diff options
Diffstat (limited to 'arch/s390')
127 files changed, 2739 insertions, 1560 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index c59d2b54df49..c60e699e99f5 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -21,7 +21,7 @@ config ARCH_PROC_KCORE_TEXT def_bool y config GENERIC_HWEIGHT - def_bool y + def_bool !HAVE_MARCH_Z196_FEATURES config GENERIC_BUG def_bool y if BUG @@ -142,6 +142,7 @@ config S390 select FUNCTION_ALIGNMENT_8B if CC_IS_GCC select FUNCTION_ALIGNMENT_16B if !CC_IS_GCC select GENERIC_ALLOCATOR + select GENERIC_CPU_DEVICES select GENERIC_CPU_AUTOPROBE select GENERIC_CPU_VULNERABILITIES select GENERIC_ENTRY @@ -158,6 +159,7 @@ config S390 select HAVE_ARCH_KASAN select HAVE_ARCH_KASAN_VMALLOC select HAVE_ARCH_KCSAN + select HAVE_ARCH_KMSAN select HAVE_ARCH_KFENCE select HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET select HAVE_ARCH_SECCOMP_FILTER @@ -602,6 +604,19 @@ config RANDOMIZE_BASE as a security feature that deters exploit attempts relying on knowledge of the location of kernel internals. +config RANDOMIZE_IDENTITY_BASE + bool "Randomize the address of the identity mapping base" + depends on RANDOMIZE_BASE + default DEBUG_VM + help + The identity mapping base address is pinned to zero by default. + Allow randomization of that base to expose otherwise missed + notion of physical and virtual addresses of data structures. + That does not have any impact on the base address at which the + kernel image is loaded. + + If unsure, say N + config KERNEL_IMAGE_BASE hex "Kernel image base address" range 0x100000 0x1FFFFFE0000000 if !KASAN @@ -797,17 +812,6 @@ config HAVE_PNETID menu "Virtualization" -config PROTECTED_VIRTUALIZATION_GUEST - def_bool n - prompt "Protected virtualization guest support" - help - Select this option, if you want to be able to run this - kernel as a protected virtualization KVM guest. - Protected virtualization capable machines have a mini hypervisor - located at machine level (an ultravisor). With help of the - Ultravisor, KVM will be able to run "protected" VMs, special - VMs whose memory and management data are unavailable to KVM. - config PFAULT def_bool y prompt "Pseudo page fault support" diff --git a/arch/s390/Makefile b/arch/s390/Makefile index f2b21c7a70ef..7fd57398221e 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -36,7 +36,7 @@ KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO_DWARF4), $(call cc-option KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_CC_NO_ARRAY_BOUNDS),-Wno-array-bounds) UTS_MACHINE := s390x -STACK_SIZE := $(if $(CONFIG_KASAN),65536,16384) +STACK_SIZE := $(if $(CONFIG_KASAN),65536,$(if $(CONFIG_KMSAN),65536,16384)) 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 c2978cb03b36..91a30e017d65 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -46,9 +46,9 @@ * /proc entries (sysctl) */ static const char appldata_proc_name[APPLDATA_PROC_NAME_LENGTH] = "appldata"; -static int appldata_timer_handler(struct ctl_table *ctl, int write, +static int appldata_timer_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos); -static int appldata_interval_handler(struct ctl_table *ctl, int write, +static int appldata_interval_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos); static struct ctl_table_header *appldata_sysctl_header; @@ -199,7 +199,7 @@ static void __appldata_vtimer_setup(int cmd) * Start/Stop timer, show status of timer (0 = not active, 1 = active) */ static int -appldata_timer_handler(struct ctl_table *ctl, int write, +appldata_timer_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { int timer_active = appldata_timer_active; @@ -232,7 +232,7 @@ appldata_timer_handler(struct ctl_table *ctl, int write, * current timer interval. */ static int -appldata_interval_handler(struct ctl_table *ctl, int write, +appldata_interval_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { int interval = appldata_interval; @@ -262,7 +262,7 @@ appldata_interval_handler(struct ctl_table *ctl, int write, * monitoring (0 = not in process, 1 = in process) */ static int -appldata_generic_handler(struct ctl_table *ctl, int write, +appldata_generic_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { struct appldata_ops *ops = NULL, *tmp_ops; diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile index 070c9b2e905f..4f476884d340 100644 --- a/arch/s390/boot/Makefile +++ b/arch/s390/boot/Makefile @@ -3,11 +3,13 @@ # Makefile for the linux s390-specific parts of the memory manager. # +# Tooling runtimes are unavailable and cannot be linked for early boot code KCOV_INSTRUMENT := n GCOV_PROFILE := n UBSAN_SANITIZE := n KASAN_SANITIZE := n KCSAN_SANITIZE := n +KMSAN_SANITIZE := n KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR) KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR) @@ -37,11 +39,11 @@ CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char obj-y := head.o als.o startup.o physmem_info.o ipl_parm.o ipl_report.o vmem.o obj-y += string.o ebcdic.o sclp_early_core.o mem.o ipl_vmparm.o cmdline.o -obj-y += version.o pgm_check_info.o ctype.o ipl_data.o relocs.o -obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE)) += uv.o +obj-y += version.o pgm_check_info.o ctype.o ipl_data.o relocs.o alternative.o uv.o obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o obj-y += $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) info.o obj-$(CONFIG_KERNEL_ZSTD) += clz_ctz.o +obj-$(CONFIG_KMSAN) += kmsan.o obj-all := $(obj-y) piggy.o syms.o targets := bzImage section_cmp.boot.data section_cmp.boot.preserved.data $(obj-y) diff --git a/arch/s390/boot/alternative.c b/arch/s390/boot/alternative.c new file mode 100644 index 000000000000..abc08d2c873d --- /dev/null +++ b/arch/s390/boot/alternative.c @@ -0,0 +1,3 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "../kernel/alternative.c" diff --git a/arch/s390/boot/boot.h b/arch/s390/boot/boot.h index 18027fdc92b0..83e2ce050b6c 100644 --- a/arch/s390/boot/boot.h +++ b/arch/s390/boot/boot.h @@ -30,6 +30,8 @@ struct vmlinux_info { unsigned long init_mm_off; unsigned long swapper_pg_dir_off; unsigned long invalid_pg_dir_off; + unsigned long alt_instructions; + unsigned long alt_instructions_end; #ifdef CONFIG_KASAN unsigned long kasan_early_shadow_page_off; unsigned long kasan_early_shadow_pte_off; @@ -89,8 +91,10 @@ extern char _end[], _decompressor_end[]; extern unsigned char _compressed_start[]; extern unsigned char _compressed_end[]; extern struct vmlinux_info _vmlinux_info; + #define vmlinux _vmlinux_info +#define __lowcore_pa(x) ((unsigned long)(x) % sizeof(struct lowcore)) #define __abs_lowcore_pa(x) (((unsigned long)(x) - __abs_lowcore) % sizeof(struct lowcore)) #define __kernel_va(x) ((void *)((unsigned long)(x) - __kaslr_offset_phys + __kaslr_offset)) #define __kernel_pa(x) ((unsigned long)(x) - __kaslr_offset + __kaslr_offset_phys) diff --git a/arch/s390/boot/install.sh b/arch/s390/boot/install.sh index a13dd2f2aa1c..fa41486258ee 100755 --- a/arch/s390/boot/install.sh +++ b/arch/s390/boot/install.sh @@ -15,6 +15,8 @@ # $3 - kernel map file # $4 - default install path (blank if root directory) +set -e + echo "Warning: '${INSTALLKERNEL}' command not available - additional " \ "bootloader config required" >&2 if [ -f "$4/vmlinuz-$1" ]; then mv -- "$4/vmlinuz-$1" "$4/vmlinuz-$1.old"; fi diff --git a/arch/s390/boot/ipl_parm.c b/arch/s390/boot/ipl_parm.c index b24de9aabf7d..1773b72a6a7b 100644 --- a/arch/s390/boot/ipl_parm.c +++ b/arch/s390/boot/ipl_parm.c @@ -3,6 +3,7 @@ #include <linux/init.h> #include <linux/ctype.h> #include <linux/pgtable.h> +#include <asm/abs_lowcore.h> #include <asm/page-states.h> #include <asm/ebcdic.h> #include <asm/sclp.h> @@ -51,11 +52,11 @@ static inline int __diag308(unsigned long subcode, void *addr) : [r1] "+&d" (r1.pair), [reg1] "=&d" (reg1), [reg2] "=&a" (reg2), - "+Q" (S390_lowcore.program_new_psw), + "+Q" (get_lowcore()->program_new_psw), "=Q" (old) : [subcode] "d" (subcode), [psw_old] "a" (&old), - [psw_pgm] "a" (&S390_lowcore.program_new_psw) + [psw_pgm] "a" (&get_lowcore()->program_new_psw) : "cc", "memory"); return r1.odd; } @@ -310,5 +311,7 @@ void parse_boot_command_line(void) prot_virt_host = 1; } #endif + if (!strcmp(param, "relocate_lowcore") && test_facility(193)) + relocate_lowcore = 1; } } diff --git a/arch/s390/boot/ipl_report.c b/arch/s390/boot/ipl_report.c index 1803035e68d2..d00898852a88 100644 --- a/arch/s390/boot/ipl_report.c +++ b/arch/s390/boot/ipl_report.c @@ -106,7 +106,7 @@ int read_ipl_report(void) * the IPL parameter list, then align the address to a double * word boundary. */ - tmp = (unsigned long) S390_lowcore.ipl_parmblock_ptr; + tmp = (unsigned long)get_lowcore()->ipl_parmblock_ptr; pl_hdr = (struct ipl_pl_hdr *) tmp; tmp = (tmp + pl_hdr->len + 7) & -8UL; rl_hdr = (struct ipl_rl_hdr *) tmp; diff --git a/arch/s390/boot/kmsan.c b/arch/s390/boot/kmsan.c new file mode 100644 index 000000000000..e7b3ac48143e --- /dev/null +++ b/arch/s390/boot/kmsan.c @@ -0,0 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 +#include <linux/kmsan-checks.h> + +void kmsan_unpoison_memory(const void *address, size_t size) +{ +} diff --git a/arch/s390/boot/pgm_check_info.c b/arch/s390/boot/pgm_check_info.c index ea96275b0380..5352b3d356da 100644 --- a/arch/s390/boot/pgm_check_info.c +++ b/arch/s390/boot/pgm_check_info.c @@ -145,22 +145,22 @@ void print_stacktrace(unsigned long sp) void print_pgm_check_info(void) { - unsigned long *gpregs = (unsigned long *)S390_lowcore.gpregs_save_area; - struct psw_bits *psw = &psw_bits(S390_lowcore.psw_save_area); + unsigned long *gpregs = (unsigned long *)get_lowcore()->gpregs_save_area; + struct psw_bits *psw = &psw_bits(get_lowcore()->psw_save_area); decompressor_printk("Linux version %s\n", kernel_version); if (!is_prot_virt_guest() && early_command_line[0]) decompressor_printk("Kernel command line: %s\n", early_command_line); decompressor_printk("Kernel fault: interruption code %04x ilc:%x\n", - S390_lowcore.pgm_code, S390_lowcore.pgm_ilc >> 1); + get_lowcore()->pgm_code, get_lowcore()->pgm_ilc >> 1); if (kaslr_enabled()) { decompressor_printk("Kernel random base: %lx\n", __kaslr_offset); decompressor_printk("Kernel random base phys: %lx\n", __kaslr_offset_phys); } decompressor_printk("PSW : %016lx %016lx (%pS)\n", - S390_lowcore.psw_save_area.mask, - S390_lowcore.psw_save_area.addr, - (void *)S390_lowcore.psw_save_area.addr); + get_lowcore()->psw_save_area.mask, + get_lowcore()->psw_save_area.addr, + (void *)get_lowcore()->psw_save_area.addr); decompressor_printk( " R:%x T:%x IO:%x EX:%x Key:%x M:%x W:%x P:%x AS:%x CC:%x PM:%x RI:%x EA:%x\n", psw->per, psw->dat, psw->io, psw->ext, psw->key, psw->mcheck, @@ -174,8 +174,8 @@ void print_pgm_check_info(void) gpregs[8], gpregs[9], gpregs[10], gpregs[11]); decompressor_printk(" %016lx %016lx %016lx %016lx\n", gpregs[12], gpregs[13], gpregs[14], gpregs[15]); - print_stacktrace(S390_lowcore.gpregs_save_area[15]); + print_stacktrace(get_lowcore()->gpregs_save_area[15]); decompressor_printk("Last Breaking-Event-Address:\n"); - decompressor_printk(" [<%016lx>] %pS\n", (unsigned long)S390_lowcore.pgm_last_break, - (void *)S390_lowcore.pgm_last_break); + decompressor_printk(" [<%016lx>] %pS\n", (unsigned long)get_lowcore()->pgm_last_break, + (void *)get_lowcore()->pgm_last_break); } diff --git a/arch/s390/boot/physmem_info.c b/arch/s390/boot/physmem_info.c index 0cf79826eef9..4c9ad8258f7e 100644 --- a/arch/s390/boot/physmem_info.c +++ b/arch/s390/boot/physmem_info.c @@ -81,11 +81,11 @@ static int __diag260(unsigned long rx1, unsigned long rx2) [reg2] "=&a" (reg2), [rc] "+&d" (rc), [ry] "+&d" (ry), - "+Q" (S390_lowcore.program_new_psw), + "+Q" (get_lowcore()->program_new_psw), "=Q" (old) : [rx] "d" (rx.pair), [psw_old] "a" (&old), - [psw_pgm] "a" (&S390_lowcore.program_new_psw) + [psw_pgm] "a" (&get_lowcore()->program_new_psw) : "cc", "memory"); return rc == 0 ? ry : -1; } @@ -129,10 +129,10 @@ static int tprot(unsigned long addr) : [reg1] "=&d" (reg1), [reg2] "=&a" (reg2), [rc] "+&d" (rc), - "=Q" (S390_lowcore.program_new_psw.addr), + "=Q" (get_lowcore()->program_new_psw.addr), "=Q" (old) : [psw_old] "a" (&old), - [psw_pgm] "a" (&S390_lowcore.program_new_psw), + [psw_pgm] "a" (&get_lowcore()->program_new_psw), [addr] "a" (addr) : "cc", "memory"); return rc; diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c index 5a36d5538dae..c73b5118ad42 100644 --- a/arch/s390/boot/startup.c +++ b/arch/s390/boot/startup.c @@ -30,6 +30,7 @@ unsigned long __bootdata_preserved(vmemmap_size); unsigned long __bootdata_preserved(MODULES_VADDR); unsigned long __bootdata_preserved(MODULES_END); unsigned long __bootdata_preserved(max_mappable); +int __bootdata_preserved(relocate_lowcore); u64 __bootdata_preserved(stfle_fac_list[16]); struct oldmem_data __bootdata_preserved(oldmem_data); @@ -78,10 +79,10 @@ static int cmma_test_essa(void) [reg2] "=&a" (reg2), [rc] "+&d" (rc), [tmp] "=&d" (tmp), - "+Q" (S390_lowcore.program_new_psw), + "+Q" (get_lowcore()->program_new_psw), "=Q" (old) : [psw_old] "a" (&old), - [psw_pgm] "a" (&S390_lowcore.program_new_psw), + [psw_pgm] "a" (&get_lowcore()->program_new_psw), [cmd] "i" (ESSA_GET_STATE) : "cc", "memory"); return rc; @@ -101,10 +102,10 @@ static void cmma_init(void) static void setup_lpp(void) { - S390_lowcore.current_pid = 0; - S390_lowcore.lpp = LPP_MAGIC; + get_lowcore()->current_pid = 0; + get_lowcore()->lpp = LPP_MAGIC; if (test_facility(40)) - lpp(&S390_lowcore.lpp); + lpp(&get_lowcore()->lpp); } #ifdef CONFIG_KERNEL_UNCOMPRESSED @@ -161,7 +162,7 @@ static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, loc = (long)*reloc + phys_offset; if (loc < min_addr || loc > max_addr) error("64-bit relocation outside of kernel!\n"); - *(u64 *)loc += offset - __START_KERNEL; + *(u64 *)loc += offset; } } @@ -176,7 +177,7 @@ static void kaslr_adjust_got(unsigned long offset) */ for (entry = (u64 *)vmlinux.got_start; entry < (u64 *)vmlinux.got_end; entry++) { if (*entry) - *entry += offset - __START_KERNEL; + *entry += offset; } } @@ -251,7 +252,7 @@ static unsigned long setup_kernel_memory_layout(unsigned long kernel_size) vmemmap_size = SECTION_ALIGN_UP(pages) * sizeof(struct page); /* choose kernel address space layout: 4 or 3 levels. */ - BUILD_BUG_ON(!IS_ALIGNED(__START_KERNEL, THREAD_SIZE)); + BUILD_BUG_ON(!IS_ALIGNED(TEXT_OFFSET, THREAD_SIZE)); BUILD_BUG_ON(!IS_ALIGNED(__NO_KASLR_START_KERNEL, THREAD_SIZE)); BUILD_BUG_ON(__NO_KASLR_END_KERNEL > _REGION1_SIZE); vsize = get_vmem_size(ident_map_size, vmemmap_size, vmalloc_size, _REGION3_SIZE); @@ -304,11 +305,18 @@ static unsigned long setup_kernel_memory_layout(unsigned long kernel_size) MODULES_END = round_down(kernel_start, _SEGMENT_SIZE); MODULES_VADDR = MODULES_END - MODULES_LEN; VMALLOC_END = MODULES_VADDR; + if (IS_ENABLED(CONFIG_KMSAN)) + VMALLOC_END -= MODULES_LEN * 2; /* allow vmalloc area to occupy up to about 1/2 of the rest virtual space left */ vsize = (VMALLOC_END - FIXMAP_SIZE) / 2; vsize = round_down(vsize, _SEGMENT_SIZE); vmalloc_size = min(vmalloc_size, vsize); + if (IS_ENABLED(CONFIG_KMSAN)) { + /* take 2/3 of vmalloc area for KMSAN shadow and origins */ + vmalloc_size = round_down(vmalloc_size / 3, _SEGMENT_SIZE); + VMALLOC_END -= vmalloc_size * 2; + } VMALLOC_START = VMALLOC_END - vmalloc_size; __memcpy_real_area = round_down(VMALLOC_START - MEMCPY_REAL_SIZE, PAGE_SIZE); @@ -333,7 +341,8 @@ static unsigned long setup_kernel_memory_layout(unsigned long kernel_size) BUILD_BUG_ON(MAX_DCSS_ADDR > (1UL << MAX_PHYSMEM_BITS)); max_mappable = max(ident_map_size, MAX_DCSS_ADDR); max_mappable = min(max_mappable, vmemmap_start); - __identity_base = round_down(vmemmap_start - max_mappable, rte_size); + if (IS_ENABLED(CONFIG_RANDOMIZE_IDENTITY_BASE)) + __identity_base = round_down(vmemmap_start - max_mappable, rte_size); return asce_limit; } @@ -369,6 +378,8 @@ static void kaslr_adjust_vmlinux_info(long offset) vmlinux.init_mm_off += offset; vmlinux.swapper_pg_dir_off += offset; vmlinux.invalid_pg_dir_off += offset; + vmlinux.alt_instructions += offset; + vmlinux.alt_instructions_end += offset; #ifdef CONFIG_KASAN vmlinux.kasan_early_shadow_page_off += offset; vmlinux.kasan_early_shadow_pte_off += offset; @@ -378,31 +389,25 @@ static void kaslr_adjust_vmlinux_info(long offset) #endif } -static void fixup_vmlinux_info(void) -{ - vmlinux.entry -= __START_KERNEL; - kaslr_adjust_vmlinux_info(-__START_KERNEL); -} - void startup_kernel(void) { - unsigned long kernel_size = vmlinux.image_size + vmlinux.bss_size; - unsigned long nokaslr_offset_phys, kaslr_large_page_offset; - unsigned long amode31_lma = 0; + unsigned long vmlinux_size = vmlinux.image_size + vmlinux.bss_size; + unsigned long nokaslr_text_lma, text_lma = 0, amode31_lma = 0; + unsigned long kernel_size = TEXT_OFFSET + vmlinux_size; + unsigned long kaslr_large_page_offset; unsigned long max_physmem_end; unsigned long asce_limit; unsigned long safe_addr; psw_t psw; - fixup_vmlinux_info(); setup_lpp(); /* * Non-randomized kernel physical start address must be _SEGMENT_SIZE * aligned (see blow). */ - nokaslr_offset_phys = ALIGN(mem_safe_offset(), _SEGMENT_SIZE); - safe_addr = PAGE_ALIGN(nokaslr_offset_phys + kernel_size); + nokaslr_text_lma = ALIGN(mem_safe_offset(), _SEGMENT_SIZE); + safe_addr = PAGE_ALIGN(nokaslr_text_lma + vmlinux_size); /* * Reserve decompressor memory together with decompression heap, @@ -446,16 +451,27 @@ void startup_kernel(void) */ kaslr_large_page_offset = __kaslr_offset & ~_SEGMENT_MASK; if (kaslr_enabled()) { - unsigned long end = ident_map_size - kaslr_large_page_offset; + unsigned long size = vmlinux_size + kaslr_large_page_offset; - __kaslr_offset_phys = randomize_within_range(kernel_size, _SEGMENT_SIZE, 0, end); + text_lma = randomize_within_range(size, _SEGMENT_SIZE, TEXT_OFFSET, ident_map_size); } - if (!__kaslr_offset_phys) - __kaslr_offset_phys = nokaslr_offset_phys; - __kaslr_offset_phys |= kaslr_large_page_offset; + if (!text_lma) + text_lma = nokaslr_text_lma; + text_lma |= kaslr_large_page_offset; + + /* + * [__kaslr_offset_phys..__kaslr_offset_phys + TEXT_OFFSET] region is + * never accessed via the kernel image mapping as per the linker script: + * + * . = TEXT_OFFSET; + * + * Therefore, this region could be used for something else and does + * not need to be reserved. See how it is skipped in setup_vmem(). + */ + __kaslr_offset_phys = text_lma - TEXT_OFFSET; kaslr_adjust_vmlinux_info(__kaslr_offset_phys); - physmem_reserve(RR_VMLINUX, __kaslr_offset_phys, kernel_size); - deploy_kernel((void *)__kaslr_offset_phys); + physmem_reserve(RR_VMLINUX, text_lma, vmlinux_size); + deploy_kernel((void *)text_lma); /* vmlinux decompression is done, shrink reserved low memory */ physmem_reserve(RR_DECOMPRESSOR, 0, (unsigned long)_decompressor_end); @@ -471,10 +487,14 @@ void startup_kernel(void) * before the kernel started. Therefore, in case the two sections * overlap there is no risk of corrupting any data. */ - if (kaslr_enabled()) - amode31_lma = randomize_within_range(vmlinux.amode31_size, PAGE_SIZE, 0, SZ_2G); + if (kaslr_enabled()) { + unsigned long amode31_min; + + amode31_min = (unsigned long)_decompressor_end; + amode31_lma = randomize_within_range(vmlinux.amode31_size, PAGE_SIZE, amode31_min, SZ_2G); + } if (!amode31_lma) - amode31_lma = __kaslr_offset_phys - vmlinux.amode31_size; + amode31_lma = text_lma - vmlinux.amode31_size; physmem_reserve(RR_AMODE31, amode31_lma, vmlinux.amode31_size); /* @@ -490,18 +510,21 @@ void startup_kernel(void) * - copy_bootdata() must follow setup_vmem() to propagate changes * to bootdata made by setup_vmem() */ - clear_bss_section(__kaslr_offset_phys); - kaslr_adjust_relocs(__kaslr_offset_phys, __kaslr_offset_phys + vmlinux.image_size, + clear_bss_section(text_lma); + kaslr_adjust_relocs(text_lma, text_lma + vmlinux.image_size, __kaslr_offset, __kaslr_offset_phys); kaslr_adjust_got(__kaslr_offset); setup_vmem(__kaslr_offset, __kaslr_offset + kernel_size, asce_limit); copy_bootdata(); + __apply_alternatives((struct alt_instr *)_vmlinux_info.alt_instructions, + (struct alt_instr *)_vmlinux_info.alt_instructions_end, + ALT_CTX_EARLY); /* * Save KASLR offset for early dumps, before vmcore_info is set. * Mark as uneven to distinguish from real vmcore_info pointer. */ - S390_lowcore.vmcore_info = __kaslr_offset_phys ? __kaslr_offset_phys | 0x1UL : 0; + get_lowcore()->vmcore_info = __kaslr_offset_phys ? __kaslr_offset_phys | 0x1UL : 0; /* * Jump to the decompressed kernel entry point and switch DAT mode on. diff --git a/arch/s390/boot/string.c b/arch/s390/boot/string.c index faccb33b462c..f6b9b1df48a8 100644 --- a/arch/s390/boot/string.c +++ b/arch/s390/boot/string.c @@ -1,11 +1,18 @@ // SPDX-License-Identifier: GPL-2.0 +#define IN_BOOT_STRING_C 1 #include <linux/ctype.h> #include <linux/kernel.h> #include <linux/errno.h> #undef CONFIG_KASAN #undef CONFIG_KASAN_GENERIC +#undef CONFIG_KMSAN #include "../lib/string.c" +/* + * Duplicate some functions from the common lib/string.c + * instead of fully including it. + */ + int strncmp(const char *cs, const char *ct, size_t count) { unsigned char c1, c2; @@ -22,6 +29,15 @@ int strncmp(const char *cs, const char *ct, size_t count) return 0; } +void *memset64(uint64_t *s, uint64_t v, size_t count) +{ + uint64_t *xs = s; + + while (count--) + *xs++ = v; + return s; +} + char *skip_spaces(const char *str) { while (isspace(*str)) diff --git a/arch/s390/boot/uv.c b/arch/s390/boot/uv.c index 1e66d2cbb096..318e6ba95bfd 100644 --- a/arch/s390/boot/uv.c +++ b/arch/s390/boot/uv.c @@ -8,12 +8,8 @@ #include "uv.h" /* will be used in arch/s390/kernel/uv.c */ -#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST int __bootdata_preserved(prot_virt_guest); -#endif -#if IS_ENABLED(CONFIG_KVM) int __bootdata_preserved(prot_virt_host); -#endif struct uv_info __bootdata_preserved(uv_info); void uv_query_info(void) @@ -53,14 +49,11 @@ void uv_query_info(void) uv_info.max_secrets = uvcb.max_secrets; } -#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST if (test_bit_inv(BIT_UVC_CMD_SET_SHARED_ACCESS, (unsigned long *)uvcb.inst_calls_list) && test_bit_inv(BIT_UVC_CMD_REMOVE_SHARED_ACCESS, (unsigned long *)uvcb.inst_calls_list)) prot_virt_guest = 1; -#endif } -#if IS_ENABLED(CONFIG_KVM) unsigned long adjust_to_uv_max(unsigned long limit) { if (is_prot_virt_host() && uv_info.max_sec_stor_addr) @@ -92,4 +85,3 @@ void sanitize_prot_virt_host(void) { prot_virt_host = is_prot_virt_host_capable(); } -#endif diff --git a/arch/s390/boot/uv.h b/arch/s390/boot/uv.h index 0f3070856f8d..da4a4a8d48e0 100644 --- a/arch/s390/boot/uv.h +++ b/arch/s390/boot/uv.h @@ -2,21 +2,8 @@ #ifndef BOOT_UV_H #define BOOT_UV_H -#if IS_ENABLED(CONFIG_KVM) unsigned long adjust_to_uv_max(unsigned long limit); void sanitize_prot_virt_host(void); -#else -static inline unsigned long adjust_to_uv_max(unsigned long limit) -{ - return limit; -} -static inline void sanitize_prot_virt_host(void) {} -#endif - -#if defined(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) || IS_ENABLED(CONFIG_KVM) void uv_query_info(void); -#else -static inline void uv_query_info(void) {} -#endif #endif /* BOOT_UV_H */ diff --git a/arch/s390/boot/vmem.c b/arch/s390/boot/vmem.c index 40cfce2687c4..145035f84a0e 100644 --- a/arch/s390/boot/vmem.c +++ b/arch/s390/boot/vmem.c @@ -26,6 +26,7 @@ atomic_long_t __bootdata_preserved(direct_pages_count[PG_DIRECT_MAP_MAX]); enum populate_mode { POPULATE_NONE, POPULATE_DIRECT, + POPULATE_LOWCORE, POPULATE_ABS_LOWCORE, POPULATE_IDENTITY, POPULATE_KERNEL, @@ -89,7 +90,7 @@ static void kasan_populate_shadow(unsigned long kernel_start, unsigned long kern } memgap_start = end; } - kasan_populate(kernel_start, kernel_end, POPULATE_KASAN_MAP_SHADOW); + kasan_populate(kernel_start + TEXT_OFFSET, kernel_end, POPULATE_KASAN_MAP_SHADOW); kasan_populate(0, (unsigned long)__identity_va(0), POPULATE_KASAN_ZERO_SHADOW); kasan_populate(AMODE31_START, AMODE31_END, POPULATE_KASAN_ZERO_SHADOW); if (IS_ENABLED(CONFIG_KASAN_VMALLOC)) { @@ -242,6 +243,8 @@ static unsigned long _pa(unsigned long addr, unsigned long size, enum populate_m return -1; case POPULATE_DIRECT: return addr; + case POPULATE_LOWCORE: + return __lowcore_pa(addr); case POPULATE_ABS_LOWCORE: return __abs_lowcore_pa(addr); case POPULATE_KERNEL: @@ -418,6 +421,7 @@ static void pgtable_populate(unsigned long addr, unsigned long end, enum populat void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned long asce_limit) { + unsigned long lowcore_address = 0; unsigned long start, end; unsigned long asce_type; unsigned long asce_bits; @@ -455,18 +459,33 @@ void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned l __arch_set_page_dat((void *)swapper_pg_dir, 1UL << CRST_ALLOC_ORDER); __arch_set_page_dat((void *)invalid_pg_dir, 1UL << CRST_ALLOC_ORDER); + if (relocate_lowcore) + lowcore_address = LOWCORE_ALT_ADDRESS; + /* * To allow prefixing the lowcore must be mapped with 4KB pages. * To prevent creation of a large page at address 0 first map * the lowcore and create the identity mapping only afterwards. */ - pgtable_populate(0, sizeof(struct lowcore), POPULATE_DIRECT); + pgtable_populate(lowcore_address, + lowcore_address + sizeof(struct lowcore), + POPULATE_LOWCORE); for_each_physmem_usable_range(i, &start, &end) { pgtable_populate((unsigned long)__identity_va(start), (unsigned long)__identity_va(end), POPULATE_IDENTITY); } - pgtable_populate(kernel_start, kernel_end, POPULATE_KERNEL); + + /* + * [kernel_start..kernel_start + TEXT_OFFSET] region is never + * accessed as per the linker script: + * + * . = TEXT_OFFSET; + * + * Therefore, skip mapping TEXT_OFFSET bytes to prevent access to + * [__kaslr_offset_phys..__kaslr_offset_phys + TEXT_OFFSET] region. + */ + pgtable_populate(kernel_start + TEXT_OFFSET, kernel_end, POPULATE_KERNEL); pgtable_populate(AMODE31_START, AMODE31_END, POPULATE_DIRECT); pgtable_populate(__abs_lowcore, __abs_lowcore + sizeof(struct lowcore), POPULATE_ABS_LOWCORE); @@ -476,13 +495,13 @@ void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned l kasan_populate_shadow(kernel_start, kernel_end); - S390_lowcore.kernel_asce.val = swapper_pg_dir | asce_bits; - S390_lowcore.user_asce = s390_invalid_asce; + get_lowcore()->kernel_asce.val = swapper_pg_dir | asce_bits; + get_lowcore()->user_asce = s390_invalid_asce; - local_ctl_load(1, &S390_lowcore.kernel_asce); - local_ctl_load(7, &S390_lowcore.user_asce); - local_ctl_load(13, &S390_lowcore.kernel_asce); + local_ctl_load(1, &get_lowcore()->kernel_asce); + local_ctl_load(7, &get_lowcore()->user_asce); + local_ctl_load(13, &get_lowcore()->kernel_asce); - init_mm.context.asce = S390_lowcore.kernel_asce.val; + init_mm.context.asce = get_lowcore()->kernel_asce.val; init_mm.pgd = init_mm_pgd; } diff --git a/arch/s390/boot/vmlinux.lds.S b/arch/s390/boot/vmlinux.lds.S index a750711d44c8..66670212a361 100644 --- a/arch/s390/boot/vmlinux.lds.S +++ b/arch/s390/boot/vmlinux.lds.S @@ -109,7 +109,12 @@ SECTIONS #ifdef CONFIG_KERNEL_UNCOMPRESSED . = ALIGN(PAGE_SIZE); . += AMODE31_SIZE; /* .amode31 section */ - . = ALIGN(1 << 20); /* _SEGMENT_SIZE */ + + /* + * Make sure the location counter is not less than TEXT_OFFSET. + * _SEGMENT_SIZE is not available, use ALIGN(1 << 20) instead. + */ + . = MAX(TEXT_OFFSET, ALIGN(1 << 20)); #else . = ALIGN(8); #endif diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index f3602414a961..ea63a7342f5f 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -55,7 +55,6 @@ CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m CONFIG_VFIO_AP=m -CONFIG_PROTECTED_VIRTUALIZATION_GUEST=y CONFIG_CMM=m CONFIG_APPLDATA_BASE=y CONFIG_S390_HYPFS_FS=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index d0d8925fdf09..d8b28ff8ff45 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -53,7 +53,6 @@ CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m CONFIG_VFIO_AP=m -CONFIG_PROTECTED_VIRTUALIZATION_GUEST=y CONFIG_CMM=m CONFIG_APPLDATA_BASE=y CONFIG_S390_HYPFS_FS=y diff --git a/arch/s390/crypto/crc32-vx.c b/arch/s390/crypto/crc32-vx.c index 74f17c905d12..89a10337e6ea 100644 --- a/arch/s390/crypto/crc32-vx.c +++ b/arch/s390/crypto/crc32-vx.c @@ -297,6 +297,7 @@ module_cpu_feature_match(S390_CPU_FEATURE_VXRS, crc_vx_mod_init); module_exit(crc_vx_mod_exit); MODULE_AUTHOR("Hendrik Brueckner <brueckner@linux.vnet.ibm.com>"); +MODULE_DESCRIPTION("CRC-32 algorithms using z/Architecture Vector Extension Facility"); MODULE_LICENSE("GPL"); MODULE_ALIAS_CRYPTO("crc32"); diff --git a/arch/s390/hypfs/hypfs_dbfs.c b/arch/s390/hypfs/hypfs_dbfs.c index 4024599eb448..0e855c5e91c5 100644 --- a/arch/s390/hypfs/hypfs_dbfs.c +++ b/arch/s390/hypfs/hypfs_dbfs.c @@ -39,7 +39,9 @@ static ssize_t dbfs_read(struct file *file, char __user *buf, return 0; df = file_inode(file)->i_private; - mutex_lock(&df->lock); + if (mutex_lock_interruptible(&df->lock)) + return -ERESTARTSYS; + data = hypfs_dbfs_data_alloc(df); if (!data) { mutex_unlock(&df->lock); diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c index 279b7bba4d43..26a009f9c49e 100644 --- a/arch/s390/hypfs/hypfs_diag.c +++ b/arch/s390/hypfs/hypfs_diag.c @@ -140,11 +140,22 @@ fail_alloc: int diag204_store(void *buf, int pages) { + unsigned long subcode; int rc; - rc = diag204((unsigned long)diag204_store_sc | - (unsigned long)diag204_get_info_type(), pages, buf); - return rc < 0 ? -EOPNOTSUPP : 0; + subcode = diag204_get_info_type(); + subcode |= diag204_store_sc; + if (diag204_has_bif()) + subcode |= DIAG204_BIF_BIT; + while (1) { + rc = diag204(subcode, pages, buf); + if (rc != -EBUSY) + break; + if (signal_pending(current)) + return -ERESTARTSYS; + schedule_timeout_interruptible(DIAG204_BUSY_WAIT); + } + return rc < 0 ? rc : 0; } struct dbfs_d204_hdr { diff --git a/arch/s390/include/asm/abs_lowcore.h b/arch/s390/include/asm/abs_lowcore.h index 6f264b79e377..d20df8c923fc 100644 --- a/arch/s390/include/asm/abs_lowcore.h +++ b/arch/s390/include/asm/abs_lowcore.h @@ -2,6 +2,7 @@ #ifndef _ASM_S390_ABS_LOWCORE_H #define _ASM_S390_ABS_LOWCORE_H +#include <asm/sections.h> #include <asm/lowcore.h> #define ABS_LOWCORE_MAP_SIZE (NR_CPUS * sizeof(struct lowcore)) @@ -24,4 +25,11 @@ static inline void put_abs_lowcore(struct lowcore *lc) put_cpu(); } +extern int __bootdata_preserved(relocate_lowcore); + +static inline int have_relocated_lowcore(void) +{ + return relocate_lowcore; +} + #endif /* _ASM_S390_ABS_LOWCORE_H */ diff --git a/arch/s390/include/asm/alternative-asm.h b/arch/s390/include/asm/alternative-asm.h deleted file mode 100644 index 608f6287ca9c..000000000000 --- a/arch/s390/include/asm/alternative-asm.h +++ /dev/null @@ -1,57 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_S390_ALTERNATIVE_ASM_H -#define _ASM_S390_ALTERNATIVE_ASM_H - -#ifdef __ASSEMBLY__ - -/* - * Issue one struct alt_instr descriptor entry (need to put it into - * the section .altinstructions, see below). This entry contains - * enough information for the alternatives patching code to patch an - * instruction. See apply_alternatives(). - */ -.macro alt_entry orig_start, orig_end, alt_start, alt_end, feature - .long \orig_start - . - .long \alt_start - . - .word \feature - .byte \orig_end - \orig_start - .org . - ( \orig_end - \orig_start ) & 1 - .org . - ( \orig_end - \orig_start ) + ( \alt_end - \alt_start ) - .org . - ( \alt_end - \alt_start ) + ( \orig_end - \orig_start ) -.endm - -/* - * Define an alternative between two instructions. If @feature is - * present, early code in apply_alternatives() replaces @oldinstr with - * @newinstr. - */ -.macro ALTERNATIVE oldinstr, newinstr, feature - .pushsection .altinstr_replacement,"ax" -770: \newinstr -771: .popsection -772: \oldinstr -773: .pushsection .altinstructions,"a" - alt_entry 772b, 773b, 770b, 771b, \feature - .popsection -.endm - -/* - * Define an alternative between two instructions. If @feature is - * present, early code in apply_alternatives() replaces @oldinstr with - * @newinstr. - */ -.macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2 - .pushsection .altinstr_replacement,"ax" -770: \newinstr1 -771: \newinstr2 -772: .popsection -773: \oldinstr -774: .pushsection .altinstructions,"a" - alt_entry 773b, 774b, 770b, 771b,\feature1 - alt_entry 773b, 774b, 771b, 772b,\feature2 - .popsection -.endm - -#endif /* __ASSEMBLY__ */ - -#endif /* _ASM_S390_ALTERNATIVE_ASM_H */ diff --git a/arch/s390/include/asm/alternative.h b/arch/s390/include/asm/alternative.h index dd93b92c3ab6..de980c938a3e 100644 --- a/arch/s390/include/asm/alternative.h +++ b/arch/s390/include/asm/alternative.h @@ -2,6 +2,58 @@ #ifndef _ASM_S390_ALTERNATIVE_H #define _ASM_S390_ALTERNATIVE_H +/* + * Each alternative comes with a 32 bit feature field: + * union { + * u32 feature; + * struct { + * u32 ctx : 4; + * u32 type : 8; + * u32 data : 20; + * }; + * } + * + * @ctx is a bitfield, where only one bit must be set. Each bit defines + * in which context an alternative is supposed to be applied to the + * kernel image: + * + * - from the decompressor before the kernel itself is executed + * - from early kernel code from within the kernel + * + * @type is a number which defines the type and with that the type + * specific alternative patching. + * + * @data is additional type specific information which defines if an + * alternative should be applied. + */ + +#define ALT_CTX_EARLY 1 +#define ALT_CTX_LATE 2 +#define ALT_CTX_ALL (ALT_CTX_EARLY | ALT_CTX_LATE) + +#define ALT_TYPE_FACILITY 0 +#define ALT_TYPE_SPEC 1 +#define ALT_TYPE_LOWCORE 2 + +#define ALT_DATA_SHIFT 0 +#define ALT_TYPE_SHIFT 20 +#define ALT_CTX_SHIFT 28 + +#define ALT_FACILITY_EARLY(facility) (ALT_CTX_EARLY << ALT_CTX_SHIFT | \ + ALT_TYPE_FACILITY << ALT_TYPE_SHIFT | \ + (facility) << ALT_DATA_SHIFT) + +#define ALT_FACILITY(facility) (ALT_CTX_LATE << ALT_CTX_SHIFT | \ + ALT_TYPE_FACILITY << ALT_TYPE_SHIFT | \ + (facility) << ALT_DATA_SHIFT) + +#define ALT_SPEC(facility) (ALT_CTX_LATE << ALT_CTX_SHIFT | \ + ALT_TYPE_SPEC << ALT_TYPE_SHIFT | \ + (facility) << ALT_DATA_SHIFT) + +#define ALT_LOWCORE (ALT_CTX_EARLY << ALT_CTX_SHIFT | \ + ALT_TYPE_LOWCORE << ALT_TYPE_SHIFT) + #ifndef __ASSEMBLY__ #include <linux/types.h> @@ -11,12 +63,30 @@ struct alt_instr { s32 instr_offset; /* original instruction */ s32 repl_offset; /* offset to replacement instruction */ - u16 facility; /* facility bit set for replacement */ + union { + u32 feature; /* feature required for replacement */ + struct { + u32 ctx : 4; /* context */ + u32 type : 8; /* type of alternative */ + u32 data : 20; /* patching information */ + }; + }; u8 instrlen; /* length of original instruction */ } __packed; -void apply_alternative_instructions(void); -void apply_alternatives(struct alt_instr *start, struct alt_instr *end); +extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; + +void __apply_alternatives(struct alt_instr *start, struct alt_instr *end, unsigned int ctx); + +static inline void apply_alternative_instructions(void) +{ + __apply_alternatives(__alt_instructions, __alt_instructions_end, ALT_CTX_LATE); +} + +static inline void apply_alternatives(struct alt_instr *start, struct alt_instr *end) +{ + __apply_alternatives(start, end, ALT_CTX_ALL); +} /* * +---------------------------------+ @@ -48,10 +118,10 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end); #define OLDINSTR(oldinstr) \ "661:\n\t" oldinstr "\n662:\n" -#define ALTINSTR_ENTRY(facility, num) \ +#define ALTINSTR_ENTRY(feature, num) \ "\t.long 661b - .\n" /* old instruction */ \ "\t.long " b_altinstr(num)"b - .\n" /* alt instruction */ \ - "\t.word " __stringify(facility) "\n" /* facility bit */ \ + "\t.long " __stringify(feature) "\n" /* feature */ \ "\t.byte " oldinstr_len "\n" /* instruction len */ \ "\t.org . - (" oldinstr_len ") & 1\n" \ "\t.org . - (" oldinstr_len ") + (" altinstr_len(num) ")\n" \ @@ -61,24 +131,24 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end); b_altinstr(num)":\n\t" altinstr "\n" e_altinstr(num) ":\n" /* alternative assembly primitive: */ -#define ALTERNATIVE(oldinstr, altinstr, facility) \ +#define ALTERNATIVE(oldinstr, altinstr, feature) \ ".pushsection .altinstr_replacement, \"ax\"\n" \ ALTINSTR_REPLACEMENT(altinstr, 1) \ ".popsection\n" \ OLDINSTR(oldinstr) \ ".pushsection .altinstructions,\"a\"\n" \ - ALTINSTR_ENTRY(facility, 1) \ + ALTINSTR_ENTRY(feature, 1) \ ".popsection\n" -#define ALTERNATIVE_2(oldinstr, altinstr1, facility1, altinstr2, facility2)\ +#define ALTERNATIVE_2(oldinstr, altinstr1, feature1, altinstr2, feature2)\ ".pushsection .altinstr_replacement, \"ax\"\n" \ ALTINSTR_REPLACEMENT(altinstr1, 1) \ ALTINSTR_REPLACEMENT(altinstr2, 2) \ ".popsection\n" \ OLDINSTR(oldinstr) \ ".pushsection .altinstructions,\"a\"\n" \ - ALTINSTR_ENTRY(facility1, 1) \ - ALTINSTR_ENTRY(facility2, 2) \ + ALTINSTR_ENTRY(feature1, 1) \ + ALTINSTR_ENTRY(feature2, 2) \ ".popsection\n" /* @@ -93,12 +163,12 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end); * For non barrier like inlines please define new variants * without volatile and memory clobber. */ -#define alternative(oldinstr, altinstr, facility) \ - asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, facility) : : : "memory") +#define alternative(oldinstr, altinstr, feature) \ + asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, feature) : : : "memory") -#define alternative_2(oldinstr, altinstr1, facility1, altinstr2, facility2) \ - asm_inline volatile(ALTERNATIVE_2(oldinstr, altinstr1, facility1, \ - altinstr2, facility2) ::: "memory") +#define alternative_2(oldinstr, altinstr1, feature1, altinstr2, feature2) \ + asm_inline volatile(ALTERNATIVE_2(oldinstr, altinstr1, feature1, \ + altinstr2, feature2) ::: "memory") /* Alternative inline assembly with input. */ #define alternative_input(oldinstr, newinstr, feature, input...) \ @@ -106,8 +176,8 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end); : : input) /* Like alternative_input, but with a single output argument */ -#define alternative_io(oldinstr, altinstr, facility, output, input...) \ - asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, facility) \ +#define alternative_io(oldinstr, altinstr, feature, output, input...) \ + asm_inline volatile(ALTERNATIVE(oldinstr, altinstr, feature) \ : output : input) /* Use this macro if more than one output parameter is needed. */ @@ -116,6 +186,56 @@ void apply_alternatives(struct alt_instr *start, struct alt_instr *end); /* Use this macro if clobbers are needed without inputs. */ #define ASM_NO_INPUT_CLOBBER(clobber...) : clobber +#else /* __ASSEMBLY__ */ + +/* + * Issue one struct alt_instr descriptor entry (need to put it into + * the section .altinstructions, see below). This entry contains + * enough information for the alternatives patching code to patch an + * instruction. See apply_alternatives(). + */ +.macro alt_entry orig_start, orig_end, alt_start, alt_end, feature + .long \orig_start - . + .long \alt_start - . + .long \feature + .byte \orig_end - \orig_start + .org . - ( \orig_end - \orig_start ) & 1 + .org . - ( \orig_end - \orig_start ) + ( \alt_end - \alt_start ) + .org . - ( \alt_end - \alt_start ) + ( \orig_end - \orig_start ) +.endm + +/* + * Define an alternative between two instructions. If @feature is + * present, early code in apply_alternatives() replaces @oldinstr with + * @newinstr. + */ +.macro ALTERNATIVE oldinstr, newinstr, feature + .pushsection .altinstr_replacement,"ax" +770: \newinstr +771: .popsection +772: \oldinstr +773: .pushsection .altinstructions,"a" + alt_entry 772b, 773b, 770b, 771b, \feature + .popsection +.endm + +/* + * Define an alternative between two instructions. If @feature is + * present, early code in apply_alternatives() replaces @oldinstr with + * @newinstr. + */ +.macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2 + .pushsection .altinstr_replacement,"ax" +770: \newinstr1 +771: \newinstr2 +772: .popsection +773: \oldinstr +774: .pushsection .altinstructions,"a" + alt_entry 773b, 774b, 770b, 771b,\feature1 + alt_entry 773b, 774b, 771b, 772b,\feature2 + .popsection +.endm + #endif /* __ASSEMBLY__ */ #endif /* _ASM_S390_ALTERNATIVE_H */ diff --git a/arch/s390/include/asm/arch_hweight.h b/arch/s390/include/asm/arch_hweight.h new file mode 100644 index 000000000000..50e23ce854e5 --- /dev/null +++ b/arch/s390/include/asm/arch_hweight.h @@ -0,0 +1,76 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _ASM_S390_ARCH_HWEIGHT_H +#define _ASM_S390_ARCH_HWEIGHT_H + +#include <linux/types.h> + +static __always_inline unsigned long popcnt_z196(unsigned long w) +{ + unsigned long cnt; + + asm volatile(".insn rrf,0xb9e10000,%[cnt],%[w],0,0" + : [cnt] "=d" (cnt) + : [w] "d" (w) + : "cc"); + return cnt; +} + +static __always_inline unsigned long popcnt_z15(unsigned long w) +{ + unsigned long cnt; + + asm volatile(".insn rrf,0xb9e10000,%[cnt],%[w],8,0" + : [cnt] "=d" (cnt) + : [w] "d" (w) + : "cc"); + return cnt; +} + +static __always_inline unsigned long __arch_hweight64(__u64 w) +{ + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z15_FEATURES)) + return popcnt_z15(w); + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z196_FEATURES)) { + w = popcnt_z196(w); + w += w >> 32; + w += w >> 16; + w += w >> 8; + return w & 0xff; + } + return __sw_hweight64(w); +} + +static __always_inline unsigned int __arch_hweight32(unsigned int w) +{ + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z15_FEATURES)) + return popcnt_z15(w); + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z196_FEATURES)) { + w = popcnt_z196(w); + w += w >> 16; + w += w >> 8; + return w & 0xff; + } + return __sw_hweight32(w); +} + +static __always_inline unsigned int __arch_hweight16(unsigned int w) +{ + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z15_FEATURES)) + return popcnt_z15((unsigned short)w); + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z196_FEATURES)) { + w = popcnt_z196(w); + w += w >> 8; + return w & 0xff; + } + return __sw_hweight16(w); +} + +static __always_inline unsigned int __arch_hweight8(unsigned int w) +{ + if (IS_ENABLED(CONFIG_HAVE_MARCH_Z196_FEATURES)) + return popcnt_z196((unsigned char)w); + return __sw_hweight8(w); +} + +#endif /* _ASM_S390_ARCH_HWEIGHT_H */ diff --git a/arch/s390/include/asm/atomic_ops.h b/arch/s390/include/asm/atomic_ops.h index 7fa5f96a553a..742c7919cbcd 100644 --- a/arch/s390/include/asm/atomic_ops.h +++ b/arch/s390/include/asm/atomic_ops.h @@ -8,21 +8,29 @@ #ifndef __ARCH_S390_ATOMIC_OPS__ #define __ARCH_S390_ATOMIC_OPS__ +#include <linux/limits.h> + static __always_inline int __atomic_read(const atomic_t *v) { int c; asm volatile( - " l %0,%1\n" - : "=d" (c) : "R" (v->counter)); + " l %[c],%[counter]\n" + : [c] "=d" (c) : [counter] "R" (v->counter)); return c; } static __always_inline void __atomic_set(atomic_t *v, int i) { - asm volatile( - " st %1,%0\n" - : "=R" (v->counter) : "d" (i)); + if (__builtin_constant_p(i) && i >= S16_MIN && i <= S16_MAX) { + asm volatile( + " mvhi %[counter], %[i]\n" + : [counter] "=Q" (v->counter) : [i] "K" (i)); + } else { + asm volatile( + " st %[i],%[counter]\n" + : [counter] "=R" (v->counter) : [i] "d" (i)); + } } static __always_inline s64 __atomic64_read(const atomic64_t *v) @@ -30,16 +38,22 @@ static __always_inline s64 __atomic64_read(const atomic64_t *v) s64 c; asm volatile( - " lg %0,%1\n" - : "=d" (c) : "RT" (v->counter)); + " lg %[c],%[counter]\n" + : [c] "=d" (c) : [counter] "RT" (v->counter)); return c; } static __always_inline void __atomic64_set(atomic64_t *v, s64 i) { - asm volatile( - " stg %1,%0\n" - : "=RT" (v->counter) : "d" (i)); + if (__builtin_constant_p(i) && i >= S16_MIN && i <= S16_MAX) { + asm volatile( + " mvghi %[counter], %[i]\n" + : [counter] "=Q" (v->counter) : [i] "K" (i)); + } else { + asm volatile( + " stg %[i],%[counter]\n" + : [counter] "=RT" (v->counter) : [i] "d" (i)); + } } #ifdef CONFIG_HAVE_MARCH_Z196_FEATURES @@ -164,26 +178,55 @@ static __always_inline int __atomic_cmpxchg(int *ptr, int old, int new) return old; } +static __always_inline long __atomic64_cmpxchg(long *ptr, long old, long new) +{ + asm volatile( + " csg %[old],%[new],%[ptr]" + : [old] "+d" (old), [ptr] "+QS" (*ptr) + : [new] "d" (new) + : "cc", "memory"); + return old; +} + +/* GCC versions before 14.2.0 may die with an ICE in some configurations. */ +#if defined(__GCC_ASM_FLAG_OUTPUTS__) && !(IS_ENABLED(CONFIG_CC_IS_GCC) && (GCC_VERSION < 140200)) + static __always_inline bool __atomic_cmpxchg_bool(int *ptr, int old, int new) { - int old_expected = old; + int cc; asm volatile( " cs %[old],%[new],%[ptr]" - : [old] "+d" (old), [ptr] "+Q" (*ptr) + : [old] "+d" (old), [ptr] "+Q" (*ptr), "=@cc" (cc) : [new] "d" (new) - : "cc", "memory"); - return old == old_expected; + : "memory"); + return cc == 0; } -static __always_inline long __atomic64_cmpxchg(long *ptr, long old, long new) +static __always_inline bool __atomic64_cmpxchg_bool(long *ptr, long old, long new) { + int cc; + asm volatile( " csg %[old],%[new],%[ptr]" - : [old] "+d" (old), [ptr] "+QS" (*ptr) + : [old] "+d" (old), [ptr] "+QS" (*ptr), "=@cc" (cc) + : [new] "d" (new) + : "memory"); + return cc == 0; +} + +#else /* __GCC_ASM_FLAG_OUTPUTS__ */ + +static __always_inline bool __atomic_cmpxchg_bool(int *ptr, int old, int new) +{ + int old_expected = old; + + asm volatile( + " cs %[old],%[new],%[ptr]" + : [old] "+d" (old), [ptr] "+Q" (*ptr) : [new] "d" (new) : "cc", "memory"); - return old; + return old == old_expected; } static __always_inline bool __atomic64_cmpxchg_bool(long *ptr, long old, long new) @@ -198,4 +241,6 @@ static __always_inline bool __atomic64_cmpxchg_bool(long *ptr, long old, long ne return old == old_expected; } +#endif /* __GCC_ASM_FLAG_OUTPUTS__ */ + #endif /* __ARCH_S390_ATOMIC_OPS__ */ diff --git a/arch/s390/include/asm/bitops.h b/arch/s390/include/asm/bitops.h index c467dffa8c12..54a079cd39ed 100644 --- a/arch/s390/include/asm/bitops.h +++ b/arch/s390/include/asm/bitops.h @@ -379,8 +379,9 @@ static inline int fls(unsigned int word) return fls64(word); } +#include <asm/arch_hweight.h> +#include <asm-generic/bitops/const_hweight.h> #include <asm-generic/bitops/ffz.h> -#include <asm-generic/bitops/hweight.h> #include <asm-generic/bitops/sched.h> #include <asm-generic/bitops/le.h> #include <asm-generic/bitops/ext2-atomic-setbit.h> diff --git a/arch/s390/include/asm/ccwdev.h b/arch/s390/include/asm/ccwdev.h index 436365ff6c19..e3afcece375e 100644 --- a/arch/s390/include/asm/ccwdev.h +++ b/arch/s390/include/asm/ccwdev.h @@ -210,7 +210,7 @@ extern void ccw_device_get_id(struct ccw_device *, struct ccw_dev_id *); #define get_ccwdev_lock(x) (x)->ccwlock #define to_ccwdev(n) container_of(n, struct ccw_device, dev) -#define to_ccwdrv(n) container_of(n, struct ccw_driver, driver) +#define to_ccwdrv(n) container_of_const(n, struct ccw_driver, driver) extern struct ccw_device *ccw_device_create_console(struct ccw_driver *); extern void ccw_device_destroy_console(struct ccw_device *); diff --git a/arch/s390/include/asm/checksum.h b/arch/s390/include/asm/checksum.h index b89159591ca0..46f5c9660616 100644 --- a/arch/s390/include/asm/checksum.h +++ b/arch/s390/include/asm/checksum.h @@ -13,6 +13,7 @@ #define _S390_CHECKSUM_H #include <linux/instrumented.h> +#include <linux/kmsan-checks.h> #include <linux/in6.h> static inline __wsum cksm(const void *buff, int len, __wsum sum) @@ -23,6 +24,7 @@ static inline __wsum cksm(const void *buff, int len, __wsum sum) }; instrument_read(buff, len); + kmsan_check_memory(buff, len); asm volatile("\n" "0: cksm %[sum],%[rp]\n" " jo 0b\n" diff --git a/arch/s390/include/asm/cpacf.h b/arch/s390/include/asm/cpacf.h index c786538e397c..dae8843b164f 100644 --- a/arch/s390/include/asm/cpacf.h +++ b/arch/s390/include/asm/cpacf.h @@ -12,6 +12,7 @@ #define _ASM_S390_CPACF_H #include <asm/facility.h> +#include <linux/kmsan-checks.h> /* * Instruction opcodes for the CPACF instructions @@ -542,6 +543,8 @@ static inline void cpacf_trng(u8 *ucbuf, unsigned long ucbuf_len, : [ucbuf] "+&d" (u.pair), [cbuf] "+&d" (c.pair) : [fc] "K" (CPACF_PRNO_TRNG), [opc] "i" (CPACF_PRNO) : "cc", "memory", "0"); + kmsan_unpoison_memory(ucbuf, ucbuf_len); + kmsan_unpoison_memory(cbuf, cbuf_len); } /** diff --git a/arch/s390/include/asm/cpu_mf.h b/arch/s390/include/asm/cpu_mf.h index a0de5b9b02ea..9e4bbc3e53f8 100644 --- a/arch/s390/include/asm/cpu_mf.h +++ b/arch/s390/include/asm/cpu_mf.h @@ -10,6 +10,7 @@ #define _ASM_S390_CPU_MF_H #include <linux/errno.h> +#include <linux/kmsan-checks.h> #include <asm/asm-extable.h> #include <asm/facility.h> @@ -239,6 +240,11 @@ static __always_inline int stcctm(enum stcctm_ctr_set set, u64 range, u64 *dest) : "=d" (cc) : "Q" (*dest), "d" (range), "i" (set) : "cc", "memory"); + /* + * If cc == 2, less than RANGE counters are stored, but it's not easy + * to tell how many. Always unpoison the whole range for simplicity. + */ + kmsan_unpoison_memory(dest, range * sizeof(u64)); return cc; } diff --git a/arch/s390/include/asm/current.h b/arch/s390/include/asm/current.h index 68f84315277c..d03a922c641e 100644 --- a/arch/s390/include/asm/current.h +++ b/arch/s390/include/asm/current.h @@ -14,6 +14,6 @@ struct task_struct; -#define current ((struct task_struct *const)S390_lowcore.current_task) +#define current ((struct task_struct *const)get_lowcore()->current_task) #endif /* !(_S390_CURRENT_H) */ diff --git a/arch/s390/include/asm/dat-bits.h b/arch/s390/include/asm/dat-bits.h new file mode 100644 index 000000000000..8d65eec2f124 --- /dev/null +++ b/arch/s390/include/asm/dat-bits.h @@ -0,0 +1,170 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * DAT table and related structures + * + * Copyright IBM Corp. 2024 + * + */ + +#ifndef _S390_DAT_BITS_H +#define _S390_DAT_BITS_H + +union asce { + unsigned long val; + struct { + unsigned long rsto: 52;/* Region- or Segment-Table Origin */ + unsigned long : 2; + unsigned long g : 1; /* Subspace Group control */ + unsigned long p : 1; /* Private Space control */ + unsigned long s : 1; /* Storage-Alteration-Event control */ + unsigned long x : 1; /* Space-Switch-Event control */ + unsigned long r : 1; /* Real-Space control */ + unsigned long : 1; + unsigned long dt : 2; /* Designation-Type control */ + unsigned long tl : 2; /* Region- or Segment-Table Length */ + }; +}; + +enum { + ASCE_TYPE_SEGMENT = 0, + ASCE_TYPE_REGION3 = 1, + ASCE_TYPE_REGION2 = 2, + ASCE_TYPE_REGION1 = 3 +}; + +union region1_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Second-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Second-Table Length */ + }; +}; + +union region2_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Third-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Third-Table Length */ + }; +}; + +struct region3_table_entry_fc0 { + unsigned long sto: 52;/* Segment-Table Origin */ + unsigned long : 1; + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Segment-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Segment-Table Length */ +}; + +struct region3_table_entry_fc1 { + unsigned long rfaa: 33;/* Region-Frame Absolute Address */ + unsigned long : 14; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc : 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long iep : 1; /* Instruction-Execution-Protection */ + unsigned long : 2; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union region3_table_entry { + unsigned long val; + struct region3_table_entry_fc0 fc0; + struct region3_table_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc: 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr: 1; /* Common-Region Bit */ + unsigned long tt: 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +struct segment_table_entry_fc0 { + unsigned long pto: 53;/* Page-Table Origin */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 3; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +struct segment_table_entry_fc1 { + unsigned long sfaa: 44;/* Segment-Frame Absolute Address */ + unsigned long : 3; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc : 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long iep : 1; /* Instruction-Execution-Protection */ + unsigned long : 2; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union segment_table_entry { + unsigned long val; + struct segment_table_entry_fc0 fc0; + struct segment_table_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc: 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs: 1; /* Common-Segment Bit */ + unsigned long tt: 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +union page_table_entry { + unsigned long val; + struct { + unsigned long pfra: 52;/* Page-Frame Real Address */ + unsigned long z : 1; /* Zero Bit */ + unsigned long i : 1; /* Page-Invalid Bit */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long iep : 1; /* Instruction-Execution-Protection */ + unsigned long : 8; + }; +}; + +enum { + TABLE_TYPE_SEGMENT = 0, + TABLE_TYPE_REGION3 = 1, + TABLE_TYPE_REGION2 = 2, + TABLE_TYPE_REGION1 = 3 +}; + +#endif /* _S390_DAT_BITS_H */ diff --git a/arch/s390/include/asm/diag.h b/arch/s390/include/asm/diag.h index 20b94220113b..c0d43512f4fc 100644 --- a/arch/s390/include/asm/diag.h +++ b/arch/s390/include/asm/diag.h @@ -12,6 +12,7 @@ #include <linux/if_ether.h> #include <linux/percpu.h> #include <asm/asm-extable.h> +#include <asm/sclp.h> #include <asm/cio.h> enum diag_stat_enum { @@ -117,6 +118,8 @@ enum diag204_sc { }; #define DIAG204_SUBCODE_MASK 0xffff +#define DIAG204_BIF_BIT 0x80000000 +#define DIAG204_BUSY_WAIT (HZ / 10) /* The two available diag 204 data formats */ enum diag204_format { @@ -326,6 +329,11 @@ union diag318_info { }; }; +static inline bool diag204_has_bif(void) +{ + return sclp.has_diag204_bif; +} + int diag204(unsigned long subcode, unsigned long size, void *addr); int diag224(void *ptr); int diag26c(void *req, void *resp, enum diag26c_sc subcode); diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h index 70a30ae258b7..8f2c23cc52b6 100644 --- a/arch/s390/include/asm/elf.h +++ b/arch/s390/include/asm/elf.h @@ -91,6 +91,14 @@ /* Keep this the last entry. */ #define R_390_NUM 61 +/* + * HWCAP flags - for AT_HWCAP + * + * Bits 32-63 are reserved for use by libc. + * Bit 31 is reserved and will be used by libc to determine if a second + * argument is passed to IFUNC resolvers. This will be implemented when + * there is a need for AT_HWCAP2. + */ enum { HWCAP_NR_ESAN3 = 0, HWCAP_NR_ZARCH = 1, diff --git a/arch/s390/include/asm/facility.h b/arch/s390/include/asm/facility.h index 796007125dff..b7d234838a36 100644 --- a/arch/s390/include/asm/facility.h +++ b/arch/s390/include/asm/facility.h @@ -20,7 +20,6 @@ #define MAX_FACILITY_BIT (sizeof(stfle_fac_list) * 8) extern u64 stfle_fac_list[16]; -extern u64 alt_stfle_fac_list[16]; static inline void __set_facility(unsigned long nr, void *facilities) { @@ -92,8 +91,8 @@ static inline void __stfle(u64 *stfle_fac_list, int size) asm volatile( " stfl 0(0)\n" - : "=m" (S390_lowcore.stfl_fac_list)); - stfl_fac_list = S390_lowcore.stfl_fac_list; + : "=m" (get_lowcore()->stfl_fac_list)); + stfl_fac_list = get_lowcore()->stfl_fac_list; memcpy(stfle_fac_list, &stfl_fac_list, 4); nr = 4; /* bytes stored by stfl */ if (stfl_fac_list & 0x01000000) { diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h index 77e479d44f1e..fbadca645af7 100644 --- a/arch/s390/include/asm/ftrace.h +++ b/arch/s390/include/asm/ftrace.h @@ -2,7 +2,6 @@ #ifndef _ASM_S390_FTRACE_H #define _ASM_S390_FTRACE_H -#define HAVE_FUNCTION_GRAPH_RET_ADDR_PTR #define ARCH_SUPPORTS_FTRACE_OPS 1 #define MCOUNT_INSN_SIZE 6 diff --git a/arch/s390/include/asm/hardirq.h b/arch/s390/include/asm/hardirq.h index 58668ffb5488..a5b45388c91f 100644 --- a/arch/s390/include/asm/hardirq.h +++ b/arch/s390/include/asm/hardirq.h @@ -13,9 +13,9 @@ #include <asm/lowcore.h> -#define local_softirq_pending() (S390_lowcore.softirq_pending) -#define set_softirq_pending(x) (S390_lowcore.softirq_pending = (x)) -#define or_softirq_pending(x) (S390_lowcore.softirq_pending |= (x)) +#define local_softirq_pending() (get_lowcore()->softirq_pending) +#define set_softirq_pending(x) (get_lowcore()->softirq_pending = (x)) +#define or_softirq_pending(x) (get_lowcore()->softirq_pending |= (x)) #define __ARCH_IRQ_STAT #define __ARCH_IRQ_EXIT_IRQS_DISABLED diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h index ce5f4fe8be4d..cf1b5d6fb1a6 100644 --- a/arch/s390/include/asm/hugetlb.h +++ b/arch/s390/include/asm/hugetlb.h @@ -19,7 +19,7 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte, unsigned long sz); void __set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte); -pte_t huge_ptep_get(pte_t *ptep); +pte_t huge_ptep_get(struct mm_struct *mm, unsigned long addr, pte_t *ptep); pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep); @@ -64,7 +64,7 @@ static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, pte_t pte, int dirty) { - int changed = !pte_same(huge_ptep_get(ptep), pte); + int changed = !pte_same(huge_ptep_get(vma->vm_mm, addr, ptep), pte); if (changed) { huge_ptep_get_and_clear(vma->vm_mm, addr, ptep); __set_huge_pte_at(vma->vm_mm, addr, ptep, pte); diff --git a/arch/s390/include/asm/irqflags.h b/arch/s390/include/asm/irqflags.h index 02427b205c11..bcab456dfb80 100644 --- a/arch/s390/include/asm/irqflags.h +++ b/arch/s390/include/asm/irqflags.h @@ -37,12 +37,18 @@ static __always_inline void __arch_local_irq_ssm(unsigned long flags) asm volatile("ssm %0" : : "Q" (flags) : "memory"); } -static __always_inline unsigned long arch_local_save_flags(void) +#ifdef CONFIG_KMSAN +#define arch_local_irq_attributes noinline notrace __no_sanitize_memory __maybe_unused +#else +#define arch_local_irq_attributes __always_inline +#endif + +static arch_local_irq_attributes unsigned long arch_local_save_flags(void) { return __arch_local_irq_stnsm(0xff); } -static __always_inline unsigned long arch_local_irq_save(void) +static arch_local_irq_attributes unsigned long arch_local_irq_save(void) { return __arch_local_irq_stnsm(0xfc); } @@ -52,7 +58,12 @@ static __always_inline void arch_local_irq_disable(void) arch_local_irq_save(); } -static __always_inline void arch_local_irq_enable(void) +static arch_local_irq_attributes void arch_local_irq_enable_external(void) +{ + __arch_local_irq_stosm(0x01); +} + +static arch_local_irq_attributes void arch_local_irq_enable(void) { __arch_local_irq_stosm(0x03); } diff --git a/arch/s390/include/asm/kmsan.h b/arch/s390/include/asm/kmsan.h new file mode 100644 index 000000000000..f73e181d09ae --- /dev/null +++ b/arch/s390/include/asm/kmsan.h @@ -0,0 +1,59 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_S390_KMSAN_H +#define _ASM_S390_KMSAN_H + +#include <asm/lowcore.h> +#include <asm/page.h> +#include <linux/kmsan.h> +#include <linux/mmzone.h> +#include <linux/stddef.h> + +#ifndef MODULE + +static inline bool is_lowcore_addr(void *addr) +{ + return addr >= (void *)get_lowcore() && + addr < (void *)(get_lowcore() + 1); +} + +static inline void *arch_kmsan_get_meta_or_null(void *addr, bool is_origin) +{ + if (is_lowcore_addr(addr)) { + /* + * Different lowcores accessed via S390_lowcore are described + * by the same struct page. Resolve the prefix manually in + * order to get a distinct struct page. + */ + addr += (void *)lowcore_ptr[raw_smp_processor_id()] - + (void *)get_lowcore(); + if (KMSAN_WARN_ON(is_lowcore_addr(addr))) + return NULL; + return kmsan_get_metadata(addr, is_origin); + } + return NULL; +} + +static inline bool kmsan_virt_addr_valid(void *addr) +{ + bool ret; + + /* + * pfn_valid() relies on RCU, and may call into the scheduler on exiting + * the critical section. However, this would result in recursion with + * KMSAN. Therefore, disable preemption here, and re-enable preemption + * below while suppressing reschedules to avoid recursion. + * + * Note, this sacrifices occasionally breaking scheduling guarantees. + * Although, a kernel compiled with KMSAN has already given up on any + * performance guarantees due to being heavily instrumented. + */ + preempt_disable(); + ret = virt_addr_valid(addr); + preempt_enable_no_resched(); + + return ret; +} + +#endif /* !MODULE */ + +#endif /* _ASM_S390_KMSAN_H */ diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 9281063636a7..8e77afbed58e 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -15,7 +15,6 @@ #include <linux/hrtimer.h> #include <linux/interrupt.h> #include <linux/kvm_types.h> -#include <linux/kvm_host.h> #include <linux/kvm.h> #include <linux/seqlock.h> #include <linux/module.h> @@ -1030,11 +1029,12 @@ void kvm_arch_crypto_clear_masks(struct kvm *kvm); void kvm_arch_crypto_set_masks(struct kvm *kvm, unsigned long *apm, unsigned long *aqm, unsigned long *adm); -int __sie64a(phys_addr_t sie_block_phys, struct kvm_s390_sie_block *sie_block, u64 *rsa); +int __sie64a(phys_addr_t sie_block_phys, struct kvm_s390_sie_block *sie_block, u64 *rsa, + unsigned long gasce); -static inline int sie64a(struct kvm_s390_sie_block *sie_block, u64 *rsa) +static inline int sie64a(struct kvm_s390_sie_block *sie_block, u64 *rsa, unsigned long gasce) { - return __sie64a(virt_to_phys(sie_block), sie_block, rsa); + return __sie64a(virt_to_phys(sie_block), sie_block, rsa, gasce); } extern char sie_exit; @@ -1046,7 +1046,6 @@ extern int kvm_s390_gisc_register(struct kvm *kvm, u32 gisc); extern int kvm_s390_gisc_unregister(struct kvm *kvm, u32 gisc); static inline void kvm_arch_sync_events(struct kvm *kvm) {} -static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {} static inline void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *slot) {} static inline void kvm_arch_memslots_updated(struct kvm *kvm, u64 gen) {} diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index 8c5f16857539..183ac29afaf8 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -14,10 +14,15 @@ #include <asm/ctlreg.h> #include <asm/cpu.h> #include <asm/types.h> +#include <asm/alternative.h> #define LC_ORDER 1 #define LC_PAGES 2 +#define LOWCORE_ALT_ADDRESS _AC(0x70000, UL) + +#ifndef __ASSEMBLY__ + struct pgm_tdb { u64 data[32]; }; @@ -97,8 +102,7 @@ struct lowcore { __u64 save_area_async[8]; /* 0x0240 */ __u64 save_area_restart[1]; /* 0x0280 */ - /* CPU flags. */ - __u64 cpu_flags; /* 0x0288 */ + __u64 pcpu; /* 0x0288 */ /* Return psws. */ psw_t return_psw; /* 0x0290 */ @@ -213,7 +217,17 @@ struct lowcore { __u8 pad_0x1900[0x2000-0x1900]; /* 0x1900 */ } __packed __aligned(8192); -#define S390_lowcore (*((struct lowcore *) 0)) +static __always_inline struct lowcore *get_lowcore(void) +{ + struct lowcore *lc; + + if (__is_defined(__DECOMPRESSOR)) + return NULL; + asm(ALTERNATIVE("llilh %[lc],0", "llilh %[lc],%[alt]", ALT_LOWCORE) + : [lc] "=d" (lc) + : [alt] "i" (LOWCORE_ALT_ADDRESS >> 16)); + return lc; +} extern struct lowcore *lowcore_ptr[]; @@ -222,4 +236,19 @@ static inline void set_prefix(__u32 address) asm volatile("spx %0" : : "Q" (address) : "memory"); } +#else /* __ASSEMBLY__ */ + +.macro GET_LC reg + ALTERNATIVE "llilh \reg,0", \ + __stringify(llilh \reg, LOWCORE_ALT_ADDRESS >> 16), \ + ALT_LOWCORE +.endm + +.macro STMG_LC start, end, savearea + ALTERNATIVE "stmg \start, \end, \savearea", \ + __stringify(stmg \start, \end, LOWCORE_ALT_ADDRESS + \savearea), \ + ALT_LOWCORE +.endm + +#endif /* __ASSEMBLY__ */ #endif /* _ASM_S390_LOWCORE_H */ diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index a7789a9f6218..d56eb0a1f37b 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -76,9 +76,9 @@ static inline void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct * int cpu = smp_processor_id(); if (next == &init_mm) - S390_lowcore.user_asce = s390_invalid_asce; + get_lowcore()->user_asce = s390_invalid_asce; else - S390_lowcore.user_asce.val = next->context.asce; + get_lowcore()->user_asce.val = next->context.asce; cpumask_set_cpu(cpu, &next->context.cpu_attach_mask); /* Clear previous user-ASCE from CR7 */ local_ctl_load(7, &s390_invalid_asce); @@ -111,7 +111,7 @@ static inline void finish_arch_post_lock_switch(void) __tlb_flush_mm_lazy(mm); preempt_enable(); } - local_ctl_load(7, &S390_lowcore.user_asce); + local_ctl_load(7, &get_lowcore()->user_asce); } #define activate_mm activate_mm @@ -120,7 +120,7 @@ static inline void activate_mm(struct mm_struct *prev, { switch_mm(prev, next, current); cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); - local_ctl_load(7, &S390_lowcore.user_asce); + local_ctl_load(7, &get_lowcore()->user_asce); } #include <asm-generic/mmu_context.h> diff --git a/arch/s390/include/asm/nospec-branch.h b/arch/s390/include/asm/nospec-branch.h index b9c1f3cae842..192835a3e24d 100644 --- a/arch/s390/include/asm/nospec-branch.h +++ b/arch/s390/include/asm/nospec-branch.h @@ -5,8 +5,17 @@ #ifndef __ASSEMBLY__ #include <linux/types.h> +#include <asm/facility.h> extern int nospec_disable; +extern int nobp; + +static inline bool nobp_enabled(void) +{ + if (__is_defined(__DECOMPRESSOR)) + return false; + return nobp && test_facility(82); +} void nospec_init_branches(void); void nospec_auto_detect(void); diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h index 224ff9d433ea..16e4caa931f1 100644 --- a/arch/s390/include/asm/page.h +++ b/arch/s390/include/asm/page.h @@ -162,6 +162,7 @@ static inline int page_reset_referenced(unsigned long addr) #define _PAGE_ACC_BITS 0xf0 /* HW access control bits */ struct page; +struct folio; void arch_free_page(struct page *page, int order); void arch_alloc_page(struct page *page, int order); @@ -173,10 +174,10 @@ static inline int devmem_is_allowed(unsigned long pfn) #define HAVE_ARCH_FREE_PAGE #define HAVE_ARCH_ALLOC_PAGE -#if IS_ENABLED(CONFIG_PGSTE) +int arch_make_folio_accessible(struct folio *folio); +#define HAVE_ARCH_MAKE_FOLIO_ACCESSIBLE int arch_make_page_accessible(struct page *page); #define HAVE_ARCH_MAKE_PAGE_ACCESSIBLE -#endif struct vm_layout { unsigned long kaslr_offset; @@ -247,7 +248,9 @@ static inline unsigned long __phys_addr(unsigned long x, bool is_31bit) #define pfn_to_phys(pfn) ((pfn) << PAGE_SHIFT) #define phys_to_page(phys) pfn_to_page(phys_to_pfn(phys)) +#define phys_to_folio(phys) page_folio(phys_to_page(phys)) #define page_to_phys(page) pfn_to_phys(page_to_pfn(page)) +#define folio_to_phys(page) pfn_to_phys(folio_pfn(folio)) static inline void *pfn_to_virt(unsigned long pfn) { @@ -276,8 +279,9 @@ static inline unsigned long virt_to_pfn(const void *kaddr) #define AMODE31_SIZE (3 * PAGE_SIZE) #define KERNEL_IMAGE_SIZE (512 * 1024 * 1024) -#define __START_KERNEL 0x100000 #define __NO_KASLR_START_KERNEL CONFIG_KERNEL_IMAGE_BASE #define __NO_KASLR_END_KERNEL (__NO_KASLR_START_KERNEL + KERNEL_IMAGE_SIZE) +#define TEXT_OFFSET 0x100000 + #endif /* _S390_PAGE_H */ diff --git a/arch/s390/include/asm/pai.h b/arch/s390/include/asm/pai.h index 3f609565734b..25f2077ba3c9 100644 --- a/arch/s390/include/asm/pai.h +++ b/arch/s390/include/asm/pai.h @@ -55,11 +55,11 @@ static __always_inline void pai_kernel_enter(struct pt_regs *regs) return; if (!static_branch_unlikely(&pai_key)) return; - if (!S390_lowcore.ccd) + if (!get_lowcore()->ccd) return; if (!user_mode(regs)) return; - WRITE_ONCE(S390_lowcore.ccd, S390_lowcore.ccd | PAI_CRYPTO_KERNEL_OFFSET); + WRITE_ONCE(get_lowcore()->ccd, get_lowcore()->ccd | PAI_CRYPTO_KERNEL_OFFSET); } static __always_inline void pai_kernel_exit(struct pt_regs *regs) @@ -68,18 +68,15 @@ static __always_inline void pai_kernel_exit(struct pt_regs *regs) return; if (!static_branch_unlikely(&pai_key)) return; - if (!S390_lowcore.ccd) + if (!get_lowcore()->ccd) return; if (!user_mode(regs)) return; - WRITE_ONCE(S390_lowcore.ccd, S390_lowcore.ccd & ~PAI_CRYPTO_KERNEL_OFFSET); + WRITE_ONCE(get_lowcore()->ccd, get_lowcore()->ccd & ~PAI_CRYPTO_KERNEL_OFFSET); } -enum paievt_mode { - PAI_MODE_NONE, - PAI_MODE_SAMPLING, - PAI_MODE_COUNTING, -}; - #define PAI_SAVE_AREA(x) ((x)->hw.event_base) +#define PAI_CPU_MASK(x) ((x)->hw.addr_filters) +#define PAI_SWLIST(x) (&(x)->hw.tp_list) + #endif diff --git a/arch/s390/include/asm/percpu.h b/arch/s390/include/asm/percpu.h index 264095dd84bc..89a28740b6ab 100644 --- a/arch/s390/include/asm/percpu.h +++ b/arch/s390/include/asm/percpu.h @@ -9,7 +9,7 @@ * s390 uses its own implementation for per cpu data, the offset of * the cpu local data area is cached in the cpu's lowcore memory. */ -#define __my_cpu_offset S390_lowcore.percpu_offset +#define __my_cpu_offset get_lowcore()->percpu_offset /* * For 64 bit module code, the module may be more than 4G above the diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 70b6ee557eb2..3fa280d0672a 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -107,6 +107,18 @@ static inline int is_module_addr(void *addr) return 1; } +#ifdef CONFIG_KMSAN +#define KMSAN_VMALLOC_SIZE (VMALLOC_END - VMALLOC_START) +#define KMSAN_VMALLOC_SHADOW_START VMALLOC_END +#define KMSAN_VMALLOC_SHADOW_END (KMSAN_VMALLOC_SHADOW_START + KMSAN_VMALLOC_SIZE) +#define KMSAN_VMALLOC_ORIGIN_START KMSAN_VMALLOC_SHADOW_END +#define KMSAN_VMALLOC_ORIGIN_END (KMSAN_VMALLOC_ORIGIN_START + KMSAN_VMALLOC_SIZE) +#define KMSAN_MODULES_SHADOW_START KMSAN_VMALLOC_ORIGIN_END +#define KMSAN_MODULES_SHADOW_END (KMSAN_MODULES_SHADOW_START + MODULES_LEN) +#define KMSAN_MODULES_ORIGIN_START KMSAN_MODULES_SHADOW_END +#define KMSAN_MODULES_ORIGIN_END (KMSAN_MODULES_ORIGIN_START + MODULES_LEN) +#endif + #ifdef CONFIG_RANDOMIZE_BASE #define KASLR_LEN (1UL << 31) #else @@ -609,7 +621,15 @@ static inline void csp(unsigned int *ptr, unsigned int old, unsigned int new) : "cc"); } -static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new) +/** + * cspg() - Compare and Swap and Purge (CSPG) + * @ptr: Pointer to the value to be exchanged + * @old: The expected old value + * @new: The new value + * + * Return: True if compare and swap was successful, otherwise false. + */ +static inline bool cspg(unsigned long *ptr, unsigned long old, unsigned long new) { union register_pair r1 = { .even = old, .odd = new, }; unsigned long address = (unsigned long)ptr | 1; @@ -619,6 +639,7 @@ static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new : [r1] "+&d" (r1.pair), "+m" (*ptr) : [address] "d" (address) : "cc"); + return old == r1.even; } #define CRDTE_DTT_PAGE 0x00UL @@ -627,7 +648,18 @@ static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new #define CRDTE_DTT_REGION2 0x18UL #define CRDTE_DTT_REGION1 0x1cUL -static inline void crdte(unsigned long old, unsigned long new, +/** + * crdte() - Compare and Replace DAT Table Entry + * @old: The expected old value + * @new: The new value + * @table: Pointer to the value to be exchanged + * @dtt: Table type of the table to be exchanged + * @address: The address mapped by the entry to be replaced + * @asce: The ASCE of this entry + * + * Return: True if compare and replace was successful, otherwise false. + */ +static inline bool crdte(unsigned long old, unsigned long new, unsigned long *table, unsigned long dtt, unsigned long address, unsigned long asce) { @@ -638,6 +670,7 @@ static inline void crdte(unsigned long old, unsigned long new, : [r1] "+&d" (r1.pair) : [r2] "d" (r2.pair), [asce] "a" (asce) : "memory", "cc"); + return old == r1.even; } /* @@ -1167,7 +1200,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, res = ptep_xchg_lazy(mm, addr, ptep, __pte(_PAGE_INVALID)); /* At this point the reference through the mapping is still present */ if (mm_is_protected(mm) && pte_present(res)) - uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + uv_convert_from_secure_pte(res); return res; } @@ -1185,7 +1218,7 @@ static inline pte_t ptep_clear_flush(struct vm_area_struct *vma, res = ptep_xchg_direct(vma->vm_mm, addr, ptep, __pte(_PAGE_INVALID)); /* At this point the reference through the mapping is still present */ if (mm_is_protected(vma->vm_mm) && pte_present(res)) - uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + uv_convert_from_secure_pte(res); return res; } @@ -1217,14 +1250,14 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, * The notifier should have destroyed all protected vCPUs at this * point, so the destroy should be successful. */ - if (full && !uv_destroy_owned_page(pte_val(res) & PAGE_MASK)) + if (full && !uv_destroy_pte(res)) return res; /* * If something went wrong and the page could not be destroyed, or * if this is not a mm teardown, the slower export is used as * fallback instead. */ - uv_convert_owned_from_secure(pte_val(res) & PAGE_MASK); + uv_convert_from_secure_pte(res); return res; } diff --git a/arch/s390/include/asm/preempt.h b/arch/s390/include/asm/preempt.h index 0e3da500e98c..3ae5f31c665d 100644 --- a/arch/s390/include/asm/preempt.h +++ b/arch/s390/include/asm/preempt.h @@ -14,7 +14,7 @@ static __always_inline int preempt_count(void) { - return READ_ONCE(S390_lowcore.preempt_count) & ~PREEMPT_NEED_RESCHED; + return READ_ONCE(get_lowcore()->preempt_count) & ~PREEMPT_NEED_RESCHED; } static __always_inline void preempt_count_set(int pc) @@ -22,26 +22,26 @@ static __always_inline void preempt_count_set(int pc) int old, new; do { - old = READ_ONCE(S390_lowcore.preempt_count); + old = READ_ONCE(get_lowcore()->preempt_count); new = (old & PREEMPT_NEED_RESCHED) | (pc & ~PREEMPT_NEED_RESCHED); - } while (__atomic_cmpxchg(&S390_lowcore.preempt_count, + } while (__atomic_cmpxchg(&get_lowcore()->preempt_count, old, new) != old); } static __always_inline void set_preempt_need_resched(void) { - __atomic_and(~PREEMPT_NEED_RESCHED, &S390_lowcore.preempt_count); + __atomic_and(~PREEMPT_NEED_RESCHED, &get_lowcore()->preempt_count); } static __always_inline void clear_preempt_need_resched(void) { - __atomic_or(PREEMPT_NEED_RESCHED, &S390_lowcore.preempt_count); + __atomic_or(PREEMPT_NEED_RESCHED, &get_lowcore()->preempt_count); } static __always_inline bool test_preempt_need_resched(void) { - return !(READ_ONCE(S390_lowcore.preempt_count) & PREEMPT_NEED_RESCHED); + return !(READ_ONCE(get_lowcore()->preempt_count) & PREEMPT_NEED_RESCHED); } static __always_inline void __preempt_count_add(int val) @@ -52,11 +52,11 @@ static __always_inline void __preempt_count_add(int val) */ if (!IS_ENABLED(CONFIG_PROFILE_ALL_BRANCHES)) { if (__builtin_constant_p(val) && (val >= -128) && (val <= 127)) { - __atomic_add_const(val, &S390_lowcore.preempt_count); + __atomic_add_const(val, &get_lowcore()->preempt_count); return; } } - __atomic_add(val, &S390_lowcore.preempt_count); + __atomic_add(val, &get_lowcore()->preempt_count); } static __always_inline void __preempt_count_sub(int val) @@ -66,12 +66,12 @@ static __always_inline void __preempt_count_sub(int val) static __always_inline bool __preempt_count_dec_and_test(void) { - return __atomic_add(-1, &S390_lowcore.preempt_count) == 1; + return __atomic_add(-1, &get_lowcore()->preempt_count) == 1; } static __always_inline bool should_resched(int preempt_offset) { - return unlikely(READ_ONCE(S390_lowcore.preempt_count) == + return unlikely(READ_ONCE(get_lowcore()->preempt_count) == preempt_offset); } @@ -81,12 +81,12 @@ static __always_inline bool should_resched(int preempt_offset) static __always_inline int preempt_count(void) { - return READ_ONCE(S390_lowcore.preempt_count); + return READ_ONCE(get_lowcore()->preempt_count); } static __always_inline void preempt_count_set(int pc) { - S390_lowcore.preempt_count = pc; + get_lowcore()->preempt_count = pc; } static __always_inline void set_preempt_need_resched(void) @@ -104,17 +104,17 @@ static __always_inline bool test_preempt_need_resched(void) static __always_inline void __preempt_count_add(int val) { - S390_lowcore.preempt_count += val; + get_lowcore()->preempt_count += val; } static __always_inline void __preempt_count_sub(int val) { - S390_lowcore.preempt_count -= val; + get_lowcore()->preempt_count -= val; } static __always_inline bool __preempt_count_dec_and_test(void) { - return !--S390_lowcore.preempt_count && tif_need_resched(); + return !--get_lowcore()->preempt_count && tif_need_resched(); } static __always_inline bool should_resched(int preempt_offset) diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 07ad5a1df878..5ecd442535b9 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -14,13 +14,11 @@ #include <linux/bits.h> -#define CIF_SIE 0 /* CPU needs SIE exit cleanup */ #define CIF_NOHZ_DELAY 2 /* delay HZ disable for a tick */ #define CIF_ENABLED_WAIT 5 /* in enabled wait state */ #define CIF_MCCK_GUEST 6 /* machine check happening in guest */ #define CIF_DEDICATED_CPU 7 /* this CPU is dedicated */ -#define _CIF_SIE BIT(CIF_SIE) #define _CIF_NOHZ_DELAY BIT(CIF_NOHZ_DELAY) #define _CIF_ENABLED_WAIT BIT(CIF_ENABLED_WAIT) #define _CIF_MCCK_GUEST BIT(CIF_MCCK_GUEST) @@ -42,21 +40,37 @@ #include <asm/irqflags.h> #include <asm/alternative.h> +struct pcpu { + unsigned long ec_mask; /* bit mask for ec_xxx functions */ + unsigned long ec_clk; /* sigp timestamp for ec_xxx */ + unsigned long flags; /* per CPU flags */ + signed char state; /* physical cpu state */ + signed char polarization; /* physical polarization */ + u16 address; /* physical cpu address */ +}; + +DECLARE_PER_CPU(struct pcpu, pcpu_devices); + typedef long (*sys_call_ptr_t)(struct pt_regs *regs); +static __always_inline struct pcpu *this_pcpu(void) +{ + return (struct pcpu *)(get_lowcore()->pcpu); +} + static __always_inline void set_cpu_flag(int flag) { - S390_lowcore.cpu_flags |= (1UL << flag); + this_pcpu()->flags |= (1UL << flag); } static __always_inline void clear_cpu_flag(int flag) { - S390_lowcore.cpu_flags &= ~(1UL << flag); + this_pcpu()->flags &= ~(1UL << flag); } static __always_inline bool test_cpu_flag(int flag) { - return S390_lowcore.cpu_flags & (1UL << flag); + return this_pcpu()->flags & (1UL << flag); } static __always_inline bool test_and_set_cpu_flag(int flag) @@ -81,9 +95,7 @@ static __always_inline bool test_and_clear_cpu_flag(int flag) */ static __always_inline bool test_cpu_flag_of(int flag, int cpu) { - struct lowcore *lc = lowcore_ptr[cpu]; - - return lc->cpu_flags & (1UL << flag); + return per_cpu(pcpu_devices, cpu).flags & (1UL << flag); } #define arch_needs_cpu() test_cpu_flag(CIF_NOHZ_DELAY) @@ -269,7 +281,7 @@ static __always_inline unsigned long __current_stack_pointer(void) static __always_inline bool on_thread_stack(void) { - unsigned long ksp = S390_lowcore.kernel_stack; + unsigned long ksp = get_lowcore()->kernel_stack; return !((ksp ^ current_stack_pointer) & ~(THREAD_SIZE - 1)); } @@ -405,7 +417,7 @@ static __always_inline bool regs_irqs_disabled(struct pt_regs *regs) static __always_inline void bpon(void) { - asm volatile(ALTERNATIVE("nop", ".insn rrf,0xb2e80000,0,0,13,0", 82)); + asm volatile(ALTERNATIVE("nop", ".insn rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82))); } #endif /* __ASSEMBLY__ */ diff --git a/arch/s390/include/asm/runtime-const.h b/arch/s390/include/asm/runtime-const.h new file mode 100644 index 000000000000..17878b1d048c --- /dev/null +++ b/arch/s390/include/asm/runtime-const.h @@ -0,0 +1,77 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_S390_RUNTIME_CONST_H +#define _ASM_S390_RUNTIME_CONST_H + +#include <linux/uaccess.h> + +#define runtime_const_ptr(sym) \ +({ \ + typeof(sym) __ret; \ + \ + asm_inline( \ + "0: iihf %[__ret],%[c1]\n" \ + " iilf %[__ret],%[c2]\n" \ + ".pushsection runtime_ptr_" #sym ",\"a\"\n" \ + ".long 0b - .\n" \ + ".popsection" \ + : [__ret] "=d" (__ret) \ + : [c1] "i" (0x01234567UL), \ + [c2] "i" (0x89abcdefUL)); \ + __ret; \ +}) + +#define runtime_const_shift_right_32(val, sym) \ +({ \ + unsigned int __ret = (val); \ + \ + asm_inline( \ + "0: srl %[__ret],12\n" \ + ".pushsection runtime_shift_" #sym ",\"a\"\n" \ + ".long 0b - .\n" \ + ".popsection" \ + : [__ret] "+d" (__ret)); \ + __ret; \ +}) + +#define runtime_const_init(type, sym) do { \ + extern s32 __start_runtime_##type##_##sym[]; \ + extern s32 __stop_runtime_##type##_##sym[]; \ + \ + runtime_const_fixup(__runtime_fixup_##type, \ + (unsigned long)(sym), \ + __start_runtime_##type##_##sym, \ + __stop_runtime_##type##_##sym); \ +} while (0) + +/* 32-bit immediate for iihf and iilf in bits in I2 field */ +static inline void __runtime_fixup_32(u32 *p, unsigned int val) +{ + s390_kernel_write(p, &val, sizeof(val)); +} + +static inline void __runtime_fixup_ptr(void *where, unsigned long val) +{ + __runtime_fixup_32(where + 2, val >> 32); + __runtime_fixup_32(where + 8, val); +} + +/* Immediate value is lower 12 bits of D2 field of srl */ +static inline void __runtime_fixup_shift(void *where, unsigned long val) +{ + u32 insn = *(u32 *)where; + + insn &= 0xfffff000; + insn |= (val & 63); + s390_kernel_write(where, &insn, sizeof(insn)); +} + +static inline void runtime_const_fixup(void (*fn)(void *, unsigned long), + unsigned long val, s32 *start, s32 *end) +{ + while (start < end) { + fn(*start + (void *)start, val); + start++; + } +} + +#endif /* _ASM_S390_RUNTIME_CONST_H */ diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h index 5742d23bba13..da3dad18fe50 100644 --- a/arch/s390/include/asm/sclp.h +++ b/arch/s390/include/asm/sclp.h @@ -84,6 +84,7 @@ struct sclp_info { unsigned char has_ibs : 1; unsigned char has_skey : 1; unsigned char has_kss : 1; + unsigned char has_diag204_bif : 1; unsigned char has_gisaf : 1; unsigned char has_diag318 : 1; unsigned char has_diag320 : 1; diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index 32f70873e2b7..8505737712ee 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -77,24 +77,24 @@ extern unsigned long max_mappable; /* The Write Back bit position in the physaddr is given by the SLPC PCI */ extern unsigned long mio_wb_bit_mask; -#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) - -#define MACHINE_HAS_DIAG9C (S390_lowcore.machine_flags & MACHINE_FLAG_DIAG9C) -#define MACHINE_HAS_ESOP (S390_lowcore.machine_flags & MACHINE_FLAG_ESOP) -#define MACHINE_HAS_IDTE (S390_lowcore.machine_flags & MACHINE_FLAG_IDTE) -#define MACHINE_HAS_EDAT1 (S390_lowcore.machine_flags & MACHINE_FLAG_EDAT1) -#define MACHINE_HAS_EDAT2 (S390_lowcore.machine_flags & MACHINE_FLAG_EDAT2) -#define MACHINE_HAS_TOPOLOGY (S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY) -#define MACHINE_HAS_TE (S390_lowcore.machine_flags & MACHINE_FLAG_TE) -#define MACHINE_HAS_TLB_LC (S390_lowcore.machine_flags & MACHINE_FLAG_TLB_LC) -#define MACHINE_HAS_TLB_GUEST (S390_lowcore.machine_flags & MACHINE_FLAG_TLB_GUEST) -#define MACHINE_HAS_NX (S390_lowcore.machine_flags & MACHINE_FLAG_NX) -#define MACHINE_HAS_GS (S390_lowcore.machine_flags & MACHINE_FLAG_GS) -#define MACHINE_HAS_SCC (S390_lowcore.machine_flags & MACHINE_FLAG_SCC) -#define MACHINE_HAS_PCI_MIO (S390_lowcore.machine_flags & MACHINE_FLAG_PCI_MIO) -#define MACHINE_HAS_RDP (S390_lowcore.machine_flags & MACHINE_FLAG_RDP) +#define MACHINE_IS_VM (get_lowcore()->machine_flags & MACHINE_FLAG_VM) +#define MACHINE_IS_KVM (get_lowcore()->machine_flags & MACHINE_FLAG_KVM) +#define MACHINE_IS_LPAR (get_lowcore()->machine_flags & MACHINE_FLAG_LPAR) + +#define MACHINE_HAS_DIAG9C (get_lowcore()->machine_flags & MACHINE_FLAG_DIAG9C) +#define MACHINE_HAS_ESOP (get_lowcore()->machine_flags & MACHINE_FLAG_ESOP) +#define MACHINE_HAS_IDTE (get_lowcore()->machine_flags & MACHINE_FLAG_IDTE) +#define MACHINE_HAS_EDAT1 (get_lowcore()->machine_flags & MACHINE_FLAG_EDAT1) +#define MACHINE_HAS_EDAT2 (get_lowcore()->machine_flags & MACHINE_FLAG_EDAT2) +#define MACHINE_HAS_TOPOLOGY (get_lowcore()->machine_flags & MACHINE_FLAG_TOPOLOGY) +#define MACHINE_HAS_TE (get_lowcore()->machine_flags & MACHINE_FLAG_TE) +#define MACHINE_HAS_TLB_LC (get_lowcore()->machine_flags & MACHINE_FLAG_TLB_LC) +#define MACHINE_HAS_TLB_GUEST (get_lowcore()->machine_flags & MACHINE_FLAG_TLB_GUEST) +#define MACHINE_HAS_NX (get_lowcore()->machine_flags & MACHINE_FLAG_NX) +#define MACHINE_HAS_GS (get_lowcore()->machine_flags & MACHINE_FLAG_GS) +#define MACHINE_HAS_SCC (get_lowcore()->machine_flags & MACHINE_FLAG_SCC) +#define MACHINE_HAS_PCI_MIO (get_lowcore()->machine_flags & MACHINE_FLAG_PCI_MIO) +#define MACHINE_HAS_RDP (get_lowcore()->machine_flags & MACHINE_FLAG_RDP) /* * Console mode. Override with conmode= diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h index 6e5b1b4b19a9..cd835f4fb11a 100644 --- a/arch/s390/include/asm/smp.h +++ b/arch/s390/include/asm/smp.h @@ -11,7 +11,7 @@ #include <asm/lowcore.h> #include <asm/processor.h> -#define raw_smp_processor_id() (S390_lowcore.cpu_nr) +#define raw_smp_processor_id() (get_lowcore()->cpu_nr) extern struct mutex smp_cpu_state_mutex; extern unsigned int smp_cpu_mt_shift; @@ -24,7 +24,6 @@ extern int __cpu_up(unsigned int cpu, struct task_struct *tidle); extern void arch_send_call_function_single_ipi(int cpu); extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); -extern void smp_call_online_cpu(void (*func)(void *), void *); extern void smp_call_ipl_cpu(void (*func)(void *), void *); extern void smp_emergency_stop(void); @@ -59,7 +58,7 @@ static inline void smp_cpus_done(unsigned int max_cpus) { } -extern int smp_rescan_cpus(void); +extern int smp_rescan_cpus(bool early); extern void __noreturn cpu_die(void); extern void __cpu_die(unsigned int cpu); extern int __cpu_disable(void); diff --git a/arch/s390/include/asm/softirq_stack.h b/arch/s390/include/asm/softirq_stack.h index 1ac5115d3115..42d61296bbad 100644 --- a/arch/s390/include/asm/softirq_stack.h +++ b/arch/s390/include/asm/softirq_stack.h @@ -8,7 +8,7 @@ #ifdef CONFIG_SOFTIRQ_ON_OWN_STACK static inline void do_softirq_own_stack(void) { - call_on_stack(0, S390_lowcore.async_stack, void, __do_softirq); + call_on_stack(0, get_lowcore()->async_stack, void, __do_softirq); } #endif #endif /* __ASM_S390_SOFTIRQ_STACK_H */ diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 37127cd7749e..77d5e804af93 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -16,7 +16,7 @@ #include <asm/processor.h> #include <asm/alternative.h> -#define SPINLOCK_LOCKVAL (S390_lowcore.spinlock_lockval) +#define SPINLOCK_LOCKVAL (get_lowcore()->spinlock_lockval) extern int spin_retry; @@ -79,7 +79,7 @@ static inline void arch_spin_unlock(arch_spinlock_t *lp) typecheck(int, lp->lock); kcsan_release(); asm_inline volatile( - ALTERNATIVE("nop", ".insn rre,0xb2fa0000,7,0", 49) /* NIAI 7 */ + ALTERNATIVE("nop", ".insn rre,0xb2fa0000,7,0", ALT_FACILITY(49)) /* NIAI 7 */ " sth %1,%0\n" : "=R" (((unsigned short *) &lp->lock)[1]) : "d" (0) : "cc", "memory"); diff --git a/arch/s390/include/asm/stacktrace.h b/arch/s390/include/asm/stacktrace.h index 85b6738b826a..1d5ca13dc90f 100644 --- a/arch/s390/include/asm/stacktrace.h +++ b/arch/s390/include/asm/stacktrace.h @@ -65,6 +65,7 @@ struct stack_frame { unsigned long sie_reason; unsigned long sie_flags; unsigned long sie_control_block_phys; + unsigned long sie_guest_asce; }; }; unsigned long gprs[10]; diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h index 351685de53d2..2ab868cbae6c 100644 --- a/arch/s390/include/asm/string.h +++ b/arch/s390/include/asm/string.h @@ -15,15 +15,12 @@ #define __HAVE_ARCH_MEMCPY /* gcc builtin & arch function */ #define __HAVE_ARCH_MEMMOVE /* gcc builtin & arch function */ #define __HAVE_ARCH_MEMSET /* gcc builtin & arch function */ -#define __HAVE_ARCH_MEMSET16 /* arch function */ -#define __HAVE_ARCH_MEMSET32 /* arch function */ -#define __HAVE_ARCH_MEMSET64 /* arch function */ void *memcpy(void *dest, const void *src, size_t n); void *memset(void *s, int c, size_t n); void *memmove(void *dest, const void *src, size_t n); -#ifndef CONFIG_KASAN +#if !defined(CONFIG_KASAN) && !defined(CONFIG_KMSAN) #define __HAVE_ARCH_MEMCHR /* inline & arch function */ #define __HAVE_ARCH_MEMCMP /* arch function */ #define __HAVE_ARCH_MEMSCAN /* inline & arch function */ @@ -36,6 +33,9 @@ void *memmove(void *dest, const void *src, size_t n); #define __HAVE_ARCH_STRNCPY /* arch function */ #define __HAVE_ARCH_STRNLEN /* inline & arch function */ #define __HAVE_ARCH_STRSTR /* arch function */ +#define __HAVE_ARCH_MEMSET16 /* arch function */ +#define __HAVE_ARCH_MEMSET32 /* arch function */ +#define __HAVE_ARCH_MEMSET64 /* arch function */ /* Prototypes for non-inlined arch strings functions. */ int memcmp(const void *s1, const void *s2, size_t n); @@ -44,7 +44,7 @@ size_t strlcat(char *dest, const char *src, size_t n); char *strncat(char *dest, const char *src, size_t n); char *strncpy(char *dest, const char *src, size_t n); char *strstr(const char *s1, const char *s2); -#endif /* !CONFIG_KASAN */ +#endif /* !defined(CONFIG_KASAN) && !defined(CONFIG_KMSAN) */ #undef __HAVE_ARCH_STRCHR #undef __HAVE_ARCH_STRNCHR @@ -74,20 +74,30 @@ 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); +#ifdef __HAVE_ARCH_MEMSET16 static inline void *memset16(uint16_t *s, uint16_t v, size_t count) { return __memset16(s, v, count * sizeof(v)); } +#endif +#ifdef __HAVE_ARCH_MEMSET32 static inline void *memset32(uint32_t *s, uint32_t v, size_t count) { return __memset32(s, v, count * sizeof(v)); } +#endif +#ifdef __HAVE_ARCH_MEMSET64 +#ifdef IN_BOOT_STRING_C +void *memset64(uint64_t *s, uint64_t v, size_t count); +#else static inline void *memset64(uint64_t *s, uint64_t v, size_t count) { return __memset64(s, v, count * sizeof(v)); } +#endif +#endif #if !defined(IN_ARCH_STRING_C) && (!defined(CONFIG_FORTIFY_SOURCE) || defined(__NO_FORTIFY)) diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h index a674c7d25da5..00ac01874a12 100644 --- a/arch/s390/include/asm/thread_info.h +++ b/arch/s390/include/asm/thread_info.h @@ -16,7 +16,7 @@ /* * General size of kernel stacks */ -#ifdef CONFIG_KASAN +#if defined(CONFIG_KASAN) || defined(CONFIG_KMSAN) #define THREAD_SIZE_ORDER 4 #else #define THREAD_SIZE_ORDER 2 @@ -40,6 +40,7 @@ struct thread_info { unsigned long flags; /* low level flags */ unsigned long syscall_work; /* SYSCALL_WORK_ flags */ unsigned int cpu; /* current CPU */ + unsigned char sie; /* running in SIE context */ }; /* diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h index 4d646659a5f5..640901f2fbc3 100644 --- a/arch/s390/include/asm/timex.h +++ b/arch/s390/include/asm/timex.h @@ -161,16 +161,16 @@ static inline unsigned long local_tick_disable(void) { unsigned long old; - old = S390_lowcore.clock_comparator; - S390_lowcore.clock_comparator = clock_comparator_max; - set_clock_comparator(S390_lowcore.clock_comparator); + old = get_lowcore()->clock_comparator; + get_lowcore()->clock_comparator = clock_comparator_max; + set_clock_comparator(get_lowcore()->clock_comparator); return old; } static inline void local_tick_enable(unsigned long comp) { - S390_lowcore.clock_comparator = comp; - set_clock_comparator(S390_lowcore.clock_comparator); + get_lowcore()->clock_comparator = comp; + set_clock_comparator(get_lowcore()->clock_comparator); } #define CLOCK_TICK_RATE 1193180 /* Underlying HZ */ diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h index 81ae8a98e7ec..a81f897a81ce 100644 --- a/arch/s390/include/asm/uaccess.h +++ b/arch/s390/include/asm/uaccess.h @@ -18,6 +18,7 @@ #include <asm/extable.h> #include <asm/facility.h> #include <asm-generic/access_ok.h> +#include <linux/instrumented.h> void debug_user_asce(int exit); @@ -78,13 +79,24 @@ union oac { int __noreturn __put_user_bad(void); -#define __put_user_asm(to, from, size) \ -({ \ +#ifdef CONFIG_KMSAN +#define get_put_user_noinstr_attributes \ + noinline __maybe_unused __no_sanitize_memory +#else +#define get_put_user_noinstr_attributes __always_inline +#endif + +#define DEFINE_PUT_USER(type) \ +static get_put_user_noinstr_attributes int \ +__put_user_##type##_noinstr(unsigned type __user *to, \ + unsigned type *from, \ + unsigned long size) \ +{ \ union oac __oac_spec = { \ .oac1.as = PSW_BITS_AS_SECONDARY, \ .oac1.a = 1, \ }; \ - int __rc; \ + int rc; \ \ asm volatile( \ " lr 0,%[spec]\n" \ @@ -93,12 +105,28 @@ int __noreturn __put_user_bad(void); "2:\n" \ EX_TABLE_UA_STORE(0b, 2b, %[rc]) \ EX_TABLE_UA_STORE(1b, 2b, %[rc]) \ - : [rc] "=&d" (__rc), [_to] "+Q" (*(to)) \ + : [rc] "=&d" (rc), [_to] "+Q" (*(to)) \ : [_size] "d" (size), [_from] "Q" (*(from)), \ [spec] "d" (__oac_spec.val) \ : "cc", "0"); \ - __rc; \ -}) + return rc; \ +} \ + \ +static __always_inline int \ +__put_user_##type(unsigned type __user *to, unsigned type *from, \ + unsigned long size) \ +{ \ + int rc; \ + \ + rc = __put_user_##type##_noinstr(to, from, size); \ + instrument_put_user(*from, to, size); \ + return rc; \ +} + +DEFINE_PUT_USER(char); +DEFINE_PUT_USER(short); +DEFINE_PUT_USER(int); +DEFINE_PUT_USER(long); static __always_inline int __put_user_fn(void *x, void __user *ptr, unsigned long size) { @@ -106,24 +134,24 @@ static __always_inline int __put_user_fn(void *x, void __user *ptr, unsigned lon switch (size) { case 1: - rc = __put_user_asm((unsigned char __user *)ptr, - (unsigned char *)x, - size); + rc = __put_user_char((unsigned char __user *)ptr, + (unsigned char *)x, + size); break; case 2: - rc = __put_user_asm((unsigned short __user *)ptr, - (unsigned short *)x, - size); + rc = __put_user_short((unsigned short __user *)ptr, + (unsigned short *)x, + size); break; case 4: - rc = __put_user_asm((unsigned int __user *)ptr, + rc = __put_user_int((unsigned int __user *)ptr, (unsigned int *)x, size); break; case 8: - rc = __put_user_asm((unsigned long __user *)ptr, - (unsigned long *)x, - size); + rc = __put_user_long((unsigned long __user *)ptr, + (unsigned long *)x, + size); break; default: __put_user_bad(); @@ -134,13 +162,17 @@ static __always_inline int __put_user_fn(void *x, void __user *ptr, unsigned lon int __noreturn __get_user_bad(void); -#define __get_user_asm(to, from, size) \ -({ \ +#define DEFINE_GET_USER(type) \ +static get_put_user_noinstr_attributes int \ +__get_user_##type##_noinstr(unsigned type *to, \ + unsigned type __user *from, \ + unsigned long size) \ +{ \ union oac __oac_spec = { \ .oac2.as = PSW_BITS_AS_SECONDARY, \ .oac2.a = 1, \ }; \ - int __rc; \ + int rc; \ \ asm volatile( \ " lr 0,%[spec]\n" \ @@ -149,13 +181,29 @@ int __noreturn __get_user_bad(void); "2:\n" \ EX_TABLE_UA_LOAD_MEM(0b, 2b, %[rc], %[_to], %[_ksize]) \ EX_TABLE_UA_LOAD_MEM(1b, 2b, %[rc], %[_to], %[_ksize]) \ - : [rc] "=&d" (__rc), "=Q" (*(to)) \ + : [rc] "=&d" (rc), "=Q" (*(to)) \ : [_size] "d" (size), [_from] "Q" (*(from)), \ [spec] "d" (__oac_spec.val), [_to] "a" (to), \ [_ksize] "K" (size) \ : "cc", "0"); \ - __rc; \ -}) + return rc; \ +} \ + \ +static __always_inline int \ +__get_user_##type(unsigned type *to, unsigned type __user *from, \ + unsigned long size) \ +{ \ + int rc; \ + \ + rc = __get_user_##type##_noinstr(to, from, size); \ + instrument_get_user(*to); \ + return rc; \ +} + +DEFINE_GET_USER(char); +DEFINE_GET_USER(short); +DEFINE_GET_USER(int); +DEFINE_GET_USER(long); static __always_inline int __get_user_fn(void *x, const void __user *ptr, unsigned long size) { @@ -163,24 +211,24 @@ static __always_inline int __get_user_fn(void *x, const void __user *ptr, unsign switch (size) { case 1: - rc = __get_user_asm((unsigned char *)x, - (unsigned char __user *)ptr, - size); + rc = __get_user_char((unsigned char *)x, + (unsigned char __user *)ptr, + size); break; case 2: - rc = __get_user_asm((unsigned short *)x, - (unsigned short __user *)ptr, - size); + rc = __get_user_short((unsigned short *)x, + (unsigned short __user *)ptr, + size); break; case 4: - rc = __get_user_asm((unsigned int *)x, + rc = __get_user_int((unsigned int *)x, (unsigned int __user *)ptr, size); break; case 8: - rc = __get_user_asm((unsigned long *)x, - (unsigned long __user *)ptr, - size); + rc = __get_user_long((unsigned long *)x, + (unsigned long __user *)ptr, + size); break; default: __get_user_bad(); @@ -284,7 +332,14 @@ static inline unsigned long __must_check clear_user(void __user *to, unsigned lo return __clear_user(to, n); } -void *s390_kernel_write(void *dst, const void *src, size_t size); +void *__s390_kernel_write(void *dst, const void *src, size_t size); + +static inline void *s390_kernel_write(void *dst, const void *src, size_t size) +{ + if (__is_defined(__DECOMPRESSOR)) + return memcpy(dst, src, size); + return __s390_kernel_write(dst, src, size); +} int __noreturn __put_kernel_bad(void); diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h index 4260bc5ce7f8..70fc671397da 100644 --- a/arch/s390/include/asm/unistd.h +++ b/arch/s390/include/asm/unistd.h @@ -35,6 +35,5 @@ #define __ARCH_WANT_SYS_FORK #define __ARCH_WANT_SYS_VFORK #define __ARCH_WANT_SYS_CLONE -#define __ARCH_WANT_SYS_CLONE3 #endif /* _ASM_S390_UNISTD_H_ */ diff --git a/arch/s390/include/asm/uv.h b/arch/s390/include/asm/uv.h index 0e7bd3873907..153d93468b77 100644 --- a/arch/s390/include/asm/uv.h +++ b/arch/s390/include/asm/uv.h @@ -414,7 +414,6 @@ static inline bool uv_has_feature(u8 feature_bit) return test_bit_inv(feature_bit, &uv_info.uv_feature_indications); } -#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST extern int prot_virt_guest; static inline int is_prot_virt_guest(void) @@ -442,7 +441,10 @@ static inline int share(unsigned long addr, u16 cmd) if (!uv_call(0, (u64)&uvcb)) return 0; - return -EINVAL; + pr_err("%s UVC failed (rc: 0x%x, rrc: 0x%x), possible hypervisor bug.\n", + uvcb.header.cmd == UVC_CMD_SET_SHARED_ACCESS ? "Share" : "Unshare", + uvcb.header.rc, uvcb.header.rrc); + panic("System security cannot be guaranteed unless the system panics now.\n"); } /* @@ -466,13 +468,6 @@ static inline int uv_remove_shared(unsigned long addr) return share(addr, UVC_CMD_REMOVE_SHARED_ACCESS); } -#else -#define is_prot_virt_guest() 0 -static inline int uv_set_shared(unsigned long addr) { return 0; } -static inline int uv_remove_shared(unsigned long addr) { return 0; } -#endif - -#if IS_ENABLED(CONFIG_KVM) extern int prot_virt_host; static inline int is_prot_virt_host(void) @@ -483,35 +478,11 @@ static inline int is_prot_virt_host(void) int uv_pin_shared(unsigned long paddr); int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb); int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr); -int uv_destroy_owned_page(unsigned long paddr); -int uv_convert_from_secure(unsigned long paddr); -int uv_convert_owned_from_secure(unsigned long paddr); +int uv_destroy_folio(struct folio *folio); +int uv_destroy_pte(pte_t pte); +int uv_convert_from_secure_pte(pte_t pte); int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr); void setup_uv(void); -#else -#define is_prot_virt_host() 0 -static inline void setup_uv(void) {} - -static inline int uv_pin_shared(unsigned long paddr) -{ - return 0; -} - -static inline int uv_destroy_owned_page(unsigned long paddr) -{ - return 0; -} - -static inline int uv_convert_from_secure(unsigned long paddr) -{ - return 0; -} - -static inline int uv_convert_owned_from_secure(unsigned long paddr) -{ - return 0; -} -#endif #endif /* _ASM_S390_UV_H */ diff --git a/arch/s390/include/asm/vtime.h b/arch/s390/include/asm/vtime.h index 561c91c1a87c..9d25fb35a042 100644 --- a/arch/s390/include/asm/vtime.h +++ b/arch/s390/include/asm/vtime.h @@ -4,16 +4,20 @@ static inline void update_timer_sys(void) { - S390_lowcore.system_timer += S390_lowcore.last_update_timer - S390_lowcore.exit_timer; - S390_lowcore.user_timer += S390_lowcore.exit_timer - S390_lowcore.sys_enter_timer; - S390_lowcore.last_update_timer = S390_lowcore.sys_enter_timer; + struct lowcore *lc = get_lowcore(); + + lc->system_timer += lc->last_update_timer - lc->exit_timer; + lc->user_timer += lc->exit_timer - lc->sys_enter_timer; + lc->last_update_timer = lc->sys_enter_timer; } static inline void update_timer_mcck(void) { - S390_lowcore.system_timer += S390_lowcore.last_update_timer - S390_lowcore.exit_timer; - S390_lowcore.user_timer += S390_lowcore.exit_timer - S390_lowcore.mcck_enter_timer; - S390_lowcore.last_update_timer = S390_lowcore.mcck_enter_timer; + struct lowcore *lc = get_lowcore(); + + lc->system_timer += lc->last_update_timer - lc->exit_timer; + lc->user_timer += lc->exit_timer - lc->mcck_enter_timer; + lc->last_update_timer = lc->mcck_enter_timer; } #endif /* _S390_VTIME_H */ diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 7241fa194709..e47a4be54ff8 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -43,7 +43,7 @@ obj-y += sysinfo.o lgr.o os_info.o ctlreg.o obj-y += runtime_instr.o cache.o fpu.o dumpstack.o guarded_storage.o sthyi.o obj-y += entry.o reipl.o kdebugfs.o alternative.o obj-y += nospec-branch.o ipl_vmparm.o machine_kexec_reloc.o unwind_bc.o -obj-y += smp.o text_amode31.o stacktrace.o abs_lowcore.o facility.o +obj-y += smp.o text_amode31.o stacktrace.o abs_lowcore.o facility.o uv.o extra-y += vmlinux.lds @@ -80,7 +80,6 @@ obj-$(CONFIG_PERF_EVENTS) += perf_cpum_cf_events.o perf_regs.o obj-$(CONFIG_PERF_EVENTS) += perf_pai_crypto.o perf_pai_ext.o obj-$(CONFIG_TRACEPOINTS) += trace.o -obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE)) += uv.o # vdso obj-y += vdso64/ diff --git a/arch/s390/kernel/abs_lowcore.c b/arch/s390/kernel/abs_lowcore.c index f9efc54ec4b7..09cd24cbe74e 100644 --- a/arch/s390/kernel/abs_lowcore.c +++ b/arch/s390/kernel/abs_lowcore.c @@ -4,6 +4,7 @@ #include <asm/abs_lowcore.h> unsigned long __bootdata_preserved(__abs_lowcore); +int __bootdata_preserved(relocate_lowcore); int abs_lowcore_map(int cpu, struct lowcore *lc, bool alloc) { diff --git a/arch/s390/kernel/alternative.c b/arch/s390/kernel/alternative.c index 1ac5f707dd70..8d5d0de35de0 100644 --- a/arch/s390/kernel/alternative.c +++ b/arch/s390/kernel/alternative.c @@ -1,68 +1,41 @@ // SPDX-License-Identifier: GPL-2.0 -#include <linux/module.h> -#include <linux/cpu.h> -#include <linux/smp.h> -#include <asm/text-patching.h> + +#include <linux/uaccess.h> +#include <asm/nospec-branch.h> +#include <asm/abs_lowcore.h> #include <asm/alternative.h> #include <asm/facility.h> -#include <asm/nospec-branch.h> - -static int __initdata_or_module alt_instr_disabled; - -static int __init disable_alternative_instructions(char *str) -{ - alt_instr_disabled = 1; - return 0; -} - -early_param("noaltinstr", disable_alternative_instructions); -static void __init_or_module __apply_alternatives(struct alt_instr *start, - struct alt_instr *end) +void __apply_alternatives(struct alt_instr *start, struct alt_instr *end, unsigned int ctx) { - struct alt_instr *a; u8 *instr, *replacement; + struct alt_instr *a; + bool replace; /* * The scan order should be from start to end. A later scanned * alternative code can overwrite previously scanned alternative code. */ for (a = start; a < end; a++) { + if (!(a->ctx & ctx)) + continue; + switch (a->type) { + case ALT_TYPE_FACILITY: + replace = test_facility(a->data); + break; + case ALT_TYPE_SPEC: + replace = nobp_enabled(); + break; + case ALT_TYPE_LOWCORE: + replace = have_relocated_lowcore(); + break; + default: + replace = false; + } + if (!replace) + continue; instr = (u8 *)&a->instr_offset + a->instr_offset; replacement = (u8 *)&a->repl_offset + a->repl_offset; - - if (!__test_facility(a->facility, alt_stfle_fac_list)) - continue; s390_kernel_write(instr, replacement, a->instrlen); } } - -void __init_or_module apply_alternatives(struct alt_instr *start, - struct alt_instr *end) -{ - if (!alt_instr_disabled) - __apply_alternatives(start, end); -} - -extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; -void __init apply_alternative_instructions(void) -{ - apply_alternatives(__alt_instructions, __alt_instructions_end); -} - -static void do_sync_core(void *info) -{ - sync_core(); -} - -void text_poke_sync(void) -{ - on_each_cpu(do_sync_core, NULL, 1); -} - -void text_poke_sync_lock(void) -{ - cpus_read_lock(); - text_poke_sync(); - cpus_read_unlock(); -} diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index f55979f64d49..ffa0dd2dbaac 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -28,6 +28,7 @@ int main(void) BLANK(); /* thread info offsets */ OFFSET(__TI_flags, task_struct, thread_info.flags); + OFFSET(__TI_sie, task_struct, thread_info.sie); BLANK(); /* pt_regs offsets */ OFFSET(__PT_PSW, pt_regs, psw); @@ -63,6 +64,7 @@ int main(void) OFFSET(__SF_SIE_REASON, stack_frame, sie_reason); OFFSET(__SF_SIE_FLAGS, stack_frame, sie_flags); OFFSET(__SF_SIE_CONTROL_PHYS, stack_frame, sie_control_block_phys); + OFFSET(__SF_SIE_GUEST_ASCE, stack_frame, sie_guest_asce); DEFINE(STACK_FRAME_OVERHEAD, sizeof(struct stack_frame)); BLANK(); OFFSET(__SFUSER_BACKCHAIN, stack_frame_user, back_chain); @@ -113,7 +115,7 @@ int main(void) OFFSET(__LC_SAVE_AREA_SYNC, lowcore, save_area_sync); OFFSET(__LC_SAVE_AREA_ASYNC, lowcore, save_area_async); OFFSET(__LC_SAVE_AREA_RESTART, lowcore, save_area_restart); - OFFSET(__LC_CPU_FLAGS, lowcore, cpu_flags); + OFFSET(__LC_PCPU, lowcore, pcpu); OFFSET(__LC_RETURN_PSW, lowcore, return_psw); OFFSET(__LC_RETURN_MCCK_PSW, lowcore, return_mcck_psw); OFFSET(__LC_SYS_ENTER_TIMER, lowcore, sys_enter_timer); @@ -185,5 +187,7 @@ int main(void) #endif OFFSET(__FTRACE_REGS_PT_REGS, ftrace_regs, regs); DEFINE(__FTRACE_REGS_SIZE, sizeof(struct ftrace_regs)); + + OFFSET(__PCPU_FLAGS, pcpu, flags); return 0; } diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 85328a0ef3b6..bce50ca75ea7 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -954,7 +954,7 @@ static int debug_active = 1; * always allow read, allow write only if debug_stoppable is set or * if debug_active is already off */ -static int s390dbf_procactive(struct ctl_table *table, int write, +static int s390dbf_procactive(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { if (!write || debug_stoppable || !debug_active) diff --git a/arch/s390/kernel/diag.c b/arch/s390/kernel/diag.c index 8dee9aa0ec95..ac7b8c8e3133 100644 --- a/arch/s390/kernel/diag.c +++ b/arch/s390/kernel/diag.c @@ -185,6 +185,8 @@ int diag14(unsigned long rx, unsigned long ry1, unsigned long subcode) } EXPORT_SYMBOL(diag14); +#define DIAG204_BUSY_RC 8 + static inline int __diag204(unsigned long *subcode, unsigned long size, void *addr) { union register_pair rp = { .even = *subcode, .odd = size }; @@ -215,16 +217,18 @@ int diag204(unsigned long subcode, unsigned long size, void *addr) { if (addr) { if (WARN_ON_ONCE(!is_vmalloc_addr(addr))) - return -1; + return -EINVAL; if (WARN_ON_ONCE(!IS_ALIGNED((unsigned long)addr, PAGE_SIZE))) - return -1; + return -EINVAL; } if ((subcode & DIAG204_SUBCODE_MASK) == DIAG204_SUBC_STIB4) addr = (void *)pfn_to_phys(vmalloc_to_pfn(addr)); diag_stat_inc(DIAG_STAT_X204); size = __diag204(&subcode, size, addr); - if (subcode) - return -1; + if (subcode == DIAG204_BUSY_RC) + return -EBUSY; + else if (subcode) + return -EOPNOTSUPP; return size; } EXPORT_SYMBOL(diag204); @@ -278,12 +282,14 @@ int diag224(void *ptr) int rc = -EOPNOTSUPP; diag_stat_inc(DIAG_STAT_X224); - asm volatile( - " diag %1,%2,0x224\n" - "0: lhi %0,0x0\n" + asm volatile("\n" + " diag %[type],%[addr],0x224\n" + "0: lhi %[rc],0\n" "1:\n" EX_TABLE(0b,1b) - : "+d" (rc) :"d" (0), "d" (addr) : "memory"); + : [rc] "+d" (rc) + , "=m" (*(struct { char buf[PAGE_SIZE]; } *)ptr) + : [type] "d" (0), [addr] "d" (addr)); return rc; } EXPORT_SYMBOL(diag224); diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c index d2012635b093..1ecd0580561f 100644 --- a/arch/s390/kernel/dumpstack.c +++ b/arch/s390/kernel/dumpstack.c @@ -61,28 +61,28 @@ static bool in_task_stack(unsigned long sp, struct task_struct *task, static bool in_irq_stack(unsigned long sp, struct stack_info *info) { - unsigned long stack = S390_lowcore.async_stack - STACK_INIT_OFFSET; + unsigned long stack = get_lowcore()->async_stack - STACK_INIT_OFFSET; return in_stack(sp, info, STACK_TYPE_IRQ, stack); } static bool in_nodat_stack(unsigned long sp, struct stack_info *info) { - unsigned long stack = S390_lowcore.nodat_stack - STACK_INIT_OFFSET; + unsigned long stack = get_lowcore()->nodat_stack - STACK_INIT_OFFSET; return in_stack(sp, info, STACK_TYPE_NODAT, stack); } static bool in_mcck_stack(unsigned long sp, struct stack_info *info) { - unsigned long stack = S390_lowcore.mcck_stack - STACK_INIT_OFFSET; + unsigned long stack = get_lowcore()->mcck_stack - STACK_INIT_OFFSET; return in_stack(sp, info, STACK_TYPE_MCCK, stack); } static bool in_restart_stack(unsigned long sp, struct stack_info *info) { - unsigned long stack = S390_lowcore.restart_stack - STACK_INIT_OFFSET; + unsigned long stack = get_lowcore()->restart_stack - STACK_INIT_OFFSET; return in_stack(sp, info, STACK_TYPE_RESTART, stack); } diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index c666271433fb..14d324865e33 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -48,6 +48,7 @@ decompressor_handled_param(dfltcc); decompressor_handled_param(facilities); decompressor_handled_param(nokaslr); decompressor_handled_param(cmma); +decompressor_handled_param(relocate_lowcore); #if IS_ENABLED(CONFIG_KVM) decompressor_handled_param(prot_virt); #endif @@ -72,7 +73,7 @@ static void __init reset_tod_clock(void) memset(&tod_clock_base, 0, sizeof(tod_clock_base)); tod_clock_base.tod = TOD_UNIX_EPOCH; - S390_lowcore.last_update_clock = TOD_UNIX_EPOCH; + get_lowcore()->last_update_clock = TOD_UNIX_EPOCH; } /* @@ -99,7 +100,7 @@ static noinline __init void detect_machine_type(void) /* Check current-configuration-level */ if (stsi(NULL, 0, 0, 0) <= 2) { - S390_lowcore.machine_flags |= MACHINE_FLAG_LPAR; + get_lowcore()->machine_flags |= MACHINE_FLAG_LPAR; return; } /* Get virtual-machine cpu information. */ @@ -108,9 +109,9 @@ static noinline __init void detect_machine_type(void) /* Detect known hypervisors */ if (!memcmp(vmms->vm[0].cpi, "\xd2\xe5\xd4", 3)) - S390_lowcore.machine_flags |= MACHINE_FLAG_KVM; + get_lowcore()->machine_flags |= MACHINE_FLAG_KVM; else if (!memcmp(vmms->vm[0].cpi, "\xa9\x61\xe5\xd4", 4)) - S390_lowcore.machine_flags |= MACHINE_FLAG_VM; + get_lowcore()->machine_flags |= MACHINE_FLAG_VM; } /* Remove leading, trailing and double whitespace. */ @@ -166,7 +167,7 @@ static __init void setup_topology(void) if (!test_facility(11)) return; - S390_lowcore.machine_flags |= MACHINE_FLAG_TOPOLOGY; + get_lowcore()->machine_flags |= MACHINE_FLAG_TOPOLOGY; for (max_mnest = 6; max_mnest > 1; max_mnest--) { if (stsi(&sysinfo_page, 15, 1, max_mnest) == 0) break; @@ -186,15 +187,8 @@ static noinline __init void setup_lowcore_early(void) psw.addr = (unsigned long)early_pgm_check_handler; psw.mask = PSW_KERNEL_BITS; - S390_lowcore.program_new_psw = psw; - S390_lowcore.preempt_count = INIT_PREEMPT_COUNT; -} - -static noinline __init void setup_facility_list(void) -{ - memcpy(alt_stfle_fac_list, stfle_fac_list, sizeof(alt_stfle_fac_list)); - if (!IS_ENABLED(CONFIG_KERNEL_NOBP)) - __clear_facility(82, alt_stfle_fac_list); + get_lowcore()->program_new_psw = psw; + get_lowcore()->preempt_count = INIT_PREEMPT_COUNT; } static __init void detect_diag9c(void) @@ -211,43 +205,43 @@ static __init void detect_diag9c(void) EX_TABLE(0b,1b) : "=d" (rc) : "0" (-EOPNOTSUPP), "d" (cpu_address) : "cc"); if (!rc) - S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG9C; + get_lowcore()->machine_flags |= MACHINE_FLAG_DIAG9C; } static __init void detect_machine_facilities(void) { if (test_facility(8)) { - S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT1; + get_lowcore()->machine_flags |= MACHINE_FLAG_EDAT1; system_ctl_set_bit(0, CR0_EDAT_BIT); } if (test_facility(78)) - S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT2; + get_lowcore()->machine_flags |= MACHINE_FLAG_EDAT2; if (test_facility(3)) - S390_lowcore.machine_flags |= MACHINE_FLAG_IDTE; + get_lowcore()->machine_flags |= MACHINE_FLAG_IDTE; if (test_facility(50) && test_facility(73)) { - S390_lowcore.machine_flags |= MACHINE_FLAG_TE; + get_lowcore()->machine_flags |= MACHINE_FLAG_TE; system_ctl_set_bit(0, CR0_TRANSACTIONAL_EXECUTION_BIT); } if (test_facility(51)) - S390_lowcore.machine_flags |= MACHINE_FLAG_TLB_LC; + get_lowcore()->machine_flags |= MACHINE_FLAG_TLB_LC; if (test_facility(129)) system_ctl_set_bit(0, CR0_VECTOR_BIT); if (test_facility(130)) - S390_lowcore.machine_flags |= MACHINE_FLAG_NX; + get_lowcore()->machine_flags |= MACHINE_FLAG_NX; if (test_facility(133)) - S390_lowcore.machine_flags |= MACHINE_FLAG_GS; + get_lowcore()->machine_flags |= MACHINE_FLAG_GS; if (test_facility(139) && (tod_clock_base.tod >> 63)) { /* Enabled signed clock comparator comparisons */ - S390_lowcore.machine_flags |= MACHINE_FLAG_SCC; + get_lowcore()->machine_flags |= MACHINE_FLAG_SCC; clock_comparator_max = -1ULL >> 1; system_ctl_set_bit(0, CR0_CLOCK_COMPARATOR_SIGN_BIT); } if (IS_ENABLED(CONFIG_PCI) && test_facility(153)) { - S390_lowcore.machine_flags |= MACHINE_FLAG_PCI_MIO; + get_lowcore()->machine_flags |= MACHINE_FLAG_PCI_MIO; /* the control bit is set during PCI initialization */ } if (test_facility(194)) - S390_lowcore.machine_flags |= MACHINE_FLAG_RDP; + get_lowcore()->machine_flags |= MACHINE_FLAG_RDP; } static inline void save_vector_registers(void) @@ -291,7 +285,6 @@ void __init startup_init(void) lockdep_off(); sort_amode31_extable(); setup_lowcore_early(); - setup_facility_list(); detect_machine_type(); setup_arch_string(); setup_boot_command_line(); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 60cf917a7122..749410cfdbc0 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -12,7 +12,7 @@ #include <linux/init.h> #include <linux/linkage.h> #include <asm/asm-extable.h> -#include <asm/alternative-asm.h> +#include <asm/alternative.h> #include <asm/processor.h> #include <asm/cache.h> #include <asm/dwarf.h> @@ -28,49 +28,54 @@ #include <asm/setup.h> #include <asm/nmi.h> #include <asm/nospec-insn.h> +#include <asm/lowcore.h> _LPP_OFFSET = __LC_LPP .macro STBEAR address - ALTERNATIVE "nop", ".insn s,0xb2010000,\address", 193 + ALTERNATIVE "nop", ".insn s,0xb2010000,\address", ALT_FACILITY(193) .endm .macro LBEAR address - ALTERNATIVE "nop", ".insn s,0xb2000000,\address", 193 + ALTERNATIVE "nop", ".insn s,0xb2000000,\address", ALT_FACILITY(193) .endm - .macro LPSWEY address,lpswe - ALTERNATIVE "b \lpswe; nopr", ".insn siy,0xeb0000000071,\address,0", 193 + .macro LPSWEY address, lpswe + ALTERNATIVE_2 "b \lpswe;nopr", \ + ".insn siy,0xeb0000000071,\address,0", ALT_FACILITY_EARLY(193), \ + __stringify(.insn siy,0xeb0000000071,LOWCORE_ALT_ADDRESS+\address,0), \ + ALT_LOWCORE .endm - .macro MBEAR reg - ALTERNATIVE "brcl 0,0", __stringify(mvc __PT_LAST_BREAK(8,\reg),__LC_LAST_BREAK), 193 + .macro MBEAR reg, lowcore + ALTERNATIVE "brcl 0,0", __stringify(mvc __PT_LAST_BREAK(8,\reg),__LC_LAST_BREAK(\lowcore)),\ + ALT_FACILITY(193) .endm - .macro CHECK_STACK savearea + .macro CHECK_STACK savearea, lowcore #ifdef CONFIG_CHECK_STACK tml %r15,THREAD_SIZE - CONFIG_STACK_GUARD - lghi %r14,\savearea + la %r14,\savearea(\lowcore) jz stack_overflow #endif .endm - .macro CHECK_VMAP_STACK savearea,oklabel + .macro CHECK_VMAP_STACK savearea, lowcore, oklabel #ifdef CONFIG_VMAP_STACK lgr %r14,%r15 nill %r14,0x10000 - THREAD_SIZE oill %r14,STACK_INIT_OFFSET - clg %r14,__LC_KERNEL_STACK + clg %r14,__LC_KERNEL_STACK(\lowcore) je \oklabel - clg %r14,__LC_ASYNC_STACK + clg %r14,__LC_ASYNC_STACK(\lowcore) je \oklabel - clg %r14,__LC_MCCK_STACK + clg %r14,__LC_MCCK_STACK(\lowcore) je \oklabel - clg %r14,__LC_NODAT_STACK + clg %r14,__LC_NODAT_STACK(\lowcore) je \oklabel - clg %r14,__LC_RESTART_STACK + clg %r14,__LC_RESTART_STACK(\lowcore) je \oklabel - lghi %r14,\savearea + la %r14,\savearea(\lowcore) j stack_overflow #else j \oklabel @@ -100,30 +105,31 @@ _LPP_OFFSET = __LC_LPP .endm .macro BPOFF - ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,12,0", 82 + ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,12,0", ALT_SPEC(82) .endm .macro BPON - ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,13,0", 82 + ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82) .endm .macro BPENTER tif_ptr,tif_mask ALTERNATIVE "TSTMSK \tif_ptr,\tif_mask; jz .+8; .insn rrf,0xb2e80000,0,0,13,0", \ - "j .+12; nop; nop", 82 + "j .+12; nop; nop", ALT_SPEC(82) .endm .macro BPEXIT tif_ptr,tif_mask TSTMSK \tif_ptr,\tif_mask ALTERNATIVE "jz .+8; .insn rrf,0xb2e80000,0,0,12,0", \ - "jnz .+8; .insn rrf,0xb2e80000,0,0,13,0", 82 + "jnz .+8; .insn rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82) .endm #if IS_ENABLED(CONFIG_KVM) - .macro SIEEXIT sie_control - lg %r9,\sie_control # get control block pointer - ni __SIE_PROG0C+3(%r9),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce - ni __LC_CPU_FLAGS+7,255-_CIF_SIE + .macro SIEEXIT sie_control,lowcore + lg %r9,\sie_control # get control block pointer + ni __SIE_PROG0C+3(%r9),0xfe # no longer in SIE + lctlg %c1,%c1,__LC_KERNEL_ASCE(\lowcore) # load primary asce + lg %r9,__LC_CURRENT(\lowcore) + mvi __TI_sie(%r9),0 larl %r9,sie_exit # skip forward to sie_exit .endm #endif @@ -163,13 +169,14 @@ SYM_FUNC_START(__switch_to_asm) stg %r15,__THREAD_ksp(%r1,%r2) # store kernel stack of prev lg %r15,0(%r4,%r3) # start of kernel stack of next agr %r15,%r5 # end of kernel stack of next - stg %r3,__LC_CURRENT # store task struct of next - stg %r15,__LC_KERNEL_STACK # store end of kernel stack + GET_LC %r13 + stg %r3,__LC_CURRENT(%r13) # store task struct of next + stg %r15,__LC_KERNEL_STACK(%r13) # store end of kernel stack lg %r15,__THREAD_ksp(%r1,%r3) # load kernel stack of next aghi %r3,__TASK_pid - mvc __LC_CURRENT_PID(4,%r0),0(%r3) # store pid of next + mvc __LC_CURRENT_PID(4,%r13),0(%r3) # store pid of next + ALTERNATIVE "nop", "lpp _LPP_OFFSET(%r13)", ALT_FACILITY(40) lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task - ALTERNATIVE "nop", "lpp _LPP_OFFSET", 40 BR_EX %r14 SYM_FUNC_END(__switch_to_asm) @@ -179,22 +186,21 @@ SYM_FUNC_END(__switch_to_asm) * %r2 pointer to sie control block phys * %r3 pointer to sie control block virt * %r4 guest register save area + * %r5 guest asce */ SYM_FUNC_START(__sie64a) stmg %r6,%r14,__SF_GPRS(%r15) # save kernel registers - lg %r12,__LC_CURRENT + GET_LC %r13 + lg %r14,__LC_CURRENT(%r13) stg %r2,__SF_SIE_CONTROL_PHYS(%r15) # save sie block physical.. stg %r3,__SF_SIE_CONTROL(%r15) # ...and virtual addresses stg %r4,__SF_SIE_SAVEAREA(%r15) # save guest register save area + stg %r5,__SF_SIE_GUEST_ASCE(%r15) # save guest asce xc __SF_SIE_REASON(8,%r15),__SF_SIE_REASON(%r15) # reason code = 0 - mvc __SF_SIE_FLAGS(8,%r15),__TI_flags(%r12) # copy thread flags + mvc __SF_SIE_FLAGS(8,%r15),__TI_flags(%r14) # copy thread flags lmg %r0,%r13,0(%r4) # load guest gprs 0-13 - lg %r14,__LC_GMAP # get gmap pointer - ltgr %r14,%r14 - jz .Lsie_gmap - oi __LC_CPU_FLAGS+7,_CIF_SIE - lctlg %c1,%c1,__GMAP_ASCE(%r14) # load primary asce -.Lsie_gmap: + mvi __TI_sie(%r14),1 + lctlg %c1,%c1,__SF_SIE_GUEST_ASCE(%r15) # load primary asce lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer oi __SIE_PROG0C+3(%r14),1 # we are going into SIE now tm __SIE_PROG20+3(%r14),3 # last exit... @@ -212,8 +218,10 @@ SYM_FUNC_START(__sie64a) .Lsie_skip: lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE - lctlg %c1,%c1,__LC_KERNEL_ASCE # load primary asce - ni __LC_CPU_FLAGS+7,255-_CIF_SIE + GET_LC %r14 + lctlg %c1,%c1,__LC_KERNEL_ASCE(%r14) # load primary asce + lg %r14,__LC_CURRENT(%r14) + mvi __TI_sie(%r14),0 # some program checks are suppressing. C code (e.g. do_protection_exception) # will rewind the PSW by the ILC, which is often 4 bytes in case of SIE. There # are some corner cases (e.g. runtime instrumentation) where ILC is unpredictable. @@ -256,14 +264,15 @@ EXPORT_SYMBOL(sie_exit) */ SYM_CODE_START(system_call) - stpt __LC_SYS_ENTER_TIMER - stmg %r8,%r15,__LC_SAVE_AREA_SYNC + STMG_LC %r8,%r15,__LC_SAVE_AREA_SYNC + GET_LC %r13 + stpt __LC_SYS_ENTER_TIMER(%r13) BPOFF lghi %r14,0 .Lsysc_per: - STBEAR __LC_LAST_BREAK - lctlg %c1,%c1,__LC_KERNEL_ASCE - lg %r15,__LC_KERNEL_STACK + STBEAR __LC_LAST_BREAK(%r13) + lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13) + lg %r15,__LC_KERNEL_STACK(%r13) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) stmg %r0,%r7,STACK_FRAME_OVERHEAD+__PT_R0(%r15) # clear user controlled register to prevent speculative use @@ -278,17 +287,17 @@ SYM_CODE_START(system_call) xgr %r10,%r10 xgr %r11,%r11 la %r2,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs - mvc __PT_R8(64,%r2),__LC_SAVE_AREA_SYNC - MBEAR %r2 + mvc __PT_R8(64,%r2),__LC_SAVE_AREA_SYNC(%r13) + MBEAR %r2,%r13 lgr %r3,%r14 brasl %r14,__do_syscall STACKLEAK_ERASE - lctlg %c1,%c1,__LC_USER_ASCE - mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) + lctlg %c1,%c1,__LC_USER_ASCE(%r13) + mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) BPON LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15) + stpt __LC_EXIT_TIMER(%r13) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) - stpt __LC_EXIT_TIMER LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE SYM_CODE_END(system_call) @@ -299,12 +308,13 @@ SYM_CODE_START(ret_from_fork) lgr %r3,%r11 brasl %r14,__ret_from_fork STACKLEAK_ERASE - lctlg %c1,%c1,__LC_USER_ASCE - mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) + GET_LC %r13 + lctlg %c1,%c1,__LC_USER_ASCE(%r13) + mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) BPON LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15) + stpt __LC_EXIT_TIMER(%r13) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) - stpt __LC_EXIT_TIMER LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE SYM_CODE_END(ret_from_fork) @@ -313,39 +323,40 @@ SYM_CODE_END(ret_from_fork) */ SYM_CODE_START(pgm_check_handler) - stpt __LC_SYS_ENTER_TIMER + STMG_LC %r8,%r15,__LC_SAVE_AREA_SYNC + GET_LC %r13 + stpt __LC_SYS_ENTER_TIMER(%r13) BPOFF - stmg %r8,%r15,__LC_SAVE_AREA_SYNC lgr %r10,%r15 - lmg %r8,%r9,__LC_PGM_OLD_PSW + lmg %r8,%r9,__LC_PGM_OLD_PSW(%r13) tmhh %r8,0x0001 # coming from user space? jno .Lpgm_skip_asce - lctlg %c1,%c1,__LC_KERNEL_ASCE + lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13) j 3f # -> fault in user space .Lpgm_skip_asce: 1: tmhh %r8,0x4000 # PER bit set in old PSW ? jnz 2f # -> enabled, can't be a double fault - tm __LC_PGM_ILC+3,0x80 # check for per exception + tm __LC_PGM_ILC+3(%r13),0x80 # check for per exception jnz .Lpgm_svcper # -> single stepped svc -2: CHECK_STACK __LC_SAVE_AREA_SYNC +2: CHECK_STACK __LC_SAVE_AREA_SYNC,%r13 aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) # CHECK_VMAP_STACK branches to stack_overflow or 4f - CHECK_VMAP_STACK __LC_SAVE_AREA_SYNC,4f -3: lg %r15,__LC_KERNEL_STACK + CHECK_VMAP_STACK __LC_SAVE_AREA_SYNC,%r13,4f +3: lg %r15,__LC_KERNEL_STACK(%r13) 4: la %r11,STACK_FRAME_OVERHEAD(%r15) xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) stmg %r0,%r7,__PT_R0(%r11) - mvc __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC - mvc __PT_LAST_BREAK(8,%r11),__LC_PGM_LAST_BREAK + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC(%r13) + mvc __PT_LAST_BREAK(8,%r11),__LC_PGM_LAST_BREAK(%r13) stctg %c1,%c1,__PT_CR1(%r11) #if IS_ENABLED(CONFIG_KVM) - ltg %r12,__LC_GMAP + ltg %r12,__LC_GMAP(%r13) jz 5f clc __GMAP_ASCE(8,%r12), __PT_CR1(%r11) jne 5f BPENTER __SF_SIE_FLAGS(%r10),_TIF_ISOLATE_BP_GUEST - SIEEXIT __SF_SIE_CONTROL(%r10) + SIEEXIT __SF_SIE_CONTROL(%r10),%r13 #endif 5: stmg %r8,%r9,__PT_PSW(%r11) # clear user controlled registers to prevent speculative use @@ -361,11 +372,11 @@ SYM_CODE_START(pgm_check_handler) tmhh %r8,0x0001 # returning to user space? jno .Lpgm_exit_kernel STACKLEAK_ERASE - lctlg %c1,%c1,__LC_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE(%r13) BPON - stpt __LC_EXIT_TIMER + stpt __LC_EXIT_TIMER(%r13) .Lpgm_exit_kernel: - mvc __LC_RETURN_PSW(16),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) + mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15) LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15) lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15) LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE @@ -374,11 +385,11 @@ SYM_CODE_START(pgm_check_handler) # single stepped system call # .Lpgm_svcper: - mvc __LC_RETURN_PSW(8),__LC_SVC_NEW_PSW + mvc __LC_RETURN_PSW(8,%r13),__LC_SVC_NEW_PSW(%r13) larl %r14,.Lsysc_per - stg %r14,__LC_RETURN_PSW+8 + stg %r14,__LC_RETURN_PSW+8(%r13) lghi %r14,1 - LBEAR __LC_PGM_LAST_BREAK + LBEAR __LC_PGM_LAST_BREAK(%r13) LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE # branch to .Lsysc_per SYM_CODE_END(pgm_check_handler) @@ -387,25 +398,27 @@ SYM_CODE_END(pgm_check_handler) */ .macro INT_HANDLER name,lc_old_psw,handler SYM_CODE_START(\name) - stckf __LC_INT_CLOCK - stpt __LC_SYS_ENTER_TIMER - STBEAR __LC_LAST_BREAK + STMG_LC %r8,%r15,__LC_SAVE_AREA_ASYNC + GET_LC %r13 + stckf __LC_INT_CLOCK(%r13) + stpt __LC_SYS_ENTER_TIMER(%r13) + STBEAR __LC_LAST_BREAK(%r13) BPOFF - stmg %r8,%r15,__LC_SAVE_AREA_ASYNC - lmg %r8,%r9,\lc_old_psw + lmg %r8,%r9,\lc_old_psw(%r13) tmhh %r8,0x0001 # interrupting from user ? jnz 1f #if IS_ENABLED(CONFIG_KVM) - TSTMSK __LC_CPU_FLAGS,_CIF_SIE + lg %r10,__LC_CURRENT(%r13) + tm __TI_sie(%r10),0xff jz 0f BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST - SIEEXIT __SF_SIE_CONTROL(%r15) + SIEEXIT __SF_SIE_CONTROL(%r15),%r13 #endif -0: CHECK_STACK __LC_SAVE_AREA_ASYNC +0: CHECK_STACK __LC_SAVE_AREA_ASYNC,%r13 aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) j 2f -1: lctlg %c1,%c1,__LC_KERNEL_ASCE - lg %r15,__LC_KERNEL_STACK +1: lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13) + lg %r15,__LC_KERNEL_STACK(%r13) 2: xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) la %r11,STACK_FRAME_OVERHEAD(%r15) stmg %r0,%r7,__PT_R0(%r11) @@ -419,18 +432,18 @@ SYM_CODE_START(\name) xgr %r7,%r7 xgr %r10,%r10 xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) - mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC - MBEAR %r11 + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC(%r13) + MBEAR %r11,%r13 stmg %r8,%r9,__PT_PSW(%r11) lgr %r2,%r11 # pass pointer to pt_regs brasl %r14,\handler - mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) + mvc __LC_RETURN_PSW(16,%r13),__PT_PSW(%r11) tmhh %r8,0x0001 # returning to user ? jno 2f STACKLEAK_ERASE - lctlg %c1,%c1,__LC_USER_ASCE + lctlg %c1,%c1,__LC_USER_ASCE(%r13) BPON - stpt __LC_EXIT_TIMER + stpt __LC_EXIT_TIMER(%r13) 2: LBEAR __PT_LAST_BREAK(%r11) lmg %r0,%r15,__PT_R0(%r11) LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE @@ -445,35 +458,37 @@ INT_HANDLER io_int_handler,__LC_IO_OLD_PSW,do_io_irq */ SYM_CODE_START(mcck_int_handler) BPOFF - lmg %r8,%r9,__LC_MCK_OLD_PSW - TSTMSK __LC_MCCK_CODE,MCCK_CODE_SYSTEM_DAMAGE + GET_LC %r13 + lmg %r8,%r9,__LC_MCK_OLD_PSW(%r13) + TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_SYSTEM_DAMAGE jo .Lmcck_panic # yes -> rest of mcck code invalid - TSTMSK __LC_MCCK_CODE,MCCK_CODE_CR_VALID + TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_CR_VALID jno .Lmcck_panic # control registers invalid -> panic ptlb - lghi %r14,__LC_CPU_TIMER_SAVE_AREA - mvc __LC_MCCK_ENTER_TIMER(8),0(%r14) - TSTMSK __LC_MCCK_CODE,MCCK_CODE_CPU_TIMER_VALID + lay %r14,__LC_CPU_TIMER_SAVE_AREA(%r13) + mvc __LC_MCCK_ENTER_TIMER(8,%r13),0(%r14) + TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_CPU_TIMER_VALID jo 3f - la %r14,__LC_SYS_ENTER_TIMER - clc 0(8,%r14),__LC_EXIT_TIMER + la %r14,__LC_SYS_ENTER_TIMER(%r13) + clc 0(8,%r14),__LC_EXIT_TIMER(%r13) jl 1f - la %r14,__LC_EXIT_TIMER -1: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER + la %r14,__LC_EXIT_TIMER(%r13) +1: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER(%r13) jl 2f - la %r14,__LC_LAST_UPDATE_TIMER + la %r14,__LC_LAST_UPDATE_TIMER(%r13) 2: spt 0(%r14) - mvc __LC_MCCK_ENTER_TIMER(8),0(%r14) -3: TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_MWP_VALID + mvc __LC_MCCK_ENTER_TIMER(8,%r13),0(%r14) +3: TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_PSW_MWP_VALID jno .Lmcck_panic tmhh %r8,0x0001 # interrupting from user ? jnz .Lmcck_user - TSTMSK __LC_MCCK_CODE,MCCK_CODE_PSW_IA_VALID + TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_PSW_IA_VALID jno .Lmcck_panic #if IS_ENABLED(CONFIG_KVM) - TSTMSK __LC_CPU_FLAGS,_CIF_SIE + lg %r10,__LC_CURRENT(%r13) + tm __TI_sie(%r10),0xff jz .Lmcck_user - # Need to compare the address instead of a CIF_SIE* flag. + # Need to compare the address instead of __TI_SIE flag. # Otherwise there would be a race between setting the flag # and entering SIE (or leaving and clearing the flag). This # would cause machine checks targeted at the guest to be @@ -482,18 +497,19 @@ SYM_CODE_START(mcck_int_handler) clgrjl %r9,%r14, 4f larl %r14,.Lsie_leave clgrjhe %r9,%r14, 4f - oi __LC_CPU_FLAGS+7, _CIF_MCCK_GUEST + lg %r10,__LC_PCPU + oi __PCPU_FLAGS+7(%r10), _CIF_MCCK_GUEST 4: BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST - SIEEXIT __SF_SIE_CONTROL(%r15) + SIEEXIT __SF_SIE_CONTROL(%r15),%r13 #endif .Lmcck_user: - lg %r15,__LC_MCCK_STACK + lg %r15,__LC_MCCK_STACK(%r13) la %r11,STACK_FRAME_OVERHEAD(%r15) stctg %c1,%c1,__PT_CR1(%r11) - lctlg %c1,%c1,__LC_KERNEL_ASCE + lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) - lghi %r14,__LC_GPREGS_SAVE_AREA+64 - stmg %r0,%r7,__PT_R0(%r11) + lay %r14,__LC_GPREGS_SAVE_AREA(%r13) + mvc __PT_R0(128,%r11),0(%r14) # clear user controlled registers to prevent speculative use xgr %r0,%r0 xgr %r1,%r1 @@ -503,7 +519,6 @@ SYM_CODE_START(mcck_int_handler) xgr %r6,%r6 xgr %r7,%r7 xgr %r10,%r10 - mvc __PT_R8(64,%r11),0(%r14) stmg %r8,%r9,__PT_PSW(%r11) xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) @@ -511,12 +526,13 @@ SYM_CODE_START(mcck_int_handler) brasl %r14,s390_do_machine_check lctlg %c1,%c1,__PT_CR1(%r11) lmg %r0,%r10,__PT_R0(%r11) - mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW - tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? + mvc __LC_RETURN_MCCK_PSW(16,%r13),__PT_PSW(%r11) # move return PSW + tm __LC_RETURN_MCCK_PSW+1(%r13),0x01 # returning to user ? jno 0f BPON - stpt __LC_EXIT_TIMER -0: ALTERNATIVE "nop", __stringify(lghi %r12,__LC_LAST_BREAK_SAVE_AREA),193 + stpt __LC_EXIT_TIMER(%r13) +0: ALTERNATIVE "brcl 0,0", __stringify(lay %r12,__LC_LAST_BREAK_SAVE_AREA(%r13)),\ + ALT_FACILITY(193) LBEAR 0(%r12) lmg %r11,%r15,__PT_R11(%r11) LPSWEY __LC_RETURN_MCCK_PSW,__LC_RETURN_MCCK_LPSWE @@ -552,7 +568,7 @@ SYM_CODE_START(mcck_int_handler) SYM_CODE_END(mcck_int_handler) SYM_CODE_START(restart_int_handler) - ALTERNATIVE "nop", "lpp _LPP_OFFSET", 40 + ALTERNATIVE "nop", "lpp _LPP_OFFSET", ALT_FACILITY(40) stg %r15,__LC_SAVE_AREA_RESTART TSTMSK __LC_RESTART_FLAGS,RESTART_FLAG_CTLREGS,4 jz 0f @@ -560,15 +576,17 @@ SYM_CODE_START(restart_int_handler) 0: larl %r15,daton_psw lpswe 0(%r15) # turn dat on, keep irqs off .Ldaton: - lg %r15,__LC_RESTART_STACK + GET_LC %r15 + lg %r15,__LC_RESTART_STACK(%r15) 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 + GET_LC %r13 + mvc STACK_FRAME_OVERHEAD+__PT_R15(8,%r15),__LC_SAVE_AREA_RESTART(%r13) + mvc STACK_FRAME_OVERHEAD+__PT_PSW(16,%r15),__LC_RST_OLD_PSW(%r13) xc 0(STACK_FRAME_OVERHEAD,%r15),0(%r15) - lg %r1,__LC_RESTART_FN # load fn, parm & source cpu - lg %r2,__LC_RESTART_DATA - lgf %r3,__LC_RESTART_SOURCE + lg %r1,__LC_RESTART_FN(%r13) # load fn, parm & source cpu + lg %r2,__LC_RESTART_DATA(%r13) + lgf %r3,__LC_RESTART_SOURCE(%r13) ltgr %r3,%r3 # test source cpu address jm 1f # negative -> skip source stop 0: sigp %r4,%r3,SIGP_SENSE # sigp sense to source cpu @@ -590,7 +608,8 @@ SYM_CODE_END(restart_int_handler) * Setup a pt_regs so that show_trace can provide a good call trace. */ SYM_CODE_START(stack_overflow) - lg %r15,__LC_NODAT_STACK # change to panic stack + GET_LC %r15 + lg %r15,__LC_NODAT_STACK(%r15) # 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/fpu.c b/arch/s390/kernel/fpu.c index fa90bbdc5ef9..6f2e87920288 100644 --- a/arch/s390/kernel/fpu.c +++ b/arch/s390/kernel/fpu.c @@ -113,7 +113,7 @@ void load_fpu_state(struct fpu *state, int flags) int mask; if (flags & KERNEL_FPC) - fpu_lfpc(&state->fpc); + fpu_lfpc_safe(&state->fpc); if (!cpu_has_vx()) { if (flags & KERNEL_VXR_V0V7) load_fp_regs_vx(state->vxrs); diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c index ddf2ee47cb87..0bd6adc40a34 100644 --- a/arch/s390/kernel/ftrace.c +++ b/arch/s390/kernel/ftrace.c @@ -12,6 +12,7 @@ #include <linux/ftrace.h> #include <linux/kernel.h> #include <linux/types.h> +#include <linux/kmsan-checks.h> #include <linux/kprobes.h> #include <linux/execmem.h> #include <trace/syscall.h> @@ -303,6 +304,7 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip, if (bit < 0) return; + kmsan_unpoison_memory(fregs, sizeof(*fregs)); regs = ftrace_get_regs(fregs); p = get_kprobe((kprobe_opcode_t *)ip); if (!regs || unlikely(!p) || kprobe_disabled(p)) diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index 45413b04efc5..396034b2fe67 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -10,6 +10,7 @@ #include <linux/init.h> #include <linux/linkage.h> +#include <asm/lowcore.h> #include <asm/asm-offsets.h> #include <asm/thread_info.h> #include <asm/page.h> @@ -18,14 +19,15 @@ __HEAD SYM_CODE_START(startup_continue) larl %r1,tod_clock_base - mvc 0(16,%r1),__LC_BOOT_CLOCK + GET_LC %r2 + mvc 0(16,%r1),__LC_BOOT_CLOCK(%r2) # # Setup stack # larl %r14,init_task - stg %r14,__LC_CURRENT + stg %r14,__LC_CURRENT(%r2) larl %r15,init_thread_union+STACK_INIT_OFFSET - stg %r15,__LC_KERNEL_STACK + stg %r15,__LC_KERNEL_STACK(%r2) brasl %r14,sclp_early_adjust_va # allow sclp_early_printk brasl %r14,startup_init # s390 specific early init brasl %r14,start_kernel # common init code diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c index af9c97c0ad73..39cb8d0ae348 100644 --- a/arch/s390/kernel/idle.c +++ b/arch/s390/kernel/idle.c @@ -24,6 +24,7 @@ static DEFINE_PER_CPU(struct s390_idle_data, s390_idle); void account_idle_time_irq(void) { struct s390_idle_data *idle = this_cpu_ptr(&s390_idle); + struct lowcore *lc = get_lowcore(); unsigned long idle_time; u64 cycles_new[8]; int i; @@ -34,13 +35,13 @@ void account_idle_time_irq(void) this_cpu_add(mt_cycles[i], cycles_new[i] - idle->mt_cycles_enter[i]); } - idle_time = S390_lowcore.int_clock - idle->clock_idle_enter; + idle_time = lc->int_clock - idle->clock_idle_enter; - S390_lowcore.steal_timer += idle->clock_idle_enter - S390_lowcore.last_update_clock; - S390_lowcore.last_update_clock = S390_lowcore.int_clock; + lc->steal_timer += idle->clock_idle_enter - lc->last_update_clock; + lc->last_update_clock = lc->int_clock; - S390_lowcore.system_timer += S390_lowcore.last_update_timer - idle->timer_idle_enter; - S390_lowcore.last_update_timer = S390_lowcore.sys_enter_timer; + lc->system_timer += lc->last_update_timer - idle->timer_idle_enter; + lc->last_update_timer = lc->sys_enter_timer; /* Account time spent with enabled wait psw loaded as idle time. */ WRITE_ONCE(idle->idle_time, READ_ONCE(idle->idle_time) + idle_time); diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 3a7d6e172211..f17bb7bf9392 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -2112,7 +2112,7 @@ void do_restart(void *arg) tracing_off(); debug_locks_off(); lgr_info_log(); - smp_call_online_cpu(__do_restart, arg); + smp_call_ipl_cpu(__do_restart, arg); } /* on halt */ diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c index 9acc6630abd3..1af5a08d72ab 100644 --- a/arch/s390/kernel/irq.c +++ b/arch/s390/kernel/irq.c @@ -100,8 +100,8 @@ static const struct irq_class irqclass_sub_desc[] = { static void do_IRQ(struct pt_regs *regs, int irq) { - if (tod_after_eq(S390_lowcore.int_clock, - S390_lowcore.clock_comparator)) + if (tod_after_eq(get_lowcore()->int_clock, + get_lowcore()->clock_comparator)) /* Serve timer interrupts first. */ clock_comparator_work(); generic_handle_irq(irq); @@ -111,7 +111,7 @@ static int on_async_stack(void) { unsigned long frame = current_frame_address(); - return ((S390_lowcore.async_stack ^ frame) & ~(THREAD_SIZE - 1)) == 0; + return ((get_lowcore()->async_stack ^ frame) & ~(THREAD_SIZE - 1)) == 0; } static void do_irq_async(struct pt_regs *regs, int irq) @@ -119,7 +119,7 @@ static void do_irq_async(struct pt_regs *regs, int irq) if (on_async_stack()) { do_IRQ(regs, irq); } else { - call_on_stack(2, S390_lowcore.async_stack, void, do_IRQ, + call_on_stack(2, get_lowcore()->async_stack, void, do_IRQ, struct pt_regs *, regs, int, irq); } } @@ -153,8 +153,8 @@ void noinstr do_io_irq(struct pt_regs *regs) set_cpu_flag(CIF_NOHZ_DELAY); do { - regs->tpi_info = S390_lowcore.tpi_info; - if (S390_lowcore.tpi_info.adapter_IO) + regs->tpi_info = get_lowcore()->tpi_info; + if (get_lowcore()->tpi_info.adapter_IO) do_irq_async(regs, THIN_INTERRUPT); else do_irq_async(regs, IO_INTERRUPT); @@ -183,9 +183,9 @@ void noinstr do_ext_irq(struct pt_regs *regs) current->thread.last_break = regs->last_break; } - regs->int_code = S390_lowcore.ext_int_code_addr; - regs->int_parm = S390_lowcore.ext_params; - regs->int_parm_long = S390_lowcore.ext_params2; + regs->int_code = get_lowcore()->ext_int_code_addr; + regs->int_parm = get_lowcore()->ext_params; + regs->int_parm_long = get_lowcore()->ext_params2; from_idle = test_and_clear_cpu_flag(CIF_ENABLED_WAIT); if (from_idle) diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 3aee98efc374..8f681ccfb83a 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -52,7 +52,7 @@ static void __do_machine_kdump(void *data) purgatory = (purgatory_t)image->start; /* store_status() saved the prefix register to lowcore */ - prefix = (unsigned long) S390_lowcore.prefixreg_save_area; + prefix = (unsigned long)get_lowcore()->prefixreg_save_area; /* Now do the reset */ s390_reset_system(); @@ -62,7 +62,7 @@ static void __do_machine_kdump(void *data) * This need to be done *after* s390_reset_system set the * prefix register of this CPU to zero */ - memcpy(absolute_pointer(__LC_FPREGS_SAVE_AREA), + memcpy(absolute_pointer(get_lowcore()->floating_pt_save_area), phys_to_virt(prefix + __LC_FPREGS_SAVE_AREA), 512); call_nodat(1, int, purgatory, int, 1); @@ -91,7 +91,7 @@ static noinline void __machine_kdump(void *image) continue; } /* Store status of the boot CPU */ - mcesa = __va(S390_lowcore.mcesad & MCESA_ORIGIN_MASK); + mcesa = __va(get_lowcore()->mcesad & MCESA_ORIGIN_MASK); if (cpu_has_vx()) save_vx_regs((__vector128 *) mcesa->vector_save_area); if (MACHINE_HAS_GS) { diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index 230d010bac9b..fbd218b6fc8e 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -117,6 +117,7 @@ static __always_inline char *u64_to_hex(char *dest, u64 val) static notrace void s390_handle_damage(void) { + struct lowcore *lc = get_lowcore(); union ctlreg0 cr0, cr0_new; char message[100]; psw_t psw_save; @@ -125,7 +126,7 @@ static notrace void s390_handle_damage(void) smp_emergency_stop(); diag_amode31_ops.diag308_reset(); ptr = nmi_puts(message, "System stopped due to unrecoverable machine check, code: 0x"); - u64_to_hex(ptr, S390_lowcore.mcck_interruption_code); + u64_to_hex(ptr, lc->mcck_interruption_code); /* * Disable low address protection and make machine check new PSW a @@ -135,17 +136,17 @@ static notrace void s390_handle_damage(void) cr0_new = cr0; cr0_new.lap = 0; local_ctl_load(0, &cr0_new.reg); - psw_save = S390_lowcore.mcck_new_psw; - psw_bits(S390_lowcore.mcck_new_psw).io = 0; - psw_bits(S390_lowcore.mcck_new_psw).ext = 0; - psw_bits(S390_lowcore.mcck_new_psw).wait = 1; + psw_save = lc->mcck_new_psw; + psw_bits(lc->mcck_new_psw).io = 0; + psw_bits(lc->mcck_new_psw).ext = 0; + psw_bits(lc->mcck_new_psw).wait = 1; sclp_emergency_printk(message); /* * Restore machine check new PSW and control register 0 to original * values. This makes possible system dump analysis easier. */ - S390_lowcore.mcck_new_psw = psw_save; + lc->mcck_new_psw = psw_save; local_ctl_load(0, &cr0.reg); disabled_wait(); while (1); @@ -226,7 +227,7 @@ static bool notrace nmi_registers_valid(union mci mci) /* * Set the clock comparator register to the next expected value. */ - set_clock_comparator(S390_lowcore.clock_comparator); + set_clock_comparator(get_lowcore()->clock_comparator); if (!mci.gr || !mci.fp || !mci.fc) return false; /* @@ -252,7 +253,7 @@ static bool notrace nmi_registers_valid(union mci mci) * check handling must take care of this. The host values are saved by * KVM and are not affected. */ - cr2.reg = S390_lowcore.cregs_save_area[2]; + cr2.reg = get_lowcore()->cregs_save_area[2]; if (cr2.gse && !mci.gs && !test_cpu_flag(CIF_MCCK_GUEST)) return false; if (!mci.ms || !mci.pm || !mci.ia) @@ -278,11 +279,10 @@ static void notrace s390_backup_mcck_info(struct pt_regs *regs) sie_page = container_of(sie_block, struct sie_page, sie_block); mcck_backup = &sie_page->mcck_info; - mcck_backup->mcic = S390_lowcore.mcck_interruption_code & + mcck_backup->mcic = get_lowcore()->mcck_interruption_code & ~(MCCK_CODE_CP | MCCK_CODE_EXT_DAMAGE); - mcck_backup->ext_damage_code = S390_lowcore.external_damage_code; - mcck_backup->failing_storage_address - = S390_lowcore.failing_storage_address; + mcck_backup->ext_damage_code = get_lowcore()->external_damage_code; + mcck_backup->failing_storage_address = get_lowcore()->failing_storage_address; } NOKPROBE_SYMBOL(s390_backup_mcck_info); @@ -302,6 +302,7 @@ void notrace s390_do_machine_check(struct pt_regs *regs) static int ipd_count; static DEFINE_SPINLOCK(ipd_lock); static unsigned long long last_ipd; + struct lowcore *lc = get_lowcore(); struct mcck_struct *mcck; unsigned long long tmp; irqentry_state_t irq_state; @@ -314,7 +315,7 @@ void notrace s390_do_machine_check(struct pt_regs *regs) if (user_mode(regs)) update_timer_mcck(); inc_irq_stat(NMI_NMI); - mci.val = S390_lowcore.mcck_interruption_code; + mci.val = lc->mcck_interruption_code; mcck = this_cpu_ptr(&cpu_mcck); /* @@ -382,9 +383,9 @@ void notrace s390_do_machine_check(struct pt_regs *regs) } if (mci.ed && mci.ec) { /* External damage */ - if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC)) + if (lc->external_damage_code & (1U << ED_STP_SYNC)) mcck->stp_queue |= stp_sync_check(); - if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND)) + if (lc->external_damage_code & (1U << ED_STP_ISLAND)) mcck->stp_queue |= stp_island_check(); mcck_pending = 1; } diff --git a/arch/s390/kernel/nospec-branch.c b/arch/s390/kernel/nospec-branch.c index 9b8c24ebb008..e11ec15960a1 100644 --- a/arch/s390/kernel/nospec-branch.c +++ b/arch/s390/kernel/nospec-branch.c @@ -4,6 +4,8 @@ #include <linux/cpu.h> #include <asm/nospec-branch.h> +int nobp = IS_ENABLED(CONFIG_KERNEL_NOBP); + static int __init nobp_setup_early(char *str) { bool enabled; @@ -17,11 +19,11 @@ static int __init nobp_setup_early(char *str) * The user explicitly requested nobp=1, enable it and * disable the expoline support. */ - __set_facility(82, alt_stfle_fac_list); + nobp = 1; if (IS_ENABLED(CONFIG_EXPOLINE)) nospec_disable = 1; } else { - __clear_facility(82, alt_stfle_fac_list); + nobp = 0; } return 0; } @@ -29,7 +31,7 @@ early_param("nobp", nobp_setup_early); static int __init nospec_setup_early(char *str) { - __clear_facility(82, alt_stfle_fac_list); + nobp = 0; return 0; } early_param("nospec", nospec_setup_early); @@ -40,7 +42,7 @@ static int __init nospec_report(void) pr_info("Spectre V2 mitigation: etokens\n"); if (nospec_uses_trampoline()) pr_info("Spectre V2 mitigation: execute trampolines\n"); - if (__test_facility(82, alt_stfle_fac_list)) + if (nobp_enabled()) pr_info("Spectre V2 mitigation: limited branch prediction\n"); return 0; } @@ -66,14 +68,14 @@ void __init nospec_auto_detect(void) */ if (__is_defined(CC_USING_EXPOLINE)) nospec_disable = 1; - __clear_facility(82, alt_stfle_fac_list); + nobp = 0; } else if (__is_defined(CC_USING_EXPOLINE)) { /* * The kernel has been compiled with expolines. * Keep expolines enabled and disable nobp. */ nospec_disable = 0; - __clear_facility(82, alt_stfle_fac_list); + nobp = 0; } /* * If the kernel has not been compiled with expolines the @@ -86,7 +88,7 @@ static int __init spectre_v2_setup_early(char *str) { if (str && !strncmp(str, "on", 2)) { nospec_disable = 0; - __clear_facility(82, alt_stfle_fac_list); + nobp = 0; } if (str && !strncmp(str, "off", 3)) nospec_disable = 1; diff --git a/arch/s390/kernel/nospec-sysfs.c b/arch/s390/kernel/nospec-sysfs.c index 52d4353188ad..a95188818637 100644 --- a/arch/s390/kernel/nospec-sysfs.c +++ b/arch/s390/kernel/nospec-sysfs.c @@ -17,7 +17,7 @@ ssize_t cpu_show_spectre_v2(struct device *dev, return sprintf(buf, "Mitigation: etokens\n"); if (nospec_uses_trampoline()) return sprintf(buf, "Mitigation: execute trampolines\n"); - if (__test_facility(82, alt_stfle_fac_list)) + if (nobp_enabled()) return sprintf(buf, "Mitigation: limited branch prediction\n"); return sprintf(buf, "Vulnerable\n"); } diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c index 1434642e9cba..6968be98af11 100644 --- a/arch/s390/kernel/perf_cpum_cf.c +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -556,25 +556,31 @@ static int cfdiag_diffctr(struct cpu_cf_events *cpuhw, unsigned long auth) struct cf_trailer_entry *trailer_start, *trailer_stop; struct cf_ctrset_entry *ctrstart, *ctrstop; size_t offset = 0; + int i; - auth &= (1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1; - do { + for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i) { ctrstart = (struct cf_ctrset_entry *)(cpuhw->start + offset); ctrstop = (struct cf_ctrset_entry *)(cpuhw->stop + offset); + /* Counter set not authorized */ + if (!(auth & cpumf_ctr_ctl[i])) + continue; + /* Counter set size zero was not saved */ + if (!cpum_cf_read_setsize(i)) + continue; + if (memcmp(ctrstop, ctrstart, sizeof(*ctrstop))) { pr_err_once("cpum_cf_diag counter set compare error " "in set %i\n", ctrstart->set); return 0; } - auth &= ~cpumf_ctr_ctl[ctrstart->set]; if (ctrstart->def == CF_DIAG_CTRSET_DEF) { cfdiag_diffctrset((u64 *)(ctrstart + 1), (u64 *)(ctrstop + 1), ctrstart->ctr); offset += ctrstart->ctr * sizeof(u64) + sizeof(*ctrstart); } - } while (ctrstart->def && auth); + } /* Save time_stamp from start of event in stop's trailer */ trailer_start = (struct cf_trailer_entry *)(cpuhw->start + offset); diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 06efad5b4f93..736c1d9632dd 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -1022,7 +1022,7 @@ static void cpumsf_pmu_enable(struct pmu *pmu) } /* Load current program parameter */ - lpp(&S390_lowcore.lpp); + lpp(&get_lowcore()->lpp); debug_sprintf_event(sfdbg, 6, "%s: es %i cs %i ed %i cd %i " "interval %#lx tear %#lx dear %#lx\n", __func__, diff --git a/arch/s390/kernel/perf_pai_crypto.c b/arch/s390/kernel/perf_pai_crypto.c index 4ad472d130a3..2f5a20e300f6 100644 --- a/arch/s390/kernel/perf_pai_crypto.c +++ b/arch/s390/kernel/perf_pai_crypto.c @@ -36,8 +36,8 @@ struct paicrypt_map { struct pai_userdata *save; /* Page to store no-zero counters */ unsigned int active_events; /* # of PAI crypto users */ refcount_t refcnt; /* Reference count mapped buffers */ - enum paievt_mode mode; /* Type of event */ struct perf_event *event; /* Perf event for sampling */ + struct list_head syswide_list; /* List system-wide sampling events */ }; struct paicrypt_mapptr { @@ -84,20 +84,16 @@ static DEFINE_MUTEX(pai_reserve_mutex); /* Adjust usage counters and remove allocated memory when all users are * gone. */ -static void paicrypt_event_destroy(struct perf_event *event) +static void paicrypt_event_destroy_cpu(struct perf_event *event, int cpu) { - struct paicrypt_mapptr *mp = per_cpu_ptr(paicrypt_root.mapptr, - event->cpu); + struct paicrypt_mapptr *mp = per_cpu_ptr(paicrypt_root.mapptr, cpu); struct paicrypt_map *cpump = mp->mapptr; - static_branch_dec(&pai_key); mutex_lock(&pai_reserve_mutex); - debug_sprintf_event(cfm_dbg, 5, "%s event %#llx cpu %d users %d" - " mode %d refcnt %u\n", __func__, - event->attr.config, event->cpu, - cpump->active_events, cpump->mode, + debug_sprintf_event(cfm_dbg, 5, "%s event %#llx cpu %d users %d " + "refcnt %u\n", __func__, event->attr.config, + event->cpu, cpump->active_events, refcount_read(&cpump->refcnt)); - free_page(PAI_SAVE_AREA(event)); if (refcount_dec_and_test(&cpump->refcnt)) { debug_sprintf_event(cfm_dbg, 4, "%s page %#lx save %p\n", __func__, (unsigned long)cpump->page, @@ -111,6 +107,23 @@ static void paicrypt_event_destroy(struct perf_event *event) mutex_unlock(&pai_reserve_mutex); } +static void paicrypt_event_destroy(struct perf_event *event) +{ + int cpu; + + static_branch_dec(&pai_key); + free_page(PAI_SAVE_AREA(event)); + if (event->cpu == -1) { + struct cpumask *mask = PAI_CPU_MASK(event); + + for_each_cpu(cpu, mask) + paicrypt_event_destroy_cpu(event, cpu); + kfree(mask); + } else { + paicrypt_event_destroy_cpu(event, event->cpu); + } +} + static u64 paicrypt_getctr(unsigned long *page, int nr, bool kernel) { if (kernel) @@ -156,23 +169,15 @@ static u64 paicrypt_getall(struct perf_event *event) return sum; } -/* Used to avoid races in checking concurrent access of counting and - * sampling for crypto events - * - * Only one instance of event pai_crypto/CRYPTO_ALL/ for sampling is - * allowed and when this event is running, no counting event is allowed. - * Several counting events are allowed in parallel, but no sampling event - * is allowed while one (or more) counting events are running. - * +/* Check concurrent access of counting and sampling for crypto events. * This function is called in process context and it is save to block. * When the event initialization functions fails, no other call back will * be invoked. * * Allocate the memory for the event. */ -static struct paicrypt_map *paicrypt_busy(struct perf_event *event) +static struct paicrypt_map *paicrypt_busy(struct perf_event *event, int cpu) { - struct perf_event_attr *a = &event->attr; struct paicrypt_map *cpump = NULL; struct paicrypt_mapptr *mp; int rc; @@ -185,7 +190,7 @@ static struct paicrypt_map *paicrypt_busy(struct perf_event *event) goto unlock; /* Allocate node for this event */ - mp = per_cpu_ptr(paicrypt_root.mapptr, event->cpu); + mp = per_cpu_ptr(paicrypt_root.mapptr, cpu); cpump = mp->mapptr; if (!cpump) { /* Paicrypt_map allocated? */ cpump = kzalloc(sizeof(*cpump), GFP_KERNEL); @@ -193,25 +198,9 @@ static struct paicrypt_map *paicrypt_busy(struct perf_event *event) rc = -ENOMEM; goto free_root; } + INIT_LIST_HEAD(&cpump->syswide_list); } - if (a->sample_period) { /* Sampling requested */ - if (cpump->mode != PAI_MODE_NONE) - rc = -EBUSY; /* ... sampling/counting active */ - } else { /* Counting requested */ - if (cpump->mode == PAI_MODE_SAMPLING) - rc = -EBUSY; /* ... and sampling active */ - } - /* - * This error case triggers when there is a conflict: - * Either sampling requested and counting already active, or visa - * versa. Therefore the struct paicrypto_map for this CPU is - * needed or the error could not have occurred. Only adjust root - * node refcount. - */ - if (rc) - goto free_root; - /* Allocate memory for counter page and counter extraction. * Only the first counting event has to allocate a page. */ @@ -235,26 +224,58 @@ static struct paicrypt_map *paicrypt_busy(struct perf_event *event) /* Set mode and reference count */ rc = 0; refcount_set(&cpump->refcnt, 1); - cpump->mode = a->sample_period ? PAI_MODE_SAMPLING : PAI_MODE_COUNTING; mp->mapptr = cpump; - debug_sprintf_event(cfm_dbg, 5, "%s sample_period %#llx users %d" - " mode %d refcnt %u page %#lx save %p rc %d\n", - __func__, a->sample_period, cpump->active_events, - cpump->mode, refcount_read(&cpump->refcnt), + debug_sprintf_event(cfm_dbg, 5, "%s users %d refcnt %u page %#lx " + "save %p rc %d\n", __func__, cpump->active_events, + refcount_read(&cpump->refcnt), (unsigned long)cpump->page, cpump->save, rc); goto unlock; free_paicrypt_map: + /* Undo memory allocation */ kfree(cpump); mp->mapptr = NULL; free_root: paicrypt_root_free(); - unlock: mutex_unlock(&pai_reserve_mutex); return rc ? ERR_PTR(rc) : cpump; } +static int paicrypt_event_init_all(struct perf_event *event) +{ + struct paicrypt_map *cpump; + struct cpumask *maskptr; + int cpu, rc = -ENOMEM; + + maskptr = kzalloc(sizeof(*maskptr), GFP_KERNEL); + if (!maskptr) + goto out; + + for_each_online_cpu(cpu) { + cpump = paicrypt_busy(event, cpu); + if (IS_ERR(cpump)) { + for_each_cpu(cpu, maskptr) + paicrypt_event_destroy_cpu(event, cpu); + kfree(maskptr); + rc = PTR_ERR(cpump); + goto out; + } + cpumask_set_cpu(cpu, maskptr); + } + + /* + * On error all cpumask are freed and all events have been destroyed. + * Save of which CPUs data structures have been allocated for. + * Release them in paicrypt_event_destroy call back function + * for this event. + */ + PAI_CPU_MASK(event) = maskptr; + rc = 0; +out: + return rc; +} + /* Might be called on different CPU than the one the event is intended for. */ static int paicrypt_event_init(struct perf_event *event) { @@ -269,10 +290,7 @@ static int paicrypt_event_init(struct perf_event *event) if (a->config < PAI_CRYPTO_BASE || a->config > PAI_CRYPTO_BASE + paicrypt_cnt) return -EINVAL; - /* Allow only CPU wide operation, no process context for now. */ - if ((event->attach_state & PERF_ATTACH_TASK) || event->cpu == -1) - return -ENOENT; - /* Allow only CRYPTO_ALL for sampling. */ + /* Allow only CRYPTO_ALL for sampling */ if (a->sample_period && a->config != PAI_CRYPTO_BASE) return -EINVAL; /* Get a page to store last counter values for sampling */ @@ -284,13 +302,17 @@ static int paicrypt_event_init(struct perf_event *event) } } - cpump = paicrypt_busy(event); - if (IS_ERR(cpump)) { + if (event->cpu >= 0) { + cpump = paicrypt_busy(event, event->cpu); + if (IS_ERR(cpump)) + rc = PTR_ERR(cpump); + } else { + rc = paicrypt_event_init_all(event); + } + if (rc) { free_page(PAI_SAVE_AREA(event)); - rc = PTR_ERR(cpump); goto out; } - event->destroy = paicrypt_event_destroy; if (a->sample_period) { @@ -331,8 +353,14 @@ static void paicrypt_start(struct perf_event *event, int flags) sum = paicrypt_getall(event); /* Get current value */ local64_set(&event->hw.prev_count, sum); } else { /* Sampling */ - cpump->event = event; - perf_sched_cb_inc(event->pmu); + memcpy((void *)PAI_SAVE_AREA(event), cpump->page, PAGE_SIZE); + /* Enable context switch callback for system-wide sampling */ + if (!(event->attach_state & PERF_ATTACH_TASK)) { + list_add_tail(PAI_SWLIST(event), &cpump->syswide_list); + perf_sched_cb_inc(event->pmu); + } else { + cpump->event = event; + } } } @@ -344,7 +372,7 @@ static int paicrypt_add(struct perf_event *event, int flags) if (++cpump->active_events == 1) { ccd = virt_to_phys(cpump->page) | PAI_CRYPTO_KERNEL_OFFSET; - WRITE_ONCE(S390_lowcore.ccd, ccd); + WRITE_ONCE(get_lowcore()->ccd, ccd); local_ctl_set_bit(0, CR0_CRYPTOGRAPHY_COUNTER_BIT); } if (flags & PERF_EF_START) @@ -353,6 +381,7 @@ static int paicrypt_add(struct perf_event *event, int flags) return 0; } +static void paicrypt_have_sample(struct perf_event *, struct paicrypt_map *); static void paicrypt_stop(struct perf_event *event, int flags) { struct paicrypt_mapptr *mp = this_cpu_ptr(paicrypt_root.mapptr); @@ -361,8 +390,13 @@ static void paicrypt_stop(struct perf_event *event, int flags) if (!event->attr.sample_period) { /* Counting */ paicrypt_read(event); } else { /* Sampling */ - perf_sched_cb_dec(event->pmu); - cpump->event = NULL; + if (!(event->attach_state & PERF_ATTACH_TASK)) { + perf_sched_cb_dec(event->pmu); + list_del(PAI_SWLIST(event)); + } else { + paicrypt_have_sample(event, cpump); + cpump->event = NULL; + } } event->hw.state = PERF_HES_STOPPED; } @@ -375,7 +409,7 @@ static void paicrypt_del(struct perf_event *event, int flags) paicrypt_stop(event, PERF_EF_UPDATE); if (--cpump->active_events == 0) { local_ctl_clear_bit(0, CR0_CRYPTOGRAPHY_COUNTER_BIT); - WRITE_ONCE(S390_lowcore.ccd, 0); + WRITE_ONCE(get_lowcore()->ccd, 0); } } @@ -455,23 +489,30 @@ static int paicrypt_push_sample(size_t rawsize, struct paicrypt_map *cpump, } /* Check if there is data to be saved on schedule out of a task. */ -static int paicrypt_have_sample(void) +static void paicrypt_have_sample(struct perf_event *event, + struct paicrypt_map *cpump) { - struct paicrypt_mapptr *mp = this_cpu_ptr(paicrypt_root.mapptr); - struct paicrypt_map *cpump = mp->mapptr; - struct perf_event *event = cpump->event; size_t rawsize; - int rc = 0; if (!event) /* No event active */ - return 0; + return; rawsize = paicrypt_copy(cpump->save, cpump->page, (unsigned long *)PAI_SAVE_AREA(event), - cpump->event->attr.exclude_user, - cpump->event->attr.exclude_kernel); + event->attr.exclude_user, + event->attr.exclude_kernel); if (rawsize) /* No incremented counters */ - rc = paicrypt_push_sample(rawsize, cpump, event); - return rc; + paicrypt_push_sample(rawsize, cpump, event); +} + +/* Check if there is data to be saved on schedule out of a task. */ +static void paicrypt_have_samples(void) +{ + struct paicrypt_mapptr *mp = this_cpu_ptr(paicrypt_root.mapptr); + struct paicrypt_map *cpump = mp->mapptr; + struct perf_event *event; + + list_for_each_entry(event, &cpump->syswide_list, hw.tp_list) + paicrypt_have_sample(event, cpump); } /* Called on schedule-in and schedule-out. No access to event structure, @@ -480,10 +521,10 @@ static int paicrypt_have_sample(void) static void paicrypt_sched_task(struct perf_event_pmu_context *pmu_ctx, bool sched_in) { /* We started with a clean page on event installation. So read out - * results on schedule_out and if page was dirty, clear values. + * results on schedule_out and if page was dirty, save old values. */ if (!sched_in) - paicrypt_have_sample(); + paicrypt_have_samples(); } /* Attribute definitions for paicrypt interface. As with other CPU @@ -527,7 +568,7 @@ static const struct attribute_group *paicrypt_attr_groups[] = { /* Performance monitoring unit for mapped counters */ static struct pmu paicrypt = { - .task_ctx_nr = perf_invalid_context, + .task_ctx_nr = perf_hw_context, .event_init = paicrypt_event_init, .add = paicrypt_add, .del = paicrypt_del, diff --git a/arch/s390/kernel/perf_pai_ext.c b/arch/s390/kernel/perf_pai_ext.c index a6da7e0cc7a6..6295531b39a2 100644 --- a/arch/s390/kernel/perf_pai_ext.c +++ b/arch/s390/kernel/perf_pai_ext.c @@ -47,11 +47,11 @@ struct paiext_cb { /* PAI extension 1 control block */ struct paiext_map { unsigned long *area; /* Area for CPU to store counters */ struct pai_userdata *save; /* Area to store non-zero counters */ - enum paievt_mode mode; /* Type of event */ unsigned int active_events; /* # of PAI Extension users */ refcount_t refcnt; struct perf_event *event; /* Perf event for sampling */ struct paiext_cb *paiext_cb; /* PAI extension control block area */ + struct list_head syswide_list; /* List system-wide sampling events */ }; struct paiext_mapptr { @@ -70,6 +70,8 @@ static void paiext_root_free(void) free_percpu(paiext_root.mapptr); paiext_root.mapptr = NULL; } + debug_sprintf_event(paiext_dbg, 5, "%s root.refcount %d\n", __func__, + refcount_read(&paiext_root.refcnt)); } /* On initialization of first event also allocate per CPU data dynamically. @@ -115,20 +117,34 @@ static void paiext_free(struct paiext_mapptr *mp) } /* Release the PMU if event is the last perf event */ -static void paiext_event_destroy(struct perf_event *event) +static void paiext_event_destroy_cpu(struct perf_event *event, int cpu) { - struct paiext_mapptr *mp = per_cpu_ptr(paiext_root.mapptr, event->cpu); + struct paiext_mapptr *mp = per_cpu_ptr(paiext_root.mapptr, cpu); struct paiext_map *cpump = mp->mapptr; - free_page(PAI_SAVE_AREA(event)); mutex_lock(&paiext_reserve_mutex); if (refcount_dec_and_test(&cpump->refcnt)) /* Last reference gone */ paiext_free(mp); paiext_root_free(); mutex_unlock(&paiext_reserve_mutex); - debug_sprintf_event(paiext_dbg, 4, "%s cpu %d mapptr %p\n", __func__, - event->cpu, mp->mapptr); +} + +static void paiext_event_destroy(struct perf_event *event) +{ + int cpu; + + free_page(PAI_SAVE_AREA(event)); + if (event->cpu == -1) { + struct cpumask *mask = PAI_CPU_MASK(event); + for_each_cpu(cpu, mask) + paiext_event_destroy_cpu(event, cpu); + kfree(mask); + } else { + paiext_event_destroy_cpu(event, event->cpu); + } + debug_sprintf_event(paiext_dbg, 4, "%s cpu %d\n", __func__, + event->cpu); } /* Used to avoid races in checking concurrent access of counting and @@ -145,19 +161,18 @@ static void paiext_event_destroy(struct perf_event *event) * * Allocate the memory for the event. */ -static int paiext_alloc(struct perf_event_attr *a, struct perf_event *event) +static int paiext_alloc_cpu(struct perf_event *event, int cpu) { struct paiext_mapptr *mp; struct paiext_map *cpump; int rc; mutex_lock(&paiext_reserve_mutex); - rc = paiext_root_alloc(); if (rc) goto unlock; - mp = per_cpu_ptr(paiext_root.mapptr, event->cpu); + mp = per_cpu_ptr(paiext_root.mapptr, cpu); cpump = mp->mapptr; if (!cpump) { /* Paiext_map allocated? */ rc = -ENOMEM; @@ -185,24 +200,13 @@ static int paiext_alloc(struct perf_event_attr *a, struct perf_event *event) paiext_free(mp); goto undo; } + INIT_LIST_HEAD(&cpump->syswide_list); refcount_set(&cpump->refcnt, 1); - cpump->mode = a->sample_period ? PAI_MODE_SAMPLING - : PAI_MODE_COUNTING; + rc = 0; } else { - /* Multiple invocation, check what is active. - * Supported are multiple counter events or only one sampling - * event concurrently at any one time. - */ - if (cpump->mode == PAI_MODE_SAMPLING || - (cpump->mode == PAI_MODE_COUNTING && a->sample_period)) { - rc = -EBUSY; - goto undo; - } refcount_inc(&cpump->refcnt); } - rc = 0; - undo: if (rc) { /* Error in allocation of event, decrement anchor. Since @@ -217,6 +221,38 @@ unlock: return rc; } +static int paiext_alloc(struct perf_event *event) +{ + struct cpumask *maskptr; + int cpu, rc = -ENOMEM; + + maskptr = kzalloc(sizeof(*maskptr), GFP_KERNEL); + if (!maskptr) + goto out; + + for_each_online_cpu(cpu) { + rc = paiext_alloc_cpu(event, cpu); + if (rc) { + for_each_cpu(cpu, maskptr) + paiext_event_destroy_cpu(event, cpu); + kfree(maskptr); + goto out; + } + cpumask_set_cpu(cpu, maskptr); + } + + /* + * On error all cpumask are freed and all events have been destroyed. + * Save of which CPUs data structures have been allocated for. + * Release them in paicrypt_event_destroy call back function + * for this event. + */ + PAI_CPU_MASK(event) = maskptr; + rc = 0; +out: + return rc; +} + /* The PAI extension 1 control block supports up to 128 entries. Return * the index within PAIE1_CB given the event number. Also validate event * number. @@ -246,9 +282,6 @@ static int paiext_event_init(struct perf_event *event) rc = paiext_event_valid(event); if (rc) return rc; - /* Allow only CPU wide operation, no process context for now. */ - if ((event->attach_state & PERF_ATTACH_TASK) || event->cpu == -1) - return -ENOENT; /* Allow only event NNPA_ALL for sampling. */ if (a->sample_period && a->config != PAI_NNPA_BASE) return -EINVAL; @@ -262,7 +295,10 @@ static int paiext_event_init(struct perf_event *event) return -ENOMEM; } - rc = paiext_alloc(a, event); + if (event->cpu >= 0) + rc = paiext_alloc_cpu(event, event->cpu); + else + rc = paiext_alloc(event); if (rc) { free_page(PAI_SAVE_AREA(event)); return rc; @@ -334,8 +370,15 @@ static void paiext_start(struct perf_event *event, int flags) sum = paiext_getall(event); /* Get current value */ local64_set(&event->hw.prev_count, sum); } else { /* Sampling */ - cpump->event = event; - perf_sched_cb_inc(event->pmu); + memcpy((void *)PAI_SAVE_AREA(event), cpump->area, + PAIE1_CTRBLOCK_SZ); + /* Enable context switch callback for system-wide sampling */ + if (!(event->attach_state & PERF_ATTACH_TASK)) { + list_add_tail(PAI_SWLIST(event), &cpump->syswide_list); + perf_sched_cb_inc(event->pmu); + } else { + cpump->event = event; + } } } @@ -346,12 +389,10 @@ static int paiext_add(struct perf_event *event, int flags) struct paiext_cb *pcb = cpump->paiext_cb; if (++cpump->active_events == 1) { - S390_lowcore.aicd = virt_to_phys(cpump->paiext_cb); + get_lowcore()->aicd = virt_to_phys(cpump->paiext_cb); pcb->acc = virt_to_phys(cpump->area) | 0x1; /* Enable CPU instruction lookup for PAIE1 control block */ local_ctl_set_bit(0, CR0_PAI_EXTENSION_BIT); - debug_sprintf_event(paiext_dbg, 4, "%s 1508 %llx acc %llx\n", - __func__, S390_lowcore.aicd, pcb->acc); } if (flags & PERF_EF_START) paiext_start(event, PERF_EF_RELOAD); @@ -359,6 +400,7 @@ static int paiext_add(struct perf_event *event, int flags) return 0; } +static void paiext_have_sample(struct perf_event *, struct paiext_map *); static void paiext_stop(struct perf_event *event, int flags) { struct paiext_mapptr *mp = this_cpu_ptr(paiext_root.mapptr); @@ -367,8 +409,13 @@ static void paiext_stop(struct perf_event *event, int flags) if (!event->attr.sample_period) { /* Counting */ paiext_read(event); } else { /* Sampling */ - perf_sched_cb_dec(event->pmu); - cpump->event = NULL; + if (!(event->attach_state & PERF_ATTACH_TASK)) { + list_del(PAI_SWLIST(event)); + perf_sched_cb_dec(event->pmu); + } else { + paiext_have_sample(event, cpump); + cpump->event = NULL; + } } event->hw.state = PERF_HES_STOPPED; } @@ -384,9 +431,7 @@ static void paiext_del(struct perf_event *event, int flags) /* Disable CPU instruction lookup for PAIE1 control block */ local_ctl_clear_bit(0, CR0_PAI_EXTENSION_BIT); pcb->acc = 0; - S390_lowcore.aicd = 0; - debug_sprintf_event(paiext_dbg, 4, "%s 1508 %llx acc %llx\n", - __func__, S390_lowcore.aicd, pcb->acc); + get_lowcore()->aicd = 0; } } @@ -470,21 +515,28 @@ static int paiext_push_sample(size_t rawsize, struct paiext_map *cpump, } /* Check if there is data to be saved on schedule out of a task. */ -static int paiext_have_sample(void) +static void paiext_have_sample(struct perf_event *event, + struct paiext_map *cpump) { - struct paiext_mapptr *mp = this_cpu_ptr(paiext_root.mapptr); - struct paiext_map *cpump = mp->mapptr; - struct perf_event *event = cpump->event; size_t rawsize; - int rc = 0; if (!event) - return 0; + return; rawsize = paiext_copy(cpump->save, cpump->area, (unsigned long *)PAI_SAVE_AREA(event)); if (rawsize) /* Incremented counters */ - rc = paiext_push_sample(rawsize, cpump, event); - return rc; + paiext_push_sample(rawsize, cpump, event); +} + +/* Check if there is data to be saved on schedule out of a task. */ +static void paiext_have_samples(void) +{ + struct paiext_mapptr *mp = this_cpu_ptr(paiext_root.mapptr); + struct paiext_map *cpump = mp->mapptr; + struct perf_event *event; + + list_for_each_entry(event, &cpump->syswide_list, hw.tp_list) + paiext_have_sample(event, cpump); } /* Called on schedule-in and schedule-out. No access to event structure, @@ -493,10 +545,10 @@ static int paiext_have_sample(void) static void paiext_sched_task(struct perf_event_pmu_context *pmu_ctx, bool sched_in) { /* We started with a clean page on event installation. So read out - * results on schedule_out and if page was dirty, clear values. + * results on schedule_out and if page was dirty, save old values. */ if (!sched_in) - paiext_have_sample(); + paiext_have_samples(); } /* Attribute definitions for pai extension1 interface. As with other CPU @@ -542,7 +594,7 @@ static const struct attribute_group *paiext_attr_groups[] = { /* Performance monitoring unit for mapped counters */ static struct pmu paiext = { - .task_ctx_nr = perf_invalid_context, + .task_ctx_nr = perf_hw_context, .event_init = paiext_event_init, .add = paiext_add, .del = paiext_del, diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index d8740631df4b..9637aee43c40 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -71,10 +71,10 @@ void flush_thread(void) void arch_setup_new_exec(void) { - if (S390_lowcore.current_pid != current->pid) { - S390_lowcore.current_pid = current->pid; + if (get_lowcore()->current_pid != current->pid) { + get_lowcore()->current_pid = current->pid; if (test_facility(40)) - lpp(&S390_lowcore.lpp); + lpp(&get_lowcore()->lpp); } } diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c index 65c1464eea4f..5ce9a795a0fe 100644 --- a/arch/s390/kernel/processor.c +++ b/arch/s390/kernel/processor.c @@ -17,7 +17,8 @@ #include <linux/mm_types.h> #include <linux/delay.h> #include <linux/cpu.h> - +#include <linux/smp.h> +#include <asm/text-patching.h> #include <asm/diag.h> #include <asm/facility.h> #include <asm/elf.h> @@ -79,6 +80,23 @@ void notrace stop_machine_yield(const struct cpumask *cpumask) } } +static void do_sync_core(void *info) +{ + sync_core(); +} + +void text_poke_sync(void) +{ + on_each_cpu(do_sync_core, NULL, 1); +} + +void text_poke_sync_lock(void) +{ + cpus_read_lock(); + text_poke_sync(); + cpus_read_unlock(); +} + /* * cpu_init - initializes state that is per-CPU. */ diff --git a/arch/s390/kernel/reipl.S b/arch/s390/kernel/reipl.S index 88087a32ebc6..69fcaf54d5ca 100644 --- a/arch/s390/kernel/reipl.S +++ b/arch/s390/kernel/reipl.S @@ -9,6 +9,7 @@ #include <asm/asm-offsets.h> #include <asm/nospec-insn.h> #include <asm/sigp.h> +#include <asm/lowcore.h> GEN_BR_THUNK %r9 @@ -20,20 +21,15 @@ # r3 = Parameter for function # SYM_CODE_START(store_status) - /* Save register one and load save area base */ - stg %r1,__LC_SAVE_AREA_RESTART + STMG_LC %r0,%r15,__LC_GPREGS_SAVE_AREA /* General purpose registers */ - lghi %r1,__LC_GPREGS_SAVE_AREA - stmg %r0,%r15,0(%r1) - mvc 8(8,%r1),__LC_SAVE_AREA_RESTART + GET_LC %r13 /* Control registers */ - lghi %r1,__LC_CREGS_SAVE_AREA - stctg %c0,%c15,0(%r1) + stctg %c0,%c15,__LC_CREGS_SAVE_AREA(%r13) /* Access registers */ - lghi %r1,__LC_AREGS_SAVE_AREA - stam %a0,%a15,0(%r1) + stamy %a0,%a15,__LC_AREGS_SAVE_AREA(%r13) /* Floating point registers */ - lghi %r1,__LC_FPREGS_SAVE_AREA + lay %r1,__LC_FPREGS_SAVE_AREA(%r13) std %f0, 0x00(%r1) std %f1, 0x08(%r1) std %f2, 0x10(%r1) @@ -51,21 +47,21 @@ SYM_CODE_START(store_status) std %f14,0x70(%r1) std %f15,0x78(%r1) /* Floating point control register */ - lghi %r1,__LC_FP_CREG_SAVE_AREA + lay %r1,__LC_FP_CREG_SAVE_AREA(%r13) stfpc 0(%r1) /* CPU timer */ - lghi %r1,__LC_CPU_TIMER_SAVE_AREA + lay %r1,__LC_CPU_TIMER_SAVE_AREA(%r13) stpt 0(%r1) /* Store prefix register */ - lghi %r1,__LC_PREFIX_SAVE_AREA + lay %r1,__LC_PREFIX_SAVE_AREA(%r13) stpx 0(%r1) /* Clock comparator - seven bytes */ - lghi %r1,__LC_CLOCK_COMP_SAVE_AREA larl %r4,clkcmp stckc 0(%r4) + lay %r1,__LC_CLOCK_COMP_SAVE_AREA(%r13) mvc 1(7,%r1),1(%r4) /* Program status word */ - lghi %r1,__LC_PSW_SAVE_AREA + lay %r1,__LC_PSW_SAVE_AREA(%r13) epsw %r4,%r5 st %r4,0(%r1) st %r5,4(%r1) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 90c2c786bb35..a3fea683b227 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -149,13 +149,12 @@ unsigned long __bootdata_preserved(max_mappable); struct physmem_info __bootdata(physmem_info); struct vm_layout __bootdata_preserved(vm_layout); -EXPORT_SYMBOL_GPL(vm_layout); +EXPORT_SYMBOL(vm_layout); int __bootdata_preserved(__kaslr_enabled); unsigned int __bootdata_preserved(zlib_dfltcc_support); EXPORT_SYMBOL(zlib_dfltcc_support); u64 __bootdata_preserved(stfle_fac_list[16]); EXPORT_SYMBOL(stfle_fac_list); -u64 alt_stfle_fac_list[16]; struct oldmem_data __bootdata_preserved(oldmem_data); unsigned long VMALLOC_START; @@ -406,6 +405,7 @@ static void __init setup_lowcore(void) panic("%s: Failed to allocate %zu bytes align=%zx\n", __func__, sizeof(*lc), sizeof(*lc)); + lc->pcpu = (unsigned long)per_cpu_ptr(&pcpu_devices, 0); lc->restart_psw.mask = PSW_KERNEL_BITS & ~PSW_MASK_DAT; lc->restart_psw.addr = __pa(restart_int_handler); lc->external_new_psw.mask = PSW_KERNEL_BITS; @@ -421,16 +421,16 @@ static void __init setup_lowcore(void) lc->clock_comparator = clock_comparator_max; lc->current_task = (unsigned long)&init_task; lc->lpp = LPP_MAGIC; - lc->machine_flags = S390_lowcore.machine_flags; - lc->preempt_count = S390_lowcore.preempt_count; + lc->machine_flags = get_lowcore()->machine_flags; + lc->preempt_count = get_lowcore()->preempt_count; nmi_alloc_mcesa_early(&lc->mcesad); - lc->sys_enter_timer = S390_lowcore.sys_enter_timer; - lc->exit_timer = S390_lowcore.exit_timer; - lc->user_timer = S390_lowcore.user_timer; - lc->system_timer = S390_lowcore.system_timer; - lc->steal_timer = S390_lowcore.steal_timer; - lc->last_update_timer = S390_lowcore.last_update_timer; - lc->last_update_clock = S390_lowcore.last_update_clock; + lc->sys_enter_timer = get_lowcore()->sys_enter_timer; + lc->exit_timer = get_lowcore()->exit_timer; + lc->user_timer = get_lowcore()->user_timer; + lc->system_timer = get_lowcore()->system_timer; + lc->steal_timer = get_lowcore()->steal_timer; + lc->last_update_timer = get_lowcore()->last_update_timer; + lc->last_update_clock = get_lowcore()->last_update_clock; /* * Allocate the global restart stack which is the same for * all CPUs in case *one* of them does a PSW restart. @@ -439,7 +439,7 @@ static void __init setup_lowcore(void) lc->mcck_stack = stack_alloc_early() + STACK_INIT_OFFSET; lc->async_stack = stack_alloc_early() + STACK_INIT_OFFSET; lc->nodat_stack = stack_alloc_early() + STACK_INIT_OFFSET; - lc->kernel_stack = S390_lowcore.kernel_stack; + lc->kernel_stack = get_lowcore()->kernel_stack; /* * Set up PSW restart to call ipl.c:do_restart(). Copy the relevant * restart data to the absolute zero lowcore. This is necessary if @@ -455,8 +455,8 @@ static void __init setup_lowcore(void) lc->return_lpswe = gen_lpswe(__LC_RETURN_PSW); lc->return_mcck_lpswe = gen_lpswe(__LC_RETURN_MCCK_PSW); lc->preempt_count = PREEMPT_DISABLED; - lc->kernel_asce = S390_lowcore.kernel_asce; - lc->user_asce = S390_lowcore.user_asce; + lc->kernel_asce = get_lowcore()->kernel_asce; + lc->user_asce = get_lowcore()->user_asce; system_ctlreg_init_save_area(lc); abs_lc = get_abs_lowcore(); @@ -734,7 +734,23 @@ static void __init memblock_add_physmem_info(void) } /* - * Reserve memory used for lowcore/command line/kernel image. + * Reserve memory used for lowcore. + */ +static void __init reserve_lowcore(void) +{ + void *lowcore_start = get_lowcore(); + void *lowcore_end = lowcore_start + sizeof(struct lowcore); + void *start, *end; + + if ((void *)__identity_base < lowcore_end) { + start = max(lowcore_start, (void *)__identity_base); + end = min(lowcore_end, (void *)(__identity_base + ident_map_size)); + memblock_reserve(__pa(start), __pa(end)); + } +} + +/* + * Reserve memory used for absolute lowcore/command line/kernel image. */ static void __init reserve_kernel(void) { @@ -889,6 +905,9 @@ void __init setup_arch(char **cmdline_p) else pr_info("Linux is running as a guest in 64-bit mode\n"); + if (have_relocated_lowcore()) + pr_info("Lowcore relocated to 0x%px\n", get_lowcore()); + log_component_list(); /* Have one command line that is parsed and saved in /proc/cmdline */ @@ -915,6 +934,7 @@ void __init setup_arch(char **cmdline_p) /* Do some memory reservations *before* memory is added to memblock */ reserve_pgtables(); + reserve_lowcore(); reserve_kernel(); reserve_initrd(); reserve_certificate_list(); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 0324649aae0a..fbba37ec53cf 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -74,18 +74,15 @@ enum { CPU_STATE_CONFIGURED, }; -static DEFINE_PER_CPU(struct cpu *, cpu_device); - -struct pcpu { - unsigned long ec_mask; /* bit mask for ec_xxx functions */ - unsigned long ec_clk; /* sigp timestamp for ec_xxx */ - signed char state; /* physical cpu state */ - signed char polarization; /* physical polarization */ - u16 address; /* physical cpu address */ -}; - static u8 boot_core_type; -static struct pcpu pcpu_devices[NR_CPUS]; +DEFINE_PER_CPU(struct pcpu, pcpu_devices); +/* + * Pointer to the pcpu area of the boot CPU. This is required when a restart + * interrupt is triggered on an offline CPU. For that case accessing percpu + * data with the common primitives does not work, since the percpu offset is + * stored in a non existent lowcore. + */ +static struct pcpu *ipl_pcpu; unsigned int smp_cpu_mt_shift; EXPORT_SYMBOL(smp_cpu_mt_shift); @@ -176,8 +173,8 @@ static struct pcpu *pcpu_find_address(const struct cpumask *mask, u16 address) int cpu; for_each_cpu(cpu, mask) - if (pcpu_devices[cpu].address == address) - return pcpu_devices + cpu; + if (per_cpu(pcpu_devices, cpu).address == address) + return &per_cpu(pcpu_devices, cpu); return NULL; } @@ -203,7 +200,7 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) mcck_stack = stack_alloc(); if (!lc || !nodat_stack || !async_stack || !mcck_stack) goto out; - memcpy(lc, &S390_lowcore, 512); + memcpy(lc, get_lowcore(), 512); memset((char *) lc + 512, 0, sizeof(*lc) - 512); lc->async_stack = async_stack + STACK_INIT_OFFSET; lc->nodat_stack = nodat_stack + STACK_INIT_OFFSET; @@ -232,13 +229,11 @@ out: return -ENOMEM; } -static void pcpu_free_lowcore(struct pcpu *pcpu) +static void pcpu_free_lowcore(struct pcpu *pcpu, int cpu) { unsigned long async_stack, nodat_stack, mcck_stack; struct lowcore *lc; - int cpu; - cpu = pcpu - pcpu_devices; lc = lowcore_ptr[cpu]; nodat_stack = lc->nodat_stack - STACK_INIT_OFFSET; async_stack = lc->async_stack - STACK_INIT_OFFSET; @@ -261,13 +256,14 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) cpumask_set_cpu(cpu, &init_mm.context.cpu_attach_mask); cpumask_set_cpu(cpu, mm_cpumask(&init_mm)); lc->cpu_nr = cpu; + lc->pcpu = (unsigned long)pcpu; lc->restart_flags = RESTART_FLAG_CTLREGS; lc->spinlock_lockval = arch_spin_lockval(cpu); lc->spinlock_index = 0; lc->percpu_offset = __per_cpu_offset[cpu]; - lc->kernel_asce = S390_lowcore.kernel_asce; + lc->kernel_asce = get_lowcore()->kernel_asce; lc->user_asce = s390_invalid_asce; - lc->machine_flags = S390_lowcore.machine_flags; + lc->machine_flags = get_lowcore()->machine_flags; lc->user_timer = lc->system_timer = lc->steal_timer = lc->avg_steal_timer = 0; abs_lc = get_abs_lowcore(); @@ -279,12 +275,10 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) arch_spin_lock_setup(cpu); } -static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk) +static void pcpu_attach_task(int cpu, struct task_struct *tsk) { struct lowcore *lc; - int cpu; - cpu = pcpu - pcpu_devices; lc = lowcore_ptr[cpu]; lc->kernel_stack = (unsigned long)task_stack_page(tsk) + STACK_INIT_OFFSET; lc->current_task = (unsigned long)tsk; @@ -298,18 +292,16 @@ static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk) lc->steal_timer = 0; } -static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data) +static void pcpu_start_fn(int cpu, void (*func)(void *), void *data) { struct lowcore *lc; - int cpu; - cpu = pcpu - pcpu_devices; lc = lowcore_ptr[cpu]; lc->restart_stack = lc->kernel_stack; lc->restart_fn = (unsigned long) func; lc->restart_data = (unsigned long) data; lc->restart_source = -1U; - pcpu_sigp_retry(pcpu, SIGP_RESTART, 0); + pcpu_sigp_retry(per_cpu_ptr(&pcpu_devices, cpu), SIGP_RESTART, 0); } typedef void (pcpu_delegate_fn)(void *); @@ -322,14 +314,14 @@ static void __pcpu_delegate(pcpu_delegate_fn *func, void *data) func(data); /* should not return */ } -static void pcpu_delegate(struct pcpu *pcpu, +static void pcpu_delegate(struct pcpu *pcpu, int cpu, pcpu_delegate_fn *func, void *data, unsigned long stack) { struct lowcore *lc, *abs_lc; unsigned int source_cpu; - lc = lowcore_ptr[pcpu - pcpu_devices]; + lc = lowcore_ptr[cpu]; source_cpu = stap(); if (pcpu->address == source_cpu) { @@ -379,38 +371,22 @@ static int pcpu_set_smt(unsigned int mtid) smp_cpu_mt_shift = 0; while (smp_cpu_mtid >= (1U << smp_cpu_mt_shift)) smp_cpu_mt_shift++; - pcpu_devices[0].address = stap(); + per_cpu(pcpu_devices, 0).address = stap(); } return cc; } /* - * Call function on an online CPU. - */ -void smp_call_online_cpu(void (*func)(void *), void *data) -{ - struct pcpu *pcpu; - - /* Use the current cpu if it is online. */ - pcpu = pcpu_find_address(cpu_online_mask, stap()); - if (!pcpu) - /* Use the first online cpu. */ - pcpu = pcpu_devices + cpumask_first(cpu_online_mask); - pcpu_delegate(pcpu, func, data, (unsigned long) restart_stack); -} - -/* * Call function on the ipl CPU. */ void smp_call_ipl_cpu(void (*func)(void *), void *data) { struct lowcore *lc = lowcore_ptr[0]; - if (pcpu_devices[0].address == stap()) - lc = &S390_lowcore; + if (ipl_pcpu->address == stap()) + lc = get_lowcore(); - pcpu_delegate(&pcpu_devices[0], func, data, - lc->nodat_stack); + pcpu_delegate(ipl_pcpu, 0, func, data, lc->nodat_stack); } int smp_find_processor_id(u16 address) @@ -418,21 +394,21 @@ int smp_find_processor_id(u16 address) int cpu; for_each_present_cpu(cpu) - if (pcpu_devices[cpu].address == address) + if (per_cpu(pcpu_devices, cpu).address == address) return cpu; return -1; } void schedule_mcck_handler(void) { - pcpu_ec_call(pcpu_devices + smp_processor_id(), ec_mcck_pending); + pcpu_ec_call(this_cpu_ptr(&pcpu_devices), ec_mcck_pending); } bool notrace arch_vcpu_is_preempted(int cpu) { if (test_cpu_flag_of(CIF_ENABLED_WAIT, cpu)) return false; - if (pcpu_running(pcpu_devices + cpu)) + if (pcpu_running(per_cpu_ptr(&pcpu_devices, cpu))) return false; return true; } @@ -444,7 +420,7 @@ void notrace smp_yield_cpu(int cpu) return; diag_stat_inc_norecursion(DIAG_STAT_X09C); asm volatile("diag %0,0,0x9c" - : : "d" (pcpu_devices[cpu].address)); + : : "d" (per_cpu(pcpu_devices, cpu).address)); } EXPORT_SYMBOL_GPL(smp_yield_cpu); @@ -465,7 +441,7 @@ void notrace smp_emergency_stop(void) end = get_tod_clock() + (1000000UL << 12); for_each_cpu(cpu, &cpumask) { - struct pcpu *pcpu = pcpu_devices + cpu; + struct pcpu *pcpu = per_cpu_ptr(&pcpu_devices, cpu); set_bit(ec_stop_cpu, &pcpu->ec_mask); while (__pcpu_sigp(pcpu->address, SIGP_EMERGENCY_SIGNAL, 0, NULL) == SIGP_CC_BUSY && @@ -474,7 +450,7 @@ void notrace smp_emergency_stop(void) } while (get_tod_clock() < end) { for_each_cpu(cpu, &cpumask) - if (pcpu_stopped(pcpu_devices + cpu)) + if (pcpu_stopped(per_cpu_ptr(&pcpu_devices, cpu))) cpumask_clear_cpu(cpu, &cpumask); if (cpumask_empty(&cpumask)) break; @@ -489,6 +465,7 @@ NOKPROBE_SYMBOL(smp_emergency_stop); */ void smp_send_stop(void) { + struct pcpu *pcpu; int cpu; /* Disable all interrupts/machine checks */ @@ -504,8 +481,9 @@ void smp_send_stop(void) for_each_online_cpu(cpu) { if (cpu == smp_processor_id()) continue; - pcpu_sigp_retry(pcpu_devices + cpu, SIGP_STOP, 0); - while (!pcpu_stopped(pcpu_devices + cpu)) + pcpu = per_cpu_ptr(&pcpu_devices, cpu); + pcpu_sigp_retry(pcpu, SIGP_STOP, 0); + while (!pcpu_stopped(pcpu)) cpu_relax(); } } @@ -519,7 +497,7 @@ static void smp_handle_ext_call(void) unsigned long bits; /* handle bit signal external calls */ - bits = xchg(&pcpu_devices[smp_processor_id()].ec_mask, 0); + bits = this_cpu_xchg(pcpu_devices.ec_mask, 0); if (test_bit(ec_stop_cpu, &bits)) smp_stop_cpu(); if (test_bit(ec_schedule, &bits)) @@ -544,12 +522,12 @@ void arch_send_call_function_ipi_mask(const struct cpumask *mask) int cpu; for_each_cpu(cpu, mask) - pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single); + pcpu_ec_call(per_cpu_ptr(&pcpu_devices, cpu), ec_call_function_single); } void arch_send_call_function_single_ipi(int cpu) { - pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single); + pcpu_ec_call(per_cpu_ptr(&pcpu_devices, cpu), ec_call_function_single); } /* @@ -559,13 +537,13 @@ void arch_send_call_function_single_ipi(int cpu) */ void arch_smp_send_reschedule(int cpu) { - pcpu_ec_call(pcpu_devices + cpu, ec_schedule); + pcpu_ec_call(per_cpu_ptr(&pcpu_devices, cpu), ec_schedule); } #ifdef CONFIG_IRQ_WORK void arch_irq_work_raise(void) { - pcpu_ec_call(pcpu_devices + smp_processor_id(), ec_irq_work); + pcpu_ec_call(this_cpu_ptr(&pcpu_devices), ec_irq_work); } #endif @@ -577,7 +555,7 @@ int smp_store_status(int cpu) struct pcpu *pcpu; unsigned long pa; - pcpu = pcpu_devices + cpu; + pcpu = per_cpu_ptr(&pcpu_devices, cpu); lc = lowcore_ptr[cpu]; pa = __pa(&lc->floating_pt_save_area); if (__pcpu_sigp_relax(pcpu->address, SIGP_STORE_STATUS_AT_ADDRESS, @@ -685,17 +663,17 @@ void __init smp_save_dump_secondary_cpus(void) void smp_cpu_set_polarization(int cpu, int val) { - pcpu_devices[cpu].polarization = val; + per_cpu(pcpu_devices, cpu).polarization = val; } int smp_cpu_get_polarization(int cpu) { - return pcpu_devices[cpu].polarization; + return per_cpu(pcpu_devices, cpu).polarization; } int smp_cpu_get_cpu_address(int cpu) { - return pcpu_devices[cpu].address; + return per_cpu(pcpu_devices, cpu).address; } static void __ref smp_get_core_info(struct sclp_core_info *info, int early) @@ -719,8 +697,6 @@ static void __ref smp_get_core_info(struct sclp_core_info *info, int early) } } -static int smp_add_present_cpu(int cpu); - static int smp_add_core(struct sclp_core_entry *core, cpumask_t *avail, bool configured, bool early) { @@ -736,7 +712,7 @@ static int smp_add_core(struct sclp_core_entry *core, cpumask_t *avail, for (i = 0; (i <= smp_cpu_mtid) && (cpu < nr_cpu_ids); i++) { if (pcpu_find_address(cpu_present_mask, address + i)) continue; - pcpu = pcpu_devices + cpu; + pcpu = per_cpu_ptr(&pcpu_devices, cpu); pcpu->address = address + i; if (configured) pcpu->state = CPU_STATE_CONFIGURED; @@ -744,7 +720,7 @@ static int smp_add_core(struct sclp_core_entry *core, cpumask_t *avail, pcpu->state = CPU_STATE_STANDBY; smp_cpu_set_polarization(cpu, POLARIZATION_UNKNOWN); set_cpu_present(cpu, true); - if (!early && smp_add_present_cpu(cpu) != 0) + if (!early && arch_register_cpu(cpu)) set_cpu_present(cpu, false); else nr++; @@ -771,7 +747,7 @@ static int __smp_rescan_cpus(struct sclp_core_info *info, bool early) * that all SMT threads get subsequent logical CPU numbers. */ if (early) { - core_id = pcpu_devices[0].address >> smp_cpu_mt_shift; + core_id = per_cpu(pcpu_devices, 0).address >> smp_cpu_mt_shift; for (i = 0; i < info->configured; i++) { core = &info->core[i]; if (core->core_id == core_id) { @@ -831,9 +807,6 @@ void __init smp_detect_cpus(void) s_cpus += smp_cpu_mtid + 1; } pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus); - - /* Add CPUs present at boot */ - __smp_rescan_cpus(info, true); memblock_free(info, sizeof(*info)); } @@ -842,15 +815,16 @@ void __init smp_detect_cpus(void) */ static void smp_start_secondary(void *cpuvoid) { + struct lowcore *lc = get_lowcore(); int cpu = raw_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 = -1U; - S390_lowcore.restart_flags = 0; - restore_access_regs(S390_lowcore.access_regs_save_area); + lc->last_update_clock = get_tod_clock(); + lc->restart_stack = (unsigned long)restart_stack; + lc->restart_fn = (unsigned long)do_restart; + lc->restart_data = 0; + lc->restart_source = -1U; + lc->restart_flags = 0; + restore_access_regs(lc->access_regs_save_area); cpu_init(); rcutree_report_cpu_starting(cpu); init_cpu_timer(); @@ -873,7 +847,7 @@ static void smp_start_secondary(void *cpuvoid) /* Upping and downing of CPUs */ int __cpu_up(unsigned int cpu, struct task_struct *tidle) { - struct pcpu *pcpu = pcpu_devices + cpu; + struct pcpu *pcpu = per_cpu_ptr(&pcpu_devices, cpu); int rc; if (pcpu->state != CPU_STATE_CONFIGURED) @@ -891,8 +865,8 @@ int __cpu_up(unsigned int cpu, struct task_struct *tidle) */ system_ctlreg_lock(); pcpu_prepare_secondary(pcpu, cpu); - pcpu_attach_task(pcpu, tidle); - pcpu_start_fn(pcpu, smp_start_secondary, NULL); + pcpu_attach_task(cpu, tidle); + pcpu_start_fn(cpu, smp_start_secondary, NULL); /* Wait until cpu puts itself in the online & active maps */ while (!cpu_online(cpu)) cpu_relax(); @@ -937,18 +911,19 @@ void __cpu_die(unsigned int cpu) struct pcpu *pcpu; /* Wait until target cpu is down */ - pcpu = pcpu_devices + cpu; + pcpu = per_cpu_ptr(&pcpu_devices, cpu); while (!pcpu_stopped(pcpu)) cpu_relax(); - pcpu_free_lowcore(pcpu); + pcpu_free_lowcore(pcpu, cpu); cpumask_clear_cpu(cpu, mm_cpumask(&init_mm)); cpumask_clear_cpu(cpu, &init_mm.context.cpu_attach_mask); + pcpu->flags = 0; } void __noreturn cpu_die(void) { idle_task_exit(); - pcpu_sigp_retry(pcpu_devices + smp_processor_id(), SIGP_STOP, 0); + pcpu_sigp_retry(this_cpu_ptr(&pcpu_devices), SIGP_STOP, 0); for (;;) ; } @@ -973,24 +948,29 @@ void __init smp_prepare_cpus(unsigned int max_cpus) if (register_external_irq(EXT_IRQ_EXTERNAL_CALL, do_ext_call_interrupt)) panic("Couldn't request external interrupt 0x1202"); system_ctl_set_bit(0, 13); + smp_rescan_cpus(true); } void __init smp_prepare_boot_cpu(void) { - struct pcpu *pcpu = pcpu_devices; + struct lowcore *lc = get_lowcore(); WARN_ON(!cpu_present(0) || !cpu_online(0)); - pcpu->state = CPU_STATE_CONFIGURED; - S390_lowcore.percpu_offset = __per_cpu_offset[0]; + lc->percpu_offset = __per_cpu_offset[0]; + ipl_pcpu = per_cpu_ptr(&pcpu_devices, 0); + ipl_pcpu->state = CPU_STATE_CONFIGURED; + lc->pcpu = (unsigned long)ipl_pcpu; smp_cpu_set_polarization(0, POLARIZATION_UNKNOWN); } void __init smp_setup_processor_id(void) { - pcpu_devices[0].address = stap(); - S390_lowcore.cpu_nr = 0; - S390_lowcore.spinlock_lockval = arch_spin_lockval(0); - S390_lowcore.spinlock_index = 0; + struct lowcore *lc = get_lowcore(); + + lc->cpu_nr = 0; + per_cpu(pcpu_devices, 0).address = stap(); + lc->spinlock_lockval = arch_spin_lockval(0); + lc->spinlock_index = 0; } /* @@ -1010,7 +990,7 @@ static ssize_t cpu_configure_show(struct device *dev, ssize_t count; mutex_lock(&smp_cpu_state_mutex); - count = sprintf(buf, "%d\n", pcpu_devices[dev->id].state); + count = sprintf(buf, "%d\n", per_cpu(pcpu_devices, dev->id).state); mutex_unlock(&smp_cpu_state_mutex); return count; } @@ -1036,7 +1016,7 @@ static ssize_t cpu_configure_store(struct device *dev, for (i = 0; i <= smp_cpu_mtid; i++) if (cpu_online(cpu + i)) goto out; - pcpu = pcpu_devices + cpu; + pcpu = per_cpu_ptr(&pcpu_devices, cpu); rc = 0; switch (val) { case 0: @@ -1048,7 +1028,7 @@ static ssize_t cpu_configure_store(struct device *dev, for (i = 0; i <= smp_cpu_mtid; i++) { if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i)) continue; - pcpu[i].state = CPU_STATE_STANDBY; + per_cpu(pcpu_devices, cpu + i).state = CPU_STATE_STANDBY; smp_cpu_set_polarization(cpu + i, POLARIZATION_UNKNOWN); } @@ -1063,7 +1043,7 @@ static ssize_t cpu_configure_store(struct device *dev, for (i = 0; i <= smp_cpu_mtid; i++) { if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i)) continue; - pcpu[i].state = CPU_STATE_CONFIGURED; + per_cpu(pcpu_devices, cpu + i).state = CPU_STATE_CONFIGURED; smp_cpu_set_polarization(cpu + i, POLARIZATION_UNKNOWN); } @@ -1082,7 +1062,7 @@ static DEVICE_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store); static ssize_t show_cpu_address(struct device *dev, struct device_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", pcpu_devices[dev->id].address); + return sprintf(buf, "%d\n", per_cpu(pcpu_devices, dev->id).address); } static DEVICE_ATTR(address, 0444, show_cpu_address, NULL); @@ -1108,35 +1088,34 @@ static struct attribute_group cpu_online_attr_group = { static int smp_cpu_online(unsigned int cpu) { - struct device *s = &per_cpu(cpu_device, cpu)->dev; + struct cpu *c = per_cpu_ptr(&cpu_devices, cpu); - return sysfs_create_group(&s->kobj, &cpu_online_attr_group); + return sysfs_create_group(&c->dev.kobj, &cpu_online_attr_group); } static int smp_cpu_pre_down(unsigned int cpu) { - struct device *s = &per_cpu(cpu_device, cpu)->dev; + struct cpu *c = per_cpu_ptr(&cpu_devices, cpu); - sysfs_remove_group(&s->kobj, &cpu_online_attr_group); + sysfs_remove_group(&c->dev.kobj, &cpu_online_attr_group); return 0; } -static int smp_add_present_cpu(int cpu) +bool arch_cpu_is_hotpluggable(int cpu) +{ + return !!cpu; +} + +int arch_register_cpu(int cpu) { - struct device *s; - struct cpu *c; + struct cpu *c = per_cpu_ptr(&cpu_devices, cpu); int rc; - c = kzalloc(sizeof(*c), GFP_KERNEL); - if (!c) - return -ENOMEM; - per_cpu(cpu_device, cpu) = c; - s = &c->dev; - c->hotpluggable = !!cpu; + c->hotpluggable = arch_cpu_is_hotpluggable(cpu); rc = register_cpu(c, cpu); if (rc) goto out; - rc = sysfs_create_group(&s->kobj, &cpu_common_attr_group); + rc = sysfs_create_group(&c->dev.kobj, &cpu_common_attr_group); if (rc) goto out_cpu; rc = topology_cpu_init(c); @@ -1145,14 +1124,14 @@ static int smp_add_present_cpu(int cpu) return 0; out_topology: - sysfs_remove_group(&s->kobj, &cpu_common_attr_group); + sysfs_remove_group(&c->dev.kobj, &cpu_common_attr_group); out_cpu: unregister_cpu(c); out: return rc; } -int __ref smp_rescan_cpus(void) +int __ref smp_rescan_cpus(bool early) { struct sclp_core_info *info; int nr; @@ -1161,7 +1140,7 @@ int __ref smp_rescan_cpus(void) if (!info) return -ENOMEM; smp_get_core_info(info, 0); - nr = __smp_rescan_cpus(info, false); + nr = __smp_rescan_cpus(info, early); kfree(info); if (nr) topology_schedule_update(); @@ -1178,7 +1157,7 @@ static ssize_t __ref rescan_store(struct device *dev, rc = lock_device_hotplug_sysfs(); if (rc) return rc; - rc = smp_rescan_cpus(); + rc = smp_rescan_cpus(false); unlock_device_hotplug(); return rc ? rc : count; } @@ -1187,7 +1166,7 @@ static DEVICE_ATTR_WO(rescan); static int __init s390_smp_init(void) { struct device *dev_root; - int cpu, rc = 0; + int rc; dev_root = bus_get_dev_root(&cpu_subsys); if (dev_root) { @@ -1196,17 +1175,9 @@ static int __init s390_smp_init(void) if (rc) return rc; } - - for_each_present_cpu(cpu) { - rc = smp_add_present_cpu(cpu); - if (rc) - goto out; - } - rc = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "s390/smp:online", smp_cpu_online, smp_cpu_pre_down); rc = rc <= 0 ? rc : 0; -out: return rc; } subsys_initcall(s390_smp_init); diff --git a/arch/s390/kernel/sthyi.c b/arch/s390/kernel/sthyi.c index 30bb20461db4..1cf2ad04f8e9 100644 --- a/arch/s390/kernel/sthyi.c +++ b/arch/s390/kernel/sthyi.c @@ -300,33 +300,56 @@ static struct diag204_x_part_block *lpar_cpu_inf(struct lpar_cpu_inf *part_inf, return (struct diag204_x_part_block *)&block->cpus[i]; } -static void fill_diag(struct sthyi_sctns *sctns) +static void *diag204_get_data(bool diag204_allow_busy) { - int i, r, pages; - bool this_lpar; + unsigned long subcode; void *diag204_buf; - void *diag224_buf = NULL; - struct diag204_x_info_blk_hdr *ti_hdr; - struct diag204_x_part_block *part_block; - struct diag204_x_phys_block *phys_block; - struct lpar_cpu_inf lpar_inf = {}; - - /* Errors are handled through the validity bits in the response. */ - pages = diag204((unsigned long)DIAG204_SUBC_RSI | - (unsigned long)DIAG204_INFO_EXT, 0, NULL); - if (pages <= 0) - return; - + int pages, rc; + + subcode = DIAG204_SUBC_RSI; + subcode |= DIAG204_INFO_EXT; + pages = diag204(subcode, 0, NULL); + if (pages < 0) + return ERR_PTR(pages); + if (pages == 0) + return ERR_PTR(-ENODATA); diag204_buf = __vmalloc_node(array_size(pages, PAGE_SIZE), PAGE_SIZE, GFP_KERNEL, NUMA_NO_NODE, __builtin_return_address(0)); if (!diag204_buf) - return; + return ERR_PTR(-ENOMEM); + subcode = DIAG204_SUBC_STIB7; + subcode |= DIAG204_INFO_EXT; + if (diag204_has_bif() && diag204_allow_busy) + subcode |= DIAG204_BIF_BIT; + rc = diag204(subcode, pages, diag204_buf); + if (rc < 0) { + vfree(diag204_buf); + return ERR_PTR(rc); + } + return diag204_buf; +} - r = diag204((unsigned long)DIAG204_SUBC_STIB7 | - (unsigned long)DIAG204_INFO_EXT, pages, diag204_buf); - if (r < 0) - goto out; +static bool is_diag204_cached(struct sthyi_sctns *sctns) +{ + /* + * Check if validity bits are set when diag204 data + * is gathered. + */ + if (sctns->par.infpval1) + return true; + return false; +} + +static void fill_diag(struct sthyi_sctns *sctns, void *diag204_buf) +{ + int i; + bool this_lpar; + void *diag224_buf = NULL; + struct diag204_x_info_blk_hdr *ti_hdr; + struct diag204_x_part_block *part_block; + struct diag204_x_phys_block *phys_block; + struct lpar_cpu_inf lpar_inf = {}; diag224_buf = (void *)__get_free_page(GFP_KERNEL | GFP_DMA); if (!diag224_buf || diag224(diag224_buf)) @@ -392,7 +415,6 @@ static void fill_diag(struct sthyi_sctns *sctns) out: free_page((unsigned long)diag224_buf); - vfree(diag204_buf); } static int sthyi(u64 vaddr, u64 *rc) @@ -414,19 +436,31 @@ static int sthyi(u64 vaddr, u64 *rc) static int fill_dst(void *dst, u64 *rc) { + void *diag204_buf; + struct sthyi_sctns *sctns = (struct sthyi_sctns *)dst; /* * If the facility is on, we don't want to emulate the instruction. * We ask the hypervisor to provide the data. */ - if (test_facility(74)) + if (test_facility(74)) { + memset(dst, 0, PAGE_SIZE); return sthyi((u64)dst, rc); - + } + /* + * When emulating, if diag204 returns BUSY don't reset dst buffer + * and use cached data. + */ + *rc = 0; + diag204_buf = diag204_get_data(is_diag204_cached(sctns)); + if (IS_ERR(diag204_buf)) + return PTR_ERR(diag204_buf); + memset(dst, 0, PAGE_SIZE); fill_hdr(sctns); fill_stsi(sctns); - fill_diag(sctns); - *rc = 0; + fill_diag(sctns, diag204_buf); + vfree(diag204_buf); return 0; } @@ -445,11 +479,14 @@ static int sthyi_update_cache(u64 *rc) { int r; - memset(sthyi_cache.info, 0, PAGE_SIZE); r = fill_dst(sthyi_cache.info, rc); - if (r) - return r; - sthyi_cache.end = jiffies + CACHE_VALID_JIFFIES; + if (r == 0) { + sthyi_cache.end = jiffies + CACHE_VALID_JIFFIES; + } else if (r == -EBUSY) { + /* mark as expired and return 0 to keep using cached data */ + sthyi_cache.end = jiffies - 1; + r = 0; + } return r; } diff --git a/arch/s390/kernel/syscall.c b/arch/s390/kernel/syscall.c index 50cbcbbaa03d..5ec28028315b 100644 --- a/arch/s390/kernel/syscall.c +++ b/arch/s390/kernel/syscall.c @@ -124,8 +124,8 @@ void noinstr __do_syscall(struct pt_regs *regs, int per_trap) { add_random_kstack_offset(); enter_from_user_mode(regs); - regs->psw = S390_lowcore.svc_old_psw; - regs->int_code = S390_lowcore.svc_int_code; + regs->psw = get_lowcore()->svc_old_psw; + regs->int_code = get_lowcore()->svc_int_code; update_timer_sys(); if (static_branch_likely(&cpu_has_bear)) current->thread.last_break = regs->last_break; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index fb9f31f36628..b713effe0579 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -131,7 +131,7 @@ void clock_comparator_work(void) { struct clock_event_device *cd; - S390_lowcore.clock_comparator = clock_comparator_max; + get_lowcore()->clock_comparator = clock_comparator_max; cd = this_cpu_ptr(&comparators); cd->event_handler(cd); } @@ -139,8 +139,8 @@ void clock_comparator_work(void) static int s390_next_event(unsigned long delta, struct clock_event_device *evt) { - S390_lowcore.clock_comparator = get_tod_clock() + delta; - set_clock_comparator(S390_lowcore.clock_comparator); + get_lowcore()->clock_comparator = get_tod_clock() + delta; + set_clock_comparator(get_lowcore()->clock_comparator); return 0; } @@ -153,8 +153,8 @@ void init_cpu_timer(void) struct clock_event_device *cd; int cpu; - S390_lowcore.clock_comparator = clock_comparator_max; - set_clock_comparator(S390_lowcore.clock_comparator); + get_lowcore()->clock_comparator = clock_comparator_max; + set_clock_comparator(get_lowcore()->clock_comparator); cpu = smp_processor_id(); cd = &per_cpu(comparators, cpu); @@ -184,8 +184,8 @@ static void clock_comparator_interrupt(struct ext_code ext_code, unsigned long param64) { inc_irq_stat(IRQEXT_CLK); - if (S390_lowcore.clock_comparator == clock_comparator_max) - set_clock_comparator(S390_lowcore.clock_comparator); + if (get_lowcore()->clock_comparator == clock_comparator_max) + set_clock_comparator(get_lowcore()->clock_comparator); } static void stp_timing_alert(struct stp_irq_parm *); @@ -408,12 +408,12 @@ static void clock_sync_global(long delta) static void clock_sync_local(long delta) { /* Add the delta to the clock comparator. */ - if (S390_lowcore.clock_comparator != clock_comparator_max) { - S390_lowcore.clock_comparator += delta; - set_clock_comparator(S390_lowcore.clock_comparator); + if (get_lowcore()->clock_comparator != clock_comparator_max) { + get_lowcore()->clock_comparator += delta; + set_clock_comparator(get_lowcore()->clock_comparator); } /* Adjust the last_update_clock time-stamp. */ - S390_lowcore.last_update_clock += delta; + get_lowcore()->last_update_clock += delta; } /* Single threaded workqueue used for stp sync events */ diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 89e91b8ce842..22029ecae1c5 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -320,16 +320,10 @@ static int __arch_update_cpu_topology(void) int arch_update_cpu_topology(void) { - struct device *dev; - int cpu, rc; + int rc; rc = __arch_update_cpu_topology(); on_each_cpu(__arch_update_dedicated_flag, NULL, 0); - for_each_online_cpu(cpu) { - dev = get_cpu_device(cpu); - if (dev) - kobject_uevent(&dev->kobj, KOBJ_CHANGE); - } return rc; } @@ -600,7 +594,7 @@ static int __init topology_setup(char *str) } early_param("topology", topology_setup); -static int topology_ctl_handler(struct ctl_table *ctl, int write, +static int topology_ctl_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { int enabled = topology_is_enabled(); diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index 52578b5cecbd..160b2acba8db 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -27,6 +27,7 @@ #include <linux/uaccess.h> #include <linux/cpu.h> #include <linux/entry-common.h> +#include <linux/kmsan.h> #include <asm/asm-extable.h> #include <asm/vtime.h> #include <asm/fpu.h> @@ -262,6 +263,11 @@ static void monitor_event_exception(struct pt_regs *regs) void kernel_stack_overflow(struct pt_regs *regs) { + /* + * Normally regs are unpoisoned by the generic entry code, but + * kernel_stack_overflow() is a rare case that is called bypassing it. + */ + kmsan_unpoison_entry_regs(regs); bust_spinlocks(1); printk("Kernel stack overflow.\n"); show_regs(regs); @@ -288,15 +294,16 @@ static void __init test_monitor_call(void) void __init trap_init(void) { + struct lowcore *lc = get_lowcore(); unsigned long flags; struct ctlreg cr0; local_irq_save(flags); cr0 = local_ctl_clear_bit(0, CR0_LOW_ADDRESS_PROTECTION_BIT); - psw_bits(S390_lowcore.external_new_psw).mcheck = 1; - psw_bits(S390_lowcore.program_new_psw).mcheck = 1; - psw_bits(S390_lowcore.svc_new_psw).mcheck = 1; - psw_bits(S390_lowcore.io_new_psw).mcheck = 1; + psw_bits(lc->external_new_psw).mcheck = 1; + psw_bits(lc->program_new_psw).mcheck = 1; + psw_bits(lc->svc_new_psw).mcheck = 1; + psw_bits(lc->io_new_psw).mcheck = 1; local_ctl_load(0, &cr0); local_irq_restore(flags); local_mcck_enable(); @@ -307,11 +314,12 @@ static void (*pgm_check_table[128])(struct pt_regs *regs); void noinstr __do_pgm_check(struct pt_regs *regs) { - unsigned int trapnr; + struct lowcore *lc = get_lowcore(); irqentry_state_t state; + unsigned int trapnr; - regs->int_code = S390_lowcore.pgm_int_code; - regs->int_parm_long = S390_lowcore.trans_exc_code; + regs->int_code = lc->pgm_int_code; + regs->int_parm_long = lc->trans_exc_code; state = irqentry_enter(regs); @@ -324,19 +332,19 @@ void noinstr __do_pgm_check(struct pt_regs *regs) current->thread.last_break = regs->last_break; } - if (S390_lowcore.pgm_code & 0x0200) { + if (lc->pgm_code & 0x0200) { /* transaction abort */ - current->thread.trap_tdb = S390_lowcore.pgm_tdb; + current->thread.trap_tdb = lc->pgm_tdb; } - if (S390_lowcore.pgm_code & PGM_INT_CODE_PER) { + if (lc->pgm_code & PGM_INT_CODE_PER) { if (user_mode(regs)) { struct per_event *ev = ¤t->thread.per_event; set_thread_flag(TIF_PER_TRAP); - ev->address = S390_lowcore.per_address; - ev->cause = S390_lowcore.per_code_combined; - ev->paid = S390_lowcore.per_access_id; + ev->address = lc->per_address; + ev->cause = lc->per_code_combined; + ev->paid = lc->per_access_id; } else { /* PER event in kernel is kprobes */ __arch_local_irq_ssm(regs->psw.mask & ~PSW_MASK_PER); diff --git a/arch/s390/kernel/unwind_bc.c b/arch/s390/kernel/unwind_bc.c index 0ece156fdd7c..cd44be2b6ce8 100644 --- a/arch/s390/kernel/unwind_bc.c +++ b/arch/s390/kernel/unwind_bc.c @@ -49,6 +49,8 @@ static inline bool is_final_pt_regs(struct unwind_state *state, READ_ONCE_NOCHECK(regs->psw.mask) & PSW_MASK_PSTATE; } +/* Avoid KMSAN false positives from touching uninitialized frames. */ +__no_kmsan_checks bool unwind_next_frame(struct unwind_state *state) { struct stack_info *info = &state->stack_info; @@ -118,6 +120,8 @@ out_stop: } EXPORT_SYMBOL_GPL(unwind_next_frame); +/* Avoid KMSAN false positives from touching uninitialized frames. */ +__no_kmsan_checks void __unwind_start(struct unwind_state *state, struct task_struct *task, struct pt_regs *regs, unsigned long first_frame) { diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index 265fea37e030..36db065c7cf7 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -18,11 +18,22 @@ #include <asm/sections.h> #include <asm/uv.h> +#if !IS_ENABLED(CONFIG_KVM) +unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr) +{ + return 0; +} + +int gmap_fault(struct gmap *gmap, unsigned long gaddr, + unsigned int fault_flags) +{ + return 0; +} +#endif + /* the bootdata_preserved fields come from ones in arch/s390/boot/uv.c */ -#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST int __bootdata_preserved(prot_virt_guest); EXPORT_SYMBOL(prot_virt_guest); -#endif /* * uv_info contains both host and guest information but it's currently only @@ -35,7 +46,6 @@ EXPORT_SYMBOL(prot_virt_guest); struct uv_info __bootdata_preserved(uv_info); EXPORT_SYMBOL(uv_info); -#if IS_ENABLED(CONFIG_KVM) int __bootdata_preserved(prot_virt_host); EXPORT_SYMBOL(prot_virt_host); @@ -110,7 +120,7 @@ EXPORT_SYMBOL_GPL(uv_pin_shared); * * @paddr: Absolute host address of page to be destroyed */ -static int uv_destroy_page(unsigned long paddr) +static int uv_destroy(unsigned long paddr) { struct uv_cb_cfs uvcb = { .header.cmd = UVC_CMD_DESTR_SEC_STOR, @@ -131,28 +141,40 @@ static int uv_destroy_page(unsigned long paddr) } /* - * The caller must already hold a reference to the page + * The caller must already hold a reference to the folio */ -int uv_destroy_owned_page(unsigned long paddr) +int uv_destroy_folio(struct folio *folio) { - struct page *page = phys_to_page(paddr); int rc; - get_page(page); - rc = uv_destroy_page(paddr); + /* See gmap_make_secure(): large folios cannot be secure */ + if (unlikely(folio_test_large(folio))) + return 0; + + folio_get(folio); + rc = uv_destroy(folio_to_phys(folio)); if (!rc) - clear_bit(PG_arch_1, &page->flags); - put_page(page); + clear_bit(PG_arch_1, &folio->flags); + folio_put(folio); return rc; } /* + * The present PTE still indirectly holds a folio reference through the mapping. + */ +int uv_destroy_pte(pte_t pte) +{ + VM_WARN_ON(!pte_present(pte)); + return uv_destroy_folio(pfn_folio(pte_pfn(pte))); +} + +/* * Requests the Ultravisor to encrypt a guest page and make it * accessible to the host for paging (export). * * @paddr: Absolute host address of page to be exported */ -int uv_convert_from_secure(unsigned long paddr) +static int uv_convert_from_secure(unsigned long paddr) { struct uv_cb_cfs uvcb = { .header.cmd = UVC_CMD_CONV_FROM_SEC_STOR, @@ -166,22 +188,34 @@ int uv_convert_from_secure(unsigned long paddr) } /* - * The caller must already hold a reference to the page + * The caller must already hold a reference to the folio. */ -int uv_convert_owned_from_secure(unsigned long paddr) +static int uv_convert_from_secure_folio(struct folio *folio) { - struct page *page = phys_to_page(paddr); int rc; - get_page(page); - rc = uv_convert_from_secure(paddr); + /* See gmap_make_secure(): large folios cannot be secure */ + if (unlikely(folio_test_large(folio))) + return 0; + + folio_get(folio); + rc = uv_convert_from_secure(folio_to_phys(folio)); if (!rc) - clear_bit(PG_arch_1, &page->flags); - put_page(page); + clear_bit(PG_arch_1, &folio->flags); + folio_put(folio); return rc; } /* + * The present PTE still indirectly holds a folio reference through the mapping. + */ +int uv_convert_from_secure_pte(pte_t pte) +{ + VM_WARN_ON(!pte_present(pte)); + return uv_convert_from_secure_folio(pfn_folio(pte_pfn(pte))); +} + +/* * Calculate the expected ref_count for a folio that would otherwise have no * further pins. This was cribbed from similar functions in other places in * the kernel, but with some slight modifications. We know that a secure @@ -267,6 +301,36 @@ static bool should_export_before_import(struct uv_cb_header *uvcb, struct mm_str } /* + * Drain LRU caches: the local one on first invocation and the ones of all + * CPUs on successive invocations. Returns "true" on the first invocation. + */ +static bool drain_lru(bool *drain_lru_called) +{ + /* + * If we have tried a local drain and the folio refcount + * still does not match our expected safe value, try with a + * system wide drain. This is needed if the pagevecs holding + * the page are on a different CPU. + */ + if (*drain_lru_called) { + lru_add_drain_all(); + /* We give up here, don't retry immediately. */ + return false; + } + /* + * We are here if the folio refcount does not match the + * expected safe value. The main culprits are usually + * pagevecs. With lru_add_drain() we drain the pagevecs + * on the local CPU so that hopefully the refcount will + * reach the expected safe value. + */ + lru_add_drain(); + *drain_lru_called = true; + /* The caller should try again immediately */ + return true; +} + +/* * Requests the Ultravisor to make a page accessible to a guest. * If it's brought in the first time, it will be cleared. If * it has been exported before, it will be decrypted and integrity @@ -275,7 +339,7 @@ static bool should_export_before_import(struct uv_cb_header *uvcb, struct mm_str int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb) { struct vm_area_struct *vma; - bool local_drain = false; + bool drain_lru_called = false; spinlock_t *ptelock; unsigned long uaddr; struct folio *folio; @@ -308,52 +372,63 @@ again: goto out; if (pte_present(*ptep) && !(pte_val(*ptep) & _PAGE_INVALID) && pte_write(*ptep)) { folio = page_folio(pte_page(*ptep)); - rc = -EINVAL; - if (folio_test_large(folio)) - goto unlock; rc = -EAGAIN; - if (folio_trylock(folio)) { + if (folio_test_large(folio)) { + rc = -E2BIG; + } else if (folio_trylock(folio)) { if (should_export_before_import(uvcb, gmap->mm)) uv_convert_from_secure(PFN_PHYS(folio_pfn(folio))); rc = make_folio_secure(folio, uvcb); folio_unlock(folio); } + + /* + * Once we drop the PTL, the folio may get unmapped and + * freed immediately. We need a temporary reference. + */ + if (rc == -EAGAIN || rc == -E2BIG) + folio_get(folio); } -unlock: pte_unmap_unlock(ptep, ptelock); out: mmap_read_unlock(gmap->mm); - if (rc == -EAGAIN) { + switch (rc) { + case -E2BIG: + folio_lock(folio); + rc = split_folio(folio); + folio_unlock(folio); + folio_put(folio); + + switch (rc) { + case 0: + /* Splitting succeeded, try again immediately. */ + goto again; + case -EAGAIN: + /* Additional folio references. */ + if (drain_lru(&drain_lru_called)) + goto again; + return -EAGAIN; + case -EBUSY: + /* Unexpected race. */ + return -EAGAIN; + } + WARN_ON_ONCE(1); + return -ENXIO; + case -EAGAIN: /* * If we are here because the UVC returned busy or partial * completion, this is just a useless check, but it is safe. */ folio_wait_writeback(folio); - } else if (rc == -EBUSY) { - /* - * If we have tried a local drain and the folio refcount - * still does not match our expected safe value, try with a - * system wide drain. This is needed if the pagevecs holding - * the page are on a different CPU. - */ - if (local_drain) { - lru_add_drain_all(); - /* We give up here, and let the caller try again */ - return -EAGAIN; - } - /* - * We are here if the folio refcount does not match the - * expected safe value. The main culprits are usually - * pagevecs. With lru_add_drain() we drain the pagevecs - * on the local CPU so that hopefully the refcount will - * reach the expected safe value. - */ - lru_add_drain(); - local_drain = true; - /* And now we try again immediately after draining */ - goto again; - } else if (rc == -ENXIO) { + folio_put(folio); + return -EAGAIN; + case -EBUSY: + /* Additional folio references. */ + if (drain_lru(&drain_lru_called)) + goto again; + return -EAGAIN; + case -ENXIO: if (gmap_fault(gmap, gaddr, FAULT_FLAG_WRITE)) return -EFAULT; return -EAGAIN; @@ -388,6 +463,7 @@ int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr) { struct vm_area_struct *vma; unsigned long uaddr; + struct folio *folio; struct page *page; int rc; @@ -411,7 +487,8 @@ int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr) page = follow_page(vma, uaddr, FOLL_WRITE | FOLL_GET); if (IS_ERR_OR_NULL(page)) goto out; - rc = uv_destroy_owned_page(page_to_phys(page)); + folio = page_folio(page); + rc = uv_destroy_folio(folio); /* * Fault handlers can race; it is possible that two CPUs will fault * on the same secure page. One CPU can destroy the page, reboot, @@ -422,8 +499,8 @@ int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr) * we instead try to export the page. */ if (rc) - rc = uv_convert_owned_from_secure(page_to_phys(page)); - put_page(page); + rc = uv_convert_from_secure_folio(folio); + folio_put(folio); out: mmap_read_unlock(gmap->mm); return rc; @@ -431,50 +508,51 @@ out: EXPORT_SYMBOL_GPL(gmap_destroy_page); /* - * To be called with the page locked or with an extra reference! This will - * prevent gmap_make_secure from touching the page concurrently. Having 2 - * parallel make_page_accessible is fine, as the UV calls will become a - * no-op if the page is already exported. + * To be called with the folio locked or with an extra reference! This will + * prevent gmap_make_secure from touching the folio concurrently. Having 2 + * parallel arch_make_folio_accessible is fine, as the UV calls will become a + * no-op if the folio is already exported. */ -int arch_make_page_accessible(struct page *page) +int arch_make_folio_accessible(struct folio *folio) { int rc = 0; - /* Hugepage cannot be protected, so nothing to do */ - if (PageHuge(page)) + /* See gmap_make_secure(): large folios cannot be secure */ + if (unlikely(folio_test_large(folio))) return 0; /* - * PG_arch_1 is used in 3 places: - * 1. for kernel page tables during early boot - * 2. for storage keys of huge pages and KVM - * 3. As an indication that this page might be secure. This can + * PG_arch_1 is used in 2 places: + * 1. for storage keys of hugetlb folios and KVM + * 2. As an indication that this small folio might be secure. This can * overindicate, e.g. we set the bit before calling * convert_to_secure. - * As secure pages are never huge, all 3 variants can co-exists. + * As secure pages are never large folios, both variants can co-exists. */ - if (!test_bit(PG_arch_1, &page->flags)) + if (!test_bit(PG_arch_1, &folio->flags)) return 0; - rc = uv_pin_shared(page_to_phys(page)); + rc = uv_pin_shared(folio_to_phys(folio)); if (!rc) { - clear_bit(PG_arch_1, &page->flags); + clear_bit(PG_arch_1, &folio->flags); return 0; } - rc = uv_convert_from_secure(page_to_phys(page)); + rc = uv_convert_from_secure(folio_to_phys(folio)); if (!rc) { - clear_bit(PG_arch_1, &page->flags); + clear_bit(PG_arch_1, &folio->flags); return 0; } return rc; } -EXPORT_SYMBOL_GPL(arch_make_page_accessible); +EXPORT_SYMBOL_GPL(arch_make_folio_accessible); -#endif - -#if defined(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) || IS_ENABLED(CONFIG_KVM) +int arch_make_page_accessible(struct page *page) +{ + return arch_make_folio_accessible(page_folio(page)); +} +EXPORT_SYMBOL_GPL(arch_make_page_accessible); static ssize_t uv_query_facilities(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { @@ -650,24 +728,13 @@ static struct attribute_group uv_query_attr_group = { static ssize_t uv_is_prot_virt_guest(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - int val = 0; - -#ifdef CONFIG_PROTECTED_VIRTUALIZATION_GUEST - val = prot_virt_guest; -#endif - return sysfs_emit(buf, "%d\n", val); + return sysfs_emit(buf, "%d\n", prot_virt_guest); } static ssize_t uv_is_prot_virt_host(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - int val = 0; - -#if IS_ENABLED(CONFIG_KVM) - val = prot_virt_host; -#endif - - return sysfs_emit(buf, "%d\n", val); + return sysfs_emit(buf, "%d\n", prot_virt_host); } static struct kobj_attribute uv_prot_virt_guest = @@ -719,4 +786,3 @@ out_kobj: return rc; } device_initcall(uv_info_init); -#endif diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index a1ce3925ec71..ae5d0a9d6911 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -39,7 +39,7 @@ PHDRS { SECTIONS { - . = __START_KERNEL; + . = TEXT_OFFSET; .text : { _stext = .; /* Start of text section */ _text = .; /* Text and read-only data */ @@ -59,14 +59,6 @@ SECTIONS } :text = 0x0700 RO_DATA(PAGE_SIZE) - .data.rel.ro : { - *(.data.rel.ro .data.rel.ro.*) - } - .got : { - __got_start = .; - *(.got) - __got_end = .; - } . = ALIGN(PAGE_SIZE); _sdata = .; /* Start of data section */ @@ -80,6 +72,15 @@ SECTIONS . = ALIGN(PAGE_SIZE); __end_ro_after_init = .; + .data.rel.ro : { + *(.data.rel.ro .data.rel.ro.*) + } + .got : { + __got_start = .; + *(.got) + __got_end = .; + } + RW_DATA(0x100, PAGE_SIZE, THREAD_SIZE) .data.rel : { *(.data.rel*) @@ -190,6 +191,9 @@ SECTIONS . = ALIGN(PAGE_SIZE); INIT_DATA_SECTION(0x100) + RUNTIME_CONST(shift, d_hash_shift) + RUNTIME_CONST(ptr, dentry_hashtable) + PERCPU_SECTION(0x100) . = ALIGN(PAGE_SIZE); @@ -219,6 +223,8 @@ SECTIONS QUAD(init_mm) QUAD(swapper_pg_dir) QUAD(invalid_pg_dir) + QUAD(__alt_instructions) + QUAD(__alt_instructions_end) #ifdef CONFIG_KASAN QUAD(kasan_early_shadow_page) QUAD(kasan_early_shadow_pte) diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index ffc1db0cbf9c..234a0ba30510 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -35,14 +35,15 @@ static DEFINE_PER_CPU(u64, mt_scaling_jiffies); static inline void set_vtimer(u64 expires) { + struct lowcore *lc = get_lowcore(); u64 timer; asm volatile( " stpt %0\n" /* Store current cpu timer value */ " spt %1" /* Set new value imm. afterwards */ : "=Q" (timer) : "Q" (expires)); - S390_lowcore.system_timer += S390_lowcore.last_update_timer - timer; - S390_lowcore.last_update_timer = expires; + lc->system_timer += lc->last_update_timer - timer; + lc->last_update_timer = expires; } static inline int virt_timer_forward(u64 elapsed) @@ -117,22 +118,23 @@ static void account_system_index_scaled(struct task_struct *p, u64 cputime, static int do_account_vtime(struct task_struct *tsk) { u64 timer, clock, user, guest, system, hardirq, softirq; + struct lowcore *lc = get_lowcore(); - timer = S390_lowcore.last_update_timer; - clock = S390_lowcore.last_update_clock; + timer = lc->last_update_timer; + clock = lc->last_update_clock; asm volatile( " stpt %0\n" /* Store current cpu timer value */ " stckf %1" /* Store current tod clock value */ - : "=Q" (S390_lowcore.last_update_timer), - "=Q" (S390_lowcore.last_update_clock) + : "=Q" (lc->last_update_timer), + "=Q" (lc->last_update_clock) : : "cc"); - clock = S390_lowcore.last_update_clock - clock; - timer -= S390_lowcore.last_update_timer; + clock = lc->last_update_clock - clock; + timer -= lc->last_update_timer; if (hardirq_count()) - S390_lowcore.hardirq_timer += timer; + lc->hardirq_timer += timer; else - S390_lowcore.system_timer += timer; + lc->system_timer += timer; /* Update MT utilization calculation */ if (smp_cpu_mtid && @@ -141,16 +143,16 @@ static int do_account_vtime(struct task_struct *tsk) /* Calculate cputime delta */ user = update_tsk_timer(&tsk->thread.user_timer, - READ_ONCE(S390_lowcore.user_timer)); + READ_ONCE(lc->user_timer)); guest = update_tsk_timer(&tsk->thread.guest_timer, - READ_ONCE(S390_lowcore.guest_timer)); + READ_ONCE(lc->guest_timer)); system = update_tsk_timer(&tsk->thread.system_timer, - READ_ONCE(S390_lowcore.system_timer)); + READ_ONCE(lc->system_timer)); hardirq = update_tsk_timer(&tsk->thread.hardirq_timer, - READ_ONCE(S390_lowcore.hardirq_timer)); + READ_ONCE(lc->hardirq_timer)); softirq = update_tsk_timer(&tsk->thread.softirq_timer, - READ_ONCE(S390_lowcore.softirq_timer)); - S390_lowcore.steal_timer += + READ_ONCE(lc->softirq_timer)); + lc->steal_timer += clock - user - guest - system - hardirq - softirq; /* Push account value */ @@ -176,17 +178,19 @@ static int do_account_vtime(struct task_struct *tsk) void vtime_task_switch(struct task_struct *prev) { + struct lowcore *lc = get_lowcore(); + do_account_vtime(prev); - prev->thread.user_timer = S390_lowcore.user_timer; - prev->thread.guest_timer = S390_lowcore.guest_timer; - prev->thread.system_timer = S390_lowcore.system_timer; - prev->thread.hardirq_timer = S390_lowcore.hardirq_timer; - prev->thread.softirq_timer = S390_lowcore.softirq_timer; - S390_lowcore.user_timer = current->thread.user_timer; - S390_lowcore.guest_timer = current->thread.guest_timer; - S390_lowcore.system_timer = current->thread.system_timer; - S390_lowcore.hardirq_timer = current->thread.hardirq_timer; - S390_lowcore.softirq_timer = current->thread.softirq_timer; + prev->thread.user_timer = lc->user_timer; + prev->thread.guest_timer = lc->guest_timer; + prev->thread.system_timer = lc->system_timer; + prev->thread.hardirq_timer = lc->hardirq_timer; + prev->thread.softirq_timer = lc->softirq_timer; + lc->user_timer = current->thread.user_timer; + lc->guest_timer = current->thread.guest_timer; + lc->system_timer = current->thread.system_timer; + lc->hardirq_timer = current->thread.hardirq_timer; + lc->softirq_timer = current->thread.softirq_timer; } /* @@ -196,28 +200,29 @@ void vtime_task_switch(struct task_struct *prev) */ void vtime_flush(struct task_struct *tsk) { + struct lowcore *lc = get_lowcore(); u64 steal, avg_steal; if (do_account_vtime(tsk)) virt_timer_expire(); - steal = S390_lowcore.steal_timer; - avg_steal = S390_lowcore.avg_steal_timer; + steal = lc->steal_timer; + avg_steal = lc->avg_steal_timer; if ((s64) steal > 0) { - S390_lowcore.steal_timer = 0; + lc->steal_timer = 0; account_steal_time(cputime_to_nsecs(steal)); avg_steal += steal; } - S390_lowcore.avg_steal_timer = avg_steal / 2; + lc->avg_steal_timer = avg_steal / 2; } static u64 vtime_delta(void) { - u64 timer = S390_lowcore.last_update_timer; - - S390_lowcore.last_update_timer = get_cpu_timer(); + struct lowcore *lc = get_lowcore(); + u64 timer = lc->last_update_timer; - return timer - S390_lowcore.last_update_timer; + lc->last_update_timer = get_cpu_timer(); + return timer - lc->last_update_timer; } /* @@ -226,12 +231,13 @@ static u64 vtime_delta(void) */ void vtime_account_kernel(struct task_struct *tsk) { + struct lowcore *lc = get_lowcore(); u64 delta = vtime_delta(); if (tsk->flags & PF_VCPU) - S390_lowcore.guest_timer += delta; + lc->guest_timer += delta; else - S390_lowcore.system_timer += delta; + lc->system_timer += delta; virt_timer_forward(delta); } @@ -241,7 +247,7 @@ void vtime_account_softirq(struct task_struct *tsk) { u64 delta = vtime_delta(); - S390_lowcore.softirq_timer += delta; + get_lowcore()->softirq_timer += delta; virt_timer_forward(delta); } @@ -250,7 +256,7 @@ void vtime_account_hardirq(struct task_struct *tsk) { u64 delta = vtime_delta(); - S390_lowcore.hardirq_timer += delta; + get_lowcore()->hardirq_timer += delta; virt_timer_forward(delta); } diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 5bf3d94e9dda..e65f597e3044 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -14,167 +14,10 @@ #include <asm/access-regs.h> #include <asm/fault.h> #include <asm/gmap.h> +#include <asm/dat-bits.h> #include "kvm-s390.h" #include "gaccess.h" -union asce { - unsigned long val; - struct { - unsigned long origin : 52; /* Region- or Segment-Table Origin */ - unsigned long : 2; - unsigned long g : 1; /* Subspace Group Control */ - unsigned long p : 1; /* Private Space Control */ - unsigned long s : 1; /* Storage-Alteration-Event Control */ - unsigned long x : 1; /* Space-Switch-Event Control */ - unsigned long r : 1; /* Real-Space Control */ - unsigned long : 1; - unsigned long dt : 2; /* Designation-Type Control */ - unsigned long tl : 2; /* Region- or Segment-Table Length */ - }; -}; - -enum { - ASCE_TYPE_SEGMENT = 0, - ASCE_TYPE_REGION3 = 1, - ASCE_TYPE_REGION2 = 2, - ASCE_TYPE_REGION1 = 3 -}; - -union region1_table_entry { - unsigned long val; - struct { - unsigned long rto: 52;/* Region-Table Origin */ - unsigned long : 2; - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long : 1; - unsigned long tf : 2; /* Region-Second-Table Offset */ - unsigned long i : 1; /* Region-Invalid Bit */ - unsigned long : 1; - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long tl : 2; /* Region-Second-Table Length */ - }; -}; - -union region2_table_entry { - unsigned long val; - struct { - unsigned long rto: 52;/* Region-Table Origin */ - unsigned long : 2; - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long : 1; - unsigned long tf : 2; /* Region-Third-Table Offset */ - unsigned long i : 1; /* Region-Invalid Bit */ - unsigned long : 1; - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long tl : 2; /* Region-Third-Table Length */ - }; -}; - -struct region3_table_entry_fc0 { - unsigned long sto: 52;/* Segment-Table Origin */ - unsigned long : 1; - unsigned long fc : 1; /* Format-Control */ - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long : 1; - unsigned long tf : 2; /* Segment-Table Offset */ - unsigned long i : 1; /* Region-Invalid Bit */ - unsigned long cr : 1; /* Common-Region Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long tl : 2; /* Segment-Table Length */ -}; - -struct region3_table_entry_fc1 { - unsigned long rfaa : 33; /* Region-Frame Absolute Address */ - unsigned long : 14; - unsigned long av : 1; /* ACCF-Validity Control */ - unsigned long acc: 4; /* Access-Control Bits */ - unsigned long f : 1; /* Fetch-Protection Bit */ - unsigned long fc : 1; /* Format-Control */ - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long iep: 1; /* Instruction-Execution-Protection */ - unsigned long : 2; - unsigned long i : 1; /* Region-Invalid Bit */ - unsigned long cr : 1; /* Common-Region Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long : 2; -}; - -union region3_table_entry { - unsigned long val; - struct region3_table_entry_fc0 fc0; - struct region3_table_entry_fc1 fc1; - struct { - unsigned long : 53; - unsigned long fc : 1; /* Format-Control */ - unsigned long : 4; - unsigned long i : 1; /* Region-Invalid Bit */ - unsigned long cr : 1; /* Common-Region Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long : 2; - }; -}; - -struct segment_entry_fc0 { - unsigned long pto: 53;/* Page-Table Origin */ - unsigned long fc : 1; /* Format-Control */ - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long : 3; - unsigned long i : 1; /* Segment-Invalid Bit */ - unsigned long cs : 1; /* Common-Segment Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long : 2; -}; - -struct segment_entry_fc1 { - unsigned long sfaa : 44; /* Segment-Frame Absolute Address */ - unsigned long : 3; - unsigned long av : 1; /* ACCF-Validity Control */ - unsigned long acc: 4; /* Access-Control Bits */ - unsigned long f : 1; /* Fetch-Protection Bit */ - unsigned long fc : 1; /* Format-Control */ - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long iep: 1; /* Instruction-Execution-Protection */ - unsigned long : 2; - unsigned long i : 1; /* Segment-Invalid Bit */ - unsigned long cs : 1; /* Common-Segment Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long : 2; -}; - -union segment_table_entry { - unsigned long val; - struct segment_entry_fc0 fc0; - struct segment_entry_fc1 fc1; - struct { - unsigned long : 53; - unsigned long fc : 1; /* Format-Control */ - unsigned long : 4; - unsigned long i : 1; /* Segment-Invalid Bit */ - unsigned long cs : 1; /* Common-Segment Bit */ - unsigned long tt : 2; /* Table-Type Bits */ - unsigned long : 2; - }; -}; - -enum { - TABLE_TYPE_SEGMENT = 0, - TABLE_TYPE_REGION3 = 1, - TABLE_TYPE_REGION2 = 2, - TABLE_TYPE_REGION1 = 3 -}; - -union page_table_entry { - unsigned long val; - struct { - unsigned long pfra : 52; /* Page-Frame Real Address */ - unsigned long z : 1; /* Zero Bit */ - unsigned long i : 1; /* Page-Invalid Bit */ - unsigned long p : 1; /* DAT-Protection Bit */ - unsigned long iep: 1; /* Instruction-Execution-Protection */ - unsigned long : 8; - }; -}; - /* * vaddress union in order to easily decode a virtual address into its * region first index, region second index etc. parts. @@ -632,7 +475,7 @@ static unsigned long guest_translate(struct kvm_vcpu *vcpu, unsigned long gva, iep = ctlreg0.iep && test_kvm_facility(vcpu->kvm, 130); if (asce.r) goto real_address; - ptr = asce.origin * PAGE_SIZE; + ptr = asce.rsto * PAGE_SIZE; switch (asce.dt) { case ASCE_TYPE_REGION1: if (vaddr.rfx01 > asce.tl) @@ -1379,7 +1222,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr, parent = sg->parent; vaddr.addr = saddr; asce.val = sg->orig_asce; - ptr = asce.origin * PAGE_SIZE; + ptr = asce.rsto * PAGE_SIZE; if (asce.r) { *fake = 1; ptr = 0; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 54b5b2565df8..0fd96860fc45 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -2997,14 +2997,9 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) break; } case KVM_CREATE_IRQCHIP: { - struct kvm_irq_routing_entry routing; - r = -EINVAL; - if (kvm->arch.use_irqchip) { - /* Set up dummy routing. */ - memset(&routing, 0, sizeof(routing)); - r = kvm_set_irq_routing(kvm, &routing, 0, 0); - } + if (kvm->arch.use_irqchip) + r = 0; break; } case KVM_SET_DEVICE_ATTR: { @@ -4080,7 +4075,7 @@ static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start, bool kvm_arch_no_poll(struct kvm_vcpu *vcpu) { /* do not poll with more than halt_poll_max_steal percent of steal time */ - if (S390_lowcore.avg_steal_timer * 100 / (TICK_USEC << 12) >= + if (get_lowcore()->avg_steal_timer * 100 / (TICK_USEC << 12) >= READ_ONCE(halt_poll_max_steal)) { vcpu->stat.halt_no_poll_steal++; return true; @@ -4830,7 +4825,8 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) sizeof(sie_page->pv_grregs)); } exit_reason = sie64a(vcpu->arch.sie_block, - vcpu->run->s.regs.gprs); + vcpu->run->s.regs.gprs, + gmap_get_enabled()->asce); if (kvm_s390_pv_cpu_is_protected(vcpu)) { memcpy(vcpu->run->s.regs.gprs, sie_page->pv_grregs, @@ -5032,7 +5028,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) if (vcpu->kvm->arch.pv.dumping) return -EINVAL; - if (kvm_run->immediate_exit) + if (!vcpu->wants_to_run) return -EINTR; if (kvm_run->kvm_valid_regs & ~KVM_SYNC_S390_VALID_FIELDS || @@ -5749,6 +5745,9 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm, { gpa_t size; + if (kvm_is_ucontrol(kvm)) + return -EINVAL; + /* When we are protected, we should not change the memory slots */ if (kvm_s390_pv_get_handle(kvm)) return -EINVAL; diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index bf8534218af3..e680c6bf0c9d 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -267,7 +267,12 @@ static inline unsigned long kvm_s390_get_gfn_end(struct kvm_memslots *slots) static inline u32 kvm_s390_get_gisa_desc(struct kvm *kvm) { - u32 gd = virt_to_phys(kvm->arch.gisa_int.origin); + u32 gd; + + if (!kvm->arch.gisa_int.origin) + return 0; + + gd = virt_to_phys(kvm->arch.gisa_int.origin); if (gd && sclp.has_gisaf) gd |= GISA_FORMAT1; diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index c9ecae830634..89cafea4c41f 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -1150,7 +1150,7 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) vcpu->arch.sie_block->prog0c |= PROG_IN_SIE; barrier(); if (!kvm_s390_vcpu_sie_inhibited(vcpu)) - rc = sie64a(scb_s, vcpu->run->s.regs.gprs); + rc = sie64a(scb_s, vcpu->run->s.regs.gprs, gmap_get_enabled()->asce); barrier(); vcpu->arch.sie_block->prog0c &= ~PROG_IN_SIE; @@ -1304,10 +1304,24 @@ static int vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) if (rc == -EAGAIN) rc = 0; - if (rc || scb_s->icptcode || signal_pending(current) || + + /* + * Exit the loop if the guest needs to process the intercept + */ + if (rc || scb_s->icptcode) + break; + + /* + * Exit the loop if the host needs to process an intercept, + * but rewind the PSW to re-enter SIE once that's completed + * instead of passing a "no action" intercept to the guest. + */ + if (signal_pending(current) || kvm_s390_vcpu_has_irq(vcpu, 0) || - kvm_s390_vcpu_sie_inhibited(vcpu)) + kvm_s390_vcpu_sie_inhibited(vcpu)) { + kvm_s390_rewind_psw(vcpu, 4); break; + } cond_resched(); } @@ -1426,8 +1440,10 @@ int kvm_s390_handle_vsie(struct kvm_vcpu *vcpu) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); if (signal_pending(current) || kvm_s390_vcpu_has_irq(vcpu, 0) || - kvm_s390_vcpu_sie_inhibited(vcpu)) + kvm_s390_vcpu_sie_inhibited(vcpu)) { + kvm_s390_rewind_psw(vcpu, 4); return 0; + } vsie_page = get_vsie_page(vcpu->kvm, scb_addr); if (IS_ERR(vsie_page)) diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c index 81c53440b3e6..9f86ad8fa8b4 100644 --- a/arch/s390/lib/spinlock.c +++ b/arch/s390/lib/spinlock.c @@ -75,7 +75,7 @@ static inline int arch_load_niai4(int *lock) int owner; asm_inline volatile( - ALTERNATIVE("nop", ".insn rre,0xb2fa0000,4,0", 49) /* NIAI 4 */ + ALTERNATIVE("nop", ".insn rre,0xb2fa0000,4,0", ALT_FACILITY(49)) /* NIAI 4 */ " l %0,%1\n" : "=d" (owner) : "Q" (*lock) : "memory"); return owner; @@ -86,7 +86,7 @@ static inline int arch_cmpxchg_niai8(int *lock, int old, int new) int expected = old; asm_inline volatile( - ALTERNATIVE("nop", ".insn rre,0xb2fa0000,8,0", 49) /* NIAI 8 */ + ALTERNATIVE("nop", ".insn rre,0xb2fa0000,8,0", ALT_FACILITY(49)) /* NIAI 8 */ " cs %0,%3,%1\n" : "=d" (old), "=Q" (*lock) : "0" (old), "d" (new), "Q" (*lock) @@ -119,7 +119,7 @@ static inline void arch_spin_lock_queued(arch_spinlock_t *lp) struct spin_wait *node, *next; int lockval, ix, node_id, tail_id, old, new, owner, count; - ix = S390_lowcore.spinlock_index++; + ix = get_lowcore()->spinlock_index++; barrier(); lockval = SPINLOCK_LOCKVAL; /* cpu + 1 */ node = this_cpu_ptr(&spin_wait[ix]); @@ -205,7 +205,7 @@ static inline void arch_spin_lock_queued(arch_spinlock_t *lp) } out: - S390_lowcore.spinlock_index--; + get_lowcore()->spinlock_index--; } static inline void arch_spin_lock_classic(arch_spinlock_t *lp) diff --git a/arch/s390/lib/test_kprobes.c b/arch/s390/lib/test_kprobes.c index 9e62d62812e5..9021298c3e8a 100644 --- a/arch/s390/lib/test_kprobes.c +++ b/arch/s390/lib/test_kprobes.c @@ -72,4 +72,5 @@ static struct kunit_suite kprobes_test_suite = { kunit_test_suites(&kprobes_test_suite); +MODULE_DESCRIPTION("KUnit tests for kprobes"); MODULE_LICENSE("GPL"); diff --git a/arch/s390/lib/test_modules.c b/arch/s390/lib/test_modules.c index 9894009fc1f2..f96b6a3737e7 100644 --- a/arch/s390/lib/test_modules.c +++ b/arch/s390/lib/test_modules.c @@ -29,4 +29,5 @@ static struct kunit_suite modules_test_suite = { kunit_test_suites(&modules_test_suite); +MODULE_DESCRIPTION("KUnit test that modules with many relocations are loaded properly"); MODULE_LICENSE("GPL"); diff --git a/arch/s390/lib/test_unwind.c b/arch/s390/lib/test_unwind.c index 2848e3fb2ff5..8b7f981e6f34 100644 --- a/arch/s390/lib/test_unwind.c +++ b/arch/s390/lib/test_unwind.c @@ -356,7 +356,7 @@ static noinline int unwindme_func2(struct unwindme *u) if (u->flags & UWM_SWITCH_STACK) { local_irq_save(flags); local_mcck_save(mflags); - rc = call_on_stack(1, S390_lowcore.nodat_stack, + rc = call_on_stack(1, get_lowcore()->nodat_stack, int, unwindme_func3, struct unwindme *, u); local_mcck_restore(mflags); local_irq_restore(flags); @@ -519,4 +519,5 @@ static struct kunit_suite test_unwind_suite = { kunit_test_suites(&test_unwind_suite); +MODULE_DESCRIPTION("KUnit test for unwind_for_each_frame"); MODULE_LICENSE("GPL"); diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c index 61d8dcd95bbc..c7c269d5c491 100644 --- a/arch/s390/lib/uaccess.c +++ b/arch/s390/lib/uaccess.c @@ -21,13 +21,13 @@ void debug_user_asce(int exit) local_ctl_store(1, &cr1); local_ctl_store(7, &cr7); - if (cr1.val == S390_lowcore.kernel_asce.val && cr7.val == S390_lowcore.user_asce.val) + if (cr1.val == get_lowcore()->kernel_asce.val && cr7.val == get_lowcore()->user_asce.val) return; panic("incorrect ASCE on kernel %s\n" "cr1: %016lx cr7: %016lx\n" "kernel: %016lx user: %016lx\n", exit ? "exit" : "entry", cr1.val, cr7.val, - S390_lowcore.kernel_asce.val, S390_lowcore.user_asce.val); + get_lowcore()->kernel_asce.val, get_lowcore()->user_asce.val); } #endif /*CONFIG_DEBUG_ENTRY */ diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c index f8b13f247646..75d15bf41d97 100644 --- a/arch/s390/mm/cmm.c +++ b/arch/s390/mm/cmm.c @@ -243,7 +243,7 @@ static int cmm_skip_blanks(char *cp, char **endp) return str != cp; } -static int cmm_pages_handler(struct ctl_table *ctl, int write, +static int cmm_pages_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { long nr = cmm_get_pages(); @@ -262,7 +262,7 @@ static int cmm_pages_handler(struct ctl_table *ctl, int write, return 0; } -static int cmm_timed_pages_handler(struct ctl_table *ctl, int write, +static int cmm_timed_pages_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { @@ -282,7 +282,7 @@ static int cmm_timed_pages_handler(struct ctl_table *ctl, int write, return 0; } -static int cmm_timeout_handler(struct ctl_table *ctl, int write, +static int cmm_timeout_handler(const struct ctl_table *ctl, int write, void *buffer, size_t *lenp, loff_t *ppos) { char buf[64], *p; @@ -427,4 +427,5 @@ static void __exit cmm_exit(void) } module_exit(cmm_exit); +MODULE_DESCRIPTION("Cooperative memory management interface"); MODULE_LICENSE("GPL"); diff --git a/arch/s390/mm/dump_pagetables.c b/arch/s390/mm/dump_pagetables.c index ffd07ed7b4af..0a67fcee4414 100644 --- a/arch/s390/mm/dump_pagetables.c +++ b/arch/s390/mm/dump_pagetables.c @@ -3,6 +3,7 @@ #include <linux/ptdump.h> #include <linux/seq_file.h> #include <linux/debugfs.h> +#include <linux/sort.h> #include <linux/mm.h> #include <linux/kfence.h> #include <linux/kasan.h> @@ -15,13 +16,15 @@ static unsigned long max_addr; struct addr_marker { + int is_start; unsigned long start_address; const char *name; }; enum address_markers_idx { - IDENTITY_BEFORE_NR = 0, - IDENTITY_BEFORE_END_NR, + KVA_NR = 0, + LOWCORE_START_NR, + LOWCORE_END_NR, AMODE31_START_NR, AMODE31_END_NR, KERNEL_START_NR, @@ -30,12 +33,22 @@ enum address_markers_idx { KFENCE_START_NR, KFENCE_END_NR, #endif - IDENTITY_AFTER_NR, - IDENTITY_AFTER_END_NR, + IDENTITY_START_NR, + IDENTITY_END_NR, VMEMMAP_NR, VMEMMAP_END_NR, VMALLOC_NR, VMALLOC_END_NR, +#ifdef CONFIG_KMSAN + KMSAN_VMALLOC_SHADOW_START_NR, + KMSAN_VMALLOC_SHADOW_END_NR, + KMSAN_VMALLOC_ORIGIN_START_NR, + KMSAN_VMALLOC_ORIGIN_END_NR, + KMSAN_MODULES_SHADOW_START_NR, + KMSAN_MODULES_SHADOW_END_NR, + KMSAN_MODULES_ORIGIN_START_NR, + KMSAN_MODULES_ORIGIN_END_NR, +#endif MODULES_NR, MODULES_END_NR, ABS_LOWCORE_NR, @@ -49,33 +62,44 @@ enum address_markers_idx { }; static struct addr_marker address_markers[] = { - [IDENTITY_BEFORE_NR] = {0, "Identity Mapping Start"}, - [IDENTITY_BEFORE_END_NR] = {(unsigned long)_stext, "Identity Mapping End"}, - [AMODE31_START_NR] = {0, "Amode31 Area Start"}, - [AMODE31_END_NR] = {0, "Amode31 Area End"}, - [KERNEL_START_NR] = {(unsigned long)_stext, "Kernel Image Start"}, - [KERNEL_END_NR] = {(unsigned long)_end, "Kernel Image End"}, + [KVA_NR] = {0, 0, "Kernel Virtual Address Space"}, + [LOWCORE_START_NR] = {1, 0, "Lowcore Start"}, + [LOWCORE_END_NR] = {0, 0, "Lowcore End"}, + [IDENTITY_START_NR] = {1, 0, "Identity Mapping Start"}, + [IDENTITY_END_NR] = {0, 0, "Identity Mapping End"}, + [AMODE31_START_NR] = {1, 0, "Amode31 Area Start"}, + [AMODE31_END_NR] = {0, 0, "Amode31 Area End"}, + [KERNEL_START_NR] = {1, (unsigned long)_stext, "Kernel Image Start"}, + [KERNEL_END_NR] = {0, (unsigned long)_end, "Kernel Image End"}, #ifdef CONFIG_KFENCE - [KFENCE_START_NR] = {0, "KFence Pool Start"}, - [KFENCE_END_NR] = {0, "KFence Pool End"}, + [KFENCE_START_NR] = {1, 0, "KFence Pool Start"}, + [KFENCE_END_NR] = {0, 0, "KFence Pool End"}, +#endif + [VMEMMAP_NR] = {1, 0, "vmemmap Area Start"}, + [VMEMMAP_END_NR] = {0, 0, "vmemmap Area End"}, + [VMALLOC_NR] = {1, 0, "vmalloc Area Start"}, + [VMALLOC_END_NR] = {0, 0, "vmalloc Area End"}, +#ifdef CONFIG_KMSAN + [KMSAN_VMALLOC_SHADOW_START_NR] = {1, 0, "Kmsan vmalloc Shadow Start"}, + [KMSAN_VMALLOC_SHADOW_END_NR] = {0, 0, "Kmsan vmalloc Shadow End"}, + [KMSAN_VMALLOC_ORIGIN_START_NR] = {1, 0, "Kmsan vmalloc Origins Start"}, + [KMSAN_VMALLOC_ORIGIN_END_NR] = {0, 0, "Kmsan vmalloc Origins End"}, + [KMSAN_MODULES_SHADOW_START_NR] = {1, 0, "Kmsan Modules Shadow Start"}, + [KMSAN_MODULES_SHADOW_END_NR] = {0, 0, "Kmsan Modules Shadow End"}, + [KMSAN_MODULES_ORIGIN_START_NR] = {1, 0, "Kmsan Modules Origins Start"}, + [KMSAN_MODULES_ORIGIN_END_NR] = {0, 0, "Kmsan Modules Origins End"}, #endif - [IDENTITY_AFTER_NR] = {(unsigned long)_end, "Identity Mapping Start"}, - [IDENTITY_AFTER_END_NR] = {0, "Identity Mapping End"}, - [VMEMMAP_NR] = {0, "vmemmap Area Start"}, - [VMEMMAP_END_NR] = {0, "vmemmap Area End"}, - [VMALLOC_NR] = {0, "vmalloc Area Start"}, - [VMALLOC_END_NR] = {0, "vmalloc Area End"}, - [MODULES_NR] = {0, "Modules Area Start"}, - [MODULES_END_NR] = {0, "Modules Area End"}, - [ABS_LOWCORE_NR] = {0, "Lowcore Area Start"}, - [ABS_LOWCORE_END_NR] = {0, "Lowcore Area End"}, - [MEMCPY_REAL_NR] = {0, "Real Memory Copy Area Start"}, - [MEMCPY_REAL_END_NR] = {0, "Real Memory Copy Area End"}, + [MODULES_NR] = {1, 0, "Modules Area Start"}, + [MODULES_END_NR] = {0, 0, "Modules Area End"}, + [ABS_LOWCORE_NR] = {1, 0, "Lowcore Area Start"}, + [ABS_LOWCORE_END_NR] = {0, 0, "Lowcore Area End"}, + [MEMCPY_REAL_NR] = {1, 0, "Real Memory Copy Area Start"}, + [MEMCPY_REAL_END_NR] = {0, 0, "Real Memory Copy Area End"}, #ifdef CONFIG_KASAN - [KASAN_SHADOW_START_NR] = {KASAN_SHADOW_START, "Kasan Shadow Start"}, - [KASAN_SHADOW_END_NR] = {KASAN_SHADOW_END, "Kasan Shadow End"}, + [KASAN_SHADOW_START_NR] = {1, KASAN_SHADOW_START, "Kasan Shadow Start"}, + [KASAN_SHADOW_END_NR] = {0, KASAN_SHADOW_END, "Kasan Shadow End"}, #endif - { -1, NULL } + {1, -1UL, NULL} }; struct pg_state { @@ -143,6 +167,19 @@ static void note_prot_wx(struct pg_state *st, unsigned long addr) st->wx_pages += (addr - st->start_address) / PAGE_SIZE; } +static void note_page_update_state(struct pg_state *st, unsigned long addr, unsigned int prot, int level) +{ + struct seq_file *m = st->seq; + + while (addr >= st->marker[1].start_address) { + st->marker++; + pt_dump_seq_printf(m, "---[ %s ]---\n", st->marker->name); + } + st->start_address = addr; + st->current_prot = prot; + st->level = level; +} + static void note_page(struct ptdump_state *pt_st, unsigned long addr, int level, u64 val) { int width = sizeof(unsigned long) * 2; @@ -166,9 +203,7 @@ static void note_page(struct ptdump_state *pt_st, unsigned long addr, int level, addr = max_addr; if (st->level == -1) { pt_dump_seq_printf(m, "---[ %s ]---\n", st->marker->name); - st->start_address = addr; - st->current_prot = prot; - st->level = level; + note_page_update_state(st, addr, prot, level); } else if (prot != st->current_prot || level != st->level || addr >= st->marker[1].start_address) { note_prot_wx(st, addr); @@ -182,13 +217,7 @@ static void note_page(struct ptdump_state *pt_st, unsigned long addr, int level, } pt_dump_seq_printf(m, "%9lu%c ", delta, *unit); print_prot(m, st->current_prot, st->level); - while (addr >= st->marker[1].start_address) { - st->marker++; - pt_dump_seq_printf(m, "---[ %s ]---\n", st->marker->name); - } - st->start_address = addr; - st->current_prot = prot; - st->level = level; + note_page_update_state(st, addr, prot, level); } } @@ -260,22 +289,25 @@ static int ptdump_show(struct seq_file *m, void *v) DEFINE_SHOW_ATTRIBUTE(ptdump); #endif /* CONFIG_PTDUMP_DEBUGFS */ -/* - * Heapsort from lib/sort.c is not a stable sorting algorithm, do a simple - * insertion sort to preserve the original order of markers with the same - * start address. - */ -static void sort_address_markers(void) +static int ptdump_cmp(const void *a, const void *b) { - struct addr_marker tmp; - int i, j; + const struct addr_marker *ama = a; + const struct addr_marker *amb = b; - for (i = 1; i < ARRAY_SIZE(address_markers) - 1; i++) { - tmp = address_markers[i]; - for (j = i - 1; j >= 0 && address_markers[j].start_address > tmp.start_address; j--) - address_markers[j + 1] = address_markers[j]; - address_markers[j + 1] = tmp; - } + if (ama->start_address > amb->start_address) + return 1; + if (ama->start_address < amb->start_address) + return -1; + /* + * If the start addresses of two markers are identical consider the + * marker which defines the start of an area higher than the one which + * defines the end of an area. This keeps pairs of markers sorted. + */ + if (ama->is_start) + return 1; + if (amb->is_start) + return -1; + return 0; } static int pt_dump_init(void) @@ -283,14 +315,19 @@ static int pt_dump_init(void) #ifdef CONFIG_KFENCE unsigned long kfence_start = (unsigned long)__kfence_pool; #endif + unsigned long lowcore = (unsigned long)get_lowcore(); + /* * Figure out the maximum virtual address being accessible with the * kernel ASCE. We need this to keep the page table walker functions * from accessing non-existent entries. */ - max_addr = (S390_lowcore.kernel_asce.val & _REGION_ENTRY_TYPE_MASK) >> 2; + max_addr = (get_lowcore()->kernel_asce.val & _REGION_ENTRY_TYPE_MASK) >> 2; max_addr = 1UL << (max_addr * 11 + 31); - address_markers[IDENTITY_AFTER_END_NR].start_address = ident_map_size; + address_markers[LOWCORE_START_NR].start_address = lowcore; + address_markers[LOWCORE_END_NR].start_address = lowcore + sizeof(struct lowcore); + address_markers[IDENTITY_START_NR].start_address = __identity_base; + address_markers[IDENTITY_END_NR].start_address = __identity_base + ident_map_size; address_markers[AMODE31_START_NR].start_address = (unsigned long)__samode31; address_markers[AMODE31_END_NR].start_address = (unsigned long)__eamode31; address_markers[MODULES_NR].start_address = MODULES_VADDR; @@ -307,7 +344,18 @@ static int pt_dump_init(void) address_markers[KFENCE_START_NR].start_address = kfence_start; address_markers[KFENCE_END_NR].start_address = kfence_start + KFENCE_POOL_SIZE; #endif - sort_address_markers(); +#ifdef CONFIG_KMSAN + address_markers[KMSAN_VMALLOC_SHADOW_START_NR].start_address = KMSAN_VMALLOC_SHADOW_START; + address_markers[KMSAN_VMALLOC_SHADOW_END_NR].start_address = KMSAN_VMALLOC_SHADOW_END; + address_markers[KMSAN_VMALLOC_ORIGIN_START_NR].start_address = KMSAN_VMALLOC_ORIGIN_START; + address_markers[KMSAN_VMALLOC_ORIGIN_END_NR].start_address = KMSAN_VMALLOC_ORIGIN_END; + address_markers[KMSAN_MODULES_SHADOW_START_NR].start_address = KMSAN_MODULES_SHADOW_START; + address_markers[KMSAN_MODULES_SHADOW_END_NR].start_address = KMSAN_MODULES_SHADOW_END; + address_markers[KMSAN_MODULES_ORIGIN_START_NR].start_address = KMSAN_MODULES_ORIGIN_START; + address_markers[KMSAN_MODULES_ORIGIN_END_NR].start_address = KMSAN_MODULES_ORIGIN_END; +#endif + sort(address_markers, ARRAY_SIZE(address_markers) - 1, + sizeof(address_markers[0]), ptdump_cmp, NULL); #ifdef CONFIG_PTDUMP_DEBUGFS debugfs_create_file("kernel_page_tables", 0400, NULL, NULL, &ptdump_fops); #endif /* CONFIG_PTDUMP_DEBUGFS */ diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 65747f15dbec..8e149ef5e89b 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -74,7 +74,7 @@ static enum fault_type get_fault_type(struct pt_regs *regs) return USER_FAULT; if (!IS_ENABLED(CONFIG_PGSTE)) return KERNEL_FAULT; - gmap = (struct gmap *)S390_lowcore.gmap; + gmap = (struct gmap *)get_lowcore()->gmap; if (gmap && gmap->asce == regs->cr1) return GMAP_FAULT; return KERNEL_FAULT; @@ -182,15 +182,15 @@ static void dump_fault_info(struct pt_regs *regs) pr_cont("mode while using "); switch (get_fault_type(regs)) { case USER_FAULT: - asce = S390_lowcore.user_asce.val; + asce = get_lowcore()->user_asce.val; pr_cont("user "); break; case GMAP_FAULT: - asce = ((struct gmap *)S390_lowcore.gmap)->asce; + asce = ((struct gmap *)get_lowcore()->gmap)->asce; pr_cont("gmap "); break; case KERNEL_FAULT: - asce = S390_lowcore.kernel_asce.val; + asce = get_lowcore()->kernel_asce.val; pr_cont("kernel "); break; default: @@ -351,7 +351,7 @@ lock_mmap: mmap_read_lock(mm); gmap = NULL; if (IS_ENABLED(CONFIG_PGSTE) && type == GMAP_FAULT) { - gmap = (struct gmap *)S390_lowcore.gmap; + gmap = (struct gmap *)get_lowcore()->gmap; current->thread.gmap_addr = address; current->thread.gmap_write_flag = !!(flags & FAULT_FLAG_WRITE); current->thread.gmap_int_code = regs->int_code & 0xffff; @@ -433,12 +433,13 @@ error: handle_fault_error_nolock(regs, 0); else do_sigsegv(regs, SEGV_MAPERR); - } else if (fault & VM_FAULT_SIGBUS) { + } else if (fault & (VM_FAULT_SIGBUS | VM_FAULT_HWPOISON)) { if (!user_mode(regs)) handle_fault_error_nolock(regs, 0); else do_sigbus(regs); } else { + pr_emerg("Unexpected fault flags: %08x\n", fault); BUG(); } } @@ -492,6 +493,7 @@ void do_secure_storage_access(struct pt_regs *regs) unsigned long addr = get_fault_address(regs); struct vm_area_struct *vma; struct mm_struct *mm; + struct folio *folio; struct page *page; struct gmap *gmap; int rc; @@ -521,7 +523,7 @@ void do_secure_storage_access(struct pt_regs *regs) switch (get_fault_type(regs)) { case GMAP_FAULT: mm = current->mm; - gmap = (struct gmap *)S390_lowcore.gmap; + gmap = (struct gmap *)get_lowcore()->gmap; mmap_read_lock(mm); addr = __gmap_translate(gmap, addr); mmap_read_unlock(mm); @@ -539,17 +541,18 @@ void do_secure_storage_access(struct pt_regs *regs) mmap_read_unlock(mm); break; } - if (arch_make_page_accessible(page)) + folio = page_folio(page); + if (arch_make_folio_accessible(folio)) send_sig(SIGSEGV, current, 0); - put_page(page); + folio_put(folio); mmap_read_unlock(mm); break; case KERNEL_FAULT: - page = phys_to_page(addr); - if (unlikely(!try_get_page(page))) + folio = phys_to_folio(addr); + if (unlikely(!folio_try_get(folio))) break; - rc = arch_make_page_accessible(page); - put_page(page); + rc = arch_make_folio_accessible(folio); + folio_put(folio); if (rc) BUG(); break; @@ -561,7 +564,7 @@ NOKPROBE_SYMBOL(do_secure_storage_access); void do_non_secure_storage_access(struct pt_regs *regs) { - struct gmap *gmap = (struct gmap *)S390_lowcore.gmap; + struct gmap *gmap = (struct gmap *)get_lowcore()->gmap; unsigned long gaddr = get_fault_address(regs); if (WARN_ON_ONCE(get_fault_type(regs) != GMAP_FAULT)) @@ -573,7 +576,7 @@ NOKPROBE_SYMBOL(do_non_secure_storage_access); void do_secure_storage_violation(struct pt_regs *regs) { - struct gmap *gmap = (struct gmap *)S390_lowcore.gmap; + struct gmap *gmap = (struct gmap *)get_lowcore()->gmap; unsigned long gaddr = get_fault_address(regs); /* diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 474a25ca5c48..eb0b51a36be0 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -287,7 +287,7 @@ EXPORT_SYMBOL_GPL(gmap_remove); */ void gmap_enable(struct gmap *gmap) { - S390_lowcore.gmap = (unsigned long) gmap; + get_lowcore()->gmap = (unsigned long)gmap; } EXPORT_SYMBOL_GPL(gmap_enable); @@ -297,7 +297,7 @@ EXPORT_SYMBOL_GPL(gmap_enable); */ void gmap_disable(struct gmap *gmap) { - S390_lowcore.gmap = 0UL; + get_lowcore()->gmap = 0UL; } EXPORT_SYMBOL_GPL(gmap_disable); @@ -308,7 +308,7 @@ EXPORT_SYMBOL_GPL(gmap_disable); */ struct gmap *gmap_get_enabled(void) { - return (struct gmap *) S390_lowcore.gmap; + return (struct gmap *)get_lowcore()->gmap; } EXPORT_SYMBOL_GPL(gmap_get_enabled); @@ -2733,7 +2733,7 @@ static int __s390_enable_skey_hugetlb(pte_t *pte, unsigned long addr, { pmd_t *pmd = (pmd_t *)pte; unsigned long start, end; - struct page *page = pmd_page(*pmd); + struct folio *folio = page_folio(pmd_page(*pmd)); /* * The write check makes sure we do not set a key on shared @@ -2748,7 +2748,7 @@ static int __s390_enable_skey_hugetlb(pte_t *pte, unsigned long addr, start = pmd_val(*pmd) & HPAGE_MASK; end = start + HPAGE_SIZE; __storage_key_init_range(start, end); - set_bit(PG_arch_1, &page->flags); + set_bit(PG_arch_1, &folio->flags); cond_resched(); return 0; } @@ -2841,13 +2841,15 @@ static const struct mm_walk_ops gather_pages_ops = { */ void s390_uv_destroy_pfns(unsigned long count, unsigned long *pfns) { + struct folio *folio; unsigned long i; for (i = 0; i < count; i++) { + folio = pfn_folio(pfns[i]); /* we always have an extra reference */ - uv_destroy_owned_page(pfn_to_phys(pfns[i])); + uv_destroy_folio(folio); /* get rid of the extra reference */ - put_page(pfn_to_page(pfns[i])); + folio_put(folio); cond_resched(); } } diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 2675aab4acc7..ded0eff58a19 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -121,7 +121,7 @@ static inline pte_t __rste_to_pte(unsigned long rste) static void clear_huge_pte_skeys(struct mm_struct *mm, unsigned long rste) { - struct page *page; + struct folio *folio; unsigned long size, paddr; if (!mm_uses_skeys(mm) || @@ -129,16 +129,16 @@ static void clear_huge_pte_skeys(struct mm_struct *mm, unsigned long rste) return; if ((rste & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) { - page = pud_page(__pud(rste)); + folio = page_folio(pud_page(__pud(rste))); size = PUD_SIZE; paddr = rste & PUD_MASK; } else { - page = pmd_page(__pmd(rste)); + folio = page_folio(pmd_page(__pmd(rste))); size = PMD_SIZE; paddr = rste & PMD_MASK; } - if (!test_and_set_bit(PG_arch_1, &page->flags)) + if (!test_and_set_bit(PG_arch_1, &folio->flags)) __storage_key_init_range(paddr, paddr + size); } @@ -169,7 +169,7 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, __set_huge_pte_at(mm, addr, ptep, pte); } -pte_t huge_ptep_get(pte_t *ptep) +pte_t huge_ptep_get(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { return __rste_to_pte(pte_val(*ptep)); } @@ -177,7 +177,7 @@ pte_t huge_ptep_get(pte_t *ptep) pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - pte_t pte = huge_ptep_get(ptep); + pte_t pte = huge_ptep_get(mm, addr, ptep); pmd_t *pmdp = (pmd_t *) ptep; pud_t *pudp = (pud_t *) ptep; diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index e769d2726f4e..e3d258f9e726 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -62,6 +62,7 @@ EXPORT_SYMBOL(zero_page_mask); static void __init setup_zero_pages(void) { + unsigned long total_pages = PHYS_PFN(memblock_phys_mem_size() - memblock_reserved_size()); unsigned int order; struct page *page; int i; @@ -70,7 +71,7 @@ static void __init setup_zero_pages(void) order = 7; /* Limit number of empty zero pages for small memory sizes */ - while (order > 2 && (totalram_pages() >> 10) < (1UL << order)) + while (order > 2 && (total_pages >> 10) < (1UL << order)) order--; empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order); @@ -107,6 +108,8 @@ void mark_rodata_ro(void) { unsigned long size = __end_ro_after_init - __start_ro_after_init; + if (MACHINE_HAS_NX) + system_ctl_set_bit(0, CR0_INSTRUCTION_EXEC_PROTECTION_BIT); __set_memory_ro(__start_ro_after_init, __end_ro_after_init); pr_info("Write protected read-only-after-init data: %luk\n", size >> 10); } @@ -169,13 +172,6 @@ void __init mem_init(void) setup_zero_pages(); /* Setup zeroed pages. */ } -void free_initmem(void) -{ - set_memory_rwnx((unsigned long)_sinittext, - (unsigned long)(_einittext - _sinittext) >> PAGE_SHIFT); - free_initmem_default(POISON_FREE_INITMEM); -} - unsigned long memory_block_size_bytes(void) { /* diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c index 632c3a55feed..28a18c42ba99 100644 --- a/arch/s390/mm/maccess.c +++ b/arch/s390/mm/maccess.c @@ -48,7 +48,7 @@ static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t siz } /* - * s390_kernel_write - write to kernel memory bypassing DAT + * __s390_kernel_write - write to kernel memory bypassing DAT * @dst: destination address * @src: source address * @size: number of bytes to copy @@ -61,7 +61,7 @@ static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t siz */ static DEFINE_SPINLOCK(s390_kernel_write_lock); -notrace void *s390_kernel_write(void *dst, const void *src, size_t size) +notrace void *__s390_kernel_write(void *dst, const void *src, size_t size) { void *tmp = dst; unsigned long flags; diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c index 01bc8fad64d6..5f805ad42d4c 100644 --- a/arch/s390/mm/pageattr.c +++ b/arch/s390/mm/pageattr.c @@ -75,7 +75,7 @@ static void pgt_set(unsigned long *old, unsigned long new, unsigned long addr, break; } table = (unsigned long *)((unsigned long)old & mask); - crdte(*old, new, table, dtt, addr, S390_lowcore.kernel_asce.val); + crdte(*old, new, table, dtt, addr, get_lowcore()->kernel_asce.val); } else if (MACHINE_HAS_IDTE) { cspg(old, *old, new); } else { diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c index abb629d7e131..f691e0fb66a2 100644 --- a/arch/s390/mm/pgalloc.c +++ b/arch/s390/mm/pgalloc.c @@ -55,6 +55,8 @@ unsigned long *crst_table_alloc(struct mm_struct *mm) void crst_table_free(struct mm_struct *mm, unsigned long *table) { + if (!table) + return; pagetable_free(virt_to_ptdesc(table)); } @@ -64,8 +66,8 @@ static void __crst_table_upgrade(void *arg) /* change all active ASCEs to avoid the creation of new TLBs */ if (current->active_mm == mm) { - S390_lowcore.user_asce.val = mm->context.asce; - local_ctl_load(7, &S390_lowcore.user_asce); + get_lowcore()->user_asce.val = mm->context.asce; + local_ctl_load(7, &get_lowcore()->user_asce); } __tlb_flush_local(); } @@ -262,6 +264,8 @@ static unsigned long *base_crst_alloc(unsigned long val) static void base_crst_free(unsigned long *table) { + if (!table) + return; pagetable_free(virt_to_ptdesc(table)); } diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index 41c714e21292..665b8228afeb 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -661,7 +661,6 @@ void __init vmem_map_init(void) { __set_memory_rox(_stext, _etext); __set_memory_ro(_etext, __end_rodata); - __set_memory_rox(_sinittext, _einittext); __set_memory_rox(__stext_amode31, __etext_amode31); /* * If the BEAR-enhancement facility is not installed the first @@ -670,16 +669,8 @@ void __init vmem_map_init(void) */ if (!static_key_enabled(&cpu_has_bear)) set_memory_x(0, 1); - if (debug_pagealloc_enabled()) { - /* - * Use RELOC_HIDE() as long as __va(0) translates to NULL, - * since performing pointer arithmetic on a NULL pointer - * has undefined behavior and generates compiler warnings. - */ - __set_memory_4k(__va(0), RELOC_HIDE(__va(0), ident_map_size)); - } - if (MACHINE_HAS_NX) - system_ctl_set_bit(0, CR0_INSTRUCTION_EXEC_PROTECTION_BIT); + if (debug_pagealloc_enabled()) + __set_memory_4k(__va(0), __va(0) + ident_map_size); pr_info("Write protected kernel read-only data: %luk\n", (unsigned long)(__end_rodata - _stext) >> 10); } diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c index 4be8f5cadd02..9d440a0b729e 100644 --- a/arch/s390/net/bpf_jit_comp.c +++ b/arch/s390/net/bpf_jit_comp.c @@ -31,11 +31,12 @@ #include <asm/nospec-branch.h> #include <asm/set_memory.h> #include <asm/text-patching.h> +#include <asm/unwind.h> #include "bpf_jit.h" struct bpf_jit { u32 seen; /* Flags to remember seen eBPF instructions */ - u32 seen_reg[16]; /* Array to remember which registers are used */ + u16 seen_regs; /* Mask to remember which registers are used */ u32 *addrs; /* Array with relative instruction addresses */ u8 *prg_buf; /* Start of program */ int size; /* Size of program and literal pool */ @@ -53,6 +54,8 @@ struct bpf_jit { int excnt; /* Number of exception table entries */ int prologue_plt_ret; /* Return address for prologue hotpatch PLT */ int prologue_plt; /* Start of prologue hotpatch PLT */ + int kern_arena; /* Pool offset of kernel arena address */ + u64 user_arena; /* User arena address */ }; #define SEEN_MEM BIT(0) /* use mem[] for temporary storage */ @@ -60,6 +63,8 @@ struct bpf_jit { #define SEEN_FUNC BIT(2) /* calls C functions */ #define SEEN_STACK (SEEN_FUNC | SEEN_MEM) +#define NVREGS 0xffc0 /* %r6-%r15 */ + /* * s390 registers */ @@ -118,8 +123,8 @@ static inline void reg_set_seen(struct bpf_jit *jit, u32 b1) { u32 r1 = reg2hex[b1]; - if (r1 >= 6 && r1 <= 15 && !jit->seen_reg[r1]) - jit->seen_reg[r1] = 1; + if (r1 >= 6 && r1 <= 15) + jit->seen_regs |= (1 << r1); } #define REG_SET_SEEN(b1) \ @@ -127,8 +132,6 @@ static inline void reg_set_seen(struct bpf_jit *jit, u32 b1) reg_set_seen(jit, b1); \ }) -#define REG_SEEN(b1) jit->seen_reg[reg2hex[(b1)]] - /* * EMIT macros for code generation */ @@ -436,12 +439,12 @@ static void restore_regs(struct bpf_jit *jit, u32 rs, u32 re, u32 stack_depth) /* * Return first seen register (from start) */ -static int get_start(struct bpf_jit *jit, int start) +static int get_start(u16 seen_regs, int start) { int i; for (i = start; i <= 15; i++) { - if (jit->seen_reg[i]) + if (seen_regs & (1 << i)) return i; } return 0; @@ -450,15 +453,15 @@ static int get_start(struct bpf_jit *jit, int start) /* * Return last seen register (from start) (gap >= 2) */ -static int get_end(struct bpf_jit *jit, int start) +static int get_end(u16 seen_regs, int start) { int i; for (i = start; i < 15; i++) { - if (!jit->seen_reg[i] && !jit->seen_reg[i + 1]) + if (!(seen_regs & (3 << i))) return i - 1; } - return jit->seen_reg[15] ? 15 : 14; + return (seen_regs & (1 << 15)) ? 15 : 14; } #define REGS_SAVE 1 @@ -467,8 +470,10 @@ static int get_end(struct bpf_jit *jit, int start) * Save and restore clobbered registers (6-15) on stack. * We save/restore registers in chunks with gap >= 2 registers. */ -static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth) +static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth, + u16 extra_regs) { + u16 seen_regs = jit->seen_regs | extra_regs; const int last = 15, save_restore_size = 6; int re = 6, rs; @@ -482,10 +487,10 @@ static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth) } do { - rs = get_start(jit, re); + rs = get_start(seen_regs, re); if (!rs) break; - re = get_end(jit, rs + 1); + re = get_end(seen_regs, rs + 1); if (op == REGS_SAVE) save_regs(jit, rs, re); else @@ -570,8 +575,21 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp, } /* Tail calls have to skip above initialization */ jit->tail_call_start = jit->prg; - /* Save registers */ - save_restore_regs(jit, REGS_SAVE, stack_depth); + if (fp->aux->exception_cb) { + /* + * Switch stack, the new address is in the 2nd parameter. + * + * Arrange the restoration of %r6-%r15 in the epilogue. + * Do not restore them now, the prog does not need them. + */ + /* lgr %r15,%r3 */ + EMIT4(0xb9040000, REG_15, REG_3); + jit->seen_regs |= NVREGS; + } else { + /* Save registers */ + save_restore_regs(jit, REGS_SAVE, stack_depth, + fp->aux->exception_boundary ? NVREGS : 0); + } /* Setup literal pool */ if (is_first_pass(jit) || (jit->seen & SEEN_LITERAL)) { if (!is_first_pass(jit) && @@ -647,7 +665,7 @@ static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth) /* Load exit code: lgr %r2,%b0 */ EMIT4(0xb9040000, REG_2, BPF_REG_0); /* Restore registers */ - save_restore_regs(jit, REGS_RESTORE, stack_depth); + save_restore_regs(jit, REGS_RESTORE, stack_depth, 0); if (nospec_uses_trampoline()) { jit->r14_thunk_ip = jit->prg; /* Generate __s390_indirect_jump_r14 thunk */ @@ -667,50 +685,111 @@ static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth) jit->prg += sizeof(struct bpf_plt); } -static int get_probe_mem_regno(const u8 *insn) -{ - /* - * insn must point to llgc, llgh, llgf, lg, lgb, lgh or lgf, which have - * destination register at the same position. - */ - if (insn[0] != 0xe3) /* common prefix */ - return -1; - if (insn[5] != 0x90 && /* llgc */ - insn[5] != 0x91 && /* llgh */ - insn[5] != 0x16 && /* llgf */ - insn[5] != 0x04 && /* lg */ - insn[5] != 0x77 && /* lgb */ - insn[5] != 0x15 && /* lgh */ - insn[5] != 0x14) /* lgf */ - return -1; - return insn[1] >> 4; -} - bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs) { regs->psw.addr = extable_fixup(x); - regs->gprs[x->data] = 0; + if (x->data != -1) + regs->gprs[x->data] = 0; return true; } -static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp, - int probe_prg, int nop_prg) +/* + * A single BPF probe instruction + */ +struct bpf_jit_probe { + int prg; /* JITed instruction offset */ + int nop_prg; /* JITed nop offset */ + int reg; /* Register to clear on exception */ + int arena_reg; /* Register to use for arena addressing */ +}; + +static void bpf_jit_probe_init(struct bpf_jit_probe *probe) +{ + probe->prg = -1; + probe->nop_prg = -1; + probe->reg = -1; + probe->arena_reg = REG_0; +} + +/* + * Handlers of certain exceptions leave psw.addr pointing to the instruction + * directly after the failing one. Therefore, create two exception table + * entries and also add a nop in case two probing instructions come directly + * after each other. + */ +static void bpf_jit_probe_emit_nop(struct bpf_jit *jit, + struct bpf_jit_probe *probe) +{ + if (probe->prg == -1 || probe->nop_prg != -1) + /* The probe is not armed or nop is already emitted. */ + return; + + probe->nop_prg = jit->prg; + /* bcr 0,%0 */ + _EMIT2(0x0700); +} + +static void bpf_jit_probe_load_pre(struct bpf_jit *jit, struct bpf_insn *insn, + struct bpf_jit_probe *probe) +{ + if (BPF_MODE(insn->code) != BPF_PROBE_MEM && + BPF_MODE(insn->code) != BPF_PROBE_MEMSX && + BPF_MODE(insn->code) != BPF_PROBE_MEM32) + return; + + if (BPF_MODE(insn->code) == BPF_PROBE_MEM32) { + /* lgrl %r1,kern_arena */ + EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena); + probe->arena_reg = REG_W1; + } + probe->prg = jit->prg; + probe->reg = reg2hex[insn->dst_reg]; +} + +static void bpf_jit_probe_store_pre(struct bpf_jit *jit, struct bpf_insn *insn, + struct bpf_jit_probe *probe) +{ + if (BPF_MODE(insn->code) != BPF_PROBE_MEM32) + return; + + /* lgrl %r1,kern_arena */ + EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena); + probe->arena_reg = REG_W1; + probe->prg = jit->prg; +} + +static void bpf_jit_probe_atomic_pre(struct bpf_jit *jit, + struct bpf_insn *insn, + struct bpf_jit_probe *probe) +{ + if (BPF_MODE(insn->code) != BPF_PROBE_ATOMIC) + return; + + /* lgrl %r1,kern_arena */ + EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena); + /* agr %r1,%dst */ + EMIT4(0xb9080000, REG_W1, insn->dst_reg); + probe->arena_reg = REG_W1; + probe->prg = jit->prg; +} + +static int bpf_jit_probe_post(struct bpf_jit *jit, struct bpf_prog *fp, + struct bpf_jit_probe *probe) { struct exception_table_entry *ex; - int reg, prg; + int i, prg; s64 delta; u8 *insn; - int i; + if (probe->prg == -1) + /* The probe is not armed. */ + return 0; + bpf_jit_probe_emit_nop(jit, probe); if (!fp->aux->extable) /* Do nothing during early JIT passes. */ return 0; - insn = jit->prg_buf + probe_prg; - reg = get_probe_mem_regno(insn); - if (WARN_ON_ONCE(reg < 0)) - /* JIT bug - unexpected probe instruction. */ - return -1; - if (WARN_ON_ONCE(probe_prg + insn_length(*insn) != nop_prg)) + insn = jit->prg_buf + probe->prg; + if (WARN_ON_ONCE(probe->prg + insn_length(*insn) != probe->nop_prg)) /* JIT bug - gap between probe and nop instructions. */ return -1; for (i = 0; i < 2; i++) { @@ -719,23 +798,24 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp, return -1; ex = &fp->aux->extable[jit->excnt]; /* Add extable entries for probe and nop instructions. */ - prg = i == 0 ? probe_prg : nop_prg; + prg = i == 0 ? probe->prg : probe->nop_prg; delta = jit->prg_buf + prg - (u8 *)&ex->insn; if (WARN_ON_ONCE(delta < INT_MIN || delta > INT_MAX)) /* JIT bug - code and extable must be close. */ return -1; ex->insn = delta; /* - * Always land on the nop. Note that extable infrastructure - * ignores fixup field, it is handled by ex_handler_bpf(). + * Land on the current instruction. Note that the extable + * infrastructure ignores the fixup field; it is handled by + * ex_handler_bpf(). */ - delta = jit->prg_buf + nop_prg - (u8 *)&ex->fixup; + delta = jit->prg_buf + jit->prg - (u8 *)&ex->fixup; if (WARN_ON_ONCE(delta < INT_MIN || delta > INT_MAX)) /* JIT bug - landing pad and extable must be close. */ return -1; ex->fixup = delta; ex->type = EX_TYPE_BPF; - ex->data = reg; + ex->data = probe->reg; jit->excnt++; } return 0; @@ -782,19 +862,15 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, s32 branch_oc_off = insn->off; u32 dst_reg = insn->dst_reg; u32 src_reg = insn->src_reg; + struct bpf_jit_probe probe; int last, insn_count = 1; u32 *addrs = jit->addrs; s32 imm = insn->imm; s16 off = insn->off; - int probe_prg = -1; unsigned int mask; - int nop_prg; int err; - if (BPF_CLASS(insn->code) == BPF_LDX && - (BPF_MODE(insn->code) == BPF_PROBE_MEM || - BPF_MODE(insn->code) == BPF_PROBE_MEMSX)) - probe_prg = jit->prg; + bpf_jit_probe_init(&probe); switch (insn->code) { /* @@ -823,6 +899,22 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, } break; case BPF_ALU64 | BPF_MOV | BPF_X: + if (insn_is_cast_user(insn)) { + int patch_brc; + + /* ltgr %dst,%src */ + EMIT4(0xb9020000, dst_reg, src_reg); + /* brc 8,0f */ + patch_brc = jit->prg; + EMIT4_PCREL_RIC(0xa7040000, 8, 0); + /* iihf %dst,user_arena>>32 */ + EMIT6_IMM(0xc0080000, dst_reg, jit->user_arena >> 32); + /* 0: */ + if (jit->prg_buf) + *(u16 *)(jit->prg_buf + patch_brc + 2) = + (jit->prg - patch_brc) >> 1; + break; + } switch (insn->off) { case 0: /* DST = SRC */ /* lgr %dst,%src */ @@ -1366,51 +1458,99 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, * BPF_ST(X) */ case BPF_STX | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = src_reg */ - /* stcy %src,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0072, src_reg, dst_reg, REG_0, off); + case BPF_STX | BPF_PROBE_MEM32 | BPF_B: + bpf_jit_probe_store_pre(jit, insn, &probe); + /* stcy %src,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0072, src_reg, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_STX | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = src */ - /* sthy %src,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0070, src_reg, dst_reg, REG_0, off); + case BPF_STX | BPF_PROBE_MEM32 | BPF_H: + bpf_jit_probe_store_pre(jit, insn, &probe); + /* sthy %src,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0070, src_reg, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_STX | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = src */ - /* sty %src,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0050, src_reg, dst_reg, REG_0, off); + case BPF_STX | BPF_PROBE_MEM32 | BPF_W: + bpf_jit_probe_store_pre(jit, insn, &probe); + /* sty %src,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0050, src_reg, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_STX | BPF_MEM | BPF_DW: /* (u64 *)(dst + off) = src */ - /* stg %src,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0024, src_reg, dst_reg, REG_0, off); + case BPF_STX | BPF_PROBE_MEM32 | BPF_DW: + bpf_jit_probe_store_pre(jit, insn, &probe); + /* stg %src,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, src_reg, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_ST | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = imm */ + case BPF_ST | BPF_PROBE_MEM32 | BPF_B: /* lhi %w0,imm */ EMIT4_IMM(0xa7080000, REG_W0, (u8) imm); - /* stcy %w0,off(dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0072, REG_W0, dst_reg, REG_0, off); + bpf_jit_probe_store_pre(jit, insn, &probe); + /* stcy %w0,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0072, REG_W0, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_ST | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = imm */ + case BPF_ST | BPF_PROBE_MEM32 | BPF_H: /* lhi %w0,imm */ EMIT4_IMM(0xa7080000, REG_W0, (u16) imm); - /* sthy %w0,off(dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0070, REG_W0, dst_reg, REG_0, off); + bpf_jit_probe_store_pre(jit, insn, &probe); + /* sthy %w0,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0070, REG_W0, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_ST | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = imm */ + case BPF_ST | BPF_PROBE_MEM32 | BPF_W: /* llilf %w0,imm */ EMIT6_IMM(0xc00f0000, REG_W0, (u32) imm); - /* sty %w0,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0050, REG_W0, dst_reg, REG_0, off); + bpf_jit_probe_store_pre(jit, insn, &probe); + /* sty %w0,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0050, REG_W0, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_ST | BPF_MEM | BPF_DW: /* *(u64 *)(dst + off) = imm */ + case BPF_ST | BPF_PROBE_MEM32 | BPF_DW: /* lgfi %w0,imm */ EMIT6_IMM(0xc0010000, REG_W0, imm); - /* stg %w0,off(%dst) */ - EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W0, dst_reg, REG_0, off); + bpf_jit_probe_store_pre(jit, insn, &probe); + /* stg %w0,off(%dst,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W0, dst_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; /* @@ -1418,15 +1558,30 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, */ case BPF_STX | BPF_ATOMIC | BPF_DW: case BPF_STX | BPF_ATOMIC | BPF_W: + case BPF_STX | BPF_PROBE_ATOMIC | BPF_DW: + case BPF_STX | BPF_PROBE_ATOMIC | BPF_W: { bool is32 = BPF_SIZE(insn->code) == BPF_W; + /* + * Unlike loads and stores, atomics have only a base register, + * but no index register. For the non-arena case, simply use + * %dst as a base. For the arena case, use the work register + * %r1: first, load the arena base into it, and then add %dst + * to it. + */ + probe.arena_reg = dst_reg; + switch (insn->imm) { -/* {op32|op64} {%w0|%src},%src,off(%dst) */ #define EMIT_ATOMIC(op32, op64) do { \ + bpf_jit_probe_atomic_pre(jit, insn, &probe); \ + /* {op32|op64} {%w0|%src},%src,off(%arena) */ \ EMIT6_DISP_LH(0xeb000000, is32 ? (op32) : (op64), \ (insn->imm & BPF_FETCH) ? src_reg : REG_W0, \ - src_reg, dst_reg, off); \ + src_reg, probe.arena_reg, off); \ + err = bpf_jit_probe_post(jit, fp, &probe); \ + if (err < 0) \ + return err; \ if (insn->imm & BPF_FETCH) { \ /* bcr 14,0 - see atomic_fetch_{add,and,or,xor}() */ \ _EMIT2(0x07e0); \ @@ -1455,25 +1610,50 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, EMIT_ATOMIC(0x00f7, 0x00e7); break; #undef EMIT_ATOMIC - case BPF_XCHG: - /* {ly|lg} %w0,off(%dst) */ + case BPF_XCHG: { + struct bpf_jit_probe load_probe = probe; + int loop_start; + + bpf_jit_probe_atomic_pre(jit, insn, &load_probe); + /* {ly|lg} %w0,off(%arena) */ EMIT6_DISP_LH(0xe3000000, is32 ? 0x0058 : 0x0004, REG_W0, REG_0, - dst_reg, off); - /* 0: {csy|csg} %w0,%src,off(%dst) */ + load_probe.arena_reg, off); + bpf_jit_probe_emit_nop(jit, &load_probe); + /* Reuse {ly|lg}'s arena_reg for {csy|csg}. */ + if (load_probe.prg != -1) { + probe.prg = jit->prg; + probe.arena_reg = load_probe.arena_reg; + } + loop_start = jit->prg; + /* 0: {csy|csg} %w0,%src,off(%arena) */ EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030, - REG_W0, src_reg, dst_reg, off); + REG_W0, src_reg, probe.arena_reg, off); + bpf_jit_probe_emit_nop(jit, &probe); /* brc 4,0b */ - EMIT4_PCREL_RIC(0xa7040000, 4, jit->prg - 6); + EMIT4_PCREL_RIC(0xa7040000, 4, loop_start); /* {llgfr|lgr} %src,%w0 */ EMIT4(is32 ? 0xb9160000 : 0xb9040000, src_reg, REG_W0); + /* Both probes should land here on exception. */ + err = bpf_jit_probe_post(jit, fp, &load_probe); + if (err < 0) + return err; + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; if (is32 && insn_is_zext(&insn[1])) insn_count = 2; break; + } case BPF_CMPXCHG: - /* 0: {csy|csg} %b0,%src,off(%dst) */ + bpf_jit_probe_atomic_pre(jit, insn, &probe); + /* 0: {csy|csg} %b0,%src,off(%arena) */ EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030, - BPF_REG_0, src_reg, dst_reg, off); + BPF_REG_0, src_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; break; default: pr_err("Unknown atomic operation %02x\n", insn->imm); @@ -1488,51 +1668,87 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, */ case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEM | BPF_B: - /* llgc %dst,0(off,%src) */ - EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg, REG_0, off); + case BPF_LDX | BPF_PROBE_MEM32 | BPF_B: + bpf_jit_probe_load_pre(jit, insn, &probe); + /* llgc %dst,off(%src,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; if (insn_is_zext(&insn[1])) insn_count = 2; break; case BPF_LDX | BPF_MEMSX | BPF_B: /* dst = *(s8 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEMSX | BPF_B: - /* lgb %dst,0(off,%src) */ + bpf_jit_probe_load_pre(jit, insn, &probe); + /* lgb %dst,off(%src) */ EMIT6_DISP_LH(0xe3000000, 0x0077, dst_reg, src_reg, REG_0, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEM | BPF_H: - /* llgh %dst,0(off,%src) */ - EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg, REG_0, off); + case BPF_LDX | BPF_PROBE_MEM32 | BPF_H: + bpf_jit_probe_load_pre(jit, insn, &probe); + /* llgh %dst,off(%src,%arena) */ + EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; if (insn_is_zext(&insn[1])) insn_count = 2; break; case BPF_LDX | BPF_MEMSX | BPF_H: /* dst = *(s16 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEMSX | BPF_H: - /* lgh %dst,0(off,%src) */ + bpf_jit_probe_load_pre(jit, insn, &probe); + /* lgh %dst,off(%src) */ EMIT6_DISP_LH(0xe3000000, 0x0015, dst_reg, src_reg, REG_0, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; jit->seen |= SEEN_MEM; break; case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEM | BPF_W: + case BPF_LDX | BPF_PROBE_MEM32 | BPF_W: + bpf_jit_probe_load_pre(jit, insn, &probe); /* llgf %dst,off(%src) */ jit->seen |= SEEN_MEM; - EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg, REG_0, off); + EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; if (insn_is_zext(&insn[1])) insn_count = 2; break; case BPF_LDX | BPF_MEMSX | BPF_W: /* dst = *(s32 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEMSX | BPF_W: + bpf_jit_probe_load_pre(jit, insn, &probe); /* lgf %dst,off(%src) */ jit->seen |= SEEN_MEM; EMIT6_DISP_LH(0xe3000000, 0x0014, dst_reg, src_reg, REG_0, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; break; case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */ case BPF_LDX | BPF_PROBE_MEM | BPF_DW: - /* lg %dst,0(off,%src) */ + case BPF_LDX | BPF_PROBE_MEM32 | BPF_DW: + bpf_jit_probe_load_pre(jit, insn, &probe); + /* lg %dst,off(%src,%arena) */ jit->seen |= SEEN_MEM; - EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg, REG_0, off); + EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg, + probe.arena_reg, off); + err = bpf_jit_probe_post(jit, fp, &probe); + if (err < 0) + return err; break; /* * BPF_JMP / CALL @@ -1647,7 +1863,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, /* * Restore registers before calling function */ - save_restore_regs(jit, REGS_RESTORE, stack_depth); + save_restore_regs(jit, REGS_RESTORE, stack_depth, 0); /* * goto *(prog->bpf_func + tail_call_start); @@ -1897,22 +2113,6 @@ branch_oc: return -1; } - if (probe_prg != -1) { - /* - * Handlers of certain exceptions leave psw.addr pointing to - * the instruction directly after the failing one. Therefore, - * create two exception table entries and also add a nop in - * case two probing instructions come directly after each - * other. - */ - nop_prg = jit->prg; - /* bcr 0,%0 */ - _EMIT2(0x0700); - err = bpf_jit_probe_mem(jit, fp, probe_prg, nop_prg); - if (err < 0) - return err; - } - return insn_count; } @@ -1958,12 +2158,18 @@ static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp, bool extra_pass, u32 stack_depth) { int i, insn_count, lit32_size, lit64_size; + u64 kern_arena; jit->lit32 = jit->lit32_start; jit->lit64 = jit->lit64_start; jit->prg = 0; jit->excnt = 0; + kern_arena = bpf_arena_get_kern_vm_start(fp->aux->arena); + if (kern_arena) + jit->kern_arena = _EMIT_CONST_U64(kern_arena); + jit->user_arena = bpf_arena_get_user_vm_start(fp->aux->arena); + bpf_jit_prologue(jit, fp, stack_depth); if (bpf_set_addr(jit, 0) < 0) return -1; @@ -2011,9 +2217,25 @@ static struct bpf_binary_header *bpf_jit_alloc(struct bpf_jit *jit, struct bpf_prog *fp) { struct bpf_binary_header *header; + struct bpf_insn *insn; u32 extable_size; u32 code_size; + int i; + + for (i = 0; i < fp->len; i++) { + insn = &fp->insnsi[i]; + if (BPF_CLASS(insn->code) == BPF_STX && + BPF_MODE(insn->code) == BPF_PROBE_ATOMIC && + (BPF_SIZE(insn->code) == BPF_DW || + BPF_SIZE(insn->code) == BPF_W) && + insn->imm == BPF_XCHG) + /* + * bpf_jit_insn() emits a load and a compare-and-swap, + * both of which need to be probed. + */ + fp->aux->num_exentries += 1; + } /* We need two entries per insn. */ fp->aux->num_exentries *= 2; @@ -2689,3 +2911,52 @@ bool bpf_jit_supports_subprog_tailcalls(void) { return true; } + +bool bpf_jit_supports_arena(void) +{ + return true; +} + +bool bpf_jit_supports_insn(struct bpf_insn *insn, bool in_arena) +{ + /* + * Currently the verifier uses this function only to check which + * atomic stores to arena are supported, and they all are. + */ + return true; +} + +bool bpf_jit_supports_exceptions(void) +{ + /* + * Exceptions require unwinding support, which is always available, + * because the kernel is always built with backchain. + */ + return true; +} + +void arch_bpf_stack_walk(bool (*consume_fn)(void *, u64, u64, u64), + void *cookie) +{ + unsigned long addr, prev_addr = 0; + struct unwind_state state; + + unwind_for_each_frame(&state, NULL, NULL, 0) { + addr = unwind_get_return_address(&state); + if (!addr) + break; + /* + * addr is a return address and state.sp is the value of %r15 + * at this address. exception_cb needs %r15 at entry to the + * function containing addr, so take the next state.sp. + * + * There is no bp, and the exception_cb prog does not need one + * to perform a quasi-longjmp. The common code requires a + * non-zero bp, so pass sp there as well. + */ + if (prev_addr && !consume_fn(cookie, prev_addr, state.sp, + state.sp)) + break; + prev_addr = addr; + } +} diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index 0de0f6e405b5..cff4838fad21 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -1064,7 +1064,7 @@ char * __init pcibios_setup(char *str) return NULL; } if (!strcmp(str, "nomio")) { - S390_lowcore.machine_flags &= ~MACHINE_FLAG_PCI_MIO; + get_lowcore()->machine_flags &= ~MACHINE_FLAG_PCI_MIO; return NULL; } if (!strcmp(str, "force_floating")) { diff --git a/arch/s390/pci/pci_irq.c b/arch/s390/pci/pci_irq.c index 0ef83b6ac0db..84482a921332 100644 --- a/arch/s390/pci/pci_irq.c +++ b/arch/s390/pci/pci_irq.c @@ -268,33 +268,20 @@ static void zpci_floating_irq_handler(struct airq_struct *airq, } } -int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) +static int __alloc_airq(struct zpci_dev *zdev, int msi_vecs, + unsigned long *bit) { - struct zpci_dev *zdev = to_zpci(pdev); - unsigned int hwirq, msi_vecs, cpu; - unsigned long bit; - struct msi_desc *msi; - struct msi_msg msg; - int cpu_addr; - int rc, irq; - - zdev->aisb = -1UL; - zdev->msi_first_bit = -1U; - if (type == PCI_CAP_ID_MSI && nvec > 1) - return 1; - msi_vecs = min_t(unsigned int, nvec, zdev->max_msi); - if (irq_delivery == DIRECTED) { /* Allocate cpu vector bits */ - bit = airq_iv_alloc(zpci_ibv[0], msi_vecs); - if (bit == -1UL) + *bit = airq_iv_alloc(zpci_ibv[0], msi_vecs); + if (*bit == -1UL) return -EIO; } else { /* Allocate adapter summary indicator bit */ - bit = airq_iv_alloc_bit(zpci_sbv); - if (bit == -1UL) + *bit = airq_iv_alloc_bit(zpci_sbv); + if (*bit == -1UL) return -EIO; - zdev->aisb = bit; + zdev->aisb = *bit; /* Create adapter interrupt vector */ zdev->aibv = airq_iv_create(msi_vecs, AIRQ_IV_DATA | AIRQ_IV_BITLOCK, NULL); @@ -302,27 +289,66 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) return -ENOMEM; /* Wire up shortcut pointer */ - zpci_ibv[bit] = zdev->aibv; + zpci_ibv[*bit] = zdev->aibv; /* Each function has its own interrupt vector */ - bit = 0; + *bit = 0; } + return 0; +} - /* Request MSI interrupts */ +int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) +{ + unsigned int hwirq, msi_vecs, irqs_per_msi, i, cpu; + struct zpci_dev *zdev = to_zpci(pdev); + struct msi_desc *msi; + struct msi_msg msg; + unsigned long bit; + int cpu_addr; + int rc, irq; + + zdev->aisb = -1UL; + zdev->msi_first_bit = -1U; + + msi_vecs = min_t(unsigned int, nvec, zdev->max_msi); + if (msi_vecs < nvec) { + pr_info("%s requested %d irqs, allocate system limit of %d", + pci_name(pdev), nvec, zdev->max_msi); + } + + rc = __alloc_airq(zdev, msi_vecs, &bit); + if (rc < 0) + return rc; + + /* + * Request MSI interrupts: + * When using MSI, nvec_used interrupt sources and their irq + * descriptors are controlled through one msi descriptor. + * Thus the outer loop over msi descriptors shall run only once, + * while two inner loops iterate over the interrupt vectors. + * When using MSI-X, each interrupt vector/irq descriptor + * is bound to exactly one msi descriptor (nvec_used is one). + * So the inner loops are executed once, while the outer iterates + * over the MSI-X descriptors. + */ hwirq = bit; msi_for_each_desc(msi, &pdev->dev, MSI_DESC_NOTASSOCIATED) { - rc = -EIO; if (hwirq - bit >= msi_vecs) break; - irq = __irq_alloc_descs(-1, 0, 1, 0, THIS_MODULE, - (irq_delivery == DIRECTED) ? - msi->affinity : NULL); + irqs_per_msi = min_t(unsigned int, msi_vecs, msi->nvec_used); + irq = __irq_alloc_descs(-1, 0, irqs_per_msi, 0, THIS_MODULE, + (irq_delivery == DIRECTED) ? + msi->affinity : NULL); if (irq < 0) return -ENOMEM; - rc = irq_set_msi_desc(irq, msi); - if (rc) - return rc; - irq_set_chip_and_handler(irq, &zpci_irq_chip, - handle_percpu_irq); + + for (i = 0; i < irqs_per_msi; i++) { + rc = irq_set_msi_desc_off(irq, i, msi); + if (rc) + return rc; + irq_set_chip_and_handler(irq + i, &zpci_irq_chip, + handle_percpu_irq); + } + msg.data = hwirq - bit; if (irq_delivery == DIRECTED) { if (msi->affinity) @@ -335,31 +361,35 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) msg.address_lo |= (cpu_addr << 8); for_each_possible_cpu(cpu) { - airq_iv_set_data(zpci_ibv[cpu], hwirq, irq); + for (i = 0; i < irqs_per_msi; i++) + airq_iv_set_data(zpci_ibv[cpu], + hwirq + i, irq + i); } } else { msg.address_lo = zdev->msi_addr & 0xffffffff; - airq_iv_set_data(zdev->aibv, hwirq, irq); + for (i = 0; i < irqs_per_msi; i++) + airq_iv_set_data(zdev->aibv, hwirq + i, irq + i); } msg.address_hi = zdev->msi_addr >> 32; pci_write_msi_msg(irq, &msg); - hwirq++; + hwirq += irqs_per_msi; } zdev->msi_first_bit = bit; - zdev->msi_nr_irqs = msi_vecs; + zdev->msi_nr_irqs = hwirq - bit; rc = zpci_set_irq(zdev); if (rc) return rc; - return (msi_vecs == nvec) ? 0 : msi_vecs; + return (zdev->msi_nr_irqs == nvec) ? 0 : zdev->msi_nr_irqs; } void arch_teardown_msi_irqs(struct pci_dev *pdev) { struct zpci_dev *zdev = to_zpci(pdev); struct msi_desc *msi; + unsigned int i; int rc; /* Disable interrupts */ @@ -369,8 +399,10 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev) /* Release MSI interrupts */ msi_for_each_desc(msi, &pdev->dev, MSI_DESC_ASSOCIATED) { - irq_set_msi_desc(msi->irq, NULL); - irq_free_desc(msi->irq); + for (i = 0; i < msi->nvec_used; i++) { + irq_set_msi_desc(msi->irq + i, NULL); + irq_free_desc(msi->irq + i); + } msi->msg.address_lo = 0; msi->msg.address_hi = 0; msi->msg.data = 0; diff --git a/arch/s390/tools/relocs.c b/arch/s390/tools/relocs.c index a74dbd5c9896..30a732c808f3 100644 --- a/arch/s390/tools/relocs.c +++ b/arch/s390/tools/relocs.c @@ -280,7 +280,7 @@ static int do_reloc(struct section *sec, Elf_Rel *rel) case R_390_GOTOFF64: break; case R_390_64: - add_reloc(&relocs64, offset - ehdr.e_entry); + add_reloc(&relocs64, offset); break; default: die("Unsupported relocation type: %d\n", r_type); |