diff options
| author | Linus Torvalds | 2020-12-14 16:22:26 -0800 |
|---|---|---|
| committer | Linus Torvalds | 2020-12-14 16:22:26 -0800 |
| commit | 586592478b1fa8bb8cd6875a9191468e9b1a8b13 (patch) | |
| tree | b93ea8074b11ffa3c10ec4c0ed910479a6564bb0 /arch | |
| parent | 0b03beface02d519693edb8020f9811c67d5c88f (diff) | |
| parent | 343dbdb7cb8997a2cb0fd804d6563b8a6de8d49b (diff) | |
Merge tag 's390-5.11-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Heiko Carstens:
- Add support for the hugetlb_cma command line option to allocate
gigantic hugepages using CMA
- Add arch_get_random_long() support.
- Add ap bus userspace notifications.
- Increase default size of vmalloc area to 512GB and otherwise let it
increase dynamically by the size of physical memory. This should fix
all occurrences where the vmalloc area was not large enough.
- Completely get rid of set_fs() (aka select SET_FS) and rework address
space handling while doing that; making address space handling much
more simple.
- Reimplement getcpu vdso syscall in C.
- Add support for extended SCLP responses (> 4k). This allows e.g. to
handle also potential large system configurations.
- Simplify KASAN by removing 3-level page table support and only
supporting 4-levels from now on.
- Improve debug-ability of the kernel decompressor code, which now
prints also stack traces and symbols in case of problems to the
console.
- Remove more power management leftovers.
- Other various fixes and improvements all over the place.
* tag 's390-5.11-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (62 commits)
s390/mm: add support to allocate gigantic hugepages using CMA
s390/crypto: add arch_get_random_long() support
s390/smp: perform initial CPU reset also for SMT siblings
s390/mm: use invalid asce for user space when switching to init_mm
s390/idle: fix accounting with machine checks
s390/idle: add missing mt_cycles calculation
s390/boot: add build-id to decompressor
s390/kexec_file: fix diag308 subcode when loading crash kernel
s390/cio: fix use-after-free in ccw_device_destroy_console
s390/cio: remove pm support from ccw bus driver
s390/cio: remove pm support from css-bus driver
s390/cio: remove pm support from IO subchannel drivers
s390/cio: remove pm support from chsc subchannel driver
s390/vmur: remove unused pm related functions
s390/tape: remove unsupported PM functions
s390/cio: remove pm support from eadm-sch drivers
s390: remove pm support from console drivers
s390/dasd: remove unused pm related functions
s390/zfcp: remove pm support from zfcp driver
s390/ap: let bus_register() add the AP bus sysfs attributes
...
Diffstat (limited to 'arch')
68 files changed, 908 insertions, 940 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 4a2a12be04c9..a60cc523d810 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -53,8 +53,7 @@ config ARCH_SUPPORTS_UPROBES config KASAN_SHADOW_OFFSET hex depends on KASAN - default 0x18000000000000 if KASAN_S390_4_LEVEL_PAGING - default 0x30000000000 + default 0x18000000000000 config S390 def_bool y @@ -191,7 +190,6 @@ config S390 select PCI_DOMAINS if PCI select PCI_MSI if PCI select PCI_MSI_ARCH_FALLBACKS if PCI_MSI - select SET_FS select SPARSE_IRQ select SYSCTL_EXCEPTION_TRACE select THREAD_INFO_IN_TASK @@ -714,7 +712,7 @@ if PCI config PCI_NR_FUNCTIONS int "Maximum number of PCI functions (1-4096)" range 1 4096 - default "128" + default "512" help This allows you to specify the maximum number of PCI functions which this kernel will support. diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug index ab48b694ade8..6bfaceebbbc0 100644 --- a/arch/s390/Kconfig.debug +++ b/arch/s390/Kconfig.debug @@ -5,3 +5,11 @@ config TRACE_IRQFLAGS_SUPPORT config EARLY_PRINTK def_bool y + +config DEBUG_USER_ASCE + bool "Debug User ASCE" + help + Check on exit to user space that address space control + elements are setup correctly. + + If unsure, say N. diff --git a/arch/s390/Makefile b/arch/s390/Makefile index ba94b03c8b2f..8db267d2a543 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -25,7 +25,7 @@ KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__ KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf)) KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY -KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float +KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float -mbackchain KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables KBUILD_CFLAGS_DECOMPRESSOR += -ffreestanding KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning, address-of-packed-member) diff --git a/arch/s390/boot/boot.h b/arch/s390/boot/boot.h index 2ea603f70c3b..8b50967f5804 100644 --- a/arch/s390/boot/boot.h +++ b/arch/s390/boot/boot.h @@ -2,20 +2,32 @@ #ifndef BOOT_BOOT_H #define BOOT_BOOT_H +#include <linux/types.h> + +#define BOOT_STACK_OFFSET 0x8000 + +#ifndef __ASSEMBLY__ + +#include <linux/compiler.h> + void startup_kernel(void); -void detect_memory(void); +unsigned long detect_memory(void); +bool is_ipl_block_dump(void); void store_ipl_parmblock(void); void setup_boot_command_line(void); void parse_boot_command_line(void); -void setup_memory_end(void); void verify_facilities(void); void print_missing_facilities(void); void print_pgm_check_info(void); unsigned long get_random_base(unsigned long safe_addr); +void __printf(1, 2) decompressor_printk(const char *fmt, ...); -extern int kaslr_enabled; extern const char kernel_version[]; +extern unsigned long memory_limit; +extern int vmalloc_size_set; +extern int kaslr_enabled; unsigned long read_ipl_report(unsigned long safe_offset); +#endif /* __ASSEMBLY__ */ #endif /* BOOT_BOOT_H */ diff --git a/arch/s390/boot/compressed/.gitignore b/arch/s390/boot/compressed/.gitignore index 765a08f1bd77..01d93832cf4a 100644 --- a/arch/s390/boot/compressed/.gitignore +++ b/arch/s390/boot/compressed/.gitignore @@ -1,3 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only vmlinux vmlinux.lds +vmlinux.syms diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile index b235ed95a3d8..de18dab518bb 100644 --- a/arch/s390/boot/compressed/Makefile +++ b/arch/s390/boot/compressed/Makefile @@ -10,21 +10,39 @@ GCOV_PROFILE := n UBSAN_SANITIZE := n KASAN_SANITIZE := n -obj-y := $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) piggy.o info.o +obj-y := $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) info.o +obj-all := $(obj-y) piggy.o syms.o targets := vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 targets += vmlinux.bin.xz vmlinux.bin.lzma vmlinux.bin.lzo vmlinux.bin.lz4 -targets += info.bin $(obj-y) +targets += info.bin syms.bin vmlinux.syms $(obj-all) KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR) KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR) OBJCOPYFLAGS := OBJECTS := $(addprefix $(obj)/,$(obj-y)) +OBJECTS_ALL := $(addprefix $(obj)/,$(obj-all)) -LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup -T -$(obj)/vmlinux: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS) FORCE +LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup --build-id=sha1 -T +$(obj)/vmlinux: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS_ALL) FORCE $(call if_changed,ld) +LDFLAGS_vmlinux.syms := --oformat $(LD_BFD) -e startup -T +$(obj)/vmlinux.syms: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS) FORCE + $(call if_changed,ld) + +quiet_cmd_dumpsyms = DUMPSYMS $< +define cmd_dumpsyms + $(NM) -n -S --format=bsd "$<" | $(PERL) -ne '/(\w+)\s+(\w+)\s+[tT]\s+(\w+)/ and printf "%x %x %s\0",hex $$1,hex $$2,$$3' > "$@" +endef + +$(obj)/syms.bin: $(obj)/vmlinux.syms FORCE + $(call if_changed,dumpsyms) + +OBJCOPYFLAGS_syms.o := -I binary -O elf64-s390 -B s390:64-bit --rename-section .data=.decompressor.syms +$(obj)/syms.o: $(obj)/syms.bin FORCE + $(call if_changed,objcopy) + OBJCOPYFLAGS_info.bin := -O binary --only-section=.vmlinux.info --set-section-flags .vmlinux.info=load $(obj)/info.bin: vmlinux FORCE $(call if_changed,objcopy) diff --git a/arch/s390/boot/compressed/decompressor.h b/arch/s390/boot/compressed/decompressor.h index c15eb7114d83..41f0ad97a4db 100644 --- a/arch/s390/boot/compressed/decompressor.h +++ b/arch/s390/boot/compressed/decompressor.h @@ -2,8 +2,10 @@ #ifndef BOOT_COMPRESSED_DECOMPRESSOR_H #define BOOT_COMPRESSED_DECOMPRESSOR_H +#include <linux/stddef.h> + #ifdef CONFIG_KERNEL_UNCOMPRESSED -static inline void *decompress_kernel(void) {} +static inline void *decompress_kernel(void) { return NULL; } #else void *decompress_kernel(void); #endif diff --git a/arch/s390/boot/compressed/vmlinux.lds.S b/arch/s390/boot/compressed/vmlinux.lds.S index 9427e2cd0c15..27a09c1c78f6 100644 --- a/arch/s390/boot/compressed/vmlinux.lds.S +++ b/arch/s390/boot/compressed/vmlinux.lds.S @@ -27,6 +27,7 @@ SECTIONS *(.rodata.*) _erodata = . ; } + NOTES .data : { _data = . ; *(.data) @@ -82,6 +83,14 @@ SECTIONS *(.vmlinux.info) } + .decompressor.syms : { + . += 1; /* make sure we have \0 before the first entry */ + . = ALIGN(2); + _decompressor_syms_start = .; + *(.decompressor.syms) + _decompressor_syms_end = .; + } + #ifdef CONFIG_KERNEL_UNCOMPRESSED . = 0x100000; #else diff --git a/arch/s390/boot/head.S b/arch/s390/boot/head.S index 1a2c2b1ed964..dacb7813f982 100644 --- a/arch/s390/boot/head.S +++ b/arch/s390/boot/head.S @@ -28,6 +28,7 @@ #include <asm/thread_info.h> #include <asm/page.h> #include <asm/ptrace.h> +#include "boot.h" #define ARCH_OFFSET 4 @@ -62,8 +63,12 @@ __HEAD .org __LC_RST_NEW_PSW # 0x1a0 .quad 0,iplstart + .org __LC_EXT_NEW_PSW # 0x1b0 + .quad 0x0002000180000000,0x1b0 # disabled wait .org __LC_PGM_NEW_PSW # 0x1d0 .quad 0x0000000180000000,startup_pgm_check_handler + .org __LC_IO_NEW_PSW # 0x1f0 + .quad 0x0002000180000000,0x1f0 # disabled wait .org 0x200 @@ -275,8 +280,8 @@ iplstart: # or linload or SALIPL # .org 0x10000 -ENTRY(startup) - j .Lep_startup_normal +SYM_CODE_START(startup) + j startup_normal .org EP_OFFSET # # This is a list of s390 kernel entry points. At address 0x1000f the number of @@ -290,9 +295,9 @@ ENTRY(startup) # kdump startup-code at 0x10010, running in 64 bit absolute addressing mode # .org 0x10010 -ENTRY(startup_kdump) - j .Lep_startup_kdump -.Lep_startup_normal: + j startup_kdump +SYM_CODE_END(startup) +SYM_CODE_START_LOCAL(startup_normal) mvi __LC_AR_MODE_ID,1 # set esame flag slr %r0,%r0 # set cpuid to zero lhi %r1,2 # mode 2 = esame (dump) @@ -303,6 +308,9 @@ ENTRY(startup_kdump) sam64 # switch to 64 bit addressing mode basr %r13,0 # get base .LPG0: + mvc __LC_EXT_NEW_PSW(16),.Lext_new_psw-.LPG0(%r13) + mvc __LC_PGM_NEW_PSW(16),.Lpgm_new_psw-.LPG0(%r13) + mvc __LC_IO_NEW_PSW(16),.Lio_new_psw-.LPG0(%r13) xc 0x200(256),0x200 # partially clear lowcore xc 0x300(256),0x300 xc 0xe00(256),0xe00 @@ -315,12 +323,18 @@ ENTRY(startup_kdump) l %r15,.Lstack-.LPG0(%r13) brasl %r14,verify_facilities brasl %r14,startup_kernel +SYM_CODE_END(startup_normal) .Lstack: - .long 0x8000 + (1<<(PAGE_SHIFT+BOOT_STACK_ORDER)) - STACK_FRAME_OVERHEAD + .long BOOT_STACK_OFFSET + BOOT_STACK_SIZE - STACK_FRAME_OVERHEAD .align 8 6: .long 0x7fffffff,0xffffffff - +.Lext_new_psw: + .quad 0x0002000180000000,0x1b0 # disabled wait +.Lpgm_new_psw: + .quad 0x0000000180000000,startup_pgm_check_handler +.Lio_new_psw: + .quad 0x0002000180000000,0x1f0 # disabled wait .Lctl: .quad 0x04040000 # cr0: AFP registers & secondary space .quad 0 # cr1: primary space segment table .quad .Lduct # cr2: dispatchable unit control table @@ -359,7 +373,7 @@ ENTRY(startup_kdump) # It simply saves general/control registers and psw in # the save area and does disabled wait with a faulty address. # -ENTRY(startup_pgm_check_handler) +SYM_CODE_START_LOCAL(startup_pgm_check_handler) stmg %r8,%r15,__LC_SAVE_AREA_SYNC la %r8,4095 stctg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r8) @@ -378,9 +392,9 @@ ENTRY(startup_pgm_check_handler) la %r8,4095 lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r8) lpswe __LC_RETURN_PSW # disabled wait +SYM_CODE_END(startup_pgm_check_handler) .Ldump_info_stack: .long 0x5000 + PAGE_SIZE - STACK_FRAME_OVERHEAD -ENDPROC(startup_pgm_check_handler) # # params at 10400 (setup.h) diff --git a/arch/s390/boot/head_kdump.S b/arch/s390/boot/head_kdump.S index 174d6959bf5b..f015469e7db9 100644 --- a/arch/s390/boot/head_kdump.S +++ b/arch/s390/boot/head_kdump.S @@ -19,8 +19,7 @@ # Note: This code has to be position independent # -.align 2 -.Lep_startup_kdump: +SYM_CODE_START_LOCAL(startup_kdump) lhi %r1,2 # mode 2 = esame (dump) sigp %r1,%r0,SIGP_SET_ARCHITECTURE # Switch to esame mode sam64 # Switch to 64 bit addressing @@ -87,14 +86,15 @@ startup_kdump_relocated: basr %r13,0 0: lpswe .Lrestart_psw-0b(%r13) # Start new kernel... +SYM_CODE_END(startup_kdump) .align 8 .Lrestart_psw: .quad 0x0000000080000000,0x0000000000000000 + startup #else -.align 2 -.Lep_startup_kdump: +SYM_CODE_START_LOCAL(startup_kdump) larl %r13,startup_kdump_crash lpswe 0(%r13) +SYM_CODE_END(startup_kdump) .align 8 startup_kdump_crash: .quad 0x0002000080000000,0x0000000000000000 + startup_kdump_crash diff --git a/arch/s390/boot/ipl_parm.c b/arch/s390/boot/ipl_parm.c index f94b91d72620..d372a45fe10e 100644 --- a/arch/s390/boot/ipl_parm.c +++ b/arch/s390/boot/ipl_parm.c @@ -17,10 +17,10 @@ int __bootdata_preserved(ipl_block_valid); unsigned int __bootdata_preserved(zlib_dfltcc_support) = ZLIB_DFLTCC_FULL; unsigned long __bootdata(vmalloc_size) = VMALLOC_DEFAULT_SIZE; -unsigned long __bootdata(memory_end); -int __bootdata(memory_end_set); int __bootdata(noexec_disabled); +unsigned long memory_limit; +int vmalloc_size_set; int kaslr_enabled; static inline int __diag308(unsigned long subcode, void *addr) @@ -57,6 +57,17 @@ void store_ipl_parmblock(void) ipl_block_valid = 1; } +bool is_ipl_block_dump(void) +{ + if (ipl_block.pb0_hdr.pbt == IPL_PBT_FCP && + ipl_block.fcp.opt == IPL_PB0_FCP_OPT_DUMP) + return true; + if (ipl_block.pb0_hdr.pbt == IPL_PBT_NVME && + ipl_block.nvme.opt == IPL_PB0_NVME_OPT_DUMP) + return true; + return false; +} + static size_t scpdata_length(const u8 *buf, size_t count) { while (count) { @@ -237,13 +248,13 @@ void parse_boot_command_line(void) while (*args) { args = next_arg(args, ¶m, &val); - if (!strcmp(param, "mem") && val) { - memory_end = round_down(memparse(val, NULL), PAGE_SIZE); - memory_end_set = 1; - } + if (!strcmp(param, "mem") && val) + memory_limit = round_down(memparse(val, NULL), PAGE_SIZE); - if (!strcmp(param, "vmalloc") && val) + if (!strcmp(param, "vmalloc") && val) { vmalloc_size = round_up(memparse(val, NULL), PAGE_SIZE); + vmalloc_size_set = 1; + } if (!strcmp(param, "dfltcc") && val) { if (!strcmp(val, "off")) @@ -279,27 +290,3 @@ void parse_boot_command_line(void) #endif } } - -static inline bool is_ipl_block_dump(void) -{ - if (ipl_block.pb0_hdr.pbt == IPL_PBT_FCP && - ipl_block.fcp.opt == IPL_PB0_FCP_OPT_DUMP) - return true; - if (ipl_block.pb0_hdr.pbt == IPL_PBT_NVME && - ipl_block.nvme.opt == IPL_PB0_NVME_OPT_DUMP) - return true; - return false; -} - -void setup_memory_end(void) -{ -#ifdef CONFIG_CRASH_DUMP - if (OLDMEM_BASE) { - kaslr_enabled = 0; - } else if (ipl_block_valid && is_ipl_block_dump()) { - kaslr_enabled = 0; - if (!sclp_early_get_hsa_size(&memory_end) && memory_end) - memory_end_set = 1; - } -#endif -} diff --git a/arch/s390/boot/kaslr.c b/arch/s390/boot/kaslr.c index d844a5ef9089..0dd48fbdbaa4 100644 --- a/arch/s390/boot/kaslr.c +++ b/arch/s390/boot/kaslr.c @@ -7,6 +7,7 @@ #include <asm/cpacf.h> #include <asm/timex.h> #include <asm/sclp.h> +#include <asm/kasan.h> #include "compressed/decompressor.h" #include "boot.h" @@ -176,8 +177,14 @@ unsigned long get_random_base(unsigned long safe_addr) unsigned long kasan_needs; int i; - if (memory_end_set) - memory_limit = min(memory_limit, memory_end); + memory_limit = min(memory_limit, ident_map_size); + + /* + * Avoid putting kernel in the end of physical memory + * which kasan will use for shadow memory and early pgtable + * mapping allocations. + */ + memory_limit -= kasan_estimate_memory_needs(memory_limit); if (IS_ENABLED(CONFIG_BLK_DEV_INITRD) && INITRD_START && INITRD_SIZE) { if (safe_addr < INITRD_START + INITRD_SIZE) @@ -185,28 +192,6 @@ unsigned long get_random_base(unsigned long safe_addr) } safe_addr = ALIGN(safe_addr, THREAD_SIZE); - if ((IS_ENABLED(CONFIG_KASAN))) { - /* - * Estimate kasan memory requirements, which it will reserve - * at the very end of available physical memory. To estimate - * that, we take into account that kasan would require - * 1/8 of available physical memory (for shadow memory) + - * creating page tables for the whole memory + shadow memory - * region (1 + 1/8). To keep page tables estimates simple take - * the double of combined ptes size. - */ - memory_limit = get_mem_detect_end(); - if (memory_end_set && memory_limit > memory_end) - memory_limit = memory_end; - - /* for shadow memory */ - kasan_needs = memory_limit / 8; - /* for paging structures */ - kasan_needs += (memory_limit + kasan_needs) / PAGE_SIZE / - _PAGE_ENTRIES * _PAGE_TABLE_SIZE * 2; - memory_limit -= kasan_needs; - } - kernel_size = vmlinux.image_size + vmlinux.bss_size; if (safe_addr + kernel_size > memory_limit) return 0; diff --git a/arch/s390/boot/mem_detect.c b/arch/s390/boot/mem_detect.c index 62e7c13ce85c..40168e59abd3 100644 --- a/arch/s390/boot/mem_detect.c +++ b/arch/s390/boot/mem_detect.c @@ -8,7 +8,6 @@ #include "compressed/decompressor.h" #include "boot.h" -unsigned long __bootdata(max_physmem_end); struct mem_detect_info __bootdata(mem_detect); /* up to 256 storage elements, 1020 subincrements each */ @@ -149,27 +148,29 @@ static void search_mem_end(void) add_mem_detect_block(0, (offset + 1) << 20); } -void detect_memory(void) +unsigned long detect_memory(void) { + unsigned long max_physmem_end; + sclp_early_get_memsize(&max_physmem_end); if (!sclp_early_read_storage_info()) { mem_detect.info_source = MEM_DETECT_SCLP_STOR_INFO; - return; + return max_physmem_end; } if (!diag260()) { mem_detect.info_source = MEM_DETECT_DIAG260; - return; + return max_physmem_end; } if (max_physmem_end) { add_mem_detect_block(0, max_physmem_end); mem_detect.info_source = MEM_DETECT_SCLP_READ_INFO; - return; + return max_physmem_end; } search_mem_end(); mem_detect.info_source = MEM_DETECT_BIN_SEARCH; - max_physmem_end = get_mem_detect_end(); + return get_mem_detect_end(); } diff --git a/arch/s390/boot/pgm_check_info.c b/arch/s390/boot/pgm_check_info.c index a3c9862bcede..3a46abed2549 100644 --- a/arch/s390/boot/pgm_check_info.c +++ b/arch/s390/boot/pgm_check_info.c @@ -1,99 +1,181 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/kernel.h> #include <linux/string.h> +#include <linux/ctype.h> +#include <asm/stacktrace.h> +#include <asm/boot_data.h> #include <asm/lowcore.h> #include <asm/setup.h> #include <asm/sclp.h> +#include <asm/uv.h> +#include <stdarg.h> #include "boot.h" const char hex_asc[] = "0123456789abcdef"; -#define add_val_as_hex(dst, val) \ - __add_val_as_hex(dst, (const unsigned char *)&val, sizeof(val)) +static char *as_hex(char *dst, unsigned long val, int pad) +{ + char *p, *end = p = dst + max(pad, (int)__fls(val | 1) / 4 + 1); + + for (*p-- = 0; p >= dst; val >>= 4) + *p-- = hex_asc[val & 0x0f]; + return end; +} -static char *__add_val_as_hex(char *dst, const unsigned char *src, size_t count) +static char *symstart(char *p) { - while (count--) - dst = hex_byte_pack(dst, *src++); - return dst; + while (*p) + p--; + return p + 1; } -static char *add_str(char *dst, char *src) +extern char _decompressor_syms_start[], _decompressor_syms_end[]; +static noinline char *findsym(unsigned long ip, unsigned short *off, unsigned short *len) { - strcpy(dst, src); - return dst + strlen(dst); + /* symbol entries are in a form "10000 c4 startup\0" */ + char *a = _decompressor_syms_start; + char *b = _decompressor_syms_end; + unsigned long start; + unsigned long size; + char *pivot; + char *endp; + + while (a < b) { + pivot = symstart(a + (b - a) / 2); + start = simple_strtoull(pivot, &endp, 16); + size = simple_strtoull(endp + 1, &endp, 16); + if (ip < start) { + b = pivot; + continue; + } + if (ip > start + size) { + a = pivot + strlen(pivot) + 1; + continue; + } + *off = ip - start; + *len = size; + return endp + 1; + } + return NULL; } -void print_pgm_check_info(void) +static noinline char *strsym(void *ip) { - struct psw_bits *psw = &psw_bits(S390_lowcore.psw_save_area); - unsigned short ilc = S390_lowcore.pgm_ilc >> 1; - char buf[256]; - int row, col; + static char buf[64]; + unsigned short off; + unsigned short len; char *p; - add_str(buf, "Linux version "); - strlcat(buf, kernel_version, sizeof(buf) - 1); - strlcat(buf, "\n", sizeof(buf)); - sclp_early_printk(buf); + p = findsym((unsigned long)ip, &off, &len); + if (p) { + strncpy(buf, p, sizeof(buf)); + /* reserve 15 bytes for offset/len in symbol+0x1234/0x1234 */ + p = buf + strnlen(buf, sizeof(buf) - 15); + strcpy(p, "+0x"); + p = as_hex(p + 3, off, 0); + strcpy(p, "/0x"); + as_hex(p + 3, len, 0); + } else { + as_hex(buf, (unsigned long)ip, 16); + } + return buf; +} - p = add_str(buf, "Kernel fault: interruption code "); - p = add_val_as_hex(buf + strlen(buf), S390_lowcore.pgm_code); - p = add_str(p, " ilc:"); - *p++ = hex_asc_lo(ilc); - add_str(p, "\n"); - sclp_early_printk(buf); +void decompressor_printk(const char *fmt, ...) +{ + char buf[1024] = { 0 }; + char *end = buf + sizeof(buf) - 1; /* make sure buf is 0 terminated */ + unsigned long pad; + char *p = buf; + va_list args; - if (kaslr_enabled) { - p = add_str(buf, "Kernel random base: "); - p = add_val_as_hex(p, __kaslr_offset); - add_str(p, "\n"); - sclp_early_printk(buf); + va_start(args, fmt); + for (; p < end && *fmt; fmt++) { + if (*fmt != '%') { + *p++ = *fmt; + continue; + } + pad = isdigit(*++fmt) ? simple_strtol(fmt, (char **)&fmt, 10) : 0; + switch (*fmt) { + case 's': + p = buf + strlcat(buf, va_arg(args, char *), sizeof(buf)); + break; + case 'p': + if (*++fmt != 'S') + goto out; + p = buf + strlcat(buf, strsym(va_arg(args, void *)), sizeof(buf)); + break; + case 'l': + if (*++fmt != 'x' || end - p <= max(sizeof(long) * 2, pad)) + goto out; + p = as_hex(p, va_arg(args, unsigned long), pad); + break; + case 'x': + if (end - p <= max(sizeof(int) * 2, pad)) + goto out; + p = as_hex(p, va_arg(args, unsigned int), pad); + break; + default: + goto out; + } } - - p = add_str(buf, "PSW : "); - p = add_val_as_hex(p, S390_lowcore.psw_save_area.mask); - p = add_str(p, " "); - p = add_val_as_hex(p, S390_lowcore.psw_save_area.addr); - add_str(p, "\n"); +out: + va_end(args); sclp_early_printk(buf); +} - p = add_str(buf, " R:"); - *p++ = hex_asc_lo(psw->per); - p = add_str(p, " T:"); - *p++ = hex_asc_lo(psw->dat); - p = add_str(p, " IO:"); - *p++ = hex_asc_lo(psw->io); - p = add_str(p, " EX:"); - *p++ = hex_asc_lo(psw->ext); - p = add_str(p, " Key:"); - *p++ = hex_asc_lo(psw->key); - p = add_str(p, " M:"); - *p++ = hex_asc_lo(psw->mcheck); - p = add_str(p, " W:"); - *p++ = hex_asc_lo(psw->wait); - p = add_str(p, " P:"); - *p++ = hex_asc_lo(psw->pstate); - p = add_str(p, " AS:"); - *p++ = hex_asc_lo(psw->as); - p = add_str(p, " CC:"); - *p++ = hex_asc_lo(psw->cc); - p = add_str(p, " PM:"); - *p++ = hex_asc_lo(psw->pm); - p = add_str(p, " RI:"); - *p++ = hex_asc_lo(psw->ri); - p = add_str(p, " EA:"); - *p++ = hex_asc_lo(psw->eaba); - add_str(p, "\n"); - sclp_early_printk(buf); +static noinline void print_stacktrace(void) +{ + struct stack_info boot_stack = { STACK_TYPE_TASK, BOOT_STACK_OFFSET, + BOOT_STACK_OFFSET + BOOT_STACK_SIZE }; + unsigned long sp = S390_lowcore.gpregs_save_area[15]; + bool first = true; - for (row = 0; row < 4; row++) { - p = add_str(buf, row == 0 ? "GPRS:" : " "); - for (col = 0; col < 4; col++) { - p = add_str(p, " "); - p = add_val_as_hex(p, S390_lowcore.gpregs_save_area[row * 4 + col]); - } - add_str(p, "\n"); - sclp_early_printk(buf); + decompressor_printk("Call Trace:\n"); + while (!(sp & 0x7) && on_stack(&boot_stack, sp, sizeof(struct stack_frame))) { + struct stack_frame *sf = (struct stack_frame *)sp; + + decompressor_printk(first ? "(sp:%016lx [<%016lx>] %pS)\n" : + " sp:%016lx [<%016lx>] %pS\n", + sp, sf->gprs[8], (void *)sf->gprs[8]); + if (sf->back_chain <= sp) + break; + sp = sf->back_chain; + first = false; } } + +void print_pgm_check_info(void) +{ + unsigned long *gpregs = (unsigned long *)S390_lowcore.gpregs_save_area; + struct psw_bits *psw = &psw_bits(S390_lowcore.psw_save_area); + + decompressor_printk("Linux version %s\n", kernel_version); + if (!is_prot_virt_guest() && early_command_line[0]) + decompressor_printk("Kernel command line: %s\n", early_command_line); + decompressor_printk("Kernel fault: interruption code %04x ilc:%x\n", + S390_lowcore.pgm_code, S390_lowcore.pgm_ilc >> 1); + if (kaslr_enabled) + decompressor_printk("Kernel random base: %lx\n", __kaslr_offset); + decompressor_printk("PSW : %016lx %016lx (%pS)\n", + S390_lowcore.psw_save_area.mask, + S390_lowcore.psw_save_area.addr, + (void *)S390_lowcore.psw_save_area.addr); + decompressor_printk( + " R:%x T:%x IO:%x EX:%x Key:%x M:%x W:%x P:%x AS:%x CC:%x PM:%x RI:%x EA:%x\n", + psw->per, psw->dat, psw->io, psw->ext, psw->key, psw->mcheck, + psw->wait, psw->pstate, psw->as, psw->cc, psw->pm, psw->ri, + psw->eaba); + decompressor_printk("GPRS: %016lx %016lx %016lx %016lx\n", + gpregs[0], gpregs[1], gpregs[2], gpregs[3]); + decompressor_printk(" %016lx %016lx %016lx %016lx\n", + gpregs[4], gpregs[5], gpregs[6], gpregs[7]); + decompressor_printk(" %016lx %016lx %016lx %016lx\n", + gpregs[8], gpregs[9], gpregs[10], gpregs[11]); + decompressor_printk(" %016lx %016lx %016lx %016lx\n", + gpregs[12], gpregs[13], gpregs[14], gpregs[15]); + print_stacktrace(); + decompressor_printk("Last Breaking-Event-Address:\n"); + decompressor_printk(" [<%016lx>] %pS\n", (unsigned long)S390_lowcore.breaking_event_addr, + (void *)S390_lowcore.breaking_event_addr); +} diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c index cc96b04cc0ba..05f8eefa3dcf 100644 --- a/arch/s390/boot/startup.c +++ b/arch/s390/boot/startup.c @@ -1,7 +1,9 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/string.h> #include <linux/elf.h> +#include <asm/boot_data.h> #include <asm/sections.h> +#include <asm/cpu_mf.h> #include <asm/setup.h> #include <asm/kexec.h> #include <asm/sclp.h> @@ -13,6 +15,7 @@ extern char __boot_data_start[], __boot_data_end[]; extern char __boot_data_preserved_start[], __boot_data_preserved_end[]; unsigned long __bootdata_preserved(__kaslr_offset); +unsigned long __bootdata(ident_map_size); /* * Some code and data needs to stay below 2 GB, even when the kernel would be @@ -58,6 +61,14 @@ void error(char *x) disabled_wait(); } +static void setup_lpp(void) +{ + S390_lowcore.current_pid = 0; + S390_lowcore.lpp = LPP_MAGIC; + if (test_facility(40)) + lpp(&S390_lowcore.lpp); +} + #ifdef CONFIG_KERNEL_UNCOMPRESSED unsigned long mem_safe_offset(void) { @@ -119,6 +130,46 @@ static void handle_relocs(unsigned long offset) } /* + * Merge information from several sources into a single ident_map_size value. + * "ident_map_size" represents the upper limit of physical memory we may ever + * reach. It might not be all online memory, but also include standby (offline) + * memory. "ident_map_size" could be lower then actual standby or even online + * memory present, due to limiting factors. We should never go above this limit. + * It is the size of our identity mapping. + * + * Consider the following factors: + * 1. max_physmem_end - end of physical memory online or standby. + * Always <= end of the last online memory block (get_mem_detect_end()). + * 2. CONFIG_MAX_PHYSMEM_BITS - the maximum size of physical memory the + * kernel is able to support. + * 3. "mem=" kernel command line option which limits physical memory usage. + * 4. OLDMEM_BASE which is a kdump memory limit when the kernel is executed as + * crash kernel. + * 5. "hsa" size which is a memory limit when the kernel is executed during + * zfcp/nvme dump. + */ +static void setup_ident_map_size(unsigned long max_physmem_end) +{ + unsigned long hsa_size; + + ident_map_size = max_physmem_end; + if (memory_limit) + ident_map_size = min(ident_map_size, memory_limit); + ident_map_size = min(ident_map_size, 1UL << MAX_PHYSMEM_BITS); + +#ifdef CONFIG_CRASH_DUMP + if (OLDMEM_BASE) { + kaslr_enabled = 0; + ident_map_size = min(ident_map_size, OLDMEM_SIZE); + } else if (ipl_block_valid && is_ipl_block_dump()) { + kaslr_enabled = 0; + if (!sclp_early_get_hsa_size(&hsa_size) && hsa_size) + ident_map_size = min(ident_map_size, hsa_size); + } +#endif +} + +/* * This function clears the BSS section of the decompressed Linux kernel and NOT the decompressor's. */ static void clear_bss_section(void) @@ -126,12 +177,27 @@ static void clear_bss_section(void) memset((void *)vmlinux.default_lma + vmlinux.image_size, 0, vmlinux.bss_size); } +/* + * Set vmalloc area size to an 8th of (potential) physical memory + * size, unless size has been set by kernel command line parameter. + */ +static void setup_vmalloc_size(void) +{ + unsigned long size; + + if (vmalloc_size_set) + return; + size = round_up(ident_map_size / 8, _SEGMENT_SIZE); + vmalloc_size = max(size, vmalloc_size); +} + void startup_kernel(void) { unsigned long random_lma; unsigned long safe_addr; void *img; + setup_lpp(); store_ipl_parmblock(); safe_addr = mem_safe_offset(); safe_addr = read_ipl_report(safe_addr); @@ -140,8 +206,8 @@ void startup_kernel(void) sclp_early_read_info(); setup_boot_command_line(); parse_boot_command_line(); - setup_memory_end(); - detect_memory(); + setup_ident_map_size(detect_memory()); + setup_vmalloc_size(); random_lma = __kaslr_offset = 0; if (IS_ENABLED(CONFIG_RANDOMIZE_BASE) && kaslr_enabled) { diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index fe6f529ac82c..c52113a238b1 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -826,6 +826,7 @@ CONFIG_FTRACE_SYSCALLS=y CONFIG_BLK_DEV_IO_TRACE=y CONFIG_BPF_KPROBE_OVERRIDE=y CONFIG_HIST_TRIGGERS=y +CONFIG_DEBUG_USER_ASCE=y CONFIG_NOTIFIER_ERROR_INJECTION=m CONFIG_NETDEV_NOTIFIER_ERROR_INJECT=m CONFIG_FAULT_INJECTION=y diff --git a/arch/s390/crypto/arch_random.c b/arch/s390/crypto/arch_random.c index dd95cdbd22ce..7b947728d57e 100644 --- a/arch/s390/crypto/arch_random.c +++ b/arch/s390/crypto/arch_random.c @@ -2,7 +2,7 @@ /* * s390 arch random implementation. * - * Copyright IBM Corp. 2017, 2018 + * Copyright IBM Corp. 2017, 2020 * Author(s): Harald Freudenberger * * The s390_arch_random_generate() function may be called from random.c @@ -33,6 +33,7 @@ #include <linux/slab.h> #include <linux/static_key.h> #include <linux/workqueue.h> +#include <linux/moduleparam.h> #include <asm/cpacf.h> DEFINE_STATIC_KEY_FALSE(s390_arch_random_available); @@ -99,6 +100,113 @@ static void arch_rng_refill_buffer(struct work_struct *unused) queue_delayed_work(system_long_wq, &arch_rng_work, delay); } +/* + * Here follows the implementation of s390_arch_get_random_long(). + * + * The random longs to be pulled by arch_get_random_long() are + * prepared in an 4K buffer which is filled from the NIST 800-90 + * compliant s390 drbg. By default the random long buffer is refilled + * 256 times before the drbg itself needs a reseed. The reseed of the + * drbg is done with 32 bytes fetched from the high quality (but slow) + * trng which is assumed to deliver 100% entropy. So the 32 * 8 = 256 + * bits of entropy are spread over 256 * 4KB = 1MB serving 131072 + * arch_get_random_long() invocations before reseeded. + * + * How often the 4K random long buffer is refilled with the drbg + * before the drbg is reseeded can be adjusted. There is a module + * parameter 's390_arch_rnd_long_drbg_reseed' accessible via + * /sys/module/arch_random/parameters/rndlong_drbg_reseed + * or as kernel command line parameter + * arch_random.rndlong_drbg_reseed=<value> + * This parameter tells how often the drbg fills the 4K buffer before + * it is re-seeded by fresh entropy from the trng. + * A value of 16 results in reseeding the drbg at every 16 * 4 KB = 64 + * KB with 32 bytes of fresh entropy pulled from the trng. So a value + * of 16 would result in 256 bits entropy per 64 KB. + * A value of 256 results in 1MB of drbg output before a reseed of the + * drbg is done. So this would spread the 256 bits of entropy among 1MB. + * Setting this parameter to 0 forces the reseed to take place every + * time the 4K buffer is depleted, so the entropy rises to 256 bits + * entropy per 4K or 0.5 bit entropy per arch_get_random_long(). With + * setting this parameter to negative values all this effort is + * disabled, arch_get_random long() returns false and thus indicating + * that the arch_get_random_long() feature is disabled at all. + */ + +static unsigned long rndlong_buf[512]; +static DEFINE_SPINLOCK(rndlong_lock); +static int rndlong_buf_index; + +static int rndlong_drbg_reseed = 256; +module_param_named(rndlong_drbg_reseed, rndlong_drbg_reseed, int, 0600); +MODULE_PARM_DESC(rndlong_drbg_reseed, "s390 arch_get_random_long() drbg reseed"); + +static inline void refill_rndlong_buf(void) +{ + static u8 prng_ws[240]; + static int drbg_counter; + + if (--drbg_counter < 0) { + /* need to re-seed the drbg */ + u8 seed[32]; + + /* fetch seed from trng */ + cpacf_trng(NULL, 0, seed, sizeof(seed)); + /* seed drbg */ + memset(prng_ws, 0, sizeof(prng_ws)); + cpacf_prno(CPACF_PRNO_SHA512_DRNG_SEED, + &prng_ws, NULL, 0, seed, sizeof(seed)); + /* re-init counter for drbg */ + drbg_counter = rndlong_drbg_reseed; + } + + /* fill the arch_get_random_long buffer from drbg */ + cpacf_prno(CPACF_PRNO_SHA512_DRNG_GEN, &prng_ws, + (u8 *) rndlong_buf, sizeof(rndlong_buf), + NULL, 0); +} + +bool s390_arch_get_random_long(unsigned long *v) +{ + bool rc = false; + unsigned long flags; + + /* arch_get_random_long() disabled ? */ + if (rndlong_drbg_reseed < 0) + return false; + + /* try to lock the random long lock */ + if (!spin_trylock_irqsave(&rndlong_lock, flags)) + return false; + + if (--rndlong_buf_index >= 0) { + /* deliver next long value from the buffer */ + *v = rndlong_buf[rndlong_buf_index]; + rc = true; + goto out; + } + + /* buffer is depleted and needs refill */ + if (in_interrupt()) { + /* delay refill in interrupt context to next caller */ + rndlong_buf_index = 0; + goto out; + } + + /* refill random long buffer */ + refill_rndlong_buf(); + rndlong_buf_index = ARRAY_SIZE(rndlong_buf); + + /* and provide one random long */ + *v = rndlong_buf[--rndlong_buf_index]; + rc = true; + +out: + spin_unlock_irqrestore(&rndlong_lock, flags); + return rc; +} +EXPORT_SYMBOL(s390_arch_get_random_long); + static int __init s390_arch_random_init(void) { /* all the needed PRNO subfunctions available ? */ diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c index 5057773f82e9..b2f219ec379c 100644 --- a/arch/s390/crypto/prng.c +++ b/arch/s390/crypto/prng.c @@ -674,20 +674,6 @@ static const struct file_operations prng_tdes_fops = { .llseek = noop_llseek, }; -static struct miscdevice prng_sha512_dev = { - .name = "prandom", - .minor = MISC_DYNAMIC_MINOR, - .mode = 0644, - .fops = &prng_sha512_fops, -}; -static struct miscdevice prng_tdes_dev = { - .name = "prandom", - .minor = MISC_DYNAMIC_MINOR, - .mode = 0644, - .fops = &prng_tdes_fops, -}; - - /* chunksize attribute (ro) */ static ssize_t prng_chunksize_show(struct device *dev, struct device_attribute *attr, @@ -801,18 +787,30 @@ static struct attribute *prng_sha512_dev_attrs[] = { &dev_attr_strength.attr, NULL }; +ATTRIBUTE_GROUPS(prng_sha512_dev); + static struct attribute *prng_tdes_dev_attrs[] = { &dev_attr_chunksize.attr, &dev_attr_byte_counter.attr, &dev_attr_mode.attr, NULL }; +ATTRIBUTE_GROUPS(prng_tdes_dev); -static struct attribute_group prng_sha512_dev_attr_group = { - .attrs = prng_sha512_dev_attrs +static struct miscdevice prng_sha512_dev = { + .name = "prandom", + .minor = MISC_DYNAMIC_MINOR, + .mode = 0644, + .fops = &prng_sha512_fops, + .groups = prng_sha512_dev_groups, }; -static struct attribute_group prng_tdes_dev_attr_group = { - .attrs = prng_tdes_dev_attrs + +static struct miscdevice prng_tdes_dev = { + .name = "prandom", + .minor = MISC_DYNAMIC_MINOR, + .mode = 0644, + .fops = &prng_tdes_fops, + .groups = prng_tdes_dev_groups, }; @@ -867,13 +865,6 @@ static int __init prng_init(void) prng_sha512_deinstantiate(); goto out; } - ret = sysfs_create_group(&prng_sha512_dev.this_device->kobj, - &prng_sha512_dev_attr_group); - if (ret) { - misc_deregister(&prng_sha512_dev); - prng_sha512_deinstantiate(); - goto out; - } } else { @@ -898,14 +889,6 @@ static int __init prng_init(void) prng_tdes_deinstantiate(); goto out; } - ret = sysfs_create_group(&prng_tdes_dev.this_device->kobj, - &prng_tdes_dev_attr_group); - if (ret) { - misc_deregister(&prng_tdes_dev); - prng_tdes_deinstantiate(); - goto out; - } - } out: @@ -916,13 +899,9 @@ out: static void __exit prng_exit(void) { if (prng_mode == PRNG_MODE_SHA512) { - sysfs_remove_group(&prng_sha512_dev.this_device->kobj, - &prng_sha512_dev_attr_group); misc_deregister(&prng_sha512_dev); prng_sha512_deinstantiate(); } else { - sysfs_remove_group(&prng_tdes_dev.this_device->kobj, - &prng_tdes_dev_attr_group); misc_deregister(&prng_tdes_dev); prng_tdes_deinstantiate(); } diff --git a/arch/s390/include/asm/archrandom.h b/arch/s390/include/asm/archrandom.h index de61ce562052..5dc712fde3c7 100644 --- a/arch/s390/include/asm/archrandom.h +++ b/arch/s390/include/asm/archrandom.h @@ -2,7 +2,7 @@ /* * Kernel interface for the s390 arch_random_* functions * - * Copyright IBM Corp. 2017 + * Copyright IBM Corp. 2017, 2020 * * Author: Harald Freudenberger <freude@de.ibm.com> * @@ -19,10 +19,13 @@ DECLARE_STATIC_KEY_FALSE(s390_arch_random_available); extern atomic64_t s390_arch_random_counter; +bool s390_arch_get_random_long(unsigned long *v); bool s390_arch_random_generate(u8 *buf, unsigned int nbytes); static inline bool __must_check arch_get_random_long(unsigned long *v) { + if (static_branch_likely(&s390_arch_random_available)) + return s390_arch_get_random_long(v); return false; } diff --git a/arch/s390/include/asm/ccwdev.h b/arch/s390/include/asm/ccwdev.h index c0be5fe1ddba..e3e2ab0acf83 100644 --- a/arch/s390/include/asm/ccwdev.h +++ b/arch/s390/include/asm/ccwdev.h @@ -115,7 +115,7 @@ enum uc_todo { }; /** - * struct ccw driver - device driver for channel attached devices + * struct ccw_driver - device driver for channel attached devices * @ids: ids supported by this driver * @probe: function called on probe * @remove: function called on remove @@ -124,11 +124,6 @@ enum uc_todo { * @notify: notify driver of device state changes * @path_event: notify driver of channel path events * @shutdown: called at device shutdown - * @prepare: prepare for pm state transition - * @complete: undo work done in @prepare - * @freeze: callback for freezing during hibernation snapshotting - * @thaw: undo work done in @freeze - * @restore: callback for restoring after hibernation * @uc_handler: callback for unit check handler * @driver: embedded device driver structure * @int_class: interruption class to use for accounting interrupts @@ -142,11 +137,6 @@ struct ccw_driver { int (*notify) (struct ccw_device *, int); void (*path_event) (struct ccw_device *, int *); void (*shutdown) (struct ccw_device *); - int (*prepare) (struct ccw_device *); - void (*complete) (struct ccw_device *); - int (*freeze)(struct ccw_device *); - int (*thaw) (struct ccw_device *); - int (*restore)(struct ccw_device *); enum uc_todo (*uc_handler) (struct ccw_device *, struct irb *); struct device_driver driver; enum interruption_class int_class; diff --git a/arch/s390/include/asm/cio.h b/arch/s390/include/asm/cio.h index 5c58756d6476..23dceb8d0453 100644 --- a/arch/s390/include/asm/cio.h +++ b/arch/s390/include/asm/cio.h @@ -329,7 +329,7 @@ struct ccw_dev_id { }; /** - * ccw_device_id_is_equal() - compare two ccw_dev_ids + * ccw_dev_id_is_equal() - compare two ccw_dev_ids * @dev_id1: a ccw_dev_id * @dev_id2: another ccw_dev_id * Returns: diff --git a/arch/s390/include/asm/delay.h b/arch/s390/include/asm/delay.h index 898323fd93d2..4a08379cd1eb 100644 --- a/arch/s390/include/asm/delay.h +++ b/arch/s390/include/asm/delay.h @@ -13,6 +13,7 @@ #ifndef _S390_DELAY_H #define _S390_DELAY_H +void udelay_enable(void); void __ndelay(unsigned long long nsecs); void __udelay(unsigned long long usecs); void udelay_simple(unsigned long long usecs); diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h index 68d362f8d6c1..695c61989f97 100644 --- a/arch/s390/include/asm/ftrace.h +++ b/arch/s390/include/asm/ftrace.h @@ -2,16 +2,9 @@ #ifndef _ASM_S390_FTRACE_H #define _ASM_S390_FTRACE_H +#define HAVE_FUNCTION_GRAPH_RET_ADDR_PTR #define ARCH_SUPPORTS_FTRACE_OPS 1 - -#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT) #define MCOUNT_INSN_SIZE 6 -#else -#define MCOUNT_INSN_SIZE 24 -#define MCOUNT_RETURN_FIXUP 18 -#endif - -#define HAVE_FUNCTION_GRAPH_RET_ADDR_PTR #ifndef __ASSEMBLY__ @@ -22,7 +15,6 @@ #define ftrace_return_address(n) __builtin_return_address(n) #endif -void _mcount(void); void ftrace_caller(void); extern char ftrace_graph_caller_end; @@ -30,12 +22,20 @@ extern unsigned long ftrace_plt; struct dyn_arch_ftrace { }; -#define MCOUNT_ADDR ((unsigned long)_mcount) +#define MCOUNT_ADDR 0 #define FTRACE_ADDR ((unsigned long)ftrace_caller) #define KPROBE_ON_FTRACE_NOP 0 #define KPROBE_ON_FTRACE_CALL 1 +struct module; +struct dyn_ftrace; +/* + * Either -mhotpatch or -mnop-mcount is used - no explicit init is required + */ +static inline int ftrace_init_nop(struct module *mod, struct dyn_ftrace *rec) { return 0; } +#define ftrace_init_nop ftrace_init_nop + static inline unsigned long ftrace_call_adjust(unsigned long addr) { return addr; @@ -49,28 +49,17 @@ struct ftrace_insn { static inline void ftrace_generate_nop_insn(struct ftrace_insn *insn) { #ifdef CONFIG_FUNCTION_TRACER -#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT) /* brcl 0,0 */ insn->opc = 0xc004; insn->disp = 0; -#else - /* jg .+24 */ - insn->opc = 0xc0f4; - insn->disp = MCOUNT_INSN_SIZE / 2; -#endif #endif } static inline int is_ftrace_nop(struct ftrace_insn *insn) { #ifdef CONFIG_FUNCTION_TRACER -#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT) if (insn->disp == 0) return 1; -#else - if (insn->disp == MCOUNT_INSN_SIZE / 2) - return 1; -#endif #endif return 0; } diff --git a/arch/s390/include/asm/futex.h b/arch/s390/include/asm/futex.h index 26f9144562c9..c22debfcebf1 100644 --- a/arch/s390/include/asm/futex.h +++ b/arch/s390/include/asm/futex.h @@ -26,9 +26,7 @@ static inline int arch_futex_atomic_op_inuser(int op, int oparg, int *oval, u32 __user *uaddr) { int oldval = 0, newval, ret; - mm_segment_t old_fs; - old_fs = enable_sacf_uaccess(); switch (op) { case FUTEX_OP_SET: __futex_atomic_op("lr %2,%5\n", @@ -53,7 +51,6 @@ static inline int arch_futex_atomic_op_inuser(int op, int oparg, int *oval, default: ret = -ENOSYS; } - disable_sacf_uaccess(old_fs); if (!ret) *oval = oldval; @@ -64,10 +61,8 @@ static inline int arch_futex_atomic_op_inuser(int op, int oparg, int *oval, static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, u32 oldval, u32 newval) { - mm_segment_t old_fs; int ret; - old_fs = enable_sacf_uaccess(); asm volatile( " sacf 256\n" "0: cs %1,%4,0(%5)\n" @@ -77,7 +72,6 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, : "=d" (ret), "+d" (oldval), "=m" (*uaddr) : "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr) : "cc", "memory"); - disable_sacf_uaccess(old_fs); *uval = oldval; return ret; } diff --git a/arch/s390/include/asm/kasan.h b/arch/s390/include/asm/kasan.h index e9bf486de136..76f351bd6645 100644 --- a/arch/s390/include/asm/kasan.h +++ b/arch/s390/include/asm/kasan.h @@ -2,28 +2,51 @@ #ifndef __ASM_KASAN_H #define __ASM_KASAN_H +#include <asm/pgtable.h> + #ifdef CONFIG_KASAN #define KASAN_SHADOW_SCALE_SHIFT 3 -#ifdef CONFIG_KASAN_S390_4_LEVEL_PAGING #define KASAN_SHADOW_SIZE \ (_AC(1, UL) << (_REGION1_SHIFT - KASAN_SHADOW_SCALE_SHIFT)) -#else -#define KASAN_SHADOW_SIZE \ - (_AC(1, UL) << (_REGION2_SHIFT - KASAN_SHADOW_SCALE_SHIFT)) -#endif #define KASAN_SHADOW_OFFSET _AC(CONFIG_KASAN_SHADOW_OFFSET, UL) #define KASAN_SHADOW_START KASAN_SHADOW_OFFSET #define KASAN_SHADOW_END (KASAN_SHADOW_START + KASAN_SHADOW_SIZE) extern void kasan_early_init(void); -extern void kasan_copy_shadow(pgd_t *dst); +extern void kasan_copy_shadow_mapping(void); extern void kasan_free_early_identity(void); extern unsigned long kasan_vmax; + +/* + * Estimate kasan memory requirements, which it will reserve + * at the very end of available physical memory. To estimate + * that, we take into account that kasan would require + * 1/8 of available physical memory (for shadow memory) + + * creating page tables for the whole memory + shadow memory + * region (1 + 1/8). To keep page tables estimates simple take + * the double of combined ptes size. + * + * physmem parameter has to be already adjusted if not entire physical memory + * would be used (e.g. due to effect of "mem=" option). + */ +static inline unsigned long kasan_estimate_memory_needs(unsigned long physmem) +{ + unsigned long kasan_needs; + unsigned long pages; + /* for shadow memory */ + kasan_needs = round_up(physmem / 8, PAGE_SIZE); + /* for paging structures */ + pages = DIV_ROUND_UP(physmem + kasan_needs, PAGE_SIZE); + kasan_needs += DIV_ROUND_UP(pages, _PAGE_ENTRIES) * _PAGE_TABLE_SIZE * 2; + + return kasan_needs; +} #else static inline void kasan_early_init(void) { } -static inline void kasan_copy_shadow(pgd_t *dst) { } +static inline void kasan_copy_shadow_mapping(void) { } static inline void kasan_free_early_identity(void) { } +static inline unsigned long kasan_estimate_memory_needs(unsigned long physmem) { return 0; } #endif #endif diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index 612ed3c6d581..69ce9191eaf1 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -116,7 +116,7 @@ struct lowcore { /* Address space pointer. */ __u64 kernel_asce; /* 0x0380 */ __u64 user_asce; /* 0x0388 */ - __u64 vdso_asce; /* 0x0390 */ + __u8 pad_0x0390[0x0398-0x0390]; /* 0x0390 */ /* * The lpp and current_pid fields form a @@ -134,7 +134,7 @@ struct lowcore { __u32 spinlock_index; /* 0x03b0 */ __u32 fpu_flags; /* 0x03b4 */ __u64 percpu_offset; /* 0x03b8 */ - __u64 vdso_per_cpu_data; /* 0x03c0 */ + __u8 pad_0x03c0[0x03c8-0x03c0]; /* 0x03c0 */ __u64 machine_flags; /* 0x03c8 */ __u64 gmap; /* 0x03d0 */ __u8 pad_0x03d8[0x0400-0x03d8]; /* 0x03d8 */ diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index c9f3d8a52756..5dc49c467319 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -71,39 +71,18 @@ static inline int init_new_context(struct task_struct *tsk, #define destroy_context(mm) do { } while (0) -static inline void set_user_asce(struct mm_struct *mm) -{ - S390_lowcore.user_asce = mm->context.asce; - __ctl_load(S390_lowcore.user_asce, 1, 1); - clear_cpu_flag(CIF_ASCE_PRIMARY); -} - -static inline void clear_user_asce(void) -{ - S390_lowcore.user_asce = S390_lowcore.kernel_asce; - __ctl_load(S390_lowcore.kernel_asce, 1, 1); - set_cpu_flag(CIF_ASCE_PRIMARY); -} - -mm_segment_t enable_sacf_uaccess(void); -void disable_sacf_uaccess(mm_segment_t old_fs); - static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, struct task_struct *tsk) { int cpu = smp_processor_id(); - S390_lowcore.user_asce = next->context.asce; + if (next == &init_mm) + S390_lowcore.user_asce = s390_invalid_asce; + else + S390_lowcore.user_asce = next->context.asce; cpumask_set_cpu(cpu, &next->context.cpu_attach_mask); - /* Clear previous user-ASCE from CR1 and CR7 */ - if (!test_cpu_flag(CIF_ASCE_PRIMARY)) { - __ctl_load(S390_lowcore.kernel_asce, 1, 1); - set_cpu_flag(CIF_ASCE_PRIMARY); - } - if (test_cpu_flag(CIF_ASCE_SECONDARY)) { - __ctl_load(S390_lowcore.vdso_asce, 7, 7); - clear_cpu_flag(CIF_ASCE_SECONDARY); - } + /* Clear previous user-ASCE from CR7 */ + __ctl_load(s390_invalid_asce, 7, 7); if (prev != next) cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask); } @@ -122,7 +101,7 @@ static inline void finish_arch_post_lock_switch(void) __tlb_flush_mm_lazy(mm); preempt_enable(); } - set_fs(current->thread.mm_segment); + __ctl_load(S390_lowcore.user_asce, 7, 7); } #define enter_lazy_tlb(mm,tsk) do { } while (0) @@ -133,7 +112,7 @@ static inline void activate_mm(struct mm_struct *prev, { switch_mm(prev, next, current); cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); - set_user_asce(next); + __ctl_load(S390_lowcore.user_asce, 7, 7); } #endif /* __S390_MMU_CONTEXT_H */ diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index b5dbae78969b..794746a32806 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -23,6 +23,7 @@ extern pgd_t swapper_pg_dir[]; extern void paging_init(void); +extern unsigned long s390_invalid_asce; enum { PG_DIRECT_MAP_4K = 0, @@ -79,15 +80,15 @@ extern unsigned long zero_page_mask; /* * The vmalloc and module area will always be on the topmost area of the - * kernel mapping. We reserve 128GB (64bit) for vmalloc and modules. - * On 64 bit kernels we have a 2GB area at the top of the vmalloc area where - * modules will reside. That makes sure that inter module branches always - * happen without trampolines and in addition the placement within a 2GB frame - * is branch prediction unit friendly. + * kernel mapping. 512GB are reserved for vmalloc by default. + * At the top of the vmalloc area a 2GB area is reserved where modules + * will reside. That makes sure that inter module branches always + * happen without trampolines and in addition the placement within a + * 2GB frame is branch prediction unit friendly. */ extern unsigned long VMALLOC_START; extern unsigned long VMALLOC_END; -#define VMALLOC_DEFAULT_SIZE ((128UL << 30) - MODULES_LEN) +#define VMALLOC_DEFAULT_SIZE ((512UL << 30) - MODULES_LEN) extern struct page *vmemmap; extern unsigned long vmemmap_size; diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 962da04234af..6b7269f51f83 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -14,8 +14,6 @@ #include <linux/bits.h> -#define CIF_ASCE_PRIMARY 0 /* primary asce needs fixup / uaccess */ -#define CIF_ASCE_SECONDARY 1 /* secondary asce needs fixup / uaccess */ #define CIF_NOHZ_DELAY 2 /* delay HZ disable for a tick */ #define CIF_FPU 3 /* restore FPU registers */ #define CIF_IGNORE_IRQ 4 /* ignore interrupt (for udelay) */ @@ -23,8 +21,6 @@ #define CIF_MCCK_GUEST 6 /* machine check happening in guest */ #define CIF_DEDICATED_CPU 7 /* this CPU is dedicated */ -#define _CIF_ASCE_PRIMARY BIT(CIF_ASCE_PRIMARY) -#define _CIF_ASCE_SECONDARY BIT(CIF_ASCE_SECONDARY) #define _CIF_NOHZ_DELAY BIT(CIF_NOHZ_DELAY) #define _CIF_FPU BIT(CIF_FPU) #define _CIF_IGNORE_IRQ BIT(CIF_IGNORE_IRQ) @@ -102,8 +98,6 @@ extern void __bpon(void); #define HAVE_ARCH_PICK_MMAP_LAYOUT -typedef unsigned int mm_segment_t; - /* * Thread structure */ @@ -116,7 +110,6 @@ struct thread_struct { unsigned long hardirq_timer; /* task cputime in hardirq context */ unsigned long softirq_timer; /* task cputime in softirq context */ unsigned long sys_call_table; /* system call table address */ - mm_segment_t mm_segment; unsigned long gmap_addr; /* address of last gmap fault. */ unsigned int gmap_write_flag; /* gmap fault write indication */ unsigned int gmap_int_code; /* int code of last gmap fault */ @@ -318,14 +311,10 @@ static __always_inline void __noreturn disabled_wait(void) } /* - * Basic Machine Check/Program Check Handler. + * Basic Program Check Handler. */ - extern void s390_base_pgm_handler(void); -extern void s390_base_ext_handler(void); - extern void (*s390_base_pgm_handler_fn)(void); -extern void (*s390_base_ext_handler_fn)(void); #define ARCH_LOW_ADDRESS_LIMIT 0x7fffffffUL diff --git a/arch/s390/include/asm/ptrace.h b/arch/s390/include/asm/ptrace.h index 16b3e4396312..73ca7f7cac33 100644 --- a/arch/s390/include/asm/ptrace.h +++ b/arch/s390/include/asm/ptrace.h @@ -87,6 +87,7 @@ struct pt_regs unsigned int int_parm; unsigned long int_parm_long; unsigned long flags; + unsigned long cr1; }; /* diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h index a7bdd128d85b..5763769a39b6 100644 --- a/arch/s390/include/asm/sclp.h +++ b/arch/s390/include/asm/sclp.h @@ -12,7 +12,12 @@ #include <asm/cpu.h> #define SCLP_CHP_INFO_MASK_SIZE 32 -#define SCLP_MAX_CORES 256 +#define EARLY_SCCB_SIZE PAGE_SIZE +#define SCLP_MAX_CORES 512 +/* 144 + 16 * SCLP_MAX_CORES + 2 * (SCLP_MAX_CORES - 1) */ +#define EXT_SCCB_READ_SCP (3 * PAGE_SIZE) +/* 24 + 16 * SCLP_MAX_CORES */ +#define EXT_SCCB_READ_CPU (3 * PAGE_SIZE) struct sclp_chp_info { u8 recognized[SCLP_CHP_INFO_MASK_SIZE]; diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index bdb242a1544e..3e388fa208d4 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -16,8 +16,6 @@ #define EARLY_SCCB_OFFSET 0x11000 #define HEAD_END 0x12000 -#define EARLY_SCCB_SIZE PAGE_SIZE - /* * Machine features detected in early.c */ @@ -88,10 +86,8 @@ extern unsigned int zlib_dfltcc_support; #define ZLIB_DFLTCC_FULL_DEBUG 4 extern int noexec_disabled; -extern int memory_end_set; -extern unsigned long memory_end; +extern unsigned long ident_map_size; extern unsigned long vmalloc_size; -extern unsigned long max_physmem_end; /* The Write Back bit position in the physaddr is given by the SLPC PCI */ extern unsigned long mio_wb_bit_mask; diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h index 13a04fcf7762..ce788f3e534d 100644 --- a/arch/s390/include/asm/thread_info.h +++ b/arch/s390/include/asm/thread_info.h @@ -18,7 +18,7 @@ #else #define THREAD_SIZE_ORDER 2 #endif -#define BOOT_STACK_ORDER 2 +#define BOOT_STACK_SIZE (PAGE_SIZE << 2) #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER) #ifndef __ASSEMBLY__ diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h index 289aaff4d365..c8e244ecdfde 100644 --- a/arch/s390/include/asm/timex.h +++ b/arch/s390/include/asm/timex.h @@ -49,6 +49,13 @@ static inline void set_clock_comparator(__u64 time) asm volatile("sckc %0" : : "Q" (time)); } +static inline void set_tod_programmable_field(u16 val) +{ + register unsigned long reg0 asm("0") = val; + + asm volatile("sckpf" : : "d" (reg0)); +} + void clock_comparator_work(void); void __init time_early_init(void); diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h index c868e7ee49b3..c6707885e7c2 100644 --- a/arch/s390/include/asm/uaccess.h +++ b/arch/s390/include/asm/uaccess.h @@ -18,23 +18,7 @@ #include <asm/extable.h> #include <asm/facility.h> -/* - * The fs value determines whether argument validity checking should be - * performed or not. If get_fs() == USER_DS, checking is performed, with - * get_fs() == KERNEL_DS, checking is bypassed. - * - * For historical reasons, these macros are grossly misnamed. - */ - -#define KERNEL_DS (0) -#define KERNEL_DS_SACF (1) -#define USER_DS (2) -#define USER_DS_SACF (3) - -#define get_fs() (current->thread.mm_segment) -#define uaccess_kernel() ((get_fs() & 2) == KERNEL_DS) - -void set_fs(mm_segment_t fs); +void debug_user_asce(void); static inline int __range_ok(unsigned long addr, unsigned long size) { @@ -88,7 +72,7 @@ int __get_user_bad(void) __attribute__((noreturn)); static __always_inline int __put_user_fn(void *x, void __user *ptr, unsigned long size) { - unsigned long spec = 0x010000UL; + unsigned long spec = 0x810000UL; int rc; switch (size) { @@ -121,7 +105,7 @@ static __always_inline int __put_user_fn(void *x, void __user *ptr, unsigned lon static __always_inline int __get_user_fn(void *x, const void __user *ptr, unsigned long size) { - unsigned long spec = 0x01UL; + unsigned long spec = 0x81UL; int rc; switch (size) { diff --git a/arch/s390/include/asm/vdso.h b/arch/s390/include/asm/vdso.h index 29b44a930e71..f65590889054 100644 --- a/arch/s390/include/asm/vdso.h +++ b/arch/s390/include/asm/vdso.h @@ -12,32 +12,9 @@ #ifndef __ASSEMBLY__ -/* - * Note about the vdso_data and vdso_per_cpu_data structures: - * - * NEVER USE THEM IN USERSPACE CODE DIRECTLY. The layout of the - * structure is supposed to be known only to the function in the vdso - * itself and may change without notice. - */ - -struct vdso_per_cpu_data { - /* - * Note: node_id and cpu_nr must be at adjacent memory locations. - * VDSO userspace must read both values with a single instruction. - */ - union { - __u64 getcpu_val; - struct { - __u32 node_id; - __u32 cpu_nr; - }; - }; -}; - extern struct vdso_data *vdso_data; -int vdso_alloc_per_cpu(struct lowcore *lowcore); -void vdso_free_per_cpu(struct lowcore *lowcore); +void vdso_getcpu_init(void); #endif /* __ASSEMBLY__ */ #endif /* __S390_VDSO_H__ */ diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 483051e10db3..79724d861dc9 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -13,7 +13,6 @@ #include <linux/purgatory.h> #include <linux/pgtable.h> #include <asm/idle.h> -#include <asm/vdso.h> #include <asm/gmap.h> #include <asm/nmi.h> #include <asm/stacktrace.h> @@ -48,6 +47,7 @@ int main(void) OFFSET(__PT_INT_PARM, pt_regs, int_parm); OFFSET(__PT_INT_PARM_LONG, pt_regs, int_parm_long); OFFSET(__PT_FLAGS, pt_regs, flags); + OFFSET(__PT_CR1, pt_regs, cr1); DEFINE(__PT_SIZE, sizeof(struct pt_regs)); BLANK(); /* stack_frame offsets */ @@ -59,8 +59,6 @@ int main(void) OFFSET(__SF_SIE_REASON, stack_frame, empty1[3]); OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[4]); BLANK(); - OFFSET(__VDSO_GETCPU_VAL, vdso_per_cpu_data, getcpu_val); - BLANK(); /* idle data offsets */ OFFSET(__CLOCK_IDLE_ENTER, s390_idle_data, clock_idle_enter); OFFSET(__CLOCK_IDLE_EXIT, s390_idle_data, clock_idle_exit); @@ -138,12 +136,11 @@ int main(void) OFFSET(__LC_RESTART_FN, lowcore, restart_fn); OFFSET(__LC_RESTART_DATA, lowcore, restart_data); OFFSET(__LC_RESTART_SOURCE, lowcore, restart_source); + OFFSET(__LC_KERNEL_ASCE, lowcore, kernel_asce); OFFSET(__LC_USER_ASCE, lowcore, user_asce); - OFFSET(__LC_VDSO_ASCE, lowcore, vdso_asce); OFFSET(__LC_LPP, lowcore, lpp); OFFSET(__LC_CURRENT_PID, lowcore, current_pid); OFFSET(__LC_PERCPU_OFFSET, lowcore, percpu_offset); - OFFSET(__LC_VDSO_PER_CPU, lowcore, vdso_per_cpu_data); OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags); OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count); OFFSET(__LC_GMAP, lowcore, gmap); diff --git a/arch/s390/kernel/base.S b/arch/s390/kernel/base.S index b79e0fd571f8..d255c69c1779 100644 --- a/arch/s390/kernel/base.S +++ b/arch/s390/kernel/base.S @@ -11,32 +11,10 @@ #include <asm/asm-offsets.h> #include <asm/nospec-insn.h> #include <asm/ptrace.h> -#include <asm/sigp.h> GEN_BR_THUNK %r9 GEN_BR_THUNK %r14 -ENTRY(s390_base_ext_handler) - stmg %r0,%r15,__LC_SAVE_AREA_ASYNC - basr %r13,0 -0: aghi %r15,-STACK_FRAME_OVERHEAD - larl %r1,s390_base_ext_handler_fn - lg %r9,0(%r1) - ltgr %r9,%r9 - jz 1f - BASR_EX %r14,%r9 -1: lmg %r0,%r15,__LC_SAVE_AREA_ASYNC - ni __LC_EXT_OLD_PSW+1,0xfd # clear wait state bit - lpswe __LC_EXT_OLD_PSW -ENDPROC(s390_base_ext_handler) - - .section .bss - .align 8 - .globl s390_base_ext_handler_fn -s390_base_ext_handler_fn: - .quad 0 - .previous - ENTRY(s390_base_pgm_handler) stmg %r0,%r15,__LC_SAVE_AREA_SYNC basr %r13,0 diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 705844f73934..cc89763a4d3c 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -169,12 +169,10 @@ static noinline __init void setup_lowcore_early(void) { psw_t psw; + psw.addr = (unsigned long)s390_base_pgm_handler; psw.mask = PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA; if (IS_ENABLED(CONFIG_KASAN)) psw.mask |= PSW_MASK_DAT; - psw.addr = (unsigned long) s390_base_ext_handler; - S390_lowcore.external_new_psw = psw; - psw.addr = (unsigned long) s390_base_pgm_handler; S390_lowcore.program_new_psw = psw; s390_base_pgm_handler_fn = early_pgm_check_handler; S390_lowcore.preempt_count = INIT_PREEMPT_COUNT; diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 92beb1444644..8bb9ebb71c4b 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -55,7 +55,7 @@ _TIF_WORK = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ _TIF_UPROBE | _TIF_GUARDED_STORAGE | _TIF_PATCH_PENDING) _TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ _TIF_SYSCALL_TRACEPOINT) -_CIF_WORK = (_CIF_ASCE_PRIMARY | _CIF_ASCE_SECONDARY | _CIF_FPU) +_CIF_WORK = (_CIF_FPU) _PIF_WORK = (_PIF_PER_TRAP | _PIF_SYSCALL_RESTART) _LPP_OFFSET = __LC_LPP @@ -90,6 +90,12 @@ _LPP_OFFSET = __LC_LPP #endif .endm + .macro DEBUG_USER_ASCE +#ifdef CONFIG_DEBUG_USER_ASCE + brasl %r14,debug_user_asce +#endif + .endm + .macro CHECK_VMAP_STACK savearea,oklabel #ifdef CONFIG_VMAP_STACK lgr %r14,%r15 @@ -110,9 +116,9 @@ _LPP_OFFSET = __LC_LPP #endif .endm - .macro SWITCH_ASYNC savearea,timer + .macro SWITCH_ASYNC savearea,timer,clock tmhh %r8,0x0001 # interrupting from user ? - jnz 2f + jnz 4f #if IS_ENABLED(CONFIG_KVM) lgr %r14,%r9 larl %r13,.Lsie_gmap @@ -125,10 +131,26 @@ _LPP_OFFSET = __LC_LPP #endif 0: larl %r13,.Lpsw_idle_exit cgr %r13,%r9 - jne 1f + jne 3f - mvc __CLOCK_IDLE_EXIT(8,%r2), __LC_INT_CLOCK - mvc __TIMER_IDLE_EXIT(8,%r2), __LC_ASYNC_ENTER_TIMER + larl %r1,smp_cpu_mtid + llgf %r1,0(%r1) + ltgr %r1,%r1 + jz 2f # no SMT, skip mt_cycles calculation + .insn rsy,0xeb0000000017,%r1,5,__SF_EMPTY+80(%r15) + larl %r3,mt_cycles + ag %r3,__LC_PERCPU_OFFSET + la %r4,__SF_EMPTY+16(%r15) +1: lg %r0,0(%r3) + slg %r0,0(%r4) + alg %r0,64(%r4) + stg %r0,0(%r3) + la %r3,8(%r3) + la %r4,8(%r4) + brct %r1,1b + +2: mvc __CLOCK_IDLE_EXIT(8,%r2), \clock + mvc __TIMER_IDLE_EXIT(8,%r2), \timer # account system time going idle ni __LC_CPU_FLAGS+7,255-_CIF_ENABLED_WAIT @@ -146,17 +168,17 @@ _LPP_OFFSET = __LC_LPP mvc __LC_LAST_UPDATE_TIMER(8),__TIMER_IDLE_EXIT(%r2) nihh %r8,0xfcfd # clear wait state and irq bits -1: lg %r14,__LC_ASYNC_STACK # are we already on the target stack? +3: lg %r14,__LC_ASYNC_STACK # are we already on the target stack? slgr %r14,%r15 srag %r14,%r14,STACK_SHIFT - jnz 3f + jnz 5f CHECK_STACK \savearea aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) - j 4f -2: UPDATE_VTIME %r14,%r15,\timer + j 6f +4: UPDATE_VTIME %r14,%r15,\timer BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP -3: lg %r15,__LC_ASYNC_STACK # load async stack -4: la %r11,STACK_FRAME_OVERHEAD(%r15) +5: lg %r15,__LC_ASYNC_STACK # load async stack +6: la %r11,STACK_FRAME_OVERHEAD(%r15) .endm .macro UPDATE_VTIME w1,w2,enter_timer @@ -327,7 +349,7 @@ ENTRY(sie64a) BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) .Lsie_skip: ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce .Lsie_done: # some program checks are suppressing. C code (e.g. do_protection_exception) # will rewind the PSW by the ILC, which is often 4 bytes in case of SIE. There @@ -380,6 +402,7 @@ ENTRY(system_call) lg %r12,__LC_CURRENT lghi %r14,_PIF_SYSCALL .Lsysc_per: + lctlg %c1,%c1,__LC_KERNEL_ASCE lghi %r13,__TASK_thread lg %r15,__LC_KERNEL_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs @@ -427,11 +450,9 @@ ENTRY(system_call) jnz .Lsysc_work TSTMSK __TI_flags(%r12),_TIF_WORK jnz .Lsysc_work # check for work - TSTMSK __LC_CPU_FLAGS,(_CIF_WORK-_CIF_FPU) - jnz .Lsysc_work + DEBUG_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP -.Lsysc_restore: - DISABLE_INTS TSTMSK __LC_CPU_FLAGS, _CIF_FPU jz .Lsysc_skip_fpu brasl %r14,load_fpu_regs @@ -469,8 +490,6 @@ ENTRY(system_call) jo .Lsysc_sigpending TSTMSK __TI_flags(%r12),_TIF_NOTIFY_RESUME jo .Lsysc_notify_resume - TSTMSK __LC_CPU_FLAGS,(_CIF_ASCE_PRIMARY|_CIF_ASCE_SECONDARY) - jnz .Lsysc_asce j .Lsysc_return # @@ -481,26 +500,6 @@ ENTRY(system_call) jg schedule # -# _CIF_ASCE_PRIMARY and/or _CIF_ASCE_SECONDARY set, load user space asce -# -.Lsysc_asce: - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_SECONDARY - lctlg %c7,%c7,__LC_VDSO_ASCE # load secondary asce - TSTMSK __LC_CPU_FLAGS,_CIF_ASCE_PRIMARY - jz .Lsysc_return -#ifndef CONFIG_HAVE_MARCH_Z10_FEATURES - tm __LC_STFLE_FAC_LIST+3,0x10 # has MVCOS ? - jnz .Lsysc_set_fs_fixup - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_PRIMARY - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce - j .Lsysc_return -.Lsysc_set_fs_fixup: -#endif - larl %r14,.Lsysc_return - jg set_fs_fixup - - -# # _TIF_SIGPENDING is set, call do_signal # .Lsysc_sigpending: @@ -636,8 +635,11 @@ ENTRY(pgm_check_handler) 0: lg %r12,__LC_CURRENT lghi %r11,0 lmg %r8,%r9,__LC_PGM_OLD_PSW - tmhh %r8,0x0001 # test problem state bit - jnz 3f # -> fault in user space + tmhh %r8,0x0001 # coming from user space? + jno .Lpgm_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE + j 3f +.Lpgm_skip_asce: #if IS_ENABLED(CONFIG_KVM) # cleanup critical section for program checks in sie64a lgr %r14,%r9 @@ -648,7 +650,7 @@ ENTRY(pgm_check_handler) jhe 1f lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce larl %r9,sie_exit # skip forward to sie_exit lghi %r11,_PIF_GUEST_FAULT #endif @@ -709,10 +711,20 @@ ENTRY(pgm_check_handler) .Lpgm_return: LOCKDEP_SYS_EXIT tm __PT_PSW+1(%r11),0x01 # returning to user ? - jno .Lsysc_restore + jno .Lpgm_restore TSTMSK __PT_FLAGS(%r11),_PIF_SYSCALL jo .Lsysc_do_syscall j .Lsysc_tif +.Lpgm_restore: + DISABLE_INTS + TSTMSK __LC_CPU_FLAGS, _CIF_FPU + jz .Lpgm_skip_fpu + brasl %r14,load_fpu_regs +.Lpgm_skip_fpu: + mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) + stpt __LC_EXIT_TIMER + lmg %r0,%r15,__PT_R0(%r11) + b __LC_RETURN_LPSWE # # PER event in supervisor state, must be kprobes @@ -745,7 +757,7 @@ ENTRY(io_int_handler) stmg %r8,%r15,__LC_SAVE_AREA_ASYNC lg %r12,__LC_CURRENT lmg %r8,%r9,__LC_IO_OLD_PSW - SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER,__LC_INT_CLOCK stmg %r0,%r7,__PT_R0(%r11) # clear user controlled registers to prevent speculative use xgr %r0,%r0 @@ -759,6 +771,10 @@ ENTRY(io_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC stmg %r8,%r9,__PT_PSW(%r11) + tm __PT_PSW+1(%r11),0x01 # coming from user space? + jno .Lio_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE +.Lio_skip_asce: mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) TSTMSK __LC_CPU_FLAGS,_CIF_IGNORE_IRQ @@ -790,6 +806,8 @@ ENTRY(io_int_handler) mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) tm __PT_PSW+1(%r11),0x01 # returning to user ? jno .Lio_exit_kernel + DEBUG_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE BPEXIT __TI_flags(%r12),_TIF_ISOLATE_BP stpt __LC_EXIT_TIMER .Lio_exit_kernel: @@ -855,30 +873,9 @@ ENTRY(io_int_handler) jo .Lio_guarded_storage TSTMSK __LC_CPU_FLAGS,_CIF_FPU jo .Lio_vxrs - TSTMSK __LC_CPU_FLAGS,(_CIF_ASCE_PRIMARY|_CIF_ASCE_SECONDARY) - jnz .Lio_asce j .Lio_return # -# _CIF_ASCE_PRIMARY and/or CIF_ASCE_SECONDARY set, load user space asce -# -.Lio_asce: - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_SECONDARY - lctlg %c7,%c7,__LC_VDSO_ASCE # load secondary asce - TSTMSK __LC_CPU_FLAGS,_CIF_ASCE_PRIMARY - jz .Lio_return -#ifndef CONFIG_HAVE_MARCH_Z10_FEATURES - tm __LC_STFLE_FAC_LIST+3,0x10 # has MVCOS ? - jnz .Lio_set_fs_fixup - ni __LC_CPU_FLAGS+7,255-_CIF_ASCE_PRIMARY - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce - j .Lio_return -.Lio_set_fs_fixup: -#endif - larl %r14,.Lio_return - jg set_fs_fixup - -# # CIF_FPU is set, restore floating-point controls and floating-point registers. # .Lio_vxrs: @@ -945,7 +942,7 @@ ENTRY(ext_int_handler) stmg %r8,%r15,__LC_SAVE_AREA_ASYNC lg %r12,__LC_CURRENT lmg %r8,%r9,__LC_EXT_OLD_PSW - SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER,__LC_INT_CLOCK stmg %r0,%r7,__PT_R0(%r11) # clear user controlled registers to prevent speculative use xgr %r0,%r0 @@ -959,6 +956,10 @@ ENTRY(ext_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC stmg %r8,%r9,__PT_PSW(%r11) + tm __PT_PSW+1(%r11),0x01 # coming from user space? + jno .Lext_skip_asce + lctlg %c1,%c1,__LC_KERNEL_ASCE +.Lext_skip_asce: lghi %r1,__LC_EXT_PARAMS2 mvc __PT_INT_CODE(4,%r11),__LC_EXT_CPU_ADDR mvc __PT_INT_PARM(4,%r11),__LC_EXT_PARAMS @@ -1167,7 +1168,7 @@ ENTRY(mcck_int_handler) TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_IA_VALID jno .Lmcck_panic 4: ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off - SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER + SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER,__LC_MCCK_CLOCK .Lmcck_skip: lghi %r14,__LC_GPREGS_SAVE_AREA+64 stmg %r0,%r7,__PT_R0(%r11) @@ -1183,6 +1184,9 @@ ENTRY(mcck_int_handler) xgr %r10,%r10 mvc __PT_R8(64,%r11),0(%r14) stmg %r8,%r9,__PT_PSW(%r11) + la %r14,4095 + mvc __PT_CR1(8,%r11),__LC_CREGS_SAVE_AREA-4095+8(%r14) + lctlg %c1,%c1,__LC_KERNEL_ASCE xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) lgr %r2,%r11 # pass pointer to pt_regs @@ -1198,6 +1202,7 @@ ENTRY(mcck_int_handler) brasl %r14,s390_handle_mcck TRACE_IRQS_ON .Lmcck_return: + lctlg %c1,%c1,__PT_CR1(%r11) lmg %r0,%r10,__PT_R0(%r11) mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? @@ -1274,7 +1279,7 @@ ENDPROC(stack_overflow) 1: BPENTER __SF_SIE_FLAGS(%r15),(_TIF_ISOLATE_BP|_TIF_ISOLATE_BP_GUEST) lg %r9,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r9),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + lctlg %c1,%c1,__LC_KERNEL_ASCE larl %r9,sie_exit # skip forward to sie_exit BR_EX %r14,%r11 diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h index d2ca3fe51f8e..a16c33b32ab0 100644 --- a/arch/s390/kernel/entry.h +++ b/arch/s390/kernel/entry.h @@ -83,7 +83,6 @@ long sys_s390_sthyi(unsigned long function_code, void __user *buffer, u64 __user DECLARE_PER_CPU(u64, mt_cycles[8]); void gs_load_bc_cb(struct pt_regs *regs); -void set_fs_fixup(void); unsigned long stack_alloc(void); void stack_free(unsigned long stack); diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c index b388e87a08bf..ebc1284a618b 100644 --- a/arch/s390/kernel/ftrace.c +++ b/arch/s390/kernel/ftrace.c @@ -22,56 +22,26 @@ #include "entry.h" /* - * The mcount code looks like this: - * stg %r14,8(%r15) # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 - * Total length is 24 bytes. Only the first instruction will be patched - * by ftrace_make_call / ftrace_make_nop. - * The enabled ftrace code block looks like this: + * To generate function prologue either gcc's hotpatch feature (since gcc 4.8) + * or a combination of -pg -mrecord-mcount -mnop-mcount -mfentry flags + * (since gcc 9 / clang 10) is used. + * In both cases the original and also the disabled function prologue contains + * only a single six byte instruction and looks like this: + * > brcl 0,0 # offset 0 + * To enable ftrace the code gets patched like above and afterwards looks + * like this: * > brasl %r0,ftrace_caller # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 + * + * The instruction will be patched by ftrace_make_call / ftrace_make_nop. * The ftrace function gets called with a non-standard C function call ABI * where r0 contains the return address. It is also expected that the called * function only clobbers r0 and r1, but restores r2-r15. * For module code we can't directly jump to ftrace caller, but need a * trampoline (ftrace_plt), which clobbers also r1. - * The return point of the ftrace function has offset 24, so execution - * continues behind the mcount block. - * The disabled ftrace code block looks like this: - * > jg .+24 # offset 0 - * larl %r1,<&counter> # offset 6 - * brasl %r14,_mcount # offset 12 - * lg %r14,8(%r15) # offset 18 - * The jg instruction branches to offset 24 to skip as many instructions - * as possible. - * In case we use gcc's hotpatch feature the original and also the disabled - * function prologue contains only a single six byte instruction and looks - * like this: - * > brcl 0,0 # offset 0 - * To enable ftrace the code gets patched like above and afterwards looks - * like this: - * > brasl %r0,ftrace_caller # offset 0 */ unsigned long ftrace_plt; -static inline void ftrace_generate_orig_insn(struct ftrace_insn *insn) -{ -#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT) - /* brcl 0,0 */ - insn->opc = 0xc004; - insn->disp = 0; -#else - /* stg r14,8(r15) */ - insn->opc = 0xe3e0; - insn->disp = 0xf0080024; -#endif -} - int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr, unsigned long addr) { @@ -85,15 +55,10 @@ int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, if (copy_from_kernel_nofault(&old, (void *) rec->ip, sizeof(old))) return -EFAULT; - if (addr == MCOUNT_ADDR) { - /* Initial code replacement */ - ftrace_generate_orig_insn(&orig); - ftrace_generate_nop_insn(&new); - } else { - /* Replace ftrace call with a nop. */ - ftrace_generate_call_insn(&orig, rec->ip); - ftrace_generate_nop_insn(&new); - } + /* Replace ftrace call with a nop. */ + ftrace_generate_call_insn(&orig, rec->ip); + ftrace_generate_nop_insn(&new); + /* Verify that the to be replaced code matches what we expect. */ if (memcmp(&orig, &old, sizeof(old))) return -EINVAL; diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 8b88dbbda7df..0c253886da78 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -18,12 +18,7 @@ __HEAD ENTRY(startup_continue) - tm __LC_STFLE_FAC_LIST+5,0x80 # LPP available ? - jz 0f - xc __LC_LPP+1(7,0),__LC_LPP+1 # clear lpp and current_pid - mvi __LC_LPP,0x80 # and set LPP_MAGIC - .insn s,0xb2800000,__LC_LPP # load program parameter -0: larl %r1,tod_clock_base + larl %r1,tod_clock_base mvc 0(16,%r1),__LC_BOOT_CLOCK larl %r13,.LPG1 # get base # diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S index 7458dcfd6464..faf64c2f90f5 100644 --- a/arch/s390/kernel/mcount.S +++ b/arch/s390/kernel/mcount.S @@ -33,11 +33,6 @@ ENDPROC(ftrace_stub) #define TRACED_FUNC_FRAME_SIZE STACK_FRAME_OVERHEAD #endif -ENTRY(_mcount) - BR_EX %r14 -ENDPROC(_mcount) -EXPORT_SYMBOL(_mcount) - ENTRY(ftrace_caller) .globl ftrace_regs_caller .set ftrace_regs_caller,ftrace_caller @@ -46,9 +41,6 @@ ENTRY(ftrace_caller) ipm %r14 # don't put any instructions sllg %r14,%r14,16 # clobbering CC before this point lgr %r1,%r15 -#if !(defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT)) - aghi %r0,MCOUNT_RETURN_FIXUP -#endif # allocate stack frame for ftrace_caller to contain traced function aghi %r15,-TRACED_FUNC_FRAME_SIZE stg %r1,__SF_BACKCHAIN(%r15) diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index ec801d3bbb37..bc3ca54edfb4 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -94,7 +94,6 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp, /* Save access registers to new thread structure. */ save_access_regs(&p->thread.acrs[0]); /* start new process with ar4 pointing to the correct address space */ - p->thread.mm_segment = get_fs(); /* Don't copy debug registers */ memset(&p->thread.per_user, 0, sizeof(p->thread.per_user)); memset(&p->thread.per_event, 0, sizeof(p->thread.per_event)); @@ -208,16 +207,3 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) ret = PAGE_ALIGN(mm->brk + brk_rnd()); return (ret > mm->brk) ? ret : mm->brk; } - -void set_fs_fixup(void) -{ - struct pt_regs *regs = current_pt_regs(); - static bool warned; - - set_fs(USER_DS); - if (warned) - return; - WARN(1, "Unbalanced set_fs - int code: 0x%x\n", regs->int_code); - show_registers(regs); - warned = true; -} diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 4d843e64496f..1f16a03be995 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -49,6 +49,7 @@ #include <linux/memory.h> #include <linux/compat.h> #include <linux/start_kernel.h> +#include <linux/hugetlb.h> #include <asm/boot_data.h> #include <asm/ipl.h> @@ -94,10 +95,8 @@ char elf_platform[ELF_PLATFORM_SIZE]; unsigned long int_hwcap = 0; int __bootdata(noexec_disabled); -int __bootdata(memory_end_set); -unsigned long __bootdata(memory_end); +unsigned long __bootdata(ident_map_size); unsigned long __bootdata(vmalloc_size); -unsigned long __bootdata(max_physmem_end); struct mem_detect_info __bootdata(mem_detect); struct exception_table_entry *__bootdata_preserved(__start_dma_ex_table); @@ -336,6 +335,7 @@ int __init arch_early_irq_init(void) if (!stack) panic("Couldn't allocate async stack"); S390_lowcore.async_stack = stack + STACK_INIT_OFFSET; + udelay_enable(); return 0; } @@ -556,24 +556,25 @@ static void __init setup_resources(void) #endif } -static void __init setup_memory_end(void) +static void __init setup_ident_map_size(void) { unsigned long vmax, tmp; /* Choose kernel address space layout: 3 or 4 levels. */ - tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE; + tmp = ident_map_size / PAGE_SIZE; tmp = tmp * (sizeof(struct page) + PAGE_SIZE); if (tmp + vmalloc_size + MODULES_LEN <= _REGION2_SIZE) vmax = _REGION2_SIZE; /* 3-level kernel page table */ else vmax = _REGION1_SIZE; /* 4-level kernel page table */ + /* module area is at the end of the kernel address space. */ + MODULES_END = vmax; if (is_prot_virt_host()) - adjust_to_uv_max(&vmax); + adjust_to_uv_max(&MODULES_END); #ifdef CONFIG_KASAN - vmax = kasan_vmax; + vmax = _REGION1_SIZE; + MODULES_END = kasan_vmax; #endif - /* module area is at the end of the kernel address space. */ - MODULES_END = vmax; MODULES_VADDR = MODULES_END - MODULES_LEN; VMALLOC_END = MODULES_VADDR; VMALLOC_START = VMALLOC_END - vmalloc_size; @@ -587,22 +588,22 @@ static void __init setup_memory_end(void) tmp = min(tmp, 1UL << MAX_PHYSMEM_BITS); vmemmap = (struct page *) tmp; - /* Take care that memory_end is set and <= vmemmap */ - memory_end = min(memory_end ?: max_physmem_end, (unsigned long)vmemmap); + /* Take care that ident_map_size <= vmemmap */ + ident_map_size = min(ident_map_size, (unsigned long)vmemmap); #ifdef CONFIG_KASAN - memory_end = min(memory_end, KASAN_SHADOW_START); + ident_map_size = min(ident_map_size, KASAN_SHADOW_START); #endif - vmemmap_size = SECTION_ALIGN_UP(memory_end / PAGE_SIZE) * sizeof(struct page); + vmemmap_size = SECTION_ALIGN_UP(ident_map_size / PAGE_SIZE) * sizeof(struct page); #ifdef CONFIG_KASAN /* move vmemmap above kasan shadow only if stands in a way */ if (KASAN_SHADOW_END > (unsigned long)vmemmap && (unsigned long)vmemmap + vmemmap_size > KASAN_SHADOW_START) vmemmap = max(vmemmap, (struct page *)KASAN_SHADOW_END); #endif - max_pfn = max_low_pfn = PFN_DOWN(memory_end); - memblock_remove(memory_end, ULONG_MAX); + max_pfn = max_low_pfn = PFN_DOWN(ident_map_size); + memblock_remove(ident_map_size, ULONG_MAX); - pr_notice("The maximum memory size is %luMB\n", memory_end >> 20); + pr_notice("The maximum memory size is %luMB\n", ident_map_size >> 20); } #ifdef CONFIG_CRASH_DUMP @@ -632,12 +633,11 @@ static struct notifier_block kdump_mem_nb = { #endif /* - * Make sure that the area behind memory_end is protected + * Make sure that the area above identity mapping is protected */ -static void __init reserve_memory_end(void) +static void __init reserve_above_ident_map(void) { - if (memory_end_set) - memblock_reserve(memory_end, ULONG_MAX); + memblock_reserve(ident_map_size, ULONG_MAX); } /* @@ -674,7 +674,7 @@ static void __init reserve_crashkernel(void) phys_addr_t low, high; int rc; - rc = parse_crashkernel(boot_command_line, memory_end, &crash_size, + rc = parse_crashkernel(boot_command_line, ident_map_size, &crash_size, &crash_base); crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN); @@ -1128,7 +1128,7 @@ void __init setup_arch(char **cmdline_p) setup_control_program_code(); /* Do some memory reservations *before* memory is added to memblock */ - reserve_memory_end(); + reserve_above_ident_map(); reserve_oldmem(); reserve_kernel(); reserve_initrd(); @@ -1143,10 +1143,12 @@ void __init setup_arch(char **cmdline_p) remove_oldmem(); setup_uv(); - setup_memory_end(); + setup_ident_map_size(); setup_memory(); - dma_contiguous_reserve(memory_end); + dma_contiguous_reserve(ident_map_size); vmcp_cma_reserve(); + if (MACHINE_HAS_EDAT2) + hugetlb_cma_reserve(PUD_SHIFT - PAGE_SHIFT); check_initrd(); reserve_crashkernel(); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 390d97daa2b3..27c763014114 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -47,7 +47,6 @@ #include <asm/vtimer.h> #include <asm/lowcore.h> #include <asm/sclp.h> -#include <asm/vdso.h> #include <asm/debug.h> #include <asm/os_info.h> #include <asm/sigp.h> @@ -55,6 +54,7 @@ #include <asm/nmi.h> #include <asm/stacktrace.h> #include <asm/topology.h> +#include <asm/vdso.h> #include "entry.h" enum { @@ -217,14 +217,10 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) lc->return_mcck_lpswe = gen_lpswe(__LC_RETURN_MCCK_PSW); if (nmi_alloc_per_cpu(lc)) goto out_async; - if (vdso_alloc_per_cpu(lc)) - goto out_mcesa; lowcore_ptr[cpu] = lc; pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, (u32)(unsigned long) lc); return 0; -out_mcesa: - nmi_free_per_cpu(lc); out_async: stack_free(async_stack); out: @@ -245,7 +241,6 @@ static void pcpu_free_lowcore(struct pcpu *pcpu) pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, 0); lowcore_ptr[pcpu - pcpu_devices] = NULL; - vdso_free_per_cpu(pcpu->lowcore); nmi_free_per_cpu(pcpu->lowcore); stack_free(async_stack); if (pcpu == &pcpu_devices[0]) @@ -265,13 +260,13 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) lc->spinlock_index = 0; lc->percpu_offset = __per_cpu_offset[cpu]; lc->kernel_asce = S390_lowcore.kernel_asce; - lc->user_asce = S390_lowcore.kernel_asce; + lc->user_asce = s390_invalid_asce; lc->machine_flags = S390_lowcore.machine_flags; lc->user_timer = lc->system_timer = lc->steal_timer = lc->avg_steal_timer = 0; __ctl_store(lc->cregs_save_area, 0, 15); lc->cregs_save_area[1] = lc->kernel_asce; - lc->cregs_save_area[7] = lc->vdso_asce; + lc->cregs_save_area[7] = lc->user_asce; save_access_regs((unsigned int *) lc->access_regs_save_area); memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list, sizeof(lc->stfle_fac_list)); @@ -859,13 +854,12 @@ static void smp_init_secondary(void) S390_lowcore.last_update_clock = get_tod_clock(); restore_access_regs(S390_lowcore.access_regs_save_area); - set_cpu_flag(CIF_ASCE_PRIMARY); - set_cpu_flag(CIF_ASCE_SECONDARY); cpu_init(); rcu_cpu_starting(cpu); preempt_disable(); init_cpu_timer(); vtime_init(); + vdso_getcpu_init(); pfault_init(); notify_cpu_starting(cpu); if (topology_cpu_dedicated(cpu)) @@ -896,24 +890,12 @@ static void __no_sanitize_address smp_start_secondary(void *cpuvoid) /* Upping and downing of CPUs */ int __cpu_up(unsigned int cpu, struct task_struct *tidle) { - struct pcpu *pcpu; - int base, i, rc; + struct pcpu *pcpu = pcpu_devices + cpu; + int rc; - pcpu = pcpu_devices + cpu; if (pcpu->state != CPU_STATE_CONFIGURED) return -EIO; - base = smp_get_base_cpu(cpu); - for (i = 0; i <= smp_cpu_mtid; i++) { - if (base + i < nr_cpu_ids) - if (cpu_online(base + i)) - break; - } - /* - * If this is the first CPU of the core to get online - * do an initial CPU reset. - */ - if (i > smp_cpu_mtid && - pcpu_sigp_retry(pcpu_devices + base, SIGP_INITIAL_CPU_RESET, 0) != + if (pcpu_sigp_retry(pcpu, SIGP_INITIAL_CPU_RESET, 0) != SIGP_CC_ORDER_CODE_ACCEPTED) return -EIO; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 0ac30ee2c633..c59cb44fbb7d 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -927,41 +927,25 @@ static ssize_t online_store(struct device *dev, */ static DEVICE_ATTR_RW(online); -static struct device_attribute *stp_attributes[] = { - &dev_attr_ctn_id, - &dev_attr_ctn_type, - &dev_attr_dst_offset, - &dev_attr_leap_seconds, - &dev_attr_online, - &dev_attr_leap_seconds_scheduled, - &dev_attr_stratum, - &dev_attr_time_offset, - &dev_attr_time_zone_offset, - &dev_attr_timing_mode, - &dev_attr_timing_state, +static struct attribute *stp_dev_attrs[] = { + &dev_attr_ctn_id.attr, + &dev_attr_ctn_type.attr, + &dev_attr_dst_offset.attr, + &dev_attr_leap_seconds.attr, + &dev_attr_online.attr, + &dev_attr_leap_seconds_scheduled.attr, + &dev_attr_stratum.attr, + &dev_attr_time_offset.attr, + &dev_attr_time_zone_offset.attr, + &dev_attr_timing_mode.attr, + &dev_attr_timing_state.attr, NULL }; +ATTRIBUTE_GROUPS(stp_dev); static int __init stp_init_sysfs(void) { - struct device_attribute **attr; - int rc; - - rc = subsys_system_register(&stp_subsys, NULL); - if (rc) - goto out; - for (attr = stp_attributes; *attr; attr++) { - rc = device_create_file(stp_subsys.dev_root, *attr); - if (rc) - goto out_unreg; - } - return 0; -out_unreg: - for (; attr >= stp_attributes; attr--) - device_remove_file(stp_subsys.dev_root, *attr); - bus_unregister(&stp_subsys); -out: - return rc; + return subsys_system_register(&stp_subsys, stp_dev_groups); } device_initcall(stp_init_sysfs); diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c index f9da5b149141..aef2edff9959 100644 --- a/arch/s390/kernel/vdso.c +++ b/arch/s390/kernel/vdso.c @@ -29,6 +29,7 @@ #include <asm/sections.h> #include <asm/vdso.h> #include <asm/facility.h> +#include <asm/timex.h> extern char vdso64_start, vdso64_end; static void *vdso64_kbase = &vdso64_start; @@ -99,60 +100,10 @@ static union { u8 page[PAGE_SIZE]; } vdso_data_store __page_aligned_data; struct vdso_data *vdso_data = (struct vdso_data *)&vdso_data_store.data; -/* - * Allocate/free per cpu vdso data. - */ -#define SEGMENT_ORDER 2 -int vdso_alloc_per_cpu(struct lowcore *lowcore) +void vdso_getcpu_init(void) { - unsigned long segment_table, page_table, page_frame; - struct vdso_per_cpu_data *vd; - - segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER); - page_table = get_zeroed_page(GFP_KERNEL); - page_frame = get_zeroed_page(GFP_KERNEL); - if (!segment_table || !page_table || !page_frame) - goto out; - arch_set_page_dat(virt_to_page(segment_table), SEGMENT_ORDER); - arch_set_page_dat(virt_to_page(page_table), 0); - - /* Initialize per-cpu vdso data page */ - vd = (struct vdso_per_cpu_data *) page_frame; - vd->cpu_nr = lowcore->cpu_nr; - vd->node_id = cpu_to_node(vd->cpu_nr); - - /* Set up page table for the vdso address space */ - memset64((u64 *)segment_table, _SEGMENT_ENTRY_EMPTY, _CRST_ENTRIES); - memset64((u64 *)page_table, _PAGE_INVALID, PTRS_PER_PTE); - - *(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table; - *(unsigned long *) page_table = _PAGE_PROTECT + page_frame; - - lowcore->vdso_asce = segment_table + - _ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT; - lowcore->vdso_per_cpu_data = page_frame; - - return 0; - -out: - free_page(page_frame); - free_page(page_table); - free_pages(segment_table, SEGMENT_ORDER); - return -ENOMEM; -} - -void vdso_free_per_cpu(struct lowcore *lowcore) -{ - unsigned long segment_table, page_table, page_frame; - - segment_table = lowcore->vdso_asce & PAGE_MASK; - page_table = *(unsigned long *) segment_table; - page_frame = *(unsigned long *) page_table; - - free_page(page_frame); - free_page(page_table); - free_pages(segment_table, SEGMENT_ORDER); + set_tod_programmable_field(smp_processor_id()); } /* @@ -225,6 +176,7 @@ static int __init vdso_init(void) { int i; + vdso_getcpu_init(); /* Calculate the size of the 64 bit vDSO */ vdso64_pages = ((&vdso64_end - &vdso64_start + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; @@ -240,8 +192,6 @@ static int __init vdso_init(void) } vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data); vdso64_pagelist[vdso64_pages] = NULL; - if (vdso_alloc_per_cpu(&S390_lowcore)) - BUG(); get_page(virt_to_page(vdso_data)); diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 13cc5a3f9abf..a6e0fb6b91d6 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -6,8 +6,9 @@ ARCH_REL_TYPE_ABS := R_390_COPY|R_390_GLOB_DAT|R_390_JMP_SLOT|R_390_RELATIVE ARCH_REL_TYPE_ABS += R_390_GOT|R_390_PLT include $(srctree)/lib/vdso/Makefile -obj-vdso64 = vdso_user_wrapper.o note.o getcpu.o -obj-cvdso64 = vdso64_generic.o +obj-vdso64 = vdso_user_wrapper.o note.o +obj-cvdso64 = vdso64_generic.o getcpu.o +CFLAGS_REMOVE_getcpu.o = -pg $(CC_FLAGS_FTRACE) $(CC_FLAGS_EXPOLINE) CFLAGS_REMOVE_vdso64_generic.o = -pg $(CC_FLAGS_FTRACE) $(CC_FLAGS_EXPOLINE) # Build rules diff --git a/arch/s390/kernel/vdso64/getcpu.S b/arch/s390/kernel/vdso64/getcpu.S deleted file mode 100644 index 3c04f7328500..000000000000 --- a/arch/s390/kernel/vdso64/getcpu.S +++ /dev/null @@ -1,31 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * Userland implementation of getcpu() for 64 bits processes in a - * s390 kernel for use in the vDSO - * - * Copyright IBM Corp. 2016 - * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> - */ -#include <asm/vdso.h> -#include <asm/asm-offsets.h> -#include <asm/dwarf.h> - - .text - .align 4 - .globl __kernel_getcpu - .type __kernel_getcpu,@function -__kernel_getcpu: - CFI_STARTPROC - sacf 256 - lm %r4,%r5,__VDSO_GETCPU_VAL(%r0) - sacf 0 - ltgr %r2,%r2 - jz 2f - st %r5,0(%r2) -2: ltgr %r3,%r3 - jz 3f - st %r4,0(%r3) -3: lghi %r2,0 - br %r14 - CFI_ENDPROC - .size __kernel_getcpu,.-__kernel_getcpu diff --git a/arch/s390/kernel/vdso64/getcpu.c b/arch/s390/kernel/vdso64/getcpu.c new file mode 100644 index 000000000000..5b2bc7494d5b --- /dev/null +++ b/arch/s390/kernel/vdso64/getcpu.c @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright IBM Corp. 2020 */ + +#include <linux/compiler.h> +#include <linux/getcpu.h> +#include <asm/timex.h> +#include "vdso.h" + +int __s390_vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused) +{ + __u16 todval[8]; + + /* CPU number is stored in the programmable field of the TOD clock */ + get_tod_clock_ext((char *)todval); + if (cpu) + *cpu = todval[7]; + /* NUMA node is always zero */ + if (node) + *node = 0; + return 0; +} diff --git a/arch/s390/kernel/vdso64/vdso.h b/arch/s390/kernel/vdso64/vdso.h new file mode 100644 index 000000000000..34c7a2312f9d --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __ARCH_S390_KERNEL_VDSO64_VDSO_H +#define __ARCH_S390_KERNEL_VDSO64_VDSO_H + +#include <vdso/datapage.h> + +struct getcpu_cache; + +int __s390_vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused); +int __s390_vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz); +int __s390_vdso_clock_gettime(clockid_t clock, struct __kernel_timespec *ts); +int __s390_vdso_clock_getres(clockid_t clock, struct __kernel_timespec *ts); + +#endif /* __ARCH_S390_KERNEL_VDSO64_VDSO_H */ diff --git a/arch/s390/kernel/vdso64/vdso64.lds.S b/arch/s390/kernel/vdso64/vdso64.lds.S index 7ddb116b5e2e..7bde3909290f 100644 --- a/arch/s390/kernel/vdso64/vdso64.lds.S +++ b/arch/s390/kernel/vdso64/vdso64.lds.S @@ -136,7 +136,6 @@ VERSION __kernel_clock_gettime; __kernel_clock_getres; __kernel_getcpu; - local: *; }; } diff --git a/arch/s390/kernel/vdso64/vdso64_generic.c b/arch/s390/kernel/vdso64/vdso64_generic.c index a8cef7e4d137..a9aa75643c08 100644 --- a/arch/s390/kernel/vdso64/vdso64_generic.c +++ b/arch/s390/kernel/vdso64/vdso64_generic.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include "../../../../lib/vdso/gettimeofday.c" +#include "vdso.h" int __s390_vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz) diff --git a/arch/s390/kernel/vdso64/vdso_user_wrapper.S b/arch/s390/kernel/vdso64/vdso_user_wrapper.S index a775d7e52872..f773505c7e63 100644 --- a/arch/s390/kernel/vdso64/vdso_user_wrapper.S +++ b/arch/s390/kernel/vdso64/vdso_user_wrapper.S @@ -36,3 +36,4 @@ __kernel_\func: vdso_func gettimeofday vdso_func clock_getres vdso_func clock_gettime +vdso_func getcpu diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 177ccfbda40a..4c0e19145cc6 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -10,7 +10,8 @@ * Put .bss..swapper_pg_dir as the first thing in .bss. This will * make sure it has 16k alignment. */ -#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir) +#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir) \ + *(.bss..invalid_pg_dir) /* Handle ro_after_init data on our own. */ #define RO_AFTER_INIT_DATA diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c index 8c0c68e7770e..68d61f2835df 100644 --- a/arch/s390/lib/delay.c +++ b/arch/s390/lib/delay.c @@ -13,11 +13,19 @@ #include <linux/export.h> #include <linux/irqflags.h> #include <linux/interrupt.h> +#include <linux/jump_label.h> #include <linux/irq.h> #include <asm/vtimer.h> #include <asm/div64.h> #include <asm/idle.h> +static DEFINE_STATIC_KEY_FALSE(udelay_ready); + +void __init udelay_enable(void) +{ + static_branch_enable(&udelay_ready); +} + void __delay(unsigned long loops) { /* @@ -76,6 +84,11 @@ void __udelay(unsigned long long usecs) { unsigned long flags; + if (!static_branch_likely(&udelay_ready)) { + udelay_simple(usecs); + return; + } + preempt_disable(); local_irq_save(flags); if (in_irq()) { diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c index 0267405ab7c6..e8f642446fed 100644 --- a/arch/s390/lib/uaccess.c +++ b/arch/s390/lib/uaccess.c @@ -16,6 +16,22 @@ #include <asm/mmu_context.h> #include <asm/facility.h> +#ifdef CONFIG_DEBUG_USER_ASCE +void debug_user_asce(void) +{ + unsigned long cr1, cr7; + + __ctl_store(cr1, 1, 1); + __ctl_store(cr7, 7, 7); + if (cr1 == S390_lowcore.kernel_asce && cr7 == S390_lowcore.user_asce) + return; + panic("incorrect ASCE on kernel exit\n" + "cr1: %016lx cr7: %016lx\n" + "kernel: %016llx user: %016llx\n", + cr1, cr7, S390_lowcore.kernel_asce, S390_lowcore.user_asce); +} +#endif /*CONFIG_DEBUG_USER_ASCE */ + #ifndef CONFIG_HAVE_MARCH_Z10_FEATURES static DEFINE_STATIC_KEY_FALSE(have_mvcos); @@ -40,71 +56,10 @@ static inline int copy_with_mvcos(void) } #endif -void set_fs(mm_segment_t fs) -{ - current->thread.mm_segment = fs; - if (fs == USER_DS) { - __ctl_load(S390_lowcore.user_asce, 1, 1); - clear_cpu_flag(CIF_ASCE_PRIMARY); - } else { - __ctl_load(S390_lowcore.kernel_asce, 1, 1); - set_cpu_flag(CIF_ASCE_PRIMARY); - } - if (fs & 1) { - if (fs == USER_DS_SACF) - __ctl_load(S390_lowcore.user_asce, 7, 7); - else - __ctl_load(S390_lowcore.kernel_asce, 7, 7); - set_cpu_flag(CIF_ASCE_SECONDARY); - } -} -EXPORT_SYMBOL(set_fs); - -mm_segment_t enable_sacf_uaccess(void) -{ - mm_segment_t old_fs; - unsigned long asce, cr; - unsigned long flags; - - old_fs = current->thread.mm_segment; - if (old_fs & 1) - return old_fs; - /* protect against a concurrent page table upgrade */ - local_irq_save(flags); - current->thread.mm_segment |= 1; - asce = S390_lowcore.kernel_asce; - if (likely(old_fs == USER_DS)) { - __ctl_store(cr, 1, 1); - if (cr != S390_lowcore.kernel_asce) { - __ctl_load(S390_lowcore.kernel_asce, 1, 1); - set_cpu_flag(CIF_ASCE_PRIMARY); - } - asce = S390_lowcore.user_asce; - } - __ctl_store(cr, 7, 7); - if (cr != asce) { - __ctl_load(asce, 7, 7); - set_cpu_flag(CIF_ASCE_SECONDARY); - } - local_irq_restore(flags); - return old_fs; -} -EXPORT_SYMBOL(enable_sacf_uaccess); - -void disable_sacf_uaccess(mm_segment_t old_fs) -{ - current->thread.mm_segment = old_fs; - if (old_fs == USER_DS && test_facility(27)) { - __ctl_load(S390_lowcore.user_asce, 1, 1); - clear_cpu_flag(CIF_ASCE_PRIMARY); - } -} -EXPORT_SYMBOL(disable_sacf_uaccess); - static inline unsigned long copy_from_user_mvcos(void *x, const void __user *ptr, unsigned long size) { - register unsigned long reg0 asm("0") = 0x01UL; + register unsigned long reg0 asm("0") = 0x81UL; unsigned long tmp1, tmp2; tmp1 = -4096UL; @@ -135,9 +90,7 @@ static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr, unsigned long size) { unsigned long tmp1, tmp2; - mm_segment_t old_fs; - old_fs = enable_sacf_uaccess(); tmp1 = -256UL; asm volatile( " sacf 0\n" @@ -164,7 +117,6 @@ static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr, EX_TABLE(7b,3b) EX_TABLE(8b,3b) EX_TABLE(9b,6b) : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) : : "cc", "memory"); - disable_sacf_uaccess(old_fs); return size; } @@ -179,7 +131,7 @@ EXPORT_SYMBOL(raw_copy_from_user); static inline unsigned long copy_to_user_mvcos(void __user *ptr, const void *x, unsigned long size) { - register unsigned long reg0 asm("0") = 0x010000UL; + register unsigned long reg0 asm("0") = 0x810000UL; unsigned long tmp1, tmp2; tmp1 = -4096UL; @@ -210,9 +162,7 @@ static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x, unsigned long size) { unsigned long tmp1, tmp2; - mm_segment_t old_fs; - old_fs = enable_sacf_uaccess(); tmp1 = -256UL; asm volatile( " sacf 0\n" @@ -239,7 +189,6 @@ static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x, EX_TABLE(7b,3b) EX_TABLE(8b,3b) EX_TABLE(9b,6b) : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) : : "cc", "memory"); - disable_sacf_uaccess(old_fs); return size; } @@ -254,7 +203,7 @@ EXPORT_SYMBOL(raw_copy_to_user); static inline unsigned long copy_in_user_mvcos(void __user *to, const void __user *from, unsigned long size) { - register unsigned long reg0 asm("0") = 0x010001UL; + register unsigned long reg0 asm("0") = 0x810081UL; unsigned long tmp1, tmp2; tmp1 = -4096UL; @@ -277,10 +226,8 @@ static inline unsigned long copy_in_user_mvcos(void __user *to, const void __use static inline unsigned long copy_in_user_mvc(void __user *to, const void __user *from, unsigned long size) { - mm_segment_t old_fs; unsigned long tmp1; - old_fs = enable_sacf_uaccess(); asm volatile( " sacf 256\n" " aghi %0,-1\n" @@ -304,7 +251,6 @@ static inline unsigned long copy_in_user_mvc(void __user *to, const void __user EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b) : "+a" (size), "+a" (to), "+a" (from), "=a" (tmp1) : : "cc", "memory"); - disable_sacf_uaccess(old_fs); return size; } @@ -318,7 +264,7 @@ EXPORT_SYMBOL(raw_copy_in_user); static inline unsigned long clear_user_mvcos(void __user *to, unsigned long size) { - register unsigned long reg0 asm("0") = 0x010000UL; + register unsigned long reg0 asm("0") = 0x810000UL; unsigned long tmp1, tmp2; tmp1 = -4096UL; @@ -346,10 +292,8 @@ static inline unsigned long clear_user_mvcos(void __user *to, unsigned long size static inline unsigned long clear_user_xc(void __user *to, unsigned long size) { - mm_segment_t old_fs; unsigned long tmp1, tmp2; - old_fs = enable_sacf_uaccess(); asm volatile( " sacf 256\n" " aghi %0,-1\n" @@ -378,7 +322,6 @@ static inline unsigned long clear_user_xc(void __user *to, unsigned long size) EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b) : "+a" (size), "+a" (to), "=a" (tmp1), "=a" (tmp2) : : "cc", "memory"); - disable_sacf_uaccess(old_fs); return size; } @@ -414,15 +357,9 @@ static inline unsigned long strnlen_user_srst(const char __user *src, unsigned long __strnlen_user(const char __user *src, unsigned long size) { - mm_segment_t old_fs; - unsigned long len; - if (unlikely(!size)) return 0; - old_fs = enable_sacf_uaccess(); - len = strnlen_user_srst(src, size); - disable_sacf_uaccess(old_fs); - return len; + return strnlen_user_srst(src, size); } EXPORT_SYMBOL(__strnlen_user); diff --git a/arch/s390/mm/dump_pagetables.c b/arch/s390/mm/dump_pagetables.c index 8f9ff7e7187d..e40a30647d99 100644 --- a/arch/s390/mm/dump_pagetables.c +++ b/arch/s390/mm/dump_pagetables.c @@ -255,7 +255,7 @@ static int pt_dump_init(void) */ max_addr = (S390_lowcore.kernel_asce & _REGION_ENTRY_TYPE_MASK) >> 2; max_addr = 1UL << (max_addr * 11 + 31); - address_markers[IDENTITY_AFTER_END_NR].start_address = memory_end; + address_markers[IDENTITY_AFTER_END_NR].start_address = ident_map_size; address_markers[MODULES_NR].start_address = MODULES_VADDR; address_markers[MODULES_END_NR].start_address = MODULES_END; address_markers[VMEMMAP_NR].start_address = (unsigned long) vmemmap; diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 996884dcc9fd..b8210103de14 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -53,7 +53,6 @@ enum fault_type { KERNEL_FAULT, USER_FAULT, - VDSO_FAULT, GMAP_FAULT, }; @@ -77,22 +76,16 @@ static enum fault_type get_fault_type(struct pt_regs *regs) trans_exc_code = regs->int_parm_long & 3; if (likely(trans_exc_code == 0)) { /* primary space exception */ - if (IS_ENABLED(CONFIG_PGSTE) && - test_pt_regs_flag(regs, PIF_GUEST_FAULT)) - return GMAP_FAULT; - if (current->thread.mm_segment == USER_DS) + if (user_mode(regs)) return USER_FAULT; - return KERNEL_FAULT; - } - if (trans_exc_code == 2) { - /* secondary space exception */ - if (current->thread.mm_segment & 1) { - if (current->thread.mm_segment == USER_DS_SACF) - return USER_FAULT; + if (!IS_ENABLED(CONFIG_PGSTE)) return KERNEL_FAULT; - } - return VDSO_FAULT; + if (test_pt_regs_flag(regs, PIF_GUEST_FAULT)) + return GMAP_FAULT; + return KERNEL_FAULT; } + if (trans_exc_code == 2) + return USER_FAULT; if (trans_exc_code == 1) { /* access register mode, not used in the kernel */ return USER_FAULT; @@ -188,10 +181,6 @@ static void dump_fault_info(struct pt_regs *regs) asce = S390_lowcore.user_asce; pr_cont("user "); break; - case VDSO_FAULT: - asce = S390_lowcore.vdso_asce; - pr_cont("vdso "); - break; case GMAP_FAULT: asce = ((struct gmap *) S390_lowcore.gmap)->asce; pr_cont("gmap "); @@ -414,9 +403,6 @@ static inline vm_fault_t do_exception(struct pt_regs *regs, int access) switch (type) { case KERNEL_FAULT: goto out; - case VDSO_FAULT: - fault = VM_FAULT_BADMAP; - goto out; case USER_FAULT: case GMAP_FAULT: if (faulthandler_disabled() || !mm) @@ -834,7 +820,6 @@ void do_secure_storage_access(struct pt_regs *regs) if (rc) BUG(); break; - case VDSO_FAULT: case GMAP_FAULT: default: do_fault_error(regs, VM_READ | VM_WRITE, VM_FAULT_BADMAP); diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 77767850d0d0..73a163065b95 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -49,6 +49,9 @@ #include <linux/virtio_config.h> pgd_t swapper_pg_dir[PTRS_PER_PGD] __section(".bss..swapper_pg_dir"); +static pgd_t invalid_pg_dir[PTRS_PER_PGD] __section(".bss..invalid_pg_dir"); + +unsigned long s390_invalid_asce; unsigned long empty_zero_page, zero_page_mask; EXPORT_SYMBOL(empty_zero_page); @@ -92,6 +95,9 @@ void __init paging_init(void) unsigned long pgd_type, asce_bits; psw_t psw; + s390_invalid_asce = (unsigned long)invalid_pg_dir; + s390_invalid_asce |= _ASCE_TYPE_REGION3 | _ASCE_TABLE_LENGTH; + crst_table_init((unsigned long *)invalid_pg_dir, _REGION3_ENTRY_EMPTY); init_mm.pgd = swapper_pg_dir; if (VMALLOC_END > _REGION2_SIZE) { asce_bits = _ASCE_TYPE_REGION2 | _ASCE_TABLE_LENGTH; @@ -102,14 +108,14 @@ void __init paging_init(void) } init_mm.context.asce = (__pa(init_mm.pgd) & PAGE_MASK) | asce_bits; S390_lowcore.kernel_asce = init_mm.context.asce; - S390_lowcore.user_asce = S390_lowcore.kernel_asce; + S390_lowcore.user_asce = s390_invalid_asce; crst_table_init((unsigned long *) init_mm.pgd, pgd_type); vmem_map_init(); - kasan_copy_shadow(init_mm.pgd); + kasan_copy_shadow_mapping(); /* enable virtual mapping in kernel mode */ __ctl_load(S390_lowcore.kernel_asce, 1, 1); - __ctl_load(S390_lowcore.kernel_asce, 7, 7); + __ctl_load(S390_lowcore.user_asce, 7, 7); __ctl_load(S390_lowcore.kernel_asce, 13, 13); psw.mask = __extract_psw(); psw_bits(psw).dat = 1; diff --git a/arch/s390/mm/kasan_init.c b/arch/s390/mm/kasan_init.c index 5646b39c728a..db4d303aaaa9 100644 --- a/arch/s390/mm/kasan_init.c +++ b/arch/s390/mm/kasan_init.c @@ -87,7 +87,7 @@ enum populate_mode { POPULATE_ZERO_SHADOW, POPULATE_SHALLOW }; -static void __init kasan_early_vmemmap_populate(unsigned long address, +static void __init kasan_early_pgtable_populate(unsigned long address, unsigned long end, enum populate_mode mode) { @@ -123,8 +123,7 @@ static void __init kasan_early_vmemmap_populate(unsigned long address, pgd_populate(&init_mm, pg_dir, p4_dir); } - if (IS_ENABLED(CONFIG_KASAN_S390_4_LEVEL_PAGING) && - mode == POPULATE_SHALLOW) { + if (mode == POPULATE_SHALLOW) { address = (address + P4D_SIZE) & P4D_MASK; continue; } @@ -143,12 +142,6 @@ static void __init kasan_early_vmemmap_populate(unsigned long address, p4d_populate(&init_mm, p4_dir, pu_dir); } - if (!IS_ENABLED(CONFIG_KASAN_S390_4_LEVEL_PAGING) && - mode == POPULATE_SHALLOW) { - address = (address + PUD_SIZE) & PUD_MASK; - continue; - } - pu_dir = pud_offset(p4_dir, address); if (pud_none(*pu_dir)) { if (mode == POPULATE_ZERO_SHADOW && @@ -281,7 +274,6 @@ void __init kasan_early_init(void) unsigned long shadow_alloc_size; unsigned long vmax_unlimited; unsigned long initrd_end; - unsigned long asce_type; unsigned long memsize; unsigned long pgt_prot = pgprot_val(PAGE_KERNEL_RO); pte_t pte_z; @@ -297,32 +289,26 @@ void __init kasan_early_init(void) memsize = get_mem_detect_end(); if (!memsize) kasan_early_panic("cannot detect physical memory size\n"); - /* respect mem= cmdline parameter */ - if (memory_end_set && memsize > memory_end) - memsize = memory_end; - if (IS_ENABLED(CONFIG_CRASH_DUMP) && OLDMEM_BASE) - memsize = min(memsize, OLDMEM_SIZE); - memsize = min(memsize, KASAN_SHADOW_START); - - if (IS_ENABLED(CONFIG_KASAN_S390_4_LEVEL_PAGING)) { - /* 4 level paging */ - BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_START, P4D_SIZE)); - BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_END, P4D_SIZE)); - crst_table_init((unsigned long *)early_pg_dir, - _REGION2_ENTRY_EMPTY); - untracked_mem_end = kasan_vmax = vmax_unlimited = _REGION1_SIZE; - if (has_uv_sec_stor_limit()) - kasan_vmax = min(vmax_unlimited, uv_info.max_sec_stor_addr); - asce_type = _ASCE_TYPE_REGION2; - } else { - /* 3 level paging */ - BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_START, PUD_SIZE)); - BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_END, PUD_SIZE)); - crst_table_init((unsigned long *)early_pg_dir, - _REGION3_ENTRY_EMPTY); - untracked_mem_end = kasan_vmax = vmax_unlimited = _REGION2_SIZE; - asce_type = _ASCE_TYPE_REGION3; - } + /* + * Kasan currently supports standby memory but only if it follows + * online memory (default allocation), i.e. no memory holes. + * - memsize represents end of online memory + * - ident_map_size represents online + standby and memory limits + * accounted. + * Kasan maps "memsize" right away. + * [0, memsize] - as identity mapping + * [__sha(0), __sha(memsize)] - shadow memory for identity mapping + * The rest [memsize, ident_map_size] if memsize < ident_map_size + * could be mapped/unmapped dynamically later during memory hotplug. + */ + memsize = min(memsize, ident_map_size); + + BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_START, P4D_SIZE)); + BUILD_BUG_ON(!IS_ALIGNED(KASAN_SHADOW_END, P4D_SIZE)); + crst_table_init((unsigned long *)early_pg_dir, _REGION2_ENTRY_EMPTY); + untracked_mem_end = kasan_vmax = vmax_unlimited = _REGION1_SIZE; + if (has_uv_sec_stor_limit()) + kasan_vmax = min(vmax_unlimited, uv_info.max_sec_stor_addr); /* init kasan zero shadow */ crst_table_init((unsigned long *)kasan_early_shadow_p4d, @@ -388,27 +374,25 @@ void __init kasan_early_init(void) * +-----------------+ +- shadow end ---+ */ /* populate kasan shadow (for identity mapping and zero page mapping) */ - kasan_early_vmemmap_populate(__sha(0), __sha(memsize), POPULATE_MAP); + kasan_early_pgtable_populate(__sha(0), __sha(memsize), POPULATE_MAP); if (IS_ENABLED(CONFIG_MODULES)) untracked_mem_end = kasan_vmax - MODULES_LEN; if (IS_ENABLED(CONFIG_KASAN_VMALLOC)) { untracked_mem_end = kasan_vmax - vmalloc_size - MODULES_LEN; /* shallowly populate kasan shadow for vmalloc and modules */ - kasan_early_vmemmap_populate(__sha(untracked_mem_end), - __sha(kasan_vmax), POPULATE_SHALLOW); + kasan_early_pgtable_populate(__sha(untracked_mem_end), __sha(kasan_vmax), + POPULATE_SHALLOW); } /* populate kasan shadow for untracked memory */ - kasan_early_vmemmap_populate(__sha(max_physmem_end), - __sha(untracked_mem_end), + kasan_early_pgtable_populate(__sha(ident_map_size), __sha(untracked_mem_end), POPULATE_ZERO_SHADOW); - kasan_early_vmemmap_populate(__sha(kasan_vmax), - __sha(vmax_unlimited), + kasan_early_pgtable_populate(__sha(kasan_vmax), __sha(vmax_unlimited), POPULATE_ZERO_SHADOW); /* memory allocated for identity mapping structs will be freed later */ pgalloc_freeable = pgalloc_pos; /* populate identity mapping */ - kasan_early_vmemmap_populate(0, memsize, POPULATE_ONE2ONE); - kasan_set_pgd(early_pg_dir, asce_type); + kasan_early_pgtable_populate(0, memsize, POPULATE_ONE2ONE); + kasan_set_pgd(early_pg_dir, _ASCE_TYPE_REGION2); kasan_enable_dat(); /* enable kasan */ init_task.kasan_depth = 0; @@ -416,7 +400,7 @@ void __init kasan_early_init(void) sclp_early_printk("KernelAddressSanitizer initialized\n"); } -void __init kasan_copy_shadow(pgd_t *pg_dir) +void __init kasan_copy_shadow_mapping(void) { /* * At this point we are still running on early pages setup early_pg_dir, @@ -428,24 +412,13 @@ void __init kasan_copy_shadow(pgd_t *pg_dir) pgd_t *pg_dir_dst; p4d_t *p4_dir_src; p4d_t *p4_dir_dst; - pud_t *pu_dir_src; - pud_t *pu_dir_dst; pg_dir_src = pgd_offset_raw(early_pg_dir, KASAN_SHADOW_START); - pg_dir_dst = pgd_offset_raw(pg_dir, KASAN_SHADOW_START); + pg_dir_dst = pgd_offset_raw(init_mm.pgd, KASAN_SHADOW_START); p4_dir_src = p4d_offset(pg_dir_src, KASAN_SHADOW_START); p4_dir_dst = p4d_offset(pg_dir_dst, KASAN_SHADOW_START); - if (!p4d_folded(*p4_dir_src)) { - /* 4 level paging */ - memcpy(p4_dir_dst, p4_dir_src, - (KASAN_SHADOW_SIZE >> P4D_SHIFT) * sizeof(p4d_t)); - return; - } - /* 3 level paging */ - pu_dir_src = pud_offset(p4_dir_src, KASAN_SHADOW_START); - pu_dir_dst = pud_offset(p4_dir_dst, KASAN_SHADOW_START); - memcpy(pu_dir_dst, pu_dir_src, - (KASAN_SHADOW_SIZE >> PUD_SHIFT) * sizeof(pud_t)); + memcpy(p4_dir_dst, p4_dir_src, + (KASAN_SHADOW_SIZE >> P4D_SHIFT) * sizeof(p4d_t)); } void __init kasan_free_early_identity(void) diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c index 11d2c8395e2a..4e87c819ddea 100644 --- a/arch/s390/mm/pgalloc.c +++ b/arch/s390/mm/pgalloc.c @@ -70,19 +70,10 @@ static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; - /* we must change all active ASCEs to avoid the creation of new TLBs */ + /* change all active ASCEs to avoid the creation of new TLBs */ if (current->active_mm == mm) { S390_lowcore.user_asce = mm->context.asce; - if (current->thread.mm_segment == USER_DS) { - __ctl_load(S390_lowcore.user_asce, 1, 1); - /* Mark user-ASCE present in CR1 */ - clear_cpu_flag(CIF_ASCE_PRIMARY); - } - if (current->thread.mm_segment == USER_DS_SACF) { - __ctl_load(S390_lowcore.user_asce, 7, 7); - /* enable_sacf_uaccess does all or nothing */ - WARN_ON(!test_cpu_flag(CIF_ASCE_SECONDARY)); - } + __ctl_load(S390_lowcore.user_asce, 7, 7); } __tlb_flush_local(); } diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index b239f2ba93b0..01f3a5f58e64 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -76,20 +76,20 @@ static void vmem_pte_free(unsigned long *table) /* * The unused vmemmap range, which was not yet memset(PAGE_UNUSED) ranges - * from unused_pmd_start to next PMD_SIZE boundary. + * from unused_sub_pmd_start to next PMD_SIZE boundary. */ -static unsigned long unused_pmd_start; +static unsigned long unused_sub_pmd_start; -static void vmemmap_flush_unused_pmd(void) +static void vmemmap_flush_unused_sub_pmd(void) { - if (!unused_pmd_start) + if (!unused_sub_pmd_start) return; - memset(__va(unused_pmd_start), PAGE_UNUSED, - ALIGN(unused_pmd_start, PMD_SIZE) - unused_pmd_start); - unused_pmd_start = 0; + memset(__va(unused_sub_pmd_start), PAGE_UNUSED, + ALIGN(unused_sub_pmd_start, PMD_SIZE) - unused_sub_pmd_start); + unused_sub_pmd_start = 0; } -static void __vmemmap_use_sub_pmd(unsigned long start, unsigned long end) +static void vmemmap_mark_sub_pmd_used(unsigned long start, unsigned long end) { /* * As we expect to add in the same granularity as we remove, it's @@ -106,24 +106,24 @@ static void vmemmap_use_sub_pmd(unsigned long start, unsigned long end) * We only optimize if the new used range directly follows the * previously unused range (esp., when populating consecutive sections). */ - if (unused_pmd_start == start) { - unused_pmd_start = end; - if (likely(IS_ALIGNED(unused_pmd_start, PMD_SIZE))) - unused_pmd_start = 0; + if (unused_sub_pmd_start == start) { + unused_sub_pmd_start = end; + if (likely(IS_ALIGNED(unused_sub_pmd_start, PMD_SIZE))) + unused_sub_pmd_start = 0; return; } - vmemmap_flush_unused_pmd(); - __vmemmap_use_sub_pmd(start, end); + vmemmap_flush_unused_sub_pmd(); + vmemmap_mark_sub_pmd_used(start, end); } static void vmemmap_use_new_sub_pmd(unsigned long start, unsigned long end) { void *page = __va(ALIGN_DOWN(start, PMD_SIZE)); - vmemmap_flush_unused_pmd(); + vmemmap_flush_unused_sub_pmd(); /* Could be our memmap page is filled with PAGE_UNUSED already ... */ - __vmemmap_use_sub_pmd(start, end); + vmemmap_mark_sub_pmd_used(start, end); /* Mark the unused parts of the new memmap page PAGE_UNUSED. */ if (!IS_ALIGNED(start, PMD_SIZE)) @@ -134,7 +134,7 @@ static void vmemmap_use_new_sub_pmd(unsigned long start, unsigned long end) * unused range in the populated PMD. */ if (!IS_ALIGNED(end, PMD_SIZE)) - unused_pmd_start = end; + unused_sub_pmd_start = end; } /* Returns true if the PMD is completely unused and can be freed. */ @@ -142,7 +142,7 @@ static bool vmemmap_unuse_sub_pmd(unsigned long start, unsigned long end) { void *page = __va(ALIGN_DOWN(start, PMD_SIZE)); - vmemmap_flush_unused_pmd(); + vmemmap_flush_unused_sub_pmd(); memset(__va(start), PAGE_UNUSED, end - start); return !memchr_inv(page, PAGE_UNUSED, PMD_SIZE); } @@ -223,7 +223,7 @@ static int __ref modify_pmd_table(pud_t *pud, unsigned long addr, if (!add) { if (pmd_none(*pmd)) continue; - if (pmd_large(*pmd) && !add) { + if (pmd_large(*pmd)) { if (IS_ALIGNED(addr, PMD_SIZE) && IS_ALIGNED(next, PMD_SIZE)) { if (!direct) diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index 570016ae8bcd..41df8fcfddde 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -851,8 +851,10 @@ static int __init pci_base_init(void) if (!s390_pci_probe) return 0; - if (!test_facility(69) || !test_facility(71)) + if (!test_facility(69) || !test_facility(71)) { + pr_info("PCI is not supported because CPU facilities 69 or 71 are not available\n"); return 0; + } if (test_facility(153) && !s390_pci_no_mio) { static_branch_enable(&have_mio); diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c index 401cf670a243..de3bdbed8881 100644 --- a/arch/s390/pci/pci_mmio.c +++ b/arch/s390/pci/pci_mmio.c @@ -93,12 +93,10 @@ static inline int __memcpy_toio_inuser(void __iomem *dst, { int size, rc = 0; u8 status = 0; - mm_segment_t old_fs; if (!src) return -EINVAL; - old_fs = enable_sacf_uaccess(); while (n > 0) { size = zpci_get_max_write_size((u64 __force) dst, (u64 __force) src, n, @@ -113,39 +111,20 @@ static inline int __memcpy_toio_inuser(void __iomem *dst, dst += size; n -= size; } - disable_sacf_uaccess(old_fs); if (rc) zpci_err_mmio(rc, status, (__force u64) dst); return rc; } -static long get_pfn(unsigned long user_addr, unsigned long access, - unsigned long *pfn) -{ - struct vm_area_struct *vma; - long ret; - - mmap_read_lock(current->mm); - ret = -EINVAL; - vma = find_vma(current->mm, user_addr); - if (!vma) - goto out; - ret = -EACCES; - if (!(vma->vm_flags & access)) - goto out; - ret = follow_pfn(vma, user_addr, pfn); -out: - mmap_read_unlock(current->mm); - return ret; -} - SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, const void __user *, user_buffer, size_t, length) { u8 local_buf[64]; void __iomem *io_addr; void *buf; - unsigned long pfn; + struct vm_area_struct *vma; + pte_t *ptep; + spinlock_t *ptl; long ret; if (!zpci_is_enabled()) @@ -158,7 +137,7 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, * We only support write access to MIO capable devices if we are on * a MIO enabled system. Otherwise we would have to check for every * address if it is a special ZPCI_ADDR and would have to do - * a get_pfn() which we don't need for MIO capable devices. Currently + * a pfn lookup which we don't need for MIO capable devices. Currently * ISM devices are the only devices without MIO support and there is no * known need for accessing these from userspace. */ @@ -176,21 +155,37 @@ SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, } else buf = local_buf; - ret = get_pfn(mmio_addr, VM_WRITE, &pfn); + ret = -EFAULT; + if (copy_from_user(buf, user_buffer, length)) + goto out_free; + + mmap_read_lock(current->mm); + ret = -EINVAL; + vma = find_vma(current->mm, mmio_addr); + if (!vma) + goto out_unlock_mmap; + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + goto out_unlock_mmap; + ret = -EACCES; + if (!(vma->vm_flags & VM_WRITE)) + goto out_unlock_mmap; + + ret = follow_pte_pmd(vma->vm_mm, mmio_addr, NULL, &ptep, NULL, &ptl); if (ret) - goto out; - io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | + goto out_unlock_mmap; + + io_addr = (void __iomem *)((pte_pfn(*ptep) << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); - ret = -EFAULT; if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) - goto out; - - if (copy_from_user(buf, user_buffer, length)) - goto out; + goto out_unlock_pt; ret = zpci_memcpy_toio(io_addr, buf, length); -out: +out_unlock_pt: + pte_unmap_unlock(ptep, ptl); +out_unlock_mmap: + mmap_read_unlock(current->mm); +out_free: if (buf != local_buf) kfree(buf); return ret; @@ -248,9 +243,7 @@ static inline int __memcpy_fromio_inuser(void __user *dst, { int size, rc = 0; u8 status; - mm_segment_t old_fs; - old_fs = enable_sacf_uaccess(); while (n > 0) { size = zpci_get_max_write_size((u64 __force) src, (u64 __force) dst, n, @@ -262,7 +255,6 @@ static inline int __memcpy_fromio_inuser(void __user *dst, dst += size; n -= size; } - disable_sacf_uaccess(old_fs); if (rc) zpci_err_mmio(rc, status, (__force u64) dst); return rc; @@ -274,7 +266,9 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, u8 local_buf[64]; void __iomem *io_addr; void *buf; - unsigned long pfn; + struct vm_area_struct *vma; + pte_t *ptep; + spinlock_t *ptl; long ret; if (!zpci_is_enabled()) @@ -287,7 +281,7 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, * We only support read access to MIO capable devices if we are on * a MIO enabled system. Otherwise we would have to check for every * address if it is a special ZPCI_ADDR and would have to do - * a get_pfn() which we don't need for MIO capable devices. Currently + * a pfn lookup which we don't need for MIO capable devices. Currently * ISM devices are the only devices without MIO support and there is no * known need for accessing these from userspace. */ @@ -306,22 +300,38 @@ SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, buf = local_buf; } - ret = get_pfn(mmio_addr, VM_READ, &pfn); + mmap_read_lock(current->mm); + ret = -EINVAL; + vma = find_vma(current->mm, mmio_addr); + if (!vma) + goto out_unlock_mmap; + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + goto out_unlock_mmap; + ret = -EACCES; + if (!(vma->vm_flags & VM_WRITE)) + goto out_unlock_mmap; + + ret = follow_pte_pmd(vma->vm_mm, mmio_addr, NULL, &ptep, NULL, &ptl); if (ret) - goto out; - io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); + goto out_unlock_mmap; + + io_addr = (void __iomem *)((pte_pfn(*ptep) << PAGE_SHIFT) | + (mmio_addr & ~PAGE_MASK)); if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) { ret = -EFAULT; - goto out; + goto out_unlock_pt; } ret = zpci_memcpy_fromio(buf, io_addr, length); - if (ret) - goto out; - if (copy_to_user(user_buffer, buf, length)) + +out_unlock_pt: + pte_unmap_unlock(ptep, ptl); +out_unlock_mmap: + mmap_read_unlock(current->mm); + + if (!ret && copy_to_user(user_buffer, buf, length)) ret = -EFAULT; -out: if (buf != local_buf) kfree(buf); return ret; diff --git a/arch/s390/purgatory/head.S b/arch/s390/purgatory/head.S index 5a10ce34b95d..3d1c31e0cf3d 100644 --- a/arch/s390/purgatory/head.S +++ b/arch/s390/purgatory/head.S @@ -62,14 +62,15 @@ jh 10b .endm -.macro START_NEXT_KERNEL base +.macro START_NEXT_KERNEL base subcode lg %r4,kernel_entry-\base(%r13) lg %r5,load_psw_mask-\base(%r13) ogr %r4,%r5 stg %r4,0(%r0) xgr %r0,%r0 - diag %r0,%r0,0x308 + lghi %r1,\subcode + diag %r0,%r1,0x308 .endm .text @@ -123,7 +124,7 @@ ENTRY(purgatory_start) je .start_crash_kernel /* start normal kernel */ - START_NEXT_KERNEL .base_crash + START_NEXT_KERNEL .base_crash 0 .return_old_kernel: lmg %r6,%r15,gprregs-.base_crash(%r13) @@ -227,7 +228,7 @@ ENTRY(purgatory_start) MEMCPY %r9,%r10,%r11 /* start crash kernel */ - START_NEXT_KERNEL .base_dst + START_NEXT_KERNEL .base_dst 1 load_psw_mask: |