diff options
Diffstat (limited to 'arch/s390')
93 files changed, 2195 insertions, 808 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 9a9c7a6fe925..039a3417dfc4 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -56,6 +56,12 @@ config PCI_QUIRKS config ARCH_SUPPORTS_UPROBES def_bool y +config KASAN_SHADOW_OFFSET + hex + depends on KASAN + default 0x18000000000000 if KASAN_S390_4_LEVEL_PAGING + default 0x30000000000 + config S390 def_bool y select ARCH_BINFMT_ELF_STATE @@ -120,11 +126,14 @@ config S390 select HAVE_ALIGNED_STRUCT_PAGE if SLUB select HAVE_ARCH_AUDITSYSCALL select HAVE_ARCH_JUMP_LABEL + select HAVE_ARCH_JUMP_LABEL_RELATIVE + select HAVE_ARCH_KASAN select CPU_NO_EFFICIENT_FFS if !HAVE_MARCH_Z9_109_FEATURES select HAVE_ARCH_SECCOMP_FILTER select HAVE_ARCH_SOFT_DIRTY select HAVE_ARCH_TRACEHOOK select HAVE_ARCH_TRANSPARENT_HUGEPAGE + select HAVE_ARCH_VMAP_STACK select HAVE_EBPF_JIT if PACK_STACK && HAVE_MARCH_Z196_FEATURES select HAVE_CMPXCHG_DOUBLE select HAVE_CMPXCHG_LOCAL @@ -649,6 +658,7 @@ config PACK_STACK config CHECK_STACK def_bool y + depends on !VMAP_STACK prompt "Detect kernel stack overflow" help This option enables the compiler option -mstack-guard and diff --git a/arch/s390/Makefile b/arch/s390/Makefile index ee65185bbc80..0b33577932c3 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -27,7 +27,7 @@ KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-option,-ffreestanding) KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),-g) KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO_DWARF4), $(call cc-option, -gdwarf-4,)) UTS_MACHINE := s390x -STACK_SIZE := 16384 +STACK_SIZE := $(if $(CONFIG_KASAN),32768,16384) CHECKFLAGS += -D__s390__ -D__s390x__ export LD_BFD diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 9bf8489df6e6..e4b58240ec53 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -137,6 +137,14 @@ static void appldata_work_fn(struct work_struct *work) mutex_unlock(&appldata_ops_mutex); } +static struct appldata_product_id appldata_id = { + .prod_nr = {0xD3, 0xC9, 0xD5, 0xE4, + 0xE7, 0xD2, 0xD9}, /* "LINUXKR" */ + .prod_fn = 0xD5D3, /* "NL" */ + .version_nr = 0xF2F6, /* "26" */ + .release_nr = 0xF0F1, /* "01" */ +}; + /* * appldata_diag() * @@ -145,17 +153,22 @@ static void appldata_work_fn(struct work_struct *work) int appldata_diag(char record_nr, u16 function, unsigned long buffer, u16 length, char *mod_lvl) { - struct appldata_product_id id = { - .prod_nr = {0xD3, 0xC9, 0xD5, 0xE4, - 0xE7, 0xD2, 0xD9}, /* "LINUXKR" */ - .prod_fn = 0xD5D3, /* "NL" */ - .version_nr = 0xF2F6, /* "26" */ - .release_nr = 0xF0F1, /* "01" */ - }; + struct appldata_parameter_list *parm_list; + struct appldata_product_id *id; + int rc; - id.record_nr = record_nr; - id.mod_lvl = (mod_lvl[0]) << 8 | mod_lvl[1]; - return appldata_asm(&id, function, (void *) buffer, length); + parm_list = kmalloc(sizeof(*parm_list), GFP_KERNEL); + id = kmemdup(&appldata_id, sizeof(appldata_id), GFP_KERNEL); + rc = -ENOMEM; + if (parm_list && id) { + id->record_nr = record_nr; + id->mod_lvl = (mod_lvl[0]) << 8 | mod_lvl[1]; + rc = appldata_asm(parm_list, id, function, + (void *) buffer, length); + } + kfree(id); + kfree(parm_list); + return rc; } /************************ timer, work, DIAG <END> ****************************/ diff --git a/arch/s390/boot/.gitignore b/arch/s390/boot/.gitignore index 017d5912ad2d..16ff906e4610 100644 --- a/arch/s390/boot/.gitignore +++ b/arch/s390/boot/.gitignore @@ -1,2 +1,3 @@ image bzImage +section_cmp.* diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile index 9e6668ee93de..d5ad724f5c96 100644 --- a/arch/s390/boot/Makefile +++ b/arch/s390/boot/Makefile @@ -6,6 +6,7 @@ KCOV_INSTRUMENT := n GCOV_PROFILE := n UBSAN_SANITIZE := n +KASAN_SANITIZE := n KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR) KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR) @@ -27,15 +28,32 @@ endif CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char -obj-y := head.o als.o ebcdic.o sclp_early_core.o mem.o -targets := bzImage startup.a $(obj-y) +obj-y := head.o als.o startup.o mem_detect.o ipl_parm.o string.o ebcdic.o +obj-y += sclp_early_core.o mem.o ipl_vmparm.o cmdline.o ctype.o +targets := bzImage startup.a section_cmp.boot.data $(obj-y) subdir- := compressed OBJECTS := $(addprefix $(obj)/,$(obj-y)) -$(obj)/bzImage: $(obj)/compressed/vmlinux FORCE +quiet_cmd_section_cmp = SECTCMP $* +define cmd_section_cmp + s1=`$(OBJDUMP) -t -j "$*" "$<" | sort | \ + sed -n "/0000000000000000/! s/.*\s$*\s\+//p" | sha256sum`; \ + s2=`$(OBJDUMP) -t -j "$*" "$(word 2,$^)" | sort | \ + sed -n "/0000000000000000/! s/.*\s$*\s\+//p" | sha256sum`; \ + if [ "$$s1" != "$$s2" ]; then \ + echo "error: section $* differs between $< and $(word 2,$^)" >&2; \ + exit 1; \ + fi; \ + touch $@ +endef + +$(obj)/bzImage: $(obj)/compressed/vmlinux $(obj)/section_cmp.boot.data FORCE $(call if_changed,objcopy) +$(obj)/section_cmp%: vmlinux $(obj)/compressed/vmlinux FORCE + $(call if_changed,section_cmp) + $(obj)/compressed/vmlinux: $(obj)/startup.a FORCE $(Q)$(MAKE) $(build)=$(obj)/compressed $@ diff --git a/arch/s390/boot/boot.h b/arch/s390/boot/boot.h new file mode 100644 index 000000000000..fc41e2277ea8 --- /dev/null +++ b/arch/s390/boot/boot.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef BOOT_BOOT_H +#define BOOT_BOOT_H + +void startup_kernel(void); +void detect_memory(void); +void store_ipl_parmblock(void); +void setup_boot_command_line(void); +void setup_memory_end(void); + +#endif /* BOOT_BOOT_H */ diff --git a/arch/s390/boot/cmdline.c b/arch/s390/boot/cmdline.c new file mode 100644 index 000000000000..73d826cdbdeb --- /dev/null +++ b/arch/s390/boot/cmdline.c @@ -0,0 +1,2 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "../../../lib/cmdline.c" diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile index 04609478d18b..593039620487 100644 --- a/arch/s390/boot/compressed/Makefile +++ b/arch/s390/boot/compressed/Makefile @@ -8,14 +8,16 @@ KCOV_INSTRUMENT := n GCOV_PROFILE := n UBSAN_SANITIZE := n +KASAN_SANITIZE := n -obj-y := $(if $(CONFIG_KERNEL_UNCOMPRESSED),,head.o misc.o) piggy.o +obj-y := $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) piggy.o info.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 += vmlinux.scr.lds $(obj-y) $(if $(CONFIG_KERNEL_UNCOMPRESSED),,sizes.h) +targets += info.bin $(obj-y) KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR) KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR) +OBJCOPYFLAGS := OBJECTS := $(addprefix $(obj)/,$(obj-y)) @@ -23,23 +25,16 @@ LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup -T $(obj)/vmlinux: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS) $(call if_changed,ld) -# extract required uncompressed vmlinux symbols and adjust them to reflect offsets inside vmlinux.bin -sed-sizes := -e 's/^\([0-9a-fA-F]*\) . \(__bss_start\|_end\)$$/\#define SZ\2 (0x\1 - 0x100000)/p' - -quiet_cmd_sizes = GEN $@ - cmd_sizes = $(NM) $< | sed -n $(sed-sizes) > $@ - -$(obj)/sizes.h: vmlinux - $(call if_changed,sizes) - -AFLAGS_head.o += -I$(objtree)/$(obj) -$(obj)/head.o: $(obj)/sizes.h +OBJCOPYFLAGS_info.bin := -O binary --only-section=.vmlinux.info +$(obj)/info.bin: vmlinux FORCE + $(call if_changed,objcopy) -CFLAGS_misc.o += -I$(objtree)/$(obj) -$(obj)/misc.o: $(obj)/sizes.h +OBJCOPYFLAGS_info.o := -I binary -O elf64-s390 -B s390:64-bit --rename-section .data=.vmlinux.info +$(obj)/info.o: $(obj)/info.bin FORCE + $(call if_changed,objcopy) -OBJCOPYFLAGS_vmlinux.bin := -R .comment -S -$(obj)/vmlinux.bin: vmlinux +OBJCOPYFLAGS_vmlinux.bin := -O binary --remove-section=.comment --remove-section=.vmlinux.info -S +$(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) vmlinux.bin.all-y := $(obj)/vmlinux.bin @@ -64,10 +59,10 @@ $(obj)/vmlinux.bin.lzo: $(vmlinux.bin.all-y) $(obj)/vmlinux.bin.xz: $(vmlinux.bin.all-y) $(call if_changed,xzkern) -LDFLAGS_piggy.o := -r --format binary --oformat $(LD_BFD) -T -$(obj)/piggy.o: $(obj)/vmlinux.scr.lds $(obj)/vmlinux.bin$(suffix-y) - $(call if_changed,ld) +OBJCOPYFLAGS_piggy.o := -I binary -O elf64-s390 -B s390:64-bit --rename-section .data=.vmlinux.bin.compressed +$(obj)/piggy.o: $(obj)/vmlinux.bin$(suffix-y) FORCE + $(call if_changed,objcopy) -chkbss := $(filter-out $(obj)/misc.o $(obj)/piggy.o,$(OBJECTS)) +chkbss := $(filter-out $(obj)/piggy.o $(obj)/info.o,$(OBJECTS)) chkbss-target := $(obj)/vmlinux.bin include $(srctree)/arch/s390/scripts/Makefile.chkbss diff --git a/arch/s390/boot/compressed/decompressor.c b/arch/s390/boot/compressed/decompressor.c new file mode 100644 index 000000000000..45046630c56a --- /dev/null +++ b/arch/s390/boot/compressed/decompressor.c @@ -0,0 +1,85 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Definitions and wrapper functions for kernel decompressor + * + * Copyright IBM Corp. 2010 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/string.h> +#include <asm/page.h> +#include "decompressor.h" + +/* + * gzip declarations + */ +#define STATIC static +#define STATIC_RW_DATA static __section(.data) + +#undef memset +#undef memcpy +#undef memmove +#define memmove memmove +#define memzero(s, n) memset((s), 0, (n)) + +/* Symbols defined by linker scripts */ +extern char _end[]; +extern unsigned char _compressed_start[]; +extern unsigned char _compressed_end[]; + +#ifdef CONFIG_HAVE_KERNEL_BZIP2 +#define HEAP_SIZE 0x400000 +#else +#define HEAP_SIZE 0x10000 +#endif + +static unsigned long free_mem_ptr = (unsigned long) _end; +static unsigned long free_mem_end_ptr = (unsigned long) _end + HEAP_SIZE; + +#ifdef CONFIG_KERNEL_GZIP +#include "../../../../lib/decompress_inflate.c" +#endif + +#ifdef CONFIG_KERNEL_BZIP2 +#include "../../../../lib/decompress_bunzip2.c" +#endif + +#ifdef CONFIG_KERNEL_LZ4 +#include "../../../../lib/decompress_unlz4.c" +#endif + +#ifdef CONFIG_KERNEL_LZMA +#include "../../../../lib/decompress_unlzma.c" +#endif + +#ifdef CONFIG_KERNEL_LZO +#include "../../../../lib/decompress_unlzo.c" +#endif + +#ifdef CONFIG_KERNEL_XZ +#include "../../../../lib/decompress_unxz.c" +#endif + +#define decompress_offset ALIGN((unsigned long)_end + HEAP_SIZE, PAGE_SIZE) + +unsigned long mem_safe_offset(void) +{ + /* + * due to 4MB HEAD_SIZE for bzip2 + * 'decompress_offset + vmlinux.image_size' could be larger than + * kernel at final position + its .bss, so take the larger of two + */ + return max(decompress_offset + vmlinux.image_size, + vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size); +} + +void *decompress_kernel(void) +{ + void *output = (void *)decompress_offset; + + __decompress(_compressed_start, _compressed_end - _compressed_start, + NULL, NULL, output, 0, NULL, error); + return output; +} diff --git a/arch/s390/boot/compressed/decompressor.h b/arch/s390/boot/compressed/decompressor.h new file mode 100644 index 000000000000..e1c1f2ec60f4 --- /dev/null +++ b/arch/s390/boot/compressed/decompressor.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef BOOT_COMPRESSED_DECOMPRESSOR_H +#define BOOT_COMPRESSED_DECOMPRESSOR_H + +#ifdef CONFIG_KERNEL_UNCOMPRESSED +static inline void *decompress_kernel(void) {} +#else +void *decompress_kernel(void); +#endif +unsigned long mem_safe_offset(void); +void error(char *m); + +struct vmlinux_info { + unsigned long default_lma; + void (*entry)(void); + unsigned long image_size; /* does not include .bss */ + unsigned long bss_size; /* uncompressed image .bss size */ + unsigned long bootdata_off; + unsigned long bootdata_size; +}; + +extern char _vmlinux_info[]; +#define vmlinux (*(struct vmlinux_info *)_vmlinux_info) + +#endif /* BOOT_COMPRESSED_DECOMPRESSOR_H */ diff --git a/arch/s390/boot/compressed/head.S b/arch/s390/boot/compressed/head.S deleted file mode 100644 index df8dbbc17bcc..000000000000 --- a/arch/s390/boot/compressed/head.S +++ /dev/null @@ -1,52 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * Startup glue code to uncompress the kernel - * - * Copyright IBM Corp. 2010 - * - * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> - */ - -#include <linux/init.h> -#include <linux/linkage.h> -#include <asm/asm-offsets.h> -#include <asm/thread_info.h> -#include <asm/page.h> -#include "sizes.h" - -__HEAD -ENTRY(startup_decompressor) - basr %r13,0 # get base -.LPG1: - # setup stack - lg %r15,.Lstack-.LPG1(%r13) - aghi %r15,-160 - brasl %r14,decompress_kernel - # Set up registers for memory mover. We move the decompressed image to - # 0x100000, where startup_continue of the decompressed image is supposed - # to be. - lgr %r4,%r2 - lg %r2,.Loffset-.LPG1(%r13) - lg %r3,.Lmvsize-.LPG1(%r13) - lgr %r5,%r3 - # Move the memory mover someplace safe so it doesn't overwrite itself. - la %r1,0x200 - mvc 0(mover_end-mover,%r1),mover-.LPG1(%r13) - # When the memory mover is done we pass control to - # arch/s390/kernel/head64.S:startup_continue which lives at 0x100000 in - # the decompressed image. - lgr %r6,%r2 - br %r1 -mover: - mvcle %r2,%r4,0 - jo mover - br %r6 -mover_end: - - .align 8 -.Lstack: - .quad 0x8000 + (1<<(PAGE_SHIFT+THREAD_SIZE_ORDER)) -.Loffset: - .quad 0x100000 -.Lmvsize: - .quad SZ__bss_start diff --git a/arch/s390/boot/compressed/misc.c b/arch/s390/boot/compressed/misc.c deleted file mode 100644 index f66ad73c205b..000000000000 --- a/arch/s390/boot/compressed/misc.c +++ /dev/null @@ -1,116 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * Definitions and wrapper functions for kernel decompressor - * - * Copyright IBM Corp. 2010 - * - * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> - */ - -#include <linux/uaccess.h> -#include <asm/page.h> -#include <asm/sclp.h> -#include <asm/ipl.h> -#include "sizes.h" - -/* - * gzip declarations - */ -#define STATIC static - -#undef memset -#undef memcpy -#undef memmove -#define memmove memmove -#define memzero(s, n) memset((s), 0, (n)) - -/* Symbols defined by linker scripts */ -extern char input_data[]; -extern int input_len; -extern char _end[]; -extern char _bss[], _ebss[]; - -static void error(char *m); - -static unsigned long free_mem_ptr; -static unsigned long free_mem_end_ptr; - -#ifdef CONFIG_HAVE_KERNEL_BZIP2 -#define HEAP_SIZE 0x400000 -#else -#define HEAP_SIZE 0x10000 -#endif - -#ifdef CONFIG_KERNEL_GZIP -#include "../../../../lib/decompress_inflate.c" -#endif - -#ifdef CONFIG_KERNEL_BZIP2 -#include "../../../../lib/decompress_bunzip2.c" -#endif - -#ifdef CONFIG_KERNEL_LZ4 -#include "../../../../lib/decompress_unlz4.c" -#endif - -#ifdef CONFIG_KERNEL_LZMA -#include "../../../../lib/decompress_unlzma.c" -#endif - -#ifdef CONFIG_KERNEL_LZO -#include "../../../../lib/decompress_unlzo.c" -#endif - -#ifdef CONFIG_KERNEL_XZ -#include "../../../../lib/decompress_unxz.c" -#endif - -static int puts(const char *s) -{ - sclp_early_printk(s); - return 0; -} - -static void error(char *x) -{ - unsigned long long psw = 0x000a0000deadbeefULL; - - puts("\n\n"); - puts(x); - puts("\n\n -- System halted"); - - asm volatile("lpsw %0" : : "Q" (psw)); -} - -unsigned long decompress_kernel(void) -{ - void *output, *kernel_end; - - output = (void *) ALIGN((unsigned long) _end + HEAP_SIZE, PAGE_SIZE); - kernel_end = output + SZ__bss_start; - -#ifdef CONFIG_BLK_DEV_INITRD - /* - * Move the initrd right behind the end of the decompressed - * kernel image. This also prevents initrd corruption caused by - * bss clearing since kernel_end will always be located behind the - * current bss section.. - */ - if (INITRD_START && INITRD_SIZE && kernel_end > (void *) INITRD_START) { - memmove(kernel_end, (void *) INITRD_START, INITRD_SIZE); - INITRD_START = (unsigned long) kernel_end; - } -#endif - - /* - * Clear bss section. free_mem_ptr and free_mem_end_ptr need to be - * initialized afterwards since they reside in bss. - */ - memset(_bss, 0, _ebss - _bss); - free_mem_ptr = (unsigned long) _end; - free_mem_end_ptr = free_mem_ptr + HEAP_SIZE; - - __decompress(input_data, input_len, NULL, NULL, output, 0, NULL, error); - return (unsigned long) output; -} - diff --git a/arch/s390/boot/compressed/vmlinux.lds.S b/arch/s390/boot/compressed/vmlinux.lds.S index b16ac8b3c439..7efc3938f595 100644 --- a/arch/s390/boot/compressed/vmlinux.lds.S +++ b/arch/s390/boot/compressed/vmlinux.lds.S @@ -1,5 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ #include <asm-generic/vmlinux.lds.h> +#include <asm/vmlinux.lds.h> OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") OUTPUT_ARCH(s390:64-bit) @@ -8,9 +9,6 @@ ENTRY(startup) SECTIONS { - /* Be careful parts of head_64.S assume startup_32 is at - * address 0. - */ . = 0; .head.text : { _head = . ; @@ -26,7 +24,7 @@ SECTIONS .rodata : { _rodata = . ; *(.rodata) /* read-only data */ - *(EXCLUDE_FILE (*piggy.o) .rodata.compressed) + *(.rodata.*) _erodata = . ; } .data : { @@ -35,14 +33,28 @@ SECTIONS *(.data.*) _edata = . ; } - startup_continue = 0x100000; + BOOT_DATA + + /* + * uncompressed image info used by the decompressor it should match + * struct vmlinux_info. It comes from .vmlinux.info section of + * uncompressed vmlinux in a form of info.o + */ + . = ALIGN(8); + .vmlinux.info : { + _vmlinux_info = .; + *(.vmlinux.info) + } + #ifdef CONFIG_KERNEL_UNCOMPRESSED . = 0x100000; #else . = ALIGN(8); #endif .rodata.compressed : { - *(.rodata.compressed) + _compressed_start = .; + *(.vmlinux.bin.compressed) + _compressed_end = .; } . = ALIGN(256); .bss : { diff --git a/arch/s390/boot/compressed/vmlinux.scr.lds.S b/arch/s390/boot/compressed/vmlinux.scr.lds.S deleted file mode 100644 index ff01d18c9222..000000000000 --- a/arch/s390/boot/compressed/vmlinux.scr.lds.S +++ /dev/null @@ -1,15 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -SECTIONS -{ - .rodata.compressed : { -#ifndef CONFIG_KERNEL_UNCOMPRESSED - input_len = .; - LONG(input_data_end - input_data) input_data = .; -#endif - *(.data) -#ifndef CONFIG_KERNEL_UNCOMPRESSED - output_len = . - 4; - input_data_end = .; -#endif - } -} diff --git a/arch/s390/boot/ctype.c b/arch/s390/boot/ctype.c new file mode 100644 index 000000000000..2495810b47e3 --- /dev/null +++ b/arch/s390/boot/ctype.c @@ -0,0 +1,2 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "../../../lib/ctype.c" diff --git a/arch/s390/boot/head.S b/arch/s390/boot/head.S index f721913b73f1..ce2cbbc41742 100644 --- a/arch/s390/boot/head.S +++ b/arch/s390/boot/head.S @@ -60,6 +60,9 @@ __HEAD .long 0x02000690,0x60000050 .long 0x020006e0,0x20000050 + .org 0x1a0 + .quad 0,iplstart + .org 0x200 # @@ -308,16 +311,11 @@ ENTRY(startup_kdump) spt 6f-.LPG0(%r13) mvc __LC_LAST_UPDATE_TIMER(8),6f-.LPG0(%r13) l %r15,.Lstack-.LPG0(%r13) - ahi %r15,-STACK_FRAME_OVERHEAD brasl %r14,verify_facilities -#ifdef CONFIG_KERNEL_UNCOMPRESSED - jg startup_continue -#else - jg startup_decompressor -#endif + brasl %r14,startup_kernel .Lstack: - .long 0x8000 + (1<<(PAGE_SHIFT+THREAD_SIZE_ORDER)) + .long 0x8000 + (1<<(PAGE_SHIFT+BOOT_STACK_ORDER)) - STACK_FRAME_OVERHEAD .align 8 6: .long 0x7fffffff,0xffffffff diff --git a/arch/s390/boot/ipl_parm.c b/arch/s390/boot/ipl_parm.c new file mode 100644 index 000000000000..9dab596be98e --- /dev/null +++ b/arch/s390/boot/ipl_parm.c @@ -0,0 +1,182 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/init.h> +#include <linux/ctype.h> +#include <asm/ebcdic.h> +#include <asm/sclp.h> +#include <asm/sections.h> +#include <asm/boot_data.h> +#include "boot.h" + +char __bootdata(early_command_line)[COMMAND_LINE_SIZE]; +struct ipl_parameter_block __bootdata(early_ipl_block); +int __bootdata(early_ipl_block_valid); + +unsigned long __bootdata(memory_end); +int __bootdata(memory_end_set); +int __bootdata(noexec_disabled); + +static inline int __diag308(unsigned long subcode, void *addr) +{ + register unsigned long _addr asm("0") = (unsigned long)addr; + register unsigned long _rc asm("1") = 0; + unsigned long reg1, reg2; + psw_t old = S390_lowcore.program_new_psw; + + asm volatile( + " epsw %0,%1\n" + " st %0,%[psw_pgm]\n" + " st %1,%[psw_pgm]+4\n" + " larl %0,1f\n" + " stg %0,%[psw_pgm]+8\n" + " diag %[addr],%[subcode],0x308\n" + "1: nopr %%r7\n" + : "=&d" (reg1), "=&a" (reg2), + [psw_pgm] "=Q" (S390_lowcore.program_new_psw), + [addr] "+d" (_addr), "+d" (_rc) + : [subcode] "d" (subcode) + : "cc", "memory"); + S390_lowcore.program_new_psw = old; + return _rc; +} + +void store_ipl_parmblock(void) +{ + int rc; + + rc = __diag308(DIAG308_STORE, &early_ipl_block); + if (rc == DIAG308_RC_OK && + early_ipl_block.hdr.version <= IPL_MAX_SUPPORTED_VERSION) + early_ipl_block_valid = 1; +} + +static size_t scpdata_length(const char *buf, size_t count) +{ + while (count) { + if (buf[count - 1] != '\0' && buf[count - 1] != ' ') + break; + count--; + } + return count; +} + +static size_t ipl_block_get_ascii_scpdata(char *dest, size_t size, + const struct ipl_parameter_block *ipb) +{ + size_t count; + size_t i; + int has_lowercase; + + count = min(size - 1, scpdata_length(ipb->ipl_info.fcp.scp_data, + ipb->ipl_info.fcp.scp_data_len)); + if (!count) + goto out; + + has_lowercase = 0; + for (i = 0; i < count; i++) { + if (!isascii(ipb->ipl_info.fcp.scp_data[i])) { + count = 0; + goto out; + } + if (!has_lowercase && islower(ipb->ipl_info.fcp.scp_data[i])) + has_lowercase = 1; + } + + if (has_lowercase) + memcpy(dest, ipb->ipl_info.fcp.scp_data, count); + else + for (i = 0; i < count; i++) + dest[i] = tolower(ipb->ipl_info.fcp.scp_data[i]); +out: + dest[count] = '\0'; + return count; +} + +static void append_ipl_block_parm(void) +{ + char *parm, *delim; + size_t len, rc = 0; + + len = strlen(early_command_line); + + delim = early_command_line + len; /* '\0' character position */ + parm = early_command_line + len + 1; /* append right after '\0' */ + + switch (early_ipl_block.hdr.pbt) { + case DIAG308_IPL_TYPE_CCW: + rc = ipl_block_get_ascii_vmparm( + parm, COMMAND_LINE_SIZE - len - 1, &early_ipl_block); + break; + case DIAG308_IPL_TYPE_FCP: + rc = ipl_block_get_ascii_scpdata( + parm, COMMAND_LINE_SIZE - len - 1, &early_ipl_block); + break; + } + if (rc) { + if (*parm == '=') + memmove(early_command_line, parm + 1, rc); + else + *delim = ' '; /* replace '\0' with space */ + } +} + +static inline int has_ebcdic_char(const char *str) +{ + int i; + + for (i = 0; str[i]; i++) + if (str[i] & 0x80) + return 1; + return 0; +} + +void setup_boot_command_line(void) +{ + COMMAND_LINE[ARCH_COMMAND_LINE_SIZE - 1] = 0; + /* convert arch command line to ascii if necessary */ + if (has_ebcdic_char(COMMAND_LINE)) + EBCASC(COMMAND_LINE, ARCH_COMMAND_LINE_SIZE); + /* copy arch command line */ + strcpy(early_command_line, strim(COMMAND_LINE)); + + /* append IPL PARM data to the boot command line */ + if (early_ipl_block_valid) + append_ipl_block_parm(); +} + +static char command_line_buf[COMMAND_LINE_SIZE] __section(.data); +static void parse_mem_opt(void) +{ + char *param, *val; + bool enabled; + char *args; + int rc; + + args = strcpy(command_line_buf, early_command_line); + while (*args) { + args = next_arg(args, ¶m, &val); + + if (!strcmp(param, "mem")) { + memory_end = memparse(val, NULL); + memory_end_set = 1; + } + + if (!strcmp(param, "noexec")) { + rc = kstrtobool(val, &enabled); + if (!rc && !enabled) + noexec_disabled = 1; + } + } +} + +void setup_memory_end(void) +{ + parse_mem_opt(); +#ifdef CONFIG_CRASH_DUMP + if (!OLDMEM_BASE && early_ipl_block_valid && + early_ipl_block.hdr.pbt == DIAG308_IPL_TYPE_FCP && + early_ipl_block.ipl_info.fcp.opt == DIAG308_IPL_OPT_DUMP) { + if (!sclp_early_get_hsa_size(&memory_end) && memory_end) + memory_end_set = 1; + } +#endif +} diff --git a/arch/s390/boot/ipl_vmparm.c b/arch/s390/boot/ipl_vmparm.c new file mode 100644 index 000000000000..8dacd5fadfd7 --- /dev/null +++ b/arch/s390/boot/ipl_vmparm.c @@ -0,0 +1,2 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "../kernel/ipl_vmparm.c" diff --git a/arch/s390/boot/mem_detect.c b/arch/s390/boot/mem_detect.c new file mode 100644 index 000000000000..4cb771ba13fa --- /dev/null +++ b/arch/s390/boot/mem_detect.c @@ -0,0 +1,182 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/errno.h> +#include <linux/init.h> +#include <asm/sclp.h> +#include <asm/sections.h> +#include <asm/mem_detect.h> +#include <asm/sparsemem.h> +#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 */ +#define ENTRIES_EXTENDED_MAX \ + (256 * (1020 / 2) * sizeof(struct mem_detect_block)) + +/* + * To avoid corrupting old kernel memory during dump, find lowest memory + * chunk possible either right after the kernel end (decompressed kernel) or + * after initrd (if it is present and there is no hole between the kernel end + * and initrd) + */ +static void *mem_detect_alloc_extended(void) +{ + unsigned long offset = ALIGN(mem_safe_offset(), sizeof(u64)); + + if (IS_ENABLED(BLK_DEV_INITRD) && INITRD_START && INITRD_SIZE && + INITRD_START < offset + ENTRIES_EXTENDED_MAX) + offset = ALIGN(INITRD_START + INITRD_SIZE, sizeof(u64)); + + return (void *)offset; +} + +static struct mem_detect_block *__get_mem_detect_block_ptr(u32 n) +{ + if (n < MEM_INLINED_ENTRIES) + return &mem_detect.entries[n]; + if (unlikely(!mem_detect.entries_extended)) + mem_detect.entries_extended = mem_detect_alloc_extended(); + return &mem_detect.entries_extended[n - MEM_INLINED_ENTRIES]; +} + +/* + * sequential calls to add_mem_detect_block with adjacent memory areas + * are merged together into single memory block. + */ +void add_mem_detect_block(u64 start, u64 end) +{ + struct mem_detect_block *block; + + if (mem_detect.count) { + block = __get_mem_detect_block_ptr(mem_detect.count - 1); + if (block->end == start) { + block->end = end; + return; + } + } + + block = __get_mem_detect_block_ptr(mem_detect.count); + block->start = start; + block->end = end; + mem_detect.count++; +} + +static unsigned long get_mem_detect_end(void) +{ + if (mem_detect.count) + return __get_mem_detect_block_ptr(mem_detect.count - 1)->end; + return 0; +} + +static int __diag260(unsigned long rx1, unsigned long rx2) +{ + register unsigned long _rx1 asm("2") = rx1; + register unsigned long _rx2 asm("3") = rx2; + register unsigned long _ry asm("4") = 0x10; /* storage configuration */ + int rc = -1; /* fail */ + unsigned long reg1, reg2; + psw_t old = S390_lowcore.program_new_psw; + + asm volatile( + " epsw %0,%1\n" + " st %0,%[psw_pgm]\n" + " st %1,%[psw_pgm]+4\n" + " larl %0,1f\n" + " stg %0,%[psw_pgm]+8\n" + " diag %[rx],%[ry],0x260\n" + " ipm %[rc]\n" + " srl %[rc],28\n" + "1:\n" + : "=&d" (reg1), "=&a" (reg2), + [psw_pgm] "=Q" (S390_lowcore.program_new_psw), + [rc] "+&d" (rc), [ry] "+d" (_ry) + : [rx] "d" (_rx1), "d" (_rx2) + : "cc", "memory"); + S390_lowcore.program_new_psw = old; + return rc == 0 ? _ry : -1; +} + +static int diag260(void) +{ + int rc, i; + + struct { + unsigned long start; + unsigned long end; + } storage_extents[8] __aligned(16); /* VM supports up to 8 extends */ + + memset(storage_extents, 0, sizeof(storage_extents)); + rc = __diag260((unsigned long)storage_extents, sizeof(storage_extents)); + if (rc == -1) + return -1; + + for (i = 0; i < min_t(int, rc, ARRAY_SIZE(storage_extents)); i++) + add_mem_detect_block(storage_extents[i].start, storage_extents[i].end + 1); + return 0; +} + +static int tprot(unsigned long addr) +{ + unsigned long pgm_addr; + int rc = -EFAULT; + psw_t old = S390_lowcore.program_new_psw; + + S390_lowcore.program_new_psw.mask = __extract_psw(); + asm volatile( + " larl %[pgm_addr],1f\n" + " stg %[pgm_addr],%[psw_pgm_addr]\n" + " tprot 0(%[addr]),0\n" + " ipm %[rc]\n" + " srl %[rc],28\n" + "1:\n" + : [pgm_addr] "=&d"(pgm_addr), + [psw_pgm_addr] "=Q"(S390_lowcore.program_new_psw.addr), + [rc] "+&d"(rc) + : [addr] "a"(addr) + : "cc", "memory"); + S390_lowcore.program_new_psw = old; + return rc; +} + +static void search_mem_end(void) +{ + unsigned long range = 1 << (MAX_PHYSMEM_BITS - 20); /* in 1MB blocks */ + unsigned long offset = 0; + unsigned long pivot; + + while (range > 1) { + range >>= 1; + pivot = offset + range; + if (!tprot(pivot << 20)) + offset = pivot; + } + + add_mem_detect_block(0, (offset + 1) << 20); +} + +void detect_memory(void) +{ + sclp_early_get_memsize(&max_physmem_end); + + if (!sclp_early_read_storage_info()) { + mem_detect.info_source = MEM_DETECT_SCLP_STOR_INFO; + return; + } + + if (!diag260()) { + mem_detect.info_source = MEM_DETECT_DIAG260; + return; + } + + if (max_physmem_end) { + add_mem_detect_block(0, max_physmem_end); + mem_detect.info_source = MEM_DETECT_SCLP_READ_INFO; + return; + } + + search_mem_end(); + mem_detect.info_source = MEM_DETECT_BIN_SEARCH; + max_physmem_end = get_mem_detect_end(); +} diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c new file mode 100644 index 000000000000..4d441317cdeb --- /dev/null +++ b/arch/s390/boot/startup.c @@ -0,0 +1,64 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/string.h> +#include <asm/setup.h> +#include <asm/sclp.h> +#include "compressed/decompressor.h" +#include "boot.h" + +extern char __boot_data_start[], __boot_data_end[]; + +void error(char *x) +{ + sclp_early_printk("\n\n"); + sclp_early_printk(x); + sclp_early_printk("\n\n -- System halted"); + + disabled_wait(0xdeadbeef); +} + +#ifdef CONFIG_KERNEL_UNCOMPRESSED +unsigned long mem_safe_offset(void) +{ + return vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size; +} +#endif + +static void rescue_initrd(void) +{ + unsigned long min_initrd_addr; + + if (!IS_ENABLED(CONFIG_BLK_DEV_INITRD)) + return; + if (!INITRD_START || !INITRD_SIZE) + return; + min_initrd_addr = mem_safe_offset(); + if (min_initrd_addr <= INITRD_START) + return; + memmove((void *)min_initrd_addr, (void *)INITRD_START, INITRD_SIZE); + INITRD_START = min_initrd_addr; +} + +static void copy_bootdata(void) +{ + if (__boot_data_end - __boot_data_start != vmlinux.bootdata_size) + error(".boot.data section size mismatch"); + memcpy((void *)vmlinux.bootdata_off, __boot_data_start, vmlinux.bootdata_size); +} + +void startup_kernel(void) +{ + void *img; + + rescue_initrd(); + sclp_early_read_info(); + store_ipl_parmblock(); + setup_boot_command_line(); + setup_memory_end(); + detect_memory(); + if (!IS_ENABLED(CONFIG_KERNEL_UNCOMPRESSED)) { + img = decompress_kernel(); + memmove((void *)vmlinux.default_lma, img, vmlinux.image_size); + } + copy_bootdata(); + vmlinux.entry(); +} diff --git a/arch/s390/boot/string.c b/arch/s390/boot/string.c new file mode 100644 index 000000000000..25aca07898ba --- /dev/null +++ b/arch/s390/boot/string.c @@ -0,0 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/ctype.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include "../lib/string.c" + +int strncmp(const char *cs, const char *ct, size_t count) +{ + unsigned char c1, c2; + + while (count) { + c1 = *cs++; + c2 = *ct++; + if (c1 != c2) + return c1 < c2 ? -1 : 1; + if (!c1) + break; + count--; + } + return 0; +} + +char *skip_spaces(const char *str) +{ + while (isspace(*str)) + ++str; + return (char *)str; +} + +char *strim(char *s) +{ + size_t size; + char *end; + + size = strlen(s); + if (!size) + return s; + + end = s + size - 1; + while (end >= s && isspace(*end)) + end--; + *(end + 1) = '\0'; + + return skip_spaces(s); +} + +/* Works only for digits and letters, but small and fast */ +#define TOLOWER(x) ((x) | 0x20) + +static unsigned int simple_guess_base(const char *cp) +{ + if (cp[0] == '0') { + if (TOLOWER(cp[1]) == 'x' && isxdigit(cp[2])) + return 16; + else + return 8; + } else { + return 10; + } +} + +/** + * simple_strtoull - convert a string to an unsigned long long + * @cp: The start of the string + * @endp: A pointer to the end of the parsed string will be placed here + * @base: The number base to use + */ + +unsigned long long simple_strtoull(const char *cp, char **endp, + unsigned int base) +{ + unsigned long long result = 0; + + if (!base) + base = simple_guess_base(cp); + + if (base == 16 && cp[0] == '0' && TOLOWER(cp[1]) == 'x') + cp += 2; + + while (isxdigit(*cp)) { + unsigned int value; + + value = isdigit(*cp) ? *cp - '0' : TOLOWER(*cp) - 'a' + 10; + if (value >= base) + break; + result = result * base + value; + cp++; + } + if (endp) + *endp = (char *)cp; + + return result; +} + +long simple_strtol(const char *cp, char **endp, unsigned int base) +{ + if (*cp == '-') + return -simple_strtoull(cp + 1, endp, base); + + return simple_strtoull(cp, endp, base); +} + +int kstrtobool(const char *s, bool *res) +{ + if (!s) + return -EINVAL; + + switch (s[0]) { + case 'y': + case 'Y': + case '1': + *res = true; + return 0; + case 'n': + case 'N': + case '0': + *res = false; + return 0; + case 'o': + case 'O': + switch (s[1]) { + case 'n': + case 'N': + *res = true; + return 0; + case 'f': + case 'F': + *res = false; + return 0; + default: + break; + } + default: + break; + } + + return -EINVAL; +} diff --git a/arch/s390/crypto/paes_s390.c b/arch/s390/crypto/paes_s390.c index 80b27294c1de..e8d9fa54569c 100644 --- a/arch/s390/crypto/paes_s390.c +++ b/arch/s390/crypto/paes_s390.c @@ -30,26 +30,31 @@ static DEFINE_SPINLOCK(ctrblk_lock); static cpacf_mask_t km_functions, kmc_functions, kmctr_functions; +struct key_blob { + __u8 key[MAXKEYBLOBSIZE]; + unsigned int keylen; +}; + struct s390_paes_ctx { - struct pkey_seckey sk; + struct key_blob kb; struct pkey_protkey pk; unsigned long fc; }; struct s390_pxts_ctx { - struct pkey_seckey sk[2]; + struct key_blob kb[2]; struct pkey_protkey pk[2]; unsigned long fc; }; -static inline int __paes_convert_key(struct pkey_seckey *sk, +static inline int __paes_convert_key(struct key_blob *kb, struct pkey_protkey *pk) { int i, ret; /* try three times in case of failure */ for (i = 0; i < 3; i++) { - ret = pkey_skey2pkey(sk, pk); + ret = pkey_keyblob2pkey(kb->key, kb->keylen, pk); if (ret == 0) break; } @@ -61,7 +66,7 @@ static int __paes_set_key(struct s390_paes_ctx *ctx) { unsigned long fc; - if (__paes_convert_key(&ctx->sk, &ctx->pk)) + if (__paes_convert_key(&ctx->kb, &ctx->pk)) return -EINVAL; /* Pick the correct function code based on the protected key type */ @@ -80,10 +85,8 @@ static int ecb_paes_set_key(struct crypto_tfm *tfm, const u8 *in_key, { struct s390_paes_ctx *ctx = crypto_tfm_ctx(tfm); - if (key_len != SECKEYBLOBSIZE) - return -EINVAL; - - memcpy(ctx->sk.seckey, in_key, SECKEYBLOBSIZE); + memcpy(ctx->kb.key, in_key, key_len); + ctx->kb.keylen = key_len; if (__paes_set_key(ctx)) { tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; return -EINVAL; @@ -147,8 +150,8 @@ static struct crypto_alg ecb_paes_alg = { .cra_list = LIST_HEAD_INIT(ecb_paes_alg.cra_list), .cra_u = { .blkcipher = { - .min_keysize = SECKEYBLOBSIZE, - .max_keysize = SECKEYBLOBSIZE, + .min_keysize = MINKEYBLOBSIZE, + .max_keysize = MAXKEYBLOBSIZE, .setkey = ecb_paes_set_key, .encrypt = ecb_paes_encrypt, .decrypt = ecb_paes_decrypt, @@ -160,7 +163,7 @@ static int __cbc_paes_set_key(struct s390_paes_ctx *ctx) { unsigned long fc; - if (__paes_convert_key(&ctx->sk, &ctx->pk)) + if (__paes_convert_key(&ctx->kb, &ctx->pk)) return -EINVAL; /* Pick the correct function code based on the protected key type */ @@ -179,7 +182,8 @@ static int cbc_paes_set_key(struct crypto_tfm *tfm, const u8 *in_key, { struct s390_paes_ctx *ctx = crypto_tfm_ctx(tfm); - memcpy(ctx->sk.seckey, in_key, SECKEYBLOBSIZE); + memcpy(ctx->kb.key, in_key, key_len); + ctx->kb.keylen = key_len; if (__cbc_paes_set_key(ctx)) { tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; return -EINVAL; @@ -208,7 +212,7 @@ static int cbc_paes_crypt(struct blkcipher_desc *desc, unsigned long modifier, walk->dst.virt.addr, walk->src.virt.addr, n); if (k) ret = blkcipher_walk_done(desc, walk, nbytes - k); - if (n < k) { + if (k < n) { if (__cbc_paes_set_key(ctx) != 0) return blkcipher_walk_done(desc, walk, -EIO); memcpy(param.key, ctx->pk.protkey, MAXPROTKEYSIZE); @@ -250,8 +254,8 @@ static struct crypto_alg cbc_paes_alg = { .cra_list = LIST_HEAD_INIT(cbc_paes_alg.cra_list), .cra_u = { .blkcipher = { - .min_keysize = SECKEYBLOBSIZE, - .max_keysize = SECKEYBLOBSIZE, + .min_keysize = MINKEYBLOBSIZE, + .max_keysize = MAXKEYBLOBSIZE, .ivsize = AES_BLOCK_SIZE, .setkey = cbc_paes_set_key, .encrypt = cbc_paes_encrypt, @@ -264,8 +268,8 @@ static int __xts_paes_set_key(struct s390_pxts_ctx *ctx) { unsigned long fc; - if (__paes_convert_key(&ctx->sk[0], &ctx->pk[0]) || - __paes_convert_key(&ctx->sk[1], &ctx->pk[1])) + if (__paes_convert_key(&ctx->kb[0], &ctx->pk[0]) || + __paes_convert_key(&ctx->kb[1], &ctx->pk[1])) return -EINVAL; if (ctx->pk[0].type != ctx->pk[1].type) @@ -287,10 +291,16 @@ static int xts_paes_set_key(struct crypto_tfm *tfm, const u8 *in_key, { struct s390_pxts_ctx *ctx = crypto_tfm_ctx(tfm); u8 ckey[2 * AES_MAX_KEY_SIZE]; - unsigned int ckey_len; + unsigned int ckey_len, keytok_len; + + if (key_len % 2) + return -EINVAL; - memcpy(ctx->sk[0].seckey, in_key, SECKEYBLOBSIZE); - memcpy(ctx->sk[1].seckey, in_key + SECKEYBLOBSIZE, SECKEYBLOBSIZE); + keytok_len = key_len / 2; + memcpy(ctx->kb[0].key, in_key, keytok_len); + ctx->kb[0].keylen = keytok_len; + memcpy(ctx->kb[1].key, in_key + keytok_len, keytok_len); + ctx->kb[1].keylen = keytok_len; if (__xts_paes_set_key(ctx)) { tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; return -EINVAL; @@ -386,8 +396,8 @@ static struct crypto_alg xts_paes_alg = { .cra_list = LIST_HEAD_INIT(xts_paes_alg.cra_list), .cra_u = { .blkcipher = { - .min_keysize = 2 * SECKEYBLOBSIZE, - .max_keysize = 2 * SECKEYBLOBSIZE, + .min_keysize = 2 * MINKEYBLOBSIZE, + .max_keysize = 2 * MAXKEYBLOBSIZE, .ivsize = AES_BLOCK_SIZE, .setkey = xts_paes_set_key, .encrypt = xts_paes_encrypt, @@ -400,7 +410,7 @@ static int __ctr_paes_set_key(struct s390_paes_ctx *ctx) { unsigned long fc; - if (__paes_convert_key(&ctx->sk, &ctx->pk)) + if (__paes_convert_key(&ctx->kb, &ctx->pk)) return -EINVAL; /* Pick the correct function code based on the protected key type */ @@ -420,7 +430,8 @@ static int ctr_paes_set_key(struct crypto_tfm *tfm, const u8 *in_key, { struct s390_paes_ctx *ctx = crypto_tfm_ctx(tfm); - memcpy(ctx->sk.seckey, in_key, key_len); + memcpy(ctx->kb.key, in_key, key_len); + ctx->kb.keylen = key_len; if (__ctr_paes_set_key(ctx)) { tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; return -EINVAL; @@ -532,8 +543,8 @@ static struct crypto_alg ctr_paes_alg = { .cra_list = LIST_HEAD_INIT(ctr_paes_alg.cra_list), .cra_u = { .blkcipher = { - .min_keysize = SECKEYBLOBSIZE, - .max_keysize = SECKEYBLOBSIZE, + .min_keysize = MINKEYBLOBSIZE, + .max_keysize = MAXKEYBLOBSIZE, .ivsize = AES_BLOCK_SIZE, .setkey = ctr_paes_set_key, .encrypt = ctr_paes_encrypt, diff --git a/arch/s390/defconfig b/arch/s390/defconfig index f40600eb1762..20add000dd6d 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -232,6 +232,7 @@ CONFIG_CRYPTO_USER_API_HASH=m CONFIG_CRYPTO_USER_API_SKCIPHER=m CONFIG_CRYPTO_USER_API_RNG=m CONFIG_ZCRYPT=m +CONFIG_ZCRYPT_MULTIDEVNODES=y CONFIG_PKEY=m CONFIG_CRYPTO_PAES_S390=m CONFIG_CRYPTO_SHA1_S390=m diff --git a/arch/s390/hypfs/hypfs_sprp.c b/arch/s390/hypfs/hypfs_sprp.c index 5d85a039391c..601b70786dc8 100644 --- a/arch/s390/hypfs/hypfs_sprp.c +++ b/arch/s390/hypfs/hypfs_sprp.c @@ -68,40 +68,44 @@ static int hypfs_sprp_create(void **data_ptr, void **free_ptr, size_t *size) static int __hypfs_sprp_ioctl(void __user *user_area) { - struct hypfs_diag304 diag304; + struct hypfs_diag304 *diag304; unsigned long cmd; void __user *udata; void *data; int rc; - if (copy_from_user(&diag304, user_area, sizeof(diag304))) - return -EFAULT; - if ((diag304.args[0] >> 8) != 0 || diag304.args[1] > DIAG304_CMD_MAX) - return -EINVAL; - + rc = -ENOMEM; data = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA); - if (!data) - return -ENOMEM; - - udata = (void __user *)(unsigned long) diag304.data; - if (diag304.args[1] == DIAG304_SET_WEIGHTS || - diag304.args[1] == DIAG304_SET_CAPPING) - if (copy_from_user(data, udata, PAGE_SIZE)) { - rc = -EFAULT; + diag304 = kzalloc(sizeof(*diag304), GFP_KERNEL); + if (!data || !diag304) + goto out; + + rc = -EFAULT; + if (copy_from_user(diag304, user_area, sizeof(*diag304))) + goto out; + rc = -EINVAL; + if ((diag304->args[0] >> 8) != 0 || diag304->args[1] > DIAG304_CMD_MAX) + goto out; + + rc = -EFAULT; + udata = (void __user *)(unsigned long) diag304->data; + if (diag304->args[1] == DIAG304_SET_WEIGHTS || + diag304->args[1] == DIAG304_SET_CAPPING) + if (copy_from_user(data, udata, PAGE_SIZE)) goto out; - } - cmd = *(unsigned long *) &diag304.args[0]; - diag304.rc = hypfs_sprp_diag304(data, cmd); + cmd = *(unsigned long *) &diag304->args[0]; + diag304->rc = hypfs_sprp_diag304(data, cmd); - if (diag304.args[1] == DIAG304_QUERY_PRP) + if (diag304->args[1] == DIAG304_QUERY_PRP) if (copy_to_user(udata, data, PAGE_SIZE)) { rc = -EFAULT; goto out; } - rc = copy_to_user(user_area, &diag304, sizeof(diag304)) ? -EFAULT : 0; + rc = copy_to_user(user_area, diag304, sizeof(*diag304)) ? -EFAULT : 0; out: + kfree(diag304); free_page((unsigned long) data); return rc; } diff --git a/arch/s390/include/asm/appldata.h b/arch/s390/include/asm/appldata.h index 4afbb5938726..c5bd9f4437e5 100644 --- a/arch/s390/include/asm/appldata.h +++ b/arch/s390/include/asm/appldata.h @@ -40,26 +40,27 @@ struct appldata_product_id { u16 mod_lvl; /* modification level */ } __attribute__ ((packed)); -static inline int appldata_asm(struct appldata_product_id *id, + +static inline int appldata_asm(struct appldata_parameter_list *parm_list, + struct appldata_product_id *id, unsigned short fn, void *buffer, unsigned short length) { - struct appldata_parameter_list parm_list; int ry; if (!MACHINE_IS_VM) return -EOPNOTSUPP; - parm_list.diag = 0xdc; - parm_list.function = fn; - parm_list.parlist_length = sizeof(parm_list); - parm_list.buffer_length = length; - parm_list.product_id_addr = (unsigned long) id; - parm_list.buffer_addr = virt_to_phys(buffer); + parm_list->diag = 0xdc; + parm_list->function = fn; + parm_list->parlist_length = sizeof(*parm_list); + parm_list->buffer_length = length; + parm_list->product_id_addr = (unsigned long) id; + parm_list->buffer_addr = virt_to_phys(buffer); diag_stat_inc(DIAG_STAT_X0DC); asm volatile( " diag %1,%0,0xdc" : "=d" (ry) - : "d" (&parm_list), "m" (parm_list), "m" (*id) + : "d" (parm_list), "m" (*parm_list), "m" (*id) : "cc"); return ry; } diff --git a/arch/s390/include/asm/boot_data.h b/arch/s390/include/asm/boot_data.h new file mode 100644 index 000000000000..2d999ccb977a --- /dev/null +++ b/arch/s390/include/asm/boot_data.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_S390_BOOT_DATA_H + +#include <asm/setup.h> +#include <asm/ipl.h> + +extern char early_command_line[COMMAND_LINE_SIZE]; +extern struct ipl_parameter_block early_ipl_block; +extern int early_ipl_block_valid; + +#endif /* _ASM_S390_BOOT_DATA_H */ diff --git a/arch/s390/include/asm/ccwgroup.h b/arch/s390/include/asm/ccwgroup.h index 860cab7479c3..7293c139dd79 100644 --- a/arch/s390/include/asm/ccwgroup.h +++ b/arch/s390/include/asm/ccwgroup.h @@ -64,6 +64,8 @@ extern int ccwgroup_driver_register (struct ccwgroup_driver *cdriver); extern void ccwgroup_driver_unregister (struct ccwgroup_driver *cdriver); int ccwgroup_create_dev(struct device *root, struct ccwgroup_driver *gdrv, int num_devices, const char *buf); +struct ccwgroup_device *get_ccwgroupdev_by_busid(struct ccwgroup_driver *gdrv, + char *bus_id); extern int ccwgroup_set_online(struct ccwgroup_device *gdev); extern int ccwgroup_set_offline(struct ccwgroup_device *gdev); diff --git a/arch/s390/include/asm/facility.h b/arch/s390/include/asm/facility.h index 99c8ce30b3cd..e78cda94456b 100644 --- a/arch/s390/include/asm/facility.h +++ b/arch/s390/include/asm/facility.h @@ -64,11 +64,10 @@ static inline int test_facility(unsigned long nr) * @stfle_fac_list: array where facility list can be stored * @size: size of passed in array in double words */ -static inline void stfle(u64 *stfle_fac_list, int size) +static inline void __stfle(u64 *stfle_fac_list, int size) { unsigned long nr; - preempt_disable(); asm volatile( " stfl 0(0)\n" : "=m" (S390_lowcore.stfl_fac_list)); @@ -85,6 +84,12 @@ static inline void stfle(u64 *stfle_fac_list, int size) nr = (reg0 + 1) * 8; /* # bytes stored by stfle */ } memset((char *) stfle_fac_list + nr, 0, size * 8 - nr); +} + +static inline void stfle(u64 *stfle_fac_list, int size) +{ + preempt_disable(); + __stfle(stfle_fac_list, size); preempt_enable(); } diff --git a/arch/s390/include/asm/ipl.h b/arch/s390/include/asm/ipl.h index ae5135704616..a8389e2d2f03 100644 --- a/arch/s390/include/asm/ipl.h +++ b/arch/s390/include/asm/ipl.h @@ -89,8 +89,8 @@ void __init save_area_add_vxrs(struct save_area *, __vector128 *vxrs); extern void s390_reset_system(void); extern void ipl_store_parameters(void); -extern size_t append_ipl_vmparm(char *, size_t); -extern size_t append_ipl_scpdata(char *, size_t); +extern size_t ipl_block_get_ascii_vmparm(char *dest, size_t size, + const struct ipl_parameter_block *ipb); enum ipl_type { IPL_TYPE_UNKNOWN = 1, diff --git a/arch/s390/include/asm/jump_label.h b/arch/s390/include/asm/jump_label.h index 40f651292aa7..e2d3e6c43395 100644 --- a/arch/s390/include/asm/jump_label.h +++ b/arch/s390/include/asm/jump_label.h @@ -14,41 +14,33 @@ * We use a brcl 0,2 instruction for jump labels at compile time so it * can be easily distinguished from a hotpatch generated instruction. */ -static __always_inline bool arch_static_branch(struct static_key *key, bool branch) +static inline bool arch_static_branch(struct static_key *key, bool branch) { - asm_volatile_goto("0: brcl 0,"__stringify(JUMP_LABEL_NOP_OFFSET)"\n" - ".pushsection __jump_table, \"aw\"\n" - ".balign 8\n" - ".quad 0b, %l[label], %0\n" - ".popsection\n" - : : "X" (&((char *)key)[branch]) : : label); - + asm_volatile_goto("0: brcl 0,"__stringify(JUMP_LABEL_NOP_OFFSET)"\n" + ".pushsection __jump_table,\"aw\"\n" + ".balign 8\n" + ".long 0b-.,%l[label]-.\n" + ".quad %0-.\n" + ".popsection\n" + : : "X" (&((char *)key)[branch]) : : label); return false; label: return true; } -static __always_inline bool arch_static_branch_jump(struct static_key *key, bool branch) +static inline bool arch_static_branch_jump(struct static_key *key, bool branch) { - asm_volatile_goto("0: brcl 15, %l[label]\n" - ".pushsection __jump_table, \"aw\"\n" - ".balign 8\n" - ".quad 0b, %l[label], %0\n" - ".popsection\n" - : : "X" (&((char *)key)[branch]) : : label); - + asm_volatile_goto("0: brcl 15,%l[label]\n" + ".pushsection __jump_table,\"aw\"\n" + ".balign 8\n" + ".long 0b-.,%l[label]-.\n" + ".quad %0-.\n" + ".popsection\n" + : : "X" (&((char *)key)[branch]) : : label); return false; label: return true; } -typedef unsigned long jump_label_t; - -struct jump_entry { - jump_label_t code; - jump_label_t target; - jump_label_t key; -}; - #endif /* __ASSEMBLY__ */ #endif diff --git a/arch/s390/include/asm/kasan.h b/arch/s390/include/asm/kasan.h new file mode 100644 index 000000000000..70930fe5c496 --- /dev/null +++ b/arch/s390/include/asm/kasan.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#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_free_early_identity(void); +#else +static inline void kasan_early_init(void) { } +static inline void kasan_copy_shadow(pgd_t *dst) { } +static inline void kasan_free_early_identity(void) { } +#endif + +#endif diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index 406d940173ab..cc0947e08b6f 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -102,9 +102,9 @@ struct lowcore { __u64 current_task; /* 0x0338 */ __u64 kernel_stack; /* 0x0340 */ - /* Interrupt, panic and restart stack. */ + /* Interrupt, DAT-off and restartstack. */ __u64 async_stack; /* 0x0348 */ - __u64 panic_stack; /* 0x0350 */ + __u64 nodat_stack; /* 0x0350 */ __u64 restart_stack; /* 0x0358 */ /* Restart function and parameter. */ diff --git a/arch/s390/include/asm/mem_detect.h b/arch/s390/include/asm/mem_detect.h new file mode 100644 index 000000000000..6114b92ab667 --- /dev/null +++ b/arch/s390/include/asm/mem_detect.h @@ -0,0 +1,82 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_S390_MEM_DETECT_H +#define _ASM_S390_MEM_DETECT_H + +#include <linux/types.h> + +enum mem_info_source { + MEM_DETECT_NONE = 0, + MEM_DETECT_SCLP_STOR_INFO, + MEM_DETECT_DIAG260, + MEM_DETECT_SCLP_READ_INFO, + MEM_DETECT_BIN_SEARCH +}; + +struct mem_detect_block { + u64 start; + u64 end; +}; + +/* + * Storage element id is defined as 1 byte (up to 256 storage elements). + * In practise only storage element id 0 and 1 are used). + * According to architecture one storage element could have as much as + * 1020 subincrements. 255 mem_detect_blocks are embedded in mem_detect_info. + * If more mem_detect_blocks are required, a block of memory from already + * known mem_detect_block is taken (entries_extended points to it). + */ +#define MEM_INLINED_ENTRIES 255 /* (PAGE_SIZE - 16) / 16 */ + +struct mem_detect_info { + u32 count; + u8 info_source; + struct mem_detect_block entries[MEM_INLINED_ENTRIES]; + struct mem_detect_block *entries_extended; +}; +extern struct mem_detect_info mem_detect; + +void add_mem_detect_block(u64 start, u64 end); + +static inline int __get_mem_detect_block(u32 n, unsigned long *start, + unsigned long *end) +{ + if (n >= mem_detect.count) { + *start = 0; + *end = 0; + return -1; + } + + if (n < MEM_INLINED_ENTRIES) { + *start = (unsigned long)mem_detect.entries[n].start; + *end = (unsigned long)mem_detect.entries[n].end; + } else { + *start = (unsigned long)mem_detect.entries_extended[n - MEM_INLINED_ENTRIES].start; + *end = (unsigned long)mem_detect.entries_extended[n - MEM_INLINED_ENTRIES].end; + } + return 0; +} + +/** + * for_each_mem_detect_block - early online memory range iterator + * @i: an integer used as loop variable + * @p_start: ptr to unsigned long for start address of the range + * @p_end: ptr to unsigned long for end address of the range + * + * Walks over detected online memory ranges. + */ +#define for_each_mem_detect_block(i, p_start, p_end) \ + for (i = 0, __get_mem_detect_block(i, p_start, p_end); \ + i < mem_detect.count; \ + i++, __get_mem_detect_block(i, p_start, p_end)) + +static inline void get_mem_detect_reserved(unsigned long *start, + unsigned long *size) +{ + *start = (unsigned long)mem_detect.entries_extended; + if (mem_detect.count > MEM_INLINED_ENTRIES) + *size = (mem_detect.count - MEM_INLINED_ENTRIES) * sizeof(struct mem_detect_block); + else + *size = 0; +} + +#endif diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index f31a15044c24..bcfb6371086f 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -16,7 +16,13 @@ typedef struct { unsigned long asce; unsigned long asce_limit; unsigned long vdso_base; - /* The mmu context allocates 4K page tables. */ + /* + * The following bitfields need a down_write on the mm + * semaphore when they are written to. As they are only + * written once, they can be read without a lock. + * + * The mmu context allocates 4K page tables. + */ unsigned int alloc_pgste:1; /* The mmu context uses extended page tables. */ unsigned int has_pgste:1; @@ -26,6 +32,8 @@ typedef struct { unsigned int uses_cmm:1; /* The gmaps associated with this context are allowed to use huge pages. */ unsigned int allow_gmap_hpage_1m:1; + /* The mmu context is for compat task */ + unsigned int compat_mm:1; } mm_context_t; #define INIT_MM_CONTEXT(name) \ diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index 0717ee76885d..dbd689d556ce 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -25,6 +25,7 @@ static inline int init_new_context(struct task_struct *tsk, atomic_set(&mm->context.flush_count, 0); mm->context.gmap_asce = 0; mm->context.flush_mm = 0; + mm->context.compat_mm = 0; #ifdef CONFIG_PGSTE mm->context.alloc_pgste = page_table_allocate_pgste || test_thread_flag(TIF_PGSTE) || diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h index 41e3908b397f..a4d38092530a 100644 --- a/arch/s390/include/asm/page.h +++ b/arch/s390/include/asm/page.h @@ -161,6 +161,7 @@ static inline int devmem_is_allowed(unsigned long pfn) #define virt_to_pfn(kaddr) (__pa(kaddr) >> PAGE_SHIFT) #define pfn_to_virt(pfn) __va((pfn) << PAGE_SHIFT) +#define pfn_to_kaddr(pfn) pfn_to_virt(pfn) #define virt_to_page(kaddr) pfn_to_page(virt_to_pfn(kaddr)) #define page_to_virt(page) pfn_to_virt(page_to_pfn(page)) diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 0e7cb0dc9c33..411d435e7a7d 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -341,6 +341,8 @@ static inline int is_module_addr(void *addr) #define PTRS_PER_P4D _CRST_ENTRIES #define PTRS_PER_PGD _CRST_ENTRIES +#define MAX_PTRS_PER_P4D PTRS_PER_P4D + /* * Segment table and region3 table entry encoding * (R = read-only, I = invalid, y = young bit): @@ -466,6 +468,12 @@ static inline int is_module_addr(void *addr) _SEGMENT_ENTRY_YOUNG | \ _SEGMENT_ENTRY_PROTECT | \ _SEGMENT_ENTRY_NOEXEC) +#define SEGMENT_KERNEL_EXEC __pgprot(_SEGMENT_ENTRY | \ + _SEGMENT_ENTRY_LARGE | \ + _SEGMENT_ENTRY_READ | \ + _SEGMENT_ENTRY_WRITE | \ + _SEGMENT_ENTRY_YOUNG | \ + _SEGMENT_ENTRY_DIRTY) /* * Region3 entry (large page) protection definitions. @@ -599,6 +607,14 @@ static inline int pgd_bad(pgd_t pgd) return (pgd_val(pgd) & mask) != 0; } +static inline unsigned long pgd_pfn(pgd_t pgd) +{ + unsigned long origin_mask; + + origin_mask = _REGION_ENTRY_ORIGIN; + return (pgd_val(pgd) & origin_mask) >> PAGE_SHIFT; +} + static inline int p4d_folded(p4d_t p4d) { return (p4d_val(p4d) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R2; @@ -1171,6 +1187,7 @@ static inline pte_t mk_pte(struct page *page, pgprot_t pgprot) #define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) #define pgd_offset_k(address) pgd_offset(&init_mm, address) +#define pgd_offset_raw(pgd, addr) ((pgd) + pgd_index(addr)) #define pmd_deref(pmd) (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) #define pud_deref(pud) (pud_val(pud) & _REGION_ENTRY_ORIGIN) @@ -1210,7 +1227,8 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address) #define pmd_page(pmd) pfn_to_page(pmd_pfn(pmd)) #define pud_page(pud) pfn_to_page(pud_pfn(pud)) -#define p4d_page(pud) pfn_to_page(p4d_pfn(p4d)) +#define p4d_page(p4d) pfn_to_page(p4d_pfn(p4d)) +#define pgd_page(pgd) pfn_to_page(pgd_pfn(pgd)) /* Find an entry in the lowest level page table.. */ #define pte_offset(pmd, addr) ((pte_t *) pmd_deref(*(pmd)) + pte_index(addr)) diff --git a/arch/s390/include/asm/pkey.h b/arch/s390/include/asm/pkey.h index 053117ba7328..9b6e79077866 100644 --- a/arch/s390/include/asm/pkey.h +++ b/arch/s390/include/asm/pkey.h @@ -109,4 +109,30 @@ int pkey_verifykey(const struct pkey_seckey *seckey, u16 *pcardnr, u16 *pdomain, u16 *pkeysize, u32 *pattributes); +/* + * In-kernel API: Generate (AES) random protected key. + * @param keytype one of the PKEY_KEYTYPE values + * @param protkey pointer to buffer receiving the protected key + * @return 0 on success, negative errno value on failure + */ +int pkey_genprotkey(__u32 keytype, struct pkey_protkey *protkey); + +/* + * In-kernel API: Verify an (AES) protected key. + * @param protkey pointer to buffer containing the protected key to verify + * @return 0 on success, negative errno value on failure. In case the protected + * key is not valid -EKEYREJECTED is returned + */ +int pkey_verifyprotkey(const struct pkey_protkey *protkey); + +/* + * In-kernel API: Transform an key blob (of any type) into a protected key. + * @param key pointer to a buffer containing the key blob + * @param keylen size of the key blob in bytes + * @param protkey pointer to buffer receiving the protected key + * @return 0 on success, negative errno value on failure + */ +int pkey_keyblob2pkey(const __u8 *key, __u32 keylen, + struct pkey_protkey *protkey); + #endif /* _KAPI_PKEY_H */ diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 7f2953c15c37..34768e6ef4fb 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -242,7 +242,7 @@ static inline unsigned long current_stack_pointer(void) return sp; } -static inline unsigned short stap(void) +static __no_sanitize_address_or_inline unsigned short stap(void) { unsigned short cpu_address; @@ -250,6 +250,55 @@ static inline unsigned short stap(void) return cpu_address; } +#define CALL_ARGS_0() \ + register unsigned long r2 asm("2") +#define CALL_ARGS_1(arg1) \ + register unsigned long r2 asm("2") = (unsigned long)(arg1) +#define CALL_ARGS_2(arg1, arg2) \ + CALL_ARGS_1(arg1); \ + register unsigned long r3 asm("3") = (unsigned long)(arg2) +#define CALL_ARGS_3(arg1, arg2, arg3) \ + CALL_ARGS_2(arg1, arg2); \ + register unsigned long r4 asm("4") = (unsigned long)(arg3) +#define CALL_ARGS_4(arg1, arg2, arg3, arg4) \ + CALL_ARGS_3(arg1, arg2, arg3); \ + register unsigned long r4 asm("5") = (unsigned long)(arg4) +#define CALL_ARGS_5(arg1, arg2, arg3, arg4, arg5) \ + CALL_ARGS_4(arg1, arg2, arg3, arg4); \ + register unsigned long r4 asm("6") = (unsigned long)(arg5) + +#define CALL_FMT_0 +#define CALL_FMT_1 CALL_FMT_0, "0" (r2) +#define CALL_FMT_2 CALL_FMT_1, "d" (r3) +#define CALL_FMT_3 CALL_FMT_2, "d" (r4) +#define CALL_FMT_4 CALL_FMT_3, "d" (r5) +#define CALL_FMT_5 CALL_FMT_4, "d" (r6) + +#define CALL_CLOBBER_5 "0", "1", "14", "cc", "memory" +#define CALL_CLOBBER_4 CALL_CLOBBER_5 +#define CALL_CLOBBER_3 CALL_CLOBBER_4, "5" +#define CALL_CLOBBER_2 CALL_CLOBBER_3, "4" +#define CALL_CLOBBER_1 CALL_CLOBBER_2, "3" +#define CALL_CLOBBER_0 CALL_CLOBBER_1 + +#define CALL_ON_STACK(fn, stack, nr, args...) \ +({ \ + CALL_ARGS_##nr(args); \ + unsigned long prev; \ + \ + asm volatile( \ + " la %[_prev],0(15)\n" \ + " la 15,0(%[_stack])\n" \ + " stg %[_prev],%[_bc](15)\n" \ + " brasl 14,%[_fn]\n" \ + " la 15,0(%[_prev])\n" \ + : "+&d" (r2), [_prev] "=&a" (prev) \ + : [_stack] "a" (stack), \ + [_bc] "i" (offsetof(struct stack_frame, back_chain)), \ + [_fn] "X" (fn) CALL_FMT_##nr : CALL_CLOBBER_##nr); \ + r2; \ +}) + /* * Give up the time slice of the virtual PU. */ @@ -287,7 +336,7 @@ static inline void __load_psw(psw_t psw) * Set PSW mask to specified value, while leaving the * PSW addr pointing to the next instruction. */ -static inline void __load_psw_mask(unsigned long mask) +static __no_sanitize_address_or_inline void __load_psw_mask(unsigned long mask) { unsigned long addr; psw_t psw; diff --git a/arch/s390/include/asm/qdio.h b/arch/s390/include/asm/qdio.h index 9c9970a5dfb1..d46edde7e458 100644 --- a/arch/s390/include/asm/qdio.h +++ b/arch/s390/include/asm/qdio.h @@ -252,13 +252,11 @@ struct slsb { * (for communication with upper layer programs) * (only required for use with completion queues) * @flags: flags indicating state of buffer - * @aob: pointer to QAOB used for the particular SBAL * @user: pointer to upper layer program's state information related to SBAL * (stored in user1 data of QAOB) */ struct qdio_outbuf_state { u8 flags; - struct qaob *aob; void *user; }; diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h index 3cae9168f63c..0cd4bda85eb1 100644 --- a/arch/s390/include/asm/sclp.h +++ b/arch/s390/include/asm/sclp.h @@ -95,6 +95,7 @@ extern struct sclp_info sclp; struct zpci_report_error_header { u8 version; /* Interface version byte */ u8 action; /* Action qualifier byte + * 0: Adapter Reset Request * 1: Deconfigure and repair action requested * (OpenCrypto Problem Call Home) * 2: Informational Report @@ -104,12 +105,17 @@ struct zpci_report_error_header { u8 data[0]; /* Subsequent Data passed verbatim to SCLP ET 24 */ } __packed; +int sclp_early_read_info(void); +int sclp_early_read_storage_info(void); int sclp_early_get_core_info(struct sclp_core_info *info); void sclp_early_get_ipl_info(struct sclp_ipl_info *info); void sclp_early_detect(void); void sclp_early_printk(const char *s); -void __sclp_early_printk(const char *s, unsigned int len); +void sclp_early_printk_force(const char *s); +void __sclp_early_printk(const char *s, unsigned int len, unsigned int force); +int sclp_early_get_memsize(unsigned long *mem); +int sclp_early_get_hsa_size(unsigned long *hsa_size); int _sclp_get_core_info(struct sclp_core_info *info); int sclp_core_configure(u8 core); int sclp_core_deconfigure(u8 core); diff --git a/arch/s390/include/asm/sections.h b/arch/s390/include/asm/sections.h index 724faede8ac5..7afe4620685c 100644 --- a/arch/s390/include/asm/sections.h +++ b/arch/s390/include/asm/sections.h @@ -4,4 +4,16 @@ #include <asm-generic/sections.h> +/* + * .boot.data section contains variables "shared" between the decompressor and + * the decompressed kernel. The decompressor will store values in them, and + * copy over to the decompressed image before starting it. + * + * Each variable end up in its own intermediate section .boot.data.<var name>, + * those sections are later sorted by alignment + name and merged together into + * final .boot.data section, which should be identical in the decompressor and + * the decompressed kernel (that is checked during the build). + */ +#define __bootdata(var) __section(.boot.data.var) var + #endif diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index 1d66016f4170..efda97804aa4 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -65,12 +65,11 @@ #define OLDMEM_SIZE (*(unsigned long *) (OLDMEM_SIZE_OFFSET)) #define COMMAND_LINE ((char *) (COMMAND_LINE_OFFSET)) +extern int noexec_disabled; extern int memory_end_set; extern unsigned long memory_end; extern unsigned long max_physmem_end; -extern void detect_memory_memblock(void); - #define MACHINE_IS_VM (S390_lowcore.machine_flags & MACHINE_FLAG_VM) #define MACHINE_IS_KVM (S390_lowcore.machine_flags & MACHINE_FLAG_KVM) #define MACHINE_IS_LPAR (S390_lowcore.machine_flags & MACHINE_FLAG_LPAR) diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h index 50f26fc9acb2..116cc15a4b8a 100644 --- a/arch/s390/include/asm/string.h +++ b/arch/s390/include/asm/string.h @@ -53,6 +53,27 @@ char *strstr(const char *s1, const char *s2); #undef __HAVE_ARCH_STRSEP #undef __HAVE_ARCH_STRSPN +#if defined(CONFIG_KASAN) && !defined(__SANITIZE_ADDRESS__) + +extern void *__memcpy(void *dest, const void *src, size_t n); +extern void *__memset(void *s, int c, size_t n); +extern void *__memmove(void *dest, const void *src, size_t n); + +/* + * For files that are not instrumented (e.g. mm/slub.c) we + * should use not instrumented version of mem* functions. + */ + +#define memcpy(dst, src, len) __memcpy(dst, src, len) +#define memmove(dst, src, len) __memmove(dst, src, len) +#define memset(s, c, n) __memset(s, c, n) + +#ifndef __NO_FORTIFY +#define __NO_FORTIFY /* FORTIFY_SOURCE uses __builtin_memcpy, etc. */ +#endif + +#endif /* defined(CONFIG_KASAN) && !defined(__SANITIZE_ADDRESS__) */ + void *__memset16(uint16_t *s, uint16_t v, size_t count); void *__memset32(uint32_t *s, uint32_t v, size_t count); void *__memset64(uint64_t *s, uint64_t v, size_t count); diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h index 3c883c368eb0..27248f42a03c 100644 --- a/arch/s390/include/asm/thread_info.h +++ b/arch/s390/include/asm/thread_info.h @@ -11,19 +11,24 @@ #include <linux/const.h> /* - * Size of kernel stack for each process + * General size of kernel stacks */ +#ifdef CONFIG_KASAN +#define THREAD_SIZE_ORDER 3 +#else #define THREAD_SIZE_ORDER 2 -#define ASYNC_ORDER 2 - +#endif +#define BOOT_STACK_ORDER 2 #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER) -#define ASYNC_SIZE (PAGE_SIZE << ASYNC_ORDER) #ifndef __ASSEMBLY__ #include <asm/lowcore.h> #include <asm/page.h> #include <asm/processor.h> +#define STACK_INIT_OFFSET \ + (THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs)) + /* * low level task data that entry.S needs immediate access to * - this struct should fit entirely inside of one cache line diff --git a/arch/s390/include/asm/vmlinux.lds.h b/arch/s390/include/asm/vmlinux.lds.h new file mode 100644 index 000000000000..2d127f900352 --- /dev/null +++ b/arch/s390/include/asm/vmlinux.lds.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include <asm/page.h> + +/* + * .boot.data section is shared between the decompressor code and the + * decompressed kernel. The decompressor will store values in it, and copy + * over to the decompressed image before starting it. + * + * .boot.data variables are kept in separate .boot.data.<var name> sections, + * which are sorted by alignment first, then by name before being merged + * into single .boot.data section. This way big holes cased by page aligned + * structs are avoided and linker produces consistent result. + */ +#define BOOT_DATA \ + . = ALIGN(PAGE_SIZE); \ + .boot.data : { \ + __boot_data_start = .; \ + *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.boot.data*))) \ + __boot_data_end = .; \ + } diff --git a/arch/s390/include/uapi/asm/Kbuild b/arch/s390/include/uapi/asm/Kbuild index e364873e0d10..dc38a90cf091 100644 --- a/arch/s390/include/uapi/asm/Kbuild +++ b/arch/s390/include/uapi/asm/Kbuild @@ -18,3 +18,4 @@ generic-y += shmbuf.h generic-y += sockios.h generic-y += swab.h generic-y += termbits.h +generic-y += siginfo.h
\ No newline at end of file diff --git a/arch/s390/include/uapi/asm/pkey.h b/arch/s390/include/uapi/asm/pkey.h index 6f84a53c3270..c0e86ce4a00b 100644 --- a/arch/s390/include/uapi/asm/pkey.h +++ b/arch/s390/include/uapi/asm/pkey.h @@ -21,9 +21,13 @@ #define PKEY_IOCTL_MAGIC 'p' #define SECKEYBLOBSIZE 64 /* secure key blob size is always 64 bytes */ +#define PROTKEYBLOBSIZE 80 /* protected key blob size is always 80 bytes */ #define MAXPROTKEYSIZE 64 /* a protected key blob may be up to 64 bytes */ #define MAXCLRKEYSIZE 32 /* a clear key value may be up to 32 bytes */ +#define MINKEYBLOBSIZE SECKEYBLOBSIZE /* Minimum size of a key blob */ +#define MAXKEYBLOBSIZE PROTKEYBLOBSIZE /* Maximum size of a key blob */ + /* defines for the type field within the pkey_protkey struct */ #define PKEY_KEYTYPE_AES_128 1 #define PKEY_KEYTYPE_AES_192 2 @@ -129,4 +133,34 @@ struct pkey_verifykey { #define PKEY_VERIFY_ATTR_AES 0x00000001 /* key is an AES key */ #define PKEY_VERIFY_ATTR_OLD_MKVP 0x00000100 /* key has old MKVP value */ +/* + * Generate (AES) random protected key. + */ +struct pkey_genprotk { + __u32 keytype; /* in: key type to generate */ + struct pkey_protkey protkey; /* out: the protected key */ +}; + +#define PKEY_GENPROTK _IOWR(PKEY_IOCTL_MAGIC, 0x08, struct pkey_genprotk) + +/* + * Verify an (AES) protected key. + */ +struct pkey_verifyprotk { + struct pkey_protkey protkey; /* in: the protected key to verify */ +}; + +#define PKEY_VERIFYPROTK _IOW(PKEY_IOCTL_MAGIC, 0x09, struct pkey_verifyprotk) + +/* + * Transform an key blob (of any type) into a protected key + */ +struct pkey_kblob2pkey { + __u8 __user *key; /* in: the key blob */ + __u32 keylen; /* in: the key blob length */ + struct pkey_protkey protkey; /* out: the protected key */ +}; + +#define PKEY_KBLOB2PROTK _IOWR(PKEY_IOCTL_MAGIC, 0x0A, struct pkey_kblob2pkey) + #endif /* _UAPI_PKEY_H */ diff --git a/arch/s390/include/uapi/asm/siginfo.h b/arch/s390/include/uapi/asm/siginfo.h deleted file mode 100644 index 6984820f2f1c..000000000000 --- a/arch/s390/include/uapi/asm/siginfo.h +++ /dev/null @@ -1,17 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -/* - * S390 version - * - * Derived from "include/asm-i386/siginfo.h" - */ - -#ifndef _S390_SIGINFO_H -#define _S390_SIGINFO_H - -#ifdef __s390x__ -#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) -#endif - -#include <asm-generic/siginfo.h> - -#endif diff --git a/arch/s390/include/uapi/asm/zcrypt.h b/arch/s390/include/uapi/asm/zcrypt.h index 2bb1f3bb98ac..42c81a95e97b 100644 --- a/arch/s390/include/uapi/asm/zcrypt.h +++ b/arch/s390/include/uapi/asm/zcrypt.h @@ -2,9 +2,9 @@ /* * include/asm-s390/zcrypt.h * - * zcrypt 2.1.0 (user-visible header) + * zcrypt 2.2.1 (user-visible header) * - * Copyright IBM Corp. 2001, 2006 + * Copyright IBM Corp. 2001, 2018 * Author(s): Robert Burroughs * Eric Rossman (edrossma@us.ibm.com) * @@ -15,12 +15,15 @@ #define __ASM_S390_ZCRYPT_H #define ZCRYPT_VERSION 2 -#define ZCRYPT_RELEASE 1 +#define ZCRYPT_RELEASE 2 #define ZCRYPT_VARIANT 1 #include <linux/ioctl.h> #include <linux/compiler.h> +/* Name of the zcrypt device driver. */ +#define ZCRYPT_NAME "zcrypt" + /** * struct ica_rsa_modexpo * @@ -310,6 +313,16 @@ struct zcrypt_device_matrix_ext { #define ZCRYPT_PERDEV_REQCNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x5a, int[MAX_ZDEV_CARDIDS_EXT]) /* + * Support for multiple zcrypt device nodes. + */ + +/* Nr of minor device node numbers to allocate. */ +#define ZCRYPT_MAX_MINOR_NODES 256 + +/* Max amount of possible ioctls */ +#define MAX_ZDEV_IOCTLS (1 << _IOC_NRBITS) + +/* * Only deprecated defines, structs and ioctls below this line. */ diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index dbfd1730e631..386b1abb217b 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -23,6 +23,10 @@ KCOV_INSTRUMENT_early_nobss.o := n UBSAN_SANITIZE_early.o := n UBSAN_SANITIZE_early_nobss.o := n +KASAN_SANITIZE_early_nobss.o := n +KASAN_SANITIZE_ipl.o := n +KASAN_SANITIZE_machine_kexec.o := n + # # Passing null pointers is ok for smp code, since we access the lowcore here. # @@ -47,7 +51,7 @@ obj-y += debug.o irq.o ipl.o dis.o diag.o vdso.o early_nobss.o obj-y += sysinfo.o jump_label.o lgr.o os_info.o machine_kexec.o pgm_check.o obj-y += runtime_instr.o cache.o fpu.o dumpstack.o guarded_storage.o sthyi.o obj-y += entry.o reipl.o relocate_kernel.o kdebugfs.o alternative.o -obj-y += nospec-branch.o +obj-y += nospec-branch.o ipl_vmparm.o extra-y += head64.o vmlinux.lds diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 66e830f1c7bf..164bec175628 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -159,7 +159,7 @@ int main(void) OFFSET(__LC_CURRENT, lowcore, current_task); OFFSET(__LC_KERNEL_STACK, lowcore, kernel_stack); OFFSET(__LC_ASYNC_STACK, lowcore, async_stack); - OFFSET(__LC_PANIC_STACK, lowcore, panic_stack); + OFFSET(__LC_NODAT_STACK, lowcore, nodat_stack); OFFSET(__LC_RESTART_STACK, lowcore, restart_stack); OFFSET(__LC_RESTART_FN, lowcore, restart_fn); OFFSET(__LC_RESTART_DATA, lowcore, restart_data); diff --git a/arch/s390/kernel/base.S b/arch/s390/kernel/base.S index b65874b0b412..f268fca67e82 100644 --- a/arch/s390/kernel/base.S +++ b/arch/s390/kernel/base.S @@ -18,7 +18,7 @@ ENTRY(s390_base_mcck_handler) basr %r13,0 -0: lg %r15,__LC_PANIC_STACK # load panic stack +0: lg %r15,__LC_NODAT_STACK # load panic stack aghi %r15,-STACK_FRAME_OVERHEAD larl %r1,s390_base_mcck_handler_fn lg %r9,0(%r1) diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c index 5b23c4f6e50c..cb7f55bbe06e 100644 --- a/arch/s390/kernel/dumpstack.c +++ b/arch/s390/kernel/dumpstack.c @@ -30,7 +30,7 @@ * The stack trace can start at any of the three stacks and can potentially * touch all of them. The order is: panic stack, async stack, sync stack. */ -static unsigned long +static unsigned long __no_sanitize_address __dump_trace(dump_trace_func_t func, void *data, unsigned long sp, unsigned long low, unsigned long high) { @@ -77,11 +77,11 @@ void dump_trace(dump_trace_func_t func, void *data, struct task_struct *task, frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); #ifdef CONFIG_CHECK_STACK sp = __dump_trace(func, data, sp, - S390_lowcore.panic_stack + frame_size - PAGE_SIZE, - S390_lowcore.panic_stack + frame_size); + S390_lowcore.nodat_stack + frame_size - THREAD_SIZE, + S390_lowcore.nodat_stack + frame_size); #endif sp = __dump_trace(func, data, sp, - S390_lowcore.async_stack + frame_size - ASYNC_SIZE, + S390_lowcore.async_stack + frame_size - THREAD_SIZE, S390_lowcore.async_stack + frame_size); task = task ?: current; __dump_trace(func, data, sp, @@ -124,7 +124,7 @@ void show_registers(struct pt_regs *regs) char *mode; mode = user_mode(regs) ? "User" : "Krnl"; - printk("%s PSW : %p %p", mode, (void *)regs->psw.mask, (void *)regs->psw.addr); + printk("%s PSW : %px %px", mode, (void *)regs->psw.mask, (void *)regs->psw.addr); if (!user_mode(regs)) pr_cont(" (%pSR)", (void *)regs->psw.addr); pr_cont("\n"); diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 5b28b434f8a1..af5c2b3f7065 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -29,10 +29,9 @@ #include <asm/cpcmd.h> #include <asm/sclp.h> #include <asm/facility.h> +#include <asm/boot_data.h> #include "entry.h" -static void __init setup_boot_command_line(void); - /* * Initialize storage key for kernel pages */ @@ -284,51 +283,11 @@ static int __init cad_setup(char *str) } early_param("cad", cad_setup); -/* Set up boot command line */ -static void __init append_to_cmdline(size_t (*ipl_data)(char *, size_t)) -{ - char *parm, *delim; - size_t rc, len; - - len = strlen(boot_command_line); - - delim = boot_command_line + len; /* '\0' character position */ - parm = boot_command_line + len + 1; /* append right after '\0' */ - - rc = ipl_data(parm, COMMAND_LINE_SIZE - len - 1); - if (rc) { - if (*parm == '=') - memmove(boot_command_line, parm + 1, rc); - else - *delim = ' '; /* replace '\0' with space */ - } -} - -static inline int has_ebcdic_char(const char *str) -{ - int i; - - for (i = 0; str[i]; i++) - if (str[i] & 0x80) - return 1; - return 0; -} - +char __bootdata(early_command_line)[COMMAND_LINE_SIZE]; static void __init setup_boot_command_line(void) { - COMMAND_LINE[ARCH_COMMAND_LINE_SIZE - 1] = 0; - /* convert arch command line to ascii if necessary */ - if (has_ebcdic_char(COMMAND_LINE)) - EBCASC(COMMAND_LINE, ARCH_COMMAND_LINE_SIZE); /* copy arch command line */ - strlcpy(boot_command_line, strstrip(COMMAND_LINE), - ARCH_COMMAND_LINE_SIZE); - - /* append IPL PARM data to the boot command line */ - if (MACHINE_IS_VM) - append_to_cmdline(append_ipl_vmparm); - - append_to_cmdline(append_ipl_scpdata); + strlcpy(boot_command_line, early_command_line, ARCH_COMMAND_LINE_SIZE); } static void __init check_image_bootable(void) diff --git a/arch/s390/kernel/early_nobss.c b/arch/s390/kernel/early_nobss.c index 2d84fc48df3a..8d73f7fae16e 100644 --- a/arch/s390/kernel/early_nobss.c +++ b/arch/s390/kernel/early_nobss.c @@ -13,8 +13,8 @@ #include <linux/string.h> #include <asm/sections.h> #include <asm/lowcore.h> -#include <asm/setup.h> #include <asm/timex.h> +#include <asm/kasan.h> #include "entry.h" static void __init reset_tod_clock(void) @@ -32,26 +32,6 @@ static void __init reset_tod_clock(void) S390_lowcore.last_update_clock = TOD_UNIX_EPOCH; } -static void __init rescue_initrd(void) -{ - unsigned long min_initrd_addr = (unsigned long) _end + (4UL << 20); - - /* - * Just like in case of IPL from VM reader we make sure there is a - * gap of 4MB between end of kernel and start of initrd. - * That way we can also be sure that saving an NSS will succeed, - * which however only requires different segments. - */ - if (!IS_ENABLED(CONFIG_BLK_DEV_INITRD)) - return; - if (!INITRD_START || !INITRD_SIZE) - return; - if (INITRD_START >= min_initrd_addr) - return; - memmove((void *) min_initrd_addr, (void *) INITRD_START, INITRD_SIZE); - INITRD_START = min_initrd_addr; -} - static void __init clear_bss_section(void) { memset(__bss_start, 0, __bss_stop - __bss_start); @@ -60,6 +40,6 @@ static void __init clear_bss_section(void) void __init startup_init_nobss(void) { reset_tod_clock(); - rescue_initrd(); clear_bss_section(); + kasan_early_init(); } diff --git a/arch/s390/kernel/early_printk.c b/arch/s390/kernel/early_printk.c index 9431784d7796..40c1dfec944e 100644 --- a/arch/s390/kernel/early_printk.c +++ b/arch/s390/kernel/early_printk.c @@ -10,7 +10,7 @@ static void sclp_early_write(struct console *con, const char *s, unsigned int len) { - __sclp_early_printk(s, len); + __sclp_early_printk(s, len, 0); } static struct console sclp_early_console = { diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 150130c897c3..724fba4d09d2 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -85,14 +85,34 @@ _LPP_OFFSET = __LC_LPP #endif .endm - .macro CHECK_STACK stacksize,savearea + .macro CHECK_STACK savearea #ifdef CONFIG_CHECK_STACK - tml %r15,\stacksize - CONFIG_STACK_GUARD + tml %r15,STACK_SIZE - CONFIG_STACK_GUARD lghi %r14,\savearea jz stack_overflow #endif .endm + .macro CHECK_VMAP_STACK savearea,oklabel +#ifdef CONFIG_VMAP_STACK + lgr %r14,%r15 + nill %r14,0x10000 - STACK_SIZE + oill %r14,STACK_INIT + clg %r14,__LC_KERNEL_STACK + je \oklabel + clg %r14,__LC_ASYNC_STACK + je \oklabel + clg %r14,__LC_NODAT_STACK + je \oklabel + clg %r14,__LC_RESTART_STACK + je \oklabel + lghi %r14,\savearea + j stack_overflow +#else + j \oklabel +#endif + .endm + .macro SWITCH_ASYNC savearea,timer tmhh %r8,0x0001 # interrupting from user ? jnz 1f @@ -104,11 +124,11 @@ _LPP_OFFSET = __LC_LPP brasl %r14,cleanup_critical tmhh %r8,0x0001 # retest problem state after cleanup jnz 1f -0: lg %r14,__LC_ASYNC_STACK # are we already on the async stack? +0: lg %r14,__LC_ASYNC_STACK # are we already on the target stack? slgr %r14,%r15 srag %r14,%r14,STACK_SHIFT jnz 2f - CHECK_STACK 1<<STACK_SHIFT,\savearea + CHECK_STACK \savearea aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) j 3f 1: UPDATE_VTIME %r14,%r15,\timer @@ -600,9 +620,10 @@ ENTRY(pgm_check_handler) jnz 1f # -> enabled, can't be a double fault tm __LC_PGM_ILC+3,0x80 # check for per exception jnz .Lpgm_svcper # -> single stepped svc -1: CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC +1: CHECK_STACK __LC_SAVE_AREA_SYNC aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) - j 4f + # CHECK_VMAP_STACK branches to stack_overflow or 4f + CHECK_VMAP_STACK __LC_SAVE_AREA_SYNC,4f 2: UPDATE_VTIME %r14,%r15,__LC_SYNC_ENTER_TIMER BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP lg %r15,__LC_KERNEL_STACK @@ -1136,7 +1157,8 @@ ENTRY(mcck_int_handler) jnz 4f TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_IA_VALID jno .Lmcck_panic -4: SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER +4: ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off + SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER .Lmcck_skip: lghi %r14,__LC_GPREGS_SAVE_AREA+64 stmg %r0,%r7,__PT_R0(%r11) @@ -1163,7 +1185,6 @@ ENTRY(mcck_int_handler) xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) la %r11,STACK_FRAME_OVERHEAD(%r1) lgr %r15,%r1 - ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off TSTMSK __LC_CPU_FLAGS,_CIF_MCCK_PENDING jno .Lmcck_return TRACE_IRQS_OFF @@ -1182,7 +1203,7 @@ ENTRY(mcck_int_handler) lpswe __LC_RETURN_MCCK_PSW .Lmcck_panic: - lg %r15,__LC_PANIC_STACK + lg %r15,__LC_NODAT_STACK la %r11,STACK_FRAME_OVERHEAD(%r15) j .Lmcck_skip @@ -1193,12 +1214,10 @@ ENTRY(restart_int_handler) ALTERNATIVE "", ".insn s,0xb2800000,_LPP_OFFSET", 40 stg %r15,__LC_SAVE_AREA_RESTART lg %r15,__LC_RESTART_STACK - aghi %r15,-__PT_SIZE # create pt_regs on stack - xc 0(__PT_SIZE,%r15),0(%r15) - stmg %r0,%r14,__PT_R0(%r15) - mvc __PT_R15(8,%r15),__LC_SAVE_AREA_RESTART - mvc __PT_PSW(16,%r15),__LC_RST_OLD_PSW # store restart old psw - aghi %r15,-STACK_FRAME_OVERHEAD # create stack frame on stack + xc STACK_FRAME_OVERHEAD(__PT_SIZE,%r15),STACK_FRAME_OVERHEAD(%r15) + stmg %r0,%r14,STACK_FRAME_OVERHEAD+__PT_R0(%r15) + mvc STACK_FRAME_OVERHEAD+__PT_R15(8,%r15),__LC_SAVE_AREA_RESTART + mvc STACK_FRAME_OVERHEAD+__PT_PSW(16,%r15),__LC_RST_OLD_PSW xc 0(STACK_FRAME_OVERHEAD,%r15),0(%r15) lg %r1,__LC_RESTART_FN # load fn, parm & source cpu lg %r2,__LC_RESTART_DATA @@ -1216,14 +1235,14 @@ ENTRY(restart_int_handler) .section .kprobes.text, "ax" -#ifdef CONFIG_CHECK_STACK +#if defined(CONFIG_CHECK_STACK) || defined(CONFIG_VMAP_STACK) /* * The synchronous or the asynchronous stack overflowed. We are dead. * No need to properly save the registers, we are going to panic anyway. * Setup a pt_regs so that show_trace can provide a good call trace. */ stack_overflow: - lg %r15,__LC_PANIC_STACK # change to panic stack + lg %r15,__LC_NODAT_STACK # change to panic stack la %r11,STACK_FRAME_OVERHEAD(%r15) stmg %r0,%r7,__PT_R0(%r11) stmg %r8,%r9,__PT_PSW(%r11) diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h index 472fa2f1a4a5..c3816ae108b0 100644 --- a/arch/s390/kernel/entry.h +++ b/arch/s390/kernel/entry.h @@ -86,4 +86,7 @@ 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); + #endif /* _ENTRY_H */ diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 6d14ad42ba88..57bba24b1c27 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -14,6 +14,7 @@ #include <asm/asm-offsets.h> #include <asm/thread_info.h> #include <asm/page.h> +#include <asm/ptrace.h> __HEAD ENTRY(startup_continue) @@ -35,10 +36,7 @@ ENTRY(startup_continue) # larl %r14,init_task stg %r14,__LC_CURRENT - larl %r15,init_thread_union - aghi %r15,1<<(PAGE_SHIFT+THREAD_SIZE_ORDER) # init_task_union + THREAD_SIZE - stg %r15,__LC_KERNEL_STACK # set end of kernel stack - aghi %r15,-160 + larl %r15,init_thread_union+THREAD_SIZE-STACK_FRAME_OVERHEAD # # Early setup functions that may not rely on an initialized bss section, # like moving the initrd. Returns with an initialized bss section. diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 4296d7e61fb6..18a5d6317acc 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -29,6 +29,8 @@ #include <asm/checksum.h> #include <asm/debug.h> #include <asm/os_info.h> +#include <asm/sections.h> +#include <asm/boot_data.h> #include "entry.h" #define IPL_PARM_BLOCK_VERSION 0 @@ -117,6 +119,9 @@ static char *dump_type_str(enum dump_type type) } } +struct ipl_parameter_block __bootdata(early_ipl_block); +int __bootdata(early_ipl_block_valid); + static int ipl_block_valid; static struct ipl_parameter_block ipl_block; @@ -151,6 +156,8 @@ static inline int __diag308(unsigned long subcode, void *addr) int diag308(unsigned long subcode, void *addr) { + if (IS_ENABLED(CONFIG_KASAN)) + __arch_local_irq_stosm(0x04); /* enable DAT */ diag_stat_inc(DIAG_STAT_X308); return __diag308(subcode, addr); } @@ -262,115 +269,16 @@ static ssize_t ipl_type_show(struct kobject *kobj, struct kobj_attribute *attr, static struct kobj_attribute sys_ipl_type_attr = __ATTR_RO(ipl_type); -/* VM IPL PARM routines */ -static size_t reipl_get_ascii_vmparm(char *dest, size_t size, - const struct ipl_parameter_block *ipb) -{ - int i; - size_t len; - char has_lowercase = 0; - - len = 0; - if ((ipb->ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID) && - (ipb->ipl_info.ccw.vm_parm_len > 0)) { - - len = min_t(size_t, size - 1, ipb->ipl_info.ccw.vm_parm_len); - memcpy(dest, ipb->ipl_info.ccw.vm_parm, len); - /* If at least one character is lowercase, we assume mixed - * case; otherwise we convert everything to lowercase. - */ - for (i = 0; i < len; i++) - if ((dest[i] > 0x80 && dest[i] < 0x8a) || /* a-i */ - (dest[i] > 0x90 && dest[i] < 0x9a) || /* j-r */ - (dest[i] > 0xa1 && dest[i] < 0xaa)) { /* s-z */ - has_lowercase = 1; - break; - } - if (!has_lowercase) - EBC_TOLOWER(dest, len); - EBCASC(dest, len); - } - dest[len] = 0; - - return len; -} - -size_t append_ipl_vmparm(char *dest, size_t size) -{ - size_t rc; - - rc = 0; - if (ipl_block_valid && ipl_block.hdr.pbt == DIAG308_IPL_TYPE_CCW) - rc = reipl_get_ascii_vmparm(dest, size, &ipl_block); - else - dest[0] = 0; - return rc; -} - static ssize_t ipl_vm_parm_show(struct kobject *kobj, struct kobj_attribute *attr, char *page) { char parm[DIAG308_VMPARM_SIZE + 1] = {}; - append_ipl_vmparm(parm, sizeof(parm)); + if (ipl_block_valid && (ipl_block.hdr.pbt == DIAG308_IPL_TYPE_CCW)) + ipl_block_get_ascii_vmparm(parm, sizeof(parm), &ipl_block); return sprintf(page, "%s\n", parm); } -static size_t scpdata_length(const char* buf, size_t count) -{ - while (count) { - if (buf[count - 1] != '\0' && buf[count - 1] != ' ') - break; - count--; - } - return count; -} - -static size_t reipl_append_ascii_scpdata(char *dest, size_t size, - const struct ipl_parameter_block *ipb) -{ - size_t count; - size_t i; - int has_lowercase; - - count = min(size - 1, scpdata_length(ipb->ipl_info.fcp.scp_data, - ipb->ipl_info.fcp.scp_data_len)); - if (!count) - goto out; - - has_lowercase = 0; - for (i = 0; i < count; i++) { - if (!isascii(ipb->ipl_info.fcp.scp_data[i])) { - count = 0; - goto out; - } - if (!has_lowercase && islower(ipb->ipl_info.fcp.scp_data[i])) - has_lowercase = 1; - } - - if (has_lowercase) - memcpy(dest, ipb->ipl_info.fcp.scp_data, count); - else - for (i = 0; i < count; i++) - dest[i] = tolower(ipb->ipl_info.fcp.scp_data[i]); -out: - dest[count] = '\0'; - return count; -} - -size_t append_ipl_scpdata(char *dest, size_t len) -{ - size_t rc; - - rc = 0; - if (ipl_block_valid && ipl_block.hdr.pbt == DIAG308_IPL_TYPE_FCP) - rc = reipl_append_ascii_scpdata(dest, len, &ipl_block); - else - dest[0] = 0; - return rc; -} - - static struct kobj_attribute sys_ipl_vm_parm_attr = __ATTR(parm, S_IRUGO, ipl_vm_parm_show, NULL); @@ -564,7 +472,7 @@ static ssize_t reipl_generic_vmparm_show(struct ipl_parameter_block *ipb, { char vmparm[DIAG308_VMPARM_SIZE + 1] = {}; - reipl_get_ascii_vmparm(vmparm, sizeof(vmparm), ipb); + ipl_block_get_ascii_vmparm(vmparm, sizeof(vmparm), ipb); return sprintf(page, "%s\n", vmparm); } @@ -1769,11 +1677,10 @@ void __init setup_ipl(void) void __init ipl_store_parameters(void) { - int rc; - - rc = diag308(DIAG308_STORE, &ipl_block); - if (rc == DIAG308_RC_OK && ipl_block.hdr.version <= IPL_MAX_SUPPORTED_VERSION) + if (early_ipl_block_valid) { + memcpy(&ipl_block, &early_ipl_block, sizeof(ipl_block)); ipl_block_valid = 1; + } } void s390_reset_system(void) diff --git a/arch/s390/kernel/ipl_vmparm.c b/arch/s390/kernel/ipl_vmparm.c new file mode 100644 index 000000000000..411838c0a0af --- /dev/null +++ b/arch/s390/kernel/ipl_vmparm.c @@ -0,0 +1,36 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <asm/ebcdic.h> +#include <asm/ipl.h> + +/* VM IPL PARM routines */ +size_t ipl_block_get_ascii_vmparm(char *dest, size_t size, + const struct ipl_parameter_block *ipb) +{ + int i; + size_t len; + char has_lowercase = 0; + + len = 0; + if ((ipb->ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID) && + (ipb->ipl_info.ccw.vm_parm_len > 0)) { + + len = min_t(size_t, size - 1, ipb->ipl_info.ccw.vm_parm_len); + memcpy(dest, ipb->ipl_info.ccw.vm_parm, len); + /* If at least one character is lowercase, we assume mixed + * case; otherwise we convert everything to lowercase. + */ + for (i = 0; i < len; i++) + if ((dest[i] > 0x80 && dest[i] < 0x8a) || /* a-i */ + (dest[i] > 0x90 && dest[i] < 0x9a) || /* j-r */ + (dest[i] > 0xa1 && dest[i] < 0xaa)) { /* s-z */ + has_lowercase = 1; + break; + } + if (!has_lowercase) + EBC_TOLOWER(dest, len); + EBCASC(dest, len); + } + dest[len] = 0; + + return len; +} diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c index 3d17c41074ca..0e8d68bac82c 100644 --- a/arch/s390/kernel/irq.c +++ b/arch/s390/kernel/irq.c @@ -172,15 +172,7 @@ void do_softirq_own_stack(void) /* Check against async. stack address range. */ new = S390_lowcore.async_stack; if (((new - old) >> (PAGE_SHIFT + THREAD_SIZE_ORDER)) != 0) { - /* Need to switch to the async. stack. */ - new -= STACK_FRAME_OVERHEAD; - ((struct stack_frame *) new)->back_chain = old; - asm volatile(" la 15,0(%0)\n" - " brasl 14,__do_softirq\n" - " la 15,0(%1)\n" - : : "a" (new), "a" (old) - : "0", "1", "2", "3", "4", "5", "14", - "cc", "memory" ); + CALL_ON_STACK(__do_softirq, new, 0); } else { /* We are already on the async stack. */ __do_softirq(); diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c index 43f8430fb67d..50a1798604a8 100644 --- a/arch/s390/kernel/jump_label.c +++ b/arch/s390/kernel/jump_label.c @@ -33,13 +33,13 @@ static void jump_label_make_branch(struct jump_entry *entry, struct insn *insn) { /* brcl 15,offset */ insn->opcode = 0xc0f4; - insn->offset = (entry->target - entry->code) >> 1; + insn->offset = (jump_entry_target(entry) - jump_entry_code(entry)) >> 1; } static void jump_label_bug(struct jump_entry *entry, struct insn *expected, struct insn *new) { - unsigned char *ipc = (unsigned char *)entry->code; + unsigned char *ipc = (unsigned char *)jump_entry_code(entry); unsigned char *ipe = (unsigned char *)expected; unsigned char *ipn = (unsigned char *)new; @@ -59,6 +59,7 @@ static void __jump_label_transform(struct jump_entry *entry, enum jump_label_type type, int init) { + void *code = (void *)jump_entry_code(entry); struct insn old, new; if (type == JUMP_LABEL_JMP) { @@ -69,13 +70,13 @@ static void __jump_label_transform(struct jump_entry *entry, jump_label_make_nop(entry, &new); } if (init) { - if (memcmp((void *)entry->code, &orignop, sizeof(orignop))) + if (memcmp(code, &orignop, sizeof(orignop))) jump_label_bug(entry, &orignop, &new); } else { - if (memcmp((void *)entry->code, &old, sizeof(old))) + if (memcmp(code, &old, sizeof(old))) jump_label_bug(entry, &old, &new); } - s390_kernel_write((void *)entry->code, &new, sizeof(new)); + s390_kernel_write(code, &new, sizeof(new)); } static int __sm_arch_jump_label_transform(void *data) diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index b7020e721ae3..cb582649aba6 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -142,18 +142,27 @@ static noinline void __machine_kdump(void *image) } #endif +static unsigned long do_start_kdump(unsigned long addr) +{ + struct kimage *image = (struct kimage *) addr; + int (*start_kdump)(int) = (void *)image->start; + int rc; + + __arch_local_irq_stnsm(0xfb); /* disable DAT */ + rc = start_kdump(0); + __arch_local_irq_stosm(0x04); /* enable DAT */ + return rc; +} + /* * Check if kdump checksums are valid: We call purgatory with parameter "0" */ static bool kdump_csum_valid(struct kimage *image) { #ifdef CONFIG_CRASH_DUMP - int (*start_kdump)(int) = (void *)image->start; int rc; - __arch_local_irq_stnsm(0xfb); /* disable DAT */ - rc = start_kdump(0); - __arch_local_irq_stosm(0x04); /* enable DAT */ + rc = CALL_ON_STACK(do_start_kdump, S390_lowcore.nodat_stack, 1, image); return rc == 0; #else return false; diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c index d298d3cb46d0..31889db609e9 100644 --- a/arch/s390/kernel/module.c +++ b/arch/s390/kernel/module.c @@ -16,6 +16,7 @@ #include <linux/fs.h> #include <linux/string.h> #include <linux/kernel.h> +#include <linux/kasan.h> #include <linux/moduleloader.h> #include <linux/bug.h> #include <asm/alternative.h> @@ -32,12 +33,18 @@ void *module_alloc(unsigned long size) { + void *p; + if (PAGE_ALIGN(size) > MODULES_LEN) return NULL; - return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, - GFP_KERNEL, PAGE_KERNEL_EXEC, - 0, NUMA_NO_NODE, - __builtin_return_address(0)); + p = __vmalloc_node_range(size, MODULE_ALIGN, MODULES_VADDR, MODULES_END, + GFP_KERNEL, PAGE_KERNEL_EXEC, 0, NUMA_NO_NODE, + __builtin_return_address(0)); + if (p && (kasan_module_alloc(p, size) < 0)) { + vfree(p); + return NULL; + } + return p; } void module_arch_freeing_init(struct module *mod) diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 5c53e977be62..7bf604ff50a1 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -2045,14 +2045,17 @@ static int __init init_cpum_sampling_pmu(void) } sfdbg = debug_register(KMSG_COMPONENT, 2, 1, 80); - if (!sfdbg) + if (!sfdbg) { pr_err("Registering for s390dbf failed\n"); + return -ENOMEM; + } debug_register_view(sfdbg, &debug_sprintf_view); err = register_external_irq(EXT_IRQ_MEASURE_ALERT, cpumf_measurement_alert); if (err) { pr_cpumsf_err(RS_INIT_FAILURE_ALRT); + debug_unregister(sfdbg); goto out; } @@ -2061,6 +2064,7 @@ static int __init init_cpum_sampling_pmu(void) pr_cpumsf_err(RS_INIT_FAILURE_PERF); unregister_external_irq(EXT_IRQ_MEASURE_ALERT, cpumf_measurement_alert); + debug_unregister(sfdbg); goto out; } diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index c637c12f9e37..a2e952b66248 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -49,6 +49,7 @@ #include <linux/crash_dump.h> #include <linux/memory.h> #include <linux/compat.h> +#include <linux/start_kernel.h> #include <asm/ipl.h> #include <asm/facility.h> @@ -69,6 +70,7 @@ #include <asm/numa.h> #include <asm/alternative.h> #include <asm/nospec-branch.h> +#include <asm/mem_detect.h> #include "entry.h" /* @@ -88,9 +90,11 @@ char elf_platform[ELF_PLATFORM_SIZE]; unsigned long int_hwcap = 0; -int __initdata memory_end_set; -unsigned long __initdata memory_end; -unsigned long __initdata max_physmem_end; +int __bootdata(noexec_disabled); +int __bootdata(memory_end_set); +unsigned long __bootdata(memory_end); +unsigned long __bootdata(max_physmem_end); +struct mem_detect_info __bootdata(mem_detect); unsigned long VMALLOC_START; EXPORT_SYMBOL(VMALLOC_START); @@ -283,15 +287,6 @@ void machine_power_off(void) void (*pm_power_off)(void) = machine_power_off; EXPORT_SYMBOL_GPL(pm_power_off); -static int __init early_parse_mem(char *p) -{ - memory_end = memparse(p, &p); - memory_end &= PAGE_MASK; - memory_end_set = 1; - return 0; -} -early_param("mem", early_parse_mem); - static int __init parse_vmalloc(char *arg) { if (!arg) @@ -303,6 +298,78 @@ early_param("vmalloc", parse_vmalloc); void *restart_stack __section(.data); +unsigned long stack_alloc(void) +{ +#ifdef CONFIG_VMAP_STACK + return (unsigned long) + __vmalloc_node_range(THREAD_SIZE, THREAD_SIZE, + VMALLOC_START, VMALLOC_END, + THREADINFO_GFP, + PAGE_KERNEL, 0, NUMA_NO_NODE, + __builtin_return_address(0)); +#else + return __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER); +#endif +} + +void stack_free(unsigned long stack) +{ +#ifdef CONFIG_VMAP_STACK + vfree((void *) stack); +#else + free_pages(stack, THREAD_SIZE_ORDER); +#endif +} + +int __init arch_early_irq_init(void) +{ + unsigned long stack; + + stack = __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER); + if (!stack) + panic("Couldn't allocate async stack"); + S390_lowcore.async_stack = stack + STACK_INIT_OFFSET; + return 0; +} + +static int __init async_stack_realloc(void) +{ + unsigned long old, new; + + old = S390_lowcore.async_stack - STACK_INIT_OFFSET; + new = stack_alloc(); + if (!new) + panic("Couldn't allocate async stack"); + S390_lowcore.async_stack = new + STACK_INIT_OFFSET; + free_pages(old, THREAD_SIZE_ORDER); + return 0; +} +early_initcall(async_stack_realloc); + +void __init arch_call_rest_init(void) +{ + struct stack_frame *frame; + unsigned long stack; + + stack = stack_alloc(); + if (!stack) + panic("Couldn't allocate kernel stack"); + current->stack = (void *) stack; +#ifdef CONFIG_VMAP_STACK + current->stack_vm_area = (void *) stack; +#endif + set_task_stack_end_magic(current); + stack += STACK_INIT_OFFSET; + S390_lowcore.kernel_stack = stack; + frame = (struct stack_frame *) stack; + memset(frame, 0, sizeof(*frame)); + /* Branch to rest_init on the new stack, never returns */ + asm volatile( + " la 15,0(%[_frame])\n" + " jg rest_init\n" + : : [_frame] "a" (frame)); +} + static void __init setup_lowcore(void) { struct lowcore *lc; @@ -329,14 +396,8 @@ static void __init setup_lowcore(void) PSW_MASK_DAT | PSW_MASK_MCHECK; lc->io_new_psw.addr = (unsigned long) io_int_handler; lc->clock_comparator = clock_comparator_max; - lc->kernel_stack = ((unsigned long) &init_thread_union) + lc->nodat_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); - lc->async_stack = (unsigned long) - memblock_virt_alloc(ASYNC_SIZE, ASYNC_SIZE) - + ASYNC_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); - lc->panic_stack = (unsigned long) - memblock_virt_alloc(PAGE_SIZE, PAGE_SIZE) - + PAGE_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); lc->current_task = (unsigned long)&init_task; lc->lpp = LPP_MAGIC; lc->machine_flags = S390_lowcore.machine_flags; @@ -357,8 +418,12 @@ static void __init setup_lowcore(void) lc->last_update_timer = S390_lowcore.last_update_timer; lc->last_update_clock = S390_lowcore.last_update_clock; - restart_stack = memblock_virt_alloc(ASYNC_SIZE, ASYNC_SIZE); - restart_stack += ASYNC_SIZE; + /* + * Allocate the global restart stack which is the same for + * all CPUs in cast *one* of them does a PSW restart. + */ + restart_stack = memblock_virt_alloc(THREAD_SIZE, THREAD_SIZE); + restart_stack += STACK_INIT_OFFSET; /* * Set up PSW restart to call ipl.c:do_restart(). Copy the relevant @@ -467,19 +532,26 @@ static void __init setup_memory_end(void) { unsigned long vmax, vmalloc_size, tmp; - /* Choose kernel address space layout: 2, 3, or 4 levels. */ + /* Choose kernel address space layout: 3 or 4 levels. */ vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN; - tmp = (memory_end ?: max_physmem_end) / 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 */ + if (IS_ENABLED(CONFIG_KASAN)) { + vmax = IS_ENABLED(CONFIG_KASAN_S390_4_LEVEL_PAGING) + ? _REGION1_SIZE + : _REGION2_SIZE; + } else { + tmp = (memory_end ?: max_physmem_end) / 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; MODULES_VADDR = MODULES_END - MODULES_LEN; VMALLOC_END = MODULES_VADDR; - VMALLOC_START = vmax - vmalloc_size; + VMALLOC_START = VMALLOC_END - vmalloc_size; /* Split remaining virtual space between 1:1 mapping & vmemmap array */ tmp = VMALLOC_START / (PAGE_SIZE + sizeof(struct page)); @@ -491,7 +563,12 @@ static void __init setup_memory_end(void) vmemmap = (struct page *) tmp; /* Take care that memory_end is set and <= vmemmap */ - memory_end = min(memory_end ?: max_physmem_end, tmp); + memory_end = min(memory_end ?: max_physmem_end, (unsigned long)vmemmap); +#ifdef CONFIG_KASAN + /* fit in kasan shadow memory region between 1:1 and vmemmap */ + memory_end = min(memory_end, 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); @@ -532,17 +609,8 @@ static struct notifier_block kdump_mem_nb = { */ static void reserve_memory_end(void) { -#ifdef CONFIG_CRASH_DUMP - if (ipl_info.type == IPL_TYPE_FCP_DUMP && - !OLDMEM_BASE && sclp.hsa_size) { - memory_end = sclp.hsa_size; - memory_end &= PAGE_MASK; - memory_end_set = 1; - } -#endif - if (!memory_end_set) - return; - memblock_reserve(memory_end, ULONG_MAX); + if (memory_end_set) + memblock_reserve(memory_end, ULONG_MAX); } /* @@ -649,6 +717,62 @@ static void __init reserve_initrd(void) #endif } +static void __init reserve_mem_detect_info(void) +{ + unsigned long start, size; + + get_mem_detect_reserved(&start, &size); + if (size) + memblock_reserve(start, size); +} + +static void __init free_mem_detect_info(void) +{ + unsigned long start, size; + + get_mem_detect_reserved(&start, &size); + if (size) + memblock_free(start, size); +} + +static void __init memblock_physmem_add(phys_addr_t start, phys_addr_t size) +{ + memblock_dbg("memblock_physmem_add: [%#016llx-%#016llx]\n", + start, start + size - 1); + memblock_add_range(&memblock.memory, start, size, 0, 0); + memblock_add_range(&memblock.physmem, start, size, 0, 0); +} + +static const char * __init get_mem_info_source(void) +{ + switch (mem_detect.info_source) { + case MEM_DETECT_SCLP_STOR_INFO: + return "sclp storage info"; + case MEM_DETECT_DIAG260: + return "diag260"; + case MEM_DETECT_SCLP_READ_INFO: + return "sclp read info"; + case MEM_DETECT_BIN_SEARCH: + return "binary search"; + } + return "none"; +} + +static void __init memblock_add_mem_detect_info(void) +{ + unsigned long start, end; + int i; + + memblock_dbg("physmem info source: %s (%hhd)\n", + get_mem_info_source(), mem_detect.info_source); + /* keep memblock lists close to the kernel */ + memblock_set_bottom_up(true); + for_each_mem_detect_block(i, &start, &end) + memblock_physmem_add(start, end - start); + memblock_set_bottom_up(false); + memblock_dump_all(); +} + /* * Check for initrd being in usable memory */ @@ -913,11 +1037,13 @@ void __init setup_arch(char **cmdline_p) reserve_oldmem(); reserve_kernel(); reserve_initrd(); + reserve_mem_detect_info(); memblock_allow_resize(); /* Get information about *all* installed memory */ - detect_memory_memblock(); + memblock_add_mem_detect_info(); + free_mem_detect_info(); remove_oldmem(); /* diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 2f8f7d7dd9a8..1b3188f57b58 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -186,36 +186,34 @@ static void pcpu_ec_call(struct pcpu *pcpu, int ec_bit) pcpu_sigp_retry(pcpu, order, 0); } -#define ASYNC_FRAME_OFFSET (ASYNC_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE) -#define PANIC_FRAME_OFFSET (PAGE_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE) - static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) { - unsigned long async_stack, panic_stack; + unsigned long async_stack, nodat_stack; struct lowcore *lc; if (pcpu != &pcpu_devices[0]) { pcpu->lowcore = (struct lowcore *) __get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER); - async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER); - panic_stack = __get_free_page(GFP_KERNEL); - if (!pcpu->lowcore || !panic_stack || !async_stack) + nodat_stack = __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER); + if (!pcpu->lowcore || !nodat_stack) goto out; } else { - async_stack = pcpu->lowcore->async_stack - ASYNC_FRAME_OFFSET; - panic_stack = pcpu->lowcore->panic_stack - PANIC_FRAME_OFFSET; + nodat_stack = pcpu->lowcore->nodat_stack - STACK_INIT_OFFSET; } + async_stack = stack_alloc(); + if (!async_stack) + goto out; lc = pcpu->lowcore; memcpy(lc, &S390_lowcore, 512); memset((char *) lc + 512, 0, sizeof(*lc) - 512); - lc->async_stack = async_stack + ASYNC_FRAME_OFFSET; - lc->panic_stack = panic_stack + PANIC_FRAME_OFFSET; + lc->async_stack = async_stack + STACK_INIT_OFFSET; + lc->nodat_stack = nodat_stack + STACK_INIT_OFFSET; lc->cpu_nr = cpu; lc->spinlock_lockval = arch_spin_lockval(cpu); lc->spinlock_index = 0; lc->br_r1_trampoline = 0x07f1; /* br %r1 */ if (nmi_alloc_per_cpu(lc)) - goto out; + goto out_async; if (vdso_alloc_per_cpu(lc)) goto out_mcesa; lowcore_ptr[cpu] = lc; @@ -224,10 +222,11 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) out_mcesa: nmi_free_per_cpu(lc); +out_async: + stack_free(async_stack); out: if (pcpu != &pcpu_devices[0]) { - free_page(panic_stack); - free_pages(async_stack, ASYNC_ORDER); + free_pages(nodat_stack, THREAD_SIZE_ORDER); free_pages((unsigned long) pcpu->lowcore, LC_ORDER); } return -ENOMEM; @@ -237,15 +236,21 @@ out: static void pcpu_free_lowcore(struct pcpu *pcpu) { + unsigned long async_stack, nodat_stack, lowcore; + + nodat_stack = pcpu->lowcore->nodat_stack - STACK_INIT_OFFSET; + async_stack = pcpu->lowcore->async_stack - STACK_INIT_OFFSET; + lowcore = (unsigned long) pcpu->lowcore; + 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]) return; - free_page(pcpu->lowcore->panic_stack-PANIC_FRAME_OFFSET); - free_pages(pcpu->lowcore->async_stack-ASYNC_FRAME_OFFSET, ASYNC_ORDER); - free_pages((unsigned long) pcpu->lowcore, LC_ORDER); + free_pages(nodat_stack, THREAD_SIZE_ORDER); + free_pages(lowcore, LC_ORDER); } #endif /* CONFIG_HOTPLUG_CPU */ @@ -293,7 +298,7 @@ static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data) { struct lowcore *lc = pcpu->lowcore; - lc->restart_stack = lc->kernel_stack; + lc->restart_stack = lc->nodat_stack; lc->restart_fn = (unsigned long) func; lc->restart_data = (unsigned long) data; lc->restart_source = -1UL; @@ -303,15 +308,21 @@ static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data) /* * Call function via PSW restart on pcpu and stop the current cpu. */ -static void pcpu_delegate(struct pcpu *pcpu, void (*func)(void *), - void *data, unsigned long stack) +static void __pcpu_delegate(void (*func)(void*), void *data) +{ + func(data); /* should not return */ +} + +static void __no_sanitize_address pcpu_delegate(struct pcpu *pcpu, + void (*func)(void *), + void *data, unsigned long stack) { struct lowcore *lc = lowcore_ptr[pcpu - pcpu_devices]; unsigned long source_cpu = stap(); - __load_psw_mask(PSW_KERNEL_BITS); + __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT); if (pcpu->address == source_cpu) - func(data); /* should not return */ + CALL_ON_STACK(__pcpu_delegate, stack, 2, func, data); /* Stop target cpu (if func returns this stops the current cpu). */ pcpu_sigp_retry(pcpu, SIGP_STOP, 0); /* Restart func on the target cpu and stop the current cpu. */ @@ -372,8 +383,7 @@ void smp_call_online_cpu(void (*func)(void *), void *data) void smp_call_ipl_cpu(void (*func)(void *), void *data) { pcpu_delegate(&pcpu_devices[0], func, data, - pcpu_devices->lowcore->panic_stack - - PANIC_FRAME_OFFSET + PAGE_SIZE); + pcpu_devices->lowcore->nodat_stack); } int smp_find_processor_id(u16 address) @@ -791,37 +801,42 @@ void __init smp_detect_cpus(void) memblock_free_early((unsigned long)info, sizeof(*info)); } -/* - * Activate a secondary processor. - */ -static void smp_start_secondary(void *cpuvoid) +static void smp_init_secondary(void) { int cpu = smp_processor_id(); S390_lowcore.last_update_clock = get_tod_clock(); - S390_lowcore.restart_stack = (unsigned long) restart_stack; - S390_lowcore.restart_fn = (unsigned long) do_restart; - S390_lowcore.restart_data = 0; - S390_lowcore.restart_source = -1UL; restore_access_regs(S390_lowcore.access_regs_save_area); - __ctl_load(S390_lowcore.cregs_save_area, 0, 15); - __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT); cpu_init(); preempt_disable(); init_cpu_timer(); vtime_init(); pfault_init(); - notify_cpu_starting(cpu); + notify_cpu_starting(smp_processor_id()); if (topology_cpu_dedicated(cpu)) set_cpu_flag(CIF_DEDICATED_CPU); else clear_cpu_flag(CIF_DEDICATED_CPU); - set_cpu_online(cpu, true); + set_cpu_online(smp_processor_id(), true); inc_irq_stat(CPU_RST); local_irq_enable(); cpu_startup_entry(CPUHP_AP_ONLINE_IDLE); } +/* + * Activate a secondary processor. + */ +static void __no_sanitize_address smp_start_secondary(void *cpuvoid) +{ + S390_lowcore.restart_stack = (unsigned long) restart_stack; + S390_lowcore.restart_fn = (unsigned long) do_restart; + S390_lowcore.restart_data = 0; + S390_lowcore.restart_source = -1UL; + __ctl_load(S390_lowcore.cregs_save_area, 0, 15); + __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT); + CALL_ON_STACK(smp_init_secondary, S390_lowcore.kernel_stack, 0); +} + /* Upping and downing of CPUs */ int __cpu_up(unsigned int cpu, struct task_struct *tidle) { diff --git a/arch/s390/kernel/sthyi.c b/arch/s390/kernel/sthyi.c index 0859cde36f75..888cc2f166db 100644 --- a/arch/s390/kernel/sthyi.c +++ b/arch/s390/kernel/sthyi.c @@ -183,17 +183,19 @@ static void fill_hdr(struct sthyi_sctns *sctns) static void fill_stsi_mac(struct sthyi_sctns *sctns, struct sysinfo_1_1_1 *sysinfo) { + sclp_ocf_cpc_name_copy(sctns->mac.infmname); + if (*(u64 *)sctns->mac.infmname != 0) + sctns->mac.infmval1 |= MAC_NAME_VLD; + if (stsi(sysinfo, 1, 1, 1)) return; - sclp_ocf_cpc_name_copy(sctns->mac.infmname); - memcpy(sctns->mac.infmtype, sysinfo->type, sizeof(sctns->mac.infmtype)); memcpy(sctns->mac.infmmanu, sysinfo->manufacturer, sizeof(sctns->mac.infmmanu)); memcpy(sctns->mac.infmpman, sysinfo->plant, sizeof(sctns->mac.infmpman)); memcpy(sctns->mac.infmseq, sysinfo->sequence, sizeof(sctns->mac.infmseq)); - sctns->mac.infmval1 |= MAC_ID_VLD | MAC_NAME_VLD; + sctns->mac.infmval1 |= MAC_ID_VLD; } static void fill_stsi_par(struct sthyi_sctns *sctns, diff --git a/arch/s390/kernel/swsusp.S b/arch/s390/kernel/swsusp.S index a049a7b9d6e8..537f97fde37f 100644 --- a/arch/s390/kernel/swsusp.S +++ b/arch/s390/kernel/swsusp.S @@ -29,10 +29,11 @@ .section .text ENTRY(swsusp_arch_suspend) - stmg %r6,%r15,__SF_GPRS(%r15) + lg %r1,__LC_NODAT_STACK + aghi %r1,-STACK_FRAME_OVERHEAD + stmg %r6,%r15,__SF_GPRS(%r1) + stg %r15,__SF_BACKCHAIN(%r1) lgr %r1,%r15 - aghi %r15,-STACK_FRAME_OVERHEAD - stg %r1,__SF_BACKCHAIN(%r15) /* Store FPU registers */ brasl %r14,save_fpu_regs @@ -197,13 +198,9 @@ pgm_check_entry: brc 2,3b /* busy, try again */ /* Suspend CPU not available -> panic */ - larl %r15,init_thread_union - ahi %r15,1<<(PAGE_SHIFT+THREAD_SIZE_ORDER) + larl %r15,init_thread_union+THREAD_SIZE-STACK_FRAME_OVERHEAD larl %r2,.Lpanic_string - lghi %r1,0 - sam31 - sigp %r1,%r0,SIGP_SET_ARCHITECTURE - brasl %r14,sclp_early_printk + brasl %r14,sclp_early_printk_force larl %r3,.Ldisabled_wait_31 lpsw 0(%r3) 4: diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c index 3031cc6dd0ab..ec31b48a42a5 100644 --- a/arch/s390/kernel/vdso.c +++ b/arch/s390/kernel/vdso.c @@ -56,7 +56,7 @@ static vm_fault_t vdso_fault(const struct vm_special_mapping *sm, vdso_pagelist = vdso64_pagelist; vdso_pages = vdso64_pages; #ifdef CONFIG_COMPAT - if (is_compat_task()) { + if (vma->vm_mm->context.compat_mm) { vdso_pagelist = vdso32_pagelist; vdso_pages = vdso32_pages; } @@ -77,7 +77,7 @@ static int vdso_mremap(const struct vm_special_mapping *sm, vdso_pages = vdso64_pages; #ifdef CONFIG_COMPAT - if (is_compat_task()) + if (vma->vm_mm->context.compat_mm) vdso_pages = vdso32_pages; #endif @@ -224,8 +224,10 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) vdso_pages = vdso64_pages; #ifdef CONFIG_COMPAT - if (is_compat_task()) + if (is_compat_task()) { vdso_pages = vdso32_pages; + mm->context.compat_mm = 1; + } #endif /* * vDSO has a problem and was disabled, just don't "enable" it for diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile index c5c856f320bc..eb8aebea3ea7 100644 --- a/arch/s390/kernel/vdso32/Makefile +++ b/arch/s390/kernel/vdso32/Makefile @@ -28,9 +28,10 @@ obj-y += vdso32_wrapper.o extra-y += vdso32.lds CPPFLAGS_vdso32.lds += -P -C -U$(ARCH) -# Disable gcov profiling and ubsan for VDSO code +# Disable gcov profiling, ubsan and kasan for VDSO code GCOV_PROFILE := n UBSAN_SANITIZE := n +KASAN_SANITIZE := n # Force dependency (incbin is bad) $(obj)/vdso32_wrapper.o : $(obj)/vdso32.so diff --git a/arch/s390/kernel/vdso32/clock_gettime.S b/arch/s390/kernel/vdso32/clock_gettime.S index a9418bf975db..ada5c11a16e5 100644 --- a/arch/s390/kernel/vdso32/clock_gettime.S +++ b/arch/s390/kernel/vdso32/clock_gettime.S @@ -10,6 +10,7 @@ #include <asm/asm-offsets.h> #include <asm/unistd.h> #include <asm/dwarf.h> +#include <asm/ptrace.h> .text .align 4 @@ -18,8 +19,8 @@ __kernel_clock_gettime: CFI_STARTPROC ahi %r15,-16 - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD basr %r5,0 0: al %r5,21f-0b(%r5) /* get &_vdso_data */ chi %r2,__CLOCK_REALTIME_COARSE @@ -72,13 +73,13 @@ __kernel_clock_gettime: st %r1,4(%r3) /* store tp->tv_nsec */ lhi %r2,0 ahi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 /* CLOCK_MONOTONIC_COARSE */ - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD 9: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ tml %r4,0x0001 /* pending update ? loop */ jnz 9b @@ -158,17 +159,17 @@ __kernel_clock_gettime: st %r1,4(%r3) /* store tp->tv_nsec */ lhi %r2,0 ahi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 /* Fallback to system call */ - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD 19: lhi %r1,__NR_clock_gettime svc 0 ahi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 CFI_ENDPROC diff --git a/arch/s390/kernel/vdso32/gettimeofday.S b/arch/s390/kernel/vdso32/gettimeofday.S index 3c0db0fa6ad9..b23063fbc892 100644 --- a/arch/s390/kernel/vdso32/gettimeofday.S +++ b/arch/s390/kernel/vdso32/gettimeofday.S @@ -10,6 +10,7 @@ #include <asm/asm-offsets.h> #include <asm/unistd.h> #include <asm/dwarf.h> +#include <asm/ptrace.h> .text .align 4 @@ -19,7 +20,7 @@ __kernel_gettimeofday: CFI_STARTPROC ahi %r15,-16 CFI_ADJUST_CFA_OFFSET 16 - CFI_VAL_OFFSET 15, -160 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD basr %r5,0 0: al %r5,13f-0b(%r5) /* get &_vdso_data */ 1: ltr %r3,%r3 /* check if tz is NULL */ diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 15b1ceafc4c1..a22b2cf86eec 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -28,9 +28,10 @@ obj-y += vdso64_wrapper.o extra-y += vdso64.lds CPPFLAGS_vdso64.lds += -P -C -U$(ARCH) -# Disable gcov profiling and ubsan for VDSO code +# Disable gcov profiling, ubsan and kasan for VDSO code GCOV_PROFILE := n UBSAN_SANITIZE := n +KASAN_SANITIZE := n # Force dependency (incbin is bad) $(obj)/vdso64_wrapper.o : $(obj)/vdso64.so diff --git a/arch/s390/kernel/vdso64/clock_gettime.S b/arch/s390/kernel/vdso64/clock_gettime.S index fac3ab5ec83a..9d2ee79b90f2 100644 --- a/arch/s390/kernel/vdso64/clock_gettime.S +++ b/arch/s390/kernel/vdso64/clock_gettime.S @@ -10,6 +10,7 @@ #include <asm/asm-offsets.h> #include <asm/unistd.h> #include <asm/dwarf.h> +#include <asm/ptrace.h> .text .align 4 @@ -18,8 +19,8 @@ __kernel_clock_gettime: CFI_STARTPROC aghi %r15,-16 - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD larl %r5,_vdso_data cghi %r2,__CLOCK_REALTIME_COARSE je 4f @@ -56,13 +57,13 @@ __kernel_clock_gettime: stg %r1,8(%r3) /* store tp->tv_nsec */ lghi %r2,0 aghi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 /* CLOCK_MONOTONIC_COARSE */ - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD 3: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ tmll %r4,0x0001 /* pending update ? loop */ jnz 3b @@ -115,13 +116,13 @@ __kernel_clock_gettime: stg %r1,8(%r3) /* store tp->tv_nsec */ lghi %r2,0 aghi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 /* CPUCLOCK_VIRT for this thread */ - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD 9: lghi %r4,0 icm %r0,15,__VDSO_ECTG_OK(%r5) jz 12f @@ -142,17 +143,17 @@ __kernel_clock_gettime: stg %r4,8(%r3) lghi %r2,0 aghi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 /* Fallback to system call */ - CFI_DEF_CFA_OFFSET 176 - CFI_VAL_OFFSET 15, -160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD 12: lghi %r1,__NR_clock_gettime svc 0 aghi %r15,16 - CFI_DEF_CFA_OFFSET 160 + CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD CFI_RESTORE 15 br %r14 CFI_ENDPROC diff --git a/arch/s390/kernel/vdso64/gettimeofday.S b/arch/s390/kernel/vdso64/gettimeofday.S index 6e1f0b421695..aebe10dc7c99 100644 --- a/arch/s390/kernel/vdso64/gettimeofday.S +++ b/arch/s390/kernel/vdso64/gettimeofday.S @@ -10,6 +10,7 @@ #include <asm/asm-offsets.h> #include <asm/unistd.h> #include <asm/dwarf.h> +#include <asm/ptrace.h> .text .align 4 @@ -19,7 +20,7 @@ __kernel_gettimeofday: CFI_STARTPROC aghi %r15,-16 CFI_ADJUST_CFA_OFFSET 16 - CFI_VAL_OFFSET 15, -160 + CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD larl %r5,_vdso_data 0: ltgr %r3,%r3 /* check if tz is NULL */ je 1f diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index b43f8d33a369..21eb7407d51b 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -16,6 +16,7 @@ #define RO_AFTER_INIT_DATA #include <asm-generic/vmlinux.lds.h> +#include <asm/vmlinux.lds.h> OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") OUTPUT_ARCH(s390:64-bit) @@ -64,6 +65,7 @@ SECTIONS __start_ro_after_init = .; .data..ro_after_init : { *(.data..ro_after_init) + JUMP_TABLE_DATA } EXCEPTION_TABLE(16) . = ALIGN(PAGE_SIZE); @@ -134,6 +136,8 @@ SECTIONS __nospec_return_end = . ; } + BOOT_DATA + /* early.c uses stsi, which requires page aligned data. */ . = ALIGN(PAGE_SIZE); INIT_DATA_SECTION(0x100) @@ -146,6 +150,19 @@ SECTIONS _end = . ; + /* + * uncompressed image info used by the decompressor + * it should match struct vmlinux_info + */ + .vmlinux.info 0 : { + QUAD(_stext) /* default_lma */ + QUAD(startup_continue) /* entry */ + QUAD(__bss_start - _stext) /* image_size */ + QUAD(__bss_stop - __bss_start) /* bss_size */ + QUAD(__boot_data_start) /* bootdata_off */ + QUAD(__boot_data_end - __boot_data_start) /* bootdata_size */ + } + /* Debugging sections. */ STABS_DEBUG DWARF_DEBUG diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 91ad4a9425c0..ac5da6b0b862 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -481,7 +481,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) break; case KVM_CAP_S390_HPAGE_1M: r = 0; - if (hpage) + if (hpage && !kvm_is_ucontrol(kvm)) r = 1; break; case KVM_CAP_S390_MEM_OP: @@ -691,11 +691,13 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) mutex_lock(&kvm->lock); if (kvm->created_vcpus) r = -EBUSY; - else if (!hpage || kvm->arch.use_cmma) + else if (!hpage || kvm->arch.use_cmma || kvm_is_ucontrol(kvm)) r = -EINVAL; else { r = 0; + down_write(&kvm->mm->mmap_sem); kvm->mm->context.allow_gmap_hpage_1m = 1; + up_write(&kvm->mm->mmap_sem); /* * We might have to create fake 4k page * tables. To avoid that the hardware works on diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index d68f10441a16..8679bd74d337 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -280,9 +280,11 @@ retry: goto retry; } } - if (rc) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); up_read(¤t->mm->mmap_sem); + if (rc == -EFAULT) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + if (rc < 0) + return rc; vcpu->run->s.regs.gprs[reg1] &= ~0xff; vcpu->run->s.regs.gprs[reg1] |= key; return 0; @@ -324,9 +326,11 @@ retry: goto retry; } } - if (rc < 0) - return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); up_read(¤t->mm->mmap_sem); + if (rc == -EFAULT) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + if (rc < 0) + return rc; kvm_s390_set_psw_cc(vcpu, rc); return 0; } @@ -390,12 +394,12 @@ static int handle_sske(struct kvm_vcpu *vcpu) FAULT_FLAG_WRITE, &unlocked); rc = !rc ? -EAGAIN : rc; } + up_read(¤t->mm->mmap_sem); if (rc == -EFAULT) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); - - up_read(¤t->mm->mmap_sem); - if (rc >= 0) - start += PAGE_SIZE; + if (rc < 0) + return rc; + start += PAGE_SIZE; } if (m3 & (SSKE_MC | SSKE_MR)) { @@ -1002,13 +1006,15 @@ static int handle_pfmf(struct kvm_vcpu *vcpu) FAULT_FLAG_WRITE, &unlocked); rc = !rc ? -EAGAIN : rc; } + up_read(¤t->mm->mmap_sem); if (rc == -EFAULT) return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); - - up_read(¤t->mm->mmap_sem); - if (rc >= 0) - start += PAGE_SIZE; + if (rc == -EAGAIN) + continue; + if (rc < 0) + return rc; } + start += PAGE_SIZE; } if (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) { if (psw_bits(vcpu->arch.sie_block->gpsw).eaba == PSW_BITS_AMODE_64BIT) { diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 63844b95c22c..a2b28cd1e3fe 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -173,7 +173,8 @@ static int shadow_crycb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) return set_validity_icpt(scb_s, 0x0039U); /* copy only the wrapping keys */ - if (read_guest_real(vcpu, crycb_addr + 72, &vsie_page->crycb, 56)) + if (read_guest_real(vcpu, crycb_addr + 72, + vsie_page->crycb.dea_wrapping_key_mask, 56)) return set_validity_icpt(scb_s, 0x0035U); scb_s->ecb3 |= ecb3_flags; diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile index 57ab40188d4b..5418d10dc2a8 100644 --- a/arch/s390/lib/Makefile +++ b/arch/s390/lib/Makefile @@ -9,5 +9,9 @@ lib-$(CONFIG_SMP) += spinlock.o lib-$(CONFIG_KPROBES) += probes.o lib-$(CONFIG_UPROBES) += probes.o +# Instrumenting memory accesses to __user data (in different address space) +# produce false positives +KASAN_SANITIZE_uaccess.o := n + chkbss := mem.o include $(srctree)/arch/s390/scripts/Makefile.chkbss diff --git a/arch/s390/lib/mem.S b/arch/s390/lib/mem.S index 40c4d59c926e..53008da05190 100644 --- a/arch/s390/lib/mem.S +++ b/arch/s390/lib/mem.S @@ -14,7 +14,8 @@ /* * void *memmove(void *dest, const void *src, size_t n) */ -ENTRY(memmove) +WEAK(memmove) +ENTRY(__memmove) ltgr %r4,%r4 lgr %r1,%r2 jz .Lmemmove_exit @@ -47,6 +48,7 @@ ENTRY(memmove) BR_EX %r14 .Lmemmove_mvc: mvc 0(1,%r1),0(%r3) +ENDPROC(__memmove) EXPORT_SYMBOL(memmove) /* @@ -64,7 +66,8 @@ EXPORT_SYMBOL(memmove) * return __builtin_memset(s, c, n); * } */ -ENTRY(memset) +WEAK(memset) +ENTRY(__memset) ltgr %r4,%r4 jz .Lmemset_exit ltgr %r3,%r3 @@ -108,6 +111,7 @@ ENTRY(memset) xc 0(1,%r1),0(%r1) .Lmemset_mvc: mvc 1(1,%r1),0(%r1) +ENDPROC(__memset) EXPORT_SYMBOL(memset) /* @@ -115,7 +119,8 @@ EXPORT_SYMBOL(memset) * * void *memcpy(void *dest, const void *src, size_t n) */ -ENTRY(memcpy) +WEAK(memcpy) +ENTRY(__memcpy) ltgr %r4,%r4 jz .Lmemcpy_exit aghi %r4,-1 @@ -136,6 +141,7 @@ ENTRY(memcpy) j .Lmemcpy_remainder .Lmemcpy_mvc: mvc 0(1,%r1),0(%r3) +ENDPROC(__memcpy) EXPORT_SYMBOL(memcpy) /* diff --git a/arch/s390/mm/Makefile b/arch/s390/mm/Makefile index 33fe418506bc..f5880bfd1b0c 100644 --- a/arch/s390/mm/Makefile +++ b/arch/s390/mm/Makefile @@ -4,10 +4,12 @@ # obj-y := init.o fault.o extmem.o mmap.o vmem.o maccess.o -obj-y += page-states.o gup.o pageattr.o mem_detect.o -obj-y += pgtable.o pgalloc.o +obj-y += page-states.o gup.o pageattr.o pgtable.o pgalloc.o obj-$(CONFIG_CMM) += cmm.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_S390_PTDUMP) += dump_pagetables.o obj-$(CONFIG_PGSTE) += gmap.o + +KASAN_SANITIZE_kasan_init.o := n +obj-$(CONFIG_KASAN) += kasan_init.o diff --git a/arch/s390/mm/dump_pagetables.c b/arch/s390/mm/dump_pagetables.c index 7cdea2ec51e9..363f6470d742 100644 --- a/arch/s390/mm/dump_pagetables.c +++ b/arch/s390/mm/dump_pagetables.c @@ -3,6 +3,8 @@ #include <linux/debugfs.h> #include <linux/sched.h> #include <linux/mm.h> +#include <linux/kasan.h> +#include <asm/kasan.h> #include <asm/sections.h> #include <asm/pgtable.h> @@ -17,18 +19,26 @@ enum address_markers_idx { IDENTITY_NR = 0, KERNEL_START_NR, KERNEL_END_NR, +#ifdef CONFIG_KASAN + KASAN_SHADOW_START_NR, + KASAN_SHADOW_END_NR, +#endif VMEMMAP_NR, VMALLOC_NR, MODULES_NR, }; static struct addr_marker address_markers[] = { - [IDENTITY_NR] = {0, "Identity Mapping"}, - [KERNEL_START_NR] = {(unsigned long)_stext, "Kernel Image Start"}, - [KERNEL_END_NR] = {(unsigned long)_end, "Kernel Image End"}, - [VMEMMAP_NR] = {0, "vmemmap Area"}, - [VMALLOC_NR] = {0, "vmalloc Area"}, - [MODULES_NR] = {0, "Modules Area"}, + [IDENTITY_NR] = {0, "Identity Mapping"}, + [KERNEL_START_NR] = {(unsigned long)_stext, "Kernel Image Start"}, + [KERNEL_END_NR] = {(unsigned long)_end, "Kernel Image End"}, +#ifdef CONFIG_KASAN + [KASAN_SHADOW_START_NR] = {KASAN_SHADOW_START, "Kasan Shadow Start"}, + [KASAN_SHADOW_END_NR] = {KASAN_SHADOW_END, "Kasan Shadow End"}, +#endif + [VMEMMAP_NR] = {0, "vmemmap Area"}, + [VMALLOC_NR] = {0, "vmalloc Area"}, + [MODULES_NR] = {0, "Modules Area"}, { -1, NULL } }; @@ -80,7 +90,7 @@ static void note_page(struct seq_file *m, struct pg_state *st, } else if (prot != cur || level != st->level || st->current_address >= st->marker[1].start_address) { /* Print the actual finished series */ - seq_printf(m, "0x%0*lx-0x%0*lx", + seq_printf(m, "0x%0*lx-0x%0*lx ", width, st->start_address, width, st->current_address); delta = (st->current_address - st->start_address) >> 10; @@ -90,7 +100,7 @@ static void note_page(struct seq_file *m, struct pg_state *st, } seq_printf(m, "%9lu%c ", delta, *unit); print_prot(m, st->current_prot, st->level); - if (st->current_address >= st->marker[1].start_address) { + while (st->current_address >= st->marker[1].start_address) { st->marker++; seq_printf(m, "---[ %s ]---\n", st->marker->name); } @@ -100,6 +110,17 @@ static void note_page(struct seq_file *m, struct pg_state *st, } } +#ifdef CONFIG_KASAN +static void note_kasan_zero_page(struct seq_file *m, struct pg_state *st) +{ + unsigned int prot; + + prot = pte_val(*kasan_zero_pte) & + (_PAGE_PROTECT | _PAGE_INVALID | _PAGE_NOEXEC); + note_page(m, st, prot, 4); +} +#endif + /* * The actual page table walker functions. In order to keep the * implementation of print_prot() short, we only check and pass @@ -132,6 +153,13 @@ static void walk_pmd_level(struct seq_file *m, struct pg_state *st, pmd_t *pmd; int i; +#ifdef CONFIG_KASAN + if ((pud_val(*pud) & PAGE_MASK) == __pa(kasan_zero_pmd)) { + note_kasan_zero_page(m, st); + return; + } +#endif + for (i = 0; i < PTRS_PER_PMD && addr < max_addr; i++) { st->current_address = addr; pmd = pmd_offset(pud, addr); @@ -156,6 +184,13 @@ static void walk_pud_level(struct seq_file *m, struct pg_state *st, pud_t *pud; int i; +#ifdef CONFIG_KASAN + if ((p4d_val(*p4d) & PAGE_MASK) == __pa(kasan_zero_pud)) { + note_kasan_zero_page(m, st); + return; + } +#endif + for (i = 0; i < PTRS_PER_PUD && addr < max_addr; i++) { st->current_address = addr; pud = pud_offset(p4d, addr); @@ -179,6 +214,13 @@ static void walk_p4d_level(struct seq_file *m, struct pg_state *st, p4d_t *p4d; int i; +#ifdef CONFIG_KASAN + if ((pgd_val(*pgd) & PAGE_MASK) == __pa(kasan_zero_p4d)) { + note_kasan_zero_page(m, st); + return; + } +#endif + for (i = 0; i < PTRS_PER_P4D && addr < max_addr; i++) { st->current_address = addr; p4d = p4d_offset(pgd, addr); diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 72af23bacbb5..2b8f32f56e0c 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -636,17 +636,19 @@ struct pfault_refbk { u64 reserved; } __attribute__ ((packed, aligned(8))); +static struct pfault_refbk pfault_init_refbk = { + .refdiagc = 0x258, + .reffcode = 0, + .refdwlen = 5, + .refversn = 2, + .refgaddr = __LC_LPP, + .refselmk = 1ULL << 48, + .refcmpmk = 1ULL << 48, + .reserved = __PF_RES_FIELD +}; + int pfault_init(void) { - struct pfault_refbk refbk = { - .refdiagc = 0x258, - .reffcode = 0, - .refdwlen = 5, - .refversn = 2, - .refgaddr = __LC_LPP, - .refselmk = 1ULL << 48, - .refcmpmk = 1ULL << 48, - .reserved = __PF_RES_FIELD }; int rc; if (pfault_disable) @@ -658,18 +660,20 @@ int pfault_init(void) "1: la %0,8\n" "2:\n" EX_TABLE(0b,1b) - : "=d" (rc) : "a" (&refbk), "m" (refbk) : "cc"); + : "=d" (rc) + : "a" (&pfault_init_refbk), "m" (pfault_init_refbk) : "cc"); return rc; } +static struct pfault_refbk pfault_fini_refbk = { + .refdiagc = 0x258, + .reffcode = 1, + .refdwlen = 5, + .refversn = 2, +}; + void pfault_fini(void) { - struct pfault_refbk refbk = { - .refdiagc = 0x258, - .reffcode = 1, - .refdwlen = 5, - .refversn = 2, - }; if (pfault_disable) return; @@ -678,7 +682,7 @@ void pfault_fini(void) " diag %0,0,0x258\n" "0: nopr %%r7\n" EX_TABLE(0b,0b) - : : "a" (&refbk), "m" (refbk) : "cc"); + : : "a" (&pfault_fini_refbk), "m" (pfault_fini_refbk) : "cc"); } static DEFINE_SPINLOCK(pfault_lock); diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index bb44990c8212..911c7ded35f1 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -708,11 +708,13 @@ void gmap_discard(struct gmap *gmap, unsigned long from, unsigned long to) vmaddr |= gaddr & ~PMD_MASK; /* Find vma in the parent mm */ vma = find_vma(gmap->mm, vmaddr); + if (!vma) + continue; /* * We do not discard pages that are backed by * hugetlbfs, so we don't have to refault them. */ - if (vma && is_vm_hugetlb_page(vma)) + if (is_vm_hugetlb_page(vma)) continue; size = min(to - gaddr, PMD_SIZE - (gaddr & ~PMD_MASK)); zap_page_range(vma, vmaddr, size); diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 3fa3e5323612..92d7a153e72a 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -42,6 +42,7 @@ #include <asm/ctl_reg.h> #include <asm/sclp.h> #include <asm/set_memory.h> +#include <asm/kasan.h> pgd_t swapper_pg_dir[PTRS_PER_PGD] __section(.bss..swapper_pg_dir); @@ -98,8 +99,9 @@ void __init paging_init(void) S390_lowcore.user_asce = S390_lowcore.kernel_asce; crst_table_init((unsigned long *) init_mm.pgd, pgd_type); vmem_map_init(); + kasan_copy_shadow(init_mm.pgd); - /* enable virtual mapping in kernel mode */ + /* 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.kernel_asce, 13, 13); @@ -107,6 +109,7 @@ void __init paging_init(void) psw_bits(psw).dat = 1; psw_bits(psw).as = PSW_BITS_AS_HOME; __load_psw_mask(psw.mask); + kasan_free_early_identity(); sparse_memory_present_with_active_regions(MAX_NUMNODES); sparse_init(); diff --git a/arch/s390/mm/kasan_init.c b/arch/s390/mm/kasan_init.c new file mode 100644 index 000000000000..acb9645b762b --- /dev/null +++ b/arch/s390/mm/kasan_init.c @@ -0,0 +1,387 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/kasan.h> +#include <linux/sched/task.h> +#include <linux/memblock.h> +#include <asm/pgalloc.h> +#include <asm/pgtable.h> +#include <asm/kasan.h> +#include <asm/mem_detect.h> +#include <asm/processor.h> +#include <asm/sclp.h> +#include <asm/facility.h> +#include <asm/sections.h> +#include <asm/setup.h> + +static unsigned long segment_pos __initdata; +static unsigned long segment_low __initdata; +static unsigned long pgalloc_pos __initdata; +static unsigned long pgalloc_low __initdata; +static unsigned long pgalloc_freeable __initdata; +static bool has_edat __initdata; +static bool has_nx __initdata; + +#define __sha(x) ((unsigned long)kasan_mem_to_shadow((void *)x)) + +static pgd_t early_pg_dir[PTRS_PER_PGD] __initdata __aligned(PAGE_SIZE); + +static void __init kasan_early_panic(const char *reason) +{ + sclp_early_printk("The Linux kernel failed to boot with the KernelAddressSanitizer:\n"); + sclp_early_printk(reason); + disabled_wait(0); +} + +static void * __init kasan_early_alloc_segment(void) +{ + segment_pos -= _SEGMENT_SIZE; + + if (segment_pos < segment_low) + kasan_early_panic("out of memory during initialisation\n"); + + return (void *)segment_pos; +} + +static void * __init kasan_early_alloc_pages(unsigned int order) +{ + pgalloc_pos -= (PAGE_SIZE << order); + + if (pgalloc_pos < pgalloc_low) + kasan_early_panic("out of memory during initialisation\n"); + + return (void *)pgalloc_pos; +} + +static void * __init kasan_early_crst_alloc(unsigned long val) +{ + unsigned long *table; + + table = kasan_early_alloc_pages(CRST_ALLOC_ORDER); + if (table) + crst_table_init(table, val); + return table; +} + +static pte_t * __init kasan_early_pte_alloc(void) +{ + static void *pte_leftover; + pte_t *pte; + + BUILD_BUG_ON(_PAGE_TABLE_SIZE * 2 != PAGE_SIZE); + + if (!pte_leftover) { + pte_leftover = kasan_early_alloc_pages(0); + pte = pte_leftover + _PAGE_TABLE_SIZE; + } else { + pte = pte_leftover; + pte_leftover = NULL; + } + memset64((u64 *)pte, _PAGE_INVALID, PTRS_PER_PTE); + return pte; +} + +enum populate_mode { + POPULATE_ONE2ONE, + POPULATE_MAP, + POPULATE_ZERO_SHADOW +}; +static void __init kasan_early_vmemmap_populate(unsigned long address, + unsigned long end, + enum populate_mode mode) +{ + unsigned long pgt_prot_zero, pgt_prot, sgt_prot; + pgd_t *pg_dir; + p4d_t *p4_dir; + pud_t *pu_dir; + pmd_t *pm_dir; + pte_t *pt_dir; + + pgt_prot_zero = pgprot_val(PAGE_KERNEL_RO); + if (!has_nx) + pgt_prot_zero &= ~_PAGE_NOEXEC; + pgt_prot = pgprot_val(PAGE_KERNEL_EXEC); + sgt_prot = pgprot_val(SEGMENT_KERNEL_EXEC); + + while (address < end) { + pg_dir = pgd_offset_k(address); + if (pgd_none(*pg_dir)) { + if (mode == POPULATE_ZERO_SHADOW && + IS_ALIGNED(address, PGDIR_SIZE) && + end - address >= PGDIR_SIZE) { + pgd_populate(&init_mm, pg_dir, kasan_zero_p4d); + address = (address + PGDIR_SIZE) & PGDIR_MASK; + continue; + } + p4_dir = kasan_early_crst_alloc(_REGION2_ENTRY_EMPTY); + pgd_populate(&init_mm, pg_dir, p4_dir); + } + + p4_dir = p4d_offset(pg_dir, address); + if (p4d_none(*p4_dir)) { + if (mode == POPULATE_ZERO_SHADOW && + IS_ALIGNED(address, P4D_SIZE) && + end - address >= P4D_SIZE) { + p4d_populate(&init_mm, p4_dir, kasan_zero_pud); + address = (address + P4D_SIZE) & P4D_MASK; + continue; + } + pu_dir = kasan_early_crst_alloc(_REGION3_ENTRY_EMPTY); + p4d_populate(&init_mm, p4_dir, pu_dir); + } + + pu_dir = pud_offset(p4_dir, address); + if (pud_none(*pu_dir)) { + if (mode == POPULATE_ZERO_SHADOW && + IS_ALIGNED(address, PUD_SIZE) && + end - address >= PUD_SIZE) { + pud_populate(&init_mm, pu_dir, kasan_zero_pmd); + address = (address + PUD_SIZE) & PUD_MASK; + continue; + } + pm_dir = kasan_early_crst_alloc(_SEGMENT_ENTRY_EMPTY); + pud_populate(&init_mm, pu_dir, pm_dir); + } + + pm_dir = pmd_offset(pu_dir, address); + if (pmd_none(*pm_dir)) { + if (mode == POPULATE_ZERO_SHADOW && + IS_ALIGNED(address, PMD_SIZE) && + end - address >= PMD_SIZE) { + pmd_populate(&init_mm, pm_dir, kasan_zero_pte); + address = (address + PMD_SIZE) & PMD_MASK; + continue; + } + /* the first megabyte of 1:1 is mapped with 4k pages */ + if (has_edat && address && end - address >= PMD_SIZE && + mode != POPULATE_ZERO_SHADOW) { + void *page; + + if (mode == POPULATE_ONE2ONE) { + page = (void *)address; + } else { + page = kasan_early_alloc_segment(); + memset(page, 0, _SEGMENT_SIZE); + } + pmd_val(*pm_dir) = __pa(page) | sgt_prot; + address = (address + PMD_SIZE) & PMD_MASK; + continue; + } + + pt_dir = kasan_early_pte_alloc(); + pmd_populate(&init_mm, pm_dir, pt_dir); + } else if (pmd_large(*pm_dir)) { + address = (address + PMD_SIZE) & PMD_MASK; + continue; + } + + pt_dir = pte_offset_kernel(pm_dir, address); + if (pte_none(*pt_dir)) { + void *page; + + switch (mode) { + case POPULATE_ONE2ONE: + page = (void *)address; + pte_val(*pt_dir) = __pa(page) | pgt_prot; + break; + case POPULATE_MAP: + page = kasan_early_alloc_pages(0); + memset(page, 0, PAGE_SIZE); + pte_val(*pt_dir) = __pa(page) | pgt_prot; + break; + case POPULATE_ZERO_SHADOW: + page = kasan_zero_page; + pte_val(*pt_dir) = __pa(page) | pgt_prot_zero; + break; + } + } + address += PAGE_SIZE; + } +} + +static void __init kasan_set_pgd(pgd_t *pgd, unsigned long asce_type) +{ + unsigned long asce_bits; + + asce_bits = asce_type | _ASCE_TABLE_LENGTH; + S390_lowcore.kernel_asce = (__pa(pgd) & PAGE_MASK) | asce_bits; + S390_lowcore.user_asce = S390_lowcore.kernel_asce; + + __ctl_load(S390_lowcore.kernel_asce, 1, 1); + __ctl_load(S390_lowcore.kernel_asce, 7, 7); + __ctl_load(S390_lowcore.kernel_asce, 13, 13); +} + +static void __init kasan_enable_dat(void) +{ + psw_t psw; + + psw.mask = __extract_psw(); + psw_bits(psw).dat = 1; + psw_bits(psw).as = PSW_BITS_AS_HOME; + __load_psw_mask(psw.mask); +} + +static void __init kasan_early_detect_facilities(void) +{ + __stfle(S390_lowcore.stfle_fac_list, + ARRAY_SIZE(S390_lowcore.stfle_fac_list)); + if (test_facility(8)) { + has_edat = true; + __ctl_set_bit(0, 23); + } + if (!noexec_disabled && test_facility(130)) { + has_nx = true; + __ctl_set_bit(0, 20); + } +} + +static unsigned long __init get_mem_detect_end(void) +{ + unsigned long start; + unsigned long end; + + if (mem_detect.count) { + __get_mem_detect_block(mem_detect.count - 1, &start, &end); + return end; + } + return 0; +} + +void __init kasan_early_init(void) +{ + unsigned long untracked_mem_end; + unsigned long shadow_alloc_size; + unsigned long initrd_end; + unsigned long asce_type; + unsigned long memsize; + unsigned long vmax; + unsigned long pgt_prot = pgprot_val(PAGE_KERNEL_RO); + pte_t pte_z; + pmd_t pmd_z = __pmd(__pa(kasan_zero_pte) | _SEGMENT_ENTRY); + pud_t pud_z = __pud(__pa(kasan_zero_pmd) | _REGION3_ENTRY); + p4d_t p4d_z = __p4d(__pa(kasan_zero_pud) | _REGION2_ENTRY); + + kasan_early_detect_facilities(); + if (!has_nx) + pgt_prot &= ~_PAGE_NOEXEC; + pte_z = __pte(__pa(kasan_zero_page) | pgt_prot); + + 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; + 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 = vmax = _REGION1_SIZE; + 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 = vmax = _REGION2_SIZE; + asce_type = _ASCE_TYPE_REGION3; + } + + /* init kasan zero shadow */ + crst_table_init((unsigned long *)kasan_zero_p4d, p4d_val(p4d_z)); + crst_table_init((unsigned long *)kasan_zero_pud, pud_val(pud_z)); + crst_table_init((unsigned long *)kasan_zero_pmd, pmd_val(pmd_z)); + memset64((u64 *)kasan_zero_pte, pte_val(pte_z), PTRS_PER_PTE); + + shadow_alloc_size = memsize >> KASAN_SHADOW_SCALE_SHIFT; + pgalloc_low = round_up((unsigned long)_end, _SEGMENT_SIZE); + if (IS_ENABLED(CONFIG_BLK_DEV_INITRD)) { + initrd_end = + round_up(INITRD_START + INITRD_SIZE, _SEGMENT_SIZE); + pgalloc_low = max(pgalloc_low, initrd_end); + } + + if (pgalloc_low + shadow_alloc_size > memsize) + kasan_early_panic("out of memory during initialisation\n"); + + if (has_edat) { + segment_pos = round_down(memsize, _SEGMENT_SIZE); + segment_low = segment_pos - shadow_alloc_size; + pgalloc_pos = segment_low; + } else { + pgalloc_pos = memsize; + } + init_mm.pgd = early_pg_dir; + /* + * Current memory layout: + * +- 0 -------------+ +- shadow start -+ + * | 1:1 ram mapping | /| 1/8 ram | + * +- end of ram ----+ / +----------------+ + * | ... gap ... |/ | kasan | + * +- shadow start --+ | zero | + * | 1/8 addr space | | page | + * +- shadow end -+ | mapping | + * | ... gap ... |\ | (untracked) | + * +- modules vaddr -+ \ +----------------+ + * | 2Gb | \| unmapped | allocated per module + * +-----------------+ +- shadow end ---+ + */ + /* populate kasan shadow (for identity mapping and zero page mapping) */ + kasan_early_vmemmap_populate(__sha(0), __sha(memsize), POPULATE_MAP); + if (IS_ENABLED(CONFIG_MODULES)) + untracked_mem_end = vmax - MODULES_LEN; + kasan_early_vmemmap_populate(__sha(max_physmem_end), + __sha(untracked_mem_end), + 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_enable_dat(); + /* enable kasan */ + init_task.kasan_depth = 0; + memblock_reserve(pgalloc_pos, memsize - pgalloc_pos); + sclp_early_printk("KernelAddressSanitizer initialized\n"); +} + +void __init kasan_copy_shadow(pgd_t *pg_dir) +{ + /* + * At this point we are still running on early pages setup early_pg_dir, + * while swapper_pg_dir has just been initialized with identity mapping. + * Carry over shadow memory region from early_pg_dir to swapper_pg_dir. + */ + + pgd_t *pg_dir_src; + 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); + 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)); +} + +void __init kasan_free_early_identity(void) +{ + memblock_free(pgalloc_pos, pgalloc_freeable - pgalloc_pos); +} diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c index 7be06475809b..97b3ee53852b 100644 --- a/arch/s390/mm/maccess.c +++ b/arch/s390/mm/maccess.c @@ -89,10 +89,8 @@ static int __memcpy_real(void *dest, void *src, size_t count) return rc; } -/* - * Copy memory in real mode (kernel to kernel) - */ -int memcpy_real(void *dest, void *src, size_t count) +static unsigned long _memcpy_real(unsigned long dest, unsigned long src, + unsigned long count) { int irqs_disabled, rc; unsigned long flags; @@ -103,7 +101,7 @@ int memcpy_real(void *dest, void *src, size_t count) irqs_disabled = arch_irqs_disabled_flags(flags); if (!irqs_disabled) trace_hardirqs_off(); - rc = __memcpy_real(dest, src, count); + rc = __memcpy_real((void *) dest, (void *) src, (size_t) count); if (!irqs_disabled) trace_hardirqs_on(); __arch_local_irq_ssm(flags); @@ -111,6 +109,23 @@ int memcpy_real(void *dest, void *src, size_t count) } /* + * Copy memory in real mode (kernel to kernel) + */ +int memcpy_real(void *dest, void *src, size_t count) +{ + if (S390_lowcore.nodat_stack != 0) + return CALL_ON_STACK(_memcpy_real, S390_lowcore.nodat_stack, + 3, dest, src, count); + /* + * This is a really early memcpy_real call, the stacks are + * not set up yet. Just call _memcpy_real on the early boot + * stack + */ + return _memcpy_real((unsigned long) dest,(unsigned long) src, + (unsigned long) count); +} + +/* * Copy memory in absolute mode (kernel to kernel) */ void memcpy_absolute(void *dest, void *src, size_t count) diff --git a/arch/s390/mm/mem_detect.c b/arch/s390/mm/mem_detect.c deleted file mode 100644 index 21f6c82c8296..000000000000 --- a/arch/s390/mm/mem_detect.c +++ /dev/null @@ -1,62 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * Copyright IBM Corp. 2008, 2009 - * - * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> - */ - -#include <linux/kernel.h> -#include <linux/memblock.h> -#include <linux/init.h> -#include <linux/debugfs.h> -#include <linux/seq_file.h> -#include <asm/ipl.h> -#include <asm/sclp.h> -#include <asm/setup.h> - -#define CHUNK_READ_WRITE 0 -#define CHUNK_READ_ONLY 1 - -static inline void memblock_physmem_add(phys_addr_t start, phys_addr_t size) -{ - memblock_dbg("memblock_physmem_add: [%#016llx-%#016llx]\n", - start, start + size - 1); - memblock_add_range(&memblock.memory, start, size, 0, 0); - memblock_add_range(&memblock.physmem, start, size, 0, 0); -} - -void __init detect_memory_memblock(void) -{ - unsigned long memsize, rnmax, rzm, addr, size; - int type; - - rzm = sclp.rzm; - rnmax = sclp.rnmax; - memsize = rzm * rnmax; - if (!rzm) - rzm = 1UL << 17; - max_physmem_end = memsize; - addr = 0; - /* keep memblock lists close to the kernel */ - memblock_set_bottom_up(true); - do { - size = 0; - /* assume lowcore is writable */ - type = addr ? tprot(addr) : CHUNK_READ_WRITE; - do { - size += rzm; - if (max_physmem_end && addr + size >= max_physmem_end) - break; - } while (type == tprot(addr + size)); - if (type == CHUNK_READ_WRITE || type == CHUNK_READ_ONLY) { - if (max_physmem_end && (addr + size > max_physmem_end)) - size = max_physmem_end - addr; - memblock_physmem_add(addr, size); - } - addr += size; - } while (addr < max_physmem_end); - memblock_set_bottom_up(false); - if (!max_physmem_end) - max_physmem_end = memblock_end_of_DRAM(); - memblock_dump_all(); -} diff --git a/arch/s390/purgatory/head.S b/arch/s390/purgatory/head.S index 2e3707b12edd..5a10ce34b95d 100644 --- a/arch/s390/purgatory/head.S +++ b/arch/s390/purgatory/head.S @@ -11,6 +11,7 @@ #include <asm/asm-offsets.h> #include <asm/page.h> #include <asm/sigp.h> +#include <asm/ptrace.h> /* The purgatory is the code running between two kernels. It's main purpose * is to verify that the next kernel was not corrupted after load and to @@ -88,8 +89,7 @@ ENTRY(purgatory_start) .base_crash: /* Setup stack */ - larl %r15,purgatory_end - aghi %r15,-160 + larl %r15,purgatory_end-STACK_FRAME_OVERHEAD /* If the next kernel is KEXEC_TYPE_CRASH the purgatory is called * directly with a flag passed in %r2 whether the purgatory shall do |