b/kernel/module.c index e2d09d604ca0..d9eae45d0145 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2209,7 +2209,7 @@ static int m_show(struct seq_file *m, void *p) Where refcount is a number or -, and deps is a comma-separated list of depends or -. */ -struct seq_operations modules_op = { +const struct seq_operations modules_op = { .start = m_start, .next = m_next, .stop = m_stop, diff --git a/kernel/power/user.c b/kernel/power/user.c index 069732e5695c..89443b85163b 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -383,7 +383,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, return error; } -static struct file_operations snapshot_fops = { +static const struct file_operations snapshot_fops = { .open = snapshot_open, .release = snapshot_release, .read = snapshot_read, diff --git a/kernel/profile.c b/kernel/profile.c index 0961d93e1d91..fb5e03d57e9d 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -501,7 +501,7 @@ static ssize_t write_profile(struct file *file, const char __user *buf, return count; } -static struct file_operations proc_profile_operations = { +static const struct file_operations proc_profile_operations = { .read = read_profile, .write = write_profile, }; diff --git a/kernel/relay.c b/kernel/relay.c index 2b92e8ece85b..75a3a9a7efc2 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1013,7 +1013,7 @@ static ssize_t relay_file_sendfile(struct file *filp, actor, &desc); } -struct file_operations relay_file_operations = { +const struct file_operations relay_file_operations = { .open = relay_file_open, .poll = relay_file_poll, .mmap = relay_file_mmap, diff --git a/kernel/resource.c b/kernel/resource.c index 6de60c12143e..7b9a497419d9 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -88,7 +88,7 @@ static int r_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations resource_op = { +static const struct seq_operations resource_op = { .start = r_start, .next = r_next, .stop = r_stop, @@ -115,14 +115,14 @@ static int iomem_open(struct inode *inode, struct file *file) return res; } -static struct file_operations proc_ioports_operations = { +static const struct file_operations proc_ioports_operations = { .open = ioports_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, }; -static struct file_operations proc_iomem_operations = { +static const struct file_operations proc_iomem_operations = { .open = iomem_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/sched.c b/kernel/sched.c index b43cef02ce5b..f385eff4682d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -505,7 +505,7 @@ static int schedstat_open(struct inode *inode, struct file *file) return res; } -struct file_operations proc_schedstat_operations = { +const struct file_operations proc_schedstat_operations = { .open = schedstat_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 6d7147cf66bb..758dbbf972a5 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -170,7 +170,7 @@ static ssize_t proc_readsys(struct file *, char __user *, size_t, loff_t *); static ssize_t proc_writesys(struct file *, const char __user *, size_t, loff_t *); static int proc_opensys(struct inode *, struct file *); -struct file_operations proc_sys_file_operations = { +const struct file_operations proc_sys_file_operations = { .open = proc_opensys, .read = proc_readsys, .write = proc_writesys, diff --git a/mm/mempolicy.c b/mm/mempolicy.c index ad864f8708b0..b917d6fdc1bb 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1707,8 +1707,8 @@ void mpol_rebind_mm(struct mm_struct *mm, nodemask_t *new) * Display pages allocated per node and memory policy via /proc. */ -static const char *policy_types[] = { "default", "prefer", "bind", - "interleave" }; +static const char * const policy_types[] = + { "default", "prefer", "bind", "interleave" }; /* * Convert a mempolicy into a string. diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 27ec7a1b8022..cace22b3ac25 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -83,7 +83,7 @@ int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { EXPORT_SYMBOL(totalram_pages); -static char *zone_names[MAX_NR_ZONES] = { +static char * const zone_names[MAX_NR_ZONES] = { "DMA", #ifdef CONFIG_ZONE_DMA32 "DMA32", diff --git a/mm/shmem.c b/mm/shmem.c index 007653680a75..c820b4f77b8d 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -177,7 +177,7 @@ static inline void shmem_unacct_blocks(unsigned long flags, long pages) static struct super_operations shmem_ops; static const struct address_space_operations shmem_aops; -static struct file_operations shmem_file_operations; +static const struct file_operations shmem_file_operations; static struct inode_operations shmem_inode_operations; static struct inode_operations shmem_dir_inode_operations; static struct inode_operations shmem_special_inode_operations; @@ -2319,7 +2319,7 @@ static const struct address_space_operations shmem_aops = { .migratepage = migrate_page, }; -static struct file_operations shmem_file_operations = { +static const struct file_operations shmem_file_operations = { .mmap = shmem_mmap, #ifdef CONFIG_TMPFS .llseek = generic_file_llseek, diff --git a/mm/slab.c b/mm/slab.c index 86f5d6e995bb..068cb4503c15 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -4142,7 +4142,7 @@ static int s_show(struct seq_file *m, void *p) * + further values on SMP and with statistics enabled */ -struct seq_operations slabinfo_op = { +const struct seq_operations slabinfo_op = { .start = s_start, .next = s_next, .stop = s_stop, @@ -4340,7 +4340,7 @@ static int leaks_show(struct seq_file *m, void *p) return 0; } -struct seq_operations slabstats_op = { +const struct seq_operations slabstats_op = { .start = leaks_start, .next = s_next, .stop = s_stop, diff --git a/mm/swapfile.c b/mm/swapfile.c index 55242363de64..c5431072f422 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1368,7 +1368,7 @@ static int swap_show(struct seq_file *swap, void *v) return 0; } -static struct seq_operations swaps_op = { +static const struct seq_operations swaps_op = { .start = swap_start, .next = swap_next, .stop = swap_stop, @@ -1380,7 +1380,7 @@ static int swaps_open(struct inode *inode, struct file *file) return seq_open(file, &swaps_op); } -static struct file_operations proc_swaps_operations = { +static const struct file_operations proc_swaps_operations = { .open = swaps_open, .read = seq_read, .llseek = seq_lseek, diff --git a/mm/vmstat.c b/mm/vmstat.c index ef4176843d9e..dc005a0c96ae 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -430,7 +430,7 @@ static int frag_show(struct seq_file *m, void *arg) return 0; } -struct seq_operations fragmentation_op = { +const struct seq_operations fragmentation_op = { .start = frag_start, .next = frag_next, .stop = frag_stop, @@ -452,7 +452,7 @@ struct seq_operations fragmentation_op = { #define TEXTS_FOR_ZONES(xx) xx "_dma", TEXT_FOR_DMA32(xx) xx "_normal", \ TEXT_FOR_HIGHMEM(xx) -static char *vmstat_text[] = { +static const char * const vmstat_text[] = { /* Zoned VM counters */ "nr_anon_pages", "nr_mapped", @@ -597,7 +597,7 @@ static int zoneinfo_show(struct seq_file *m, void *arg) return 0; } -struct seq_operations zoneinfo_op = { +const struct seq_operations zoneinfo_op = { .start = frag_start, /* iterate over all zones. The same as in * fragmentation. */ .next = frag_next, @@ -660,7 +660,7 @@ static void vmstat_stop(struct seq_file *m, void *arg) m->private = NULL; } -struct seq_operations vmstat_op = { +const struct seq_operations vmstat_op = { .start = vmstat_start, .next = vmstat_next, .stop = vmstat_stop, -- cgit v1.2.3 From 85916f8166b59eeac63d2b4f7f1df8de849334b4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Wed, 6 Dec 2006 20:40:41 -0800 Subject: [PATCH] Kexec / Kdump: Unify elf note code The elf note saving code is currently duplicated over several architectures. This cleanup patch simply adds code to a common file and then replaces the arch-specific code with calls to the newly added code. The only drawback with this approach is that s390 doesn't fully support kexec-on-panic which for that arch leads to introduction of unused code. Signed-off-by: Magnus Damm Cc: Vivek Goyal Cc: Andi Kleen Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/crash.c | 66 ++----------------------------------------- arch/powerpc/kernel/crash.c | 59 ++------------------------------------ arch/x86_64/kernel/crash.c | 69 ++------------------------------------------- include/linux/kexec.h | 1 + kernel/kexec.c | 56 ++++++++++++++++++++++++++++++++++++ 5 files changed, 63 insertions(+), 188 deletions(-) (limited to 'include/linux') diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index 144b43288965..a5e0e990ea95 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -31,68 +31,6 @@ /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, - size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that, so there is no need to invent something new. - */ - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - if (!buf) - return; - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - -static void crash_save_self(struct pt_regs *regs) -{ - int cpu; - - cpu = safe_smp_processor_id(); - crash_save_this_cpu(regs, cpu); -} - #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) static atomic_t waiting_for_crash_ipi; @@ -121,7 +59,7 @@ static int crash_nmi_callback(struct notifier_block *self, crash_fixup_ss_esp(&fixed_regs, regs); regs = &fixed_regs; } - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); disable_local_APIC(); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -195,5 +133,5 @@ void machine_crash_shutdown(struct pt_regs *regs) #if defined(CONFIG_X86_IO_APIC) disable_IO_APIC(); #endif - crash_save_self(regs); + crash_save_cpu(regs, safe_smp_processor_id()); } diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 89b03c8da9d2..d3f2080d2eee 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -46,61 +46,6 @@ int crashing_cpu = -1; static cpumask_t cpus_in_crash = CPU_MASK_NONE; cpumask_t cpus_in_sr = CPU_MASK_NONE; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, - size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that that no need to invent something new. - */ - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - if (!buf) - return; - - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - #ifdef CONFIG_SMP static atomic_t enter_on_soft_reset = ATOMIC_INIT(0); @@ -113,7 +58,7 @@ void crash_ipi_callback(struct pt_regs *regs) hard_irq_disable(); if (!cpu_isset(cpu, cpus_in_crash)) - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); cpu_set(cpu, cpus_in_crash); /* @@ -306,7 +251,7 @@ void default_machine_crash_shutdown(struct pt_regs *regs) * such that another IPI will not be sent. */ crashing_cpu = smp_processor_id(); - crash_save_this_cpu(regs, crashing_cpu); + crash_save_cpu(regs, crashing_cpu); crash_kexec_prepare_cpus(crashing_cpu); cpu_set(crashing_cpu, cpus_in_crash); if (ppc_md.kexec_cpu_down) diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c index 3525f884af82..95a7a2c13131 100644 --- a/arch/x86_64/kernel/crash.c +++ b/arch/x86_64/kernel/crash.c @@ -28,71 +28,6 @@ /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, - void *data, size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) +3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - -static void crash_save_this_cpu(struct pt_regs *regs, int cpu) -{ - struct elf_prstatus prstatus; - u32 *buf; - - if ((cpu < 0) || (cpu >= NR_CPUS)) - return; - - /* Using ELF notes here is opportunistic. - * I need a well defined structure format - * for the data I pass, and I need tags - * on the data to indicate what information I have - * squirrelled away. ELF notes happen to provide - * all of that, no need to invent something new. - */ - - buf = (u32*)per_cpu_ptr(crash_notes, cpu); - - if (!buf) - return; - - memset(&prstatus, 0, sizeof(prstatus)); - prstatus.pr_pid = current->pid; - elf_core_copy_regs(&prstatus.pr_reg, regs); - buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, - sizeof(prstatus)); - final_note(buf); -} - -static void crash_save_self(struct pt_regs *regs) -{ - int cpu; - - cpu = smp_processor_id(); - crash_save_this_cpu(regs, cpu); -} - #ifdef CONFIG_SMP static atomic_t waiting_for_crash_ipi; @@ -117,7 +52,7 @@ static int crash_nmi_callback(struct notifier_block *self, return NOTIFY_STOP; local_irq_disable(); - crash_save_this_cpu(regs, cpu); + crash_save_cpu(regs, cpu); disable_local_APIC(); atomic_dec(&waiting_for_crash_ipi); /* Assume hlt works */ @@ -196,5 +131,5 @@ void machine_crash_shutdown(struct pt_regs *regs) disable_IO_APIC(); - crash_save_self(regs); + crash_save_cpu(regs, smp_processor_id()); } diff --git a/include/linux/kexec.h b/include/linux/kexec.h index a4ede62b339d..e3abcec6c51c 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -105,6 +105,7 @@ extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); extern void crash_kexec(struct pt_regs *); int kexec_should_crash(struct task_struct *); +void crash_save_cpu(struct pt_regs *regs, int cpu); extern struct kimage *kexec_image; extern struct kimage *kexec_crash_image; diff --git a/kernel/kexec.c b/kernel/kexec.c index d43692cf2321..afbbbe981be2 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -20,6 +20,8 @@ #include #include #include +#include +#include #include #include @@ -1066,6 +1068,60 @@ void crash_kexec(struct pt_regs *regs) } } +static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, + size_t data_len) +{ + struct elf_note note; + + note.n_namesz = strlen(name) + 1; + note.n_descsz = data_len; + note.n_type = type; + memcpy(buf, ¬e, sizeof(note)); + buf += (sizeof(note) + 3)/4; + memcpy(buf, name, note.n_namesz); + buf += (note.n_namesz + 3)/4; + memcpy(buf, data, note.n_descsz); + buf += (note.n_descsz + 3)/4; + + return buf; +} + +static void final_note(u32 *buf) +{ + struct elf_note note; + + note.n_namesz = 0; + note.n_descsz = 0; + note.n_type = 0; + memcpy(buf, ¬e, sizeof(note)); +} + +void crash_save_cpu(struct pt_regs *regs, int cpu) +{ + struct elf_prstatus prstatus; + u32 *buf; + + if ((cpu < 0) || (cpu >= NR_CPUS)) + return; + + /* Using ELF notes here is opportunistic. + * I need a well defined structure format + * for the data I pass, and I need tags + * on the data to indicate what information I have + * squirrelled away. ELF notes happen to provide + * all of that, so there is no need to invent something new. + */ + buf = (u32*)per_cpu_ptr(crash_notes, cpu); + if (!buf) + return; + memset(&prstatus, 0, sizeof(prstatus)); + prstatus.pr_pid = current->pid; + elf_core_copy_regs(&prstatus.pr_reg, regs); + buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, + sizeof(prstatus)); + final_note(buf); +} + static int __init crash_notes_memory_init(void) { /* Allocate memory for saving cpu registers. */ -- cgit v1.2.3 From 97d2a80584b30b5cd32da411deca1986ef61877a Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Wed, 6 Dec 2006 20:40:45 -0800 Subject: [PATCH] aio: remove ki_retried debugging member Remove the ki_retried member from struct kiocb. I think the idea was bounced around a while back, but Arnaldo pointed out another reason that we should dig it up when he pointed out that the last cacheline of struct kiocb only contains 4 bytes. By removing the debugging member, we save more than the 8 byte on 64 bit machines. Signed-off-by: Benjamin LaHaise Acked-by: Ken Chen Acked-by: Zach Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 15 --------------- include/linux/aio.h | 2 -- 2 files changed, 17 deletions(-) (limited to 'include/linux') diff --git a/fs/aio.c b/fs/aio.c index f02ad2677fc5..d3a6ec2c9627 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -666,17 +666,6 @@ static ssize_t aio_run_iocb(struct kiocb *iocb) ssize_t (*retry)(struct kiocb *); ssize_t ret; - if (iocb->ki_retried++ > 1024*1024) { - printk("Maximal retry count. Bytes done %Zd\n", - iocb->ki_nbytes - iocb->ki_left); - return -EAGAIN; - } - - if (!(iocb->ki_retried & 0xff)) { - pr_debug("%ld retry: %zd of %zd\n", iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); - } - if (!(retry = iocb->ki_retry)) { printk("aio_run_iocb: iocb->ki_retry = NULL\n"); return 0; @@ -1005,9 +994,6 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) kunmap_atomic(ring, KM_IRQ1); pr_debug("added to ring %p at [%lu]\n", iocb, tail); - - pr_debug("%ld retries: %zd of %zd\n", iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); @@ -1590,7 +1576,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, req->ki_opcode = iocb->aio_lio_opcode; init_waitqueue_func_entry(&req->ki_wait, aio_wake_function); INIT_LIST_HEAD(&req->ki_wait.task_list); - req->ki_retried = 0; ret = aio_setup_iocb(req); diff --git a/include/linux/aio.h b/include/linux/aio.h index 9e350fd44d77..3372ec6bf53a 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -111,7 +111,6 @@ struct kiocb { size_t ki_nbytes; /* copy of iocb->aio_nbytes */ char __user *ki_buf; /* remaining iocb->aio_buf */ size_t ki_left; /* remaining bytes */ - long ki_retried; /* just for testing */ struct iovec ki_inline_vec; /* inline vector */ struct iovec *ki_iovec; unsigned long ki_nr_segs; @@ -238,7 +237,6 @@ do { \ } while (0) #define io_wait_to_kiocb(wait) container_of(wait, struct kiocb, ki_wait) -#define is_retried_kiocb(iocb) ((iocb)->ki_retried > 1) #include -- cgit v1.2.3 From 6b39bb6548d60b9a18826134b5ccd5c3cef85fe2 Mon Sep 17 00:00:00 2001 From: Paul Clements Date: Wed, 6 Dec 2006 20:40:53 -0800 Subject: [PATCH] nbd: show nbd client pid in sysfs Allow nbd to expose the nbd-client daemon's PID in /sys/block/nbd/pid. This is helpful for tracking connection status of a device and for determining which nbd devices are currently in use. Signed-off-by: Paul Clements Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/nbd.c | 16 ++++++++++++++++ include/linux/nbd.h | 1 + 2 files changed, 17 insertions(+) (limited to 'include/linux') diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 9d1035e8d9d8..7bf2cfbd6285 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -355,14 +355,30 @@ harderror: return NULL; } +static ssize_t pid_show(struct gendisk *disk, char *page) +{ + return sprintf(page, "%ld\n", + (long) ((struct nbd_device *)disk->private_data)->pid); +} + +static struct disk_attribute pid_attr = { + .attr = { .name = "pid", .mode = S_IRUGO }, + .show = pid_show, +}; + static void nbd_do_it(struct nbd_device *lo) { struct request *req; BUG_ON(lo->magic != LO_MAGIC); + lo->pid = current->pid; + sysfs_create_file(&lo->disk->kobj, &pid_attr.attr); + while ((req = nbd_read_stat(lo)) != NULL) nbd_end_request(req); + + sysfs_remove_file(&lo->disk->kobj, &pid_attr.attr); return; } diff --git a/include/linux/nbd.h b/include/linux/nbd.h index d6b6dc09ad97..0f3e69302540 100644 --- a/include/linux/nbd.h +++ b/include/linux/nbd.h @@ -64,6 +64,7 @@ struct nbd_device { struct gendisk *disk; int blksize; u64 bytesize; + pid_t pid; /* pid of nbd-client, if attached */ }; #endif -- cgit v1.2.3 From 759643b874907e76ae81e34df62f41ab6683f5c2 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 6 Dec 2006 20:40:59 -0800 Subject: [PATCH] IPMI: pass sysfs name from lower level driver Pass in the sysfs name from the lower-level IPMI driver, as the coming IPMI serial driver will need that to link properly from the serial device sysfs directory. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 34 ++++++++++++++++++++++++++++------ drivers/char/ipmi/ipmi_si_intf.c | 1 + include/linux/ipmi_smi.h | 1 + 3 files changed, 30 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 761ed2699204..6a77b264eb2c 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -205,6 +205,7 @@ struct ipmi_smi struct bmc_device *bmc; char *my_dev_name; + char *sysfs_name; /* This is the lower-layer's sender routine. */ struct ipmi_smi_handlers *handlers; @@ -2004,7 +2005,11 @@ static void ipmi_bmc_unregister(ipmi_smi_t intf) { struct bmc_device *bmc = intf->bmc; - sysfs_remove_link(&intf->si_dev->kobj, "bmc"); + if (intf->sysfs_name) { + sysfs_remove_link(&intf->si_dev->kobj, intf->sysfs_name); + kfree(intf->sysfs_name); + intf->sysfs_name = NULL; + } if (intf->my_dev_name) { sysfs_remove_link(&bmc->dev->dev.kobj, intf->my_dev_name); kfree(intf->my_dev_name); @@ -2140,7 +2145,8 @@ out: return err; } -static int ipmi_bmc_register(ipmi_smi_t intf) +static int ipmi_bmc_register(ipmi_smi_t intf, int ifnum, + const char *sysfs_name) { int rv; struct bmc_device *bmc = intf->bmc; @@ -2257,29 +2263,44 @@ static int ipmi_bmc_register(ipmi_smi_t intf) * create symlink from system interface device to bmc device * and back. */ + intf->sysfs_name = kstrdup(sysfs_name, GFP_KERNEL); + if (!intf->sysfs_name) { + rv = -ENOMEM; + printk(KERN_ERR + "ipmi_msghandler: allocate link to BMC: %d\n", + rv); + goto out_err; + } + rv = sysfs_create_link(&intf->si_dev->kobj, - &bmc->dev->dev.kobj, "bmc"); + &bmc->dev->dev.kobj, intf->sysfs_name); if (rv) { + kfree(intf->sysfs_name); + intf->sysfs_name = NULL; printk(KERN_ERR "ipmi_msghandler: Unable to create bmc symlink: %d\n", rv); goto out_err; } - size = snprintf(dummy, 0, "ipmi%d", intf->intf_num); + size = snprintf(dummy, 0, "ipmi%d", ifnum); intf->my_dev_name = kmalloc(size+1, GFP_KERNEL); if (!intf->my_dev_name) { + kfree(intf->sysfs_name); + intf->sysfs_name = NULL; rv = -ENOMEM; printk(KERN_ERR "ipmi_msghandler: allocate link from BMC: %d\n", rv); goto out_err; } - snprintf(intf->my_dev_name, size+1, "ipmi%d", intf->intf_num); + snprintf(intf->my_dev_name, size+1, "ipmi%d", ifnum); rv = sysfs_create_link(&bmc->dev->dev.kobj, &intf->si_dev->kobj, intf->my_dev_name); if (rv) { + kfree(intf->sysfs_name); + intf->sysfs_name = NULL; kfree(intf->my_dev_name); intf->my_dev_name = NULL; printk(KERN_ERR @@ -2464,6 +2485,7 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, void *send_info, struct ipmi_device_id *device_id, struct device *si_dev, + const char *sysfs_name, unsigned char slave_addr) { int i, j; @@ -2579,7 +2601,7 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, if (rv == 0) rv = add_proc_entries(intf, i); - rv = ipmi_bmc_register(intf); + rv = ipmi_bmc_register(intf, i, sysfs_name); out: if (rv) { diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index bb1fac104fda..89cdb928061c 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2362,6 +2362,7 @@ static int try_smi_init(struct smi_info *new_smi) new_smi, &new_smi->device_id, new_smi->dev, + "bmc", new_smi->slave_addr); if (rv) { printk(KERN_ERR diff --git a/include/linux/ipmi_smi.h b/include/linux/ipmi_smi.h index 6d9c7e4da472..2cc960da4176 100644 --- a/include/linux/ipmi_smi.h +++ b/include/linux/ipmi_smi.h @@ -173,6 +173,7 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, void *send_info, struct ipmi_device_id *device_id, struct device *dev, + const char *sysfs_name, unsigned char slave_addr); /* -- cgit v1.2.3 From b9675136e2ad95156fb93be6155f17590bb26fd7 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 6 Dec 2006 20:41:02 -0800 Subject: [PATCH] IPMI: Add maintenance mode Some commands and operations on a BMC can cause the BMC to "go away" for a while. This can cause the automatic flag processing and other things of that nature to timeout and generate annoying logs, or possibly cause other bad things to happen when in firmware update mode. Add detection of those commands (cold reset, warm reset, and any firmware command) and turns off automatic processing for 30 seconds. It also add a manual override either way. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_devintf.c | 25 ++++++++ drivers/char/ipmi/ipmi_msghandler.c | 117 +++++++++++++++++++++++++++++++++++- drivers/char/ipmi/ipmi_si_intf.c | 9 +++ include/linux/ipmi.h | 45 ++++++++++++++ include/linux/ipmi_msgdefs.h | 5 ++ include/linux/ipmi_smi.h | 7 +++ 6 files changed, 207 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/char/ipmi/ipmi_devintf.c b/drivers/char/ipmi/ipmi_devintf.c index 81fcf0ce21d1..375d3378eecd 100644 --- a/drivers/char/ipmi/ipmi_devintf.c +++ b/drivers/char/ipmi/ipmi_devintf.c @@ -596,6 +596,31 @@ static int ipmi_ioctl(struct inode *inode, rv = 0; break; } + + case IPMICTL_GET_MAINTENANCE_MODE_CMD: + { + int mode; + + mode = ipmi_get_maintenance_mode(priv->user); + if (copy_to_user(arg, &mode, sizeof(mode))) { + rv = -EFAULT; + break; + } + rv = 0; + break; + } + + case IPMICTL_SET_MAINTENANCE_MODE_CMD: + { + int mode; + + if (copy_from_user(&mode, arg, sizeof(mode))) { + rv = -EFAULT; + break; + } + rv = ipmi_set_maintenance_mode(priv->user, mode); + break; + } } return rv; diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 03f32611831d..ff0e68f0386b 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -48,7 +48,7 @@ #define PFX "IPMI message handler: " -#define IPMI_DRIVER_VERSION "39.0" +#define IPMI_DRIVER_VERSION "39.1" static struct ipmi_recv_msg *ipmi_alloc_recv_msg(void); static int ipmi_init_msghandler(void); @@ -59,6 +59,9 @@ static int initialized = 0; static struct proc_dir_entry *proc_ipmi_root = NULL; #endif /* CONFIG_PROC_FS */ +/* Remain in auto-maintenance mode for this amount of time (in ms). */ +#define IPMI_MAINTENANCE_MODE_TIMEOUT 30000 + #define MAX_EVENTS_IN_QUEUE 25 /* Don't let a message sit in a queue forever, always time it with at lest @@ -262,6 +265,12 @@ struct ipmi_smi unsigned char local_sel_device; unsigned char local_event_generator; + /* For handling of maintenance mode. */ + int maintenance_mode; + int maintenance_mode_enable; + int auto_maintenance_timeout; + spinlock_t maintenance_mode_lock; /* Used in a timer... */ + /* A cheap hack, if this is non-null and a message to an interface comes in with a NULL user, call this routine with it. Note that the message will still be freed by the @@ -985,6 +994,65 @@ int ipmi_get_my_LUN(ipmi_user_t user, return 0; } +int ipmi_get_maintenance_mode(ipmi_user_t user) +{ + int mode; + unsigned long flags; + + spin_lock_irqsave(&user->intf->maintenance_mode_lock, flags); + mode = user->intf->maintenance_mode; + spin_unlock_irqrestore(&user->intf->maintenance_mode_lock, flags); + + return mode; +} +EXPORT_SYMBOL(ipmi_get_maintenance_mode); + +static void maintenance_mode_update(ipmi_smi_t intf) +{ + if (intf->handlers->set_maintenance_mode) + intf->handlers->set_maintenance_mode( + intf->send_info, intf->maintenance_mode_enable); +} + +int ipmi_set_maintenance_mode(ipmi_user_t user, int mode) +{ + int rv = 0; + unsigned long flags; + ipmi_smi_t intf = user->intf; + + spin_lock_irqsave(&intf->maintenance_mode_lock, flags); + if (intf->maintenance_mode != mode) { + switch (mode) { + case IPMI_MAINTENANCE_MODE_AUTO: + intf->maintenance_mode = mode; + intf->maintenance_mode_enable + = (intf->auto_maintenance_timeout > 0); + break; + + case IPMI_MAINTENANCE_MODE_OFF: + intf->maintenance_mode = mode; + intf->maintenance_mode_enable = 0; + break; + + case IPMI_MAINTENANCE_MODE_ON: + intf->maintenance_mode = mode; + intf->maintenance_mode_enable = 1; + break; + + default: + rv = -EINVAL; + goto out_unlock; + } + + maintenance_mode_update(intf); + } + out_unlock: + spin_unlock_irqrestore(&intf->maintenance_mode_lock, flags); + + return rv; +} +EXPORT_SYMBOL(ipmi_set_maintenance_mode); + int ipmi_set_gets_events(ipmi_user_t user, int val) { unsigned long flags; @@ -1322,6 +1390,24 @@ static int i_ipmi_request(ipmi_user_t user, goto out_err; } + if (((msg->netfn == IPMI_NETFN_APP_REQUEST) + && ((msg->cmd == IPMI_COLD_RESET_CMD) + || (msg->cmd == IPMI_WARM_RESET_CMD))) + || (msg->netfn == IPMI_NETFN_FIRMWARE_REQUEST)) + { + spin_lock_irqsave(&intf->maintenance_mode_lock, flags); + intf->auto_maintenance_timeout + = IPMI_MAINTENANCE_MODE_TIMEOUT; + if (!intf->maintenance_mode + && !intf->maintenance_mode_enable) + { + intf->maintenance_mode_enable = 1; + maintenance_mode_update(intf); + } + spin_unlock_irqrestore(&intf->maintenance_mode_lock, + flags); + } + if ((msg->data_len + 2) > IPMI_MAX_MSG_LENGTH) { spin_lock_irqsave(&intf->counter_lock, flags); intf->sent_invalid_commands++; @@ -2605,6 +2691,7 @@ int ipmi_register_smi(struct ipmi_smi_handlers *handlers, INIT_LIST_HEAD(&intf->waiting_events); intf->waiting_events_count = 0; mutex_init(&intf->cmd_rcvrs_mutex); + spin_lock_init(&intf->maintenance_mode_lock); INIT_LIST_HEAD(&intf->cmd_rcvrs); init_waitqueue_head(&intf->waitq); @@ -3593,6 +3680,30 @@ static void ipmi_timeout_handler(long timeout_period) list_for_each_entry_safe(msg, msg2, &timeouts, link) deliver_err_response(msg, IPMI_TIMEOUT_COMPLETION_CODE); + + /* + * Maintenance mode handling. Check the timeout + * optimistically before we claim the lock. It may + * mean a timeout gets missed occasionally, but that + * only means the timeout gets extended by one period + * in that case. No big deal, and it avoids the lock + * most of the time. + */ + if (intf->auto_maintenance_timeout > 0) { + spin_lock_irqsave(&intf->maintenance_mode_lock, flags); + if (intf->auto_maintenance_timeout > 0) { + intf->auto_maintenance_timeout + -= timeout_period; + if (!intf->maintenance_mode + && (intf->auto_maintenance_timeout <= 0)) + { + intf->maintenance_mode_enable = 0; + maintenance_mode_update(intf); + } + } + spin_unlock_irqrestore(&intf->maintenance_mode_lock, + flags); + } } rcu_read_unlock(); } @@ -3606,6 +3717,10 @@ static void ipmi_request_event(void) /* Called from the timer, no need to check if handlers is * valid. */ list_for_each_entry_rcu(intf, &ipmi_interfaces, link) { + /* No event requests when in maintenance mode. */ + if (intf->maintenance_mode_enable) + continue; + handlers = intf->handlers; if (handlers) handlers->request_events(intf->send_info); diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 89cdb928061c..f04bee76ba2b 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -949,12 +949,21 @@ static int smi_start_processing(void *send_info, return 0; } +static void set_maintenance_mode(void *send_info, int enable) +{ + struct smi_info *smi_info = send_info; + + if (!enable) + atomic_set(&smi_info->req_events, 0); +} + static struct ipmi_smi_handlers handlers = { .owner = THIS_MODULE, .start_processing = smi_start_processing, .sender = sender, .request_events = request_events, + .set_maintenance_mode = set_maintenance_mode, .set_run_to_completion = set_run_to_completion, .poll = poll, }; diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 796ca009fd46..7a9db390c56a 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -208,6 +208,15 @@ struct kernel_ipmi_msg code as the first byte of the incoming data, unlike a response. */ +/* + * Modes for ipmi_set_maint_mode() and the userland IOCTL. The AUTO + * setting is the default and means it will be set on certain + * commands. Hard setting it on and off will override automatic + * operation. + */ +#define IPMI_MAINTENANCE_MODE_AUTO 0 +#define IPMI_MAINTENANCE_MODE_OFF 1 +#define IPMI_MAINTENANCE_MODE_ON 2 #ifdef __KERNEL__ @@ -373,6 +382,35 @@ int ipmi_unregister_for_cmd(ipmi_user_t user, unsigned char cmd, unsigned int chans); +/* + * Go into a mode where the driver will not autonomously attempt to do + * things with the interface. It will still respond to attentions and + * interrupts, and it will expect that commands will complete. It + * will not automatcially check for flags, events, or things of that + * nature. + * + * This is primarily used for firmware upgrades. The idea is that + * when you go into firmware upgrade mode, you do this operation + * and the driver will not attempt to do anything but what you tell + * it or what the BMC asks for. + * + * Note that if you send a command that resets the BMC, the driver + * will still expect a response from that command. So the BMC should + * reset itself *after* the response is sent. Resetting before the + * response is just silly. + * + * If in auto maintenance mode, the driver will automatically go into + * maintenance mode for 30 seconds if it sees a cold reset, a warm + * reset, or a firmware NetFN. This means that code that uses only + * firmware NetFN commands to do upgrades will work automatically + * without change, assuming it sends a message every 30 seconds or + * less. + * + * See the IPMI_MAINTENANCE_MODE_xxx defines for what the mode means. + */ +int ipmi_get_maintenance_mode(ipmi_user_t user); +int ipmi_set_maintenance_mode(ipmi_user_t user, int mode); + /* * Allow run-to-completion mode to be set for the interface of * a specific user. @@ -656,4 +694,11 @@ struct ipmi_timing_parms #define IPMICTL_GET_TIMING_PARMS_CMD _IOR(IPMI_IOC_MAGIC, 23, \ struct ipmi_timing_parms) +/* + * Set the maintenance mode. See ipmi_set_maintenance_mode() above + * for a description of what this does. + */ +#define IPMICTL_GET_MAINTENANCE_MODE_CMD _IOR(IPMI_IOC_MAGIC, 30, int) +#define IPMICTL_SET_MAINTENANCE_MODE_CMD _IOW(IPMI_IOC_MAGIC, 31, int) + #endif /* __LINUX_IPMI_H */ diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index 4d04d8b58a0a..8d6759cc1a71 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -46,6 +46,8 @@ #define IPMI_NETFN_APP_REQUEST 0x06 #define IPMI_NETFN_APP_RESPONSE 0x07 #define IPMI_GET_DEVICE_ID_CMD 0x01 +#define IPMI_COLD_RESET_CMD 0x02 +#define IPMI_WARM_RESET_CMD 0x03 #define IPMI_CLEAR_MSG_FLAGS_CMD 0x30 #define IPMI_GET_DEVICE_GUID_CMD 0x08 #define IPMI_GET_MSG_FLAGS_CMD 0x31 @@ -60,6 +62,9 @@ #define IPMI_NETFN_STORAGE_RESPONSE 0x0b #define IPMI_ADD_SEL_ENTRY_CMD 0x44 +#define IPMI_NETFN_FIRMWARE_REQUEST 0x08 +#define IPMI_NETFN_FIRMWARE_RESPONSE 0x09 + /* The default slave address */ #define IPMI_BMC_SLAVE_ADDR 0x20 diff --git a/include/linux/ipmi_smi.h b/include/linux/ipmi_smi.h index 2cc960da4176..c0633108d05d 100644 --- a/include/linux/ipmi_smi.h +++ b/include/linux/ipmi_smi.h @@ -115,6 +115,13 @@ struct ipmi_smi_handlers poll for operations during things like crash dumps. */ void (*poll)(void *send_info); + /* Enable/disable firmware maintenance mode. Note that this + is *not* the modes defined, this is simply an on/off + setting. The message handler does the mode handling. Note + that this is called from interupt context, so it cannot + block. */ + void (*set_maintenance_mode)(void *send_info, int enable); + /* Tell the handler that we are using it/not using it. The message handler get the modules that this handler belongs to; this function lets the SMI claim any modules that it -- cgit v1.2.3 From 4d7cbac7c870ca66d8fb27d68188efbb5de2dffa Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Wed, 6 Dec 2006 20:41:14 -0800 Subject: [PATCH] IPMI: Fix BT long busy The IPMI BT subdriver has been patched to survive "long busy" timeouts seen during firmware upgrades and resets. The patch never returns the HOSED state, synthesizes response messages with meaningful completion codes, and recovers gracefully when the hardware finishes the long busy. The subdriver now issues a "Get BT Capabilities" command and properly uses those results. More informative completion codes are returned on error from transaction starts; this logic was propogated to the KCS and SMIC subdrivers. Finally, indent and other style quirks were normalized. Signed-off-by: Rocky Craig Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_bt_sm.c | 641 ++++++++++++++++++++++++--------------- drivers/char/ipmi/ipmi_kcs_sm.c | 14 +- drivers/char/ipmi/ipmi_si_intf.c | 12 +- drivers/char/ipmi/ipmi_smic_sm.c | 14 +- include/linux/ipmi_msgdefs.h | 8 +- 5 files changed, 423 insertions(+), 266 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/ipmi/ipmi_bt_sm.c b/drivers/char/ipmi/ipmi_bt_sm.c index 0030cd8e2e95..6c59baa887a8 100644 --- a/drivers/char/ipmi/ipmi_bt_sm.c +++ b/drivers/char/ipmi/ipmi_bt_sm.c @@ -33,11 +33,13 @@ #include /* for completion codes */ #include "ipmi_si_sm.h" -static int bt_debug = 0x00; /* Production value 0, see following flags */ +#define BT_DEBUG_OFF 0 /* Used in production */ +#define BT_DEBUG_ENABLE 1 /* Generic messages */ +#define BT_DEBUG_MSG 2 /* Prints all request/response buffers */ +#define BT_DEBUG_STATES 4 /* Verbose look at state changes */ + +static int bt_debug = BT_DEBUG_OFF; -#define BT_DEBUG_ENABLE 1 -#define BT_DEBUG_MSG 2 -#define BT_DEBUG_STATES 4 module_param(bt_debug, int, 0644); MODULE_PARM_DESC(bt_debug, "debug bitmask, 1=enable, 2=messages, 4=states"); @@ -47,38 +49,54 @@ MODULE_PARM_DESC(bt_debug, "debug bitmask, 1=enable, 2=messages, 4=states"); Since the Open IPMI architecture is single-message oriented at this stage, the queue depth of BT is of no concern. */ -#define BT_NORMAL_TIMEOUT 5000000 /* seconds in microseconds */ -#define BT_RETRY_LIMIT 2 -#define BT_RESET_DELAY 6000000 /* 6 seconds after warm reset */ +#define BT_NORMAL_TIMEOUT 5 /* seconds */ +#define BT_NORMAL_RETRY_LIMIT 2 +#define BT_RESET_DELAY 6 /* seconds after warm reset */ + +/* States are written in chronological order and usually cover + multiple rows of the state table discussion in the IPMI spec. */ enum bt_states { - BT_STATE_IDLE, + BT_STATE_IDLE = 0, /* Order is critical in this list */ BT_STATE_XACTION_START, BT_STATE_WRITE_BYTES, - BT_STATE_WRITE_END, BT_STATE_WRITE_CONSUME, - BT_STATE_B2H_WAIT, - BT_STATE_READ_END, - BT_STATE_RESET1, /* These must come last */ + BT_STATE_READ_WAIT, + BT_STATE_CLEAR_B2H, + BT_STATE_READ_BYTES, + BT_STATE_RESET1, /* These must come last */ BT_STATE_RESET2, BT_STATE_RESET3, BT_STATE_RESTART, - BT_STATE_HOSED + BT_STATE_PRINTME, + BT_STATE_CAPABILITIES_BEGIN, + BT_STATE_CAPABILITIES_END, + BT_STATE_LONG_BUSY /* BT doesn't get hosed :-) */ }; +/* Macros seen at the end of state "case" blocks. They help with legibility + and debugging. */ + +#define BT_STATE_CHANGE(X,Y) { bt->state = X; return Y; } + +#define BT_SI_SM_RETURN(Y) { last_printed = BT_STATE_PRINTME; return Y; } + struct si_sm_data { enum bt_states state; - enum bt_states last_state; /* assist printing and resets */ unsigned char seq; /* BT sequence number */ struct si_sm_io *io; - unsigned char write_data[IPMI_MAX_MSG_LENGTH]; - int write_count; - unsigned char read_data[IPMI_MAX_MSG_LENGTH]; - int read_count; - int truncated; - long timeout; - unsigned int error_retries; /* end of "common" fields */ + unsigned char write_data[IPMI_MAX_MSG_LENGTH]; + int write_count; + unsigned char read_data[IPMI_MAX_MSG_LENGTH]; + int read_count; + int truncated; + long timeout; /* microseconds countdown */ + int error_retries; /* end of "common" fields */ int nonzero_status; /* hung BMCs stay all 0 */ + enum bt_states complete; /* to divert the state machine */ + int BT_CAP_outreqs; + long BT_CAP_req2rsp; + int BT_CAP_retries; /* Recommended retries */ }; #define BT_CLR_WR_PTR 0x01 /* See IPMI 1.5 table 11.6.4 */ @@ -111,86 +129,118 @@ struct si_sm_data { static char *state2txt(unsigned char state) { switch (state) { - case BT_STATE_IDLE: return("IDLE"); - case BT_STATE_XACTION_START: return("XACTION"); - case BT_STATE_WRITE_BYTES: return("WR_BYTES"); - case BT_STATE_WRITE_END: return("WR_END"); - case BT_STATE_WRITE_CONSUME: return("WR_CONSUME"); - case BT_STATE_B2H_WAIT: return("B2H_WAIT"); - case BT_STATE_READ_END: return("RD_END"); - case BT_STATE_RESET1: return("RESET1"); - case BT_STATE_RESET2: return("RESET2"); - case BT_STATE_RESET3: return("RESET3"); - case BT_STATE_RESTART: return("RESTART"); - case BT_STATE_HOSED: return("HOSED"); + case BT_STATE_IDLE: return("IDLE"); + case BT_STATE_XACTION_START: return("XACTION"); + case BT_STATE_WRITE_BYTES: return("WR_BYTES"); + case BT_STATE_WRITE_CONSUME: return("WR_CONSUME"); + case BT_STATE_READ_WAIT: return("RD_WAIT"); + case BT_STATE_CLEAR_B2H: return("CLEAR_B2H"); + case BT_STATE_READ_BYTES: return("RD_BYTES"); + case BT_STATE_RESET1: return("RESET1"); + case BT_STATE_RESET2: return("RESET2"); + case BT_STATE_RESET3: return("RESET3"); + case BT_STATE_RESTART: return("RESTART"); + case BT_STATE_LONG_BUSY: return("LONG_BUSY"); + case BT_STATE_CAPABILITIES_BEGIN: return("CAP_BEGIN"); + case BT_STATE_CAPABILITIES_END: return("CAP_END"); } return("BAD STATE"); } #define STATE2TXT state2txt(bt->state) -static char *status2txt(unsigned char status, char *buf) +static char *status2txt(unsigned char status) { + /* + * This cannot be called by two threads at the same time and + * the buffer is always consumed immediately, so the static is + * safe to use. + */ + static char buf[40]; + strcpy(buf, "[ "); - if (status & BT_B_BUSY) strcat(buf, "B_BUSY "); - if (status & BT_H_BUSY) strcat(buf, "H_BUSY "); - if (status & BT_OEM0) strcat(buf, "OEM0 "); - if (status & BT_SMS_ATN) strcat(buf, "SMS "); - if (status & BT_B2H_ATN) strcat(buf, "B2H "); - if (status & BT_H2B_ATN) strcat(buf, "H2B "); + if (status & BT_B_BUSY) + strcat(buf, "B_BUSY "); + if (status & BT_H_BUSY) + strcat(buf, "H_BUSY "); + if (status & BT_OEM0) + strcat(buf, "OEM0 "); + if (status & BT_SMS_ATN) + strcat(buf, "SMS "); + if (status & BT_B2H_ATN) + strcat(buf, "B2H "); + if (status & BT_H2B_ATN) + strcat(buf, "H2B "); strcat(buf, "]"); return buf; } -#define STATUS2TXT(buf) status2txt(status, buf) +#define STATUS2TXT status2txt(status) + +/* called externally at insmod time, and internally on cleanup */ -/* This will be called from within this module on a hosed condition */ -#define FIRST_SEQ 0 static unsigned int bt_init_data(struct si_sm_data *bt, struct si_sm_io *io) { - bt->state = BT_STATE_IDLE; - bt->last_state = BT_STATE_IDLE; - bt->seq = FIRST_SEQ; - bt->io = io; - bt->write_count = 0; - bt->read_count = 0; - bt->error_retries = 0; - bt->nonzero_status = 0; - bt->truncated = 0; - bt->timeout = BT_NORMAL_TIMEOUT; + memset(bt, 0, sizeof(struct si_sm_data)); + if (bt->io != io) { /* external: one-time only things */ + bt->io = io; + bt->seq = 0; + } + bt->state = BT_STATE_IDLE; /* start here */ + bt->complete = BT_STATE_IDLE; /* end here */ + bt->BT_CAP_req2rsp = BT_NORMAL_TIMEOUT * 1000000; + bt->BT_CAP_retries = BT_NORMAL_RETRY_LIMIT; + /* BT_CAP_outreqs == zero is a flag to read BT Capabilities */ return 3; /* We claim 3 bytes of space; ought to check SPMI table */ } +/* Jam a completion code (probably an error) into a response */ + +static void force_result(struct si_sm_data *bt, unsigned char completion_code) +{ + bt->read_data[0] = 4; /* # following bytes */ + bt->read_data[1] = bt->write_data[1] | 4; /* Odd NetFn/LUN */ + bt->read_data[2] = bt->write_data[2]; /* seq (ignored) */ + bt->read_data[3] = bt->write_data[3]; /* Command */ + bt->read_data[4] = completion_code; + bt->read_count = 5; +} + +/* The upper state machine starts here */ + static int bt_start_transaction(struct si_sm_data *bt, unsigned char *data, unsigned int size) { unsigned int i; - if ((size < 2) || (size > (IPMI_MAX_MSG_LENGTH - 2))) - return -1; + if (size < 2) + return IPMI_REQ_LEN_INVALID_ERR; + if (size > IPMI_MAX_MSG_LENGTH) + return IPMI_REQ_LEN_EXCEEDED_ERR; - if ((bt->state != BT_STATE_IDLE) && (bt->state != BT_STATE_HOSED)) - return -2; + if (bt->state == BT_STATE_LONG_BUSY) + return IPMI_NODE_BUSY_ERR; + + if (bt->state != BT_STATE_IDLE) + return IPMI_NOT_IN_MY_STATE_ERR; if (bt_debug & BT_DEBUG_MSG) { - printk(KERN_WARNING "+++++++++++++++++++++++++++++++++++++\n"); - printk(KERN_WARNING "BT: write seq=0x%02X:", bt->seq); + printk(KERN_WARNING "BT: +++++++++++++++++ New command\n"); + printk(KERN_WARNING "BT: NetFn/LUN CMD [%d data]:", size - 2); for (i = 0; i < size; i ++) - printk (" %02x", data[i]); + printk (" %02x", data[i]); printk("\n"); } bt->write_data[0] = size + 1; /* all data plus seq byte */ bt->write_data[1] = *data; /* NetFn/LUN */ - bt->write_data[2] = bt->seq; + bt->write_data[2] = bt->seq++; memcpy(bt->write_data + 3, data + 1, size - 1); bt->write_count = size + 2; - bt->error_retries = 0; bt->nonzero_status = 0; - bt->read_count = 0; bt->truncated = 0; bt->state = BT_STATE_XACTION_START; - bt->last_state = BT_STATE_IDLE; - bt->timeout = BT_NORMAL_TIMEOUT; + bt->timeout = bt->BT_CAP_req2rsp; + force_result(bt, IPMI_ERR_UNSPECIFIED); return 0; } @@ -198,38 +248,30 @@ static int bt_start_transaction(struct si_sm_data *bt, it calls this. Strip out the length and seq bytes. */ static int bt_get_result(struct si_sm_data *bt, - unsigned char *data, - unsigned int length) + unsigned char *data, + unsigned int length) { int i, msg_len; msg_len = bt->read_count - 2; /* account for length & seq */ - /* Always NetFn, Cmd, cCode */ if (msg_len < 3 || msg_len > IPMI_MAX_MSG_LENGTH) { - printk(KERN_DEBUG "BT results: bad msg_len = %d\n", msg_len); - data[0] = bt->write_data[1] | 0x4; /* Kludge a response */ - data[1] = bt->write_data[3]; - data[2] = IPMI_ERR_UNSPECIFIED; + force_result(bt, IPMI_ERR_UNSPECIFIED); msg_len = 3; - } else { - data[0] = bt->read_data[1]; - data[1] = bt->read_data[3]; - if (length < msg_len) - bt->truncated = 1; - if (bt->truncated) { /* can be set in read_all_bytes() */ - data[2] = IPMI_ERR_MSG_TRUNCATED; - msg_len = 3; - } else - memcpy(data + 2, bt->read_data + 4, msg_len - 2); + } + data[0] = bt->read_data[1]; + data[1] = bt->read_data[3]; + if (length < msg_len || bt->truncated) { + data[2] = IPMI_ERR_MSG_TRUNCATED; + msg_len = 3; + } else + memcpy(data + 2, bt->read_data + 4, msg_len - 2); - if (bt_debug & BT_DEBUG_MSG) { - printk (KERN_WARNING "BT: res (raw)"); - for (i = 0; i < msg_len; i++) - printk(" %02x", data[i]); - printk ("\n"); - } + if (bt_debug & BT_DEBUG_MSG) { + printk (KERN_WARNING "BT: result %d bytes:", msg_len); + for (i = 0; i < msg_len; i++) + printk(" %02x", data[i]); + printk ("\n"); } - bt->read_count = 0; /* paranoia */ return msg_len; } @@ -238,22 +280,40 @@ static int bt_get_result(struct si_sm_data *bt, static void reset_flags(struct si_sm_data *bt) { + if (bt_debug) + printk(KERN_WARNING "IPMI BT: flag reset %s\n", + status2txt(BT_STATUS)); if (BT_STATUS & BT_H_BUSY) - BT_CONTROL(BT_H_BUSY); - if (BT_STATUS & BT_B_BUSY) - BT_CONTROL(BT_B_BUSY); - BT_CONTROL(BT_CLR_WR_PTR); - BT_CONTROL(BT_SMS_ATN); - - if (BT_STATUS & BT_B2H_ATN) { - int i; - BT_CONTROL(BT_H_BUSY); - BT_CONTROL(BT_B2H_ATN); - BT_CONTROL(BT_CLR_RD_PTR); - for (i = 0; i < IPMI_MAX_MSG_LENGTH + 2; i++) - BMC2HOST; - BT_CONTROL(BT_H_BUSY); - } + BT_CONTROL(BT_H_BUSY); /* force clear */ + BT_CONTROL(BT_CLR_WR_PTR); /* always reset */ + BT_CONTROL(BT_SMS_ATN); /* always clear */ + BT_INTMASK_W(BT_BMC_HWRST); +} + +/* Get rid of an unwanted/stale response. This should only be needed for + BMCs that support multiple outstanding requests. */ + +static void drain_BMC2HOST(struct si_sm_data *bt) +{ + int i, size; + + if (!(BT_STATUS & BT_B2H_ATN)) /* Not signalling a response */ + return; + + BT_CONTROL(BT_H_BUSY); /* now set */ + BT_CONTROL(BT_B2H_ATN); /* always clear */ + BT_STATUS; /* pause */ + BT_CONTROL(BT_B2H_ATN); /* some BMCs are stubborn */ + BT_CONTROL(BT_CLR_RD_PTR); /* always reset */ + if (bt_debug) + printk(KERN_WARNING "IPMI BT: stale response %s; ", + status2txt(BT_STATUS)); + size = BMC2HOST; + for (i = 0; i < size ; i++) + BMC2HOST; + BT_CONTROL(BT_H_BUSY); /* now clear */ + if (bt_debug) + printk("drained %d bytes\n", size + 1); } static inline void write_all_bytes(struct si_sm_data *bt) @@ -261,201 +321,256 @@ static inline void write_all_bytes(struct si_sm_data *bt) int i; if (bt_debug & BT_DEBUG_MSG) { - printk(KERN_WARNING "BT: write %d bytes seq=0x%02X", + printk(KERN_WARNING "BT: write %d bytes seq=0x%02X", bt->write_count, bt->seq); for (i = 0; i < bt->write_count; i++) printk (" %02x", bt->write_data[i]); printk ("\n"); } for (i = 0; i < bt->write_count; i++) - HOST2BMC(bt->write_data[i]); + HOST2BMC(bt->write_data[i]); } static inline int read_all_bytes(struct si_sm_data *bt) { unsigned char i; + /* length is "framing info", minimum = 4: NetFn, Seq, Cmd, cCode. + Keep layout of first four bytes aligned with write_data[] */ + bt->read_data[0] = BMC2HOST; bt->read_count = bt->read_data[0]; - if (bt_debug & BT_DEBUG_MSG) - printk(KERN_WARNING "BT: read %d bytes:", bt->read_count); - /* minimum: length, NetFn, Seq, Cmd, cCode == 5 total, or 4 more - following the length byte. */ if (bt->read_count < 4 || bt->read_count >= IPMI_MAX_MSG_LENGTH) { if (bt_debug & BT_DEBUG_MSG) - printk("bad length %d\n", bt->read_count); + printk(KERN_WARNING "BT: bad raw rsp len=%d\n", + bt->read_count); bt->truncated = 1; return 1; /* let next XACTION START clean it up */ } for (i = 1; i <= bt->read_count; i++) - bt->read_data[i] = BMC2HOST; - bt->read_count++; /* account for the length byte */ + bt->read_data[i] = BMC2HOST; + bt->read_count++; /* Account internally for length byte */ if (bt_debug & BT_DEBUG_MSG) { - for (i = 0; i < bt->read_count; i++) + int max = bt->read_count; + + printk(KERN_WARNING "BT: got %d bytes seq=0x%02X", + max, bt->read_data[2]); + if (max > 16) + max = 16; + for (i = 0; i < max; i++) printk (" %02x", bt->read_data[i]); - printk ("\n"); + printk ("%s\n", bt->read_count == max ? "" : " ..."); } - if (bt->seq != bt->write_data[2]) /* idiot check */ - printk(KERN_DEBUG "BT: internal error: sequence mismatch\n"); - /* per the spec, the (NetFn, Seq, Cmd) tuples should match */ - if ((bt->read_data[3] == bt->write_data[3]) && /* Cmd */ - (bt->read_data[2] == bt->write_data[2]) && /* Sequence */ - ((bt->read_data[1] & 0xF8) == (bt->write_data[1] & 0xF8))) + /* per the spec, the (NetFn[1], Seq[2], Cmd[3]) tuples must match */ + if ((bt->read_data[3] == bt->write_data[3]) && + (bt->read_data[2] == bt->write_data[2]) && + ((bt->read_data[1] & 0xF8) == (bt->write_data[1] & 0xF8))) return 1; if (bt_debug & BT_DEBUG_MSG) - printk(KERN_WARNING "BT: bad packet: " + printk(KERN_WARNING "IPMI BT: bad packet: " "want 0x(%02X, %02X, %02X) got (%02X, %02X, %02X)\n", - bt->write_data[1], bt->write_data[2], bt->write_data[3], + bt->write_data[1] | 0x04, bt->write_data[2], bt->write_data[3], bt->read_data[1], bt->read_data[2], bt->read_data[3]); return 0; } -/* Modifies bt->state appropriately, need to get into the bt_event() switch */ +/* Restart if retries are left, or return an error completion code */ -static void error_recovery(struct si_sm_data *bt, char *reason) +static enum si_sm_result error_recovery(struct si_sm_data *bt, + unsigned char status, + unsigned char cCode) { - unsigned char status; - char buf[40]; /* For getting status */ + char *reason; - bt->timeout = BT_NORMAL_TIMEOUT; /* various places want to retry */ + bt->timeout = bt->BT_CAP_req2rsp; - status = BT_STATUS; - printk(KERN_DEBUG "BT: %s in %s %s\n", reason, STATE2TXT, - STATUS2TXT(buf)); + switch (cCode) { + case IPMI_TIMEOUT_ERR: + reason = "timeout"; + break; + default: + reason = "internal error"; + break; + } + + printk(KERN_WARNING "IPMI BT: %s in %s %s ", /* open-ended line */ + reason, STATE2TXT, STATUS2TXT); + /* Per the IPMI spec, retries are based on the sequence number + known only to this module, so manage a restart here. */ (bt->error_retries)++; - if (bt->error_retries > BT_RETRY_LIMIT) { - printk(KERN_DEBUG "retry limit (%d) exceeded\n", BT_RETRY_LIMIT); - bt->state = BT_STATE_HOSED; - if (!bt->nonzero_status) - printk(KERN_ERR "IPMI: BT stuck, try power cycle\n"); - else if (bt->error_retries <= BT_RETRY_LIMIT + 1) { - printk(KERN_DEBUG "IPMI: BT reset (takes 5 secs)\n"); - bt->state = BT_STATE_RESET1; - } - return; + if (bt->error_retries < bt->BT_CAP_retries) { + printk("%d retries left\n", + bt->BT_CAP_retries - bt->error_retries); + bt->state = BT_STATE_RESTART; + return SI_SM_CALL_WITHOUT_DELAY; } - /* Sometimes the BMC queues get in an "off-by-one" state...*/ - if ((bt->state == BT_STATE_B2H_WAIT) && (status & BT_B2H_ATN)) { - printk(KERN_DEBUG "retry B2H_WAIT\n"); - return; + printk("failed %d retries, sending error response\n", + bt->BT_CAP_retries); + if (!bt->nonzero_status) + printk(KERN_ERR "IPMI BT: stuck, try power cycle\n"); + + /* this is most likely during insmod */ + else if (bt->seq <= (unsigned char)(bt->BT_CAP_retries & 0xFF)) { + printk(KERN_WARNING "IPMI: BT reset (takes 5 secs)\n"); + bt->state = BT_STATE_RESET1; + return SI_SM_CALL_WITHOUT_DELAY; } - printk(KERN_DEBUG "restart command\n"); - bt->state = BT_STATE_RESTART; + /* Concoct a useful error message, set up the next state, and + be done with this sequence. */ + + bt->state = BT_STATE_IDLE; + switch (cCode) { + case IPMI_TIMEOUT_ERR: + if (status & BT_B_BUSY) { + cCode = IPMI_NODE_BUSY_ERR; + bt->state = BT_STATE_LONG_BUSY; + } + break; + default: + break; + } + force_result(bt, cCode); + return SI_SM_TRANSACTION_COMPLETE; } -/* Check the status and (possibly) advance the BT state machine. The - default return is SI_SM_CALL_WITH_DELAY. */ +/* Check status and (usually) take action and change this state machine. */ static enum si_sm_result bt_event(struct si_sm_data *bt, long time) { - unsigned char status; - char buf[40]; /* For getting status */ + unsigned char status, BT_CAP[8]; + static enum bt_states last_printed = BT_STATE_PRINTME; int i; status = BT_STATUS; bt->nonzero_status |= status; - - if ((bt_debug & BT_DEBUG_STATES) && (bt->state != bt->last_state)) + if ((bt_debug & BT_DEBUG_STATES) && (bt->state != last_printed)) { printk(KERN_WARNING "BT: %s %s TO=%ld - %ld \n", STATE2TXT, - STATUS2TXT(buf), + STATUS2TXT, bt->timeout, time); - bt->last_state = bt->state; + last_printed = bt->state; + } - if (bt->state == BT_STATE_HOSED) - return SI_SM_HOSED; + /* Commands that time out may still (eventually) provide a response. + This stale response will get in the way of a new response so remove + it if possible (hopefully during IDLE). Even if it comes up later + it will be rejected by its (now-forgotten) seq number. */ + + if ((bt->state < BT_STATE_WRITE_BYTES) && (status & BT_B2H_ATN)) { + drain_BMC2HOST(bt); + BT_SI_SM_RETURN(SI_SM_CALL_WITH_DELAY); + } - if (bt->state != BT_STATE_IDLE) { /* do timeout test */ + if ((bt->state != BT_STATE_IDLE) && + (bt->state < BT_STATE_PRINTME)) { /* check timeout */ bt->timeout -= time; - if ((bt->timeout < 0) && (bt->state < BT_STATE_RESET1)) { - error_recovery(bt, "timed out"); - return SI_SM_CALL_WITHOUT_DELAY; - } + if ((bt->timeout < 0) && (bt->state < BT_STATE_RESET1)) + return error_recovery(bt, + status, + IPMI_TIMEOUT_ERR); } switch (bt->state) { - case BT_STATE_IDLE: /* check for asynchronous messages */ + /* Idle state first checks for asynchronous messages from another + channel, then does some opportunistic housekeeping. */ + + case BT_STATE_IDLE: if (status & BT_SMS_ATN) { BT_CONTROL(BT_SMS_ATN); /* clear it */ return SI_SM_ATTN; } - return SI_SM_IDLE; - case BT_STATE_XACTION_START: - if (status & BT_H_BUSY) { + if (status & BT_H_BUSY) /* clear a leftover H_BUSY */ BT_CONTROL(BT_H_BUSY); - break; - } - if (status & BT_B2H_ATN) - break; - bt->state = BT_STATE_WRITE_BYTES; - return SI_SM_CALL_WITHOUT_DELAY; /* for logging */ - case BT_STATE_WRITE_BYTES: + /* Read BT capabilities if it hasn't been done yet */ + if (!bt->BT_CAP_outreqs) + BT_STATE_CHANGE(BT_STATE_CAPABILITIES_BEGIN, + SI_SM_CALL_WITHOUT_DELAY); + bt->timeout = bt->BT_CAP_req2rsp; + BT_SI_SM_RETURN(SI_SM_IDLE); + + case BT_STATE_XACTION_START: if (status & (BT_B_BUSY | BT_H2B_ATN)) - break; + BT_SI_SM_RETURN(SI_SM_CALL_WITH_DELAY); + if (BT_STATUS & BT_H_BUSY) + BT_CONTROL(BT_H_BUSY); /* force clear */ + BT_STATE_CHANGE(BT_STATE_WRITE_BYTES, + SI_SM_CALL_WITHOUT_DELAY); + + case BT_STATE_WRITE_BYTES: + if (status & BT_H_BUSY) + BT_CONTROL(BT_H_BUSY); /* clear */ BT_CONTROL(BT_CLR_WR_PTR); write_all_bytes(bt); - BT_CONTROL(BT_H2B_ATN); /* clears too fast to catch? */ - bt->state = BT_STATE_WRITE_CONSUME; - return SI_SM_CALL_WITHOUT_DELAY; /* it MIGHT sail through */ - - case BT_STATE_WRITE_CONSUME: /* BMCs usually blow right thru here */ - if (status & (BT_H2B_ATN | BT_B_BUSY)) - break; - bt->state = BT_STATE_B2H_WAIT; - /* fall through with status */ - - /* Stay in BT_STATE_B2H_WAIT until a packet matches. However, spinning - hard here, constantly reading status, seems to hold off the - generation of B2H_ATN so ALWAYS return CALL_WITH_DELAY. */ - - case BT_STATE_B2H_WAIT: - if (!(status & BT_B2H_ATN)) - break; - - /* Assume ordered, uncached writes: no need to wait */ - if (!(status & BT_H_BUSY)) - BT_CONTROL(BT_H_BUSY); /* set */ - BT_CONTROL(BT_B2H_ATN); /* clear it, ACK to the BMC */ - BT_CONTROL(BT_CLR_RD_PTR); /* reset the queue */ - i = read_all_bytes(bt); - BT_CONTROL(BT_H_BUSY); /* clear */ - if (!i) /* Try this state again */ - break; - bt->state = BT_STATE_READ_END; - return SI_SM_CALL_WITHOUT_DELAY; /* for logging */ - - case BT_STATE_READ_END: - - /* I could wait on BT_H_BUSY to go clear for a truly clean - exit. However, this is already done in XACTION_START - and the (possible) extra loop/status/possible wait affects - performance. So, as long as it works, just ignore H_BUSY */ - -#ifdef MAKE_THIS_TRUE_IF_NECESSARY + BT_CONTROL(BT_H2B_ATN); /* can clear too fast to catch */ + BT_STATE_CHANGE(BT_STATE_WRITE_CONSUME, + SI_SM_CALL_WITHOUT_DELAY); - if (status & BT_H_BUSY) - break; -#endif - bt->seq++; - bt->state = BT_STATE_IDLE; - return SI_SM_TRANSACTION_COMPLETE; + case BT_STATE_WRITE_CONSUME: + if (status & (BT_B_BUSY | BT_H2B_ATN)) + BT_SI_SM_RETURN(SI_SM_CALL_WITH_DELAY); + BT_STATE_CHANGE(BT_STATE_READ_WAIT, + SI_SM_CALL_WITHOUT_DELAY); + + /* Spinning hard can suppress B2H_ATN and force a timeout */ + + case BT_STATE_READ_WAIT: + if (!(status & BT_B2H_ATN)) + BT_SI_SM_RETURN(SI_SM_CALL_WITH_DELAY); + BT_CONTROL(BT_H_BUSY); /* set */ + + /* Uncached, ordered writes should just proceeed serially but + some BMCs don't clear B2H_ATN with one hit. Fast-path a + workaround without too much penalty to the general case. */ + + BT_CONTROL(BT_B2H_ATN); /* clear it to ACK the BMC */ + BT_STATE_CHANGE(BT_STATE_CLEAR_B2H, + SI_SM_CALL_WITHOUT_DELAY); + + case BT_STATE_CLEAR_B2H: + if (status & BT_B2H_ATN) { /* keep hitting it */ + BT_CONTROL(BT_B2H_ATN); + BT_SI_SM_RETURN(SI_SM_CALL_WITH_DELAY); + } + BT_STATE_CHANGE(BT_STATE_READ_BYTES, + SI_SM_CALL_WITHOUT_DELAY); + + case BT_STATE_READ_BYTES: + if (!(status & BT_H_BUSY)) /* check in case of retry */ + BT_CONTROL(BT_H_BUSY); + BT_CONTROL(BT_CLR_RD_PTR); /* start of BMC2HOST buffer */ + i = read_all_bytes(bt); /* true == packet seq match */ + BT_CONTROL(BT_H_BUSY); /* NOW clear */ + if (!i) /* Not my message */ + BT_STATE_CHANGE(BT_STATE_READ_WAIT, + SI_SM_CALL_WITHOUT_DELAY); + bt->state = bt->complete; + return bt->state == BT_STATE_IDLE ? /* where to next? */ + SI_SM_TRANSACTION_COMPLETE : /* normal */ + SI_SM_CALL_WITHOUT_DELAY; /* Startup magic */ + + case BT_STATE_LONG_BUSY: /* For example: after FW update */ + if (!(status & BT_B_BUSY)) { + reset_flags(bt); /* next state is now IDLE */ + bt_init_data(bt, bt->io); + } + return SI_SM_CALL_WITH_DELAY; /* No repeat printing */ case BT_STATE_RESET1: - reset_flags(bt); - bt->timeout = BT_RESET_DELAY; - bt->state = BT_STATE_RESET2; - break; + reset_flags(bt); + drain_BMC2HOST(bt); + BT_STATE_CHANGE(BT_STATE_RESET2, + SI_SM_CALL_WITH_DELAY); case BT_STATE_RESET2: /* Send a soft reset */ BT_CONTROL(BT_CLR_WR_PTR); @@ -464,29 +579,59 @@ static enum si_sm_result bt_event(struct si_sm_data *bt, long time) HOST2BMC(42); /* Sequence number */ HOST2BMC(3); /* Cmd == Soft reset */ BT_CONTROL(BT_H2B_ATN); - bt->state = BT_STATE_RESET3; - break; + bt->timeout = BT_RESET_DELAY * 1000000; + BT_STATE_CHANGE(BT_STATE_RESET3, + SI_SM_CALL_WITH_DELAY); - case BT_STATE_RESET3: + case BT_STATE_RESET3: /* Hold off everything for a bit */ if (bt->timeout > 0) - return SI_SM_CALL_WITH_DELAY; - bt->state = BT_STATE_RESTART; /* printk in debug modes */ - break; + return SI_SM_CALL_WITH_DELAY; + drain_BMC2HOST(bt); + BT_STATE_CHANGE(BT_STATE_RESTART, + SI_SM_CALL_WITH_DELAY); - case BT_STATE_RESTART: /* don't reset retries! */ - reset_flags(bt); - bt->write_data[2] = ++bt->seq; + case BT_STATE_RESTART: /* don't reset retries or seq! */ bt->read_count = 0; bt->nonzero_status = 0; - bt->timeout = BT_NORMAL_TIMEOUT; - bt->state = BT_STATE_XACTION_START; - break; - - default: /* HOSED is supposed to be caught much earlier */ - error_recovery(bt, "internal logic error"); - break; - } - return SI_SM_CALL_WITH_DELAY; + bt->timeout = bt->BT_CAP_req2rsp; + BT_STATE_CHANGE(BT_STATE_XACTION_START, + SI_SM_CALL_WITH_DELAY); + + /* Get BT Capabilities, using timing of upper level state machine. + Set outreqs to prevent infinite loop on timeout. */ + case BT_STATE_CAPABILITIES_BEGIN: + bt->BT_CAP_outreqs = 1; + { + unsigned char GetBT_CAP[] = { 0x18, 0x36 }; + bt->state = BT_STATE_IDLE; + bt_start_transaction(bt, GetBT_CAP, sizeof(GetBT_CAP)); + } + bt->complete = BT_STATE_CAPABILITIES_END; + BT_STATE_CHANGE(BT_STATE_XACTION_START, + SI_SM_CALL_WITH_DELAY); + + case BT_STATE_CAPABILITIES_END: + i = bt_get_result(bt, BT_CAP, sizeof(BT_CAP)); + bt_init_data(bt, bt->io); + if ((i == 8) && !BT_CAP[2]) { + bt->BT_CAP_outreqs = BT_CAP[3]; + bt->BT_CAP_req2rsp = BT_CAP[6] * 1000000; + bt->BT_CAP_retries = BT_CAP[7]; + } else + printk(KERN_WARNING "IPMI BT: using default values\n"); + if (!bt->BT_CAP_outreqs) + bt->BT_CAP_outreqs = 1; + printk(KERN_WARNING "IPMI BT: req2rsp=%ld secs retries=%d\n", + bt->BT_CAP_req2rsp / 1000000L, bt->BT_CAP_retries); + bt->timeout = bt->BT_CAP_req2rsp; + return SI_SM_CALL_WITHOUT_DELAY; + + default: /* should never occur */ + return error_recovery(bt, + status, + IPMI_ERR_UNSPECIFIED); + } + return SI_SM_CALL_WITH_DELAY; } static int bt_detect(struct si_sm_data *bt) @@ -497,7 +642,7 @@ static int bt_detect(struct si_sm_data *bt) test that first. The calling routine uses negative logic. */ if ((BT_STATUS == 0xFF) && (BT_INTMASK_R == 0xFF)) - return 1; + return 1; reset_flags(bt); return 0; } @@ -513,11 +658,11 @@ static int bt_size(void) struct si_sm_handlers bt_smi_handlers = { - .init_data = bt_init_data, - .start_transaction = bt_start_transaction, - .get_result = bt_get_result, - .event = bt_event, - .detect = bt_detect, - .cleanup = bt_cleanup, - .size = bt_size, + .init_data = bt_init_data, + .start_transaction = bt_start_transaction, + .get_result = bt_get_result, + .event = bt_event, + .detect = bt_detect, + .cleanup = bt_cleanup, + .size = bt_size, }; diff --git a/drivers/char/ipmi/ipmi_kcs_sm.c b/drivers/char/ipmi/ipmi_kcs_sm.c index 2062675f9e99..fb46979567e3 100644 --- a/drivers/char/ipmi/ipmi_kcs_sm.c +++ b/drivers/char/ipmi/ipmi_kcs_sm.c @@ -261,12 +261,14 @@ static int start_kcs_transaction(struct si_sm_data *kcs, unsigned char *data, { unsigned int i; - if ((size < 2) || (size > MAX_KCS_WRITE_SIZE)) { - return -1; - } - if ((kcs->state != KCS_IDLE) && (kcs->state != KCS_HOSED)) { - return -2; - } + if (size < 2) + return IPMI_REQ_LEN_INVALID_ERR; + if (size > MAX_KCS_WRITE_SIZE) + return IPMI_REQ_LEN_EXCEEDED_ERR; + + if ((kcs->state != KCS_IDLE) && (kcs->state != KCS_HOSED)) + return IPMI_NOT_IN_MY_STATE_ERR; + if (kcs_debug & KCS_DEBUG_MSG) { printk(KERN_DEBUG "start_kcs_transaction -"); for (i = 0; i < size; i ++) { diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index c7de2e86a6dc..81a0c89598e7 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -247,14 +247,18 @@ static void deliver_recv_msg(struct smi_info *smi_info, spin_lock(&(smi_info->si_lock)); } -static void return_hosed_msg(struct smi_info *smi_info) +static void return_hosed_msg(struct smi_info *smi_info, int cCode) { struct ipmi_smi_msg *msg = smi_info->curr_msg; + if (cCode < 0 || cCode > IPMI_ERR_UNSPECIFIED) + cCode = IPMI_ERR_UNSPECIFIED; + /* else use it as is */ + /* Make it a reponse */ msg->rsp[0] = msg->data[0] | 4; msg->rsp[1] = msg->data[1]; - msg->rsp[2] = IPMI_ERR_UNSPECIFIED; + msg->rsp[2] = cCode; msg->rsp_size = 3; smi_info->curr_msg = NULL; @@ -305,7 +309,7 @@ static enum si_sm_result start_next_msg(struct smi_info *smi_info) smi_info->curr_msg->data, smi_info->curr_msg->data_size); if (err) { - return_hosed_msg(smi_info); + return_hosed_msg(smi_info, err); } rv = SI_SM_CALL_WITHOUT_DELAY; @@ -647,7 +651,7 @@ static enum si_sm_result smi_event_handler(struct smi_info *smi_info, /* If we were handling a user message, format a response to send to the upper layer to tell it about the error. */ - return_hosed_msg(smi_info); + return_hosed_msg(smi_info, IPMI_ERR_UNSPECIFIED); } si_sm_result = smi_info->handlers->event(smi_info->si_sm, 0); } diff --git a/drivers/char/ipmi/ipmi_smic_sm.c b/drivers/char/ipmi/ipmi_smic_sm.c index 39d7e5ef1a2b..e64ea7d25d24 100644 --- a/drivers/char/ipmi/ipmi_smic_sm.c +++ b/drivers/char/ipmi/ipmi_smic_sm.c @@ -141,12 +141,14 @@ static int start_smic_transaction(struct si_sm_data *smic, { unsigned int i; - if ((size < 2) || (size > MAX_SMIC_WRITE_SIZE)) { - return -1; - } - if ((smic->state != SMIC_IDLE) && (smic->state != SMIC_HOSED)) { - return -2; - } + if (size < 2) + return IPMI_REQ_LEN_INVALID_ERR; + if (size > MAX_SMIC_WRITE_SIZE) + return IPMI_REQ_LEN_EXCEEDED_ERR; + + if ((smic->state != SMIC_IDLE) && (smic->state != SMIC_HOSED)) + return IPMI_NOT_IN_MY_STATE_ERR; + if (smic_debug & SMIC_DEBUG_MSG) { printk(KERN_INFO "start_smic_transaction -"); for (i = 0; i < size; i ++) { diff --git a/include/linux/ipmi_msgdefs.h b/include/linux/ipmi_msgdefs.h index 8d6759cc1a71..b56a158d587a 100644 --- a/include/linux/ipmi_msgdefs.h +++ b/include/linux/ipmi_msgdefs.h @@ -71,14 +71,18 @@ /* The BT interface on high-end HP systems supports up to 255 bytes in * one transfer. Its "virtual" BMC supports some commands that are longer * than 128 bytes. Use the full 256, plus NetFn/LUN, Cmd, cCode, plus - * some overhead. It would be nice to base this on the "BT Capabilities" - * but that's too hard to propagate to the rest of the driver. */ + * some overhead; it's not worth the effort to dynamically size this based + * on the results of the "Get BT Capabilities" command. */ #define IPMI_MAX_MSG_LENGTH 272 /* multiple of 16 */ #define IPMI_CC_NO_ERROR 0x00 #define IPMI_NODE_BUSY_ERR 0xc0 #define IPMI_INVALID_COMMAND_ERR 0xc1 +#define IPMI_TIMEOUT_ERR 0xc3 #define IPMI_ERR_MSG_TRUNCATED 0xc6 +#define IPMI_REQ_LEN_INVALID_ERR 0xc7 +#define IPMI_REQ_LEN_EXCEEDED_ERR 0xc8 +#define IPMI_NOT_IN_MY_STATE_ERR 0xd5 /* IPMI 2.0 */ #define IPMI_LOST_ARBITRATION_ERR 0x81 #define IPMI_BUS_ERR 0x82 #define IPMI_NAK_ON_WRITE_ERR 0x83 -- cgit v1.2.3 From 6cf24f031bc97cb5a7c9df3b6e73c45b628b2b28 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 6 Dec 2006 20:41:39 -0800 Subject: [PATCH] elf.h: forward declare struct file In file included from include/asm/patch.h:14, from arch/ia64/kernel/patch.c:10: include/linux/elf.h:375: warning: "struct file" declared inside parameter list include/linux/elf.h:375: warning: its scope is only this definition or declaration, which is probably not what you want Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/elf.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/elf.h b/include/linux/elf.h index b403516d5c3d..60713e6ea297 100644 --- a/include/linux/elf.h +++ b/include/linux/elf.h @@ -6,6 +6,8 @@ #include #include +struct file; + #ifndef elf_read_implies_exec /* Executables for which elf_read_implies_exec() returns TRUE will have the READ_IMPLIES_EXEC personality flag set automatically. -- cgit v1.2.3 From 68380b581383c028830f79ec2670f4a193854aa6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 7 Dec 2006 09:28:19 -0800 Subject: Add "run_scheduled_work()" workqueue function This allows workqueue users to run just their own pending work, rather than wait for the whole workqueue to finish running. This solves the deadlock with networking libphy that was due to other workqueue entries possibly needing a lock that was held by the routine that wanted to flush its own work. It's not wonderful: if you absolutely need to synchronize with the work function having been executed, any user strictly speaking should have its own completion tracking logic, since when we run things explicitly by hand, the generic workqueue layer can no longer help us synchronize. Also, this is strictly only usable for work that has been scheduled without any delayed timers. You can not mix the new interface with schedule_delayed_work(). But it's better than what we had currently. Acked-by: Maciej W. Rozycki Signed-off-by: Linus Torvalds --- drivers/net/phy/phy.c | 3 +- include/linux/workqueue.h | 1 + kernel/workqueue.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 75 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 4044bb1ada86..e175f3910b18 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -587,8 +587,7 @@ int phy_stop_interrupts(struct phy_device *phydev) * Finish any pending work; we might have been scheduled * to be called from keventd ourselves, though. */ - if (!current_is_keventd()) - flush_scheduled_work(); + run_scheduled_work(&phydev->phy_queue); free_irq(phydev->irq, phydev); diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index f0cb1df7b475..edef8d50b26b 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -162,6 +162,7 @@ extern int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, extern void FASTCALL(flush_workqueue(struct workqueue_struct *wq)); extern int FASTCALL(schedule_work(struct work_struct *work)); +extern int FASTCALL(run_scheduled_work(struct work_struct *work)); extern int FASTCALL(schedule_delayed_work(struct delayed_work *work, unsigned long delay)); extern int schedule_delayed_work_on(int cpu, struct delayed_work *work, unsigned long delay); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index c5257316f4b9..6b186750e9be 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -108,6 +108,79 @@ static inline void *get_wq_data(struct work_struct *work) return (void *) (work->management & WORK_STRUCT_WQ_DATA_MASK); } +static int __run_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) +{ + int ret = 0; + unsigned long flags; + + spin_lock_irqsave(&cwq->lock, flags); + /* + * We need to re-validate the work info after we've gotten + * the cpu_workqueue lock. We can run the work now iff: + * + * - the wq_data still matches the cpu_workqueue_struct + * - AND the work is still marked pending + * - AND the work is still on a list (which will be this + * workqueue_struct list) + * + * All these conditions are important, because we + * need to protect against the work being run right + * now on another CPU (all but the last one might be + * true if it's currently running and has not been + * released yet, for example). + */ + if (get_wq_data(work) == cwq + && work_pending(work) + && !list_empty(&work->entry)) { + work_func_t f = work->func; + list_del_init(&work->entry); + spin_unlock_irqrestore(&cwq->lock, flags); + + if (!test_bit(WORK_STRUCT_NOAUTOREL, &work->management)) + work_release(work); + f(work); + + spin_lock_irqsave(&cwq->lock, flags); + cwq->remove_sequence++; + wake_up(&cwq->work_done); + ret = 1; + } + spin_unlock_irqrestore(&cwq->lock, flags); + return ret; +} + +/** + * run_scheduled_work - run scheduled work synchronously + * @work: work to run + * + * This checks if the work was pending, and runs it + * synchronously if so. It returns a boolean to indicate + * whether it had any scheduled work to run or not. + * + * NOTE! This _only_ works for normal work_structs. You + * CANNOT use this for delayed work, because the wq data + * for delayed work will not point properly to the per- + * CPU workqueue struct, but will change! + */ +int fastcall run_scheduled_work(struct work_struct *work) +{ + for (;;) { + struct cpu_workqueue_struct *cwq; + + if (!work_pending(work)) + return 0; + if (list_empty(&work->entry)) + return 0; + /* NOTE! This depends intimately on __queue_work! */ + cwq = get_wq_data(work); + if (!cwq) + return 0; + if (__run_work(cwq, work)) + return 1; + } +} +EXPORT_SYMBOL(run_scheduled_work); + /* Preempt must be disabled. */ static void __queue_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) -- cgit v1.2.3 From a79561134f38de12dce14ed72138f38e55ef53fc Mon Sep 17 00:00:00 2001 From: Zou Nan hai Date: Thu, 7 Dec 2006 09:51:35 -0800 Subject: [IA64] IA64 Kexec/kdump Changes and updates. 1. Remove fake rendz path and related code according to discuss with Khalid Aziz. 2. fc.i offset fix in relocate_kernel.S. 3. iospic shutdown code eoi and mask race fix from Fujitsu. 4. Warm boot hook in machine_kexec to SN SAL code from Jack Steiner. 5. Send slave to SAL slave loop patch from Jay Lan. 6. Kdump on non-recoverable MCA event patch from Jay Lan 7. Use CTL_UNNUMBERED in kdump_on_init sysctl. Signed-off-by: Zou Nan hai Signed-off-by: Tony Luck --- arch/ia64/Kconfig | 23 +++ arch/ia64/kernel/Makefile | 1 + arch/ia64/kernel/crash.c | 245 +++++++++++++++++++++++++++ arch/ia64/kernel/efi.c | 65 +++++++- arch/ia64/kernel/entry.S | 2 +- arch/ia64/kernel/iosapic.c | 21 +++ arch/ia64/kernel/machine_kexec.c | 133 +++++++++++++++ arch/ia64/kernel/mca.c | 5 + arch/ia64/kernel/relocate_kernel.S | 334 +++++++++++++++++++++++++++++++++++++ arch/ia64/kernel/setup.c | 38 +++++ arch/ia64/kernel/smp.c | 28 +++- arch/ia64/sn/kernel/setup.c | 8 + include/asm-ia64/kexec.h | 47 ++++++ include/asm-ia64/machvec.h | 5 + include/asm-ia64/machvec_sn2.h | 2 + include/asm-ia64/meminit.h | 3 +- include/asm-ia64/sn/sn_sal.h | 9 + include/linux/kexec.h | 5 + kernel/kexec.c | 1 + 19 files changed, 969 insertions(+), 6 deletions(-) create mode 100644 arch/ia64/kernel/crash.c create mode 100644 arch/ia64/kernel/machine_kexec.c create mode 100644 arch/ia64/kernel/relocate_kernel.S create mode 100644 include/asm-ia64/kexec.h (limited to 'include/linux') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 683b12c6f76c..75d839715b2f 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -434,6 +434,29 @@ config IA64_ESI source "drivers/sn/Kconfig" +config KEXEC + bool "kexec system call (EXPERIMENTAL)" + depends on EXPERIMENTAL && !IA64_HP_SIM && (!SMP || HOTPLUG_CPU) + help + kexec is a system call that implements the ability to shutdown your + current kernel, and to start another kernel. It is like a reboot + but it is indepedent of the system firmware. And like a reboot + you can start any kernel with it, not just Linux. + + The name comes from the similiarity to the exec system call. + + It is an ongoing process to be certain the hardware in a machine + is properly shutdown, so do not be surprised if this code does not + initially work for you. It may help to enable device hotplugging + support. As of this writing the exact hardware interface is + strongly in flux, so no good recommendation can be made. + +config CRASH_DUMP + bool "kernel crash dumps (EXPERIMENTAL)" + depends on EXPERIMENTAL && IA64_MCA_RECOVERY && !IA64_HP_SIM && (!SMP || HOTPLUG_CPU) + help + Generate crash dump after being started by kexec. + source "drivers/firmware/Kconfig" source "fs/Kconfig.binfmt" diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index cfa099b04cda..8ae384eb5357 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -28,6 +28,7 @@ obj-$(CONFIG_IA64_CYCLONE) += cyclone.o obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_IA64_MCA_RECOVERY) += mca_recovery.o obj-$(CONFIG_KPROBES) += kprobes.o jprobes.o +obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR) += uncached.o obj-$(CONFIG_AUDIT) += audit.o obj-$(CONFIG_PCI_MSI) += msi_ia64.o diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c new file mode 100644 index 000000000000..0aabedf95dad --- /dev/null +++ b/arch/ia64/kernel/crash.c @@ -0,0 +1,245 @@ +/* + * arch/ia64/kernel/crash.c + * + * Architecture specific (ia64) functions for kexec based crash dumps. + * + * Created by: Khalid Aziz + * Copyright (C) 2005 Hewlett-Packard Development Company, L.P. + * Copyright (C) 2005 Intel Corp Zou Nan hai + * + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +int kdump_status[NR_CPUS]; +atomic_t kdump_cpu_freezed; +atomic_t kdump_in_progress; +int kdump_on_init = 1; +ssize_t +copy_oldmem_page(unsigned long pfn, char *buf, + size_t csize, unsigned long offset, int userbuf) +{ + void *vaddr; + + if (!csize) + return 0; + vaddr = __va(pfn<n_namesz = strlen(name) + 1; + note->n_descsz = data_len; + note->n_type = type; + buf += (sizeof(*note) + 3)/4; + memcpy(buf, name, note->n_namesz); + buf += (note->n_namesz + 3)/4; + memcpy(buf, data, data_len); + buf += (data_len + 3)/4; + return buf; +} + +static void +final_note(void *buf) +{ + memset(buf, 0, sizeof(struct elf_note)); +} + +extern void ia64_dump_cpu_regs(void *); + +static DEFINE_PER_CPU(struct elf_prstatus, elf_prstatus); + +void +crash_save_this_cpu() +{ + void *buf; + unsigned long cfm, sof, sol; + + int cpu = smp_processor_id(); + struct elf_prstatus *prstatus = &per_cpu(elf_prstatus, cpu); + + elf_greg_t *dst = (elf_greg_t *)&(prstatus->pr_reg); + memset(prstatus, 0, sizeof(*prstatus)); + prstatus->pr_pid = current->pid; + + ia64_dump_cpu_regs(dst); + cfm = dst[43]; + sol = (cfm >> 7) & 0x7f; + sof = cfm & 0x7f; + dst[46] = (unsigned long)ia64_rse_skip_regs((unsigned long *)dst[46], + sof - sol); + + buf = (u64 *) per_cpu_ptr(crash_notes, cpu); + if (!buf) + return; + buf = append_elf_note(buf, "CORE", NT_PRSTATUS, prstatus, + sizeof(*prstatus)); + final_note(buf); +} + +static int +kdump_wait_cpu_freeze(void) +{ + int cpu_num = num_online_cpus() - 1; + int timeout = 1000; + while(timeout-- > 0) { + if (atomic_read(&kdump_cpu_freezed) == cpu_num) + return 0; + udelay(1000); + } + return 1; +} + +void +machine_crash_shutdown(struct pt_regs *pt) +{ + /* This function is only called after the system + * has paniced or is otherwise in a critical state. + * The minimum amount of code to allow a kexec'd kernel + * to run successfully needs to happen here. + * + * In practice this means shooting down the other cpus in + * an SMP system. + */ + kexec_disable_iosapic(); +#ifdef CONFIG_SMP + kdump_smp_send_stop(); + if (kdump_wait_cpu_freeze() && kdump_on_init) { + //not all cpu response to IPI, send INIT to freeze them + kdump_smp_send_init(); + } +#endif +} + +static void +machine_kdump_on_init(void) +{ + local_irq_disable(); + kexec_disable_iosapic(); + machine_kexec(ia64_kimage); +} + +void +kdump_cpu_freeze(struct unw_frame_info *info, void *arg) +{ + int cpuid; + local_irq_disable(); + cpuid = smp_processor_id(); + crash_save_this_cpu(); + current->thread.ksp = (__u64)info->sw - 16; + atomic_inc(&kdump_cpu_freezed); + kdump_status[cpuid] = 1; + mb(); + if (cpuid == 0) { + for (;;) + cpu_relax(); + } else + ia64_jump_to_sal(&sal_boot_rendez_state[cpuid]); +} + +static int +kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data) +{ + struct ia64_mca_notify_die *nd; + struct die_args *args = data; + + if (!kdump_on_init) + return NOTIFY_DONE; + + if (val != DIE_INIT_MONARCH_ENTER && + val != DIE_INIT_SLAVE_ENTER && + val != DIE_MCA_RENDZVOUS_LEAVE && + val != DIE_MCA_MONARCH_LEAVE) + return NOTIFY_DONE; + + nd = (struct ia64_mca_notify_die *)args->err; + /* Reason code 1 means machine check rendezous*/ + if ((val == DIE_INIT_MONARCH_ENTER || DIE_INIT_SLAVE_ENTER) && + nd->sos->rv_rc == 1) + return NOTIFY_DONE; + + switch (val) { + case DIE_INIT_MONARCH_ENTER: + machine_kdump_on_init(); + break; + case DIE_INIT_SLAVE_ENTER: + unw_init_running(kdump_cpu_freeze, NULL); + break; + case DIE_MCA_RENDZVOUS_LEAVE: + if (atomic_read(&kdump_in_progress)) + unw_init_running(kdump_cpu_freeze, NULL); + break; + case DIE_MCA_MONARCH_LEAVE: + /* die_register->signr indicate if MCA is recoverable */ + if (!args->signr) + machine_kdump_on_init(); + break; + } + return NOTIFY_DONE; +} + +#ifdef CONFIG_SYSCTL +static ctl_table kdump_on_init_table[] = { + { + .ctl_name = CTL_UNNUMBERED, + .procname = "kdump_on_init", + .data = &kdump_on_init, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0 } +}; + +static ctl_table sys_table[] = { + { + .ctl_name = CTL_KERN, + .procname = "kernel", + .mode = 0555, + .child = kdump_on_init_table, + }, + { .ctl_name = 0 } +}; +#endif + +static int +machine_crash_setup(void) +{ + char *from = strstr(saved_command_line, "elfcorehdr="); + static struct notifier_block kdump_init_notifier_nb = { + .notifier_call = kdump_init_notifier, + }; + int ret; + if (from) + elfcorehdr_addr = memparse(from+11, &from); + saved_max_pfn = (unsigned long)-1; + if((ret = register_die_notifier(&kdump_init_notifier_nb)) != 0) + return ret; +#ifdef CONFIG_SYSCTL + register_sysctl_table(sys_table, 0); +#endif + return 0; +} + +__initcall(machine_crash_setup); + diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index bb8770a177b5..9b96e7dbaf67 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include @@ -41,7 +42,7 @@ extern efi_status_t efi_call_phys (void *, ...); struct efi efi; EXPORT_SYMBOL(efi); static efi_runtime_services_t *runtime; -static unsigned long mem_limit = ~0UL, max_addr = ~0UL; +static unsigned long mem_limit = ~0UL, max_addr = ~0UL, min_addr = 0UL; #define efi_call_virt(f, args...) (*(f))(args) @@ -421,6 +422,8 @@ efi_init (void) mem_limit = memparse(cp + 4, &cp); } else if (memcmp(cp, "max_addr=", 9) == 0) { max_addr = GRANULEROUNDDOWN(memparse(cp + 9, &cp)); + } else if (memcmp(cp, "min_addr=", 9) == 0) { + min_addr = GRANULEROUNDDOWN(memparse(cp + 9, &cp)); } else { while (*cp != ' ' && *cp) ++cp; @@ -428,6 +431,8 @@ efi_init (void) ++cp; } } + if (min_addr != 0UL) + printk(KERN_INFO "Ignoring memory below %luMB\n", min_addr >> 20); if (max_addr != ~0UL) printk(KERN_INFO "Ignoring memory above %luMB\n", max_addr >> 20); @@ -894,7 +899,8 @@ find_memmap_space (void) as = max(contig_low, md->phys_addr); ae = min(contig_high, efi_md_end(md)); - /* keep within max_addr= command line arg */ + /* keep within max_addr= and min_addr= command line arg */ + as = max(as, min_addr); ae = min(ae, max_addr); if (ae <= as) continue; @@ -1004,7 +1010,8 @@ efi_memmap_init(unsigned long *s, unsigned long *e) } else ae = efi_md_end(md); - /* keep within max_addr= command line arg */ + /* keep within max_addr= and min_addr= command line arg */ + as = max(as, min_addr); ae = min(ae, max_addr); if (ae <= as) continue; @@ -1116,6 +1123,58 @@ efi_initialize_iomem_resources(struct resource *code_resource, */ insert_resource(res, code_resource); insert_resource(res, data_resource); +#ifdef CONFIG_KEXEC + insert_resource(res, &efi_memmap_res); + insert_resource(res, &boot_param_res); + if (crashk_res.end > crashk_res.start) + insert_resource(res, &crashk_res); +#endif } } } + +#ifdef CONFIG_KEXEC +/* find a block of memory aligned to 64M exclude reserved regions + rsvd_regions are sorted + */ +unsigned long +kdump_find_rsvd_region (unsigned long size, + struct rsvd_region *r, int n) +{ + int i; + u64 start, end; + u64 alignment = 1UL << _PAGE_SIZE_64M; + void *efi_map_start, *efi_map_end, *p; + efi_memory_desc_t *md; + u64 efi_desc_size; + + efi_map_start = __va(ia64_boot_param->efi_memmap); + efi_map_end = efi_map_start + ia64_boot_param->efi_memmap_size; + efi_desc_size = ia64_boot_param->efi_memdesc_size; + + for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) { + md = p; + if (!efi_wb(md)) + continue; + start = ALIGN(md->phys_addr, alignment); + end = efi_md_end(md); + for (i = 0; i < n; i++) { + if (__pa(r[i].start) >= start && __pa(r[i].end) < end) { + if (__pa(r[i].start) > start + size) + return start; + start = ALIGN(__pa(r[i].end), alignment); + if (i < n-1 && __pa(r[i+1].start) < start + size) + continue; + else + break; + } + } + if (end > start + size) + return start; + } + + printk(KERN_WARNING "Cannot reserve 0x%lx byte of memory for crashdump\n", + size); + return ~0UL; +} +#endif diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 3390b7c5a63f..15234ed3a341 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1575,7 +1575,7 @@ sys_call_table: data8 sys_mq_timedreceive // 1265 data8 sys_mq_notify data8 sys_mq_getsetattr - data8 sys_ni_syscall // reserved for kexec_load + data8 sys_kexec_load data8 sys_ni_syscall // reserved for vserver data8 sys_waitid // 1270 data8 sys_add_key diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 60d64950e3c2..0fc5fb7865cf 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -288,6 +288,27 @@ nop (unsigned int irq) /* do nothing... */ } + +#ifdef CONFIG_KEXEC +void +kexec_disable_iosapic(void) +{ + struct iosapic_intr_info *info; + struct iosapic_rte_info *rte; + u8 vec = 0; + for (info = iosapic_intr_info; info < + iosapic_intr_info + IA64_NUM_VECTORS; ++info, ++vec) { + list_for_each_entry(rte, &info->rtes, + rte_list) { + iosapic_write(rte->addr, + IOSAPIC_RTE_LOW(rte->rte_index), + IOSAPIC_MASK|vec); + iosapic_eoi(rte->addr, vec); + } + } +} +#endif + static void mask_irq (unsigned int irq) { diff --git a/arch/ia64/kernel/machine_kexec.c b/arch/ia64/kernel/machine_kexec.c new file mode 100644 index 000000000000..468233fa2cee --- /dev/null +++ b/arch/ia64/kernel/machine_kexec.c @@ -0,0 +1,133 @@ +/* + * arch/ia64/kernel/machine_kexec.c + * + * Handle transition of Linux booting another kernel + * Copyright (C) 2005 Hewlett-Packard Development Comapny, L.P. + * Copyright (C) 2005 Khalid Aziz + * Copyright (C) 2006 Intel Corp, Zou Nan hai + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +typedef void (*relocate_new_kernel_t)(unsigned long, unsigned long, + struct ia64_boot_param *, unsigned long); + +struct kimage *ia64_kimage; + +struct resource efi_memmap_res = { + .name = "EFI Memory Map", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + +struct resource boot_param_res = { + .name = "Boot parameter", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_MEM +}; + + +/* + * Do what every setup is needed on image and the + * reboot code buffer to allow us to avoid allocations + * later. + */ +int machine_kexec_prepare(struct kimage *image) +{ + void *control_code_buffer; + const unsigned long *func; + + func = (unsigned long *)&relocate_new_kernel; + /* Pre-load control code buffer to minimize work in kexec path */ + control_code_buffer = page_address(image->control_code_page); + memcpy((void *)control_code_buffer, (const void *)func[0], + relocate_new_kernel_size); + flush_icache_range((unsigned long)control_code_buffer, + (unsigned long)control_code_buffer + relocate_new_kernel_size); + ia64_kimage = image; + + return 0; +} + +void machine_kexec_cleanup(struct kimage *image) +{ +} + +void machine_shutdown(void) +{ + int cpu; + + for_each_online_cpu(cpu) { + if (cpu != smp_processor_id()) + cpu_down(cpu); + } + kexec_disable_iosapic(); +} + +/* + * Do not allocate memory (or fail in any way) in machine_kexec(). + * We are past the point of no return, committed to rebooting now. + */ +extern void *efi_get_pal_addr(void); +static void ia64_machine_kexec(struct unw_frame_info *info, void *arg) +{ + struct kimage *image = arg; + relocate_new_kernel_t rnk; + void *pal_addr = efi_get_pal_addr(); + unsigned long code_addr = (unsigned long)page_address(image->control_code_page); + unsigned long vector; + int ii; + + if (image->type == KEXEC_TYPE_CRASH) { + crash_save_this_cpu(); + current->thread.ksp = (__u64)info->sw - 16; + } + + /* Interrupts aren't acceptable while we reboot */ + local_irq_disable(); + + /* Mask CMC and Performance Monitor interrupts */ + ia64_setreg(_IA64_REG_CR_PMV, 1 << 16); + ia64_setreg(_IA64_REG_CR_CMCV, 1 << 16); + + /* Mask ITV and Local Redirect Registers */ + ia64_set_itv(1 << 16); + ia64_set_lrr0(1 << 16); + ia64_set_lrr1(1 << 16); + + /* terminate possible nested in-service interrupts */ + for (ii = 0; ii < 16; ii++) + ia64_eoi(); + + /* unmask TPR and clear any pending interrupts */ + ia64_setreg(_IA64_REG_CR_TPR, 0); + ia64_srlz_d(); + vector = ia64_get_ivr(); + while (vector != IA64_SPURIOUS_INT_VECTOR) { + ia64_eoi(); + vector = ia64_get_ivr(); + } + platform_kernel_launch_event(); + rnk = (relocate_new_kernel_t)&code_addr; + (*rnk)(image->head, image->start, ia64_boot_param, + GRANULEROUNDDOWN((unsigned long) pal_addr)); + BUG(); +} + +void machine_kexec(struct kimage *image) +{ + unw_init_running(ia64_machine_kexec, image); + for(;;); +} diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 6bedd97570ca..87c1c4f42872 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -82,6 +82,7 @@ #include #include #include +#include #include #include @@ -1238,6 +1239,10 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, } else { /* Dump buffered message to console */ ia64_mlogbuf_finish(1); +#ifdef CONFIG_CRASH_DUMP + atomic_set(&kdump_in_progress, 1); + monarch_cpu = -1; +#endif } if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, (long)&nd, 0, recover) == NOTIFY_STOP) diff --git a/arch/ia64/kernel/relocate_kernel.S b/arch/ia64/kernel/relocate_kernel.S new file mode 100644 index 000000000000..ae473e3f2a0d --- /dev/null +++ b/arch/ia64/kernel/relocate_kernel.S @@ -0,0 +1,334 @@ +/* + * arch/ia64/kernel/relocate_kernel.S + * + * Relocate kexec'able kernel and start it + * + * Copyright (C) 2005 Hewlett-Packard Development Company, L.P. + * Copyright (C) 2005 Khalid Aziz + * Copyright (C) 2005 Intel Corp, Zou Nan hai + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ +#include +#include +#include +#include +#include + + /* Must be relocatable PIC code callable as a C function + */ +GLOBAL_ENTRY(relocate_new_kernel) + .prologue + alloc r31=ar.pfs,4,0,0,0 + .body +.reloc_entry: +{ + rsm psr.i| psr.ic + mov r2=ip +} + ;; +{ + flushrs // must be first insn in group + srlz.i +} + ;; + dep r2=0,r2,61,3 //to physical address + ;; + //first switch to physical mode + add r3=1f-.reloc_entry, r2 + movl r16 = IA64_PSR_AC|IA64_PSR_BN|IA64_PSR_IC + mov ar.rsc=0 // put RSE in enforced lazy mode + ;; + add sp=(memory_stack_end - 16 - .reloc_entry),r2 + add r8=(register_stack - .reloc_entry),r2 + ;; + mov r18=ar.rnat + mov ar.bspstore=r8 + ;; + mov cr.ipsr=r16 + mov cr.iip=r3 + mov cr.ifs=r0 + srlz.i + ;; + mov ar.rnat=r18 + rfi + ;; +1: + //physical mode code begin + mov b6=in1 + dep r28=0,in2,61,3 //to physical address + + // purge all TC entries +#define O(member) IA64_CPUINFO_##member##_OFFSET + GET_THIS_PADDR(r2, cpu_info) // load phys addr of cpu_info into r2 + ;; + addl r17=O(PTCE_STRIDE),r2 + addl r2=O(PTCE_BASE),r2 + ;; + ld8 r18=[r2],(O(PTCE_COUNT)-O(PTCE_BASE));; // r18=ptce_base + ld4 r19=[r2],4 // r19=ptce_count[0] + ld4 r21=[r17],4 // r21=ptce_stride[0] + ;; + ld4 r20=[r2] // r20=ptce_count[1] + ld4 r22=[r17] // r22=ptce_stride[1] + mov r24=r0 + ;; + adds r20=-1,r20 + ;; +#undef O +2: + cmp.ltu p6,p7=r24,r19 +(p7) br.cond.dpnt.few 4f + mov ar.lc=r20 +3: + ptc.e r18 + ;; + add r18=r22,r18 + br.cloop.sptk.few 3b + ;; + add r18=r21,r18 + add r24=1,r24 + ;; + br.sptk.few 2b +4: + srlz.i + ;; + //purge TR entry for kernel text and data + movl r16=KERNEL_START + mov r18=KERNEL_TR_PAGE_SHIFT<<2 + ;; + ptr.i r16, r18 + ptr.d r16, r18 + ;; + srlz.i + ;; + + // purge TR entry for percpu data + movl r16=PERCPU_ADDR + mov r18=PERCPU_PAGE_SHIFT<<2 + ;; + ptr.d r16,r18 + ;; + srlz.d + ;; + + // purge TR entry for pal code + mov r16=in3 + mov r18=IA64_GRANULE_SHIFT<<2 + ;; + ptr.i r16,r18 + ;; + srlz.i + ;; + + // purge TR entry for stack + mov r16=IA64_KR(CURRENT_STACK) + ;; + shl r16=r16,IA64_GRANULE_SHIFT + movl r19=PAGE_OFFSET + ;; + add r16=r19,r16 + mov r18=IA64_GRANULE_SHIFT<<2 + ;; + ptr.d r16,r18 + ;; + srlz.i + ;; + + //copy segments + movl r16=PAGE_MASK + mov r30=in0 // in0 is page_list + br.sptk.few .dest_page + ;; +.loop: + ld8 r30=[in0], 8;; +.dest_page: + tbit.z p0, p6=r30, 0;; // 0x1 dest page +(p6) and r17=r30, r16 +(p6) br.cond.sptk.few .loop;; + + tbit.z p0, p6=r30, 1;; // 0x2 indirect page +(p6) and in0=r30, r16 +(p6) br.cond.sptk.few .loop;; + + tbit.z p0, p6=r30, 2;; // 0x4 end flag +(p6) br.cond.sptk.few .end_loop;; + + tbit.z p6, p0=r30, 3;; // 0x8 source page +(p6) br.cond.sptk.few .loop + + and r18=r30, r16 + + // simple copy page, may optimize later + movl r14=PAGE_SIZE/8 - 1;; + mov ar.lc=r14;; +1: + ld8 r14=[r18], 8;; + st8 [r17]=r14;; + fc.i r17 + add r17=8, r17 + br.ctop.sptk.few 1b + br.sptk.few .loop + ;; + +.end_loop: + sync.i // for fc.i + ;; + srlz.i + ;; + srlz.d + ;; + br.call.sptk.many b0=b6;; + +.align 32 +memory_stack: + .fill 8192, 1, 0 +memory_stack_end: +register_stack: + .fill 8192, 1, 0 +register_stack_end: +relocate_new_kernel_end: +END(relocate_new_kernel) + +.global relocate_new_kernel_size +relocate_new_kernel_size: + data8 relocate_new_kernel_end - relocate_new_kernel + +GLOBAL_ENTRY(ia64_dump_cpu_regs) + .prologue + alloc loc0=ar.pfs,1,2,0,0 + .body + mov ar.rsc=0 // put RSE in enforced lazy mode + add loc1=4*8, in0 // save r4 and r5 first + ;; +{ + flushrs // flush dirty regs to backing store + srlz.i +} + st8 [loc1]=r4, 8 + ;; + st8 [loc1]=r5, 8 + ;; + add loc1=32*8, in0 + mov r4=ar.rnat + ;; + st8 [in0]=r0, 8 // r0 + st8 [loc1]=r4, 8 // rnat + mov r5=pr + ;; + st8 [in0]=r1, 8 // r1 + st8 [loc1]=r5, 8 // pr + mov r4=b0 + ;; + st8 [in0]=r2, 8 // r2 + st8 [loc1]=r4, 8 // b0 + mov r5=b1; + ;; + st8 [in0]=r3, 24 // r3 + st8 [loc1]=r5, 8 // b1 + mov r4=b2 + ;; + st8 [in0]=r6, 8 // r6 + st8 [loc1]=r4, 8 // b2 + mov r5=b3 + ;; + st8 [in0]=r7, 8 // r7 + st8 [loc1]=r5, 8 // b3 + mov r4=b4 + ;; + st8 [in0]=r8, 8 // r8 + st8 [loc1]=r4, 8 // b4 + mov r5=b5 + ;; + st8 [in0]=r9, 8 // r9 + st8 [loc1]=r5, 8 // b5 + mov r4=b6 + ;; + st8 [in0]=r10, 8 // r10 + st8 [loc1]=r5, 8 // b6 + mov r5=b7 + ;; + st8 [in0]=r11, 8 // r11 + st8 [loc1]=r5, 8 // b7 + mov r4=b0 + ;; + st8 [in0]=r12, 8 // r12 + st8 [loc1]=r4, 8 // ip + mov r5=loc0 + ;; + st8 [in0]=r13, 8 // r13 + extr.u r5=r5, 0, 38 // ar.pfs.pfm + mov r4=r0 // user mask + ;; + st8 [in0]=r14, 8 // r14 + st8 [loc1]=r5, 8 // cfm + ;; + st8 [in0]=r15, 8 // r15 + st8 [loc1]=r4, 8 // user mask + mov r5=ar.rsc + ;; + st8 [in0]=r16, 8 // r16 + st8 [loc1]=r5, 8 // ar.rsc + mov r4=ar.bsp + ;; + st8 [in0]=r17, 8 // r17 + st8 [loc1]=r4, 8 // ar.bsp + mov r5=ar.bspstore + ;; + st8 [in0]=r18, 8 // r18 + st8 [loc1]=r5, 8 // ar.bspstore + mov r4=ar.rnat + ;; + st8 [in0]=r19, 8 // r19 + st8 [loc1]=r4, 8 // ar.rnat + mov r5=ar.ccv + ;; + st8 [in0]=r20, 8 // r20 + st8 [loc1]=r5, 8 // ar.ccv + mov r4=ar.unat + ;; + st8 [in0]=r21, 8 // r21 + st8 [loc1]=r4, 8 // ar.unat + mov r5 = ar.fpsr + ;; + st8 [in0]=r22, 8 // r22 + st8 [loc1]=r5, 8 // ar.fpsr + mov r4 = ar.unat + ;; + st8 [in0]=r23, 8 // r23 + st8 [loc1]=r4, 8 // unat + mov r5 = ar.fpsr + ;; + st8 [in0]=r24, 8 // r24 + st8 [loc1]=r5, 8 // fpsr + mov r4 = ar.pfs + ;; + st8 [in0]=r25, 8 // r25 + st8 [loc1]=r4, 8 // ar.pfs + mov r5 = ar.lc + ;; + st8 [in0]=r26, 8 // r26 + st8 [loc1]=r5, 8 // ar.lc + mov r4 = ar.ec + ;; + st8 [in0]=r27, 8 // r27 + st8 [loc1]=r4, 8 // ar.ec + mov r5 = ar.csd + ;; + st8 [in0]=r28, 8 // r28 + st8 [loc1]=r5, 8 // ar.csd + mov r4 = ar.ssd + ;; + st8 [in0]=r29, 8 // r29 + st8 [loc1]=r4, 8 // ar.ssd + ;; + st8 [in0]=r30, 8 // r30 + ;; + st8 [in0]=r31, 8 // r31 + mov ar.pfs=loc0 + ;; + br.ret.sptk.many rp +END(ia64_dump_cpu_regs) + + diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index d10404a41756..14e1200376a9 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -43,6 +43,8 @@ #include #include #include +#include +#include #include #include @@ -252,6 +254,41 @@ reserve_memory (void) efi_memmap_init(&rsvd_region[n].start, &rsvd_region[n].end); n++; +#ifdef CONFIG_KEXEC + /* crashkernel=size@offset specifies the size to reserve for a crash + * kernel.(offset is ingored for keep compatibility with other archs) + * By reserving this memory we guarantee that linux never set's it + * up as a DMA target.Useful for holding code to do something + * appropriate after a kernel panic. + */ + { + char *from = strstr(saved_command_line, "crashkernel="); + unsigned long base, size; + if (from) { + size = memparse(from + 12, &from); + if (size) { + sort_regions(rsvd_region, n); + base = kdump_find_rsvd_region(size, + rsvd_region, n); + if (base != ~0UL) { + rsvd_region[n].start = + (unsigned long)__va(base); + rsvd_region[n].end = + (unsigned long)__va(base + size); + n++; + crashk_res.start = base; + crashk_res.end = base + size - 1; + } + } + } + efi_memmap_res.start = ia64_boot_param->efi_memmap; + efi_memmap_res.end = efi_memmap_res.start + + ia64_boot_param->efi_memmap_size; + boot_param_res.start = __pa(ia64_boot_param); + boot_param_res.end = boot_param_res.start + + sizeof(*ia64_boot_param); + } +#endif /* end of memory marker */ rsvd_region[n].start = ~0UL; rsvd_region[n].end = ~0UL; @@ -263,6 +300,7 @@ reserve_memory (void) sort_regions(rsvd_region, num_rsvd_regions); } + /** * find_initrd - get initrd parameters from the boot parameter structure * diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 6ab95ceaf9d4..b1b9aa4364b9 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -66,6 +67,7 @@ static volatile struct call_data_struct *call_data; #define IPI_CALL_FUNC 0 #define IPI_CPU_STOP 1 +#define IPI_KDUMP_CPU_STOP 3 /* This needs to be cacheline aligned because it is written to by *other* CPUs. */ static DEFINE_PER_CPU(u64, ipi_operation) ____cacheline_aligned; @@ -155,7 +157,11 @@ handle_IPI (int irq, void *dev_id) case IPI_CPU_STOP: stop_this_cpu(); break; - +#ifdef CONFIG_CRASH_DUMP + case IPI_KDUMP_CPU_STOP: + unw_init_running(kdump_cpu_freeze, NULL); + break; +#endif default: printk(KERN_CRIT "Unknown IPI on CPU %d: %lu\n", this_cpu, which); break; @@ -213,6 +219,26 @@ send_IPI_self (int op) send_IPI_single(smp_processor_id(), op); } +#ifdef CONFIG_CRASH_DUMP +void +kdump_smp_send_stop() +{ + send_IPI_allbutself(IPI_KDUMP_CPU_STOP); +} + +void +kdump_smp_send_init() +{ + unsigned int cpu, self_cpu; + self_cpu = smp_processor_id(); + for_each_online_cpu(cpu) { + if (cpu != self_cpu) { + if(kdump_status[cpu] == 0) + platform_send_ipi(cpu, 0, IA64_IPI_DM_INIT, 0); + } + } +} +#endif /* * Called with preeemption disabled. */ diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 1d009f93244d..a934ad069425 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -769,5 +769,13 @@ int sn_prom_feature_available(int id) return 0; return test_bit(id, sn_prom_features); } + +void +sn_kernel_launch_event(void) +{ + /* ignore status until we understand possible failure, if any*/ + if (ia64_sn_kernel_launch_event()) + printk(KERN_ERR "KEXEC is not supported in this PROM, Please update the PROM.\n"); +} EXPORT_SYMBOL(sn_prom_feature_available); diff --git a/include/asm-ia64/kexec.h b/include/asm-ia64/kexec.h new file mode 100644 index 000000000000..01c36b004747 --- /dev/null +++ b/include/asm-ia64/kexec.h @@ -0,0 +1,47 @@ +#ifndef _ASM_IA64_KEXEC_H +#define _ASM_IA64_KEXEC_H + + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL) +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL) +/* Maximum address we can use for the control code buffer */ +#define KEXEC_CONTROL_MEMORY_LIMIT TASK_SIZE + +#define KEXEC_CONTROL_CODE_SIZE (8192 + 8192 + 4096) + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_IA_64 + +#define MAX_NOTE_BYTES 1024 + +#define kexec_flush_icache_page(page) do { \ + unsigned long page_addr = (unsigned long)page_address(page); \ + flush_icache_range(page_addr, page_addr + PAGE_SIZE); \ + } while(0) + +extern struct kimage *ia64_kimage; +DECLARE_PER_CPU(u64, ia64_mca_pal_base); +const extern unsigned int relocate_new_kernel_size; +extern void relocate_new_kernel(unsigned long, unsigned long, + struct ia64_boot_param *, unsigned long); +static inline void +crash_setup_regs(struct pt_regs *newregs, struct pt_regs *oldregs) +{ +} +extern struct resource efi_memmap_res; +extern struct resource boot_param_res; +extern void kdump_smp_send_stop(void); +extern void kdump_smp_send_init(void); +extern void kexec_disable_iosapic(void); +extern void crash_save_this_cpu(void); +struct rsvd_region; +extern unsigned long kdump_find_rsvd_region(unsigned long size, + struct rsvd_region *rsvd_regions, int n); +extern void kdump_cpu_freeze(struct unw_frame_info *info, void *arg); +extern int kdump_status[]; +extern atomic_t kdump_cpu_freezed; +extern atomic_t kdump_in_progress; + +#endif /* _ASM_IA64_KEXEC_H */ diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h index 8f784f8e45b0..a3891eb3f217 100644 --- a/include/asm-ia64/machvec.h +++ b/include/asm-ia64/machvec.h @@ -37,6 +37,7 @@ typedef int ia64_mv_pci_legacy_write_t (struct pci_bus *, u16 port, u32 val, u8 size); typedef void ia64_mv_migrate_t(struct task_struct * task); typedef void ia64_mv_pci_fixup_bus_t (struct pci_bus *); +typedef void ia64_mv_kernel_launch_event_t(void); /* DMA-mapping interface: */ typedef void ia64_mv_dma_init (void); @@ -218,6 +219,7 @@ struct ia64_machine_vector { ia64_mv_setup_msi_irq_t *setup_msi_irq; ia64_mv_teardown_msi_irq_t *teardown_msi_irq; ia64_mv_pci_fixup_bus_t *pci_fixup_bus; + ia64_mv_kernel_launch_event_t *kernel_launch_event; } __attribute__((__aligned__(16))); /* align attrib? see above comment */ #define MACHVEC_INIT(name) \ @@ -318,6 +320,9 @@ extern ia64_mv_dma_supported swiotlb_dma_supported; #ifndef platform_tlb_migrate_finish # define platform_tlb_migrate_finish machvec_noop_mm #endif +#ifndef platform_kernel_launch_event +# define platform_kernel_launch_event machvec_noop +#endif #ifndef platform_dma_init # define platform_dma_init swiotlb_init #endif diff --git a/include/asm-ia64/machvec_sn2.h b/include/asm-ia64/machvec_sn2.h index 83325f6db03e..eaa2fce0fecd 100644 --- a/include/asm-ia64/machvec_sn2.h +++ b/include/asm-ia64/machvec_sn2.h @@ -67,6 +67,7 @@ extern ia64_mv_dma_sync_sg_for_device sn_dma_sync_sg_for_device; extern ia64_mv_dma_mapping_error sn_dma_mapping_error; extern ia64_mv_dma_supported sn_dma_supported; extern ia64_mv_migrate_t sn_migrate; +extern ia64_mv_kernel_launch_event_t sn_kernel_launch_event; extern ia64_mv_setup_msi_irq_t sn_setup_msi_irq; extern ia64_mv_teardown_msi_irq_t sn_teardown_msi_irq; extern ia64_mv_pci_fixup_bus_t sn_pci_fixup_bus; @@ -121,6 +122,7 @@ extern ia64_mv_pci_fixup_bus_t sn_pci_fixup_bus; #define platform_dma_mapping_error sn_dma_mapping_error #define platform_dma_supported sn_dma_supported #define platform_migrate sn_migrate +#define platform_kernel_launch_event sn_kernel_launch_event #ifdef CONFIG_PCI_MSI #define platform_setup_msi_irq sn_setup_msi_irq #define platform_teardown_msi_irq sn_teardown_msi_irq diff --git a/include/asm-ia64/meminit.h b/include/asm-ia64/meminit.h index c3b1f862e6e7..c8df75901083 100644 --- a/include/asm-ia64/meminit.h +++ b/include/asm-ia64/meminit.h @@ -15,11 +15,12 @@ * - initrd (optional) * - command line string * - kernel code & data + * - crash dumping code reserved region * - Kernel memory map built from EFI memory map * * More could be added if necessary */ -#define IA64_MAX_RSVD_REGIONS 6 +#define IA64_MAX_RSVD_REGIONS 7 struct rsvd_region { unsigned long start; /* virtual address of beginning of element */ diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index be5d83ad7cb1..2c4004eb5a68 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -88,6 +88,8 @@ #define SN_SAL_INJECT_ERROR 0x02000067 #define SN_SAL_SET_CPU_NUMBER 0x02000068 +#define SN_SAL_KERNEL_LAUNCH_EVENT 0x02000069 + /* * Service-specific constants */ @@ -1155,4 +1157,11 @@ ia64_sn_set_cpu_number(int cpu) SAL_CALL_NOLOCK(rv, SN_SAL_SET_CPU_NUMBER, cpu, 0, 0, 0, 0, 0, 0); return rv.status; } +static inline int +ia64_sn_kernel_launch_event(void) +{ + struct ia64_sal_retval rv; + SAL_CALL_NOLOCK(rv, SN_SAL_KERNEL_LAUNCH_EVENT, 0, 0, 0, 0, 0, 0, 0); + return rv.status; +} #endif /* _ASM_IA64_SN_SN_SAL_H */ diff --git a/include/linux/kexec.h b/include/linux/kexec.h index a4ede62b339d..e14cd388c65b 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -108,6 +108,10 @@ int kexec_should_crash(struct task_struct *); extern struct kimage *kexec_image; extern struct kimage *kexec_crash_image; +#ifndef kexec_flush_icache_page +#define kexec_flush_icache_page(page) +#endif + #define KEXEC_ON_CRASH 0x00000001 #define KEXEC_ARCH_MASK 0xffff0000 @@ -133,6 +137,7 @@ extern struct resource crashk_res; typedef u32 note_buf_t[MAX_NOTE_BYTES/4]; extern note_buf_t *crash_notes; + #else /* !CONFIG_KEXEC */ struct pt_regs; struct task_struct; diff --git a/kernel/kexec.c b/kernel/kexec.c index fcdd5d2bc3f4..05aada293592 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -851,6 +851,7 @@ static int kimage_load_crash_segment(struct kimage *image, memset(ptr + uchunk, 0, mchunk - uchunk); } result = copy_from_user(ptr, buf, uchunk); + kexec_flush_icache_page(page); kunmap(page); if (result) { result = (result < 0) ? result : -EIO; -- cgit v1.2.3 From 24ec839c431eb79bb8f6abc00c4e1eb3b8c4d517 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 8 Dec 2006 02:36:04 -0800 Subject: [PATCH] tty: ->signal->tty locking Fix the locking of signal->tty. Use ->sighand->siglock to protect ->signal->tty; this lock is already used by most other members of ->signal/->sighand. And unless we are 'current' or the tasklist_lock is held we need ->siglock to access ->signal anyway. (NOTE: sys_unshare() is broken wrt ->sighand locking rules) Note that tty_mutex is held over tty destruction, so while holding tty_mutex any tty pointer remains valid. Otherwise the lifetime of ttys are governed by their open file handles. This leaves some holes for tty access from signal->tty (or any other non file related tty access). It solves the tty SLAB scribbles we were seeing. (NOTE: the change from group_send_sig_info to __group_send_sig_info needs to be examined by someone familiar with the security framework, I think it is safe given the SEND_SIG_PRIV from other __group_send_sig_info invocations) [schwidefsky@de.ibm.com: 3270 fix] [akpm@osdl.org: various post-viro fixes] Signed-off-by: Peter Zijlstra Acked-by: Alan Cox Cc: Oleg Nesterov Cc: Prarit Bhargava Cc: Chris Wright Cc: Roland McGrath Cc: Stephen Smalley Cc: James Morris Cc: "David S. Miller" Cc: Jeff Dike Cc: Martin Schwidefsky Cc: Jan Kara Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/sparc64/solaris/misc.c | 4 +- arch/um/kernel/exec.c | 7 +- drivers/char/tty_io.c | 237 ++++++++++++++++++++++++++------------------ drivers/s390/char/fs3270.c | 12 ++- fs/dquot.c | 14 +-- fs/open.c | 1 + include/linux/tty.h | 11 +- kernel/acct.c | 9 +- kernel/auditsc.c | 2 + kernel/exit.c | 4 +- kernel/sys.c | 6 +- security/selinux/hooks.c | 11 +- 12 files changed, 186 insertions(+), 132 deletions(-) (limited to 'include/linux') diff --git a/arch/sparc64/solaris/misc.c b/arch/sparc64/solaris/misc.c index 9ed997982f8d..e84241d5f7f4 100644 --- a/arch/sparc64/solaris/misc.c +++ b/arch/sparc64/solaris/misc.c @@ -423,9 +423,7 @@ asmlinkage int solaris_procids(int cmd, s32 pid, s32 pgid) Solaris setpgrp and setsid? */ ret = sys_setpgid(0, 0); if (ret) return ret; - mutex_lock(&tty_mutex); - current->signal->tty = NULL; - mutex_unlock(&tty_mutex); + proc_clear_tty(current); return process_group(current); } case 2: /* getsid */ diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c index 0561c43b4685..8d56ec6cca79 100644 --- a/arch/um/kernel/exec.c +++ b/arch/um/kernel/exec.c @@ -39,12 +39,13 @@ static long execve1(char *file, char __user * __user *argv, char __user *__user *env) { long error; + struct tty_struct *tty; #ifdef CONFIG_TTY_LOG mutex_lock(&tty_mutex); - task_lock(current); /* FIXME: is this needed ? */ - log_exec(argv, current->signal->tty); - task_unlock(current); + tty = get_current_tty(); + if (tty) + log_exec(argv, tty); mutex_unlock(&tty_mutex); #endif error = do_execve(file, argv, env, ¤t->thread.regs); diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index b3cfc8bc613c..0c856a6f3677 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -126,7 +126,7 @@ EXPORT_SYMBOL(tty_std_termios); LIST_HEAD(tty_drivers); /* linked list of tty drivers */ -/* Semaphore to protect creating and releasing a tty. This is shared with +/* Mutex to protect creating and releasing a tty. This is shared with vt.c for deeply disgusting hack reasons */ DEFINE_MUTEX(tty_mutex); EXPORT_SYMBOL(tty_mutex); @@ -250,7 +250,7 @@ static int check_tty_count(struct tty_struct *tty, const char *routine) "!= #fd's(%d) in %s\n", tty->name, tty->count, count, routine); return count; - } + } #endif return 0; } @@ -259,18 +259,6 @@ static int check_tty_count(struct tty_struct *tty, const char *routine) * Tty buffer allocation management */ - -/** - * tty_buffer_free_all - free buffers used by a tty - * @tty: tty to free from - * - * Remove all the buffers pending on a tty whether queued with data - * or in the free ring. Must be called when the tty is no longer in use - * - * Locking: none - */ - - /** * tty_buffer_free_all - free buffers used by a tty * @tty: tty to free from @@ -614,7 +602,7 @@ EXPORT_SYMBOL_GPL(tty_prepare_flip_string_flags); * they are not on hot paths so a little discipline won't do * any harm. * - * Locking: takes termios_sem + * Locking: takes termios_mutex */ static void tty_set_termios_ldisc(struct tty_struct *tty, int num) @@ -915,7 +903,7 @@ static void tty_ldisc_enable(struct tty_struct *tty) * context. * * Locking: takes tty_ldisc_lock. - * called functions take termios_sem + * called functions take termios_mutex */ static int tty_set_ldisc(struct tty_struct *tty, int ldisc) @@ -1267,12 +1255,12 @@ EXPORT_SYMBOL_GPL(tty_ldisc_flush); * * Locking: * BKL - * redirect lock for undoing redirection - * file list lock for manipulating list of ttys - * tty_ldisc_lock from called functions - * termios_sem resetting termios data - * tasklist_lock to walk task list for hangup event - * + * redirect lock for undoing redirection + * file list lock for manipulating list of ttys + * tty_ldisc_lock from called functions + * termios_mutex resetting termios data + * tasklist_lock to walk task list for hangup event + * ->siglock to protect ->signal/->sighand */ static void do_tty_hangup(struct work_struct *work) { @@ -1354,14 +1342,18 @@ static void do_tty_hangup(struct work_struct *work) read_lock(&tasklist_lock); if (tty->session > 0) { do_each_task_pid(tty->session, PIDTYPE_SID, p) { + spin_lock_irq(&p->sighand->siglock); if (p->signal->tty == tty) p->signal->tty = NULL; - if (!p->signal->leader) + if (!p->signal->leader) { + spin_unlock_irq(&p->sighand->siglock); continue; - group_send_sig_info(SIGHUP, SEND_SIG_PRIV, p); - group_send_sig_info(SIGCONT, SEND_SIG_PRIV, p); + } + __group_send_sig_info(SIGHUP, SEND_SIG_PRIV, p); + __group_send_sig_info(SIGCONT, SEND_SIG_PRIV, p); if (tty->pgrp > 0) p->signal->tty_old_pgrp = tty->pgrp; + spin_unlock_irq(&p->sighand->siglock); } while_each_task_pid(tty->session, PIDTYPE_SID, p); } read_unlock(&tasklist_lock); @@ -1453,6 +1445,14 @@ int tty_hung_up_p(struct file * filp) EXPORT_SYMBOL(tty_hung_up_p); +static void session_clear_tty(pid_t session) +{ + struct task_struct *p; + do_each_task_pid(session, PIDTYPE_SID, p) { + proc_clear_tty(p); + } while_each_task_pid(session, PIDTYPE_SID, p); +} + /** * disassociate_ctty - disconnect controlling tty * @on_exit: true if exiting so need to "hang up" the session @@ -1469,31 +1469,35 @@ EXPORT_SYMBOL(tty_hung_up_p); * The argument on_exit is set to 1 if called when a process is * exiting; it is 0 if called by the ioctl TIOCNOTTY. * - * Locking: tty_mutex is taken to protect current->signal->tty + * Locking: * BKL is taken for hysterical raisins - * Tasklist lock is taken (under tty_mutex) to walk process - * lists for the session. + * tty_mutex is taken to protect tty + * ->siglock is taken to protect ->signal/->sighand + * tasklist_lock is taken to walk process list for sessions + * ->siglock is taken to protect ->signal/->sighand */ void disassociate_ctty(int on_exit) { struct tty_struct *tty; - struct task_struct *p; int tty_pgrp = -1; + int session; lock_kernel(); mutex_lock(&tty_mutex); - tty = current->signal->tty; + tty = get_current_tty(); if (tty) { tty_pgrp = tty->pgrp; mutex_unlock(&tty_mutex); + /* XXX: here we race, there is nothing protecting tty */ if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY) tty_vhangup(tty); } else { - if (current->signal->tty_old_pgrp) { - kill_pg(current->signal->tty_old_pgrp, SIGHUP, on_exit); - kill_pg(current->signal->tty_old_pgrp, SIGCONT, on_exit); + pid_t old_pgrp = current->signal->tty_old_pgrp; + if (old_pgrp) { + kill_pg(old_pgrp, SIGHUP, on_exit); + kill_pg(old_pgrp, SIGCONT, on_exit); } mutex_unlock(&tty_mutex); unlock_kernel(); @@ -1505,19 +1509,29 @@ void disassociate_ctty(int on_exit) kill_pg(tty_pgrp, SIGCONT, on_exit); } - /* Must lock changes to tty_old_pgrp */ - mutex_lock(&tty_mutex); + spin_lock_irq(¤t->sighand->siglock); current->signal->tty_old_pgrp = 0; - tty->session = 0; - tty->pgrp = -1; + session = current->signal->session; + spin_unlock_irq(¤t->sighand->siglock); + + mutex_lock(&tty_mutex); + /* It is possible that do_tty_hangup has free'd this tty */ + tty = get_current_tty(); + if (tty) { + tty->session = 0; + tty->pgrp = 0; + } else { +#ifdef TTY_DEBUG_HANGUP + printk(KERN_DEBUG "error attempted to write to tty [0x%p]" + " = NULL", tty); +#endif + } + mutex_unlock(&tty_mutex); /* Now clear signal->tty under the lock */ read_lock(&tasklist_lock); - do_each_task_pid(current->signal->session, PIDTYPE_SID, p) { - p->signal->tty = NULL; - } while_each_task_pid(current->signal->session, PIDTYPE_SID, p); + session_clear_tty(session); read_unlock(&tasklist_lock); - mutex_unlock(&tty_mutex); unlock_kernel(); } @@ -2337,16 +2351,10 @@ static void release_dev(struct file * filp) * tty. */ if (tty_closing || o_tty_closing) { - struct task_struct *p; - read_lock(&tasklist_lock); - do_each_task_pid(tty->session, PIDTYPE_SID, p) { - p->signal->tty = NULL; - } while_each_task_pid(tty->session, PIDTYPE_SID, p); + session_clear_tty(tty->session); if (o_tty) - do_each_task_pid(o_tty->session, PIDTYPE_SID, p) { - p->signal->tty = NULL; - } while_each_task_pid(o_tty->session, PIDTYPE_SID, p); + session_clear_tty(o_tty->session); read_unlock(&tasklist_lock); } @@ -2443,9 +2451,9 @@ static void release_dev(struct file * filp) * The termios state of a pty is reset on first open so that * settings don't persist across reuse. * - * Locking: tty_mutex protects current->signal->tty, get_tty_driver and - * init_dev work. tty->count should protect the rest. - * task_lock is held to update task details for sessions + * Locking: tty_mutex protects tty, get_tty_driver and init_dev work. + * tty->count should protect the rest. + * ->siglock protects ->signal/->sighand */ static int tty_open(struct inode * inode, struct file * filp) @@ -2467,12 +2475,13 @@ retry_open: mutex_lock(&tty_mutex); if (device == MKDEV(TTYAUX_MAJOR,0)) { - if (!current->signal->tty) { + tty = get_current_tty(); + if (!tty) { mutex_unlock(&tty_mutex); return -ENXIO; } - driver = current->signal->tty->driver; - index = current->signal->tty->index; + driver = tty->driver; + index = tty->index; filp->f_flags |= O_NONBLOCK; /* Don't let /dev/tty block */ /* noctty = 1; */ goto got_driver; @@ -2547,17 +2556,16 @@ got_driver: filp->f_op = &tty_fops; goto retry_open; } + + mutex_lock(&tty_mutex); + spin_lock_irq(¤t->sighand->siglock); if (!noctty && current->signal->leader && !current->signal->tty && - tty->session == 0) { - task_lock(current); - current->signal->tty = tty; - task_unlock(current); - current->signal->tty_old_pgrp = 0; - tty->session = current->signal->session; - tty->pgrp = process_group(current); - } + tty->session == 0) + __proc_set_tty(current, tty); + spin_unlock_irq(¤t->sighand->siglock); + mutex_unlock(&tty_mutex); return 0; } @@ -2747,7 +2755,7 @@ static int tiocsti(struct tty_struct *tty, char __user *p) * * Copies the kernel idea of the window size into the user buffer. * - * Locking: tty->termios_sem is taken to ensure the winsize data + * Locking: tty->termios_mutex is taken to ensure the winsize data * is consistent. */ @@ -2774,8 +2782,8 @@ static int tiocgwinsz(struct tty_struct *tty, struct winsize __user * arg) * Locking: * Called function use the console_sem is used to ensure we do * not try and resize the console twice at once. - * The tty->termios_sem is used to ensure we don't double - * resize and get confused. Lock order - tty->termios.sem before + * The tty->termios_mutex is used to ensure we don't double + * resize and get confused. Lock order - tty->termios_mutex before * console sem */ @@ -2880,25 +2888,28 @@ static int fionbio(struct file *file, int __user *p) * leader to set this tty as the controlling tty for the session. * * Locking: - * Takes tasklist lock internally to walk sessions - * Takes task_lock() when updating signal->tty * Takes tty_mutex() to protect tty instance - * + * Takes tasklist_lock internally to walk sessions + * Takes ->siglock() when updating signal->tty */ static int tiocsctty(struct tty_struct *tty, int arg) { - struct task_struct *p; - + int ret = 0; if (current->signal->leader && (current->signal->session == tty->session)) - return 0; + return ret; + + mutex_lock(&tty_mutex); /* * The process must be a session leader and * not have a controlling tty already. */ - if (!current->signal->leader || current->signal->tty) - return -EPERM; + if (!current->signal->leader || current->signal->tty) { + ret = -EPERM; + goto unlock; + } + if (tty->session > 0) { /* * This tty is already the controlling @@ -2908,24 +2919,18 @@ static int tiocsctty(struct tty_struct *tty, int arg) /* * Steal it away */ - read_lock(&tasklist_lock); - do_each_task_pid(tty->session, PIDTYPE_SID, p) { - p->signal->tty = NULL; - } while_each_task_pid(tty->session, PIDTYPE_SID, p); + session_clear_tty(tty->session); read_unlock(&tasklist_lock); - } else - return -EPERM; + } else { + ret = -EPERM; + goto unlock; + } } - mutex_lock(&tty_mutex); - task_lock(current); - current->signal->tty = tty; - task_unlock(current); + proc_set_tty(current, tty); +unlock: mutex_unlock(&tty_mutex); - current->signal->tty_old_pgrp = 0; - tty->session = current->signal->session; - tty->pgrp = process_group(current); - return 0; + return ret; } /** @@ -2937,7 +2942,7 @@ static int tiocsctty(struct tty_struct *tty, int arg) * Obtain the process group of the tty. If there is no process group * return an error. * - * Locking: none. Reference to ->signal->tty is safe. + * Locking: none. Reference to current->signal->tty is safe. */ static int tiocgpgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p) @@ -2995,7 +3000,7 @@ static int tiocspgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t * Obtain the session id of the tty. If there is no session * return an error. * - * Locking: none. Reference to ->signal->tty is safe. + * Locking: none. Reference to current->signal->tty is safe. */ static int tiocgsid(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p) @@ -3214,14 +3219,11 @@ int tty_ioctl(struct inode * inode, struct file * file, clear_bit(TTY_EXCLUSIVE, &tty->flags); return 0; case TIOCNOTTY: - /* FIXME: taks lock or tty_mutex ? */ if (current->signal->tty != tty) return -ENOTTY; if (current->signal->leader) disassociate_ctty(0); - task_lock(current); - current->signal->tty = NULL; - task_unlock(current); + proc_clear_tty(current); return 0; case TIOCSCTTY: return tiocsctty(tty, arg); @@ -3321,7 +3323,7 @@ static void __do_SAK(struct work_struct *work) if (!tty) return; - session = tty->session; + session = tty->session; /* We don't want an ldisc switch during this */ disc = tty_ldisc_ref(tty); @@ -3834,9 +3836,52 @@ int tty_unregister_driver(struct tty_driver *driver) cdev_del(&driver->cdev); return 0; } - EXPORT_SYMBOL(tty_unregister_driver); +dev_t tty_devnum(struct tty_struct *tty) +{ + return MKDEV(tty->driver->major, tty->driver->minor_start) + tty->index; +} +EXPORT_SYMBOL(tty_devnum); + +void proc_clear_tty(struct task_struct *p) +{ + spin_lock_irq(&p->sighand->siglock); + p->signal->tty = NULL; + spin_unlock_irq(&p->sighand->siglock); +} +EXPORT_SYMBOL(proc_clear_tty); + +void __proc_set_tty(struct task_struct *tsk, struct tty_struct *tty) +{ + if (tty) { + tty->session = tsk->signal->session; + tty->pgrp = process_group(tsk); + } + tsk->signal->tty = tty; + tsk->signal->tty_old_pgrp = 0; +} + +void proc_set_tty(struct task_struct *tsk, struct tty_struct *tty) +{ + spin_lock_irq(&tsk->sighand->siglock); + __proc_set_tty(tsk, tty); + spin_unlock_irq(&tsk->sighand->siglock); +} + +struct tty_struct *get_current_tty(void) +{ + struct tty_struct *tty; + WARN_ON_ONCE(!mutex_is_locked(&tty_mutex)); + tty = current->signal->tty; + /* + * session->tty can be changed/cleared from under us, make sure we + * issue the load. The obtained pointer, when not NULL, is valid as + * long as we hold tty_mutex. + */ + barrier(); + return tty; +} /* * Initialize the console device. This is called *early*, so diff --git a/drivers/s390/char/fs3270.c b/drivers/s390/char/fs3270.c index 78f8bda81dae..ef205ddd31a4 100644 --- a/drivers/s390/char/fs3270.c +++ b/drivers/s390/char/fs3270.c @@ -424,11 +424,15 @@ fs3270_open(struct inode *inode, struct file *filp) minor = iminor(filp->f_dentry->d_inode); /* Check for minor 0 multiplexer. */ if (minor == 0) { - if (!current->signal->tty) + struct tty_struct *tty; + mutex_lock(&tty_mutex); + tty = get_current_tty(); + if (!tty || tty->driver->major != IBM_TTY3270_MAJOR) { + mutex_unlock(&tty_mutex); return -ENODEV; - if (current->signal->tty->driver->major != IBM_TTY3270_MAJOR) - return -ENODEV; - minor = current->signal->tty->index + RAW3270_FIRSTMINOR; + } + minor = tty->index + RAW3270_FIRSTMINOR; + mutex_unlock(&tty_mutex); } /* Check if some other program is already using fullscreen mode. */ fp = (struct fs3270 *) raw3270_find_view(&fs3270_fn, minor); diff --git a/fs/dquot.c b/fs/dquot.c index f9cd5e23ebdf..89066b19124d 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -828,6 +828,7 @@ static inline int need_print_warning(struct dquot *dquot) static void print_warning(struct dquot *dquot, const char warntype) { char *msg = NULL; + struct tty_struct *tty; int flag = (warntype == BHARDWARN || warntype == BSOFTLONGWARN) ? DQ_BLKS_B : ((warntype == IHARDWARN || warntype == ISOFTLONGWARN) ? DQ_INODES_B : 0); @@ -835,14 +836,15 @@ static void print_warning(struct dquot *dquot, const char warntype) return; mutex_lock(&tty_mutex); - if (!current->signal->tty) + tty = get_current_tty(); + if (!tty) goto out_lock; - tty_write_message(current->signal->tty, dquot->dq_sb->s_id); + tty_write_message(tty, dquot->dq_sb->s_id); if (warntype == ISOFTWARN || warntype == BSOFTWARN) - tty_write_message(current->signal->tty, ": warning, "); + tty_write_message(tty, ": warning, "); else - tty_write_message(current->signal->tty, ": write failed, "); - tty_write_message(current->signal->tty, quotatypes[dquot->dq_type]); + tty_write_message(tty, ": write failed, "); + tty_write_message(tty, quotatypes[dquot->dq_type]); switch (warntype) { case IHARDWARN: msg = " file limit reached.\r\n"; @@ -863,7 +865,7 @@ static void print_warning(struct dquot *dquot, const char warntype) msg = " block quota exceeded.\r\n"; break; } - tty_write_message(current->signal->tty, msg); + tty_write_message(tty, msg); out_lock: mutex_unlock(&tty_mutex); } diff --git a/fs/open.c b/fs/open.c index 89e0c237a636..3b56192816ca 100644 --- a/fs/open.c +++ b/fs/open.c @@ -1087,6 +1087,7 @@ EXPORT_SYMBOL(sys_close); asmlinkage long sys_vhangup(void) { if (capable(CAP_SYS_TTY_CONFIG)) { + /* XXX: this needs locking */ tty_vhangup(current->signal->tty); return 0; } diff --git a/include/linux/tty.h b/include/linux/tty.h index f717f0898238..1d29999a3439 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -309,6 +309,12 @@ extern void tty_ldisc_flush(struct tty_struct *tty); extern int tty_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg); +extern dev_t tty_devnum(struct tty_struct *tty); +extern void proc_clear_tty(struct task_struct *p); +extern void __proc_set_tty(struct task_struct *tsk, struct tty_struct *tty); +extern void proc_set_tty(struct task_struct *tsk, struct tty_struct *tty); +extern struct tty_struct *get_current_tty(void); + extern struct mutex tty_mutex; /* n_tty.c */ @@ -335,10 +341,5 @@ extern void console_print(const char *); extern int vt_ioctl(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg); -static inline dev_t tty_devnum(struct tty_struct *tty) -{ - return MKDEV(tty->driver->major, tty->driver->minor_start) + tty->index; -} - #endif /* __KERNEL__ */ #endif diff --git a/kernel/acct.c b/kernel/acct.c index dc12db8600e7..ca5619039367 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -428,6 +428,7 @@ static void do_acct_process(struct file *file) u64 elapsed; u64 run_time; struct timespec uptime; + struct tty_struct *tty; /* * First check to see if there is enough free_space to continue @@ -485,12 +486,8 @@ static void do_acct_process(struct file *file) #endif mutex_lock(&tty_mutex); - /* FIXME: Whoever is responsible for current->signal locking needs - to use the same locking all over the kernel and document it */ - read_lock(&tasklist_lock); - ac.ac_tty = current->signal->tty ? - old_encode_dev(tty_devnum(current->signal->tty)) : 0; - read_unlock(&tasklist_lock); + tty = get_current_tty(); + ac.ac_tty = tty ? old_encode_dev(tty_devnum(tty)) : 0; mutex_unlock(&tty_mutex); spin_lock_irq(¤t->sighand->siglock); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 40722e26de98..b6cb802fbcd1 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -826,10 +826,12 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts context->return_code); mutex_lock(&tty_mutex); + read_lock(&tasklist_lock); if (tsk->signal && tsk->signal->tty && tsk->signal->tty->name) tty = tsk->signal->tty->name; else tty = "(none)"; + read_unlock(&tasklist_lock); audit_log_format(ab, " a0=%lx a1=%lx a2=%lx a3=%lx items=%d" " ppid=%d pid=%d auid=%u uid=%u gid=%u" diff --git a/kernel/exit.c b/kernel/exit.c index 4e3f919edc48..fa235779b6a3 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -384,9 +384,7 @@ void daemonize(const char *name, ...) exit_mm(current); set_special_pids(1, 1); - mutex_lock(&tty_mutex); - current->signal->tty = NULL; - mutex_unlock(&tty_mutex); + proc_clear_tty(current); /* Block and flush all signals */ sigfillset(&blocked); diff --git a/kernel/sys.c b/kernel/sys.c index a0c1a29a507f..1ac2d1c5d84e 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1484,7 +1484,6 @@ asmlinkage long sys_setsid(void) pid_t session; int err = -EPERM; - mutex_lock(&tty_mutex); write_lock_irq(&tasklist_lock); /* Fail if I am already a session leader */ @@ -1504,12 +1503,15 @@ asmlinkage long sys_setsid(void) group_leader->signal->leader = 1; __set_special_pids(session, session); + + spin_lock(&group_leader->sighand->siglock); group_leader->signal->tty = NULL; group_leader->signal->tty_old_pgrp = 0; + spin_unlock(&group_leader->sighand->siglock); + err = process_group(group_leader); out: write_unlock_irq(&tasklist_lock); - mutex_unlock(&tty_mutex); return err; } diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 44e9cd470543..f5df8c70a9b5 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1695,9 +1695,10 @@ static inline void flush_unauthorized_files(struct files_struct * files) struct tty_struct *tty; struct fdtable *fdt; long j = -1; + int drop_tty = 0; mutex_lock(&tty_mutex); - tty = current->signal->tty; + tty = get_current_tty(); if (tty) { file_list_lock(); file = list_entry(tty->tty_files.next, typeof(*file), f_u.fu_list); @@ -1710,12 +1711,14 @@ static inline void flush_unauthorized_files(struct files_struct * files) struct inode *inode = file->f_dentry->d_inode; if (inode_has_perm(current, inode, FILE__READ | FILE__WRITE, NULL)) { - /* Reset controlling tty. */ - current->signal->tty = NULL; - current->signal->tty_old_pgrp = 0; + drop_tty = 1; } } file_list_unlock(); + + /* Reset controlling tty. */ + if (drop_tty) + proc_set_tty(current, NULL); } mutex_unlock(&tty_mutex); -- cgit v1.2.3 From ae424ae4b5bcd820ad6ee6f0b986c4e14ed4d6cf Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 8 Dec 2006 02:36:08 -0800 Subject: [PATCH] make set_special_pids() static Make set_special_pids() static, the only caller is daemonize(). Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 - kernel/exit.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index dede82c63445..5e8a0ba61749 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1240,7 +1240,6 @@ extern struct mm_struct init_mm; #define find_task_by_pid(nr) find_task_by_pid_type(PIDTYPE_PID, nr) extern struct task_struct *find_task_by_pid_type(int type, int pid); -extern void set_special_pids(pid_t session, pid_t pgrp); extern void __set_special_pids(pid_t session, pid_t pgrp); /* per-UID process charging. */ diff --git a/kernel/exit.c b/kernel/exit.c index fa235779b6a3..fa0495e167e9 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -314,7 +314,7 @@ void __set_special_pids(pid_t session, pid_t pgrp) } } -void set_special_pids(pid_t session, pid_t pgrp) +static void set_special_pids(pid_t session, pid_t pgrp) { write_lock_irq(&tasklist_lock); __set_special_pids(session, pgrp); -- cgit v1.2.3 From 0a0fc9601dd1024ec7171993bf075a789246e1ed Mon Sep 17 00:00:00 2001 From: Thomas Maier Date: Fri, 8 Dec 2006 02:36:11 -0800 Subject: [PATCH] pktcdvd: bio write congestion using congestion_wait() This adds a bio write queue congestion control to the pktcdvd driver with fixed on/off marks. It prevents that the driver consumes a unlimited amount of write requests. [akpm@osdl.org: sync with congestion_wait() renaming] Signed-off-by: Thomas Maier Cc: Peter Osterlund Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/pktcdvd.c | 32 ++++++++++++++++++++++++++++++++ include/linux/pktcdvd.h | 9 +++++++++ 2 files changed, 41 insertions(+) (limited to 'include/linux') diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 7e7b892f621a..91fbe7fd6d0b 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -84,6 +84,8 @@ static struct pktcdvd_device *pkt_devs[MAX_WRITERS]; static struct proc_dir_entry *pkt_proc; static int pktdev_major; +static int write_congestion_on = PKT_WRITE_CONGESTION_ON; +static int write_congestion_off = PKT_WRITE_CONGESTION_OFF; static struct mutex ctl_mutex; /* Serialize open/close/setup/teardown */ static mempool_t *psd_pool; @@ -894,6 +896,7 @@ static int pkt_handle_queue(struct pktcdvd_device *pd) sector_t zone = 0; /* Suppress gcc warning */ struct pkt_rb_node *node, *first_node; struct rb_node *n; + int wakeup; VPRINTK("handle_queue\n"); @@ -966,7 +969,13 @@ try_next_bio: pkt->write_size += bio->bi_size / CD_FRAMESIZE; spin_unlock(&pkt->lock); } + /* check write congestion marks, and if bio_queue_size is + below, wake up any waiters */ + wakeup = (pd->write_congestion_on > 0 + && pd->bio_queue_size <= pd->write_congestion_off); spin_unlock(&pd->lock); + if (wakeup) + blk_clear_queue_congested(pd->disk->queue, WRITE); pkt->sleep_time = max(PACKET_WAIT_TIME, 1); pkt_set_state(pkt, PACKET_WAITING_STATE); @@ -2179,6 +2188,23 @@ static int pkt_make_request(request_queue_t *q, struct bio *bio) } spin_unlock(&pd->cdrw.active_list_lock); + /* + * Test if there is enough room left in the bio work queue + * (queue size >= congestion on mark). + * If not, wait till the work queue size is below the congestion off mark. + */ + spin_lock(&pd->lock); + if (pd->write_congestion_on > 0 + && pd->bio_queue_size >= pd->write_congestion_on) { + blk_set_queue_congested(q, WRITE); + do { + spin_unlock(&pd->lock); + congestion_wait(WRITE, HZ); + spin_lock(&pd->lock); + } while(pd->bio_queue_size > pd->write_congestion_off); + } + spin_unlock(&pd->lock); + /* * No matching packet found. Store the bio in the work queue. */ @@ -2298,6 +2324,9 @@ static int pkt_seq_show(struct seq_file *m, void *p) seq_printf(m, "\tstate:\t\t\ti:%d ow:%d rw:%d ww:%d rec:%d fin:%d\n", states[0], states[1], states[2], states[3], states[4], states[5]); + seq_printf(m, "\twrite congestion marks:\toff=%d on=%d\n", + pd->write_congestion_off, + pd->write_congestion_on); return 0; } @@ -2474,6 +2503,9 @@ static int pkt_setup_dev(dev_t dev, dev_t* pkt_dev) init_waitqueue_head(&pd->wqueue); pd->bio_queue = RB_ROOT; + pd->write_congestion_on = write_congestion_on; + pd->write_congestion_off = write_congestion_off; + disk = alloc_disk(1); if (!disk) goto out_mem; diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h index 8a94c717c266..9b1a185fb1e5 100644 --- a/include/linux/pktcdvd.h +++ b/include/linux/pktcdvd.h @@ -112,6 +112,12 @@ struct pkt_ctrl_command { #include #include + +/* default bio write queue congestion marks */ +#define PKT_WRITE_CONGESTION_ON 10000 +#define PKT_WRITE_CONGESTION_OFF 9000 + + struct packet_settings { __u32 size; /* packet size in (512 byte) sectors */ @@ -271,6 +277,9 @@ struct pktcdvd_device struct packet_iosched iosched; struct gendisk *disk; + + int write_congestion_off; + int write_congestion_on; }; #endif /* __KERNEL__ */ -- cgit v1.2.3 From 32694850a91bd4fedcdd4a46292f870588be81d1 Mon Sep 17 00:00:00 2001 From: Thomas Maier Date: Fri, 8 Dec 2006 02:36:12 -0800 Subject: [PATCH] pktcdvd: add sysfs and debugfs interface Add a sysfs and debugfs interface to the pktcdvd driver. Look into the Documentation/ABI/testing/* files in the patch for more info. Signed-off-by: Thomas Maier Signed-off-by: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/ABI/testing/debugfs-pktcdvd | 20 ++ Documentation/ABI/testing/sysfs-class-pktcdvd | 72 +++++ Documentation/cdrom/packet-writing.txt | 35 ++ drivers/block/pktcdvd.c | 443 +++++++++++++++++++++++++- include/linux/pktcdvd.h | 18 +- 5 files changed, 583 insertions(+), 5 deletions(-) create mode 100644 Documentation/ABI/testing/debugfs-pktcdvd create mode 100644 Documentation/ABI/testing/sysfs-class-pktcdvd (limited to 'include/linux') diff --git a/Documentation/ABI/testing/debugfs-pktcdvd b/Documentation/ABI/testing/debugfs-pktcdvd new file mode 100644 index 000000000000..03dbd883cc41 --- /dev/null +++ b/Documentation/ABI/testing/debugfs-pktcdvd @@ -0,0 +1,20 @@ +What: /debug/pktcdvd/pktcdvd[0-7] +Date: Oct. 2006 +KernelVersion: 2.6.19 +Contact: Thomas Maier +Description: + +debugfs interface +----------------- + +The pktcdvd module (packet writing driver) creates +these files in debugfs: + +/debug/pktcdvd/pktcdvd[0-7]/ + info (0444) Lots of human readable driver + statistics and infos. Multiple lines! + +Example: +------- + +cat /debug/pktcdvd/pktcdvd0/info diff --git a/Documentation/ABI/testing/sysfs-class-pktcdvd b/Documentation/ABI/testing/sysfs-class-pktcdvd new file mode 100644 index 000000000000..c4c55edc9a5c --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-pktcdvd @@ -0,0 +1,72 @@ +What: /sys/class/pktcdvd/ +Date: Oct. 2006 +KernelVersion: 2.6.19 +Contact: Thomas Maier +Description: + +sysfs interface +--------------- + +The pktcdvd module (packet writing driver) creates +these files in the sysfs: +( is in format major:minor ) + +/sys/class/pktcdvd/ + add (0200) Write a block device id (major:minor) + to create a new pktcdvd device and map + it to the block device. + + remove (0200) Write the pktcdvd device id (major:minor) + to it to remove the pktcdvd device. + + device_map (0444) Shows the device mapping in format: + pktcdvd[0-7] + +/sys/class/pktcdvd/pktcdvd[0-7]/ + dev (0444) Device id + uevent (0200) To send an uevent. + +/sys/class/pktcdvd/pktcdvd[0-7]/stat/ + packets_started (0444) Number of started packets. + packets_finished (0444) Number of finished packets. + + kb_written (0444) kBytes written. + kb_read (0444) kBytes read. + kb_read_gather (0444) kBytes read to fill write packets. + + reset (0200) Write any value to it to reset + pktcdvd device statistic values, like + bytes read/written. + +/sys/class/pktcdvd/pktcdvd[0-7]/write_queue/ + size (0444) Contains the size of the bio write + queue. + + congestion_off (0644) If bio write queue size is below + this mark, accept new bio requests + from the block layer. + + congestion_on (0644) If bio write queue size is higher + as this mark, do no longer accept + bio write requests from the block + layer and wait till the pktcdvd + device has processed enough bio's + so that bio write queue size is + below congestion off mark. + A value of <= 0 disables congestion + control. + + +Example: +-------- +To use the pktcdvd sysfs interface directly, you can do: + +# create a new pktcdvd device mapped to /dev/hdc +echo "22:0" >/sys/class/pktcdvd/add +cat /sys/class/pktcdvd/device_map +# assuming device pktcdvd0 was created, look at stat's +cat /sys/class/pktcdvd/pktcdvd0/stat/kb_written +# print the device id of the mapped block device +fgrep pktcdvd0 /sys/class/pktcdvd/device_map +# remove device, using pktcdvd0 device id 253:0 +echo "253:0" >/sys/class/pktcdvd/remove diff --git a/Documentation/cdrom/packet-writing.txt b/Documentation/cdrom/packet-writing.txt index 3d44c561fe6d..7715d2247c4d 100644 --- a/Documentation/cdrom/packet-writing.txt +++ b/Documentation/cdrom/packet-writing.txt @@ -90,6 +90,41 @@ Notes to create an ext2 filesystem on the disc. +Using the pktcdvd sysfs interface +--------------------------------- + +Since Linux 2.6.19, the pktcdvd module has a sysfs interface +and can be controlled by it. For example the "pktcdvd" tool uses +this interface. (see http://people.freenet.de/BalaGi#pktcdvd ) + +"pktcdvd" works similar to "pktsetup", e.g.: + + # pktcdvd -a dev_name /dev/hdc + # mkudffs /dev/pktcdvd/dev_name + # mount -t udf -o rw,noatime /dev/pktcdvd/dev_name /dvdram + # cp files /dvdram + # umount /dvdram + # pktcdvd -r dev_name + + +For a description of the sysfs interface look into the file: + + Documentation/ABI/testing/sysfs-block-pktcdvd + + +Using the pktcdvd debugfs interface +----------------------------------- + +To read pktcdvd device infos in human readable form, do: + + # cat /debug/pktcdvd/pktcdvd[0-7]/info + +For a description of the debugfs interface look into the file: + + Documentation/ABI/testing/debugfs-pktcdvd + + + Links ----- diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 91fbe7fd6d0b..7c95c762950f 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -60,6 +60,8 @@ #include #include #include +#include +#include #include @@ -89,6 +91,419 @@ static int write_congestion_off = PKT_WRITE_CONGESTION_OFF; static struct mutex ctl_mutex; /* Serialize open/close/setup/teardown */ static mempool_t *psd_pool; +static struct class *class_pktcdvd = NULL; /* /sys/class/pktcdvd */ +static struct dentry *pkt_debugfs_root = NULL; /* /debug/pktcdvd */ + +/* forward declaration */ +static int pkt_setup_dev(dev_t dev, dev_t* pkt_dev); +static int pkt_remove_dev(dev_t pkt_dev); +static int pkt_seq_show(struct seq_file *m, void *p); + + + +/* + * create and register a pktcdvd kernel object. + */ +static struct pktcdvd_kobj* pkt_kobj_create(struct pktcdvd_device *pd, + const char* name, + struct kobject* parent, + struct kobj_type* ktype) +{ + struct pktcdvd_kobj *p; + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) + return NULL; + kobject_set_name(&p->kobj, "%s", name); + p->kobj.parent = parent; + p->kobj.ktype = ktype; + p->pd = pd; + if (kobject_register(&p->kobj) != 0) + return NULL; + return p; +} +/* + * remove a pktcdvd kernel object. + */ +static void pkt_kobj_remove(struct pktcdvd_kobj *p) +{ + if (p) + kobject_unregister(&p->kobj); +} +/* + * default release function for pktcdvd kernel objects. + */ +static void pkt_kobj_release(struct kobject *kobj) +{ + kfree(to_pktcdvdkobj(kobj)); +} + + +/********************************************************** + * + * sysfs interface for pktcdvd + * by (C) 2006 Thomas Maier + * + **********************************************************/ + +#define DEF_ATTR(_obj,_name,_mode) \ + static struct attribute _obj = { \ + .name = _name, .owner = THIS_MODULE, .mode = _mode } + +/********************************************************** + /sys/class/pktcdvd/pktcdvd[0-7]/ + stat/reset + stat/packets_started + stat/packets_finished + stat/kb_written + stat/kb_read + stat/kb_read_gather + write_queue/size + write_queue/congestion_off + write_queue/congestion_on + **********************************************************/ + +DEF_ATTR(kobj_pkt_attr_st1, "reset", 0200); +DEF_ATTR(kobj_pkt_attr_st2, "packets_started", 0444); +DEF_ATTR(kobj_pkt_attr_st3, "packets_finished", 0444); +DEF_ATTR(kobj_pkt_attr_st4, "kb_written", 0444); +DEF_ATTR(kobj_pkt_attr_st5, "kb_read", 0444); +DEF_ATTR(kobj_pkt_attr_st6, "kb_read_gather", 0444); + +static struct attribute *kobj_pkt_attrs_stat[] = { + &kobj_pkt_attr_st1, + &kobj_pkt_attr_st2, + &kobj_pkt_attr_st3, + &kobj_pkt_attr_st4, + &kobj_pkt_attr_st5, + &kobj_pkt_attr_st6, + NULL +}; + +DEF_ATTR(kobj_pkt_attr_wq1, "size", 0444); +DEF_ATTR(kobj_pkt_attr_wq2, "congestion_off", 0644); +DEF_ATTR(kobj_pkt_attr_wq3, "congestion_on", 0644); + +static struct attribute *kobj_pkt_attrs_wqueue[] = { + &kobj_pkt_attr_wq1, + &kobj_pkt_attr_wq2, + &kobj_pkt_attr_wq3, + NULL +}; + +/* declares a char buffer[64] _dbuf, copies data from + * _b with length _l into it and ensures that _dbuf ends + * with a \0 character. + */ +#define DECLARE_BUF_AS_STRING(_dbuf, _b, _l) \ + char _dbuf[64]; int dlen = (_l) < 0 ? 0 : (_l); \ + if (dlen >= sizeof(_dbuf)) dlen = sizeof(_dbuf)-1; \ + memcpy(_dbuf, _b, dlen); _dbuf[dlen] = 0 + +static ssize_t kobj_pkt_show(struct kobject *kobj, + struct attribute *attr, char *data) +{ + struct pktcdvd_device *pd = to_pktcdvdkobj(kobj)->pd; + int n = 0; + int v; + if (strcmp(attr->name, "packets_started") == 0) { + n = sprintf(data, "%lu\n", pd->stats.pkt_started); + + } else if (strcmp(attr->name, "packets_finished") == 0) { + n = sprintf(data, "%lu\n", pd->stats.pkt_ended); + + } else if (strcmp(attr->name, "kb_written") == 0) { + n = sprintf(data, "%lu\n", pd->stats.secs_w >> 1); + + } else if (strcmp(attr->name, "kb_read") == 0) { + n = sprintf(data, "%lu\n", pd->stats.secs_r >> 1); + + } else if (strcmp(attr->name, "kb_read_gather") == 0) { + n = sprintf(data, "%lu\n", pd->stats.secs_rg >> 1); + + } else if (strcmp(attr->name, "size") == 0) { + spin_lock(&pd->lock); + v = pd->bio_queue_size; + spin_unlock(&pd->lock); + n = sprintf(data, "%d\n", v); + + } else if (strcmp(attr->name, "congestion_off") == 0) { + spin_lock(&pd->lock); + v = pd->write_congestion_off; + spin_unlock(&pd->lock); + n = sprintf(data, "%d\n", v); + + } else if (strcmp(attr->name, "congestion_on") == 0) { + spin_lock(&pd->lock); + v = pd->write_congestion_on; + spin_unlock(&pd->lock); + n = sprintf(data, "%d\n", v); + } + return n; +} + +static void init_write_congestion_marks(int* lo, int* hi) +{ + if (*hi > 0) { + *hi = max(*hi, 500); + *hi = min(*hi, 1000000); + if (*lo <= 0) + *lo = *hi - 100; + else { + *lo = min(*lo, *hi - 100); + *lo = max(*lo, 100); + } + } else { + *hi = -1; + *lo = -1; + } +} + +static ssize_t kobj_pkt_store(struct kobject *kobj, + struct attribute *attr, + const char *data, size_t len) +{ + struct pktcdvd_device *pd = to_pktcdvdkobj(kobj)->pd; + int val; + DECLARE_BUF_AS_STRING(dbuf, data, len); /* ensure sscanf scans a string */ + + if (strcmp(attr->name, "reset") == 0 && dlen > 0) { + pd->stats.pkt_started = 0; + pd->stats.pkt_ended = 0; + pd->stats.secs_w = 0; + pd->stats.secs_rg = 0; + pd->stats.secs_r = 0; + + } else if (strcmp(attr->name, "congestion_off") == 0 + && sscanf(dbuf, "%d", &val) == 1) { + spin_lock(&pd->lock); + pd->write_congestion_off = val; + init_write_congestion_marks(&pd->write_congestion_off, + &pd->write_congestion_on); + spin_unlock(&pd->lock); + + } else if (strcmp(attr->name, "congestion_on") == 0 + && sscanf(dbuf, "%d", &val) == 1) { + spin_lock(&pd->lock); + pd->write_congestion_on = val; + init_write_congestion_marks(&pd->write_congestion_off, + &pd->write_congestion_on); + spin_unlock(&pd->lock); + } + return len; +} + +static struct sysfs_ops kobj_pkt_ops = { + .show = kobj_pkt_show, + .store = kobj_pkt_store +}; +static struct kobj_type kobj_pkt_type_stat = { + .release = pkt_kobj_release, + .sysfs_ops = &kobj_pkt_ops, + .default_attrs = kobj_pkt_attrs_stat +}; +static struct kobj_type kobj_pkt_type_wqueue = { + .release = pkt_kobj_release, + .sysfs_ops = &kobj_pkt_ops, + .default_attrs = kobj_pkt_attrs_wqueue +}; + +static void pkt_sysfs_dev_new(struct pktcdvd_device *pd) +{ + if (class_pktcdvd) { + pd->clsdev = class_device_create(class_pktcdvd, + NULL, pd->pkt_dev, + NULL, "%s", pd->name); + if (IS_ERR(pd->clsdev)) + pd->clsdev = NULL; + } + if (pd->clsdev) { + pd->kobj_stat = pkt_kobj_create(pd, "stat", + &pd->clsdev->kobj, + &kobj_pkt_type_stat); + pd->kobj_wqueue = pkt_kobj_create(pd, "write_queue", + &pd->clsdev->kobj, + &kobj_pkt_type_wqueue); + } +} + +static void pkt_sysfs_dev_remove(struct pktcdvd_device *pd) +{ + pkt_kobj_remove(pd->kobj_stat); + pkt_kobj_remove(pd->kobj_wqueue); + if (class_pktcdvd) + class_device_destroy(class_pktcdvd, pd->pkt_dev); +} + + +/******************************************************************** + /sys/class/pktcdvd/ + add map block device + remove unmap packet dev + device_map show mappings + *******************************************************************/ + +static void class_pktcdvd_release(struct class *cls) +{ + kfree(cls); +} +static ssize_t class_pktcdvd_show_map(struct class *c, char *data) +{ + int n = 0; + int idx; + mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING); + for (idx = 0; idx < MAX_WRITERS; idx++) { + struct pktcdvd_device *pd = pkt_devs[idx]; + if (!pd) + continue; + n += sprintf(data+n, "%s %u:%u %u:%u\n", + pd->name, + MAJOR(pd->pkt_dev), MINOR(pd->pkt_dev), + MAJOR(pd->bdev->bd_dev), + MINOR(pd->bdev->bd_dev)); + } + mutex_unlock(&ctl_mutex); + return n; +} + +static ssize_t class_pktcdvd_store_add(struct class *c, const char *buf, + size_t count) +{ + unsigned int major, minor; + DECLARE_BUF_AS_STRING(dbuf, buf, count); + if (sscanf(dbuf, "%u:%u", &major, &minor) == 2) { + pkt_setup_dev(MKDEV(major, minor), NULL); + return count; + } + return -EINVAL; +} + +static ssize_t class_pktcdvd_store_remove(struct class *c, const char *buf, + size_t count) +{ + unsigned int major, minor; + DECLARE_BUF_AS_STRING(dbuf, buf, count); + if (sscanf(dbuf, "%u:%u", &major, &minor) == 2) { + pkt_remove_dev(MKDEV(major, minor)); + return count; + } + return -EINVAL; +} + +static struct class_attribute class_pktcdvd_attrs[] = { + __ATTR(add, 0200, NULL, class_pktcdvd_store_add), + __ATTR(remove, 0200, NULL, class_pktcdvd_store_remove), + __ATTR(device_map, 0444, class_pktcdvd_show_map, NULL), + __ATTR_NULL +}; + + +static int pkt_sysfs_init(void) +{ + int ret = 0; + + /* + * create control files in sysfs + * /sys/class/pktcdvd/... + */ + class_pktcdvd = kzalloc(sizeof(*class_pktcdvd), GFP_KERNEL); + if (!class_pktcdvd) + return -ENOMEM; + class_pktcdvd->name = DRIVER_NAME; + class_pktcdvd->owner = THIS_MODULE; + class_pktcdvd->class_release = class_pktcdvd_release; + class_pktcdvd->class_attrs = class_pktcdvd_attrs; + ret = class_register(class_pktcdvd); + if (ret) { + kfree(class_pktcdvd); + class_pktcdvd = NULL; + printk(DRIVER_NAME": failed to create class pktcdvd\n"); + return ret; + } + return 0; +} + +static void pkt_sysfs_cleanup(void) +{ + if (class_pktcdvd) + class_destroy(class_pktcdvd); + class_pktcdvd = NULL; +} + +/******************************************************************** + entries in debugfs + + /debugfs/pktcdvd[0-7]/ + info + + *******************************************************************/ + +static int pkt_debugfs_seq_show(struct seq_file *m, void *p) +{ + return pkt_seq_show(m, p); +} + +static int pkt_debugfs_fops_open(struct inode *inode, struct file *file) +{ + return single_open(file, pkt_debugfs_seq_show, inode->i_private); +} + +static struct file_operations debug_fops = { + .open = pkt_debugfs_fops_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .owner = THIS_MODULE, +}; + +static void pkt_debugfs_dev_new(struct pktcdvd_device *pd) +{ + if (!pkt_debugfs_root) + return; + pd->dfs_f_info = NULL; + pd->dfs_d_root = debugfs_create_dir(pd->name, pkt_debugfs_root); + if (IS_ERR(pd->dfs_d_root)) { + pd->dfs_d_root = NULL; + return; + } + pd->dfs_f_info = debugfs_create_file("info", S_IRUGO, + pd->dfs_d_root, pd, &debug_fops); + if (IS_ERR(pd->dfs_f_info)) { + pd->dfs_f_info = NULL; + return; + } +} + +static void pkt_debugfs_dev_remove(struct pktcdvd_device *pd) +{ + if (!pkt_debugfs_root) + return; + if (pd->dfs_f_info) + debugfs_remove(pd->dfs_f_info); + pd->dfs_f_info = NULL; + if (pd->dfs_d_root) + debugfs_remove(pd->dfs_d_root); + pd->dfs_d_root = NULL; +} + +static void pkt_debugfs_init(void) +{ + pkt_debugfs_root = debugfs_create_dir(DRIVER_NAME, NULL); + if (IS_ERR(pkt_debugfs_root)) { + pkt_debugfs_root = NULL; + return; + } +} + +static void pkt_debugfs_cleanup(void) +{ + if (!pkt_debugfs_root) + return; + debugfs_remove(pkt_debugfs_root); + pkt_debugfs_root = NULL; +} + +/* ----------------------------------------------------------*/ + static void pkt_bio_finished(struct pktcdvd_device *pd) { @@ -2527,6 +2942,9 @@ static int pkt_setup_dev(dev_t dev, dev_t* pkt_dev) add_disk(disk); + pkt_sysfs_dev_new(pd); + pkt_debugfs_dev_new(pd); + pkt_devs[idx] = pd; if (pkt_dev) *pkt_dev = pd->pkt_dev; @@ -2577,6 +2995,11 @@ static int pkt_remove_dev(dev_t pkt_dev) if (!IS_ERR(pd->cdrw.thread)) kthread_stop(pd->cdrw.thread); + pkt_devs[idx] = NULL; + + pkt_debugfs_dev_remove(pd); + pkt_sysfs_dev_remove(pd); + blkdev_put(pd->bdev); remove_proc_entry(pd->name, pkt_proc); @@ -2586,7 +3009,6 @@ static int pkt_remove_dev(dev_t pkt_dev) blk_cleanup_queue(pd->disk->queue); put_disk(pd->disk); - pkt_devs[idx] = NULL; mempool_destroy(pd->rb_pool); kfree(pd); @@ -2670,6 +3092,8 @@ static int __init pkt_init(void) { int ret; + mutex_init(&ctl_mutex); + psd_pool = mempool_create_kmalloc_pool(PSD_POOL_SIZE, sizeof(struct packet_stacked_data)); if (!psd_pool) @@ -2683,18 +3107,25 @@ static int __init pkt_init(void) if (!pktdev_major) pktdev_major = ret; + ret = pkt_sysfs_init(); + if (ret) + goto out; + + pkt_debugfs_init(); + ret = misc_register(&pkt_misc); if (ret) { printk(DRIVER_NAME": Unable to register misc device\n"); - goto out; + goto out_misc; } - mutex_init(&ctl_mutex); - pkt_proc = proc_mkdir(DRIVER_NAME, proc_root_driver); return 0; +out_misc: + pkt_debugfs_cleanup(); + pkt_sysfs_cleanup(); out: unregister_blkdev(pktdev_major, DRIVER_NAME); out2: @@ -2706,6 +3137,10 @@ static void __exit pkt_exit(void) { remove_proc_entry(DRIVER_NAME, proc_root_driver); misc_deregister(&pkt_misc); + + pkt_debugfs_cleanup(); + pkt_sysfs_cleanup(); + unregister_blkdev(pktdev_major, DRIVER_NAME); mempool_destroy(psd_pool); } diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h index 9b1a185fb1e5..5ea4f05683f6 100644 --- a/include/linux/pktcdvd.h +++ b/include/linux/pktcdvd.h @@ -111,7 +111,8 @@ struct pkt_ctrl_command { #include #include #include - +#include +#include /* default bio write queue congestion marks */ #define PKT_WRITE_CONGESTION_ON 10000 @@ -247,6 +248,14 @@ struct packet_stacked_data }; #define PSD_POOL_SIZE 64 +struct pktcdvd_kobj +{ + struct kobject kobj; + struct pktcdvd_device *pd; +}; +#define to_pktcdvdkobj(_k) \ + ((struct pktcdvd_kobj*)container_of(_k,struct pktcdvd_kobj,kobj)) + struct pktcdvd_device { struct block_device *bdev; /* dev attached */ @@ -280,6 +289,13 @@ struct pktcdvd_device int write_congestion_off; int write_congestion_on; + + struct class_device *clsdev; /* sysfs pktcdvd[0-7] class dev */ + struct pktcdvd_kobj *kobj_stat; /* sysfs pktcdvd[0-7]/stat/ */ + struct pktcdvd_kobj *kobj_wqueue; /* sysfs pktcdvd[0-7]/write_queue/ */ + + struct dentry *dfs_d_root; /* debugfs: devname directory */ + struct dentry *dfs_f_info; /* debugfs: info file */ }; #endif /* __KERNEL__ */ -- cgit v1.2.3 From 2e7b651df113c8a463853e4169951c52c39f9d19 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 8 Dec 2006 02:36:13 -0800 Subject: [PATCH] remove the old bd_mutex lockdep annotation Remove the old complex and crufty bd_mutex annotation. Signed-off-by: Peter Zijlstra Cc: Neil Brown Cc: Ingo Molnar Cc: Arjan van de Ven Cc: Jason Baron Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- block/ioctl.c | 4 +- drivers/md/md.c | 6 +- fs/block_dev.c | 180 ++++++++++++++++------------------------------------- include/linux/fs.h | 17 ----- 4 files changed, 60 insertions(+), 147 deletions(-) (limited to 'include/linux') diff --git a/block/ioctl.c b/block/ioctl.c index 58aab630dfc1..9fddb99c3efb 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -72,7 +72,7 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user bdevp = bdget_disk(disk, part); if (!bdevp) return -ENOMEM; - mutex_lock_nested(&bdevp->bd_mutex, BD_MUTEX_PARTITION); + mutex_lock(&bdevp->bd_mutex); if (bdevp->bd_openers) { mutex_unlock(&bdevp->bd_mutex); bdput(bdevp); @@ -82,7 +82,7 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user fsync_bdev(bdevp); invalidate_bdev(bdevp, 0); - mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_WHOLE); + mutex_lock(&bdev->bd_mutex); delete_partition(disk, part); mutex_unlock(&bdev->bd_mutex); mutex_unlock(&bdevp->bd_mutex); diff --git a/drivers/md/md.c b/drivers/md/md.c index 6c4345bde07e..d2bcb410e3ab 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1413,7 +1413,7 @@ static int lock_rdev(mdk_rdev_t *rdev, dev_t dev) struct block_device *bdev; char b[BDEVNAME_SIZE]; - bdev = open_partition_by_devnum(dev, FMODE_READ|FMODE_WRITE); + bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE); if (IS_ERR(bdev)) { printk(KERN_ERR "md: could not open %s.\n", __bdevname(dev, b)); @@ -1423,7 +1423,7 @@ static int lock_rdev(mdk_rdev_t *rdev, dev_t dev) if (err) { printk(KERN_ERR "md: could not bd_claim %s.\n", bdevname(bdev, b)); - blkdev_put_partition(bdev); + blkdev_put(bdev); return err; } rdev->bdev = bdev; @@ -1437,7 +1437,7 @@ static void unlock_rdev(mdk_rdev_t *rdev) if (!bdev) MD_BUG(); bd_release(bdev); - blkdev_put_partition(bdev); + blkdev_put(bdev); } void md_autodetect_dev(dev_t dev); diff --git a/fs/block_dev.c b/fs/block_dev.c index 13816b4d76f6..36c38f48a4ed 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -762,7 +762,7 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, if (!bo) return -ENOMEM; - mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); + mutex_lock(&bdev->bd_mutex); res = bd_claim(bdev, holder); if (res == 0) { found = find_bd_holder(bdev, bo); @@ -796,7 +796,7 @@ static void bd_release_from_kobject(struct block_device *bdev, if (!kobj) return; - mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); + mutex_lock(&bdev->bd_mutex); bd_release(bdev); if ((bo = del_bd_holder(bdev, kobj))) free_bd_holder(bo); @@ -854,22 +854,6 @@ struct block_device *open_by_devnum(dev_t dev, unsigned mode) EXPORT_SYMBOL(open_by_devnum); -static int -blkdev_get_partition(struct block_device *bdev, mode_t mode, unsigned flags); - -struct block_device *open_partition_by_devnum(dev_t dev, unsigned mode) -{ - struct block_device *bdev = bdget(dev); - int err = -ENOMEM; - int flags = mode & FMODE_WRITE ? O_RDWR : O_RDONLY; - if (bdev) - err = blkdev_get_partition(bdev, mode, flags); - return err ? ERR_PTR(err) : bdev; -} - -EXPORT_SYMBOL(open_partition_by_devnum); - - /* * This routine checks whether a removable media has been changed, * and invalidates all buffer-cache-entries in that case. This @@ -916,66 +900,7 @@ void bd_set_size(struct block_device *bdev, loff_t size) } EXPORT_SYMBOL(bd_set_size); -static int __blkdev_put(struct block_device *bdev, unsigned int subclass) -{ - int ret = 0; - struct inode *bd_inode = bdev->bd_inode; - struct gendisk *disk = bdev->bd_disk; - - mutex_lock_nested(&bdev->bd_mutex, subclass); - lock_kernel(); - if (!--bdev->bd_openers) { - sync_blockdev(bdev); - kill_bdev(bdev); - } - if (bdev->bd_contains == bdev) { - if (disk->fops->release) - ret = disk->fops->release(bd_inode, NULL); - } else { - mutex_lock_nested(&bdev->bd_contains->bd_mutex, - subclass + 1); - bdev->bd_contains->bd_part_count--; - mutex_unlock(&bdev->bd_contains->bd_mutex); - } - if (!bdev->bd_openers) { - struct module *owner = disk->fops->owner; - - put_disk(disk); - module_put(owner); - - if (bdev->bd_contains != bdev) { - kobject_put(&bdev->bd_part->kobj); - bdev->bd_part = NULL; - } - bdev->bd_disk = NULL; - bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; - if (bdev != bdev->bd_contains) - __blkdev_put(bdev->bd_contains, subclass + 1); - bdev->bd_contains = NULL; - } - unlock_kernel(); - mutex_unlock(&bdev->bd_mutex); - bdput(bdev); - return ret; -} - -int blkdev_put(struct block_device *bdev) -{ - return __blkdev_put(bdev, BD_MUTEX_NORMAL); -} -EXPORT_SYMBOL(blkdev_put); - -int blkdev_put_partition(struct block_device *bdev) -{ - return __blkdev_put(bdev, BD_MUTEX_PARTITION); -} -EXPORT_SYMBOL(blkdev_put_partition); - -static int -blkdev_get_whole(struct block_device *bdev, mode_t mode, unsigned flags); - -static int -do_open(struct block_device *bdev, struct file *file, unsigned int subclass) +static int do_open(struct block_device *bdev, struct file *file) { struct module *owner = NULL; struct gendisk *disk; @@ -992,8 +917,7 @@ do_open(struct block_device *bdev, struct file *file, unsigned int subclass) } owner = disk->fops->owner; - mutex_lock_nested(&bdev->bd_mutex, subclass); - + mutex_lock(&bdev->bd_mutex); if (!bdev->bd_openers) { bdev->bd_disk = disk; bdev->bd_contains = bdev; @@ -1020,11 +944,11 @@ do_open(struct block_device *bdev, struct file *file, unsigned int subclass) ret = -ENOMEM; if (!whole) goto out_first; - ret = blkdev_get_whole(whole, file->f_mode, file->f_flags); + ret = blkdev_get(whole, file->f_mode, file->f_flags); if (ret) goto out_first; bdev->bd_contains = whole; - mutex_lock_nested(&whole->bd_mutex, BD_MUTEX_WHOLE); + mutex_lock(&whole->bd_mutex); whole->bd_part_count++; p = disk->part[part - 1]; bdev->bd_inode->i_data.backing_dev_info = @@ -1052,8 +976,7 @@ do_open(struct block_device *bdev, struct file *file, unsigned int subclass) if (bdev->bd_invalidated) rescan_partitions(bdev->bd_disk, bdev); } else { - mutex_lock_nested(&bdev->bd_contains->bd_mutex, - BD_MUTEX_WHOLE); + mutex_lock(&bdev->bd_contains->bd_mutex); bdev->bd_contains->bd_part_count++; mutex_unlock(&bdev->bd_contains->bd_mutex); } @@ -1067,7 +990,7 @@ out_first: bdev->bd_disk = NULL; bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) - __blkdev_put(bdev->bd_contains, BD_MUTEX_WHOLE); + blkdev_put(bdev->bd_contains); bdev->bd_contains = NULL; put_disk(disk); module_put(owner); @@ -1094,49 +1017,11 @@ int blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags) fake_file.f_dentry = &fake_dentry; fake_dentry.d_inode = bdev->bd_inode; - return do_open(bdev, &fake_file, BD_MUTEX_NORMAL); + return do_open(bdev, &fake_file); } EXPORT_SYMBOL(blkdev_get); -static int -blkdev_get_whole(struct block_device *bdev, mode_t mode, unsigned flags) -{ - /* - * This crockload is due to bad choice of ->open() type. - * It will go away. - * For now, block device ->open() routine must _not_ - * examine anything in 'inode' argument except ->i_rdev. - */ - struct file fake_file = {}; - struct dentry fake_dentry = {}; - fake_file.f_mode = mode; - fake_file.f_flags = flags; - fake_file.f_dentry = &fake_dentry; - fake_dentry.d_inode = bdev->bd_inode; - - return do_open(bdev, &fake_file, BD_MUTEX_WHOLE); -} - -static int -blkdev_get_partition(struct block_device *bdev, mode_t mode, unsigned flags) -{ - /* - * This crockload is due to bad choice of ->open() type. - * It will go away. - * For now, block device ->open() routine must _not_ - * examine anything in 'inode' argument except ->i_rdev. - */ - struct file fake_file = {}; - struct dentry fake_dentry = {}; - fake_file.f_mode = mode; - fake_file.f_flags = flags; - fake_file.f_dentry = &fake_dentry; - fake_dentry.d_inode = bdev->bd_inode; - - return do_open(bdev, &fake_file, BD_MUTEX_PARTITION); -} - static int blkdev_open(struct inode * inode, struct file * filp) { struct block_device *bdev; @@ -1154,7 +1039,7 @@ static int blkdev_open(struct inode * inode, struct file * filp) if (bdev == NULL) return -ENOMEM; - res = do_open(bdev, filp, BD_MUTEX_NORMAL); + res = do_open(bdev, filp); if (res) return res; @@ -1168,6 +1053,51 @@ static int blkdev_open(struct inode * inode, struct file * filp) return res; } +int blkdev_put(struct block_device *bdev) +{ + int ret = 0; + struct inode *bd_inode = bdev->bd_inode; + struct gendisk *disk = bdev->bd_disk; + + mutex_lock(&bdev->bd_mutex); + lock_kernel(); + if (!--bdev->bd_openers) { + sync_blockdev(bdev); + kill_bdev(bdev); + } + if (bdev->bd_contains == bdev) { + if (disk->fops->release) + ret = disk->fops->release(bd_inode, NULL); + } else { + mutex_lock(&bdev->bd_contains->bd_mutex); + bdev->bd_contains->bd_part_count--; + mutex_unlock(&bdev->bd_contains->bd_mutex); + } + if (!bdev->bd_openers) { + struct module *owner = disk->fops->owner; + + put_disk(disk); + module_put(owner); + + if (bdev->bd_contains != bdev) { + kobject_put(&bdev->bd_part->kobj); + bdev->bd_part = NULL; + } + bdev->bd_disk = NULL; + bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; + if (bdev != bdev->bd_contains) { + blkdev_put(bdev->bd_contains); + } + bdev->bd_contains = NULL; + } + unlock_kernel(); + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return ret; +} + +EXPORT_SYMBOL(blkdev_put); + static int blkdev_close(struct inode * inode, struct file * filp) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); diff --git a/include/linux/fs.h b/include/linux/fs.h index 70b99fbb560b..45f2cabb8c75 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -481,21 +481,6 @@ struct block_device { unsigned long bd_private; }; -/* - * bdev->bd_mutex nesting subclasses for the lock validator: - * - * 0: normal - * 1: 'whole' - * 2: 'partition' - */ -enum bdev_bd_mutex_lock_class -{ - BD_MUTEX_NORMAL, - BD_MUTEX_WHOLE, - BD_MUTEX_PARTITION -}; - - /* * Radix-tree tags, for tagging dirty and writeback pages within the pagecache * radix trees @@ -1499,7 +1484,6 @@ extern void bd_set_size(struct block_device *, loff_t size); extern void bd_forget(struct inode *inode); extern void bdput(struct block_device *); extern struct block_device *open_by_devnum(dev_t, unsigned); -extern struct block_device *open_partition_by_devnum(dev_t, unsigned); extern const struct address_space_operations def_blk_aops; #else static inline void bd_forget(struct inode *inode) {} @@ -1517,7 +1501,6 @@ extern int blkdev_driver_ioctl(struct inode *inode, struct file *file, extern long compat_blkdev_ioctl(struct file *, unsigned, unsigned long); extern int blkdev_get(struct block_device *, mode_t, unsigned); extern int blkdev_put(struct block_device *); -extern int blkdev_put_partition(struct block_device *); extern int bd_claim(struct block_device *, void *); extern void bd_release(struct block_device *); #ifdef CONFIG_SYSFS -- cgit v1.2.3 From d63a5a74dee87883fda6b7d170244acaac5b05e8 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 8 Dec 2006 02:36:17 -0800 Subject: [PATCH] lockdep: avoid lockdep warning in md md_open takes ->reconfig_mutex which causes lockdep to complain. This (normally) doesn't have deadlock potential as the possible conflict is with a reconfig_mutex in a different device. I say "normally" because if a loop were created in the array->member hierarchy a deadlock could happen. However that causes bigger problems than a deadlock and should be fixed independently. So we flag the lock in md_open as a nested lock. This requires defining mutex_lock_interruptible_nested. Cc: Ingo Molnar Acked-by: Peter Zijlstra Acked-by: Ingo Molnar Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 2 +- include/linux/mutex.h | 2 ++ kernel/mutex.c | 9 +++++++++ 3 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/md/md.c b/drivers/md/md.c index d2bcb410e3ab..3ce7a5d81d81 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -4423,7 +4423,7 @@ static int md_open(struct inode *inode, struct file *file) mddev_t *mddev = inode->i_bdev->bd_disk->private_data; int err; - if ((err = mddev_lock(mddev))) + if ((err = mutex_lock_interruptible_nested(&mddev->reconfig_mutex, 1))) goto out; err = 0; diff --git a/include/linux/mutex.h b/include/linux/mutex.h index b2b91c477563..a7544afd7582 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -125,8 +125,10 @@ extern int fastcall mutex_lock_interruptible(struct mutex *lock); #ifdef CONFIG_DEBUG_LOCK_ALLOC extern void mutex_lock_nested(struct mutex *lock, unsigned int subclass); +extern int mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass); #else # define mutex_lock_nested(lock, subclass) mutex_lock(lock) +# define mutex_lock_interruptible_nested(lock, subclass) mutex_lock_interruptible(lock) #endif /* diff --git a/kernel/mutex.c b/kernel/mutex.c index 8c71cf72a497..e7cbbb82765b 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -206,6 +206,15 @@ mutex_lock_nested(struct mutex *lock, unsigned int subclass) } EXPORT_SYMBOL_GPL(mutex_lock_nested); + +int __sched +mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass) +{ + might_sleep(); + return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, subclass); +} + +EXPORT_SYMBOL_GPL(mutex_lock_interruptible_nested); #endif /* -- cgit v1.2.3 From 7664c5a1da4711bb6383117f51b94c8dc8f3f1cd Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 8 Dec 2006 02:36:19 -0800 Subject: [PATCH] Generic BUG implementation This patch adds common handling for kernel BUGs, for use by architectures as they wish. The code is derived from arch/powerpc. The advantages of having common BUG handling are: - consistent BUG reporting across architectures - shared implementation of out-of-line file/line data - implement CONFIG_DEBUG_BUGVERBOSE consistently This means that in inline impact of BUG is just the illegal instruction itself, which is an improvement for i386 and x86-64. A BUG is represented in the instruction stream as an illegal instruction, which has file/line information associated with it. This extra information is stored in the __bug_table section in the ELF file. When the kernel gets an illegal instruction, it first confirms it might possibly be from a BUG (ie, in kernel mode, the right illegal instruction). It then calls report_bug(). This searches __bug_table for a matching instruction pointer, and if found, prints the corresponding file/line information. If report_bug() determines that it wasn't a BUG which caused the trap, it returns BUG_TRAP_TYPE_NONE. Some architectures (powerpc) implement WARN using the same mechanism; if the illegal instruction was the result of a WARN, then report_bug(Q) returns CONFIG_DEBUG_BUGVERBOSE; otherwise it returns BUG_TRAP_TYPE_BUG. lib/bug.c keeps a list of loaded modules which can be searched for __bug_table entries. The architecture must call module_bug_finalize()/module_bug_cleanup() from its corresponding module_finalize/cleanup functions. Unsetting CONFIG_DEBUG_BUGVERBOSE will reduce the kernel size by some amount. At the very least, filename and line information will not be recorded for each but, but architectures may decide to store no extra information per BUG at all. Unfortunately, gcc doesn't have a general way to mark an asm() as noreturn, so architectures will generally have to include an infinite loop (or similar) in the BUG code, so that gcc knows execution won't continue beyond that point. gcc does have a __builtin_trap() operator which may be useful to achieve the same effect, unfortunately it cannot be used to actually implement the BUG itself, because there's no way to get the instruction's address for use in generating the __bug_table entry. [randy.dunlap@oracle.com: Handle BUG=n, GENERIC_BUG=n to prevent build errors] [bunk@stusta.de: include/linux/bug.h must always #include Cc: Andi Kleen Cc: Hugh Dickens Cc: Michael Ellerman Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Rusty Russell Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-generic/bug.h | 16 ++++ include/asm-generic/vmlinux.lds.h | 8 ++ include/linux/bug.h | 47 +++++++++++ include/linux/module.h | 7 ++ lib/Kconfig.debug | 2 +- lib/Makefile | 2 + lib/bug.c | 163 ++++++++++++++++++++++++++++++++++++++ 7 files changed, 244 insertions(+), 1 deletion(-) create mode 100644 include/linux/bug.h create mode 100644 lib/bug.c (limited to 'include/linux') diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index c92ae0f166ff..47e3561638b1 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -4,6 +4,22 @@ #include #ifdef CONFIG_BUG + +#ifdef CONFIG_GENERIC_BUG +#ifndef __ASSEMBLY__ +struct bug_entry { + unsigned long bug_addr; +#ifdef CONFIG_DEBUG_BUGVERBOSE + const char *file; + unsigned short line; +#endif + unsigned short flags; +}; +#endif /* __ASSEMBLY__ */ + +#define BUGFLAG_WARNING (1<<0) +#endif /* CONFIG_GENERIC_BUG */ + #ifndef HAVE_ARCH_BUG #define BUG() do { \ printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __FUNCTION__); \ diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 4d4c62d11059..6e9fcebbf89f 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -218,6 +218,14 @@ .stab.indexstr 0 : { *(.stab.indexstr) } \ .comment 0 : { *(.comment) } +#define BUG_TABLE \ + . = ALIGN(8); \ + __bug_table : AT(ADDR(__bug_table) - LOAD_OFFSET) { \ + __start___bug_table = .; \ + *(__bug_table) \ + __stop___bug_table = .; \ + } + #define NOTES \ .notes : { *(.note.*) } :note diff --git a/include/linux/bug.h b/include/linux/bug.h new file mode 100644 index 000000000000..42aa0a54b6f4 --- /dev/null +++ b/include/linux/bug.h @@ -0,0 +1,47 @@ +#ifndef _LINUX_BUG_H +#define _LINUX_BUG_H + +#include +#include + +enum bug_trap_type { + BUG_TRAP_TYPE_NONE = 0, + BUG_TRAP_TYPE_WARN = 1, + BUG_TRAP_TYPE_BUG = 2, +}; + +#ifdef CONFIG_GENERIC_BUG +#include + +static inline int is_warning_bug(const struct bug_entry *bug) +{ + return bug->flags & BUGFLAG_WARNING; +} + +const struct bug_entry *find_bug(unsigned long bugaddr); + +enum bug_trap_type report_bug(unsigned long bug_addr); + +int module_bug_finalize(const Elf_Ehdr *, const Elf_Shdr *, + struct module *); +void module_bug_cleanup(struct module *); + +/* These are defined by the architecture */ +int is_valid_bugaddr(unsigned long addr); + +#else /* !CONFIG_GENERIC_BUG */ + +static inline enum bug_trap_type report_bug(unsigned long bug_addr) +{ + return BUG_TRAP_TYPE_BUG; +} +static inline int module_bug_finalize(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + struct module *mod) +{ + return 0; +} +static inline void module_bug_cleanup(struct module *mod) {} + +#endif /* CONFIG_GENERIC_BUG */ +#endif /* _LINUX_BUG_H */ diff --git a/include/linux/module.h b/include/linux/module.h index d33df2408e05..10f771a49997 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -319,6 +319,13 @@ struct module unsigned int taints; /* same bits as kernel:tainted */ +#ifdef CONFIG_GENERIC_BUG + /* Support for BUG */ + struct list_head bug_list; + struct bug_entry *bug_table; + unsigned num_bugs; +#endif + #ifdef CONFIG_MODULE_UNLOAD /* Reference counts */ struct module_ref ref[NR_CPUS]; diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index b75fed737f25..ee46fb335d24 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -285,7 +285,7 @@ config DEBUG_HIGHMEM config DEBUG_BUGVERBOSE bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED depends on BUG - depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV || SUPERH + depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV || SUPERH || GENERIC_BUG default !EMBEDDED help Say Y here to make BUG() panics output the file name and line number diff --git a/lib/Makefile b/lib/Makefile index fea8f9035f07..7ab486b883a8 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -55,6 +55,8 @@ obj-$(CONFIG_AUDIT_GENERIC) += audit.o obj-$(CONFIG_SWIOTLB) += swiotlb.o +lib-$(CONFIG_GENERIC_BUG) += bug.o + hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/bug.c b/lib/bug.c new file mode 100644 index 000000000000..014b582c5c4b --- /dev/null +++ b/lib/bug.c @@ -0,0 +1,163 @@ +/* + Generic support for BUG() + + This respects the following config options: + + CONFIG_BUG - emit BUG traps. Nothing happens without this. + CONFIG_GENERIC_BUG - enable this code. + CONFIG_DEBUG_BUGVERBOSE - emit full file+line information for each BUG + + CONFIG_BUG and CONFIG_DEBUG_BUGVERBOSE are potentially user-settable + (though they're generally always on). + + CONFIG_GENERIC_BUG is set by each architecture using this code. + + To use this, your architecture must: + + 1. Set up the config options: + - Enable CONFIG_GENERIC_BUG if CONFIG_BUG + + 2. Implement BUG (and optionally BUG_ON, WARN, WARN_ON) + - Define HAVE_ARCH_BUG + - Implement BUG() to generate a faulting instruction + - NOTE: struct bug_entry does not have "file" or "line" entries + when CONFIG_DEBUG_BUGVERBOSE is not enabled, so you must generate + the values accordingly. + + 3. Implement the trap + - In the illegal instruction trap handler (typically), verify + that the fault was in kernel mode, and call report_bug() + - report_bug() will return whether it was a false alarm, a warning, + or an actual bug. + - You must implement the is_valid_bugaddr(bugaddr) callback which + returns true if the eip is a real kernel address, and it points + to the expected BUG trap instruction. + + Jeremy Fitzhardinge 2006 + */ +#include +#include +#include + +extern const struct bug_entry __start___bug_table[], __stop___bug_table[]; + +#ifdef CONFIG_MODULES +static LIST_HEAD(module_bug_list); + +static const struct bug_entry *module_find_bug(unsigned long bugaddr) +{ + struct module *mod; + + list_for_each_entry(mod, &module_bug_list, bug_list) { + const struct bug_entry *bug = mod->bug_table; + unsigned i; + + for (i = 0; i < mod->num_bugs; ++i, ++bug) + if (bugaddr == bug->bug_addr) + return bug; + } + return NULL; +} + +int module_bug_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, + struct module *mod) +{ + char *secstrings; + unsigned int i; + + mod->bug_table = NULL; + mod->num_bugs = 0; + + /* Find the __bug_table section, if present */ + secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; + for (i = 1; i < hdr->e_shnum; i++) { + if (strcmp(secstrings+sechdrs[i].sh_name, "__bug_table")) + continue; + mod->bug_table = (void *) sechdrs[i].sh_addr; + mod->num_bugs = sechdrs[i].sh_size / sizeof(struct bug_entry); + break; + } + + /* + * Strictly speaking this should have a spinlock to protect against + * traversals, but since we only traverse on BUG()s, a spinlock + * could potentially lead to deadlock and thus be counter-productive. + */ + list_add(&mod->bug_list, &module_bug_list); + + return 0; +} + +void module_bug_cleanup(struct module *mod) +{ + list_del(&mod->bug_list); +} + +#else + +static inline const struct bug_entry *module_find_bug(unsigned long bugaddr) +{ + return NULL; +} +#endif + +const struct bug_entry *find_bug(unsigned long bugaddr) +{ + const struct bug_entry *bug; + + for (bug = __start___bug_table; bug < __stop___bug_table; ++bug) + if (bugaddr == bug->bug_addr) + return bug; + + return module_find_bug(bugaddr); +} + +enum bug_trap_type report_bug(unsigned long bugaddr) +{ + const struct bug_entry *bug; + const char *file; + unsigned line, warning; + + if (!is_valid_bugaddr(bugaddr)) + return BUG_TRAP_TYPE_NONE; + + bug = find_bug(bugaddr); + + printk(KERN_EMERG "------------[ cut here ]------------\n"); + + file = NULL; + line = 0; + warning = 0; + + if (bug) { +#ifdef CONFIG_DEBUG_BUGVERBOSE + file = bug->file; + line = bug->line; +#endif + warning = (bug->flags & BUGFLAG_WARNING) != 0; + } + + if (warning) { + /* this is a WARN_ON rather than BUG/BUG_ON */ + if (file) + printk(KERN_ERR "Badness at %s:%u\n", + file, line); + else + printk(KERN_ERR "Badness at %p " + "[verbose debug info unavailable]\n", + (void *)bugaddr); + + dump_stack(); + return BUG_TRAP_TYPE_WARN; + } + + if (file) + printk(KERN_CRIT "kernel BUG at %s:%u!\n", + file, line); + else + printk(KERN_CRIT "Kernel BUG at %p " + "[verbose debug info unavailable]\n", + (void *)bugaddr); + + return BUG_TRAP_TYPE_BUG; +} -- cgit v1.2.3 From a5cfc1ec58a07074dacb6aa8c79eff864c966d12 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:36:25 -0800 Subject: [PATCH] bit reverse library This patch provides two bit reverse functions and bit reverse table. - reverse the order of bits in a u32 value u8 bitrev8(u8 x); - reverse the order of bits in a u32 value u32 bitrev32(u32 x); - byte reverse table const u8 byte_rev_table[256]; Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bitrev.h | 15 ++++++++++++++ lib/Kconfig | 3 +++ lib/Makefile | 1 + lib/bitrev.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 73 insertions(+) create mode 100644 include/linux/bitrev.h create mode 100644 lib/bitrev.c (limited to 'include/linux') diff --git a/include/linux/bitrev.h b/include/linux/bitrev.h new file mode 100644 index 000000000000..05e540d6963a --- /dev/null +++ b/include/linux/bitrev.h @@ -0,0 +1,15 @@ +#ifndef _LINUX_BITREV_H +#define _LINUX_BITREV_H + +#include + +extern u8 const byte_rev_table[256]; + +static inline u8 bitrev8(u8 byte) +{ + return byte_rev_table[byte]; +} + +extern u32 bitrev32(u32 in); + +#endif /* _LINUX_BITREV_H */ diff --git a/lib/Kconfig b/lib/Kconfig index 734ce95a93d1..0d683559080e 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -4,6 +4,9 @@ menu "Library routines" +config BITREVERSE + tristate + config CRC_CCITT tristate "CRC-CCITT functions" help diff --git a/lib/Makefile b/lib/Makefile index 7ab486b883a8..6f675c6149a7 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -35,6 +35,7 @@ ifneq ($(CONFIG_HAVE_DEC_LOCK),y) lib-y += dec_and_lock.o endif +obj-$(CONFIG_BITREVERSE) += bitrev.o obj-$(CONFIG_CRC_CCITT) += crc-ccitt.o obj-$(CONFIG_CRC16) += crc16.o obj-$(CONFIG_CRC32) += crc32.o diff --git a/lib/bitrev.c b/lib/bitrev.c new file mode 100644 index 000000000000..f4e1c49c825a --- /dev/null +++ b/lib/bitrev.c @@ -0,0 +1,54 @@ +#include +#include +#include + +const u8 byte_rev_table[256] = { + 0x00, 0x80, 0x40, 0xc0, 0x20, 0xa0, 0x60, 0xe0, + 0x10, 0x90, 0x50, 0xd0, 0x30, 0xb0, 0x70, 0xf0, + 0x08, 0x88, 0x48, 0xc8, 0x28, 0xa8, 0x68, 0xe8, + 0x18, 0x98, 0x58, 0xd8, 0x38, 0xb8, 0x78, 0xf8, + 0x04, 0x84, 0x44, 0xc4, 0x24, 0xa4, 0x64, 0xe4, + 0x14, 0x94, 0x54, 0xd4, 0x34, 0xb4, 0x74, 0xf4, + 0x0c, 0x8c, 0x4c, 0xcc, 0x2c, 0xac, 0x6c, 0xec, + 0x1c, 0x9c, 0x5c, 0xdc, 0x3c, 0xbc, 0x7c, 0xfc, + 0x02, 0x82, 0x42, 0xc2, 0x22, 0xa2, 0x62, 0xe2, + 0x12, 0x92, 0x52, 0xd2, 0x32, 0xb2, 0x72, 0xf2, + 0x0a, 0x8a, 0x4a, 0xca, 0x2a, 0xaa, 0x6a, 0xea, + 0x1a, 0x9a, 0x5a, 0xda, 0x3a, 0xba, 0x7a, 0xfa, + 0x06, 0x86, 0x46, 0xc6, 0x26, 0xa6, 0x66, 0xe6, + 0x16, 0x96, 0x56, 0xd6, 0x36, 0xb6, 0x76, 0xf6, + 0x0e, 0x8e, 0x4e, 0xce, 0x2e, 0xae, 0x6e, 0xee, + 0x1e, 0x9e, 0x5e, 0xde, 0x3e, 0xbe, 0x7e, 0xfe, + 0x01, 0x81, 0x41, 0xc1, 0x21, 0xa1, 0x61, 0xe1, + 0x11, 0x91, 0x51, 0xd1, 0x31, 0xb1, 0x71, 0xf1, + 0x09, 0x89, 0x49, 0xc9, 0x29, 0xa9, 0x69, 0xe9, + 0x19, 0x99, 0x59, 0xd9, 0x39, 0xb9, 0x79, 0xf9, + 0x05, 0x85, 0x45, 0xc5, 0x25, 0xa5, 0x65, 0xe5, + 0x15, 0x95, 0x55, 0xd5, 0x35, 0xb5, 0x75, 0xf5, + 0x0d, 0x8d, 0x4d, 0xcd, 0x2d, 0xad, 0x6d, 0xed, + 0x1d, 0x9d, 0x5d, 0xdd, 0x3d, 0xbd, 0x7d, 0xfd, + 0x03, 0x83, 0x43, 0xc3, 0x23, 0xa3, 0x63, 0xe3, + 0x13, 0x93, 0x53, 0xd3, 0x33, 0xb3, 0x73, 0xf3, + 0x0b, 0x8b, 0x4b, 0xcb, 0x2b, 0xab, 0x6b, 0xeb, + 0x1b, 0x9b, 0x5b, 0xdb, 0x3b, 0xbb, 0x7b, 0xfb, + 0x07, 0x87, 0x47, 0xc7, 0x27, 0xa7, 0x67, 0xe7, + 0x17, 0x97, 0x57, 0xd7, 0x37, 0xb7, 0x77, 0xf7, + 0x0f, 0x8f, 0x4f, 0xcf, 0x2f, 0xaf, 0x6f, 0xef, + 0x1f, 0x9f, 0x5f, 0xdf, 0x3f, 0xbf, 0x7f, 0xff, +}; +EXPORT_SYMBOL_GPL(byte_rev_table); + +static __always_inline u16 bitrev16(u16 x) +{ + return (bitrev8(x & 0xff) << 8) | bitrev8(x >> 8); +} + +/** + * bitrev32 - reverse the order of bits in a u32 value + * @x: value to be bit-reversed + */ +u32 bitrev32(u32 x) +{ + return (bitrev16(x & 0xffff) << 16) | bitrev16(x >> 16); +} +EXPORT_SYMBOL(bitrev32); -- cgit v1.2.3 From 906d66df18faa4aac8d898ae6920d1014694a932 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:36:25 -0800 Subject: [PATCH] crc32: replace bitreverse by bitrev32 This patch replaces bitreverse() by bitrev32. The only users of bitreverse() are crc32 itself and via-velocity. Cc: Jeff Garzik Cc: Matt Domsch Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/net/via-velocity.c | 2 +- include/linux/crc32.h | 4 ++-- lib/Kconfig | 1 + lib/crc32.c | 28 +++++----------------------- 4 files changed, 9 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index 74f894795a1b..4587f23f4e4b 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -3132,7 +3132,7 @@ static u16 wol_calc_crc(int size, u8 * pattern, u8 *mask_pattern) } /* Finally, invert the result once to get the correct data */ crc = ~crc; - return bitreverse(crc) >> 16; + return bitrev32(crc) >> 16; } /** diff --git a/include/linux/crc32.h b/include/linux/crc32.h index 56c0645789a9..e20dd1f9b40a 100644 --- a/include/linux/crc32.h +++ b/include/linux/crc32.h @@ -6,10 +6,10 @@ #define _LINUX_CRC32_H #include +#include extern u32 crc32_le(u32 crc, unsigned char const *p, size_t len); extern u32 crc32_be(u32 crc, unsigned char const *p, size_t len); -extern u32 bitreverse(u32 in); #define crc32(seed, data, length) crc32_le(seed, (unsigned char const *)data, length) @@ -21,7 +21,7 @@ extern u32 bitreverse(u32 in); * is in bit nr 0], thus it must be reversed before use. Except for * nics that bit swap the result internally... */ -#define ether_crc(length, data) bitreverse(crc32_le(~0, data, length)) +#define ether_crc(length, data) bitrev32(crc32_le(~0, data, length)) #define ether_crc_le(length, data) crc32_le(~0, data, length) #endif /* _LINUX_CRC32_H */ diff --git a/lib/Kconfig b/lib/Kconfig index 0d683559080e..47b172df3e31 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -26,6 +26,7 @@ config CRC16 config CRC32 tristate "CRC32 functions" default y + select BITREVERSE help This option is provided for the case where no in-kernel-tree modules require CRC32 functions, but a module built outside the diff --git a/lib/crc32.c b/lib/crc32.c index 285fd9bc61be..bfc33314c720 100644 --- a/lib/crc32.c +++ b/lib/crc32.c @@ -235,23 +235,8 @@ u32 __attribute_pure__ crc32_be(u32 crc, unsigned char const *p, size_t len) } #endif -/** - * bitreverse - reverse the order of bits in a u32 value - * @x: value to be bit-reversed - */ -u32 bitreverse(u32 x) -{ - x = (x >> 16) | (x << 16); - x = (x >> 8 & 0x00ff00ff) | (x << 8 & 0xff00ff00); - x = (x >> 4 & 0x0f0f0f0f) | (x << 4 & 0xf0f0f0f0); - x = (x >> 2 & 0x33333333) | (x << 2 & 0xcccccccc); - x = (x >> 1 & 0x55555555) | (x << 1 & 0xaaaaaaaa); - return x; -} - EXPORT_SYMBOL(crc32_le); EXPORT_SYMBOL(crc32_be); -EXPORT_SYMBOL(bitreverse); /* * A brief CRC tutorial. @@ -400,10 +385,7 @@ buf_dump(char const *prefix, unsigned char const *buf, size_t len) static void bytereverse(unsigned char *buf, size_t len) { while (len--) { - unsigned char x = *buf; - x = (x >> 4) | (x << 4); - x = (x >> 2 & 0x33) | (x << 2 & 0xcc); - x = (x >> 1 & 0x55) | (x << 1 & 0xaa); + unsigned char x = bitrev8(*buf); *buf++ = x; } } @@ -460,11 +442,11 @@ static u32 test_step(u32 init, unsigned char *buf, size_t len) /* Now swap it around for the other test */ bytereverse(buf, len + 4); - init = bitreverse(init); - crc2 = bitreverse(crc1); - if (crc1 != bitreverse(crc2)) + init = bitrev32(init); + crc2 = bitrev32(crc1); + if (crc1 != bitrev32(crc2)) printf("\nBit reversal fail: 0x%08x -> 0x%08x -> 0x%08x\n", - crc1, crc2, bitreverse(crc2)); + crc1, crc2, bitrev32(crc2)); crc1 = crc32_le(init, buf, len); if (crc1 != crc2) printf("\nCRC endianness fail: 0x%08x != 0x%08x\n", crc1, -- cgit v1.2.3 From 42cf11939becc717bd125d121a1a23415106a099 Mon Sep 17 00:00:00 2001 From: Josef "Jeff" Sipek Date: Fri, 8 Dec 2006 02:36:31 -0800 Subject: [PATCH] fsstack: Introduce fsstack_copy_{attr,inode}_* Introduce several fsstack_copy_* functions which allow stackable filesystems (such as eCryptfs and Unionfs) to easily copy over (currently only) inode attributes. This prevents code duplication and allows for code reuse. [akpm@osdl.org: Remove unneeded wrapper] [bunk@stusta.de: fs/stack.c should #include ] Signed-off-by: Josef "Jeff" Sipek Cc: Michael Halcrow Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Makefile | 3 ++- fs/stack.c | 40 ++++++++++++++++++++++++++++++++++++++++ include/linux/fs_stack.h | 31 +++++++++++++++++++++++++++++++ 3 files changed, 73 insertions(+), 1 deletion(-) create mode 100644 fs/stack.c create mode 100644 include/linux/fs_stack.h (limited to 'include/linux') diff --git a/fs/Makefile b/fs/Makefile index 9a5ce9323bfd..b9ffa63f77fc 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -10,7 +10,8 @@ obj-y := open.o read_write.o file_table.o super.o \ ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \ attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \ seq_file.o xattr.o libfs.o fs-writeback.o \ - pnode.o drop_caches.o splice.o sync.o utimes.o + pnode.o drop_caches.o splice.o sync.o utimes.o \ + stack.o ifeq ($(CONFIG_BLOCK),y) obj-y += buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o diff --git a/fs/stack.c b/fs/stack.c new file mode 100644 index 000000000000..5ddbc34535f9 --- /dev/null +++ b/fs/stack.c @@ -0,0 +1,40 @@ +#include +#include +#include + +/* does _NOT_ require i_mutex to be held. + * + * This function cannot be inlined since i_size_{read,write} is rather + * heavy-weight on 32-bit systems + */ +void fsstack_copy_inode_size(struct inode *dst, const struct inode *src) +{ + i_size_write(dst, i_size_read((struct inode *)src)); + dst->i_blocks = src->i_blocks; +} +EXPORT_SYMBOL_GPL(fsstack_copy_inode_size); + +/* copy all attributes; get_nlinks is optional way to override the i_nlink + * copying + */ +void fsstack_copy_attr_all(struct inode *dest, const struct inode *src, + int (*get_nlinks)(struct inode *)) +{ + if (!get_nlinks) + dest->i_nlink = src->i_nlink; + else + dest->i_nlink = (*get_nlinks)(dest); + + dest->i_mode = src->i_mode; + dest->i_uid = src->i_uid; + dest->i_gid = src->i_gid; + dest->i_rdev = src->i_rdev; + dest->i_atime = src->i_atime; + dest->i_mtime = src->i_mtime; + dest->i_ctime = src->i_ctime; + dest->i_blkbits = src->i_blkbits; + dest->i_flags = src->i_flags; + + fsstack_copy_inode_size(dest, src); +} +EXPORT_SYMBOL_GPL(fsstack_copy_attr_all); diff --git a/include/linux/fs_stack.h b/include/linux/fs_stack.h new file mode 100644 index 000000000000..bb516ceeefc9 --- /dev/null +++ b/include/linux/fs_stack.h @@ -0,0 +1,31 @@ +#ifndef _LINUX_FS_STACK_H +#define _LINUX_FS_STACK_H + +/* This file defines generic functions used primarily by stackable + * filesystems; none of these functions require i_mutex to be held. + */ + +#include + +/* externs for fs/stack.c */ +extern void fsstack_copy_attr_all(struct inode *dest, const struct inode *src, + int (*get_nlinks)(struct inode *)); + +extern void fsstack_copy_inode_size(struct inode *dst, const struct inode *src); + +/* inlines */ +static inline void fsstack_copy_attr_atime(struct inode *dest, + const struct inode *src) +{ + dest->i_atime = src->i_atime; +} + +static inline void fsstack_copy_attr_times(struct inode *dest, + const struct inode *src) +{ + dest->i_atime = src->i_atime; + dest->i_mtime = src->i_mtime; + dest->i_ctime = src->i_ctime; +} + +#endif /* _LINUX_FS_STACK_H */ -- cgit v1.2.3 From fec6d055da71fb02a76f9c2c12427fa79974018b Mon Sep 17 00:00:00 2001 From: Josef "Jeff" Sipek Date: Fri, 8 Dec 2006 02:36:32 -0800 Subject: [PATCH] struct path: rename Reiserfs's struct path Rename Reiserfs's struct path to struct treepath to prevent name collision between it and struct path from fs/namei.c. Signed-off-by: Josef "Jeff" Sipek Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/bitmap.c | 2 +- fs/reiserfs/fix_node.c | 6 +++--- fs/reiserfs/inode.c | 12 ++++++------ fs/reiserfs/namei.c | 6 +++--- fs/reiserfs/stree.c | 42 ++++++++++++++++++++--------------------- fs/reiserfs/tail_conversion.c | 4 ++-- include/linux/reiserfs_fs.h | 44 +++++++++++++++++++++---------------------- 7 files changed, 58 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index e3d466a228d4..b286ccb08587 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -708,7 +708,7 @@ static void oid_groups(reiserfs_blocknr_hint_t * hint) */ static int get_left_neighbor(reiserfs_blocknr_hint_t * hint) { - struct path *path; + struct treepath *path; struct buffer_head *bh; struct item_head *ih; int pos_in_item; diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c index 6d0e554daa9d..0ee35c6c9b72 100644 --- a/fs/reiserfs/fix_node.c +++ b/fs/reiserfs/fix_node.c @@ -957,7 +957,7 @@ static int get_far_parent(struct tree_balance *p_s_tb, { struct buffer_head *p_s_parent; INITIALIZE_PATH(s_path_to_neighbor_father); - struct path *p_s_path = p_s_tb->tb_path; + struct treepath *p_s_path = p_s_tb->tb_path; struct cpu_key s_lr_father_key; int n_counter, n_position = INT_MAX, @@ -1074,7 +1074,7 @@ static int get_far_parent(struct tree_balance *p_s_tb, */ static int get_parents(struct tree_balance *p_s_tb, int n_h) { - struct path *p_s_path = p_s_tb->tb_path; + struct treepath *p_s_path = p_s_tb->tb_path; int n_position, n_ret_value, n_path_offset = PATH_H_PATH_OFFSET(p_s_tb->tb_path, n_h); @@ -1885,7 +1885,7 @@ static int check_balance(int mode, static int get_direct_parent(struct tree_balance *p_s_tb, int n_h) { struct buffer_head *p_s_bh; - struct path *p_s_path = p_s_tb->tb_path; + struct treepath *p_s_path = p_s_tb->tb_path; int n_position, n_path_offset = PATH_H_PATH_OFFSET(p_s_tb->tb_path, n_h); diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 254239e6f9e3..f3d1c4a77979 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -207,7 +207,7 @@ static int file_capable(struct inode *inode, long block) } /*static*/ int restart_transaction(struct reiserfs_transaction_handle *th, - struct inode *inode, struct path *path) + struct inode *inode, struct treepath *path) { struct super_block *s = th->t_super; int len = th->t_blocks_allocated; @@ -570,7 +570,7 @@ static inline int _allocate_block(struct reiserfs_transaction_handle *th, long block, struct inode *inode, b_blocknr_t * allocated_block_nr, - struct path *path, int flags) + struct treepath *path, int flags) { BUG_ON(!th->t_trans_id); @@ -1107,7 +1107,7 @@ static inline ulong to_fake_used_blocks(struct inode *inode, int sd_size) // // called by read_locked_inode -static void init_inode(struct inode *inode, struct path *path) +static void init_inode(struct inode *inode, struct treepath *path) { struct buffer_head *bh; struct item_head *ih; @@ -1284,7 +1284,7 @@ static void inode2sd_v1(void *sd, struct inode *inode, loff_t size) /* NOTE, you must prepare the buffer head before sending it here, ** and then log it after the call */ -static void update_stat_data(struct path *path, struct inode *inode, +static void update_stat_data(struct treepath *path, struct inode *inode, loff_t size) { struct buffer_head *bh; @@ -1653,7 +1653,7 @@ int reiserfs_write_inode(struct inode *inode, int do_sync) containing "." and ".." entries */ static int reiserfs_new_directory(struct reiserfs_transaction_handle *th, struct inode *inode, - struct item_head *ih, struct path *path, + struct item_head *ih, struct treepath *path, struct inode *dir) { struct super_block *sb = th->t_super; @@ -1712,7 +1712,7 @@ static int reiserfs_new_directory(struct reiserfs_transaction_handle *th, containing the body of symlink */ static int reiserfs_new_symlink(struct reiserfs_transaction_handle *th, struct inode *inode, /* Inode of symlink */ struct item_head *ih, - struct path *path, const char *symname, + struct treepath *path, const char *symname, int item_len) { struct super_block *sb = th->t_super; diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index abde1edc2235..23f5cd5bbf56 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -54,7 +54,7 @@ static int bin_search_in_dir_item(struct reiserfs_dir_entry *de, loff_t off) // comment? maybe something like set de to point to what the path points to? static inline void set_de_item_location(struct reiserfs_dir_entry *de, - struct path *path) + struct treepath *path) { de->de_bh = get_last_bh(path); de->de_ih = get_ih(path); @@ -113,7 +113,7 @@ entry position in the item /* The function is NOT SCHEDULE-SAFE! */ int search_by_entry_key(struct super_block *sb, const struct cpu_key *key, - struct path *path, struct reiserfs_dir_entry *de) + struct treepath *path, struct reiserfs_dir_entry *de) { int retval; @@ -282,7 +282,7 @@ static int linear_search_in_dir_item(struct cpu_key *key, // may return NAME_FOUND, NAME_FOUND_INVISIBLE, NAME_NOT_FOUND // FIXME: should add something like IOERROR static int reiserfs_find_entry(struct inode *dir, const char *name, int namelen, - struct path *path_to_entry, + struct treepath *path_to_entry, struct reiserfs_dir_entry *de) { struct cpu_key key_to_search; diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index 5240abe1a709..47e7027ea39f 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -244,7 +244,7 @@ static const struct reiserfs_key MAX_KEY = { of the path, and going upwards. We must check the path's validity at each step. If the key is not in the path, there is no delimiting key in the tree (buffer is first or last buffer in tree), and in this case we return a special key, either MIN_KEY or MAX_KEY. */ -static inline const struct reiserfs_key *get_lkey(const struct path +static inline const struct reiserfs_key *get_lkey(const struct treepath *p_s_chk_path, const struct super_block *p_s_sb) @@ -290,7 +290,7 @@ static inline const struct reiserfs_key *get_lkey(const struct path } /* Get delimiting key of the buffer at the path and its right neighbor. */ -inline const struct reiserfs_key *get_rkey(const struct path *p_s_chk_path, +inline const struct reiserfs_key *get_rkey(const struct treepath *p_s_chk_path, const struct super_block *p_s_sb) { int n_position, n_path_offset = p_s_chk_path->path_length; @@ -337,7 +337,7 @@ inline const struct reiserfs_key *get_rkey(const struct path *p_s_chk_path, the path. These delimiting keys are stored at least one level above that buffer in the tree. If the buffer is the first or last node in the tree order then one of the delimiting keys may be absent, and in this case get_lkey and get_rkey return a special key which is MIN_KEY or MAX_KEY. */ -static inline int key_in_buffer(struct path *p_s_chk_path, /* Path which should be checked. */ +static inline int key_in_buffer(struct treepath *p_s_chk_path, /* Path which should be checked. */ const struct cpu_key *p_s_key, /* Key which should be checked. */ struct super_block *p_s_sb /* Super block pointer. */ ) @@ -374,7 +374,7 @@ inline void decrement_bcount(struct buffer_head *p_s_bh) } /* Decrement b_count field of the all buffers in the path. */ -void decrement_counters_in_path(struct path *p_s_search_path) +void decrement_counters_in_path(struct treepath *p_s_search_path) { int n_path_offset = p_s_search_path->path_length; @@ -391,7 +391,7 @@ void decrement_counters_in_path(struct path *p_s_search_path) p_s_search_path->path_length = ILLEGAL_PATH_ELEMENT_OFFSET; } -int reiserfs_check_path(struct path *p) +int reiserfs_check_path(struct treepath *p) { RFALSE(p->path_length != ILLEGAL_PATH_ELEMENT_OFFSET, "path not properly relsed"); @@ -403,7 +403,7 @@ int reiserfs_check_path(struct path *p) ** ** only called from fix_nodes() */ -void pathrelse_and_restore(struct super_block *s, struct path *p_s_search_path) +void pathrelse_and_restore(struct super_block *s, struct treepath *p_s_search_path) { int n_path_offset = p_s_search_path->path_length; @@ -421,7 +421,7 @@ void pathrelse_and_restore(struct super_block *s, struct path *p_s_search_path) } /* Release all buffers in the path. */ -void pathrelse(struct path *p_s_search_path) +void pathrelse(struct treepath *p_s_search_path) { int n_path_offset = p_s_search_path->path_length; @@ -602,7 +602,7 @@ static void search_by_key_reada(struct super_block *s, correctness of the bottom of the path */ /* The function is NOT SCHEDULE-SAFE! */ int search_by_key(struct super_block *p_s_sb, const struct cpu_key *p_s_key, /* Key to search. */ - struct path *p_s_search_path, /* This structure was + struct treepath *p_s_search_path,/* This structure was allocated and initialized by the calling function. It is filled up @@ -813,7 +813,7 @@ int search_by_key(struct super_block *p_s_sb, const struct cpu_key *p_s_key, /* /* The function is NOT SCHEDULE-SAFE! */ int search_for_position_by_key(struct super_block *p_s_sb, /* Pointer to the super block. */ const struct cpu_key *p_cpu_key, /* Key to search (cpu variable) */ - struct path *p_s_search_path /* Filled up by this function. */ + struct treepath *p_s_search_path /* Filled up by this function. */ ) { struct item_head *p_le_ih; /* pointer to on-disk structure */ @@ -884,7 +884,7 @@ int search_for_position_by_key(struct super_block *p_s_sb, /* Pointer to the sup } /* Compare given item and item pointed to by the path. */ -int comp_items(const struct item_head *stored_ih, const struct path *p_s_path) +int comp_items(const struct item_head *stored_ih, const struct treepath *p_s_path) { struct buffer_head *p_s_bh; struct item_head *ih; @@ -911,7 +911,7 @@ int comp_items(const struct item_head *stored_ih, const struct path *p_s_path) #define block_in_use(bh) (buffer_locked(bh) || (held_by_others(bh))) // prepare for delete or cut of direct item -static inline int prepare_for_direct_item(struct path *path, +static inline int prepare_for_direct_item(struct treepath *path, struct item_head *le_ih, struct inode *inode, loff_t new_file_length, int *cut_size) @@ -952,7 +952,7 @@ static inline int prepare_for_direct_item(struct path *path, return M_CUT; /* Cut from this item. */ } -static inline int prepare_for_direntry_item(struct path *path, +static inline int prepare_for_direntry_item(struct treepath *path, struct item_head *le_ih, struct inode *inode, loff_t new_file_length, @@ -987,7 +987,7 @@ static inline int prepare_for_direntry_item(struct path *path, In case of file truncate calculate whether this item must be deleted/truncated or last unformatted node of this item will be converted to a direct item. This function returns a determination of what balance mode the calling function should employ. */ -static char prepare_for_delete_or_cut(struct reiserfs_transaction_handle *th, struct inode *inode, struct path *p_s_path, const struct cpu_key *p_s_item_key, int *p_n_removed, /* Number of unformatted nodes which were removed +static char prepare_for_delete_or_cut(struct reiserfs_transaction_handle *th, struct inode *inode, struct treepath *p_s_path, const struct cpu_key *p_s_item_key, int *p_n_removed, /* Number of unformatted nodes which were removed from end of the file. */ int *p_n_cut_size, unsigned long long n_new_file_length /* MAX_KEY_OFFSET in case of delete. */ ) @@ -1125,7 +1125,7 @@ static int calc_deleted_bytes_number(struct tree_balance *p_s_tb, char c_mode) static void init_tb_struct(struct reiserfs_transaction_handle *th, struct tree_balance *p_s_tb, struct super_block *p_s_sb, - struct path *p_s_path, int n_size) + struct treepath *p_s_path, int n_size) { BUG_ON(!th->t_trans_id); @@ -1176,7 +1176,7 @@ char head2type(struct item_head *ih) #endif /* Delete object item. */ -int reiserfs_delete_item(struct reiserfs_transaction_handle *th, struct path *p_s_path, /* Path to the deleted item. */ +int reiserfs_delete_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_path, /* Path to the deleted item. */ const struct cpu_key *p_s_item_key, /* Key to search for the deleted item. */ struct inode *p_s_inode, /* inode is here just to update i_blocks and quotas */ struct buffer_head *p_s_un_bh) @@ -1468,7 +1468,7 @@ static void unmap_buffers(struct page *page, loff_t pos) static int maybe_indirect_to_direct(struct reiserfs_transaction_handle *th, struct inode *p_s_inode, struct page *page, - struct path *p_s_path, + struct treepath *p_s_path, const struct cpu_key *p_s_item_key, loff_t n_new_file_size, char *p_c_mode) { @@ -1503,7 +1503,7 @@ static int maybe_indirect_to_direct(struct reiserfs_transaction_handle *th, pointer being converted. Therefore we have to delete inserted direct item(s) */ static void indirect_to_direct_roll_back(struct reiserfs_transaction_handle *th, - struct inode *inode, struct path *path) + struct inode *inode, struct treepath *path) { struct cpu_key tail_key; int tail_len; @@ -1545,7 +1545,7 @@ static void indirect_to_direct_roll_back(struct reiserfs_transaction_handle *th, /* (Truncate or cut entry) or delete object item. Returns < 0 on failure */ int reiserfs_cut_from_item(struct reiserfs_transaction_handle *th, - struct path *p_s_path, + struct treepath *p_s_path, struct cpu_key *p_s_item_key, struct inode *p_s_inode, struct page *page, loff_t n_new_file_size) @@ -1920,7 +1920,7 @@ int reiserfs_do_truncate(struct reiserfs_transaction_handle *th, struct inode *p #ifdef CONFIG_REISERFS_CHECK // this makes sure, that we __append__, not overwrite or add holes -static void check_research_for_paste(struct path *path, +static void check_research_for_paste(struct treepath *path, const struct cpu_key *p_s_key) { struct item_head *found_ih = get_ih(path); @@ -1954,7 +1954,7 @@ static void check_research_for_paste(struct path *path, #endif /* config reiserfs check */ /* Paste bytes to the existing item. Returns bytes number pasted into the item. */ -int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, struct path *p_s_search_path, /* Path to the pasted item. */ +int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_search_path, /* Path to the pasted item. */ const struct cpu_key *p_s_key, /* Key to search for the needed item. */ struct inode *inode, /* Inode item belongs to */ const char *p_c_body, /* Pointer to the bytes to paste. */ @@ -2036,7 +2036,7 @@ int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, struct path } /* Insert new item into the buffer at the path. */ -int reiserfs_insert_item(struct reiserfs_transaction_handle *th, struct path *p_s_path, /* Path to the inserteded item. */ +int reiserfs_insert_item(struct reiserfs_transaction_handle *th, struct treepath *p_s_path, /* Path to the inserteded item. */ const struct cpu_key *key, struct item_head *p_s_ih, /* Pointer to the item header to insert. */ struct inode *inode, const char *p_c_body) { /* Pointer to the bytes to insert. */ diff --git a/fs/reiserfs/tail_conversion.c b/fs/reiserfs/tail_conversion.c index 36f108fc1cf5..f8121a1147e8 100644 --- a/fs/reiserfs/tail_conversion.c +++ b/fs/reiserfs/tail_conversion.c @@ -15,7 +15,7 @@ /* path points to first direct item of the file regarless of how many of them are there */ int direct2indirect(struct reiserfs_transaction_handle *th, struct inode *inode, - struct path *path, struct buffer_head *unbh, + struct treepath *path, struct buffer_head *unbh, loff_t tail_offset) { struct super_block *sb = inode->i_sb; @@ -171,7 +171,7 @@ void reiserfs_unmap_buffer(struct buffer_head *bh) what we expect from it (number of cut bytes). But when tail remains in the unformatted node, we set mode to SKIP_BALANCING and unlock inode */ -int indirect2direct(struct reiserfs_transaction_handle *th, struct inode *p_s_inode, struct page *page, struct path *p_s_path, /* path to the indirect item. */ +int indirect2direct(struct reiserfs_transaction_handle *th, struct inode *p_s_inode, struct page *page, struct treepath *p_s_path, /* path to the indirect item. */ const struct cpu_key *p_s_item_key, /* Key to look for unformatted node pointer to be cut. */ loff_t n_new_file_size, /* New file size. */ char *p_c_mode) diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index d0e4dce33ad5..c3fc6caaad3f 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1159,7 +1159,7 @@ znodes are the way! */ #define PATH_READA 0x1 /* do read ahead */ #define PATH_READA_BACK 0x2 /* read backwards */ -struct path { +struct treepath { int path_length; /* Length of the array above. */ int reada; struct path_element path_elements[EXTENDED_MAX_HEIGHT]; /* Array of the path elements. */ @@ -1169,7 +1169,7 @@ struct path { #define pos_in_item(path) ((path)->pos_in_item) #define INITIALIZE_PATH(var) \ -struct path var = {.path_length = ILLEGAL_PATH_ELEMENT_OFFSET, .reada = 0,} +struct treepath var = {.path_length = ILLEGAL_PATH_ELEMENT_OFFSET, .reada = 0,} /* Get path element by path and path position. */ #define PATH_OFFSET_PELEMENT(p_s_path,n_offset) ((p_s_path)->path_elements +(n_offset)) @@ -1327,7 +1327,7 @@ struct tree_balance { int need_balance_dirty; struct super_block *tb_sb; struct reiserfs_transaction_handle *transaction_handle; - struct path *tb_path; + struct treepath *tb_path; struct buffer_head *L[MAX_HEIGHT]; /* array of left neighbors of nodes in the path */ struct buffer_head *R[MAX_HEIGHT]; /* array of right neighbors of nodes in the path */ struct buffer_head *FL[MAX_HEIGHT]; /* array of fathers of the left neighbors */ @@ -1793,41 +1793,41 @@ static inline void copy_key(struct reiserfs_key *to, memcpy(to, from, KEY_SIZE); } -int comp_items(const struct item_head *stored_ih, const struct path *p_s_path); -const struct reiserfs_key *get_rkey(const struct path *p_s_chk_path, +int comp_items(const struct item_head *stored_ih, const struct treepath *p_s_path); +const struct reiserfs_key *get_rkey(const struct treepath *p_s_chk_path, const struct super_block *p_s_sb); int search_by_key(struct super_block *, const struct cpu_key *, - struct path *, int); + struct treepath *, int); #define search_item(s,key,path) search_by_key (s, key, path, DISK_LEAF_NODE_LEVEL) int search_for_position_by_key(struct super_block *p_s_sb, const struct cpu_key *p_s_cpu_key, - struct path *p_s_search_path); + struct treepath *p_s_search_path); extern void decrement_bcount(struct buffer_head *p_s_bh); -void decrement_counters_in_path(struct path *p_s_search_path); -void pathrelse(struct path *p_s_search_path); -int reiserfs_check_path(struct path *p); -void pathrelse_and_restore(struct super_block *s, struct path *p_s_search_path); +void decrement_counters_in_path(struct treepath *p_s_search_path); +void pathrelse(struct treepath *p_s_search_path); +int reiserfs_check_path(struct treepath *p); +void pathrelse_and_restore(struct super_block *s, struct treepath *p_s_search_path); int reiserfs_insert_item(struct reiserfs_transaction_handle *th, - struct path *path, + struct treepath *path, const struct cpu_key *key, struct item_head *ih, struct inode *inode, const char *body); int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, - struct path *path, + struct treepath *path, const struct cpu_key *key, struct inode *inode, const char *body, int paste_size); int reiserfs_cut_from_item(struct reiserfs_transaction_handle *th, - struct path *path, + struct treepath *path, struct cpu_key *key, struct inode *inode, struct page *page, loff_t new_file_size); int reiserfs_delete_item(struct reiserfs_transaction_handle *th, - struct path *path, + struct treepath *path, const struct cpu_key *key, struct inode *inode, struct buffer_head *p_s_un_bh); @@ -1858,7 +1858,7 @@ void padd_item(char *item, int total_length, int length); #define GET_BLOCK_NO_DANGLE 16 /* don't leave any transactions running */ int restart_transaction(struct reiserfs_transaction_handle *th, - struct inode *inode, struct path *path); + struct inode *inode, struct treepath *path); void reiserfs_read_locked_inode(struct inode *inode, struct reiserfs_iget_args *args); int reiserfs_find_actor(struct inode *inode, void *p); @@ -1905,7 +1905,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr); /* namei.c */ void set_de_name_and_namelen(struct reiserfs_dir_entry *de); int search_by_entry_key(struct super_block *sb, const struct cpu_key *key, - struct path *path, struct reiserfs_dir_entry *de); + struct treepath *path, struct reiserfs_dir_entry *de); struct dentry *reiserfs_get_parent(struct dentry *); /* procfs.c */ @@ -1956,9 +1956,9 @@ extern const struct file_operations reiserfs_dir_operations; /* tail_conversion.c */ int direct2indirect(struct reiserfs_transaction_handle *, struct inode *, - struct path *, struct buffer_head *, loff_t); + struct treepath *, struct buffer_head *, loff_t); int indirect2direct(struct reiserfs_transaction_handle *, struct inode *, - struct page *, struct path *, const struct cpu_key *, + struct page *, struct treepath *, const struct cpu_key *, loff_t, char *); void reiserfs_unmap_buffer(struct buffer_head *); @@ -2045,7 +2045,7 @@ struct __reiserfs_blocknr_hint { struct inode *inode; /* inode passed to allocator, if we allocate unf. nodes */ long block; /* file offset, in blocks */ struct in_core_key key; - struct path *path; /* search path, used by allocator to deternine search_start by + struct treepath *path; /* search path, used by allocator to deternine search_start by * various ways */ struct reiserfs_transaction_handle *th; /* transaction handle is needed to log super blocks and * bitmap blocks changes */ @@ -2101,7 +2101,7 @@ static inline int reiserfs_new_form_blocknrs(struct tree_balance *tb, static inline int reiserfs_new_unf_blocknrs(struct reiserfs_transaction_handle *th, struct inode *inode, b_blocknr_t * new_blocknrs, - struct path *path, long block) + struct treepath *path, long block) { reiserfs_blocknr_hint_t hint = { .th = th, @@ -2118,7 +2118,7 @@ static inline int reiserfs_new_unf_blocknrs(struct reiserfs_transaction_handle static inline int reiserfs_new_unf_blocknrs2(struct reiserfs_transaction_handle *th, struct inode *inode, b_blocknr_t * new_blocknrs, - struct path *path, long block) + struct treepath *path, long block) { reiserfs_blocknr_hint_t hint = { .th = th, -- cgit v1.2.3 From 346f20ff6020ffa11d40b789069079c56a444ae6 Mon Sep 17 00:00:00 2001 From: Josef "Jeff" Sipek Date: Fri, 8 Dec 2006 02:36:34 -0800 Subject: [PATCH] struct path: move struct path from fs/namei.c into include/linux Moved struct path from fs/namei.c to include/linux/namei.h. This allows many places in the VFS, as well as any stackable filesystem to easily keep track of dentry-vfsmount pairs. Signed-off-by: Josef "Jeff" Sipek Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/namei.c | 5 ----- include/linux/namei.h | 5 +++++ 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/namei.c b/fs/namei.c index db1bca26d88c..8c2db88bb20d 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -572,11 +572,6 @@ fail: return PTR_ERR(link); } -struct path { - struct vfsmount *mnt; - struct dentry *dentry; -}; - static inline void dput_path(struct path *path, struct nameidata *nd) { dput(path->dentry); diff --git a/include/linux/namei.h b/include/linux/namei.h index f5f19606effb..d39a5a67e979 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -29,6 +29,11 @@ struct nameidata { } intent; }; +struct path { + struct vfsmount *mnt; + struct dentry *dentry; +}; + /* * Type of the last component on LOOKUP_PARENT */ -- cgit v1.2.3 From 0f7fc9e4d03987fe29f6dd4aa67e4c56eb7ecb05 Mon Sep 17 00:00:00 2001 From: Josef "Jeff" Sipek Date: Fri, 8 Dec 2006 02:36:35 -0800 Subject: [PATCH] VFS: change struct file to use struct path This patch changes struct file to use struct path instead of having independent pointers to struct dentry and struct vfsmount, and converts all users of f_{dentry,vfsmnt} in fs/ to use f_path.{dentry,mnt}. Additionally, it adds two #define's to make the transition easier for users of the f_dentry and f_vfsmnt. Signed-off-by: Josef "Jeff" Sipek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_aout.c | 8 ++++---- fs/binfmt_elf.c | 2 +- fs/binfmt_elf_fdpic.c | 4 ++-- fs/binfmt_flat.c | 2 +- fs/binfmt_misc.c | 10 +++++----- fs/block_dev.c | 4 ++-- fs/compat.c | 12 ++++++------ fs/compat_ioctl.c | 2 +- fs/dnotify.c | 4 ++-- fs/dquot.c | 4 ++-- fs/eventpoll.c | 4 ++-- fs/exec.c | 10 +++++----- fs/fcntl.c | 2 +- fs/file_table.c | 10 +++++----- fs/inode.c | 2 +- fs/inotify_user.c | 6 +++--- fs/ioctl.c | 14 +++++++------- fs/libfs.c | 14 +++++++------- fs/locks.c | 32 ++++++++++++++++---------------- fs/namei.c | 10 +++++----- fs/open.c | 26 +++++++++++++------------- fs/pipe.c | 28 ++++++++++++++-------------- fs/read_write.c | 20 ++++++++++---------- fs/readdir.c | 2 +- fs/seq_file.c | 2 +- fs/splice.c | 18 +++++++++--------- fs/stat.c | 2 +- fs/super.c | 2 +- fs/sync.c | 4 ++-- fs/xattr.c | 8 ++++---- include/linux/fs.h | 10 ++++++---- include/linux/fsnotify.h | 2 +- 32 files changed, 141 insertions(+), 139 deletions(-) (limited to 'include/linux') diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index 517e111bb7ef..813a887cd2b3 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -274,7 +274,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) if ((N_MAGIC(ex) != ZMAGIC && N_MAGIC(ex) != OMAGIC && N_MAGIC(ex) != QMAGIC && N_MAGIC(ex) != NMAGIC) || N_TRSIZE(ex) || N_DRSIZE(ex) || - i_size_read(bprm->file->f_dentry->d_inode) < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) { + i_size_read(bprm->file->f_path.dentry->d_inode) < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) { return -ENOEXEC; } @@ -389,7 +389,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) { printk(KERN_WARNING "fd_offset is not page aligned. Please convert program: %s\n", - bprm->file->f_dentry->d_name.name); + bprm->file->f_path.dentry->d_name.name); error_time = jiffies; } @@ -469,7 +469,7 @@ static int load_aout_library(struct file *file) int retval; struct exec ex; - inode = file->f_dentry->d_inode; + inode = file->f_path.dentry->d_inode; retval = -ENOEXEC; error = kernel_read(file, 0, (char *) &ex, sizeof(ex)); @@ -506,7 +506,7 @@ static int load_aout_library(struct file *file) { printk(KERN_WARNING "N_TXTOFF is not page aligned. Please convert library: %s\n", - file->f_dentry->d_name.name); + file->f_path.dentry->d_name.name); error_time = jiffies; } down_write(¤t->mm->mmap_sem); diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index be5869d34999..c6dbb4a7ec78 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1190,7 +1190,7 @@ static int maydump(struct vm_area_struct *vma) /* Dump shared memory only if mapped from an anonymous file. */ if (vma->vm_flags & VM_SHARED) - return vma->vm_file->f_dentry->d_inode->i_nlink == 0; + return vma->vm_file->f_path.dentry->d_inode->i_nlink == 0; /* If it hasn't been written to, don't write it out */ if (!vma->anon_vma) diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index ed9a61c6beb3..9f0b7efc3df5 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -855,7 +855,7 @@ static int elf_fdpic_map_file(struct elf_fdpic_params *params, dynamic_error: printk("ELF FDPIC %s with invalid DYNAMIC section (inode=%lu)\n", - what, file->f_dentry->d_inode->i_ino); + what, file->f_path.dentry->d_inode->i_ino); return -ELIBBAD; } @@ -1186,7 +1186,7 @@ static int maydump(struct vm_area_struct *vma) /* Dump shared memory only if mapped from an anonymous file. */ if (vma->vm_flags & VM_SHARED) { - if (vma->vm_file->f_dentry->d_inode->i_nlink == 0) { + if (vma->vm_file->f_path.dentry->d_inode->i_nlink == 0) { kdcore("%08lx: %08lx: no (share)", vma->vm_start, vma->vm_flags); return 1; } diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index a62fd4018a20..ae8595d49856 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -429,7 +429,7 @@ static int load_flat_file(struct linux_binprm * bprm, int ret; hdr = ((struct flat_hdr *) bprm->buf); /* exec-header */ - inode = bprm->file->f_dentry->d_inode; + inode = bprm->file->f_path.dentry->d_inode; text_len = ntohl(hdr->data_start); data_len = ntohl(hdr->data_end) - ntohl(hdr->data_start); diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 1713c48fef54..00687ea62738 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -542,7 +542,7 @@ static void kill_node(Node *e) static ssize_t bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos) { - Node *e = file->f_dentry->d_inode->i_private; + Node *e = file->f_path.dentry->d_inode->i_private; loff_t pos = *ppos; ssize_t res; char *page; @@ -576,7 +576,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { struct dentry *root; - Node *e = file->f_dentry->d_inode->i_private; + Node *e = file->f_path.dentry->d_inode->i_private; int res = parse_command(buffer, count); switch (res) { @@ -584,7 +584,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, break; case 2: set_bit(Enabled, &e->flags); break; - case 3: root = dget(file->f_vfsmnt->mnt_sb->s_root); + case 3: root = dget(file->f_path.mnt->mnt_sb->s_root); mutex_lock(&root->d_inode->i_mutex); kill_node(e); @@ -610,7 +610,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, Node *e; struct inode *inode; struct dentry *root, *dentry; - struct super_block *sb = file->f_vfsmnt->mnt_sb; + struct super_block *sb = file->f_path.mnt->mnt_sb; int err = 0; e = create_entry(buffer, count); @@ -699,7 +699,7 @@ static ssize_t bm_status_write(struct file * file, const char __user * buffer, switch (res) { case 1: enabled = 0; break; case 2: enabled = 1; break; - case 3: root = dget(file->f_vfsmnt->mnt_sb->s_root); + case 3: root = dget(file->f_path.mnt->mnt_sb->s_root); mutex_lock(&root->d_inode->i_mutex); while (!list_empty(&entries)) diff --git a/fs/block_dev.c b/fs/block_dev.c index f3c3a44dd8b6..197f93921847 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -190,7 +190,7 @@ static int blkdev_commit_write(struct file *file, struct page *page, unsigned fr /* * private llseek: - * for a block special file file->f_dentry->d_inode->i_size is zero + * for a block special file file->f_path.dentry->d_inode->i_size is zero * so we compute the size by hand (just as in block_read/write above) */ static loff_t block_llseek(struct file *file, loff_t offset, int origin) @@ -1013,7 +1013,7 @@ static int __blkdev_get(struct block_device *bdev, mode_t mode, unsigned flags, struct dentry fake_dentry = {}; fake_file.f_mode = mode; fake_file.f_flags = flags; - fake_file.f_dentry = &fake_dentry; + fake_file.f_path.dentry = &fake_dentry; fake_dentry.d_inode = bdev->bd_inode; return do_open(bdev, &fake_file, for_part); diff --git a/fs/compat.c b/fs/compat.c index a7e3f162fb15..b766964a625c 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -232,7 +232,7 @@ asmlinkage long compat_sys_fstatfs(unsigned int fd, struct compat_statfs __user file = fget(fd); if (!file) goto out; - error = vfs_statfs(file->f_dentry, &tmp); + error = vfs_statfs(file->f_path.dentry, &tmp); if (!error) error = put_compat_statfs(buf, &tmp); fput(file); @@ -303,7 +303,7 @@ asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct c file = fget(fd); if (!file) goto out; - error = vfs_statfs(file->f_dentry, &tmp); + error = vfs_statfs(file->f_path.dentry, &tmp); if (!error) error = put_compat_statfs64(buf, &tmp); fput(file); @@ -365,7 +365,7 @@ static void compat_ioctl_error(struct file *filp, unsigned int fd, /* find the name of the device. */ path = (char *)__get_free_page(GFP_KERNEL); if (path) { - fn = d_path(filp->f_dentry, filp->f_vfsmnt, path, PAGE_SIZE); + fn = d_path(filp->f_path.dentry, filp->f_path.mnt, path, PAGE_SIZE); if (IS_ERR(fn)) fn = "?"; } @@ -416,7 +416,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, case FIBMAP: case FIGETBSZ: case FIONREAD: - if (S_ISREG(filp->f_dentry->d_inode->i_mode)) + if (S_ISREG(filp->f_path.dentry->d_inode->i_mode)) break; /*FALL THROUGH*/ @@ -438,7 +438,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, goto found_handler; } - if (S_ISSOCK(filp->f_dentry->d_inode->i_mode) && + if (S_ISSOCK(filp->f_path.dentry->d_inode->i_mode) && cmd >= SIOCDEVPRIVATE && cmd <= (SIOCDEVPRIVATE + 15)) { error = siocdevprivate_ioctl(fd, cmd, arg); } else { @@ -1259,7 +1259,7 @@ out: if (iov != iovstack) kfree(iov); if ((ret + (type == READ)) > 0) { - struct dentry *dentry = file->f_dentry; + struct dentry *dentry = file->f_path.dentry; if (type == READ) fsnotify_access(dentry); else diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index bcc3caf5d820..c81c958b3e1d 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1177,7 +1177,7 @@ static int cdrom_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long ar static int vt_check(struct file *file) { struct tty_struct *tty; - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file->f_path.dentry->d_inode; if (file->f_op->ioctl != tty_ioctl) return -EINVAL; diff --git a/fs/dnotify.c b/fs/dnotify.c index 1f26a2b9eee1..936409fcd939 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -42,7 +42,7 @@ void dnotify_flush(struct file *filp, fl_owner_t id) struct dnotify_struct **prev; struct inode *inode; - inode = filp->f_dentry->d_inode; + inode = filp->f_path.dentry->d_inode; if (!S_ISDIR(inode->i_mode)) return; spin_lock(&inode->i_lock); @@ -74,7 +74,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) } if (!dir_notify_enable) return -EINVAL; - inode = filp->f_dentry->d_inode; + inode = filp->f_path.dentry->d_inode; if (!S_ISDIR(inode->i_mode)) return -ENOTDIR; dn = kmem_cache_alloc(dn_cache, GFP_KERNEL); diff --git a/fs/dquot.c b/fs/dquot.c index 89066b19124d..0952cc474d9a 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -694,9 +694,9 @@ restart: file_list_lock(); list_for_each(p, &sb->s_files) { struct file *filp = list_entry(p, struct file, f_u.fu_list); - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; if (filp->f_mode & FMODE_WRITE && dqinit_needed(inode, type)) { - struct dentry *dentry = dget(filp->f_dentry); + struct dentry *dentry = dget(filp->f_path.dentry); file_list_unlock(); sb->dq_op->initialize(inode, type); dput(dentry); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 88a6f8d0b88e..3ae644e7e860 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -795,8 +795,8 @@ static int ep_getfd(int *efd, struct inode **einode, struct file **efile, goto eexit_4; dentry->d_op = &eventpollfs_dentry_operations; d_add(dentry, inode); - file->f_vfsmnt = mntget(eventpoll_mnt); - file->f_dentry = dentry; + file->f_path.mnt = mntget(eventpoll_mnt); + file->f_path.dentry = dentry; file->f_mapping = inode->i_mapping; file->f_pos = 0; diff --git a/fs/exec.c b/fs/exec.c index add0e03c3ea9..60433e2254a4 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -912,7 +912,7 @@ EXPORT_SYMBOL(flush_old_exec); int prepare_binprm(struct linux_binprm *bprm) { int mode; - struct inode * inode = bprm->file->f_dentry->d_inode; + struct inode * inode = bprm->file->f_path.dentry->d_inode; int retval; mode = inode->i_mode; @@ -922,7 +922,7 @@ int prepare_binprm(struct linux_binprm *bprm) bprm->e_uid = current->euid; bprm->e_gid = current->egid; - if(!(bprm->file->f_vfsmnt->mnt_flags & MNT_NOSUID)) { + if(!(bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID)) { /* Set-uid? */ if (mode & S_ISUID) { current->personality &= ~PER_CLEAR_ON_SETID; @@ -1519,10 +1519,10 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) 0600); if (IS_ERR(file)) goto fail_unlock; - inode = file->f_dentry->d_inode; + inode = file->f_path.dentry->d_inode; if (inode->i_nlink > 1) goto close_fail; /* multiple links - don't dump */ - if (!ispipe && d_unhashed(file->f_dentry)) + if (!ispipe && d_unhashed(file->f_path.dentry)) goto close_fail; /* AK: actually i see no reason to not allow this for named pipes etc., @@ -1533,7 +1533,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) goto close_fail; if (!file->f_op->write) goto close_fail; - if (!ispipe && do_truncate(file->f_dentry, 0, 0, file) != 0) + if (!ispipe && do_truncate(file->f_path.dentry, 0, 0, file) != 0) goto close_fail; retval = binfmt->core_dump(signr, regs, file); diff --git a/fs/fcntl.c b/fs/fcntl.c index 4740d35e52cd..2bdaef35da54 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -204,7 +204,7 @@ asmlinkage long sys_dup(unsigned int fildes) static int setfl(int fd, struct file * filp, unsigned long arg) { - struct inode * inode = filp->f_dentry->d_inode; + struct inode * inode = filp->f_path.dentry->d_inode; int error = 0; /* diff --git a/fs/file_table.c b/fs/file_table.c index 24f25a057d9c..4c17a18d8c10 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -152,8 +152,8 @@ EXPORT_SYMBOL(fput); */ void fastcall __fput(struct file *file) { - struct dentry *dentry = file->f_dentry; - struct vfsmount *mnt = file->f_vfsmnt; + struct dentry *dentry = file->f_path.dentry; + struct vfsmount *mnt = file->f_path.mnt; struct inode *inode = dentry->d_inode; might_sleep(); @@ -176,8 +176,8 @@ void fastcall __fput(struct file *file) put_write_access(inode); put_pid(file->f_owner.pid); file_kill(file); - file->f_dentry = NULL; - file->f_vfsmnt = NULL; + file->f_path.dentry = NULL; + file->f_path.mnt = NULL; file_free(file); dput(dentry); mntput(mnt); @@ -271,7 +271,7 @@ int fs_may_remount_ro(struct super_block *sb) file_list_lock(); list_for_each(p, &sb->s_files) { struct file *file = list_entry(p, struct file, f_u.fu_list); - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file->f_path.dentry->d_inode; /* File with pending delete? */ if (inode->i_nlink == 0) diff --git a/fs/inode.c b/fs/inode.c index 9ecccab7326d..d00de182ecb9 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1200,7 +1200,7 @@ EXPORT_SYMBOL(touch_atime); void file_update_time(struct file *file) { - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file->f_path.dentry->d_inode; struct timespec now; int sync_it = 0; diff --git a/fs/inotify_user.c b/fs/inotify_user.c index e1956e6f116c..55f6da55b7c0 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -570,9 +570,9 @@ asmlinkage long sys_inotify_init(void) dev->ih = ih; filp->f_op = &inotify_fops; - filp->f_vfsmnt = mntget(inotify_mnt); - filp->f_dentry = dget(inotify_mnt->mnt_root); - filp->f_mapping = filp->f_dentry->d_inode->i_mapping; + filp->f_path.mnt = mntget(inotify_mnt); + filp->f_path.dentry = dget(inotify_mnt->mnt_root); + filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping; filp->f_mode = FMODE_READ; filp->f_flags = O_RDONLY; filp->private_data = dev; diff --git a/fs/ioctl.c b/fs/ioctl.c index 4b7660b09ac0..ff61772ceedd 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -31,7 +31,7 @@ static long do_ioctl(struct file *filp, unsigned int cmd, goto out; } else if (filp->f_op->ioctl) { lock_kernel(); - error = filp->f_op->ioctl(filp->f_dentry->d_inode, + error = filp->f_op->ioctl(filp->f_path.dentry->d_inode, filp, cmd, arg); unlock_kernel(); } @@ -45,7 +45,7 @@ static int file_ioctl(struct file *filp, unsigned int cmd, { int error; int block; - struct inode * inode = filp->f_dentry->d_inode; + struct inode * inode = filp->f_path.dentry->d_inode; int __user *p = (int __user *)arg; switch (cmd) { @@ -137,17 +137,17 @@ int vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, unsigned lon break; case FIOQSIZE: - if (S_ISDIR(filp->f_dentry->d_inode->i_mode) || - S_ISREG(filp->f_dentry->d_inode->i_mode) || - S_ISLNK(filp->f_dentry->d_inode->i_mode)) { - loff_t res = inode_get_bytes(filp->f_dentry->d_inode); + if (S_ISDIR(filp->f_path.dentry->d_inode->i_mode) || + S_ISREG(filp->f_path.dentry->d_inode->i_mode) || + S_ISLNK(filp->f_path.dentry->d_inode->i_mode)) { + loff_t res = inode_get_bytes(filp->f_path.dentry->d_inode); error = copy_to_user((loff_t __user *)arg, &res, sizeof(res)) ? -EFAULT : 0; } else error = -ENOTTY; break; default: - if (S_ISREG(filp->f_dentry->d_inode->i_mode)) + if (S_ISREG(filp->f_path.dentry->d_inode->i_mode)) error = file_ioctl(filp, cmd, arg); else error = do_ioctl(filp, cmd, arg); diff --git a/fs/libfs.c b/fs/libfs.c index bd08e0e64a8c..503898d5c4a7 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -63,7 +63,7 @@ int dcache_dir_open(struct inode *inode, struct file *file) { static struct qstr cursor_name = {.len = 1, .name = "."}; - file->private_data = d_alloc(file->f_dentry, &cursor_name); + file->private_data = d_alloc(file->f_path.dentry, &cursor_name); return file->private_data ? 0 : -ENOMEM; } @@ -76,7 +76,7 @@ int dcache_dir_close(struct inode *inode, struct file *file) loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) { - mutex_lock(&file->f_dentry->d_inode->i_mutex); + mutex_lock(&file->f_path.dentry->d_inode->i_mutex); switch (origin) { case 1: offset += file->f_pos; @@ -84,7 +84,7 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) if (offset >= 0) break; default: - mutex_unlock(&file->f_dentry->d_inode->i_mutex); + mutex_unlock(&file->f_path.dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -96,8 +96,8 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) spin_lock(&dcache_lock); list_del(&cursor->d_u.d_child); - p = file->f_dentry->d_subdirs.next; - while (n && p != &file->f_dentry->d_subdirs) { + p = file->f_path.dentry->d_subdirs.next; + while (n && p != &file->f_path.dentry->d_subdirs) { struct dentry *next; next = list_entry(p, struct dentry, d_u.d_child); if (!d_unhashed(next) && next->d_inode) @@ -108,7 +108,7 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) spin_unlock(&dcache_lock); } } - mutex_unlock(&file->f_dentry->d_inode->i_mutex); + mutex_unlock(&file->f_path.dentry->d_inode->i_mutex); return offset; } @@ -126,7 +126,7 @@ static inline unsigned char dt_type(struct inode *inode) int dcache_readdir(struct file * filp, void * dirent, filldir_t filldir) { - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct dentry *cursor = filp->private_data; struct list_head *p, *q = &cursor->d_u.d_child; ino_t ino; diff --git a/fs/locks.c b/fs/locks.c index 1cb0c57fedbd..52a81005dab4 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -321,7 +321,7 @@ static int flock_to_posix_lock(struct file *filp, struct file_lock *fl, start = filp->f_pos; break; case SEEK_END: - start = i_size_read(filp->f_dentry->d_inode); + start = i_size_read(filp->f_path.dentry->d_inode); break; default: return -EINVAL; @@ -371,7 +371,7 @@ static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl, start = filp->f_pos; break; case SEEK_END: - start = i_size_read(filp->f_dentry->d_inode); + start = i_size_read(filp->f_path.dentry->d_inode); break; default: return -EINVAL; @@ -672,7 +672,7 @@ posix_test_lock(struct file *filp, struct file_lock *fl, struct file_lock *cfl; lock_kernel(); - for (cfl = filp->f_dentry->d_inode->i_flock; cfl; cfl = cfl->fl_next) { + for (cfl = filp->f_path.dentry->d_inode->i_flock; cfl; cfl = cfl->fl_next) { if (!IS_POSIX(cfl)) continue; if (posix_locks_conflict(cfl, fl)) @@ -734,7 +734,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) { struct file_lock *new_fl = NULL; struct file_lock **before; - struct inode * inode = filp->f_dentry->d_inode; + struct inode * inode = filp->f_path.dentry->d_inode; int error = 0; int found = 0; @@ -1018,7 +1018,7 @@ static int __posix_lock_file_conf(struct inode *inode, struct file_lock *request */ int posix_lock_file(struct file *filp, struct file_lock *fl) { - return __posix_lock_file_conf(filp->f_dentry->d_inode, fl, NULL); + return __posix_lock_file_conf(filp->f_path.dentry->d_inode, fl, NULL); } EXPORT_SYMBOL(posix_lock_file); @@ -1033,7 +1033,7 @@ EXPORT_SYMBOL(posix_lock_file); int posix_lock_file_conf(struct file *filp, struct file_lock *fl, struct file_lock *conflock) { - return __posix_lock_file_conf(filp->f_dentry->d_inode, fl, conflock); + return __posix_lock_file_conf(filp->f_path.dentry->d_inode, fl, conflock); } EXPORT_SYMBOL(posix_lock_file_conf); @@ -1333,8 +1333,8 @@ int fcntl_getlease(struct file *filp) int type = F_UNLCK; lock_kernel(); - time_out_leases(filp->f_dentry->d_inode); - for (fl = filp->f_dentry->d_inode->i_flock; fl && IS_LEASE(fl); + time_out_leases(filp->f_path.dentry->d_inode); + for (fl = filp->f_path.dentry->d_inode->i_flock; fl && IS_LEASE(fl); fl = fl->fl_next) { if (fl->fl_file == filp) { type = fl->fl_type & ~F_INPROGRESS; @@ -1359,7 +1359,7 @@ int fcntl_getlease(struct file *filp) static int __setlease(struct file *filp, long arg, struct file_lock **flp) { struct file_lock *fl, **before, **my_before = NULL, *lease; - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; int error, rdlease_count = 0, wrlease_count = 0; @@ -1448,7 +1448,7 @@ out: int setlease(struct file *filp, long arg, struct file_lock **lease) { - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; int error; @@ -1482,7 +1482,7 @@ EXPORT_SYMBOL(setlease); int fcntl_setlease(unsigned int fd, struct file *filp, long arg) { struct file_lock fl, *flp = &fl; - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; int error; @@ -1692,7 +1692,7 @@ int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd, if (copy_from_user(&flock, l, sizeof(flock))) goto out; - inode = filp->f_dentry->d_inode; + inode = filp->f_path.dentry->d_inode; /* Don't allow mandatory locks on files that may be memory mapped * and shared. @@ -1835,7 +1835,7 @@ int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd, if (copy_from_user(&flock, l, sizeof(flock))) goto out; - inode = filp->f_dentry->d_inode; + inode = filp->f_path.dentry->d_inode; /* Don't allow mandatory locks on files that may be memory mapped * and shared. @@ -1922,7 +1922,7 @@ void locks_remove_posix(struct file *filp, fl_owner_t owner) * posix_lock_file(). Another process could be setting a lock on this * file at the same time, but we wouldn't remove that lock anyway. */ - if (!filp->f_dentry->d_inode->i_flock) + if (!filp->f_path.dentry->d_inode->i_flock) return; lock.fl_type = F_UNLCK; @@ -1951,7 +1951,7 @@ EXPORT_SYMBOL(locks_remove_posix); */ void locks_remove_flock(struct file *filp) { - struct inode * inode = filp->f_dentry->d_inode; + struct inode * inode = filp->f_path.dentry->d_inode; struct file_lock *fl; struct file_lock **before; @@ -2020,7 +2020,7 @@ static void lock_get_status(char* out, struct file_lock *fl, int id, char *pfx) struct inode *inode = NULL; if (fl->fl_file != NULL) - inode = fl->fl_file->f_dentry->d_inode; + inode = fl->fl_file->f_path.dentry->d_inode; out += sprintf(out, "%d:%s ", id, pfx); if (IS_POSIX(fl)) { diff --git a/fs/namei.c b/fs/namei.c index 8c2db88bb20d..e4f108f08230 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -297,7 +297,7 @@ int vfs_permission(struct nameidata *nd, int mask) */ int file_permission(struct file *file, int mask) { - return permission(file->f_dentry->d_inode, mask, NULL); + return permission(file->f_path.dentry->d_inode, mask, NULL); } /* @@ -333,7 +333,7 @@ int get_write_access(struct inode * inode) int deny_write_access(struct file * file) { - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file->f_path.dentry->d_inode; spin_lock(&inode->i_lock); if (atomic_read(&inode->i_writecount) > 0) { @@ -368,7 +368,7 @@ void path_release_on_umount(struct nameidata *nd) */ void release_open_intent(struct nameidata *nd) { - if (nd->intent.open.file->f_dentry == NULL) + if (nd->intent.open.file->f_path.dentry == NULL) put_filp(nd->intent.open.file); else fput(nd->intent.open.file); @@ -1138,7 +1138,7 @@ static int fastcall do_path_lookup(int dfd, const char *name, if (!file) goto out_fail; - dentry = file->f_dentry; + dentry = file->f_path.dentry; retval = -ENOTDIR; if (!S_ISDIR(dentry->d_inode->i_mode)) @@ -1148,7 +1148,7 @@ static int fastcall do_path_lookup(int dfd, const char *name, if (retval) goto fput_fail; - nd->mnt = mntget(file->f_vfsmnt); + nd->mnt = mntget(file->f_path.mnt); nd->dentry = dget(dentry); fput_light(file, fput_needed); diff --git a/fs/open.c b/fs/open.c index 3b56192816ca..0d94319e8681 100644 --- a/fs/open.c +++ b/fs/open.c @@ -165,7 +165,7 @@ asmlinkage long sys_fstatfs(unsigned int fd, struct statfs __user * buf) file = fget(fd); if (!file) goto out; - error = vfs_statfs_native(file->f_dentry, &tmp); + error = vfs_statfs_native(file->f_path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; fput(file); @@ -186,7 +186,7 @@ asmlinkage long sys_fstatfs64(unsigned int fd, size_t sz, struct statfs64 __user file = fget(fd); if (!file) goto out; - error = vfs_statfs64(file->f_dentry, &tmp); + error = vfs_statfs64(file->f_path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; fput(file); @@ -302,7 +302,7 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small) if (file->f_flags & O_LARGEFILE) small = 0; - dentry = file->f_dentry; + dentry = file->f_path.dentry; inode = dentry->d_inode; error = -EINVAL; if (!S_ISREG(inode->i_mode) || !(file->f_mode & FMODE_WRITE)) @@ -448,8 +448,8 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!file) goto out; - dentry = file->f_dentry; - mnt = file->f_vfsmnt; + dentry = file->f_path.dentry; + mnt = file->f_path.mnt; inode = dentry->d_inode; error = -ENOTDIR; @@ -503,7 +503,7 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) if (!file) goto out; - dentry = file->f_dentry; + dentry = file->f_path.dentry; inode = dentry->d_inode; audit_inode(NULL, inode); @@ -662,7 +662,7 @@ asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group) if (!file) goto out; - dentry = file->f_dentry; + dentry = file->f_path.dentry; audit_inode(NULL, dentry->d_inode); error = chown_common(dentry, user, group); fput(file); @@ -688,8 +688,8 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, } f->f_mapping = inode->i_mapping; - f->f_dentry = dentry; - f->f_vfsmnt = mnt; + f->f_path.dentry = dentry; + f->f_path.mnt = mnt; f->f_pos = 0; f->f_op = fops_get(inode->i_fop); file_move(f, &inode->i_sb->s_files); @@ -723,8 +723,8 @@ cleanup_all: if (f->f_mode & FMODE_WRITE) put_write_access(inode); file_kill(f); - f->f_dentry = NULL; - f->f_vfsmnt = NULL; + f->f_path.dentry = NULL; + f->f_path.mnt = NULL; cleanup_file: put_filp(f); dput(dentry); @@ -822,7 +822,7 @@ struct file *nameidata_to_filp(struct nameidata *nd, int flags) /* Pick up the filp from the open intent */ filp = nd->intent.open.file; /* Has the filesystem initialised the file for us? */ - if (filp->f_dentry == NULL) + if (filp->f_path.dentry == NULL) filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL); else path_release(nd); @@ -965,7 +965,7 @@ long do_sys_open(int dfd, const char __user *filename, int flags, int mode) put_unused_fd(fd); fd = PTR_ERR(f); } else { - fsnotify_open(f->f_dentry); + fsnotify_open(f->f_path.dentry); fd_install(fd, f); } } diff --git a/fs/pipe.c b/fs/pipe.c index ae36b89b1a37..f8b6bdcb879a 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -222,7 +222,7 @@ pipe_read(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t pos) { struct file *filp = iocb->ki_filp; - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe; int do_wakeup; ssize_t ret; @@ -335,7 +335,7 @@ pipe_write(struct kiocb *iocb, const struct iovec *_iov, unsigned long nr_segs, loff_t ppos) { struct file *filp = iocb->ki_filp; - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe; ssize_t ret; int do_wakeup; @@ -520,7 +520,7 @@ static int pipe_ioctl(struct inode *pino, struct file *filp, unsigned int cmd, unsigned long arg) { - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe; int count, buf, nrbufs; @@ -548,7 +548,7 @@ static unsigned int pipe_poll(struct file *filp, poll_table *wait) { unsigned int mask; - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe = inode->i_pipe; int nrbufs; @@ -601,7 +601,7 @@ pipe_release(struct inode *inode, int decr, int decw) static int pipe_read_fasync(int fd, struct file *filp, int on) { - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; int retval; mutex_lock(&inode->i_mutex); @@ -618,7 +618,7 @@ pipe_read_fasync(int fd, struct file *filp, int on) static int pipe_write_fasync(int fd, struct file *filp, int on) { - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; int retval; mutex_lock(&inode->i_mutex); @@ -635,7 +635,7 @@ pipe_write_fasync(int fd, struct file *filp, int on) static int pipe_rdwr_fasync(int fd, struct file *filp, int on) { - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = filp->f_path.dentry->d_inode; struct pipe_inode_info *pipe = inode->i_pipe; int retval; @@ -914,8 +914,8 @@ struct file *create_write_pipe(void) */ dentry->d_flags &= ~DCACHE_UNHASHED; d_instantiate(dentry, inode); - f->f_vfsmnt = mntget(pipe_mnt); - f->f_dentry = dentry; + f->f_path.mnt = mntget(pipe_mnt); + f->f_path.dentry = dentry; f->f_mapping = inode->i_mapping; f->f_flags = O_WRONLY; @@ -935,8 +935,8 @@ struct file *create_write_pipe(void) void free_write_pipe(struct file *f) { - mntput(f->f_vfsmnt); - dput(f->f_dentry); + mntput(f->f_path.mnt); + dput(f->f_path.dentry); put_filp(f); } @@ -947,9 +947,9 @@ struct file *create_read_pipe(struct file *wrf) return ERR_PTR(-ENFILE); /* Grab pipe from the writer */ - f->f_vfsmnt = mntget(wrf->f_vfsmnt); - f->f_dentry = dget(wrf->f_dentry); - f->f_mapping = wrf->f_dentry->d_inode->i_mapping; + f->f_path.mnt = mntget(wrf->f_path.mnt); + f->f_path.dentry = dget(wrf->f_path.dentry); + f->f_mapping = wrf->f_path.dentry->d_inode->i_mapping; f->f_pos = 0; f->f_flags = O_RDONLY; diff --git a/fs/read_write.c b/fs/read_write.c index f792000a28e6..1d3dda4fa70c 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -64,13 +64,13 @@ loff_t remote_llseek(struct file *file, loff_t offset, int origin) lock_kernel(); switch (origin) { case 2: - offset += i_size_read(file->f_dentry->d_inode); + offset += i_size_read(file->f_path.dentry->d_inode); break; case 1: offset += file->f_pos; } retval = -EINVAL; - if (offset>=0 && offset<=file->f_dentry->d_inode->i_sb->s_maxbytes) { + if (offset>=0 && offset<=file->f_path.dentry->d_inode->i_sb->s_maxbytes) { if (offset != file->f_pos) { file->f_pos = offset; file->f_version = 0; @@ -95,7 +95,7 @@ loff_t default_llseek(struct file *file, loff_t offset, int origin) lock_kernel(); switch (origin) { case 2: - offset += i_size_read(file->f_dentry->d_inode); + offset += i_size_read(file->f_path.dentry->d_inode); break; case 1: offset += file->f_pos; @@ -203,7 +203,7 @@ int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count if (unlikely((pos < 0) || (loff_t) (pos + count) < 0)) goto Einval; - inode = file->f_dentry->d_inode; + inode = file->f_path.dentry->d_inode; if (unlikely(inode->i_flock && MANDATORY_LOCK(inode))) { int retval = locks_mandatory_area( read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE, @@ -273,7 +273,7 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos) else ret = do_sync_read(file, buf, count, pos); if (ret > 0) { - fsnotify_access(file->f_dentry); + fsnotify_access(file->f_path.dentry); current->rchar += ret; } current->syscr++; @@ -331,7 +331,7 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_ else ret = do_sync_write(file, buf, count, pos); if (ret > 0) { - fsnotify_modify(file->f_dentry); + fsnotify_modify(file->f_path.dentry); current->wchar += ret; } current->syscw++; @@ -628,9 +628,9 @@ out: kfree(iov); if ((ret + (type == READ)) > 0) { if (type == READ) - fsnotify_access(file->f_dentry); + fsnotify_access(file->f_path.dentry); else - fsnotify_modify(file->f_dentry); + fsnotify_modify(file->f_path.dentry); } return ret; } @@ -722,7 +722,7 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (!(in_file->f_mode & FMODE_READ)) goto fput_in; retval = -EINVAL; - in_inode = in_file->f_dentry->d_inode; + in_inode = in_file->f_path.dentry->d_inode; if (!in_inode) goto fput_in; if (!in_file->f_op || !in_file->f_op->sendfile) @@ -754,7 +754,7 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, retval = -EINVAL; if (!out_file->f_op || !out_file->f_op->sendpage) goto fput_out; - out_inode = out_file->f_dentry->d_inode; + out_inode = out_file->f_path.dentry->d_inode; retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count); if (retval < 0) goto fput_out; diff --git a/fs/readdir.c b/fs/readdir.c index bff3ee58e2f8..f39f5b313252 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -21,7 +21,7 @@ int vfs_readdir(struct file *file, filldir_t filler, void *buf) { - struct inode *inode = file->f_dentry->d_inode; + struct inode *inode = file->f_path.dentry->d_inode; int res = -ENOTDIR; if (!file->f_op || !file->f_op->readdir) goto out; diff --git a/fs/seq_file.c b/fs/seq_file.c index 10690aa401c7..0ac22af7afe5 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -269,7 +269,7 @@ EXPORT_SYMBOL(seq_lseek); /** * seq_release - free the structures associated with sequential file. * @file: file in question - * @inode: file->f_dentry->d_inode + * @inode: file->f_path.dentry->d_inode * * Frees the structures associated with sequential file; can be used * as ->f_op->release() if you don't have private data to destroy. diff --git a/fs/splice.c b/fs/splice.c index da74583a00ee..bbd0aeb3f68e 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -844,7 +844,7 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; int err; - err = remove_suid(out->f_dentry); + err = remove_suid(out->f_path.dentry); if (unlikely(err)) return err; @@ -890,10 +890,10 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; int err; - err = should_remove_suid(out->f_dentry); + err = should_remove_suid(out->f_path.dentry); if (unlikely(err)) { mutex_lock(&inode->i_mutex); - err = __remove_suid(out->f_dentry, err); + err = __remove_suid(out->f_path.dentry, err); mutex_unlock(&inode->i_mutex); if (err) return err; @@ -1008,7 +1008,7 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, * randomly drop data for eg socket -> socket splicing. Use the * piped splicing for that! */ - i_mode = in->f_dentry->d_inode->i_mode; + i_mode = in->f_path.dentry->d_inode->i_mode; if (unlikely(!S_ISREG(i_mode) && !S_ISBLK(i_mode))) return -EINVAL; @@ -1132,7 +1132,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, loff_t offset, *off; long ret; - pipe = pipe_info(in->f_dentry->d_inode); + pipe = pipe_info(in->f_path.dentry->d_inode); if (pipe) { if (off_in) return -ESPIPE; @@ -1153,7 +1153,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, return ret; } - pipe = pipe_info(out->f_dentry->d_inode); + pipe = pipe_info(out->f_path.dentry->d_inode); if (pipe) { if (off_out) return -ESPIPE; @@ -1321,7 +1321,7 @@ static long do_vmsplice(struct file *file, const struct iovec __user *iov, .ops = &user_page_pipe_buf_ops, }; - pipe = pipe_info(file->f_dentry->d_inode); + pipe = pipe_info(file->f_path.dentry->d_inode); if (!pipe) return -EBADF; if (unlikely(nr_segs > UIO_MAXIOV)) @@ -1549,8 +1549,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, static long do_tee(struct file *in, struct file *out, size_t len, unsigned int flags) { - struct pipe_inode_info *ipipe = pipe_info(in->f_dentry->d_inode); - struct pipe_inode_info *opipe = pipe_info(out->f_dentry->d_inode); + struct pipe_inode_info *ipipe = pipe_info(in->f_path.dentry->d_inode); + struct pipe_inode_info *opipe = pipe_info(out->f_path.dentry->d_inode); int ret = -EINVAL; /* diff --git a/fs/stat.c b/fs/stat.c index a0ebfc7f8a64..38a8cb2a28de 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -102,7 +102,7 @@ int vfs_fstat(unsigned int fd, struct kstat *stat) int error = -EBADF; if (f) { - error = vfs_getattr(f->f_vfsmnt, f->f_dentry, stat); + error = vfs_getattr(f->f_path.mnt, f->f_path.dentry, stat); fput(f); } return error; diff --git a/fs/super.c b/fs/super.c index 84c320f6ad7e..f961e0307997 100644 --- a/fs/super.c +++ b/fs/super.c @@ -570,7 +570,7 @@ static void mark_files_ro(struct super_block *sb) file_list_lock(); list_for_each_entry(f, &sb->s_files, f_u.fu_list) { - if (S_ISREG(f->f_dentry->d_inode->i_mode) && file_count(f)) + if (S_ISREG(f->f_path.dentry->d_inode->i_mode) && file_count(f)) f->f_mode &= ~FMODE_WRITE; } file_list_unlock(); diff --git a/fs/sync.c b/fs/sync.c index 865f32be386e..d0feff61e6aa 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -94,7 +94,7 @@ long do_fsync(struct file *file, int datasync) * livelocks in fsync_buffers_list(). */ mutex_lock(&mapping->host->i_mutex); - err = file->f_op->fsync(file, file->f_dentry, datasync); + err = file->f_op->fsync(file, file->f_path.dentry, datasync); if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); @@ -223,7 +223,7 @@ asmlinkage long sys_sync_file_range(int fd, loff_t offset, loff_t nbytes, if (!file) goto out; - i_mode = file->f_dentry->d_inode->i_mode; + i_mode = file->f_path.dentry->d_inode->i_mode; ret = -ESPIPE; if (!S_ISREG(i_mode) && !S_ISBLK(i_mode) && !S_ISDIR(i_mode) && !S_ISLNK(i_mode)) diff --git a/fs/xattr.c b/fs/xattr.c index 0901bdc2ce24..38646132ab0e 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -268,7 +268,7 @@ sys_fsetxattr(int fd, char __user *name, void __user *value, f = fget(fd); if (!f) return error; - dentry = f->f_dentry; + dentry = f->f_path.dentry; audit_inode(NULL, dentry->d_inode); error = setxattr(dentry, name, value, size, flags); fput(f); @@ -351,7 +351,7 @@ sys_fgetxattr(int fd, char __user *name, void __user *value, size_t size) f = fget(fd); if (!f) return error; - error = getxattr(f->f_dentry, name, value, size); + error = getxattr(f->f_path.dentry, name, value, size); fput(f); return error; } @@ -423,7 +423,7 @@ sys_flistxattr(int fd, char __user *list, size_t size) f = fget(fd); if (!f) return error; - error = listxattr(f->f_dentry, list, size); + error = listxattr(f->f_path.dentry, list, size); fput(f); return error; } @@ -484,7 +484,7 @@ sys_fremovexattr(int fd, char __user *name) f = fget(fd); if (!f) return error; - dentry = f->f_dentry; + dentry = f->f_path.dentry; audit_inode(NULL, dentry->d_inode); error = removexattr(dentry, name); fput(f); diff --git a/include/linux/fs.h b/include/linux/fs.h index 45f2cabb8c75..adce6e1d70c2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -269,6 +269,7 @@ extern int dir_notify_enable; #include #include #include +#include #include #include #include @@ -711,8 +712,9 @@ struct file { struct list_head fu_list; struct rcu_head fu_rcuhead; } f_u; - struct dentry *f_dentry; - struct vfsmount *f_vfsmnt; + struct path f_path; +#define f_dentry f_path.dentry +#define f_vfsmnt f_path.mnt const struct file_operations *f_op; atomic_t f_count; unsigned int f_flags; @@ -1224,7 +1226,7 @@ extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry); static inline void file_accessed(struct file *file) { if (!(file->f_flags & O_NOATIME)) - touch_atime(file->f_vfsmnt, file->f_dentry); + touch_atime(file->f_path.mnt, file->f_path.dentry); } int sync_inode(struct inode *inode, struct writeback_control *wbc); @@ -1615,7 +1617,7 @@ static inline void put_write_access(struct inode * inode) static inline void allow_write_access(struct file *file) { if (file) - atomic_inc(&file->f_dentry->d_inode->i_writecount); + atomic_inc(&file->f_path.dentry->d_inode->i_writecount); } extern int do_pipe(int *); extern struct file *create_read_pipe(struct file *f); diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index d4f219ffaa5d..dfc4e4f68da4 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -164,7 +164,7 @@ static inline void fsnotify_open(struct dentry *dentry) */ static inline void fsnotify_close(struct file *file) { - struct dentry *dentry = file->f_dentry; + struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; const char *name = dentry->d_name.name; mode_t mode = file->f_mode; -- cgit v1.2.3 From 225a719f79fbc4d0cd9d9ebc5b2e3ac0e95845aa Mon Sep 17 00:00:00 2001 From: Josef Sipek Date: Fri, 8 Dec 2006 02:37:18 -0800 Subject: [PATCH] struct path: convert lockd Signed-off-by: Josef Sipek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/clntlock.c | 2 +- fs/lockd/clntproc.c | 2 +- fs/lockd/svclock.c | 16 ++++++++-------- fs/lockd/svcsubs.c | 2 +- include/linux/lockd/lockd.h | 2 +- 5 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index b85a0ad2cfb6..92681c9e9b20 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -126,7 +126,7 @@ __be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock continue; if (!nlm_cmp_addr(&block->b_host->h_addr, addr)) continue; - if (nfs_compare_fh(NFS_FH(fl_blocked->fl_file->f_dentry->d_inode) ,fh) != 0) + if (nfs_compare_fh(NFS_FH(fl_blocked->fl_file->f_path.dentry->d_inode) ,fh) != 0) continue; /* Alright, we found a lock. Set the return status * and wake up the caller diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 497c3cd59d52..80a1a6dccc8f 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -129,7 +129,7 @@ static void nlmclnt_setlockargs(struct nlm_rqst *req, struct file_lock *fl) nlmclnt_next_cookie(&argp->cookie); argp->state = nsm_local_state; - memcpy(&lock->fh, NFS_FH(fl->fl_file->f_dentry->d_inode), sizeof(struct nfs_fh)); + memcpy(&lock->fh, NFS_FH(fl->fl_file->f_path.dentry->d_inode), sizeof(struct nfs_fh)); lock->caller = utsname()->nodename; lock->oh.data = req->a_owner; lock->oh.len = snprintf(req->a_owner, sizeof(req->a_owner), "%u@%s", diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 7e219b938552..5c054b20fd5e 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -343,8 +343,8 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, __be32 ret; dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n", - file->f_file->f_dentry->d_inode->i_sb->s_id, - file->f_file->f_dentry->d_inode->i_ino, + file->f_file->f_path.dentry->d_inode->i_sb->s_id, + file->f_file->f_path.dentry->d_inode->i_ino, lock->fl.fl_type, lock->fl.fl_pid, (long long)lock->fl.fl_start, (long long)lock->fl.fl_end, @@ -420,8 +420,8 @@ nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, struct nlm_lock *conflock) { dprintk("lockd: nlmsvc_testlock(%s/%ld, ty=%d, %Ld-%Ld)\n", - file->f_file->f_dentry->d_inode->i_sb->s_id, - file->f_file->f_dentry->d_inode->i_ino, + file->f_file->f_path.dentry->d_inode->i_sb->s_id, + file->f_file->f_path.dentry->d_inode->i_ino, lock->fl.fl_type, (long long)lock->fl.fl_start, (long long)lock->fl.fl_end); @@ -454,8 +454,8 @@ nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) int error; dprintk("lockd: nlmsvc_unlock(%s/%ld, pi=%d, %Ld-%Ld)\n", - file->f_file->f_dentry->d_inode->i_sb->s_id, - file->f_file->f_dentry->d_inode->i_ino, + file->f_file->f_path.dentry->d_inode->i_sb->s_id, + file->f_file->f_path.dentry->d_inode->i_ino, lock->fl.fl_pid, (long long)lock->fl.fl_start, (long long)lock->fl.fl_end); @@ -483,8 +483,8 @@ nlmsvc_cancel_blocked(struct nlm_file *file, struct nlm_lock *lock) int status = 0; dprintk("lockd: nlmsvc_cancel(%s/%ld, pi=%d, %Ld-%Ld)\n", - file->f_file->f_dentry->d_inode->i_sb->s_id, - file->f_file->f_dentry->d_inode->i_ino, + file->f_file->f_path.dentry->d_inode->i_sb->s_id, + file->f_file->f_path.dentry->d_inode->i_ino, lock->fl.fl_pid, (long long)lock->fl.fl_start, (long long)lock->fl.fl_end); diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index e83024e16042..c0df00c74ce3 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -43,7 +43,7 @@ static inline void nlm_debug_print_fh(char *msg, struct nfs_fh *f) static inline void nlm_debug_print_file(char *msg, struct nlm_file *file) { - struct inode *inode = file->f_file->f_dentry->d_inode; + struct inode *inode = file->f_file->f_path.dentry->d_inode; dprintk("lockd: %s %s/%ld\n", msg, inode->i_sb->s_id, inode->i_ino); diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 8c39654549d8..0c962b82a9de 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -206,7 +206,7 @@ void nlmsvc_invalidate_all(void); static __inline__ struct inode * nlmsvc_file_inode(struct nlm_file *file) { - return file->f_file->f_dentry->d_inode; + return file->f_file->f_path.dentry->d_inode; } /* -- cgit v1.2.3 From f0d1b0b30d250a07627ad8b9fbbb5c7cc08422e8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Dec 2006 02:37:49 -0800 Subject: [PATCH] LOG2: Implement a general integer log2 facility in the kernel This facility provides three entry points: ilog2() Log base 2 of unsigned long ilog2_u32() Log base 2 of u32 ilog2_u64() Log base 2 of u64 These facilities can either be used inside functions on dynamic data: int do_something(long q) { ...; y = ilog2(x) ...; } Or can be used to statically initialise global variables with constant values: unsigned n = ilog2(27); When performing static initialisation, the compiler will report "error: initializer element is not constant" if asked to take a log of zero or of something not reducible to a constant. They treat negative numbers as unsigned. When not dealing with a constant, they fall back to using fls() which permits them to use arch-specific log calculation instructions - such as BSR on x86/x86_64 or SCAN on FRV - if available. [akpm@osdl.org: MMC fix] Signed-off-by: David Howells Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Herbert Xu Cc: David Howells Cc: Wojtek Kaniewski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/Kconfig | 8 ++ arch/arm/Kconfig | 8 ++ arch/arm26/Kconfig | 8 ++ arch/avr32/Kconfig | 8 ++ arch/cris/Kconfig | 8 ++ arch/frv/Kconfig | 8 ++ arch/h8300/Kconfig | 8 ++ arch/i386/Kconfig.cpu | 8 ++ arch/ia64/Kconfig | 8 ++ arch/m32r/Kconfig | 8 ++ arch/m68k/Kconfig | 8 ++ arch/m68knommu/Kconfig | 8 ++ arch/mips/Kconfig | 8 ++ arch/parisc/Kconfig | 8 ++ arch/powerpc/Kconfig | 8 ++ arch/ppc/Kconfig | 8 ++ arch/s390/Kconfig | 8 ++ arch/sh/Kconfig | 8 ++ arch/sh64/Kconfig | 8 ++ arch/sparc/Kconfig | 8 ++ arch/sparc64/Kconfig | 8 ++ arch/v850/Kconfig | 8 ++ arch/x86_64/Kconfig | 8 ++ arch/xtensa/Kconfig | 8 ++ drivers/infiniband/hw/mthca/mthca_provider.c | 4 +- drivers/infiniband/hw/mthca/mthca_qp.c | 4 +- drivers/infiniband/hw/mthca/mthca_srq.c | 4 +- drivers/infiniband/ulp/iser/iser_memory.c | 4 +- drivers/md/dm-crypt.c | 2 +- drivers/mmc/tifm_sd.c | 2 +- fs/ext2/super.c | 6 +- fs/ext3/super.c | 6 +- include/asm-frv/bitops.h | 44 +++++++++ include/linux/kernel.h | 9 +- include/linux/log2.h | 131 +++++++++++++++++++++++++++ mm/page_alloc.c | 4 +- 36 files changed, 384 insertions(+), 28 deletions(-) create mode 100644 include/linux/log2.h (limited to 'include/linux') diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 7e55ea66c6d4..84caf50725b5 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -25,6 +25,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 8c05d4321ae9..aa1d400d721a 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -74,6 +74,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/arm26/Kconfig b/arch/arm26/Kconfig index c14fe918bc4c..74eba8b5a8ca 100644 --- a/arch/arm26/Kconfig +++ b/arch/arm26/Kconfig @@ -41,6 +41,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index 5f1694eea842..bb059a4e1df9 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -45,6 +45,14 @@ config GENERIC_TIME config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_BUST_SPINLOCK bool diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index 6a1238a29d6c..3474309e049c 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -16,6 +16,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index cf1c446e003a..7561d7b72e75 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -41,6 +41,14 @@ config TIME_LOW_RES bool default y +config ARCH_HAS_ILOG2_U32 + bool + default y + +config ARCH_HAS_ILOG2_U64 + bool + default y + mainmenu "Fujitsu FR-V Kernel Configuration" source "init/Kconfig" diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig index cabf0bfffc53..34a84bc4baf5 100644 --- a/arch/h8300/Kconfig +++ b/arch/h8300/Kconfig @@ -29,6 +29,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index 821fd269ca58..2aecfba4ac4f 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -248,6 +248,14 @@ config RWSEM_XCHGADD_ALGORITHM depends on !M386 default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_CALIBRATE_DELAY bool default y diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 75d839715b2f..fcacfe291b9b 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -34,6 +34,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index 41fd490af3b4..f383dab973f5 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -214,6 +214,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index 7bc14461a6ac..70a577c89c7c 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -17,6 +17,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index aa70dde54228..25993c2a8fbb 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -25,6 +25,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default n +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index d8af858fe3f5..57af8d8cf46f 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -819,6 +819,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index d2101237442e..0f9ff618c6d7 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -25,6 +25,14 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_FIND_NEXT_BIT bool default y diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 291c95ac4b31..97c7a212cdda 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -41,6 +41,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index f76a146cff13..692b5ba53209 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -19,6 +19,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default y + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 583d9ff0a571..45e47bfb68a9 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -22,6 +22,14 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config GENERIC_HWEIGHT bool default y diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index d83d64af31f2..8e24c40662e3 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -59,6 +59,14 @@ config LOCKDEP_SUPPORT bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" menu "System type" diff --git a/arch/sh64/Kconfig b/arch/sh64/Kconfig index 58c678e06667..7bc0744b7ab6 100644 --- a/arch/sh64/Kconfig +++ b/arch/sh64/Kconfig @@ -39,6 +39,14 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_ISA_DMA bool +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source init/Kconfig menu "System type" diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 92a7c8a636d3..d0dec1ea2eed 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -166,6 +166,14 @@ config ARCH_MAY_HAVE_PC_FDC bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config SUN_PM bool default y diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig index b627f8dbcaad..d391d11f245a 100644 --- a/arch/sparc64/Kconfig +++ b/arch/sparc64/Kconfig @@ -34,6 +34,14 @@ config ARCH_MAY_HAVE_PC_FDC bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + config AUDIT_ARCH bool default y diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig index 37ec644603ab..bcf825875d17 100644 --- a/arch/v850/Kconfig +++ b/arch/v850/Kconfig @@ -38,6 +38,14 @@ config TIME_LOW_RES bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + # Turn off some random 386 crap that can affect device config config ISA bool diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 3254a616c69e..3ac581d17202 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -101,6 +101,14 @@ config GENERIC_BUG default y depends on BUG +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index c1e69a1f92a4..9eccfbd1b536 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -34,6 +34,14 @@ config GENERIC_HARDIRQS bool default y +config ARCH_HAS_ILOG2_U32 + bool + default n + +config ARCH_HAS_ILOG2_U64 + bool + default n + source "init/Kconfig" menu "Processor type and features" diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 21422a3336ad..7ec7c4b937f9 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -124,7 +124,7 @@ static int mthca_query_device(struct ib_device *ibdev, props->max_map_per_fmr = 255; else props->max_map_per_fmr = - (1 << (32 - long_log2(mdev->limits.num_mpts))) - 1; + (1 << (32 - ilog2(mdev->limits.num_mpts))) - 1; err = 0; out: @@ -816,7 +816,7 @@ static int mthca_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *uda lkey = ucmd.lkey; } - ret = mthca_RESIZE_CQ(dev, cq->cqn, lkey, long_log2(entries), &status); + ret = mthca_RESIZE_CQ(dev, cq->cqn, lkey, ilog2(entries), &status); if (status) ret = -EINVAL; diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 33e3ba7937f1..d844a2569b47 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -636,11 +636,11 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, if (mthca_is_memfree(dev)) { if (qp->rq.max) - qp_context->rq_size_stride = long_log2(qp->rq.max) << 3; + qp_context->rq_size_stride = ilog2(qp->rq.max) << 3; qp_context->rq_size_stride |= qp->rq.wqe_shift - 4; if (qp->sq.max) - qp_context->sq_size_stride = long_log2(qp->sq.max) << 3; + qp_context->sq_size_stride = ilog2(qp->sq.max) << 3; qp_context->sq_size_stride |= qp->sq.wqe_shift - 4; } diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 34d2c4768962..10684da33d58 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -120,7 +120,7 @@ static void mthca_arbel_init_srq_context(struct mthca_dev *dev, memset(context, 0, sizeof *context); - logsize = long_log2(srq->max); + logsize = ilog2(srq->max); context->state_logsize_srqn = cpu_to_be32(logsize << 24 | srq->srqn); context->lkey = cpu_to_be32(srq->mr.ibmr.lkey); context->db_index = cpu_to_be32(srq->db_index); @@ -213,7 +213,7 @@ int mthca_alloc_srq(struct mthca_dev *dev, struct mthca_pd *pd, if (!mthca_is_memfree(dev) && (ds > dev->limits.max_desc_sz)) return -EINVAL; - srq->wqe_shift = long_log2(ds); + srq->wqe_shift = ilog2(ds); srq->srqn = mthca_alloc(&dev->srq_table.alloc); if (srq->srqn == -1) diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index 5e122501fd80..3aedd59b8a84 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -114,7 +114,7 @@ int iser_start_rdma_unaligned_sg(struct iscsi_iser_cmd_task *iser_ctask, if (cmd_data_len > ISER_KMALLOC_THRESHOLD) mem = (void *)__get_free_pages(GFP_NOIO, - long_log2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); + ilog2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); else mem = kmalloc(cmd_data_len, GFP_NOIO); @@ -211,7 +211,7 @@ void iser_finalize_rdma_unaligned_sg(struct iscsi_iser_cmd_task *iser_ctask, if (cmd_data_len > ISER_KMALLOC_THRESHOLD) free_pages((unsigned long)mem_copy->copy_buf, - long_log2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); + ilog2(roundup_pow_of_two(cmd_data_len)) - PAGE_SHIFT); else kfree(mem_copy->copy_buf); diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index a1086ee8cccd..96152868525b 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -220,7 +220,7 @@ static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti, const char *opts) { unsigned int bs = crypto_blkcipher_blocksize(cc->tfm); - int log = long_log2(bs); + int log = ilog2(bs); /* we need to calculate how far we must shift the sector count * to get the cipher block count, we use this shift in _gen */ diff --git a/drivers/mmc/tifm_sd.c b/drivers/mmc/tifm_sd.c index e846499a004c..f18ad998b3cb 100644 --- a/drivers/mmc/tifm_sd.c +++ b/drivers/mmc/tifm_sd.c @@ -387,7 +387,7 @@ static void tifm_sd_prepare_data(struct tifm_sd *card, struct mmc_command *cmd) writel(TIFM_FIFO_INT_SETALL, sock->addr + SOCK_DMA_FIFO_INT_ENABLE_CLEAR); - writel(long_log2(cmd->data->blksz) - 2, + writel(ilog2(cmd->data->blksz) - 2, sock->addr + SOCK_FIFO_PAGE_SIZE); writel(TIFM_FIFO_ENABLE, sock->addr + SOCK_FIFO_CONTROL); writel(TIFM_FIFO_INTMASK, sock->addr + SOCK_DMA_FIFO_INT_ENABLE_SET); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 255cef5f7420..6347c2dbdd81 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -597,8 +597,6 @@ static int ext2_check_descriptors (struct super_block * sb) return 1; } -#define log2(n) ffz(~(n)) - /* * Maximal file size. There is a direct, and {,double-,triple-}indirect * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. @@ -834,9 +832,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); sbi->s_addr_per_block_bits = - log2 (EXT2_ADDR_PER_BLOCK(sb)); + ilog2 (EXT2_ADDR_PER_BLOCK(sb)); sbi->s_desc_per_block_bits = - log2 (EXT2_DESC_PER_BLOCK(sb)); + ilog2 (EXT2_DESC_PER_BLOCK(sb)); if (sb->s_magic != EXT2_SUPER_MAGIC) goto cantfind_ext2; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 580b8a6ca979..b34886734a44 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1347,8 +1347,6 @@ static void ext3_orphan_cleanup (struct super_block * sb, sb->s_flags = s_flags; /* Restore MS_RDONLY status */ } -#define log2(n) ffz(~(n)) - /* * Maximal file size. There is a direct, and {,double-,triple-}indirect * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. @@ -1597,8 +1595,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) sbi->s_desc_per_block = blocksize / sizeof(struct ext3_group_desc); sbi->s_sbh = bh; sbi->s_mount_state = le16_to_cpu(es->s_state); - sbi->s_addr_per_block_bits = log2(EXT3_ADDR_PER_BLOCK(sb)); - sbi->s_desc_per_block_bits = log2(EXT3_DESC_PER_BLOCK(sb)); + sbi->s_addr_per_block_bits = ilog2(EXT3_ADDR_PER_BLOCK(sb)); + sbi->s_desc_per_block_bits = ilog2(EXT3_DESC_PER_BLOCK(sb)); for (i=0; i < 4; i++) sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); sbi->s_def_hash_version = es->s_def_hash_version; diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h index 1f70d47148bd..f8560edf59ff 100644 --- a/include/asm-frv/bitops.h +++ b/include/asm-frv/bitops.h @@ -256,6 +256,50 @@ int __ffs(unsigned long x) return 31 - bit; } +/* + * special slimline version of fls() for calculating ilog2_u32() + * - note: no protection against n == 0 + */ +#define ARCH_HAS_ILOG2_U32 +static inline __attribute__((const)) +int __ilog2_u32(u32 n) +{ + int bit; + asm("scan %1,gr0,%0" : "=r"(bit) : "r"(n)); + return 31 - bit; +} + +/* + * special slimline version of fls64() for calculating ilog2_u64() + * - note: no protection against n == 0 + */ +#define ARCH_HAS_ILOG2_U64 +static inline __attribute__((const)) +int __ilog2_u64(u64 n) +{ + union { + u64 ll; + struct { u32 h, l; }; + } _; + int bit, x, y; + + _.ll = n; + + asm(" subcc %3,gr0,gr0,icc0 \n" + " ckeq icc0,cc4 \n" + " cscan.p %3,gr0,%0 ,cc4,0 \n" + " setlos #63,%1 \n" + " cscan.p %4,gr0,%0 ,cc4,1 \n" + " setlos #31,%2 \n" + " csub.p %1,%0,%0 ,cc4,0 \n" + " csub %2,%0,%0 ,cc4,1 \n" + : "=&r"(bit), "=r"(x), "=r"(y) + : "0r"(_.h), "r"(_.l) + : "icc0", "cc4" + ); + return bit; +} + #include #include diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 6738283ac385..3710cce16642 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -157,14 +158,6 @@ static inline int printk(const char *s, ...) { return 0; } unsigned long int_sqrt(unsigned long); -static inline int __attribute_pure__ long_log2(unsigned long x) -{ - int r = 0; - for (x >>= 1; x > 0; x >>= 1) - r++; - return r; -} - static inline unsigned long __attribute_const__ roundup_pow_of_two(unsigned long x) { diff --git a/include/linux/log2.h b/include/linux/log2.h new file mode 100644 index 000000000000..3979c60325ff --- /dev/null +++ b/include/linux/log2.h @@ -0,0 +1,131 @@ +/* Integer base 2 logarithm calculation + * + * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#ifndef _LINUX_LOG2_H +#define _LINUX_LOG2_H + +#include +#include + +/* + * deal with unrepresentable constant logarithms + */ +extern __attribute__((const, noreturn)) +int ____ilog2_NaN(void); + +/* + * non-constant log of base 2 calculators + * - the arch may override these in asm/bitops.h if they can be implemented + * more efficiently than using fls() and fls64() + * - the arch is not required to handle n==0 if implementing the fallback + */ +#ifndef CONFIG_ARCH_HAS_ILOG2_U32 +static inline __attribute__((const)) +int __ilog2_u32(u32 n) +{ + return fls(n) - 1; +} +#endif + +#ifndef CONFIG_ARCH_HAS_ILOG2_U64 +static inline __attribute__((const)) +int __ilog2_u64(u64 n) +{ + return fls64(n) - 1; +} +#endif + +/** + * ilog2 - log of base 2 of 32-bit or a 64-bit unsigned value + * @n - parameter + * + * constant-capable log of base 2 calculation + * - this can be used to initialise global variables from constant data, hence + * the massive ternary operator construction + * + * selects the appropriately-sized optimised version depending on sizeof(n) + */ +#define ilog2(n) \ +( \ + __builtin_constant_p(n) ? ( \ + (n) < 1 ? ____ilog2_NaN() : \ + (n) & (1ULL << 63) ? 63 : \ + (n) & (1ULL << 62) ? 62 : \ + (n) & (1ULL << 61) ? 61 : \ + (n) & (1ULL << 60) ? 60 : \ + (n) & (1ULL << 59) ? 59 : \ + (n) & (1ULL << 58) ? 58 : \ + (n) & (1ULL << 57) ? 57 : \ + (n) & (1ULL << 56) ? 56 : \ + (n) & (1ULL << 55) ? 55 : \ + (n) & (1ULL << 54) ? 54 : \ + (n) & (1ULL << 53) ? 53 : \ + (n) & (1ULL << 52) ? 52 : \ + (n) & (1ULL << 51) ? 51 : \ + (n) & (1ULL << 50) ? 50 : \ + (n) & (1ULL << 49) ? 49 : \ + (n) & (1ULL << 48) ? 48 : \ + (n) & (1ULL << 47) ? 47 : \ + (n) & (1ULL << 46) ? 46 : \ + (n) & (1ULL << 45) ? 45 : \ + (n) & (1ULL << 44) ? 44 : \ + (n) & (1ULL << 43) ? 43 : \ + (n) & (1ULL << 42) ? 42 : \ + (n) & (1ULL << 41) ? 41 : \ + (n) & (1ULL << 40) ? 40 : \ + (n) & (1ULL << 39) ? 39 : \ + (n) & (1ULL << 38) ? 38 : \ + (n) & (1ULL << 37) ? 37 : \ + (n) & (1ULL << 36) ? 36 : \ + (n) & (1ULL << 35) ? 35 : \ + (n) & (1ULL << 34) ? 34 : \ + (n) & (1ULL << 33) ? 33 : \ + (n) & (1ULL << 32) ? 32 : \ + (n) & (1ULL << 31) ? 31 : \ + (n) & (1ULL << 30) ? 30 : \ + (n) & (1ULL << 29) ? 29 : \ + (n) & (1ULL << 28) ? 28 : \ + (n) & (1ULL << 27) ? 27 : \ + (n) & (1ULL << 26) ? 26 : \ + (n) & (1ULL << 25) ? 25 : \ + (n) & (1ULL << 24) ? 24 : \ + (n) & (1ULL << 23) ? 23 : \ + (n) & (1ULL << 22) ? 22 : \ + (n) & (1ULL << 21) ? 21 : \ + (n) & (1ULL << 20) ? 20 : \ + (n) & (1ULL << 19) ? 19 : \ + (n) & (1ULL << 18) ? 18 : \ + (n) & (1ULL << 17) ? 17 : \ + (n) & (1ULL << 16) ? 16 : \ + (n) & (1ULL << 15) ? 15 : \ + (n) & (1ULL << 14) ? 14 : \ + (n) & (1ULL << 13) ? 13 : \ + (n) & (1ULL << 12) ? 12 : \ + (n) & (1ULL << 11) ? 11 : \ + (n) & (1ULL << 10) ? 10 : \ + (n) & (1ULL << 9) ? 9 : \ + (n) & (1ULL << 8) ? 8 : \ + (n) & (1ULL << 7) ? 7 : \ + (n) & (1ULL << 6) ? 6 : \ + (n) & (1ULL << 5) ? 5 : \ + (n) & (1ULL << 4) ? 4 : \ + (n) & (1ULL << 3) ? 3 : \ + (n) & (1ULL << 2) ? 2 : \ + (n) & (1ULL << 1) ? 1 : \ + (n) & (1ULL << 0) ? 0 : \ + ____ilog2_NaN() \ + ) : \ + (sizeof(n) <= 4) ? \ + __ilog2_u32(n) : \ + __ilog2_u64(n) \ + ) + +#endif /* _LINUX_LOG2_H */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index cace22b3ac25..18f0e044c43d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3244,7 +3244,7 @@ void *__init alloc_large_system_hash(const char *tablename, if (numentries > max) numentries = max; - log2qty = long_log2(numentries); + log2qty = ilog2(numentries); do { size = bucketsize << log2qty; @@ -3266,7 +3266,7 @@ void *__init alloc_large_system_hash(const char *tablename, printk("%s hash table entries: %d (order: %d, %lu bytes)\n", tablename, (1U << log2qty), - long_log2(size) - PAGE_SHIFT, + ilog2(size) - PAGE_SHIFT, size); if (_hash_shift) -- cgit v1.2.3 From 312a0c170945b49f319960afd2e492c05f9dd551 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 8 Dec 2006 02:37:51 -0800 Subject: [PATCH] LOG2: Alter roundup_pow_of_two() so that it can use a ilog2() on a constant Alter roundup_pow_of_two() so that it can make use of ilog2() on a constant to produce a constant value, retaining the ability for an arch to override it in the non-const case. This permits the function to be used to initialise variables. Signed-off-by: David Howells Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 6 ------ include/linux/log2.h | 26 ++++++++++++++++++++++++++ 2 files changed, 26 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 3710cce16642..e8bfac34d2ba 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -158,12 +158,6 @@ static inline int printk(const char *s, ...) { return 0; } unsigned long int_sqrt(unsigned long); -static inline unsigned long -__attribute_const__ roundup_pow_of_two(unsigned long x) -{ - return 1UL << fls_long(x - 1); -} - extern int printk_ratelimit(void); extern int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst); extern bool printk_timed_ratelimit(unsigned long *caller_jiffies, diff --git a/include/linux/log2.h b/include/linux/log2.h index 3979c60325ff..d02e1a547a7e 100644 --- a/include/linux/log2.h +++ b/include/linux/log2.h @@ -43,6 +43,15 @@ int __ilog2_u64(u64 n) } #endif +/* + * round up to nearest power of two + */ +static inline __attribute__((const)) +unsigned long __roundup_pow_of_two(unsigned long n) +{ + return 1UL << fls_long(n - 1); +} + /** * ilog2 - log of base 2 of 32-bit or a 64-bit unsigned value * @n - parameter @@ -128,4 +137,21 @@ int __ilog2_u64(u64 n) __ilog2_u64(n) \ ) +/** + * roundup_pow_of_two - round the given value up to nearest power of two + * @n - parameter + * + * round the given balue up to the nearest power of two + * - the result is undefined when n == 0 + * - this can be used to initialise global variables from constant data + */ +#define roundup_pow_of_two(n) \ +( \ + __builtin_constant_p(n) ? ( \ + (n == 1) ? 0 : \ + (1UL << (ilog2((n) - 1) + 1)) \ + ) : \ + __roundup_pow_of_two(n) \ + ) + #endif /* _LINUX_LOG2_H */ -- cgit v1.2.3 From 937949d9edbf4049bd41af6c9f92c26280584564 Mon Sep 17 00:00:00 2001 From: Cedric Le Goater Date: Fri, 8 Dec 2006 02:37:54 -0800 Subject: [PATCH] add process_session() helper routine Replace occurences of task->signal->session by a new process_session() helper routine. It will be useful for pid namespaces to abstract the session pid number. Signed-off-by: Cedric Le Goater Cc: Kirill Korotaev Cc: Eric W. Biederman Cc: Herbert Poetzl Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/mips/kernel/irixelf.c | 2 +- drivers/char/mxser.c | 2 +- drivers/char/rocket.c | 2 +- drivers/char/tty_io.c | 12 ++++++------ fs/binfmt_elf.c | 4 ++-- fs/binfmt_elf_fdpic.c | 4 ++-- include/linux/sched.h | 5 +++++ kernel/exit.c | 19 ++++++++++--------- kernel/fork.c | 4 ++-- kernel/signal.c | 2 +- kernel/sys.c | 8 ++++---- 11 files changed, 35 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/arch/mips/kernel/irixelf.c b/arch/mips/kernel/irixelf.c index 1bbefbf43373..37cad5de515c 100644 --- a/arch/mips/kernel/irixelf.c +++ b/arch/mips/kernel/irixelf.c @@ -1145,7 +1145,7 @@ static int irix_core_dump(long signr, struct pt_regs * regs, struct file *file) psinfo.pr_pid = prstatus.pr_pid = current->pid; psinfo.pr_ppid = prstatus.pr_ppid = current->parent->pid; psinfo.pr_pgrp = prstatus.pr_pgrp = process_group(current); - psinfo.pr_sid = prstatus.pr_sid = current->signal->session; + psinfo.pr_sid = prstatus.pr_sid = process_session(current); if (current->pid == current->tgid) { /* * This is the record for the group leader. Add in the diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c index 5ed2486b7581..b1cc89c89820 100644 --- a/drivers/char/mxser.c +++ b/drivers/char/mxser.c @@ -994,7 +994,7 @@ static int mxser_open(struct tty_struct *tty, struct file *filp) mxser_change_speed(info, NULL); } - info->session = current->signal->session; + info->session = process_session(current); info->pgrp = process_group(current); /* diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c index bac80056f7e0..4fdf52e9f3b1 100644 --- a/drivers/char/rocket.c +++ b/drivers/char/rocket.c @@ -1017,7 +1017,7 @@ static int rp_open(struct tty_struct *tty, struct file *filp) /* * Info->count is now 1; so it's safe to sleep now. */ - info->session = current->signal->session; + info->session = process_session(current); info->pgrp = process_group(current); if ((info->flags & ROCKET_INITIALIZED) == 0) { diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 8a2d26e98ac4..4ebba7ca1dc9 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -1511,7 +1511,7 @@ void disassociate_ctty(int on_exit) spin_lock_irq(¤t->sighand->siglock); current->signal->tty_old_pgrp = 0; - session = current->signal->session; + session = process_session(current); spin_unlock_irq(¤t->sighand->siglock); mutex_lock(&tty_mutex); @@ -2897,7 +2897,7 @@ static int tiocsctty(struct tty_struct *tty, int arg) { int ret = 0; if (current->signal->leader && - (current->signal->session == tty->session)) + (process_session(current) == tty->session)) return ret; mutex_lock(&tty_mutex); @@ -2979,13 +2979,13 @@ static int tiocspgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t return retval; if (!current->signal->tty || (current->signal->tty != real_tty) || - (real_tty->session != current->signal->session)) + (real_tty->session != process_session(current))) return -ENOTTY; if (get_user(pgrp, p)) return -EFAULT; if (pgrp < 0) return -EINVAL; - if (session_of_pgrp(pgrp) != current->signal->session) + if (session_of_pgrp(pgrp) != process_session(current)) return -EPERM; real_tty->pgrp = pgrp; return 0; @@ -3338,7 +3338,7 @@ static void __do_SAK(struct work_struct *work) /* Kill the entire session */ do_each_task_pid(session, PIDTYPE_SID, p) { printk(KERN_NOTICE "SAK: killed process %d" - " (%s): p->signal->session==tty->session\n", + " (%s): process_session(p)==tty->session\n", p->pid, p->comm); send_sig(SIGKILL, p, 1); } while_each_task_pid(session, PIDTYPE_SID, p); @@ -3348,7 +3348,7 @@ static void __do_SAK(struct work_struct *work) do_each_thread(g, p) { if (p->signal->tty == tty) { printk(KERN_NOTICE "SAK: killed process %d" - " (%s): p->signal->session==tty->session\n", + " (%s): process_session(p)==tty->session\n", p->pid, p->comm); send_sig(SIGKILL, p, 1); continue; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index c6dbb4a7ec78..d3adfd353ff9 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1317,7 +1317,7 @@ static void fill_prstatus(struct elf_prstatus *prstatus, prstatus->pr_pid = p->pid; prstatus->pr_ppid = p->parent->pid; prstatus->pr_pgrp = process_group(p); - prstatus->pr_sid = p->signal->session; + prstatus->pr_sid = process_session(p); if (thread_group_leader(p)) { /* * This is the record for the group leader. Add in the @@ -1363,7 +1363,7 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, psinfo->pr_pid = p->pid; psinfo->pr_ppid = p->parent->pid; psinfo->pr_pgrp = process_group(p); - psinfo->pr_sid = p->signal->session; + psinfo->pr_sid = process_session(p); i = p->state ? ffz(~p->state) + 1 : 0; psinfo->pr_state = i; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 9f0b7efc3df5..76f06f6bc2f6 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -1322,7 +1322,7 @@ static void fill_prstatus(struct elf_prstatus *prstatus, prstatus->pr_pid = p->pid; prstatus->pr_ppid = p->parent->pid; prstatus->pr_pgrp = process_group(p); - prstatus->pr_sid = p->signal->session; + prstatus->pr_sid = process_session(p); if (thread_group_leader(p)) { /* * This is the record for the group leader. Add in the @@ -1371,7 +1371,7 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, psinfo->pr_pid = p->pid; psinfo->pr_ppid = p->parent->pid; psinfo->pr_pgrp = process_group(p); - psinfo->pr_sid = p->signal->session; + psinfo->pr_sid = process_session(p); i = p->state ? ffz(~p->state) + 1 : 0; psinfo->pr_state = i; diff --git a/include/linux/sched.h b/include/linux/sched.h index 5e8a0ba61749..270d864a8ff1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1047,6 +1047,11 @@ static inline pid_t process_group(struct task_struct *tsk) return tsk->signal->pgrp; } +static inline pid_t process_session(struct task_struct *tsk) +{ + return tsk->signal->session; +} + static inline struct pid *task_pid(struct task_struct *task) { return task->pids[PIDTYPE_PID].pid; diff --git a/kernel/exit.c b/kernel/exit.c index fa0495e167e9..8d289bfc13d1 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -193,14 +193,14 @@ int session_of_pgrp(int pgrp) read_lock(&tasklist_lock); do_each_task_pid(pgrp, PIDTYPE_PGID, p) { - if (p->signal->session > 0) { - sid = p->signal->session; + if (process_session(p) > 0) { + sid = process_session(p); goto out; } } while_each_task_pid(pgrp, PIDTYPE_PGID, p); p = find_task_by_pid(pgrp); if (p) - sid = p->signal->session; + sid = process_session(p); out: read_unlock(&tasklist_lock); @@ -225,8 +225,8 @@ static int will_become_orphaned_pgrp(int pgrp, struct task_struct *ignored_task) || p->exit_state || is_init(p->real_parent)) continue; - if (process_group(p->real_parent) != pgrp - && p->real_parent->signal->session == p->signal->session) { + if (process_group(p->real_parent) != pgrp && + process_session(p->real_parent) == process_session(p)) { ret = 0; break; } @@ -302,7 +302,7 @@ void __set_special_pids(pid_t session, pid_t pgrp) { struct task_struct *curr = current->group_leader; - if (curr->signal->session != session) { + if (process_session(curr) != session) { detach_pid(curr, PIDTYPE_SID); curr->signal->session = session; attach_pid(curr, PIDTYPE_SID, session); @@ -647,10 +647,11 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced) * outside, so the child pgrp is now orphaned. */ if ((process_group(p) != process_group(father)) && - (p->signal->session == father->signal->session)) { + (process_session(p) == process_session(father))) { int pgrp = process_group(p); - if (will_become_orphaned_pgrp(pgrp, NULL) && has_stopped_jobs(pgrp)) { + if (will_become_orphaned_pgrp(pgrp, NULL) && + has_stopped_jobs(pgrp)) { __kill_pg_info(SIGHUP, SEND_SIG_PRIV, pgrp); __kill_pg_info(SIGCONT, SEND_SIG_PRIV, pgrp); } @@ -784,7 +785,7 @@ static void exit_notify(struct task_struct *tsk) t = tsk->real_parent; if ((process_group(t) != process_group(tsk)) && - (t->signal->session == tsk->signal->session) && + (process_session(t) == process_session(tsk)) && will_become_orphaned_pgrp(process_group(tsk), tsk) && has_stopped_jobs(process_group(tsk))) { __kill_pg_info(SIGHUP, SEND_SIG_PRIV, process_group(tsk)); diff --git a/kernel/fork.c b/kernel/fork.c index 597707819327..298c4d6ab512 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1259,9 +1259,9 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (thread_group_leader(p)) { p->signal->tty = current->signal->tty; p->signal->pgrp = process_group(current); - p->signal->session = current->signal->session; + p->signal->session = process_session(current); attach_pid(p, PIDTYPE_PGID, process_group(p)); - attach_pid(p, PIDTYPE_SID, p->signal->session); + attach_pid(p, PIDTYPE_SID, process_session(p)); list_add_tail_rcu(&p->tasks, &init_task.tasks); __get_cpu_var(process_counts)++; diff --git a/kernel/signal.c b/kernel/signal.c index ec81defde339..9eac4db60eda 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -583,7 +583,7 @@ static int check_kill_permission(int sig, struct siginfo *info, error = -EPERM; if ((info == SEND_SIG_NOINFO || (!is_si_special(info) && SI_FROMUSER(info))) && ((sig != SIGCONT) || - (current->signal->session != t->signal->session)) + (process_session(current) != process_session(t))) && (current->euid ^ t->suid) && (current->euid ^ t->uid) && (current->uid ^ t->suid) && (current->uid ^ t->uid) && !capable(CAP_KILL)) diff --git a/kernel/sys.c b/kernel/sys.c index 1ac2d1c5d84e..4f9d23a3095f 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1381,7 +1381,7 @@ asmlinkage long sys_setpgid(pid_t pid, pid_t pgid) if (p->real_parent == group_leader) { err = -EPERM; - if (p->signal->session != group_leader->signal->session) + if (process_session(p) != process_session(group_leader)) goto out; err = -EACCES; if (p->did_exec) @@ -1400,7 +1400,7 @@ asmlinkage long sys_setpgid(pid_t pid, pid_t pgid) struct task_struct *p; do_each_task_pid(pgid, PIDTYPE_PGID, p) { - if (p->signal->session == group_leader->signal->session) + if (process_session(p) == process_session(group_leader)) goto ok_pgid; } while_each_task_pid(pgid, PIDTYPE_PGID, p); goto out; @@ -1459,7 +1459,7 @@ asmlinkage long sys_getpgrp(void) asmlinkage long sys_getsid(pid_t pid) { if (!pid) - return current->signal->session; + return process_session(current); else { int retval; struct task_struct *p; @@ -1471,7 +1471,7 @@ asmlinkage long sys_getsid(pid_t pid) if (p) { retval = security_task_getsid(p); if (!retval) - retval = p->signal->session; + retval = process_session(p); } read_unlock(&tasklist_lock); return retval; -- cgit v1.2.3 From 1ec320afdc9552c92191d5f89fcd1ebe588334ca Mon Sep 17 00:00:00 2001 From: Cedric Le Goater Date: Fri, 8 Dec 2006 02:37:55 -0800 Subject: [PATCH] add process_session() helper routine: deprecate old field Add an anonymous union and ((deprecated)) to catch direct usage of the session field. [akpm@osdl.org: fix various missed conversions] [jdike@addtoit.com: fix UML bug] Signed-off-by: Jeff Dike Cc: Cedric Le Goater Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 2 +- fs/proc/array.c | 2 +- include/linux/init_task.h | 11 ++++++----- include/linux/sched.h | 19 +++++++++++++++++-- kernel/exit.c | 2 +- kernel/fork.c | 2 +- 6 files changed, 27 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 4ebba7ca1dc9..48cee2004e97 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -3855,7 +3855,7 @@ EXPORT_SYMBOL(proc_clear_tty); void __proc_set_tty(struct task_struct *tsk, struct tty_struct *tty) { if (tty) { - tty->session = tsk->signal->session; + tty->session = process_session(tsk); tty->pgrp = process_group(tsk); } tsk->signal->tty = tty; diff --git a/fs/proc/array.c b/fs/proc/array.c index b0cd014a39bd..70e4fab117b1 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -381,7 +381,7 @@ static int do_task_stat(struct task_struct *task, char * buffer, int whole) stime = cputime_add(stime, sig->stime); } - sid = sig->session; + sid = signal_session(sig); pgid = process_group(task); ppid = rcu_dereference(task->real_parent)->tgid; diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 733790d4f7db..848a68af3d42 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -57,17 +57,18 @@ .cpu_vm_mask = CPU_MASK_ALL, \ } -#define INIT_SIGNALS(sig) { \ - .count = ATOMIC_INIT(1), \ +#define INIT_SIGNALS(sig) { \ + .count = ATOMIC_INIT(1), \ .wait_chldexit = __WAIT_QUEUE_HEAD_INITIALIZER(sig.wait_chldexit),\ - .shared_pending = { \ + .shared_pending = { \ .list = LIST_HEAD_INIT(sig.shared_pending.list), \ - .signal = {{0}}}, \ + .signal = {{0}}}, \ .posix_timers = LIST_HEAD_INIT(sig.posix_timers), \ .cpu_timers = INIT_CPU_TIMERS(sig.cpu_timers), \ .rlim = INIT_RLIMITS, \ .pgrp = 1, \ - .session = 1, \ + .tty_old_pgrp = 0, \ + { .__session = 1}, \ } extern struct nsproxy init_nsproxy; diff --git a/include/linux/sched.h b/include/linux/sched.h index 270d864a8ff1..6fec1d419714 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -436,7 +436,12 @@ struct signal_struct { /* job control IDs */ pid_t pgrp; pid_t tty_old_pgrp; - pid_t session; + + union { + pid_t session __deprecated; + pid_t __session; + }; + /* boolean value for session group leader */ int leader; @@ -1047,9 +1052,19 @@ static inline pid_t process_group(struct task_struct *tsk) return tsk->signal->pgrp; } +static inline pid_t signal_session(struct signal_struct *sig) +{ + return sig->__session; +} + static inline pid_t process_session(struct task_struct *tsk) { - return tsk->signal->session; + return signal_session(tsk->signal); +} + +static inline void set_signal_session(struct signal_struct *sig, pid_t session) +{ + sig->__session = session; } static inline struct pid *task_pid(struct task_struct *task) diff --git a/kernel/exit.c b/kernel/exit.c index 8d289bfc13d1..6267a6cc6113 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -304,7 +304,7 @@ void __set_special_pids(pid_t session, pid_t pgrp) if (process_session(curr) != session) { detach_pid(curr, PIDTYPE_SID); - curr->signal->session = session; + set_signal_session(curr->signal, session); attach_pid(curr, PIDTYPE_SID, session); } if (process_group(curr) != pgrp) { diff --git a/kernel/fork.c b/kernel/fork.c index 298c4d6ab512..60d2644bfe85 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1259,7 +1259,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (thread_group_leader(p)) { p->signal->tty = current->signal->tty; p->signal->pgrp = process_group(current); - p->signal->session = process_session(current); + set_signal_session(p->signal, process_session(current)); attach_pid(p, PIDTYPE_PGID, process_group(p)); attach_pid(p, PIDTYPE_SID, process_session(p)); -- cgit v1.2.3 From 6b3286ed1169d74fea401367d6d4d6c6ec758a81 Mon Sep 17 00:00:00 2001 From: Kirill Korotaev Date: Fri, 8 Dec 2006 02:37:56 -0800 Subject: [PATCH] rename struct namespace to struct mnt_namespace Rename 'struct namespace' to 'struct mnt_namespace' to avoid confusion with other namespaces being developped for the containers : pid, uts, ipc, etc. 'namespace' variables and attributes are also renamed to 'mnt_ns' Signed-off-by: Kirill Korotaev Signed-off-by: Cedric Le Goater Cc: Eric W. Biederman Cc: Herbert Poetzl Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/afs/mntpt.c | 2 +- fs/namespace.c | 113 +++++++++++++++++++++--------------------- fs/nfs/getroot.c | 2 +- fs/pnode.c | 2 +- fs/pnode.h | 2 +- fs/proc/base.c | 36 +++++++------- fs/reiserfs/super.c | 2 +- include/linux/init_task.h | 2 +- include/linux/mnt_namespace.h | 42 ++++++++++++++++ include/linux/mount.h | 4 +- include/linux/namespace.h | 42 ---------------- include/linux/nsproxy.h | 4 +- kernel/exit.c | 2 +- kernel/fork.c | 21 ++++---- kernel/kmod.c | 2 +- kernel/nsproxy.c | 16 +++--- 16 files changed, 148 insertions(+), 146 deletions(-) create mode 100644 include/linux/mnt_namespace.h delete mode 100644 include/linux/namespace.h (limited to 'include/linux') diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index f33b1a81a761..8f74e8450826 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include "super.h" #include "cell.h" #include "volume.h" diff --git a/fs/namespace.c b/fs/namespace.c index b00ac84ebbdd..fde8553faa76 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include #include #include @@ -133,10 +133,10 @@ struct vfsmount *lookup_mnt(struct vfsmount *mnt, struct dentry *dentry) static inline int check_mnt(struct vfsmount *mnt) { - return mnt->mnt_namespace == current->nsproxy->namespace; + return mnt->mnt_ns == current->nsproxy->mnt_ns; } -static void touch_namespace(struct namespace *ns) +static void touch_mnt_namespace(struct mnt_namespace *ns) { if (ns) { ns->event = ++event; @@ -144,7 +144,7 @@ static void touch_namespace(struct namespace *ns) } } -static void __touch_namespace(struct namespace *ns) +static void __touch_mnt_namespace(struct mnt_namespace *ns) { if (ns && ns->event != event) { ns->event = event; @@ -187,19 +187,19 @@ static void commit_tree(struct vfsmount *mnt) struct vfsmount *parent = mnt->mnt_parent; struct vfsmount *m; LIST_HEAD(head); - struct namespace *n = parent->mnt_namespace; + struct mnt_namespace *n = parent->mnt_ns; BUG_ON(parent == mnt); list_add_tail(&head, &mnt->mnt_list); list_for_each_entry(m, &head, mnt_list) - m->mnt_namespace = n; + m->mnt_ns = n; list_splice(&head, n->list.prev); list_add_tail(&mnt->mnt_hash, mount_hashtable + hash(parent, mnt->mnt_mountpoint)); list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); - touch_namespace(n); + touch_mnt_namespace(n); } static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) @@ -320,7 +320,7 @@ EXPORT_SYMBOL(mnt_unpin); /* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) { - struct namespace *n = m->private; + struct mnt_namespace *n = m->private; struct list_head *p; loff_t l = *pos; @@ -333,7 +333,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) static void *m_next(struct seq_file *m, void *v, loff_t *pos) { - struct namespace *n = m->private; + struct mnt_namespace *n = m->private; struct list_head *p = ((struct vfsmount *)v)->mnt_list.next; (*pos)++; return p == &n->list ? NULL : list_entry(p, struct vfsmount, mnt_list); @@ -526,8 +526,8 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) list_for_each_entry(p, kill, mnt_hash) { list_del_init(&p->mnt_expire); list_del_init(&p->mnt_list); - __touch_namespace(p->mnt_namespace); - p->mnt_namespace = NULL; + __touch_mnt_namespace(p->mnt_ns); + p->mnt_ns = NULL; list_del_init(&p->mnt_child); if (p->mnt_parent != p) p->mnt_mountpoint->d_mounted--; @@ -830,7 +830,7 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, if (parent_nd) { detach_mnt(source_mnt, parent_nd); attach_mnt(source_mnt, nd); - touch_namespace(current->nsproxy->namespace); + touch_mnt_namespace(current->nsproxy->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); commit_tree(source_mnt); @@ -1145,9 +1145,9 @@ static void expire_mount(struct vfsmount *mnt, struct list_head *mounts, */ if (!propagate_mount_busy(mnt, 2)) { /* delete from the namespace */ - touch_namespace(mnt->mnt_namespace); + touch_mnt_namespace(mnt->mnt_ns); list_del_init(&mnt->mnt_list); - mnt->mnt_namespace = NULL; + mnt->mnt_ns = NULL; umount_tree(mnt, 1, umounts); spin_unlock(&vfsmount_lock); } else { @@ -1168,7 +1168,7 @@ static void expire_mount(struct vfsmount *mnt, struct list_head *mounts, */ static void expire_mount_list(struct list_head *graveyard, struct list_head *mounts) { - struct namespace *namespace; + struct mnt_namespace *ns; struct vfsmount *mnt; while (!list_empty(graveyard)) { @@ -1178,10 +1178,10 @@ static void expire_mount_list(struct list_head *graveyard, struct list_head *mou /* don't do anything if the namespace is dead - all the * vfsmounts from it are going away anyway */ - namespace = mnt->mnt_namespace; - if (!namespace || !namespace->root) + ns = mnt->mnt_ns; + if (!ns || !ns->root) continue; - get_namespace(namespace); + get_mnt_ns(ns); spin_unlock(&vfsmount_lock); down_write(&namespace_sem); @@ -1189,7 +1189,7 @@ static void expire_mount_list(struct list_head *graveyard, struct list_head *mou up_write(&namespace_sem); release_mounts(&umounts); mntput(mnt); - put_namespace(namespace); + put_mnt_ns(ns); spin_lock(&vfsmount_lock); } } @@ -1439,14 +1439,15 @@ dput_out: * Allocate a new namespace structure and populate it with contents * copied from the namespace of the passed in task structure. */ -struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs) +struct mnt_namespace *dup_mnt_ns(struct task_struct *tsk, + struct fs_struct *fs) { - struct namespace *namespace = tsk->nsproxy->namespace; - struct namespace *new_ns; + struct mnt_namespace *mnt_ns = tsk->nsproxy->mnt_ns; + struct mnt_namespace *new_ns; struct vfsmount *rootmnt = NULL, *pwdmnt = NULL, *altrootmnt = NULL; struct vfsmount *p, *q; - new_ns = kmalloc(sizeof(struct namespace), GFP_KERNEL); + new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL); if (!new_ns) return NULL; @@ -1457,7 +1458,7 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs) down_write(&namespace_sem); /* First pass: copy the tree topology */ - new_ns->root = copy_tree(namespace->root, namespace->root->mnt_root, + new_ns->root = copy_tree(mnt_ns->root, mnt_ns->root->mnt_root, CL_COPY_ALL | CL_EXPIRE); if (!new_ns->root) { up_write(&namespace_sem); @@ -1473,10 +1474,10 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs) * as belonging to new namespace. We have already acquired a private * fs_struct, so tsk->fs->lock is not needed. */ - p = namespace->root; + p = mnt_ns->root; q = new_ns->root; while (p) { - q->mnt_namespace = new_ns; + q->mnt_ns = new_ns; if (fs) { if (p == fs->rootmnt) { rootmnt = p; @@ -1491,7 +1492,7 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs) fs->altrootmnt = mntget(q); } } - p = next_mnt(p, namespace->root); + p = next_mnt(p, mnt_ns->root); q = next_mnt(q, new_ns->root); } up_write(&namespace_sem); @@ -1506,16 +1507,16 @@ struct namespace *dup_namespace(struct task_struct *tsk, struct fs_struct *fs) return new_ns; } -int copy_namespace(int flags, struct task_struct *tsk) +int copy_mnt_ns(int flags, struct task_struct *tsk) { - struct namespace *namespace = tsk->nsproxy->namespace; - struct namespace *new_ns; + struct mnt_namespace *ns = tsk->nsproxy->mnt_ns; + struct mnt_namespace *new_ns; int err = 0; - if (!namespace) + if (!ns) return 0; - get_namespace(namespace); + get_mnt_ns(ns); if (!(flags & CLONE_NEWNS)) return 0; @@ -1525,16 +1526,16 @@ int copy_namespace(int flags, struct task_struct *tsk) goto out; } - new_ns = dup_namespace(tsk, tsk->fs); + new_ns = dup_mnt_ns(tsk, tsk->fs); if (!new_ns) { err = -ENOMEM; goto out; } - tsk->nsproxy->namespace = new_ns; + tsk->nsproxy->mnt_ns = new_ns; out: - put_namespace(namespace); + put_mnt_ns(ns); return err; } @@ -1754,7 +1755,7 @@ asmlinkage long sys_pivot_root(const char __user * new_root, detach_mnt(user_nd.mnt, &root_parent); attach_mnt(user_nd.mnt, &old_nd); /* mount old root on put_old */ attach_mnt(new_nd.mnt, &root_parent); /* mount new_root on / */ - touch_namespace(current->nsproxy->namespace); + touch_mnt_namespace(current->nsproxy->mnt_ns); spin_unlock(&vfsmount_lock); chroot_fs_refs(&user_nd, &new_nd); security_sb_post_pivotroot(&user_nd, &new_nd); @@ -1779,27 +1780,27 @@ out3: static void __init init_mount_tree(void) { struct vfsmount *mnt; - struct namespace *namespace; + struct mnt_namespace *ns; mnt = do_kern_mount("rootfs", 0, "rootfs", NULL); if (IS_ERR(mnt)) panic("Can't create rootfs"); - namespace = kmalloc(sizeof(*namespace), GFP_KERNEL); - if (!namespace) + ns = kmalloc(sizeof(*ns), GFP_KERNEL); + if (!ns) panic("Can't allocate initial namespace"); - atomic_set(&namespace->count, 1); - INIT_LIST_HEAD(&namespace->list); - init_waitqueue_head(&namespace->poll); - namespace->event = 0; - list_add(&mnt->mnt_list, &namespace->list); - namespace->root = mnt; - mnt->mnt_namespace = namespace; - - init_task.nsproxy->namespace = namespace; - get_namespace(namespace); - - set_fs_pwd(current->fs, namespace->root, namespace->root->mnt_root); - set_fs_root(current->fs, namespace->root, namespace->root->mnt_root); + atomic_set(&ns->count, 1); + INIT_LIST_HEAD(&ns->list); + init_waitqueue_head(&ns->poll); + ns->event = 0; + list_add(&mnt->mnt_list, &ns->list); + ns->root = mnt; + mnt->mnt_ns = ns; + + init_task.nsproxy->mnt_ns = ns; + get_mnt_ns(ns); + + set_fs_pwd(current->fs, ns->root, ns->root->mnt_root); + set_fs_root(current->fs, ns->root, ns->root->mnt_root); } void __init mnt_init(unsigned long mempages) @@ -1860,11 +1861,11 @@ void __init mnt_init(unsigned long mempages) init_mount_tree(); } -void __put_namespace(struct namespace *namespace) +void __put_mnt_ns(struct mnt_namespace *ns) { - struct vfsmount *root = namespace->root; + struct vfsmount *root = ns->root; LIST_HEAD(umount_list); - namespace->root = NULL; + ns->root = NULL; spin_unlock(&vfsmount_lock); down_write(&namespace_sem); spin_lock(&vfsmount_lock); @@ -1872,5 +1873,5 @@ void __put_namespace(struct namespace *namespace) spin_unlock(&vfsmount_lock); up_write(&namespace_sem); release_mounts(&umount_list); - kfree(namespace); + kfree(ns); } diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 20c6f39ea38a..8391bd7a83ce 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -31,7 +31,7 @@ #include #include #include -#include +#include #include #include diff --git a/fs/pnode.c b/fs/pnode.c index da42ee61c1df..56aacead8362 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -6,7 +6,7 @@ * Author : Ram Pai (linuxram@us.ibm.com) * */ -#include +#include #include #include #include "pnode.h" diff --git a/fs/pnode.h b/fs/pnode.h index 020e1bb60fdb..d45bd8ec36bf 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -13,7 +13,7 @@ #define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED) #define IS_MNT_SLAVE(mnt) (mnt->mnt_master) -#define IS_MNT_NEW(mnt) (!mnt->mnt_namespace) +#define IS_MNT_NEW(mnt) (!mnt->mnt_ns) #define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED) #define IS_MNT_UNBINDABLE(mnt) (mnt->mnt_flags & MNT_UNBINDABLE) diff --git a/fs/proc/base.c b/fs/proc/base.c index a71f1755bb57..a3b5074118a7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -59,7 +59,7 @@ #include #include #include -#include +#include #include #include #include @@ -365,33 +365,33 @@ struct proc_mounts { static int mounts_open(struct inode *inode, struct file *file) { struct task_struct *task = get_proc_task(inode); - struct namespace *namespace = NULL; + struct mnt_namespace *ns = NULL; struct proc_mounts *p; int ret = -EINVAL; if (task) { task_lock(task); - namespace = task->nsproxy->namespace; - if (namespace) - get_namespace(namespace); + ns = task->nsproxy->mnt_ns; + if (ns) + get_mnt_ns(ns); task_unlock(task); put_task_struct(task); } - if (namespace) { + if (ns) { ret = -ENOMEM; p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL); if (p) { file->private_data = &p->m; ret = seq_open(file, &mounts_op); if (!ret) { - p->m.private = namespace; - p->event = namespace->event; + p->m.private = ns; + p->event = ns->event; return 0; } kfree(p); } - put_namespace(namespace); + put_mnt_ns(ns); } return ret; } @@ -399,15 +399,15 @@ static int mounts_open(struct inode *inode, struct file *file) static int mounts_release(struct inode *inode, struct file *file) { struct seq_file *m = file->private_data; - struct namespace *namespace = m->private; - put_namespace(namespace); + struct mnt_namespace *ns = m->private; + put_mnt_ns(ns); return seq_release(inode, file); } static unsigned mounts_poll(struct file *file, poll_table *wait) { struct proc_mounts *p = file->private_data; - struct namespace *ns = p->m.private; + struct mnt_namespace *ns = p->m.private; unsigned res = 0; poll_wait(file, &ns->poll, wait); @@ -437,21 +437,21 @@ static int mountstats_open(struct inode *inode, struct file *file) if (!ret) { struct seq_file *m = file->private_data; - struct namespace *namespace = NULL; + struct mnt_namespace *mnt_ns = NULL; struct task_struct *task = get_proc_task(inode); if (task) { task_lock(task); if (task->nsproxy) - namespace = task->nsproxy->namespace; - if (namespace) - get_namespace(namespace); + mnt_ns = task->nsproxy->mnt_ns; + if (mnt_ns) + get_mnt_ns(mnt_ns); task_unlock(task); put_task_struct(task); } - if (namespace) - m->private = namespace; + if (mnt_ns) + m->private = mnt_ns; else { seq_release(inode, file); ret = -EINVAL; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 7fb5fb036f90..58ad4551a7c1 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -23,7 +23,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 848a68af3d42..5c4989172f7e 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -76,7 +76,7 @@ extern struct nsproxy init_nsproxy; .count = ATOMIC_INIT(1), \ .nslock = __SPIN_LOCK_UNLOCKED(nsproxy.nslock), \ .uts_ns = &init_uts_ns, \ - .namespace = NULL, \ + .mnt_ns = NULL, \ INIT_IPC_NS(ipc_ns) \ } diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h new file mode 100644 index 000000000000..4af0b1fc282a --- /dev/null +++ b/include/linux/mnt_namespace.h @@ -0,0 +1,42 @@ +#ifndef _NAMESPACE_H_ +#define _NAMESPACE_H_ +#ifdef __KERNEL__ + +#include +#include +#include + +struct mnt_namespace { + atomic_t count; + struct vfsmount * root; + struct list_head list; + wait_queue_head_t poll; + int event; +}; + +extern int copy_mnt_ns(int, struct task_struct *); +extern void __put_mnt_ns(struct mnt_namespace *ns); +extern struct mnt_namespace *dup_mnt_ns(struct task_struct *, + struct fs_struct *); + +static inline void put_mnt_ns(struct mnt_namespace *ns) +{ + if (atomic_dec_and_lock(&ns->count, &vfsmount_lock)) + /* releases vfsmount_lock */ + __put_mnt_ns(ns); +} + +static inline void exit_mnt_ns(struct task_struct *p) +{ + struct mnt_namespace *ns = p->nsproxy->mnt_ns; + if (ns) + put_mnt_ns(ns); +} + +static inline void get_mnt_ns(struct mnt_namespace *ns) +{ + atomic_inc(&ns->count); +} + +#endif +#endif diff --git a/include/linux/mount.h b/include/linux/mount.h index 403d1a97c512..e357dc86a4de 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -20,7 +20,7 @@ struct super_block; struct vfsmount; struct dentry; -struct namespace; +struct mnt_namespace; #define MNT_NOSUID 0x01 #define MNT_NODEV 0x02 @@ -52,7 +52,7 @@ struct vfsmount { struct list_head mnt_slave_list;/* list of slave mounts */ struct list_head mnt_slave; /* slave list entry */ struct vfsmount *mnt_master; /* slave is on master->mnt_slave_list */ - struct namespace *mnt_namespace; /* containing namespace */ + struct mnt_namespace *mnt_ns; /* containing namespace */ int mnt_pinned; }; diff --git a/include/linux/namespace.h b/include/linux/namespace.h deleted file mode 100644 index d137009f0b2b..000000000000 --- a/include/linux/namespace.h +++ /dev/null @@ -1,42 +0,0 @@ -#ifndef _NAMESPACE_H_ -#define _NAMESPACE_H_ -#ifdef __KERNEL__ - -#include -#include -#include - -struct namespace { - atomic_t count; - struct vfsmount * root; - struct list_head list; - wait_queue_head_t poll; - int event; -}; - -extern int copy_namespace(int, struct task_struct *); -extern void __put_namespace(struct namespace *namespace); -extern struct namespace *dup_namespace(struct task_struct *, struct fs_struct *); - -static inline void put_namespace(struct namespace *namespace) -{ - if (atomic_dec_and_lock(&namespace->count, &vfsmount_lock)) - /* releases vfsmount_lock */ - __put_namespace(namespace); -} - -static inline void exit_namespace(struct task_struct *p) -{ - struct namespace *namespace = p->nsproxy->namespace; - if (namespace) { - put_namespace(namespace); - } -} - -static inline void get_namespace(struct namespace *namespace) -{ - atomic_inc(&namespace->count); -} - -#endif -#endif diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index 971d1c6dfc4b..0aba1b1a39c7 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -4,7 +4,7 @@ #include #include -struct namespace; +struct mnt_namespace; struct uts_namespace; struct ipc_namespace; @@ -25,7 +25,7 @@ struct nsproxy { spinlock_t nslock; struct uts_namespace *uts_ns; struct ipc_namespace *ipc_ns; - struct namespace *namespace; + struct mnt_namespace *mnt_ns; }; extern struct nsproxy init_nsproxy; diff --git a/kernel/exit.c b/kernel/exit.c index 6267a6cc6113..28d9feedfd27 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/fork.c b/kernel/fork.c index 60d2644bfe85..8c859eef8e6a 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include #include #include @@ -1525,17 +1525,18 @@ static int unshare_fs(unsigned long unshare_flags, struct fs_struct **new_fsp) } /* - * Unshare the namespace structure if it is being shared + * Unshare the mnt_namespace structure if it is being shared */ -static int unshare_namespace(unsigned long unshare_flags, struct namespace **new_nsp, struct fs_struct *new_fs) +static int unshare_mnt_namespace(unsigned long unshare_flags, + struct mnt_namespace **new_nsp, struct fs_struct *new_fs) { - struct namespace *ns = current->nsproxy->namespace; + struct mnt_namespace *ns = current->nsproxy->mnt_ns; if ((unshare_flags & CLONE_NEWNS) && ns) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; - *new_nsp = dup_namespace(current, new_fs ? new_fs : current->fs); + *new_nsp = dup_mnt_ns(current, new_fs ? new_fs : current->fs); if (!*new_nsp) return -ENOMEM; } @@ -1623,7 +1624,7 @@ asmlinkage long sys_unshare(unsigned long unshare_flags) { int err = 0; struct fs_struct *fs, *new_fs = NULL; - struct namespace *ns, *new_ns = NULL; + struct mnt_namespace *ns, *new_ns = NULL; struct sighand_struct *new_sigh = NULL; struct mm_struct *mm, *new_mm = NULL, *active_mm = NULL; struct files_struct *fd, *new_fd = NULL; @@ -1645,7 +1646,7 @@ asmlinkage long sys_unshare(unsigned long unshare_flags) goto bad_unshare_out; if ((err = unshare_fs(unshare_flags, &new_fs))) goto bad_unshare_cleanup_thread; - if ((err = unshare_namespace(unshare_flags, &new_ns, new_fs))) + if ((err = unshare_mnt_namespace(unshare_flags, &new_ns, new_fs))) goto bad_unshare_cleanup_fs; if ((err = unshare_sighand(unshare_flags, &new_sigh))) goto bad_unshare_cleanup_ns; @@ -1686,8 +1687,8 @@ asmlinkage long sys_unshare(unsigned long unshare_flags) } if (new_ns) { - ns = current->nsproxy->namespace; - current->nsproxy->namespace = new_ns; + ns = current->nsproxy->mnt_ns; + current->nsproxy->mnt_ns = new_ns; new_ns = ns; } @@ -1748,7 +1749,7 @@ bad_unshare_cleanup_sigh: bad_unshare_cleanup_ns: if (new_ns) - put_namespace(new_ns); + put_mnt_ns(new_ns); bad_unshare_cleanup_fs: if (new_fs) diff --git a/kernel/kmod.c b/kernel/kmod.c index 8d2bea09a4ec..3a7379aa31ca 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -25,7 +25,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 674aceb7335a..bd9cb435dfe0 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #include struct nsproxy init_nsproxy = INIT_NSPROXY(init_nsproxy); @@ -60,8 +60,8 @@ struct nsproxy *dup_namespaces(struct nsproxy *orig) struct nsproxy *ns = clone_namespaces(orig); if (ns) { - if (ns->namespace) - get_namespace(ns->namespace); + if (ns->mnt_ns) + get_mnt_ns(ns->mnt_ns); if (ns->uts_ns) get_uts_ns(ns->uts_ns); if (ns->ipc_ns) @@ -97,7 +97,7 @@ int copy_namespaces(int flags, struct task_struct *tsk) tsk->nsproxy = new_ns; - err = copy_namespace(flags, tsk); + err = copy_mnt_ns(flags, tsk); if (err) goto out_ns; @@ -117,8 +117,8 @@ out_ipc: if (new_ns->uts_ns) put_uts_ns(new_ns->uts_ns); out_uts: - if (new_ns->namespace) - put_namespace(new_ns->namespace); + if (new_ns->mnt_ns) + put_mnt_ns(new_ns->mnt_ns); out_ns: tsk->nsproxy = old_ns; kfree(new_ns); @@ -127,8 +127,8 @@ out_ns: void free_nsproxy(struct nsproxy *ns) { - if (ns->namespace) - put_namespace(ns->namespace); + if (ns->mnt_ns) + put_mnt_ns(ns->mnt_ns); if (ns->uts_ns) put_uts_ns(ns->uts_ns); if (ns->ipc_ns) -- cgit v1.2.3 From 373beb35cd6b625e0ba4ad98baace12310a26aa8 Mon Sep 17 00:00:00 2001 From: Cedric Le Goater Date: Fri, 8 Dec 2006 02:37:57 -0800 Subject: [PATCH] identifier to nsproxy Add an identifier to nsproxy. The default init_ns_proxy has identifier 0 and allocated nsproxies are given -1. This identifier will be used by a new syscall sys_bind_ns. Signed-off-by: Cedric Le Goater Cc: Kirill Korotaev Cc: Eric W. Biederman Cc: Herbert Poetzl Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init_task.h | 1 + include/linux/nsproxy.h | 1 + kernel/nsproxy.c | 4 +++- 3 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 5c4989172f7e..90c5f9a07730 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -75,6 +75,7 @@ extern struct nsproxy init_nsproxy; #define INIT_NSPROXY(nsproxy) { \ .count = ATOMIC_INIT(1), \ .nslock = __SPIN_LOCK_UNLOCKED(nsproxy.nslock), \ + .id = 0, \ .uts_ns = &init_uts_ns, \ .mnt_ns = NULL, \ INIT_IPC_NS(ipc_ns) \ diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index 0aba1b1a39c7..27f37c1ec1d9 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -23,6 +23,7 @@ struct ipc_namespace; struct nsproxy { atomic_t count; spinlock_t nslock; + unsigned long id; struct uts_namespace *uts_ns; struct ipc_namespace *ipc_ns; struct mnt_namespace *mnt_ns; diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index bd9cb435dfe0..f223c15c18e9 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -45,8 +45,10 @@ static inline struct nsproxy *clone_namespaces(struct nsproxy *orig) struct nsproxy *ns; ns = kmemdup(orig, sizeof(struct nsproxy), GFP_KERNEL); - if (ns) + if (ns) { atomic_set(&ns->count, 1); + ns->id = -1; + } return ns; } -- cgit v1.2.3 From 61a58c6c238cc81f7742b8cc84212cc55fb57747 Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Fri, 8 Dec 2006 02:37:58 -0800 Subject: [PATCH] rename struct pspace to struct pid_namespace Rename struct pspace to struct pid_namespace for consistency with other namespaces (uts_namespace and ipc_namespace). Also rename include/linux/pspace.h to include/linux/pid_namespace.h and variables from pspace to pid_ns. Signed-off-by: Sukadev Bhattiprolu Signed-off-by: Cedric Le Goater Cc: Kirill Korotaev Cc: Eric W. Biederman Cc: Herbert Poetzl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_misc.c | 4 ++-- include/linux/pid_namespace.h | 23 ++++++++++++++++++++ include/linux/pspace.h | 23 -------------------- kernel/pid.c | 49 ++++++++++++++++++++++--------------------- 4 files changed, 50 insertions(+), 49 deletions(-) create mode 100644 include/linux/pid_namespace.h delete mode 100644 include/linux/pspace.h (limited to 'include/linux') diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 9397ff62553e..89ccfa16b000 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -45,7 +45,7 @@ #include #include #include -#include +#include #include #include #include @@ -92,7 +92,7 @@ static int loadavg_read_proc(char *page, char **start, off_t off, LOAD_INT(a), LOAD_FRAC(a), LOAD_INT(b), LOAD_FRAC(b), LOAD_INT(c), LOAD_FRAC(c), - nr_running(), nr_threads, init_pspace.last_pid); + nr_running(), nr_threads, init_pid_ns.last_pid); return proc_calc_metrics(page, start, off, count, eof, len); } diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h new file mode 100644 index 000000000000..54d79095e295 --- /dev/null +++ b/include/linux/pid_namespace.h @@ -0,0 +1,23 @@ +#ifndef _LINUX_PID_NS_H +#define _LINUX_PID_NS_H + +#include +#include +#include +#include + +struct pidmap { + atomic_t nr_free; + void *page; +}; + +#define PIDMAP_ENTRIES ((PID_MAX_LIMIT + 8*PAGE_SIZE - 1)/PAGE_SIZE/8) + +struct pid_namespace { + struct pidmap pidmap[PIDMAP_ENTRIES]; + int last_pid; +}; + +extern struct pid_namespace init_pid_ns; + +#endif /* _LINUX_PID_NS_H */ diff --git a/include/linux/pspace.h b/include/linux/pspace.h deleted file mode 100644 index 91d48b8b2d99..000000000000 --- a/include/linux/pspace.h +++ /dev/null @@ -1,23 +0,0 @@ -#ifndef _LINUX_PSPACE_H -#define _LINUX_PSPACE_H - -#include -#include -#include -#include - -struct pidmap { - atomic_t nr_free; - void *page; -}; - -#define PIDMAP_ENTRIES ((PID_MAX_LIMIT + 8*PAGE_SIZE - 1)/PAGE_SIZE/8) - -struct pspace { - struct pidmap pidmap[PIDMAP_ENTRIES]; - int last_pid; -}; - -extern struct pspace init_pspace; - -#endif /* _LINUX_PSPACE_H */ diff --git a/kernel/pid.c b/kernel/pid.c index a48879b0b921..25807e1b98dd 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -26,7 +26,7 @@ #include #include #include -#include +#include #define pid_hashfn(nr) hash_long((unsigned long)nr, pidhash_shift) static struct hlist_head *pid_hash; @@ -43,9 +43,10 @@ int pid_max_max = PID_MAX_LIMIT; #define BITS_PER_PAGE (PAGE_SIZE*8) #define BITS_PER_PAGE_MASK (BITS_PER_PAGE-1) -static inline int mk_pid(struct pspace *pspace, struct pidmap *map, int off) +static inline int mk_pid(struct pid_namespace *pid_ns, + struct pidmap *map, int off) { - return (map - pspace->pidmap)*BITS_PER_PAGE + off; + return (map - pid_ns->pidmap)*BITS_PER_PAGE + off; } #define find_next_offset(map, off) \ @@ -57,7 +58,7 @@ static inline int mk_pid(struct pspace *pspace, struct pidmap *map, int off) * value does not cause lots of bitmaps to be allocated, but * the scheme scales to up to 4 million PIDs, runtime. */ -struct pspace init_pspace = { +struct pid_namespace init_pid_ns = { .pidmap = { [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } }, @@ -80,25 +81,25 @@ struct pspace init_pspace = { static __cacheline_aligned_in_smp DEFINE_SPINLOCK(pidmap_lock); -static fastcall void free_pidmap(struct pspace *pspace, int pid) +static fastcall void free_pidmap(struct pid_namespace *pid_ns, int pid) { - struct pidmap *map = pspace->pidmap + pid / BITS_PER_PAGE; + struct pidmap *map = pid_ns->pidmap + pid / BITS_PER_PAGE; int offset = pid & BITS_PER_PAGE_MASK; clear_bit(offset, map->page); atomic_inc(&map->nr_free); } -static int alloc_pidmap(struct pspace *pspace) +static int alloc_pidmap(struct pid_namespace *pid_ns) { - int i, offset, max_scan, pid, last = pspace->last_pid; + int i, offset, max_scan, pid, last = pid_ns->last_pid; struct pidmap *map; pid = last + 1; if (pid >= pid_max) pid = RESERVED_PIDS; offset = pid & BITS_PER_PAGE_MASK; - map = &pspace->pidmap[pid/BITS_PER_PAGE]; + map = &pid_ns->pidmap[pid/BITS_PER_PAGE]; max_scan = (pid_max + BITS_PER_PAGE - 1)/BITS_PER_PAGE - !offset; for (i = 0; i <= max_scan; ++i) { if (unlikely(!map->page)) { @@ -120,11 +121,11 @@ static int alloc_pidmap(struct pspace *pspace) do { if (!test_and_set_bit(offset, map->page)) { atomic_dec(&map->nr_free); - pspace->last_pid = pid; + pid_ns->last_pid = pid; return pid; } offset = find_next_offset(map, offset); - pid = mk_pid(pspace, map, offset); + pid = mk_pid(pid_ns, map, offset); /* * find_next_offset() found a bit, the pid from it * is in-bounds, and if we fell back to the last @@ -135,34 +136,34 @@ static int alloc_pidmap(struct pspace *pspace) (i != max_scan || pid < last || !((last+1) & BITS_PER_PAGE_MASK))); } - if (map < &pspace->pidmap[(pid_max-1)/BITS_PER_PAGE]) { + if (map < &pid_ns->pidmap[(pid_max-1)/BITS_PER_PAGE]) { ++map; offset = 0; } else { - map = &pspace->pidmap[0]; + map = &pid_ns->pidmap[0]; offset = RESERVED_PIDS; if (unlikely(last == offset)) break; } - pid = mk_pid(pspace, map, offset); + pid = mk_pid(pid_ns, map, offset); } return -1; } -static int next_pidmap(struct pspace *pspace, int last) +static int next_pidmap(struct pid_namespace *pid_ns, int last) { int offset; struct pidmap *map, *end; offset = (last + 1) & BITS_PER_PAGE_MASK; - map = &pspace->pidmap[(last + 1)/BITS_PER_PAGE]; - end = &pspace->pidmap[PIDMAP_ENTRIES]; + map = &pid_ns->pidmap[(last + 1)/BITS_PER_PAGE]; + end = &pid_ns->pidmap[PIDMAP_ENTRIES]; for (; map < end; map++, offset = 0) { if (unlikely(!map->page)) continue; offset = find_next_bit((map)->page, BITS_PER_PAGE, offset); if (offset < BITS_PER_PAGE) - return mk_pid(pspace, map, offset); + return mk_pid(pid_ns, map, offset); } return -1; } @@ -192,7 +193,7 @@ fastcall void free_pid(struct pid *pid) hlist_del_rcu(&pid->pid_chain); spin_unlock_irqrestore(&pidmap_lock, flags); - free_pidmap(&init_pspace, pid->nr); + free_pidmap(&init_pid_ns, pid->nr); call_rcu(&pid->rcu, delayed_put_pid); } @@ -206,7 +207,7 @@ struct pid *alloc_pid(void) if (!pid) goto out; - nr = alloc_pidmap(&init_pspace); + nr = alloc_pidmap(&init_pid_ns); if (nr < 0) goto out_free; @@ -348,7 +349,7 @@ struct pid *find_ge_pid(int nr) pid = find_pid(nr); if (pid) break; - nr = next_pidmap(&init_pspace, nr); + nr = next_pidmap(&init_pid_ns, nr); } while (nr > 0); return pid; @@ -382,10 +383,10 @@ void __init pidhash_init(void) void __init pidmap_init(void) { - init_pspace.pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL); + init_pid_ns.pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL); /* Reserve PID 0. We never call free_pidmap(0) */ - set_bit(0, init_pspace.pidmap[0].page); - atomic_dec(&init_pspace.pidmap[0].nr_free); + set_bit(0, init_pid_ns.pidmap[0].page); + atomic_dec(&init_pid_ns.pidmap[0].nr_free); pid_cachep = kmem_cache_create("pid", sizeof(struct pid), __alignof__(struct pid), -- cgit v1.2.3 From 9a575a92db3312a40cdf0b0406d88de88ad9741e Mon Sep 17 00:00:00 2001 From: Cedric Le Goater Date: Fri, 8 Dec 2006 02:37:59 -0800 Subject: [PATCH] to nsproxy Add the pid namespace framework to the nsproxy object. The copy of the pid namespace only increases the refcount on the global pid namespace, init_pid_ns, and unshare is not implemented. There is no configuration option to activate or deactivate this feature because this not relevant for the moment. Signed-off-by: Cedric Le Goater Cc: Kirill Korotaev Cc: Eric W. Biederman Cc: Herbert Poetzl Cc: Sukadev Bhattiprolu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init_task.h | 2 ++ include/linux/nsproxy.h | 2 ++ include/linux/pid_namespace.h | 20 ++++++++++++++++++-- kernel/nsproxy.c | 26 +++++++++++++++++++------- kernel/pid.c | 23 +++++++++++++++++++++++ 5 files changed, 64 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 90c5f9a07730..7272ff9ee77c 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -7,6 +7,7 @@ #include #include #include +#include #define INIT_FDTABLE \ { \ @@ -73,6 +74,7 @@ extern struct nsproxy init_nsproxy; #define INIT_NSPROXY(nsproxy) { \ + .pid_ns = &init_pid_ns, \ .count = ATOMIC_INIT(1), \ .nslock = __SPIN_LOCK_UNLOCKED(nsproxy.nslock), \ .id = 0, \ diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index 27f37c1ec1d9..fdfb0e44912f 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -7,6 +7,7 @@ struct mnt_namespace; struct uts_namespace; struct ipc_namespace; +struct pid_namespace; /* * A structure to contain pointers to all per-process @@ -27,6 +28,7 @@ struct nsproxy { struct uts_namespace *uts_ns; struct ipc_namespace *ipc_ns; struct mnt_namespace *mnt_ns; + struct pid_namespace *pid_ns; }; extern struct nsproxy init_nsproxy; diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index 54d79095e295..76e7c6b2cf33 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -5,6 +5,8 @@ #include #include #include +#include +#include struct pidmap { atomic_t nr_free; @@ -14,10 +16,24 @@ struct pidmap { #define PIDMAP_ENTRIES ((PID_MAX_LIMIT + 8*PAGE_SIZE - 1)/PAGE_SIZE/8) struct pid_namespace { - struct pidmap pidmap[PIDMAP_ENTRIES]; - int last_pid; + struct kref kref; + struct pidmap pidmap[PIDMAP_ENTRIES]; + int last_pid; }; extern struct pid_namespace init_pid_ns; +static inline void get_pid_ns(struct pid_namespace *ns) +{ + kref_get(&ns->kref); +} + +extern int copy_pid_ns(int flags, struct task_struct *tsk); +extern void free_pid_ns(struct kref *kref); + +static inline void put_pid_ns(struct pid_namespace *ns) +{ + kref_put(&ns->kref, free_pid_ns); +} + #endif /* _LINUX_PID_NS_H */ diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index f223c15c18e9..e2ce748e96af 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -19,6 +19,7 @@ #include #include #include +#include struct nsproxy init_nsproxy = INIT_NSPROXY(init_nsproxy); @@ -68,6 +69,8 @@ struct nsproxy *dup_namespaces(struct nsproxy *orig) get_uts_ns(ns->uts_ns); if (ns->ipc_ns) get_ipc_ns(ns->ipc_ns); + if (ns->pid_ns) + get_pid_ns(ns->pid_ns); } return ns; @@ -111,10 +114,17 @@ int copy_namespaces(int flags, struct task_struct *tsk) if (err) goto out_ipc; + err = copy_pid_ns(flags, tsk); + if (err) + goto out_pid; + out: put_nsproxy(old_ns); return err; +out_pid: + if (new_ns->ipc_ns) + put_ipc_ns(new_ns->ipc_ns); out_ipc: if (new_ns->uts_ns) put_uts_ns(new_ns->uts_ns); @@ -129,11 +139,13 @@ out_ns: void free_nsproxy(struct nsproxy *ns) { - if (ns->mnt_ns) - put_mnt_ns(ns->mnt_ns); - if (ns->uts_ns) - put_uts_ns(ns->uts_ns); - if (ns->ipc_ns) - put_ipc_ns(ns->ipc_ns); - kfree(ns); + if (ns->mnt_ns) + put_mnt_ns(ns->mnt_ns); + if (ns->uts_ns) + put_uts_ns(ns->uts_ns); + if (ns->ipc_ns) + put_ipc_ns(ns->ipc_ns); + if (ns->pid_ns) + put_pid_ns(ns->pid_ns); + kfree(ns); } diff --git a/kernel/pid.c b/kernel/pid.c index 25807e1b98dd..5319b9f2fc5e 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -59,6 +59,9 @@ static inline int mk_pid(struct pid_namespace *pid_ns, * the scheme scales to up to 4 million PIDs, runtime. */ struct pid_namespace init_pid_ns = { + .kref = { + .refcount = ATOMIC_INIT(2), + }, .pidmap = { [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } }, @@ -356,6 +359,26 @@ struct pid *find_ge_pid(int nr) } EXPORT_SYMBOL_GPL(find_get_pid); +int copy_pid_ns(int flags, struct task_struct *tsk) +{ + struct pid_namespace *old_ns = tsk->nsproxy->pid_ns; + int err = 0; + + if (!old_ns) + return 0; + + get_pid_ns(old_ns); + return err; +} + +void free_pid_ns(struct kref *kref) +{ + struct pid_namespace *ns; + + ns = container_of(kref, struct pid_namespace, kref); + kfree(ns); +} + /* * The pid hash table is scaled according to the amount of memory in the * machine. From a minimum of 16 slots up to 4096 slots at one gigabyte or -- cgit v1.2.3 From 84d737866e2babdeab0c6b18ea155c6a649663b8 Mon Sep 17 00:00:00 2001 From: Sukadev Bhattiprolu Date: Fri, 8 Dec 2006 02:38:01 -0800 Subject: [PATCH] add child reaper to pid_namespace Add a per pid_namespace child-reaper. This is needed so processes are reaped within the same pid space and do not spill over to the parent pid space. Its also needed so containers preserve existing semantic that pid == 1 would reap orphaned children. This is based on Eric Biederman's patch: http://lkml.org/lkml/2006/2/6/285 Signed-off-by: Sukadev Bhattiprolu Signed-off-by: Cedric Le Goater Cc: Kirill Korotaev Cc: Eric W. Biederman Cc: Herbert Poetzl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 5 +++-- include/linux/pid.h | 5 +++-- include/linux/pid_namespace.h | 6 ++++++ include/linux/sched.h | 1 - init/main.c | 5 ++--- kernel/exit.c | 23 +++++++++++++++-------- kernel/pid.c | 3 ++- kernel/signal.c | 11 +++++++++-- 8 files changed, 40 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/fs/exec.c b/fs/exec.c index 60433e2254a4..12d8cd461b41 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -620,8 +621,8 @@ static int de_thread(struct task_struct *tsk) * Reparenting needs write_lock on tasklist_lock, * so it is safe to do it under read_lock. */ - if (unlikely(tsk->group_leader == child_reaper)) - child_reaper = tsk; + if (unlikely(tsk->group_leader == child_reaper(tsk))) + tsk->nsproxy->pid_ns->child_reaper = tsk; zap_other_threads(tsk); read_unlock(&tasklist_lock); diff --git a/include/linux/pid.h b/include/linux/pid.h index 2c0007d17218..4dec047b1837 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -35,8 +35,9 @@ enum pid_type * * Holding a reference to struct pid solves both of these problems. * It is small so holding a reference does not consume a lot of - * resources, and since a new struct pid is allocated when the numeric - * pid value is reused we don't mistakenly refer to new processes. + * resources, and since a new struct pid is allocated when the numeric pid + * value is reused (when pids wrap around) we don't mistakenly refer to new + * processes. */ struct pid diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index 76e7c6b2cf33..d2a9d419f01f 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -19,6 +19,7 @@ struct pid_namespace { struct kref kref; struct pidmap pidmap[PIDMAP_ENTRIES]; int last_pid; + struct task_struct *child_reaper; }; extern struct pid_namespace init_pid_ns; @@ -36,4 +37,9 @@ static inline void put_pid_ns(struct pid_namespace *ns) kref_put(&ns->kref, free_pid_ns); } +static inline struct task_struct *child_reaper(struct task_struct *tsk) +{ + return tsk->nsproxy->pid_ns->child_reaper; +} + #endif /* _LINUX_PID_NS_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 6fec1d419714..f0317edea141 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1400,7 +1400,6 @@ extern NORET_TYPE void do_group_exit(int); extern void daemonize(const char *, ...); extern int allow_signal(int); extern int disallow_signal(int); -extern struct task_struct *child_reaper; extern int do_execve(char *, char __user * __user *, char __user * __user *, struct pt_regs *); extern long do_fork(unsigned long, unsigned long, struct pt_regs *, unsigned long, int __user *, int __user *); diff --git a/init/main.c b/init/main.c index 4cdcd06e6d78..036f97c0c34c 100644 --- a/init/main.c +++ b/init/main.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include @@ -626,8 +627,6 @@ static int __init initcall_debug_setup(char *str) } __setup("initcall_debug", initcall_debug_setup); -struct task_struct *child_reaper = &init_task; - extern initcall_t __initcall_start[], __initcall_end[]; static void __init do_initcalls(void) @@ -727,7 +726,7 @@ static int init(void * unused) * assumptions about where in the task array this * can be found. */ - child_reaper = current; + init_pid_ns.child_reaper = current; cad_pid = task_pid(current); diff --git a/kernel/exit.c b/kernel/exit.c index 28d9feedfd27..fd0e067952ab 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -48,7 +49,6 @@ #include extern void sem_exit (void); -extern struct task_struct *child_reaper; static void exit_mm(struct task_struct * tsk); @@ -260,7 +260,8 @@ static int has_stopped_jobs(int pgrp) } /** - * reparent_to_init - Reparent the calling kernel thread to the init task. + * reparent_to_init - Reparent the calling kernel thread to the init task + * of the pid space that the thread belongs to. * * If a kernel thread is launched as a result of a system call, or if * it ever exits, it should generally reparent itself to init so that @@ -278,8 +279,8 @@ static void reparent_to_init(void) ptrace_unlink(current); /* Reparent to init */ remove_parent(current); - current->parent = child_reaper; - current->real_parent = child_reaper; + current->parent = child_reaper(current); + current->real_parent = child_reaper(current); add_parent(current); /* Set the exit signal to SIGCHLD so we signal init on exit */ @@ -662,7 +663,8 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced) * When we die, we re-parent all our children. * Try to give them to another thread in our thread * group, and if no such member exists, give it to - * the global child reaper process (ie "init") + * the child reaper process (ie "init") in our pid + * space. */ static void forget_original_parent(struct task_struct *father, struct list_head *to_release) @@ -673,7 +675,7 @@ forget_original_parent(struct task_struct *father, struct list_head *to_release) do { reaper = next_thread(reaper); if (reaper == father) { - reaper = child_reaper; + reaper = child_reaper(father); break; } } while (reaper->exit_state); @@ -859,8 +861,13 @@ fastcall NORET_TYPE void do_exit(long code) panic("Aiee, killing interrupt handler!"); if (unlikely(!tsk->pid)) panic("Attempted to kill the idle task!"); - if (unlikely(tsk == child_reaper)) - panic("Attempted to kill init!"); + if (unlikely(tsk == child_reaper(tsk))) { + if (tsk->nsproxy->pid_ns != &init_pid_ns) + tsk->nsproxy->pid_ns->child_reaper = init_pid_ns.child_reaper; + else + panic("Attempted to kill init!"); + } + if (unlikely(current->ptrace & PT_TRACE_EXIT)) { current->ptrace_message = code; diff --git a/kernel/pid.c b/kernel/pid.c index 1d9cc268b499..2efe9d8d367b 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -65,7 +65,8 @@ struct pid_namespace init_pid_ns = { .pidmap = { [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } }, - .last_pid = 0 + .last_pid = 0, + .child_reaper = &init_task }; /* diff --git a/kernel/signal.c b/kernel/signal.c index 9eac4db60eda..1921ffdc5e77 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -24,6 +24,9 @@ #include #include #include +#include +#include + #include #include #include @@ -1877,8 +1880,12 @@ relock: if (sig_kernel_ignore(signr)) /* Default is nothing. */ continue; - /* Init gets no signals it doesn't want. */ - if (current == child_reaper) + /* + * Init of a pid space gets no signals it doesn't want from + * within that pid space. It can of course get signals from + * its parent pid space. + */ + if (current == child_reaper(current)) continue; if (sig_kernel_stop(signr)) { -- cgit v1.2.3 From 3306ce3d0554e2e59cc429b7133e17e1513307cb Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 8 Dec 2006 02:38:13 -0800 Subject: [PATCH] Char: mxser_new, upgrade to 1.9.1 Change cloned experimental driver according to original 1.9.1 moxa driver. Some int->ulong conversions, outb ~UART_IER_THRI constant. Remove commented stuff. I also added printk line with info, if somebody wants to test it, he may contact me as I can potentially debug the driver with him or just to confirm it works properly. Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/mxser_new.c | 396 +++++++++++++++++++++++------------------------ include/linux/pci_ids.h | 3 + 2 files changed, 195 insertions(+), 204 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/mxser_new.c b/drivers/char/mxser_new.c index 993e3a26c406..befff0548da3 100644 --- a/drivers/char/mxser_new.c +++ b/drivers/char/mxser_new.c @@ -1,7 +1,7 @@ /* * mxser.c -- MOXA Smartio/Industio family multiport serial driver. * - * Copyright (C) 1999-2001 Moxa Technologies (support@moxa.com.tw). + * Copyright (C) 1999-2006 Moxa Technologies (support@moxa.com.tw). * * This code is loosely based on the Linux serial driver, written by * Linus Torvalds, Theodore T'so and others. @@ -20,15 +20,6 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * Original release 10/26/00 - * - * 02/06/01 Support MOXA Industio family boards. - * 02/06/01 Support TIOCGICOUNT. - * 02/06/01 Fix the problem for connecting to serial mouse. - * 02/06/01 Fix the problem for H/W flow control. - * 02/06/01 Fix the compling warning when CONFIG_PCI - * don't be defined. - * * Fed through a cleanup, indent and remove of non 2.6 code by Alan Cox * . The original 1.8 code is available on www.moxa.com. * - Fixed x86_64 cleanness @@ -66,7 +57,7 @@ #include "mxser_new.h" -#define MXSER_VERSION "1.8" +#define MXSER_VERSION "1.9.1" #define MXSERMAJOR 174 #define MXSERCUMAJOR 175 @@ -76,7 +67,7 @@ #define MXSER_BOARDS 4 /* Max. boards */ #define MXSER_PORTS 32 /* Max. ports */ #define MXSER_PORTS_PER_BOARD 8 /* Max. ports per board */ -#define MXSER_ISR_PASS_LIMIT 256 +#define MXSER_ISR_PASS_LIMIT 99999L #define MXSER_ERR_IOADDR -1 #define MXSER_ERR_IRQ -2 @@ -125,6 +116,9 @@ enum { MXSER_BOARD_CP118U, MXSER_BOARD_CP102UL, MXSER_BOARD_CP102U, + MXSER_BOARD_CP118EL, + MXSER_BOARD_CP168EL, + MXSER_BOARD_CP104EL }; static char *mxser_brdname[] = { @@ -149,6 +143,9 @@ static char *mxser_brdname[] = { "CP-118U series", "CP-102UL series", "CP-102U series", + "CP-118EL series", + "CP-168EL series", + "CP-104EL series" }; static int mxser_numports[] = { @@ -173,6 +170,9 @@ static int mxser_numports[] = { 8, /* CP118U */ 2, /* CP102UL */ 2, /* CP102U */ + 8, /* CP118EL */ + 8, /* CP168EL */ + 4 /* CP104EL */ }; #define UART_TYPE_NUM 2 @@ -205,22 +205,43 @@ static const struct mxpciuart_info Gpci_uart_info[UART_INFO_NUM] = { #ifdef CONFIG_PCI static struct pci_device_id mxser_pcibrds[] = { - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_C168, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_C168_PCI}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_C104, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_C104_PCI}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP132, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP132}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP114, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP114}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CT114, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CT114}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP102}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP104U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP104U}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP168U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP168U}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP132U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP132U}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP134U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP134U}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP104JU, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP104JU}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_RC7000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_RC7000}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP118U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP118U}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102UL, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP102UL}, - {PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102U, PCI_ANY_ID, PCI_ANY_ID, 0, 0, MXSER_BOARD_CP102U}, - {0} + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_C168), + .driver_data = MXSER_BOARD_C168_PCI }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_C104), + .driver_data = MXSER_BOARD_C104_PCI }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP132), + .driver_data = MXSER_BOARD_CP132 }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP114), + .driver_data = MXSER_BOARD_CP114 }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CT114), + .driver_data = MXSER_BOARD_CT114 }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102), + .driver_data = MXSER_BOARD_CP102 }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP104U), + .driver_data = MXSER_BOARD_CP104U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP168U), + .driver_data = MXSER_BOARD_CP168U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP132U), + .driver_data = MXSER_BOARD_CP132U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP134U), + .driver_data = MXSER_BOARD_CP134U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP104JU), + .driver_data = MXSER_BOARD_CP104JU }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_RC7000), + .driver_data = MXSER_BOARD_RC7000 }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP118U), + .driver_data = MXSER_BOARD_CP118U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102UL), + .driver_data = MXSER_BOARD_CP102UL }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP102U), + .driver_data = MXSER_BOARD_CP102U }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP118EL), + .driver_data = MXSER_BOARD_CP118EL }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP168EL), + .driver_data = MXSER_BOARD_CP168EL }, + { PCI_DEVICE(PCI_VENDOR_ID_MOXA, PCI_DEVICE_ID_MOXA_CP104EL), + .driver_data = MXSER_BOARD_CP104EL }, + { } }; MODULE_DEVICE_TABLE(pci, mxser_pcibrds); @@ -245,7 +266,6 @@ MODULE_AUTHOR("Casper Yang"); MODULE_DESCRIPTION("MOXA Smartio/Industio Family Multiport Board Device Driver"); module_param_array(ioaddr, int, NULL, 0); module_param(ttymajor, int, 0); -module_param(calloutmajor, int, 0); module_param(verbose, bool, 0); MODULE_LICENSE("GPL"); @@ -285,23 +305,23 @@ struct mxser_hwconf { int board_type; int ports; int irq; - int vector; - int vector_mask; + unsigned long vector; + unsigned long vector_mask; int uart_type; - int ioaddr[MXSER_PORTS_PER_BOARD]; + unsigned long ioaddr[MXSER_PORTS_PER_BOARD]; int baud_base[MXSER_PORTS_PER_BOARD]; moxa_pci_info pciInfo; int IsMoxaMustChipFlag; /* add by Victor Yu. 08-30-2002 */ int MaxCanSetBaudRate[MXSER_PORTS_PER_BOARD]; /* add by Victor Yu. 09-04-2002 */ - int opmode_ioaddr[MXSER_PORTS_PER_BOARD]; /* add by Victor Yu. 01-05-2004 */ + unsigned long opmode_ioaddr[MXSER_PORTS_PER_BOARD]; /* add by Victor Yu. 01-05-2004 */ }; struct mxser_struct { int port; - int base; /* port base address */ + unsigned long base; /* port base address */ int irq; /* port using irq no. */ - int vector; /* port irq vector */ - int vectormask; /* port vector mask */ + unsigned long vector; /* port irq vector */ + unsigned long vectormask; /* port vector mask */ int rx_high_water; int rx_trigger; /* Rx fifo trigger level */ int rx_low_water; @@ -337,7 +357,7 @@ struct mxser_struct { int timeout; int IsMoxaMustChipFlag; /* add by Victor Yu. 08-30-2002 */ int MaxCanSetBaudRate; /* add by Victor Yu. 09-04-2002 */ - int opmode_ioaddr; /* add by Victor Yu. 01-05-2004 */ + unsigned long opmode_ioaddr; /* add by Victor Yu. 01-05-2004 */ unsigned char stop_rx; unsigned char ldisc_stop_rx; long realbaud; @@ -485,6 +505,9 @@ static int __init mxser_module_init(void) if (verbose) printk(KERN_DEBUG "Loading module mxser ...\n"); + printk(KERN_INFO "This is mxser driver version 1.9.1 and needs TESTING." + "If your are willing to test this driver, please report to " + "jirislaby@gmail.com. Thanks.\n"); ret = mxser_init(); if (verbose) printk(KERN_DEBUG "Done.\n"); @@ -636,7 +659,7 @@ static int mxser_get_PCI_conf(int busnum, int devnum, int board_type, struct mxs { int i, j; /* unsigned int val; */ - unsigned int ioaddress; + unsigned long ioaddress; struct pci_dev *pdev = hwconf->pciInfo.pdev; /* io address */ @@ -790,7 +813,7 @@ static int mxser_init(void) /* Start finding ISA boards from module arg */ for (b = 0; b < MXSER_BOARDS && m < MXSER_BOARDS; b++) { - int cap; + unsigned long cap; if (!(cap = ioaddr[b])) continue; @@ -928,12 +951,10 @@ static void mxser_do_softint(void *private_) tty = info->tty; - if (tty) { - if (test_and_clear_bit(MXSER_EVENT_TXLOW, &info->event)) - tty_wakeup(tty); - if (test_and_clear_bit(MXSER_EVENT_HANGUP, &info->event)) - tty_hangup(tty); - } + if (test_and_clear_bit(MXSER_EVENT_TXLOW, &info->event)) + tty_wakeup(tty); + if (test_and_clear_bit(MXSER_EVENT_HANGUP, &info->event)) + tty_hangup(tty); } static unsigned char mxser_get_msr(int baseaddr, int mode, int port, struct mxser_struct *info) @@ -979,6 +1000,7 @@ static int mxser_open(struct tty_struct *tty, struct file *filp) /* * Start up serial port */ + info->count++; retval = mxser_startup(info); if (retval) return retval; @@ -987,8 +1009,6 @@ static int mxser_open(struct tty_struct *tty, struct file *filp) if (retval) return retval; - info->count++; - if ((info->count == 1) && (info->flags & ASYNC_SPLIT_TERMIOS)) { if (tty->driver->subtype == SERIAL_TYPE_NORMAL) *tty->termios = info->normal_termios; @@ -1150,11 +1170,13 @@ static int mxser_write(struct tty_struct *tty, const unsigned char *buf, int cou total += c; } - if (info->xmit_cnt && !tty->stopped && !(info->IER & UART_IER_THRI)) { + if (info->xmit_cnt && !tty->stopped + /*&& !(info->IER & UART_IER_THRI)*/) { if (!tty->hw_stopped || (info->type == PORT_16550A) || (info->IsMoxaMustChipFlag)) { spin_lock_irqsave(&info->slock, flags); + outb(info->IER & ~UART_IER_THRI, info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); spin_unlock_irqrestore(&info->slock, flags); @@ -1179,11 +1201,12 @@ static void mxser_put_char(struct tty_struct *tty, unsigned char ch) info->xmit_head &= SERIAL_XMIT_SIZE - 1; info->xmit_cnt++; spin_unlock_irqrestore(&info->slock, flags); - if (!tty->stopped && !(info->IER & UART_IER_THRI)) { + if (!tty->stopped /*&& !(info->IER & UART_IER_THRI)*/) { if (!tty->hw_stopped || (info->type == PORT_16550A) || info->IsMoxaMustChipFlag) { spin_lock_irqsave(&info->slock, flags); + outb(info->IER & ~UART_IER_THRI, info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); spin_unlock_irqrestore(&info->slock, flags); @@ -1208,6 +1231,7 @@ static void mxser_flush_chars(struct tty_struct *tty) spin_lock_irqsave(&info->slock, flags); + outb(info->IER & ~UART_IER_THRI, info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); @@ -1228,7 +1252,12 @@ static int mxser_write_room(struct tty_struct *tty) static int mxser_chars_in_buffer(struct tty_struct *tty) { struct mxser_struct *info = tty->driver_data; - return info->xmit_cnt; + int len = info->xmit_cnt; + + if (!(inb(info->base + UART_LSR) & UART_LSR_THRE)) + len++; + + return len; } static void mxser_flush_buffer(struct tty_struct *tty) @@ -1270,7 +1299,8 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file, unsigned int c /* following add by Victor Yu. 01-05-2004 */ if (cmd == MOXA_SET_OP_MODE || cmd == MOXA_GET_OP_MODE) { - int opmode, p; + int p; + unsigned long opmode; static unsigned char ModeMask[] = { 0xfc, 0xf3, 0xcf, 0x3f }; int shiftbit; unsigned char val, mask; @@ -1572,9 +1602,8 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp) return -EFAULT; return 0; case MOXA_ASPP_MON_EXT: { - int status; - int opmode, p; - int shiftbit; + int status, p, shiftbit; + unsigned long opmode; unsigned cflag, iflag; for (i = 0; i < MXSER_PORTS; i++) { @@ -1654,73 +1683,52 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp) static void mxser_stoprx(struct tty_struct *tty) { struct mxser_struct *info = tty->driver_data; - /* unsigned long flags; */ info->ldisc_stop_rx = 1; if (I_IXOFF(tty)) { - /* MX_LOCK(&info->slock); */ /* following add by Victor Yu. 09-02-2002 */ if (info->IsMoxaMustChipFlag) { info->IER &= ~MOXA_MUST_RECV_ISR; outb(info->IER, info->base + UART_IER); - } else { - /* above add by Victor Yu. 09-02-2002 */ + } else if (!(info->flags & ASYNC_CLOSING)) { info->x_char = STOP_CHAR(tty); - /* mask by Victor Yu. 09-02-2002 */ - /* outb(info->IER, 0); */ outb(0, info->base + UART_IER); info->IER |= UART_IER_THRI; - /* force Tx interrupt */ outb(info->IER, info->base + UART_IER); - } /* add by Victor Yu. 09-02-2002 */ - /* MX_UNLOCK(&info->slock); */ + } } if (info->tty->termios->c_cflag & CRTSCTS) { - /* MX_LOCK(&info->slock); */ info->MCR &= ~UART_MCR_RTS; outb(info->MCR, info->base + UART_MCR); - /* MX_UNLOCK(&info->slock); */ } } static void mxser_startrx(struct tty_struct *tty) { struct mxser_struct *info = tty->driver_data; - /* unsigned long flags; */ info->ldisc_stop_rx = 0; if (I_IXOFF(tty)) { if (info->x_char) info->x_char = 0; else { - /* MX_LOCK(&info->slock); */ - /* following add by Victor Yu. 09-02-2002 */ if (info->IsMoxaMustChipFlag) { info->IER |= MOXA_MUST_RECV_ISR; outb(info->IER, info->base + UART_IER); - } else { - /* above add by Victor Yu. 09-02-2002 */ - + } else if (!(info->flags & ASYNC_CLOSING)) { info->x_char = START_CHAR(tty); - /* mask by Victor Yu. 09-02-2002 */ - /* outb(info->IER, 0); */ - /* add by Victor Yu. 09-02-2002 */ outb(0, info->base + UART_IER); - /* force Tx interrupt */ info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); - } /* add by Victor Yu. 09-02-2002 */ - /* MX_UNLOCK(&info->slock); */ + } } } if (info->tty->termios->c_cflag & CRTSCTS) { - /* MX_LOCK(&info->slock); */ info->MCR |= UART_MCR_RTS; outb(info->MCR, info->base + UART_MCR); - /* MX_UNLOCK(&info->slock); */ } } @@ -1730,22 +1738,22 @@ static void mxser_startrx(struct tty_struct *tty) */ static void mxser_throttle(struct tty_struct *tty) { - /* struct mxser_struct *info = tty->driver_data; */ - /* unsigned long flags; */ + struct mxser_struct *info = tty->driver_data; + unsigned long flags; - /* MX_LOCK(&info->slock); */ + spin_lock_irqsave(&info->slock, flags); mxser_stoprx(tty); - /* MX_UNLOCK(&info->slock); */ + spin_unlock_irqrestore(&info->slock, flags); } static void mxser_unthrottle(struct tty_struct *tty) { - /* struct mxser_struct *info = tty->driver_data; */ - /* unsigned long flags; */ + struct mxser_struct *info = tty->driver_data; + unsigned long flags; - /* MX_LOCK(&info->slock); */ + spin_lock_irqsave(&info->slock, flags); mxser_startrx(tty); - /* MX_UNLOCK(&info->slock); */ + spin_unlock_irqrestore(&info->slock, flags); } static void mxser_set_termios(struct tty_struct *tty, struct termios *old_termios) @@ -1807,7 +1815,9 @@ static void mxser_start(struct tty_struct *tty) unsigned long flags; spin_lock_irqsave(&info->slock, flags); - if (info->xmit_cnt && info->xmit_buf && !(info->IER & UART_IER_THRI)) { + if (info->xmit_cnt && info->xmit_buf + /* && !(info->IER & UART_IER_THRI) */) { + outb(info->IER & ~UART_IER_THRI, info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); } @@ -1927,6 +1937,7 @@ static irqreturn_t mxser_interrupt(int irq, void *dev_id, struct pt_regs *regs) struct mxser_struct *port; int max, irqbits, bits, msr; int pass_counter = 0; + unsigned int int_cnt; int handled = IRQ_NONE; port = NULL; @@ -1957,90 +1968,77 @@ static irqreturn_t mxser_interrupt(int irq, void *dev_id, struct pt_regs *regs) continue; info = port + i; - /* following add by Victor Yu. 09-13-2002 */ - iir = inb(info->base + UART_IIR); - if (iir & UART_IIR_NO_INT) - continue; - iir &= MOXA_MUST_IIR_MASK; - if (!info->tty) { - status = inb(info->base + UART_LSR); - outb(0x27, info->base + UART_FCR); - inb(info->base + UART_MSR); - continue; - } - /* above add by Victor Yu. 09-13-2002 */ - /* - if (info->tty->flip.count < TTY_FLIPBUF_SIZE / 4) { - info->IER |= MOXA_MUST_RECV_ISR; - outb(info->IER, info->base + UART_IER); - } - */ - - - /* mask by Victor Yu. 09-13-2002 - if ( !info->tty || - (inb(info->base + UART_IIR) & UART_IIR_NO_INT) ) - continue; - */ - /* mask by Victor Yu. 09-02-2002 - status = inb(info->base + UART_LSR) & info->read_status_mask; - */ - - /* following add by Victor Yu. 09-02-2002 */ - status = inb(info->base + UART_LSR); + int_cnt = 0; + do { + /* following add by Victor Yu. 09-13-2002 */ + iir = inb(info->base + UART_IIR); + if (iir & UART_IIR_NO_INT) + break; + iir &= MOXA_MUST_IIR_MASK; + if (!info->tty) { + status = inb(info->base + UART_LSR); + outb(0x27, info->base + UART_FCR); + inb(info->base + UART_MSR); + break; + } + /* above add by Victor Yu. 09-13-2002 */ - if (status & UART_LSR_PE) - info->err_shadow |= NPPI_NOTIFY_PARITY; - if (status & UART_LSR_FE) - info->err_shadow |= NPPI_NOTIFY_FRAMING; - if (status & UART_LSR_OE) - info->err_shadow |= NPPI_NOTIFY_HW_OVERRUN; - if (status & UART_LSR_BI) - info->err_shadow |= NPPI_NOTIFY_BREAK; + spin_lock(&info->slock); + /* following add by Victor Yu. 09-02-2002 */ + status = inb(info->base + UART_LSR); - if (info->IsMoxaMustChipFlag) { - /* - if ( (status & 0x02) && !(status & 0x01) ) { - outb(info->base+UART_FCR, 0x23); - continue; - } - */ - if (iir == MOXA_MUST_IIR_GDA || - iir == MOXA_MUST_IIR_RDA || - iir == MOXA_MUST_IIR_RTO || - iir == MOXA_MUST_IIR_LSR) - mxser_receive_chars(info, &status); + if (status & UART_LSR_PE) + info->err_shadow |= NPPI_NOTIFY_PARITY; + if (status & UART_LSR_FE) + info->err_shadow |= NPPI_NOTIFY_FRAMING; + if (status & UART_LSR_OE) + info->err_shadow |= + NPPI_NOTIFY_HW_OVERRUN; + if (status & UART_LSR_BI) + info->err_shadow |= NPPI_NOTIFY_BREAK; + + if (info->IsMoxaMustChipFlag) { + /* + if ( (status & 0x02) && !(status & 0x01) ) { + outb(info->base+UART_FCR, 0x23); + continue; + } + */ + if (iir == MOXA_MUST_IIR_GDA || + iir == MOXA_MUST_IIR_RDA || + iir == MOXA_MUST_IIR_RTO || + iir == MOXA_MUST_IIR_LSR) + mxser_receive_chars(info, + &status); - } else { - /* above add by Victor Yu. 09-02-2002 */ + } else { + /* above add by Victor Yu. 09-02-2002 */ - status &= info->read_status_mask; - if (status & UART_LSR_DR) - mxser_receive_chars(info, &status); - } - msr = inb(info->base + UART_MSR); - if (msr & UART_MSR_ANY_DELTA) { - mxser_check_modem_status(info, msr); - } - /* following add by Victor Yu. 09-13-2002 */ - if (info->IsMoxaMustChipFlag) { - if ((iir == 0x02) && (status & UART_LSR_THRE)) { - mxser_transmit_chars(info); + status &= info->read_status_mask; + if (status & UART_LSR_DR) + mxser_receive_chars(info, + &status); } - } else { - /* above add by Victor Yu. 09-13-2002 */ + msr = inb(info->base + UART_MSR); + if (msr & UART_MSR_ANY_DELTA) + mxser_check_modem_status(info, msr); + + /* following add by Victor Yu. 09-13-2002 */ + if (info->IsMoxaMustChipFlag) { + if (iir == 0x02 && (status & + UART_LSR_THRE)) + mxser_transmit_chars(info); + } else { + /* above add by Victor Yu. 09-13-2002 */ - if (status & UART_LSR_THRE) { -/* 8-2-99 by William - if ( info->x_char || (info->xmit_cnt > 0) ) -*/ - mxser_transmit_chars(info); + if (status & UART_LSR_THRE) + mxser_transmit_chars(info); } - } + spin_unlock(&info->slock); + } while (int_cnt++ < MXSER_ISR_PASS_LIMIT); } - if (pass_counter++ > MXSER_ISR_PASS_LIMIT) { + if (pass_counter++ > MXSER_ISR_PASS_LIMIT) break; /* Prevent infinite loops */ - } } irq_stop: @@ -2070,9 +2068,8 @@ static void mxser_receive_chars(struct mxser_struct *info, int *status) /* following add by Victor Yu. 09-02-2002 */ if (info->IsMoxaMustChipFlag != MOXA_OTHER_UART) { - if (*status & UART_LSR_SPECIAL) { + if (*status & UART_LSR_SPECIAL) goto intr_old; - } /* following add by Victor Yu. 02-11-2004 */ if (info->IsMoxaMustChipFlag == MOXA_MUST_MU860_HWID && (*status & MOXA_MUST_LSR_RERR)) @@ -2097,12 +2094,6 @@ static void mxser_receive_chars(struct mxser_struct *info, int *status) ch = inb(info->base + UART_RX); tty_insert_flip_char(tty, ch, 0); cnt++; - /* - if ((cnt >= HI_WATER) && (info->stop_rx == 0)) { - mxser_stoprx(tty); - info->stop_rx = 1; - break; - } */ } goto end_intr; } @@ -2112,17 +2103,11 @@ static void mxser_receive_chars(struct mxser_struct *info, int *status) do { if (max-- < 0) break; - /* - if ((cnt >= HI_WATER) && (info->stop_rx == 0)) { - mxser_stoprx(tty); - info->stop_rx=1; - break; - } - */ ch = inb(info->base + UART_RX); /* following add by Victor Yu. 09-02-2002 */ - if (info->IsMoxaMustChipFlag && (*status & UART_LSR_OE) /*&& !(*status&UART_LSR_DR) */ ) + if (info->IsMoxaMustChipFlag && (*status & UART_LSR_OE) + /*&& !(*status&UART_LSR_DR) */) outb(0x23, info->base + UART_FCR); *status &= info->read_status_mask; /* above add by Victor Yu. 09-02-2002 */ @@ -2136,26 +2121,25 @@ static void mxser_receive_chars(struct mxser_struct *info, int *status) flag = TTY_BREAK; /* added by casper 1/11/2000 */ info->icount.brk++; -/* */ + if (info->flags & ASYNC_SAK) do_SAK(tty); } else if (*status & UART_LSR_PE) { flag = TTY_PARITY; /* added by casper 1/11/2000 */ info->icount.parity++; -/* */ } else if (*status & UART_LSR_FE) { flag = TTY_FRAME; /* added by casper 1/11/2000 */ info->icount.frame++; -/* */ } else if (*status & UART_LSR_OE) { flag = TTY_OVERRUN; /* added by casper 1/11/2000 */ info->icount.overrun++; -/* */ - } - } + } else + flags = TTY_BREAK; + } else + flags = 0; tty_insert_flip_char(tty, ch, flag); cnt++; if (cnt >= recv_room) { @@ -2171,7 +2155,6 @@ static void mxser_receive_chars(struct mxser_struct *info, int *status) /* following add by Victor Yu. 09-02-2002 */ if (info->IsMoxaMustChipFlag) break; - /* above add by Victor Yu. 09-02-2002 */ /* mask by Victor Yu. 09-02-2002 *status = inb(info->base + UART_LSR) & info->read_status_mask; @@ -2206,24 +2189,25 @@ static void mxser_transmit_chars(struct mxser_struct *info) /* added by casper 1/11/2000 */ info->icount.tx++; -/* */ - spin_unlock_irqrestore(&info->slock, flags); - return; + goto unlock; } - if (info->xmit_buf == 0) { - spin_unlock_irqrestore(&info->slock, flags); - return; - } + if (info->xmit_buf == 0) + goto unlock; - if ((info->xmit_cnt <= 0) || info->tty->stopped || - (info->tty->hw_stopped && + if (info->xmit_cnt == 0) { + if (info->xmit_cnt < WAKEUP_CHARS) { /* XXX what's this for?? */ + set_bit(MXSER_EVENT_TXLOW, &info->event); + schedule_work(&info->tqueue); + } + goto unlock; + } + if (info->tty->stopped || (info->tty->hw_stopped && (info->type != PORT_16550A) && (!info->IsMoxaMustChipFlag))) { info->IER &= ~UART_IER_THRI; outb(info->IER, info->base + UART_IER); - spin_unlock_irqrestore(&info->slock, flags); - return; + goto unlock; } cnt = info->xmit_cnt; @@ -2240,11 +2224,9 @@ static void mxser_transmit_chars(struct mxser_struct *info) /* added by James 03-12-2004. */ info->mon_data.txcnt += (cnt - info->xmit_cnt); info->mon_data.up_txcnt += (cnt - info->xmit_cnt); -/* (above) added by James. */ /* added by casper 1/11/2000 */ info->icount.tx += (cnt - info->xmit_cnt); -/* */ if (info->xmit_cnt < WAKEUP_CHARS) { set_bit(MXSER_EVENT_TXLOW, &info->event); @@ -2254,6 +2236,7 @@ static void mxser_transmit_chars(struct mxser_struct *info) info->IER &= ~UART_IER_THRI; outb(info->IER, info->base + UART_IER); } +unlock: spin_unlock_irqrestore(&info->slock, flags); } @@ -2284,16 +2267,19 @@ static void mxser_check_modem_status(struct mxser_struct *info, int status) if ((info->type != PORT_16550A) && (!info->IsMoxaMustChipFlag)) { + outb(info->IER & ~UART_IER_THRI, + info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); } set_bit(MXSER_EVENT_TXLOW, &info->event); - schedule_work(&info->tqueue); } + schedule_work(&info->tqueue); + } } else { if (!(status & UART_MSR_CTS)) { info->tty->hw_stopped = 1; - if ((info->type != PORT_16550A) && - (!info->IsMoxaMustChipFlag)) { + if (info->type != PORT_16550A && + !info->IsMoxaMustChipFlag) { info->IER &= ~UART_IER_THRI; outb(info->IER, info->base + UART_IER); } @@ -2645,8 +2631,10 @@ static int mxser_change_speed(struct mxser_struct *info, struct termios *old_ter if (info->tty->hw_stopped) { if (status & UART_MSR_CTS) { info->tty->hw_stopped = 0; - if ((info->type != PORT_16550A) && - (!info->IsMoxaMustChipFlag)) { + if (info->type != PORT_16550A && + !info->IsMoxaMustChipFlag) { + outb(info->IER & ~UART_IER_THRI, + info->base + UART_IER); info->IER |= UART_IER_THRI; outb(info->IER, info->base + UART_IER); } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 4d972bbef316..51180dba9a98 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1785,14 +1785,17 @@ #define PCI_DEVICE_ID_MOXA_C104 0x1040 #define PCI_DEVICE_ID_MOXA_CP104U 0x1041 #define PCI_DEVICE_ID_MOXA_CP104JU 0x1042 +#define PCI_DEVICE_ID_MOXA_CP104EL 0x1043 #define PCI_DEVICE_ID_MOXA_CT114 0x1140 #define PCI_DEVICE_ID_MOXA_CP114 0x1141 #define PCI_DEVICE_ID_MOXA_CP118U 0x1180 +#define PCI_DEVICE_ID_MOXA_CP118EL 0x1181 #define PCI_DEVICE_ID_MOXA_CP132 0x1320 #define PCI_DEVICE_ID_MOXA_CP132U 0x1321 #define PCI_DEVICE_ID_MOXA_CP134U 0x1340 #define PCI_DEVICE_ID_MOXA_C168 0x1680 #define PCI_DEVICE_ID_MOXA_CP168U 0x1681 +#define PCI_DEVICE_ID_MOXA_CP168EL 0x1682 #define PCI_VENDOR_ID_CCD 0x1397 #define PCI_DEVICE_ID_CCD_2BD0 0x2bd0 -- cgit v1.2.3 From ca7ed0f22f4876986b2eb1fbd80ba78e07fd69d5 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 8 Dec 2006 02:38:39 -0800 Subject: [PATCH] Char: stallion, kill typedefs Typedefs are considered ugly in the kernel. Eliminate them. Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/stallion.c | 458 +++++++++++++++++++++++------------------------ include/linux/stallion.h | 24 +-- 2 files changed, 239 insertions(+), 243 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index 6a6e5124a5fa..b108ac6e3d83 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -88,16 +88,14 @@ * PCI BIOS32 support is compiled into the kernel. */ -typedef struct { +static struct stlconf { int brdtype; int ioaddr1; int ioaddr2; unsigned long memaddr; int irq; int irqtype; -} stlconf_t; - -static stlconf_t stl_brdconf[] = { +} stl_brdconf[] = { /*{ BRD_EASYIO, 0x2a0, 0, 0, 10, 0 },*/ }; @@ -154,8 +152,8 @@ static struct termios stl_deftermios = { */ static comstats_t stl_comstats; static combrd_t stl_brdstats; -static stlbrd_t stl_dummybrd; -static stlport_t stl_dummyport; +static struct stlbrd stl_dummybrd; +static struct stlport stl_dummyport; /* * Define global place to put buffer overflow characters. @@ -164,7 +162,7 @@ static char stl_unwanted[SC26198_RXFIFOSIZE]; /*****************************************************************************/ -static stlbrd_t *stl_brds[STL_MAXBRDS]; +static struct stlbrd *stl_brds[STL_MAXBRDS]; /* * Per board state flags. Used with the state field of the board struct. @@ -243,12 +241,10 @@ static char **stl_brdsp[] = { * parse any module arguments. */ -typedef struct stlbrdtype { +static struct { char *name; int type; -} stlbrdtype_t; - -static stlbrdtype_t stl_brdstr[] = { +} stl_brdstr[] = { { "easyio", BRD_EASYIO }, { "eio", BRD_EASYIO }, { "20", BRD_EASYIO }, @@ -458,7 +454,7 @@ static unsigned int stl_baudrates[] = { */ static void stl_argbrds(void); -static int stl_parsebrd(stlconf_t *confp, char **argp); +static int stl_parsebrd(struct stlconf *confp, char **argp); static unsigned long stl_atol(char *str); @@ -482,31 +478,31 @@ static void stl_waituntilsent(struct tty_struct *tty, int timeout); static void stl_sendxchar(struct tty_struct *tty, char ch); static void stl_hangup(struct tty_struct *tty); static int stl_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, unsigned long arg); -static int stl_portinfo(stlport_t *portp, int portnr, char *pos); +static int stl_portinfo(struct stlport *portp, int portnr, char *pos); static int stl_readproc(char *page, char **start, off_t off, int count, int *eof, void *data); -static int stl_brdinit(stlbrd_t *brdp); -static int stl_initports(stlbrd_t *brdp, stlpanel_t *panelp); -static int stl_getserial(stlport_t *portp, struct serial_struct __user *sp); -static int stl_setserial(stlport_t *portp, struct serial_struct __user *sp); +static int stl_brdinit(struct stlbrd *brdp); +static int stl_initports(struct stlbrd *brdp, struct stlpanel *panelp); +static int stl_getserial(struct stlport *portp, struct serial_struct __user *sp); +static int stl_setserial(struct stlport *portp, struct serial_struct __user *sp); static int stl_getbrdstats(combrd_t __user *bp); -static int stl_getportstats(stlport_t *portp, comstats_t __user *cp); -static int stl_clrportstats(stlport_t *portp, comstats_t __user *cp); -static int stl_getportstruct(stlport_t __user *arg); -static int stl_getbrdstruct(stlbrd_t __user *arg); -static int stl_waitcarrier(stlport_t *portp, struct file *filp); -static int stl_eiointr(stlbrd_t *brdp); -static int stl_echatintr(stlbrd_t *brdp); -static int stl_echmcaintr(stlbrd_t *brdp); -static int stl_echpciintr(stlbrd_t *brdp); -static int stl_echpci64intr(stlbrd_t *brdp); +static int stl_getportstats(struct stlport *portp, comstats_t __user *cp); +static int stl_clrportstats(struct stlport *portp, comstats_t __user *cp); +static int stl_getportstruct(struct stlport __user *arg); +static int stl_getbrdstruct(struct stlbrd __user *arg); +static int stl_waitcarrier(struct stlport *portp, struct file *filp); +static int stl_eiointr(struct stlbrd *brdp); +static int stl_echatintr(struct stlbrd *brdp); +static int stl_echmcaintr(struct stlbrd *brdp); +static int stl_echpciintr(struct stlbrd *brdp); +static int stl_echpci64intr(struct stlbrd *brdp); static void stl_offintr(struct work_struct *); -static stlbrd_t *stl_allocbrd(void); -static stlport_t *stl_getport(int brdnr, int panelnr, int portnr); +static struct stlbrd *stl_allocbrd(void); +static struct stlport *stl_getport(int brdnr, int panelnr, int portnr); static inline int stl_initbrds(void); -static inline int stl_initeio(stlbrd_t *brdp); -static inline int stl_initech(stlbrd_t *brdp); +static inline int stl_initeio(struct stlbrd *brdp); +static inline int stl_initech(struct stlbrd *brdp); static inline int stl_getbrdnr(void); #ifdef CONFIG_PCI @@ -517,59 +513,59 @@ static inline int stl_initpcibrd(int brdtype, struct pci_dev *devp); /* * CD1400 uart specific handling functions. */ -static void stl_cd1400setreg(stlport_t *portp, int regnr, int value); -static int stl_cd1400getreg(stlport_t *portp, int regnr); -static int stl_cd1400updatereg(stlport_t *portp, int regnr, int value); -static int stl_cd1400panelinit(stlbrd_t *brdp, stlpanel_t *panelp); -static void stl_cd1400portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *portp); -static void stl_cd1400setport(stlport_t *portp, struct termios *tiosp); -static int stl_cd1400getsignals(stlport_t *portp); -static void stl_cd1400setsignals(stlport_t *portp, int dtr, int rts); -static void stl_cd1400ccrwait(stlport_t *portp); -static void stl_cd1400enablerxtx(stlport_t *portp, int rx, int tx); -static void stl_cd1400startrxtx(stlport_t *portp, int rx, int tx); -static void stl_cd1400disableintrs(stlport_t *portp); -static void stl_cd1400sendbreak(stlport_t *portp, int len); -static void stl_cd1400flowctrl(stlport_t *portp, int state); -static void stl_cd1400sendflow(stlport_t *portp, int state); -static void stl_cd1400flush(stlport_t *portp); -static int stl_cd1400datastate(stlport_t *portp); -static void stl_cd1400eiointr(stlpanel_t *panelp, unsigned int iobase); -static void stl_cd1400echintr(stlpanel_t *panelp, unsigned int iobase); -static void stl_cd1400txisr(stlpanel_t *panelp, int ioaddr); -static void stl_cd1400rxisr(stlpanel_t *panelp, int ioaddr); -static void stl_cd1400mdmisr(stlpanel_t *panelp, int ioaddr); - -static inline int stl_cd1400breakisr(stlport_t *portp, int ioaddr); +static void stl_cd1400setreg(struct stlport *portp, int regnr, int value); +static int stl_cd1400getreg(struct stlport *portp, int regnr); +static int stl_cd1400updatereg(struct stlport *portp, int regnr, int value); +static int stl_cd1400panelinit(struct stlbrd *brdp, struct stlpanel *panelp); +static void stl_cd1400portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); +static void stl_cd1400setport(struct stlport *portp, struct termios *tiosp); +static int stl_cd1400getsignals(struct stlport *portp); +static void stl_cd1400setsignals(struct stlport *portp, int dtr, int rts); +static void stl_cd1400ccrwait(struct stlport *portp); +static void stl_cd1400enablerxtx(struct stlport *portp, int rx, int tx); +static void stl_cd1400startrxtx(struct stlport *portp, int rx, int tx); +static void stl_cd1400disableintrs(struct stlport *portp); +static void stl_cd1400sendbreak(struct stlport *portp, int len); +static void stl_cd1400flowctrl(struct stlport *portp, int state); +static void stl_cd1400sendflow(struct stlport *portp, int state); +static void stl_cd1400flush(struct stlport *portp); +static int stl_cd1400datastate(struct stlport *portp); +static void stl_cd1400eiointr(struct stlpanel *panelp, unsigned int iobase); +static void stl_cd1400echintr(struct stlpanel *panelp, unsigned int iobase); +static void stl_cd1400txisr(struct stlpanel *panelp, int ioaddr); +static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr); +static void stl_cd1400mdmisr(struct stlpanel *panelp, int ioaddr); + +static inline int stl_cd1400breakisr(struct stlport *portp, int ioaddr); /* * SC26198 uart specific handling functions. */ -static void stl_sc26198setreg(stlport_t *portp, int regnr, int value); -static int stl_sc26198getreg(stlport_t *portp, int regnr); -static int stl_sc26198updatereg(stlport_t *portp, int regnr, int value); -static int stl_sc26198getglobreg(stlport_t *portp, int regnr); -static int stl_sc26198panelinit(stlbrd_t *brdp, stlpanel_t *panelp); -static void stl_sc26198portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *portp); -static void stl_sc26198setport(stlport_t *portp, struct termios *tiosp); -static int stl_sc26198getsignals(stlport_t *portp); -static void stl_sc26198setsignals(stlport_t *portp, int dtr, int rts); -static void stl_sc26198enablerxtx(stlport_t *portp, int rx, int tx); -static void stl_sc26198startrxtx(stlport_t *portp, int rx, int tx); -static void stl_sc26198disableintrs(stlport_t *portp); -static void stl_sc26198sendbreak(stlport_t *portp, int len); -static void stl_sc26198flowctrl(stlport_t *portp, int state); -static void stl_sc26198sendflow(stlport_t *portp, int state); -static void stl_sc26198flush(stlport_t *portp); -static int stl_sc26198datastate(stlport_t *portp); -static void stl_sc26198wait(stlport_t *portp); -static void stl_sc26198txunflow(stlport_t *portp, struct tty_struct *tty); -static void stl_sc26198intr(stlpanel_t *panelp, unsigned int iobase); -static void stl_sc26198txisr(stlport_t *port); -static void stl_sc26198rxisr(stlport_t *port, unsigned int iack); -static void stl_sc26198rxbadch(stlport_t *portp, unsigned char status, char ch); -static void stl_sc26198rxbadchars(stlport_t *portp); -static void stl_sc26198otherisr(stlport_t *port, unsigned int iack); +static void stl_sc26198setreg(struct stlport *portp, int regnr, int value); +static int stl_sc26198getreg(struct stlport *portp, int regnr); +static int stl_sc26198updatereg(struct stlport *portp, int regnr, int value); +static int stl_sc26198getglobreg(struct stlport *portp, int regnr); +static int stl_sc26198panelinit(struct stlbrd *brdp, struct stlpanel *panelp); +static void stl_sc26198portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); +static void stl_sc26198setport(struct stlport *portp, struct termios *tiosp); +static int stl_sc26198getsignals(struct stlport *portp); +static void stl_sc26198setsignals(struct stlport *portp, int dtr, int rts); +static void stl_sc26198enablerxtx(struct stlport *portp, int rx, int tx); +static void stl_sc26198startrxtx(struct stlport *portp, int rx, int tx); +static void stl_sc26198disableintrs(struct stlport *portp); +static void stl_sc26198sendbreak(struct stlport *portp, int len); +static void stl_sc26198flowctrl(struct stlport *portp, int state); +static void stl_sc26198sendflow(struct stlport *portp, int state); +static void stl_sc26198flush(struct stlport *portp); +static int stl_sc26198datastate(struct stlport *portp); +static void stl_sc26198wait(struct stlport *portp); +static void stl_sc26198txunflow(struct stlport *portp, struct tty_struct *tty); +static void stl_sc26198intr(struct stlpanel *panelp, unsigned int iobase); +static void stl_sc26198txisr(struct stlport *port); +static void stl_sc26198rxisr(struct stlport *port, unsigned int iack); +static void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char ch); +static void stl_sc26198rxbadchars(struct stlport *portp); +static void stl_sc26198otherisr(struct stlport *port, unsigned int iack); /*****************************************************************************/ @@ -577,20 +573,20 @@ static void stl_sc26198otherisr(stlport_t *port, unsigned int iack); * Generic UART support structure. */ typedef struct uart { - int (*panelinit)(stlbrd_t *brdp, stlpanel_t *panelp); - void (*portinit)(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *portp); - void (*setport)(stlport_t *portp, struct termios *tiosp); - int (*getsignals)(stlport_t *portp); - void (*setsignals)(stlport_t *portp, int dtr, int rts); - void (*enablerxtx)(stlport_t *portp, int rx, int tx); - void (*startrxtx)(stlport_t *portp, int rx, int tx); - void (*disableintrs)(stlport_t *portp); - void (*sendbreak)(stlport_t *portp, int len); - void (*flowctrl)(stlport_t *portp, int state); - void (*sendflow)(stlport_t *portp, int state); - void (*flush)(stlport_t *portp); - int (*datastate)(stlport_t *portp); - void (*intr)(stlpanel_t *panelp, unsigned int iobase); + int (*panelinit)(struct stlbrd *brdp, struct stlpanel *panelp); + void (*portinit)(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); + void (*setport)(struct stlport *portp, struct termios *tiosp); + int (*getsignals)(struct stlport *portp); + void (*setsignals)(struct stlport *portp, int dtr, int rts); + void (*enablerxtx)(struct stlport *portp, int rx, int tx); + void (*startrxtx)(struct stlport *portp, int rx, int tx); + void (*disableintrs)(struct stlport *portp); + void (*sendbreak)(struct stlport *portp, int len); + void (*flowctrl)(struct stlport *portp, int state); + void (*sendflow)(struct stlport *portp, int state); + void (*flush)(struct stlport *portp); + int (*datastate)(struct stlport *portp); + void (*intr)(struct stlpanel *panelp, unsigned int iobase); } uart_t; /* @@ -730,9 +726,9 @@ static int __init stallion_module_init(void) static void __exit stallion_module_exit(void) { - stlbrd_t *brdp; - stlpanel_t *panelp; - stlport_t *portp; + struct stlbrd *brdp; + struct stlpanel *panelp; + struct stlport *portp; int i, j, k; pr_debug("cleanup_module()\n"); @@ -802,8 +798,8 @@ module_exit(stallion_module_exit); static void stl_argbrds(void) { - stlconf_t conf; - stlbrd_t *brdp; + struct stlconf conf; + struct stlbrd *brdp; int i; pr_debug("stl_argbrds()\n"); @@ -867,7 +863,7 @@ static unsigned long stl_atol(char *str) * Parse the supplied argument string, into the board conf struct. */ -static int stl_parsebrd(stlconf_t *confp, char **argp) +static int stl_parsebrd(struct stlconf *confp, char **argp) { char *sp; int i; @@ -911,14 +907,14 @@ static int stl_parsebrd(stlconf_t *confp, char **argp) * Allocate a new board structure. Fill out the basic info in it. */ -static stlbrd_t *stl_allocbrd(void) +static struct stlbrd *stl_allocbrd(void) { - stlbrd_t *brdp; + struct stlbrd *brdp; - brdp = kzalloc(sizeof(stlbrd_t), GFP_KERNEL); + brdp = kzalloc(sizeof(struct stlbrd), GFP_KERNEL); if (!brdp) { printk("STALLION: failed to allocate memory (size=%Zd)\n", - sizeof(stlbrd_t)); + sizeof(struct stlbrd)); return NULL; } @@ -930,8 +926,8 @@ static stlbrd_t *stl_allocbrd(void) static int stl_open(struct tty_struct *tty, struct file *filp) { - stlport_t *portp; - stlbrd_t *brdp; + struct stlport *portp; + struct stlbrd *brdp; unsigned int minordev; int brdnr, panelnr, portnr, rc; @@ -1020,7 +1016,7 @@ static int stl_open(struct tty_struct *tty, struct file *filp) * maybe because if we are clocal then we don't need to wait... */ -static int stl_waitcarrier(stlport_t *portp, struct file *filp) +static int stl_waitcarrier(struct stlport *portp, struct file *filp) { unsigned long flags; int rc, doclocal; @@ -1074,7 +1070,7 @@ static int stl_waitcarrier(stlport_t *portp, struct file *filp) static void stl_close(struct tty_struct *tty, struct file *filp) { - stlport_t *portp; + struct stlport *portp; unsigned long flags; pr_debug("stl_close(tty=%p,filp=%p)\n", tty, filp); @@ -1154,7 +1150,7 @@ static void stl_close(struct tty_struct *tty, struct file *filp) static int stl_write(struct tty_struct *tty, const unsigned char *buf, int count) { - stlport_t *portp; + struct stlport *portp; unsigned int len, stlen; unsigned char *chbuf; char *head, *tail; @@ -1211,7 +1207,7 @@ static int stl_write(struct tty_struct *tty, const unsigned char *buf, int count static void stl_putchar(struct tty_struct *tty, unsigned char ch) { - stlport_t *portp; + struct stlport *portp; unsigned int len; char *head, *tail; @@ -1249,7 +1245,7 @@ static void stl_putchar(struct tty_struct *tty, unsigned char ch) static void stl_flushchars(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_flushchars(tty=%p)\n", tty); @@ -1268,7 +1264,7 @@ static void stl_flushchars(struct tty_struct *tty) static int stl_writeroom(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; char *head, *tail; pr_debug("stl_writeroom(tty=%p)\n", tty); @@ -1299,7 +1295,7 @@ static int stl_writeroom(struct tty_struct *tty) static int stl_charsinbuffer(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; unsigned int size; char *head, *tail; @@ -1327,10 +1323,10 @@ static int stl_charsinbuffer(struct tty_struct *tty) * Generate the serial struct info. */ -static int stl_getserial(stlport_t *portp, struct serial_struct __user *sp) +static int stl_getserial(struct stlport *portp, struct serial_struct __user *sp) { struct serial_struct sio; - stlbrd_t *brdp; + struct stlbrd *brdp; pr_debug("stl_getserial(portp=%p,sp=%p)\n", portp, sp); @@ -1366,7 +1362,7 @@ static int stl_getserial(stlport_t *portp, struct serial_struct __user *sp) * just quietly ignore any requests to change irq, etc. */ -static int stl_setserial(stlport_t *portp, struct serial_struct __user *sp) +static int stl_setserial(struct stlport *portp, struct serial_struct __user *sp) { struct serial_struct sio; @@ -1396,7 +1392,7 @@ static int stl_setserial(stlport_t *portp, struct serial_struct __user *sp) static int stl_tiocmget(struct tty_struct *tty, struct file *file) { - stlport_t *portp; + struct stlport *portp; if (tty == NULL) return -ENODEV; @@ -1412,7 +1408,7 @@ static int stl_tiocmget(struct tty_struct *tty, struct file *file) static int stl_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear) { - stlport_t *portp; + struct stlport *portp; int rts = -1, dtr = -1; if (tty == NULL) @@ -1438,7 +1434,7 @@ static int stl_tiocmset(struct tty_struct *tty, struct file *file, static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg) { - stlport_t *portp; + struct stlport *portp; unsigned int ival; int rc; void __user *argp = (void __user *)arg; @@ -1503,7 +1499,7 @@ static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd static void stl_settermios(struct tty_struct *tty, struct termios *old) { - stlport_t *portp; + struct stlport *portp; struct termios *tiosp; pr_debug("stl_settermios(tty=%p,old=%p)\n", tty, old); @@ -1539,7 +1535,7 @@ static void stl_settermios(struct tty_struct *tty, struct termios *old) static void stl_throttle(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_throttle(tty=%p)\n", tty); @@ -1559,7 +1555,7 @@ static void stl_throttle(struct tty_struct *tty) static void stl_unthrottle(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_unthrottle(tty=%p)\n", tty); @@ -1580,7 +1576,7 @@ static void stl_unthrottle(struct tty_struct *tty) static void stl_stop(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_stop(tty=%p)\n", tty); @@ -1600,7 +1596,7 @@ static void stl_stop(struct tty_struct *tty) static void stl_start(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_start(tty=%p)\n", tty); @@ -1622,7 +1618,7 @@ static void stl_start(struct tty_struct *tty) static void stl_hangup(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_hangup(tty=%p)\n", tty); @@ -1656,7 +1652,7 @@ static void stl_hangup(struct tty_struct *tty) static void stl_flushbuffer(struct tty_struct *tty) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_flushbuffer(tty=%p)\n", tty); @@ -1674,7 +1670,7 @@ static void stl_flushbuffer(struct tty_struct *tty) static void stl_breakctl(struct tty_struct *tty, int state) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_breakctl(tty=%p,state=%d)\n", tty, state); @@ -1691,7 +1687,7 @@ static void stl_breakctl(struct tty_struct *tty, int state) static void stl_waituntilsent(struct tty_struct *tty, int timeout) { - stlport_t *portp; + struct stlport *portp; unsigned long tend; pr_debug("stl_waituntilsent(tty=%p,timeout=%d)\n", tty, timeout); @@ -1719,7 +1715,7 @@ static void stl_waituntilsent(struct tty_struct *tty, int timeout) static void stl_sendxchar(struct tty_struct *tty, char ch) { - stlport_t *portp; + struct stlport *portp; pr_debug("stl_sendxchar(tty=%p,ch=%x)\n", tty, ch); @@ -1747,7 +1743,7 @@ static void stl_sendxchar(struct tty_struct *tty, char ch) * short then padded with spaces). */ -static int stl_portinfo(stlport_t *portp, int portnr, char *pos) +static int stl_portinfo(struct stlport *portp, int portnr, char *pos) { char *sp; int sigs, cnt; @@ -1793,9 +1789,9 @@ static int stl_portinfo(stlport_t *portp, int portnr, char *pos) static int stl_readproc(char *page, char **start, off_t off, int count, int *eof, void *data) { - stlbrd_t *brdp; - stlpanel_t *panelp; - stlport_t *portp; + struct stlbrd *brdp; + struct stlpanel *panelp; + struct stlport *portp; int brdnr, panelnr, portnr, totalport; int curoff, maxoff; char *pos; @@ -1876,7 +1872,7 @@ stl_readdone: static irqreturn_t stl_intr(int irq, void *dev_id) { - stlbrd_t *brdp = dev_id; + struct stlbrd *brdp = dev_id; pr_debug("stl_intr(brdp=%p,irq=%d)\n", brdp, irq); @@ -1889,9 +1885,9 @@ static irqreturn_t stl_intr(int irq, void *dev_id) * Interrupt service routine for EasyIO board types. */ -static int stl_eiointr(stlbrd_t *brdp) +static int stl_eiointr(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int iobase; int handled = 0; @@ -1912,9 +1908,9 @@ static int stl_eiointr(stlbrd_t *brdp) * Interrupt service routine for ECH-AT board types. */ -static int stl_echatintr(stlbrd_t *brdp) +static int stl_echatintr(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int ioaddr; int bnknr; int handled = 0; @@ -1943,9 +1939,9 @@ static int stl_echatintr(stlbrd_t *brdp) * Interrupt service routine for ECH-MCA board types. */ -static int stl_echmcaintr(stlbrd_t *brdp) +static int stl_echmcaintr(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int ioaddr; int bnknr; int handled = 0; @@ -1969,9 +1965,9 @@ static int stl_echmcaintr(stlbrd_t *brdp) * Interrupt service routine for ECH-PCI board types. */ -static int stl_echpciintr(stlbrd_t *brdp) +static int stl_echpciintr(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int ioaddr; int bnknr, recheck; int handled = 0; @@ -2000,9 +1996,9 @@ static int stl_echpciintr(stlbrd_t *brdp) * Interrupt service routine for ECH-8/64-PCI board types. */ -static int stl_echpci64intr(stlbrd_t *brdp) +static int stl_echpci64intr(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int ioaddr; int bnknr; int handled = 0; @@ -2028,7 +2024,7 @@ static int stl_echpci64intr(stlbrd_t *brdp) */ static void stl_offintr(struct work_struct *work) { - stlport_t *portp = container_of(work, stlport_t, tqueue); + struct stlport *portp = container_of(work, struct stlport, tqueue); struct tty_struct *tty; unsigned int oldsigs; @@ -2065,9 +2061,9 @@ static void stl_offintr(struct work_struct *work) * Initialize all the ports on a panel. */ -static int __init stl_initports(stlbrd_t *brdp, stlpanel_t *panelp) +static int __init stl_initports(struct stlbrd *brdp, struct stlpanel *panelp) { - stlport_t *portp; + struct stlport *portp; int chipmask, i; pr_debug("stl_initports(brdp=%p,panelp=%p)\n", brdp, panelp); @@ -2079,10 +2075,10 @@ static int __init stl_initports(stlbrd_t *brdp, stlpanel_t *panelp) * each ports data structures. */ for (i = 0; (i < panelp->nrports); i++) { - portp = kzalloc(sizeof(stlport_t), GFP_KERNEL); + portp = kzalloc(sizeof(struct stlport), GFP_KERNEL); if (!portp) { printk("STALLION: failed to allocate memory " - "(size=%Zd)\n", sizeof(stlport_t)); + "(size=%Zd)\n", sizeof(struct stlport)); break; } @@ -2114,9 +2110,9 @@ static int __init stl_initports(stlbrd_t *brdp, stlpanel_t *panelp) * Try to find and initialize an EasyIO board. */ -static inline int stl_initeio(stlbrd_t *brdp) +static inline int stl_initeio(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int status; char *name; int rc; @@ -2213,10 +2209,10 @@ static inline int stl_initeio(stlbrd_t *brdp) * can complete the setup. */ - panelp = kzalloc(sizeof(stlpanel_t), GFP_KERNEL); + panelp = kzalloc(sizeof(struct stlpanel), GFP_KERNEL); if (!panelp) { printk(KERN_WARNING "STALLION: failed to allocate memory " - "(size=%Zd)\n", sizeof(stlpanel_t)); + "(size=%Zd)\n", sizeof(struct stlpanel)); return -ENOMEM; } @@ -2255,9 +2251,9 @@ static inline int stl_initeio(stlbrd_t *brdp) * dealing with all types of ECH board. */ -static inline int stl_initech(stlbrd_t *brdp) +static inline int stl_initech(struct stlbrd *brdp) { - stlpanel_t *panelp; + struct stlpanel *panelp; unsigned int status, nxtid, ioaddr, conflict; int panelnr, banknr, i; char *name; @@ -2385,10 +2381,10 @@ static inline int stl_initech(stlbrd_t *brdp) status = inb(ioaddr + ECH_PNLSTATUS); if ((status & ECH_PNLIDMASK) != nxtid) break; - panelp = kzalloc(sizeof(stlpanel_t), GFP_KERNEL); + panelp = kzalloc(sizeof(struct stlpanel), GFP_KERNEL); if (!panelp) { printk("STALLION: failed to allocate memory " - "(size=%Zd)\n", sizeof(stlpanel_t)); + "(size=%Zd)\n", sizeof(struct stlpanel)); break; } panelp->magic = STL_PANELMAGIC; @@ -2466,7 +2462,7 @@ static inline int stl_initech(stlbrd_t *brdp) * since the initial search and setup is very different. */ -static int __init stl_brdinit(stlbrd_t *brdp) +static int __init stl_brdinit(struct stlbrd *brdp) { int i; @@ -2540,7 +2536,7 @@ static inline int stl_getbrdnr(void) static inline int stl_initpcibrd(int brdtype, struct pci_dev *devp) { - stlbrd_t *brdp; + struct stlbrd *brdp; pr_debug("stl_initpcibrd(brdtype=%d,busnr=%x,devnr=%x)\n", brdtype, devp->bus->number, devp->devfn); @@ -2638,8 +2634,8 @@ static inline int stl_findpcibrds(void) static inline int stl_initbrds(void) { - stlbrd_t *brdp; - stlconf_t *confp; + struct stlbrd *brdp; + struct stlconf *confp; int i; pr_debug("stl_initbrds()\n"); @@ -2688,8 +2684,8 @@ static inline int stl_initbrds(void) static int stl_getbrdstats(combrd_t __user *bp) { - stlbrd_t *brdp; - stlpanel_t *panelp; + struct stlbrd *brdp; + struct stlpanel *panelp; int i; if (copy_from_user(&stl_brdstats, bp, sizeof(combrd_t))) @@ -2726,10 +2722,10 @@ static int stl_getbrdstats(combrd_t __user *bp) * Resolve the referenced port number into a port struct pointer. */ -static stlport_t *stl_getport(int brdnr, int panelnr, int portnr) +static struct stlport *stl_getport(int brdnr, int panelnr, int portnr) { - stlbrd_t *brdp; - stlpanel_t *panelp; + struct stlbrd *brdp; + struct stlpanel *panelp; if ((brdnr < 0) || (brdnr >= STL_MAXBRDS)) return(NULL); @@ -2754,7 +2750,7 @@ static stlport_t *stl_getport(int brdnr, int panelnr, int portnr) * what port to get stats for (used through board control device). */ -static int stl_getportstats(stlport_t *portp, comstats_t __user *cp) +static int stl_getportstats(struct stlport *portp, comstats_t __user *cp) { unsigned char *head, *tail; unsigned long flags; @@ -2812,7 +2808,7 @@ static int stl_getportstats(stlport_t *portp, comstats_t __user *cp) * Clear the port stats structure. We also return it zeroed out... */ -static int stl_clrportstats(stlport_t *portp, comstats_t __user *cp) +static int stl_clrportstats(struct stlport *portp, comstats_t __user *cp) { if (!portp) { if (copy_from_user(&stl_comstats, cp, sizeof(comstats_t))) @@ -2837,17 +2833,17 @@ static int stl_clrportstats(stlport_t *portp, comstats_t __user *cp) * Return the entire driver ports structure to a user app. */ -static int stl_getportstruct(stlport_t __user *arg) +static int stl_getportstruct(struct stlport __user *arg) { - stlport_t *portp; + struct stlport *portp; - if (copy_from_user(&stl_dummyport, arg, sizeof(stlport_t))) + if (copy_from_user(&stl_dummyport, arg, sizeof(struct stlport))) return -EFAULT; portp = stl_getport(stl_dummyport.brdnr, stl_dummyport.panelnr, stl_dummyport.portnr); if (!portp) return -ENODEV; - return copy_to_user(arg, portp, sizeof(stlport_t)) ? -EFAULT : 0; + return copy_to_user(arg, portp, sizeof(struct stlport)) ? -EFAULT : 0; } /*****************************************************************************/ @@ -2856,18 +2852,18 @@ static int stl_getportstruct(stlport_t __user *arg) * Return the entire driver board structure to a user app. */ -static int stl_getbrdstruct(stlbrd_t __user *arg) +static int stl_getbrdstruct(struct stlbrd __user *arg) { - stlbrd_t *brdp; + struct stlbrd *brdp; - if (copy_from_user(&stl_dummybrd, arg, sizeof(stlbrd_t))) + if (copy_from_user(&stl_dummybrd, arg, sizeof(struct stlbrd))) return -EFAULT; if ((stl_dummybrd.brdnr < 0) || (stl_dummybrd.brdnr >= STL_MAXBRDS)) return -ENODEV; brdp = stl_brds[stl_dummybrd.brdnr]; if (!brdp) return(-ENODEV); - return copy_to_user(arg, brdp, sizeof(stlbrd_t)) ? -EFAULT : 0; + return copy_to_user(arg, brdp, sizeof(struct stlbrd)) ? -EFAULT : 0; } /*****************************************************************************/ @@ -2997,19 +2993,19 @@ static int __init stl_init(void) * (Maybe should make this inline...) */ -static int stl_cd1400getreg(stlport_t *portp, int regnr) +static int stl_cd1400getreg(struct stlport *portp, int regnr) { outb((regnr + portp->uartaddr), portp->ioaddr); return inb(portp->ioaddr + EREG_DATA); } -static void stl_cd1400setreg(stlport_t *portp, int regnr, int value) +static void stl_cd1400setreg(struct stlport *portp, int regnr, int value) { outb((regnr + portp->uartaddr), portp->ioaddr); outb(value, portp->ioaddr + EREG_DATA); } -static int stl_cd1400updatereg(stlport_t *portp, int regnr, int value) +static int stl_cd1400updatereg(struct stlport *portp, int regnr, int value) { outb((regnr + portp->uartaddr), portp->ioaddr); if (inb(portp->ioaddr + EREG_DATA) != value) { @@ -3027,7 +3023,7 @@ static int stl_cd1400updatereg(stlport_t *portp, int regnr, int value) * identical when dealing with ports. */ -static int stl_cd1400panelinit(stlbrd_t *brdp, stlpanel_t *panelp) +static int stl_cd1400panelinit(struct stlbrd *brdp, struct stlpanel *panelp) { unsigned int gfrcr; int chipmask, i, j; @@ -3084,7 +3080,7 @@ static int stl_cd1400panelinit(stlbrd_t *brdp, stlpanel_t *panelp) * Initialize hardware specific port registers. */ -static void stl_cd1400portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *portp) +static void stl_cd1400portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp) { unsigned long flags; pr_debug("stl_cd1400portinit(brdp=%p,panelp=%p,portp=%p)\n", brdp, @@ -3115,7 +3111,7 @@ static void stl_cd1400portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *po * since it won't usually take too long to be ready. */ -static void stl_cd1400ccrwait(stlport_t *portp) +static void stl_cd1400ccrwait(struct stlport *portp) { int i; @@ -3136,9 +3132,9 @@ static void stl_cd1400ccrwait(stlport_t *portp) * settings. */ -static void stl_cd1400setport(stlport_t *portp, struct termios *tiosp) +static void stl_cd1400setport(struct stlport *portp, struct termios *tiosp) { - stlbrd_t *brdp; + struct stlbrd *brdp; unsigned long flags; unsigned int clkdiv, baudrate; unsigned char cor1, cor2, cor3; @@ -3359,7 +3355,7 @@ static void stl_cd1400setport(stlport_t *portp, struct termios *tiosp) * Set the state of the DTR and RTS signals. */ -static void stl_cd1400setsignals(stlport_t *portp, int dtr, int rts) +static void stl_cd1400setsignals(struct stlport *portp, int dtr, int rts) { unsigned char msvr1, msvr2; unsigned long flags; @@ -3391,7 +3387,7 @@ static void stl_cd1400setsignals(stlport_t *portp, int dtr, int rts) * Return the state of the signals. */ -static int stl_cd1400getsignals(stlport_t *portp) +static int stl_cd1400getsignals(struct stlport *portp) { unsigned char msvr1, msvr2; unsigned long flags; @@ -3427,7 +3423,7 @@ static int stl_cd1400getsignals(stlport_t *portp) * Enable/Disable the Transmitter and/or Receiver. */ -static void stl_cd1400enablerxtx(stlport_t *portp, int rx, int tx) +static void stl_cd1400enablerxtx(struct stlport *portp, int rx, int tx) { unsigned char ccr; unsigned long flags; @@ -3461,7 +3457,7 @@ static void stl_cd1400enablerxtx(stlport_t *portp, int rx, int tx) * Start/stop the Transmitter and/or Receiver. */ -static void stl_cd1400startrxtx(stlport_t *portp, int rx, int tx) +static void stl_cd1400startrxtx(struct stlport *portp, int rx, int tx) { unsigned char sreron, sreroff; unsigned long flags; @@ -3498,7 +3494,7 @@ static void stl_cd1400startrxtx(stlport_t *portp, int rx, int tx) * Disable all interrupts from this port. */ -static void stl_cd1400disableintrs(stlport_t *portp) +static void stl_cd1400disableintrs(struct stlport *portp) { unsigned long flags; @@ -3514,7 +3510,7 @@ static void stl_cd1400disableintrs(stlport_t *portp) /*****************************************************************************/ -static void stl_cd1400sendbreak(stlport_t *portp, int len) +static void stl_cd1400sendbreak(struct stlport *portp, int len) { unsigned long flags; @@ -3539,7 +3535,7 @@ static void stl_cd1400sendbreak(stlport_t *portp, int len) * Take flow control actions... */ -static void stl_cd1400flowctrl(stlport_t *portp, int state) +static void stl_cd1400flowctrl(struct stlport *portp, int state) { struct tty_struct *tty; unsigned long flags; @@ -3601,7 +3597,7 @@ static void stl_cd1400flowctrl(stlport_t *portp, int state) * Send a flow control character... */ -static void stl_cd1400sendflow(stlport_t *portp, int state) +static void stl_cd1400sendflow(struct stlport *portp, int state) { struct tty_struct *tty; unsigned long flags; @@ -3634,7 +3630,7 @@ static void stl_cd1400sendflow(stlport_t *portp, int state) /*****************************************************************************/ -static void stl_cd1400flush(stlport_t *portp) +static void stl_cd1400flush(struct stlport *portp) { unsigned long flags; @@ -3663,7 +3659,7 @@ static void stl_cd1400flush(stlport_t *portp) * maintains the busy port flag. */ -static int stl_cd1400datastate(stlport_t *portp) +static int stl_cd1400datastate(struct stlport *portp) { pr_debug("stl_cd1400datastate(portp=%p)\n", portp); @@ -3679,7 +3675,7 @@ static int stl_cd1400datastate(stlport_t *portp) * Interrupt service routine for cd1400 EasyIO boards. */ -static void stl_cd1400eiointr(stlpanel_t *panelp, unsigned int iobase) +static void stl_cd1400eiointr(struct stlpanel *panelp, unsigned int iobase) { unsigned char svrtype; @@ -3709,7 +3705,7 @@ static void stl_cd1400eiointr(stlpanel_t *panelp, unsigned int iobase) * Interrupt service routine for cd1400 panels. */ -static void stl_cd1400echintr(stlpanel_t *panelp, unsigned int iobase) +static void stl_cd1400echintr(struct stlpanel *panelp, unsigned int iobase) { unsigned char svrtype; @@ -3735,7 +3731,7 @@ static void stl_cd1400echintr(stlpanel_t *panelp, unsigned int iobase) * this is the only way to generate them on the cd1400. */ -static inline int stl_cd1400breakisr(stlport_t *portp, int ioaddr) +static inline int stl_cd1400breakisr(struct stlport *portp, int ioaddr) { if (portp->brklen == 1) { outb((COR2 + portp->uartaddr), ioaddr); @@ -3777,9 +3773,9 @@ static inline int stl_cd1400breakisr(stlport_t *portp, int ioaddr) * be NULL if the buffer has been freed. */ -static void stl_cd1400txisr(stlpanel_t *panelp, int ioaddr) +static void stl_cd1400txisr(struct stlpanel *panelp, int ioaddr) { - stlport_t *portp; + struct stlport *portp; int len, stlen; char *head, *tail; unsigned char ioack, srer; @@ -3856,9 +3852,9 @@ stl_txalldone: * shutdown a port not in user context. Need to handle this case. */ -static void stl_cd1400rxisr(stlpanel_t *panelp, int ioaddr) +static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr) { - stlport_t *portp; + struct stlport *portp; struct tty_struct *tty; unsigned int ioack, len, buflen; unsigned char status; @@ -3954,9 +3950,9 @@ stl_rxalldone: * processing routine. */ -static void stl_cd1400mdmisr(stlpanel_t *panelp, int ioaddr) +static void stl_cd1400mdmisr(struct stlpanel *panelp, int ioaddr) { - stlport_t *portp; + struct stlport *portp; unsigned int ioack; unsigned char misr; @@ -3992,19 +3988,19 @@ static void stl_cd1400mdmisr(stlpanel_t *panelp, int ioaddr) * (Maybe should make this inline...) */ -static int stl_sc26198getreg(stlport_t *portp, int regnr) +static int stl_sc26198getreg(struct stlport *portp, int regnr) { outb((regnr | portp->uartaddr), (portp->ioaddr + XP_ADDR)); return inb(portp->ioaddr + XP_DATA); } -static void stl_sc26198setreg(stlport_t *portp, int regnr, int value) +static void stl_sc26198setreg(struct stlport *portp, int regnr, int value) { outb((regnr | portp->uartaddr), (portp->ioaddr + XP_ADDR)); outb(value, (portp->ioaddr + XP_DATA)); } -static int stl_sc26198updatereg(stlport_t *portp, int regnr, int value) +static int stl_sc26198updatereg(struct stlport *portp, int regnr, int value) { outb((regnr | portp->uartaddr), (portp->ioaddr + XP_ADDR)); if (inb(portp->ioaddr + XP_DATA) != value) { @@ -4020,14 +4016,14 @@ static int stl_sc26198updatereg(stlport_t *portp, int regnr, int value) * Functions to get and set the sc26198 global registers. */ -static int stl_sc26198getglobreg(stlport_t *portp, int regnr) +static int stl_sc26198getglobreg(struct stlport *portp, int regnr) { outb(regnr, (portp->ioaddr + XP_ADDR)); return inb(portp->ioaddr + XP_DATA); } #if 0 -static void stl_sc26198setglobreg(stlport_t *portp, int regnr, int value) +static void stl_sc26198setglobreg(struct stlport *portp, int regnr, int value) { outb(regnr, (portp->ioaddr + XP_ADDR)); outb(value, (portp->ioaddr + XP_DATA)); @@ -4042,7 +4038,7 @@ static void stl_sc26198setglobreg(stlport_t *portp, int regnr, int value) * identical when dealing with ports. */ -static int stl_sc26198panelinit(stlbrd_t *brdp, stlpanel_t *panelp) +static int stl_sc26198panelinit(struct stlbrd *brdp, struct stlpanel *panelp) { int chipmask, i; int nrchips, ioaddr; @@ -4087,7 +4083,7 @@ static int stl_sc26198panelinit(stlbrd_t *brdp, stlpanel_t *panelp) * Initialize hardware specific port registers. */ -static void stl_sc26198portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *portp) +static void stl_sc26198portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp) { pr_debug("stl_sc26198portinit(brdp=%p,panelp=%p,portp=%p)\n", brdp, panelp, portp); @@ -4113,9 +4109,9 @@ static void stl_sc26198portinit(stlbrd_t *brdp, stlpanel_t *panelp, stlport_t *p * settings. */ -static void stl_sc26198setport(stlport_t *portp, struct termios *tiosp) +static void stl_sc26198setport(struct stlport *portp, struct termios *tiosp) { - stlbrd_t *brdp; + struct stlbrd *brdp; unsigned long flags; unsigned int baudrate; unsigned char mr0, mr1, mr2, clk; @@ -4308,7 +4304,7 @@ static void stl_sc26198setport(stlport_t *portp, struct termios *tiosp) * Set the state of the DTR and RTS signals. */ -static void stl_sc26198setsignals(stlport_t *portp, int dtr, int rts) +static void stl_sc26198setsignals(struct stlport *portp, int dtr, int rts) { unsigned char iopioron, iopioroff; unsigned long flags; @@ -4341,7 +4337,7 @@ static void stl_sc26198setsignals(stlport_t *portp, int dtr, int rts) * Return the state of the signals. */ -static int stl_sc26198getsignals(stlport_t *portp) +static int stl_sc26198getsignals(struct stlport *portp) { unsigned char ipr; unsigned long flags; @@ -4370,7 +4366,7 @@ static int stl_sc26198getsignals(stlport_t *portp) * Enable/Disable the Transmitter and/or Receiver. */ -static void stl_sc26198enablerxtx(stlport_t *portp, int rx, int tx) +static void stl_sc26198enablerxtx(struct stlport *portp, int rx, int tx) { unsigned char ccr; unsigned long flags; @@ -4401,7 +4397,7 @@ static void stl_sc26198enablerxtx(stlport_t *portp, int rx, int tx) * Start/stop the Transmitter and/or Receiver. */ -static void stl_sc26198startrxtx(stlport_t *portp, int rx, int tx) +static void stl_sc26198startrxtx(struct stlport *portp, int rx, int tx) { unsigned char imr; unsigned long flags; @@ -4434,7 +4430,7 @@ static void stl_sc26198startrxtx(stlport_t *portp, int rx, int tx) * Disable all interrupts from this port. */ -static void stl_sc26198disableintrs(stlport_t *portp) +static void stl_sc26198disableintrs(struct stlport *portp) { unsigned long flags; @@ -4450,7 +4446,7 @@ static void stl_sc26198disableintrs(stlport_t *portp) /*****************************************************************************/ -static void stl_sc26198sendbreak(stlport_t *portp, int len) +static void stl_sc26198sendbreak(struct stlport *portp, int len) { unsigned long flags; @@ -4474,7 +4470,7 @@ static void stl_sc26198sendbreak(stlport_t *portp, int len) * Take flow control actions... */ -static void stl_sc26198flowctrl(stlport_t *portp, int state) +static void stl_sc26198flowctrl(struct stlport *portp, int state) { struct tty_struct *tty; unsigned long flags; @@ -4543,7 +4539,7 @@ static void stl_sc26198flowctrl(stlport_t *portp, int state) * Send a flow control character. */ -static void stl_sc26198sendflow(stlport_t *portp, int state) +static void stl_sc26198sendflow(struct stlport *portp, int state) { struct tty_struct *tty; unsigned long flags; @@ -4582,7 +4578,7 @@ static void stl_sc26198sendflow(stlport_t *portp, int state) /*****************************************************************************/ -static void stl_sc26198flush(stlport_t *portp) +static void stl_sc26198flush(struct stlport *portp) { unsigned long flags; @@ -4610,7 +4606,7 @@ static void stl_sc26198flush(stlport_t *portp) * check the port statusy register to be sure. */ -static int stl_sc26198datastate(stlport_t *portp) +static int stl_sc26198datastate(struct stlport *portp) { unsigned long flags; unsigned char sr; @@ -4638,7 +4634,7 @@ static int stl_sc26198datastate(stlport_t *portp) * to process a command... */ -static void stl_sc26198wait(stlport_t *portp) +static void stl_sc26198wait(struct stlport *portp) { int i; @@ -4659,7 +4655,7 @@ static void stl_sc26198wait(stlport_t *portp) * automatic flow control modes of the sc26198. */ -static inline void stl_sc26198txunflow(stlport_t *portp, struct tty_struct *tty) +static inline void stl_sc26198txunflow(struct stlport *portp, struct tty_struct *tty) { unsigned char mr0; @@ -4677,9 +4673,9 @@ static inline void stl_sc26198txunflow(stlport_t *portp, struct tty_struct *tty) * Interrupt service routine for sc26198 panels. */ -static void stl_sc26198intr(stlpanel_t *panelp, unsigned int iobase) +static void stl_sc26198intr(struct stlpanel *panelp, unsigned int iobase) { - stlport_t *portp; + struct stlport *portp; unsigned int iack; spin_lock(&brd_lock); @@ -4715,7 +4711,7 @@ static void stl_sc26198intr(stlpanel_t *panelp, unsigned int iobase) * be NULL if the buffer has been freed. */ -static void stl_sc26198txisr(stlport_t *portp) +static void stl_sc26198txisr(struct stlport *portp) { unsigned int ioaddr; unsigned char mr0; @@ -4776,7 +4772,7 @@ static void stl_sc26198txisr(stlport_t *portp) * shutdown a port not in user context. Need to handle this case. */ -static void stl_sc26198rxisr(stlport_t *portp, unsigned int iack) +static void stl_sc26198rxisr(struct stlport *portp, unsigned int iack) { struct tty_struct *tty; unsigned int len, buflen, ioaddr; @@ -4830,7 +4826,7 @@ static void stl_sc26198rxisr(stlport_t *portp, unsigned int iack) * Process an RX bad character. */ -static inline void stl_sc26198rxbadch(stlport_t *portp, unsigned char status, char ch) +static inline void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char ch) { struct tty_struct *tty; unsigned int ioaddr; @@ -4888,7 +4884,7 @@ static inline void stl_sc26198rxbadch(stlport_t *portp, unsigned char status, ch * the FIFO). */ -static void stl_sc26198rxbadchars(stlport_t *portp) +static void stl_sc26198rxbadchars(struct stlport *portp) { unsigned char status, mr1; char ch; @@ -4921,7 +4917,7 @@ static void stl_sc26198rxbadchars(stlport_t *portp) * processing time. */ -static void stl_sc26198otherisr(stlport_t *portp, unsigned int iack) +static void stl_sc26198otherisr(struct stlport *portp, unsigned int iack) { unsigned char cir, ipr, xisr; diff --git a/include/linux/stallion.h b/include/linux/stallion.h index 13a37f137ea2..ef5270b83d02 100644 --- a/include/linux/stallion.h +++ b/include/linux/stallion.h @@ -52,11 +52,11 @@ * protection - since "write" code only needs to change the head, and * interrupt code only needs to change the tail. */ -typedef struct { +struct stlrq { char *buf; char *head; char *tail; -} stlrq_t; +}; /* * Port, panel and board structures to hold status info about each. @@ -67,7 +67,7 @@ typedef struct { * is associated with, this makes it (fairly) easy to get back to the * board/panel info for a port. */ -typedef struct stlport { +struct stlport { unsigned long magic; int portnr; int panelnr; @@ -97,10 +97,10 @@ typedef struct stlport { wait_queue_head_t close_wait; struct work_struct tqueue; comstats_t stats; - stlrq_t tx; -} stlport_t; + struct stlrq tx; +}; -typedef struct stlpanel { +struct stlpanel { unsigned long magic; int panelnr; int brdnr; @@ -111,10 +111,10 @@ typedef struct stlpanel { void (*isr)(struct stlpanel *panelp, unsigned int iobase); unsigned int hwid; unsigned int ackmask; - stlport_t *ports[STL_PORTSPERPANEL]; -} stlpanel_t; + struct stlport *ports[STL_PORTSPERPANEL]; +}; -typedef struct stlbrd { +struct stlbrd { unsigned long magic; int brdnr; int brdtype; @@ -136,9 +136,9 @@ typedef struct stlbrd { unsigned long clk; unsigned int bnkpageaddr[STL_MAXBANKS]; unsigned int bnkstataddr[STL_MAXBANKS]; - stlpanel_t *bnk2panel[STL_MAXBANKS]; - stlpanel_t *panels[STL_MAXPANELS]; -} stlbrd_t; + struct stlpanel *bnk2panel[STL_MAXBANKS]; + struct stlpanel *panels[STL_MAXPANELS]; +}; /* -- cgit v1.2.3 From edc6afc5496875a640bef0913604be7550c1795d Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 8 Dec 2006 02:38:44 -0800 Subject: [PATCH] tty: switch to ktermios and new framework This is the core of the switch to the new framework. I've split it from the driver patches which are mostly search/replace and would encourage people to give this one a good hard stare. The references to BOTHER and ISHIFT are the termios values that must be defined by a platform once it wants to turn on "new style" ioctl support. The code patches here ensure that providing 1. The termios overlays the ktermios in memory 2. The only new kernel only fields are c_ispeed/c_ospeed (or none) the existing behaviour is retained. This is true for the patches at this point in time. Future patches will define BOTHER, ISHIFT and enable newer termios structures for each architecture, and once they are all done some of the ifdefs also vanish. [akpm@osdl.org: warning fix] [akpm@osdl.org: IRDA fix] Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 137 ++++++-------------- drivers/char/tty_ioctl.c | 258 +++++++++++++++++++++++++++++++++++-- include/linux/tty.h | 9 +- include/linux/tty_driver.h | 12 +- include/linux/tty_ldisc.h | 4 +- include/net/irda/ircomm_tty.h | 2 +- net/irda/ircomm/ircomm_tty_ioctl.c | 2 +- 7 files changed, 297 insertions(+), 127 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 48cee2004e97..4044c864fdd4 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -109,13 +109,15 @@ #define TTY_PARANOIA_CHECK 1 #define CHECK_TTY_COUNT 1 -struct termios tty_std_termios = { /* for the benefit of tty drivers */ +struct ktermios tty_std_termios = { /* for the benefit of tty drivers */ .c_iflag = ICRNL | IXON, .c_oflag = OPOST | ONLCR, .c_cflag = B38400 | CS8 | CREAD | HUPCL, .c_lflag = ISIG | ICANON | ECHO | ECHOE | ECHOK | ECHOCTL | ECHOKE | IEXTEN, - .c_cc = INIT_C_CC + .c_cc = INIT_C_CC, + .c_ispeed = 38400, + .c_ospeed = 38400 }; EXPORT_SYMBOL(tty_std_termios); @@ -1239,6 +1241,22 @@ void tty_ldisc_flush(struct tty_struct *tty) } EXPORT_SYMBOL_GPL(tty_ldisc_flush); + +/** + * tty_reset_termios - reset terminal state + * @tty: tty to reset + * + * Restore a terminal to the driver default state + */ + +static void tty_reset_termios(struct tty_struct *tty) +{ + mutex_lock(&tty->termios_mutex); + *tty->termios = tty->driver->init_termios; + tty->termios->c_ispeed = tty_termios_input_baud_rate(tty->termios); + tty->termios->c_ospeed = tty_termios_baud_rate(tty->termios); + mutex_unlock(&tty->termios_mutex); +} /** * do_tty_hangup - actual handler for hangup events @@ -1327,11 +1345,7 @@ static void do_tty_hangup(struct work_struct *work) * N_TTY. */ if (tty->driver->flags & TTY_DRIVER_RESET_TERMIOS) - { - mutex_lock(&tty->termios_mutex); - *tty->termios = tty->driver->init_termios; - mutex_unlock(&tty->termios_mutex); - } + tty_reset_termios(tty); /* Defer ldisc switch */ /* tty_deferred_ldisc_switch(N_TTY); @@ -1870,8 +1884,8 @@ static int init_dev(struct tty_driver *driver, int idx, struct tty_struct **ret_tty) { struct tty_struct *tty, *o_tty; - struct termios *tp, **tp_loc, *o_tp, **o_tp_loc; - struct termios *ltp, **ltp_loc, *o_ltp, **o_ltp_loc; + struct ktermios *tp, **tp_loc, *o_tp, **o_tp_loc; + struct ktermios *ltp, **ltp_loc, *o_ltp, **o_ltp_loc; int retval = 0; /* check whether we're reopening an existing tty */ @@ -1918,7 +1932,7 @@ static int init_dev(struct tty_driver *driver, int idx, } if (!*tp_loc) { - tp = (struct termios *) kmalloc(sizeof(struct termios), + tp = (struct ktermios *) kmalloc(sizeof(struct ktermios), GFP_KERNEL); if (!tp) goto free_mem_out; @@ -1926,11 +1940,11 @@ static int init_dev(struct tty_driver *driver, int idx, } if (!*ltp_loc) { - ltp = (struct termios *) kmalloc(sizeof(struct termios), + ltp = (struct ktermios *) kmalloc(sizeof(struct ktermios), GFP_KERNEL); if (!ltp) goto free_mem_out; - memset(ltp, 0, sizeof(struct termios)); + memset(ltp, 0, sizeof(struct ktermios)); } if (driver->type == TTY_DRIVER_TYPE_PTY) { @@ -1951,19 +1965,19 @@ static int init_dev(struct tty_driver *driver, int idx, } if (!*o_tp_loc) { - o_tp = (struct termios *) - kmalloc(sizeof(struct termios), GFP_KERNEL); + o_tp = (struct ktermios *) + kmalloc(sizeof(struct ktermios), GFP_KERNEL); if (!o_tp) goto free_mem_out; *o_tp = driver->other->init_termios; } if (!*o_ltp_loc) { - o_ltp = (struct termios *) - kmalloc(sizeof(struct termios), GFP_KERNEL); + o_ltp = (struct ktermios *) + kmalloc(sizeof(struct ktermios), GFP_KERNEL); if (!o_ltp) goto free_mem_out; - memset(o_ltp, 0, sizeof(struct termios)); + memset(o_ltp, 0, sizeof(struct ktermios)); } /* @@ -2002,6 +2016,9 @@ static int init_dev(struct tty_driver *driver, int idx, *ltp_loc = ltp; tty->termios = *tp_loc; tty->termios_locked = *ltp_loc; + /* Compatibility until drivers always set this */ + tty->termios->c_ispeed = tty_termios_input_baud_rate(tty->termios); + tty->termios->c_ospeed = tty_termios_baud_rate(tty->termios); driver->refcount++; tty->count++; @@ -2104,7 +2121,7 @@ release_mem_out: static void release_mem(struct tty_struct *tty, int idx) { struct tty_struct *o_tty; - struct termios *tp; + struct ktermios *tp; int devpts = tty->driver->flags & TTY_DRIVER_DEVPTS_MEM; if ((o_tty = tty->link) != NULL) { @@ -3458,84 +3475,6 @@ static void flush_to_ldisc(struct work_struct *work) tty_ldisc_deref(disc); } -/* - * Routine which returns the baud rate of the tty - * - * Note that the baud_table needs to be kept in sync with the - * include/asm/termbits.h file. - */ -static int baud_table[] = { - 0, 50, 75, 110, 134, 150, 200, 300, 600, 1200, 1800, 2400, 4800, - 9600, 19200, 38400, 57600, 115200, 230400, 460800, -#ifdef __sparc__ - 76800, 153600, 307200, 614400, 921600 -#else - 500000, 576000, 921600, 1000000, 1152000, 1500000, 2000000, - 2500000, 3000000, 3500000, 4000000 -#endif -}; - -static int n_baud_table = ARRAY_SIZE(baud_table); - -/** - * tty_termios_baud_rate - * @termios: termios structure - * - * Convert termios baud rate data into a speed. This should be called - * with the termios lock held if this termios is a terminal termios - * structure. May change the termios data. - * - * Locking: none - */ - -int tty_termios_baud_rate(struct termios *termios) -{ - unsigned int cbaud; - - cbaud = termios->c_cflag & CBAUD; - - if (cbaud & CBAUDEX) { - cbaud &= ~CBAUDEX; - - if (cbaud < 1 || cbaud + 15 > n_baud_table) - termios->c_cflag &= ~CBAUDEX; - else - cbaud += 15; - } - return baud_table[cbaud]; -} - -EXPORT_SYMBOL(tty_termios_baud_rate); - -/** - * tty_get_baud_rate - get tty bit rates - * @tty: tty to query - * - * Returns the baud rate as an integer for this terminal. The - * termios lock must be held by the caller and the terminal bit - * flags may be updated. - * - * Locking: none - */ - -int tty_get_baud_rate(struct tty_struct *tty) -{ - int baud = tty_termios_baud_rate(tty->termios); - - if (baud == 38400 && tty->alt_speed) { - if (!tty->warned) { - printk(KERN_WARNING "Use of setserial/setrocket to " - "set SPD_* flags is deprecated\n"); - tty->warned = 1; - } - baud = tty->alt_speed; - } - - return baud; -} - -EXPORT_SYMBOL(tty_get_baud_rate); - /** * tty_flip_buffer_push - terminal * @tty: tty to push @@ -3758,8 +3697,8 @@ int tty_register_driver(struct tty_driver *driver) if (p) { driver->ttys = (struct tty_struct **)p; - driver->termios = (struct termios **)(p + driver->num); - driver->termios_locked = (struct termios **)(p + driver->num * 2); + driver->termios = (struct ktermios **)(p + driver->num); + driver->termios_locked = (struct ktermios **)(p + driver->num * 2); } else { driver->ttys = NULL; driver->termios = NULL; @@ -3798,7 +3737,7 @@ EXPORT_SYMBOL(tty_register_driver); int tty_unregister_driver(struct tty_driver *driver) { int i; - struct termios *tp; + struct ktermios *tp; void *p; if (driver->refcount) diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c index 3b6fa7b0be8b..0ffba4e911ca 100644 --- a/drivers/char/tty_ioctl.c +++ b/drivers/char/tty_ioctl.c @@ -36,6 +36,7 @@ #define TERMIOS_FLUSH 1 #define TERMIOS_WAIT 2 #define TERMIOS_TERMIO 4 +#define TERMIOS_OLD 8 /** @@ -84,9 +85,9 @@ stop_waiting: EXPORT_SYMBOL(tty_wait_until_sent); -static void unset_locked_termios(struct termios *termios, - struct termios *old, - struct termios *locked) +static void unset_locked_termios(struct ktermios *termios, + struct ktermios *old, + struct ktermios *locked) { int i; @@ -105,8 +106,204 @@ static void unset_locked_termios(struct termios *termios, for (i=0; i < NCCS; i++) termios->c_cc[i] = locked->c_cc[i] ? old->c_cc[i] : termios->c_cc[i]; + /* FIXME: What should we do for i/ospeed */ } +/* + * Routine which returns the baud rate of the tty + * + * Note that the baud_table needs to be kept in sync with the + * include/asm/termbits.h file. + */ +static const speed_t baud_table[] = { + 0, 50, 75, 110, 134, 150, 200, 300, 600, 1200, 1800, 2400, 4800, + 9600, 19200, 38400, 57600, 115200, 230400, 460800, +#ifdef __sparc__ + 76800, 153600, 307200, 614400, 921600 +#else + 500000, 576000, 921600, 1000000, 1152000, 1500000, 2000000, + 2500000, 3000000, 3500000, 4000000 +#endif +}; + +#ifndef __sparc__ +static const tcflag_t baud_bits[] = { + B0, B50, B75, B110, B134, B150, B200, B300, B600, + B1200, B1800, B2400, B4800, B9600, B19200, B38400, + B57600, B115200, B230400, B460800, B500000, B576000, + B921600, B1000000, B1152000, B1500000, B2000000, B2500000, + B3000000, B3500000, B4000000 +}; +#else +static const tcflag_t baud_bits[] = { + B0, B50, B75, B110, B134, B150, B200, B300, B600, + B1200, B1800, B2400, B4800, B9600, B19200, B38400, + B57600, B115200, B230400, B460800, B76800, B153600, + B307200, B614400, B921600 +}; +#endif + +static int n_baud_table = ARRAY_SIZE(baud_table); + +/** + * tty_termios_baud_rate + * @termios: termios structure + * + * Convert termios baud rate data into a speed. This should be called + * with the termios lock held if this termios is a terminal termios + * structure. May change the termios data. Device drivers can call this + * function but should use ->c_[io]speed directly as they are updated. + * + * Locking: none + */ + +speed_t tty_termios_baud_rate(struct ktermios *termios) +{ + unsigned int cbaud; + + cbaud = termios->c_cflag & CBAUD; + +#ifdef BOTHER + /* Magic token for arbitary speed via c_ispeed/c_ospeed */ + if (cbaud == BOTHER) + return termios->c_ospeed; +#endif + if (cbaud & CBAUDEX) { + cbaud &= ~CBAUDEX; + + if (cbaud < 1 || cbaud + 15 > n_baud_table) + termios->c_cflag &= ~CBAUDEX; + else + cbaud += 15; + } + return baud_table[cbaud]; +} + +EXPORT_SYMBOL(tty_termios_baud_rate); + +/** + * tty_termios_input_baud_rate + * @termios: termios structure + * + * Convert termios baud rate data into a speed. This should be called + * with the termios lock held if this termios is a terminal termios + * structure. May change the termios data. Device drivers can call this + * function but should use ->c_[io]speed directly as they are updated. + * + * Locking: none + */ + +speed_t tty_termios_input_baud_rate(struct ktermios *termios) +{ +#ifdef IBSHIFT + unsigned int cbaud = (termios->c_cflag >> IBSHIFT) & CBAUD; + + if (cbaud == B0) + return tty_termios_baud_rate(termios); + + /* Magic token for arbitary speed via c_ispeed*/ + if (cbaud == BOTHER) + return termios->c_ispeed; + + if (cbaud & CBAUDEX) { + cbaud &= ~CBAUDEX; + + if (cbaud < 1 || cbaud + 15 > n_baud_table) + termios->c_cflag &= ~(CBAUDEX << IBSHIFT); + else + cbaud += 15; + } + return baud_table[cbaud]; +#else + return tty_termios_baud_rate(termios); +#endif +} + +EXPORT_SYMBOL(tty_termios_input_baud_rate); + +#ifdef BOTHER + +/** + * tty_termios_encode_baud_rate + * @termios: termios structure + * @ispeed: input speed + * @ospeed: output speed + * + * Encode the speeds set into the passed termios structure. This is + * used as a library helper for drivers os that they can report back + * the actual speed selected when it differs from the speed requested + * + * For now input and output speed must agree. + * + * Locking: Caller should hold termios lock. This is already held + * when calling this function from the driver termios handler. + */ + +void tty_termios_encode_baud_rate(struct ktermios *termios, speed_t ibaud, speed_t obaud) +{ + int i = 0; + int ifound = 0, ofound = 0; + + termios->c_ispeed = ibaud; + termios->c_ospeed = obaud; + + termios->c_cflag &= ~CBAUD; + /* Identical speed means no input encoding (ie B0 << IBSHIFT)*/ + if (termios->c_ispeed == termios->c_ospeed) + ifound = 1; + + do { + if (obaud == baud_table[i]) { + termios->c_cflag |= baud_bits[i]; + ofound = 1; + /* So that if ibaud == obaud we don't set it */ + continue; + } + if (ibaud == baud_table[i]) { + termios->c_cflag |= (baud_bits[i] << IBSHIFT); + ifound = 1; + } + } + while(++i < n_baud_table); + if (!ofound) + termios->c_cflag |= BOTHER; + if (!ifound) + termios->c_cflag |= (BOTHER << IBSHIFT); +} + +EXPORT_SYMBOL_GPL(tty_termios_encode_baud_rate); + +#endif + +/** + * tty_get_baud_rate - get tty bit rates + * @tty: tty to query + * + * Returns the baud rate as an integer for this terminal. The + * termios lock must be held by the caller and the terminal bit + * flags may be updated. + * + * Locking: none + */ + +speed_t tty_get_baud_rate(struct tty_struct *tty) +{ + speed_t baud = tty_termios_baud_rate(tty->termios); + + if (baud == 38400 && tty->alt_speed) { + if (!tty->warned) { + printk(KERN_WARNING "Use of setserial/setrocket to " + "set SPD_* flags is deprecated\n"); + tty->warned = 1; + } + baud = tty->alt_speed; + } + + return baud; +} + +EXPORT_SYMBOL(tty_get_baud_rate); + /** * change_termios - update termios values * @tty: tty to update @@ -119,10 +316,10 @@ static void unset_locked_termios(struct termios *termios, * Locking: termios_sem */ -static void change_termios(struct tty_struct * tty, struct termios * new_termios) +static void change_termios(struct tty_struct * tty, struct ktermios * new_termios) { int canon_change; - struct termios old_termios = *tty->termios; + struct ktermios old_termios = *tty->termios; struct tty_ldisc *ld; /* @@ -195,7 +392,7 @@ static void change_termios(struct tty_struct * tty, struct termios * new_termios static int set_termios(struct tty_struct * tty, void __user *arg, int opt) { - struct termios tmp_termios; + struct ktermios tmp_termios; struct tty_ldisc *ld; int retval = tty_check_change(tty); @@ -203,16 +400,28 @@ static int set_termios(struct tty_struct * tty, void __user *arg, int opt) return retval; if (opt & TERMIOS_TERMIO) { - memcpy(&tmp_termios, tty->termios, sizeof(struct termios)); + memcpy(&tmp_termios, tty->termios, sizeof(struct ktermios)); if (user_termio_to_kernel_termios(&tmp_termios, (struct termio __user *)arg)) return -EFAULT; +#ifdef TCGETS2 + } else if (opt & TERMIOS_OLD) { + memcpy(&tmp_termios, tty->termios, sizeof(struct termios)); + if (user_termios_to_kernel_termios_1(&tmp_termios, + (struct termios_v1 __user *)arg)) + return -EFAULT; +#endif } else { if (user_termios_to_kernel_termios(&tmp_termios, (struct termios __user *)arg)) return -EFAULT; } + /* If old style Bfoo values are used then load c_ispeed/c_ospeed with the real speed + so its unconditionally usable */ + tmp_termios.c_ispeed = tty_termios_input_baud_rate(&tmp_termios); + tmp_termios.c_ospeed = tty_termios_baud_rate(&tmp_termios); + ld = tty_ldisc_ref(tty); if (ld != NULL) { @@ -286,8 +495,8 @@ static int get_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb) struct sgttyb tmp; mutex_lock(&tty->termios_mutex); - tmp.sg_ispeed = 0; - tmp.sg_ospeed = 0; + tmp.sg_ispeed = tty->c_ispeed; + tmp.sg_ospeed = tty->c_ospeed; tmp.sg_erase = tty->termios->c_cc[VERASE]; tmp.sg_kill = tty->termios->c_cc[VKILL]; tmp.sg_flags = get_sgflags(tty); @@ -296,7 +505,7 @@ static int get_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb) return copy_to_user(sgttyb, &tmp, sizeof(tmp)) ? -EFAULT : 0; } -static void set_sgflags(struct termios * termios, int flags) +static void set_sgflags(struct ktermios * termios, int flags) { termios->c_iflag = ICRNL | IXON; termios->c_oflag = 0; @@ -337,7 +546,7 @@ static int set_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb) { int retval; struct sgttyb tmp; - struct termios termios; + struct ktermios termios; retval = tty_check_change(tty); if (retval) @@ -351,6 +560,10 @@ static int set_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb) termios.c_cc[VERASE] = tmp.sg_erase; termios.c_cc[VKILL] = tmp.sg_kill; set_sgflags(&termios, tmp.sg_flags); + /* Try and encode into Bfoo format */ +#ifdef BOTHER + tty_termios_encode_baud_rate(&termios, termios.c_ispeed, termios.c_ospeed); +#endif mutex_unlock(&tty->termios_mutex); change_termios(tty, &termios); return 0; @@ -481,16 +694,33 @@ int n_tty_ioctl(struct tty_struct * tty, struct file * file, case TIOCSLTC: return set_ltchars(real_tty, p); #endif + case TCSETSF: + return set_termios(real_tty, p, TERMIOS_FLUSH | TERMIOS_WAIT | TERMIOS_OLD); + case TCSETSW: + return set_termios(real_tty, p, TERMIOS_WAIT | TERMIOS_OLD); + case TCSETS: + return set_termios(real_tty, p, TERMIOS_OLD); +#ifndef TCGETS2 case TCGETS: if (kernel_termios_to_user_termios((struct termios __user *)arg, real_tty->termios)) return -EFAULT; return 0; - case TCSETSF: +#else + case TCGETS: + if (kernel_termios_to_user_termios_1((struct termios_v1 __user *)arg, real_tty->termios)) + return -EFAULT; + return 0; + case TCGETS2: + if (kernel_termios_to_user_termios((struct termios __user *)arg, real_tty->termios)) + return -EFAULT; + return 0; + case TCSETSF2: return set_termios(real_tty, p, TERMIOS_FLUSH | TERMIOS_WAIT); - case TCSETSW: + case TCSETSW2: return set_termios(real_tty, p, TERMIOS_WAIT); - case TCSETS: + case TCSETS2: return set_termios(real_tty, p, 0); +#endif case TCGETA: return get_termio(real_tty, p); case TCSETAF: diff --git a/include/linux/tty.h b/include/linux/tty.h index 1d29999a3439..65cbcf22c31e 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -175,7 +175,7 @@ struct tty_struct { int index; struct tty_ldisc ldisc; struct mutex termios_mutex; - struct termios *termios, *termios_locked; + struct ktermios *termios, *termios_locked; char name[64]; int pgrp; int session; @@ -258,7 +258,7 @@ struct tty_struct { extern void tty_write_flush(struct tty_struct *); -extern struct termios tty_std_termios; +extern struct ktermios tty_std_termios; extern int kmsg_redirect; @@ -293,8 +293,9 @@ extern int tty_hung_up_p(struct file * filp); extern void do_SAK(struct tty_struct *tty); extern void disassociate_ctty(int priv); extern void tty_flip_buffer_push(struct tty_struct *tty); -extern int tty_get_baud_rate(struct tty_struct *tty); -extern int tty_termios_baud_rate(struct termios *termios); +extern speed_t tty_get_baud_rate(struct tty_struct *tty); +extern speed_t tty_termios_baud_rate(struct ktermios *termios); +extern speed_t tty_termios_input_baud_rate(struct ktermios *termios); extern struct tty_ldisc *tty_ldisc_ref(struct tty_struct *); extern void tty_ldisc_deref(struct tty_ldisc *); diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h index 5c8473bb6882..659487e3ebeb 100644 --- a/include/linux/tty_driver.h +++ b/include/linux/tty_driver.h @@ -53,7 +53,7 @@ * device-specific ioctl's. If the ioctl number passed in cmd * is not recognized by the driver, it should return ENOIOCTLCMD. * - * void (*set_termios)(struct tty_struct *tty, struct termios * old); + * void (*set_termios)(struct tty_struct *tty, struct ktermios * old); * * This routine allows the tty driver to be notified when * device's termios settings have changed. Note that a @@ -132,7 +132,7 @@ struct tty_operations { int (*chars_in_buffer)(struct tty_struct *tty); int (*ioctl)(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg); - void (*set_termios)(struct tty_struct *tty, struct termios * old); + void (*set_termios)(struct tty_struct *tty, struct ktermios * old); void (*throttle)(struct tty_struct * tty); void (*unthrottle)(struct tty_struct * tty); void (*stop)(struct tty_struct *tty); @@ -165,7 +165,7 @@ struct tty_driver { int num; /* number of devices allocated */ short type; /* type of tty driver */ short subtype; /* subtype of tty driver */ - struct termios init_termios; /* Initial termios */ + struct ktermios init_termios; /* Initial termios */ int flags; /* tty driver flags */ int refcount; /* for loadable tty drivers */ struct proc_dir_entry *proc_entry; /* /proc fs entry */ @@ -175,8 +175,8 @@ struct tty_driver { * Pointer to the tty data structures */ struct tty_struct **ttys; - struct termios **termios; - struct termios **termios_locked; + struct ktermios **termios; + struct ktermios **termios_locked; void *driver_state; /* only used for the PTY driver */ /* @@ -193,7 +193,7 @@ struct tty_driver { int (*chars_in_buffer)(struct tty_struct *tty); int (*ioctl)(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg); - void (*set_termios)(struct tty_struct *tty, struct termios * old); + void (*set_termios)(struct tty_struct *tty, struct ktermios * old); void (*throttle)(struct tty_struct * tty); void (*unthrottle)(struct tty_struct * tty); void (*stop)(struct tty_struct *tty); diff --git a/include/linux/tty_ldisc.h b/include/linux/tty_ldisc.h index 83c6e6c10ebb..d75932e27710 100644 --- a/include/linux/tty_ldisc.h +++ b/include/linux/tty_ldisc.h @@ -59,7 +59,7 @@ * low-level driver can "grab" an ioctl request before the line * discpline has a chance to see it. * - * void (*set_termios)(struct tty_struct *tty, struct termios * old); + * void (*set_termios)(struct tty_struct *tty, struct ktermios * old); * * This function notifies the line discpline that a change has * been made to the termios structure. @@ -118,7 +118,7 @@ struct tty_ldisc { const unsigned char * buf, size_t nr); int (*ioctl)(struct tty_struct * tty, struct file * file, unsigned int cmd, unsigned long arg); - void (*set_termios)(struct tty_struct *tty, struct termios * old); + void (*set_termios)(struct tty_struct *tty, struct ktermios * old); unsigned int (*poll)(struct tty_struct *, struct file *, struct poll_table_struct *); int (*hangup)(struct tty_struct *tty); diff --git a/include/net/irda/ircomm_tty.h b/include/net/irda/ircomm_tty.h index 87699cb4ef8c..8dabdd603fe1 100644 --- a/include/net/irda/ircomm_tty.h +++ b/include/net/irda/ircomm_tty.h @@ -126,7 +126,7 @@ extern int ircomm_tty_tiocmset(struct tty_struct *tty, struct file *file, extern int ircomm_tty_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg); extern void ircomm_tty_set_termios(struct tty_struct *tty, - struct termios *old_termios); + struct ktermios *old_termios); extern hashbin_t *ircomm_tty; #endif diff --git a/net/irda/ircomm/ircomm_tty_ioctl.c b/net/irda/ircomm/ircomm_tty_ioctl.c index 197e3e7ed7e2..75e39ea599d8 100644 --- a/net/irda/ircomm/ircomm_tty_ioctl.c +++ b/net/irda/ircomm/ircomm_tty_ioctl.c @@ -146,7 +146,7 @@ static void ircomm_tty_change_speed(struct ircomm_tty_cb *self) * do something rational. */ void ircomm_tty_set_termios(struct tty_struct *tty, - struct termios *old_termios) + struct ktermios *old_termios) { struct ircomm_tty_cb *self = (struct ircomm_tty_cb *) tty->driver_data; unsigned int cflag = tty->termios->c_cflag; -- cgit v1.2.3 From 606d099cdd1080bbb50ea50dc52d98252f8f10a1 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Fri, 8 Dec 2006 02:38:45 -0800 Subject: [PATCH] tty: switch to ktermios This is the grungy swap all the occurrences in the right places patch that goes with the updates. At this point we have the same functionality as before (except that sgttyb() returns speeds not zero) and are ready to begin turning new stuff on providing nobody reports lots of bugs If you are a tty driver author converting an out of tree driver the only impact should be termios->ktermios name changes for the speed/property setting functions from your upper layers. If you are implementing your own TCGETS function before then your driver was broken already and its about to get a whole lot more painful for you so please fix it 8) Also fill in c_ispeed/ospeed on init for most devices, although the current code will do this for you anyway but I'd like eventually to lose that extra paranoia [akpm@osdl.org: bluetooth fix] [mp3@de.ibm.com: sclp fix] [mp3@de.ibm.com: warning fix for tty3270] [hugh@veritas.com: fix tty_ioctl powerpc build] [jdike@addtoit.com: uml: fix ->set_termios declaration] Signed-off-by: Alan Cox Signed-off-by: Martin Peschke Acked-by: Peter Oberparleiter Cc: Cornelia Huck Signed-off-by: Hugh Dickins Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/drivers/line.c | 2 +- arch/um/include/line.h | 2 +- drivers/char/amiserial.c | 6 +++--- drivers/char/cyclades.c | 2 +- drivers/char/epca.c | 16 +++++++++++----- drivers/char/esp.c | 2 +- drivers/char/generic_serial.c | 4 ++-- drivers/char/hvcs.c | 6 ++++-- drivers/char/hvsi.c | 2 ++ drivers/char/ip2/ip2main.c | 10 +++++----- drivers/char/isicom.c | 2 +- drivers/char/istallion.c | 14 ++++++++------ drivers/char/moxa.c | 18 ++++++++++-------- drivers/char/mxser.c | 18 ++++++++++-------- drivers/char/mxser_new.c | 8 ++++---- drivers/char/n_r3964.c | 4 ++-- drivers/char/n_tty.c | 2 +- drivers/char/pcmcia/synclink_cs.c | 2 +- drivers/char/pty.c | 10 +++++++++- drivers/char/riscom8.c | 4 +++- drivers/char/rocket.c | 8 +++++--- drivers/char/ser_a2232.c | 2 ++ drivers/char/serial167.c | 2 +- drivers/char/specialix.c | 9 +++++++-- drivers/char/stallion.c | 20 +++++++++++--------- drivers/char/sx.c | 2 ++ drivers/char/synclink.c | 4 +++- drivers/char/synclink_gt.c | 6 ++++-- drivers/char/synclinkmp.c | 6 ++++-- drivers/char/tty_ioctl.c | 4 ++-- drivers/char/vme_scc.c | 2 ++ drivers/isdn/capi/capi.c | 2 +- drivers/isdn/gigaset/interface.c | 4 ++-- drivers/isdn/i4l/isdn_tty.c | 2 +- drivers/net/irda/irtty-sir.c | 4 ++-- drivers/s390/char/sclp_tty.c | 2 -- drivers/s390/char/tty3270.c | 2 +- drivers/serial/21285.c | 4 ++-- drivers/serial/68328serial.c | 2 +- drivers/serial/68360serial.c | 2 +- drivers/serial/8250.c | 4 ++-- drivers/serial/amba-pl010.c | 4 ++-- drivers/serial/amba-pl011.c | 4 ++-- drivers/serial/atmel_serial.c | 2 +- drivers/serial/clps711x.c | 4 ++-- drivers/serial/crisv10.c | 8 +++++--- drivers/serial/crisv10.h | 4 ++-- drivers/serial/dz.c | 4 ++-- drivers/serial/icom.c | 4 ++-- drivers/serial/imx.c | 4 ++-- drivers/serial/ioc3_serial.c | 4 ++-- drivers/serial/ioc4_serial.c | 6 +++--- drivers/serial/ip22zilog.c | 4 ++-- drivers/serial/jsm/jsm_tty.c | 10 +++++----- drivers/serial/m32r_sio.c | 2 +- drivers/serial/mcfserial.c | 2 +- drivers/serial/mpc52xx_uart.c | 4 ++-- drivers/serial/mpsc.c | 4 ++-- drivers/serial/mux.c | 4 ++-- drivers/serial/netx-serial.c | 4 ++-- drivers/serial/pmac_zilog.c | 10 +++++----- drivers/serial/pmac_zilog.h | 2 +- drivers/serial/pxa.c | 4 ++-- drivers/serial/s3c2410.c | 4 ++-- drivers/serial/sa1100.c | 4 ++-- drivers/serial/serial_core.c | 21 +++++++++++---------- drivers/serial/serial_lh7a40x.c | 4 ++-- drivers/serial/serial_txx9.c | 4 ++-- drivers/serial/sh-sci.c | 4 ++-- drivers/serial/sn_console.c | 4 ++-- drivers/serial/sunhv.c | 4 ++-- drivers/serial/sunsab.c | 4 ++-- drivers/serial/sunsu.c | 4 ++-- drivers/serial/sunzilog.c | 4 ++-- drivers/serial/uartlite.c | 4 ++-- drivers/serial/v850e_uart.c | 4 ++-- drivers/serial/vr41xx_siu.c | 4 ++-- drivers/tc/zs.c | 2 +- drivers/usb/class/cdc-acm.c | 4 ++-- drivers/usb/gadget/serial.c | 4 ++-- drivers/usb/serial/ark3116.c | 4 ++-- drivers/usb/serial/belkin_sa.c | 4 ++-- drivers/usb/serial/console.c | 2 +- drivers/usb/serial/cp2101.c | 4 ++-- drivers/usb/serial/cypress_m8.c | 10 +++++----- drivers/usb/serial/digi_acceleport.c | 6 +++--- drivers/usb/serial/empeg.c | 4 ++-- drivers/usb/serial/ftdi_sio.c | 4 ++-- drivers/usb/serial/io_edgeport.c | 8 ++++---- drivers/usb/serial/io_ti.c | 6 +++--- drivers/usb/serial/ir-usb.c | 4 ++-- drivers/usb/serial/keyspan.c | 2 +- drivers/usb/serial/keyspan.h | 2 +- drivers/usb/serial/keyspan_pda.c | 2 +- drivers/usb/serial/kl5kusb105.c | 6 +++--- drivers/usb/serial/kobil_sct.c | 10 +++++----- drivers/usb/serial/mct_u232.c | 4 ++-- drivers/usb/serial/mos7720.c | 4 ++-- drivers/usb/serial/mos7840.c | 4 ++-- drivers/usb/serial/option.c | 4 ++-- drivers/usb/serial/pl2303.c | 4 ++-- drivers/usb/serial/sierra.c | 2 +- drivers/usb/serial/ti_usb_3410_5052.c | 4 ++-- drivers/usb/serial/usb-serial.c | 2 +- drivers/usb/serial/visor.c | 4 ++-- drivers/usb/serial/whiteheat.c | 6 +++--- include/asm-generic/termios.h | 4 ++-- include/asm-powerpc/termbits.h | 2 +- include/linux/generic_serial.h | 2 +- include/linux/isdn.h | 8 ++++---- include/linux/serial_core.h | 8 ++++---- include/linux/usb/serial.h | 2 +- net/bluetooth/rfcomm/tty.c | 4 ++-- 113 files changed, 301 insertions(+), 253 deletions(-) (limited to 'include/linux') diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index aa3090d05a8f..83301e1ef67c 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -246,7 +246,7 @@ out_up: return ret; } -void line_set_termios(struct tty_struct *tty, struct termios * old) +void line_set_termios(struct tty_struct *tty, struct ktermios * old) { /* nothing */ } diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 214ee76c40df..5f232ae89fbb 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -76,7 +76,7 @@ extern int line_setup(struct line *lines, unsigned int sizeof_lines, extern int line_write(struct tty_struct *tty, const unsigned char *buf, int len); extern void line_put_char(struct tty_struct *tty, unsigned char ch); -extern void line_set_termios(struct tty_struct *tty, struct termios * old); +extern void line_set_termios(struct tty_struct *tty, struct ktermios * old); extern int line_chars_in_buffer(struct tty_struct *tty); extern void line_flush_buffer(struct tty_struct *tty); extern void line_flush_chars(struct tty_struct *tty); diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c index 66086fa2d59a..feb4ac802a0d 100644 --- a/drivers/char/amiserial.c +++ b/drivers/char/amiserial.c @@ -104,7 +104,7 @@ static struct async_struct *IRQ_ports; static unsigned char current_ctl_bits; -static void change_speed(struct async_struct *info, struct termios *old); +static void change_speed(struct async_struct *info, struct ktermios *old); static void rs_wait_until_sent(struct tty_struct *tty, int timeout); @@ -694,7 +694,7 @@ static void shutdown(struct async_struct * info) * the specified baud rate for a serial port. */ static void change_speed(struct async_struct *info, - struct termios *old_termios) + struct ktermios *old_termios) { int quot = 0, baud_base, baud; unsigned cflag, cval = 0; @@ -1365,7 +1365,7 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file, return 0; } -static void rs_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct async_struct *info = (struct async_struct *)tty->driver_data; unsigned long flags; diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c index acb2de5e3a98..3bb4e534c14e 100644 --- a/drivers/char/cyclades.c +++ b/drivers/char/cyclades.c @@ -4101,7 +4101,7 @@ cy_ioctl(struct tty_struct *tty, struct file * file, * where old == NULL, and try to do something rational. */ static void -cy_set_termios(struct tty_struct *tty, struct termios * old_termios) +cy_set_termios(struct tty_struct *tty, struct ktermios * old_termios) { struct cyclades_port *info = (struct cyclades_port *)tty->driver_data; diff --git a/drivers/char/epca.c b/drivers/char/epca.c index 7c71eb779802..a0f822c9d74d 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -199,7 +199,7 @@ static int pc_ioctl(struct tty_struct *, struct file *, unsigned int, unsigned long); static int info_ioctl(struct tty_struct *, struct file *, unsigned int, unsigned long); -static void pc_set_termios(struct tty_struct *, struct termios *); +static void pc_set_termios(struct tty_struct *, struct ktermios *); static void do_softint(struct work_struct *work); static void pc_stop(struct tty_struct *); static void pc_start(struct tty_struct *); @@ -1236,6 +1236,8 @@ static int __init pc_init(void) pc_driver->init_termios.c_oflag = 0; pc_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | CLOCAL | HUPCL; pc_driver->init_termios.c_lflag = 0; + pc_driver->init_termios.c_ispeed = 9600; + pc_driver->init_termios.c_ospeed = 9600; pc_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(pc_driver, &pc_ops); @@ -1250,6 +1252,8 @@ static int __init pc_init(void) pc_info->init_termios.c_oflag = 0; pc_info->init_termios.c_lflag = 0; pc_info->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL; + pc_info->init_termios.c_ispeed = 9600; + pc_info->init_termios.c_ospeed = 9600; pc_info->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(pc_info, &info_ops); @@ -1999,7 +2003,7 @@ static void epcaparam(struct tty_struct *tty, struct channel *ch) { /* Begin epcaparam */ unsigned int cmdHead; - struct termios *ts; + struct ktermios *ts; struct board_chan __iomem *bc; unsigned mval, hflow, cflag, iflag; @@ -2114,7 +2118,7 @@ static void receive_data(struct channel *ch) { /* Begin receive_data */ unchar *rptr; - struct termios *ts = NULL; + struct ktermios *ts = NULL; struct tty_struct *tty; struct board_chan __iomem *bc; int dataToRead, wrapgap, bytesAvailable; @@ -2362,12 +2366,14 @@ static int pc_ioctl(struct tty_struct *tty, struct file * file, switch (cmd) { /* Begin switch cmd */ +#if 0 /* Handled by calling layer properly */ case TCGETS: - if (copy_to_user(argp, tty->termios, sizeof(struct termios))) + if (copy_to_user(argp, tty->termios, sizeof(struct ktermios))) return -EFAULT; return 0; case TCGETA: return get_termio(tty, argp); +#endif case TCSBRK: /* SVID version: non-zero arg --> no break */ retval = tty_check_change(tty); if (retval) @@ -2536,7 +2542,7 @@ static int pc_ioctl(struct tty_struct *tty, struct file * file, /* --------------------- Begin pc_set_termios ----------------------- */ -static void pc_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void pc_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { /* Begin pc_set_termios */ struct channel *ch; diff --git a/drivers/char/esp.c b/drivers/char/esp.c index 93b551962513..d1bfbaa2aa02 100644 --- a/drivers/char/esp.c +++ b/drivers/char/esp.c @@ -1915,7 +1915,7 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file, return 0; } -static void rs_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct esp_struct *info = (struct esp_struct *)tty->driver_data; unsigned long flags; diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c index 87127e49c0db..e769811e7417 100644 --- a/drivers/char/generic_serial.c +++ b/drivers/char/generic_serial.c @@ -718,11 +718,11 @@ static unsigned int gs_baudrates[] = { void gs_set_termios (struct tty_struct * tty, - struct termios * old_termios) + struct ktermios * old_termios) { struct gs_port *port; int baudrate, tmp, rv; - struct termios *tiosp; + struct ktermios *tiosp; func_enter(); diff --git a/drivers/char/hvcs.c b/drivers/char/hvcs.c index d090622f1dea..207f7343ba60 100644 --- a/drivers/char/hvcs.c +++ b/drivers/char/hvcs.c @@ -192,11 +192,13 @@ MODULE_VERSION(HVCS_DRIVER_VERSION); * that will cause echoing or we'll go into recursive loop echoing chars back * and forth with the console drivers. */ -static struct termios hvcs_tty_termios = { +static struct ktermios hvcs_tty_termios = { .c_iflag = IGNBRK | IGNPAR, .c_oflag = OPOST, .c_cflag = B38400 | CS8 | CREAD | HUPCL, - .c_cc = INIT_C_CC + .c_cc = INIT_C_CC, + .c_ispeed = 38400, + .c_ospeed = 38400 }; /* diff --git a/drivers/char/hvsi.c b/drivers/char/hvsi.c index 82a41d5b4ed0..d7806834fc17 100644 --- a/drivers/char/hvsi.c +++ b/drivers/char/hvsi.c @@ -1161,6 +1161,8 @@ static int __init hvsi_init(void) hvsi_driver->type = TTY_DRIVER_TYPE_SYSTEM; hvsi_driver->init_termios = tty_std_termios; hvsi_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL; + hvsi_driver->init_termios.c_ispeed = 9600; + hvsi_driver->init_termios.c_ospeed = 9600; hvsi_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(hvsi_driver, &hvsi_ops); diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c index 6810b7bbcbb7..7c70310a49b5 100644 --- a/drivers/char/ip2/ip2main.c +++ b/drivers/char/ip2/ip2main.c @@ -177,7 +177,7 @@ static int ip2_write_room(PTTY); static int ip2_chars_in_buf(PTTY); static void ip2_flush_buffer(PTTY); static int ip2_ioctl(PTTY, struct file *, UINT, ULONG); -static void ip2_set_termios(PTTY, struct termios *); +static void ip2_set_termios(PTTY, struct ktermios *); static void ip2_set_line_discipline(PTTY); static void ip2_throttle(PTTY); static void ip2_unthrottle(PTTY); @@ -198,7 +198,7 @@ static void do_status(struct work_struct *); static void ip2_wait_until_sent(PTTY,int); -static void set_params (i2ChanStrPtr, struct termios *); +static void set_params (i2ChanStrPtr, struct ktermios *); static int get_serial_info(i2ChanStrPtr, struct serial_struct __user *); static int set_serial_info(i2ChanStrPtr, struct serial_struct __user *); @@ -2398,7 +2398,7 @@ set_serial_info( i2ChanStrPtr pCh, struct serial_struct __user *new_info ) /* */ /******************************************************************************/ static void -ip2_set_termios( PTTY tty, struct termios *old_termios ) +ip2_set_termios( PTTY tty, struct ktermios *old_termios ) { i2ChanStrPtr pCh = (i2ChanStrPtr)tty->driver_data; @@ -2440,11 +2440,11 @@ ip2_set_line_discipline ( PTTY tty ) /* change. */ /******************************************************************************/ static void -set_params( i2ChanStrPtr pCh, struct termios *o_tios ) +set_params( i2ChanStrPtr pCh, struct ktermios *o_tios ) { tcflag_t cflag, iflag, lflag; char stop_char, start_char; - struct termios dummy; + struct ktermios dummy; lflag = pCh->pTTY->termios->c_lflag; cflag = pCh->pTTY->termios->c_cflag; diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c index 1637c1d9a4ba..29e28b756336 100644 --- a/drivers/char/isicom.c +++ b/drivers/char/isicom.c @@ -1399,7 +1399,7 @@ static int isicom_ioctl(struct tty_struct *tty, struct file *filp, /* set_termios et all */ static void isicom_set_termios(struct tty_struct *tty, - struct termios *old_termios) + struct ktermios *old_termios) { struct isi_port *port = tty->driver_data; diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index b6f0d036a760..0ef2523733ee 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -194,9 +194,11 @@ static struct tty_struct *stli_txcooktty; * with this termios initially. Basically all it defines is a raw port * at 9600 baud, 8 data bits, no parity, 1 stop bit. */ -static struct termios stli_deftermios = { +static struct ktermios stli_deftermios = { .c_cflag = (B9600 | CS8 | CREAD | HUPCL | CLOCAL), .c_cc = INIT_C_CC, + .c_ispeed = 9600, + .c_ospeed = 9600, }; /* @@ -639,7 +641,7 @@ static void stli_flushchars(struct tty_struct *tty); static int stli_writeroom(struct tty_struct *tty); static int stli_charsinbuffer(struct tty_struct *tty); static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg); -static void stli_settermios(struct tty_struct *tty, struct termios *old); +static void stli_settermios(struct tty_struct *tty, struct ktermios *old); static void stli_throttle(struct tty_struct *tty); static void stli_unthrottle(struct tty_struct *tty); static void stli_stop(struct tty_struct *tty); @@ -669,7 +671,7 @@ static int stli_cmdwait(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, v static void stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback); static void __stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback); static void stli_dodelaycmd(stliport_t *portp, cdkctrl_t __iomem *cp); -static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct termios *tiosp); +static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct ktermios *tiosp); static void stli_mkasysigs(asysigs_t *sp, int dtr, int rts); static long stli_mktiocm(unsigned long sigvalue); static void stli_read(stlibrd_t *brdp, stliport_t *portp); @@ -1889,11 +1891,11 @@ static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cm * Looks like it is true for the current ttys implementation..!! */ -static void stli_settermios(struct tty_struct *tty, struct termios *old) +static void stli_settermios(struct tty_struct *tty, struct ktermios *old) { stliport_t *portp; stlibrd_t *brdp; - struct termios *tiosp; + struct ktermios *tiosp; asyport_t aport; if (tty == NULL) @@ -2730,7 +2732,7 @@ static void stli_poll(unsigned long arg) * the slave. */ -static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct termios *tiosp) +static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct ktermios *tiosp) { memset(pp, 0, sizeof(asyport_t)); diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c index 8b316953173d..f391a24a1b44 100644 --- a/drivers/char/moxa.c +++ b/drivers/char/moxa.c @@ -234,7 +234,7 @@ static void moxa_put_char(struct tty_struct *, unsigned char); static int moxa_ioctl(struct tty_struct *, struct file *, unsigned int, unsigned long); static void moxa_throttle(struct tty_struct *); static void moxa_unthrottle(struct tty_struct *); -static void moxa_set_termios(struct tty_struct *, struct termios *); +static void moxa_set_termios(struct tty_struct *, struct ktermios *); static void moxa_stop(struct tty_struct *); static void moxa_start(struct tty_struct *); static void moxa_hangup(struct tty_struct *); @@ -261,7 +261,7 @@ static void MoxaPortEnable(int); static void MoxaPortDisable(int); static long MoxaPortGetMaxBaud(int); static long MoxaPortSetBaud(int, long); -static int MoxaPortSetTermio(int, struct termios *, speed_t); +static int MoxaPortSetTermio(int, struct ktermios *, speed_t); static int MoxaPortGetLineOut(int, int *, int *); static void MoxaPortLineCtrl(int, int, int); static void MoxaPortFlowCtrl(int, int, int, int, int, int); @@ -355,6 +355,8 @@ static int __init moxa_init(void) moxaDriver->init_termios.c_oflag = 0; moxaDriver->init_termios.c_cflag = B9600 | CS8 | CREAD | CLOCAL | HUPCL; moxaDriver->init_termios.c_lflag = 0; + moxaDriver->init_termios.c_ispeed = 9600; + moxaDriver->init_termios.c_ospeed = 9600; moxaDriver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(moxaDriver, &moxa_ops); @@ -864,7 +866,7 @@ static void moxa_unthrottle(struct tty_struct *tty) } static void moxa_set_termios(struct tty_struct *tty, - struct termios *old_termios) + struct ktermios *old_termios) { struct moxa_str *ch = (struct moxa_str *) tty->driver_data; @@ -978,7 +980,7 @@ static void moxa_poll(unsigned long ignored) static void set_tty_param(struct tty_struct *tty) { - register struct termios *ts; + register struct ktermios *ts; struct moxa_str *ch; int rts, cts, txflow, rxflow, xany; @@ -1149,7 +1151,7 @@ static void shut_down(struct moxa_str *ch) static void receive_data(struct moxa_str *ch) { struct tty_struct *tp; - struct termios *ts; + struct ktermios *ts; unsigned long flags; ts = NULL; @@ -1912,9 +1914,9 @@ int MoxaPortsOfCard(int cardno) * * Function 12: Configure the port. * Syntax: - * int MoxaPortSetTermio(int port, struct termios *termio, speed_t baud); + * int MoxaPortSetTermio(int port, struct ktermios *termio, speed_t baud); * int port : port number (0 - 127) - * struct termios * termio : termio structure pointer + * struct ktermios * termio : termio structure pointer * speed_t baud : baud rate * * return: -1 : this port is invalid or termio == NULL @@ -2195,7 +2197,7 @@ long MoxaPortSetBaud(int port, long baud) return (baud); } -int MoxaPortSetTermio(int port, struct termios *termio, speed_t baud) +int MoxaPortSetTermio(int port, struct ktermios *termio, speed_t baud) { void __iomem *ofsAddr; tcflag_t cflag; diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c index 2dc49be144e6..c063359baf78 100644 --- a/drivers/char/mxser.c +++ b/drivers/char/mxser.c @@ -328,8 +328,8 @@ struct mxser_struct { int xmit_tail; int xmit_cnt; struct work_struct tqueue; - struct termios normal_termios; - struct termios callout_termios; + struct ktermios normal_termios; + struct ktermios callout_termios; wait_queue_head_t open_wait; wait_queue_head_t close_wait; wait_queue_head_t delta_msr_wait; @@ -364,8 +364,8 @@ static int mxserBoardCAP[MXSER_BOARDS] = { static struct tty_driver *mxvar_sdriver; static struct mxser_struct mxvar_table[MXSER_PORTS]; static struct tty_struct *mxvar_tty[MXSER_PORTS + 1]; -static struct termios *mxvar_termios[MXSER_PORTS + 1]; -static struct termios *mxvar_termios_locked[MXSER_PORTS + 1]; +static struct ktermios *mxvar_termios[MXSER_PORTS + 1]; +static struct ktermios *mxvar_termios_locked[MXSER_PORTS + 1]; static struct mxser_log mxvar_log; static int mxvar_diagflag; static unsigned char mxser_msr[MXSER_PORTS + 1]; @@ -402,7 +402,7 @@ static int mxser_ioctl(struct tty_struct *, struct file *, uint, ulong); static int mxser_ioctl_special(unsigned int, void __user *); static void mxser_throttle(struct tty_struct *); static void mxser_unthrottle(struct tty_struct *); -static void mxser_set_termios(struct tty_struct *, struct termios *); +static void mxser_set_termios(struct tty_struct *, struct ktermios *); static void mxser_stop(struct tty_struct *); static void mxser_start(struct tty_struct *); static void mxser_hangup(struct tty_struct *); @@ -414,7 +414,7 @@ static void mxser_check_modem_status(struct mxser_struct *, int); static int mxser_block_til_ready(struct tty_struct *, struct file *, struct mxser_struct *); static int mxser_startup(struct mxser_struct *); static void mxser_shutdown(struct mxser_struct *); -static int mxser_change_speed(struct mxser_struct *, struct termios *old_termios); +static int mxser_change_speed(struct mxser_struct *, struct ktermios *old_termios); static int mxser_get_serial_info(struct mxser_struct *, struct serial_struct __user *); static int mxser_set_serial_info(struct mxser_struct *, struct serial_struct __user *); static int mxser_get_lsr_info(struct mxser_struct *, unsigned int __user *); @@ -726,6 +726,8 @@ static int mxser_init(void) mxvar_sdriver->subtype = SERIAL_TYPE_NORMAL; mxvar_sdriver->init_termios = tty_std_termios; mxvar_sdriver->init_termios.c_cflag = B9600|CS8|CREAD|HUPCL|CLOCAL; + mxvar_sdriver->init_termios.c_ispeed = 9600; + mxvar_sdriver->init_termios.c_ospeed = 9600; mxvar_sdriver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(mxvar_sdriver, &mxser_ops); mxvar_sdriver->ttys = mxvar_tty; @@ -1749,7 +1751,7 @@ static void mxser_unthrottle(struct tty_struct *tty) /* MX_UNLOCK(&info->slock); */ } -static void mxser_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void mxser_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct mxser_struct *info = tty->driver_data; unsigned long flags; @@ -2541,7 +2543,7 @@ static void mxser_shutdown(struct mxser_struct *info) * This routine is called to set the UART divisor registers to match * the specified baud rate for a serial port. */ -static int mxser_change_speed(struct mxser_struct *info, struct termios *old_termios) +static int mxser_change_speed(struct mxser_struct *info, struct ktermios *old_termios) { unsigned cflag, cval, fcr; int ret = 0; diff --git a/drivers/char/mxser_new.c b/drivers/char/mxser_new.c index 369d2742d6be..efa8076c33e0 100644 --- a/drivers/char/mxser_new.c +++ b/drivers/char/mxser_new.c @@ -266,8 +266,8 @@ struct mxser_port { int xmit_tail; int xmit_cnt; - struct termios normal_termios; - struct termios callout_termios; + struct ktermios normal_termios; + struct ktermios callout_termios; struct mxser_mon mon_data; @@ -512,7 +512,7 @@ static int mxser_set_baud(struct mxser_port *info, long newspd) * the specified baud rate for a serial port. */ static int mxser_change_speed(struct mxser_port *info, - struct termios *old_termios) + struct ktermios *old_termios) { unsigned cflag, cval, fcr; int ret = 0; @@ -1966,7 +1966,7 @@ static void mxser_start(struct tty_struct *tty) spin_unlock_irqrestore(&info->slock, flags); } -static void mxser_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void mxser_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct mxser_port *info = tty->driver_data; unsigned long flags; diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c index 203dc2b661d5..103d338f21e2 100644 --- a/drivers/char/n_r3964.c +++ b/drivers/char/n_r3964.c @@ -142,7 +142,7 @@ static ssize_t r3964_write(struct tty_struct * tty, struct file * file, const unsigned char * buf, size_t nr); static int r3964_ioctl(struct tty_struct * tty, struct file * file, unsigned int cmd, unsigned long arg); -static void r3964_set_termios(struct tty_struct *tty, struct termios * old); +static void r3964_set_termios(struct tty_struct *tty, struct ktermios * old); static unsigned int r3964_poll(struct tty_struct * tty, struct file * file, struct poll_table_struct *wait); static void r3964_receive_buf(struct tty_struct *tty, const unsigned char *cp, @@ -1347,7 +1347,7 @@ static int r3964_ioctl(struct tty_struct * tty, struct file * file, } } -static void r3964_set_termios(struct tty_struct *tty, struct termios * old) +static void r3964_set_termios(struct tty_struct *tty, struct ktermios * old) { TRACE_L("set_termios"); } diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c index 603b9ade5eb0..e96a00fe1389 100644 --- a/drivers/char/n_tty.c +++ b/drivers/char/n_tty.c @@ -994,7 +994,7 @@ int is_ignored(int sig) * when the ldisc is closed. */ -static void n_tty_set_termios(struct tty_struct *tty, struct termios * old) +static void n_tty_set_termios(struct tty_struct *tty, struct ktermios * old) { if (!tty) return; diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c index 74d21c1c104f..5152cedd8878 100644 --- a/drivers/char/pcmcia/synclink_cs.c +++ b/drivers/char/pcmcia/synclink_cs.c @@ -2375,7 +2375,7 @@ static int ioctl_common(MGSLPC_INFO *info, unsigned int cmd, unsigned long arg) * tty pointer to tty structure * termios pointer to buffer to hold returned old termios */ -static void mgslpc_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void mgslpc_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { MGSLPC_INFO *info = (MGSLPC_INFO *)tty->driver_data; unsigned long flags; diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 80d3eedd7f96..c07a1b5cd05d 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -218,7 +218,7 @@ out: return retval; } -static void pty_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void pty_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { tty->termios->c_cflag &= ~(CSIZE | PARENB); tty->termios->c_cflag |= (CS8 | CREAD); @@ -272,6 +272,8 @@ static void __init legacy_pty_init(void) pty_driver->init_termios.c_oflag = 0; pty_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; pty_driver->init_termios.c_lflag = 0; + pty_driver->init_termios.c_ispeed = 38400; + pty_driver->init_termios.c_ospeed = 38400; pty_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW; pty_driver->other = pty_slave_driver; tty_set_operations(pty_driver, &pty_ops); @@ -286,6 +288,8 @@ static void __init legacy_pty_init(void) pty_slave_driver->subtype = PTY_TYPE_SLAVE; pty_slave_driver->init_termios = tty_std_termios; pty_slave_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; + pty_slave_driver->init_termios.c_ispeed = 38400; + pty_slave_driver->init_termios.c_ospeed = 38400; pty_slave_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW; pty_slave_driver->other = pty_driver; @@ -366,6 +370,8 @@ static void __init unix98_pty_init(void) ptm_driver->init_termios.c_oflag = 0; ptm_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; ptm_driver->init_termios.c_lflag = 0; + ptm_driver->init_termios.c_ispeed = 38400; + ptm_driver->init_termios.c_ospeed = 38400; ptm_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM; ptm_driver->other = pts_driver; @@ -381,6 +387,8 @@ static void __init unix98_pty_init(void) pts_driver->subtype = PTY_TYPE_SLAVE; pts_driver->init_termios = tty_std_termios; pts_driver->init_termios.c_cflag = B38400 | CS8 | CREAD; + pts_driver->init_termios.c_ispeed = 38400; + pts_driver->init_termios.c_ospeed = 38400; pts_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM; pts_driver->other = ptm_driver; diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c index 0a77bfcd5b5e..e2a94bfb2a43 100644 --- a/drivers/char/riscom8.c +++ b/drivers/char/riscom8.c @@ -1539,7 +1539,7 @@ static void rc_hangup(struct tty_struct * tty) wake_up_interruptible(&port->open_wait); } -static void rc_set_termios(struct tty_struct * tty, struct termios * old_termios) +static void rc_set_termios(struct tty_struct * tty, struct ktermios * old_termios) { struct riscom_port *port = (struct riscom_port *)tty->driver_data; unsigned long flags; @@ -1614,6 +1614,8 @@ static inline int rc_init_drivers(void) riscom_driver->init_termios = tty_std_termios; riscom_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + riscom_driver->init_termios.c_ispeed = 9600; + riscom_driver->init_termios.c_ospeed = 9600; riscom_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(riscom_driver, &riscom_ops); if ((error = tty_register_driver(riscom_driver))) { diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c index 4fdf52e9f3b1..e94a62e30fc4 100644 --- a/drivers/char/rocket.c +++ b/drivers/char/rocket.c @@ -712,7 +712,7 @@ static void init_r_port(int board, int aiop, int chan, struct pci_dev *pci_dev) * user mode into the driver (exception handler). *info CD manipulation is spinlock protected. */ static void configure_r_port(struct r_port *info, - struct termios *old_termios) + struct ktermios *old_termios) { unsigned cflag; unsigned long flags; @@ -1194,7 +1194,7 @@ static void rp_close(struct tty_struct *tty, struct file *filp) } static void rp_set_termios(struct tty_struct *tty, - struct termios *old_termios) + struct ktermios *old_termios) { struct r_port *info = (struct r_port *) tty->driver_data; CHANNEL_t *cp; @@ -2214,7 +2214,7 @@ static int __init init_PCI(int boards_found) int count = 0; /* Work through the PCI device list, pulling out ours */ - while ((dev = pci_find_device(PCI_VENDOR_ID_RP, PCI_ANY_ID, dev))) { + while ((dev = pci_get_device(PCI_VENDOR_ID_RP, PCI_ANY_ID, dev))) { if (register_PCI(count + boards_found, dev)) count++; } @@ -2436,6 +2436,8 @@ static int __init rp_init(void) rocket_driver->init_termios = tty_std_termios; rocket_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + rocket_driver->init_termios.c_ispeed = 9600; + rocket_driver->init_termios.c_ospeed = 9600; #ifdef ROCKET_SOFT_FLOW rocket_driver->flags |= TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; #endif diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c index 4217d38caef9..75de5f66517a 100644 --- a/drivers/char/ser_a2232.c +++ b/drivers/char/ser_a2232.c @@ -695,6 +695,8 @@ static int a2232_init_drivers(void) a2232_driver->init_termios = tty_std_termios; a2232_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + a2232_driver->init_termios.c_ispeed = 9600; + a2232_driver->init_termios.c_ospeed = 9600; a2232_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(a2232_driver, &a2232_ops); if ((error = tty_register_driver(a2232_driver))) { diff --git a/drivers/char/serial167.c b/drivers/char/serial167.c index 9ba13af234be..af50d32ae2c7 100644 --- a/drivers/char/serial167.c +++ b/drivers/char/serial167.c @@ -1695,7 +1695,7 @@ cy_ioctl(struct tty_struct *tty, struct file * file, static void -cy_set_termios(struct tty_struct *tty, struct termios * old_termios) +cy_set_termios(struct tty_struct *tty, struct ktermios * old_termios) { struct cyclades_port *info = (struct cyclades_port *)tty->driver_data; diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index 99137ab66b62..20946f5127e0 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -2311,7 +2311,7 @@ static void sx_hangup(struct tty_struct * tty) } -static void sx_set_termios(struct tty_struct * tty, struct termios * old_termios) +static void sx_set_termios(struct tty_struct * tty, struct ktermios * old_termios) { struct specialix_port *port = (struct specialix_port *)tty->driver_data; unsigned long flags; @@ -2400,6 +2400,8 @@ static int sx_init_drivers(void) specialix_driver->init_termios = tty_std_termios; specialix_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + specialix_driver->init_termios.c_ispeed = 9600; + specialix_driver->init_termios.c_ospeed = 9600; specialix_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(specialix_driver, &sx_ops); @@ -2475,7 +2477,7 @@ static int __init specialix_init(void) i++; continue; } - pdev = pci_find_device (PCI_VENDOR_ID_SPECIALIX, + pdev = pci_get_device (PCI_VENDOR_ID_SPECIALIX, PCI_DEVICE_ID_SPECIALIX_IO8, pdev); if (!pdev) break; @@ -2491,6 +2493,9 @@ static int __init specialix_init(void) if (!sx_probe(&sx_board[i])) found ++; } + /* May exit pci_get sequence early with lots of boards */ + if (pdev != NULL) + pci_dev_put(pdev); } #endif diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index a547c0c8fb2a..71bfdccfb42e 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -141,9 +141,11 @@ static struct tty_driver *stl_serial; * with this termios initially. Basically all it defines is a raw port * at 9600, 8 data bits, 1 stop bit. */ -static struct termios stl_deftermios = { +static struct ktermios stl_deftermios = { .c_cflag = (B9600 | CS8 | CREAD | HUPCL | CLOCAL), .c_cc = INIT_C_CC, + .c_ispeed = 9600, + .c_ospeed = 9600, }; /* @@ -464,7 +466,7 @@ static int stl_cd1400getreg(struct stlport *portp, int regnr); static int stl_cd1400updatereg(struct stlport *portp, int regnr, int value); static int stl_cd1400panelinit(struct stlbrd *brdp, struct stlpanel *panelp); static void stl_cd1400portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); -static void stl_cd1400setport(struct stlport *portp, struct termios *tiosp); +static void stl_cd1400setport(struct stlport *portp, struct ktermios *tiosp); static int stl_cd1400getsignals(struct stlport *portp); static void stl_cd1400setsignals(struct stlport *portp, int dtr, int rts); static void stl_cd1400ccrwait(struct stlport *portp); @@ -493,7 +495,7 @@ static int stl_sc26198updatereg(struct stlport *portp, int regnr, int value); static int stl_sc26198getglobreg(struct stlport *portp, int regnr); static int stl_sc26198panelinit(struct stlbrd *brdp, struct stlpanel *panelp); static void stl_sc26198portinit(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); -static void stl_sc26198setport(struct stlport *portp, struct termios *tiosp); +static void stl_sc26198setport(struct stlport *portp, struct ktermios *tiosp); static int stl_sc26198getsignals(struct stlport *portp); static void stl_sc26198setsignals(struct stlport *portp, int dtr, int rts); static void stl_sc26198enablerxtx(struct stlport *portp, int rx, int tx); @@ -521,7 +523,7 @@ static void stl_sc26198otherisr(struct stlport *port, unsigned int iack); typedef struct uart { int (*panelinit)(struct stlbrd *brdp, struct stlpanel *panelp); void (*portinit)(struct stlbrd *brdp, struct stlpanel *panelp, struct stlport *portp); - void (*setport)(struct stlport *portp, struct termios *tiosp); + void (*setport)(struct stlport *portp, struct ktermios *tiosp); int (*getsignals)(struct stlport *portp); void (*setsignals)(struct stlport *portp, int dtr, int rts); void (*enablerxtx)(struct stlport *portp, int rx, int tx); @@ -1427,10 +1429,10 @@ static void stl_start(struct tty_struct *tty) /*****************************************************************************/ -static void stl_settermios(struct tty_struct *tty, struct termios *old) +static void stl_settermios(struct tty_struct *tty, struct ktermios *old) { struct stlport *portp; - struct termios *tiosp; + struct ktermios *tiosp; pr_debug("stl_settermios(tty=%p,old=%p)\n", tty, old); @@ -2468,7 +2470,7 @@ static int __init stl_findpcibrds(void) pr_debug("stl_findpcibrds()\n"); for (i = 0; (i < stl_nrpcibrds); i++) - while ((dev = pci_find_device(stl_pcibrds[i].vendid, + while ((dev = pci_get_device(stl_pcibrds[i].vendid, stl_pcibrds[i].devid, dev))) { /* @@ -2947,7 +2949,7 @@ static void stl_cd1400ccrwait(struct stlport *portp) * settings. */ -static void stl_cd1400setport(struct stlport *portp, struct termios *tiosp) +static void stl_cd1400setport(struct stlport *portp, struct ktermios *tiosp) { struct stlbrd *brdp; unsigned long flags; @@ -3924,7 +3926,7 @@ static void stl_sc26198portinit(struct stlbrd *brdp, struct stlpanel *panelp, st * settings. */ -static void stl_sc26198setport(struct stlport *portp, struct termios *tiosp) +static void stl_sc26198setport(struct stlport *portp, struct ktermios *tiosp) { struct stlbrd *brdp; unsigned long flags; diff --git a/drivers/char/sx.c b/drivers/char/sx.c index cc10af08cb05..e9bc147c32fe 100644 --- a/drivers/char/sx.c +++ b/drivers/char/sx.c @@ -2263,6 +2263,8 @@ static int sx_init_drivers(void) sx_driver->init_termios = tty_std_termios; sx_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + sx_driver->init_termios.c_ispeed = 9600; + sx_driver->init_termios.c_ospeed = 9600; sx_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(sx_driver, &sx_ops); diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index 645187b9141e..acc6fab601cc 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -3060,7 +3060,7 @@ static int mgsl_ioctl_common(struct mgsl_struct *info, unsigned int cmd, unsigne * * Return Value: None */ -static void mgsl_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void mgsl_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct mgsl_struct *info = (struct mgsl_struct *)tty->driver_data; unsigned long flags; @@ -4405,6 +4405,8 @@ static int mgsl_init_tty(void) serial_driver->init_termios = tty_std_termios; serial_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + serial_driver->init_termios.c_ispeed = 9600; + serial_driver->init_termios.c_ospeed = 9600; serial_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(serial_driver, &mgsl_ops); if ((rc = tty_register_driver(serial_driver)) < 0) { diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index e4730a7312b5..792c79c315e0 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -151,7 +151,7 @@ static struct tty_driver *serial_driver; static int open(struct tty_struct *tty, struct file * filp); static void close(struct tty_struct *tty, struct file * filp); static void hangup(struct tty_struct *tty); -static void set_termios(struct tty_struct *tty, struct termios *old_termios); +static void set_termios(struct tty_struct *tty, struct ktermios *old_termios); static int write(struct tty_struct *tty, const unsigned char *buf, int count); static void put_char(struct tty_struct *tty, unsigned char ch); @@ -816,7 +816,7 @@ static void hangup(struct tty_struct *tty) wake_up_interruptible(&info->open_wait); } -static void set_termios(struct tty_struct *tty, struct termios *old_termios) +static void set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct slgt_info *info = tty->driver_data; unsigned long flags; @@ -3546,6 +3546,8 @@ static int __init slgt_init(void) serial_driver->init_termios = tty_std_termios; serial_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + serial_driver->init_termios.c_ispeed = 9600; + serial_driver->init_termios.c_ospeed = 9600; serial_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(serial_driver, &ops); if ((rc = tty_register_driver(serial_driver)) < 0) { diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index 20a96ef250be..53e8ccf94fe3 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -519,7 +519,7 @@ static struct tty_driver *serial_driver; static int open(struct tty_struct *tty, struct file * filp); static void close(struct tty_struct *tty, struct file * filp); static void hangup(struct tty_struct *tty); -static void set_termios(struct tty_struct *tty, struct termios *old_termios); +static void set_termios(struct tty_struct *tty, struct ktermios *old_termios); static int write(struct tty_struct *tty, const unsigned char *buf, int count); static void put_char(struct tty_struct *tty, unsigned char ch); @@ -918,7 +918,7 @@ static void hangup(struct tty_struct *tty) /* Set new termios settings */ -static void set_termios(struct tty_struct *tty, struct termios *old_termios) +static void set_termios(struct tty_struct *tty, struct ktermios *old_termios) { SLMP_INFO *info = (SLMP_INFO *)tty->driver_data; unsigned long flags; @@ -4034,6 +4034,8 @@ static int __init synclinkmp_init(void) serial_driver->init_termios = tty_std_termios; serial_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + serial_driver->init_termios.c_ispeed = 9600; + serial_driver->init_termios.c_ospeed = 9600; serial_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(serial_driver, &ops); if ((rc = tty_register_driver(serial_driver)) < 0) { diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c index 0ffba4e911ca..30486df9fd3f 100644 --- a/drivers/char/tty_ioctl.c +++ b/drivers/char/tty_ioctl.c @@ -495,8 +495,8 @@ static int get_sgttyb(struct tty_struct * tty, struct sgttyb __user * sgttyb) struct sgttyb tmp; mutex_lock(&tty->termios_mutex); - tmp.sg_ispeed = tty->c_ispeed; - tmp.sg_ospeed = tty->c_ospeed; + tmp.sg_ispeed = tty->termios->c_ispeed; + tmp.sg_ospeed = tty->termios->c_ospeed; tmp.sg_erase = tty->termios->c_cc[VERASE]; tmp.sg_kill = tty->termios->c_cc[VKILL]; tmp.sg_flags = get_sgflags(tty); diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c index d0b94dd1af6d..e01317cb1a0e 100644 --- a/drivers/char/vme_scc.c +++ b/drivers/char/vme_scc.c @@ -153,6 +153,8 @@ static int scc_init_drivers(void) scc_driver->init_termios = tty_std_termios; scc_driver->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + scc_driver->init_termios.c_ispeed = 9600; + scc_driver->init_termios.c_ospeed = 9600; scc_driver->flags = TTY_DRIVER_REAL_RAW; tty_set_operations(scc_driver, &scc_ops); diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c index 787a630a8e56..0475a54df83a 100644 --- a/drivers/isdn/capi/capi.c +++ b/drivers/isdn/capi/capi.c @@ -1203,7 +1203,7 @@ static int capinc_tty_ioctl(struct tty_struct *tty, struct file * file, return error; } -static void capinc_tty_set_termios(struct tty_struct *tty, struct termios * old) +static void capinc_tty_set_termios(struct tty_struct *tty, struct ktermios * old) { #ifdef _DEBUG_TTYFUNCS printk(KERN_DEBUG "capinc_tty_set_termios\n"); diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index 7edea015867e..458b6462f937 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -127,7 +127,7 @@ static int if_write_room(struct tty_struct *tty); static int if_chars_in_buffer(struct tty_struct *tty); static void if_throttle(struct tty_struct *tty); static void if_unthrottle(struct tty_struct *tty); -static void if_set_termios(struct tty_struct *tty, struct termios *old); +static void if_set_termios(struct tty_struct *tty, struct ktermios *old); static int if_tiocmget(struct tty_struct *tty, struct file *file); static int if_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear); @@ -490,7 +490,7 @@ static void if_unthrottle(struct tty_struct *tty) mutex_unlock(&cs->mutex); } -static void if_set_termios(struct tty_struct *tty, struct termios *old) +static void if_set_termios(struct tty_struct *tty, struct ktermios *old) { struct cardstate *cs; unsigned int iflag; diff --git a/drivers/isdn/i4l/isdn_tty.c b/drivers/isdn/i4l/isdn_tty.c index 2b91bb07fc7f..fc80afe555b9 100644 --- a/drivers/isdn/i4l/isdn_tty.c +++ b/drivers/isdn/i4l/isdn_tty.c @@ -1464,7 +1464,7 @@ isdn_tty_ioctl(struct tty_struct *tty, struct file *file, } static void -isdn_tty_set_termios(struct tty_struct *tty, struct termios *old_termios) +isdn_tty_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { modem_info *info = (modem_info *) tty->driver_data; diff --git a/drivers/net/irda/irtty-sir.c b/drivers/net/irda/irtty-sir.c index 6a98b7ae4975..ad1857364d51 100644 --- a/drivers/net/irda/irtty-sir.c +++ b/drivers/net/irda/irtty-sir.c @@ -117,7 +117,7 @@ static int irtty_change_speed(struct sir_dev *dev, unsigned speed) { struct sirtty_cb *priv = dev->priv; struct tty_struct *tty; - struct termios old_termios; + struct ktermios old_termios; int cflag; IRDA_ASSERT(priv != NULL, return -1;); @@ -318,7 +318,7 @@ static void irtty_write_wakeup(struct tty_struct *tty) static inline void irtty_stop_receiver(struct tty_struct *tty, int stop) { - struct termios old_termios; + struct ktermios old_termios; int cflag; lock_kernel(); diff --git a/drivers/s390/char/sclp_tty.c b/drivers/s390/char/sclp_tty.c index 6f43e04dbefd..2d173e5c8a09 100644 --- a/drivers/s390/char/sclp_tty.c +++ b/drivers/s390/char/sclp_tty.c @@ -60,8 +60,6 @@ static unsigned short int sclp_tty_chars_count; struct tty_driver *sclp_tty_driver; -extern struct termios tty_std_termios; - static struct sclp_ioctls sclp_ioctls; static struct sclp_ioctls sclp_ioctls_init = { diff --git a/drivers/s390/char/tty3270.c b/drivers/s390/char/tty3270.c index 4717c3611601..09844621edc0 100644 --- a/drivers/s390/char/tty3270.c +++ b/drivers/s390/char/tty3270.c @@ -1659,7 +1659,7 @@ tty3270_flush_buffer(struct tty_struct *tty) * Check for visible/invisible input switches */ static void -tty3270_set_termios(struct tty_struct *tty, struct termios *old) +tty3270_set_termios(struct tty_struct *tty, struct ktermios *old) { struct tty3270 *tp; int new; diff --git a/drivers/serial/21285.c b/drivers/serial/21285.c index 6a1a568ca649..facb67855619 100644 --- a/drivers/serial/21285.c +++ b/drivers/serial/21285.c @@ -214,8 +214,8 @@ static void serial21285_shutdown(struct uart_port *port) } static void -serial21285_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +serial21285_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned long flags; unsigned int baud, quot, h_lcr; diff --git a/drivers/serial/68328serial.c b/drivers/serial/68328serial.c index 9b8b585513ec..cad426c9711e 100644 --- a/drivers/serial/68328serial.c +++ b/drivers/serial/68328serial.c @@ -1061,7 +1061,7 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file, return 0; } -static void rs_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct m68k_serial *info = (struct m68k_serial *)tty->driver_data; diff --git a/drivers/serial/68360serial.c b/drivers/serial/68360serial.c index 634ecca36a77..68817a7d8c0d 100644 --- a/drivers/serial/68360serial.c +++ b/drivers/serial/68360serial.c @@ -1523,7 +1523,7 @@ static int rs_360_ioctl(struct tty_struct *tty, struct file * file, /* FIX UP modem control here someday...... */ -static void rs_360_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void rs_360_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { ser_info_t *info = (ser_info_t *)tty->driver_data; diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index e34bd03cfce7..51f3c739f7e1 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -1763,8 +1763,8 @@ static unsigned int serial8250_get_divisor(struct uart_port *port, unsigned int } static void -serial8250_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +serial8250_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_8250_port *up = (struct uart_8250_port *)port; unsigned char cval, fcr = 0; diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index 4d3626ef4643..61db6973755a 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -345,8 +345,8 @@ static void pl010_shutdown(struct uart_port *port) } static void -pl010_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +pl010_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int lcr_h, old_cr; unsigned long flags; diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index d503625730df..9a3b374b2a08 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -412,8 +412,8 @@ static void pl011_shutdown(struct uart_port *port) } static void -pl011_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +pl011_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int lcr_h, old_cr; unsigned long flags; diff --git a/drivers/serial/atmel_serial.c b/drivers/serial/atmel_serial.c index 9217ee6c7865..ed7f7209ea59 100644 --- a/drivers/serial/atmel_serial.c +++ b/drivers/serial/atmel_serial.c @@ -478,7 +478,7 @@ static void atmel_serial_pm(struct uart_port *port, unsigned int state, unsigned /* * Change the port parameters */ -static void atmel_set_termios(struct uart_port *port, struct termios * termios, struct termios * old) +static void atmel_set_termios(struct uart_port *port, struct ktermios * termios, struct ktermios * old) { unsigned long flags; unsigned int mode, imr, quot, baud; diff --git a/drivers/serial/clps711x.c b/drivers/serial/clps711x.c index 598012714882..23827189ec0e 100644 --- a/drivers/serial/clps711x.c +++ b/drivers/serial/clps711x.c @@ -286,8 +286,8 @@ static void clps711xuart_shutdown(struct uart_port *port) } static void -clps711xuart_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +clps711xuart_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int ubrlcr, baud, quot; unsigned long flags; diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c index 7a24e53546c7..42b050c46abe 100644 --- a/drivers/serial/crisv10.c +++ b/drivers/serial/crisv10.c @@ -804,8 +804,8 @@ static struct e100_serial rs_table[] = { #define NR_PORTS (sizeof(rs_table)/sizeof(struct e100_serial)) -static struct termios *serial_termios[NR_PORTS]; -static struct termios *serial_termios_locked[NR_PORTS]; +static struct ktermios *serial_termios[NR_PORTS]; +static struct ktermios *serial_termios_locked[NR_PORTS]; #ifdef CONFIG_ETRAX_SERIAL_FAST_TIMER static struct fast_timer fast_timers[NR_PORTS]; #endif @@ -4223,7 +4223,7 @@ rs_ioctl(struct tty_struct *tty, struct file * file, } static void -rs_set_termios(struct tty_struct *tty, struct termios *old_termios) +rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct e100_serial *info = (struct e100_serial *)tty->driver_data; @@ -4877,6 +4877,8 @@ rs_init(void) driver->init_termios = tty_std_termios; driver->init_termios.c_cflag = B115200 | CS8 | CREAD | HUPCL | CLOCAL; /* is normally B9600 default... */ + driver->init_termios.c_ispeed = 115200; + driver->init_termios.c_ospeed = 115200; driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; driver->termios = serial_termios; driver->termios_locked = serial_termios_locked; diff --git a/drivers/serial/crisv10.h b/drivers/serial/crisv10.h index f30b93d6ef79..4a23340663aa 100644 --- a/drivers/serial/crisv10.h +++ b/drivers/serial/crisv10.h @@ -93,8 +93,8 @@ struct e100_serial { struct work_struct work; struct async_icount icount; /* error-statistics etc.*/ - struct termios normal_termios; - struct termios callout_termios; + struct ktermios normal_termios; + struct ktermios callout_termios; #ifdef DECLARE_WAITQUEUE wait_queue_head_t open_wait; wait_queue_head_t close_wait; diff --git a/drivers/serial/dz.c b/drivers/serial/dz.c index af1544f3356f..587d87b9eb3c 100644 --- a/drivers/serial/dz.c +++ b/drivers/serial/dz.c @@ -461,8 +461,8 @@ static void dz_break_ctl(struct uart_port *uport, int break_state) spin_unlock_irqrestore(&uport->lock, flags); } -static void dz_set_termios(struct uart_port *uport, struct termios *termios, - struct termios *old_termios) +static void dz_set_termios(struct uart_port *uport, struct ktermios *termios, + struct ktermios *old_termios) { struct dz_port *dport = (struct dz_port *)uport; unsigned long flags; diff --git a/drivers/serial/icom.c b/drivers/serial/icom.c index 8aa0f641866b..7d623003e65e 100644 --- a/drivers/serial/icom.c +++ b/drivers/serial/icom.c @@ -1087,8 +1087,8 @@ static void icom_close(struct uart_port *port) } static void icom_set_termios(struct uart_port *port, - struct termios *termios, - struct termios *old_termios) + struct ktermios *termios, + struct ktermios *old_termios) { int baud; unsigned cflag, iflag; diff --git a/drivers/serial/imx.c b/drivers/serial/imx.c index ee5c782597dd..e216dcf29376 100644 --- a/drivers/serial/imx.c +++ b/drivers/serial/imx.c @@ -459,8 +459,8 @@ static void imx_shutdown(struct uart_port *port) } static void -imx_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +imx_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct imx_port *sport = (struct imx_port *)port; unsigned long flags; diff --git a/drivers/serial/ioc3_serial.c b/drivers/serial/ioc3_serial.c index 2308d26c8629..9cc0be932316 100644 --- a/drivers/serial/ioc3_serial.c +++ b/drivers/serial/ioc3_serial.c @@ -950,7 +950,7 @@ static void transmit_chars(struct uart_port *the_port) */ static void ioc3_change_speed(struct uart_port *the_port, - struct termios *new_termios, struct termios *old_termios) + struct ktermios *new_termios, struct ktermios *old_termios) { struct ioc3_port *port = get_ioc3_port(the_port); unsigned int cflag; @@ -1853,7 +1853,7 @@ static int ic3_startup(struct uart_port *the_port) */ static void ic3_set_termios(struct uart_port *the_port, - struct termios *termios, struct termios *old_termios) + struct ktermios *termios, struct ktermios *old_termios) { unsigned long port_flags; diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index 711bd1511439..c862f67c985a 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -1681,7 +1681,7 @@ static void transmit_chars(struct uart_port *the_port) */ static void ioc4_change_speed(struct uart_port *the_port, - struct termios *new_termios, struct termios *old_termios) + struct ktermios *new_termios, struct ktermios *old_termios) { struct ioc4_port *port = get_ioc4_port(the_port, 0); int baud, bits; @@ -1802,7 +1802,7 @@ static inline int ic4_startup_local(struct uart_port *the_port) ioc4_set_proto(port, the_port->mapbase); /* set the speed of the serial port */ - ioc4_change_speed(the_port, info->tty->termios, (struct termios *)0); + ioc4_change_speed(the_port, info->tty->termios, (struct ktermios *)0); return 0; } @@ -2570,7 +2570,7 @@ static int ic4_startup(struct uart_port *the_port) */ static void ic4_set_termios(struct uart_port *the_port, - struct termios *termios, struct termios *old_termios) + struct ktermios *termios, struct ktermios *old_termios) { unsigned long port_flags; diff --git a/drivers/serial/ip22zilog.c b/drivers/serial/ip22zilog.c index dca6c1bde8f9..0746c9446ae0 100644 --- a/drivers/serial/ip22zilog.c +++ b/drivers/serial/ip22zilog.c @@ -840,8 +840,8 @@ ip22zilog_convert_to_zs(struct uart_ip22zilog_port *up, unsigned int cflag, /* The port lock is not held. */ static void -ip22zilog_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +ip22zilog_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_ip22zilog_port *up = (struct uart_ip22zilog_port *) port; unsigned long flags; diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c index f8262e6ad8d3..7cf1c60027f8 100644 --- a/drivers/serial/jsm/jsm_tty.c +++ b/drivers/serial/jsm/jsm_tty.c @@ -142,7 +142,7 @@ static void jsm_tty_send_xchar(struct uart_port *port, char ch) { unsigned long lock_flags; struct jsm_channel *channel = (struct jsm_channel *)port; - struct termios *termios; + struct ktermios *termios; spin_lock_irqsave(&port->lock, lock_flags); termios = port->info->tty->termios; @@ -180,7 +180,7 @@ static int jsm_tty_open(struct uart_port *port) struct jsm_board *brd; int rc = 0; struct jsm_channel *channel = (struct jsm_channel *)port; - struct termios *termios; + struct ktermios *termios; /* Get board pointer from our array of majors we have allocated */ brd = channel->ch_bd; @@ -269,7 +269,7 @@ static int jsm_tty_open(struct uart_port *port) static void jsm_tty_close(struct uart_port *port) { struct jsm_board *bd; - struct termios *ts; + struct ktermios *ts; struct jsm_channel *channel = (struct jsm_channel *)port; jsm_printk(CLOSE, INFO, &channel->ch_bd->pci_dev, "start\n"); @@ -302,8 +302,8 @@ static void jsm_tty_close(struct uart_port *port) } static void jsm_tty_set_termios(struct uart_port *port, - struct termios *termios, - struct termios *old_termios) + struct ktermios *termios, + struct ktermios *old_termios) { unsigned long lock_flags; struct jsm_channel *channel = (struct jsm_channel *)port; diff --git a/drivers/serial/m32r_sio.c b/drivers/serial/m32r_sio.c index 7656a35f5e2f..6e09c8b395e8 100644 --- a/drivers/serial/m32r_sio.c +++ b/drivers/serial/m32r_sio.c @@ -699,7 +699,7 @@ static unsigned int m32r_sio_get_divisor(struct uart_port *port, } static void m32r_sio_set_termios(struct uart_port *port, - struct termios *termios, struct termios *old) + struct ktermios *termios, struct ktermios *old) { struct uart_sio_port *up = (struct uart_sio_port *)port; unsigned char cval = 0; diff --git a/drivers/serial/mcfserial.c b/drivers/serial/mcfserial.c index 3db206d29b33..08430961a895 100644 --- a/drivers/serial/mcfserial.c +++ b/drivers/serial/mcfserial.c @@ -1132,7 +1132,7 @@ static int mcfrs_ioctl(struct tty_struct *tty, struct file * file, return 0; } -static void mcfrs_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void mcfrs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct mcf_serial *info = (struct mcf_serial *)tty->driver_data; diff --git a/drivers/serial/mpc52xx_uart.c b/drivers/serial/mpc52xx_uart.c index 6dd579ed9777..9d11a75663e6 100644 --- a/drivers/serial/mpc52xx_uart.c +++ b/drivers/serial/mpc52xx_uart.c @@ -270,8 +270,8 @@ mpc52xx_uart_shutdown(struct uart_port *port) } static void -mpc52xx_uart_set_termios(struct uart_port *port, struct termios *new, - struct termios *old) +mpc52xx_uart_set_termios(struct uart_port *port, struct ktermios *new, + struct ktermios *old) { struct mpc52xx_psc __iomem *psc = PSC(port); unsigned long flags; diff --git a/drivers/serial/mpsc.c b/drivers/serial/mpsc.c index 29823bd60fb0..3d2fcc57b1ce 100644 --- a/drivers/serial/mpsc.c +++ b/drivers/serial/mpsc.c @@ -1440,8 +1440,8 @@ mpsc_shutdown(struct uart_port *port) } static void -mpsc_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +mpsc_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct mpsc_port_info *pi = (struct mpsc_port_info *)port; u32 baud; diff --git a/drivers/serial/mux.c b/drivers/serial/mux.c index 8ad1b8c5ec5d..ccb8fa1800a5 100644 --- a/drivers/serial/mux.c +++ b/drivers/serial/mux.c @@ -273,8 +273,8 @@ static void mux_shutdown(struct uart_port *port) * The Serial Mux does not support this function. */ static void -mux_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +mux_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { } diff --git a/drivers/serial/netx-serial.c b/drivers/serial/netx-serial.c index 062bad457b1a..b56f7db45031 100644 --- a/drivers/serial/netx-serial.c +++ b/drivers/serial/netx-serial.c @@ -337,8 +337,8 @@ static void netx_shutdown(struct uart_port *port) } static void -netx_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +netx_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int baud, quot; unsigned char old_cr; diff --git a/drivers/serial/pmac_zilog.c b/drivers/serial/pmac_zilog.c index bf9809ed9c0b..752ef07516b9 100644 --- a/drivers/serial/pmac_zilog.c +++ b/drivers/serial/pmac_zilog.c @@ -1262,8 +1262,8 @@ static void pmz_irda_setup(struct uart_pmac_port *uap, unsigned long *baud) } -static void __pmz_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void __pmz_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_pmac_port *uap = to_pmz(port); unsigned long baud; @@ -1273,7 +1273,7 @@ static void __pmz_set_termios(struct uart_port *port, struct termios *termios, if (ZS_IS_ASLEEP(uap)) return; - memcpy(&uap->termios_cache, termios, sizeof(struct termios)); + memcpy(&uap->termios_cache, termios, sizeof(struct ktermios)); /* XXX Check which revs of machines actually allow 1 and 4Mb speeds * on the IR dongle. Note that the IRTTY driver currently doesn't know @@ -1313,8 +1313,8 @@ static void __pmz_set_termios(struct uart_port *port, struct termios *termios, } /* The port lock is not held. */ -static void pmz_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void pmz_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_pmac_port *uap = to_pmz(port); unsigned long flags; diff --git a/drivers/serial/pmac_zilog.h b/drivers/serial/pmac_zilog.h index c03f9bfacdd8..570b0d925e83 100644 --- a/drivers/serial/pmac_zilog.h +++ b/drivers/serial/pmac_zilog.h @@ -60,7 +60,7 @@ struct uart_pmac_port { volatile struct dbdma_regs __iomem *tx_dma_regs; volatile struct dbdma_regs __iomem *rx_dma_regs; - struct termios termios_cache; + struct ktermios termios_cache; }; #define to_pmz(p) ((struct uart_pmac_port *)(p)) diff --git a/drivers/serial/pxa.c b/drivers/serial/pxa.c index 415fe9633a9b..d403aaa55092 100644 --- a/drivers/serial/pxa.c +++ b/drivers/serial/pxa.c @@ -433,8 +433,8 @@ static void serial_pxa_shutdown(struct uart_port *port) } static void -serial_pxa_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +serial_pxa_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_pxa_port *up = (struct uart_pxa_port *)port; unsigned char cval, fcr = 0; diff --git a/drivers/serial/s3c2410.c b/drivers/serial/s3c2410.c index 8dfc2dd058ca..3ba9208ebd0c 100644 --- a/drivers/serial/s3c2410.c +++ b/drivers/serial/s3c2410.c @@ -738,8 +738,8 @@ static unsigned int s3c24xx_serial_getclk(struct uart_port *port, } static void s3c24xx_serial_set_termios(struct uart_port *port, - struct termios *termios, - struct termios *old) + struct ktermios *termios, + struct ktermios *old) { struct s3c2410_uartcfg *cfg = s3c24xx_port_to_cfg(port); struct s3c24xx_uart_port *ourport = to_ourport(port); diff --git a/drivers/serial/sa1100.c b/drivers/serial/sa1100.c index d4065266b6fc..58a83c27e14b 100644 --- a/drivers/serial/sa1100.c +++ b/drivers/serial/sa1100.c @@ -408,8 +408,8 @@ static void sa1100_shutdown(struct uart_port *port) } static void -sa1100_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +sa1100_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct sa1100_port *sport = (struct sa1100_port *)port; unsigned long flags; diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index c67b05e9a451..f84982e508c7 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -65,7 +65,7 @@ static struct lock_class_key port_lock_key; #define uart_console(port) (0) #endif -static void uart_change_speed(struct uart_state *state, struct termios *old_termios); +static void uart_change_speed(struct uart_state *state, struct ktermios *old_termios); static void uart_wait_until_sent(struct tty_struct *tty, int timeout); static void uart_change_pm(struct uart_state *state, int pm_state); @@ -338,8 +338,8 @@ EXPORT_SYMBOL(uart_update_timeout); * we're actually going to be using. */ unsigned int -uart_get_baud_rate(struct uart_port *port, struct termios *termios, - struct termios *old, unsigned int min, unsigned int max) +uart_get_baud_rate(struct uart_port *port, struct ktermios *termios, + struct ktermios *old, unsigned int min, unsigned int max) { unsigned int try, baud, altbaud = 38400; upf_t flags = port->flags & UPF_SPD_MASK; @@ -421,11 +421,11 @@ uart_get_divisor(struct uart_port *port, unsigned int baud) EXPORT_SYMBOL(uart_get_divisor); static void -uart_change_speed(struct uart_state *state, struct termios *old_termios) +uart_change_speed(struct uart_state *state, struct ktermios *old_termios) { struct tty_struct *tty = state->info->tty; struct uart_port *port = state->port; - struct termios *termios; + struct ktermios *termios; /* * If we have no tty, termios, or the port does not exist, @@ -1139,7 +1139,7 @@ uart_ioctl(struct tty_struct *tty, struct file *filp, unsigned int cmd, return ret; } -static void uart_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void uart_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct uart_state *state = tty->driver_data; unsigned long flags; @@ -1866,7 +1866,7 @@ int __init uart_set_options(struct uart_port *port, struct console *co, int baud, int parity, int bits, int flow) { - struct termios termios; + struct ktermios termios; int i; /* @@ -1876,7 +1876,7 @@ uart_set_options(struct uart_port *port, struct console *co, spin_lock_init(&port->lock); lockdep_set_class(&port->lock, &port_lock_key); - memset(&termios, 0, sizeof(struct termios)); + memset(&termios, 0, sizeof(struct ktermios)); termios.c_cflag = CREAD | HUPCL | CLOCAL; @@ -1991,12 +1991,12 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port) * Re-enable the console device after suspending. */ if (uart_console(port)) { - struct termios termios; + struct ktermios termios; /* * First try to use the console cflag setting. */ - memset(&termios, 0, sizeof(struct termios)); + memset(&termios, 0, sizeof(struct ktermios)); termios.c_cflag = port->cons->cflag; /* @@ -2189,6 +2189,7 @@ int uart_register_driver(struct uart_driver *drv) normal->subtype = SERIAL_TYPE_NORMAL; normal->init_termios = tty_std_termios; normal->init_termios.c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL; + normal->init_termios.c_ispeed = normal->init_termios.c_ospeed = 9600; normal->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; normal->driver_state = drv; tty_set_operations(normal, &uart_ops); diff --git a/drivers/serial/serial_lh7a40x.c b/drivers/serial/serial_lh7a40x.c index 5e1ac356bbb0..eb18d429752d 100644 --- a/drivers/serial/serial_lh7a40x.c +++ b/drivers/serial/serial_lh7a40x.c @@ -348,8 +348,8 @@ static void lh7a40xuart_shutdown (struct uart_port* port) } static void lh7a40xuart_set_termios (struct uart_port* port, - struct termios* termios, - struct termios* old) + struct ktermios* termios, + struct ktermios* old) { unsigned int con; unsigned int inten; diff --git a/drivers/serial/serial_txx9.c b/drivers/serial/serial_txx9.c index 2a48289ac722..7186a82c4759 100644 --- a/drivers/serial/serial_txx9.c +++ b/drivers/serial/serial_txx9.c @@ -556,8 +556,8 @@ static void serial_txx9_shutdown(struct uart_port *port) } static void -serial_txx9_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +serial_txx9_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_txx9_port *up = (struct uart_txx9_port *)port; unsigned int cval, fcr = 0; diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c index 3b5f19ec2126..9031b57f12dd 100644 --- a/drivers/serial/sh-sci.c +++ b/drivers/serial/sh-sci.c @@ -943,8 +943,8 @@ static void sci_shutdown(struct uart_port *port) s->disable(port); } -static void sci_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void sci_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct sci_port *s = &sci_ports[port->line]; unsigned int status, baud, smr_val; diff --git a/drivers/serial/sn_console.c b/drivers/serial/sn_console.c index 956b2cf08e1e..253ceb895ca7 100644 --- a/drivers/serial/sn_console.c +++ b/drivers/serial/sn_console.c @@ -361,8 +361,8 @@ static int snp_startup(struct uart_port *port) * */ static void -snp_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +snp_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { } diff --git a/drivers/serial/sunhv.c b/drivers/serial/sunhv.c index 03941d27d15d..40d48566215c 100644 --- a/drivers/serial/sunhv.c +++ b/drivers/serial/sunhv.c @@ -281,8 +281,8 @@ static void sunhv_shutdown(struct uart_port *port) } /* port->lock is not held. */ -static void sunhv_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void sunhv_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int baud = uart_get_baud_rate(port, termios, old, 0, 4000000); unsigned int quot = uart_get_divisor(port, baud); diff --git a/drivers/serial/sunsab.c b/drivers/serial/sunsab.c index 08a7cd6a3a0c..493d5bbb661b 100644 --- a/drivers/serial/sunsab.c +++ b/drivers/serial/sunsab.c @@ -786,8 +786,8 @@ static void sunsab_convert_to_sab(struct uart_sunsab_port *up, unsigned int cfla } /* port->lock is not held. */ -static void sunsab_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void sunsab_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_sunsab_port *up = (struct uart_sunsab_port *) port; unsigned long flags; diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c index c577faea60e8..564592b2b2ba 100644 --- a/drivers/serial/sunsu.c +++ b/drivers/serial/sunsu.c @@ -893,8 +893,8 @@ sunsu_change_speed(struct uart_port *port, unsigned int cflag, } static void -sunsu_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +sunsu_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned int baud, quot; diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c index b2cc703b2b9e..75de919a9471 100644 --- a/drivers/serial/sunzilog.c +++ b/drivers/serial/sunzilog.c @@ -922,8 +922,8 @@ sunzilog_convert_to_zs(struct uart_sunzilog_port *up, unsigned int cflag, /* The port lock is not held. */ static void -sunzilog_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +sunzilog_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { struct uart_sunzilog_port *up = (struct uart_sunzilog_port *) port; unsigned long flags; diff --git a/drivers/serial/uartlite.c b/drivers/serial/uartlite.c index 83690653b78b..92eba893559d 100644 --- a/drivers/serial/uartlite.c +++ b/drivers/serial/uartlite.c @@ -214,8 +214,8 @@ static void ulite_shutdown(struct uart_port *port) free_irq(port->irq, port); } -static void ulite_set_termios(struct uart_port *port, struct termios *termios, - struct termios *old) +static void ulite_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned long flags; unsigned int baud; diff --git a/drivers/serial/v850e_uart.c b/drivers/serial/v850e_uart.c index 28f3bbff87bf..dd98aca6ed08 100644 --- a/drivers/serial/v850e_uart.c +++ b/drivers/serial/v850e_uart.c @@ -404,8 +404,8 @@ static void v850e_uart_shutdown (struct uart_port *port) } static void -v850e_uart_set_termios (struct uart_port *port, struct termios *termios, - struct termios *old) +v850e_uart_set_termios (struct uart_port *port, struct ktermios *termios, + struct ktermios *old) { unsigned cflags = termios->c_cflag; diff --git a/drivers/serial/vr41xx_siu.c b/drivers/serial/vr41xx_siu.c index fd51f8182dec..cf0e663b42ed 100644 --- a/drivers/serial/vr41xx_siu.c +++ b/drivers/serial/vr41xx_siu.c @@ -562,8 +562,8 @@ static void siu_shutdown(struct uart_port *port) free_irq(port->irq, port); } -static void siu_set_termios(struct uart_port *port, struct termios *new, - struct termios *old) +static void siu_set_termios(struct uart_port *port, struct ktermios *new, + struct ktermios *old) { tcflag_t c_cflag, c_iflag; uint8_t lcr, fcr, ier; diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c index 792becdfe6f8..fc3197273663 100644 --- a/drivers/tc/zs.c +++ b/drivers/tc/zs.c @@ -1238,7 +1238,7 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file, return 0; } -static void rs_set_termios(struct tty_struct *tty, struct termios *old_termios) +static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios) { struct dec_serial *info = (struct dec_serial *)tty->driver_data; int was_stopped; diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 7f1fa956dcdb..98199628e394 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -677,10 +677,10 @@ static const __u8 acm_tty_size[] = { 5, 6, 7, 8 }; -static void acm_tty_set_termios(struct tty_struct *tty, struct termios *termios_old) +static void acm_tty_set_termios(struct tty_struct *tty, struct ktermios *termios_old) { struct acm *acm = tty->driver_data; - struct termios *termios = tty->termios; + struct ktermios *termios = tty->termios; struct usb_cdc_line_coding newline; int newctrl = acm->ctrlout; diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c index 208e55a667ac..5516c59ed5ec 100644 --- a/drivers/usb/gadget/serial.c +++ b/drivers/usb/gadget/serial.c @@ -200,7 +200,7 @@ static void gs_unthrottle(struct tty_struct * tty); static void gs_break(struct tty_struct *tty, int break_state); static int gs_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg); -static void gs_set_termios(struct tty_struct *tty, struct termios *old); +static void gs_set_termios(struct tty_struct *tty, struct ktermios *old); static int gs_send(struct gs_dev *dev); static int gs_send_packet(struct gs_dev *dev, char *packet, @@ -1077,7 +1077,7 @@ static int gs_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, /* * gs_set_termios */ -static void gs_set_termios(struct tty_struct *tty, struct termios *old) +static void gs_set_termios(struct tty_struct *tty, struct ktermios *old) { } diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c index 863966c1c5ac..5261cd22ee6b 100644 --- a/drivers/usb/serial/ark3116.c +++ b/drivers/usb/serial/ark3116.c @@ -156,7 +156,7 @@ cleanup: } static void ark3116_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct usb_serial *serial = port->serial; struct ark3116_private *priv = usb_get_serial_port_data(port); @@ -326,7 +326,7 @@ static void ark3116_set_termios(struct usb_serial_port *port, static int ark3116_open(struct usb_serial_port *port, struct file *filp) { - struct termios tmp_termios; + struct ktermios tmp_termios; struct usb_serial *serial = port->serial; char *buf; int result = 0; diff --git a/drivers/usb/serial/belkin_sa.c b/drivers/usb/serial/belkin_sa.c index 8835bb58ca9b..38b4dae319ee 100644 --- a/drivers/usb/serial/belkin_sa.c +++ b/drivers/usb/serial/belkin_sa.c @@ -92,7 +92,7 @@ static void belkin_sa_shutdown (struct usb_serial *serial); static int belkin_sa_open (struct usb_serial_port *port, struct file *filp); static void belkin_sa_close (struct usb_serial_port *port, struct file *filp); static void belkin_sa_read_int_callback (struct urb *urb); -static void belkin_sa_set_termios (struct usb_serial_port *port, struct termios * old); +static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios * old); static int belkin_sa_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); static void belkin_sa_break_ctl (struct usb_serial_port *port, int break_state ); static int belkin_sa_tiocmget (struct usb_serial_port *port, struct file *file); @@ -333,7 +333,7 @@ exit: __FUNCTION__, retval); } -static void belkin_sa_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void belkin_sa_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { struct usb_serial *serial = port->serial; struct belkin_sa_private *priv = usb_get_serial_port_data(port); diff --git a/drivers/usb/serial/console.c b/drivers/usb/serial/console.c index 7167728d764c..9386e216d681 100644 --- a/drivers/usb/serial/console.c +++ b/drivers/usb/serial/console.c @@ -65,7 +65,7 @@ static int usb_console_setup(struct console *co, char *options) struct usb_serial_port *port; int retval = 0; struct tty_struct *tty; - struct termios *termios; + struct ktermios *termios; dbg ("%s", __FUNCTION__); diff --git a/drivers/usb/serial/cp2101.c b/drivers/usb/serial/cp2101.c index f95d42c0d16a..2f9b7ac32663 100644 --- a/drivers/usb/serial/cp2101.c +++ b/drivers/usb/serial/cp2101.c @@ -41,7 +41,7 @@ static int cp2101_open(struct usb_serial_port*, struct file*); static void cp2101_cleanup(struct usb_serial_port*); static void cp2101_close(struct usb_serial_port*, struct file*); static void cp2101_get_termios(struct usb_serial_port*); -static void cp2101_set_termios(struct usb_serial_port*, struct termios*); +static void cp2101_set_termios(struct usb_serial_port*, struct ktermios*); static int cp2101_tiocmget (struct usb_serial_port *, struct file *); static int cp2101_tiocmset (struct usb_serial_port *, struct file *, unsigned int, unsigned int); @@ -506,7 +506,7 @@ static void cp2101_get_termios (struct usb_serial_port *port) } static void cp2101_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { unsigned int cflag, old_cflag=0; int baud=0, bits; diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c index 093f303b3189..a1fdb85b8c0a 100644 --- a/drivers/usb/serial/cypress_m8.c +++ b/drivers/usb/serial/cypress_m8.c @@ -143,7 +143,7 @@ struct cypress_private { wait_queue_head_t delta_msr_wait; /* used for TIOCMIWAIT */ char prev_status, diff_status; /* used for TIOCMIWAIT */ /* we pass a pointer to this as the arguement sent to cypress_set_termios old_termios */ - struct termios tmp_termios; /* stores the old termios settings */ + struct ktermios tmp_termios; /* stores the old termios settings */ }; /* write buffer structure */ @@ -165,7 +165,7 @@ static int cypress_write (struct usb_serial_port *port, const unsigned char *b static void cypress_send (struct usb_serial_port *port); static int cypress_write_room (struct usb_serial_port *port); static int cypress_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); -static void cypress_set_termios (struct usb_serial_port *port, struct termios * old); +static void cypress_set_termios (struct usb_serial_port *port, struct ktermios * old); static int cypress_tiocmget (struct usb_serial_port *port, struct file *file); static int cypress_tiocmset (struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear); static int cypress_chars_in_buffer (struct usb_serial_port *port); @@ -949,13 +949,13 @@ static int cypress_ioctl (struct usb_serial_port *port, struct file * file, unsi switch (cmd) { case TIOCGSERIAL: - if (copy_to_user((void __user *)arg, port->tty->termios, sizeof(struct termios))) { + if (copy_to_user((void __user *)arg, port->tty->termios, sizeof(struct ktermios))) { return -EFAULT; } return (0); break; case TIOCSSERIAL: - if (copy_from_user(port->tty->termios, (void __user *)arg, sizeof(struct termios))) { + if (copy_from_user(port->tty->termios, (void __user *)arg, sizeof(struct ktermios))) { return -EFAULT; } /* here we need to call cypress_set_termios to invoke the new settings */ @@ -1019,7 +1019,7 @@ static int cypress_ioctl (struct usb_serial_port *port, struct file * file, unsi static void cypress_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct cypress_private *priv = usb_get_serial_port_data(port); struct tty_struct *tty; diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index 83d0e21145b0..9d9ea874639c 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -449,7 +449,7 @@ static int digi_transmit_idle( struct usb_serial_port *port, static void digi_rx_throttle (struct usb_serial_port *port); static void digi_rx_unthrottle (struct usb_serial_port *port); static void digi_set_termios( struct usb_serial_port *port, - struct termios *old_termios ); + struct ktermios *old_termios ); static void digi_break_ctl( struct usb_serial_port *port, int break_state ); static int digi_ioctl( struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg ); @@ -976,7 +976,7 @@ dbg( "digi_rx_unthrottle: TOP: port=%d", priv->dp_port_num ); static void digi_set_termios( struct usb_serial_port *port, - struct termios *old_termios ) + struct ktermios *old_termios ) { struct digi_port *priv = usb_get_serial_port_data(port); @@ -1463,7 +1463,7 @@ static int digi_open( struct usb_serial_port *port, struct file *filp ) int ret; unsigned char buf[32]; struct digi_port *priv = usb_get_serial_port_data(port); - struct termios not_termios; + struct ktermios not_termios; unsigned long flags = 0; diff --git a/drivers/usb/serial/empeg.c b/drivers/usb/serial/empeg.c index 4ce10a831953..92beeb19795f 100644 --- a/drivers/usb/serial/empeg.c +++ b/drivers/usb/serial/empeg.c @@ -92,7 +92,7 @@ static int empeg_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); -static void empeg_set_termios (struct usb_serial_port *port, struct termios *old_termios); +static void empeg_set_termios (struct usb_serial_port *port, struct ktermios *old_termios); static void empeg_write_bulk_callback (struct urb *urb); static void empeg_read_bulk_callback (struct urb *urb); @@ -442,7 +442,7 @@ static int empeg_ioctl (struct usb_serial_port *port, struct file * file, unsign } -static void empeg_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void empeg_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { dbg("%s - port %d", __FUNCTION__, port->number); diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 72e4d48f51e9..41b0ad2d56ac 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -595,7 +595,7 @@ static int ftdi_chars_in_buffer (struct usb_serial_port *port); static void ftdi_write_bulk_callback (struct urb *urb); static void ftdi_read_bulk_callback (struct urb *urb); static void ftdi_process_read (struct work_struct *work); -static void ftdi_set_termios (struct usb_serial_port *port, struct termios * old); +static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios * old); static int ftdi_tiocmget (struct usb_serial_port *port, struct file *file); static int ftdi_tiocmset (struct usb_serial_port *port, struct file * file, unsigned int set, unsigned int clear); static int ftdi_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); @@ -1880,7 +1880,7 @@ static void ftdi_break_ctl( struct usb_serial_port *port, int break_state ) * WARNING: set_termios calls this with old_termios in kernel space */ -static void ftdi_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void ftdi_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { /* ftdi_termios */ struct usb_device *dev = port->serial->dev; unsigned int cflag = port->tty->termios->c_cflag; diff --git a/drivers/usb/serial/io_edgeport.c b/drivers/usb/serial/io_edgeport.c index d06547a13f28..f623d58370a4 100644 --- a/drivers/usb/serial/io_edgeport.c +++ b/drivers/usb/serial/io_edgeport.c @@ -229,7 +229,7 @@ static int edge_write_room (struct usb_serial_port *port); static int edge_chars_in_buffer (struct usb_serial_port *port); static void edge_throttle (struct usb_serial_port *port); static void edge_unthrottle (struct usb_serial_port *port); -static void edge_set_termios (struct usb_serial_port *port, struct termios *old_termios); +static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old_termios); static int edge_ioctl (struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg); static void edge_break (struct usb_serial_port *port, int break_state); static int edge_tiocmget (struct usb_serial_port *port, struct file *file); @@ -257,7 +257,7 @@ static void handle_new_lsr (struct edgeport_port *edge_port, __u8 lsrData, __u8 static int send_iosp_ext_cmd (struct edgeport_port *edge_port, __u8 command, __u8 param); static int calc_baud_rate_divisor (int baud_rate, int *divisor); static int send_cmd_write_baud_rate (struct edgeport_port *edge_port, int baudRate); -static void change_port_settings (struct edgeport_port *edge_port, struct termios *old_termios); +static void change_port_settings (struct edgeport_port *edge_port, struct ktermios *old_termios); static int send_cmd_write_uart_register (struct edgeport_port *edge_port, __u8 regNum, __u8 regValue); static int write_cmd_usb (struct edgeport_port *edge_port, unsigned char *buffer, int writeLength); static void send_more_port_data (struct edgeport_serial *edge_serial, struct edgeport_port *edge_port); @@ -1431,7 +1431,7 @@ static void edge_unthrottle (struct usb_serial_port *port) * SerialSetTermios * this function is called by the tty driver when it wants to change the termios structure *****************************************************************************/ -static void edge_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { struct edgeport_port *edge_port = usb_get_serial_port_data(port); struct tty_struct *tty = port->tty; @@ -2412,7 +2412,7 @@ static int send_cmd_write_uart_register (struct edgeport_port *edge_port, __u8 r #ifndef CMSPAR #define CMSPAR 0 #endif -static void change_port_settings (struct edgeport_port *edge_port, struct termios *old_termios) +static void change_port_settings (struct edgeport_port *edge_port, struct ktermios *old_termios) { struct tty_struct *tty; int baud; diff --git a/drivers/usb/serial/io_ti.c b/drivers/usb/serial/io_ti.c index ee0c921e1520..2da2684e0809 100644 --- a/drivers/usb/serial/io_ti.c +++ b/drivers/usb/serial/io_ti.c @@ -238,7 +238,7 @@ static void edge_tty_recv(struct device *dev, struct tty_struct *tty, unsigned c static void stop_read(struct edgeport_port *edge_port); static int restart_read(struct edgeport_port *edge_port); -static void edge_set_termios (struct usb_serial_port *port, struct termios *old_termios); +static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old_termios); static void edge_send(struct usb_serial_port *port); /* circular buffer */ @@ -2361,7 +2361,7 @@ static int restart_read(struct edgeport_port *edge_port) return status; } -static void change_port_settings (struct edgeport_port *edge_port, struct termios *old_termios) +static void change_port_settings (struct edgeport_port *edge_port, struct ktermios *old_termios) { struct ump_uart_config *config; struct tty_struct *tty; @@ -2512,7 +2512,7 @@ static void change_port_settings (struct edgeport_port *edge_port, struct termio return; } -static void edge_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void edge_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { struct edgeport_port *edge_port = usb_get_serial_port_data(port); struct tty_struct *tty = port->tty; diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c index 331bf81556fc..8fdf486e3465 100644 --- a/drivers/usb/serial/ir-usb.c +++ b/drivers/usb/serial/ir-usb.c @@ -107,7 +107,7 @@ static void ir_close (struct usb_serial_port *port, struct file *filep); static int ir_write (struct usb_serial_port *port, const unsigned char *buf, int count); static void ir_write_bulk_callback (struct urb *urb); static void ir_read_bulk_callback (struct urb *urb); -static void ir_set_termios (struct usb_serial_port *port, struct termios *old_termios); +static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_termios); static u8 ir_baud = 0; static u8 ir_xbof = 0; @@ -497,7 +497,7 @@ static void ir_read_bulk_callback (struct urb *urb) return; } -static void ir_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void ir_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { unsigned char *transfer_buffer; unsigned int cflag; diff --git a/drivers/usb/serial/keyspan.c b/drivers/usb/serial/keyspan.c index 7639652cec42..9d2fdfd6865f 100644 --- a/drivers/usb/serial/keyspan.c +++ b/drivers/usb/serial/keyspan.c @@ -264,7 +264,7 @@ static void keyspan_break_ctl (struct usb_serial_port *port, int break_state) static void keyspan_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { int baud_rate, device_port; struct keyspan_port_private *p_priv; diff --git a/drivers/usb/serial/keyspan.h b/drivers/usb/serial/keyspan.h index 7472ed6bf626..6413d73c139c 100644 --- a/drivers/usb/serial/keyspan.h +++ b/drivers/usb/serial/keyspan.h @@ -59,7 +59,7 @@ static int keyspan_ioctl (struct usb_serial_port *port, unsigned int cmd, unsigned long arg); static void keyspan_set_termios (struct usb_serial_port *port, - struct termios *old); + struct ktermios *old); static void keyspan_break_ctl (struct usb_serial_port *port, int break_state); static int keyspan_tiocmget (struct usb_serial_port *port, diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c index e09a0bfe6231..126b9703bbaf 100644 --- a/drivers/usb/serial/keyspan_pda.c +++ b/drivers/usb/serial/keyspan_pda.c @@ -365,7 +365,7 @@ static void keyspan_pda_break_ctl (struct usb_serial_port *port, int break_state static void keyspan_pda_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct usb_serial *serial = port->serial; unsigned int cflag = port->tty->termios->c_cflag; diff --git a/drivers/usb/serial/kl5kusb105.c b/drivers/usb/serial/kl5kusb105.c index 17e205699c2b..73d755df4840 100644 --- a/drivers/usb/serial/kl5kusb105.c +++ b/drivers/usb/serial/kl5kusb105.c @@ -86,7 +86,7 @@ static int klsi_105_write_room (struct usb_serial_port *port); static void klsi_105_read_bulk_callback (struct urb *urb); static void klsi_105_set_termios (struct usb_serial_port *port, - struct termios * old); + struct ktermios *old); static int klsi_105_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, @@ -164,7 +164,7 @@ struct klsi_105_port_settings { #define URB_TRANSFER_BUFFER_SIZE 64 struct klsi_105_private { struct klsi_105_port_settings cfg; - struct termios termios; + struct ktermios termios; unsigned long line_state; /* modem line settings */ /* write pool */ struct urb * write_urb_pool[NUM_URBS]; @@ -688,7 +688,7 @@ static void klsi_105_read_bulk_callback (struct urb *urb) static void klsi_105_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct klsi_105_private *priv = usb_get_serial_port_data(port); unsigned int iflag = port->tty->termios->c_iflag; diff --git a/drivers/usb/serial/kobil_sct.c b/drivers/usb/serial/kobil_sct.c index 237289920f03..e284d6c0fd35 100644 --- a/drivers/usb/serial/kobil_sct.c +++ b/drivers/usb/serial/kobil_sct.c @@ -136,7 +136,7 @@ struct kobil_private { int cur_pos; // index of the next char to send in buf __u16 device_type; int line_state; - struct termios internal_termios; + struct ktermios internal_termios; }; @@ -624,11 +624,11 @@ static int kobil_ioctl(struct usb_serial_port *port, struct file *file, switch (cmd) { case TCGETS: // 0x5401 - if (!access_ok(VERIFY_WRITE, user_arg, sizeof(struct termios))) { + if (!access_ok(VERIFY_WRITE, user_arg, sizeof(struct ktermios))) { dbg("%s - port %d Error in access_ok", __FUNCTION__, port->number); return -EFAULT; } - if (kernel_termios_to_user_termios((struct termios __user *)arg, + if (kernel_termios_to_user_termios((struct ktermios __user *)arg, &priv->internal_termios)) return -EFAULT; return 0; @@ -638,12 +638,12 @@ static int kobil_ioctl(struct usb_serial_port *port, struct file *file, dbg("%s - port %d Error: port->tty->termios is NULL", __FUNCTION__, port->number); return -ENOTTY; } - if (!access_ok(VERIFY_READ, user_arg, sizeof(struct termios))) { + if (!access_ok(VERIFY_READ, user_arg, sizeof(struct ktermios))) { dbg("%s - port %d Error in access_ok", __FUNCTION__, port->number); return -EFAULT; } if (user_termios_to_kernel_termios(&priv->internal_termios, - (struct termios __user *)arg)) + (struct ktermios __user *)arg)) return -EFAULT; settings = kzalloc(50, GFP_KERNEL); diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c index a906e500a02b..38b1d17e06ef 100644 --- a/drivers/usb/serial/mct_u232.c +++ b/drivers/usb/serial/mct_u232.c @@ -98,7 +98,7 @@ static void mct_u232_close (struct usb_serial_port *port, struct file *filp); static void mct_u232_read_int_callback (struct urb *urb); static void mct_u232_set_termios (struct usb_serial_port *port, - struct termios * old); + struct ktermios * old); static int mct_u232_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, @@ -556,7 +556,7 @@ exit: } /* mct_u232_read_int_callback */ static void mct_u232_set_termios (struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct usb_serial *serial = port->serial; struct mct_u232_private *priv = usb_get_serial_port_data(port); diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c index 70f93b18292f..e55f4ed81d7b 100644 --- a/drivers/usb/serial/mos7720.c +++ b/drivers/usb/serial/mos7720.c @@ -1014,7 +1014,7 @@ static int send_cmd_write_baud_rate(struct moschip_port *mos7720_port, * the specified new settings. */ static void change_port_settings(struct moschip_port *mos7720_port, - struct termios *old_termios) + struct ktermios *old_termios) { struct usb_serial_port *port; struct usb_serial *serial; @@ -1203,7 +1203,7 @@ static void change_port_settings(struct moschip_port *mos7720_port, * termios structure. */ static void mos7720_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { int status; unsigned int cflag; diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c index 5432c6340086..8cc728a49e41 100644 --- a/drivers/usb/serial/mos7840.c +++ b/drivers/usb/serial/mos7840.c @@ -1931,7 +1931,7 @@ static int mos7840_send_cmd_write_baud_rate(struct moschip_port *mos7840_port, *****************************************************************************/ static void mos7840_change_port_settings(struct moschip_port *mos7840_port, - struct termios *old_termios) + struct ktermios *old_termios) { struct tty_struct *tty; int baud; @@ -2118,7 +2118,7 @@ static void mos7840_change_port_settings(struct moschip_port *mos7840_port, *****************************************************************************/ static void mos7840_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { int status; unsigned int cflag; diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 130afbbd3fca..0ae4098718c3 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -59,7 +59,7 @@ static int option_chars_in_buffer(struct usb_serial_port *port); static int option_ioctl(struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg); static void option_set_termios(struct usb_serial_port *port, - struct termios *old); + struct ktermios *old); static void option_break_ctl(struct usb_serial_port *port, int break_state); static int option_tiocmget(struct usb_serial_port *port, struct file *file); static int option_tiocmset(struct usb_serial_port *port, struct file *file, @@ -230,7 +230,7 @@ static void option_break_ctl(struct usb_serial_port *port, int break_state) } static void option_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { dbg("%s", __FUNCTION__); diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c index bc800c8787a8..d124d780e42e 100644 --- a/drivers/usb/serial/pl2303.c +++ b/drivers/usb/serial/pl2303.c @@ -455,7 +455,7 @@ static int pl2303_chars_in_buffer(struct usb_serial_port *port) } static void pl2303_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct usb_serial *serial = port->serial; struct pl2303_private *priv = usb_get_serial_port_data(port); @@ -687,7 +687,7 @@ static void pl2303_close(struct usb_serial_port *port, struct file *filp) static int pl2303_open(struct usb_serial_port *port, struct file *filp) { - struct termios tmp_termios; + struct ktermios tmp_termios; struct usb_serial *serial = port->serial; struct pl2303_private *priv = usb_get_serial_port_data(port); unsigned char *buf; diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c index 4b5097fa48d7..6d8e91e00ecf 100644 --- a/drivers/usb/serial/sierra.c +++ b/drivers/usb/serial/sierra.c @@ -145,7 +145,7 @@ static void sierra_break_ctl(struct usb_serial_port *port, int break_state) } static void sierra_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { dbg("%s", __FUNCTION__); diff --git a/drivers/usb/serial/ti_usb_3410_5052.c b/drivers/usb/serial/ti_usb_3410_5052.c index ae98d8cbdbb8..f42eb9ea6405 100644 --- a/drivers/usb/serial/ti_usb_3410_5052.c +++ b/drivers/usb/serial/ti_usb_3410_5052.c @@ -161,7 +161,7 @@ static void ti_throttle(struct usb_serial_port *port); static void ti_unthrottle(struct usb_serial_port *port); static int ti_ioctl(struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg); static void ti_set_termios(struct usb_serial_port *port, - struct termios *old_termios); + struct ktermios *old_termios); static int ti_tiocmget(struct usb_serial_port *port, struct file *file); static int ti_tiocmset(struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear); @@ -881,7 +881,7 @@ static int ti_ioctl(struct usb_serial_port *port, struct file *file, static void ti_set_termios(struct usb_serial_port *port, - struct termios *old_termios) + struct ktermios *old_termios) { struct ti_port *tport = usb_get_serial_port_data(port); struct tty_struct *tty = port->tty; diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c index 3d5072f14b8d..716f6806cc89 100644 --- a/drivers/usb/serial/usb-serial.c +++ b/drivers/usb/serial/usb-serial.c @@ -397,7 +397,7 @@ exit: return retval; } -static void serial_set_termios (struct tty_struct *tty, struct termios * old) +static void serial_set_termios (struct tty_struct *tty, struct ktermios * old) { struct usb_serial_port *port = tty->driver_data; diff --git a/drivers/usb/serial/visor.c b/drivers/usb/serial/visor.c index eef5eaa5fa0b..b09f06096056 100644 --- a/drivers/usb/serial/visor.c +++ b/drivers/usb/serial/visor.c @@ -46,7 +46,7 @@ static int visor_probe (struct usb_serial *serial, const struct usb_device_id static int visor_calc_num_ports(struct usb_serial *serial); static void visor_shutdown (struct usb_serial *serial); static int visor_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); -static void visor_set_termios (struct usb_serial_port *port, struct termios *old_termios); +static void visor_set_termios (struct usb_serial_port *port, struct ktermios *old_termios); static void visor_write_bulk_callback (struct urb *urb); static void visor_read_bulk_callback (struct urb *urb); static void visor_read_int_callback (struct urb *urb); @@ -916,7 +916,7 @@ static int visor_ioctl (struct usb_serial_port *port, struct file * file, unsign /* This function is all nice and good, but we don't change anything based on it :) */ -static void visor_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void visor_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { unsigned int cflag; diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c index 154c7d290597..dc45e58e2b8c 100644 --- a/drivers/usb/serial/whiteheat.c +++ b/drivers/usb/serial/whiteheat.c @@ -145,7 +145,7 @@ static void whiteheat_close (struct usb_serial_port *port, struct file *filp); static int whiteheat_write (struct usb_serial_port *port, const unsigned char *buf, int count); static int whiteheat_write_room (struct usb_serial_port *port); static int whiteheat_ioctl (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); -static void whiteheat_set_termios (struct usb_serial_port *port, struct termios * old); +static void whiteheat_set_termios (struct usb_serial_port *port, struct ktermios * old); static int whiteheat_tiocmget (struct usb_serial_port *port, struct file *file); static int whiteheat_tiocmset (struct usb_serial_port *port, struct file *file, unsigned int set, unsigned int clear); static void whiteheat_break_ctl (struct usb_serial_port *port, int break_state); @@ -597,7 +597,7 @@ static void whiteheat_shutdown (struct usb_serial *serial) static int whiteheat_open (struct usb_serial_port *port, struct file *filp) { int retval = 0; - struct termios old_term; + struct ktermios old_term; dbg("%s - port %d", __FUNCTION__, port->number); @@ -870,7 +870,7 @@ static int whiteheat_ioctl (struct usb_serial_port *port, struct file * file, un } -static void whiteheat_set_termios (struct usb_serial_port *port, struct termios *old_termios) +static void whiteheat_set_termios (struct usb_serial_port *port, struct ktermios *old_termios) { dbg("%s -port %d", __FUNCTION__, port->number); diff --git a/include/asm-generic/termios.h b/include/asm-generic/termios.h index 1e58ca39592c..3769e6bd63b1 100644 --- a/include/asm-generic/termios.h +++ b/include/asm-generic/termios.h @@ -11,7 +11,7 @@ /* * Translate a "termio" structure into a "termios". Ugh. */ -static inline int user_termio_to_kernel_termios(struct termios *termios, +static inline int user_termio_to_kernel_termios(struct ktermios *termios, struct termio __user *termio) { unsigned short tmp; @@ -48,7 +48,7 @@ static inline int user_termio_to_kernel_termios(struct termios *termios, * Translate a "termios" structure into a "termio". Ugh. */ static inline int kernel_termios_to_user_termio(struct termio __user *termio, - struct termios *termios) + struct ktermios *termios) { if (put_user(termios->c_iflag, &termio->c_iflag) < 0 || put_user(termios->c_oflag, &termio->c_oflag) < 0 || diff --git a/include/asm-powerpc/termbits.h b/include/asm-powerpc/termbits.h index b572f21b32c4..5e79198f7d18 100644 --- a/include/asm-powerpc/termbits.h +++ b/include/asm-powerpc/termbits.h @@ -37,8 +37,8 @@ struct ktermios { tcflag_t c_oflag; /* output mode flags */ tcflag_t c_cflag; /* control mode flags */ tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ cc_t c_cc[NCCS]; /* control characters */ + cc_t c_line; /* line discipline (== c_cc[19]) */ speed_t c_ispeed; /* input speed */ speed_t c_ospeed; /* output speed */ }; diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h index e25384561955..5412da28fa47 100644 --- a/include/linux/generic_serial.h +++ b/include/linux/generic_serial.h @@ -91,7 +91,7 @@ void gs_hangup(struct tty_struct *tty); int gs_block_til_ready(void *port, struct file *filp); void gs_close(struct tty_struct *tty, struct file *filp); void gs_set_termios (struct tty_struct * tty, - struct termios * old_termios); + struct ktermios * old_termios); int gs_init_port(struct gs_port *port); int gs_setserial(struct gs_port *port, struct serial_struct __user *sp); int gs_getserial(struct gs_port *port, struct serial_struct __user *sp); diff --git a/include/linux/isdn.h b/include/linux/isdn.h index 62991148d5a5..3c7875b7ab5b 100644 --- a/include/linux/isdn.h +++ b/include/linux/isdn.h @@ -511,8 +511,8 @@ typedef struct modem_info { #endif struct tty_struct *tty; /* Pointer to corresponding tty */ atemu emu; /* AT-emulator data */ - struct termios normal_termios; /* For saving termios structs */ - struct termios callout_termios; + struct ktermios normal_termios; /* For saving termios structs */ + struct ktermios callout_termios; wait_queue_head_t open_wait, close_wait; struct semaphore write_sem; spinlock_t readlock; @@ -525,8 +525,8 @@ typedef struct _isdn_modem { int refcount; /* Number of opens */ struct tty_driver *tty_modem; /* tty-device */ struct tty_struct *modem_table[ISDN_MAX_CHANNELS]; /* ?? copied from Orig */ - struct termios *modem_termios[ISDN_MAX_CHANNELS]; - struct termios *modem_termios_locked[ISDN_MAX_CHANNELS]; + struct ktermios *modem_termios[ISDN_MAX_CHANNELS]; + struct ktermios *modem_termios_locked[ISDN_MAX_CHANNELS]; modem_info info[ISDN_MAX_CHANNELS]; /* Private data */ } isdn_modem_t; diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 827672136646..cf23813cbec2 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -166,8 +166,8 @@ struct uart_ops { void (*break_ctl)(struct uart_port *, int ctl); int (*startup)(struct uart_port *); void (*shutdown)(struct uart_port *); - void (*set_termios)(struct uart_port *, struct termios *new, - struct termios *old); + void (*set_termios)(struct uart_port *, struct ktermios *new, + struct ktermios *old); void (*pm)(struct uart_port *, unsigned int state, unsigned int oldstate); int (*set_wake)(struct uart_port *, unsigned int state); @@ -361,8 +361,8 @@ void uart_write_wakeup(struct uart_port *port); */ void uart_update_timeout(struct uart_port *port, unsigned int cflag, unsigned int baud); -unsigned int uart_get_baud_rate(struct uart_port *port, struct termios *termios, - struct termios *old, unsigned int min, +unsigned int uart_get_baud_rate(struct uart_port *port, struct ktermios *termios, + struct ktermios *old, unsigned int min, unsigned int max); unsigned int uart_get_divisor(struct uart_port *port, unsigned int baud); diff --git a/include/linux/usb/serial.h b/include/linux/usb/serial.h index 91b3ea2bbb14..10f99e5f1a97 100644 --- a/include/linux/usb/serial.h +++ b/include/linux/usb/serial.h @@ -218,7 +218,7 @@ struct usb_serial_driver { int (*write) (struct usb_serial_port *port, const unsigned char *buf, int count); int (*write_room) (struct usb_serial_port *port); int (*ioctl) (struct usb_serial_port *port, struct file * file, unsigned int cmd, unsigned long arg); - void (*set_termios) (struct usb_serial_port *port, struct termios * old); + void (*set_termios) (struct usb_serial_port *port, struct ktermios * old); void (*break_ctl) (struct usb_serial_port *port, int break_state); int (*chars_in_buffer) (struct usb_serial_port *port); void (*throttle) (struct usb_serial_port *port); diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c index 1fb5d42f37ae..e0e0d09023b2 100644 --- a/net/bluetooth/rfcomm/tty.c +++ b/net/bluetooth/rfcomm/tty.c @@ -752,9 +752,9 @@ static int rfcomm_tty_ioctl(struct tty_struct *tty, struct file *filp, unsigned return -ENOIOCTLCMD; } -static void rfcomm_tty_set_termios(struct tty_struct *tty, struct termios *old) +static void rfcomm_tty_set_termios(struct tty_struct *tty, struct ktermios *old) { - struct termios *new = (struct termios *) tty->termios; + struct ktermios *new = tty->termios; int old_baud_rate = tty_termios_baud_rate(old); int new_baud_rate = tty_termios_baud_rate(new); -- cgit v1.2.3 From 6b2c9457bb377bf843f0a3ca2f4eb2ef69c67985 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 8 Dec 2006 02:39:15 -0800 Subject: [PATCH] Char: stallion, variables cleanup - fix `gcc -W' un/signed warnings by converting some ints -> uints. - move 3 global variables into functions, where are they used. Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/stallion.c | 67 ++++++++++++++++++++++-------------------------- include/linux/stallion.h | 28 ++++++++++---------- 2 files changed, 45 insertions(+), 50 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index 2db2e9fbb5c6..97c7dc9c0cf0 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -63,7 +63,7 @@ #define BRD_EASYIOPCI 28 struct stlconf { - int brdtype; + unsigned int brdtype; int ioaddr1; int ioaddr2; unsigned long memaddr; @@ -120,15 +120,6 @@ static struct ktermios stl_deftermios = { .c_ospeed = 9600, }; -/* - * Define global stats structures. Not used often, and can be - * re-used for each stats call. - */ -static comstats_t stl_comstats; -static combrd_t stl_brdstats; -static struct stlbrd stl_dummybrd; -static struct stlport stl_dummyport; - /* * Define global place to put buffer overflow characters. */ @@ -200,7 +191,7 @@ static char *stl_brdnames[] = { * load line. These allow for easy board definitions, and easy * modification of the io, memory and irq resoucres. */ -static int stl_nargs = 0; +static unsigned int stl_nargs; static char *board0[4]; static char *board1[4]; static char *board2[4]; @@ -632,7 +623,7 @@ static struct class *stallion_class; static int __init stl_parsebrd(struct stlconf *confp, char **argp) { char *sp; - int i; + unsigned int i; pr_debug("stl_parsebrd(confp=%p,argp=%p)\n", confp, argp); @@ -694,8 +685,8 @@ static int stl_open(struct tty_struct *tty, struct file *filp) { struct stlport *portp; struct stlbrd *brdp; - unsigned int minordev; - int brdnr, panelnr, portnr, rc; + unsigned int minordev, brdnr, panelnr; + int portnr, rc; pr_debug("stl_open(tty=%p,filp=%p): device=%s\n", tty, filp, tty->name); @@ -1556,8 +1547,8 @@ static int stl_readproc(char *page, char **start, off_t off, int count, int *eof struct stlbrd *brdp; struct stlpanel *panelp; struct stlport *portp; - int brdnr, panelnr, portnr, totalport; - int curoff, maxoff; + unsigned int brdnr, panelnr, portnr; + int totalport, curoff, maxoff; char *pos; pr_debug("stl_readproc(page=%p,start=%p,off=%lx,count=%d,eof=%p," @@ -1675,8 +1666,7 @@ static int stl_eiointr(struct stlbrd *brdp) static int stl_echatintr(struct stlbrd *brdp) { struct stlpanel *panelp; - unsigned int ioaddr; - int bnknr; + unsigned int ioaddr, bnknr; int handled = 0; outb((brdp->ioctrlval | ECH_BRDENABLE), brdp->ioctrl); @@ -1706,8 +1696,7 @@ static int stl_echatintr(struct stlbrd *brdp) static int stl_echmcaintr(struct stlbrd *brdp) { struct stlpanel *panelp; - unsigned int ioaddr; - int bnknr; + unsigned int ioaddr, bnknr; int handled = 0; while (inb(brdp->iostatus) & ECH_INTRPEND) { @@ -1732,8 +1721,7 @@ static int stl_echmcaintr(struct stlbrd *brdp) static int stl_echpciintr(struct stlbrd *brdp) { struct stlpanel *panelp; - unsigned int ioaddr; - int bnknr, recheck; + unsigned int ioaddr, bnknr, recheck; int handled = 0; while (1) { @@ -1763,8 +1751,7 @@ static int stl_echpciintr(struct stlbrd *brdp) static int stl_echpci64intr(struct stlbrd *brdp) { struct stlpanel *panelp; - unsigned int ioaddr; - int bnknr; + unsigned int ioaddr, bnknr; int handled = 0; while (inb(brdp->ioctrl) & 0x1) { @@ -1826,8 +1813,9 @@ static void stl_offintr(struct work_struct *work) static int __devinit stl_initports(struct stlbrd *brdp, struct stlpanel *panelp) { - struct stlport *portp; - int chipmask, i; + struct stlport *portp; + unsigned int i; + int chipmask; pr_debug("stl_initports(brdp=%p,panelp=%p)\n", brdp, panelp); @@ -2052,8 +2040,8 @@ err: static int __devinit stl_initech(struct stlbrd *brdp) { struct stlpanel *panelp; - unsigned int status, nxtid, ioaddr, conflict; - int panelnr, banknr, i, retval; + unsigned int status, nxtid, ioaddr, conflict, panelnr, banknr, i; + int retval; char *name; pr_debug("stl_initech(brdp=%p)\n", brdp); @@ -2337,7 +2325,7 @@ err: static int __devinit stl_getbrdnr(void) { - int i; + unsigned int i; for (i = 0; i < STL_MAXBRDS; i++) if (stl_brds[i] == NULL) { @@ -2361,7 +2349,7 @@ static int __devinit stl_pciprobe(struct pci_dev *pdev, { struct stlbrd *brdp; unsigned int brdtype = ent->driver_data; - int retval = -ENODEV; + int brdnr, retval = -ENODEV; if ((pdev->class >> 8) == PCI_CLASS_STORAGE_IDE) goto err; @@ -2378,13 +2366,14 @@ static int __devinit stl_pciprobe(struct pci_dev *pdev, goto err; } mutex_lock(&stl_brdslock); - brdp->brdnr = stl_getbrdnr(); - if (brdp->brdnr < 0) { + brdnr = stl_getbrdnr(); + if (brdnr < 0) { dev_err(&pdev->dev, "too many boards found, " "maximum supported %d\n", STL_MAXBRDS); mutex_unlock(&stl_brdslock); goto err_fr; } + brdp->brdnr = (unsigned int)brdnr; stl_brds[brdp->brdnr] = brdp; mutex_unlock(&stl_brdslock); @@ -2460,9 +2449,10 @@ static struct pci_driver stl_pcidriver = { static int stl_getbrdstats(combrd_t __user *bp) { + combrd_t stl_brdstats; struct stlbrd *brdp; struct stlpanel *panelp; - int i; + unsigned int i; if (copy_from_user(&stl_brdstats, bp, sizeof(combrd_t))) return -EFAULT; @@ -2508,12 +2498,12 @@ static struct stlport *stl_getport(int brdnr, int panelnr, int portnr) brdp = stl_brds[brdnr]; if (brdp == NULL) return NULL; - if (panelnr < 0 || panelnr >= brdp->nrpanels) + if (panelnr < 0 || (unsigned int)panelnr >= brdp->nrpanels) return NULL; panelp = brdp->panels[panelnr]; if (panelp == NULL) return NULL; - if (portnr < 0 || portnr >= panelp->nrports) + if (portnr < 0 || (unsigned int)portnr >= panelp->nrports) return NULL; return panelp->ports[portnr]; } @@ -2528,6 +2518,7 @@ static struct stlport *stl_getport(int brdnr, int panelnr, int portnr) static int stl_getportstats(struct stlport *portp, comstats_t __user *cp) { + comstats_t stl_comstats; unsigned char *head, *tail; unsigned long flags; @@ -2585,6 +2576,8 @@ static int stl_getportstats(struct stlport *portp, comstats_t __user *cp) static int stl_clrportstats(struct stlport *portp, comstats_t __user *cp) { + comstats_t stl_comstats; + if (!portp) { if (copy_from_user(&stl_comstats, cp, sizeof(comstats_t))) return -EFAULT; @@ -2610,6 +2603,7 @@ static int stl_clrportstats(struct stlport *portp, comstats_t __user *cp) static int stl_getportstruct(struct stlport __user *arg) { + struct stlport stl_dummyport; struct stlport *portp; if (copy_from_user(&stl_dummyport, arg, sizeof(struct stlport))) @@ -2629,11 +2623,12 @@ static int stl_getportstruct(struct stlport __user *arg) static int stl_getbrdstruct(struct stlbrd __user *arg) { + struct stlbrd stl_dummybrd; struct stlbrd *brdp; if (copy_from_user(&stl_dummybrd, arg, sizeof(struct stlbrd))) return -EFAULT; - if ((stl_dummybrd.brdnr < 0) || (stl_dummybrd.brdnr >= STL_MAXBRDS)) + if (stl_dummybrd.brdnr >= STL_MAXBRDS) return -ENODEV; brdp = stl_brds[stl_dummybrd.brdnr]; if (!brdp) diff --git a/include/linux/stallion.h b/include/linux/stallion.h index ef5270b83d02..4a0a329beafb 100644 --- a/include/linux/stallion.h +++ b/include/linux/stallion.h @@ -69,12 +69,12 @@ struct stlrq { */ struct stlport { unsigned long magic; - int portnr; - int panelnr; - int brdnr; + unsigned int portnr; + unsigned int panelnr; + unsigned int brdnr; int ioaddr; int uartaddr; - int pagenr; + unsigned int pagenr; long istate; int flags; int baud_base; @@ -102,10 +102,10 @@ struct stlport { struct stlpanel { unsigned long magic; - int panelnr; - int brdnr; - int pagenr; - int nrports; + unsigned int panelnr; + unsigned int brdnr; + unsigned int pagenr; + unsigned int nrports; int iobase; void *uartp; void (*isr)(struct stlpanel *panelp, unsigned int iobase); @@ -116,12 +116,12 @@ struct stlpanel { struct stlbrd { unsigned long magic; - int brdnr; - int brdtype; - int state; - int nrpanels; - int nrports; - int nrbnks; + unsigned int brdnr; + unsigned int brdtype; + unsigned int state; + unsigned int nrpanels; + unsigned int nrports; + unsigned int nrbnks; int irq; int irqtype; int (*isr)(struct stlbrd *brdp); -- cgit v1.2.3 From 1f8ec435e3516eb831bb98110cc2b2b28057154b Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 8 Dec 2006 02:39:18 -0800 Subject: [PATCH] Char: istallion, eliminate typedefs Use only struct instead of defining a new type . Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/istallion.c | 414 +++++++++++++++++++++++----------------------- include/linux/istallion.h | 12 +- 2 files changed, 212 insertions(+), 214 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index 2f56bc7d147b..75b1e0cbf586 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -101,14 +101,14 @@ * interrupt is required. */ -typedef struct { +struct stlconf { int brdtype; int ioaddr1; int ioaddr2; unsigned long memaddr; int irq; int irqtype; -} stlconf_t; +}; static int stli_nrbrds; @@ -185,13 +185,13 @@ static struct ktermios stli_deftermios = { */ static comstats_t stli_comstats; static combrd_t stli_brdstats; -static asystats_t stli_cdkstats; -static stlibrd_t stli_dummybrd; -static stliport_t stli_dummyport; +static struct asystats stli_cdkstats; +static struct stlibrd stli_dummybrd; +static struct stliport stli_dummyport; /*****************************************************************************/ -static stlibrd_t *stli_brds[STL_MAXBRDS]; +static struct stlibrd *stli_brds[STL_MAXBRDS]; static int stli_shared; @@ -284,12 +284,10 @@ static char **stli_brdsp[] = { * parse any module arguments. */ -typedef struct stlibrdtype { +static struct stlibrdtype { char *name; int type; -} stlibrdtype_t; - -static stlibrdtype_t stli_brdstr[] = { +} stli_brdstr[] = { { "stallion", BRD_STALLION }, { "1", BRD_STALLION }, { "brumby", BRD_BRUMBY }, @@ -594,7 +592,7 @@ static struct pci_driver stli_pcidriver; * Prototype all functions in this driver! */ -static int stli_parsebrd(stlconf_t *confp, char **argp); +static int stli_parsebrd(struct stlconf *confp, char **argp); static int stli_init(void); static int stli_open(struct tty_struct *tty, struct file *filp); static void stli_close(struct tty_struct *tty, struct file *filp); @@ -614,82 +612,82 @@ static void stli_breakctl(struct tty_struct *tty, int state); static void stli_waituntilsent(struct tty_struct *tty, int timeout); static void stli_sendxchar(struct tty_struct *tty, char ch); static void stli_hangup(struct tty_struct *tty); -static int stli_portinfo(stlibrd_t *brdp, stliport_t *portp, int portnr, char *pos); +static int stli_portinfo(struct stlibrd *brdp, struct stliport *portp, int portnr, char *pos); -static int stli_brdinit(stlibrd_t *brdp); -static int stli_startbrd(stlibrd_t *brdp); +static int stli_brdinit(struct stlibrd *brdp); +static int stli_startbrd(struct stlibrd *brdp); static ssize_t stli_memread(struct file *fp, char __user *buf, size_t count, loff_t *offp); static ssize_t stli_memwrite(struct file *fp, const char __user *buf, size_t count, loff_t *offp); static int stli_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, unsigned long arg); -static void stli_brdpoll(stlibrd_t *brdp, cdkhdr_t __iomem *hdrp); +static void stli_brdpoll(struct stlibrd *brdp, cdkhdr_t __iomem *hdrp); static void stli_poll(unsigned long arg); -static int stli_hostcmd(stlibrd_t *brdp, stliport_t *portp); -static int stli_initopen(stlibrd_t *brdp, stliport_t *portp); -static int stli_rawopen(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, int wait); -static int stli_rawclose(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, int wait); -static int stli_waitcarrier(stlibrd_t *brdp, stliport_t *portp, struct file *filp); +static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp); +static int stli_initopen(struct stlibrd *brdp, struct stliport *portp); +static int stli_rawopen(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait); +static int stli_rawclose(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait); +static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct file *filp); static void stli_dohangup(struct work_struct *); -static int stli_setport(stliport_t *portp); -static int stli_cmdwait(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback); -static void stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback); -static void __stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback); -static void stli_dodelaycmd(stliport_t *portp, cdkctrl_t __iomem *cp); -static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct ktermios *tiosp); +static int stli_setport(struct stliport *portp); +static int stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback); +static void stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback); +static void __stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback); +static void stli_dodelaycmd(struct stliport *portp, cdkctrl_t __iomem *cp); +static void stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermios *tiosp); static void stli_mkasysigs(asysigs_t *sp, int dtr, int rts); static long stli_mktiocm(unsigned long sigvalue); -static void stli_read(stlibrd_t *brdp, stliport_t *portp); -static int stli_getserial(stliport_t *portp, struct serial_struct __user *sp); -static int stli_setserial(stliport_t *portp, struct serial_struct __user *sp); +static void stli_read(struct stlibrd *brdp, struct stliport *portp); +static int stli_getserial(struct stliport *portp, struct serial_struct __user *sp); +static int stli_setserial(struct stliport *portp, struct serial_struct __user *sp); static int stli_getbrdstats(combrd_t __user *bp); -static int stli_getportstats(stliport_t *portp, comstats_t __user *cp); -static int stli_portcmdstats(stliport_t *portp); -static int stli_clrportstats(stliport_t *portp, comstats_t __user *cp); -static int stli_getportstruct(stliport_t __user *arg); -static int stli_getbrdstruct(stlibrd_t __user *arg); -static stlibrd_t *stli_allocbrd(void); - -static void stli_ecpinit(stlibrd_t *brdp); -static void stli_ecpenable(stlibrd_t *brdp); -static void stli_ecpdisable(stlibrd_t *brdp); -static void __iomem *stli_ecpgetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_ecpreset(stlibrd_t *brdp); -static void stli_ecpintr(stlibrd_t *brdp); -static void stli_ecpeiinit(stlibrd_t *brdp); -static void stli_ecpeienable(stlibrd_t *brdp); -static void stli_ecpeidisable(stlibrd_t *brdp); -static void __iomem *stli_ecpeigetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_ecpeireset(stlibrd_t *brdp); -static void stli_ecpmcenable(stlibrd_t *brdp); -static void stli_ecpmcdisable(stlibrd_t *brdp); -static void __iomem *stli_ecpmcgetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_ecpmcreset(stlibrd_t *brdp); -static void stli_ecppciinit(stlibrd_t *brdp); -static void __iomem *stli_ecppcigetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_ecppcireset(stlibrd_t *brdp); - -static void stli_onbinit(stlibrd_t *brdp); -static void stli_onbenable(stlibrd_t *brdp); -static void stli_onbdisable(stlibrd_t *brdp); -static void __iomem *stli_onbgetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_onbreset(stlibrd_t *brdp); -static void stli_onbeinit(stlibrd_t *brdp); -static void stli_onbeenable(stlibrd_t *brdp); -static void stli_onbedisable(stlibrd_t *brdp); -static void __iomem *stli_onbegetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_onbereset(stlibrd_t *brdp); -static void stli_bbyinit(stlibrd_t *brdp); -static void __iomem *stli_bbygetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_bbyreset(stlibrd_t *brdp); -static void stli_stalinit(stlibrd_t *brdp); -static void __iomem *stli_stalgetmemptr(stlibrd_t *brdp, unsigned long offset, int line); -static void stli_stalreset(stlibrd_t *brdp); - -static stliport_t *stli_getport(int brdnr, int panelnr, int portnr); - -static int stli_initecp(stlibrd_t *brdp); -static int stli_initonb(stlibrd_t *brdp); -static int stli_eisamemprobe(stlibrd_t *brdp); -static int stli_initports(stlibrd_t *brdp); +static int stli_getportstats(struct stliport *portp, comstats_t __user *cp); +static int stli_portcmdstats(struct stliport *portp); +static int stli_clrportstats(struct stliport *portp, comstats_t __user *cp); +static int stli_getportstruct(struct stliport __user *arg); +static int stli_getbrdstruct(struct stlibrd __user *arg); +static struct stlibrd *stli_allocbrd(void); + +static void stli_ecpinit(struct stlibrd *brdp); +static void stli_ecpenable(struct stlibrd *brdp); +static void stli_ecpdisable(struct stlibrd *brdp); +static void __iomem *stli_ecpgetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_ecpreset(struct stlibrd *brdp); +static void stli_ecpintr(struct stlibrd *brdp); +static void stli_ecpeiinit(struct stlibrd *brdp); +static void stli_ecpeienable(struct stlibrd *brdp); +static void stli_ecpeidisable(struct stlibrd *brdp); +static void __iomem *stli_ecpeigetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_ecpeireset(struct stlibrd *brdp); +static void stli_ecpmcenable(struct stlibrd *brdp); +static void stli_ecpmcdisable(struct stlibrd *brdp); +static void __iomem *stli_ecpmcgetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_ecpmcreset(struct stlibrd *brdp); +static void stli_ecppciinit(struct stlibrd *brdp); +static void __iomem *stli_ecppcigetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_ecppcireset(struct stlibrd *brdp); + +static void stli_onbinit(struct stlibrd *brdp); +static void stli_onbenable(struct stlibrd *brdp); +static void stli_onbdisable(struct stlibrd *brdp); +static void __iomem *stli_onbgetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_onbreset(struct stlibrd *brdp); +static void stli_onbeinit(struct stlibrd *brdp); +static void stli_onbeenable(struct stlibrd *brdp); +static void stli_onbedisable(struct stlibrd *brdp); +static void __iomem *stli_onbegetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_onbereset(struct stlibrd *brdp); +static void stli_bbyinit(struct stlibrd *brdp); +static void __iomem *stli_bbygetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_bbyreset(struct stlibrd *brdp); +static void stli_stalinit(struct stlibrd *brdp); +static void __iomem *stli_stalgetmemptr(struct stlibrd *brdp, unsigned long offset, int line); +static void stli_stalreset(struct stlibrd *brdp); + +static struct stliport *stli_getport(int brdnr, int panelnr, int portnr); + +static int stli_initecp(struct stlibrd *brdp); +static int stli_initonb(struct stlibrd *brdp); +static int stli_eisamemprobe(struct stlibrd *brdp); +static int stli_initports(struct stlibrd *brdp); /*****************************************************************************/ @@ -727,9 +725,9 @@ static int stli_timeron; static struct class *istallion_class; -static void stli_cleanup_ports(stlibrd_t *brdp) +static void stli_cleanup_ports(struct stlibrd *brdp) { - stliport_t *portp; + struct stliport *portp; unsigned int j; for (j = 0; j < STL_MAXPORTS; j++) { @@ -756,7 +754,7 @@ static int __init istallion_module_init(void) static void __exit istallion_module_exit(void) { - stlibrd_t *brdp; + struct stlibrd *brdp; int i; printk(KERN_INFO "Unloading %s: version %s\n", stli_drvtitle, @@ -811,7 +809,7 @@ module_exit(istallion_module_exit); * Parse the supplied argument string, into the board conf struct. */ -static int stli_parsebrd(stlconf_t *confp, char **argp) +static int stli_parsebrd(struct stlconf *confp, char **argp) { char *sp; int i; @@ -843,8 +841,8 @@ static int stli_parsebrd(stlconf_t *confp, char **argp) static int stli_open(struct tty_struct *tty, struct file *filp) { - stlibrd_t *brdp; - stliport_t *portp; + struct stlibrd *brdp; + struct stliport *portp; unsigned int minordev; int brdnr, portnr, rc; @@ -938,8 +936,8 @@ static int stli_open(struct tty_struct *tty, struct file *filp) static void stli_close(struct tty_struct *tty, struct file *filp) { - stlibrd_t *brdp; - stliport_t *portp; + struct stlibrd *brdp; + struct stliport *portp; unsigned long flags; portp = tty->driver_data; @@ -1016,7 +1014,7 @@ static void stli_close(struct tty_struct *tty, struct file *filp) * this still all happens pretty quickly. */ -static int stli_initopen(stlibrd_t *brdp, stliport_t *portp) +static int stli_initopen(struct stlibrd *brdp, struct stliport *portp) { struct tty_struct *tty; asynotify_t nt; @@ -1064,7 +1062,7 @@ static int stli_initopen(stlibrd_t *brdp, stliport_t *portp) * to overlap. */ -static int stli_rawopen(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, int wait) +static int stli_rawopen(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait) { cdkhdr_t __iomem *hdrp; cdkctrl_t __iomem *cp; @@ -1135,7 +1133,7 @@ static int stli_rawopen(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, i * wait is true then must have user context (to sleep). */ -static int stli_rawclose(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, int wait) +static int stli_rawclose(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait) { cdkhdr_t __iomem *hdrp; cdkctrl_t __iomem *cp; @@ -1199,7 +1197,7 @@ static int stli_rawclose(stlibrd_t *brdp, stliport_t *portp, unsigned long arg, * to complete (as opposed to initiating the command then returning). */ -static int stli_cmdwait(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback) +static int stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback) { wait_event_interruptible(portp->raw_wait, !test_bit(ST_CMDING, &portp->state)); @@ -1225,9 +1223,9 @@ static int stli_cmdwait(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, v * waiting for the command to complete - so must have user context. */ -static int stli_setport(stliport_t *portp) +static int stli_setport(struct stliport *portp) { - stlibrd_t *brdp; + struct stlibrd *brdp; asyport_t aport; if (portp == NULL) @@ -1251,7 +1249,7 @@ static int stli_setport(stliport_t *portp) * maybe because if we are clocal then we don't need to wait... */ -static int stli_waitcarrier(stlibrd_t *brdp, stliport_t *portp, struct file *filp) +static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct file *filp) { unsigned long flags; int rc, doclocal; @@ -1316,8 +1314,8 @@ static int stli_write(struct tty_struct *tty, const unsigned char *buf, int coun unsigned char __iomem *bits; unsigned char __iomem *shbuf; unsigned char *chbuf; - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned int len, stlen, head, tail, size; unsigned long flags; @@ -1423,8 +1421,8 @@ static void stli_flushchars(struct tty_struct *tty) unsigned char __iomem *bits; cdkasy_t __iomem *ap; struct tty_struct *cooktty; - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned int len, stlen, head, tail, size, count, cooksize; unsigned char *buf; unsigned char __iomem *shbuf; @@ -1511,8 +1509,8 @@ static void stli_flushchars(struct tty_struct *tty) static int stli_writeroom(struct tty_struct *tty) { cdkasyrq_t __iomem *rp; - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned int head, tail, len; unsigned long flags; @@ -1564,8 +1562,8 @@ static int stli_writeroom(struct tty_struct *tty) static int stli_charsinbuffer(struct tty_struct *tty) { cdkasyrq_t __iomem *rp; - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned int head, tail, len; unsigned long flags; @@ -1602,10 +1600,10 @@ static int stli_charsinbuffer(struct tty_struct *tty) * Generate the serial struct info. */ -static int stli_getserial(stliport_t *portp, struct serial_struct __user *sp) +static int stli_getserial(struct stliport *portp, struct serial_struct __user *sp) { struct serial_struct sio; - stlibrd_t *brdp; + struct stlibrd *brdp; memset(&sio, 0, sizeof(struct serial_struct)); sio.type = PORT_UNKNOWN; @@ -1635,7 +1633,7 @@ static int stli_getserial(stliport_t *portp, struct serial_struct __user *sp) * just quietly ignore any requests to change irq, etc. */ -static int stli_setserial(stliport_t *portp, struct serial_struct __user *sp) +static int stli_setserial(struct stliport *portp, struct serial_struct __user *sp) { struct serial_struct sio; int rc; @@ -1666,8 +1664,8 @@ static int stli_setserial(stliport_t *portp, struct serial_struct __user *sp) static int stli_tiocmget(struct tty_struct *tty, struct file *file) { - stliport_t *portp = tty->driver_data; - stlibrd_t *brdp; + struct stliport *portp = tty->driver_data; + struct stlibrd *brdp; int rc; if (portp == NULL) @@ -1690,8 +1688,8 @@ static int stli_tiocmget(struct tty_struct *tty, struct file *file) static int stli_tiocmset(struct tty_struct *tty, struct file *file, unsigned int set, unsigned int clear) { - stliport_t *portp = tty->driver_data; - stlibrd_t *brdp; + struct stliport *portp = tty->driver_data; + struct stlibrd *brdp; int rts = -1, dtr = -1; if (portp == NULL) @@ -1721,8 +1719,8 @@ static int stli_tiocmset(struct tty_struct *tty, struct file *file, static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg) { - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned int ival; int rc; void __user *argp = (void __user *)arg; @@ -1798,8 +1796,8 @@ static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cm static void stli_settermios(struct tty_struct *tty, struct ktermios *old) { - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; struct ktermios *tiosp; asyport_t aport; @@ -1844,7 +1842,7 @@ static void stli_settermios(struct tty_struct *tty, struct ktermios *old) static void stli_throttle(struct tty_struct *tty) { - stliport_t *portp = tty->driver_data; + struct stliport *portp = tty->driver_data; if (portp == NULL) return; set_bit(ST_RXSTOP, &portp->state); @@ -1860,7 +1858,7 @@ static void stli_throttle(struct tty_struct *tty) static void stli_unthrottle(struct tty_struct *tty) { - stliport_t *portp = tty->driver_data; + struct stliport *portp = tty->driver_data; if (portp == NULL) return; clear_bit(ST_RXSTOP, &portp->state); @@ -1899,7 +1897,7 @@ static void stli_start(struct tty_struct *tty) static void stli_dohangup(struct work_struct *ugly_api) { - stliport_t *portp = container_of(ugly_api, stliport_t, tqhangup); + struct stliport *portp = container_of(ugly_api, struct stliport, tqhangup); if (portp->tty != NULL) { tty_hangup(portp->tty); } @@ -1916,8 +1914,8 @@ static void stli_dohangup(struct work_struct *ugly_api) static void stli_hangup(struct tty_struct *tty) { - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned long flags; portp = tty->driver_data; @@ -1969,8 +1967,8 @@ static void stli_hangup(struct tty_struct *tty) static void stli_flushbuffer(struct tty_struct *tty) { - stliport_t *portp; - stlibrd_t *brdp; + struct stliport *portp; + struct stlibrd *brdp; unsigned long ftype, flags; portp = tty->driver_data; @@ -2006,8 +2004,8 @@ static void stli_flushbuffer(struct tty_struct *tty) static void stli_breakctl(struct tty_struct *tty, int state) { - stlibrd_t *brdp; - stliport_t *portp; + struct stlibrd *brdp; + struct stliport *portp; long arg; portp = tty->driver_data; @@ -2027,7 +2025,7 @@ static void stli_breakctl(struct tty_struct *tty, int state) static void stli_waituntilsent(struct tty_struct *tty, int timeout) { - stliport_t *portp; + struct stliport *portp; unsigned long tend; if (tty == NULL) @@ -2053,8 +2051,8 @@ static void stli_waituntilsent(struct tty_struct *tty, int timeout) static void stli_sendxchar(struct tty_struct *tty, char ch) { - stlibrd_t *brdp; - stliport_t *portp; + struct stlibrd *brdp; + struct stliport *portp; asyctrl_t actrl; portp = tty->driver_data; @@ -2088,7 +2086,7 @@ static void stli_sendxchar(struct tty_struct *tty, char ch) * short then padded with spaces). */ -static int stli_portinfo(stlibrd_t *brdp, stliport_t *portp, int portnr, char *pos) +static int stli_portinfo(struct stlibrd *brdp, struct stliport *portp, int portnr, char *pos) { char *sp, *uart; int rc, cnt; @@ -2151,8 +2149,8 @@ static int stli_portinfo(stlibrd_t *brdp, stliport_t *portp, int portnr, char *p static int stli_readproc(char *page, char **start, off_t off, int count, int *eof, void *data) { - stlibrd_t *brdp; - stliport_t *portp; + struct stlibrd *brdp; + struct stliport *portp; int brdnr, portnr, totalport; int curoff, maxoff; char *pos; @@ -2223,7 +2221,7 @@ stli_readdone: * entry point) */ -static void __stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback) +static void __stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback) { cdkhdr_t __iomem *hdrp; cdkctrl_t __iomem *cp; @@ -2259,7 +2257,7 @@ static void __stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd spin_unlock_irqrestore(&brd_lock, flags); } -static void stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, void *arg, int size, int copyback) +static void stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback) { unsigned long flags; @@ -2278,7 +2276,7 @@ static void stli_sendcmd(stlibrd_t *brdp, stliport_t *portp, unsigned long cmd, * more chars to unload. */ -static void stli_read(stlibrd_t *brdp, stliport_t *portp) +static void stli_read(struct stlibrd *brdp, struct stliport *portp) { cdkasyrq_t __iomem *rp; char __iomem *shbuf; @@ -2340,7 +2338,7 @@ static void stli_read(stlibrd_t *brdp, stliport_t *portp) * difficult to deal with them here. */ -static void stli_dodelaycmd(stliport_t *portp, cdkctrl_t __iomem *cp) +static void stli_dodelaycmd(struct stliport *portp, cdkctrl_t __iomem *cp) { int cmd; @@ -2388,7 +2386,7 @@ static void stli_dodelaycmd(stliport_t *portp, cdkctrl_t __iomem *cp) * then port is still busy, otherwise no longer busy. */ -static int stli_hostcmd(stlibrd_t *brdp, stliport_t *portp) +static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp) { cdkasy_t __iomem *ap; cdkctrl_t __iomem *cp; @@ -2535,9 +2533,9 @@ static int stli_hostcmd(stlibrd_t *brdp, stliport_t *portp) * at the cdk header structure. */ -static void stli_brdpoll(stlibrd_t *brdp, cdkhdr_t __iomem *hdrp) +static void stli_brdpoll(struct stlibrd *brdp, cdkhdr_t __iomem *hdrp) { - stliport_t *portp; + struct stliport *portp; unsigned char hostbits[(STL_MAXCHANS / 8) + 1]; unsigned char slavebits[(STL_MAXCHANS / 8) + 1]; unsigned char __iomem *slavep; @@ -2604,7 +2602,7 @@ static void stli_brdpoll(stlibrd_t *brdp, cdkhdr_t __iomem *hdrp) static void stli_poll(unsigned long arg) { cdkhdr_t __iomem *hdrp; - stlibrd_t *brdp; + struct stlibrd *brdp; int brdnr; stli_timerlist.expires = STLI_TIMEOUT; @@ -2637,7 +2635,7 @@ static void stli_poll(unsigned long arg) * the slave. */ -static void stli_mkasyport(stliport_t *portp, asyport_t *pp, struct ktermios *tiosp) +static void stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermios *tiosp) { memset(pp, 0, sizeof(asyport_t)); @@ -2786,13 +2784,13 @@ static long stli_mktiocm(unsigned long sigvalue) * we need to do here is set up the appropriate per port data structures. */ -static int stli_initports(stlibrd_t *brdp) +static int stli_initports(struct stlibrd *brdp) { - stliport_t *portp; + struct stliport *portp; int i, panelnr, panelport; for (i = 0, panelnr = 0, panelport = 0; (i < brdp->nrports); i++) { - portp = kzalloc(sizeof(stliport_t), GFP_KERNEL); + portp = kzalloc(sizeof(struct stliport), GFP_KERNEL); if (!portp) { printk("STALLION: failed to allocate port structure\n"); continue; @@ -2826,7 +2824,7 @@ static int stli_initports(stlibrd_t *brdp) * All the following routines are board specific hardware operations. */ -static void stli_ecpinit(stlibrd_t *brdp) +static void stli_ecpinit(struct stlibrd *brdp) { unsigned long memconf; @@ -2841,21 +2839,21 @@ static void stli_ecpinit(stlibrd_t *brdp) /*****************************************************************************/ -static void stli_ecpenable(stlibrd_t *brdp) +static void stli_ecpenable(struct stlibrd *brdp) { outb(ECP_ATENABLE, (brdp->iobase + ECP_ATCONFR)); } /*****************************************************************************/ -static void stli_ecpdisable(stlibrd_t *brdp) +static void stli_ecpdisable(struct stlibrd *brdp) { outb(ECP_ATDISABLE, (brdp->iobase + ECP_ATCONFR)); } /*****************************************************************************/ -static void __iomem *stli_ecpgetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_ecpgetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -2876,7 +2874,7 @@ static void __iomem *stli_ecpgetmemptr(stlibrd_t *brdp, unsigned long offset, in /*****************************************************************************/ -static void stli_ecpreset(stlibrd_t *brdp) +static void stli_ecpreset(struct stlibrd *brdp) { outb(ECP_ATSTOP, (brdp->iobase + ECP_ATCONFR)); udelay(10); @@ -2886,7 +2884,7 @@ static void stli_ecpreset(stlibrd_t *brdp) /*****************************************************************************/ -static void stli_ecpintr(stlibrd_t *brdp) +static void stli_ecpintr(struct stlibrd *brdp) { outb(0x1, brdp->iobase); } @@ -2897,7 +2895,7 @@ static void stli_ecpintr(stlibrd_t *brdp) * The following set of functions act on ECP EISA boards. */ -static void stli_ecpeiinit(stlibrd_t *brdp) +static void stli_ecpeiinit(struct stlibrd *brdp) { unsigned long memconf; @@ -2915,21 +2913,21 @@ static void stli_ecpeiinit(stlibrd_t *brdp) /*****************************************************************************/ -static void stli_ecpeienable(stlibrd_t *brdp) +static void stli_ecpeienable(struct stlibrd *brdp) { outb(ECP_EIENABLE, (brdp->iobase + ECP_EICONFR)); } /*****************************************************************************/ -static void stli_ecpeidisable(stlibrd_t *brdp) +static void stli_ecpeidisable(struct stlibrd *brdp) { outb(ECP_EIDISABLE, (brdp->iobase + ECP_EICONFR)); } /*****************************************************************************/ -static void __iomem *stli_ecpeigetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_ecpeigetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -2953,7 +2951,7 @@ static void __iomem *stli_ecpeigetmemptr(stlibrd_t *brdp, unsigned long offset, /*****************************************************************************/ -static void stli_ecpeireset(stlibrd_t *brdp) +static void stli_ecpeireset(struct stlibrd *brdp) { outb(ECP_EISTOP, (brdp->iobase + ECP_EICONFR)); udelay(10); @@ -2967,21 +2965,21 @@ static void stli_ecpeireset(stlibrd_t *brdp) * The following set of functions act on ECP MCA boards. */ -static void stli_ecpmcenable(stlibrd_t *brdp) +static void stli_ecpmcenable(struct stlibrd *brdp) { outb(ECP_MCENABLE, (brdp->iobase + ECP_MCCONFR)); } /*****************************************************************************/ -static void stli_ecpmcdisable(stlibrd_t *brdp) +static void stli_ecpmcdisable(struct stlibrd *brdp) { outb(ECP_MCDISABLE, (brdp->iobase + ECP_MCCONFR)); } /*****************************************************************************/ -static void __iomem *stli_ecpmcgetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_ecpmcgetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -3002,7 +3000,7 @@ static void __iomem *stli_ecpmcgetmemptr(stlibrd_t *brdp, unsigned long offset, /*****************************************************************************/ -static void stli_ecpmcreset(stlibrd_t *brdp) +static void stli_ecpmcreset(struct stlibrd *brdp) { outb(ECP_MCSTOP, (brdp->iobase + ECP_MCCONFR)); udelay(10); @@ -3016,7 +3014,7 @@ static void stli_ecpmcreset(stlibrd_t *brdp) * The following set of functions act on ECP PCI boards. */ -static void stli_ecppciinit(stlibrd_t *brdp) +static void stli_ecppciinit(struct stlibrd *brdp) { outb(ECP_PCISTOP, (brdp->iobase + ECP_PCICONFR)); udelay(10); @@ -3026,7 +3024,7 @@ static void stli_ecppciinit(stlibrd_t *brdp) /*****************************************************************************/ -static void __iomem *stli_ecppcigetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_ecppcigetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -3047,7 +3045,7 @@ static void __iomem *stli_ecppcigetmemptr(stlibrd_t *brdp, unsigned long offset, /*****************************************************************************/ -static void stli_ecppcireset(stlibrd_t *brdp) +static void stli_ecppcireset(struct stlibrd *brdp) { outb(ECP_PCISTOP, (brdp->iobase + ECP_PCICONFR)); udelay(10); @@ -3061,7 +3059,7 @@ static void stli_ecppcireset(stlibrd_t *brdp) * The following routines act on ONboards. */ -static void stli_onbinit(stlibrd_t *brdp) +static void stli_onbinit(struct stlibrd *brdp) { unsigned long memconf; @@ -3078,21 +3076,21 @@ static void stli_onbinit(stlibrd_t *brdp) /*****************************************************************************/ -static void stli_onbenable(stlibrd_t *brdp) +static void stli_onbenable(struct stlibrd *brdp) { outb((brdp->enabval | ONB_ATENABLE), (brdp->iobase + ONB_ATCONFR)); } /*****************************************************************************/ -static void stli_onbdisable(stlibrd_t *brdp) +static void stli_onbdisable(struct stlibrd *brdp) { outb((brdp->enabval | ONB_ATDISABLE), (brdp->iobase + ONB_ATCONFR)); } /*****************************************************************************/ -static void __iomem *stli_onbgetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_onbgetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; @@ -3109,7 +3107,7 @@ static void __iomem *stli_onbgetmemptr(stlibrd_t *brdp, unsigned long offset, in /*****************************************************************************/ -static void stli_onbreset(stlibrd_t *brdp) +static void stli_onbreset(struct stlibrd *brdp) { outb(ONB_ATSTOP, (brdp->iobase + ONB_ATCONFR)); udelay(10); @@ -3123,7 +3121,7 @@ static void stli_onbreset(stlibrd_t *brdp) * The following routines act on ONboard EISA. */ -static void stli_onbeinit(stlibrd_t *brdp) +static void stli_onbeinit(struct stlibrd *brdp) { unsigned long memconf; @@ -3143,21 +3141,21 @@ static void stli_onbeinit(stlibrd_t *brdp) /*****************************************************************************/ -static void stli_onbeenable(stlibrd_t *brdp) +static void stli_onbeenable(struct stlibrd *brdp) { outb(ONB_EIENABLE, (brdp->iobase + ONB_EICONFR)); } /*****************************************************************************/ -static void stli_onbedisable(stlibrd_t *brdp) +static void stli_onbedisable(struct stlibrd *brdp) { outb(ONB_EIDISABLE, (brdp->iobase + ONB_EICONFR)); } /*****************************************************************************/ -static void __iomem *stli_onbegetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_onbegetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -3181,7 +3179,7 @@ static void __iomem *stli_onbegetmemptr(stlibrd_t *brdp, unsigned long offset, i /*****************************************************************************/ -static void stli_onbereset(stlibrd_t *brdp) +static void stli_onbereset(struct stlibrd *brdp) { outb(ONB_EISTOP, (brdp->iobase + ONB_EICONFR)); udelay(10); @@ -3195,7 +3193,7 @@ static void stli_onbereset(stlibrd_t *brdp) * The following routines act on Brumby boards. */ -static void stli_bbyinit(stlibrd_t *brdp) +static void stli_bbyinit(struct stlibrd *brdp) { outb(BBY_ATSTOP, (brdp->iobase + BBY_ATCONFR)); udelay(10); @@ -3207,7 +3205,7 @@ static void stli_bbyinit(stlibrd_t *brdp) /*****************************************************************************/ -static void __iomem *stli_bbygetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_bbygetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { void __iomem *ptr; unsigned char val; @@ -3222,7 +3220,7 @@ static void __iomem *stli_bbygetmemptr(stlibrd_t *brdp, unsigned long offset, in /*****************************************************************************/ -static void stli_bbyreset(stlibrd_t *brdp) +static void stli_bbyreset(struct stlibrd *brdp) { outb(BBY_ATSTOP, (brdp->iobase + BBY_ATCONFR)); udelay(10); @@ -3236,7 +3234,7 @@ static void stli_bbyreset(stlibrd_t *brdp) * The following routines act on original old Stallion boards. */ -static void stli_stalinit(stlibrd_t *brdp) +static void stli_stalinit(struct stlibrd *brdp) { outb(0x1, brdp->iobase); mdelay(1000); @@ -3244,7 +3242,7 @@ static void stli_stalinit(stlibrd_t *brdp) /*****************************************************************************/ -static void __iomem *stli_stalgetmemptr(stlibrd_t *brdp, unsigned long offset, int line) +static void __iomem *stli_stalgetmemptr(struct stlibrd *brdp, unsigned long offset, int line) { BUG_ON(offset > brdp->memsize); return brdp->membase + (offset % STAL_PAGESIZE); @@ -3252,7 +3250,7 @@ static void __iomem *stli_stalgetmemptr(stlibrd_t *brdp, unsigned long offset, i /*****************************************************************************/ -static void stli_stalreset(stlibrd_t *brdp) +static void stli_stalreset(struct stlibrd *brdp) { u32 __iomem *vecp; @@ -3269,7 +3267,7 @@ static void stli_stalreset(stlibrd_t *brdp) * board types. */ -static int stli_initecp(stlibrd_t *brdp) +static int stli_initecp(struct stlibrd *brdp) { cdkecpsig_t sig; cdkecpsig_t __iomem *sigsp; @@ -3415,7 +3413,7 @@ static int stli_initecp(stlibrd_t *brdp) * This handles only these board types. */ -static int stli_initonb(stlibrd_t *brdp) +static int stli_initonb(struct stlibrd *brdp) { cdkonbsig_t sig; cdkonbsig_t __iomem *sigsp; @@ -3566,13 +3564,13 @@ static int stli_initonb(stlibrd_t *brdp) * read in the memory map, and get the show on the road... */ -static int stli_startbrd(stlibrd_t *brdp) +static int stli_startbrd(struct stlibrd *brdp) { cdkhdr_t __iomem *hdrp; cdkmem_t __iomem *memp; cdkasy_t __iomem *ap; unsigned long flags; - stliport_t *portp; + struct stliport *portp; int portnr, nrdevs, i, rc = 0; u32 memoff; @@ -3673,7 +3671,7 @@ stli_donestartup: * Probe and initialize the specified board. */ -static int __devinit stli_brdinit(stlibrd_t *brdp) +static int __devinit stli_brdinit(struct stlibrd *brdp) { stli_brds[brdp->brdnr] = brdp; @@ -3720,7 +3718,7 @@ static int __devinit stli_brdinit(stlibrd_t *brdp) * might be. This is a bit if hack, but it is the best we can do. */ -static int stli_eisamemprobe(stlibrd_t *brdp) +static int stli_eisamemprobe(struct stlibrd *brdp) { cdkecpsig_t ecpsig, __iomem *ecpsigp; cdkonbsig_t onbsig, __iomem *onbsigp; @@ -3835,7 +3833,7 @@ static int stli_getbrdnr(void) static int stli_findeisabrds(void) { - stlibrd_t *brdp; + struct stlibrd *brdp; unsigned int iobase, eid; int i; @@ -3912,7 +3910,7 @@ static int stli_findeisabrds(void) static int __devinit stli_pciprobe(struct pci_dev *pdev, const struct pci_device_id *ent) { - stlibrd_t *brdp; + struct stlibrd *brdp; int retval = -EIO; retval = pci_enable_device(pdev); @@ -3951,7 +3949,7 @@ err: static void stli_pciremove(struct pci_dev *pdev) { - stlibrd_t *brdp = pci_get_drvdata(pdev); + struct stlibrd *brdp = pci_get_drvdata(pdev); stli_cleanup_ports(brdp); @@ -3975,14 +3973,14 @@ static struct pci_driver stli_pcidriver = { * Allocate a new board structure. Fill out the basic info in it. */ -static stlibrd_t *stli_allocbrd(void) +static struct stlibrd *stli_allocbrd(void) { - stlibrd_t *brdp; + struct stlibrd *brdp; - brdp = kzalloc(sizeof(stlibrd_t), GFP_KERNEL); + brdp = kzalloc(sizeof(struct stlibrd), GFP_KERNEL); if (!brdp) { printk(KERN_ERR "STALLION: failed to allocate memory " - "(size=%Zd)\n", sizeof(stlibrd_t)); + "(size=%Zd)\n", sizeof(struct stlibrd)); return NULL; } brdp->magic = STLI_BOARDMAGIC; @@ -3998,8 +3996,8 @@ static stlibrd_t *stli_allocbrd(void) static int stli_initbrds(void) { - stlibrd_t *brdp, *nxtbrdp; - stlconf_t conf; + struct stlibrd *brdp, *nxtbrdp; + struct stlconf conf; int i, j, retval; for (stli_nrbrds = 0; stli_nrbrds < ARRAY_SIZE(stli_brdsp); @@ -4075,7 +4073,7 @@ static ssize_t stli_memread(struct file *fp, char __user *buf, size_t count, lof { unsigned long flags; void __iomem *memptr; - stlibrd_t *brdp; + struct stlibrd *brdp; int brdnr, size, n; void *p; loff_t off = *offp; @@ -4138,7 +4136,7 @@ static ssize_t stli_memwrite(struct file *fp, const char __user *buf, size_t cou { unsigned long flags; void __iomem *memptr; - stlibrd_t *brdp; + struct stlibrd *brdp; char __user *chbuf; int brdnr, size, n; void *p; @@ -4199,7 +4197,7 @@ out: static int stli_getbrdstats(combrd_t __user *bp) { - stlibrd_t *brdp; + struct stlibrd *brdp; int i; if (copy_from_user(&stli_brdstats, bp, sizeof(combrd_t))) @@ -4236,9 +4234,9 @@ static int stli_getbrdstats(combrd_t __user *bp) * Resolve the referenced port number into a port struct pointer. */ -static stliport_t *stli_getport(int brdnr, int panelnr, int portnr) +static struct stliport *stli_getport(int brdnr, int panelnr, int portnr) { - stlibrd_t *brdp; + struct stlibrd *brdp; int i; if (brdnr < 0 || brdnr >= STL_MAXBRDS) @@ -4261,10 +4259,10 @@ static stliport_t *stli_getport(int brdnr, int panelnr, int portnr) * what port to get stats for (used through board control device). */ -static int stli_portcmdstats(stliport_t *portp) +static int stli_portcmdstats(struct stliport *portp) { unsigned long flags; - stlibrd_t *brdp; + struct stlibrd *brdp; int rc; memset(&stli_comstats, 0, sizeof(comstats_t)); @@ -4335,9 +4333,9 @@ static int stli_portcmdstats(stliport_t *portp) * what port to get stats for (used through board control device). */ -static int stli_getportstats(stliport_t *portp, comstats_t __user *cp) +static int stli_getportstats(struct stliport *portp, comstats_t __user *cp) { - stlibrd_t *brdp; + struct stlibrd *brdp; int rc; if (!portp) { @@ -4366,9 +4364,9 @@ static int stli_getportstats(stliport_t *portp, comstats_t __user *cp) * Clear the port stats structure. We also return it zeroed out... */ -static int stli_clrportstats(stliport_t *portp, comstats_t __user *cp) +static int stli_clrportstats(struct stliport *portp, comstats_t __user *cp) { - stlibrd_t *brdp; + struct stlibrd *brdp; int rc; if (!portp) { @@ -4405,17 +4403,17 @@ static int stli_clrportstats(stliport_t *portp, comstats_t __user *cp) * Return the entire driver ports structure to a user app. */ -static int stli_getportstruct(stliport_t __user *arg) +static int stli_getportstruct(struct stliport __user *arg) { - stliport_t *portp; + struct stliport *portp; - if (copy_from_user(&stli_dummyport, arg, sizeof(stliport_t))) + if (copy_from_user(&stli_dummyport, arg, sizeof(struct stliport))) return -EFAULT; portp = stli_getport(stli_dummyport.brdnr, stli_dummyport.panelnr, stli_dummyport.portnr); if (!portp) return -ENODEV; - if (copy_to_user(arg, portp, sizeof(stliport_t))) + if (copy_to_user(arg, portp, sizeof(struct stliport))) return -EFAULT; return 0; } @@ -4426,18 +4424,18 @@ static int stli_getportstruct(stliport_t __user *arg) * Return the entire driver board structure to a user app. */ -static int stli_getbrdstruct(stlibrd_t __user *arg) +static int stli_getbrdstruct(struct stlibrd __user *arg) { - stlibrd_t *brdp; + struct stlibrd *brdp; - if (copy_from_user(&stli_dummybrd, arg, sizeof(stlibrd_t))) + if (copy_from_user(&stli_dummybrd, arg, sizeof(struct stlibrd))) return -EFAULT; if ((stli_dummybrd.brdnr < 0) || (stli_dummybrd.brdnr >= STL_MAXBRDS)) return -ENODEV; brdp = stli_brds[stli_dummybrd.brdnr]; if (!brdp) return -ENODEV; - if (copy_to_user(arg, brdp, sizeof(stlibrd_t))) + if (copy_to_user(arg, brdp, sizeof(struct stlibrd))) return -EFAULT; return 0; } @@ -4452,7 +4450,7 @@ static int stli_getbrdstruct(stlibrd_t __user *arg) static int stli_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, unsigned long arg) { - stlibrd_t *brdp; + struct stlibrd *brdp; int brdnr, rc, done; void __user *argp = (void __user *)arg; diff --git a/include/linux/istallion.h b/include/linux/istallion.h index b55e2a035605..af2c32d16d45 100644 --- a/include/linux/istallion.h +++ b/include/linux/istallion.h @@ -49,7 +49,7 @@ * communication with the slave board will always be on a per port * basis. */ -typedef struct { +struct stliport { unsigned long magic; int portnr; int panelnr; @@ -72,7 +72,7 @@ typedef struct { wait_queue_head_t close_wait; wait_queue_head_t raw_wait; struct work_struct tqhangup; - asysigs_t asig; + struct asysigs asig; unsigned long addr; unsigned long rxoffset; unsigned long txoffset; @@ -83,13 +83,13 @@ typedef struct { unsigned char reqbit; unsigned char portidx; unsigned char portbit; -} stliport_t; +}; /* * Use a structure of function pointers to do board level operations. * These include, enable/disable, paging shared memory, interrupting, etc. */ -typedef struct stlibrd { +struct stlibrd { unsigned long magic; int brdnr; int brdtype; @@ -116,8 +116,8 @@ typedef struct stlibrd { void __iomem *(*getmemptr)(struct stlibrd *brdp, unsigned long offset, int line); void (*intr)(struct stlibrd *brdp); void (*reset)(struct stlibrd *brdp); - stliport_t *ports[STL_MAXPORTS]; -} stlibrd_t; + struct stliport *ports[STL_MAXPORTS]; +}; /* -- cgit v1.2.3 From 1328d737f510e9933a621f66aa8de81c02b647a7 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 8 Dec 2006 02:39:19 -0800 Subject: [PATCH] Char: istallion, variables cleanup - wipe gcc -W warnings by int -> uint conversion - move 2 global variables into their local place Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/istallion.c | 100 +++++++++++++++++++++++++--------------------- include/linux/istallion.h | 24 +++++------ 2 files changed, 67 insertions(+), 57 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c index 75b1e0cbf586..28e9230e887a 100644 --- a/drivers/char/istallion.c +++ b/drivers/char/istallion.c @@ -110,7 +110,7 @@ struct stlconf { int irqtype; }; -static int stli_nrbrds; +static unsigned int stli_nrbrds; /* stli_lock must NOT be taken holding brd_lock */ static spinlock_t stli_lock; /* TTY logic lock */ @@ -186,8 +186,6 @@ static struct ktermios stli_deftermios = { static comstats_t stli_comstats; static combrd_t stli_brdstats; static struct asystats stli_cdkstats; -static struct stlibrd stli_dummybrd; -static struct stliport stli_dummyport; /*****************************************************************************/ @@ -682,7 +680,7 @@ static void stli_stalinit(struct stlibrd *brdp); static void __iomem *stli_stalgetmemptr(struct stlibrd *brdp, unsigned long offset, int line); static void stli_stalreset(struct stlibrd *brdp); -static struct stliport *stli_getport(int brdnr, int panelnr, int portnr); +static struct stliport *stli_getport(unsigned int brdnr, unsigned int panelnr, unsigned int portnr); static int stli_initecp(struct stlibrd *brdp); static int stli_initonb(struct stlibrd *brdp); @@ -755,6 +753,7 @@ static int __init istallion_module_init(void) static void __exit istallion_module_exit(void) { struct stlibrd *brdp; + unsigned int j; int i; printk(KERN_INFO "Unloading %s: version %s\n", stli_drvtitle, @@ -777,8 +776,8 @@ static void __exit istallion_module_exit(void) return; } put_tty_driver(stli_serial); - for (i = 0; i < 4; i++) - class_device_destroy(istallion_class, MKDEV(STL_SIOMEMMAJOR, i)); + for (j = 0; j < 4; j++) + class_device_destroy(istallion_class, MKDEV(STL_SIOMEMMAJOR, j)); class_destroy(istallion_class); if ((i = unregister_chrdev(STL_SIOMEMMAJOR, "staliomem"))) printk("STALLION: failed to un-register serial memory device, " @@ -786,8 +785,8 @@ static void __exit istallion_module_exit(void) kfree(stli_txcookbuf); - for (i = 0; (i < stli_nrbrds); i++) { - if ((brdp = stli_brds[i]) == NULL) + for (j = 0; (j < stli_nrbrds); j++) { + if ((brdp = stli_brds[j]) == NULL) continue; stli_cleanup_ports(brdp); @@ -796,7 +795,7 @@ static void __exit istallion_module_exit(void) if (brdp->iosize > 0) release_region(brdp->iobase, brdp->iosize); kfree(brdp); - stli_brds[i] = NULL; + stli_brds[j] = NULL; } } @@ -811,8 +810,8 @@ module_exit(istallion_module_exit); static int stli_parsebrd(struct stlconf *confp, char **argp) { + unsigned int i; char *sp; - int i; if (argp[0] == NULL || *argp[0] == 0) return 0; @@ -843,8 +842,8 @@ static int stli_open(struct tty_struct *tty, struct file *filp) { struct stlibrd *brdp; struct stliport *portp; - unsigned int minordev; - int brdnr, portnr, rc; + unsigned int minordev, brdnr, portnr; + int rc; minordev = tty->index; brdnr = MINOR2BRD(minordev); @@ -856,7 +855,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp) if ((brdp->state & BST_STARTED) == 0) return -ENODEV; portnr = MINOR2PORT(minordev); - if ((portnr < 0) || (portnr > brdp->nrports)) + if (portnr > brdp->nrports) return -ENODEV; portp = brdp->ports[portnr]; @@ -1232,7 +1231,7 @@ static int stli_setport(struct stliport *portp) return -ENODEV; if (portp->tty == NULL) return -ENODEV; - if (portp->brdnr < 0 && portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return -ENODEV; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1324,7 +1323,7 @@ static int stli_write(struct tty_struct *tty, const unsigned char *buf, int coun portp = tty->driver_data; if (portp == NULL) return 0; - if ((portp->brdnr < 0) || (portp->brdnr >= stli_nrbrds)) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1446,7 +1445,7 @@ static void stli_flushchars(struct tty_struct *tty) portp = tty->driver_data; if (portp == NULL) return; - if ((portp->brdnr < 0) || (portp->brdnr >= stli_nrbrds)) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1524,7 +1523,7 @@ static int stli_writeroom(struct tty_struct *tty) portp = tty->driver_data; if (portp == NULL) return 0; - if ((portp->brdnr < 0) || (portp->brdnr >= stli_nrbrds)) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1572,7 +1571,7 @@ static int stli_charsinbuffer(struct tty_struct *tty) portp = tty->driver_data; if (portp == NULL) return 0; - if ((portp->brdnr < 0) || (portp->brdnr >= stli_nrbrds)) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1670,7 +1669,7 @@ static int stli_tiocmget(struct tty_struct *tty, struct file *file) if (portp == NULL) return -ENODEV; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1694,7 +1693,7 @@ static int stli_tiocmset(struct tty_struct *tty, struct file *file, if (portp == NULL) return -ENODEV; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1728,7 +1727,7 @@ static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cm portp = tty->driver_data; if (portp == NULL) return -ENODEV; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return 0; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1806,7 +1805,7 @@ static void stli_settermios(struct tty_struct *tty, struct ktermios *old) portp = tty->driver_data; if (portp == NULL) return; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1921,7 +1920,7 @@ static void stli_hangup(struct tty_struct *tty) portp = tty->driver_data; if (portp == NULL) return; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -1974,7 +1973,7 @@ static void stli_flushbuffer(struct tty_struct *tty) portp = tty->driver_data; if (portp == NULL) return; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -2011,7 +2010,7 @@ static void stli_breakctl(struct tty_struct *tty, int state) portp = tty->driver_data; if (portp == NULL) return; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -2058,7 +2057,7 @@ static void stli_sendxchar(struct tty_struct *tty, char ch) portp = tty->driver_data; if (portp == NULL) return; - if (portp->brdnr < 0 || portp->brdnr >= stli_nrbrds) + if (portp->brdnr >= stli_nrbrds) return; brdp = stli_brds[portp->brdnr]; if (brdp == NULL) @@ -2151,7 +2150,7 @@ static int stli_readproc(char *page, char **start, off_t off, int count, int *eo { struct stlibrd *brdp; struct stliport *portp; - int brdnr, portnr, totalport; + unsigned int brdnr, portnr, totalport; int curoff, maxoff; char *pos; @@ -2603,7 +2602,7 @@ static void stli_poll(unsigned long arg) { cdkhdr_t __iomem *hdrp; struct stlibrd *brdp; - int brdnr; + unsigned int brdnr; stli_timerlist.expires = STLI_TIMEOUT; add_timer(&stli_timerlist); @@ -2787,7 +2786,7 @@ static long stli_mktiocm(unsigned long sigvalue) static int stli_initports(struct stlibrd *brdp) { struct stliport *portp; - int i, panelnr, panelport; + unsigned int i, panelnr, panelport; for (i = 0, panelnr = 0, panelport = 0; (i < brdp->nrports); i++) { portp = kzalloc(sizeof(struct stliport), GFP_KERNEL); @@ -3570,8 +3569,9 @@ static int stli_startbrd(struct stlibrd *brdp) cdkmem_t __iomem *memp; cdkasy_t __iomem *ap; unsigned long flags; + unsigned int portnr, nrdevs, i; struct stliport *portp; - int portnr, nrdevs, i, rc = 0; + int rc = 0; u32 memoff; spin_lock_irqsave(&brd_lock, flags); @@ -3807,7 +3807,7 @@ static int stli_eisamemprobe(struct stlibrd *brdp) static int stli_getbrdnr(void) { - int i; + unsigned int i; for (i = 0; i < STL_MAXBRDS; i++) { if (!stli_brds[i]) { @@ -3834,8 +3834,8 @@ static int stli_getbrdnr(void) static int stli_findeisabrds(void) { struct stlibrd *brdp; - unsigned int iobase, eid; - int i; + unsigned int iobase, eid, i; + int brdnr; /* * Firstly check if this is an EISA system. If this is not an EISA system then @@ -3874,8 +3874,10 @@ static int stli_findeisabrds(void) */ if ((brdp = stli_allocbrd()) == NULL) return -ENOMEM; - if ((brdp->brdnr = stli_getbrdnr()) < 0) + brdnr = stli_getbrdnr(); + if (brdnr < 0) return -ENOMEM; + brdp->brdnr = (unsigned int)brdnr; eid = inb(iobase + 0xc82); if (eid == ECP_EISAID) brdp->brdtype = BRD_ECPE; @@ -3911,7 +3913,7 @@ static int __devinit stli_pciprobe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct stlibrd *brdp; - int retval = -EIO; + int brdnr, retval = -EIO; retval = pci_enable_device(pdev); if (retval) @@ -3921,12 +3923,14 @@ static int __devinit stli_pciprobe(struct pci_dev *pdev, retval = -ENOMEM; goto err; } - if ((brdp->brdnr = stli_getbrdnr()) < 0) { /* TODO: locking */ + brdnr = stli_getbrdnr(); + if (brdnr < 0) { /* TODO: locking */ printk(KERN_INFO "STALLION: too many boards found, " "maximum supported %d\n", STL_MAXBRDS); retval = -EIO; goto err_fr; } + brdp->brdnr = (unsigned int)brdnr; brdp->brdtype = BRD_ECPPCI; /* * We have all resources from the board, so lets setup the actual @@ -3998,7 +4002,8 @@ static int stli_initbrds(void) { struct stlibrd *brdp, *nxtbrdp; struct stlconf conf; - int i, j, retval; + unsigned int i, j; + int retval; for (stli_nrbrds = 0; stli_nrbrds < ARRAY_SIZE(stli_brdsp); stli_nrbrds++) { @@ -4074,7 +4079,8 @@ static ssize_t stli_memread(struct file *fp, char __user *buf, size_t count, lof unsigned long flags; void __iomem *memptr; struct stlibrd *brdp; - int brdnr, size, n; + unsigned int brdnr; + int size, n; void *p; loff_t off = *offp; @@ -4138,7 +4144,8 @@ static ssize_t stli_memwrite(struct file *fp, const char __user *buf, size_t cou void __iomem *memptr; struct stlibrd *brdp; char __user *chbuf; - int brdnr, size, n; + unsigned int brdnr; + int size, n; void *p; loff_t off = *offp; @@ -4198,7 +4205,7 @@ out: static int stli_getbrdstats(combrd_t __user *bp) { struct stlibrd *brdp; - int i; + unsigned int i; if (copy_from_user(&stli_brdstats, bp, sizeof(combrd_t))) return -EFAULT; @@ -4234,19 +4241,20 @@ static int stli_getbrdstats(combrd_t __user *bp) * Resolve the referenced port number into a port struct pointer. */ -static struct stliport *stli_getport(int brdnr, int panelnr, int portnr) +static struct stliport *stli_getport(unsigned int brdnr, unsigned int panelnr, + unsigned int portnr) { struct stlibrd *brdp; - int i; + unsigned int i; - if (brdnr < 0 || brdnr >= STL_MAXBRDS) + if (brdnr >= STL_MAXBRDS) return NULL; brdp = stli_brds[brdnr]; if (brdp == NULL) return NULL; for (i = 0; (i < panelnr); i++) portnr += brdp->panels[i]; - if ((portnr < 0) || (portnr >= brdp->nrports)) + if (portnr >= brdp->nrports) return NULL; return brdp->ports[portnr]; } @@ -4405,6 +4413,7 @@ static int stli_clrportstats(struct stliport *portp, comstats_t __user *cp) static int stli_getportstruct(struct stliport __user *arg) { + struct stliport stli_dummyport; struct stliport *portp; if (copy_from_user(&stli_dummyport, arg, sizeof(struct stliport))) @@ -4426,11 +4435,12 @@ static int stli_getportstruct(struct stliport __user *arg) static int stli_getbrdstruct(struct stlibrd __user *arg) { + struct stlibrd stli_dummybrd; struct stlibrd *brdp; if (copy_from_user(&stli_dummybrd, arg, sizeof(struct stlibrd))) return -EFAULT; - if ((stli_dummybrd.brdnr < 0) || (stli_dummybrd.brdnr >= STL_MAXBRDS)) + if (stli_dummybrd.brdnr >= STL_MAXBRDS) return -ENODEV; brdp = stli_brds[stli_dummybrd.brdnr]; if (!brdp) diff --git a/include/linux/istallion.h b/include/linux/istallion.h index af2c32d16d45..106a5e85e5c4 100644 --- a/include/linux/istallion.h +++ b/include/linux/istallion.h @@ -51,11 +51,11 @@ */ struct stliport { unsigned long magic; - int portnr; - int panelnr; - int brdnr; + unsigned int portnr; + unsigned int panelnr; + unsigned int brdnr; unsigned long state; - int devnr; + unsigned int devnr; int flags; int baud_base; int custom_divisor; @@ -91,23 +91,23 @@ struct stliport { */ struct stlibrd { unsigned long magic; - int brdnr; - int brdtype; - int state; - int nrpanels; - int nrports; - int nrdevs; + unsigned int brdnr; + unsigned int brdtype; + unsigned int state; + unsigned int nrpanels; + unsigned int nrports; + unsigned int nrdevs; unsigned int iobase; int iosize; unsigned long memaddr; void __iomem *membase; - int memsize; + unsigned long memsize; int pagesize; int hostoffset; int slaveoffset; int bitsize; int enabval; - int panels[STL_MAXPANELS]; + unsigned int panels[STL_MAXPANELS]; int panelids[STL_MAXPANELS]; void (*init)(struct stlibrd *brdp); void (*enable)(struct stlibrd *brdp); -- cgit v1.2.3 From 6ff1cb355e628f8fc55fa2d01e269e5e1bbc2fe9 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:39:43 -0800 Subject: [PATCH] fault-injection capabilities infrastructure This patch provides base functions implement to fault-injection capabilities. - The function should_fail() is taken from failmalloc-1.0 (http://www.nongnu.org/failmalloc/) [akpm@osdl.org: cleanups, comments, add __init] Cc: Signed-off-by: Akinobu Mita Signed-off-by: Don Mullis Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fault-inject.h | 69 ++++++++++++++++ lib/Kconfig.debug | 12 +++ lib/Makefile | 1 + lib/fault-inject.c | 183 +++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 265 insertions(+) create mode 100644 include/linux/fault-inject.h create mode 100644 lib/fault-inject.c (limited to 'include/linux') diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h new file mode 100644 index 000000000000..4df4902bc8d8 --- /dev/null +++ b/include/linux/fault-inject.h @@ -0,0 +1,69 @@ +#ifndef _LINUX_FAULT_INJECT_H +#define _LINUX_FAULT_INJECT_H + +#ifdef CONFIG_FAULT_INJECTION + +#include +#include +#include + +/* + * For explanation of the elements of this struct, see + * Documentation/fault-injection/fault-injection.txt + */ +struct fault_attr { + unsigned long probability; + unsigned long interval; + atomic_t times; + atomic_t space; + unsigned long verbose; + + unsigned long count; + +#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS + + struct { + struct dentry *dir; + + struct dentry *probability_file; + struct dentry *interval_file; + struct dentry *times_file; + struct dentry *space_file; + struct dentry *verbose_file; + } dentries; + +#endif +}; + +#define FAULT_ATTR_INITIALIZER { \ + .interval = 1, \ + .times = ATOMIC_INIT(1), \ + } + +#define DECLARE_FAULT_ATTR(name) struct fault_attr name = FAULT_ATTR_INITIALIZER +int setup_fault_attr(struct fault_attr *attr, char *str); +void should_fail_srandom(unsigned long entropy); +int should_fail(struct fault_attr *attr, ssize_t size); + +#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS + +int init_fault_attr_dentries(struct fault_attr *attr, const char *name); +void cleanup_fault_attr_dentries(struct fault_attr *attr); + +#else /* CONFIG_FAULT_INJECTION_DEBUG_FS */ + +static inline int init_fault_attr_dentries(struct fault_attr *attr, + const char *name) +{ + return -ENODEV; +} + +static inline void cleanup_fault_attr_dentries(struct fault_attr *attr) +{ +} + +#endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */ + +#endif /* CONFIG_FAULT_INJECTION */ + +#endif /* _LINUX_FAULT_INJECT_H */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 2c133c098607..35228b3e7e92 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -413,3 +413,15 @@ config LKDTM Documentation on how to use the module can be found in drivers/misc/lkdtm.c + +config FAULT_INJECTION + bool + +config FAULT_INJECTION_DEBUG_FS + bool "Debugfs entries for fault-injection capabilities" + depends on FAULT_INJECTION && SYSFS + select DEBUG_FS + help + This option enable configuration of fault-injection capabilities via + debugfs. + diff --git a/lib/Makefile b/lib/Makefile index 6f675c6149a7..2d6106af53cd 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_SMP) += percpu_counter.o obj-$(CONFIG_AUDIT_GENERIC) += audit.o obj-$(CONFIG_SWIOTLB) += swiotlb.o +obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/lib/fault-inject.c b/lib/fault-inject.c new file mode 100644 index 000000000000..a7cb3afd132a --- /dev/null +++ b/lib/fault-inject.c @@ -0,0 +1,183 @@ +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * setup_fault_attr() is a helper function for various __setup handlers, so it + * returns 0 on error, because that is what __setup handlers do. + */ +int __init setup_fault_attr(struct fault_attr *attr, char *str) +{ + unsigned long probability; + unsigned long interval; + int times; + int space; + + /* ",,," */ + if (sscanf(str, "%lu,%lu,%d,%d", + &interval, &probability, &space, ×) < 4) { + printk(KERN_WARNING + "FAULT_INJECTION: failed to parse arguments\n"); + return 0; + } + + attr->probability = probability; + attr->interval = interval; + atomic_set(&attr->times, times); + atomic_set(&attr->space, space); + + return 1; +} + +static void fail_dump(struct fault_attr *attr) +{ + if (attr->verbose > 0) + printk(KERN_NOTICE "FAULT_INJECTION: forcing a failure\n"); + if (attr->verbose > 1) + dump_stack(); +} + +#define atomic_dec_not_zero(v) atomic_add_unless((v), -1, 0) + +/* + * This code is stolen from failmalloc-1.0 + * http://www.nongnu.org/failmalloc/ + */ + +int should_fail(struct fault_attr *attr, ssize_t size) +{ + if (atomic_read(&attr->times) == 0) + return 0; + + if (atomic_read(&attr->space) > size) { + atomic_sub(size, &attr->space); + return 0; + } + + if (attr->interval > 1) { + attr->count++; + if (attr->count % attr->interval) + return 0; + } + + if (attr->probability > random32() % 100) + goto fail; + + return 0; + +fail: + fail_dump(attr); + + if (atomic_read(&attr->times) != -1) + atomic_dec_not_zero(&attr->times); + + return 1; +} + +#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS + +static void debugfs_ul_set(void *data, u64 val) +{ + *(unsigned long *)data = val; +} + +static u64 debugfs_ul_get(void *data) +{ + return *(unsigned long *)data; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_ul, debugfs_ul_get, debugfs_ul_set, "%llu\n"); + +static struct dentry *debugfs_create_ul(const char *name, mode_t mode, + struct dentry *parent, unsigned long *value) +{ + return debugfs_create_file(name, mode, parent, value, &fops_ul); +} + +static void debugfs_atomic_t_set(void *data, u64 val) +{ + atomic_set((atomic_t *)data, val); +} + +static u64 debugfs_atomic_t_get(void *data) +{ + return atomic_read((atomic_t *)data); +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_atomic_t, debugfs_atomic_t_get, + debugfs_atomic_t_set, "%lld\n"); + +static struct dentry *debugfs_create_atomic_t(const char *name, mode_t mode, + struct dentry *parent, atomic_t *value) +{ + return debugfs_create_file(name, mode, parent, value, &fops_atomic_t); +} + +void cleanup_fault_attr_dentries(struct fault_attr *attr) +{ + debugfs_remove(attr->dentries.probability_file); + attr->dentries.probability_file = NULL; + + debugfs_remove(attr->dentries.interval_file); + attr->dentries.interval_file = NULL; + + debugfs_remove(attr->dentries.times_file); + attr->dentries.times_file = NULL; + + debugfs_remove(attr->dentries.space_file); + attr->dentries.space_file = NULL; + + debugfs_remove(attr->dentries.verbose_file); + attr->dentries.verbose_file = NULL; + + if (attr->dentries.dir) + WARN_ON(!simple_empty(attr->dentries.dir)); + + debugfs_remove(attr->dentries.dir); + attr->dentries.dir = NULL; +} + +int init_fault_attr_dentries(struct fault_attr *attr, const char *name) +{ + mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; + struct dentry *dir; + + memset(&attr->dentries, 0, sizeof(attr->dentries)); + + dir = debugfs_create_dir(name, NULL); + if (!dir) + goto fail; + attr->dentries.dir = dir; + + attr->dentries.probability_file = + debugfs_create_ul("probability", mode, dir, &attr->probability); + + attr->dentries.interval_file = + debugfs_create_ul("interval", mode, dir, &attr->interval); + + attr->dentries.times_file = + debugfs_create_atomic_t("times", mode, dir, &attr->times); + + attr->dentries.space_file = + debugfs_create_atomic_t("space", mode, dir, &attr->space); + + attr->dentries.verbose_file = + debugfs_create_ul("verbose", mode, dir, &attr->verbose); + + if (!attr->dentries.probability_file || !attr->dentries.interval_file + || !attr->dentries.times_file || !attr->dentries.space_file + || !attr->dentries.verbose_file) + goto fail; + + return 0; +fail: + cleanup_fault_attr_dentries(attr); + return -ENOMEM; +} + +#endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */ -- cgit v1.2.3 From c17bb4951752d3e0f49cd1ea9d2e868422f9e0d6 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:39:46 -0800 Subject: [PATCH] fault-injection capability for disk IO This patch provides fault-injection capability for disk IO. Boot option: fail_make_request=,,, -- specifies the interval of failures. -- specifies how often it should fail in percent. -- specifies the size of free space where disk IO can be issued safely in bytes. -- specifies how many times failures may happen at most. Debugfs: /debug/fail_make_request/interval /debug/fail_make_request/probability /debug/fail_make_request/specifies /debug/fail_make_request/times Example: fail_make_request=10,100,0,-1 echo 1 > /sys/blocks/hda/hda1/make-it-fail generic_make_request() on /dev/hda1 fails once per 10 times. Cc: Jens Axboe Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- block/genhd.c | 31 +++++++++++++++++++++++++++++++ block/ll_rw_blk.c | 40 ++++++++++++++++++++++++++++++++++++++++ fs/partitions/check.c | 27 +++++++++++++++++++++++++++ include/linux/genhd.h | 4 ++++ lib/Kconfig.debug | 7 +++++++ 5 files changed, 109 insertions(+) (limited to 'include/linux') diff --git a/block/genhd.c b/block/genhd.c index 653919d50cd4..457fdac4c17d 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -417,6 +417,34 @@ static struct disk_attribute disk_attr_stat = { .show = disk_stats_read }; +#ifdef CONFIG_FAIL_MAKE_REQUEST + +static ssize_t disk_fail_store(struct gendisk * disk, + const char *buf, size_t count) +{ + int i; + + if (count > 0 && sscanf(buf, "%d", &i) > 0) { + if (i == 0) + disk->flags &= ~GENHD_FL_FAIL; + else + disk->flags |= GENHD_FL_FAIL; + } + + return count; +} +static ssize_t disk_fail_read(struct gendisk * disk, char *page) +{ + return sprintf(page, "%d\n", disk->flags & GENHD_FL_FAIL ? 1 : 0); +} +static struct disk_attribute disk_attr_fail = { + .attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR }, + .store = disk_fail_store, + .show = disk_fail_read +}; + +#endif + static struct attribute * default_attrs[] = { &disk_attr_uevent.attr, &disk_attr_dev.attr, @@ -424,6 +452,9 @@ static struct attribute * default_attrs[] = { &disk_attr_removable.attr, &disk_attr_size.attr, &disk_attr_stat.attr, +#ifdef CONFIG_FAIL_MAKE_REQUEST + &disk_attr_fail.attr, +#endif NULL, }; diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 31512cd9f3ad..4f83fd922377 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -28,6 +28,7 @@ #include #include #include +#include /* * for max sense size @@ -3056,6 +3057,42 @@ static void handle_bad_sector(struct bio *bio) set_bit(BIO_EOF, &bio->bi_flags); } +#ifdef CONFIG_FAIL_MAKE_REQUEST + +static DECLARE_FAULT_ATTR(fail_make_request); + +static int __init setup_fail_make_request(char *str) +{ + return setup_fault_attr(&fail_make_request, str); +} +__setup("fail_make_request=", setup_fail_make_request); + +static int should_fail_request(struct bio *bio) +{ + if ((bio->bi_bdev->bd_disk->flags & GENHD_FL_FAIL) || + (bio->bi_bdev->bd_part && bio->bi_bdev->bd_part->make_it_fail)) + return should_fail(&fail_make_request, bio->bi_size); + + return 0; +} + +static int __init fail_make_request_debugfs(void) +{ + return init_fault_attr_dentries(&fail_make_request, + "fail_make_request"); +} + +late_initcall(fail_make_request_debugfs); + +#else /* CONFIG_FAIL_MAKE_REQUEST */ + +static inline int should_fail_request(struct bio *bio) +{ + return 0; +} + +#endif /* CONFIG_FAIL_MAKE_REQUEST */ + /** * generic_make_request: hand a buffer to its device driver for I/O * @bio: The bio describing the location in memory and on the device. @@ -3141,6 +3178,9 @@ end_io: if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) goto end_io; + if (should_fail_request(bio)) + goto end_io; + /* * If this device has partitions, remap block n * of partition p to block n+start(p) of the disk. diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 1901137f4eca..3d73d94d93a7 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -276,12 +276,39 @@ static struct part_attribute part_attr_stat = { .show = part_stat_read }; +#ifdef CONFIG_FAIL_MAKE_REQUEST + +static ssize_t part_fail_store(struct hd_struct * p, + const char *buf, size_t count) +{ + int i; + + if (count > 0 && sscanf(buf, "%d", &i) > 0) + p->make_it_fail = (i == 0) ? 0 : 1; + + return count; +} +static ssize_t part_fail_read(struct hd_struct * p, char *page) +{ + return sprintf(page, "%d\n", p->make_it_fail); +} +static struct part_attribute part_attr_fail = { + .attr = {.name = "make-it-fail", .mode = S_IRUGO | S_IWUSR }, + .store = part_fail_store, + .show = part_fail_read +}; + +#endif + static struct attribute * default_attrs[] = { &part_attr_uevent.attr, &part_attr_dev.attr, &part_attr_start.attr, &part_attr_size.attr, &part_attr_stat.attr, +#ifdef CONFIG_FAIL_MAKE_REQUEST + &part_attr_fail.attr, +#endif NULL, }; diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 41f276fdd185..0a022b2f63fc 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -83,6 +83,9 @@ struct hd_struct { struct kobject *holder_dir; unsigned ios[2], sectors[2]; /* READs and WRITEs */ int policy, partno; +#ifdef CONFIG_FAIL_MAKE_REQUEST + int make_it_fail; +#endif }; #define GENHD_FL_REMOVABLE 1 @@ -90,6 +93,7 @@ struct hd_struct { #define GENHD_FL_CD 8 #define GENHD_FL_UP 16 #define GENHD_FL_SUPPRESS_PARTITION_INFO 32 +#define GENHD_FL_FAIL 64 struct disk_stats { unsigned long sectors[2]; /* READs and WRITEs */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index c66b7b400c9f..336241bf0ced 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -431,6 +431,13 @@ config FAIL_PAGE_ALLOC help This option provides fault-injection capabilitiy for alloc_pages(). +config FAIL_MAKE_REQUEST + bool "Fault-injection capabilitiy for disk IO" + depends on DEBUG_KERNEL + select FAULT_INJECTION + help + This option provides fault-injection capabilitiy to disk IO. + config FAULT_INJECTION_DEBUG_FS bool "Debugfs entries for fault-injection capabilities" depends on FAULT_INJECTION && SYSFS -- cgit v1.2.3 From f4f154fd920b2178382a6a24a236348e4429ebc1 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:39:47 -0800 Subject: [PATCH] fault injection: process filtering for fault-injection capabilities This patch provides process filtering feature. The process filter allows failing only permitted processes by /proc//make-it-fail Please see the example that demostrates how to inject slab allocation failures into module init/cleanup code in Documentation/fault-injection/fault-injection.txt Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 65 ++++++++++++++++++++++++++++++++++++++++++++ include/linux/fault-inject.h | 2 ++ include/linux/sched.h | 3 ++ lib/fault-inject.c | 17 +++++++++++- 4 files changed, 86 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/proc/base.c b/fs/proc/base.c index a3b5074118a7..fd959d5b5a80 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -853,6 +853,65 @@ static struct file_operations proc_seccomp_operations = { }; #endif /* CONFIG_SECCOMP */ +#ifdef CONFIG_FAULT_INJECTION +static ssize_t proc_fault_inject_read(struct file * file, char __user * buf, + size_t count, loff_t *ppos) +{ + struct task_struct *task = get_proc_task(file->f_dentry->d_inode); + char buffer[PROC_NUMBUF]; + size_t len; + int make_it_fail; + loff_t __ppos = *ppos; + + if (!task) + return -ESRCH; + make_it_fail = task->make_it_fail; + put_task_struct(task); + + len = snprintf(buffer, sizeof(buffer), "%i\n", make_it_fail); + if (__ppos >= len) + return 0; + if (count > len-__ppos) + count = len-__ppos; + if (copy_to_user(buf, buffer + __ppos, count)) + return -EFAULT; + *ppos = __ppos + count; + return count; +} + +static ssize_t proc_fault_inject_write(struct file * file, + const char __user * buf, size_t count, loff_t *ppos) +{ + struct task_struct *task; + char buffer[PROC_NUMBUF], *end; + int make_it_fail; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + make_it_fail = simple_strtol(buffer, &end, 0); + if (*end == '\n') + end++; + task = get_proc_task(file->f_dentry->d_inode); + if (!task) + return -ESRCH; + task->make_it_fail = make_it_fail; + put_task_struct(task); + if (end - buffer == 0) + return -EIO; + return end - buffer; +} + +static struct file_operations proc_fault_inject_operations = { + .read = proc_fault_inject_read, + .write = proc_fault_inject_write, +}; +#endif + static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd) { struct inode *inode = dentry->d_inode; @@ -1793,6 +1852,9 @@ static struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, loginuid), #endif +#ifdef CONFIG_FAULT_INJECTION + REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), +#endif }; static int proc_tgid_base_readdir(struct file * filp, @@ -2068,6 +2130,9 @@ static struct pid_entry tid_base_stuff[] = { #ifdef CONFIG_AUDITSYSCALL REG("loginuid", S_IWUSR|S_IRUGO, loginuid), #endif +#ifdef CONFIG_FAULT_INJECTION + REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), +#endif }; static int proc_tid_base_readdir(struct file * filp, diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h index 4df4902bc8d8..a525f9b9f015 100644 --- a/include/linux/fault-inject.h +++ b/include/linux/fault-inject.h @@ -17,6 +17,7 @@ struct fault_attr { atomic_t times; atomic_t space; unsigned long verbose; + u32 task_filter; unsigned long count; @@ -30,6 +31,7 @@ struct fault_attr { struct dentry *times_file; struct dentry *space_file; struct dentry *verbose_file; + struct dentry *task_filter_file; } dentries; #endif diff --git a/include/linux/sched.h b/include/linux/sched.h index f0317edea141..ad9c46071ff8 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1045,6 +1045,9 @@ struct task_struct { #ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info *delays; #endif +#ifdef CONFIG_FAULT_INJECTION + int make_it_fail; +#endif }; static inline pid_t process_group(struct task_struct *tsk) diff --git a/lib/fault-inject.c b/lib/fault-inject.c index a7cb3afd132a..03468609d701 100644 --- a/lib/fault-inject.c +++ b/lib/fault-inject.c @@ -5,6 +5,7 @@ #include #include #include +#include #include /* @@ -44,6 +45,11 @@ static void fail_dump(struct fault_attr *attr) #define atomic_dec_not_zero(v) atomic_add_unless((v), -1, 0) +static int fail_task(struct fault_attr *attr, struct task_struct *task) +{ + return !in_interrupt() && task->make_it_fail; +} + /* * This code is stolen from failmalloc-1.0 * http://www.nongnu.org/failmalloc/ @@ -51,6 +57,9 @@ static void fail_dump(struct fault_attr *attr) int should_fail(struct fault_attr *attr, ssize_t size) { + if (attr->task_filter && !fail_task(attr, current)) + return 0; + if (atomic_read(&attr->times) == 0) return 0; @@ -135,6 +144,9 @@ void cleanup_fault_attr_dentries(struct fault_attr *attr) debugfs_remove(attr->dentries.verbose_file); attr->dentries.verbose_file = NULL; + debugfs_remove(attr->dentries.task_filter_file); + attr->dentries.task_filter_file = NULL; + if (attr->dentries.dir) WARN_ON(!simple_empty(attr->dentries.dir)); @@ -169,9 +181,12 @@ int init_fault_attr_dentries(struct fault_attr *attr, const char *name) attr->dentries.verbose_file = debugfs_create_ul("verbose", mode, dir, &attr->verbose); + attr->dentries.task_filter_file = debugfs_create_bool("task-filter", + mode, dir, &attr->task_filter); + if (!attr->dentries.probability_file || !attr->dentries.interval_file || !attr->dentries.times_file || !attr->dentries.space_file - || !attr->dentries.verbose_file) + || !attr->dentries.verbose_file || !attr->dentries.task_filter_file) goto fail; return 0; -- cgit v1.2.3 From 329409aeda064c4aff00c51f837fcd3bbdaeeba6 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Fri, 8 Dec 2006 02:39:48 -0800 Subject: [PATCH] fault injection: stacktrace filtering This patch provides stacktrace filtering feature. The stacktrace filter allows failing only for the caller you are interested in. For example someone may want to inject kmalloc() failures into only e100 module. they want to inject not only direct kmalloc() call, but also indirect allocation, too. - e100_poll --> netif_receive_skb --> packet_rcv_spkt --> skb_clone --> kmem_cache_alloc This patch enables to detect function calls like this by stacktrace and inject failures. The script Documentaion/fault-injection/failmodule.sh helps it. The range of text section of loaded e100 is expected to be [/sys/module/e100/sections/.text, /sys/module/e100/sections/.exit.text) So failmodule.sh stores these values into /debug/failslab/address-start and /debug/failslab/address-end. The maximum stacktrace depth is specified by /debug/failslab/stacktrace-depth. Please see the example that demonstrates how to inject slab allocation failures only for a specific module in Documentation/fault-injection/fault-injection.txt [dwm@meer.net: reject failure if any caller lies within specified range] Signed-off-by: Akinobu Mita Signed-off-by: Don Mullis Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/fault-injection/fault-injection.txt | 12 ++- include/linux/fault-inject.h | 12 +++ lib/Kconfig.debug | 5 + lib/fault-inject.c | 126 +++++++++++++++++++++- 4 files changed, 150 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/Documentation/fault-injection/fault-injection.txt b/Documentation/fault-injection/fault-injection.txt index cf075c20eda0..6d6e5ac5ea92 100644 --- a/Documentation/fault-injection/fault-injection.txt +++ b/Documentation/fault-injection/fault-injection.txt @@ -73,13 +73,17 @@ configuration of fault-injection capabilities. Any positive value limits failures to only processes indicated by /proc//make-it-fail==1. -- /debug/*/address-start: -- /debug/*/address-end: +- /debug/*/require-start: +- /debug/*/require-end: +- /debug/*/reject-start: +- /debug/*/reject-end: specifies the range of virtual addresses tested during stacktrace walking. Failure is injected only if some caller - in the walked stacktrace lies within this range. - Default is [0,ULONG_MAX) (whole of virtual address space). + in the walked stacktrace lies within the required range, and + none lies within the rejected range. + Default required range is [0,ULONG_MAX) (whole of virtual address space). + Default rejected range is [0,0). - /debug/*/stacktrace-depth: diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h index a525f9b9f015..9bb584e89399 100644 --- a/include/linux/fault-inject.h +++ b/include/linux/fault-inject.h @@ -18,6 +18,11 @@ struct fault_attr { atomic_t space; unsigned long verbose; u32 task_filter; + unsigned long stacktrace_depth; + unsigned long require_start; + unsigned long require_end; + unsigned long reject_start; + unsigned long reject_end; unsigned long count; @@ -32,6 +37,11 @@ struct fault_attr { struct dentry *space_file; struct dentry *verbose_file; struct dentry *task_filter_file; + struct dentry *stacktrace_depth_file; + struct dentry *require_start_file; + struct dentry *require_end_file; + struct dentry *reject_start_file; + struct dentry *reject_end_file; } dentries; #endif @@ -40,6 +50,8 @@ struct fault_attr { #define FAULT_ATTR_INITIALIZER { \ .interval = 1, \ .times = ATOMIC_INIT(1), \ + .require_end = ULONG_MAX, \ + .stacktrace_depth = 32, \ } #define DECLARE_FAULT_ATTR(name) struct fault_attr name = FAULT_ATTR_INITIALIZER diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 336241bf0ced..1449ca1bf572 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -416,6 +416,11 @@ config LKDTM config FAULT_INJECTION bool + select STACKTRACE + select FRAME_POINTER + help + Provide fault-injection framework. + For more details, see Documentation/fault-injection/. config FAILSLAB bool "Fault-injection capabilitiy for kmalloc" diff --git a/lib/fault-inject.c b/lib/fault-inject.c index 03468609d701..361c6e9cd77f 100644 --- a/lib/fault-inject.c +++ b/lib/fault-inject.c @@ -6,6 +6,9 @@ #include #include #include +#include +#include +#include #include /* @@ -50,6 +53,86 @@ static int fail_task(struct fault_attr *attr, struct task_struct *task) return !in_interrupt() && task->make_it_fail; } +#ifdef CONFIG_STACK_UNWIND + +static asmlinkage int fail_stacktrace_callback(struct unwind_frame_info *info, + void *arg) +{ + int depth; + struct fault_attr *attr = arg; + bool found = (attr->require_start == 0 && attr->require_end == ULONG_MAX); + + for (depth = 0; depth < attr->stacktrace_depth + && unwind(info) == 0 && UNW_PC(info); depth++) { + if (arch_unw_user_mode(info)) + break; + if (attr->reject_start <= UNW_PC(info) && + UNW_PC(info) < attr->reject_end) + return 0; + if (attr->require_start <= UNW_PC(info) && + UNW_PC(info) < attr->require_end) + found = 1; + } + return found; +} + +static int fail_stacktrace(struct fault_attr *attr) +{ + struct unwind_frame_info info; + + return unwind_init_running(&info, fail_stacktrace_callback, attr); +} + +#elif defined(CONFIG_STACKTRACE) + +#define MAX_STACK_TRACE_DEPTH 32 + +static int fail_stacktrace(struct fault_attr *attr) +{ + struct stack_trace trace; + int depth = attr->stacktrace_depth; + unsigned long entries[MAX_STACK_TRACE_DEPTH]; + int n; + bool found = (attr->require_start == 0 && attr->require_end == ULONG_MAX); + + if (depth == 0) + return found; + + trace.nr_entries = 0; + trace.entries = entries; + trace.max_entries = (depth < MAX_STACK_TRACE_DEPTH) ? + depth : MAX_STACK_TRACE_DEPTH; + trace.skip = 1; + trace.all_contexts = 0; + + save_stack_trace(&trace, NULL); + for (n = 0; n < trace.nr_entries; n++) { + if (attr->reject_start <= entries[n] && + entries[n] < attr->reject_end) + return 0; + if (attr->require_start <= entries[n] && + entries[n] < attr->require_end) + found = 1; + } + return found; +} + +#else + +static inline int fail_stacktrace(struct fault_attr *attr) +{ + static int firsttime = 1; + + if (firsttime) { + printk(KERN_WARNING + "This architecture does not implement save_stack_trace()\n"); + firsttime = 0; + } + return 0; +} + +#endif + /* * This code is stolen from failmalloc-1.0 * http://www.nongnu.org/failmalloc/ @@ -60,6 +143,9 @@ int should_fail(struct fault_attr *attr, ssize_t size) if (attr->task_filter && !fail_task(attr, current)) return 0; + if (!fail_stacktrace(attr)) + return 0; + if (atomic_read(&attr->times) == 0) return 0; @@ -147,6 +233,21 @@ void cleanup_fault_attr_dentries(struct fault_attr *attr) debugfs_remove(attr->dentries.task_filter_file); attr->dentries.task_filter_file = NULL; + debugfs_remove(attr->dentries.stacktrace_depth_file); + attr->dentries.stacktrace_depth_file = NULL; + + debugfs_remove(attr->dentries.require_start_file); + attr->dentries.require_start_file = NULL; + + debugfs_remove(attr->dentries.require_end_file); + attr->dentries.require_end_file = NULL; + + debugfs_remove(attr->dentries.reject_start_file); + attr->dentries.reject_start_file = NULL; + + debugfs_remove(attr->dentries.reject_end_file); + attr->dentries.reject_end_file = NULL; + if (attr->dentries.dir) WARN_ON(!simple_empty(attr->dentries.dir)); @@ -184,9 +285,32 @@ int init_fault_attr_dentries(struct fault_attr *attr, const char *name) attr->dentries.task_filter_file = debugfs_create_bool("task-filter", mode, dir, &attr->task_filter); + attr->dentries.stacktrace_depth_file = + debugfs_create_ul("stacktrace-depth", mode, dir, + &attr->stacktrace_depth); + + attr->dentries.require_start_file = + debugfs_create_ul("require-start", mode, dir, &attr->require_start); + + attr->dentries.require_end_file = + debugfs_create_ul("require-end", mode, dir, &attr->require_end); + + attr->dentries.reject_start_file = + debugfs_create_ul("reject-start", mode, dir, &attr->reject_start); + + attr->dentries.reject_end_file = + debugfs_create_ul("reject-end", mode, dir, &attr->reject_end); + + if (!attr->dentries.probability_file || !attr->dentries.interval_file || !attr->dentries.times_file || !attr->dentries.space_file - || !attr->dentries.verbose_file || !attr->dentries.task_filter_file) + || !attr->dentries.verbose_file || !attr->dentries.task_filter_file + || !attr->dentries.stacktrace_depth_file + || !attr->dentries.require_start_file + || !attr->dentries.require_end_file + || !attr->dentries.reject_start_file + || !attr->dentries.reject_end_file + ) goto fail; return 0; -- cgit v1.2.3 From 08b3df2d16cbebf7d72c09dcbc071696c14d07e3 Mon Sep 17 00:00:00 2001 From: Don Mullis Date: Fri, 8 Dec 2006 02:39:51 -0800 Subject: [PATCH] fault-injection: Use bool-true-false throughout Use bool-true-false throughout. Signed-off-by: Don Mullis Cc: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fault-inject.h | 2 +- lib/fault-inject.c | 40 +++++++++++++++++++--------------------- 2 files changed, 20 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h index 9bb584e89399..c77067916b7e 100644 --- a/include/linux/fault-inject.h +++ b/include/linux/fault-inject.h @@ -57,7 +57,7 @@ struct fault_attr { #define DECLARE_FAULT_ATTR(name) struct fault_attr name = FAULT_ATTR_INITIALIZER int setup_fault_attr(struct fault_attr *attr, char *str); void should_fail_srandom(unsigned long entropy); -int should_fail(struct fault_attr *attr, ssize_t size); +bool should_fail(struct fault_attr *attr, ssize_t size); #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS diff --git a/lib/fault-inject.c b/lib/fault-inject.c index 361c6e9cd77f..244782940349 100644 --- a/lib/fault-inject.c +++ b/lib/fault-inject.c @@ -48,7 +48,7 @@ static void fail_dump(struct fault_attr *attr) #define atomic_dec_not_zero(v) atomic_add_unless((v), -1, 0) -static int fail_task(struct fault_attr *attr, struct task_struct *task) +static bool fail_task(struct fault_attr *attr, struct task_struct *task) { return !in_interrupt() && task->make_it_fail; } @@ -68,15 +68,15 @@ static asmlinkage int fail_stacktrace_callback(struct unwind_frame_info *info, break; if (attr->reject_start <= UNW_PC(info) && UNW_PC(info) < attr->reject_end) - return 0; + return false; if (attr->require_start <= UNW_PC(info) && UNW_PC(info) < attr->require_end) - found = 1; + found = true; } return found; } -static int fail_stacktrace(struct fault_attr *attr) +static bool fail_stacktrace(struct fault_attr *attr) { struct unwind_frame_info info; @@ -85,9 +85,7 @@ static int fail_stacktrace(struct fault_attr *attr) #elif defined(CONFIG_STACKTRACE) -#define MAX_STACK_TRACE_DEPTH 32 - -static int fail_stacktrace(struct fault_attr *attr) +static bool fail_stacktrace(struct fault_attr *attr) { struct stack_trace trace; int depth = attr->stacktrace_depth; @@ -109,26 +107,26 @@ static int fail_stacktrace(struct fault_attr *attr) for (n = 0; n < trace.nr_entries; n++) { if (attr->reject_start <= entries[n] && entries[n] < attr->reject_end) - return 0; + return false; if (attr->require_start <= entries[n] && entries[n] < attr->require_end) - found = 1; + found = true; } return found; } #else -static inline int fail_stacktrace(struct fault_attr *attr) +static inline bool fail_stacktrace(struct fault_attr *attr) { - static int firsttime = 1; + static bool firsttime = true; if (firsttime) { printk(KERN_WARNING "This architecture does not implement save_stack_trace()\n"); - firsttime = 0; + firsttime = false; } - return 0; + return false; } #endif @@ -138,32 +136,32 @@ static inline int fail_stacktrace(struct fault_attr *attr) * http://www.nongnu.org/failmalloc/ */ -int should_fail(struct fault_attr *attr, ssize_t size) +bool should_fail(struct fault_attr *attr, ssize_t size) { if (attr->task_filter && !fail_task(attr, current)) - return 0; + return false; if (!fail_stacktrace(attr)) - return 0; + return false; if (atomic_read(&attr->times) == 0) - return 0; + return false; if (atomic_read(&attr->space) > size) { atomic_sub(size, &attr->space); - return 0; + return false; } if (attr->interval > 1) { attr->count++; if (attr->count % attr->interval) - return 0; + return false; } if (attr->probability > random32() % 100) goto fail; - return 0; + return false; fail: fail_dump(attr); @@ -171,7 +169,7 @@ fail: if (atomic_read(&attr->times) != -1) atomic_dec_not_zero(&attr->times); - return 1; + return true; } #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS -- cgit v1.2.3 From 6b1b60f41eef3ba7b188fd72f1d6de478aafd93c Mon Sep 17 00:00:00 2001 From: Don Mullis Date: Fri, 8 Dec 2006 02:39:53 -0800 Subject: [PATCH] fault-injection: defaults likely to please a new user Assign defaults most likely to please a new user: 1) generate some logging output (verbose=2) 2) avoid injecting failures likely to lock up UI (ignore_gfp_wait=1, ignore_gfp_highmem=1) Signed-off-by: Don Mullis Cc: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fault-inject.h | 1 + mm/page_alloc.c | 2 ++ mm/slab.c | 1 + 3 files changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h index c77067916b7e..32368c4f0326 100644 --- a/include/linux/fault-inject.h +++ b/include/linux/fault-inject.h @@ -52,6 +52,7 @@ struct fault_attr { .times = ATOMIC_INIT(1), \ .require_end = ULONG_MAX, \ .stacktrace_depth = 32, \ + .verbose = 2, \ } #define DECLARE_FAULT_ATTR(name) struct fault_attr name = FAULT_ATTR_INITIALIZER diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0cc8b4376e91..e6b17b2989e0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -910,6 +910,8 @@ static struct fail_page_alloc_attr { } fail_page_alloc = { .attr = FAULT_ATTR_INITIALIZER, + .ignore_gfp_wait = 1, + .ignore_gfp_highmem = 1, }; static int __init setup_fail_page_alloc(char *str) diff --git a/mm/slab.c b/mm/slab.c index 47011e2ef3c9..56af694c9e6a 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -3102,6 +3102,7 @@ static struct failslab_attr { } failslab = { .attr = FAULT_ATTR_INITIALIZER, + .ignore_gfp_wait = 1, }; static int __init setup_failslab(char *str) -- cgit v1.2.3 From adf6b206546414fd006098d027e81f2b576ea2aa Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Fri, 8 Dec 2006 02:40:27 -0800 Subject: [PATCH] fbcmap.c: mark structs const or __read_mostly - Mark the default colormaps read-only, as nobody should be allowed to modify them - Additionally mark color values as __read_mostly since they will only be modified (very seldom) by fb_invert_cmaps() - Add named C99-initializers in fb_cmap structs and use the ARRAY_SIZE() macro Signed-off-by: Helge Deller Acked-by: James Simmons Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/fbcmap.c | 55 +++++++++++++++++++++++++------------------------- include/linux/fb.h | 6 +++--- 2 files changed, 31 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c index e8b135f3d80d..148108afdd51 100644 --- a/drivers/video/fbcmap.c +++ b/drivers/video/fbcmap.c @@ -18,63 +18,64 @@ #include -static u16 red2[] = { +static u16 red2[] __read_mostly = { 0x0000, 0xaaaa }; -static u16 green2[] = { +static u16 green2[] __read_mostly = { 0x0000, 0xaaaa }; -static u16 blue2[] = { +static u16 blue2[] __read_mostly = { 0x0000, 0xaaaa }; -static u16 red4[] = { +static u16 red4[] __read_mostly = { 0x0000, 0xaaaa, 0x5555, 0xffff }; -static u16 green4[] = { +static u16 green4[] __read_mostly = { 0x0000, 0xaaaa, 0x5555, 0xffff }; -static u16 blue4[] = { +static u16 blue4[] __read_mostly = { 0x0000, 0xaaaa, 0x5555, 0xffff }; -static u16 red8[] = { +static u16 red8[] __read_mostly = { 0x0000, 0x0000, 0x0000, 0x0000, 0xaaaa, 0xaaaa, 0xaaaa, 0xaaaa }; -static u16 green8[] = { +static u16 green8[] __read_mostly = { 0x0000, 0x0000, 0xaaaa, 0xaaaa, 0x0000, 0x0000, 0x5555, 0xaaaa }; -static u16 blue8[] = { +static u16 blue8[] __read_mostly = { 0x0000, 0xaaaa, 0x0000, 0xaaaa, 0x0000, 0xaaaa, 0x0000, 0xaaaa }; -static u16 red16[] = { +static u16 red16[] __read_mostly = { 0x0000, 0x0000, 0x0000, 0x0000, 0xaaaa, 0xaaaa, 0xaaaa, 0xaaaa, 0x5555, 0x5555, 0x5555, 0x5555, 0xffff, 0xffff, 0xffff, 0xffff }; -static u16 green16[] = { +static u16 green16[] __read_mostly = { 0x0000, 0x0000, 0xaaaa, 0xaaaa, 0x0000, 0x0000, 0x5555, 0xaaaa, 0x5555, 0x5555, 0xffff, 0xffff, 0x5555, 0x5555, 0xffff, 0xffff }; -static u16 blue16[] = { +static u16 blue16[] __read_mostly = { 0x0000, 0xaaaa, 0x0000, 0xaaaa, 0x0000, 0xaaaa, 0x0000, 0xaaaa, 0x5555, 0xffff, 0x5555, 0xffff, 0x5555, 0xffff, 0x5555, 0xffff }; -static struct fb_cmap default_2_colors = { - 0, 2, red2, green2, blue2, NULL +static const struct fb_cmap default_2_colors = { + .len=2, .red=red2, .green=green2, .blue=blue2 }; -static struct fb_cmap default_8_colors = { - 0, 8, red8, green8, blue8, NULL +static const struct fb_cmap default_8_colors = { + .len=8, .red=red8, .green=green8, .blue=blue8 }; -static struct fb_cmap default_4_colors = { - 0, 4, red4, green4, blue4, NULL +static const struct fb_cmap default_4_colors = { + .len=4, .red=red4, .green=green4, .blue=blue4 }; -static struct fb_cmap default_16_colors = { - 0, 16, red16, green16, blue16, NULL +static const struct fb_cmap default_16_colors = { + .len=16, .red=red16, .green=green16, .blue=blue16 }; + /** * fb_alloc_cmap - allocate a colormap * @cmap: frame buffer colormap structure @@ -146,7 +147,7 @@ void fb_dealloc_cmap(struct fb_cmap *cmap) * Copy contents of colormap from @from to @to. */ -int fb_copy_cmap(struct fb_cmap *from, struct fb_cmap *to) +int fb_copy_cmap(const struct fb_cmap *from, struct fb_cmap *to) { int tooff = 0, fromoff = 0; int size; @@ -170,7 +171,7 @@ int fb_copy_cmap(struct fb_cmap *from, struct fb_cmap *to) return 0; } -int fb_cmap_to_user(struct fb_cmap *from, struct fb_cmap_user *to) +int fb_cmap_to_user(const struct fb_cmap *from, struct fb_cmap_user *to) { int tooff = 0, fromoff = 0; int size; @@ -282,7 +283,7 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) * */ -struct fb_cmap *fb_default_cmap(int len) +const struct fb_cmap *fb_default_cmap(int len) { if (len <= 2) return &default_2_colors; @@ -305,22 +306,22 @@ void fb_invert_cmaps(void) { u_int i; - for (i = 0; i < 2; i++) { + for (i = 0; i < ARRAY_SIZE(red2); i++) { red2[i] = ~red2[i]; green2[i] = ~green2[i]; blue2[i] = ~blue2[i]; } - for (i = 0; i < 4; i++) { + for (i = 0; i < ARRAY_SIZE(red4); i++) { red4[i] = ~red4[i]; green4[i] = ~green4[i]; blue4[i] = ~blue4[i]; } - for (i = 0; i < 8; i++) { + for (i = 0; i < ARRAY_SIZE(red8); i++) { red8[i] = ~red8[i]; green8[i] = ~green8[i]; blue8[i] = ~blue8[i]; } - for (i = 0; i < 16; i++) { + for (i = 0; i < ARRAY_SIZE(red16); i++) { red16[i] = ~red16[i]; green16[i] = ~green16[i]; blue16[i] = ~blue16[i]; diff --git a/include/linux/fb.h b/include/linux/fb.h index fa23e0671bb3..6fe56aaa6685 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -970,11 +970,11 @@ extern struct fb_videomode *fb_find_best_display(struct fb_monspecs *specs, /* drivers/video/fbcmap.c */ extern int fb_alloc_cmap(struct fb_cmap *cmap, int len, int transp); extern void fb_dealloc_cmap(struct fb_cmap *cmap); -extern int fb_copy_cmap(struct fb_cmap *from, struct fb_cmap *to); -extern int fb_cmap_to_user(struct fb_cmap *from, struct fb_cmap_user *to); +extern int fb_copy_cmap(const struct fb_cmap *from, struct fb_cmap *to); +extern int fb_cmap_to_user(const struct fb_cmap *from, struct fb_cmap_user *to); extern int fb_set_cmap(struct fb_cmap *cmap, struct fb_info *fb_info); extern int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *fb_info); -extern struct fb_cmap *fb_default_cmap(int len); +extern const struct fb_cmap *fb_default_cmap(int len); extern void fb_invert_cmaps(void); struct fb_videomode { -- cgit v1.2.3 From a3d77d35be6f416a250c528c3ed5c70013a915e8 Mon Sep 17 00:00:00 2001 From: Kiyoshi Ueda Date: Fri, 8 Dec 2006 02:41:04 -0800 Subject: [PATCH] dm: suspend: parameter change Change the interface of dm_suspend() so that we can pass several options without increasing the number of parameters. The existing 'do_lockfs' integer parameter is replaced by a flag DM_SUSPEND_LOCKFS_FLAG. There is no functional change to the code. Test results: I have tested 'dmsetup suspend' command with/without the '--nolockfs' option and confirmed the do_lockfs value is correctly set. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Alasdair G Kergon Cc: dm-devel@redhat.com Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm-ioctl.c | 12 ++++++------ drivers/md/dm.c | 3 ++- drivers/md/dm.h | 5 +++++ include/linux/device-mapper.h | 2 +- 4 files changed, 14 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 4510ad8f971c..6d7a3d0c8f88 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -765,7 +765,7 @@ out: static int do_suspend(struct dm_ioctl *param) { int r = 0; - int do_lockfs = 1; + unsigned suspend_flags = DM_SUSPEND_LOCKFS_FLAG; struct mapped_device *md; md = find_device(param); @@ -773,10 +773,10 @@ static int do_suspend(struct dm_ioctl *param) return -ENXIO; if (param->flags & DM_SKIP_LOCKFS_FLAG) - do_lockfs = 0; + suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; if (!dm_suspended(md)) - r = dm_suspend(md, do_lockfs); + r = dm_suspend(md, suspend_flags); if (!r) r = __dev_status(md, param); @@ -788,7 +788,7 @@ static int do_suspend(struct dm_ioctl *param) static int do_resume(struct dm_ioctl *param) { int r = 0; - int do_lockfs = 1; + unsigned suspend_flags = DM_SUSPEND_LOCKFS_FLAG; struct hash_cell *hc; struct mapped_device *md; struct dm_table *new_map; @@ -814,9 +814,9 @@ static int do_resume(struct dm_ioctl *param) if (new_map) { /* Suspend if it isn't already suspended */ if (param->flags & DM_SKIP_LOCKFS_FLAG) - do_lockfs = 0; + suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; if (!dm_suspended(md)) - dm_suspend(md, do_lockfs); + dm_suspend(md, suspend_flags); r = dm_swap_table(md, new_map); if (r) { diff --git a/drivers/md/dm.c b/drivers/md/dm.c index dd50e30b6dcf..b42e71bb9e67 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1272,12 +1272,13 @@ static void unlock_fs(struct mapped_device *md) * dm_bind_table, dm_suspend must be called to flush any in * flight bios and ensure that any further io gets deferred. */ -int dm_suspend(struct mapped_device *md, int do_lockfs) +int dm_suspend(struct mapped_device *md, unsigned suspend_flags) { struct dm_table *map = NULL; DECLARE_WAITQUEUE(wait, current); struct bio *def; int r = -EINVAL; + int do_lockfs = suspend_flags & DM_SUSPEND_LOCKFS_FLAG ? 1 : 0; down(&md->suspend_lock); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index a48ec5e3c1f4..81c98d6e35e9 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -32,6 +32,11 @@ #define SECTOR_SHIFT 9 +/* + * Suspend feature flags + */ +#define DM_SUSPEND_LOCKFS_FLAG (1 << 0) + /* * List of devices that a metadevice uses and should open/close. */ diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 03ef41c1eaac..9194ff2d08c0 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -173,7 +173,7 @@ void *dm_get_mdptr(struct mapped_device *md); /* * A device can still be used while suspended, but I/O is deferred. */ -int dm_suspend(struct mapped_device *md, int with_lockfs); +int dm_suspend(struct mapped_device *md, unsigned suspend_flags); int dm_resume(struct mapped_device *md); /* -- cgit v1.2.3 From 45cbcd798354251b99694086af9d57c99e89bb43 Mon Sep 17 00:00:00 2001 From: Kiyoshi Ueda Date: Fri, 8 Dec 2006 02:41:05 -0800 Subject: [PATCH] dm: map and endio return code clarification Tighten the use of return values from the target map and end_io functions. Values of 2 and above are now explictly reserved for future use. There are no existing targets using such values. The patch has no effect on existing behaviour. o Reserve return values of 2 and above from target map functions. Any positive value currently indicates "mapping complete", but all existing drivers use the value 1. We now make that a requirement so we can assign new meaning to higher values in future. The new definition of return values from target map functions is: < 0 : error = 0 : The target will handle the io (DM_MAPIO_SUBMITTED). = 1 : Mapping completed (DM_MAPIO_REMAPPED). > 1 : Reserved (undefined). Previously this was the same as '= 1'. o Reserve return values of 2 and above from target end_io functions for similar reasons. DM_ENDIO_INCOMPLETE is introduced for a return value of 1. Test results: I have tested by using the multipath target. I/Os succeed when valid paths exist. I/Os are queued in the multipath target when there are no valid paths and queue_if_no_path is set. I/Os fail when there are no valid paths and queue_if_no_path is not set. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Alasdair G Kergon Cc: dm-devel@redhat.com Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm.c | 13 ++++++++++--- drivers/md/dm.h | 11 +++++++++++ include/linux/device-mapper.h | 2 +- 3 files changed, 22 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index b42e71bb9e67..d8544e1a4c1f 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -482,9 +482,13 @@ static int clone_endio(struct bio *bio, unsigned int done, int error) r = endio(tio->ti, bio, error, &tio->info); if (r < 0) error = r; - else if (r > 0) - /* the target wants another shot at the io */ + else if (r == DM_ENDIO_INCOMPLETE) + /* The target will handle the io */ return 1; + else if (r) { + DMWARN("unimplemented target endio return value: %d", r); + BUG(); + } } dec_pending(tio->io, error); @@ -542,7 +546,7 @@ static void __map_bio(struct dm_target *ti, struct bio *clone, atomic_inc(&tio->io->io_count); sector = clone->bi_sector; r = ti->type->map(ti, clone, &tio->info); - if (r > 0) { + if (r == DM_MAPIO_REMAPPED) { /* the bio has been remapped so dispatch it */ blk_add_trace_remap(bdev_get_queue(clone->bi_bdev), clone, @@ -560,6 +564,9 @@ static void __map_bio(struct dm_target *ti, struct bio *clone, clone->bi_private = md->bs; bio_put(clone); free_tio(md, tio); + } else if (r) { + DMWARN("unimplemented target map return value: %d", r); + BUG(); } } diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 81c98d6e35e9..293d5ce62a21 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -32,6 +32,17 @@ #define SECTOR_SHIFT 9 +/* + * Definitions of return values from target end_io function. + */ +#define DM_ENDIO_INCOMPLETE 1 + +/* + * Definitions of return values from target map function. + */ +#define DM_MAPIO_SUBMITTED 0 +#define DM_MAPIO_REMAPPED 1 + /* * Suspend feature flags */ diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 9194ff2d08c0..2e5c42346c38 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -39,7 +39,7 @@ typedef void (*dm_dtr_fn) (struct dm_target *ti); * The map function must return: * < 0: error * = 0: The target will handle the io by resubmitting it later - * > 0: simple remap complete + * = 1: simple remap complete */ typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio, union map_info *map_context); -- cgit v1.2.3 From 81fdb096dbcedcc3b94c7e47b59362b5214891e2 Mon Sep 17 00:00:00 2001 From: Kiyoshi Ueda Date: Fri, 8 Dec 2006 02:41:07 -0800 Subject: [PATCH] dm: ioctl: add noflush suspend Provide a dm ioctl option to request noflush suspending. (See next patch for what this is for.) As the interface is extended, the version number is incremented. Other than accepting the new option through the interface, There is no change to existing behaviour. Test results: Confirmed the option is given from user-space correctly. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Alasdair G Kergon Cc: dm-devel@redhat.com Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm-ioctl.c | 4 ++++ drivers/md/dm.h | 1 + include/linux/dm-ioctl.h | 9 +++++++-- 3 files changed, 12 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 6d7a3d0c8f88..cd6a184536a1 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -774,6 +774,8 @@ static int do_suspend(struct dm_ioctl *param) if (param->flags & DM_SKIP_LOCKFS_FLAG) suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; + if (param->flags & DM_NOFLUSH_FLAG) + suspend_flags |= DM_SUSPEND_NOFLUSH_FLAG; if (!dm_suspended(md)) r = dm_suspend(md, suspend_flags); @@ -815,6 +817,8 @@ static int do_resume(struct dm_ioctl *param) /* Suspend if it isn't already suspended */ if (param->flags & DM_SKIP_LOCKFS_FLAG) suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; + if (param->flags & DM_NOFLUSH_FLAG) + suspend_flags |= DM_SUSPEND_NOFLUSH_FLAG; if (!dm_suspended(md)) dm_suspend(md, suspend_flags); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 293d5ce62a21..c307ca9a4c33 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -47,6 +47,7 @@ * Suspend feature flags */ #define DM_SUSPEND_LOCKFS_FLAG (1 << 0) +#define DM_SUSPEND_NOFLUSH_FLAG (1 << 1) /* * List of devices that a metadevice uses and should open/close. diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h index 8853fc4d1c5e..b93486107821 100644 --- a/include/linux/dm-ioctl.h +++ b/include/linux/dm-ioctl.h @@ -285,9 +285,9 @@ typedef char ioctl_struct[308]; #define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl) #define DM_VERSION_MAJOR 4 -#define DM_VERSION_MINOR 10 +#define DM_VERSION_MINOR 11 #define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2006-09-14)" +#define DM_VERSION_EXTRA "-ioctl (2006-10-12)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ @@ -323,4 +323,9 @@ typedef char ioctl_struct[308]; */ #define DM_SKIP_LOCKFS_FLAG (1 << 10) /* In */ +/* + * Set this to suspend without flushing queued ios. + */ +#define DM_NOFLUSH_FLAG (1 << 11) /* In */ + #endif /* _LINUX_DM_IOCTL_H */ -- cgit v1.2.3 From 2e93ccc1933d08d32d9bde3784c3823e67b9b030 Mon Sep 17 00:00:00 2001 From: Kiyoshi Ueda Date: Fri, 8 Dec 2006 02:41:09 -0800 Subject: [PATCH] dm: suspend: add noflush pushback In device-mapper I/O is sometimes queued within targets for later processing. For example the multipath target can be configured to store I/O when no paths are available instead of returning it -EIO. This patch allows the device-mapper core to instruct a target to transfer the contents of any such in-target queue back into the core. This frees up the resources used by the target so the core can replace that target with an alternative one and then resend the I/O to it. Without this patch the only way to change the target in such circumstances involves returning the I/O with an error back to the filesystem/application. In the multipath case, this patch will let us add new paths for existing I/O to try after all the existing paths have failed. DMF_NOFLUSH_SUSPENDING ---------------------- If the DM_NOFLUSH_FLAG ioctl option is specified at suspend time, the DMF_NOFLUSH_SUSPENDING flag is set in md->flags during dm_suspend(). It is always cleared before dm_suspend() returns. The flag must be visible while the target is flushing pending I/Os so it is set before presuspend where the flush starts and unset after the wait for md->pending where the flush ends. Target drivers can check this flag by calling dm_noflush_suspending(). DM_MAPIO_REQUEUE / DM_ENDIO_REQUEUE ----------------------------------- A target's map() function can now return DM_MAPIO_REQUEUE to request the device mapper core queue the bio. Similarly, a target's end_io() function can return DM_ENDIO_REQUEUE to request the same. This has been labelled 'pushback'. The __map_bio() and clone_endio() functions in the core treat these return values as errors and call dec_pending() to end the I/O. dec_pending ----------- dec_pending() saves the pushback request in struct dm_io->error. Once all the split clones have ended, dec_pending() will put the original bio on the md->pushback list. Note that this supercedes any I/O errors. It is possible for the suspend with DM_NOFLUSH_FLAG to be aborted while in progress (e.g. by user interrupt). dec_pending() checks for this and returns -EIO if it happened. pushdback list and pushback_lock -------------------------------- The bio is queued on md->pushback temporarily in dec_pending(), and after all pending I/Os return, md->pushback is merged into md->deferred in dm_suspend() for re-issuing at resume time. md->pushback_lock protects md->pushback. The lock should be held with irq disabled because dec_pending() can be called from interrupt context. Queueing bios to md->pushback in dec_pending() must be done atomically with the check for DMF_NOFLUSH_SUSPENDING flag. So md->pushback_lock is held when checking the flag. Otherwise dec_pending() may queue a bio to md->pushback after the interrupted dm_suspend() flushes md->pushback. Then the bio would be left in md->pushback. Flag setting in dm_suspend() can be done without md->pushback_lock because the flag is checked only after presuspend and the set value is already made visible via the target's presuspend function. The flag can be checked without md->pushback_lock (e.g. the first part of the dec_pending() or target drivers), because the flag is checked again with md->pushback_lock held when the bio is really queued to md->pushback as described above. So even if the flag is cleared after the lockless checkings, the bio isn't left in md->pushback but returned to applications with -EIO. Other notes on the current patch -------------------------------- - md->pushback is added to the struct mapped_device instead of using md->deferred directly because md->io_lock which protects md->deferred is rw_semaphore and can't be used in interrupt context like dec_pending(), and md->io_lock protects the DMF_BLOCK_IO flag of md->flags too. - Don't issue lock_fs() in dm_suspend() if the DM_NOFLUSH_FLAG ioctl option is specified, because I/Os generated by lock_fs() would be pushed back and never return if there were no valid devices. - If an error occurs in dm_suspend() after the DMF_NOFLUSH_SUSPENDING flag is set, md->pushback must be flushed because I/Os may be queued to the list already. (flush_and_out label in dm_suspend()) Test results ------------ I have tested using multipath target with the next patch. The following tests are for regression/compatibility: - I/Os succeed when valid paths exist; - I/Os fail when there are no valid paths and queue_if_no_path is not set; - I/Os are queued in the multipath target when there are no valid paths and queue_if_no_path is set; - The queued I/Os above fail when suspend is issued without the DM_NOFLUSH_FLAG ioctl option. I/Os spanning 2 multipath targets also fail. The following tests are for the normal code path of new pushback feature: - Queued I/Os in the multipath target are flushed from the target but don't return when suspend is issued with the DM_NOFLUSH_FLAG ioctl option; - The I/Os above are queued in the multipath target again when resume is issued without path recovery; - The I/Os above succeed when resume is issued after path recovery or table load; - Queued I/Os in the multipath target succeed when resume is issued with the DM_NOFLUSH_FLAG ioctl option after table load. I/Os spanning 2 multipath targets also succeed. The following tests are for the error paths of the new pushback feature: - When the bdget_disk() fails in dm_suspend(), the DMF_NOFLUSH_SUSPENDING flag is cleared and I/Os already queued to the pushback list are flushed properly. - When suspend with the DM_NOFLUSH_FLAG ioctl option is interrupted, o I/Os which had already been queued to the pushback list at the time don't return, and are re-issued at resume time; o I/Os which hadn't been returned at the time return with EIO. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Alasdair G Kergon Cc: dm-devel@redhat.com Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm-bio-list.h | 14 ++++++ drivers/md/dm.c | 105 ++++++++++++++++++++++++++++++++++++++---- drivers/md/dm.h | 2 + include/linux/device-mapper.h | 3 ++ 4 files changed, 114 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-bio-list.h b/drivers/md/dm-bio-list.h index bbf4615f0e30..da4349649f7f 100644 --- a/drivers/md/dm-bio-list.h +++ b/drivers/md/dm-bio-list.h @@ -44,6 +44,20 @@ static inline void bio_list_merge(struct bio_list *bl, struct bio_list *bl2) bl->tail = bl2->tail; } +static inline void bio_list_merge_head(struct bio_list *bl, + struct bio_list *bl2) +{ + if (!bl2->head) + return; + + if (bl->head) + bl2->tail->bi_next = bl->head; + else + bl->tail = bl2->tail; + + bl->head = bl2->head; +} + static inline struct bio *bio_list_pop(struct bio_list *bl) { struct bio *bio = bl->head; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index d8544e1a4c1f..fe7c56e10435 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -68,10 +68,12 @@ union map_info *dm_get_mapinfo(struct bio *bio) #define DMF_FROZEN 2 #define DMF_FREEING 3 #define DMF_DELETING 4 +#define DMF_NOFLUSH_SUSPENDING 5 struct mapped_device { struct rw_semaphore io_lock; struct semaphore suspend_lock; + spinlock_t pushback_lock; rwlock_t map_lock; atomic_t holders; atomic_t open_count; @@ -90,6 +92,7 @@ struct mapped_device { atomic_t pending; wait_queue_head_t wait; struct bio_list deferred; + struct bio_list pushback; /* * The current mapping. @@ -444,23 +447,50 @@ int dm_set_geometry(struct mapped_device *md, struct hd_geometry *geo) * you this clearly demarcated crap. *---------------------------------------------------------------*/ +static int __noflush_suspending(struct mapped_device *md) +{ + return test_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); +} + /* * Decrements the number of outstanding ios that a bio has been * cloned into, completing the original io if necc. */ static void dec_pending(struct dm_io *io, int error) { - if (error) + unsigned long flags; + + /* Push-back supersedes any I/O errors */ + if (error && !(io->error > 0 && __noflush_suspending(io->md))) io->error = error; if (atomic_dec_and_test(&io->io_count)) { + if (io->error == DM_ENDIO_REQUEUE) { + /* + * Target requested pushing back the I/O. + * This must be handled before the sleeper on + * suspend queue merges the pushback list. + */ + spin_lock_irqsave(&io->md->pushback_lock, flags); + if (__noflush_suspending(io->md)) + bio_list_add(&io->md->pushback, io->bio); + else + /* noflush suspend was interrupted. */ + io->error = -EIO; + spin_unlock_irqrestore(&io->md->pushback_lock, flags); + } + if (end_io_acct(io)) /* nudge anyone waiting on suspend queue */ wake_up(&io->md->wait); - blk_add_trace_bio(io->md->queue, io->bio, BLK_TA_COMPLETE); + if (io->error != DM_ENDIO_REQUEUE) { + blk_add_trace_bio(io->md->queue, io->bio, + BLK_TA_COMPLETE); + + bio_endio(io->bio, io->bio->bi_size, io->error); + } - bio_endio(io->bio, io->bio->bi_size, io->error); free_io(io->md, io); } } @@ -480,7 +510,11 @@ static int clone_endio(struct bio *bio, unsigned int done, int error) if (endio) { r = endio(tio->ti, bio, error, &tio->info); - if (r < 0) + if (r < 0 || r == DM_ENDIO_REQUEUE) + /* + * error and requeue request are handled + * in dec_pending(). + */ error = r; else if (r == DM_ENDIO_INCOMPLETE) /* The target will handle the io */ @@ -554,8 +588,8 @@ static void __map_bio(struct dm_target *ti, struct bio *clone, clone->bi_sector); generic_make_request(clone); - } else if (r < 0) { - /* error the io and bail out */ + } else if (r < 0 || r == DM_MAPIO_REQUEUE) { + /* error the io and bail out, or requeue it if needed */ md = tio->io->md; dec_pending(tio->io, r); /* @@ -952,6 +986,7 @@ static struct mapped_device *alloc_dev(int minor) memset(md, 0, sizeof(*md)); init_rwsem(&md->io_lock); init_MUTEX(&md->suspend_lock); + spin_lock_init(&md->pushback_lock); rwlock_init(&md->map_lock); atomic_set(&md->holders, 1); atomic_set(&md->open_count, 0); @@ -1282,10 +1317,12 @@ static void unlock_fs(struct mapped_device *md) int dm_suspend(struct mapped_device *md, unsigned suspend_flags) { struct dm_table *map = NULL; + unsigned long flags; DECLARE_WAITQUEUE(wait, current); struct bio *def; int r = -EINVAL; int do_lockfs = suspend_flags & DM_SUSPEND_LOCKFS_FLAG ? 1 : 0; + int noflush = suspend_flags & DM_SUSPEND_NOFLUSH_FLAG ? 1 : 0; down(&md->suspend_lock); @@ -1294,6 +1331,13 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) map = dm_get_table(md); + /* + * DMF_NOFLUSH_SUSPENDING must be set before presuspend. + * This flag is cleared before dm_suspend returns. + */ + if (noflush) + set_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); + /* This does not get reverted if there's an error later. */ dm_table_presuspend_targets(map); @@ -1301,11 +1345,14 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) if (!md->suspended_bdev) { DMWARN("bdget failed in dm_suspend"); r = -ENOMEM; - goto out; + goto flush_and_out; } - /* Flush I/O to the device. */ - if (do_lockfs) { + /* + * Flush I/O to the device. + * noflush supersedes do_lockfs, because lock_fs() needs to flush I/Os. + */ + if (do_lockfs && !noflush) { r = lock_fs(md); if (r) goto out; @@ -1341,6 +1388,14 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) down_write(&md->io_lock); remove_wait_queue(&md->wait, &wait); + if (noflush) { + spin_lock_irqsave(&md->pushback_lock, flags); + clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); + bio_list_merge_head(&md->deferred, &md->pushback); + bio_list_init(&md->pushback); + spin_unlock_irqrestore(&md->pushback_lock, flags); + } + /* were we interrupted ? */ r = -EINTR; if (atomic_read(&md->pending)) { @@ -1349,7 +1404,7 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) __flush_deferred_io(md, def); up_write(&md->io_lock); unlock_fs(md); - goto out; + goto out; /* pushback list is already flushed, so skip flush */ } up_write(&md->io_lock); @@ -1359,6 +1414,25 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) r = 0; +flush_and_out: + if (r && noflush) { + /* + * Because there may be already I/Os in the pushback list, + * flush them before return. + */ + down_write(&md->io_lock); + + spin_lock_irqsave(&md->pushback_lock, flags); + clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); + bio_list_merge_head(&md->deferred, &md->pushback); + bio_list_init(&md->pushback); + spin_unlock_irqrestore(&md->pushback_lock, flags); + + def = bio_list_get(&md->deferred); + __flush_deferred_io(md, def); + up_write(&md->io_lock); + } + out: if (r && md->suspended_bdev) { bdput(md->suspended_bdev); @@ -1445,6 +1519,17 @@ int dm_suspended(struct mapped_device *md) return test_bit(DMF_SUSPENDED, &md->flags); } +int dm_noflush_suspending(struct dm_target *ti) +{ + struct mapped_device *md = dm_table_get_md(ti->table); + int r = __noflush_suspending(md); + + dm_put(md); + + return r; +} +EXPORT_SYMBOL_GPL(dm_noflush_suspending); + static struct block_device_operations dm_blk_dops = { .open = dm_blk_open, .release = dm_blk_close, diff --git a/drivers/md/dm.h b/drivers/md/dm.h index c307ca9a4c33..2f796b1436b2 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -36,12 +36,14 @@ * Definitions of return values from target end_io function. */ #define DM_ENDIO_INCOMPLETE 1 +#define DM_ENDIO_REQUEUE 2 /* * Definitions of return values from target map function. */ #define DM_MAPIO_SUBMITTED 0 #define DM_MAPIO_REMAPPED 1 +#define DM_MAPIO_REQUEUE DM_ENDIO_REQUEUE /* * Suspend feature flags diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 2e5c42346c38..499f5373e213 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -40,6 +40,7 @@ typedef void (*dm_dtr_fn) (struct dm_target *ti); * < 0: error * = 0: The target will handle the io by resubmitting it later * = 1: simple remap complete + * = 2: The target wants to push back the io */ typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio, union map_info *map_context); @@ -50,6 +51,7 @@ typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio, * 0 : ended successfully * 1 : for some reason the io has still not completed (eg, * multipath target might want to requeue a failed io). + * 2 : The target wants to push back the io */ typedef int (*dm_endio_fn) (struct dm_target *ti, struct bio *bio, int error, @@ -188,6 +190,7 @@ int dm_wait_event(struct mapped_device *md, int event_nr); const char *dm_device_name(struct mapped_device *md); struct gendisk *dm_disk(struct mapped_device *md); int dm_suspended(struct mapped_device *md); +int dm_noflush_suspending(struct dm_target *ti); /* * Geometry functions. -- cgit v1.2.3 From dde5845a529ff753364a6d1aea61180946270bfa Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:40:44 +0100 Subject: [PATCH] Generic HID layer - code split The "big main" split of USB HID code into generic HID code and USB-transport specific HID handling. Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-core.c | 940 +++++++++++++++++++++++++++++++++++++ drivers/hid/hid-input.c | 840 +++++++++++++++++++++++++++++++++ drivers/usb/input/hid-core.c | 1039 ++++------------------------------------- drivers/usb/input/hid-debug.h | 757 ------------------------------ drivers/usb/input/hid-ff.c | 2 +- drivers/usb/input/hid-input.c | 844 +-------------------------------- drivers/usb/input/hid-lgff.c | 2 +- drivers/usb/input/hid-pidff.c | 4 +- drivers/usb/input/hid-tmff.c | 2 +- drivers/usb/input/hid-zpff.c | 2 +- drivers/usb/input/hid.h | 540 --------------------- drivers/usb/input/hiddev.c | 3 +- include/linux/hid-debug.h | 757 ++++++++++++++++++++++++++++++ include/linux/hid.h | 540 +++++++++++++++++++++ 14 files changed, 3183 insertions(+), 3089 deletions(-) create mode 100644 drivers/hid/hid-core.c create mode 100644 drivers/hid/hid-input.c delete mode 100644 drivers/usb/input/hid-debug.h delete mode 100644 drivers/usb/input/hid.h create mode 100644 include/linux/hid-debug.h create mode 100644 include/linux/hid.h (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c new file mode 100644 index 000000000000..689ae16adf33 --- /dev/null +++ b/drivers/hid/hid-core.c @@ -0,0 +1,940 @@ +/* + * USB HID support for Linux + * + * Copyright (c) 1999 Andreas Gal + * Copyright (c) 2000-2005 Vojtech Pavlik + * Copyright (c) 2005 Michael Haboustak for Concept2, Inc + * Copyright (c) 2006 Jiri Kosina + */ + +/* + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#undef DEBUG +#undef DEBUG_DATA + +#include + +#include +#include + +/* + * Version Information + */ + +#define DRIVER_VERSION "v2.6" +#define DRIVER_AUTHOR "Andreas Gal, Vojtech Pavlik" +#define DRIVER_DESC "USB HID core driver" +#define DRIVER_LICENSE "GPL" + +/* + * Module parameters. + */ + +static unsigned int hid_mousepoll_interval; +module_param_named(mousepoll, hid_mousepoll_interval, uint, 0644); +MODULE_PARM_DESC(mousepoll, "Polling interval of mice"); + +/* + * Register a new report for a device. + */ + +static struct hid_report *hid_register_report(struct hid_device *device, unsigned type, unsigned id) +{ + struct hid_report_enum *report_enum = device->report_enum + type; + struct hid_report *report; + + if (report_enum->report_id_hash[id]) + return report_enum->report_id_hash[id]; + + if (!(report = kzalloc(sizeof(struct hid_report), GFP_KERNEL))) + return NULL; + + if (id != 0) + report_enum->numbered = 1; + + report->id = id; + report->type = type; + report->size = 0; + report->device = device; + report_enum->report_id_hash[id] = report; + + list_add_tail(&report->list, &report_enum->report_list); + + return report; +} + +/* + * Register a new field for this report. + */ + +static struct hid_field *hid_register_field(struct hid_report *report, unsigned usages, unsigned values) +{ + struct hid_field *field; + + if (report->maxfield == HID_MAX_FIELDS) { + dbg("too many fields in report"); + return NULL; + } + + if (!(field = kzalloc(sizeof(struct hid_field) + usages * sizeof(struct hid_usage) + + values * sizeof(unsigned), GFP_KERNEL))) return NULL; + + field->index = report->maxfield++; + report->field[field->index] = field; + field->usage = (struct hid_usage *)(field + 1); + field->value = (unsigned *)(field->usage + usages); + field->report = report; + + return field; +} + +/* + * Open a collection. The type/usage is pushed on the stack. + */ + +static int open_collection(struct hid_parser *parser, unsigned type) +{ + struct hid_collection *collection; + unsigned usage; + + usage = parser->local.usage[0]; + + if (parser->collection_stack_ptr == HID_COLLECTION_STACK_SIZE) { + dbg("collection stack overflow"); + return -1; + } + + if (parser->device->maxcollection == parser->device->collection_size) { + collection = kmalloc(sizeof(struct hid_collection) * + parser->device->collection_size * 2, GFP_KERNEL); + if (collection == NULL) { + dbg("failed to reallocate collection array"); + return -1; + } + memcpy(collection, parser->device->collection, + sizeof(struct hid_collection) * + parser->device->collection_size); + memset(collection + parser->device->collection_size, 0, + sizeof(struct hid_collection) * + parser->device->collection_size); + kfree(parser->device->collection); + parser->device->collection = collection; + parser->device->collection_size *= 2; + } + + parser->collection_stack[parser->collection_stack_ptr++] = + parser->device->maxcollection; + + collection = parser->device->collection + + parser->device->maxcollection++; + collection->type = type; + collection->usage = usage; + collection->level = parser->collection_stack_ptr - 1; + + if (type == HID_COLLECTION_APPLICATION) + parser->device->maxapplication++; + + return 0; +} + +/* + * Close a collection. + */ + +static int close_collection(struct hid_parser *parser) +{ + if (!parser->collection_stack_ptr) { + dbg("collection stack underflow"); + return -1; + } + parser->collection_stack_ptr--; + return 0; +} + +/* + * Climb up the stack, search for the specified collection type + * and return the usage. + */ + +static unsigned hid_lookup_collection(struct hid_parser *parser, unsigned type) +{ + int n; + for (n = parser->collection_stack_ptr - 1; n >= 0; n--) + if (parser->device->collection[parser->collection_stack[n]].type == type) + return parser->device->collection[parser->collection_stack[n]].usage; + return 0; /* we know nothing about this usage type */ +} + +/* + * Add a usage to the temporary parser table. + */ + +static int hid_add_usage(struct hid_parser *parser, unsigned usage) +{ + if (parser->local.usage_index >= HID_MAX_USAGES) { + dbg("usage index exceeded"); + return -1; + } + parser->local.usage[parser->local.usage_index] = usage; + parser->local.collection_index[parser->local.usage_index] = + parser->collection_stack_ptr ? + parser->collection_stack[parser->collection_stack_ptr - 1] : 0; + parser->local.usage_index++; + return 0; +} + +/* + * Register a new field for this report. + */ + +static int hid_add_field(struct hid_parser *parser, unsigned report_type, unsigned flags) +{ + struct hid_report *report; + struct hid_field *field; + int usages; + unsigned offset; + int i; + + if (!(report = hid_register_report(parser->device, report_type, parser->global.report_id))) { + dbg("hid_register_report failed"); + return -1; + } + + if (parser->global.logical_maximum < parser->global.logical_minimum) { + dbg("logical range invalid %d %d", parser->global.logical_minimum, parser->global.logical_maximum); + return -1; + } + + offset = report->size; + report->size += parser->global.report_size * parser->global.report_count; + + if (!parser->local.usage_index) /* Ignore padding fields */ + return 0; + + usages = max_t(int, parser->local.usage_index, parser->global.report_count); + + if ((field = hid_register_field(report, usages, parser->global.report_count)) == NULL) + return 0; + + field->physical = hid_lookup_collection(parser, HID_COLLECTION_PHYSICAL); + field->logical = hid_lookup_collection(parser, HID_COLLECTION_LOGICAL); + field->application = hid_lookup_collection(parser, HID_COLLECTION_APPLICATION); + + for (i = 0; i < usages; i++) { + int j = i; + /* Duplicate the last usage we parsed if we have excess values */ + if (i >= parser->local.usage_index) + j = parser->local.usage_index - 1; + field->usage[i].hid = parser->local.usage[j]; + field->usage[i].collection_index = + parser->local.collection_index[j]; + } + + field->maxusage = usages; + field->flags = flags; + field->report_offset = offset; + field->report_type = report_type; + field->report_size = parser->global.report_size; + field->report_count = parser->global.report_count; + field->logical_minimum = parser->global.logical_minimum; + field->logical_maximum = parser->global.logical_maximum; + field->physical_minimum = parser->global.physical_minimum; + field->physical_maximum = parser->global.physical_maximum; + field->unit_exponent = parser->global.unit_exponent; + field->unit = parser->global.unit; + + return 0; +} + +/* + * Read data value from item. + */ + +static u32 item_udata(struct hid_item *item) +{ + switch (item->size) { + case 1: return item->data.u8; + case 2: return item->data.u16; + case 4: return item->data.u32; + } + return 0; +} + +static s32 item_sdata(struct hid_item *item) +{ + switch (item->size) { + case 1: return item->data.s8; + case 2: return item->data.s16; + case 4: return item->data.s32; + } + return 0; +} + +/* + * Process a global item. + */ + +static int hid_parser_global(struct hid_parser *parser, struct hid_item *item) +{ + switch (item->tag) { + + case HID_GLOBAL_ITEM_TAG_PUSH: + + if (parser->global_stack_ptr == HID_GLOBAL_STACK_SIZE) { + dbg("global enviroment stack overflow"); + return -1; + } + + memcpy(parser->global_stack + parser->global_stack_ptr++, + &parser->global, sizeof(struct hid_global)); + return 0; + + case HID_GLOBAL_ITEM_TAG_POP: + + if (!parser->global_stack_ptr) { + dbg("global enviroment stack underflow"); + return -1; + } + + memcpy(&parser->global, parser->global_stack + --parser->global_stack_ptr, + sizeof(struct hid_global)); + return 0; + + case HID_GLOBAL_ITEM_TAG_USAGE_PAGE: + parser->global.usage_page = item_udata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_LOGICAL_MINIMUM: + parser->global.logical_minimum = item_sdata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_LOGICAL_MAXIMUM: + if (parser->global.logical_minimum < 0) + parser->global.logical_maximum = item_sdata(item); + else + parser->global.logical_maximum = item_udata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_PHYSICAL_MINIMUM: + parser->global.physical_minimum = item_sdata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_PHYSICAL_MAXIMUM: + if (parser->global.physical_minimum < 0) + parser->global.physical_maximum = item_sdata(item); + else + parser->global.physical_maximum = item_udata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_UNIT_EXPONENT: + parser->global.unit_exponent = item_sdata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_UNIT: + parser->global.unit = item_udata(item); + return 0; + + case HID_GLOBAL_ITEM_TAG_REPORT_SIZE: + if ((parser->global.report_size = item_udata(item)) > 32) { + dbg("invalid report_size %d", parser->global.report_size); + return -1; + } + return 0; + + case HID_GLOBAL_ITEM_TAG_REPORT_COUNT: + if ((parser->global.report_count = item_udata(item)) > HID_MAX_USAGES) { + dbg("invalid report_count %d", parser->global.report_count); + return -1; + } + return 0; + + case HID_GLOBAL_ITEM_TAG_REPORT_ID: + if ((parser->global.report_id = item_udata(item)) == 0) { + dbg("report_id 0 is invalid"); + return -1; + } + return 0; + + default: + dbg("unknown global tag 0x%x", item->tag); + return -1; + } +} + +/* + * Process a local item. + */ + +static int hid_parser_local(struct hid_parser *parser, struct hid_item *item) +{ + __u32 data; + unsigned n; + + if (item->size == 0) { + dbg("item data expected for local item"); + return -1; + } + + data = item_udata(item); + + switch (item->tag) { + + case HID_LOCAL_ITEM_TAG_DELIMITER: + + if (data) { + /* + * We treat items before the first delimiter + * as global to all usage sets (branch 0). + * In the moment we process only these global + * items and the first delimiter set. + */ + if (parser->local.delimiter_depth != 0) { + dbg("nested delimiters"); + return -1; + } + parser->local.delimiter_depth++; + parser->local.delimiter_branch++; + } else { + if (parser->local.delimiter_depth < 1) { + dbg("bogus close delimiter"); + return -1; + } + parser->local.delimiter_depth--; + } + return 1; + + case HID_LOCAL_ITEM_TAG_USAGE: + + if (parser->local.delimiter_branch > 1) { + dbg("alternative usage ignored"); + return 0; + } + + if (item->size <= 2) + data = (parser->global.usage_page << 16) + data; + + return hid_add_usage(parser, data); + + case HID_LOCAL_ITEM_TAG_USAGE_MINIMUM: + + if (parser->local.delimiter_branch > 1) { + dbg("alternative usage ignored"); + return 0; + } + + if (item->size <= 2) + data = (parser->global.usage_page << 16) + data; + + parser->local.usage_minimum = data; + return 0; + + case HID_LOCAL_ITEM_TAG_USAGE_MAXIMUM: + + if (parser->local.delimiter_branch > 1) { + dbg("alternative usage ignored"); + return 0; + } + + if (item->size <= 2) + data = (parser->global.usage_page << 16) + data; + + for (n = parser->local.usage_minimum; n <= data; n++) + if (hid_add_usage(parser, n)) { + dbg("hid_add_usage failed\n"); + return -1; + } + return 0; + + default: + + dbg("unknown local item tag 0x%x", item->tag); + return 0; + } + return 0; +} + +/* + * Process a main item. + */ + +static int hid_parser_main(struct hid_parser *parser, struct hid_item *item) +{ + __u32 data; + int ret; + + data = item_udata(item); + + switch (item->tag) { + case HID_MAIN_ITEM_TAG_BEGIN_COLLECTION: + ret = open_collection(parser, data & 0xff); + break; + case HID_MAIN_ITEM_TAG_END_COLLECTION: + ret = close_collection(parser); + break; + case HID_MAIN_ITEM_TAG_INPUT: + ret = hid_add_field(parser, HID_INPUT_REPORT, data); + break; + case HID_MAIN_ITEM_TAG_OUTPUT: + ret = hid_add_field(parser, HID_OUTPUT_REPORT, data); + break; + case HID_MAIN_ITEM_TAG_FEATURE: + ret = hid_add_field(parser, HID_FEATURE_REPORT, data); + break; + default: + dbg("unknown main item tag 0x%x", item->tag); + ret = 0; + } + + memset(&parser->local, 0, sizeof(parser->local)); /* Reset the local parser environment */ + + return ret; +} + +/* + * Process a reserved item. + */ + +static int hid_parser_reserved(struct hid_parser *parser, struct hid_item *item) +{ + dbg("reserved item type, tag 0x%x", item->tag); + return 0; +} + +/* + * Free a report and all registered fields. The field->usage and + * field->value table's are allocated behind the field, so we need + * only to free(field) itself. + */ + +static void hid_free_report(struct hid_report *report) +{ + unsigned n; + + for (n = 0; n < report->maxfield; n++) + kfree(report->field[n]); + kfree(report); +} + +/* + * Free a device structure, all reports, and all fields. + */ + +static void hid_free_device(struct hid_device *device) +{ + unsigned i,j; + + for (i = 0; i < HID_REPORT_TYPES; i++) { + struct hid_report_enum *report_enum = device->report_enum + i; + + for (j = 0; j < 256; j++) { + struct hid_report *report = report_enum->report_id_hash[j]; + if (report) + hid_free_report(report); + } + } + + kfree(device->rdesc); + kfree(device); +} + +/* + * Fetch a report description item from the data stream. We support long + * items, though they are not used yet. + */ + +static u8 *fetch_item(__u8 *start, __u8 *end, struct hid_item *item) +{ + u8 b; + + if ((end - start) <= 0) + return NULL; + + b = *start++; + + item->type = (b >> 2) & 3; + item->tag = (b >> 4) & 15; + + if (item->tag == HID_ITEM_TAG_LONG) { + + item->format = HID_ITEM_FORMAT_LONG; + + if ((end - start) < 2) + return NULL; + + item->size = *start++; + item->tag = *start++; + + if ((end - start) < item->size) + return NULL; + + item->data.longdata = start; + start += item->size; + return start; + } + + item->format = HID_ITEM_FORMAT_SHORT; + item->size = b & 3; + + switch (item->size) { + + case 0: + return start; + + case 1: + if ((end - start) < 1) + return NULL; + item->data.u8 = *start++; + return start; + + case 2: + if ((end - start) < 2) + return NULL; + item->data.u16 = le16_to_cpu(get_unaligned((__le16*)start)); + start = (__u8 *)((__le16 *)start + 1); + return start; + + case 3: + item->size++; + if ((end - start) < 4) + return NULL; + item->data.u32 = le32_to_cpu(get_unaligned((__le32*)start)); + start = (__u8 *)((__le32 *)start + 1); + return start; + } + + return NULL; +} + +/* + * Parse a report description into a hid_device structure. Reports are + * enumerated, fields are attached to these reports. + */ + +static struct hid_device *hid_parse_report(__u8 *start, unsigned size) +{ + struct hid_device *device; + struct hid_parser *parser; + struct hid_item item; + __u8 *end; + unsigned i; + static int (*dispatch_type[])(struct hid_parser *parser, + struct hid_item *item) = { + hid_parser_main, + hid_parser_global, + hid_parser_local, + hid_parser_reserved + }; + + if (!(device = kzalloc(sizeof(struct hid_device), GFP_KERNEL))) + return NULL; + + if (!(device->collection = kzalloc(sizeof(struct hid_collection) * + HID_DEFAULT_NUM_COLLECTIONS, GFP_KERNEL))) { + kfree(device); + return NULL; + } + device->collection_size = HID_DEFAULT_NUM_COLLECTIONS; + + for (i = 0; i < HID_REPORT_TYPES; i++) + INIT_LIST_HEAD(&device->report_enum[i].report_list); + + if (!(device->rdesc = (__u8 *)kmalloc(size, GFP_KERNEL))) { + kfree(device->collection); + kfree(device); + return NULL; + } + memcpy(device->rdesc, start, size); + device->rsize = size; + + if (!(parser = kzalloc(sizeof(struct hid_parser), GFP_KERNEL))) { + kfree(device->rdesc); + kfree(device->collection); + kfree(device); + return NULL; + } + parser->device = device; + + end = start + size; + while ((start = fetch_item(start, end, &item)) != NULL) { + + if (item.format != HID_ITEM_FORMAT_SHORT) { + dbg("unexpected long global item"); + kfree(device->collection); + hid_free_device(device); + kfree(parser); + return NULL; + } + + if (dispatch_type[item.type](parser, &item)) { + dbg("item %u %u %u %u parsing failed\n", + item.format, (unsigned)item.size, (unsigned)item.type, (unsigned)item.tag); + kfree(device->collection); + hid_free_device(device); + kfree(parser); + return NULL; + } + + if (start == end) { + if (parser->collection_stack_ptr) { + dbg("unbalanced collection at end of report description"); + kfree(device->collection); + hid_free_device(device); + kfree(parser); + return NULL; + } + if (parser->local.delimiter_depth) { + dbg("unbalanced delimiter at end of report description"); + kfree(device->collection); + hid_free_device(device); + kfree(parser); + return NULL; + } + kfree(parser); + return device; + } + } + + dbg("item fetching failed at offset %d\n", (int)(end - start)); + kfree(device->collection); + hid_free_device(device); + kfree(parser); + return NULL; +} + +/* + * Convert a signed n-bit integer to signed 32-bit integer. Common + * cases are done through the compiler, the screwed things has to be + * done by hand. + */ + +static s32 snto32(__u32 value, unsigned n) +{ + switch (n) { + case 8: return ((__s8)value); + case 16: return ((__s16)value); + case 32: return ((__s32)value); + } + return value & (1 << (n - 1)) ? value | (-1 << n) : value; +} + +/* + * Convert a signed 32-bit integer to a signed n-bit integer. + */ + +static u32 s32ton(__s32 value, unsigned n) +{ + s32 a = value >> (n - 1); + if (a && a != -1) + return value < 0 ? 1 << (n - 1) : (1 << (n - 1)) - 1; + return value & ((1 << n) - 1); +} + +/* + * Extract/implement a data field from/to a little endian report (bit array). + * + * Code sort-of follows HID spec: + * http://www.usb.org/developers/devclass_docs/HID1_11.pdf + * + * While the USB HID spec allows unlimited length bit fields in "report + * descriptors", most devices never use more than 16 bits. + * One model of UPS is claimed to report "LINEV" as a 32-bit field. + * Search linux-kernel and linux-usb-devel archives for "hid-core extract". + */ + +static __inline__ __u32 extract(__u8 *report, unsigned offset, unsigned n) +{ + u64 x; + + WARN_ON(n > 32); + + report += offset >> 3; /* adjust byte index */ + offset &= 7; /* now only need bit offset into one byte */ + x = get_unaligned((u64 *) report); + x = le64_to_cpu(x); + x = (x >> offset) & ((1ULL << n) - 1); /* extract bit field */ + return (u32) x; +} + +/* + * "implement" : set bits in a little endian bit stream. + * Same concepts as "extract" (see comments above). + * The data mangled in the bit stream remains in little endian + * order the whole time. It make more sense to talk about + * endianness of register values by considering a register + * a "cached" copy of the little endiad bit stream. + */ +static __inline__ void implement(__u8 *report, unsigned offset, unsigned n, __u32 value) +{ + u64 x; + u64 m = (1ULL << n) - 1; + + WARN_ON(n > 32); + + WARN_ON(value > m); + value &= m; + + report += offset >> 3; + offset &= 7; + + x = get_unaligned((u64 *)report); + x &= cpu_to_le64(~(m << offset)); + x |= cpu_to_le64(((u64) value) << offset); + put_unaligned(x, (u64 *) report); +} + +/* + * Search an array for a value. + */ + +static __inline__ int search(__s32 *array, __s32 value, unsigned n) +{ + while (n--) { + if (*array++ == value) + return 0; + } + return -1; +} + +static void hid_process_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value, int interrupt) +{ + hid_dump_input(usage, value); + if (hid->claimed & HID_CLAIMED_INPUT) + hidinput_hid_event(hid, field, usage, value); + if (hid->claimed & HID_CLAIMED_HIDDEV && interrupt) + hiddev_hid_event(hid, field, usage, value); +} + +/* + * Analyse a received field, and fetch the data from it. The field + * content is stored for next report processing (we do differential + * reporting to the layer). + */ + +static void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt) +{ + unsigned n; + unsigned count = field->report_count; + unsigned offset = field->report_offset; + unsigned size = field->report_size; + __s32 min = field->logical_minimum; + __s32 max = field->logical_maximum; + __s32 *value; + + if (!(value = kmalloc(sizeof(__s32) * count, GFP_ATOMIC))) + return; + + for (n = 0; n < count; n++) { + + value[n] = min < 0 ? snto32(extract(data, offset + n * size, size), size) : + extract(data, offset + n * size, size); + + if (!(field->flags & HID_MAIN_ITEM_VARIABLE) /* Ignore report if ErrorRollOver */ + && value[n] >= min && value[n] <= max + && field->usage[value[n] - min].hid == HID_UP_KEYBOARD + 1) + goto exit; + } + + for (n = 0; n < count; n++) { + + if (HID_MAIN_ITEM_VARIABLE & field->flags) { + hid_process_event(hid, field, &field->usage[n], value[n], interrupt); + continue; + } + + if (field->value[n] >= min && field->value[n] <= max + && field->usage[field->value[n] - min].hid + && search(value, field->value[n], count)) + hid_process_event(hid, field, &field->usage[field->value[n] - min], 0, interrupt); + + if (value[n] >= min && value[n] <= max + && field->usage[value[n] - min].hid + && search(field->value, value[n], count)) + hid_process_event(hid, field, &field->usage[value[n] - min], 1, interrupt); + } + + memcpy(field->value, value, count * sizeof(__s32)); +exit: + kfree(value); +} + + +/* + * Output the field into the report. + */ + +static void hid_output_field(struct hid_field *field, __u8 *data) +{ + unsigned count = field->report_count; + unsigned offset = field->report_offset; + unsigned size = field->report_size; + unsigned n; + + for (n = 0; n < count; n++) { + if (field->logical_minimum < 0) /* signed values */ + implement(data, offset + n * size, size, s32ton(field->value[n], size)); + else /* unsigned values */ + implement(data, offset + n * size, size, field->value[n]); + } +} + +/* + * Create a report. + */ + +static void hid_output_report(struct hid_report *report, __u8 *data) +{ + unsigned n; + + if (report->id > 0) + *data++ = report->id; + + for (n = 0; n < report->maxfield; n++) + hid_output_field(report->field[n], data); +} + +/* + * Set a field value. The report this field belongs to has to be + * created and transferred to the device, to set this value in the + * device. + */ + +int hid_set_field(struct hid_field *field, unsigned offset, __s32 value) +{ + unsigned size = field->report_size; + + hid_dump_input(field->usage + offset, value); + + if (offset >= field->report_count) { + dbg("offset (%d) exceeds report_count (%d)", offset, field->report_count); + hid_dump_field(field, 8); + return -1; + } + if (field->logical_minimum < 0) { + if (value != snto32(s32ton(value, size), size)) { + dbg("value %d is out of range", value); + return -1; + } + } + field->value[offset] = value; + return 0; +} + diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c new file mode 100644 index 000000000000..d459005062e0 --- /dev/null +++ b/drivers/hid/hid-input.c @@ -0,0 +1,840 @@ +/* + * $Id: hid-input.c,v 1.2 2002/04/23 00:59:25 rdamazio Exp $ + * + * Copyright (c) 2000-2001 Vojtech Pavlik + * Copyright (c) 2006 Jiri Kosina + * + * HID to Linux Input mapping + */ + +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Should you need to contact me, the author, you can do so either by + * e-mail - mail your message to , or by paper mail: + * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic + */ + +#include +#include +#include +#include + +#undef DEBUG + +#include + +#define unk KEY_UNKNOWN + +static const unsigned char hid_keyboard[256] = { + 0, 0, 0, 0, 30, 48, 46, 32, 18, 33, 34, 35, 23, 36, 37, 38, + 50, 49, 24, 25, 16, 19, 31, 20, 22, 47, 17, 45, 21, 44, 2, 3, + 4, 5, 6, 7, 8, 9, 10, 11, 28, 1, 14, 15, 57, 12, 13, 26, + 27, 43, 43, 39, 40, 41, 51, 52, 53, 58, 59, 60, 61, 62, 63, 64, + 65, 66, 67, 68, 87, 88, 99, 70,119,110,102,104,111,107,109,106, + 105,108,103, 69, 98, 55, 74, 78, 96, 79, 80, 81, 75, 76, 77, 71, + 72, 73, 82, 83, 86,127,116,117,183,184,185,186,187,188,189,190, + 191,192,193,194,134,138,130,132,128,129,131,137,133,135,136,113, + 115,114,unk,unk,unk,121,unk, 89, 93,124, 92, 94, 95,unk,unk,unk, + 122,123, 90, 91, 85,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, + unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, + unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, + unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, + unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, + 29, 42, 56,125, 97, 54,100,126,164,166,165,163,161,115,114,113, + 150,158,159,128,136,177,178,176,142,152,173,140,unk,unk,unk,unk +}; + +static const struct { + __s32 x; + __s32 y; +} hid_hat_to_axis[] = {{ 0, 0}, { 0,-1}, { 1,-1}, { 1, 0}, { 1, 1}, { 0, 1}, {-1, 1}, {-1, 0}, {-1,-1}}; + +#define map_abs(c) do { usage->code = c; usage->type = EV_ABS; bit = input->absbit; max = ABS_MAX; } while (0) +#define map_rel(c) do { usage->code = c; usage->type = EV_REL; bit = input->relbit; max = REL_MAX; } while (0) +#define map_key(c) do { usage->code = c; usage->type = EV_KEY; bit = input->keybit; max = KEY_MAX; } while (0) +#define map_led(c) do { usage->code = c; usage->type = EV_LED; bit = input->ledbit; max = LED_MAX; } while (0) + +#define map_abs_clear(c) do { map_abs(c); clear_bit(c, bit); } while (0) +#define map_key_clear(c) do { map_key(c); clear_bit(c, bit); } while (0) + +#ifdef CONFIG_USB_HIDINPUT_POWERBOOK + +struct hidinput_key_translation { + u16 from; + u16 to; + u8 flags; +}; + +#define POWERBOOK_FLAG_FKEY 0x01 + +static struct hidinput_key_translation powerbook_fn_keys[] = { + { KEY_BACKSPACE, KEY_DELETE }, + { KEY_F1, KEY_BRIGHTNESSDOWN, POWERBOOK_FLAG_FKEY }, + { KEY_F2, KEY_BRIGHTNESSUP, POWERBOOK_FLAG_FKEY }, + { KEY_F3, KEY_MUTE, POWERBOOK_FLAG_FKEY }, + { KEY_F4, KEY_VOLUMEDOWN, POWERBOOK_FLAG_FKEY }, + { KEY_F5, KEY_VOLUMEUP, POWERBOOK_FLAG_FKEY }, + { KEY_F6, KEY_NUMLOCK, POWERBOOK_FLAG_FKEY }, + { KEY_F7, KEY_SWITCHVIDEOMODE, POWERBOOK_FLAG_FKEY }, + { KEY_F8, KEY_KBDILLUMTOGGLE, POWERBOOK_FLAG_FKEY }, + { KEY_F9, KEY_KBDILLUMDOWN, POWERBOOK_FLAG_FKEY }, + { KEY_F10, KEY_KBDILLUMUP, POWERBOOK_FLAG_FKEY }, + { KEY_UP, KEY_PAGEUP }, + { KEY_DOWN, KEY_PAGEDOWN }, + { KEY_LEFT, KEY_HOME }, + { KEY_RIGHT, KEY_END }, + { } +}; + +static struct hidinput_key_translation powerbook_numlock_keys[] = { + { KEY_J, KEY_KP1 }, + { KEY_K, KEY_KP2 }, + { KEY_L, KEY_KP3 }, + { KEY_U, KEY_KP4 }, + { KEY_I, KEY_KP5 }, + { KEY_O, KEY_KP6 }, + { KEY_7, KEY_KP7 }, + { KEY_8, KEY_KP8 }, + { KEY_9, KEY_KP9 }, + { KEY_M, KEY_KP0 }, + { KEY_DOT, KEY_KPDOT }, + { KEY_SLASH, KEY_KPPLUS }, + { KEY_SEMICOLON, KEY_KPMINUS }, + { KEY_P, KEY_KPASTERISK }, + { KEY_MINUS, KEY_KPEQUAL }, + { KEY_0, KEY_KPSLASH }, + { KEY_F6, KEY_NUMLOCK }, + { KEY_KPENTER, KEY_KPENTER }, + { KEY_BACKSPACE, KEY_BACKSPACE }, + { } +}; + +static struct hidinput_key_translation powerbook_iso_keyboard[] = { + { KEY_GRAVE, KEY_102ND }, + { KEY_102ND, KEY_GRAVE }, + { } +}; + + +static int usbhid_pb_fnmode = 1; +module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); +MODULE_PARM_DESC(pb_fnmode, + "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); + +static struct hidinput_key_translation *find_translation(struct hidinput_key_translation *table, u16 from) +{ + struct hidinput_key_translation *trans; + + /* Look for the translation */ + for (trans = table; trans->from; trans++) + if (trans->from == from) + return trans; + + return NULL; +} + +static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, + struct hid_usage *usage, __s32 value) +{ + struct hidinput_key_translation *trans; + + if (usage->code == KEY_FN) { + if (value) hid->quirks |= HID_QUIRK_POWERBOOK_FN_ON; + else hid->quirks &= ~HID_QUIRK_POWERBOOK_FN_ON; + + input_event(input, usage->type, usage->code, value); + + return 1; + } + + if (usbhid_pb_fnmode) { + int do_translate; + + trans = find_translation(powerbook_fn_keys, usage->code); + if (trans) { + if (test_bit(usage->code, hid->pb_pressed_fn)) + do_translate = 1; + else if (trans->flags & POWERBOOK_FLAG_FKEY) + do_translate = + (usbhid_pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || + (usbhid_pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); + else + do_translate = (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON); + + if (do_translate) { + if (value) + set_bit(usage->code, hid->pb_pressed_fn); + else + clear_bit(usage->code, hid->pb_pressed_fn); + + input_event(input, usage->type, trans->to, value); + + return 1; + } + } + + if (test_bit(usage->code, hid->pb_pressed_numlock) || + test_bit(LED_NUML, input->led)) { + trans = find_translation(powerbook_numlock_keys, usage->code); + + if (trans) { + if (value) + set_bit(usage->code, hid->pb_pressed_numlock); + else + clear_bit(usage->code, hid->pb_pressed_numlock); + + input_event(input, usage->type, trans->to, value); + } + + return 1; + } + } + + if (hid->quirks & HID_QUIRK_POWERBOOK_ISO_KEYBOARD) { + trans = find_translation(powerbook_iso_keyboard, usage->code); + if (trans) { + input_event(input, usage->type, trans->to, value); + return 1; + } + } + + return 0; +} + +static void hidinput_pb_setup(struct input_dev *input) +{ + struct hidinput_key_translation *trans; + + set_bit(KEY_NUMLOCK, input->keybit); + + /* Enable all needed keys */ + for (trans = powerbook_fn_keys; trans->from; trans++) + set_bit(trans->to, input->keybit); + + for (trans = powerbook_numlock_keys; trans->from; trans++) + set_bit(trans->to, input->keybit); + + for (trans = powerbook_iso_keyboard; trans->from; trans++) + set_bit(trans->to, input->keybit); + +} +#else +static inline int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, + struct hid_usage *usage, __s32 value) +{ + return 0; +} + +static inline void hidinput_pb_setup(struct input_dev *input) +{ +} +#endif + +static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_field *field, + struct hid_usage *usage) +{ + struct input_dev *input = hidinput->input; + struct hid_device *device = input->private; + int max = 0, code; + unsigned long *bit = NULL; + + field->hidinput = hidinput; + +#ifdef DEBUG + printk(KERN_DEBUG "Mapping: "); + resolv_usage(usage->hid); + printk(" ---> "); +#endif + + if (field->flags & HID_MAIN_ITEM_CONSTANT) + goto ignore; + + switch (usage->hid & HID_USAGE_PAGE) { + + case HID_UP_UNDEFINED: + goto ignore; + + case HID_UP_KEYBOARD: + + set_bit(EV_REP, input->evbit); + + if ((usage->hid & HID_USAGE) < 256) { + if (!hid_keyboard[usage->hid & HID_USAGE]) goto ignore; + map_key_clear(hid_keyboard[usage->hid & HID_USAGE]); + } else + map_key(KEY_UNKNOWN); + + break; + + case HID_UP_BUTTON: + + code = ((usage->hid - 1) & 0xf); + + switch (field->application) { + case HID_GD_MOUSE: + case HID_GD_POINTER: code += 0x110; break; + case HID_GD_JOYSTICK: code += 0x120; break; + case HID_GD_GAMEPAD: code += 0x130; break; + default: + switch (field->physical) { + case HID_GD_MOUSE: + case HID_GD_POINTER: code += 0x110; break; + case HID_GD_JOYSTICK: code += 0x120; break; + case HID_GD_GAMEPAD: code += 0x130; break; + default: code += 0x100; + } + } + + map_key(code); + break; + + + case HID_UP_SIMULATION: + + switch (usage->hid & 0xffff) { + case 0xba: map_abs(ABS_RUDDER); break; + case 0xbb: map_abs(ABS_THROTTLE); break; + case 0xc4: map_abs(ABS_GAS); break; + case 0xc5: map_abs(ABS_BRAKE); break; + case 0xc8: map_abs(ABS_WHEEL); break; + default: goto ignore; + } + break; + + case HID_UP_GENDESK: + + if ((usage->hid & 0xf0) == 0x80) { /* SystemControl */ + switch (usage->hid & 0xf) { + case 0x1: map_key_clear(KEY_POWER); break; + case 0x2: map_key_clear(KEY_SLEEP); break; + case 0x3: map_key_clear(KEY_WAKEUP); break; + default: goto unknown; + } + break; + } + + if ((usage->hid & 0xf0) == 0x90) { /* D-pad */ + switch (usage->hid) { + case HID_GD_UP: usage->hat_dir = 1; break; + case HID_GD_DOWN: usage->hat_dir = 5; break; + case HID_GD_RIGHT: usage->hat_dir = 3; break; + case HID_GD_LEFT: usage->hat_dir = 7; break; + default: goto unknown; + } + if (field->dpad) { + map_abs(field->dpad); + goto ignore; + } + map_abs(ABS_HAT0X); + break; + } + + switch (usage->hid) { + + /* These usage IDs map directly to the usage codes. */ + case HID_GD_X: case HID_GD_Y: case HID_GD_Z: + case HID_GD_RX: case HID_GD_RY: case HID_GD_RZ: + case HID_GD_SLIDER: case HID_GD_DIAL: case HID_GD_WHEEL: + if (field->flags & HID_MAIN_ITEM_RELATIVE) + map_rel(usage->hid & 0xf); + else + map_abs(usage->hid & 0xf); + break; + + case HID_GD_HATSWITCH: + usage->hat_min = field->logical_minimum; + usage->hat_max = field->logical_maximum; + map_abs(ABS_HAT0X); + break; + + case HID_GD_START: map_key_clear(BTN_START); break; + case HID_GD_SELECT: map_key_clear(BTN_SELECT); break; + + default: goto unknown; + } + + break; + + case HID_UP_LED: + if (((usage->hid - 1) & 0xffff) >= LED_MAX) + goto ignore; + map_led((usage->hid - 1) & 0xffff); + break; + + case HID_UP_DIGITIZER: + + switch (usage->hid & 0xff) { + + case 0x30: /* TipPressure */ + if (!test_bit(BTN_TOUCH, input->keybit)) { + device->quirks |= HID_QUIRK_NOTOUCH; + set_bit(EV_KEY, input->evbit); + set_bit(BTN_TOUCH, input->keybit); + } + + map_abs_clear(ABS_PRESSURE); + break; + + case 0x32: /* InRange */ + switch (field->physical & 0xff) { + case 0x21: map_key(BTN_TOOL_MOUSE); break; + case 0x22: map_key(BTN_TOOL_FINGER); break; + default: map_key(BTN_TOOL_PEN); break; + } + break; + + case 0x3c: /* Invert */ + map_key_clear(BTN_TOOL_RUBBER); + break; + + case 0x33: /* Touch */ + case 0x42: /* TipSwitch */ + case 0x43: /* TipSwitch2 */ + device->quirks &= ~HID_QUIRK_NOTOUCH; + map_key_clear(BTN_TOUCH); + break; + + case 0x44: /* BarrelSwitch */ + map_key_clear(BTN_STYLUS); + break; + + default: goto unknown; + } + break; + + case HID_UP_CONSUMER: /* USB HUT v1.1, pages 56-62 */ + + switch (usage->hid & HID_USAGE) { + case 0x000: goto ignore; + case 0x034: map_key_clear(KEY_SLEEP); break; + case 0x036: map_key_clear(BTN_MISC); break; + case 0x045: map_key_clear(KEY_RADIO); break; + case 0x08a: map_key_clear(KEY_WWW); break; + case 0x08d: map_key_clear(KEY_PROGRAM); break; + case 0x095: map_key_clear(KEY_HELP); break; + case 0x09c: map_key_clear(KEY_CHANNELUP); break; + case 0x09d: map_key_clear(KEY_CHANNELDOWN); break; + case 0x0b0: map_key_clear(KEY_PLAY); break; + case 0x0b1: map_key_clear(KEY_PAUSE); break; + case 0x0b2: map_key_clear(KEY_RECORD); break; + case 0x0b3: map_key_clear(KEY_FASTFORWARD); break; + case 0x0b4: map_key_clear(KEY_REWIND); break; + case 0x0b5: map_key_clear(KEY_NEXTSONG); break; + case 0x0b6: map_key_clear(KEY_PREVIOUSSONG); break; + case 0x0b7: map_key_clear(KEY_STOPCD); break; + case 0x0b8: map_key_clear(KEY_EJECTCD); break; + case 0x0cd: map_key_clear(KEY_PLAYPAUSE); break; + case 0x0e0: map_abs_clear(ABS_VOLUME); break; + case 0x0e2: map_key_clear(KEY_MUTE); break; + case 0x0e5: map_key_clear(KEY_BASSBOOST); break; + case 0x0e9: map_key_clear(KEY_VOLUMEUP); break; + case 0x0ea: map_key_clear(KEY_VOLUMEDOWN); break; + case 0x183: map_key_clear(KEY_CONFIG); break; + case 0x18a: map_key_clear(KEY_MAIL); break; + case 0x192: map_key_clear(KEY_CALC); break; + case 0x194: map_key_clear(KEY_FILE); break; + case 0x1a7: map_key_clear(KEY_DOCUMENTS); break; + case 0x201: map_key_clear(KEY_NEW); break; + case 0x207: map_key_clear(KEY_SAVE); break; + case 0x208: map_key_clear(KEY_PRINT); break; + case 0x209: map_key_clear(KEY_PROPS); break; + case 0x21a: map_key_clear(KEY_UNDO); break; + case 0x21b: map_key_clear(KEY_COPY); break; + case 0x21c: map_key_clear(KEY_CUT); break; + case 0x21d: map_key_clear(KEY_PASTE); break; + case 0x221: map_key_clear(KEY_FIND); break; + case 0x223: map_key_clear(KEY_HOMEPAGE); break; + case 0x224: map_key_clear(KEY_BACK); break; + case 0x225: map_key_clear(KEY_FORWARD); break; + case 0x226: map_key_clear(KEY_STOP); break; + case 0x227: map_key_clear(KEY_REFRESH); break; + case 0x22a: map_key_clear(KEY_BOOKMARKS); break; + case 0x233: map_key_clear(KEY_SCROLLUP); break; + case 0x234: map_key_clear(KEY_SCROLLDOWN); break; + case 0x238: map_rel(REL_HWHEEL); break; + case 0x279: map_key_clear(KEY_REDO); break; + case 0x289: map_key_clear(KEY_REPLY); break; + case 0x28b: map_key_clear(KEY_FORWARDMAIL); break; + case 0x28c: map_key_clear(KEY_SEND); break; + + /* Reported on a Cherry Cymotion keyboard */ + case 0x301: map_key_clear(KEY_PROG1); break; + case 0x302: map_key_clear(KEY_PROG2); break; + case 0x303: map_key_clear(KEY_PROG3); break; + + default: goto ignore; + } + break; + + case HID_UP_HPVENDOR: /* Reported on a Dutch layout HP5308 */ + + set_bit(EV_REP, input->evbit); + switch (usage->hid & HID_USAGE) { + case 0x021: map_key_clear(KEY_PRINT); break; + case 0x070: map_key_clear(KEY_HP); break; + case 0x071: map_key_clear(KEY_CAMERA); break; + case 0x072: map_key_clear(KEY_SOUND); break; + case 0x073: map_key_clear(KEY_QUESTION); break; + case 0x080: map_key_clear(KEY_EMAIL); break; + case 0x081: map_key_clear(KEY_CHAT); break; + case 0x082: map_key_clear(KEY_SEARCH); break; + case 0x083: map_key_clear(KEY_CONNECT); break; + case 0x084: map_key_clear(KEY_FINANCE); break; + case 0x085: map_key_clear(KEY_SPORT); break; + case 0x086: map_key_clear(KEY_SHOP); break; + default: goto ignore; + } + break; + + case HID_UP_MSVENDOR: + goto ignore; + + case HID_UP_CUSTOM: /* Reported on Logitech and Powerbook USB keyboards */ + + set_bit(EV_REP, input->evbit); + switch(usage->hid & HID_USAGE) { + case 0x003: + /* The fn key on Apple PowerBooks */ + map_key_clear(KEY_FN); + hidinput_pb_setup(input); + break; + + default: goto ignore; + } + break; + + case HID_UP_LOGIVENDOR: /* Reported on Logitech Ultra X Media Remote */ + + set_bit(EV_REP, input->evbit); + switch(usage->hid & HID_USAGE) { + case 0x004: map_key_clear(KEY_AGAIN); break; + case 0x00d: map_key_clear(KEY_HOME); break; + case 0x024: map_key_clear(KEY_SHUFFLE); break; + case 0x025: map_key_clear(KEY_TV); break; + case 0x026: map_key_clear(KEY_MENU); break; + case 0x031: map_key_clear(KEY_AUDIO); break; + case 0x032: map_key_clear(KEY_TEXT); break; + case 0x033: map_key_clear(KEY_LAST); break; + case 0x047: map_key_clear(KEY_MP3); break; + case 0x048: map_key_clear(KEY_DVD); break; + case 0x049: map_key_clear(KEY_MEDIA); break; + case 0x04a: map_key_clear(KEY_VIDEO); break; + case 0x04b: map_key_clear(KEY_ANGLE); break; + case 0x04c: map_key_clear(KEY_LANGUAGE); break; + case 0x04d: map_key_clear(KEY_SUBTITLE); break; + case 0x051: map_key_clear(KEY_RED); break; + case 0x052: map_key_clear(KEY_CLOSE); break; + default: goto ignore; + } + break; + + case HID_UP_PID: + + switch(usage->hid & HID_USAGE) { + case 0xa4: map_key_clear(BTN_DEAD); break; + default: goto ignore; + } + break; + + default: + unknown: + if (field->report_size == 1) { + if (field->report->type == HID_OUTPUT_REPORT) { + map_led(LED_MISC); + break; + } + map_key(BTN_MISC); + break; + } + if (field->flags & HID_MAIN_ITEM_RELATIVE) { + map_rel(REL_MISC); + break; + } + map_abs(ABS_MISC); + break; + } + + if (device->quirks & HID_QUIRK_MIGHTYMOUSE) { + if (usage->hid == HID_GD_Z) + map_rel(REL_HWHEEL); + else if (usage->code == BTN_1) + map_key(BTN_2); + else if (usage->code == BTN_2) + map_key(BTN_1); + } + + if ((device->quirks & (HID_QUIRK_2WHEEL_MOUSE_HACK_7 | HID_QUIRK_2WHEEL_MOUSE_HACK_5)) && + (usage->type == EV_REL) && (usage->code == REL_WHEEL)) + set_bit(REL_HWHEEL, bit); + + if (((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005)) + || ((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) + goto ignore; + + if ((device->quirks & HID_QUIRK_BAD_RELATIVE_KEYS) && + usage->type == EV_KEY && (field->flags & HID_MAIN_ITEM_RELATIVE)) + field->flags &= ~HID_MAIN_ITEM_RELATIVE; + + set_bit(usage->type, input->evbit); + + while (usage->code <= max && test_and_set_bit(usage->code, bit)) + usage->code = find_next_zero_bit(bit, max + 1, usage->code); + + if (usage->code > max) + goto ignore; + + + if (usage->type == EV_ABS) { + + int a = field->logical_minimum; + int b = field->logical_maximum; + + if ((device->quirks & HID_QUIRK_BADPAD) && (usage->code == ABS_X || usage->code == ABS_Y)) { + a = field->logical_minimum = 0; + b = field->logical_maximum = 255; + } + + if (field->application == HID_GD_GAMEPAD || field->application == HID_GD_JOYSTICK) + input_set_abs_params(input, usage->code, a, b, (b - a) >> 8, (b - a) >> 4); + else input_set_abs_params(input, usage->code, a, b, 0, 0); + + } + + if (usage->type == EV_ABS && + (usage->hat_min < usage->hat_max || usage->hat_dir)) { + int i; + for (i = usage->code; i < usage->code + 2 && i <= max; i++) { + input_set_abs_params(input, i, -1, 1, 0, 0); + set_bit(i, input->absbit); + } + if (usage->hat_dir && !field->dpad) + field->dpad = usage->code; + } + +#ifdef DEBUG + resolv_event(usage->type, usage->code); + printk("\n"); +#endif + return; + +ignore: +#ifdef DEBUG + printk("IGNORED\n"); +#endif + return; +} + +void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value) +{ + struct input_dev *input; + int *quirks = &hid->quirks; + + if (!field->hidinput) + return; + + input = field->hidinput->input; + + if (!usage->type) + return; + + if (((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005)) + || ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) { + if (value) hid->quirks |= HID_QUIRK_2WHEEL_MOUSE_HACK_ON; + else hid->quirks &= ~HID_QUIRK_2WHEEL_MOUSE_HACK_ON; + return; + } + + if ((hid->quirks & HID_QUIRK_INVERT_HWHEEL) && (usage->code == REL_HWHEEL)) { + input_event(input, usage->type, usage->code, -value); + return; + } + + if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_ON) && (usage->code == REL_WHEEL)) { + input_event(input, usage->type, REL_HWHEEL, value); + return; + } + + if ((hid->quirks & HID_QUIRK_POWERBOOK_HAS_FN) && hidinput_pb_event(hid, input, usage, value)) + return; + + if (usage->hat_min < usage->hat_max || usage->hat_dir) { + int hat_dir = usage->hat_dir; + if (!hat_dir) + hat_dir = (value - usage->hat_min) * 8 / (usage->hat_max - usage->hat_min + 1) + 1; + if (hat_dir < 0 || hat_dir > 8) hat_dir = 0; + input_event(input, usage->type, usage->code , hid_hat_to_axis[hat_dir].x); + input_event(input, usage->type, usage->code + 1, hid_hat_to_axis[hat_dir].y); + return; + } + + if (usage->hid == (HID_UP_DIGITIZER | 0x003c)) { /* Invert */ + *quirks = value ? (*quirks | HID_QUIRK_INVERT) : (*quirks & ~HID_QUIRK_INVERT); + return; + } + + if (usage->hid == (HID_UP_DIGITIZER | 0x0032)) { /* InRange */ + if (value) { + input_event(input, usage->type, (*quirks & HID_QUIRK_INVERT) ? BTN_TOOL_RUBBER : usage->code, 1); + return; + } + input_event(input, usage->type, usage->code, 0); + input_event(input, usage->type, BTN_TOOL_RUBBER, 0); + return; + } + + if (usage->hid == (HID_UP_DIGITIZER | 0x0030) && (*quirks & HID_QUIRK_NOTOUCH)) { /* Pressure */ + int a = field->logical_minimum; + int b = field->logical_maximum; + input_event(input, EV_KEY, BTN_TOUCH, value > a + ((b - a) >> 3)); + } + + if (usage->hid == (HID_UP_PID | 0x83UL)) { /* Simultaneous Effects Max */ + dbg("Maximum Effects - %d",value); + return; + } + + if (usage->hid == (HID_UP_PID | 0x7fUL)) { + dbg("PID Pool Report\n"); + return; + } + + if ((usage->type == EV_KEY) && (usage->code == 0)) /* Key 0 is "unassigned", not KEY_UNKNOWN */ + return; + + input_event(input, usage->type, usage->code, value); + + if ((field->flags & HID_MAIN_ITEM_RELATIVE) && (usage->type == EV_KEY)) + input_event(input, usage->type, usage->code, 0); +} + +void hidinput_report_event(struct hid_device *hid, struct hid_report *report) +{ + struct hid_input *hidinput; + + list_for_each_entry(hidinput, &hid->inputs, list) + input_sync(hidinput->input); +} + +static int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field) +{ + struct hid_report *report; + int i, j; + + list_for_each_entry(report, &hid->report_enum[HID_OUTPUT_REPORT].report_list, list) { + for (i = 0; i < report->maxfield; i++) { + *field = report->field[i]; + for (j = 0; j < (*field)->maxusage; j++) + if ((*field)->usage[j].type == type && (*field)->usage[j].code == code) + return j; + } + } + return -1; +} + +/* + * Register the input device; print a message. + * Configure the input layer interface + * Read all reports and initialize the absolute field values. + */ + +int hidinput_connect(struct hid_device *hid) +{ + struct usb_device *dev = hid->dev; + struct hid_report *report; + struct hid_input *hidinput = NULL; + struct input_dev *input_dev; + int i, j, k; + + INIT_LIST_HEAD(&hid->inputs); + + for (i = 0; i < hid->maxcollection; i++) + if (hid->collection[i].type == HID_COLLECTION_APPLICATION || + hid->collection[i].type == HID_COLLECTION_PHYSICAL) + if (IS_INPUT_APPLICATION(hid->collection[i].usage)) + break; + + if (i == hid->maxcollection) + return -1; + + for (k = HID_INPUT_REPORT; k <= HID_OUTPUT_REPORT; k++) + list_for_each_entry(report, &hid->report_enum[k].report_list, list) { + + if (!report->maxfield) + continue; + + if (!hidinput) { + hidinput = kzalloc(sizeof(*hidinput), GFP_KERNEL); + input_dev = input_allocate_device(); + if (!hidinput || !input_dev) { + kfree(hidinput); + input_free_device(input_dev); + err("Out of memory during hid input probe"); + return -1; + } + + input_dev->private = hid; + input_dev->event = hidinput_input_event; + input_dev->open = hidinput_open; + input_dev->close = hidinput_close; + + input_dev->name = hid->name; + input_dev->phys = hid->phys; + input_dev->uniq = hid->uniq; + usb_to_input_id(dev, &input_dev->id); + input_dev->cdev.dev = &hid->intf->dev; + + hidinput->input = input_dev; + list_add_tail(&hidinput->list, &hid->inputs); + } + + for (i = 0; i < report->maxfield; i++) + for (j = 0; j < report->field[i]->maxusage; j++) + hidinput_configure_usage(hidinput, report->field[i], + report->field[i]->usage + j); + + if (hid->quirks & HID_QUIRK_MULTI_INPUT) { + /* This will leave hidinput NULL, so that it + * allocates another one if we have more inputs on + * the same interface. Some devices (e.g. Happ's + * UGCI) cram a lot of unrelated inputs into the + * same interface. */ + hidinput->report = report; + input_register_device(hidinput->input); + hidinput = NULL; + } + } + + /* This only gets called when we are a single-input (most of the + * time). IOW, not a HID_QUIRK_MULTI_INPUT. The hid_ff_init() is + * only useful in this case, and not for multi-input quirks. */ + if (hidinput) { + hid_ff_init(hid); + input_register_device(hidinput->input); + } + + return 0; +} + +void hidinput_disconnect(struct hid_device *hid) +{ + struct hid_input *hidinput, *next; + + list_for_each_entry_safe(hidinput, next, &hid->inputs, list) { + list_del(&hidinput->list); + input_unregister_device(hidinput->input); + kfree(hidinput); + } +} diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 0811c39bd14f..06e169b6a17e 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -4,6 +4,7 @@ * Copyright (c) 1999 Andreas Gal * Copyright (c) 2000-2005 Vojtech Pavlik * Copyright (c) 2005 Michael Haboustak for Concept2, Inc + * Copyright (c) 2006 Jiri Kosina */ /* @@ -32,7 +33,7 @@ #include -#include "hid.h" +#include #include /* @@ -54,887 +55,6 @@ static unsigned int hid_mousepoll_interval; module_param_named(mousepoll, hid_mousepoll_interval, uint, 0644); MODULE_PARM_DESC(mousepoll, "Polling interval of mice"); -/* - * Register a new report for a device. - */ - -static struct hid_report *hid_register_report(struct hid_device *device, unsigned type, unsigned id) -{ - struct hid_report_enum *report_enum = device->report_enum + type; - struct hid_report *report; - - if (report_enum->report_id_hash[id]) - return report_enum->report_id_hash[id]; - - if (!(report = kzalloc(sizeof(struct hid_report), GFP_KERNEL))) - return NULL; - - if (id != 0) - report_enum->numbered = 1; - - report->id = id; - report->type = type; - report->size = 0; - report->device = device; - report_enum->report_id_hash[id] = report; - - list_add_tail(&report->list, &report_enum->report_list); - - return report; -} - -/* - * Register a new field for this report. - */ - -static struct hid_field *hid_register_field(struct hid_report *report, unsigned usages, unsigned values) -{ - struct hid_field *field; - - if (report->maxfield == HID_MAX_FIELDS) { - dbg("too many fields in report"); - return NULL; - } - - if (!(field = kzalloc(sizeof(struct hid_field) + usages * sizeof(struct hid_usage) - + values * sizeof(unsigned), GFP_KERNEL))) return NULL; - - field->index = report->maxfield++; - report->field[field->index] = field; - field->usage = (struct hid_usage *)(field + 1); - field->value = (unsigned *)(field->usage + usages); - field->report = report; - - return field; -} - -/* - * Open a collection. The type/usage is pushed on the stack. - */ - -static int open_collection(struct hid_parser *parser, unsigned type) -{ - struct hid_collection *collection; - unsigned usage; - - usage = parser->local.usage[0]; - - if (parser->collection_stack_ptr == HID_COLLECTION_STACK_SIZE) { - dbg("collection stack overflow"); - return -1; - } - - if (parser->device->maxcollection == parser->device->collection_size) { - collection = kmalloc(sizeof(struct hid_collection) * - parser->device->collection_size * 2, GFP_KERNEL); - if (collection == NULL) { - dbg("failed to reallocate collection array"); - return -1; - } - memcpy(collection, parser->device->collection, - sizeof(struct hid_collection) * - parser->device->collection_size); - memset(collection + parser->device->collection_size, 0, - sizeof(struct hid_collection) * - parser->device->collection_size); - kfree(parser->device->collection); - parser->device->collection = collection; - parser->device->collection_size *= 2; - } - - parser->collection_stack[parser->collection_stack_ptr++] = - parser->device->maxcollection; - - collection = parser->device->collection + - parser->device->maxcollection++; - collection->type = type; - collection->usage = usage; - collection->level = parser->collection_stack_ptr - 1; - - if (type == HID_COLLECTION_APPLICATION) - parser->device->maxapplication++; - - return 0; -} - -/* - * Close a collection. - */ - -static int close_collection(struct hid_parser *parser) -{ - if (!parser->collection_stack_ptr) { - dbg("collection stack underflow"); - return -1; - } - parser->collection_stack_ptr--; - return 0; -} - -/* - * Climb up the stack, search for the specified collection type - * and return the usage. - */ - -static unsigned hid_lookup_collection(struct hid_parser *parser, unsigned type) -{ - int n; - for (n = parser->collection_stack_ptr - 1; n >= 0; n--) - if (parser->device->collection[parser->collection_stack[n]].type == type) - return parser->device->collection[parser->collection_stack[n]].usage; - return 0; /* we know nothing about this usage type */ -} - -/* - * Add a usage to the temporary parser table. - */ - -static int hid_add_usage(struct hid_parser *parser, unsigned usage) -{ - if (parser->local.usage_index >= HID_MAX_USAGES) { - dbg("usage index exceeded"); - return -1; - } - parser->local.usage[parser->local.usage_index] = usage; - parser->local.collection_index[parser->local.usage_index] = - parser->collection_stack_ptr ? - parser->collection_stack[parser->collection_stack_ptr - 1] : 0; - parser->local.usage_index++; - return 0; -} - -/* - * Register a new field for this report. - */ - -static int hid_add_field(struct hid_parser *parser, unsigned report_type, unsigned flags) -{ - struct hid_report *report; - struct hid_field *field; - int usages; - unsigned offset; - int i; - - if (!(report = hid_register_report(parser->device, report_type, parser->global.report_id))) { - dbg("hid_register_report failed"); - return -1; - } - - if (parser->global.logical_maximum < parser->global.logical_minimum) { - dbg("logical range invalid %d %d", parser->global.logical_minimum, parser->global.logical_maximum); - return -1; - } - - offset = report->size; - report->size += parser->global.report_size * parser->global.report_count; - - if (!parser->local.usage_index) /* Ignore padding fields */ - return 0; - - usages = max_t(int, parser->local.usage_index, parser->global.report_count); - - if ((field = hid_register_field(report, usages, parser->global.report_count)) == NULL) - return 0; - - field->physical = hid_lookup_collection(parser, HID_COLLECTION_PHYSICAL); - field->logical = hid_lookup_collection(parser, HID_COLLECTION_LOGICAL); - field->application = hid_lookup_collection(parser, HID_COLLECTION_APPLICATION); - - for (i = 0; i < usages; i++) { - int j = i; - /* Duplicate the last usage we parsed if we have excess values */ - if (i >= parser->local.usage_index) - j = parser->local.usage_index - 1; - field->usage[i].hid = parser->local.usage[j]; - field->usage[i].collection_index = - parser->local.collection_index[j]; - } - - field->maxusage = usages; - field->flags = flags; - field->report_offset = offset; - field->report_type = report_type; - field->report_size = parser->global.report_size; - field->report_count = parser->global.report_count; - field->logical_minimum = parser->global.logical_minimum; - field->logical_maximum = parser->global.logical_maximum; - field->physical_minimum = parser->global.physical_minimum; - field->physical_maximum = parser->global.physical_maximum; - field->unit_exponent = parser->global.unit_exponent; - field->unit = parser->global.unit; - - return 0; -} - -/* - * Read data value from item. - */ - -static u32 item_udata(struct hid_item *item) -{ - switch (item->size) { - case 1: return item->data.u8; - case 2: return item->data.u16; - case 4: return item->data.u32; - } - return 0; -} - -static s32 item_sdata(struct hid_item *item) -{ - switch (item->size) { - case 1: return item->data.s8; - case 2: return item->data.s16; - case 4: return item->data.s32; - } - return 0; -} - -/* - * Process a global item. - */ - -static int hid_parser_global(struct hid_parser *parser, struct hid_item *item) -{ - switch (item->tag) { - - case HID_GLOBAL_ITEM_TAG_PUSH: - - if (parser->global_stack_ptr == HID_GLOBAL_STACK_SIZE) { - dbg("global enviroment stack overflow"); - return -1; - } - - memcpy(parser->global_stack + parser->global_stack_ptr++, - &parser->global, sizeof(struct hid_global)); - return 0; - - case HID_GLOBAL_ITEM_TAG_POP: - - if (!parser->global_stack_ptr) { - dbg("global enviroment stack underflow"); - return -1; - } - - memcpy(&parser->global, parser->global_stack + --parser->global_stack_ptr, - sizeof(struct hid_global)); - return 0; - - case HID_GLOBAL_ITEM_TAG_USAGE_PAGE: - parser->global.usage_page = item_udata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_LOGICAL_MINIMUM: - parser->global.logical_minimum = item_sdata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_LOGICAL_MAXIMUM: - if (parser->global.logical_minimum < 0) - parser->global.logical_maximum = item_sdata(item); - else - parser->global.logical_maximum = item_udata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_PHYSICAL_MINIMUM: - parser->global.physical_minimum = item_sdata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_PHYSICAL_MAXIMUM: - if (parser->global.physical_minimum < 0) - parser->global.physical_maximum = item_sdata(item); - else - parser->global.physical_maximum = item_udata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_UNIT_EXPONENT: - parser->global.unit_exponent = item_sdata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_UNIT: - parser->global.unit = item_udata(item); - return 0; - - case HID_GLOBAL_ITEM_TAG_REPORT_SIZE: - if ((parser->global.report_size = item_udata(item)) > 32) { - dbg("invalid report_size %d", parser->global.report_size); - return -1; - } - return 0; - - case HID_GLOBAL_ITEM_TAG_REPORT_COUNT: - if ((parser->global.report_count = item_udata(item)) > HID_MAX_USAGES) { - dbg("invalid report_count %d", parser->global.report_count); - return -1; - } - return 0; - - case HID_GLOBAL_ITEM_TAG_REPORT_ID: - if ((parser->global.report_id = item_udata(item)) == 0) { - dbg("report_id 0 is invalid"); - return -1; - } - return 0; - - default: - dbg("unknown global tag 0x%x", item->tag); - return -1; - } -} - -/* - * Process a local item. - */ - -static int hid_parser_local(struct hid_parser *parser, struct hid_item *item) -{ - __u32 data; - unsigned n; - - if (item->size == 0) { - dbg("item data expected for local item"); - return -1; - } - - data = item_udata(item); - - switch (item->tag) { - - case HID_LOCAL_ITEM_TAG_DELIMITER: - - if (data) { - /* - * We treat items before the first delimiter - * as global to all usage sets (branch 0). - * In the moment we process only these global - * items and the first delimiter set. - */ - if (parser->local.delimiter_depth != 0) { - dbg("nested delimiters"); - return -1; - } - parser->local.delimiter_depth++; - parser->local.delimiter_branch++; - } else { - if (parser->local.delimiter_depth < 1) { - dbg("bogus close delimiter"); - return -1; - } - parser->local.delimiter_depth--; - } - return 1; - - case HID_LOCAL_ITEM_TAG_USAGE: - - if (parser->local.delimiter_branch > 1) { - dbg("alternative usage ignored"); - return 0; - } - - if (item->size <= 2) - data = (parser->global.usage_page << 16) + data; - - return hid_add_usage(parser, data); - - case HID_LOCAL_ITEM_TAG_USAGE_MINIMUM: - - if (parser->local.delimiter_branch > 1) { - dbg("alternative usage ignored"); - return 0; - } - - if (item->size <= 2) - data = (parser->global.usage_page << 16) + data; - - parser->local.usage_minimum = data; - return 0; - - case HID_LOCAL_ITEM_TAG_USAGE_MAXIMUM: - - if (parser->local.delimiter_branch > 1) { - dbg("alternative usage ignored"); - return 0; - } - - if (item->size <= 2) - data = (parser->global.usage_page << 16) + data; - - for (n = parser->local.usage_minimum; n <= data; n++) - if (hid_add_usage(parser, n)) { - dbg("hid_add_usage failed\n"); - return -1; - } - return 0; - - default: - - dbg("unknown local item tag 0x%x", item->tag); - return 0; - } - return 0; -} - -/* - * Process a main item. - */ - -static int hid_parser_main(struct hid_parser *parser, struct hid_item *item) -{ - __u32 data; - int ret; - - data = item_udata(item); - - switch (item->tag) { - case HID_MAIN_ITEM_TAG_BEGIN_COLLECTION: - ret = open_collection(parser, data & 0xff); - break; - case HID_MAIN_ITEM_TAG_END_COLLECTION: - ret = close_collection(parser); - break; - case HID_MAIN_ITEM_TAG_INPUT: - ret = hid_add_field(parser, HID_INPUT_REPORT, data); - break; - case HID_MAIN_ITEM_TAG_OUTPUT: - ret = hid_add_field(parser, HID_OUTPUT_REPORT, data); - break; - case HID_MAIN_ITEM_TAG_FEATURE: - ret = hid_add_field(parser, HID_FEATURE_REPORT, data); - break; - default: - dbg("unknown main item tag 0x%x", item->tag); - ret = 0; - } - - memset(&parser->local, 0, sizeof(parser->local)); /* Reset the local parser environment */ - - return ret; -} - -/* - * Process a reserved item. - */ - -static int hid_parser_reserved(struct hid_parser *parser, struct hid_item *item) -{ - dbg("reserved item type, tag 0x%x", item->tag); - return 0; -} - -/* - * Free a report and all registered fields. The field->usage and - * field->value table's are allocated behind the field, so we need - * only to free(field) itself. - */ - -static void hid_free_report(struct hid_report *report) -{ - unsigned n; - - for (n = 0; n < report->maxfield; n++) - kfree(report->field[n]); - kfree(report); -} - -/* - * Free a device structure, all reports, and all fields. - */ - -static void hid_free_device(struct hid_device *device) -{ - unsigned i,j; - - for (i = 0; i < HID_REPORT_TYPES; i++) { - struct hid_report_enum *report_enum = device->report_enum + i; - - for (j = 0; j < 256; j++) { - struct hid_report *report = report_enum->report_id_hash[j]; - if (report) - hid_free_report(report); - } - } - - kfree(device->rdesc); - kfree(device); -} - -/* - * Fetch a report description item from the data stream. We support long - * items, though they are not used yet. - */ - -static u8 *fetch_item(__u8 *start, __u8 *end, struct hid_item *item) -{ - u8 b; - - if ((end - start) <= 0) - return NULL; - - b = *start++; - - item->type = (b >> 2) & 3; - item->tag = (b >> 4) & 15; - - if (item->tag == HID_ITEM_TAG_LONG) { - - item->format = HID_ITEM_FORMAT_LONG; - - if ((end - start) < 2) - return NULL; - - item->size = *start++; - item->tag = *start++; - - if ((end - start) < item->size) - return NULL; - - item->data.longdata = start; - start += item->size; - return start; - } - - item->format = HID_ITEM_FORMAT_SHORT; - item->size = b & 3; - - switch (item->size) { - - case 0: - return start; - - case 1: - if ((end - start) < 1) - return NULL; - item->data.u8 = *start++; - return start; - - case 2: - if ((end - start) < 2) - return NULL; - item->data.u16 = le16_to_cpu(get_unaligned((__le16*)start)); - start = (__u8 *)((__le16 *)start + 1); - return start; - - case 3: - item->size++; - if ((end - start) < 4) - return NULL; - item->data.u32 = le32_to_cpu(get_unaligned((__le32*)start)); - start = (__u8 *)((__le32 *)start + 1); - return start; - } - - return NULL; -} - -/* - * Parse a report description into a hid_device structure. Reports are - * enumerated, fields are attached to these reports. - */ - -static struct hid_device *hid_parse_report(__u8 *start, unsigned size) -{ - struct hid_device *device; - struct hid_parser *parser; - struct hid_item item; - __u8 *end; - unsigned i; - static int (*dispatch_type[])(struct hid_parser *parser, - struct hid_item *item) = { - hid_parser_main, - hid_parser_global, - hid_parser_local, - hid_parser_reserved - }; - - if (!(device = kzalloc(sizeof(struct hid_device), GFP_KERNEL))) - return NULL; - - if (!(device->collection = kzalloc(sizeof(struct hid_collection) * - HID_DEFAULT_NUM_COLLECTIONS, GFP_KERNEL))) { - kfree(device); - return NULL; - } - device->collection_size = HID_DEFAULT_NUM_COLLECTIONS; - - for (i = 0; i < HID_REPORT_TYPES; i++) - INIT_LIST_HEAD(&device->report_enum[i].report_list); - - if (!(device->rdesc = (__u8 *)kmalloc(size, GFP_KERNEL))) { - kfree(device->collection); - kfree(device); - return NULL; - } - memcpy(device->rdesc, start, size); - device->rsize = size; - - if (!(parser = kzalloc(sizeof(struct hid_parser), GFP_KERNEL))) { - kfree(device->rdesc); - kfree(device->collection); - kfree(device); - return NULL; - } - parser->device = device; - - end = start + size; - while ((start = fetch_item(start, end, &item)) != NULL) { - - if (item.format != HID_ITEM_FORMAT_SHORT) { - dbg("unexpected long global item"); - kfree(device->collection); - hid_free_device(device); - kfree(parser); - return NULL; - } - - if (dispatch_type[item.type](parser, &item)) { - dbg("item %u %u %u %u parsing failed\n", - item.format, (unsigned)item.size, (unsigned)item.type, (unsigned)item.tag); - kfree(device->collection); - hid_free_device(device); - kfree(parser); - return NULL; - } - - if (start == end) { - if (parser->collection_stack_ptr) { - dbg("unbalanced collection at end of report description"); - kfree(device->collection); - hid_free_device(device); - kfree(parser); - return NULL; - } - if (parser->local.delimiter_depth) { - dbg("unbalanced delimiter at end of report description"); - kfree(device->collection); - hid_free_device(device); - kfree(parser); - return NULL; - } - kfree(parser); - return device; - } - } - - dbg("item fetching failed at offset %d\n", (int)(end - start)); - kfree(device->collection); - hid_free_device(device); - kfree(parser); - return NULL; -} - -/* - * Convert a signed n-bit integer to signed 32-bit integer. Common - * cases are done through the compiler, the screwed things has to be - * done by hand. - */ - -static s32 snto32(__u32 value, unsigned n) -{ - switch (n) { - case 8: return ((__s8)value); - case 16: return ((__s16)value); - case 32: return ((__s32)value); - } - return value & (1 << (n - 1)) ? value | (-1 << n) : value; -} - -/* - * Convert a signed 32-bit integer to a signed n-bit integer. - */ - -static u32 s32ton(__s32 value, unsigned n) -{ - s32 a = value >> (n - 1); - if (a && a != -1) - return value < 0 ? 1 << (n - 1) : (1 << (n - 1)) - 1; - return value & ((1 << n) - 1); -} - -/* - * Extract/implement a data field from/to a little endian report (bit array). - * - * Code sort-of follows HID spec: - * http://www.usb.org/developers/devclass_docs/HID1_11.pdf - * - * While the USB HID spec allows unlimited length bit fields in "report - * descriptors", most devices never use more than 16 bits. - * One model of UPS is claimed to report "LINEV" as a 32-bit field. - * Search linux-kernel and linux-usb-devel archives for "hid-core extract". - */ - -static __inline__ __u32 extract(__u8 *report, unsigned offset, unsigned n) -{ - u64 x; - - WARN_ON(n > 32); - - report += offset >> 3; /* adjust byte index */ - offset &= 7; /* now only need bit offset into one byte */ - x = get_unaligned((u64 *) report); - x = le64_to_cpu(x); - x = (x >> offset) & ((1ULL << n) - 1); /* extract bit field */ - return (u32) x; -} - -/* - * "implement" : set bits in a little endian bit stream. - * Same concepts as "extract" (see comments above). - * The data mangled in the bit stream remains in little endian - * order the whole time. It make more sense to talk about - * endianness of register values by considering a register - * a "cached" copy of the little endiad bit stream. - */ -static __inline__ void implement(__u8 *report, unsigned offset, unsigned n, __u32 value) -{ - u64 x; - u64 m = (1ULL << n) - 1; - - WARN_ON(n > 32); - - WARN_ON(value > m); - value &= m; - - report += offset >> 3; - offset &= 7; - - x = get_unaligned((u64 *)report); - x &= cpu_to_le64(~(m << offset)); - x |= cpu_to_le64(((u64) value) << offset); - put_unaligned(x, (u64 *) report); -} - -/* - * Search an array for a value. - */ - -static __inline__ int search(__s32 *array, __s32 value, unsigned n) -{ - while (n--) { - if (*array++ == value) - return 0; - } - return -1; -} - -static void hid_process_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value, int interrupt) -{ - hid_dump_input(usage, value); - if (hid->claimed & HID_CLAIMED_INPUT) - hidinput_hid_event(hid, field, usage, value); - if (hid->claimed & HID_CLAIMED_HIDDEV && interrupt) - hiddev_hid_event(hid, field, usage, value); -} - -/* - * Analyse a received field, and fetch the data from it. The field - * content is stored for next report processing (we do differential - * reporting to the layer). - */ - -static void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt) -{ - unsigned n; - unsigned count = field->report_count; - unsigned offset = field->report_offset; - unsigned size = field->report_size; - __s32 min = field->logical_minimum; - __s32 max = field->logical_maximum; - __s32 *value; - - if (!(value = kmalloc(sizeof(__s32) * count, GFP_ATOMIC))) - return; - - for (n = 0; n < count; n++) { - - value[n] = min < 0 ? snto32(extract(data, offset + n * size, size), size) : - extract(data, offset + n * size, size); - - if (!(field->flags & HID_MAIN_ITEM_VARIABLE) /* Ignore report if ErrorRollOver */ - && value[n] >= min && value[n] <= max - && field->usage[value[n] - min].hid == HID_UP_KEYBOARD + 1) - goto exit; - } - - for (n = 0; n < count; n++) { - - if (HID_MAIN_ITEM_VARIABLE & field->flags) { - hid_process_event(hid, field, &field->usage[n], value[n], interrupt); - continue; - } - - if (field->value[n] >= min && field->value[n] <= max - && field->usage[field->value[n] - min].hid - && search(value, field->value[n], count)) - hid_process_event(hid, field, &field->usage[field->value[n] - min], 0, interrupt); - - if (value[n] >= min && value[n] <= max - && field->usage[value[n] - min].hid - && search(field->value, value[n], count)) - hid_process_event(hid, field, &field->usage[value[n] - min], 1, interrupt); - } - - memcpy(field->value, value, count * sizeof(__s32)); -exit: - kfree(value); -} - -static int hid_input_report(int type, struct urb *urb, int interrupt) -{ - struct hid_device *hid = urb->context; - struct hid_report_enum *report_enum = hid->report_enum + type; - u8 *data = urb->transfer_buffer; - int len = urb->actual_length; - struct hid_report *report; - int n, size; - - if (!len) { - dbg("empty report"); - return -1; - } - -#ifdef DEBUG_DATA - printk(KERN_DEBUG __FILE__ ": report (size %u) (%snumbered)\n", len, report_enum->numbered ? "" : "un"); -#endif - - n = 0; /* Normally report number is 0 */ - if (report_enum->numbered) { /* Device uses numbered reports, data[0] is report number */ - n = *data++; - len--; - } - -#ifdef DEBUG_DATA - { - int i; - printk(KERN_DEBUG __FILE__ ": report %d (size %u) = ", n, len); - for (i = 0; i < len; i++) - printk(" %02x", data[i]); - printk("\n"); - } -#endif - - if (!(report = report_enum->report_id_hash[n])) { - dbg("undefined report_id %d received", n); - return -1; - } - - size = ((report->size - 1) >> 3) + 1; - - if (len < size) { - dbg("report %d is too short, (%d < %d)", report->id, len, size); - memset(data + len, 0, size - len); - } - - if (hid->claimed & HID_CLAIMED_HIDDEV) - hiddev_report_event(hid, report); - - for (n = 0; n < report->maxfield; n++) - hid_input_field(hid, report->field[n], data, interrupt); - - if (hid->claimed & HID_CLAIMED_INPUT) - hidinput_report_event(hid, report); - - return 0; -} - /* * Input submission and I/O error handler. */ @@ -1044,6 +164,65 @@ done: spin_unlock_irqrestore(&hid->inlock, flags); } + +static int hid_input_report(int type, struct urb *urb, int interrupt) +{ + struct hid_device *hid = urb->context; + struct hid_report_enum *report_enum = hid->report_enum + type; + u8 *data = urb->transfer_buffer; + int len = urb->actual_length; + struct hid_report *report; + int n, size; + + if (!len) { + dbg("empty report"); + return -1; + } + +#ifdef DEBUG_DATA + printk(KERN_DEBUG __FILE__ ": report (size %u) (%snumbered)\n", len, report_enum->numbered ? "" : "un"); +#endif + + n = 0; /* Normally report number is 0 */ + if (report_enum->numbered) { /* Device uses numbered reports, data[0] is report number */ + n = *data++; + len--; + } + +#ifdef DEBUG_DATA + { + int i; + printk(KERN_DEBUG __FILE__ ": report %d (size %u) = ", n, len); + for (i = 0; i < len; i++) + printk(" %02x", data[i]); + printk("\n"); + } +#endif + + if (!(report = report_enum->report_id_hash[n])) { + dbg("undefined report_id %d received", n); + return -1; + } + + size = ((report->size - 1) >> 3) + 1; + + if (len < size) { + dbg("report %d is too short, (%d < %d)", report->id, len, size); + memset(data + len, 0, size - len); + } + + if (hid->claimed & HID_CLAIMED_HIDDEV) + hiddev_report_event(hid, report); + + for (n = 0; n < report->maxfield; n++) + hid_input_field(hid, report->field[n], data, interrupt); + + if (hid->claimed & HID_CLAIMED_INPUT) + hidinput_report_event(hid, report); + + return 0; +} + /* * Input interrupt completion handler. */ @@ -1092,67 +271,6 @@ static void hid_irq_in(struct urb *urb) } } -/* - * Output the field into the report. - */ - -static void hid_output_field(struct hid_field *field, __u8 *data) -{ - unsigned count = field->report_count; - unsigned offset = field->report_offset; - unsigned size = field->report_size; - unsigned n; - - for (n = 0; n < count; n++) { - if (field->logical_minimum < 0) /* signed values */ - implement(data, offset + n * size, size, s32ton(field->value[n], size)); - else /* unsigned values */ - implement(data, offset + n * size, size, field->value[n]); - } -} - -/* - * Create a report. - */ - -static void hid_output_report(struct hid_report *report, __u8 *data) -{ - unsigned n; - - if (report->id > 0) - *data++ = report->id; - - for (n = 0; n < report->maxfield; n++) - hid_output_field(report->field[n], data); -} - -/* - * Set a field value. The report this field belongs to has to be - * created and transferred to the device, to set this value in the - * device. - */ - -int hid_set_field(struct hid_field *field, unsigned offset, __s32 value) -{ - unsigned size = field->report_size; - - hid_dump_input(field->usage + offset, value); - - if (offset >= field->report_count) { - dbg("offset (%d) exceeds report_count (%d)", offset, field->report_count); - hid_dump_field(field, 8); - return -1; - } - if (field->logical_minimum < 0) { - if (value != snto32(s32ton(value, size), size)) { - dbg("value %d is out of range", value); - return -1; - } - } - field->value[offset] = value; - return 0; -} - /* * Find a report field with a specified HID usage. */ @@ -1379,6 +497,29 @@ void hid_submit_report(struct hid_device *hid, struct hid_report *report, unsign spin_unlock_irqrestore(&hid->ctrllock, flags); } +static int hidinput_input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) +{ + struct hid_device *hid = dev->private; + struct hid_field *field; + int offset; + + if (type == EV_FF) + return input_ff_event(dev, type, code, value); + + if (type != EV_LED) + return -1; + + if ((offset = hidinput_find_field(hid, type, code, &field)) == -1) { + warn("event field not found"); + return -1; + } + + hid_set_field(field, offset, value); + hid_submit_report(hid, field->report, USB_DIR_OUT); + + return 0; +} + int hid_wait_io(struct hid_device *hid) { if (!wait_event_timeout(hid->wait, (!test_bit(HID_CTRL_RUNNING, &hid->iofl) && @@ -1428,6 +569,18 @@ void hid_close(struct hid_device *hid) usb_kill_urb(hid->urbin); } +static int hidinput_open(struct input_dev *dev) +{ + struct hid_device *hid = dev->private; + return hid_open(hid); +} + +static void hidinput_close(struct input_dev *dev) +{ + struct hid_device *hid = dev->private; + hid_close(hid); +} + #define USB_VENDOR_ID_PANJIT 0x134c #define USB_VENDOR_ID_TURBOX 0x062a diff --git a/drivers/usb/input/hid-debug.h b/drivers/usb/input/hid-debug.h deleted file mode 100644 index f04d6d75c098..000000000000 --- a/drivers/usb/input/hid-debug.h +++ /dev/null @@ -1,757 +0,0 @@ -/* - * $Id: hid-debug.h,v 1.8 2001/09/25 09:37:57 vojtech Exp $ - * - * (c) 1999 Andreas Gal - * (c) 2000-2001 Vojtech Pavlik - * - * Some debug stuff for the HID parser. - */ - -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Should you need to contact me, the author, you can do so either by - * e-mail - mail your message to , or by paper mail: - * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic - */ - -#include - -struct hid_usage_entry { - unsigned page; - unsigned usage; - char *description; -}; - -static const struct hid_usage_entry hid_usage_table[] = { - { 0, 0, "Undefined" }, - { 1, 0, "GenericDesktop" }, - {0, 0x01, "Pointer"}, - {0, 0x02, "Mouse"}, - {0, 0x04, "Joystick"}, - {0, 0x05, "GamePad"}, - {0, 0x06, "Keyboard"}, - {0, 0x07, "Keypad"}, - {0, 0x08, "MultiAxis"}, - {0, 0x30, "X"}, - {0, 0x31, "Y"}, - {0, 0x32, "Z"}, - {0, 0x33, "Rx"}, - {0, 0x34, "Ry"}, - {0, 0x35, "Rz"}, - {0, 0x36, "Slider"}, - {0, 0x37, "Dial"}, - {0, 0x38, "Wheel"}, - {0, 0x39, "HatSwitch"}, - {0, 0x3a, "CountedBuffer"}, - {0, 0x3b, "ByteCount"}, - {0, 0x3c, "MotionWakeup"}, - {0, 0x3d, "Start"}, - {0, 0x3e, "Select"}, - {0, 0x40, "Vx"}, - {0, 0x41, "Vy"}, - {0, 0x42, "Vz"}, - {0, 0x43, "Vbrx"}, - {0, 0x44, "Vbry"}, - {0, 0x45, "Vbrz"}, - {0, 0x46, "Vno"}, - {0, 0x80, "SystemControl"}, - {0, 0x81, "SystemPowerDown"}, - {0, 0x82, "SystemSleep"}, - {0, 0x83, "SystemWakeUp"}, - {0, 0x84, "SystemContextMenu"}, - {0, 0x85, "SystemMainMenu"}, - {0, 0x86, "SystemAppMenu"}, - {0, 0x87, "SystemMenuHelp"}, - {0, 0x88, "SystemMenuExit"}, - {0, 0x89, "SystemMenuSelect"}, - {0, 0x8a, "SystemMenuRight"}, - {0, 0x8b, "SystemMenuLeft"}, - {0, 0x8c, "SystemMenuUp"}, - {0, 0x8d, "SystemMenuDown"}, - {0, 0x90, "D-PadUp"}, - {0, 0x91, "D-PadDown"}, - {0, 0x92, "D-PadRight"}, - {0, 0x93, "D-PadLeft"}, - { 2, 0, "Simulation" }, - {0, 0xb0, "Aileron"}, - {0, 0xb1, "AileronTrim"}, - {0, 0xb2, "Anti-Torque"}, - {0, 0xb3, "Autopilot"}, - {0, 0xb4, "Chaff"}, - {0, 0xb5, "Collective"}, - {0, 0xb6, "DiveBrake"}, - {0, 0xb7, "ElectronicCountermeasures"}, - {0, 0xb8, "Elevator"}, - {0, 0xb9, "ElevatorTrim"}, - {0, 0xba, "Rudder"}, - {0, 0xbb, "Throttle"}, - {0, 0xbc, "FlightCommunications"}, - {0, 0xbd, "FlareRelease"}, - {0, 0xbe, "LandingGear"}, - {0, 0xbf, "ToeBrake"}, - { 7, 0, "Keyboard" }, - { 8, 0, "LED" }, - {0, 0x01, "NumLock"}, - {0, 0x02, "CapsLock"}, - {0, 0x03, "ScrollLock"}, - {0, 0x04, "Compose"}, - {0, 0x05, "Kana"}, - {0, 0x4b, "GenericIndicator"}, - { 9, 0, "Button" }, - { 10, 0, "Ordinal" }, - { 12, 0, "Consumer" }, - {0, 0x238, "HorizontalWheel"}, - { 13, 0, "Digitizers" }, - {0, 0x01, "Digitizer"}, - {0, 0x02, "Pen"}, - {0, 0x03, "LightPen"}, - {0, 0x04, "TouchScreen"}, - {0, 0x05, "TouchPad"}, - {0, 0x20, "Stylus"}, - {0, 0x21, "Puck"}, - {0, 0x22, "Finger"}, - {0, 0x30, "TipPressure"}, - {0, 0x31, "BarrelPressure"}, - {0, 0x32, "InRange"}, - {0, 0x33, "Touch"}, - {0, 0x34, "UnTouch"}, - {0, 0x35, "Tap"}, - {0, 0x39, "TabletFunctionKey"}, - {0, 0x3a, "ProgramChangeKey"}, - {0, 0x3c, "Invert"}, - {0, 0x42, "TipSwitch"}, - {0, 0x43, "SecondaryTipSwitch"}, - {0, 0x44, "BarrelSwitch"}, - {0, 0x45, "Eraser"}, - {0, 0x46, "TabletPick"}, - { 15, 0, "PhysicalInterfaceDevice" }, - {0, 0x00, "Undefined"}, - {0, 0x01, "Physical_Interface_Device"}, - {0, 0x20, "Normal"}, - {0, 0x21, "Set_Effect_Report"}, - {0, 0x22, "Effect_Block_Index"}, - {0, 0x23, "Parameter_Block_Offset"}, - {0, 0x24, "ROM_Flag"}, - {0, 0x25, "Effect_Type"}, - {0, 0x26, "ET_Constant_Force"}, - {0, 0x27, "ET_Ramp"}, - {0, 0x28, "ET_Custom_Force_Data"}, - {0, 0x30, "ET_Square"}, - {0, 0x31, "ET_Sine"}, - {0, 0x32, "ET_Triangle"}, - {0, 0x33, "ET_Sawtooth_Up"}, - {0, 0x34, "ET_Sawtooth_Down"}, - {0, 0x40, "ET_Spring"}, - {0, 0x41, "ET_Damper"}, - {0, 0x42, "ET_Inertia"}, - {0, 0x43, "ET_Friction"}, - {0, 0x50, "Duration"}, - {0, 0x51, "Sample_Period"}, - {0, 0x52, "Gain"}, - {0, 0x53, "Trigger_Button"}, - {0, 0x54, "Trigger_Repeat_Interval"}, - {0, 0x55, "Axes_Enable"}, - {0, 0x56, "Direction_Enable"}, - {0, 0x57, "Direction"}, - {0, 0x58, "Type_Specific_Block_Offset"}, - {0, 0x59, "Block_Type"}, - {0, 0x5A, "Set_Envelope_Report"}, - {0, 0x5B, "Attack_Level"}, - {0, 0x5C, "Attack_Time"}, - {0, 0x5D, "Fade_Level"}, - {0, 0x5E, "Fade_Time"}, - {0, 0x5F, "Set_Condition_Report"}, - {0, 0x60, "CP_Offset"}, - {0, 0x61, "Positive_Coefficient"}, - {0, 0x62, "Negative_Coefficient"}, - {0, 0x63, "Positive_Saturation"}, - {0, 0x64, "Negative_Saturation"}, - {0, 0x65, "Dead_Band"}, - {0, 0x66, "Download_Force_Sample"}, - {0, 0x67, "Isoch_Custom_Force_Enable"}, - {0, 0x68, "Custom_Force_Data_Report"}, - {0, 0x69, "Custom_Force_Data"}, - {0, 0x6A, "Custom_Force_Vendor_Defined_Data"}, - {0, 0x6B, "Set_Custom_Force_Report"}, - {0, 0x6C, "Custom_Force_Data_Offset"}, - {0, 0x6D, "Sample_Count"}, - {0, 0x6E, "Set_Periodic_Report"}, - {0, 0x6F, "Offset"}, - {0, 0x70, "Magnitude"}, - {0, 0x71, "Phase"}, - {0, 0x72, "Period"}, - {0, 0x73, "Set_Constant_Force_Report"}, - {0, 0x74, "Set_Ramp_Force_Report"}, - {0, 0x75, "Ramp_Start"}, - {0, 0x76, "Ramp_End"}, - {0, 0x77, "Effect_Operation_Report"}, - {0, 0x78, "Effect_Operation"}, - {0, 0x79, "Op_Effect_Start"}, - {0, 0x7A, "Op_Effect_Start_Solo"}, - {0, 0x7B, "Op_Effect_Stop"}, - {0, 0x7C, "Loop_Count"}, - {0, 0x7D, "Device_Gain_Report"}, - {0, 0x7E, "Device_Gain"}, - {0, 0x7F, "PID_Pool_Report"}, - {0, 0x80, "RAM_Pool_Size"}, - {0, 0x81, "ROM_Pool_Size"}, - {0, 0x82, "ROM_Effect_Block_Count"}, - {0, 0x83, "Simultaneous_Effects_Max"}, - {0, 0x84, "Pool_Alignment"}, - {0, 0x85, "PID_Pool_Move_Report"}, - {0, 0x86, "Move_Source"}, - {0, 0x87, "Move_Destination"}, - {0, 0x88, "Move_Length"}, - {0, 0x89, "PID_Block_Load_Report"}, - {0, 0x8B, "Block_Load_Status"}, - {0, 0x8C, "Block_Load_Success"}, - {0, 0x8D, "Block_Load_Full"}, - {0, 0x8E, "Block_Load_Error"}, - {0, 0x8F, "Block_Handle"}, - {0, 0x90, "PID_Block_Free_Report"}, - {0, 0x91, "Type_Specific_Block_Handle"}, - {0, 0x92, "PID_State_Report"}, - {0, 0x94, "Effect_Playing"}, - {0, 0x95, "PID_Device_Control_Report"}, - {0, 0x96, "PID_Device_Control"}, - {0, 0x97, "DC_Enable_Actuators"}, - {0, 0x98, "DC_Disable_Actuators"}, - {0, 0x99, "DC_Stop_All_Effects"}, - {0, 0x9A, "DC_Device_Reset"}, - {0, 0x9B, "DC_Device_Pause"}, - {0, 0x9C, "DC_Device_Continue"}, - {0, 0x9F, "Device_Paused"}, - {0, 0xA0, "Actuators_Enabled"}, - {0, 0xA4, "Safety_Switch"}, - {0, 0xA5, "Actuator_Override_Switch"}, - {0, 0xA6, "Actuator_Power"}, - {0, 0xA7, "Start_Delay"}, - {0, 0xA8, "Parameter_Block_Size"}, - {0, 0xA9, "Device_Managed_Pool"}, - {0, 0xAA, "Shared_Parameter_Blocks"}, - {0, 0xAB, "Create_New_Effect_Report"}, - {0, 0xAC, "RAM_Pool_Available"}, - { 0x84, 0, "Power Device" }, - { 0x84, 0x02, "PresentStatus" }, - { 0x84, 0x03, "ChangeStatus" }, - { 0x84, 0x04, "UPS" }, - { 0x84, 0x05, "PowerSupply" }, - { 0x84, 0x10, "BatterySystem" }, - { 0x84, 0x11, "BatterySystemID" }, - { 0x84, 0x12, "Battery" }, - { 0x84, 0x13, "BatteryID" }, - { 0x84, 0x14, "Charger" }, - { 0x84, 0x15, "ChargerID" }, - { 0x84, 0x16, "PowerConverter" }, - { 0x84, 0x17, "PowerConverterID" }, - { 0x84, 0x18, "OutletSystem" }, - { 0x84, 0x19, "OutletSystemID" }, - { 0x84, 0x1a, "Input" }, - { 0x84, 0x1b, "InputID" }, - { 0x84, 0x1c, "Output" }, - { 0x84, 0x1d, "OutputID" }, - { 0x84, 0x1e, "Flow" }, - { 0x84, 0x1f, "FlowID" }, - { 0x84, 0x20, "Outlet" }, - { 0x84, 0x21, "OutletID" }, - { 0x84, 0x22, "Gang" }, - { 0x84, 0x24, "PowerSummary" }, - { 0x84, 0x25, "PowerSummaryID" }, - { 0x84, 0x30, "Voltage" }, - { 0x84, 0x31, "Current" }, - { 0x84, 0x32, "Frequency" }, - { 0x84, 0x33, "ApparentPower" }, - { 0x84, 0x35, "PercentLoad" }, - { 0x84, 0x40, "ConfigVoltage" }, - { 0x84, 0x41, "ConfigCurrent" }, - { 0x84, 0x43, "ConfigApparentPower" }, - { 0x84, 0x53, "LowVoltageTransfer" }, - { 0x84, 0x54, "HighVoltageTransfer" }, - { 0x84, 0x56, "DelayBeforeStartup" }, - { 0x84, 0x57, "DelayBeforeShutdown" }, - { 0x84, 0x58, "Test" }, - { 0x84, 0x5a, "AudibleAlarmControl" }, - { 0x84, 0x60, "Present" }, - { 0x84, 0x61, "Good" }, - { 0x84, 0x62, "InternalFailure" }, - { 0x84, 0x65, "Overload" }, - { 0x84, 0x66, "OverCharged" }, - { 0x84, 0x67, "OverTemperature" }, - { 0x84, 0x68, "ShutdownRequested" }, - { 0x84, 0x69, "ShutdownImminent" }, - { 0x84, 0x6b, "SwitchOn/Off" }, - { 0x84, 0x6c, "Switchable" }, - { 0x84, 0x6d, "Used" }, - { 0x84, 0x6e, "Boost" }, - { 0x84, 0x73, "CommunicationLost" }, - { 0x84, 0xfd, "iManufacturer" }, - { 0x84, 0xfe, "iProduct" }, - { 0x84, 0xff, "iSerialNumber" }, - { 0x85, 0, "Battery System" }, - { 0x85, 0x01, "SMBBatteryMode" }, - { 0x85, 0x02, "SMBBatteryStatus" }, - { 0x85, 0x03, "SMBAlarmWarning" }, - { 0x85, 0x04, "SMBChargerMode" }, - { 0x85, 0x05, "SMBChargerStatus" }, - { 0x85, 0x06, "SMBChargerSpecInfo" }, - { 0x85, 0x07, "SMBSelectorState" }, - { 0x85, 0x08, "SMBSelectorPresets" }, - { 0x85, 0x09, "SMBSelectorInfo" }, - { 0x85, 0x29, "RemainingCapacityLimit" }, - { 0x85, 0x2c, "CapacityMode" }, - { 0x85, 0x42, "BelowRemainingCapacityLimit" }, - { 0x85, 0x44, "Charging" }, - { 0x85, 0x45, "Discharging" }, - { 0x85, 0x4b, "NeedReplacement" }, - { 0x85, 0x66, "RemainingCapacity" }, - { 0x85, 0x68, "RunTimeToEmpty" }, - { 0x85, 0x6a, "AverageTimeToFull" }, - { 0x85, 0x83, "DesignCapacity" }, - { 0x85, 0x85, "ManufacturerDate" }, - { 0x85, 0x89, "iDeviceChemistry" }, - { 0x85, 0x8b, "Rechargable" }, - { 0x85, 0x8f, "iOEMInformation" }, - { 0x85, 0x8d, "CapacityGranularity1" }, - { 0x85, 0xd0, "ACPresent" }, - /* pages 0xff00 to 0xffff are vendor-specific */ - { 0xffff, 0, "Vendor-specific-FF" }, - { 0, 0, NULL } -}; - -static void resolv_usage_page(unsigned page) { - const struct hid_usage_entry *p; - - for (p = hid_usage_table; p->description; p++) - if (p->page == page) { - printk("%s", p->description); - return; - } - printk("%04x", page); -} - -static void resolv_usage(unsigned usage) { - const struct hid_usage_entry *p; - - resolv_usage_page(usage >> 16); - printk("."); - for (p = hid_usage_table; p->description; p++) - if (p->page == (usage >> 16)) { - for(++p; p->description && p->usage != 0; p++) - if (p->usage == (usage & 0xffff)) { - printk("%s", p->description); - return; - } - break; - } - printk("%04x", usage & 0xffff); -} - -__inline__ static void tab(int n) { - while (n--) printk(" "); -} - -static void hid_dump_field(struct hid_field *field, int n) { - int j; - - if (field->physical) { - tab(n); - printk("Physical("); - resolv_usage(field->physical); printk(")\n"); - } - if (field->logical) { - tab(n); - printk("Logical("); - resolv_usage(field->logical); printk(")\n"); - } - tab(n); printk("Usage(%d)\n", field->maxusage); - for (j = 0; j < field->maxusage; j++) { - tab(n+2);resolv_usage(field->usage[j].hid); printk("\n"); - } - if (field->logical_minimum != field->logical_maximum) { - tab(n); printk("Logical Minimum(%d)\n", field->logical_minimum); - tab(n); printk("Logical Maximum(%d)\n", field->logical_maximum); - } - if (field->physical_minimum != field->physical_maximum) { - tab(n); printk("Physical Minimum(%d)\n", field->physical_minimum); - tab(n); printk("Physical Maximum(%d)\n", field->physical_maximum); - } - if (field->unit_exponent) { - tab(n); printk("Unit Exponent(%d)\n", field->unit_exponent); - } - if (field->unit) { - char *systems[5] = { "None", "SI Linear", "SI Rotation", "English Linear", "English Rotation" }; - char *units[5][8] = { - { "None", "None", "None", "None", "None", "None", "None", "None" }, - { "None", "Centimeter", "Gram", "Seconds", "Kelvin", "Ampere", "Candela", "None" }, - { "None", "Radians", "Gram", "Seconds", "Kelvin", "Ampere", "Candela", "None" }, - { "None", "Inch", "Slug", "Seconds", "Fahrenheit", "Ampere", "Candela", "None" }, - { "None", "Degrees", "Slug", "Seconds", "Fahrenheit", "Ampere", "Candela", "None" } - }; - - int i; - int sys; - __u32 data = field->unit; - - /* First nibble tells us which system we're in. */ - sys = data & 0xf; - data >>= 4; - - if(sys > 4) { - tab(n); printk("Unit(Invalid)\n"); - } - else { - int earlier_unit = 0; - - tab(n); printk("Unit(%s : ", systems[sys]); - - for (i=1 ; i>= 4; - if (nibble != 0) { - if(earlier_unit++ > 0) - printk("*"); - printk("%s", units[sys][i]); - if(nibble != 1) { - /* This is a _signed_ nibble(!) */ - - int val = nibble & 0x7; - if(nibble & 0x08) - val = -((0x7 & ~val) +1); - printk("^%d", val); - } - } - } - printk(")\n"); - } - } - tab(n); printk("Report Size(%u)\n", field->report_size); - tab(n); printk("Report Count(%u)\n", field->report_count); - tab(n); printk("Report Offset(%u)\n", field->report_offset); - - tab(n); printk("Flags( "); - j = field->flags; - printk("%s", HID_MAIN_ITEM_CONSTANT & j ? "Constant " : ""); - printk("%s", HID_MAIN_ITEM_VARIABLE & j ? "Variable " : "Array "); - printk("%s", HID_MAIN_ITEM_RELATIVE & j ? "Relative " : "Absolute "); - printk("%s", HID_MAIN_ITEM_WRAP & j ? "Wrap " : ""); - printk("%s", HID_MAIN_ITEM_NONLINEAR & j ? "NonLinear " : ""); - printk("%s", HID_MAIN_ITEM_NO_PREFERRED & j ? "NoPrefferedState " : ""); - printk("%s", HID_MAIN_ITEM_NULL_STATE & j ? "NullState " : ""); - printk("%s", HID_MAIN_ITEM_VOLATILE & j ? "Volatile " : ""); - printk("%s", HID_MAIN_ITEM_BUFFERED_BYTE & j ? "BufferedByte " : ""); - printk(")\n"); -} - -static void __attribute__((unused)) hid_dump_device(struct hid_device *device) { - struct hid_report_enum *report_enum; - struct hid_report *report; - struct list_head *list; - unsigned i,k; - static char *table[] = {"INPUT", "OUTPUT", "FEATURE"}; - - for (i = 0; i < HID_REPORT_TYPES; i++) { - report_enum = device->report_enum + i; - list = report_enum->report_list.next; - while (list != &report_enum->report_list) { - report = (struct hid_report *) list; - tab(2); - printk("%s", table[i]); - if (report->id) - printk("(%d)", report->id); - printk("[%s]", table[report->type]); - printk("\n"); - for (k = 0; k < report->maxfield; k++) { - tab(4); - printk("Field(%d)\n", k); - hid_dump_field(report->field[k], 6); - } - list = list->next; - } - } -} - -static void __attribute__((unused)) hid_dump_input(struct hid_usage *usage, __s32 value) { - printk("hid-debug: input "); - resolv_usage(usage->hid); - printk(" = %d\n", value); -} - - -static char *events[EV_MAX + 1] = { - [EV_SYN] = "Sync", [EV_KEY] = "Key", - [EV_REL] = "Relative", [EV_ABS] = "Absolute", - [EV_MSC] = "Misc", [EV_LED] = "LED", - [EV_SND] = "Sound", [EV_REP] = "Repeat", - [EV_FF] = "ForceFeedback", [EV_PWR] = "Power", - [EV_FF_STATUS] = "ForceFeedbackStatus", -}; - -static char *syncs[2] = { - [SYN_REPORT] = "Report", [SYN_CONFIG] = "Config", -}; -static char *keys[KEY_MAX + 1] = { - [KEY_RESERVED] = "Reserved", [KEY_ESC] = "Esc", - [KEY_1] = "1", [KEY_2] = "2", - [KEY_3] = "3", [KEY_4] = "4", - [KEY_5] = "5", [KEY_6] = "6", - [KEY_7] = "7", [KEY_8] = "8", - [KEY_9] = "9", [KEY_0] = "0", - [KEY_MINUS] = "Minus", [KEY_EQUAL] = "Equal", - [KEY_BACKSPACE] = "Backspace", [KEY_TAB] = "Tab", - [KEY_Q] = "Q", [KEY_W] = "W", - [KEY_E] = "E", [KEY_R] = "R", - [KEY_T] = "T", [KEY_Y] = "Y", - [KEY_U] = "U", [KEY_I] = "I", - [KEY_O] = "O", [KEY_P] = "P", - [KEY_LEFTBRACE] = "LeftBrace", [KEY_RIGHTBRACE] = "RightBrace", - [KEY_ENTER] = "Enter", [KEY_LEFTCTRL] = "LeftControl", - [KEY_A] = "A", [KEY_S] = "S", - [KEY_D] = "D", [KEY_F] = "F", - [KEY_G] = "G", [KEY_H] = "H", - [KEY_J] = "J", [KEY_K] = "K", - [KEY_L] = "L", [KEY_SEMICOLON] = "Semicolon", - [KEY_APOSTROPHE] = "Apostrophe", [KEY_GRAVE] = "Grave", - [KEY_LEFTSHIFT] = "LeftShift", [KEY_BACKSLASH] = "BackSlash", - [KEY_Z] = "Z", [KEY_X] = "X", - [KEY_C] = "C", [KEY_V] = "V", - [KEY_B] = "B", [KEY_N] = "N", - [KEY_M] = "M", [KEY_COMMA] = "Comma", - [KEY_DOT] = "Dot", [KEY_SLASH] = "Slash", - [KEY_RIGHTSHIFT] = "RightShift", [KEY_KPASTERISK] = "KPAsterisk", - [KEY_LEFTALT] = "LeftAlt", [KEY_SPACE] = "Space", - [KEY_CAPSLOCK] = "CapsLock", [KEY_F1] = "F1", - [KEY_F2] = "F2", [KEY_F3] = "F3", - [KEY_F4] = "F4", [KEY_F5] = "F5", - [KEY_F6] = "F6", [KEY_F7] = "F7", - [KEY_F8] = "F8", [KEY_F9] = "F9", - [KEY_F10] = "F10", [KEY_NUMLOCK] = "NumLock", - [KEY_SCROLLLOCK] = "ScrollLock", [KEY_KP7] = "KP7", - [KEY_KP8] = "KP8", [KEY_KP9] = "KP9", - [KEY_KPMINUS] = "KPMinus", [KEY_KP4] = "KP4", - [KEY_KP5] = "KP5", [KEY_KP6] = "KP6", - [KEY_KPPLUS] = "KPPlus", [KEY_KP1] = "KP1", - [KEY_KP2] = "KP2", [KEY_KP3] = "KP3", - [KEY_KP0] = "KP0", [KEY_KPDOT] = "KPDot", - [KEY_ZENKAKUHANKAKU] = "Zenkaku/Hankaku", [KEY_102ND] = "102nd", - [KEY_F11] = "F11", [KEY_F12] = "F12", - [KEY_RO] = "RO", [KEY_KATAKANA] = "Katakana", - [KEY_HIRAGANA] = "HIRAGANA", [KEY_HENKAN] = "Henkan", - [KEY_KATAKANAHIRAGANA] = "Katakana/Hiragana", [KEY_MUHENKAN] = "Muhenkan", - [KEY_KPJPCOMMA] = "KPJpComma", [KEY_KPENTER] = "KPEnter", - [KEY_RIGHTCTRL] = "RightCtrl", [KEY_KPSLASH] = "KPSlash", - [KEY_SYSRQ] = "SysRq", [KEY_RIGHTALT] = "RightAlt", - [KEY_LINEFEED] = "LineFeed", [KEY_HOME] = "Home", - [KEY_UP] = "Up", [KEY_PAGEUP] = "PageUp", - [KEY_LEFT] = "Left", [KEY_RIGHT] = "Right", - [KEY_END] = "End", [KEY_DOWN] = "Down", - [KEY_PAGEDOWN] = "PageDown", [KEY_INSERT] = "Insert", - [KEY_DELETE] = "Delete", [KEY_MACRO] = "Macro", - [KEY_MUTE] = "Mute", [KEY_VOLUMEDOWN] = "VolumeDown", - [KEY_VOLUMEUP] = "VolumeUp", [KEY_POWER] = "Power", - [KEY_KPEQUAL] = "KPEqual", [KEY_KPPLUSMINUS] = "KPPlusMinus", - [KEY_PAUSE] = "Pause", [KEY_KPCOMMA] = "KPComma", - [KEY_HANGUEL] = "Hangeul", [KEY_HANJA] = "Hanja", - [KEY_YEN] = "Yen", [KEY_LEFTMETA] = "LeftMeta", - [KEY_RIGHTMETA] = "RightMeta", [KEY_COMPOSE] = "Compose", - [KEY_STOP] = "Stop", [KEY_AGAIN] = "Again", - [KEY_PROPS] = "Props", [KEY_UNDO] = "Undo", - [KEY_FRONT] = "Front", [KEY_COPY] = "Copy", - [KEY_OPEN] = "Open", [KEY_PASTE] = "Paste", - [KEY_FIND] = "Find", [KEY_CUT] = "Cut", - [KEY_HELP] = "Help", [KEY_MENU] = "Menu", - [KEY_CALC] = "Calc", [KEY_SETUP] = "Setup", - [KEY_SLEEP] = "Sleep", [KEY_WAKEUP] = "WakeUp", - [KEY_FILE] = "File", [KEY_SENDFILE] = "SendFile", - [KEY_DELETEFILE] = "DeleteFile", [KEY_XFER] = "X-fer", - [KEY_PROG1] = "Prog1", [KEY_PROG2] = "Prog2", - [KEY_WWW] = "WWW", [KEY_MSDOS] = "MSDOS", - [KEY_COFFEE] = "Coffee", [KEY_DIRECTION] = "Direction", - [KEY_CYCLEWINDOWS] = "CycleWindows", [KEY_MAIL] = "Mail", - [KEY_BOOKMARKS] = "Bookmarks", [KEY_COMPUTER] = "Computer", - [KEY_BACK] = "Back", [KEY_FORWARD] = "Forward", - [KEY_CLOSECD] = "CloseCD", [KEY_EJECTCD] = "EjectCD", - [KEY_EJECTCLOSECD] = "EjectCloseCD", [KEY_NEXTSONG] = "NextSong", - [KEY_PLAYPAUSE] = "PlayPause", [KEY_PREVIOUSSONG] = "PreviousSong", - [KEY_STOPCD] = "StopCD", [KEY_RECORD] = "Record", - [KEY_REWIND] = "Rewind", [KEY_PHONE] = "Phone", - [KEY_ISO] = "ISOKey", [KEY_CONFIG] = "Config", - [KEY_HOMEPAGE] = "HomePage", [KEY_REFRESH] = "Refresh", - [KEY_EXIT] = "Exit", [KEY_MOVE] = "Move", - [KEY_EDIT] = "Edit", [KEY_SCROLLUP] = "ScrollUp", - [KEY_SCROLLDOWN] = "ScrollDown", [KEY_KPLEFTPAREN] = "KPLeftParenthesis", - [KEY_KPRIGHTPAREN] = "KPRightParenthesis", [KEY_NEW] = "New", - [KEY_REDO] = "Redo", [KEY_F13] = "F13", - [KEY_F14] = "F14", [KEY_F15] = "F15", - [KEY_F16] = "F16", [KEY_F17] = "F17", - [KEY_F18] = "F18", [KEY_F19] = "F19", - [KEY_F20] = "F20", [KEY_F21] = "F21", - [KEY_F22] = "F22", [KEY_F23] = "F23", - [KEY_F24] = "F24", [KEY_PLAYCD] = "PlayCD", - [KEY_PAUSECD] = "PauseCD", [KEY_PROG3] = "Prog3", - [KEY_PROG4] = "Prog4", [KEY_SUSPEND] = "Suspend", - [KEY_CLOSE] = "Close", [KEY_PLAY] = "Play", - [KEY_FASTFORWARD] = "FastForward", [KEY_BASSBOOST] = "BassBoost", - [KEY_PRINT] = "Print", [KEY_HP] = "HP", - [KEY_CAMERA] = "Camera", [KEY_SOUND] = "Sound", - [KEY_QUESTION] = "Question", [KEY_EMAIL] = "Email", - [KEY_CHAT] = "Chat", [KEY_SEARCH] = "Search", - [KEY_CONNECT] = "Connect", [KEY_FINANCE] = "Finance", - [KEY_SPORT] = "Sport", [KEY_SHOP] = "Shop", - [KEY_ALTERASE] = "AlternateErase", [KEY_CANCEL] = "Cancel", - [KEY_BRIGHTNESSDOWN] = "BrightnessDown", [KEY_BRIGHTNESSUP] = "BrightnessUp", - [KEY_MEDIA] = "Media", [KEY_UNKNOWN] = "Unknown", - [BTN_0] = "Btn0", [BTN_1] = "Btn1", - [BTN_2] = "Btn2", [BTN_3] = "Btn3", - [BTN_4] = "Btn4", [BTN_5] = "Btn5", - [BTN_6] = "Btn6", [BTN_7] = "Btn7", - [BTN_8] = "Btn8", [BTN_9] = "Btn9", - [BTN_LEFT] = "LeftBtn", [BTN_RIGHT] = "RightBtn", - [BTN_MIDDLE] = "MiddleBtn", [BTN_SIDE] = "SideBtn", - [BTN_EXTRA] = "ExtraBtn", [BTN_FORWARD] = "ForwardBtn", - [BTN_BACK] = "BackBtn", [BTN_TASK] = "TaskBtn", - [BTN_TRIGGER] = "Trigger", [BTN_THUMB] = "ThumbBtn", - [BTN_THUMB2] = "ThumbBtn2", [BTN_TOP] = "TopBtn", - [BTN_TOP2] = "TopBtn2", [BTN_PINKIE] = "PinkieBtn", - [BTN_BASE] = "BaseBtn", [BTN_BASE2] = "BaseBtn2", - [BTN_BASE3] = "BaseBtn3", [BTN_BASE4] = "BaseBtn4", - [BTN_BASE5] = "BaseBtn5", [BTN_BASE6] = "BaseBtn6", - [BTN_DEAD] = "BtnDead", [BTN_A] = "BtnA", - [BTN_B] = "BtnB", [BTN_C] = "BtnC", - [BTN_X] = "BtnX", [BTN_Y] = "BtnY", - [BTN_Z] = "BtnZ", [BTN_TL] = "BtnTL", - [BTN_TR] = "BtnTR", [BTN_TL2] = "BtnTL2", - [BTN_TR2] = "BtnTR2", [BTN_SELECT] = "BtnSelect", - [BTN_START] = "BtnStart", [BTN_MODE] = "BtnMode", - [BTN_THUMBL] = "BtnThumbL", [BTN_THUMBR] = "BtnThumbR", - [BTN_TOOL_PEN] = "ToolPen", [BTN_TOOL_RUBBER] = "ToolRubber", - [BTN_TOOL_BRUSH] = "ToolBrush", [BTN_TOOL_PENCIL] = "ToolPencil", - [BTN_TOOL_AIRBRUSH] = "ToolAirbrush", [BTN_TOOL_FINGER] = "ToolFinger", - [BTN_TOOL_MOUSE] = "ToolMouse", [BTN_TOOL_LENS] = "ToolLens", - [BTN_TOUCH] = "Touch", [BTN_STYLUS] = "Stylus", - [BTN_STYLUS2] = "Stylus2", [BTN_TOOL_DOUBLETAP] = "ToolDoubleTap", - [BTN_TOOL_TRIPLETAP] = "ToolTripleTap", [BTN_GEAR_DOWN] = "WheelBtn", - [BTN_GEAR_UP] = "Gear up", [KEY_OK] = "Ok", - [KEY_SELECT] = "Select", [KEY_GOTO] = "Goto", - [KEY_CLEAR] = "Clear", [KEY_POWER2] = "Power2", - [KEY_OPTION] = "Option", [KEY_INFO] = "Info", - [KEY_TIME] = "Time", [KEY_VENDOR] = "Vendor", - [KEY_ARCHIVE] = "Archive", [KEY_PROGRAM] = "Program", - [KEY_CHANNEL] = "Channel", [KEY_FAVORITES] = "Favorites", - [KEY_EPG] = "EPG", [KEY_PVR] = "PVR", - [KEY_MHP] = "MHP", [KEY_LANGUAGE] = "Language", - [KEY_TITLE] = "Title", [KEY_SUBTITLE] = "Subtitle", - [KEY_ANGLE] = "Angle", [KEY_ZOOM] = "Zoom", - [KEY_MODE] = "Mode", [KEY_KEYBOARD] = "Keyboard", - [KEY_SCREEN] = "Screen", [KEY_PC] = "PC", - [KEY_TV] = "TV", [KEY_TV2] = "TV2", - [KEY_VCR] = "VCR", [KEY_VCR2] = "VCR2", - [KEY_SAT] = "Sat", [KEY_SAT2] = "Sat2", - [KEY_CD] = "CD", [KEY_TAPE] = "Tape", - [KEY_RADIO] = "Radio", [KEY_TUNER] = "Tuner", - [KEY_PLAYER] = "Player", [KEY_TEXT] = "Text", - [KEY_DVD] = "DVD", [KEY_AUX] = "Aux", - [KEY_MP3] = "MP3", [KEY_AUDIO] = "Audio", - [KEY_VIDEO] = "Video", [KEY_DIRECTORY] = "Directory", - [KEY_LIST] = "List", [KEY_MEMO] = "Memo", - [KEY_CALENDAR] = "Calendar", [KEY_RED] = "Red", - [KEY_GREEN] = "Green", [KEY_YELLOW] = "Yellow", - [KEY_BLUE] = "Blue", [KEY_CHANNELUP] = "ChannelUp", - [KEY_CHANNELDOWN] = "ChannelDown", [KEY_FIRST] = "First", - [KEY_LAST] = "Last", [KEY_AB] = "AB", - [KEY_NEXT] = "Next", [KEY_RESTART] = "Restart", - [KEY_SLOW] = "Slow", [KEY_SHUFFLE] = "Shuffle", - [KEY_BREAK] = "Break", [KEY_PREVIOUS] = "Previous", - [KEY_DIGITS] = "Digits", [KEY_TEEN] = "TEEN", - [KEY_TWEN] = "TWEN", [KEY_DEL_EOL] = "DeleteEOL", - [KEY_DEL_EOS] = "DeleteEOS", [KEY_INS_LINE] = "InsertLine", - [KEY_DEL_LINE] = "DeleteLine", - [KEY_SEND] = "Send", [KEY_REPLY] = "Reply", - [KEY_FORWARDMAIL] = "ForwardMail", [KEY_SAVE] = "Save", - [KEY_DOCUMENTS] = "Documents", - [KEY_FN] = "Fn", [KEY_FN_ESC] = "Fn+ESC", - [KEY_FN_1] = "Fn+1", [KEY_FN_2] = "Fn+2", - [KEY_FN_B] = "Fn+B", [KEY_FN_D] = "Fn+D", - [KEY_FN_E] = "Fn+E", [KEY_FN_F] = "Fn+F", - [KEY_FN_S] = "Fn+S", - [KEY_FN_F1] = "Fn+F1", [KEY_FN_F2] = "Fn+F2", - [KEY_FN_F3] = "Fn+F3", [KEY_FN_F4] = "Fn+F4", - [KEY_FN_F5] = "Fn+F5", [KEY_FN_F6] = "Fn+F6", - [KEY_FN_F7] = "Fn+F7", [KEY_FN_F8] = "Fn+F8", - [KEY_FN_F9] = "Fn+F9", [KEY_FN_F10] = "Fn+F10", - [KEY_FN_F11] = "Fn+F11", [KEY_FN_F12] = "Fn+F12", - [KEY_KBDILLUMTOGGLE] = "KbdIlluminationToggle", - [KEY_KBDILLUMDOWN] = "KbdIlluminationDown", - [KEY_KBDILLUMUP] = "KbdIlluminationUp", - [KEY_SWITCHVIDEOMODE] = "SwitchVideoMode", -}; - -static char *relatives[REL_MAX + 1] = { - [REL_X] = "X", [REL_Y] = "Y", - [REL_Z] = "Z", [REL_HWHEEL] = "HWheel", - [REL_DIAL] = "Dial", [REL_WHEEL] = "Wheel", - [REL_MISC] = "Misc", -}; - -static char *absolutes[ABS_MAX + 1] = { - [ABS_X] = "X", [ABS_Y] = "Y", - [ABS_Z] = "Z", [ABS_RX] = "Rx", - [ABS_RY] = "Ry", [ABS_RZ] = "Rz", - [ABS_THROTTLE] = "Throttle", [ABS_RUDDER] = "Rudder", - [ABS_WHEEL] = "Wheel", [ABS_GAS] = "Gas", - [ABS_BRAKE] = "Brake", [ABS_HAT0X] = "Hat0X", - [ABS_HAT0Y] = "Hat0Y", [ABS_HAT1X] = "Hat1X", - [ABS_HAT1Y] = "Hat1Y", [ABS_HAT2X] = "Hat2X", - [ABS_HAT2Y] = "Hat2Y", [ABS_HAT3X] = "Hat3X", - [ABS_HAT3Y] = "Hat 3Y", [ABS_PRESSURE] = "Pressure", - [ABS_DISTANCE] = "Distance", [ABS_TILT_X] = "XTilt", - [ABS_TILT_Y] = "YTilt", [ABS_TOOL_WIDTH] = "Tool Width", - [ABS_VOLUME] = "Volume", [ABS_MISC] = "Misc", -}; - -static char *misc[MSC_MAX + 1] = { - [MSC_SERIAL] = "Serial", [MSC_PULSELED] = "Pulseled", - [MSC_GESTURE] = "Gesture", [MSC_RAW] = "RawData" -}; - -static char *leds[LED_MAX + 1] = { - [LED_NUML] = "NumLock", [LED_CAPSL] = "CapsLock", - [LED_SCROLLL] = "ScrollLock", [LED_COMPOSE] = "Compose", - [LED_KANA] = "Kana", [LED_SLEEP] = "Sleep", - [LED_SUSPEND] = "Suspend", [LED_MUTE] = "Mute", - [LED_MISC] = "Misc", -}; - -static char *repeats[REP_MAX + 1] = { - [REP_DELAY] = "Delay", [REP_PERIOD] = "Period" -}; - -static char *sounds[SND_MAX + 1] = { - [SND_CLICK] = "Click", [SND_BELL] = "Bell", - [SND_TONE] = "Tone" -}; - -static char **names[EV_MAX + 1] = { - [EV_SYN] = syncs, [EV_KEY] = keys, - [EV_REL] = relatives, [EV_ABS] = absolutes, - [EV_MSC] = misc, [EV_LED] = leds, - [EV_SND] = sounds, [EV_REP] = repeats, -}; - -static void __attribute__((unused)) resolv_event(__u8 type, __u16 code) { - - printk("%s.%s", events[type] ? events[type] : "?", - names[type] ? (names[type][code] ? names[type][code] : "?") : "?"); -} diff --git a/drivers/usb/input/hid-ff.c b/drivers/usb/input/hid-ff.c index a8fc46c721c5..4187f4ee9a9c 100644 --- a/drivers/usb/input/hid-ff.c +++ b/drivers/usb/input/hid-ff.c @@ -32,7 +32,7 @@ #undef DEBUG #include -#include "hid.h" +#include /* * This table contains pointers to initializers. To add support for new diff --git a/drivers/usb/input/hid-input.c b/drivers/usb/input/hid-input.c index 3a7e5fbff025..8756eb3263de 100644 --- a/drivers/usb/input/hid-input.c +++ b/drivers/usb/input/hid-input.c @@ -4,6 +4,7 @@ * Copyright (c) 2000-2001 Vojtech Pavlik * * USB HID to Linux Input mapping + * */ /* @@ -26,847 +27,4 @@ * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic */ -#include -#include -#include -#include - -#undef DEBUG - -#include "hid.h" - -#define unk KEY_UNKNOWN - -static const unsigned char hid_keyboard[256] = { - 0, 0, 0, 0, 30, 48, 46, 32, 18, 33, 34, 35, 23, 36, 37, 38, - 50, 49, 24, 25, 16, 19, 31, 20, 22, 47, 17, 45, 21, 44, 2, 3, - 4, 5, 6, 7, 8, 9, 10, 11, 28, 1, 14, 15, 57, 12, 13, 26, - 27, 43, 43, 39, 40, 41, 51, 52, 53, 58, 59, 60, 61, 62, 63, 64, - 65, 66, 67, 68, 87, 88, 99, 70,119,110,102,104,111,107,109,106, - 105,108,103, 69, 98, 55, 74, 78, 96, 79, 80, 81, 75, 76, 77, 71, - 72, 73, 82, 83, 86,127,116,117,183,184,185,186,187,188,189,190, - 191,192,193,194,134,138,130,132,128,129,131,137,133,135,136,113, - 115,114,unk,unk,unk,121,unk, 89, 93,124, 92, 94, 95,unk,unk,unk, - 122,123, 90, 91, 85,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, - unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, - unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, - unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, - unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk, - 29, 42, 56,125, 97, 54,100,126,164,166,165,163,161,115,114,113, - 150,158,159,128,136,177,178,176,142,152,173,140,unk,unk,unk,unk -}; - -static const struct { - __s32 x; - __s32 y; -} hid_hat_to_axis[] = {{ 0, 0}, { 0,-1}, { 1,-1}, { 1, 0}, { 1, 1}, { 0, 1}, {-1, 1}, {-1, 0}, {-1,-1}}; - -#define map_abs(c) do { usage->code = c; usage->type = EV_ABS; bit = input->absbit; max = ABS_MAX; } while (0) -#define map_rel(c) do { usage->code = c; usage->type = EV_REL; bit = input->relbit; max = REL_MAX; } while (0) -#define map_key(c) do { usage->code = c; usage->type = EV_KEY; bit = input->keybit; max = KEY_MAX; } while (0) -#define map_led(c) do { usage->code = c; usage->type = EV_LED; bit = input->ledbit; max = LED_MAX; } while (0) - -#define map_abs_clear(c) do { map_abs(c); clear_bit(c, bit); } while (0) -#define map_key_clear(c) do { map_key(c); clear_bit(c, bit); } while (0) - -#ifdef CONFIG_USB_HIDINPUT_POWERBOOK - -struct hidinput_key_translation { - u16 from; - u16 to; - u8 flags; -}; - -#define POWERBOOK_FLAG_FKEY 0x01 - -static struct hidinput_key_translation powerbook_fn_keys[] = { - { KEY_BACKSPACE, KEY_DELETE }, - { KEY_F1, KEY_BRIGHTNESSDOWN, POWERBOOK_FLAG_FKEY }, - { KEY_F2, KEY_BRIGHTNESSUP, POWERBOOK_FLAG_FKEY }, - { KEY_F3, KEY_MUTE, POWERBOOK_FLAG_FKEY }, - { KEY_F4, KEY_VOLUMEDOWN, POWERBOOK_FLAG_FKEY }, - { KEY_F5, KEY_VOLUMEUP, POWERBOOK_FLAG_FKEY }, - { KEY_F6, KEY_NUMLOCK, POWERBOOK_FLAG_FKEY }, - { KEY_F7, KEY_SWITCHVIDEOMODE, POWERBOOK_FLAG_FKEY }, - { KEY_F8, KEY_KBDILLUMTOGGLE, POWERBOOK_FLAG_FKEY }, - { KEY_F9, KEY_KBDILLUMDOWN, POWERBOOK_FLAG_FKEY }, - { KEY_F10, KEY_KBDILLUMUP, POWERBOOK_FLAG_FKEY }, - { KEY_UP, KEY_PAGEUP }, - { KEY_DOWN, KEY_PAGEDOWN }, - { KEY_LEFT, KEY_HOME }, - { KEY_RIGHT, KEY_END }, - { } -}; - -static struct hidinput_key_translation powerbook_numlock_keys[] = { - { KEY_J, KEY_KP1 }, - { KEY_K, KEY_KP2 }, - { KEY_L, KEY_KP3 }, - { KEY_U, KEY_KP4 }, - { KEY_I, KEY_KP5 }, - { KEY_O, KEY_KP6 }, - { KEY_7, KEY_KP7 }, - { KEY_8, KEY_KP8 }, - { KEY_9, KEY_KP9 }, - { KEY_M, KEY_KP0 }, - { KEY_DOT, KEY_KPDOT }, - { KEY_SLASH, KEY_KPPLUS }, - { KEY_SEMICOLON, KEY_KPMINUS }, - { KEY_P, KEY_KPASTERISK }, - { KEY_MINUS, KEY_KPEQUAL }, - { KEY_0, KEY_KPSLASH }, - { KEY_F6, KEY_NUMLOCK }, - { KEY_KPENTER, KEY_KPENTER }, - { KEY_BACKSPACE, KEY_BACKSPACE }, - { } -}; - -static struct hidinput_key_translation powerbook_iso_keyboard[] = { - { KEY_GRAVE, KEY_102ND }, - { KEY_102ND, KEY_GRAVE }, - { } -}; - -static int usbhid_pb_fnmode = 1; -module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); -MODULE_PARM_DESC(pb_fnmode, - "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); - -static struct hidinput_key_translation *find_translation(struct hidinput_key_translation *table, u16 from) -{ - struct hidinput_key_translation *trans; - - /* Look for the translation */ - for (trans = table; trans->from; trans++) - if (trans->from == from) - return trans; - - return NULL; -} - -static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, - struct hid_usage *usage, __s32 value) -{ - struct hidinput_key_translation *trans; - - if (usage->code == KEY_FN) { - if (value) hid->quirks |= HID_QUIRK_POWERBOOK_FN_ON; - else hid->quirks &= ~HID_QUIRK_POWERBOOK_FN_ON; - - input_event(input, usage->type, usage->code, value); - - return 1; - } - - if (usbhid_pb_fnmode) { - int do_translate; - - trans = find_translation(powerbook_fn_keys, usage->code); - if (trans) { - if (test_bit(usage->code, hid->pb_pressed_fn)) - do_translate = 1; - else if (trans->flags & POWERBOOK_FLAG_FKEY) - do_translate = - (usbhid_pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || - (usbhid_pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); - else - do_translate = (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON); - - if (do_translate) { - if (value) - set_bit(usage->code, hid->pb_pressed_fn); - else - clear_bit(usage->code, hid->pb_pressed_fn); - - input_event(input, usage->type, trans->to, value); - - return 1; - } - } - - if (test_bit(usage->code, hid->pb_pressed_numlock) || - test_bit(LED_NUML, input->led)) { - trans = find_translation(powerbook_numlock_keys, usage->code); - - if (trans) { - if (value) - set_bit(usage->code, hid->pb_pressed_numlock); - else - clear_bit(usage->code, hid->pb_pressed_numlock); - - input_event(input, usage->type, trans->to, value); - } - - return 1; - } - } - - if (hid->quirks & HID_QUIRK_POWERBOOK_ISO_KEYBOARD) { - trans = find_translation(powerbook_iso_keyboard, usage->code); - if (trans) { - input_event(input, usage->type, trans->to, value); - return 1; - } - } - - return 0; -} - -static void hidinput_pb_setup(struct input_dev *input) -{ - struct hidinput_key_translation *trans; - - set_bit(KEY_NUMLOCK, input->keybit); - - /* Enable all needed keys */ - for (trans = powerbook_fn_keys; trans->from; trans++) - set_bit(trans->to, input->keybit); - - for (trans = powerbook_numlock_keys; trans->from; trans++) - set_bit(trans->to, input->keybit); - - for (trans = powerbook_iso_keyboard; trans->from; trans++) - set_bit(trans->to, input->keybit); -} -#else -static inline int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, - struct hid_usage *usage, __s32 value) -{ - return 0; -} - -static inline void hidinput_pb_setup(struct input_dev *input) -{ -} -#endif - -static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_field *field, - struct hid_usage *usage) -{ - struct input_dev *input = hidinput->input; - struct hid_device *device = input->private; - int max = 0, code; - unsigned long *bit = NULL; - - field->hidinput = hidinput; - -#ifdef DEBUG - printk(KERN_DEBUG "Mapping: "); - resolv_usage(usage->hid); - printk(" ---> "); -#endif - - if (field->flags & HID_MAIN_ITEM_CONSTANT) - goto ignore; - - switch (usage->hid & HID_USAGE_PAGE) { - - case HID_UP_UNDEFINED: - goto ignore; - - case HID_UP_KEYBOARD: - - set_bit(EV_REP, input->evbit); - - if ((usage->hid & HID_USAGE) < 256) { - if (!hid_keyboard[usage->hid & HID_USAGE]) goto ignore; - map_key_clear(hid_keyboard[usage->hid & HID_USAGE]); - } else - map_key(KEY_UNKNOWN); - - break; - - case HID_UP_BUTTON: - - code = ((usage->hid - 1) & 0xf); - - switch (field->application) { - case HID_GD_MOUSE: - case HID_GD_POINTER: code += 0x110; break; - case HID_GD_JOYSTICK: code += 0x120; break; - case HID_GD_GAMEPAD: code += 0x130; break; - default: - switch (field->physical) { - case HID_GD_MOUSE: - case HID_GD_POINTER: code += 0x110; break; - case HID_GD_JOYSTICK: code += 0x120; break; - case HID_GD_GAMEPAD: code += 0x130; break; - default: code += 0x100; - } - } - - map_key(code); - break; - - - case HID_UP_SIMULATION: - - switch (usage->hid & 0xffff) { - case 0xba: map_abs(ABS_RUDDER); break; - case 0xbb: map_abs(ABS_THROTTLE); break; - case 0xc4: map_abs(ABS_GAS); break; - case 0xc5: map_abs(ABS_BRAKE); break; - case 0xc8: map_abs(ABS_WHEEL); break; - default: goto ignore; - } - break; - - case HID_UP_GENDESK: - - if ((usage->hid & 0xf0) == 0x80) { /* SystemControl */ - switch (usage->hid & 0xf) { - case 0x1: map_key_clear(KEY_POWER); break; - case 0x2: map_key_clear(KEY_SLEEP); break; - case 0x3: map_key_clear(KEY_WAKEUP); break; - default: goto unknown; - } - break; - } - - if ((usage->hid & 0xf0) == 0x90) { /* D-pad */ - switch (usage->hid) { - case HID_GD_UP: usage->hat_dir = 1; break; - case HID_GD_DOWN: usage->hat_dir = 5; break; - case HID_GD_RIGHT: usage->hat_dir = 3; break; - case HID_GD_LEFT: usage->hat_dir = 7; break; - default: goto unknown; - } - if (field->dpad) { - map_abs(field->dpad); - goto ignore; - } - map_abs(ABS_HAT0X); - break; - } - - switch (usage->hid) { - - /* These usage IDs map directly to the usage codes. */ - case HID_GD_X: case HID_GD_Y: case HID_GD_Z: - case HID_GD_RX: case HID_GD_RY: case HID_GD_RZ: - case HID_GD_SLIDER: case HID_GD_DIAL: case HID_GD_WHEEL: - if (field->flags & HID_MAIN_ITEM_RELATIVE) - map_rel(usage->hid & 0xf); - else - map_abs(usage->hid & 0xf); - break; - - case HID_GD_HATSWITCH: - usage->hat_min = field->logical_minimum; - usage->hat_max = field->logical_maximum; - map_abs(ABS_HAT0X); - break; - - case HID_GD_START: map_key_clear(BTN_START); break; - case HID_GD_SELECT: map_key_clear(BTN_SELECT); break; - - default: goto unknown; - } - - break; - - case HID_UP_LED: - if (((usage->hid - 1) & 0xffff) >= LED_MAX) - goto ignore; - map_led((usage->hid - 1) & 0xffff); - break; - - case HID_UP_DIGITIZER: - - switch (usage->hid & 0xff) { - - case 0x30: /* TipPressure */ - if (!test_bit(BTN_TOUCH, input->keybit)) { - device->quirks |= HID_QUIRK_NOTOUCH; - set_bit(EV_KEY, input->evbit); - set_bit(BTN_TOUCH, input->keybit); - } - - map_abs_clear(ABS_PRESSURE); - break; - - case 0x32: /* InRange */ - switch (field->physical & 0xff) { - case 0x21: map_key(BTN_TOOL_MOUSE); break; - case 0x22: map_key(BTN_TOOL_FINGER); break; - default: map_key(BTN_TOOL_PEN); break; - } - break; - - case 0x3c: /* Invert */ - map_key_clear(BTN_TOOL_RUBBER); - break; - - case 0x33: /* Touch */ - case 0x42: /* TipSwitch */ - case 0x43: /* TipSwitch2 */ - device->quirks &= ~HID_QUIRK_NOTOUCH; - map_key_clear(BTN_TOUCH); - break; - - case 0x44: /* BarrelSwitch */ - map_key_clear(BTN_STYLUS); - break; - - default: goto unknown; - } - break; - - case HID_UP_CONSUMER: /* USB HUT v1.1, pages 56-62 */ - - switch (usage->hid & HID_USAGE) { - case 0x000: goto ignore; - case 0x034: map_key_clear(KEY_SLEEP); break; - case 0x036: map_key_clear(BTN_MISC); break; - case 0x045: map_key_clear(KEY_RADIO); break; - case 0x08a: map_key_clear(KEY_WWW); break; - case 0x08d: map_key_clear(KEY_PROGRAM); break; - case 0x095: map_key_clear(KEY_HELP); break; - case 0x09c: map_key_clear(KEY_CHANNELUP); break; - case 0x09d: map_key_clear(KEY_CHANNELDOWN); break; - case 0x0b0: map_key_clear(KEY_PLAY); break; - case 0x0b1: map_key_clear(KEY_PAUSE); break; - case 0x0b2: map_key_clear(KEY_RECORD); break; - case 0x0b3: map_key_clear(KEY_FASTFORWARD); break; - case 0x0b4: map_key_clear(KEY_REWIND); break; - case 0x0b5: map_key_clear(KEY_NEXTSONG); break; - case 0x0b6: map_key_clear(KEY_PREVIOUSSONG); break; - case 0x0b7: map_key_clear(KEY_STOPCD); break; - case 0x0b8: map_key_clear(KEY_EJECTCD); break; - case 0x0cd: map_key_clear(KEY_PLAYPAUSE); break; - case 0x0e0: map_abs_clear(ABS_VOLUME); break; - case 0x0e2: map_key_clear(KEY_MUTE); break; - case 0x0e5: map_key_clear(KEY_BASSBOOST); break; - case 0x0e9: map_key_clear(KEY_VOLUMEUP); break; - case 0x0ea: map_key_clear(KEY_VOLUMEDOWN); break; - case 0x183: map_key_clear(KEY_CONFIG); break; - case 0x18a: map_key_clear(KEY_MAIL); break; - case 0x192: map_key_clear(KEY_CALC); break; - case 0x194: map_key_clear(KEY_FILE); break; - case 0x1a7: map_key_clear(KEY_DOCUMENTS); break; - case 0x201: map_key_clear(KEY_NEW); break; - case 0x207: map_key_clear(KEY_SAVE); break; - case 0x208: map_key_clear(KEY_PRINT); break; - case 0x209: map_key_clear(KEY_PROPS); break; - case 0x21a: map_key_clear(KEY_UNDO); break; - case 0x21b: map_key_clear(KEY_COPY); break; - case 0x21c: map_key_clear(KEY_CUT); break; - case 0x21d: map_key_clear(KEY_PASTE); break; - case 0x221: map_key_clear(KEY_FIND); break; - case 0x223: map_key_clear(KEY_HOMEPAGE); break; - case 0x224: map_key_clear(KEY_BACK); break; - case 0x225: map_key_clear(KEY_FORWARD); break; - case 0x226: map_key_clear(KEY_STOP); break; - case 0x227: map_key_clear(KEY_REFRESH); break; - case 0x22a: map_key_clear(KEY_BOOKMARKS); break; - case 0x233: map_key_clear(KEY_SCROLLUP); break; - case 0x234: map_key_clear(KEY_SCROLLDOWN); break; - case 0x238: map_rel(REL_HWHEEL); break; - case 0x279: map_key_clear(KEY_REDO); break; - case 0x289: map_key_clear(KEY_REPLY); break; - case 0x28b: map_key_clear(KEY_FORWARDMAIL); break; - case 0x28c: map_key_clear(KEY_SEND); break; - - /* Reported on a Cherry Cymotion keyboard */ - case 0x301: map_key_clear(KEY_PROG1); break; - case 0x302: map_key_clear(KEY_PROG2); break; - case 0x303: map_key_clear(KEY_PROG3); break; - - default: goto ignore; - } - break; - - case HID_UP_HPVENDOR: /* Reported on a Dutch layout HP5308 */ - - set_bit(EV_REP, input->evbit); - switch (usage->hid & HID_USAGE) { - case 0x021: map_key_clear(KEY_PRINT); break; - case 0x070: map_key_clear(KEY_HP); break; - case 0x071: map_key_clear(KEY_CAMERA); break; - case 0x072: map_key_clear(KEY_SOUND); break; - case 0x073: map_key_clear(KEY_QUESTION); break; - case 0x080: map_key_clear(KEY_EMAIL); break; - case 0x081: map_key_clear(KEY_CHAT); break; - case 0x082: map_key_clear(KEY_SEARCH); break; - case 0x083: map_key_clear(KEY_CONNECT); break; - case 0x084: map_key_clear(KEY_FINANCE); break; - case 0x085: map_key_clear(KEY_SPORT); break; - case 0x086: map_key_clear(KEY_SHOP); break; - default: goto ignore; - } - break; - - case HID_UP_MSVENDOR: - goto ignore; - - case HID_UP_CUSTOM: /* Reported on Logitech and Powerbook USB keyboards */ - - set_bit(EV_REP, input->evbit); - switch(usage->hid & HID_USAGE) { - case 0x003: - /* The fn key on Apple PowerBooks */ - map_key_clear(KEY_FN); - hidinput_pb_setup(input); - break; - - default: goto ignore; - } - break; - - case HID_UP_LOGIVENDOR: /* Reported on Logitech Ultra X Media Remote */ - - set_bit(EV_REP, input->evbit); - switch(usage->hid & HID_USAGE) { - case 0x004: map_key_clear(KEY_AGAIN); break; - case 0x00d: map_key_clear(KEY_HOME); break; - case 0x024: map_key_clear(KEY_SHUFFLE); break; - case 0x025: map_key_clear(KEY_TV); break; - case 0x026: map_key_clear(KEY_MENU); break; - case 0x031: map_key_clear(KEY_AUDIO); break; - case 0x032: map_key_clear(KEY_TEXT); break; - case 0x033: map_key_clear(KEY_LAST); break; - case 0x047: map_key_clear(KEY_MP3); break; - case 0x048: map_key_clear(KEY_DVD); break; - case 0x049: map_key_clear(KEY_MEDIA); break; - case 0x04a: map_key_clear(KEY_VIDEO); break; - case 0x04b: map_key_clear(KEY_ANGLE); break; - case 0x04c: map_key_clear(KEY_LANGUAGE); break; - case 0x04d: map_key_clear(KEY_SUBTITLE); break; - case 0x051: map_key_clear(KEY_RED); break; - case 0x052: map_key_clear(KEY_CLOSE); break; - default: goto ignore; - } - break; - - case HID_UP_PID: - - switch(usage->hid & HID_USAGE) { - case 0xa4: map_key_clear(BTN_DEAD); break; - default: goto ignore; - } - break; - - default: - unknown: - if (field->report_size == 1) { - if (field->report->type == HID_OUTPUT_REPORT) { - map_led(LED_MISC); - break; - } - map_key(BTN_MISC); - break; - } - if (field->flags & HID_MAIN_ITEM_RELATIVE) { - map_rel(REL_MISC); - break; - } - map_abs(ABS_MISC); - break; - } - - if (device->quirks & HID_QUIRK_MIGHTYMOUSE) { - if (usage->hid == HID_GD_Z) - map_rel(REL_HWHEEL); - else if (usage->code == BTN_1) - map_key(BTN_2); - else if (usage->code == BTN_2) - map_key(BTN_1); - } - - if ((device->quirks & (HID_QUIRK_2WHEEL_MOUSE_HACK_7 | HID_QUIRK_2WHEEL_MOUSE_HACK_5)) && - (usage->type == EV_REL) && (usage->code == REL_WHEEL)) - set_bit(REL_HWHEEL, bit); - - if (((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005)) - || ((device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) - goto ignore; - - if ((device->quirks & HID_QUIRK_BAD_RELATIVE_KEYS) && - usage->type == EV_KEY && (field->flags & HID_MAIN_ITEM_RELATIVE)) - field->flags &= ~HID_MAIN_ITEM_RELATIVE; - - set_bit(usage->type, input->evbit); - - while (usage->code <= max && test_and_set_bit(usage->code, bit)) - usage->code = find_next_zero_bit(bit, max + 1, usage->code); - - if (usage->code > max) - goto ignore; - - - if (usage->type == EV_ABS) { - - int a = field->logical_minimum; - int b = field->logical_maximum; - - if ((device->quirks & HID_QUIRK_BADPAD) && (usage->code == ABS_X || usage->code == ABS_Y)) { - a = field->logical_minimum = 0; - b = field->logical_maximum = 255; - } - - if (field->application == HID_GD_GAMEPAD || field->application == HID_GD_JOYSTICK) - input_set_abs_params(input, usage->code, a, b, (b - a) >> 8, (b - a) >> 4); - else input_set_abs_params(input, usage->code, a, b, 0, 0); - - } - - if (usage->type == EV_ABS && - (usage->hat_min < usage->hat_max || usage->hat_dir)) { - int i; - for (i = usage->code; i < usage->code + 2 && i <= max; i++) { - input_set_abs_params(input, i, -1, 1, 0, 0); - set_bit(i, input->absbit); - } - if (usage->hat_dir && !field->dpad) - field->dpad = usage->code; - } - -#ifdef DEBUG - resolv_event(usage->type, usage->code); - printk("\n"); -#endif - return; - -ignore: -#ifdef DEBUG - printk("IGNORED\n"); -#endif - return; -} - -void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value) -{ - struct input_dev *input; - int *quirks = &hid->quirks; - - if (!field->hidinput) - return; - - input = field->hidinput->input; - - if (!usage->type) - return; - - if (((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_5) && (usage->hid == 0x00090005)) - || ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_7) && (usage->hid == 0x00090007))) { - if (value) hid->quirks |= HID_QUIRK_2WHEEL_MOUSE_HACK_ON; - else hid->quirks &= ~HID_QUIRK_2WHEEL_MOUSE_HACK_ON; - return; - } - - if ((hid->quirks & HID_QUIRK_INVERT_HWHEEL) && (usage->code == REL_HWHEEL)) { - input_event(input, usage->type, usage->code, -value); - return; - } - - if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_ON) && (usage->code == REL_WHEEL)) { - input_event(input, usage->type, REL_HWHEEL, value); - return; - } - - if ((hid->quirks & HID_QUIRK_POWERBOOK_HAS_FN) && hidinput_pb_event(hid, input, usage, value)) - return; - - if (usage->hat_min < usage->hat_max || usage->hat_dir) { - int hat_dir = usage->hat_dir; - if (!hat_dir) - hat_dir = (value - usage->hat_min) * 8 / (usage->hat_max - usage->hat_min + 1) + 1; - if (hat_dir < 0 || hat_dir > 8) hat_dir = 0; - input_event(input, usage->type, usage->code , hid_hat_to_axis[hat_dir].x); - input_event(input, usage->type, usage->code + 1, hid_hat_to_axis[hat_dir].y); - return; - } - - if (usage->hid == (HID_UP_DIGITIZER | 0x003c)) { /* Invert */ - *quirks = value ? (*quirks | HID_QUIRK_INVERT) : (*quirks & ~HID_QUIRK_INVERT); - return; - } - - if (usage->hid == (HID_UP_DIGITIZER | 0x0032)) { /* InRange */ - if (value) { - input_event(input, usage->type, (*quirks & HID_QUIRK_INVERT) ? BTN_TOOL_RUBBER : usage->code, 1); - return; - } - input_event(input, usage->type, usage->code, 0); - input_event(input, usage->type, BTN_TOOL_RUBBER, 0); - return; - } - - if (usage->hid == (HID_UP_DIGITIZER | 0x0030) && (*quirks & HID_QUIRK_NOTOUCH)) { /* Pressure */ - int a = field->logical_minimum; - int b = field->logical_maximum; - input_event(input, EV_KEY, BTN_TOUCH, value > a + ((b - a) >> 3)); - } - - if (usage->hid == (HID_UP_PID | 0x83UL)) { /* Simultaneous Effects Max */ - dbg("Maximum Effects - %d",value); - return; - } - - if (usage->hid == (HID_UP_PID | 0x7fUL)) { - dbg("PID Pool Report\n"); - return; - } - - if ((usage->type == EV_KEY) && (usage->code == 0)) /* Key 0 is "unassigned", not KEY_UNKNOWN */ - return; - - input_event(input, usage->type, usage->code, value); - - if ((field->flags & HID_MAIN_ITEM_RELATIVE) && (usage->type == EV_KEY)) - input_event(input, usage->type, usage->code, 0); -} - -void hidinput_report_event(struct hid_device *hid, struct hid_report *report) -{ - struct hid_input *hidinput; - - list_for_each_entry(hidinput, &hid->inputs, list) - input_sync(hidinput->input); -} - -static int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field) -{ - struct hid_report *report; - int i, j; - - list_for_each_entry(report, &hid->report_enum[HID_OUTPUT_REPORT].report_list, list) { - for (i = 0; i < report->maxfield; i++) { - *field = report->field[i]; - for (j = 0; j < (*field)->maxusage; j++) - if ((*field)->usage[j].type == type && (*field)->usage[j].code == code) - return j; - } - } - return -1; -} - -static int hidinput_input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) -{ - struct hid_device *hid = dev->private; - struct hid_field *field; - int offset; - - if (type == EV_FF) - return input_ff_event(dev, type, code, value); - - if (type != EV_LED) - return -1; - - if ((offset = hidinput_find_field(hid, type, code, &field)) == -1) { - warn("event field not found"); - return -1; - } - - hid_set_field(field, offset, value); - hid_submit_report(hid, field->report, USB_DIR_OUT); - - return 0; -} - -static int hidinput_open(struct input_dev *dev) -{ - struct hid_device *hid = dev->private; - return hid_open(hid); -} - -static void hidinput_close(struct input_dev *dev) -{ - struct hid_device *hid = dev->private; - hid_close(hid); -} - -/* - * Register the input device; print a message. - * Configure the input layer interface - * Read all reports and initialize the absolute field values. - */ - -int hidinput_connect(struct hid_device *hid) -{ - struct usb_device *dev = hid->dev; - struct hid_report *report; - struct hid_input *hidinput = NULL; - struct input_dev *input_dev; - int i, j, k; - - INIT_LIST_HEAD(&hid->inputs); - - for (i = 0; i < hid->maxcollection; i++) - if (hid->collection[i].type == HID_COLLECTION_APPLICATION || - hid->collection[i].type == HID_COLLECTION_PHYSICAL) - if (IS_INPUT_APPLICATION(hid->collection[i].usage)) - break; - - if (i == hid->maxcollection) - return -1; - - for (k = HID_INPUT_REPORT; k <= HID_OUTPUT_REPORT; k++) - list_for_each_entry(report, &hid->report_enum[k].report_list, list) { - - if (!report->maxfield) - continue; - - if (!hidinput) { - hidinput = kzalloc(sizeof(*hidinput), GFP_KERNEL); - input_dev = input_allocate_device(); - if (!hidinput || !input_dev) { - kfree(hidinput); - input_free_device(input_dev); - err("Out of memory during hid input probe"); - return -1; - } - - input_dev->private = hid; - input_dev->event = hidinput_input_event; - input_dev->open = hidinput_open; - input_dev->close = hidinput_close; - - input_dev->name = hid->name; - input_dev->phys = hid->phys; - input_dev->uniq = hid->uniq; - usb_to_input_id(dev, &input_dev->id); - input_dev->cdev.dev = &hid->intf->dev; - - hidinput->input = input_dev; - list_add_tail(&hidinput->list, &hid->inputs); - } - - for (i = 0; i < report->maxfield; i++) - for (j = 0; j < report->field[i]->maxusage; j++) - hidinput_configure_usage(hidinput, report->field[i], - report->field[i]->usage + j); - - if (hid->quirks & HID_QUIRK_MULTI_INPUT) { - /* This will leave hidinput NULL, so that it - * allocates another one if we have more inputs on - * the same interface. Some devices (e.g. Happ's - * UGCI) cram a lot of unrelated inputs into the - * same interface. */ - hidinput->report = report; - input_register_device(hidinput->input); - hidinput = NULL; - } - } - - /* This only gets called when we are a single-input (most of the - * time). IOW, not a HID_QUIRK_MULTI_INPUT. The hid_ff_init() is - * only useful in this case, and not for multi-input quirks. */ - if (hidinput) { - hid_ff_init(hid); - input_register_device(hidinput->input); - } - - return 0; -} - -void hidinput_disconnect(struct hid_device *hid) -{ - struct hid_input *hidinput, *next; - list_for_each_entry_safe(hidinput, next, &hid->inputs, list) { - list_del(&hidinput->list); - input_unregister_device(hidinput->input); - kfree(hidinput); - } -} diff --git a/drivers/usb/input/hid-lgff.c b/drivers/usb/input/hid-lgff.c index 93da222b6da8..e977ba3d17e0 100644 --- a/drivers/usb/input/hid-lgff.c +++ b/drivers/usb/input/hid-lgff.c @@ -29,7 +29,7 @@ #include #include -#include "hid.h" +#include struct device_type { u16 idVendor; diff --git a/drivers/usb/input/hid-pidff.c b/drivers/usb/input/hid-pidff.c index 5420c13eb8eb..b4caea3864e3 100644 --- a/drivers/usb/input/hid-pidff.c +++ b/drivers/usb/input/hid-pidff.c @@ -28,7 +28,9 @@ #include #include -#include "hid.h" +#include + +#include "usbhid.h" #define PID_EFFECTS_MAX 64 diff --git a/drivers/usb/input/hid-tmff.c b/drivers/usb/input/hid-tmff.c index 2d5be4c318ac..1cd1418ad6ac 100644 --- a/drivers/usb/input/hid-tmff.c +++ b/drivers/usb/input/hid-tmff.c @@ -32,7 +32,7 @@ #undef DEBUG #include -#include "hid.h" +#include /* Usages for thrustmaster devices I know about */ #define THRUSTMASTER_USAGE_RUMBLE_LR (HID_UP_GENDESK | 0xbb) diff --git a/drivers/usb/input/hid-zpff.c b/drivers/usb/input/hid-zpff.c index d2ce3214572c..af1bfae39dce 100644 --- a/drivers/usb/input/hid-zpff.c +++ b/drivers/usb/input/hid-zpff.c @@ -27,7 +27,7 @@ #include #include -#include "hid.h" +#include struct zpff_device { struct hid_report *report; diff --git a/drivers/usb/input/hid.h b/drivers/usb/input/hid.h deleted file mode 100644 index 76ad68d9edfd..000000000000 --- a/drivers/usb/input/hid.h +++ /dev/null @@ -1,540 +0,0 @@ -#ifndef __HID_H -#define __HID_H - -/* - * $Id: hid.h,v 1.24 2001/12/27 10:37:41 vojtech Exp $ - * - * Copyright (c) 1999 Andreas Gal - * Copyright (c) 2000-2001 Vojtech Pavlik - */ - -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Should you need to contact me, the author, you can do so either by - * e-mail - mail your message to , or by paper mail: - * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic - */ - -#include -#include -#include -#include -#include - -/* - * USB HID (Human Interface Device) interface class code - */ - -#define USB_INTERFACE_CLASS_HID 3 - -/* - * USB HID interface subclass and protocol codes - */ - -#define USB_INTERFACE_SUBCLASS_BOOT 1 -#define USB_INTERFACE_PROTOCOL_KEYBOARD 1 -#define USB_INTERFACE_PROTOCOL_MOUSE 2 - -/* - * HID class requests - */ - -#define HID_REQ_GET_REPORT 0x01 -#define HID_REQ_GET_IDLE 0x02 -#define HID_REQ_GET_PROTOCOL 0x03 -#define HID_REQ_SET_REPORT 0x09 -#define HID_REQ_SET_IDLE 0x0A -#define HID_REQ_SET_PROTOCOL 0x0B - -/* - * HID class descriptor types - */ - -#define HID_DT_HID (USB_TYPE_CLASS | 0x01) -#define HID_DT_REPORT (USB_TYPE_CLASS | 0x02) -#define HID_DT_PHYSICAL (USB_TYPE_CLASS | 0x03) - -/* - * We parse each description item into this structure. Short items data - * values are expanded to 32-bit signed int, long items contain a pointer - * into the data area. - */ - -struct hid_item { - unsigned format; - __u8 size; - __u8 type; - __u8 tag; - union { - __u8 u8; - __s8 s8; - __u16 u16; - __s16 s16; - __u32 u32; - __s32 s32; - __u8 *longdata; - } data; -}; - -/* - * HID report item format - */ - -#define HID_ITEM_FORMAT_SHORT 0 -#define HID_ITEM_FORMAT_LONG 1 - -/* - * Special tag indicating long items - */ - -#define HID_ITEM_TAG_LONG 15 - -/* - * HID report descriptor item type (prefix bit 2,3) - */ - -#define HID_ITEM_TYPE_MAIN 0 -#define HID_ITEM_TYPE_GLOBAL 1 -#define HID_ITEM_TYPE_LOCAL 2 -#define HID_ITEM_TYPE_RESERVED 3 - -/* - * HID report descriptor main item tags - */ - -#define HID_MAIN_ITEM_TAG_INPUT 8 -#define HID_MAIN_ITEM_TAG_OUTPUT 9 -#define HID_MAIN_ITEM_TAG_FEATURE 11 -#define HID_MAIN_ITEM_TAG_BEGIN_COLLECTION 10 -#define HID_MAIN_ITEM_TAG_END_COLLECTION 12 - -/* - * HID report descriptor main item contents - */ - -#define HID_MAIN_ITEM_CONSTANT 0x001 -#define HID_MAIN_ITEM_VARIABLE 0x002 -#define HID_MAIN_ITEM_RELATIVE 0x004 -#define HID_MAIN_ITEM_WRAP 0x008 -#define HID_MAIN_ITEM_NONLINEAR 0x010 -#define HID_MAIN_ITEM_NO_PREFERRED 0x020 -#define HID_MAIN_ITEM_NULL_STATE 0x040 -#define HID_MAIN_ITEM_VOLATILE 0x080 -#define HID_MAIN_ITEM_BUFFERED_BYTE 0x100 - -/* - * HID report descriptor collection item types - */ - -#define HID_COLLECTION_PHYSICAL 0 -#define HID_COLLECTION_APPLICATION 1 -#define HID_COLLECTION_LOGICAL 2 - -/* - * HID report descriptor global item tags - */ - -#define HID_GLOBAL_ITEM_TAG_USAGE_PAGE 0 -#define HID_GLOBAL_ITEM_TAG_LOGICAL_MINIMUM 1 -#define HID_GLOBAL_ITEM_TAG_LOGICAL_MAXIMUM 2 -#define HID_GLOBAL_ITEM_TAG_PHYSICAL_MINIMUM 3 -#define HID_GLOBAL_ITEM_TAG_PHYSICAL_MAXIMUM 4 -#define HID_GLOBAL_ITEM_TAG_UNIT_EXPONENT 5 -#define HID_GLOBAL_ITEM_TAG_UNIT 6 -#define HID_GLOBAL_ITEM_TAG_REPORT_SIZE 7 -#define HID_GLOBAL_ITEM_TAG_REPORT_ID 8 -#define HID_GLOBAL_ITEM_TAG_REPORT_COUNT 9 -#define HID_GLOBAL_ITEM_TAG_PUSH 10 -#define HID_GLOBAL_ITEM_TAG_POP 11 - -/* - * HID report descriptor local item tags - */ - -#define HID_LOCAL_ITEM_TAG_USAGE 0 -#define HID_LOCAL_ITEM_TAG_USAGE_MINIMUM 1 -#define HID_LOCAL_ITEM_TAG_USAGE_MAXIMUM 2 -#define HID_LOCAL_ITEM_TAG_DESIGNATOR_INDEX 3 -#define HID_LOCAL_ITEM_TAG_DESIGNATOR_MINIMUM 4 -#define HID_LOCAL_ITEM_TAG_DESIGNATOR_MAXIMUM 5 -#define HID_LOCAL_ITEM_TAG_STRING_INDEX 7 -#define HID_LOCAL_ITEM_TAG_STRING_MINIMUM 8 -#define HID_LOCAL_ITEM_TAG_STRING_MAXIMUM 9 -#define HID_LOCAL_ITEM_TAG_DELIMITER 10 - -/* - * HID usage tables - */ - -#define HID_USAGE_PAGE 0xffff0000 - -#define HID_UP_UNDEFINED 0x00000000 -#define HID_UP_GENDESK 0x00010000 -#define HID_UP_SIMULATION 0x00020000 -#define HID_UP_KEYBOARD 0x00070000 -#define HID_UP_LED 0x00080000 -#define HID_UP_BUTTON 0x00090000 -#define HID_UP_ORDINAL 0x000a0000 -#define HID_UP_CONSUMER 0x000c0000 -#define HID_UP_DIGITIZER 0x000d0000 -#define HID_UP_PID 0x000f0000 -#define HID_UP_HPVENDOR 0xff7f0000 -#define HID_UP_MSVENDOR 0xff000000 -#define HID_UP_CUSTOM 0x00ff0000 -#define HID_UP_LOGIVENDOR 0xffbc0000 - -#define HID_USAGE 0x0000ffff - -#define HID_GD_POINTER 0x00010001 -#define HID_GD_MOUSE 0x00010002 -#define HID_GD_JOYSTICK 0x00010004 -#define HID_GD_GAMEPAD 0x00010005 -#define HID_GD_KEYBOARD 0x00010006 -#define HID_GD_KEYPAD 0x00010007 -#define HID_GD_MULTIAXIS 0x00010008 -#define HID_GD_X 0x00010030 -#define HID_GD_Y 0x00010031 -#define HID_GD_Z 0x00010032 -#define HID_GD_RX 0x00010033 -#define HID_GD_RY 0x00010034 -#define HID_GD_RZ 0x00010035 -#define HID_GD_SLIDER 0x00010036 -#define HID_GD_DIAL 0x00010037 -#define HID_GD_WHEEL 0x00010038 -#define HID_GD_HATSWITCH 0x00010039 -#define HID_GD_BUFFER 0x0001003a -#define HID_GD_BYTECOUNT 0x0001003b -#define HID_GD_MOTION 0x0001003c -#define HID_GD_START 0x0001003d -#define HID_GD_SELECT 0x0001003e -#define HID_GD_VX 0x00010040 -#define HID_GD_VY 0x00010041 -#define HID_GD_VZ 0x00010042 -#define HID_GD_VBRX 0x00010043 -#define HID_GD_VBRY 0x00010044 -#define HID_GD_VBRZ 0x00010045 -#define HID_GD_VNO 0x00010046 -#define HID_GD_FEATURE 0x00010047 -#define HID_GD_UP 0x00010090 -#define HID_GD_DOWN 0x00010091 -#define HID_GD_RIGHT 0x00010092 -#define HID_GD_LEFT 0x00010093 - -/* - * HID report types --- Ouch! HID spec says 1 2 3! - */ - -#define HID_INPUT_REPORT 0 -#define HID_OUTPUT_REPORT 1 -#define HID_FEATURE_REPORT 2 - -/* - * HID device quirks. - */ - -#define HID_QUIRK_INVERT 0x00000001 -#define HID_QUIRK_NOTOUCH 0x00000002 -#define HID_QUIRK_IGNORE 0x00000004 -#define HID_QUIRK_NOGET 0x00000008 -#define HID_QUIRK_HIDDEV 0x00000010 -#define HID_QUIRK_BADPAD 0x00000020 -#define HID_QUIRK_MULTI_INPUT 0x00000040 -#define HID_QUIRK_2WHEEL_MOUSE_HACK_7 0x00000080 -#define HID_QUIRK_2WHEEL_MOUSE_HACK_5 0x00000100 -#define HID_QUIRK_2WHEEL_MOUSE_HACK_ON 0x00000200 -#define HID_QUIRK_MIGHTYMOUSE 0x00000400 -#define HID_QUIRK_CYMOTION 0x00000800 -#define HID_QUIRK_POWERBOOK_HAS_FN 0x00001000 -#define HID_QUIRK_POWERBOOK_FN_ON 0x00002000 -#define HID_QUIRK_INVERT_HWHEEL 0x00004000 -#define HID_QUIRK_POWERBOOK_ISO_KEYBOARD 0x00008000 -#define HID_QUIRK_BAD_RELATIVE_KEYS 0x00010000 - -/* - * This is the global environment of the parser. This information is - * persistent for main-items. The global environment can be saved and - * restored with PUSH/POP statements. - */ - -struct hid_global { - unsigned usage_page; - __s32 logical_minimum; - __s32 logical_maximum; - __s32 physical_minimum; - __s32 physical_maximum; - __s32 unit_exponent; - unsigned unit; - unsigned report_id; - unsigned report_size; - unsigned report_count; -}; - -/* - * This is the local environment. It is persistent up the next main-item. - */ - -#define HID_MAX_DESCRIPTOR_SIZE 4096 -#define HID_MAX_USAGES 1024 -#define HID_DEFAULT_NUM_COLLECTIONS 16 - -struct hid_local { - unsigned usage[HID_MAX_USAGES]; /* usage array */ - unsigned collection_index[HID_MAX_USAGES]; /* collection index array */ - unsigned usage_index; - unsigned usage_minimum; - unsigned delimiter_depth; - unsigned delimiter_branch; -}; - -/* - * This is the collection stack. We climb up the stack to determine - * application and function of each field. - */ - -struct hid_collection { - unsigned type; - unsigned usage; - unsigned level; -}; - -struct hid_usage { - unsigned hid; /* hid usage code */ - unsigned collection_index; /* index into collection array */ - /* hidinput data */ - __u16 code; /* input driver code */ - __u8 type; /* input driver type */ - __s8 hat_min; /* hat switch fun */ - __s8 hat_max; /* ditto */ - __s8 hat_dir; /* ditto */ -}; - -struct hid_input; - -struct hid_field { - unsigned physical; /* physical usage for this field */ - unsigned logical; /* logical usage for this field */ - unsigned application; /* application usage for this field */ - struct hid_usage *usage; /* usage table for this function */ - unsigned maxusage; /* maximum usage index */ - unsigned flags; /* main-item flags (i.e. volatile,array,constant) */ - unsigned report_offset; /* bit offset in the report */ - unsigned report_size; /* size of this field in the report */ - unsigned report_count; /* number of this field in the report */ - unsigned report_type; /* (input,output,feature) */ - __s32 *value; /* last known value(s) */ - __s32 logical_minimum; - __s32 logical_maximum; - __s32 physical_minimum; - __s32 physical_maximum; - __s32 unit_exponent; - unsigned unit; - struct hid_report *report; /* associated report */ - unsigned index; /* index into report->field[] */ - /* hidinput data */ - struct hid_input *hidinput; /* associated input structure */ - __u16 dpad; /* dpad input code */ -}; - -#define HID_MAX_FIELDS 64 - -struct hid_report { - struct list_head list; - unsigned id; /* id of this report */ - unsigned type; /* report type */ - struct hid_field *field[HID_MAX_FIELDS]; /* fields of the report */ - unsigned maxfield; /* maximum valid field index */ - unsigned size; /* size of the report (bits) */ - struct hid_device *device; /* associated device */ -}; - -struct hid_report_enum { - unsigned numbered; - struct list_head report_list; - struct hid_report *report_id_hash[256]; -}; - -#define HID_REPORT_TYPES 3 - -#define HID_MIN_BUFFER_SIZE 64 /* make sure there is at least a packet size of space */ -#define HID_MAX_BUFFER_SIZE 4096 /* 4kb */ -#define HID_CONTROL_FIFO_SIZE 256 /* to init devices with >100 reports */ -#define HID_OUTPUT_FIFO_SIZE 64 - -struct hid_control_fifo { - unsigned char dir; - struct hid_report *report; -}; - -#define HID_CLAIMED_INPUT 1 -#define HID_CLAIMED_HIDDEV 2 - -#define HID_CTRL_RUNNING 1 -#define HID_OUT_RUNNING 2 -#define HID_IN_RUNNING 3 -#define HID_RESET_PENDING 4 -#define HID_SUSPENDED 5 -#define HID_CLEAR_HALT 6 - -struct hid_input { - struct list_head list; - struct hid_report *report; - struct input_dev *input; -}; - -struct hid_device { /* device report descriptor */ - __u8 *rdesc; - unsigned rsize; - struct hid_collection *collection; /* List of HID collections */ - unsigned collection_size; /* Number of allocated hid_collections */ - unsigned maxcollection; /* Number of parsed collections */ - unsigned maxapplication; /* Number of applications */ - unsigned version; /* HID version */ - unsigned country; /* HID country */ - struct hid_report_enum report_enum[HID_REPORT_TYPES]; - - struct usb_device *dev; /* USB device */ - struct usb_interface *intf; /* USB interface */ - int ifnum; /* USB interface number */ - - unsigned long iofl; /* I/O flags (CTRL_RUNNING, OUT_RUNNING) */ - struct timer_list io_retry; /* Retry timer */ - unsigned long stop_retry; /* Time to give up, in jiffies */ - unsigned int retry_delay; /* Delay length in ms */ - struct work_struct reset_work; /* Task context for resets */ - - unsigned int bufsize; /* URB buffer size */ - - struct urb *urbin; /* Input URB */ - char *inbuf; /* Input buffer */ - dma_addr_t inbuf_dma; /* Input buffer dma */ - spinlock_t inlock; /* Input fifo spinlock */ - - struct urb *urbctrl; /* Control URB */ - struct usb_ctrlrequest *cr; /* Control request struct */ - dma_addr_t cr_dma; /* Control request struct dma */ - struct hid_control_fifo ctrl[HID_CONTROL_FIFO_SIZE]; /* Control fifo */ - unsigned char ctrlhead, ctrltail; /* Control fifo head & tail */ - char *ctrlbuf; /* Control buffer */ - dma_addr_t ctrlbuf_dma; /* Control buffer dma */ - spinlock_t ctrllock; /* Control fifo spinlock */ - - struct urb *urbout; /* Output URB */ - struct hid_report *out[HID_CONTROL_FIFO_SIZE]; /* Output pipe fifo */ - unsigned char outhead, outtail; /* Output pipe fifo head & tail */ - char *outbuf; /* Output buffer */ - dma_addr_t outbuf_dma; /* Output buffer dma */ - spinlock_t outlock; /* Output fifo spinlock */ - - unsigned claimed; /* Claimed by hidinput, hiddev? */ - unsigned quirks; /* Various quirks the device can pull on us */ - - struct list_head inputs; /* The list of inputs */ - void *hiddev; /* The hiddev structure */ - int minor; /* Hiddev minor number */ - - wait_queue_head_t wait; /* For sleeping */ - - int open; /* is the device open by anyone? */ - char name[128]; /* Device name */ - char phys[64]; /* Device physical location */ - char uniq[64]; /* Device unique identifier (serial #) */ - -#ifdef CONFIG_USB_HIDINPUT_POWERBOOK - unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; - unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; -#endif -}; - -#define HID_GLOBAL_STACK_SIZE 4 -#define HID_COLLECTION_STACK_SIZE 4 - -struct hid_parser { - struct hid_global global; - struct hid_global global_stack[HID_GLOBAL_STACK_SIZE]; - unsigned global_stack_ptr; - struct hid_local local; - unsigned collection_stack[HID_COLLECTION_STACK_SIZE]; - unsigned collection_stack_ptr; - struct hid_device *device; -}; - -struct hid_class_descriptor { - __u8 bDescriptorType; - __u16 wDescriptorLength; -} __attribute__ ((packed)); - -struct hid_descriptor { - __u8 bLength; - __u8 bDescriptorType; - __u16 bcdHID; - __u8 bCountryCode; - __u8 bNumDescriptors; - - struct hid_class_descriptor desc[1]; -} __attribute__ ((packed)); - -#ifdef DEBUG -#include "hid-debug.h" -#else -#define hid_dump_input(a,b) do { } while (0) -#define hid_dump_device(c) do { } while (0) -#define hid_dump_field(a,b) do { } while (0) -#define resolv_usage(a) do { } while (0) -#define resolv_event(a,b) do { } while (0) -#endif - -#endif - -#ifdef CONFIG_USB_HIDINPUT -/* Applications from HID Usage Tables 4/8/99 Version 1.1 */ -/* We ignore a few input applications that are not widely used */ -#define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001)) -extern void hidinput_hid_event(struct hid_device *, struct hid_field *, struct hid_usage *, __s32); -extern void hidinput_report_event(struct hid_device *hid, struct hid_report *report); -extern int hidinput_connect(struct hid_device *); -extern void hidinput_disconnect(struct hid_device *); -#else -#define IS_INPUT_APPLICATION(a) (0) -static inline void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value) { } -static inline void hidinput_report_event(struct hid_device *hid, struct hid_report *report) { } -static inline int hidinput_connect(struct hid_device *hid) { return -ENODEV; } -static inline void hidinput_disconnect(struct hid_device *hid) { } -#endif - -int hid_open(struct hid_device *); -void hid_close(struct hid_device *); -int hid_set_field(struct hid_field *, unsigned, __s32); -void hid_submit_report(struct hid_device *, struct hid_report *, unsigned char dir); -void hid_init_reports(struct hid_device *hid); -int hid_wait_io(struct hid_device* hid); - - -#ifdef CONFIG_HID_FF -int hid_ff_init(struct hid_device *hid); - -int hid_lgff_init(struct hid_device *hid); -int hid_tmff_init(struct hid_device *hid); -int hid_zpff_init(struct hid_device *hid); -#ifdef CONFIG_HID_PID -int hid_pidff_init(struct hid_device *hid); -#else -static inline int hid_pidff_init(struct hid_device *hid) { return -ENODEV; } -#endif - -#else -static inline int hid_ff_init(struct hid_device *hid) { return -1; } -#endif - diff --git a/drivers/usb/input/hiddev.c b/drivers/usb/input/hiddev.c index 7dc14d0cacc1..cbd3b60d93bb 100644 --- a/drivers/usb/input/hiddev.c +++ b/drivers/usb/input/hiddev.c @@ -32,8 +32,9 @@ #include #include #include -#include "hid.h" +#include #include +#include "usbhid.h" #ifdef CONFIG_USB_DYNAMIC_MINORS #define HIDDEV_MINOR_BASE 0 diff --git a/include/linux/hid-debug.h b/include/linux/hid-debug.h new file mode 100644 index 000000000000..f04d6d75c098 --- /dev/null +++ b/include/linux/hid-debug.h @@ -0,0 +1,757 @@ +/* + * $Id: hid-debug.h,v 1.8 2001/09/25 09:37:57 vojtech Exp $ + * + * (c) 1999 Andreas Gal + * (c) 2000-2001 Vojtech Pavlik + * + * Some debug stuff for the HID parser. + */ + +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Should you need to contact me, the author, you can do so either by + * e-mail - mail your message to , or by paper mail: + * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic + */ + +#include + +struct hid_usage_entry { + unsigned page; + unsigned usage; + char *description; +}; + +static const struct hid_usage_entry hid_usage_table[] = { + { 0, 0, "Undefined" }, + { 1, 0, "GenericDesktop" }, + {0, 0x01, "Pointer"}, + {0, 0x02, "Mouse"}, + {0, 0x04, "Joystick"}, + {0, 0x05, "GamePad"}, + {0, 0x06, "Keyboard"}, + {0, 0x07, "Keypad"}, + {0, 0x08, "MultiAxis"}, + {0, 0x30, "X"}, + {0, 0x31, "Y"}, + {0, 0x32, "Z"}, + {0, 0x33, "Rx"}, + {0, 0x34, "Ry"}, + {0, 0x35, "Rz"}, + {0, 0x36, "Slider"}, + {0, 0x37, "Dial"}, + {0, 0x38, "Wheel"}, + {0, 0x39, "HatSwitch"}, + {0, 0x3a, "CountedBuffer"}, + {0, 0x3b, "ByteCount"}, + {0, 0x3c, "MotionWakeup"}, + {0, 0x3d, "Start"}, + {0, 0x3e, "Select"}, + {0, 0x40, "Vx"}, + {0, 0x41, "Vy"}, + {0, 0x42, "Vz"}, + {0, 0x43, "Vbrx"}, + {0, 0x44, "Vbry"}, + {0, 0x45, "Vbrz"}, + {0, 0x46, "Vno"}, + {0, 0x80, "SystemControl"}, + {0, 0x81, "SystemPowerDown"}, + {0, 0x82, "SystemSleep"}, + {0, 0x83, "SystemWakeUp"}, + {0, 0x84, "SystemContextMenu"}, + {0, 0x85, "SystemMainMenu"}, + {0, 0x86, "SystemAppMenu"}, + {0, 0x87, "SystemMenuHelp"}, + {0, 0x88, "SystemMenuExit"}, + {0, 0x89, "SystemMenuSelect"}, + {0, 0x8a, "SystemMenuRight"}, + {0, 0x8b, "SystemMenuLeft"}, + {0, 0x8c, "SystemMenuUp"}, + {0, 0x8d, "SystemMenuDown"}, + {0, 0x90, "D-PadUp"}, + {0, 0x91, "D-PadDown"}, + {0, 0x92, "D-PadRight"}, + {0, 0x93, "D-PadLeft"}, + { 2, 0, "Simulation" }, + {0, 0xb0, "Aileron"}, + {0, 0xb1, "AileronTrim"}, + {0, 0xb2, "Anti-Torque"}, + {0, 0xb3, "Autopilot"}, + {0, 0xb4, "Chaff"}, + {0, 0xb5, "Collective"}, + {0, 0xb6, "DiveBrake"}, + {0, 0xb7, "ElectronicCountermeasures"}, + {0, 0xb8, "Elevator"}, + {0, 0xb9, "ElevatorTrim"}, + {0, 0xba, "Rudder"}, + {0, 0xbb, "Throttle"}, + {0, 0xbc, "FlightCommunications"}, + {0, 0xbd, "FlareRelease"}, + {0, 0xbe, "LandingGear"}, + {0, 0xbf, "ToeBrake"}, + { 7, 0, "Keyboard" }, + { 8, 0, "LED" }, + {0, 0x01, "NumLock"}, + {0, 0x02, "CapsLock"}, + {0, 0x03, "ScrollLock"}, + {0, 0x04, "Compose"}, + {0, 0x05, "Kana"}, + {0, 0x4b, "GenericIndicator"}, + { 9, 0, "Button" }, + { 10, 0, "Ordinal" }, + { 12, 0, "Consumer" }, + {0, 0x238, "HorizontalWheel"}, + { 13, 0, "Digitizers" }, + {0, 0x01, "Digitizer"}, + {0, 0x02, "Pen"}, + {0, 0x03, "LightPen"}, + {0, 0x04, "TouchScreen"}, + {0, 0x05, "TouchPad"}, + {0, 0x20, "Stylus"}, + {0, 0x21, "Puck"}, + {0, 0x22, "Finger"}, + {0, 0x30, "TipPressure"}, + {0, 0x31, "BarrelPressure"}, + {0, 0x32, "InRange"}, + {0, 0x33, "Touch"}, + {0, 0x34, "UnTouch"}, + {0, 0x35, "Tap"}, + {0, 0x39, "TabletFunctionKey"}, + {0, 0x3a, "ProgramChangeKey"}, + {0, 0x3c, "Invert"}, + {0, 0x42, "TipSwitch"}, + {0, 0x43, "SecondaryTipSwitch"}, + {0, 0x44, "BarrelSwitch"}, + {0, 0x45, "Eraser"}, + {0, 0x46, "TabletPick"}, + { 15, 0, "PhysicalInterfaceDevice" }, + {0, 0x00, "Undefined"}, + {0, 0x01, "Physical_Interface_Device"}, + {0, 0x20, "Normal"}, + {0, 0x21, "Set_Effect_Report"}, + {0, 0x22, "Effect_Block_Index"}, + {0, 0x23, "Parameter_Block_Offset"}, + {0, 0x24, "ROM_Flag"}, + {0, 0x25, "Effect_Type"}, + {0, 0x26, "ET_Constant_Force"}, + {0, 0x27, "ET_Ramp"}, + {0, 0x28, "ET_Custom_Force_Data"}, + {0, 0x30, "ET_Square"}, + {0, 0x31, "ET_Sine"}, + {0, 0x32, "ET_Triangle"}, + {0, 0x33, "ET_Sawtooth_Up"}, + {0, 0x34, "ET_Sawtooth_Down"}, + {0, 0x40, "ET_Spring"}, + {0, 0x41, "ET_Damper"}, + {0, 0x42, "ET_Inertia"}, + {0, 0x43, "ET_Friction"}, + {0, 0x50, "Duration"}, + {0, 0x51, "Sample_Period"}, + {0, 0x52, "Gain"}, + {0, 0x53, "Trigger_Button"}, + {0, 0x54, "Trigger_Repeat_Interval"}, + {0, 0x55, "Axes_Enable"}, + {0, 0x56, "Direction_Enable"}, + {0, 0x57, "Direction"}, + {0, 0x58, "Type_Specific_Block_Offset"}, + {0, 0x59, "Block_Type"}, + {0, 0x5A, "Set_Envelope_Report"}, + {0, 0x5B, "Attack_Level"}, + {0, 0x5C, "Attack_Time"}, + {0, 0x5D, "Fade_Level"}, + {0, 0x5E, "Fade_Time"}, + {0, 0x5F, "Set_Condition_Report"}, + {0, 0x60, "CP_Offset"}, + {0, 0x61, "Positive_Coefficient"}, + {0, 0x62, "Negative_Coefficient"}, + {0, 0x63, "Positive_Saturation"}, + {0, 0x64, "Negative_Saturation"}, + {0, 0x65, "Dead_Band"}, + {0, 0x66, "Download_Force_Sample"}, + {0, 0x67, "Isoch_Custom_Force_Enable"}, + {0, 0x68, "Custom_Force_Data_Report"}, + {0, 0x69, "Custom_Force_Data"}, + {0, 0x6A, "Custom_Force_Vendor_Defined_Data"}, + {0, 0x6B, "Set_Custom_Force_Report"}, + {0, 0x6C, "Custom_Force_Data_Offset"}, + {0, 0x6D, "Sample_Count"}, + {0, 0x6E, "Set_Periodic_Report"}, + {0, 0x6F, "Offset"}, + {0, 0x70, "Magnitude"}, + {0, 0x71, "Phase"}, + {0, 0x72, "Period"}, + {0, 0x73, "Set_Constant_Force_Report"}, + {0, 0x74, "Set_Ramp_Force_Report"}, + {0, 0x75, "Ramp_Start"}, + {0, 0x76, "Ramp_End"}, + {0, 0x77, "Effect_Operation_Report"}, + {0, 0x78, "Effect_Operation"}, + {0, 0x79, "Op_Effect_Start"}, + {0, 0x7A, "Op_Effect_Start_Solo"}, + {0, 0x7B, "Op_Effect_Stop"}, + {0, 0x7C, "Loop_Count"}, + {0, 0x7D, "Device_Gain_Report"}, + {0, 0x7E, "Device_Gain"}, + {0, 0x7F, "PID_Pool_Report"}, + {0, 0x80, "RAM_Pool_Size"}, + {0, 0x81, "ROM_Pool_Size"}, + {0, 0x82, "ROM_Effect_Block_Count"}, + {0, 0x83, "Simultaneous_Effects_Max"}, + {0, 0x84, "Pool_Alignment"}, + {0, 0x85, "PID_Pool_Move_Report"}, + {0, 0x86, "Move_Source"}, + {0, 0x87, "Move_Destination"}, + {0, 0x88, "Move_Length"}, + {0, 0x89, "PID_Block_Load_Report"}, + {0, 0x8B, "Block_Load_Status"}, + {0, 0x8C, "Block_Load_Success"}, + {0, 0x8D, "Block_Load_Full"}, + {0, 0x8E, "Block_Load_Error"}, + {0, 0x8F, "Block_Handle"}, + {0, 0x90, "PID_Block_Free_Report"}, + {0, 0x91, "Type_Specific_Block_Handle"}, + {0, 0x92, "PID_State_Report"}, + {0, 0x94, "Effect_Playing"}, + {0, 0x95, "PID_Device_Control_Report"}, + {0, 0x96, "PID_Device_Control"}, + {0, 0x97, "DC_Enable_Actuators"}, + {0, 0x98, "DC_Disable_Actuators"}, + {0, 0x99, "DC_Stop_All_Effects"}, + {0, 0x9A, "DC_Device_Reset"}, + {0, 0x9B, "DC_Device_Pause"}, + {0, 0x9C, "DC_Device_Continue"}, + {0, 0x9F, "Device_Paused"}, + {0, 0xA0, "Actuators_Enabled"}, + {0, 0xA4, "Safety_Switch"}, + {0, 0xA5, "Actuator_Override_Switch"}, + {0, 0xA6, "Actuator_Power"}, + {0, 0xA7, "Start_Delay"}, + {0, 0xA8, "Parameter_Block_Size"}, + {0, 0xA9, "Device_Managed_Pool"}, + {0, 0xAA, "Shared_Parameter_Blocks"}, + {0, 0xAB, "Create_New_Effect_Report"}, + {0, 0xAC, "RAM_Pool_Available"}, + { 0x84, 0, "Power Device" }, + { 0x84, 0x02, "PresentStatus" }, + { 0x84, 0x03, "ChangeStatus" }, + { 0x84, 0x04, "UPS" }, + { 0x84, 0x05, "PowerSupply" }, + { 0x84, 0x10, "BatterySystem" }, + { 0x84, 0x11, "BatterySystemID" }, + { 0x84, 0x12, "Battery" }, + { 0x84, 0x13, "BatteryID" }, + { 0x84, 0x14, "Charger" }, + { 0x84, 0x15, "ChargerID" }, + { 0x84, 0x16, "PowerConverter" }, + { 0x84, 0x17, "PowerConverterID" }, + { 0x84, 0x18, "OutletSystem" }, + { 0x84, 0x19, "OutletSystemID" }, + { 0x84, 0x1a, "Input" }, + { 0x84, 0x1b, "InputID" }, + { 0x84, 0x1c, "Output" }, + { 0x84, 0x1d, "OutputID" }, + { 0x84, 0x1e, "Flow" }, + { 0x84, 0x1f, "FlowID" }, + { 0x84, 0x20, "Outlet" }, + { 0x84, 0x21, "OutletID" }, + { 0x84, 0x22, "Gang" }, + { 0x84, 0x24, "PowerSummary" }, + { 0x84, 0x25, "PowerSummaryID" }, + { 0x84, 0x30, "Voltage" }, + { 0x84, 0x31, "Current" }, + { 0x84, 0x32, "Frequency" }, + { 0x84, 0x33, "ApparentPower" }, + { 0x84, 0x35, "PercentLoad" }, + { 0x84, 0x40, "ConfigVoltage" }, + { 0x84, 0x41, "ConfigCurrent" }, + { 0x84, 0x43, "ConfigApparentPower" }, + { 0x84, 0x53, "LowVoltageTransfer" }, + { 0x84, 0x54, "HighVoltageTransfer" }, + { 0x84, 0x56, "DelayBeforeStartup" }, + { 0x84, 0x57, "DelayBeforeShutdown" }, + { 0x84, 0x58, "Test" }, + { 0x84, 0x5a, "AudibleAlarmControl" }, + { 0x84, 0x60, "Present" }, + { 0x84, 0x61, "Good" }, + { 0x84, 0x62, "InternalFailure" }, + { 0x84, 0x65, "Overload" }, + { 0x84, 0x66, "OverCharged" }, + { 0x84, 0x67, "OverTemperature" }, + { 0x84, 0x68, "ShutdownRequested" }, + { 0x84, 0x69, "ShutdownImminent" }, + { 0x84, 0x6b, "SwitchOn/Off" }, + { 0x84, 0x6c, "Switchable" }, + { 0x84, 0x6d, "Used" }, + { 0x84, 0x6e, "Boost" }, + { 0x84, 0x73, "CommunicationLost" }, + { 0x84, 0xfd, "iManufacturer" }, + { 0x84, 0xfe, "iProduct" }, + { 0x84, 0xff, "iSerialNumber" }, + { 0x85, 0, "Battery System" }, + { 0x85, 0x01, "SMBBatteryMode" }, + { 0x85, 0x02, "SMBBatteryStatus" }, + { 0x85, 0x03, "SMBAlarmWarning" }, + { 0x85, 0x04, "SMBChargerMode" }, + { 0x85, 0x05, "SMBChargerStatus" }, + { 0x85, 0x06, "SMBChargerSpecInfo" }, + { 0x85, 0x07, "SMBSelectorState" }, + { 0x85, 0x08, "SMBSelectorPresets" }, + { 0x85, 0x09, "SMBSelectorInfo" }, + { 0x85, 0x29, "RemainingCapacityLimit" }, + { 0x85, 0x2c, "CapacityMode" }, + { 0x85, 0x42, "BelowRemainingCapacityLimit" }, + { 0x85, 0x44, "Charging" }, + { 0x85, 0x45, "Discharging" }, + { 0x85, 0x4b, "NeedReplacement" }, + { 0x85, 0x66, "RemainingCapacity" }, + { 0x85, 0x68, "RunTimeToEmpty" }, + { 0x85, 0x6a, "AverageTimeToFull" }, + { 0x85, 0x83, "DesignCapacity" }, + { 0x85, 0x85, "ManufacturerDate" }, + { 0x85, 0x89, "iDeviceChemistry" }, + { 0x85, 0x8b, "Rechargable" }, + { 0x85, 0x8f, "iOEMInformation" }, + { 0x85, 0x8d, "CapacityGranularity1" }, + { 0x85, 0xd0, "ACPresent" }, + /* pages 0xff00 to 0xffff are vendor-specific */ + { 0xffff, 0, "Vendor-specific-FF" }, + { 0, 0, NULL } +}; + +static void resolv_usage_page(unsigned page) { + const struct hid_usage_entry *p; + + for (p = hid_usage_table; p->description; p++) + if (p->page == page) { + printk("%s", p->description); + return; + } + printk("%04x", page); +} + +static void resolv_usage(unsigned usage) { + const struct hid_usage_entry *p; + + resolv_usage_page(usage >> 16); + printk("."); + for (p = hid_usage_table; p->description; p++) + if (p->page == (usage >> 16)) { + for(++p; p->description && p->usage != 0; p++) + if (p->usage == (usage & 0xffff)) { + printk("%s", p->description); + return; + } + break; + } + printk("%04x", usage & 0xffff); +} + +__inline__ static void tab(int n) { + while (n--) printk(" "); +} + +static void hid_dump_field(struct hid_field *field, int n) { + int j; + + if (field->physical) { + tab(n); + printk("Physical("); + resolv_usage(field->physical); printk(")\n"); + } + if (field->logical) { + tab(n); + printk("Logical("); + resolv_usage(field->logical); printk(")\n"); + } + tab(n); printk("Usage(%d)\n", field->maxusage); + for (j = 0; j < field->maxusage; j++) { + tab(n+2);resolv_usage(field->usage[j].hid); printk("\n"); + } + if (field->logical_minimum != field->logical_maximum) { + tab(n); printk("Logical Minimum(%d)\n", field->logical_minimum); + tab(n); printk("Logical Maximum(%d)\n", field->logical_maximum); + } + if (field->physical_minimum != field->physical_maximum) { + tab(n); printk("Physical Minimum(%d)\n", field->physical_minimum); + tab(n); printk("Physical Maximum(%d)\n", field->physical_maximum); + } + if (field->unit_exponent) { + tab(n); printk("Unit Exponent(%d)\n", field->unit_exponent); + } + if (field->unit) { + char *systems[5] = { "None", "SI Linear", "SI Rotation", "English Linear", "English Rotation" }; + char *units[5][8] = { + { "None", "None", "None", "None", "None", "None", "None", "None" }, + { "None", "Centimeter", "Gram", "Seconds", "Kelvin", "Ampere", "Candela", "None" }, + { "None", "Radians", "Gram", "Seconds", "Kelvin", "Ampere", "Candela", "None" }, + { "None", "Inch", "Slug", "Seconds", "Fahrenheit", "Ampere", "Candela", "None" }, + { "None", "Degrees", "Slug", "Seconds", "Fahrenheit", "Ampere", "Candela", "None" } + }; + + int i; + int sys; + __u32 data = field->unit; + + /* First nibble tells us which system we're in. */ + sys = data & 0xf; + data >>= 4; + + if(sys > 4) { + tab(n); printk("Unit(Invalid)\n"); + } + else { + int earlier_unit = 0; + + tab(n); printk("Unit(%s : ", systems[sys]); + + for (i=1 ; i>= 4; + if (nibble != 0) { + if(earlier_unit++ > 0) + printk("*"); + printk("%s", units[sys][i]); + if(nibble != 1) { + /* This is a _signed_ nibble(!) */ + + int val = nibble & 0x7; + if(nibble & 0x08) + val = -((0x7 & ~val) +1); + printk("^%d", val); + } + } + } + printk(")\n"); + } + } + tab(n); printk("Report Size(%u)\n", field->report_size); + tab(n); printk("Report Count(%u)\n", field->report_count); + tab(n); printk("Report Offset(%u)\n", field->report_offset); + + tab(n); printk("Flags( "); + j = field->flags; + printk("%s", HID_MAIN_ITEM_CONSTANT & j ? "Constant " : ""); + printk("%s", HID_MAIN_ITEM_VARIABLE & j ? "Variable " : "Array "); + printk("%s", HID_MAIN_ITEM_RELATIVE & j ? "Relative " : "Absolute "); + printk("%s", HID_MAIN_ITEM_WRAP & j ? "Wrap " : ""); + printk("%s", HID_MAIN_ITEM_NONLINEAR & j ? "NonLinear " : ""); + printk("%s", HID_MAIN_ITEM_NO_PREFERRED & j ? "NoPrefferedState " : ""); + printk("%s", HID_MAIN_ITEM_NULL_STATE & j ? "NullState " : ""); + printk("%s", HID_MAIN_ITEM_VOLATILE & j ? "Volatile " : ""); + printk("%s", HID_MAIN_ITEM_BUFFERED_BYTE & j ? "BufferedByte " : ""); + printk(")\n"); +} + +static void __attribute__((unused)) hid_dump_device(struct hid_device *device) { + struct hid_report_enum *report_enum; + struct hid_report *report; + struct list_head *list; + unsigned i,k; + static char *table[] = {"INPUT", "OUTPUT", "FEATURE"}; + + for (i = 0; i < HID_REPORT_TYPES; i++) { + report_enum = device->report_enum + i; + list = report_enum->report_list.next; + while (list != &report_enum->report_list) { + report = (struct hid_report *) list; + tab(2); + printk("%s", table[i]); + if (report->id) + printk("(%d)", report->id); + printk("[%s]", table[report->type]); + printk("\n"); + for (k = 0; k < report->maxfield; k++) { + tab(4); + printk("Field(%d)\n", k); + hid_dump_field(report->field[k], 6); + } + list = list->next; + } + } +} + +static void __attribute__((unused)) hid_dump_input(struct hid_usage *usage, __s32 value) { + printk("hid-debug: input "); + resolv_usage(usage->hid); + printk(" = %d\n", value); +} + + +static char *events[EV_MAX + 1] = { + [EV_SYN] = "Sync", [EV_KEY] = "Key", + [EV_REL] = "Relative", [EV_ABS] = "Absolute", + [EV_MSC] = "Misc", [EV_LED] = "LED", + [EV_SND] = "Sound", [EV_REP] = "Repeat", + [EV_FF] = "ForceFeedback", [EV_PWR] = "Power", + [EV_FF_STATUS] = "ForceFeedbackStatus", +}; + +static char *syncs[2] = { + [SYN_REPORT] = "Report", [SYN_CONFIG] = "Config", +}; +static char *keys[KEY_MAX + 1] = { + [KEY_RESERVED] = "Reserved", [KEY_ESC] = "Esc", + [KEY_1] = "1", [KEY_2] = "2", + [KEY_3] = "3", [KEY_4] = "4", + [KEY_5] = "5", [KEY_6] = "6", + [KEY_7] = "7", [KEY_8] = "8", + [KEY_9] = "9", [KEY_0] = "0", + [KEY_MINUS] = "Minus", [KEY_EQUAL] = "Equal", + [KEY_BACKSPACE] = "Backspace", [KEY_TAB] = "Tab", + [KEY_Q] = "Q", [KEY_W] = "W", + [KEY_E] = "E", [KEY_R] = "R", + [KEY_T] = "T", [KEY_Y] = "Y", + [KEY_U] = "U", [KEY_I] = "I", + [KEY_O] = "O", [KEY_P] = "P", + [KEY_LEFTBRACE] = "LeftBrace", [KEY_RIGHTBRACE] = "RightBrace", + [KEY_ENTER] = "Enter", [KEY_LEFTCTRL] = "LeftControl", + [KEY_A] = "A", [KEY_S] = "S", + [KEY_D] = "D", [KEY_F] = "F", + [KEY_G] = "G", [KEY_H] = "H", + [KEY_J] = "J", [KEY_K] = "K", + [KEY_L] = "L", [KEY_SEMICOLON] = "Semicolon", + [KEY_APOSTROPHE] = "Apostrophe", [KEY_GRAVE] = "Grave", + [KEY_LEFTSHIFT] = "LeftShift", [KEY_BACKSLASH] = "BackSlash", + [KEY_Z] = "Z", [KEY_X] = "X", + [KEY_C] = "C", [KEY_V] = "V", + [KEY_B] = "B", [KEY_N] = "N", + [KEY_M] = "M", [KEY_COMMA] = "Comma", + [KEY_DOT] = "Dot", [KEY_SLASH] = "Slash", + [KEY_RIGHTSHIFT] = "RightShift", [KEY_KPASTERISK] = "KPAsterisk", + [KEY_LEFTALT] = "LeftAlt", [KEY_SPACE] = "Space", + [KEY_CAPSLOCK] = "CapsLock", [KEY_F1] = "F1", + [KEY_F2] = "F2", [KEY_F3] = "F3", + [KEY_F4] = "F4", [KEY_F5] = "F5", + [KEY_F6] = "F6", [KEY_F7] = "F7", + [KEY_F8] = "F8", [KEY_F9] = "F9", + [KEY_F10] = "F10", [KEY_NUMLOCK] = "NumLock", + [KEY_SCROLLLOCK] = "ScrollLock", [KEY_KP7] = "KP7", + [KEY_KP8] = "KP8", [KEY_KP9] = "KP9", + [KEY_KPMINUS] = "KPMinus", [KEY_KP4] = "KP4", + [KEY_KP5] = "KP5", [KEY_KP6] = "KP6", + [KEY_KPPLUS] = "KPPlus", [KEY_KP1] = "KP1", + [KEY_KP2] = "KP2", [KEY_KP3] = "KP3", + [KEY_KP0] = "KP0", [KEY_KPDOT] = "KPDot", + [KEY_ZENKAKUHANKAKU] = "Zenkaku/Hankaku", [KEY_102ND] = "102nd", + [KEY_F11] = "F11", [KEY_F12] = "F12", + [KEY_RO] = "RO", [KEY_KATAKANA] = "Katakana", + [KEY_HIRAGANA] = "HIRAGANA", [KEY_HENKAN] = "Henkan", + [KEY_KATAKANAHIRAGANA] = "Katakana/Hiragana", [KEY_MUHENKAN] = "Muhenkan", + [KEY_KPJPCOMMA] = "KPJpComma", [KEY_KPENTER] = "KPEnter", + [KEY_RIGHTCTRL] = "RightCtrl", [KEY_KPSLASH] = "KPSlash", + [KEY_SYSRQ] = "SysRq", [KEY_RIGHTALT] = "RightAlt", + [KEY_LINEFEED] = "LineFeed", [KEY_HOME] = "Home", + [KEY_UP] = "Up", [KEY_PAGEUP] = "PageUp", + [KEY_LEFT] = "Left", [KEY_RIGHT] = "Right", + [KEY_END] = "End", [KEY_DOWN] = "Down", + [KEY_PAGEDOWN] = "PageDown", [KEY_INSERT] = "Insert", + [KEY_DELETE] = "Delete", [KEY_MACRO] = "Macro", + [KEY_MUTE] = "Mute", [KEY_VOLUMEDOWN] = "VolumeDown", + [KEY_VOLUMEUP] = "VolumeUp", [KEY_POWER] = "Power", + [KEY_KPEQUAL] = "KPEqual", [KEY_KPPLUSMINUS] = "KPPlusMinus", + [KEY_PAUSE] = "Pause", [KEY_KPCOMMA] = "KPComma", + [KEY_HANGUEL] = "Hangeul", [KEY_HANJA] = "Hanja", + [KEY_YEN] = "Yen", [KEY_LEFTMETA] = "LeftMeta", + [KEY_RIGHTMETA] = "RightMeta", [KEY_COMPOSE] = "Compose", + [KEY_STOP] = "Stop", [KEY_AGAIN] = "Again", + [KEY_PROPS] = "Props", [KEY_UNDO] = "Undo", + [KEY_FRONT] = "Front", [KEY_COPY] = "Copy", + [KEY_OPEN] = "Open", [KEY_PASTE] = "Paste", + [KEY_FIND] = "Find", [KEY_CUT] = "Cut", + [KEY_HELP] = "Help", [KEY_MENU] = "Menu", + [KEY_CALC] = "Calc", [KEY_SETUP] = "Setup", + [KEY_SLEEP] = "Sleep", [KEY_WAKEUP] = "WakeUp", + [KEY_FILE] = "File", [KEY_SENDFILE] = "SendFile", + [KEY_DELETEFILE] = "DeleteFile", [KEY_XFER] = "X-fer", + [KEY_PROG1] = "Prog1", [KEY_PROG2] = "Prog2", + [KEY_WWW] = "WWW", [KEY_MSDOS] = "MSDOS", + [KEY_COFFEE] = "Coffee", [KEY_DIRECTION] = "Direction", + [KEY_CYCLEWINDOWS] = "CycleWindows", [KEY_MAIL] = "Mail", + [KEY_BOOKMARKS] = "Bookmarks", [KEY_COMPUTER] = "Computer", + [KEY_BACK] = "Back", [KEY_FORWARD] = "Forward", + [KEY_CLOSECD] = "CloseCD", [KEY_EJECTCD] = "EjectCD", + [KEY_EJECTCLOSECD] = "EjectCloseCD", [KEY_NEXTSONG] = "NextSong", + [KEY_PLAYPAUSE] = "PlayPause", [KEY_PREVIOUSSONG] = "PreviousSong", + [KEY_STOPCD] = "StopCD", [KEY_RECORD] = "Record", + [KEY_REWIND] = "Rewind", [KEY_PHONE] = "Phone", + [KEY_ISO] = "ISOKey", [KEY_CONFIG] = "Config", + [KEY_HOMEPAGE] = "HomePage", [KEY_REFRESH] = "Refresh", + [KEY_EXIT] = "Exit", [KEY_MOVE] = "Move", + [KEY_EDIT] = "Edit", [KEY_SCROLLUP] = "ScrollUp", + [KEY_SCROLLDOWN] = "ScrollDown", [KEY_KPLEFTPAREN] = "KPLeftParenthesis", + [KEY_KPRIGHTPAREN] = "KPRightParenthesis", [KEY_NEW] = "New", + [KEY_REDO] = "Redo", [KEY_F13] = "F13", + [KEY_F14] = "F14", [KEY_F15] = "F15", + [KEY_F16] = "F16", [KEY_F17] = "F17", + [KEY_F18] = "F18", [KEY_F19] = "F19", + [KEY_F20] = "F20", [KEY_F21] = "F21", + [KEY_F22] = "F22", [KEY_F23] = "F23", + [KEY_F24] = "F24", [KEY_PLAYCD] = "PlayCD", + [KEY_PAUSECD] = "PauseCD", [KEY_PROG3] = "Prog3", + [KEY_PROG4] = "Prog4", [KEY_SUSPEND] = "Suspend", + [KEY_CLOSE] = "Close", [KEY_PLAY] = "Play", + [KEY_FASTFORWARD] = "FastForward", [KEY_BASSBOOST] = "BassBoost", + [KEY_PRINT] = "Print", [KEY_HP] = "HP", + [KEY_CAMERA] = "Camera", [KEY_SOUND] = "Sound", + [KEY_QUESTION] = "Question", [KEY_EMAIL] = "Email", + [KEY_CHAT] = "Chat", [KEY_SEARCH] = "Search", + [KEY_CONNECT] = "Connect", [KEY_FINANCE] = "Finance", + [KEY_SPORT] = "Sport", [KEY_SHOP] = "Shop", + [KEY_ALTERASE] = "AlternateErase", [KEY_CANCEL] = "Cancel", + [KEY_BRIGHTNESSDOWN] = "BrightnessDown", [KEY_BRIGHTNESSUP] = "BrightnessUp", + [KEY_MEDIA] = "Media", [KEY_UNKNOWN] = "Unknown", + [BTN_0] = "Btn0", [BTN_1] = "Btn1", + [BTN_2] = "Btn2", [BTN_3] = "Btn3", + [BTN_4] = "Btn4", [BTN_5] = "Btn5", + [BTN_6] = "Btn6", [BTN_7] = "Btn7", + [BTN_8] = "Btn8", [BTN_9] = "Btn9", + [BTN_LEFT] = "LeftBtn", [BTN_RIGHT] = "RightBtn", + [BTN_MIDDLE] = "MiddleBtn", [BTN_SIDE] = "SideBtn", + [BTN_EXTRA] = "ExtraBtn", [BTN_FORWARD] = "ForwardBtn", + [BTN_BACK] = "BackBtn", [BTN_TASK] = "TaskBtn", + [BTN_TRIGGER] = "Trigger", [BTN_THUMB] = "ThumbBtn", + [BTN_THUMB2] = "ThumbBtn2", [BTN_TOP] = "TopBtn", + [BTN_TOP2] = "TopBtn2", [BTN_PINKIE] = "PinkieBtn", + [BTN_BASE] = "BaseBtn", [BTN_BASE2] = "BaseBtn2", + [BTN_BASE3] = "BaseBtn3", [BTN_BASE4] = "BaseBtn4", + [BTN_BASE5] = "BaseBtn5", [BTN_BASE6] = "BaseBtn6", + [BTN_DEAD] = "BtnDead", [BTN_A] = "BtnA", + [BTN_B] = "BtnB", [BTN_C] = "BtnC", + [BTN_X] = "BtnX", [BTN_Y] = "BtnY", + [BTN_Z] = "BtnZ", [BTN_TL] = "BtnTL", + [BTN_TR] = "BtnTR", [BTN_TL2] = "BtnTL2", + [BTN_TR2] = "BtnTR2", [BTN_SELECT] = "BtnSelect", + [BTN_START] = "BtnStart", [BTN_MODE] = "BtnMode", + [BTN_THUMBL] = "BtnThumbL", [BTN_THUMBR] = "BtnThumbR", + [BTN_TOOL_PEN] = "ToolPen", [BTN_TOOL_RUBBER] = "ToolRubber", + [BTN_TOOL_BRUSH] = "ToolBrush", [BTN_TOOL_PENCIL] = "ToolPencil", + [BTN_TOOL_AIRBRUSH] = "ToolAirbrush", [BTN_TOOL_FINGER] = "ToolFinger", + [BTN_TOOL_MOUSE] = "ToolMouse", [BTN_TOOL_LENS] = "ToolLens", + [BTN_TOUCH] = "Touch", [BTN_STYLUS] = "Stylus", + [BTN_STYLUS2] = "Stylus2", [BTN_TOOL_DOUBLETAP] = "ToolDoubleTap", + [BTN_TOOL_TRIPLETAP] = "ToolTripleTap", [BTN_GEAR_DOWN] = "WheelBtn", + [BTN_GEAR_UP] = "Gear up", [KEY_OK] = "Ok", + [KEY_SELECT] = "Select", [KEY_GOTO] = "Goto", + [KEY_CLEAR] = "Clear", [KEY_POWER2] = "Power2", + [KEY_OPTION] = "Option", [KEY_INFO] = "Info", + [KEY_TIME] = "Time", [KEY_VENDOR] = "Vendor", + [KEY_ARCHIVE] = "Archive", [KEY_PROGRAM] = "Program", + [KEY_CHANNEL] = "Channel", [KEY_FAVORITES] = "Favorites", + [KEY_EPG] = "EPG", [KEY_PVR] = "PVR", + [KEY_MHP] = "MHP", [KEY_LANGUAGE] = "Language", + [KEY_TITLE] = "Title", [KEY_SUBTITLE] = "Subtitle", + [KEY_ANGLE] = "Angle", [KEY_ZOOM] = "Zoom", + [KEY_MODE] = "Mode", [KEY_KEYBOARD] = "Keyboard", + [KEY_SCREEN] = "Screen", [KEY_PC] = "PC", + [KEY_TV] = "TV", [KEY_TV2] = "TV2", + [KEY_VCR] = "VCR", [KEY_VCR2] = "VCR2", + [KEY_SAT] = "Sat", [KEY_SAT2] = "Sat2", + [KEY_CD] = "CD", [KEY_TAPE] = "Tape", + [KEY_RADIO] = "Radio", [KEY_TUNER] = "Tuner", + [KEY_PLAYER] = "Player", [KEY_TEXT] = "Text", + [KEY_DVD] = "DVD", [KEY_AUX] = "Aux", + [KEY_MP3] = "MP3", [KEY_AUDIO] = "Audio", + [KEY_VIDEO] = "Video", [KEY_DIRECTORY] = "Directory", + [KEY_LIST] = "List", [KEY_MEMO] = "Memo", + [KEY_CALENDAR] = "Calendar", [KEY_RED] = "Red", + [KEY_GREEN] = "Green", [KEY_YELLOW] = "Yellow", + [KEY_BLUE] = "Blue", [KEY_CHANNELUP] = "ChannelUp", + [KEY_CHANNELDOWN] = "ChannelDown", [KEY_FIRST] = "First", + [KEY_LAST] = "Last", [KEY_AB] = "AB", + [KEY_NEXT] = "Next", [KEY_RESTART] = "Restart", + [KEY_SLOW] = "Slow", [KEY_SHUFFLE] = "Shuffle", + [KEY_BREAK] = "Break", [KEY_PREVIOUS] = "Previous", + [KEY_DIGITS] = "Digits", [KEY_TEEN] = "TEEN", + [KEY_TWEN] = "TWEN", [KEY_DEL_EOL] = "DeleteEOL", + [KEY_DEL_EOS] = "DeleteEOS", [KEY_INS_LINE] = "InsertLine", + [KEY_DEL_LINE] = "DeleteLine", + [KEY_SEND] = "Send", [KEY_REPLY] = "Reply", + [KEY_FORWARDMAIL] = "ForwardMail", [KEY_SAVE] = "Save", + [KEY_DOCUMENTS] = "Documents", + [KEY_FN] = "Fn", [KEY_FN_ESC] = "Fn+ESC", + [KEY_FN_1] = "Fn+1", [KEY_FN_2] = "Fn+2", + [KEY_FN_B] = "Fn+B", [KEY_FN_D] = "Fn+D", + [KEY_FN_E] = "Fn+E", [KEY_FN_F] = "Fn+F", + [KEY_FN_S] = "Fn+S", + [KEY_FN_F1] = "Fn+F1", [KEY_FN_F2] = "Fn+F2", + [KEY_FN_F3] = "Fn+F3", [KEY_FN_F4] = "Fn+F4", + [KEY_FN_F5] = "Fn+F5", [KEY_FN_F6] = "Fn+F6", + [KEY_FN_F7] = "Fn+F7", [KEY_FN_F8] = "Fn+F8", + [KEY_FN_F9] = "Fn+F9", [KEY_FN_F10] = "Fn+F10", + [KEY_FN_F11] = "Fn+F11", [KEY_FN_F12] = "Fn+F12", + [KEY_KBDILLUMTOGGLE] = "KbdIlluminationToggle", + [KEY_KBDILLUMDOWN] = "KbdIlluminationDown", + [KEY_KBDILLUMUP] = "KbdIlluminationUp", + [KEY_SWITCHVIDEOMODE] = "SwitchVideoMode", +}; + +static char *relatives[REL_MAX + 1] = { + [REL_X] = "X", [REL_Y] = "Y", + [REL_Z] = "Z", [REL_HWHEEL] = "HWheel", + [REL_DIAL] = "Dial", [REL_WHEEL] = "Wheel", + [REL_MISC] = "Misc", +}; + +static char *absolutes[ABS_MAX + 1] = { + [ABS_X] = "X", [ABS_Y] = "Y", + [ABS_Z] = "Z", [ABS_RX] = "Rx", + [ABS_RY] = "Ry", [ABS_RZ] = "Rz", + [ABS_THROTTLE] = "Throttle", [ABS_RUDDER] = "Rudder", + [ABS_WHEEL] = "Wheel", [ABS_GAS] = "Gas", + [ABS_BRAKE] = "Brake", [ABS_HAT0X] = "Hat0X", + [ABS_HAT0Y] = "Hat0Y", [ABS_HAT1X] = "Hat1X", + [ABS_HAT1Y] = "Hat1Y", [ABS_HAT2X] = "Hat2X", + [ABS_HAT2Y] = "Hat2Y", [ABS_HAT3X] = "Hat3X", + [ABS_HAT3Y] = "Hat 3Y", [ABS_PRESSURE] = "Pressure", + [ABS_DISTANCE] = "Distance", [ABS_TILT_X] = "XTilt", + [ABS_TILT_Y] = "YTilt", [ABS_TOOL_WIDTH] = "Tool Width", + [ABS_VOLUME] = "Volume", [ABS_MISC] = "Misc", +}; + +static char *misc[MSC_MAX + 1] = { + [MSC_SERIAL] = "Serial", [MSC_PULSELED] = "Pulseled", + [MSC_GESTURE] = "Gesture", [MSC_RAW] = "RawData" +}; + +static char *leds[LED_MAX + 1] = { + [LED_NUML] = "NumLock", [LED_CAPSL] = "CapsLock", + [LED_SCROLLL] = "ScrollLock", [LED_COMPOSE] = "Compose", + [LED_KANA] = "Kana", [LED_SLEEP] = "Sleep", + [LED_SUSPEND] = "Suspend", [LED_MUTE] = "Mute", + [LED_MISC] = "Misc", +}; + +static char *repeats[REP_MAX + 1] = { + [REP_DELAY] = "Delay", [REP_PERIOD] = "Period" +}; + +static char *sounds[SND_MAX + 1] = { + [SND_CLICK] = "Click", [SND_BELL] = "Bell", + [SND_TONE] = "Tone" +}; + +static char **names[EV_MAX + 1] = { + [EV_SYN] = syncs, [EV_KEY] = keys, + [EV_REL] = relatives, [EV_ABS] = absolutes, + [EV_MSC] = misc, [EV_LED] = leds, + [EV_SND] = sounds, [EV_REP] = repeats, +}; + +static void __attribute__((unused)) resolv_event(__u8 type, __u16 code) { + + printk("%s.%s", events[type] ? events[type] : "?", + names[type] ? (names[type][code] ? names[type][code] : "?") : "?"); +} diff --git a/include/linux/hid.h b/include/linux/hid.h new file mode 100644 index 000000000000..ee567ae6fec1 --- /dev/null +++ b/include/linux/hid.h @@ -0,0 +1,540 @@ +#ifndef __HID_H +#define __HID_H + +/* + * $Id: hid.h,v 1.24 2001/12/27 10:37:41 vojtech Exp $ + * + * Copyright (c) 1999 Andreas Gal + * Copyright (c) 2000-2001 Vojtech Pavlik + * Copyright (c) 2006 Jiri Kosina + */ + +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Should you need to contact me, the author, you can do so either by + * e-mail - mail your message to , or by paper mail: + * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic + */ + +#include +#include +#include +#include +#include +#include + +/* + * USB HID (Human Interface Device) interface class code + */ + +#define USB_INTERFACE_CLASS_HID 3 + +/* + * USB HID interface subclass and protocol codes + */ + +#define USB_INTERFACE_SUBCLASS_BOOT 1 +#define USB_INTERFACE_PROTOCOL_KEYBOARD 1 +#define USB_INTERFACE_PROTOCOL_MOUSE 2 + +/* + * HID class requests + */ + +#define HID_REQ_GET_REPORT 0x01 +#define HID_REQ_GET_IDLE 0x02 +#define HID_REQ_GET_PROTOCOL 0x03 +#define HID_REQ_SET_REPORT 0x09 +#define HID_REQ_SET_IDLE 0x0A +#define HID_REQ_SET_PROTOCOL 0x0B + +/* + * HID class descriptor types + */ + +#define HID_DT_HID (USB_TYPE_CLASS | 0x01) +#define HID_DT_REPORT (USB_TYPE_CLASS | 0x02) +#define HID_DT_PHYSICAL (USB_TYPE_CLASS | 0x03) + +/* + * We parse each description item into this structure. Short items data + * values are expanded to 32-bit signed int, long items contain a pointer + * into the data area. + */ + +struct hid_item { + unsigned format; + __u8 size; + __u8 type; + __u8 tag; + union { + __u8 u8; + __s8 s8; + __u16 u16; + __s16 s16; + __u32 u32; + __s32 s32; + __u8 *longdata; + } data; +}; + +/* + * HID report item format + */ + +#define HID_ITEM_FORMAT_SHORT 0 +#define HID_ITEM_FORMAT_LONG 1 + +/* + * Special tag indicating long items + */ + +#define HID_ITEM_TAG_LONG 15 + +/* + * HID report descriptor item type (prefix bit 2,3) + */ + +#define HID_ITEM_TYPE_MAIN 0 +#define HID_ITEM_TYPE_GLOBAL 1 +#define HID_ITEM_TYPE_LOCAL 2 +#define HID_ITEM_TYPE_RESERVED 3 + +/* + * HID report descriptor main item tags + */ + +#define HID_MAIN_ITEM_TAG_INPUT 8 +#define HID_MAIN_ITEM_TAG_OUTPUT 9 +#define HID_MAIN_ITEM_TAG_FEATURE 11 +#define HID_MAIN_ITEM_TAG_BEGIN_COLLECTION 10 +#define HID_MAIN_ITEM_TAG_END_COLLECTION 12 + +/* + * HID report descriptor main item contents + */ + +#define HID_MAIN_ITEM_CONSTANT 0x001 +#define HID_MAIN_ITEM_VARIABLE 0x002 +#define HID_MAIN_ITEM_RELATIVE 0x004 +#define HID_MAIN_ITEM_WRAP 0x008 +#define HID_MAIN_ITEM_NONLINEAR 0x010 +#define HID_MAIN_ITEM_NO_PREFERRED 0x020 +#define HID_MAIN_ITEM_NULL_STATE 0x040 +#define HID_MAIN_ITEM_VOLATILE 0x080 +#define HID_MAIN_ITEM_BUFFERED_BYTE 0x100 + +/* + * HID report descriptor collection item types + */ + +#define HID_COLLECTION_PHYSICAL 0 +#define HID_COLLECTION_APPLICATION 1 +#define HID_COLLECTION_LOGICAL 2 + +/* + * HID report descriptor global item tags + */ + +#define HID_GLOBAL_ITEM_TAG_USAGE_PAGE 0 +#define HID_GLOBAL_ITEM_TAG_LOGICAL_MINIMUM 1 +#define HID_GLOBAL_ITEM_TAG_LOGICAL_MAXIMUM 2 +#define HID_GLOBAL_ITEM_TAG_PHYSICAL_MINIMUM 3 +#define HID_GLOBAL_ITEM_TAG_PHYSICAL_MAXIMUM 4 +#define HID_GLOBAL_ITEM_TAG_UNIT_EXPONENT 5 +#define HID_GLOBAL_ITEM_TAG_UNIT 6 +#define HID_GLOBAL_ITEM_TAG_REPORT_SIZE 7 +#define HID_GLOBAL_ITEM_TAG_REPORT_ID 8 +#define HID_GLOBAL_ITEM_TAG_REPORT_COUNT 9 +#define HID_GLOBAL_ITEM_TAG_PUSH 10 +#define HID_GLOBAL_ITEM_TAG_POP 11 + +/* + * HID report descriptor local item tags + */ + +#define HID_LOCAL_ITEM_TAG_USAGE 0 +#define HID_LOCAL_ITEM_TAG_USAGE_MINIMUM 1 +#define HID_LOCAL_ITEM_TAG_USAGE_MAXIMUM 2 +#define HID_LOCAL_ITEM_TAG_DESIGNATOR_INDEX 3 +#define HID_LOCAL_ITEM_TAG_DESIGNATOR_MINIMUM 4 +#define HID_LOCAL_ITEM_TAG_DESIGNATOR_MAXIMUM 5 +#define HID_LOCAL_ITEM_TAG_STRING_INDEX 7 +#define HID_LOCAL_ITEM_TAG_STRING_MINIMUM 8 +#define HID_LOCAL_ITEM_TAG_STRING_MAXIMUM 9 +#define HID_LOCAL_ITEM_TAG_DELIMITER 10 + +/* + * HID usage tables + */ + +#define HID_USAGE_PAGE 0xffff0000 + +#define HID_UP_UNDEFINED 0x00000000 +#define HID_UP_GENDESK 0x00010000 +#define HID_UP_SIMULATION 0x00020000 +#define HID_UP_KEYBOARD 0x00070000 +#define HID_UP_LED 0x00080000 +#define HID_UP_BUTTON 0x00090000 +#define HID_UP_ORDINAL 0x000a0000 +#define HID_UP_CONSUMER 0x000c0000 +#define HID_UP_DIGITIZER 0x000d0000 +#define HID_UP_PID 0x000f0000 +#define HID_UP_HPVENDOR 0xff7f0000 +#define HID_UP_MSVENDOR 0xff000000 +#define HID_UP_CUSTOM 0x00ff0000 +#define HID_UP_LOGIVENDOR 0xffbc0000 + +#define HID_USAGE 0x0000ffff + +#define HID_GD_POINTER 0x00010001 +#define HID_GD_MOUSE 0x00010002 +#define HID_GD_JOYSTICK 0x00010004 +#define HID_GD_GAMEPAD 0x00010005 +#define HID_GD_KEYBOARD 0x00010006 +#define HID_GD_KEYPAD 0x00010007 +#define HID_GD_MULTIAXIS 0x00010008 +#define HID_GD_X 0x00010030 +#define HID_GD_Y 0x00010031 +#define HID_GD_Z 0x00010032 +#define HID_GD_RX 0x00010033 +#define HID_GD_RY 0x00010034 +#define HID_GD_RZ 0x00010035 +#define HID_GD_SLIDER 0x00010036 +#define HID_GD_DIAL 0x00010037 +#define HID_GD_WHEEL 0x00010038 +#define HID_GD_HATSWITCH 0x00010039 +#define HID_GD_BUFFER 0x0001003a +#define HID_GD_BYTECOUNT 0x0001003b +#define HID_GD_MOTION 0x0001003c +#define HID_GD_START 0x0001003d +#define HID_GD_SELECT 0x0001003e +#define HID_GD_VX 0x00010040 +#define HID_GD_VY 0x00010041 +#define HID_GD_VZ 0x00010042 +#define HID_GD_VBRX 0x00010043 +#define HID_GD_VBRY 0x00010044 +#define HID_GD_VBRZ 0x00010045 +#define HID_GD_VNO 0x00010046 +#define HID_GD_FEATURE 0x00010047 +#define HID_GD_UP 0x00010090 +#define HID_GD_DOWN 0x00010091 +#define HID_GD_RIGHT 0x00010092 +#define HID_GD_LEFT 0x00010093 + +/* + * HID report types --- Ouch! HID spec says 1 2 3! + */ + +#define HID_INPUT_REPORT 0 +#define HID_OUTPUT_REPORT 1 +#define HID_FEATURE_REPORT 2 + +/* + * HID device quirks. + */ + +#define HID_QUIRK_INVERT 0x00000001 +#define HID_QUIRK_NOTOUCH 0x00000002 +#define HID_QUIRK_IGNORE 0x00000004 +#define HID_QUIRK_NOGET 0x00000008 +#define HID_QUIRK_HIDDEV 0x00000010 +#define HID_QUIRK_BADPAD 0x00000020 +#define HID_QUIRK_MULTI_INPUT 0x00000040 +#define HID_QUIRK_2WHEEL_MOUSE_HACK_7 0x00000080 +#define HID_QUIRK_2WHEEL_MOUSE_HACK_5 0x00000100 +#define HID_QUIRK_2WHEEL_MOUSE_HACK_ON 0x00000200 +#define HID_QUIRK_MIGHTYMOUSE 0x00000400 +#define HID_QUIRK_CYMOTION 0x00000800 +#define HID_QUIRK_POWERBOOK_HAS_FN 0x00001000 +#define HID_QUIRK_POWERBOOK_FN_ON 0x00002000 +#define HID_QUIRK_INVERT_HWHEEL 0x00004000 +#define HID_QUIRK_POWERBOOK_ISO_KEYBOARD 0x00008000 +#define HID_QUIRK_BAD_RELATIVE_KEYS 0x00010000 + +/* + * This is the global environment of the parser. This information is + * persistent for main-items. The global environment can be saved and + * restored with PUSH/POP statements. + */ + +struct hid_global { + unsigned usage_page; + __s32 logical_minimum; + __s32 logical_maximum; + __s32 physical_minimum; + __s32 physical_maximum; + __s32 unit_exponent; + unsigned unit; + unsigned report_id; + unsigned report_size; + unsigned report_count; +}; + +/* + * This is the local environment. It is persistent up the next main-item. + */ + +#define HID_MAX_DESCRIPTOR_SIZE 4096 +#define HID_MAX_USAGES 1024 +#define HID_DEFAULT_NUM_COLLECTIONS 16 + +struct hid_local { + unsigned usage[HID_MAX_USAGES]; /* usage array */ + unsigned collection_index[HID_MAX_USAGES]; /* collection index array */ + unsigned usage_index; + unsigned usage_minimum; + unsigned delimiter_depth; + unsigned delimiter_branch; +}; + +/* + * This is the collection stack. We climb up the stack to determine + * application and function of each field. + */ + +struct hid_collection { + unsigned type; + unsigned usage; + unsigned level; +}; + +struct hid_usage { + unsigned hid; /* hid usage code */ + unsigned collection_index; /* index into collection array */ + /* hidinput data */ + __u16 code; /* input driver code */ + __u8 type; /* input driver type */ + __s8 hat_min; /* hat switch fun */ + __s8 hat_max; /* ditto */ + __s8 hat_dir; /* ditto */ +}; + +struct hid_input; + +struct hid_field { + unsigned physical; /* physical usage for this field */ + unsigned logical; /* logical usage for this field */ + unsigned application; /* application usage for this field */ + struct hid_usage *usage; /* usage table for this function */ + unsigned maxusage; /* maximum usage index */ + unsigned flags; /* main-item flags (i.e. volatile,array,constant) */ + unsigned report_offset; /* bit offset in the report */ + unsigned report_size; /* size of this field in the report */ + unsigned report_count; /* number of this field in the report */ + unsigned report_type; /* (input,output,feature) */ + __s32 *value; /* last known value(s) */ + __s32 logical_minimum; + __s32 logical_maximum; + __s32 physical_minimum; + __s32 physical_maximum; + __s32 unit_exponent; + unsigned unit; + struct hid_report *report; /* associated report */ + unsigned index; /* index into report->field[] */ + /* hidinput data */ + struct hid_input *hidinput; /* associated input structure */ + __u16 dpad; /* dpad input code */ +}; + +#define HID_MAX_FIELDS 64 + +struct hid_report { + struct list_head list; + unsigned id; /* id of this report */ + unsigned type; /* report type */ + struct hid_field *field[HID_MAX_FIELDS]; /* fields of the report */ + unsigned maxfield; /* maximum valid field index */ + unsigned size; /* size of the report (bits) */ + struct hid_device *device; /* associated device */ +}; + +struct hid_report_enum { + unsigned numbered; + struct list_head report_list; + struct hid_report *report_id_hash[256]; +}; + +#define HID_REPORT_TYPES 3 + +#define HID_MIN_BUFFER_SIZE 64 /* make sure there is at least a packet size of space */ +#define HID_MAX_BUFFER_SIZE 4096 /* 4kb */ +#define HID_CONTROL_FIFO_SIZE 256 /* to init devices with >100 reports */ +#define HID_OUTPUT_FIFO_SIZE 64 + +struct hid_control_fifo { + unsigned char dir; + struct hid_report *report; +}; + +#define HID_CLAIMED_INPUT 1 +#define HID_CLAIMED_HIDDEV 2 + +#define HID_CTRL_RUNNING 1 +#define HID_OUT_RUNNING 2 +#define HID_IN_RUNNING 3 +#define HID_RESET_PENDING 4 +#define HID_SUSPENDED 5 +#define HID_CLEAR_HALT 6 + +struct hid_input { + struct list_head list; + struct hid_report *report; + struct input_dev *input; +}; + +struct hid_device { /* device report descriptor */ + __u8 *rdesc; + unsigned rsize; + struct hid_collection *collection; /* List of HID collections */ + unsigned collection_size; /* Number of allocated hid_collections */ + unsigned maxcollection; /* Number of parsed collections */ + unsigned maxapplication; /* Number of applications */ + unsigned version; /* HID version */ + unsigned country; /* HID country */ + struct hid_report_enum report_enum[HID_REPORT_TYPES]; + + struct usb_device *dev; /* USB device */ + struct usb_interface *intf; /* USB interface */ + int ifnum; /* USB interface number */ + + unsigned long iofl; /* I/O flags (CTRL_RUNNING, OUT_RUNNING) */ + struct timer_list io_retry; /* Retry timer */ + unsigned long stop_retry; /* Time to give up, in jiffies */ + unsigned int retry_delay; /* Delay length in ms */ + struct work_struct reset_work; /* Task context for resets */ + + unsigned int bufsize; /* URB buffer size */ + + struct urb *urbin; /* Input URB */ + char *inbuf; /* Input buffer */ + dma_addr_t inbuf_dma; /* Input buffer dma */ + spinlock_t inlock; /* Input fifo spinlock */ + + struct urb *urbctrl; /* Control URB */ + struct usb_ctrlrequest *cr; /* Control request struct */ + dma_addr_t cr_dma; /* Control request struct dma */ + struct hid_control_fifo ctrl[HID_CONTROL_FIFO_SIZE]; /* Control fifo */ + unsigned char ctrlhead, ctrltail; /* Control fifo head & tail */ + char *ctrlbuf; /* Control buffer */ + dma_addr_t ctrlbuf_dma; /* Control buffer dma */ + spinlock_t ctrllock; /* Control fifo spinlock */ + + struct urb *urbout; /* Output URB */ + struct hid_report *out[HID_CONTROL_FIFO_SIZE]; /* Output pipe fifo */ + unsigned char outhead, outtail; /* Output pipe fifo head & tail */ + char *outbuf; /* Output buffer */ + dma_addr_t outbuf_dma; /* Output buffer dma */ + spinlock_t outlock; /* Output fifo spinlock */ + + unsigned claimed; /* Claimed by hidinput, hiddev? */ + unsigned quirks; /* Various quirks the device can pull on us */ + + struct list_head inputs; /* The list of inputs */ + void *hiddev; /* The hiddev structure */ + int minor; /* Hiddev minor number */ + + wait_queue_head_t wait; /* For sleeping */ + + int open; /* is the device open by anyone? */ + char name[128]; /* Device name */ + char phys[64]; /* Device physical location */ + char uniq[64]; /* Device unique identifier (serial #) */ + +#ifdef CONFIG_USB_HIDINPUT_POWERBOOK + unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; + unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; +#endif +}; + +#define HID_GLOBAL_STACK_SIZE 4 +#define HID_COLLECTION_STACK_SIZE 4 + +struct hid_parser { + struct hid_global global; + struct hid_global global_stack[HID_GLOBAL_STACK_SIZE]; + unsigned global_stack_ptr; + struct hid_local local; + unsigned collection_stack[HID_COLLECTION_STACK_SIZE]; + unsigned collection_stack_ptr; + struct hid_device *device; +}; + +struct hid_class_descriptor { + __u8 bDescriptorType; + __u16 wDescriptorLength; +} __attribute__ ((packed)); + +struct hid_descriptor { + __u8 bLength; + __u8 bDescriptorType; + __u16 bcdHID; + __u8 bCountryCode; + __u8 bNumDescriptors; + + struct hid_class_descriptor desc[1]; +} __attribute__ ((packed)); + +#ifdef DEBUG +#include "hid-debug.h" +#else +#define hid_dump_input(a,b) do { } while (0) +#define hid_dump_device(c) do { } while (0) +#define hid_dump_field(a,b) do { } while (0) +#define resolv_usage(a) do { } while (0) +#define resolv_event(a,b) do { } while (0) +#endif + +#ifdef CONFIG_HID +/* Applications from HID Usage Tables 4/8/99 Version 1.1 */ +/* We ignore a few input applications that are not widely used */ +#define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001)) +extern void hidinput_hid_event(struct hid_device *, struct hid_field *, struct hid_usage *, __s32); +extern void hidinput_report_event(struct hid_device *hid, struct hid_report *report); +extern int hidinput_connect(struct hid_device *); +extern void hidinput_disconnect(struct hid_device *); +#else +#define IS_INPUT_APPLICATION(a) (0) +static inline void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value) { } +static inline void hidinput_report_event(struct hid_device *hid, struct hid_report *report) { } +static inline int hidinput_connect(struct hid_device *hid) { return -ENODEV; } +static inline void hidinput_disconnect(struct hid_device *hid) { } +#endif + +int hid_set_field(struct hid_field *, unsigned, __s32); +int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field); +void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt); +void hid_output_report(struct hid_report *report, __u8 *data); +void hid_free_device(struct hid_device *device); +struct hid_device *hid_parse_report(__u8 *start, unsigned size); + +#ifdef CONFIG_HID_FF +int hid_ff_init(struct hid_device *hid); + +int hid_lgff_init(struct hid_device *hid); +int hid_tmff_init(struct hid_device *hid); +int hid_zpff_init(struct hid_device *hid); +#ifdef CONFIG_HID_PID +int hid_pidff_init(struct hid_device *hid); +#else +static inline int hid_pidff_init(struct hid_device *hid) { return -ENODEV; } +#endif + +#else +static inline int hid_ff_init(struct hid_device *hid) { return -1; } +#endif +#endif + -- cgit v1.2.3 From 229695e51efc4ed5e04ab471c82591d0f432909d Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:40:53 +0100 Subject: [PATCH] Generic HID layer - API - fixed generic API (added neccessary EXPORT_SYMBOL, fixed hid.h to provide correct prototypes) - extended hid_device with open/close/event function pointers to driver-specific functions - added driver specific driver_data to hid_device Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-core.c | 22 ++++++++++++---------- drivers/hid/hid-input.c | 19 +++++++++++++------ drivers/usb/input/hid-core.c | 9 ++++++++- drivers/usb/input/hid-ff.c | 2 ++ drivers/usb/input/hiddev.c | 2 +- include/linux/hid.h | 36 ++++++++++++++++++++++++++++-------- 6 files changed, 64 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 689ae16adf33..8474a7923322 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -1,5 +1,5 @@ /* - * USB HID support for Linux + * HID support for Linux * * Copyright (c) 1999 Andreas Gal * Copyright (c) 2000-2005 Vojtech Pavlik @@ -31,8 +31,6 @@ #undef DEBUG #undef DEBUG_DATA -#include - #include #include @@ -538,7 +536,7 @@ static void hid_free_report(struct hid_report *report) * Free a device structure, all reports, and all fields. */ -static void hid_free_device(struct hid_device *device) +void hid_free_device(struct hid_device *device) { unsigned i,j; @@ -555,6 +553,7 @@ static void hid_free_device(struct hid_device *device) kfree(device->rdesc); kfree(device); } +EXPORT_SYMBOL_GPL(hid_free_device); /* * Fetch a report description item from the data stream. We support long @@ -629,7 +628,7 @@ static u8 *fetch_item(__u8 *start, __u8 *end, struct hid_item *item) * enumerated, fields are attached to these reports. */ -static struct hid_device *hid_parse_report(__u8 *start, unsigned size) +struct hid_device *hid_parse_report(__u8 *start, unsigned size) { struct hid_device *device; struct hid_parser *parser; @@ -719,6 +718,7 @@ static struct hid_device *hid_parse_report(__u8 *start, unsigned size) kfree(parser); return NULL; } +EXPORT_SYMBOL_GPL(hid_parse_report); /* * Convert a signed n-bit integer to signed 32-bit integer. Common @@ -767,10 +767,10 @@ static __inline__ __u32 extract(__u8 *report, unsigned offset, unsigned n) WARN_ON(n > 32); report += offset >> 3; /* adjust byte index */ - offset &= 7; /* now only need bit offset into one byte */ + offset &= 7; /* now only need bit offset into one byte */ x = get_unaligned((u64 *) report); x = le64_to_cpu(x); - x = (x >> offset) & ((1ULL << n) - 1); /* extract bit field */ + x = (x >> offset) & ((1ULL << n) - 1); /* extract bit field */ return (u32) x; } @@ -829,7 +829,7 @@ static void hid_process_event(struct hid_device *hid, struct hid_field *field, s * reporting to the layer). */ -static void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt) +void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt) { unsigned n; unsigned count = field->report_count; @@ -875,7 +875,7 @@ static void hid_input_field(struct hid_device *hid, struct hid_field *field, __u exit: kfree(value); } - +EXPORT_SYMBOL_GPL(hid_input_field); /* * Output the field into the report. @@ -900,7 +900,7 @@ static void hid_output_field(struct hid_field *field, __u8 *data) * Create a report. */ -static void hid_output_report(struct hid_report *report, __u8 *data) +void hid_output_report(struct hid_report *report, __u8 *data) { unsigned n; @@ -910,6 +910,7 @@ static void hid_output_report(struct hid_report *report, __u8 *data) for (n = 0; n < report->maxfield; n++) hid_output_field(report->field[n], data); } +EXPORT_SYMBOL_GPL(hid_output_report); /* * Set a field value. The report this field belongs to has to be @@ -937,4 +938,5 @@ int hid_set_field(struct hid_field *field, unsigned offset, __s32 value) field->value[offset] = value; return 0; } +EXPORT_SYMBOL_GPL(hid_set_field); diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index d459005062e0..6d3d80ba9582 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -727,8 +727,9 @@ void hidinput_report_event(struct hid_device *hid, struct hid_report *report) list_for_each_entry(hidinput, &hid->inputs, list) input_sync(hidinput->input); } +EXPORT_SYMBOL_GPL(hidinput_report_event); -static int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field) +int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field) { struct hid_report *report; int i, j; @@ -743,6 +744,7 @@ static int hidinput_find_field(struct hid_device *hid, unsigned int type, unsign } return -1; } +EXPORT_SYMBOL_GPL(hidinput_find_field); /* * Register the input device; print a message. @@ -752,7 +754,6 @@ static int hidinput_find_field(struct hid_device *hid, unsigned int type, unsign int hidinput_connect(struct hid_device *hid) { - struct usb_device *dev = hid->dev; struct hid_report *report; struct hid_input *hidinput = NULL; struct input_dev *input_dev; @@ -786,14 +787,17 @@ int hidinput_connect(struct hid_device *hid) } input_dev->private = hid; - input_dev->event = hidinput_input_event; - input_dev->open = hidinput_open; - input_dev->close = hidinput_close; + input_dev->event = hid->hidinput_input_event; + input_dev->open = hid->hidinput_open; + input_dev->close = hid->hidinput_close; input_dev->name = hid->name; input_dev->phys = hid->phys; input_dev->uniq = hid->uniq; - usb_to_input_id(dev, &input_dev->id); + input_dev->id.bustype = hid->bus; + input_dev->id.vendor = hid->vendor; + input_dev->id.product = hid->product; + input_dev->id.version = hid->version; input_dev->cdev.dev = &hid->intf->dev; hidinput->input = input_dev; @@ -827,6 +831,7 @@ int hidinput_connect(struct hid_device *hid) return 0; } +EXPORT_SYMBOL_GPL(hidinput_connect); void hidinput_disconnect(struct hid_device *hid) { @@ -838,3 +843,5 @@ void hidinput_disconnect(struct hid_device *hid) kfree(hidinput); } } +EXPORT_SYMBOL_GPL(hidinput_disconnect); + diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 06e169b6a17e..462947f74135 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -497,7 +497,7 @@ void hid_submit_report(struct hid_device *hid, struct hid_report *report, unsign spin_unlock_irqrestore(&hid->ctrllock, flags); } -static int hidinput_input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) +static int usb_hidinput_input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) { struct hid_device *hid = dev->private; struct hid_field *field; @@ -1231,6 +1231,10 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) le16_to_cpu(dev->descriptor.idVendor), le16_to_cpu(dev->descriptor.idProduct)); + hid->bus = BUS_USB; + hid->vendor = dev->descriptor.idVendor; + hid->product = dev->descriptor.idProduct; + usb_make_path(dev, hid->phys, sizeof(hid->phys)); strlcat(hid->phys, "/input", sizeof(hid->phys)); len = strlen(hid->phys); @@ -1250,6 +1254,9 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) hid->urbctrl->setup_dma = hid->cr_dma; hid->urbctrl->transfer_dma = hid->ctrlbuf_dma; hid->urbctrl->transfer_flags |= (URB_NO_TRANSFER_DMA_MAP | URB_NO_SETUP_DMA_MAP); + hid->hidinput_input_event = usb_hidinput_input_event; + hid->hidinput_open = hidinput_open; + hid->hidinput_close = hidinput_close; return hid; diff --git a/drivers/usb/input/hid-ff.c b/drivers/usb/input/hid-ff.c index 4187f4ee9a9c..7ecdafa8eb7b 100644 --- a/drivers/usb/input/hid-ff.c +++ b/drivers/usb/input/hid-ff.c @@ -79,3 +79,5 @@ int hid_ff_init(struct hid_device* hid) return init->init(hid); } +EXPORT_SYMBOL_GPL(hid_ff_init); + diff --git a/drivers/usb/input/hiddev.c b/drivers/usb/input/hiddev.c index cbd3b60d93bb..07d7996575cd 100644 --- a/drivers/usb/input/hiddev.c +++ b/drivers/usb/input/hiddev.c @@ -197,7 +197,7 @@ void hiddev_hid_event(struct hid_device *hid, struct hid_field *field, hiddev_send_event(hid, &uref); } - +EXPORT_SYMBOL_GPL(hiddev_hid_event); void hiddev_report_event(struct hid_device *hid, struct hid_report *report) { diff --git a/include/linux/hid.h b/include/linux/hid.h index ee567ae6fec1..0473b45b73b8 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -403,11 +403,17 @@ struct hid_device { /* device report descriptor */ unsigned collection_size; /* Number of allocated hid_collections */ unsigned maxcollection; /* Number of parsed collections */ unsigned maxapplication; /* Number of applications */ + unsigned short bus; /* BUS ID */ + unsigned short vendor; /* Vendor ID */ + unsigned short product; /* Product ID */ unsigned version; /* HID version */ unsigned country; /* HID country */ struct hid_report_enum report_enum[HID_REPORT_TYPES]; - struct usb_device *dev; /* USB device */ + struct usb_device *dev; /* device */ + + /* USB specific fields */ + struct usb_interface *intf; /* USB interface */ int ifnum; /* USB interface number */ @@ -454,6 +460,13 @@ struct hid_device { /* device report descriptor */ char phys[64]; /* Device physical location */ char uniq[64]; /* Device unique identifier (serial #) */ + void *driver_data; + + /* device-specific function pointers */ + int (*hidinput_input_event) (struct input_dev *, unsigned int, unsigned int, int); + int (*hidinput_open) (struct input_dev *); + void (*hidinput_close) (struct input_dev *); + #ifdef CONFIG_USB_HIDINPUT_POWERBOOK unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; @@ -502,17 +515,15 @@ struct hid_descriptor { /* Applications from HID Usage Tables 4/8/99 Version 1.1 */ /* We ignore a few input applications that are not widely used */ #define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001)) +#else +#define IS_INPUT_APPLICATION(a) (0) +#endif + +/* HID core API */ extern void hidinput_hid_event(struct hid_device *, struct hid_field *, struct hid_usage *, __s32); extern void hidinput_report_event(struct hid_device *hid, struct hid_report *report); extern int hidinput_connect(struct hid_device *); extern void hidinput_disconnect(struct hid_device *); -#else -#define IS_INPUT_APPLICATION(a) (0) -static inline void hidinput_hid_event(struct hid_device *hid, struct hid_field *field, struct hid_usage *usage, __s32 value) { } -static inline void hidinput_report_event(struct hid_device *hid, struct hid_report *report) { } -static inline int hidinput_connect(struct hid_device *hid) { return -ENODEV; } -static inline void hidinput_disconnect(struct hid_device *hid) { } -#endif int hid_set_field(struct hid_field *, unsigned, __s32); int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field); @@ -536,5 +547,14 @@ static inline int hid_pidff_init(struct hid_device *hid) { return -ENODEV; } #else static inline int hid_ff_init(struct hid_device *hid) { return -1; } #endif +#ifdef DEBUG +#define dbg(format, arg...) printk(KERN_DEBUG "%s: " format "\n" , \ + __FILE__ , ## arg) +#else +#define dbg(format, arg...) do {} while (0) +#endif + +#define err(format, arg...) printk(KERN_ERR "%s: " format "\n" , \ + __FILE__ , ## arg) #endif -- cgit v1.2.3 From 4916b3a57fc94664677d439b911b8aaf86c7ec23 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:41:03 +0100 Subject: [PATCH] Generic HID layer - USB API - 'dev' in struct hid_device changed from struct usb_device to struct device and fixed all the users - renamed functions which are part of USB HID API from 'hid_*' to 'usbhid_*' - force feedback initialization moved from common part into USB-specific driver - added usbhid.h header for USB HID API users - removed USB-specific fields from struct hid_device and moved them to new usbhid_device, which is pointed to by hid_device->driver_data - fixed all USB users to use this new structure Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-input.c | 10 +- drivers/usb/input/hid-core.c | 419 +++++++++++++++++++++++------------------- drivers/usb/input/hid-ff.c | 4 +- drivers/usb/input/hid-lgff.c | 5 +- drivers/usb/input/hid-pidff.c | 54 +++--- drivers/usb/input/hid-tmff.c | 3 +- drivers/usb/input/hid-zpff.c | 5 +- drivers/usb/input/hiddev.c | 31 ++-- drivers/usb/input/usbhid.h | 84 +++++++++ include/linux/hid.h | 36 +--- 10 files changed, 370 insertions(+), 281 deletions(-) create mode 100644 drivers/usb/input/usbhid.h (limited to 'include/linux') diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index 6d3d80ba9582..e542ef971c46 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -798,8 +798,7 @@ int hidinput_connect(struct hid_device *hid) input_dev->id.vendor = hid->vendor; input_dev->id.product = hid->product; input_dev->id.version = hid->version; - input_dev->cdev.dev = &hid->intf->dev; - + input_dev->cdev.dev = hid->dev; hidinput->input = input_dev; list_add_tail(&hidinput->list, &hid->inputs); } @@ -821,13 +820,8 @@ int hidinput_connect(struct hid_device *hid) } } - /* This only gets called when we are a single-input (most of the - * time). IOW, not a HID_QUIRK_MULTI_INPUT. The hid_ff_init() is - * only useful in this case, and not for multi-input quirks. */ - if (hidinput) { - hid_ff_init(hid); + if (hidinput) input_register_device(hidinput->input); - } return 0; } diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 462947f74135..0991c4b751a6 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -35,6 +35,7 @@ #include #include +#include "usbhid.h" /* * Version Information @@ -66,15 +67,16 @@ static int hid_start_in(struct hid_device *hid) { unsigned long flags; int rc = 0; + struct usbhid_device *usbhid = hid->driver_data; - spin_lock_irqsave(&hid->inlock, flags); - if (hid->open > 0 && !test_bit(HID_SUSPENDED, &hid->iofl) && - !test_and_set_bit(HID_IN_RUNNING, &hid->iofl)) { - rc = usb_submit_urb(hid->urbin, GFP_ATOMIC); + spin_lock_irqsave(&usbhid->inlock, flags); + if (hid->open > 0 && !test_bit(HID_SUSPENDED, &usbhid->iofl) && + !test_and_set_bit(HID_IN_RUNNING, &usbhid->iofl)) { + rc = usb_submit_urb(usbhid->urbin, GFP_ATOMIC); if (rc != 0) - clear_bit(HID_IN_RUNNING, &hid->iofl); + clear_bit(HID_IN_RUNNING, &usbhid->iofl); } - spin_unlock_irqrestore(&hid->inlock, flags); + spin_unlock_irqrestore(&usbhid->inlock, flags); return rc; } @@ -82,8 +84,9 @@ static int hid_start_in(struct hid_device *hid) static void hid_retry_timeout(unsigned long _hid) { struct hid_device *hid = (struct hid_device *) _hid; + struct usbhid_device *usbhid = hid->driver_data; - dev_dbg(&hid->intf->dev, "retrying intr urb\n"); + dev_dbg(&usbhid->intf->dev, "retrying intr urb\n"); if (hid_start_in(hid)) hid_io_error(hid); } @@ -91,38 +94,39 @@ static void hid_retry_timeout(unsigned long _hid) /* Workqueue routine to reset the device or clear a halt */ static void hid_reset(struct work_struct *work) { - struct hid_device *hid = - container_of(work, struct hid_device, reset_work); + struct usbhid_device *usbhid = + container_of(work, struct usbhid_device, reset_work); + struct hid_device *hid = usbhid->hid; int rc_lock, rc = 0; - if (test_bit(HID_CLEAR_HALT, &hid->iofl)) { - dev_dbg(&hid->intf->dev, "clear halt\n"); - rc = usb_clear_halt(hid->dev, hid->urbin->pipe); - clear_bit(HID_CLEAR_HALT, &hid->iofl); + if (test_bit(HID_CLEAR_HALT, &usbhid->iofl)) { + dev_dbg(&usbhid->intf->dev, "clear halt\n"); + rc = usb_clear_halt(to_usb_device(hid->dev), usbhid->urbin->pipe); + clear_bit(HID_CLEAR_HALT, &usbhid->iofl); hid_start_in(hid); } - else if (test_bit(HID_RESET_PENDING, &hid->iofl)) { - dev_dbg(&hid->intf->dev, "resetting device\n"); - rc = rc_lock = usb_lock_device_for_reset(hid->dev, hid->intf); + else if (test_bit(HID_RESET_PENDING, &usbhid->iofl)) { + dev_dbg(&usbhid->intf->dev, "resetting device\n"); + rc = rc_lock = usb_lock_device_for_reset(to_usb_device(hid->dev), usbhid->intf); if (rc_lock >= 0) { - rc = usb_reset_composite_device(hid->dev, hid->intf); + rc = usb_reset_composite_device(to_usb_device(hid->dev), usbhid->intf); if (rc_lock) - usb_unlock_device(hid->dev); + usb_unlock_device(to_usb_device(hid->dev)); } - clear_bit(HID_RESET_PENDING, &hid->iofl); + clear_bit(HID_RESET_PENDING, &usbhid->iofl); } switch (rc) { case 0: - if (!test_bit(HID_IN_RUNNING, &hid->iofl)) + if (!test_bit(HID_IN_RUNNING, &usbhid->iofl)) hid_io_error(hid); break; default: err("can't reset device, %s-%s/input%d, status %d", - hid->dev->bus->bus_name, - hid->dev->devpath, - hid->ifnum, rc); + to_usb_device(hid->dev)->bus->bus_name, + to_usb_device(hid->dev)->devpath, + usbhid->ifnum, rc); /* FALLTHROUGH */ case -EHOSTUNREACH: case -ENODEV: @@ -135,33 +139,34 @@ static void hid_reset(struct work_struct *work) static void hid_io_error(struct hid_device *hid) { unsigned long flags; + struct usbhid_device *usbhid = hid->driver_data; - spin_lock_irqsave(&hid->inlock, flags); + spin_lock_irqsave(&usbhid->inlock, flags); /* Stop when disconnected */ - if (usb_get_intfdata(hid->intf) == NULL) + if (usb_get_intfdata(usbhid->intf) == NULL) goto done; /* When an error occurs, retry at increasing intervals */ - if (hid->retry_delay == 0) { - hid->retry_delay = 13; /* Then 26, 52, 104, 104, ... */ - hid->stop_retry = jiffies + msecs_to_jiffies(1000); - } else if (hid->retry_delay < 100) - hid->retry_delay *= 2; + if (usbhid->retry_delay == 0) { + usbhid->retry_delay = 13; /* Then 26, 52, 104, 104, ... */ + usbhid->stop_retry = jiffies + msecs_to_jiffies(1000); + } else if (usbhid->retry_delay < 100) + usbhid->retry_delay *= 2; - if (time_after(jiffies, hid->stop_retry)) { + if (time_after(jiffies, usbhid->stop_retry)) { /* Retries failed, so do a port reset */ - if (!test_and_set_bit(HID_RESET_PENDING, &hid->iofl)) { - schedule_work(&hid->reset_work); + if (!test_and_set_bit(HID_RESET_PENDING, &usbhid->iofl)) { + schedule_work(&usbhid->reset_work); goto done; } } - mod_timer(&hid->io_retry, - jiffies + msecs_to_jiffies(hid->retry_delay)); + mod_timer(&usbhid->io_retry, + jiffies + msecs_to_jiffies(usbhid->retry_delay)); done: - spin_unlock_irqrestore(&hid->inlock, flags); + spin_unlock_irqrestore(&usbhid->inlock, flags); } @@ -230,28 +235,29 @@ static int hid_input_report(int type, struct urb *urb, int interrupt) static void hid_irq_in(struct urb *urb) { struct hid_device *hid = urb->context; + struct usbhid_device *usbhid = hid->driver_data; int status; switch (urb->status) { case 0: /* success */ - hid->retry_delay = 0; + usbhid->retry_delay = 0; hid_input_report(HID_INPUT_REPORT, urb, 1); break; case -EPIPE: /* stall */ - clear_bit(HID_IN_RUNNING, &hid->iofl); - set_bit(HID_CLEAR_HALT, &hid->iofl); - schedule_work(&hid->reset_work); + clear_bit(HID_IN_RUNNING, &usbhid->iofl); + set_bit(HID_CLEAR_HALT, &usbhid->iofl); + schedule_work(&usbhid->reset_work); return; case -ECONNRESET: /* unlink */ case -ENOENT: case -ESHUTDOWN: /* unplug */ - clear_bit(HID_IN_RUNNING, &hid->iofl); + clear_bit(HID_IN_RUNNING, &usbhid->iofl); return; case -EILSEQ: /* protocol error or unplug */ case -EPROTO: /* protocol error or unplug */ case -ETIME: /* protocol error or unplug */ case -ETIMEDOUT: /* Should never happen, but... */ - clear_bit(HID_IN_RUNNING, &hid->iofl); + clear_bit(HID_IN_RUNNING, &usbhid->iofl); hid_io_error(hid); return; default: /* error */ @@ -260,12 +266,12 @@ static void hid_irq_in(struct urb *urb) status = usb_submit_urb(urb, GFP_ATOMIC); if (status) { - clear_bit(HID_IN_RUNNING, &hid->iofl); + clear_bit(HID_IN_RUNNING, &usbhid->iofl); if (status != -EPERM) { err("can't resubmit intr, %s-%s/input%d, status %d", - hid->dev->bus->bus_name, - hid->dev->devpath, - hid->ifnum, status); + to_usb_device(hid->dev)->bus->bus_name, + to_usb_device(hid->dev)->devpath, + usbhid->ifnum, status); hid_io_error(hid); } } @@ -291,16 +297,17 @@ struct hid_field *hid_find_field_by_usage(struct hid_device *hid, __u32 wanted_u static int hid_submit_out(struct hid_device *hid) { struct hid_report *report; + struct usbhid_device *usbhid = hid->driver_data; - report = hid->out[hid->outtail]; + report = usbhid->out[usbhid->outtail]; - hid_output_report(report, hid->outbuf); - hid->urbout->transfer_buffer_length = ((report->size - 1) >> 3) + 1 + (report->id > 0); - hid->urbout->dev = hid->dev; + hid_output_report(report, usbhid->outbuf); + usbhid->urbout->transfer_buffer_length = ((report->size - 1) >> 3) + 1 + (report->id > 0); + usbhid->urbout->dev = to_usb_device(hid->dev); dbg("submitting out urb"); - if (usb_submit_urb(hid->urbout, GFP_ATOMIC)) { + if (usb_submit_urb(usbhid->urbout, GFP_ATOMIC)) { err("usb_submit_urb(out) failed"); return -1; } @@ -313,42 +320,43 @@ static int hid_submit_ctrl(struct hid_device *hid) struct hid_report *report; unsigned char dir; int len; + struct usbhid_device *usbhid = hid->driver_data; - report = hid->ctrl[hid->ctrltail].report; - dir = hid->ctrl[hid->ctrltail].dir; + report = usbhid->ctrl[usbhid->ctrltail].report; + dir = usbhid->ctrl[usbhid->ctrltail].dir; len = ((report->size - 1) >> 3) + 1 + (report->id > 0); if (dir == USB_DIR_OUT) { - hid_output_report(report, hid->ctrlbuf); - hid->urbctrl->pipe = usb_sndctrlpipe(hid->dev, 0); - hid->urbctrl->transfer_buffer_length = len; + hid_output_report(report, usbhid->ctrlbuf); + usbhid->urbctrl->pipe = usb_sndctrlpipe(to_usb_device(hid->dev), 0); + usbhid->urbctrl->transfer_buffer_length = len; } else { int maxpacket, padlen; - hid->urbctrl->pipe = usb_rcvctrlpipe(hid->dev, 0); - maxpacket = usb_maxpacket(hid->dev, hid->urbctrl->pipe, 0); + usbhid->urbctrl->pipe = usb_rcvctrlpipe(to_usb_device(hid->dev), 0); + maxpacket = usb_maxpacket(to_usb_device(hid->dev), usbhid->urbctrl->pipe, 0); if (maxpacket > 0) { padlen = (len + maxpacket - 1) / maxpacket; padlen *= maxpacket; - if (padlen > hid->bufsize) - padlen = hid->bufsize; + if (padlen > usbhid->bufsize) + padlen = usbhid->bufsize; } else padlen = 0; - hid->urbctrl->transfer_buffer_length = padlen; + usbhid->urbctrl->transfer_buffer_length = padlen; } - hid->urbctrl->dev = hid->dev; + usbhid->urbctrl->dev = to_usb_device(hid->dev); - hid->cr->bRequestType = USB_TYPE_CLASS | USB_RECIP_INTERFACE | dir; - hid->cr->bRequest = (dir == USB_DIR_OUT) ? HID_REQ_SET_REPORT : HID_REQ_GET_REPORT; - hid->cr->wValue = cpu_to_le16(((report->type + 1) << 8) | report->id); - hid->cr->wIndex = cpu_to_le16(hid->ifnum); - hid->cr->wLength = cpu_to_le16(len); + usbhid->cr->bRequestType = USB_TYPE_CLASS | USB_RECIP_INTERFACE | dir; + usbhid->cr->bRequest = (dir == USB_DIR_OUT) ? HID_REQ_SET_REPORT : HID_REQ_GET_REPORT; + usbhid->cr->wValue = cpu_to_le16(((report->type + 1) << 8) | report->id); + usbhid->cr->wIndex = cpu_to_le16(usbhid->ifnum); + usbhid->cr->wLength = cpu_to_le16(len); dbg("submitting ctrl urb: %s wValue=0x%04x wIndex=0x%04x wLength=%u", - hid->cr->bRequest == HID_REQ_SET_REPORT ? "Set_Report" : "Get_Report", - hid->cr->wValue, hid->cr->wIndex, hid->cr->wLength); + usbhid->cr->bRequest == HID_REQ_SET_REPORT ? "Set_Report" : "Get_Report", + usbhid->cr->wValue, usbhid->cr->wIndex, usbhid->cr->wLength); - if (usb_submit_urb(hid->urbctrl, GFP_ATOMIC)) { + if (usb_submit_urb(usbhid->urbctrl, GFP_ATOMIC)) { err("usb_submit_urb(ctrl) failed"); return -1; } @@ -363,6 +371,7 @@ static int hid_submit_ctrl(struct hid_device *hid) static void hid_irq_out(struct urb *urb) { struct hid_device *hid = urb->context; + struct usbhid_device *usbhid = hid->driver_data; unsigned long flags; int unplug = 0; @@ -380,24 +389,24 @@ static void hid_irq_out(struct urb *urb) warn("output irq status %d received", urb->status); } - spin_lock_irqsave(&hid->outlock, flags); + spin_lock_irqsave(&usbhid->outlock, flags); if (unplug) - hid->outtail = hid->outhead; + usbhid->outtail = usbhid->outhead; else - hid->outtail = (hid->outtail + 1) & (HID_OUTPUT_FIFO_SIZE - 1); + usbhid->outtail = (usbhid->outtail + 1) & (HID_OUTPUT_FIFO_SIZE - 1); - if (hid->outhead != hid->outtail) { + if (usbhid->outhead != usbhid->outtail) { if (hid_submit_out(hid)) { - clear_bit(HID_OUT_RUNNING, &hid->iofl); + clear_bit(HID_OUT_RUNNING, &usbhid->iofl); wake_up(&hid->wait); } - spin_unlock_irqrestore(&hid->outlock, flags); + spin_unlock_irqrestore(&usbhid->outlock, flags); return; } - clear_bit(HID_OUT_RUNNING, &hid->iofl); - spin_unlock_irqrestore(&hid->outlock, flags); + clear_bit(HID_OUT_RUNNING, &usbhid->iofl); + spin_unlock_irqrestore(&usbhid->outlock, flags); wake_up(&hid->wait); } @@ -408,15 +417,16 @@ static void hid_irq_out(struct urb *urb) static void hid_ctrl(struct urb *urb) { struct hid_device *hid = urb->context; + struct usbhid_device *usbhid = hid->driver_data; unsigned long flags; int unplug = 0; - spin_lock_irqsave(&hid->ctrllock, flags); + spin_lock_irqsave(&usbhid->ctrllock, flags); switch (urb->status) { case 0: /* success */ - if (hid->ctrl[hid->ctrltail].dir == USB_DIR_IN) - hid_input_report(hid->ctrl[hid->ctrltail].report->type, urb, 0); + if (usbhid->ctrl[usbhid->ctrltail].dir == USB_DIR_IN) + hid_input_report(usbhid->ctrl[usbhid->ctrltail].report->type, urb, 0); break; case -ESHUTDOWN: /* unplug */ unplug = 1; @@ -431,70 +441,71 @@ static void hid_ctrl(struct urb *urb) } if (unplug) - hid->ctrltail = hid->ctrlhead; + usbhid->ctrltail = usbhid->ctrlhead; else - hid->ctrltail = (hid->ctrltail + 1) & (HID_CONTROL_FIFO_SIZE - 1); + usbhid->ctrltail = (usbhid->ctrltail + 1) & (HID_CONTROL_FIFO_SIZE - 1); - if (hid->ctrlhead != hid->ctrltail) { + if (usbhid->ctrlhead != usbhid->ctrltail) { if (hid_submit_ctrl(hid)) { - clear_bit(HID_CTRL_RUNNING, &hid->iofl); + clear_bit(HID_CTRL_RUNNING, &usbhid->iofl); wake_up(&hid->wait); } - spin_unlock_irqrestore(&hid->ctrllock, flags); + spin_unlock_irqrestore(&usbhid->ctrllock, flags); return; } - clear_bit(HID_CTRL_RUNNING, &hid->iofl); - spin_unlock_irqrestore(&hid->ctrllock, flags); + clear_bit(HID_CTRL_RUNNING, &usbhid->iofl); + spin_unlock_irqrestore(&usbhid->ctrllock, flags); wake_up(&hid->wait); } -void hid_submit_report(struct hid_device *hid, struct hid_report *report, unsigned char dir) +void usbhid_submit_report(struct hid_device *hid, struct hid_report *report, unsigned char dir) { int head; unsigned long flags; + struct usbhid_device *usbhid = hid->driver_data; if ((hid->quirks & HID_QUIRK_NOGET) && dir == USB_DIR_IN) return; - if (hid->urbout && dir == USB_DIR_OUT && report->type == HID_OUTPUT_REPORT) { + if (usbhid->urbout && dir == USB_DIR_OUT && report->type == HID_OUTPUT_REPORT) { - spin_lock_irqsave(&hid->outlock, flags); + spin_lock_irqsave(&usbhid->outlock, flags); - if ((head = (hid->outhead + 1) & (HID_OUTPUT_FIFO_SIZE - 1)) == hid->outtail) { - spin_unlock_irqrestore(&hid->outlock, flags); + if ((head = (usbhid->outhead + 1) & (HID_OUTPUT_FIFO_SIZE - 1)) == usbhid->outtail) { + spin_unlock_irqrestore(&usbhid->outlock, flags); warn("output queue full"); return; } - hid->out[hid->outhead] = report; - hid->outhead = head; + usbhid->out[usbhid->outhead] = report; + usbhid->outhead = head; - if (!test_and_set_bit(HID_OUT_RUNNING, &hid->iofl)) + if (!test_and_set_bit(HID_OUT_RUNNING, &usbhid->iofl)) if (hid_submit_out(hid)) - clear_bit(HID_OUT_RUNNING, &hid->iofl); + clear_bit(HID_OUT_RUNNING, &usbhid->iofl); - spin_unlock_irqrestore(&hid->outlock, flags); + spin_unlock_irqrestore(&usbhid->outlock, flags); return; } - spin_lock_irqsave(&hid->ctrllock, flags); + spin_lock_irqsave(&usbhid->ctrllock, flags); - if ((head = (hid->ctrlhead + 1) & (HID_CONTROL_FIFO_SIZE - 1)) == hid->ctrltail) { - spin_unlock_irqrestore(&hid->ctrllock, flags); + if ((head = (usbhid->ctrlhead + 1) & (HID_CONTROL_FIFO_SIZE - 1)) == usbhid->ctrltail) { + spin_unlock_irqrestore(&usbhid->ctrllock, flags); warn("control queue full"); return; } - hid->ctrl[hid->ctrlhead].report = report; - hid->ctrl[hid->ctrlhead].dir = dir; - hid->ctrlhead = head; + usbhid->ctrl[usbhid->ctrlhead].report = report; + usbhid->ctrl[usbhid->ctrlhead].dir = dir; + usbhid->ctrlhead = head; - if (!test_and_set_bit(HID_CTRL_RUNNING, &hid->iofl)) + if (!test_and_set_bit(HID_CTRL_RUNNING, &usbhid->iofl)) if (hid_submit_ctrl(hid)) - clear_bit(HID_CTRL_RUNNING, &hid->iofl); + clear_bit(HID_CTRL_RUNNING, &usbhid->iofl); - spin_unlock_irqrestore(&hid->ctrllock, flags); + spin_unlock_irqrestore(&usbhid->ctrllock, flags); } static int usb_hidinput_input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value) @@ -515,15 +526,17 @@ static int usb_hidinput_input_event(struct input_dev *dev, unsigned int type, un } hid_set_field(field, offset, value); - hid_submit_report(hid, field->report, USB_DIR_OUT); + usbhid_submit_report(hid, field->report, USB_DIR_OUT); return 0; } -int hid_wait_io(struct hid_device *hid) +int usbhid_wait_io(struct hid_device *hid) { - if (!wait_event_timeout(hid->wait, (!test_bit(HID_CTRL_RUNNING, &hid->iofl) && - !test_bit(HID_OUT_RUNNING, &hid->iofl)), + struct usbhid_device *usbhid = hid->driver_data; + + if (!wait_event_timeout(hid->wait, (!test_bit(HID_CTRL_RUNNING, &usbhid->iofl) && + !test_bit(HID_OUT_RUNNING, &usbhid->iofl)), 10*HZ)) { dbg("timeout waiting for ctrl or out queue to clear"); return -1; @@ -555,7 +568,7 @@ static int hid_get_class_descriptor(struct usb_device *dev, int ifnum, return result; } -int hid_open(struct hid_device *hid) +int usbhid_open(struct hid_device *hid) { ++hid->open; if (hid_start_in(hid)) @@ -563,22 +576,24 @@ int hid_open(struct hid_device *hid) return 0; } -void hid_close(struct hid_device *hid) +void usbhid_close(struct hid_device *hid) { + struct usbhid_device *usbhid = hid->driver_data; + if (!--hid->open) - usb_kill_urb(hid->urbin); + usb_kill_urb(usbhid->urbin); } static int hidinput_open(struct input_dev *dev) { struct hid_device *hid = dev->private; - return hid_open(hid); + return usbhid_open(hid); } static void hidinput_close(struct input_dev *dev) { struct hid_device *hid = dev->private; - hid_close(hid); + usbhid_close(hid); } #define USB_VENDOR_ID_PANJIT 0x134c @@ -590,26 +605,27 @@ static void hidinput_close(struct input_dev *dev) * Initialize all reports */ -void hid_init_reports(struct hid_device *hid) +void usbhid_init_reports(struct hid_device *hid) { struct hid_report *report; + struct usbhid_device *usbhid = hid->driver_data; int err, ret; list_for_each_entry(report, &hid->report_enum[HID_INPUT_REPORT].report_list, list) - hid_submit_report(hid, report, USB_DIR_IN); + usbhid_submit_report(hid, report, USB_DIR_IN); list_for_each_entry(report, &hid->report_enum[HID_FEATURE_REPORT].report_list, list) - hid_submit_report(hid, report, USB_DIR_IN); + usbhid_submit_report(hid, report, USB_DIR_IN); err = 0; - ret = hid_wait_io(hid); + ret = usbhid_wait_io(hid); while (ret) { err |= ret; - if (test_bit(HID_CTRL_RUNNING, &hid->iofl)) - usb_kill_urb(hid->urbctrl); - if (test_bit(HID_OUT_RUNNING, &hid->iofl)) - usb_kill_urb(hid->urbout); - ret = hid_wait_io(hid); + if (test_bit(HID_CTRL_RUNNING, &usbhid->iofl)) + usb_kill_urb(usbhid->urbctrl); + if (test_bit(HID_OUT_RUNNING, &usbhid->iofl)) + usb_kill_urb(usbhid->urbout); + ret = usbhid_wait_io(hid); } if (err) @@ -1022,13 +1038,15 @@ static void hid_find_max_report(struct hid_device *hid, unsigned int type, int * static int hid_alloc_buffers(struct usb_device *dev, struct hid_device *hid) { - if (!(hid->inbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->inbuf_dma))) + struct usbhid_device *usbhid = hid->driver_data; + + if (!(usbhid->inbuf = usb_buffer_alloc(dev, usbhid->bufsize, GFP_ATOMIC, &usbhid->inbuf_dma))) return -1; - if (!(hid->outbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->outbuf_dma))) + if (!(usbhid->outbuf = usb_buffer_alloc(dev, usbhid->bufsize, GFP_ATOMIC, &usbhid->outbuf_dma))) return -1; - if (!(hid->cr = usb_buffer_alloc(dev, sizeof(*(hid->cr)), GFP_ATOMIC, &hid->cr_dma))) + if (!(usbhid->cr = usb_buffer_alloc(dev, sizeof(*(usbhid->cr)), GFP_ATOMIC, &usbhid->cr_dma))) return -1; - if (!(hid->ctrlbuf = usb_buffer_alloc(dev, hid->bufsize, GFP_ATOMIC, &hid->ctrlbuf_dma))) + if (!(usbhid->ctrlbuf = usb_buffer_alloc(dev, usbhid->bufsize, GFP_ATOMIC, &usbhid->ctrlbuf_dma))) return -1; return 0; @@ -1036,14 +1054,16 @@ static int hid_alloc_buffers(struct usb_device *dev, struct hid_device *hid) static void hid_free_buffers(struct usb_device *dev, struct hid_device *hid) { - if (hid->inbuf) - usb_buffer_free(dev, hid->bufsize, hid->inbuf, hid->inbuf_dma); - if (hid->outbuf) - usb_buffer_free(dev, hid->bufsize, hid->outbuf, hid->outbuf_dma); - if (hid->cr) - usb_buffer_free(dev, sizeof(*(hid->cr)), hid->cr, hid->cr_dma); - if (hid->ctrlbuf) - usb_buffer_free(dev, hid->bufsize, hid->ctrlbuf, hid->ctrlbuf_dma); + struct usbhid_device *usbhid = hid->driver_data; + + if (usbhid->inbuf) + usb_buffer_free(dev, usbhid->bufsize, usbhid->inbuf, usbhid->inbuf_dma); + if (usbhid->outbuf) + usb_buffer_free(dev, usbhid->bufsize, usbhid->outbuf, usbhid->outbuf_dma); + if (usbhid->cr) + usb_buffer_free(dev, sizeof(*(usbhid->cr)), usbhid->cr, usbhid->cr_dma); + if (usbhid->ctrlbuf) + usb_buffer_free(dev, usbhid->bufsize, usbhid->ctrlbuf, usbhid->ctrlbuf_dma); } /* @@ -1069,6 +1089,7 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) unsigned quirks = 0, rsize = 0; char *rdesc; int n, len, insize = 0; + struct usbhid_device *usbhid; /* Ignore all Wacom devices */ if (le16_to_cpu(dev->descriptor.idVendor) == USB_VENDOR_ID_WACOM) @@ -1138,13 +1159,19 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) kfree(rdesc); hid->quirks = quirks; - hid->bufsize = HID_MIN_BUFFER_SIZE; - hid_find_max_report(hid, HID_INPUT_REPORT, &hid->bufsize); - hid_find_max_report(hid, HID_OUTPUT_REPORT, &hid->bufsize); - hid_find_max_report(hid, HID_FEATURE_REPORT, &hid->bufsize); + if (!(usbhid = kzalloc(sizeof(struct usbhid_device), GFP_KERNEL))) + goto fail; + + hid->driver_data = usbhid; + usbhid->hid = hid; + + usbhid->bufsize = HID_MIN_BUFFER_SIZE; + hid_find_max_report(hid, HID_INPUT_REPORT, &usbhid->bufsize); + hid_find_max_report(hid, HID_OUTPUT_REPORT, &usbhid->bufsize); + hid_find_max_report(hid, HID_FEATURE_REPORT, &usbhid->bufsize); - if (hid->bufsize > HID_MAX_BUFFER_SIZE) - hid->bufsize = HID_MAX_BUFFER_SIZE; + if (usbhid->bufsize > HID_MAX_BUFFER_SIZE) + usbhid->bufsize = HID_MAX_BUFFER_SIZE; hid_find_max_report(hid, HID_INPUT_REPORT, &insize); @@ -1173,47 +1200,47 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) interval = hid_mousepoll_interval; if (usb_endpoint_dir_in(endpoint)) { - if (hid->urbin) + if (usbhid->urbin) continue; - if (!(hid->urbin = usb_alloc_urb(0, GFP_KERNEL))) + if (!(usbhid->urbin = usb_alloc_urb(0, GFP_KERNEL))) goto fail; pipe = usb_rcvintpipe(dev, endpoint->bEndpointAddress); - usb_fill_int_urb(hid->urbin, dev, pipe, hid->inbuf, insize, + usb_fill_int_urb(usbhid->urbin, dev, pipe, usbhid->inbuf, insize, hid_irq_in, hid, interval); - hid->urbin->transfer_dma = hid->inbuf_dma; - hid->urbin->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; + usbhid->urbin->transfer_dma = usbhid->inbuf_dma; + usbhid->urbin->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; } else { - if (hid->urbout) + if (usbhid->urbout) continue; - if (!(hid->urbout = usb_alloc_urb(0, GFP_KERNEL))) + if (!(usbhid->urbout = usb_alloc_urb(0, GFP_KERNEL))) goto fail; pipe = usb_sndintpipe(dev, endpoint->bEndpointAddress); - usb_fill_int_urb(hid->urbout, dev, pipe, hid->outbuf, 0, + usb_fill_int_urb(usbhid->urbout, dev, pipe, usbhid->outbuf, 0, hid_irq_out, hid, interval); - hid->urbout->transfer_dma = hid->outbuf_dma; - hid->urbout->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; + usbhid->urbout->transfer_dma = usbhid->outbuf_dma; + usbhid->urbout->transfer_flags |= URB_NO_TRANSFER_DMA_MAP; } } - if (!hid->urbin) { + if (!usbhid->urbin) { err("couldn't find an input interrupt endpoint"); goto fail; } init_waitqueue_head(&hid->wait); - INIT_WORK(&hid->reset_work, hid_reset); - setup_timer(&hid->io_retry, hid_retry_timeout, (unsigned long) hid); + INIT_WORK(&usbhid->reset_work, hid_reset); + setup_timer(&usbhid->io_retry, hid_retry_timeout, (unsigned long) hid); - spin_lock_init(&hid->inlock); - spin_lock_init(&hid->outlock); - spin_lock_init(&hid->ctrllock); + spin_lock_init(&usbhid->inlock); + spin_lock_init(&usbhid->outlock); + spin_lock_init(&usbhid->ctrllock); hid->version = le16_to_cpu(hdesc->bcdHID); hid->country = hdesc->bCountryCode; - hid->dev = dev; - hid->intf = intf; - hid->ifnum = interface->desc.bInterfaceNumber; + hid->dev = &dev->dev; + usbhid->intf = intf; + usbhid->ifnum = interface->desc.bInterfaceNumber; hid->name[0] = 0; @@ -1245,15 +1272,15 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) if (usb_string(dev, dev->descriptor.iSerialNumber, hid->uniq, 64) <= 0) hid->uniq[0] = 0; - hid->urbctrl = usb_alloc_urb(0, GFP_KERNEL); - if (!hid->urbctrl) + usbhid->urbctrl = usb_alloc_urb(0, GFP_KERNEL); + if (!usbhid->urbctrl) goto fail; - usb_fill_control_urb(hid->urbctrl, dev, 0, (void *) hid->cr, - hid->ctrlbuf, 1, hid_ctrl, hid); - hid->urbctrl->setup_dma = hid->cr_dma; - hid->urbctrl->transfer_dma = hid->ctrlbuf_dma; - hid->urbctrl->transfer_flags |= (URB_NO_TRANSFER_DMA_MAP | URB_NO_SETUP_DMA_MAP); + usb_fill_control_urb(usbhid->urbctrl, dev, 0, (void *) usbhid->cr, + usbhid->ctrlbuf, 1, hid_ctrl, hid); + usbhid->urbctrl->setup_dma = usbhid->cr_dma; + usbhid->urbctrl->transfer_dma = usbhid->ctrlbuf_dma; + usbhid->urbctrl->transfer_flags |= (URB_NO_TRANSFER_DMA_MAP | URB_NO_SETUP_DMA_MAP); hid->hidinput_input_event = usb_hidinput_input_event; hid->hidinput_open = hidinput_open; hid->hidinput_close = hidinput_close; @@ -1261,9 +1288,9 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) return hid; fail: - usb_free_urb(hid->urbin); - usb_free_urb(hid->urbout); - usb_free_urb(hid->urbctrl); + usb_free_urb(usbhid->urbin); + usb_free_urb(usbhid->urbout); + usb_free_urb(usbhid->urbctrl); hid_free_buffers(dev, hid); hid_free_device(hid); @@ -1273,18 +1300,21 @@ fail: static void hid_disconnect(struct usb_interface *intf) { struct hid_device *hid = usb_get_intfdata (intf); + struct usbhid_device *usbhid; if (!hid) return; - spin_lock_irq(&hid->inlock); /* Sync with error handler */ + usbhid = hid->driver_data; + + spin_lock_irq(&usbhid->inlock); /* Sync with error handler */ usb_set_intfdata(intf, NULL); - spin_unlock_irq(&hid->inlock); - usb_kill_urb(hid->urbin); - usb_kill_urb(hid->urbout); - usb_kill_urb(hid->urbctrl); + spin_unlock_irq(&usbhid->inlock); + usb_kill_urb(usbhid->urbin); + usb_kill_urb(usbhid->urbout); + usb_kill_urb(usbhid->urbctrl); - del_timer_sync(&hid->io_retry); + del_timer_sync(&usbhid->io_retry); flush_scheduled_work(); if (hid->claimed & HID_CLAIMED_INPUT) @@ -1292,11 +1322,11 @@ static void hid_disconnect(struct usb_interface *intf) if (hid->claimed & HID_CLAIMED_HIDDEV) hiddev_disconnect(hid); - usb_free_urb(hid->urbin); - usb_free_urb(hid->urbctrl); - usb_free_urb(hid->urbout); + usb_free_urb(usbhid->urbin); + usb_free_urb(usbhid->urbctrl); + usb_free_urb(usbhid->urbout); - hid_free_buffers(hid->dev, hid); + hid_free_buffers(to_usb_device(hid->dev), hid); hid_free_device(hid); } @@ -1313,7 +1343,7 @@ static int hid_probe(struct usb_interface *intf, const struct usb_device_id *id) if (!(hid = usb_hid_configure(intf))) return -ENODEV; - hid_init_reports(hid); + usbhid_init_reports(hid); hid_dump_device(hid); if (!hidinput_connect(hid)) @@ -1329,6 +1359,13 @@ static int hid_probe(struct usb_interface *intf, const struct usb_device_id *id) return -ENODEV; } + /* This only gets called when we are a single-input (most of the + * time). IOW, not a HID_QUIRK_MULTI_INPUT. The hid_ff_init() is + * only useful in this case, and not for multi-input quirks. */ + if ((hid->claimed & HID_CLAIMED_INPUT) && + !(hid->quirks & HID_QUIRK_MULTI_INPUT)) + hid_ff_init(hid); + printk(KERN_INFO); if (hid->claimed & HID_CLAIMED_INPUT) @@ -1359,12 +1396,13 @@ static int hid_probe(struct usb_interface *intf, const struct usb_device_id *id) static int hid_suspend(struct usb_interface *intf, pm_message_t message) { struct hid_device *hid = usb_get_intfdata (intf); + struct usbhid_device *usbhid = hid->driver_data; - spin_lock_irq(&hid->inlock); /* Sync with error handler */ - set_bit(HID_SUSPENDED, &hid->iofl); - spin_unlock_irq(&hid->inlock); - del_timer(&hid->io_retry); - usb_kill_urb(hid->urbin); + spin_lock_irq(&usbhid->inlock); /* Sync with error handler */ + set_bit(HID_SUSPENDED, &usbhid->iofl); + spin_unlock_irq(&usbhid->inlock); + del_timer(&usbhid->io_retry); + usb_kill_urb(usbhid->urbin); dev_dbg(&intf->dev, "suspend\n"); return 0; } @@ -1372,10 +1410,11 @@ static int hid_suspend(struct usb_interface *intf, pm_message_t message) static int hid_resume(struct usb_interface *intf) { struct hid_device *hid = usb_get_intfdata (intf); + struct usbhid_device *usbhid = hid->driver_data; int status; - clear_bit(HID_SUSPENDED, &hid->iofl); - hid->retry_delay = 0; + clear_bit(HID_SUSPENDED, &usbhid->iofl); + usbhid->retry_delay = 0; status = hid_start_in(hid); dev_dbg(&intf->dev, "resume status %d\n", status); return status; diff --git a/drivers/usb/input/hid-ff.c b/drivers/usb/input/hid-ff.c index 7ecdafa8eb7b..f8f660ee3fac 100644 --- a/drivers/usb/input/hid-ff.c +++ b/drivers/usb/input/hid-ff.c @@ -70,8 +70,8 @@ static struct hid_ff_initializer inits[] = { int hid_ff_init(struct hid_device* hid) { struct hid_ff_initializer *init; - int vendor = le16_to_cpu(hid->dev->descriptor.idVendor); - int product = le16_to_cpu(hid->dev->descriptor.idProduct); + int vendor = le16_to_cpu(to_usb_device(hid->dev)->descriptor.idVendor); + int product = le16_to_cpu(to_usb_device(hid->dev)->descriptor.idProduct); for (init = inits; init->idVendor; init++) if (init->idVendor == vendor && init->idProduct == product) diff --git a/drivers/usb/input/hid-lgff.c b/drivers/usb/input/hid-lgff.c index e977ba3d17e0..e47466268565 100644 --- a/drivers/usb/input/hid-lgff.c +++ b/drivers/usb/input/hid-lgff.c @@ -30,6 +30,7 @@ #include #include #include +#include "usbhid.h" struct device_type { u16 idVendor; @@ -75,7 +76,7 @@ static int hid_lgff_play(struct input_dev *dev, void *data, struct ff_effect *ef report->field[0]->value[2] = x; report->field[0]->value[3] = y; dbg("(x, y)=(%04x, %04x)", x, y); - hid_submit_report(hid, report, USB_DIR_OUT); + usbhid_submit_report(hid, report, USB_DIR_OUT); break; case FF_RUMBLE: @@ -90,7 +91,7 @@ static int hid_lgff_play(struct input_dev *dev, void *data, struct ff_effect *ef report->field[0]->value[2] = left; report->field[0]->value[3] = right; dbg("(left, right)=(%04x, %04x)", left, right); - hid_submit_report(hid, report, USB_DIR_OUT); + usbhid_submit_report(hid, report, USB_DIR_OUT); break; } return 0; diff --git a/drivers/usb/input/hid-pidff.c b/drivers/usb/input/hid-pidff.c index b4caea3864e3..cbd2d53fefff 100644 --- a/drivers/usb/input/hid-pidff.c +++ b/drivers/usb/input/hid-pidff.c @@ -262,7 +262,7 @@ static void pidff_set_envelope_report(struct pidff_device *pidff, debug("attack %u => %d", envelope->attack_level, pidff->set_envelope[PID_ATTACK_LEVEL].value[0]); - hid_submit_report(pidff->hid, pidff->reports[PID_SET_ENVELOPE], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_ENVELOPE], USB_DIR_OUT); } @@ -289,7 +289,7 @@ static void pidff_set_constant_force_report(struct pidff_device *pidff, pidff_set_signed(&pidff->set_constant[PID_MAGNITUDE], effect->u.constant.level); - hid_submit_report(pidff->hid, pidff->reports[PID_SET_CONSTANT], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_CONSTANT], USB_DIR_OUT); } @@ -324,7 +324,7 @@ static void pidff_set_effect_report(struct pidff_device *pidff, pidff->effect_direction); pidff->set_effect[PID_START_DELAY].value[0] = effect->replay.delay; - hid_submit_report(pidff->hid, pidff->reports[PID_SET_EFFECT], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_EFFECT], USB_DIR_OUT); } @@ -356,7 +356,7 @@ static void pidff_set_periodic_report(struct pidff_device *pidff, pidff_set(&pidff->set_periodic[PID_PHASE], effect->u.periodic.phase); pidff->set_periodic[PID_PERIOD].value[0] = effect->u.periodic.period; - hid_submit_report(pidff->hid, pidff->reports[PID_SET_PERIODIC], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_PERIODIC], USB_DIR_OUT); } @@ -398,8 +398,8 @@ static void pidff_set_condition_report(struct pidff_device *pidff, effect->u.condition[i].left_saturation); pidff_set(&pidff->set_condition[PID_DEAD_BAND], effect->u.condition[i].deadband); - hid_wait_io(pidff->hid); - hid_submit_report(pidff->hid, pidff->reports[PID_SET_CONDITION], + usbhid_wait_io(pidff->hid); + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_CONDITION], USB_DIR_OUT); } } @@ -440,7 +440,7 @@ static void pidff_set_ramp_force_report(struct pidff_device *pidff, effect->u.ramp.start_level); pidff_set_signed(&pidff->set_ramp[PID_RAMP_END], effect->u.ramp.end_level); - hid_submit_report(pidff->hid, pidff->reports[PID_SET_RAMP], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_RAMP], USB_DIR_OUT); } @@ -465,19 +465,19 @@ static int pidff_request_effect_upload(struct pidff_device *pidff, int efnum) int j; pidff->create_new_effect_type->value[0] = efnum; - hid_submit_report(pidff->hid, pidff->reports[PID_CREATE_NEW_EFFECT], + usbhid_submit_report(pidff->hid, pidff->reports[PID_CREATE_NEW_EFFECT], USB_DIR_OUT); debug("create_new_effect sent, type: %d", efnum); pidff->block_load[PID_EFFECT_BLOCK_INDEX].value[0] = 0; pidff->block_load_status->value[0] = 0; - hid_wait_io(pidff->hid); + usbhid_wait_io(pidff->hid); for (j = 0; j < 60; j++) { debug("pid_block_load requested"); - hid_submit_report(pidff->hid, pidff->reports[PID_BLOCK_LOAD], + usbhid_submit_report(pidff->hid, pidff->reports[PID_BLOCK_LOAD], USB_DIR_IN); - hid_wait_io(pidff->hid); + usbhid_wait_io(pidff->hid); if (pidff->block_load_status->value[0] == pidff->status_id[PID_BLOCK_LOAD_SUCCESS]) { debug("device reported free memory: %d bytes", @@ -513,8 +513,8 @@ static void pidff_playback_pid(struct pidff_device *pidff, int pid_id, int n) pidff->effect_operation[PID_LOOP_COUNT].value[0] = n; } - hid_wait_io(pidff->hid); - hid_submit_report(pidff->hid, pidff->reports[PID_EFFECT_OPERATION], + usbhid_wait_io(pidff->hid); + usbhid_submit_report(pidff->hid, pidff->reports[PID_EFFECT_OPERATION], USB_DIR_OUT); } @@ -536,7 +536,7 @@ static int pidff_playback(struct input_dev *dev, int effect_id, int value) static void pidff_erase_pid(struct pidff_device *pidff, int pid_id) { pidff->block_free[PID_EFFECT_BLOCK_INDEX].value[0] = pid_id; - hid_submit_report(pidff->hid, pidff->reports[PID_BLOCK_FREE], + usbhid_submit_report(pidff->hid, pidff->reports[PID_BLOCK_FREE], USB_DIR_OUT); } @@ -716,7 +716,7 @@ static void pidff_set_gain(struct input_dev *dev, u16 gain) struct pidff_device *pidff = dev->ff->private; pidff_set(&pidff->device_gain[PID_DEVICE_GAIN_FIELD], gain); - hid_submit_report(pidff->hid, pidff->reports[PID_DEVICE_GAIN], + usbhid_submit_report(pidff->hid, pidff->reports[PID_DEVICE_GAIN], USB_DIR_OUT); } @@ -741,7 +741,7 @@ static void pidff_autocenter(struct pidff_device *pidff, u16 magnitude) pidff_set(&pidff->set_effect[PID_GAIN], magnitude); pidff->set_effect[PID_START_DELAY].value[0] = 0; - hid_submit_report(pidff->hid, pidff->reports[PID_SET_EFFECT], + usbhid_submit_report(pidff->hid, pidff->reports[PID_SET_EFFECT], USB_DIR_OUT); } @@ -1165,19 +1165,19 @@ static void pidff_reset(struct pidff_device *pidff) pidff->device_control->value[0] = pidff->control_id[PID_RESET]; /* We reset twice as sometimes hid_wait_io isn't waiting long enough */ - hid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); - hid_wait_io(hid); - hid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); - hid_wait_io(hid); + usbhid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); + usbhid_wait_io(hid); + usbhid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); + usbhid_wait_io(hid); pidff->device_control->value[0] = pidff->control_id[PID_ENABLE_ACTUATORS]; - hid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); - hid_wait_io(hid); + usbhid_submit_report(hid, pidff->reports[PID_DEVICE_CONTROL], USB_DIR_OUT); + usbhid_wait_io(hid); /* pool report is sometimes messed up, refetch it */ - hid_submit_report(hid, pidff->reports[PID_POOL], USB_DIR_IN); - hid_wait_io(hid); + usbhid_submit_report(hid, pidff->reports[PID_POOL], USB_DIR_IN); + usbhid_wait_io(hid); if (pidff->pool[PID_SIMULTANEOUS_MAX].value) { int sim_effects = pidff->pool[PID_SIMULTANEOUS_MAX].value[0]; @@ -1189,9 +1189,9 @@ static void pidff_reset(struct pidff_device *pidff) break; } debug("pid_pool requested again"); - hid_submit_report(hid, pidff->reports[PID_POOL], + usbhid_submit_report(hid, pidff->reports[PID_POOL], USB_DIR_IN); - hid_wait_io(hid); + usbhid_wait_io(hid); } } } @@ -1277,7 +1277,7 @@ int hid_pidff_init(struct hid_device *hid) if (test_bit(FF_GAIN, dev->ffbit)) { pidff_set(&pidff->device_gain[PID_DEVICE_GAIN_FIELD], 0xffff); - hid_submit_report(pidff->hid, pidff->reports[PID_DEVICE_GAIN], + usbhid_submit_report(pidff->hid, pidff->reports[PID_DEVICE_GAIN], USB_DIR_OUT); } diff --git a/drivers/usb/input/hid-tmff.c b/drivers/usb/input/hid-tmff.c index 1cd1418ad6ac..ab67331620d0 100644 --- a/drivers/usb/input/hid-tmff.c +++ b/drivers/usb/input/hid-tmff.c @@ -33,6 +33,7 @@ #include #include +#include "usbhid.h" /* Usages for thrustmaster devices I know about */ #define THRUSTMASTER_USAGE_RUMBLE_LR (HID_UP_GENDESK | 0xbb) @@ -70,7 +71,7 @@ static int hid_tmff_play(struct input_dev *dev, void *data, struct ff_effect *ef tmff->rumble->value[0] = left; tmff->rumble->value[1] = right; dbg("(left,right)=(%08x, %08x)", left, right); - hid_submit_report(hid, tmff->report, USB_DIR_OUT); + usbhid_submit_report(hid, tmff->report, USB_DIR_OUT); return 0; } diff --git a/drivers/usb/input/hid-zpff.c b/drivers/usb/input/hid-zpff.c index af1bfae39dce..7bd8238ca212 100644 --- a/drivers/usb/input/hid-zpff.c +++ b/drivers/usb/input/hid-zpff.c @@ -28,6 +28,7 @@ #include #include #include +#include "usbhid.h" struct zpff_device { struct hid_report *report; @@ -56,7 +57,7 @@ static int hid_zpff_play(struct input_dev *dev, void *data, zpff->report->field[2]->value[0] = left; zpff->report->field[3]->value[0] = right; debug("running with 0x%02x 0x%02x", left, right); - hid_submit_report(hid, zpff->report, USB_DIR_OUT); + usbhid_submit_report(hid, zpff->report, USB_DIR_OUT); return 0; } @@ -101,7 +102,7 @@ int hid_zpff_init(struct hid_device *hid) zpff->report->field[1]->value[0] = 0x02; zpff->report->field[2]->value[0] = 0x00; zpff->report->field[3]->value[0] = 0x00; - hid_submit_report(hid, zpff->report, USB_DIR_OUT); + usbhid_submit_report(hid, zpff->report, USB_DIR_OUT); printk(KERN_INFO "Force feedback for Zeroplus based devices by " "Anssi Hannula \n"); diff --git a/drivers/usb/input/hiddev.c b/drivers/usb/input/hiddev.c index 07d7996575cd..0c2647eb9eec 100644 --- a/drivers/usb/input/hiddev.c +++ b/drivers/usb/input/hiddev.c @@ -240,7 +240,7 @@ static int hiddev_release(struct inode * inode, struct file * file) if (!--list->hiddev->open) { if (list->hiddev->exist) - hid_close(list->hiddev->hid); + usbhid_close(list->hiddev->hid); else kfree(list->hiddev); } @@ -271,7 +271,7 @@ static int hiddev_open(struct inode *inode, struct file *file) if (!list->hiddev->open++) if (list->hiddev->exist) - hid_open(hiddev_table[i]->hid); + usbhid_open(hiddev_table[i]->hid); return 0; } @@ -383,7 +383,7 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd struct hiddev_list *list = file->private_data; struct hiddev *hiddev = list->hiddev; struct hid_device *hid = hiddev->hid; - struct usb_device *dev = hid->dev; + struct usb_device *dev = to_usb_device(hid->dev); struct hiddev_collection_info cinfo; struct hiddev_report_info rinfo; struct hiddev_field_info finfo; @@ -392,6 +392,7 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd struct hiddev_devinfo dinfo; struct hid_report *report; struct hid_field *field; + struct usbhid_device *usbhid = hid->driver_data; void __user *user_arg = (void __user *)arg; int i; @@ -421,7 +422,7 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd dinfo.bustype = BUS_USB; dinfo.busnum = dev->bus->busnum; dinfo.devnum = dev->devnum; - dinfo.ifnum = hid->ifnum; + dinfo.ifnum = usbhid->ifnum; dinfo.vendor = le16_to_cpu(dev->descriptor.idVendor); dinfo.product = le16_to_cpu(dev->descriptor.idProduct); dinfo.version = le16_to_cpu(dev->descriptor.bcdDevice); @@ -480,7 +481,7 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd } case HIDIOCINITREPORT: - hid_init_reports(hid); + usbhid_init_reports(hid); return 0; @@ -494,8 +495,8 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd if ((report = hiddev_lookup_report(hid, &rinfo)) == NULL) return -EINVAL; - hid_submit_report(hid, report, USB_DIR_IN); - hid_wait_io(hid); + usbhid_submit_report(hid, report, USB_DIR_IN); + usbhid_wait_io(hid); return 0; @@ -509,8 +510,8 @@ static int hiddev_ioctl(struct inode *inode, struct file *file, unsigned int cmd if ((report = hiddev_lookup_report(hid, &rinfo)) == NULL) return -EINVAL; - hid_submit_report(hid, report, USB_DIR_OUT); - hid_wait_io(hid); + usbhid_submit_report(hid, report, USB_DIR_OUT); + usbhid_wait_io(hid); return 0; @@ -746,6 +747,7 @@ static struct usb_class_driver hiddev_class = { int hiddev_connect(struct hid_device *hid) { struct hiddev *hiddev; + struct usbhid_device *usbhid = hid->driver_data; int i; int retval; @@ -761,7 +763,7 @@ int hiddev_connect(struct hid_device *hid) if (!(hiddev = kzalloc(sizeof(struct hiddev), GFP_KERNEL))) return -1; - retval = usb_register_dev(hid->intf, &hiddev_class); + retval = usb_register_dev(usbhid->intf, &hiddev_class); if (retval) { err("Not able to get a minor for this device."); kfree(hiddev); @@ -773,10 +775,10 @@ int hiddev_connect(struct hid_device *hid) hiddev->hid = hid; hiddev->exist = 1; - hid->minor = hid->intf->minor; + hid->minor = usbhid->intf->minor; hid->hiddev = hiddev; - hiddev_table[hid->intf->minor - HIDDEV_MINOR_BASE] = hiddev; + hiddev_table[usbhid->intf->minor - HIDDEV_MINOR_BASE] = hiddev; return 0; } @@ -789,14 +791,15 @@ static struct usb_class_driver hiddev_class; void hiddev_disconnect(struct hid_device *hid) { struct hiddev *hiddev = hid->hiddev; + struct usbhid_device *usbhid = hid->driver_data; hiddev->exist = 0; hiddev_table[hiddev->hid->minor - HIDDEV_MINOR_BASE] = NULL; - usb_deregister_dev(hiddev->hid->intf, &hiddev_class); + usb_deregister_dev(usbhid->intf, &hiddev_class); if (hiddev->open) { - hid_close(hiddev->hid); + usbhid_close(hiddev->hid); wake_up_interruptible(&hiddev->wait); } else { kfree(hiddev); diff --git a/drivers/usb/input/usbhid.h b/drivers/usb/input/usbhid.h new file mode 100644 index 000000000000..830107e5251f --- /dev/null +++ b/drivers/usb/input/usbhid.h @@ -0,0 +1,84 @@ +#ifndef __USBHID_H +#define __USBHID_H + +/* + * Copyright (c) 1999 Andreas Gal + * Copyright (c) 2000-2001 Vojtech Pavlik + * Copyright (c) 2006 Jiri Kosina + */ + +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include +#include +#include +#include +#include +#include + +/* API provided by hid-core.c for USB HID drivers */ +int usbhid_wait_io(struct hid_device* hid); +void usbhid_close(struct hid_device *hid); +int usbhid_open(struct hid_device *hid); +void usbhid_init_reports(struct hid_device *hid); +void usbhid_submit_report(struct hid_device *hid, struct hid_report *report, unsigned char dir); + +/* + * USB-specific HID struct, to be pointed to + * from struct hid_device->driver_data + */ + +struct usbhid_device { + struct hid_device *hid; /* pointer to corresponding HID dev */ + + struct usb_interface *intf; /* USB interface */ + int ifnum; /* USB interface number */ + + unsigned int bufsize; /* URB buffer size */ + + struct urb *urbin; /* Input URB */ + char *inbuf; /* Input buffer */ + dma_addr_t inbuf_dma; /* Input buffer dma */ + spinlock_t inlock; /* Input fifo spinlock */ + + struct urb *urbctrl; /* Control URB */ + struct usb_ctrlrequest *cr; /* Control request struct */ + dma_addr_t cr_dma; /* Control request struct dma */ + struct hid_control_fifo ctrl[HID_CONTROL_FIFO_SIZE]; /* Control fifo */ + unsigned char ctrlhead, ctrltail; /* Control fifo head & tail */ + char *ctrlbuf; /* Control buffer */ + dma_addr_t ctrlbuf_dma; /* Control buffer dma */ + spinlock_t ctrllock; /* Control fifo spinlock */ + + struct urb *urbout; /* Output URB */ + struct hid_report *out[HID_CONTROL_FIFO_SIZE]; /* Output pipe fifo */ + unsigned char outhead, outtail; /* Output pipe fifo head & tail */ + char *outbuf; /* Output buffer */ + dma_addr_t outbuf_dma; /* Output buffer dma */ + spinlock_t outlock; /* Output fifo spinlock */ + + unsigned long iofl; /* I/O flags (CTRL_RUNNING, OUT_RUNNING) */ + struct timer_list io_retry; /* Retry timer */ + unsigned long stop_retry; /* Time to give up, in jiffies */ + unsigned int retry_delay; /* Delay length in ms */ + struct work_struct reset_work; /* Task context for resets */ + +}; + +#endif + diff --git a/include/linux/hid.h b/include/linux/hid.h index 0473b45b73b8..fc6f74228da3 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -410,41 +410,7 @@ struct hid_device { /* device report descriptor */ unsigned country; /* HID country */ struct hid_report_enum report_enum[HID_REPORT_TYPES]; - struct usb_device *dev; /* device */ - - /* USB specific fields */ - - struct usb_interface *intf; /* USB interface */ - int ifnum; /* USB interface number */ - - unsigned long iofl; /* I/O flags (CTRL_RUNNING, OUT_RUNNING) */ - struct timer_list io_retry; /* Retry timer */ - unsigned long stop_retry; /* Time to give up, in jiffies */ - unsigned int retry_delay; /* Delay length in ms */ - struct work_struct reset_work; /* Task context for resets */ - - unsigned int bufsize; /* URB buffer size */ - - struct urb *urbin; /* Input URB */ - char *inbuf; /* Input buffer */ - dma_addr_t inbuf_dma; /* Input buffer dma */ - spinlock_t inlock; /* Input fifo spinlock */ - - struct urb *urbctrl; /* Control URB */ - struct usb_ctrlrequest *cr; /* Control request struct */ - dma_addr_t cr_dma; /* Control request struct dma */ - struct hid_control_fifo ctrl[HID_CONTROL_FIFO_SIZE]; /* Control fifo */ - unsigned char ctrlhead, ctrltail; /* Control fifo head & tail */ - char *ctrlbuf; /* Control buffer */ - dma_addr_t ctrlbuf_dma; /* Control buffer dma */ - spinlock_t ctrllock; /* Control fifo spinlock */ - - struct urb *urbout; /* Output URB */ - struct hid_report *out[HID_CONTROL_FIFO_SIZE]; /* Output pipe fifo */ - unsigned char outhead, outtail; /* Output pipe fifo head & tail */ - char *outbuf; /* Output buffer */ - dma_addr_t outbuf_dma; /* Output buffer dma */ - spinlock_t outlock; /* Output fifo spinlock */ + struct device *dev; /* device */ unsigned claimed; /* Claimed by hidinput, hiddev? */ unsigned quirks; /* Various quirks the device can pull on us */ -- cgit v1.2.3 From aa938f7974b82cfd9ee955031987344f332b7c77 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:41:10 +0100 Subject: [PATCH] Generic HID layer - hiddev - hiddev is USB-only (agreed with Marcel Holtmann that Bluetooth currently doesn't need it, and future planned interface (rawhid) will be more flexible and usable) - both HID and USB-hid can be now compiled as modules (wasn't possible before hiddev was fully separated from generic HID layer) Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-core.c | 6 ++++-- drivers/usb/input/hid-core.c | 3 +++ include/linux/hid.h | 8 ++++---- 3 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 8474a7923322..1dd9e4f0df1e 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -819,8 +819,8 @@ static void hid_process_event(struct hid_device *hid, struct hid_field *field, s hid_dump_input(usage, value); if (hid->claimed & HID_CLAIMED_INPUT) hidinput_hid_event(hid, field, usage, value); - if (hid->claimed & HID_CLAIMED_HIDDEV && interrupt) - hiddev_hid_event(hid, field, usage, value); + if (hid->claimed & HID_CLAIMED_HIDDEV && interrupt && hid->hiddev_hid_event) + hid->hiddev_hid_event(hid, field, usage, value); } /* @@ -940,3 +940,5 @@ int hid_set_field(struct hid_field *field, unsigned offset, __s32 value) } EXPORT_SYMBOL_GPL(hid_set_field); +MODULE_LICENSE(DRIVER_LICENSE); + diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 0991c4b751a6..4fc828041d0f 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1284,6 +1284,9 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) hid->hidinput_input_event = usb_hidinput_input_event; hid->hidinput_open = hidinput_open; hid->hidinput_close = hidinput_close; +#ifdef CONFIG_USB_HIDDEV + hid->hiddev_hid_event = hiddev_hid_event; +#endif return hid; diff --git a/include/linux/hid.h b/include/linux/hid.h index fc6f74228da3..5a969a137b85 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -433,6 +433,10 @@ struct hid_device { /* device report descriptor */ int (*hidinput_open) (struct input_dev *); void (*hidinput_close) (struct input_dev *); + /* hiddev event handler */ + void (*hiddev_hid_event) (struct hid_device *, struct hid_field *field, + struct hid_usage *, __s32); + #ifdef CONFIG_USB_HIDINPUT_POWERBOOK unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; @@ -477,13 +481,9 @@ struct hid_descriptor { #define resolv_event(a,b) do { } while (0) #endif -#ifdef CONFIG_HID /* Applications from HID Usage Tables 4/8/99 Version 1.1 */ /* We ignore a few input applications that are not widely used */ #define IS_INPUT_APPLICATION(a) (((a >= 0x00010000) && (a <= 0x00010008)) || (a == 0x00010080) || (a == 0x000c0001)) -#else -#define IS_INPUT_APPLICATION(a) (0) -#endif /* HID core API */ extern void hidinput_hid_event(struct hid_device *, struct hid_field *, struct hid_usage *, __s32); -- cgit v1.2.3 From aa8de2f038baec993f07ef66fb3e94481d1ec22b Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:41:17 +0100 Subject: [PATCH] Generic HID layer - input and event reporting hid_input_report() was needlessly USB-specific in USB HID. This patch makes the function independent of HID implementation and fixes all the current users. Bluetooth patches comply with this prototype. Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-core.c | 59 ++++++++++++++++++++++++++++++++++++++ drivers/usb/input/hid-core.c | 67 ++++---------------------------------------- drivers/usb/input/hiddev.c | 1 + include/linux/hid.h | 3 +- 4 files changed, 68 insertions(+), 62 deletions(-) (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index 1dd9e4f0df1e..18c2b3cf6bcc 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -940,5 +940,64 @@ int hid_set_field(struct hid_field *field, unsigned offset, __s32 value) } EXPORT_SYMBOL_GPL(hid_set_field); +int hid_input_report(struct hid_device *hid, int type, u8 *data, int size, int interrupt) +{ + struct hid_report_enum *report_enum = hid->report_enum + type; + struct hid_report *report; + int n, rsize; + + if (!hid) + return -ENODEV; + + if (!size) { + dbg("empty report"); + return -1; + } + +#ifdef DEBUG_DATA + printk(KERN_DEBUG __FILE__ ": report (size %u) (%snumbered)\n", len, report_enum->numbered ? "" : "un"); +#endif + + n = 0; /* Normally report number is 0 */ + if (report_enum->numbered) { /* Device uses numbered reports, data[0] is report number */ + n = *data++; + size--; + } + +#ifdef DEBUG_DATA + { + int i; + printk(KERN_DEBUG __FILE__ ": report %d (size %u) = ", n, size); + for (i = 0; i < size; i++) + printk(" %02x", data[i]); + printk("\n"); + } +#endif + + if (!(report = report_enum->report_id_hash[n])) { + dbg("undefined report_id %d received", n); + return -1; + } + + rsize = ((report->size - 1) >> 3) + 1; + + if (size < rsize) { + dbg("report %d is too short, (%d < %d)", report->id, size, rsize); + return -1; + } + + if ((hid->claimed & HID_CLAIMED_HIDDEV) && hid->hiddev_report_event) + hid->hiddev_report_event(hid, report); + + for (n = 0; n < report->maxfield; n++) + hid_input_field(hid, report->field[n], data, interrupt); + + if (hid->claimed & HID_CLAIMED_INPUT) + hidinput_report_event(hid, report); + + return 0; +} +EXPORT_SYMBOL_GPL(hid_input_report); + MODULE_LICENSE(DRIVER_LICENSE); diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 4fc828041d0f..a20ff61624b1 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -169,65 +169,6 @@ done: spin_unlock_irqrestore(&usbhid->inlock, flags); } - -static int hid_input_report(int type, struct urb *urb, int interrupt) -{ - struct hid_device *hid = urb->context; - struct hid_report_enum *report_enum = hid->report_enum + type; - u8 *data = urb->transfer_buffer; - int len = urb->actual_length; - struct hid_report *report; - int n, size; - - if (!len) { - dbg("empty report"); - return -1; - } - -#ifdef DEBUG_DATA - printk(KERN_DEBUG __FILE__ ": report (size %u) (%snumbered)\n", len, report_enum->numbered ? "" : "un"); -#endif - - n = 0; /* Normally report number is 0 */ - if (report_enum->numbered) { /* Device uses numbered reports, data[0] is report number */ - n = *data++; - len--; - } - -#ifdef DEBUG_DATA - { - int i; - printk(KERN_DEBUG __FILE__ ": report %d (size %u) = ", n, len); - for (i = 0; i < len; i++) - printk(" %02x", data[i]); - printk("\n"); - } -#endif - - if (!(report = report_enum->report_id_hash[n])) { - dbg("undefined report_id %d received", n); - return -1; - } - - size = ((report->size - 1) >> 3) + 1; - - if (len < size) { - dbg("report %d is too short, (%d < %d)", report->id, len, size); - memset(data + len, 0, size - len); - } - - if (hid->claimed & HID_CLAIMED_HIDDEV) - hiddev_report_event(hid, report); - - for (n = 0; n < report->maxfield; n++) - hid_input_field(hid, report->field[n], data, interrupt); - - if (hid->claimed & HID_CLAIMED_INPUT) - hidinput_report_event(hid, report); - - return 0; -} - /* * Input interrupt completion handler. */ @@ -241,7 +182,9 @@ static void hid_irq_in(struct urb *urb) switch (urb->status) { case 0: /* success */ usbhid->retry_delay = 0; - hid_input_report(HID_INPUT_REPORT, urb, 1); + hid_input_report(urb->context, HID_INPUT_REPORT, + urb->transfer_buffer, + urb->actual_length, 1); break; case -EPIPE: /* stall */ clear_bit(HID_IN_RUNNING, &usbhid->iofl); @@ -426,7 +369,8 @@ static void hid_ctrl(struct urb *urb) switch (urb->status) { case 0: /* success */ if (usbhid->ctrl[usbhid->ctrltail].dir == USB_DIR_IN) - hid_input_report(usbhid->ctrl[usbhid->ctrltail].report->type, urb, 0); + hid_input_report(urb->context, usbhid->ctrl[usbhid->ctrltail].report->type, + urb->transfer_buffer, urb->actual_length, 0); break; case -ESHUTDOWN: /* unplug */ unplug = 1; @@ -1286,6 +1230,7 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) hid->hidinput_close = hidinput_close; #ifdef CONFIG_USB_HIDDEV hid->hiddev_hid_event = hiddev_hid_event; + hid->hiddev_report_event = hiddev_report_event; #endif return hid; diff --git a/drivers/usb/input/hiddev.c b/drivers/usb/input/hiddev.c index 0c2647eb9eec..114d6c9f64b1 100644 --- a/drivers/usb/input/hiddev.c +++ b/drivers/usb/input/hiddev.c @@ -214,6 +214,7 @@ void hiddev_report_event(struct hid_device *hid, struct hid_report *report) hiddev_send_event(hid, &uref); } + /* * fasync file op */ diff --git a/include/linux/hid.h b/include/linux/hid.h index 5a969a137b85..342b4e639acb 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -436,7 +436,7 @@ struct hid_device { /* device report descriptor */ /* hiddev event handler */ void (*hiddev_hid_event) (struct hid_device *, struct hid_field *field, struct hid_usage *, __s32); - + void (*hiddev_report_event) (struct hid_device *, struct hid_report *); #ifdef CONFIG_USB_HIDINPUT_POWERBOOK unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; @@ -492,6 +492,7 @@ extern int hidinput_connect(struct hid_device *); extern void hidinput_disconnect(struct hid_device *); int hid_set_field(struct hid_field *, unsigned, __s32); +int hid_input_report(struct hid_device *, int type, u8 *, int, int); int hidinput_find_field(struct hid_device *hid, unsigned int type, unsigned int code, struct hid_field **field); void hid_input_field(struct hid_device *hid, struct hid_field *field, __u8 *data, int interrupt); void hid_output_report(struct hid_report *report, __u8 *data); -- cgit v1.2.3 From 4c2ae844b5ef85fd4b571c9c91ac48afa6ef2dfc Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Fri, 8 Dec 2006 18:41:22 +0100 Subject: [PATCH] Generic HID layer - pb_fnmode pb_fnmode parameter has to be passed to usbhid, both for compatibility reasons and also because it logically belongs there. Also removes empty hid-input.c file in drivers/usb/input. Signed-off-by: Jiri Kosina Signed-off-by: Marcel Holtmann Cc: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/hid/hid-input.c | 12 +++--------- drivers/usb/input/hid-core.c | 8 ++++++++ drivers/usb/input/hid-input.c | 30 ------------------------------ include/linux/hid.h | 1 + 4 files changed, 12 insertions(+), 39 deletions(-) delete mode 100644 drivers/usb/input/hid-input.c (limited to 'include/linux') diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index e542ef971c46..14cdf09316ce 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -128,12 +128,6 @@ static struct hidinput_key_translation powerbook_iso_keyboard[] = { { } }; - -static int usbhid_pb_fnmode = 1; -module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); -MODULE_PARM_DESC(pb_fnmode, - "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); - static struct hidinput_key_translation *find_translation(struct hidinput_key_translation *table, u16 from) { struct hidinput_key_translation *trans; @@ -160,7 +154,7 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, return 1; } - if (usbhid_pb_fnmode) { + if (hid->pb_fnmode) { int do_translate; trans = find_translation(powerbook_fn_keys, usage->code); @@ -169,8 +163,8 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, do_translate = 1; else if (trans->flags & POWERBOOK_FLAG_FKEY) do_translate = - (usbhid_pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || - (usbhid_pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); + (hid->pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || + (hid->pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); else do_translate = (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON); diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index a20ff61624b1..89fa6885709b 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -56,6 +56,11 @@ static unsigned int hid_mousepoll_interval; module_param_named(mousepoll, hid_mousepoll_interval, uint, 0644); MODULE_PARM_DESC(mousepoll, "Polling interval of mice"); +static int usbhid_pb_fnmode = 1; +module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); +MODULE_PARM_DESC(pb_fnmode, + "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); + /* * Input submission and I/O error handler. */ @@ -1232,6 +1237,9 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) hid->hiddev_hid_event = hiddev_hid_event; hid->hiddev_report_event = hiddev_report_event; #endif +#ifdef CONFIG_USB_HIDINPUT_POWERBOOK + hid->pb_fnmode = usbhid_pb_fnmode; +#endif return hid; diff --git a/drivers/usb/input/hid-input.c b/drivers/usb/input/hid-input.c deleted file mode 100644 index 8756eb3263de..000000000000 --- a/drivers/usb/input/hid-input.c +++ /dev/null @@ -1,30 +0,0 @@ -/* - * $Id: hid-input.c,v 1.2 2002/04/23 00:59:25 rdamazio Exp $ - * - * Copyright (c) 2000-2001 Vojtech Pavlik - * - * USB HID to Linux Input mapping - * - */ - -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Should you need to contact me, the author, you can do so either by - * e-mail - mail your message to , or by paper mail: - * Vojtech Pavlik, Simunkova 1594, Prague 8, 182 00 Czech Republic - */ - - diff --git a/include/linux/hid.h b/include/linux/hid.h index 342b4e639acb..770120add15a 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -438,6 +438,7 @@ struct hid_device { /* device report descriptor */ struct hid_usage *, __s32); void (*hiddev_report_event) (struct hid_device *, struct hid_report *); #ifdef CONFIG_USB_HIDINPUT_POWERBOOK + unsigned int pb_fnmode; unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; #endif -- cgit v1.2.3 From 3644f0cee77494190452de132e82245107939284 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 7 Dec 2006 15:08:17 -0800 Subject: [NET]: Convert hh_lock to seqlock. The hard header cache is in the main output path, so using seqlock instead of reader/writer lock should reduce overhead. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- include/net/neighbour.h | 18 ++++++++++++++++++ net/core/neighbour.c | 11 ++++++----- net/ipv4/ip_output.c | 14 +++----------- net/ipv6/ip6_output.c | 17 ++++------------- 5 files changed, 32 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c57088f575a3..631cec4ff5e1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -199,7 +199,7 @@ struct hh_cache */ u16 hh_len; /* length of header */ int (*hh_output)(struct sk_buff *skb); - rwlock_t hh_lock; + seqlock_t hh_lock; /* cached hardware header; allow for machine alignment needs. */ #define HH_DATA_MOD 16 diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 23967031ddb7..3725b93c52f3 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -309,6 +309,24 @@ static inline int neigh_event_send(struct neighbour *neigh, struct sk_buff *skb) return 0; } +static inline int neigh_hh_output(struct hh_cache *hh, struct sk_buff *skb) +{ + unsigned seq; + int hh_len; + + do { + int hh_alen; + + seq = read_seqbegin(&hh->hh_lock); + hh_len = hh->hh_len; + hh_alen = HH_DATA_ALIGN(hh_len); + memcpy(skb->data - hh_alen, hh->hh_data, hh_alen); + } while (read_seqretry(&hh->hh_lock, seq)); + + skb_push(skb, hh_len); + return hh->hh_output(skb); +} + static inline struct neighbour * __neigh_lookup(struct neigh_table *tbl, const void *pkey, struct net_device *dev, int creat) { diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 0ab1987b9348..e7300b6b4079 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -577,9 +577,10 @@ void neigh_destroy(struct neighbour *neigh) while ((hh = neigh->hh) != NULL) { neigh->hh = hh->hh_next; hh->hh_next = NULL; - write_lock_bh(&hh->hh_lock); + + write_seqlock_bh(&hh->hh_lock); hh->hh_output = neigh_blackhole; - write_unlock_bh(&hh->hh_lock); + write_sequnlock_bh(&hh->hh_lock); if (atomic_dec_and_test(&hh->hh_refcnt)) kfree(hh); } @@ -897,9 +898,9 @@ static void neigh_update_hhs(struct neighbour *neigh) if (update) { for (hh = neigh->hh; hh; hh = hh->hh_next) { - write_lock_bh(&hh->hh_lock); + write_seqlock_bh(&hh->hh_lock); update(hh, neigh->dev, neigh->ha); - write_unlock_bh(&hh->hh_lock); + write_sequnlock_bh(&hh->hh_lock); } } } @@ -1089,7 +1090,7 @@ static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, break; if (!hh && (hh = kzalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) { - rwlock_init(&hh->hh_lock); + seqlock_init(&hh->hh_lock); hh->hh_type = protocol; atomic_set(&hh->hh_refcnt, 0); hh->hh_next = NULL; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index a35209d517ad..f071f84808fa 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -164,7 +164,6 @@ EXPORT_SYMBOL_GPL(ip_build_and_send_pkt); static inline int ip_finish_output2(struct sk_buff *skb) { struct dst_entry *dst = skb->dst; - struct hh_cache *hh = dst->hh; struct net_device *dev = dst->dev; int hh_len = LL_RESERVED_SPACE(dev); @@ -183,16 +182,9 @@ static inline int ip_finish_output2(struct sk_buff *skb) skb = skb2; } - if (hh) { - int hh_alen; - - read_lock_bh(&hh->hh_lock); - hh_alen = HH_DATA_ALIGN(hh->hh_len); - memcpy(skb->data - hh_alen, hh->hh_data, hh_alen); - read_unlock_bh(&hh->hh_lock); - skb_push(skb, hh->hh_len); - return hh->hh_output(skb); - } else if (dst->neighbour) + if (dst->hh) + return neigh_hh_output(dst->hh, skb); + else if (dst->neighbour) return dst->neighbour->output(skb); if (net_ratelimit()) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index e9212c7ff5cf..7b7bd44fbf47 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -72,20 +72,11 @@ static __inline__ void ipv6_select_ident(struct sk_buff *skb, struct frag_hdr *f static inline int ip6_output_finish(struct sk_buff *skb) { - struct dst_entry *dst = skb->dst; - struct hh_cache *hh = dst->hh; - - if (hh) { - int hh_alen; - - read_lock_bh(&hh->hh_lock); - hh_alen = HH_DATA_ALIGN(hh->hh_len); - memcpy(skb->data - hh_alen, hh->hh_data, hh_alen); - read_unlock_bh(&hh->hh_lock); - skb_push(skb, hh->hh_len); - return hh->hh_output(skb); - } else if (dst->neighbour) + + if (dst->hh) + return neigh_hh_output(dst->hh, skb); + else if (dst->neighbour) return dst->neighbour->output(skb); IP6_INC_STATS_BH(ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); -- cgit v1.2.3 From e07bca84cd9d31f76ed655d51e68b6a0ca15f162 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 7 Dec 2006 23:49:45 -0800 Subject: [NETLINK]: Restore API compatibility of address and neighbour bits Restore API compatibility due to bits moved from rtnetlink.h to separate headers. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 493297acdae8..4a629ea70cc4 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -3,6 +3,8 @@ #include #include +#include +#include /**** * Routing/neighbour discovery messages. -- cgit v1.2.3 From f0490980a152958d25ce9762bfb296d8fd4c5512 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 8 Dec 2006 00:08:43 -0800 Subject: [NET]: Force a cache line split in hh_cache in SMP. hh_lock was converted from rwlock to seqlock by Stephen. To have a 100% benefit of this change, I suggest to place read mostly fields of hh_cache in a separate cache line, because hh_refcnt may be changed quite frequently on some busy machines. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 631cec4ff5e1..6be767c76b37 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -193,7 +193,14 @@ struct hh_cache { struct hh_cache *hh_next; /* Next entry */ atomic_t hh_refcnt; /* number of users */ - __be16 hh_type; /* protocol identifier, f.e ETH_P_IP +/* + * We want hh_output, hh_len, hh_lock and hh_data be a in a separate + * cache line on SMP. + * They are mostly read, but hh_refcnt may be changed quite frequently, + * incurring cache line ping pongs. + */ + __be16 hh_type ____cacheline_aligned_in_smp; + /* protocol identifier, f.e ETH_P_IP * NOTE: For VLANs, this will be the * encapuslated type. --BLG */ -- cgit v1.2.3 From 93366c537b3426261cac4db27acc10a99cd91b06 Mon Sep 17 00:00:00 2001 From: J Hadi Salim Date: Fri, 8 Dec 2006 00:12:15 -0800 Subject: [XFRM]: Fix XFRMGRP_REPORT to use correct multicast group. XFRMGRP_REPORT uses 0x10 which is a group that belongs to events. The correct value is 0x20. We should really be using xfrm_nlgroups going forward; it was tempting to delete the definition of XFRMGRP_REPORT but it would break at least iproute2. Signed-off-by: J Hadi Salim Signed-off-by: David S. Miller --- include/linux/xfrm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 088ba8113f7e..9529ea1ae392 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -357,7 +357,7 @@ struct xfrm_user_report { #define XFRMGRP_EXPIRE 2 #define XFRMGRP_SA 4 #define XFRMGRP_POLICY 8 -#define XFRMGRP_REPORT 0x10 +#define XFRMGRP_REPORT 0x20 #endif enum xfrm_nlgroups { -- cgit v1.2.3 From d3dcc077bf88806201093f86325ec656e4dbfbce Mon Sep 17 00:00:00 2001 From: David S. Miller Date: Fri, 8 Dec 2006 17:05:13 -0800 Subject: [NETLINK]: Put {IFA,IFLA}_{RTA,PAYLOAD} macros back for userspace. GLIBC uses them etc. They are guarded by ifndef __KERNEL__ so nobody will start accidently using them in the kernel again, it's just for userspace. Signed-off-by: David S. Miller --- include/linux/if_addr.h | 6 ++++++ include/linux/if_link.h | 6 ++++++ 2 files changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/if_addr.h b/include/linux/if_addr.h index dbe8f6120a40..d557e4ce9b6b 100644 --- a/include/linux/if_addr.h +++ b/include/linux/if_addr.h @@ -52,4 +52,10 @@ struct ifa_cacheinfo __u32 tstamp; /* updated timestamp, hundredths of seconds */ }; +/* backwards compatibility for userspace */ +#ifndef __KERNEL__ +#define IFA_RTA(r) ((struct rtattr*)(((char*)(r)) + NLMSG_ALIGN(sizeof(struct ifaddrmsg)))) +#define IFA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ifaddrmsg)) +#endif + #endif diff --git a/include/linux/if_link.h b/include/linux/if_link.h index e963a077e6f5..35ed3b5467f3 100644 --- a/include/linux/if_link.h +++ b/include/linux/if_link.h @@ -82,6 +82,12 @@ enum #define IFLA_MAX (__IFLA_MAX - 1) +/* backwards compatibility for userspace */ +#ifndef __KERNEL__ +#define IFLA_RTA(r) ((struct rtattr*)(((char*)(r)) + NLMSG_ALIGN(sizeof(struct ifinfomsg)))) +#define IFLA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ifinfomsg)) +#endif + /* ifi_flags. IFF_* flags. -- cgit v1.2.3 From 9c4dfadbde3cfa78b92c28597125b8c41d36ffd0 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Sat, 4 Nov 2006 09:22:27 -0300 Subject: V4L/DVB (4796): A couple of V4L2 defines needed by Cafe Camara driver Two defines for V4L2, needed by the Cafe camera driver: 1) Add the RGB444 image format 2) Add the "init" internal command which is separate from "reset". Signed-off-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab --- include/linux/videodev2.h | 1 + include/media/v4l2-common.h | 4 ++++ 2 files changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h index df5c4654360d..5cb380a559fd 100644 --- a/include/linux/videodev2.h +++ b/include/linux/videodev2.h @@ -244,6 +244,7 @@ struct v4l2_pix_format #define V4L2_PIX_FMT_YYUV v4l2_fourcc('Y','Y','U','V') /* 16 YUV 4:2:2 */ #define V4L2_PIX_FMT_HI240 v4l2_fourcc('H','I','2','4') /* 8 8-bit color */ #define V4L2_PIX_FMT_HM12 v4l2_fourcc('H','M','1','2') /* 8 YUV 4:2:0 16x16 macroblocks */ +#define V4L2_PIX_FMT_RGB444 v4l2_fourcc('R','4','4','4') /* 16 xxxxrrrr ggggbbbb */ /* see http://www.siliconimaging.com/RGB%20Bayer.htm */ #define V4L2_PIX_FMT_SBGGR8 v4l2_fourcc('B','A','8','1') /* 8 BGBG.. GRGR.. */ diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h index aecc946980a3..8263ea00ca6f 100644 --- a/include/media/v4l2-common.h +++ b/include/media/v4l2-common.h @@ -251,4 +251,8 @@ struct v4l2_crystal_freq { If the frequency is not supported, then -EINVAL is returned. */ #define VIDIOC_INT_S_CRYSTAL_FREQ _IOW ('d', 113, struct v4l2_crystal_freq) +/* Initialize the sensor registors to some sort of reasonable + default values. */ +#define VIDIOC_INT_INIT _IOW ('d', 114, u32) + #endif /* V4L2_COMMON_H_ */ -- cgit v1.2.3 From d905b382d797a213e15868cbf3204f50ed52e30b Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Sat, 4 Nov 2006 09:25:53 -0300 Subject: V4L/DVB (4797): Marvell 88ALP01 "cafe" driver A driver for the Marvell M88ALP01 "CAFE" CMOS integrated camera controller. This driver has been renamed "cafe_ccic" since my previous patch set. Signed-off-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab --- Documentation/video4linux/cafe_ccic | 54 + drivers/media/video/Kconfig | 9 + drivers/media/video/Makefile | 2 + drivers/media/video/cafe_ccic-regs.h | 160 +++ drivers/media/video/cafe_ccic.c | 2237 ++++++++++++++++++++++++++++++++++ include/linux/i2c-id.h | 1 + 6 files changed, 2463 insertions(+) create mode 100644 Documentation/video4linux/cafe_ccic create mode 100644 drivers/media/video/cafe_ccic-regs.h create mode 100644 drivers/media/video/cafe_ccic.c (limited to 'include/linux') diff --git a/Documentation/video4linux/cafe_ccic b/Documentation/video4linux/cafe_ccic new file mode 100644 index 000000000000..88821022a5de --- /dev/null +++ b/Documentation/video4linux/cafe_ccic @@ -0,0 +1,54 @@ +"cafe_ccic" is a driver for the Marvell 88ALP01 "cafe" CMOS camera +controller. This is the controller found in first-generation OLPC systems, +and this driver was written with support from the OLPC project. + +Current status: the core driver works. It can generate data in YUV422, +RGB565, and RGB444 formats. (Anybody looking at the code will see RGB32 as +well, but that is a debugging aid which will be removed shortly). VGA and +QVGA modes work; CIF is there but the colors remain funky. Only the OV7670 +sensor is known to work with this controller at this time. + +To try it out: either of these commands will work: + + mplayer tv:// -tv driver=v4l2:width=640:height=480 -nosound + mplayer tv:// -tv driver=v4l2:width=640:height=480:outfmt=bgr16 -nosound + +The "xawtv" utility also works; gqcam does not, for unknown reasons. + +There are a few load-time options, most of which can be changed after +loading via sysfs as well: + + - alloc_bufs_at_load: Normally, the driver will not allocate any DMA + buffers until the time comes to transfer data. If this option is set, + then worst-case-sized buffers will be allocated at module load time. + This option nails down the memory for the life of the module, but + perhaps decreases the chances of an allocation failure later on. + + - dma_buf_size: The size of DMA buffers to allocate. Note that this + option is only consulted for load-time allocation; when buffers are + allocated at run time, they will be sized appropriately for the current + camera settings. + + - n_dma_bufs: The controller can cycle through either two or three DMA + buffers. Normally, the driver tries to use three buffers; on faster + systems, however, it will work well with only two. + + - min_buffers: The minimum number of streaming I/O buffers that the driver + will consent to work with. Default is one, but, on slower systems, + better behavior with mplayer can be achieved by setting to a higher + value (like six). + + - max_buffers: The maximum number of streaming I/O buffers; default is + ten. That number was carefully picked out of a hat and should not be + assumed to actually mean much of anything. + + - flip: If this boolean parameter is set, the sensor will be instructed to + invert the video image. Whether it makes sense is determined by how + your particular camera is mounted. + +Work is ongoing with this driver, stay tuned. + +jon + +Jonathan Corbet +corbet@lwn.net diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index b8fde5cf4735..4ea1d0ebf5fd 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -670,6 +670,15 @@ config VIDEO_M32R_AR_M64278 To compile this driver as a module, choose M here: the module will be called arv. +config VIDEO_CAFE_CCIC + tristate "Marvell 88ALP01 (Cafe) CMOS Camera Controller support" + depends on I2C && VIDEO_V4L2 + select VIDEO_OV7670 + ---help--- + This is a video4linux2 driver for the Marvell 88ALP01 integrated + CMOS camera controller. This is the controller found on first- + generation OLPC systems. + # # USB Multimedia device configuration # diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile index af57abce8a6e..8ff787a4cf6a 100644 --- a/drivers/media/video/Makefile +++ b/drivers/media/video/Makefile @@ -92,6 +92,8 @@ obj-$(CONFIG_VIDEO_UPD64031A) += upd64031a.o obj-$(CONFIG_VIDEO_UPD64083) += upd64083.o obj-$(CONFIG_VIDEO_CX2341X) += cx2341x.o +obj-$(CONFIG_VIDEO_CAFE_CCIC) += cafe_ccic.o + obj-$(CONFIG_USB_DABUSB) += dabusb.o obj-$(CONFIG_USB_OV511) += ov511.o obj-$(CONFIG_USB_SE401) += se401.o diff --git a/drivers/media/video/cafe_ccic-regs.h b/drivers/media/video/cafe_ccic-regs.h new file mode 100644 index 000000000000..b2c22a0d6643 --- /dev/null +++ b/drivers/media/video/cafe_ccic-regs.h @@ -0,0 +1,160 @@ +/* + * Register definitions for the m88alp01 camera interface. Offsets in bytes + * as given in the spec. + * + * Copyright 2006 One Laptop Per Child Association, Inc. + * + * Written by Jonathan Corbet, corbet@lwn.net. + * + * This file may be distributed under the terms of the GNU General + * Public License, version 2. + */ +#define REG_Y0BAR 0x00 +#define REG_Y1BAR 0x04 +#define REG_Y2BAR 0x08 +/* ... */ + +#define REG_IMGPITCH 0x24 /* Image pitch register */ +#define IMGP_YP_SHFT 2 /* Y pitch params */ +#define IMGP_YP_MASK 0x00003ffc /* Y pitch field */ +#define IMGP_UVP_SHFT 18 /* UV pitch (planar) */ +#define IMGP_UVP_MASK 0x3ffc0000 +#define REG_IRQSTATRAW 0x28 /* RAW IRQ Status */ +#define IRQ_EOF0 0x00000001 /* End of frame 0 */ +#define IRQ_EOF1 0x00000002 /* End of frame 1 */ +#define IRQ_EOF2 0x00000004 /* End of frame 2 */ +#define IRQ_SOF0 0x00000008 /* Start of frame 0 */ +#define IRQ_SOF1 0x00000010 /* Start of frame 1 */ +#define IRQ_SOF2 0x00000020 /* Start of frame 2 */ +#define IRQ_OVERFLOW 0x00000040 /* FIFO overflow */ +#define IRQ_TWSIW 0x00010000 /* TWSI (smbus) write */ +#define IRQ_TWSIR 0x00020000 /* TWSI read */ +#define IRQ_TWSIE 0x00040000 /* TWSI error */ +#define TWSIIRQS (IRQ_TWSIW|IRQ_TWSIR|IRQ_TWSIE) +#define FRAMEIRQS (IRQ_EOF0|IRQ_EOF1|IRQ_EOF2|IRQ_SOF0|IRQ_SOF1|IRQ_SOF2) +#define ALLIRQS (TWSIIRQS|FRAMEIRQS|IRQ_OVERFLOW) +#define REG_IRQMASK 0x2c /* IRQ mask - same bits as IRQSTAT */ +#define REG_IRQSTAT 0x30 /* IRQ status / clear */ + +#define REG_IMGSIZE 0x34 /* Image size */ +#define IMGSZ_V_MASK 0x1fff0000 +#define IMGSZ_V_SHIFT 16 +#define IMGSZ_H_MASK 0x00003fff +#define REG_IMGOFFSET 0x38 /* IMage offset */ + +#define REG_CTRL0 0x3c /* Control 0 */ +#define C0_ENABLE 0x00000001 /* Makes the whole thing go */ + +/* Mask for all the format bits */ +#define C0_DF_MASK 0x00fffffc /* Bits 2-23 */ + +/* RGB ordering */ +#define C0_RGB4_RGBX 0x00000000 +#define C0_RGB4_XRGB 0x00000004 +#define C0_RGB4_BGRX 0x00000008 +#define C0_RGB4_XBGR 0x0000000c +#define C0_RGB5_RGGB 0x00000000 +#define C0_RGB5_GRBG 0x00000004 +#define C0_RGB5_GBRG 0x00000008 +#define C0_RGB5_BGGR 0x0000000c + +/* Spec has two fields for DIN and DOUT, but they must match, so + combine them here. */ +#define C0_DF_YUV 0x00000000 /* Data is YUV */ +#define C0_DF_RGB 0x000000a0 /* ... RGB */ +#define C0_DF_BAYER 0x00000140 /* ... Bayer */ +/* 8-8-8 must be missing from the below - ask */ +#define C0_RGBF_565 0x00000000 +#define C0_RGBF_444 0x00000800 +#define C0_RGB_BGR 0x00001000 /* Blue comes first */ +#define C0_YUV_PLANAR 0x00000000 /* YUV 422 planar format */ +#define C0_YUV_PACKED 0x00008000 /* YUV 422 packed */ +#define C0_YUV_420PL 0x0000a000 /* YUV 420 planar */ +/* Think that 420 packed must be 111 - ask */ +#define C0_YUVE_YUYV 0x00000000 /* Y1CbY0Cr */ +#define C0_YUVE_YVYU 0x00010000 /* Y1CrY0Cb */ +#define C0_YUVE_VYUY 0x00020000 /* CrY1CbY0 */ +#define C0_YUVE_UYVY 0x00030000 /* CbY1CrY0 */ +#define C0_YUVE_XYUV 0x00000000 /* 420: .YUV */ +#define C0_YUVE_XYVU 0x00010000 /* 420: .YVU */ +#define C0_YUVE_XUVY 0x00020000 /* 420: .UVY */ +#define C0_YUVE_XVUY 0x00030000 /* 420: .VUY */ +/* Bayer bits 18,19 if needed */ +#define C0_HPOL_LOW 0x01000000 /* HSYNC polarity active low */ +#define C0_VPOL_LOW 0x02000000 /* VSYNC polarity active low */ +#define C0_VCLK_LOW 0x04000000 /* VCLK on falling edge */ +#define C0_DOWNSCALE 0x08000000 /* Enable downscaler */ +#define C0_SIFM_MASK 0xc0000000 /* SIF mode bits */ +#define C0_SIF_HVSYNC 0x00000000 /* Use H/VSYNC */ +#define CO_SOF_NOSYNC 0x40000000 /* Use inband active signaling */ + + +#define REG_CTRL1 0x40 /* Control 1 */ +#define C1_444ALPHA 0x00f00000 /* Alpha field in RGB444 */ +#define C1_ALPHA_SHFT 20 +#define C1_DMAB32 0x00000000 /* 32-byte DMA burst */ +#define C1_DMAB16 0x02000000 /* 16-byte DMA burst */ +#define C1_DMAB64 0x04000000 /* 64-byte DMA burst */ +#define C1_DMAB_MASK 0x06000000 +#define C1_TWOBUFS 0x08000000 /* Use only two DMA buffers */ +#define C1_PWRDWN 0x10000000 /* Power down */ + +#define REG_CLKCTRL 0x88 /* Clock control */ +#define CLK_DIV_MASK 0x0000ffff /* Upper bits RW "reserved" */ + +#define REG_GPR 0xb4 /* General purpose register. This + controls inputs to the power and reset + pins on the OV7670 used with OLPC; + other deployments could differ. */ +#define GPR_C1EN 0x00000020 /* Pad 1 (power down) enable */ +#define GPR_C0EN 0x00000010 /* Pad 0 (reset) enable */ +#define GPR_C1 0x00000002 /* Control 1 value */ +/* + * Control 0 is wired to reset on OLPC machines. For ov7x sensors, + * it is active low, for 0v6x, instead, it's active high. What + * fun. + */ +#define GPR_C0 0x00000001 /* Control 0 value */ + +#define REG_TWSIC0 0xb8 /* TWSI (smbus) control 0 */ +#define TWSIC0_EN 0x00000001 /* TWSI enable */ +#define TWSIC0_MODE 0x00000002 /* 1 = 16-bit, 0 = 8-bit */ +#define TWSIC0_SID 0x000003fc /* Slave ID */ +#define TWSIC0_SID_SHIFT 2 +#define TWSIC0_CLKDIV 0x0007fc00 /* Clock divider */ +#define TWSIC0_MASKACK 0x00400000 /* Mask ack from sensor */ +#define TWSIC0_OVMAGIC 0x00800000 /* Make it work on OV sensors */ + +#define REG_TWSIC1 0xbc /* TWSI control 1 */ +#define TWSIC1_DATA 0x0000ffff /* Data to/from camchip */ +#define TWSIC1_ADDR 0x00ff0000 /* Address (register) */ +#define TWSIC1_ADDR_SHIFT 16 +#define TWSIC1_READ 0x01000000 /* Set for read op */ +#define TWSIC1_WSTAT 0x02000000 /* Write status */ +#define TWSIC1_RVALID 0x04000000 /* Read data valid */ +#define TWSIC1_ERROR 0x08000000 /* Something screwed up */ + + +#define REG_UBAR 0xc4 /* Upper base address register */ + +/* + * Here's the weird global control registers which are said to live + * way up here. + */ +#define REG_GL_CSR 0x3004 /* Control/status register */ +#define GCSR_SRS 0x00000001 /* SW Reset set */ +#define GCSR_SRC 0x00000002 /* SW Reset clear */ +#define GCSR_MRS 0x00000004 /* Master reset set */ +#define GCSR_MRC 0x00000008 /* HW Reset clear */ +#define GCSR_CCIC_EN 0x00004000 /* CCIC Clock enable */ +#define REG_GL_IMASK 0x300c /* Interrupt mask register */ +#define GIMSK_CCIC_EN 0x00000004 /* CCIC Interrupt enable */ + +#define REG_LEN REG_GL_IMASK + 4 + + +/* + * Useful stuff that probably belongs somewhere global. + */ +#define VGA_WIDTH 640 +#define VGA_HEIGHT 480 diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c new file mode 100644 index 000000000000..2026c4983b26 --- /dev/null +++ b/drivers/media/video/cafe_ccic.c @@ -0,0 +1,2237 @@ +/* + * A driver for the CMOS camera controller in the Marvell 88ALP01 "cafe" + * multifunction chip. Currently works with the Omnivision OV7670 + * sensor. + * + * Copyright 2006 One Laptop Per Child Association, Inc. + * + * Written by Jonathan Corbet, corbet@lwn.net. + * + * This file may be distributed under the terms of the GNU General + * Public License, version 2. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include "cafe_ccic-regs.h" + +#define CAFE_VERSION 0x000001 + + +/* + * Parameters. + */ +MODULE_AUTHOR("Jonathan Corbet "); +MODULE_DESCRIPTION("Marvell 88ALP01 CMOS Camera Controller driver"); +MODULE_LICENSE("GPL"); +MODULE_SUPPORTED_DEVICE("Video"); + +/* + * Internal DMA buffer management. Since the controller cannot do S/G I/O, + * we must have physically contiguous buffers to bring frames into. + * These parameters control how many buffers we use, whether we + * allocate them at load time (better chance of success, but nails down + * memory) or when somebody tries to use the camera (riskier), and, + * for load-time allocation, how big they should be. + * + * The controller can cycle through three buffers. We could use + * more by flipping pointers around, but it probably makes little + * sense. + */ + +#define MAX_DMA_BUFS 3 +static int alloc_bufs_at_load = 0; +module_param(alloc_bufs_at_load, bool, 0444); +MODULE_PARM_DESC(alloc_bufs_at_load, + "Non-zero value causes DMA buffers to be allocated at module " + "load time. This increases the chances of successfully getting " + "those buffers, but at the cost of nailing down the memory from " + "the outset."); + +static int n_dma_bufs = 3; +module_param(n_dma_bufs, uint, 0644); +MODULE_PARM_DESC(n_dma_bufs, + "The number of DMA buffers to allocate. Can be either two " + "(saves memory, makes timing tighter) or three."); + +static int dma_buf_size = VGA_WIDTH * VGA_HEIGHT * 2; /* Worst case */ +module_param(dma_buf_size, uint, 0444); +MODULE_PARM_DESC(dma_buf_size, + "The size of the allocated DMA buffers. If actual operating " + "parameters require larger buffers, an attempt to reallocate " + "will be made."); + +static int min_buffers = 1; +module_param(min_buffers, uint, 0644); +MODULE_PARM_DESC(min_buffers, + "The minimum number of streaming I/O buffers we are willing " + "to work with."); + +static int max_buffers = 10; +module_param(max_buffers, uint, 0644); +MODULE_PARM_DESC(max_buffers, + "The maximum number of streaming I/O buffers an application " + "will be allowed to allocate. These buffers are big and live " + "in vmalloc space."); + +static int flip = 0; +module_param(flip, bool, 0444); +MODULE_PARM_DESC(flip, + "If set, the sensor will be instructed to flip the image " + "vertically."); + + +enum cafe_state { + S_NOTREADY, /* Not yet initialized */ + S_IDLE, /* Just hanging around */ + S_FLAKED, /* Some sort of problem */ + S_SINGLEREAD, /* In read() */ + S_SPECREAD, /* Speculative read (for future read()) */ + S_STREAMING /* Streaming data */ +}; + +/* + * Tracking of streaming I/O buffers. + */ +struct cafe_sio_buffer { + struct list_head list; + struct v4l2_buffer v4lbuf; + char *buffer; /* Where it lives in kernel space */ + int mapcount; + struct cafe_camera *cam; +}; + +/* + * A description of one of our devices. + * Locking: controlled by s_mutex. Certain fields, however, require + * the dev_lock spinlock; they are marked as such by comments. + * dev_lock is also required for access to device registers. + */ +struct cafe_camera +{ + enum cafe_state state; + unsigned long flags; /* Buffer status, mainly (dev_lock) */ + int users; /* How many open FDs */ + struct file *owner; /* Who has data access (v4l2) */ + + /* + * Subsystem structures. + */ + struct pci_dev *pdev; + struct video_device v4ldev; + struct i2c_adapter i2c_adapter; + struct i2c_client *sensor; + + unsigned char __iomem *regs; + struct list_head dev_list; /* link to other devices */ + + /* DMA buffers */ + unsigned int nbufs; /* How many are alloc'd */ + int next_buf; /* Next to consume (dev_lock) */ + unsigned int dma_buf_size; /* allocated size */ + void *dma_bufs[MAX_DMA_BUFS]; /* Internal buffer addresses */ + dma_addr_t dma_handles[MAX_DMA_BUFS]; /* Buffer bus addresses */ + unsigned int specframes; /* Unconsumed spec frames (dev_lock) */ + unsigned int sequence; /* Frame sequence number */ + unsigned int buf_seq[MAX_DMA_BUFS]; /* Sequence for individual buffers */ + + /* Streaming buffers */ + unsigned int n_sbufs; /* How many we have */ + struct cafe_sio_buffer *sb_bufs; /* The array of housekeeping structs */ + struct list_head sb_avail; /* Available for data (we own) (dev_lock) */ + struct list_head sb_full; /* With data (user space owns) (dev_lock) */ + struct tasklet_struct s_tasklet; + + /* Current operating parameters */ + enum v4l2_chip_ident sensor_type; /* Currently ov7670 only */ + struct v4l2_pix_format pix_format; + + /* Locks */ + struct mutex s_mutex; /* Access to this structure */ + spinlock_t dev_lock; /* Access to device */ + + /* Misc */ + wait_queue_head_t smbus_wait; /* Waiting on i2c events */ + wait_queue_head_t iowait; /* Waiting on frame data */ +#ifdef CONFIG_VIDEO_ADV_DEBUG + struct dentry *dfs_regs; + struct dentry *dfs_cam_regs; +#endif +}; + +/* + * Status flags. Always manipulated with bit operations. + */ +#define CF_BUF0_VALID 0 /* Buffers valid - first three */ +#define CF_BUF1_VALID 1 +#define CF_BUF2_VALID 2 +#define CF_DMA_ACTIVE 3 /* A frame is incoming */ +#define CF_CONFIG_NEEDED 4 /* Must configure hardware */ + + + +/* + * Start over with DMA buffers - dev_lock needed. + */ +static void cafe_reset_buffers(struct cafe_camera *cam) +{ + int i; + + cam->next_buf = -1; + for (i = 0; i < cam->nbufs; i++) + clear_bit(i, &cam->flags); + cam->specframes = 0; +} + +static inline int cafe_needs_config(struct cafe_camera *cam) +{ + return test_bit(CF_CONFIG_NEEDED, &cam->flags); +} + +static void cafe_set_config_needed(struct cafe_camera *cam, int needed) +{ + if (needed) + set_bit(CF_CONFIG_NEEDED, &cam->flags); + else + clear_bit(CF_CONFIG_NEEDED, &cam->flags); +} + + + + +/* + * Debugging and related. + */ +#define cam_err(cam, fmt, arg...) \ + dev_err(&(cam)->pdev->dev, fmt, ##arg); +#define cam_warn(cam, fmt, arg...) \ + dev_warn(&(cam)->pdev->dev, fmt, ##arg); +#define cam_dbg(cam, fmt, arg...) \ + dev_dbg(&(cam)->pdev->dev, fmt, ##arg); + + +/* ---------------------------------------------------------------------*/ +/* + * We keep a simple list of known devices to search at open time. + */ +static LIST_HEAD(cafe_dev_list); +static DEFINE_MUTEX(cafe_dev_list_lock); + +static void cafe_add_dev(struct cafe_camera *cam) +{ + mutex_lock(&cafe_dev_list_lock); + list_add_tail(&cam->dev_list, &cafe_dev_list); + mutex_unlock(&cafe_dev_list_lock); +} + +static void cafe_remove_dev(struct cafe_camera *cam) +{ + mutex_lock(&cafe_dev_list_lock); + list_del(&cam->dev_list); + mutex_unlock(&cafe_dev_list_lock); +} + +static struct cafe_camera *cafe_find_dev(int minor) +{ + struct cafe_camera *cam; + + mutex_lock(&cafe_dev_list_lock); + list_for_each_entry(cam, &cafe_dev_list, dev_list) { + if (cam->v4ldev.minor == minor) + goto done; + } + cam = NULL; + done: + mutex_unlock(&cafe_dev_list_lock); + return cam; +} + + +static struct cafe_camera *cafe_find_by_pdev(struct pci_dev *pdev) +{ + struct cafe_camera *cam; + + mutex_lock(&cafe_dev_list_lock); + list_for_each_entry(cam, &cafe_dev_list, dev_list) { + if (cam->pdev == pdev) + goto done; + } + cam = NULL; + done: + mutex_unlock(&cafe_dev_list_lock); + return cam; +} + + +/* ------------------------------------------------------------------------ */ +/* + * Device register I/O + */ +static inline void cafe_reg_write(struct cafe_camera *cam, unsigned int reg, + unsigned int val) +{ + iowrite32(val, cam->regs + reg); +} + +static inline unsigned int cafe_reg_read(struct cafe_camera *cam, + unsigned int reg) +{ + return ioread32(cam->regs + reg); +} + + +static inline void cafe_reg_write_mask(struct cafe_camera *cam, unsigned int reg, + unsigned int val, unsigned int mask) +{ + unsigned int v = cafe_reg_read(cam, reg); + + v = (v & ~mask) | (val & mask); + cafe_reg_write(cam, reg, v); +} + +static inline void cafe_reg_clear_bit(struct cafe_camera *cam, + unsigned int reg, unsigned int val) +{ + cafe_reg_write_mask(cam, reg, 0, val); +} + +static inline void cafe_reg_set_bit(struct cafe_camera *cam, + unsigned int reg, unsigned int val) +{ + cafe_reg_write_mask(cam, reg, val, val); +} + + + +/* -------------------------------------------------------------------- */ +/* + * The I2C/SMBUS interface to the camera itself starts here. The + * controller handles SMBUS itself, presenting a relatively simple register + * interface; all we have to do is to tell it where to route the data. + */ +#define CAFE_SMBUS_TIMEOUT (HZ) /* generous */ + +static int cafe_smbus_write_done(struct cafe_camera *cam) +{ + unsigned long flags; + int c1; + + /* + * We must delay after the interrupt, or the controller gets confused + * and never does give us good status. Fortunately, we don't do this + * often. + */ + udelay(20); + spin_lock_irqsave(&cam->dev_lock, flags); + c1 = cafe_reg_read(cam, REG_TWSIC1); + spin_unlock_irqrestore(&cam->dev_lock, flags); + return (c1 & (TWSIC1_WSTAT|TWSIC1_ERROR)) != TWSIC1_WSTAT; +} + +static int cafe_smbus_write_data(struct cafe_camera *cam, + u16 addr, u8 command, u8 value) +{ + unsigned int rval; + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + rval = TWSIC0_EN | ((addr << TWSIC0_SID_SHIFT) & TWSIC0_SID); + rval |= TWSIC0_OVMAGIC; /* Make OV sensors work */ + /* + * Marvell sez set clkdiv to all 1's for now. + */ + rval |= TWSIC0_CLKDIV; + cafe_reg_write(cam, REG_TWSIC0, rval); + (void) cafe_reg_read(cam, REG_TWSIC1); /* force write */ + rval = value | ((command << TWSIC1_ADDR_SHIFT) & TWSIC1_ADDR); + cafe_reg_write(cam, REG_TWSIC1, rval); + spin_unlock_irqrestore(&cam->dev_lock, flags); + msleep(2); /* Required or things flake */ + + wait_event_timeout(cam->smbus_wait, cafe_smbus_write_done(cam), + CAFE_SMBUS_TIMEOUT); + spin_lock_irqsave(&cam->dev_lock, flags); + rval = cafe_reg_read(cam, REG_TWSIC1); + spin_unlock_irqrestore(&cam->dev_lock, flags); + + if (rval & TWSIC1_WSTAT) { + cam_err(cam, "SMBUS write (%02x/%02x/%02x) timed out\n", addr, + command, value); + return -EIO; + } + if (rval & TWSIC1_ERROR) { + cam_err(cam, "SMBUS write (%02x/%02x/%02x) error\n", addr, + command, value); + return -EIO; + } + return 0; +} + + + +static int cafe_smbus_read_done(struct cafe_camera *cam) +{ + unsigned long flags; + int c1; + + /* + * We must delay after the interrupt, or the controller gets confused + * and never does give us good status. Fortunately, we don't do this + * often. + */ + udelay(20); + spin_lock_irqsave(&cam->dev_lock, flags); + c1 = cafe_reg_read(cam, REG_TWSIC1); + spin_unlock_irqrestore(&cam->dev_lock, flags); + return c1 & (TWSIC1_RVALID|TWSIC1_ERROR); +} + + + +static int cafe_smbus_read_data(struct cafe_camera *cam, + u16 addr, u8 command, u8 *value) +{ + unsigned int rval; + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + rval = TWSIC0_EN | ((addr << TWSIC0_SID_SHIFT) & TWSIC0_SID); + rval |= TWSIC0_OVMAGIC; /* Make OV sensors work */ + /* + * Marvel sez set clkdiv to all 1's for now. + */ + rval |= TWSIC0_CLKDIV; + cafe_reg_write(cam, REG_TWSIC0, rval); + (void) cafe_reg_read(cam, REG_TWSIC1); /* force write */ + rval = TWSIC1_READ | ((command << TWSIC1_ADDR_SHIFT) & TWSIC1_ADDR); + cafe_reg_write(cam, REG_TWSIC1, rval); + spin_unlock_irqrestore(&cam->dev_lock, flags); + + wait_event_timeout(cam->smbus_wait, + cafe_smbus_read_done(cam), CAFE_SMBUS_TIMEOUT); + spin_lock_irqsave(&cam->dev_lock, flags); + rval = cafe_reg_read(cam, REG_TWSIC1); + spin_unlock_irqrestore(&cam->dev_lock, flags); + + if (rval & TWSIC1_ERROR) { + cam_err(cam, "SMBUS read (%02x/%02x) error\n", addr, command); + return -EIO; + } + if (! (rval & TWSIC1_RVALID)) { + cam_err(cam, "SMBUS read (%02x/%02x) timed out\n", addr, + command); + return -EIO; + } + *value = rval & 0xff; + return 0; +} + +/* + * Perform a transfer over SMBUS. This thing is called under + * the i2c bus lock, so we shouldn't race with ourselves... + */ +static int cafe_smbus_xfer(struct i2c_adapter *adapter, u16 addr, + unsigned short flags, char rw, u8 command, + int size, union i2c_smbus_data *data) +{ + struct cafe_camera *cam = i2c_get_adapdata(adapter); + int ret = -EINVAL; + + /* + * Refuse to talk to anything but OV cam chips. We should + * never even see an attempt to do so, but one never knows. + */ + if (cam->sensor && addr != cam->sensor->addr) { + cam_err(cam, "funky smbus addr %d\n", addr); + return -EINVAL; + } + /* + * This interface would appear to only do byte data ops. OK + * it can do word too, but the cam chip has no use for that. + */ + if (size != I2C_SMBUS_BYTE_DATA) { + cam_err(cam, "funky xfer size %d\n", size); + return -EINVAL; + } + + if (rw == I2C_SMBUS_WRITE) + ret = cafe_smbus_write_data(cam, addr, command, data->byte); + else if (rw == I2C_SMBUS_READ) + ret = cafe_smbus_read_data(cam, addr, command, &data->byte); + return ret; +} + + +static void cafe_smbus_enable_irq(struct cafe_camera *cam) +{ + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_reg_set_bit(cam, REG_IRQMASK, TWSIIRQS); + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + +static u32 cafe_smbus_func(struct i2c_adapter *adapter) +{ + return I2C_FUNC_SMBUS_READ_BYTE_DATA | + I2C_FUNC_SMBUS_WRITE_BYTE_DATA; +} + +static struct i2c_algorithm cafe_smbus_algo = { + .smbus_xfer = cafe_smbus_xfer, + .functionality = cafe_smbus_func +}; + +/* Somebody is on the bus */ +static int cafe_cam_init(struct cafe_camera *cam); + +static int cafe_smbus_attach(struct i2c_client *client) +{ + struct cafe_camera *cam = i2c_get_adapdata(client->adapter); + + /* + * Don't talk to chips we don't recognize. + */ + cam_err(cam, "smbus_attach id = %d\n", client->driver->id); + if (client->driver->id == I2C_DRIVERID_OV7670) { + cam->sensor = client; + return cafe_cam_init(cam); + } + return -EINVAL; +} + +static int cafe_smbus_detach(struct i2c_client *client) +{ + struct cafe_camera *cam = i2c_get_adapdata(client->adapter); + + if (cam->sensor == client) + cam->sensor = NULL; /* Bummer, no camera */ + return 0; +} + +static int cafe_smbus_setup(struct cafe_camera *cam) +{ + struct i2c_adapter *adap = &cam->i2c_adapter; + int ret; + + cafe_smbus_enable_irq(cam); + adap->id = I2C_HW_SMBUS_CAFE; + adap->class = I2C_CLASS_CAM_DIGITAL; + adap->owner = THIS_MODULE; + adap->client_register = cafe_smbus_attach; + adap->client_unregister = cafe_smbus_detach; + adap->algo = &cafe_smbus_algo; + strcpy(adap->name, "cafe_ccic"); + i2c_set_adapdata(adap, cam); + ret = i2c_add_adapter(adap); + if (ret) + printk(KERN_ERR "Unable to register cafe i2c adapter\n"); + return ret; +} + +static void cafe_smbus_shutdown(struct cafe_camera *cam) +{ + i2c_del_adapter(&cam->i2c_adapter); +} + + +/* ------------------------------------------------------------------- */ +/* + * Deal with the controller. + */ + +/* + * Do everything we think we need to have the interface operating + * according to the desired format. + */ +static void cafe_ctlr_dma(struct cafe_camera *cam) +{ + /* + * Store the first two Y buffers (we aren't supporting + * planar formats for now, so no UV bufs). Then either + * set the third if it exists, or tell the controller + * to just use two. + */ + cafe_reg_write(cam, REG_Y0BAR, cam->dma_handles[0]); + cafe_reg_write(cam, REG_Y1BAR, cam->dma_handles[1]); + if (cam->nbufs > 2) { + cafe_reg_write(cam, REG_Y2BAR, cam->dma_handles[2]); + cafe_reg_clear_bit(cam, REG_CTRL1, C1_TWOBUFS); + } + else + cafe_reg_set_bit(cam, REG_CTRL1, C1_TWOBUFS); + cafe_reg_write(cam, REG_UBAR, 0); /* 32 bits only for now */ +} + +static void cafe_ctlr_image(struct cafe_camera *cam) +{ + int imgsz; + struct v4l2_pix_format *fmt = &cam->pix_format; + + imgsz = ((fmt->height << IMGSZ_V_SHIFT) & IMGSZ_V_MASK) | + (fmt->bytesperline & IMGSZ_H_MASK); + cafe_reg_write(cam, REG_IMGSIZE, imgsz); + cafe_reg_write(cam, REG_IMGOFFSET, 0); + /* YPITCH just drops the last two bits */ + cafe_reg_write_mask(cam, REG_IMGPITCH, fmt->bytesperline, + IMGP_YP_MASK); + /* + * Tell the controller about the image format we are using. + */ + switch (cam->pix_format.pixelformat) { + case V4L2_PIX_FMT_YUYV: + cafe_reg_write_mask(cam, REG_CTRL0, + C0_DF_YUV|C0_YUV_PACKED|C0_YUVE_YUYV, + C0_DF_MASK); + break; + + /* + * For "fake rgb32" get the image pitch right. + */ + case V4L2_PIX_FMT_RGB32: + cafe_reg_write_mask(cam, REG_IMGPITCH, fmt->bytesperline/2, + IMGP_YP_MASK); + imgsz = ((fmt->height << IMGSZ_V_SHIFT) & IMGSZ_V_MASK) | + ((fmt->bytesperline/2) & IMGSZ_H_MASK); + cafe_reg_write(cam, REG_IMGSIZE, imgsz); + /* fall into ... */ + case V4L2_PIX_FMT_RGB444: + cafe_reg_write_mask(cam, REG_CTRL0, + C0_DF_RGB|C0_RGBF_444|C0_RGB4_XRGB, + C0_DF_MASK); + /* Alpha value? */ + break; + + case V4L2_PIX_FMT_RGB565: + cafe_reg_write_mask(cam, REG_CTRL0, + C0_DF_RGB|C0_RGBF_565|C0_RGB5_BGGR, + C0_DF_MASK); + break; + + default: + cam_err(cam, "Unknown format %x\n", cam->pix_format.pixelformat); + break; + } + /* + * Make sure it knows we want to use hsync/vsync. + */ + cafe_reg_write_mask(cam, REG_CTRL0, C0_SIF_HVSYNC, + C0_SIFM_MASK); +} + + +/* + * Configure the controller for operation; caller holds the + * device mutex. + */ +static int cafe_ctlr_configure(struct cafe_camera *cam) +{ + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_ctlr_dma(cam); + cafe_ctlr_image(cam); + cafe_set_config_needed(cam, 0); + spin_unlock_irqrestore(&cam->dev_lock, flags); + return 0; +} + +static void cafe_ctlr_irq_enable(struct cafe_camera *cam) +{ + /* + * Clear any pending interrupts, since we do not + * expect to have I/O active prior to enabling. + */ + cafe_reg_write(cam, REG_IRQSTAT, FRAMEIRQS); + cafe_reg_set_bit(cam, REG_IRQMASK, FRAMEIRQS); +} + +static void cafe_ctlr_irq_disable(struct cafe_camera *cam) +{ + cafe_reg_clear_bit(cam, REG_IRQMASK, FRAMEIRQS); +} + +/* + * Make the controller start grabbing images. Everything must + * be set up before doing this. + */ +static void cafe_ctlr_start(struct cafe_camera *cam) +{ + /* set_bit performs a read, so no other barrier should be + needed here */ + cafe_reg_set_bit(cam, REG_CTRL0, C0_ENABLE); +} + +static void cafe_ctlr_stop(struct cafe_camera *cam) +{ + cafe_reg_clear_bit(cam, REG_CTRL0, C0_ENABLE); +} + +static void cafe_ctlr_init(struct cafe_camera *cam) +{ + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + /* + * Added magic to bring up the hardware on the B-Test board + */ + cafe_reg_write(cam, 0x3038, 0x8); + cafe_reg_write(cam, 0x315c, 0x80008); + /* + * Go through the dance needed to wake the device up. + * Note that these registers are global and shared + * with the NAND and SD devices. Interaction between the + * three still needs to be examined. + */ + cafe_reg_write(cam, REG_GL_CSR, GCSR_SRS|GCSR_MRS); /* Needed? */ + cafe_reg_write(cam, REG_GL_CSR, GCSR_SRC|GCSR_MRC); + cafe_reg_write(cam, REG_GL_CSR, GCSR_SRC|GCSR_MRS); + mdelay(5); /* FIXME revisit this */ + cafe_reg_write(cam, REG_GL_CSR, GCSR_CCIC_EN|GCSR_SRC|GCSR_MRC); + cafe_reg_set_bit(cam, REG_GL_IMASK, GIMSK_CCIC_EN); + /* + * Make sure it's not powered down. + */ + cafe_reg_clear_bit(cam, REG_CTRL1, C1_PWRDWN); + /* + * Turn off the enable bit. It sure should be off anyway, + * but it's good to be sure. + */ + cafe_reg_clear_bit(cam, REG_CTRL0, C0_ENABLE); + /* + * Mask all interrupts. + */ + cafe_reg_write(cam, REG_IRQMASK, 0); + /* + * Clock the sensor appropriately. Controller clock should + * be 48MHz, sensor "typical" value is half that. + */ + cafe_reg_write_mask(cam, REG_CLKCTRL, 2, CLK_DIV_MASK); + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + + +/* + * Stop the controller, and don't return until we're really sure that no + * further DMA is going on. + */ +static void cafe_ctlr_stop_dma(struct cafe_camera *cam) +{ + unsigned long flags; + + /* + * Theory: stop the camera controller (whether it is operating + * or not). Delay briefly just in case we race with the SOF + * interrupt, then wait until no DMA is active. + */ + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_ctlr_stop(cam); + spin_unlock_irqrestore(&cam->dev_lock, flags); + mdelay(1); + wait_event_timeout(cam->iowait, + !test_bit(CF_DMA_ACTIVE, &cam->flags), HZ); + if (test_bit(CF_DMA_ACTIVE, &cam->flags)) + cam_err(cam, "Timeout waiting for DMA to end\n"); + /* This would be bad news - what now? */ + spin_lock_irqsave(&cam->dev_lock, flags); + cam->state = S_IDLE; + cafe_ctlr_irq_disable(cam); + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + +/* + * Power up and down. + */ +static void cafe_ctlr_power_up(struct cafe_camera *cam) +{ + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_reg_clear_bit(cam, REG_CTRL1, C1_PWRDWN); + /* + * Put the sensor into operational mode (assumes OLPC-style + * wiring). Control 0 is reset - set to 1 to operate. + * Control 1 is power down, set to 0 to operate. + */ + cafe_reg_write(cam, REG_GPR, GPR_C1EN|GPR_C0EN|GPR_C1); + mdelay(1); /* Marvell says 1ms will do it */ + cafe_reg_write(cam, REG_GPR, GPR_C1EN|GPR_C0EN|GPR_C0); + mdelay(1); /* Enough? */ + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + +static void cafe_ctlr_power_down(struct cafe_camera *cam) +{ + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_reg_write(cam, REG_GPR, GPR_C1EN|GPR_C0EN|GPR_C1); + cafe_reg_set_bit(cam, REG_CTRL1, C1_PWRDWN); + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + +/* -------------------------------------------------------------------- */ +/* + * Communications with the sensor. + */ + +static int __cafe_cam_cmd(struct cafe_camera *cam, int cmd, void *arg) +{ + struct i2c_client *sc = cam->sensor; + int ret; + + if (sc == NULL || sc->driver == NULL || sc->driver->command == NULL) + return -EINVAL; + ret = sc->driver->command(sc, cmd, arg); + if (ret == -EPERM) /* Unsupported command */ + return 0; + return ret; +} + +static int __cafe_cam_reset(struct cafe_camera *cam) +{ + int zero = 0; + return __cafe_cam_cmd(cam, VIDIOC_INT_RESET, &zero); +} + +/* + * We have found the sensor on the i2c. Let's try to have a + * conversation. + */ +static int cafe_cam_init(struct cafe_camera *cam) +{ + int ret; + + mutex_lock(&cam->s_mutex); + if (cam->state != S_NOTREADY) + cam_warn(cam, "Cam init with device in funky state %d", + cam->state); + ret = __cafe_cam_reset(cam); + if (ret) + goto out; + ret = __cafe_cam_cmd(cam, VIDIOC_INT_G_CHIP_IDENT, &cam->sensor_type); + if (ret) + goto out; +// if (cam->sensor->addr != OV7xx0_SID) { + if (cam->sensor_type != V4L2_IDENT_OV7670) { + cam_err(cam, "Unsupported sensor type %d", cam->sensor->addr); + ret = -EINVAL; + goto out; + } +/* Get/set parameters? */ + ret = 0; + cam->state = S_IDLE; + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + +/* + * Configure the sensor to match the parameters we have. Caller should + * hold s_mutex + */ +static int cafe_cam_set_flip(struct cafe_camera *cam) +{ + struct v4l2_control ctrl; + + memset(&ctrl, 0, sizeof(ctrl)); + ctrl.id = V4L2_CID_VFLIP; + ctrl.value = flip; + return __cafe_cam_cmd(cam, VIDIOC_S_CTRL, &ctrl); +} + + +static int cafe_cam_configure(struct cafe_camera *cam) +{ + struct v4l2_format fmt; + int ret, zero = 0; + + if (cam->state != S_IDLE) + return -EINVAL; + fmt.fmt.pix = cam->pix_format; + ret = __cafe_cam_cmd(cam, VIDIOC_INT_INIT, &zero); + if (ret == 0) + ret = __cafe_cam_cmd(cam, VIDIOC_S_FMT, &fmt); + /* + * OV7670 does weird things if flip is set *before* format... + */ + ret += cafe_cam_set_flip(cam); + return ret; +} + +/* -------------------------------------------------------------------- */ +/* + * DMA buffer management. These functions need s_mutex held. + */ + +/* FIXME: this is inefficient as hell, since dma_alloc_coherent just + * does a get_free_pages() call, and we waste a good chunk of an orderN + * allocation. Should try to allocate the whole set in one chunk. + */ +static int cafe_alloc_dma_bufs(struct cafe_camera *cam, int loadtime) +{ + int i; + + cafe_set_config_needed(cam, 1); + if (loadtime) + cam->dma_buf_size = dma_buf_size; + else { + cam->dma_buf_size = cam->pix_format.sizeimage; + if (cam->pix_format.pixelformat == V4L2_PIX_FMT_RGB32) + cam->dma_buf_size /= 2; + } + if (n_dma_bufs > 3) + n_dma_bufs = 3; + + cam->nbufs = 0; + for (i = 0; i < n_dma_bufs; i++) { + cam->dma_bufs[i] = dma_alloc_coherent(&cam->pdev->dev, + cam->dma_buf_size, cam->dma_handles + i, + GFP_KERNEL); + if (cam->dma_bufs[i] == NULL) { + cam_warn(cam, "Failed to allocate DMA buffer\n"); + break; + } + /* For debug, remove eventually */ + memset(cam->dma_bufs[i], 0xcc, cam->dma_buf_size); + (cam->nbufs)++; + } + + switch (cam->nbufs) { + case 1: + dma_free_coherent(&cam->pdev->dev, cam->dma_buf_size, + cam->dma_bufs[0], cam->dma_handles[0]); + cam->nbufs = 0; + case 0: + cam_err(cam, "Insufficient DMA buffers, cannot operate\n"); + return -ENOMEM; + + case 2: + if (n_dma_bufs > 2) + cam_warn(cam, "Will limp along with only 2 buffers\n"); + break; + } + return 0; +} + +static void cafe_free_dma_bufs(struct cafe_camera *cam) +{ + int i; + + for (i = 0; i < cam->nbufs; i++) { + dma_free_coherent(&cam->pdev->dev, cam->dma_buf_size, + cam->dma_bufs[i], cam->dma_handles[i]); + cam->dma_bufs[i] = NULL; + } + cam->nbufs = 0; +} + + + + + +/* ----------------------------------------------------------------------- */ +/* + * Here starts the V4L2 interface code. + */ + +/* + * Read an image from the device. + */ +static ssize_t cafe_deliver_buffer(struct cafe_camera *cam, + char __user *buffer, size_t len, loff_t *pos) +{ + int bufno; + unsigned long flags; + + spin_lock_irqsave(&cam->dev_lock, flags); + if (cam->next_buf < 0) { + cam_err(cam, "deliver_buffer: No next buffer\n"); + spin_unlock_irqrestore(&cam->dev_lock, flags); + return -EIO; + } + bufno = cam->next_buf; + clear_bit(bufno, &cam->flags); + if (++(cam->next_buf) >= cam->nbufs) + cam->next_buf = 0; + if (! test_bit(cam->next_buf, &cam->flags)) + cam->next_buf = -1; + cam->specframes = 0; + spin_unlock_irqrestore(&cam->dev_lock, flags); + + if (len > cam->pix_format.sizeimage) + len = cam->pix_format.sizeimage; + if (copy_to_user(buffer, cam->dma_bufs[bufno], len)) + return -EFAULT; + (*pos) += len; + return len; +} + +/* + * Get everything ready, and start grabbing frames. + */ +static int cafe_read_setup(struct cafe_camera *cam, enum cafe_state state) +{ + int ret; + unsigned long flags; + + /* + * Configuration. If we still don't have DMA buffers, + * make one last, desperate attempt. + */ + if (cam->nbufs == 0) + if (cafe_alloc_dma_bufs(cam, 0)) + return -ENOMEM; + + if (cafe_needs_config(cam)) { + cafe_cam_configure(cam); + ret = cafe_ctlr_configure(cam); + if (ret) + return ret; + } + + /* + * Turn it loose. + */ + spin_lock_irqsave(&cam->dev_lock, flags); + cafe_reset_buffers(cam); + cafe_ctlr_irq_enable(cam); + cam->state = state; + cafe_ctlr_start(cam); + spin_unlock_irqrestore(&cam->dev_lock, flags); + return 0; +} + + +static ssize_t cafe_v4l_read(struct file *filp, + char __user *buffer, size_t len, loff_t *pos) +{ + struct cafe_camera *cam = filp->private_data; + int ret; + + /* + * Perhaps we're in speculative read mode and already + * have data? + */ + mutex_lock(&cam->s_mutex); + if (cam->state == S_SPECREAD) { + if (cam->next_buf >= 0) { + ret = cafe_deliver_buffer(cam, buffer, len, pos); + if (ret != 0) + goto out_unlock; + } + } else if (cam->state == S_FLAKED || cam->state == S_NOTREADY) { + ret = -EIO; + goto out_unlock; + } else if (cam->state != S_IDLE) { + ret = -EBUSY; + goto out_unlock; + } + + /* + * v4l2: multiple processes can open the device, but only + * one gets to grab data from it. + */ + if (cam->owner && cam->owner != filp) { + ret = -EBUSY; + goto out_unlock; + } + cam->owner = filp; + + /* + * Do setup if need be. + */ + if (cam->state != S_SPECREAD) { + ret = cafe_read_setup(cam, S_SINGLEREAD); + if (ret) + goto out_unlock; + } + /* + * Wait for something to happen. This should probably + * be interruptible (FIXME). + */ + wait_event_timeout(cam->iowait, cam->next_buf >= 0, HZ); + if (cam->next_buf < 0) { + cam_err(cam, "read() operation timed out\n"); + cafe_ctlr_stop_dma(cam); + ret = -EIO; + goto out_unlock; + } + /* + * Give them their data and we should be done. + */ + ret = cafe_deliver_buffer(cam, buffer, len, pos); + + out_unlock: + mutex_unlock(&cam->s_mutex); + return ret; +} + + + + + + + + +/* + * Streaming I/O support. + */ + + + +static int cafe_vidioc_streamon(struct file *filp, void *priv, + enum v4l2_buf_type type) +{ + struct cafe_camera *cam = filp->private_data; + int ret = -EINVAL; + + if (type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + goto out; + mutex_lock(&cam->s_mutex); + if (cam->state != S_IDLE || cam->n_sbufs == 0) + goto out_unlock; + + cam->sequence = 0; + ret = cafe_read_setup(cam, S_STREAMING); + + out_unlock: + mutex_unlock(&cam->s_mutex); + out: + return ret; +} + + +static int cafe_vidioc_streamoff(struct file *filp, void *priv, + enum v4l2_buf_type type) +{ + struct cafe_camera *cam = filp->private_data; + int ret = -EINVAL; + + if (type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + goto out; + mutex_lock(&cam->s_mutex); + if (cam->state != S_STREAMING) + goto out_unlock; + + cafe_ctlr_stop_dma(cam); + ret = 0; + + out_unlock: + mutex_unlock(&cam->s_mutex); + out: + return ret; +} + + + +static int cafe_setup_siobuf(struct cafe_camera *cam, int index) +{ + struct cafe_sio_buffer *buf = cam->sb_bufs + index; + + INIT_LIST_HEAD(&buf->list); + buf->v4lbuf.length = PAGE_ALIGN(cam->pix_format.sizeimage); + buf->buffer = vmalloc_user(buf->v4lbuf.length); + if (buf->buffer == NULL) + return -ENOMEM; + buf->mapcount = 0; + buf->cam = cam; + + buf->v4lbuf.index = index; + buf->v4lbuf.type = V4L2_BUF_TYPE_VIDEO_CAPTURE; + buf->v4lbuf.field = V4L2_FIELD_NONE; + buf->v4lbuf.memory = V4L2_MEMORY_MMAP; + /* + * Offset: must be 32-bit even on a 64-bit system. video-buf + * just uses the length times the index, but the spec warns + * against doing just that - vma merging problems. So we + * leave a gap between each pair of buffers. + */ + buf->v4lbuf.m.offset = 2*index*buf->v4lbuf.length; + return 0; +} + +static int cafe_free_sio_buffers(struct cafe_camera *cam) +{ + int i; + + /* + * If any buffers are mapped, we cannot free them at all. + */ + for (i = 0; i < cam->n_sbufs; i++) + if (cam->sb_bufs[i].mapcount > 0) + return -EBUSY; + /* + * OK, let's do it. + */ + for (i = 0; i < cam->n_sbufs; i++) + vfree(cam->sb_bufs[i].buffer); + cam->n_sbufs = 0; + kfree(cam->sb_bufs); + cam->sb_bufs = NULL; + INIT_LIST_HEAD(&cam->sb_avail); + INIT_LIST_HEAD(&cam->sb_full); + return 0; +} + + + +static int cafe_vidioc_reqbufs(struct file *filp, void *priv, + struct v4l2_requestbuffers *req) +{ + struct cafe_camera *cam = filp->private_data; + int ret; + + /* + * Make sure it's something we can do. User pointers could be + * implemented without great pain, but that's not been done yet. + */ + if (req->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + return -EINVAL; + if (req->memory != V4L2_MEMORY_MMAP) + return -EINVAL; + /* + * If they ask for zero buffers, they really want us to stop streaming + * (if it's happening) and free everything. Should we check owner? + */ + mutex_lock(&cam->s_mutex); + if (req->count == 0) { + if (cam->state == S_STREAMING) + cafe_ctlr_stop_dma(cam); + ret = cafe_free_sio_buffers (cam); + goto out; + } + /* + * Device needs to be idle and working. We *could* try to do the + * right thing in S_SPECREAD by shutting things down, but it + * probably doesn't matter. + */ + if (cam->state != S_IDLE || (cam->owner && cam->owner != filp)) { + ret = -EBUSY; + goto out; + } + cam->owner = filp; + + if (req->count < min_buffers) + req->count = min_buffers; + else if (req->count > max_buffers) + req->count = max_buffers; + if (cam->n_sbufs > 0) { + ret = cafe_free_sio_buffers(cam); + if (ret) + goto out; + } + + cam->sb_bufs = kzalloc(req->count*sizeof(struct cafe_sio_buffer), + GFP_KERNEL); + if (cam->sb_bufs == NULL) { + ret = -ENOMEM; + goto out; + } + for (cam->n_sbufs = 0; cam->n_sbufs < req->count; (cam->n_sbufs++)) { + ret = cafe_setup_siobuf(cam, cam->n_sbufs); + if (ret) + break; + } + + if (cam->n_sbufs == 0) /* no luck at all - ret already set */ + kfree(cam->sb_bufs); + else + ret = 0; + req->count = cam->n_sbufs; /* In case of partial success */ + + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + + +static int cafe_vidioc_querybuf(struct file *filp, void *priv, + struct v4l2_buffer *buf) +{ + struct cafe_camera *cam = filp->private_data; + int ret = -EINVAL; + + mutex_lock(&cam->s_mutex); + if (buf->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + goto out; + if (buf->index < 0 || buf->index >= cam->n_sbufs) + goto out; + *buf = cam->sb_bufs[buf->index].v4lbuf; + ret = 0; + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + +static int cafe_vidioc_qbuf(struct file *filp, void *priv, + struct v4l2_buffer *buf) +{ + struct cafe_camera *cam = filp->private_data; + struct cafe_sio_buffer *sbuf; + int ret = -EINVAL; + unsigned long flags; + + mutex_lock(&cam->s_mutex); + if (buf->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + goto out; + if (buf->index < 0 || buf->index >= cam->n_sbufs) + goto out; + sbuf = cam->sb_bufs + buf->index; + if (sbuf->v4lbuf.flags & V4L2_BUF_FLAG_QUEUED) { + ret = 0; /* Already queued?? */ + goto out; + } + if (sbuf->v4lbuf.flags & V4L2_BUF_FLAG_DONE) { + /* Spec doesn't say anything, seems appropriate tho */ + ret = -EBUSY; + goto out; + } + sbuf->v4lbuf.flags |= V4L2_BUF_FLAG_QUEUED; + spin_lock_irqsave(&cam->dev_lock, flags); + list_add(&sbuf->list, &cam->sb_avail); + spin_unlock_irqrestore(&cam->dev_lock, flags); + ret = 0; + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + +static int cafe_vidioc_dqbuf(struct file *filp, void *priv, + struct v4l2_buffer *buf) +{ + struct cafe_camera *cam = filp->private_data; + struct cafe_sio_buffer *sbuf; + int ret = -EINVAL; + unsigned long flags; + + mutex_lock(&cam->s_mutex); + if (buf->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + goto out_unlock; + if (cam->state != S_STREAMING) + goto out_unlock; + if (list_empty(&cam->sb_full) && filp->f_flags & O_NONBLOCK) { + ret = -EAGAIN; + goto out_unlock; + } + + while (list_empty(&cam->sb_full) && cam->state == S_STREAMING) { + mutex_unlock(&cam->s_mutex); + if (wait_event_interruptible(cam->iowait, + !list_empty(&cam->sb_full))) { + ret = -ERESTARTSYS; + goto out; + } + mutex_lock(&cam->s_mutex); + } + + if (cam->state != S_STREAMING) + ret = -EINTR; + else { + spin_lock_irqsave(&cam->dev_lock, flags); + /* Should probably recheck !list_empty() here */ + sbuf = list_entry(cam->sb_full.next, + struct cafe_sio_buffer, list); + list_del_init(&sbuf->list); + spin_unlock_irqrestore(&cam->dev_lock, flags); + sbuf->v4lbuf.flags &= ~V4L2_BUF_FLAG_DONE; + *buf = sbuf->v4lbuf; + ret = 0; + } + + out_unlock: + mutex_unlock(&cam->s_mutex); + out: + return ret; +} + + + +static void cafe_v4l_vm_open(struct vm_area_struct *vma) +{ + struct cafe_sio_buffer *sbuf = vma->vm_private_data; + /* + * Locking: done under mmap_sem, so we don't need to + * go back to the camera lock here. + */ + sbuf->mapcount++; +} + + +static void cafe_v4l_vm_close(struct vm_area_struct *vma) +{ + struct cafe_sio_buffer *sbuf = vma->vm_private_data; + + mutex_lock(&sbuf->cam->s_mutex); + sbuf->mapcount--; + /* Docs say we should stop I/O too... */ + if (sbuf->mapcount == 0) + sbuf->v4lbuf.flags &= ~V4L2_BUF_FLAG_MAPPED; + mutex_unlock(&sbuf->cam->s_mutex); +} + +static struct vm_operations_struct cafe_v4l_vm_ops = { + .open = cafe_v4l_vm_open, + .close = cafe_v4l_vm_close +}; + + +static int cafe_v4l_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct cafe_camera *cam = filp->private_data; + unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; + int ret = -EINVAL; + int i; + struct cafe_sio_buffer *sbuf = NULL; + + if (! (vma->vm_flags & VM_WRITE) || ! (vma->vm_flags & VM_SHARED)) + return -EINVAL; + /* + * Find the buffer they are looking for. + */ + mutex_lock(&cam->s_mutex); + for (i = 0; i < cam->n_sbufs; i++) + if (cam->sb_bufs[i].v4lbuf.m.offset == offset) { + sbuf = cam->sb_bufs + i; + break; + } + if (sbuf == NULL) + goto out; + + ret = remap_vmalloc_range(vma, sbuf->buffer, 0); + if (ret) + goto out; + vma->vm_flags |= VM_DONTEXPAND; + vma->vm_private_data = sbuf; + vma->vm_ops = &cafe_v4l_vm_ops; + sbuf->v4lbuf.flags |= V4L2_BUF_FLAG_MAPPED; + cafe_v4l_vm_open(vma); + ret = 0; + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + + + +static int cafe_v4l_open(struct inode *inode, struct file *filp) +{ + struct cafe_camera *cam; + + cam = cafe_find_dev(iminor(inode)); + if (cam == NULL) + return -ENODEV; + filp->private_data = cam; + + mutex_lock(&cam->s_mutex); + if (cam->users == 0) { + cafe_ctlr_power_up(cam); + __cafe_cam_reset(cam); + cafe_set_config_needed(cam, 1); + /* FIXME make sure this is complete */ + } + (cam->users)++; + mutex_unlock(&cam->s_mutex); + return 0; +} + + +static int cafe_v4l_release(struct inode *inode, struct file *filp) +{ + struct cafe_camera *cam = filp->private_data; + + mutex_lock(&cam->s_mutex); + (cam->users)--; + if (filp == cam->owner) { + cafe_ctlr_stop_dma(cam); + cafe_free_sio_buffers(cam); + cam->owner = NULL; + } + if (cam->users == 0) + cafe_ctlr_power_down(cam); + mutex_unlock(&cam->s_mutex); + return 0; +} + + + +static unsigned int cafe_v4l_poll(struct file *filp, + struct poll_table_struct *pt) +{ + struct cafe_camera *cam = filp->private_data; + + poll_wait(filp, &cam->iowait, pt); + if (cam->next_buf >= 0) + return POLLIN | POLLRDNORM; + return 0; +} + + + +static int cafe_vidioc_queryctrl(struct file *filp, void *priv, + struct v4l2_queryctrl *qc) +{ + struct cafe_camera *cam = filp->private_data; + int ret; + + mutex_lock(&cam->s_mutex); + ret = __cafe_cam_cmd(cam, VIDIOC_QUERYCTRL, qc); + mutex_unlock(&cam->s_mutex); + return ret; +} + + +static int cafe_vidioc_g_ctrl(struct file *filp, void *priv, + struct v4l2_control *ctrl) +{ + struct cafe_camera *cam = filp->private_data; + int ret; + + mutex_lock(&cam->s_mutex); + ret = __cafe_cam_cmd(cam, VIDIOC_G_CTRL, ctrl); + mutex_unlock(&cam->s_mutex); + return ret; +} + + +static int cafe_vidioc_s_ctrl(struct file *filp, void *priv, + struct v4l2_control *ctrl) +{ + struct cafe_camera *cam = filp->private_data; + int ret; + + mutex_lock(&cam->s_mutex); + ret = __cafe_cam_cmd(cam, VIDIOC_S_CTRL, ctrl); + mutex_unlock(&cam->s_mutex); + return ret; +} + + + + + +static int cafe_vidioc_querycap(struct file *file, void *priv, + struct v4l2_capability *cap) +{ + strcpy(cap->driver, "cafe_ccic"); + strcpy(cap->card, "cafe_ccic"); + cap->version = CAFE_VERSION; + cap->capabilities = V4L2_CAP_VIDEO_CAPTURE | + V4L2_CAP_READWRITE | V4L2_CAP_STREAMING; + return 0; +} + + +/* + * The default format we use until somebody says otherwise. + */ +static struct v4l2_pix_format cafe_def_pix_format = { + .width = VGA_WIDTH, + .height = VGA_HEIGHT, + .pixelformat = V4L2_PIX_FMT_YUYV, + .field = V4L2_FIELD_NONE, + .bytesperline = VGA_WIDTH*2, + .sizeimage = VGA_WIDTH*VGA_HEIGHT*2, +}; + +static int cafe_vidioc_enum_fmt_cap(struct file *filp, + void *priv, struct v4l2_fmtdesc *fmt) +{ + struct cafe_camera *cam = priv; + int ret; + + if (fmt->type != V4L2_BUF_TYPE_VIDEO_CAPTURE) + return -EINVAL; + mutex_lock(&cam->s_mutex); + ret = __cafe_cam_cmd(cam, VIDIOC_ENUM_FMT, fmt); + mutex_unlock(&cam->s_mutex); + return ret; +} + + +static int cafe_vidioc_try_fmt_cap (struct file *filp, void *priv, + struct v4l2_format *fmt) +{ + struct cafe_camera *cam = priv; + int ret; + + mutex_lock(&cam->s_mutex); + ret = __cafe_cam_cmd(cam, VIDIOC_TRY_FMT, fmt); + mutex_unlock(&cam->s_mutex); + return ret; +} + +static int cafe_vidioc_s_fmt_cap(struct file *filp, void *priv, + struct v4l2_format *fmt) +{ + struct cafe_camera *cam = priv; + int ret; + + /* + * Can't do anything if the device is not idle + * Also can't if there are streaming buffers in place. + */ + if (cam->state != S_IDLE || cam->n_sbufs > 0) + return -EBUSY; + /* + * See if the formatting works in principle. + */ + ret = cafe_vidioc_try_fmt_cap(filp, priv, fmt); + if (ret) + return ret; + /* + * Now we start to change things for real, so let's do it + * under lock. + */ + mutex_lock(&cam->s_mutex); + cam->pix_format = fmt->fmt.pix; + /* + * Make sure we have appropriate DMA buffers. + */ + ret = -ENOMEM; + if (cam->nbufs > 0 && cam->dma_buf_size < cam->pix_format.sizeimage) + cafe_free_dma_bufs(cam); + if (cam->nbufs == 0) { + if (cafe_alloc_dma_bufs(cam, 0)) + goto out; + } + /* + * It looks like this might work, so let's program the sensor. + */ + ret = cafe_cam_configure(cam); + if (! ret) + ret = cafe_ctlr_configure(cam); + out: + mutex_unlock(&cam->s_mutex); + return ret; +} + +/* + * Return our stored notion of how the camera is/should be configured. + * The V4l2 spec wants us to be smarter, and actually get this from + * the camera (and not mess with it at open time). Someday. + */ +static int cafe_vidioc_g_fmt_cap(struct file *filp, void *priv, + struct v4l2_format *f) +{ + struct cafe_camera *cam = priv; + + f->fmt.pix = cam->pix_format; + return 0; +} + +/* + * We only have one input - the sensor - so minimize the nonsense here. + */ +static int cafe_vidioc_enum_input(struct file *filp, void *priv, + struct v4l2_input *input) +{ + if (input->index != 0) + return -EINVAL; + + input->type = V4L2_INPUT_TYPE_CAMERA; + input->std = V4L2_STD_ALL; /* Not sure what should go here */ + strcpy(input->name, "Camera"); + return 0; +} + +static int cafe_vidioc_g_input(struct file *filp, void *priv, unsigned int *i) +{ + *i = 0; + return 0; +} + +static int cafe_vidioc_s_input(struct file *filp, void *priv, unsigned int i) +{ + if (i != 0) + return -EINVAL; + return 0; +} + +/* from vivi.c */ +static int cafe_vidioc_s_std(struct file *filp, void *priv, v4l2_std_id a) +{ + return 0; +} + + +/* + * The TV Norm stuff is weird - we're a camera with little to do with TV, + * really. The following is what vivi does. + */ +static struct v4l2_tvnorm cafe_tvnorm[] = { + { + .name = "NTSC-M", + .id = V4L2_STD_NTSC_M, + } +}; + + +void cafe_v4l_dev_release(struct video_device *vd) +{ + struct cafe_camera *cam = container_of(vd, struct cafe_camera, v4ldev); + + kfree(cam); +} + + +/* + * This template device holds all of those v4l2 methods; we + * clone it for specific real devices. + */ + +static struct file_operations cafe_v4l_fops = { + .owner = THIS_MODULE, + .open = cafe_v4l_open, + .release = cafe_v4l_release, + .read = cafe_v4l_read, + .poll = cafe_v4l_poll, + .mmap = cafe_v4l_mmap, + .ioctl = video_ioctl2, + .llseek = no_llseek, +}; + +static struct video_device cafe_v4l_template = { + .name = "cafe", + .type = VFL_TYPE_GRABBER, + .type2 = VID_TYPE_CAPTURE, + .minor = -1, /* Get one dynamically */ + .tvnorms = cafe_tvnorm, + .tvnormsize = 1, + .current_norm = V4L2_STD_NTSC_M, /* make mplayer happy */ + + .fops = &cafe_v4l_fops, + .release = cafe_v4l_dev_release, + + .vidioc_querycap = cafe_vidioc_querycap, + .vidioc_enum_fmt_cap = cafe_vidioc_enum_fmt_cap, + .vidioc_try_fmt_cap = cafe_vidioc_try_fmt_cap, + .vidioc_s_fmt_cap = cafe_vidioc_s_fmt_cap, + .vidioc_g_fmt_cap = cafe_vidioc_g_fmt_cap, + .vidioc_enum_input = cafe_vidioc_enum_input, + .vidioc_g_input = cafe_vidioc_g_input, + .vidioc_s_input = cafe_vidioc_s_input, + .vidioc_s_std = cafe_vidioc_s_std, + .vidioc_reqbufs = cafe_vidioc_reqbufs, + .vidioc_querybuf = cafe_vidioc_querybuf, + .vidioc_qbuf = cafe_vidioc_qbuf, + .vidioc_dqbuf = cafe_vidioc_dqbuf, + .vidioc_streamon = cafe_vidioc_streamon, + .vidioc_streamoff = cafe_vidioc_streamoff, + .vidioc_queryctrl = cafe_vidioc_queryctrl, + .vidioc_g_ctrl = cafe_vidioc_g_ctrl, + .vidioc_s_ctrl = cafe_vidioc_s_ctrl, + /* Do cropping someday */ +}; + + + + + + + +/* ---------------------------------------------------------------------- */ +/* + * Interrupt handler stuff + */ + +/* + * Create RGB32 from RGB444 so it can be displayed before the applications + * know about the latter format. + */ +static void cafe_fake_rgb32(struct cafe_camera *cam, char *dest, char *src) +{ + int i; + u16 *ssrc = (u16 *) src; + + /* RGB444 version */ + for (i = 0; i < cam->pix_format.sizeimage; i += 4) { + // dest[0] = (*ssrc & 0xf000) >> 8; + dest[0] = (*ssrc & 0x000f) << 4; + dest[1] = (*ssrc & 0x00f0); + dest[2] = (*ssrc & 0x0f00) >> 4; + dest[3] = (*ssrc & 0xf000); /* Alpha */ + dest += 4; + ssrc++; + } +} + + +static void cafe_frame_tasklet(unsigned long data) +{ + struct cafe_camera *cam = (struct cafe_camera *) data; + int i; + unsigned long flags; + struct cafe_sio_buffer *sbuf; + + spin_lock_irqsave(&cam->dev_lock, flags); + for (i = 0; i < cam->nbufs; i++) { + int bufno = cam->next_buf; + if (bufno < 0) { /* "will never happen" */ + cam_err(cam, "No valid bufs in tasklet!\n"); + break; + } + if (++(cam->next_buf) >= cam->nbufs) + cam->next_buf = 0; + if (! test_bit(bufno, &cam->flags)) + continue; + if (list_empty(&cam->sb_avail)) + break; /* Leave it valid, hope for better later */ + clear_bit(bufno, &cam->flags); + /* + * We could perhaps drop the spinlock during this + * big copy. Something to consider. + */ + sbuf = list_entry(cam->sb_avail.next, + struct cafe_sio_buffer, list); + if (cam->pix_format.pixelformat == V4L2_PIX_FMT_RGB32) + cafe_fake_rgb32(cam, sbuf->buffer, cam->dma_bufs[bufno]); + else + memcpy(sbuf->buffer, cam->dma_bufs[bufno], + cam->pix_format.sizeimage); + sbuf->v4lbuf.bytesused = cam->pix_format.sizeimage; + sbuf->v4lbuf.sequence = cam->buf_seq[bufno]; + sbuf->v4lbuf.flags &= ~V4L2_BUF_FLAG_QUEUED; + sbuf->v4lbuf.flags |= V4L2_BUF_FLAG_DONE; + list_move_tail(&sbuf->list, &cam->sb_full); + } + if (! list_empty(&cam->sb_full)) + wake_up(&cam->iowait); + spin_unlock_irqrestore(&cam->dev_lock, flags); +} + + + +static void cafe_frame_complete(struct cafe_camera *cam, int frame) +{ + /* + * Basic frame housekeeping. + */ + if (test_bit(frame, &cam->flags) && printk_ratelimit()) + cam_err(cam, "Frame overrun on %d, frames lost\n", frame); + set_bit(frame, &cam->flags); + clear_bit(CF_DMA_ACTIVE, &cam->flags); + if (cam->next_buf < 0) + cam->next_buf = frame; + cam->buf_seq[frame] = ++(cam->sequence); + + switch (cam->state) { + /* + * If in single read mode, try going speculative. + */ + case S_SINGLEREAD: + cam->state = S_SPECREAD; + cam->specframes = 0; + wake_up(&cam->iowait); + break; + + /* + * If we are already doing speculative reads, and nobody is + * reading them, just stop. + */ + case S_SPECREAD: + if (++(cam->specframes) >= cam->nbufs) { + cafe_ctlr_stop(cam); + cafe_ctlr_irq_disable(cam); + cam->state = S_IDLE; + } + wake_up(&cam->iowait); + break; + /* + * For the streaming case, we defer the real work to the + * camera tasklet. + * + * FIXME: if the application is not consuming the buffers, + * we should eventually put things on hold and restart in + * vidioc_dqbuf(). + */ + case S_STREAMING: + tasklet_schedule(&cam->s_tasklet); + break; + + default: + cam_err(cam, "Frame interrupt in non-operational state\n"); + break; + } +} + + + + +static void cafe_frame_irq(struct cafe_camera *cam, unsigned int irqs) +{ + unsigned int frame; + + cafe_reg_write(cam, REG_IRQSTAT, FRAMEIRQS); /* Clear'em all */ + /* + * Handle any frame completions. There really should + * not be more than one of these, or we have fallen + * far behind. + */ + for (frame = 0; frame < cam->nbufs; frame++) + if (irqs & (IRQ_EOF0 << frame)) + cafe_frame_complete(cam, frame); + /* + * If a frame starts, note that we have DMA active. This + * code assumes that we won't get multiple frame interrupts + * at once; may want to rethink that. + */ + if (irqs & (IRQ_SOF0 | IRQ_SOF1 | IRQ_SOF2)) + set_bit(CF_DMA_ACTIVE, &cam->flags); +} + + + +static irqreturn_t cafe_irq(int irq, void *data) +{ + struct cafe_camera *cam = data; + unsigned int irqs; + + spin_lock(&cam->dev_lock); + irqs = cafe_reg_read(cam, REG_IRQSTAT); + if ((irqs & ALLIRQS) == 0) { + spin_unlock(&cam->dev_lock); + return IRQ_NONE; + } + if (irqs & FRAMEIRQS) + cafe_frame_irq(cam, irqs); + if (irqs & TWSIIRQS) { + cafe_reg_write(cam, REG_IRQSTAT, TWSIIRQS); + wake_up(&cam->smbus_wait); + } + spin_unlock(&cam->dev_lock); + return IRQ_HANDLED; +} + + +/* -------------------------------------------------------------------------- */ +#ifdef CONFIG_VIDEO_ADV_DEBUG +/* + * Debugfs stuff. + */ + +static char cafe_debug_buf[1024]; +static struct dentry *cafe_dfs_root; + +static void cafe_dfs_setup(void) +{ + cafe_dfs_root = debugfs_create_dir("cafe_ccic", NULL); + if (IS_ERR(cafe_dfs_root)) { + cafe_dfs_root = NULL; /* Never mind */ + printk(KERN_NOTICE "cafe_ccic unable to set up debugfs\n"); + } +} + +static void cafe_dfs_shutdown(void) +{ + if (cafe_dfs_root) + debugfs_remove(cafe_dfs_root); +} + +static int cafe_dfs_open(struct inode *inode, struct file *file) +{ + file->private_data = inode->i_private; + return 0; +} + +static ssize_t cafe_dfs_read_regs(struct file *file, + char __user *buf, size_t count, loff_t *ppos) +{ + struct cafe_camera *cam = file->private_data; + char *s = cafe_debug_buf; + int offset; + + for (offset = 0; offset < 0x44; offset += 4) + s += sprintf(s, "%02x: %08x\n", offset, + cafe_reg_read(cam, offset)); + for (offset = 0x88; offset <= 0x90; offset += 4) + s += sprintf(s, "%02x: %08x\n", offset, + cafe_reg_read(cam, offset)); + for (offset = 0xb4; offset <= 0xbc; offset += 4) + s += sprintf(s, "%02x: %08x\n", offset, + cafe_reg_read(cam, offset)); + for (offset = 0x3000; offset <= 0x300c; offset += 4) + s += sprintf(s, "%04x: %08x\n", offset, + cafe_reg_read(cam, offset)); + return simple_read_from_buffer(buf, count, ppos, cafe_debug_buf, + s - cafe_debug_buf); +} + +static struct file_operations cafe_dfs_reg_ops = { + .owner = THIS_MODULE, + .read = cafe_dfs_read_regs, + .open = cafe_dfs_open +}; + +static ssize_t cafe_dfs_read_cam(struct file *file, + char __user *buf, size_t count, loff_t *ppos) +{ + struct cafe_camera *cam = file->private_data; + char *s = cafe_debug_buf; + int offset; + + if (! cam->sensor) + return -EINVAL; + for (offset = 0x0; offset < 0x8a; offset++) + { + u8 v; + + cafe_smbus_read_data(cam, cam->sensor->addr, offset, &v); + s += sprintf(s, "%02x: %02x\n", offset, v); + } + return simple_read_from_buffer(buf, count, ppos, cafe_debug_buf, + s - cafe_debug_buf); +} + +static struct file_operations cafe_dfs_cam_ops = { + .owner = THIS_MODULE, + .read = cafe_dfs_read_cam, + .open = cafe_dfs_open +}; + + + +static void cafe_dfs_cam_setup(struct cafe_camera *cam) +{ + char fname[40]; + + if (!cafe_dfs_root) + return; + sprintf(fname, "regs-%d", cam->v4ldev.minor); + cam->dfs_regs = debugfs_create_file(fname, 0444, cafe_dfs_root, + cam, &cafe_dfs_reg_ops); + sprintf(fname, "cam-%d", cam->v4ldev.minor); + cam->dfs_cam_regs = debugfs_create_file(fname, 0444, cafe_dfs_root, + cam, &cafe_dfs_cam_ops); +} + + +static void cafe_dfs_cam_shutdown(struct cafe_camera *cam) +{ + if (! IS_ERR(cam->dfs_regs)) + debugfs_remove(cam->dfs_regs); + if (! IS_ERR(cam->dfs_cam_regs)) + debugfs_remove(cam->dfs_cam_regs); +} + +#else + +#define cafe_dfs_setup() +#define cafe_dfs_shutdown() +#define cafe_dfs_cam_setup(cam) +#define cafe_dfs_cam_shutdown(cam) +#endif /* CONFIG_VIDEO_ADV_DEBUG */ + + + + +/* ------------------------------------------------------------------------*/ +/* + * PCI interface stuff. + */ + +static int cafe_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + int ret; + u16 classword; + struct cafe_camera *cam; + /* + * Make sure we have a camera here - we'll get calls for + * the other cafe devices as well. + */ + pci_read_config_word(pdev, PCI_CLASS_DEVICE, &classword); + if (classword != PCI_CLASS_MULTIMEDIA_VIDEO) + return -ENODEV; + /* + * Start putting together one of our big camera structures. + */ + ret = -ENOMEM; + cam = kzalloc(sizeof(struct cafe_camera), GFP_KERNEL); + if (cam == NULL) + goto out; + mutex_init(&cam->s_mutex); + mutex_lock(&cam->s_mutex); + spin_lock_init(&cam->dev_lock); + cam->state = S_NOTREADY; + cafe_set_config_needed(cam, 1); + init_waitqueue_head(&cam->smbus_wait); + init_waitqueue_head(&cam->iowait); + cam->pdev = pdev; + cam->pix_format = cafe_def_pix_format; + INIT_LIST_HEAD(&cam->dev_list); + INIT_LIST_HEAD(&cam->sb_avail); + INIT_LIST_HEAD(&cam->sb_full); + tasklet_init(&cam->s_tasklet, cafe_frame_tasklet, (unsigned long) cam); + /* + * Get set up on the PCI bus. + */ + ret = pci_enable_device(pdev); + if (ret) + goto out_free; + pci_set_master(pdev); + + ret = -EIO; + cam->regs = pci_iomap(pdev, 0, 0); + if (! cam->regs) { + printk(KERN_ERR "Unable to ioremap cafe-ccic regs\n"); + goto out_free; + } + ret = request_irq(pdev->irq, cafe_irq, IRQF_SHARED, "cafe-ccic", cam); + if (ret) + goto out_iounmap; + cafe_ctlr_init(cam); + cafe_ctlr_power_up(cam); + /* + * Set up I2C/SMBUS communications + */ + mutex_unlock(&cam->s_mutex); /* attach can deadlock */ + ret = cafe_smbus_setup(cam); + if (ret) + goto out_freeirq; + /* + * Get the v4l2 setup done. + */ + mutex_lock(&cam->s_mutex); + cam->v4ldev = cafe_v4l_template; + cam->v4ldev.debug = 0; +// cam->v4ldev.debug = V4L2_DEBUG_IOCTL_ARG; + ret = video_register_device(&cam->v4ldev, VFL_TYPE_GRABBER, -1); + if (ret) + goto out_smbus; + /* + * If so requested, try to get our DMA buffers now. + */ + if (alloc_bufs_at_load) { + if (cafe_alloc_dma_bufs(cam, 1)) + cam_warn(cam, "Unable to alloc DMA buffers at load" + " will try again later."); + } + + cafe_dfs_cam_setup(cam); + mutex_unlock(&cam->s_mutex); + cafe_add_dev(cam); + return 0; + + out_smbus: + cafe_smbus_shutdown(cam); + out_freeirq: + cafe_ctlr_power_down(cam); + free_irq(pdev->irq, cam); + out_iounmap: + pci_iounmap(pdev, cam->regs); + out_free: + kfree(cam); + out: + return ret; +} + + +/* + * Shut down an initialized device + */ +static void cafe_shutdown(struct cafe_camera *cam) +{ +/* FIXME: Make sure we take care of everything here */ + cafe_dfs_cam_shutdown(cam); + if (cam->n_sbufs > 0) + /* What if they are still mapped? Shouldn't be, but... */ + cafe_free_sio_buffers(cam); + cafe_remove_dev(cam); + cafe_ctlr_stop_dma(cam); + cafe_ctlr_power_down(cam); + cafe_smbus_shutdown(cam); + cafe_free_dma_bufs(cam); + free_irq(cam->pdev->irq, cam); + pci_iounmap(cam->pdev, cam->regs); + video_unregister_device(&cam->v4ldev); + /* kfree(cam); done in v4l_release () */ +} + + +static void cafe_pci_remove(struct pci_dev *pdev) +{ + struct cafe_camera *cam = cafe_find_by_pdev(pdev); + + if (cam == NULL) { + cam_warn(cam, "pci_remove on unknown pdev %p\n", pdev); + return; + } + mutex_lock(&cam->s_mutex); + if (cam->users > 0) + cam_warn(cam, "Removing a device with users!\n"); + cafe_shutdown(cam); +/* No unlock - it no longer exists */ +} + + + + +static struct pci_device_id cafe_ids[] = { + { PCI_DEVICE(0x1148, 0x4340) }, /* Temporary ID on devel board */ + { PCI_DEVICE(0x11ab, 0x4100) }, /* Eventual real ID */ + { PCI_DEVICE(0x11ab, 0x4102) }, /* Really eventual real ID */ + { 0, } +}; + +MODULE_DEVICE_TABLE(pci, cafe_ids); + +static struct pci_driver cafe_pci_driver = { + .name = "cafe1000-ccic", + .id_table = cafe_ids, + .probe = cafe_pci_probe, + .remove = cafe_pci_remove, +}; + + + + +static int __init cafe_init(void) +{ + int ret; + + printk(KERN_NOTICE "Marvell M88ALP01 'CAFE' Camera Controller version %d\n", + CAFE_VERSION); + cafe_dfs_setup(); + ret = pci_register_driver(&cafe_pci_driver); + if (ret) { + printk(KERN_ERR "Unable to register cafe_ccic driver\n"); + goto out; + } + request_module("ov7670"); /* FIXME want something more general */ + ret = 0; + + out: + return ret; +} + + +static void __exit cafe_exit(void) +{ + pci_unregister_driver(&cafe_pci_driver); + cafe_dfs_shutdown(); +} + +module_init(cafe_init); +module_exit(cafe_exit); diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 0a8f750cbede..a32db163cbc0 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -250,6 +250,7 @@ #define I2C_HW_SMBUS_OV518 0x04000f /* OV518(+) USB 1.1 webcam ICs */ #define I2C_HW_SMBUS_OV519 0x040010 /* OV519 USB 1.1 webcam IC */ #define I2C_HW_SMBUS_OVFX2 0x040011 /* Cypress/OmniVision FX2 webcam */ +#define I2C_HW_SMBUS_CAFE 0x040012 /* Marvell 88ALP01 "CAFE" cam */ /* --- ISA pseudo-adapter */ #define I2C_HW_ISA 0x050000 -- cgit v1.2.3 From 111f33564e19b2b5f70e3df9a8f92c08c1c91fd9 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Sat, 4 Nov 2006 09:26:00 -0300 Subject: V4L/DVB (4798): OmniVision OV7670 driver This patch adds a V4L2 driver for the OmniVision OV7670 camera. Signed-off-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/Kconfig | 8 + drivers/media/video/Makefile | 1 + drivers/media/video/ov7670.c | 1002 ++++++++++++++++++++++++++++++++++++++++++ include/linux/i2c-id.h | 1 + include/media/v4l2-common.h | 3 + 5 files changed, 1015 insertions(+) create mode 100644 drivers/media/video/ov7670.c (limited to 'include/linux') diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index 4ea1d0ebf5fd..c95735cd688c 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -184,6 +184,14 @@ config VIDEO_KS0127 To compile this driver as a module, choose M here: the module will be called ks0127. +config VIDEO_OV7670 + tristate "OmniVision OV7670 sensor support" + depends on I2C && VIDEO_V4L2 + ---help--- + This is a Video4Linux2 sensor-level driver for the OmniVision + OV7670 VGA camera. It currently only works with the M88ALP01 + controller. + config VIDEO_SAA7110 tristate "Philips SAA7110 video decoder" depends on VIDEO_V4L1 && I2C diff --git a/drivers/media/video/Makefile b/drivers/media/video/Makefile index 8ff787a4cf6a..2adb56d01ef6 100644 --- a/drivers/media/video/Makefile +++ b/drivers/media/video/Makefile @@ -93,6 +93,7 @@ obj-$(CONFIG_VIDEO_UPD64083) += upd64083.o obj-$(CONFIG_VIDEO_CX2341X) += cx2341x.o obj-$(CONFIG_VIDEO_CAFE_CCIC) += cafe_ccic.o +obj-$(CONFIG_VIDEO_OV7670) += ov7670.o obj-$(CONFIG_USB_DABUSB) += dabusb.o obj-$(CONFIG_USB_OV511) += ov511.o diff --git a/drivers/media/video/ov7670.c b/drivers/media/video/ov7670.c new file mode 100644 index 000000000000..382aa83b73c5 --- /dev/null +++ b/drivers/media/video/ov7670.c @@ -0,0 +1,1002 @@ +/* + * A V4L2 driver for OmniVision OV7670 cameras. + * + * Copyright 2006 One Laptop Per Child Association, Inc. Written + * by Jonathan Corbet with substantial inspiration from Mark + * McClelland's ovcamchip code. + * + * This file may be distributed under the terms of the GNU General + * Public License, version 2. + */ +#include +#include +#include +#include +#include +#include +#include +#include + + +MODULE_AUTHOR("Jonathan Corbet 0) + *value = (unsigned char) ret; + return ret; +} + + +static int ov7670_write(struct i2c_client *c, unsigned char reg, + unsigned char value) +{ + return i2c_smbus_write_byte_data(c, reg, value); +} + + +/* + * Write a list of register settings; ff/ff stops the process. + */ +static int ov7670_write_array(struct i2c_client *c, struct regval_list *vals) +{ + while (vals->reg_num != 0xff || vals->value != 0xff) { + int ret = ov7670_write(c, vals->reg_num, vals->value); + if (ret < 0) + return ret; + vals++; + } + return 0; +} + + +/* + * Stuff that knows about the sensor. + */ +static void ov7670_reset(struct i2c_client *client) +{ + ov7670_write(client, REG_COM7, COM7_RESET); + msleep(1); +} + + +static int ov7670_init(struct i2c_client *client) +{ + return ov7670_write_array(client, ov7670_default_regs); +} + + + +static int ov7670_detect(struct i2c_client *client) +{ + unsigned char v; + int ret; + + ret = ov7670_init(client); + if (ret < 0) + return ret; + ret = ov7670_read(client, REG_MIDH, &v); + if (ret < 0) + return ret; + if (v != 0x7f) /* OV manuf. id. */ + return -ENODEV; + ret = ov7670_read(client, REG_MIDL, &v); + if (ret < 0) + return ret; + if (v != 0xa2) + return -ENODEV; + /* + * OK, we know we have an OmniVision chip...but which one? + */ + ret = ov7670_read(client, REG_PID, &v); + if (ret < 0) + return ret; + if (v != 0x76) /* PID + VER = 0x76 / 0x73 */ + return -ENODEV; + ret = ov7670_read(client, REG_VER, &v); + if (ret < 0) + return ret; + if (v != 0x73) /* PID + VER = 0x76 / 0x73 */ + return -ENODEV; + return 0; +} + + + + + +static struct ov7670_format_struct { + __u8 *desc; + __u32 pixelformat; + struct regval_list *regs; +} ov7670_formats[] = { + { + .desc = "YUYV 4:2:2", + .pixelformat = V4L2_PIX_FMT_YUYV, + .regs = ov7670_fmt_yuv422, + }, + { + .desc = "RGB 444", + .pixelformat = V4L2_PIX_FMT_RGB444, + .regs = ov7670_fmt_rgb444, + }, + { + .desc = "RGB 565", + .pixelformat = V4L2_PIX_FMT_RGB565, + .regs = ov7670_fmt_rgb565, + }, + /* + * Pretend we do RGB32. This is here on the assumption that the + * upper layer will reformat RGB444 appropriately. + * + * The entire purpose for this thing's existence is to enable easy + * display of RGB444 for debugging purposes. It will come out soon. + */ + { + .desc = "RGB32 (faked)", + .pixelformat = V4L2_PIX_FMT_RGB32, + .regs = ov7670_fmt_rgb444, + }, +}; +#define N_OV7670_FMTS (sizeof(ov7670_formats)/sizeof(ov7670_formats[0])) + +/* + * All formats we support are 2 bytes/pixel. + */ +#define BYTES_PER_PIXEL 2 + +/* + * Then there is the issue of window sizes. Try to capture the info here. + */ +static struct ov7670_win_size { + int width; + int height; + unsigned char com7_bit; + int hstart; /* Start/stop values for the camera. Note */ + int hstop; /* that they do not always make complete */ + int vstart; /* sense to humans, but evidently the sensor */ + int vstop; /* will do the right thing... */ +/* h/vref stuff */ +} ov7670_win_sizes[] = { + /* VGA */ + { + .width = VGA_WIDTH, + .height = VGA_HEIGHT, + .com7_bit = COM7_FMT_VGA, + .hstart = 158, /* These values from */ + .hstop = 14, /* Omnivision */ + .vstart = 10, + .vstop = 490, + }, + /* CIF */ + { + .width = CIF_WIDTH, + .height = CIF_HEIGHT, + .com7_bit = COM7_FMT_CIF, + .hstart = 170, /* Empirically determined */ + .hstop = 90, + .vstart = 14, + .vstop = 494, + }, + /* QVGA */ + { + .width = QVGA_WIDTH, + .height = QVGA_HEIGHT, + .com7_bit = COM7_FMT_QVGA, + .hstart = 164, /* Empirically determined */ + .hstop = 20, + .vstart = 14, + .vstop = 494, + }, +}; + +#define N_WIN_SIZES (sizeof(ov7670_win_sizes)/sizeof(ov7670_win_sizes[0])) + + +/* + * Store a set of start/stop values into the camera. + */ +static int ov7670_set_hw(struct i2c_client *client, int hstart, int hstop, + int vstart, int vstop) +{ + int ret; + unsigned char v; +/* + * Horizontal: 11 bits, top 8 live in hstart and hstop. Bottom 3 of + * hstart are in href[2:0], bottom 3 of hstop in href[5:3]. There is + * a mystery "edge offset" value in the top two bits of href. + */ + ret = ov7670_write(client, REG_HSTART, (hstart >> 3) & 0xff); + ret += ov7670_write(client, REG_HSTOP, (hstop >> 3) & 0xff); + ret += ov7670_read(client, REG_HREF, &v); + v = (v & 0xc0) | ((hstop & 0x7) << 3) | (hstart & 0x7); + msleep(10); + ret += ov7670_write(client, REG_HREF, v); +/* + * Vertical: similar arrangement, but only 10 bits. + */ + ret += ov7670_write(client, REG_VSTART, (vstart >> 2) & 0xff); + ret += ov7670_write(client, REG_VSTOP, (vstop >> 2) & 0xff); + ret += ov7670_read(client, REG_VREF, &v); + v = (v & 0xf0) | ((vstop & 0x3) << 2) | (vstart & 0x3); + msleep(10); + ret += ov7670_write(client, REG_VREF, v); + return ret; +} + + +static int ov7670_enum_fmt(struct i2c_client *c, struct v4l2_fmtdesc *fmt) +{ + struct ov7670_format_struct *ofmt; + + if (fmt->index >= N_OV7670_FMTS) + return -EINVAL; + + ofmt = ov7670_formats + fmt->index; + fmt->flags = 0; + strcpy(fmt->description, ofmt->desc); + fmt->pixelformat = ofmt->pixelformat; + return 0; +} + + +static int ov7670_try_fmt(struct i2c_client *c, struct v4l2_format *fmt, + struct ov7670_format_struct **ret_fmt, + struct ov7670_win_size **ret_wsize) +{ + int index; + struct ov7670_win_size *wsize; + struct v4l2_pix_format *pix = &fmt->fmt.pix; + + for (index = 0; index < N_OV7670_FMTS; index++) + if (ov7670_formats[index].pixelformat == pix->pixelformat) + break; + if (index >= N_OV7670_FMTS) + return -EINVAL; + if (ret_fmt != NULL) + *ret_fmt = ov7670_formats + index; + /* + * Fields: the OV devices claim to be progressive. + */ + if (pix->field == V4L2_FIELD_ANY) + pix->field = V4L2_FIELD_NONE; + else if (pix->field != V4L2_FIELD_NONE) + return -EINVAL; + /* + * Round requested image size down to the nearest + * we support, but not below the smallest. + */ + for (wsize = ov7670_win_sizes; wsize < ov7670_win_sizes + N_WIN_SIZES; + wsize++) + if (pix->width >= wsize->width && pix->height >= wsize->height) + break; + if (wsize > ov7670_win_sizes + N_WIN_SIZES) + wsize--; /* Take the smallest one */ + if (ret_wsize != NULL) + *ret_wsize = wsize; + /* + * Note the size we'll actually handle. + */ + pix->width = wsize->width; + pix->height = wsize->height; + pix->bytesperline = pix->width*BYTES_PER_PIXEL; + if (pix->pixelformat == V4L2_PIX_FMT_RGB32) + pix->bytesperline *= 2; + pix->sizeimage = pix->height*pix->bytesperline; + return 0; + +} + +/* + * Set a format. + */ +static int ov7670_s_fmt(struct i2c_client *c, struct v4l2_format *fmt) +{ + int ret; + struct ov7670_format_struct *ovfmt; + struct ov7670_win_size *wsize; + unsigned char com7; + + ret = ov7670_try_fmt(c, fmt, &ovfmt, &wsize); + if (ret) + return ret; + /* + * COM7 is a pain in the ass, it doesn't like to be read then + * quickly written afterward. But we have everything we need + * to set it absolutely here, as long as the format-specific + * register sets list it first. + */ + com7 = ovfmt->regs[0].value; + com7 |= wsize->com7_bit; + ov7670_write(c, REG_COM7, com7); + /* + * Now write the rest of the array. Also store start/stops + */ + ov7670_write_array(c, ovfmt->regs + 1); + ov7670_set_hw(c, wsize->hstart, wsize->hstop, wsize->vstart, + wsize->vstop); + return 0; +} + +/* + * Code for dealing with controls. + */ + +/* + * Some weird registers seem to store values in a sign/magnitude format! + */ +static unsigned char ov7670_sm_to_abs(unsigned char v) +{ + if ((v & 0x80) == 0) + return v + 128; + else + return 128 - (v & 0x7f); +} + + +static unsigned char ov7670_abs_to_sm(unsigned char v) +{ + if (v > 127) + return v & 0x7f; + else + return (128 - v) | 0x80; +} + +static int ov7670_t_brightness(struct i2c_client *client, unsigned char value) +{ + unsigned char com8; + int ret; + + ov7670_read(client, REG_COM8, &com8); + com8 &= ~COM8_AEC; + ov7670_write(client, REG_COM8, com8); + value = ov7670_abs_to_sm(value); + ret = ov7670_write(client, REG_BRIGHT, value); + return ret; +} + +static int ov7670_q_brightness(struct i2c_client *client, unsigned char *value) +{ + int ret; + ret = ov7670_read(client, REG_BRIGHT, value); + *value = ov7670_sm_to_abs(*value); + return ret; +} + +static int ov7670_t_contrast(struct i2c_client *client, unsigned char value) +{ + return ov7670_write(client, REG_CONTRAS, value); +} + +static int ov7670_q_contrast(struct i2c_client *client, unsigned char *value) +{ + return ov7670_read(client, REG_CONTRAS, value); +} + +static int ov7670_q_hflip(struct i2c_client *client, unsigned char *value) +{ + int ret; + unsigned char v; + + ret = ov7670_read(client, REG_MVFP, &v); + *value = (v & MVFP_MIRROR) == MVFP_MIRROR; + return ret; +} + + +static int ov7670_t_hflip(struct i2c_client *client, unsigned char value) +{ + unsigned char v; + int ret; + + ret = ov7670_read(client, REG_MVFP, &v); + if (value) + v |= MVFP_MIRROR; + else + v &= ~MVFP_MIRROR; + msleep(10); /* FIXME */ + ret += ov7670_write(client, REG_MVFP, v); + return ret; +} + + + +static int ov7670_q_vflip(struct i2c_client *client, unsigned char *value) +{ + int ret; + unsigned char v; + + ret = ov7670_read(client, REG_MVFP, &v); + *value = (v & MVFP_FLIP) == MVFP_FLIP; + return ret; +} + + +static int ov7670_t_vflip(struct i2c_client *client, unsigned char value) +{ + unsigned char v; + int ret; + + ret = ov7670_read(client, REG_MVFP, &v); + if (value) + v |= MVFP_FLIP; + else + v &= ~MVFP_FLIP; + msleep(10); /* FIXME */ + ret += ov7670_write(client, REG_MVFP, v); + return ret; +} + + +static struct ov7670_control { + struct v4l2_queryctrl qc; + int (*query)(struct i2c_client *c, unsigned char *value); + int (*tweak)(struct i2c_client *c, unsigned char value); +} ov7670_controls[] = +{ + { + .qc = { + .id = V4L2_CID_BRIGHTNESS, + .type = V4L2_CTRL_TYPE_INTEGER, + .name = "Brightness", + .minimum = 0, + .maximum = 255, + .step = 1, + .default_value = 0x80, + .flags = V4L2_CTRL_FLAG_SLIDER + }, + .tweak = ov7670_t_brightness, + .query = ov7670_q_brightness, + }, + { + .qc = { + .id = V4L2_CID_CONTRAST, + .type = V4L2_CTRL_TYPE_INTEGER, + .name = "Contrast", + .minimum = 0, + .maximum = 127, + .step = 1, + .default_value = 0x40, /* XXX ov7670 spec */ + .flags = V4L2_CTRL_FLAG_SLIDER + }, + .tweak = ov7670_t_contrast, + .query = ov7670_q_contrast, + }, + { + .qc = { + .id = V4L2_CID_VFLIP, + .type = V4L2_CTRL_TYPE_BOOLEAN, + .name = "Vertical flip", + .minimum = 0, + .maximum = 1, + .step = 1, + .default_value = 0, + }, + .tweak = ov7670_t_vflip, + .query = ov7670_q_vflip, + }, + { + .qc = { + .id = V4L2_CID_HFLIP, + .type = V4L2_CTRL_TYPE_BOOLEAN, + .name = "Horizontal mirror", + .minimum = 0, + .maximum = 1, + .step = 1, + .default_value = 0, + }, + .tweak = ov7670_t_hflip, + .query = ov7670_q_hflip, + }, +}; +#define N_CONTROLS (sizeof(ov7670_controls)/sizeof(ov7670_controls[0])) + +static struct ov7670_control *ov7670_find_control(__u32 id) +{ + int i; + + for (i = 0; i < N_CONTROLS; i++) + if (ov7670_controls[i].qc.id == id) + return ov7670_controls + i; + return NULL; +} + + +static int ov7670_queryctrl(struct i2c_client *client, + struct v4l2_queryctrl *qc) +{ + struct ov7670_control *ctrl = ov7670_find_control(qc->id); + + if (ctrl == NULL) + return -EINVAL; + *qc = ctrl->qc; + return 0; +} + +static int ov7670_g_ctrl(struct i2c_client *client, struct v4l2_control *ctrl) +{ + struct ov7670_control *octrl = ov7670_find_control(ctrl->id); + int ret; + unsigned char v; + + if (octrl == NULL) + return -EINVAL; + ret = octrl->query(client, &v); + if (ret >= 0) { + ctrl->value = v; + return 0; + } + return ret; +} + +static int ov7670_s_ctrl(struct i2c_client *client, struct v4l2_control *ctrl) +{ + struct ov7670_control *octrl = ov7670_find_control(ctrl->id); + + if (octrl == NULL) + return -EINVAL; + return octrl->tweak(client, ctrl->value); +} + + + + + + + +/* + * Basic i2c stuff. + */ +static struct i2c_driver ov7670_driver; + +static int ov7670_attach(struct i2c_adapter *adapter) +{ + int ret; + struct i2c_client *client; + + printk(KERN_ERR "ov7670 attach, id = %d\n", adapter->id); + /* + * For now: only deal with adapters we recognize. + */ + if (adapter->id != I2C_HW_SMBUS_CAFE) + return -ENODEV; + + printk(KERN_ERR "ov7670 accepting\n"); + client = kzalloc(sizeof (struct i2c_client), GFP_KERNEL); + if (! client) + return -ENOMEM; + client->adapter = adapter; + client->addr = OV7670_I2C_ADDR; + client->driver = &ov7670_driver, + strcpy(client->name, "OV7670"); + /* Do we need clientdata? */ + + /* + * Make sure it's an ov7670 + */ + ret = ov7670_detect(client); + printk(KERN_ERR "detect result is %d\n", ret); + if (ret) + goto out_free; + i2c_attach_client(client); + return 0; + + out_free: + kfree(client); + return ret; +} + + +static int ov7670_detach(struct i2c_client *client) +{ + i2c_detach_client(client); + kfree(client); + return 0; +} + + +static int ov7670_command(struct i2c_client *client, unsigned int cmd, + void *arg) +{ + switch (cmd) { + case VIDIOC_INT_G_CHIP_IDENT: + * (enum v4l2_chip_ident *) arg = V4L2_IDENT_OV7670; + return 0; + + case VIDIOC_INT_RESET: + ov7670_reset(client); + return 0; + + case VIDIOC_INT_INIT: + return ov7670_init(client); + + case VIDIOC_ENUM_FMT: + return ov7670_enum_fmt(client, (struct v4l2_fmtdesc *) arg); + case VIDIOC_TRY_FMT: + return ov7670_try_fmt(client, (struct v4l2_format *) arg, NULL, NULL); + case VIDIOC_S_FMT: + return ov7670_s_fmt(client, (struct v4l2_format *) arg); + case VIDIOC_QUERYCTRL: + return ov7670_queryctrl(client, (struct v4l2_queryctrl *) arg); + case VIDIOC_S_CTRL: + return ov7670_s_ctrl(client, (struct v4l2_control *) arg); + case VIDIOC_G_CTRL: + return ov7670_g_ctrl(client, (struct v4l2_control *) arg); + /* Todo: + g/s_parm + initialization + */ + } + return -EINVAL; +} + + + +static struct i2c_driver ov7670_driver = { + .driver = { + .name = "ov7670", + }, + .id = I2C_DRIVERID_OV7670, + .class = I2C_CLASS_CAM_DIGITAL, + .attach_adapter = ov7670_attach, + .detach_client = ov7670_detach, + .command = ov7670_command, +}; + + +/* + * Module initialization + */ +static int __init ov7670_mod_init(void) +{ + printk(KERN_NOTICE "OmniVision ov7670 sensor driver, at your service\n"); + return i2c_add_driver(&ov7670_driver); +} + +static void __exit ov7670_mod_exit(void) +{ + i2c_del_driver(&ov7670_driver); +} + +module_init(ov7670_mod_init); +module_exit(ov7670_mod_exit); diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index a32db163cbc0..0f4cf34b6fa2 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -159,6 +159,7 @@ #define I2C_DRIVERID_ASB100 1043 #define I2C_DRIVERID_FSCHER 1046 #define I2C_DRIVERID_W83L785TS 1047 +#define I2C_DRIVERID_OV7670 1048 /* Omnivision 7670 camera */ /* * ---- Adapter types ---------------------------------------------------- diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h index 8263ea00ca6f..91b19921f958 100644 --- a/include/media/v4l2-common.h +++ b/include/media/v4l2-common.h @@ -144,6 +144,9 @@ enum v4l2_chip_ident { V4L2_IDENT_CX25841 = 241, V4L2_IDENT_CX25842 = 242, V4L2_IDENT_CX25843 = 243, + + /* OmniVision sensors - range 250-299 */ + V4L2_IDENT_OV7670 = 250, }; /* audio ioctls */ -- cgit v1.2.3 From 5c543eff6cc658f46241f1ccb77436d65abbf445 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 10 Dec 2006 02:18:58 -0800 Subject: [PATCH] freezer.h uses task_struct fields freezer.h uses task_struct fields so it should include sched.h. CC [M] fs/jfs/jfs_txnmgr.o In file included from fs/jfs/jfs_txnmgr.c:49: include/linux/freezer.h: In function 'frozen': include/linux/freezer.h:9: error: dereferencing pointer to incomplete type include/linux/freezer.h:9: error: 'PF_FROZEN' undeclared (first use in this function) include/linux/freezer.h:9: error: (Each undeclared identifier is reported only once include/linux/freezer.h:9: error: for each function it appears in.) include/linux/freezer.h: In function 'freezing': include/linux/freezer.h:17: error: dereferencing pointer to incomplete type include/linux/freezer.h:17: error: 'PF_FREEZE' undeclared (first use in this function) include/linux/freezer.h: In function 'freeze': include/linux/freezer.h:26: error: dereferencing pointer to incomplete type include/linux/freezer.h:26: error: 'PF_FREEZE' undeclared (first use in this function) include/linux/freezer.h: In function 'do_not_freeze': include/linux/freezer.h:34: error: dereferencing pointer to incomplete type include/linux/freezer.h:34: error: 'PF_FREEZE' undeclared (first use in this function) include/linux/freezer.h: In function 'thaw_process': include/linux/freezer.h:43: error: dereferencing pointer to incomplete type include/linux/freezer.h:43: error: 'PF_FROZEN' undeclared (first use in this function) include/linux/freezer.h:44: warning: implicit declaration of function 'wake_up_process' include/linux/freezer.h: In function 'frozen_process': include/linux/freezer.h:55: error: dereferencing pointer to incomplete type include/linux/freezer.h:55: error: dereferencing pointer to incomplete type include/linux/freezer.h:55: error: 'PF_FREEZE' undeclared (first use in this function) include/linux/freezer.h:55: error: 'PF_FROZEN' undeclared (first use in this function) fs/jfs/jfs_txnmgr.c: In function 'freezing': include/linux/freezer.h:18: warning: control reaches end of non-void function make[2]: *** [fs/jfs/jfs_txnmgr.o] Error 1 Signed-off-by: Randy Dunlap Acked-by: Dave Kleikamp Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/freezer.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 6e05e3e7ce39..393063096134 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -1,5 +1,7 @@ /* Freezer declarations */ +#include + #ifdef CONFIG_PM /* * Check if a process has been frozen -- cgit v1.2.3 From 884b4aaaa242a2db8c8252796f0118164a680ab5 Mon Sep 17 00:00:00 2001 From: Scott Wood Date: Sun, 10 Dec 2006 02:19:00 -0800 Subject: [PATCH] rtc: Add rtc_merge_alarm() Add rtc_merge_alarm(), which can be used by rtc drivers to turn a partially specified alarm expiry (i.e. most significant fields set to -1, as with the RTC_ALM_SET ioctl()) into a fully specified expiry. If the most significant specified field is earlier than the current time, the least significant unspecified field is incremented. Signed-off-by: Scott Wood Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-lib.c | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/rtc.h | 1 + 2 files changed, 82 insertions(+) (limited to 'include/linux') diff --git a/drivers/rtc/rtc-lib.c b/drivers/rtc/rtc-lib.c index ba795a4db1e9..7bbc26a34bd2 100644 --- a/drivers/rtc/rtc-lib.c +++ b/drivers/rtc/rtc-lib.c @@ -117,4 +117,85 @@ int rtc_tm_to_time(struct rtc_time *tm, unsigned long *time) } EXPORT_SYMBOL(rtc_tm_to_time); + +/* Merge the valid (i.e. non-negative) fields of alarm into the current + * time. If the valid alarm fields are earlier than the equivalent + * fields in the time, carry one into the least significant invalid + * field, so that the alarm expiry is in the future. It assumes that the + * least significant invalid field is more significant than the most + * significant valid field, and that the seconds field is valid. + * + * This is used by alarms that take relative (rather than absolute) + * times, and/or have a simple binary second counter instead of + * day/hour/minute/sec registers. + */ +void rtc_merge_alarm(struct rtc_time *now, struct rtc_time *alarm) +{ + int *alarmp = &alarm->tm_sec; + int *timep = &now->tm_sec; + int carry_into, i; + + /* Ignore everything past the 6th element (tm_year). */ + for (i = 5; i > 0; i--) { + if (alarmp[i] < 0) + alarmp[i] = timep[i]; + else + break; + } + + /* No carry needed if all fields are valid. */ + if (i == 5) + return; + + for (carry_into = i + 1; i >= 0; i--) { + if (alarmp[i] < timep[i]) + break; + + if (alarmp[i] > timep[i]) + return; + } + + switch (carry_into) { + case 1: + alarm->tm_min++; + + if (alarm->tm_min < 60) + return; + + alarm->tm_min = 0; + /* fall-through */ + + case 2: + alarm->tm_hour++; + + if (alarm->tm_hour < 60) + return; + + alarm->tm_hour = 0; + /* fall-through */ + + case 3: + alarm->tm_mday++; + + if (alarm->tm_mday <= rtc_days_in_month[alarm->tm_mon]) + return; + + alarm->tm_mday = 1; + /* fall-through */ + + case 4: + alarm->tm_mon++; + + if (alarm->tm_mon <= 12) + return; + + alarm->tm_mon = 1; + /* fall-through */ + + case 5: + alarm->tm_year++; + } +} +EXPORT_SYMBOL(rtc_merge_alarm); + MODULE_LICENSE("GPL"); diff --git a/include/linux/rtc.h b/include/linux/rtc.h index 09ff4c3e2713..5e22d4510d11 100644 --- a/include/linux/rtc.h +++ b/include/linux/rtc.h @@ -106,6 +106,7 @@ extern int rtc_year_days(unsigned int day, unsigned int month, unsigned int year extern int rtc_valid_tm(struct rtc_time *tm); extern int rtc_tm_to_time(struct rtc_time *tm, unsigned long *time); extern void rtc_time_to_tm(unsigned long time, struct rtc_time *tm); +extern void rtc_merge_alarm(struct rtc_time *now, struct rtc_time *alarm); #include #include -- cgit v1.2.3 From 1f29bcd739972f71f2fd5d5d265daf3e1208fa5e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sun, 10 Dec 2006 02:19:10 -0800 Subject: [PATCH] sysctl: remove unused "context" param Signed-off-by: Alexey Dobriyan Cc: Andi Kleen Cc: "David S. Miller" Cc: David Howells Cc: Ralf Baechle Cc: "Eric W. Biederman" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/pm.c | 6 +++--- arch/mips/lasat/sysctl.c | 17 +++++++-------- arch/x86_64/kernel/vsyscall.c | 3 +-- drivers/char/random.c | 2 +- include/linux/sysctl.h | 5 ++--- include/net/ip.h | 3 +-- include/net/sctp/sctp.h | 2 +- kernel/sysctl.c | 47 ++++++++++++++++++++---------------------- net/decnet/dn_dev.c | 6 ++---- net/decnet/sysctl_net_decnet.c | 6 ++---- net/ipv4/devinet.c | 3 +-- net/ipv4/route.c | 3 +-- net/ipv4/sysctl_net_ipv4.c | 16 ++++++-------- net/ipv6/addrconf.c | 3 +-- net/ipv6/ndisc.c | 9 +++----- 15 files changed, 55 insertions(+), 76 deletions(-) (limited to 'include/linux') diff --git a/arch/frv/kernel/pm.c b/arch/frv/kernel/pm.c index c1d9fc8f1a85..ee677ced7b68 100644 --- a/arch/frv/kernel/pm.c +++ b/arch/frv/kernel/pm.c @@ -223,7 +223,7 @@ static int cmode_procctl(ctl_table *ctl, int write, struct file *filp, static int cmode_sysctl(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (oldval && oldlenp) { size_t oldlen; @@ -326,7 +326,7 @@ static int p0_procctl(ctl_table *ctl, int write, struct file *filp, static int p0_sysctl(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (oldval && oldlenp) { size_t oldlen; @@ -370,7 +370,7 @@ static int cm_procctl(ctl_table *ctl, int write, struct file *filp, static int cm_sysctl(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (oldval && oldlenp) { size_t oldlen; diff --git a/arch/mips/lasat/sysctl.c b/arch/mips/lasat/sysctl.c index da35d4555491..12878359f2c8 100644 --- a/arch/mips/lasat/sysctl.c +++ b/arch/mips/lasat/sysctl.c @@ -40,12 +40,12 @@ static DEFINE_MUTEX(lasat_info_mutex); /* Strategy function to write EEPROM after changing string entry */ int sysctl_lasatstring(ctl_table *table, int *name, int nlen, void *oldval, size_t *oldlenp, - void *newval, size_t newlen, void **context) + void *newval, size_t newlen) { int r; mutex_lock(&lasat_info_mutex); r = sysctl_string(table, name, - nlen, oldval, oldlenp, newval, newlen, context); + nlen, oldval, oldlenp, newval, newlen); if (r < 0) { mutex_unlock(&lasat_info_mutex); return r; @@ -119,11 +119,11 @@ int proc_dolasatrtc(ctl_table *table, int write, struct file *filp, /* Sysctl for setting the IP addresses */ int sysctl_lasat_intvec(ctl_table *table, int *name, int nlen, void *oldval, size_t *oldlenp, - void *newval, size_t newlen, void **context) + void *newval, size_t newlen) { int r; mutex_lock(&lasat_info_mutex); - r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen, context); + r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen); if (r < 0) { mutex_unlock(&lasat_info_mutex); return r; @@ -139,14 +139,14 @@ int sysctl_lasat_intvec(ctl_table *table, int *name, int nlen, /* Same for RTC */ int sysctl_lasat_rtc(ctl_table *table, int *name, int nlen, void *oldval, size_t *oldlenp, - void *newval, size_t newlen, void **context) + void *newval, size_t newlen) { int r; mutex_lock(&lasat_info_mutex); rtctmp = ds1603_read(); if (rtctmp < 0) rtctmp = 0; - r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen, context); + r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen); if (r < 0) { mutex_unlock(&lasat_info_mutex); return r; @@ -251,13 +251,12 @@ int proc_lasat_ip(ctl_table *table, int write, struct file *filp, static int sysctl_lasat_eeprom_value(ctl_table *table, int *name, int nlen, void *oldval, size_t *oldlenp, - void *newval, size_t newlen, - void **context) + void *newval, size_t newlen) { int r; mutex_lock(&lasat_info_mutex); - r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen, context); + r = sysctl_intvec(table, name, nlen, oldval, oldlenp, newval, newlen); if (r < 0) { mutex_unlock(&lasat_info_mutex); return r; diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c index 4a673f5397a0..2433d6fc68b1 100644 --- a/arch/x86_64/kernel/vsyscall.c +++ b/arch/x86_64/kernel/vsyscall.c @@ -225,8 +225,7 @@ out: static int vsyscall_sysctl_nostrat(ctl_table *t, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } diff --git a/drivers/char/random.c b/drivers/char/random.c index 092a01cc02da..13d0b1350a62 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1203,7 +1203,7 @@ static int proc_do_uuid(ctl_table *table, int write, struct file *filp, static int uuid_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { unsigned char tmp_uuid[16], *uuid; unsigned int len; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 94316a98e0d0..6d8846e7be6d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -918,8 +918,7 @@ typedef struct ctl_table ctl_table; typedef int ctl_handler (ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context); + void __user *newval, size_t newlen); typedef int proc_handler (ctl_table *ctl, int write, struct file * filp, void __user *buffer, size_t *lenp, loff_t *ppos); @@ -950,7 +949,7 @@ extern int do_sysctl (int __user *name, int nlen, extern int do_sysctl_strategy (ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void ** context); + void __user *newval, size_t newlen); extern ctl_handler sysctl_string; extern ctl_handler sysctl_intvec; diff --git a/include/net/ip.h b/include/net/ip.h index 83cb9ac5554e..053f02b5cb89 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -376,8 +376,7 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write, size_t *lenp, loff_t *ppos); int ipv4_doint_and_flush_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context); + void __user *newval, size_t newlen); #ifdef CONFIG_PROC_FS extern int ip_misc_proc_init(void); #endif diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 215461f18db1..c818f87122af 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -368,7 +368,7 @@ static inline void sctp_sysctl_register(void) { return; } static inline void sctp_sysctl_unregister(void) { return; } static inline int sctp_sysctl_jiffies_ms(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) { + void __user *newval, size_t newlen) { return -ENOSYS; } #endif diff --git a/kernel/sysctl.c b/kernel/sysctl.c index c86445a62af2..130c5ec9ee0b 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -133,7 +133,7 @@ extern int max_lock_depth; #ifdef CONFIG_SYSCTL_SYSCALL static int parse_table(int __user *, int, void __user *, size_t __user *, - void __user *, size_t, ctl_table *, void **); + void __user *, size_t, ctl_table *); #endif static int proc_do_uts_string(ctl_table *table, int write, struct file *filp, @@ -141,12 +141,12 @@ static int proc_do_uts_string(ctl_table *table, int write, struct file *filp, static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context); + void __user *newval, size_t newlen); #ifdef CONFIG_SYSVIPC static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context); + void __user *newval, size_t newlen); #endif #ifdef CONFIG_PROC_SYSCTL @@ -1243,7 +1243,6 @@ int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *ol do { struct ctl_table_header *head = list_entry(tmp, struct ctl_table_header, ctl_entry); - void *context = NULL; if (!use_table(head)) continue; @@ -1251,9 +1250,7 @@ int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *ol spin_unlock(&sysctl_lock); error = parse_table(name, nlen, oldval, oldlenp, - newval, newlen, head->ctl_table, - &context); - kfree(context); + newval, newlen, head->ctl_table); spin_lock(&sysctl_lock); unuse_table(head); @@ -1309,7 +1306,7 @@ static inline int ctl_perm(ctl_table *table, int op) static int parse_table(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen, - ctl_table *table, void **context) + ctl_table *table) { int n; repeat: @@ -1329,7 +1326,7 @@ repeat: error = table->strategy( table, name, nlen, oldval, oldlenp, - newval, newlen, context); + newval, newlen); if (error) return error; } @@ -1340,7 +1337,7 @@ repeat: } error = do_sysctl_strategy(table, name, nlen, oldval, oldlenp, - newval, newlen, context); + newval, newlen); return error; } } @@ -1351,7 +1348,7 @@ repeat: int do_sysctl_strategy (ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { int op = 0, rc; size_t len; @@ -1365,7 +1362,7 @@ int do_sysctl_strategy (ctl_table *table, if (table->strategy) { rc = table->strategy(table, name, nlen, oldval, oldlenp, - newval, newlen, context); + newval, newlen); if (rc < 0) return rc; if (rc > 0) @@ -2473,7 +2470,7 @@ int proc_doulongvec_ms_jiffies_minmax(ctl_table *table, int write, /* The generic string strategy routine: */ int sysctl_string(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (!table->data || !table->maxlen) return -ENOTDIR; @@ -2519,7 +2516,7 @@ int sysctl_string(ctl_table *table, int __user *name, int nlen, */ int sysctl_intvec(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (newval && newlen) { @@ -2555,7 +2552,7 @@ int sysctl_intvec(ctl_table *table, int __user *name, int nlen, /* Strategy function to convert jiffies to seconds */ int sysctl_jiffies(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (oldval) { size_t olen; @@ -2583,7 +2580,7 @@ int sysctl_jiffies(ctl_table *table, int __user *name, int nlen, /* Strategy function to convert jiffies to seconds */ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { if (oldval) { size_t olen; @@ -2612,7 +2609,7 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, /* The generic string strategy routine: */ static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { struct ctl_table uts_table; int r, write; @@ -2620,7 +2617,7 @@ static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen, memcpy(&uts_table, table, sizeof(uts_table)); uts_table.data = get_uts(table, write); r = sysctl_string(&uts_table, name, nlen, - oldval, oldlenp, newval, newlen, context); + oldval, oldlenp, newval, newlen); put_uts(table, write, uts_table.data); return r; } @@ -2629,7 +2626,7 @@ static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen, /* The generic sysctl ipc data routine. */ static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { size_t len; void *data; @@ -2704,41 +2701,41 @@ out: int sysctl_string(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } int sysctl_intvec(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } int sysctl_jiffies(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, void **context) + void __user *newval, size_t newlen) { return -ENOSYS; } diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 0b9d4c955154..fc6f3c023a54 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -167,8 +167,7 @@ static int dn_forwarding_proc(ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); static int dn_forwarding_sysctl(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context); + void __user *newval, size_t newlen); static struct dn_dev_sysctl_table { struct ctl_table_header *sysctl_header; @@ -347,8 +346,7 @@ static int dn_forwarding_proc(ctl_table *table, int write, static int dn_forwarding_sysctl(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { #ifdef CONFIG_DECNET_ROUTER struct net_device *dev = table->extra1; diff --git a/net/decnet/sysctl_net_decnet.c b/net/decnet/sysctl_net_decnet.c index e246f054f368..a4065eb1341e 100644 --- a/net/decnet/sysctl_net_decnet.c +++ b/net/decnet/sysctl_net_decnet.c @@ -134,8 +134,7 @@ static int parse_addr(__le16 *addr, char *str) static int dn_node_address_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { size_t len; __le16 addr; @@ -220,8 +219,7 @@ static int dn_node_address_handler(ctl_table *table, int write, static int dn_def_dev_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { size_t len; struct net_device *dev; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 2fd899160f85..84bed40273ad 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1303,8 +1303,7 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write, int ipv4_doint_and_flush_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { int *valp = table->data; int new; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 11c167118e87..1aaff0a2e098 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2872,8 +2872,7 @@ static int ipv4_sysctl_rtcache_flush_strategy(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, - size_t newlen, - void **context) + size_t newlen) { int delay; if (newlen != sizeof(int)) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index dfcf47f10f88..fabf69a9108c 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -51,8 +51,7 @@ int ipv4_sysctl_forward(ctl_table *ctl, int write, struct file * filp, static int ipv4_sysctl_forward_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { int *valp = table->data; int new; @@ -111,8 +110,7 @@ static int proc_tcp_congestion_control(ctl_table *ctl, int write, struct file * static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { char val[TCP_CA_NAME_MAX]; ctl_table tbl = { @@ -122,8 +120,7 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int ret; tcp_get_default_congestion_control(val); - ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, - context); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen); if (ret == 0 && newval && newlen) ret = tcp_set_default_congestion_control(val); return ret; @@ -169,8 +166,8 @@ static int proc_allowed_congestion_control(ctl_table *ctl, static int strategy_allowed_congestion_control(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, + size_t newlen) { ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX }; int ret; @@ -180,8 +177,7 @@ static int strategy_allowed_congestion_control(ctl_table *table, int __user *nam return -ENOMEM; tcp_get_available_congestion_control(tbl.data, tbl.maxlen); - ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen, - context); + ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen); if (ret == 0 && newval && newlen) ret = tcp_set_allowed_congestion_control(tbl.data); kfree(tbl.data); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a5e8d207a51b..9b0a90643151 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3656,8 +3656,7 @@ static int addrconf_sysctl_forward_strategy(ctl_table *table, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { int *valp = table->data; int new; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 56ea92837307..6a9f616de37d 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1667,8 +1667,7 @@ int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, struct file * f static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, - void __user *newval, size_t newlen, - void **context) + void __user *newval, size_t newlen) { struct net_device *dev = ctl->extra1; struct inet6_dev *idev; @@ -1681,14 +1680,12 @@ static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, switch (ctl->ctl_name) { case NET_NEIGH_REACHABLE_TIME: ret = sysctl_jiffies(ctl, name, nlen, - oldval, oldlenp, newval, newlen, - context); + oldval, oldlenp, newval, newlen); break; case NET_NEIGH_RETRANS_TIME_MS: case NET_NEIGH_REACHABLE_TIME_MS: ret = sysctl_ms_jiffies(ctl, name, nlen, - oldval, oldlenp, newval, newlen, - context); + oldval, oldlenp, newval, newlen); break; default: ret = 0; -- cgit v1.2.3 From 58f64d83c37f5073a01573d27043c9c0ccc764f1 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Sun, 10 Dec 2006 02:19:11 -0800 Subject: [PATCH] Fix noise in futex.h There are some kernel-only bits in the middle of which should be removed in what we export to userspace. Signed-off-by: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/Kbuild | 2 +- include/linux/futex.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index e618b25b5add..a1b04d8a1d01 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -61,7 +61,6 @@ header-y += fd.h header-y += fdreg.h header-y += fib_rules.h header-y += fuse.h -header-y += futex.h header-y += genetlink.h header-y += gen_stats.h header-y += gigaset_dev.h @@ -203,6 +202,7 @@ unifdef-y += fb.h unifdef-y += fcntl.h unifdef-y += filter.h unifdef-y += flat.h +unifdef-y += futex.h unifdef-y += fs.h unifdef-y += gameport.h unifdef-y += generic_serial.h diff --git a/include/linux/futex.h b/include/linux/futex.h index d097b5b72bc6..3f153b4e156c 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -93,6 +93,7 @@ struct robust_list_head { */ #define ROBUST_LIST_LIMIT 2048 +#ifdef __KERNEL__ long do_futex(u32 __user *uaddr, int op, u32 val, unsigned long timeout, u32 __user *uaddr2, u32 val2, u32 val3); @@ -110,6 +111,7 @@ static inline void exit_pi_state_list(struct task_struct *curr) { } #endif +#endif /* __KERNEL__ */ #define FUTEX_OP_SET 0 /* *(int *)UADDR2 = OPARG; */ #define FUTEX_OP_ADD 1 /* *(int *)UADDR2 += OPARG; */ -- cgit v1.2.3 From 7c3ab7381e79dfc7db14a67c6f4f3285664e1ec2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 10 Dec 2006 02:19:19 -0800 Subject: [PATCH] io-accounting: core statistics The present per-task IO accounting isn't very useful. It simply counts the number of bytes passed into read() and write(). So if a process reads 1MB from an already-cached file, it is accused of having performed 1MB of I/O, which is wrong. (David Wright had some comments on the applicability of the present logical IO accounting: For billing purposes it is useless but for workload analysis it is very useful read_bytes/read_calls average read request size write_bytes/write_calls average write request size read_bytes/read_blocks ie logical/physical can indicate hit rate or thrashing write_bytes/write_blocks ie logical/physical guess since pdflush writes can be missed I often look for logical larger than physical to see filesystem cache problems. And the bytes/cpusec can help find applications that are dominating the cache and causing slow interactive response from page cache contention. I want to find the IO intensive applications and make sure they are doing efficient IO. Thus the acctcms(sysV) or csacms command would give the high IO commands). This patchset adds new accounting which tries to be more accurate. We account for three things: reads: attempt to count the number of bytes which this process really did cause to be fetched from the storage layer. Done at the submit_bio() level, so it is accurate for block-backed filesystems. I also attempt to wire up NFS and CIFS. writes: attempt to count the number of bytes which this process caused to be sent to the storage layer. This is done at page-dirtying time. The big inaccuracy here is truncate. If a process writes 1MB to a file and then deletes the file, it will in fact perform no writeout. But it will have been accounted as having caused 1MB of write. So... cancelled_writes: account the number of bytes which this process caused to not happen, by truncating pagecache. We _could_ just subtract this from the process's `write' accounting. But that means that some processes would be reported to have done negative amounts of write IO, which is silly. So we just report the raw number and punt this decision up to userspace. Now, we _could_ account for writes at the physical I/O level. But - This would require that we track memory-dirtying tasks at the per-page level (would require a new pointer in struct page). - It would mean that IO statistics for a process are usually only available long after that process has exitted. Which means that we probably cannot communicate this info via taskstats. This patch: Wire up the kernel-private data structures and the accessor functions to manipulate them. Cc: Jay Lan Cc: Shailabh Nagar Cc: Balbir Singh Cc: Chris Sturtivant Cc: Tony Ernst Cc: Guillaume Thouvenin Cc: David Wright Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 ++ include/linux/task_io_accounting.h | 37 ++++++++++++++++++++++++++ include/linux/task_io_accounting_ops.h | 47 ++++++++++++++++++++++++++++++++++ init/Kconfig | 9 +++++++ kernel/fork.c | 2 ++ 5 files changed, 97 insertions(+) create mode 100644 include/linux/task_io_accounting.h create mode 100644 include/linux/task_io_accounting_ops.h (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index ad9c46071ff8..1208feab46e0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -82,6 +82,7 @@ struct sched_param { #include #include #include +#include #include @@ -1013,6 +1014,7 @@ struct task_struct { wait_queue_t *io_wait; /* i/o counters(bytes read/written, #syscalls */ u64 rchar, wchar, syscr, syscw; + struct task_io_accounting ioac; #if defined(CONFIG_TASK_XACCT) u64 acct_rss_mem1; /* accumulated rss usage */ u64 acct_vm_mem1; /* accumulated virtual memory usage */ diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h new file mode 100644 index 000000000000..44d00e9cceea --- /dev/null +++ b/include/linux/task_io_accounting.h @@ -0,0 +1,37 @@ +/* + * task_io_accounting: a structure which is used for recording a single task's + * IO statistics. + * + * Don't include this header file directly - it is designed to be dragged in via + * sched.h. + * + * Blame akpm@osdl.org for all this. + */ + +#ifdef CONFIG_TASK_IO_ACCOUNTING +struct task_io_accounting { + /* + * The number of bytes which this task has caused to be read from + * storage. + */ + u64 read_bytes; + + /* + * The number of bytes which this task has caused, or shall cause to be + * written to disk. + */ + u64 write_bytes; + + /* + * A task can cause "negative" IO too. If this task truncates some + * dirty pagecache, some IO which another task has been accounted for + * (in its write_bytes) will not be happening. We _could_ just + * subtract that from the truncating task's write_bytes, but there is + * information loss in doing that. + */ + u64 cancelled_write_bytes; +}; +#else +struct task_io_accounting { +}; +#endif diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h new file mode 100644 index 000000000000..df2a319106b2 --- /dev/null +++ b/include/linux/task_io_accounting_ops.h @@ -0,0 +1,47 @@ +/* + * Task I/O accounting operations + */ +#ifndef __TASK_IO_ACCOUNTING_OPS_INCLUDED +#define __TASK_IO_ACCOUNTING_OPS_INCLUDED + +#ifdef CONFIG_TASK_IO_ACCOUNTING +static inline void task_io_account_read(size_t bytes) +{ + current->ioac.read_bytes += bytes; +} + +static inline void task_io_account_write(size_t bytes) +{ + current->ioac.write_bytes += bytes; +} + +static inline void task_io_account_cancelled_write(size_t bytes) +{ + current->ioac.cancelled_write_bytes += bytes; +} + +static inline void task_io_accounting_init(struct task_struct *tsk) +{ + memset(&tsk->ioac, 0, sizeof(tsk->ioac)); +} + +#else + +static inline void task_io_account_read(size_t bytes) +{ +} + +static inline void task_io_account_write(size_t bytes) +{ +} + +static inline void task_io_account_cancelled_write(size_t bytes) +{ +} + +static inline void task_io_accounting_init(struct task_struct *tsk) +{ +} + +#endif /* CONFIG_TASK_IO_ACCOUNTING */ +#endif /* __TASK_IO_ACCOUNTING_OPS_INCLUDED */ diff --git a/init/Kconfig b/init/Kconfig index 14d484606fab..9edf103b3ec3 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -304,6 +304,15 @@ config TASK_XACCT Say N if unsure. +config TASK_IO_ACCOUNTING + bool "Enable per-task storage I/O accounting (EXPERIMENTAL)" + depends on TASK_XACCT + help + Collect information on the number of bytes of storage I/O which this + task has caused. + + Say N if unsure. + config SYSCTL bool diff --git a/kernel/fork.c b/kernel/fork.c index 8c859eef8e6a..086e172d0d3d 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -1055,6 +1056,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->wchar = 0; /* I/O counter: bytes written */ p->syscr = 0; /* I/O counter: read syscalls */ p->syscw = 0; /* I/O counter: write syscalls */ + task_io_accounting_init(p); acct_clear_integrals(p); p->it_virt_expires = cputime_zero; -- cgit v1.2.3 From f2f1f8a3b86ccc5e998dc70a3ba35af199fdbc58 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 10 Dec 2006 02:19:50 -0800 Subject: [PATCH] cleanup taskstats.h Fix weird whitespace mangling in taskstats.h Cc: Jay Lan Cc: Shailabh Nagar Cc: Balbir Singh Cc: Chris Sturtivant Cc: Tony Ernst Cc: Guillaume Thouvenin Cc: David Wright Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats.h | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h index 45248806ae9c..15c64037be1b 100644 --- a/include/linux/taskstats.h +++ b/include/linux/taskstats.h @@ -115,31 +115,31 @@ struct taskstats { __u64 ac_majflt; /* Major Page Fault Count */ /* Basic Accounting Fields end */ - /* Extended accounting fields start */ + /* Extended accounting fields start */ /* Accumulated RSS usage in duration of a task, in MBytes-usecs. * The current rss usage is added to this counter every time * a tick is charged to a task's system time. So, at the end we * will have memory usage multiplied by system time. Thus an * average usage per system time unit can be calculated. */ - __u64 coremem; /* accumulated RSS usage in MB-usec */ + __u64 coremem; /* accumulated RSS usage in MB-usec */ /* Accumulated virtual memory usage in duration of a task. * Same as acct_rss_mem1 above except that we keep track of VM usage. */ - __u64 virtmem; /* accumulated VM usage in MB-usec */ + __u64 virtmem; /* accumulated VM usage in MB-usec */ /* High watermark of RSS and virtual memory usage in duration of * a task, in KBytes. */ - __u64 hiwater_rss; /* High-watermark of RSS usage, in KB */ - __u64 hiwater_vm; /* High-water VM usage, in KB */ + __u64 hiwater_rss; /* High-watermark of RSS usage, in KB */ + __u64 hiwater_vm; /* High-water VM usage, in KB */ /* The following four fields are I/O statistics of a task. */ - __u64 read_char; /* bytes read */ - __u64 write_char; /* bytes written */ - __u64 read_syscalls; /* read syscalls */ - __u64 write_syscalls; /* write syscalls */ - /* Extended accounting fields end */ + __u64 read_char; /* bytes read */ + __u64 write_char; /* bytes written */ + __u64 read_syscalls; /* read syscalls */ + __u64 write_syscalls; /* write syscalls */ + /* Extended accounting fields end */ }; -- cgit v1.2.3 From 4a7864ca638e0a38307962ee8ef122822a351b65 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sun, 10 Dec 2006 02:19:53 -0800 Subject: [PATCH] io-accounting: via taskstats Deliver IO accounting via taskstats. Cc: Jay Lan Cc: Shailabh Nagar Cc: Balbir Singh Cc: Chris Sturtivant Cc: Tony Ernst Cc: Guillaume Thouvenin Cc: David Wright Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/taskstats.h | 8 +++++++- kernel/tsacct.c | 9 +++++++++ 2 files changed, 16 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h index 15c64037be1b..3fced4798255 100644 --- a/include/linux/taskstats.h +++ b/include/linux/taskstats.h @@ -31,7 +31,7 @@ */ -#define TASKSTATS_VERSION 2 +#define TASKSTATS_VERSION 3 #define TS_COMM_LEN 32 /* should be >= TASK_COMM_LEN * in linux/sched.h */ @@ -140,6 +140,12 @@ struct taskstats { __u64 read_syscalls; /* read syscalls */ __u64 write_syscalls; /* write syscalls */ /* Extended accounting fields end */ + +#define TASKSTATS_HAS_IO_ACCOUNTING + /* Per-task storage I/O accounting starts */ + __u64 read_bytes; /* bytes of read I/O */ + __u64 write_bytes; /* bytes of write I/O */ + __u64 cancelled_write_bytes; /* bytes of cancelled write I/O */ }; diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 96f77013d3f0..baacc3691415 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -96,6 +96,15 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) stats->write_char = p->wchar; stats->read_syscalls = p->syscr; stats->write_syscalls = p->syscw; +#ifdef CONFIG_TASK_IO_ACCOUNTING + stats->read_bytes = p->ioac.read_bytes; + stats->write_bytes = p->ioac.write_bytes; + stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes; +#else + stats->read_bytes = 0; + stats->write_bytes = 0; + stats->cancelled_write_bytes = 0; +#endif } #undef KB #undef MB -- cgit v1.2.3 From ac7d550499e225efb51a53d0b00667f26b93bdff Mon Sep 17 00:00:00 2001 From: Siddha, Suresh B Date: Sun, 10 Dec 2006 02:20:12 -0800 Subject: [PATCH] sched domain: increase the SMT busy rebalance interval With SMT, if the logical processor is busy, load balance happens for every 8msec(min)-16msec(max). There is no need to do this often, as this is just for fairness(to maintain uniform runqueue lengths) and default time slice anyhow is 100msec. Appended patch increases this interval to 64msec(min)-128msec(max) when the logical processor is busy. Signed-off-by: Suresh Siddha Cc: Nick Piggin Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/topology.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/topology.h b/include/linux/topology.h index da508d1998e4..b93bb6cc6cc2 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -93,7 +93,7 @@ .groups = NULL, \ .min_interval = 1, \ .max_interval = 2, \ - .busy_factor = 8, \ + .busy_factor = 64, \ .imbalance_pct = 110, \ .cache_nice_tries = 0, \ .per_cpu_gain = 25, \ -- cgit v1.2.3 From c9819f4593e8d052b41a89f47140f5c5e7e30582 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Sun, 10 Dec 2006 02:20:25 -0800 Subject: [PATCH] sched: use softirq for load balancing Call rebalance_tick (renamed to run_rebalance_domains) from a newly introduced softirq. We calculate the earliest time for each layer of sched domains to be rescanned (this is the rescan time for idle) and use the earliest of those to schedule the softirq via a new field "next_balance" added to struct rq. Signed-off-by: Christoph Lameter Cc: Peter Williams Cc: Nick Piggin Cc: Christoph Lameter Cc: "Siddha, Suresh B" Cc: "Chen, Kenneth W" Acked-by: Ingo Molnar Cc: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/interrupt.h | 3 ++- kernel/sched.c | 22 +++++++++++++++++----- 2 files changed, 19 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index de7593f4e895..e36e86c869fb 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -231,7 +231,8 @@ enum NET_TX_SOFTIRQ, NET_RX_SOFTIRQ, BLOCK_SOFTIRQ, - TASKLET_SOFTIRQ + TASKLET_SOFTIRQ, + SCHED_SOFTIRQ, }; /* softirq mask and active fields moved to irq_cpustat_t in diff --git a/kernel/sched.c b/kernel/sched.c index 14a8d9050cd4..0a3e748d737d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -227,6 +227,7 @@ struct rq { unsigned long expired_timestamp; unsigned long long timestamp_last_tick; struct task_struct *curr, *idle; + unsigned long next_balance; struct mm_struct *prev_mm; struct prio_array *active, *expired, arrays[2]; int best_expired_prio; @@ -2858,7 +2859,7 @@ static void update_load(struct rq *this_rq) } /* - * rebalance_tick will get called every timer tick, on every CPU. + * run_rebalance_domains is triggered when needed from the scheduler tick. * * It checks each scheduling domain to see if it is due to be balanced, * and initiates a balancing operation if so. @@ -2866,9 +2867,10 @@ static void update_load(struct rq *this_rq) * Balancing parameters are set up in arch_init_sched_domains. */ -static void -rebalance_tick(int this_cpu, struct rq *this_rq) +static void run_rebalance_domains(struct softirq_action *h) { + int this_cpu = smp_processor_id(); + struct rq *this_rq = cpu_rq(this_cpu); unsigned long interval; struct sched_domain *sd; /* @@ -2877,6 +2879,8 @@ rebalance_tick(int this_cpu, struct rq *this_rq) */ enum idle_type idle = !this_rq->nr_running ? SCHED_IDLE : NOT_IDLE; + /* Earliest time when we have to call run_rebalance_domains again */ + unsigned long next_balance = jiffies + 60*HZ; for_each_domain(this_cpu, sd) { if (!(sd->flags & SD_LOAD_BALANCE)) @@ -2891,7 +2895,7 @@ rebalance_tick(int this_cpu, struct rq *this_rq) if (unlikely(!interval)) interval = 1; - if (jiffies - sd->last_balance >= interval) { + if (time_after_eq(jiffies, sd->last_balance + interval)) { if (load_balance(this_cpu, this_rq, sd, idle)) { /* * We've pulled tasks over so either we're no @@ -2902,7 +2906,10 @@ rebalance_tick(int this_cpu, struct rq *this_rq) } sd->last_balance += interval; } + if (time_after(next_balance, sd->last_balance + interval)) + next_balance = sd->last_balance + interval; } + this_rq->next_balance = next_balance; } #else /* @@ -3155,7 +3162,8 @@ void scheduler_tick(void) task_running_tick(rq, p); #ifdef CONFIG_SMP update_load(rq); - rebalance_tick(cpu, rq); + if (time_after_eq(jiffies, rq->next_balance)) + raise_softirq(SCHED_SOFTIRQ); #endif } @@ -6859,6 +6867,10 @@ void __init sched_init(void) set_load_weight(&init_task); +#ifdef CONFIG_SMP + open_softirq(SCHED_SOFTIRQ, run_rebalance_domains, NULL); +#endif + #ifdef CONFIG_RT_MUTEXES plist_head_init(&init_task.pi_waiters, &init_task.pi_lock); #endif -- cgit v1.2.3 From 08c183f31bdbb709f177f6d3110d5f288ea33933 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Sun, 10 Dec 2006 02:20:29 -0800 Subject: [PATCH] sched: add option to serialize load balancing Large sched domains can be very expensive to scan. Add an option SD_SERIALIZE to the sched domain flags. If that flag is set then we make sure that no other such domain is being balanced. [akpm@osdl.org: build fix] Signed-off-by: Christoph Lameter Cc: Peter Williams Cc: Nick Piggin Cc: Christoph Lameter Cc: "Siddha, Suresh B" Cc: "Chen, Kenneth W" Acked-by: Ingo Molnar Cc: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/topology.h | 1 + include/asm-ia64/topology.h | 1 + include/asm-powerpc/topology.h | 1 + include/asm-x86_64/topology.h | 1 + include/linux/sched.h | 1 + include/linux/topology.h | 3 ++- kernel/sched.c | 9 +++++++++ 7 files changed, 16 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h index 978d09596130..ac58580ad664 100644 --- a/include/asm-i386/topology.h +++ b/include/asm-i386/topology.h @@ -89,6 +89,7 @@ static inline int node_to_first_cpu(int node) .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ | SD_BALANCE_FORK \ + | SD_SERIALIZE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/asm-ia64/topology.h b/include/asm-ia64/topology.h index a6e38565ab4c..22ed6749557e 100644 --- a/include/asm-ia64/topology.h +++ b/include/asm-ia64/topology.h @@ -101,6 +101,7 @@ void build_cpu_to_node_map(void); .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ | SD_BALANCE_FORK \ + | SD_SERIALIZE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 64, \ diff --git a/include/asm-powerpc/topology.h b/include/asm-powerpc/topology.h index 50c014007de7..6610495f5f16 100644 --- a/include/asm-powerpc/topology.h +++ b/include/asm-powerpc/topology.h @@ -66,6 +66,7 @@ static inline int pcibus_to_node(struct pci_bus *bus) | SD_BALANCE_EXEC \ | SD_BALANCE_NEWIDLE \ | SD_WAKE_IDLE \ + | SD_SERIALIZE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/asm-x86_64/topology.h b/include/asm-x86_64/topology.h index 5c8f49280dbc..2facec5914d2 100644 --- a/include/asm-x86_64/topology.h +++ b/include/asm-x86_64/topology.h @@ -47,6 +47,7 @@ extern int __node_distance(int, int); .flags = SD_LOAD_BALANCE \ | SD_BALANCE_FORK \ | SD_BALANCE_EXEC \ + | SD_SERIALIZE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 1208feab46e0..ea92e5c89089 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -648,6 +648,7 @@ enum idle_type #define SD_SHARE_CPUPOWER 128 /* Domain members share cpu power */ #define SD_POWERSAVINGS_BALANCE 256 /* Balance for power savings */ #define SD_SHARE_PKG_RESOURCES 512 /* Domain members share cpu pkg resources */ +#define SD_SERIALIZE 1024 /* Only a single load balancing instance */ #define BALANCE_FOR_MC_POWER \ (sched_smt_power_savings ? SD_POWERSAVINGS_BALANCE : 0) diff --git a/include/linux/topology.h b/include/linux/topology.h index b93bb6cc6cc2..6c5a6e6e813b 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -194,7 +194,8 @@ .wake_idx = 0, /* unused */ \ .forkexec_idx = 0, /* unused */ \ .per_cpu_gain = 100, \ - .flags = SD_LOAD_BALANCE, \ + .flags = SD_LOAD_BALANCE \ + | SD_SERIALIZE, \ .last_balance = jiffies, \ .balance_interval = 64, \ .nr_balance_failed = 0, \ diff --git a/kernel/sched.c b/kernel/sched.c index 0a4a26b21f69..2b2b780939c9 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2880,6 +2880,7 @@ static void update_load(struct rq *this_rq) * * Balancing parameters are set up in arch_init_sched_domains. */ +static DEFINE_SPINLOCK(balancing); static void run_rebalance_domains(struct softirq_action *h) { @@ -2909,6 +2910,11 @@ static void run_rebalance_domains(struct softirq_action *h) if (unlikely(!interval)) interval = 1; + if (sd->flags & SD_SERIALIZE) { + if (!spin_trylock(&balancing)) + goto out; + } + if (time_after_eq(jiffies, sd->last_balance + interval)) { if (load_balance(this_cpu, this_rq, sd, idle)) { /* @@ -2920,6 +2926,9 @@ static void run_rebalance_domains(struct softirq_action *h) } sd->last_balance = jiffies; } + if (sd->flags & SD_SERIALIZE) + spin_unlock(&balancing); +out: if (time_after(next_balance, sd->last_balance + interval)) next_balance = sd->last_balance + interval; } -- cgit v1.2.3 From 783609c6cb4eaa23f2ac5c968a44483584ec133f Mon Sep 17 00:00:00 2001 From: Siddha, Suresh B Date: Sun, 10 Dec 2006 02:20:33 -0800 Subject: [PATCH] sched: decrease number of load balances Currently at a particular domain, each cpu in the sched group will do a load balance at the frequency of balance_interval. More the cores and threads, more the cpus will be in each sched group at SMP and NUMA domain. And we endup spending quite a bit of time doing load balancing in those domains. Fix this by making only one cpu(first idle cpu or first cpu in the group if all the cpus are busy) in the sched group do the load balance at that particular sched domain and this load will slowly percolate down to the other cpus with in that group(when they do load balancing at lower domains). Signed-off-by: Suresh Siddha Cc: Christoph Lameter Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 + kernel/sched.c | 59 ++++++++++++++++++++++++++++++++++++++++----------- 2 files changed, 48 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index ea92e5c89089..72d6927d29ed 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -707,6 +707,7 @@ struct sched_domain { unsigned long lb_hot_gained[MAX_IDLE_TYPES]; unsigned long lb_nobusyg[MAX_IDLE_TYPES]; unsigned long lb_nobusyq[MAX_IDLE_TYPES]; + unsigned long lb_stopbalance[MAX_IDLE_TYPES]; /* Active load balancing */ unsigned long alb_cnt; diff --git a/kernel/sched.c b/kernel/sched.c index 15ce772a471a..4e453431c61a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -428,7 +428,7 @@ static inline void task_rq_unlock(struct rq *rq, unsigned long *flags) * bump this up when changing the output format or the meaning of an existing * format, so that tools can adapt (or abort) */ -#define SCHEDSTAT_VERSION 12 +#define SCHEDSTAT_VERSION 13 static int show_schedstat(struct seq_file *seq, void *v) { @@ -466,7 +466,7 @@ static int show_schedstat(struct seq_file *seq, void *v) seq_printf(seq, "domain%d %s", dcnt++, mask_str); for (itype = SCHED_IDLE; itype < MAX_IDLE_TYPES; itype++) { - seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu", + seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu", sd->lb_cnt[itype], sd->lb_balanced[itype], sd->lb_failed[itype], @@ -474,7 +474,8 @@ static int show_schedstat(struct seq_file *seq, void *v) sd->lb_gained[itype], sd->lb_hot_gained[itype], sd->lb_nobusyq[itype], - sd->lb_nobusyg[itype]); + sd->lb_nobusyg[itype], + sd->lb_stopbalance[itype]); } seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu\n", sd->alb_cnt, sd->alb_failed, sd->alb_pushed, @@ -2249,7 +2250,7 @@ out: static struct sched_group * find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long *imbalance, enum idle_type idle, int *sd_idle, - cpumask_t *cpus) + cpumask_t *cpus, int *balance) { struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; @@ -2278,10 +2279,14 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long load, group_capacity; int local_group; int i; + unsigned int balance_cpu = -1, first_idle_cpu = 0; unsigned long sum_nr_running, sum_weighted_load; local_group = cpu_isset(this_cpu, group->cpumask); + if (local_group) + balance_cpu = first_cpu(group->cpumask); + /* Tally up the load of all CPUs in the group */ sum_weighted_load = sum_nr_running = avg_load = 0; @@ -2297,9 +2302,14 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, *sd_idle = 0; /* Bias balancing toward cpus of our domain */ - if (local_group) + if (local_group) { + if (idle_cpu(i) && !first_idle_cpu) { + first_idle_cpu = 1; + balance_cpu = i; + } + load = target_load(i, load_idx); - else + } else load = source_load(i, load_idx); avg_load += load; @@ -2307,6 +2317,16 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, sum_weighted_load += rq->raw_weighted_load; } + /* + * First idle cpu or the first cpu(busiest) in this sched group + * is eligible for doing load balancing at this and above + * domains. + */ + if (local_group && balance_cpu != this_cpu && balance) { + *balance = 0; + goto ret; + } + total_load += avg_load; total_pwr += group->cpu_power; @@ -2498,8 +2518,8 @@ out_balanced: *imbalance = min_load_per_task; return group_min; } -ret: #endif +ret: *imbalance = 0; return NULL; } @@ -2550,7 +2570,8 @@ static inline unsigned long minus_1_or_zero(unsigned long n) * tasks if there is an imbalance. */ static int load_balance(int this_cpu, struct rq *this_rq, - struct sched_domain *sd, enum idle_type idle) + struct sched_domain *sd, enum idle_type idle, + int *balance) { int nr_moved, all_pinned = 0, active_balance = 0, sd_idle = 0; struct sched_group *group; @@ -2573,7 +2594,13 @@ static int load_balance(int this_cpu, struct rq *this_rq, redo: group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, - &cpus); + &cpus, balance); + + if (*balance == 0) { + schedstat_inc(sd, lb_stopbalance[idle]); + goto out_balanced; + } + if (!group) { schedstat_inc(sd, lb_nobusyg[idle]); goto out_balanced; @@ -2715,7 +2742,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) schedstat_inc(sd, lb_cnt[NEWLY_IDLE]); redo: group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, - &sd_idle, &cpus); + &sd_idle, &cpus, NULL); if (!group) { schedstat_inc(sd, lb_nobusyg[NEWLY_IDLE]); goto out_balanced; @@ -2885,7 +2912,7 @@ static DEFINE_SPINLOCK(balancing); static void run_rebalance_domains(struct softirq_action *h) { - int this_cpu = smp_processor_id(); + int this_cpu = smp_processor_id(), balance = 1; struct rq *this_rq = cpu_rq(this_cpu); unsigned long interval; struct sched_domain *sd; @@ -2917,7 +2944,7 @@ static void run_rebalance_domains(struct softirq_action *h) } if (time_after_eq(jiffies, sd->last_balance + interval)) { - if (load_balance(this_cpu, this_rq, sd, idle)) { + if (load_balance(this_cpu, this_rq, sd, idle, &balance)) { /* * We've pulled tasks over so either we're no * longer idle, or one of our SMT siblings is @@ -2932,6 +2959,14 @@ static void run_rebalance_domains(struct softirq_action *h) out: if (time_after(next_balance, sd->last_balance + interval)) next_balance = sd->last_balance + interval; + + /* + * Stop the load balance at this level. There is another + * CPU in our sched group which is doing load balancing more + * actively. + */ + if (!balance) + break; } this_rq->next_balance = next_balance; } -- cgit v1.2.3 From 06066714f6016cffcb249f6ab21b7919de1bc859 Mon Sep 17 00:00:00 2001 From: Chen, Kenneth W Date: Sun, 10 Dec 2006 02:20:35 -0800 Subject: [PATCH] sched: remove lb_stopbalance counter Remove scheduler stats lb_stopbalance counter. This counter can be calculated by: lb_balanced - lb_nobusyg - lb_nobusyq. There is no need to create gazillion counters while we can derive the value. Signed-off-by: Ken Chen Signed-off-by: Suresh Siddha Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 1 - kernel/sched.c | 11 ++++------- 2 files changed, 4 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 72d6927d29ed..ea92e5c89089 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -707,7 +707,6 @@ struct sched_domain { unsigned long lb_hot_gained[MAX_IDLE_TYPES]; unsigned long lb_nobusyg[MAX_IDLE_TYPES]; unsigned long lb_nobusyq[MAX_IDLE_TYPES]; - unsigned long lb_stopbalance[MAX_IDLE_TYPES]; /* Active load balancing */ unsigned long alb_cnt; diff --git a/kernel/sched.c b/kernel/sched.c index 4e453431c61a..66e44b5b53d2 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -428,7 +428,7 @@ static inline void task_rq_unlock(struct rq *rq, unsigned long *flags) * bump this up when changing the output format or the meaning of an existing * format, so that tools can adapt (or abort) */ -#define SCHEDSTAT_VERSION 13 +#define SCHEDSTAT_VERSION 14 static int show_schedstat(struct seq_file *seq, void *v) { @@ -466,7 +466,7 @@ static int show_schedstat(struct seq_file *seq, void *v) seq_printf(seq, "domain%d %s", dcnt++, mask_str); for (itype = SCHED_IDLE; itype < MAX_IDLE_TYPES; itype++) { - seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu", + seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu", sd->lb_cnt[itype], sd->lb_balanced[itype], sd->lb_failed[itype], @@ -474,8 +474,7 @@ static int show_schedstat(struct seq_file *seq, void *v) sd->lb_gained[itype], sd->lb_hot_gained[itype], sd->lb_nobusyq[itype], - sd->lb_nobusyg[itype], - sd->lb_stopbalance[itype]); + sd->lb_nobusyg[itype]); } seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu\n", sd->alb_cnt, sd->alb_failed, sd->alb_pushed, @@ -2596,10 +2595,8 @@ redo: group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, &cpus, balance); - if (*balance == 0) { - schedstat_inc(sd, lb_stopbalance[idle]); + if (*balance == 0) goto out_balanced; - } if (!group) { schedstat_inc(sd, lb_nobusyg[idle]); -- cgit v1.2.3 From ee2f344b33b507af23610c8fdfdde38d7c10fb33 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Sun, 10 Dec 2006 02:20:39 -0800 Subject: [PATCH] ide-cd: Handle strange interrupt on the Intel ESB2 The ESB2 appears to emit spurious DMA interrupts when configured for native mode and handling ATAPI devices. Stratus were able to pin this bug down and produce a patch. This is a rework which applies the fixup only to the ESB2 (for now). We can apply it to other chips later if the same problem is found. This code has been tested and confirmed to fix the problem on the tested systems. Signed-off-by: Alan Cox (Most of the hard work done by Stratus however) Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/ide/ide-cd.c | 7 +++++++ drivers/ide/pci/piix.c | 4 ++++ include/linux/ide.h | 1 + 3 files changed, 12 insertions(+) (limited to 'include/linux') diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index 88214943d00a..5969cec58dc1 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -687,8 +687,15 @@ static void ide_dump_status_no_sense(ide_drive_t *drive, const char *msg, u8 sta static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret) { struct request *rq = HWGROUP(drive)->rq; + ide_hwif_t *hwif = HWIF(drive); int stat, err, sense_key; + /* We may have bogus DMA interrupts in PIO state here */ + if (HWIF(drive)->dma_status && hwif->atapi_irq_bogon) { + stat = hwif->INB(hwif->dma_status); + /* Should we force the bit as well ? */ + hwif->OUTB(stat, hwif->dma_status); + } /* Check for errors. */ stat = HWIF(drive)->INB(IDE_STATUS_REG); if (stat_ret) diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c index cdc3aab9ebcb..b1e9a8eba6b6 100644 --- a/drivers/ide/pci/piix.c +++ b/drivers/ide/pci/piix.c @@ -505,6 +505,10 @@ static void __devinit init_hwif_piix(ide_hwif_t *hwif) /* This is a painful system best to let it self tune for now */ return; } + /* ESB2 appears to generate spurious DMA interrupts in PIO mode + when in native mode */ + if (hwif->pci_dev->device == PCI_DEVICE_ID_INTEL_ESB2_18) + hwif->atapi_irq_bogon = 1; hwif->autodma = 0; hwif->tuneproc = &piix_tune_drive; diff --git a/include/linux/ide.h b/include/linux/ide.h index 9c2050293f17..64e070f62a87 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -796,6 +796,7 @@ typedef struct hwif_s { unsigned sg_mapped : 1; /* sg_table and sg_nents are ready */ unsigned no_io_32bit : 1; /* 1 = can not do 32-bit IO ops */ unsigned err_stops_fifo : 1; /* 1=data FIFO is cleared by an error */ + unsigned atapi_irq_bogon : 1; /* Generates spurious DMA interrupts in PIO mode */ struct device gendev; struct completion gendev_rel_comp; /* To deal with device release() */ -- cgit v1.2.3 From 46031f9a38a9773021f1872abc713d62467ac22e Mon Sep 17 00:00:00 2001 From: Raz Ben-Jehuda(caro) Date: Sun, 10 Dec 2006 02:20:47 -0800 Subject: [PATCH] md: allow reads that have bypassed the cache to be retried on failure If a bypass-the-cache read fails, we simply try again through the cache. If it fails again it will trigger normal recovery precedures. update 1: From: NeilBrown 1/ chunk_aligned_read and retry_aligned_read assume that data_disks == raid_disks - 1 which is not true for raid6. So when an aligned read request bypasses the cache, we can get the wrong data. 2/ The cloned bio is being used-after-free in raid5_align_endio (to test BIO_UPTODATE). 3/ We forgot to add rdev->data_offset when submitting a bio for aligned-read 4/ clone_bio calls blk_recount_segments and then we change bi_bdev, so we need to invalidate the segment counts. 5/ We don't de-reference the rdev when the read completes. This means we need to record the rdev to so it is still available in the end_io routine. Fortunately bi_next in the original bio is unused at this point so we can stuff it in there. 6/ We leak a cloned bio if the target rdev is not usable. From: NeilBrown update 2: 1/ When aligned requests fail (read error) they need to be retried via the normal method (stripe cache). As we cannot be sure that we can process a single read in one go (we may not be able to allocate all the stripes needed) we store a bio-being-retried and a list of bioes-that-still-need-to-be-retried. When find a bio that needs to be retried, we should add it to the list, not to single-bio... 2/ We were never incrementing 'scnt' when resubmitting failed aligned requests. [akpm@osdl.org: build fix] Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid5.c | 165 +++++++++++++++++++++++++++++++++++++++++++-- include/linux/raid/raid5.h | 3 + 2 files changed, 163 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 269b7771a30b..2ac2e56a1a40 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -134,6 +134,8 @@ static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh) if (!test_bit(STRIPE_EXPANDING, &sh->state)) { list_add_tail(&sh->lru, &conf->inactive_list); wake_up(&conf->wait_for_stripe); + if (conf->retry_read_aligned) + md_wakeup_thread(conf->mddev->thread); } } } @@ -2644,19 +2646,81 @@ static int in_chunk_boundary(mddev_t *mddev, struct bio *bio) ((sector & (chunk_sectors - 1)) + bio_sectors); } +/* + * add bio to the retry LIFO ( in O(1) ... we are in interrupt ) + * later sampled by raid5d. + */ +static void add_bio_to_retry(struct bio *bi,raid5_conf_t *conf) +{ + unsigned long flags; + + spin_lock_irqsave(&conf->device_lock, flags); + + bi->bi_next = conf->retry_read_aligned_list; + conf->retry_read_aligned_list = bi; + + spin_unlock_irqrestore(&conf->device_lock, flags); + md_wakeup_thread(conf->mddev->thread); +} + + +static struct bio *remove_bio_from_retry(raid5_conf_t *conf) +{ + struct bio *bi; + + bi = conf->retry_read_aligned; + if (bi) { + conf->retry_read_aligned = NULL; + return bi; + } + bi = conf->retry_read_aligned_list; + if(bi) { + conf->retry_read_aligned = bi->bi_next; + bi->bi_next = NULL; + bi->bi_phys_segments = 1; /* biased count of active stripes */ + bi->bi_hw_segments = 0; /* count of processed stripes */ + } + + return bi; +} + + /* * The "raid5_align_endio" should check if the read succeeded and if it * did, call bio_endio on the original bio (having bio_put the new bio * first). * If the read failed.. */ -int raid5_align_endio(struct bio *bi, unsigned int bytes , int error) +static int raid5_align_endio(struct bio *bi, unsigned int bytes, int error) { struct bio* raid_bi = bi->bi_private; + mddev_t *mddev; + raid5_conf_t *conf; + int uptodate = test_bit(BIO_UPTODATE, &bi->bi_flags); + mdk_rdev_t *rdev; + if (bi->bi_size) return 1; bio_put(bi); - bio_endio(raid_bi, bytes, error); + + mddev = raid_bi->bi_bdev->bd_disk->queue->queuedata; + conf = mddev_to_conf(mddev); + rdev = (void*)raid_bi->bi_next; + raid_bi->bi_next = NULL; + + rdev_dec_pending(rdev, conf->mddev); + + if (!error && uptodate) { + bio_endio(raid_bi, bytes, 0); + if (atomic_dec_and_test(&conf->active_aligned_reads)) + wake_up(&conf->wait_for_stripe); + return 0; + } + + + PRINTK("raid5_align_endio : io error...handing IO for a retry\n"); + + add_bio_to_retry(raid_bi, conf); return 0; } @@ -2665,7 +2729,7 @@ static int chunk_aligned_read(request_queue_t *q, struct bio * raid_bio) mddev_t *mddev = q->queuedata; raid5_conf_t *conf = mddev_to_conf(mddev); const unsigned int raid_disks = conf->raid_disks; - const unsigned int data_disks = raid_disks - 1; + const unsigned int data_disks = raid_disks - conf->max_degraded; unsigned int dd_idx, pd_idx; struct bio* align_bi; mdk_rdev_t *rdev; @@ -2699,13 +2763,25 @@ static int chunk_aligned_read(request_queue_t *q, struct bio * raid_bio) rcu_read_lock(); rdev = rcu_dereference(conf->disks[dd_idx].rdev); if (rdev && test_bit(In_sync, &rdev->flags)) { - align_bi->bi_bdev = rdev->bdev; atomic_inc(&rdev->nr_pending); rcu_read_unlock(); + raid_bio->bi_next = (void*)rdev; + align_bi->bi_bdev = rdev->bdev; + align_bi->bi_flags &= ~(1 << BIO_SEG_VALID); + align_bi->bi_sector += rdev->data_offset; + + spin_lock_irq(&conf->device_lock); + wait_event_lock_irq(conf->wait_for_stripe, + conf->quiesce == 0, + conf->device_lock, /* nothing */); + atomic_inc(&conf->active_aligned_reads); + spin_unlock_irq(&conf->device_lock); + generic_make_request(align_bi); return 1; } else { rcu_read_unlock(); + bio_put(align_bi); return 0; } } @@ -3050,6 +3126,72 @@ static inline sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *ski return STRIPE_SECTORS; } +static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio) +{ + /* We may not be able to submit a whole bio at once as there + * may not be enough stripe_heads available. + * We cannot pre-allocate enough stripe_heads as we may need + * more than exist in the cache (if we allow ever large chunks). + * So we do one stripe head at a time and record in + * ->bi_hw_segments how many have been done. + * + * We *know* that this entire raid_bio is in one chunk, so + * it will be only one 'dd_idx' and only need one call to raid5_compute_sector. + */ + struct stripe_head *sh; + int dd_idx, pd_idx; + sector_t sector, logical_sector, last_sector; + int scnt = 0; + int remaining; + int handled = 0; + + logical_sector = raid_bio->bi_sector & ~((sector_t)STRIPE_SECTORS-1); + sector = raid5_compute_sector( logical_sector, + conf->raid_disks, + conf->raid_disks - conf->max_degraded, + &dd_idx, + &pd_idx, + conf); + last_sector = raid_bio->bi_sector + (raid_bio->bi_size>>9); + + for (; logical_sector < last_sector; + logical_sector += STRIPE_SECTORS, scnt++) { + + if (scnt < raid_bio->bi_hw_segments) + /* already done this stripe */ + continue; + + sh = get_active_stripe(conf, sector, conf->raid_disks, pd_idx, 1); + + if (!sh) { + /* failed to get a stripe - must wait */ + raid_bio->bi_hw_segments = scnt; + conf->retry_read_aligned = raid_bio; + return handled; + } + + set_bit(R5_ReadError, &sh->dev[dd_idx].flags); + add_stripe_bio(sh, raid_bio, dd_idx, 0); + handle_stripe(sh, NULL); + release_stripe(sh); + handled++; + } + spin_lock_irq(&conf->device_lock); + remaining = --raid_bio->bi_phys_segments; + spin_unlock_irq(&conf->device_lock); + if (remaining == 0) { + int bytes = raid_bio->bi_size; + + raid_bio->bi_size = 0; + raid_bio->bi_end_io(raid_bio, bytes, 0); + } + if (atomic_dec_and_test(&conf->active_aligned_reads)) + wake_up(&conf->wait_for_stripe); + return handled; +} + + + /* * This is our raid5 kernel thread. * @@ -3071,6 +3213,7 @@ static void raid5d (mddev_t *mddev) spin_lock_irq(&conf->device_lock); while (1) { struct list_head *first; + struct bio *bio; if (conf->seq_flush != conf->seq_write) { int seq = conf->seq_flush; @@ -3087,6 +3230,16 @@ static void raid5d (mddev_t *mddev) !list_empty(&conf->delayed_list)) raid5_activate_delayed(conf); + while ((bio = remove_bio_from_retry(conf))) { + int ok; + spin_unlock_irq(&conf->device_lock); + ok = retry_aligned_read(conf, bio); + spin_lock_irq(&conf->device_lock); + if (!ok) + break; + handled++; + } + if (list_empty(&conf->handle_list)) break; @@ -3274,6 +3427,7 @@ static int run(mddev_t *mddev) INIT_LIST_HEAD(&conf->inactive_list); atomic_set(&conf->active_stripes, 0); atomic_set(&conf->preread_active_stripes, 0); + atomic_set(&conf->active_aligned_reads, 0); PRINTK("raid5: run(%s) called.\n", mdname(mddev)); @@ -3796,7 +3950,8 @@ static void raid5_quiesce(mddev_t *mddev, int state) spin_lock_irq(&conf->device_lock); conf->quiesce = 1; wait_event_lock_irq(conf->wait_for_stripe, - atomic_read(&conf->active_stripes) == 0, + atomic_read(&conf->active_stripes) == 0 && + atomic_read(&conf->active_aligned_reads) == 0, conf->device_lock, /* nothing */); spin_unlock_irq(&conf->device_lock); break; diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h index 03636d7918fe..d8286db60b96 100644 --- a/include/linux/raid/raid5.h +++ b/include/linux/raid/raid5.h @@ -227,7 +227,10 @@ struct raid5_private_data { struct list_head handle_list; /* stripes needing handling */ struct list_head delayed_list; /* stripes that have plugged requests */ struct list_head bitmap_list; /* stripes delaying awaiting bitmap update */ + struct bio *retry_read_aligned; /* currently retrying aligned bios */ + struct bio *retry_read_aligned_list; /* aligned bios retry list */ atomic_t preread_active_stripes; /* stripes with scheduled io */ + atomic_t active_aligned_reads; atomic_t reshape_stripes; /* stripes with pending writes for reshape */ /* unfortunately we need two cache names as we temporarily have -- cgit v1.2.3 From bbea9f69668a3d0cf9feba15a724cd02896f8675 Mon Sep 17 00:00:00 2001 From: Vadim Lobanov Date: Sun, 10 Dec 2006 02:21:12 -0800 Subject: [PATCH] fdtable: Make fdarray and fdsets equal in size Currently, each fdtable supports three dynamically-sized arrays of data: the fdarray and two fdsets. The code allows the number of fds supported by the fdarray (fdtable->max_fds) to differ from the number of fds supported by each of the fdsets (fdtable->max_fdset). In practice, it is wasteful for these two sizes to differ: whenever we hit a limit on the smaller-capacity structure, we will reallocate the entire fdtable and all the dynamic arrays within it, so any delta in the memory used by the larger-capacity structure will never be touched at all. Rather than hogging this excess, we shouldn't even allocate it in the first place, and keep the capacities of the fdarray and the fdsets equal. This patch removes fdtable->max_fdset. As an added bonus, most of the supporting code becomes simpler. Signed-off-by: Vadim Lobanov Cc: Christoph Hellwig Cc: Al Viro Cc: Dipankar Sarma Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/osf_sys.c | 6 ++--- arch/mips/kernel/kspd.c | 2 +- fs/compat.c | 10 ++++---- fs/exec.c | 2 +- fs/fcntl.c | 5 ++-- fs/file.c | 56 ++++++++++++++++++--------------------------- fs/open.c | 3 +-- fs/select.c | 10 ++++---- include/linux/file.h | 6 ----- include/linux/init_task.h | 1 - kernel/exit.c | 2 +- kernel/fork.c | 24 +++++-------------- security/selinux/hooks.c | 2 +- 13 files changed, 48 insertions(+), 81 deletions(-) (limited to 'include/linux') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index fb804043b320..be133f1f75a4 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -979,7 +979,7 @@ osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, long timeout; int ret = -EINVAL; struct fdtable *fdt; - int max_fdset; + int max_fds; timeout = MAX_SCHEDULE_TIMEOUT; if (tvp) { @@ -1003,9 +1003,9 @@ osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, rcu_read_lock(); fdt = files_fdtable(current->files); - max_fdset = fdt->max_fdset; + max_fds = fdt->max_fds; rcu_read_unlock(); - if (n < 0 || n > max_fdset) + if (n < 0 || n > max_fds) goto out_nofds; /* diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index 2c82412b9efe..5929f883e46b 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -301,7 +301,7 @@ static void sp_cleanup(void) for (;;) { unsigned long set; i = j * __NFDBITS; - if (i >= fdt->max_fdset || i >= fdt->max_fds) + if (i >= fdt->max_fds) break; set = fdt->open_fds->fds_bits[j++]; while (set) { diff --git a/fs/compat.c b/fs/compat.c index b766964a625c..0ec70e3cee0a 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1679,19 +1679,19 @@ int compat_core_sys_select(int n, compat_ulong_t __user *inp, { fd_set_bits fds; char *bits; - int size, max_fdset, ret = -EINVAL; + int size, max_fds, ret = -EINVAL; struct fdtable *fdt; if (n < 0) goto out_nofds; - /* max_fdset can increase, so grab it once to avoid race */ + /* max_fds can increase, so grab it once to avoid race */ rcu_read_lock(); fdt = files_fdtable(current->files); - max_fdset = fdt->max_fdset; + max_fds = fdt->max_fds; rcu_read_unlock(); - if (n > max_fdset) - n = max_fdset; + if (n > max_fds) + n = max_fds; /* * We need 6 bitmaps (in/out/ex for both incoming and outgoing), diff --git a/fs/exec.c b/fs/exec.c index 12d8cd461b41..11fe93f7363c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -783,7 +783,7 @@ static void flush_old_files(struct files_struct * files) j++; i = j * __NFDBITS; fdt = files_fdtable(files); - if (i >= fdt->max_fds || i >= fdt->max_fdset) + if (i >= fdt->max_fds) break; set = fdt->close_on_exec->fds_bits[j]; if (!set) diff --git a/fs/fcntl.c b/fs/fcntl.c index 2bdaef35da54..8e382a5d51bd 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -77,10 +77,9 @@ repeat: start = files->next_fd; newfd = start; - if (start < fdt->max_fdset) { + if (start < fdt->max_fds) newfd = find_next_zero_bit(fdt->open_fds->fds_bits, - fdt->max_fdset, start); - } + fdt->max_fds, start); error = -EMFILE; if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) diff --git a/fs/file.c b/fs/file.c index 51aef675470f..fb3d2038dc21 100644 --- a/fs/file.c +++ b/fs/file.c @@ -68,8 +68,8 @@ void free_fd_array(struct file **array, int num) static void __free_fdtable(struct fdtable *fdt) { - free_fdset(fdt->open_fds, fdt->max_fdset); - free_fdset(fdt->close_on_exec, fdt->max_fdset); + free_fdset(fdt->open_fds, fdt->max_fds); + free_fdset(fdt->close_on_exec, fdt->max_fds); free_fd_array(fdt->fd, fdt->max_fds); kfree(fdt); } @@ -98,7 +98,7 @@ static void free_fdtable_rcu(struct rcu_head *rcu) struct fdtable_defer *fddef; BUG_ON(!fdt); - fdset_size = fdt->max_fdset / 8; + fdset_size = fdt->max_fds / 8; fdarray_size = fdt->max_fds * sizeof(struct file *); if (fdt->free_files) { @@ -110,13 +110,11 @@ static void free_fdtable_rcu(struct rcu_head *rcu) kmem_cache_free(files_cachep, fdt->free_files); return; } - if (fdt->max_fdset <= EMBEDDED_FD_SET_SIZE && - fdt->max_fds <= NR_OPEN_DEFAULT) { + if (fdt->max_fds <= NR_OPEN_DEFAULT) /* * The fdtable was embedded */ return; - } if (fdset_size <= PAGE_SIZE && fdarray_size <= PAGE_SIZE) { kfree(fdt->open_fds); kfree(fdt->close_on_exec); @@ -136,9 +134,7 @@ static void free_fdtable_rcu(struct rcu_head *rcu) void free_fdtable(struct fdtable *fdt) { - if (fdt->free_files || - fdt->max_fdset > EMBEDDED_FD_SET_SIZE || - fdt->max_fds > NR_OPEN_DEFAULT) + if (fdt->free_files || fdt->max_fds > NR_OPEN_DEFAULT) call_rcu(&fdt->rcu, free_fdtable_rcu); } @@ -151,12 +147,11 @@ static void copy_fdtable(struct fdtable *nfdt, struct fdtable *fdt) int i; int count; - BUG_ON(nfdt->max_fdset < fdt->max_fdset); BUG_ON(nfdt->max_fds < fdt->max_fds); /* Copy the existing tables and install the new pointers */ - i = fdt->max_fdset / (sizeof(unsigned long) * 8); - count = (nfdt->max_fdset - fdt->max_fdset) / 8; + i = fdt->max_fds / (sizeof(unsigned long) * 8); + count = (nfdt->max_fds - fdt->max_fds) / 8; /* * Don't copy the entire array if the current fdset is @@ -164,9 +159,9 @@ static void copy_fdtable(struct fdtable *nfdt, struct fdtable *fdt) */ if (i) { memcpy (nfdt->open_fds, fdt->open_fds, - fdt->max_fdset/8); + fdt->max_fds/8); memcpy (nfdt->close_on_exec, fdt->close_on_exec, - fdt->max_fdset/8); + fdt->max_fds/8); memset (&nfdt->open_fds->fds_bits[i], 0, count); memset (&nfdt->close_on_exec->fds_bits[i], 0, count); } @@ -201,7 +196,7 @@ fd_set * alloc_fdset(int num) void free_fdset(fd_set *array, int num) { - if (num <= EMBEDDED_FD_SET_SIZE) /* Don't free an embedded fdset */ + if (num <= NR_OPEN_DEFAULT) /* Don't free an embedded fdset */ return; else if (num <= 8 * PAGE_SIZE) kfree(array); @@ -220,18 +215,6 @@ static struct fdtable *alloc_fdtable(int nr) if (!fdt) goto out; - nfds = max_t(int, 8 * L1_CACHE_BYTES, roundup_pow_of_two(nr + 1)); - if (nfds > NR_OPEN) - nfds = NR_OPEN; - - new_openset = alloc_fdset(nfds); - new_execset = alloc_fdset(nfds); - if (!new_openset || !new_execset) - goto out; - fdt->open_fds = new_openset; - fdt->close_on_exec = new_execset; - fdt->max_fdset = nfds; - nfds = NR_OPEN_DEFAULT; /* * Expand to the max in easy steps, and keep expanding it until @@ -251,15 +234,21 @@ static struct fdtable *alloc_fdtable(int nr) nfds = NR_OPEN; } } while (nfds <= nr); + + new_openset = alloc_fdset(nfds); + new_execset = alloc_fdset(nfds); + if (!new_openset || !new_execset) + goto out; + fdt->open_fds = new_openset; + fdt->close_on_exec = new_execset; + new_fds = alloc_fd_array(nfds); if (!new_fds) - goto out2; + goto out; fdt->fd = new_fds; fdt->max_fds = nfds; fdt->free_files = NULL; return fdt; -out2: - nfds = fdt->max_fdset; out: free_fdset(new_openset, nfds); free_fdset(new_execset, nfds); @@ -290,7 +279,7 @@ static int expand_fdtable(struct files_struct *files, int nr) * we dropped the lock */ cur_fdt = files_fdtable(files); - if (nr >= cur_fdt->max_fds || nr >= cur_fdt->max_fdset) { + if (nr >= cur_fdt->max_fds) { /* Continue as planned */ copy_fdtable(new_fdt, cur_fdt); rcu_assign_pointer(files->fdt, new_fdt); @@ -316,11 +305,10 @@ int expand_files(struct files_struct *files, int nr) fdt = files_fdtable(files); /* Do we need to expand? */ - if (nr < fdt->max_fdset && nr < fdt->max_fds) + if (nr < fdt->max_fds) return 0; /* Can we expand? */ - if (fdt->max_fdset >= NR_OPEN || fdt->max_fds >= NR_OPEN || - nr >= NR_OPEN) + if (nr >= NR_OPEN) return -EMFILE; /* All good, so we try */ diff --git a/fs/open.c b/fs/open.c index 0d94319e8681..c989fb4cf7b9 100644 --- a/fs/open.c +++ b/fs/open.c @@ -864,8 +864,7 @@ int get_unused_fd(void) repeat: fdt = files_fdtable(files); - fd = find_next_zero_bit(fdt->open_fds->fds_bits, - fdt->max_fdset, + fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds, files->next_fd); /* diff --git a/fs/select.c b/fs/select.c index dcbc1112b7ec..fe0893afd931 100644 --- a/fs/select.c +++ b/fs/select.c @@ -311,7 +311,7 @@ static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, { fd_set_bits fds; void *bits; - int ret, max_fdset; + int ret, max_fds; unsigned int size; struct fdtable *fdt; /* Allocate small arguments on the stack to save memory and be faster */ @@ -321,13 +321,13 @@ static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, if (n < 0) goto out_nofds; - /* max_fdset can increase, so grab it once to avoid race */ + /* max_fds can increase, so grab it once to avoid race */ rcu_read_lock(); fdt = files_fdtable(current->files); - max_fdset = fdt->max_fdset; + max_fds = fdt->max_fds; rcu_read_unlock(); - if (n > max_fdset) - n = max_fdset; + if (n > max_fds) + n = max_fds; /* * We need 6 bitmaps (in/out/ex for both incoming and outgoing), diff --git a/include/linux/file.h b/include/linux/file.h index 6e77b9177f9e..02be4012225b 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -26,14 +26,8 @@ struct embedded_fd_set { unsigned long fds_bits[1]; }; -/* - * More than this number of fds: we use a separately allocated fd_set - */ -#define EMBEDDED_FD_SET_SIZE (BITS_PER_BYTE * sizeof(struct embedded_fd_set)) - struct fdtable { unsigned int max_fds; - int max_fdset; struct file ** fd; /* current fd array */ fd_set *close_on_exec; fd_set *open_fds; diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 7272ff9ee77c..58c18daab65d 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -12,7 +12,6 @@ #define INIT_FDTABLE \ { \ .max_fds = NR_OPEN_DEFAULT, \ - .max_fdset = EMBEDDED_FD_SET_SIZE, \ .fd = &init_files.fd_array[0], \ .close_on_exec = (fd_set *)&init_files.close_on_exec_init, \ .open_fds = (fd_set *)&init_files.open_fds_init, \ diff --git a/kernel/exit.c b/kernel/exit.c index 03e64fe4a14a..5f77e76b4f97 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -425,7 +425,7 @@ static void close_files(struct files_struct * files) for (;;) { unsigned long set; i = j * __NFDBITS; - if (i >= fdt->max_fdset || i >= fdt->max_fds) + if (i >= fdt->max_fds) break; set = fdt->open_fds->fds_bits[j++]; while (set) { diff --git a/kernel/fork.c b/kernel/fork.c index 30eab4f063cd..aba595424f78 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -614,7 +614,7 @@ static inline int copy_fs(unsigned long clone_flags, struct task_struct * tsk) static int count_open_files(struct fdtable *fdt) { - int size = fdt->max_fdset; + int size = fdt->max_fds; int i; /* Find the last open fd */ @@ -641,7 +641,6 @@ static struct files_struct *alloc_files(void) newf->next_fd = 0; fdt = &newf->fdtab; fdt->max_fds = NR_OPEN_DEFAULT; - fdt->max_fdset = EMBEDDED_FD_SET_SIZE; fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init; fdt->open_fds = (fd_set *)&newf->open_fds_init; fdt->fd = &newf->fd_array[0]; @@ -662,7 +661,7 @@ static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) { struct files_struct *newf; struct file **old_fds, **new_fds; - int open_files, size, i, expand; + int open_files, size, i; struct fdtable *old_fdt, *new_fdt; *errorp = -ENOMEM; @@ -673,25 +672,14 @@ static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) spin_lock(&oldf->file_lock); old_fdt = files_fdtable(oldf); new_fdt = files_fdtable(newf); - size = old_fdt->max_fdset; open_files = count_open_files(old_fdt); - expand = 0; /* - * Check whether we need to allocate a larger fd array or fd set. - * Note: we're not a clone task, so the open count won't change. + * Check whether we need to allocate a larger fd array and fd set. + * Note: we're not a clone task, so the open count won't change. */ - if (open_files > new_fdt->max_fdset) { - new_fdt->max_fdset = 0; - expand = 1; - } if (open_files > new_fdt->max_fds) { new_fdt->max_fds = 0; - expand = 1; - } - - /* if the old fdset gets grown now, we'll only copy up to "size" fds */ - if (expand) { spin_unlock(&oldf->file_lock); spin_lock(&newf->file_lock); *errorp = expand_files(newf, open_files-1); @@ -739,8 +727,8 @@ static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) /* This is long word aligned thus could use a optimized version */ memset(new_fds, 0, size); - if (new_fdt->max_fdset > open_files) { - int left = (new_fdt->max_fdset-open_files)/8; + if (new_fdt->max_fds > open_files) { + int left = (new_fdt->max_fds-open_files)/8; int start = open_files / (8 * sizeof(unsigned long)); memset(&new_fdt->open_fds->fds_bits[start], 0, left); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 3753416eb9b9..65fb5e8ea941 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1734,7 +1734,7 @@ static inline void flush_unauthorized_files(struct files_struct * files) j++; i = j * __NFDBITS; fdt = files_fdtable(files); - if (i >= fdt->max_fds || i >= fdt->max_fdset) + if (i >= fdt->max_fds) break; set = fdt->open_fds->fds_bits[j]; if (!set) -- cgit v1.2.3 From 4fd45812cbe875a620c86a096a5d46c742694b7e Mon Sep 17 00:00:00 2001 From: Vadim Lobanov Date: Sun, 10 Dec 2006 02:21:17 -0800 Subject: [PATCH] fdtable: Remove the free_files field An fdtable can either be embedded inside a files_struct or standalone (after being expanded). When an fdtable is being discarded after all RCU references to it have expired, we must either free it directly, in the standalone case, or free the files_struct it is contained within, in the embedded case. Currently the free_files field controls this behavior, but we can get rid of it entirely, as all the necessary information is already recorded. We can distinguish embedded and standalone fdtables using max_fds, and if it is embedded we can divine the relevant files_struct using container_of(). Signed-off-by: Vadim Lobanov Cc: Christoph Hellwig Cc: Al Viro Cc: Dipankar Sarma Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/file.c | 27 ++++++++------------------- include/linux/file.h | 3 +-- include/linux/init_task.h | 1 - kernel/exit.c | 6 ++---- kernel/fork.c | 1 - 5 files changed, 11 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/fs/file.c b/fs/file.c index fb3d2038dc21..17e6a55521e2 100644 --- a/fs/file.c +++ b/fs/file.c @@ -91,7 +91,7 @@ static void free_fdtable_work(struct work_struct *work) } } -static void free_fdtable_rcu(struct rcu_head *rcu) +void free_fdtable_rcu(struct rcu_head *rcu) { struct fdtable *fdt = container_of(rcu, struct fdtable, rcu); int fdset_size, fdarray_size; @@ -101,20 +101,15 @@ static void free_fdtable_rcu(struct rcu_head *rcu) fdset_size = fdt->max_fds / 8; fdarray_size = fdt->max_fds * sizeof(struct file *); - if (fdt->free_files) { + if (fdt->max_fds <= NR_OPEN_DEFAULT) { /* - * The this fdtable was embedded in the files structure - * and the files structure itself was getting destroyed. - * It is now safe to free the files structure. + * This fdtable is embedded in the files structure and that + * structure itself is getting destroyed. */ - kmem_cache_free(files_cachep, fdt->free_files); + kmem_cache_free(files_cachep, + container_of(fdt, struct files_struct, fdtab)); return; } - if (fdt->max_fds <= NR_OPEN_DEFAULT) - /* - * The fdtable was embedded - */ - return; if (fdset_size <= PAGE_SIZE && fdarray_size <= PAGE_SIZE) { kfree(fdt->open_fds); kfree(fdt->close_on_exec); @@ -132,12 +127,6 @@ static void free_fdtable_rcu(struct rcu_head *rcu) } } -void free_fdtable(struct fdtable *fdt) -{ - if (fdt->free_files || fdt->max_fds > NR_OPEN_DEFAULT) - call_rcu(&fdt->rcu, free_fdtable_rcu); -} - /* * Expand the fdset in the files_struct. Called with the files spinlock * held for write. @@ -247,7 +236,6 @@ static struct fdtable *alloc_fdtable(int nr) goto out; fdt->fd = new_fds; fdt->max_fds = nfds; - fdt->free_files = NULL; return fdt; out: free_fdset(new_openset, nfds); @@ -283,7 +271,8 @@ static int expand_fdtable(struct files_struct *files, int nr) /* Continue as planned */ copy_fdtable(new_fdt, cur_fdt); rcu_assign_pointer(files->fdt, new_fdt); - free_fdtable(cur_fdt); + if (cur_fdt->max_fds > NR_OPEN_DEFAULT) + call_rcu(&cur_fdt->rcu, free_fdtable_rcu); } else { /* Somebody else expanded, so undo our attempt */ __free_fdtable(new_fdt); diff --git a/include/linux/file.h b/include/linux/file.h index 02be4012225b..319118f275b0 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -32,7 +32,6 @@ struct fdtable { fd_set *close_on_exec; fd_set *open_fds; struct rcu_head rcu; - struct files_struct *free_files; struct fdtable *next; }; @@ -84,7 +83,7 @@ extern fd_set *alloc_fdset(int); extern void free_fdset(fd_set *, int); extern int expand_files(struct files_struct *, int nr); -extern void free_fdtable(struct fdtable *fdt); +extern void free_fdtable_rcu(struct rcu_head *rcu); extern void __init files_defer_init(void); static inline struct file * fcheck_files(struct files_struct *files, unsigned int fd) diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 58c18daab65d..b5315150199e 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -16,7 +16,6 @@ .close_on_exec = (fd_set *)&init_files.close_on_exec_init, \ .open_fds = (fd_set *)&init_files.open_fds_init, \ .rcu = RCU_HEAD_INIT, \ - .free_files = NULL, \ .next = NULL, \ } diff --git a/kernel/exit.c b/kernel/exit.c index 5f77e76b4f97..122fadb972fc 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -466,11 +466,9 @@ void fastcall put_files_struct(struct files_struct *files) * you can free files immediately. */ fdt = files_fdtable(files); - if (fdt == &files->fdtab) - fdt->free_files = files; - else + if (fdt != &files->fdtab) kmem_cache_free(files_cachep, files); - free_fdtable(fdt); + call_rcu(&fdt->rcu, free_fdtable_rcu); } } diff --git a/kernel/fork.c b/kernel/fork.c index aba595424f78..d16c566eb645 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -645,7 +645,6 @@ static struct files_struct *alloc_files(void) fdt->open_fds = (fd_set *)&newf->open_fds_init; fdt->fd = &newf->fd_array[0]; INIT_RCU_HEAD(&fdt->rcu); - fdt->free_files = NULL; fdt->next = NULL; rcu_assign_pointer(newf->fdt, fdt); out: -- cgit v1.2.3 From 5466b456ed6748e0bfe02831e570004d4c04c1d7 Mon Sep 17 00:00:00 2001 From: Vadim Lobanov Date: Sun, 10 Dec 2006 02:21:22 -0800 Subject: [PATCH] fdtable: Implement new pagesize-based fdtable allocator This patch provides an improved fdtable allocation scheme, useful for expanding fdtable file descriptor entries. The main focus is on the fdarray, as its memory usage grows 128 times faster than that of an fdset. The allocation algorithm sizes the fdarray in such a way that its memory usage increases in easy page-sized chunks. The overall algorithm expands the allowed size in powers of two, in order to amortize the cost of invoking vmalloc() for larger allocation sizes. Namely, the following sizes for the fdarray are considered, and the smallest that accommodates the requested fd count is chosen: pagesize / 4 pagesize / 2 pagesize <- memory allocator switch point pagesize * 2 pagesize * 4 ...etc... Unlike the current implementation, this allocation scheme does not require a loop to compute the optimal fdarray size, and can be done in efficient straightline code. Furthermore, since the fdarray overflows the pagesize boundary long before any of the fdsets do, it makes sense to optimize run-time by allocating both fdsets in a single swoop. Even together, they will still be, by far, smaller than the fdarray. The fdtable->open_fds is now used as the anchor for the fdset memory allocation. Signed-off-by: Vadim Lobanov Cc: Christoph Hellwig Cc: Al Viro Cc: Dipankar Sarma Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/file.c | 208 ++++++++++++++++++--------------------------------- include/linux/file.h | 6 -- 2 files changed, 72 insertions(+), 142 deletions(-) (limited to 'include/linux') diff --git a/fs/file.c b/fs/file.c index 17e6a55521e2..857fa49e984c 100644 --- a/fs/file.c +++ b/fs/file.c @@ -32,46 +32,28 @@ struct fdtable_defer { */ static DEFINE_PER_CPU(struct fdtable_defer, fdtable_defer_list); - -/* - * Allocate an fd array, using kmalloc or vmalloc. - * Note: the array isn't cleared at allocation time. - */ -struct file ** alloc_fd_array(int num) +static inline void * alloc_fdmem(unsigned int size) { - struct file **new_fds; - int size = num * sizeof(struct file *); - if (size <= PAGE_SIZE) - new_fds = (struct file **) kmalloc(size, GFP_KERNEL); - else - new_fds = (struct file **) vmalloc(size); - return new_fds; + return kmalloc(size, GFP_KERNEL); + else + return vmalloc(size); } -void free_fd_array(struct file **array, int num) +static inline void free_fdarr(struct fdtable *fdt) { - int size = num * sizeof(struct file *); - - if (!array) { - printk (KERN_ERR "free_fd_array: array = 0 (num = %d)\n", num); - return; - } - - if (num <= NR_OPEN_DEFAULT) /* Don't free the embedded fd array! */ - return; - else if (size <= PAGE_SIZE) - kfree(array); + if (fdt->max_fds <= (PAGE_SIZE / sizeof(struct file *))) + kfree(fdt->fd); else - vfree(array); + vfree(fdt->fd); } -static void __free_fdtable(struct fdtable *fdt) +static inline void free_fdset(struct fdtable *fdt) { - free_fdset(fdt->open_fds, fdt->max_fds); - free_fdset(fdt->close_on_exec, fdt->max_fds); - free_fd_array(fdt->fd, fdt->max_fds); - kfree(fdt); + if (fdt->max_fds <= (PAGE_SIZE * BITS_PER_BYTE / 2)) + kfree(fdt->open_fds); + else + vfree(fdt->open_fds); } static void free_fdtable_work(struct work_struct *work) @@ -86,7 +68,9 @@ static void free_fdtable_work(struct work_struct *work) spin_unlock_bh(&f->lock); while(fdt) { struct fdtable *next = fdt->next; - __free_fdtable(fdt); + vfree(fdt->fd); + free_fdset(fdt); + kfree(fdt); fdt = next; } } @@ -94,12 +78,9 @@ static void free_fdtable_work(struct work_struct *work) void free_fdtable_rcu(struct rcu_head *rcu) { struct fdtable *fdt = container_of(rcu, struct fdtable, rcu); - int fdset_size, fdarray_size; struct fdtable_defer *fddef; BUG_ON(!fdt); - fdset_size = fdt->max_fds / 8; - fdarray_size = fdt->max_fds * sizeof(struct file *); if (fdt->max_fds <= NR_OPEN_DEFAULT) { /* @@ -110,10 +91,9 @@ void free_fdtable_rcu(struct rcu_head *rcu) container_of(fdt, struct files_struct, fdtab)); return; } - if (fdset_size <= PAGE_SIZE && fdarray_size <= PAGE_SIZE) { - kfree(fdt->open_fds); - kfree(fdt->close_on_exec); + if (fdt->max_fds <= (PAGE_SIZE / sizeof(struct file *))) { kfree(fdt->fd); + kfree(fdt->open_fds); kfree(fdt); } else { fddef = &get_cpu_var(fdtable_defer_list); @@ -131,116 +111,70 @@ void free_fdtable_rcu(struct rcu_head *rcu) * Expand the fdset in the files_struct. Called with the files spinlock * held for write. */ -static void copy_fdtable(struct fdtable *nfdt, struct fdtable *fdt) +static void copy_fdtable(struct fdtable *nfdt, struct fdtable *ofdt) { - int i; - int count; - - BUG_ON(nfdt->max_fds < fdt->max_fds); - /* Copy the existing tables and install the new pointers */ - - i = fdt->max_fds / (sizeof(unsigned long) * 8); - count = (nfdt->max_fds - fdt->max_fds) / 8; - - /* - * Don't copy the entire array if the current fdset is - * not yet initialised. - */ - if (i) { - memcpy (nfdt->open_fds, fdt->open_fds, - fdt->max_fds/8); - memcpy (nfdt->close_on_exec, fdt->close_on_exec, - fdt->max_fds/8); - memset (&nfdt->open_fds->fds_bits[i], 0, count); - memset (&nfdt->close_on_exec->fds_bits[i], 0, count); - } + unsigned int cpy, set; - /* Don't copy/clear the array if we are creating a new - fd array for fork() */ - if (fdt->max_fds) { - memcpy(nfdt->fd, fdt->fd, - fdt->max_fds * sizeof(struct file *)); - /* clear the remainder of the array */ - memset(&nfdt->fd[fdt->max_fds], 0, - (nfdt->max_fds - fdt->max_fds) * - sizeof(struct file *)); - } -} - -/* - * Allocate an fdset array, using kmalloc or vmalloc. - * Note: the array isn't cleared at allocation time. - */ -fd_set * alloc_fdset(int num) -{ - fd_set *new_fdset; - int size = num / 8; - - if (size <= PAGE_SIZE) - new_fdset = (fd_set *) kmalloc(size, GFP_KERNEL); - else - new_fdset = (fd_set *) vmalloc(size); - return new_fdset; -} - -void free_fdset(fd_set *array, int num) -{ - if (num <= NR_OPEN_DEFAULT) /* Don't free an embedded fdset */ + BUG_ON(nfdt->max_fds < ofdt->max_fds); + if (ofdt->max_fds == 0) return; - else if (num <= 8 * PAGE_SIZE) - kfree(array); - else - vfree(array); + + cpy = ofdt->max_fds * sizeof(struct file *); + set = (nfdt->max_fds - ofdt->max_fds) * sizeof(struct file *); + memcpy(nfdt->fd, ofdt->fd, cpy); + memset((char *)(nfdt->fd) + cpy, 0, set); + + cpy = ofdt->max_fds / BITS_PER_BYTE; + set = (nfdt->max_fds - ofdt->max_fds) / BITS_PER_BYTE; + memcpy(nfdt->open_fds, ofdt->open_fds, cpy); + memset((char *)(nfdt->open_fds) + cpy, 0, set); + memcpy(nfdt->close_on_exec, ofdt->close_on_exec, cpy); + memset((char *)(nfdt->close_on_exec) + cpy, 0, set); } -static struct fdtable *alloc_fdtable(int nr) +static struct fdtable * alloc_fdtable(unsigned int nr) { - struct fdtable *fdt = NULL; - int nfds = 0; - fd_set *new_openset = NULL, *new_execset = NULL; - struct file **new_fds; - - fdt = kzalloc(sizeof(*fdt), GFP_KERNEL); - if (!fdt) - goto out; + struct fdtable *fdt; + char *data; - nfds = NR_OPEN_DEFAULT; /* - * Expand to the max in easy steps, and keep expanding it until - * we have enough for the requested fd array size. + * Figure out how many fds we actually want to support in this fdtable. + * Allocation steps are keyed to the size of the fdarray, since it + * grows far faster than any of the other dynamic data. We try to fit + * the fdarray into comfortable page-tuned chunks: starting at 1024B + * and growing in powers of two from there on. */ - do { -#if NR_OPEN_DEFAULT < 256 - if (nfds < 256) - nfds = 256; - else -#endif - if (nfds < (PAGE_SIZE / sizeof(struct file *))) - nfds = PAGE_SIZE / sizeof(struct file *); - else { - nfds = nfds * 2; - if (nfds > NR_OPEN) - nfds = NR_OPEN; - } - } while (nfds <= nr); - - new_openset = alloc_fdset(nfds); - new_execset = alloc_fdset(nfds); - if (!new_openset || !new_execset) - goto out; - fdt->open_fds = new_openset; - fdt->close_on_exec = new_execset; + nr /= (1024 / sizeof(struct file *)); + nr = roundup_pow_of_two(nr + 1); + nr *= (1024 / sizeof(struct file *)); + if (nr > NR_OPEN) + nr = NR_OPEN; - new_fds = alloc_fd_array(nfds); - if (!new_fds) + fdt = kmalloc(sizeof(struct fdtable), GFP_KERNEL); + if (!fdt) goto out; - fdt->fd = new_fds; - fdt->max_fds = nfds; + fdt->max_fds = nr; + data = alloc_fdmem(nr * sizeof(struct file *)); + if (!data) + goto out_fdt; + fdt->fd = (struct file **)data; + data = alloc_fdmem(max_t(unsigned int, + 2 * nr / BITS_PER_BYTE, L1_CACHE_BYTES)); + if (!data) + goto out_arr; + fdt->open_fds = (fd_set *)data; + data += nr / BITS_PER_BYTE; + fdt->close_on_exec = (fd_set *)data; + INIT_RCU_HEAD(&fdt->rcu); + fdt->next = NULL; + return fdt; -out: - free_fdset(new_openset, nfds); - free_fdset(new_execset, nfds); + +out_arr: + free_fdarr(fdt); +out_fdt: kfree(fdt); +out: return NULL; } @@ -275,7 +209,9 @@ static int expand_fdtable(struct files_struct *files, int nr) call_rcu(&cur_fdt->rcu, free_fdtable_rcu); } else { /* Somebody else expanded, so undo our attempt */ - __free_fdtable(new_fdt); + free_fdarr(new_fdt); + free_fdset(new_fdt); + kfree(new_fdt); } return 1; } diff --git a/include/linux/file.h b/include/linux/file.h index 319118f275b0..edca361f2ab4 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -76,12 +76,6 @@ extern int get_unused_fd(void); extern void FASTCALL(put_unused_fd(unsigned int fd)); struct kmem_cache; -extern struct file ** alloc_fd_array(int); -extern void free_fd_array(struct file **, int); - -extern fd_set *alloc_fdset(int); -extern void free_fdset(fd_set *, int); - extern int expand_files(struct files_struct *, int nr); extern void free_fdtable_rcu(struct rcu_head *rcu); extern void __init files_defer_init(void); -- cgit v1.2.3 From 4c36a5dec25fb344ad76b11860da3a8b50bd1248 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sun, 10 Dec 2006 02:21:24 -0800 Subject: [PATCH] round_jiffies infrastructure Introduce a round_jiffies() function as well as a round_jiffies_relative() function. These functions round a jiffies value to the next whole second. The primary purpose of this rounding is to cause all "we don't care exactly when" timers to happen at the same jiffy. This avoids multiple timers firing within the second for no real reason; with dynamic ticks these extra timers cause wakeups from deep sleep CPU sleep states and thus waste power. The exact wakeup moment is skewed by the cpu number, to avoid all cpus from waking up at the exact same time (and hitting the same lock/cachelines there) [akpm@osdl.org: fix variable type] Signed-off-by: Arjan van de Ven Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/timer.h | 6 +++ kernel/timer.c | 132 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 138 insertions(+) (limited to 'include/linux') diff --git a/include/linux/timer.h b/include/linux/timer.h index c982304dbafd..eeef6643d4c6 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -98,4 +98,10 @@ extern void run_local_timers(void); struct hrtimer; extern int it_real_fn(struct hrtimer *); +unsigned long __round_jiffies(unsigned long j, int cpu); +unsigned long __round_jiffies_relative(unsigned long j, int cpu); +unsigned long round_jiffies(unsigned long j); +unsigned long round_jiffies_relative(unsigned long j); + + #endif diff --git a/kernel/timer.c b/kernel/timer.c index c1c7fbcffec1..b1f40f256eb0 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -80,6 +80,138 @@ tvec_base_t boot_tvec_bases; EXPORT_SYMBOL(boot_tvec_bases); static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases; +/** + * __round_jiffies - function to round jiffies to a full second + * @j: the time in (absolute) jiffies that should be rounded + * @cpu: the processor number on which the timeout will happen + * + * __round_jiffies rounds an absolute time in the future (in jiffies) + * up or down to (approximately) full seconds. This is useful for timers + * for which the exact time they fire does not matter too much, as long as + * they fire approximately every X seconds. + * + * By rounding these timers to whole seconds, all such timers will fire + * at the same time, rather than at various times spread out. The goal + * of this is to have the CPU wake up less, which saves power. + * + * The exact rounding is skewed for each processor to avoid all + * processors firing at the exact same time, which could lead + * to lock contention or spurious cache line bouncing. + * + * The return value is the rounded version of the "j" parameter. + */ +unsigned long __round_jiffies(unsigned long j, int cpu) +{ + int rem; + unsigned long original = j; + + /* + * We don't want all cpus firing their timers at once hitting the + * same lock or cachelines, so we skew each extra cpu with an extra + * 3 jiffies. This 3 jiffies came originally from the mm/ code which + * already did this. + * The skew is done by adding 3*cpunr, then round, then subtract this + * extra offset again. + */ + j += cpu * 3; + + rem = j % HZ; + + /* + * If the target jiffie is just after a whole second (which can happen + * due to delays of the timer irq, long irq off times etc etc) then + * we should round down to the whole second, not up. Use 1/4th second + * as cutoff for this rounding as an extreme upper bound for this. + */ + if (rem < HZ/4) /* round down */ + j = j - rem; + else /* round up */ + j = j - rem + HZ; + + /* now that we have rounded, subtract the extra skew again */ + j -= cpu * 3; + + if (j <= jiffies) /* rounding ate our timeout entirely; */ + return original; + return j; +} +EXPORT_SYMBOL_GPL(__round_jiffies); + +/** + * __round_jiffies_relative - function to round jiffies to a full second + * @j: the time in (relative) jiffies that should be rounded + * @cpu: the processor number on which the timeout will happen + * + * __round_jiffies_relative rounds a time delta in the future (in jiffies) + * up or down to (approximately) full seconds. This is useful for timers + * for which the exact time they fire does not matter too much, as long as + * they fire approximately every X seconds. + * + * By rounding these timers to whole seconds, all such timers will fire + * at the same time, rather than at various times spread out. The goal + * of this is to have the CPU wake up less, which saves power. + * + * The exact rounding is skewed for each processor to avoid all + * processors firing at the exact same time, which could lead + * to lock contention or spurious cache line bouncing. + * + * The return value is the rounded version of the "j" parameter. + */ +unsigned long __round_jiffies_relative(unsigned long j, int cpu) +{ + /* + * In theory the following code can skip a jiffy in case jiffies + * increments right between the addition and the later subtraction. + * However since the entire point of this function is to use approximate + * timeouts, it's entirely ok to not handle that. + */ + return __round_jiffies(j + jiffies, cpu) - jiffies; +} +EXPORT_SYMBOL_GPL(__round_jiffies_relative); + +/** + * round_jiffies - function to round jiffies to a full second + * @j: the time in (absolute) jiffies that should be rounded + * + * round_jiffies rounds an absolute time in the future (in jiffies) + * up or down to (approximately) full seconds. This is useful for timers + * for which the exact time they fire does not matter too much, as long as + * they fire approximately every X seconds. + * + * By rounding these timers to whole seconds, all such timers will fire + * at the same time, rather than at various times spread out. The goal + * of this is to have the CPU wake up less, which saves power. + * + * The return value is the rounded version of the "j" parameter. + */ +unsigned long round_jiffies(unsigned long j) +{ + return __round_jiffies(j, raw_smp_processor_id()); +} +EXPORT_SYMBOL_GPL(round_jiffies); + +/** + * round_jiffies_relative - function to round jiffies to a full second + * @j: the time in (relative) jiffies that should be rounded + * + * round_jiffies_relative rounds a time delta in the future (in jiffies) + * up or down to (approximately) full seconds. This is useful for timers + * for which the exact time they fire does not matter too much, as long as + * they fire approximately every X seconds. + * + * By rounding these timers to whole seconds, all such timers will fire + * at the same time, rather than at various times spread out. The goal + * of this is to have the CPU wake up less, which saves power. + * + * The return value is the rounded version of the "j" parameter. + */ +unsigned long round_jiffies_relative(unsigned long j) +{ + return __round_jiffies_relative(j, raw_smp_processor_id()); +} +EXPORT_SYMBOL_GPL(round_jiffies_relative); + + static inline void set_running_timer(tvec_base_t *base, struct timer_list *timer) { -- cgit v1.2.3 From f5f1a24a2caa299bb7d294aee92d7dd3410d9ed7 Mon Sep 17 00:00:00 2001 From: Daniel Walker Date: Sun, 10 Dec 2006 02:21:33 -0800 Subject: [PATCH] clocksource: small cleanup Mostly changing alignment. Just some general cleanup. [akpm@osdl.org: build fix] Signed-off-by: Daniel Walker Acked-by: John Stultz Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/clocksource/acpi_pm.c | 6 +++--- include/linux/clocksource.h | 2 +- kernel/time/clocksource.c | 6 +++--- kernel/timer.c | 16 +++++++++++----- 4 files changed, 18 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/clocksource/acpi_pm.c b/drivers/clocksource/acpi_pm.c index 8ab61ef97b4c..b6bcdbbf57b3 100644 --- a/drivers/clocksource/acpi_pm.c +++ b/drivers/clocksource/acpi_pm.c @@ -77,11 +77,11 @@ static struct clocksource clocksource_acpi_pm = { #ifdef CONFIG_PCI -static int acpi_pm_good; +static int __devinitdata acpi_pm_good; static int __init acpi_pm_good_setup(char *__str) { - acpi_pm_good = 1; - return 1; + acpi_pm_good = 1; + return 1; } __setup("acpi_pm_good", acpi_pm_good_setup); diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index d852024ed095..1622d23a8dc3 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -159,7 +159,7 @@ static inline s64 cyc2ns(struct clocksource *cs, cycle_t cycles) * Unless you're the timekeeping code, you should not be using this! */ static inline void clocksource_calculate_interval(struct clocksource *c, - unsigned long length_nsec) + unsigned long length_nsec) { u64 tmp; diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 3651b34cc355..22504afc0d34 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -156,7 +156,7 @@ int clocksource_register(struct clocksource *c) /* check if clocksource is already registered */ if (is_registered_source(c)) { printk("register_clocksource: Cannot register %s. " - "Already registered!", c->name); + "Already registered!", c->name); ret = -EBUSY; } else { /* register it */ @@ -276,10 +276,10 @@ sysfs_show_available_clocksources(struct sys_device *dev, char *buf) * Sysfs setup bits: */ static SYSDEV_ATTR(current_clocksource, 0600, sysfs_show_current_clocksources, - sysfs_override_clocksource); + sysfs_override_clocksource); static SYSDEV_ATTR(available_clocksource, 0600, - sysfs_show_available_clocksources, NULL); + sysfs_show_available_clocksources, NULL); static struct sysdev_class clocksource_sysclass = { set_kset_name("clocksource"), diff --git a/kernel/timer.c b/kernel/timer.c index b1f40f256eb0..0256ab443d8a 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -846,7 +846,7 @@ static int change_clocksource(void) clock = new; clock->cycle_last = now; printk(KERN_INFO "Time: %s clocksource has been installed.\n", - clock->name); + clock->name); return 1; } else if (clock->update_callback) { return clock->update_callback(); @@ -854,7 +854,10 @@ static int change_clocksource(void) return 0; } #else -#define change_clocksource() (0) +static inline int change_clocksource(void) +{ + return 0; +} #endif /** @@ -952,7 +955,8 @@ device_initcall(timekeeping_init_device); * If the error is already larger, we look ahead even further * to compensate for late or lost adjustments. */ -static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, s64 *offset) +static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, + s64 *offset) { s64 tick_error, i; u32 look_ahead, adj; @@ -976,7 +980,8 @@ static __always_inline int clocksource_bigadjust(s64 error, s64 *interval, s64 * * Now calculate the error in (1 << look_ahead) ticks, but first * remove the single look ahead already included in the error. */ - tick_error = current_tick_length() >> (TICK_LENGTH_SHIFT - clock->shift + 1); + tick_error = current_tick_length() >> + (TICK_LENGTH_SHIFT - clock->shift + 1); tick_error -= clock->xtime_interval >> 1; error = ((error - tick_error) >> look_ahead) + tick_error; @@ -1028,7 +1033,8 @@ static void clocksource_adjust(struct clocksource *clock, s64 offset) clock->mult += adj; clock->xtime_interval += interval; clock->xtime_nsec -= offset; - clock->error -= (interval - offset) << (TICK_LENGTH_SHIFT - clock->shift); + clock->error -= (interval - offset) << + (TICK_LENGTH_SHIFT - clock->shift); } /** -- cgit v1.2.3 From 6aa8b732ca01c3d7a54e93f4d701b8aabbe60fb7 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 10 Dec 2006 02:21:36 -0800 Subject: [PATCH] kvm: userspace interface web site: http://kvm.sourceforge.net mailing list: kvm-devel@lists.sourceforge.net (http://lists.sourceforge.net/lists/listinfo/kvm-devel) The following patchset adds a driver for Intel's hardware virtualization extensions to the x86 architecture. The driver adds a character device (/dev/kvm) that exposes the virtualization capabilities to userspace. Using this driver, a process can run a virtual machine (a "guest") in a fully virtualized PC containing its own virtual hard disks, network adapters, and display. Using this driver, one can start multiple virtual machines on a host. Each virtual machine is a process on the host; a virtual cpu is a thread in that process. kill(1), nice(1), top(1) work as expected. In effect, the driver adds a third execution mode to the existing two: we now have kernel mode, user mode, and guest mode. Guest mode has its own address space mapping guest physical memory (which is accessible to user mode by mmap()ing /dev/kvm). Guest mode has no access to any I/O devices; any such access is intercepted and directed to user mode for emulation. The driver supports i386 and x86_64 hosts and guests. All combinations are allowed except x86_64 guest on i386 host. For i386 guests and hosts, both pae and non-pae paging modes are supported. SMP hosts and UP guests are supported. At the moment only Intel hardware is supported, but AMD virtualization support is being worked on. Performance currently is non-stellar due to the naive implementation of the mmu virtualization, which throws away most of the shadow page table entries every context switch. We plan to address this in two ways: - cache shadow page tables across tlb flushes - wait until AMD and Intel release processors with nested page tables Currently a virtual desktop is responsive but consumes a lot of CPU. Under Windows I tried playing pinball and watching a few flash movies; with a recent CPU one can hardly feel the virtualization. Linux/X is slower, probably due to X being in a separate process. In addition to the driver, you need a slightly modified qemu to provide I/O device emulation and the BIOS. Caveats (akpm: might no longer be true): - The Windows install currently bluescreens due to a problem with the virtual APIC. We are working on a fix. A temporary workaround is to use an existing image or install through qemu - Windows 64-bit does not work. That's also true for qemu, so it's probably a problem with the device model. [bero@arklinux.org: build fix] [simon.kagstrom@bth.se: build fix, other fixes] [uril@qumranet.com: KVM: Expose interrupt bitmap] [akpm@osdl.org: i386 build fix] [mingo@elte.hu: i386 fixes] [rdreier@cisco.com: add log levels to all printks] [randy.dunlap@oracle.com: Fix sparse NULL and C99 struct init warnings] [anthony@codemonkey.ws: KVM: AMD SVM: 32-bit host support] Signed-off-by: Yaniv Kamay Signed-off-by: Avi Kivity Cc: Simon Kagstrom Cc: Bernhard Rosenkraenzer Signed-off-by: Uri Lublin Cc: Ingo Molnar Cc: Roland Dreier Signed-off-by: Randy Dunlap Signed-off-by: Anthony Liguori Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/Kconfig | 2 + drivers/Makefile | 1 + drivers/kvm/Kconfig | 33 + drivers/kvm/Makefile | 10 + drivers/kvm/kvm.h | 551 +++++++++++ drivers/kvm/kvm_main.c | 1935 ++++++++++++++++++++++++++++++++++++ drivers/kvm/kvm_svm.h | 44 + drivers/kvm/kvm_vmx.h | 14 + drivers/kvm/mmu.c | 699 +++++++++++++ drivers/kvm/paging_tmpl.h | 397 ++++++++ drivers/kvm/segment_descriptor.h | 17 + drivers/kvm/svm.c | 1677 +++++++++++++++++++++++++++++++ drivers/kvm/svm.h | 315 ++++++ drivers/kvm/vmx.c | 2002 ++++++++++++++++++++++++++++++++++++++ drivers/kvm/vmx.h | 296 ++++++ drivers/kvm/x86_emulate.c | 1409 +++++++++++++++++++++++++++ drivers/kvm/x86_emulate.h | 185 ++++ include/linux/kvm.h | 227 +++++ 18 files changed, 9814 insertions(+) create mode 100644 drivers/kvm/Kconfig create mode 100644 drivers/kvm/Makefile create mode 100644 drivers/kvm/kvm.h create mode 100644 drivers/kvm/kvm_main.c create mode 100644 drivers/kvm/kvm_svm.h create mode 100644 drivers/kvm/kvm_vmx.h create mode 100644 drivers/kvm/mmu.c create mode 100644 drivers/kvm/paging_tmpl.h create mode 100644 drivers/kvm/segment_descriptor.h create mode 100644 drivers/kvm/svm.c create mode 100644 drivers/kvm/svm.h create mode 100644 drivers/kvm/vmx.c create mode 100644 drivers/kvm/vmx.h create mode 100644 drivers/kvm/x86_emulate.c create mode 100644 drivers/kvm/x86_emulate.h create mode 100644 include/linux/kvm.h (limited to 'include/linux') diff --git a/drivers/Kconfig b/drivers/Kconfig index 4929e923b5c6..e7da9fa724ec 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -80,4 +80,6 @@ source "drivers/rtc/Kconfig" source "drivers/dma/Kconfig" +source "drivers/kvm/Kconfig" + endmenu diff --git a/drivers/Makefile b/drivers/Makefile index 50f76da598c9..0dd96d1afd39 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -43,6 +43,7 @@ obj-$(CONFIG_SPI) += spi/ obj-$(CONFIG_PCCARD) += pcmcia/ obj-$(CONFIG_DIO) += dio/ obj-$(CONFIG_SBUS) += sbus/ +obj-$(CONFIG_KVM) += kvm/ obj-$(CONFIG_ZORRO) += zorro/ obj-$(CONFIG_MAC) += macintosh/ obj-$(CONFIG_ATA_OVER_ETH) += block/aoe/ diff --git a/drivers/kvm/Kconfig b/drivers/kvm/Kconfig new file mode 100644 index 000000000000..36412e90f09b --- /dev/null +++ b/drivers/kvm/Kconfig @@ -0,0 +1,33 @@ +# +# KVM configuration +# +config KVM + tristate "Kernel-based Virtual Machine (KVM) support" + depends on X86 && EXPERIMENTAL + ---help--- + Support hosting fully virtualized guest machines using hardware + virtualization extensions. You will need a fairly recent + processor equipped with virtualization extensions. You will also + need to select one or more of the processor modules below. + + This module provides access to the hardware capabilities through + a character device node named /dev/kvm. + + To compile this as a module, choose M here: the module + will be called kvm. + + If unsure, say N. + +config KVM_INTEL + tristate "KVM for Intel processors support" + depends on KVM + ---help--- + Provides support for KVM on Intel processors equipped with the VT + extensions. + +config KVM_AMD + tristate "KVM for AMD processors support" + depends on KVM + ---help--- + Provides support for KVM on AMD processors equipped with the AMD-V + (SVM) extensions. diff --git a/drivers/kvm/Makefile b/drivers/kvm/Makefile new file mode 100644 index 000000000000..c0a789fa9d65 --- /dev/null +++ b/drivers/kvm/Makefile @@ -0,0 +1,10 @@ +# +# Makefile for Kernel-based Virtual Machine module +# + +kvm-objs := kvm_main.o mmu.o x86_emulate.o +obj-$(CONFIG_KVM) += kvm.o +kvm-intel-objs = vmx.o +obj-$(CONFIG_KVM_INTEL) += kvm-intel.o +kvm-amd-objs = svm.o +obj-$(CONFIG_KVM_AMD) += kvm-amd.o diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h new file mode 100644 index 000000000000..5785d0870ab6 --- /dev/null +++ b/drivers/kvm/kvm.h @@ -0,0 +1,551 @@ +#ifndef __KVM_H +#define __KVM_H + +/* + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + */ + +#include +#include +#include +#include +#include + +#include "vmx.h" +#include + +#define CR0_PE_MASK (1ULL << 0) +#define CR0_TS_MASK (1ULL << 3) +#define CR0_NE_MASK (1ULL << 5) +#define CR0_WP_MASK (1ULL << 16) +#define CR0_NW_MASK (1ULL << 29) +#define CR0_CD_MASK (1ULL << 30) +#define CR0_PG_MASK (1ULL << 31) + +#define CR3_WPT_MASK (1ULL << 3) +#define CR3_PCD_MASK (1ULL << 4) + +#define CR3_RESEVED_BITS 0x07ULL +#define CR3_L_MODE_RESEVED_BITS (~((1ULL << 40) - 1) | 0x0fe7ULL) +#define CR3_FLAGS_MASK ((1ULL << 5) - 1) + +#define CR4_VME_MASK (1ULL << 0) +#define CR4_PSE_MASK (1ULL << 4) +#define CR4_PAE_MASK (1ULL << 5) +#define CR4_PGE_MASK (1ULL << 7) +#define CR4_VMXE_MASK (1ULL << 13) + +#define KVM_GUEST_CR0_MASK \ + (CR0_PG_MASK | CR0_PE_MASK | CR0_WP_MASK | CR0_NE_MASK \ + | CR0_NW_MASK | CR0_CD_MASK) +#define KVM_VM_CR0_ALWAYS_ON \ + (CR0_PG_MASK | CR0_PE_MASK | CR0_WP_MASK | CR0_NE_MASK) +#define KVM_GUEST_CR4_MASK \ + (CR4_PSE_MASK | CR4_PAE_MASK | CR4_PGE_MASK | CR4_VMXE_MASK | CR4_VME_MASK) +#define KVM_PMODE_VM_CR4_ALWAYS_ON (CR4_VMXE_MASK | CR4_PAE_MASK) +#define KVM_RMODE_VM_CR4_ALWAYS_ON (CR4_VMXE_MASK | CR4_PAE_MASK | CR4_VME_MASK) + +#define INVALID_PAGE (~(hpa_t)0) +#define UNMAPPED_GVA (~(gpa_t)0) + +#define KVM_MAX_VCPUS 1 +#define KVM_MEMORY_SLOTS 4 +#define KVM_NUM_MMU_PAGES 256 + +#define FX_IMAGE_SIZE 512 +#define FX_IMAGE_ALIGN 16 +#define FX_BUF_SIZE (2 * FX_IMAGE_SIZE + FX_IMAGE_ALIGN) + +#define DE_VECTOR 0 +#define DF_VECTOR 8 +#define TS_VECTOR 10 +#define NP_VECTOR 11 +#define SS_VECTOR 12 +#define GP_VECTOR 13 +#define PF_VECTOR 14 + +#define SELECTOR_TI_MASK (1 << 2) +#define SELECTOR_RPL_MASK 0x03 + +#define IOPL_SHIFT 12 + +/* + * Address types: + * + * gva - guest virtual address + * gpa - guest physical address + * gfn - guest frame number + * hva - host virtual address + * hpa - host physical address + * hfn - host frame number + */ + +typedef unsigned long gva_t; +typedef u64 gpa_t; +typedef unsigned long gfn_t; + +typedef unsigned long hva_t; +typedef u64 hpa_t; +typedef unsigned long hfn_t; + +struct kvm_mmu_page { + struct list_head link; + hpa_t page_hpa; + unsigned long slot_bitmap; /* One bit set per slot which has memory + * in this shadow page. + */ + int global; /* Set if all ptes in this page are global */ + u64 *parent_pte; +}; + +struct vmcs { + u32 revision_id; + u32 abort; + char data[0]; +}; + +#define vmx_msr_entry kvm_msr_entry + +struct kvm_vcpu; + +/* + * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level + * 32-bit). The kvm_mmu structure abstracts the details of the current mmu + * mode. + */ +struct kvm_mmu { + void (*new_cr3)(struct kvm_vcpu *vcpu); + int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err); + void (*inval_page)(struct kvm_vcpu *vcpu, gva_t gva); + void (*free)(struct kvm_vcpu *vcpu); + gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva); + hpa_t root_hpa; + int root_level; + int shadow_root_level; +}; + +struct kvm_guest_debug { + int enabled; + unsigned long bp[4]; + int singlestep; +}; + +enum { + VCPU_REGS_RAX = 0, + VCPU_REGS_RCX = 1, + VCPU_REGS_RDX = 2, + VCPU_REGS_RBX = 3, + VCPU_REGS_RSP = 4, + VCPU_REGS_RBP = 5, + VCPU_REGS_RSI = 6, + VCPU_REGS_RDI = 7, +#ifdef __x86_64__ + VCPU_REGS_R8 = 8, + VCPU_REGS_R9 = 9, + VCPU_REGS_R10 = 10, + VCPU_REGS_R11 = 11, + VCPU_REGS_R12 = 12, + VCPU_REGS_R13 = 13, + VCPU_REGS_R14 = 14, + VCPU_REGS_R15 = 15, +#endif + NR_VCPU_REGS +}; + +enum { + VCPU_SREG_CS, + VCPU_SREG_DS, + VCPU_SREG_ES, + VCPU_SREG_FS, + VCPU_SREG_GS, + VCPU_SREG_SS, + VCPU_SREG_TR, + VCPU_SREG_LDTR, +}; + +struct kvm_vcpu { + struct kvm *kvm; + union { + struct vmcs *vmcs; + struct vcpu_svm *svm; + }; + struct mutex mutex; + int cpu; + int launched; + unsigned long irq_summary; /* bit vector: 1 per word in irq_pending */ +#define NR_IRQ_WORDS KVM_IRQ_BITMAP_SIZE(unsigned long) + unsigned long irq_pending[NR_IRQ_WORDS]; + unsigned long regs[NR_VCPU_REGS]; /* for rsp: vcpu_load_rsp_rip() */ + unsigned long rip; /* needs vcpu_load_rsp_rip() */ + + unsigned long cr0; + unsigned long cr2; + unsigned long cr3; + unsigned long cr4; + unsigned long cr8; + u64 shadow_efer; + u64 apic_base; + int nmsrs; + struct vmx_msr_entry *guest_msrs; + struct vmx_msr_entry *host_msrs; + + struct list_head free_pages; + struct kvm_mmu_page page_header_buf[KVM_NUM_MMU_PAGES]; + struct kvm_mmu mmu; + + struct kvm_guest_debug guest_debug; + + char fx_buf[FX_BUF_SIZE]; + char *host_fx_image; + char *guest_fx_image; + + int mmio_needed; + int mmio_read_completed; + int mmio_is_write; + int mmio_size; + unsigned char mmio_data[8]; + gpa_t mmio_phys_addr; + + struct { + int active; + u8 save_iopl; + struct kvm_save_segment { + u16 selector; + unsigned long base; + u32 limit; + u32 ar; + } tr, es, ds, fs, gs; + } rmode; +}; + +struct kvm_memory_slot { + gfn_t base_gfn; + unsigned long npages; + unsigned long flags; + struct page **phys_mem; + unsigned long *dirty_bitmap; +}; + +struct kvm { + spinlock_t lock; /* protects everything except vcpus */ + int nmemslots; + struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS]; + struct list_head active_mmu_pages; + struct kvm_vcpu vcpus[KVM_MAX_VCPUS]; + int memory_config_version; + int busy; +}; + +struct kvm_stat { + u32 pf_fixed; + u32 pf_guest; + u32 tlb_flush; + u32 invlpg; + + u32 exits; + u32 io_exits; + u32 mmio_exits; + u32 signal_exits; + u32 irq_exits; +}; + +struct descriptor_table { + u16 limit; + unsigned long base; +} __attribute__((packed)); + +struct kvm_arch_ops { + int (*cpu_has_kvm_support)(void); /* __init */ + int (*disabled_by_bios)(void); /* __init */ + void (*hardware_enable)(void *dummy); /* __init */ + void (*hardware_disable)(void *dummy); + int (*hardware_setup)(void); /* __init */ + void (*hardware_unsetup)(void); /* __exit */ + + int (*vcpu_create)(struct kvm_vcpu *vcpu); + void (*vcpu_free)(struct kvm_vcpu *vcpu); + + struct kvm_vcpu *(*vcpu_load)(struct kvm_vcpu *vcpu); + void (*vcpu_put)(struct kvm_vcpu *vcpu); + + int (*set_guest_debug)(struct kvm_vcpu *vcpu, + struct kvm_debug_guest *dbg); + int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata); + int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); + u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg); + void (*get_segment)(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg); + void (*set_segment)(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg); + int (*is_long_mode)(struct kvm_vcpu *vcpu); + void (*get_cs_db_l_bits)(struct kvm_vcpu *vcpu, int *db, int *l); + void (*set_cr0)(struct kvm_vcpu *vcpu, unsigned long cr0); + void (*set_cr0_no_modeswitch)(struct kvm_vcpu *vcpu, + unsigned long cr0); + void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long cr3); + void (*set_cr4)(struct kvm_vcpu *vcpu, unsigned long cr4); + void (*set_efer)(struct kvm_vcpu *vcpu, u64 efer); + void (*get_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt); + void (*set_idt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt); + void (*get_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt); + void (*set_gdt)(struct kvm_vcpu *vcpu, struct descriptor_table *dt); + unsigned long (*get_dr)(struct kvm_vcpu *vcpu, int dr); + void (*set_dr)(struct kvm_vcpu *vcpu, int dr, unsigned long value, + int *exception); + void (*cache_regs)(struct kvm_vcpu *vcpu); + void (*decache_regs)(struct kvm_vcpu *vcpu); + unsigned long (*get_rflags)(struct kvm_vcpu *vcpu); + void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); + + void (*invlpg)(struct kvm_vcpu *vcpu, gva_t addr); + void (*tlb_flush)(struct kvm_vcpu *vcpu); + void (*inject_page_fault)(struct kvm_vcpu *vcpu, + unsigned long addr, u32 err_code); + + void (*inject_gp)(struct kvm_vcpu *vcpu, unsigned err_code); + + int (*run)(struct kvm_vcpu *vcpu, struct kvm_run *run); + int (*vcpu_setup)(struct kvm_vcpu *vcpu); + void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu); +}; + +extern struct kvm_stat kvm_stat; +extern struct kvm_arch_ops *kvm_arch_ops; + +#define kvm_printf(kvm, fmt ...) printk(KERN_DEBUG fmt) +#define vcpu_printf(vcpu, fmt...) kvm_printf(vcpu->kvm, fmt) + +int kvm_init_arch(struct kvm_arch_ops *ops, struct module *module); +void kvm_exit_arch(void); + +void kvm_mmu_destroy(struct kvm_vcpu *vcpu); +int kvm_mmu_init(struct kvm_vcpu *vcpu); + +int kvm_mmu_reset_context(struct kvm_vcpu *vcpu); +void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot); + +hpa_t gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa); +#define HPA_MSB ((sizeof(hpa_t) * 8) - 1) +#define HPA_ERR_MASK ((hpa_t)1 << HPA_MSB) +static inline int is_error_hpa(hpa_t hpa) { return hpa >> HPA_MSB; } +hpa_t gva_to_hpa(struct kvm_vcpu *vcpu, gva_t gva); + +void kvm_emulator_want_group7_invlpg(void); + +extern hpa_t bad_page_address; + +static inline struct page *gfn_to_page(struct kvm_memory_slot *slot, gfn_t gfn) +{ + return slot->phys_mem[gfn - slot->base_gfn]; +} + +struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn); +void mark_page_dirty(struct kvm *kvm, gfn_t gfn); + +enum emulation_result { + EMULATE_DONE, /* no further processing */ + EMULATE_DO_MMIO, /* kvm_run filled with mmio request */ + EMULATE_FAIL, /* can't emulate this instruction */ +}; + +int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run, + unsigned long cr2, u16 error_code); +void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); +void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); +void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw, + unsigned long *rflags); + +unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr); +void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long value, + unsigned long *rflags); + +struct x86_emulate_ctxt; + +int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address); +int emulate_clts(struct kvm_vcpu *vcpu); +int emulator_get_dr(struct x86_emulate_ctxt* ctxt, int dr, + unsigned long *dest); +int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, + unsigned long value); + +void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); +void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr0); +void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr0); +void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr0); +void lmsw(struct kvm_vcpu *vcpu, unsigned long msw); + +#ifdef __x86_64__ +void set_efer(struct kvm_vcpu *vcpu, u64 efer); +#endif + +void fx_init(struct kvm_vcpu *vcpu); + +void load_msrs(struct vmx_msr_entry *e, int n); +void save_msrs(struct vmx_msr_entry *e, int n); +void kvm_resched(struct kvm_vcpu *vcpu); + +int kvm_read_guest(struct kvm_vcpu *vcpu, + gva_t addr, + unsigned long size, + void *dest); + +int kvm_write_guest(struct kvm_vcpu *vcpu, + gva_t addr, + unsigned long size, + void *data); + +unsigned long segment_base(u16 selector); + +static inline struct page *_gfn_to_page(struct kvm *kvm, gfn_t gfn) +{ + struct kvm_memory_slot *slot = gfn_to_memslot(kvm, gfn); + return (slot) ? slot->phys_mem[gfn - slot->base_gfn] : NULL; +} + +static inline int is_pae(struct kvm_vcpu *vcpu) +{ + return vcpu->cr4 & CR4_PAE_MASK; +} + +static inline int is_pse(struct kvm_vcpu *vcpu) +{ + return vcpu->cr4 & CR4_PSE_MASK; +} + +static inline int is_paging(struct kvm_vcpu *vcpu) +{ + return vcpu->cr0 & CR0_PG_MASK; +} + +static inline int memslot_id(struct kvm *kvm, struct kvm_memory_slot *slot) +{ + return slot - kvm->memslots; +} + +static inline struct kvm_mmu_page *page_header(hpa_t shadow_page) +{ + struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT); + + return (struct kvm_mmu_page *)page->private; +} + +static inline u16 read_fs(void) +{ + u16 seg; + asm ("mov %%fs, %0" : "=g"(seg)); + return seg; +} + +static inline u16 read_gs(void) +{ + u16 seg; + asm ("mov %%gs, %0" : "=g"(seg)); + return seg; +} + +static inline u16 read_ldt(void) +{ + u16 ldt; + asm ("sldt %0" : "=g"(ldt)); + return ldt; +} + +static inline void load_fs(u16 sel) +{ + asm ("mov %0, %%fs" : : "rm"(sel)); +} + +static inline void load_gs(u16 sel) +{ + asm ("mov %0, %%gs" : : "rm"(sel)); +} + +#ifndef load_ldt +static inline void load_ldt(u16 sel) +{ + asm ("lldt %0" : : "g"(sel)); +} +#endif + +static inline void get_idt(struct descriptor_table *table) +{ + asm ("sidt %0" : "=m"(*table)); +} + +static inline void get_gdt(struct descriptor_table *table) +{ + asm ("sgdt %0" : "=m"(*table)); +} + +static inline unsigned long read_tr_base(void) +{ + u16 tr; + asm ("str %0" : "=g"(tr)); + return segment_base(tr); +} + +#ifdef __x86_64__ +static inline unsigned long read_msr(unsigned long msr) +{ + u64 value; + + rdmsrl(msr, value); + return value; +} +#endif + +static inline void fx_save(void *image) +{ + asm ("fxsave (%0)":: "r" (image)); +} + +static inline void fx_restore(void *image) +{ + asm ("fxrstor (%0)":: "r" (image)); +} + +static inline void fpu_init(void) +{ + asm ("finit"); +} + +static inline u32 get_rdx_init_val(void) +{ + return 0x600; /* P6 family */ +} + +#define ASM_VMX_VMCLEAR_RAX ".byte 0x66, 0x0f, 0xc7, 0x30" +#define ASM_VMX_VMLAUNCH ".byte 0x0f, 0x01, 0xc2" +#define ASM_VMX_VMRESUME ".byte 0x0f, 0x01, 0xc3" +#define ASM_VMX_VMPTRLD_RAX ".byte 0x0f, 0xc7, 0x30" +#define ASM_VMX_VMREAD_RDX_RAX ".byte 0x0f, 0x78, 0xd0" +#define ASM_VMX_VMWRITE_RAX_RDX ".byte 0x0f, 0x79, 0xd0" +#define ASM_VMX_VMWRITE_RSP_RDX ".byte 0x0f, 0x79, 0xd4" +#define ASM_VMX_VMXOFF ".byte 0x0f, 0x01, 0xc4" +#define ASM_VMX_VMXON_RAX ".byte 0xf3, 0x0f, 0xc7, 0x30" + +#define MSR_IA32_TIME_STAMP_COUNTER 0x010 + +#define TSS_IOPB_BASE_OFFSET 0x66 +#define TSS_BASE_SIZE 0x68 +#define TSS_IOPB_SIZE (65536 / 8) +#define TSS_REDIRECTION_SIZE (256 / 8) +#define RMODE_TSS_SIZE (TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1) + +#ifdef __x86_64__ + +/* + * When emulating 32-bit mode, cr3 is only 32 bits even on x86_64. Therefore + * we need to allocate shadow page tables in the first 4GB of memory, which + * happens to fit the DMA32 zone. + */ +#define GFP_KVM_MMU (GFP_KERNEL | __GFP_DMA32) + +#else + +#define GFP_KVM_MMU GFP_KERNEL + +#endif + +#endif diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c new file mode 100644 index 000000000000..b6b8a41b5ec8 --- /dev/null +++ b/drivers/kvm/kvm_main.c @@ -0,0 +1,1935 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * + * This module enables machines with Intel VT-x extensions to run virtual + * machines without emulation or binary translation. + * + * Copyright (C) 2006 Qumranet, Inc. + * + * Authors: + * Avi Kivity + * Yaniv Kamay + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include "kvm.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "x86_emulate.h" +#include "segment_descriptor.h" + +MODULE_AUTHOR("Qumranet"); +MODULE_LICENSE("GPL"); + +struct kvm_arch_ops *kvm_arch_ops; +struct kvm_stat kvm_stat; +EXPORT_SYMBOL_GPL(kvm_stat); + +static struct kvm_stats_debugfs_item { + const char *name; + u32 *data; + struct dentry *dentry; +} debugfs_entries[] = { + { "pf_fixed", &kvm_stat.pf_fixed }, + { "pf_guest", &kvm_stat.pf_guest }, + { "tlb_flush", &kvm_stat.tlb_flush }, + { "invlpg", &kvm_stat.invlpg }, + { "exits", &kvm_stat.exits }, + { "io_exits", &kvm_stat.io_exits }, + { "mmio_exits", &kvm_stat.mmio_exits }, + { "signal_exits", &kvm_stat.signal_exits }, + { "irq_exits", &kvm_stat.irq_exits }, + { 0, 0 } +}; + +static struct dentry *debugfs_dir; + +#define MAX_IO_MSRS 256 + +#define CR0_RESEVED_BITS 0xffffffff1ffaffc0ULL +#define LMSW_GUEST_MASK 0x0eULL +#define CR4_RESEVED_BITS (~((1ULL << 11) - 1)) +#define CR8_RESEVED_BITS (~0x0fULL) +#define EFER_RESERVED_BITS 0xfffffffffffff2fe + +struct vmx_msr_entry *find_msr_entry(struct kvm_vcpu *vcpu, u32 msr) +{ + int i; + + for (i = 0; i < vcpu->nmsrs; ++i) + if (vcpu->guest_msrs[i].index == msr) + return &vcpu->guest_msrs[i]; + return 0; +} +EXPORT_SYMBOL_GPL(find_msr_entry); + +#ifdef __x86_64__ +// LDT or TSS descriptor in the GDT. 16 bytes. +struct segment_descriptor_64 { + struct segment_descriptor s; + u32 base_higher; + u32 pad_zero; +}; + +#endif + +unsigned long segment_base(u16 selector) +{ + struct descriptor_table gdt; + struct segment_descriptor *d; + unsigned long table_base; + typedef unsigned long ul; + unsigned long v; + + if (selector == 0) + return 0; + + asm ("sgdt %0" : "=m"(gdt)); + table_base = gdt.base; + + if (selector & 4) { /* from ldt */ + u16 ldt_selector; + + asm ("sldt %0" : "=g"(ldt_selector)); + table_base = segment_base(ldt_selector); + } + d = (struct segment_descriptor *)(table_base + (selector & ~7)); + v = d->base_low | ((ul)d->base_mid << 16) | ((ul)d->base_high << 24); +#ifdef __x86_64__ + if (d->system == 0 + && (d->type == 2 || d->type == 9 || d->type == 11)) + v |= ((ul)((struct segment_descriptor_64 *)d)->base_higher) << 32; +#endif + return v; +} +EXPORT_SYMBOL_GPL(segment_base); + +int kvm_read_guest(struct kvm_vcpu *vcpu, + gva_t addr, + unsigned long size, + void *dest) +{ + unsigned char *host_buf = dest; + unsigned long req_size = size; + + while (size) { + hpa_t paddr; + unsigned now; + unsigned offset; + hva_t guest_buf; + + paddr = gva_to_hpa(vcpu, addr); + + if (is_error_hpa(paddr)) + break; + + guest_buf = (hva_t)kmap_atomic( + pfn_to_page(paddr >> PAGE_SHIFT), + KM_USER0); + offset = addr & ~PAGE_MASK; + guest_buf |= offset; + now = min(size, PAGE_SIZE - offset); + memcpy(host_buf, (void*)guest_buf, now); + host_buf += now; + addr += now; + size -= now; + kunmap_atomic((void *)(guest_buf & PAGE_MASK), KM_USER0); + } + return req_size - size; +} +EXPORT_SYMBOL_GPL(kvm_read_guest); + +int kvm_write_guest(struct kvm_vcpu *vcpu, + gva_t addr, + unsigned long size, + void *data) +{ + unsigned char *host_buf = data; + unsigned long req_size = size; + + while (size) { + hpa_t paddr; + unsigned now; + unsigned offset; + hva_t guest_buf; + + paddr = gva_to_hpa(vcpu, addr); + + if (is_error_hpa(paddr)) + break; + + guest_buf = (hva_t)kmap_atomic( + pfn_to_page(paddr >> PAGE_SHIFT), KM_USER0); + offset = addr & ~PAGE_MASK; + guest_buf |= offset; + now = min(size, PAGE_SIZE - offset); + memcpy((void*)guest_buf, host_buf, now); + host_buf += now; + addr += now; + size -= now; + kunmap_atomic((void *)(guest_buf & PAGE_MASK), KM_USER0); + } + return req_size - size; +} +EXPORT_SYMBOL_GPL(kvm_write_guest); + +static int vcpu_slot(struct kvm_vcpu *vcpu) +{ + return vcpu - vcpu->kvm->vcpus; +} + +/* + * Switches to specified vcpu, until a matching vcpu_put() + */ +static struct kvm_vcpu *vcpu_load(struct kvm *kvm, int vcpu_slot) +{ + struct kvm_vcpu *vcpu = &kvm->vcpus[vcpu_slot]; + + mutex_lock(&vcpu->mutex); + if (unlikely(!vcpu->vmcs)) { + mutex_unlock(&vcpu->mutex); + return 0; + } + return kvm_arch_ops->vcpu_load(vcpu); +} + +static void vcpu_put(struct kvm_vcpu *vcpu) +{ + kvm_arch_ops->vcpu_put(vcpu); + put_cpu(); + mutex_unlock(&vcpu->mutex); +} + +static int kvm_dev_open(struct inode *inode, struct file *filp) +{ + struct kvm *kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL); + int i; + + if (!kvm) + return -ENOMEM; + + spin_lock_init(&kvm->lock); + INIT_LIST_HEAD(&kvm->active_mmu_pages); + for (i = 0; i < KVM_MAX_VCPUS; ++i) { + struct kvm_vcpu *vcpu = &kvm->vcpus[i]; + + mutex_init(&vcpu->mutex); + vcpu->mmu.root_hpa = INVALID_PAGE; + INIT_LIST_HEAD(&vcpu->free_pages); + } + filp->private_data = kvm; + return 0; +} + +/* + * Free any memory in @free but not in @dont. + */ +static void kvm_free_physmem_slot(struct kvm_memory_slot *free, + struct kvm_memory_slot *dont) +{ + int i; + + if (!dont || free->phys_mem != dont->phys_mem) + if (free->phys_mem) { + for (i = 0; i < free->npages; ++i) + __free_page(free->phys_mem[i]); + vfree(free->phys_mem); + } + + if (!dont || free->dirty_bitmap != dont->dirty_bitmap) + vfree(free->dirty_bitmap); + + free->phys_mem = 0; + free->npages = 0; + free->dirty_bitmap = 0; +} + +static void kvm_free_physmem(struct kvm *kvm) +{ + int i; + + for (i = 0; i < kvm->nmemslots; ++i) + kvm_free_physmem_slot(&kvm->memslots[i], 0); +} + +static void kvm_free_vcpu(struct kvm_vcpu *vcpu) +{ + kvm_arch_ops->vcpu_free(vcpu); + kvm_mmu_destroy(vcpu); +} + +static void kvm_free_vcpus(struct kvm *kvm) +{ + unsigned int i; + + for (i = 0; i < KVM_MAX_VCPUS; ++i) + kvm_free_vcpu(&kvm->vcpus[i]); +} + +static int kvm_dev_release(struct inode *inode, struct file *filp) +{ + struct kvm *kvm = filp->private_data; + + kvm_free_vcpus(kvm); + kvm_free_physmem(kvm); + kfree(kvm); + return 0; +} + +static void inject_gp(struct kvm_vcpu *vcpu) +{ + kvm_arch_ops->inject_gp(vcpu, 0); +} + +static int pdptrs_have_reserved_bits_set(struct kvm_vcpu *vcpu, + unsigned long cr3) +{ + gfn_t pdpt_gfn = cr3 >> PAGE_SHIFT; + unsigned offset = (cr3 & (PAGE_SIZE-1)) >> 5; + int i; + u64 pdpte; + u64 *pdpt; + struct kvm_memory_slot *memslot; + + spin_lock(&vcpu->kvm->lock); + memslot = gfn_to_memslot(vcpu->kvm, pdpt_gfn); + /* FIXME: !memslot - emulate? 0xff? */ + pdpt = kmap_atomic(gfn_to_page(memslot, pdpt_gfn), KM_USER0); + + for (i = 0; i < 4; ++i) { + pdpte = pdpt[offset + i]; + if ((pdpte & 1) && (pdpte & 0xfffffff0000001e6ull)) + break; + } + + kunmap_atomic(pdpt, KM_USER0); + spin_unlock(&vcpu->kvm->lock); + + return i != 4; +} + +void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +{ + if (cr0 & CR0_RESEVED_BITS) { + printk(KERN_DEBUG "set_cr0: 0x%lx #GP, reserved bits 0x%lx\n", + cr0, vcpu->cr0); + inject_gp(vcpu); + return; + } + + if ((cr0 & CR0_NW_MASK) && !(cr0 & CR0_CD_MASK)) { + printk(KERN_DEBUG "set_cr0: #GP, CD == 0 && NW == 1\n"); + inject_gp(vcpu); + return; + } + + if ((cr0 & CR0_PG_MASK) && !(cr0 & CR0_PE_MASK)) { + printk(KERN_DEBUG "set_cr0: #GP, set PG flag " + "and a clear PE flag\n"); + inject_gp(vcpu); + return; + } + + if (!is_paging(vcpu) && (cr0 & CR0_PG_MASK)) { +#ifdef __x86_64__ + if ((vcpu->shadow_efer & EFER_LME)) { + int cs_db, cs_l; + + if (!is_pae(vcpu)) { + printk(KERN_DEBUG "set_cr0: #GP, start paging " + "in long mode while PAE is disabled\n"); + inject_gp(vcpu); + return; + } + kvm_arch_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); + if (cs_l) { + printk(KERN_DEBUG "set_cr0: #GP, start paging " + "in long mode while CS.L == 1\n"); + inject_gp(vcpu); + return; + + } + } else +#endif + if (is_pae(vcpu) && + pdptrs_have_reserved_bits_set(vcpu, vcpu->cr3)) { + printk(KERN_DEBUG "set_cr0: #GP, pdptrs " + "reserved bits\n"); + inject_gp(vcpu); + return; + } + + } + + kvm_arch_ops->set_cr0(vcpu, cr0); + vcpu->cr0 = cr0; + + spin_lock(&vcpu->kvm->lock); + kvm_mmu_reset_context(vcpu); + spin_unlock(&vcpu->kvm->lock); + return; +} +EXPORT_SYMBOL_GPL(set_cr0); + +void lmsw(struct kvm_vcpu *vcpu, unsigned long msw) +{ + set_cr0(vcpu, (vcpu->cr0 & ~0x0ful) | (msw & 0x0f)); +} +EXPORT_SYMBOL_GPL(lmsw); + +void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +{ + if (cr4 & CR4_RESEVED_BITS) { + printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n"); + inject_gp(vcpu); + return; + } + + if (kvm_arch_ops->is_long_mode(vcpu)) { + if (!(cr4 & CR4_PAE_MASK)) { + printk(KERN_DEBUG "set_cr4: #GP, clearing PAE while " + "in long mode\n"); + inject_gp(vcpu); + return; + } + } else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & CR4_PAE_MASK) + && pdptrs_have_reserved_bits_set(vcpu, vcpu->cr3)) { + printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n"); + inject_gp(vcpu); + } + + if (cr4 & CR4_VMXE_MASK) { + printk(KERN_DEBUG "set_cr4: #GP, setting VMXE\n"); + inject_gp(vcpu); + return; + } + kvm_arch_ops->set_cr4(vcpu, cr4); + spin_lock(&vcpu->kvm->lock); + kvm_mmu_reset_context(vcpu); + spin_unlock(&vcpu->kvm->lock); +} +EXPORT_SYMBOL_GPL(set_cr4); + +void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +{ + if (kvm_arch_ops->is_long_mode(vcpu)) { + if ( cr3 & CR3_L_MODE_RESEVED_BITS) { + printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n"); + inject_gp(vcpu); + return; + } + } else { + if (cr3 & CR3_RESEVED_BITS) { + printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n"); + inject_gp(vcpu); + return; + } + if (is_paging(vcpu) && is_pae(vcpu) && + pdptrs_have_reserved_bits_set(vcpu, cr3)) { + printk(KERN_DEBUG "set_cr3: #GP, pdptrs " + "reserved bits\n"); + inject_gp(vcpu); + return; + } + } + + vcpu->cr3 = cr3; + spin_lock(&vcpu->kvm->lock); + vcpu->mmu.new_cr3(vcpu); + spin_unlock(&vcpu->kvm->lock); +} +EXPORT_SYMBOL_GPL(set_cr3); + +void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) +{ + if ( cr8 & CR8_RESEVED_BITS) { + printk(KERN_DEBUG "set_cr8: #GP, reserved bits 0x%lx\n", cr8); + inject_gp(vcpu); + return; + } + vcpu->cr8 = cr8; +} +EXPORT_SYMBOL_GPL(set_cr8); + +void fx_init(struct kvm_vcpu *vcpu) +{ + struct __attribute__ ((__packed__)) fx_image_s { + u16 control; //fcw + u16 status; //fsw + u16 tag; // ftw + u16 opcode; //fop + u64 ip; // fpu ip + u64 operand;// fpu dp + u32 mxcsr; + u32 mxcsr_mask; + + } *fx_image; + + fx_save(vcpu->host_fx_image); + fpu_init(); + fx_save(vcpu->guest_fx_image); + fx_restore(vcpu->host_fx_image); + + fx_image = (struct fx_image_s *)vcpu->guest_fx_image; + fx_image->mxcsr = 0x1f80; + memset(vcpu->guest_fx_image + sizeof(struct fx_image_s), + 0, FX_IMAGE_SIZE - sizeof(struct fx_image_s)); +} +EXPORT_SYMBOL_GPL(fx_init); + +/* + * Creates some virtual cpus. Good luck creating more than one. + */ +static int kvm_dev_ioctl_create_vcpu(struct kvm *kvm, int n) +{ + int r; + struct kvm_vcpu *vcpu; + + r = -EINVAL; + if (n < 0 || n >= KVM_MAX_VCPUS) + goto out; + + vcpu = &kvm->vcpus[n]; + + mutex_lock(&vcpu->mutex); + + if (vcpu->vmcs) { + mutex_unlock(&vcpu->mutex); + return -EEXIST; + } + + vcpu->host_fx_image = (char*)ALIGN((hva_t)vcpu->fx_buf, + FX_IMAGE_ALIGN); + vcpu->guest_fx_image = vcpu->host_fx_image + FX_IMAGE_SIZE; + + vcpu->cpu = -1; /* First load will set up TR */ + vcpu->kvm = kvm; + r = kvm_arch_ops->vcpu_create(vcpu); + if (r < 0) + goto out_free_vcpus; + + kvm_arch_ops->vcpu_load(vcpu); + + r = kvm_arch_ops->vcpu_setup(vcpu); + if (r >= 0) + r = kvm_mmu_init(vcpu); + + vcpu_put(vcpu); + + if (r < 0) + goto out_free_vcpus; + + return 0; + +out_free_vcpus: + kvm_free_vcpu(vcpu); + mutex_unlock(&vcpu->mutex); +out: + return r; +} + +/* + * Allocate some memory and give it an address in the guest physical address + * space. + * + * Discontiguous memory is allowed, mostly for framebuffers. + */ +static int kvm_dev_ioctl_set_memory_region(struct kvm *kvm, + struct kvm_memory_region *mem) +{ + int r; + gfn_t base_gfn; + unsigned long npages; + unsigned long i; + struct kvm_memory_slot *memslot; + struct kvm_memory_slot old, new; + int memory_config_version; + + r = -EINVAL; + /* General sanity checks */ + if (mem->memory_size & (PAGE_SIZE - 1)) + goto out; + if (mem->guest_phys_addr & (PAGE_SIZE - 1)) + goto out; + if (mem->slot >= KVM_MEMORY_SLOTS) + goto out; + if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr) + goto out; + + memslot = &kvm->memslots[mem->slot]; + base_gfn = mem->guest_phys_addr >> PAGE_SHIFT; + npages = mem->memory_size >> PAGE_SHIFT; + + if (!npages) + mem->flags &= ~KVM_MEM_LOG_DIRTY_PAGES; + +raced: + spin_lock(&kvm->lock); + + memory_config_version = kvm->memory_config_version; + new = old = *memslot; + + new.base_gfn = base_gfn; + new.npages = npages; + new.flags = mem->flags; + + /* Disallow changing a memory slot's size. */ + r = -EINVAL; + if (npages && old.npages && npages != old.npages) + goto out_unlock; + + /* Check for overlaps */ + r = -EEXIST; + for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { + struct kvm_memory_slot *s = &kvm->memslots[i]; + + if (s == memslot) + continue; + if (!((base_gfn + npages <= s->base_gfn) || + (base_gfn >= s->base_gfn + s->npages))) + goto out_unlock; + } + /* + * Do memory allocations outside lock. memory_config_version will + * detect any races. + */ + spin_unlock(&kvm->lock); + + /* Deallocate if slot is being removed */ + if (!npages) + new.phys_mem = 0; + + /* Free page dirty bitmap if unneeded */ + if (!(new.flags & KVM_MEM_LOG_DIRTY_PAGES)) + new.dirty_bitmap = 0; + + r = -ENOMEM; + + /* Allocate if a slot is being created */ + if (npages && !new.phys_mem) { + new.phys_mem = vmalloc(npages * sizeof(struct page *)); + + if (!new.phys_mem) + goto out_free; + + memset(new.phys_mem, 0, npages * sizeof(struct page *)); + for (i = 0; i < npages; ++i) { + new.phys_mem[i] = alloc_page(GFP_HIGHUSER + | __GFP_ZERO); + if (!new.phys_mem[i]) + goto out_free; + } + } + + /* Allocate page dirty bitmap if needed */ + if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) { + unsigned dirty_bytes = ALIGN(npages, BITS_PER_LONG) / 8; + + new.dirty_bitmap = vmalloc(dirty_bytes); + if (!new.dirty_bitmap) + goto out_free; + memset(new.dirty_bitmap, 0, dirty_bytes); + } + + spin_lock(&kvm->lock); + + if (memory_config_version != kvm->memory_config_version) { + spin_unlock(&kvm->lock); + kvm_free_physmem_slot(&new, &old); + goto raced; + } + + r = -EAGAIN; + if (kvm->busy) + goto out_unlock; + + if (mem->slot >= kvm->nmemslots) + kvm->nmemslots = mem->slot + 1; + + *memslot = new; + ++kvm->memory_config_version; + + spin_unlock(&kvm->lock); + + for (i = 0; i < KVM_MAX_VCPUS; ++i) { + struct kvm_vcpu *vcpu; + + vcpu = vcpu_load(kvm, i); + if (!vcpu) + continue; + kvm_mmu_reset_context(vcpu); + vcpu_put(vcpu); + } + + kvm_free_physmem_slot(&old, &new); + return 0; + +out_unlock: + spin_unlock(&kvm->lock); +out_free: + kvm_free_physmem_slot(&new, &old); +out: + return r; +} + +/* + * Get (and clear) the dirty memory log for a memory slot. + */ +static int kvm_dev_ioctl_get_dirty_log(struct kvm *kvm, + struct kvm_dirty_log *log) +{ + struct kvm_memory_slot *memslot; + int r, i; + int n; + unsigned long any = 0; + + spin_lock(&kvm->lock); + + /* + * Prevent changes to guest memory configuration even while the lock + * is not taken. + */ + ++kvm->busy; + spin_unlock(&kvm->lock); + r = -EINVAL; + if (log->slot >= KVM_MEMORY_SLOTS) + goto out; + + memslot = &kvm->memslots[log->slot]; + r = -ENOENT; + if (!memslot->dirty_bitmap) + goto out; + + n = ALIGN(memslot->npages, 8) / 8; + + for (i = 0; !any && i < n; ++i) + any = memslot->dirty_bitmap[i]; + + r = -EFAULT; + if (copy_to_user(log->dirty_bitmap, memslot->dirty_bitmap, n)) + goto out; + + + if (any) { + spin_lock(&kvm->lock); + kvm_mmu_slot_remove_write_access(kvm, log->slot); + spin_unlock(&kvm->lock); + memset(memslot->dirty_bitmap, 0, n); + for (i = 0; i < KVM_MAX_VCPUS; ++i) { + struct kvm_vcpu *vcpu = vcpu_load(kvm, i); + + if (!vcpu) + continue; + kvm_arch_ops->tlb_flush(vcpu); + vcpu_put(vcpu); + } + } + + r = 0; + +out: + spin_lock(&kvm->lock); + --kvm->busy; + spin_unlock(&kvm->lock); + return r; +} + +struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) +{ + int i; + + for (i = 0; i < kvm->nmemslots; ++i) { + struct kvm_memory_slot *memslot = &kvm->memslots[i]; + + if (gfn >= memslot->base_gfn + && gfn < memslot->base_gfn + memslot->npages) + return memslot; + } + return 0; +} +EXPORT_SYMBOL_GPL(gfn_to_memslot); + +void mark_page_dirty(struct kvm *kvm, gfn_t gfn) +{ + int i; + struct kvm_memory_slot *memslot = 0; + unsigned long rel_gfn; + + for (i = 0; i < kvm->nmemslots; ++i) { + memslot = &kvm->memslots[i]; + + if (gfn >= memslot->base_gfn + && gfn < memslot->base_gfn + memslot->npages) { + + if (!memslot || !memslot->dirty_bitmap) + return; + + rel_gfn = gfn - memslot->base_gfn; + + /* avoid RMW */ + if (!test_bit(rel_gfn, memslot->dirty_bitmap)) + set_bit(rel_gfn, memslot->dirty_bitmap); + return; + } + } +} + +static int emulator_read_std(unsigned long addr, + unsigned long *val, + unsigned int bytes, + struct x86_emulate_ctxt *ctxt) +{ + struct kvm_vcpu *vcpu = ctxt->vcpu; + void *data = val; + + while (bytes) { + gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr); + unsigned offset = addr & (PAGE_SIZE-1); + unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset); + unsigned long pfn; + struct kvm_memory_slot *memslot; + void *page; + + if (gpa == UNMAPPED_GVA) + return X86EMUL_PROPAGATE_FAULT; + pfn = gpa >> PAGE_SHIFT; + memslot = gfn_to_memslot(vcpu->kvm, pfn); + if (!memslot) + return X86EMUL_UNHANDLEABLE; + page = kmap_atomic(gfn_to_page(memslot, pfn), KM_USER0); + + memcpy(data, page + offset, tocopy); + + kunmap_atomic(page, KM_USER0); + + bytes -= tocopy; + data += tocopy; + addr += tocopy; + } + + return X86EMUL_CONTINUE; +} + +static int emulator_write_std(unsigned long addr, + unsigned long val, + unsigned int bytes, + struct x86_emulate_ctxt *ctxt) +{ + printk(KERN_ERR "emulator_write_std: addr %lx n %d\n", + addr, bytes); + return X86EMUL_UNHANDLEABLE; +} + +static int emulator_read_emulated(unsigned long addr, + unsigned long *val, + unsigned int bytes, + struct x86_emulate_ctxt *ctxt) +{ + struct kvm_vcpu *vcpu = ctxt->vcpu; + + if (vcpu->mmio_read_completed) { + memcpy(val, vcpu->mmio_data, bytes); + vcpu->mmio_read_completed = 0; + return X86EMUL_CONTINUE; + } else if (emulator_read_std(addr, val, bytes, ctxt) + == X86EMUL_CONTINUE) + return X86EMUL_CONTINUE; + else { + gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr); + if (gpa == UNMAPPED_GVA) + return vcpu_printf(vcpu, "not present\n"), X86EMUL_PROPAGATE_FAULT; + vcpu->mmio_needed = 1; + vcpu->mmio_phys_addr = gpa; + vcpu->mmio_size = bytes; + vcpu->mmio_is_write = 0; + + return X86EMUL_UNHANDLEABLE; + } +} + +static int emulator_write_emulated(unsigned long addr, + unsigned long val, + unsigned int bytes, + struct x86_emulate_ctxt *ctxt) +{ + struct kvm_vcpu *vcpu = ctxt->vcpu; + gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr); + + if (gpa == UNMAPPED_GVA) + return X86EMUL_PROPAGATE_FAULT; + + vcpu->mmio_needed = 1; + vcpu->mmio_phys_addr = gpa; + vcpu->mmio_size = bytes; + vcpu->mmio_is_write = 1; + memcpy(vcpu->mmio_data, &val, bytes); + + return X86EMUL_CONTINUE; +} + +static int emulator_cmpxchg_emulated(unsigned long addr, + unsigned long old, + unsigned long new, + unsigned int bytes, + struct x86_emulate_ctxt *ctxt) +{ + static int reported; + + if (!reported) { + reported = 1; + printk(KERN_WARNING "kvm: emulating exchange as write\n"); + } + return emulator_write_emulated(addr, new, bytes, ctxt); +} + +static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) +{ + return kvm_arch_ops->get_segment_base(vcpu, seg); +} + +int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address) +{ + spin_lock(&vcpu->kvm->lock); + vcpu->mmu.inval_page(vcpu, address); + spin_unlock(&vcpu->kvm->lock); + kvm_arch_ops->invlpg(vcpu, address); + return X86EMUL_CONTINUE; +} + +int emulate_clts(struct kvm_vcpu *vcpu) +{ + unsigned long cr0 = vcpu->cr0; + + cr0 &= ~CR0_TS_MASK; + kvm_arch_ops->set_cr0(vcpu, cr0); + return X86EMUL_CONTINUE; +} + +int emulator_get_dr(struct x86_emulate_ctxt* ctxt, int dr, unsigned long *dest) +{ + struct kvm_vcpu *vcpu = ctxt->vcpu; + + switch (dr) { + case 0 ... 3: + *dest = kvm_arch_ops->get_dr(vcpu, dr); + return X86EMUL_CONTINUE; + default: + printk(KERN_DEBUG "%s: unexpected dr %u\n", + __FUNCTION__, dr); + return X86EMUL_UNHANDLEABLE; + } +} + +int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) +{ + unsigned long mask = (ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U; + int exception; + + kvm_arch_ops->set_dr(ctxt->vcpu, dr, value & mask, &exception); + if (exception) { + /* FIXME: better handling */ + return X86EMUL_UNHANDLEABLE; + } + return X86EMUL_CONTINUE; +} + +static void report_emulation_failure(struct x86_emulate_ctxt *ctxt) +{ + static int reported; + u8 opcodes[4]; + unsigned long rip = ctxt->vcpu->rip; + unsigned long rip_linear; + + rip_linear = rip + get_segment_base(ctxt->vcpu, VCPU_SREG_CS); + + if (reported) + return; + + emulator_read_std(rip_linear, (void *)opcodes, 4, ctxt); + + printk(KERN_ERR "emulation failed but !mmio_needed?" + " rip %lx %02x %02x %02x %02x\n", + rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]); + reported = 1; +} + +struct x86_emulate_ops emulate_ops = { + .read_std = emulator_read_std, + .write_std = emulator_write_std, + .read_emulated = emulator_read_emulated, + .write_emulated = emulator_write_emulated, + .cmpxchg_emulated = emulator_cmpxchg_emulated, +}; + +int emulate_instruction(struct kvm_vcpu *vcpu, + struct kvm_run *run, + unsigned long cr2, + u16 error_code) +{ + struct x86_emulate_ctxt emulate_ctxt; + int r; + int cs_db, cs_l; + + kvm_arch_ops->cache_regs(vcpu); + + kvm_arch_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); + + emulate_ctxt.vcpu = vcpu; + emulate_ctxt.eflags = kvm_arch_ops->get_rflags(vcpu); + emulate_ctxt.cr2 = cr2; + emulate_ctxt.mode = (emulate_ctxt.eflags & X86_EFLAGS_VM) + ? X86EMUL_MODE_REAL : cs_l + ? X86EMUL_MODE_PROT64 : cs_db + ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16; + + if (emulate_ctxt.mode == X86EMUL_MODE_PROT64) { + emulate_ctxt.cs_base = 0; + emulate_ctxt.ds_base = 0; + emulate_ctxt.es_base = 0; + emulate_ctxt.ss_base = 0; + } else { + emulate_ctxt.cs_base = get_segment_base(vcpu, VCPU_SREG_CS); + emulate_ctxt.ds_base = get_segment_base(vcpu, VCPU_SREG_DS); + emulate_ctxt.es_base = get_segment_base(vcpu, VCPU_SREG_ES); + emulate_ctxt.ss_base = get_segment_base(vcpu, VCPU_SREG_SS); + } + + emulate_ctxt.gs_base = get_segment_base(vcpu, VCPU_SREG_GS); + emulate_ctxt.fs_base = get_segment_base(vcpu, VCPU_SREG_FS); + + vcpu->mmio_is_write = 0; + r = x86_emulate_memop(&emulate_ctxt, &emulate_ops); + + if ((r || vcpu->mmio_is_write) && run) { + run->mmio.phys_addr = vcpu->mmio_phys_addr; + memcpy(run->mmio.data, vcpu->mmio_data, 8); + run->mmio.len = vcpu->mmio_size; + run->mmio.is_write = vcpu->mmio_is_write; + } + + if (r) { + if (!vcpu->mmio_needed) { + report_emulation_failure(&emulate_ctxt); + return EMULATE_FAIL; + } + return EMULATE_DO_MMIO; + } + + kvm_arch_ops->decache_regs(vcpu); + kvm_arch_ops->set_rflags(vcpu, emulate_ctxt.eflags); + + if (vcpu->mmio_is_write) + return EMULATE_DO_MMIO; + + return EMULATE_DONE; +} +EXPORT_SYMBOL_GPL(emulate_instruction); + +static u64 mk_cr_64(u64 curr_cr, u32 new_val) +{ + return (curr_cr & ~((1ULL << 32) - 1)) | new_val; +} + +void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base) +{ + struct descriptor_table dt = { limit, base }; + + kvm_arch_ops->set_gdt(vcpu, &dt); +} + +void realmode_lidt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base) +{ + struct descriptor_table dt = { limit, base }; + + kvm_arch_ops->set_idt(vcpu, &dt); +} + +void realmode_lmsw(struct kvm_vcpu *vcpu, unsigned long msw, + unsigned long *rflags) +{ + lmsw(vcpu, msw); + *rflags = kvm_arch_ops->get_rflags(vcpu); +} + +unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr) +{ + switch (cr) { + case 0: + return vcpu->cr0; + case 2: + return vcpu->cr2; + case 3: + return vcpu->cr3; + case 4: + return vcpu->cr4; + default: + vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr); + return 0; + } +} + +void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val, + unsigned long *rflags) +{ + switch (cr) { + case 0: + set_cr0(vcpu, mk_cr_64(vcpu->cr0, val)); + *rflags = kvm_arch_ops->get_rflags(vcpu); + break; + case 2: + vcpu->cr2 = val; + break; + case 3: + set_cr3(vcpu, val); + break; + case 4: + set_cr4(vcpu, mk_cr_64(vcpu->cr4, val)); + break; + default: + vcpu_printf(vcpu, "%s: unexpected cr %u\n", __FUNCTION__, cr); + } +} + +/* + * Reads an msr value (of 'msr_index') into 'pdata'. + * Returns 0 on success, non-0 otherwise. + * Assumes vcpu_load() was already called. + */ +static int get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) +{ + return kvm_arch_ops->get_msr(vcpu, msr_index, pdata); +} + +#ifdef __x86_64__ + +void set_efer(struct kvm_vcpu *vcpu, u64 efer) +{ + struct vmx_msr_entry *msr; + + if (efer & EFER_RESERVED_BITS) { + printk(KERN_DEBUG "set_efer: 0x%llx #GP, reserved bits\n", + efer); + inject_gp(vcpu); + return; + } + + if (is_paging(vcpu) + && (vcpu->shadow_efer & EFER_LME) != (efer & EFER_LME)) { + printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n"); + inject_gp(vcpu); + return; + } + + efer &= ~EFER_LMA; + efer |= vcpu->shadow_efer & EFER_LMA; + + vcpu->shadow_efer = efer; + + msr = find_msr_entry(vcpu, MSR_EFER); + + if (!(efer & EFER_LMA)) + efer &= ~EFER_LME; + msr->data = efer; +} +EXPORT_SYMBOL_GPL(set_efer); + +#endif + +/* + * Writes msr value into into the appropriate "register". + * Returns 0 on success, non-0 otherwise. + * Assumes vcpu_load() was already called. + */ +static int set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) +{ + return kvm_arch_ops->set_msr(vcpu, msr_index, data); +} + +void kvm_resched(struct kvm_vcpu *vcpu) +{ + vcpu_put(vcpu); + cond_resched(); + /* Cannot fail - no vcpu unplug yet. */ + vcpu_load(vcpu->kvm, vcpu_slot(vcpu)); +} +EXPORT_SYMBOL_GPL(kvm_resched); + +void load_msrs(struct vmx_msr_entry *e, int n) +{ + int i; + + for (i = 0; i < n; ++i) + wrmsrl(e[i].index, e[i].data); +} +EXPORT_SYMBOL_GPL(load_msrs); + +void save_msrs(struct vmx_msr_entry *e, int n) +{ + int i; + + for (i = 0; i < n; ++i) + rdmsrl(e[i].index, e[i].data); +} +EXPORT_SYMBOL_GPL(save_msrs); + +static int kvm_dev_ioctl_run(struct kvm *kvm, struct kvm_run *kvm_run) +{ + struct kvm_vcpu *vcpu; + int r; + + if (kvm_run->vcpu < 0 || kvm_run->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + + vcpu = vcpu_load(kvm, kvm_run->vcpu); + if (!vcpu) + return -ENOENT; + + if (kvm_run->emulated) { + kvm_arch_ops->skip_emulated_instruction(vcpu); + kvm_run->emulated = 0; + } + + if (kvm_run->mmio_completed) { + memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8); + vcpu->mmio_read_completed = 1; + } + + vcpu->mmio_needed = 0; + + r = kvm_arch_ops->run(vcpu, kvm_run); + + vcpu_put(vcpu); + return r; +} + +static int kvm_dev_ioctl_get_regs(struct kvm *kvm, struct kvm_regs *regs) +{ + struct kvm_vcpu *vcpu; + + if (regs->vcpu < 0 || regs->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + + vcpu = vcpu_load(kvm, regs->vcpu); + if (!vcpu) + return -ENOENT; + + kvm_arch_ops->cache_regs(vcpu); + + regs->rax = vcpu->regs[VCPU_REGS_RAX]; + regs->rbx = vcpu->regs[VCPU_REGS_RBX]; + regs->rcx = vcpu->regs[VCPU_REGS_RCX]; + regs->rdx = vcpu->regs[VCPU_REGS_RDX]; + regs->rsi = vcpu->regs[VCPU_REGS_RSI]; + regs->rdi = vcpu->regs[VCPU_REGS_RDI]; + regs->rsp = vcpu->regs[VCPU_REGS_RSP]; + regs->rbp = vcpu->regs[VCPU_REGS_RBP]; +#ifdef __x86_64__ + regs->r8 = vcpu->regs[VCPU_REGS_R8]; + regs->r9 = vcpu->regs[VCPU_REGS_R9]; + regs->r10 = vcpu->regs[VCPU_REGS_R10]; + regs->r11 = vcpu->regs[VCPU_REGS_R11]; + regs->r12 = vcpu->regs[VCPU_REGS_R12]; + regs->r13 = vcpu->regs[VCPU_REGS_R13]; + regs->r14 = vcpu->regs[VCPU_REGS_R14]; + regs->r15 = vcpu->regs[VCPU_REGS_R15]; +#endif + + regs->rip = vcpu->rip; + regs->rflags = kvm_arch_ops->get_rflags(vcpu); + + /* + * Don't leak debug flags in case they were set for guest debugging + */ + if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep) + regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); + + vcpu_put(vcpu); + + return 0; +} + +static int kvm_dev_ioctl_set_regs(struct kvm *kvm, struct kvm_regs *regs) +{ + struct kvm_vcpu *vcpu; + + if (regs->vcpu < 0 || regs->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + + vcpu = vcpu_load(kvm, regs->vcpu); + if (!vcpu) + return -ENOENT; + + vcpu->regs[VCPU_REGS_RAX] = regs->rax; + vcpu->regs[VCPU_REGS_RBX] = regs->rbx; + vcpu->regs[VCPU_REGS_RCX] = regs->rcx; + vcpu->regs[VCPU_REGS_RDX] = regs->rdx; + vcpu->regs[VCPU_REGS_RSI] = regs->rsi; + vcpu->regs[VCPU_REGS_RDI] = regs->rdi; + vcpu->regs[VCPU_REGS_RSP] = regs->rsp; + vcpu->regs[VCPU_REGS_RBP] = regs->rbp; +#ifdef __x86_64__ + vcpu->regs[VCPU_REGS_R8] = regs->r8; + vcpu->regs[VCPU_REGS_R9] = regs->r9; + vcpu->regs[VCPU_REGS_R10] = regs->r10; + vcpu->regs[VCPU_REGS_R11] = regs->r11; + vcpu->regs[VCPU_REGS_R12] = regs->r12; + vcpu->regs[VCPU_REGS_R13] = regs->r13; + vcpu->regs[VCPU_REGS_R14] = regs->r14; + vcpu->regs[VCPU_REGS_R15] = regs->r15; +#endif + + vcpu->rip = regs->rip; + kvm_arch_ops->set_rflags(vcpu, regs->rflags); + + kvm_arch_ops->decache_regs(vcpu); + + vcpu_put(vcpu); + + return 0; +} + +static void get_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + return kvm_arch_ops->get_segment(vcpu, var, seg); +} + +static int kvm_dev_ioctl_get_sregs(struct kvm *kvm, struct kvm_sregs *sregs) +{ + struct kvm_vcpu *vcpu; + struct descriptor_table dt; + + if (sregs->vcpu < 0 || sregs->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + vcpu = vcpu_load(kvm, sregs->vcpu); + if (!vcpu) + return -ENOENT; + + get_segment(vcpu, &sregs->cs, VCPU_SREG_CS); + get_segment(vcpu, &sregs->ds, VCPU_SREG_DS); + get_segment(vcpu, &sregs->es, VCPU_SREG_ES); + get_segment(vcpu, &sregs->fs, VCPU_SREG_FS); + get_segment(vcpu, &sregs->gs, VCPU_SREG_GS); + get_segment(vcpu, &sregs->ss, VCPU_SREG_SS); + + get_segment(vcpu, &sregs->tr, VCPU_SREG_TR); + get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); + + kvm_arch_ops->get_idt(vcpu, &dt); + sregs->idt.limit = dt.limit; + sregs->idt.base = dt.base; + kvm_arch_ops->get_gdt(vcpu, &dt); + sregs->gdt.limit = dt.limit; + sregs->gdt.base = dt.base; + + sregs->cr0 = vcpu->cr0; + sregs->cr2 = vcpu->cr2; + sregs->cr3 = vcpu->cr3; + sregs->cr4 = vcpu->cr4; + sregs->cr8 = vcpu->cr8; + sregs->efer = vcpu->shadow_efer; + sregs->apic_base = vcpu->apic_base; + + memcpy(sregs->interrupt_bitmap, vcpu->irq_pending, + sizeof sregs->interrupt_bitmap); + + vcpu_put(vcpu); + + return 0; +} + +static void set_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + return kvm_arch_ops->set_segment(vcpu, var, seg); +} + +static int kvm_dev_ioctl_set_sregs(struct kvm *kvm, struct kvm_sregs *sregs) +{ + struct kvm_vcpu *vcpu; + int mmu_reset_needed = 0; + int i; + struct descriptor_table dt; + + if (sregs->vcpu < 0 || sregs->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + vcpu = vcpu_load(kvm, sregs->vcpu); + if (!vcpu) + return -ENOENT; + + set_segment(vcpu, &sregs->cs, VCPU_SREG_CS); + set_segment(vcpu, &sregs->ds, VCPU_SREG_DS); + set_segment(vcpu, &sregs->es, VCPU_SREG_ES); + set_segment(vcpu, &sregs->fs, VCPU_SREG_FS); + set_segment(vcpu, &sregs->gs, VCPU_SREG_GS); + set_segment(vcpu, &sregs->ss, VCPU_SREG_SS); + + set_segment(vcpu, &sregs->tr, VCPU_SREG_TR); + set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); + + dt.limit = sregs->idt.limit; + dt.base = sregs->idt.base; + kvm_arch_ops->set_idt(vcpu, &dt); + dt.limit = sregs->gdt.limit; + dt.base = sregs->gdt.base; + kvm_arch_ops->set_gdt(vcpu, &dt); + + vcpu->cr2 = sregs->cr2; + mmu_reset_needed |= vcpu->cr3 != sregs->cr3; + vcpu->cr3 = sregs->cr3; + + vcpu->cr8 = sregs->cr8; + + mmu_reset_needed |= vcpu->shadow_efer != sregs->efer; +#ifdef __x86_64__ + kvm_arch_ops->set_efer(vcpu, sregs->efer); +#endif + vcpu->apic_base = sregs->apic_base; + + mmu_reset_needed |= vcpu->cr0 != sregs->cr0; + kvm_arch_ops->set_cr0_no_modeswitch(vcpu, sregs->cr0); + + mmu_reset_needed |= vcpu->cr4 != sregs->cr4; + kvm_arch_ops->set_cr4(vcpu, sregs->cr4); + + if (mmu_reset_needed) + kvm_mmu_reset_context(vcpu); + + memcpy(vcpu->irq_pending, sregs->interrupt_bitmap, + sizeof vcpu->irq_pending); + vcpu->irq_summary = 0; + for (i = 0; i < NR_IRQ_WORDS; ++i) + if (vcpu->irq_pending[i]) + __set_bit(i, &vcpu->irq_summary); + + vcpu_put(vcpu); + + return 0; +} + +/* + * List of msr numbers which we expose to userspace through KVM_GET_MSRS + * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST. + */ +static u32 msrs_to_save[] = { + MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP, + MSR_K6_STAR, +#ifdef __x86_64__ + MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR, +#endif + MSR_IA32_TIME_STAMP_COUNTER, +}; + + +/* + * Adapt set_msr() to msr_io()'s calling convention + */ +static int do_set_msr(struct kvm_vcpu *vcpu, unsigned index, u64 *data) +{ + return set_msr(vcpu, index, *data); +} + +/* + * Read or write a bunch of msrs. All parameters are kernel addresses. + * + * @return number of msrs set successfully. + */ +static int __msr_io(struct kvm *kvm, struct kvm_msrs *msrs, + struct kvm_msr_entry *entries, + int (*do_msr)(struct kvm_vcpu *vcpu, + unsigned index, u64 *data)) +{ + struct kvm_vcpu *vcpu; + int i; + + if (msrs->vcpu < 0 || msrs->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + + vcpu = vcpu_load(kvm, msrs->vcpu); + if (!vcpu) + return -ENOENT; + + for (i = 0; i < msrs->nmsrs; ++i) + if (do_msr(vcpu, entries[i].index, &entries[i].data)) + break; + + vcpu_put(vcpu); + + return i; +} + +/* + * Read or write a bunch of msrs. Parameters are user addresses. + * + * @return number of msrs set successfully. + */ +static int msr_io(struct kvm *kvm, struct kvm_msrs __user *user_msrs, + int (*do_msr)(struct kvm_vcpu *vcpu, + unsigned index, u64 *data), + int writeback) +{ + struct kvm_msrs msrs; + struct kvm_msr_entry *entries; + int r, n; + unsigned size; + + r = -EFAULT; + if (copy_from_user(&msrs, user_msrs, sizeof msrs)) + goto out; + + r = -E2BIG; + if (msrs.nmsrs >= MAX_IO_MSRS) + goto out; + + r = -ENOMEM; + size = sizeof(struct kvm_msr_entry) * msrs.nmsrs; + entries = vmalloc(size); + if (!entries) + goto out; + + r = -EFAULT; + if (copy_from_user(entries, user_msrs->entries, size)) + goto out_free; + + r = n = __msr_io(kvm, &msrs, entries, do_msr); + if (r < 0) + goto out_free; + + r = -EFAULT; + if (writeback && copy_to_user(user_msrs->entries, entries, size)) + goto out_free; + + r = n; + +out_free: + vfree(entries); +out: + return r; +} + +/* + * Translate a guest virtual address to a guest physical address. + */ +static int kvm_dev_ioctl_translate(struct kvm *kvm, struct kvm_translation *tr) +{ + unsigned long vaddr = tr->linear_address; + struct kvm_vcpu *vcpu; + gpa_t gpa; + + vcpu = vcpu_load(kvm, tr->vcpu); + if (!vcpu) + return -ENOENT; + spin_lock(&kvm->lock); + gpa = vcpu->mmu.gva_to_gpa(vcpu, vaddr); + tr->physical_address = gpa; + tr->valid = gpa != UNMAPPED_GVA; + tr->writeable = 1; + tr->usermode = 0; + spin_unlock(&kvm->lock); + vcpu_put(vcpu); + + return 0; +} + +static int kvm_dev_ioctl_interrupt(struct kvm *kvm, struct kvm_interrupt *irq) +{ + struct kvm_vcpu *vcpu; + + if (irq->vcpu < 0 || irq->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + if (irq->irq < 0 || irq->irq >= 256) + return -EINVAL; + vcpu = vcpu_load(kvm, irq->vcpu); + if (!vcpu) + return -ENOENT; + + set_bit(irq->irq, vcpu->irq_pending); + set_bit(irq->irq / BITS_PER_LONG, &vcpu->irq_summary); + + vcpu_put(vcpu); + + return 0; +} + +static int kvm_dev_ioctl_debug_guest(struct kvm *kvm, + struct kvm_debug_guest *dbg) +{ + struct kvm_vcpu *vcpu; + int r; + + if (dbg->vcpu < 0 || dbg->vcpu >= KVM_MAX_VCPUS) + return -EINVAL; + vcpu = vcpu_load(kvm, dbg->vcpu); + if (!vcpu) + return -ENOENT; + + r = kvm_arch_ops->set_guest_debug(vcpu, dbg); + + vcpu_put(vcpu); + + return r; +} + +static long kvm_dev_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm *kvm = filp->private_data; + int r = -EINVAL; + + switch (ioctl) { + case KVM_CREATE_VCPU: { + r = kvm_dev_ioctl_create_vcpu(kvm, arg); + if (r) + goto out; + break; + } + case KVM_RUN: { + struct kvm_run kvm_run; + + r = -EFAULT; + if (copy_from_user(&kvm_run, (void *)arg, sizeof kvm_run)) + goto out; + r = kvm_dev_ioctl_run(kvm, &kvm_run); + if (r < 0) + goto out; + r = -EFAULT; + if (copy_to_user((void *)arg, &kvm_run, sizeof kvm_run)) + goto out; + r = 0; + break; + } + case KVM_GET_REGS: { + struct kvm_regs kvm_regs; + + r = -EFAULT; + if (copy_from_user(&kvm_regs, (void *)arg, sizeof kvm_regs)) + goto out; + r = kvm_dev_ioctl_get_regs(kvm, &kvm_regs); + if (r) + goto out; + r = -EFAULT; + if (copy_to_user((void *)arg, &kvm_regs, sizeof kvm_regs)) + goto out; + r = 0; + break; + } + case KVM_SET_REGS: { + struct kvm_regs kvm_regs; + + r = -EFAULT; + if (copy_from_user(&kvm_regs, (void *)arg, sizeof kvm_regs)) + goto out; + r = kvm_dev_ioctl_set_regs(kvm, &kvm_regs); + if (r) + goto out; + r = 0; + break; + } + case KVM_GET_SREGS: { + struct kvm_sregs kvm_sregs; + + r = -EFAULT; + if (copy_from_user(&kvm_sregs, (void *)arg, sizeof kvm_sregs)) + goto out; + r = kvm_dev_ioctl_get_sregs(kvm, &kvm_sregs); + if (r) + goto out; + r = -EFAULT; + if (copy_to_user((void *)arg, &kvm_sregs, sizeof kvm_sregs)) + goto out; + r = 0; + break; + } + case KVM_SET_SREGS: { + struct kvm_sregs kvm_sregs; + + r = -EFAULT; + if (copy_from_user(&kvm_sregs, (void *)arg, sizeof kvm_sregs)) + goto out; + r = kvm_dev_ioctl_set_sregs(kvm, &kvm_sregs); + if (r) + goto out; + r = 0; + break; + } + case KVM_TRANSLATE: { + struct kvm_translation tr; + + r = -EFAULT; + if (copy_from_user(&tr, (void *)arg, sizeof tr)) + goto out; + r = kvm_dev_ioctl_translate(kvm, &tr); + if (r) + goto out; + r = -EFAULT; + if (copy_to_user((void *)arg, &tr, sizeof tr)) + goto out; + r = 0; + break; + } + case KVM_INTERRUPT: { + struct kvm_interrupt irq; + + r = -EFAULT; + if (copy_from_user(&irq, (void *)arg, sizeof irq)) + goto out; + r = kvm_dev_ioctl_interrupt(kvm, &irq); + if (r) + goto out; + r = 0; + break; + } + case KVM_DEBUG_GUEST: { + struct kvm_debug_guest dbg; + + r = -EFAULT; + if (copy_from_user(&dbg, (void *)arg, sizeof dbg)) + goto out; + r = kvm_dev_ioctl_debug_guest(kvm, &dbg); + if (r) + goto out; + r = 0; + break; + } + case KVM_SET_MEMORY_REGION: { + struct kvm_memory_region kvm_mem; + + r = -EFAULT; + if (copy_from_user(&kvm_mem, (void *)arg, sizeof kvm_mem)) + goto out; + r = kvm_dev_ioctl_set_memory_region(kvm, &kvm_mem); + if (r) + goto out; + break; + } + case KVM_GET_DIRTY_LOG: { + struct kvm_dirty_log log; + + r = -EFAULT; + if (copy_from_user(&log, (void *)arg, sizeof log)) + goto out; + r = kvm_dev_ioctl_get_dirty_log(kvm, &log); + if (r) + goto out; + break; + } + case KVM_GET_MSRS: + r = msr_io(kvm, (void __user *)arg, get_msr, 1); + break; + case KVM_SET_MSRS: + r = msr_io(kvm, (void __user *)arg, do_set_msr, 0); + break; + case KVM_GET_MSR_INDEX_LIST: { + struct kvm_msr_list __user *user_msr_list = (void __user *)arg; + struct kvm_msr_list msr_list; + unsigned n; + + r = -EFAULT; + if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list)) + goto out; + n = msr_list.nmsrs; + msr_list.nmsrs = ARRAY_SIZE(msrs_to_save); + if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list)) + goto out; + r = -E2BIG; + if (n < ARRAY_SIZE(msrs_to_save)) + goto out; + r = -EFAULT; + if (copy_to_user(user_msr_list->indices, &msrs_to_save, + sizeof msrs_to_save)) + goto out; + r = 0; + } + default: + ; + } +out: + return r; +} + +static struct page *kvm_dev_nopage(struct vm_area_struct *vma, + unsigned long address, + int *type) +{ + struct kvm *kvm = vma->vm_file->private_data; + unsigned long pgoff; + struct kvm_memory_slot *slot; + struct page *page; + + *type = VM_FAULT_MINOR; + pgoff = ((address - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + slot = gfn_to_memslot(kvm, pgoff); + if (!slot) + return NOPAGE_SIGBUS; + page = gfn_to_page(slot, pgoff); + if (!page) + return NOPAGE_SIGBUS; + get_page(page); + return page; +} + +static struct vm_operations_struct kvm_dev_vm_ops = { + .nopage = kvm_dev_nopage, +}; + +static int kvm_dev_mmap(struct file *file, struct vm_area_struct *vma) +{ + vma->vm_ops = &kvm_dev_vm_ops; + return 0; +} + +static struct file_operations kvm_chardev_ops = { + .open = kvm_dev_open, + .release = kvm_dev_release, + .unlocked_ioctl = kvm_dev_ioctl, + .compat_ioctl = kvm_dev_ioctl, + .mmap = kvm_dev_mmap, +}; + +static struct miscdevice kvm_dev = { + MISC_DYNAMIC_MINOR, + "kvm", + &kvm_chardev_ops, +}; + +static int kvm_reboot(struct notifier_block *notifier, unsigned long val, + void *v) +{ + if (val == SYS_RESTART) { + /* + * Some (well, at least mine) BIOSes hang on reboot if + * in vmx root mode. + */ + printk(KERN_INFO "kvm: exiting hardware virtualization\n"); + on_each_cpu(kvm_arch_ops->hardware_disable, 0, 0, 1); + } + return NOTIFY_OK; +} + +static struct notifier_block kvm_reboot_notifier = { + .notifier_call = kvm_reboot, + .priority = 0, +}; + +static __init void kvm_init_debug(void) +{ + struct kvm_stats_debugfs_item *p; + + debugfs_dir = debugfs_create_dir("kvm", 0); + for (p = debugfs_entries; p->name; ++p) + p->dentry = debugfs_create_u32(p->name, 0444, debugfs_dir, + p->data); +} + +static void kvm_exit_debug(void) +{ + struct kvm_stats_debugfs_item *p; + + for (p = debugfs_entries; p->name; ++p) + debugfs_remove(p->dentry); + debugfs_remove(debugfs_dir); +} + +hpa_t bad_page_address; + +int kvm_init_arch(struct kvm_arch_ops *ops, struct module *module) +{ + int r; + + kvm_arch_ops = ops; + + if (!kvm_arch_ops->cpu_has_kvm_support()) { + printk(KERN_ERR "kvm: no hardware support\n"); + return -EOPNOTSUPP; + } + if (kvm_arch_ops->disabled_by_bios()) { + printk(KERN_ERR "kvm: disabled by bios\n"); + return -EOPNOTSUPP; + } + + r = kvm_arch_ops->hardware_setup(); + if (r < 0) + return r; + + on_each_cpu(kvm_arch_ops->hardware_enable, 0, 0, 1); + register_reboot_notifier(&kvm_reboot_notifier); + + kvm_chardev_ops.owner = module; + + r = misc_register(&kvm_dev); + if (r) { + printk (KERN_ERR "kvm: misc device register failed\n"); + goto out_free; + } + + return r; + +out_free: + unregister_reboot_notifier(&kvm_reboot_notifier); + on_each_cpu(kvm_arch_ops->hardware_disable, 0, 0, 1); + kvm_arch_ops->hardware_unsetup(); + return r; +} + +void kvm_exit_arch(void) +{ + misc_deregister(&kvm_dev); + + unregister_reboot_notifier(&kvm_reboot_notifier); + on_each_cpu(kvm_arch_ops->hardware_disable, 0, 0, 1); + kvm_arch_ops->hardware_unsetup(); +} + +static __init int kvm_init(void) +{ + static struct page *bad_page; + int r = 0; + + kvm_init_debug(); + + if ((bad_page = alloc_page(GFP_KERNEL)) == NULL) { + r = -ENOMEM; + goto out; + } + + bad_page_address = page_to_pfn(bad_page) << PAGE_SHIFT; + memset(__va(bad_page_address), 0, PAGE_SIZE); + + return r; + +out: + kvm_exit_debug(); + return r; +} + +static __exit void kvm_exit(void) +{ + kvm_exit_debug(); + __free_page(pfn_to_page(bad_page_address >> PAGE_SHIFT)); +} + +module_init(kvm_init) +module_exit(kvm_exit) + +EXPORT_SYMBOL_GPL(kvm_init_arch); +EXPORT_SYMBOL_GPL(kvm_exit_arch); diff --git a/drivers/kvm/kvm_svm.h b/drivers/kvm/kvm_svm.h new file mode 100644 index 000000000000..7d7f2aa10960 --- /dev/null +++ b/drivers/kvm/kvm_svm.h @@ -0,0 +1,44 @@ +#ifndef __KVM_SVM_H +#define __KVM_SVM_H + +#include +#include +#include + +#include "svm.h" +#include "kvm.h" + +static const u32 host_save_msrs[] = { +#ifdef __x86_64__ + MSR_STAR, MSR_LSTAR, MSR_CSTAR, MSR_SYSCALL_MASK, MSR_KERNEL_GS_BASE, + MSR_FS_BASE, MSR_GS_BASE, +#endif + MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP, + MSR_IA32_DEBUGCTLMSR, /*MSR_IA32_LASTBRANCHFROMIP, + MSR_IA32_LASTBRANCHTOIP, MSR_IA32_LASTINTFROMIP,MSR_IA32_LASTINTTOIP,*/ +}; + +#define NR_HOST_SAVE_MSRS (sizeof(host_save_msrs) / sizeof(*host_save_msrs)) +#define NUM_DB_REGS 4 + +struct vcpu_svm { + struct vmcb *vmcb; + unsigned long vmcb_pa; + struct svm_cpu_data *svm_data; + uint64_t asid_generation; + + unsigned long cr0; + unsigned long cr4; + unsigned long db_regs[NUM_DB_REGS]; + + u64 next_rip; + + u64 host_msrs[NR_HOST_SAVE_MSRS]; + unsigned long host_cr2; + unsigned long host_db_regs[NUM_DB_REGS]; + unsigned long host_dr6; + unsigned long host_dr7; +}; + +#endif + diff --git a/drivers/kvm/kvm_vmx.h b/drivers/kvm/kvm_vmx.h new file mode 100644 index 000000000000..87e12d2bfa16 --- /dev/null +++ b/drivers/kvm/kvm_vmx.h @@ -0,0 +1,14 @@ +#ifndef __KVM_VMX_H +#define __KVM_VMX_H + +#ifdef __x86_64__ +/* + * avoid save/load MSR_SYSCALL_MASK and MSR_LSTAR by std vt + * mechanism (cpu bug AA24) + */ +#define NR_BAD_MSRS 2 +#else +#define NR_BAD_MSRS 0 +#endif + +#endif diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c new file mode 100644 index 000000000000..4e29d9b7211c --- /dev/null +++ b/drivers/kvm/mmu.c @@ -0,0 +1,699 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * + * This module enables machines with Intel VT-x extensions to run virtual + * machines without emulation or binary translation. + * + * MMU support + * + * Copyright (C) 2006 Qumranet, Inc. + * + * Authors: + * Yaniv Kamay + * Avi Kivity + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ +#include +#include +#include +#include +#include +#include + +#include "vmx.h" +#include "kvm.h" + +#define pgprintk(x...) do { } while (0) + +#define ASSERT(x) \ + if (!(x)) { \ + printk(KERN_WARNING "assertion failed %s:%d: %s\n", \ + __FILE__, __LINE__, #x); \ + } + +#define PT64_ENT_PER_PAGE 512 +#define PT32_ENT_PER_PAGE 1024 + +#define PT_WRITABLE_SHIFT 1 + +#define PT_PRESENT_MASK (1ULL << 0) +#define PT_WRITABLE_MASK (1ULL << PT_WRITABLE_SHIFT) +#define PT_USER_MASK (1ULL << 2) +#define PT_PWT_MASK (1ULL << 3) +#define PT_PCD_MASK (1ULL << 4) +#define PT_ACCESSED_MASK (1ULL << 5) +#define PT_DIRTY_MASK (1ULL << 6) +#define PT_PAGE_SIZE_MASK (1ULL << 7) +#define PT_PAT_MASK (1ULL << 7) +#define PT_GLOBAL_MASK (1ULL << 8) +#define PT64_NX_MASK (1ULL << 63) + +#define PT_PAT_SHIFT 7 +#define PT_DIR_PAT_SHIFT 12 +#define PT_DIR_PAT_MASK (1ULL << PT_DIR_PAT_SHIFT) + +#define PT32_DIR_PSE36_SIZE 4 +#define PT32_DIR_PSE36_SHIFT 13 +#define PT32_DIR_PSE36_MASK (((1ULL << PT32_DIR_PSE36_SIZE) - 1) << PT32_DIR_PSE36_SHIFT) + + +#define PT32_PTE_COPY_MASK \ + (PT_PRESENT_MASK | PT_PWT_MASK | PT_PCD_MASK | \ + PT_ACCESSED_MASK | PT_DIRTY_MASK | PT_PAT_MASK | \ + PT_GLOBAL_MASK ) + +#define PT32_NON_PTE_COPY_MASK \ + (PT_PRESENT_MASK | PT_PWT_MASK | PT_PCD_MASK | \ + PT_ACCESSED_MASK | PT_DIRTY_MASK) + + +#define PT64_PTE_COPY_MASK \ + (PT64_NX_MASK | PT32_PTE_COPY_MASK) + +#define PT64_NON_PTE_COPY_MASK \ + (PT64_NX_MASK | PT32_NON_PTE_COPY_MASK) + + + +#define PT_FIRST_AVAIL_BITS_SHIFT 9 +#define PT64_SECOND_AVAIL_BITS_SHIFT 52 + +#define PT_SHADOW_PS_MARK (1ULL << PT_FIRST_AVAIL_BITS_SHIFT) +#define PT_SHADOW_IO_MARK (1ULL << PT_FIRST_AVAIL_BITS_SHIFT) + +#define PT_SHADOW_WRITABLE_SHIFT (PT_FIRST_AVAIL_BITS_SHIFT + 1) +#define PT_SHADOW_WRITABLE_MASK (1ULL << PT_SHADOW_WRITABLE_SHIFT) + +#define PT_SHADOW_USER_SHIFT (PT_SHADOW_WRITABLE_SHIFT + 1) +#define PT_SHADOW_USER_MASK (1ULL << (PT_SHADOW_USER_SHIFT)) + +#define PT_SHADOW_BITS_OFFSET (PT_SHADOW_WRITABLE_SHIFT - PT_WRITABLE_SHIFT) + +#define VALID_PAGE(x) ((x) != INVALID_PAGE) + +#define PT64_LEVEL_BITS 9 + +#define PT64_LEVEL_SHIFT(level) \ + ( PAGE_SHIFT + (level - 1) * PT64_LEVEL_BITS ) + +#define PT64_LEVEL_MASK(level) \ + (((1ULL << PT64_LEVEL_BITS) - 1) << PT64_LEVEL_SHIFT(level)) + +#define PT64_INDEX(address, level)\ + (((address) >> PT64_LEVEL_SHIFT(level)) & ((1 << PT64_LEVEL_BITS) - 1)) + + +#define PT32_LEVEL_BITS 10 + +#define PT32_LEVEL_SHIFT(level) \ + ( PAGE_SHIFT + (level - 1) * PT32_LEVEL_BITS ) + +#define PT32_LEVEL_MASK(level) \ + (((1ULL << PT32_LEVEL_BITS) - 1) << PT32_LEVEL_SHIFT(level)) + +#define PT32_INDEX(address, level)\ + (((address) >> PT32_LEVEL_SHIFT(level)) & ((1 << PT32_LEVEL_BITS) - 1)) + + +#define PT64_BASE_ADDR_MASK (((1ULL << 52) - 1) & PAGE_MASK) +#define PT64_DIR_BASE_ADDR_MASK \ + (PT64_BASE_ADDR_MASK & ~((1ULL << (PAGE_SHIFT + PT64_LEVEL_BITS)) - 1)) + +#define PT32_BASE_ADDR_MASK PAGE_MASK +#define PT32_DIR_BASE_ADDR_MASK \ + (PAGE_MASK & ~((1ULL << (PAGE_SHIFT + PT32_LEVEL_BITS)) - 1)) + + +#define PFERR_PRESENT_MASK (1U << 0) +#define PFERR_WRITE_MASK (1U << 1) +#define PFERR_USER_MASK (1U << 2) + +#define PT64_ROOT_LEVEL 4 +#define PT32_ROOT_LEVEL 2 +#define PT32E_ROOT_LEVEL 3 + +#define PT_DIRECTORY_LEVEL 2 +#define PT_PAGE_TABLE_LEVEL 1 + +static int is_write_protection(struct kvm_vcpu *vcpu) +{ + return vcpu->cr0 & CR0_WP_MASK; +} + +static int is_cpuid_PSE36(void) +{ + return 1; +} + +static int is_present_pte(unsigned long pte) +{ + return pte & PT_PRESENT_MASK; +} + +static int is_writeble_pte(unsigned long pte) +{ + return pte & PT_WRITABLE_MASK; +} + +static int is_io_pte(unsigned long pte) +{ + return pte & PT_SHADOW_IO_MARK; +} + +static void kvm_mmu_free_page(struct kvm_vcpu *vcpu, hpa_t page_hpa) +{ + struct kvm_mmu_page *page_head = page_header(page_hpa); + + list_del(&page_head->link); + page_head->page_hpa = page_hpa; + list_add(&page_head->link, &vcpu->free_pages); +} + +static int is_empty_shadow_page(hpa_t page_hpa) +{ + u32 *pos; + u32 *end; + for (pos = __va(page_hpa), end = pos + PAGE_SIZE / sizeof(u32); + pos != end; pos++) + if (*pos != 0) + return 0; + return 1; +} + +static hpa_t kvm_mmu_alloc_page(struct kvm_vcpu *vcpu, u64 *parent_pte) +{ + struct kvm_mmu_page *page; + + if (list_empty(&vcpu->free_pages)) + return INVALID_PAGE; + + page = list_entry(vcpu->free_pages.next, struct kvm_mmu_page, link); + list_del(&page->link); + list_add(&page->link, &vcpu->kvm->active_mmu_pages); + ASSERT(is_empty_shadow_page(page->page_hpa)); + page->slot_bitmap = 0; + page->global = 1; + page->parent_pte = parent_pte; + return page->page_hpa; +} + +static void page_header_update_slot(struct kvm *kvm, void *pte, gpa_t gpa) +{ + int slot = memslot_id(kvm, gfn_to_memslot(kvm, gpa >> PAGE_SHIFT)); + struct kvm_mmu_page *page_head = page_header(__pa(pte)); + + __set_bit(slot, &page_head->slot_bitmap); +} + +hpa_t safe_gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa) +{ + hpa_t hpa = gpa_to_hpa(vcpu, gpa); + + return is_error_hpa(hpa) ? bad_page_address | (gpa & ~PAGE_MASK): hpa; +} + +hpa_t gpa_to_hpa(struct kvm_vcpu *vcpu, gpa_t gpa) +{ + struct kvm_memory_slot *slot; + struct page *page; + + ASSERT((gpa & HPA_ERR_MASK) == 0); + slot = gfn_to_memslot(vcpu->kvm, gpa >> PAGE_SHIFT); + if (!slot) + return gpa | HPA_ERR_MASK; + page = gfn_to_page(slot, gpa >> PAGE_SHIFT); + return ((hpa_t)page_to_pfn(page) << PAGE_SHIFT) + | (gpa & (PAGE_SIZE-1)); +} + +hpa_t gva_to_hpa(struct kvm_vcpu *vcpu, gva_t gva) +{ + gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva); + + if (gpa == UNMAPPED_GVA) + return UNMAPPED_GVA; + return gpa_to_hpa(vcpu, gpa); +} + + +static void release_pt_page_64(struct kvm_vcpu *vcpu, hpa_t page_hpa, + int level) +{ + ASSERT(vcpu); + ASSERT(VALID_PAGE(page_hpa)); + ASSERT(level <= PT64_ROOT_LEVEL && level > 0); + + if (level == 1) + memset(__va(page_hpa), 0, PAGE_SIZE); + else { + u64 *pos; + u64 *end; + + for (pos = __va(page_hpa), end = pos + PT64_ENT_PER_PAGE; + pos != end; pos++) { + u64 current_ent = *pos; + + *pos = 0; + if (is_present_pte(current_ent)) + release_pt_page_64(vcpu, + current_ent & + PT64_BASE_ADDR_MASK, + level - 1); + } + } + kvm_mmu_free_page(vcpu, page_hpa); +} + +static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) +{ +} + +static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, hpa_t p) +{ + int level = PT32E_ROOT_LEVEL; + hpa_t table_addr = vcpu->mmu.root_hpa; + + for (; ; level--) { + u32 index = PT64_INDEX(v, level); + u64 *table; + + ASSERT(VALID_PAGE(table_addr)); + table = __va(table_addr); + + if (level == 1) { + mark_page_dirty(vcpu->kvm, v >> PAGE_SHIFT); + page_header_update_slot(vcpu->kvm, table, v); + table[index] = p | PT_PRESENT_MASK | PT_WRITABLE_MASK | + PT_USER_MASK; + return 0; + } + + if (table[index] == 0) { + hpa_t new_table = kvm_mmu_alloc_page(vcpu, + &table[index]); + + if (!VALID_PAGE(new_table)) { + pgprintk("nonpaging_map: ENOMEM\n"); + return -ENOMEM; + } + + if (level == PT32E_ROOT_LEVEL) + table[index] = new_table | PT_PRESENT_MASK; + else + table[index] = new_table | PT_PRESENT_MASK | + PT_WRITABLE_MASK | PT_USER_MASK; + } + table_addr = table[index] & PT64_BASE_ADDR_MASK; + } +} + +static void nonpaging_flush(struct kvm_vcpu *vcpu) +{ + hpa_t root = vcpu->mmu.root_hpa; + + ++kvm_stat.tlb_flush; + pgprintk("nonpaging_flush\n"); + ASSERT(VALID_PAGE(root)); + release_pt_page_64(vcpu, root, vcpu->mmu.shadow_root_level); + root = kvm_mmu_alloc_page(vcpu, NULL); + ASSERT(VALID_PAGE(root)); + vcpu->mmu.root_hpa = root; + if (is_paging(vcpu)) + root |= (vcpu->cr3 & (CR3_PCD_MASK | CR3_WPT_MASK)); + kvm_arch_ops->set_cr3(vcpu, root); + kvm_arch_ops->tlb_flush(vcpu); +} + +static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr) +{ + return vaddr; +} + +static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva, + u32 error_code) +{ + int ret; + gpa_t addr = gva; + + ASSERT(vcpu); + ASSERT(VALID_PAGE(vcpu->mmu.root_hpa)); + + for (;;) { + hpa_t paddr; + + paddr = gpa_to_hpa(vcpu , addr & PT64_BASE_ADDR_MASK); + + if (is_error_hpa(paddr)) + return 1; + + ret = nonpaging_map(vcpu, addr & PAGE_MASK, paddr); + if (ret) { + nonpaging_flush(vcpu); + continue; + } + break; + } + return ret; +} + +static void nonpaging_inval_page(struct kvm_vcpu *vcpu, gva_t addr) +{ +} + +static void nonpaging_free(struct kvm_vcpu *vcpu) +{ + hpa_t root; + + ASSERT(vcpu); + root = vcpu->mmu.root_hpa; + if (VALID_PAGE(root)) + release_pt_page_64(vcpu, root, vcpu->mmu.shadow_root_level); + vcpu->mmu.root_hpa = INVALID_PAGE; +} + +static int nonpaging_init_context(struct kvm_vcpu *vcpu) +{ + struct kvm_mmu *context = &vcpu->mmu; + + context->new_cr3 = nonpaging_new_cr3; + context->page_fault = nonpaging_page_fault; + context->inval_page = nonpaging_inval_page; + context->gva_to_gpa = nonpaging_gva_to_gpa; + context->free = nonpaging_free; + context->root_level = PT32E_ROOT_LEVEL; + context->shadow_root_level = PT32E_ROOT_LEVEL; + context->root_hpa = kvm_mmu_alloc_page(vcpu, NULL); + ASSERT(VALID_PAGE(context->root_hpa)); + kvm_arch_ops->set_cr3(vcpu, context->root_hpa); + return 0; +} + + +static void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu) +{ + struct kvm_mmu_page *page, *npage; + + list_for_each_entry_safe(page, npage, &vcpu->kvm->active_mmu_pages, + link) { + if (page->global) + continue; + + if (!page->parent_pte) + continue; + + *page->parent_pte = 0; + release_pt_page_64(vcpu, page->page_hpa, 1); + } + ++kvm_stat.tlb_flush; + kvm_arch_ops->tlb_flush(vcpu); +} + +static void paging_new_cr3(struct kvm_vcpu *vcpu) +{ + kvm_mmu_flush_tlb(vcpu); +} + +static void mark_pagetable_nonglobal(void *shadow_pte) +{ + page_header(__pa(shadow_pte))->global = 0; +} + +static inline void set_pte_common(struct kvm_vcpu *vcpu, + u64 *shadow_pte, + gpa_t gaddr, + int dirty, + u64 access_bits) +{ + hpa_t paddr; + + *shadow_pte |= access_bits << PT_SHADOW_BITS_OFFSET; + if (!dirty) + access_bits &= ~PT_WRITABLE_MASK; + + if (access_bits & PT_WRITABLE_MASK) + mark_page_dirty(vcpu->kvm, gaddr >> PAGE_SHIFT); + + *shadow_pte |= access_bits; + + paddr = gpa_to_hpa(vcpu, gaddr & PT64_BASE_ADDR_MASK); + + if (!(*shadow_pte & PT_GLOBAL_MASK)) + mark_pagetable_nonglobal(shadow_pte); + + if (is_error_hpa(paddr)) { + *shadow_pte |= gaddr; + *shadow_pte |= PT_SHADOW_IO_MARK; + *shadow_pte &= ~PT_PRESENT_MASK; + } else { + *shadow_pte |= paddr; + page_header_update_slot(vcpu->kvm, shadow_pte, gaddr); + } +} + +static void inject_page_fault(struct kvm_vcpu *vcpu, + u64 addr, + u32 err_code) +{ + kvm_arch_ops->inject_page_fault(vcpu, addr, err_code); +} + +static inline int fix_read_pf(u64 *shadow_ent) +{ + if ((*shadow_ent & PT_SHADOW_USER_MASK) && + !(*shadow_ent & PT_USER_MASK)) { + /* + * If supervisor write protect is disabled, we shadow kernel + * pages as user pages so we can trap the write access. + */ + *shadow_ent |= PT_USER_MASK; + *shadow_ent &= ~PT_WRITABLE_MASK; + + return 1; + + } + return 0; +} + +static int may_access(u64 pte, int write, int user) +{ + + if (user && !(pte & PT_USER_MASK)) + return 0; + if (write && !(pte & PT_WRITABLE_MASK)) + return 0; + return 1; +} + +/* + * Remove a shadow pte. + */ +static void paging_inval_page(struct kvm_vcpu *vcpu, gva_t addr) +{ + hpa_t page_addr = vcpu->mmu.root_hpa; + int level = vcpu->mmu.shadow_root_level; + + ++kvm_stat.invlpg; + + for (; ; level--) { + u32 index = PT64_INDEX(addr, level); + u64 *table = __va(page_addr); + + if (level == PT_PAGE_TABLE_LEVEL ) { + table[index] = 0; + return; + } + + if (!is_present_pte(table[index])) + return; + + page_addr = table[index] & PT64_BASE_ADDR_MASK; + + if (level == PT_DIRECTORY_LEVEL && + (table[index] & PT_SHADOW_PS_MARK)) { + table[index] = 0; + release_pt_page_64(vcpu, page_addr, PT_PAGE_TABLE_LEVEL); + + kvm_arch_ops->tlb_flush(vcpu); + return; + } + } +} + +static void paging_free(struct kvm_vcpu *vcpu) +{ + nonpaging_free(vcpu); +} + +#define PTTYPE 64 +#include "paging_tmpl.h" +#undef PTTYPE + +#define PTTYPE 32 +#include "paging_tmpl.h" +#undef PTTYPE + +static int paging64_init_context(struct kvm_vcpu *vcpu) +{ + struct kvm_mmu *context = &vcpu->mmu; + + ASSERT(is_pae(vcpu)); + context->new_cr3 = paging_new_cr3; + context->page_fault = paging64_page_fault; + context->inval_page = paging_inval_page; + context->gva_to_gpa = paging64_gva_to_gpa; + context->free = paging_free; + context->root_level = PT64_ROOT_LEVEL; + context->shadow_root_level = PT64_ROOT_LEVEL; + context->root_hpa = kvm_mmu_alloc_page(vcpu, NULL); + ASSERT(VALID_PAGE(context->root_hpa)); + kvm_arch_ops->set_cr3(vcpu, context->root_hpa | + (vcpu->cr3 & (CR3_PCD_MASK | CR3_WPT_MASK))); + return 0; +} + +static int paging32_init_context(struct kvm_vcpu *vcpu) +{ + struct kvm_mmu *context = &vcpu->mmu; + + context->new_cr3 = paging_new_cr3; + context->page_fault = paging32_page_fault; + context->inval_page = paging_inval_page; + context->gva_to_gpa = paging32_gva_to_gpa; + context->free = paging_free; + context->root_level = PT32_ROOT_LEVEL; + context->shadow_root_level = PT32E_ROOT_LEVEL; + context->root_hpa = kvm_mmu_alloc_page(vcpu, NULL); + ASSERT(VALID_PAGE(context->root_hpa)); + kvm_arch_ops->set_cr3(vcpu, context->root_hpa | + (vcpu->cr3 & (CR3_PCD_MASK | CR3_WPT_MASK))); + return 0; +} + +static int paging32E_init_context(struct kvm_vcpu *vcpu) +{ + int ret; + + if ((ret = paging64_init_context(vcpu))) + return ret; + + vcpu->mmu.root_level = PT32E_ROOT_LEVEL; + vcpu->mmu.shadow_root_level = PT32E_ROOT_LEVEL; + return 0; +} + +static int init_kvm_mmu(struct kvm_vcpu *vcpu) +{ + ASSERT(vcpu); + ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa)); + + if (!is_paging(vcpu)) + return nonpaging_init_context(vcpu); + else if (kvm_arch_ops->is_long_mode(vcpu)) + return paging64_init_context(vcpu); + else if (is_pae(vcpu)) + return paging32E_init_context(vcpu); + else + return paging32_init_context(vcpu); +} + +static void destroy_kvm_mmu(struct kvm_vcpu *vcpu) +{ + ASSERT(vcpu); + if (VALID_PAGE(vcpu->mmu.root_hpa)) { + vcpu->mmu.free(vcpu); + vcpu->mmu.root_hpa = INVALID_PAGE; + } +} + +int kvm_mmu_reset_context(struct kvm_vcpu *vcpu) +{ + destroy_kvm_mmu(vcpu); + return init_kvm_mmu(vcpu); +} + +static void free_mmu_pages(struct kvm_vcpu *vcpu) +{ + while (!list_empty(&vcpu->free_pages)) { + struct kvm_mmu_page *page; + + page = list_entry(vcpu->free_pages.next, + struct kvm_mmu_page, link); + list_del(&page->link); + __free_page(pfn_to_page(page->page_hpa >> PAGE_SHIFT)); + page->page_hpa = INVALID_PAGE; + } +} + +static int alloc_mmu_pages(struct kvm_vcpu *vcpu) +{ + int i; + + ASSERT(vcpu); + + for (i = 0; i < KVM_NUM_MMU_PAGES; i++) { + struct page *page; + struct kvm_mmu_page *page_header = &vcpu->page_header_buf[i]; + + INIT_LIST_HEAD(&page_header->link); + if ((page = alloc_page(GFP_KVM_MMU)) == NULL) + goto error_1; + page->private = (unsigned long)page_header; + page_header->page_hpa = (hpa_t)page_to_pfn(page) << PAGE_SHIFT; + memset(__va(page_header->page_hpa), 0, PAGE_SIZE); + list_add(&page_header->link, &vcpu->free_pages); + } + return 0; + +error_1: + free_mmu_pages(vcpu); + return -ENOMEM; +} + +int kvm_mmu_init(struct kvm_vcpu *vcpu) +{ + int r; + + ASSERT(vcpu); + ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa)); + ASSERT(list_empty(&vcpu->free_pages)); + + if ((r = alloc_mmu_pages(vcpu))) + return r; + + if ((r = init_kvm_mmu(vcpu))) { + free_mmu_pages(vcpu); + return r; + } + return 0; +} + +void kvm_mmu_destroy(struct kvm_vcpu *vcpu) +{ + ASSERT(vcpu); + + destroy_kvm_mmu(vcpu); + free_mmu_pages(vcpu); +} + +void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot) +{ + struct kvm_mmu_page *page; + + list_for_each_entry(page, &kvm->active_mmu_pages, link) { + int i; + u64 *pt; + + if (!test_bit(slot, &page->slot_bitmap)) + continue; + + pt = __va(page->page_hpa); + for (i = 0; i < PT64_ENT_PER_PAGE; ++i) + /* avoid RMW */ + if (pt[i] & PT_WRITABLE_MASK) + pt[i] &= ~PT_WRITABLE_MASK; + + } +} diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h new file mode 100644 index 000000000000..765c2e1a048e --- /dev/null +++ b/drivers/kvm/paging_tmpl.h @@ -0,0 +1,397 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * + * This module enables machines with Intel VT-x extensions to run virtual + * machines without emulation or binary translation. + * + * MMU support + * + * Copyright (C) 2006 Qumranet, Inc. + * + * Authors: + * Yaniv Kamay + * Avi Kivity + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +/* + * We need the mmu code to access both 32-bit and 64-bit guest ptes, + * so the code in this file is compiled twice, once per pte size. + */ + +#if PTTYPE == 64 + #define pt_element_t u64 + #define guest_walker guest_walker64 + #define FNAME(name) paging##64_##name + #define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK + #define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK + #define PT_INDEX(addr, level) PT64_INDEX(addr, level) + #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) + #define PT_LEVEL_MASK(level) PT64_LEVEL_MASK(level) + #define PT_PTE_COPY_MASK PT64_PTE_COPY_MASK + #define PT_NON_PTE_COPY_MASK PT64_NON_PTE_COPY_MASK +#elif PTTYPE == 32 + #define pt_element_t u32 + #define guest_walker guest_walker32 + #define FNAME(name) paging##32_##name + #define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK + #define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK + #define PT_INDEX(addr, level) PT32_INDEX(addr, level) + #define SHADOW_PT_INDEX(addr, level) PT64_INDEX(addr, level) + #define PT_LEVEL_MASK(level) PT32_LEVEL_MASK(level) + #define PT_PTE_COPY_MASK PT32_PTE_COPY_MASK + #define PT_NON_PTE_COPY_MASK PT32_NON_PTE_COPY_MASK +#else + #error Invalid PTTYPE value +#endif + +/* + * The guest_walker structure emulates the behavior of the hardware page + * table walker. + */ +struct guest_walker { + int level; + pt_element_t *table; + pt_element_t inherited_ar; +}; + +static void FNAME(init_walker)(struct guest_walker *walker, + struct kvm_vcpu *vcpu) +{ + hpa_t hpa; + struct kvm_memory_slot *slot; + + walker->level = vcpu->mmu.root_level; + slot = gfn_to_memslot(vcpu->kvm, + (vcpu->cr3 & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT); + hpa = safe_gpa_to_hpa(vcpu, vcpu->cr3 & PT64_BASE_ADDR_MASK); + walker->table = kmap_atomic(pfn_to_page(hpa >> PAGE_SHIFT), KM_USER0); + + ASSERT((!kvm_arch_ops->is_long_mode(vcpu) && is_pae(vcpu)) || + (vcpu->cr3 & ~(PAGE_MASK | CR3_FLAGS_MASK)) == 0); + + walker->table = (pt_element_t *)( (unsigned long)walker->table | + (unsigned long)(vcpu->cr3 & ~(PAGE_MASK | CR3_FLAGS_MASK)) ); + walker->inherited_ar = PT_USER_MASK | PT_WRITABLE_MASK; +} + +static void FNAME(release_walker)(struct guest_walker *walker) +{ + kunmap_atomic(walker->table, KM_USER0); +} + +static void FNAME(set_pte)(struct kvm_vcpu *vcpu, u64 guest_pte, + u64 *shadow_pte, u64 access_bits) +{ + ASSERT(*shadow_pte == 0); + access_bits &= guest_pte; + *shadow_pte = (guest_pte & PT_PTE_COPY_MASK); + set_pte_common(vcpu, shadow_pte, guest_pte & PT_BASE_ADDR_MASK, + guest_pte & PT_DIRTY_MASK, access_bits); +} + +static void FNAME(set_pde)(struct kvm_vcpu *vcpu, u64 guest_pde, + u64 *shadow_pte, u64 access_bits, + int index) +{ + gpa_t gaddr; + + ASSERT(*shadow_pte == 0); + access_bits &= guest_pde; + gaddr = (guest_pde & PT_DIR_BASE_ADDR_MASK) + PAGE_SIZE * index; + if (PTTYPE == 32 && is_cpuid_PSE36()) + gaddr |= (guest_pde & PT32_DIR_PSE36_MASK) << + (32 - PT32_DIR_PSE36_SHIFT); + *shadow_pte = (guest_pde & (PT_NON_PTE_COPY_MASK | PT_GLOBAL_MASK)) | + ((guest_pde & PT_DIR_PAT_MASK) >> + (PT_DIR_PAT_SHIFT - PT_PAT_SHIFT)); + set_pte_common(vcpu, shadow_pte, gaddr, + guest_pde & PT_DIRTY_MASK, access_bits); +} + +/* + * Fetch a guest pte from a specific level in the paging hierarchy. + */ +static pt_element_t *FNAME(fetch_guest)(struct kvm_vcpu *vcpu, + struct guest_walker *walker, + int level, + gva_t addr) +{ + + ASSERT(level > 0 && level <= walker->level); + + for (;;) { + int index = PT_INDEX(addr, walker->level); + hpa_t paddr; + + ASSERT(((unsigned long)walker->table & PAGE_MASK) == + ((unsigned long)&walker->table[index] & PAGE_MASK)); + if (level == walker->level || + !is_present_pte(walker->table[index]) || + (walker->level == PT_DIRECTORY_LEVEL && + (walker->table[index] & PT_PAGE_SIZE_MASK) && + (PTTYPE == 64 || is_pse(vcpu)))) + return &walker->table[index]; + if (walker->level != 3 || kvm_arch_ops->is_long_mode(vcpu)) + walker->inherited_ar &= walker->table[index]; + paddr = safe_gpa_to_hpa(vcpu, walker->table[index] & PT_BASE_ADDR_MASK); + kunmap_atomic(walker->table, KM_USER0); + walker->table = kmap_atomic(pfn_to_page(paddr >> PAGE_SHIFT), + KM_USER0); + --walker->level; + } +} + +/* + * Fetch a shadow pte for a specific level in the paging hierarchy. + */ +static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, + struct guest_walker *walker) +{ + hpa_t shadow_addr; + int level; + u64 *prev_shadow_ent = NULL; + + shadow_addr = vcpu->mmu.root_hpa; + level = vcpu->mmu.shadow_root_level; + + for (; ; level--) { + u32 index = SHADOW_PT_INDEX(addr, level); + u64 *shadow_ent = ((u64 *)__va(shadow_addr)) + index; + pt_element_t *guest_ent; + + if (is_present_pte(*shadow_ent) || is_io_pte(*shadow_ent)) { + if (level == PT_PAGE_TABLE_LEVEL) + return shadow_ent; + shadow_addr = *shadow_ent & PT64_BASE_ADDR_MASK; + prev_shadow_ent = shadow_ent; + continue; + } + + if (PTTYPE == 32 && level > PT32_ROOT_LEVEL) { + ASSERT(level == PT32E_ROOT_LEVEL); + guest_ent = FNAME(fetch_guest)(vcpu, walker, + PT32_ROOT_LEVEL, addr); + } else + guest_ent = FNAME(fetch_guest)(vcpu, walker, + level, addr); + + if (!is_present_pte(*guest_ent)) + return NULL; + + /* Don't set accessed bit on PAE PDPTRs */ + if (vcpu->mmu.root_level != 3 || walker->level != 3) + *guest_ent |= PT_ACCESSED_MASK; + + if (level == PT_PAGE_TABLE_LEVEL) { + + if (walker->level == PT_DIRECTORY_LEVEL) { + if (prev_shadow_ent) + *prev_shadow_ent |= PT_SHADOW_PS_MARK; + FNAME(set_pde)(vcpu, *guest_ent, shadow_ent, + walker->inherited_ar, + PT_INDEX(addr, PT_PAGE_TABLE_LEVEL)); + } else { + ASSERT(walker->level == PT_PAGE_TABLE_LEVEL); + FNAME(set_pte)(vcpu, *guest_ent, shadow_ent, walker->inherited_ar); + } + return shadow_ent; + } + + shadow_addr = kvm_mmu_alloc_page(vcpu, shadow_ent); + if (!VALID_PAGE(shadow_addr)) + return ERR_PTR(-ENOMEM); + if (!kvm_arch_ops->is_long_mode(vcpu) && level == 3) + *shadow_ent = shadow_addr | + (*guest_ent & (PT_PRESENT_MASK | PT_PWT_MASK | PT_PCD_MASK)); + else { + *shadow_ent = shadow_addr | + (*guest_ent & PT_NON_PTE_COPY_MASK); + *shadow_ent |= (PT_WRITABLE_MASK | PT_USER_MASK); + } + prev_shadow_ent = shadow_ent; + } +} + +/* + * The guest faulted for write. We need to + * + * - check write permissions + * - update the guest pte dirty bit + * - update our own dirty page tracking structures + */ +static int FNAME(fix_write_pf)(struct kvm_vcpu *vcpu, + u64 *shadow_ent, + struct guest_walker *walker, + gva_t addr, + int user) +{ + pt_element_t *guest_ent; + int writable_shadow; + gfn_t gfn; + + if (is_writeble_pte(*shadow_ent)) + return 0; + + writable_shadow = *shadow_ent & PT_SHADOW_WRITABLE_MASK; + if (user) { + /* + * User mode access. Fail if it's a kernel page or a read-only + * page. + */ + if (!(*shadow_ent & PT_SHADOW_USER_MASK) || !writable_shadow) + return 0; + ASSERT(*shadow_ent & PT_USER_MASK); + } else + /* + * Kernel mode access. Fail if it's a read-only page and + * supervisor write protection is enabled. + */ + if (!writable_shadow) { + if (is_write_protection(vcpu)) + return 0; + *shadow_ent &= ~PT_USER_MASK; + } + + guest_ent = FNAME(fetch_guest)(vcpu, walker, PT_PAGE_TABLE_LEVEL, addr); + + if (!is_present_pte(*guest_ent)) { + *shadow_ent = 0; + return 0; + } + + gfn = (*guest_ent & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT; + mark_page_dirty(vcpu->kvm, gfn); + *shadow_ent |= PT_WRITABLE_MASK; + *guest_ent |= PT_DIRTY_MASK; + + return 1; +} + +/* + * Page fault handler. There are several causes for a page fault: + * - there is no shadow pte for the guest pte + * - write access through a shadow pte marked read only so that we can set + * the dirty bit + * - write access to a shadow pte marked read only so we can update the page + * dirty bitmap, when userspace requests it + * - mmio access; in this case we will never install a present shadow pte + * - normal guest page fault due to the guest pte marked not present, not + * writable, or not executable + * + * Returns: 1 if we need to emulate the instruction, 0 otherwise + */ +static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, + u32 error_code) +{ + int write_fault = error_code & PFERR_WRITE_MASK; + int pte_present = error_code & PFERR_PRESENT_MASK; + int user_fault = error_code & PFERR_USER_MASK; + struct guest_walker walker; + u64 *shadow_pte; + int fixed; + + /* + * Look up the shadow pte for the faulting address. + */ + for (;;) { + FNAME(init_walker)(&walker, vcpu); + shadow_pte = FNAME(fetch)(vcpu, addr, &walker); + if (IS_ERR(shadow_pte)) { /* must be -ENOMEM */ + nonpaging_flush(vcpu); + FNAME(release_walker)(&walker); + continue; + } + break; + } + + /* + * The page is not mapped by the guest. Let the guest handle it. + */ + if (!shadow_pte) { + inject_page_fault(vcpu, addr, error_code); + FNAME(release_walker)(&walker); + return 0; + } + + /* + * Update the shadow pte. + */ + if (write_fault) + fixed = FNAME(fix_write_pf)(vcpu, shadow_pte, &walker, addr, + user_fault); + else + fixed = fix_read_pf(shadow_pte); + + FNAME(release_walker)(&walker); + + /* + * mmio: emulate if accessible, otherwise its a guest fault. + */ + if (is_io_pte(*shadow_pte)) { + if (may_access(*shadow_pte, write_fault, user_fault)) + return 1; + pgprintk("%s: io work, no access\n", __FUNCTION__); + inject_page_fault(vcpu, addr, + error_code | PFERR_PRESENT_MASK); + return 0; + } + + /* + * pte not present, guest page fault. + */ + if (pte_present && !fixed) { + inject_page_fault(vcpu, addr, error_code); + return 0; + } + + ++kvm_stat.pf_fixed; + + return 0; +} + +static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr) +{ + struct guest_walker walker; + pt_element_t guest_pte; + gpa_t gpa; + + FNAME(init_walker)(&walker, vcpu); + guest_pte = *FNAME(fetch_guest)(vcpu, &walker, PT_PAGE_TABLE_LEVEL, + vaddr); + FNAME(release_walker)(&walker); + + if (!is_present_pte(guest_pte)) + return UNMAPPED_GVA; + + if (walker.level == PT_DIRECTORY_LEVEL) { + ASSERT((guest_pte & PT_PAGE_SIZE_MASK)); + ASSERT(PTTYPE == 64 || is_pse(vcpu)); + + gpa = (guest_pte & PT_DIR_BASE_ADDR_MASK) | (vaddr & + (PT_LEVEL_MASK(PT_PAGE_TABLE_LEVEL) | ~PAGE_MASK)); + + if (PTTYPE == 32 && is_cpuid_PSE36()) + gpa |= (guest_pte & PT32_DIR_PSE36_MASK) << + (32 - PT32_DIR_PSE36_SHIFT); + } else { + gpa = (guest_pte & PT_BASE_ADDR_MASK); + gpa |= (vaddr & ~PAGE_MASK); + } + + return gpa; +} + +#undef pt_element_t +#undef guest_walker +#undef FNAME +#undef PT_BASE_ADDR_MASK +#undef PT_INDEX +#undef SHADOW_PT_INDEX +#undef PT_LEVEL_MASK +#undef PT_PTE_COPY_MASK +#undef PT_NON_PTE_COPY_MASK +#undef PT_DIR_BASE_ADDR_MASK diff --git a/drivers/kvm/segment_descriptor.h b/drivers/kvm/segment_descriptor.h new file mode 100644 index 000000000000..71fdf458619a --- /dev/null +++ b/drivers/kvm/segment_descriptor.h @@ -0,0 +1,17 @@ +struct segment_descriptor { + u16 limit_low; + u16 base_low; + u8 base_mid; + u8 type : 4; + u8 system : 1; + u8 dpl : 2; + u8 present : 1; + u8 limit_high : 4; + u8 avl : 1; + u8 long_mode : 1; + u8 default_op : 1; + u8 granularity : 1; + u8 base_high; +} __attribute__((packed)); + + diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c new file mode 100644 index 000000000000..a33a89c68138 --- /dev/null +++ b/drivers/kvm/svm.c @@ -0,0 +1,1677 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * + * AMD SVM support + * + * Copyright (C) 2006 Qumranet, Inc. + * + * Authors: + * Yaniv Kamay + * Avi Kivity + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include +#include +#include +#include + +#include "kvm_svm.h" +#include "x86_emulate.h" + +MODULE_AUTHOR("Qumranet"); +MODULE_LICENSE("GPL"); + +#define IOPM_ALLOC_ORDER 2 +#define MSRPM_ALLOC_ORDER 1 + +#define DB_VECTOR 1 +#define UD_VECTOR 6 +#define GP_VECTOR 13 + +#define DR7_GD_MASK (1 << 13) +#define DR6_BD_MASK (1 << 13) +#define CR4_DE_MASK (1UL << 3) + +#define SEG_TYPE_LDT 2 +#define SEG_TYPE_BUSY_TSS16 3 + +#define KVM_EFER_LMA (1 << 10) +#define KVM_EFER_LME (1 << 8) + +unsigned long iopm_base; +unsigned long msrpm_base; + +struct kvm_ldttss_desc { + u16 limit0; + u16 base0; + unsigned base1 : 8, type : 5, dpl : 2, p : 1; + unsigned limit1 : 4, zero0 : 3, g : 1, base2 : 8; + u32 base3; + u32 zero1; +} __attribute__((packed)); + +struct svm_cpu_data { + int cpu; + + uint64_t asid_generation; + uint32_t max_asid; + uint32_t next_asid; + struct kvm_ldttss_desc *tss_desc; + + struct page *save_area; +}; + +static DEFINE_PER_CPU(struct svm_cpu_data *, svm_data); + +struct svm_init_data { + int cpu; + int r; +}; + +static u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000}; + +#define NUM_MSR_MAPS (sizeof(msrpm_ranges) / sizeof(*msrpm_ranges)) +#define MSRS_RANGE_SIZE 2048 +#define MSRS_IN_RANGE (MSRS_RANGE_SIZE * 8 / 2) + +#define MAX_INST_SIZE 15 + +static unsigned get_addr_size(struct kvm_vcpu *vcpu) +{ + struct vmcb_save_area *sa = &vcpu->svm->vmcb->save; + u16 cs_attrib; + + if (!(sa->cr0 & CR0_PE_MASK) || (sa->rflags & X86_EFLAGS_VM)) + return 2; + + cs_attrib = sa->cs.attrib; + + return (cs_attrib & SVM_SELECTOR_L_MASK) ? 8 : + (cs_attrib & SVM_SELECTOR_DB_MASK) ? 4 : 2; +} + +static inline u8 pop_irq(struct kvm_vcpu *vcpu) +{ + int word_index = __ffs(vcpu->irq_summary); + int bit_index = __ffs(vcpu->irq_pending[word_index]); + int irq = word_index * BITS_PER_LONG + bit_index; + + clear_bit(bit_index, &vcpu->irq_pending[word_index]); + if (!vcpu->irq_pending[word_index]) + clear_bit(word_index, &vcpu->irq_summary); + return irq; +} + +static inline void push_irq(struct kvm_vcpu *vcpu, u8 irq) +{ + set_bit(irq, vcpu->irq_pending); + set_bit(irq / BITS_PER_LONG, &vcpu->irq_summary); +} + +static inline void clgi(void) +{ + asm volatile (SVM_CLGI); +} + +static inline void stgi(void) +{ + asm volatile (SVM_STGI); +} + +static inline void invlpga(unsigned long addr, u32 asid) +{ + asm volatile (SVM_INVLPGA :: "a"(addr), "c"(asid)); +} + +static inline unsigned long kvm_read_cr2(void) +{ + unsigned long cr2; + + asm volatile ("mov %%cr2, %0" : "=r" (cr2)); + return cr2; +} + +static inline void kvm_write_cr2(unsigned long val) +{ + asm volatile ("mov %0, %%cr2" :: "r" (val)); +} + +static inline unsigned long read_dr6(void) +{ + unsigned long dr6; + + asm volatile ("mov %%dr6, %0" : "=r" (dr6)); + return dr6; +} + +static inline void write_dr6(unsigned long val) +{ + asm volatile ("mov %0, %%dr6" :: "r" (val)); +} + +static inline unsigned long read_dr7(void) +{ + unsigned long dr7; + + asm volatile ("mov %%dr7, %0" : "=r" (dr7)); + return dr7; +} + +static inline void write_dr7(unsigned long val) +{ + asm volatile ("mov %0, %%dr7" :: "r" (val)); +} + +static inline int svm_is_long_mode(struct kvm_vcpu *vcpu) +{ + return vcpu->svm->vmcb->save.efer & KVM_EFER_LMA; +} + +static inline void force_new_asid(struct kvm_vcpu *vcpu) +{ + vcpu->svm->asid_generation--; +} + +static inline void flush_guest_tlb(struct kvm_vcpu *vcpu) +{ + force_new_asid(vcpu); +} + +static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) +{ + if (!(efer & KVM_EFER_LMA)) + efer &= ~KVM_EFER_LME; + + vcpu->svm->vmcb->save.efer = efer | MSR_EFER_SVME_MASK; + vcpu->shadow_efer = efer; +} + +static void svm_inject_gp(struct kvm_vcpu *vcpu, unsigned error_code) +{ + vcpu->svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | + SVM_EVTINJ_VALID_ERR | + SVM_EVTINJ_TYPE_EXEPT | + GP_VECTOR; + vcpu->svm->vmcb->control.event_inj_err = error_code; +} + +static void inject_ud(struct kvm_vcpu *vcpu) +{ + vcpu->svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | + SVM_EVTINJ_TYPE_EXEPT | + UD_VECTOR; +} + +static void inject_db(struct kvm_vcpu *vcpu) +{ + vcpu->svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | + SVM_EVTINJ_TYPE_EXEPT | + DB_VECTOR; +} + +static int is_page_fault(uint32_t info) +{ + info &= SVM_EVTINJ_VEC_MASK | SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID; + return info == (PF_VECTOR | SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_EXEPT); +} + +static int is_external_interrupt(u32 info) +{ + info &= SVM_EVTINJ_TYPE_MASK | SVM_EVTINJ_VALID; + return info == (SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR); +} + +static void skip_emulated_instruction(struct kvm_vcpu *vcpu) +{ + if (!vcpu->svm->next_rip) { + printk(KERN_DEBUG "%s: NOP\n", __FUNCTION__); + return; + } + if (vcpu->svm->next_rip - vcpu->svm->vmcb->save.rip > 15) { + printk(KERN_ERR "%s: ip 0x%llx next 0x%llx\n", + __FUNCTION__, + vcpu->svm->vmcb->save.rip, + vcpu->svm->next_rip); + } + + vcpu->rip = vcpu->svm->vmcb->save.rip = vcpu->svm->next_rip; + vcpu->svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK; +} + +static int has_svm(void) +{ + uint32_t eax, ebx, ecx, edx; + + if (current_cpu_data.x86_vendor != X86_VENDOR_AMD) { + printk(KERN_INFO "has_svm: not amd\n"); + return 0; + } + + cpuid(0x80000000, &eax, &ebx, &ecx, &edx); + if (eax < SVM_CPUID_FUNC) { + printk(KERN_INFO "has_svm: can't execute cpuid_8000000a\n"); + return 0; + } + + cpuid(0x80000001, &eax, &ebx, &ecx, &edx); + if (!(ecx & (1 << SVM_CPUID_FEATURE_SHIFT))) { + printk(KERN_DEBUG "has_svm: svm not available\n"); + return 0; + } + return 1; +} + +static void svm_hardware_disable(void *garbage) +{ + struct svm_cpu_data *svm_data + = per_cpu(svm_data, raw_smp_processor_id()); + + if (svm_data) { + uint64_t efer; + + wrmsrl(MSR_VM_HSAVE_PA, 0); + rdmsrl(MSR_EFER, efer); + wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK); + per_cpu(svm_data, raw_smp_processor_id()) = 0; + __free_page(svm_data->save_area); + kfree(svm_data); + } +} + +static void svm_hardware_enable(void *garbage) +{ + + struct svm_cpu_data *svm_data; + uint64_t efer; +#ifdef __x86_64__ + struct desc_ptr gdt_descr; +#else + struct Xgt_desc_struct gdt_descr; +#endif + struct desc_struct *gdt; + int me = raw_smp_processor_id(); + + if (!has_svm()) { + printk(KERN_ERR "svm_cpu_init: err EOPNOTSUPP on %d\n", me); + return; + } + svm_data = per_cpu(svm_data, me); + + if (!svm_data) { + printk(KERN_ERR "svm_cpu_init: svm_data is NULL on %d\n", + me); + return; + } + + svm_data->asid_generation = 1; + svm_data->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1; + svm_data->next_asid = svm_data->max_asid + 1; + + asm volatile ( "sgdt %0" : "=m"(gdt_descr) ); + gdt = (struct desc_struct *)gdt_descr.address; + svm_data->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS); + + rdmsrl(MSR_EFER, efer); + wrmsrl(MSR_EFER, efer | MSR_EFER_SVME_MASK); + + wrmsrl(MSR_VM_HSAVE_PA, + page_to_pfn(svm_data->save_area) << PAGE_SHIFT); +} + +static int svm_cpu_init(int cpu) +{ + struct svm_cpu_data *svm_data; + int r; + + svm_data = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL); + if (!svm_data) + return -ENOMEM; + svm_data->cpu = cpu; + svm_data->save_area = alloc_page(GFP_KERNEL); + r = -ENOMEM; + if (!svm_data->save_area) + goto err_1; + + per_cpu(svm_data, cpu) = svm_data; + + return 0; + +err_1: + kfree(svm_data); + return r; + +} + +static int set_msr_interception(u32 *msrpm, unsigned msr, + int read, int write) +{ + int i; + + for (i = 0; i < NUM_MSR_MAPS; i++) { + if (msr >= msrpm_ranges[i] && + msr < msrpm_ranges[i] + MSRS_IN_RANGE) { + u32 msr_offset = (i * MSRS_IN_RANGE + msr - + msrpm_ranges[i]) * 2; + + u32 *base = msrpm + (msr_offset / 32); + u32 msr_shift = msr_offset % 32; + u32 mask = ((write) ? 0 : 2) | ((read) ? 0 : 1); + *base = (*base & ~(0x3 << msr_shift)) | + (mask << msr_shift); + return 1; + } + } + printk(KERN_DEBUG "%s: not found 0x%x\n", __FUNCTION__, msr); + return 0; +} + +static __init int svm_hardware_setup(void) +{ + int cpu; + struct page *iopm_pages; + struct page *msrpm_pages; + void *msrpm_va; + int r; + + + iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER); + + if (!iopm_pages) + return -ENOMEM; + memset(page_address(iopm_pages), 0xff, + PAGE_SIZE * (1 << IOPM_ALLOC_ORDER)); + iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT; + + + msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER); + + r = -ENOMEM; + if (!msrpm_pages) + goto err_1; + + msrpm_va = page_address(msrpm_pages); + memset(msrpm_va, 0xff, PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER)); + msrpm_base = page_to_pfn(msrpm_pages) << PAGE_SHIFT; + +#ifdef __x86_64__ + set_msr_interception(msrpm_va, MSR_GS_BASE, 1, 1); + set_msr_interception(msrpm_va, MSR_FS_BASE, 1, 1); + set_msr_interception(msrpm_va, MSR_KERNEL_GS_BASE, 1, 1); + set_msr_interception(msrpm_va, MSR_STAR, 1, 1); + set_msr_interception(msrpm_va, MSR_LSTAR, 1, 1); + set_msr_interception(msrpm_va, MSR_CSTAR, 1, 1); + set_msr_interception(msrpm_va, MSR_SYSCALL_MASK, 1, 1); +#endif + set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_CS, 1, 1); + set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_ESP, 1, 1); + set_msr_interception(msrpm_va, MSR_IA32_SYSENTER_EIP, 1, 1); + + for_each_online_cpu(cpu) { + r = svm_cpu_init(cpu); + if (r) + goto err_2; + } + return 0; + +err_2: + __free_pages(msrpm_pages, MSRPM_ALLOC_ORDER); + msrpm_base = 0; +err_1: + __free_pages(iopm_pages, IOPM_ALLOC_ORDER); + iopm_base = 0; + return r; +} + +static __exit void svm_hardware_unsetup(void) +{ + __free_pages(pfn_to_page(msrpm_base >> PAGE_SHIFT), MSRPM_ALLOC_ORDER); + __free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), IOPM_ALLOC_ORDER); + iopm_base = msrpm_base = 0; +} + +static void init_seg(struct vmcb_seg *seg) +{ + seg->selector = 0; + seg->attrib = SVM_SELECTOR_P_MASK | SVM_SELECTOR_S_MASK | + SVM_SELECTOR_WRITE_MASK; /* Read/Write Data Segment */ + seg->limit = 0xffff; + seg->base = 0; +} + +static void init_sys_seg(struct vmcb_seg *seg, uint32_t type) +{ + seg->selector = 0; + seg->attrib = SVM_SELECTOR_P_MASK | type; + seg->limit = 0xffff; + seg->base = 0; +} + +static int svm_vcpu_setup(struct kvm_vcpu *vcpu) +{ + return 0; +} + +static void init_vmcb(struct vmcb *vmcb) +{ + struct vmcb_control_area *control = &vmcb->control; + struct vmcb_save_area *save = &vmcb->save; + u64 tsc; + + control->intercept_cr_read = INTERCEPT_CR0_MASK | + INTERCEPT_CR3_MASK | + INTERCEPT_CR4_MASK; + + control->intercept_cr_write = INTERCEPT_CR0_MASK | + INTERCEPT_CR3_MASK | + INTERCEPT_CR4_MASK; + + control->intercept_dr_read = INTERCEPT_DR0_MASK | + INTERCEPT_DR1_MASK | + INTERCEPT_DR2_MASK | + INTERCEPT_DR3_MASK; + + control->intercept_dr_write = INTERCEPT_DR0_MASK | + INTERCEPT_DR1_MASK | + INTERCEPT_DR2_MASK | + INTERCEPT_DR3_MASK | + INTERCEPT_DR5_MASK | + INTERCEPT_DR7_MASK; + + control->intercept_exceptions = 1 << PF_VECTOR; + + + control->intercept = (1ULL << INTERCEPT_INTR) | + (1ULL << INTERCEPT_NMI) | + /* + * selective cr0 intercept bug? + * 0: 0f 22 d8 mov %eax,%cr3 + * 3: 0f 20 c0 mov %cr0,%eax + * 6: 0d 00 00 00 80 or $0x80000000,%eax + * b: 0f 22 c0 mov %eax,%cr0 + * set cr3 ->interception + * get cr0 ->interception + * set cr0 -> no interception + */ + /* (1ULL << INTERCEPT_SELECTIVE_CR0) | */ + (1ULL << INTERCEPT_CPUID) | + (1ULL << INTERCEPT_HLT) | + (1ULL << INTERCEPT_INVLPG) | + (1ULL << INTERCEPT_INVLPGA) | + (1ULL << INTERCEPT_IOIO_PROT) | + (1ULL << INTERCEPT_MSR_PROT) | + (1ULL << INTERCEPT_TASK_SWITCH) | + (1ULL << INTERCEPT_VMRUN) | + (1ULL << INTERCEPT_VMMCALL) | + (1ULL << INTERCEPT_VMLOAD) | + (1ULL << INTERCEPT_VMSAVE) | + (1ULL << INTERCEPT_STGI) | + (1ULL << INTERCEPT_CLGI) | + (1ULL << INTERCEPT_SKINIT); + + control->iopm_base_pa = iopm_base; + control->msrpm_base_pa = msrpm_base; + rdtscll(tsc); + control->tsc_offset = -tsc; + control->int_ctl = V_INTR_MASKING_MASK; + + init_seg(&save->es); + init_seg(&save->ss); + init_seg(&save->ds); + init_seg(&save->fs); + init_seg(&save->gs); + + save->cs.selector = 0xf000; + /* Executable/Readable Code Segment */ + save->cs.attrib = SVM_SELECTOR_READ_MASK | SVM_SELECTOR_P_MASK | + SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK; + save->cs.limit = 0xffff; + save->cs.base = 0xffff0000; + + save->gdtr.limit = 0xffff; + save->idtr.limit = 0xffff; + + init_sys_seg(&save->ldtr, SEG_TYPE_LDT); + init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16); + + save->efer = MSR_EFER_SVME_MASK; + + save->dr6 = 0xffff0ff0; + save->dr7 = 0x400; + save->rflags = 2; + save->rip = 0x0000fff0; + + /* + * cr0 val on cpu init should be 0x60000010, we enable cpu + * cache by default. the orderly way is to enable cache in bios. + */ + save->cr0 = 0x00000010 | CR0_PG_MASK; + save->cr4 = CR4_PAE_MASK; + /* rdx = ?? */ +} + +static int svm_create_vcpu(struct kvm_vcpu *vcpu) +{ + struct page *page; + int r; + + r = -ENOMEM; + vcpu->svm = kzalloc(sizeof *vcpu->svm, GFP_KERNEL); + if (!vcpu->svm) + goto out1; + page = alloc_page(GFP_KERNEL); + if (!page) + goto out2; + + vcpu->svm->vmcb = page_address(page); + memset(vcpu->svm->vmcb, 0, PAGE_SIZE); + vcpu->svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT; + vcpu->svm->cr0 = 0x00000010; + vcpu->svm->asid_generation = 0; + memset(vcpu->svm->db_regs, 0, sizeof(vcpu->svm->db_regs)); + init_vmcb(vcpu->svm->vmcb); + + return 0; + +out2: + kfree(vcpu->svm); +out1: + return r; +} + +static void svm_free_vcpu(struct kvm_vcpu *vcpu) +{ + if (!vcpu->svm) + return; + if (vcpu->svm->vmcb) + __free_page(pfn_to_page(vcpu->svm->vmcb_pa >> PAGE_SHIFT)); + kfree(vcpu->svm); +} + +static struct kvm_vcpu *svm_vcpu_load(struct kvm_vcpu *vcpu) +{ + get_cpu(); + return vcpu; +} + +static void svm_vcpu_put(struct kvm_vcpu *vcpu) +{ + put_cpu(); +} + +static void svm_cache_regs(struct kvm_vcpu *vcpu) +{ + vcpu->regs[VCPU_REGS_RAX] = vcpu->svm->vmcb->save.rax; + vcpu->regs[VCPU_REGS_RSP] = vcpu->svm->vmcb->save.rsp; + vcpu->rip = vcpu->svm->vmcb->save.rip; +} + +static void svm_decache_regs(struct kvm_vcpu *vcpu) +{ + vcpu->svm->vmcb->save.rax = vcpu->regs[VCPU_REGS_RAX]; + vcpu->svm->vmcb->save.rsp = vcpu->regs[VCPU_REGS_RSP]; + vcpu->svm->vmcb->save.rip = vcpu->rip; +} + +static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu) +{ + return vcpu->svm->vmcb->save.rflags; +} + +static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) +{ + vcpu->svm->vmcb->save.rflags = rflags; +} + +static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg) +{ + struct vmcb_save_area *save = &vcpu->svm->vmcb->save; + + switch (seg) { + case VCPU_SREG_CS: return &save->cs; + case VCPU_SREG_DS: return &save->ds; + case VCPU_SREG_ES: return &save->es; + case VCPU_SREG_FS: return &save->fs; + case VCPU_SREG_GS: return &save->gs; + case VCPU_SREG_SS: return &save->ss; + case VCPU_SREG_TR: return &save->tr; + case VCPU_SREG_LDTR: return &save->ldtr; + } + BUG(); + return 0; +} + +static u64 svm_get_segment_base(struct kvm_vcpu *vcpu, int seg) +{ + struct vmcb_seg *s = svm_seg(vcpu, seg); + + return s->base; +} + +static void svm_get_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + struct vmcb_seg *s = svm_seg(vcpu, seg); + + var->base = s->base; + var->limit = s->limit; + var->selector = s->selector; + var->type = s->attrib & SVM_SELECTOR_TYPE_MASK; + var->s = (s->attrib >> SVM_SELECTOR_S_SHIFT) & 1; + var->dpl = (s->attrib >> SVM_SELECTOR_DPL_SHIFT) & 3; + var->present = (s->attrib >> SVM_SELECTOR_P_SHIFT) & 1; + var->avl = (s->attrib >> SVM_SELECTOR_AVL_SHIFT) & 1; + var->l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1; + var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1; + var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1; + var->unusable = !var->present; +} + +static void svm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) +{ + struct vmcb_seg *s = svm_seg(vcpu, VCPU_SREG_CS); + + *db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1; + *l = (s->attrib >> SVM_SELECTOR_L_SHIFT) & 1; +} + +static void svm_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + dt->limit = vcpu->svm->vmcb->save.ldtr.limit; + dt->base = vcpu->svm->vmcb->save.ldtr.base; +} + +static void svm_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + vcpu->svm->vmcb->save.ldtr.limit = dt->limit; + vcpu->svm->vmcb->save.ldtr.base = dt->base ; +} + +static void svm_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + dt->limit = vcpu->svm->vmcb->save.gdtr.limit; + dt->base = vcpu->svm->vmcb->save.gdtr.base; +} + +static void svm_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + vcpu->svm->vmcb->save.gdtr.limit = dt->limit; + vcpu->svm->vmcb->save.gdtr.base = dt->base ; +} + +static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +{ +#ifdef __x86_64__ + if (vcpu->shadow_efer & KVM_EFER_LME) { + if (!is_paging(vcpu) && (cr0 & CR0_PG_MASK)) { + vcpu->shadow_efer |= KVM_EFER_LMA; + vcpu->svm->vmcb->save.efer |= KVM_EFER_LMA | KVM_EFER_LME; + } + + if (is_paging(vcpu) && !(cr0 & CR0_PG_MASK) ) { + vcpu->shadow_efer &= ~KVM_EFER_LMA; + vcpu->svm->vmcb->save.efer &= ~(KVM_EFER_LMA | KVM_EFER_LME); + } + } +#endif + vcpu->svm->cr0 = cr0; + vcpu->svm->vmcb->save.cr0 = cr0 | CR0_PG_MASK; + vcpu->cr0 = cr0; +} + +static void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +{ + vcpu->cr4 = cr4; + vcpu->svm->vmcb->save.cr4 = cr4 | CR4_PAE_MASK; +} + +static void svm_set_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + struct vmcb_seg *s = svm_seg(vcpu, seg); + + s->base = var->base; + s->limit = var->limit; + s->selector = var->selector; + if (var->unusable) + s->attrib = 0; + else { + s->attrib = (var->type & SVM_SELECTOR_TYPE_MASK); + s->attrib |= (var->s & 1) << SVM_SELECTOR_S_SHIFT; + s->attrib |= (var->dpl & 3) << SVM_SELECTOR_DPL_SHIFT; + s->attrib |= (var->present & 1) << SVM_SELECTOR_P_SHIFT; + s->attrib |= (var->avl & 1) << SVM_SELECTOR_AVL_SHIFT; + s->attrib |= (var->l & 1) << SVM_SELECTOR_L_SHIFT; + s->attrib |= (var->db & 1) << SVM_SELECTOR_DB_SHIFT; + s->attrib |= (var->g & 1) << SVM_SELECTOR_G_SHIFT; + } + if (seg == VCPU_SREG_CS) + vcpu->svm->vmcb->save.cpl + = (vcpu->svm->vmcb->save.cs.attrib + >> SVM_SELECTOR_DPL_SHIFT) & 3; + +} + +/* FIXME: + + vcpu->svm->vmcb->control.int_ctl &= ~V_TPR_MASK; + vcpu->svm->vmcb->control.int_ctl |= (sregs->cr8 & V_TPR_MASK); + +*/ + +static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) +{ + return -EOPNOTSUPP; +} + +static void load_host_msrs(struct kvm_vcpu *vcpu) +{ + int i; + + for ( i = 0; i < NR_HOST_SAVE_MSRS; i++) + wrmsrl(host_save_msrs[i], vcpu->svm->host_msrs[i]); +} + +static void save_host_msrs(struct kvm_vcpu *vcpu) +{ + int i; + + for ( i = 0; i < NR_HOST_SAVE_MSRS; i++) + rdmsrl(host_save_msrs[i], vcpu->svm->host_msrs[i]); +} + +static void new_asid(struct kvm_vcpu *vcpu, struct svm_cpu_data *svm_data) +{ + if (svm_data->next_asid > svm_data->max_asid) { + ++svm_data->asid_generation; + svm_data->next_asid = 1; + vcpu->svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ALL_ASID; + } + + vcpu->cpu = svm_data->cpu; + vcpu->svm->asid_generation = svm_data->asid_generation; + vcpu->svm->vmcb->control.asid = svm_data->next_asid++; +} + +static void svm_invlpg(struct kvm_vcpu *vcpu, gva_t address) +{ + invlpga(address, vcpu->svm->vmcb->control.asid); // is needed? +} + +static unsigned long svm_get_dr(struct kvm_vcpu *vcpu, int dr) +{ + return vcpu->svm->db_regs[dr]; +} + +static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value, + int *exception) +{ + *exception = 0; + + if (vcpu->svm->vmcb->save.dr7 & DR7_GD_MASK) { + vcpu->svm->vmcb->save.dr7 &= ~DR7_GD_MASK; + vcpu->svm->vmcb->save.dr6 |= DR6_BD_MASK; + *exception = DB_VECTOR; + return; + } + + switch (dr) { + case 0 ... 3: + vcpu->svm->db_regs[dr] = value; + return; + case 4 ... 5: + if (vcpu->cr4 & CR4_DE_MASK) { + *exception = UD_VECTOR; + return; + } + case 7: { + if (value & ~((1ULL << 32) - 1)) { + *exception = GP_VECTOR; + return; + } + vcpu->svm->vmcb->save.dr7 = value; + return; + } + default: + printk(KERN_DEBUG "%s: unexpected dr %u\n", + __FUNCTION__, dr); + *exception = UD_VECTOR; + return; + } +} + +static int pf_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 exit_int_info = vcpu->svm->vmcb->control.exit_int_info; + u64 fault_address; + u32 error_code; + enum emulation_result er; + + if (is_external_interrupt(exit_int_info)) + push_irq(vcpu, exit_int_info & SVM_EVTINJ_VEC_MASK); + + spin_lock(&vcpu->kvm->lock); + + fault_address = vcpu->svm->vmcb->control.exit_info_2; + error_code = vcpu->svm->vmcb->control.exit_info_1; + if (!vcpu->mmu.page_fault(vcpu, fault_address, error_code)) { + spin_unlock(&vcpu->kvm->lock); + return 1; + } + er = emulate_instruction(vcpu, kvm_run, fault_address, error_code); + spin_unlock(&vcpu->kvm->lock); + + switch (er) { + case EMULATE_DONE: + return 1; + case EMULATE_DO_MMIO: + ++kvm_stat.mmio_exits; + kvm_run->exit_reason = KVM_EXIT_MMIO; + return 0; + case EMULATE_FAIL: + vcpu_printf(vcpu, "%s: emulate fail\n", __FUNCTION__); + break; + default: + BUG(); + } + + kvm_run->exit_reason = KVM_EXIT_UNKNOWN; + return 0; +} + +static int io_get_override(struct kvm_vcpu *vcpu, + struct vmcb_seg **seg, + int *addr_override) +{ + u8 inst[MAX_INST_SIZE]; + unsigned ins_length; + gva_t rip; + int i; + + rip = vcpu->svm->vmcb->save.rip; + ins_length = vcpu->svm->next_rip - rip; + rip += vcpu->svm->vmcb->save.cs.base; + + if (ins_length > MAX_INST_SIZE) + printk(KERN_DEBUG + "%s: inst length err, cs base 0x%llx rip 0x%llx " + "next rip 0x%llx ins_length %u\n", + __FUNCTION__, + vcpu->svm->vmcb->save.cs.base, + vcpu->svm->vmcb->save.rip, + vcpu->svm->vmcb->control.exit_info_2, + ins_length); + + if (kvm_read_guest(vcpu, rip, ins_length, inst) != ins_length) + /* #PF */ + return 0; + + *addr_override = 0; + *seg = 0; + for (i = 0; i < ins_length; i++) + switch (inst[i]) { + case 0xf0: + case 0xf2: + case 0xf3: + case 0x66: + continue; + case 0x67: + *addr_override = 1; + continue; + case 0x2e: + *seg = &vcpu->svm->vmcb->save.cs; + continue; + case 0x36: + *seg = &vcpu->svm->vmcb->save.ss; + continue; + case 0x3e: + *seg = &vcpu->svm->vmcb->save.ds; + continue; + case 0x26: + *seg = &vcpu->svm->vmcb->save.es; + continue; + case 0x64: + *seg = &vcpu->svm->vmcb->save.fs; + continue; + case 0x65: + *seg = &vcpu->svm->vmcb->save.gs; + continue; + default: + return 1; + } + printk(KERN_DEBUG "%s: unexpected\n", __FUNCTION__); + return 0; +} + +static unsigned long io_adress(struct kvm_vcpu *vcpu, int ins, u64 *address) +{ + unsigned long addr_mask; + unsigned long *reg; + struct vmcb_seg *seg; + int addr_override; + struct vmcb_save_area *save_area = &vcpu->svm->vmcb->save; + u16 cs_attrib = save_area->cs.attrib; + unsigned addr_size = get_addr_size(vcpu); + + if (!io_get_override(vcpu, &seg, &addr_override)) + return 0; + + if (addr_override) + addr_size = (addr_size == 2) ? 4: (addr_size >> 1); + + if (ins) { + reg = &vcpu->regs[VCPU_REGS_RDI]; + seg = &vcpu->svm->vmcb->save.es; + } else { + reg = &vcpu->regs[VCPU_REGS_RSI]; + seg = (seg) ? seg : &vcpu->svm->vmcb->save.ds; + } + + addr_mask = ~0ULL >> (64 - (addr_size * 8)); + + if ((cs_attrib & SVM_SELECTOR_L_MASK) && + !(vcpu->svm->vmcb->save.rflags & X86_EFLAGS_VM)) { + *address = (*reg & addr_mask); + return addr_mask; + } + + if (!(seg->attrib & SVM_SELECTOR_P_SHIFT)) { + svm_inject_gp(vcpu, 0); + return 0; + } + + *address = (*reg & addr_mask) + seg->base; + return addr_mask; +} + +static int io_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 io_info = vcpu->svm->vmcb->control.exit_info_1; //address size bug? + int _in = io_info & SVM_IOIO_TYPE_MASK; + + ++kvm_stat.io_exits; + + vcpu->svm->next_rip = vcpu->svm->vmcb->control.exit_info_2; + + kvm_run->exit_reason = KVM_EXIT_IO; + kvm_run->io.port = io_info >> 16; + kvm_run->io.direction = (_in) ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT; + kvm_run->io.size = ((io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT); + kvm_run->io.string = (io_info & SVM_IOIO_STR_MASK) != 0; + kvm_run->io.rep = (io_info & SVM_IOIO_REP_MASK) != 0; + + if (kvm_run->io.string) { + unsigned addr_mask; + + addr_mask = io_adress(vcpu, _in, &kvm_run->io.address); + if (!addr_mask) { + printk(KERN_DEBUG "%s: get io address failed\n", __FUNCTION__); + return 1; + } + + if (kvm_run->io.rep) { + kvm_run->io.count = vcpu->regs[VCPU_REGS_RCX] & addr_mask; + kvm_run->io.string_down = (vcpu->svm->vmcb->save.rflags + & X86_EFLAGS_DF) != 0; + } + } else { + kvm_run->io.value = vcpu->svm->vmcb->save.rax; + } + return 0; +} + + +static int nop_on_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + return 1; +} + +static int halt_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + vcpu->svm->next_rip = vcpu->svm->vmcb->save.rip + 1; + skip_emulated_instruction(vcpu); + if (vcpu->irq_summary && (vcpu->svm->vmcb->save.rflags & X86_EFLAGS_IF)) + return 1; + + kvm_run->exit_reason = KVM_EXIT_HLT; + return 0; +} + +static int invalid_op_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + inject_ud(vcpu); + return 1; +} + +static int task_switch_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + printk(KERN_DEBUG "%s: task swiche is unsupported\n", __FUNCTION__); + kvm_run->exit_reason = KVM_EXIT_UNKNOWN; + return 0; +} + +static int cpuid_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + vcpu->svm->next_rip = vcpu->svm->vmcb->save.rip + 2; + kvm_run->exit_reason = KVM_EXIT_CPUID; + return 0; +} + +static int emulate_on_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + if (emulate_instruction(vcpu, 0, 0, 0) != EMULATE_DONE) + printk(KERN_ERR "%s: failed\n", __FUNCTION__); + return 1; +} + +static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data) +{ + switch (ecx) { + case MSR_IA32_MC0_CTL: + case MSR_IA32_MCG_STATUS: + case MSR_IA32_MCG_CAP: + case MSR_IA32_MC0_MISC: + case MSR_IA32_MC0_MISC+4: + case MSR_IA32_MC0_MISC+8: + case MSR_IA32_MC0_MISC+12: + case MSR_IA32_MC0_MISC+16: + case MSR_IA32_UCODE_REV: + /* MTRR registers */ + case 0xfe: + case 0x200 ... 0x2ff: + *data = 0; + break; + case MSR_IA32_TIME_STAMP_COUNTER: { + u64 tsc; + + rdtscll(tsc); + *data = vcpu->svm->vmcb->control.tsc_offset + tsc; + break; + } + case MSR_EFER: + *data = vcpu->shadow_efer; + break; + case MSR_IA32_APICBASE: + *data = vcpu->apic_base; + break; +#ifdef __x86_64__ + case MSR_STAR: + *data = vcpu->svm->vmcb->save.star; + break; + case MSR_LSTAR: + *data = vcpu->svm->vmcb->save.lstar; + break; + case MSR_CSTAR: + *data = vcpu->svm->vmcb->save.cstar; + break; + case MSR_KERNEL_GS_BASE: + *data = vcpu->svm->vmcb->save.kernel_gs_base; + break; + case MSR_SYSCALL_MASK: + *data = vcpu->svm->vmcb->save.sfmask; + break; +#endif + case MSR_IA32_SYSENTER_CS: + *data = vcpu->svm->vmcb->save.sysenter_cs; + break; + case MSR_IA32_SYSENTER_EIP: + *data = vcpu->svm->vmcb->save.sysenter_eip; + break; + case MSR_IA32_SYSENTER_ESP: + *data = vcpu->svm->vmcb->save.sysenter_esp; + break; + default: + printk(KERN_ERR "kvm: unhandled rdmsr: 0x%x\n", ecx); + return 1; + } + return 0; +} + +static int rdmsr_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 ecx = vcpu->regs[VCPU_REGS_RCX]; + u64 data; + + if (svm_get_msr(vcpu, ecx, &data)) + svm_inject_gp(vcpu, 0); + else { + vcpu->svm->vmcb->save.rax = data & 0xffffffff; + vcpu->regs[VCPU_REGS_RDX] = data >> 32; + vcpu->svm->next_rip = vcpu->svm->vmcb->save.rip + 2; + skip_emulated_instruction(vcpu); + } + return 1; +} + +static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data) +{ + switch (ecx) { +#ifdef __x86_64__ + case MSR_EFER: + set_efer(vcpu, data); + break; +#endif + case MSR_IA32_MC0_STATUS: + printk(KERN_WARNING "%s: MSR_IA32_MC0_STATUS 0x%llx, nop\n" + , __FUNCTION__, data); + break; + case MSR_IA32_TIME_STAMP_COUNTER: { + u64 tsc; + + rdtscll(tsc); + vcpu->svm->vmcb->control.tsc_offset = data - tsc; + break; + } + case MSR_IA32_UCODE_REV: + case MSR_IA32_UCODE_WRITE: + case 0x200 ... 0x2ff: /* MTRRs */ + break; + case MSR_IA32_APICBASE: + vcpu->apic_base = data; + break; +#ifdef __x86_64___ + case MSR_STAR: + vcpu->svm->vmcb->save.star = data; + break; + case MSR_LSTAR: + vcpu->svm->vmcb->save.lstar = data; + break; + case MSR_CSTAR: + vcpu->svm->vmcb->save.cstar = data; + break; + case MSR_KERNEL_GS_BASE: + vcpu->svm->vmcb->save.kernel_gs_base = data; + break; + case MSR_SYSCALL_MASK: + vcpu->svm->vmcb->save.sfmask = data; + break; +#endif + case MSR_IA32_SYSENTER_CS: + vcpu->svm->vmcb->save.sysenter_cs = data; + break; + case MSR_IA32_SYSENTER_EIP: + vcpu->svm->vmcb->save.sysenter_eip = data; + break; + case MSR_IA32_SYSENTER_ESP: + vcpu->svm->vmcb->save.sysenter_esp = data; + break; + default: + printk(KERN_ERR "kvm: unhandled wrmsr: %x\n", ecx); + return 1; + } + return 0; +} + +static int wrmsr_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 ecx = vcpu->regs[VCPU_REGS_RCX]; + u64 data = (vcpu->svm->vmcb->save.rax & -1u) + | ((u64)(vcpu->regs[VCPU_REGS_RDX] & -1u) << 32); + vcpu->svm->next_rip = vcpu->svm->vmcb->save.rip + 2; + if (svm_set_msr(vcpu, ecx, data)) + svm_inject_gp(vcpu, 0); + else + skip_emulated_instruction(vcpu); + return 1; +} + +static int msr_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + if (vcpu->svm->vmcb->control.exit_info_1) + return wrmsr_interception(vcpu, kvm_run); + else + return rdmsr_interception(vcpu, kvm_run); +} + +static int (*svm_exit_handlers[])(struct kvm_vcpu *vcpu, + struct kvm_run *kvm_run) = { + [SVM_EXIT_READ_CR0] = emulate_on_interception, + [SVM_EXIT_READ_CR3] = emulate_on_interception, + [SVM_EXIT_READ_CR4] = emulate_on_interception, + /* for now: */ + [SVM_EXIT_WRITE_CR0] = emulate_on_interception, + [SVM_EXIT_WRITE_CR3] = emulate_on_interception, + [SVM_EXIT_WRITE_CR4] = emulate_on_interception, + [SVM_EXIT_READ_DR0] = emulate_on_interception, + [SVM_EXIT_READ_DR1] = emulate_on_interception, + [SVM_EXIT_READ_DR2] = emulate_on_interception, + [SVM_EXIT_READ_DR3] = emulate_on_interception, + [SVM_EXIT_WRITE_DR0] = emulate_on_interception, + [SVM_EXIT_WRITE_DR1] = emulate_on_interception, + [SVM_EXIT_WRITE_DR2] = emulate_on_interception, + [SVM_EXIT_WRITE_DR3] = emulate_on_interception, + [SVM_EXIT_WRITE_DR5] = emulate_on_interception, + [SVM_EXIT_WRITE_DR7] = emulate_on_interception, + [SVM_EXIT_EXCP_BASE + PF_VECTOR] = pf_interception, + [SVM_EXIT_INTR] = nop_on_interception, + [SVM_EXIT_NMI] = nop_on_interception, + [SVM_EXIT_SMI] = nop_on_interception, + [SVM_EXIT_INIT] = nop_on_interception, + /* [SVM_EXIT_CR0_SEL_WRITE] = emulate_on_interception, */ + [SVM_EXIT_CPUID] = cpuid_interception, + [SVM_EXIT_HLT] = halt_interception, + [SVM_EXIT_INVLPG] = emulate_on_interception, + [SVM_EXIT_INVLPGA] = invalid_op_interception, + [SVM_EXIT_IOIO] = io_interception, + [SVM_EXIT_MSR] = msr_interception, + [SVM_EXIT_TASK_SWITCH] = task_switch_interception, + [SVM_EXIT_VMRUN] = invalid_op_interception, + [SVM_EXIT_VMMCALL] = invalid_op_interception, + [SVM_EXIT_VMLOAD] = invalid_op_interception, + [SVM_EXIT_VMSAVE] = invalid_op_interception, + [SVM_EXIT_STGI] = invalid_op_interception, + [SVM_EXIT_CLGI] = invalid_op_interception, + [SVM_EXIT_SKINIT] = invalid_op_interception, +}; + + +static int handle_exit(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 exit_code = vcpu->svm->vmcb->control.exit_code; + + kvm_run->exit_type = KVM_EXIT_TYPE_VM_EXIT; + + if (is_external_interrupt(vcpu->svm->vmcb->control.exit_int_info) && + exit_code != SVM_EXIT_EXCP_BASE + PF_VECTOR) + printk(KERN_ERR "%s: unexpected exit_ini_info 0x%x " + "exit_code 0x%x\n", + __FUNCTION__, vcpu->svm->vmcb->control.exit_int_info, + exit_code); + + if (exit_code >= sizeof(svm_exit_handlers) / sizeof(*svm_exit_handlers) + || svm_exit_handlers[exit_code] == 0) { + kvm_run->exit_reason = KVM_EXIT_UNKNOWN; + printk(KERN_ERR "%s: 0x%x @ 0x%llx cr0 0x%lx rflags 0x%llx\n", + __FUNCTION__, + exit_code, + vcpu->svm->vmcb->save.rip, + vcpu->cr0, + vcpu->svm->vmcb->save.rflags); + return 0; + } + + return svm_exit_handlers[exit_code](vcpu, kvm_run); +} + +static void reload_tss(struct kvm_vcpu *vcpu) +{ + int cpu = raw_smp_processor_id(); + + struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu); + svm_data->tss_desc->type = 9; //available 32/64-bit TSS + load_TR_desc(); +} + +static void pre_svm_run(struct kvm_vcpu *vcpu) +{ + int cpu = raw_smp_processor_id(); + + struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu); + + vcpu->svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING; + if (vcpu->cpu != cpu || + vcpu->svm->asid_generation != svm_data->asid_generation) + new_asid(vcpu, svm_data); +} + + +static inline void kvm_try_inject_irq(struct kvm_vcpu *vcpu) +{ + struct vmcb_control_area *control; + + if (!vcpu->irq_summary) + return; + + control = &vcpu->svm->vmcb->control; + + control->int_vector = pop_irq(vcpu); + control->int_ctl &= ~V_INTR_PRIO_MASK; + control->int_ctl |= V_IRQ_MASK | + ((/*control->int_vector >> 4*/ 0xf) << V_INTR_PRIO_SHIFT); +} + +static void kvm_reput_irq(struct kvm_vcpu *vcpu) +{ + struct vmcb_control_area *control = &vcpu->svm->vmcb->control; + + if (control->int_ctl & V_IRQ_MASK) { + control->int_ctl &= ~V_IRQ_MASK; + push_irq(vcpu, control->int_vector); + } +} + +static void save_db_regs(unsigned long *db_regs) +{ +#ifdef __x86_64__ + asm ("mov %%dr0, %%rax \n\t" + "mov %%rax, %[dr0] \n\t" + "mov %%dr1, %%rax \n\t" + "mov %%rax, %[dr1] \n\t" + "mov %%dr2, %%rax \n\t" + "mov %%rax, %[dr2] \n\t" + "mov %%dr3, %%rax \n\t" + "mov %%rax, %[dr3] \n\t" + : [dr0] "=m"(db_regs[0]), + [dr1] "=m"(db_regs[1]), + [dr2] "=m"(db_regs[2]), + [dr3] "=m"(db_regs[3]) + : : "rax"); +#else + asm ("mov %%dr0, %%eax \n\t" + "mov %%eax, %[dr0] \n\t" + "mov %%dr1, %%eax \n\t" + "mov %%eax, %[dr1] \n\t" + "mov %%dr2, %%eax \n\t" + "mov %%eax, %[dr2] \n\t" + "mov %%dr3, %%eax \n\t" + "mov %%eax, %[dr3] \n\t" + : [dr0] "=m"(db_regs[0]), + [dr1] "=m"(db_regs[1]), + [dr2] "=m"(db_regs[2]), + [dr3] "=m"(db_regs[3]) + : : "eax"); +#endif +} + +static void load_db_regs(unsigned long *db_regs) +{ + asm volatile ("mov %[dr0], %%dr0 \n\t" + "mov %[dr1], %%dr1 \n\t" + "mov %[dr2], %%dr2 \n\t" + "mov %[dr3], %%dr3 \n\t" + : + : [dr0] "r"(db_regs[0]), + [dr1] "r"(db_regs[1]), + [dr2] "r"(db_regs[2]), + [dr3] "r"(db_regs[3]) +#ifdef __x86_64__ + : "rax"); +#else + : "eax"); +#endif +} + +static int svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u16 fs_selector; + u16 gs_selector; + u16 ldt_selector; + +again: + kvm_try_inject_irq(vcpu); + + clgi(); + + pre_svm_run(vcpu); + + save_host_msrs(vcpu); + fs_selector = read_fs(); + gs_selector = read_gs(); + ldt_selector = read_ldt(); + vcpu->svm->host_cr2 = kvm_read_cr2(); + vcpu->svm->host_dr6 = read_dr6(); + vcpu->svm->host_dr7 = read_dr7(); + vcpu->svm->vmcb->save.cr2 = vcpu->cr2; + + if (vcpu->svm->vmcb->save.dr7 & 0xff) { + write_dr7(0); + save_db_regs(vcpu->svm->host_db_regs); + load_db_regs(vcpu->svm->db_regs); + } + asm volatile ( +#ifdef __x86_64__ + "push %%rbx; push %%rcx; push %%rdx;" + "push %%rsi; push %%rdi; push %%rbp;" + "push %%r8; push %%r9; push %%r10; push %%r11;" + "push %%r12; push %%r13; push %%r14; push %%r15;" +#else + "push %%ebx; push %%ecx; push %%edx;" + "push %%esi; push %%edi; push %%ebp;" +#endif + +#ifdef __x86_64__ + "mov %c[rbx](%[vcpu]), %%rbx \n\t" + "mov %c[rcx](%[vcpu]), %%rcx \n\t" + "mov %c[rdx](%[vcpu]), %%rdx \n\t" + "mov %c[rsi](%[vcpu]), %%rsi \n\t" + "mov %c[rdi](%[vcpu]), %%rdi \n\t" + "mov %c[rbp](%[vcpu]), %%rbp \n\t" + "mov %c[r8](%[vcpu]), %%r8 \n\t" + "mov %c[r9](%[vcpu]), %%r9 \n\t" + "mov %c[r10](%[vcpu]), %%r10 \n\t" + "mov %c[r11](%[vcpu]), %%r11 \n\t" + "mov %c[r12](%[vcpu]), %%r12 \n\t" + "mov %c[r13](%[vcpu]), %%r13 \n\t" + "mov %c[r14](%[vcpu]), %%r14 \n\t" + "mov %c[r15](%[vcpu]), %%r15 \n\t" +#else + "mov %c[rbx](%[vcpu]), %%ebx \n\t" + "mov %c[rcx](%[vcpu]), %%ecx \n\t" + "mov %c[rdx](%[vcpu]), %%edx \n\t" + "mov %c[rsi](%[vcpu]), %%esi \n\t" + "mov %c[rdi](%[vcpu]), %%edi \n\t" + "mov %c[rbp](%[vcpu]), %%ebp \n\t" +#endif + +#ifdef __x86_64__ + /* Enter guest mode */ + "push %%rax \n\t" + "mov %c[svm](%[vcpu]), %%rax \n\t" + "mov %c[vmcb](%%rax), %%rax \n\t" + SVM_VMLOAD "\n\t" + SVM_VMRUN "\n\t" + SVM_VMSAVE "\n\t" + "pop %%rax \n\t" +#else + /* Enter guest mode */ + "push %%eax \n\t" + "mov %c[svm](%[vcpu]), %%eax \n\t" + "mov %c[vmcb](%%eax), %%eax \n\t" + SVM_VMLOAD "\n\t" + SVM_VMRUN "\n\t" + SVM_VMSAVE "\n\t" + "pop %%eax \n\t" +#endif + + /* Save guest registers, load host registers */ +#ifdef __x86_64__ + "mov %%rbx, %c[rbx](%[vcpu]) \n\t" + "mov %%rcx, %c[rcx](%[vcpu]) \n\t" + "mov %%rdx, %c[rdx](%[vcpu]) \n\t" + "mov %%rsi, %c[rsi](%[vcpu]) \n\t" + "mov %%rdi, %c[rdi](%[vcpu]) \n\t" + "mov %%rbp, %c[rbp](%[vcpu]) \n\t" + "mov %%r8, %c[r8](%[vcpu]) \n\t" + "mov %%r9, %c[r9](%[vcpu]) \n\t" + "mov %%r10, %c[r10](%[vcpu]) \n\t" + "mov %%r11, %c[r11](%[vcpu]) \n\t" + "mov %%r12, %c[r12](%[vcpu]) \n\t" + "mov %%r13, %c[r13](%[vcpu]) \n\t" + "mov %%r14, %c[r14](%[vcpu]) \n\t" + "mov %%r15, %c[r15](%[vcpu]) \n\t" + + "pop %%r15; pop %%r14; pop %%r13; pop %%r12;" + "pop %%r11; pop %%r10; pop %%r9; pop %%r8;" + "pop %%rbp; pop %%rdi; pop %%rsi;" + "pop %%rdx; pop %%rcx; pop %%rbx; \n\t" +#else + "mov %%ebx, %c[rbx](%[vcpu]) \n\t" + "mov %%ecx, %c[rcx](%[vcpu]) \n\t" + "mov %%edx, %c[rdx](%[vcpu]) \n\t" + "mov %%esi, %c[rsi](%[vcpu]) \n\t" + "mov %%edi, %c[rdi](%[vcpu]) \n\t" + "mov %%ebp, %c[rbp](%[vcpu]) \n\t" + + "pop %%ebp; pop %%edi; pop %%esi;" + "pop %%edx; pop %%ecx; pop %%ebx; \n\t" +#endif + : + : [vcpu]"a"(vcpu), + [svm]"i"(offsetof(struct kvm_vcpu, svm)), + [vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)), + [rbx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RBX])), + [rcx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RCX])), + [rdx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RDX])), + [rsi]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RSI])), + [rdi]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RDI])), + [rbp]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RBP])) +#ifdef __x86_64__ + ,[r8 ]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R8 ])), + [r9 ]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R9 ])), + [r10]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R10])), + [r11]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R11])), + [r12]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R12])), + [r13]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R13])), + [r14]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R14])), + [r15]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R15])) +#endif + : "cc", "memory" ); + + if ((vcpu->svm->vmcb->save.dr7 & 0xff)) + load_db_regs(vcpu->svm->host_db_regs); + + vcpu->cr2 = vcpu->svm->vmcb->save.cr2; + + write_dr6(vcpu->svm->host_dr6); + write_dr7(vcpu->svm->host_dr7); + kvm_write_cr2(vcpu->svm->host_cr2); + + load_fs(fs_selector); + load_gs(gs_selector); + load_ldt(ldt_selector); + load_host_msrs(vcpu); + + reload_tss(vcpu); + + stgi(); + + kvm_reput_irq(vcpu); + + vcpu->svm->next_rip = 0; + + if (vcpu->svm->vmcb->control.exit_code == SVM_EXIT_ERR) { + kvm_run->exit_type = KVM_EXIT_TYPE_FAIL_ENTRY; + kvm_run->exit_reason = vcpu->svm->vmcb->control.exit_code; + return 0; + } + + if (handle_exit(vcpu, kvm_run)) { + if (signal_pending(current)) { + ++kvm_stat.signal_exits; + return -EINTR; + } + kvm_resched(vcpu); + goto again; + } + return 0; +} + +static void svm_flush_tlb(struct kvm_vcpu *vcpu) +{ + force_new_asid(vcpu); +} + +static void svm_set_cr3(struct kvm_vcpu *vcpu, unsigned long root) +{ + vcpu->svm->vmcb->save.cr3 = root; + force_new_asid(vcpu); +} + +static void svm_inject_page_fault(struct kvm_vcpu *vcpu, + unsigned long addr, + uint32_t err_code) +{ + uint32_t exit_int_info = vcpu->svm->vmcb->control.exit_int_info; + + ++kvm_stat.pf_guest; + + if (is_page_fault(exit_int_info)) { + + vcpu->svm->vmcb->control.event_inj_err = 0; + vcpu->svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | + SVM_EVTINJ_VALID_ERR | + SVM_EVTINJ_TYPE_EXEPT | + DF_VECTOR; + return; + } + vcpu->cr2 = addr; + vcpu->svm->vmcb->save.cr2 = addr; + vcpu->svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | + SVM_EVTINJ_VALID_ERR | + SVM_EVTINJ_TYPE_EXEPT | + PF_VECTOR; + vcpu->svm->vmcb->control.event_inj_err = err_code; +} + + +static int is_disabled(void) +{ + return 0; +} + +static struct kvm_arch_ops svm_arch_ops = { + .cpu_has_kvm_support = has_svm, + .disabled_by_bios = is_disabled, + .hardware_setup = svm_hardware_setup, + .hardware_unsetup = svm_hardware_unsetup, + .hardware_enable = svm_hardware_enable, + .hardware_disable = svm_hardware_disable, + + .vcpu_create = svm_create_vcpu, + .vcpu_free = svm_free_vcpu, + + .vcpu_load = svm_vcpu_load, + .vcpu_put = svm_vcpu_put, + + .set_guest_debug = svm_guest_debug, + .get_msr = svm_get_msr, + .set_msr = svm_set_msr, + .get_segment_base = svm_get_segment_base, + .get_segment = svm_get_segment, + .set_segment = svm_set_segment, + .is_long_mode = svm_is_long_mode, + .get_cs_db_l_bits = svm_get_cs_db_l_bits, + .set_cr0 = svm_set_cr0, + .set_cr0_no_modeswitch = svm_set_cr0, + .set_cr3 = svm_set_cr3, + .set_cr4 = svm_set_cr4, + .set_efer = svm_set_efer, + .get_idt = svm_get_idt, + .set_idt = svm_set_idt, + .get_gdt = svm_get_gdt, + .set_gdt = svm_set_gdt, + .get_dr = svm_get_dr, + .set_dr = svm_set_dr, + .cache_regs = svm_cache_regs, + .decache_regs = svm_decache_regs, + .get_rflags = svm_get_rflags, + .set_rflags = svm_set_rflags, + + .invlpg = svm_invlpg, + .tlb_flush = svm_flush_tlb, + .inject_page_fault = svm_inject_page_fault, + + .inject_gp = svm_inject_gp, + + .run = svm_vcpu_run, + .skip_emulated_instruction = skip_emulated_instruction, + .vcpu_setup = svm_vcpu_setup, +}; + +static int __init svm_init(void) +{ + kvm_emulator_want_group7_invlpg(); + kvm_init_arch(&svm_arch_ops, THIS_MODULE); + return 0; +} + +static void __exit svm_exit(void) +{ + kvm_exit_arch(); +} + +module_init(svm_init) +module_exit(svm_exit) diff --git a/drivers/kvm/svm.h b/drivers/kvm/svm.h new file mode 100644 index 000000000000..df731c3fb588 --- /dev/null +++ b/drivers/kvm/svm.h @@ -0,0 +1,315 @@ +#ifndef __SVM_H +#define __SVM_H + +enum { + INTERCEPT_INTR, + INTERCEPT_NMI, + INTERCEPT_SMI, + INTERCEPT_INIT, + INTERCEPT_VINTR, + INTERCEPT_SELECTIVE_CR0, + INTERCEPT_STORE_IDTR, + INTERCEPT_STORE_GDTR, + INTERCEPT_STORE_LDTR, + INTERCEPT_STORE_TR, + INTERCEPT_LOAD_IDTR, + INTERCEPT_LOAD_GDTR, + INTERCEPT_LOAD_LDTR, + INTERCEPT_LOAD_TR, + INTERCEPT_RDTSC, + INTERCEPT_RDPMC, + INTERCEPT_PUSHF, + INTERCEPT_POPF, + INTERCEPT_CPUID, + INTERCEPT_RSM, + INTERCEPT_IRET, + INTERCEPT_INTn, + INTERCEPT_INVD, + INTERCEPT_PAUSE, + INTERCEPT_HLT, + INTERCEPT_INVLPG, + INTERCEPT_INVLPGA, + INTERCEPT_IOIO_PROT, + INTERCEPT_MSR_PROT, + INTERCEPT_TASK_SWITCH, + INTERCEPT_FERR_FREEZE, + INTERCEPT_SHUTDOWN, + INTERCEPT_VMRUN, + INTERCEPT_VMMCALL, + INTERCEPT_VMLOAD, + INTERCEPT_VMSAVE, + INTERCEPT_STGI, + INTERCEPT_CLGI, + INTERCEPT_SKINIT, + INTERCEPT_RDTSCP, + INTERCEPT_ICEBP, + INTERCEPT_WBINVD, +}; + + +struct __attribute__ ((__packed__)) vmcb_control_area { + u16 intercept_cr_read; + u16 intercept_cr_write; + u16 intercept_dr_read; + u16 intercept_dr_write; + u32 intercept_exceptions; + u64 intercept; + u8 reserved_1[44]; + u64 iopm_base_pa; + u64 msrpm_base_pa; + u64 tsc_offset; + u32 asid; + u8 tlb_ctl; + u8 reserved_2[3]; + u32 int_ctl; + u32 int_vector; + u32 int_state; + u8 reserved_3[4]; + u32 exit_code; + u32 exit_code_hi; + u64 exit_info_1; + u64 exit_info_2; + u32 exit_int_info; + u32 exit_int_info_err; + u64 nested_ctl; + u8 reserved_4[16]; + u32 event_inj; + u32 event_inj_err; + u64 nested_cr3; + u64 lbr_ctl; + u8 reserved_5[832]; +}; + + +#define TLB_CONTROL_DO_NOTHING 0 +#define TLB_CONTROL_FLUSH_ALL_ASID 1 + +#define V_TPR_MASK 0x0f + +#define V_IRQ_SHIFT 8 +#define V_IRQ_MASK (1 << V_IRQ_SHIFT) + +#define V_INTR_PRIO_SHIFT 16 +#define V_INTR_PRIO_MASK (0x0f << V_INTR_PRIO_SHIFT) + +#define V_IGN_TPR_SHIFT 20 +#define V_IGN_TPR_MASK (1 << V_IGN_TPR_SHIFT) + +#define V_INTR_MASKING_SHIFT 24 +#define V_INTR_MASKING_MASK (1 << V_INTR_MASKING_SHIFT) + +#define SVM_INTERRUPT_SHADOW_MASK 1 + +#define SVM_IOIO_STR_SHIFT 2 +#define SVM_IOIO_REP_SHIFT 3 +#define SVM_IOIO_SIZE_SHIFT 4 +#define SVM_IOIO_ASIZE_SHIFT 7 + +#define SVM_IOIO_TYPE_MASK 1 +#define SVM_IOIO_STR_MASK (1 << SVM_IOIO_STR_SHIFT) +#define SVM_IOIO_REP_MASK (1 << SVM_IOIO_REP_SHIFT) +#define SVM_IOIO_SIZE_MASK (7 << SVM_IOIO_SIZE_SHIFT) +#define SVM_IOIO_ASIZE_MASK (7 << SVM_IOIO_ASIZE_SHIFT) + +struct __attribute__ ((__packed__)) vmcb_seg { + u16 selector; + u16 attrib; + u32 limit; + u64 base; +}; + +struct __attribute__ ((__packed__)) vmcb_save_area { + struct vmcb_seg es; + struct vmcb_seg cs; + struct vmcb_seg ss; + struct vmcb_seg ds; + struct vmcb_seg fs; + struct vmcb_seg gs; + struct vmcb_seg gdtr; + struct vmcb_seg ldtr; + struct vmcb_seg idtr; + struct vmcb_seg tr; + u8 reserved_1[43]; + u8 cpl; + u8 reserved_2[4]; + u64 efer; + u8 reserved_3[112]; + u64 cr4; + u64 cr3; + u64 cr0; + u64 dr7; + u64 dr6; + u64 rflags; + u64 rip; + u8 reserved_4[88]; + u64 rsp; + u8 reserved_5[24]; + u64 rax; + u64 star; + u64 lstar; + u64 cstar; + u64 sfmask; + u64 kernel_gs_base; + u64 sysenter_cs; + u64 sysenter_esp; + u64 sysenter_eip; + u64 cr2; + u8 reserved_6[32]; + u64 g_pat; + u64 dbgctl; + u64 br_from; + u64 br_to; + u64 last_excp_from; + u64 last_excp_to; +}; + +struct __attribute__ ((__packed__)) vmcb { + struct vmcb_control_area control; + struct vmcb_save_area save; +}; + +#define SVM_CPUID_FEATURE_SHIFT 2 +#define SVM_CPUID_FUNC 0x8000000a + +#define MSR_EFER_SVME_MASK (1ULL << 12) +#define MSR_VM_HSAVE_PA 0xc0010117ULL + +#define SVM_SELECTOR_S_SHIFT 4 +#define SVM_SELECTOR_DPL_SHIFT 5 +#define SVM_SELECTOR_P_SHIFT 7 +#define SVM_SELECTOR_AVL_SHIFT 8 +#define SVM_SELECTOR_L_SHIFT 9 +#define SVM_SELECTOR_DB_SHIFT 10 +#define SVM_SELECTOR_G_SHIFT 11 + +#define SVM_SELECTOR_TYPE_MASK (0xf) +#define SVM_SELECTOR_S_MASK (1 << SVM_SELECTOR_S_SHIFT) +#define SVM_SELECTOR_DPL_MASK (3 << SVM_SELECTOR_DPL_SHIFT) +#define SVM_SELECTOR_P_MASK (1 << SVM_SELECTOR_P_SHIFT) +#define SVM_SELECTOR_AVL_MASK (1 << SVM_SELECTOR_AVL_SHIFT) +#define SVM_SELECTOR_L_MASK (1 << SVM_SELECTOR_L_SHIFT) +#define SVM_SELECTOR_DB_MASK (1 << SVM_SELECTOR_DB_SHIFT) +#define SVM_SELECTOR_G_MASK (1 << SVM_SELECTOR_G_SHIFT) + +#define SVM_SELECTOR_WRITE_MASK (1 << 1) +#define SVM_SELECTOR_READ_MASK SVM_SELECTOR_WRITE_MASK +#define SVM_SELECTOR_CODE_MASK (1 << 3) + +#define INTERCEPT_CR0_MASK 1 +#define INTERCEPT_CR3_MASK (1 << 3) +#define INTERCEPT_CR4_MASK (1 << 4) + +#define INTERCEPT_DR0_MASK 1 +#define INTERCEPT_DR1_MASK (1 << 1) +#define INTERCEPT_DR2_MASK (1 << 2) +#define INTERCEPT_DR3_MASK (1 << 3) +#define INTERCEPT_DR4_MASK (1 << 4) +#define INTERCEPT_DR5_MASK (1 << 5) +#define INTERCEPT_DR6_MASK (1 << 6) +#define INTERCEPT_DR7_MASK (1 << 7) + +#define SVM_EVTINJ_VEC_MASK 0xff + +#define SVM_EVTINJ_TYPE_SHIFT 8 +#define SVM_EVTINJ_TYPE_MASK (7 << SVM_EVTINJ_TYPE_SHIFT) + +#define SVM_EVTINJ_TYPE_INTR (0 << SVM_EVTINJ_TYPE_SHIFT) +#define SVM_EVTINJ_TYPE_NMI (2 << SVM_EVTINJ_TYPE_SHIFT) +#define SVM_EVTINJ_TYPE_EXEPT (3 << SVM_EVTINJ_TYPE_SHIFT) +#define SVM_EVTINJ_TYPE_SOFT (4 << SVM_EVTINJ_TYPE_SHIFT) + +#define SVM_EVTINJ_VALID (1 << 31) +#define SVM_EVTINJ_VALID_ERR (1 << 11) + +#define SVM_EXITINTINFO_VEC_MASK SVM_EVTINJ_VEC_MASK + +#define SVM_EXITINTINFO_TYPE_INTR SVM_EVTINJ_TYPE_INTR +#define SVM_EXITINTINFO_TYPE_NMI SVM_EVTINJ_TYPE_NMI +#define SVM_EXITINTINFO_TYPE_EXEPT SVM_EVTINJ_TYPE_EXEPT +#define SVM_EXITINTINFO_TYPE_SOFT SVM_EVTINJ_TYPE_SOFT + +#define SVM_EXITINTINFO_VALID SVM_EVTINJ_VALID +#define SVM_EXITINTINFO_VALID_ERR SVM_EVTINJ_VALID_ERR + +#define SVM_EXIT_READ_CR0 0x000 +#define SVM_EXIT_READ_CR3 0x003 +#define SVM_EXIT_READ_CR4 0x004 +#define SVM_EXIT_READ_CR8 0x008 +#define SVM_EXIT_WRITE_CR0 0x010 +#define SVM_EXIT_WRITE_CR3 0x013 +#define SVM_EXIT_WRITE_CR4 0x014 +#define SVM_EXIT_WRITE_CR8 0x018 +#define SVM_EXIT_READ_DR0 0x020 +#define SVM_EXIT_READ_DR1 0x021 +#define SVM_EXIT_READ_DR2 0x022 +#define SVM_EXIT_READ_DR3 0x023 +#define SVM_EXIT_READ_DR4 0x024 +#define SVM_EXIT_READ_DR5 0x025 +#define SVM_EXIT_READ_DR6 0x026 +#define SVM_EXIT_READ_DR7 0x027 +#define SVM_EXIT_WRITE_DR0 0x030 +#define SVM_EXIT_WRITE_DR1 0x031 +#define SVM_EXIT_WRITE_DR2 0x032 +#define SVM_EXIT_WRITE_DR3 0x033 +#define SVM_EXIT_WRITE_DR4 0x034 +#define SVM_EXIT_WRITE_DR5 0x035 +#define SVM_EXIT_WRITE_DR6 0x036 +#define SVM_EXIT_WRITE_DR7 0x037 +#define SVM_EXIT_EXCP_BASE 0x040 +#define SVM_EXIT_INTR 0x060 +#define SVM_EXIT_NMI 0x061 +#define SVM_EXIT_SMI 0x062 +#define SVM_EXIT_INIT 0x063 +#define SVM_EXIT_VINTR 0x064 +#define SVM_EXIT_CR0_SEL_WRITE 0x065 +#define SVM_EXIT_IDTR_READ 0x066 +#define SVM_EXIT_GDTR_READ 0x067 +#define SVM_EXIT_LDTR_READ 0x068 +#define SVM_EXIT_TR_READ 0x069 +#define SVM_EXIT_IDTR_WRITE 0x06a +#define SVM_EXIT_GDTR_WRITE 0x06b +#define SVM_EXIT_LDTR_WRITE 0x06c +#define SVM_EXIT_TR_WRITE 0x06d +#define SVM_EXIT_RDTSC 0x06e +#define SVM_EXIT_RDPMC 0x06f +#define SVM_EXIT_PUSHF 0x070 +#define SVM_EXIT_POPF 0x071 +#define SVM_EXIT_CPUID 0x072 +#define SVM_EXIT_RSM 0x073 +#define SVM_EXIT_IRET 0x074 +#define SVM_EXIT_SWINT 0x075 +#define SVM_EXIT_INVD 0x076 +#define SVM_EXIT_PAUSE 0x077 +#define SVM_EXIT_HLT 0x078 +#define SVM_EXIT_INVLPG 0x079 +#define SVM_EXIT_INVLPGA 0x07a +#define SVM_EXIT_IOIO 0x07b +#define SVM_EXIT_MSR 0x07c +#define SVM_EXIT_TASK_SWITCH 0x07d +#define SVM_EXIT_FERR_FREEZE 0x07e +#define SVM_EXIT_SHUTDOWN 0x07f +#define SVM_EXIT_VMRUN 0x080 +#define SVM_EXIT_VMMCALL 0x081 +#define SVM_EXIT_VMLOAD 0x082 +#define SVM_EXIT_VMSAVE 0x083 +#define SVM_EXIT_STGI 0x084 +#define SVM_EXIT_CLGI 0x085 +#define SVM_EXIT_SKINIT 0x086 +#define SVM_EXIT_RDTSCP 0x087 +#define SVM_EXIT_ICEBP 0x088 +#define SVM_EXIT_WBINVD 0x089 +#define SVM_EXIT_NPF 0x400 + +#define SVM_EXIT_ERR -1 + +#define SVM_CR0_SELECTIVE_MASK (1 << 3 | 1) // TS and MP + +#define SVM_VMLOAD ".byte 0x0f, 0x01, 0xda" +#define SVM_VMRUN ".byte 0x0f, 0x01, 0xd8" +#define SVM_VMSAVE ".byte 0x0f, 0x01, 0xdb" +#define SVM_CLGI ".byte 0x0f, 0x01, 0xdd" +#define SVM_STGI ".byte 0x0f, 0x01, 0xdc" +#define SVM_INVLPGA ".byte 0x0f, 0x01, 0xdf" + +#endif + diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c new file mode 100644 index 000000000000..bda7a7ae2167 --- /dev/null +++ b/drivers/kvm/vmx.c @@ -0,0 +1,2002 @@ +/* + * Kernel-based Virtual Machine driver for Linux + * + * This module enables machines with Intel VT-x extensions to run virtual + * machines without emulation or binary translation. + * + * Copyright (C) 2006 Qumranet, Inc. + * + * Authors: + * Avi Kivity + * Yaniv Kamay + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include "kvm.h" +#include "vmx.h" +#include "kvm_vmx.h" +#include +#include +#include +#include + +#include "segment_descriptor.h" + +#define MSR_IA32_FEATURE_CONTROL 0x03a + +MODULE_AUTHOR("Qumranet"); +MODULE_LICENSE("GPL"); + +static DEFINE_PER_CPU(struct vmcs *, vmxarea); +static DEFINE_PER_CPU(struct vmcs *, current_vmcs); + +#ifdef __x86_64__ +#define HOST_IS_64 1 +#else +#define HOST_IS_64 0 +#endif + +static struct vmcs_descriptor { + int size; + int order; + u32 revision_id; +} vmcs_descriptor; + +#define VMX_SEGMENT_FIELD(seg) \ + [VCPU_SREG_##seg] = { \ + .selector = GUEST_##seg##_SELECTOR, \ + .base = GUEST_##seg##_BASE, \ + .limit = GUEST_##seg##_LIMIT, \ + .ar_bytes = GUEST_##seg##_AR_BYTES, \ + } + +static struct kvm_vmx_segment_field { + unsigned selector; + unsigned base; + unsigned limit; + unsigned ar_bytes; +} kvm_vmx_segment_fields[] = { + VMX_SEGMENT_FIELD(CS), + VMX_SEGMENT_FIELD(DS), + VMX_SEGMENT_FIELD(ES), + VMX_SEGMENT_FIELD(FS), + VMX_SEGMENT_FIELD(GS), + VMX_SEGMENT_FIELD(SS), + VMX_SEGMENT_FIELD(TR), + VMX_SEGMENT_FIELD(LDTR), +}; + +static const u32 vmx_msr_index[] = { +#ifdef __x86_64__ + MSR_SYSCALL_MASK, MSR_LSTAR, MSR_CSTAR, MSR_KERNEL_GS_BASE, +#endif + MSR_EFER, MSR_K6_STAR, +}; +#define NR_VMX_MSR (sizeof(vmx_msr_index) / sizeof(*vmx_msr_index)) + +struct vmx_msr_entry *find_msr_entry(struct kvm_vcpu *vcpu, u32 msr); + +static inline int is_page_fault(u32 intr_info) +{ + return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK | + INTR_INFO_VALID_MASK)) == + (INTR_TYPE_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK); +} + +static inline int is_external_interrupt(u32 intr_info) +{ + return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VALID_MASK)) + == (INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK); +} + +static void vmcs_clear(struct vmcs *vmcs) +{ + u64 phys_addr = __pa(vmcs); + u8 error; + + asm volatile (ASM_VMX_VMCLEAR_RAX "; setna %0" + : "=g"(error) : "a"(&phys_addr), "m"(phys_addr) + : "cc", "memory"); + if (error) + printk(KERN_ERR "kvm: vmclear fail: %p/%llx\n", + vmcs, phys_addr); +} + +static void __vcpu_clear(void *arg) +{ + struct kvm_vcpu *vcpu = arg; + int cpu = smp_processor_id(); + + if (vcpu->cpu == cpu) + vmcs_clear(vcpu->vmcs); + if (per_cpu(current_vmcs, cpu) == vcpu->vmcs) + per_cpu(current_vmcs, cpu) = NULL; +} + +static unsigned long vmcs_readl(unsigned long field) +{ + unsigned long value; + + asm volatile (ASM_VMX_VMREAD_RDX_RAX + : "=a"(value) : "d"(field) : "cc"); + return value; +} + +static u16 vmcs_read16(unsigned long field) +{ + return vmcs_readl(field); +} + +static u32 vmcs_read32(unsigned long field) +{ + return vmcs_readl(field); +} + +static u64 vmcs_read64(unsigned long field) +{ +#ifdef __x86_64__ + return vmcs_readl(field); +#else + return vmcs_readl(field) | ((u64)vmcs_readl(field+1) << 32); +#endif +} + +static void vmcs_writel(unsigned long field, unsigned long value) +{ + u8 error; + + asm volatile (ASM_VMX_VMWRITE_RAX_RDX "; setna %0" + : "=q"(error) : "a"(value), "d"(field) : "cc" ); + if (error) + printk(KERN_ERR "vmwrite error: reg %lx value %lx (err %d)\n", + field, value, vmcs_read32(VM_INSTRUCTION_ERROR)); +} + +static void vmcs_write16(unsigned long field, u16 value) +{ + vmcs_writel(field, value); +} + +static void vmcs_write32(unsigned long field, u32 value) +{ + vmcs_writel(field, value); +} + +static void vmcs_write64(unsigned long field, u64 value) +{ +#ifdef __x86_64__ + vmcs_writel(field, value); +#else + vmcs_writel(field, value); + asm volatile (""); + vmcs_writel(field+1, value >> 32); +#endif +} + +/* + * Switches to specified vcpu, until a matching vcpu_put(), but assumes + * vcpu mutex is already taken. + */ +static struct kvm_vcpu *vmx_vcpu_load(struct kvm_vcpu *vcpu) +{ + u64 phys_addr = __pa(vcpu->vmcs); + int cpu; + + cpu = get_cpu(); + + if (vcpu->cpu != cpu) { + smp_call_function(__vcpu_clear, vcpu, 0, 1); + vcpu->launched = 0; + } + + if (per_cpu(current_vmcs, cpu) != vcpu->vmcs) { + u8 error; + + per_cpu(current_vmcs, cpu) = vcpu->vmcs; + asm volatile (ASM_VMX_VMPTRLD_RAX "; setna %0" + : "=g"(error) : "a"(&phys_addr), "m"(phys_addr) + : "cc"); + if (error) + printk(KERN_ERR "kvm: vmptrld %p/%llx fail\n", + vcpu->vmcs, phys_addr); + } + + if (vcpu->cpu != cpu) { + struct descriptor_table dt; + unsigned long sysenter_esp; + + vcpu->cpu = cpu; + /* + * Linux uses per-cpu TSS and GDT, so set these when switching + * processors. + */ + vmcs_writel(HOST_TR_BASE, read_tr_base()); /* 22.2.4 */ + get_gdt(&dt); + vmcs_writel(HOST_GDTR_BASE, dt.base); /* 22.2.4 */ + + rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp); + vmcs_writel(HOST_IA32_SYSENTER_ESP, sysenter_esp); /* 22.2.3 */ + } + return vcpu; +} + +static void vmx_vcpu_put(struct kvm_vcpu *vcpu) +{ + put_cpu(); +} + +static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) +{ + return vmcs_readl(GUEST_RFLAGS); +} + +static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) +{ + vmcs_writel(GUEST_RFLAGS, rflags); +} + +static void skip_emulated_instruction(struct kvm_vcpu *vcpu) +{ + unsigned long rip; + u32 interruptibility; + + rip = vmcs_readl(GUEST_RIP); + rip += vmcs_read32(VM_EXIT_INSTRUCTION_LEN); + vmcs_writel(GUEST_RIP, rip); + + /* + * We emulated an instruction, so temporary interrupt blocking + * should be removed, if set. + */ + interruptibility = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); + if (interruptibility & 3) + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + interruptibility & ~3); +} + +static void vmx_inject_gp(struct kvm_vcpu *vcpu, unsigned error_code) +{ + printk(KERN_DEBUG "inject_general_protection: rip 0x%lx\n", + vmcs_readl(GUEST_RIP)); + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + GP_VECTOR | + INTR_TYPE_EXCEPTION | + INTR_INFO_DELIEVER_CODE_MASK | + INTR_INFO_VALID_MASK); +} + +/* + * reads and returns guest's timestamp counter "register" + * guest_tsc = host_tsc + tsc_offset -- 21.3 + */ +static u64 guest_read_tsc(void) +{ + u64 host_tsc, tsc_offset; + + rdtscll(host_tsc); + tsc_offset = vmcs_read64(TSC_OFFSET); + return host_tsc + tsc_offset; +} + +/* + * writes 'guest_tsc' into guest's timestamp counter "register" + * guest_tsc = host_tsc + tsc_offset ==> tsc_offset = guest_tsc - host_tsc + */ +static void guest_write_tsc(u64 guest_tsc) +{ + u64 host_tsc; + + rdtscll(host_tsc); + vmcs_write64(TSC_OFFSET, guest_tsc - host_tsc); +} + +static void reload_tss(void) +{ +#ifndef __x86_64__ + + /* + * VT restores TR but not its size. Useless. + */ + struct descriptor_table gdt; + struct segment_descriptor *descs; + + get_gdt(&gdt); + descs = (void *)gdt.base; + descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ + load_TR_desc(); +#endif +} + +/* + * Reads an msr value (of 'msr_index') into 'pdata'. + * Returns 0 on success, non-0 otherwise. + * Assumes vcpu_load() was already called. + */ +static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) +{ + u64 data; + struct vmx_msr_entry *msr; + + if (!pdata) { + printk(KERN_ERR "BUG: get_msr called with NULL pdata\n"); + return -EINVAL; + } + + switch (msr_index) { +#ifdef __x86_64__ + case MSR_FS_BASE: + data = vmcs_readl(GUEST_FS_BASE); + break; + case MSR_GS_BASE: + data = vmcs_readl(GUEST_GS_BASE); + break; + case MSR_EFER: + data = vcpu->shadow_efer; + break; +#endif + case MSR_IA32_TIME_STAMP_COUNTER: + data = guest_read_tsc(); + break; + case MSR_IA32_SYSENTER_CS: + data = vmcs_read32(GUEST_SYSENTER_CS); + break; + case MSR_IA32_SYSENTER_EIP: + data = vmcs_read32(GUEST_SYSENTER_EIP); + break; + case MSR_IA32_SYSENTER_ESP: + data = vmcs_read32(GUEST_SYSENTER_ESP); + break; + case MSR_IA32_MC0_CTL: + case MSR_IA32_MCG_STATUS: + case MSR_IA32_MCG_CAP: + case MSR_IA32_MC0_MISC: + case MSR_IA32_MC0_MISC+4: + case MSR_IA32_MC0_MISC+8: + case MSR_IA32_MC0_MISC+12: + case MSR_IA32_MC0_MISC+16: + case MSR_IA32_UCODE_REV: + /* MTRR registers */ + case 0xfe: + case 0x200 ... 0x2ff: + data = 0; + break; + case MSR_IA32_APICBASE: + data = vcpu->apic_base; + break; + default: + msr = find_msr_entry(vcpu, msr_index); + if (!msr) { + printk(KERN_ERR "kvm: unhandled rdmsr: %x\n", msr_index); + return 1; + } + data = msr->data; + break; + } + + *pdata = data; + return 0; +} + +/* + * Writes msr value into into the appropriate "register". + * Returns 0 on success, non-0 otherwise. + * Assumes vcpu_load() was already called. + */ +static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) +{ + struct vmx_msr_entry *msr; + switch (msr_index) { +#ifdef __x86_64__ + case MSR_FS_BASE: + vmcs_writel(GUEST_FS_BASE, data); + break; + case MSR_GS_BASE: + vmcs_writel(GUEST_GS_BASE, data); + break; +#endif + case MSR_IA32_SYSENTER_CS: + vmcs_write32(GUEST_SYSENTER_CS, data); + break; + case MSR_IA32_SYSENTER_EIP: + vmcs_write32(GUEST_SYSENTER_EIP, data); + break; + case MSR_IA32_SYSENTER_ESP: + vmcs_write32(GUEST_SYSENTER_ESP, data); + break; +#ifdef __x86_64 + case MSR_EFER: + set_efer(vcpu, data); + break; + case MSR_IA32_MC0_STATUS: + printk(KERN_WARNING "%s: MSR_IA32_MC0_STATUS 0x%llx, nop\n" + , __FUNCTION__, data); + break; +#endif + case MSR_IA32_TIME_STAMP_COUNTER: { + guest_write_tsc(data); + break; + } + case MSR_IA32_UCODE_REV: + case MSR_IA32_UCODE_WRITE: + case 0x200 ... 0x2ff: /* MTRRs */ + break; + case MSR_IA32_APICBASE: + vcpu->apic_base = data; + break; + default: + msr = find_msr_entry(vcpu, msr_index); + if (!msr) { + printk(KERN_ERR "kvm: unhandled wrmsr: 0x%x\n", msr_index); + return 1; + } + msr->data = data; + break; + } + + return 0; +} + +/* + * Sync the rsp and rip registers into the vcpu structure. This allows + * registers to be accessed by indexing vcpu->regs. + */ +static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu) +{ + vcpu->regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP); + vcpu->rip = vmcs_readl(GUEST_RIP); +} + +/* + * Syncs rsp and rip back into the vmcs. Should be called after possible + * modification. + */ +static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu) +{ + vmcs_writel(GUEST_RSP, vcpu->regs[VCPU_REGS_RSP]); + vmcs_writel(GUEST_RIP, vcpu->rip); +} + +static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) +{ + unsigned long dr7 = 0x400; + u32 exception_bitmap; + int old_singlestep; + + exception_bitmap = vmcs_read32(EXCEPTION_BITMAP); + old_singlestep = vcpu->guest_debug.singlestep; + + vcpu->guest_debug.enabled = dbg->enabled; + if (vcpu->guest_debug.enabled) { + int i; + + dr7 |= 0x200; /* exact */ + for (i = 0; i < 4; ++i) { + if (!dbg->breakpoints[i].enabled) + continue; + vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address; + dr7 |= 2 << (i*2); /* global enable */ + dr7 |= 0 << (i*4+16); /* execution breakpoint */ + } + + exception_bitmap |= (1u << 1); /* Trap debug exceptions */ + + vcpu->guest_debug.singlestep = dbg->singlestep; + } else { + exception_bitmap &= ~(1u << 1); /* Ignore debug exceptions */ + vcpu->guest_debug.singlestep = 0; + } + + if (old_singlestep && !vcpu->guest_debug.singlestep) { + unsigned long flags; + + flags = vmcs_readl(GUEST_RFLAGS); + flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); + vmcs_writel(GUEST_RFLAGS, flags); + } + + vmcs_write32(EXCEPTION_BITMAP, exception_bitmap); + vmcs_writel(GUEST_DR7, dr7); + + return 0; +} + +static __init int cpu_has_kvm_support(void) +{ + unsigned long ecx = cpuid_ecx(1); + return test_bit(5, &ecx); /* CPUID.1:ECX.VMX[bit 5] -> VT */ +} + +static __init int vmx_disabled_by_bios(void) +{ + u64 msr; + + rdmsrl(MSR_IA32_FEATURE_CONTROL, msr); + return (msr & 5) == 1; /* locked but not enabled */ +} + +static __init void hardware_enable(void *garbage) +{ + int cpu = raw_smp_processor_id(); + u64 phys_addr = __pa(per_cpu(vmxarea, cpu)); + u64 old; + + rdmsrl(MSR_IA32_FEATURE_CONTROL, old); + if ((old & 5) == 0) + /* enable and lock */ + wrmsrl(MSR_IA32_FEATURE_CONTROL, old | 5); + write_cr4(read_cr4() | CR4_VMXE); /* FIXME: not cpu hotplug safe */ + asm volatile (ASM_VMX_VMXON_RAX : : "a"(&phys_addr), "m"(phys_addr) + : "memory", "cc"); +} + +static void hardware_disable(void *garbage) +{ + asm volatile (ASM_VMX_VMXOFF : : : "cc"); +} + +static __init void setup_vmcs_descriptor(void) +{ + u32 vmx_msr_low, vmx_msr_high; + + rdmsr(MSR_IA32_VMX_BASIC_MSR, vmx_msr_low, vmx_msr_high); + vmcs_descriptor.size = vmx_msr_high & 0x1fff; + vmcs_descriptor.order = get_order(vmcs_descriptor.size); + vmcs_descriptor.revision_id = vmx_msr_low; +}; + +static struct vmcs *alloc_vmcs_cpu(int cpu) +{ + int node = cpu_to_node(cpu); + struct page *pages; + struct vmcs *vmcs; + + pages = alloc_pages_node(node, GFP_KERNEL, vmcs_descriptor.order); + if (!pages) + return NULL; + vmcs = page_address(pages); + memset(vmcs, 0, vmcs_descriptor.size); + vmcs->revision_id = vmcs_descriptor.revision_id; /* vmcs revision id */ + return vmcs; +} + +static struct vmcs *alloc_vmcs(void) +{ + return alloc_vmcs_cpu(smp_processor_id()); +} + +static void free_vmcs(struct vmcs *vmcs) +{ + free_pages((unsigned long)vmcs, vmcs_descriptor.order); +} + +static __exit void free_kvm_area(void) +{ + int cpu; + + for_each_online_cpu(cpu) + free_vmcs(per_cpu(vmxarea, cpu)); +} + +extern struct vmcs *alloc_vmcs_cpu(int cpu); + +static __init int alloc_kvm_area(void) +{ + int cpu; + + for_each_online_cpu(cpu) { + struct vmcs *vmcs; + + vmcs = alloc_vmcs_cpu(cpu); + if (!vmcs) { + free_kvm_area(); + return -ENOMEM; + } + + per_cpu(vmxarea, cpu) = vmcs; + } + return 0; +} + +static __init int hardware_setup(void) +{ + setup_vmcs_descriptor(); + return alloc_kvm_area(); +} + +static __exit void hardware_unsetup(void) +{ + free_kvm_area(); +} + +static void update_exception_bitmap(struct kvm_vcpu *vcpu) +{ + if (vcpu->rmode.active) + vmcs_write32(EXCEPTION_BITMAP, ~0); + else + vmcs_write32(EXCEPTION_BITMAP, 1 << PF_VECTOR); +} + +static void fix_pmode_dataseg(int seg, struct kvm_save_segment *save) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + + if (vmcs_readl(sf->base) == save->base) { + vmcs_write16(sf->selector, save->selector); + vmcs_writel(sf->base, save->base); + vmcs_write32(sf->limit, save->limit); + vmcs_write32(sf->ar_bytes, save->ar); + } else { + u32 dpl = (vmcs_read16(sf->selector) & SELECTOR_RPL_MASK) + << AR_DPL_SHIFT; + vmcs_write32(sf->ar_bytes, 0x93 | dpl); + } +} + +static void enter_pmode(struct kvm_vcpu *vcpu) +{ + unsigned long flags; + + vcpu->rmode.active = 0; + + vmcs_writel(GUEST_TR_BASE, vcpu->rmode.tr.base); + vmcs_write32(GUEST_TR_LIMIT, vcpu->rmode.tr.limit); + vmcs_write32(GUEST_TR_AR_BYTES, vcpu->rmode.tr.ar); + + flags = vmcs_readl(GUEST_RFLAGS); + flags &= ~(IOPL_MASK | X86_EFLAGS_VM); + flags |= (vcpu->rmode.save_iopl << IOPL_SHIFT); + vmcs_writel(GUEST_RFLAGS, flags); + + vmcs_writel(GUEST_CR4, (vmcs_readl(GUEST_CR4) & ~CR4_VME_MASK) | + (vmcs_readl(CR4_READ_SHADOW) & CR4_VME_MASK)); + + update_exception_bitmap(vcpu); + + fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->rmode.es); + fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->rmode.ds); + fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->rmode.gs); + fix_pmode_dataseg(VCPU_SREG_FS, &vcpu->rmode.fs); + + vmcs_write16(GUEST_SS_SELECTOR, 0); + vmcs_write32(GUEST_SS_AR_BYTES, 0x93); + + vmcs_write16(GUEST_CS_SELECTOR, + vmcs_read16(GUEST_CS_SELECTOR) & ~SELECTOR_RPL_MASK); + vmcs_write32(GUEST_CS_AR_BYTES, 0x9b); +} + +static int rmode_tss_base(struct kvm* kvm) +{ + gfn_t base_gfn = kvm->memslots[0].base_gfn + kvm->memslots[0].npages - 3; + return base_gfn << PAGE_SHIFT; +} + +static void fix_rmode_seg(int seg, struct kvm_save_segment *save) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + + save->selector = vmcs_read16(sf->selector); + save->base = vmcs_readl(sf->base); + save->limit = vmcs_read32(sf->limit); + save->ar = vmcs_read32(sf->ar_bytes); + vmcs_write16(sf->selector, vmcs_readl(sf->base) >> 4); + vmcs_write32(sf->limit, 0xffff); + vmcs_write32(sf->ar_bytes, 0xf3); +} + +static void enter_rmode(struct kvm_vcpu *vcpu) +{ + unsigned long flags; + + vcpu->rmode.active = 1; + + vcpu->rmode.tr.base = vmcs_readl(GUEST_TR_BASE); + vmcs_writel(GUEST_TR_BASE, rmode_tss_base(vcpu->kvm)); + + vcpu->rmode.tr.limit = vmcs_read32(GUEST_TR_LIMIT); + vmcs_write32(GUEST_TR_LIMIT, RMODE_TSS_SIZE - 1); + + vcpu->rmode.tr.ar = vmcs_read32(GUEST_TR_AR_BYTES); + vmcs_write32(GUEST_TR_AR_BYTES, 0x008b); + + flags = vmcs_readl(GUEST_RFLAGS); + vcpu->rmode.save_iopl = (flags & IOPL_MASK) >> IOPL_SHIFT; + + flags |= IOPL_MASK | X86_EFLAGS_VM; + + vmcs_writel(GUEST_RFLAGS, flags); + vmcs_writel(GUEST_CR4, vmcs_readl(GUEST_CR4) | CR4_VME_MASK); + update_exception_bitmap(vcpu); + + vmcs_write16(GUEST_SS_SELECTOR, vmcs_readl(GUEST_SS_BASE) >> 4); + vmcs_write32(GUEST_SS_LIMIT, 0xffff); + vmcs_write32(GUEST_SS_AR_BYTES, 0xf3); + + vmcs_write32(GUEST_CS_AR_BYTES, 0xf3); + vmcs_write16(GUEST_CS_SELECTOR, vmcs_readl(GUEST_CS_BASE) >> 4); + + fix_rmode_seg(VCPU_SREG_ES, &vcpu->rmode.es); + fix_rmode_seg(VCPU_SREG_DS, &vcpu->rmode.ds); + fix_rmode_seg(VCPU_SREG_GS, &vcpu->rmode.gs); + fix_rmode_seg(VCPU_SREG_FS, &vcpu->rmode.fs); +} + +#ifdef __x86_64__ + +static void enter_lmode(struct kvm_vcpu *vcpu) +{ + u32 guest_tr_ar; + + guest_tr_ar = vmcs_read32(GUEST_TR_AR_BYTES); + if ((guest_tr_ar & AR_TYPE_MASK) != AR_TYPE_BUSY_64_TSS) { + printk(KERN_DEBUG "%s: tss fixup for long mode. \n", + __FUNCTION__); + vmcs_write32(GUEST_TR_AR_BYTES, + (guest_tr_ar & ~AR_TYPE_MASK) + | AR_TYPE_BUSY_64_TSS); + } + + vcpu->shadow_efer |= EFER_LMA; + + find_msr_entry(vcpu, MSR_EFER)->data |= EFER_LMA | EFER_LME; + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) + | VM_ENTRY_CONTROLS_IA32E_MASK); +} + +static void exit_lmode(struct kvm_vcpu *vcpu) +{ + vcpu->shadow_efer &= ~EFER_LMA; + + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) + & ~VM_ENTRY_CONTROLS_IA32E_MASK); +} + +#endif + +static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +{ + if (vcpu->rmode.active && (cr0 & CR0_PE_MASK)) + enter_pmode(vcpu); + + if (!vcpu->rmode.active && !(cr0 & CR0_PE_MASK)) + enter_rmode(vcpu); + +#ifdef __x86_64__ + if (vcpu->shadow_efer & EFER_LME) { + if (!is_paging(vcpu) && (cr0 & CR0_PG_MASK)) + enter_lmode(vcpu); + if (is_paging(vcpu) && !(cr0 & CR0_PG_MASK)) + exit_lmode(vcpu); + } +#endif + + vmcs_writel(CR0_READ_SHADOW, cr0); + vmcs_writel(GUEST_CR0, + (cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON); + vcpu->cr0 = cr0; +} + +/* + * Used when restoring the VM to avoid corrupting segment registers + */ +static void vmx_set_cr0_no_modeswitch(struct kvm_vcpu *vcpu, unsigned long cr0) +{ + vcpu->rmode.active = ((cr0 & CR0_PE_MASK) == 0); + update_exception_bitmap(vcpu); + vmcs_writel(CR0_READ_SHADOW, cr0); + vmcs_writel(GUEST_CR0, + (cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON); + vcpu->cr0 = cr0; +} + +static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +{ + vmcs_writel(GUEST_CR3, cr3); +} + +static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +{ + vmcs_writel(CR4_READ_SHADOW, cr4); + vmcs_writel(GUEST_CR4, cr4 | (vcpu->rmode.active ? + KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON)); + vcpu->cr4 = cr4; +} + +#ifdef __x86_64__ + +static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer) +{ + struct vmx_msr_entry *msr = find_msr_entry(vcpu, MSR_EFER); + + vcpu->shadow_efer = efer; + if (efer & EFER_LMA) { + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) | + VM_ENTRY_CONTROLS_IA32E_MASK); + msr->data = efer; + + } else { + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) & + ~VM_ENTRY_CONTROLS_IA32E_MASK); + + msr->data = efer & ~EFER_LME; + } +} + +#endif + +static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + + return vmcs_readl(sf->base); +} + +static void vmx_get_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + u32 ar; + + var->base = vmcs_readl(sf->base); + var->limit = vmcs_read32(sf->limit); + var->selector = vmcs_read16(sf->selector); + ar = vmcs_read32(sf->ar_bytes); + if (ar & AR_UNUSABLE_MASK) + ar = 0; + var->type = ar & 15; + var->s = (ar >> 4) & 1; + var->dpl = (ar >> 5) & 3; + var->present = (ar >> 7) & 1; + var->avl = (ar >> 12) & 1; + var->l = (ar >> 13) & 1; + var->db = (ar >> 14) & 1; + var->g = (ar >> 15) & 1; + var->unusable = (ar >> 16) & 1; +} + +static void vmx_set_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + u32 ar; + + vmcs_writel(sf->base, var->base); + vmcs_write32(sf->limit, var->limit); + vmcs_write16(sf->selector, var->selector); + if (var->unusable) + ar = 1 << 16; + else { + ar = var->type & 15; + ar |= (var->s & 1) << 4; + ar |= (var->dpl & 3) << 5; + ar |= (var->present & 1) << 7; + ar |= (var->avl & 1) << 12; + ar |= (var->l & 1) << 13; + ar |= (var->db & 1) << 14; + ar |= (var->g & 1) << 15; + } + vmcs_write32(sf->ar_bytes, ar); +} + +static int vmx_is_long_mode(struct kvm_vcpu *vcpu) +{ + return vmcs_read32(VM_ENTRY_CONTROLS) & VM_ENTRY_CONTROLS_IA32E_MASK; +} + +static void vmx_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) +{ + u32 ar = vmcs_read32(GUEST_CS_AR_BYTES); + + *db = (ar >> 14) & 1; + *l = (ar >> 13) & 1; +} + +static void vmx_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + dt->limit = vmcs_read32(GUEST_IDTR_LIMIT); + dt->base = vmcs_readl(GUEST_IDTR_BASE); +} + +static void vmx_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + vmcs_write32(GUEST_IDTR_LIMIT, dt->limit); + vmcs_writel(GUEST_IDTR_BASE, dt->base); +} + +static void vmx_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + dt->limit = vmcs_read32(GUEST_GDTR_LIMIT); + dt->base = vmcs_readl(GUEST_GDTR_BASE); +} + +static void vmx_set_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) +{ + vmcs_write32(GUEST_GDTR_LIMIT, dt->limit); + vmcs_writel(GUEST_GDTR_BASE, dt->base); +} + +static int init_rmode_tss(struct kvm* kvm) +{ + struct page *p1, *p2, *p3; + gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT; + char *page; + + p1 = _gfn_to_page(kvm, fn++); + p2 = _gfn_to_page(kvm, fn++); + p3 = _gfn_to_page(kvm, fn); + + if (!p1 || !p2 || !p3) { + kvm_printf(kvm,"%s: gfn_to_page failed\n", __FUNCTION__); + return 0; + } + + page = kmap_atomic(p1, KM_USER0); + memset(page, 0, PAGE_SIZE); + *(u16*)(page + 0x66) = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE; + kunmap_atomic(page, KM_USER0); + + page = kmap_atomic(p2, KM_USER0); + memset(page, 0, PAGE_SIZE); + kunmap_atomic(page, KM_USER0); + + page = kmap_atomic(p3, KM_USER0); + memset(page, 0, PAGE_SIZE); + *(page + RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1) = ~0; + kunmap_atomic(page, KM_USER0); + + return 1; +} + +static void vmcs_write32_fixedbits(u32 msr, u32 vmcs_field, u32 val) +{ + u32 msr_high, msr_low; + + rdmsr(msr, msr_low, msr_high); + + val &= msr_high; + val |= msr_low; + vmcs_write32(vmcs_field, val); +} + +static void seg_setup(int seg) +{ + struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; + + vmcs_write16(sf->selector, 0); + vmcs_writel(sf->base, 0); + vmcs_write32(sf->limit, 0xffff); + vmcs_write32(sf->ar_bytes, 0x93); +} + +/* + * Sets up the vmcs for emulated real mode. + */ +static int vmx_vcpu_setup(struct kvm_vcpu *vcpu) +{ + u32 host_sysenter_cs; + u32 junk; + unsigned long a; + struct descriptor_table dt; + int i; + int ret = 0; + int nr_good_msrs; + extern asmlinkage void kvm_vmx_return(void); + + if (!init_rmode_tss(vcpu->kvm)) { + ret = -ENOMEM; + goto out; + } + + memset(vcpu->regs, 0, sizeof(vcpu->regs)); + vcpu->regs[VCPU_REGS_RDX] = get_rdx_init_val(); + vcpu->cr8 = 0; + vcpu->apic_base = 0xfee00000 | + /*for vcpu 0*/ MSR_IA32_APICBASE_BSP | + MSR_IA32_APICBASE_ENABLE; + + fx_init(vcpu); + + /* + * GUEST_CS_BASE should really be 0xffff0000, but VT vm86 mode + * insists on having GUEST_CS_BASE == GUEST_CS_SELECTOR << 4. Sigh. + */ + vmcs_write16(GUEST_CS_SELECTOR, 0xf000); + vmcs_writel(GUEST_CS_BASE, 0x000f0000); + vmcs_write32(GUEST_CS_LIMIT, 0xffff); + vmcs_write32(GUEST_CS_AR_BYTES, 0x9b); + + seg_setup(VCPU_SREG_DS); + seg_setup(VCPU_SREG_ES); + seg_setup(VCPU_SREG_FS); + seg_setup(VCPU_SREG_GS); + seg_setup(VCPU_SREG_SS); + + vmcs_write16(GUEST_TR_SELECTOR, 0); + vmcs_writel(GUEST_TR_BASE, 0); + vmcs_write32(GUEST_TR_LIMIT, 0xffff); + vmcs_write32(GUEST_TR_AR_BYTES, 0x008b); + + vmcs_write16(GUEST_LDTR_SELECTOR, 0); + vmcs_writel(GUEST_LDTR_BASE, 0); + vmcs_write32(GUEST_LDTR_LIMIT, 0xffff); + vmcs_write32(GUEST_LDTR_AR_BYTES, 0x00082); + + vmcs_write32(GUEST_SYSENTER_CS, 0); + vmcs_writel(GUEST_SYSENTER_ESP, 0); + vmcs_writel(GUEST_SYSENTER_EIP, 0); + + vmcs_writel(GUEST_RFLAGS, 0x02); + vmcs_writel(GUEST_RIP, 0xfff0); + vmcs_writel(GUEST_RSP, 0); + + vmcs_writel(GUEST_CR3, 0); + + //todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 + vmcs_writel(GUEST_DR7, 0x400); + + vmcs_writel(GUEST_GDTR_BASE, 0); + vmcs_write32(GUEST_GDTR_LIMIT, 0xffff); + + vmcs_writel(GUEST_IDTR_BASE, 0); + vmcs_write32(GUEST_IDTR_LIMIT, 0xffff); + + vmcs_write32(GUEST_ACTIVITY_STATE, 0); + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0); + vmcs_write32(GUEST_PENDING_DBG_EXCEPTIONS, 0); + + /* I/O */ + vmcs_write64(IO_BITMAP_A, 0); + vmcs_write64(IO_BITMAP_B, 0); + + guest_write_tsc(0); + + vmcs_write64(VMCS_LINK_POINTER, -1ull); /* 22.3.1.5 */ + + /* Special registers */ + vmcs_write64(GUEST_IA32_DEBUGCTL, 0); + + /* Control */ + vmcs_write32_fixedbits(MSR_IA32_VMX_PINBASED_CTLS_MSR, + PIN_BASED_VM_EXEC_CONTROL, + PIN_BASED_EXT_INTR_MASK /* 20.6.1 */ + | PIN_BASED_NMI_EXITING /* 20.6.1 */ + ); + vmcs_write32_fixedbits(MSR_IA32_VMX_PROCBASED_CTLS_MSR, + CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_HLT_EXITING /* 20.6.2 */ + | CPU_BASED_CR8_LOAD_EXITING /* 20.6.2 */ + | CPU_BASED_CR8_STORE_EXITING /* 20.6.2 */ + | CPU_BASED_UNCOND_IO_EXITING /* 20.6.2 */ + | CPU_BASED_INVDPG_EXITING + | CPU_BASED_MOV_DR_EXITING + | CPU_BASED_USE_TSC_OFFSETING /* 21.3 */ + ); + + vmcs_write32(EXCEPTION_BITMAP, 1 << PF_VECTOR); + vmcs_write32(PAGE_FAULT_ERROR_CODE_MASK, 0); + vmcs_write32(PAGE_FAULT_ERROR_CODE_MATCH, 0); + vmcs_write32(CR3_TARGET_COUNT, 0); /* 22.2.1 */ + + vmcs_writel(HOST_CR0, read_cr0()); /* 22.2.3 */ + vmcs_writel(HOST_CR4, read_cr4()); /* 22.2.3, 22.2.5 */ + vmcs_writel(HOST_CR3, read_cr3()); /* 22.2.3 FIXME: shadow tables */ + + vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS); /* 22.2.4 */ + vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ + vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS); /* 22.2.4 */ + vmcs_write16(HOST_FS_SELECTOR, read_fs()); /* 22.2.4 */ + vmcs_write16(HOST_GS_SELECTOR, read_gs()); /* 22.2.4 */ + vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ +#ifdef __x86_64__ + rdmsrl(MSR_FS_BASE, a); + vmcs_writel(HOST_FS_BASE, a); /* 22.2.4 */ + rdmsrl(MSR_GS_BASE, a); + vmcs_writel(HOST_GS_BASE, a); /* 22.2.4 */ +#else + vmcs_writel(HOST_FS_BASE, 0); /* 22.2.4 */ + vmcs_writel(HOST_GS_BASE, 0); /* 22.2.4 */ +#endif + + vmcs_write16(HOST_TR_SELECTOR, GDT_ENTRY_TSS*8); /* 22.2.4 */ + + get_idt(&dt); + vmcs_writel(HOST_IDTR_BASE, dt.base); /* 22.2.4 */ + + + vmcs_writel(HOST_RIP, (unsigned long)kvm_vmx_return); /* 22.2.5 */ + + rdmsr(MSR_IA32_SYSENTER_CS, host_sysenter_cs, junk); + vmcs_write32(HOST_IA32_SYSENTER_CS, host_sysenter_cs); + rdmsrl(MSR_IA32_SYSENTER_ESP, a); + vmcs_writel(HOST_IA32_SYSENTER_ESP, a); /* 22.2.3 */ + rdmsrl(MSR_IA32_SYSENTER_EIP, a); + vmcs_writel(HOST_IA32_SYSENTER_EIP, a); /* 22.2.3 */ + + ret = -ENOMEM; + vcpu->guest_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!vcpu->guest_msrs) + goto out; + vcpu->host_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!vcpu->host_msrs) + goto out_free_guest_msrs; + + for (i = 0; i < NR_VMX_MSR; ++i) { + u32 index = vmx_msr_index[i]; + u32 data_low, data_high; + u64 data; + int j = vcpu->nmsrs; + + if (rdmsr_safe(index, &data_low, &data_high) < 0) + continue; + data = data_low | ((u64)data_high << 32); + vcpu->host_msrs[j].index = index; + vcpu->host_msrs[j].reserved = 0; + vcpu->host_msrs[j].data = data; + vcpu->guest_msrs[j] = vcpu->host_msrs[j]; + ++vcpu->nmsrs; + } + printk(KERN_DEBUG "kvm: msrs: %d\n", vcpu->nmsrs); + + nr_good_msrs = vcpu->nmsrs - NR_BAD_MSRS; + vmcs_writel(VM_ENTRY_MSR_LOAD_ADDR, + virt_to_phys(vcpu->guest_msrs + NR_BAD_MSRS)); + vmcs_writel(VM_EXIT_MSR_STORE_ADDR, + virt_to_phys(vcpu->guest_msrs + NR_BAD_MSRS)); + vmcs_writel(VM_EXIT_MSR_LOAD_ADDR, + virt_to_phys(vcpu->host_msrs + NR_BAD_MSRS)); + vmcs_write32_fixedbits(MSR_IA32_VMX_EXIT_CTLS_MSR, VM_EXIT_CONTROLS, + (HOST_IS_64 << 9)); /* 22.2,1, 20.7.1 */ + vmcs_write32(VM_EXIT_MSR_STORE_COUNT, nr_good_msrs); /* 22.2.2 */ + vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, nr_good_msrs); /* 22.2.2 */ + vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, nr_good_msrs); /* 22.2.2 */ + + + /* 22.2.1, 20.8.1 */ + vmcs_write32_fixedbits(MSR_IA32_VMX_ENTRY_CTLS_MSR, + VM_ENTRY_CONTROLS, 0); + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, 0); /* 22.2.1 */ + + vmcs_writel(VIRTUAL_APIC_PAGE_ADDR, 0); + vmcs_writel(TPR_THRESHOLD, 0); + + vmcs_writel(CR0_GUEST_HOST_MASK, KVM_GUEST_CR0_MASK); + vmcs_writel(CR4_GUEST_HOST_MASK, KVM_GUEST_CR4_MASK); + + vcpu->cr0 = 0x60000010; + vmx_set_cr0(vcpu, vcpu->cr0); // enter rmode + vmx_set_cr4(vcpu, 0); +#ifdef __x86_64__ + vmx_set_efer(vcpu, 0); +#endif + + return 0; + +out_free_guest_msrs: + kfree(vcpu->guest_msrs); +out: + return ret; +} + +static void inject_rmode_irq(struct kvm_vcpu *vcpu, int irq) +{ + u16 ent[2]; + u16 cs; + u16 ip; + unsigned long flags; + unsigned long ss_base = vmcs_readl(GUEST_SS_BASE); + u16 sp = vmcs_readl(GUEST_RSP); + u32 ss_limit = vmcs_read32(GUEST_SS_LIMIT); + + if (sp > ss_limit || sp - 6 > sp) { + vcpu_printf(vcpu, "%s: #SS, rsp 0x%lx ss 0x%lx limit 0x%x\n", + __FUNCTION__, + vmcs_readl(GUEST_RSP), + vmcs_readl(GUEST_SS_BASE), + vmcs_read32(GUEST_SS_LIMIT)); + return; + } + + if (kvm_read_guest(vcpu, irq * sizeof(ent), sizeof(ent), &ent) != + sizeof(ent)) { + vcpu_printf(vcpu, "%s: read guest err\n", __FUNCTION__); + return; + } + + flags = vmcs_readl(GUEST_RFLAGS); + cs = vmcs_readl(GUEST_CS_BASE) >> 4; + ip = vmcs_readl(GUEST_RIP); + + + if (kvm_write_guest(vcpu, ss_base + sp - 2, 2, &flags) != 2 || + kvm_write_guest(vcpu, ss_base + sp - 4, 2, &cs) != 2 || + kvm_write_guest(vcpu, ss_base + sp - 6, 2, &ip) != 2) { + vcpu_printf(vcpu, "%s: write guest err\n", __FUNCTION__); + return; + } + + vmcs_writel(GUEST_RFLAGS, flags & + ~( X86_EFLAGS_IF | X86_EFLAGS_AC | X86_EFLAGS_TF)); + vmcs_write16(GUEST_CS_SELECTOR, ent[1]) ; + vmcs_writel(GUEST_CS_BASE, ent[1] << 4); + vmcs_writel(GUEST_RIP, ent[0]); + vmcs_writel(GUEST_RSP, (vmcs_readl(GUEST_RSP) & ~0xffff) | (sp - 6)); +} + +static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) +{ + int word_index = __ffs(vcpu->irq_summary); + int bit_index = __ffs(vcpu->irq_pending[word_index]); + int irq = word_index * BITS_PER_LONG + bit_index; + + clear_bit(bit_index, &vcpu->irq_pending[word_index]); + if (!vcpu->irq_pending[word_index]) + clear_bit(word_index, &vcpu->irq_summary); + + if (vcpu->rmode.active) { + inject_rmode_irq(vcpu, irq); + return; + } + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + irq | INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK); +} + +static void kvm_try_inject_irq(struct kvm_vcpu *vcpu) +{ + if ((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) + && (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0) + /* + * Interrupts enabled, and not blocked by sti or mov ss. Good. + */ + kvm_do_inject_irq(vcpu); + else + /* + * Interrupts blocked. Wait for unblock. + */ + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, + vmcs_read32(CPU_BASED_VM_EXEC_CONTROL) + | CPU_BASED_VIRTUAL_INTR_PENDING); +} + +static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu) +{ + struct kvm_guest_debug *dbg = &vcpu->guest_debug; + + set_debugreg(dbg->bp[0], 0); + set_debugreg(dbg->bp[1], 1); + set_debugreg(dbg->bp[2], 2); + set_debugreg(dbg->bp[3], 3); + + if (dbg->singlestep) { + unsigned long flags; + + flags = vmcs_readl(GUEST_RFLAGS); + flags |= X86_EFLAGS_TF | X86_EFLAGS_RF; + vmcs_writel(GUEST_RFLAGS, flags); + } +} + +static int handle_rmode_exception(struct kvm_vcpu *vcpu, + int vec, u32 err_code) +{ + if (!vcpu->rmode.active) + return 0; + + if (vec == GP_VECTOR && err_code == 0) + if (emulate_instruction(vcpu, NULL, 0, 0) == EMULATE_DONE) + return 1; + return 0; +} + +static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 intr_info, error_code; + unsigned long cr2, rip; + u32 vect_info; + enum emulation_result er; + + vect_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); + intr_info = vmcs_read32(VM_EXIT_INTR_INFO); + + if ((vect_info & VECTORING_INFO_VALID_MASK) && + !is_page_fault(intr_info)) { + printk(KERN_ERR "%s: unexpected, vectoring info 0x%x " + "intr info 0x%x\n", __FUNCTION__, vect_info, intr_info); + } + + if (is_external_interrupt(vect_info)) { + int irq = vect_info & VECTORING_INFO_VECTOR_MASK; + set_bit(irq, vcpu->irq_pending); + set_bit(irq / BITS_PER_LONG, &vcpu->irq_summary); + } + + if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) { /* nmi */ + asm ("int $2"); + return 1; + } + error_code = 0; + rip = vmcs_readl(GUEST_RIP); + if (intr_info & INTR_INFO_DELIEVER_CODE_MASK) + error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE); + if (is_page_fault(intr_info)) { + cr2 = vmcs_readl(EXIT_QUALIFICATION); + + spin_lock(&vcpu->kvm->lock); + if (!vcpu->mmu.page_fault(vcpu, cr2, error_code)) { + spin_unlock(&vcpu->kvm->lock); + return 1; + } + + er = emulate_instruction(vcpu, kvm_run, cr2, error_code); + spin_unlock(&vcpu->kvm->lock); + + switch (er) { + case EMULATE_DONE: + return 1; + case EMULATE_DO_MMIO: + ++kvm_stat.mmio_exits; + kvm_run->exit_reason = KVM_EXIT_MMIO; + return 0; + case EMULATE_FAIL: + vcpu_printf(vcpu, "%s: emulate fail\n", __FUNCTION__); + break; + default: + BUG(); + } + } + + if (vcpu->rmode.active && + handle_rmode_exception(vcpu, intr_info & INTR_INFO_VECTOR_MASK, + error_code)) + return 1; + + if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) == (INTR_TYPE_EXCEPTION | 1)) { + kvm_run->exit_reason = KVM_EXIT_DEBUG; + return 0; + } + kvm_run->exit_reason = KVM_EXIT_EXCEPTION; + kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK; + kvm_run->ex.error_code = error_code; + return 0; +} + +static int handle_external_interrupt(struct kvm_vcpu *vcpu, + struct kvm_run *kvm_run) +{ + ++kvm_stat.irq_exits; + return 1; +} + + +static int get_io_count(struct kvm_vcpu *vcpu, u64 *count) +{ + u64 inst; + gva_t rip; + int countr_size; + int i, n; + + if ((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_VM)) { + countr_size = 2; + } else { + u32 cs_ar = vmcs_read32(GUEST_CS_AR_BYTES); + + countr_size = (cs_ar & AR_L_MASK) ? 8: + (cs_ar & AR_DB_MASK) ? 4: 2; + } + + rip = vmcs_readl(GUEST_RIP); + if (countr_size != 8) + rip += vmcs_readl(GUEST_CS_BASE); + + n = kvm_read_guest(vcpu, rip, sizeof(inst), &inst); + + for (i = 0; i < n; i++) { + switch (((u8*)&inst)[i]) { + case 0xf0: + case 0xf2: + case 0xf3: + case 0x2e: + case 0x36: + case 0x3e: + case 0x26: + case 0x64: + case 0x65: + case 0x66: + break; + case 0x67: + countr_size = (countr_size == 2) ? 4: (countr_size >> 1); + default: + goto done; + } + } + return 0; +done: + countr_size *= 8; + *count = vcpu->regs[VCPU_REGS_RCX] & (~0ULL >> (64 - countr_size)); + return 1; +} + +static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification; + + ++kvm_stat.io_exits; + exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + kvm_run->exit_reason = KVM_EXIT_IO; + if (exit_qualification & 8) + kvm_run->io.direction = KVM_EXIT_IO_IN; + else + kvm_run->io.direction = KVM_EXIT_IO_OUT; + kvm_run->io.size = (exit_qualification & 7) + 1; + kvm_run->io.string = (exit_qualification & 16) != 0; + kvm_run->io.string_down + = (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_DF) != 0; + kvm_run->io.rep = (exit_qualification & 32) != 0; + kvm_run->io.port = exit_qualification >> 16; + if (kvm_run->io.string) { + if (!get_io_count(vcpu, &kvm_run->io.count)) + return 1; + kvm_run->io.address = vmcs_readl(GUEST_LINEAR_ADDRESS); + } else + kvm_run->io.value = vcpu->regs[VCPU_REGS_RAX]; /* rax */ + return 0; +} + +static int handle_invlpg(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 address = vmcs_read64(EXIT_QUALIFICATION); + int instruction_length = vmcs_read32(VM_EXIT_INSTRUCTION_LEN); + spin_lock(&vcpu->kvm->lock); + vcpu->mmu.inval_page(vcpu, address); + spin_unlock(&vcpu->kvm->lock); + vmcs_writel(GUEST_RIP, vmcs_readl(GUEST_RIP) + instruction_length); + return 1; +} + +static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification; + int cr; + int reg; + + exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + cr = exit_qualification & 15; + reg = (exit_qualification >> 8) & 15; + switch ((exit_qualification >> 4) & 3) { + case 0: /* mov to cr */ + switch (cr) { + case 0: + vcpu_load_rsp_rip(vcpu); + set_cr0(vcpu, vcpu->regs[reg]); + skip_emulated_instruction(vcpu); + return 1; + case 3: + vcpu_load_rsp_rip(vcpu); + set_cr3(vcpu, vcpu->regs[reg]); + skip_emulated_instruction(vcpu); + return 1; + case 4: + vcpu_load_rsp_rip(vcpu); + set_cr4(vcpu, vcpu->regs[reg]); + skip_emulated_instruction(vcpu); + return 1; + case 8: + vcpu_load_rsp_rip(vcpu); + set_cr8(vcpu, vcpu->regs[reg]); + skip_emulated_instruction(vcpu); + return 1; + }; + break; + case 1: /*mov from cr*/ + switch (cr) { + case 3: + vcpu_load_rsp_rip(vcpu); + vcpu->regs[reg] = vcpu->cr3; + vcpu_put_rsp_rip(vcpu); + skip_emulated_instruction(vcpu); + return 1; + case 8: + printk(KERN_DEBUG "handle_cr: read CR8 " + "cpu erratum AA15\n"); + vcpu_load_rsp_rip(vcpu); + vcpu->regs[reg] = vcpu->cr8; + vcpu_put_rsp_rip(vcpu); + skip_emulated_instruction(vcpu); + return 1; + } + break; + case 3: /* lmsw */ + lmsw(vcpu, (exit_qualification >> LMSW_SOURCE_DATA_SHIFT) & 0x0f); + + skip_emulated_instruction(vcpu); + return 1; + default: + break; + } + kvm_run->exit_reason = 0; + printk(KERN_ERR "kvm: unhandled control register: op %d cr %d\n", + (int)(exit_qualification >> 4) & 3, cr); + return 0; +} + +static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u64 exit_qualification; + unsigned long val; + int dr, reg; + + /* + * FIXME: this code assumes the host is debugging the guest. + * need to deal with guest debugging itself too. + */ + exit_qualification = vmcs_read64(EXIT_QUALIFICATION); + dr = exit_qualification & 7; + reg = (exit_qualification >> 8) & 15; + vcpu_load_rsp_rip(vcpu); + if (exit_qualification & 16) { + /* mov from dr */ + switch (dr) { + case 6: + val = 0xffff0ff0; + break; + case 7: + val = 0x400; + break; + default: + val = 0; + } + vcpu->regs[reg] = val; + } else { + /* mov to dr */ + } + vcpu_put_rsp_rip(vcpu); + skip_emulated_instruction(vcpu); + return 1; +} + +static int handle_cpuid(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + kvm_run->exit_reason = KVM_EXIT_CPUID; + return 0; +} + +static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 ecx = vcpu->regs[VCPU_REGS_RCX]; + u64 data; + + if (vmx_get_msr(vcpu, ecx, &data)) { + vmx_inject_gp(vcpu, 0); + return 1; + } + + /* FIXME: handling of bits 32:63 of rax, rdx */ + vcpu->regs[VCPU_REGS_RAX] = data & -1u; + vcpu->regs[VCPU_REGS_RDX] = (data >> 32) & -1u; + skip_emulated_instruction(vcpu); + return 1; +} + +static int handle_wrmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 ecx = vcpu->regs[VCPU_REGS_RCX]; + u64 data = (vcpu->regs[VCPU_REGS_RAX] & -1u) + | ((u64)(vcpu->regs[VCPU_REGS_RDX] & -1u) << 32); + + if (vmx_set_msr(vcpu, ecx, data) != 0) { + vmx_inject_gp(vcpu, 0); + return 1; + } + + skip_emulated_instruction(vcpu); + return 1; +} + +static int handle_interrupt_window(struct kvm_vcpu *vcpu, + struct kvm_run *kvm_run) +{ + /* Turn off interrupt window reporting. */ + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, + vmcs_read32(CPU_BASED_VM_EXEC_CONTROL) + & ~CPU_BASED_VIRTUAL_INTR_PENDING); + return 1; +} + +static int handle_halt(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + skip_emulated_instruction(vcpu); + if (vcpu->irq_summary && (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF)) + return 1; + + kvm_run->exit_reason = KVM_EXIT_HLT; + return 0; +} + +/* + * The exit handlers return 1 if the exit was handled fully and guest execution + * may resume. Otherwise they set the kvm_run parameter to indicate what needs + * to be done to userspace and return 0. + */ +static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, + struct kvm_run *kvm_run) = { + [EXIT_REASON_EXCEPTION_NMI] = handle_exception, + [EXIT_REASON_EXTERNAL_INTERRUPT] = handle_external_interrupt, + [EXIT_REASON_IO_INSTRUCTION] = handle_io, + [EXIT_REASON_INVLPG] = handle_invlpg, + [EXIT_REASON_CR_ACCESS] = handle_cr, + [EXIT_REASON_DR_ACCESS] = handle_dr, + [EXIT_REASON_CPUID] = handle_cpuid, + [EXIT_REASON_MSR_READ] = handle_rdmsr, + [EXIT_REASON_MSR_WRITE] = handle_wrmsr, + [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, + [EXIT_REASON_HLT] = handle_halt, +}; + +static const int kvm_vmx_max_exit_handlers = + sizeof(kvm_vmx_exit_handlers) / sizeof(*kvm_vmx_exit_handlers); + +/* + * The guest has exited. See if we can fix it or if we need userspace + * assistance. + */ +static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) +{ + u32 vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); + u32 exit_reason = vmcs_read32(VM_EXIT_REASON); + + if ( (vectoring_info & VECTORING_INFO_VALID_MASK) && + exit_reason != EXIT_REASON_EXCEPTION_NMI ) + printk(KERN_WARNING "%s: unexpected, valid vectoring info and " + "exit reason is 0x%x\n", __FUNCTION__, exit_reason); + kvm_run->instruction_length = vmcs_read32(VM_EXIT_INSTRUCTION_LEN); + if (exit_reason < kvm_vmx_max_exit_handlers + && kvm_vmx_exit_handlers[exit_reason]) + return kvm_vmx_exit_handlers[exit_reason](vcpu, kvm_run); + else { + kvm_run->exit_reason = KVM_EXIT_UNKNOWN; + kvm_run->hw.hardware_exit_reason = exit_reason; + } + return 0; +} + +static int vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u8 fail; + u16 fs_sel, gs_sel, ldt_sel; + int fs_gs_ldt_reload_needed; + +again: + /* + * Set host fs and gs selectors. Unfortunately, 22.2.3 does not + * allow segment selectors with cpl > 0 or ti == 1. + */ + fs_sel = read_fs(); + gs_sel = read_gs(); + ldt_sel = read_ldt(); + fs_gs_ldt_reload_needed = (fs_sel & 7) | (gs_sel & 7) | ldt_sel; + if (!fs_gs_ldt_reload_needed) { + vmcs_write16(HOST_FS_SELECTOR, fs_sel); + vmcs_write16(HOST_GS_SELECTOR, gs_sel); + } else { + vmcs_write16(HOST_FS_SELECTOR, 0); + vmcs_write16(HOST_GS_SELECTOR, 0); + } + +#ifdef __x86_64__ + vmcs_writel(HOST_FS_BASE, read_msr(MSR_FS_BASE)); + vmcs_writel(HOST_GS_BASE, read_msr(MSR_GS_BASE)); +#else + vmcs_writel(HOST_FS_BASE, segment_base(fs_sel)); + vmcs_writel(HOST_GS_BASE, segment_base(gs_sel)); +#endif + + if (vcpu->irq_summary && + !(vmcs_read32(VM_ENTRY_INTR_INFO_FIELD) & INTR_INFO_VALID_MASK)) + kvm_try_inject_irq(vcpu); + + if (vcpu->guest_debug.enabled) + kvm_guest_debug_pre(vcpu); + + fx_save(vcpu->host_fx_image); + fx_restore(vcpu->guest_fx_image); + + save_msrs(vcpu->host_msrs, vcpu->nmsrs); + load_msrs(vcpu->guest_msrs, NR_BAD_MSRS); + + asm ( + /* Store host registers */ + "pushf \n\t" +#ifdef __x86_64__ + "push %%rax; push %%rbx; push %%rdx;" + "push %%rsi; push %%rdi; push %%rbp;" + "push %%r8; push %%r9; push %%r10; push %%r11;" + "push %%r12; push %%r13; push %%r14; push %%r15;" + "push %%rcx \n\t" + ASM_VMX_VMWRITE_RSP_RDX "\n\t" +#else + "pusha; push %%ecx \n\t" + ASM_VMX_VMWRITE_RSP_RDX "\n\t" +#endif + /* Check if vmlaunch of vmresume is needed */ + "cmp $0, %1 \n\t" + /* Load guest registers. Don't clobber flags. */ +#ifdef __x86_64__ + "mov %c[cr2](%3), %%rax \n\t" + "mov %%rax, %%cr2 \n\t" + "mov %c[rax](%3), %%rax \n\t" + "mov %c[rbx](%3), %%rbx \n\t" + "mov %c[rdx](%3), %%rdx \n\t" + "mov %c[rsi](%3), %%rsi \n\t" + "mov %c[rdi](%3), %%rdi \n\t" + "mov %c[rbp](%3), %%rbp \n\t" + "mov %c[r8](%3), %%r8 \n\t" + "mov %c[r9](%3), %%r9 \n\t" + "mov %c[r10](%3), %%r10 \n\t" + "mov %c[r11](%3), %%r11 \n\t" + "mov %c[r12](%3), %%r12 \n\t" + "mov %c[r13](%3), %%r13 \n\t" + "mov %c[r14](%3), %%r14 \n\t" + "mov %c[r15](%3), %%r15 \n\t" + "mov %c[rcx](%3), %%rcx \n\t" /* kills %3 (rcx) */ +#else + "mov %c[cr2](%3), %%eax \n\t" + "mov %%eax, %%cr2 \n\t" + "mov %c[rax](%3), %%eax \n\t" + "mov %c[rbx](%3), %%ebx \n\t" + "mov %c[rdx](%3), %%edx \n\t" + "mov %c[rsi](%3), %%esi \n\t" + "mov %c[rdi](%3), %%edi \n\t" + "mov %c[rbp](%3), %%ebp \n\t" + "mov %c[rcx](%3), %%ecx \n\t" /* kills %3 (ecx) */ +#endif + /* Enter guest mode */ + "jne launched \n\t" + ASM_VMX_VMLAUNCH "\n\t" + "jmp kvm_vmx_return \n\t" + "launched: " ASM_VMX_VMRESUME "\n\t" + ".globl kvm_vmx_return \n\t" + "kvm_vmx_return: " + /* Save guest registers, load host registers, keep flags */ +#ifdef __x86_64__ + "xchg %3, 0(%%rsp) \n\t" + "mov %%rax, %c[rax](%3) \n\t" + "mov %%rbx, %c[rbx](%3) \n\t" + "pushq 0(%%rsp); popq %c[rcx](%3) \n\t" + "mov %%rdx, %c[rdx](%3) \n\t" + "mov %%rsi, %c[rsi](%3) \n\t" + "mov %%rdi, %c[rdi](%3) \n\t" + "mov %%rbp, %c[rbp](%3) \n\t" + "mov %%r8, %c[r8](%3) \n\t" + "mov %%r9, %c[r9](%3) \n\t" + "mov %%r10, %c[r10](%3) \n\t" + "mov %%r11, %c[r11](%3) \n\t" + "mov %%r12, %c[r12](%3) \n\t" + "mov %%r13, %c[r13](%3) \n\t" + "mov %%r14, %c[r14](%3) \n\t" + "mov %%r15, %c[r15](%3) \n\t" + "mov %%cr2, %%rax \n\t" + "mov %%rax, %c[cr2](%3) \n\t" + "mov 0(%%rsp), %3 \n\t" + + "pop %%rcx; pop %%r15; pop %%r14; pop %%r13; pop %%r12;" + "pop %%r11; pop %%r10; pop %%r9; pop %%r8;" + "pop %%rbp; pop %%rdi; pop %%rsi;" + "pop %%rdx; pop %%rbx; pop %%rax \n\t" +#else + "xchg %3, 0(%%esp) \n\t" + "mov %%eax, %c[rax](%3) \n\t" + "mov %%ebx, %c[rbx](%3) \n\t" + "pushl 0(%%esp); popl %c[rcx](%3) \n\t" + "mov %%edx, %c[rdx](%3) \n\t" + "mov %%esi, %c[rsi](%3) \n\t" + "mov %%edi, %c[rdi](%3) \n\t" + "mov %%ebp, %c[rbp](%3) \n\t" + "mov %%cr2, %%eax \n\t" + "mov %%eax, %c[cr2](%3) \n\t" + "mov 0(%%esp), %3 \n\t" + + "pop %%ecx; popa \n\t" +#endif + "setbe %0 \n\t" + "popf \n\t" + : "=g" (fail) + : "r"(vcpu->launched), "d"((unsigned long)HOST_RSP), + "c"(vcpu), + [rax]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RAX])), + [rbx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RBX])), + [rcx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RCX])), + [rdx]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RDX])), + [rsi]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RSI])), + [rdi]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RDI])), + [rbp]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_RBP])), +#ifdef __x86_64__ + [r8 ]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R8 ])), + [r9 ]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R9 ])), + [r10]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R10])), + [r11]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R11])), + [r12]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R12])), + [r13]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R13])), + [r14]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R14])), + [r15]"i"(offsetof(struct kvm_vcpu, regs[VCPU_REGS_R15])), +#endif + [cr2]"i"(offsetof(struct kvm_vcpu, cr2)) + : "cc", "memory" ); + + ++kvm_stat.exits; + + save_msrs(vcpu->guest_msrs, NR_BAD_MSRS); + load_msrs(vcpu->host_msrs, NR_BAD_MSRS); + + fx_save(vcpu->guest_fx_image); + fx_restore(vcpu->host_fx_image); + +#ifndef __x86_64__ + asm ("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); +#endif + + kvm_run->exit_type = 0; + if (fail) { + kvm_run->exit_type = KVM_EXIT_TYPE_FAIL_ENTRY; + kvm_run->exit_reason = vmcs_read32(VM_INSTRUCTION_ERROR); + } else { + if (fs_gs_ldt_reload_needed) { + load_ldt(ldt_sel); + load_fs(fs_sel); + /* + * If we have to reload gs, we must take care to + * preserve our gs base. + */ + local_irq_disable(); + load_gs(gs_sel); +#ifdef __x86_64__ + wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE)); +#endif + local_irq_enable(); + + reload_tss(); + } + vcpu->launched = 1; + kvm_run->exit_type = KVM_EXIT_TYPE_VM_EXIT; + if (kvm_handle_exit(kvm_run, vcpu)) { + /* Give scheduler a change to reschedule. */ + if (signal_pending(current)) { + ++kvm_stat.signal_exits; + return -EINTR; + } + kvm_resched(vcpu); + goto again; + } + } + return 0; +} + +static void vmx_flush_tlb(struct kvm_vcpu *vcpu) +{ + vmcs_writel(GUEST_CR3, vmcs_readl(GUEST_CR3)); +} + +static void vmx_inject_page_fault(struct kvm_vcpu *vcpu, + unsigned long addr, + u32 err_code) +{ + u32 vect_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); + + ++kvm_stat.pf_guest; + + if (is_page_fault(vect_info)) { + printk(KERN_DEBUG "inject_page_fault: " + "double fault 0x%lx @ 0x%lx\n", + addr, vmcs_readl(GUEST_RIP)); + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, 0); + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + DF_VECTOR | + INTR_TYPE_EXCEPTION | + INTR_INFO_DELIEVER_CODE_MASK | + INTR_INFO_VALID_MASK); + return; + } + vcpu->cr2 = addr; + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, err_code); + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + PF_VECTOR | + INTR_TYPE_EXCEPTION | + INTR_INFO_DELIEVER_CODE_MASK | + INTR_INFO_VALID_MASK); + +} + +static void vmx_free_vmcs(struct kvm_vcpu *vcpu) +{ + if (vcpu->vmcs) { + on_each_cpu(__vcpu_clear, vcpu, 0, 1); + free_vmcs(vcpu->vmcs); + vcpu->vmcs = NULL; + } +} + +static void vmx_free_vcpu(struct kvm_vcpu *vcpu) +{ + vmx_free_vmcs(vcpu); +} + +static int vmx_create_vcpu(struct kvm_vcpu *vcpu) +{ + struct vmcs *vmcs; + + vmcs = alloc_vmcs(); + if (!vmcs) + return -ENOMEM; + vmcs_clear(vmcs); + vcpu->vmcs = vmcs; + vcpu->launched = 0; + return 0; +} + +static struct kvm_arch_ops vmx_arch_ops = { + .cpu_has_kvm_support = cpu_has_kvm_support, + .disabled_by_bios = vmx_disabled_by_bios, + .hardware_setup = hardware_setup, + .hardware_unsetup = hardware_unsetup, + .hardware_enable = hardware_enable, + .hardware_disable = hardware_disable, + + .vcpu_create = vmx_create_vcpu, + .vcpu_free = vmx_free_vcpu, + + .vcpu_load = vmx_vcpu_load, + .vcpu_put = vmx_vcpu_put, + + .set_guest_debug = set_guest_debug, + .get_msr = vmx_get_msr, + .set_msr = vmx_set_msr, + .get_segment_base = vmx_get_segment_base, + .get_segment = vmx_get_segment, + .set_segment = vmx_set_segment, + .is_long_mode = vmx_is_long_mode, + .get_cs_db_l_bits = vmx_get_cs_db_l_bits, + .set_cr0 = vmx_set_cr0, + .set_cr0_no_modeswitch = vmx_set_cr0_no_modeswitch, + .set_cr3 = vmx_set_cr3, + .set_cr4 = vmx_set_cr4, +#ifdef __x86_64__ + .set_efer = vmx_set_efer, +#endif + .get_idt = vmx_get_idt, + .set_idt = vmx_set_idt, + .get_gdt = vmx_get_gdt, + .set_gdt = vmx_set_gdt, + .cache_regs = vcpu_load_rsp_rip, + .decache_regs = vcpu_put_rsp_rip, + .get_rflags = vmx_get_rflags, + .set_rflags = vmx_set_rflags, + + .tlb_flush = vmx_flush_tlb, + .inject_page_fault = vmx_inject_page_fault, + + .inject_gp = vmx_inject_gp, + + .run = vmx_vcpu_run, + .skip_emulated_instruction = skip_emulated_instruction, + .vcpu_setup = vmx_vcpu_setup, +}; + +static int __init vmx_init(void) +{ + kvm_init_arch(&vmx_arch_ops, THIS_MODULE); + return 0; +} + +static void __exit vmx_exit(void) +{ + kvm_exit_arch(); +} + +module_init(vmx_init) +module_exit(vmx_exit) diff --git a/drivers/kvm/vmx.h b/drivers/kvm/vmx.h new file mode 100644 index 000000000000..797278341581 --- /dev/null +++ b/drivers/kvm/vmx.h @@ -0,0 +1,296 @@ +#ifndef VMX_H +#define VMX_H + +/* + * vmx.h: VMX Architecture related definitions + * Copyright (c) 2004, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + * A few random additions are: + * Copyright (C) 2006 Qumranet + * Avi Kivity + * Yaniv Kamay + * + */ + +#define CPU_BASED_VIRTUAL_INTR_PENDING 0x00000004 +#define CPU_BASED_USE_TSC_OFFSETING 0x00000008 +#define CPU_BASED_HLT_EXITING 0x00000080 +#define CPU_BASED_INVDPG_EXITING 0x00000200 +#define CPU_BASED_MWAIT_EXITING 0x00000400 +#define CPU_BASED_RDPMC_EXITING 0x00000800 +#define CPU_BASED_RDTSC_EXITING 0x00001000 +#define CPU_BASED_CR8_LOAD_EXITING 0x00080000 +#define CPU_BASED_CR8_STORE_EXITING 0x00100000 +#define CPU_BASED_TPR_SHADOW 0x00200000 +#define CPU_BASED_MOV_DR_EXITING 0x00800000 +#define CPU_BASED_UNCOND_IO_EXITING 0x01000000 +#define CPU_BASED_ACTIVATE_IO_BITMAP 0x02000000 +#define CPU_BASED_MSR_BITMAPS 0x10000000 +#define CPU_BASED_MONITOR_EXITING 0x20000000 +#define CPU_BASED_PAUSE_EXITING 0x40000000 + +#define PIN_BASED_EXT_INTR_MASK 0x1 +#define PIN_BASED_NMI_EXITING 0x8 + +#define VM_EXIT_ACK_INTR_ON_EXIT 0x00008000 +#define VM_EXIT_HOST_ADD_SPACE_SIZE 0x00000200 + + +/* VMCS Encodings */ +enum vmcs_field { + GUEST_ES_SELECTOR = 0x00000800, + GUEST_CS_SELECTOR = 0x00000802, + GUEST_SS_SELECTOR = 0x00000804, + GUEST_DS_SELECTOR = 0x00000806, + GUEST_FS_SELECTOR = 0x00000808, + GUEST_GS_SELECTOR = 0x0000080a, + GUEST_LDTR_SELECTOR = 0x0000080c, + GUEST_TR_SELECTOR = 0x0000080e, + HOST_ES_SELECTOR = 0x00000c00, + HOST_CS_SELECTOR = 0x00000c02, + HOST_SS_SELECTOR = 0x00000c04, + HOST_DS_SELECTOR = 0x00000c06, + HOST_FS_SELECTOR = 0x00000c08, + HOST_GS_SELECTOR = 0x00000c0a, + HOST_TR_SELECTOR = 0x00000c0c, + IO_BITMAP_A = 0x00002000, + IO_BITMAP_A_HIGH = 0x00002001, + IO_BITMAP_B = 0x00002002, + IO_BITMAP_B_HIGH = 0x00002003, + MSR_BITMAP = 0x00002004, + MSR_BITMAP_HIGH = 0x00002005, + VM_EXIT_MSR_STORE_ADDR = 0x00002006, + VM_EXIT_MSR_STORE_ADDR_HIGH = 0x00002007, + VM_EXIT_MSR_LOAD_ADDR = 0x00002008, + VM_EXIT_MSR_LOAD_ADDR_HIGH = 0x00002009, + VM_ENTRY_MSR_LOAD_ADDR = 0x0000200a, + VM_ENTRY_MSR_LOAD_ADDR_HIGH = 0x0000200b, + TSC_OFFSET = 0x00002010, + TSC_OFFSET_HIGH = 0x00002011, + VIRTUAL_APIC_PAGE_ADDR = 0x00002012, + VIRTUAL_APIC_PAGE_ADDR_HIGH = 0x00002013, + VMCS_LINK_POINTER = 0x00002800, + VMCS_LINK_POINTER_HIGH = 0x00002801, + GUEST_IA32_DEBUGCTL = 0x00002802, + GUEST_IA32_DEBUGCTL_HIGH = 0x00002803, + PIN_BASED_VM_EXEC_CONTROL = 0x00004000, + CPU_BASED_VM_EXEC_CONTROL = 0x00004002, + EXCEPTION_BITMAP = 0x00004004, + PAGE_FAULT_ERROR_CODE_MASK = 0x00004006, + PAGE_FAULT_ERROR_CODE_MATCH = 0x00004008, + CR3_TARGET_COUNT = 0x0000400a, + VM_EXIT_CONTROLS = 0x0000400c, + VM_EXIT_MSR_STORE_COUNT = 0x0000400e, + VM_EXIT_MSR_LOAD_COUNT = 0x00004010, + VM_ENTRY_CONTROLS = 0x00004012, + VM_ENTRY_MSR_LOAD_COUNT = 0x00004014, + VM_ENTRY_INTR_INFO_FIELD = 0x00004016, + VM_ENTRY_EXCEPTION_ERROR_CODE = 0x00004018, + VM_ENTRY_INSTRUCTION_LEN = 0x0000401a, + TPR_THRESHOLD = 0x0000401c, + SECONDARY_VM_EXEC_CONTROL = 0x0000401e, + VM_INSTRUCTION_ERROR = 0x00004400, + VM_EXIT_REASON = 0x00004402, + VM_EXIT_INTR_INFO = 0x00004404, + VM_EXIT_INTR_ERROR_CODE = 0x00004406, + IDT_VECTORING_INFO_FIELD = 0x00004408, + IDT_VECTORING_ERROR_CODE = 0x0000440a, + VM_EXIT_INSTRUCTION_LEN = 0x0000440c, + VMX_INSTRUCTION_INFO = 0x0000440e, + GUEST_ES_LIMIT = 0x00004800, + GUEST_CS_LIMIT = 0x00004802, + GUEST_SS_LIMIT = 0x00004804, + GUEST_DS_LIMIT = 0x00004806, + GUEST_FS_LIMIT = 0x00004808, + GUEST_GS_LIMIT = 0x0000480a, + GUEST_LDTR_LIMIT = 0x0000480c, + GUEST_TR_LIMIT = 0x0000480e, + GUEST_GDTR_LIMIT = 0x00004810, + GUEST_IDTR_LIMIT = 0x00004812, + GUEST_ES_AR_BYTES = 0x00004814, + GUEST_CS_AR_BYTES = 0x00004816, + GUEST_SS_AR_BYTES = 0x00004818, + GUEST_DS_AR_BYTES = 0x0000481a, + GUEST_FS_AR_BYTES = 0x0000481c, + GUEST_GS_AR_BYTES = 0x0000481e, + GUEST_LDTR_AR_BYTES = 0x00004820, + GUEST_TR_AR_BYTES = 0x00004822, + GUEST_INTERRUPTIBILITY_INFO = 0x00004824, + GUEST_ACTIVITY_STATE = 0X00004826, + GUEST_SYSENTER_CS = 0x0000482A, + HOST_IA32_SYSENTER_CS = 0x00004c00, + CR0_GUEST_HOST_MASK = 0x00006000, + CR4_GUEST_HOST_MASK = 0x00006002, + CR0_READ_SHADOW = 0x00006004, + CR4_READ_SHADOW = 0x00006006, + CR3_TARGET_VALUE0 = 0x00006008, + CR3_TARGET_VALUE1 = 0x0000600a, + CR3_TARGET_VALUE2 = 0x0000600c, + CR3_TARGET_VALUE3 = 0x0000600e, + EXIT_QUALIFICATION = 0x00006400, + GUEST_LINEAR_ADDRESS = 0x0000640a, + GUEST_CR0 = 0x00006800, + GUEST_CR3 = 0x00006802, + GUEST_CR4 = 0x00006804, + GUEST_ES_BASE = 0x00006806, + GUEST_CS_BASE = 0x00006808, + GUEST_SS_BASE = 0x0000680a, + GUEST_DS_BASE = 0x0000680c, + GUEST_FS_BASE = 0x0000680e, + GUEST_GS_BASE = 0x00006810, + GUEST_LDTR_BASE = 0x00006812, + GUEST_TR_BASE = 0x00006814, + GUEST_GDTR_BASE = 0x00006816, + GUEST_IDTR_BASE = 0x00006818, + GUEST_DR7 = 0x0000681a, + GUEST_RSP = 0x0000681c, + GUEST_RIP = 0x0000681e, + GUEST_RFLAGS = 0x00006820, + GUEST_PENDING_DBG_EXCEPTIONS = 0x00006822, + GUEST_SYSENTER_ESP = 0x00006824, + GUEST_SYSENTER_EIP = 0x00006826, + HOST_CR0 = 0x00006c00, + HOST_CR3 = 0x00006c02, + HOST_CR4 = 0x00006c04, + HOST_FS_BASE = 0x00006c06, + HOST_GS_BASE = 0x00006c08, + HOST_TR_BASE = 0x00006c0a, + HOST_GDTR_BASE = 0x00006c0c, + HOST_IDTR_BASE = 0x00006c0e, + HOST_IA32_SYSENTER_ESP = 0x00006c10, + HOST_IA32_SYSENTER_EIP = 0x00006c12, + HOST_RSP = 0x00006c14, + HOST_RIP = 0x00006c16, +}; + +#define VMX_EXIT_REASONS_FAILED_VMENTRY 0x80000000 + +#define EXIT_REASON_EXCEPTION_NMI 0 +#define EXIT_REASON_EXTERNAL_INTERRUPT 1 + +#define EXIT_REASON_PENDING_INTERRUPT 7 + +#define EXIT_REASON_TASK_SWITCH 9 +#define EXIT_REASON_CPUID 10 +#define EXIT_REASON_HLT 12 +#define EXIT_REASON_INVLPG 14 +#define EXIT_REASON_RDPMC 15 +#define EXIT_REASON_RDTSC 16 +#define EXIT_REASON_VMCALL 18 +#define EXIT_REASON_VMCLEAR 19 +#define EXIT_REASON_VMLAUNCH 20 +#define EXIT_REASON_VMPTRLD 21 +#define EXIT_REASON_VMPTRST 22 +#define EXIT_REASON_VMREAD 23 +#define EXIT_REASON_VMRESUME 24 +#define EXIT_REASON_VMWRITE 25 +#define EXIT_REASON_VMOFF 26 +#define EXIT_REASON_VMON 27 +#define EXIT_REASON_CR_ACCESS 28 +#define EXIT_REASON_DR_ACCESS 29 +#define EXIT_REASON_IO_INSTRUCTION 30 +#define EXIT_REASON_MSR_READ 31 +#define EXIT_REASON_MSR_WRITE 32 +#define EXIT_REASON_MWAIT_INSTRUCTION 36 + +/* + * Interruption-information format + */ +#define INTR_INFO_VECTOR_MASK 0xff /* 7:0 */ +#define INTR_INFO_INTR_TYPE_MASK 0x700 /* 10:8 */ +#define INTR_INFO_DELIEVER_CODE_MASK 0x800 /* 11 */ +#define INTR_INFO_VALID_MASK 0x80000000 /* 31 */ + +#define VECTORING_INFO_VECTOR_MASK INTR_INFO_VECTOR_MASK +#define VECTORING_INFO_TYPE_MASK INTR_INFO_INTR_TYPE_MASK +#define VECTORING_INFO_DELIEVER_CODE_MASK INTR_INFO_DELIEVER_CODE_MASK +#define VECTORING_INFO_VALID_MASK INTR_INFO_VALID_MASK + +#define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */ +#define INTR_TYPE_EXCEPTION (3 << 8) /* processor exception */ + +/* + * Exit Qualifications for MOV for Control Register Access + */ +#define CONTROL_REG_ACCESS_NUM 0x7 /* 2:0, number of control register */ +#define CONTROL_REG_ACCESS_TYPE 0x30 /* 5:4, access type */ +#define CONTROL_REG_ACCESS_REG 0xf00 /* 10:8, general purpose register */ +#define LMSW_SOURCE_DATA_SHIFT 16 +#define LMSW_SOURCE_DATA (0xFFFF << LMSW_SOURCE_DATA_SHIFT) /* 16:31 lmsw source */ +#define REG_EAX (0 << 8) +#define REG_ECX (1 << 8) +#define REG_EDX (2 << 8) +#define REG_EBX (3 << 8) +#define REG_ESP (4 << 8) +#define REG_EBP (5 << 8) +#define REG_ESI (6 << 8) +#define REG_EDI (7 << 8) +#define REG_R8 (8 << 8) +#define REG_R9 (9 << 8) +#define REG_R10 (10 << 8) +#define REG_R11 (11 << 8) +#define REG_R12 (12 << 8) +#define REG_R13 (13 << 8) +#define REG_R14 (14 << 8) +#define REG_R15 (15 << 8) + +/* + * Exit Qualifications for MOV for Debug Register Access + */ +#define DEBUG_REG_ACCESS_NUM 0x7 /* 2:0, number of debug register */ +#define DEBUG_REG_ACCESS_TYPE 0x10 /* 4, direction of access */ +#define TYPE_MOV_TO_DR (0 << 4) +#define TYPE_MOV_FROM_DR (1 << 4) +#define DEBUG_REG_ACCESS_REG 0xf00 /* 11:8, general purpose register */ + + +/* segment AR */ +#define SEGMENT_AR_L_MASK (1 << 13) + +/* entry controls */ +#define VM_ENTRY_CONTROLS_IA32E_MASK (1 << 9) + +#define AR_TYPE_ACCESSES_MASK 1 +#define AR_TYPE_READABLE_MASK (1 << 1) +#define AR_TYPE_WRITEABLE_MASK (1 << 2) +#define AR_TYPE_CODE_MASK (1 << 3) +#define AR_TYPE_MASK 0x0f +#define AR_TYPE_BUSY_64_TSS 11 +#define AR_TYPE_BUSY_32_TSS 11 +#define AR_TYPE_BUSY_16_TSS 3 +#define AR_TYPE_LDT 2 + +#define AR_UNUSABLE_MASK (1 << 16) +#define AR_S_MASK (1 << 4) +#define AR_P_MASK (1 << 7) +#define AR_L_MASK (1 << 13) +#define AR_DB_MASK (1 << 14) +#define AR_G_MASK (1 << 15) +#define AR_DPL_SHIFT 5 +#define AR_DPL(ar) (((ar) >> AR_DPL_SHIFT) & 3) + +#define AR_RESERVD_MASK 0xfffe0f00 + +#define CR4_VMXE 0x2000 + +#define MSR_IA32_VMX_BASIC_MSR 0x480 +#define MSR_IA32_FEATURE_CONTROL 0x03a +#define MSR_IA32_VMX_PINBASED_CTLS_MSR 0x481 +#define MSR_IA32_VMX_PROCBASED_CTLS_MSR 0x482 +#define MSR_IA32_VMX_EXIT_CTLS_MSR 0x483 +#define MSR_IA32_VMX_ENTRY_CTLS_MSR 0x484 + +#endif diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c new file mode 100644 index 000000000000..7e838bf0592d --- /dev/null +++ b/drivers/kvm/x86_emulate.c @@ -0,0 +1,1409 @@ +/****************************************************************************** + * x86_emulate.c + * + * Generic x86 (32-bit and 64-bit) instruction decoder and emulator. + * + * Copyright (c) 2005 Keir Fraser + * + * Linux coding style, mod r/m decoder, segment base fixes, real-mode + * privieged instructions: + * + * Copyright (C) 2006 Qumranet + * + * Avi Kivity + * Yaniv Kamay + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4 + */ + +#ifndef __KERNEL__ +#include +#include +#include +#define DPRINTF(_f, _a ...) printf( _f , ## _a ) +#else +#include "kvm.h" +#define DPRINTF(x...) do {} while (0) +#endif +#include "x86_emulate.h" +#include + +/* + * Opcode effective-address decode tables. + * Note that we only emulate instructions that have at least one memory + * operand (excluding implicit stack references). We assume that stack + * references and instruction fetches will never occur in special memory + * areas that require emulation. So, for example, 'mov ,' need + * not be handled. + */ + +/* Operand sizes: 8-bit operands or specified/overridden size. */ +#define ByteOp (1<<0) /* 8-bit operands. */ +/* Destination operand type. */ +#define ImplicitOps (1<<1) /* Implicit in opcode. No generic decode. */ +#define DstReg (2<<1) /* Register operand. */ +#define DstMem (3<<1) /* Memory operand. */ +#define DstMask (3<<1) +/* Source operand type. */ +#define SrcNone (0<<3) /* No source operand. */ +#define SrcImplicit (0<<3) /* Source operand is implicit in the opcode. */ +#define SrcReg (1<<3) /* Register operand. */ +#define SrcMem (2<<3) /* Memory operand. */ +#define SrcMem16 (3<<3) /* Memory operand (16-bit). */ +#define SrcMem32 (4<<3) /* Memory operand (32-bit). */ +#define SrcImm (5<<3) /* Immediate operand. */ +#define SrcImmByte (6<<3) /* 8-bit sign-extended immediate operand. */ +#define SrcMask (7<<3) +/* Generic ModRM decode. */ +#define ModRM (1<<6) +/* Destination is only written; never read. */ +#define Mov (1<<7) + +static u8 opcode_table[256] = { + /* 0x00 - 0x07 */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x08 - 0x0F */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x10 - 0x17 */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x18 - 0x1F */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x20 - 0x27 */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x28 - 0x2F */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x30 - 0x37 */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x38 - 0x3F */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstReg | SrcMem | ModRM, DstReg | SrcMem | ModRM, + 0, 0, 0, 0, + /* 0x40 - 0x4F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x50 - 0x5F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x60 - 0x6F */ + 0, 0, 0, DstReg | SrcMem32 | ModRM | Mov /* movsxd (x86/64) */ , + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x70 - 0x7F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x80 - 0x87 */ + ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImm | ModRM, + ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM, + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, + /* 0x88 - 0x8F */ + ByteOp | DstMem | SrcReg | ModRM | Mov, DstMem | SrcReg | ModRM | Mov, + ByteOp | DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + 0, 0, 0, DstMem | SrcNone | ModRM | Mov, + /* 0x90 - 0x9F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xA0 - 0xA7 */ + ByteOp | DstReg | SrcMem | Mov, DstReg | SrcMem | Mov, + ByteOp | DstMem | SrcReg | Mov, DstMem | SrcReg | Mov, + ByteOp | ImplicitOps | Mov, ImplicitOps | Mov, + ByteOp | ImplicitOps, ImplicitOps, + /* 0xA8 - 0xAF */ + 0, 0, ByteOp | ImplicitOps | Mov, ImplicitOps | Mov, + ByteOp | ImplicitOps | Mov, ImplicitOps | Mov, + ByteOp | ImplicitOps, ImplicitOps, + /* 0xB0 - 0xBF */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xC0 - 0xC7 */ + ByteOp | DstMem | SrcImm | ModRM, DstMem | SrcImmByte | ModRM, 0, 0, + 0, 0, ByteOp | DstMem | SrcImm | ModRM | Mov, + DstMem | SrcImm | ModRM | Mov, + /* 0xC8 - 0xCF */ + 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xD0 - 0xD7 */ + ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM, + ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM, + 0, 0, 0, 0, + /* 0xD8 - 0xDF */ + 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xE0 - 0xEF */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xF0 - 0xF7 */ + 0, 0, 0, 0, + 0, 0, ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM, + /* 0xF8 - 0xFF */ + 0, 0, 0, 0, + 0, 0, ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM +}; + +static u8 twobyte_table[256] = { + /* 0x00 - 0x0F */ + 0, SrcMem | ModRM | DstReg, 0, 0, 0, 0, ImplicitOps, 0, + 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, + /* 0x10 - 0x1F */ + 0, 0, 0, 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, 0, 0, 0, 0, 0, + /* 0x20 - 0x2F */ + ModRM | ImplicitOps, ModRM, ModRM | ImplicitOps, ModRM, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x30 - 0x3F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x40 - 0x47 */ + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + /* 0x48 - 0x4F */ + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem | ModRM | Mov, + /* 0x50 - 0x5F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x60 - 0x6F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x70 - 0x7F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x80 - 0x8F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0x90 - 0x9F */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xA0 - 0xA7 */ + 0, 0, 0, DstMem | SrcReg | ModRM, 0, 0, 0, 0, + /* 0xA8 - 0xAF */ + 0, 0, 0, DstMem | SrcReg | ModRM, 0, 0, 0, 0, + /* 0xB0 - 0xB7 */ + ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, 0, + DstMem | SrcReg | ModRM, + 0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem16 | ModRM | Mov, + /* 0xB8 - 0xBF */ + 0, 0, DstMem | SrcImmByte | ModRM, DstMem | SrcReg | ModRM, + 0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov, + DstReg | SrcMem16 | ModRM | Mov, + /* 0xC0 - 0xCF */ + 0, 0, 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xD0 - 0xDF */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xE0 - 0xEF */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + /* 0xF0 - 0xFF */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 +}; + +/* + * Tell the emulator that of the Group 7 instructions (sgdt, lidt, etc.) we + * are interested only in invlpg and not in any of the rest. + * + * invlpg is a special instruction in that the data it references may not + * be mapped. + */ +void kvm_emulator_want_group7_invlpg(void) +{ + twobyte_table[1] &= ~SrcMem; +} +EXPORT_SYMBOL_GPL(kvm_emulator_want_group7_invlpg); + +/* Type, address-of, and value of an instruction's operand. */ +struct operand { + enum { OP_REG, OP_MEM, OP_IMM } type; + unsigned int bytes; + unsigned long val, orig_val, *ptr; +}; + +/* EFLAGS bit definitions. */ +#define EFLG_OF (1<<11) +#define EFLG_DF (1<<10) +#define EFLG_SF (1<<7) +#define EFLG_ZF (1<<6) +#define EFLG_AF (1<<4) +#define EFLG_PF (1<<2) +#define EFLG_CF (1<<0) + +/* + * Instruction emulation: + * Most instructions are emulated directly via a fragment of inline assembly + * code. This allows us to save/restore EFLAGS and thus very easily pick up + * any modified flags. + */ + +#if defined(__x86_64__) +#define _LO32 "k" /* force 32-bit operand */ +#define _STK "%%rsp" /* stack pointer */ +#elif defined(__i386__) +#define _LO32 "" /* force 32-bit operand */ +#define _STK "%%esp" /* stack pointer */ +#endif + +/* + * These EFLAGS bits are restored from saved value during emulation, and + * any changes are written back to the saved value after emulation. + */ +#define EFLAGS_MASK (EFLG_OF|EFLG_SF|EFLG_ZF|EFLG_AF|EFLG_PF|EFLG_CF) + +/* Before executing instruction: restore necessary bits in EFLAGS. */ +#define _PRE_EFLAGS(_sav, _msk, _tmp) \ + /* EFLAGS = (_sav & _msk) | (EFLAGS & ~_msk); */ \ + "push %"_sav"; " \ + "movl %"_msk",%"_LO32 _tmp"; " \ + "andl %"_LO32 _tmp",("_STK"); " \ + "pushf; " \ + "notl %"_LO32 _tmp"; " \ + "andl %"_LO32 _tmp",("_STK"); " \ + "pop %"_tmp"; " \ + "orl %"_LO32 _tmp",("_STK"); " \ + "popf; " \ + /* _sav &= ~msk; */ \ + "movl %"_msk",%"_LO32 _tmp"; " \ + "notl %"_LO32 _tmp"; " \ + "andl %"_LO32 _tmp",%"_sav"; " + +/* After executing instruction: write-back necessary bits in EFLAGS. */ +#define _POST_EFLAGS(_sav, _msk, _tmp) \ + /* _sav |= EFLAGS & _msk; */ \ + "pushf; " \ + "pop %"_tmp"; " \ + "andl %"_msk",%"_LO32 _tmp"; " \ + "orl %"_LO32 _tmp",%"_sav"; " + +/* Raw emulation: instruction has two explicit operands. */ +#define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \ + do { \ + unsigned long _tmp; \ + \ + switch ((_dst).bytes) { \ + case 2: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","4","2") \ + _op"w %"_wx"3,%1; " \ + _POST_EFLAGS("0","4","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : _wy ((_src).val), "i" (EFLAGS_MASK) ); \ + break; \ + case 4: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","4","2") \ + _op"l %"_lx"3,%1; " \ + _POST_EFLAGS("0","4","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : _ly ((_src).val), "i" (EFLAGS_MASK) ); \ + break; \ + case 8: \ + __emulate_2op_8byte(_op, _src, _dst, \ + _eflags, _qx, _qy); \ + break; \ + } \ + } while (0) + +#define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \ + do { \ + unsigned long _tmp; \ + switch ( (_dst).bytes ) \ + { \ + case 1: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","4","2") \ + _op"b %"_bx"3,%1; " \ + _POST_EFLAGS("0","4","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : _by ((_src).val), "i" (EFLAGS_MASK) ); \ + break; \ + default: \ + __emulate_2op_nobyte(_op, _src, _dst, _eflags, \ + _wx, _wy, _lx, _ly, _qx, _qy); \ + break; \ + } \ + } while (0) + +/* Source operand is byte-sized and may be restricted to just %cl. */ +#define emulate_2op_SrcB(_op, _src, _dst, _eflags) \ + __emulate_2op(_op, _src, _dst, _eflags, \ + "b", "c", "b", "c", "b", "c", "b", "c") + +/* Source operand is byte, word, long or quad sized. */ +#define emulate_2op_SrcV(_op, _src, _dst, _eflags) \ + __emulate_2op(_op, _src, _dst, _eflags, \ + "b", "q", "w", "r", _LO32, "r", "", "r") + +/* Source operand is word, long or quad sized. */ +#define emulate_2op_SrcV_nobyte(_op, _src, _dst, _eflags) \ + __emulate_2op_nobyte(_op, _src, _dst, _eflags, \ + "w", "r", _LO32, "r", "", "r") + +/* Instruction has only one explicit operand (no source operand). */ +#define emulate_1op(_op, _dst, _eflags) \ + do { \ + unsigned long _tmp; \ + \ + switch ( (_dst).bytes ) \ + { \ + case 1: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","3","2") \ + _op"b %1; " \ + _POST_EFLAGS("0","3","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : "i" (EFLAGS_MASK) ); \ + break; \ + case 2: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","3","2") \ + _op"w %1; " \ + _POST_EFLAGS("0","3","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : "i" (EFLAGS_MASK) ); \ + break; \ + case 4: \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","3","2") \ + _op"l %1; " \ + _POST_EFLAGS("0","3","2") \ + : "=m" (_eflags), "=m" ((_dst).val), \ + "=&r" (_tmp) \ + : "i" (EFLAGS_MASK) ); \ + break; \ + case 8: \ + __emulate_1op_8byte(_op, _dst, _eflags); \ + break; \ + } \ + } while (0) + +/* Emulate an instruction with quadword operands (x86/64 only). */ +#if defined(__x86_64__) +#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy) \ + do { \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","4","2") \ + _op"q %"_qx"3,%1; " \ + _POST_EFLAGS("0","4","2") \ + : "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \ + : _qy ((_src).val), "i" (EFLAGS_MASK) ); \ + } while (0) + +#define __emulate_1op_8byte(_op, _dst, _eflags) \ + do { \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0","3","2") \ + _op"q %1; " \ + _POST_EFLAGS("0","3","2") \ + : "=m" (_eflags), "=m" ((_dst).val), "=&r" (_tmp) \ + : "i" (EFLAGS_MASK) ); \ + } while (0) + +#elif defined(__i386__) +#define __emulate_2op_8byte(_op, _src, _dst, _eflags, _qx, _qy) +#define __emulate_1op_8byte(_op, _dst, _eflags) +#endif /* __i386__ */ + +/* Fetch next part of the instruction being emulated. */ +#define insn_fetch(_type, _size, _eip) \ +({ unsigned long _x; \ + rc = ops->read_std((unsigned long)(_eip) + ctxt->cs_base, &_x, \ + (_size), ctxt); \ + if ( rc != 0 ) \ + goto done; \ + (_eip) += (_size); \ + (_type)_x; \ +}) + +/* Access/update address held in a register, based on addressing mode. */ +#define register_address(base, reg) \ + ((base) + ((ad_bytes == sizeof(unsigned long)) ? (reg) : \ + ((reg) & ((1UL << (ad_bytes << 3)) - 1)))) + +#define register_address_increment(reg, inc) \ + do { \ + /* signed type ensures sign extension to long */ \ + int _inc = (inc); \ + if ( ad_bytes == sizeof(unsigned long) ) \ + (reg) += _inc; \ + else \ + (reg) = ((reg) & ~((1UL << (ad_bytes << 3)) - 1)) | \ + (((reg) + _inc) & ((1UL << (ad_bytes << 3)) - 1)); \ + } while (0) + +void *decode_register(u8 modrm_reg, unsigned long *regs, + int highbyte_regs) +{ + void *p; + + p = ®s[modrm_reg]; + if (highbyte_regs && modrm_reg >= 4 && modrm_reg < 8) + p = (unsigned char *)®s[modrm_reg & 3] + 1; + return p; +} + +static int read_descriptor(struct x86_emulate_ctxt *ctxt, + struct x86_emulate_ops *ops, + void *ptr, + u16 *size, unsigned long *address, int op_bytes) +{ + int rc; + + if (op_bytes == 2) + op_bytes = 3; + *address = 0; + rc = ops->read_std((unsigned long)ptr, (unsigned long *)size, 2, ctxt); + if (rc) + return rc; + rc = ops->read_std((unsigned long)ptr + 2, address, op_bytes, ctxt); + return rc; +} + +int +x86_emulate_memop(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) +{ + u8 b, d, sib, twobyte = 0, rex_prefix = 0; + u8 modrm, modrm_mod = 0, modrm_reg = 0, modrm_rm = 0; + unsigned long *override_base = NULL; + unsigned int op_bytes, ad_bytes, lock_prefix = 0, rep_prefix = 0, i; + int rc = 0; + struct operand src, dst; + unsigned long cr2 = ctxt->cr2; + int mode = ctxt->mode; + unsigned long modrm_ea; + int use_modrm_ea, index_reg = 0, base_reg = 0, scale, rip_relative = 0; + + /* Shadow copy of register state. Committed on successful emulation. */ + unsigned long _regs[NR_VCPU_REGS]; + unsigned long _eip = ctxt->vcpu->rip, _eflags = ctxt->eflags; + unsigned long modrm_val = 0; + + memcpy(_regs, ctxt->vcpu->regs, sizeof _regs); + + switch (mode) { + case X86EMUL_MODE_REAL: + case X86EMUL_MODE_PROT16: + op_bytes = ad_bytes = 2; + break; + case X86EMUL_MODE_PROT32: + op_bytes = ad_bytes = 4; + break; +#ifdef __x86_64__ + case X86EMUL_MODE_PROT64: + op_bytes = 4; + ad_bytes = 8; + break; +#endif + default: + return -1; + } + + /* Legacy prefixes. */ + for (i = 0; i < 8; i++) { + switch (b = insn_fetch(u8, 1, _eip)) { + case 0x66: /* operand-size override */ + op_bytes ^= 6; /* switch between 2/4 bytes */ + break; + case 0x67: /* address-size override */ + if (mode == X86EMUL_MODE_PROT64) + ad_bytes ^= 12; /* switch between 4/8 bytes */ + else + ad_bytes ^= 6; /* switch between 2/4 bytes */ + break; + case 0x2e: /* CS override */ + override_base = &ctxt->cs_base; + break; + case 0x3e: /* DS override */ + override_base = &ctxt->ds_base; + break; + case 0x26: /* ES override */ + override_base = &ctxt->es_base; + break; + case 0x64: /* FS override */ + override_base = &ctxt->fs_base; + break; + case 0x65: /* GS override */ + override_base = &ctxt->gs_base; + break; + case 0x36: /* SS override */ + override_base = &ctxt->ss_base; + break; + case 0xf0: /* LOCK */ + lock_prefix = 1; + break; + case 0xf3: /* REP/REPE/REPZ */ + rep_prefix = 1; + break; + case 0xf2: /* REPNE/REPNZ */ + break; + default: + goto done_prefixes; + } + } + +done_prefixes: + + /* REX prefix. */ + if ((mode == X86EMUL_MODE_PROT64) && ((b & 0xf0) == 0x40)) { + rex_prefix = b; + if (b & 8) + op_bytes = 8; /* REX.W */ + modrm_reg = (b & 4) << 1; /* REX.R */ + index_reg = (b & 2) << 2; /* REX.X */ + modrm_rm = base_reg = (b & 1) << 3; /* REG.B */ + b = insn_fetch(u8, 1, _eip); + } + + /* Opcode byte(s). */ + d = opcode_table[b]; + if (d == 0) { + /* Two-byte opcode? */ + if (b == 0x0f) { + twobyte = 1; + b = insn_fetch(u8, 1, _eip); + d = twobyte_table[b]; + } + + /* Unrecognised? */ + if (d == 0) + goto cannot_emulate; + } + + /* ModRM and SIB bytes. */ + if (d & ModRM) { + modrm = insn_fetch(u8, 1, _eip); + modrm_mod |= (modrm & 0xc0) >> 6; + modrm_reg |= (modrm & 0x38) >> 3; + modrm_rm |= (modrm & 0x07); + modrm_ea = 0; + use_modrm_ea = 1; + + if (modrm_mod == 3) { + modrm_val = *(unsigned long *) + decode_register(modrm_rm, _regs, d & ByteOp); + goto modrm_done; + } + + if (ad_bytes == 2) { + unsigned bx = _regs[VCPU_REGS_RBX]; + unsigned bp = _regs[VCPU_REGS_RBP]; + unsigned si = _regs[VCPU_REGS_RSI]; + unsigned di = _regs[VCPU_REGS_RDI]; + + /* 16-bit ModR/M decode. */ + switch (modrm_mod) { + case 0: + if (modrm_rm == 6) + modrm_ea += insn_fetch(u16, 2, _eip); + break; + case 1: + modrm_ea += insn_fetch(s8, 1, _eip); + break; + case 2: + modrm_ea += insn_fetch(u16, 2, _eip); + break; + } + switch (modrm_rm) { + case 0: + modrm_ea += bx + si; + break; + case 1: + modrm_ea += bx + di; + break; + case 2: + modrm_ea += bp + si; + break; + case 3: + modrm_ea += bp + di; + break; + case 4: + modrm_ea += si; + break; + case 5: + modrm_ea += di; + break; + case 6: + if (modrm_mod != 0) + modrm_ea += bp; + break; + case 7: + modrm_ea += bx; + break; + } + if (modrm_rm == 2 || modrm_rm == 3 || + (modrm_rm == 6 && modrm_mod != 0)) + if (!override_base) + override_base = &ctxt->ss_base; + modrm_ea = (u16)modrm_ea; + } else { + /* 32/64-bit ModR/M decode. */ + switch (modrm_rm) { + case 4: + case 12: + sib = insn_fetch(u8, 1, _eip); + index_reg |= (sib >> 3) & 7; + base_reg |= sib & 7; + scale = sib >> 6; + + switch (base_reg) { + case 5: + if (modrm_mod != 0) + modrm_ea += _regs[base_reg]; + else + modrm_ea += insn_fetch(s32, 4, _eip); + break; + default: + modrm_ea += _regs[base_reg]; + } + switch (index_reg) { + case 4: + break; + default: + modrm_ea += _regs[index_reg] << scale; + + } + break; + case 5: + if (modrm_mod != 0) + modrm_ea += _regs[modrm_rm]; + else if (mode == X86EMUL_MODE_PROT64) + rip_relative = 1; + break; + default: + modrm_ea += _regs[modrm_rm]; + break; + } + switch (modrm_mod) { + case 0: + if (modrm_rm == 5) + modrm_ea += insn_fetch(s32, 4, _eip); + break; + case 1: + modrm_ea += insn_fetch(s8, 1, _eip); + break; + case 2: + modrm_ea += insn_fetch(s32, 4, _eip); + break; + } + } + if (!override_base) + override_base = &ctxt->ds_base; + if (mode == X86EMUL_MODE_PROT64 && + override_base != &ctxt->fs_base && + override_base != &ctxt->gs_base) + override_base = NULL; + + if (override_base) + modrm_ea += *override_base; + + if (rip_relative) { + modrm_ea += _eip; + switch (d & SrcMask) { + case SrcImmByte: + modrm_ea += 1; + break; + case SrcImm: + if (d & ByteOp) + modrm_ea += 1; + else + if (op_bytes == 8) + modrm_ea += 4; + else + modrm_ea += op_bytes; + } + } + if (ad_bytes != 8) + modrm_ea = (u32)modrm_ea; + cr2 = modrm_ea; + modrm_done: + ; + } + + /* Decode and fetch the destination operand: register or memory. */ + switch (d & DstMask) { + case ImplicitOps: + /* Special instructions do their own operand decoding. */ + goto special_insn; + case DstReg: + dst.type = OP_REG; + if ((d & ByteOp) + && !(twobyte_table && (b == 0xb6 || b == 0xb7))) { + dst.ptr = decode_register(modrm_reg, _regs, + (rex_prefix == 0)); + dst.val = *(u8 *) dst.ptr; + dst.bytes = 1; + } else { + dst.ptr = decode_register(modrm_reg, _regs, 0); + switch ((dst.bytes = op_bytes)) { + case 2: + dst.val = *(u16 *)dst.ptr; + break; + case 4: + dst.val = *(u32 *)dst.ptr; + break; + case 8: + dst.val = *(u64 *)dst.ptr; + break; + } + } + break; + case DstMem: + dst.type = OP_MEM; + dst.ptr = (unsigned long *)cr2; + dst.bytes = (d & ByteOp) ? 1 : op_bytes; + if (!(d & Mov) && /* optimisation - avoid slow emulated read */ + ((rc = ops->read_emulated((unsigned long)dst.ptr, + &dst.val, dst.bytes, ctxt)) != 0)) + goto done; + break; + } + dst.orig_val = dst.val; + + /* + * Decode and fetch the source operand: register, memory + * or immediate. + */ + switch (d & SrcMask) { + case SrcNone: + break; + case SrcReg: + src.type = OP_REG; + if (d & ByteOp) { + src.ptr = decode_register(modrm_reg, _regs, + (rex_prefix == 0)); + src.val = src.orig_val = *(u8 *) src.ptr; + src.bytes = 1; + } else { + src.ptr = decode_register(modrm_reg, _regs, 0); + switch ((src.bytes = op_bytes)) { + case 2: + src.val = src.orig_val = *(u16 *) src.ptr; + break; + case 4: + src.val = src.orig_val = *(u32 *) src.ptr; + break; + case 8: + src.val = src.orig_val = *(u64 *) src.ptr; + break; + } + } + break; + case SrcMem16: + src.bytes = 2; + goto srcmem_common; + case SrcMem32: + src.bytes = 4; + goto srcmem_common; + case SrcMem: + src.bytes = (d & ByteOp) ? 1 : op_bytes; + srcmem_common: + src.type = OP_MEM; + src.ptr = (unsigned long *)cr2; + if ((rc = ops->read_emulated((unsigned long)src.ptr, + &src.val, src.bytes, ctxt)) != 0) + goto done; + src.orig_val = src.val; + break; + case SrcImm: + src.type = OP_IMM; + src.ptr = (unsigned long *)_eip; + src.bytes = (d & ByteOp) ? 1 : op_bytes; + if (src.bytes == 8) + src.bytes = 4; + /* NB. Immediates are sign-extended as necessary. */ + switch (src.bytes) { + case 1: + src.val = insn_fetch(s8, 1, _eip); + break; + case 2: + src.val = insn_fetch(s16, 2, _eip); + break; + case 4: + src.val = insn_fetch(s32, 4, _eip); + break; + } + break; + case SrcImmByte: + src.type = OP_IMM; + src.ptr = (unsigned long *)_eip; + src.bytes = 1; + src.val = insn_fetch(s8, 1, _eip); + break; + } + + if (twobyte) + goto twobyte_insn; + + switch (b) { + case 0x00 ... 0x05: + add: /* add */ + emulate_2op_SrcV("add", src, dst, _eflags); + break; + case 0x08 ... 0x0d: + or: /* or */ + emulate_2op_SrcV("or", src, dst, _eflags); + break; + case 0x10 ... 0x15: + adc: /* adc */ + emulate_2op_SrcV("adc", src, dst, _eflags); + break; + case 0x18 ... 0x1d: + sbb: /* sbb */ + emulate_2op_SrcV("sbb", src, dst, _eflags); + break; + case 0x20 ... 0x25: + and: /* and */ + emulate_2op_SrcV("and", src, dst, _eflags); + break; + case 0x28 ... 0x2d: + sub: /* sub */ + emulate_2op_SrcV("sub", src, dst, _eflags); + break; + case 0x30 ... 0x35: + xor: /* xor */ + emulate_2op_SrcV("xor", src, dst, _eflags); + break; + case 0x38 ... 0x3d: + cmp: /* cmp */ + emulate_2op_SrcV("cmp", src, dst, _eflags); + break; + case 0x63: /* movsxd */ + if (mode != X86EMUL_MODE_PROT64) + goto cannot_emulate; + dst.val = (s32) src.val; + break; + case 0x80 ... 0x83: /* Grp1 */ + switch (modrm_reg) { + case 0: + goto add; + case 1: + goto or; + case 2: + goto adc; + case 3: + goto sbb; + case 4: + goto and; + case 5: + goto sub; + case 6: + goto xor; + case 7: + goto cmp; + } + break; + case 0x84 ... 0x85: + test: /* test */ + emulate_2op_SrcV("test", src, dst, _eflags); + break; + case 0x86 ... 0x87: /* xchg */ + /* Write back the register source. */ + switch (dst.bytes) { + case 1: + *(u8 *) src.ptr = (u8) dst.val; + break; + case 2: + *(u16 *) src.ptr = (u16) dst.val; + break; + case 4: + *src.ptr = (u32) dst.val; + break; /* 64b reg: zero-extend */ + case 8: + *src.ptr = dst.val; + break; + } + /* + * Write back the memory destination with implicit LOCK + * prefix. + */ + dst.val = src.val; + lock_prefix = 1; + break; + case 0xa0 ... 0xa1: /* mov */ + dst.ptr = (unsigned long *)&_regs[VCPU_REGS_RAX]; + dst.val = src.val; + _eip += ad_bytes; /* skip src displacement */ + break; + case 0xa2 ... 0xa3: /* mov */ + dst.val = (unsigned long)_regs[VCPU_REGS_RAX]; + _eip += ad_bytes; /* skip dst displacement */ + break; + case 0x88 ... 0x8b: /* mov */ + case 0xc6 ... 0xc7: /* mov (sole member of Grp11) */ + dst.val = src.val; + break; + case 0x8f: /* pop (sole member of Grp1a) */ + /* 64-bit mode: POP always pops a 64-bit operand. */ + if (mode == X86EMUL_MODE_PROT64) + dst.bytes = 8; + if ((rc = ops->read_std(register_address(ctxt->ss_base, + _regs[VCPU_REGS_RSP]), + &dst.val, dst.bytes, ctxt)) != 0) + goto done; + register_address_increment(_regs[VCPU_REGS_RSP], dst.bytes); + break; + case 0xc0 ... 0xc1: + grp2: /* Grp2 */ + switch (modrm_reg) { + case 0: /* rol */ + emulate_2op_SrcB("rol", src, dst, _eflags); + break; + case 1: /* ror */ + emulate_2op_SrcB("ror", src, dst, _eflags); + break; + case 2: /* rcl */ + emulate_2op_SrcB("rcl", src, dst, _eflags); + break; + case 3: /* rcr */ + emulate_2op_SrcB("rcr", src, dst, _eflags); + break; + case 4: /* sal/shl */ + case 6: /* sal/shl */ + emulate_2op_SrcB("sal", src, dst, _eflags); + break; + case 5: /* shr */ + emulate_2op_SrcB("shr", src, dst, _eflags); + break; + case 7: /* sar */ + emulate_2op_SrcB("sar", src, dst, _eflags); + break; + } + break; + case 0xd0 ... 0xd1: /* Grp2 */ + src.val = 1; + goto grp2; + case 0xd2 ... 0xd3: /* Grp2 */ + src.val = _regs[VCPU_REGS_RCX]; + goto grp2; + case 0xf6 ... 0xf7: /* Grp3 */ + switch (modrm_reg) { + case 0 ... 1: /* test */ + /* + * Special case in Grp3: test has an immediate + * source operand. + */ + src.type = OP_IMM; + src.ptr = (unsigned long *)_eip; + src.bytes = (d & ByteOp) ? 1 : op_bytes; + if (src.bytes == 8) + src.bytes = 4; + switch (src.bytes) { + case 1: + src.val = insn_fetch(s8, 1, _eip); + break; + case 2: + src.val = insn_fetch(s16, 2, _eip); + break; + case 4: + src.val = insn_fetch(s32, 4, _eip); + break; + } + goto test; + case 2: /* not */ + dst.val = ~dst.val; + break; + case 3: /* neg */ + emulate_1op("neg", dst, _eflags); + break; + default: + goto cannot_emulate; + } + break; + case 0xfe ... 0xff: /* Grp4/Grp5 */ + switch (modrm_reg) { + case 0: /* inc */ + emulate_1op("inc", dst, _eflags); + break; + case 1: /* dec */ + emulate_1op("dec", dst, _eflags); + break; + case 6: /* push */ + /* 64-bit mode: PUSH always pushes a 64-bit operand. */ + if (mode == X86EMUL_MODE_PROT64) { + dst.bytes = 8; + if ((rc = ops->read_std((unsigned long)dst.ptr, + &dst.val, 8, + ctxt)) != 0) + goto done; + } + register_address_increment(_regs[VCPU_REGS_RSP], + -dst.bytes); + if ((rc = ops->write_std( + register_address(ctxt->ss_base, + _regs[VCPU_REGS_RSP]), + dst.val, dst.bytes, ctxt)) != 0) + goto done; + dst.val = dst.orig_val; /* skanky: disable writeback */ + break; + default: + goto cannot_emulate; + } + break; + } + +writeback: + if ((d & Mov) || (dst.orig_val != dst.val)) { + switch (dst.type) { + case OP_REG: + /* The 4-byte case *is* correct: in 64-bit mode we zero-extend. */ + switch (dst.bytes) { + case 1: + *(u8 *)dst.ptr = (u8)dst.val; + break; + case 2: + *(u16 *)dst.ptr = (u16)dst.val; + break; + case 4: + *dst.ptr = (u32)dst.val; + break; /* 64b: zero-ext */ + case 8: + *dst.ptr = dst.val; + break; + } + break; + case OP_MEM: + if (lock_prefix) + rc = ops->cmpxchg_emulated((unsigned long)dst. + ptr, dst.orig_val, + dst.val, dst.bytes, + ctxt); + else + rc = ops->write_emulated((unsigned long)dst.ptr, + dst.val, dst.bytes, + ctxt); + if (rc != 0) + goto done; + default: + break; + } + } + + /* Commit shadow register state. */ + memcpy(ctxt->vcpu->regs, _regs, sizeof _regs); + ctxt->eflags = _eflags; + ctxt->vcpu->rip = _eip; + +done: + return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; + +special_insn: + if (twobyte) + goto twobyte_special_insn; + if (rep_prefix) { + if (_regs[VCPU_REGS_RCX] == 0) { + ctxt->vcpu->rip = _eip; + goto done; + } + _regs[VCPU_REGS_RCX]--; + _eip = ctxt->vcpu->rip; + } + switch (b) { + case 0xa4 ... 0xa5: /* movs */ + dst.type = OP_MEM; + dst.bytes = (d & ByteOp) ? 1 : op_bytes; + dst.ptr = (unsigned long *)register_address(ctxt->es_base, + _regs[VCPU_REGS_RDI]); + if ((rc = ops->read_emulated(register_address( + override_base ? *override_base : ctxt->ds_base, + _regs[VCPU_REGS_RSI]), &dst.val, dst.bytes, ctxt)) != 0) + goto done; + register_address_increment(_regs[VCPU_REGS_RSI], + (_eflags & EFLG_DF) ? -dst.bytes : dst.bytes); + register_address_increment(_regs[VCPU_REGS_RDI], + (_eflags & EFLG_DF) ? -dst.bytes : dst.bytes); + break; + case 0xa6 ... 0xa7: /* cmps */ + DPRINTF("Urk! I don't handle CMPS.\n"); + goto cannot_emulate; + case 0xaa ... 0xab: /* stos */ + dst.type = OP_MEM; + dst.bytes = (d & ByteOp) ? 1 : op_bytes; + dst.ptr = (unsigned long *)cr2; + dst.val = _regs[VCPU_REGS_RAX]; + register_address_increment(_regs[VCPU_REGS_RDI], + (_eflags & EFLG_DF) ? -dst.bytes : dst.bytes); + break; + case 0xac ... 0xad: /* lods */ + dst.type = OP_REG; + dst.bytes = (d & ByteOp) ? 1 : op_bytes; + dst.ptr = (unsigned long *)&_regs[VCPU_REGS_RAX]; + if ((rc = ops->read_emulated(cr2, &dst.val, dst.bytes, ctxt)) != 0) + goto done; + register_address_increment(_regs[VCPU_REGS_RSI], + (_eflags & EFLG_DF) ? -dst.bytes : dst.bytes); + break; + case 0xae ... 0xaf: /* scas */ + DPRINTF("Urk! I don't handle SCAS.\n"); + goto cannot_emulate; + } + goto writeback; + +twobyte_insn: + switch (b) { + case 0x01: /* lgdt, lidt, lmsw */ + switch (modrm_reg) { + u16 size; + unsigned long address; + + case 2: /* lgdt */ + rc = read_descriptor(ctxt, ops, src.ptr, + &size, &address, op_bytes); + if (rc) + goto done; + realmode_lgdt(ctxt->vcpu, size, address); + break; + case 3: /* lidt */ + rc = read_descriptor(ctxt, ops, src.ptr, + &size, &address, op_bytes); + if (rc) + goto done; + realmode_lidt(ctxt->vcpu, size, address); + break; + case 4: /* smsw */ + if (modrm_mod != 3) + goto cannot_emulate; + *(u16 *)&_regs[modrm_rm] + = realmode_get_cr(ctxt->vcpu, 0); + break; + case 6: /* lmsw */ + if (modrm_mod != 3) + goto cannot_emulate; + realmode_lmsw(ctxt->vcpu, (u16)modrm_val, &_eflags); + break; + case 7: /* invlpg*/ + emulate_invlpg(ctxt->vcpu, cr2); + break; + default: + goto cannot_emulate; + } + break; + case 0x21: /* mov from dr to reg */ + if (modrm_mod != 3) + goto cannot_emulate; + rc = emulator_get_dr(ctxt, modrm_reg, &_regs[modrm_rm]); + break; + case 0x23: /* mov from reg to dr */ + if (modrm_mod != 3) + goto cannot_emulate; + rc = emulator_set_dr(ctxt, modrm_reg, _regs[modrm_rm]); + break; + case 0x40 ... 0x4f: /* cmov */ + dst.val = dst.orig_val = src.val; + d &= ~Mov; /* default to no move */ + /* + * First, assume we're decoding an even cmov opcode + * (lsb == 0). + */ + switch ((b & 15) >> 1) { + case 0: /* cmovo */ + d |= (_eflags & EFLG_OF) ? Mov : 0; + break; + case 1: /* cmovb/cmovc/cmovnae */ + d |= (_eflags & EFLG_CF) ? Mov : 0; + break; + case 2: /* cmovz/cmove */ + d |= (_eflags & EFLG_ZF) ? Mov : 0; + break; + case 3: /* cmovbe/cmovna */ + d |= (_eflags & (EFLG_CF | EFLG_ZF)) ? Mov : 0; + break; + case 4: /* cmovs */ + d |= (_eflags & EFLG_SF) ? Mov : 0; + break; + case 5: /* cmovp/cmovpe */ + d |= (_eflags & EFLG_PF) ? Mov : 0; + break; + case 7: /* cmovle/cmovng */ + d |= (_eflags & EFLG_ZF) ? Mov : 0; + /* fall through */ + case 6: /* cmovl/cmovnge */ + d |= (!(_eflags & EFLG_SF) != + !(_eflags & EFLG_OF)) ? Mov : 0; + break; + } + /* Odd cmov opcodes (lsb == 1) have inverted sense. */ + d ^= (b & 1) ? Mov : 0; + break; + case 0xb0 ... 0xb1: /* cmpxchg */ + /* + * Save real source value, then compare EAX against + * destination. + */ + src.orig_val = src.val; + src.val = _regs[VCPU_REGS_RAX]; + emulate_2op_SrcV("cmp", src, dst, _eflags); + /* Always write back. The question is: where to? */ + d |= Mov; + if (_eflags & EFLG_ZF) { + /* Success: write back to memory. */ + dst.val = src.orig_val; + } else { + /* Failure: write the value we saw to EAX. */ + dst.type = OP_REG; + dst.ptr = (unsigned long *)&_regs[VCPU_REGS_RAX]; + } + break; + case 0xa3: + bt: /* bt */ + src.val &= (dst.bytes << 3) - 1; /* only subword offset */ + emulate_2op_SrcV_nobyte("bt", src, dst, _eflags); + break; + case 0xb3: + btr: /* btr */ + src.val &= (dst.bytes << 3) - 1; /* only subword offset */ + emulate_2op_SrcV_nobyte("btr", src, dst, _eflags); + break; + case 0xab: + bts: /* bts */ + src.val &= (dst.bytes << 3) - 1; /* only subword offset */ + emulate_2op_SrcV_nobyte("bts", src, dst, _eflags); + break; + case 0xb6 ... 0xb7: /* movzx */ + dst.bytes = op_bytes; + dst.val = (d & ByteOp) ? (u8) src.val : (u16) src.val; + break; + case 0xbb: + btc: /* btc */ + src.val &= (dst.bytes << 3) - 1; /* only subword offset */ + emulate_2op_SrcV_nobyte("btc", src, dst, _eflags); + break; + case 0xba: /* Grp8 */ + switch (modrm_reg & 3) { + case 0: + goto bt; + case 1: + goto bts; + case 2: + goto btr; + case 3: + goto btc; + } + break; + case 0xbe ... 0xbf: /* movsx */ + dst.bytes = op_bytes; + dst.val = (d & ByteOp) ? (s8) src.val : (s16) src.val; + break; + } + goto writeback; + +twobyte_special_insn: + /* Disable writeback. */ + dst.orig_val = dst.val; + switch (b) { + case 0x0d: /* GrpP (prefetch) */ + case 0x18: /* Grp16 (prefetch/nop) */ + break; + case 0x06: + emulate_clts(ctxt->vcpu); + break; + case 0x20: /* mov cr, reg */ + if (modrm_mod != 3) + goto cannot_emulate; + _regs[modrm_rm] = realmode_get_cr(ctxt->vcpu, modrm_reg); + break; + case 0x22: /* mov reg, cr */ + if (modrm_mod != 3) + goto cannot_emulate; + realmode_set_cr(ctxt->vcpu, modrm_reg, modrm_val, &_eflags); + break; + case 0xc7: /* Grp9 (cmpxchg8b) */ +#if defined(__i386__) + { + unsigned long old_lo, old_hi; + if (((rc = ops->read_emulated(cr2 + 0, &old_lo, 4, + ctxt)) != 0) + || ((rc = ops->read_emulated(cr2 + 4, &old_hi, 4, + ctxt)) != 0)) + goto done; + if ((old_lo != _regs[VCPU_REGS_RAX]) + || (old_hi != _regs[VCPU_REGS_RDI])) { + _regs[VCPU_REGS_RAX] = old_lo; + _regs[VCPU_REGS_RDX] = old_hi; + _eflags &= ~EFLG_ZF; + } else if (ops->cmpxchg8b_emulated == NULL) { + rc = X86EMUL_UNHANDLEABLE; + goto done; + } else { + if ((rc = ops->cmpxchg8b_emulated(cr2, old_lo, + old_hi, + _regs[VCPU_REGS_RBX], + _regs[VCPU_REGS_RCX], + ctxt)) != 0) + goto done; + _eflags |= EFLG_ZF; + } + break; + } +#elif defined(__x86_64__) + { + unsigned long old, new; + if ((rc = ops->read_emulated(cr2, &old, 8, ctxt)) != 0) + goto done; + if (((u32) (old >> 0) != (u32) _regs[VCPU_REGS_RAX]) || + ((u32) (old >> 32) != (u32) _regs[VCPU_REGS_RDX])) { + _regs[VCPU_REGS_RAX] = (u32) (old >> 0); + _regs[VCPU_REGS_RDX] = (u32) (old >> 32); + _eflags &= ~EFLG_ZF; + } else { + new = (_regs[VCPU_REGS_RCX] << 32) | (u32) _regs[VCPU_REGS_RBX]; + if ((rc = ops->cmpxchg_emulated(cr2, old, + new, 8, ctxt)) != 0) + goto done; + _eflags |= EFLG_ZF; + } + break; + } +#endif + } + goto writeback; + +cannot_emulate: + DPRINTF("Cannot emulate %02x\n", b); + return -1; +} + +#ifdef __XEN__ + +#include +#include + +int +x86_emulate_read_std(unsigned long addr, + unsigned long *val, + unsigned int bytes, struct x86_emulate_ctxt *ctxt) +{ + unsigned int rc; + + *val = 0; + + if ((rc = copy_from_user((void *)val, (void *)addr, bytes)) != 0) { + propagate_page_fault(addr + bytes - rc, 0); /* read fault */ + return X86EMUL_PROPAGATE_FAULT; + } + + return X86EMUL_CONTINUE; +} + +int +x86_emulate_write_std(unsigned long addr, + unsigned long val, + unsigned int bytes, struct x86_emulate_ctxt *ctxt) +{ + unsigned int rc; + + if ((rc = copy_to_user((void *)addr, (void *)&val, bytes)) != 0) { + propagate_page_fault(addr + bytes - rc, PGERR_write_access); + return X86EMUL_PROPAGATE_FAULT; + } + + return X86EMUL_CONTINUE; +} + +#endif diff --git a/drivers/kvm/x86_emulate.h b/drivers/kvm/x86_emulate.h new file mode 100644 index 000000000000..658b58de30fc --- /dev/null +++ b/drivers/kvm/x86_emulate.h @@ -0,0 +1,185 @@ +/****************************************************************************** + * x86_emulate.h + * + * Generic x86 (32-bit and 64-bit) instruction decoder and emulator. + * + * Copyright (c) 2005 Keir Fraser + * + * From: xen-unstable 10676:af9809f51f81a3c43f276f00c81a52ef558afda4 + */ + +#ifndef __X86_EMULATE_H__ +#define __X86_EMULATE_H__ + +struct x86_emulate_ctxt; + +/* + * x86_emulate_ops: + * + * These operations represent the instruction emulator's interface to memory. + * There are two categories of operation: those that act on ordinary memory + * regions (*_std), and those that act on memory regions known to require + * special treatment or emulation (*_emulated). + * + * The emulator assumes that an instruction accesses only one 'emulated memory' + * location, that this location is the given linear faulting address (cr2), and + * that this is one of the instruction's data operands. Instruction fetches and + * stack operations are assumed never to access emulated memory. The emulator + * automatically deduces which operand of a string-move operation is accessing + * emulated memory, and assumes that the other operand accesses normal memory. + * + * NOTES: + * 1. The emulator isn't very smart about emulated vs. standard memory. + * 'Emulated memory' access addresses should be checked for sanity. + * 'Normal memory' accesses may fault, and the caller must arrange to + * detect and handle reentrancy into the emulator via recursive faults. + * Accesses may be unaligned and may cross page boundaries. + * 2. If the access fails (cannot emulate, or a standard access faults) then + * it is up to the memop to propagate the fault to the guest VM via + * some out-of-band mechanism, unknown to the emulator. The memop signals + * failure by returning X86EMUL_PROPAGATE_FAULT to the emulator, which will + * then immediately bail. + * 3. Valid access sizes are 1, 2, 4 and 8 bytes. On x86/32 systems only + * cmpxchg8b_emulated need support 8-byte accesses. + * 4. The emulator cannot handle 64-bit mode emulation on an x86/32 system. + */ +/* Access completed successfully: continue emulation as normal. */ +#define X86EMUL_CONTINUE 0 +/* Access is unhandleable: bail from emulation and return error to caller. */ +#define X86EMUL_UNHANDLEABLE 1 +/* Terminate emulation but return success to the caller. */ +#define X86EMUL_PROPAGATE_FAULT 2 /* propagate a generated fault to guest */ +#define X86EMUL_RETRY_INSTR 2 /* retry the instruction for some reason */ +#define X86EMUL_CMPXCHG_FAILED 2 /* cmpxchg did not see expected value */ +struct x86_emulate_ops { + /* + * read_std: Read bytes of standard (non-emulated/special) memory. + * Used for instruction fetch, stack operations, and others. + * @addr: [IN ] Linear address from which to read. + * @val: [OUT] Value read from memory, zero-extended to 'u_long'. + * @bytes: [IN ] Number of bytes to read from memory. + */ + int (*read_std)(unsigned long addr, + unsigned long *val, + unsigned int bytes, struct x86_emulate_ctxt * ctxt); + + /* + * write_std: Write bytes of standard (non-emulated/special) memory. + * Used for stack operations, and others. + * @addr: [IN ] Linear address to which to write. + * @val: [IN ] Value to write to memory (low-order bytes used as + * required). + * @bytes: [IN ] Number of bytes to write to memory. + */ + int (*write_std)(unsigned long addr, + unsigned long val, + unsigned int bytes, struct x86_emulate_ctxt * ctxt); + + /* + * read_emulated: Read bytes from emulated/special memory area. + * @addr: [IN ] Linear address from which to read. + * @val: [OUT] Value read from memory, zero-extended to 'u_long'. + * @bytes: [IN ] Number of bytes to read from memory. + */ + int (*read_emulated) (unsigned long addr, + unsigned long *val, + unsigned int bytes, + struct x86_emulate_ctxt * ctxt); + + /* + * write_emulated: Read bytes from emulated/special memory area. + * @addr: [IN ] Linear address to which to write. + * @val: [IN ] Value to write to memory (low-order bytes used as + * required). + * @bytes: [IN ] Number of bytes to write to memory. + */ + int (*write_emulated) (unsigned long addr, + unsigned long val, + unsigned int bytes, + struct x86_emulate_ctxt * ctxt); + + /* + * cmpxchg_emulated: Emulate an atomic (LOCKed) CMPXCHG operation on an + * emulated/special memory area. + * @addr: [IN ] Linear address to access. + * @old: [IN ] Value expected to be current at @addr. + * @new: [IN ] Value to write to @addr. + * @bytes: [IN ] Number of bytes to access using CMPXCHG. + */ + int (*cmpxchg_emulated) (unsigned long addr, + unsigned long old, + unsigned long new, + unsigned int bytes, + struct x86_emulate_ctxt * ctxt); + + /* + * cmpxchg8b_emulated: Emulate an atomic (LOCKed) CMPXCHG8B operation on an + * emulated/special memory area. + * @addr: [IN ] Linear address to access. + * @old: [IN ] Value expected to be current at @addr. + * @new: [IN ] Value to write to @addr. + * NOTES: + * 1. This function is only ever called when emulating a real CMPXCHG8B. + * 2. This function is *never* called on x86/64 systems. + * 2. Not defining this function (i.e., specifying NULL) is equivalent + * to defining a function that always returns X86EMUL_UNHANDLEABLE. + */ + int (*cmpxchg8b_emulated) (unsigned long addr, + unsigned long old_lo, + unsigned long old_hi, + unsigned long new_lo, + unsigned long new_hi, + struct x86_emulate_ctxt * ctxt); +}; + +struct cpu_user_regs; + +struct x86_emulate_ctxt { + /* Register state before/after emulation. */ + struct kvm_vcpu *vcpu; + + /* Linear faulting address (if emulating a page-faulting instruction). */ + unsigned long eflags; + unsigned long cr2; + + /* Emulated execution mode, represented by an X86EMUL_MODE value. */ + int mode; + + unsigned long cs_base; + unsigned long ds_base; + unsigned long es_base; + unsigned long ss_base; + unsigned long gs_base; + unsigned long fs_base; +}; + +/* Execution mode, passed to the emulator. */ +#define X86EMUL_MODE_REAL 0 /* Real mode. */ +#define X86EMUL_MODE_PROT16 2 /* 16-bit protected mode. */ +#define X86EMUL_MODE_PROT32 4 /* 32-bit protected mode. */ +#define X86EMUL_MODE_PROT64 8 /* 64-bit (long) mode. */ + +/* Host execution mode. */ +#if defined(__i386__) +#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT32 +#elif defined(__x86_64__) +#define X86EMUL_MODE_HOST X86EMUL_MODE_PROT64 +#endif + +/* + * x86_emulate_memop: Emulate an instruction that faulted attempting to + * read/write a 'special' memory area. + * Returns -1 on failure, 0 on success. + */ +int x86_emulate_memop(struct x86_emulate_ctxt *ctxt, + struct x86_emulate_ops *ops); + +/* + * Given the 'reg' portion of a ModRM byte, and a register block, return a + * pointer into the block that addresses the relevant register. + * @highbyte_regs specifies whether to decode AH,CH,DH,BH. + */ +void *decode_register(u8 modrm_reg, unsigned long *regs, + int highbyte_regs); + +#endif /* __X86_EMULATE_H__ */ diff --git a/include/linux/kvm.h b/include/linux/kvm.h new file mode 100644 index 000000000000..5bb2c3c585c1 --- /dev/null +++ b/include/linux/kvm.h @@ -0,0 +1,227 @@ +#ifndef __LINUX_KVM_H +#define __LINUX_KVM_H + +/* + * Userspace interface for /dev/kvm - kernel based virtual machine + * + * Note: this interface is considered experimental and may change without + * notice. + */ + +#include +#include + +/* + * Architectural interrupt line count, and the size of the bitmap needed + * to hold them. + */ +#define KVM_NR_INTERRUPTS 256 +#define KVM_IRQ_BITMAP_SIZE_BYTES ((KVM_NR_INTERRUPTS + 7) / 8) +#define KVM_IRQ_BITMAP_SIZE(type) (KVM_IRQ_BITMAP_SIZE_BYTES / sizeof(type)) + + +/* for KVM_CREATE_MEMORY_REGION */ +struct kvm_memory_region { + __u32 slot; + __u32 flags; + __u64 guest_phys_addr; + __u64 memory_size; /* bytes */ +}; + +/* for kvm_memory_region::flags */ +#define KVM_MEM_LOG_DIRTY_PAGES 1UL + + +#define KVM_EXIT_TYPE_FAIL_ENTRY 1 +#define KVM_EXIT_TYPE_VM_EXIT 2 + +enum kvm_exit_reason { + KVM_EXIT_UNKNOWN = 0, + KVM_EXIT_EXCEPTION = 1, + KVM_EXIT_IO = 2, + KVM_EXIT_CPUID = 3, + KVM_EXIT_DEBUG = 4, + KVM_EXIT_HLT = 5, + KVM_EXIT_MMIO = 6, +}; + +/* for KVM_RUN */ +struct kvm_run { + /* in */ + __u32 vcpu; + __u32 emulated; /* skip current instruction */ + __u32 mmio_completed; /* mmio request completed */ + + /* out */ + __u32 exit_type; + __u32 exit_reason; + __u32 instruction_length; + union { + /* KVM_EXIT_UNKNOWN */ + struct { + __u32 hardware_exit_reason; + } hw; + /* KVM_EXIT_EXCEPTION */ + struct { + __u32 exception; + __u32 error_code; + } ex; + /* KVM_EXIT_IO */ + struct { +#define KVM_EXIT_IO_IN 0 +#define KVM_EXIT_IO_OUT 1 + __u8 direction; + __u8 size; /* bytes */ + __u8 string; + __u8 string_down; + __u8 rep; + __u8 pad; + __u16 port; + __u64 count; + union { + __u64 address; + __u32 value; + }; + } io; + struct { + } debug; + /* KVM_EXIT_MMIO */ + struct { + __u64 phys_addr; + __u8 data[8]; + __u32 len; + __u8 is_write; + } mmio; + }; +}; + +/* for KVM_GET_REGS and KVM_SET_REGS */ +struct kvm_regs { + /* in */ + __u32 vcpu; + __u32 padding; + + /* out (KVM_GET_REGS) / in (KVM_SET_REGS) */ + __u64 rax, rbx, rcx, rdx; + __u64 rsi, rdi, rsp, rbp; + __u64 r8, r9, r10, r11; + __u64 r12, r13, r14, r15; + __u64 rip, rflags; +}; + +struct kvm_segment { + __u64 base; + __u32 limit; + __u16 selector; + __u8 type; + __u8 present, dpl, db, s, l, g, avl; + __u8 unusable; + __u8 padding; +}; + +struct kvm_dtable { + __u64 base; + __u16 limit; + __u16 padding[3]; +}; + +/* for KVM_GET_SREGS and KVM_SET_SREGS */ +struct kvm_sregs { + /* in */ + __u32 vcpu; + __u32 padding; + + /* out (KVM_GET_SREGS) / in (KVM_SET_SREGS) */ + struct kvm_segment cs, ds, es, fs, gs, ss; + struct kvm_segment tr, ldt; + struct kvm_dtable gdt, idt; + __u64 cr0, cr2, cr3, cr4, cr8; + __u64 efer; + __u64 apic_base; + __u64 interrupt_bitmap[KVM_IRQ_BITMAP_SIZE(__u64)]; +}; + +struct kvm_msr_entry { + __u32 index; + __u32 reserved; + __u64 data; +}; + +/* for KVM_GET_MSRS and KVM_SET_MSRS */ +struct kvm_msrs { + __u32 vcpu; + __u32 nmsrs; /* number of msrs in entries */ + + struct kvm_msr_entry entries[0]; +}; + +/* for KVM_GET_MSR_INDEX_LIST */ +struct kvm_msr_list { + __u32 nmsrs; /* number of msrs in entries */ + __u32 indices[0]; +}; + +/* for KVM_TRANSLATE */ +struct kvm_translation { + /* in */ + __u64 linear_address; + __u32 vcpu; + __u32 padding; + + /* out */ + __u64 physical_address; + __u8 valid; + __u8 writeable; + __u8 usermode; +}; + +/* for KVM_INTERRUPT */ +struct kvm_interrupt { + /* in */ + __u32 vcpu; + __u32 irq; +}; + +struct kvm_breakpoint { + __u32 enabled; + __u32 padding; + __u64 address; +}; + +/* for KVM_DEBUG_GUEST */ +struct kvm_debug_guest { + /* int */ + __u32 vcpu; + __u32 enabled; + struct kvm_breakpoint breakpoints[4]; + __u32 singlestep; +}; + +/* for KVM_GET_DIRTY_LOG */ +struct kvm_dirty_log { + __u32 slot; + __u32 padding; + union { + void __user *dirty_bitmap; /* one bit per page */ + __u64 padding; + }; +}; + +#define KVMIO 0xAE + +#define KVM_RUN _IOWR(KVMIO, 2, struct kvm_run) +#define KVM_GET_REGS _IOWR(KVMIO, 3, struct kvm_regs) +#define KVM_SET_REGS _IOW(KVMIO, 4, struct kvm_regs) +#define KVM_GET_SREGS _IOWR(KVMIO, 5, struct kvm_sregs) +#define KVM_SET_SREGS _IOW(KVMIO, 6, struct kvm_sregs) +#define KVM_TRANSLATE _IOWR(KVMIO, 7, struct kvm_translation) +#define KVM_INTERRUPT _IOW(KVMIO, 8, struct kvm_interrupt) +#define KVM_DEBUG_GUEST _IOW(KVMIO, 9, struct kvm_debug_guest) +#define KVM_SET_MEMORY_REGION _IOW(KVMIO, 10, struct kvm_memory_region) +#define KVM_CREATE_VCPU _IOW(KVMIO, 11, int /* vcpu_slot */) +#define KVM_GET_DIRTY_LOG _IOW(KVMIO, 12, struct kvm_dirty_log) +#define KVM_GET_MSRS _IOWR(KVMIO, 13, struct kvm_msrs) +#define KVM_SET_MSRS _IOWR(KVMIO, 14, struct kvm_msrs) +#define KVM_GET_MSR_INDEX_LIST _IOWR(KVMIO, 15, struct kvm_msr_list) + +#endif -- cgit v1.2.3 From 36cfb5ccfa39ddd030926d08fcf80ba553c88737 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 10 Dec 2006 21:21:28 +0100 Subject: i2c: Update the list of driver IDs * A chip driver ID was assigned to the Radeon, while it is an adapter so it needs an i2c adapter ID. * The SAA7191 is a video decoder, not encoder. * The icspll driver is dead, and will never be ported to Linux 2.6. Signed-off-by: Jean Delvare --- include/linux/i2c-id.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 0a8f750cbede..36f0c5b0cbf7 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -82,9 +82,8 @@ #define I2C_DRIVERID_STM41T00 52 /* real time clock */ #define I2C_DRIVERID_UDA1342 53 /* UDA1342 audio codec */ #define I2C_DRIVERID_ADV7170 54 /* video encoder */ -#define I2C_DRIVERID_RADEON 55 /* I2C bus on Radeon boards */ #define I2C_DRIVERID_MAX1617 56 /* temp sensor */ -#define I2C_DRIVERID_SAA7191 57 /* video encoder */ +#define I2C_DRIVERID_SAA7191 57 /* video decoder */ #define I2C_DRIVERID_INDYCAM 58 /* SGI IndyCam */ #define I2C_DRIVERID_BT832 59 /* CMOS camera video processor */ #define I2C_DRIVERID_TDA9887 60 /* TDA988x IF-PLL demodulator */ @@ -132,7 +131,6 @@ #define I2C_DRIVERID_ADM1021 1008 #define I2C_DRIVERID_ADM9240 1009 #define I2C_DRIVERID_LTC1710 1010 -#define I2C_DRIVERID_ICSPLL 1012 #define I2C_DRIVERID_BT869 1013 #define I2C_DRIVERID_MAXILIFE 1014 #define I2C_DRIVERID_MATORB 1015 -- cgit v1.2.3 From 51fd554b6547b74c7e6d1f52885ba8532b531023 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 10 Dec 2006 21:21:29 +0100 Subject: i2c: Delete the broken i2c-ite bus driver The rest of the ITE8172 support was already removed from MIPS tree. Signed-off-by: Jean Delvare Signed-off-by: Yoichi Yuasa Acked-by: Ralf Baechle --- Documentation/feature-removal-schedule.txt | 11 - drivers/i2c/algos/Kconfig | 11 - drivers/i2c/algos/Makefile | 1 - drivers/i2c/algos/i2c-algo-ite.c | 806 ----------------------------- drivers/i2c/algos/i2c-algo-ite.h | 117 ----- drivers/i2c/busses/Kconfig | 12 - drivers/i2c/busses/Makefile | 1 - drivers/i2c/busses/i2c-ite.c | 278 ---------- include/linux/i2c-algo-ite.h | 72 --- include/linux/i2c-id.h | 3 - 10 files changed, 1312 deletions(-) delete mode 100644 drivers/i2c/algos/i2c-algo-ite.c delete mode 100644 drivers/i2c/algos/i2c-algo-ite.h delete mode 100644 drivers/i2c/busses/i2c-ite.c delete mode 100644 include/linux/i2c-algo-ite.h (limited to 'include/linux') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 46f2a559b27c..031029e89fd9 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -216,17 +216,6 @@ Who: Thomas Gleixner --------------------------- -What: i2c-ite and i2c-algo-ite drivers -When: September 2006 -Why: These drivers never compiled since they were added to the kernel - tree 5 years ago. This feature removal can be reevaluated if - someone shows interest in the drivers, fixes them and takes over - maintenance. - http://marc.theaimsgroup.com/?l=linux-mips&m=115040510817448 -Who: Jean Delvare - ---------------------------- - What: Bridge netfilter deferred IPv4/IPv6 output hook calling When: January 2007 Why: The deferred output hooks are a layering violation causing unusual diff --git a/drivers/i2c/algos/Kconfig b/drivers/i2c/algos/Kconfig index c034820615bb..af0203409dd1 100644 --- a/drivers/i2c/algos/Kconfig +++ b/drivers/i2c/algos/Kconfig @@ -38,17 +38,6 @@ config I2C_ALGOPCA This support is also available as a module. If so, the module will be called i2c-algo-pca. -config I2C_ALGOITE - tristate "ITE I2C Algorithm" - depends on MIPS_ITE8172 && I2C - help - This supports the use of the ITE8172 I2C interface found on some MIPS - systems. Say Y if you have one of these. You should also say Y for - the ITE I2C peripheral driver support below. - - This support is also available as a module. If so, the module - will be called i2c-algo-ite. - config I2C_ALGO8XX tristate "MPC8xx CPM I2C interface" depends on 8xx && I2C diff --git a/drivers/i2c/algos/Makefile b/drivers/i2c/algos/Makefile index 208be04a3dbd..cac1051bd4f1 100644 --- a/drivers/i2c/algos/Makefile +++ b/drivers/i2c/algos/Makefile @@ -5,7 +5,6 @@ obj-$(CONFIG_I2C_ALGOBIT) += i2c-algo-bit.o obj-$(CONFIG_I2C_ALGOPCF) += i2c-algo-pcf.o obj-$(CONFIG_I2C_ALGOPCA) += i2c-algo-pca.o -obj-$(CONFIG_I2C_ALGOITE) += i2c-algo-ite.o obj-$(CONFIG_I2C_ALGO_SGI) += i2c-algo-sgi.o ifeq ($(CONFIG_I2C_DEBUG_ALGO),y) diff --git a/drivers/i2c/algos/i2c-algo-ite.c b/drivers/i2c/algos/i2c-algo-ite.c deleted file mode 100644 index 70d8eefb5efc..000000000000 --- a/drivers/i2c/algos/i2c-algo-ite.c +++ /dev/null @@ -1,806 +0,0 @@ -/* - ------------------------------------------------------------------------- - i2c-algo-ite.c i2c driver algorithms for ITE adapters - - Hai-Pao Fan, MontaVista Software, Inc. - hpfan@mvista.com or source@mvista.com - - Copyright 2000 MontaVista Software Inc. - - --------------------------------------------------------------------------- - This file was highly leveraged from i2c-algo-pcf.c, which was created - by Simon G. Vogl and Hans Berglund: - - - Copyright (C) 1995-1997 Simon G. Vogl - 1998-2000 Hans Berglund - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -/* ------------------------------------------------------------------------- */ - -/* With some changes from Kyösti Mälkki and - Frodo Looijaard ,and also from Martin Bailey - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "i2c-algo-ite.h" - -#define PM_DSR IT8172_PCI_IO_BASE + IT_PM_DSR -#define PM_IBSR IT8172_PCI_IO_BASE + IT_PM_DSR + 0x04 -#define GPIO_CCR IT8172_PCI_IO_BASE + IT_GPCCR - -#define DEB2(x) if (i2c_debug>=2) x -#define DEB3(x) if (i2c_debug>=3) x /* print several statistical values*/ -#define DEF_TIMEOUT 16 - - -/* module parameters: - */ -static int i2c_debug; -static int iic_test; /* see if the line-setting functions work */ - -/* --- setting states on the bus with the right timing: --------------- */ - -#define get_clock(adap) adap->getclock(adap->data) -#define iic_outw(adap, reg, val) adap->setiic(adap->data, reg, val) -#define iic_inw(adap, reg) adap->getiic(adap->data, reg) - - -/* --- other auxiliary functions -------------------------------------- */ - -static void iic_start(struct i2c_algo_iic_data *adap) -{ - iic_outw(adap,ITE_I2CHCR,ITE_CMD); -} - -static void iic_stop(struct i2c_algo_iic_data *adap) -{ - iic_outw(adap,ITE_I2CHCR,0); - iic_outw(adap,ITE_I2CHSR,ITE_I2CHSR_TDI); -} - -static void iic_reset(struct i2c_algo_iic_data *adap) -{ - iic_outw(adap, PM_IBSR, iic_inw(adap, PM_IBSR) | 0x80); -} - - -static int wait_for_bb(struct i2c_algo_iic_data *adap) -{ - int timeout = DEF_TIMEOUT; - short status; - - status = iic_inw(adap, ITE_I2CHSR); -#ifndef STUB_I2C - while (timeout-- && (status & ITE_I2CHSR_HB)) { - udelay(1000); /* How much is this? */ - status = iic_inw(adap, ITE_I2CHSR); - } -#endif - if (timeout<=0) { - printk(KERN_ERR "Timeout, host is busy\n"); - iic_reset(adap); - } - return(timeout<=0); -} - -/* After we issue a transaction on the IIC bus, this function - * is called. It puts this process to sleep until we get an interrupt from - * from the controller telling us that the transaction we requested in complete. - */ -static int wait_for_pin(struct i2c_algo_iic_data *adap, short *status) { - - int timeout = DEF_TIMEOUT; - - timeout = wait_for_bb(adap); - if (timeout) { - DEB2(printk("Timeout waiting for host not busy\n");) - return -EIO; - } - timeout = DEF_TIMEOUT; - - *status = iic_inw(adap, ITE_I2CHSR); -#ifndef STUB_I2C - while (timeout-- && !(*status & ITE_I2CHSR_TDI)) { - adap->waitforpin(); - *status = iic_inw(adap, ITE_I2CHSR); - } -#endif - if (timeout <= 0) - return(-1); - else - return(0); -} - -static int wait_for_fe(struct i2c_algo_iic_data *adap, short *status) -{ - int timeout = DEF_TIMEOUT; - - *status = iic_inw(adap, ITE_I2CFSR); -#ifndef STUB_I2C - while (timeout-- && (*status & ITE_I2CFSR_FE)) { - udelay(1000); - iic_inw(adap, ITE_I2CFSR); - } -#endif - if (timeout <= 0) - return(-1); - else - return(0); -} - -static int iic_init (struct i2c_algo_iic_data *adap) -{ - short i; - - /* Clear bit 7 to set I2C to normal operation mode */ - i=iic_inw(adap, PM_DSR)& 0xff7f; - iic_outw(adap, PM_DSR, i); - - /* set IT_GPCCR port C bit 2&3 as function 2 */ - i = iic_inw(adap, GPIO_CCR) & 0xfc0f; - iic_outw(adap,GPIO_CCR,i); - - /* Clear slave address/sub-address */ - iic_outw(adap,ITE_I2CSAR, 0); - iic_outw(adap,ITE_I2CSSAR, 0); - - /* Set clock counter register */ - iic_outw(adap,ITE_I2CCKCNT, get_clock(adap)); - - /* Set START/reSTART/STOP time registers */ - iic_outw(adap,ITE_I2CSHDR, 0x0a); - iic_outw(adap,ITE_I2CRSUR, 0x0a); - iic_outw(adap,ITE_I2CPSUR, 0x0a); - - /* Enable interrupts on completing the current transaction */ - iic_outw(adap,ITE_I2CHCR, ITE_I2CHCR_IE | ITE_I2CHCR_HCE); - - /* Clear transfer count */ - iic_outw(adap,ITE_I2CFBCR, 0x0); - - DEB2(printk("iic_init: Initialized IIC on ITE 0x%x\n", - iic_inw(adap, ITE_I2CHSR))); - return 0; -} - - -/* - * Sanity check for the adapter hardware - check the reaction of - * the bus lines only if it seems to be idle. - */ -static int test_bus(struct i2c_algo_iic_data *adap, char *name) { -#if 0 - int scl,sda; - sda=getsda(adap); - if (adap->getscl==NULL) { - printk("test_bus: Warning: Adapter can't read from clock line - skipping test.\n"); - return 0; - } - scl=getscl(adap); - printk("test_bus: Adapter: %s scl: %d sda: %d -- testing...\n", - name,getscl(adap),getsda(adap)); - if (!scl || !sda ) { - printk("test_bus: %s seems to be busy.\n",adap->name); - goto bailout; - } - sdalo(adap); - printk("test_bus:1 scl: %d sda: %d\n", getscl(adap), - getsda(adap)); - if ( 0 != getsda(adap) ) { - printk("test_bus: %s SDA stuck high!\n",name); - sdahi(adap); - goto bailout; - } - if ( 0 == getscl(adap) ) { - printk("test_bus: %s SCL unexpected low while pulling SDA low!\n", - name); - goto bailout; - } - sdahi(adap); - printk("test_bus:2 scl: %d sda: %d\n", getscl(adap), - getsda(adap)); - if ( 0 == getsda(adap) ) { - printk("test_bus: %s SDA stuck low!\n",name); - sdahi(adap); - goto bailout; - } - if ( 0 == getscl(adap) ) { - printk("test_bus: %s SCL unexpected low while SDA high!\n", - adap->name); - goto bailout; - } - scllo(adap); - printk("test_bus:3 scl: %d sda: %d\n", getscl(adap), - getsda(adap)); - if ( 0 != getscl(adap) ) { - - sclhi(adap); - goto bailout; - } - if ( 0 == getsda(adap) ) { - printk("test_bus: %s SDA unexpected low while pulling SCL low!\n", - name); - goto bailout; - } - sclhi(adap); - printk("test_bus:4 scl: %d sda: %d\n", getscl(adap), - getsda(adap)); - if ( 0 == getscl(adap) ) { - printk("test_bus: %s SCL stuck low!\n",name); - sclhi(adap); - goto bailout; - } - if ( 0 == getsda(adap) ) { - printk("test_bus: %s SDA unexpected low while SCL high!\n", - name); - goto bailout; - } - printk("test_bus: %s passed test.\n",name); - return 0; -bailout: - sdahi(adap); - sclhi(adap); - return -ENODEV; -#endif - return (0); -} - -/* ----- Utility functions - */ - - -/* Verify the device we want to talk to on the IIC bus really exists. */ -static inline int try_address(struct i2c_algo_iic_data *adap, - unsigned int addr, int retries) -{ - int i, ret = -1; - short status; - - for (i=0;iudelay); - } - DEB2(if (i) printk("try_address: needed %d retries for 0x%x\n",i, - addr)); - return ret; -} - - -static int iic_sendbytes(struct i2c_adapter *i2c_adap,const char *buf, - int count) -{ - struct i2c_algo_iic_data *adap = i2c_adap->algo_data; - int wrcount=0, timeout; - short status; - int loops, remainder, i, j; - union { - char byte[2]; - unsigned short word; - } tmp; - - iic_outw(adap, ITE_I2CSSAR, (unsigned short)buf[wrcount++]); - count--; - if (count == 0) - return -EIO; - - loops = count / 32; /* 32-byte FIFO */ - remainder = count % 32; - - if(loops) { - for(i=0; iname); - return -EREMOTEIO; /* got a better one ?? */ - } - if (status & ITE_I2CHSR_DB) { - iic_stop(adap); - printk("iic_sendbytes: %s write error - no ack.\n", i2c_adap->name); - return -EREMOTEIO; /* got a better one ?? */ - } - } - } - if(remainder) { - iic_outw(adap, ITE_I2CFBCR, remainder); - for(i=0; iname); - return -EREMOTEIO; /* got a better one ?? */ - } -#ifndef STUB_I2C - if (status & ITE_I2CHSR_DB) { - iic_stop(adap); - printk("iic_sendbytes: %s write error - no ack.\n", i2c_adap->name); - return -EREMOTEIO; /* got a better one ?? */ - } -#endif - } - iic_stop(adap); - return wrcount; -} - - -static int iic_readbytes(struct i2c_adapter *i2c_adap, char *buf, int count, - int sread) -{ - int rdcount=0, i, timeout; - short status; - struct i2c_algo_iic_data *adap = i2c_adap->algo_data; - int loops, remainder, j; - union { - char byte[2]; - unsigned short word; - } tmp; - - loops = count / 32; /* 32-byte FIFO */ - remainder = count % 32; - - if(loops) { - for(i=0; iname); - return (-1); - } -#ifndef STUB_I2C - if (status & ITE_I2CHSR_DB) { - iic_stop(adap); - printk("iic_readbytes: %s read error - no ack.\n", i2c_adap->name); - return (-1); - } -#endif - - timeout = wait_for_fe(adap, &status); - if(timeout) { - iic_stop(adap); - printk("iic_readbytes: %s FIFO is empty\n", i2c_adap->name); - return (-1); - } - - for(j=0; j<32/2; j++) { - tmp.word = iic_inw(adap, ITE_I2CFDR); - buf[rdcount++] = tmp.byte[1]; - buf[rdcount++] = tmp.byte[0]; - } - - /* status FIFO underrun */ - iic_inw(adap, ITE_I2CFSR); - - } - } - - - if(remainder) { - remainder=(remainder+1)/2 * 2; - iic_outw(adap, ITE_I2CFBCR, remainder); - if (sread) - iic_outw(adap, ITE_I2CHCR, ITE_SREAD); - else - iic_outw(adap, ITE_I2CHCR, ITE_READ); /* Issue READ command */ - - timeout = wait_for_pin(adap, &status); - if(timeout) { - iic_stop(adap); - printk("iic_readbytes: %s read timeout.\n", i2c_adap->name); - return (-1); - } -#ifndef STUB_I2C - if (status & ITE_I2CHSR_DB) { - iic_stop(adap); - printk("iic_readbytes: %s read error - no ack.\n", i2c_adap->name); - return (-1); - } -#endif - timeout = wait_for_fe(adap, &status); - if(timeout) { - iic_stop(adap); - printk("iic_readbytes: %s FIFO is empty\n", i2c_adap->name); - return (-1); - } - - for(i=0; i<(remainder+1)/2; i++) { - tmp.word = iic_inw(adap, ITE_I2CFDR); - buf[rdcount++] = tmp.byte[1]; - buf[rdcount++] = tmp.byte[0]; - } - - /* status FIFO underrun */ - iic_inw(adap, ITE_I2CFSR); - - } - - iic_stop(adap); - return rdcount; -} - - -/* This function implements combined transactions. Combined - * transactions consist of combinations of reading and writing blocks of data. - * Each transfer (i.e. a read or a write) is separated by a repeated start - * condition. - */ -#if 0 -static int iic_combined_transaction(struct i2c_adapter *i2c_adap, struct i2c_msg *msgs, int num) -{ - int i; - struct i2c_msg *pmsg; - int ret; - - DEB2(printk("Beginning combined transaction\n")); - - for(i=0; i<(num-1); i++) { - pmsg = &msgs[i]; - if(pmsg->flags & I2C_M_RD) { - DEB2(printk(" This one is a read\n")); - ret = iic_readbytes(i2c_adap, pmsg->buf, pmsg->len, IIC_COMBINED_XFER); - } - else if(!(pmsg->flags & I2C_M_RD)) { - DEB2(printk("This one is a write\n")); - ret = iic_sendbytes(i2c_adap, pmsg->buf, pmsg->len, IIC_COMBINED_XFER); - } - } - /* Last read or write segment needs to be terminated with a stop */ - pmsg = &msgs[i]; - - if(pmsg->flags & I2C_M_RD) { - DEB2(printk("Doing the last read\n")); - ret = iic_readbytes(i2c_adap, pmsg->buf, pmsg->len, IIC_SINGLE_XFER); - } - else if(!(pmsg->flags & I2C_M_RD)) { - DEB2(printk("Doing the last write\n")); - ret = iic_sendbytes(i2c_adap, pmsg->buf, pmsg->len, IIC_SINGLE_XFER); - } - - return ret; -} -#endif - - -/* Whenever we initiate a transaction, the first byte clocked - * onto the bus after the start condition is the address (7 bit) of the - * device we want to talk to. This function manipulates the address specified - * so that it makes sense to the hardware when written to the IIC peripheral. - * - * Note: 10 bit addresses are not supported in this driver, although they are - * supported by the hardware. This functionality needs to be implemented. - */ -static inline int iic_doAddress(struct i2c_algo_iic_data *adap, - struct i2c_msg *msg, int retries) -{ - unsigned short flags = msg->flags; - unsigned int addr; - int ret; - -/* Ten bit addresses not supported right now */ - if ( (flags & I2C_M_TEN) ) { -#if 0 - addr = 0xf0 | (( msg->addr >> 7) & 0x03); - DEB2(printk("addr0: %d\n",addr)); - ret = try_address(adap, addr, retries); - if (ret!=1) { - printk("iic_doAddress: died at extended address code.\n"); - return -EREMOTEIO; - } - iic_outw(adap,msg->addr & 0x7f); - if (ret != 1) { - printk("iic_doAddress: died at 2nd address code.\n"); - return -EREMOTEIO; - } - if ( flags & I2C_M_RD ) { - i2c_repstart(adap); - addr |= 0x01; - ret = try_address(adap, addr, retries); - if (ret!=1) { - printk("iic_doAddress: died at extended address code.\n"); - return -EREMOTEIO; - } - } -#endif - } else { - - addr = ( msg->addr << 1 ); - -#if 0 - if (flags & I2C_M_RD ) - addr |= 1; - if (flags & I2C_M_REV_DIR_ADDR ) - addr ^= 1; -#endif - - if (iic_inw(adap, ITE_I2CSAR) != addr) { - iic_outw(adap, ITE_I2CSAR, addr); - ret = try_address(adap, addr, retries); - if (ret!=1) { - printk("iic_doAddress: died at address code.\n"); - return -EREMOTEIO; - } - } - - } - - return 0; -} - - -/* Description: Prepares the controller for a transaction (clearing status - * registers, data buffers, etc), and then calls either iic_readbytes or - * iic_sendbytes to do the actual transaction. - * - * still to be done: Before we issue a transaction, we should - * verify that the bus is not busy or in some unknown state. - */ -static int iic_xfer(struct i2c_adapter *i2c_adap, - struct i2c_msg *msgs, - int num) -{ - struct i2c_algo_iic_data *adap = i2c_adap->algo_data; - struct i2c_msg *pmsg; - int i = 0; - int ret, timeout; - - pmsg = &msgs[i]; - - if(!pmsg->len) { - DEB2(printk("iic_xfer: read/write length is 0\n");) - return -EIO; - } - if(!(pmsg->flags & I2C_M_RD) && (!(pmsg->len)%2) ) { - DEB2(printk("iic_xfer: write buffer length is not odd\n");) - return -EIO; - } - - /* Wait for any pending transfers to complete */ - timeout = wait_for_bb(adap); - if (timeout) { - DEB2(printk("iic_xfer: Timeout waiting for host not busy\n");) - return -EIO; - } - - /* Flush FIFO */ - iic_outw(adap, ITE_I2CFCR, ITE_I2CFCR_FLUSH); - - /* Load address */ - ret = iic_doAddress(adap, pmsg, i2c_adap->retries); - if (ret) - return -EIO; - -#if 0 - /* Combined transaction (read and write) */ - if(num > 1) { - DEB2(printk("iic_xfer: Call combined transaction\n")); - ret = iic_combined_transaction(i2c_adap, msgs, num); - } -#endif - - DEB3(printk("iic_xfer: Msg %d, addr=0x%x, flags=0x%x, len=%d\n", - i, msgs[i].addr, msgs[i].flags, msgs[i].len);) - - if(pmsg->flags & I2C_M_RD) /* Read */ - ret = iic_readbytes(i2c_adap, pmsg->buf, pmsg->len, 0); - else { /* Write */ - udelay(1000); - ret = iic_sendbytes(i2c_adap, pmsg->buf, pmsg->len); - } - - if (ret != pmsg->len) - DEB3(printk("iic_xfer: error or fail on read/write %d bytes.\n",ret)); - else - DEB3(printk("iic_xfer: read/write %d bytes.\n",ret)); - - return ret; -} - - -/* Implements device specific ioctls. Higher level ioctls can - * be found in i2c-core.c and are typical of any i2c controller (specifying - * slave address, timeouts, etc). These ioctls take advantage of any hardware - * features built into the controller for which this algorithm-adapter set - * was written. These ioctls allow you to take control of the data and clock - * lines and set the either high or low, - * similar to a GPIO pin. - */ -static int algo_control(struct i2c_adapter *adapter, - unsigned int cmd, unsigned long arg) -{ - - struct i2c_algo_iic_data *adap = adapter->algo_data; - struct i2c_iic_msg s_msg; - char *buf; - int ret; - - if (cmd == I2C_SREAD) { - if(copy_from_user(&s_msg, (struct i2c_iic_msg *)arg, - sizeof(struct i2c_iic_msg))) - return -EFAULT; - buf = kmalloc(s_msg.len, GFP_KERNEL); - if (buf== NULL) - return -ENOMEM; - - /* Flush FIFO */ - iic_outw(adap, ITE_I2CFCR, ITE_I2CFCR_FLUSH); - - /* Load address */ - iic_outw(adap, ITE_I2CSAR,s_msg.addr<<1); - iic_outw(adap, ITE_I2CSSAR,s_msg.waddr & 0xff); - - ret = iic_readbytes(adapter, buf, s_msg.len, 1); - if (ret>=0) { - if(copy_to_user( s_msg.buf, buf, s_msg.len) ) - ret = -EFAULT; - } - kfree(buf); - } - return 0; -} - - -static u32 iic_func(struct i2c_adapter *adap) -{ - return I2C_FUNC_SMBUS_EMUL | I2C_FUNC_10BIT_ADDR | - I2C_FUNC_PROTOCOL_MANGLING; -} - -/* -----exported algorithm data: ------------------------------------- */ - -static struct i2c_algorithm iic_algo = { - .master_xfer = iic_xfer, - .algo_control = algo_control, /* ioctl */ - .functionality = iic_func, -}; - - -/* - * registering functions to load algorithms at runtime - */ -int i2c_iic_add_bus(struct i2c_adapter *adap) -{ - struct i2c_algo_iic_data *iic_adap = adap->algo_data; - - if (iic_test) { - int ret = test_bus(iic_adap, adap->name); - if (ret<0) - return -ENODEV; - } - - DEB2(printk("i2c-algo-ite: hw routines for %s registered.\n", - adap->name)); - - /* register new adapter to i2c module... */ - adap->algo = &iic_algo; - - adap->timeout = 100; /* default values, should */ - adap->retries = 3; /* be replaced by defines */ - adap->flags = 0; - - iic_init(iic_adap); - return i2c_add_adapter(adap); -} - - -int i2c_iic_del_bus(struct i2c_adapter *adap) -{ - int res; - if ((res = i2c_del_adapter(adap)) < 0) - return res; - DEB2(printk("i2c-algo-ite: adapter unregistered: %s\n",adap->name)); - - return 0; -} - - -int __init i2c_algo_iic_init (void) -{ - printk(KERN_INFO "ITE iic (i2c) algorithm module\n"); - return 0; -} - - -void i2c_algo_iic_exit(void) -{ - return; -} - - -EXPORT_SYMBOL(i2c_iic_add_bus); -EXPORT_SYMBOL(i2c_iic_del_bus); - -/* The MODULE_* macros resolve to nothing if MODULES is not defined - * when this file is compiled. - */ -MODULE_AUTHOR("MontaVista Software "); -MODULE_DESCRIPTION("ITE iic algorithm"); -MODULE_LICENSE("GPL"); - -module_param(iic_test, bool, 0); -module_param(i2c_debug, int, S_IRUGO | S_IWUSR); - -MODULE_PARM_DESC(iic_test, "Test if the I2C bus is available"); -MODULE_PARM_DESC(i2c_debug, - "debug level - 0 off; 1 normal; 2,3 more verbose; 9 iic-protocol"); - - -/* This function resolves to init_module (the function invoked when a module - * is loaded via insmod) when this file is compiled with MODULES defined. - * Otherwise (i.e. if you want this driver statically linked to the kernel), - * a pointer to this function is stored in a table and called - * during the initialization of the kernel (in do_basic_setup in /init/main.c) - * - * All this functionality is complements of the macros defined in linux/init.h - */ -module_init(i2c_algo_iic_init); - - -/* If MODULES is defined when this file is compiled, then this function will - * resolved to cleanup_module. - */ -module_exit(i2c_algo_iic_exit); diff --git a/drivers/i2c/algos/i2c-algo-ite.h b/drivers/i2c/algos/i2c-algo-ite.h deleted file mode 100644 index a8ca3c9b546a..000000000000 --- a/drivers/i2c/algos/i2c-algo-ite.h +++ /dev/null @@ -1,117 +0,0 @@ -/* - -------------------------------------------------------------------- - i2c-ite.h: Global defines for the I2C controller on board the - ITE MIPS processor. - -------------------------------------------------------------------- - Hai-Pao Fan, MontaVista Software, Inc. - hpfan@mvista.com or source@mvista.com - - Copyright 2001 MontaVista Software Inc. - - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; either version 2 of the License, or (at your - * option) any later version. - * - * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED - * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF - * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN - * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT - * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF - * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON - * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF - * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 675 Mass Ave, Cambridge, MA 02139, USA. - - */ - -#ifndef I2C_ITE_H -#define I2C_ITE_H 1 - -#include - -/* I2C Registers */ -#define ITE_I2CHCR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x30 -#define ITE_I2CHSR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x34 -#define ITE_I2CSAR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x38 -#define ITE_I2CSSAR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x3c -#define ITE_I2CCKCNT IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x48 -#define ITE_I2CSHDR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x4c -#define ITE_I2CRSUR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x50 -#define ITE_I2CPSUR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x54 - -#define ITE_I2CFDR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x70 -#define ITE_I2CFBCR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x74 -#define ITE_I2CFCR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x78 -#define ITE_I2CFSR IT8172_PCI_IO_BASE + IT_I2C_BASE + 0x7c - - -/* Host Control Register ITE_I2CHCR */ -#define ITE_I2CHCR_HCE 0x01 /* Enable I2C Host Controller */ -#define ITE_I2CHCR_IE 0x02 /* Enable the interrupt after completing - the current transaction */ -#define ITE_I2CHCR_CP_W 0x00 /* bit2-4 000 - Write */ -#define ITE_I2CHCR_CP_R 0x08 /* 010 - Current address read */ -#define ITE_I2CHCR_CP_S 0x10 /* 100 - Sequential read */ -#define ITE_I2CHCR_ST 0x20 /* Initiates the I2C host controller to execute - the command and send the data programmed in - all required registers to I2C bus */ -#define ITE_CMD ITE_I2CHCR_HCE | ITE_I2CHCR_IE | ITE_I2CHCR_ST -#define ITE_WRITE ITE_CMD | ITE_I2CHCR_CP_W -#define ITE_READ ITE_CMD | ITE_I2CHCR_CP_R -#define ITE_SREAD ITE_CMD | ITE_I2CHCR_CP_S - -/* Host Status Register ITE_I2CHSR */ -#define ITE_I2CHSR_DB 0x01 /* Device is busy, receives NACK response except - in the first and last bytes */ -#define ITE_I2CHSR_DNE 0x02 /* Target address on I2C bus does not exist */ -#define ITE_I2CHSR_TDI 0x04 /* R/W Transaction on I2C bus was completed */ -#define ITE_I2CHSR_HB 0x08 /* Host controller is processing transactions */ -#define ITE_I2CHSR_FER 0x10 /* Error occurs in the FIFO */ - -/* Slave Address Register ITE_I2CSAR */ -#define ITE_I2CSAR_SA_MASK 0xfe /* Target I2C device address */ -#define ITE_I2CSAR_ASO 0x0100 /* Output 1/0 to I2CAS port when the - next slave address is addressed */ - -/* Slave Sub-address Register ITE_I2CSSAR */ -#define ITE_I2CSSAR_SUBA_MASK 0xff /* Target I2C device sub-address */ - -/* Clock Counter Register ITE_I2CCKCNT */ -#define ITE_I2CCKCNT_STOP 0x00 /* stop I2C clock */ -#define ITE_I2CCKCNT_HPCC_MASK 0x7f /* SCL high period counter */ -#define ITE_I2CCKCNT_LPCC_MASK 0x7f00 /* SCL low period counter */ - -/* START Hold Time Register ITE_I2CSHDR */ -/* value is counted based on 16 MHz internal clock */ -#define ITE_I2CSHDR_FM 0x0a /* START condition at fast mode */ -#define ITE_I2CSHDR_SM 0x47 /* START contition at standard mode */ - -/* (Repeated) START Setup Time Register ITE_I2CRSUR */ -/* value is counted based on 16 MHz internal clock */ -#define ITE_I2CRSUR_FM 0x0a /* repeated START condition at fast mode */ -#define ITE_I2CRSUR_SM 0x50 /* repeated START condition at standard mode */ - -/* STOP setup Time Register ITE_I2CPSUR */ - -/* FIFO Data Register ITE_I2CFDR */ -#define ITE_I2CFDR_MASK 0xff - -/* FIFO Byte Count Register ITE_I2CFBCR */ -#define ITE_I2CFBCR_MASK 0x3f - -/* FIFO Control Register ITE_I2CFCR */ -#define ITE_I2CFCR_FLUSH 0x01 /* Flush FIFO and reset the FIFO point - and I2CFSR */ -/* FIFO Status Register ITE_I2CFSR */ -#define ITE_I2CFSR_FO 0x01 /* FIFO is overrun when write */ -#define ITE_I2CFSR_FU 0x02 /* FIFO is underrun when read */ -#define ITE_I2CFSR_FF 0x04 /* FIFO is full when write */ -#define ITE_I2CFSR_FE 0x08 /* FIFO is empty when read */ - -#endif /* I2C_ITE_H */ diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index 90f91d039ee2..7c9ab337d180 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -209,18 +209,6 @@ config I2C_ISA tristate depends on I2C -config I2C_ITE - tristate "ITE I2C Adapter" - depends on I2C && MIPS_ITE8172 - select I2C_ALGOITE - help - This supports the ITE8172 I2C peripheral found on some MIPS - systems. Say Y if you have one of these. You should also say Y for - the ITE I2C driver algorithm support above. - - This support is also available as a module. If so, the module - will be called i2c-ite. - config I2C_IXP4XX tristate "IXP4xx GPIO-Based I2C Interface" depends on I2C && ARCH_IXP4XX diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile index 493c87289b62..bf5fd078b292 100644 --- a/drivers/i2c/busses/Makefile +++ b/drivers/i2c/busses/Makefile @@ -16,7 +16,6 @@ obj-$(CONFIG_I2C_I810) += i2c-i810.o obj-$(CONFIG_I2C_IBM_IIC) += i2c-ibm_iic.o obj-$(CONFIG_I2C_IOP3XX) += i2c-iop3xx.o obj-$(CONFIG_I2C_ISA) += i2c-isa.o -obj-$(CONFIG_I2C_ITE) += i2c-ite.o obj-$(CONFIG_I2C_IXP2000) += i2c-ixp2000.o obj-$(CONFIG_I2C_IXP4XX) += i2c-ixp4xx.o obj-$(CONFIG_I2C_POWERMAC) += i2c-powermac.o diff --git a/drivers/i2c/busses/i2c-ite.c b/drivers/i2c/busses/i2c-ite.c deleted file mode 100644 index f7d71869b3b9..000000000000 --- a/drivers/i2c/busses/i2c-ite.c +++ /dev/null @@ -1,278 +0,0 @@ -/* - ------------------------------------------------------------------------- - i2c-adap-ite.c i2c-hw access for the IIC peripheral on the ITE MIPS system - ------------------------------------------------------------------------- - Hai-Pao Fan, MontaVista Software, Inc. - hpfan@mvista.com or source@mvista.com - - Copyright 2001 MontaVista Software Inc. - - ---------------------------------------------------------------------------- - This file was highly leveraged from i2c-elektor.c, which was created - by Simon G. Vogl and Hans Berglund: - - - Copyright (C) 1995-97 Simon G. Vogl - 1998-99 Hans Berglund - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -/* ------------------------------------------------------------------------- */ - -/* With some changes from Kyösti Mälkki and even - Frodo Looijaard */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include "../i2c-ite.h" - -#define DEFAULT_BASE 0x14014030 -#define ITE_IIC_IO_SIZE 0x40 -#define DEFAULT_IRQ 0 -#define DEFAULT_CLOCK 0x1b0e /* default 16MHz/(27+14) = 400KHz */ -#define DEFAULT_OWN 0x55 - -static int base; -static int irq; -static int clock; -static int own; - -static struct iic_ite gpi; -static wait_queue_head_t iic_wait; -static int iic_pending; -static spinlock_t lock; - -/* ----- local functions ---------------------------------------------- */ - -static void iic_ite_setiic(void *data, int ctl, short val) -{ - unsigned long j = jiffies + 10; - - pr_debug(" Write 0x%02x to 0x%x\n",(unsigned short)val, ctl&0xff); -#ifdef DEBUG - while (time_before(jiffies, j)) - schedule(); -#endif - outw(val,ctl); -} - -static short iic_ite_getiic(void *data, int ctl) -{ - short val; - - val = inw(ctl); - pr_debug("Read 0x%02x from 0x%x\n",(unsigned short)val, ctl&0xff); - return (val); -} - -/* Return our slave address. This is the address - * put on the I2C bus when another master on the bus wants to address us - * as a slave - */ -static int iic_ite_getown(void *data) -{ - return (gpi.iic_own); -} - - -static int iic_ite_getclock(void *data) -{ - return (gpi.iic_clock); -} - - -/* Put this process to sleep. We will wake up when the - * IIC controller interrupts. - */ -static void iic_ite_waitforpin(void) { - DEFINE_WAIT(wait); - int timeout = 2; - unsigned long flags; - - /* If interrupts are enabled (which they are), then put the process to - * sleep. This process will be awakened by two events -- either the - * the IIC peripheral interrupts or the timeout expires. - * If interrupts are not enabled then delay for a reasonable amount - * of time and return. - */ - if (gpi.iic_irq > 0) { - spin_lock_irqsave(&lock, flags); - if (iic_pending == 0) { - spin_unlock_irqrestore(&lock, flags); - prepare_to_wait(&iic_wait, &wait, TASK_INTERRUPTIBLE); - if (schedule_timeout(timeout*HZ)) { - spin_lock_irqsave(&lock, flags); - if (iic_pending == 1) { - iic_pending = 0; - } - spin_unlock_irqrestore(&lock, flags); - } - finish_wait(&iic_wait, &wait); - } else { - iic_pending = 0; - spin_unlock_irqrestore(&lock, flags); - } - } else { - udelay(100); - } -} - - -static irqreturn_t iic_ite_handler(int this_irq, void *dev_id) -{ - spin_lock(&lock); - iic_pending = 1; - spin_unlock(&lock); - - wake_up_interruptible(&iic_wait); - - return IRQ_HANDLED; -} - - -/* Lock the region of memory where I/O registers exist. Request our - * interrupt line and register its associated handler. - */ -static int iic_hw_resrc_init(void) -{ - if (!request_region(gpi.iic_base, ITE_IIC_IO_SIZE, "i2c")) - return -ENODEV; - - if (gpi.iic_irq <= 0) - return 0; - - if (request_irq(gpi.iic_irq, iic_ite_handler, 0, "ITE IIC", 0) < 0) - gpi.iic_irq = 0; - else - enable_irq(gpi.iic_irq); - - return 0; -} - - -static void iic_ite_release(void) -{ - if (gpi.iic_irq > 0) { - disable_irq(gpi.iic_irq); - free_irq(gpi.iic_irq, 0); - } - release_region(gpi.iic_base , 2); -} - -/* ------------------------------------------------------------------------ - * Encapsulate the above functions in the correct operations structure. - * This is only done when more than one hardware adapter is supported. - */ -static struct i2c_algo_iic_data iic_ite_data = { - NULL, - iic_ite_setiic, - iic_ite_getiic, - iic_ite_getown, - iic_ite_getclock, - iic_ite_waitforpin, - 80, 80, 100, /* waits, timeout */ -}; - -static struct i2c_adapter iic_ite_ops = { - .owner = THIS_MODULE, - .id = I2C_HW_I_IIC, - .algo_data = &iic_ite_data, - .name = "ITE IIC adapter", -}; - -/* Called when the module is loaded. This function starts the - * cascade of calls up through the hierarchy of i2c modules (i.e. up to the - * algorithm layer and into to the core layer) - */ -static int __init iic_ite_init(void) -{ - - struct iic_ite *piic = &gpi; - - printk(KERN_INFO "Initialize ITE IIC adapter module\n"); - if (base == 0) - piic->iic_base = DEFAULT_BASE; - else - piic->iic_base = base; - - if (irq == 0) - piic->iic_irq = DEFAULT_IRQ; - else - piic->iic_irq = irq; - - if (clock == 0) - piic->iic_clock = DEFAULT_CLOCK; - else - piic->iic_clock = clock; - - if (own == 0) - piic->iic_own = DEFAULT_OWN; - else - piic->iic_own = own; - - iic_ite_data.data = (void *)piic; - init_waitqueue_head(&iic_wait); - spin_lock_init(&lock); - if (iic_hw_resrc_init() == 0) { - if (i2c_iic_add_bus(&iic_ite_ops) < 0) - return -ENODEV; - } else { - return -ENODEV; - } - printk(KERN_INFO " found device at %#x irq %d.\n", - piic->iic_base, piic->iic_irq); - return 0; -} - - -static void iic_ite_exit(void) -{ - i2c_iic_del_bus(&iic_ite_ops); - iic_ite_release(); -} - -/* If modules is NOT defined when this file is compiled, then the MODULE_* - * macros will resolve to nothing - */ -MODULE_AUTHOR("MontaVista Software "); -MODULE_DESCRIPTION("I2C-Bus adapter routines for ITE IIC bus adapter"); -MODULE_LICENSE("GPL"); - -module_param(base, int, 0); -module_param(irq, int, 0); -module_param(clock, int, 0); -module_param(own, int, 0); - - -/* Called when module is loaded or when kernel is initialized. - * If MODULES is defined when this file is compiled, then this function will - * resolve to init_module (the function called when insmod is invoked for a - * module). Otherwise, this function is called early in the boot, when the - * kernel is intialized. Check out /include/init.h to see how this works. - */ -module_init(iic_ite_init); - -/* Resolves to module_cleanup when MODULES is defined. */ -module_exit(iic_ite_exit); diff --git a/include/linux/i2c-algo-ite.h b/include/linux/i2c-algo-ite.h deleted file mode 100644 index 0073fe96c76e..000000000000 --- a/include/linux/i2c-algo-ite.h +++ /dev/null @@ -1,72 +0,0 @@ -/* ------------------------------------------------------------------------- */ -/* i2c-algo-ite.h i2c driver algorithms for ITE IIC adapters */ -/* ------------------------------------------------------------------------- */ -/* Copyright (C) 1995-97 Simon G. Vogl - 1998-99 Hans Berglund - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -/* ------------------------------------------------------------------------- */ - -/* With some changes from Kyösti Mälkki and even - Frodo Looijaard */ - -/* Modifications by MontaVista Software, 2001 - Changes made to support the ITE IIC peripheral */ - - -#ifndef I2C_ALGO_ITE_H -#define I2C_ALGO_ITE_H 1 - -#include - -/* Example of a sequential read request: - struct i2c_iic_msg s_msg; - - s_msg.addr=device_address; - s_msg.len=length; - s_msg.buf=buffer; - s_msg.waddr=word_address; - ioctl(file,I2C_SREAD, &s_msg); - */ -#define I2C_SREAD 0x780 /* SREAD ioctl command */ - -struct i2c_iic_msg { - __u16 addr; /* device address */ - __u16 waddr; /* word address */ - short len; /* msg length */ - char *buf; /* pointer to msg data */ -}; - -#ifdef __KERNEL__ -struct i2c_adapter; - -struct i2c_algo_iic_data { - void *data; /* private data for lolevel routines */ - void (*setiic) (void *data, int ctl, int val); - int (*getiic) (void *data, int ctl); - int (*getown) (void *data); - int (*getclock) (void *data); - void (*waitforpin) (void); - - /* local settings */ - int udelay; - int mdelay; - int timeout; -}; - -int i2c_iic_add_bus(struct i2c_adapter *); -int i2c_iic_del_bus(struct i2c_adapter *); -#endif /* __KERNEL__ */ -#endif /* I2C_ALGO_ITE_H */ diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 36f0c5b0cbf7..00da370bbcd6 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -210,9 +210,6 @@ /* --- MPC8xx PowerPC adapters */ #define I2C_HW_MPC8XX_EPON 0x110000 /* Eponymous MPC8xx I2C adapter */ -/* --- ITE based algorithms */ -#define I2C_HW_I_IIC 0x080000 /* controller on the ITE */ - /* --- PowerPC on-chip adapters */ #define I2C_HW_OCP 0x120000 /* IBM on-chip I2C adapter */ -- cgit v1.2.3 From 41561f28e76a47dc6de0a954da85d0b5c42874eb Mon Sep 17 00:00:00 2001 From: Vitaly Wool Date: Sun, 10 Dec 2006 21:21:29 +0100 Subject: i2c: New Philips PNX bus driver New I2C bus driver for Philips ARM boards (Philips IP3204 I2C IP block). This I2C controller can be found on (at least) PNX010x, PNX52xx and PNX4008 Philips boards. Signed-off-by: Vitaly Wool Signed-off-by: Jean Delvare --- arch/arm/mach-pnx4008/Makefile | 2 +- arch/arm/mach-pnx4008/i2c.c | 167 +++++++++ drivers/i2c/busses/Kconfig | 19 + drivers/i2c/busses/Makefile | 1 + drivers/i2c/busses/i2c-pnx.c | 708 +++++++++++++++++++++++++++++++++++++ include/asm-arm/arch-pnx4008/i2c.h | 67 ++++ include/linux/i2c-pnx.h | 43 +++ 7 files changed, 1006 insertions(+), 1 deletion(-) create mode 100644 arch/arm/mach-pnx4008/i2c.c create mode 100644 drivers/i2c/busses/i2c-pnx.c create mode 100644 include/asm-arm/arch-pnx4008/i2c.h create mode 100644 include/linux/i2c-pnx.h (limited to 'include/linux') diff --git a/arch/arm/mach-pnx4008/Makefile b/arch/arm/mach-pnx4008/Makefile index b457ca0a431a..777564c90a12 100644 --- a/arch/arm/mach-pnx4008/Makefile +++ b/arch/arm/mach-pnx4008/Makefile @@ -2,7 +2,7 @@ # Makefile for the linux kernel. # -obj-y := core.o irq.o time.o clock.o gpio.o serial.o dma.o +obj-y := core.o irq.o time.o clock.o gpio.o serial.o dma.o i2c.o obj-m := obj-n := obj- := diff --git a/arch/arm/mach-pnx4008/i2c.c b/arch/arm/mach-pnx4008/i2c.c new file mode 100644 index 000000000000..6f308827c4fe --- /dev/null +++ b/arch/arm/mach-pnx4008/i2c.c @@ -0,0 +1,167 @@ +/* + * I2C initialization for PNX4008. + * + * Author: Vitaly Wool + * + * 2005-2006 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ + +#include +#include +#include +#include +#include +#include +#include + +static int set_clock_run(struct platform_device *pdev) +{ + struct clk *clk; + char name[10]; + int retval = 0; + + snprintf(name, 10, "i2c%d_ck", pdev->id); + clk = clk_get(&pdev->dev, name); + if (!IS_ERR(clk)) { + clk_set_rate(clk, 1); + clk_put(clk); + } else + retval = -ENOENT; + + return retval; +} + +static int set_clock_stop(struct platform_device *pdev) +{ + struct clk *clk; + char name[10]; + int retval = 0; + + snprintf(name, 10, "i2c%d_ck", pdev->id); + clk = clk_get(&pdev->dev, name); + if (!IS_ERR(clk)) { + clk_set_rate(clk, 0); + clk_put(clk); + } else + retval = -ENOENT; + + return retval; +} + +static int i2c_pnx_suspend(struct platform_device *pdev, pm_message_t state) +{ + int retval = 0; +#ifdef CONFIG_PM + retval = set_clock_run(pdev); +#endif + return retval; +} + +static int i2c_pnx_resume(struct platform_device *pdev) +{ + int retval = 0; +#ifdef CONFIG_PM + retval = set_clock_run(pdev); +#endif + return retval; +} + +static u32 calculate_input_freq(struct platform_device *pdev) +{ + return HCLK_MHZ; +} + + +static struct i2c_pnx_algo_data pnx_algo_data0 = { + .base = PNX4008_I2C1_BASE, + .irq = I2C_1_INT, +}; + +static struct i2c_pnx_algo_data pnx_algo_data1 = { + .base = PNX4008_I2C2_BASE, + .irq = I2C_2_INT, +}; + +static struct i2c_pnx_algo_data pnx_algo_data2 = { + .base = (PNX4008_USB_CONFIG_BASE + 0x300), + .irq = USB_I2C_INT, +}; + +static struct i2c_adapter pnx_adapter0 = { + .name = I2C_CHIP_NAME "0", + .algo_data = &pnx_algo_data0, +}; +static struct i2c_adapter pnx_adapter1 = { + .name = I2C_CHIP_NAME "1", + .algo_data = &pnx_algo_data1, +}; + +static struct i2c_adapter pnx_adapter2 = { + .name = "USB-I2C", + .algo_data = &pnx_algo_data2, +}; + +static struct i2c_pnx_data i2c0_data = { + .suspend = i2c_pnx_suspend, + .resume = i2c_pnx_resume, + .calculate_input_freq = calculate_input_freq, + .set_clock_run = set_clock_run, + .set_clock_stop = set_clock_stop, + .adapter = &pnx_adapter0, +}; + +static struct i2c_pnx_data i2c1_data = { + .suspend = i2c_pnx_suspend, + .resume = i2c_pnx_resume, + .calculate_input_freq = calculate_input_freq, + .set_clock_run = set_clock_run, + .set_clock_stop = set_clock_stop, + .adapter = &pnx_adapter1, +}; + +static struct i2c_pnx_data i2c2_data = { + .suspend = i2c_pnx_suspend, + .resume = i2c_pnx_resume, + .calculate_input_freq = calculate_input_freq, + .set_clock_run = set_clock_run, + .set_clock_stop = set_clock_stop, + .adapter = &pnx_adapter2, +}; + +static struct platform_device i2c0_device = { + .name = "pnx-i2c", + .id = 0, + .dev = { + .platform_data = &i2c0_data, + }, +}; + +static struct platform_device i2c1_device = { + .name = "pnx-i2c", + .id = 1, + .dev = { + .platform_data = &i2c1_data, + }, +}; + +static struct platform_device i2c2_device = { + .name = "pnx-i2c", + .id = 2, + .dev = { + .platform_data = &i2c2_data, + }, +}; + +static struct platform_device *devices[] __initdata = { + &i2c0_device, + &i2c1_device, + &i2c2_device, +}; + +void __init pnx4008_register_i2c_devices(void) +{ + platform_add_devices(devices, ARRAY_SIZE(devices)); +} diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index 7c9ab337d180..291da25e84d1 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -536,4 +536,23 @@ config I2C_MV64XXX This driver can also be built as a module. If so, the module will be called i2c-mv64xxx. +config I2C_PNX + tristate "I2C bus support for Philips PNX targets" + depends on ARCH_PNX4008 && I2C + help + This driver supports the Philips IP3204 I2C IP block master and/or + slave controller + + This driver can also be built as a module. If so, the module + will be called i2c-pnx. + +config I2C_PNX_EARLY + bool "Early initialization for I2C on PNXxxxx" + depends on I2C_PNX=y + help + Under certain circumstances one may need to make sure I2C on PNXxxxx + is initialized earlier than some other driver that depends on it + (for instance, that might be USB in case of PNX4008). With this + option turned on you can guarantee that. + endmenu diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile index bf5fd078b292..626ac71af959 100644 --- a/drivers/i2c/busses/Makefile +++ b/drivers/i2c/busses/Makefile @@ -28,6 +28,7 @@ obj-$(CONFIG_I2C_PARPORT) += i2c-parport.o obj-$(CONFIG_I2C_PARPORT_LIGHT) += i2c-parport-light.o obj-$(CONFIG_I2C_PCA_ISA) += i2c-pca-isa.o obj-$(CONFIG_I2C_PIIX4) += i2c-piix4.o +obj-$(CONFIG_I2C_PNX) += i2c-pnx.o obj-$(CONFIG_I2C_PROSAVAGE) += i2c-prosavage.o obj-$(CONFIG_I2C_PXA) += i2c-pxa.o obj-$(CONFIG_I2C_RPXLITE) += i2c-rpx.o diff --git a/drivers/i2c/busses/i2c-pnx.c b/drivers/i2c/busses/i2c-pnx.c new file mode 100644 index 000000000000..de0bca77e926 --- /dev/null +++ b/drivers/i2c/busses/i2c-pnx.c @@ -0,0 +1,708 @@ +/* + * Provides I2C support for Philips PNX010x/PNX4008 boards. + * + * Authors: Dennis Kovalev + * Vitaly Wool + * + * 2004-2006 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define I2C_PNX_TIMEOUT 10 /* msec */ +#define I2C_PNX_SPEED_KHZ 100 +#define I2C_PNX_REGION_SIZE 0x100 +#define PNX_DEFAULT_FREQ 13 /* MHz */ + +static inline int wait_timeout(long timeout, struct i2c_pnx_algo_data *data) +{ + while (timeout > 0 && + (ioread32(I2C_REG_STS(data)) & mstatus_active)) { + mdelay(1); + timeout--; + } + return (timeout <= 0); +} + +static inline int wait_reset(long timeout, struct i2c_pnx_algo_data *data) +{ + while (timeout > 0 && + (ioread32(I2C_REG_CTL(data)) & mcntrl_reset)) { + mdelay(1); + timeout--; + } + return (timeout <= 0); +} + +static inline void i2c_pnx_arm_timer(struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *data = adap->algo_data; + struct timer_list *timer = &data->mif.timer; + int expires = I2C_PNX_TIMEOUT / (1000 / HZ); + + del_timer_sync(timer); + + dev_dbg(&adap->dev, "Timer armed at %lu plus %u jiffies.\n", + jiffies, expires); + + timer->expires = jiffies + expires; + timer->data = (unsigned long)adap; + + add_timer(timer); +} + +/** + * i2c_pnx_start - start a device + * @slave_addr: slave address + * @adap: pointer to adapter structure + * + * Generate a START signal in the desired mode. + */ +static int i2c_pnx_start(unsigned char slave_addr, struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + + dev_dbg(&adap->dev, "%s(): addr 0x%x mode %d\n", __FUNCTION__, + slave_addr, alg_data->mif.mode); + + /* Check for 7 bit slave addresses only */ + if (slave_addr & ~0x7f) { + dev_err(&adap->dev, "%s: Invalid slave address %x. " + "Only 7-bit addresses are supported\n", + adap->name, slave_addr); + return -EINVAL; + } + + /* First, make sure bus is idle */ + if (wait_timeout(I2C_PNX_TIMEOUT, alg_data)) { + /* Somebody else is monopolizing the bus */ + dev_err(&adap->dev, "%s: Bus busy. Slave addr = %02x, " + "cntrl = %x, stat = %x\n", + adap->name, slave_addr, + ioread32(I2C_REG_CTL(alg_data)), + ioread32(I2C_REG_STS(alg_data))); + return -EBUSY; + } else if (ioread32(I2C_REG_STS(alg_data)) & mstatus_afi) { + /* Sorry, we lost the bus */ + dev_err(&adap->dev, "%s: Arbitration failure. " + "Slave addr = %02x\n", adap->name, slave_addr); + return -EIO; + } + + /* + * OK, I2C is enabled and we have the bus. + * Clear the current TDI and AFI status flags. + */ + iowrite32(ioread32(I2C_REG_STS(alg_data)) | mstatus_tdi | mstatus_afi, + I2C_REG_STS(alg_data)); + + dev_dbg(&adap->dev, "%s(): sending %#x\n", __FUNCTION__, + (slave_addr << 1) | start_bit | alg_data->mif.mode); + + /* Write the slave address, START bit and R/W bit */ + iowrite32((slave_addr << 1) | start_bit | alg_data->mif.mode, + I2C_REG_TX(alg_data)); + + dev_dbg(&adap->dev, "%s(): exit\n", __FUNCTION__); + + return 0; +} + +/** + * i2c_pnx_stop - stop a device + * @adap: pointer to I2C adapter structure + * + * Generate a STOP signal to terminate the master transaction. + */ +static void i2c_pnx_stop(struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + /* Only 1 msec max timeout due to interrupt context */ + long timeout = 1000; + + dev_dbg(&adap->dev, "%s(): entering: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + /* Write a STOP bit to TX FIFO */ + iowrite32(0xff | stop_bit, I2C_REG_TX(alg_data)); + + /* Wait until the STOP is seen. */ + while (timeout > 0 && + (ioread32(I2C_REG_STS(alg_data)) & mstatus_active)) { + /* may be called from interrupt context */ + udelay(1); + timeout--; + } + + dev_dbg(&adap->dev, "%s(): exiting: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); +} + +/** + * i2c_pnx_master_xmit - transmit data to slave + * @adap: pointer to I2C adapter structure + * + * Sends one byte of data to the slave + */ +static int i2c_pnx_master_xmit(struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + u32 val; + + dev_dbg(&adap->dev, "%s(): entering: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + if (alg_data->mif.len > 0) { + /* We still have something to talk about... */ + val = *alg_data->mif.buf++; + + if (alg_data->mif.len == 1) { + val |= stop_bit; + if (!alg_data->last) + val |= start_bit; + } + + alg_data->mif.len--; + iowrite32(val, I2C_REG_TX(alg_data)); + + dev_dbg(&adap->dev, "%s(): xmit %#x [%d]\n", __FUNCTION__, + val, alg_data->mif.len + 1); + + if (alg_data->mif.len == 0) { + if (alg_data->last) { + /* Wait until the STOP is seen. */ + if (wait_timeout(I2C_PNX_TIMEOUT, alg_data)) + dev_err(&adap->dev, "The bus is still " + "active after timeout\n"); + } + /* Disable master interrupts */ + iowrite32(ioread32(I2C_REG_CTL(alg_data)) & + ~(mcntrl_afie | mcntrl_naie | mcntrl_drmie), + I2C_REG_CTL(alg_data)); + + del_timer_sync(&alg_data->mif.timer); + + dev_dbg(&adap->dev, "%s(): Waking up xfer routine.\n", + __FUNCTION__); + + complete(&alg_data->mif.complete); + } + } else if (alg_data->mif.len == 0) { + /* zero-sized transfer */ + i2c_pnx_stop(adap); + + /* Disable master interrupts. */ + iowrite32(ioread32(I2C_REG_CTL(alg_data)) & + ~(mcntrl_afie | mcntrl_naie | mcntrl_drmie), + I2C_REG_CTL(alg_data)); + + /* Stop timer. */ + del_timer_sync(&alg_data->mif.timer); + dev_dbg(&adap->dev, "%s(): Waking up xfer routine after " + "zero-xfer.\n", __FUNCTION__); + + complete(&alg_data->mif.complete); + } + + dev_dbg(&adap->dev, "%s(): exiting: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + return 0; +} + +/** + * i2c_pnx_master_rcv - receive data from slave + * @adap: pointer to I2C adapter structure + * + * Reads one byte data from the slave + */ +static int i2c_pnx_master_rcv(struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + unsigned int val = 0; + u32 ctl = 0; + + dev_dbg(&adap->dev, "%s(): entering: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + /* Check, whether there is already data, + * or we didn't 'ask' for it yet. + */ + if (ioread32(I2C_REG_STS(alg_data)) & mstatus_rfe) { + dev_dbg(&adap->dev, "%s(): Write dummy data to fill " + "Rx-fifo...\n", __FUNCTION__); + + if (alg_data->mif.len == 1) { + /* Last byte, do not acknowledge next rcv. */ + val |= stop_bit; + if (!alg_data->last) + val |= start_bit; + + /* + * Enable interrupt RFDAIE (data in Rx fifo), + * and disable DRMIE (need data for Tx) + */ + ctl = ioread32(I2C_REG_CTL(alg_data)); + ctl |= mcntrl_rffie | mcntrl_daie; + ctl &= ~mcntrl_drmie; + iowrite32(ctl, I2C_REG_CTL(alg_data)); + } + + /* + * Now we'll 'ask' for data: + * For each byte we want to receive, we must + * write a (dummy) byte to the Tx-FIFO. + */ + iowrite32(val, I2C_REG_TX(alg_data)); + + return 0; + } + + /* Handle data. */ + if (alg_data->mif.len > 0) { + val = ioread32(I2C_REG_RX(alg_data)); + *alg_data->mif.buf++ = (u8) (val & 0xff); + dev_dbg(&adap->dev, "%s(): rcv 0x%x [%d]\n", __FUNCTION__, val, + alg_data->mif.len); + + alg_data->mif.len--; + if (alg_data->mif.len == 0) { + if (alg_data->last) + /* Wait until the STOP is seen. */ + if (wait_timeout(I2C_PNX_TIMEOUT, alg_data)) + dev_err(&adap->dev, "The bus is still " + "active after timeout\n"); + + /* Disable master interrupts */ + ctl = ioread32(I2C_REG_CTL(alg_data)); + ctl &= ~(mcntrl_afie | mcntrl_naie | mcntrl_rffie | + mcntrl_drmie | mcntrl_daie); + iowrite32(ctl, I2C_REG_CTL(alg_data)); + + /* Kill timer. */ + del_timer_sync(&alg_data->mif.timer); + complete(&alg_data->mif.complete); + } + } + + dev_dbg(&adap->dev, "%s(): exiting: stat = %04x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + return 0; +} + +static irqreturn_t +i2c_pnx_interrupt(int irq, void *dev_id, struct pt_regs *regs) +{ + u32 stat, ctl; + struct i2c_adapter *adap = dev_id; + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + + dev_dbg(&adap->dev, "%s(): mstat = %x mctrl = %x, mode = %d\n", + __FUNCTION__, + ioread32(I2C_REG_STS(alg_data)), + ioread32(I2C_REG_CTL(alg_data)), + alg_data->mif.mode); + stat = ioread32(I2C_REG_STS(alg_data)); + + /* let's see what kind of event this is */ + if (stat & mstatus_afi) { + /* We lost arbitration in the midst of a transfer */ + alg_data->mif.ret = -EIO; + + /* Disable master interrupts. */ + ctl = ioread32(I2C_REG_CTL(alg_data)); + ctl &= ~(mcntrl_afie | mcntrl_naie | mcntrl_rffie | + mcntrl_drmie); + iowrite32(ctl, I2C_REG_CTL(alg_data)); + + /* Stop timer, to prevent timeout. */ + del_timer_sync(&alg_data->mif.timer); + complete(&alg_data->mif.complete); + } else if (stat & mstatus_nai) { + /* Slave did not acknowledge, generate a STOP */ + dev_dbg(&adap->dev, "%s(): " + "Slave did not acknowledge, generating a STOP.\n", + __FUNCTION__); + i2c_pnx_stop(adap); + + /* Disable master interrupts. */ + ctl = ioread32(I2C_REG_CTL(alg_data)); + ctl &= ~(mcntrl_afie | mcntrl_naie | mcntrl_rffie | + mcntrl_drmie); + iowrite32(ctl, I2C_REG_CTL(alg_data)); + + /* Our return value. */ + alg_data->mif.ret = -EIO; + + /* Stop timer, to prevent timeout. */ + del_timer_sync(&alg_data->mif.timer); + complete(&alg_data->mif.complete); + } else { + /* + * Two options: + * - Master Tx needs data. + * - There is data in the Rx-fifo + * The latter is only the case if we have requested for data, + * via a dummy write. (See 'i2c_pnx_master_rcv'.) + * We therefore check, as a sanity check, whether that interrupt + * has been enabled. + */ + if ((stat & mstatus_drmi) || !(stat & mstatus_rfe)) { + if (alg_data->mif.mode == I2C_SMBUS_WRITE) { + i2c_pnx_master_xmit(adap); + } else if (alg_data->mif.mode == I2C_SMBUS_READ) { + i2c_pnx_master_rcv(adap); + } + } + } + + /* Clear TDI and AFI bits */ + stat = ioread32(I2C_REG_STS(alg_data)); + iowrite32(stat | mstatus_tdi | mstatus_afi, I2C_REG_STS(alg_data)); + + dev_dbg(&adap->dev, "%s(): exiting, stat = %x ctrl = %x.\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data)), + ioread32(I2C_REG_CTL(alg_data))); + + return IRQ_HANDLED; +} + +static void i2c_pnx_timeout(unsigned long data) +{ + struct i2c_adapter *adap = (struct i2c_adapter *)data; + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + u32 ctl; + + dev_err(&adap->dev, "Master timed out. stat = %04x, cntrl = %04x. " + "Resetting master...\n", + ioread32(I2C_REG_STS(alg_data)), + ioread32(I2C_REG_CTL(alg_data))); + + /* Reset master and disable interrupts */ + ctl = ioread32(I2C_REG_CTL(alg_data)); + ctl &= ~(mcntrl_afie | mcntrl_naie | mcntrl_rffie | mcntrl_drmie); + iowrite32(ctl, I2C_REG_CTL(alg_data)); + + ctl |= mcntrl_reset; + iowrite32(ctl, I2C_REG_CTL(alg_data)); + wait_reset(I2C_PNX_TIMEOUT, alg_data); + alg_data->mif.ret = -EIO; + complete(&alg_data->mif.complete); +} + +static inline void bus_reset_if_active(struct i2c_adapter *adap) +{ + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + u32 stat; + + if ((stat = ioread32(I2C_REG_STS(alg_data))) & mstatus_active) { + dev_err(&adap->dev, + "%s: Bus is still active after xfer. Reset it...\n", + adap->name); + iowrite32(ioread32(I2C_REG_CTL(alg_data)) | mcntrl_reset, + I2C_REG_CTL(alg_data)); + wait_reset(I2C_PNX_TIMEOUT, alg_data); + } else if (!(stat & mstatus_rfe) || !(stat & mstatus_tfe)) { + /* If there is data in the fifo's after transfer, + * flush fifo's by reset. + */ + iowrite32(ioread32(I2C_REG_CTL(alg_data)) | mcntrl_reset, + I2C_REG_CTL(alg_data)); + wait_reset(I2C_PNX_TIMEOUT, alg_data); + } else if (stat & mstatus_nai) { + iowrite32(ioread32(I2C_REG_CTL(alg_data)) | mcntrl_reset, + I2C_REG_CTL(alg_data)); + wait_reset(I2C_PNX_TIMEOUT, alg_data); + } +} + +/** + * i2c_pnx_xfer - generic transfer entry point + * @adap: pointer to I2C adapter structure + * @msgs: array of messages + * @num: number of messages + * + * Initiates the transfer + */ +static int +i2c_pnx_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num) +{ + struct i2c_msg *pmsg; + int rc = 0, completed = 0, i; + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + u32 stat = ioread32(I2C_REG_STS(alg_data)); + + dev_dbg(&adap->dev, "%s(): entering: %d messages, stat = %04x.\n", + __FUNCTION__, num, ioread32(I2C_REG_STS(alg_data))); + + bus_reset_if_active(adap); + + /* Process transactions in a loop. */ + for (i = 0; rc >= 0 && i < num; i++) { + u8 addr; + + pmsg = &msgs[i]; + addr = pmsg->addr; + + if (pmsg->flags & I2C_M_TEN) { + dev_err(&adap->dev, + "%s: 10 bits addr not supported!\n", + adap->name); + rc = -EINVAL; + break; + } + + alg_data->mif.buf = pmsg->buf; + alg_data->mif.len = pmsg->len; + alg_data->mif.mode = (pmsg->flags & I2C_M_RD) ? + I2C_SMBUS_READ : I2C_SMBUS_WRITE; + alg_data->mif.ret = 0; + alg_data->last = (i == num - 1); + + dev_dbg(&adap->dev, "%s(): mode %d, %d bytes\n", __FUNCTION__, + alg_data->mif.mode, + alg_data->mif.len); + + i2c_pnx_arm_timer(adap); + + /* initialize the completion var */ + init_completion(&alg_data->mif.complete); + + /* Enable master interrupt */ + iowrite32(ioread32(I2C_REG_CTL(alg_data)) | mcntrl_afie | + mcntrl_naie | mcntrl_drmie, + I2C_REG_CTL(alg_data)); + + /* Put start-code and slave-address on the bus. */ + rc = i2c_pnx_start(addr, adap); + if (rc < 0) + break; + + /* Wait for completion */ + wait_for_completion(&alg_data->mif.complete); + + if (!(rc = alg_data->mif.ret)) + completed++; + dev_dbg(&adap->dev, "%s(): Complete, return code = %d.\n", + __FUNCTION__, rc); + + /* Clear TDI and AFI bits in case they are set. */ + if ((stat = ioread32(I2C_REG_STS(alg_data))) & mstatus_tdi) { + dev_dbg(&adap->dev, + "%s: TDI still set... clearing now.\n", + adap->name); + iowrite32(stat, I2C_REG_STS(alg_data)); + } + if ((stat = ioread32(I2C_REG_STS(alg_data))) & mstatus_afi) { + dev_dbg(&adap->dev, + "%s: AFI still set... clearing now.\n", + adap->name); + iowrite32(stat, I2C_REG_STS(alg_data)); + } + } + + bus_reset_if_active(adap); + + /* Cleanup to be sure... */ + alg_data->mif.buf = NULL; + alg_data->mif.len = 0; + + dev_dbg(&adap->dev, "%s(): exiting, stat = %x\n", + __FUNCTION__, ioread32(I2C_REG_STS(alg_data))); + + if (completed != num) + return ((rc < 0) ? rc : -EREMOTEIO); + + return num; +} + +static u32 i2c_pnx_func(struct i2c_adapter *adapter) +{ + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; +} + +static struct i2c_algorithm pnx_algorithm = { + .master_xfer = i2c_pnx_xfer, + .functionality = i2c_pnx_func, +}; + +static int i2c_pnx_controller_suspend(struct platform_device *pdev, + pm_message_t state) +{ + struct i2c_pnx_data *i2c_pnx = platform_get_drvdata(pdev); + return i2c_pnx->suspend(pdev, state); +} + +static int i2c_pnx_controller_resume(struct platform_device *pdev) +{ + struct i2c_pnx_data *i2c_pnx = platform_get_drvdata(pdev); + return i2c_pnx->resume(pdev); +} + +static int __devinit i2c_pnx_probe(struct platform_device *pdev) +{ + unsigned long tmp; + int ret = 0; + struct i2c_pnx_algo_data *alg_data; + int freq_mhz; + struct i2c_pnx_data *i2c_pnx = pdev->dev.platform_data; + + if (!i2c_pnx || !i2c_pnx->adapter) { + dev_err(&pdev->dev, "%s: no platform data supplied\n", + __FUNCTION__); + ret = -EINVAL; + goto out; + } + + platform_set_drvdata(pdev, i2c_pnx); + + if (i2c_pnx->calculate_input_freq) + freq_mhz = i2c_pnx->calculate_input_freq(pdev); + else { + freq_mhz = PNX_DEFAULT_FREQ; + dev_info(&pdev->dev, "Setting bus frequency to default value: " + "%d MHz", freq_mhz); + } + + i2c_pnx->adapter->algo = &pnx_algorithm; + + alg_data = i2c_pnx->adapter->algo_data; + init_timer(&alg_data->mif.timer); + alg_data->mif.timer.function = i2c_pnx_timeout; + alg_data->mif.timer.data = (unsigned long)i2c_pnx->adapter; + + /* Register I/O resource */ + if (!request_region(alg_data->base, I2C_PNX_REGION_SIZE, pdev->name)) { + dev_err(&pdev->dev, + "I/O region 0x%08x for I2C already in use.\n", + alg_data->base); + ret = -ENODEV; + goto out_drvdata; + } + + if (!(alg_data->ioaddr = + (u32)ioremap(alg_data->base, I2C_PNX_REGION_SIZE))) { + dev_err(&pdev->dev, "Couldn't ioremap I2C I/O region\n"); + ret = -ENOMEM; + goto out_release; + } + + i2c_pnx->set_clock_run(pdev); + + /* + * Clock Divisor High This value is the number of system clocks + * the serial clock (SCL) will be high. + * For example, if the system clock period is 50 ns and the maximum + * desired serial period is 10000 ns (100 kHz), then CLKHI would be + * set to 0.5*(f_sys/f_i2c)-2=0.5*(20e6/100e3)-2=98. The actual value + * programmed into CLKHI will vary from this slightly due to + * variations in the output pad's rise and fall times as well as + * the deglitching filter length. + */ + + tmp = ((freq_mhz * 1000) / I2C_PNX_SPEED_KHZ) / 2 - 2; + iowrite32(tmp, I2C_REG_CKH(alg_data)); + iowrite32(tmp, I2C_REG_CKL(alg_data)); + + iowrite32(mcntrl_reset, I2C_REG_CTL(alg_data)); + if (wait_reset(I2C_PNX_TIMEOUT, alg_data)) { + ret = -ENODEV; + goto out_unmap; + } + init_completion(&alg_data->mif.complete); + + ret = request_irq(alg_data->irq, i2c_pnx_interrupt, + 0, pdev->name, i2c_pnx->adapter); + if (ret) + goto out_clock; + + /* Register this adapter with the I2C subsystem */ + i2c_pnx->adapter->dev.parent = &pdev->dev; + ret = i2c_add_adapter(i2c_pnx->adapter); + if (ret < 0) { + dev_err(&pdev->dev, "I2C: Failed to add bus\n"); + goto out_irq; + } + + dev_dbg(&pdev->dev, "%s: Master at %#8x, irq %d.\n", + i2c_pnx->adapter->name, alg_data->base, alg_data->irq); + + return 0; + +out_irq: + free_irq(alg_data->irq, alg_data); +out_clock: + i2c_pnx->set_clock_stop(pdev); +out_unmap: + iounmap((void *)alg_data->ioaddr); +out_release: + release_region(alg_data->base, I2C_PNX_REGION_SIZE); +out_drvdata: + platform_set_drvdata(pdev, NULL); +out: + return ret; +} + +static int __devexit i2c_pnx_remove(struct platform_device *pdev) +{ + struct i2c_pnx_data *i2c_pnx = platform_get_drvdata(pdev); + struct i2c_adapter *adap = i2c_pnx->adapter; + struct i2c_pnx_algo_data *alg_data = adap->algo_data; + + free_irq(alg_data->irq, alg_data); + i2c_del_adapter(adap); + i2c_pnx->set_clock_stop(pdev); + iounmap((void *)alg_data->ioaddr); + release_region(alg_data->base, I2C_PNX_REGION_SIZE); + platform_set_drvdata(pdev, NULL); + + return 0; +} + +static struct platform_driver i2c_pnx_driver = { + .driver = { + .name = "pnx-i2c", + .owner = THIS_MODULE, + }, + .probe = i2c_pnx_probe, + .remove = __devexit_p(i2c_pnx_remove), + .suspend = i2c_pnx_controller_suspend, + .resume = i2c_pnx_controller_resume, +}; + +static int __init i2c_adap_pnx_init(void) +{ + return platform_driver_register(&i2c_pnx_driver); +} + +static void __exit i2c_adap_pnx_exit(void) +{ + platform_driver_unregister(&i2c_pnx_driver); +} + +MODULE_AUTHOR("Vitaly Wool, Dennis Kovalev "); +MODULE_DESCRIPTION("I2C driver for Philips IP3204-based I2C busses"); +MODULE_LICENSE("GPL"); + +#ifdef CONFIG_I2C_PNX_EARLY +/* We need to make sure I2C is initialized before USB */ +subsys_initcall(i2c_adap_pnx_init); +#else +mudule_init(i2c_adap_pnx_init); +#endif +module_exit(i2c_adap_pnx_exit); diff --git a/include/asm-arm/arch-pnx4008/i2c.h b/include/asm-arm/arch-pnx4008/i2c.h new file mode 100644 index 000000000000..92e8d65006f7 --- /dev/null +++ b/include/asm-arm/arch-pnx4008/i2c.h @@ -0,0 +1,67 @@ +/* + * PNX4008-specific tweaks for I2C IP3204 block + * + * Author: Vitaly Wool + * + * 2005 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ + +#ifndef __ASM_ARCH_I2C_H__ +#define __ASM_ARCH_I2C_H__ + +#include +#include + +enum { + mstatus_tdi = 0x00000001, + mstatus_afi = 0x00000002, + mstatus_nai = 0x00000004, + mstatus_drmi = 0x00000008, + mstatus_active = 0x00000020, + mstatus_scl = 0x00000040, + mstatus_sda = 0x00000080, + mstatus_rff = 0x00000100, + mstatus_rfe = 0x00000200, + mstatus_tff = 0x00000400, + mstatus_tfe = 0x00000800, +}; + +enum { + mcntrl_tdie = 0x00000001, + mcntrl_afie = 0x00000002, + mcntrl_naie = 0x00000004, + mcntrl_drmie = 0x00000008, + mcntrl_daie = 0x00000020, + mcntrl_rffie = 0x00000040, + mcntrl_tffie = 0x00000080, + mcntrl_reset = 0x00000100, + mcntrl_cdbmode = 0x00000400, +}; + +enum { + rw_bit = 1 << 0, + start_bit = 1 << 8, + stop_bit = 1 << 9, +}; + +#define I2C_REG_RX(a) ((a)->ioaddr) /* Rx FIFO reg (RO) */ +#define I2C_REG_TX(a) ((a)->ioaddr) /* Tx FIFO reg (WO) */ +#define I2C_REG_STS(a) ((a)->ioaddr + 0x04) /* Status reg (RO) */ +#define I2C_REG_CTL(a) ((a)->ioaddr + 0x08) /* Ctl reg */ +#define I2C_REG_CKL(a) ((a)->ioaddr + 0x0c) /* Clock divider low */ +#define I2C_REG_CKH(a) ((a)->ioaddr + 0x10) /* Clock divider high */ +#define I2C_REG_ADR(a) ((a)->ioaddr + 0x14) /* I2C address */ +#define I2C_REG_RFL(a) ((a)->ioaddr + 0x18) /* Rx FIFO level (RO) */ +#define I2C_REG_TFL(a) ((a)->ioaddr + 0x1c) /* Tx FIFO level (RO) */ +#define I2C_REG_RXB(a) ((a)->ioaddr + 0x20) /* Num of bytes Rx-ed (RO) */ +#define I2C_REG_TXB(a) ((a)->ioaddr + 0x24) /* Num of bytes Tx-ed (RO) */ +#define I2C_REG_TXS(a) ((a)->ioaddr + 0x28) /* Tx slave FIFO (RO) */ +#define I2C_REG_STFL(a) ((a)->ioaddr + 0x2c) /* Tx slave FIFO level (RO) */ + +#define HCLK_MHZ 13 +#define I2C_CHIP_NAME "PNX4008-I2C" + +#endif /* __ASM_ARCH_I2C_H___ */ diff --git a/include/linux/i2c-pnx.h b/include/linux/i2c-pnx.h new file mode 100644 index 000000000000..e6e9c814da61 --- /dev/null +++ b/include/linux/i2c-pnx.h @@ -0,0 +1,43 @@ +/* + * Header file for I2C support on PNX010x/4008. + * + * Author: Dennis Kovalev + * + * 2004-2006 (c) MontaVista Software, Inc. This file is licensed under + * the terms of the GNU General Public License version 2. This program + * is licensed "as is" without any warranty of any kind, whether express + * or implied. + */ + +#ifndef __I2C_PNX_H__ +#define __I2C_PNX_H__ + +#include + +struct i2c_pnx_mif { + int ret; /* Return value */ + int mode; /* Interface mode */ + struct completion complete; /* I/O completion */ + struct timer_list timer; /* Timeout */ + char * buf; /* Data buffer */ + int len; /* Length of data buffer */ +}; + +struct i2c_pnx_algo_data { + u32 base; + u32 ioaddr; + int irq; + struct i2c_pnx_mif mif; + int last; +}; + +struct i2c_pnx_data { + int (*suspend) (struct platform_device *pdev, pm_message_t state); + int (*resume) (struct platform_device *pdev); + u32 (*calculate_input_freq) (struct platform_device *pdev); + int (*set_clock_run) (struct platform_device *pdev); + int (*set_clock_stop) (struct platform_device *pdev); + struct i2c_adapter *adapter; +}; + +#endif /* __I2C_PNX_H__ */ -- cgit v1.2.3 From 6ea23039cb1cc7c379eb5fba0ed2c53291e9bea7 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Sun, 10 Dec 2006 21:21:30 +0100 Subject: i2c: Add support for nested i2c bus locking This patch adds the 'level' field into the i2c_adapter structure, which is used to represent the 'logical' level of nesting for the purposes of lockdep. This field is then used in the i2c_transfer() function, to acquire the per-adapter bus_lock with correct nesting level. Signed-off-by: Jiri Kosina Signed-off-by: Jean Delvare --- drivers/i2c/i2c-core.c | 2 +- include/linux/i2c.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 7ca81f42d14b..79eaa12474dd 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -603,7 +603,7 @@ int i2c_transfer(struct i2c_adapter * adap, struct i2c_msg *msgs, int num) } #endif - mutex_lock(&adap->bus_lock); + mutex_lock_nested(&adap->bus_lock, adap->level); ret = adap->algo->master_xfer(adap,msgs,num); mutex_unlock(&adap->bus_lock); diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 9b5d04768c2c..08df4169b411 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -216,6 +216,7 @@ struct i2c_adapter { int (*client_unregister)(struct i2c_client *); /* data fields that are valid for all devices */ + u8 level; /* nesting level for lockdep */ struct mutex bus_lock; struct mutex clist_lock; -- cgit v1.2.3 From 438d6c2c015cf63bf7e9bdc2033d435433ac8455 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Sun, 10 Dec 2006 21:21:31 +0100 Subject: i2c: Whitespace cleanups Remove extraneous whitespace from various i2c headers and core files, like space-before-tab and whitespace at end of line. Signed-off-by: David Brownell Signed-off-by: Jean Delvare --- drivers/i2c/i2c-core.c | 54 ++++++++++++++++---------------- drivers/i2c/i2c-dev.c | 34 ++++++++++---------- include/linux/i2c-algo-bit.h | 4 +-- include/linux/i2c-algo-pcf.h | 2 +- include/linux/i2c-id.h | 10 +++--- include/linux/i2c.h | 74 ++++++++++++++++++++++---------------------- 6 files changed, 89 insertions(+), 89 deletions(-) (limited to 'include/linux') diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 79eaa12474dd..64f11e00651c 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -127,7 +127,7 @@ static ssize_t show_client_name(struct device *dev, struct device_attribute *att return sprintf(buf, "%s\n", client->name); } -/* +/* * We can't use the DEVICE_ATTR() macro here as we want the same filename for a * different type of a device. So beware if the DEVICE_ATTR() macro ever * changes, this definition will also have to change. @@ -139,8 +139,8 @@ static struct device_attribute dev_attr_client_name = { /* --------------------------------------------------- - * registering functions - * --------------------------------------------------- + * registering functions + * --------------------------------------------------- */ /* ----- @@ -314,7 +314,7 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver) res = driver_register(&driver->driver); if (res) return res; - + mutex_lock(&core_lists); list_add_tail(&driver->list,&drivers); @@ -338,13 +338,13 @@ int i2c_del_driver(struct i2c_driver *driver) struct list_head *item1, *item2, *_n; struct i2c_client *client; struct i2c_adapter *adap; - + int res = 0; mutex_lock(&core_lists); /* Have a look at each adapter, if clients of this driver are still - * attached. If so, detach them to be able to kill the driver + * attached. If so, detach them to be able to kill the driver * afterwards. */ list_for_each(item1,&adapters) { @@ -419,14 +419,14 @@ int i2c_attach_client(struct i2c_client *client) goto out_unlock; } list_add_tail(&client->list,&adapter->clients); - + client->usage_count = 0; client->dev.parent = &client->adapter->dev; client->dev.driver = &client->driver->driver; client->dev.bus = &i2c_bus_type; client->dev.release = &i2c_client_release; - + snprintf(&client->dev.bus_id[0], sizeof(client->dev.bus_id), "%d-%04x", i2c_adapter_id(adapter), client->addr); dev_dbg(&adapter->dev, "client [%s] registered with bus id %s\n", @@ -467,7 +467,7 @@ int i2c_detach_client(struct i2c_client *client) { struct i2c_adapter *adapter = client->adapter; int res = 0; - + if (client->usage_count > 0) { dev_warn(&client->dev, "Client [%s] still busy, " "can't detach\n", client->name); @@ -535,10 +535,10 @@ int i2c_release_client(struct i2c_client *client) __FUNCTION__); return -EPERM; } - + client->usage_count--; i2c_dec_use_client(client); - + return 0; } @@ -624,7 +624,7 @@ int i2c_master_send(struct i2c_client *client,const char *buf ,int count) msg.flags = client->flags & I2C_M_TEN; msg.len = count; msg.buf = (char *)buf; - + ret = i2c_transfer(adap, &msg, 1); /* If everything went ok (i.e. 1 msg transmitted), return #bytes @@ -757,7 +757,7 @@ int i2c_probe(struct i2c_adapter *adapter, if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_QUICK)) { if (address_data->probe[0] == I2C_CLIENT_END && address_data->normal_i2c[0] == I2C_CLIENT_END) - return 0; + return 0; dev_warn(&adapter->dev, "SMBus Quick command not supported, " "can't probe for chips\n"); @@ -817,7 +817,7 @@ int i2c_probe(struct i2c_adapter *adapter, struct i2c_adapter* i2c_get_adapter(int id) { struct i2c_adapter *adapter; - + mutex_lock(&core_lists); adapter = (struct i2c_adapter *)idr_find(&i2c_adapter_idr, id); if (adapter && !try_module_get(adapter->owner)) @@ -834,14 +834,14 @@ void i2c_put_adapter(struct i2c_adapter *adap) /* The SMBus parts */ -#define POLY (0x1070U << 3) +#define POLY (0x1070U << 3) static u8 crc8(u16 data) { int i; - + for(i = 0; i < 8; i++) { - if (data & 0x8000) + if (data & 0x8000) data = data ^ POLY; data = data << 1; } @@ -891,13 +891,13 @@ static int i2c_smbus_check_pec(u8 cpec, struct i2c_msg *msg) rpec, cpec); return -1; } - return 0; + return 0; } s32 i2c_smbus_write_quick(struct i2c_client *client, u8 value) { return i2c_smbus_xfer(client->adapter,client->addr,client->flags, - value,0,I2C_SMBUS_QUICK,NULL); + value,0,I2C_SMBUS_QUICK,NULL); } s32 i2c_smbus_read_byte(struct i2c_client *client) @@ -996,11 +996,11 @@ s32 i2c_smbus_write_i2c_block_data(struct i2c_client *client, u8 command, I2C_SMBUS_I2C_BLOCK_DATA, &data); } -/* Simulate a SMBus command using the i2c protocol +/* Simulate a SMBus command using the i2c protocol No checking of parameters is done! */ -static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, +static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, unsigned short flags, - char read_write, u8 command, int size, + char read_write, u8 command, int size, union i2c_smbus_data * data) { /* So we need to generate a series of msgs. In the case of writing, we @@ -1010,7 +1010,7 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, unsigned char msgbuf0[I2C_SMBUS_BLOCK_MAX+3]; unsigned char msgbuf1[I2C_SMBUS_BLOCK_MAX+2]; int num = read_write == I2C_SMBUS_READ?2:1; - struct i2c_msg msg[2] = { { addr, flags, 1, msgbuf0 }, + struct i2c_msg msg[2] = { { addr, flags, 1, msgbuf0 }, { addr, flags | I2C_M_RD, 0, msgbuf1 } }; int i; @@ -1103,14 +1103,14 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, if (i) { /* Compute PEC if first message is a write */ if (!(msg[0].flags & I2C_M_RD)) { - if (num == 1) /* Write only */ + if (num == 1) /* Write only */ i2c_smbus_add_pec(&msg[0]); else /* Write followed by read */ partial_pec = i2c_smbus_msg_pec(0, &msg[0]); } /* Ask for PEC if last message is a read */ if (msg[num-1].flags & I2C_M_RD) - msg[num-1].len++; + msg[num-1].len++; } if (i2c_transfer(adapter, msg, num) < 0) @@ -1130,7 +1130,7 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, case I2C_SMBUS_BYTE_DATA: data->byte = msgbuf1[0]; break; - case I2C_SMBUS_WORD_DATA: + case I2C_SMBUS_WORD_DATA: case I2C_SMBUS_PROC_CALL: data->word = msgbuf1[0] | (msgbuf1[1] << 8); break; @@ -1146,7 +1146,7 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr, s32 i2c_smbus_xfer(struct i2c_adapter * adapter, u16 addr, unsigned short flags, - char read_write, u8 command, int size, + char read_write, u8 command, int size, union i2c_smbus_data * data) { s32 res; diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c index d66cefc63df7..cf9381eae8da 100644 --- a/drivers/i2c/i2c-dev.c +++ b/drivers/i2c/i2c-dev.c @@ -1,5 +1,5 @@ /* - i2c-dev.c - i2c-bus driver, char device interface + i2c-dev.c - i2c-bus driver, char device interface Copyright (C) 1995-97 Simon G. Vogl Copyright (C) 1998-99 Frodo Looijaard @@ -172,7 +172,7 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, switch ( cmd ) { case I2C_SLAVE: case I2C_SLAVE_FORCE: - if ((arg > 0x3ff) || + if ((arg > 0x3ff) || (((client->flags & I2C_M_TEN) == 0) && arg > 0x7f)) return -EINVAL; if ((cmd == I2C_SLAVE) && i2c_check_addr(client->adapter,arg)) @@ -196,8 +196,8 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, return put_user(funcs, (unsigned long __user *)arg); case I2C_RDWR: - if (copy_from_user(&rdwr_arg, - (struct i2c_rdwr_ioctl_data __user *)arg, + if (copy_from_user(&rdwr_arg, + (struct i2c_rdwr_ioctl_data __user *)arg, sizeof(rdwr_arg))) return -EFAULT; @@ -205,9 +205,9 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, * be sent at once */ if (rdwr_arg.nmsgs > I2C_RDRW_IOCTL_MAX_MSGS) return -EINVAL; - + rdwr_pa = (struct i2c_msg *) - kmalloc(rdwr_arg.nmsgs * sizeof(struct i2c_msg), + kmalloc(rdwr_arg.nmsgs * sizeof(struct i2c_msg), GFP_KERNEL); if (rdwr_pa == NULL) return -ENOMEM; @@ -277,9 +277,9 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, (struct i2c_smbus_ioctl_data __user *) arg, sizeof(struct i2c_smbus_ioctl_data))) return -EFAULT; - if ((data_arg.size != I2C_SMBUS_BYTE) && + if ((data_arg.size != I2C_SMBUS_BYTE) && (data_arg.size != I2C_SMBUS_QUICK) && - (data_arg.size != I2C_SMBUS_BYTE_DATA) && + (data_arg.size != I2C_SMBUS_BYTE_DATA) && (data_arg.size != I2C_SMBUS_WORD_DATA) && (data_arg.size != I2C_SMBUS_PROC_CALL) && (data_arg.size != I2C_SMBUS_BLOCK_DATA) && @@ -290,11 +290,11 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, data_arg.size); return -EINVAL; } - /* Note that I2C_SMBUS_READ and I2C_SMBUS_WRITE are 0 and 1, + /* Note that I2C_SMBUS_READ and I2C_SMBUS_WRITE are 0 and 1, so the check is valid if size==I2C_SMBUS_QUICK too. */ - if ((data_arg.read_write != I2C_SMBUS_READ) && + if ((data_arg.read_write != I2C_SMBUS_READ) && (data_arg.read_write != I2C_SMBUS_WRITE)) { - dev_dbg(&client->adapter->dev, + dev_dbg(&client->adapter->dev, "read_write out of range (%x) in ioctl I2C_SMBUS.\n", data_arg.read_write); return -EINVAL; @@ -303,7 +303,7 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, /* Note that command values are always valid! */ if ((data_arg.size == I2C_SMBUS_QUICK) || - ((data_arg.size == I2C_SMBUS_BYTE) && + ((data_arg.size == I2C_SMBUS_BYTE) && (data_arg.read_write == I2C_SMBUS_WRITE))) /* These are special: we do not use data */ return i2c_smbus_xfer(client->adapter, client->addr, @@ -321,14 +321,14 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, if ((data_arg.size == I2C_SMBUS_BYTE_DATA) || (data_arg.size == I2C_SMBUS_BYTE)) datasize = sizeof(data_arg.data->byte); - else if ((data_arg.size == I2C_SMBUS_WORD_DATA) || + else if ((data_arg.size == I2C_SMBUS_WORD_DATA) || (data_arg.size == I2C_SMBUS_PROC_CALL)) datasize = sizeof(data_arg.data->word); else /* size == smbus block, i2c block, or block proc. call */ datasize = sizeof(data_arg.data->block); - if ((data_arg.size == I2C_SMBUS_PROC_CALL) || - (data_arg.size == I2C_SMBUS_BLOCK_PROC_CALL) || + if ((data_arg.size == I2C_SMBUS_PROC_CALL) || + (data_arg.size == I2C_SMBUS_BLOCK_PROC_CALL) || (data_arg.read_write == I2C_SMBUS_WRITE)) { if (copy_from_user(&temp, data_arg.data, datasize)) return -EFAULT; @@ -336,8 +336,8 @@ static int i2cdev_ioctl(struct inode *inode, struct file *file, res = i2c_smbus_xfer(client->adapter,client->addr,client->flags, data_arg.read_write, data_arg.command,data_arg.size,&temp); - if (! res && ((data_arg.size == I2C_SMBUS_PROC_CALL) || - (data_arg.size == I2C_SMBUS_BLOCK_PROC_CALL) || + if (! res && ((data_arg.size == I2C_SMBUS_PROC_CALL) || + (data_arg.size == I2C_SMBUS_BLOCK_PROC_CALL) || (data_arg.read_write == I2C_SMBUS_READ))) { if (copy_to_user(data_arg.data, &temp, datasize)) return -EFAULT; diff --git a/include/linux/i2c-algo-bit.h b/include/linux/i2c-algo-bit.h index c8f8df25c7e0..a99e28a86c3b 100644 --- a/include/linux/i2c-algo-bit.h +++ b/include/linux/i2c-algo-bit.h @@ -26,9 +26,9 @@ /* --- Defines for bit-adapters --------------------------------------- */ /* - * This struct contains the hw-dependent functions of bit-style adapters to + * This struct contains the hw-dependent functions of bit-style adapters to * manipulate the line states, and to init any hw-specific features. This is - * only used if you have more than one hw-type of adapter running. + * only used if you have more than one hw-type of adapter running. */ struct i2c_algo_bit_data { void *data; /* private data for lowlevel routines */ diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h index 9908f3fc4839..7d7045cbb344 100644 --- a/include/linux/i2c-algo-pcf.h +++ b/include/linux/i2c-algo-pcf.h @@ -31,7 +31,7 @@ struct i2c_algo_pcf_data { int (*getpcf) (void *data, int ctl); int (*getown) (void *data); int (*getclock) (void *data); - void (*waitforpin) (void); + void (*waitforpin) (void); /* local settings */ int udelay; diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 00da370bbcd6..5b97f1cce7a5 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -1,7 +1,7 @@ /* ------------------------------------------------------------------------- */ -/* */ +/* */ /* i2c-id.h - identifier values for i2c drivers and adapters */ -/* */ +/* */ /* ------------------------------------------------------------------------- */ /* Copyright (C) 1995-1999 Simon G. Vogl @@ -40,10 +40,10 @@ #define I2C_DRIVERID_SAA7120 11 /* video encoder */ #define I2C_DRIVERID_SAA7121 12 /* video encoder */ #define I2C_DRIVERID_SAA7185B 13 /* video encoder */ -#define I2C_DRIVERID_CH7003 14 /* digital pc to tv encoder */ +#define I2C_DRIVERID_CH7003 14 /* digital pc to tv encoder */ #define I2C_DRIVERID_PCF8574A 15 /* i2c expander - 8 bit in/out */ #define I2C_DRIVERID_PCF8582C 16 /* eeprom */ -#define I2C_DRIVERID_AT24Cxx 17 /* eeprom 1/2/4/8/16 K */ +#define I2C_DRIVERID_AT24Cxx 17 /* eeprom 1/2/4/8/16 K */ #define I2C_DRIVERID_TEA6300 18 /* audio mixer */ #define I2C_DRIVERID_BT829 19 /* pc to tv encoder */ #define I2C_DRIVERID_TDA9850 20 /* audio mixer */ @@ -162,7 +162,7 @@ * ---- Adapter types ---------------------------------------------------- */ -/* --- Bit algorithm adapters */ +/* --- Bit algorithm adapters */ #define I2C_HW_B_LP 0x010000 /* Parallel port Philips style */ #define I2C_HW_B_SER 0x010002 /* Serial line interface */ #define I2C_HW_B_BT848 0x010005 /* BT848 video boards */ diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 08df4169b411..71e50d3e492f 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -1,7 +1,7 @@ /* ------------------------------------------------------------------------- */ -/* */ +/* */ /* i2c.h - definitions for the i2c-bus interface */ -/* */ +/* */ /* ------------------------------------------------------------------------- */ /* Copyright (C) 1995-2000 Simon G. Vogl @@ -27,7 +27,7 @@ #define _LINUX_I2C_H #include -#ifdef __KERNEL__ +#ifdef __KERNEL__ #include #include #include @@ -53,8 +53,8 @@ union i2c_smbus_data; /* * The master routines are the ones normally used to transmit data to devices - * on a bus (or read from them). Apart from two basic transfer functions to - * transmit one message at a time, a more complex version can be used to + * on a bus (or read from them). Apart from two basic transfer functions to + * transmit one message at a time, a more complex version can be used to * transmit an arbitrary number of messages without interruption. */ extern int i2c_master_send(struct i2c_client *,const char* ,int); @@ -67,10 +67,10 @@ extern int i2c_transfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num) /* This is the very generalized SMBus access routine. You probably do not want to use this, though; one of the functions below may be much easier, - and probably just as fast. + and probably just as fast. Note that we use i2c_adapter here, because you do not need a specific smbus adapter to call this function. */ -extern s32 i2c_smbus_xfer (struct i2c_adapter * adapter, u16 addr, +extern s32 i2c_smbus_xfer (struct i2c_adapter * adapter, u16 addr, unsigned short flags, char read_write, u8 command, int size, union i2c_smbus_data * data); @@ -112,14 +112,14 @@ struct i2c_driver { /* Notifies the driver that a new bus has appeared. This routine * can be used by the driver to test if the bus meets its conditions - * & seek for the presence of the chip(s) it supports. If found, it + * & seek for the presence of the chip(s) it supports. If found, it * registers the client(s) that are on the bus to the i2c admin. via * i2c_attach_client. */ int (*attach_adapter)(struct i2c_adapter *); int (*detach_adapter)(struct i2c_adapter *); - /* tells the driver that a client is about to be deleted & gives it + /* tells the driver that a client is about to be deleted & gives it * the chance to remove its private data. Also, if the client struct * has been dynamically allocated by the driver in the function above, * it must be freed here. @@ -139,13 +139,13 @@ struct i2c_driver { #define I2C_NAME_SIZE 50 /* - * i2c_client identifies a single device (i.e. chip) that is connected to an + * i2c_client identifies a single device (i.e. chip) that is connected to an * i2c bus. The behaviour is defined by the routines of the driver. This * function is mainly used for lookup & other admin. functions. */ struct i2c_client { unsigned int flags; /* div., see below */ - unsigned short addr; /* chip address - NOTE: 7bit */ + unsigned short addr; /* chip address - NOTE: 7bit */ /* addresses are stored in the */ /* _LOWER_ 7 bits */ struct i2c_adapter *adapter; /* the adapter we sit on */ @@ -182,14 +182,14 @@ static inline void i2c_set_clientdata (struct i2c_client *dev, void *data) */ struct i2c_algorithm { /* If an adapter algorithm can't do I2C-level access, set master_xfer - to NULL. If an adapter algorithm can do SMBus access, set + to NULL. If an adapter algorithm can do SMBus access, set smbus_xfer. If set to NULL, the SMBus protocol is simulated using common I2C messages */ /* master_xfer should return the number of messages successfully processed, or a negative value on error */ - int (*master_xfer)(struct i2c_adapter *adap,struct i2c_msg *msgs, + int (*master_xfer)(struct i2c_adapter *adap,struct i2c_msg *msgs, int num); - int (*smbus_xfer) (struct i2c_adapter *adap, u16 addr, + int (*smbus_xfer) (struct i2c_adapter *adap, u16 addr, unsigned short flags, char read_write, u8 command, int size, union i2c_smbus_data * data); @@ -317,7 +317,7 @@ extern int i2c_check_addr (struct i2c_adapter *adapter, int addr); * It will only call found_proc if some client is connected at the * specific address (unless a 'force' matched); */ -extern int i2c_probe(struct i2c_adapter *adapter, +extern int i2c_probe(struct i2c_adapter *adapter, struct i2c_client_address_data *address_data, int (*found_proc) (struct i2c_adapter *, int, int)); @@ -353,15 +353,15 @@ static inline int i2c_adapter_id(struct i2c_adapter *adap) */ struct i2c_msg { __u16 addr; /* slave address */ - __u16 flags; + __u16 flags; #define I2C_M_TEN 0x10 /* we have a ten bit chip address */ #define I2C_M_RD 0x01 #define I2C_M_NOSTART 0x4000 #define I2C_M_REV_DIR_ADDR 0x2000 #define I2C_M_IGNORE_NAK 0x1000 #define I2C_M_NO_RD_ACK 0x0800 - __u16 len; /* msg length */ - __u8 *buf; /* pointer to msg data */ + __u16 len; /* msg length */ + __u8 *buf; /* pointer to msg data */ }; /* To determine what functionality is present */ @@ -371,16 +371,16 @@ struct i2c_msg { #define I2C_FUNC_PROTOCOL_MANGLING 0x00000004 /* I2C_M_{REV_DIR_ADDR,NOSTART,..} */ #define I2C_FUNC_SMBUS_HWPEC_CALC 0x00000008 /* SMBus 2.0 */ #define I2C_FUNC_SMBUS_BLOCK_PROC_CALL 0x00008000 /* SMBus 2.0 */ -#define I2C_FUNC_SMBUS_QUICK 0x00010000 -#define I2C_FUNC_SMBUS_READ_BYTE 0x00020000 -#define I2C_FUNC_SMBUS_WRITE_BYTE 0x00040000 -#define I2C_FUNC_SMBUS_READ_BYTE_DATA 0x00080000 -#define I2C_FUNC_SMBUS_WRITE_BYTE_DATA 0x00100000 -#define I2C_FUNC_SMBUS_READ_WORD_DATA 0x00200000 -#define I2C_FUNC_SMBUS_WRITE_WORD_DATA 0x00400000 -#define I2C_FUNC_SMBUS_PROC_CALL 0x00800000 -#define I2C_FUNC_SMBUS_READ_BLOCK_DATA 0x01000000 -#define I2C_FUNC_SMBUS_WRITE_BLOCK_DATA 0x02000000 +#define I2C_FUNC_SMBUS_QUICK 0x00010000 +#define I2C_FUNC_SMBUS_READ_BYTE 0x00020000 +#define I2C_FUNC_SMBUS_WRITE_BYTE 0x00040000 +#define I2C_FUNC_SMBUS_READ_BYTE_DATA 0x00080000 +#define I2C_FUNC_SMBUS_WRITE_BYTE_DATA 0x00100000 +#define I2C_FUNC_SMBUS_READ_WORD_DATA 0x00200000 +#define I2C_FUNC_SMBUS_WRITE_WORD_DATA 0x00400000 +#define I2C_FUNC_SMBUS_PROC_CALL 0x00800000 +#define I2C_FUNC_SMBUS_READ_BLOCK_DATA 0x01000000 +#define I2C_FUNC_SMBUS_WRITE_BLOCK_DATA 0x02000000 #define I2C_FUNC_SMBUS_READ_I2C_BLOCK 0x04000000 /* I2C-like block xfer */ #define I2C_FUNC_SMBUS_WRITE_I2C_BLOCK 0x08000000 /* w/ 1-byte reg. addr. */ #define I2C_FUNC_SMBUS_READ_I2C_BLOCK_2 0x10000000 /* I2C-like block xfer */ @@ -407,10 +407,10 @@ struct i2c_msg { I2C_FUNC_SMBUS_WRITE_BLOCK_DATA | \ I2C_FUNC_SMBUS_I2C_BLOCK) -/* - * Data for SMBus Messages +/* + * Data for SMBus Messages */ -#define I2C_SMBUS_BLOCK_MAX 32 /* As specified in SMBus standard */ +#define I2C_SMBUS_BLOCK_MAX 32 /* As specified in SMBus standard */ union i2c_smbus_data { __u8 byte; __u16 word; @@ -422,11 +422,11 @@ union i2c_smbus_data { #define I2C_SMBUS_READ 1 #define I2C_SMBUS_WRITE 0 -/* SMBus transaction types (size parameter in the above functions) +/* SMBus transaction types (size parameter in the above functions) Note: these no longer correspond to the (arbitrary) PIIX4 internal codes! */ #define I2C_SMBUS_QUICK 0 #define I2C_SMBUS_BYTE 1 -#define I2C_SMBUS_BYTE_DATA 2 +#define I2C_SMBUS_BYTE_DATA 2 #define I2C_SMBUS_WORD_DATA 3 #define I2C_SMBUS_PROC_CALL 4 #define I2C_SMBUS_BLOCK_DATA 5 @@ -435,15 +435,15 @@ union i2c_smbus_data { /* ----- commands for the ioctl like i2c_command call: - * note that additional calls are defined in the algorithm and hw - * dependent layers - these can be listed here, or see the + * note that additional calls are defined in the algorithm and hw + * dependent layers - these can be listed here, or see the * corresponding header files. */ /* -> bit-adapter specific ioctls */ #define I2C_RETRIES 0x0701 /* number of times a device address */ /* should be polled when not */ - /* acknowledging */ -#define I2C_TIMEOUT 0x0702 /* set timeout - call with int */ + /* acknowledging */ +#define I2C_TIMEOUT 0x0702 /* set timeout - call with int */ /* this is for i2c-dev.c */ -- cgit v1.2.3 From 3269711b76ba27b78862c48398b0d313ccaa99c2 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sun, 10 Dec 2006 21:21:33 +0100 Subject: i2c: Discard the i2c algo del_bus wrappers They are all only calling i2c_del_adapter, so we may as well do it directly. Signed-off-by: Jean Delvare --- drivers/acorn/char/i2c.c | 2 +- drivers/i2c/algos/i2c-algo-bit.c | 8 -------- drivers/i2c/algos/i2c-algo-pca.c | 7 ------- drivers/i2c/algos/i2c-algo-pcf.c | 8 -------- drivers/i2c/algos/i2c-algo-sgi.c | 8 -------- drivers/i2c/busses/i2c-elektor.c | 2 +- drivers/i2c/busses/i2c-hydra.c | 2 +- drivers/i2c/busses/i2c-i810.c | 6 +++--- drivers/i2c/busses/i2c-ixp2000.c | 2 +- drivers/i2c/busses/i2c-ixp4xx.c | 2 +- drivers/i2c/busses/i2c-parport-light.c | 2 +- drivers/i2c/busses/i2c-parport.c | 2 +- drivers/i2c/busses/i2c-pca-isa.c | 2 +- drivers/i2c/busses/i2c-prosavage.c | 2 +- drivers/i2c/busses/i2c-savage4.c | 2 +- drivers/i2c/busses/i2c-via.c | 2 +- drivers/i2c/busses/i2c-voodoo3.c | 6 +++--- drivers/i2c/busses/scx200_i2c.c | 2 +- drivers/ieee1394/pcilynx.c | 2 +- drivers/media/dvb/pluto2/pluto2.c | 8 ++++---- drivers/media/video/bt8xx/bttv-i2c.c | 6 +----- drivers/media/video/cx88/cx88-core.c | 2 +- drivers/media/video/cx88/cx88-vp3054-i2c.c | 2 +- drivers/media/video/vino.c | 2 +- drivers/media/video/zoran_card.c | 2 +- drivers/video/aty/radeon_i2c.c | 8 ++++---- drivers/video/i810/i810-i2c.c | 6 +++--- drivers/video/intelfb/intelfb_i2c.c | 4 ++-- drivers/video/matrox/i2c-matroxfb.c | 2 +- drivers/video/nvidia/nv_i2c.c | 6 +++--- drivers/video/riva/rivafb-i2c.c | 6 +++--- drivers/video/savage/savagefb-i2c.c | 2 +- include/linux/i2c-algo-bit.h | 1 - include/linux/i2c-algo-pca.h | 1 - include/linux/i2c-algo-pcf.h | 1 - include/linux/i2c-algo-sgi.h | 1 - 36 files changed, 45 insertions(+), 84 deletions(-) (limited to 'include/linux') diff --git a/drivers/acorn/char/i2c.c b/drivers/acorn/char/i2c.c index bdb9c8b78ed8..9e584a7af434 100644 --- a/drivers/acorn/char/i2c.c +++ b/drivers/acorn/char/i2c.c @@ -360,7 +360,7 @@ static int __init i2c_ioc_init(void) if (ret >= 0){ ret = misc_register(&rtc_dev); if(ret < 0) - i2c_bit_del_bus(&ioc_ops); + i2c_del_adapter(&ioc_ops); } return ret; diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c index 21c36bfb5e6b..95aa5395a5be 100644 --- a/drivers/i2c/algos/i2c-algo-bit.c +++ b/drivers/i2c/algos/i2c-algo-bit.c @@ -540,15 +540,7 @@ int i2c_bit_add_bus(struct i2c_adapter *adap) return i2c_add_adapter(adap); } - - -int i2c_bit_del_bus(struct i2c_adapter *adap) -{ - return i2c_del_adapter(adap); -} - EXPORT_SYMBOL(i2c_bit_add_bus); -EXPORT_SYMBOL(i2c_bit_del_bus); MODULE_AUTHOR("Simon G. Vogl "); MODULE_DESCRIPTION("I2C-Bus bit-banging algorithm"); diff --git a/drivers/i2c/algos/i2c-algo-pca.c b/drivers/i2c/algos/i2c-algo-pca.c index 9081c9fbcd29..36fdf971f080 100644 --- a/drivers/i2c/algos/i2c-algo-pca.c +++ b/drivers/i2c/algos/i2c-algo-pca.c @@ -381,14 +381,7 @@ int i2c_pca_add_bus(struct i2c_adapter *adap) return rval; } - -int i2c_pca_del_bus(struct i2c_adapter *adap) -{ - return i2c_del_adapter(adap); -} - EXPORT_SYMBOL(i2c_pca_add_bus); -EXPORT_SYMBOL(i2c_pca_del_bus); MODULE_AUTHOR("Ian Campbell "); MODULE_DESCRIPTION("I2C-Bus PCA9564 algorithm"); diff --git a/drivers/i2c/algos/i2c-algo-pcf.c b/drivers/i2c/algos/i2c-algo-pcf.c index 3b2003398966..ecb2c2d7d540 100644 --- a/drivers/i2c/algos/i2c-algo-pcf.c +++ b/drivers/i2c/algos/i2c-algo-pcf.c @@ -486,15 +486,7 @@ int i2c_pcf_add_bus(struct i2c_adapter *adap) return rval; } - - -int i2c_pcf_del_bus(struct i2c_adapter *adap) -{ - return i2c_del_adapter(adap); -} - EXPORT_SYMBOL(i2c_pcf_add_bus); -EXPORT_SYMBOL(i2c_pcf_del_bus); MODULE_AUTHOR("Hans Berglund "); MODULE_DESCRIPTION("I2C-Bus PCF8584 algorithm"); diff --git a/drivers/i2c/algos/i2c-algo-sgi.c b/drivers/i2c/algos/i2c-algo-sgi.c index 490d99997fd0..ac2d5053078a 100644 --- a/drivers/i2c/algos/i2c-algo-sgi.c +++ b/drivers/i2c/algos/i2c-algo-sgi.c @@ -171,15 +171,7 @@ int i2c_sgi_add_bus(struct i2c_adapter *adap) return i2c_add_adapter(adap); } - - -int i2c_sgi_del_bus(struct i2c_adapter *adap) -{ - return i2c_del_adapter(adap); -} - EXPORT_SYMBOL(i2c_sgi_add_bus); -EXPORT_SYMBOL(i2c_sgi_del_bus); MODULE_AUTHOR("Ladislav Michl "); MODULE_DESCRIPTION("I2C-Bus SGI algorithm"); diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c index a591fe685f06..834967464814 100644 --- a/drivers/i2c/busses/i2c-elektor.c +++ b/drivers/i2c/busses/i2c-elektor.c @@ -293,7 +293,7 @@ static int __init i2c_pcfisa_init(void) static void i2c_pcfisa_exit(void) { - i2c_pcf_del_bus(&pcf_isa_ops); + i2c_del_adapter(&pcf_isa_ops); if (irq > 0) { disable_irq(irq); diff --git a/drivers/i2c/busses/i2c-hydra.c b/drivers/i2c/busses/i2c-hydra.c index 457d48a0ab9d..9832f773651d 100644 --- a/drivers/i2c/busses/i2c-hydra.c +++ b/drivers/i2c/busses/i2c-hydra.c @@ -146,7 +146,7 @@ static int __devinit hydra_probe(struct pci_dev *dev, static void __devexit hydra_remove(struct pci_dev *dev) { pdregw(hydra_bit_data.data, 0); /* clear SCLK_OE and SDAT_OE */ - i2c_bit_del_bus(&hydra_adap); + i2c_del_adapter(&hydra_adap); iounmap(hydra_bit_data.data); release_mem_region(pci_resource_start(dev, 0)+ offsetof(struct Hydra, CachePD), 4); diff --git a/drivers/i2c/busses/i2c-i810.c b/drivers/i2c/busses/i2c-i810.c index b66fb6bb1870..10c98bc88aa6 100644 --- a/drivers/i2c/busses/i2c-i810.c +++ b/drivers/i2c/busses/i2c-i810.c @@ -219,14 +219,14 @@ static int __devinit i810_probe(struct pci_dev *dev, const struct pci_device_id return retval; retval = i2c_bit_add_bus(&i810_ddc_adapter); if (retval) - i2c_bit_del_bus(&i810_i2c_adapter); + i2c_del_adapter(&i810_i2c_adapter); return retval; } static void __devexit i810_remove(struct pci_dev *dev) { - i2c_bit_del_bus(&i810_ddc_adapter); - i2c_bit_del_bus(&i810_i2c_adapter); + i2c_del_adapter(&i810_ddc_adapter); + i2c_del_adapter(&i810_i2c_adapter); iounmap(ioaddr); } diff --git a/drivers/i2c/busses/i2c-ixp2000.c b/drivers/i2c/busses/i2c-ixp2000.c index dd3f4cd3aa68..efa3ecc5522a 100644 --- a/drivers/i2c/busses/i2c-ixp2000.c +++ b/drivers/i2c/busses/i2c-ixp2000.c @@ -90,7 +90,7 @@ static int ixp2000_i2c_remove(struct platform_device *plat_dev) platform_set_drvdata(plat_dev, NULL); - i2c_bit_del_bus(&drv_data->adapter); + i2c_del_adapter(&drv_data->adapter); kfree(drv_data); diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c index 68fe863f9d54..08e89b83984a 100644 --- a/drivers/i2c/busses/i2c-ixp4xx.c +++ b/drivers/i2c/busses/i2c-ixp4xx.c @@ -91,7 +91,7 @@ static int ixp4xx_i2c_remove(struct platform_device *plat_dev) platform_set_drvdata(plat_dev, NULL); - i2c_bit_del_bus(&drv_data->adapter); + i2c_del_adapter(&drv_data->adapter); kfree(drv_data); diff --git a/drivers/i2c/busses/i2c-parport-light.c b/drivers/i2c/busses/i2c-parport-light.c index 5eb2bd294fd9..4bc42810b9aa 100644 --- a/drivers/i2c/busses/i2c-parport-light.c +++ b/drivers/i2c/busses/i2c-parport-light.c @@ -163,7 +163,7 @@ static void __exit i2c_parport_exit(void) if (adapter_parm[type].init.val) line_set(0, &adapter_parm[type].init); - i2c_bit_del_bus(&parport_adapter); + i2c_del_adapter(&parport_adapter); release_region(base, 3); } diff --git a/drivers/i2c/busses/i2c-parport.c b/drivers/i2c/busses/i2c-parport.c index 48a829431c7b..66696a40c7b5 100644 --- a/drivers/i2c/busses/i2c-parport.c +++ b/drivers/i2c/busses/i2c-parport.c @@ -218,7 +218,7 @@ static void i2c_parport_detach (struct parport *port) if (adapter_parm[type].init.val) line_set(port, 0, &adapter_parm[type].init); - i2c_bit_del_bus(&adapter->adapter); + i2c_del_adapter(&adapter->adapter); parport_unregister_device(adapter->pdev); if (prev) prev->next = adapter->next; diff --git a/drivers/i2c/busses/i2c-pca-isa.c b/drivers/i2c/busses/i2c-pca-isa.c index 407840b6a260..cc6536a19eca 100644 --- a/drivers/i2c/busses/i2c-pca-isa.c +++ b/drivers/i2c/busses/i2c-pca-isa.c @@ -156,7 +156,7 @@ static int __init pca_isa_init(void) static void pca_isa_exit(void) { - i2c_pca_del_bus(&pca_isa_ops); + i2c_del_adapter(&pca_isa_ops); if (irq > 0) { disable_irq(irq); diff --git a/drivers/i2c/busses/i2c-prosavage.c b/drivers/i2c/busses/i2c-prosavage.c index 7745e21874a8..07c1f1e27df1 100644 --- a/drivers/i2c/busses/i2c-prosavage.c +++ b/drivers/i2c/busses/i2c-prosavage.c @@ -212,7 +212,7 @@ static void prosavage_remove(struct pci_dev *dev) if (chip->i2c_bus[i].adap_ok == 0) continue; - ret = i2c_bit_del_bus(&chip->i2c_bus[i].adap); + ret = i2c_del_adapter(&chip->i2c_bus[i].adap); if (ret) { dev_err(&dev->dev, "%s not removed\n", chip->i2c_bus[i].adap.name); diff --git a/drivers/i2c/busses/i2c-savage4.c b/drivers/i2c/busses/i2c-savage4.c index 209f47ea1750..844b4ff90893 100644 --- a/drivers/i2c/busses/i2c-savage4.c +++ b/drivers/i2c/busses/i2c-savage4.c @@ -173,7 +173,7 @@ static int __devinit savage4_probe(struct pci_dev *dev, const struct pci_device_ static void __devexit savage4_remove(struct pci_dev *dev) { - i2c_bit_del_bus(&savage4_i2c_adapter); + i2c_del_adapter(&savage4_i2c_adapter); iounmap(ioaddr); } diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c index 910e200ad500..15d7e00e47e6 100644 --- a/drivers/i2c/busses/i2c-via.c +++ b/drivers/i2c/busses/i2c-via.c @@ -151,7 +151,7 @@ static int __devinit vt586b_probe(struct pci_dev *dev, const struct pci_device_i static void __devexit vt586b_remove(struct pci_dev *dev) { - i2c_bit_del_bus(&vt586b_adapter); + i2c_del_adapter(&vt586b_adapter); release_region(I2C_DIR, IOSPACE); pm_io_base = 0; } diff --git a/drivers/i2c/busses/i2c-voodoo3.c b/drivers/i2c/busses/i2c-voodoo3.c index 6c8d25183382..b0377b81744b 100644 --- a/drivers/i2c/busses/i2c-voodoo3.c +++ b/drivers/i2c/busses/i2c-voodoo3.c @@ -211,14 +211,14 @@ static int __devinit voodoo3_probe(struct pci_dev *dev, const struct pci_device_ return retval; retval = i2c_bit_add_bus(&voodoo3_ddc_adapter); if (retval) - i2c_bit_del_bus(&voodoo3_i2c_adapter); + i2c_del_adapter(&voodoo3_i2c_adapter); return retval; } static void __devexit voodoo3_remove(struct pci_dev *dev) { - i2c_bit_del_bus(&voodoo3_i2c_adapter); - i2c_bit_del_bus(&voodoo3_ddc_adapter); + i2c_del_adapter(&voodoo3_i2c_adapter); + i2c_del_adapter(&voodoo3_ddc_adapter); iounmap(ioaddr); } diff --git a/drivers/i2c/busses/scx200_i2c.c b/drivers/i2c/busses/scx200_i2c.c index 8ddbae4fafe6..6cd96e43aa72 100644 --- a/drivers/i2c/busses/scx200_i2c.c +++ b/drivers/i2c/busses/scx200_i2c.c @@ -116,7 +116,7 @@ static int scx200_i2c_init(void) static void scx200_i2c_cleanup(void) { - i2c_bit_del_bus(&scx200_i2c_ops); + i2c_del_adapter(&scx200_i2c_ops); } module_init(scx200_i2c_init); diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c index 13a617917bf2..fbb7f14ec509 100644 --- a/drivers/ieee1394/pcilynx.c +++ b/drivers/ieee1394/pcilynx.c @@ -1485,7 +1485,7 @@ static int __devinit add_card(struct pci_dev *dev, } - i2c_bit_del_bus(i2c_ad); + i2c_del_adapter(i2c_ad); kfree(i2c_ad); } } diff --git a/drivers/media/dvb/pluto2/pluto2.c b/drivers/media/dvb/pluto2/pluto2.c index 8e4ce101eb22..ffda71dfdd65 100644 --- a/drivers/media/dvb/pluto2/pluto2.c +++ b/drivers/media/dvb/pluto2/pluto2.c @@ -650,7 +650,7 @@ static int __devinit pluto2_probe(struct pci_dev *pdev, /* dvb */ ret = dvb_register_adapter(&pluto->dvb_adapter, DRIVER_NAME, THIS_MODULE, &pdev->dev); if (ret < 0) - goto err_i2c_bit_del_bus; + goto err_i2c_del_adapter; dvb_adapter = &pluto->dvb_adapter; @@ -712,8 +712,8 @@ err_dvb_dmx_release: dvb_dmx_release(dvbdemux); err_dvb_unregister_adapter: dvb_unregister_adapter(dvb_adapter); -err_i2c_bit_del_bus: - i2c_bit_del_bus(&pluto->i2c_adap); +err_i2c_del_adapter: + i2c_del_adapter(&pluto->i2c_adap); err_pluto_hw_exit: pluto_hw_exit(pluto); err_free_irq: @@ -748,7 +748,7 @@ static void __devexit pluto2_remove(struct pci_dev *pdev) dvb_dmxdev_release(&pluto->dmxdev); dvb_dmx_release(dvbdemux); dvb_unregister_adapter(dvb_adapter); - i2c_bit_del_bus(&pluto->i2c_adap); + i2c_del_adapter(&pluto->i2c_adap); pluto_hw_exit(pluto); free_irq(pdev->irq, pluto); pci_iounmap(pdev, pluto->io_mem); diff --git a/drivers/media/video/bt8xx/bttv-i2c.c b/drivers/media/video/bt8xx/bttv-i2c.c index 70de6c96e201..62b873076e09 100644 --- a/drivers/media/video/bt8xx/bttv-i2c.c +++ b/drivers/media/video/bt8xx/bttv-i2c.c @@ -479,11 +479,7 @@ int __devexit fini_bttv_i2c(struct bttv *btv) if (0 != btv->i2c_rc) return 0; - if (btv->use_i2c_hw) { - return i2c_del_adapter(&btv->c.i2c_adap); - } else { - return i2c_bit_del_bus(&btv->c.i2c_adap); - } + return i2c_del_adapter(&btv->c.i2c_adap); } /* diff --git a/drivers/media/video/cx88/cx88-core.c b/drivers/media/video/cx88/cx88-core.c index 4b655f2ef278..453af5e943ff 100644 --- a/drivers/media/video/cx88/cx88-core.c +++ b/drivers/media/video/cx88/cx88-core.c @@ -1153,7 +1153,7 @@ void cx88_core_put(struct cx88_core *core, struct pci_dev *pci) mutex_lock(&devlist); cx88_ir_fini(core); if (0 == core->i2c_rc) - i2c_bit_del_bus(&core->i2c_adap); + i2c_del_adapter(&core->i2c_adap); list_del(&core->devlist); iounmap(core->lmmio); cx88_devcount--; diff --git a/drivers/media/video/cx88/cx88-vp3054-i2c.c b/drivers/media/video/cx88/cx88-vp3054-i2c.c index 2b4f1970c7df..6068c9bf82cd 100644 --- a/drivers/media/video/cx88/cx88-vp3054-i2c.c +++ b/drivers/media/video/cx88/cx88-vp3054-i2c.c @@ -168,7 +168,7 @@ void vp3054_i2c_remove(struct cx8802_dev *dev) dev->core->board != CX88_BOARD_DNTV_LIVE_DVB_T_PRO) return; - i2c_bit_del_bus(&vp3054_i2c->adap); + i2c_del_adapter(&vp3054_i2c->adap); kfree(vp3054_i2c); } diff --git a/drivers/media/video/vino.c b/drivers/media/video/vino.c index 6b6dff4d236a..a373c142e742 100644 --- a/drivers/media/video/vino.c +++ b/drivers/media/video/vino.c @@ -782,7 +782,7 @@ static int vino_i2c_add_bus(void) static int vino_i2c_del_bus(void) { - return i2c_sgi_del_bus(&vino_i2c_adapter); + return i2c_del_adapter(&vino_i2c_adapter); } static int i2c_camera_command(unsigned int cmd, void *arg) diff --git a/drivers/media/video/zoran_card.c b/drivers/media/video/zoran_card.c index 653822ce391c..4d1eb2fba34a 100644 --- a/drivers/media/video/zoran_card.c +++ b/drivers/media/video/zoran_card.c @@ -849,7 +849,7 @@ zoran_register_i2c (struct zoran *zr) static void zoran_unregister_i2c (struct zoran *zr) { - i2c_bit_del_bus((&zr->i2c_adapter)); + i2c_del_adapter(&zr->i2c_adapter); } /* Check a zoran_params struct for correctness, insert default params */ diff --git a/drivers/video/aty/radeon_i2c.c b/drivers/video/aty/radeon_i2c.c index 869725a13c21..e7c5b219ad1b 100644 --- a/drivers/video/aty/radeon_i2c.c +++ b/drivers/video/aty/radeon_i2c.c @@ -120,19 +120,19 @@ void radeon_create_i2c_busses(struct radeonfb_info *rinfo) void radeon_delete_i2c_busses(struct radeonfb_info *rinfo) { if (rinfo->i2c[0].rinfo) - i2c_bit_del_bus(&rinfo->i2c[0].adapter); + i2c_del_adapter(&rinfo->i2c[0].adapter); rinfo->i2c[0].rinfo = NULL; if (rinfo->i2c[1].rinfo) - i2c_bit_del_bus(&rinfo->i2c[1].adapter); + i2c_del_adapter(&rinfo->i2c[1].adapter); rinfo->i2c[1].rinfo = NULL; if (rinfo->i2c[2].rinfo) - i2c_bit_del_bus(&rinfo->i2c[2].adapter); + i2c_del_adapter(&rinfo->i2c[2].adapter); rinfo->i2c[2].rinfo = NULL; if (rinfo->i2c[3].rinfo) - i2c_bit_del_bus(&rinfo->i2c[3].adapter); + i2c_del_adapter(&rinfo->i2c[3].adapter); rinfo->i2c[3].rinfo = NULL; } diff --git a/drivers/video/i810/i810-i2c.c b/drivers/video/i810/i810-i2c.c index b952e4504abe..961f4d404467 100644 --- a/drivers/video/i810/i810-i2c.c +++ b/drivers/video/i810/i810-i2c.c @@ -137,15 +137,15 @@ void i810_create_i2c_busses(struct i810fb_par *par) void i810_delete_i2c_busses(struct i810fb_par *par) { if (par->chan[0].par) - i2c_bit_del_bus(&par->chan[0].adapter); + i2c_del_adapter(&par->chan[0].adapter); par->chan[0].par = NULL; if (par->chan[1].par) - i2c_bit_del_bus(&par->chan[1].adapter); + i2c_del_adapter(&par->chan[1].adapter); par->chan[1].par = NULL; if (par->chan[2].par) - i2c_bit_del_bus(&par->chan[2].adapter); + i2c_del_adapter(&par->chan[2].adapter); par->chan[2].par = NULL; } diff --git a/drivers/video/intelfb/intelfb_i2c.c b/drivers/video/intelfb/intelfb_i2c.c index 5686e2164e39..33bc41f50540 100644 --- a/drivers/video/intelfb/intelfb_i2c.c +++ b/drivers/video/intelfb/intelfb_i2c.c @@ -188,11 +188,11 @@ void intelfb_delete_i2c_busses(struct intelfb_info *dinfo) for (i = 0; i < MAX_OUTPUTS; i++) { if (dinfo->output[i].i2c_bus.dinfo) { - i2c_bit_del_bus(&dinfo->output[i].i2c_bus.adapter); + i2c_del_adapter(&dinfo->output[i].i2c_bus.adapter); dinfo->output[i].i2c_bus.dinfo = NULL; } if (dinfo->output[i].ddc_bus.dinfo) { - i2c_bit_del_bus(&dinfo->output[i].ddc_bus.adapter); + i2c_del_adapter(&dinfo->output[i].ddc_bus.adapter); dinfo->output[i].ddc_bus.dinfo = NULL; } } diff --git a/drivers/video/matrox/i2c-matroxfb.c b/drivers/video/matrox/i2c-matroxfb.c index 795c1a99a680..797b42305b0f 100644 --- a/drivers/video/matrox/i2c-matroxfb.c +++ b/drivers/video/matrox/i2c-matroxfb.c @@ -124,7 +124,7 @@ static int i2c_bus_reg(struct i2c_bit_adapter* b, struct matrox_fb_info* minfo, static void i2c_bit_bus_del(struct i2c_bit_adapter* b) { if (b->initialized) { - i2c_bit_del_bus(&b->adapter); + i2c_del_adapter(&b->adapter); b->initialized = 0; } } diff --git a/drivers/video/nvidia/nv_i2c.c b/drivers/video/nvidia/nv_i2c.c index 442e85328341..8454adf2d178 100644 --- a/drivers/video/nvidia/nv_i2c.c +++ b/drivers/video/nvidia/nv_i2c.c @@ -147,15 +147,15 @@ void nvidia_create_i2c_busses(struct nvidia_par *par) void nvidia_delete_i2c_busses(struct nvidia_par *par) { if (par->chan[0].par) - i2c_bit_del_bus(&par->chan[0].adapter); + i2c_del_adapter(&par->chan[0].adapter); par->chan[0].par = NULL; if (par->chan[1].par) - i2c_bit_del_bus(&par->chan[1].adapter); + i2c_del_adapter(&par->chan[1].adapter); par->chan[1].par = NULL; if (par->chan[2].par) - i2c_bit_del_bus(&par->chan[2].adapter); + i2c_del_adapter(&par->chan[2].adapter); par->chan[2].par = NULL; } diff --git a/drivers/video/riva/rivafb-i2c.c b/drivers/video/riva/rivafb-i2c.c index c15b259af644..01b85e3b0ae1 100644 --- a/drivers/video/riva/rivafb-i2c.c +++ b/drivers/video/riva/rivafb-i2c.c @@ -144,15 +144,15 @@ void riva_create_i2c_busses(struct riva_par *par) void riva_delete_i2c_busses(struct riva_par *par) { if (par->chan[0].par) - i2c_bit_del_bus(&par->chan[0].adapter); + i2c_del_adapter(&par->chan[0].adapter); par->chan[0].par = NULL; if (par->chan[1].par) - i2c_bit_del_bus(&par->chan[1].adapter); + i2c_del_adapter(&par->chan[1].adapter); par->chan[1].par = NULL; if (par->chan[2].par) - i2c_bit_del_bus(&par->chan[2].adapter); + i2c_del_adapter(&par->chan[2].adapter); par->chan[2].par = NULL; } diff --git a/drivers/video/savage/savagefb-i2c.c b/drivers/video/savage/savagefb-i2c.c index cef5bf591cdf..1411f3b6a009 100644 --- a/drivers/video/savage/savagefb-i2c.c +++ b/drivers/video/savage/savagefb-i2c.c @@ -208,7 +208,7 @@ void savagefb_delete_i2c_busses(struct fb_info *info) struct savagefb_par *par = info->par; if (par->chan.par) - i2c_bit_del_bus(&par->chan.adapter); + i2c_del_adapter(&par->chan.adapter); par->chan.par = NULL; } diff --git a/include/linux/i2c-algo-bit.h b/include/linux/i2c-algo-bit.h index a99e28a86c3b..937da70cb4c4 100644 --- a/include/linux/i2c-algo-bit.h +++ b/include/linux/i2c-algo-bit.h @@ -44,6 +44,5 @@ struct i2c_algo_bit_data { }; int i2c_bit_add_bus(struct i2c_adapter *); -int i2c_bit_del_bus(struct i2c_adapter *); #endif /* _LINUX_I2C_ALGO_BIT_H */ diff --git a/include/linux/i2c-algo-pca.h b/include/linux/i2c-algo-pca.h index 226693e0d88b..fce47c051bb1 100644 --- a/include/linux/i2c-algo-pca.h +++ b/include/linux/i2c-algo-pca.h @@ -10,6 +10,5 @@ struct i2c_algo_pca_data { }; int i2c_pca_add_bus(struct i2c_adapter *); -int i2c_pca_del_bus(struct i2c_adapter *); #endif /* _LINUX_I2C_ALGO_PCA_H */ diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h index 7d7045cbb344..994eb86f882c 100644 --- a/include/linux/i2c-algo-pcf.h +++ b/include/linux/i2c-algo-pcf.h @@ -39,6 +39,5 @@ struct i2c_algo_pcf_data { }; int i2c_pcf_add_bus(struct i2c_adapter *); -int i2c_pcf_del_bus(struct i2c_adapter *); #endif /* _LINUX_I2C_ALGO_PCF_H */ diff --git a/include/linux/i2c-algo-sgi.h b/include/linux/i2c-algo-sgi.h index 4a0113d64064..3b7715024e69 100644 --- a/include/linux/i2c-algo-sgi.h +++ b/include/linux/i2c-algo-sgi.h @@ -22,6 +22,5 @@ struct i2c_algo_sgi_data { }; int i2c_sgi_add_bus(struct i2c_adapter *); -int i2c_sgi_del_bus(struct i2c_adapter *); #endif /* I2C_ALGO_SGI_H */ -- cgit v1.2.3 From d10f73480b991da2aa1c000ed38eda3e4a987292 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Sun, 10 Dec 2006 23:26:16 -0600 Subject: [PPC] Fix compile failure do to introduction of PHY_POLL PHY_POLL is defined in include it in so board code will have it defined. Signed-off-by: Kumar Gala --- include/linux/fsl_devices.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h index 3da29e2d524a..abb64c437f6f 100644 --- a/include/linux/fsl_devices.h +++ b/include/linux/fsl_devices.h @@ -19,6 +19,7 @@ #define _FSL_DEVICE_H_ #include +#include /* * Some conventions on how we handle peripherals on Freescale chips -- cgit v1.2.3 From 8993780a6e44fb4e7ed34e33458506a775356c6e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 11 Dec 2006 09:28:46 -0800 Subject: Make SLES9 "get_kernel_version" work on the kernel binary again As reported by Andy Whitcroft, at least the SLES9 initrd build process depends on getting the kernel version from the kernel binary. It does that by simply trawling the binary and looking for the signature of the "linux_banner" string (the string "Linux version " to be exact. Which is really broken in itself, but whatever..) That got broken when the string was changed to allow /proc/version to change the UTS release information dynamically, and "get_kernel_version" thus returned "%s" (see commit a2ee8649ba6d71416712e798276bf7c40b64e6e5: "[PATCH] Fix linux banner utsname information"). This just restores "linux_banner" as a static string, which should fix the version finding. And /proc/version simply uses a different string. To avoid wasting even that miniscule amount of memory, the early boot string should really be marked __initdata, but that just causes the same bug in SLES9 to re-appear, since it will then find other occurrences of "Linux version " first. Cc: Andy Whitcroft Acked-by: Herbert Poetzl Cc: Andi Kleen Cc: Andrew Morton Cc: Steve Fox Acked-by: Olaf Hering Signed-off-by: Linus Torvalds --- fs/proc/proc_misc.c | 12 ++++++++++-- include/linux/kernel.h | 2 -- init/main.c | 8 +++++++- init/version.c | 5 ----- 4 files changed, 17 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index dc3e580d1dca..92ea7743fe8f 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include #include @@ -253,8 +254,15 @@ static int version_read_proc(char *page, char **start, off_t off, { int len; - len = sprintf(page, linux_banner, - utsname()->release, utsname()->version); + /* FIXED STRING! Don't touch! */ + len = snprintf(page, PAGE_SIZE, + "%s version %s" + " (" LINUX_COMPILE_BY "@" LINUX_COMPILE_HOST ")" + " (" LINUX_COMPILER ")" + " %s\n", + utsname()->sysname, + utsname()->release, + utsname()->version); return proc_calc_metrics(page, start, off, count, eof, len); } diff --git a/include/linux/kernel.h b/include/linux/kernel.h index e8bfac34d2ba..b0c4a05a4b0c 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -17,8 +17,6 @@ #include #include -extern const char linux_banner[]; - #define INT_MAX ((int)(~0U>>1)) #define INT_MIN (-INT_MAX - 1) #define UINT_MAX (~0U) diff --git a/init/main.c b/init/main.c index 036f97c0c34c..fcd9ddc3ccf5 100644 --- a/init/main.c +++ b/init/main.c @@ -483,6 +483,12 @@ void __init __attribute__((weak)) smp_setup_processor_id(void) { } +static const char linux_banner[] = + "Linux version " UTS_RELEASE + " (" LINUX_COMPILE_BY "@" LINUX_COMPILE_HOST ")" + " (" LINUX_COMPILER ")" + " " UTS_VERSION "\n"; + asmlinkage void __init start_kernel(void) { char * command_line; @@ -509,7 +515,7 @@ asmlinkage void __init start_kernel(void) boot_cpu_init(); page_address_init(); printk(KERN_NOTICE); - printk(linux_banner, UTS_RELEASE, UTS_VERSION); + printk(linux_banner); setup_arch(&command_line); unwind_setup(); setup_per_cpu_areas(); diff --git a/init/version.c b/init/version.c index 2a5dfcd1c2e6..9d96d36501ca 100644 --- a/init/version.c +++ b/init/version.c @@ -33,8 +33,3 @@ struct uts_namespace init_uts_ns = { }, }; EXPORT_SYMBOL_GPL(init_uts_ns); - -const char linux_banner[] = - "Linux version %s (" LINUX_COMPILE_BY "@" - LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") %s\n"; - -- cgit v1.2.3 From 8d610dd52dd1da696e199e4b4545f33a2a5de5c6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 11 Dec 2006 12:12:04 -0800 Subject: Make sure we populate the initroot filesystem late enough We should not initialize rootfs before all the core initializers have run. So do it as a separate stage just before starting the regular driver initializers. Signed-off-by: Linus Torvalds --- include/asm-generic/vmlinux.lds.h | 1 + include/linux/init.h | 1 + init/initramfs.c | 6 ++++-- init/main.c | 7 ------- 4 files changed, 6 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 6e9fcebbf89f..7437ccaada77 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -242,6 +242,7 @@ *(.initcall4s.init) \ *(.initcall5.init) \ *(.initcall5s.init) \ + *(.initcallrootfs.init) \ *(.initcall6.init) \ *(.initcall6s.init) \ *(.initcall7.init) \ diff --git a/include/linux/init.h b/include/linux/init.h index 5eb5d24b7680..5a593a1dec1e 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -111,6 +111,7 @@ extern void setup_arch(char **); #define subsys_initcall_sync(fn) __define_initcall("4s",fn,4s) #define fs_initcall(fn) __define_initcall("5",fn,5) #define fs_initcall_sync(fn) __define_initcall("5s",fn,5s) +#define rootfs_initcall(fn) __define_initcall("rootfs",fn,rootfs) #define device_initcall(fn) __define_initcall("6",fn,6) #define device_initcall_sync(fn) __define_initcall("6s",fn,6s) #define late_initcall(fn) __define_initcall("7",fn,7) diff --git a/init/initramfs.c b/init/initramfs.c index 85f04037ade1..4fa0f7977de1 100644 --- a/init/initramfs.c +++ b/init/initramfs.c @@ -526,7 +526,7 @@ static void __init free_initrd(void) #endif -void __init populate_rootfs(void) +static int __init populate_rootfs(void) { char *err = unpack_to_rootfs(__initramfs_start, __initramfs_end - __initramfs_start, 0); @@ -544,7 +544,7 @@ void __init populate_rootfs(void) unpack_to_rootfs((char *)initrd_start, initrd_end - initrd_start, 0); free_initrd(); - return; + return 0; } printk("it isn't (%s); looks like an initrd\n", err); fd = sys_open("/initrd.image", O_WRONLY|O_CREAT, 0700); @@ -565,4 +565,6 @@ void __init populate_rootfs(void) #endif } #endif + return 0; } +rootfs_initcall(populate_rootfs); diff --git a/init/main.c b/init/main.c index fcd9ddc3ccf5..e3f0bb20b4dd 100644 --- a/init/main.c +++ b/init/main.c @@ -94,7 +94,6 @@ extern void pidmap_init(void); extern void prio_tree_init(void); extern void radix_tree_init(void); extern void free_initmem(void); -extern void populate_rootfs(void); extern void driver_init(void); extern void prepare_namespace(void); #ifdef CONFIG_ACPI @@ -745,12 +744,6 @@ static int init(void * unused) cpuset_init_smp(); - /* - * Do this before initcalls, because some drivers want to access - * firmware files. - */ - populate_rootfs(); - do_basic_setup(); /* -- cgit v1.2.3 From 1a21e49a8d60f588c1276f765198b14d5688a778 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Sun, 10 Dec 2006 00:02:12 -0200 Subject: [DCCP] ccid3: Finer-grained resolution of sending rates This patch * resolves a bug where packets smaller than 32/64 bytes resulted in sending rates of 0 * supports all sending rates from 1/64 bytes/second up to 4Gbyte/second * simplifies the present overflow problems in calculations Current sending rate X and the cached value X_recv of the receiver-estimated sending rate are both scaled by 64 (2^6) in order to * cope with low sending rates (minimally 1 byte/second) * allow upgrading to use a packets-per-second implementation of CCID 3 * avoid calculation errors due to integer arithmetic cut-off The patch implements a revised strategy from http://www.mail-archive.com/dccp@vger.kernel.org/msg01040.html The only difference with regard to that strategy is that t_ipi is already used in the calculation of the nofeedback timeout, which saves one division. Signed-off-by: Gerrit Renker Acked-by: Ian McDonald Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/tfrc.h | 8 ++++-- net/dccp/ccids/ccid3.c | 76 ++++++++++++++++++++++++++++++-------------------- net/dccp/ccids/ccid3.h | 33 +++++++++++++++++----- 3 files changed, 78 insertions(+), 39 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tfrc.h b/include/linux/tfrc.h index 31a9b25276fe..8a8462b4a4dd 100644 --- a/include/linux/tfrc.h +++ b/include/linux/tfrc.h @@ -37,10 +37,14 @@ struct tfrc_rx_info { * @tfrctx_p: current loss event rate (5.4) * @tfrctx_rto: estimate of RTO, equals 4*RTT (4.3) * @tfrctx_ipi: inter-packet interval (4.6) + * + * Note: X and X_recv are both maintained in units of 64 * bytes/second. This + * enables a finer resolution of sending rates and avoids problems with + * integer arithmetic; u32 is not sufficient as scaling consumes 6 bits. */ struct tfrc_tx_info { - __u32 tfrctx_x; - __u32 tfrctx_x_recv; + __u64 tfrctx_x; + __u64 tfrctx_x_recv; __u32 tfrctx_x_calc; __u32 tfrctx_rtt; __u32 tfrctx_p; diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index c54663f21fdd..aa355d4cfc8a 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -108,8 +108,9 @@ static inline void ccid3_update_send_time(struct ccid3_hc_tx_sock *hctx) { timeval_sub_usecs(&hctx->ccid3hctx_t_nom, hctx->ccid3hctx_t_ipi); - /* Calculate new t_ipi (inter packet interval) by t_ipi = s / X_inst */ - hctx->ccid3hctx_t_ipi = usecs_div(hctx->ccid3hctx_s, hctx->ccid3hctx_x); + /* Calculate new t_ipi = s / X_inst (X_inst is in 64 * bytes/second) */ + hctx->ccid3hctx_t_ipi = scaled_div(hctx->ccid3hctx_s, + hctx->ccid3hctx_x >> 6); /* Update nominal send time with regard to the new t_ipi */ timeval_add_usecs(&hctx->ccid3hctx_t_nom, hctx->ccid3hctx_t_ipi); @@ -128,26 +129,33 @@ static inline void ccid3_update_send_time(struct ccid3_hc_tx_sock *hctx) * X = max(min(2 * X, 2 * X_recv), s / R); * tld = now; * + * Note: X and X_recv are both stored in units of 64 * bytes/second, to support + * fine-grained resolution of sending rates. This requires scaling by 2^6 + * throughout the code. Only X_calc is unscaled (in bytes/second). + * * If X has changed, we also update the scheduled send time t_now, * the inter-packet interval t_ipi, and the delta value. - */ + */ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) { struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); - const __u32 old_x = hctx->ccid3hctx_x; + const __u64 old_x = hctx->ccid3hctx_x; if (hctx->ccid3hctx_p > 0) { - hctx->ccid3hctx_x = max_t(u32, min(hctx->ccid3hctx_x_calc, - hctx->ccid3hctx_x_recv * 2), - hctx->ccid3hctx_s / TFRC_T_MBI); + + hctx->ccid3hctx_x = min_t(u64, hctx->ccid3hctx_x_calc << 6, + hctx->ccid3hctx_x_recv * 2 ); + hctx->ccid3hctx_x = max_t(u64, hctx->ccid3hctx_x, + (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) >= hctx->ccid3hctx_rtt) { - hctx->ccid3hctx_x = max(min(hctx->ccid3hctx_x_recv, - hctx->ccid3hctx_x ) * 2, - usecs_div(hctx->ccid3hctx_s, - hctx->ccid3hctx_rtt) ); + + hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, + hctx->ccid3hctx_x_recv), + scaled_div(hctx->ccid3hctx_s << 6, + hctx->ccid3hctx_rtt )); hctx->ccid3hctx_t_ld = *now; } @@ -194,13 +202,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) case TFRC_SSTATE_NO_FBACK: /* RFC 3448, 4.4: Halve send rate directly */ hctx->ccid3hctx_x = max_t(u32, hctx->ccid3hctx_x / 2, - hctx->ccid3hctx_s / TFRC_T_MBI); + (hctx->ccid3hctx_s << 6)/TFRC_T_MBI); - ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %d " + ccid3_pr_debug("%s, sk=%p, state=%s, updated tx rate to %u " "bytes/s\n", dccp_role(sk), sk, ccid3_tx_state_name(hctx->ccid3hctx_state), - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); /* The value of R is still undefined and so we can not recompute * the timout value. Keep initial value as per [RFC 4342, 5]. */ t_nfb = TFRC_INITIAL_TIMEOUT; @@ -209,11 +217,11 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) case TFRC_SSTATE_FBACK: /* * Check if IDLE since last timeout and recv rate is less than - * 4 packets per RTT + * 4 packets (in units of 64*bytes/sec) per RTT */ if (!hctx->ccid3hctx_idle || - (hctx->ccid3hctx_x_recv >= - 4 * usecs_div(hctx->ccid3hctx_s, hctx->ccid3hctx_rtt))) { + (hctx->ccid3hctx_x_recv >= 4 * + scaled_div(hctx->ccid3hctx_s << 6, hctx->ccid3hctx_rtt))) { struct timeval now; ccid3_pr_debug("%s, sk=%p, state=%s, not idle\n", @@ -227,17 +235,23 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) * X_recv = max(X_recv / 2, s / (2 * t_mbi)); * Else * X_recv = X_calc / 4; + * + * Note that X_recv is scaled by 2^6 while X_calc is not */ BUG_ON(hctx->ccid3hctx_p && !hctx->ccid3hctx_x_calc); if (hctx->ccid3hctx_p == 0 || - hctx->ccid3hctx_x_calc > 2 * hctx->ccid3hctx_x_recv) { - hctx->ccid3hctx_x_recv = max_t(u32, hctx->ccid3hctx_x_recv / 2, - hctx->ccid3hctx_s / (2 * TFRC_T_MBI)); + hctx->ccid3hctx_x_calc > (hctx->ccid3hctx_x_recv >> 5)) { + + hctx->ccid3hctx_x_recv = + max_t(u64, hctx->ccid3hctx_x_recv / 2, + (hctx->ccid3hctx_s << 6) / + (2*TFRC_T_MBI)); + if (hctx->ccid3hctx_p == 0) dccp_timestamp(sk, &now); } else - hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc / 4; + hctx->ccid3hctx_x_recv = hctx->ccid3hctx_x_calc << 4; /* Now recalculate X [RFC 3448, 4.3, step (4)] */ ccid3_hc_tx_update_x(sk, &now); @@ -315,9 +329,9 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_t_last_win_count = now; ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); - /* Set initial sending rate to 1 packet per second */ + /* Set initial sending rate X/s to 1pps (X is scaled by 2^6) */ ccid3_hc_tx_update_s(hctx, skb->len); - hctx->ccid3hctx_x = hctx->ccid3hctx_s; + hctx->ccid3hctx_x = hctx->ccid3hctx_s << 6; /* First timeout, according to [RFC 3448, 4.2], is 1 second */ hctx->ccid3hctx_t_ipi = USEC_PER_SEC; @@ -438,8 +452,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) return; } - /* Update receive rate */ - hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate; + /* Update receive rate in units of 64 * bytes/second */ + hctx->ccid3hctx_x_recv = opt_recv->ccid3or_receive_rate << 6; /* Update loss event rate */ pinv = opt_recv->ccid3or_loss_event_rate; @@ -475,12 +489,14 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * q is a constant, RFC 3448 recomments 0.9 */ if (hctx->ccid3hctx_state == TFRC_SSTATE_NO_FBACK) { - /* Use Larger Initial Windows [RFC 4342, sec. 5] - * We deviate in that we use `s' instead of `MSS'. */ + /* + * Larger Initial Windows [RFC 4342, sec. 5] + * We deviate in that we use `s' instead of `MSS'. + */ u16 w_init = min( 4 * hctx->ccid3hctx_s, max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; - hctx->ccid3hctx_x = usecs_div(w_init, r_sample); + hctx->ccid3hctx_x = scaled_div(w_init<< 6, r_sample); hctx->ccid3hctx_t_ld = now; ccid3_update_send_time(hctx); @@ -488,7 +504,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_pr_debug("%s(%p), s=%u, w_init=%u, " "R_sample=%ldus, X=%u\n", dccp_role(sk), sk, hctx->ccid3hctx_s, w_init, r_sample, - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_FBACK); } else { @@ -508,7 +524,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) sk, hctx->ccid3hctx_rtt, r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, - hctx->ccid3hctx_x); + (unsigned)(hctx->ccid3hctx_x >> 6)); } /* unschedule no feedback timer */ diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 07596d704ef9..cd4fc542f73e 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -75,14 +75,14 @@ enum ccid3_hc_tx_states { /** struct ccid3_hc_tx_sock - CCID3 sender half-connection socket * - * @ccid3hctx_x - Current sending rate - * @ccid3hctx_x_recv - Receive rate - * @ccid3hctx_x_calc - Calculated send rate (RFC 3448, 3.1) + * @ccid3hctx_x - Current sending rate in 64 * bytes per second + * @ccid3hctx_x_recv - Receive rate in 64 * bytes per second + * @ccid3hctx_x_calc - Calculated rate in bytes per second * @ccid3hctx_rtt - Estimate of current round trip time in usecs * @ccid3hctx_p - Current loss event rate (0-1) scaled by 1000000 - * @ccid3hctx_s - Packet size - * @ccid3hctx_t_rto - Retransmission Timeout (RFC 3448, 3.1) - * @ccid3hctx_t_ipi - Interpacket (send) interval (RFC 3448, 4.6) + * @ccid3hctx_s - Packet size in bytes + * @ccid3hctx_t_rto - Nofeedback Timer setting in usecs + * @ccid3hctx_t_ipi - Interpacket (send) interval (RFC 3448, 4.6) in usecs * @ccid3hctx_state - Sender state, one of %ccid3_hc_tx_states * @ccid3hctx_last_win_count - Last window counter sent * @ccid3hctx_t_last_win_count - Timestamp of earliest packet @@ -91,7 +91,7 @@ enum ccid3_hc_tx_states { * @ccid3hctx_idle - Flag indicating that sender is idling * @ccid3hctx_t_ld - Time last doubled during slow start * @ccid3hctx_t_nom - Nominal send time of next packet - * @ccid3hctx_delta - Send timer delta + * @ccid3hctx_delta - Send timer delta (RFC 3448, 4.6) in usecs * @ccid3hctx_hist - Packet history * @ccid3hctx_options_received - Parsed set of retrieved options */ @@ -171,4 +171,23 @@ static inline struct ccid3_hc_rx_sock *ccid3_hc_rx_sk(const struct sock *sk) return ccid_priv(dccp_sk(sk)->dccps_hc_rx_ccid); } +static inline u64 scaled_div(u64 a, u32 b) +{ + BUG_ON(b==0); + a *= 1000000; + do_div(a, b); + return a; +} + +static inline u32 scaled_div32(u64 a, u32 b) +{ + u64 result = scaled_div(a, b); + + if (result > UINT_MAX) { + DCCP_CRIT("Overflow: a(%llu)/b(%u) > ~0U", + (unsigned long long)a, b); + return UINT_MAX; + } + return result; +} #endif /* _DCCP_CCID3_H_ */ -- cgit v1.2.3 From 8109b02b5397ed52a32c116163a62a34f4768b26 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 10 Dec 2006 16:01:18 -0200 Subject: [DCCP]: Whitespace cleanups That accumulated over the last months hackaton, shame on me for not using git-apply whitespace helping hand, will do that from now on. Signed-off-by: Arnaldo Carvalho de Melo --- include/linux/dccp.h | 26 ++++++------ net/dccp/ackvec.c | 4 +- net/dccp/ccids/ccid2.c | 12 +++--- net/dccp/ccids/ccid3.c | 111 +++++++++++++++++++++++++++---------------------- net/dccp/ccids/ccid3.h | 22 +++++----- net/dccp/feat.c | 6 +-- net/dccp/input.c | 26 ++++++------ net/dccp/ipv4.c | 26 ++++++------ net/dccp/ipv6.c | 24 +++++------ net/dccp/minisocks.c | 2 +- net/dccp/output.c | 7 ++-- net/dccp/proto.c | 6 +-- net/dccp/timer.c | 14 +++---- 13 files changed, 149 insertions(+), 137 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dccp.h b/include/linux/dccp.h index ed6cc8962d87..1cb054bd93f2 100644 --- a/include/linux/dccp.h +++ b/include/linux/dccp.h @@ -176,20 +176,20 @@ enum { }; /* DCCP features (RFC 4340 section 6.4) */ - enum { - DCCPF_RESERVED = 0, - DCCPF_CCID = 1, +enum { + DCCPF_RESERVED = 0, + DCCPF_CCID = 1, DCCPF_SHORT_SEQNOS = 2, /* XXX: not yet implemented */ - DCCPF_SEQUENCE_WINDOW = 3, + DCCPF_SEQUENCE_WINDOW = 3, DCCPF_ECN_INCAPABLE = 4, /* XXX: not yet implemented */ - DCCPF_ACK_RATIO = 5, - DCCPF_SEND_ACK_VECTOR = 6, - DCCPF_SEND_NDP_COUNT = 7, + DCCPF_ACK_RATIO = 5, + DCCPF_SEND_ACK_VECTOR = 6, + DCCPF_SEND_NDP_COUNT = 7, DCCPF_MIN_CSUM_COVER = 8, DCCPF_DATA_CHECKSUM = 9, /* XXX: not yet implemented */ - /* 10-127 reserved */ - DCCPF_MIN_CCID_SPECIFIC = 128, - DCCPF_MAX_CCID_SPECIFIC = 255, + /* 10-127 reserved */ + DCCPF_MIN_CCID_SPECIFIC = 128, + DCCPF_MAX_CCID_SPECIFIC = 255, }; /* this structure is argument to DCCP_SOCKOPT_CHANGE_X */ @@ -427,7 +427,7 @@ struct dccp_service_list { }; #define DCCP_SERVICE_INVALID_VALUE htonl((__u32)-1) -#define DCCP_SERVICE_CODE_IS_ABSENT 0 +#define DCCP_SERVICE_CODE_IS_ABSENT 0 static inline int dccp_list_has_service(const struct dccp_service_list *sl, const __be32 service) @@ -436,7 +436,7 @@ static inline int dccp_list_has_service(const struct dccp_service_list *sl, u32 i = sl->dccpsl_nr; while (i--) if (sl->dccpsl_list[i] == service) - return 1; + return 1; } return 0; } @@ -511,7 +511,7 @@ struct dccp_sock { __u8 dccps_hc_tx_insert_options:1; struct timer_list dccps_xmit_timer; }; - + static inline struct dccp_sock *dccp_sk(const struct sock *sk) { return (struct dccp_sock *)sk; diff --git a/net/dccp/ackvec.c b/net/dccp/ackvec.c index 1f4727ddbdbf..a086c6312d3b 100644 --- a/net/dccp/ackvec.c +++ b/net/dccp/ackvec.c @@ -223,7 +223,7 @@ static inline int dccp_ackvec_set_buf_head_state(struct dccp_ackvec *av, gap = -new_head; } new_head += DCCP_MAX_ACKVEC_LEN; - } + } av->dccpav_buf_head = new_head; @@ -336,7 +336,7 @@ out_duplicate: void dccp_ackvector_print(const u64 ackno, const unsigned char *vector, int len) { dccp_pr_debug_cat("ACK vector len=%d, ackno=%llu |", len, - (unsigned long long)ackno); + (unsigned long long)ackno); while (len--) { const u8 state = (*vector & DCCP_ACKVEC_STATE_MASK) >> 6; diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 2555be8f4790..fd38b05d6f79 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -351,7 +351,7 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) while (seqp != hctx->ccid2hctx_seqh) { ccid2_pr_debug("out seq=%llu acked=%d time=%lu\n", - (unsigned long long)seqp->ccid2s_seq, + (unsigned long long)seqp->ccid2s_seq, seqp->ccid2s_acked, seqp->ccid2s_sent); seqp = seqp->ccid2s_next; } @@ -473,7 +473,7 @@ static inline void ccid2_new_ack(struct sock *sk, /* first measurement */ if (hctx->ccid2hctx_srtt == -1) { ccid2_pr_debug("R: %lu Time=%lu seq=%llu\n", - r, jiffies, + r, jiffies, (unsigned long long)seqp->ccid2s_seq); ccid2_change_srtt(hctx, r); hctx->ccid2hctx_rttvar = r >> 1; @@ -518,8 +518,8 @@ static inline void ccid2_new_ack(struct sock *sk, hctx->ccid2hctx_lastrtt = jiffies; ccid2_pr_debug("srtt: %ld rttvar: %ld rto: %ld (HZ=%d) R=%lu\n", - hctx->ccid2hctx_srtt, hctx->ccid2hctx_rttvar, - hctx->ccid2hctx_rto, HZ, r); + hctx->ccid2hctx_srtt, hctx->ccid2hctx_rttvar, + hctx->ccid2hctx_rto, HZ, r); hctx->ccid2hctx_sent = 0; } @@ -667,9 +667,9 @@ static void ccid2_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) /* new packet received or marked */ if (state != DCCP_ACKVEC_STATE_NOT_RECEIVED && !seqp->ccid2s_acked) { - if (state == + if (state == DCCP_ACKVEC_STATE_ECN_MARKED) { - ccid2_congestion_event(hctx, + ccid2_congestion_event(hctx, seqp); } else ccid2_new_ack(sk, seqp, diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c index 5e817e2f1ebc..fa6b75372ed7 100644 --- a/net/dccp/ccids/ccid3.c +++ b/net/dccp/ccids/ccid3.c @@ -124,18 +124,18 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) if (hctx->ccid3hctx_p > 0) { hctx->ccid3hctx_x = min(((__u64)hctx->ccid3hctx_x_calc) << 6, - hctx->ccid3hctx_x_recv * 2 ); + hctx->ccid3hctx_x_recv * 2); hctx->ccid3hctx_x = max(hctx->ccid3hctx_x, (((__u64)hctx->ccid3hctx_s) << 6) / - TFRC_T_MBI); + TFRC_T_MBI); } else if (timeval_delta(now, &hctx->ccid3hctx_t_ld) - - (suseconds_t)hctx->ccid3hctx_rtt >= 0 ) { + (suseconds_t)hctx->ccid3hctx_rtt >= 0) { hctx->ccid3hctx_x = max(2 * min(hctx->ccid3hctx_x, hctx->ccid3hctx_x_recv), scaled_div(((__u64)hctx->ccid3hctx_s) << 6, - hctx->ccid3hctx_rtt ) ); + hctx->ccid3hctx_rtt)); hctx->ccid3hctx_t_ld = *now; } @@ -144,8 +144,8 @@ static void ccid3_hc_tx_update_x(struct sock *sk, struct timeval *now) } /* - * Track the mean packet size `s' (cf. RFC 4342, 5.3 and RFC 3448, 4.1) - * @len: DCCP packet payload size in bytes + * Track the mean packet size `s' (cf. RFC 4342, 5.3 and RFC 3448, 4.1) + * @len: DCCP packet payload size in bytes */ static inline void ccid3_hc_tx_update_s(struct ccid3_hc_tx_sock *hctx, int len) { @@ -163,8 +163,8 @@ static inline void ccid3_hc_tx_update_s(struct ccid3_hc_tx_sock *hctx, int len) } /* - * Update Window Counter using the algorithm from [RFC 4342, 8.1]. - * The algorithm is not applicable if RTT < 4 microseconds. + * Update Window Counter using the algorithm from [RFC 4342, 8.1]. + * The algorithm is not applicable if RTT < 4 microseconds. */ static inline void ccid3_hc_tx_update_win_count(struct ccid3_hc_tx_sock *hctx, struct timeval *now) @@ -228,12 +228,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) */ if (!hctx->ccid3hctx_idle || (hctx->ccid3hctx_x_recv >= 4 * - scaled_div(((__u64)hctx->ccid3hctx_s) << 6, hctx->ccid3hctx_rtt))) { + scaled_div(((__u64)hctx->ccid3hctx_s) << 6, + hctx->ccid3hctx_rtt))) { struct timeval now; ccid3_pr_debug("%s(%p, state=%s), not idle\n", dccp_role(sk), sk, - ccid3_tx_state_name(hctx->ccid3hctx_state)); + ccid3_tx_state_name(hctx->ccid3hctx_state)); /* * Modify the cached value of X_recv [RFC 3448, 4.4] @@ -248,12 +249,13 @@ static void ccid3_hc_tx_no_feedback_timer(unsigned long data) BUG_ON(hctx->ccid3hctx_p && !hctx->ccid3hctx_x_calc); if (hctx->ccid3hctx_p == 0 || - hctx->ccid3hctx_x_calc > (hctx->ccid3hctx_x_recv >> 5)) { + (hctx->ccid3hctx_x_calc > + (hctx->ccid3hctx_x_recv >> 5))) { hctx->ccid3hctx_x_recv = max(hctx->ccid3hctx_x_recv / 2, (((__u64)hctx->ccid3hctx_s) << 6) / - (2*TFRC_T_MBI)); + (2 * TFRC_T_MBI)); if (hctx->ccid3hctx_p == 0) dccp_timestamp(sk, &now); @@ -316,7 +318,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) switch (hctx->ccid3hctx_state) { case TFRC_SSTATE_NO_SENT: sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, - jiffies + usecs_to_jiffies(TFRC_INITIAL_TIMEOUT)); + (jiffies + + usecs_to_jiffies(TFRC_INITIAL_TIMEOUT))); hctx->ccid3hctx_last_win_count = 0; hctx->ccid3hctx_t_last_win_count = now; ccid3_hc_tx_set_state(sk, TFRC_SSTATE_NO_FBACK); @@ -338,7 +341,7 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) case TFRC_SSTATE_FBACK: delay = timeval_delta(&hctx->ccid3hctx_t_nom, &now); /* - * Scheduling of packet transmissions [RFC 3448, 4.6] + * Scheduling of packet transmissions [RFC 3448, 4.6] * * if (t_now > t_nom - delta) * // send the packet now @@ -365,7 +368,8 @@ static int ccid3_hc_tx_send_packet(struct sock *sk, struct sk_buff *skb) return 0; } -static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, unsigned int len) +static void ccid3_hc_tx_packet_sent(struct sock *sk, int more, + unsigned int len) { struct ccid3_hc_tx_sock *hctx = ccid3_hc_tx_sk(sk); struct timeval now; @@ -415,12 +419,12 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) case TFRC_SSTATE_FBACK: /* get packet from history to look up t_recvdata */ packet = dccp_tx_hist_find_entry(&hctx->ccid3hctx_hist, - DCCP_SKB_CB(skb)->dccpd_ack_seq); + DCCP_SKB_CB(skb)->dccpd_ack_seq); if (unlikely(packet == NULL)) { DCCP_WARN("%s(%p), seqno %llu(%s) doesn't exist " "in history!\n", dccp_role(sk), sk, (unsigned long long)DCCP_SKB_CB(skb)->dccpd_ack_seq, - dccp_packet_name(DCCP_SKB_CB(skb)->dccpd_type)); + dccp_packet_name(DCCP_SKB_CB(skb)->dccpd_type)); return; } @@ -433,13 +437,13 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) if (pinv == ~0U || pinv == 0) /* see RFC 4342, 8.5 */ hctx->ccid3hctx_p = 0; else /* can not exceed 100% */ - hctx->ccid3hctx_p = 1000000 / pinv; + hctx->ccid3hctx_p = 1000000 / pinv; dccp_timestamp(sk, &now); /* * Calculate new round trip sample as per [RFC 3448, 4.3] by - * R_sample = (now - t_recvdata) - t_elapsed + * R_sample = (now - t_recvdata) - t_elapsed */ r_sample = timeval_delta(&now, &packet->dccphtx_tstamp); t_elapsed = dp->dccps_options_received.dccpor_elapsed_time * 10; @@ -465,7 +469,7 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) * Larger Initial Windows [RFC 4342, sec. 5] * We deviate in that we use `s' instead of `MSS'. */ - __u64 w_init = min( 4 * hctx->ccid3hctx_s, + __u64 w_init = min(4 * hctx->ccid3hctx_s, max(2 * hctx->ccid3hctx_s, 4380)); hctx->ccid3hctx_rtt = r_sample; hctx->ccid3hctx_x = scaled_div(w_init << 6, r_sample); @@ -475,13 +479,14 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_pr_debug("%s(%p), s=%u, w_init=%llu, " "R_sample=%dus, X=%u\n", dccp_role(sk), - sk, hctx->ccid3hctx_s, w_init, (int)r_sample, + sk, hctx->ccid3hctx_s, w_init, + (int)r_sample, (unsigned)(hctx->ccid3hctx_x >> 6)); ccid3_hc_tx_set_state(sk, TFRC_SSTATE_FBACK); } else { hctx->ccid3hctx_rtt = (9 * hctx->ccid3hctx_rtt + - (u32)r_sample ) / 10; + (u32)r_sample) / 10; /* Update sending rate (step 4 of [RFC 3448, 4.3]) */ if (hctx->ccid3hctx_p > 0) @@ -492,12 +497,13 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) ccid3_hc_tx_update_x(sk, &now); ccid3_pr_debug("%s(%p), RTT=%uus (sample=%dus), s=%u, " - "p=%u, X_calc=%u, X_recv=%u, X=%u\n", dccp_role(sk), + "p=%u, X_calc=%u, X_recv=%u, X=%u\n", + dccp_role(sk), sk, hctx->ccid3hctx_rtt, (int)r_sample, hctx->ccid3hctx_s, hctx->ccid3hctx_p, hctx->ccid3hctx_x_calc, (unsigned)(hctx->ccid3hctx_x_recv >> 6), - (unsigned)(hctx->ccid3hctx_x >> 6) ); + (unsigned)(hctx->ccid3hctx_x >> 6)); } /* unschedule no feedback timer */ @@ -507,20 +513,20 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) dccp_tx_hist_purge_older(ccid3_tx_hist, &hctx->ccid3hctx_hist, packet); /* - * As we have calculated new ipi, delta, t_nom it is possible that - * we now can send a packet, so wake up dccp_wait_for_ccid + * As we have calculated new ipi, delta, t_nom it is possible + * that we now can send a packet, so wake up dccp_wait_for_ccid */ sk->sk_write_space(sk); /* * Update timeout interval for the nofeedback timer. * We use a configuration option to increase the lower bound. - * This can help avoid triggering the nofeedback timer too often - * ('spinning') on LANs with small RTTs. + * This can help avoid triggering the nofeedback timer too + * often ('spinning') on LANs with small RTTs. */ hctx->ccid3hctx_t_rto = max_t(u32, 4 * hctx->ccid3hctx_rtt, CONFIG_IP_DCCP_CCID3_RTO * - (USEC_PER_SEC/1000) ); + (USEC_PER_SEC/1000)); /* * Schedule no feedback timer to expire in * max(t_RTO, 2 * s/X) = max(t_RTO, 2 * t_ipi) @@ -528,7 +534,8 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) t_nfb = max(hctx->ccid3hctx_t_rto, 2 * hctx->ccid3hctx_t_ipi); ccid3_pr_debug("%s(%p), Scheduled no feedback timer to " - "expire in %lu jiffies (%luus)\n", dccp_role(sk), + "expire in %lu jiffies (%luus)\n", + dccp_role(sk), sk, usecs_to_jiffies(t_nfb), t_nfb); sk_reset_timer(sk, &hctx->ccid3hctx_no_feedback_timer, @@ -538,7 +545,9 @@ static void ccid3_hc_tx_packet_recv(struct sock *sk, struct sk_buff *skb) hctx->ccid3hctx_idle = 1; break; case TFRC_SSTATE_NO_SENT: - /* XXX when implementing bidirectional rx/tx check this again */ + /* + * XXX when implementing bidirectional rx/tx check this again + */ DCCP_WARN("Illegal ACK received - no packet sent\n"); /* fall through */ case TFRC_SSTATE_TERM: /* ignore feedback when closing */ @@ -575,7 +584,8 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, dccp_role(sk), sk, len); rc = -EINVAL; } else { - opt_recv->ccid3or_loss_event_rate = ntohl(*(__be32 *)value); + opt_recv->ccid3or_loss_event_rate = + ntohl(*(__be32 *)value); ccid3_pr_debug("%s(%p), LOSS_EVENT_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_loss_event_rate); @@ -596,7 +606,8 @@ static int ccid3_hc_tx_parse_options(struct sock *sk, unsigned char option, dccp_role(sk), sk, len); rc = -EINVAL; } else { - opt_recv->ccid3or_receive_rate = ntohl(*(__be32 *)value); + opt_recv->ccid3or_receive_rate = + ntohl(*(__be32 *)value); ccid3_pr_debug("%s(%p), RECEIVE_RATE=%u\n", dccp_role(sk), sk, opt_recv->ccid3or_receive_rate); @@ -616,7 +627,8 @@ static int ccid3_hc_tx_init(struct ccid *ccid, struct sock *sk) hctx->ccid3hctx_state = TFRC_SSTATE_NO_SENT; INIT_LIST_HEAD(&hctx->ccid3hctx_hist); - hctx->ccid3hctx_no_feedback_timer.function = ccid3_hc_tx_no_feedback_timer; + hctx->ccid3hctx_no_feedback_timer.function = + ccid3_hc_tx_no_feedback_timer; hctx->ccid3hctx_no_feedback_timer.data = (unsigned long)sk; init_timer(&hctx->ccid3hctx_no_feedback_timer); @@ -754,9 +766,9 @@ static int ccid3_hc_rx_insert_options(struct sock *sk, struct sk_buff *skb) hcrx->ccid3hcrx_elapsed_time)) || dccp_insert_option_timestamp(sk, skb) || dccp_insert_option(sk, skb, TFRC_OPT_LOSS_EVENT_RATE, - &pinv, sizeof(pinv)) || + &pinv, sizeof(pinv)) || dccp_insert_option(sk, skb, TFRC_OPT_RECEIVE_RATE, - &x_recv, sizeof(x_recv))) + &x_recv, sizeof(x_recv))) return -1; return 0; @@ -827,9 +839,9 @@ found: /* * Determine the length of the first loss interval via inverse lookup. * Assume that X_recv can be computed by the throughput equation - * s - * X_recv = -------- - * R * fval + * s + * X_recv = -------- + * R * fval * Find some p such that f(p) = fval; return 1/p [RFC 3448, 6.3.1]. */ if (rtt == 0) { /* would result in divide-by-zero */ @@ -860,7 +872,7 @@ found: if (p == 0) return ~0; else - return 1000000 / p; + return 1000000 / p; } static void ccid3_hc_rx_update_li(struct sock *sk, u64 seq_loss, u8 win_loss) @@ -914,7 +926,8 @@ static int ccid3_hc_rx_detect_loss(struct sock *sk, struct dccp_rx_hist_entry *packet) { struct ccid3_hc_rx_sock *hcrx = ccid3_hc_rx_sk(sk); - struct dccp_rx_hist_entry *rx_hist = dccp_rx_hist_head(&hcrx->ccid3hcrx_hist); + struct dccp_rx_hist_entry *rx_hist = + dccp_rx_hist_head(&hcrx->ccid3hcrx_hist); u64 seqno = packet->dccphrx_seqno; u64 tmp_seqno; int loss = 0; @@ -942,7 +955,7 @@ static int ccid3_hc_rx_detect_loss(struct sock *sk, dccp_inc_seqno(&tmp_seqno); while (dccp_rx_hist_find_entry(&hcrx->ccid3hcrx_hist, tmp_seqno, &ccval)) { - hcrx->ccid3hcrx_seqno_nonloss = tmp_seqno; + hcrx->ccid3hcrx_seqno_nonloss = tmp_seqno; hcrx->ccid3hcrx_ccval_nonloss = ccval; dccp_inc_seqno(&tmp_seqno); } @@ -1044,8 +1057,8 @@ static void ccid3_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb) break; dccp_timestamp(sk, &now); - if (timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) - - (suseconds_t)hcrx->ccid3hcrx_rtt >= 0) { + if ((timeval_delta(&now, &hcrx->ccid3hcrx_tstamp_last_ack) - + (suseconds_t)hcrx->ccid3hcrx_rtt) >= 0) { hcrx->ccid3hcrx_tstamp_last_ack = now; ccid3_hc_rx_send_feedback(sk); } @@ -1118,9 +1131,9 @@ static void ccid3_hc_rx_get_info(struct sock *sk, struct tcp_info *info) BUG_ON(hcrx == NULL); - info->tcpi_ca_state = hcrx->ccid3hcrx_state; - info->tcpi_options |= TCPI_OPT_TIMESTAMPS; - info->tcpi_rcv_rtt = hcrx->ccid3hcrx_rtt; + info->tcpi_ca_state = hcrx->ccid3hcrx_state; + info->tcpi_options |= TCPI_OPT_TIMESTAMPS; + info->tcpi_rcv_rtt = hcrx->ccid3hcrx_rtt; } static void ccid3_hc_tx_get_info(struct sock *sk, struct tcp_info *info) @@ -1212,7 +1225,7 @@ static struct ccid_operations ccid3 = { .ccid_hc_rx_getsockopt = ccid3_hc_rx_getsockopt, .ccid_hc_tx_getsockopt = ccid3_hc_tx_getsockopt, }; - + #ifdef CONFIG_IP_DCCP_CCID3_DEBUG module_param(ccid3_debug, int, 0444); MODULE_PARM_DESC(ccid3_debug, "Enable debug messages"); @@ -1235,7 +1248,7 @@ static __init int ccid3_module_init(void) goto out_free_tx; rc = ccid_register(&ccid3); - if (rc != 0) + if (rc != 0) goto out_free_loss_interval_history; out: return rc; diff --git a/net/dccp/ccids/ccid3.h b/net/dccp/ccids/ccid3.h index 3e3a9cc69b10..15776a88c090 100644 --- a/net/dccp/ccids/ccid3.h +++ b/net/dccp/ccids/ccid3.h @@ -77,7 +77,7 @@ struct ccid3_options_received { /* TFRC sender states */ enum ccid3_hc_tx_states { - TFRC_SSTATE_NO_SENT = 1, + TFRC_SSTATE_NO_SENT = 1, TFRC_SSTATE_NO_FBACK, TFRC_SSTATE_FBACK, TFRC_SSTATE_TERM, @@ -96,7 +96,7 @@ enum ccid3_hc_tx_states { * @ccid3hctx_state - Sender state, one of %ccid3_hc_tx_states * @ccid3hctx_last_win_count - Last window counter sent * @ccid3hctx_t_last_win_count - Timestamp of earliest packet - * with last_win_count value sent + * with last_win_count value sent * @ccid3hctx_no_feedback_timer - Handle to no feedback timer * @ccid3hctx_idle - Flag indicating that sender is idling * @ccid3hctx_t_ld - Time last doubled during slow start @@ -115,7 +115,7 @@ struct ccid3_hc_tx_sock { #define ccid3hctx_t_rto ccid3hctx_tfrc.tfrctx_rto #define ccid3hctx_t_ipi ccid3hctx_tfrc.tfrctx_ipi u16 ccid3hctx_s; - enum ccid3_hc_tx_states ccid3hctx_state:8; + enum ccid3_hc_tx_states ccid3hctx_state:8; u8 ccid3hctx_last_win_count; u8 ccid3hctx_idle; struct timeval ccid3hctx_t_last_win_count; @@ -129,7 +129,7 @@ struct ccid3_hc_tx_sock { /* TFRC receiver states */ enum ccid3_hc_rx_states { - TFRC_RSTATE_NO_DATA = 1, + TFRC_RSTATE_NO_DATA = 1, TFRC_RSTATE_DATA, TFRC_RSTATE_TERM = 127, }; @@ -157,18 +157,18 @@ struct ccid3_hc_rx_sock { #define ccid3hcrx_x_recv ccid3hcrx_tfrc.tfrcrx_x_recv #define ccid3hcrx_rtt ccid3hcrx_tfrc.tfrcrx_rtt #define ccid3hcrx_p ccid3hcrx_tfrc.tfrcrx_p - u64 ccid3hcrx_seqno_nonloss:48, + u64 ccid3hcrx_seqno_nonloss:48, ccid3hcrx_ccval_nonloss:4, ccid3hcrx_ccval_last_counter:4; enum ccid3_hc_rx_states ccid3hcrx_state:8; - u32 ccid3hcrx_bytes_recv; - struct timeval ccid3hcrx_tstamp_last_feedback; - struct timeval ccid3hcrx_tstamp_last_ack; + u32 ccid3hcrx_bytes_recv; + struct timeval ccid3hcrx_tstamp_last_feedback; + struct timeval ccid3hcrx_tstamp_last_ack; struct list_head ccid3hcrx_hist; struct list_head ccid3hcrx_li_hist; - u16 ccid3hcrx_s; - u32 ccid3hcrx_pinv; - u32 ccid3hcrx_elapsed_time; + u16 ccid3hcrx_s; + u32 ccid3hcrx_pinv; + u32 ccid3hcrx_elapsed_time; }; static inline struct ccid3_hc_tx_sock *ccid3_hc_tx_sk(const struct sock *sk) diff --git a/net/dccp/feat.c b/net/dccp/feat.c index 4dc487f27a1f..95b6927ec653 100644 --- a/net/dccp/feat.c +++ b/net/dccp/feat.c @@ -329,7 +329,7 @@ static void dccp_feat_empty_confirm(struct dccp_minisock *dmsk, switch (type) { case DCCPO_CHANGE_L: opt->dccpop_type = DCCPO_CONFIRM_R; break; case DCCPO_CHANGE_R: opt->dccpop_type = DCCPO_CONFIRM_L; break; - default: DCCP_WARN("invalid type %d\n", type); return; + default: DCCP_WARN("invalid type %d\n", type); return; } opt->dccpop_feat = feature; @@ -427,7 +427,7 @@ int dccp_feat_confirm_recv(struct sock *sk, u8 type, u8 feature, switch (type) { case DCCPO_CONFIRM_L: t = DCCPO_CHANGE_R; break; case DCCPO_CONFIRM_R: t = DCCPO_CHANGE_L; break; - default: DCCP_WARN("invalid type %d\n", type); + default: DCCP_WARN("invalid type %d\n", type); return 1; } @@ -610,7 +610,7 @@ const char *dccp_feat_typename(const u8 type) case DCCPO_CHANGE_R: return("ChangeR"); case DCCPO_CONFIRM_R: return("ConfirmR"); /* the following case must not appear in feature negotation */ - default: dccp_pr_debug("unknown type %d [BUG!]\n", type); + default: dccp_pr_debug("unknown type %d [BUG!]\n", type); } return NULL; } diff --git a/net/dccp/input.c b/net/dccp/input.c index 4a3279cd684c..565bc80557ce 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -1,6 +1,6 @@ /* * net/dccp/input.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -82,7 +82,7 @@ static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb) * Otherwise, * Drop packet and return */ - if (dh->dccph_type == DCCP_PKT_SYNC || + if (dh->dccph_type == DCCP_PKT_SYNC || dh->dccph_type == DCCP_PKT_SYNCACK) { if (between48(DCCP_SKB_CB(skb)->dccpd_ack_seq, dp->dccps_awl, dp->dccps_awh) && @@ -185,8 +185,8 @@ static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb, dccp_rcv_close(sk, skb); return 0; case DCCP_PKT_REQUEST: - /* Step 7 - * or (S.is_server and P.type == Response) + /* Step 7 + * or (S.is_server and P.type == Response) * or (S.is_client and P.type == Request) * or (S.state >= OPEN and P.type == Request * and P.seqno >= S.OSR) @@ -274,7 +274,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, const struct dccp_hdr *dh, const unsigned len) { - /* + /* * Step 4: Prepare sequence numbers in REQUEST * If S.state == REQUEST, * If (P.type == Response or P.type == Reset) @@ -342,7 +342,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, * from the Response * / * S.state := PARTOPEN * Set PARTOPEN timer - * Continue with S.state == PARTOPEN + * Continue with S.state == PARTOPEN * / * Step 12 will send the Ack completing the * three-way handshake * / */ @@ -373,7 +373,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, */ __kfree_skb(skb); return 0; - } + } dccp_send_ack(sk); return -1; } @@ -381,7 +381,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk, out_invalid_packet: /* dccp_v4_do_rcv will send a reset */ DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_PACKET_ERROR; - return 1; + return 1; } static int dccp_rcv_respond_partopen_state_process(struct sock *sk, @@ -488,11 +488,11 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, if (dcb->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ) dccp_event_ack_recv(sk, skb); - if (dccp_msk(sk)->dccpms_send_ack_vector && + if (dccp_msk(sk)->dccpms_send_ack_vector && dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk, - DCCP_SKB_CB(skb)->dccpd_seq, - DCCP_ACKVEC_STATE_RECEIVED)) - goto discard; + DCCP_SKB_CB(skb)->dccpd_seq, + DCCP_ACKVEC_STATE_RECEIVED)) + goto discard; /* XXX see the comments in dccp_rcv_established about this */ if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER) @@ -580,7 +580,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, } } - if (!queued) { + if (!queued) { discard: __kfree_skb(skb); } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index ff81679c9f17..90c74b4adb73 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -157,7 +157,7 @@ static inline void dccp_do_pmtu_discovery(struct sock *sk, /* We don't check in the destentry if pmtu discovery is forbidden * on this route. We just assume that no packet_to_big packets * are send back when pmtu discovery is not active. - * There is a small race when the user changes this flag in the + * There is a small race when the user changes this flag in the * route, but I think that's acceptable. */ if ((dst = __sk_dst_check(sk, 0)) == NULL) @@ -467,7 +467,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk, .uli_u = { .ports = { .sport = dccp_hdr(skb)->dccph_dport, .dport = dccp_hdr(skb)->dccph_sport } - } + } }; security_skb_classify_flow(skb, &fl); @@ -595,7 +595,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct inet_request_sock *ireq; struct request_sock *req; struct dccp_request_sock *dreq; - const __be32 service = dccp_hdr_request(skb)->dccph_req_service; + const __be32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; @@ -609,7 +609,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (dccp_bad_service_code(sk, service)) { reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; goto drop; - } + } /* * TW buckets are converted to open requests without * limitations, they conserve resources and peer is @@ -644,7 +644,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) ireq->rmt_addr = skb->nh.iph->saddr; ireq->opt = NULL; - /* + /* * Step 3: Process LISTEN state * * Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookie @@ -846,15 +846,15 @@ static int dccp_v4_rcv(struct sk_buff *skb) } /* Step 2: - * Look up flow ID in table and get corresponding socket */ + * Look up flow ID in table and get corresponding socket */ sk = __inet_lookup(&dccp_hashinfo, skb->nh.iph->saddr, dh->dccph_sport, skb->nh.iph->daddr, dh->dccph_dport, inet_iif(skb)); - /* + /* * Step 2: - * If no socket ... + * If no socket ... */ if (sk == NULL) { dccp_pr_debug("failed to look up flow ID in table and " @@ -862,9 +862,9 @@ static int dccp_v4_rcv(struct sk_buff *skb) goto no_dccp_socket; } - /* + /* * Step 2: - * ... or S.state == TIMEWAIT, + * ... or S.state == TIMEWAIT, * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ @@ -876,8 +876,8 @@ static int dccp_v4_rcv(struct sk_buff *skb) /* * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage - * o if MinCsCov = 0, only packets with CsCov = 0 are accepted - * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov */ min_cov = dccp_sk(sk)->dccps_pcrlen; if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { @@ -900,7 +900,7 @@ no_dccp_socket: goto discard_it; /* * Step 2: - * If no socket ... + * If no socket ... * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c7aaa2574f52..6b91a9dd0411 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -77,7 +77,7 @@ static inline void dccp_v6_send_check(struct sock *sk, int unused_value, } static inline __u32 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr, - __be16 sport, __be16 dport ) + __be16 sport, __be16 dport ) { return secure_tcpv6_sequence_number(saddr, daddr, sport, dport); } @@ -329,7 +329,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) skb = alloc_skb(dccp_v6_ctl_socket->sk->sk_prot->max_header, GFP_ATOMIC); if (skb == NULL) - return; + return; skb_reserve(skb, dccp_v6_ctl_socket->sk->sk_prot->max_header); @@ -353,7 +353,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) dccp_csum_outgoing(skb); dh->dccph_checksum = dccp_v6_csum_finish(skb, &rxskb->nh.ipv6h->saddr, - &rxskb->nh.ipv6h->daddr); + &rxskb->nh.ipv6h->daddr); memset(&fl, 0, sizeof(fl)); ipv6_addr_copy(&fl.fl6_dst, &rxskb->nh.ipv6h->saddr); @@ -424,7 +424,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_request_sock *dreq; struct inet6_request_sock *ireq6; struct ipv6_pinfo *np = inet6_sk(sk); - const __be32 service = dccp_hdr_request(skb)->dccph_req_service; + const __be32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; @@ -437,7 +437,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (dccp_bad_service_code(sk, service)) { reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE; goto drop; - } + } /* * There are no SYN attacks on IPv6, yet... */ @@ -787,7 +787,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) * otherwise we just shortcircuit this and continue with * the new socket.. */ - if (nsk != sk) { + if (nsk != sk) { if (dccp_child_process(sk, nsk, skb)) goto reset; if (opt_skb != NULL) @@ -843,14 +843,14 @@ static int dccp_v6_rcv(struct sk_buff **pskb) DCCP_SKB_CB(skb)->dccpd_ack_seq = dccp_hdr_ack_seq(skb); /* Step 2: - * Look up flow ID in table and get corresponding socket */ + * Look up flow ID in table and get corresponding socket */ sk = __inet6_lookup(&dccp_hashinfo, &skb->nh.ipv6h->saddr, dh->dccph_sport, &skb->nh.ipv6h->daddr, ntohs(dh->dccph_dport), inet6_iif(skb)); /* * Step 2: - * If no socket ... + * If no socket ... */ if (sk == NULL) { dccp_pr_debug("failed to look up flow ID in table and " @@ -860,7 +860,7 @@ static int dccp_v6_rcv(struct sk_buff **pskb) /* * Step 2: - * ... or S.state == TIMEWAIT, + * ... or S.state == TIMEWAIT, * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ @@ -872,8 +872,8 @@ static int dccp_v6_rcv(struct sk_buff **pskb) /* * RFC 4340, sec. 9.2.1: Minimum Checksum Coverage - * o if MinCsCov = 0, only packets with CsCov = 0 are accepted - * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov + * o if MinCsCov = 0, only packets with CsCov = 0 are accepted + * o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov */ min_cov = dccp_sk(sk)->dccps_pcrlen; if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) { @@ -893,7 +893,7 @@ no_dccp_socket: goto discard_it; /* * Step 2: - * If no socket ... + * If no socket ... * Generate Reset(No Connection) unless P.type == Reset * Drop packet and return */ diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 4c9e26775f72..6656bb497c7b 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -182,7 +182,7 @@ out_free: EXPORT_SYMBOL_GPL(dccp_create_openreq_child); -/* +/* * Process an incoming packet for RESPOND sockets represented * as an request_sock. */ diff --git a/net/dccp/output.c b/net/dccp/output.c index b4df12b24161..824569659083 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -1,6 +1,6 @@ /* * net/dccp/output.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -338,7 +338,6 @@ EXPORT_SYMBOL_GPL(dccp_make_response); static struct sk_buff *dccp_make_reset(struct sock *sk, struct dst_entry *dst, const enum dccp_reset_codes code) - { struct dccp_hdr *dh; struct dccp_sock *dp = dccp_sk(sk); @@ -419,14 +418,14 @@ static inline void dccp_connect_init(struct sock *sk) dccp_sync_mss(sk, dst_mtu(dst)); - /* + /* * SWL and AWL are initially adjusted so that they are not less than * the initial Sequence Numbers received and sent, respectively: * SWL := max(GSR + 1 - floor(W/4), ISR), * AWL := max(GSS - W' + 1, ISS). * These adjustments MUST be applied only at the beginning of the * connection. - */ + */ dccp_update_gss(sk, dp->dccps_iss); dccp_set_seqno(&dp->dccps_awl, max48(dp->dccps_awl, dp->dccps_iss)); diff --git a/net/dccp/proto.c b/net/dccp/proto.c index 5ec47d9ee447..63b3fa20e14b 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -196,7 +196,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized) sk, GFP_KERNEL); dp->dccps_hc_tx_ccid = ccid_hc_tx_new(dmsk->dccpms_tx_ccid, sk, GFP_KERNEL); - if (unlikely(dp->dccps_hc_rx_ccid == NULL || + if (unlikely(dp->dccps_hc_rx_ccid == NULL || dp->dccps_hc_tx_ccid == NULL)) { ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); @@ -390,7 +390,7 @@ static int dccp_setsockopt_service(struct sock *sk, const __be32 service, struct dccp_sock *dp = dccp_sk(sk); struct dccp_service_list *sl = NULL; - if (service == DCCP_SERVICE_INVALID_VALUE || + if (service == DCCP_SERVICE_INVALID_VALUE || optlen > DCCP_SERVICE_LIST_MAX_LEN * sizeof(u32)) return -EINVAL; @@ -830,7 +830,7 @@ EXPORT_SYMBOL_GPL(inet_dccp_listen); static const unsigned char dccp_new_state[] = { /* current state: new state: action: */ [0] = DCCP_CLOSED, - [DCCP_OPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, + [DCCP_OPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, [DCCP_REQUESTING] = DCCP_CLOSED, [DCCP_PARTOPEN] = DCCP_CLOSING | DCCP_ACTION_FIN, [DCCP_LISTEN] = DCCP_CLOSED, diff --git a/net/dccp/timer.c b/net/dccp/timer.c index e8f519e7f481..e5348f369c60 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -1,6 +1,6 @@ /* * net/dccp/timer.c - * + * * An implementation of the DCCP protocol * Arnaldo Carvalho de Melo * @@ -102,13 +102,13 @@ static void dccp_retransmit_timer(struct sock *sk) * sk->sk_send_head has to have one skb with * DCCP_SKB_CB(skb)->dccpd_type set to one of the retransmittable DCCP * packet types. The only packets eligible for retransmission are: - * -- Requests in client-REQUEST state (sec. 8.1.1) - * -- Acks in client-PARTOPEN state (sec. 8.1.5) - * -- CloseReq in server-CLOSEREQ state (sec. 8.3) - * -- Close in node-CLOSING state (sec. 8.3) */ + * -- Requests in client-REQUEST state (sec. 8.1.1) + * -- Acks in client-PARTOPEN state (sec. 8.1.5) + * -- CloseReq in server-CLOSEREQ state (sec. 8.3) + * -- Close in node-CLOSING state (sec. 8.3) */ BUG_TRAP(sk->sk_send_head != NULL); - /* + /* * More than than 4MSL (8 minutes) has passed, a RESET(aborted) was * sent, no need to retransmit, this sock is dead. */ @@ -200,7 +200,7 @@ static void dccp_keepalive_timer(unsigned long data) /* Only process if socket is not in use. */ bh_lock_sock(sk); if (sock_owned_by_user(sk)) { - /* Try again later. */ + /* Try again later. */ inet_csk_reset_keepalive_timer(sk, HZ / 20); goto out; } -- cgit v1.2.3 From 2b02a1792000900ead4638f4d42fcdd742062cfa Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 5 Dec 2006 10:19:14 +0100 Subject: [PATCH] remove blk_queue_activity_fn While working on bidi support at struct request level I have found that blk_queue_activity_fn is actually never used. The only user is in ide-probe.c with this code: /* enable led activity for disk drives only */ if (drive->media == ide_disk && hwif->led_act) blk_queue_activity_fn(q, hwif->led_act, drive); And led_act is never initialized anywhere. (Looking back at older kernels it was used in the PPC arch, but was removed around 2.6.18) Unless it is all for future use off course. (this patch is against linux-2.6-block.git as off 2006/12/4) Signed-off-by: Boaz Harrosh Signed-off-by: Jens Axboe --- block/ll_rw_blk.c | 12 ------------ drivers/ide/ide-probe.c | 4 ---- include/linux/blkdev.h | 5 ----- include/linux/ide.h | 2 -- 4 files changed, 23 deletions(-) (limited to 'include/linux') diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 785e61c9a810..a541b42c08e3 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -129,13 +129,6 @@ struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev) } EXPORT_SYMBOL(blk_get_backing_dev_info); -void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data) -{ - q->activity_fn = fn; - q->activity_data = data; -} -EXPORT_SYMBOL(blk_queue_activity_fn); - /** * blk_queue_prep_rq - set a prepare_request function for queue * @q: queue @@ -238,8 +231,6 @@ void blk_queue_make_request(request_queue_t * q, make_request_fn * mfn) * by default assume old behaviour and bounce for any highmem page */ blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH); - - blk_queue_activity_fn(q, NULL, NULL); } EXPORT_SYMBOL(blk_queue_make_request); @@ -2696,9 +2687,6 @@ static inline void add_request(request_queue_t * q, struct request * req) { drive_stat_acct(req, req->nr_sectors, 1); - if (q->activity_fn) - q->activity_fn(q->activity_data, rq_data_dir(req)); - /* * elevator indicated where it wants this request to be * inserted at elevator_merge time diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index dad9c47ebb69..5a5c565a32a8 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -1000,10 +1000,6 @@ static int ide_init_queue(ide_drive_t *drive) /* needs drive->queue to be set */ ide_toggle_bounce(drive, 1); - /* enable led activity for disk drives only */ - if (drive->media == ide_disk && hwif->led_act) - blk_queue_activity_fn(q, hwif->led_act, drive); - return 0; } diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index e1c7286165ff..ea330d7b46c0 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -342,7 +342,6 @@ typedef void (unplug_fn) (request_queue_t *); struct bio_vec; typedef int (merge_bvec_fn) (request_queue_t *, struct bio *, struct bio_vec *); -typedef void (activity_fn) (void *data, int rw); typedef int (issue_flush_fn) (request_queue_t *, struct gendisk *, sector_t *); typedef void (prepare_flush_fn) (request_queue_t *, struct request *); typedef void (softirq_done_fn)(struct request *); @@ -384,7 +383,6 @@ struct request_queue prep_rq_fn *prep_rq_fn; unplug_fn *unplug_fn; merge_bvec_fn *merge_bvec_fn; - activity_fn *activity_fn; issue_flush_fn *issue_flush_fn; prepare_flush_fn *prepare_flush_fn; softirq_done_fn *softirq_done_fn; @@ -411,8 +409,6 @@ struct request_queue */ void *queuedata; - void *activity_data; - /* * queue needs bounce pages for pages above this limit */ @@ -677,7 +673,6 @@ extern void blk_sync_queue(struct request_queue *q); extern void __blk_stop_queue(request_queue_t *q); extern void blk_run_queue(request_queue_t *); extern void blk_start_queueing(request_queue_t *); -extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *); extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned long); extern int blk_rq_unmap_user(struct request *); extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, gfp_t); diff --git a/include/linux/ide.h b/include/linux/ide.h index 64e070f62a87..e26a03981a94 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -804,8 +804,6 @@ typedef struct hwif_s { void *hwif_data; /* extra hwif data */ unsigned dma; - - void (*led_act)(void *data, int rw); } ____cacheline_internodealigned_in_smp ide_hwif_t; /* -- cgit v1.2.3 From 99a3eb3845f034eb55640a3da73e5e28349678c6 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 12 Dec 2006 12:10:28 +0100 Subject: [PATCH] lockdep: fix seqlock_init() seqlock_init() needs to use spin_lock_init() for dynamic locks, so that lockdep is notified about the presence of a new lock. (this is a fallout of the recent networking merge, which started using the so-far unused seqlock_init() API.) This fix solves the following lockdep-internal warning on current -git: INFO: trying to register non-static key. the code is fine but needs lockdep annotation. turning off the locking correctness validator. __lock_acquire+0x10c/0x9f9 lock_acquire+0x56/0x72 _spin_lock+0x35/0x42 neigh_destroy+0x9d/0x12e neigh_periodic_timer+0x10a/0x15c run_timer_softirq+0x126/0x18e __do_softirq+0x6b/0xe6 do_softirq+0x64/0xd2 ksoftirqd+0x82/0x138 Signed-off-by: Ingo Molnar Signed-off-by: Linus Torvalds --- include/linux/seqlock.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h index 46000936f8f1..6b0648cfdffc 100644 --- a/include/linux/seqlock.h +++ b/include/linux/seqlock.h @@ -44,8 +44,11 @@ typedef struct { #define SEQLOCK_UNLOCKED \ __SEQLOCK_UNLOCKED(old_style_seqlock_init) -#define seqlock_init(x) \ - do { *(x) = (seqlock_t) __SEQLOCK_UNLOCKED(x); } while (0) +#define seqlock_init(x) \ + do { \ + (x)->sequence = 0; \ + spin_lock_init(&(x)->lock); \ + } while (0) #define DEFINE_SEQLOCK(x) \ seqlock_t x = __SEQLOCK_UNLOCKED(x) -- cgit v1.2.3