aboutsummaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/Kconfig1
-rw-r--r--fs/Makefile1
-rw-r--r--fs/attr.c2
-rw-r--r--fs/binfmt_aout.c38
-rw-r--r--fs/binfmt_elf.c151
-rw-r--r--fs/binfmt_elf_fdpic.c181
-rw-r--r--fs/binfmt_flat.c2
-rw-r--r--fs/compat_binfmt_elf.c2
-rw-r--r--fs/compat_ioctl.c4
-rw-r--r--fs/exec.c50
-rw-r--r--fs/fcntl.c2
-rw-r--r--fs/file.c2
-rw-r--r--fs/file_table.c2
-rw-r--r--fs/fscache/Kconfig1
-rw-r--r--fs/lockd/host.c2
-rw-r--r--fs/lockd/mon.c12
-rw-r--r--fs/lockd/svc.c2
-rw-r--r--fs/logfs/Kconfig17
-rw-r--r--fs/logfs/Makefile13
-rw-r--r--fs/logfs/compr.c95
-rw-r--r--fs/logfs/dev_bdev.c327
-rw-r--r--fs/logfs/dev_mtd.c254
-rw-r--r--fs/logfs/dir.c827
-rw-r--r--fs/logfs/file.c263
-rw-r--r--fs/logfs/gc.c730
-rw-r--r--fs/logfs/inode.c417
-rw-r--r--fs/logfs/journal.c883
-rw-r--r--fs/logfs/logfs.h724
-rw-r--r--fs/logfs/logfs_abi.h629
-rw-r--r--fs/logfs/readwrite.c2246
-rw-r--r--fs/logfs/segment.c927
-rw-r--r--fs/logfs/super.c650
-rw-r--r--fs/namei.c2
-rw-r--r--fs/nfs/callback.c2
-rw-r--r--fs/nfsd/nfs4callback.c5
-rw-r--r--fs/nfsd/nfs4recover.c4
-rw-r--r--fs/nfsd/nfs4state.c6
-rw-r--r--fs/nfsd/nfs4xdr.c2
-rw-r--r--fs/nfsd/nfsctl.c24
-rw-r--r--fs/nfsd/vfs.c153
-rw-r--r--fs/ocfs2/quota_local.c2
-rw-r--r--fs/proc/array.c4
-rw-r--r--fs/proc/generic.c33
-rw-r--r--fs/proc/task_mmu.c13
-rw-r--r--fs/select.c2
-rw-r--r--fs/seq_file.c4
-rw-r--r--fs/xfs/Makefile1
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c221
-rw-r--r--fs/xfs/linux-2.6/xfs_export.c20
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c854
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c10
-rw-r--r--fs/xfs/linux-2.6/xfs_linux.h1
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c796
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.h29
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c10
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.c16
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.h22
-rw-r--r--fs/xfs/xfs_bmap.c220
-rw-r--r--fs/xfs/xfs_fs.h3
-rw-r--r--fs/xfs/xfs_iget.c19
-rw-r--r--fs/xfs/xfs_inode.c68
-rw-r--r--fs/xfs/xfs_inode.h3
-rw-r--r--fs/xfs/xfs_inode_item.c18
-rw-r--r--fs/xfs/xfs_itable.c2
-rw-r--r--fs/xfs/xfs_log.c106
-rw-r--r--fs/xfs/xfs_log.h16
-rw-r--r--fs/xfs/xfs_mount.c69
-rw-r--r--fs/xfs/xfs_mount.h2
-rw-r--r--fs/xfs/xfs_trans.c2
-rw-r--r--fs/xfs/xfs_trans.h2
-rw-r--r--fs/xfs/xfs_trans_buf.c216
-rw-r--r--fs/xfs/xfs_vnodeops.c107
-rw-r--r--fs/xfs/xfs_vnodeops.h15
73 files changed, 10767 insertions, 1794 deletions
diff --git a/fs/Kconfig b/fs/Kconfig
index 64d44efad7a5..7405f071be67 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -177,6 +177,7 @@ source "fs/efs/Kconfig"
source "fs/jffs2/Kconfig"
# UBIFS File system configuration
source "fs/ubifs/Kconfig"
+source "fs/logfs/Kconfig"
source "fs/cramfs/Kconfig"
source "fs/squashfs/Kconfig"
source "fs/freevxfs/Kconfig"
diff --git a/fs/Makefile b/fs/Makefile
index af6d04700d9c..c3633aa46911 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -99,6 +99,7 @@ obj-$(CONFIG_NTFS_FS) += ntfs/
obj-$(CONFIG_UFS_FS) += ufs/
obj-$(CONFIG_EFS_FS) += efs/
obj-$(CONFIG_JFFS2_FS) += jffs2/
+obj-$(CONFIG_LOGFS) += logfs/
obj-$(CONFIG_UBIFS_FS) += ubifs/
obj-$(CONFIG_AFFS_FS) += affs/
obj-$(CONFIG_ROMFS_FS) += romfs/
diff --git a/fs/attr.c b/fs/attr.c
index 0a6ea54cde7d..0815e93bb487 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -81,7 +81,7 @@ int inode_newsize_ok(const struct inode *inode, loff_t offset)
if (inode->i_size < offset) {
unsigned long limit;
- limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;
+ limit = rlimit(RLIMIT_FSIZE);
if (limit != RLIM_INFINITY && offset > limit)
goto out_sig;
if (offset > inode->i_sb->s_maxbytes)
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index fdd397099172..15d80bb35d6f 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -24,6 +24,7 @@
#include <linux/binfmts.h>
#include <linux/personality.h>
#include <linux/init.h>
+#include <linux/coredump.h>
#include <asm/system.h>
#include <asm/uaccess.h>
@@ -60,26 +61,6 @@ static int set_brk(unsigned long start, unsigned long end)
}
/*
- * These are the only things you should do on a core-file: use only these
- * macros to write out all the necessary info.
- */
-
-static int dump_write(struct file *file, const void *addr, int nr)
-{
- return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
-
-#define DUMP_WRITE(addr, nr) \
- if (!dump_write(file, (void *)(addr), (nr))) \
- goto end_coredump;
-
-#define DUMP_SEEK(offset) \
-if (file->f_op->llseek) { \
- if (file->f_op->llseek(file,(offset),0) != (offset)) \
- goto end_coredump; \
-} else file->f_pos = (offset)
-
-/*
* Routine writes a core dump image in the current directory.
* Currently only a stub-function.
*
@@ -130,26 +111,31 @@ static int aout_core_dump(struct coredump_params *cprm)
set_fs(KERNEL_DS);
/* struct user */
- DUMP_WRITE(&dump,sizeof(dump));
+ if (!dump_write(file, &dump, sizeof(dump)))
+ goto end_coredump;
/* Now dump all of the user data. Include malloced stuff as well */
- DUMP_SEEK(PAGE_SIZE);
+ if (!dump_seek(cprm->file, PAGE_SIZE - sizeof(dump)))
+ goto end_coredump;
/* now we start writing out the user space info */
set_fs(USER_DS);
/* Dump the data area */
if (dump.u_dsize != 0) {
dump_start = START_DATA(dump);
dump_size = dump.u_dsize << PAGE_SHIFT;
- DUMP_WRITE(dump_start,dump_size);
+ if (!dump_write(file, dump_start, dump_size))
+ goto end_coredump;
}
/* Now prepare to dump the stack area */
if (dump.u_ssize != 0) {
dump_start = START_STACK(dump);
dump_size = dump.u_ssize << PAGE_SHIFT;
- DUMP_WRITE(dump_start,dump_size);
+ if (!dump_write(file, dump_start, dump_size))
+ goto end_coredump;
}
/* Finally dump the task struct. Not be used by gdb, but could be useful */
set_fs(KERNEL_DS);
- DUMP_WRITE(current,sizeof(*current));
+ if (!dump_write(file, current, sizeof(*current)))
+ goto end_coredump;
end_coredump:
set_fs(fs);
return has_dumped;
@@ -247,7 +233,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
* size limits imposed on them by creating programs with large
* arrays in the data or bss.
*/
- rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur;
+ rlim = rlimit(RLIMIT_DATA);
if (rlim >= RLIM_INFINITY)
rlim = ~0;
if (ex.a_data + ex.a_bss > rlim)
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index fd5b2ea5d299..535e763ab1a6 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -31,6 +31,7 @@
#include <linux/random.h>
#include <linux/elf.h>
#include <linux/utsname.h>
+#include <linux/coredump.h>
#include <asm/uaccess.h>
#include <asm/param.h>
#include <asm/page.h>
@@ -1085,36 +1086,6 @@ out:
* Modelled on fs/exec.c:aout_core_dump()
* Jeremy Fitzhardinge <jeremy@sw.oz.au>
*/
-/*
- * 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 dump_write(struct file *file, const void *addr, int nr)
-{
- return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
-
-static int dump_seek(struct file *file, loff_t off)
-{
- if (file->f_op->llseek && file->f_op->llseek != no_llseek) {
- if (file->f_op->llseek(file, off, SEEK_CUR) < 0)
- return 0;
- } else {
- char *buf = (char *)get_zeroed_page(GFP_KERNEL);
- if (!buf)
- return 0;
- while (off > 0) {
- unsigned long n = off;
- if (n > PAGE_SIZE)
- n = PAGE_SIZE;
- if (!dump_write(file, buf, n))
- return 0;
- off -= n;
- }
- free_page((unsigned long)buf);
- }
- return 1;
-}
/*
* Decide what to dump of a segment, part, all or none.
@@ -1249,11 +1220,6 @@ static int writenote(struct memelfnote *men, struct file *file,
}
#undef DUMP_WRITE
-#define DUMP_WRITE(addr, nr) \
- if ((size += (nr)) > cprm->limit || \
- !dump_write(cprm->file, (addr), (nr))) \
- goto end_coredump;
-
static void fill_elf_header(struct elfhdr *elf, int segs,
u16 machine, u32 flags, u8 osabi)
{
@@ -1872,6 +1838,34 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma,
return gate_vma;
}
+static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum,
+ elf_addr_t e_shoff, int segs)
+{
+ elf->e_shoff = e_shoff;
+ elf->e_shentsize = sizeof(*shdr4extnum);
+ elf->e_shnum = 1;
+ elf->e_shstrndx = SHN_UNDEF;
+
+ memset(shdr4extnum, 0, sizeof(*shdr4extnum));
+
+ shdr4extnum->sh_type = SHT_NULL;
+ shdr4extnum->sh_size = elf->e_shnum;
+ shdr4extnum->sh_link = elf->e_shstrndx;
+ shdr4extnum->sh_info = segs;
+}
+
+static size_t elf_core_vma_data_size(struct vm_area_struct *gate_vma,
+ unsigned long mm_flags)
+{
+ struct vm_area_struct *vma;
+ size_t size = 0;
+
+ for (vma = first_vma(current, gate_vma); vma != NULL;
+ vma = next_vma(vma, gate_vma))
+ size += vma_dump_size(vma, mm_flags);
+ return size;
+}
+
/*
* Actual dumper
*
@@ -1888,8 +1882,11 @@ static int elf_core_dump(struct coredump_params *cprm)
struct vm_area_struct *vma, *gate_vma;
struct elfhdr *elf = NULL;
loff_t offset = 0, dataoff, foffset;
- unsigned long mm_flags;
struct elf_note_info info;
+ struct elf_phdr *phdr4note = NULL;
+ struct elf_shdr *shdr4extnum = NULL;
+ Elf_Half e_phnum;
+ elf_addr_t e_shoff;
/*
* We no longer stop all VM operations.
@@ -1912,20 +1909,25 @@ static int elf_core_dump(struct coredump_params *cprm)
* Please check DEFAULT_MAX_MAP_COUNT definition when you modify here.
*/
segs = current->mm->map_count;
-#ifdef ELF_CORE_EXTRA_PHDRS
- segs += ELF_CORE_EXTRA_PHDRS;
-#endif
+ segs += elf_core_extra_phdrs();
gate_vma = get_gate_vma(current);
if (gate_vma != NULL)
segs++;
+ /* for notes section */
+ segs++;
+
+ /* If segs > PN_XNUM(0xffff), then e_phnum overflows. To avoid
+ * this, kernel supports extended numbering. Have a look at
+ * include/linux/elf.h for further information. */
+ e_phnum = segs > PN_XNUM ? PN_XNUM : segs;
+
/*
* Collect all the non-memory information about the process for the
* notes. This also sets up the file header.
*/
- if (!fill_note_info(elf, segs + 1, /* including notes section */
- &info, cprm->signr, cprm->regs))
+ if (!fill_note_info(elf, e_phnum, &info, cprm->signr, cprm->regs))
goto cleanup;
has_dumped = 1;
@@ -1934,31 +1936,47 @@ static int elf_core_dump(struct coredump_params *cprm)
fs = get_fs();
set_fs(KERNEL_DS);
- DUMP_WRITE(elf, sizeof(*elf));
offset += sizeof(*elf); /* Elf header */
- offset += (segs + 1) * sizeof(struct elf_phdr); /* Program headers */
+ offset += segs * sizeof(struct elf_phdr); /* Program headers */
foffset = offset;
/* Write notes phdr entry */
{
- struct elf_phdr phdr;
size_t sz = get_note_info_size(&info);
sz += elf_coredump_extra_notes_size();
- fill_elf_note_phdr(&phdr, sz, offset);
+ phdr4note = kmalloc(sizeof(*phdr4note), GFP_KERNEL);
+ if (!phdr4note)
+ goto end_coredump;
+
+ fill_elf_note_phdr(phdr4note, sz, offset);
offset += sz;
- DUMP_WRITE(&phdr, sizeof(phdr));
}
dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE);
- /*
- * We must use the same mm->flags while dumping core to avoid
- * inconsistency between the program headers and bodies, otherwise an
- * unusable core file can be generated.
- */
- mm_flags = current->mm->flags;
+ offset += elf_core_vma_data_size(gate_vma, cprm->mm_flags);
+ offset += elf_core_extra_data_size();
+ e_shoff = offset;
+
+ if (e_phnum == PN_XNUM) {
+ shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL);
+ if (!shdr4extnum)
+ goto end_coredump;
+ fill_extnum_info(elf, shdr4extnum, e_shoff, segs);
+ }
+
+ offset = dataoff;
+
+ size += sizeof(*elf);
+ if (size > cprm->limit || !dump_write(cprm->file, elf, sizeof(*elf)))
+ goto end_coredump;
+
+ size += sizeof(*phdr4note);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, phdr4note, sizeof(*phdr4note)))
+ goto end_coredump;
/* Write program headers for segments dump */
for (vma = first_vma(current, gate_vma); vma != NULL;
@@ -1969,7 +1987,7 @@ static int elf_core_dump(struct coredump_params *cprm)
phdr.p_offset = offset;
phdr.p_vaddr = vma->vm_start;
phdr.p_paddr = 0;
- phdr.p_filesz = vma_dump_size(vma, mm_flags);
+ phdr.p_filesz = vma_dump_size(vma, cprm->mm_flags);
phdr.p_memsz = vma->vm_end - vma->vm_start;
offset += phdr.p_filesz;
phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0;
@@ -1979,12 +1997,14 @@ static int elf_core_dump(struct coredump_params *cprm)
phdr.p_flags |= PF_X;
phdr.p_align = ELF_EXEC_PAGESIZE;
- DUMP_WRITE(&phdr, sizeof(phdr));
+ size += sizeof(phdr);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, &phdr, sizeof(phdr)))
+ goto end_coredump;
}
-#ifdef ELF_CORE_WRITE_EXTRA_PHDRS
- ELF_CORE_WRITE_EXTRA_PHDRS;
-#endif
+ if (!elf_core_write_extra_phdrs(cprm->file, offset, &size, cprm->limit))
+ goto end_coredump;
/* write out the notes section */
if (!write_note_info(&info, cprm->file, &foffset))
@@ -2002,7 +2022,7 @@ static int elf_core_dump(struct coredump_params *cprm)
unsigned long addr;
unsigned long end;
- end = vma->vm_start + vma_dump_size(vma, mm_flags);
+ end = vma->vm_start + vma_dump_size(vma, cprm->mm_flags);
for (addr = vma->vm_start; addr < end; addr += PAGE_SIZE) {
struct page *page;
@@ -2023,15 +2043,24 @@ static int elf_core_dump(struct coredump_params *cprm)
}
}
-#ifdef ELF_CORE_WRITE_EXTRA_DATA
- ELF_CORE_WRITE_EXTRA_DATA;
-#endif
+ if (!elf_core_write_extra_data(cprm->file, &size, cprm->limit))
+ goto end_coredump;
+
+ if (e_phnum == PN_XNUM) {
+ size += sizeof(*shdr4extnum);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, shdr4extnum,
+ sizeof(*shdr4extnum)))
+ goto end_coredump;
+ }
end_coredump:
set_fs(fs);
cleanup:
free_note_info(&info);
+ kfree(shdr4extnum);
+ kfree(phdr4note);
kfree(elf);
out:
return has_dumped;
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 18d77297ccc8..6d6a16c5e9bb 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -34,6 +34,7 @@
#include <linux/elf.h>
#include <linux/elf-fdpic.h>
#include <linux/elfcore.h>
+#include <linux/coredump.h>
#include <asm/uaccess.h>
#include <asm/param.h>
@@ -1216,26 +1217,6 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params,
#ifdef CONFIG_ELF_CORE
/*
- * 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 dump_write(struct file *file, const void *addr, int nr)
-{
- return file->f_op->write(file, addr, nr, &file->f_pos) == nr;
-}
-
-static int dump_seek(struct file *file, loff_t off)
-{
- if (file->f_op->llseek) {
- if (file->f_op->llseek(file, off, SEEK_SET) != off)
- return 0;
- } else {
- file->f_pos = off;
- }
- return 1;
-}
-
-/*
* Decide whether a segment is worth dumping; default is yes to be
* sure (missing info is worse than too much; etc).
* Personally I'd include everything, and use the coredump limit...
@@ -1313,35 +1294,35 @@ static int notesize(struct memelfnote *en)
/* #define DEBUG */
-#define DUMP_WRITE(addr, nr) \
- do { if (!dump_write(file, (addr), (nr))) return 0; } while(0)
-#define DUMP_SEEK(off) \
- do { if (!dump_seek(file, (off))) return 0; } while(0)
+#define DUMP_WRITE(addr, nr, foffset) \
+ do { if (!dump_write(file, (addr), (nr))) return 0; *foffset += (nr); } while(0)
-static int writenote(struct memelfnote *men, struct file *file)
+static int alignfile(struct file *file, loff_t *foffset)
{
- struct elf_note en;
+ static const char buf[4] = { 0, };
+ DUMP_WRITE(buf, roundup(*foffset, 4) - *foffset, foffset);
+ return 1;
+}
+static int writenote(struct memelfnote *men, struct file *file,
+ loff_t *foffset)
+{
+ struct elf_note en;
en.n_namesz = strlen(men->name) + 1;
en.n_descsz = men->datasz;
en.n_type = men->type;
- DUMP_WRITE(&en, sizeof(en));
- DUMP_WRITE(men->name, en.n_namesz);
- /* XXX - cast from long long to long to avoid need for libgcc.a */
- DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */
- DUMP_WRITE(men->data, men->datasz);
- DUMP_SEEK(roundup((unsigned long)file->f_pos, 4)); /* XXX */
+ DUMP_WRITE(&en, sizeof(en), foffset);
+ DUMP_WRITE(men->name, en.n_namesz, foffset);
+ if (!alignfile(file, foffset))
+ return 0;
+ DUMP_WRITE(men->data, men->datasz, foffset);
+ if (!alignfile(file, foffset))
+ return 0;
return 1;
}
#undef DUMP_WRITE
-#undef DUMP_SEEK
-
-#define DUMP_WRITE(addr, nr) \
- if ((size += (nr)) > cprm->limit || \
- !dump_write(cprm->file, (addr), (nr))) \
- goto end_coredump;
static inline void fill_elf_fdpic_header(struct elfhdr *elf, int segs)
{
@@ -1524,6 +1505,22 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t)
return sz;
}
+static void fill_extnum_info(struct elfhdr *elf, struct elf_shdr *shdr4extnum,
+ elf_addr_t e_shoff, int segs)
+{
+ elf->e_shoff = e_shoff;
+ elf->e_shentsize = sizeof(*shdr4extnum);
+ elf->e_shnum = 1;
+ elf->e_shstrndx = SHN_UNDEF;
+
+ memset(shdr4extnum, 0, sizeof(*shdr4extnum));
+
+ shdr4extnum->sh_type = SHT_NULL;
+ shdr4extnum->sh_size = elf->e_shnum;
+ shdr4extnum->sh_link = elf->e_shstrndx;
+ shdr4extnum->sh_info = segs;
+}
+
/*
* dump the segments for an MMU process
*/
@@ -1552,7 +1549,7 @@ static int elf_fdpic_dump_segments(struct file *file, size_t *size,
err = -EIO;
kunmap(page);
page_cache_release(page);
- } else if (!dump_seek(file, file->f_pos + PAGE_SIZE))
+ } else if (!dump_seek(file, PAGE_SIZE))
err = -EFBIG;
if (err)
goto out;
@@ -1588,6 +1585,17 @@ static int elf_fdpic_dump_segments(struct file *file, size_t *size,
}
#endif
+static size_t elf_core_vma_data_size(unsigned long mm_flags)
+{
+ struct vm_area_struct *vma;
+ size_t size = 0;
+
+ for (vma = current->mm->mmap; vma; vma->vm_next)
+ if (maydump(vma, mm_flags))
+ size += vma->vm_end - vma->vm_start;
+ return size;
+}
+
/*
* Actual dumper
*
@@ -1605,7 +1613,7 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
int i;
struct vm_area_struct *vma;
struct elfhdr *elf = NULL;
- loff_t offset = 0, dataoff;
+ loff_t offset = 0, dataoff, foffset;
int numnote;
struct memelfnote *notes = NULL;
struct elf_prstatus *prstatus = NULL; /* NT_PRSTATUS */
@@ -1618,7 +1626,10 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
#endif
int thread_status_size = 0;
elf_addr_t *auxv;
- unsigned long mm_flags;
+ struct elf_phdr *phdr4note = NULL;
+ struct elf_shdr *shdr4extnum = NULL;
+ Elf_Half e_phnum;
+ elf_addr_t e_shoff;
/*
* We no longer stop all VM operations.
@@ -1683,12 +1694,18 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
elf_core_copy_regs(&prstatus->pr_reg, cprm->regs);
segs = current->mm->map_count;
-#ifdef ELF_CORE_EXTRA_PHDRS
- segs += ELF_CORE_EXTRA_PHDRS;
-#endif
+ segs += elf_core_extra_phdrs();
+
+ /* for notes section */
+ segs++;
+
+ /* If segs > PN_XNUM(0xffff), then e_phnum overflows. To avoid
+ * this, kernel supports extended numbering. Have a look at
+ * include/linux/elf.h for further information. */
+ e_phnum = segs > PN_XNUM ? PN_XNUM : segs;
/* Set up header */
- fill_elf_fdpic_header(elf, segs + 1); /* including notes section */
+ fill_elf_fdpic_header(elf, e_phnum);
has_dumped = 1;
current->flags |= PF_DUMPCORE;
@@ -1727,13 +1744,12 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
fs = get_fs();
set_fs(KERNEL_DS);
- DUMP_WRITE(elf, sizeof(*elf));
offset += sizeof(*elf); /* Elf header */
- offset += (segs+1) * sizeof(struct elf_phdr); /* Program headers */
+ offset += segs * sizeof(struct elf_phdr); /* Program headers */
+ foffset = offset;
/* Write notes phdr entry */
{
- struct elf_phdr phdr;
int sz = 0;
for (i = 0; i < numnote; i++)
@@ -1741,20 +1757,38 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
sz += thread_status_size;
- fill_elf_note_phdr(&phdr, sz, offset);
+ phdr4note = kmalloc(sizeof(*phdr4note), GFP_KERNEL);
+ if (!phdr4note)
+ goto end_coredump;
+
+ fill_elf_note_phdr(phdr4note, sz, offset);
offset += sz;
- DUMP_WRITE(&phdr, sizeof(phdr));
}
/* Page-align dumped data */
dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE);
- /*
- * We must use the same mm->flags while dumping core to avoid
- * inconsistency between the program headers and bodies, otherwise an
- * unusable core file can be generated.
- */
- mm_flags = current->mm->flags;
+ offset += elf_core_vma_data_size(cprm->mm_flags);
+ offset += elf_core_extra_data_size();
+ e_shoff = offset;
+
+ if (e_phnum == PN_XNUM) {
+ shdr4extnum = kmalloc(sizeof(*shdr4extnum), GFP_KERNEL);
+ if (!shdr4extnum)
+ goto end_coredump;
+ fill_extnum_info(elf, shdr4extnum, e_shoff, segs);
+ }
+
+ offset = dataoff;
+
+ size += sizeof(*elf);
+ if (size > cprm->limit || !dump_write(cprm->file, elf, sizeof(*elf)))
+ goto end_coredump;
+
+ size += sizeof(*phdr4note);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, phdr4note, sizeof(*phdr4note)))
+ goto end_coredump;
/* write program headers for segments dump */
for (vma = current->mm->mmap; vma; vma = vma->vm_next) {
@@ -1767,7 +1801,7 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
phdr.p_offset = offset;
phdr.p_vaddr = vma->vm_start;
phdr.p_paddr = 0;
- phdr.p_filesz = maydump(vma, mm_flags) ? sz : 0;
+ phdr.p_filesz = maydump(vma, cprm->mm_flags) ? sz : 0;
phdr.p_memsz = sz;
offset += phdr.p_filesz;
phdr.p_flags = vma->vm_flags & VM_READ ? PF_R : 0;
@@ -1777,16 +1811,18 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
phdr.p_flags |= PF_X;
phdr.p_align = ELF_EXEC_PAGESIZE;
- DUMP_WRITE(&phdr, sizeof(phdr));
+ size += sizeof(phdr);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, &phdr, sizeof(phdr)))
+ goto end_coredump;
}
-#ifdef ELF_CORE_WRITE_EXTRA_PHDRS
- ELF_CORE_WRITE_EXTRA_PHDRS;
-#endif
+ if (!elf_core_write_extra_phdrs(cprm->file, offset, &size, cprm->limit))
+ goto end_coredump;
/* write out the notes section */
for (i = 0; i < numnote; i++)
- if (!writenote(notes + i, cprm->file))
+ if (!writenote(notes + i, cprm->file, &foffset))
goto end_coredump;
/* write out the thread status notes section */
@@ -1795,20 +1831,27 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm)
list_entry(t, struct elf_thread_status, list);
for (i = 0; i < tmp->num_notes; i++)
- if (!writenote(&tmp->notes[i], cprm->file))
+ if (!writenote(&tmp->notes[i], cprm->file, &foffset))
goto end_coredump;
}
- if (!dump_seek(cprm->file, dataoff))
+ if (!dump_seek(cprm->file, dataoff - foffset))
goto end_coredump;
if (elf_fdpic_dump_segments(cprm->file, &size, &cprm->limit,
- mm_flags) < 0)
+ cprm->mm_flags) < 0)
goto end_coredump;
-#ifdef ELF_CORE_WRITE_EXTRA_DATA
- ELF_CORE_WRITE_EXTRA_DATA;
-#endif
+ if (!elf_core_write_extra_data(cprm->file, &size, cprm->limit))
+ goto end_coredump;
+
+ if (e_phnum == PN_XNUM) {
+ size += sizeof(*shdr4extnum);
+ if (size > cprm->limit
+ || !dump_write(cprm->file, shdr4extnum,
+ sizeof(*shdr4extnum)))
+ goto end_coredump;
+ }
if (cprm->file->f_pos != offset) {
/* Sanity check */
@@ -1826,7 +1869,7 @@ cleanup:
list_del(tmp);
kfree(list_entry(tmp, struct elf_thread_status, list));
}
-
+ kfree(phdr4note);
kfree(elf);
kfree(prstatus);
kfree(psinfo);
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index 42c6b4a54445..e0e769bdca59 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -501,7 +501,7 @@ static int load_flat_file(struct linux_binprm * bprm,
* size limits imposed on them by creating programs with large
* arrays in the data or bss.
*/
- rlim = current->signal->rlim[RLIMIT_DATA].rlim_cur;
+ rlim = rlimit(RLIMIT_DATA);
if (rlim >= RLIM_INFINITY)
rlim = ~0;
if (data_len + bss_len > rlim) {
diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c
index 0adced2f296f..112e45a17e99 100644
--- a/fs/compat_binfmt_elf.c
+++ b/fs/compat_binfmt_elf.c
@@ -28,10 +28,12 @@
#undef elfhdr
#undef elf_phdr
+#undef elf_shdr
#undef elf_note
#undef elf_addr_t
#define elfhdr elf32_hdr
#define elf_phdr elf32_phdr
+#define elf_shdr elf32_shdr
#define elf_note elf32_note
#define elf_addr_t Elf32_Addr
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 0ca9ec4a79c3..6d55b61bfa79 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -545,7 +545,7 @@ static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, void __user *argp)
kcmd = MTIOCPOS;
karg = &pos;
break;
- case MTIOCGET32:
+ default: /* MTIOCGET32 */
kcmd = MTIOCGET;
karg = &get;
break;
@@ -663,7 +663,7 @@ static int raw_ioctl(unsigned fd, unsigned cmd,
switch (cmd) {
case RAW_SETBIND:
- case RAW_GETBIND: {
+ default: { /* RAW_GETBIND */
struct raw_config_request req;
mm_segment_t oldfs = get_fs();
diff --git a/fs/exec.c b/fs/exec.c
index cce6bbdbdbb1..49cdaa19e5b9 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -195,7 +195,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
* to work from.
*/
rlim = current->signal->rlim;
- if (size > rlim[RLIMIT_STACK].rlim_cur / 4) {
+ if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4) {
put_page(page);
return NULL;
}
@@ -246,6 +246,7 @@ static int __bprm_mm_init(struct linux_binprm *bprm)
vma->vm_start = vma->vm_end - PAGE_SIZE;
vma->vm_flags = VM_STACK_FLAGS;
vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
+ INIT_LIST_HEAD(&vma->anon_vma_chain);
err = insert_vm_struct(mm, vma);
if (err)
goto err;
@@ -516,7 +517,8 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift)
/*
* cover the whole range: [new_start, old_end)
*/
- vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL);
+ if (vma_adjust(vma, new_start, old_end, vma->vm_pgoff, NULL))
+ return -ENOMEM;
/*
* move the page tables downwards, on failure we rely on
@@ -547,15 +549,13 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift)
tlb_finish_mmu(tlb, new_end, old_end);
/*
- * shrink the vma to just the new range.
+ * Shrink the vma to just the new range. Always succeeds.
*/
vma_adjust(vma, new_start, new_end, vma->vm_pgoff, NULL);
return 0;
}
-#define EXTRA_STACK_VM_PAGES 20 /* random */
-
/*
* Finalizes the stack vm_area_struct. The flags and permissions are updated,
* the stack is optionally relocated, and some extra space is added.
@@ -577,7 +577,7 @@ int setup_arg_pages(struct linux_binprm *bprm,
#ifdef CONFIG_STACK_GROWSUP
/* Limit stack size to 1GB */
- stack_base = current->signal->rlim[RLIMIT_STACK].rlim_max;
+ stack_base = rlimit_max(RLIMIT_STACK);
if (stack_base > (1 << 30))
stack_base = 1 << 30;
@@ -630,7 +630,7 @@ int setup_arg_pages(struct linux_binprm *bprm,
goto out_unlock;
}
- stack_expand = EXTRA_STACK_VM_PAGES * PAGE_SIZE;
+ stack_expand = 131072UL; /* randomly 32*4k (or 2*64k) pages */
stack_size = vma->vm_end - vma->vm_start;
/*
* Align this down to a page boundary as expand_stack
@@ -718,6 +718,7 @@ static int exec_mmap(struct mm_struct *mm)
/* Notify parent that we're no longer interested in the old VM */
tsk = current;
old_mm = current->mm;
+ sync_mm_rss(tsk, old_mm);
mm_release(tsk, old_mm);
if (old_mm) {
@@ -1532,7 +1533,7 @@ static int format_corename(char *corename, long signr)
/* core limit size */
case 'c':
rc = snprintf(out_ptr, out_end - out_ptr,
- "%lu", current->signal->rlim[RLIMIT_CORE].rlim_cur);
+ "%lu", rlimit(RLIMIT_CORE));
if (rc > out_end - out_ptr)
goto out;
out_ptr += rc;
@@ -1560,12 +1561,13 @@ out:
return ispipe;
}
-static int zap_process(struct task_struct *start)
+static int zap_process(struct task_struct *start, int exit_code)
{
struct task_struct *t;
int nr = 0;
start->signal->flags = SIGNAL_GROUP_EXIT;
+ start->signal->group_exit_code = exit_code;
start->signal->group_stop_count = 0;
t = start;
@@ -1590,8 +1592,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
spin_lock_irq(&tsk->sighand->siglock);
if (!signal_group_exit(tsk->signal)) {
mm->core_state = core_state;
- tsk->signal->group_exit_code = exit_code;
- nr = zap_process(tsk);
+ nr = zap_process(tsk, exit_code);
}
spin_unlock_irq(&tsk->sighand->siglock);
if (unlikely(nr < 0))
@@ -1640,7 +1641,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
if (p->mm) {
if (unlikely(p->mm == mm)) {
lock_task_sighand(p, &flags);
- nr += zap_process(p);
+ nr += zap_process(p, exit_code);
unlock_task_sighand(p, &flags);
}
break;
@@ -1747,14 +1748,19 @@ void set_dumpable(struct mm_struct *mm, int value)
}
}
-int get_dumpable(struct mm_struct *mm)
+static int __get_dumpable(unsigned long mm_flags)
{
int ret;
- ret = mm->flags & 0x3;
+ ret = mm_flags & MMF_DUMPABLE_MASK;
return (ret >= 2) ? 2 : ret;
}
+int get_dumpable(struct mm_struct *mm)
+{
+ return __get_dumpable(mm->flags);
+}
+
static void wait_for_dump_helpers(struct file *file)
{
struct pipe_inode_info *pipe;
@@ -1797,7 +1803,13 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
struct coredump_params cprm = {
.signr = signr,
.regs = regs,
- .limit = current->signal->rlim[RLIMIT_CORE].rlim_cur,
+ .limit = rlimit(RLIMIT_CORE),
+ /*
+ * We must use the same mm->flags while dumping core to avoid
+ * inconsistency of bit flags, since this flag is not protected
+ * by any locks.
+ */
+ .mm_flags = mm->flags,
};
audit_core_dumps(signr);
@@ -1816,7 +1828,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
/*
* If another thread got here first, or we are not dumpable, bail out.
*/
- if (mm->core_state || !get_dumpable(mm)) {
+ if (mm->core_state || !__get_dumpable(cprm.mm_flags)) {
up_write(&mm->mmap_sem);
put_cred(cred);
goto fail;
@@ -1827,7 +1839,8 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
* process nor do we know its entire history. We only know it
* was tainted so we dump it as root in mode 2.
*/
- if (get_dumpable(mm) == 2) { /* Setuid core dump mode */
+ if (__get_dumpable(cprm.mm_flags) == 2) {
+ /* Setuid core dump mode */
flag = O_EXCL; /* Stop rewrite attacks */
cred->fsuid = 0; /* Dump root private */
}
@@ -1923,8 +1936,9 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
/*
* Dont allow local users get cute and trick others to coredump
* into their pre-created files:
+ * Note, this is not relevant for pipes
*/
- if (inode->i_uid != current_fsuid())
+ if (!ispipe && (inode->i_uid != current_fsuid()))
goto close_fail;
if (!cprm.file->f_op)
goto close_fail;
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 97e01dc0d95f..452d02f9075e 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -344,7 +344,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
switch (cmd) {
case F_DUPFD:
case F_DUPFD_CLOEXEC:
- if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
+ if (arg >= rlimit(RLIMIT_NOFILE))
break;
err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0);
if (err >= 0) {
diff --git a/fs/file.c b/fs/file.c
index 38039af67663..34bb7f71d994 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -257,7 +257,7 @@ int expand_files(struct files_struct *files, int nr)
* N.B. For clone tasks sharing a files structure, this test
* will limit the total number of files that can be opened.
*/
- if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
+ if (nr >= rlimit(RLIMIT_NOFILE))
return -EMFILE;
/* Do we need to expand? */
diff --git a/fs/file_table.c b/fs/file_table.c
index b98404b54383..32d12b78bac8 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -393,7 +393,9 @@ retry:
continue;
if (!(f->f_mode & FMODE_WRITE))
continue;
+ spin_lock(&f->f_lock);
f->f_mode &= ~FMODE_WRITE;
+ spin_unlock(&f->f_lock);
if (file_check_writeable(f) != 0)
continue;
file_release_write(f);
diff --git a/fs/fscache/Kconfig b/fs/fscache/Kconfig
index 864dac20a242..cc94bb9563f2 100644
--- a/fs/fscache/Kconfig
+++ b/fs/fscache/Kconfig
@@ -1,7 +1,6 @@
config FSCACHE
tristate "General filesystem local caching manager"
- depends on EXPERIMENTAL
select SLOW_WORK
help
This option enables a generic filesystem caching manager that can be
diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 4600c2037b8b..bb464d12104c 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -479,8 +479,8 @@ again: mutex_lock(&nlm_host_mutex);
}
}
}
-
mutex_unlock(&nlm_host_mutex);
+ nsm_release(nsm);
}
/*
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index f956651d0f65..fefa4df3f005 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -349,9 +349,9 @@ retry:
* nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle
* @info: pointer to NLMPROC_SM_NOTIFY arguments
*
- * Returns a matching nsm_handle if found in the nsm cache; the returned
- * nsm_handle's reference count is bumped and sm_monitored is cleared.
- * Otherwise returns NULL if some error occurred.
+ * Returns a matching nsm_handle if found in the nsm cache. The returned
+ * nsm_handle's reference count is bumped. Otherwise returns NULL if some
+ * error occurred.
*/
struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info)
{
@@ -370,12 +370,6 @@ struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info)
atomic_inc(&cached->sm_count);
spin_unlock(&nsm_lock);
- /*
- * During subsequent lock activity, force a fresh
- * notification to be set up for this host.
- */
- cached->sm_monitored = 0;
-
dprintk("lockd: host %s (%s) rebooted, cnt %d\n",
cached->sm_name, cached->sm_addrbuf,
atomic_read(&cached->sm_count));
diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c
index e50cfa3d9654..7d150517ddf0 100644
--- a/fs/lockd/svc.c
+++ b/fs/lockd/svc.c
@@ -243,11 +243,9 @@ static int make_socks(struct svc_serv *serv)
if (err < 0)
goto out_err;
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
err = create_lockd_family(serv, PF_INET6);
if (err < 0 && err != -EAFNOSUPPORT)
goto out_err;
-#endif /* CONFIG_IPV6 || CONFIG_IPV6_MODULE */
warned = 0;
return 0;
diff --git a/fs/logfs/Kconfig b/fs/logfs/Kconfig
new file mode 100644
index 000000000000..daf9a9b32dd3
--- /dev/null
+++ b/fs/logfs/Kconfig
@@ -0,0 +1,17 @@
+config LOGFS
+ tristate "LogFS file system (EXPERIMENTAL)"
+ depends on (MTD || BLOCK) && EXPERIMENTAL
+ select ZLIB_INFLATE
+ select ZLIB_DEFLATE
+ select CRC32
+ select BTREE
+ help
+ Flash filesystem aimed to scale efficiently to large devices.
+ In comparison to JFFS2 it offers significantly faster mount
+ times and potentially less RAM usage, although the latter has
+ not been measured yet.
+
+ In its current state it is still very experimental and should
+ not be used for other than testing purposes.
+
+ If unsure, say N.
diff --git a/fs/logfs/Makefile b/fs/logfs/Makefile
new file mode 100644
index 000000000000..4820027787ee
--- /dev/null
+++ b/fs/logfs/Makefile
@@ -0,0 +1,13 @@
+obj-$(CONFIG_LOGFS) += logfs.o
+
+logfs-y += compr.o
+logfs-y += dir.o
+logfs-y += file.o
+logfs-y += gc.o
+logfs-y += inode.o
+logfs-y += journal.o
+logfs-y += readwrite.o
+logfs-y += segment.o
+logfs-y += super.o
+logfs-$(CONFIG_BLOCK) += dev_bdev.o
+logfs-$(CONFIG_MTD) += dev_mtd.o
diff --git a/fs/logfs/compr.c b/fs/logfs/compr.c
new file mode 100644
index 000000000000..44bbfd249abc
--- /dev/null
+++ b/fs/logfs/compr.c
@@ -0,0 +1,95 @@
+/*
+ * fs/logfs/compr.c - compression routines
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/vmalloc.h>
+#include <linux/zlib.h>
+
+#define COMPR_LEVEL 3
+
+static DEFINE_MUTEX(compr_mutex);
+static struct z_stream_s stream;
+
+int logfs_compress(void *in, void *out, size_t inlen, size_t outlen)
+{
+ int err, ret;
+
+ ret = -EIO;
+ mutex_lock(&compr_mutex);
+ err = zlib_deflateInit(&stream, COMPR_LEVEL);
+ if (err != Z_OK)
+ goto error;
+
+ stream.next_in = in;
+ stream.avail_in = inlen;
+ stream.total_in = 0;
+ stream.next_out = out;
+ stream.avail_out = outlen;
+ stream.total_out = 0;
+
+ err = zlib_deflate(&stream, Z_FINISH);
+ if (err != Z_STREAM_END)
+ goto error;
+
+ err = zlib_deflateEnd(&stream);
+ if (err != Z_OK)
+ goto error;
+
+ if (stream.total_out >= stream.total_in)
+ goto error;
+
+ ret = stream.total_out;
+error:
+ mutex_unlock(&compr_mutex);
+ return ret;
+}
+
+int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen)
+{
+ int err, ret;
+
+ ret = -EIO;
+ mutex_lock(&compr_mutex);
+ err = zlib_inflateInit(&stream);
+ if (err != Z_OK)
+ goto error;
+
+ stream.next_in = in;
+ stream.avail_in = inlen;
+ stream.total_in = 0;
+ stream.next_out = out;
+ stream.avail_out = outlen;
+ stream.total_out = 0;
+
+ err = zlib_inflate(&stream, Z_FINISH);
+ if (err != Z_STREAM_END)
+ goto error;
+
+ err = zlib_inflateEnd(&stream);
+ if (err != Z_OK)
+ goto error;
+
+ ret = 0;
+error:
+ mutex_unlock(&compr_mutex);
+ return ret;
+}
+
+int __init logfs_compr_init(void)
+{
+ size_t size = max(zlib_deflate_workspacesize(),
+ zlib_inflate_workspacesize());
+ stream.workspace = vmalloc(size);
+ if (!stream.workspace)
+ return -ENOMEM;
+ return 0;
+}
+
+void logfs_compr_exit(void)
+{
+ vfree(stream.workspace);
+}
diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c
new file mode 100644
index 000000000000..9718c22f186d
--- /dev/null
+++ b/fs/logfs/dev_bdev.c
@@ -0,0 +1,327 @@
+/*
+ * fs/logfs/dev_bdev.c - Device access methods for block devices
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+#include <linux/buffer_head.h>
+
+#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1))
+
+static void request_complete(struct bio *bio, int err)
+{
+ complete((struct completion *)bio->bi_private);
+}
+
+static int sync_request(struct page *page, struct block_device *bdev, int rw)
+{
+ struct bio bio;
+ struct bio_vec bio_vec;
+ struct completion complete;
+
+ bio_init(&bio);
+ bio.bi_io_vec = &bio_vec;
+ bio_vec.bv_page = page;
+ bio_vec.bv_len = PAGE_SIZE;
+ bio_vec.bv_offset = 0;
+ bio.bi_vcnt = 1;
+ bio.bi_idx = 0;
+ bio.bi_size = PAGE_SIZE;
+ bio.bi_bdev = bdev;
+ bio.bi_sector = page->index * (PAGE_SIZE >> 9);
+ init_completion(&complete);
+ bio.bi_private = &complete;
+ bio.bi_end_io = request_complete;
+
+ submit_bio(rw, &bio);
+ generic_unplug_device(bdev_get_queue(bdev));
+ wait_for_completion(&complete);
+ return test_bit(BIO_UPTODATE, &bio.bi_flags) ? 0 : -EIO;
+}
+
+static int bdev_readpage(void *_sb, struct page *page)
+{
+ struct super_block *sb = _sb;
+ struct block_device *bdev = logfs_super(sb)->s_bdev;
+ int err;
+
+ err = sync_request(page, bdev, READ);
+ if (err) {
+ ClearPageUptodate(page);
+ SetPageError(page);
+ } else {
+ SetPageUptodate(page);
+ ClearPageError(page);
+ }
+ unlock_page(page);
+ return err;
+}
+
+static DECLARE_WAIT_QUEUE_HEAD(wq);
+
+static void writeseg_end_io(struct bio *bio, int err)
+{
+ const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
+ struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
+ struct super_block *sb = bio->bi_private;
+ struct logfs_super *super = logfs_super(sb);
+ struct page *page;
+
+ BUG_ON(!uptodate); /* FIXME: Retry io or write elsewhere */
+ BUG_ON(err);
+ BUG_ON(bio->bi_vcnt == 0);
+ do {
+ page = bvec->bv_page;
+ if (--bvec >= bio->bi_io_vec)
+ prefetchw(&bvec->bv_page->flags);
+
+ end_page_writeback(page);
+ } while (bvec >= bio->bi_io_vec);
+ bio_put(bio);
+ if (atomic_dec_and_test(&super->s_pending_writes))
+ wake_up(&wq);
+}
+
+static int __bdev_writeseg(struct super_block *sb, u64 ofs, pgoff_t index,
+ size_t nr_pages)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ struct bio *bio;
+ struct page *page;
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9);
+ int i;
+
+ bio = bio_alloc(GFP_NOFS, max_pages);
+ BUG_ON(!bio); /* FIXME: handle this */
+
+ for (i = 0; i < nr_pages; i++) {
+ if (i >= max_pages) {
+ /* Block layer cannot split bios :( */
+ bio->bi_vcnt = i;
+ bio->bi_idx = 0;
+ bio->bi_size = i * PAGE_SIZE;
+ bio->bi_bdev = super->s_bdev;
+ bio->bi_sector = ofs >> 9;
+ bio->bi_private = sb;
+ bio->bi_end_io = writeseg_end_io;
+ atomic_inc(&super->s_pending_writes);
+ submit_bio(WRITE, bio);
+
+ ofs += i * PAGE_SIZE;
+ index += i;
+ nr_pages -= i;
+ i = 0;
+
+ bio = bio_alloc(GFP_NOFS, max_pages);
+ BUG_ON(!bio);
+ }
+ page = find_lock_page(mapping, index + i);
+ BUG_ON(!page);
+ bio->bi_io_vec[i].bv_page = page;
+ bio->bi_io_vec[i].bv_len = PAGE_SIZE;
+ bio->bi_io_vec[i].bv_offset = 0;
+
+ BUG_ON(PageWriteback(page));
+ set_page_writeback(page);
+ unlock_page(page);
+ }
+ bio->bi_vcnt = nr_pages;
+ bio->bi_idx = 0;
+ bio->bi_size = nr_pages * PAGE_SIZE;
+ bio->bi_bdev = super->s_bdev;
+ bio->bi_sector = ofs >> 9;
+ bio->bi_private = sb;
+ bio->bi_end_io = writeseg_end_io;
+ atomic_inc(&super->s_pending_writes);
+ submit_bio(WRITE, bio);
+ return 0;
+}
+
+static void bdev_writeseg(struct super_block *sb, u64 ofs, size_t len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int head;
+
+ BUG_ON(super->s_flags & LOGFS_SB_FLAG_RO);
+
+ if (len == 0) {
+ /* This can happen when the object fit perfectly into a
+ * segment, the segment gets written per sync and subsequently
+ * closed.
+ */
+ return;
+ }
+ head = ofs & (PAGE_SIZE - 1);
+ if (head) {
+ ofs -= head;
+ len += head;
+ }
+ len = PAGE_ALIGN(len);
+ __bdev_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT);
+ generic_unplug_device(bdev_get_queue(logfs_super(sb)->s_bdev));
+}
+
+
+static void erase_end_io(struct bio *bio, int err)
+{
+ const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
+ struct super_block *sb = bio->bi_private;
+ struct logfs_super *super = logfs_super(sb);
+
+ BUG_ON(!uptodate); /* FIXME: Retry io or write elsewhere */
+ BUG_ON(err);
+ BUG_ON(bio->bi_vcnt == 0);
+ bio_put(bio);
+ if (atomic_dec_and_test(&super->s_pending_writes))
+ wake_up(&wq);
+}
+
+static int do_erase(struct super_block *sb, u64 ofs, pgoff_t index,
+ size_t nr_pages)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct bio *bio;
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ unsigned int max_pages = queue_max_hw_sectors(q) >> (PAGE_SHIFT - 9);
+ int i;
+
+ bio = bio_alloc(GFP_NOFS, max_pages);
+ BUG_ON(!bio); /* FIXME: handle this */
+
+ for (i = 0; i < nr_pages; i++) {
+ if (i >= max_pages) {
+ /* Block layer cannot split bios :( */
+ bio->bi_vcnt = i;
+ bio->bi_idx = 0;
+ bio->bi_size = i * PAGE_SIZE;
+ bio->bi_bdev = super->s_bdev;
+ bio->bi_sector = ofs >> 9;
+ bio->bi_private = sb;
+ bio->bi_end_io = erase_end_io;
+ atomic_inc(&super->s_pending_writes);
+ submit_bio(WRITE, bio);
+
+ ofs += i * PAGE_SIZE;
+ index += i;
+ nr_pages -= i;
+ i = 0;
+
+ bio = bio_alloc(GFP_NOFS, max_pages);
+ BUG_ON(!bio);
+ }
+ bio->bi_io_vec[i].bv_page = super->s_erase_page;
+ bio->bi_io_vec[i].bv_len = PAGE_SIZE;
+ bio->bi_io_vec[i].bv_offset = 0;
+ }
+ bio->bi_vcnt = nr_pages;
+ bio->bi_idx = 0;
+ bio->bi_size = nr_pages * PAGE_SIZE;
+ bio->bi_bdev = super->s_bdev;
+ bio->bi_sector = ofs >> 9;
+ bio->bi_private = sb;
+ bio->bi_end_io = erase_end_io;
+ atomic_inc(&super->s_pending_writes);
+ submit_bio(WRITE, bio);
+ return 0;
+}
+
+static int bdev_erase(struct super_block *sb, loff_t to, size_t len,
+ int ensure_write)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ BUG_ON(to & (PAGE_SIZE - 1));
+ BUG_ON(len & (PAGE_SIZE - 1));
+
+ if (super->s_flags & LOGFS_SB_FLAG_RO)
+ return -EROFS;
+
+ if (ensure_write) {
+ /*
+ * Object store doesn't care whether erases happen or not.
+ * But for the journal they are required. Otherwise a scan
+ * can find an old commit entry and assume it is the current
+ * one, travelling back in time.
+ */
+ do_erase(sb, to, to >> PAGE_SHIFT, len >> PAGE_SHIFT);
+ }
+
+ return 0;
+}
+
+static void bdev_sync(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ wait_event(wq, atomic_read(&super->s_pending_writes) == 0);
+}
+
+static struct page *bdev_find_first_sb(struct super_block *sb, u64 *ofs)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ filler_t *filler = bdev_readpage;
+
+ *ofs = 0;
+ return read_cache_page(mapping, 0, filler, sb);
+}
+
+static struct page *bdev_find_last_sb(struct super_block *sb, u64 *ofs)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ filler_t *filler = bdev_readpage;
+ u64 pos = (super->s_bdev->bd_inode->i_size & ~0xfffULL) - 0x1000;
+ pgoff_t index = pos >> PAGE_SHIFT;
+
+ *ofs = pos;
+ return read_cache_page(mapping, index, filler, sb);
+}
+
+static int bdev_write_sb(struct super_block *sb, struct page *page)
+{
+ struct block_device *bdev = logfs_super(sb)->s_bdev;
+
+ /* Nothing special to do for block devices. */
+ return sync_request(page, bdev, WRITE);
+}
+
+static void bdev_put_device(struct super_block *sb)
+{
+ close_bdev_exclusive(logfs_super(sb)->s_bdev, FMODE_READ|FMODE_WRITE);
+}
+
+static const struct logfs_device_ops bd_devops = {
+ .find_first_sb = bdev_find_first_sb,
+ .find_last_sb = bdev_find_last_sb,
+ .write_sb = bdev_write_sb,
+ .readpage = bdev_readpage,
+ .writeseg = bdev_writeseg,
+ .erase = bdev_erase,
+ .sync = bdev_sync,
+ .put_device = bdev_put_device,
+};
+
+int logfs_get_sb_bdev(struct file_system_type *type, int flags,
+ const char *devname, struct vfsmount *mnt)
+{
+ struct block_device *bdev;
+
+ bdev = open_bdev_exclusive(devname, FMODE_READ|FMODE_WRITE, type);
+ if (IS_ERR(bdev))
+ return PTR_ERR(bdev);
+
+ if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) {
+ int mtdnr = MINOR(bdev->bd_dev);
+ close_bdev_exclusive(bdev, FMODE_READ|FMODE_WRITE);
+ return logfs_get_sb_mtd(type, flags, mtdnr, mnt);
+ }
+
+ return logfs_get_sb_device(type, flags, NULL, bdev, &bd_devops, mnt);
+}
diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c
new file mode 100644
index 000000000000..cafb6ef2e05b
--- /dev/null
+++ b/fs/logfs/dev_mtd.c
@@ -0,0 +1,254 @@
+/*
+ * fs/logfs/dev_mtd.c - Device access methods for MTD
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/completion.h>
+#include <linux/mount.h>
+#include <linux/sched.h>
+
+#define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1))
+
+static int mtd_read(struct super_block *sb, loff_t ofs, size_t len, void *buf)
+{
+ struct mtd_info *mtd = logfs_super(sb)->s_mtd;
+ size_t retlen;
+ int ret;
+
+ ret = mtd->read(mtd, ofs, len, &retlen, buf);
+ BUG_ON(ret == -EINVAL);
+ if (ret)
+ return ret;
+
+ /* Not sure if we should loop instead. */
+ if (retlen != len)
+ return -EIO;
+
+ return 0;
+}
+
+static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct mtd_info *mtd = super->s_mtd;
+ size_t retlen;
+ loff_t page_start, page_end;
+ int ret;
+
+ if (super->s_flags & LOGFS_SB_FLAG_RO)
+ return -EROFS;
+
+ BUG_ON((ofs >= mtd->size) || (len > mtd->size - ofs));
+ BUG_ON(ofs != (ofs >> super->s_writeshift) << super->s_writeshift);
+ BUG_ON(len > PAGE_CACHE_SIZE);
+ page_start = ofs & PAGE_CACHE_MASK;
+ page_end = PAGE_CACHE_ALIGN(ofs + len) - 1;
+ ret = mtd->write(mtd, ofs, len, &retlen, buf);
+ if (ret || (retlen != len))
+ return -EIO;
+
+ return 0;
+}
+
+/*
+ * For as long as I can remember (since about 2001) mtd->erase has been an
+ * asynchronous interface lacking the first driver to actually use the
+ * asynchronous properties. So just to prevent the first implementor of such
+ * a thing from breaking logfs in 2350, we do the usual pointless dance to
+ * declare a completion variable and wait for completion before returning
+ * from mtd_erase(). What an excercise in futility!
+ */
+static void logfs_erase_callback(struct erase_info *ei)
+{
+ complete((struct completion *)ei->priv);
+}
+
+static int mtd_erase_mapping(struct super_block *sb, loff_t ofs, size_t len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ struct page *page;
+ pgoff_t index = ofs >> PAGE_SHIFT;
+
+ for (index = ofs >> PAGE_SHIFT; index < (ofs + len) >> PAGE_SHIFT; index++) {
+ page = find_get_page(mapping, index);
+ if (!page)
+ continue;
+ memset(page_address(page), 0xFF, PAGE_SIZE);
+ page_cache_release(page);
+ }
+ return 0;
+}
+
+static int mtd_erase(struct super_block *sb, loff_t ofs, size_t len,
+ int ensure_write)
+{
+ struct mtd_info *mtd = logfs_super(sb)->s_mtd;
+ struct erase_info ei;
+ DECLARE_COMPLETION_ONSTACK(complete);
+ int ret;
+
+ BUG_ON(len % mtd->erasesize);
+ if (logfs_super(sb)->s_flags & LOGFS_SB_FLAG_RO)
+ return -EROFS;
+
+ memset(&ei, 0, sizeof(ei));
+ ei.mtd = mtd;
+ ei.addr = ofs;
+ ei.len = len;
+ ei.callback = logfs_erase_callback;
+ ei.priv = (long)&complete;
+ ret = mtd->erase(mtd, &ei);
+ if (ret)
+ return -EIO;
+
+ wait_for_completion(&complete);
+ if (ei.state != MTD_ERASE_DONE)
+ return -EIO;
+ return mtd_erase_mapping(sb, ofs, len);
+}
+
+static void mtd_sync(struct super_block *sb)
+{
+ struct mtd_info *mtd = logfs_super(sb)->s_mtd;
+
+ if (mtd->sync)
+ mtd->sync(mtd);
+}
+
+static int mtd_readpage(void *_sb, struct page *page)
+{
+ struct super_block *sb = _sb;
+ int err;
+
+ err = mtd_read(sb, page->index << PAGE_SHIFT, PAGE_SIZE,
+ page_address(page));
+ if (err == -EUCLEAN) {
+ err = 0;
+ /* FIXME: force GC this segment */
+ }
+ if (err) {
+ ClearPageUptodate(page);
+ SetPageError(page);
+ } else {
+ SetPageUptodate(page);
+ ClearPageError(page);
+ }
+ unlock_page(page);
+ return err;
+}
+
+static struct page *mtd_find_first_sb(struct super_block *sb, u64 *ofs)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ filler_t *filler = mtd_readpage;
+ struct mtd_info *mtd = super->s_mtd;
+
+ if (!mtd->block_isbad)
+ return NULL;
+
+ *ofs = 0;
+ while (mtd->block_isbad(mtd, *ofs)) {
+ *ofs += mtd->erasesize;
+ if (*ofs >= mtd->size)
+ return NULL;
+ }
+ BUG_ON(*ofs & ~PAGE_MASK);
+ return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb);
+}
+
+static struct page *mtd_find_last_sb(struct super_block *sb, u64 *ofs)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ filler_t *filler = mtd_readpage;
+ struct mtd_info *mtd = super->s_mtd;
+
+ if (!mtd->block_isbad)
+ return NULL;
+
+ *ofs = mtd->size - mtd->erasesize;
+ while (mtd->block_isbad(mtd, *ofs)) {
+ *ofs -= mtd->erasesize;
+ if (*ofs <= 0)
+ return NULL;
+ }
+ *ofs = *ofs + mtd->erasesize - 0x1000;
+ BUG_ON(*ofs & ~PAGE_MASK);
+ return read_cache_page(mapping, *ofs >> PAGE_SHIFT, filler, sb);
+}
+
+static int __mtd_writeseg(struct super_block *sb, u64 ofs, pgoff_t index,
+ size_t nr_pages)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ struct page *page;
+ int i, err;
+
+ for (i = 0; i < nr_pages; i++) {
+ page = find_lock_page(mapping, index + i);
+ BUG_ON(!page);
+
+ err = mtd_write(sb, page->index << PAGE_SHIFT, PAGE_SIZE,
+ page_address(page));
+ unlock_page(page);
+ page_cache_release(page);
+ if (err)
+ return err;
+ }
+ return 0;
+}
+
+static void mtd_writeseg(struct super_block *sb, u64 ofs, size_t len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int head;
+
+ if (super->s_flags & LOGFS_SB_FLAG_RO)
+ return;
+
+ if (len == 0) {
+ /* This can happen when the object fit perfectly into a
+ * segment, the segment gets written per sync and subsequently
+ * closed.
+ */
+ return;
+ }
+ head = ofs & (PAGE_SIZE - 1);
+ if (head) {
+ ofs -= head;
+ len += head;
+ }
+ len = PAGE_ALIGN(len);
+ __mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT);
+}
+
+static void mtd_put_device(struct super_block *sb)
+{
+ put_mtd_device(logfs_super(sb)->s_mtd);
+}
+
+static const struct logfs_device_ops mtd_devops = {
+ .find_first_sb = mtd_find_first_sb,
+ .find_last_sb = mtd_find_last_sb,
+ .readpage = mtd_readpage,
+ .writeseg = mtd_writeseg,
+ .erase = mtd_erase,
+ .sync = mtd_sync,
+ .put_device = mtd_put_device,
+};
+
+int logfs_get_sb_mtd(struct file_system_type *type, int flags,
+ int mtdnr, struct vfsmount *mnt)
+{
+ struct mtd_info *mtd;
+ const struct logfs_device_ops *devops = &mtd_devops;
+
+ mtd = get_mtd_device(NULL, mtdnr);
+ return logfs_get_sb_device(type, flags, mtd, NULL, devops, mnt);
+}
diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c
new file mode 100644
index 000000000000..56a8bfbb0120
--- /dev/null
+++ b/fs/logfs/dir.c
@@ -0,0 +1,827 @@
+/*
+ * fs/logfs/dir.c - directory-related code
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+
+
+/*
+ * Atomic dir operations
+ *
+ * Directory operations are by default not atomic. Dentries and Inodes are
+ * created/removed/altered in seperate operations. Therefore we need to do
+ * a small amount of journaling.
+ *
+ * Create, link, mkdir, mknod and symlink all share the same function to do
+ * the work: __logfs_create. This function works in two atomic steps:
+ * 1. allocate inode (remember in journal)
+ * 2. allocate dentry (clear journal)
+ *
+ * As we can only get interrupted between the two, when the inode we just
+ * created is simply stored in the anchor. On next mount, if we were
+ * interrupted, we delete the inode. From a users point of view the
+ * operation never happened.
+ *
+ * Unlink and rmdir also share the same function: unlink. Again, this
+ * function works in two atomic steps
+ * 1. remove dentry (remember inode in journal)
+ * 2. unlink inode (clear journal)
+ *
+ * And again, on the next mount, if we were interrupted, we delete the inode.
+ * From a users point of view the operation succeeded.
+ *
+ * Rename is the real pain to deal with, harder than all the other methods
+ * combined. Depending on the circumstances we can run into three cases.
+ * A "target rename" where the target dentry already existed, a "local
+ * rename" where both parent directories are identical or a "cross-directory
+ * rename" in the remaining case.
+ *
+ * Local rename is atomic, as the old dentry is simply rewritten with a new
+ * name.
+ *
+ * Cross-directory rename works in two steps, similar to __logfs_create and
+ * logfs_unlink:
+ * 1. Write new dentry (remember old dentry in journal)
+ * 2. Remove old dentry (clear journal)
+ *
+ * Here we remember a dentry instead of an inode. On next mount, if we were
+ * interrupted, we delete the dentry. From a users point of view, the
+ * operation succeeded.
+ *
+ * Target rename works in three atomic steps:
+ * 1. Attach old inode to new dentry (remember old dentry and new inode)
+ * 2. Remove old dentry (still remember the new inode)
+ * 3. Remove victim inode
+ *
+ * Here we remember both an inode an a dentry. If we get interrupted
+ * between steps 1 and 2, we delete both the dentry and the inode. If
+ * we get interrupted between steps 2 and 3, we delete just the inode.
+ * In either case, the remaining objects are deleted on next mount. From
+ * a users point of view, the operation succeeded.
+ */
+
+static int write_dir(struct inode *dir, struct logfs_disk_dentry *dd,
+ loff_t pos)
+{
+ return logfs_inode_write(dir, dd, sizeof(*dd), pos, WF_LOCK, NULL);
+}
+
+static int write_inode(struct inode *inode)
+{
+ return __logfs_write_inode(inode, WF_LOCK);
+}
+
+static s64 dir_seek_data(struct inode *inode, s64 pos)
+{
+ s64 new_pos = logfs_seek_data(inode, pos);
+
+ return max(pos, new_pos - 1);
+}
+
+static int beyond_eof(struct inode *inode, loff_t bix)
+{
+ loff_t pos = bix << inode->i_sb->s_blocksize_bits;
+ return pos >= i_size_read(inode);
+}
+
+/*
+ * Prime value was chosen to be roughly 256 + 26. r5 hash uses 11,
+ * so short names (len <= 9) don't even occupy the complete 32bit name
+ * space. A prime >256 ensures short names quickly spread the 32bit
+ * name space. Add about 26 for the estimated amount of information
+ * of each character and pick a prime nearby, preferrably a bit-sparse
+ * one.
+ */
+static u32 hash_32(const char *s, int len, u32 seed)
+{
+ u32 hash = seed;
+ int i;
+
+ for (i = 0; i < len; i++)
+ hash = hash * 293 + s[i];
+ return hash;
+}
+
+/*
+ * We have to satisfy several conflicting requirements here. Small
+ * directories should stay fairly compact and not require too many
+ * indirect blocks. The number of possible locations for a given hash
+ * should be small to make lookup() fast. And we should try hard not
+ * to overflow the 32bit name space or nfs and 32bit host systems will
+ * be unhappy.
+ *
+ * So we use the following scheme. First we reduce the hash to 0..15
+ * and try a direct block. If that is occupied we reduce the hash to
+ * 16..255 and try an indirect block. Same for 2x and 3x indirect
+ * blocks. Lastly we reduce the hash to 0x800_0000 .. 0xffff_ffff,
+ * but use buckets containing eight entries instead of a single one.
+ *
+ * Using 16 entries should allow for a reasonable amount of hash
+ * collisions, so the 32bit name space can be packed fairly tight
+ * before overflowing. Oh and currently we don't overflow but return
+ * and error.
+ *
+ * How likely are collisions? Doing the appropriate math is beyond me
+ * and the Bronstein textbook. But running a test program to brute
+ * force collisions for a couple of days showed that on average the
+ * first collision occurs after 598M entries, with 290M being the
+ * smallest result. Obviously 21 entries could already cause a
+ * collision if all entries are carefully chosen.
+ */
+static pgoff_t hash_index(u32 hash, int round)
+{
+ u32 i0_blocks = I0_BLOCKS;
+ u32 i1_blocks = I1_BLOCKS;
+ u32 i2_blocks = I2_BLOCKS;
+ u32 i3_blocks = I3_BLOCKS;
+
+ switch (round) {
+ case 0:
+ return hash % i0_blocks;
+ case 1:
+ return i0_blocks + hash % (i1_blocks - i0_blocks);
+ case 2:
+ return i1_blocks + hash % (i2_blocks - i1_blocks);
+ case 3:
+ return i2_blocks + hash % (i3_blocks - i2_blocks);
+ case 4 ... 19:
+ return i3_blocks + 16 * (hash % (((1<<31) - i3_blocks) / 16))
+ + round - 4;
+ }
+ BUG();
+}
+
+static struct page *logfs_get_dd_page(struct inode *dir, struct dentry *dentry)
+{
+ struct qstr *name = &dentry->d_name;
+ struct page *page;
+ struct logfs_disk_dentry *dd;
+ u32 hash = hash_32(name->name, name->len, 0);
+ pgoff_t index;
+ int round;
+
+ if (name->len > LOGFS_MAX_NAMELEN)
+ return ERR_PTR(-ENAMETOOLONG);
+
+ for (round = 0; round < 20; round++) {
+ index = hash_index(hash, round);
+
+ if (beyond_eof(dir, index))
+ return NULL;
+ if (!logfs_exist_block(dir, index))
+ continue;
+ page = read_cache_page(dir->i_mapping, index,
+ (filler_t *)logfs_readpage, NULL);
+ if (IS_ERR(page))
+ return page;
+ dd = kmap_atomic(page, KM_USER0);
+ BUG_ON(dd->namelen == 0);
+
+ if (name->len != be16_to_cpu(dd->namelen) ||
+ memcmp(name->name, dd->name, name->len)) {
+ kunmap_atomic(dd, KM_USER0);
+ page_cache_release(page);
+ continue;
+ }
+
+ kunmap_atomic(dd, KM_USER0);
+ return page;
+ }
+ return NULL;
+}
+
+static int logfs_remove_inode(struct inode *inode)
+{
+ int ret;
+
+ inode->i_nlink--;
+ ret = write_inode(inode);
+ LOGFS_BUG_ON(ret, inode->i_sb);
+ return ret;
+}
+
+static void abort_transaction(struct inode *inode, struct logfs_transaction *ta)
+{
+ if (logfs_inode(inode)->li_block)
+ logfs_inode(inode)->li_block->ta = NULL;
+ kfree(ta);
+}
+
+static int logfs_unlink(struct inode *dir, struct dentry *dentry)
+{
+ struct logfs_super *super = logfs_super(dir->i_sb);
+ struct inode *inode = dentry->d_inode;
+ struct logfs_transaction *ta;
+ struct page *page;
+ pgoff_t index;
+ int ret;
+
+ ta = kzalloc(sizeof(*ta), GFP_KERNEL);
+ if (!ta)
+ return -ENOMEM;
+
+ ta->state = UNLINK_1;
+ ta->ino = inode->i_ino;
+
+ inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
+
+ page = logfs_get_dd_page(dir, dentry);
+ if (!page) {
+ kfree(ta);
+ return -ENOENT;
+ }
+ if (IS_ERR(page)) {
+ kfree(ta);
+ return PTR_ERR(page);
+ }
+ index = page->index;
+ page_cache_release(page);
+
+ mutex_lock(&super->s_dirop_mutex);
+ logfs_add_transaction(dir, ta);
+
+ ret = logfs_delete(dir, index, NULL);
+ if (!ret)
+ ret = write_inode(dir);
+
+ if (ret) {
+ abort_transaction(dir, ta);
+ printk(KERN_ERR"LOGFS: unable to delete inode\n");
+ goto out;
+ }
+
+ ta->state = UNLINK_2;
+ logfs_add_transaction(inode, ta);
+ ret = logfs_remove_inode(inode);
+out:
+ mutex_unlock(&super->s_dirop_mutex);
+ return ret;
+}
+
+static inline int logfs_empty_dir(struct inode *dir)
+{
+ u64 data;
+
+ data = logfs_seek_data(dir, 0) << dir->i_sb->s_blocksize_bits;
+ return data >= i_size_read(dir);
+}
+
+static int logfs_rmdir(struct inode *dir, struct dentry *dentry)
+{
+ struct inode *inode = dentry->d_inode;
+
+ if (!logfs_empty_dir(inode))
+ return -ENOTEMPTY;
+
+ return logfs_unlink(dir, dentry);
+}
+
+/* FIXME: readdir currently has it's own dir_walk code. I don't see a good
+ * way to combine the two copies */
+#define IMPLICIT_NODES 2
+static int __logfs_readdir(struct file *file, void *buf, filldir_t filldir)
+{
+ struct inode *dir = file->f_dentry->d_inode;
+ loff_t pos = file->f_pos - IMPLICIT_NODES;
+ struct page *page;
+ struct logfs_disk_dentry *dd;
+ int full;
+
+ BUG_ON(pos < 0);
+ for (;; pos++) {
+ if (beyond_eof(dir, pos))
+ break;
+ if (!logfs_exist_block(dir, pos)) {
+ /* deleted dentry */
+ pos = dir_seek_data(dir, pos);
+ continue;
+ }
+ page = read_cache_page(dir->i_mapping, pos,
+ (filler_t *)logfs_readpage, NULL);
+ if (IS_ERR(page))
+ return PTR_ERR(page);
+ dd = kmap_atomic(page, KM_USER0);
+ BUG_ON(dd->namelen == 0);
+
+ full = filldir(buf, (char *)dd->name, be16_to_cpu(dd->namelen),
+ pos, be64_to_cpu(dd->ino), dd->type);
+ kunmap_atomic(dd, KM_USER0);
+ page_cache_release(page);
+ if (full)
+ break;
+ }
+
+ file->f_pos = pos + IMPLICIT_NODES;
+ return 0;
+}
+
+static int logfs_readdir(struct file *file, void *buf, filldir_t filldir)
+{
+ struct inode *inode = file->f_dentry->d_inode;
+ ino_t pino = parent_ino(file->f_dentry);
+ int err;
+
+ if (file->f_pos < 0)
+ return -EINVAL;
+
+ if (file->f_pos == 0) {
+ if (filldir(buf, ".", 1, 1, inode->i_ino, DT_DIR) < 0)
+ return 0;
+ file->f_pos++;
+ }
+ if (file->f_pos == 1) {
+ if (filldir(buf, "..", 2, 2, pino, DT_DIR) < 0)
+ return 0;
+ file->f_pos++;
+ }
+
+ err = __logfs_readdir(file, buf, filldir);
+ return err;
+}
+
+static void logfs_set_name(struct logfs_disk_dentry *dd, struct qstr *name)
+{
+ dd->namelen = cpu_to_be16(name->len);
+ memcpy(dd->name, name->name, name->len);
+}
+
+static struct dentry *logfs_lookup(struct inode *dir, struct dentry *dentry,
+ struct nameidata *nd)
+{
+ struct page *page;
+ struct logfs_disk_dentry *dd;
+ pgoff_t index;
+ u64 ino = 0;
+ struct inode *inode;
+
+ page = logfs_get_dd_page(dir, dentry);
+ if (IS_ERR(page))
+ return ERR_CAST(page);
+ if (!page) {
+ d_add(dentry, NULL);
+ return NULL;
+ }
+ index = page->index;
+ dd = kmap_atomic(page, KM_USER0);
+ ino = be64_to_cpu(dd->ino);
+ kunmap_atomic(dd, KM_USER0);
+ page_cache_release(page);
+
+ inode = logfs_iget(dir->i_sb, ino);
+ if (IS_ERR(inode)) {
+ printk(KERN_ERR"LogFS: Cannot read inode #%llx for dentry (%lx, %lx)n",
+ ino, dir->i_ino, index);
+ return ERR_CAST(inode);
+ }
+ return d_splice_alias(inode, dentry);
+}
+
+static void grow_dir(struct inode *dir, loff_t index)
+{
+ index = (index + 1) << dir->i_sb->s_blocksize_bits;
+ if (i_size_read(dir) < index)
+ i_size_write(dir, index);
+}
+
+static int logfs_write_dir(struct inode *dir, struct dentry *dentry,
+ struct inode *inode)
+{
+ struct page *page;
+ struct logfs_disk_dentry *dd;
+ u32 hash = hash_32(dentry->d_name.name, dentry->d_name.len, 0);
+ pgoff_t index;
+ int round, err;
+
+ for (round = 0; round < 20; round++) {
+ index = hash_index(hash, round);
+
+ if (logfs_exist_block(dir, index))
+ continue;
+ page = find_or_create_page(dir->i_mapping, index, GFP_KERNEL);
+ if (!page)
+ return -ENOMEM;
+
+ dd = kmap_atomic(page, KM_USER0);
+ memset(dd, 0, sizeof(*dd));
+ dd->ino = cpu_to_be64(inode->i_ino);
+ dd->type = logfs_type(inode);
+ logfs_set_name(dd, &dentry->d_name);
+ kunmap_atomic(dd, KM_USER0);
+
+ err = logfs_write_buf(dir, page, WF_LOCK);
+ unlock_page(page);
+ page_cache_release(page);
+ if (!err)
+ grow_dir(dir, index);
+ return err;
+ }
+ /* FIXME: Is there a better return value? In most cases neither
+ * the filesystem nor the directory are full. But we have had
+ * too many collisions for this particular hash and no fallback.
+ */
+ return -ENOSPC;
+}
+
+static int __logfs_create(struct inode *dir, struct dentry *dentry,
+ struct inode *inode, const char *dest, long destlen)
+{
+ struct logfs_super *super = logfs_super(dir->i_sb);
+ struct logfs_inode *li = logfs_inode(inode);
+ struct logfs_transaction *ta;
+ int ret;
+
+ ta = kzalloc(sizeof(*ta), GFP_KERNEL);
+ if (!ta)
+ return -ENOMEM;
+
+ ta->state = CREATE_1;
+ ta->ino = inode->i_ino;
+ mutex_lock(&super->s_dirop_mutex);
+ logfs_add_transaction(inode, ta);
+
+ if (dest) {
+ /* symlink */
+ ret = logfs_inode_write(inode, dest, destlen, 0, WF_LOCK, NULL);
+ if (!ret)
+ ret = write_inode(inode);
+ } else {
+ /* creat/mkdir/mknod */
+ ret = write_inode(inode);
+ }
+ if (ret) {
+ abort_transaction(inode, ta);
+ li->li_flags |= LOGFS_IF_STILLBORN;
+ /* FIXME: truncate symlink */
+ inode->i_nlink--;
+ iput(inode);
+ goto out;
+ }
+
+ ta->state = CREATE_2;
+ logfs_add_transaction(dir, ta);
+ ret = logfs_write_dir(dir, dentry, inode);
+ /* sync directory */
+ if (!ret)
+ ret = write_inode(dir);
+
+ if (ret) {
+ logfs_del_transaction(dir, ta);
+ ta->state = CREATE_2;
+ logfs_add_transaction(inode, ta);
+ logfs_remove_inode(inode);
+ iput(inode);
+ goto out;
+ }
+ d_instantiate(dentry, inode);
+out:
+ mutex_unlock(&super->s_dirop_mutex);
+ return ret;
+}
+
+static int logfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+ struct inode *inode;
+
+ /*
+ * FIXME: why do we have to fill in S_IFDIR, while the mode is
+ * correct for mknod, creat, etc.? Smells like the vfs *should*
+ * do it for us but for some reason fails to do so.
+ */
+ inode = logfs_new_inode(dir, S_IFDIR | mode);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+
+ inode->i_op = &logfs_dir_iops;
+ inode->i_fop = &logfs_dir_fops;
+
+ return __logfs_create(dir, dentry, inode, NULL, 0);
+}
+
+static int logfs_create(struct inode *dir, struct dentry *dentry, int mode,
+ struct nameidata *nd)
+{
+ struct inode *inode;
+
+ inode = logfs_new_inode(dir, mode);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+
+ inode->i_op = &logfs_reg_iops;
+ inode->i_fop = &logfs_reg_fops;
+ inode->i_mapping->a_ops = &logfs_reg_aops;
+
+ return __logfs_create(dir, dentry, inode, NULL, 0);
+}
+
+static int logfs_mknod(struct inode *dir, struct dentry *dentry, int mode,
+ dev_t rdev)
+{
+ struct inode *inode;
+
+ if (dentry->d_name.len > LOGFS_MAX_NAMELEN)
+ return -ENAMETOOLONG;
+
+ inode = logfs_new_inode(dir, mode);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+
+ init_special_inode(inode, mode, rdev);
+
+ return __logfs_create(dir, dentry, inode, NULL, 0);
+}
+
+static int logfs_symlink(struct inode *dir, struct dentry *dentry,
+ const char *target)
+{
+ struct inode *inode;
+ size_t destlen = strlen(target) + 1;
+
+ if (destlen > dir->i_sb->s_blocksize)
+ return -ENAMETOOLONG;
+
+ inode = logfs_new_inode(dir, S_IFLNK | 0777);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+
+ inode->i_op = &logfs_symlink_iops;
+ inode->i_mapping->a_ops = &logfs_reg_aops;
+
+ return __logfs_create(dir, dentry, inode, target, destlen);
+}
+
+static int logfs_permission(struct inode *inode, int mask)
+{
+ return generic_permission(inode, mask, NULL);
+}
+
+static int logfs_link(struct dentry *old_dentry, struct inode *dir,
+ struct dentry *dentry)
+{
+ struct inode *inode = old_dentry->d_inode;
+
+ if (inode->i_nlink >= LOGFS_LINK_MAX)
+ return -EMLINK;
+
+ inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME;
+ atomic_inc(&inode->i_count);
+ inode->i_nlink++;
+ mark_inode_dirty_sync(inode);
+
+ return __logfs_create(dir, dentry, inode, NULL, 0);
+}
+
+static int logfs_get_dd(struct inode *dir, struct dentry *dentry,
+ struct logfs_disk_dentry *dd, loff_t *pos)
+{
+ struct page *page;
+ void *map;
+
+ page = logfs_get_dd_page(dir, dentry);
+ if (IS_ERR(page))
+ return PTR_ERR(page);
+ *pos = page->index;
+ map = kmap_atomic(page, KM_USER0);
+ memcpy(dd, map, sizeof(*dd));
+ kunmap_atomic(map, KM_USER0);
+ page_cache_release(page);
+ return 0;
+}
+
+static int logfs_delete_dd(struct inode *dir, loff_t pos)
+{
+ /*
+ * Getting called with pos somewhere beyond eof is either a goofup
+ * within this file or means someone maliciously edited the
+ * (crc-protected) journal.
+ */
+ BUG_ON(beyond_eof(dir, pos));
+ dir->i_ctime = dir->i_mtime = CURRENT_TIME;
+ log_dir(" Delete dentry (%lx, %llx)\n", dir->i_ino, pos);
+ return logfs_delete(dir, pos, NULL);
+}
+
+/*
+ * Cross-directory rename, target does not exist. Just a little nasty.
+ * Create a new dentry in the target dir, then remove the old dentry,
+ * all the while taking care to remember our operation in the journal.
+ */
+static int logfs_rename_cross(struct inode *old_dir, struct dentry *old_dentry,
+ struct inode *new_dir, struct dentry *new_dentry)
+{
+ struct logfs_super *super = logfs_super(old_dir->i_sb);
+ struct logfs_disk_dentry dd;
+ struct logfs_transaction *ta;
+ loff_t pos;
+ int err;
+
+ /* 1. locate source dd */
+ err = logfs_get_dd(old_dir, old_dentry, &dd, &pos);
+ if (err)
+ return err;
+
+ ta = kzalloc(sizeof(*ta), GFP_KERNEL);
+ if (!ta)
+ return -ENOMEM;
+
+ ta->state = CROSS_RENAME_1;
+ ta->dir = old_dir->i_ino;
+ ta->pos = pos;
+
+ /* 2. write target dd */
+ mutex_lock(&super->s_dirop_mutex);
+ logfs_add_transaction(new_dir, ta);
+ err = logfs_write_dir(new_dir, new_dentry, old_dentry->d_inode);
+ if (!err)
+ err = write_inode(new_dir);
+
+ if (err) {
+ super->s_rename_dir = 0;
+ super->s_rename_pos = 0;
+ abort_transaction(new_dir, ta);
+ goto out;
+ }
+
+ /* 3. remove source dd */
+ ta->state = CROSS_RENAME_2;
+ logfs_add_transaction(old_dir, ta);
+ err = logfs_delete_dd(old_dir, pos);
+ if (!err)
+ err = write_inode(old_dir);
+ LOGFS_BUG_ON(err, old_dir->i_sb);
+out:
+ mutex_unlock(&super->s_dirop_mutex);
+ return err;
+}
+
+static int logfs_replace_inode(struct inode *dir, struct dentry *dentry,
+ struct logfs_disk_dentry *dd, struct inode *inode)
+{
+ loff_t pos;
+ int err;
+
+ err = logfs_get_dd(dir, dentry, dd, &pos);
+ if (err)
+ return err;
+ dd->ino = cpu_to_be64(inode->i_ino);
+ dd->type = logfs_type(inode);
+
+ err = write_dir(dir, dd, pos);
+ if (err)
+ return err;
+ log_dir("Replace dentry (%lx, %llx) %s -> %llx\n", dir->i_ino, pos,
+ dd->name, be64_to_cpu(dd->ino));
+ return write_inode(dir);
+}
+
+/* Target dentry exists - the worst case. We need to attach the source
+ * inode to the target dentry, then remove the orphaned target inode and
+ * source dentry.
+ */
+static int logfs_rename_target(struct inode *old_dir, struct dentry *old_dentry,
+ struct inode *new_dir, struct dentry *new_dentry)
+{
+ struct logfs_super *super = logfs_super(old_dir->i_sb);
+ struct inode *old_inode = old_dentry->d_inode;
+ struct inode *new_inode = new_dentry->d_inode;
+ int isdir = S_ISDIR(old_inode->i_mode);
+ struct logfs_disk_dentry dd;
+ struct logfs_transaction *ta;
+ loff_t pos;
+ int err;
+
+ BUG_ON(isdir != S_ISDIR(new_inode->i_mode));
+ if (isdir) {
+ if (!logfs_empty_dir(new_inode))
+ return -ENOTEMPTY;
+ }
+
+ /* 1. locate source dd */
+ err = logfs_get_dd(old_dir, old_dentry, &dd, &pos);
+ if (err)
+ return err;
+
+ ta = kzalloc(sizeof(*ta), GFP_KERNEL);
+ if (!ta)
+ return -ENOMEM;
+
+ ta->state = TARGET_RENAME_1;
+ ta->dir = old_dir->i_ino;
+ ta->pos = pos;
+ ta->ino = new_inode->i_ino;
+
+ /* 2. attach source inode to target dd */
+ mutex_lock(&super->s_dirop_mutex);
+ logfs_add_transaction(new_dir, ta);
+ err = logfs_replace_inode(new_dir, new_dentry, &dd, old_inode);
+ if (err) {
+ super->s_rename_dir = 0;
+ super->s_rename_pos = 0;
+ super->s_victim_ino = 0;
+ abort_transaction(new_dir, ta);
+ goto out;
+ }
+
+ /* 3. remove source dd */
+ ta->state = TARGET_RENAME_2;
+ logfs_add_transaction(old_dir, ta);
+ err = logfs_delete_dd(old_dir, pos);
+ if (!err)
+ err = write_inode(old_dir);
+ LOGFS_BUG_ON(err, old_dir->i_sb);
+
+ /* 4. remove target inode */
+ ta->state = TARGET_RENAME_3;
+ logfs_add_transaction(new_inode, ta);
+ err = logfs_remove_inode(new_inode);
+
+out:
+ mutex_unlock(&super->s_dirop_mutex);
+ return err;
+}
+
+static int logfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+ struct inode *new_dir, struct dentry *new_dentry)
+{
+ if (new_dentry->d_inode)
+ return logfs_rename_target(old_dir, old_dentry,
+ new_dir, new_dentry);
+ return logfs_rename_cross(old_dir, old_dentry, new_dir, new_dentry);
+}
+
+/* No locking done here, as this is called before .get_sb() returns. */
+int logfs_replay_journal(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode;
+ u64 ino, pos;
+ int err;
+
+ if (super->s_victim_ino) {
+ /* delete victim inode */
+ ino = super->s_victim_ino;
+ printk(KERN_INFO"LogFS: delete unmapped inode #%llx\n", ino);
+ inode = logfs_iget(sb, ino);
+ if (IS_ERR(inode))
+ goto fail;
+
+ LOGFS_BUG_ON(i_size_read(inode) > 0, sb);
+ super->s_victim_ino = 0;
+ err = logfs_remove_inode(inode);
+ iput(inode);
+ if (err) {
+ super->s_victim_ino = ino;
+ goto fail;
+ }
+ }
+ if (super->s_rename_dir) {
+ /* delete old dd from rename */
+ ino = super->s_rename_dir;
+ pos = super->s_rename_pos;
+ printk(KERN_INFO"LogFS: delete unbacked dentry (%llx, %llx)\n",
+ ino, pos);
+ inode = logfs_iget(sb, ino);
+ if (IS_ERR(inode))
+ goto fail;
+
+ super->s_rename_dir = 0;
+ super->s_rename_pos = 0;
+ err = logfs_delete_dd(inode, pos);
+ iput(inode);
+ if (err) {
+ super->s_rename_dir = ino;
+ super->s_rename_pos = pos;
+ goto fail;
+ }
+ }
+ return 0;
+fail:
+ LOGFS_BUG(sb);
+ return -EIO;
+}
+
+const struct inode_operations logfs_symlink_iops = {
+ .readlink = generic_readlink,
+ .follow_link = page_follow_link_light,
+};
+
+const struct inode_operations logfs_dir_iops = {
+ .create = logfs_create,
+ .link = logfs_link,
+ .lookup = logfs_lookup,
+ .mkdir = logfs_mkdir,
+ .mknod = logfs_mknod,
+ .rename = logfs_rename,
+ .rmdir = logfs_rmdir,
+ .permission = logfs_permission,
+ .symlink = logfs_symlink,
+ .unlink = logfs_unlink,
+};
+const struct file_operations logfs_dir_fops = {
+ .fsync = logfs_fsync,
+ .ioctl = logfs_ioctl,
+ .readdir = logfs_readdir,
+ .read = generic_read_dir,
+};
diff --git a/fs/logfs/file.c b/fs/logfs/file.c
new file mode 100644
index 000000000000..370f367a933e
--- /dev/null
+++ b/fs/logfs/file.c
@@ -0,0 +1,263 @@
+/*
+ * fs/logfs/file.c - prepare_write, commit_write and friends
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/sched.h>
+#include <linux/writeback.h>
+
+static int logfs_write_begin(struct file *file, struct address_space *mapping,
+ loff_t pos, unsigned len, unsigned flags,
+ struct page **pagep, void **fsdata)
+{
+ struct inode *inode = mapping->host;
+ struct page *page;
+ pgoff_t index = pos >> PAGE_CACHE_SHIFT;
+
+ page = grab_cache_page_write_begin(mapping, index, flags);
+ if (!page)
+ return -ENOMEM;
+ *pagep = page;
+
+ if ((len == PAGE_CACHE_SIZE) || PageUptodate(page))
+ return 0;
+ if ((pos & PAGE_CACHE_MASK) >= i_size_read(inode)) {
+ unsigned start = pos & (PAGE_CACHE_SIZE - 1);
+ unsigned end = start + len;
+
+ /* Reading beyond i_size is simple: memset to zero */
+ zero_user_segments(page, 0, start, end, PAGE_CACHE_SIZE);
+ return 0;
+ }
+ return logfs_readpage_nolock(page);
+}
+
+static int logfs_write_end(struct file *file, struct address_space *mapping,
+ loff_t pos, unsigned len, unsigned copied, struct page *page,
+ void *fsdata)
+{
+ struct inode *inode = mapping->host;
+ pgoff_t index = page->index;
+ unsigned start = pos & (PAGE_CACHE_SIZE - 1);
+ unsigned end = start + copied;
+ int ret = 0;
+
+ BUG_ON(PAGE_CACHE_SIZE != inode->i_sb->s_blocksize);
+ BUG_ON(page->index > I3_BLOCKS);
+
+ if (copied < len) {
+ /*
+ * Short write of a non-initialized paged. Just tell userspace
+ * to retry the entire page.
+ */
+ if (!PageUptodate(page)) {
+ copied = 0;
+ goto out;
+ }
+ }
+ if (copied == 0)
+ goto out; /* FIXME: do we need to update inode? */
+
+ if (i_size_read(inode) < (index << PAGE_CACHE_SHIFT) + end) {
+ i_size_write(inode, (index << PAGE_CACHE_SHIFT) + end);
+ mark_inode_dirty_sync(inode);
+ }
+
+ SetPageUptodate(page);
+ if (!PageDirty(page)) {
+ if (!get_page_reserve(inode, page))
+ __set_page_dirty_nobuffers(page);
+ else
+ ret = logfs_write_buf(inode, page, WF_LOCK);
+ }
+out:
+ unlock_page(page);
+ page_cache_release(page);
+ return ret ? ret : copied;
+}
+
+int logfs_readpage(struct file *file, struct page *page)
+{
+ int ret;
+
+ ret = logfs_readpage_nolock(page);
+ unlock_page(page);
+ return ret;
+}
+
+/* Clear the page's dirty flag in the radix tree. */
+/* TODO: mucking with PageWriteback is silly. Add a generic function to clear
+ * the dirty bit from the radix tree for filesystems that don't have to wait
+ * for page writeback to finish (i.e. any compressing filesystem).
+ */
+static void clear_radix_tree_dirty(struct page *page)
+{
+ BUG_ON(PagePrivate(page) || page->private);
+ set_page_writeback(page);
+ end_page_writeback(page);
+}
+
+static int __logfs_writepage(struct page *page)
+{
+ struct inode *inode = page->mapping->host;
+ int err;
+
+ err = logfs_write_buf(inode, page, WF_LOCK);
+ if (err)
+ set_page_dirty(page);
+ else
+ clear_radix_tree_dirty(page);
+ unlock_page(page);
+ return err;
+}
+
+static int logfs_writepage(struct page *page, struct writeback_control *wbc)
+{
+ struct inode *inode = page->mapping->host;
+ loff_t i_size = i_size_read(inode);
+ pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT;
+ unsigned offset;
+ u64 bix;
+ level_t level;
+
+ log_file("logfs_writepage(%lx, %lx, %p)\n", inode->i_ino, page->index,
+ page);
+
+ logfs_unpack_index(page->index, &bix, &level);
+
+ /* Indirect blocks are never truncated */
+ if (level != 0)
+ return __logfs_writepage(page);
+
+ /*
+ * TODO: everything below is a near-verbatim copy of nobh_writepage().
+ * The relevant bits should be factored out after logfs is merged.
+ */
+
+ /* Is the page fully inside i_size? */
+ if (bix < end_index)
+ return __logfs_writepage(page);
+
+ /* Is the page fully outside i_size? (truncate in progress) */
+ offset = i_size & (PAGE_CACHE_SIZE-1);
+ if (bix > end_index || offset == 0) {
+ unlock_page(page);
+ return 0; /* don't care */
+ }
+
+ /*
+ * The page straddles i_size. It must be zeroed out on each and every
+ * writepage invokation because it may be mmapped. "A file is mapped
+ * in multiples of the page size. For a file that is not a multiple of
+ * the page size, the remaining memory is zeroed when mapped, and
+ * writes to that region are not written out to the file."
+ */
+ zero_user_segment(page, offset, PAGE_CACHE_SIZE);
+ return __logfs_writepage(page);
+}
+
+static void logfs_invalidatepage(struct page *page, unsigned long offset)
+{
+ move_page_to_btree(page);
+ BUG_ON(PagePrivate(page) || page->private);
+}
+
+static int logfs_releasepage(struct page *page, gfp_t only_xfs_uses_this)
+{
+ return 0; /* None of these are easy to release */
+}
+
+
+int logfs_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
+ unsigned long arg)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ unsigned int oldflags, flags;
+ int err;
+
+ switch (cmd) {
+ case FS_IOC_GETFLAGS:
+ flags = li->li_flags & LOGFS_FL_USER_VISIBLE;
+ return put_user(flags, (int __user *)arg);
+ case FS_IOC_SETFLAGS:
+ if (IS_RDONLY(inode))
+ return -EROFS;
+
+ if (!is_owner_or_cap(inode))
+ return -EACCES;
+
+ err = get_user(flags, (int __user *)arg);
+ if (err)
+ return err;
+
+ mutex_lock(&inode->i_mutex);
+ oldflags = li->li_flags;
+ flags &= LOGFS_FL_USER_MODIFIABLE;
+ flags |= oldflags & ~LOGFS_FL_USER_MODIFIABLE;
+ li->li_flags = flags;
+ mutex_unlock(&inode->i_mutex);
+
+ inode->i_ctime = CURRENT_TIME;
+ mark_inode_dirty_sync(inode);
+ return 0;
+
+ default:
+ return -ENOTTY;
+ }
+}
+
+int logfs_fsync(struct file *file, struct dentry *dentry, int datasync)
+{
+ struct super_block *sb = dentry->d_inode->i_sb;
+ struct logfs_super *super = logfs_super(sb);
+
+ /* FIXME: write anchor */
+ super->s_devops->sync(sb);
+ return 0;
+}
+
+static int logfs_setattr(struct dentry *dentry, struct iattr *attr)
+{
+ struct inode *inode = dentry->d_inode;
+ int err = 0;
+
+ if (attr->ia_valid & ATTR_SIZE)
+ err = logfs_truncate(inode, attr->ia_size);
+ attr->ia_valid &= ~ATTR_SIZE;
+
+ if (!err)
+ err = inode_change_ok(inode, attr);
+ if (!err)
+ err = inode_setattr(inode, attr);
+ return err;
+}
+
+const struct inode_operations logfs_reg_iops = {
+ .setattr = logfs_setattr,
+};
+
+const struct file_operations logfs_reg_fops = {
+ .aio_read = generic_file_aio_read,
+ .aio_write = generic_file_aio_write,
+ .fsync = logfs_fsync,
+ .ioctl = logfs_ioctl,
+ .llseek = generic_file_llseek,
+ .mmap = generic_file_readonly_mmap,
+ .open = generic_file_open,
+ .read = do_sync_read,
+ .write = do_sync_write,
+};
+
+const struct address_space_operations logfs_reg_aops = {
+ .invalidatepage = logfs_invalidatepage,
+ .readpage = logfs_readpage,
+ .releasepage = logfs_releasepage,
+ .set_page_dirty = __set_page_dirty_nobuffers,
+ .writepage = logfs_writepage,
+ .writepages = generic_writepages,
+ .write_begin = logfs_write_begin,
+ .write_end = logfs_write_end,
+};
diff --git a/fs/logfs/gc.c b/fs/logfs/gc.c
new file mode 100644
index 000000000000..92949f95a901
--- /dev/null
+++ b/fs/logfs/gc.c
@@ -0,0 +1,730 @@
+/*
+ * fs/logfs/gc.c - garbage collection code
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/sched.h>
+
+/*
+ * Wear leveling needs to kick in when the difference between low erase
+ * counts and high erase counts gets too big. A good value for "too big"
+ * may be somewhat below 10% of maximum erase count for the device.
+ * Why not 397, to pick a nice round number with no specific meaning? :)
+ *
+ * WL_RATELIMIT is the minimum time between two wear level events. A huge
+ * number of segments may fulfil the requirements for wear leveling at the
+ * same time. If that happens we don't want to cause a latency from hell,
+ * but just gently pick one segment every so often and minimize overhead.
+ */
+#define WL_DELTA 397
+#define WL_RATELIMIT 100
+#define MAX_OBJ_ALIASES 2600
+#define SCAN_RATIO 512 /* number of scanned segments per gc'd segment */
+#define LIST_SIZE 64 /* base size of candidate lists */
+#define SCAN_ROUNDS 128 /* maximum number of complete medium scans */
+#define SCAN_ROUNDS_HIGH 4 /* maximum number of higher-level scans */
+
+static int no_free_segments(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ return super->s_free_list.count;
+}
+
+/* journal has distance -1, top-most ifile layer distance 0 */
+static u8 root_distance(struct super_block *sb, gc_level_t __gc_level)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u8 gc_level = (__force u8)__gc_level;
+
+ switch (gc_level) {
+ case 0: /* fall through */
+ case 1: /* fall through */
+ case 2: /* fall through */
+ case 3:
+ /* file data or indirect blocks */
+ return super->s_ifile_levels + super->s_iblock_levels - gc_level;
+ case 6: /* fall through */
+ case 7: /* fall through */
+ case 8: /* fall through */
+ case 9:
+ /* inode file data or indirect blocks */
+ return super->s_ifile_levels - (gc_level - 6);
+ default:
+ printk(KERN_ERR"LOGFS: segment of unknown level %x found\n",
+ gc_level);
+ WARN_ON(1);
+ return super->s_ifile_levels + super->s_iblock_levels;
+ }
+}
+
+static int segment_is_reserved(struct super_block *sb, u32 segno)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area;
+ void *reserved;
+ int i;
+
+ /* Some segments are reserved. Just pretend they were all valid */
+ reserved = btree_lookup32(&super->s_reserved_segments, segno);
+ if (reserved)
+ return 1;
+
+ /* Currently open segments */
+ for_each_area(i) {
+ area = super->s_area[i];
+ if (area->a_is_open && area->a_segno == segno)
+ return 1;
+ }
+
+ return 0;
+}
+
+static void logfs_mark_segment_bad(struct super_block *sb, u32 segno)
+{
+ BUG();
+}
+
+/*
+ * Returns the bytes consumed by valid objects in this segment. Object headers
+ * are counted, the segment header is not.
+ */
+static u32 logfs_valid_bytes(struct super_block *sb, u32 segno, u32 *ec,
+ gc_level_t *gc_level)
+{
+ struct logfs_segment_entry se;
+ u32 ec_level;
+
+ logfs_get_segment_entry(sb, segno, &se);
+ if (se.ec_level == cpu_to_be32(BADSEG) ||
+ se.valid == cpu_to_be32(RESERVED))
+ return RESERVED;
+
+ ec_level = be32_to_cpu(se.ec_level);
+ *ec = ec_level >> 4;
+ *gc_level = GC_LEVEL(ec_level & 0xf);
+ return be32_to_cpu(se.valid);
+}
+
+static void logfs_cleanse_block(struct super_block *sb, u64 ofs, u64 ino,
+ u64 bix, gc_level_t gc_level)
+{
+ struct inode *inode;
+ int err, cookie;
+
+ inode = logfs_safe_iget(sb, ino, &cookie);
+ err = logfs_rewrite_block(inode, bix, ofs, gc_level, 0);
+ BUG_ON(err);
+ logfs_safe_iput(inode, cookie);
+}
+
+static u32 logfs_gc_segment(struct super_block *sb, u32 segno, u8 dist)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_segment_header sh;
+ struct logfs_object_header oh;
+ u64 ofs, ino, bix;
+ u32 seg_ofs, logical_segno, cleaned = 0;
+ int err, len, valid;
+ gc_level_t gc_level;
+
+ LOGFS_BUG_ON(segment_is_reserved(sb, segno), sb);
+
+ btree_insert32(&super->s_reserved_segments, segno, (void *)1, GFP_NOFS);
+ err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh);
+ BUG_ON(err);
+ gc_level = GC_LEVEL(sh.level);
+ logical_segno = be32_to_cpu(sh.segno);
+ if (sh.crc != logfs_crc32(&sh, sizeof(sh), 4)) {
+ logfs_mark_segment_bad(sb, segno);
+ cleaned = -1;
+ goto out;
+ }
+
+ for (seg_ofs = LOGFS_SEGMENT_HEADERSIZE;
+ seg_ofs + sizeof(oh) < super->s_segsize; ) {
+ ofs = dev_ofs(sb, logical_segno, seg_ofs);
+ err = wbuf_read(sb, dev_ofs(sb, segno, seg_ofs), sizeof(oh),
+ &oh);
+ BUG_ON(err);
+
+ if (!memchr_inv(&oh, 0xff, sizeof(oh)))
+ break;
+
+ if (oh.crc != logfs_crc32(&oh, sizeof(oh) - 4, 4)) {
+ logfs_mark_segment_bad(sb, segno);
+ cleaned = super->s_segsize - 1;
+ goto out;
+ }
+
+ ino = be64_to_cpu(oh.ino);
+ bix = be64_to_cpu(oh.bix);
+ len = sizeof(oh) + be16_to_cpu(oh.len);
+ valid = logfs_is_valid_block(sb, ofs, ino, bix, gc_level);
+ if (valid == 1) {
+ logfs_cleanse_block(sb, ofs, ino, bix, gc_level);
+ cleaned += len;
+ } else if (valid == 2) {
+ /* Will be invalid upon journal commit */
+ cleaned += len;
+ }
+ seg_ofs += len;
+ }
+out:
+ btree_remove32(&super->s_reserved_segments, segno);
+ return cleaned;
+}
+
+static struct gc_candidate *add_list(struct gc_candidate *cand,
+ struct candidate_list *list)
+{
+ struct rb_node **p = &list->rb_tree.rb_node;
+ struct rb_node *parent = NULL;
+ struct gc_candidate *cur;
+ int comp;
+
+ cand->list = list;
+ while (*p) {
+ parent = *p;
+ cur = rb_entry(parent, struct gc_candidate, rb_node);
+
+ if (list->sort_by_ec)
+ comp = cand->erase_count < cur->erase_count;
+ else
+ comp = cand->valid < cur->valid;
+
+ if (comp)
+ p = &parent->rb_left;
+ else
+ p = &parent->rb_right;
+ }
+ rb_link_node(&cand->rb_node, parent, p);
+ rb_insert_color(&cand->rb_node, &list->rb_tree);
+
+ if (list->count <= list->maxcount) {
+ list->count++;
+ return NULL;
+ }
+ cand = rb_entry(rb_last(&list->rb_tree), struct gc_candidate, rb_node);
+ rb_erase(&cand->rb_node, &list->rb_tree);
+ cand->list = NULL;
+ return cand;
+}
+
+static void remove_from_list(struct gc_candidate *cand)
+{
+ struct candidate_list *list = cand->list;
+
+ rb_erase(&cand->rb_node, &list->rb_tree);
+ list->count--;
+}
+
+static void free_candidate(struct super_block *sb, struct gc_candidate *cand)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ btree_remove32(&super->s_cand_tree, cand->segno);
+ kfree(cand);
+}
+
+u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec)
+{
+ struct gc_candidate *cand;
+ u32 segno;
+
+ BUG_ON(list->count == 0);
+
+ cand = rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node);
+ remove_from_list(cand);
+ segno = cand->segno;
+ if (ec)
+ *ec = cand->erase_count;
+ free_candidate(sb, cand);
+ return segno;
+}
+
+/*
+ * We have several lists to manage segments with. The reserve_list is used to
+ * deal with bad blocks. We try to keep the best (lowest ec) segments on this
+ * list.
+ * The free_list contains free segments for normal usage. It usually gets the
+ * second pick after the reserve_list. But when the free_list is running short
+ * it is more important to keep the free_list full than to keep a reserve.
+ *
+ * Segments that are not free are put onto a per-level low_list. If we have
+ * to run garbage collection, we pick a candidate from there. All segments on
+ * those lists should have at least some free space so GC will make progress.
+ *
+ * And last we have the ec_list, which is used to pick segments for wear
+ * leveling.
+ *
+ * If all appropriate lists are full, we simply free the candidate and forget
+ * about that segment for a while. We have better candidates for each purpose.
+ */
+static void __add_candidate(struct super_block *sb, struct gc_candidate *cand)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u32 full = super->s_segsize - LOGFS_SEGMENT_RESERVE;
+
+ if (cand->valid == 0) {
+ /* 100% free segments */
+ log_gc_noisy("add reserve segment %x (ec %x) at %llx\n",
+ cand->segno, cand->erase_count,
+ dev_ofs(sb, cand->segno, 0));
+ cand = add_list(cand, &super->s_reserve_list);
+ if (cand) {
+ log_gc_noisy("add free segment %x (ec %x) at %llx\n",
+ cand->segno, cand->erase_count,
+ dev_ofs(sb, cand->segno, 0));
+ cand = add_list(cand, &super->s_free_list);
+ }
+ } else {
+ /* good candidates for Garbage Collection */
+ if (cand->valid < full)
+ cand = add_list(cand, &super->s_low_list[cand->dist]);
+ /* good candidates for wear leveling,
+ * segments that were recently written get ignored */
+ if (cand)
+ cand = add_list(cand, &super->s_ec_list);
+ }
+ if (cand)
+ free_candidate(sb, cand);
+}
+
+static int add_candidate(struct super_block *sb, u32 segno, u32 valid, u32 ec,
+ u8 dist)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct gc_candidate *cand;
+
+ cand = kmalloc(sizeof(*cand), GFP_NOFS);
+ if (!cand)
+ return -ENOMEM;
+
+ cand->segno = segno;
+ cand->valid = valid;
+ cand->erase_count = ec;
+ cand->dist = dist;
+
+ btree_insert32(&super->s_cand_tree, segno, cand, GFP_NOFS);
+ __add_candidate(sb, cand);
+ return 0;
+}
+
+static void remove_segment_from_lists(struct super_block *sb, u32 segno)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct gc_candidate *cand;
+
+ cand = btree_lookup32(&super->s_cand_tree, segno);
+ if (cand) {
+ remove_from_list(cand);
+ free_candidate(sb, cand);
+ }
+}
+
+static void scan_segment(struct super_block *sb, u32 segno)
+{
+ u32 valid, ec = 0;
+ gc_level_t gc_level = 0;
+ u8 dist;
+
+ if (segment_is_reserved(sb, segno))
+ return;
+
+ remove_segment_from_lists(sb, segno);
+ valid = logfs_valid_bytes(sb, segno, &ec, &gc_level);
+ if (valid == RESERVED)
+ return;
+
+ dist = root_distance(sb, gc_level);
+ add_candidate(sb, segno, valid, ec, dist);
+}
+
+static struct gc_candidate *first_in_list(struct candidate_list *list)
+{
+ if (list->count == 0)
+ return NULL;
+ return rb_entry(rb_first(&list->rb_tree), struct gc_candidate, rb_node);
+}
+
+/*
+ * Find the best segment for garbage collection. Main criterion is
+ * the segment requiring the least effort to clean. Secondary
+ * criterion is to GC on the lowest level available.
+ *
+ * So we search the least effort segment on the lowest level first,
+ * then move up and pick another segment iff is requires significantly
+ * less effort. Hence the LOGFS_MAX_OBJECTSIZE in the comparison.
+ */
+static struct gc_candidate *get_candidate(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i, max_dist;
+ struct gc_candidate *cand = NULL, *this;
+
+ max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS);
+
+ for (i = max_dist; i >= 0; i--) {
+ this = first_in_list(&super->s_low_list[i]);
+ if (!this)
+ continue;
+ if (!cand)
+ cand = this;
+ if (this->valid + LOGFS_MAX_OBJECTSIZE <= cand->valid)
+ cand = this;
+ }
+ return cand;
+}
+
+static int __logfs_gc_once(struct super_block *sb, struct gc_candidate *cand)
+{
+ struct logfs_super *super = logfs_super(sb);
+ gc_level_t gc_level;
+ u32 cleaned, valid, segno, ec;
+ u8 dist;
+
+ if (!cand) {
+ log_gc("GC attempted, but no candidate found\n");
+ return 0;
+ }
+
+ segno = cand->segno;
+ dist = cand->dist;
+ valid = logfs_valid_bytes(sb, segno, &ec, &gc_level);
+ free_candidate(sb, cand);
+ log_gc("GC segment #%02x at %llx, %x required, %x free, %x valid, %llx free\n",
+ segno, (u64)segno << super->s_segshift,
+ dist, no_free_segments(sb), valid,
+ super->s_free_bytes);
+ cleaned = logfs_gc_segment(sb, segno, dist);
+ log_gc("GC segment #%02x complete - now %x valid\n", segno,
+ valid - cleaned);
+ BUG_ON(cleaned != valid);
+ return 1;
+}
+
+static int logfs_gc_once(struct super_block *sb)
+{
+ struct gc_candidate *cand;
+
+ cand = get_candidate(sb);
+ if (cand)
+ remove_from_list(cand);
+ return __logfs_gc_once(sb, cand);
+}
+
+/* returns 1 if a wrap occurs, 0 otherwise */
+static int logfs_scan_some(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u32 segno;
+ int i, ret = 0;
+
+ segno = super->s_sweeper;
+ for (i = SCAN_RATIO; i > 0; i--) {
+ segno++;
+ if (segno >= super->s_no_segs) {
+ segno = 0;
+ ret = 1;
+ /* Break out of the loop. We want to read a single
+ * block from the segment size on next invocation if
+ * SCAN_RATIO is set to match block size
+ */
+ break;
+ }
+
+ scan_segment(sb, segno);
+ }
+ super->s_sweeper = segno;
+ return ret;
+}
+
+/*
+ * In principle, this function should loop forever, looking for GC candidates
+ * and moving data. LogFS is designed in such a way that this loop is
+ * guaranteed to terminate.
+ *
+ * Limiting the loop to some iterations serves purely to catch cases when
+ * these guarantees have failed. An actual endless loop is an obvious bug
+ * and should be reported as such.
+ */
+static void __logfs_gc_pass(struct super_block *sb, int target)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_block *block;
+ int round, progress, last_progress = 0;
+
+ if (no_free_segments(sb) >= target &&
+ super->s_no_object_aliases < MAX_OBJ_ALIASES)
+ return;
+
+ log_gc("__logfs_gc_pass(%x)\n", target);
+ for (round = 0; round < SCAN_ROUNDS; ) {
+ if (no_free_segments(sb) >= target)
+ goto write_alias;
+
+ /* Sync in-memory state with on-medium state in case they
+ * diverged */
+ logfs_write_anchor(sb);
+ round += logfs_scan_some(sb);
+ if (no_free_segments(sb) >= target)
+ goto write_alias;
+ progress = logfs_gc_once(sb);
+ if (progress)
+ last_progress = round;
+ else if (round - last_progress > 2)
+ break;
+ continue;
+
+ /*
+ * The goto logic is nasty, I just don't know a better way to
+ * code it. GC is supposed to ensure two things:
+ * 1. Enough free segments are available.
+ * 2. The number of aliases is bounded.
+ * When 1. is achieved, we take a look at 2. and write back
+ * some alias-containing blocks, if necessary. However, after
+ * each such write we need to go back to 1., as writes can
+ * consume free segments.
+ */
+write_alias:
+ if (super->s_no_object_aliases < MAX_OBJ_ALIASES)
+ return;
+ if (list_empty(&super->s_object_alias)) {
+ /* All aliases are still in btree */
+ return;
+ }
+ log_gc("Write back one alias\n");
+ block = list_entry(super->s_object_alias.next,
+ struct logfs_block, alias_list);
+ block->ops->write_block(block);
+ /*
+ * To round off the nasty goto logic, we reset round here. It
+ * is a safety-net for GC not making any progress and limited
+ * to something reasonably small. If incremented it for every
+ * single alias, the loop could terminate rather quickly.
+ */
+ round = 0;
+ }
+ LOGFS_BUG(sb);
+}
+
+static int wl_ratelimit(struct super_block *sb, u64 *next_event)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ if (*next_event < super->s_gec) {
+ *next_event = super->s_gec + WL_RATELIMIT;
+ return 0;
+ }
+ return 1;
+}
+
+static void logfs_wl_pass(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct gc_candidate *wl_cand, *free_cand;
+
+ if (wl_ratelimit(sb, &super->s_wl_gec_ostore))
+ return;
+
+ wl_cand = first_in_list(&super->s_ec_list);
+ if (!wl_cand)
+ return;
+ free_cand = first_in_list(&super->s_free_list);
+ if (!free_cand)
+ return;
+
+ if (wl_cand->erase_count < free_cand->erase_count + WL_DELTA) {
+ remove_from_list(wl_cand);
+ __logfs_gc_once(sb, wl_cand);
+ }
+}
+
+/*
+ * The journal needs wear leveling as well. But moving the journal is an
+ * expensive operation so we try to avoid it as much as possible. And if we
+ * have to do it, we move the whole journal, not individual segments.
+ *
+ * Ratelimiting is not strictly necessary here, it mainly serves to avoid the
+ * calculations. First we check whether moving the journal would be a
+ * significant improvement. That means that a) the current journal segments
+ * have more wear than the future journal segments and b) the current journal
+ * segments have more wear than normal ostore segments.
+ * Rationale for b) is that we don't have to move the journal if it is aging
+ * less than the ostore, even if the reserve segments age even less (they are
+ * excluded from wear leveling, after all).
+ * Next we check that the superblocks have less wear than the journal. Since
+ * moving the journal requires writing the superblocks, we have to protect the
+ * superblocks even more than the journal.
+ *
+ * Also we double the acceptable wear difference, compared to ostore wear
+ * leveling. Journal data is read and rewritten rapidly, comparatively. So
+ * soft errors have much less time to accumulate and we allow the journal to
+ * be a bit worse than the ostore.
+ */
+static void logfs_journal_wl_pass(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct gc_candidate *cand;
+ u32 min_journal_ec = -1, max_reserve_ec = 0;
+ int i;
+
+ if (wl_ratelimit(sb, &super->s_wl_gec_journal))
+ return;
+
+ if (super->s_reserve_list.count < super->s_no_journal_segs) {
+ /* Reserve is not full enough to move complete journal */
+ return;
+ }
+
+ journal_for_each(i)
+ if (super->s_journal_seg[i])
+ min_journal_ec = min(min_journal_ec,
+ super->s_journal_ec[i]);
+ cand = rb_entry(rb_first(&super->s_free_list.rb_tree),
+ struct gc_candidate, rb_node);
+ max_reserve_ec = cand->erase_count;
+ for (i = 0; i < 2; i++) {
+ struct logfs_segment_entry se;
+ u32 segno = seg_no(sb, super->s_sb_ofs[i]);
+ u32 ec;
+
+ logfs_get_segment_entry(sb, segno, &se);
+ ec = be32_to_cpu(se.ec_level) >> 4;
+ max_reserve_ec = max(max_reserve_ec, ec);
+ }
+
+ if (min_journal_ec > max_reserve_ec + 2 * WL_DELTA) {
+ do_logfs_journal_wl_pass(sb);
+ }
+}
+
+void logfs_gc_pass(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ //BUG_ON(mutex_trylock(&logfs_super(sb)->s_w_mutex));
+ /* Write journal before free space is getting saturated with dirty
+ * objects.
+ */
+ if (super->s_dirty_used_bytes + super->s_dirty_free_bytes
+ + LOGFS_MAX_OBJECTSIZE >= super->s_free_bytes)
+ logfs_write_anchor(sb);
+ __logfs_gc_pass(sb, super->s_total_levels);
+ logfs_wl_pass(sb);
+ logfs_journal_wl_pass(sb);
+}
+
+static int check_area(struct super_block *sb, int i)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_area[i];
+ struct logfs_object_header oh;
+ u32 segno = area->a_segno;
+ u32 ofs = area->a_used_bytes;
+ __be32 crc;
+ int err;
+
+ if (!area->a_is_open)
+ return 0;
+
+ for (ofs = area->a_used_bytes;
+ ofs <= super->s_segsize - sizeof(oh);
+ ofs += (u32)be16_to_cpu(oh.len) + sizeof(oh)) {
+ err = wbuf_read(sb, dev_ofs(sb, segno, ofs), sizeof(oh), &oh);
+ if (err)
+ return err;
+
+ if (!memchr_inv(&oh, 0xff, sizeof(oh)))
+ break;
+
+ crc = logfs_crc32(&oh, sizeof(oh) - 4, 4);
+ if (crc != oh.crc) {
+ printk(KERN_INFO "interrupted header at %llx\n",
+ dev_ofs(sb, segno, ofs));
+ return 0;
+ }
+ }
+ if (ofs != area->a_used_bytes) {
+ printk(KERN_INFO "%x bytes unaccounted data found at %llx\n",
+ ofs - area->a_used_bytes,
+ dev_ofs(sb, segno, area->a_used_bytes));
+ area->a_used_bytes = ofs;
+ }
+ return 0;
+}
+
+int logfs_check_areas(struct super_block *sb)
+{
+ int i, err;
+
+ for_each_area(i) {
+ err = check_area(sb, i);
+ if (err)
+ return err;
+ }
+ return 0;
+}
+
+static void logfs_init_candlist(struct candidate_list *list, int maxcount,
+ int sort_by_ec)
+{
+ list->count = 0;
+ list->maxcount = maxcount;
+ list->sort_by_ec = sort_by_ec;
+ list->rb_tree = RB_ROOT;
+}
+
+int logfs_init_gc(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i;
+
+ btree_init_mempool32(&super->s_cand_tree, super->s_btree_pool);
+ logfs_init_candlist(&super->s_free_list, LIST_SIZE + SCAN_RATIO, 1);
+ logfs_init_candlist(&super->s_reserve_list,
+ super->s_bad_seg_reserve, 1);
+ for_each_area(i)
+ logfs_init_candlist(&super->s_low_list[i], LIST_SIZE, 0);
+ logfs_init_candlist(&super->s_ec_list, LIST_SIZE, 1);
+ return 0;
+}
+
+static void logfs_cleanup_list(struct super_block *sb,
+ struct candidate_list *list)
+{
+ struct gc_candidate *cand;
+
+ while (list->count) {
+ cand = rb_entry(list->rb_tree.rb_node, struct gc_candidate,
+ rb_node);
+ remove_from_list(cand);
+ free_candidate(sb, cand);
+ }
+ BUG_ON(list->rb_tree.rb_node);
+}
+
+void logfs_cleanup_gc(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i;
+
+ if (!super->s_free_list.count)
+ return;
+
+ /*
+ * FIXME: The btree may still contain a single empty node. So we
+ * call the grim visitor to clean up that mess. Btree code should
+ * do it for us, really.
+ */
+ btree_grim_visitor32(&super->s_cand_tree, 0, NULL);
+ logfs_cleanup_list(sb, &super->s_free_list);
+ logfs_cleanup_list(sb, &super->s_reserve_list);
+ for_each_area(i)
+ logfs_cleanup_list(sb, &super->s_low_list[i]);
+ logfs_cleanup_list(sb, &super->s_ec_list);
+}
diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c
new file mode 100644
index 000000000000..33ec1aeaeec4
--- /dev/null
+++ b/fs/logfs/inode.c
@@ -0,0 +1,417 @@
+/*
+ * fs/logfs/inode.c - inode handling code
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+#include <linux/writeback.h>
+#include <linux/backing-dev.h>
+
+/*
+ * How soon to reuse old inode numbers? LogFS doesn't store deleted inodes
+ * on the medium. It therefore also lacks a method to store the previous
+ * generation number for deleted inodes. Instead a single generation number
+ * is stored which will be used for new inodes. Being just a 32bit counter,
+ * this can obvious wrap relatively quickly. So we only reuse inodes if we
+ * know that a fair number of inodes can be created before we have to increment
+ * the generation again - effectively adding some bits to the counter.
+ * But being too aggressive here means we keep a very large and very sparse
+ * inode file, wasting space on indirect blocks.
+ * So what is a good value? Beats me. 64k seems moderately bad on both
+ * fronts, so let's use that for now...
+ *
+ * NFS sucks, as everyone already knows.
+ */
+#define INOS_PER_WRAP (0x10000)
+
+/*
+ * Logfs' requirement to read inodes for garbage collection makes life a bit
+ * harder. GC may have to read inodes that are in I_FREEING state, when they
+ * are being written out - and waiting for GC to make progress, naturally.
+ *
+ * So we cannot just call iget() or some variant of it, but first have to check
+ * wether the inode in question might be in I_FREEING state. Therefore we
+ * maintain our own per-sb list of "almost deleted" inodes and check against
+ * that list first. Normally this should be at most 1-2 entries long.
+ *
+ * Also, inodes have logfs-specific reference counting on top of what the vfs
+ * does. When .destroy_inode is called, normally the reference count will drop
+ * to zero and the inode gets deleted. But if GC accessed the inode, its
+ * refcount will remain nonzero and final deletion will have to wait.
+ *
+ * As a result we have two sets of functions to get/put inodes:
+ * logfs_safe_iget/logfs_safe_iput - safe to call from GC context
+ * logfs_iget/iput - normal version
+ */
+static struct kmem_cache *logfs_inode_cache;
+
+static DEFINE_SPINLOCK(logfs_inode_lock);
+
+static void logfs_inode_setops(struct inode *inode)
+{
+ switch (inode->i_mode & S_IFMT) {
+ case S_IFDIR:
+ inode->i_op = &logfs_dir_iops;
+ inode->i_fop = &logfs_dir_fops;
+ inode->i_mapping->a_ops = &logfs_reg_aops;
+ break;
+ case S_IFREG:
+ inode->i_op = &logfs_reg_iops;
+ inode->i_fop = &logfs_reg_fops;
+ inode->i_mapping->a_ops = &logfs_reg_aops;
+ break;
+ case S_IFLNK:
+ inode->i_op = &logfs_symlink_iops;
+ inode->i_mapping->a_ops = &logfs_reg_aops;
+ break;
+ case S_IFSOCK: /* fall through */
+ case S_IFBLK: /* fall through */
+ case S_IFCHR: /* fall through */
+ case S_IFIFO:
+ init_special_inode(inode, inode->i_mode, inode->i_rdev);
+ break;
+ default:
+ BUG();
+ }
+}
+
+static struct inode *__logfs_iget(struct super_block *sb, ino_t ino)
+{
+ struct inode *inode = iget_locked(sb, ino);
+ int err;
+
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+ if (!(inode->i_state & I_NEW))
+ return inode;
+
+ err = logfs_read_inode(inode);
+ if (err || inode->i_nlink == 0) {
+ /* inode->i_nlink == 0 can be true when called from
+ * block validator */
+ /* set i_nlink to 0 to prevent caching */
+ inode->i_nlink = 0;
+ logfs_inode(inode)->li_flags |= LOGFS_IF_ZOMBIE;
+ iget_failed(inode);
+ if (!err)
+ err = -ENOENT;
+ return ERR_PTR(err);
+ }
+
+ logfs_inode_setops(inode);
+ unlock_new_inode(inode);
+ return inode;
+}
+
+struct inode *logfs_iget(struct super_block *sb, ino_t ino)
+{
+ BUG_ON(ino == LOGFS_INO_MASTER);
+ BUG_ON(ino == LOGFS_INO_SEGFILE);
+ return __logfs_iget(sb, ino);
+}
+
+/*
+ * is_cached is set to 1 if we hand out a cached inode, 0 otherwise.
+ * this allows logfs_iput to do the right thing later
+ */
+struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *is_cached)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_inode *li;
+
+ if (ino == LOGFS_INO_MASTER)
+ return super->s_master_inode;
+ if (ino == LOGFS_INO_SEGFILE)
+ return super->s_segfile_inode;
+
+ spin_lock(&logfs_inode_lock);
+ list_for_each_entry(li, &super->s_freeing_list, li_freeing_list)
+ if (li->vfs_inode.i_ino == ino) {
+ li->li_refcount++;
+ spin_unlock(&logfs_inode_lock);
+ *is_cached = 1;
+ return &li->vfs_inode;
+ }
+ spin_unlock(&logfs_inode_lock);
+
+ *is_cached = 0;
+ return __logfs_iget(sb, ino);
+}
+
+static void __logfs_destroy_inode(struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ BUG_ON(li->li_block);
+ list_del(&li->li_freeing_list);
+ kmem_cache_free(logfs_inode_cache, li);
+}
+
+static void logfs_destroy_inode(struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ BUG_ON(list_empty(&li->li_freeing_list));
+ spin_lock(&logfs_inode_lock);
+ li->li_refcount--;
+ if (li->li_refcount == 0)
+ __logfs_destroy_inode(inode);
+ spin_unlock(&logfs_inode_lock);
+}
+
+void logfs_safe_iput(struct inode *inode, int is_cached)
+{
+ if (inode->i_ino == LOGFS_INO_MASTER)
+ return;
+ if (inode->i_ino == LOGFS_INO_SEGFILE)
+ return;
+
+ if (is_cached) {
+ logfs_destroy_inode(inode);
+ return;
+ }
+
+ iput(inode);
+}
+
+static void logfs_init_inode(struct super_block *sb, struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ int i;
+
+ li->li_flags = 0;
+ li->li_height = 0;
+ li->li_used_bytes = 0;
+ li->li_block = NULL;
+ inode->i_uid = 0;
+ inode->i_gid = 0;
+ inode->i_size = 0;
+ inode->i_blocks = 0;
+ inode->i_ctime = CURRENT_TIME;
+ inode->i_mtime = CURRENT_TIME;
+ inode->i_nlink = 1;
+ INIT_LIST_HEAD(&li->li_freeing_list);
+
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ li->li_data[i] = 0;
+
+ return;
+}
+
+static struct inode *logfs_alloc_inode(struct super_block *sb)
+{
+ struct logfs_inode *li;
+
+ li = kmem_cache_alloc(logfs_inode_cache, GFP_NOFS);
+ if (!li)
+ return NULL;
+ logfs_init_inode(sb, &li->vfs_inode);
+ return &li->vfs_inode;
+}
+
+/*
+ * In logfs inodes are written to an inode file. The inode file, like any
+ * other file, is managed with a inode. The inode file's inode, aka master
+ * inode, requires special handling in several respects. First, it cannot be
+ * written to the inode file, so it is stored in the journal instead.
+ *
+ * Secondly, this inode cannot be written back and destroyed before all other
+ * inodes have been written. The ordering is important. Linux' VFS is happily
+ * unaware of the ordering constraint and would ordinarily destroy the master
+ * inode at umount time while other inodes are still in use and dirty. Not
+ * good.
+ *
+ * So logfs makes sure the master inode is not written until all other inodes
+ * have been destroyed. Sadly, this method has another side-effect. The VFS
+ * will notice one remaining inode and print a frightening warning message.
+ * Worse, it is impossible to judge whether such a warning was caused by the
+ * master inode or any other inodes have leaked as well.
+ *
+ * Our attempt of solving this is with logfs_new_meta_inode() below. Its
+ * purpose is to create a new inode that will not trigger the warning if such
+ * an inode is still in use. An ugly hack, no doubt. Suggections for
+ * improvement are welcome.
+ */
+struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino)
+{
+ struct inode *inode;
+
+ inode = logfs_alloc_inode(sb);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+
+ inode->i_mode = S_IFREG;
+ inode->i_ino = ino;
+ inode->i_sb = sb;
+
+ /* This is a blatant copy of alloc_inode code. We'd need alloc_inode
+ * to be nonstatic, alas. */
+ {
+ struct address_space * const mapping = &inode->i_data;
+
+ mapping->a_ops = &logfs_reg_aops;
+ mapping->host = inode;
+ mapping->flags = 0;
+ mapping_set_gfp_mask(mapping, GFP_NOFS);
+ mapping->assoc_mapping = NULL;
+ mapping->backing_dev_info = &default_backing_dev_info;
+ inode->i_mapping = mapping;
+ inode->i_nlink = 1;
+ }
+
+ return inode;
+}
+
+struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino)
+{
+ struct inode *inode;
+ int err;
+
+ inode = logfs_new_meta_inode(sb, ino);
+ if (IS_ERR(inode))
+ return inode;
+
+ err = logfs_read_inode(inode);
+ if (err) {
+ destroy_meta_inode(inode);
+ return ERR_PTR(err);
+ }
+ logfs_inode_setops(inode);
+ return inode;
+}
+
+static int logfs_write_inode(struct inode *inode, struct writeback_control *wbc)
+{
+ int ret;
+ long flags = WF_LOCK;
+
+ /* Can only happen if creat() failed. Safe to skip. */
+ if (logfs_inode(inode)->li_flags & LOGFS_IF_STILLBORN)
+ return 0;
+
+ ret = __logfs_write_inode(inode, flags);
+ LOGFS_BUG_ON(ret, inode->i_sb);
+ return ret;
+}
+
+void destroy_meta_inode(struct inode *inode)
+{
+ if (inode) {
+ if (inode->i_data.nrpages)
+ truncate_inode_pages(&inode->i_data, 0);
+ logfs_clear_inode(inode);
+ kmem_cache_free(logfs_inode_cache, logfs_inode(inode));
+ }
+}
+
+/* called with inode_lock held */
+static void logfs_drop_inode(struct inode *inode)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ struct logfs_inode *li = logfs_inode(inode);
+
+ spin_lock(&logfs_inode_lock);
+ list_move(&li->li_freeing_list, &super->s_freeing_list);
+ spin_unlock(&logfs_inode_lock);
+ generic_drop_inode(inode);
+}
+
+static void logfs_set_ino_generation(struct super_block *sb,
+ struct inode *inode)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u64 ino;
+
+ mutex_lock(&super->s_journal_mutex);
+ ino = logfs_seek_hole(super->s_master_inode, super->s_last_ino);
+ super->s_last_ino = ino;
+ super->s_inos_till_wrap--;
+ if (super->s_inos_till_wrap < 0) {
+ super->s_last_ino = LOGFS_RESERVED_INOS;
+ super->s_generation++;
+ super->s_inos_till_wrap = INOS_PER_WRAP;
+ }
+ inode->i_ino = ino;
+ inode->i_generation = super->s_generation;
+ mutex_unlock(&super->s_journal_mutex);
+}
+
+struct inode *logfs_new_inode(struct inode *dir, int mode)
+{
+ struct super_block *sb = dir->i_sb;
+ struct inode *inode;
+
+ inode = new_inode(sb);
+ if (!inode)
+ return ERR_PTR(-ENOMEM);
+
+ logfs_init_inode(sb, inode);
+
+ /* inherit parent flags */
+ logfs_inode(inode)->li_flags |=
+ logfs_inode(dir)->li_flags & LOGFS_FL_INHERITED;
+
+ inode->i_mode = mode;
+ logfs_set_ino_generation(sb, inode);
+
+ inode->i_uid = current_fsuid();
+ inode->i_gid = current_fsgid();
+ if (dir->i_mode & S_ISGID) {
+ inode->i_gid = dir->i_gid;
+ if (S_ISDIR(mode))
+ inode->i_mode |= S_ISGID;
+ }
+
+ logfs_inode_setops(inode);
+ insert_inode_hash(inode);
+
+ return inode;
+}
+
+static void logfs_init_once(void *_li)
+{
+ struct logfs_inode *li = _li;
+ int i;
+
+ li->li_flags = 0;
+ li->li_used_bytes = 0;
+ li->li_refcount = 1;
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ li->li_data[i] = 0;
+ inode_init_once(&li->vfs_inode);
+}
+
+static int logfs_sync_fs(struct super_block *sb, int wait)
+{
+ /* FIXME: write anchor */
+ logfs_super(sb)->s_devops->sync(sb);
+ return 0;
+}
+
+const struct super_operations logfs_super_operations = {
+ .alloc_inode = logfs_alloc_inode,
+ .clear_inode = logfs_clear_inode,
+ .delete_inode = logfs_delete_inode,
+ .destroy_inode = logfs_destroy_inode,
+ .drop_inode = logfs_drop_inode,
+ .write_inode = logfs_write_inode,
+ .statfs = logfs_statfs,
+ .sync_fs = logfs_sync_fs,
+};
+
+int logfs_init_inode_cache(void)
+{
+ logfs_inode_cache = kmem_cache_create("logfs_inode_cache",
+ sizeof(struct logfs_inode), 0, SLAB_RECLAIM_ACCOUNT,
+ logfs_init_once);
+ if (!logfs_inode_cache)
+ return -ENOMEM;
+ return 0;
+}
+
+void logfs_destroy_inode_cache(void)
+{
+ kmem_cache_destroy(logfs_inode_cache);
+}
diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c
new file mode 100644
index 000000000000..6ad30a4c9052
--- /dev/null
+++ b/fs/logfs/journal.c
@@ -0,0 +1,883 @@
+/*
+ * fs/logfs/journal.c - journal handling code
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ */
+#include "logfs.h"
+
+static void logfs_calc_free(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u64 reserve, no_segs = super->s_no_segs;
+ s64 free;
+ int i;
+
+ /* superblock segments */
+ no_segs -= 2;
+ super->s_no_journal_segs = 0;
+ /* journal */
+ journal_for_each(i)
+ if (super->s_journal_seg[i]) {
+ no_segs--;
+ super->s_no_journal_segs++;
+ }
+
+ /* open segments plus one extra per level for GC */
+ no_segs -= 2 * super->s_total_levels;
+
+ free = no_segs * (super->s_segsize - LOGFS_SEGMENT_RESERVE);
+ free -= super->s_used_bytes;
+ /* just a bit extra */
+ free -= super->s_total_levels * 4096;
+
+ /* Bad blocks are 'paid' for with speed reserve - the filesystem
+ * simply gets slower as bad blocks accumulate. Until the bad blocks
+ * exceed the speed reserve - then the filesystem gets smaller.
+ */
+ reserve = super->s_bad_segments + super->s_bad_seg_reserve;
+ reserve *= super->s_segsize - LOGFS_SEGMENT_RESERVE;
+ reserve = max(reserve, super->s_speed_reserve);
+ free -= reserve;
+ if (free < 0)
+ free = 0;
+
+ super->s_free_bytes = free;
+}
+
+static void reserve_sb_and_journal(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct btree_head32 *head = &super->s_reserved_segments;
+ int i, err;
+
+ err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[0]), (void *)1,
+ GFP_KERNEL);
+ BUG_ON(err);
+
+ err = btree_insert32(head, seg_no(sb, super->s_sb_ofs[1]), (void *)1,
+ GFP_KERNEL);
+ BUG_ON(err);
+
+ journal_for_each(i) {
+ if (!super->s_journal_seg[i])
+ continue;
+ err = btree_insert32(head, super->s_journal_seg[i], (void *)1,
+ GFP_KERNEL);
+ BUG_ON(err);
+ }
+}
+
+static void read_dynsb(struct super_block *sb,
+ struct logfs_je_dynsb *dynsb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ super->s_gec = be64_to_cpu(dynsb->ds_gec);
+ super->s_sweeper = be64_to_cpu(dynsb->ds_sweeper);
+ super->s_victim_ino = be64_to_cpu(dynsb->ds_victim_ino);
+ super->s_rename_dir = be64_to_cpu(dynsb->ds_rename_dir);
+ super->s_rename_pos = be64_to_cpu(dynsb->ds_rename_pos);
+ super->s_used_bytes = be64_to_cpu(dynsb->ds_used_bytes);
+ super->s_generation = be32_to_cpu(dynsb->ds_generation);
+}
+
+static void read_anchor(struct super_block *sb,
+ struct logfs_je_anchor *da)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode = super->s_master_inode;
+ struct logfs_inode *li = logfs_inode(inode);
+ int i;
+
+ super->s_last_ino = be64_to_cpu(da->da_last_ino);
+ li->li_flags = 0;
+ li->li_height = da->da_height;
+ i_size_write(inode, be64_to_cpu(da->da_size));
+ li->li_used_bytes = be64_to_cpu(da->da_used_bytes);
+
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ li->li_data[i] = be64_to_cpu(da->da_data[i]);
+}
+
+static void read_erasecount(struct super_block *sb,
+ struct logfs_je_journal_ec *ec)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i;
+
+ journal_for_each(i)
+ super->s_journal_ec[i] = be32_to_cpu(ec->ec[i]);
+}
+
+static int read_area(struct super_block *sb, struct logfs_je_area *a)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_area[a->gc_level];
+ u64 ofs;
+ u32 writemask = ~(super->s_writesize - 1);
+
+ if (a->gc_level >= LOGFS_NO_AREAS)
+ return -EIO;
+ if (a->vim != VIM_DEFAULT)
+ return -EIO; /* TODO: close area and continue */
+
+ area->a_used_bytes = be32_to_cpu(a->used_bytes);
+ area->a_written_bytes = area->a_used_bytes & writemask;
+ area->a_segno = be32_to_cpu(a->segno);
+ if (area->a_segno)
+ area->a_is_open = 1;
+
+ ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes);
+ if (super->s_writesize > 1)
+ logfs_buf_recover(area, ofs, a + 1, super->s_writesize);
+ else
+ logfs_buf_recover(area, ofs, NULL, 0);
+ return 0;
+}
+
+static void *unpack(void *from, void *to)
+{
+ struct logfs_journal_header *jh = from;
+ void *data = from + sizeof(struct logfs_journal_header);
+ int err;
+ size_t inlen, outlen;
+
+ inlen = be16_to_cpu(jh->h_len);
+ outlen = be16_to_cpu(jh->h_datalen);
+
+ if (jh->h_compr == COMPR_NONE)
+ memcpy(to, data, inlen);
+ else {
+ err = logfs_uncompress(data, to, inlen, outlen);
+ BUG_ON(err);
+ }
+ return to;
+}
+
+static int __read_je_header(struct super_block *sb, u64 ofs,
+ struct logfs_journal_header *jh)
+{
+ struct logfs_super *super = logfs_super(sb);
+ size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize)
+ + MAX_JOURNAL_HEADER;
+ u16 type, len, datalen;
+ int err;
+
+ /* read header only */
+ err = wbuf_read(sb, ofs, sizeof(*jh), jh);
+ if (err)
+ return err;
+ type = be16_to_cpu(jh->h_type);
+ len = be16_to_cpu(jh->h_len);
+ datalen = be16_to_cpu(jh->h_datalen);
+ if (len > sb->s_blocksize)
+ return -EIO;
+ if ((type < JE_FIRST) || (type > JE_LAST))
+ return -EIO;
+ if (datalen > bufsize)
+ return -EIO;
+ return 0;
+}
+
+static int __read_je_payload(struct super_block *sb, u64 ofs,
+ struct logfs_journal_header *jh)
+{
+ u16 len;
+ int err;
+
+ len = be16_to_cpu(jh->h_len);
+ err = wbuf_read(sb, ofs + sizeof(*jh), len, jh + 1);
+ if (err)
+ return err;
+ if (jh->h_crc != logfs_crc32(jh, len + sizeof(*jh), 4)) {
+ /* Old code was confused. It forgot about the header length
+ * and stopped calculating the crc 16 bytes before the end
+ * of data - ick!
+ * FIXME: Remove this hack once the old code is fixed.
+ */
+ if (jh->h_crc == logfs_crc32(jh, len, 4))
+ WARN_ON_ONCE(1);
+ else
+ return -EIO;
+ }
+ return 0;
+}
+
+/*
+ * jh needs to be large enough to hold the complete entry, not just the header
+ */
+static int __read_je(struct super_block *sb, u64 ofs,
+ struct logfs_journal_header *jh)
+{
+ int err;
+
+ err = __read_je_header(sb, ofs, jh);
+ if (err)
+ return err;
+ return __read_je_payload(sb, ofs, jh);
+}
+
+static int read_je(struct super_block *sb, u64 ofs)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_journal_header *jh = super->s_compressed_je;
+ void *scratch = super->s_je;
+ u16 type, datalen;
+ int err;
+
+ err = __read_je(sb, ofs, jh);
+ if (err)
+ return err;
+ type = be16_to_cpu(jh->h_type);
+ datalen = be16_to_cpu(jh->h_datalen);
+
+ switch (type) {
+ case JE_DYNSB:
+ read_dynsb(sb, unpack(jh, scratch));
+ break;
+ case JE_ANCHOR:
+ read_anchor(sb, unpack(jh, scratch));
+ break;
+ case JE_ERASECOUNT:
+ read_erasecount(sb, unpack(jh, scratch));
+ break;
+ case JE_AREA:
+ read_area(sb, unpack(jh, scratch));
+ break;
+ case JE_OBJ_ALIAS:
+ err = logfs_load_object_aliases(sb, unpack(jh, scratch),
+ datalen);
+ break;
+ default:
+ WARN_ON_ONCE(1);
+ return -EIO;
+ }
+ return err;
+}
+
+static int logfs_read_segment(struct super_block *sb, u32 segno)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_journal_header *jh = super->s_compressed_je;
+ u64 ofs, seg_ofs = dev_ofs(sb, segno, 0);
+ u32 h_ofs, last_ofs = 0;
+ u16 len, datalen, last_len = 0;
+ int i, err;
+
+ /* search for most recent commit */
+ for (h_ofs = 0; h_ofs < super->s_segsize; h_ofs += sizeof(*jh)) {
+ ofs = seg_ofs + h_ofs;
+ err = __read_je_header(sb, ofs, jh);
+ if (err)
+ continue;
+ if (jh->h_type != cpu_to_be16(JE_COMMIT))
+ continue;
+ err = __read_je_payload(sb, ofs, jh);
+ if (err)
+ continue;
+ len = be16_to_cpu(jh->h_len);
+ datalen = be16_to_cpu(jh->h_datalen);
+ if ((datalen > sizeof(super->s_je_array)) ||
+ (datalen % sizeof(__be64)))
+ continue;
+ last_ofs = h_ofs;
+ last_len = datalen;
+ h_ofs += ALIGN(len, sizeof(*jh)) - sizeof(*jh);
+ }
+ /* read commit */
+ if (last_ofs == 0)
+ return -ENOENT;
+ ofs = seg_ofs + last_ofs;
+ log_journal("Read commit from %llx\n", ofs);
+ err = __read_je(sb, ofs, jh);
+ BUG_ON(err); /* We should have caught it in the scan loop already */
+ if (err)
+ return err;
+ /* uncompress */
+ unpack(jh, super->s_je_array);
+ super->s_no_je = last_len / sizeof(__be64);
+ /* iterate over array */
+ for (i = 0; i < super->s_no_je; i++) {
+ err = read_je(sb, be64_to_cpu(super->s_je_array[i]));
+ if (err)
+ return err;
+ }
+ super->s_journal_area->a_segno = segno;
+ return 0;
+}
+
+static u64 read_gec(struct super_block *sb, u32 segno)
+{
+ struct logfs_segment_header sh;
+ __be32 crc;
+ int err;
+
+ if (!segno)
+ return 0;
+ err = wbuf_read(sb, dev_ofs(sb, segno, 0), sizeof(sh), &sh);
+ if (err)
+ return 0;
+ crc = logfs_crc32(&sh, sizeof(sh), 4);
+ if (crc != sh.crc) {
+ WARN_ON(sh.gec != cpu_to_be64(0xffffffffffffffffull));
+ /* Most likely it was just erased */
+ return 0;
+ }
+ return be64_to_cpu(sh.gec);
+}
+
+static int logfs_read_journal(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u64 gec[LOGFS_JOURNAL_SEGS], max;
+ u32 segno;
+ int i, max_i;
+
+ max = 0;
+ max_i = -1;
+ journal_for_each(i) {
+ segno = super->s_journal_seg[i];
+ gec[i] = read_gec(sb, super->s_journal_seg[i]);
+ if (gec[i] > max) {
+ max = gec[i];
+ max_i = i;
+ }
+ }
+ if (max_i == -1)
+ return -EIO;
+ /* FIXME: Try older segments in case of error */
+ return logfs_read_segment(sb, super->s_journal_seg[max_i]);
+}
+
+/*
+ * First search the current segment (outer loop), then pick the next segment
+ * in the array, skipping any zero entries (inner loop).
+ */
+static void journal_get_free_segment(struct logfs_area *area)
+{
+ struct logfs_super *super = logfs_super(area->a_sb);
+ int i;
+
+ journal_for_each(i) {
+ if (area->a_segno != super->s_journal_seg[i])
+ continue;
+
+ do {
+ i++;
+ if (i == LOGFS_JOURNAL_SEGS)
+ i = 0;
+ } while (!super->s_journal_seg[i]);
+
+ area->a_segno = super->s_journal_seg[i];
+ area->a_erase_count = ++(super->s_journal_ec[i]);
+ log_journal("Journal now at %x (ec %x)\n", area->a_segno,
+ area->a_erase_count);
+ return;
+ }
+ BUG();
+}
+
+static void journal_get_erase_count(struct logfs_area *area)
+{
+ /* erase count is stored globally and incremented in
+ * journal_get_free_segment() - nothing to do here */
+}
+
+static int journal_erase_segment(struct logfs_area *area)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_segment_header sh;
+ u64 ofs;
+ int err;
+
+ err = logfs_erase_segment(sb, area->a_segno, 1);
+ if (err)
+ return err;
+
+ sh.pad = 0;
+ sh.type = SEG_JOURNAL;
+ sh.level = 0;
+ sh.segno = cpu_to_be32(area->a_segno);
+ sh.ec = cpu_to_be32(area->a_erase_count);
+ sh.gec = cpu_to_be64(logfs_super(sb)->s_gec);
+ sh.crc = logfs_crc32(&sh, sizeof(sh), 4);
+
+ /* This causes a bug in segment.c. Not yet. */
+ //logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count, 0);
+
+ ofs = dev_ofs(sb, area->a_segno, 0);
+ area->a_used_bytes = ALIGN(sizeof(sh), 16);
+ logfs_buf_write(area, ofs, &sh, sizeof(sh));
+ return 0;
+}
+
+static size_t __logfs_write_header(struct logfs_super *super,
+ struct logfs_journal_header *jh, size_t len, size_t datalen,
+ u16 type, u8 compr)
+{
+ jh->h_len = cpu_to_be16(len);
+ jh->h_type = cpu_to_be16(type);
+ jh->h_datalen = cpu_to_be16(datalen);
+ jh->h_compr = compr;
+ jh->h_pad[0] = 'H';
+ jh->h_pad[1] = 'E';
+ jh->h_pad[2] = 'A';
+ jh->h_pad[3] = 'D';
+ jh->h_pad[4] = 'R';
+ jh->h_crc = logfs_crc32(jh, len + sizeof(*jh), 4);
+ return ALIGN(len, 16) + sizeof(*jh);
+}
+
+static size_t logfs_write_header(struct logfs_super *super,
+ struct logfs_journal_header *jh, size_t datalen, u16 type)
+{
+ size_t len = datalen;
+
+ return __logfs_write_header(super, jh, len, datalen, type, COMPR_NONE);
+}
+
+static inline size_t logfs_journal_erasecount_size(struct logfs_super *super)
+{
+ return LOGFS_JOURNAL_SEGS * sizeof(__be32);
+}
+
+static void *logfs_write_erasecount(struct super_block *sb, void *_ec,
+ u16 *type, size_t *len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_je_journal_ec *ec = _ec;
+ int i;
+
+ journal_for_each(i)
+ ec->ec[i] = cpu_to_be32(super->s_journal_ec[i]);
+ *type = JE_ERASECOUNT;
+ *len = logfs_journal_erasecount_size(super);
+ return ec;
+}
+
+static void account_shadow(void *_shadow, unsigned long _sb, u64 ignore,
+ size_t ignore2)
+{
+ struct logfs_shadow *shadow = _shadow;
+ struct super_block *sb = (void *)_sb;
+ struct logfs_super *super = logfs_super(sb);
+
+ /* consume new space */
+ super->s_free_bytes -= shadow->new_len;
+ super->s_used_bytes += shadow->new_len;
+ super->s_dirty_used_bytes -= shadow->new_len;
+
+ /* free up old space */
+ super->s_free_bytes += shadow->old_len;
+ super->s_used_bytes -= shadow->old_len;
+ super->s_dirty_free_bytes -= shadow->old_len;
+
+ logfs_set_segment_used(sb, shadow->old_ofs, -shadow->old_len);
+ logfs_set_segment_used(sb, shadow->new_ofs, shadow->new_len);
+
+ log_journal("account_shadow(%llx, %llx, %x) %llx->%llx %x->%x\n",
+ shadow->ino, shadow->bix, shadow->gc_level,
+ shadow->old_ofs, shadow->new_ofs,
+ shadow->old_len, shadow->new_len);
+ mempool_free(shadow, super->s_shadow_pool);
+}
+
+static void account_shadows(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode = super->s_master_inode;
+ struct logfs_inode *li = logfs_inode(inode);
+ struct shadow_tree *tree = &super->s_shadow_tree;
+
+ btree_grim_visitor64(&tree->new, (unsigned long)sb, account_shadow);
+ btree_grim_visitor64(&tree->old, (unsigned long)sb, account_shadow);
+
+ if (li->li_block) {
+ /*
+ * We never actually use the structure, when attached to the
+ * master inode. But it is easier to always free it here than
+ * to have checks in several places elsewhere when allocating
+ * it.
+ */
+ li->li_block->ops->free_block(sb, li->li_block);
+ }
+ BUG_ON((s64)li->li_used_bytes < 0);
+}
+
+static void *__logfs_write_anchor(struct super_block *sb, void *_da,
+ u16 *type, size_t *len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_je_anchor *da = _da;
+ struct inode *inode = super->s_master_inode;
+ struct logfs_inode *li = logfs_inode(inode);
+ int i;
+
+ da->da_height = li->li_height;
+ da->da_last_ino = cpu_to_be64(super->s_last_ino);
+ da->da_size = cpu_to_be64(i_size_read(inode));
+ da->da_used_bytes = cpu_to_be64(li->li_used_bytes);
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ da->da_data[i] = cpu_to_be64(li->li_data[i]);
+ *type = JE_ANCHOR;
+ *len = sizeof(*da);
+ return da;
+}
+
+static void *logfs_write_dynsb(struct super_block *sb, void *_dynsb,
+ u16 *type, size_t *len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_je_dynsb *dynsb = _dynsb;
+
+ dynsb->ds_gec = cpu_to_be64(super->s_gec);
+ dynsb->ds_sweeper = cpu_to_be64(super->s_sweeper);
+ dynsb->ds_victim_ino = cpu_to_be64(super->s_victim_ino);
+ dynsb->ds_rename_dir = cpu_to_be64(super->s_rename_dir);
+ dynsb->ds_rename_pos = cpu_to_be64(super->s_rename_pos);
+ dynsb->ds_used_bytes = cpu_to_be64(super->s_used_bytes);
+ dynsb->ds_generation = cpu_to_be32(super->s_generation);
+ *type = JE_DYNSB;
+ *len = sizeof(*dynsb);
+ return dynsb;
+}
+
+static void write_wbuf(struct super_block *sb, struct logfs_area *area,
+ void *wbuf)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ u64 ofs;
+ pgoff_t index;
+ int page_ofs;
+ struct page *page;
+
+ ofs = dev_ofs(sb, area->a_segno,
+ area->a_used_bytes & ~(super->s_writesize - 1));
+ index = ofs >> PAGE_SHIFT;
+ page_ofs = ofs & (PAGE_SIZE - 1);
+
+ page = find_lock_page(mapping, index);
+ BUG_ON(!page);
+ memcpy(wbuf, page_address(page) + page_ofs, super->s_writesize);
+ unlock_page(page);
+}
+
+static void *logfs_write_area(struct super_block *sb, void *_a,
+ u16 *type, size_t *len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_area[super->s_sum_index];
+ struct logfs_je_area *a = _a;
+
+ a->vim = VIM_DEFAULT;
+ a->gc_level = super->s_sum_index;
+ a->used_bytes = cpu_to_be32(area->a_used_bytes);
+ a->segno = cpu_to_be32(area->a_segno);
+ if (super->s_writesize > 1)
+ write_wbuf(sb, area, a + 1);
+
+ *type = JE_AREA;
+ *len = sizeof(*a) + super->s_writesize;
+ return a;
+}
+
+static void *logfs_write_commit(struct super_block *sb, void *h,
+ u16 *type, size_t *len)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ *type = JE_COMMIT;
+ *len = super->s_no_je * sizeof(__be64);
+ return super->s_je_array;
+}
+
+static size_t __logfs_write_je(struct super_block *sb, void *buf, u16 type,
+ size_t len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ void *header = super->s_compressed_je;
+ void *data = header + sizeof(struct logfs_journal_header);
+ ssize_t compr_len, pad_len;
+ u8 compr = COMPR_ZLIB;
+
+ if (len == 0)
+ return logfs_write_header(super, header, 0, type);
+
+ compr_len = logfs_compress(buf, data, len, sb->s_blocksize);
+ if (compr_len < 0 || type == JE_ANCHOR) {
+ BUG_ON(len > sb->s_blocksize);
+ memcpy(data, buf, len);
+ compr_len = len;
+ compr = COMPR_NONE;
+ }
+
+ pad_len = ALIGN(compr_len, 16);
+ memset(data + compr_len, 0, pad_len - compr_len);
+
+ return __logfs_write_header(super, header, compr_len, len, type, compr);
+}
+
+static s64 logfs_get_free_bytes(struct logfs_area *area, size_t *bytes,
+ int must_pad)
+{
+ u32 writesize = logfs_super(area->a_sb)->s_writesize;
+ s32 ofs;
+ int ret;
+
+ ret = logfs_open_area(area, *bytes);
+ if (ret)
+ return -EAGAIN;
+
+ ofs = area->a_used_bytes;
+ area->a_used_bytes += *bytes;
+
+ if (must_pad) {
+ area->a_used_bytes = ALIGN(area->a_used_bytes, writesize);
+ *bytes = area->a_used_bytes - ofs;
+ }
+
+ return dev_ofs(area->a_sb, area->a_segno, ofs);
+}
+
+static int logfs_write_je_buf(struct super_block *sb, void *buf, u16 type,
+ size_t buf_len)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_journal_area;
+ struct logfs_journal_header *jh = super->s_compressed_je;
+ size_t len;
+ int must_pad = 0;
+ s64 ofs;
+
+ len = __logfs_write_je(sb, buf, type, buf_len);
+ if (jh->h_type == cpu_to_be16(JE_COMMIT))
+ must_pad = 1;
+
+ ofs = logfs_get_free_bytes(area, &len, must_pad);
+ if (ofs < 0)
+ return ofs;
+ logfs_buf_write(area, ofs, super->s_compressed_je, len);
+ super->s_je_array[super->s_no_je++] = cpu_to_be64(ofs);
+ return 0;
+}
+
+static int logfs_write_je(struct super_block *sb,
+ void* (*write)(struct super_block *sb, void *scratch,
+ u16 *type, size_t *len))
+{
+ void *buf;
+ size_t len;
+ u16 type;
+
+ buf = write(sb, logfs_super(sb)->s_je, &type, &len);
+ return logfs_write_je_buf(sb, buf, type, len);
+}
+
+int write_alias_journal(struct super_block *sb, u64 ino, u64 bix,
+ level_t level, int child_no, __be64 val)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_obj_alias *oa = super->s_je;
+ int err = 0, fill = super->s_je_fill;
+
+ log_aliases("logfs_write_obj_aliases #%x(%llx, %llx, %x, %x) %llx\n",
+ fill, ino, bix, level, child_no, be64_to_cpu(val));
+ oa[fill].ino = cpu_to_be64(ino);
+ oa[fill].bix = cpu_to_be64(bix);
+ oa[fill].val = val;
+ oa[fill].level = (__force u8)level;
+ oa[fill].child_no = cpu_to_be16(child_no);
+ fill++;
+ if (fill >= sb->s_blocksize / sizeof(*oa)) {
+ err = logfs_write_je_buf(sb, oa, JE_OBJ_ALIAS, sb->s_blocksize);
+ fill = 0;
+ }
+
+ super->s_je_fill = fill;
+ return err;
+}
+
+static int logfs_write_obj_aliases(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int err;
+
+ log_journal("logfs_write_obj_aliases: %d aliases to write\n",
+ super->s_no_object_aliases);
+ super->s_je_fill = 0;
+ err = logfs_write_obj_aliases_pagecache(sb);
+ if (err)
+ return err;
+
+ if (super->s_je_fill)
+ err = logfs_write_je_buf(sb, super->s_je, JE_OBJ_ALIAS,
+ super->s_je_fill
+ * sizeof(struct logfs_obj_alias));
+ return err;
+}
+
+/*
+ * Write all journal entries. The goto logic ensures that all journal entries
+ * are written whenever a new segment is used. It is ugly and potentially a
+ * bit wasteful, but robustness is more important. With this we can *always*
+ * erase all journal segments except the one containing the most recent commit.
+ */
+void logfs_write_anchor(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_journal_area;
+ int i, err;
+
+ if (!(super->s_flags & LOGFS_SB_FLAG_DIRTY))
+ return;
+ super->s_flags &= ~LOGFS_SB_FLAG_DIRTY;
+
+ BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN);
+ mutex_lock(&super->s_journal_mutex);
+
+ /* Do this first or suffer corruption */
+ logfs_sync_segments(sb);
+ account_shadows(sb);
+
+again:
+ super->s_no_je = 0;
+ for_each_area(i) {
+ if (!super->s_area[i]->a_is_open)
+ continue;
+ super->s_sum_index = i;
+ err = logfs_write_je(sb, logfs_write_area);
+ if (err)
+ goto again;
+ }
+ err = logfs_write_obj_aliases(sb);
+ if (err)
+ goto again;
+ err = logfs_write_je(sb, logfs_write_erasecount);
+ if (err)
+ goto again;
+ err = logfs_write_je(sb, __logfs_write_anchor);
+ if (err)
+ goto again;
+ err = logfs_write_je(sb, logfs_write_dynsb);
+ if (err)
+ goto again;
+ /*
+ * Order is imperative. First we sync all writes, including the
+ * non-committed journal writes. Then we write the final commit and
+ * sync the current journal segment.
+ * There is a theoretical bug here. Syncing the journal segment will
+ * write a number of journal entries and the final commit. All these
+ * are written in a single operation. If the device layer writes the
+ * data back-to-front, the commit will precede the other journal
+ * entries, leaving a race window.
+ * Two fixes are possible. Preferred is to fix the device layer to
+ * ensure writes happen front-to-back. Alternatively we can insert
+ * another logfs_sync_area() super->s_devops->sync() combo before
+ * writing the commit.
+ */
+ /*
+ * On another subject, super->s_devops->sync is usually not necessary.
+ * Unless called from sys_sync or friends, a barrier would suffice.
+ */
+ super->s_devops->sync(sb);
+ err = logfs_write_je(sb, logfs_write_commit);
+ if (err)
+ goto again;
+ log_journal("Write commit to %llx\n",
+ be64_to_cpu(super->s_je_array[super->s_no_je - 1]));
+ logfs_sync_area(area);
+ BUG_ON(area->a_used_bytes != area->a_written_bytes);
+ super->s_devops->sync(sb);
+
+ mutex_unlock(&super->s_journal_mutex);
+ return;
+}
+
+void do_logfs_journal_wl_pass(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_area *area = super->s_journal_area;
+ u32 segno, ec;
+ int i, err;
+
+ log_journal("Journal requires wear-leveling.\n");
+ /* Drop old segments */
+ journal_for_each(i)
+ if (super->s_journal_seg[i]) {
+ logfs_set_segment_unreserved(sb,
+ super->s_journal_seg[i],
+ super->s_journal_ec[i]);
+ super->s_journal_seg[i] = 0;
+ super->s_journal_ec[i] = 0;
+ }
+ /* Get new segments */
+ for (i = 0; i < super->s_no_journal_segs; i++) {
+ segno = get_best_cand(sb, &super->s_reserve_list, &ec);
+ super->s_journal_seg[i] = segno;
+ super->s_journal_ec[i] = ec;
+ logfs_set_segment_reserved(sb, segno);
+ }
+ /* Manually move journal_area */
+ area->a_segno = super->s_journal_seg[0];
+ area->a_is_open = 0;
+ area->a_used_bytes = 0;
+ /* Write journal */
+ logfs_write_anchor(sb);
+ /* Write superblocks */
+ err = logfs_write_sb(sb);
+ BUG_ON(err);
+}
+
+static const struct logfs_area_ops journal_area_ops = {
+ .get_free_segment = journal_get_free_segment,
+ .get_erase_count = journal_get_erase_count,
+ .erase_segment = journal_erase_segment,
+};
+
+int logfs_init_journal(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ size_t bufsize = max_t(size_t, sb->s_blocksize, super->s_writesize)
+ + MAX_JOURNAL_HEADER;
+ int ret = -ENOMEM;
+
+ mutex_init(&super->s_journal_mutex);
+ btree_init_mempool32(&super->s_reserved_segments, super->s_btree_pool);
+
+ super->s_je = kzalloc(bufsize, GFP_KERNEL);
+ if (!super->s_je)
+ return ret;
+
+ super->s_compressed_je = kzalloc(bufsize, GFP_KERNEL);
+ if (!super->s_compressed_je)
+ return ret;
+
+ super->s_master_inode = logfs_new_meta_inode(sb, LOGFS_INO_MASTER);
+ if (IS_ERR(super->s_master_inode))
+ return PTR_ERR(super->s_master_inode);
+
+ ret = logfs_read_journal(sb);
+ if (ret)
+ return -EIO;
+
+ reserve_sb_and_journal(sb);
+ logfs_calc_free(sb);
+
+ super->s_journal_area->a_ops = &journal_area_ops;
+ return 0;
+}
+
+void logfs_cleanup_journal(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ btree_grim_visitor32(&super->s_reserved_segments, 0, NULL);
+ destroy_meta_inode(super->s_master_inode);
+ super->s_master_inode = NULL;
+
+ kfree(super->s_compressed_je);
+ kfree(super->s_je);
+}
diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h
new file mode 100644
index 000000000000..129779431373
--- /dev/null
+++ b/fs/logfs/logfs.h
@@ -0,0 +1,724 @@
+/*
+ * fs/logfs/logfs.h
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ *
+ * Private header for logfs.
+ */
+#ifndef FS_LOGFS_LOGFS_H
+#define FS_LOGFS_LOGFS_H
+
+#undef __CHECK_ENDIAN__
+#define __CHECK_ENDIAN__
+
+#include <linux/btree.h>
+#include <linux/crc32.h>
+#include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/mempool.h>
+#include <linux/pagemap.h>
+#include <linux/mtd/mtd.h>
+#include "logfs_abi.h"
+
+#define LOGFS_DEBUG_SUPER (0x0001)
+#define LOGFS_DEBUG_SEGMENT (0x0002)
+#define LOGFS_DEBUG_JOURNAL (0x0004)
+#define LOGFS_DEBUG_DIR (0x0008)
+#define LOGFS_DEBUG_FILE (0x0010)
+#define LOGFS_DEBUG_INODE (0x0020)
+#define LOGFS_DEBUG_READWRITE (0x0040)
+#define LOGFS_DEBUG_GC (0x0080)
+#define LOGFS_DEBUG_GC_NOISY (0x0100)
+#define LOGFS_DEBUG_ALIASES (0x0200)
+#define LOGFS_DEBUG_BLOCKMOVE (0x0400)
+#define LOGFS_DEBUG_ALL (0xffffffff)
+
+#define LOGFS_DEBUG (0x01)
+/*
+ * To enable specific log messages, simply define LOGFS_DEBUG to match any
+ * or all of the above.
+ */
+#ifndef LOGFS_DEBUG
+#define LOGFS_DEBUG (0)
+#endif
+
+#define log_cond(cond, fmt, arg...) do { \
+ if (cond) \
+ printk(KERN_DEBUG fmt, ##arg); \
+} while (0)
+
+#define log_super(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SUPER, fmt, ##arg)
+#define log_segment(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_SEGMENT, fmt, ##arg)
+#define log_journal(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_JOURNAL, fmt, ##arg)
+#define log_dir(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_DIR, fmt, ##arg)
+#define log_file(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_FILE, fmt, ##arg)
+#define log_inode(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_INODE, fmt, ##arg)
+#define log_readwrite(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_READWRITE, fmt, ##arg)
+#define log_gc(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC, fmt, ##arg)
+#define log_gc_noisy(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_GC_NOISY, fmt, ##arg)
+#define log_aliases(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_ALIASES, fmt, ##arg)
+#define log_blockmove(fmt, arg...) \
+ log_cond(LOGFS_DEBUG & LOGFS_DEBUG_BLOCKMOVE, fmt, ##arg)
+
+#define PG_pre_locked PG_owner_priv_1
+#define PagePreLocked(page) test_bit(PG_pre_locked, &(page)->flags)
+#define SetPagePreLocked(page) set_bit(PG_pre_locked, &(page)->flags)
+#define ClearPagePreLocked(page) clear_bit(PG_pre_locked, &(page)->flags)
+
+/* FIXME: This should really be somewhere in the 64bit area. */
+#define LOGFS_LINK_MAX (1<<30)
+
+/* Read-only filesystem */
+#define LOGFS_SB_FLAG_RO 0x0001
+#define LOGFS_SB_FLAG_DIRTY 0x0002
+#define LOGFS_SB_FLAG_OBJ_ALIAS 0x0004
+#define LOGFS_SB_FLAG_SHUTDOWN 0x0008
+
+/* Write Control Flags */
+#define WF_LOCK 0x01 /* take write lock */
+#define WF_WRITE 0x02 /* write block */
+#define WF_DELETE 0x04 /* delete old block */
+
+typedef u8 __bitwise level_t;
+typedef u8 __bitwise gc_level_t;
+
+#define LEVEL(level) ((__force level_t)(level))
+#define GC_LEVEL(gc_level) ((__force gc_level_t)(gc_level))
+
+#define SUBLEVEL(level) ( (void)((level) == LEVEL(1)), \
+ (__force level_t)((__force u8)(level) - 1) )
+
+/**
+ * struct logfs_area - area management information
+ *
+ * @a_sb: the superblock this area belongs to
+ * @a_is_open: 1 if the area is currently open, else 0
+ * @a_segno: segment number of area
+ * @a_written_bytes: number of bytes already written back
+ * @a_used_bytes: number of used bytes
+ * @a_ops: area operations (either journal or ostore)
+ * @a_erase_count: erase count
+ * @a_level: GC level
+ */
+struct logfs_area { /* a segment open for writing */
+ struct super_block *a_sb;
+ int a_is_open;
+ u32 a_segno;
+ u32 a_written_bytes;
+ u32 a_used_bytes;
+ const struct logfs_area_ops *a_ops;
+ u32 a_erase_count;
+ gc_level_t a_level;
+};
+
+/**
+ * struct logfs_area_ops - area operations
+ *
+ * @get_free_segment: fill area->ofs with the offset of a free segment
+ * @get_erase_count: fill area->erase_count (needs area->ofs)
+ * @erase_segment: erase and setup segment
+ */
+struct logfs_area_ops {
+ void (*get_free_segment)(struct logfs_area *area);
+ void (*get_erase_count)(struct logfs_area *area);
+ int (*erase_segment)(struct logfs_area *area);
+};
+
+/**
+ * struct logfs_device_ops - device access operations
+ *
+ * @readpage: read one page (mm page)
+ * @writeseg: write one segment. may be a partial segment
+ * @erase: erase one segment
+ * @read: read from the device
+ * @erase: erase part of the device
+ */
+struct logfs_device_ops {
+ struct page *(*find_first_sb)(struct super_block *sb, u64 *ofs);
+ struct page *(*find_last_sb)(struct super_block *sb, u64 *ofs);
+ int (*write_sb)(struct super_block *sb, struct page *page);
+ int (*readpage)(void *_sb, struct page *page);
+ void (*writeseg)(struct super_block *sb, u64 ofs, size_t len);
+ int (*erase)(struct super_block *sb, loff_t ofs, size_t len,
+ int ensure_write);
+ void (*sync)(struct super_block *sb);
+ void (*put_device)(struct super_block *sb);
+};
+
+/**
+ * struct candidate_list - list of similar candidates
+ */
+struct candidate_list {
+ struct rb_root rb_tree;
+ int count;
+ int maxcount;
+ int sort_by_ec;
+};
+
+/**
+ * struct gc_candidate - "candidate" segment to be garbage collected next
+ *
+ * @list: list (either free of low)
+ * @segno: segment number
+ * @valid: number of valid bytes
+ * @erase_count: erase count of segment
+ * @dist: distance from tree root
+ *
+ * Candidates can be on two lists. The free list contains electees rather
+ * than candidates - segments that no longer contain any valid data. The
+ * low list contains candidates to be picked for GC. It should be kept
+ * short. It is not required to always pick a perfect candidate. In the
+ * worst case GC will have to move more data than absolutely necessary.
+ */
+struct gc_candidate {
+ struct rb_node rb_node;
+ struct candidate_list *list;
+ u32 segno;
+ u32 valid;
+ u32 erase_count;
+ u8 dist;
+};
+
+/**
+ * struct logfs_journal_entry - temporary structure used during journal scan
+ *
+ * @used:
+ * @version: normalized version
+ * @len: length
+ * @offset: offset
+ */
+struct logfs_journal_entry {
+ int used;
+ s16 version;
+ u16 len;
+ u16 datalen;
+ u64 offset;
+};
+
+enum transaction_state {
+ CREATE_1 = 1,
+ CREATE_2,
+ UNLINK_1,
+ UNLINK_2,
+ CROSS_RENAME_1,
+ CROSS_RENAME_2,
+ TARGET_RENAME_1,
+ TARGET_RENAME_2,
+ TARGET_RENAME_3
+};
+
+/**
+ * struct logfs_transaction - essential fields to support atomic dirops
+ *
+ * @ino: target inode
+ * @dir: inode of directory containing dentry
+ * @pos: pos of dentry in directory
+ */
+struct logfs_transaction {
+ enum transaction_state state;
+ u64 ino;
+ u64 dir;
+ u64 pos;
+};
+
+/**
+ * struct logfs_shadow - old block in the shadow of a not-yet-committed new one
+ * @old_ofs: offset of old block on medium
+ * @new_ofs: offset of new block on medium
+ * @ino: inode number
+ * @bix: block index
+ * @old_len: size of old block, including header
+ * @new_len: size of new block, including header
+ * @level: block level
+ */
+struct logfs_shadow {
+ u64 old_ofs;
+ u64 new_ofs;
+ u64 ino;
+ u64 bix;
+ int old_len;
+ int new_len;
+ gc_level_t gc_level;
+};
+
+/**
+ * struct shadow_tree
+ * @new: shadows where old_ofs==0, indexed by new_ofs
+ * @old: shadows where old_ofs!=0, indexed by old_ofs
+ */
+struct shadow_tree {
+ struct btree_head64 new;
+ struct btree_head64 old;
+};
+
+struct object_alias_item {
+ struct list_head list;
+ __be64 val;
+ int child_no;
+};
+
+/**
+ * struct logfs_block - contains any block state
+ * @type: indirect block or inode
+ * @full: number of fully populated children
+ * @partial: number of partially populated children
+ *
+ * Most blocks are directly represented by page cache pages. But when a block
+ * becomes dirty, is part of a transaction, contains aliases or is otherwise
+ * special, a struct logfs_block is allocated to track the additional state.
+ * Inodes are very similar to indirect blocks, so they can also get one of
+ * these structures added when appropriate.
+ */
+#define BLOCK_INDIRECT 1 /* Indirect block */
+#define BLOCK_INODE 2 /* Inode */
+struct logfs_block_ops;
+struct logfs_block {
+ struct list_head alias_list;
+ struct list_head item_list;
+ struct super_block *sb;
+ u64 ino;
+ u64 bix;
+ level_t level;
+ struct page *page;
+ struct inode *inode;
+ struct logfs_transaction *ta;
+ unsigned long alias_map[LOGFS_BLOCK_FACTOR / BITS_PER_LONG];
+ struct logfs_block_ops *ops;
+ int full;
+ int partial;
+ int reserved_bytes;
+};
+
+typedef int write_alias_t(struct super_block *sb, u64 ino, u64 bix,
+ level_t level, int child_no, __be64 val);
+struct logfs_block_ops {
+ void (*write_block)(struct logfs_block *block);
+ gc_level_t (*block_level)(struct logfs_block *block);
+ void (*free_block)(struct super_block *sb, struct logfs_block*block);
+ int (*write_alias)(struct super_block *sb,
+ struct logfs_block *block,
+ write_alias_t *write_one_alias);
+};
+
+struct logfs_super {
+ struct mtd_info *s_mtd; /* underlying device */
+ struct block_device *s_bdev; /* underlying device */
+ const struct logfs_device_ops *s_devops;/* device access */
+ struct inode *s_master_inode; /* inode file */
+ struct inode *s_segfile_inode; /* segment file */
+ struct inode *s_mapping_inode; /* device mapping */
+ atomic_t s_pending_writes; /* outstanting bios */
+ long s_flags;
+ mempool_t *s_btree_pool; /* for btree nodes */
+ mempool_t *s_alias_pool; /* aliases in segment.c */
+ u64 s_feature_incompat;
+ u64 s_feature_ro_compat;
+ u64 s_feature_compat;
+ u64 s_feature_flags;
+ u64 s_sb_ofs[2];
+ struct page *s_erase_page; /* for dev_bdev.c */
+ /* alias.c fields */
+ struct btree_head32 s_segment_alias; /* remapped segments */
+ int s_no_object_aliases;
+ struct list_head s_object_alias; /* remapped objects */
+ struct btree_head128 s_object_alias_tree; /* remapped objects */
+ struct mutex s_object_alias_mutex;
+ /* dir.c fields */
+ struct mutex s_dirop_mutex; /* for creat/unlink/rename */
+ u64 s_victim_ino; /* used for atomic dir-ops */
+ u64 s_rename_dir; /* source directory ino */
+ u64 s_rename_pos; /* position of source dd */
+ /* gc.c fields */
+ long s_segsize; /* size of a segment */
+ int s_segshift; /* log2 of segment size */
+ long s_segmask; /* 1 << s_segshift - 1 */
+ long s_no_segs; /* segments on device */
+ long s_no_journal_segs; /* segments used for journal */
+ long s_no_blocks; /* blocks per segment */
+ long s_writesize; /* minimum write size */
+ int s_writeshift; /* log2 of write size */
+ u64 s_size; /* filesystem size */
+ struct logfs_area *s_area[LOGFS_NO_AREAS]; /* open segment array */
+ u64 s_gec; /* global erase count */
+ u64 s_wl_gec_ostore; /* time of last wl event */
+ u64 s_wl_gec_journal; /* time of last wl event */
+ u64 s_sweeper; /* current sweeper pos */
+ u8 s_ifile_levels; /* max level of ifile */
+ u8 s_iblock_levels; /* max level of regular files */
+ u8 s_data_levels; /* # of segments to leaf block*/
+ u8 s_total_levels; /* sum of above three */
+ struct btree_head32 s_cand_tree; /* all candidates */
+ struct candidate_list s_free_list; /* 100% free segments */
+ struct candidate_list s_reserve_list; /* Bad segment reserve */
+ struct candidate_list s_low_list[LOGFS_NO_AREAS];/* good candidates */
+ struct candidate_list s_ec_list; /* wear level candidates */
+ struct btree_head32 s_reserved_segments;/* sb, journal, bad, etc. */
+ /* inode.c fields */
+ u64 s_last_ino; /* highest ino used */
+ long s_inos_till_wrap;
+ u32 s_generation; /* i_generation for new files */
+ struct list_head s_freeing_list; /* inodes being freed */
+ /* journal.c fields */
+ struct mutex s_journal_mutex;
+ void *s_je; /* journal entry to compress */
+ void *s_compressed_je; /* block to write to journal */
+ u32 s_journal_seg[LOGFS_JOURNAL_SEGS]; /* journal segments */
+ u32 s_journal_ec[LOGFS_JOURNAL_SEGS]; /* journal erasecounts */
+ u64 s_last_version;
+ struct logfs_area *s_journal_area; /* open journal segment */
+ __be64 s_je_array[64];
+ int s_no_je;
+
+ int s_sum_index; /* for the 12 summaries */
+ struct shadow_tree s_shadow_tree;
+ int s_je_fill; /* index of current je */
+ /* readwrite.c fields */
+ struct mutex s_write_mutex;
+ int s_lock_count;
+ mempool_t *s_block_pool; /* struct logfs_block pool */
+ mempool_t *s_shadow_pool; /* struct logfs_shadow pool */
+ /*
+ * Space accounting:
+ * - s_used_bytes specifies space used to store valid data objects.
+ * - s_dirty_used_bytes is space used to store non-committed data
+ * objects. Those objects have already been written themselves,
+ * but they don't become valid until all indirect blocks up to the
+ * journal have been written as well.
+ * - s_dirty_free_bytes is space used to store the old copy of a
+ * replaced object, as long as the replacement is non-committed.
+ * In other words, it is the amount of space freed when all dirty
+ * blocks are written back.
+ * - s_free_bytes is the amount of free space available for any
+ * purpose.
+ * - s_root_reserve is the amount of free space available only to
+ * the root user. Non-privileged users can no longer write once
+ * this watermark has been reached.
+ * - s_speed_reserve is space which remains unused to speed up
+ * garbage collection performance.
+ * - s_dirty_pages is the space reserved for currently dirty pages.
+ * It is a pessimistic estimate, so some/most will get freed on
+ * page writeback.
+ *
+ * s_used_bytes + s_free_bytes + s_speed_reserve = total usable size
+ */
+ u64 s_free_bytes;
+ u64 s_used_bytes;
+ u64 s_dirty_free_bytes;
+ u64 s_dirty_used_bytes;
+ u64 s_root_reserve;
+ u64 s_speed_reserve;
+ u64 s_dirty_pages;
+ /* Bad block handling:
+ * - s_bad_seg_reserve is a number of segments usually kept
+ * free. When encountering bad blocks, the affected segment's data
+ * is _temporarily_ moved to a reserved segment.
+ * - s_bad_segments is the number of known bad segments.
+ */
+ u32 s_bad_seg_reserve;
+ u32 s_bad_segments;
+};
+
+/**
+ * struct logfs_inode - in-memory inode
+ *
+ * @vfs_inode: struct inode
+ * @li_data: data pointers
+ * @li_used_bytes: number of used bytes
+ * @li_freeing_list: used to track inodes currently being freed
+ * @li_flags: inode flags
+ * @li_refcount: number of internal (GC-induced) references
+ */
+struct logfs_inode {
+ struct inode vfs_inode;
+ u64 li_data[LOGFS_EMBEDDED_FIELDS];
+ u64 li_used_bytes;
+ struct list_head li_freeing_list;
+ struct logfs_block *li_block;
+ u32 li_flags;
+ u8 li_height;
+ int li_refcount;
+};
+
+#define journal_for_each(__i) for (__i = 0; __i < LOGFS_JOURNAL_SEGS; __i++)
+#define for_each_area(__i) for (__i = 0; __i < LOGFS_NO_AREAS; __i++)
+#define for_each_area_down(__i) for (__i = LOGFS_NO_AREAS - 1; __i >= 0; __i--)
+
+/* compr.c */
+int logfs_compress(void *in, void *out, size_t inlen, size_t outlen);
+int logfs_uncompress(void *in, void *out, size_t inlen, size_t outlen);
+int __init logfs_compr_init(void);
+void logfs_compr_exit(void);
+
+/* dev_bdev.c */
+#ifdef CONFIG_BLOCK
+int logfs_get_sb_bdev(struct file_system_type *type, int flags,
+ const char *devname, struct vfsmount *mnt);
+#else
+static inline int logfs_get_sb_bdev(struct file_system_type *type, int flags,
+ const char *devname, struct vfsmount *mnt)
+{
+ return -ENODEV;
+}
+#endif
+
+/* dev_mtd.c */
+#ifdef CONFIG_MTD
+int logfs_get_sb_mtd(struct file_system_type *type, int flags,
+ int mtdnr, struct vfsmount *mnt);
+#else
+static inline int logfs_get_sb_mtd(struct file_system_type *type, int flags,
+ int mtdnr, struct vfsmount *mnt)
+{
+ return -ENODEV;
+}
+#endif
+
+/* dir.c */
+extern const struct inode_operations logfs_symlink_iops;
+extern const struct inode_operations logfs_dir_iops;
+extern const struct file_operations logfs_dir_fops;
+int logfs_replay_journal(struct super_block *sb);
+
+/* file.c */
+extern const struct inode_operations logfs_reg_iops;
+extern const struct file_operations logfs_reg_fops;
+extern const struct address_space_operations logfs_reg_aops;
+int logfs_readpage(struct file *file, struct page *page);
+int logfs_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
+ unsigned long arg);
+int logfs_fsync(struct file *file, struct dentry *dentry, int datasync);
+
+/* gc.c */
+u32 get_best_cand(struct super_block *sb, struct candidate_list *list, u32 *ec);
+void logfs_gc_pass(struct super_block *sb);
+int logfs_check_areas(struct super_block *sb);
+int logfs_init_gc(struct super_block *sb);
+void logfs_cleanup_gc(struct super_block *sb);
+
+/* inode.c */
+extern const struct super_operations logfs_super_operations;
+struct inode *logfs_iget(struct super_block *sb, ino_t ino);
+struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *cookie);
+void logfs_safe_iput(struct inode *inode, int cookie);
+struct inode *logfs_new_inode(struct inode *dir, int mode);
+struct inode *logfs_new_meta_inode(struct super_block *sb, u64 ino);
+struct inode *logfs_read_meta_inode(struct super_block *sb, u64 ino);
+int logfs_init_inode_cache(void);
+void logfs_destroy_inode_cache(void);
+void destroy_meta_inode(struct inode *inode);
+void logfs_set_blocks(struct inode *inode, u64 no);
+/* these logically belong into inode.c but actually reside in readwrite.c */
+int logfs_read_inode(struct inode *inode);
+int __logfs_write_inode(struct inode *inode, long flags);
+void logfs_delete_inode(struct inode *inode);
+void logfs_clear_inode(struct inode *inode);
+
+/* journal.c */
+void logfs_write_anchor(struct super_block *sb);
+int logfs_init_journal(struct super_block *sb);
+void logfs_cleanup_journal(struct super_block *sb);
+int write_alias_journal(struct super_block *sb, u64 ino, u64 bix,
+ level_t level, int child_no, __be64 val);
+void do_logfs_journal_wl_pass(struct super_block *sb);
+
+/* readwrite.c */
+pgoff_t logfs_pack_index(u64 bix, level_t level);
+void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level);
+int logfs_inode_write(struct inode *inode, const void *buf, size_t count,
+ loff_t bix, long flags, struct shadow_tree *shadow_tree);
+int logfs_readpage_nolock(struct page *page);
+int logfs_write_buf(struct inode *inode, struct page *page, long flags);
+int logfs_delete(struct inode *inode, pgoff_t index,
+ struct shadow_tree *shadow_tree);
+int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs,
+ gc_level_t gc_level, long flags);
+int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix,
+ gc_level_t gc_level);
+int logfs_truncate(struct inode *inode, u64 size);
+u64 logfs_seek_hole(struct inode *inode, u64 bix);
+u64 logfs_seek_data(struct inode *inode, u64 bix);
+int logfs_open_segfile(struct super_block *sb);
+int logfs_init_rw(struct super_block *sb);
+void logfs_cleanup_rw(struct super_block *sb);
+void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta);
+void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta);
+void logfs_write_block(struct logfs_block *block, long flags);
+int logfs_write_obj_aliases_pagecache(struct super_block *sb);
+void logfs_get_segment_entry(struct super_block *sb, u32 segno,
+ struct logfs_segment_entry *se);
+void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment);
+void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec,
+ gc_level_t gc_level);
+void logfs_set_segment_reserved(struct super_block *sb, u32 segno);
+void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec);
+struct logfs_block *__alloc_block(struct super_block *sb,
+ u64 ino, u64 bix, level_t level);
+void __free_block(struct super_block *sb, struct logfs_block *block);
+void btree_write_block(struct logfs_block *block);
+void initialize_block_counters(struct page *page, struct logfs_block *block,
+ __be64 *array, int page_is_empty);
+int logfs_exist_block(struct inode *inode, u64 bix);
+int get_page_reserve(struct inode *inode, struct page *page);
+extern struct logfs_block_ops indirect_block_ops;
+
+/* segment.c */
+int logfs_erase_segment(struct super_block *sb, u32 ofs, int ensure_erase);
+int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf);
+int logfs_segment_read(struct inode *inode, struct page *page, u64 ofs, u64 bix,
+ level_t level);
+int logfs_segment_write(struct inode *inode, struct page *page,
+ struct logfs_shadow *shadow);
+int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow);
+int logfs_load_object_aliases(struct super_block *sb,
+ struct logfs_obj_alias *oa, int count);
+void move_page_to_btree(struct page *page);
+int logfs_init_mapping(struct super_block *sb);
+void logfs_sync_area(struct logfs_area *area);
+void logfs_sync_segments(struct super_block *sb);
+
+/* area handling */
+int logfs_init_areas(struct super_block *sb);
+void logfs_cleanup_areas(struct super_block *sb);
+int logfs_open_area(struct logfs_area *area, size_t bytes);
+void __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len,
+ int use_filler);
+
+static inline void logfs_buf_write(struct logfs_area *area, u64 ofs,
+ void *buf, size_t len)
+{
+ __logfs_buf_write(area, ofs, buf, len, 0);
+}
+
+static inline void logfs_buf_recover(struct logfs_area *area, u64 ofs,
+ void *buf, size_t len)
+{
+ __logfs_buf_write(area, ofs, buf, len, 1);
+}
+
+/* super.c */
+struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index);
+void emergency_read_end(struct page *page);
+void logfs_crash_dump(struct super_block *sb);
+void *memchr_inv(const void *s, int c, size_t n);
+int logfs_statfs(struct dentry *dentry, struct kstatfs *stats);
+int logfs_get_sb_device(struct file_system_type *type, int flags,
+ struct mtd_info *mtd, struct block_device *bdev,
+ const struct logfs_device_ops *devops, struct vfsmount *mnt);
+int logfs_check_ds(struct logfs_disk_super *ds);
+int logfs_write_sb(struct super_block *sb);
+
+static inline struct logfs_super *logfs_super(struct super_block *sb)
+{
+ return sb->s_fs_info;
+}
+
+static inline struct logfs_inode *logfs_inode(struct inode *inode)
+{
+ return container_of(inode, struct logfs_inode, vfs_inode);
+}
+
+static inline void logfs_set_ro(struct super_block *sb)
+{
+ logfs_super(sb)->s_flags |= LOGFS_SB_FLAG_RO;
+}
+
+#define LOGFS_BUG(sb) do { \
+ struct super_block *__sb = sb; \
+ logfs_crash_dump(__sb); \
+ logfs_super(__sb)->s_flags |= LOGFS_SB_FLAG_RO; \
+ BUG(); \
+} while (0)
+
+#define LOGFS_BUG_ON(condition, sb) \
+ do { if (unlikely(condition)) LOGFS_BUG((sb)); } while (0)
+
+static inline __be32 logfs_crc32(void *data, size_t len, size_t skip)
+{
+ return cpu_to_be32(crc32(~0, data+skip, len-skip));
+}
+
+static inline u8 logfs_type(struct inode *inode)
+{
+ return (inode->i_mode >> 12) & 15;
+}
+
+static inline pgoff_t logfs_index(struct super_block *sb, u64 pos)
+{
+ return pos >> sb->s_blocksize_bits;
+}
+
+static inline u64 dev_ofs(struct super_block *sb, u32 segno, u32 ofs)
+{
+ return ((u64)segno << logfs_super(sb)->s_segshift) + ofs;
+}
+
+static inline u32 seg_no(struct super_block *sb, u64 ofs)
+{
+ return ofs >> logfs_super(sb)->s_segshift;
+}
+
+static inline u32 seg_ofs(struct super_block *sb, u64 ofs)
+{
+ return ofs & logfs_super(sb)->s_segmask;
+}
+
+static inline u64 seg_align(struct super_block *sb, u64 ofs)
+{
+ return ofs & ~logfs_super(sb)->s_segmask;
+}
+
+static inline struct logfs_block *logfs_block(struct page *page)
+{
+ return (void *)page->private;
+}
+
+static inline level_t shrink_level(gc_level_t __level)
+{
+ u8 level = (__force u8)__level;
+
+ if (level >= LOGFS_MAX_LEVELS)
+ level -= LOGFS_MAX_LEVELS;
+ return (__force level_t)level;
+}
+
+static inline gc_level_t expand_level(u64 ino, level_t __level)
+{
+ u8 level = (__force u8)__level;
+
+ if (ino == LOGFS_INO_MASTER) {
+ /* ifile has seperate areas */
+ level += LOGFS_MAX_LEVELS;
+ }
+ return (__force gc_level_t)level;
+}
+
+static inline int logfs_block_shift(struct super_block *sb, level_t level)
+{
+ level = shrink_level((__force gc_level_t)level);
+ return (__force int)level * (sb->s_blocksize_bits - 3);
+}
+
+static inline u64 logfs_block_mask(struct super_block *sb, level_t level)
+{
+ return ~0ull << logfs_block_shift(sb, level);
+}
+
+static inline struct logfs_area *get_area(struct super_block *sb,
+ gc_level_t gc_level)
+{
+ return logfs_super(sb)->s_area[(__force u8)gc_level];
+}
+
+#endif
diff --git a/fs/logfs/logfs_abi.h b/fs/logfs/logfs_abi.h
new file mode 100644
index 000000000000..f674725663fe
--- /dev/null
+++ b/fs/logfs/logfs_abi.h
@@ -0,0 +1,629 @@
+/*
+ * fs/logfs/logfs_abi.h
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ *
+ * Public header for logfs.
+ */
+#ifndef FS_LOGFS_LOGFS_ABI_H
+#define FS_LOGFS_LOGFS_ABI_H
+
+/* For out-of-kernel compiles */
+#ifndef BUILD_BUG_ON
+#define BUILD_BUG_ON(condition) /**/
+#endif
+
+#define SIZE_CHECK(type, size) \
+static inline void check_##type(void) \
+{ \
+ BUILD_BUG_ON(sizeof(struct type) != (size)); \
+}
+
+/*
+ * Throughout the logfs code, we're constantly dealing with blocks at
+ * various positions or offsets. To remove confusion, we stricly
+ * distinguish between a "position" - the logical position within a
+ * file and an "offset" - the physical location within the device.
+ *
+ * Any usage of the term offset for a logical location or position for
+ * a physical one is a bug and should get fixed.
+ */
+
+/*
+ * Block are allocated in one of several segments depending on their
+ * level. The following levels are used:
+ * 0 - regular data block
+ * 1 - i1 indirect blocks
+ * 2 - i2 indirect blocks
+ * 3 - i3 indirect blocks
+ * 4 - i4 indirect blocks
+ * 5 - i5 indirect blocks
+ * 6 - ifile data blocks
+ * 7 - ifile i1 indirect blocks
+ * 8 - ifile i2 indirect blocks
+ * 9 - ifile i3 indirect blocks
+ * 10 - ifile i4 indirect blocks
+ * 11 - ifile i5 indirect blocks
+ * Potential levels to be used in the future:
+ * 12 - gc recycled blocks, long-lived data
+ * 13 - replacement blocks, short-lived data
+ *
+ * Levels 1-11 are necessary for robust gc operations and help seperate
+ * short-lived metadata from longer-lived file data. In the future,
+ * file data should get seperated into several segments based on simple
+ * heuristics. Old data recycled during gc operation is expected to be
+ * long-lived. New data is of uncertain life expectancy. New data
+ * used to replace older blocks in existing files is expected to be
+ * short-lived.
+ */
+
+
+/* Magic numbers. 64bit for superblock, 32bit for statfs f_type */
+#define LOGFS_MAGIC 0x7a3a8e5cb9d5bf67ull
+#define LOGFS_MAGIC_U32 0xc97e8168u
+
+/*
+ * Various blocksize related macros. Blocksize is currently fixed at 4KiB.
+ * Sooner or later that should become configurable and the macros replaced
+ * by something superblock-dependent. Pointers in indirect blocks are and
+ * will remain 64bit.
+ *
+ * LOGFS_BLOCKSIZE - self-explaining
+ * LOGFS_BLOCK_FACTOR - number of pointers per indirect block
+ * LOGFS_BLOCK_BITS - log2 of LOGFS_BLOCK_FACTOR, used for shifts
+ */
+#define LOGFS_BLOCKSIZE (4096ull)
+#define LOGFS_BLOCK_FACTOR (LOGFS_BLOCKSIZE / sizeof(u64))
+#define LOGFS_BLOCK_BITS (9)
+
+/*
+ * Number of blocks at various levels of indirection. There are 16 direct
+ * block pointers plus a single indirect pointer.
+ */
+#define I0_BLOCKS (16)
+#define I1_BLOCKS LOGFS_BLOCK_FACTOR
+#define I2_BLOCKS (LOGFS_BLOCK_FACTOR * I1_BLOCKS)
+#define I3_BLOCKS (LOGFS_BLOCK_FACTOR * I2_BLOCKS)
+#define I4_BLOCKS (LOGFS_BLOCK_FACTOR * I3_BLOCKS)
+#define I5_BLOCKS (LOGFS_BLOCK_FACTOR * I4_BLOCKS)
+
+#define INDIRECT_INDEX I0_BLOCKS
+#define LOGFS_EMBEDDED_FIELDS (I0_BLOCKS + 1)
+
+/*
+ * Sizes at which files require another level of indirection. Files smaller
+ * than LOGFS_EMBEDDED_SIZE can be completely stored in the inode itself,
+ * similar like ext2 fast symlinks.
+ *
+ * Data at a position smaller than LOGFS_I0_SIZE is accessed through the
+ * direct pointers, else through the 1x indirect pointer and so forth.
+ */
+#define LOGFS_EMBEDDED_SIZE (LOGFS_EMBEDDED_FIELDS * sizeof(u64))
+#define LOGFS_I0_SIZE (I0_BLOCKS * LOGFS_BLOCKSIZE)
+#define LOGFS_I1_SIZE (I1_BLOCKS * LOGFS_BLOCKSIZE)
+#define LOGFS_I2_SIZE (I2_BLOCKS * LOGFS_BLOCKSIZE)
+#define LOGFS_I3_SIZE (I3_BLOCKS * LOGFS_BLOCKSIZE)
+#define LOGFS_I4_SIZE (I4_BLOCKS * LOGFS_BLOCKSIZE)
+#define LOGFS_I5_SIZE (I5_BLOCKS * LOGFS_BLOCKSIZE)
+
+/*
+ * Each indirect block pointer must have this flag set, if all block pointers
+ * behind it are set, i.e. there is no hole hidden in the shadow of this
+ * indirect block pointer.
+ */
+#define LOGFS_FULLY_POPULATED (1ULL << 63)
+#define pure_ofs(ofs) (ofs & ~LOGFS_FULLY_POPULATED)
+
+/*
+ * LogFS needs to seperate data into levels. Each level is defined as the
+ * maximal possible distance from the master inode (inode of the inode file).
+ * Data blocks reside on level 0, 1x indirect block on level 1, etc.
+ * Inodes reside on level 6, indirect blocks for the inode file on levels 7-11.
+ * This effort is necessary to guarantee garbage collection to always make
+ * progress.
+ *
+ * LOGFS_MAX_INDIRECT is the maximal indirection through indirect blocks,
+ * LOGFS_MAX_LEVELS is one more for the actual data level of a file. It is
+ * the maximal number of levels for one file.
+ * LOGFS_NO_AREAS is twice that, as the inode file and regular files are
+ * effectively stacked on top of each other.
+ */
+#define LOGFS_MAX_INDIRECT (5)
+#define LOGFS_MAX_LEVELS (LOGFS_MAX_INDIRECT + 1)
+#define LOGFS_NO_AREAS (2 * LOGFS_MAX_LEVELS)
+
+/* Maximum size of filenames */
+#define LOGFS_MAX_NAMELEN (255)
+
+/* Number of segments in the primary journal. */
+#define LOGFS_JOURNAL_SEGS (16)
+
+/* Maximum number of free/erased/etc. segments in journal entries */
+#define MAX_CACHED_SEGS (64)
+
+
+/*
+ * LOGFS_OBJECT_HEADERSIZE is the size of a single header in the object store,
+ * LOGFS_MAX_OBJECTSIZE the size of the largest possible object, including
+ * its header,
+ * LOGFS_SEGMENT_RESERVE is the amount of space reserved for each segment for
+ * its segment header and the padded space at the end when no further objects
+ * fit.
+ */
+#define LOGFS_OBJECT_HEADERSIZE (0x1c)
+#define LOGFS_SEGMENT_HEADERSIZE (0x18)
+#define LOGFS_MAX_OBJECTSIZE (LOGFS_OBJECT_HEADERSIZE + LOGFS_BLOCKSIZE)
+#define LOGFS_SEGMENT_RESERVE \
+ (LOGFS_SEGMENT_HEADERSIZE + LOGFS_MAX_OBJECTSIZE - 1)
+
+/*
+ * Segment types:
+ * SEG_SUPER - Data or indirect block
+ * SEG_JOURNAL - Inode
+ * SEG_OSTORE - Dentry
+ */
+enum {
+ SEG_SUPER = 0x01,
+ SEG_JOURNAL = 0x02,
+ SEG_OSTORE = 0x03,
+};
+
+/**
+ * struct logfs_segment_header - per-segment header in the ostore
+ *
+ * @crc: crc32 of header (there is no data)
+ * @pad: unused, must be 0
+ * @type: segment type, see above
+ * @level: GC level for all objects in this segment
+ * @segno: segment number
+ * @ec: erase count for this segment
+ * @gec: global erase count at time of writing
+ */
+struct logfs_segment_header {
+ __be32 crc;
+ __be16 pad;
+ __u8 type;
+ __u8 level;
+ __be32 segno;
+ __be32 ec;
+ __be64 gec;
+};
+
+SIZE_CHECK(logfs_segment_header, LOGFS_SEGMENT_HEADERSIZE);
+
+#define LOGFS_FEATURES_INCOMPAT (0ull)
+#define LOGFS_FEATURES_RO_COMPAT (0ull)
+#define LOGFS_FEATURES_COMPAT (0ull)
+
+/**
+ * struct logfs_disk_super - on-medium superblock
+ *
+ * @ds_magic: magic number, must equal LOGFS_MAGIC
+ * @ds_crc: crc32 of structure starting with the next field
+ * @ds_ifile_levels: maximum number of levels for ifile
+ * @ds_iblock_levels: maximum number of levels for regular files
+ * @ds_data_levels: number of seperate levels for data
+ * @pad0: reserved, must be 0
+ * @ds_feature_incompat: incompatible filesystem features
+ * @ds_feature_ro_compat: read-only compatible filesystem features
+ * @ds_feature_compat: compatible filesystem features
+ * @ds_flags: flags
+ * @ds_segment_shift: log2 of segment size
+ * @ds_block_shift: log2 of block size
+ * @ds_write_shift: log2 of write size
+ * @pad1: reserved, must be 0
+ * @ds_journal_seg: segments used by primary journal
+ * @ds_root_reserve: bytes reserved for the superuser
+ * @ds_speed_reserve: bytes reserved to speed up GC
+ * @ds_bad_seg_reserve: number of segments reserved to handle bad blocks
+ * @pad2: reserved, must be 0
+ * @pad3: reserved, must be 0
+ *
+ * Contains only read-only fields. Read-write fields like the amount of used
+ * space is tracked in the dynamic superblock, which is stored in the journal.
+ */
+struct logfs_disk_super {
+ struct logfs_segment_header ds_sh;
+ __be64 ds_magic;
+
+ __be32 ds_crc;
+ __u8 ds_ifile_levels;
+ __u8 ds_iblock_levels;
+ __u8 ds_data_levels;
+ __u8 ds_segment_shift;
+ __u8 ds_block_shift;
+ __u8 ds_write_shift;
+ __u8 pad0[6];
+
+ __be64 ds_filesystem_size;
+ __be32 ds_segment_size;
+ __be32 ds_bad_seg_reserve;
+
+ __be64 ds_feature_incompat;
+ __be64 ds_feature_ro_compat;
+
+ __be64 ds_feature_compat;
+ __be64 ds_feature_flags;
+
+ __be64 ds_root_reserve;
+ __be64 ds_speed_reserve;
+
+ __be32 ds_journal_seg[LOGFS_JOURNAL_SEGS];
+
+ __be64 ds_super_ofs[2];
+ __be64 pad3[8];
+};
+
+SIZE_CHECK(logfs_disk_super, 256);
+
+/*
+ * Object types:
+ * OBJ_BLOCK - Data or indirect block
+ * OBJ_INODE - Inode
+ * OBJ_DENTRY - Dentry
+ */
+enum {
+ OBJ_BLOCK = 0x04,
+ OBJ_INODE = 0x05,
+ OBJ_DENTRY = 0x06,
+};
+
+/**
+ * struct logfs_object_header - per-object header in the ostore
+ *
+ * @crc: crc32 of header, excluding data_crc
+ * @len: length of data
+ * @type: object type, see above
+ * @compr: compression type
+ * @ino: inode number
+ * @bix: block index
+ * @data_crc: crc32 of payload
+ */
+struct logfs_object_header {
+ __be32 crc;
+ __be16 len;
+ __u8 type;
+ __u8 compr;
+ __be64 ino;
+ __be64 bix;
+ __be32 data_crc;
+} __attribute__((packed));
+
+SIZE_CHECK(logfs_object_header, LOGFS_OBJECT_HEADERSIZE);
+
+/*
+ * Reserved inode numbers:
+ * LOGFS_INO_MASTER - master inode (for inode file)
+ * LOGFS_INO_ROOT - root directory
+ * LOGFS_INO_SEGFILE - per-segment used bytes and erase count
+ */
+enum {
+ LOGFS_INO_MAPPING = 0x00,
+ LOGFS_INO_MASTER = 0x01,
+ LOGFS_INO_ROOT = 0x02,
+ LOGFS_INO_SEGFILE = 0x03,
+ LOGFS_RESERVED_INOS = 0x10,
+};
+
+/*
+ * Inode flags. High bits should never be written to the medium. They are
+ * reserved for in-memory usage.
+ * Low bits should either remain in sync with the corresponding FS_*_FL or
+ * reuse slots that obviously don't make sense for logfs.
+ *
+ * LOGFS_IF_DIRTY Inode must be written back
+ * LOGFS_IF_ZOMBIE Inode has been deleted
+ * LOGFS_IF_STILLBORN -ENOSPC happened when creating inode
+ */
+#define LOGFS_IF_COMPRESSED 0x00000004 /* == FS_COMPR_FL */
+#define LOGFS_IF_DIRTY 0x20000000
+#define LOGFS_IF_ZOMBIE 0x40000000
+#define LOGFS_IF_STILLBORN 0x80000000
+
+/* Flags available to chattr */
+#define LOGFS_FL_USER_VISIBLE (LOGFS_IF_COMPRESSED)
+#define LOGFS_FL_USER_MODIFIABLE (LOGFS_IF_COMPRESSED)
+/* Flags inherited from parent directory on file/directory creation */
+#define LOGFS_FL_INHERITED (LOGFS_IF_COMPRESSED)
+
+/**
+ * struct logfs_disk_inode - on-medium inode
+ *
+ * @di_mode: file mode
+ * @di_pad: reserved, must be 0
+ * @di_flags: inode flags, see above
+ * @di_uid: user id
+ * @di_gid: group id
+ * @di_ctime: change time
+ * @di_mtime: modify time
+ * @di_refcount: reference count (aka nlink or link count)
+ * @di_generation: inode generation, for nfs
+ * @di_used_bytes: number of bytes used
+ * @di_size: file size
+ * @di_data: data pointers
+ */
+struct logfs_disk_inode {
+ __be16 di_mode;
+ __u8 di_height;
+ __u8 di_pad;
+ __be32 di_flags;
+ __be32 di_uid;
+ __be32 di_gid;
+
+ __be64 di_ctime;
+ __be64 di_mtime;
+
+ __be64 di_atime;
+ __be32 di_refcount;
+ __be32 di_generation;
+
+ __be64 di_used_bytes;
+ __be64 di_size;
+
+ __be64 di_data[LOGFS_EMBEDDED_FIELDS];
+};
+
+SIZE_CHECK(logfs_disk_inode, 200);
+
+#define INODE_POINTER_OFS \
+ (offsetof(struct logfs_disk_inode, di_data) / sizeof(__be64))
+#define INODE_USED_OFS \
+ (offsetof(struct logfs_disk_inode, di_used_bytes) / sizeof(__be64))
+#define INODE_SIZE_OFS \
+ (offsetof(struct logfs_disk_inode, di_size) / sizeof(__be64))
+#define INODE_HEIGHT_OFS (0)
+
+/**
+ * struct logfs_disk_dentry - on-medium dentry structure
+ *
+ * @ino: inode number
+ * @namelen: length of file name
+ * @type: file type, identical to bits 12..15 of mode
+ * @name: file name
+ */
+/* FIXME: add 6 bytes of padding to remove the __packed */
+struct logfs_disk_dentry {
+ __be64 ino;
+ __be16 namelen;
+ __u8 type;
+ __u8 name[LOGFS_MAX_NAMELEN];
+} __attribute__((packed));
+
+SIZE_CHECK(logfs_disk_dentry, 266);
+
+#define RESERVED 0xffffffff
+#define BADSEG 0xffffffff
+/**
+ * struct logfs_segment_entry - segment file entry
+ *
+ * @ec_level: erase count and level
+ * @valid: number of valid bytes
+ *
+ * Segment file contains one entry for every segment. ec_level contains the
+ * erasecount in the upper 28 bits and the level in the lower 4 bits. An
+ * ec_level of BADSEG (-1) identifies bad segments. valid contains the number
+ * of valid bytes or RESERVED (-1 again) if the segment is used for either the
+ * superblock or the journal, or when the segment is bad.
+ */
+struct logfs_segment_entry {
+ __be32 ec_level;
+ __be32 valid;
+};
+
+SIZE_CHECK(logfs_segment_entry, 8);
+
+/**
+ * struct logfs_journal_header - header for journal entries (JEs)
+ *
+ * @h_crc: crc32 of journal entry
+ * @h_len: length of compressed journal entry,
+ * not including header
+ * @h_datalen: length of uncompressed data
+ * @h_type: JE type
+ * @h_compr: compression type
+ * @h_pad: reserved
+ */
+struct logfs_journal_header {
+ __be32 h_crc;
+ __be16 h_len;
+ __be16 h_datalen;
+ __be16 h_type;
+ __u8 h_compr;
+ __u8 h_pad[5];
+};
+
+SIZE_CHECK(logfs_journal_header, 16);
+
+/*
+ * Life expectency of data.
+ * VIM_DEFAULT - default vim
+ * VIM_SEGFILE - for segment file only - very short-living
+ * VIM_GC - GC'd data - likely long-living
+ */
+enum logfs_vim {
+ VIM_DEFAULT = 0,
+ VIM_SEGFILE = 1,
+};
+
+/**
+ * struct logfs_je_area - wbuf header
+ *
+ * @segno: segment number of area
+ * @used_bytes: number of bytes already used
+ * @gc_level: GC level
+ * @vim: life expectancy of data
+ *
+ * "Areas" are segments currently being used for writing. There is at least
+ * one area per GC level. Several may be used to seperate long-living from
+ * short-living data. If an area with unknown vim is encountered, it can
+ * simply be closed.
+ * The write buffer immediately follow this header.
+ */
+struct logfs_je_area {
+ __be32 segno;
+ __be32 used_bytes;
+ __u8 gc_level;
+ __u8 vim;
+} __attribute__((packed));
+
+SIZE_CHECK(logfs_je_area, 10);
+
+#define MAX_JOURNAL_HEADER \
+ (sizeof(struct logfs_journal_header) + sizeof(struct logfs_je_area))
+
+/**
+ * struct logfs_je_dynsb - dynamic superblock
+ *
+ * @ds_gec: global erase count
+ * @ds_sweeper: current position of GC "sweeper"
+ * @ds_rename_dir: source directory ino (see dir.c documentation)
+ * @ds_rename_pos: position of source dd (see dir.c documentation)
+ * @ds_victim_ino: victims of incomplete dir operation (see dir.c)
+ * @ds_victim_ino: parent inode of victim (see dir.c)
+ * @ds_used_bytes: number of used bytes
+ */
+struct logfs_je_dynsb {
+ __be64 ds_gec;
+ __be64 ds_sweeper;
+
+ __be64 ds_rename_dir;
+ __be64 ds_rename_pos;
+
+ __be64 ds_victim_ino;
+ __be64 ds_victim_parent; /* XXX */
+
+ __be64 ds_used_bytes;
+ __be32 ds_generation;
+ __be32 pad;
+};
+
+SIZE_CHECK(logfs_je_dynsb, 64);
+
+/**
+ * struct logfs_je_anchor - anchor of filesystem tree, aka master inode
+ *
+ * @da_size: size of inode file
+ * @da_last_ino: last created inode
+ * @da_used_bytes: number of bytes used
+ * @da_data: data pointers
+ */
+struct logfs_je_anchor {
+ __be64 da_size;
+ __be64 da_last_ino;
+
+ __be64 da_used_bytes;
+ u8 da_height;
+ u8 pad[7];
+
+ __be64 da_data[LOGFS_EMBEDDED_FIELDS];
+};
+
+SIZE_CHECK(logfs_je_anchor, 168);
+
+/**
+ * struct logfs_je_spillout - spillout entry (from 1st to 2nd journal)
+ *
+ * @so_segment: segments used for 2nd journal
+ *
+ * Length of the array is given by h_len field in the header.
+ */
+struct logfs_je_spillout {
+ __be64 so_segment[0];
+};
+
+SIZE_CHECK(logfs_je_spillout, 0);
+
+/**
+ * struct logfs_je_journal_ec - erase counts for all journal segments
+ *
+ * @ec: erase count
+ *
+ * Length of the array is given by h_len field in the header.
+ */
+struct logfs_je_journal_ec {
+ __be32 ec[0];
+};
+
+SIZE_CHECK(logfs_je_journal_ec, 0);
+
+/**
+ * struct logfs_je_free_segments - list of free segmetns with erase count
+ */
+struct logfs_je_free_segments {
+ __be32 segno;
+ __be32 ec;
+};
+
+SIZE_CHECK(logfs_je_free_segments, 8);
+
+/**
+ * struct logfs_seg_alias - list of segment aliases
+ */
+struct logfs_seg_alias {
+ __be32 old_segno;
+ __be32 new_segno;
+};
+
+SIZE_CHECK(logfs_seg_alias, 8);
+
+/**
+ * struct logfs_obj_alias - list of object aliases
+ */
+struct logfs_obj_alias {
+ __be64 ino;
+ __be64 bix;
+ __be64 val;
+ u8 level;
+ u8 pad[5];
+ __be16 child_no;
+};
+
+SIZE_CHECK(logfs_obj_alias, 32);
+
+/**
+ * Compression types.
+ *
+ * COMPR_NONE - uncompressed
+ * COMPR_ZLIB - compressed with zlib
+ */
+enum {
+ COMPR_NONE = 0,
+ COMPR_ZLIB = 1,
+};
+
+/*
+ * Journal entries come in groups of 16. First group contains unique
+ * entries, next groups contain one entry per level
+ *
+ * JE_FIRST - smallest possible journal entry number
+ *
+ * JEG_BASE - base group, containing unique entries
+ * JE_COMMIT - commit entry, validates all previous entries
+ * JE_DYNSB - dynamic superblock, anything that ought to be in the
+ * superblock but cannot because it is read-write data
+ * JE_ANCHOR - anchor aka master inode aka inode file's inode
+ * JE_ERASECOUNT erasecounts for all journal segments
+ * JE_SPILLOUT - unused
+ * JE_SEG_ALIAS - aliases segments
+ * JE_AREA - area description
+ *
+ * JE_LAST - largest possible journal entry number
+ */
+enum {
+ JE_FIRST = 0x01,
+
+ JEG_BASE = 0x00,
+ JE_COMMIT = 0x02,
+ JE_DYNSB = 0x03,
+ JE_ANCHOR = 0x04,
+ JE_ERASECOUNT = 0x05,
+ JE_SPILLOUT = 0x06,
+ JE_OBJ_ALIAS = 0x0d,
+ JE_AREA = 0x0e,
+
+ JE_LAST = 0x0e,
+};
+
+#endif
diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c
new file mode 100644
index 000000000000..7a23b3e7c0a7
--- /dev/null
+++ b/fs/logfs/readwrite.c
@@ -0,0 +1,2246 @@
+/*
+ * fs/logfs/readwrite.c
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ *
+ *
+ * Actually contains five sets of very similar functions:
+ * read read blocks from a file
+ * seek_hole find next hole
+ * seek_data find next data block
+ * valid check whether a block still belongs to a file
+ * write write blocks to a file
+ * delete delete a block (for directories and ifile)
+ * rewrite move existing blocks of a file to a new location (gc helper)
+ * truncate truncate a file
+ */
+#include "logfs.h"
+#include <linux/sched.h>
+
+static u64 adjust_bix(u64 bix, level_t level)
+{
+ switch (level) {
+ case 0:
+ return bix;
+ case LEVEL(1):
+ return max_t(u64, bix, I0_BLOCKS);
+ case LEVEL(2):
+ return max_t(u64, bix, I1_BLOCKS);
+ case LEVEL(3):
+ return max_t(u64, bix, I2_BLOCKS);
+ case LEVEL(4):
+ return max_t(u64, bix, I3_BLOCKS);
+ case LEVEL(5):
+ return max_t(u64, bix, I4_BLOCKS);
+ default:
+ WARN_ON(1);
+ return bix;
+ }
+}
+
+static inline u64 maxbix(u8 height)
+{
+ return 1ULL << (LOGFS_BLOCK_BITS * height);
+}
+
+/**
+ * The inode address space is cut in two halves. Lower half belongs to data
+ * pages, upper half to indirect blocks. If the high bit (INDIRECT_BIT) is
+ * set, the actual block index (bix) and level can be derived from the page
+ * index.
+ *
+ * The lowest three bits of the block index are set to 0 after packing and
+ * unpacking. Since the lowest n bits (9 for 4KiB blocksize) are ignored
+ * anyway this is harmless.
+ */
+#define ARCH_SHIFT (BITS_PER_LONG - 32)
+#define INDIRECT_BIT (0x80000000UL << ARCH_SHIFT)
+#define LEVEL_SHIFT (28 + ARCH_SHIFT)
+static inline pgoff_t first_indirect_block(void)
+{
+ return INDIRECT_BIT | (1ULL << LEVEL_SHIFT);
+}
+
+pgoff_t logfs_pack_index(u64 bix, level_t level)
+{
+ pgoff_t index;
+
+ BUG_ON(bix >= INDIRECT_BIT);
+ if (level == 0)
+ return bix;
+
+ index = INDIRECT_BIT;
+ index |= (__force long)level << LEVEL_SHIFT;
+ index |= bix >> ((__force u8)level * LOGFS_BLOCK_BITS);
+ return index;
+}
+
+void logfs_unpack_index(pgoff_t index, u64 *bix, level_t *level)
+{
+ u8 __level;
+
+ if (!(index & INDIRECT_BIT)) {
+ *bix = index;
+ *level = 0;
+ return;
+ }
+
+ __level = (index & ~INDIRECT_BIT) >> LEVEL_SHIFT;
+ *level = LEVEL(__level);
+ *bix = (index << (__level * LOGFS_BLOCK_BITS)) & ~INDIRECT_BIT;
+ *bix = adjust_bix(*bix, *level);
+ return;
+}
+#undef ARCH_SHIFT
+#undef INDIRECT_BIT
+#undef LEVEL_SHIFT
+
+/*
+ * Time is stored as nanoseconds since the epoch.
+ */
+static struct timespec be64_to_timespec(__be64 betime)
+{
+ return ns_to_timespec(be64_to_cpu(betime));
+}
+
+static __be64 timespec_to_be64(struct timespec tsp)
+{
+ return cpu_to_be64((u64)tsp.tv_sec * NSEC_PER_SEC + tsp.tv_nsec);
+}
+
+static void logfs_disk_to_inode(struct logfs_disk_inode *di, struct inode*inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ int i;
+
+ inode->i_mode = be16_to_cpu(di->di_mode);
+ li->li_height = di->di_height;
+ li->li_flags = be32_to_cpu(di->di_flags);
+ inode->i_uid = be32_to_cpu(di->di_uid);
+ inode->i_gid = be32_to_cpu(di->di_gid);
+ inode->i_size = be64_to_cpu(di->di_size);
+ logfs_set_blocks(inode, be64_to_cpu(di->di_used_bytes));
+ inode->i_atime = be64_to_timespec(di->di_atime);
+ inode->i_ctime = be64_to_timespec(di->di_ctime);
+ inode->i_mtime = be64_to_timespec(di->di_mtime);
+ inode->i_nlink = be32_to_cpu(di->di_refcount);
+ inode->i_generation = be32_to_cpu(di->di_generation);
+
+ switch (inode->i_mode & S_IFMT) {
+ case S_IFSOCK: /* fall through */
+ case S_IFBLK: /* fall through */
+ case S_IFCHR: /* fall through */
+ case S_IFIFO:
+ inode->i_rdev = be64_to_cpu(di->di_data[0]);
+ break;
+ case S_IFDIR: /* fall through */
+ case S_IFREG: /* fall through */
+ case S_IFLNK:
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ li->li_data[i] = be64_to_cpu(di->di_data[i]);
+ break;
+ default:
+ BUG();
+ }
+}
+
+static void logfs_inode_to_disk(struct inode *inode, struct logfs_disk_inode*di)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ int i;
+
+ di->di_mode = cpu_to_be16(inode->i_mode);
+ di->di_height = li->li_height;
+ di->di_pad = 0;
+ di->di_flags = cpu_to_be32(li->li_flags);
+ di->di_uid = cpu_to_be32(inode->i_uid);
+ di->di_gid = cpu_to_be32(inode->i_gid);
+ di->di_size = cpu_to_be64(i_size_read(inode));
+ di->di_used_bytes = cpu_to_be64(li->li_used_bytes);
+ di->di_atime = timespec_to_be64(inode->i_atime);
+ di->di_ctime = timespec_to_be64(inode->i_ctime);
+ di->di_mtime = timespec_to_be64(inode->i_mtime);
+ di->di_refcount = cpu_to_be32(inode->i_nlink);
+ di->di_generation = cpu_to_be32(inode->i_generation);
+
+ switch (inode->i_mode & S_IFMT) {
+ case S_IFSOCK: /* fall through */
+ case S_IFBLK: /* fall through */
+ case S_IFCHR: /* fall through */
+ case S_IFIFO:
+ di->di_data[0] = cpu_to_be64(inode->i_rdev);
+ break;
+ case S_IFDIR: /* fall through */
+ case S_IFREG: /* fall through */
+ case S_IFLNK:
+ for (i = 0; i < LOGFS_EMBEDDED_FIELDS; i++)
+ di->di_data[i] = cpu_to_be64(li->li_data[i]);
+ break;
+ default:
+ BUG();
+ }
+}
+
+static void __logfs_set_blocks(struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_inode *li = logfs_inode(inode);
+
+ inode->i_blocks = ULONG_MAX;
+ if (li->li_used_bytes >> sb->s_blocksize_bits < ULONG_MAX)
+ inode->i_blocks = ALIGN(li->li_used_bytes, 512) >> 9;
+}
+
+void logfs_set_blocks(struct inode *inode, u64 bytes)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ li->li_used_bytes = bytes;
+ __logfs_set_blocks(inode);
+}
+
+static void prelock_page(struct super_block *sb, struct page *page, int lock)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ BUG_ON(!PageLocked(page));
+ if (lock) {
+ BUG_ON(PagePreLocked(page));
+ SetPagePreLocked(page);
+ } else {
+ /* We are in GC path. */
+ if (PagePreLocked(page))
+ super->s_lock_count++;
+ else
+ SetPagePreLocked(page);
+ }
+}
+
+static void preunlock_page(struct super_block *sb, struct page *page, int lock)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ BUG_ON(!PageLocked(page));
+ if (lock)
+ ClearPagePreLocked(page);
+ else {
+ /* We are in GC path. */
+ BUG_ON(!PagePreLocked(page));
+ if (super->s_lock_count)
+ super->s_lock_count--;
+ else
+ ClearPagePreLocked(page);
+ }
+}
+
+/*
+ * Logfs is prone to an AB-BA deadlock where one task tries to acquire
+ * s_write_mutex with a locked page and GC tries to get that page while holding
+ * s_write_mutex.
+ * To solve this issue logfs will ignore the page lock iff the page in question
+ * is waiting for s_write_mutex. We annotate this fact by setting PG_pre_locked
+ * in addition to PG_locked.
+ */
+static void logfs_get_wblocks(struct super_block *sb, struct page *page,
+ int lock)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ if (page)
+ prelock_page(sb, page, lock);
+
+ if (lock) {
+ mutex_lock(&super->s_write_mutex);
+ logfs_gc_pass(sb);
+ /* FIXME: We also have to check for shadowed space
+ * and mempool fill grade */
+ }
+}
+
+static void logfs_put_wblocks(struct super_block *sb, struct page *page,
+ int lock)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ if (page)
+ preunlock_page(sb, page, lock);
+ /* Order matters - we must clear PG_pre_locked before releasing
+ * s_write_mutex or we could race against another task. */
+ if (lock)
+ mutex_unlock(&super->s_write_mutex);
+}
+
+static struct page *logfs_get_read_page(struct inode *inode, u64 bix,
+ level_t level)
+{
+ return find_or_create_page(inode->i_mapping,
+ logfs_pack_index(bix, level), GFP_NOFS);
+}
+
+static void logfs_put_read_page(struct page *page)
+{
+ unlock_page(page);
+ page_cache_release(page);
+}
+
+static void logfs_lock_write_page(struct page *page)
+{
+ int loop = 0;
+
+ while (unlikely(!trylock_page(page))) {
+ if (loop++ > 0x1000) {
+ /* Has been observed once so far... */
+ printk(KERN_ERR "stack at %p\n", &loop);
+ BUG();
+ }
+ if (PagePreLocked(page)) {
+ /* Holder of page lock is waiting for us, it
+ * is safe to use this page. */
+ break;
+ }
+ /* Some other process has this page locked and has
+ * nothing to do with us. Wait for it to finish.
+ */
+ schedule();
+ }
+ BUG_ON(!PageLocked(page));
+}
+
+static struct page *logfs_get_write_page(struct inode *inode, u64 bix,
+ level_t level)
+{
+ struct address_space *mapping = inode->i_mapping;
+ pgoff_t index = logfs_pack_index(bix, level);
+ struct page *page;
+ int err;
+
+repeat:
+ page = find_get_page(mapping, index);
+ if (!page) {
+ page = __page_cache_alloc(GFP_NOFS);
+ if (!page)
+ return NULL;
+ err = add_to_page_cache_lru(page, mapping, index, GFP_NOFS);
+ if (unlikely(err)) {
+ page_cache_release(page);
+ if (err == -EEXIST)
+ goto repeat;
+ return NULL;
+ }
+ } else logfs_lock_write_page(page);
+ BUG_ON(!PageLocked(page));
+ return page;
+}
+
+static void logfs_unlock_write_page(struct page *page)
+{
+ if (!PagePreLocked(page))
+ unlock_page(page);
+}
+
+static void logfs_put_write_page(struct page *page)
+{
+ logfs_unlock_write_page(page);
+ page_cache_release(page);
+}
+
+static struct page *logfs_get_page(struct inode *inode, u64 bix, level_t level,
+ int rw)
+{
+ if (rw == READ)
+ return logfs_get_read_page(inode, bix, level);
+ else
+ return logfs_get_write_page(inode, bix, level);
+}
+
+static void logfs_put_page(struct page *page, int rw)
+{
+ if (rw == READ)
+ logfs_put_read_page(page);
+ else
+ logfs_put_write_page(page);
+}
+
+static unsigned long __get_bits(u64 val, int skip, int no)
+{
+ u64 ret = val;
+
+ ret >>= skip * no;
+ ret <<= 64 - no;
+ ret >>= 64 - no;
+ return ret;
+}
+
+static unsigned long get_bits(u64 val, level_t skip)
+{
+ return __get_bits(val, (__force int)skip, LOGFS_BLOCK_BITS);
+}
+
+static inline void init_shadow_tree(struct super_block *sb,
+ struct shadow_tree *tree)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ btree_init_mempool64(&tree->new, super->s_btree_pool);
+ btree_init_mempool64(&tree->old, super->s_btree_pool);
+}
+
+static void indirect_write_block(struct logfs_block *block)
+{
+ struct page *page;
+ struct inode *inode;
+ int ret;
+
+ page = block->page;
+ inode = page->mapping->host;
+ logfs_lock_write_page(page);
+ ret = logfs_write_buf(inode, page, 0);
+ logfs_unlock_write_page(page);
+ /*
+ * This needs some rework. Unless you want your filesystem to run
+ * completely synchronously (you don't), the filesystem will always
+ * report writes as 'successful' before the actual work has been
+ * done. The actual work gets done here and this is where any errors
+ * will show up. And there isn't much we can do about it, really.
+ *
+ * Some attempts to fix the errors (move from bad blocks, retry io,...)
+ * have already been done, so anything left should be either a broken
+ * device or a bug somewhere in logfs itself. Being relatively new,
+ * the odds currently favor a bug, so for now the line below isn't
+ * entirely tasteles.
+ */
+ BUG_ON(ret);
+}
+
+static void inode_write_block(struct logfs_block *block)
+{
+ struct inode *inode;
+ int ret;
+
+ inode = block->inode;
+ if (inode->i_ino == LOGFS_INO_MASTER)
+ logfs_write_anchor(inode->i_sb);
+ else {
+ ret = __logfs_write_inode(inode, 0);
+ /* see indirect_write_block comment */
+ BUG_ON(ret);
+ }
+}
+
+static gc_level_t inode_block_level(struct logfs_block *block)
+{
+ BUG_ON(block->inode->i_ino == LOGFS_INO_MASTER);
+ return GC_LEVEL(LOGFS_MAX_LEVELS);
+}
+
+static gc_level_t indirect_block_level(struct logfs_block *block)
+{
+ struct page *page;
+ struct inode *inode;
+ u64 bix;
+ level_t level;
+
+ page = block->page;
+ inode = page->mapping->host;
+ logfs_unpack_index(page->index, &bix, &level);
+ return expand_level(inode->i_ino, level);
+}
+
+/*
+ * This silences a false, yet annoying gcc warning. I hate it when my editor
+ * jumps into bitops.h each time I recompile this file.
+ * TODO: Complain to gcc folks about this and upgrade compiler.
+ */
+static unsigned long fnb(const unsigned long *addr,
+ unsigned long size, unsigned long offset)
+{
+ return find_next_bit(addr, size, offset);
+}
+
+static __be64 inode_val0(struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ u64 val;
+
+ /*
+ * Explicit shifting generates good code, but must match the format
+ * of the structure. Add some paranoia just in case.
+ */
+ BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_mode) != 0);
+ BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_height) != 2);
+ BUILD_BUG_ON(offsetof(struct logfs_disk_inode, di_flags) != 4);
+
+ val = (u64)inode->i_mode << 48 |
+ (u64)li->li_height << 40 |
+ (u64)li->li_flags;
+ return cpu_to_be64(val);
+}
+
+static int inode_write_alias(struct super_block *sb,
+ struct logfs_block *block, write_alias_t *write_one_alias)
+{
+ struct inode *inode = block->inode;
+ struct logfs_inode *li = logfs_inode(inode);
+ unsigned long pos;
+ u64 ino , bix;
+ __be64 val;
+ level_t level;
+ int err;
+
+ for (pos = 0; ; pos++) {
+ pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos);
+ if (pos >= LOGFS_EMBEDDED_FIELDS + INODE_POINTER_OFS)
+ return 0;
+
+ switch (pos) {
+ case INODE_HEIGHT_OFS:
+ val = inode_val0(inode);
+ break;
+ case INODE_USED_OFS:
+ val = cpu_to_be64(li->li_used_bytes);;
+ break;
+ case INODE_SIZE_OFS:
+ val = cpu_to_be64(i_size_read(inode));
+ break;
+ case INODE_POINTER_OFS ... INODE_POINTER_OFS + LOGFS_EMBEDDED_FIELDS - 1:
+ val = cpu_to_be64(li->li_data[pos - INODE_POINTER_OFS]);
+ break;
+ default:
+ BUG();
+ }
+
+ ino = LOGFS_INO_MASTER;
+ bix = inode->i_ino;
+ level = LEVEL(0);
+ err = write_one_alias(sb, ino, bix, level, pos, val);
+ if (err)
+ return err;
+ }
+}
+
+static int indirect_write_alias(struct super_block *sb,
+ struct logfs_block *block, write_alias_t *write_one_alias)
+{
+ unsigned long pos;
+ struct page *page = block->page;
+ u64 ino , bix;
+ __be64 *child, val;
+ level_t level;
+ int err;
+
+ for (pos = 0; ; pos++) {
+ pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos);
+ if (pos >= LOGFS_BLOCK_FACTOR)
+ return 0;
+
+ ino = page->mapping->host->i_ino;
+ logfs_unpack_index(page->index, &bix, &level);
+ child = kmap_atomic(page, KM_USER0);
+ val = child[pos];
+ kunmap_atomic(child, KM_USER0);
+ err = write_one_alias(sb, ino, bix, level, pos, val);
+ if (err)
+ return err;
+ }
+}
+
+int logfs_write_obj_aliases_pagecache(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_block *block;
+ int err;
+
+ list_for_each_entry(block, &super->s_object_alias, alias_list) {
+ err = block->ops->write_alias(sb, block, write_alias_journal);
+ if (err)
+ return err;
+ }
+ return 0;
+}
+
+void __free_block(struct super_block *sb, struct logfs_block *block)
+{
+ BUG_ON(!list_empty(&block->item_list));
+ list_del(&block->alias_list);
+ mempool_free(block, logfs_super(sb)->s_block_pool);
+}
+
+static void inode_free_block(struct super_block *sb, struct logfs_block *block)
+{
+ struct inode *inode = block->inode;
+
+ logfs_inode(inode)->li_block = NULL;
+ __free_block(sb, block);
+}
+
+static void indirect_free_block(struct super_block *sb,
+ struct logfs_block *block)
+{
+ ClearPagePrivate(block->page);
+ block->page->private = 0;
+ __free_block(sb, block);
+}
+
+
+static struct logfs_block_ops inode_block_ops = {
+ .write_block = inode_write_block,
+ .block_level = inode_block_level,
+ .free_block = inode_free_block,
+ .write_alias = inode_write_alias,
+};
+
+struct logfs_block_ops indirect_block_ops = {
+ .write_block = indirect_write_block,
+ .block_level = indirect_block_level,
+ .free_block = indirect_free_block,
+ .write_alias = indirect_write_alias,
+};
+
+struct logfs_block *__alloc_block(struct super_block *sb,
+ u64 ino, u64 bix, level_t level)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_block *block;
+
+ block = mempool_alloc(super->s_block_pool, GFP_NOFS);
+ memset(block, 0, sizeof(*block));
+ INIT_LIST_HEAD(&block->alias_list);
+ INIT_LIST_HEAD(&block->item_list);
+ block->sb = sb;
+ block->ino = ino;
+ block->bix = bix;
+ block->level = level;
+ return block;
+}
+
+static void alloc_inode_block(struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct logfs_block *block;
+
+ if (li->li_block)
+ return;
+
+ block = __alloc_block(inode->i_sb, LOGFS_INO_MASTER, inode->i_ino, 0);
+ block->inode = inode;
+ li->li_block = block;
+ block->ops = &inode_block_ops;
+}
+
+void initialize_block_counters(struct page *page, struct logfs_block *block,
+ __be64 *array, int page_is_empty)
+{
+ u64 ptr;
+ int i, start;
+
+ block->partial = 0;
+ block->full = 0;
+ start = 0;
+ if (page->index < first_indirect_block()) {
+ /* Counters are pointless on level 0 */
+ return;
+ }
+ if (page->index == first_indirect_block()) {
+ /* Skip unused pointers */
+ start = I0_BLOCKS;
+ block->full = I0_BLOCKS;
+ }
+ if (!page_is_empty) {
+ for (i = start; i < LOGFS_BLOCK_FACTOR; i++) {
+ ptr = be64_to_cpu(array[i]);
+ if (ptr)
+ block->partial++;
+ if (ptr & LOGFS_FULLY_POPULATED)
+ block->full++;
+ }
+ }
+}
+
+static void alloc_data_block(struct inode *inode, struct page *page)
+{
+ struct logfs_block *block;
+ u64 bix;
+ level_t level;
+
+ if (PagePrivate(page))
+ return;
+
+ logfs_unpack_index(page->index, &bix, &level);
+ block = __alloc_block(inode->i_sb, inode->i_ino, bix, level);
+ block->page = page;
+ SetPagePrivate(page);
+ page->private = (unsigned long)block;
+ block->ops = &indirect_block_ops;
+}
+
+static void alloc_indirect_block(struct inode *inode, struct page *page,
+ int page_is_empty)
+{
+ struct logfs_block *block;
+ __be64 *array;
+
+ if (PagePrivate(page))
+ return;
+
+ alloc_data_block(inode, page);
+
+ block = logfs_block(page);
+ array = kmap_atomic(page, KM_USER0);
+ initialize_block_counters(page, block, array, page_is_empty);
+ kunmap_atomic(array, KM_USER0);
+}
+
+static void block_set_pointer(struct page *page, int index, u64 ptr)
+{
+ struct logfs_block *block = logfs_block(page);
+ __be64 *array;
+ u64 oldptr;
+
+ BUG_ON(!block);
+ array = kmap_atomic(page, KM_USER0);
+ oldptr = be64_to_cpu(array[index]);
+ array[index] = cpu_to_be64(ptr);
+ kunmap_atomic(array, KM_USER0);
+ SetPageUptodate(page);
+
+ block->full += !!(ptr & LOGFS_FULLY_POPULATED)
+ - !!(oldptr & LOGFS_FULLY_POPULATED);
+ block->partial += !!ptr - !!oldptr;
+}
+
+static u64 block_get_pointer(struct page *page, int index)
+{
+ __be64 *block;
+ u64 ptr;
+
+ block = kmap_atomic(page, KM_USER0);
+ ptr = be64_to_cpu(block[index]);
+ kunmap_atomic(block, KM_USER0);
+ return ptr;
+}
+
+static int logfs_read_empty(struct page *page)
+{
+ zero_user_segment(page, 0, PAGE_CACHE_SIZE);
+ return 0;
+}
+
+static int logfs_read_direct(struct inode *inode, struct page *page)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ pgoff_t index = page->index;
+ u64 block;
+
+ block = li->li_data[index];
+ if (!block)
+ return logfs_read_empty(page);
+
+ return logfs_segment_read(inode, page, block, index, 0);
+}
+
+static int logfs_read_loop(struct inode *inode, struct page *page,
+ int rw_context)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ u64 bix, bofs = li->li_data[INDIRECT_INDEX];
+ level_t level, target_level;
+ int ret;
+ struct page *ipage;
+
+ logfs_unpack_index(page->index, &bix, &target_level);
+ if (!bofs)
+ return logfs_read_empty(page);
+
+ if (bix >= maxbix(li->li_height))
+ return logfs_read_empty(page);
+
+ for (level = LEVEL(li->li_height);
+ (__force u8)level > (__force u8)target_level;
+ level = SUBLEVEL(level)){
+ ipage = logfs_get_page(inode, bix, level, rw_context);
+ if (!ipage)
+ return -ENOMEM;
+
+ ret = logfs_segment_read(inode, ipage, bofs, bix, level);
+ if (ret) {
+ logfs_put_read_page(ipage);
+ return ret;
+ }
+
+ bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level)));
+ logfs_put_page(ipage, rw_context);
+ if (!bofs)
+ return logfs_read_empty(page);
+ }
+
+ return logfs_segment_read(inode, page, bofs, bix, 0);
+}
+
+static int logfs_read_block(struct inode *inode, struct page *page,
+ int rw_context)
+{
+ pgoff_t index = page->index;
+
+ if (index < I0_BLOCKS)
+ return logfs_read_direct(inode, page);
+ return logfs_read_loop(inode, page, rw_context);
+}
+
+static int logfs_exist_loop(struct inode *inode, u64 bix)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ u64 bofs = li->li_data[INDIRECT_INDEX];
+ level_t level;
+ int ret;
+ struct page *ipage;
+
+ if (!bofs)
+ return 0;
+ if (bix >= maxbix(li->li_height))
+ return 0;
+
+ for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) {
+ ipage = logfs_get_read_page(inode, bix, level);
+ if (!ipage)
+ return -ENOMEM;
+
+ ret = logfs_segment_read(inode, ipage, bofs, bix, level);
+ if (ret) {
+ logfs_put_read_page(ipage);
+ return ret;
+ }
+
+ bofs = block_get_pointer(ipage, get_bits(bix, SUBLEVEL(level)));
+ logfs_put_read_page(ipage);
+ if (!bofs)
+ return 0;
+ }
+
+ return 1;
+}
+
+int logfs_exist_block(struct inode *inode, u64 bix)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if (bix < I0_BLOCKS)
+ return !!li->li_data[bix];
+ return logfs_exist_loop(inode, bix);
+}
+
+static u64 seek_holedata_direct(struct inode *inode, u64 bix, int data)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ for (; bix < I0_BLOCKS; bix++)
+ if (data ^ (li->li_data[bix] == 0))
+ return bix;
+ return I0_BLOCKS;
+}
+
+static u64 seek_holedata_loop(struct inode *inode, u64 bix, int data)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ __be64 *rblock;
+ u64 increment, bofs = li->li_data[INDIRECT_INDEX];
+ level_t level;
+ int ret, slot;
+ struct page *page;
+
+ BUG_ON(!bofs);
+
+ for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)) {
+ increment = 1 << (LOGFS_BLOCK_BITS * ((__force u8)level-1));
+ page = logfs_get_read_page(inode, bix, level);
+ if (!page)
+ return bix;
+
+ ret = logfs_segment_read(inode, page, bofs, bix, level);
+ if (ret) {
+ logfs_put_read_page(page);
+ return bix;
+ }
+
+ slot = get_bits(bix, SUBLEVEL(level));
+ rblock = kmap_atomic(page, KM_USER0);
+ while (slot < LOGFS_BLOCK_FACTOR) {
+ if (data && (rblock[slot] != 0))
+ break;
+ if (!data && !(be64_to_cpu(rblock[slot]) & LOGFS_FULLY_POPULATED))
+ break;
+ slot++;
+ bix += increment;
+ bix &= ~(increment - 1);
+ }
+ if (slot >= LOGFS_BLOCK_FACTOR) {
+ kunmap_atomic(rblock, KM_USER0);
+ logfs_put_read_page(page);
+ return bix;
+ }
+ bofs = be64_to_cpu(rblock[slot]);
+ kunmap_atomic(rblock, KM_USER0);
+ logfs_put_read_page(page);
+ if (!bofs) {
+ BUG_ON(data);
+ return bix;
+ }
+ }
+ return bix;
+}
+
+/**
+ * logfs_seek_hole - find next hole starting at a given block index
+ * @inode: inode to search in
+ * @bix: block index to start searching
+ *
+ * Returns next hole. If the file doesn't contain any further holes, the
+ * block address next to eof is returned instead.
+ */
+u64 logfs_seek_hole(struct inode *inode, u64 bix)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if (bix < I0_BLOCKS) {
+ bix = seek_holedata_direct(inode, bix, 0);
+ if (bix < I0_BLOCKS)
+ return bix;
+ }
+
+ if (!li->li_data[INDIRECT_INDEX])
+ return bix;
+ else if (li->li_data[INDIRECT_INDEX] & LOGFS_FULLY_POPULATED)
+ bix = maxbix(li->li_height);
+ else {
+ bix = seek_holedata_loop(inode, bix, 0);
+ if (bix < maxbix(li->li_height))
+ return bix;
+ /* Should not happen anymore. But if some port writes semi-
+ * corrupt images (as this one used to) we might run into it.
+ */
+ WARN_ON_ONCE(bix == maxbix(li->li_height));
+ }
+
+ return bix;
+}
+
+static u64 __logfs_seek_data(struct inode *inode, u64 bix)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if (bix < I0_BLOCKS) {
+ bix = seek_holedata_direct(inode, bix, 1);
+ if (bix < I0_BLOCKS)
+ return bix;
+ }
+
+ if (bix < maxbix(li->li_height)) {
+ if (!li->li_data[INDIRECT_INDEX])
+ bix = maxbix(li->li_height);
+ else
+ return seek_holedata_loop(inode, bix, 1);
+ }
+
+ return bix;
+}
+
+/**
+ * logfs_seek_data - find next data block after a given block index
+ * @inode: inode to search in
+ * @bix: block index to start searching
+ *
+ * Returns next data block. If the file doesn't contain any further data
+ * blocks, the last block in the file is returned instead.
+ */
+u64 logfs_seek_data(struct inode *inode, u64 bix)
+{
+ struct super_block *sb = inode->i_sb;
+ u64 ret, end;
+
+ ret = __logfs_seek_data(inode, bix);
+ end = i_size_read(inode) >> sb->s_blocksize_bits;
+ if (ret >= end)
+ ret = max(bix, end);
+ return ret;
+}
+
+static int logfs_is_valid_direct(struct logfs_inode *li, u64 bix, u64 ofs)
+{
+ return pure_ofs(li->li_data[bix]) == ofs;
+}
+
+static int __logfs_is_valid_loop(struct inode *inode, u64 bix,
+ u64 ofs, u64 bofs)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ level_t level;
+ int ret;
+ struct page *page;
+
+ for (level = LEVEL(li->li_height); level != 0; level = SUBLEVEL(level)){
+ page = logfs_get_write_page(inode, bix, level);
+ BUG_ON(!page);
+
+ ret = logfs_segment_read(inode, page, bofs, bix, level);
+ if (ret) {
+ logfs_put_write_page(page);
+ return 0;
+ }
+
+ bofs = block_get_pointer(page, get_bits(bix, SUBLEVEL(level)));
+ logfs_put_write_page(page);
+ if (!bofs)
+ return 0;
+
+ if (pure_ofs(bofs) == ofs)
+ return 1;
+ }
+ return 0;
+}
+
+static int logfs_is_valid_loop(struct inode *inode, u64 bix, u64 ofs)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ u64 bofs = li->li_data[INDIRECT_INDEX];
+
+ if (!bofs)
+ return 0;
+
+ if (bix >= maxbix(li->li_height))
+ return 0;
+
+ if (pure_ofs(bofs) == ofs)
+ return 1;
+
+ return __logfs_is_valid_loop(inode, bix, ofs, bofs);
+}
+
+static int __logfs_is_valid_block(struct inode *inode, u64 bix, u64 ofs)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if ((inode->i_nlink == 0) && atomic_read(&inode->i_count) == 1)
+ return 0;
+
+ if (bix < I0_BLOCKS)
+ return logfs_is_valid_direct(li, bix, ofs);
+ return logfs_is_valid_loop(inode, bix, ofs);
+}
+
+/**
+ * logfs_is_valid_block - check whether this block is still valid
+ *
+ * @sb - superblock
+ * @ofs - block physical offset
+ * @ino - block inode number
+ * @bix - block index
+ * @level - block level
+ *
+ * Returns 0 if the block is invalid, 1 if it is valid and 2 if it will
+ * become invalid once the journal is written.
+ */
+int logfs_is_valid_block(struct super_block *sb, u64 ofs, u64 ino, u64 bix,
+ gc_level_t gc_level)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode;
+ int ret, cookie;
+
+ /* Umount closes a segment with free blocks remaining. Those
+ * blocks are by definition invalid. */
+ if (ino == -1)
+ return 0;
+
+ LOGFS_BUG_ON((u64)(u_long)ino != ino, sb);
+
+ inode = logfs_safe_iget(sb, ino, &cookie);
+ if (IS_ERR(inode))
+ goto invalid;
+
+ ret = __logfs_is_valid_block(inode, bix, ofs);
+ logfs_safe_iput(inode, cookie);
+ if (ret)
+ return ret;
+
+invalid:
+ /* Block is nominally invalid, but may still sit in the shadow tree,
+ * waiting for a journal commit.
+ */
+ if (btree_lookup64(&super->s_shadow_tree.old, ofs))
+ return 2;
+ return 0;
+}
+
+int logfs_readpage_nolock(struct page *page)
+{
+ struct inode *inode = page->mapping->host;
+ int ret = -EIO;
+
+ ret = logfs_read_block(inode, page, READ);
+
+ if (ret) {
+ ClearPageUptodate(page);
+ SetPageError(page);
+ } else {
+ SetPageUptodate(page);
+ ClearPageError(page);
+ }
+ flush_dcache_page(page);
+
+ return ret;
+}
+
+static int logfs_reserve_bytes(struct inode *inode, int bytes)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ u64 available = super->s_free_bytes + super->s_dirty_free_bytes
+ - super->s_dirty_used_bytes - super->s_dirty_pages;
+
+ if (!bytes)
+ return 0;
+
+ if (available < bytes)
+ return -ENOSPC;
+
+ if (available < bytes + super->s_root_reserve &&
+ !capable(CAP_SYS_RESOURCE))
+ return -ENOSPC;
+
+ return 0;
+}
+
+int get_page_reserve(struct inode *inode, struct page *page)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ int ret;
+
+ if (logfs_block(page) && logfs_block(page)->reserved_bytes)
+ return 0;
+
+ logfs_get_wblocks(inode->i_sb, page, WF_LOCK);
+ ret = logfs_reserve_bytes(inode, 6 * LOGFS_MAX_OBJECTSIZE);
+ if (!ret) {
+ alloc_data_block(inode, page);
+ logfs_block(page)->reserved_bytes += 6 * LOGFS_MAX_OBJECTSIZE;
+ super->s_dirty_pages += 6 * LOGFS_MAX_OBJECTSIZE;
+ }
+ logfs_put_wblocks(inode->i_sb, page, WF_LOCK);
+ return ret;
+}
+
+/*
+ * We are protected by write lock. Push victims up to superblock level
+ * and release transaction when appropriate.
+ */
+/* FIXME: This is currently called from the wrong spots. */
+static void logfs_handle_transaction(struct inode *inode,
+ struct logfs_transaction *ta)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+
+ if (!ta)
+ return;
+ logfs_inode(inode)->li_block->ta = NULL;
+
+ if (inode->i_ino != LOGFS_INO_MASTER) {
+ BUG(); /* FIXME: Yes, this needs more thought */
+ /* just remember the transaction until inode is written */
+ //BUG_ON(logfs_inode(inode)->li_transaction);
+ //logfs_inode(inode)->li_transaction = ta;
+ return;
+ }
+
+ switch (ta->state) {
+ case CREATE_1: /* fall through */
+ case UNLINK_1:
+ BUG_ON(super->s_victim_ino);
+ super->s_victim_ino = ta->ino;
+ break;
+ case CREATE_2: /* fall through */
+ case UNLINK_2:
+ BUG_ON(super->s_victim_ino != ta->ino);
+ super->s_victim_ino = 0;
+ /* transaction ends here - free it */
+ kfree(ta);
+ break;
+ case CROSS_RENAME_1:
+ BUG_ON(super->s_rename_dir);
+ BUG_ON(super->s_rename_pos);
+ super->s_rename_dir = ta->dir;
+ super->s_rename_pos = ta->pos;
+ break;
+ case CROSS_RENAME_2:
+ BUG_ON(super->s_rename_dir != ta->dir);
+ BUG_ON(super->s_rename_pos != ta->pos);
+ super->s_rename_dir = 0;
+ super->s_rename_pos = 0;
+ kfree(ta);
+ break;
+ case TARGET_RENAME_1:
+ BUG_ON(super->s_rename_dir);
+ BUG_ON(super->s_rename_pos);
+ BUG_ON(super->s_victim_ino);
+ super->s_rename_dir = ta->dir;
+ super->s_rename_pos = ta->pos;
+ super->s_victim_ino = ta->ino;
+ break;
+ case TARGET_RENAME_2:
+ BUG_ON(super->s_rename_dir != ta->dir);
+ BUG_ON(super->s_rename_pos != ta->pos);
+ BUG_ON(super->s_victim_ino != ta->ino);
+ super->s_rename_dir = 0;
+ super->s_rename_pos = 0;
+ break;
+ case TARGET_RENAME_3:
+ BUG_ON(super->s_rename_dir);
+ BUG_ON(super->s_rename_pos);
+ BUG_ON(super->s_victim_ino != ta->ino);
+ super->s_victim_ino = 0;
+ kfree(ta);
+ break;
+ default:
+ BUG();
+ }
+}
+
+/*
+ * Not strictly a reservation, but rather a check that we still have enough
+ * space to satisfy the write.
+ */
+static int logfs_reserve_blocks(struct inode *inode, int blocks)
+{
+ return logfs_reserve_bytes(inode, blocks * LOGFS_MAX_OBJECTSIZE);
+}
+
+struct write_control {
+ u64 ofs;
+ long flags;
+};
+
+static struct logfs_shadow *alloc_shadow(struct inode *inode, u64 bix,
+ level_t level, u64 old_ofs)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ struct logfs_shadow *shadow;
+
+ shadow = mempool_alloc(super->s_shadow_pool, GFP_NOFS);
+ memset(shadow, 0, sizeof(*shadow));
+ shadow->ino = inode->i_ino;
+ shadow->bix = bix;
+ shadow->gc_level = expand_level(inode->i_ino, level);
+ shadow->old_ofs = old_ofs & ~LOGFS_FULLY_POPULATED;
+ return shadow;
+}
+
+static void free_shadow(struct inode *inode, struct logfs_shadow *shadow)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+
+ mempool_free(shadow, super->s_shadow_pool);
+}
+
+/**
+ * fill_shadow_tree - Propagate shadow tree changes due to a write
+ * @inode: Inode owning the page
+ * @page: Struct page that was written
+ * @shadow: Shadow for the current write
+ *
+ * Writes in logfs can result in two semi-valid objects. The old object
+ * is still valid as long as it can be reached by following pointers on
+ * the medium. Only when writes propagate all the way up to the journal
+ * has the new object safely replaced the old one.
+ *
+ * To handle this problem, a struct logfs_shadow is used to represent
+ * every single write. It is attached to the indirect block, which is
+ * marked dirty. When the indirect block is written, its shadows are
+ * handed up to the next indirect block (or inode). Untimately they
+ * will reach the master inode and be freed upon journal commit.
+ *
+ * This function handles a single step in the propagation. It adds the
+ * shadow for the current write to the tree, along with any shadows in
+ * the page's tree, in case it was an indirect block. If a page is
+ * written, the inode parameter is left NULL, if an inode is written,
+ * the page parameter is left NULL.
+ */
+static void fill_shadow_tree(struct inode *inode, struct page *page,
+ struct logfs_shadow *shadow)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ struct logfs_block *block = logfs_block(page);
+ struct shadow_tree *tree = &super->s_shadow_tree;
+
+ if (PagePrivate(page)) {
+ if (block->alias_map)
+ super->s_no_object_aliases -= bitmap_weight(
+ block->alias_map, LOGFS_BLOCK_FACTOR);
+ logfs_handle_transaction(inode, block->ta);
+ block->ops->free_block(inode->i_sb, block);
+ }
+ if (shadow) {
+ if (shadow->old_ofs)
+ btree_insert64(&tree->old, shadow->old_ofs, shadow,
+ GFP_NOFS);
+ else
+ btree_insert64(&tree->new, shadow->new_ofs, shadow,
+ GFP_NOFS);
+
+ super->s_dirty_used_bytes += shadow->new_len;
+ super->s_dirty_free_bytes += shadow->old_len;
+ }
+}
+
+static void logfs_set_alias(struct super_block *sb, struct logfs_block *block,
+ long child_no)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ if (block->inode && block->inode->i_ino == LOGFS_INO_MASTER) {
+ /* Aliases in the master inode are pointless. */
+ return;
+ }
+
+ if (!test_bit(child_no, block->alias_map)) {
+ set_bit(child_no, block->alias_map);
+ super->s_no_object_aliases++;
+ }
+ list_move_tail(&block->alias_list, &super->s_object_alias);
+}
+
+/*
+ * Object aliases can and often do change the size and occupied space of a
+ * file. So not only do we have to change the pointers, we also have to
+ * change inode->i_size and li->li_used_bytes. Which is done by setting
+ * another two object aliases for the inode itself.
+ */
+static void set_iused(struct inode *inode, struct logfs_shadow *shadow)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if (shadow->new_len == shadow->old_len)
+ return;
+
+ alloc_inode_block(inode);
+ li->li_used_bytes += shadow->new_len - shadow->old_len;
+ __logfs_set_blocks(inode);
+ logfs_set_alias(inode->i_sb, li->li_block, INODE_USED_OFS);
+ logfs_set_alias(inode->i_sb, li->li_block, INODE_SIZE_OFS);
+}
+
+static int logfs_write_i0(struct inode *inode, struct page *page,
+ struct write_control *wc)
+{
+ struct logfs_shadow *shadow;
+ u64 bix;
+ level_t level;
+ int full, err = 0;
+
+ logfs_unpack_index(page->index, &bix, &level);
+ if (wc->ofs == 0)
+ if (logfs_reserve_blocks(inode, 1))
+ return -ENOSPC;
+
+ shadow = alloc_shadow(inode, bix, level, wc->ofs);
+ if (wc->flags & WF_WRITE)
+ err = logfs_segment_write(inode, page, shadow);
+ if (wc->flags & WF_DELETE)
+ logfs_segment_delete(inode, shadow);
+ if (err) {
+ free_shadow(inode, shadow);
+ return err;
+ }
+
+ set_iused(inode, shadow);
+ full = 1;
+ if (level != 0) {
+ alloc_indirect_block(inode, page, 0);
+ full = logfs_block(page)->full == LOGFS_BLOCK_FACTOR;
+ }
+ fill_shadow_tree(inode, page, shadow);
+ wc->ofs = shadow->new_ofs;
+ if (wc->ofs && full)
+ wc->ofs |= LOGFS_FULLY_POPULATED;
+ return 0;
+}
+
+static int logfs_write_direct(struct inode *inode, struct page *page,
+ long flags)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct write_control wc = {
+ .ofs = li->li_data[page->index],
+ .flags = flags,
+ };
+ int err;
+
+ alloc_inode_block(inode);
+
+ err = logfs_write_i0(inode, page, &wc);
+ if (err)
+ return err;
+
+ li->li_data[page->index] = wc.ofs;
+ logfs_set_alias(inode->i_sb, li->li_block,
+ page->index + INODE_POINTER_OFS);
+ return 0;
+}
+
+static int ptr_change(u64 ofs, struct page *page)
+{
+ struct logfs_block *block = logfs_block(page);
+ int empty0, empty1, full0, full1;
+
+ empty0 = ofs == 0;
+ empty1 = block->partial == 0;
+ if (empty0 != empty1)
+ return 1;
+
+ /* The !! is necessary to shrink result to int */
+ full0 = !!(ofs & LOGFS_FULLY_POPULATED);
+ full1 = block->full == LOGFS_BLOCK_FACTOR;
+ if (full0 != full1)
+ return 1;
+ return 0;
+}
+
+static int __logfs_write_rec(struct inode *inode, struct page *page,
+ struct write_control *this_wc,
+ pgoff_t bix, level_t target_level, level_t level)
+{
+ int ret, page_empty = 0;
+ int child_no = get_bits(bix, SUBLEVEL(level));
+ struct page *ipage;
+ struct write_control child_wc = {
+ .flags = this_wc->flags,
+ };
+
+ ipage = logfs_get_write_page(inode, bix, level);
+ if (!ipage)
+ return -ENOMEM;
+
+ if (this_wc->ofs) {
+ ret = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level);
+ if (ret)
+ goto out;
+ } else if (!PageUptodate(ipage)) {
+ page_empty = 1;
+ logfs_read_empty(ipage);
+ }
+
+ child_wc.ofs = block_get_pointer(ipage, child_no);
+
+ if ((__force u8)level-1 > (__force u8)target_level)
+ ret = __logfs_write_rec(inode, page, &child_wc, bix,
+ target_level, SUBLEVEL(level));
+ else
+ ret = logfs_write_i0(inode, page, &child_wc);
+
+ if (ret)
+ goto out;
+
+ alloc_indirect_block(inode, ipage, page_empty);
+ block_set_pointer(ipage, child_no, child_wc.ofs);
+ /* FIXME: first condition seems superfluous */
+ if (child_wc.ofs || logfs_block(ipage)->partial)
+ this_wc->flags |= WF_WRITE;
+ /* the condition on this_wc->ofs ensures that we won't consume extra
+ * space for indirect blocks in the future, which we cannot reserve */
+ if (!this_wc->ofs || ptr_change(this_wc->ofs, ipage))
+ ret = logfs_write_i0(inode, ipage, this_wc);
+ else
+ logfs_set_alias(inode->i_sb, logfs_block(ipage), child_no);
+out:
+ logfs_put_write_page(ipage);
+ return ret;
+}
+
+static int logfs_write_rec(struct inode *inode, struct page *page,
+ pgoff_t bix, level_t target_level, long flags)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct write_control wc = {
+ .ofs = li->li_data[INDIRECT_INDEX],
+ .flags = flags,
+ };
+ int ret;
+
+ alloc_inode_block(inode);
+
+ if (li->li_height > (__force u8)target_level)
+ ret = __logfs_write_rec(inode, page, &wc, bix, target_level,
+ LEVEL(li->li_height));
+ else
+ ret = logfs_write_i0(inode, page, &wc);
+ if (ret)
+ return ret;
+
+ if (li->li_data[INDIRECT_INDEX] != wc.ofs) {
+ li->li_data[INDIRECT_INDEX] = wc.ofs;
+ logfs_set_alias(inode->i_sb, li->li_block,
+ INDIRECT_INDEX + INODE_POINTER_OFS);
+ }
+ return ret;
+}
+
+void logfs_add_transaction(struct inode *inode, struct logfs_transaction *ta)
+{
+ alloc_inode_block(inode);
+ logfs_inode(inode)->li_block->ta = ta;
+}
+
+void logfs_del_transaction(struct inode *inode, struct logfs_transaction *ta)
+{
+ struct logfs_block *block = logfs_inode(inode)->li_block;
+
+ if (block && block->ta)
+ block->ta = NULL;
+}
+
+static int grow_inode(struct inode *inode, u64 bix, level_t level)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ u8 height = (__force u8)level;
+ struct page *page;
+ struct write_control wc = {
+ .flags = WF_WRITE,
+ };
+ int err;
+
+ BUG_ON(height > 5 || li->li_height > 5);
+ while (height > li->li_height || bix >= maxbix(li->li_height)) {
+ page = logfs_get_write_page(inode, I0_BLOCKS + 1,
+ LEVEL(li->li_height + 1));
+ if (!page)
+ return -ENOMEM;
+ logfs_read_empty(page);
+ alloc_indirect_block(inode, page, 1);
+ block_set_pointer(page, 0, li->li_data[INDIRECT_INDEX]);
+ err = logfs_write_i0(inode, page, &wc);
+ logfs_put_write_page(page);
+ if (err)
+ return err;
+ li->li_data[INDIRECT_INDEX] = wc.ofs;
+ wc.ofs = 0;
+ li->li_height++;
+ logfs_set_alias(inode->i_sb, li->li_block, INODE_HEIGHT_OFS);
+ }
+ return 0;
+}
+
+static int __logfs_write_buf(struct inode *inode, struct page *page, long flags)
+{
+ struct logfs_super *super = logfs_super(inode->i_sb);
+ pgoff_t index = page->index;
+ u64 bix;
+ level_t level;
+ int err;
+
+ flags |= WF_WRITE | WF_DELETE;
+ inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+
+ logfs_unpack_index(index, &bix, &level);
+ if (logfs_block(page) && logfs_block(page)->reserved_bytes)
+ super->s_dirty_pages -= logfs_block(page)->reserved_bytes;
+
+ if (index < I0_BLOCKS)
+ return logfs_write_direct(inode, page, flags);
+
+ bix = adjust_bix(bix, level);
+ err = grow_inode(inode, bix, level);
+ if (err)
+ return err;
+ return logfs_write_rec(inode, page, bix, level, flags);
+}
+
+int logfs_write_buf(struct inode *inode, struct page *page, long flags)
+{
+ struct super_block *sb = inode->i_sb;
+ int ret;
+
+ logfs_get_wblocks(sb, page, flags & WF_LOCK);
+ ret = __logfs_write_buf(inode, page, flags);
+ logfs_put_wblocks(sb, page, flags & WF_LOCK);
+ return ret;
+}
+
+static int __logfs_delete(struct inode *inode, struct page *page)
+{
+ long flags = WF_DELETE;
+
+ inode->i_ctime = inode->i_mtime = CURRENT_TIME;
+
+ if (page->index < I0_BLOCKS)
+ return logfs_write_direct(inode, page, flags);
+ return logfs_write_rec(inode, page, page->index, 0, flags);
+}
+
+int logfs_delete(struct inode *inode, pgoff_t index,
+ struct shadow_tree *shadow_tree)
+{
+ struct super_block *sb = inode->i_sb;
+ struct page *page;
+ int ret;
+
+ page = logfs_get_read_page(inode, index, 0);
+ if (!page)
+ return -ENOMEM;
+
+ logfs_get_wblocks(sb, page, 1);
+ ret = __logfs_delete(inode, page);
+ logfs_put_wblocks(sb, page, 1);
+
+ logfs_put_read_page(page);
+
+ return ret;
+}
+
+/* Rewrite cannot mark the inode dirty but has to write it immediatly. */
+int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs,
+ gc_level_t gc_level, long flags)
+{
+ level_t level = shrink_level(gc_level);
+ struct page *page;
+ int err;
+
+ page = logfs_get_write_page(inode, bix, level);
+ if (!page)
+ return -ENOMEM;
+
+ err = logfs_segment_read(inode, page, ofs, bix, level);
+ if (!err) {
+ if (level != 0)
+ alloc_indirect_block(inode, page, 0);
+ err = logfs_write_buf(inode, page, flags);
+ }
+ logfs_put_write_page(page);
+ return err;
+}
+
+static int truncate_data_block(struct inode *inode, struct page *page,
+ u64 ofs, struct logfs_shadow *shadow, u64 size)
+{
+ loff_t pageofs = page->index << inode->i_sb->s_blocksize_bits;
+ u64 bix;
+ level_t level;
+ int err;
+
+ /* Does truncation happen within this page? */
+ if (size <= pageofs || size - pageofs >= PAGE_SIZE)
+ return 0;
+
+ logfs_unpack_index(page->index, &bix, &level);
+ BUG_ON(level != 0);
+
+ err = logfs_segment_read(inode, page, ofs, bix, level);
+ if (err)
+ return err;
+
+ zero_user_segment(page, size - pageofs, PAGE_CACHE_SIZE);
+ return logfs_segment_write(inode, page, shadow);
+}
+
+static int logfs_truncate_i0(struct inode *inode, struct page *page,
+ struct write_control *wc, u64 size)
+{
+ struct logfs_shadow *shadow;
+ u64 bix;
+ level_t level;
+ int err = 0;
+
+ logfs_unpack_index(page->index, &bix, &level);
+ BUG_ON(level != 0);
+ shadow = alloc_shadow(inode, bix, level, wc->ofs);
+
+ err = truncate_data_block(inode, page, wc->ofs, shadow, size);
+ if (err) {
+ free_shadow(inode, shadow);
+ return err;
+ }
+
+ logfs_segment_delete(inode, shadow);
+ set_iused(inode, shadow);
+ fill_shadow_tree(inode, page, shadow);
+ wc->ofs = shadow->new_ofs;
+ return 0;
+}
+
+static int logfs_truncate_direct(struct inode *inode, u64 size)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct write_control wc;
+ struct page *page;
+ int e;
+ int err;
+
+ alloc_inode_block(inode);
+
+ for (e = I0_BLOCKS - 1; e >= 0; e--) {
+ if (size > (e+1) * LOGFS_BLOCKSIZE)
+ break;
+
+ wc.ofs = li->li_data[e];
+ if (!wc.ofs)
+ continue;
+
+ page = logfs_get_write_page(inode, e, 0);
+ if (!page)
+ return -ENOMEM;
+ err = logfs_segment_read(inode, page, wc.ofs, e, 0);
+ if (err) {
+ logfs_put_write_page(page);
+ return err;
+ }
+ err = logfs_truncate_i0(inode, page, &wc, size);
+ logfs_put_write_page(page);
+ if (err)
+ return err;
+
+ li->li_data[e] = wc.ofs;
+ }
+ return 0;
+}
+
+/* FIXME: these need to become per-sb once we support different blocksizes */
+static u64 __logfs_step[] = {
+ 1,
+ I1_BLOCKS,
+ I2_BLOCKS,
+ I3_BLOCKS,
+};
+
+static u64 __logfs_start_index[] = {
+ I0_BLOCKS,
+ I1_BLOCKS,
+ I2_BLOCKS,
+ I3_BLOCKS
+};
+
+static inline u64 logfs_step(level_t level)
+{
+ return __logfs_step[(__force u8)level];
+}
+
+static inline u64 logfs_factor(u8 level)
+{
+ return __logfs_step[level] * LOGFS_BLOCKSIZE;
+}
+
+static inline u64 logfs_start_index(level_t level)
+{
+ return __logfs_start_index[(__force u8)level];
+}
+
+static void logfs_unpack_raw_index(pgoff_t index, u64 *bix, level_t *level)
+{
+ logfs_unpack_index(index, bix, level);
+ if (*bix <= logfs_start_index(SUBLEVEL(*level)))
+ *bix = 0;
+}
+
+static int __logfs_truncate_rec(struct inode *inode, struct page *ipage,
+ struct write_control *this_wc, u64 size)
+{
+ int truncate_happened = 0;
+ int e, err = 0;
+ u64 bix, child_bix, next_bix;
+ level_t level;
+ struct page *page;
+ struct write_control child_wc = { /* FIXME: flags */ };
+
+ logfs_unpack_raw_index(ipage->index, &bix, &level);
+ err = logfs_segment_read(inode, ipage, this_wc->ofs, bix, level);
+ if (err)
+ return err;
+
+ for (e = LOGFS_BLOCK_FACTOR - 1; e >= 0; e--) {
+ child_bix = bix + e * logfs_step(SUBLEVEL(level));
+ next_bix = child_bix + logfs_step(SUBLEVEL(level));
+ if (size > next_bix * LOGFS_BLOCKSIZE)
+ break;
+
+ child_wc.ofs = pure_ofs(block_get_pointer(ipage, e));
+ if (!child_wc.ofs)
+ continue;
+
+ page = logfs_get_write_page(inode, child_bix, SUBLEVEL(level));
+ if (!page)
+ return -ENOMEM;
+
+ if ((__force u8)level > 1)
+ err = __logfs_truncate_rec(inode, page, &child_wc, size);
+ else
+ err = logfs_truncate_i0(inode, page, &child_wc, size);
+ logfs_put_write_page(page);
+ if (err)
+ return err;
+
+ truncate_happened = 1;
+ alloc_indirect_block(inode, ipage, 0);
+ block_set_pointer(ipage, e, child_wc.ofs);
+ }
+
+ if (!truncate_happened) {
+ printk("ineffectual truncate (%lx, %lx, %llx)\n", inode->i_ino, ipage->index, size);
+ return 0;
+ }
+
+ this_wc->flags = WF_DELETE;
+ if (logfs_block(ipage)->partial)
+ this_wc->flags |= WF_WRITE;
+
+ return logfs_write_i0(inode, ipage, this_wc);
+}
+
+static int logfs_truncate_rec(struct inode *inode, u64 size)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct write_control wc = {
+ .ofs = li->li_data[INDIRECT_INDEX],
+ };
+ struct page *page;
+ int err;
+
+ alloc_inode_block(inode);
+
+ if (!wc.ofs)
+ return 0;
+
+ page = logfs_get_write_page(inode, 0, LEVEL(li->li_height));
+ if (!page)
+ return -ENOMEM;
+
+ err = __logfs_truncate_rec(inode, page, &wc, size);
+ logfs_put_write_page(page);
+ if (err)
+ return err;
+
+ if (li->li_data[INDIRECT_INDEX] != wc.ofs)
+ li->li_data[INDIRECT_INDEX] = wc.ofs;
+ return 0;
+}
+
+static int __logfs_truncate(struct inode *inode, u64 size)
+{
+ int ret;
+
+ if (size >= logfs_factor(logfs_inode(inode)->li_height))
+ return 0;
+
+ ret = logfs_truncate_rec(inode, size);
+ if (ret)
+ return ret;
+
+ return logfs_truncate_direct(inode, size);
+}
+
+int logfs_truncate(struct inode *inode, u64 size)
+{
+ struct super_block *sb = inode->i_sb;
+ int err;
+
+ logfs_get_wblocks(sb, NULL, 1);
+ err = __logfs_truncate(inode, size);
+ if (!err)
+ err = __logfs_write_inode(inode, 0);
+ logfs_put_wblocks(sb, NULL, 1);
+
+ if (!err)
+ err = vmtruncate(inode, size);
+
+ /* I don't trust error recovery yet. */
+ WARN_ON(err);
+ return err;
+}
+
+static void move_page_to_inode(struct inode *inode, struct page *page)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct logfs_block *block = logfs_block(page);
+
+ if (!block)
+ return;
+
+ log_blockmove("move_page_to_inode(%llx, %llx, %x)\n",
+ block->ino, block->bix, block->level);
+ BUG_ON(li->li_block);
+ block->ops = &inode_block_ops;
+ block->inode = inode;
+ li->li_block = block;
+
+ block->page = NULL;
+ page->private = 0;
+ ClearPagePrivate(page);
+}
+
+static void move_inode_to_page(struct page *page, struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+ struct logfs_block *block = li->li_block;
+
+ if (!block)
+ return;
+
+ log_blockmove("move_inode_to_page(%llx, %llx, %x)\n",
+ block->ino, block->bix, block->level);
+ BUG_ON(PagePrivate(page));
+ block->ops = &indirect_block_ops;
+ block->page = page;
+ page->private = (unsigned long)block;
+ SetPagePrivate(page);
+
+ block->inode = NULL;
+ li->li_block = NULL;
+}
+
+int logfs_read_inode(struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *master_inode = super->s_master_inode;
+ struct page *page;
+ struct logfs_disk_inode *di;
+ u64 ino = inode->i_ino;
+
+ if (ino << sb->s_blocksize_bits > i_size_read(master_inode))
+ return -ENODATA;
+ if (!logfs_exist_block(master_inode, ino))
+ return -ENODATA;
+
+ page = read_cache_page(master_inode->i_mapping, ino,
+ (filler_t *)logfs_readpage, NULL);
+ if (IS_ERR(page))
+ return PTR_ERR(page);
+
+ di = kmap_atomic(page, KM_USER0);
+ logfs_disk_to_inode(di, inode);
+ kunmap_atomic(di, KM_USER0);
+ move_page_to_inode(inode, page);
+ page_cache_release(page);
+ return 0;
+}
+
+/* Caller must logfs_put_write_page(page); */
+static struct page *inode_to_page(struct inode *inode)
+{
+ struct inode *master_inode = logfs_super(inode->i_sb)->s_master_inode;
+ struct logfs_disk_inode *di;
+ struct page *page;
+
+ BUG_ON(inode->i_ino == LOGFS_INO_MASTER);
+
+ page = logfs_get_write_page(master_inode, inode->i_ino, 0);
+ if (!page)
+ return NULL;
+
+ di = kmap_atomic(page, KM_USER0);
+ logfs_inode_to_disk(inode, di);
+ kunmap_atomic(di, KM_USER0);
+ move_inode_to_page(page, inode);
+ return page;
+}
+
+/* Cheaper version of write_inode. All changes are concealed in
+ * aliases, which are moved back. No write to the medium happens.
+ */
+void logfs_clear_inode(struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_inode *li = logfs_inode(inode);
+ struct logfs_block *block = li->li_block;
+ struct page *page;
+
+ /* Only deleted files may be dirty at this point */
+ BUG_ON(inode->i_state & I_DIRTY && inode->i_nlink);
+ if (!block)
+ return;
+ if ((logfs_super(sb)->s_flags & LOGFS_SB_FLAG_SHUTDOWN)) {
+ block->ops->free_block(inode->i_sb, block);
+ return;
+ }
+
+ BUG_ON(inode->i_ino < LOGFS_RESERVED_INOS);
+ page = inode_to_page(inode);
+ BUG_ON(!page); /* FIXME: Use emergency page */
+ logfs_put_write_page(page);
+}
+
+static int do_write_inode(struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ struct inode *master_inode = logfs_super(sb)->s_master_inode;
+ loff_t size = (inode->i_ino + 1) << inode->i_sb->s_blocksize_bits;
+ struct page *page;
+ int err;
+
+ BUG_ON(inode->i_ino == LOGFS_INO_MASTER);
+ /* FIXME: lock inode */
+
+ if (i_size_read(master_inode) < size)
+ i_size_write(master_inode, size);
+
+ /* TODO: Tell vfs this inode is clean now */
+
+ page = inode_to_page(inode);
+ if (!page)
+ return -ENOMEM;
+
+ /* FIXME: transaction is part of logfs_block now. Is that enough? */
+ err = logfs_write_buf(master_inode, page, 0);
+ logfs_put_write_page(page);
+ return err;
+}
+
+static void logfs_mod_segment_entry(struct super_block *sb, u32 segno,
+ int write,
+ void (*change_se)(struct logfs_segment_entry *, long),
+ long arg)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode;
+ struct page *page;
+ struct logfs_segment_entry *se;
+ pgoff_t page_no;
+ int child_no;
+
+ page_no = segno >> (sb->s_blocksize_bits - 3);
+ child_no = segno & ((sb->s_blocksize >> 3) - 1);
+
+ inode = super->s_segfile_inode;
+ page = logfs_get_write_page(inode, page_no, 0);
+ BUG_ON(!page); /* FIXME: We need some reserve page for this case */
+ if (!PageUptodate(page))
+ logfs_read_block(inode, page, WRITE);
+
+ if (write)
+ alloc_indirect_block(inode, page, 0);
+ se = kmap_atomic(page, KM_USER0);
+ change_se(se + child_no, arg);
+ if (write) {
+ logfs_set_alias(sb, logfs_block(page), child_no);
+ BUG_ON((int)be32_to_cpu(se[child_no].valid) > super->s_segsize);
+ }
+ kunmap_atomic(se, KM_USER0);
+
+ logfs_put_write_page(page);
+}
+
+static void __get_segment_entry(struct logfs_segment_entry *se, long _target)
+{
+ struct logfs_segment_entry *target = (void *)_target;
+
+ *target = *se;
+}
+
+void logfs_get_segment_entry(struct super_block *sb, u32 segno,
+ struct logfs_segment_entry *se)
+{
+ logfs_mod_segment_entry(sb, segno, 0, __get_segment_entry, (long)se);
+}
+
+static void __set_segment_used(struct logfs_segment_entry *se, long increment)
+{
+ u32 valid;
+
+ valid = be32_to_cpu(se->valid);
+ valid += increment;
+ se->valid = cpu_to_be32(valid);
+}
+
+void logfs_set_segment_used(struct super_block *sb, u64 ofs, int increment)
+{
+ struct logfs_super *super = logfs_super(sb);
+ u32 segno = ofs >> super->s_segshift;
+
+ if (!increment)
+ return;
+
+ logfs_mod_segment_entry(sb, segno, 1, __set_segment_used, increment);
+}
+
+static void __set_segment_erased(struct logfs_segment_entry *se, long ec_level)
+{
+ se->ec_level = cpu_to_be32(ec_level);
+}
+
+void logfs_set_segment_erased(struct super_block *sb, u32 segno, u32 ec,
+ gc_level_t gc_level)
+{
+ u32 ec_level = ec << 4 | (__force u8)gc_level;
+
+ logfs_mod_segment_entry(sb, segno, 1, __set_segment_erased, ec_level);
+}
+
+static void __set_segment_reserved(struct logfs_segment_entry *se, long ignore)
+{
+ se->valid = cpu_to_be32(RESERVED);
+}
+
+void logfs_set_segment_reserved(struct super_block *sb, u32 segno)
+{
+ logfs_mod_segment_entry(sb, segno, 1, __set_segment_reserved, 0);
+}
+
+static void __set_segment_unreserved(struct logfs_segment_entry *se,
+ long ec_level)
+{
+ se->valid = 0;
+ se->ec_level = cpu_to_be32(ec_level);
+}
+
+void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec)
+{
+ u32 ec_level = ec << 4;
+
+ logfs_mod_segment_entry(sb, segno, 1, __set_segment_unreserved,
+ ec_level);
+}
+
+int __logfs_write_inode(struct inode *inode, long flags)
+{
+ struct super_block *sb = inode->i_sb;
+ int ret;
+
+ logfs_get_wblocks(sb, NULL, flags & WF_LOCK);
+ ret = do_write_inode(inode);
+ logfs_put_wblocks(sb, NULL, flags & WF_LOCK);
+ return ret;
+}
+
+static int do_delete_inode(struct inode *inode)
+{
+ struct super_block *sb = inode->i_sb;
+ struct inode *master_inode = logfs_super(sb)->s_master_inode;
+ struct page *page;
+ int ret;
+
+ page = logfs_get_write_page(master_inode, inode->i_ino, 0);
+ if (!page)
+ return -ENOMEM;
+
+ move_inode_to_page(page, inode);
+
+ logfs_get_wblocks(sb, page, 1);
+ ret = __logfs_delete(master_inode, page);
+ logfs_put_wblocks(sb, page, 1);
+
+ logfs_put_write_page(page);
+ return ret;
+}
+
+/*
+ * ZOMBIE inodes have already been deleted before and should remain dead,
+ * if it weren't for valid checking. No need to kill them again here.
+ */
+void logfs_delete_inode(struct inode *inode)
+{
+ struct logfs_inode *li = logfs_inode(inode);
+
+ if (!(li->li_flags & LOGFS_IF_ZOMBIE)) {
+ li->li_flags |= LOGFS_IF_ZOMBIE;
+ if (i_size_read(inode) > 0)
+ logfs_truncate(inode, 0);
+ do_delete_inode(inode);
+ }
+ truncate_inode_pages(&inode->i_data, 0);
+ clear_inode(inode);
+}
+
+void btree_write_block(struct logfs_block *block)
+{
+ struct inode *inode;
+ struct page *page;
+ int err, cookie;
+
+ inode = logfs_safe_iget(block->sb, block->ino, &cookie);
+ page = logfs_get_write_page(inode, block->bix, block->level);
+
+ err = logfs_readpage_nolock(page);
+ BUG_ON(err);
+ BUG_ON(!PagePrivate(page));
+ BUG_ON(logfs_block(page) != block);
+ err = __logfs_write_buf(inode, page, 0);
+ BUG_ON(err);
+ BUG_ON(PagePrivate(page) || page->private);
+
+ logfs_put_write_page(page);
+ logfs_safe_iput(inode, cookie);
+}
+
+/**
+ * logfs_inode_write - write inode or dentry objects
+ *
+ * @inode: parent inode (ifile or directory)
+ * @buf: object to write (inode or dentry)
+ * @n: object size
+ * @_pos: object number (file position in blocks/objects)
+ * @flags: write flags
+ * @lock: 0 if write lock is already taken, 1 otherwise
+ * @shadow_tree: shadow below this inode
+ *
+ * FIXME: All caller of this put a 200-300 byte variable on the stack,
+ * only to call here and do a memcpy from that stack variable. A good
+ * example of wasted performance and stack space.
+ */
+int logfs_inode_write(struct inode *inode, const void *buf, size_t count,
+ loff_t bix, long flags, struct shadow_tree *shadow_tree)
+{
+ loff_t pos = bix << inode->i_sb->s_blocksize_bits;
+ int err;
+ struct page *page;
+ void *pagebuf;
+
+ BUG_ON(pos & (LOGFS_BLOCKSIZE-1));
+ BUG_ON(count > LOGFS_BLOCKSIZE);
+ page = logfs_get_write_page(inode, bix, 0);
+ if (!page)
+ return -ENOMEM;
+
+ pagebuf = kmap_atomic(page, KM_USER0);
+ memcpy(pagebuf, buf, count);
+ flush_dcache_page(page);
+ kunmap_atomic(pagebuf, KM_USER0);
+
+ if (i_size_read(inode) < pos + LOGFS_BLOCKSIZE)
+ i_size_write(inode, pos + LOGFS_BLOCKSIZE);
+
+ err = logfs_write_buf(inode, page, flags);
+ logfs_put_write_page(page);
+ return err;
+}
+
+int logfs_open_segfile(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *inode;
+
+ inode = logfs_read_meta_inode(sb, LOGFS_INO_SEGFILE);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+ super->s_segfile_inode = inode;
+ return 0;
+}
+
+int logfs_init_rw(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int min_fill = 3 * super->s_no_blocks;
+
+ INIT_LIST_HEAD(&super->s_object_alias);
+ mutex_init(&super->s_write_mutex);
+ super->s_block_pool = mempool_create_kmalloc_pool(min_fill,
+ sizeof(struct logfs_block));
+ super->s_shadow_pool = mempool_create_kmalloc_pool(min_fill,
+ sizeof(struct logfs_shadow));
+ return 0;
+}
+
+void logfs_cleanup_rw(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ destroy_meta_inode(super->s_segfile_inode);
+ if (super->s_block_pool)
+ mempool_destroy(super->s_block_pool);
+ if (super->s_shadow_pool)
+ mempool_destroy(super->s_shadow_pool);
+}
diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c
new file mode 100644
index 000000000000..1a14f9910d55
--- /dev/null
+++ b/fs/logfs/segment.c
@@ -0,0 +1,927 @@
+/*
+ * fs/logfs/segment.c - Handling the Object Store
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ *
+ * Object store or ostore makes up the complete device with exception of
+ * the superblock and journal areas. Apart from its own metadata it stores
+ * three kinds of objects: inodes, dentries and blocks, both data and indirect.
+ */
+#include "logfs.h"
+
+static int logfs_mark_segment_bad(struct super_block *sb, u32 segno)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct btree_head32 *head = &super->s_reserved_segments;
+ int err;
+
+ err = btree_insert32(head, segno, (void *)1, GFP_NOFS);
+ if (err)
+ return err;
+ logfs_super(sb)->s_bad_segments++;
+ /* FIXME: write to journal */
+ return 0;
+}
+
+int logfs_erase_segment(struct super_block *sb, u32 segno, int ensure_erase)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ super->s_gec++;
+
+ return super->s_devops->erase(sb, (u64)segno << super->s_segshift,
+ super->s_segsize, ensure_erase);
+}
+
+static s64 logfs_get_free_bytes(struct logfs_area *area, size_t bytes)
+{
+ s32 ofs;
+
+ logfs_open_area(area, bytes);
+
+ ofs = area->a_used_bytes;
+ area->a_used_bytes += bytes;
+ BUG_ON(area->a_used_bytes >= logfs_super(area->a_sb)->s_segsize);
+
+ return dev_ofs(area->a_sb, area->a_segno, ofs);
+}
+
+static struct page *get_mapping_page(struct super_block *sb, pgoff_t index,
+ int use_filler)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ filler_t *filler = super->s_devops->readpage;
+ struct page *page;
+
+ BUG_ON(mapping_gfp_mask(mapping) & __GFP_FS);
+ if (use_filler)
+ page = read_cache_page(mapping, index, filler, sb);
+ else {
+ page = find_or_create_page(mapping, index, GFP_NOFS);
+ unlock_page(page);
+ }
+ return page;
+}
+
+void __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len,
+ int use_filler)
+{
+ pgoff_t index = ofs >> PAGE_SHIFT;
+ struct page *page;
+ long offset = ofs & (PAGE_SIZE-1);
+ long copylen;
+
+ /* Only logfs_wbuf_recover may use len==0 */
+ BUG_ON(!len && !use_filler);
+ do {
+ copylen = min((ulong)len, PAGE_SIZE - offset);
+
+ page = get_mapping_page(area->a_sb, index, use_filler);
+ SetPageUptodate(page);
+ BUG_ON(!page); /* FIXME: reserve a pool */
+ memcpy(page_address(page) + offset, buf, copylen);
+ SetPagePrivate(page);
+ page_cache_release(page);
+
+ buf += copylen;
+ len -= copylen;
+ offset = 0;
+ index++;
+ } while (len);
+}
+
+/*
+ * bdev_writeseg will write full pages. Memset the tail to prevent data leaks.
+ */
+static void pad_wbuf(struct logfs_area *area, int final)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct page *page;
+ u64 ofs = dev_ofs(sb, area->a_segno, area->a_used_bytes);
+ pgoff_t index = ofs >> PAGE_SHIFT;
+ long offset = ofs & (PAGE_SIZE-1);
+ u32 len = PAGE_SIZE - offset;
+
+ if (len == PAGE_SIZE) {
+ /* The math in this function can surely use some love */
+ len = 0;
+ }
+ if (len) {
+ BUG_ON(area->a_used_bytes >= super->s_segsize);
+
+ page = get_mapping_page(area->a_sb, index, 0);
+ BUG_ON(!page); /* FIXME: reserve a pool */
+ memset(page_address(page) + offset, 0xff, len);
+ SetPagePrivate(page);
+ page_cache_release(page);
+ }
+
+ if (!final)
+ return;
+
+ area->a_used_bytes += len;
+ for ( ; area->a_used_bytes < super->s_segsize;
+ area->a_used_bytes += PAGE_SIZE) {
+ /* Memset another page */
+ index++;
+ page = get_mapping_page(area->a_sb, index, 0);
+ BUG_ON(!page); /* FIXME: reserve a pool */
+ memset(page_address(page), 0xff, PAGE_SIZE);
+ SetPagePrivate(page);
+ page_cache_release(page);
+ }
+}
+
+/*
+ * We have to be careful with the alias tree. Since lookup is done by bix,
+ * it needs to be normalized, so 14, 15, 16, etc. all match when dealing with
+ * indirect blocks. So always use it through accessor functions.
+ */
+static void *alias_tree_lookup(struct super_block *sb, u64 ino, u64 bix,
+ level_t level)
+{
+ struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree;
+ pgoff_t index = logfs_pack_index(bix, level);
+
+ return btree_lookup128(head, ino, index);
+}
+
+static int alias_tree_insert(struct super_block *sb, u64 ino, u64 bix,
+ level_t level, void *val)
+{
+ struct btree_head128 *head = &logfs_super(sb)->s_object_alias_tree;
+ pgoff_t index = logfs_pack_index(bix, level);
+
+ return btree_insert128(head, ino, index, val, GFP_NOFS);
+}
+
+static int btree_write_alias(struct super_block *sb, struct logfs_block *block,
+ write_alias_t *write_one_alias)
+{
+ struct object_alias_item *item;
+ int err;
+
+ list_for_each_entry(item, &block->item_list, list) {
+ err = write_alias_journal(sb, block->ino, block->bix,
+ block->level, item->child_no, item->val);
+ if (err)
+ return err;
+ }
+ return 0;
+}
+
+static gc_level_t btree_block_level(struct logfs_block *block)
+{
+ return expand_level(block->ino, block->level);
+}
+
+static struct logfs_block_ops btree_block_ops = {
+ .write_block = btree_write_block,
+ .block_level = btree_block_level,
+ .free_block = __free_block,
+ .write_alias = btree_write_alias,
+};
+
+int logfs_load_object_aliases(struct super_block *sb,
+ struct logfs_obj_alias *oa, int count)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_block *block;
+ struct object_alias_item *item;
+ u64 ino, bix;
+ level_t level;
+ int i, err;
+
+ super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS;
+ count /= sizeof(*oa);
+ for (i = 0; i < count; i++) {
+ item = mempool_alloc(super->s_alias_pool, GFP_NOFS);
+ if (!item)
+ return -ENOMEM;
+ memset(item, 0, sizeof(*item));
+
+ super->s_no_object_aliases++;
+ item->val = oa[i].val;
+ item->child_no = be16_to_cpu(oa[i].child_no);
+
+ ino = be64_to_cpu(oa[i].ino);
+ bix = be64_to_cpu(oa[i].bix);
+ level = LEVEL(oa[i].level);
+
+ log_aliases("logfs_load_object_aliases(%llx, %llx, %x, %x) %llx\n",
+ ino, bix, level, item->child_no,
+ be64_to_cpu(item->val));
+ block = alias_tree_lookup(sb, ino, bix, level);
+ if (!block) {
+ block = __alloc_block(sb, ino, bix, level);
+ block->ops = &btree_block_ops;
+ err = alias_tree_insert(sb, ino, bix, level, block);
+ BUG_ON(err); /* mempool empty */
+ }
+ if (test_and_set_bit(item->child_no, block->alias_map)) {
+ printk(KERN_ERR"LogFS: Alias collision detected\n");
+ return -EIO;
+ }
+ list_move_tail(&block->alias_list, &super->s_object_alias);
+ list_add(&item->list, &block->item_list);
+ }
+ return 0;
+}
+
+static void kill_alias(void *_block, unsigned long ignore0,
+ u64 ignore1, u64 ignore2, size_t ignore3)
+{
+ struct logfs_block *block = _block;
+ struct super_block *sb = block->sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct object_alias_item *item;
+
+ while (!list_empty(&block->item_list)) {
+ item = list_entry(block->item_list.next, typeof(*item), list);
+ list_del(&item->list);
+ mempool_free(item, super->s_alias_pool);
+ }
+ block->ops->free_block(sb, block);
+}
+
+static int obj_type(struct inode *inode, level_t level)
+{
+ if (level == 0) {
+ if (S_ISDIR(inode->i_mode))
+ return OBJ_DENTRY;
+ if (inode->i_ino == LOGFS_INO_MASTER)
+ return OBJ_INODE;
+ }
+ return OBJ_BLOCK;
+}
+
+static int obj_len(struct super_block *sb, int obj_type)
+{
+ switch (obj_type) {
+ case OBJ_DENTRY:
+ return sizeof(struct logfs_disk_dentry);
+ case OBJ_INODE:
+ return sizeof(struct logfs_disk_inode);
+ case OBJ_BLOCK:
+ return sb->s_blocksize;
+ default:
+ BUG();
+ }
+}
+
+static int __logfs_segment_write(struct inode *inode, void *buf,
+ struct logfs_shadow *shadow, int type, int len, int compr)
+{
+ struct logfs_area *area;
+ struct super_block *sb = inode->i_sb;
+ s64 ofs;
+ struct logfs_object_header h;
+ int acc_len;
+
+ if (shadow->gc_level == 0)
+ acc_len = len;
+ else
+ acc_len = obj_len(sb, type);
+
+ area = get_area(sb, shadow->gc_level);
+ ofs = logfs_get_free_bytes(area, len + LOGFS_OBJECT_HEADERSIZE);
+ LOGFS_BUG_ON(ofs <= 0, sb);
+ /*
+ * Order is important. logfs_get_free_bytes(), by modifying the
+ * segment file, may modify the content of the very page we're about
+ * to write now. Which is fine, as long as the calculated crc and
+ * written data still match. So do the modifications _before_
+ * calculating the crc.
+ */
+
+ h.len = cpu_to_be16(len);
+ h.type = type;
+ h.compr = compr;
+ h.ino = cpu_to_be64(inode->i_ino);
+ h.bix = cpu_to_be64(shadow->bix);
+ h.crc = logfs_crc32(&h, sizeof(h) - 4, 4);
+ h.data_crc = logfs_crc32(buf, len, 0);
+
+ logfs_buf_write(area, ofs, &h, sizeof(h));
+ logfs_buf_write(area, ofs + LOGFS_OBJECT_HEADERSIZE, buf, len);
+
+ shadow->new_ofs = ofs;
+ shadow->new_len = acc_len + LOGFS_OBJECT_HEADERSIZE;
+
+ return 0;
+}
+
+static s64 logfs_segment_write_compress(struct inode *inode, void *buf,
+ struct logfs_shadow *shadow, int type, int len)
+{
+ struct super_block *sb = inode->i_sb;
+ void *compressor_buf = logfs_super(sb)->s_compressed_je;
+ ssize_t compr_len;
+ int ret;
+
+ mutex_lock(&logfs_super(sb)->s_journal_mutex);
+ compr_len = logfs_compress(buf, compressor_buf, len, len);
+
+ if (compr_len >= 0) {
+ ret = __logfs_segment_write(inode, compressor_buf, shadow,
+ type, compr_len, COMPR_ZLIB);
+ } else {
+ ret = __logfs_segment_write(inode, buf, shadow, type, len,
+ COMPR_NONE);
+ }
+ mutex_unlock(&logfs_super(sb)->s_journal_mutex);
+ return ret;
+}
+
+/**
+ * logfs_segment_write - write data block to object store
+ * @inode: inode containing data
+ *
+ * Returns an errno or zero.
+ */
+int logfs_segment_write(struct inode *inode, struct page *page,
+ struct logfs_shadow *shadow)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_super *super = logfs_super(sb);
+ int do_compress, type, len;
+ int ret;
+ void *buf;
+
+ super->s_flags |= LOGFS_SB_FLAG_DIRTY;
+ BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN);
+ do_compress = logfs_inode(inode)->li_flags & LOGFS_IF_COMPRESSED;
+ if (shadow->gc_level != 0) {
+ /* temporarily disable compression for indirect blocks */
+ do_compress = 0;
+ }
+
+ type = obj_type(inode, shrink_level(shadow->gc_level));
+ len = obj_len(sb, type);
+ buf = kmap(page);
+ if (do_compress)
+ ret = logfs_segment_write_compress(inode, buf, shadow, type,
+ len);
+ else
+ ret = __logfs_segment_write(inode, buf, shadow, type, len,
+ COMPR_NONE);
+ kunmap(page);
+
+ log_segment("logfs_segment_write(%llx, %llx, %x) %llx->%llx %x->%x\n",
+ shadow->ino, shadow->bix, shadow->gc_level,
+ shadow->old_ofs, shadow->new_ofs,
+ shadow->old_len, shadow->new_len);
+ /* this BUG_ON did catch a locking bug. useful */
+ BUG_ON(!(shadow->new_ofs & (super->s_segsize - 1)));
+ return ret;
+}
+
+int wbuf_read(struct super_block *sb, u64 ofs, size_t len, void *buf)
+{
+ pgoff_t index = ofs >> PAGE_SHIFT;
+ struct page *page;
+ long offset = ofs & (PAGE_SIZE-1);
+ long copylen;
+
+ while (len) {
+ copylen = min((ulong)len, PAGE_SIZE - offset);
+
+ page = get_mapping_page(sb, index, 1);
+ if (IS_ERR(page))
+ return PTR_ERR(page);
+ memcpy(buf, page_address(page) + offset, copylen);
+ page_cache_release(page);
+
+ buf += copylen;
+ len -= copylen;
+ offset = 0;
+ index++;
+ }
+ return 0;
+}
+
+/*
+ * The "position" of indirect blocks is ambiguous. It can be the position
+ * of any data block somewhere behind this indirect block. So we need to
+ * normalize the positions through logfs_block_mask() before comparing.
+ */
+static int check_pos(struct super_block *sb, u64 pos1, u64 pos2, level_t level)
+{
+ return (pos1 & logfs_block_mask(sb, level)) !=
+ (pos2 & logfs_block_mask(sb, level));
+}
+
+#if 0
+static int read_seg_header(struct super_block *sb, u64 ofs,
+ struct logfs_segment_header *sh)
+{
+ __be32 crc;
+ int err;
+
+ err = wbuf_read(sb, ofs, sizeof(*sh), sh);
+ if (err)
+ return err;
+ crc = logfs_crc32(sh, sizeof(*sh), 4);
+ if (crc != sh->crc) {
+ printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, "
+ "got %x\n", ofs, be32_to_cpu(sh->crc),
+ be32_to_cpu(crc));
+ return -EIO;
+ }
+ return 0;
+}
+#endif
+
+static int read_obj_header(struct super_block *sb, u64 ofs,
+ struct logfs_object_header *oh)
+{
+ __be32 crc;
+ int err;
+
+ err = wbuf_read(sb, ofs, sizeof(*oh), oh);
+ if (err)
+ return err;
+ crc = logfs_crc32(oh, sizeof(*oh) - 4, 4);
+ if (crc != oh->crc) {
+ printk(KERN_ERR"LOGFS: header crc error at %llx: expected %x, "
+ "got %x\n", ofs, be32_to_cpu(oh->crc),
+ be32_to_cpu(crc));
+ return -EIO;
+ }
+ return 0;
+}
+
+static void move_btree_to_page(struct inode *inode, struct page *page,
+ __be64 *data)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct btree_head128 *head = &super->s_object_alias_tree;
+ struct logfs_block *block;
+ struct object_alias_item *item, *next;
+
+ if (!(super->s_flags & LOGFS_SB_FLAG_OBJ_ALIAS))
+ return;
+
+ block = btree_remove128(head, inode->i_ino, page->index);
+ if (!block)
+ return;
+
+ log_blockmove("move_btree_to_page(%llx, %llx, %x)\n",
+ block->ino, block->bix, block->level);
+ list_for_each_entry_safe(item, next, &block->item_list, list) {
+ data[item->child_no] = item->val;
+ list_del(&item->list);
+ mempool_free(item, super->s_alias_pool);
+ }
+ block->page = page;
+ SetPagePrivate(page);
+ page->private = (unsigned long)block;
+ block->ops = &indirect_block_ops;
+ initialize_block_counters(page, block, data, 0);
+}
+
+/*
+ * This silences a false, yet annoying gcc warning. I hate it when my editor
+ * jumps into bitops.h each time I recompile this file.
+ * TODO: Complain to gcc folks about this and upgrade compiler.
+ */
+static unsigned long fnb(const unsigned long *addr,
+ unsigned long size, unsigned long offset)
+{
+ return find_next_bit(addr, size, offset);
+}
+
+void move_page_to_btree(struct page *page)
+{
+ struct logfs_block *block = logfs_block(page);
+ struct super_block *sb = block->sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct object_alias_item *item;
+ unsigned long pos;
+ __be64 *child;
+ int err;
+
+ if (super->s_flags & LOGFS_SB_FLAG_SHUTDOWN) {
+ block->ops->free_block(sb, block);
+ return;
+ }
+ log_blockmove("move_page_to_btree(%llx, %llx, %x)\n",
+ block->ino, block->bix, block->level);
+ super->s_flags |= LOGFS_SB_FLAG_OBJ_ALIAS;
+
+ for (pos = 0; ; pos++) {
+ pos = fnb(block->alias_map, LOGFS_BLOCK_FACTOR, pos);
+ if (pos >= LOGFS_BLOCK_FACTOR)
+ break;
+
+ item = mempool_alloc(super->s_alias_pool, GFP_NOFS);
+ BUG_ON(!item); /* mempool empty */
+ memset(item, 0, sizeof(*item));
+
+ child = kmap_atomic(page, KM_USER0);
+ item->val = child[pos];
+ kunmap_atomic(child, KM_USER0);
+ item->child_no = pos;
+ list_add(&item->list, &block->item_list);
+ }
+ block->page = NULL;
+ ClearPagePrivate(page);
+ page->private = 0;
+ block->ops = &btree_block_ops;
+ err = alias_tree_insert(block->sb, block->ino, block->bix, block->level,
+ block);
+ BUG_ON(err); /* mempool empty */
+ ClearPageUptodate(page);
+}
+
+static int __logfs_segment_read(struct inode *inode, void *buf,
+ u64 ofs, u64 bix, level_t level)
+{
+ struct super_block *sb = inode->i_sb;
+ void *compressor_buf = logfs_super(sb)->s_compressed_je;
+ struct logfs_object_header oh;
+ __be32 crc;
+ u16 len;
+ int err, block_len;
+
+ block_len = obj_len(sb, obj_type(inode, level));
+ err = read_obj_header(sb, ofs, &oh);
+ if (err)
+ goto out_err;
+
+ err = -EIO;
+ if (be64_to_cpu(oh.ino) != inode->i_ino
+ || check_pos(sb, be64_to_cpu(oh.bix), bix, level)) {
+ printk(KERN_ERR"LOGFS: (ino, bix) don't match at %llx: "
+ "expected (%lx, %llx), got (%llx, %llx)\n",
+ ofs, inode->i_ino, bix,
+ be64_to_cpu(oh.ino), be64_to_cpu(oh.bix));
+ goto out_err;
+ }
+
+ len = be16_to_cpu(oh.len);
+
+ switch (oh.compr) {
+ case COMPR_NONE:
+ err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len, buf);
+ if (err)
+ goto out_err;
+ crc = logfs_crc32(buf, len, 0);
+ if (crc != oh.data_crc) {
+ printk(KERN_ERR"LOGFS: uncompressed data crc error at "
+ "%llx: expected %x, got %x\n", ofs,
+ be32_to_cpu(oh.data_crc),
+ be32_to_cpu(crc));
+ goto out_err;
+ }
+ break;
+ case COMPR_ZLIB:
+ mutex_lock(&logfs_super(sb)->s_journal_mutex);
+ err = wbuf_read(sb, ofs + LOGFS_OBJECT_HEADERSIZE, len,
+ compressor_buf);
+ if (err) {
+ mutex_unlock(&logfs_super(sb)->s_journal_mutex);
+ goto out_err;
+ }
+ crc = logfs_crc32(compressor_buf, len, 0);
+ if (crc != oh.data_crc) {
+ printk(KERN_ERR"LOGFS: compressed data crc error at "
+ "%llx: expected %x, got %x\n", ofs,
+ be32_to_cpu(oh.data_crc),
+ be32_to_cpu(crc));
+ mutex_unlock(&logfs_super(sb)->s_journal_mutex);
+ goto out_err;
+ }
+ err = logfs_uncompress(compressor_buf, buf, len, block_len);
+ mutex_unlock(&logfs_super(sb)->s_journal_mutex);
+ if (err) {
+ printk(KERN_ERR"LOGFS: uncompress error at %llx\n", ofs);
+ goto out_err;
+ }
+ break;
+ default:
+ LOGFS_BUG(sb);
+ err = -EIO;
+ goto out_err;
+ }
+ return 0;
+
+out_err:
+ logfs_set_ro(sb);
+ printk(KERN_ERR"LOGFS: device is read-only now\n");
+ LOGFS_BUG(sb);
+ return err;
+}
+
+/**
+ * logfs_segment_read - read data block from object store
+ * @inode: inode containing data
+ * @buf: data buffer
+ * @ofs: physical data offset
+ * @bix: block index
+ * @level: block level
+ *
+ * Returns 0 on success or a negative errno.
+ */
+int logfs_segment_read(struct inode *inode, struct page *page,
+ u64 ofs, u64 bix, level_t level)
+{
+ int err;
+ void *buf;
+
+ if (PageUptodate(page))
+ return 0;
+
+ ofs &= ~LOGFS_FULLY_POPULATED;
+
+ buf = kmap(page);
+ err = __logfs_segment_read(inode, buf, ofs, bix, level);
+ if (!err) {
+ move_btree_to_page(inode, page, buf);
+ SetPageUptodate(page);
+ }
+ kunmap(page);
+ log_segment("logfs_segment_read(%lx, %llx, %x) %llx (%d)\n",
+ inode->i_ino, bix, level, ofs, err);
+ return err;
+}
+
+int logfs_segment_delete(struct inode *inode, struct logfs_shadow *shadow)
+{
+ struct super_block *sb = inode->i_sb;
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_object_header h;
+ u16 len;
+ int err;
+
+ super->s_flags |= LOGFS_SB_FLAG_DIRTY;
+ BUG_ON(super->s_flags & LOGFS_SB_FLAG_SHUTDOWN);
+ BUG_ON(shadow->old_ofs & LOGFS_FULLY_POPULATED);
+ if (!shadow->old_ofs)
+ return 0;
+
+ log_segment("logfs_segment_delete(%llx, %llx, %x) %llx->%llx %x->%x\n",
+ shadow->ino, shadow->bix, shadow->gc_level,
+ shadow->old_ofs, shadow->new_ofs,
+ shadow->old_len, shadow->new_len);
+ err = read_obj_header(sb, shadow->old_ofs, &h);
+ LOGFS_BUG_ON(err, sb);
+ LOGFS_BUG_ON(be64_to_cpu(h.ino) != inode->i_ino, sb);
+ LOGFS_BUG_ON(check_pos(sb, shadow->bix, be64_to_cpu(h.bix),
+ shrink_level(shadow->gc_level)), sb);
+
+ if (shadow->gc_level == 0)
+ len = be16_to_cpu(h.len);
+ else
+ len = obj_len(sb, h.type);
+ shadow->old_len = len + sizeof(h);
+ return 0;
+}
+
+static void freeseg(struct super_block *sb, u32 segno)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping = super->s_mapping_inode->i_mapping;
+ struct page *page;
+ u64 ofs, start, end;
+
+ start = dev_ofs(sb, segno, 0);
+ end = dev_ofs(sb, segno + 1, 0);
+ for (ofs = start; ofs < end; ofs += PAGE_SIZE) {
+ page = find_get_page(mapping, ofs >> PAGE_SHIFT);
+ if (!page)
+ continue;
+ ClearPagePrivate(page);
+ page_cache_release(page);
+ }
+}
+
+int logfs_open_area(struct logfs_area *area, size_t bytes)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_super *super = logfs_super(sb);
+ int err, closed = 0;
+
+ if (area->a_is_open && area->a_used_bytes + bytes <= super->s_segsize)
+ return 0;
+
+ if (area->a_is_open) {
+ u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes);
+ u32 len = super->s_segsize - area->a_written_bytes;
+
+ log_gc("logfs_close_area(%x)\n", area->a_segno);
+ pad_wbuf(area, 1);
+ super->s_devops->writeseg(area->a_sb, ofs, len);
+ freeseg(sb, area->a_segno);
+ closed = 1;
+ }
+
+ area->a_used_bytes = 0;
+ area->a_written_bytes = 0;
+again:
+ area->a_ops->get_free_segment(area);
+ area->a_ops->get_erase_count(area);
+
+ log_gc("logfs_open_area(%x, %x)\n", area->a_segno, area->a_level);
+ err = area->a_ops->erase_segment(area);
+ if (err) {
+ printk(KERN_WARNING "LogFS: Error erasing segment %x\n",
+ area->a_segno);
+ logfs_mark_segment_bad(sb, area->a_segno);
+ goto again;
+ }
+ area->a_is_open = 1;
+ return closed;
+}
+
+void logfs_sync_area(struct logfs_area *area)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_super *super = logfs_super(sb);
+ u64 ofs = dev_ofs(sb, area->a_segno, area->a_written_bytes);
+ u32 len = (area->a_used_bytes - area->a_written_bytes);
+
+ if (super->s_writesize)
+ len &= ~(super->s_writesize - 1);
+ if (len == 0)
+ return;
+ pad_wbuf(area, 0);
+ super->s_devops->writeseg(sb, ofs, len);
+ area->a_written_bytes += len;
+}
+
+void logfs_sync_segments(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i;
+
+ for_each_area(i)
+ logfs_sync_area(super->s_area[i]);
+}
+
+/*
+ * Pick a free segment to be used for this area. Effectively takes a
+ * candidate from the free list (not really a candidate anymore).
+ */
+static void ostore_get_free_segment(struct logfs_area *area)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_super *super = logfs_super(sb);
+
+ if (super->s_free_list.count == 0) {
+ printk(KERN_ERR"LOGFS: ran out of free segments\n");
+ LOGFS_BUG(sb);
+ }
+
+ area->a_segno = get_best_cand(sb, &super->s_free_list, NULL);
+}
+
+static void ostore_get_erase_count(struct logfs_area *area)
+{
+ struct logfs_segment_entry se;
+ u32 ec_level;
+
+ logfs_get_segment_entry(area->a_sb, area->a_segno, &se);
+ BUG_ON(se.ec_level == cpu_to_be32(BADSEG) ||
+ se.valid == cpu_to_be32(RESERVED));
+
+ ec_level = be32_to_cpu(se.ec_level);
+ area->a_erase_count = (ec_level >> 4) + 1;
+}
+
+static int ostore_erase_segment(struct logfs_area *area)
+{
+ struct super_block *sb = area->a_sb;
+ struct logfs_segment_header sh;
+ u64 ofs;
+ int err;
+
+ err = logfs_erase_segment(sb, area->a_segno, 0);
+ if (err)
+ return err;
+
+ sh.pad = 0;
+ sh.type = SEG_OSTORE;
+ sh.level = (__force u8)area->a_level;
+ sh.segno = cpu_to_be32(area->a_segno);
+ sh.ec = cpu_to_be32(area->a_erase_count);
+ sh.gec = cpu_to_be64(logfs_super(sb)->s_gec);
+ sh.crc = logfs_crc32(&sh, sizeof(sh), 4);
+
+ logfs_set_segment_erased(sb, area->a_segno, area->a_erase_count,
+ area->a_level);
+
+ ofs = dev_ofs(sb, area->a_segno, 0);
+ area->a_used_bytes = sizeof(sh);
+ logfs_buf_write(area, ofs, &sh, sizeof(sh));
+ return 0;
+}
+
+static const struct logfs_area_ops ostore_area_ops = {
+ .get_free_segment = ostore_get_free_segment,
+ .get_erase_count = ostore_get_erase_count,
+ .erase_segment = ostore_erase_segment,
+};
+
+static void free_area(struct logfs_area *area)
+{
+ if (area)
+ freeseg(area->a_sb, area->a_segno);
+ kfree(area);
+}
+
+static struct logfs_area *alloc_area(struct super_block *sb)
+{
+ struct logfs_area *area;
+
+ area = kzalloc(sizeof(*area), GFP_KERNEL);
+ if (!area)
+ return NULL;
+
+ area->a_sb = sb;
+ return area;
+}
+
+static void map_invalidatepage(struct page *page, unsigned long l)
+{
+ BUG();
+}
+
+static int map_releasepage(struct page *page, gfp_t g)
+{
+ /* Don't release these pages */
+ return 0;
+}
+
+static const struct address_space_operations mapping_aops = {
+ .invalidatepage = map_invalidatepage,
+ .releasepage = map_releasepage,
+ .set_page_dirty = __set_page_dirty_nobuffers,
+};
+
+int logfs_init_mapping(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct address_space *mapping;
+ struct inode *inode;
+
+ inode = logfs_new_meta_inode(sb, LOGFS_INO_MAPPING);
+ if (IS_ERR(inode))
+ return PTR_ERR(inode);
+ super->s_mapping_inode = inode;
+ mapping = inode->i_mapping;
+ mapping->a_ops = &mapping_aops;
+ /* Would it be possible to use __GFP_HIGHMEM as well? */
+ mapping_set_gfp_mask(mapping, GFP_NOFS);
+ return 0;
+}
+
+int logfs_init_areas(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i = -1;
+
+ super->s_alias_pool = mempool_create_kmalloc_pool(600,
+ sizeof(struct object_alias_item));
+ if (!super->s_alias_pool)
+ return -ENOMEM;
+
+ super->s_journal_area = alloc_area(sb);
+ if (!super->s_journal_area)
+ goto err;
+
+ for_each_area(i) {
+ super->s_area[i] = alloc_area(sb);
+ if (!super->s_area[i])
+ goto err;
+ super->s_area[i]->a_level = GC_LEVEL(i);
+ super->s_area[i]->a_ops = &ostore_area_ops;
+ }
+ btree_init_mempool128(&super->s_object_alias_tree,
+ super->s_btree_pool);
+ return 0;
+
+err:
+ for (i--; i >= 0; i--)
+ free_area(super->s_area[i]);
+ free_area(super->s_journal_area);
+ mempool_destroy(super->s_alias_pool);
+ return -ENOMEM;
+}
+
+void logfs_cleanup_areas(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int i;
+
+ btree_grim_visitor128(&super->s_object_alias_tree, 0, kill_alias);
+ for_each_area(i)
+ free_area(super->s_area[i]);
+ free_area(super->s_journal_area);
+ destroy_meta_inode(super->s_mapping_inode);
+}
diff --git a/fs/logfs/super.c b/fs/logfs/super.c
new file mode 100644
index 000000000000..c66beab78dee
--- /dev/null
+++ b/fs/logfs/super.c
@@ -0,0 +1,650 @@
+/*
+ * fs/logfs/super.c
+ *
+ * As should be obvious for Linux kernel code, license is GPLv2
+ *
+ * Copyright (c) 2005-2008 Joern Engel <joern@logfs.org>
+ *
+ * Generally contains mount/umount code and also serves as a dump area for
+ * any functions that don't fit elsewhere and neither justify a file of their
+ * own.
+ */
+#include "logfs.h"
+#include <linux/bio.h>
+#include <linux/mtd/mtd.h>
+#include <linux/statfs.h>
+#include <linux/buffer_head.h>
+
+static DEFINE_MUTEX(emergency_mutex);
+static struct page *emergency_page;
+
+struct page *emergency_read_begin(struct address_space *mapping, pgoff_t index)
+{
+ filler_t *filler = (filler_t *)mapping->a_ops->readpage;
+ struct page *page;
+ int err;
+
+ page = read_cache_page(mapping, index, filler, NULL);
+ if (page)
+ return page;
+
+ /* No more pages available, switch to emergency page */
+ printk(KERN_INFO"Logfs: Using emergency page\n");
+ mutex_lock(&emergency_mutex);
+ err = filler(NULL, emergency_page);
+ if (err) {
+ mutex_unlock(&emergency_mutex);
+ printk(KERN_EMERG"Logfs: Error reading emergency page\n");
+ return ERR_PTR(err);
+ }
+ return emergency_page;
+}
+
+void emergency_read_end(struct page *page)
+{
+ if (page == emergency_page)
+ mutex_unlock(&emergency_mutex);
+ else
+ page_cache_release(page);
+}
+
+static void dump_segfile(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_segment_entry se;
+ u32 segno;
+
+ for (segno = 0; segno < super->s_no_segs; segno++) {
+ logfs_get_segment_entry(sb, segno, &se);
+ printk("%3x: %6x %8x", segno, be32_to_cpu(se.ec_level),
+ be32_to_cpu(se.valid));
+ if (++segno < super->s_no_segs) {
+ logfs_get_segment_entry(sb, segno, &se);
+ printk(" %6x %8x", be32_to_cpu(se.ec_level),
+ be32_to_cpu(se.valid));
+ }
+ if (++segno < super->s_no_segs) {
+ logfs_get_segment_entry(sb, segno, &se);
+ printk(" %6x %8x", be32_to_cpu(se.ec_level),
+ be32_to_cpu(se.valid));
+ }
+ if (++segno < super->s_no_segs) {
+ logfs_get_segment_entry(sb, segno, &se);
+ printk(" %6x %8x", be32_to_cpu(se.ec_level),
+ be32_to_cpu(se.valid));
+ }
+ printk("\n");
+ }
+}
+
+/*
+ * logfs_crash_dump - dump debug information to device
+ *
+ * The LogFS superblock only occupies part of a segment. This function will
+ * write as much debug information as it can gather into the spare space.
+ */
+void logfs_crash_dump(struct super_block *sb)
+{
+ dump_segfile(sb);
+}
+
+/*
+ * TODO: move to lib/string.c
+ */
+/**
+ * memchr_inv - Find a character in an area of memory.
+ * @s: The memory area
+ * @c: The byte to search for
+ * @n: The size of the area.
+ *
+ * returns the address of the first character other than @c, or %NULL
+ * if the whole buffer contains just @c.
+ */
+void *memchr_inv(const void *s, int c, size_t n)
+{
+ const unsigned char *p = s;
+ while (n-- != 0)
+ if ((unsigned char)c != *p++)
+ return (void *)(p - 1);
+
+ return NULL;
+}
+
+/*
+ * FIXME: There should be a reserve for root, similar to ext2.
+ */
+int logfs_statfs(struct dentry *dentry, struct kstatfs *stats)
+{
+ struct super_block *sb = dentry->d_sb;
+ struct logfs_super *super = logfs_super(sb);
+
+ stats->f_type = LOGFS_MAGIC_U32;
+ stats->f_bsize = sb->s_blocksize;
+ stats->f_blocks = super->s_size >> LOGFS_BLOCK_BITS >> 3;
+ stats->f_bfree = super->s_free_bytes >> sb->s_blocksize_bits;
+ stats->f_bavail = super->s_free_bytes >> sb->s_blocksize_bits;
+ stats->f_files = 0;
+ stats->f_ffree = 0;
+ stats->f_namelen = LOGFS_MAX_NAMELEN;
+ return 0;
+}
+
+static int logfs_sb_set(struct super_block *sb, void *_super)
+{
+ struct logfs_super *super = _super;
+
+ sb->s_fs_info = super;
+ sb->s_mtd = super->s_mtd;
+ sb->s_bdev = super->s_bdev;
+ return 0;
+}
+
+static int logfs_sb_test(struct super_block *sb, void *_super)
+{
+ struct logfs_super *super = _super;
+ struct mtd_info *mtd = super->s_mtd;
+
+ if (mtd && sb->s_mtd == mtd)
+ return 1;
+ if (super->s_bdev && sb->s_bdev == super->s_bdev)
+ return 1;
+ return 0;
+}
+
+static void set_segment_header(struct logfs_segment_header *sh, u8 type,
+ u8 level, u32 segno, u32 ec)
+{
+ sh->pad = 0;
+ sh->type = type;
+ sh->level = level;
+ sh->segno = cpu_to_be32(segno);
+ sh->ec = cpu_to_be32(ec);
+ sh->gec = cpu_to_be64(segno);
+ sh->crc = logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4);
+}
+
+static void logfs_write_ds(struct super_block *sb, struct logfs_disk_super *ds,
+ u32 segno, u32 ec)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_segment_header *sh = &ds->ds_sh;
+ int i;
+
+ memset(ds, 0, sizeof(*ds));
+ set_segment_header(sh, SEG_SUPER, 0, segno, ec);
+
+ ds->ds_ifile_levels = super->s_ifile_levels;
+ ds->ds_iblock_levels = super->s_iblock_levels;
+ ds->ds_data_levels = super->s_data_levels; /* XXX: Remove */
+ ds->ds_segment_shift = super->s_segshift;
+ ds->ds_block_shift = sb->s_blocksize_bits;
+ ds->ds_write_shift = super->s_writeshift;
+ ds->ds_filesystem_size = cpu_to_be64(super->s_size);
+ ds->ds_segment_size = cpu_to_be32(super->s_segsize);
+ ds->ds_bad_seg_reserve = cpu_to_be32(super->s_bad_seg_reserve);
+ ds->ds_feature_incompat = cpu_to_be64(super->s_feature_incompat);
+ ds->ds_feature_ro_compat= cpu_to_be64(super->s_feature_ro_compat);
+ ds->ds_feature_compat = cpu_to_be64(super->s_feature_compat);
+ ds->ds_feature_flags = cpu_to_be64(super->s_feature_flags);
+ ds->ds_root_reserve = cpu_to_be64(super->s_root_reserve);
+ ds->ds_speed_reserve = cpu_to_be64(super->s_speed_reserve);
+ journal_for_each(i)
+ ds->ds_journal_seg[i] = cpu_to_be32(super->s_journal_seg[i]);
+ ds->ds_magic = cpu_to_be64(LOGFS_MAGIC);
+ ds->ds_crc = logfs_crc32(ds, sizeof(*ds),
+ LOGFS_SEGMENT_HEADERSIZE + 12);
+}
+
+static int write_one_sb(struct super_block *sb,
+ struct page *(*find_sb)(struct super_block *sb, u64 *ofs))
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_disk_super *ds;
+ struct logfs_segment_entry se;
+ struct page *page;
+ u64 ofs;
+ u32 ec, segno;
+ int err;
+
+ page = find_sb(sb, &ofs);
+ if (!page)
+ return -EIO;
+ ds = page_address(page);
+ segno = seg_no(sb, ofs);
+ logfs_get_segment_entry(sb, segno, &se);
+ ec = be32_to_cpu(se.ec_level) >> 4;
+ ec++;
+ logfs_set_segment_erased(sb, segno, ec, 0);
+ logfs_write_ds(sb, ds, segno, ec);
+ err = super->s_devops->write_sb(sb, page);
+ page_cache_release(page);
+ return err;
+}
+
+int logfs_write_sb(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int err;
+
+ /* First superblock */
+ err = write_one_sb(sb, super->s_devops->find_first_sb);
+ if (err)
+ return err;
+
+ /* Last superblock */
+ err = write_one_sb(sb, super->s_devops->find_last_sb);
+ if (err)
+ return err;
+ return 0;
+}
+
+static int ds_cmp(const void *ds0, const void *ds1)
+{
+ size_t len = sizeof(struct logfs_disk_super);
+
+ /* We know the segment headers differ, so ignore them */
+ len -= LOGFS_SEGMENT_HEADERSIZE;
+ ds0 += LOGFS_SEGMENT_HEADERSIZE;
+ ds1 += LOGFS_SEGMENT_HEADERSIZE;
+ return memcmp(ds0, ds1, len);
+}
+
+static int logfs_recover_sb(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct logfs_disk_super _ds0, *ds0 = &_ds0;
+ struct logfs_disk_super _ds1, *ds1 = &_ds1;
+ int err, valid0, valid1;
+
+ /* read first superblock */
+ err = wbuf_read(sb, super->s_sb_ofs[0], sizeof(*ds0), ds0);
+ if (err)
+ return err;
+ /* read last superblock */
+ err = wbuf_read(sb, super->s_sb_ofs[1], sizeof(*ds1), ds1);
+ if (err)
+ return err;
+ valid0 = logfs_check_ds(ds0) == 0;
+ valid1 = logfs_check_ds(ds1) == 0;
+
+ if (!valid0 && valid1) {
+ printk(KERN_INFO"First superblock is invalid - fixing.\n");
+ return write_one_sb(sb, super->s_devops->find_first_sb);
+ }
+ if (valid0 && !valid1) {
+ printk(KERN_INFO"Last superblock is invalid - fixing.\n");
+ return write_one_sb(sb, super->s_devops->find_last_sb);
+ }
+ if (valid0 && valid1 && ds_cmp(ds0, ds1)) {
+ printk(KERN_INFO"Superblocks don't match - fixing.\n");
+ return write_one_sb(sb, super->s_devops->find_last_sb);
+ }
+ /* If neither is valid now, something's wrong. Didn't we properly
+ * check them before?!? */
+ BUG_ON(!valid0 && !valid1);
+ return 0;
+}
+
+static int logfs_make_writeable(struct super_block *sb)
+{
+ int err;
+
+ /* Repair any broken superblock copies */
+ err = logfs_recover_sb(sb);
+ if (err)
+ return err;
+
+ /* Check areas for trailing unaccounted data */
+ err = logfs_check_areas(sb);
+ if (err)
+ return err;
+
+ err = logfs_open_segfile(sb);
+ if (err)
+ return err;
+
+ /* Do one GC pass before any data gets dirtied */
+ logfs_gc_pass(sb);
+
+ /* after all initializations are done, replay the journal
+ * for rw-mounts, if necessary */
+ err = logfs_replay_journal(sb);
+ if (err)
+ return err;
+
+ return 0;
+}
+
+static int logfs_get_sb_final(struct super_block *sb, struct vfsmount *mnt)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct inode *rootdir;
+ int err;
+
+ /* root dir */
+ rootdir = logfs_iget(sb, LOGFS_INO_ROOT);
+ if (IS_ERR(rootdir))
+ goto fail;
+
+ sb->s_root = d_alloc_root(rootdir);
+ if (!sb->s_root)
+ goto fail;
+
+ super->s_erase_page = alloc_pages(GFP_KERNEL, 0);
+ if (!super->s_erase_page)
+ goto fail2;
+ memset(page_address(super->s_erase_page), 0xFF, PAGE_SIZE);
+
+ /* FIXME: check for read-only mounts */
+ err = logfs_make_writeable(sb);
+ if (err)
+ goto fail3;
+
+ log_super("LogFS: Finished mounting\n");
+ simple_set_mnt(mnt, sb);
+ return 0;
+
+fail3:
+ __free_page(super->s_erase_page);
+fail2:
+ iput(rootdir);
+fail:
+ iput(logfs_super(sb)->s_master_inode);
+ return -EIO;
+}
+
+int logfs_check_ds(struct logfs_disk_super *ds)
+{
+ struct logfs_segment_header *sh = &ds->ds_sh;
+
+ if (ds->ds_magic != cpu_to_be64(LOGFS_MAGIC))
+ return -EINVAL;
+ if (sh->crc != logfs_crc32(sh, LOGFS_SEGMENT_HEADERSIZE, 4))
+ return -EINVAL;
+ if (ds->ds_crc != logfs_crc32(ds, sizeof(*ds),
+ LOGFS_SEGMENT_HEADERSIZE + 12))
+ return -EINVAL;
+ return 0;
+}
+
+static struct page *find_super_block(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct page *first, *last;
+
+ first = super->s_devops->find_first_sb(sb, &super->s_sb_ofs[0]);
+ if (!first || IS_ERR(first))
+ return NULL;
+ last = super->s_devops->find_last_sb(sb, &super->s_sb_ofs[1]);
+ if (!last || IS_ERR(first)) {
+ page_cache_release(first);
+ return NULL;
+ }
+
+ if (!logfs_check_ds(page_address(first))) {
+ page_cache_release(last);
+ return first;
+ }
+
+ /* First one didn't work, try the second superblock */
+ if (!logfs_check_ds(page_address(last))) {
+ page_cache_release(first);
+ return last;
+ }
+
+ /* Neither worked, sorry folks */
+ page_cache_release(first);
+ page_cache_release(last);
+ return NULL;
+}
+
+static int __logfs_read_sb(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+ struct page *page;
+ struct logfs_disk_super *ds;
+ int i;
+
+ page = find_super_block(sb);
+ if (!page)
+ return -EIO;
+
+ ds = page_address(page);
+ super->s_size = be64_to_cpu(ds->ds_filesystem_size);
+ super->s_root_reserve = be64_to_cpu(ds->ds_root_reserve);
+ super->s_speed_reserve = be64_to_cpu(ds->ds_speed_reserve);
+ super->s_bad_seg_reserve = be32_to_cpu(ds->ds_bad_seg_reserve);
+ super->s_segsize = 1 << ds->ds_segment_shift;
+ super->s_segmask = (1 << ds->ds_segment_shift) - 1;
+ super->s_segshift = ds->ds_segment_shift;
+ sb->s_blocksize = 1 << ds->ds_block_shift;
+ sb->s_blocksize_bits = ds->ds_block_shift;
+ super->s_writesize = 1 << ds->ds_write_shift;
+ super->s_writeshift = ds->ds_write_shift;
+ super->s_no_segs = super->s_size >> super->s_segshift;
+ super->s_no_blocks = super->s_segsize >> sb->s_blocksize_bits;
+ super->s_feature_incompat = be64_to_cpu(ds->ds_feature_incompat);
+ super->s_feature_ro_compat = be64_to_cpu(ds->ds_feature_ro_compat);
+ super->s_feature_compat = be64_to_cpu(ds->ds_feature_compat);
+ super->s_feature_flags = be64_to_cpu(ds->ds_feature_flags);
+
+ journal_for_each(i)
+ super->s_journal_seg[i] = be32_to_cpu(ds->ds_journal_seg[i]);
+
+ super->s_ifile_levels = ds->ds_ifile_levels;
+ super->s_iblock_levels = ds->ds_iblock_levels;
+ super->s_data_levels = ds->ds_data_levels;
+ super->s_total_levels = super->s_ifile_levels + super->s_iblock_levels
+ + super->s_data_levels;
+ page_cache_release(page);
+ return 0;
+}
+
+static int logfs_read_sb(struct super_block *sb, int read_only)
+{
+ struct logfs_super *super = logfs_super(sb);
+ int ret;
+
+ super->s_btree_pool = mempool_create(32, btree_alloc, btree_free, NULL);
+ if (!super->s_btree_pool)
+ return -ENOMEM;
+
+ btree_init_mempool64(&super->s_shadow_tree.new, super->s_btree_pool);
+ btree_init_mempool64(&super->s_shadow_tree.old, super->s_btree_pool);
+
+ ret = logfs_init_mapping(sb);
+ if (ret)
+ return ret;
+
+ ret = __logfs_read_sb(sb);
+ if (ret)
+ return ret;
+
+ if (super->s_feature_incompat & ~LOGFS_FEATURES_INCOMPAT)
+ return -EIO;
+ if ((super->s_feature_ro_compat & ~LOGFS_FEATURES_RO_COMPAT) &&
+ !read_only)
+ return -EIO;
+
+ mutex_init(&super->s_dirop_mutex);
+ mutex_init(&super->s_object_alias_mutex);
+ INIT_LIST_HEAD(&super->s_freeing_list);
+
+ ret = logfs_init_rw(sb);
+ if (ret)
+ return ret;
+
+ ret = logfs_init_areas(sb);
+ if (ret)
+ return ret;
+
+ ret = logfs_init_gc(sb);
+ if (ret)
+ return ret;
+
+ ret = logfs_init_journal(sb);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
+static void logfs_kill_sb(struct super_block *sb)
+{
+ struct logfs_super *super = logfs_super(sb);
+
+ log_super("LogFS: Start unmounting\n");
+ /* Alias entries slow down mount, so evict as many as possible */
+ sync_filesystem(sb);
+ logfs_write_anchor(sb);
+
+ /*
+ * From this point on alias entries are simply dropped - and any
+ * writes to the object store are considered bugs.
+ */
+ super->s_flags |= LOGFS_SB_FLAG_SHUTDOWN;
+ log_super("LogFS: Now in shutdown\n");
+ generic_shutdown_super(sb);
+
+ BUG_ON(super->s_dirty_used_bytes || super->s_dirty_free_bytes);
+
+ logfs_cleanup_gc(sb);
+ logfs_cleanup_journal(sb);
+ logfs_cleanup_areas(sb);
+ logfs_cleanup_rw(sb);
+ if (super->s_erase_page)
+ __free_page(super->s_erase_page);
+ super->s_devops->put_device(sb);
+ mempool_destroy(super->s_btree_pool);
+ mempool_destroy(super->s_alias_pool);
+ kfree(super);
+ log_super("LogFS: Finished unmounting\n");
+}
+
+int logfs_get_sb_device(struct file_system_type *type, int flags,
+ struct mtd_info *mtd, struct block_device *bdev,
+ const struct logfs_device_ops *devops, struct vfsmount *mnt)
+{
+ struct logfs_super *super;
+ struct super_block *sb;
+ int err = -ENOMEM;
+ static int mount_count;
+
+ log_super("LogFS: Start mount %x\n", mount_count++);
+ super = kzalloc(sizeof(*super), GFP_KERNEL);
+ if (!super)
+ goto err0;
+
+ super->s_mtd = mtd;
+ super->s_bdev = bdev;
+ err = -EINVAL;
+ sb = sget(type, logfs_sb_test, logfs_sb_set, super);
+ if (IS_ERR(sb))
+ goto err0;
+
+ if (sb->s_root) {
+ /* Device is already in use */
+ err = 0;
+ simple_set_mnt(mnt, sb);
+ goto err0;
+ }
+
+ super->s_devops = devops;
+
+ /*
+ * sb->s_maxbytes is limited to 8TB. On 32bit systems, the page cache
+ * only covers 16TB and the upper 8TB are used for indirect blocks.
+ * On 64bit system we could bump up the limit, but that would make
+ * the filesystem incompatible with 32bit systems.
+ */
+ sb->s_maxbytes = (1ull << 43) - 1;
+ sb->s_op = &logfs_super_operations;
+ sb->s_flags = flags | MS_NOATIME;
+
+ err = logfs_read_sb(sb, sb->s_flags & MS_RDONLY);
+ if (err)
+ goto err1;
+
+ sb->s_flags |= MS_ACTIVE;
+ err = logfs_get_sb_final(sb, mnt);
+ if (err)
+ goto err1;
+ return 0;
+
+err1:
+ up_write(&sb->s_umount);
+ deactivate_super(sb);
+ return err;
+err0:
+ kfree(super);
+ //devops->put_device(sb);
+ return err;
+}
+
+static int logfs_get_sb(struct file_system_type *type, int flags,
+ const char *devname, void *data, struct vfsmount *mnt)
+{
+ ulong mtdnr;
+
+ if (!devname)
+ return logfs_get_sb_bdev(type, flags, devname, mnt);
+ if (strncmp(devname, "mtd", 3))
+ return logfs_get_sb_bdev(type, flags, devname, mnt);
+
+ {
+ char *garbage;
+ mtdnr = simple_strtoul(devname+3, &garbage, 0);
+ if (*garbage)
+ return -EINVAL;
+ }
+
+ return logfs_get_sb_mtd(type, flags, mtdnr, mnt);
+}
+
+static struct file_system_type logfs_fs_type = {
+ .owner = THIS_MODULE,
+ .name = "logfs",
+ .get_sb = logfs_get_sb,
+ .kill_sb = logfs_kill_sb,
+ .fs_flags = FS_REQUIRES_DEV,
+
+};
+
+static int __init logfs_init(void)
+{
+ int ret;
+
+ emergency_page = alloc_pages(GFP_KERNEL, 0);
+ if (!emergency_page)
+ return -ENOMEM;
+
+ ret = logfs_compr_init();
+ if (ret)
+ goto out1;
+
+ ret = logfs_init_inode_cache();
+ if (ret)
+ goto out2;
+
+ return register_filesystem(&logfs_fs_type);
+out2:
+ logfs_compr_exit();
+out1:
+ __free_pages(emergency_page, 0);
+ return ret;
+}
+
+static void __exit logfs_exit(void)
+{
+ unregister_filesystem(&logfs_fs_type);
+ logfs_destroy_inode_cache();
+ logfs_compr_exit();
+ __free_pages(emergency_page, 0);
+}
+
+module_init(logfs_init);
+module_exit(logfs_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Joern Engel <joern@logfs.org>");
+MODULE_DESCRIPTION("scalable flash filesystem");
diff --git a/fs/namei.c b/fs/namei.c
index 3d9d2f965f84..48e60a187325 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1656,7 +1656,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path,
if (path->dentry->d_inode->i_op->follow_link)
return NULL;
error = -ENOTDIR;
- if (*want_dir & !path->dentry->d_inode->i_op->lookup)
+ if (*want_dir && !path->dentry->d_inode->i_op->lookup)
goto exit_dput;
path_to_nameidata(path, nd);
audit_inode(pathname, nd->path.dentry);
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 73ab220354df..36dfdae95123 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -118,7 +118,6 @@ nfs4_callback_up(struct svc_serv *serv)
dprintk("NFS: Callback listener port = %u (af %u)\n",
nfs_callback_tcpport, PF_INET);
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
ret = svc_create_xprt(serv, "tcp", PF_INET6,
nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS);
if (ret > 0) {
@@ -129,7 +128,6 @@ nfs4_callback_up(struct svc_serv *serv)
ret = 0;
else
goto out_err;
-#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
return svc_prepare_thread(serv, &serv->sv_pools[0]);
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index c6eed2a3b093..4bc22c763de7 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -525,6 +525,8 @@ static struct rpc_cred *callback_cred;
int set_callback_cred(void)
{
+ if (callback_cred)
+ return 0;
callback_cred = rpc_lookup_machine_cred();
if (!callback_cred)
return -ENOMEM;
@@ -542,7 +544,8 @@ void do_probe_callback(struct nfs4_client *clp)
};
int status;
- status = rpc_call_async(cb->cb_client, &msg, RPC_TASK_SOFT,
+ status = rpc_call_async(cb->cb_client, &msg,
+ RPC_TASK_SOFT | RPC_TASK_SOFTCONN,
&nfsd4_cb_probe_ops, (void *)clp);
if (status) {
warn_no_callback_path(clp, status);
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index 5a754f7b71ed..98fb98e330b4 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -119,9 +119,7 @@ out_no_tfm:
static void
nfsd4_sync_rec_dir(void)
{
- mutex_lock(&rec_dir.dentry->d_inode->i_mutex);
- nfsd_sync_dir(rec_dir.dentry);
- mutex_unlock(&rec_dir.dentry->d_inode->i_mutex);
+ vfs_fsync(NULL, rec_dir.dentry, 0);
}
int
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index f19ed866c95f..c97fddbd17db 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1998,7 +1998,9 @@ nfs4_file_downgrade(struct file *filp, unsigned int share_access)
{
if (share_access & NFS4_SHARE_ACCESS_WRITE) {
drop_file_write_access(filp);
+ spin_lock(&filp->f_lock);
filp->f_mode = (filp->f_mode | FMODE_READ) & ~FMODE_WRITE;
+ spin_unlock(&filp->f_lock);
}
}
@@ -2480,8 +2482,10 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
}
memcpy(&open->op_stateid, &stp->st_stateid, sizeof(stateid_t));
- if (nfsd4_has_session(&resp->cstate))
+ if (nfsd4_has_session(&resp->cstate)) {
open->op_stateowner->so_confirmed = 1;
+ nfsd4_create_clid_dir(open->op_stateowner->so_client);
+ }
/*
* Attempt to hand out a delegation. No error return, because the
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index bbf72d8f9fc0..78c7e24e5129 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1434,7 +1434,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
}
op->opnum = ntohl(*argp->p++);
- if (op->opnum >= OP_ACCESS && op->opnum < ops->nops)
+ if (op->opnum >= FIRST_NFS4_OP && op->opnum <= LAST_NFS4_OP)
op->status = ops->decoders[op->opnum](argp, &op->u);
else {
op->opnum = OP_ILLEGAL;
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 2604c3e70ea5..0f0e77f2012f 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -988,6 +988,7 @@ static ssize_t __write_ports_delfd(char *buf)
static ssize_t __write_ports_addxprt(char *buf)
{
char transport[16];
+ struct svc_xprt *xprt;
int port, err;
if (sscanf(buf, "%15s %4u", transport, &port) != 2)
@@ -1002,13 +1003,24 @@ static ssize_t __write_ports_addxprt(char *buf)
err = svc_create_xprt(nfsd_serv, transport,
PF_INET, port, SVC_SOCK_ANONYMOUS);
- if (err < 0) {
- /* Give a reasonable perror msg for bad transport string */
- if (err == -ENOENT)
- err = -EPROTONOSUPPORT;
- return err;
- }
+ if (err < 0)
+ goto out_err;
+
+ err = svc_create_xprt(nfsd_serv, transport,
+ PF_INET6, port, SVC_SOCK_ANONYMOUS);
+ if (err < 0 && err != -EAFNOSUPPORT)
+ goto out_close;
return 0;
+out_close:
+ xprt = svc_find_xprt(nfsd_serv, transport, PF_INET, port);
+ if (xprt != NULL) {
+ svc_close_xprt(xprt);
+ svc_xprt_put(xprt);
+ }
+out_err:
+ /* Decrease the count, but don't shut down the service */
+ nfsd_serv->sv_nrthreads--;
+ return err;
}
/*
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 8eca17df4f63..a11b0e8678ee 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -26,6 +26,8 @@
#include <linux/jhash.h>
#include <linux/ima.h>
#include <asm/uaccess.h>
+#include <linux/exportfs.h>
+#include <linux/writeback.h>
#ifdef CONFIG_NFSD_V3
#include "xdr3.h"
@@ -270,6 +272,32 @@ out:
return err;
}
+/*
+ * Commit metadata changes to stable storage.
+ */
+static int
+commit_metadata(struct svc_fh *fhp)
+{
+ struct inode *inode = fhp->fh_dentry->d_inode;
+ const struct export_operations *export_ops = inode->i_sb->s_export_op;
+ int error = 0;
+
+ if (!EX_ISSYNC(fhp->fh_export))
+ return 0;
+
+ if (export_ops->commit_metadata) {
+ error = export_ops->commit_metadata(inode);
+ } else {
+ struct writeback_control wbc = {
+ .sync_mode = WB_SYNC_ALL,
+ .nr_to_write = 0, /* metadata only */
+ };
+
+ error = sync_inode(inode, &wbc);
+ }
+
+ return error;
+}
/*
* Set various file attributes.
@@ -767,43 +795,6 @@ nfsd_close(struct file *filp)
}
/*
- * Sync a file
- * As this calls fsync (not fdatasync) there is no need for a write_inode
- * after it.
- */
-static inline int nfsd_dosync(struct file *filp, struct dentry *dp,
- const struct file_operations *fop)
-{
- struct inode *inode = dp->d_inode;
- int (*fsync) (struct file *, struct dentry *, int);
- int err;
-
- err = filemap_write_and_wait(inode->i_mapping);
- if (err == 0 && fop && (fsync = fop->fsync))
- err = fsync(filp, dp, 0);
- return err;
-}
-
-static int
-nfsd_sync(struct file *filp)
-{
- int err;
- struct inode *inode = filp->f_path.dentry->d_inode;
- dprintk("nfsd: sync file %s\n", filp->f_path.dentry->d_name.name);
- mutex_lock(&inode->i_mutex);
- err=nfsd_dosync(filp, filp->f_path.dentry, filp->f_op);
- mutex_unlock(&inode->i_mutex);
-
- return err;
-}
-
-int
-nfsd_sync_dir(struct dentry *dp)
-{
- return nfsd_dosync(NULL, dp, dp->d_inode->i_fop);
-}
-
-/*
* Obtain the readahead parameters for the file
* specified by (dev, ino).
*/
@@ -1006,7 +997,7 @@ static int wait_for_concurrent_writes(struct file *file)
if (inode->i_state & I_DIRTY) {
dprintk("nfsd: write sync %d\n", task_pid_nr(current));
- err = nfsd_sync(file);
+ err = vfs_fsync(file, file->f_path.dentry, 0);
}
last_ino = inode->i_ino;
last_dev = inode->i_sb->s_dev;
@@ -1154,8 +1145,9 @@ out:
#ifdef CONFIG_NFSD_V3
/*
* Commit all pending writes to stable storage.
- * Strictly speaking, we could sync just the indicated file region here,
- * but there's currently no way we can ask the VFS to do so.
+ *
+ * Note: we only guarantee that data that lies within the range specified
+ * by the 'offset' and 'count' parameters will be synced.
*
* 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.
@@ -1165,23 +1157,32 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp,
loff_t offset, unsigned long count)
{
struct file *file;
- __be32 err;
+ loff_t end = LLONG_MAX;
+ __be32 err = nfserr_inval;
- if ((u64)count > ~(u64)offset)
- return nfserr_inval;
+ if (offset < 0)
+ goto out;
+ if (count != 0) {
+ end = offset + (loff_t)count - 1;
+ if (end < offset)
+ goto out;
+ }
err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_WRITE, &file);
if (err)
- return err;
+ goto out;
if (EX_ISSYNC(fhp->fh_export)) {
- if (file->f_op && file->f_op->fsync) {
- err = nfserrno(nfsd_sync(file));
- } else {
+ int err2 = vfs_fsync_range(file, file->f_path.dentry,
+ offset, end, 0);
+
+ if (err2 != -EINVAL)
+ err = nfserrno(err2);
+ else
err = nfserr_notsupp;
- }
}
nfsd_close(file);
+out:
return err;
}
#endif /* CONFIG_NFSD_V3 */
@@ -1334,12 +1335,14 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
goto out_nfserr;
}
- if (EX_ISSYNC(fhp->fh_export)) {
- err = nfserrno(nfsd_sync_dir(dentry));
- write_inode_now(dchild->d_inode, 1);
- }
+ err = nfsd_create_setattr(rqstp, resfhp, iap);
- err2 = nfsd_create_setattr(rqstp, resfhp, iap);
+ /*
+ * nfsd_setattr already committed the child. Transactional filesystems
+ * had a chance to commit changes for both parent and child
+ * simultaneously making the following commit_metadata a noop.
+ */
+ err2 = nfserrno(commit_metadata(fhp));
if (err2)
err = err2;
mnt_drop_write(fhp->fh_export->ex_path.mnt);
@@ -1371,7 +1374,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
struct dentry *dentry, *dchild = NULL;
struct inode *dirp;
__be32 err;
- __be32 err2;
int host_err;
__u32 v_mtime=0, v_atime=0;
@@ -1466,11 +1468,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
if (created)
*created = 1;
- if (EX_ISSYNC(fhp->fh_export)) {
- err = nfserrno(nfsd_sync_dir(dentry));
- /* setattr will sync the child (or not) */
- }
-
nfsd_check_ignore_resizing(iap);
if (createmode == NFS3_CREATE_EXCLUSIVE) {
@@ -1485,9 +1482,13 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
}
set_attr:
- err2 = nfsd_create_setattr(rqstp, resfhp, iap);
- if (err2)
- err = err2;
+ err = nfsd_create_setattr(rqstp, resfhp, iap);
+
+ /*
+ * nfsd_setattr already committed the child (and possibly also the parent).
+ */
+ if (!err)
+ err = nfserrno(commit_metadata(fhp));
mnt_drop_write(fhp->fh_export->ex_path.mnt);
/*
@@ -1602,12 +1603,9 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
}
} else
host_err = vfs_symlink(dentry->d_inode, dnew, path);
-
- if (!host_err) {
- if (EX_ISSYNC(fhp->fh_export))
- host_err = nfsd_sync_dir(dentry);
- }
err = nfserrno(host_err);
+ if (!err)
+ err = nfserrno(commit_metadata(fhp));
fh_unlock(fhp);
mnt_drop_write(fhp->fh_export->ex_path.mnt);
@@ -1669,11 +1667,9 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
}
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;
+ err = nfserrno(commit_metadata(ffhp));
+ if (!err)
+ err = nfserrno(commit_metadata(tfhp));
} else {
if (host_err == -EXDEV && rqstp->rq_vers == 2)
err = nfserr_acces;
@@ -1769,10 +1765,10 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
goto out_dput_new;
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 = commit_metadata(tfhp);
if (!host_err)
- host_err = nfsd_sync_dir(fdentry);
+ host_err = commit_metadata(ffhp);
}
mnt_drop_write(ffhp->fh_export->ex_path.mnt);
@@ -1853,12 +1849,9 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
dput(rdentry);
- if (host_err)
- goto out_drop;
- if (EX_ISSYNC(fhp->fh_export))
- host_err = nfsd_sync_dir(dentry);
+ if (!host_err)
+ host_err = commit_metadata(fhp);
-out_drop:
mnt_drop_write(fhp->fh_export->ex_path.mnt);
out_nfserr:
err = nfserrno(host_err);
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index 21f9e71223ca..a6467f3d262e 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -457,7 +457,7 @@ static int ocfs2_recover_local_quota_file(struct inode *lqinode,
break;
}
dchunk = (struct ocfs2_local_disk_chunk *)hbh->b_data;
- for_each_bit(bit, rchunk->rc_bitmap, ol_chunk_entries(sb)) {
+ for_each_set_bit(bit, rchunk->rc_bitmap, ol_chunk_entries(sb)) {
qbh = NULL;
status = ocfs2_read_quota_block(lqinode,
ol_dqblk_block(sb, chunk, bit),
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 18e20feee251..aa8637b81028 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -273,7 +273,7 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p)
rcu_read_lock(); /* FIXME: is this correct? */
qsize = atomic_read(&__task_cred(p)->user->sigpending);
rcu_read_unlock();
- qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur;
+ qlim = task_rlimit(p, RLIMIT_SIGPENDING);
unlock_task_sighand(p, &flags);
}
@@ -420,7 +420,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
cutime = sig->cutime;
cstime = sig->cstime;
cgtime = sig->cgtime;
- rsslim = sig->rlim[RLIMIT_RSS].rlim_cur;
+ rsslim = ACCESS_ONCE(sig->rlim[RLIMIT_RSS].rlim_cur);
/* add up live thread stats at the group level */
if (whole) {
diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index 9580abeadeb3..08f4d71dacd7 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -291,19 +291,17 @@ static const struct inode_operations proc_file_inode_operations = {
* returns the struct proc_dir_entry for "/proc/tty/driver", and
* returns "serial" in residual.
*/
-static int xlate_proc_name(const char *name,
- struct proc_dir_entry **ret, const char **residual)
+static int __xlate_proc_name(const char *name, struct proc_dir_entry **ret,
+ const char **residual)
{
const char *cp = name, *next;
struct proc_dir_entry *de;
int len;
- int rtn = 0;
de = *ret;
if (!de)
de = &proc_root;
- spin_lock(&proc_subdir_lock);
while (1) {
next = strchr(cp, '/');
if (!next)
@@ -315,16 +313,25 @@ static int xlate_proc_name(const char *name,
break;
}
if (!de) {
- rtn = -ENOENT;
- goto out;
+ WARN(1, "name '%s'\n", name);
+ return -ENOENT;
}
cp += len + 1;
}
*residual = cp;
*ret = de;
-out:
+ return 0;
+}
+
+static int xlate_proc_name(const char *name, struct proc_dir_entry **ret,
+ const char **residual)
+{
+ int rv;
+
+ spin_lock(&proc_subdir_lock);
+ rv = __xlate_proc_name(name, ret, residual);
spin_unlock(&proc_subdir_lock);
- return rtn;
+ return rv;
}
static DEFINE_IDA(proc_inum_ida);
@@ -797,11 +804,13 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent)
const char *fn = name;
int len;
- if (xlate_proc_name(name, &parent, &fn) != 0)
+ spin_lock(&proc_subdir_lock);
+ if (__xlate_proc_name(name, &parent, &fn) != 0) {
+ spin_unlock(&proc_subdir_lock);
return;
+ }
len = strlen(fn);
- spin_lock(&proc_subdir_lock);
for (p = &parent->subdir; *p; p=&(*p)->next ) {
if (proc_match(len, fn, *p)) {
de = *p;
@@ -811,8 +820,10 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent)
}
}
spin_unlock(&proc_subdir_lock);
- if (!de)
+ if (!de) {
+ WARN(1, "name '%s'\n", name);
return;
+ }
spin_lock(&de->pde_unload_lock);
/*
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index f277c4a111cb..183f8ff5f400 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -16,7 +16,7 @@
void task_mem(struct seq_file *m, struct mm_struct *mm)
{
- unsigned long data, text, lib;
+ unsigned long data, text, lib, swap;
unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss;
/*
@@ -36,6 +36,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
data = mm->total_vm - mm->shared_vm - mm->stack_vm;
text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10;
lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text;
+ swap = get_mm_counter(mm, MM_SWAPENTS);
seq_printf(m,
"VmPeak:\t%8lu kB\n"
"VmSize:\t%8lu kB\n"
@@ -46,7 +47,8 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
"VmStk:\t%8lu kB\n"
"VmExe:\t%8lu kB\n"
"VmLib:\t%8lu kB\n"
- "VmPTE:\t%8lu kB\n",
+ "VmPTE:\t%8lu kB\n"
+ "VmSwap:\t%8lu kB\n",
hiwater_vm << (PAGE_SHIFT-10),
(total_vm - mm->reserved_vm) << (PAGE_SHIFT-10),
mm->locked_vm << (PAGE_SHIFT-10),
@@ -54,7 +56,8 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
total_rss << (PAGE_SHIFT-10),
data << (PAGE_SHIFT-10),
mm->stack_vm << (PAGE_SHIFT-10), text, lib,
- (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10);
+ (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10,
+ swap << (PAGE_SHIFT-10));
}
unsigned long task_vsize(struct mm_struct *mm)
@@ -65,11 +68,11 @@ unsigned long task_vsize(struct mm_struct *mm)
int task_statm(struct mm_struct *mm, int *shared, int *text,
int *data, int *resident)
{
- *shared = get_mm_counter(mm, file_rss);
+ *shared = get_mm_counter(mm, MM_FILEPAGES);
*text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK))
>> PAGE_SHIFT;
*data = mm->total_vm - mm->shared_vm;
- *resident = *shared + get_mm_counter(mm, anon_rss);
+ *resident = *shared + get_mm_counter(mm, MM_ANONPAGES);
return mm->total_vm;
}
diff --git a/fs/select.c b/fs/select.c
index fd38ce2e32e3..73715e90030f 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -821,7 +821,7 @@ int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds,
struct poll_list *walk = head;
unsigned long todo = nfds;
- if (nfds > current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
+ if (nfds > rlimit(RLIMIT_NOFILE))
return -EINVAL;
len = min_t(unsigned int, nfds, N_STACK_PPS);
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 5afd554efad3..e1f437be6c3c 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -734,7 +734,7 @@ EXPORT_SYMBOL(seq_hlist_start_head);
* seq_hlist_next - move to the next position of the hlist
* @v: the current iterator
* @head: the head of the hlist
- * @pos: the current posision
+ * @ppos: the current position
*
* Called at seq_file->op->next().
*/
@@ -800,7 +800,7 @@ EXPORT_SYMBOL(seq_hlist_start_head_rcu);
* seq_hlist_next_rcu - move to the next position of the hlist protected by RCU
* @v: the current iterator
* @head: the head of the hlist
- * @pos: the current posision
+ * @ppos: the current position
*
* Called at seq_file->op->next().
*
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 5c5a366aa332..b4769e40e8bc 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -105,7 +105,6 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \
xfs_globals.o \
xfs_ioctl.o \
xfs_iops.o \
- xfs_lrw.o \
xfs_super.o \
xfs_sync.o \
xfs_xattr.o)
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 66abe36c1213..9083357f9e44 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -39,6 +39,7 @@
#include "xfs_iomap.h"
#include "xfs_vnodeops.h"
#include "xfs_trace.h"
+#include "xfs_bmap.h"
#include <linux/mpage.h>
#include <linux/pagevec.h>
#include <linux/writeback.h>
@@ -163,14 +164,17 @@ xfs_ioend_new_eof(
}
/*
- * Update on-disk file size now that data has been written to disk.
- * The current in-memory file size is i_size. If a write is beyond
- * eof i_new_size will be the intended file size until i_size is
- * updated. If this write does not extend all the way to the valid
- * file size then restrict this update to the end of the write.
+ * Update on-disk file size now that data has been written to disk. The
+ * current in-memory file size is i_size. If a write is beyond eof i_new_size
+ * will be the intended file size until i_size is updated. If this write does
+ * not extend all the way to the valid file size then restrict this update to
+ * the end of the write.
+ *
+ * This function does not block as blocking on the inode lock in IO completion
+ * can lead to IO completion order dependency deadlocks.. If it can't get the
+ * inode ilock it will return EAGAIN. Callers must handle this.
*/
-
-STATIC void
+STATIC int
xfs_setfilesize(
xfs_ioend_t *ioend)
{
@@ -181,16 +185,40 @@ xfs_setfilesize(
ASSERT(ioend->io_type != IOMAP_READ);
if (unlikely(ioend->io_error))
- return;
+ return 0;
+
+ if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))
+ return EAGAIN;
- xfs_ilock(ip, XFS_ILOCK_EXCL);
isize = xfs_ioend_new_eof(ioend);
if (isize) {
ip->i_d.di_size = isize;
- xfs_mark_inode_dirty_sync(ip);
+ xfs_mark_inode_dirty(ip);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ return 0;
+}
+
+/*
+ * Schedule IO completion handling on a xfsdatad if this was
+ * the final hold on this ioend. If we are asked to wait,
+ * flush the workqueue.
+ */
+STATIC void
+xfs_finish_ioend(
+ xfs_ioend_t *ioend,
+ int wait)
+{
+ if (atomic_dec_and_test(&ioend->io_remaining)) {
+ struct workqueue_struct *wq;
+
+ wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
+ xfsconvertd_workqueue : xfsdatad_workqueue;
+ queue_work(wq, &ioend->io_work);
+ if (wait)
+ flush_workqueue(wq);
+ }
}
/*
@@ -198,11 +226,11 @@ xfs_setfilesize(
*/
STATIC void
xfs_end_io(
- struct work_struct *work)
+ struct work_struct *work)
{
- xfs_ioend_t *ioend =
- container_of(work, xfs_ioend_t, io_work);
- struct xfs_inode *ip = XFS_I(ioend->io_inode);
+ xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work);
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
+ int error = 0;
/*
* For unwritten extents we need to issue transactions to convert a
@@ -210,7 +238,6 @@ xfs_end_io(
*/
if (ioend->io_type == IOMAP_UNWRITTEN &&
likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {
- int error;
error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
ioend->io_size);
@@ -222,30 +249,23 @@ xfs_end_io(
* We might have to update the on-disk file size after extending
* writes.
*/
- if (ioend->io_type != IOMAP_READ)
- xfs_setfilesize(ioend);
- xfs_destroy_ioend(ioend);
-}
-
-/*
- * Schedule IO completion handling on a xfsdatad if this was
- * the final hold on this ioend. If we are asked to wait,
- * flush the workqueue.
- */
-STATIC void
-xfs_finish_ioend(
- xfs_ioend_t *ioend,
- int wait)
-{
- if (atomic_dec_and_test(&ioend->io_remaining)) {
- struct workqueue_struct *wq;
-
- wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
- xfsconvertd_workqueue : xfsdatad_workqueue;
- queue_work(wq, &ioend->io_work);
- if (wait)
- flush_workqueue(wq);
+ if (ioend->io_type != IOMAP_READ) {
+ error = xfs_setfilesize(ioend);
+ ASSERT(!error || error == EAGAIN);
}
+
+ /*
+ * If we didn't complete processing of the ioend, requeue it to the
+ * tail of the workqueue for another attempt later. Otherwise destroy
+ * it.
+ */
+ if (error == EAGAIN) {
+ atomic_inc(&ioend->io_remaining);
+ xfs_finish_ioend(ioend, 0);
+ /* ensure we don't spin on blocked ioends */
+ delay(1);
+ } else
+ xfs_destroy_ioend(ioend);
}
/*
@@ -341,7 +361,7 @@ xfs_submit_ioend_bio(
* but don't update the inode size until I/O completion.
*/
if (xfs_ioend_new_eof(ioend))
- xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
+ xfs_mark_inode_dirty(XFS_I(ioend->io_inode));
submit_bio(wbc->sync_mode == WB_SYNC_ALL ?
WRITE_SYNC_PLUG : WRITE, bio);
@@ -874,6 +894,118 @@ xfs_cluster_write(
}
}
+STATIC void
+xfs_vm_invalidatepage(
+ struct page *page,
+ unsigned long offset)
+{
+ trace_xfs_invalidatepage(page->mapping->host, page, offset);
+ block_invalidatepage(page, offset);
+}
+
+/*
+ * If the page has delalloc buffers on it, we need to punch them out before we
+ * invalidate the page. If we don't, we leave a stale delalloc mapping on the
+ * inode that can trip a BUG() in xfs_get_blocks() later on if a direct IO read
+ * is done on that same region - the delalloc extent is returned when none is
+ * supposed to be there.
+ *
+ * We prevent this by truncating away the delalloc regions on the page before
+ * invalidating it. Because they are delalloc, we can do this without needing a
+ * transaction. Indeed - if we get ENOSPC errors, we have to be able to do this
+ * truncation without a transaction as there is no space left for block
+ * reservation (typically why we see a ENOSPC in writeback).
+ *
+ * This is not a performance critical path, so for now just do the punching a
+ * buffer head at a time.
+ */
+STATIC void
+xfs_aops_discard_page(
+ struct page *page)
+{
+ struct inode *inode = page->mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct buffer_head *bh, *head;
+ loff_t offset = page_offset(page);
+ ssize_t len = 1 << inode->i_blkbits;
+
+ if (!xfs_is_delayed_page(page, IOMAP_DELAY))
+ goto out_invalidate;
+
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard on page %p, inode 0x%llx, offset %llu.",
+ page, ip->i_ino, offset);
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ bh = head = page_buffers(page);
+ do {
+ int done;
+ xfs_fileoff_t offset_fsb;
+ xfs_bmbt_irec_t imap;
+ int nimaps = 1;
+ int error;
+ xfs_fsblock_t firstblock;
+ xfs_bmap_free_t flist;
+
+ if (!buffer_delay(bh))
+ goto next_buffer;
+
+ offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset);
+
+ /*
+ * Map the range first and check that it is a delalloc extent
+ * before trying to unmap the range. Otherwise we will be
+ * trying to remove a real extent (which requires a
+ * transaction) or a hole, which is probably a bad idea...
+ */
+ error = xfs_bmapi(NULL, ip, offset_fsb, 1,
+ XFS_BMAPI_ENTIRE, NULL, 0, &imap,
+ &nimaps, NULL, NULL);
+
+ if (error) {
+ /* something screwed, just bail */
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard failed delalloc mapping lookup.");
+ break;
+ }
+ if (!nimaps) {
+ /* nothing there */
+ goto next_buffer;
+ }
+ if (imap.br_startblock != DELAYSTARTBLOCK) {
+ /* been converted, ignore */
+ goto next_buffer;
+ }
+ WARN_ON(imap.br_blockcount == 0);
+
+ /*
+ * Note: while we initialise the firstblock/flist pair, they
+ * should never be used because blocks should never be
+ * allocated or freed for a delalloc extent and hence we need
+ * don't cancel or finish them after the xfs_bunmapi() call.
+ */
+ xfs_bmap_init(&flist, &firstblock);
+ error = xfs_bunmapi(NULL, ip, offset_fsb, 1, 0, 1, &firstblock,
+ &flist, NULL, &done);
+
+ ASSERT(!flist.xbf_count && !flist.xbf_first);
+ if (error) {
+ /* something screwed, just bail */
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard unable to remove delalloc mapping.");
+ break;
+ }
+next_buffer:
+ offset += len;
+
+ } while ((bh = bh->b_this_page) != head);
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+out_invalidate:
+ xfs_vm_invalidatepage(page, 0);
+ return;
+}
+
/*
* Calling this without startio set means we are being asked to make a dirty
* page ready for freeing it's buffers. When called with startio set then
@@ -1125,7 +1257,7 @@ error:
*/
if (err != -EAGAIN) {
if (!unmapped)
- block_invalidatepage(page, 0);
+ xfs_aops_discard_page(page);
ClearPageUptodate(page);
}
return err;
@@ -1535,15 +1667,6 @@ xfs_vm_readpages(
return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks);
}
-STATIC void
-xfs_vm_invalidatepage(
- struct page *page,
- unsigned long offset)
-{
- trace_xfs_invalidatepage(page->mapping->host, page, offset);
- block_invalidatepage(page, offset);
-}
-
const struct address_space_operations xfs_address_space_operations = {
.readpage = xfs_vm_readpage,
.readpages = xfs_vm_readpages,
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index 87b8cbd23d4b..846b75aeb2ab 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -29,6 +29,7 @@
#include "xfs_vnodeops.h"
#include "xfs_bmap_btree.h"
#include "xfs_inode.h"
+#include "xfs_inode_item.h"
/*
* Note that we only accept fileids which are long enough rather than allow
@@ -215,9 +216,28 @@ xfs_fs_get_parent(
return d_obtain_alias(VFS_I(cip));
}
+STATIC int
+xfs_fs_nfs_commit_metadata(
+ struct inode *inode)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ int error = 0;
+
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+ if (xfs_ipincount(ip)) {
+ error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn,
+ XFS_LOG_SYNC, NULL);
+ }
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+
+ return error;
+}
+
const struct export_operations xfs_export_operations = {
.encode_fh = xfs_fs_encode_fh,
.fh_to_dentry = xfs_fs_fh_to_dentry,
.fh_to_parent = xfs_fs_fh_to_parent,
.get_parent = xfs_fs_get_parent,
+ .commit_metadata = xfs_fs_nfs_commit_metadata,
};
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index e4caeb28ce2e..42dd3bcfba6b 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -16,6 +16,7 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
+#include "xfs_fs.h"
#include "xfs_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
@@ -34,52 +35,279 @@
#include "xfs_dir2_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
+#include "xfs_inode_item.h"
+#include "xfs_bmap.h"
#include "xfs_error.h"
#include "xfs_rw.h"
#include "xfs_vnodeops.h"
#include "xfs_da_btree.h"
#include "xfs_ioctl.h"
+#include "xfs_trace.h"
#include <linux/dcache.h>
static const struct vm_operations_struct xfs_file_vm_ops;
-STATIC ssize_t
-xfs_file_aio_read(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
+/*
+ * xfs_iozero
+ *
+ * xfs_iozero clears the specified range of buffer supplied,
+ * and marks all the affected blocks as valid and modified. If
+ * an affected block is not allocated, it will be allocated. If
+ * an affected block is not completely overwritten, and is not
+ * valid before the operation, it will be read from disk before
+ * being partially zeroed.
+ */
+STATIC int
+xfs_iozero(
+ struct xfs_inode *ip, /* inode */
+ loff_t pos, /* offset in file */
+ size_t count) /* size of data to zero */
{
- struct file *file = iocb->ki_filp;
- int ioflags = 0;
+ struct page *page;
+ struct address_space *mapping;
+ int status;
- BUG_ON(iocb->ki_pos != pos);
- if (unlikely(file->f_flags & O_DIRECT))
- ioflags |= IO_ISDIRECT;
- if (file->f_mode & FMODE_NOCMTIME)
- ioflags |= IO_INVIS;
- return xfs_read(XFS_I(file->f_path.dentry->d_inode), iocb, iov,
- nr_segs, &iocb->ki_pos, ioflags);
+ mapping = VFS_I(ip)->i_mapping;
+ do {
+ unsigned offset, bytes;
+ void *fsdata;
+
+ offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
+ bytes = PAGE_CACHE_SIZE - offset;
+ if (bytes > count)
+ bytes = count;
+
+ status = pagecache_write_begin(NULL, mapping, pos, bytes,
+ AOP_FLAG_UNINTERRUPTIBLE,
+ &page, &fsdata);
+ if (status)
+ break;
+
+ zero_user(page, offset, bytes);
+
+ status = pagecache_write_end(NULL, mapping, pos, bytes, bytes,
+ page, fsdata);
+ WARN_ON(status <= 0); /* can't return less than zero! */
+ pos += bytes;
+ count -= bytes;
+ status = 0;
+ } while (count);
+
+ return (-status);
+}
+
+STATIC int
+xfs_file_fsync(
+ struct file *file,
+ struct dentry *dentry,
+ int datasync)
+{
+ struct xfs_inode *ip = XFS_I(dentry->d_inode);
+ struct xfs_trans *tp;
+ int error = 0;
+ int log_flushed = 0;
+
+ xfs_itrace_entry(ip);
+
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -XFS_ERROR(EIO);
+
+ xfs_iflags_clear(ip, XFS_ITRUNCATED);
+
+ /*
+ * We always need to make sure that the required inode state is safe on
+ * disk. The inode might be clean but we still might need to force the
+ * log because of committed transactions that haven't hit the disk yet.
+ * Likewise, there could be unflushed non-transactional changes to the
+ * inode core that have to go to disk and this requires us to issue
+ * a synchronous transaction to capture these changes correctly.
+ *
+ * This code relies on the assumption that if the i_update_core field
+ * of the inode is clear and the inode is unpinned then it is clean
+ * and no action is required.
+ */
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+
+ /*
+ * First check if the VFS inode is marked dirty. All the dirtying
+ * of non-transactional updates no goes through mark_inode_dirty*,
+ * which allows us to distinguish beteeen pure timestamp updates
+ * and i_size updates which need to be caught for fdatasync.
+ * After that also theck for the dirty state in the XFS inode, which
+ * might gets cleared when the inode gets written out via the AIL
+ * or xfs_iflush_cluster.
+ */
+ if (((dentry->d_inode->i_state & I_DIRTY_DATASYNC) ||
+ ((dentry->d_inode->i_state & I_DIRTY_SYNC) && !datasync)) &&
+ ip->i_update_core) {
+ /*
+ * Kick off a transaction to log the inode core to get the
+ * updates. The sync transaction will also force the log.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
+ error = xfs_trans_reserve(tp, 0,
+ XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
+ if (error) {
+ xfs_trans_cancel(tp, 0);
+ return -error;
+ }
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+
+ /*
+ * Note - it's possible that we might have pushed ourselves out
+ * of the way during trans_reserve which would flush the inode.
+ * But there's no guarantee that the inode buffer has actually
+ * gone out yet (it's delwri). Plus the buffer could be pinned
+ * anyway if it's part of an inode in another recent
+ * transaction. So we play it safe and fire off the
+ * transaction anyway.
+ */
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+ xfs_trans_ihold(tp, ip);
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ xfs_trans_set_sync(tp);
+ error = _xfs_trans_commit(tp, 0, &log_flushed);
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ } else {
+ /*
+ * Timestamps/size haven't changed since last inode flush or
+ * inode transaction commit. That means either nothing got
+ * written or a transaction committed which caught the updates.
+ * If the latter happened and the transaction hasn't hit the
+ * disk yet, the inode will be still be pinned. If it is,
+ * force the log.
+ */
+ if (xfs_ipincount(ip)) {
+ error = _xfs_log_force_lsn(ip->i_mount,
+ ip->i_itemp->ili_last_lsn,
+ XFS_LOG_SYNC, &log_flushed);
+ }
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ }
+
+ if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
+ /*
+ * If the log write didn't issue an ordered tag we need
+ * to flush the disk cache for the data device now.
+ */
+ if (!log_flushed)
+ xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
+
+ /*
+ * If this inode is on the RT dev we need to flush that
+ * cache as well.
+ */
+ if (XFS_IS_REALTIME_INODE(ip))
+ xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
+ }
+
+ return -error;
}
STATIC ssize_t
-xfs_file_aio_write(
+xfs_file_aio_read(
struct kiocb *iocb,
- const struct iovec *iov,
+ const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
+ struct inode *inode = file->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ size_t size = 0;
+ ssize_t ret = 0;
int ioflags = 0;
+ xfs_fsize_t n;
+ unsigned long seg;
+
+ XFS_STATS_INC(xs_read_calls);
BUG_ON(iocb->ki_pos != pos);
+
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= IO_ISDIRECT;
if (file->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
- return xfs_write(XFS_I(file->f_mapping->host), iocb, iov, nr_segs,
- &iocb->ki_pos, ioflags);
+
+ /* START copy & waste from filemap.c */
+ for (seg = 0; seg < nr_segs; seg++) {
+ const struct iovec *iv = &iovp[seg];
+
+ /*
+ * If any segment has a negative length, or the cumulative
+ * length ever wraps negative then return -EINVAL.
+ */
+ size += iv->iov_len;
+ if (unlikely((ssize_t)(size|iv->iov_len) < 0))
+ return XFS_ERROR(-EINVAL);
+ }
+ /* END copy & waste from filemap.c */
+
+ if (unlikely(ioflags & IO_ISDIRECT)) {
+ xfs_buftarg_t *target =
+ XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+ if ((iocb->ki_pos & target->bt_smask) ||
+ (size & target->bt_smask)) {
+ if (iocb->ki_pos == ip->i_size)
+ return 0;
+ return -XFS_ERROR(EINVAL);
+ }
+ }
+
+ n = XFS_MAXIOFFSET(mp) - iocb->ki_pos;
+ if (n <= 0 || size == 0)
+ return 0;
+
+ if (n < size)
+ size = n;
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return -EIO;
+
+ if (unlikely(ioflags & IO_ISDIRECT))
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
+ int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);
+ int iolock = XFS_IOLOCK_SHARED;
+
+ ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, iocb->ki_pos, size,
+ dmflags, &iolock);
+ if (ret) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ if (unlikely(ioflags & IO_ISDIRECT))
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
+ }
+
+ if (unlikely(ioflags & IO_ISDIRECT)) {
+ if (inode->i_mapping->nrpages) {
+ ret = -xfs_flushinval_pages(ip,
+ (iocb->ki_pos & PAGE_CACHE_MASK),
+ -1, FI_REMAPF_LOCKED);
+ }
+ mutex_unlock(&inode->i_mutex);
+ if (ret) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
+ }
+ }
+
+ trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
+
+ ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_read_bytes, ret);
+
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
}
STATIC ssize_t
@@ -87,16 +315,44 @@ xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
- size_t len,
+ size_t count,
unsigned int flags)
{
+ struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
+ struct xfs_mount *mp = ip->i_mount;
int ioflags = 0;
+ ssize_t ret;
+
+ XFS_STATS_INC(xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
- return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode),
- infilp, ppos, pipe, len, flags, ioflags);
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -EIO;
+
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
+ int iolock = XFS_IOLOCK_SHARED;
+ int error;
+
+ error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
+ FILP_DELAY_FLAG(infilp), &iolock);
+ if (error) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return -error;
+ }
+ }
+
+ trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
+
+ ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_read_bytes, ret);
+
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
}
STATIC ssize_t
@@ -104,16 +360,538 @@ xfs_file_splice_write(
struct pipe_inode_info *pipe,
struct file *outfilp,
loff_t *ppos,
- size_t len,
+ size_t count,
unsigned int flags)
{
+ struct inode *inode = outfilp->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ xfs_fsize_t isize, new_size;
int ioflags = 0;
+ ssize_t ret;
+
+ XFS_STATS_INC(xs_write_calls);
if (outfilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
- return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode),
- pipe, outfilp, ppos, len, flags, ioflags);
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -EIO;
+
+ xfs_ilock(ip, XFS_IOLOCK_EXCL);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) {
+ int iolock = XFS_IOLOCK_EXCL;
+ int error;
+
+ error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count,
+ FILP_DELAY_FLAG(outfilp), &iolock);
+ if (error) {
+ xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ return -error;
+ }
+ }
+
+ new_size = *ppos + count;
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
+
+ ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ isize = i_size_read(inode);
+ if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
+ *ppos = isize;
+
+ if (*ppos > ip->i_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (*ppos > ip->i_size)
+ ip->i_size = *ppos;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+
+ if (ip->i_new_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+ xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ return ret;
+}
+
+/*
+ * This routine is called to handle zeroing any space in the last
+ * block of the file that is beyond the EOF. We do this since the
+ * size is being increased without writing anything to that block
+ * and we don't want anyone to read the garbage on the disk.
+ */
+STATIC int /* error (positive) */
+xfs_zero_last_block(
+ xfs_inode_t *ip,
+ xfs_fsize_t offset,
+ xfs_fsize_t isize)
+{
+ xfs_fileoff_t last_fsb;
+ xfs_mount_t *mp = ip->i_mount;
+ int nimaps;
+ int zero_offset;
+ int zero_len;
+ int error = 0;
+ xfs_bmbt_irec_t imap;
+
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
+ zero_offset = XFS_B_FSB_OFFSET(mp, isize);
+ if (zero_offset == 0) {
+ /*
+ * There are no extra bytes in the last block on disk to
+ * zero, so return.
+ */
+ return 0;
+ }
+
+ last_fsb = XFS_B_TO_FSBT(mp, isize);
+ nimaps = 1;
+ error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
+ &nimaps, NULL, NULL);
+ if (error) {
+ return error;
+ }
+ ASSERT(nimaps > 0);
+ /*
+ * If the block underlying isize is just a hole, then there
+ * is nothing to zero.
+ */
+ if (imap.br_startblock == HOLESTARTBLOCK) {
+ return 0;
+ }
+ /*
+ * Zero the part of the last block beyond the EOF, and write it
+ * out sync. We need to drop the ilock while we do this so we
+ * don't deadlock when the buffer cache calls back to us.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ zero_len = mp->m_sb.sb_blocksize - zero_offset;
+ if (isize + zero_len > offset)
+ zero_len = offset - isize;
+ error = xfs_iozero(ip, isize, zero_len);
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ASSERT(error >= 0);
+ return error;
+}
+
+/*
+ * Zero any on disk space between the current EOF and the new,
+ * larger EOF. This handles the normal case of zeroing the remainder
+ * of the last block in the file and the unusual case of zeroing blocks
+ * out beyond the size of the file. This second case only happens
+ * with fixed size extents and when the system crashes before the inode
+ * size was updated but after blocks were allocated. If fill is set,
+ * then any holes in the range are filled and zeroed. If not, the holes
+ * are left alone as holes.
+ */
+
+int /* error (positive) */
+xfs_zero_eof(
+ xfs_inode_t *ip,
+ xfs_off_t offset, /* starting I/O offset */
+ xfs_fsize_t isize) /* current inode size */
+{
+ xfs_mount_t *mp = ip->i_mount;
+ xfs_fileoff_t start_zero_fsb;
+ xfs_fileoff_t end_zero_fsb;
+ xfs_fileoff_t zero_count_fsb;
+ xfs_fileoff_t last_fsb;
+ xfs_fileoff_t zero_off;
+ xfs_fsize_t zero_len;
+ int nimaps;
+ int error = 0;
+ xfs_bmbt_irec_t imap;
+
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ ASSERT(offset > isize);
+
+ /*
+ * First handle zeroing the block on which isize resides.
+ * We only zero a part of that block so it is handled specially.
+ */
+ error = xfs_zero_last_block(ip, offset, isize);
+ if (error) {
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ return error;
+ }
+
+ /*
+ * Calculate the range between the new size and the old
+ * where blocks needing to be zeroed may exist. To get the
+ * block where the last byte in the file currently resides,
+ * we need to subtract one from the size and truncate back
+ * to a block boundary. We subtract 1 in case the size is
+ * exactly on a block boundary.
+ */
+ last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
+ start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
+ end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
+ ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
+ if (last_fsb == end_zero_fsb) {
+ /*
+ * The size was only incremented on its last block.
+ * We took care of that above, so just return.
+ */
+ return 0;
+ }
+
+ ASSERT(start_zero_fsb <= end_zero_fsb);
+ while (start_zero_fsb <= end_zero_fsb) {
+ nimaps = 1;
+ zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
+ error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
+ 0, NULL, 0, &imap, &nimaps, NULL, NULL);
+ if (error) {
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ return error;
+ }
+ ASSERT(nimaps > 0);
+
+ if (imap.br_state == XFS_EXT_UNWRITTEN ||
+ imap.br_startblock == HOLESTARTBLOCK) {
+ /*
+ * This loop handles initializing pages that were
+ * partially initialized by the code below this
+ * loop. It basically zeroes the part of the page
+ * that sits on a hole and sets the page as P_HOLE
+ * and calls remapf if it is a mapped file.
+ */
+ start_zero_fsb = imap.br_startoff + imap.br_blockcount;
+ ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
+ continue;
+ }
+
+ /*
+ * There are blocks we need to zero.
+ * Drop the inode lock while we're doing the I/O.
+ * We'll still have the iolock to protect us.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
+ zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
+
+ if ((zero_off + zero_len) > offset)
+ zero_len = offset - zero_off;
+
+ error = xfs_iozero(ip, zero_off, zero_len);
+ if (error) {
+ goto out_lock;
+ }
+
+ start_zero_fsb = imap.br_startoff + imap.br_blockcount;
+ ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ }
+
+ return 0;
+
+out_lock:
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ASSERT(error >= 0);
+ return error;
+}
+
+STATIC ssize_t
+xfs_file_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ ssize_t ret = 0, error = 0;
+ int ioflags = 0;
+ xfs_fsize_t isize, new_size;
+ int iolock;
+ int eventsent = 0;
+ size_t ocount = 0, count;
+ int need_i_mutex;
+
+ XFS_STATS_INC(xs_write_calls);
+
+ BUG_ON(iocb->ki_pos != pos);
+
+ if (unlikely(file->f_flags & O_DIRECT))
+ ioflags |= IO_ISDIRECT;
+ if (file->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+
+ error = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
+ if (error)
+ return error;
+
+ count = ocount;
+ if (count == 0)
+ return 0;
+
+ xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return -EIO;
+
+relock:
+ if (ioflags & IO_ISDIRECT) {
+ iolock = XFS_IOLOCK_SHARED;
+ need_i_mutex = 0;
+ } else {
+ iolock = XFS_IOLOCK_EXCL;
+ need_i_mutex = 1;
+ mutex_lock(&inode->i_mutex);
+ }
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
+
+start:
+ error = -generic_write_checks(file, &pos, &count,
+ S_ISBLK(inode->i_mode));
+ if (error) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ goto out_unlock_mutex;
+ }
+
+ if ((DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) &&
+ !(ioflags & IO_INVIS) && !eventsent)) {
+ int dmflags = FILP_DELAY_FLAG(file);
+
+ if (need_i_mutex)
+ dmflags |= DM_FLAGS_IMUX;
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ error = XFS_SEND_DATA(ip->i_mount, DM_EVENT_WRITE, ip,
+ pos, count, dmflags, &iolock);
+ if (error) {
+ goto out_unlock_internal;
+ }
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ eventsent = 1;
+
+ /*
+ * The iolock was dropped and reacquired in XFS_SEND_DATA
+ * so we have to recheck the size when appending.
+ * We will only "goto start;" once, since having sent the
+ * event prevents another call to XFS_SEND_DATA, which is
+ * what allows the size to change in the first place.
+ */
+ if ((file->f_flags & O_APPEND) && pos != ip->i_size)
+ goto start;
+ }
+
+ if (ioflags & IO_ISDIRECT) {
+ xfs_buftarg_t *target =
+ XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+
+ if ((pos & target->bt_smask) || (count & target->bt_smask)) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ return XFS_ERROR(-EINVAL);
+ }
+
+ if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ iolock = XFS_IOLOCK_EXCL;
+ need_i_mutex = 1;
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
+ goto start;
+ }
+ }
+
+ new_size = pos + count;
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
+
+ if (likely(!(ioflags & IO_INVIS)))
+ file_update_time(file);
+
+ /*
+ * If the offset is beyond the size of the file, we have a couple
+ * of things to do. First, if there is already space allocated
+ * we need to either create holes or zero the disk or ...
+ *
+ * If there is a page where the previous size lands, we need
+ * to zero it out up to the new size.
+ */
+
+ if (pos > ip->i_size) {
+ error = xfs_zero_eof(ip, pos, ip->i_size);
+ if (error) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ goto out_unlock_internal;
+ }
+ }
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ /*
+ * If we're writing the file then make sure to clear the
+ * setuid and setgid bits if the process is not being run
+ * by root. This keeps people from modifying setuid and
+ * setgid binaries.
+ */
+ error = -file_remove_suid(file);
+ if (unlikely(error))
+ goto out_unlock_internal;
+
+ /* We can write back this queue in page reclaim */
+ current->backing_dev_info = mapping->backing_dev_info;
+
+ if ((ioflags & IO_ISDIRECT)) {
+ if (mapping->nrpages) {
+ WARN_ON(need_i_mutex == 0);
+ error = xfs_flushinval_pages(ip,
+ (pos & PAGE_CACHE_MASK),
+ -1, FI_REMAPF_LOCKED);
+ if (error)
+ goto out_unlock_internal;
+ }
+
+ if (need_i_mutex) {
+ /* demote the lock now the cached pages are gone */
+ xfs_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ mutex_unlock(&inode->i_mutex);
+
+ iolock = XFS_IOLOCK_SHARED;
+ need_i_mutex = 0;
+ }
+
+ trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
+ ret = generic_file_direct_write(iocb, iovp,
+ &nr_segs, pos, &iocb->ki_pos, count, ocount);
+
+ /*
+ * direct-io write to a hole: fall through to buffered I/O
+ * for completing the rest of the request.
+ */
+ if (ret >= 0 && ret != count) {
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ pos += ret;
+ count -= ret;
+
+ ioflags &= ~IO_ISDIRECT;
+ xfs_iunlock(ip, iolock);
+ goto relock;
+ }
+ } else {
+ int enospc = 0;
+ ssize_t ret2 = 0;
+
+write_retry:
+ trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags);
+ ret2 = generic_file_buffered_write(iocb, iovp, nr_segs,
+ pos, &iocb->ki_pos, count, ret);
+ /*
+ * if we just got an ENOSPC, flush the inode now we
+ * aren't holding any page locks and retry *once*
+ */
+ if (ret2 == -ENOSPC && !enospc) {
+ error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
+ if (error)
+ goto out_unlock_internal;
+ enospc = 1;
+ goto write_retry;
+ }
+ ret = ret2;
+ }
+
+ current->backing_dev_info = NULL;
+
+ isize = i_size_read(inode);
+ if (unlikely(ret < 0 && ret != -EFAULT && iocb->ki_pos > isize))
+ iocb->ki_pos = isize;
+
+ if (iocb->ki_pos > ip->i_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (iocb->ki_pos > ip->i_size)
+ ip->i_size = iocb->ki_pos;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+
+ if (ret == -ENOSPC &&
+ DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
+ xfs_iunlock(ip, iolock);
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+ error = XFS_SEND_NAMESP(ip->i_mount, DM_EVENT_NOSPACE, ip,
+ DM_RIGHT_NULL, ip, DM_RIGHT_NULL, NULL, NULL,
+ 0, 0, 0); /* Delay flag intentionally unused */
+ if (need_i_mutex)
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, iolock);
+ if (error)
+ goto out_unlock_internal;
+ goto start;
+ }
+
+ error = -ret;
+ if (ret <= 0)
+ goto out_unlock_internal;
+
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ /* Handle various SYNC-type writes */
+ if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
+ loff_t end = pos + ret - 1;
+ int error2;
+
+ xfs_iunlock(ip, iolock);
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+
+ error2 = filemap_write_and_wait_range(mapping, pos, end);
+ if (!error)
+ error = error2;
+ if (need_i_mutex)
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, iolock);
+
+ error2 = -xfs_file_fsync(file, file->f_path.dentry,
+ (file->f_flags & __O_SYNC) ? 0 : 1);
+ if (!error)
+ error = error2;
+ }
+
+ out_unlock_internal:
+ if (ip->i_new_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ /*
+ * If this was a direct or synchronous I/O that failed (such
+ * as ENOSPC) then part of the I/O may have been written to
+ * disk before the error occured. In this case the on-disk
+ * file size may have been adjusted beyond the in-memory file
+ * size and now needs to be truncated back.
+ */
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+ xfs_iunlock(ip, iolock);
+ out_unlock_mutex:
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+ return -error;
}
STATIC int
@@ -160,28 +938,6 @@ xfs_file_release(
return -xfs_release(XFS_I(inode));
}
-/*
- * We ignore the datasync flag here because a datasync is effectively
- * identical to an fsync. That is, datasync implies that we need to write
- * only the metadata needed to be able to access the data that is written
- * if we crash after the call completes. Hence if we are writing beyond
- * EOF we have to log the inode size change as well, which makes it a
- * full fsync. If we don't write beyond EOF, the inode core will be
- * clean in memory and so we don't need to log the inode, just like
- * fsync.
- */
-STATIC int
-xfs_file_fsync(
- struct file *file,
- struct dentry *dentry,
- int datasync)
-{
- struct xfs_inode *ip = XFS_I(dentry->d_inode);
-
- xfs_iflags_clear(ip, XFS_ITRUNCATED);
- return -xfs_fsync(ip);
-}
-
STATIC int
xfs_file_readdir(
struct file *filp,
@@ -203,9 +959,9 @@ xfs_file_readdir(
*
* Try to give it an estimate that's good enough, maybe at some
* point we can change the ->readdir prototype to include the
- * buffer size.
+ * buffer size. For now we use the current glibc buffer size.
*/
- bufsize = (size_t)min_t(loff_t, PAGE_SIZE, ip->i_d.di_size);
+ bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size);
error = xfs_readdir(ip, dirent, bufsize,
(xfs_off_t *)&filp->f_pos, filldir);
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index e8566bbf0f00..61a99608731e 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -91,6 +91,16 @@ xfs_mark_inode_dirty_sync(
mark_inode_dirty_sync(inode);
}
+void
+xfs_mark_inode_dirty(
+ xfs_inode_t *ip)
+{
+ struct inode *inode = VFS_I(ip);
+
+ if (!(inode->i_state & (I_WILL_FREE|I_FREEING|I_CLEAR)))
+ mark_inode_dirty(inode);
+}
+
/*
* Change the requested timestamp in the given inode.
* We don't lock across timestamp updates, and we don't log them but
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 5af0c81ca1ae..facfb323a706 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -88,7 +88,6 @@
#include <xfs_super.h>
#include <xfs_globals.h>
#include <xfs_fs_subr.h>
-#include <xfs_lrw.h>
#include <xfs_buf.h>
/*
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
deleted file mode 100644
index eac6f80d786d..000000000000
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ /dev/null
@@ -1,796 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * 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.
- *
- * This program is distributed in the hope that it would 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 the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_ag.h"
-#include "xfs_dir2.h"
-#include "xfs_alloc.h"
-#include "xfs_dmapi.h"
-#include "xfs_quota.h"
-#include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
-#include "xfs_bmap.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
-#include "xfs_rtalloc.h"
-#include "xfs_error.h"
-#include "xfs_itable.h"
-#include "xfs_rw.h"
-#include "xfs_attr.h"
-#include "xfs_inode_item.h"
-#include "xfs_buf_item.h"
-#include "xfs_utils.h"
-#include "xfs_iomap.h"
-#include "xfs_vnodeops.h"
-#include "xfs_trace.h"
-
-#include <linux/capability.h>
-#include <linux/writeback.h>
-
-
-/*
- * xfs_iozero
- *
- * xfs_iozero clears the specified range of buffer supplied,
- * and marks all the affected blocks as valid and modified. If
- * an affected block is not allocated, it will be allocated. If
- * an affected block is not completely overwritten, and is not
- * valid before the operation, it will be read from disk before
- * being partially zeroed.
- */
-STATIC int
-xfs_iozero(
- struct xfs_inode *ip, /* inode */
- loff_t pos, /* offset in file */
- size_t count) /* size of data to zero */
-{
- struct page *page;
- struct address_space *mapping;
- int status;
-
- mapping = VFS_I(ip)->i_mapping;
- do {
- unsigned offset, bytes;
- void *fsdata;
-
- offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
- bytes = PAGE_CACHE_SIZE - offset;
- if (bytes > count)
- bytes = count;
-
- status = pagecache_write_begin(NULL, mapping, pos, bytes,
- AOP_FLAG_UNINTERRUPTIBLE,
- &page, &fsdata);
- if (status)
- break;
-
- zero_user(page, offset, bytes);
-
- status = pagecache_write_end(NULL, mapping, pos, bytes, bytes,
- page, fsdata);
- WARN_ON(status <= 0); /* can't return less than zero! */
- pos += bytes;
- count -= bytes;
- status = 0;
- } while (count);
-
- return (-status);
-}
-
-ssize_t /* bytes read, or (-) error */
-xfs_read(
- xfs_inode_t *ip,
- struct kiocb *iocb,
- const struct iovec *iovp,
- unsigned int segs,
- loff_t *offset,
- int ioflags)
-{
- struct file *file = iocb->ki_filp;
- struct inode *inode = file->f_mapping->host;
- xfs_mount_t *mp = ip->i_mount;
- size_t size = 0;
- ssize_t ret = 0;
- xfs_fsize_t n;
- unsigned long seg;
-
-
- XFS_STATS_INC(xs_read_calls);
-
- /* START copy & waste from filemap.c */
- for (seg = 0; seg < segs; seg++) {
- const struct iovec *iv = &iovp[seg];
-
- /*
- * If any segment has a negative length, or the cumulative
- * length ever wraps negative then return -EINVAL.
- */
- size += iv->iov_len;
- if (unlikely((ssize_t)(size|iv->iov_len) < 0))
- return XFS_ERROR(-EINVAL);
- }
- /* END copy & waste from filemap.c */
-
- if (unlikely(ioflags & IO_ISDIRECT)) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(ip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
- if ((*offset & target->bt_smask) ||
- (size & target->bt_smask)) {
- if (*offset == ip->i_size) {
- return (0);
- }
- return -XFS_ERROR(EINVAL);
- }
- }
-
- n = XFS_MAXIOFFSET(mp) - *offset;
- if ((n <= 0) || (size == 0))
- return 0;
-
- if (n < size)
- size = n;
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
-
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
- int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);
- int iolock = XFS_IOLOCK_SHARED;
-
- ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *offset, size,
- dmflags, &iolock);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_unlock(&inode->i_mutex);
- return ret;
- }
- }
-
- if (unlikely(ioflags & IO_ISDIRECT)) {
- if (inode->i_mapping->nrpages)
- ret = -xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- mutex_unlock(&inode->i_mutex);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
- }
- }
-
- trace_xfs_file_read(ip, size, *offset, ioflags);
-
- iocb->ki_pos = *offset;
- ret = generic_file_aio_read(iocb, iovp, segs, *offset);
- if (ret > 0)
- XFS_STATS_ADD(xs_read_bytes, ret);
-
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
-}
-
-ssize_t
-xfs_splice_read(
- xfs_inode_t *ip,
- struct file *infilp,
- loff_t *ppos,
- struct pipe_inode_info *pipe,
- size_t count,
- int flags,
- int ioflags)
-{
- xfs_mount_t *mp = ip->i_mount;
- ssize_t ret;
-
- XFS_STATS_INC(xs_read_calls);
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return -EIO;
-
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
- int iolock = XFS_IOLOCK_SHARED;
- int error;
-
- error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
- FILP_DELAY_FLAG(infilp), &iolock);
- if (error) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return -error;
- }
- }
-
- trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
-
- ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_read_bytes, ret);
-
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
-}
-
-ssize_t
-xfs_splice_write(
- xfs_inode_t *ip,
- struct pipe_inode_info *pipe,
- struct file *outfilp,
- loff_t *ppos,
- size_t count,
- int flags,
- int ioflags)
-{
- xfs_mount_t *mp = ip->i_mount;
- ssize_t ret;
- struct inode *inode = outfilp->f_mapping->host;
- xfs_fsize_t isize, new_size;
-
- XFS_STATS_INC(xs_write_calls);
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return -EIO;
-
- xfs_ilock(ip, XFS_IOLOCK_EXCL);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) {
- int iolock = XFS_IOLOCK_EXCL;
- int error;
-
- error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count,
- FILP_DELAY_FLAG(outfilp), &iolock);
- if (error) {
- xfs_iunlock(ip, XFS_IOLOCK_EXCL);
- return -error;
- }
- }
-
- new_size = *ppos + count;
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (new_size > ip->i_size)
- ip->i_new_size = new_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
-
- ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
- *ppos = isize;
-
- if (*ppos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (*ppos > ip->i_size)
- ip->i_size = *ppos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
- xfs_iunlock(ip, XFS_IOLOCK_EXCL);
- return ret;
-}
-
-/*
- * This routine is called to handle zeroing any space in the last
- * block of the file that is beyond the EOF. We do this since the
- * size is being increased without writing anything to that block
- * and we don't want anyone to read the garbage on the disk.
- */
-STATIC int /* error (positive) */
-xfs_zero_last_block(
- xfs_inode_t *ip,
- xfs_fsize_t offset,
- xfs_fsize_t isize)
-{
- xfs_fileoff_t last_fsb;
- xfs_mount_t *mp = ip->i_mount;
- int nimaps;
- int zero_offset;
- int zero_len;
- int error = 0;
- xfs_bmbt_irec_t imap;
-
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
- zero_offset = XFS_B_FSB_OFFSET(mp, isize);
- if (zero_offset == 0) {
- /*
- * There are no extra bytes in the last block on disk to
- * zero, so return.
- */
- return 0;
- }
-
- last_fsb = XFS_B_TO_FSBT(mp, isize);
- nimaps = 1;
- error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
- &nimaps, NULL, NULL);
- if (error) {
- return error;
- }
- ASSERT(nimaps > 0);
- /*
- * If the block underlying isize is just a hole, then there
- * is nothing to zero.
- */
- if (imap.br_startblock == HOLESTARTBLOCK) {
- return 0;
- }
- /*
- * Zero the part of the last block beyond the EOF, and write it
- * out sync. We need to drop the ilock while we do this so we
- * don't deadlock when the buffer cache calls back to us.
- */
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- zero_len = mp->m_sb.sb_blocksize - zero_offset;
- if (isize + zero_len > offset)
- zero_len = offset - isize;
- error = xfs_iozero(ip, isize, zero_len);
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ASSERT(error >= 0);
- return error;
-}
-
-/*
- * Zero any on disk space between the current EOF and the new,
- * larger EOF. This handles the normal case of zeroing the remainder
- * of the last block in the file and the unusual case of zeroing blocks
- * out beyond the size of the file. This second case only happens
- * with fixed size extents and when the system crashes before the inode
- * size was updated but after blocks were allocated. If fill is set,
- * then any holes in the range are filled and zeroed. If not, the holes
- * are left alone as holes.
- */
-
-int /* error (positive) */
-xfs_zero_eof(
- xfs_inode_t *ip,
- xfs_off_t offset, /* starting I/O offset */
- xfs_fsize_t isize) /* current inode size */
-{
- xfs_mount_t *mp = ip->i_mount;
- xfs_fileoff_t start_zero_fsb;
- xfs_fileoff_t end_zero_fsb;
- xfs_fileoff_t zero_count_fsb;
- xfs_fileoff_t last_fsb;
- xfs_fileoff_t zero_off;
- xfs_fsize_t zero_len;
- int nimaps;
- int error = 0;
- xfs_bmbt_irec_t imap;
-
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- ASSERT(offset > isize);
-
- /*
- * First handle zeroing the block on which isize resides.
- * We only zero a part of that block so it is handled specially.
- */
- error = xfs_zero_last_block(ip, offset, isize);
- if (error) {
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- return error;
- }
-
- /*
- * Calculate the range between the new size and the old
- * where blocks needing to be zeroed may exist. To get the
- * block where the last byte in the file currently resides,
- * we need to subtract one from the size and truncate back
- * to a block boundary. We subtract 1 in case the size is
- * exactly on a block boundary.
- */
- last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
- start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
- end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
- ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
- if (last_fsb == end_zero_fsb) {
- /*
- * The size was only incremented on its last block.
- * We took care of that above, so just return.
- */
- return 0;
- }
-
- ASSERT(start_zero_fsb <= end_zero_fsb);
- while (start_zero_fsb <= end_zero_fsb) {
- nimaps = 1;
- zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
- error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
- 0, NULL, 0, &imap, &nimaps, NULL, NULL);
- if (error) {
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- return error;
- }
- ASSERT(nimaps > 0);
-
- if (imap.br_state == XFS_EXT_UNWRITTEN ||
- imap.br_startblock == HOLESTARTBLOCK) {
- /*
- * This loop handles initializing pages that were
- * partially initialized by the code below this
- * loop. It basically zeroes the part of the page
- * that sits on a hole and sets the page as P_HOLE
- * and calls remapf if it is a mapped file.
- */
- start_zero_fsb = imap.br_startoff + imap.br_blockcount;
- ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
- continue;
- }
-
- /*
- * There are blocks we need to zero.
- * Drop the inode lock while we're doing the I/O.
- * We'll still have the iolock to protect us.
- */
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
- zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
-
- if ((zero_off + zero_len) > offset)
- zero_len = offset - zero_off;
-
- error = xfs_iozero(ip, zero_off, zero_len);
- if (error) {
- goto out_lock;
- }
-
- start_zero_fsb = imap.br_startoff + imap.br_blockcount;
- ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- }
-
- return 0;
-
-out_lock:
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ASSERT(error >= 0);
- return error;
-}
-
-ssize_t /* bytes written, or (-) error */
-xfs_write(
- struct xfs_inode *xip,
- struct kiocb *iocb,
- const struct iovec *iovp,
- unsigned int nsegs,
- loff_t *offset,
- int ioflags)
-{
- struct file *file = iocb->ki_filp;
- struct address_space *mapping = file->f_mapping;
- struct inode *inode = mapping->host;
- unsigned long segs = nsegs;
- xfs_mount_t *mp;
- ssize_t ret = 0, error = 0;
- xfs_fsize_t isize, new_size;
- int iolock;
- int eventsent = 0;
- size_t ocount = 0, count;
- loff_t pos;
- int need_i_mutex;
-
- XFS_STATS_INC(xs_write_calls);
-
- error = generic_segment_checks(iovp, &segs, &ocount, VERIFY_READ);
- if (error)
- return error;
-
- count = ocount;
- pos = *offset;
-
- if (count == 0)
- return 0;
-
- mp = xip->i_mount;
-
- xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
-
-relock:
- if (ioflags & IO_ISDIRECT) {
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- } else {
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- }
-
- xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);
-
-start:
- error = -generic_write_checks(file, &pos, &count,
- S_ISBLK(inode->i_mode));
- if (error) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_mutex;
- }
-
- if ((DM_EVENT_ENABLED(xip, DM_EVENT_WRITE) &&
- !(ioflags & IO_INVIS) && !eventsent)) {
- int dmflags = FILP_DELAY_FLAG(file);
-
- if (need_i_mutex)
- dmflags |= DM_FLAGS_IMUX;
-
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, xip,
- pos, count, dmflags, &iolock);
- if (error) {
- goto out_unlock_internal;
- }
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- eventsent = 1;
-
- /*
- * The iolock was dropped and reacquired in XFS_SEND_DATA
- * so we have to recheck the size when appending.
- * We will only "goto start;" once, since having sent the
- * event prevents another call to XFS_SEND_DATA, which is
- * what allows the size to change in the first place.
- */
- if ((file->f_flags & O_APPEND) && pos != xip->i_size)
- goto start;
- }
-
- if (ioflags & IO_ISDIRECT) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(xip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
-
- if ((pos & target->bt_smask) || (count & target->bt_smask)) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- return XFS_ERROR(-EINVAL);
- }
-
- if (!need_i_mutex && (mapping->nrpages || pos > xip->i_size)) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);
- goto start;
- }
- }
-
- new_size = pos + count;
- if (new_size > xip->i_size)
- xip->i_new_size = new_size;
-
- if (likely(!(ioflags & IO_INVIS)))
- file_update_time(file);
-
- /*
- * If the offset is beyond the size of the file, we have a couple
- * of things to do. First, if there is already space allocated
- * we need to either create holes or zero the disk or ...
- *
- * If there is a page where the previous size lands, we need
- * to zero it out up to the new size.
- */
-
- if (pos > xip->i_size) {
- error = xfs_zero_eof(xip, pos, xip->i_size);
- if (error) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- goto out_unlock_internal;
- }
- }
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
-
- /*
- * If we're writing the file then make sure to clear the
- * setuid and setgid bits if the process is not being run
- * by root. This keeps people from modifying setuid and
- * setgid binaries.
- */
- error = -file_remove_suid(file);
- if (unlikely(error))
- goto out_unlock_internal;
-
- /* We can write back this queue in page reclaim */
- current->backing_dev_info = mapping->backing_dev_info;
-
- if ((ioflags & IO_ISDIRECT)) {
- if (mapping->nrpages) {
- WARN_ON(need_i_mutex == 0);
- error = xfs_flushinval_pages(xip,
- (pos & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- if (error)
- goto out_unlock_internal;
- }
-
- if (need_i_mutex) {
- /* demote the lock now the cached pages are gone */
- xfs_ilock_demote(xip, XFS_IOLOCK_EXCL);
- mutex_unlock(&inode->i_mutex);
-
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- }
-
- trace_xfs_file_direct_write(xip, count, *offset, ioflags);
- ret = generic_file_direct_write(iocb, iovp,
- &segs, pos, offset, count, ocount);
-
- /*
- * direct-io write to a hole: fall through to buffered I/O
- * for completing the rest of the request.
- */
- if (ret >= 0 && ret != count) {
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- pos += ret;
- count -= ret;
-
- ioflags &= ~IO_ISDIRECT;
- xfs_iunlock(xip, iolock);
- goto relock;
- }
- } else {
- int enospc = 0;
- ssize_t ret2 = 0;
-
-write_retry:
- trace_xfs_file_buffered_write(xip, count, *offset, ioflags);
- ret2 = generic_file_buffered_write(iocb, iovp, segs,
- pos, offset, count, ret);
- /*
- * if we just got an ENOSPC, flush the inode now we
- * aren't holding any page locks and retry *once*
- */
- if (ret2 == -ENOSPC && !enospc) {
- error = xfs_flush_pages(xip, 0, -1, 0, FI_NONE);
- if (error)
- goto out_unlock_internal;
- enospc = 1;
- goto write_retry;
- }
- ret = ret2;
- }
-
- current->backing_dev_info = NULL;
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
- *offset = isize;
-
- if (*offset > xip->i_size) {
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- if (*offset > xip->i_size)
- xip->i_size = *offset;
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- }
-
- if (ret == -ENOSPC &&
- DM_EVENT_ENABLED(xip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
- xfs_iunlock(xip, iolock);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, xip,
- DM_RIGHT_NULL, xip, DM_RIGHT_NULL, NULL, NULL,
- 0, 0, 0); /* Delay flag intentionally unused */
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, iolock);
- if (error)
- goto out_unlock_internal;
- goto start;
- }
-
- error = -ret;
- if (ret <= 0)
- goto out_unlock_internal;
-
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- /* Handle various SYNC-type writes */
- if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
- loff_t end = pos + ret - 1;
- int error2;
-
- xfs_iunlock(xip, iolock);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
-
- error2 = filemap_write_and_wait_range(mapping, pos, end);
- if (!error)
- error = error2;
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, iolock);
-
- error2 = xfs_fsync(xip);
- if (!error)
- error = error2;
- }
-
- out_unlock_internal:
- if (xip->i_new_size) {
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- xip->i_new_size = 0;
- /*
- * If this was a direct or synchronous I/O that failed (such
- * as ENOSPC) then part of the I/O may have been written to
- * disk before the error occured. In this case the on-disk
- * file size may have been adjusted beyond the in-memory file
- * size and now needs to be truncated back.
- */
- if (xip->i_d.di_size > xip->i_size)
- xip->i_d.di_size = xip->i_size;
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- }
- xfs_iunlock(xip, iolock);
- out_unlock_mutex:
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- return -error;
-}
-
-/*
- * If the underlying (data/log/rt) device is readonly, there are some
- * operations that cannot proceed.
- */
-int
-xfs_dev_is_read_only(
- xfs_mount_t *mp,
- char *message)
-{
- if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
- xfs_readonly_buftarg(mp->m_logdev_targp) ||
- (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
- cmn_err(CE_NOTE,
- "XFS: %s required on read-only device.", message);
- cmn_err(CE_NOTE,
- "XFS: write access unavailable, cannot proceed.");
- return EROFS;
- }
- return 0;
-}
diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h
deleted file mode 100644
index 342ae8c0d011..000000000000
--- a/fs/xfs/linux-2.6/xfs_lrw.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * 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.
- *
- * This program is distributed in the hope that it would 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 the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_LRW_H__
-#define __XFS_LRW_H__
-
-struct xfs_mount;
-struct xfs_inode;
-struct xfs_buf;
-
-extern int xfs_dev_is_read_only(struct xfs_mount *, char *);
-
-extern int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
-
-#endif /* __XFS_LRW_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index a9f6d20aff41..05cd85317f6f 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -607,7 +607,8 @@ xfssyncd(
set_freezable();
timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
for (;;) {
- timeleft = schedule_timeout_interruptible(timeleft);
+ if (list_empty(&mp->m_sync_list))
+ timeleft = schedule_timeout_interruptible(timeleft);
/* swsusp */
try_to_freeze();
if (kthread_should_stop() && list_empty(&mp->m_sync_list))
@@ -627,8 +628,7 @@ xfssyncd(
list_add_tail(&mp->m_sync_work.w_list,
&mp->m_sync_list);
}
- list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list)
- list_move(&work->w_list, &tmp);
+ list_splice_init(&mp->m_sync_list, &tmp);
spin_unlock(&mp->m_sync_lock);
list_for_each_entry_safe(work, n, &tmp, w_list) {
@@ -688,12 +688,12 @@ xfs_inode_set_reclaim_tag(
struct xfs_perag *pag;
pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
- read_lock(&pag->pag_ici_lock);
+ write_lock(&pag->pag_ici_lock);
spin_lock(&ip->i_flags_lock);
__xfs_inode_set_reclaim_tag(pag, ip);
__xfs_iflags_set(ip, XFS_IRECLAIMABLE);
spin_unlock(&ip->i_flags_lock);
- read_unlock(&pag->pag_ici_lock);
+ write_unlock(&pag->pag_ici_lock);
xfs_perag_put(pag);
}
diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c
index 856eb3c8d605..5a107601e969 100644
--- a/fs/xfs/linux-2.6/xfs_trace.c
+++ b/fs/xfs/linux-2.6/xfs_trace.c
@@ -52,22 +52,6 @@
#include "quota/xfs_dquot.h"
/*
- * Format fsblock number into a static buffer & return it.
- */
-STATIC char *xfs_fmtfsblock(xfs_fsblock_t bno)
-{
- static char rval[50];
-
- if (bno == NULLFSBLOCK)
- sprintf(rval, "NULLFSBLOCK");
- else if (isnullstartblock(bno))
- sprintf(rval, "NULLSTARTBLOCK(%lld)", startblockval(bno));
- else
- sprintf(rval, "%lld", (xfs_dfsbno_t)bno);
- return rval;
-}
-
-/*
* We include this last to have the helpers above available for the trace
* event implementations.
*/
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index a4574dcf5065..fcaa62f0799e 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -197,13 +197,13 @@ TRACE_EVENT(xfs_iext_insert,
__entry->caller_ip = caller_ip;
),
TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
- "offset %lld block %s count %lld flag %d caller %pf",
+ "offset %lld block %lld count %lld flag %d caller %pf",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->ino,
__print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
(long)__entry->idx,
__entry->startoff,
- xfs_fmtfsblock(__entry->startblock),
+ (__int64_t)__entry->startblock,
__entry->blockcount,
__entry->state,
(char *)__entry->caller_ip)
@@ -241,13 +241,13 @@ DECLARE_EVENT_CLASS(xfs_bmap_class,
__entry->caller_ip = caller_ip;
),
TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
- "offset %lld block %s count %lld flag %d caller %pf",
+ "offset %lld block %lld count %lld flag %d caller %pf",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->ino,
__print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
(long)__entry->idx,
__entry->startoff,
- xfs_fmtfsblock(__entry->startblock),
+ (__int64_t)__entry->startblock,
__entry->blockcount,
__entry->state,
(char *)__entry->caller_ip)
@@ -593,7 +593,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
TP_ARGS(dqp),
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(__be32, id)
+ __field(u32, id)
__field(unsigned, flags)
__field(unsigned, nrefs)
__field(unsigned long long, res_bcount)
@@ -606,7 +606,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
), \
TP_fast_assign(
__entry->dev = dqp->q_mount->m_super->s_dev;
- __entry->id = dqp->q_core.d_id;
+ __entry->id = be32_to_cpu(dqp->q_core.d_id);
__entry->flags = dqp->dq_flags;
__entry->nrefs = dqp->q_nrefs;
__entry->res_bcount = dqp->q_res_bcount;
@@ -622,10 +622,10 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
be64_to_cpu(dqp->q_core.d_ino_softlimit);
),
TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx "
- "bcnt 0x%llx [hard 0x%llx | soft 0x%llx] "
- "icnt 0x%llx [hard 0x%llx | soft 0x%llx]",
+ "bcnt 0x%llx bhardlimit 0x%llx bsoftlimit 0x%llx "
+ "icnt 0x%llx ihardlimit 0x%llx isoftlimit 0x%llx]",
MAJOR(__entry->dev), MINOR(__entry->dev),
- be32_to_cpu(__entry->id),
+ __entry->id,
__print_flags(__entry->flags, "|", XFS_DQ_FLAGS),
__entry->nrefs,
__entry->res_bcount,
@@ -881,7 +881,7 @@ TRACE_EVENT(name, \
), \
TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
"offset 0x%llx count %zd flags %s " \
- "startoff 0x%llx startblock %s blockcount 0x%llx", \
+ "startoff 0x%llx startblock %lld blockcount 0x%llx", \
MAJOR(__entry->dev), MINOR(__entry->dev), \
__entry->ino, \
__entry->size, \
@@ -890,7 +890,7 @@ TRACE_EVENT(name, \
__entry->count, \
__print_flags(__entry->flags, "|", BMAPI_FLAGS), \
__entry->startoff, \
- xfs_fmtfsblock(__entry->startblock), \
+ (__int64_t)__entry->startblock, \
__entry->blockcount) \
)
DEFINE_IOMAP_EVENT(xfs_iomap_enter);
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 1869fb973819..5c11e4d17010 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -2550,22 +2550,134 @@ xfs_bmap_rtalloc(
}
STATIC int
+xfs_bmap_btalloc_nullfb(
+ struct xfs_bmalloca *ap,
+ struct xfs_alloc_arg *args,
+ xfs_extlen_t *blen)
+{
+ struct xfs_mount *mp = ap->ip->i_mount;
+ struct xfs_perag *pag;
+ xfs_agnumber_t ag, startag;
+ int notinit = 0;
+ int error;
+
+ if (ap->userdata && xfs_inode_is_filestream(ap->ip))
+ args->type = XFS_ALLOCTYPE_NEAR_BNO;
+ else
+ args->type = XFS_ALLOCTYPE_START_BNO;
+ args->total = ap->total;
+
+ /*
+ * Search for an allocation group with a single extent large enough
+ * for the request. If one isn't found, then adjust the minimum
+ * allocation size to the largest space found.
+ */
+ startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
+ if (startag == NULLAGNUMBER)
+ startag = ag = 0;
+
+ pag = xfs_perag_get(mp, ag);
+ while (*blen < ap->alen) {
+ if (!pag->pagf_init) {
+ error = xfs_alloc_pagf_init(mp, args->tp, ag,
+ XFS_ALLOC_FLAG_TRYLOCK);
+ if (error) {
+ xfs_perag_put(pag);
+ return error;
+ }
+ }
+
+ /*
+ * See xfs_alloc_fix_freelist...
+ */
+ if (pag->pagf_init) {
+ xfs_extlen_t longest;
+ longest = xfs_alloc_longest_free_extent(mp, pag);
+ if (*blen < longest)
+ *blen = longest;
+ } else
+ notinit = 1;
+
+ if (xfs_inode_is_filestream(ap->ip)) {
+ if (*blen >= ap->alen)
+ break;
+
+ if (ap->userdata) {
+ /*
+ * If startag is an invalid AG, we've
+ * come here once before and
+ * xfs_filestream_new_ag picked the
+ * best currently available.
+ *
+ * Don't continue looping, since we
+ * could loop forever.
+ */
+ if (startag == NULLAGNUMBER)
+ break;
+
+ error = xfs_filestream_new_ag(ap, &ag);
+ xfs_perag_put(pag);
+ if (error)
+ return error;
+
+ /* loop again to set 'blen'*/
+ startag = NULLAGNUMBER;
+ pag = xfs_perag_get(mp, ag);
+ continue;
+ }
+ }
+ if (++ag == mp->m_sb.sb_agcount)
+ ag = 0;
+ if (ag == startag)
+ break;
+ xfs_perag_put(pag);
+ pag = xfs_perag_get(mp, ag);
+ }
+ xfs_perag_put(pag);
+
+ /*
+ * Since the above loop did a BUF_TRYLOCK, it is
+ * possible that there is space for this request.
+ */
+ if (notinit || *blen < ap->minlen)
+ args->minlen = ap->minlen;
+ /*
+ * If the best seen length is less than the request
+ * length, use the best as the minimum.
+ */
+ else if (*blen < ap->alen)
+ args->minlen = *blen;
+ /*
+ * Otherwise we've seen an extent as big as alen,
+ * use that as the minimum.
+ */
+ else
+ args->minlen = ap->alen;
+
+ /*
+ * set the failure fallback case to look in the selected
+ * AG as the stream may have moved.
+ */
+ if (xfs_inode_is_filestream(ap->ip))
+ ap->rval = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+
+ return 0;
+}
+
+STATIC int
xfs_bmap_btalloc(
xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
xfs_mount_t *mp; /* mount point structure */
xfs_alloctype_t atype = 0; /* type for allocation routines */
xfs_extlen_t align; /* minimum allocation alignment */
- xfs_agnumber_t ag;
xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
- xfs_agnumber_t startag;
+ xfs_agnumber_t ag;
xfs_alloc_arg_t args;
xfs_extlen_t blen;
xfs_extlen_t nextminlen = 0;
- xfs_perag_t *pag;
int nullfb; /* true if ap->firstblock isn't set */
int isaligned;
- int notinit;
int tryagain;
int error;
@@ -2612,103 +2724,9 @@ xfs_bmap_btalloc(
args.firstblock = ap->firstblock;
blen = 0;
if (nullfb) {
- if (ap->userdata && xfs_inode_is_filestream(ap->ip))
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
- else
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.total = ap->total;
-
- /*
- * Search for an allocation group with a single extent
- * large enough for the request.
- *
- * If one isn't found, then adjust the minimum allocation
- * size to the largest space found.
- */
- startag = ag = XFS_FSB_TO_AGNO(mp, args.fsbno);
- if (startag == NULLAGNUMBER)
- startag = ag = 0;
- notinit = 0;
- pag = xfs_perag_get(mp, ag);
- while (blen < ap->alen) {
- if (!pag->pagf_init &&
- (error = xfs_alloc_pagf_init(mp, args.tp,
- ag, XFS_ALLOC_FLAG_TRYLOCK))) {
- xfs_perag_put(pag);
- return error;
- }
- /*
- * See xfs_alloc_fix_freelist...
- */
- if (pag->pagf_init) {
- xfs_extlen_t longest;
- longest = xfs_alloc_longest_free_extent(mp, pag);
- if (blen < longest)
- blen = longest;
- } else
- notinit = 1;
-
- if (xfs_inode_is_filestream(ap->ip)) {
- if (blen >= ap->alen)
- break;
-
- if (ap->userdata) {
- /*
- * If startag is an invalid AG, we've
- * come here once before and
- * xfs_filestream_new_ag picked the
- * best currently available.
- *
- * Don't continue looping, since we
- * could loop forever.
- */
- if (startag == NULLAGNUMBER)
- break;
-
- error = xfs_filestream_new_ag(ap, &ag);
- xfs_perag_put(pag);
- if (error)
- return error;
-
- /* loop again to set 'blen'*/
- startag = NULLAGNUMBER;
- pag = xfs_perag_get(mp, ag);
- continue;
- }
- }
- if (++ag == mp->m_sb.sb_agcount)
- ag = 0;
- if (ag == startag)
- break;
- xfs_perag_put(pag);
- pag = xfs_perag_get(mp, ag);
- }
- xfs_perag_put(pag);
- /*
- * Since the above loop did a BUF_TRYLOCK, it is
- * possible that there is space for this request.
- */
- if (notinit || blen < ap->minlen)
- args.minlen = ap->minlen;
- /*
- * If the best seen length is less than the request
- * length, use the best as the minimum.
- */
- else if (blen < ap->alen)
- args.minlen = blen;
- /*
- * Otherwise we've seen an extent as big as alen,
- * use that as the minimum.
- */
- else
- args.minlen = ap->alen;
-
- /*
- * set the failure fallback case to look in the selected
- * AG as the stream may have moved.
- */
- if (xfs_inode_is_filestream(ap->ip))
- ap->rval = args.fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+ error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
+ if (error)
+ return error;
} else if (ap->low) {
if (xfs_inode_is_filestream(ap->ip))
args.type = XFS_ALLOCTYPE_FIRST_AG;
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index f52ac276277e..7cf7220e7d5f 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -292,7 +292,8 @@ typedef struct xfs_bstat {
__s32 bs_extents; /* number of extents */
__u32 bs_gen; /* generation count */
__u16 bs_projid; /* project id */
- unsigned char bs_pad[14]; /* pad space, unused */
+ __u16 bs_forkoff; /* inode fork offset in bytes */
+ unsigned char bs_pad[12]; /* pad space, unused */
__u32 bs_dmevmask; /* DMIG event mask */
__u16 bs_dmstate; /* DMIG state info */
__u16 bs_aextents; /* attribute number of extents */
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index e281eb4a1c49..6845db90818f 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -190,13 +190,12 @@ xfs_iget_cache_hit(
trace_xfs_iget_reclaim(ip);
/*
- * We need to set XFS_INEW atomically with clearing the
- * reclaimable tag so that we do have an indicator of the
- * inode still being initialized.
+ * We need to set XFS_IRECLAIM to prevent xfs_reclaim_inode
+ * from stomping over us while we recycle the inode. We can't
+ * clear the radix tree reclaimable tag yet as it requires
+ * pag_ici_lock to be held exclusive.
*/
- ip->i_flags |= XFS_INEW;
- ip->i_flags &= ~XFS_IRECLAIMABLE;
- __xfs_inode_clear_reclaim_tag(mp, pag, ip);
+ ip->i_flags |= XFS_IRECLAIM;
spin_unlock(&ip->i_flags_lock);
read_unlock(&pag->pag_ici_lock);
@@ -216,7 +215,15 @@ xfs_iget_cache_hit(
trace_xfs_iget_reclaim(ip);
goto out_error;
}
+
+ write_lock(&pag->pag_ici_lock);
+ spin_lock(&ip->i_flags_lock);
+ ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM);
+ ip->i_flags |= XFS_INEW;
+ __xfs_inode_clear_reclaim_tag(mp, pag, ip);
inode->i_state = I_NEW;
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
} else {
/* If the VFS inode is being torn down, pause and try again. */
if (!igrab(inode)) {
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index fa31360046d4..0ffd56447045 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2439,75 +2439,31 @@ xfs_idestroy_fork(
}
/*
- * Increment the pin count of the given buffer.
- * This value is protected by ipinlock spinlock in the mount structure.
+ * This is called to unpin an inode. The caller must have the inode locked
+ * in at least shared mode so that the buffer cannot be subsequently pinned
+ * once someone is waiting for it to be unpinned.
*/
-void
-xfs_ipin(
- xfs_inode_t *ip)
-{
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
- atomic_inc(&ip->i_pincount);
-}
-
-/*
- * Decrement the pin count of the given inode, and wake up
- * anyone in xfs_iwait_unpin() if the count goes to 0. The
- * inode must have been previously pinned with a call to xfs_ipin().
- */
-void
-xfs_iunpin(
- xfs_inode_t *ip)
-{
- ASSERT(atomic_read(&ip->i_pincount) > 0);
-
- if (atomic_dec_and_test(&ip->i_pincount))
- wake_up(&ip->i_ipin_wait);
-}
-
-/*
- * This is called to unpin an inode. It can be directed to wait or to return
- * immediately without waiting for the inode to be unpinned. The caller must
- * have the inode locked in at least shared mode so that the buffer cannot be
- * subsequently pinned once someone is waiting for it to be unpinned.
- */
-STATIC void
-__xfs_iunpin_wait(
- xfs_inode_t *ip,
- int wait)
+static void
+xfs_iunpin_nowait(
+ struct xfs_inode *ip)
{
- xfs_inode_log_item_t *iip = ip->i_itemp;
-
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- if (atomic_read(&ip->i_pincount) == 0)
- return;
/* Give the log a push to start the unpinning I/O */
- if (iip && iip->ili_last_lsn)
- xfs_log_force_lsn(ip->i_mount, iip->ili_last_lsn, 0);
- else
- xfs_log_force(ip->i_mount, 0);
+ xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0);
- if (wait)
- wait_event(ip->i_ipin_wait, (atomic_read(&ip->i_pincount) == 0));
}
void
xfs_iunpin_wait(
- xfs_inode_t *ip)
+ struct xfs_inode *ip)
{
- __xfs_iunpin_wait(ip, 1);
-}
-
-static inline void
-xfs_iunpin_nowait(
- xfs_inode_t *ip)
-{
- __xfs_iunpin_wait(ip, 0);
+ if (xfs_ipincount(ip)) {
+ xfs_iunpin_nowait(ip);
+ wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0));
+ }
}
-
/*
* xfs_iextents_copy()
*
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 6c912b027596..9965e40a4615 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -471,8 +471,6 @@ int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *,
int xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
void xfs_iext_realloc(xfs_inode_t *, int, int);
-void xfs_ipin(xfs_inode_t *);
-void xfs_iunpin(xfs_inode_t *);
void xfs_iunpin_wait(xfs_inode_t *);
int xfs_iflush(xfs_inode_t *, uint);
void xfs_ichgtime(xfs_inode_t *, int);
@@ -480,6 +478,7 @@ void xfs_lock_inodes(xfs_inode_t **, int, uint);
void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
void xfs_synchronize_times(xfs_inode_t *);
+void xfs_mark_inode_dirty(xfs_inode_t *);
void xfs_mark_inode_dirty_sync(xfs_inode_t *);
#define IHOLD(ip) \
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index d4dc063111f8..7bfea8540159 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -535,23 +535,23 @@ xfs_inode_item_format(
/*
* This is called to pin the inode associated with the inode log
- * item in memory so it cannot be written out. Do this by calling
- * xfs_ipin() to bump the pin count in the inode while holding the
- * inode pin lock.
+ * item in memory so it cannot be written out.
*/
STATIC void
xfs_inode_item_pin(
xfs_inode_log_item_t *iip)
{
ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL));
- xfs_ipin(iip->ili_inode);
+
+ atomic_inc(&iip->ili_inode->i_pincount);
}
/*
* This is called to unpin the inode associated with the inode log
* item which was previously pinned with a call to xfs_inode_item_pin().
- * Just call xfs_iunpin() on the inode to do this.
+ *
+ * Also wake up anyone in xfs_iunpin_wait() if the count goes to 0.
*/
/* ARGSUSED */
STATIC void
@@ -559,7 +559,11 @@ xfs_inode_item_unpin(
xfs_inode_log_item_t *iip,
int stale)
{
- xfs_iunpin(iip->ili_inode);
+ struct xfs_inode *ip = iip->ili_inode;
+
+ ASSERT(atomic_read(&ip->i_pincount) > 0);
+ if (atomic_dec_and_test(&ip->i_pincount))
+ wake_up(&ip->i_ipin_wait);
}
/* ARGSUSED */
@@ -568,7 +572,7 @@ xfs_inode_item_unpin_remove(
xfs_inode_log_item_t *iip,
xfs_trans_t *tp)
{
- xfs_iunpin(iip->ili_inode);
+ xfs_inode_item_unpin(iip, 0);
}
/*
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 3af02314c605..b1b801e4a28e 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -106,6 +106,7 @@ xfs_bulkstat_one_iget(
buf->bs_dmevmask = dic->di_dmevmask;
buf->bs_dmstate = dic->di_dmstate;
buf->bs_aextents = dic->di_anextents;
+ buf->bs_forkoff = XFS_IFORK_BOFF(ip);
switch (dic->di_format) {
case XFS_DINODE_FMT_DEV:
@@ -176,6 +177,7 @@ xfs_bulkstat_one_dinode(
buf->bs_dmevmask = be32_to_cpu(dic->di_dmevmask);
buf->bs_dmstate = be16_to_cpu(dic->di_dmstate);
buf->bs_aextents = be16_to_cpu(dic->di_anextents);
+ buf->bs_forkoff = XFS_DFORK_BOFF(dic);
switch (dic->di_format) {
case XFS_DINODE_FMT_DEV:
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 4f16be4b6ee5..e8fba92d7cd9 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -60,7 +60,7 @@ STATIC int xlog_space_left(xlog_t *log, int cycle, int bytes);
STATIC int xlog_sync(xlog_t *log, xlog_in_core_t *iclog);
STATIC void xlog_dealloc_log(xlog_t *log);
STATIC int xlog_write(xfs_mount_t *mp, xfs_log_iovec_t region[],
- int nentries, xfs_log_ticket_t tic,
+ int nentries, struct xlog_ticket *tic,
xfs_lsn_t *start_lsn,
xlog_in_core_t **commit_iclog,
uint flags);
@@ -243,14 +243,14 @@ xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type)
* out when the next write occurs.
*/
xfs_lsn_t
-xfs_log_done(xfs_mount_t *mp,
- xfs_log_ticket_t xtic,
- void **iclog,
- uint flags)
+xfs_log_done(
+ struct xfs_mount *mp,
+ struct xlog_ticket *ticket,
+ struct xlog_in_core **iclog,
+ uint flags)
{
- xlog_t *log = mp->m_log;
- xlog_ticket_t *ticket = (xfs_log_ticket_t) xtic;
- xfs_lsn_t lsn = 0;
+ struct log *log = mp->m_log;
+ xfs_lsn_t lsn = 0;
if (XLOG_FORCED_SHUTDOWN(log) ||
/*
@@ -258,8 +258,7 @@ xfs_log_done(xfs_mount_t *mp,
* If we get an error, just continue and give back the log ticket.
*/
(((ticket->t_flags & XLOG_TIC_INITED) == 0) &&
- (xlog_commit_record(mp, ticket,
- (xlog_in_core_t **)iclog, &lsn)))) {
+ (xlog_commit_record(mp, ticket, iclog, &lsn)))) {
lsn = (xfs_lsn_t) -1;
if (ticket->t_flags & XLOG_TIC_PERM_RESERV) {
flags |= XFS_LOG_REL_PERM_RESERV;
@@ -289,7 +288,7 @@ xfs_log_done(xfs_mount_t *mp,
}
return lsn;
-} /* xfs_log_done */
+}
/*
* Attaches a new iclog I/O completion callback routine during
@@ -298,11 +297,11 @@ xfs_log_done(xfs_mount_t *mp,
* executing the callback at an appropriate time.
*/
int
-xfs_log_notify(xfs_mount_t *mp, /* mount of partition */
- void *iclog_hndl, /* iclog to hang callback off */
- xfs_log_callback_t *cb)
+xfs_log_notify(
+ struct xfs_mount *mp,
+ struct xlog_in_core *iclog,
+ xfs_log_callback_t *cb)
{
- xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl;
int abortflg;
spin_lock(&iclog->ic_callback_lock);
@@ -316,16 +315,14 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */
}
spin_unlock(&iclog->ic_callback_lock);
return abortflg;
-} /* xfs_log_notify */
+}
int
-xfs_log_release_iclog(xfs_mount_t *mp,
- void *iclog_hndl)
+xfs_log_release_iclog(
+ struct xfs_mount *mp,
+ struct xlog_in_core *iclog)
{
- xlog_t *log = mp->m_log;
- xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl;
-
- if (xlog_state_release_iclog(log, iclog)) {
+ if (xlog_state_release_iclog(mp->m_log, iclog)) {
xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
return EIO;
}
@@ -344,17 +341,18 @@ xfs_log_release_iclog(xfs_mount_t *mp,
* reservation, we prevent over allocation problems.
*/
int
-xfs_log_reserve(xfs_mount_t *mp,
- int unit_bytes,
- int cnt,
- xfs_log_ticket_t *ticket,
- __uint8_t client,
- uint flags,
- uint t_type)
+xfs_log_reserve(
+ struct xfs_mount *mp,
+ int unit_bytes,
+ int cnt,
+ struct xlog_ticket **ticket,
+ __uint8_t client,
+ uint flags,
+ uint t_type)
{
- xlog_t *log = mp->m_log;
- xlog_ticket_t *internal_ticket;
- int retval = 0;
+ struct log *log = mp->m_log;
+ struct xlog_ticket *internal_ticket;
+ int retval = 0;
ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
ASSERT((flags & XFS_LOG_NOSLEEP) == 0);
@@ -367,7 +365,7 @@ xfs_log_reserve(xfs_mount_t *mp,
if (*ticket != NULL) {
ASSERT(flags & XFS_LOG_PERM_RESERV);
- internal_ticket = (xlog_ticket_t *)*ticket;
+ internal_ticket = *ticket;
trace_xfs_log_reserve(log, internal_ticket);
@@ -519,7 +517,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
xlog_in_core_t *first_iclog;
#endif
xfs_log_iovec_t reg[1];
- xfs_log_ticket_t tic = NULL;
+ xlog_ticket_t *tic = NULL;
xfs_lsn_t lsn;
int error;
@@ -656,24 +654,24 @@ xfs_log_unmount(xfs_mount_t *mp)
* transaction occur with one call to xfs_log_write().
*/
int
-xfs_log_write(xfs_mount_t * mp,
- xfs_log_iovec_t reg[],
- int nentries,
- xfs_log_ticket_t tic,
- xfs_lsn_t *start_lsn)
+xfs_log_write(
+ struct xfs_mount *mp,
+ struct xfs_log_iovec reg[],
+ int nentries,
+ struct xlog_ticket *tic,
+ xfs_lsn_t *start_lsn)
{
- int error;
- xlog_t *log = mp->m_log;
+ struct log *log = mp->m_log;
+ int error;
if (XLOG_FORCED_SHUTDOWN(log))
return XFS_ERROR(EIO);
- if ((error = xlog_write(mp, reg, nentries, tic, start_lsn, NULL, 0))) {
+ error = xlog_write(mp, reg, nentries, tic, start_lsn, NULL, 0);
+ if (error)
xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
- }
return error;
-} /* xfs_log_write */
-
+}
void
xfs_log_move_tail(xfs_mount_t *mp,
@@ -1642,16 +1640,16 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
* bytes have been written out.
*/
STATIC int
-xlog_write(xfs_mount_t * mp,
- xfs_log_iovec_t reg[],
- int nentries,
- xfs_log_ticket_t tic,
- xfs_lsn_t *start_lsn,
- xlog_in_core_t **commit_iclog,
- uint flags)
+xlog_write(
+ struct xfs_mount *mp,
+ struct xfs_log_iovec reg[],
+ int nentries,
+ struct xlog_ticket *ticket,
+ xfs_lsn_t *start_lsn,
+ struct xlog_in_core **commit_iclog,
+ uint flags)
{
xlog_t *log = mp->m_log;
- xlog_ticket_t *ticket = (xlog_ticket_t *)tic;
xlog_in_core_t *iclog = NULL; /* ptr to current in-core log */
xlog_op_header_t *logop_head; /* ptr to log operation header */
__psint_t ptr; /* copy address into data region */
@@ -1765,7 +1763,7 @@ xlog_write(xfs_mount_t * mp,
default:
xfs_fs_cmn_err(CE_WARN, mp,
"Bad XFS transaction clientid 0x%x in ticket 0x%p",
- logop_head->oh_clientid, tic);
+ logop_head->oh_clientid, ticket);
return XFS_ERROR(EIO);
}
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 7074be9d13e9..97a24c7795a4 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -110,8 +110,6 @@ typedef struct xfs_log_iovec {
uint i_type; /* type of region */
} xfs_log_iovec_t;
-typedef void* xfs_log_ticket_t;
-
/*
* Structure used to pass callback function and the function's argument
* to the log manager.
@@ -126,10 +124,12 @@ typedef struct xfs_log_callback {
#ifdef __KERNEL__
/* Log manager interfaces */
struct xfs_mount;
+struct xlog_in_core;
struct xlog_ticket;
+
xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
- xfs_log_ticket_t ticket,
- void **iclog,
+ struct xlog_ticket *ticket,
+ struct xlog_in_core **iclog,
uint flags);
int _xfs_log_force(struct xfs_mount *mp,
uint flags,
@@ -151,21 +151,21 @@ int xfs_log_mount_finish(struct xfs_mount *mp);
void xfs_log_move_tail(struct xfs_mount *mp,
xfs_lsn_t tail_lsn);
int xfs_log_notify(struct xfs_mount *mp,
- void *iclog,
+ struct xlog_in_core *iclog,
xfs_log_callback_t *callback_entry);
int xfs_log_release_iclog(struct xfs_mount *mp,
- void *iclog_hndl);
+ struct xlog_in_core *iclog);
int xfs_log_reserve(struct xfs_mount *mp,
int length,
int count,
- xfs_log_ticket_t *ticket,
+ struct xlog_ticket **ticket,
__uint8_t clientid,
uint flags,
uint t_type);
int xfs_log_write(struct xfs_mount *mp,
xfs_log_iovec_t region[],
int nentries,
- xfs_log_ticket_t ticket,
+ struct xlog_ticket *ticket,
xfs_lsn_t *start_lsn);
int xfs_log_unmount_write(struct xfs_mount *mp);
void xfs_log_unmount(struct xfs_mount *mp);
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 6afaaeb2950a..e79b56b4bca6 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -1097,13 +1097,15 @@ xfs_default_resblks(xfs_mount_t *mp)
__uint64_t resblks;
/*
- * We default to 5% or 1024 fsbs of space reserved, whichever is smaller.
- * This may drive us straight to ENOSPC on mount, but that implies
- * we were already there on the last unmount. Warn if this occurs.
+ * We default to 5% or 8192 fsbs of space reserved, whichever is
+ * smaller. This is intended to cover concurrent allocation
+ * transactions when we initially hit enospc. These each require a 4
+ * block reservation. Hence by default we cover roughly 2000 concurrent
+ * allocation reservations.
*/
resblks = mp->m_sb.sb_dblocks;
do_div(resblks, 20);
- resblks = min_t(__uint64_t, resblks, 1024);
+ resblks = min_t(__uint64_t, resblks, 8192);
return resblks;
}
@@ -1417,6 +1419,9 @@ xfs_mountfs(
* when at ENOSPC. This is needed for operations like create with
* attr, unwritten extent conversion at ENOSPC, etc. Data allocations
* are not allowed to use this reserved space.
+ *
+ * This may drive us straight to ENOSPC on mount, but that implies
+ * we were already there on the last unmount. Warn if this occurs.
*/
if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
resblks = xfs_default_resblks(mp);
@@ -1725,26 +1730,30 @@ xfs_mod_incore_sb_unlocked(
lcounter += rem;
}
} else { /* Taking blocks away */
-
lcounter += delta;
+ if (lcounter >= 0) {
+ mp->m_sb.sb_fdblocks = lcounter +
+ XFS_ALLOC_SET_ASIDE(mp);
+ return 0;
+ }
- /*
- * If were out of blocks, use any available reserved blocks if
- * were allowed to.
- */
+ /*
+ * We are out of blocks, use any available reserved
+ * blocks if were allowed to.
+ */
+ if (!rsvd)
+ return XFS_ERROR(ENOSPC);
- if (lcounter < 0) {
- if (rsvd) {
- lcounter = (long long)mp->m_resblks_avail + delta;
- if (lcounter < 0) {
- return XFS_ERROR(ENOSPC);
- }
- mp->m_resblks_avail = lcounter;
- return 0;
- } else { /* not reserved */
- return XFS_ERROR(ENOSPC);
- }
+ lcounter = (long long)mp->m_resblks_avail + delta;
+ if (lcounter >= 0) {
+ mp->m_resblks_avail = lcounter;
+ return 0;
}
+ printk_once(KERN_WARNING
+ "Filesystem \"%s\": reserve blocks depleted! "
+ "Consider increasing reserve pool size.",
+ mp->m_fsname);
+ return XFS_ERROR(ENOSPC);
}
mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp);
@@ -2052,6 +2061,26 @@ xfs_mount_log_sb(
return error;
}
+/*
+ * If the underlying (data/log/rt) device is readonly, there are some
+ * operations that cannot proceed.
+ */
+int
+xfs_dev_is_read_only(
+ struct xfs_mount *mp,
+ char *message)
+{
+ if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
+ xfs_readonly_buftarg(mp->m_logdev_targp) ||
+ (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
+ cmn_err(CE_NOTE,
+ "XFS: %s required on read-only device.", message);
+ cmn_err(CE_NOTE,
+ "XFS: write access unavailable, cannot proceed.");
+ return EROFS;
+ }
+ return 0;
+}
#ifdef HAVE_PERCPU_SB
/*
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 14dafd608230..4fa0bc7b983e 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -436,6 +436,8 @@ extern void xfs_freesb(xfs_mount_t *);
extern int xfs_fs_writable(xfs_mount_t *);
extern int xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t);
+extern int xfs_dev_is_read_only(struct xfs_mount *, char *);
+
extern int xfs_dmops_get(struct xfs_mount *);
extern void xfs_dmops_put(struct xfs_mount *);
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index be942d4e3324..f73e358bae8d 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -796,7 +796,7 @@ _xfs_trans_commit(
int sync;
#define XFS_TRANS_LOGVEC_COUNT 16
xfs_log_iovec_t log_vector_fast[XFS_TRANS_LOGVEC_COUNT];
- void *commit_iclog;
+ struct xlog_in_core *commit_iclog;
int shutdown;
commit_lsn = -1;
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index c93e3a102857..79c8bab9dfff 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -910,7 +910,7 @@ typedef struct xfs_trans {
unsigned int t_blk_res_used; /* # of resvd blocks used */
unsigned int t_rtx_res; /* # of rt extents resvd */
unsigned int t_rtx_res_used; /* # of resvd rt extents used */
- xfs_log_ticket_t t_ticket; /* log mgr ticket */
+ struct xlog_ticket *t_ticket; /* log mgr ticket */
xfs_lsn_t t_lsn; /* log seq num of start of
* transaction. */
xfs_lsn_t t_commit_lsn; /* log seq num of end of
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 5ffd544434eb..fb586360d1c9 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -46,6 +46,65 @@ STATIC xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, xfs_buftarg_t *,
STATIC xfs_buf_t *xfs_trans_buf_item_match_all(xfs_trans_t *, xfs_buftarg_t *,
xfs_daddr_t, int);
+/*
+ * Add the locked buffer to the transaction.
+ *
+ * The buffer must be locked, and it cannot be associated with any
+ * transaction.
+ *
+ * If the buffer does not yet have a buf log item associated with it,
+ * then allocate one for it. Then add the buf item to the transaction.
+ */
+STATIC void
+_xfs_trans_bjoin(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
+ int reset_recur)
+{
+ struct xfs_buf_log_item *bip;
+
+ ASSERT(XFS_BUF_ISBUSY(bp));
+ ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL);
+
+ /*
+ * The xfs_buf_log_item pointer is stored in b_fsprivate. If
+ * it doesn't have one yet, then allocate one and initialize it.
+ * The checks to see if one is there are in xfs_buf_item_init().
+ */
+ xfs_buf_item_init(bp, tp->t_mountp);
+ bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
+ ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
+ ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
+ if (reset_recur)
+ bip->bli_recur = 0;
+
+ /*
+ * Take a reference for this transaction on the buf item.
+ */
+ atomic_inc(&bip->bli_refcount);
+
+ /*
+ * Get a log_item_desc to point at the new item.
+ */
+ (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip);
+
+ /*
+ * Initialize b_fsprivate2 so we can find it with incore_match()
+ * in xfs_trans_get_buf() and friends above.
+ */
+ XFS_BUF_SET_FSPRIVATE2(bp, tp);
+
+}
+
+void
+xfs_trans_bjoin(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
+{
+ _xfs_trans_bjoin(tp, bp, 0);
+ trace_xfs_trans_bjoin(bp->b_fspriv);
+}
/*
* Get and lock the buffer for the caller if it is not already
@@ -132,40 +191,8 @@ xfs_trans_get_buf(xfs_trans_t *tp,
ASSERT(!XFS_BUF_GETERROR(bp));
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
-
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_get_buf(bip);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_get_buf(bp->b_fspriv);
return (bp);
}
@@ -210,44 +237,11 @@ xfs_trans_getsb(xfs_trans_t *tp,
}
bp = xfs_getsb(mp, flags);
- if (bp == NULL) {
+ if (bp == NULL)
return NULL;
- }
-
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, mp);
-
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
- trace_xfs_trans_getsb(bip);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_getsb(bp->b_fspriv);
return (bp);
}
@@ -425,40 +419,9 @@ xfs_trans_read_buf(
if (XFS_FORCED_SHUTDOWN(mp))
goto shutdown_abort;
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_read_buf(bp->b_fspriv);
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_read_buf(bip);
*bpp = bp;
return 0;
@@ -623,53 +586,6 @@ xfs_trans_brelse(xfs_trans_t *tp,
}
/*
- * Add the locked buffer to the transaction.
- * The buffer must be locked, and it cannot be associated with any
- * transaction.
- *
- * If the buffer does not yet have a buf log item associated with it,
- * then allocate one for it. Then add the buf item to the transaction.
- */
-void
-xfs_trans_bjoin(xfs_trans_t *tp,
- xfs_buf_t *bp)
-{
- xfs_buf_log_item_t *bip;
-
- ASSERT(XFS_BUF_ISBUSY(bp));
- ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL);
-
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * in xfs_trans_get_buf() and friends above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_bjoin(bip);
-}
-
-/*
* Mark the buffer as not needing to be unlocked when the buf item's
* IOP_UNLOCK() routine is called. The buffer must already be locked
* and associated with the given transaction.
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index ddd2c5d1b854..9d376be0ea38 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -584,113 +584,6 @@ xfs_readlink(
}
/*
- * xfs_fsync
- *
- * This is called to sync the inode and its data out to disk. We need to hold
- * the I/O lock while flushing the data, and the inode lock while flushing the
- * inode. The inode lock CANNOT be held while flushing the data, so acquire
- * after we're done with that.
- */
-int
-xfs_fsync(
- xfs_inode_t *ip)
-{
- xfs_trans_t *tp;
- int error = 0;
- int log_flushed = 0;
-
- xfs_itrace_entry(ip);
-
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return XFS_ERROR(EIO);
-
- /*
- * We always need to make sure that the required inode state is safe on
- * disk. The inode might be clean but we still might need to force the
- * log because of committed transactions that haven't hit the disk yet.
- * Likewise, there could be unflushed non-transactional changes to the
- * inode core that have to go to disk and this requires us to issue
- * a synchronous transaction to capture these changes correctly.
- *
- * This code relies on the assumption that if the update_* fields
- * of the inode are clear and the inode is unpinned then it is clean
- * and no action is required.
- */
- xfs_ilock(ip, XFS_ILOCK_SHARED);
-
- if (!ip->i_update_core) {
- /*
- * Timestamps/size haven't changed since last inode flush or
- * inode transaction commit. That means either nothing got
- * written or a transaction committed which caught the updates.
- * If the latter happened and the transaction hasn't hit the
- * disk yet, the inode will be still be pinned. If it is,
- * force the log.
- */
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
- if (xfs_ipincount(ip)) {
- if (ip->i_itemp->ili_last_lsn) {
- error = _xfs_log_force_lsn(ip->i_mount,
- ip->i_itemp->ili_last_lsn,
- XFS_LOG_SYNC, &log_flushed);
- } else {
- error = _xfs_log_force(ip->i_mount,
- XFS_LOG_SYNC, &log_flushed);
- }
- }
- } else {
- /*
- * Kick off a transaction to log the inode core to get the
- * updates. The sync transaction will also force the log.
- */
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
- tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
- error = xfs_trans_reserve(tp, 0,
- XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
- if (error) {
- xfs_trans_cancel(tp, 0);
- return error;
- }
- xfs_ilock(ip, XFS_ILOCK_EXCL);
-
- /*
- * Note - it's possible that we might have pushed ourselves out
- * of the way during trans_reserve which would flush the inode.
- * But there's no guarantee that the inode buffer has actually
- * gone out yet (it's delwri). Plus the buffer could be pinned
- * anyway if it's part of an inode in another recent
- * transaction. So we play it safe and fire off the
- * transaction anyway.
- */
- xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
- xfs_trans_ihold(tp, ip);
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- xfs_trans_set_sync(tp);
- error = _xfs_trans_commit(tp, 0, &log_flushed);
-
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
- /*
- * If the log write didn't issue an ordered tag we need
- * to flush the disk cache for the data device now.
- */
- if (!log_flushed)
- xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
-
- /*
- * If this inode is on the RT dev we need to flush that
- * cache as well.
- */
- if (XFS_IS_REALTIME_INODE(ip))
- xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
- }
-
- return error;
-}
-
-/*
* Flags for xfs_free_eofblocks
*/
#define XFS_FREE_EOF_TRYLOCK (1<<0)
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index 774f40729ca1..d8dfa8d0dadd 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -21,7 +21,6 @@ int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags);
#define XFS_ATTR_NOACL 0x08 /* Don't call xfs_acl_chmod */
int xfs_readlink(struct xfs_inode *ip, char *link);
-int xfs_fsync(struct xfs_inode *ip);
int xfs_release(struct xfs_inode *ip);
int xfs_inactive(struct xfs_inode *ip);
int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name,
@@ -50,18 +49,6 @@ int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name,
int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags);
int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize,
int flags, struct attrlist_cursor_kern *cursor);
-ssize_t xfs_read(struct xfs_inode *ip, struct kiocb *iocb,
- const struct iovec *iovp, unsigned int segs,
- loff_t *offset, int ioflags);
-ssize_t xfs_splice_read(struct xfs_inode *ip, struct file *infilp,
- loff_t *ppos, struct pipe_inode_info *pipe, size_t count,
- int flags, int ioflags);
-ssize_t xfs_splice_write(struct xfs_inode *ip,
- struct pipe_inode_info *pipe, struct file *outfilp,
- loff_t *ppos, size_t count, int flags, int ioflags);
-ssize_t xfs_write(struct xfs_inode *xip, struct kiocb *iocb,
- const struct iovec *iovp, unsigned int nsegs,
- loff_t *offset, int ioflags);
int xfs_bmap(struct xfs_inode *ip, xfs_off_t offset, ssize_t count,
int flags, struct xfs_iomap *iomapp, int *niomaps);
void xfs_tosspages(struct xfs_inode *inode, xfs_off_t first,
@@ -72,4 +59,6 @@ int xfs_flush_pages(struct xfs_inode *ip, xfs_off_t first,
xfs_off_t last, uint64_t flags, int fiopt);
int xfs_wait_on_pages(struct xfs_inode *ip, xfs_off_t first, xfs_off_t last);
+int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
+
#endif /* _XFS_VNODEOPS_H */