diff options
| author | Linus Torvalds | 2024-05-13 08:33:52 -0700 |
|---|---|---|
| committer | Linus Torvalds | 2024-05-13 08:33:52 -0700 |
| commit | d65e1a0f305ba3e7aabf6261a37bb871790d9f93 (patch) | |
| tree | 4033039381e4a35eb76158e5b5130f05ae204139 /arch | |
| parent | a38297e3fb012ddfa7ce0321a7e5a8daeb1872b6 (diff) | |
| parent | 1812dc9c334f98227c65bc9c475f16fb6840a94b (diff) | |
Merge tag 's390-6.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Alexander Gordeev:
- Store AP Query Configuration Information in a static buffer
- Rework the AP initialization and add missing cleanups to the error
path
- Swap IRQ and AP bus/device registration to avoid race conditions
- Export prot_virt_guest symbol
- Introduce AP configuration changes notifier interface to facilitate
modularization of the AP bus
- Add CONFIG_AP kernel configuration option to allow modularization of
the AP bus
- Rework CONFIG_ZCRYPT_DEBUG kernel configuration option description
and dependency and rename it to CONFIG_AP_DEBUG
- Convert sprintf() and snprintf() to sysfs_emit() in CIO code
- Adjust indentation of RELOCS command build step
- Make crypto performance counters upward compatible
- Convert make_page_secure() and gmap_make_secure() to use folio
- Rework channel-utilization-block (CUB) handling in preparation of
introducing additional CUBs
- Use attribute groups to simplify registration, removal and extension
of measurement-related channel-path sysfs attributes
- Add a per-channel-path binary "ext_measurement" sysfs attribute that
provides access to extended channel-path measurement data
- Export measurement data for all channel-measurement-groups (CMG), not
only for a specific ones. This enables support of new CMG data
formats in userspace without the need for kernel changes
- Add a per-channel-path sysfs attribute "speed_bps" that provides the
operating speed in bits per second or 0 if the operating speed is not
available
- The CIO tracepoint subchannel-type field "st" is incorrectly set to
the value of subchannel-enabled SCHIB "ena" field. Fix that
- Do not forcefully limit vmemmap starting address to MAX_PHYSMEM_BITS
- Consider the maximum physical address available to a DCSS segment
(512GB) when memory layout is set up
- Simplify the virtual memory layout setup by reducing the size of
identity mapping vs vmemmap overlap
- Swap vmalloc and Lowcore/Real Memory Copy areas in virtual memory.
This will allow to place the kernel image next to kernel modules
- Move everyting KASLR related from <asm/setup.h> to <asm/page.h>
- Put virtual memory layout information into a structure to improve
code generation
- Currently __kaslr_offset is the kernel offset in both physical and
virtual memory spaces. Uncouple these offsets to allow uncoupling of
the addresses spaces
- Currently the identity mapping base address is implicit and is always
set to zero. Make it explicit by putting into __identity_base
persistent boot variable and use it in proper context
- Introduce .amode31 section start and end macros AMODE31_START and
AMODE31_END
- Introduce OS_INFO entries that do not reference any data in memory,
but rather provide only values
- Store virtual memory layout in OS_INFO. It is read out by
makedumpfile, crash and other tools
- Store virtual memory layout in VMCORE_INFO. It is read out by crash
and other tools when /proc/kcore device is used
- Create additional PT_LOAD ELF program header that covers kernel image
only, so that vmcore tools could locate kernel text and data when
virtual and physical memory spaces are uncoupled
- Uncouple physical and virtual address spaces
- Map kernel at fixed location when KASLR mode is disabled. The
location is defined by CONFIG_KERNEL_IMAGE_BASE kernel configuration
value.
- Rework deployment of kernel image for both compressed and
uncompressed variants as defined by CONFIG_KERNEL_UNCOMPRESSED kernel
configuration value
- Move .vmlinux.relocs section in front of the compressed kernel. The
interim section rescue step is avoided as result
- Correct modules thunk offset calculation when branch target is more
than 2GB away
- Kernel modules contain their own set of expoline thunks. Now that the
kernel modules area is less than 4GB away from kernel expoline
thunks, make modules use kernel expolines. Also make EXPOLINE_EXTERN
the default if the compiler supports it
- userfaultfd can insert shared zeropages into processes running VMs,
but that is not allowed for s390. Fallback to allocating a fresh
zeroed anonymous folio and insert that instead
- Re-enable shared zeropages for non-PV and non-skeys KVM guests
- Rename hex2bitmap() to ap_hex2bitmap() and export it for external use
- Add ap_config sysfs attribute to provide the means for setting or
displaying adapters, domains and control domains assigned to a
vfio-ap mediated device in a single operation
- Make vfio_ap_mdev_link_queue() ignore duplicate link requests
- Add write support to ap_config sysfs attribute to allow atomic update
a vfio-ap mediated device state
- Document ap_config sysfs attribute
- Function os_info_old_init() is expected to be called only from a
regular kdump kernel. Enable it to be called from a stand-alone dump
kernel
- Address gcc -Warray-bounds warning and fix array size in struct
os_info
- s390 does not support SMBIOS, so drop unneeded CONFIG_DMI checks
- Use unwinder instead of __builtin_return_address() with ftrace to
prevent returning of undefined values
- Sections .hash and .gnu.hash are only created when CONFIG_PIE_BUILD
kernel is enabled. Drop these for the case CONFIG_PIE_BUILD is
disabled
- Compile kernel with -fPIC and link with -no-pie to allow kpatch
feature always succeed and drop the whole CONFIG_PIE_BUILD
option-enabled code
- Add missing virt_to_phys() converter for VSIE facility and crypto
control blocks
* tag 's390-6.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (54 commits)
Revert "s390: Relocate vmlinux ELF data to virtual address space"
KVM: s390: vsie: Use virt_to_phys for crypto control block
s390: Relocate vmlinux ELF data to virtual address space
s390: Compile kernel with -fPIC and link with -no-pie
s390: vmlinux.lds.S: Drop .hash and .gnu.hash for !CONFIG_PIE_BUILD
s390/ftrace: Use unwinder instead of __builtin_return_address()
s390/pci: Drop unneeded reference to CONFIG_DMI
s390/os_info: Fix array size in struct os_info
s390/os_info: Initialize old os_info in standalone dump kernel
docs: Update s390 vfio-ap doc for ap_config sysfs attribute
s390/vfio-ap: Add write support to sysfs attr ap_config
s390/vfio-ap: Ignore duplicate link requests in vfio_ap_mdev_link_queue
s390/vfio-ap: Add sysfs attr, ap_config, to export mdev state
s390/ap: Externalize AP bus specific bitmap reading function
s390/mm: Re-enable the shared zeropage for !PV and !skeys KVM guests
mm/userfaultfd: Do not place zeropages when zeropages are disallowed
s390/expoline: Make modules use kernel expolines
s390/nospec: Correct modules thunk offset calculation
s390/boot: Do not rescue .vmlinux.relocs section
s390/boot: Rework deployment of the kernel image
...
Diffstat (limited to 'arch')
47 files changed, 698 insertions, 455 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 8f01ada6845e..7e7fe89c9b25 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -17,6 +17,9 @@ config ARCH_HAS_ILOG2_U32 config ARCH_HAS_ILOG2_U64 def_bool n +config ARCH_PROC_KCORE_TEXT + def_bool y + config GENERIC_HWEIGHT def_bool y @@ -552,7 +555,7 @@ config EXPOLINE If unsure, say N. config EXPOLINE_EXTERN - def_bool n + def_bool y if EXPOLINE depends on EXPOLINE depends on CC_IS_GCC && GCC_VERSION >= 110200 depends on $(success,$(srctree)/arch/s390/tools/gcc-thunk-extern.sh $(CC)) @@ -590,18 +593,6 @@ config RELOCATABLE Note: this option exists only for documentation purposes, please do not remove it. -config PIE_BUILD - def_bool CC_IS_CLANG && !$(cc-option,-munaligned-symbols) - help - If the compiler is unable to generate code that can manage unaligned - symbols, the kernel is linked as a position-independent executable - (PIE) and includes dynamic relocations that are processed early - during bootup. - - For kpatch functionality, it is recommended to build the kernel - without the PIE_BUILD option. PIE_BUILD is only enabled when the - compiler lacks proper support for handling unaligned symbols. - config RANDOMIZE_BASE bool "Randomize the address of the kernel image (KASLR)" default y @@ -611,6 +602,25 @@ config RANDOMIZE_BASE as a security feature that deters exploit attempts relying on knowledge of the location of kernel internals. +config KERNEL_IMAGE_BASE + hex "Kernel image base address" + range 0x100000 0x1FFFFFE0000000 if !KASAN + range 0x100000 0x1BFFFFE0000000 if KASAN + default 0x3FFE0000000 if !KASAN + default 0x7FFFE0000000 if KASAN + help + This is the address at which the kernel image is loaded in case + Kernel Address Space Layout Randomization (KASLR) is disabled. + + In case the Protected virtualization guest support is enabled the + Ultravisor imposes a virtual address limit. If the value of this + option leads to the kernel image exceeding the Ultravisor limit, + this option is ignored and the image is loaded below the limit. + + If the value of this option leads to the kernel image overlapping + the virtual memory where other data structures are located, this + option is ignored and the image is loaded above the structures. + endmenu menu "Memory setup" @@ -724,6 +734,33 @@ config EADM_SCH To compile this driver as a module, choose M here: the module will be called eadm_sch. +config AP + def_tristate y + prompt "Support for Adjunct Processors (ap)" + help + This driver allows usage to Adjunct Processor (AP) devices via + the ap bus, cards and queues. Supported Adjunct Processors are + the CryptoExpress Cards (CEX). + + To compile this driver as a module, choose M here: the + module will be called ap. + + If unsure, say Y (default). + +config AP_DEBUG + def_bool n + prompt "Enable debug features for Adjunct Processor (ap) devices" + depends on AP + help + Say 'Y' here to enable some additional debug features for Adjunct + Processor (ap) devices. + + There will be some more sysfs attributes displayed for ap queues. + + Do not enable on production level kernel build. + + If unsure, say N. + config VFIO_CCW def_tristate n prompt "Support for VFIO-CCW subchannels" @@ -740,7 +777,7 @@ config VFIO_AP prompt "VFIO support for AP devices" depends on KVM depends on VFIO - depends on ZCRYPT + depends on AP select VFIO_MDEV help This driver grants access to Adjunct Processor (AP) devices diff --git a/arch/s390/Makefile b/arch/s390/Makefile index 2dbb2d2f22f9..f2b21c7a70ef 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -14,14 +14,9 @@ KBUILD_AFLAGS_MODULE += -fPIC KBUILD_CFLAGS_MODULE += -fPIC KBUILD_AFLAGS += -m64 KBUILD_CFLAGS += -m64 -ifdef CONFIG_PIE_BUILD -KBUILD_CFLAGS += -fPIE -LDFLAGS_vmlinux := -pie -z notext -else -KBUILD_CFLAGS += $(call cc-option,-munaligned-symbols,) -LDFLAGS_vmlinux := --emit-relocs --discard-none +KBUILD_CFLAGS += -fPIC +LDFLAGS_vmlinux := -no-pie --emit-relocs --discard-none extra_tools := relocs -endif aflags_dwarf := -Wa,-gdwarf-2 KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__ ifndef CONFIG_AS_IS_LLVM @@ -88,7 +83,6 @@ endif ifdef CONFIG_EXPOLINE ifdef CONFIG_EXPOLINE_EXTERN - KBUILD_LDFLAGS_MODULE += arch/s390/lib/expoline/expoline.o CC_FLAGS_EXPOLINE := -mindirect-branch=thunk-extern CC_FLAGS_EXPOLINE += -mfunction-return=thunk-extern else @@ -167,11 +161,6 @@ vdso_prepare: prepare0 vdso-install-y += arch/s390/kernel/vdso64/vdso64.so.dbg vdso-install-$(CONFIG_COMPAT) += arch/s390/kernel/vdso32/vdso32.so.dbg -ifdef CONFIG_EXPOLINE_EXTERN -modules_prepare: expoline_prepare -expoline_prepare: scripts - $(Q)$(MAKE) $(build)=arch/s390/lib/expoline arch/s390/lib/expoline/expoline.o -endif endif # Don't use tabs in echo arguments diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile index 294f08a8811a..070c9b2e905f 100644 --- a/arch/s390/boot/Makefile +++ b/arch/s390/boot/Makefile @@ -37,8 +37,7 @@ CFLAGS_sclp_early_core.o += -I$(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 -obj-y += $(if $(CONFIG_PIE_BUILD),machine_kexec_reloc.o,relocs.o) +obj-y += version.o pgm_check_info.o ctype.o ipl_data.o relocs.o obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE)) += uv.o obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o obj-y += $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) info.o @@ -49,9 +48,7 @@ targets := bzImage section_cmp.boot.data section_cmp.boot.preserved.data $(obj-y targets += vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 targets += vmlinux.bin.xz vmlinux.bin.lzma vmlinux.bin.lzo vmlinux.bin.lz4 targets += vmlinux.bin.zst info.bin syms.bin vmlinux.syms $(obj-all) -ifndef CONFIG_PIE_BUILD targets += relocs.S -endif OBJECTS := $(addprefix $(obj)/,$(obj-y)) OBJECTS_ALL := $(addprefix $(obj)/,$(obj-all)) @@ -110,13 +107,11 @@ OBJCOPYFLAGS_vmlinux.bin := -O binary --remove-section=.comment --remove-section $(obj)/vmlinux.bin: vmlinux FORCE $(call if_changed,objcopy) -ifndef CONFIG_PIE_BUILD CMD_RELOCS=arch/s390/tools/relocs -quiet_cmd_relocs = RELOCS $@ +quiet_cmd_relocs = RELOCS $@ cmd_relocs = $(CMD_RELOCS) $< > $@ $(obj)/relocs.S: vmlinux FORCE $(call if_changed,relocs) -endif suffix-$(CONFIG_KERNEL_GZIP) := .gz suffix-$(CONFIG_KERNEL_BZIP2) := .bz2 diff --git a/arch/s390/boot/boot.h b/arch/s390/boot/boot.h index 567d60f78bbc..18027fdc92b0 100644 --- a/arch/s390/boot/boot.h +++ b/arch/s390/boot/boot.h @@ -17,7 +17,6 @@ struct machine_info { }; struct vmlinux_info { - unsigned long default_lma; unsigned long entry; unsigned long image_size; /* does not include .bss */ unsigned long bss_size; /* uncompressed image .bss size */ @@ -25,14 +24,8 @@ struct vmlinux_info { unsigned long bootdata_size; unsigned long bootdata_preserved_off; unsigned long bootdata_preserved_size; -#ifdef CONFIG_PIE_BUILD - unsigned long dynsym_start; - unsigned long rela_dyn_start; - unsigned long rela_dyn_end; -#else unsigned long got_start; unsigned long got_end; -#endif unsigned long amode31_size; unsigned long init_mm_off; unsigned long swapper_pg_dir_off; @@ -74,10 +67,11 @@ void sclp_early_setup_buffer(void); void print_pgm_check_info(void); unsigned long randomize_within_range(unsigned long size, unsigned long align, unsigned long min, unsigned long max); -void setup_vmem(unsigned long asce_limit); +void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned long asce_limit); void __printf(1, 2) decompressor_printk(const char *fmt, ...); void print_stacktrace(unsigned long sp); void error(char *m); +int get_random(unsigned long limit, unsigned long *value); extern struct machine_info machine; @@ -98,6 +92,10 @@ extern struct vmlinux_info _vmlinux_info; #define vmlinux _vmlinux_info #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) +#define __identity_va(x) ((void *)((unsigned long)(x) + __identity_base)) +#define __identity_pa(x) ((unsigned long)(x) - __identity_base) static inline bool intersects(unsigned long addr0, unsigned long size0, unsigned long addr1, unsigned long size1) diff --git a/arch/s390/boot/decompressor.c b/arch/s390/boot/decompressor.c index d762733a0753..f478e8e9cbda 100644 --- a/arch/s390/boot/decompressor.c +++ b/arch/s390/boot/decompressor.c @@ -63,24 +63,13 @@ static unsigned long free_mem_end_ptr = (unsigned long) _end + BOOT_HEAP_SIZE; #include "../../../../lib/decompress_unzstd.c" #endif -#define decompress_offset ALIGN((unsigned long)_end + BOOT_HEAP_SIZE, PAGE_SIZE) - unsigned long mem_safe_offset(void) { - /* - * due to 4MB HEAD_SIZE for bzip2 - * 'decompress_offset + vmlinux.image_size' could be larger than - * kernel at final position + its .bss, so take the larger of two - */ - return max(decompress_offset + vmlinux.image_size, - vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size); + return ALIGN(free_mem_end_ptr, PAGE_SIZE); } -void *decompress_kernel(void) +void deploy_kernel(void *output) { - void *output = (void *)decompress_offset; - __decompress(_compressed_start, _compressed_end - _compressed_start, NULL, NULL, output, vmlinux.image_size, NULL, error); - return output; } diff --git a/arch/s390/boot/decompressor.h b/arch/s390/boot/decompressor.h index 92b81d2ea35d..4f966f06bd65 100644 --- a/arch/s390/boot/decompressor.h +++ b/arch/s390/boot/decompressor.h @@ -2,11 +2,9 @@ #ifndef BOOT_COMPRESSED_DECOMPRESSOR_H #define BOOT_COMPRESSED_DECOMPRESSOR_H -#ifdef CONFIG_KERNEL_UNCOMPRESSED -static inline void *decompress_kernel(void) { return NULL; } -#else -void *decompress_kernel(void); -#endif +#ifndef CONFIG_KERNEL_UNCOMPRESSED unsigned long mem_safe_offset(void); +void deploy_kernel(void *output); +#endif #endif /* BOOT_COMPRESSED_DECOMPRESSOR_H */ diff --git a/arch/s390/boot/kaslr.c b/arch/s390/boot/kaslr.c index 90602101e2ae..bd3bf5ef472d 100644 --- a/arch/s390/boot/kaslr.c +++ b/arch/s390/boot/kaslr.c @@ -43,7 +43,7 @@ static int check_prng(void) return PRNG_MODE_TDES; } -static int get_random(unsigned long limit, unsigned long *value) +int get_random(unsigned long limit, unsigned long *value) { struct prng_parm prng = { /* initial parameter block for tdes mode, copied from libica */ diff --git a/arch/s390/boot/pgm_check_info.c b/arch/s390/boot/pgm_check_info.c index 97244cd7a206..ea96275b0380 100644 --- a/arch/s390/boot/pgm_check_info.c +++ b/arch/s390/boot/pgm_check_info.c @@ -153,8 +153,10 @@ void print_pgm_check_info(void) decompressor_printk("Kernel command line: %s\n", early_command_line); decompressor_printk("Kernel fault: interruption code %04x ilc:%x\n", S390_lowcore.pgm_code, S390_lowcore.pgm_ilc >> 1); - if (kaslr_enabled()) + 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, diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c index 6cf89314209a..467283b112cd 100644 --- a/arch/s390/boot/startup.c +++ b/arch/s390/boot/startup.c @@ -3,6 +3,7 @@ #include <linux/elf.h> #include <asm/page-states.h> #include <asm/boot_data.h> +#include <asm/extmem.h> #include <asm/sections.h> #include <asm/maccess.h> #include <asm/cpu_mf.h> @@ -18,7 +19,7 @@ #include "boot.h" #include "uv.h" -unsigned long __bootdata_preserved(__kaslr_offset); +struct vm_layout __bootdata_preserved(vm_layout); unsigned long __bootdata_preserved(__abs_lowcore); unsigned long __bootdata_preserved(__memcpy_real_area); pte_t *__bootdata_preserved(memcpy_real_ptep); @@ -29,7 +30,6 @@ unsigned long __bootdata_preserved(vmemmap_size); unsigned long __bootdata_preserved(MODULES_VADDR); unsigned long __bootdata_preserved(MODULES_END); unsigned long __bootdata_preserved(max_mappable); -unsigned long __bootdata(ident_map_size); u64 __bootdata_preserved(stfle_fac_list[16]); u64 __bootdata_preserved(alt_stfle_fac_list[16]); @@ -109,9 +109,19 @@ static void setup_lpp(void) } #ifdef CONFIG_KERNEL_UNCOMPRESSED -unsigned long mem_safe_offset(void) +static unsigned long mem_safe_offset(void) { - return vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size; + return (unsigned long)_compressed_start; +} + +static void deploy_kernel(void *output) +{ + void *uncompressed_start = (void *)_compressed_start; + + if (output == uncompressed_start) + return; + memmove(output, uncompressed_start, vmlinux.image_size); + memset(uncompressed_start, 0, vmlinux.image_size); } #endif @@ -141,70 +151,18 @@ static void copy_bootdata(void) memcpy((void *)vmlinux.bootdata_preserved_off, __boot_data_preserved_start, vmlinux.bootdata_preserved_size); } -#ifdef CONFIG_PIE_BUILD -static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, unsigned long offset) -{ - Elf64_Rela *rela_start, *rela_end, *rela; - int r_type, r_sym, rc; - Elf64_Addr loc, val; - Elf64_Sym *dynsym; - - rela_start = (Elf64_Rela *) vmlinux.rela_dyn_start; - rela_end = (Elf64_Rela *) vmlinux.rela_dyn_end; - dynsym = (Elf64_Sym *) vmlinux.dynsym_start; - for (rela = rela_start; rela < rela_end; rela++) { - loc = rela->r_offset + offset; - val = rela->r_addend; - r_sym = ELF64_R_SYM(rela->r_info); - if (r_sym) { - if (dynsym[r_sym].st_shndx != SHN_UNDEF) - val += dynsym[r_sym].st_value + offset; - } else { - /* - * 0 == undefined symbol table index (STN_UNDEF), - * used for R_390_RELATIVE, only add KASLR offset - */ - val += offset; - } - r_type = ELF64_R_TYPE(rela->r_info); - rc = arch_kexec_do_relocs(r_type, (void *) loc, val, 0); - if (rc) - error("Unknown relocation type"); - } -} - -static void kaslr_adjust_got(unsigned long offset) {} -static void rescue_relocs(void) {} -static void free_relocs(void) {} -#else -static int *vmlinux_relocs_64_start; -static int *vmlinux_relocs_64_end; - -static void rescue_relocs(void) -{ - unsigned long size = __vmlinux_relocs_64_end - __vmlinux_relocs_64_start; - - vmlinux_relocs_64_start = (void *)physmem_alloc_top_down(RR_RELOC, size, 0); - vmlinux_relocs_64_end = (void *)vmlinux_relocs_64_start + size; - memmove(vmlinux_relocs_64_start, __vmlinux_relocs_64_start, size); -} - -static void free_relocs(void) -{ - physmem_free(RR_RELOC); -} - -static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, unsigned long offset) +static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, + unsigned long offset, unsigned long phys_offset) { int *reloc; long loc; /* Adjust R_390_64 relocations */ - for (reloc = vmlinux_relocs_64_start; reloc < vmlinux_relocs_64_end; reloc++) { - loc = (long)*reloc + offset; + for (reloc = (int *)__vmlinux_relocs_64_start; reloc < (int *)__vmlinux_relocs_64_end; reloc++) { + loc = (long)*reloc + phys_offset; if (loc < min_addr || loc > max_addr) error("64-bit relocation outside of kernel!\n"); - *(u64 *)loc += offset; + *(u64 *)loc += offset - __START_KERNEL; } } @@ -217,9 +175,8 @@ static void kaslr_adjust_got(unsigned long offset) * reason. Adjust the GOT entries. */ for (entry = (u64 *)vmlinux.got_start; entry < (u64 *)vmlinux.got_end; entry++) - *entry += offset; + *entry += offset - __START_KERNEL; } -#endif /* * Merge information from several sources into a single ident_map_size value. @@ -261,9 +218,26 @@ static void setup_ident_map_size(unsigned long max_physmem_end) #endif } -static unsigned long setup_kernel_memory_layout(void) +#define FIXMAP_SIZE round_up(MEMCPY_REAL_SIZE + ABS_LOWCORE_MAP_SIZE, sizeof(struct lowcore)) + +static unsigned long get_vmem_size(unsigned long identity_size, + unsigned long vmemmap_size, + unsigned long vmalloc_size, + unsigned long rte_size) +{ + unsigned long max_mappable, vsize; + + max_mappable = max(identity_size, MAX_DCSS_ADDR); + vsize = round_up(SZ_2G + max_mappable, rte_size) + + round_up(vmemmap_size, rte_size) + + FIXMAP_SIZE + MODULES_LEN + KASLR_LEN; + return size_add(vsize, vmalloc_size); +} + +static unsigned long setup_kernel_memory_layout(unsigned long kernel_size) { unsigned long vmemmap_start; + unsigned long kernel_start; unsigned long asce_limit; unsigned long rte_size; unsigned long pages; @@ -275,12 +249,19 @@ static unsigned long setup_kernel_memory_layout(void) vmemmap_size = SECTION_ALIGN_UP(pages) * sizeof(struct page); /* choose kernel address space layout: 4 or 3 levels. */ - vsize = round_up(ident_map_size, _REGION3_SIZE) + vmemmap_size + - MODULES_LEN + MEMCPY_REAL_SIZE + ABS_LOWCORE_MAP_SIZE; - vsize = size_add(vsize, vmalloc_size); - if (IS_ENABLED(CONFIG_KASAN) || (vsize > _REGION2_SIZE)) { + BUILD_BUG_ON(!IS_ALIGNED(__START_KERNEL, 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); + if (IS_ENABLED(CONFIG_KASAN) || __NO_KASLR_END_KERNEL > _REGION2_SIZE || + (vsize > _REGION2_SIZE && kaslr_enabled())) { asce_limit = _REGION1_SIZE; - rte_size = _REGION2_SIZE; + if (__NO_KASLR_END_KERNEL > _REGION2_SIZE) { + rte_size = _REGION2_SIZE; + vsize = get_vmem_size(ident_map_size, vmemmap_size, vmalloc_size, _REGION2_SIZE); + } else { + rte_size = _REGION3_SIZE; + } } else { asce_limit = _REGION2_SIZE; rte_size = _REGION3_SIZE; @@ -290,38 +271,67 @@ static unsigned long setup_kernel_memory_layout(void) * Forcing modules and vmalloc area under the ultravisor * secure storage limit, so that any vmalloc allocation * we do could be used to back secure guest storage. + * + * Assume the secure storage limit always exceeds _REGION2_SIZE, + * otherwise asce_limit and rte_size would have been adjusted. */ vmax = adjust_to_uv_max(asce_limit); #ifdef CONFIG_KASAN + BUILD_BUG_ON(__NO_KASLR_END_KERNEL > KASAN_SHADOW_START); /* force vmalloc and modules below kasan shadow */ vmax = min(vmax, KASAN_SHADOW_START); #endif - __memcpy_real_area = round_down(vmax - MEMCPY_REAL_SIZE, PAGE_SIZE); - __abs_lowcore = round_down(__memcpy_real_area - ABS_LOWCORE_MAP_SIZE, - sizeof(struct lowcore)); - MODULES_END = round_down(__abs_lowcore, _SEGMENT_SIZE); + vsize = min(vsize, vmax); + if (kaslr_enabled()) { + unsigned long kernel_end, kaslr_len, slots, pos; + + kaslr_len = max(KASLR_LEN, vmax - vsize); + slots = DIV_ROUND_UP(kaslr_len - kernel_size, THREAD_SIZE); + if (get_random(slots, &pos)) + pos = 0; + kernel_end = vmax - pos * THREAD_SIZE; + kernel_start = round_down(kernel_end - kernel_size, THREAD_SIZE); + } else if (vmax < __NO_KASLR_END_KERNEL || vsize > __NO_KASLR_END_KERNEL) { + kernel_start = round_down(vmax - kernel_size, THREAD_SIZE); + decompressor_printk("The kernel base address is forced to %lx\n", kernel_start); + } else { + kernel_start = __NO_KASLR_START_KERNEL; + } + __kaslr_offset = kernel_start; + + MODULES_END = round_down(kernel_start, _SEGMENT_SIZE); MODULES_VADDR = MODULES_END - MODULES_LEN; VMALLOC_END = MODULES_VADDR; /* allow vmalloc area to occupy up to about 1/2 of the rest virtual space left */ - vsize = round_down(VMALLOC_END / 2, _SEGMENT_SIZE); + vsize = (VMALLOC_END - FIXMAP_SIZE) / 2; + vsize = round_down(vsize, _SEGMENT_SIZE); vmalloc_size = min(vmalloc_size, vsize); VMALLOC_START = VMALLOC_END - vmalloc_size; + __memcpy_real_area = round_down(VMALLOC_START - MEMCPY_REAL_SIZE, PAGE_SIZE); + __abs_lowcore = round_down(__memcpy_real_area - ABS_LOWCORE_MAP_SIZE, + sizeof(struct lowcore)); + /* split remaining virtual space between 1:1 mapping & vmemmap array */ - pages = VMALLOC_START / (PAGE_SIZE + sizeof(struct page)); + pages = __abs_lowcore / (PAGE_SIZE + sizeof(struct page)); pages = SECTION_ALIGN_UP(pages); /* keep vmemmap_start aligned to a top level region table entry */ - vmemmap_start = round_down(VMALLOC_START - pages * sizeof(struct page), rte_size); - vmemmap_start = min(vmemmap_start, 1UL << MAX_PHYSMEM_BITS); - /* maximum mappable address as seen by arch_get_mappable_range() */ - max_mappable = vmemmap_start; + vmemmap_start = round_down(__abs_lowcore - pages * sizeof(struct page), rte_size); /* make sure identity map doesn't overlay with vmemmap */ ident_map_size = min(ident_map_size, vmemmap_start); vmemmap_size = SECTION_ALIGN_UP(ident_map_size / PAGE_SIZE) * sizeof(struct page); - /* make sure vmemmap doesn't overlay with vmalloc area */ - VMALLOC_START = max(vmemmap_start + vmemmap_size, VMALLOC_START); + /* make sure vmemmap doesn't overlay with absolute lowcore area */ + if (vmemmap_start + vmemmap_size > __abs_lowcore) { + vmemmap_size = SECTION_ALIGN_DOWN(ident_map_size / PAGE_SIZE) * sizeof(struct page); + ident_map_size = vmemmap_size / sizeof(struct page) * PAGE_SIZE; + } vmemmap = (struct page *)vmemmap_start; + /* maximum address for which linear mapping could be created (DCSS, memory) */ + 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); return asce_limit; } @@ -329,9 +339,9 @@ static unsigned long setup_kernel_memory_layout(void) /* * This function clears the BSS section of the decompressed Linux kernel and NOT the decompressor's. */ -static void clear_bss_section(unsigned long vmlinux_lma) +static void clear_bss_section(unsigned long kernel_start) { - memset((void *)vmlinux_lma + vmlinux.image_size, 0, vmlinux.bss_size); + memset((void *)kernel_start + vmlinux.image_size, 0, vmlinux.bss_size); } /* @@ -348,19 +358,12 @@ static void setup_vmalloc_size(void) vmalloc_size = max(size, vmalloc_size); } -static void kaslr_adjust_vmlinux_info(unsigned long offset) +static void kaslr_adjust_vmlinux_info(long offset) { - *(unsigned long *)(&vmlinux.entry) += offset; vmlinux.bootdata_off += offset; vmlinux.bootdata_preserved_off += offset; -#ifdef CONFIG_PIE_BUILD - vmlinux.rela_dyn_start += offset; - vmlinux.rela_dyn_end += offset; - vmlinux.dynsym_start += offset; -#else vmlinux.got_start += offset; vmlinux.got_end += offset; -#endif vmlinux.init_mm_off += offset; vmlinux.swapper_pg_dir_off += offset; vmlinux.invalid_pg_dir_off += offset; @@ -373,23 +376,30 @@ static void kaslr_adjust_vmlinux_info(unsigned 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 max_physmem_end; - unsigned long vmlinux_lma = 0; + unsigned long kernel_size = vmlinux.image_size + vmlinux.bss_size; + unsigned long nokaslr_offset_phys = mem_safe_offset(); unsigned long amode31_lma = 0; + unsigned long max_physmem_end; unsigned long asce_limit; unsigned long safe_addr; - void *img; psw_t psw; + fixup_vmlinux_info(); setup_lpp(); - safe_addr = mem_safe_offset(); + safe_addr = PAGE_ALIGN(nokaslr_offset_phys + kernel_size); /* - * Reserve decompressor memory together with decompression heap, buffer and - * memory which might be occupied by uncompressed kernel at default 1Mb - * position (if KASLR is off or failed). + * Reserve decompressor memory together with decompression heap, + * buffer and memory which might be occupied by uncompressed kernel + * (if KASLR is off or failed). */ physmem_reserve(RR_DECOMPRESSOR, 0, safe_addr); if (IS_ENABLED(CONFIG_BLK_DEV_INITRD) && parmarea.initrd_size) @@ -409,40 +419,39 @@ void startup_kernel(void) max_physmem_end = detect_max_physmem_end(); setup_ident_map_size(max_physmem_end); setup_vmalloc_size(); - asce_limit = setup_kernel_memory_layout(); + asce_limit = setup_kernel_memory_layout(kernel_size); /* got final ident_map_size, physmem allocations could be performed now */ physmem_set_usable_limit(ident_map_size); detect_physmem_online_ranges(max_physmem_end); save_ipl_cert_comp_list(); rescue_initrd(safe_addr, ident_map_size); - rescue_relocs(); - if (kaslr_enabled()) { - vmlinux_lma = randomize_within_range(vmlinux.image_size + vmlinux.bss_size, - THREAD_SIZE, vmlinux.default_lma, - ident_map_size); - if (vmlinux_lma) { - __kaslr_offset = vmlinux_lma - vmlinux.default_lma; - kaslr_adjust_vmlinux_info(__kaslr_offset); - } - } - vmlinux_lma = vmlinux_lma ?: vmlinux.default_lma; - physmem_reserve(RR_VMLINUX, vmlinux_lma, vmlinux.image_size + vmlinux.bss_size); - - if (!IS_ENABLED(CONFIG_KERNEL_UNCOMPRESSED)) { - img = decompress_kernel(); - memmove((void *)vmlinux_lma, img, vmlinux.image_size); - } else if (__kaslr_offset) { - img = (void *)vmlinux.default_lma; - memmove((void *)vmlinux_lma, img, vmlinux.image_size); - memset(img, 0, vmlinux.image_size); - } + if (kaslr_enabled()) + __kaslr_offset_phys = randomize_within_range(kernel_size, THREAD_SIZE, 0, ident_map_size); + if (!__kaslr_offset_phys) + __kaslr_offset_phys = nokaslr_offset_phys; + kaslr_adjust_vmlinux_info(__kaslr_offset_phys); + physmem_reserve(RR_VMLINUX, __kaslr_offset_phys, kernel_size); + deploy_kernel((void *)__kaslr_offset_phys); /* vmlinux decompression is done, shrink reserved low memory */ physmem_reserve(RR_DECOMPRESSOR, 0, (unsigned long)_decompressor_end); + + /* + * In case KASLR is enabled the randomized location of .amode31 + * section might overlap with .vmlinux.relocs section. To avoid that + * the below randomize_within_range() could have been called with + * __vmlinux_relocs_64_end as the lower range address. However, + * .amode31 section is written to by the decompressed kernel - at + * that time the contents of .vmlinux.relocs is not needed anymore. + * Conversly, .vmlinux.relocs is read only by the decompressor, even + * 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); - amode31_lma = amode31_lma ?: vmlinux.default_lma - vmlinux.amode31_size; + if (!amode31_lma) + amode31_lma = __kaslr_offset_phys - vmlinux.amode31_size; physmem_reserve(RR_AMODE31, amode31_lma, vmlinux.amode31_size); /* @@ -458,23 +467,23 @@ void startup_kernel(void) * - copy_bootdata() must follow setup_vmem() to propagate changes * to bootdata made by setup_vmem() */ - clear_bss_section(vmlinux_lma); - kaslr_adjust_relocs(vmlinux_lma, vmlinux_lma + vmlinux.image_size, __kaslr_offset); + clear_bss_section(__kaslr_offset_phys); + kaslr_adjust_relocs(__kaslr_offset_phys, __kaslr_offset_phys + vmlinux.image_size, + __kaslr_offset, __kaslr_offset_phys); kaslr_adjust_got(__kaslr_offset); - free_relocs(); - setup_vmem(asce_limit); + setup_vmem(__kaslr_offset, __kaslr_offset + kernel_size, asce_limit); copy_bootdata(); /* * 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 ? __kaslr_offset | 0x1UL : 0; + S390_lowcore.vmcore_info = __kaslr_offset_phys ? __kaslr_offset_phys | 0x1UL : 0; /* * Jump to the decompressed kernel entry point and switch DAT mode on. */ - psw.addr = vmlinux.entry; + psw.addr = __kaslr_offset + vmlinux.entry; psw.mask = PSW_KERNEL_BITS; __load_psw(psw); } diff --git a/arch/s390/boot/vmem.c b/arch/s390/boot/vmem.c index 09b10bb6e4d0..96d48b7112d4 100644 --- a/arch/s390/boot/vmem.c +++ b/arch/s390/boot/vmem.c @@ -27,6 +27,8 @@ enum populate_mode { POPULATE_NONE, POPULATE_DIRECT, POPULATE_ABS_LOWCORE, + POPULATE_IDENTITY, + POPULATE_KERNEL, #ifdef CONFIG_KASAN POPULATE_KASAN_MAP_SHADOW, POPULATE_KASAN_ZERO_SHADOW, @@ -54,7 +56,7 @@ static inline void kasan_populate(unsigned long start, unsigned long end, enum p pgtable_populate(start, end, mode); } -static void kasan_populate_shadow(void) +static void kasan_populate_shadow(unsigned long kernel_start, unsigned long kernel_end) { pmd_t pmd_z = __pmd(__pa(kasan_early_shadow_pte) | _SEGMENT_ENTRY); pud_t pud_z = __pud(__pa(kasan_early_shadow_pmd) | _REGION3_ENTRY); @@ -76,44 +78,20 @@ static void kasan_populate_shadow(void) __arch_set_page_dat(kasan_early_shadow_pmd, 1UL << CRST_ALLOC_ORDER); __arch_set_page_dat(kasan_early_shadow_pte, 1); - /* - * Current memory layout: - * +- 0 -------------+ +- shadow start -+ - * |1:1 ident mapping| /|1/8 of ident map| - * | | / | | - * +-end of ident map+ / +----------------+ - * | ... gap ... | / | kasan | - * | | / | zero page | - * +- vmalloc area -+ / | mapping | - * | vmalloc_size | / | (untracked) | - * +- modules vaddr -+ / +----------------+ - * | 2Gb |/ | unmapped | allocated per module - * +- shadow start -+ +----------------+ - * | 1/8 addr space | | zero pg mapping| (untracked) - * +- shadow end ----+---------+- shadow end ---+ - * - * Current memory layout (KASAN_VMALLOC): - * +- 0 -------------+ +- shadow start -+ - * |1:1 ident mapping| /|1/8 of ident map| - * | | / | | - * +-end of ident map+ / +----------------+ - * | ... gap ... | / | kasan zero page| (untracked) - * | | / | mapping | - * +- vmalloc area -+ / +----------------+ - * | vmalloc_size | / |shallow populate| - * +- modules vaddr -+ / +----------------+ - * | 2Gb |/ |shallow populate| - * +- shadow start -+ +----------------+ - * | 1/8 addr space | | zero pg mapping| (untracked) - * +- shadow end ----+---------+- shadow end ---+ - */ - for_each_physmem_usable_range(i, &start, &end) { - kasan_populate(start, end, POPULATE_KASAN_MAP_SHADOW); - if (memgap_start && physmem_info.info_source == MEM_DETECT_DIAG260) - kasan_populate(memgap_start, start, POPULATE_KASAN_ZERO_SHADOW); + kasan_populate((unsigned long)__identity_va(start), + (unsigned long)__identity_va(end), + POPULATE_KASAN_MAP_SHADOW); + if (memgap_start && physmem_info.info_source == MEM_DETECT_DIAG260) { + kasan_populate((unsigned long)__identity_va(memgap_start), + (unsigned long)__identity_va(start), + POPULATE_KASAN_ZERO_SHADOW); + } memgap_start = end; } + kasan_populate(kernel_start, 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)) { untracked_end = VMALLOC_START; /* shallowly populate kasan shadow for vmalloc and modules */ @@ -122,8 +100,9 @@ static void kasan_populate_shadow(void) untracked_end = MODULES_VADDR; } /* populate kasan shadow for untracked memory */ - kasan_populate(ident_map_size, untracked_end, POPULATE_KASAN_ZERO_SHADOW); - kasan_populate(MODULES_END, _REGION1_SIZE, POPULATE_KASAN_ZERO_SHADOW); + kasan_populate((unsigned long)__identity_va(ident_map_size), untracked_end, + POPULATE_KASAN_ZERO_SHADOW); + kasan_populate(kernel_end, _REGION1_SIZE, POPULATE_KASAN_ZERO_SHADOW); } static bool kasan_pgd_populate_zero_shadow(pgd_t *pgd, unsigned long addr, @@ -180,7 +159,9 @@ static bool kasan_pte_populate_zero_shadow(pte_t *pte, enum populate_mode mode) } #else -static inline void kasan_populate_shadow(void) {} +static inline void kasan_populate_shadow(unsigned long kernel_start, unsigned long kernel_end) +{ +} static inline bool kasan_pgd_populate_zero_shadow(pgd_t *pgd, unsigned long addr, unsigned long end, enum populate_mode mode) @@ -263,6 +244,10 @@ static unsigned long _pa(unsigned long addr, unsigned long size, enum populate_m return addr; case POPULATE_ABS_LOWCORE: return __abs_lowcore_pa(addr); + case POPULATE_KERNEL: + return __kernel_pa(addr); + case POPULATE_IDENTITY: + return __identity_pa(addr); #ifdef CONFIG_KASAN case POPULATE_KASAN_MAP_SHADOW: addr = physmem_alloc_top_down(RR_VMEM, size, size); @@ -274,15 +259,22 @@ static unsigned long _pa(unsigned long addr, unsigned long size, enum populate_m } } -static bool can_large_pud(pud_t *pu_dir, unsigned long addr, unsigned long end) +static bool large_allowed(enum populate_mode mode) +{ + return (mode == POPULATE_DIRECT) || (mode == POPULATE_IDENTITY); +} + +static bool can_large_pud(pud_t *pu_dir, unsigned long addr, unsigned long end, + enum populate_mode mode) { - return machine.has_edat2 && + return machine.has_edat2 && large_allowed(mode) && IS_ALIGNED(addr, PUD_SIZE) && (end - addr) >= PUD_SIZE; } -static bool can_large_pmd(pmd_t *pm_dir, unsigned long addr, unsigned long end) +static bool can_large_pmd(pmd_t *pm_dir, unsigned long addr, unsigned long end, + enum populate_mode mode) { - return machine.has_edat1 && + return machine.has_edat1 && large_allowed(mode) && IS_ALIGNED(addr, PMD_SIZE) && (end - addr) >= PMD_SIZE; } @@ -322,7 +314,7 @@ static void pgtable_pmd_populate(pud_t *pud, unsigned long addr, unsigned long e if (pmd_none(*pmd)) { if (kasan_pmd_populate_zero_shadow(pmd, addr, next, mode)) continue; - if (can_large_pmd(pmd, addr, next)) { + if (can_large_pmd(pmd, addr, next, mode)) { entry = __pmd(_pa(addr, _SEGMENT_SIZE, mode)); entry = set_pmd_bit(entry, SEGMENT_KERNEL); if (!machine.has_nx) @@ -355,7 +347,7 @@ static void pgtable_pud_populate(p4d_t *p4d, unsigned long addr, unsigned long e if (pud_none(*pud)) { if (kasan_pud_populate_zero_shadow(pud, addr, next, mode)) continue; - if (can_large_pud(pud, addr, next)) { + if (can_large_pud(pud, addr, next, mode)) { entry = __pud(_pa(addr, _REGION3_SIZE, mode)); entry = set_pud_bit(entry, REGION3_KERNEL); if (!machine.has_nx) @@ -418,11 +410,12 @@ static void pgtable_populate(unsigned long addr, unsigned long end, enum populat } } -void setup_vmem(unsigned long asce_limit) +void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned long asce_limit) { unsigned long start, end; unsigned long asce_type; unsigned long asce_bits; + pgd_t *init_mm_pgd; int i; /* @@ -433,6 +426,15 @@ void setup_vmem(unsigned long asce_limit) for_each_physmem_online_range(i, &start, &end) __arch_set_page_nodat((void *)start, (end - start) >> PAGE_SHIFT); + /* + * init_mm->pgd contains virtual address of swapper_pg_dir. + * It is unusable at this stage since DAT is yet off. Swap + * it for physical address of swapper_pg_dir and restore + * the virtual address after all page tables are created. + */ + init_mm_pgd = init_mm.pgd; + init_mm.pgd = (pgd_t *)swapper_pg_dir; + if (asce_limit == _REGION1_SIZE) { asce_type = _REGION2_ENTRY_EMPTY; asce_bits = _ASCE_TYPE_REGION2 | _ASCE_TABLE_LENGTH; @@ -453,15 +455,20 @@ void setup_vmem(unsigned long asce_limit) * the lowcore and create the identity mapping only afterwards. */ pgtable_populate(0, sizeof(struct lowcore), POPULATE_DIRECT); - for_each_physmem_usable_range(i, &start, &end) - pgtable_populate(start, end, POPULATE_DIRECT); + 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); + pgtable_populate(AMODE31_START, AMODE31_END, POPULATE_DIRECT); pgtable_populate(__abs_lowcore, __abs_lowcore + sizeof(struct lowcore), POPULATE_ABS_LOWCORE); pgtable_populate(__memcpy_real_area, __memcpy_real_area + PAGE_SIZE, POPULATE_NONE); - memcpy_real_ptep = __virt_to_kpte(__memcpy_real_area); + memcpy_real_ptep = __identity_va(__virt_to_kpte(__memcpy_real_area)); - kasan_populate_shadow(); + kasan_populate_shadow(kernel_start, kernel_end); S390_lowcore.kernel_asce.val = swapper_pg_dir | asce_bits; S390_lowcore.user_asce = s390_invalid_asce; @@ -471,4 +478,5 @@ void setup_vmem(unsigned long asce_limit) local_ctl_load(13, &S390_lowcore.kernel_asce); init_mm.context.asce = S390_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 3d7ea585ab99..1fe5a1d3ff60 100644 --- a/arch/s390/boot/vmlinux.lds.S +++ b/arch/s390/boot/vmlinux.lds.S @@ -99,8 +99,16 @@ SECTIONS _decompressor_end = .; + . = ALIGN(4); + .vmlinux.relocs : { + __vmlinux_relocs_64_start = .; + *(.vmlinux.relocs_64) + __vmlinux_relocs_64_end = .; + } + #ifdef CONFIG_KERNEL_UNCOMPRESSED - . = 0x100000; + . = ALIGN(PAGE_SIZE); + . += AMODE31_SIZE; /* .amode31 section */ #else . = ALIGN(8); #endif @@ -110,24 +118,6 @@ SECTIONS _compressed_end = .; } -#ifndef CONFIG_PIE_BUILD - /* - * When the kernel is built with CONFIG_KERNEL_UNCOMPRESSED, the entire - * uncompressed vmlinux.bin is positioned in the bzImage decompressor - * image at the default kernel LMA of 0x100000, enabling it to be - * executed in-place. However, the size of .vmlinux.relocs could be - * large enough to cause an overlap with the uncompressed kernel at the - * address 0x100000. To address this issue, .vmlinux.relocs is - * positioned after the .rodata.compressed. - */ - . = ALIGN(4); - .vmlinux.relocs : { - __vmlinux_relocs_64_start = .; - *(.vmlinux.relocs_64) - __vmlinux_relocs_64_end = .; - } -#endif - #define SB_TRAILER_SIZE 32 /* Trailer needed for Secure Boot */ . += SB_TRAILER_SIZE; /* make sure .sb.trailer does not overwrite the previous section */ diff --git a/arch/s390/include/asm/ap.h b/arch/s390/include/asm/ap.h index 43ac4a64f49b..395b02d6a133 100644 --- a/arch/s390/include/asm/ap.h +++ b/arch/s390/include/asm/ap.h @@ -223,13 +223,18 @@ static inline struct ap_queue_status ap_zapq(ap_qid_t qid, int fbit) * config info as returned by the ap_qci() function. */ struct ap_config_info { - unsigned int apsc : 1; /* S bit */ - unsigned int apxa : 1; /* N bit */ - unsigned int qact : 1; /* C bit */ - unsigned int rc8a : 1; /* R bit */ - unsigned int : 4; - unsigned int apsb : 1; /* B bit */ - unsigned int : 23; + union { + unsigned int flags; + struct { + unsigned int apsc : 1; /* S bit */ + unsigned int apxa : 1; /* N bit */ + unsigned int qact : 1; /* C bit */ + unsigned int rc8a : 1; /* R bit */ + unsigned int : 4; + unsigned int apsb : 1; /* B bit */ + unsigned int : 23; + }; + }; unsigned char na; /* max # of APs - 1 */ unsigned char nd; /* max # of Domains - 1 */ unsigned char _reserved0[10]; @@ -544,15 +549,4 @@ static inline struct ap_queue_status ap_dqap(ap_qid_t qid, return reg1.status; } -/* - * Interface to tell the AP bus code that a configuration - * change has happened. The bus code should at least do - * an ap bus resource rescan. - */ -#if IS_ENABLED(CONFIG_ZCRYPT) -void ap_bus_cfg_chg(void); -#else -static inline void ap_bus_cfg_chg(void){} -#endif - #endif /* _ASM_S390_AP_H_ */ diff --git a/arch/s390/include/asm/asm-prototypes.h b/arch/s390/include/asm/asm-prototypes.h index 56096ae26f29..f662eb4b9246 100644 --- a/arch/s390/include/asm/asm-prototypes.h +++ b/arch/s390/include/asm/asm-prototypes.h @@ -4,6 +4,7 @@ #include <linux/kvm_host.h> #include <linux/ftrace.h> #include <asm/fpu.h> +#include <asm/nospec-branch.h> #include <asm-generic/asm-prototypes.h> __int128_t __ashlti3(__int128_t a, int b); diff --git a/arch/s390/include/asm/chsc.h b/arch/s390/include/asm/chsc.h index bb48ea380c0d..bb78159d8042 100644 --- a/arch/s390/include/asm/chsc.h +++ b/arch/s390/include/asm/chsc.h @@ -11,6 +11,9 @@ #include <uapi/asm/chsc.h> +/* struct from linux/notifier.h */ +struct notifier_block; + /** * Operation codes for CHSC PNSO: * PNSO_OC_NET_BRIDGE_INFO - only addresses that are visible to a bridgeport @@ -66,4 +69,16 @@ struct chsc_pnso_area { struct chsc_pnso_naid_l2 entries[]; } __packed __aligned(PAGE_SIZE); +/* + * notifier interface - registered notifiers gets called on + * the following events: + * - ap config changed (CHSC_NOTIFY_AP_CFG) + */ +enum chsc_notify_type { + CHSC_NOTIFY_AP_CFG = 3, +}; + +int chsc_notifier_register(struct notifier_block *nb); +int chsc_notifier_unregister(struct notifier_block *nb); + #endif /* _ASM_S390_CHSC_H */ diff --git a/arch/s390/include/asm/extmem.h b/arch/s390/include/asm/extmem.h index 568fd81bb77b..e0a06060afdd 100644 --- a/arch/s390/include/asm/extmem.h +++ b/arch/s390/include/asm/extmem.h @@ -8,6 +8,13 @@ #define _ASM_S390X_DCSS_H #ifndef __ASSEMBLY__ +/* + * DCSS segment is defined as a contiguous range of pages using DEFSEG command. + * The range start and end is a page number with a value less than or equal to + * 0x7ffffff (see CP Commands and Utilities Reference). + */ +#define MAX_DCSS_ADDR (512UL * SZ_1G) + /* possible values for segment type as returned by segment_info */ #define SEG_TYPE_SW 0 #define SEG_TYPE_EW 1 diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h index 621f23d5ae30..77e479d44f1e 100644 --- a/arch/s390/include/asm/ftrace.h +++ b/arch/s390/include/asm/ftrace.h @@ -8,12 +8,8 @@ #ifndef __ASSEMBLY__ -#ifdef CONFIG_CC_IS_CLANG -/* https://llvm.org/pr41424 */ -#define ftrace_return_address(n) 0UL -#else -#define ftrace_return_address(n) __builtin_return_address(n) -#endif +unsigned long return_address(unsigned int n); +#define ftrace_return_address(n) return_address(n) void ftrace_caller(void); diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index 5cc46e0dde62..9725586f4259 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -146,7 +146,7 @@ int gmap_mprotect_notify(struct gmap *, unsigned long start, void gmap_sync_dirty_log_pmd(struct gmap *gmap, unsigned long dirty_bitmap[4], unsigned long gaddr, unsigned long vmaddr); -int gmap_mark_unmergeable(void); +int s390_disable_cow_sharing(void); void s390_unlist_old_asce(struct gmap *gmap); int s390_replace_asce(struct gmap *gmap); void s390_uv_destroy_pfns(unsigned long count, unsigned long *pfns); diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index bb1b4bef1878..4c2dc7abc285 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -32,6 +32,11 @@ typedef struct { unsigned int uses_skeys:1; /* The mmu context uses CMM. */ unsigned int uses_cmm:1; + /* + * The mmu context allows COW-sharing of memory pages (KSM, zeropage). + * Note that COW-sharing during fork() is currently always allowed. + */ + unsigned int allow_cow_sharing:1; /* The gmaps associated with this context are allowed to use huge pages. */ unsigned int allow_gmap_hpage_1m:1; } mm_context_t; diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index 929af18b0908..a7789a9f6218 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -35,6 +35,7 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.has_pgste = 0; mm->context.uses_skeys = 0; mm->context.uses_cmm = 0; + mm->context.allow_cow_sharing = 1; mm->context.allow_gmap_hpage_1m = 0; #endif switch (mm->context.asce_limit) { diff --git a/arch/s390/include/asm/nospec-branch.h b/arch/s390/include/asm/nospec-branch.h index 82725cf783c7..b9c1f3cae842 100644 --- a/arch/s390/include/asm/nospec-branch.h +++ b/arch/s390/include/asm/nospec-branch.h @@ -17,6 +17,26 @@ static inline bool nospec_uses_trampoline(void) return __is_defined(CC_USING_EXPOLINE) && !nospec_disable; } +#ifdef CONFIG_EXPOLINE_EXTERN + +void __s390_indirect_jump_r1(void); +void __s390_indirect_jump_r2(void); +void __s390_indirect_jump_r3(void); +void __s390_indirect_jump_r4(void); +void __s390_indirect_jump_r5(void); +void __s390_indirect_jump_r6(void); +void __s390_indirect_jump_r7(void); +void __s390_indirect_jump_r8(void); +void __s390_indirect_jump_r9(void); +void __s390_indirect_jump_r10(void); +void __s390_indirect_jump_r11(void); +void __s390_indirect_jump_r12(void); +void __s390_indirect_jump_r13(void); +void __s390_indirect_jump_r14(void); +void __s390_indirect_jump_r15(void); + +#endif + #endif /* __ASSEMBLY__ */ #endif /* _ASM_S390_EXPOLINE_H */ diff --git a/arch/s390/include/asm/nospec-insn.h b/arch/s390/include/asm/nospec-insn.h index 7a946c42ad13..cb15dd25bf21 100644 --- a/arch/s390/include/asm/nospec-insn.h +++ b/arch/s390/include/asm/nospec-insn.h @@ -16,24 +16,25 @@ */ .macro __THUNK_PROLOG_NAME name #ifdef CONFIG_EXPOLINE_EXTERN - .pushsection .text,"ax",@progbits - __ALIGN + SYM_CODE_START(\name) #else .pushsection .text.\name,"axG",@progbits,\name,comdat -#endif .globl \name .hidden \name .type \name,@function \name: CFI_STARTPROC +#endif .endm .macro __THUNK_EPILOG_NAME name - CFI_ENDPROC #ifdef CONFIG_EXPOLINE_EXTERN - .size \name, .-\name -#endif + SYM_CODE_END(\name) + EXPORT_SYMBOL(\name) +#else + CFI_ENDPROC .popsection +#endif .endm .macro __THUNK_PROLOG_BR r1 diff --git a/arch/s390/include/asm/os_info.h b/arch/s390/include/asm/os_info.h index a4d2e103f116..3ee9e8f5ceae 100644 --- a/arch/s390/include/asm/os_info.h +++ b/arch/s390/include/asm/os_info.h @@ -17,11 +17,25 @@ #define OS_INFO_VMCOREINFO 0 #define OS_INFO_REIPL_BLOCK 1 #define OS_INFO_FLAGS_ENTRY 2 +#define OS_INFO_RESERVED 3 +#define OS_INFO_IDENTITY_BASE 4 +#define OS_INFO_KASLR_OFFSET 5 +#define OS_INFO_KASLR_OFF_PHYS 6 +#define OS_INFO_VMEMMAP 7 +#define OS_INFO_AMODE31_START 8 +#define OS_INFO_AMODE31_END 9 +#define OS_INFO_IMAGE_START 10 +#define OS_INFO_IMAGE_END 11 +#define OS_INFO_IMAGE_PHYS 12 +#define OS_INFO_MAX 13 #define OS_INFO_FLAG_REIPL_CLEAR (1UL << 0) struct os_info_entry { - u64 addr; + union { + u64 addr; + u64 val; + }; u64 size; u32 csum; } __packed; @@ -33,17 +47,24 @@ struct os_info { u16 version_minor; u64 crashkernel_addr; u64 crashkernel_size; - struct os_info_entry entry[3]; - u8 reserved[4004]; + struct os_info_entry entry[OS_INFO_MAX]; + u8 reserved[3804]; } __packed; void os_info_init(void); -void os_info_entry_add(int nr, void *ptr, u64 len); +void os_info_entry_add_data(int nr, void *ptr, u64 len); +void os_info_entry_add_val(int nr, u64 val); void os_info_crashkernel_add(unsigned long base, unsigned long size); u32 os_info_csum(struct os_info *os_info); #ifdef CONFIG_CRASH_DUMP void *os_info_old_entry(int nr, unsigned long *size); +static inline unsigned long os_info_old_value(int nr) +{ + unsigned long size; + + return (unsigned long)os_info_old_entry(nr, &size); +} #else static inline void *os_info_old_entry(int nr, unsigned long *size) { diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h index 9381879f7ecf..224ff9d433ea 100644 --- a/arch/s390/include/asm/page.h +++ b/arch/s390/include/asm/page.h @@ -178,19 +178,52 @@ int arch_make_page_accessible(struct page *page); #define HAVE_ARCH_MAKE_PAGE_ACCESSIBLE #endif -#define __PAGE_OFFSET 0x0UL -#define PAGE_OFFSET 0x0UL +struct vm_layout { + unsigned long kaslr_offset; + unsigned long kaslr_offset_phys; + unsigned long identity_base; + unsigned long identity_size; +}; -#define __pa_nodebug(x) ((unsigned long)(x)) +extern struct vm_layout vm_layout; + +#define __kaslr_offset vm_layout.kaslr_offset +#define __kaslr_offset_phys vm_layout.kaslr_offset_phys +#define __identity_base vm_layout.identity_base +#define ident_map_size vm_layout.identity_size + +static inline unsigned long kaslr_offset(void) +{ + return __kaslr_offset; +} + +extern int __kaslr_enabled; +static inline int kaslr_enabled(void) +{ + if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) + return __kaslr_enabled; + return 0; +} + +#define __PAGE_OFFSET __identity_base +#define PAGE_OFFSET __PAGE_OFFSET #ifdef __DECOMPRESSOR +#define __pa_nodebug(x) ((unsigned long)(x)) #define __pa(x) __pa_nodebug(x) #define __pa32(x) __pa(x) #define __va(x) ((void *)(unsigned long)(x)) #else /* __DECOMPRESSOR */ +static inline unsigned long __pa_nodebug(unsigned long x) +{ + if (x < __kaslr_offset) + return x - __identity_base; + return x - __kaslr_offset + __kaslr_offset_phys; +} + #ifdef CONFIG_DEBUG_VIRTUAL unsigned long __phys_addr(unsigned long x, bool is_31bit); @@ -206,7 +239,7 @@ static inline unsigned long __phys_addr(unsigned long x, bool is_31bit) #define __pa(x) __phys_addr((unsigned long)(x), false) #define __pa32(x) __phys_addr((unsigned long)(x), true) -#define __va(x) ((void *)(unsigned long)(x)) +#define __va(x) ((void *)((unsigned long)(x) + __identity_base)) #endif /* __DECOMPRESSOR */ @@ -231,7 +264,7 @@ static inline unsigned long virt_to_pfn(const void *kaddr) #define virt_to_page(kaddr) pfn_to_page(virt_to_pfn(kaddr)) #define page_to_virt(page) pfn_to_virt(page_to_pfn(page)) -#define virt_addr_valid(kaddr) pfn_valid(phys_to_pfn(__pa_nodebug(kaddr))) +#define virt_addr_valid(kaddr) pfn_valid(phys_to_pfn(__pa_nodebug((unsigned long)(kaddr)))) #define VM_DATA_DEFAULT_FLAGS VM_DATA_FLAGS_NON_EXEC @@ -240,4 +273,11 @@ static inline unsigned long virt_to_pfn(const void *kaddr) #include <asm-generic/memory_model.h> #include <asm-generic/getorder.h> +#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) + #endif /* _S390_PAGE_H */ diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 60950e7a25f5..6f11d063d545 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -107,6 +107,12 @@ static inline int is_module_addr(void *addr) return 1; } +#ifdef CONFIG_RANDOMIZE_BASE +#define KASLR_LEN (1UL << 31) +#else +#define KASLR_LEN 0UL +#endif + /* * A 64 bit pagetable entry of S390 has following format: * | PFRA |0IPC| OS | @@ -566,10 +572,20 @@ static inline pud_t set_pud_bit(pud_t pud, pgprot_t prot) } /* - * In the case that a guest uses storage keys - * faults should no longer be backed by zero pages + * As soon as the guest uses storage keys or enables PV, we deduplicate all + * mapped shared zeropages and prevent new shared zeropages from getting + * mapped. */ -#define mm_forbids_zeropage mm_has_pgste +#define mm_forbids_zeropage mm_forbids_zeropage +static inline int mm_forbids_zeropage(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (!mm->context.allow_cow_sharing) + return 1; +#endif + return 0; +} + static inline int mm_uses_skeys(struct mm_struct *mm) { #ifdef CONFIG_PGSTE diff --git a/arch/s390/include/asm/physmem_info.h b/arch/s390/include/asm/physmem_info.h index e747b067f8db..f45cfc8bc233 100644 --- a/arch/s390/include/asm/physmem_info.h +++ b/arch/s390/include/asm/physmem_info.h @@ -22,7 +22,6 @@ enum reserved_range_type { RR_DECOMPRESSOR, RR_INITRD, RR_VMLINUX, - RR_RELOC, RR_AMODE31, RR_IPLREPORT, RR_CERT_COMP_LIST, @@ -170,4 +169,7 @@ static inline unsigned long get_physmem_reserved(enum reserved_range_type type, return *size; } +#define AMODE31_START (physmem_info.reserved[RR_AMODE31].start) +#define AMODE31_END (physmem_info.reserved[RR_AMODE31].end) + #endif diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index 03bcaa8effb2..32f70873e2b7 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -127,20 +127,6 @@ extern void (*_machine_restart)(char *command); extern void (*_machine_halt)(void); extern void (*_machine_power_off)(void); -extern unsigned long __kaslr_offset; -static inline unsigned long kaslr_offset(void) -{ - return __kaslr_offset; -} - -extern int __kaslr_enabled; -static inline int kaslr_enabled(void) -{ - if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) - return __kaslr_enabled; - return 0; -} - struct oldmem_data { unsigned long start; unsigned long size; diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index fa029d0dc28f..db2d9ba5a86d 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -11,6 +11,8 @@ CFLAGS_REMOVE_ftrace.o = $(CC_FLAGS_FTRACE) # Do not trace early setup code CFLAGS_REMOVE_early.o = $(CC_FLAGS_FTRACE) CFLAGS_REMOVE_rethook.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_stacktrace.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_unwind_bc.o = $(CC_FLAGS_FTRACE) endif diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c index d09ebb6f5262..9863ebe75019 100644 --- a/arch/s390/kernel/crash_dump.c +++ b/arch/s390/kernel/crash_dump.c @@ -465,7 +465,11 @@ static void *ehdr_init(Elf64_Ehdr *ehdr, int mem_chunk_cnt) ehdr->e_phoff = sizeof(Elf64_Ehdr); ehdr->e_ehsize = sizeof(Elf64_Ehdr); ehdr->e_phentsize = sizeof(Elf64_Phdr); - ehdr->e_phnum = mem_chunk_cnt + 1; + /* + * Number of memory chunk PT_LOAD program headers plus one kernel + * image PT_LOAD program header plus one PT_NOTE program header. + */ + ehdr->e_phnum = mem_chunk_cnt + 1 + 1; return ehdr + 1; } @@ -501,15 +505,16 @@ static int get_mem_chunk_cnt(void) */ static void loads_init(Elf64_Phdr *phdr) { + unsigned long old_identity_base = os_info_old_value(OS_INFO_IDENTITY_BASE); phys_addr_t start, end; u64 idx; for_each_physmem_range(idx, &oldmem_type, &start, &end) { - phdr->p_filesz = end - start; phdr->p_type = PT_LOAD; + phdr->p_vaddr = old_identity_base + start; phdr->p_offset = start; - phdr->p_vaddr = (unsigned long)__va(start); phdr->p_paddr = start; + phdr->p_filesz = end - start; phdr->p_memsz = end - start; phdr->p_flags = PF_R | PF_W | PF_X; phdr->p_align = PAGE_SIZE; @@ -518,6 +523,25 @@ static void loads_init(Elf64_Phdr *phdr) } /* + * Prepare PT_LOAD type program header for kernel image region + */ +static void text_init(Elf64_Phdr *phdr) +{ + unsigned long start_phys = os_info_old_value(OS_INFO_IMAGE_PHYS); + unsigned long start = os_info_old_value(OS_INFO_IMAGE_START); + unsigned long end = os_info_old_value(OS_INFO_IMAGE_END); + + phdr->p_type = PT_LOAD; + phdr->p_vaddr = start; + phdr->p_filesz = end - start; + phdr->p_memsz = end - start; + phdr->p_offset = start_phys; + phdr->p_paddr = start_phys; + phdr->p_flags = PF_R | PF_W | PF_X; + phdr->p_align = PAGE_SIZE; +} + +/* * Initialize notes (new kernel) */ static void *notes_init(Elf64_Phdr *phdr, void *ptr, u64 notes_offset) @@ -557,6 +581,8 @@ static size_t get_elfcorehdr_size(int mem_chunk_cnt) size += nt_vmcoreinfo_size(); /* nt_final */ size += sizeof(Elf64_Nhdr); + /* PT_LOAD type program header for kernel text region */ + size += sizeof(Elf64_Phdr); /* PT_LOADS */ size += mem_chunk_cnt * sizeof(Elf64_Phdr); @@ -568,7 +594,7 @@ static size_t get_elfcorehdr_size(int mem_chunk_cnt) */ int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size) { - Elf64_Phdr *phdr_notes, *phdr_loads; + Elf64_Phdr *phdr_notes, *phdr_loads, *phdr_text; size_t alloc_size; int mem_chunk_cnt; void *ptr, *hdr; @@ -606,14 +632,19 @@ int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size) /* Init program headers */ phdr_notes = ptr; ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr)); + phdr_text = ptr; + ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr)); phdr_loads = ptr; ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr) * mem_chunk_cnt); /* Init notes */ hdr_off = PTR_DIFF(ptr, hdr); ptr = notes_init(phdr_notes, ptr, ((unsigned long) hdr) + hdr_off); + /* Init kernel text program header */ + text_init(phdr_text); /* Init loads */ - hdr_off = PTR_DIFF(ptr, hdr); loads_init(phdr_loads); + /* Finalize program headers */ + hdr_off = PTR_DIFF(ptr, hdr); *addr = (unsigned long long) hdr; *size = (unsigned long long) hdr_off; BUG_ON(elfcorehdr_size > alloc_size); diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 1486350a4177..7dc54571f18e 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1209,8 +1209,8 @@ static struct attribute_group reipl_nss_attr_group = { void set_os_info_reipl_block(void) { - os_info_entry_add(OS_INFO_REIPL_BLOCK, reipl_block_actual, - reipl_block_actual->hdr.len); + os_info_entry_add_data(OS_INFO_REIPL_BLOCK, reipl_block_actual, + reipl_block_actual->hdr.len); } /* reipl type */ @@ -1940,7 +1940,7 @@ static void dump_reipl_run(struct shutdown_trigger *trigger) reipl_type == IPL_TYPE_NSS || reipl_type == IPL_TYPE_UNKNOWN) os_info_flags |= OS_INFO_FLAG_REIPL_CLEAR; - os_info_entry_add(OS_INFO_FLAGS_ENTRY, &os_info_flags, sizeof(os_info_flags)); + os_info_entry_add_data(OS_INFO_FLAGS_ENTRY, &os_info_flags, sizeof(os_info_flags)); csum = (__force unsigned int)cksm(reipl_block_actual, reipl_block_actual->hdr.len, 0); abs_lc = get_abs_lowcore(); abs_lc->ipib = __pa(reipl_block_actual); diff --git a/arch/s390/kernel/nospec-branch.c b/arch/s390/kernel/nospec-branch.c index d1b16d83e49a..9b8c24ebb008 100644 --- a/arch/s390/kernel/nospec-branch.c +++ b/arch/s390/kernel/nospec-branch.c @@ -114,10 +114,10 @@ static void __init_or_module __nospec_revert(s32 *start, s32 *end) type = BRASL_EXPOLINE; /* brasl instruction */ else continue; - thunk = instr + (*(int *)(instr + 2)) * 2; + thunk = instr + (long)(*(int *)(instr + 2)) * 2; if (thunk[0] == 0xc6 && thunk[1] == 0x00) /* exrl %r0,<target-br> */ - br = thunk + (*(int *)(thunk + 2)) * 2; + br = thunk + (long)(*(int *)(thunk + 2)) * 2; else continue; if (br[0] != 0x07 || (br[1] & 0xf0) != 0xf0) diff --git a/arch/s390/kernel/os_info.c b/arch/s390/kernel/os_info.c index a801e6bd5341..b695f980bbde 100644 --- a/arch/s390/kernel/os_info.c +++ b/arch/s390/kernel/os_info.c @@ -15,8 +15,10 @@ #include <asm/checksum.h> #include <asm/abs_lowcore.h> #include <asm/os_info.h> +#include <asm/physmem_info.h> #include <asm/maccess.h> #include <asm/asm-offsets.h> +#include <asm/ipl.h> /* * OS info structure has to be page aligned @@ -43,9 +45,9 @@ void os_info_crashkernel_add(unsigned long base, unsigned long size) } /* - * Add OS info entry and update checksum + * Add OS info data entry and update checksum */ -void os_info_entry_add(int nr, void *ptr, u64 size) +void os_info_entry_add_data(int nr, void *ptr, u64 size) { os_info.entry[nr].addr = __pa(ptr); os_info.entry[nr].size = size; @@ -54,15 +56,36 @@ void os_info_entry_add(int nr, void *ptr, u64 size) } /* + * Add OS info value entry and update checksum + */ +void os_info_entry_add_val(int nr, u64 value) +{ + os_info.entry[nr].val = value; + os_info.entry[nr].size = 0; + os_info.entry[nr].csum = 0; + os_info.csum = os_info_csum(&os_info); +} + +/* * Initialize OS info structure and set lowcore pointer */ void __init os_info_init(void) { struct lowcore *abs_lc; + BUILD_BUG_ON(sizeof(struct os_info) != PAGE_SIZE); os_info.version_major = OS_INFO_VERSION_MAJOR; os_info.version_minor = OS_INFO_VERSION_MINOR; os_info.magic = OS_INFO_MAGIC; + os_info_entry_add_val(OS_INFO_IDENTITY_BASE, __identity_base); + os_info_entry_add_val(OS_INFO_KASLR_OFFSET, kaslr_offset()); + os_info_entry_add_val(OS_INFO_KASLR_OFF_PHYS, __kaslr_offset_phys); + os_info_entry_add_val(OS_INFO_VMEMMAP, (unsigned long)vmemmap); + os_info_entry_add_val(OS_INFO_AMODE31_START, AMODE31_START); + os_info_entry_add_val(OS_INFO_AMODE31_END, AMODE31_END); + os_info_entry_add_val(OS_INFO_IMAGE_START, (unsigned long)_stext); + os_info_entry_add_val(OS_INFO_IMAGE_END, (unsigned long)_end); + os_info_entry_add_val(OS_INFO_IMAGE_PHYS, __pa_symbol(_stext)); os_info.csum = os_info_csum(&os_info); abs_lc = get_abs_lowcore(); abs_lc->os_info = __pa(&os_info); @@ -125,7 +148,7 @@ static void os_info_old_init(void) if (os_info_init) return; - if (!oldmem_data.start) + if (!oldmem_data.start && !is_ipl_type_dump()) goto fail; if (copy_oldmem_kernel(&addr, __LC_OS_INFO, sizeof(addr))) goto fail; diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c index 41ed6e0f0a2a..1434642e9cba 100644 --- a/arch/s390/kernel/perf_cpum_cf.c +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -428,7 +428,7 @@ static void cpum_cf_make_setsize(enum cpumf_ctr_set ctrset) case CPUMF_CTR_SET_CRYPTO: if (cpumf_ctr_info.csvn >= 1 && cpumf_ctr_info.csvn <= 5) ctrset_size = 16; - else if (cpumf_ctr_info.csvn == 6 || cpumf_ctr_info.csvn == 7) + else if (cpumf_ctr_info.csvn >= 6) ctrset_size = 20; break; case CPUMF_CTR_SET_EXT: diff --git a/arch/s390/kernel/perf_cpum_cf_events.c b/arch/s390/kernel/perf_cpum_cf_events.c index 0d64aafd158f..e4a6bfc91080 100644 --- a/arch/s390/kernel/perf_cpum_cf_events.c +++ b/arch/s390/kernel/perf_cpum_cf_events.c @@ -855,16 +855,11 @@ __init const struct attribute_group **cpumf_cf_event_group(void) } /* Determine version specific crypto set */ - switch (ci.csvn) { - case 1 ... 5: + csvn = none; + if (ci.csvn >= 1 && ci.csvn <= 5) csvn = cpumcf_svn_12345_pmu_event_attr; - break; - case 6 ... 7: + else if (ci.csvn >= 6) csvn = cpumcf_svn_67_pmu_event_attr; - break; - default: - csvn = none; - } /* Determine model-specific counter set(s) */ get_cpu_id(&cpu_id); diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 24ed33f044ec..cbd5290939df 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -146,10 +146,10 @@ static u32 __amode31_ref *__ctl_linkage_stack = __ctl_linkage_stack_amode31; static u32 __amode31_ref *__ctl_duct = __ctl_duct_amode31; unsigned long __bootdata_preserved(max_mappable); -unsigned long __bootdata(ident_map_size); struct physmem_info __bootdata(physmem_info); -unsigned long __bootdata_preserved(__kaslr_offset); +struct vm_layout __bootdata_preserved(vm_layout); +EXPORT_SYMBOL_GPL(vm_layout); int __bootdata_preserved(__kaslr_enabled); unsigned int __bootdata_preserved(zlib_dfltcc_support); EXPORT_SYMBOL(zlib_dfltcc_support); @@ -765,7 +765,7 @@ static void __init relocate_amode31_section(void) unsigned long amode31_size = __eamode31 - __samode31; long amode31_offset, *ptr; - amode31_offset = physmem_info.reserved[RR_AMODE31].start - (unsigned long)__samode31; + amode31_offset = AMODE31_START - (unsigned long)__samode31; pr_info("Relocating AMODE31 section of size 0x%08lx\n", amode31_size); /* Move original AMODE31 section to the new one */ diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c index 94f440e38303..7c294da45bf5 100644 --- a/arch/s390/kernel/stacktrace.c +++ b/arch/s390/kernel/stacktrace.c @@ -101,3 +101,22 @@ void arch_stack_walk_user(stack_trace_consume_fn consume_entry, void *cookie, } pagefault_enable(); } + +unsigned long return_address(unsigned int n) +{ + struct unwind_state state; + unsigned long addr; + + /* Increment to skip current stack entry */ + n++; + + unwind_for_each_frame(&state, NULL, NULL, 0) { + addr = unwind_get_return_address(&state); + if (!addr) + break; + if (!n--) + return addr; + } + return 0; +} +EXPORT_SYMBOL_GPL(return_address); diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index fc07bc39e698..265fea37e030 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -21,6 +21,7 @@ /* 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 /* @@ -181,36 +182,36 @@ int uv_convert_owned_from_secure(unsigned long paddr) } /* - * Calculate the expected ref_count for a page that would otherwise have no + * 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 - * page can not be a huge page for example. + * folio can not be a large folio, for example. */ -static int expected_page_refs(struct page *page) +static int expected_folio_refs(struct folio *folio) { int res; - res = page_mapcount(page); - if (PageSwapCache(page)) { + res = folio_mapcount(folio); + if (folio_test_swapcache(folio)) { res++; - } else if (page_mapping(page)) { + } else if (folio_mapping(folio)) { res++; - if (page_has_private(page)) + if (folio->private) res++; } return res; } -static int make_page_secure(struct page *page, struct uv_cb_header *uvcb) +static int make_folio_secure(struct folio *folio, struct uv_cb_header *uvcb) { int expected, cc = 0; - if (PageWriteback(page)) + if (folio_test_writeback(folio)) return -EAGAIN; - expected = expected_page_refs(page); - if (!page_ref_freeze(page, expected)) + expected = expected_folio_refs(folio); + if (!folio_ref_freeze(folio, expected)) return -EBUSY; - set_bit(PG_arch_1, &page->flags); + set_bit(PG_arch_1, &folio->flags); /* * If the UVC does not succeed or fail immediately, we don't want to * loop for long, or we might get stall notifications. @@ -220,9 +221,9 @@ static int make_page_secure(struct page *page, struct uv_cb_header *uvcb) * -EAGAIN and we let the callers deal with it. */ cc = __uv_call(0, (u64)uvcb); - page_ref_unfreeze(page, expected); + folio_ref_unfreeze(folio, expected); /* - * Return -ENXIO if the page was not mapped, -EINVAL for other errors. + * Return -ENXIO if the folio was not mapped, -EINVAL for other errors. * If busy or partially completed, return -EAGAIN. */ if (cc == UVC_CC_OK) @@ -277,7 +278,7 @@ int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb) bool local_drain = false; spinlock_t *ptelock; unsigned long uaddr; - struct page *page; + struct folio *folio; pte_t *ptep; int rc; @@ -306,15 +307,19 @@ again: if (!ptep) goto out; if (pte_present(*ptep) && !(pte_val(*ptep) & _PAGE_INVALID) && pte_write(*ptep)) { - page = pte_page(*ptep); + folio = page_folio(pte_page(*ptep)); + rc = -EINVAL; + if (folio_test_large(folio)) + goto unlock; rc = -EAGAIN; - if (trylock_page(page)) { + if (folio_trylock(folio)) { if (should_export_before_import(uvcb, gmap->mm)) - uv_convert_from_secure(page_to_phys(page)); - rc = make_page_secure(page, uvcb); - unlock_page(page); + uv_convert_from_secure(PFN_PHYS(folio_pfn(folio))); + rc = make_folio_secure(folio, uvcb); + folio_unlock(folio); } } +unlock: pte_unmap_unlock(ptep, ptelock); out: mmap_read_unlock(gmap->mm); @@ -324,10 +329,10 @@ out: * If we are here because the UVC returned busy or partial * completion, this is just a useless check, but it is safe. */ - wait_on_page_writeback(page); + folio_wait_writeback(folio); } else if (rc == -EBUSY) { /* - * If we have tried a local drain and the page refcount + * 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. @@ -338,7 +343,7 @@ out: return -EAGAIN; } /* - * We are here if the page refcount does not match the + * 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 diff --git a/arch/s390/kernel/vmcore_info.c b/arch/s390/kernel/vmcore_info.c index d296dfc22191..23f7d7619a99 100644 --- a/arch/s390/kernel/vmcore_info.c +++ b/arch/s390/kernel/vmcore_info.c @@ -14,7 +14,9 @@ void arch_crash_save_vmcoreinfo(void) VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS); vmcoreinfo_append_str("SAMODE31=%lx\n", (unsigned long)__samode31); vmcoreinfo_append_str("EAMODE31=%lx\n", (unsigned long)__eamode31); + vmcoreinfo_append_str("IDENTITYBASE=%lx\n", __identity_base); vmcoreinfo_append_str("KERNELOFFSET=%lx\n", kaslr_offset()); + vmcoreinfo_append_str("KERNELOFFPHYS=%lx\n", __kaslr_offset_phys); abs_lc = get_abs_lowcore(); abs_lc->vmcore_info = paddr_vmcoreinfo_note(); put_abs_lowcore(abs_lc); diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 48de296e8905..a1ce3925ec71 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -39,7 +39,7 @@ PHDRS { SECTIONS { - . = 0x100000; + . = __START_KERNEL; .text : { _stext = .; /* Start of text section */ _text = .; /* Text and read-only data */ @@ -183,7 +183,7 @@ SECTIONS .amode31.data : { *(.amode31.data) } - . = ALIGN(PAGE_SIZE); + . = _samode31 + AMODE31_SIZE; _eamode31 = .; /* early.c uses stsi, which requires page aligned data. */ @@ -192,31 +192,6 @@ SECTIONS PERCPU_SECTION(0x100) -#ifdef CONFIG_PIE_BUILD - .dynsym ALIGN(8) : { - __dynsym_start = .; - *(.dynsym) - __dynsym_end = .; - } - .rela.dyn ALIGN(8) : { - __rela_dyn_start = .; - *(.rela*) - __rela_dyn_end = .; - } - .dynamic ALIGN(8) : { - *(.dynamic) - } - .dynstr ALIGN(8) : { - *(.dynstr) - } -#endif - .hash ALIGN(8) : { - *(.hash) - } - .gnu.hash ALIGN(8) : { - *(.gnu.hash) - } - . = ALIGN(PAGE_SIZE); __init_end = .; /* freed after init ends here */ @@ -230,7 +205,6 @@ SECTIONS * it should match struct vmlinux_info */ .vmlinux.info 0 (INFO) : { - QUAD(_stext) /* default_lma */ QUAD(startup_continue) /* entry */ QUAD(__bss_start - _stext) /* image_size */ QUAD(__bss_stop - __bss_start) /* bss_size */ @@ -239,14 +213,8 @@ SECTIONS QUAD(__boot_data_preserved_start) /* bootdata_preserved_off */ QUAD(__boot_data_preserved_end - __boot_data_preserved_start) /* bootdata_preserved_size */ -#ifdef CONFIG_PIE_BUILD - QUAD(__dynsym_start) /* dynsym_start */ - QUAD(__rela_dyn_start) /* rela_dyn_start */ - QUAD(__rela_dyn_end) /* rela_dyn_end */ -#else QUAD(__got_start) /* got_start */ QUAD(__got_end) /* got_end */ -#endif QUAD(_eamode31 - _samode31) /* amode31_size */ QUAD(init_mm) QUAD(swapper_pg_dir) @@ -282,12 +250,10 @@ SECTIONS *(.plt) *(.plt.*) *(.iplt) *(.igot .igot.plt) } ASSERT(SIZEOF(.plt) == 0, "Unexpected run-time procedure linkages detected!") -#ifndef CONFIG_PIE_BUILD .rela.dyn : { *(.rela.*) *(.rela_*) } ASSERT(SIZEOF(.rela.dyn) == 0, "Unexpected run-time relocations (.rela) detected!") -#endif /* Sections to be discarded */ DISCARDS diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 7721eb522f43..82e9631cd9ef 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -2631,9 +2631,7 @@ static int kvm_s390_handle_pv(struct kvm *kvm, struct kvm_pv_cmd *cmd) if (r) break; - mmap_write_lock(current->mm); - r = gmap_mark_unmergeable(); - mmap_write_unlock(current->mm); + r = s390_disable_cow_sharing(); if (r) break; diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index b2c9f010f0fe..c9ecae830634 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -12,6 +12,7 @@ #include <linux/list.h> #include <linux/bitmap.h> #include <linux/sched/signal.h> +#include <linux/io.h> #include <asm/gmap.h> #include <asm/mmu_context.h> @@ -361,7 +362,7 @@ end: case -EACCES: return set_validity_icpt(scb_s, 0x003CU); } - scb_s->crycbd = ((__u32)(__u64) &vsie_page->crycb) | CRYCB_FORMAT2; + scb_s->crycbd = (u32)virt_to_phys(&vsie_page->crycb) | CRYCB_FORMAT2; return 0; } @@ -1005,7 +1006,7 @@ static int handle_stfle(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) if (read_guest_real(vcpu, fac, &vsie_page->fac, stfle_size() * sizeof(u64))) return set_validity_icpt(scb_s, 0x1090U); - scb_s->fac = (__u32)(__u64) &vsie_page->fac; + scb_s->fac = (u32)virt_to_phys(&vsie_page->fac); } return 0; } diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile index 90eac15ea62a..f43f897d3fc0 100644 --- a/arch/s390/lib/Makefile +++ b/arch/s390/lib/Makefile @@ -23,4 +23,4 @@ obj-$(CONFIG_S390_MODULES_SANITY_TEST_HELPERS) += test_modules_helpers.o lib-$(CONFIG_FUNCTION_ERROR_INJECTION) += error-inject.o -obj-$(CONFIG_EXPOLINE_EXTERN) += expoline/ +obj-$(CONFIG_EXPOLINE_EXTERN) += expoline.o diff --git a/arch/s390/lib/expoline/expoline.S b/arch/s390/lib/expoline.S index 92ed8409a7a4..92ed8409a7a4 100644 --- a/arch/s390/lib/expoline/expoline.S +++ b/arch/s390/lib/expoline.S diff --git a/arch/s390/lib/expoline/Makefile b/arch/s390/lib/expoline/Makefile deleted file mode 100644 index 854631d9cb03..000000000000 --- a/arch/s390/lib/expoline/Makefile +++ /dev/null @@ -1,3 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 - -obj-y += expoline.o diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 12d22a7fa32f..474a25ca5c48 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -2550,41 +2550,6 @@ static inline void thp_split_mm(struct mm_struct *mm) #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ /* - * Remove all empty zero pages from the mapping for lazy refaulting - * - This must be called after mm->context.has_pgste is set, to avoid - * future creation of zero pages - * - This must be called after THP was disabled. - * - * mm contracts with s390, that even if mm were to remove a page table, - * racing with the loop below and so causing pte_offset_map_lock() to fail, - * it will never insert a page table containing empty zero pages once - * mm_forbids_zeropage(mm) i.e. mm->context.has_pgste is set. - */ -static int __zap_zero_pages(pmd_t *pmd, unsigned long start, - unsigned long end, struct mm_walk *walk) -{ - unsigned long addr; - - for (addr = start; addr != end; addr += PAGE_SIZE) { - pte_t *ptep; - spinlock_t *ptl; - - ptep = pte_offset_map_lock(walk->mm, pmd, addr, &ptl); - if (!ptep) - break; - if (is_zero_pfn(pte_pfn(*ptep))) - ptep_xchg_direct(walk->mm, addr, ptep, __pte(_PAGE_INVALID)); - pte_unmap_unlock(ptep, ptl); - } - return 0; -} - -static const struct mm_walk_ops zap_zero_walk_ops = { - .pmd_entry = __zap_zero_pages, - .walk_lock = PGWALK_WRLOCK, -}; - -/* * switch on pgstes for its userspace process (for kvm) */ int s390_enable_sie(void) @@ -2601,22 +2566,142 @@ int s390_enable_sie(void) mm->context.has_pgste = 1; /* split thp mappings and disable thp for future mappings */ thp_split_mm(mm); - walk_page_range(mm, 0, TASK_SIZE, &zap_zero_walk_ops, NULL); mmap_write_unlock(mm); return 0; } EXPORT_SYMBOL_GPL(s390_enable_sie); -int gmap_mark_unmergeable(void) +static int find_zeropage_pte_entry(pte_t *pte, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + unsigned long *found_addr = walk->private; + + /* Return 1 of the page is a zeropage. */ + if (is_zero_pfn(pte_pfn(*pte))) { + /* + * Shared zeropage in e.g., a FS DAX mapping? We cannot do the + * right thing and likely don't care: FAULT_FLAG_UNSHARE + * currently only works in COW mappings, which is also where + * mm_forbids_zeropage() is checked. + */ + if (!is_cow_mapping(walk->vma->vm_flags)) + return -EFAULT; + + *found_addr = addr; + return 1; + } + return 0; +} + +static const struct mm_walk_ops find_zeropage_ops = { + .pte_entry = find_zeropage_pte_entry, + .walk_lock = PGWALK_WRLOCK, +}; + +/* + * Unshare all shared zeropages, replacing them by anonymous pages. Note that + * we cannot simply zap all shared zeropages, because this could later + * trigger unexpected userfaultfd missing events. + * + * This must be called after mm->context.allow_cow_sharing was + * set to 0, to avoid future mappings of shared zeropages. + * + * mm contracts with s390, that even if mm were to remove a page table, + * and racing with walk_page_range_vma() calling pte_offset_map_lock() + * would fail, it will never insert a page table containing empty zero + * pages once mm_forbids_zeropage(mm) i.e. + * mm->context.allow_cow_sharing is set to 0. + */ +static int __s390_unshare_zeropages(struct mm_struct *mm) +{ + struct vm_area_struct *vma; + VMA_ITERATOR(vmi, mm, 0); + unsigned long addr; + vm_fault_t fault; + int rc; + + for_each_vma(vmi, vma) { + /* + * We could only look at COW mappings, but it's more future + * proof to catch unexpected zeropages in other mappings and + * fail. + */ + if ((vma->vm_flags & VM_PFNMAP) || is_vm_hugetlb_page(vma)) + continue; + addr = vma->vm_start; + +retry: + rc = walk_page_range_vma(vma, addr, vma->vm_end, + &find_zeropage_ops, &addr); + if (rc < 0) + return rc; + else if (!rc) + continue; + + /* addr was updated by find_zeropage_pte_entry() */ + fault = handle_mm_fault(vma, addr, + FAULT_FLAG_UNSHARE | FAULT_FLAG_REMOTE, + NULL); + if (fault & VM_FAULT_OOM) + return -ENOMEM; + /* + * See break_ksm(): even after handle_mm_fault() returned 0, we + * must start the lookup from the current address, because + * handle_mm_fault() may back out if there's any difficulty. + * + * VM_FAULT_SIGBUS and VM_FAULT_SIGSEGV are unexpected but + * maybe they could trigger in the future on concurrent + * truncation. In that case, the shared zeropage would be gone + * and we can simply retry and make progress. + */ + cond_resched(); + goto retry; + } + + return 0; +} + +static int __s390_disable_cow_sharing(struct mm_struct *mm) { + int rc; + + if (!mm->context.allow_cow_sharing) + return 0; + + mm->context.allow_cow_sharing = 0; + + /* Replace all shared zeropages by anonymous pages. */ + rc = __s390_unshare_zeropages(mm); /* * Make sure to disable KSM (if enabled for the whole process or * individual VMAs). Note that nothing currently hinders user space * from re-enabling it. */ - return ksm_disable(current->mm); + if (!rc) + rc = ksm_disable(mm); + if (rc) + mm->context.allow_cow_sharing = 1; + return rc; +} + +/* + * Disable most COW-sharing of memory pages for the whole process: + * (1) Disable KSM and unmerge/unshare any KSM pages. + * (2) Disallow shared zeropages and unshare any zerpages that are mapped. + * + * Not that we currently don't bother with COW-shared pages that are shared + * with parent/child processes due to fork(). + */ +int s390_disable_cow_sharing(void) +{ + int rc; + + mmap_write_lock(current->mm); + rc = __s390_disable_cow_sharing(current->mm); + mmap_write_unlock(current->mm); + return rc; } -EXPORT_SYMBOL_GPL(gmap_mark_unmergeable); +EXPORT_SYMBOL_GPL(s390_disable_cow_sharing); /* * Enable storage key handling from now on and initialize the storage @@ -2685,7 +2770,7 @@ int s390_enable_skey(void) goto out_up; mm->context.uses_skeys = 1; - rc = gmap_mark_unmergeable(); + rc = __s390_disable_cow_sharing(mm); if (rc) { mm->context.uses_skeys = 0; goto out_up; diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index 85cddf904cb2..41c714e21292 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -13,7 +13,9 @@ #include <linux/slab.h> #include <linux/sort.h> #include <asm/page-states.h> +#include <asm/abs_lowcore.h> #include <asm/cacheflush.h> +#include <asm/maccess.h> #include <asm/nospec-branch.h> #include <asm/ctlreg.h> #include <asm/pgalloc.h> @@ -21,6 +23,7 @@ #include <asm/tlbflush.h> #include <asm/sections.h> #include <asm/set_memory.h> +#include <asm/physmem_info.h> static DEFINE_MUTEX(vmem_mutex); @@ -436,7 +439,7 @@ static int modify_pagetable(unsigned long start, unsigned long end, bool add, if (WARN_ON_ONCE(!PAGE_ALIGNED(start | end))) return -EINVAL; /* Don't mess with any tables not fully in 1:1 mapping & vmemmap area */ - if (WARN_ON_ONCE(end > VMALLOC_START)) + if (WARN_ON_ONCE(end > __abs_lowcore)) return -EINVAL; for (addr = start; addr < end; addr = next) { next = pgd_addr_end(addr, end); diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c index a0b872b74fe3..0f4f1e8fc480 100644 --- a/arch/s390/pci/pci_sysfs.c +++ b/arch/s390/pci/pci_sysfs.c @@ -172,7 +172,6 @@ static ssize_t uid_is_unique_show(struct device *dev, } static DEVICE_ATTR_RO(uid_is_unique); -#ifndef CONFIG_DMI /* analogous to smbios index */ static ssize_t index_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -202,7 +201,6 @@ static struct attribute_group zpci_ident_attr_group = { .attrs = zpci_ident_attrs, .is_visible = zpci_index_is_visible, }; -#endif static struct bin_attribute *zpci_bin_attrs[] = { &bin_attr_util_string, @@ -245,8 +243,6 @@ static struct attribute_group pfip_attr_group = { const struct attribute_group *zpci_attr_groups[] = { &zpci_attr_group, &pfip_attr_group, -#ifndef CONFIG_DMI &zpci_ident_attr_group, -#endif NULL, }; diff --git a/arch/s390/tools/relocs.c b/arch/s390/tools/relocs.c index 30a732c808f3..a74dbd5c9896 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); + add_reloc(&relocs64, offset - ehdr.e_entry); break; default: die("Unsupported relocation type: %d\n", r_type); |