From 515ab7c41306aad1f80a980e1936ef635c61570c Mon Sep 17 00:00:00 2001 From: Nadav Amit Date: Wed, 31 Jan 2018 13:19:12 -0800 Subject: x86/mm: Align TLB invalidation info The TLB invalidation info is allocated on the stack, which might cause it to be unaligned. Since this information may be transferred to different cores for TLB shootdown, this may cause an additional cache line to become shared. While the overhead is likely to be small, the fix is simple. We do not use __cacheline_aligned() since it also defines the section, which is inappropriate for stack variables. Signed-off-by: Nadav Amit Acked-by: Andy Lutomirski Cc: Dave Hansen Cc: Linus Torvalds Cc: Nadav Amit Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20180131211912.52064-1-namit@vmware.com Signed-off-by: Ingo Molnar --- arch/x86/mm/tlb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 8dcc0607f805..6550d37d0f06 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -613,7 +613,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, { int cpu; - struct flush_tlb_info info = { + struct flush_tlb_info info __aligned(SMP_CACHE_BYTES) = { .mm = mm, }; -- cgit v1.2.3-70-g09d2 From 1cd9c22fee3ac21db52a0997d08cf2f065d2c0c0 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 31 Jan 2018 16:54:02 +0300 Subject: x86/mm/encrypt: Move page table helpers into separate translation unit There are bunch of functions in mem_encrypt.c that operate on the identity mapping, which means they want virtual addresses to be equal to physical one, without PAGE_OFFSET shift. We also need to avoid paravirtualizaion call there. Getting this done is tricky. We cannot use usual page table helpers. It forces us to open-code a lot of things. It makes code ugly and hard to modify. We can get it work with the page table helpers, but it requires few preprocessor tricks. These tricks may have side effects for the rest of the file. Let's isolate such functions into own translation unit. Tested-by: Tom Lendacky Signed-off-by: Kirill A. Shutemov Reviewed-by: Tom Lendacky Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180131135404.40692-2-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mem_encrypt.h | 1 + arch/x86/mm/Makefile | 14 +- arch/x86/mm/mem_encrypt.c | 578 +---------------------------------- arch/x86/mm/mem_encrypt_identity.c | 597 +++++++++++++++++++++++++++++++++++++ 4 files changed, 608 insertions(+), 582 deletions(-) create mode 100644 arch/x86/mm/mem_encrypt_identity.c (limited to 'arch/x86/mm') diff --git a/arch/x86/include/asm/mem_encrypt.h b/arch/x86/include/asm/mem_encrypt.h index 22c5f3e6f820..8fe61ad21047 100644 --- a/arch/x86/include/asm/mem_encrypt.h +++ b/arch/x86/include/asm/mem_encrypt.h @@ -22,6 +22,7 @@ #ifdef CONFIG_AMD_MEM_ENCRYPT extern u64 sme_me_mask; +extern bool sev_enabled; void sme_encrypt_execute(unsigned long encrypted_kernel_vaddr, unsigned long decrypted_kernel_vaddr, diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 27e9e90a8d35..03c6c8561623 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -1,12 +1,15 @@ # SPDX-License-Identifier: GPL-2.0 -# Kernel does not boot with instrumentation of tlb.c and mem_encrypt.c -KCOV_INSTRUMENT_tlb.o := n -KCOV_INSTRUMENT_mem_encrypt.o := n +# Kernel does not boot with instrumentation of tlb.c and mem_encrypt*.c +KCOV_INSTRUMENT_tlb.o := n +KCOV_INSTRUMENT_mem_encrypt.o := n +KCOV_INSTRUMENT_mem_encrypt_identity.o := n -KASAN_SANITIZE_mem_encrypt.o := n +KASAN_SANITIZE_mem_encrypt.o := n +KASAN_SANITIZE_mem_encrypt_identity.o := n ifdef CONFIG_FUNCTION_TRACER -CFLAGS_REMOVE_mem_encrypt.o = -pg +CFLAGS_REMOVE_mem_encrypt.o = -pg +CFLAGS_REMOVE_mem_encrypt_identity.o = -pg endif obj-y := init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \ @@ -47,4 +50,5 @@ obj-$(CONFIG_RANDOMIZE_MEMORY) += kaslr.o obj-$(CONFIG_PAGE_TABLE_ISOLATION) += pti.o obj-$(CONFIG_AMD_MEM_ENCRYPT) += mem_encrypt.o +obj-$(CONFIG_AMD_MEM_ENCRYPT) += mem_encrypt_identity.o obj-$(CONFIG_AMD_MEM_ENCRYPT) += mem_encrypt_boot.o diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c index 1a53071e2e17..3a1b5fe4c2ca 100644 --- a/arch/x86/mm/mem_encrypt.c +++ b/arch/x86/mm/mem_encrypt.c @@ -25,17 +25,12 @@ #include #include #include -#include #include #include #include #include "mm_internal.h" -static char sme_cmdline_arg[] __initdata = "mem_encrypt"; -static char sme_cmdline_on[] __initdata = "on"; -static char sme_cmdline_off[] __initdata = "off"; - /* * Since SME related variables are set early in the boot process they must * reside in the .data section so as not to be zeroed out when the .bss @@ -46,7 +41,7 @@ EXPORT_SYMBOL(sme_me_mask); DEFINE_STATIC_KEY_FALSE(sev_enable_key); EXPORT_SYMBOL_GPL(sev_enable_key); -static bool sev_enabled __section(.data); +bool sev_enabled __section(.data); /* Buffer used for early in-place encryption by BSP, no locking needed */ static char sme_early_buffer[PAGE_SIZE] __aligned(PAGE_SIZE); @@ -463,574 +458,3 @@ void swiotlb_set_mem_attributes(void *vaddr, unsigned long size) /* Make the SWIOTLB buffer area decrypted */ set_memory_decrypted((unsigned long)vaddr, size >> PAGE_SHIFT); } - -struct sme_populate_pgd_data { - void *pgtable_area; - pgd_t *pgd; - - pmdval_t pmd_flags; - pteval_t pte_flags; - unsigned long paddr; - - unsigned long vaddr; - unsigned long vaddr_end; -}; - -static void __init sme_clear_pgd(struct sme_populate_pgd_data *ppd) -{ - unsigned long pgd_start, pgd_end, pgd_size; - pgd_t *pgd_p; - - pgd_start = ppd->vaddr & PGDIR_MASK; - pgd_end = ppd->vaddr_end & PGDIR_MASK; - - pgd_size = (((pgd_end - pgd_start) / PGDIR_SIZE) + 1) * sizeof(pgd_t); - - pgd_p = ppd->pgd + pgd_index(ppd->vaddr); - - memset(pgd_p, 0, pgd_size); -} - -#define PGD_FLAGS _KERNPG_TABLE_NOENC -#define P4D_FLAGS _KERNPG_TABLE_NOENC -#define PUD_FLAGS _KERNPG_TABLE_NOENC -#define PMD_FLAGS _KERNPG_TABLE_NOENC - -#define PMD_FLAGS_LARGE (__PAGE_KERNEL_LARGE_EXEC & ~_PAGE_GLOBAL) - -#define PMD_FLAGS_DEC PMD_FLAGS_LARGE -#define PMD_FLAGS_DEC_WP ((PMD_FLAGS_DEC & ~_PAGE_CACHE_MASK) | \ - (_PAGE_PAT | _PAGE_PWT)) - -#define PMD_FLAGS_ENC (PMD_FLAGS_LARGE | _PAGE_ENC) - -#define PTE_FLAGS (__PAGE_KERNEL_EXEC & ~_PAGE_GLOBAL) - -#define PTE_FLAGS_DEC PTE_FLAGS -#define PTE_FLAGS_DEC_WP ((PTE_FLAGS_DEC & ~_PAGE_CACHE_MASK) | \ - (_PAGE_PAT | _PAGE_PWT)) - -#define PTE_FLAGS_ENC (PTE_FLAGS | _PAGE_ENC) - -static pmd_t __init *sme_prepare_pgd(struct sme_populate_pgd_data *ppd) -{ - pgd_t *pgd_p; - p4d_t *p4d_p; - pud_t *pud_p; - pmd_t *pmd_p; - - pgd_p = ppd->pgd + pgd_index(ppd->vaddr); - if (native_pgd_val(*pgd_p)) { - if (IS_ENABLED(CONFIG_X86_5LEVEL)) - p4d_p = (p4d_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); - else - pud_p = (pud_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); - } else { - pgd_t pgd; - - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_p = ppd->pgtable_area; - memset(p4d_p, 0, sizeof(*p4d_p) * PTRS_PER_P4D); - ppd->pgtable_area += sizeof(*p4d_p) * PTRS_PER_P4D; - - pgd = native_make_pgd((pgdval_t)p4d_p + PGD_FLAGS); - } else { - pud_p = ppd->pgtable_area; - memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); - ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; - - pgd = native_make_pgd((pgdval_t)pud_p + PGD_FLAGS); - } - native_set_pgd(pgd_p, pgd); - } - - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_p += p4d_index(ppd->vaddr); - if (native_p4d_val(*p4d_p)) { - pud_p = (pud_t *)(native_p4d_val(*p4d_p) & ~PTE_FLAGS_MASK); - } else { - p4d_t p4d; - - pud_p = ppd->pgtable_area; - memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); - ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; - - p4d = native_make_p4d((pudval_t)pud_p + P4D_FLAGS); - native_set_p4d(p4d_p, p4d); - } - } - - pud_p += pud_index(ppd->vaddr); - if (native_pud_val(*pud_p)) { - if (native_pud_val(*pud_p) & _PAGE_PSE) - return NULL; - - pmd_p = (pmd_t *)(native_pud_val(*pud_p) & ~PTE_FLAGS_MASK); - } else { - pud_t pud; - - pmd_p = ppd->pgtable_area; - memset(pmd_p, 0, sizeof(*pmd_p) * PTRS_PER_PMD); - ppd->pgtable_area += sizeof(*pmd_p) * PTRS_PER_PMD; - - pud = native_make_pud((pmdval_t)pmd_p + PUD_FLAGS); - native_set_pud(pud_p, pud); - } - - return pmd_p; -} - -static void __init sme_populate_pgd_large(struct sme_populate_pgd_data *ppd) -{ - pmd_t *pmd_p; - - pmd_p = sme_prepare_pgd(ppd); - if (!pmd_p) - return; - - pmd_p += pmd_index(ppd->vaddr); - if (!native_pmd_val(*pmd_p) || !(native_pmd_val(*pmd_p) & _PAGE_PSE)) - native_set_pmd(pmd_p, native_make_pmd(ppd->paddr | ppd->pmd_flags)); -} - -static void __init sme_populate_pgd(struct sme_populate_pgd_data *ppd) -{ - pmd_t *pmd_p; - pte_t *pte_p; - - pmd_p = sme_prepare_pgd(ppd); - if (!pmd_p) - return; - - pmd_p += pmd_index(ppd->vaddr); - if (native_pmd_val(*pmd_p)) { - if (native_pmd_val(*pmd_p) & _PAGE_PSE) - return; - - pte_p = (pte_t *)(native_pmd_val(*pmd_p) & ~PTE_FLAGS_MASK); - } else { - pmd_t pmd; - - pte_p = ppd->pgtable_area; - memset(pte_p, 0, sizeof(*pte_p) * PTRS_PER_PTE); - ppd->pgtable_area += sizeof(*pte_p) * PTRS_PER_PTE; - - pmd = native_make_pmd((pteval_t)pte_p + PMD_FLAGS); - native_set_pmd(pmd_p, pmd); - } - - pte_p += pte_index(ppd->vaddr); - if (!native_pte_val(*pte_p)) - native_set_pte(pte_p, native_make_pte(ppd->paddr | ppd->pte_flags)); -} - -static void __init __sme_map_range_pmd(struct sme_populate_pgd_data *ppd) -{ - while (ppd->vaddr < ppd->vaddr_end) { - sme_populate_pgd_large(ppd); - - ppd->vaddr += PMD_PAGE_SIZE; - ppd->paddr += PMD_PAGE_SIZE; - } -} - -static void __init __sme_map_range_pte(struct sme_populate_pgd_data *ppd) -{ - while (ppd->vaddr < ppd->vaddr_end) { - sme_populate_pgd(ppd); - - ppd->vaddr += PAGE_SIZE; - ppd->paddr += PAGE_SIZE; - } -} - -static void __init __sme_map_range(struct sme_populate_pgd_data *ppd, - pmdval_t pmd_flags, pteval_t pte_flags) -{ - unsigned long vaddr_end; - - ppd->pmd_flags = pmd_flags; - ppd->pte_flags = pte_flags; - - /* Save original end value since we modify the struct value */ - vaddr_end = ppd->vaddr_end; - - /* If start is not 2MB aligned, create PTE entries */ - ppd->vaddr_end = ALIGN(ppd->vaddr, PMD_PAGE_SIZE); - __sme_map_range_pte(ppd); - - /* Create PMD entries */ - ppd->vaddr_end = vaddr_end & PMD_PAGE_MASK; - __sme_map_range_pmd(ppd); - - /* If end is not 2MB aligned, create PTE entries */ - ppd->vaddr_end = vaddr_end; - __sme_map_range_pte(ppd); -} - -static void __init sme_map_range_encrypted(struct sme_populate_pgd_data *ppd) -{ - __sme_map_range(ppd, PMD_FLAGS_ENC, PTE_FLAGS_ENC); -} - -static void __init sme_map_range_decrypted(struct sme_populate_pgd_data *ppd) -{ - __sme_map_range(ppd, PMD_FLAGS_DEC, PTE_FLAGS_DEC); -} - -static void __init sme_map_range_decrypted_wp(struct sme_populate_pgd_data *ppd) -{ - __sme_map_range(ppd, PMD_FLAGS_DEC_WP, PTE_FLAGS_DEC_WP); -} - -static unsigned long __init sme_pgtable_calc(unsigned long len) -{ - unsigned long p4d_size, pud_size, pmd_size, pte_size; - unsigned long total; - - /* - * Perform a relatively simplistic calculation of the pagetable - * entries that are needed. Those mappings will be covered mostly - * by 2MB PMD entries so we can conservatively calculate the required - * number of P4D, PUD and PMD structures needed to perform the - * mappings. For mappings that are not 2MB aligned, PTE mappings - * would be needed for the start and end portion of the address range - * that fall outside of the 2MB alignment. This results in, at most, - * two extra pages to hold PTE entries for each range that is mapped. - * Incrementing the count for each covers the case where the addresses - * cross entries. - */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; - p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; - pud_size = (ALIGN(len, P4D_SIZE) / P4D_SIZE) + 1; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } else { - p4d_size = 0; - pud_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } - pmd_size = (ALIGN(len, PUD_SIZE) / PUD_SIZE) + 1; - pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; - pte_size = 2 * sizeof(pte_t) * PTRS_PER_PTE; - - total = p4d_size + pud_size + pmd_size + pte_size; - - /* - * Now calculate the added pagetable structures needed to populate - * the new pagetables. - */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; - p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; - pud_size = ALIGN(total, P4D_SIZE) / P4D_SIZE; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } else { - p4d_size = 0; - pud_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } - pmd_size = ALIGN(total, PUD_SIZE) / PUD_SIZE; - pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; - - total += p4d_size + pud_size + pmd_size; - - return total; -} - -void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) -{ - unsigned long workarea_start, workarea_end, workarea_len; - unsigned long execute_start, execute_end, execute_len; - unsigned long kernel_start, kernel_end, kernel_len; - unsigned long initrd_start, initrd_end, initrd_len; - struct sme_populate_pgd_data ppd; - unsigned long pgtable_area_len; - unsigned long decrypted_base; - - if (!sme_active()) - return; - - /* - * Prepare for encrypting the kernel and initrd by building new - * pagetables with the necessary attributes needed to encrypt the - * kernel in place. - * - * One range of virtual addresses will map the memory occupied - * by the kernel and initrd as encrypted. - * - * Another range of virtual addresses will map the memory occupied - * by the kernel and initrd as decrypted and write-protected. - * - * The use of write-protect attribute will prevent any of the - * memory from being cached. - */ - - /* Physical addresses gives us the identity mapped virtual addresses */ - kernel_start = __pa_symbol(_text); - kernel_end = ALIGN(__pa_symbol(_end), PMD_PAGE_SIZE); - kernel_len = kernel_end - kernel_start; - - initrd_start = 0; - initrd_end = 0; - initrd_len = 0; -#ifdef CONFIG_BLK_DEV_INITRD - initrd_len = (unsigned long)bp->hdr.ramdisk_size | - ((unsigned long)bp->ext_ramdisk_size << 32); - if (initrd_len) { - initrd_start = (unsigned long)bp->hdr.ramdisk_image | - ((unsigned long)bp->ext_ramdisk_image << 32); - initrd_end = PAGE_ALIGN(initrd_start + initrd_len); - initrd_len = initrd_end - initrd_start; - } -#endif - - /* Set the encryption workarea to be immediately after the kernel */ - workarea_start = kernel_end; - - /* - * Calculate required number of workarea bytes needed: - * executable encryption area size: - * stack page (PAGE_SIZE) - * encryption routine page (PAGE_SIZE) - * intermediate copy buffer (PMD_PAGE_SIZE) - * pagetable structures for the encryption of the kernel - * pagetable structures for workarea (in case not currently mapped) - */ - execute_start = workarea_start; - execute_end = execute_start + (PAGE_SIZE * 2) + PMD_PAGE_SIZE; - execute_len = execute_end - execute_start; - - /* - * One PGD for both encrypted and decrypted mappings and a set of - * PUDs and PMDs for each of the encrypted and decrypted mappings. - */ - pgtable_area_len = sizeof(pgd_t) * PTRS_PER_PGD; - pgtable_area_len += sme_pgtable_calc(execute_end - kernel_start) * 2; - if (initrd_len) - pgtable_area_len += sme_pgtable_calc(initrd_len) * 2; - - /* PUDs and PMDs needed in the current pagetables for the workarea */ - pgtable_area_len += sme_pgtable_calc(execute_len + pgtable_area_len); - - /* - * The total workarea includes the executable encryption area and - * the pagetable area. The start of the workarea is already 2MB - * aligned, align the end of the workarea on a 2MB boundary so that - * we don't try to create/allocate PTE entries from the workarea - * before it is mapped. - */ - workarea_len = execute_len + pgtable_area_len; - workarea_end = ALIGN(workarea_start + workarea_len, PMD_PAGE_SIZE); - - /* - * Set the address to the start of where newly created pagetable - * structures (PGDs, PUDs and PMDs) will be allocated. New pagetable - * structures are created when the workarea is added to the current - * pagetables and when the new encrypted and decrypted kernel - * mappings are populated. - */ - ppd.pgtable_area = (void *)execute_end; - - /* - * Make sure the current pagetable structure has entries for - * addressing the workarea. - */ - ppd.pgd = (pgd_t *)native_read_cr3_pa(); - ppd.paddr = workarea_start; - ppd.vaddr = workarea_start; - ppd.vaddr_end = workarea_end; - sme_map_range_decrypted(&ppd); - - /* Flush the TLB - no globals so cr3 is enough */ - native_write_cr3(__native_read_cr3()); - - /* - * A new pagetable structure is being built to allow for the kernel - * and initrd to be encrypted. It starts with an empty PGD that will - * then be populated with new PUDs and PMDs as the encrypted and - * decrypted kernel mappings are created. - */ - ppd.pgd = ppd.pgtable_area; - memset(ppd.pgd, 0, sizeof(pgd_t) * PTRS_PER_PGD); - ppd.pgtable_area += sizeof(pgd_t) * PTRS_PER_PGD; - - /* - * A different PGD index/entry must be used to get different - * pagetable entries for the decrypted mapping. Choose the next - * PGD index and convert it to a virtual address to be used as - * the base of the mapping. - */ - decrypted_base = (pgd_index(workarea_end) + 1) & (PTRS_PER_PGD - 1); - if (initrd_len) { - unsigned long check_base; - - check_base = (pgd_index(initrd_end) + 1) & (PTRS_PER_PGD - 1); - decrypted_base = max(decrypted_base, check_base); - } - decrypted_base <<= PGDIR_SHIFT; - - /* Add encrypted kernel (identity) mappings */ - ppd.paddr = kernel_start; - ppd.vaddr = kernel_start; - ppd.vaddr_end = kernel_end; - sme_map_range_encrypted(&ppd); - - /* Add decrypted, write-protected kernel (non-identity) mappings */ - ppd.paddr = kernel_start; - ppd.vaddr = kernel_start + decrypted_base; - ppd.vaddr_end = kernel_end + decrypted_base; - sme_map_range_decrypted_wp(&ppd); - - if (initrd_len) { - /* Add encrypted initrd (identity) mappings */ - ppd.paddr = initrd_start; - ppd.vaddr = initrd_start; - ppd.vaddr_end = initrd_end; - sme_map_range_encrypted(&ppd); - /* - * Add decrypted, write-protected initrd (non-identity) mappings - */ - ppd.paddr = initrd_start; - ppd.vaddr = initrd_start + decrypted_base; - ppd.vaddr_end = initrd_end + decrypted_base; - sme_map_range_decrypted_wp(&ppd); - } - - /* Add decrypted workarea mappings to both kernel mappings */ - ppd.paddr = workarea_start; - ppd.vaddr = workarea_start; - ppd.vaddr_end = workarea_end; - sme_map_range_decrypted(&ppd); - - ppd.paddr = workarea_start; - ppd.vaddr = workarea_start + decrypted_base; - ppd.vaddr_end = workarea_end + decrypted_base; - sme_map_range_decrypted(&ppd); - - /* Perform the encryption */ - sme_encrypt_execute(kernel_start, kernel_start + decrypted_base, - kernel_len, workarea_start, (unsigned long)ppd.pgd); - - if (initrd_len) - sme_encrypt_execute(initrd_start, initrd_start + decrypted_base, - initrd_len, workarea_start, - (unsigned long)ppd.pgd); - - /* - * At this point we are running encrypted. Remove the mappings for - * the decrypted areas - all that is needed for this is to remove - * the PGD entry/entries. - */ - ppd.vaddr = kernel_start + decrypted_base; - ppd.vaddr_end = kernel_end + decrypted_base; - sme_clear_pgd(&ppd); - - if (initrd_len) { - ppd.vaddr = initrd_start + decrypted_base; - ppd.vaddr_end = initrd_end + decrypted_base; - sme_clear_pgd(&ppd); - } - - ppd.vaddr = workarea_start + decrypted_base; - ppd.vaddr_end = workarea_end + decrypted_base; - sme_clear_pgd(&ppd); - - /* Flush the TLB - no globals so cr3 is enough */ - native_write_cr3(__native_read_cr3()); -} - -void __init __nostackprotector sme_enable(struct boot_params *bp) -{ - const char *cmdline_ptr, *cmdline_arg, *cmdline_on, *cmdline_off; - unsigned int eax, ebx, ecx, edx; - unsigned long feature_mask; - bool active_by_default; - unsigned long me_mask; - char buffer[16]; - u64 msr; - - /* Check for the SME/SEV support leaf */ - eax = 0x80000000; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - if (eax < 0x8000001f) - return; - -#define AMD_SME_BIT BIT(0) -#define AMD_SEV_BIT BIT(1) - /* - * Set the feature mask (SME or SEV) based on whether we are - * running under a hypervisor. - */ - eax = 1; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - feature_mask = (ecx & BIT(31)) ? AMD_SEV_BIT : AMD_SME_BIT; - - /* - * Check for the SME/SEV feature: - * CPUID Fn8000_001F[EAX] - * - Bit 0 - Secure Memory Encryption support - * - Bit 1 - Secure Encrypted Virtualization support - * CPUID Fn8000_001F[EBX] - * - Bits 5:0 - Pagetable bit position used to indicate encryption - */ - eax = 0x8000001f; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - if (!(eax & feature_mask)) - return; - - me_mask = 1UL << (ebx & 0x3f); - - /* Check if memory encryption is enabled */ - if (feature_mask == AMD_SME_BIT) { - /* For SME, check the SYSCFG MSR */ - msr = __rdmsr(MSR_K8_SYSCFG); - if (!(msr & MSR_K8_SYSCFG_MEM_ENCRYPT)) - return; - } else { - /* For SEV, check the SEV MSR */ - msr = __rdmsr(MSR_AMD64_SEV); - if (!(msr & MSR_AMD64_SEV_ENABLED)) - return; - - /* SEV state cannot be controlled by a command line option */ - sme_me_mask = me_mask; - sev_enabled = true; - return; - } - - /* - * Fixups have not been applied to phys_base yet and we're running - * identity mapped, so we must obtain the address to the SME command - * line argument data using rip-relative addressing. - */ - asm ("lea sme_cmdline_arg(%%rip), %0" - : "=r" (cmdline_arg) - : "p" (sme_cmdline_arg)); - asm ("lea sme_cmdline_on(%%rip), %0" - : "=r" (cmdline_on) - : "p" (sme_cmdline_on)); - asm ("lea sme_cmdline_off(%%rip), %0" - : "=r" (cmdline_off) - : "p" (sme_cmdline_off)); - - if (IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT)) - active_by_default = true; - else - active_by_default = false; - - cmdline_ptr = (const char *)((u64)bp->hdr.cmd_line_ptr | - ((u64)bp->ext_cmd_line_ptr << 32)); - - cmdline_find_option(cmdline_ptr, cmdline_arg, buffer, sizeof(buffer)); - - if (!strncmp(buffer, cmdline_on, sizeof(buffer))) - sme_me_mask = me_mask; - else if (!strncmp(buffer, cmdline_off, sizeof(buffer))) - sme_me_mask = 0; - else - sme_me_mask = active_by_default ? me_mask : 0; -} diff --git a/arch/x86/mm/mem_encrypt_identity.c b/arch/x86/mm/mem_encrypt_identity.c new file mode 100644 index 000000000000..a28978a37bfa --- /dev/null +++ b/arch/x86/mm/mem_encrypt_identity.c @@ -0,0 +1,597 @@ +/* + * AMD Memory Encryption Support + * + * Copyright (C) 2016 Advanced Micro Devices, Inc. + * + * Author: Tom Lendacky + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#define DISABLE_BRANCH_PROFILING + +#include +#include + +#include +#include +#include + +#include "mm_internal.h" + +#define PGD_FLAGS _KERNPG_TABLE_NOENC +#define P4D_FLAGS _KERNPG_TABLE_NOENC +#define PUD_FLAGS _KERNPG_TABLE_NOENC +#define PMD_FLAGS _KERNPG_TABLE_NOENC + +#define PMD_FLAGS_LARGE (__PAGE_KERNEL_LARGE_EXEC & ~_PAGE_GLOBAL) + +#define PMD_FLAGS_DEC PMD_FLAGS_LARGE +#define PMD_FLAGS_DEC_WP ((PMD_FLAGS_DEC & ~_PAGE_CACHE_MASK) | \ + (_PAGE_PAT | _PAGE_PWT)) + +#define PMD_FLAGS_ENC (PMD_FLAGS_LARGE | _PAGE_ENC) + +#define PTE_FLAGS (__PAGE_KERNEL_EXEC & ~_PAGE_GLOBAL) + +#define PTE_FLAGS_DEC PTE_FLAGS +#define PTE_FLAGS_DEC_WP ((PTE_FLAGS_DEC & ~_PAGE_CACHE_MASK) | \ + (_PAGE_PAT | _PAGE_PWT)) + +#define PTE_FLAGS_ENC (PTE_FLAGS | _PAGE_ENC) + +struct sme_populate_pgd_data { + void *pgtable_area; + pgd_t *pgd; + + pmdval_t pmd_flags; + pteval_t pte_flags; + unsigned long paddr; + + unsigned long vaddr; + unsigned long vaddr_end; +}; + +static char sme_cmdline_arg[] __initdata = "mem_encrypt"; +static char sme_cmdline_on[] __initdata = "on"; +static char sme_cmdline_off[] __initdata = "off"; + +static void __init sme_clear_pgd(struct sme_populate_pgd_data *ppd) +{ + unsigned long pgd_start, pgd_end, pgd_size; + pgd_t *pgd_p; + + pgd_start = ppd->vaddr & PGDIR_MASK; + pgd_end = ppd->vaddr_end & PGDIR_MASK; + + pgd_size = (((pgd_end - pgd_start) / PGDIR_SIZE) + 1) * sizeof(pgd_t); + + pgd_p = ppd->pgd + pgd_index(ppd->vaddr); + + memset(pgd_p, 0, pgd_size); +} + +static pmd_t __init *sme_prepare_pgd(struct sme_populate_pgd_data *ppd) +{ + pgd_t *pgd_p; + p4d_t *p4d_p; + pud_t *pud_p; + pmd_t *pmd_p; + + pgd_p = ppd->pgd + pgd_index(ppd->vaddr); + if (native_pgd_val(*pgd_p)) { + if (IS_ENABLED(CONFIG_X86_5LEVEL)) + p4d_p = (p4d_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); + else + pud_p = (pud_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); + } else { + pgd_t pgd; + + if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + p4d_p = ppd->pgtable_area; + memset(p4d_p, 0, sizeof(*p4d_p) * PTRS_PER_P4D); + ppd->pgtable_area += sizeof(*p4d_p) * PTRS_PER_P4D; + + pgd = native_make_pgd((pgdval_t)p4d_p + PGD_FLAGS); + } else { + pud_p = ppd->pgtable_area; + memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); + ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; + + pgd = native_make_pgd((pgdval_t)pud_p + PGD_FLAGS); + } + native_set_pgd(pgd_p, pgd); + } + + if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + p4d_p += p4d_index(ppd->vaddr); + if (native_p4d_val(*p4d_p)) { + pud_p = (pud_t *)(native_p4d_val(*p4d_p) & ~PTE_FLAGS_MASK); + } else { + p4d_t p4d; + + pud_p = ppd->pgtable_area; + memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); + ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; + + p4d = native_make_p4d((pudval_t)pud_p + P4D_FLAGS); + native_set_p4d(p4d_p, p4d); + } + } + + pud_p += pud_index(ppd->vaddr); + if (native_pud_val(*pud_p)) { + if (native_pud_val(*pud_p) & _PAGE_PSE) + return NULL; + + pmd_p = (pmd_t *)(native_pud_val(*pud_p) & ~PTE_FLAGS_MASK); + } else { + pud_t pud; + + pmd_p = ppd->pgtable_area; + memset(pmd_p, 0, sizeof(*pmd_p) * PTRS_PER_PMD); + ppd->pgtable_area += sizeof(*pmd_p) * PTRS_PER_PMD; + + pud = native_make_pud((pmdval_t)pmd_p + PUD_FLAGS); + native_set_pud(pud_p, pud); + } + + return pmd_p; +} + +static void __init sme_populate_pgd_large(struct sme_populate_pgd_data *ppd) +{ + pmd_t *pmd_p; + + pmd_p = sme_prepare_pgd(ppd); + if (!pmd_p) + return; + + pmd_p += pmd_index(ppd->vaddr); + if (!native_pmd_val(*pmd_p) || !(native_pmd_val(*pmd_p) & _PAGE_PSE)) + native_set_pmd(pmd_p, native_make_pmd(ppd->paddr | ppd->pmd_flags)); +} + +static void __init sme_populate_pgd(struct sme_populate_pgd_data *ppd) +{ + pmd_t *pmd_p; + pte_t *pte_p; + + pmd_p = sme_prepare_pgd(ppd); + if (!pmd_p) + return; + + pmd_p += pmd_index(ppd->vaddr); + if (native_pmd_val(*pmd_p)) { + if (native_pmd_val(*pmd_p) & _PAGE_PSE) + return; + + pte_p = (pte_t *)(native_pmd_val(*pmd_p) & ~PTE_FLAGS_MASK); + } else { + pmd_t pmd; + + pte_p = ppd->pgtable_area; + memset(pte_p, 0, sizeof(*pte_p) * PTRS_PER_PTE); + ppd->pgtable_area += sizeof(*pte_p) * PTRS_PER_PTE; + + pmd = native_make_pmd((pteval_t)pte_p + PMD_FLAGS); + native_set_pmd(pmd_p, pmd); + } + + pte_p += pte_index(ppd->vaddr); + if (!native_pte_val(*pte_p)) + native_set_pte(pte_p, native_make_pte(ppd->paddr | ppd->pte_flags)); +} + +static void __init __sme_map_range_pmd(struct sme_populate_pgd_data *ppd) +{ + while (ppd->vaddr < ppd->vaddr_end) { + sme_populate_pgd_large(ppd); + + ppd->vaddr += PMD_PAGE_SIZE; + ppd->paddr += PMD_PAGE_SIZE; + } +} + +static void __init __sme_map_range_pte(struct sme_populate_pgd_data *ppd) +{ + while (ppd->vaddr < ppd->vaddr_end) { + sme_populate_pgd(ppd); + + ppd->vaddr += PAGE_SIZE; + ppd->paddr += PAGE_SIZE; + } +} + +static void __init __sme_map_range(struct sme_populate_pgd_data *ppd, + pmdval_t pmd_flags, pteval_t pte_flags) +{ + unsigned long vaddr_end; + + ppd->pmd_flags = pmd_flags; + ppd->pte_flags = pte_flags; + + /* Save original end value since we modify the struct value */ + vaddr_end = ppd->vaddr_end; + + /* If start is not 2MB aligned, create PTE entries */ + ppd->vaddr_end = ALIGN(ppd->vaddr, PMD_PAGE_SIZE); + __sme_map_range_pte(ppd); + + /* Create PMD entries */ + ppd->vaddr_end = vaddr_end & PMD_PAGE_MASK; + __sme_map_range_pmd(ppd); + + /* If end is not 2MB aligned, create PTE entries */ + ppd->vaddr_end = vaddr_end; + __sme_map_range_pte(ppd); +} + +static void __init sme_map_range_encrypted(struct sme_populate_pgd_data *ppd) +{ + __sme_map_range(ppd, PMD_FLAGS_ENC, PTE_FLAGS_ENC); +} + +static void __init sme_map_range_decrypted(struct sme_populate_pgd_data *ppd) +{ + __sme_map_range(ppd, PMD_FLAGS_DEC, PTE_FLAGS_DEC); +} + +static void __init sme_map_range_decrypted_wp(struct sme_populate_pgd_data *ppd) +{ + __sme_map_range(ppd, PMD_FLAGS_DEC_WP, PTE_FLAGS_DEC_WP); +} + +static unsigned long __init sme_pgtable_calc(unsigned long len) +{ + unsigned long p4d_size, pud_size, pmd_size, pte_size; + unsigned long total; + + /* + * Perform a relatively simplistic calculation of the pagetable + * entries that are needed. Those mappings will be covered mostly + * by 2MB PMD entries so we can conservatively calculate the required + * number of P4D, PUD and PMD structures needed to perform the + * mappings. For mappings that are not 2MB aligned, PTE mappings + * would be needed for the start and end portion of the address range + * that fall outside of the 2MB alignment. This results in, at most, + * two extra pages to hold PTE entries for each range that is mapped. + * Incrementing the count for each covers the case where the addresses + * cross entries. + */ + if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + p4d_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; + p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; + pud_size = (ALIGN(len, P4D_SIZE) / P4D_SIZE) + 1; + pud_size *= sizeof(pud_t) * PTRS_PER_PUD; + } else { + p4d_size = 0; + pud_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; + pud_size *= sizeof(pud_t) * PTRS_PER_PUD; + } + pmd_size = (ALIGN(len, PUD_SIZE) / PUD_SIZE) + 1; + pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; + pte_size = 2 * sizeof(pte_t) * PTRS_PER_PTE; + + total = p4d_size + pud_size + pmd_size + pte_size; + + /* + * Now calculate the added pagetable structures needed to populate + * the new pagetables. + */ + if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + p4d_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; + p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; + pud_size = ALIGN(total, P4D_SIZE) / P4D_SIZE; + pud_size *= sizeof(pud_t) * PTRS_PER_PUD; + } else { + p4d_size = 0; + pud_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; + pud_size *= sizeof(pud_t) * PTRS_PER_PUD; + } + pmd_size = ALIGN(total, PUD_SIZE) / PUD_SIZE; + pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; + + total += p4d_size + pud_size + pmd_size; + + return total; +} + +void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) +{ + unsigned long workarea_start, workarea_end, workarea_len; + unsigned long execute_start, execute_end, execute_len; + unsigned long kernel_start, kernel_end, kernel_len; + unsigned long initrd_start, initrd_end, initrd_len; + struct sme_populate_pgd_data ppd; + unsigned long pgtable_area_len; + unsigned long decrypted_base; + + if (!sme_active()) + return; + + /* + * Prepare for encrypting the kernel and initrd by building new + * pagetables with the necessary attributes needed to encrypt the + * kernel in place. + * + * One range of virtual addresses will map the memory occupied + * by the kernel and initrd as encrypted. + * + * Another range of virtual addresses will map the memory occupied + * by the kernel and initrd as decrypted and write-protected. + * + * The use of write-protect attribute will prevent any of the + * memory from being cached. + */ + + /* Physical addresses gives us the identity mapped virtual addresses */ + kernel_start = __pa_symbol(_text); + kernel_end = ALIGN(__pa_symbol(_end), PMD_PAGE_SIZE); + kernel_len = kernel_end - kernel_start; + + initrd_start = 0; + initrd_end = 0; + initrd_len = 0; +#ifdef CONFIG_BLK_DEV_INITRD + initrd_len = (unsigned long)bp->hdr.ramdisk_size | + ((unsigned long)bp->ext_ramdisk_size << 32); + if (initrd_len) { + initrd_start = (unsigned long)bp->hdr.ramdisk_image | + ((unsigned long)bp->ext_ramdisk_image << 32); + initrd_end = PAGE_ALIGN(initrd_start + initrd_len); + initrd_len = initrd_end - initrd_start; + } +#endif + + /* Set the encryption workarea to be immediately after the kernel */ + workarea_start = kernel_end; + + /* + * Calculate required number of workarea bytes needed: + * executable encryption area size: + * stack page (PAGE_SIZE) + * encryption routine page (PAGE_SIZE) + * intermediate copy buffer (PMD_PAGE_SIZE) + * pagetable structures for the encryption of the kernel + * pagetable structures for workarea (in case not currently mapped) + */ + execute_start = workarea_start; + execute_end = execute_start + (PAGE_SIZE * 2) + PMD_PAGE_SIZE; + execute_len = execute_end - execute_start; + + /* + * One PGD for both encrypted and decrypted mappings and a set of + * PUDs and PMDs for each of the encrypted and decrypted mappings. + */ + pgtable_area_len = sizeof(pgd_t) * PTRS_PER_PGD; + pgtable_area_len += sme_pgtable_calc(execute_end - kernel_start) * 2; + if (initrd_len) + pgtable_area_len += sme_pgtable_calc(initrd_len) * 2; + + /* PUDs and PMDs needed in the current pagetables for the workarea */ + pgtable_area_len += sme_pgtable_calc(execute_len + pgtable_area_len); + + /* + * The total workarea includes the executable encryption area and + * the pagetable area. The start of the workarea is already 2MB + * aligned, align the end of the workarea on a 2MB boundary so that + * we don't try to create/allocate PTE entries from the workarea + * before it is mapped. + */ + workarea_len = execute_len + pgtable_area_len; + workarea_end = ALIGN(workarea_start + workarea_len, PMD_PAGE_SIZE); + + /* + * Set the address to the start of where newly created pagetable + * structures (PGDs, PUDs and PMDs) will be allocated. New pagetable + * structures are created when the workarea is added to the current + * pagetables and when the new encrypted and decrypted kernel + * mappings are populated. + */ + ppd.pgtable_area = (void *)execute_end; + + /* + * Make sure the current pagetable structure has entries for + * addressing the workarea. + */ + ppd.pgd = (pgd_t *)native_read_cr3_pa(); + ppd.paddr = workarea_start; + ppd.vaddr = workarea_start; + ppd.vaddr_end = workarea_end; + sme_map_range_decrypted(&ppd); + + /* Flush the TLB - no globals so cr3 is enough */ + native_write_cr3(__native_read_cr3()); + + /* + * A new pagetable structure is being built to allow for the kernel + * and initrd to be encrypted. It starts with an empty PGD that will + * then be populated with new PUDs and PMDs as the encrypted and + * decrypted kernel mappings are created. + */ + ppd.pgd = ppd.pgtable_area; + memset(ppd.pgd, 0, sizeof(pgd_t) * PTRS_PER_PGD); + ppd.pgtable_area += sizeof(pgd_t) * PTRS_PER_PGD; + + /* + * A different PGD index/entry must be used to get different + * pagetable entries for the decrypted mapping. Choose the next + * PGD index and convert it to a virtual address to be used as + * the base of the mapping. + */ + decrypted_base = (pgd_index(workarea_end) + 1) & (PTRS_PER_PGD - 1); + if (initrd_len) { + unsigned long check_base; + + check_base = (pgd_index(initrd_end) + 1) & (PTRS_PER_PGD - 1); + decrypted_base = max(decrypted_base, check_base); + } + decrypted_base <<= PGDIR_SHIFT; + + /* Add encrypted kernel (identity) mappings */ + ppd.paddr = kernel_start; + ppd.vaddr = kernel_start; + ppd.vaddr_end = kernel_end; + sme_map_range_encrypted(&ppd); + + /* Add decrypted, write-protected kernel (non-identity) mappings */ + ppd.paddr = kernel_start; + ppd.vaddr = kernel_start + decrypted_base; + ppd.vaddr_end = kernel_end + decrypted_base; + sme_map_range_decrypted_wp(&ppd); + + if (initrd_len) { + /* Add encrypted initrd (identity) mappings */ + ppd.paddr = initrd_start; + ppd.vaddr = initrd_start; + ppd.vaddr_end = initrd_end; + sme_map_range_encrypted(&ppd); + /* + * Add decrypted, write-protected initrd (non-identity) mappings + */ + ppd.paddr = initrd_start; + ppd.vaddr = initrd_start + decrypted_base; + ppd.vaddr_end = initrd_end + decrypted_base; + sme_map_range_decrypted_wp(&ppd); + } + + /* Add decrypted workarea mappings to both kernel mappings */ + ppd.paddr = workarea_start; + ppd.vaddr = workarea_start; + ppd.vaddr_end = workarea_end; + sme_map_range_decrypted(&ppd); + + ppd.paddr = workarea_start; + ppd.vaddr = workarea_start + decrypted_base; + ppd.vaddr_end = workarea_end + decrypted_base; + sme_map_range_decrypted(&ppd); + + /* Perform the encryption */ + sme_encrypt_execute(kernel_start, kernel_start + decrypted_base, + kernel_len, workarea_start, (unsigned long)ppd.pgd); + + if (initrd_len) + sme_encrypt_execute(initrd_start, initrd_start + decrypted_base, + initrd_len, workarea_start, + (unsigned long)ppd.pgd); + + /* + * At this point we are running encrypted. Remove the mappings for + * the decrypted areas - all that is needed for this is to remove + * the PGD entry/entries. + */ + ppd.vaddr = kernel_start + decrypted_base; + ppd.vaddr_end = kernel_end + decrypted_base; + sme_clear_pgd(&ppd); + + if (initrd_len) { + ppd.vaddr = initrd_start + decrypted_base; + ppd.vaddr_end = initrd_end + decrypted_base; + sme_clear_pgd(&ppd); + } + + ppd.vaddr = workarea_start + decrypted_base; + ppd.vaddr_end = workarea_end + decrypted_base; + sme_clear_pgd(&ppd); + + /* Flush the TLB - no globals so cr3 is enough */ + native_write_cr3(__native_read_cr3()); +} + +void __init __nostackprotector sme_enable(struct boot_params *bp) +{ + const char *cmdline_ptr, *cmdline_arg, *cmdline_on, *cmdline_off; + unsigned int eax, ebx, ecx, edx; + unsigned long feature_mask; + bool active_by_default; + unsigned long me_mask; + char buffer[16]; + u64 msr; + + /* Check for the SME/SEV support leaf */ + eax = 0x80000000; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + if (eax < 0x8000001f) + return; + +#define AMD_SME_BIT BIT(0) +#define AMD_SEV_BIT BIT(1) + /* + * Set the feature mask (SME or SEV) based on whether we are + * running under a hypervisor. + */ + eax = 1; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + feature_mask = (ecx & BIT(31)) ? AMD_SEV_BIT : AMD_SME_BIT; + + /* + * Check for the SME/SEV feature: + * CPUID Fn8000_001F[EAX] + * - Bit 0 - Secure Memory Encryption support + * - Bit 1 - Secure Encrypted Virtualization support + * CPUID Fn8000_001F[EBX] + * - Bits 5:0 - Pagetable bit position used to indicate encryption + */ + eax = 0x8000001f; + ecx = 0; + native_cpuid(&eax, &ebx, &ecx, &edx); + if (!(eax & feature_mask)) + return; + + me_mask = 1UL << (ebx & 0x3f); + + /* Check if memory encryption is enabled */ + if (feature_mask == AMD_SME_BIT) { + /* For SME, check the SYSCFG MSR */ + msr = __rdmsr(MSR_K8_SYSCFG); + if (!(msr & MSR_K8_SYSCFG_MEM_ENCRYPT)) + return; + } else { + /* For SEV, check the SEV MSR */ + msr = __rdmsr(MSR_AMD64_SEV); + if (!(msr & MSR_AMD64_SEV_ENABLED)) + return; + + /* SEV state cannot be controlled by a command line option */ + sme_me_mask = me_mask; + sev_enabled = true; + return; + } + + /* + * Fixups have not been applied to phys_base yet and we're running + * identity mapped, so we must obtain the address to the SME command + * line argument data using rip-relative addressing. + */ + asm ("lea sme_cmdline_arg(%%rip), %0" + : "=r" (cmdline_arg) + : "p" (sme_cmdline_arg)); + asm ("lea sme_cmdline_on(%%rip), %0" + : "=r" (cmdline_on) + : "p" (sme_cmdline_on)); + asm ("lea sme_cmdline_off(%%rip), %0" + : "=r" (cmdline_off) + : "p" (sme_cmdline_off)); + + if (IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT)) + active_by_default = true; + else + active_by_default = false; + + cmdline_ptr = (const char *)((u64)bp->hdr.cmd_line_ptr | + ((u64)bp->ext_cmd_line_ptr << 32)); + + cmdline_find_option(cmdline_ptr, cmdline_arg, buffer, sizeof(buffer)); + + if (!strncmp(buffer, cmdline_on, sizeof(buffer))) + sme_me_mask = me_mask; + else if (!strncmp(buffer, cmdline_off, sizeof(buffer))) + sme_me_mask = 0; + else + sme_me_mask = active_by_default ? me_mask : 0; +} -- cgit v1.2.3-70-g09d2 From aad983913d77af2c3394f29b88d7bb75ebd7d172 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 31 Jan 2018 16:54:03 +0300 Subject: x86/mm/encrypt: Simplify sme_populate_pgd() and sme_populate_pgd_large() sme_populate_pgd() and sme_populate_pgd_large() operate on the identity mapping, which means they want virtual addresses to be equal to physical one, without PAGE_OFFSET shift. We also need to avoid paravirtualization call there. Getting this done is tricky. We cannot use usual page table helpers. It forces us to open-code a lot of things. It makes code ugly and hard to modify. We can get it work with the page table helpers, but it requires few preprocessor tricks. - Define __pa() and __va() to be compatible with identity mapping. - Undef CONFIG_PARAVIRT and CONFIG_PARAVIRT_SPINLOCKS before including any file. This way we can avoid paravirtualization calls. Now we can user normal page table helpers just fine. Tested-by: Tom Lendacky Signed-off-by: Kirill A. Shutemov Reviewed-by: Tom Lendacky Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180131135404.40692-3-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/mm/mem_encrypt_identity.c | 159 +++++++++++++++++-------------------- 1 file changed, 72 insertions(+), 87 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/mem_encrypt_identity.c b/arch/x86/mm/mem_encrypt_identity.c index a28978a37bfa..4b6a2e3098c5 100644 --- a/arch/x86/mm/mem_encrypt_identity.c +++ b/arch/x86/mm/mem_encrypt_identity.c @@ -12,6 +12,24 @@ #define DISABLE_BRANCH_PROFILING +/* + * Since we're dealing with identity mappings, physical and virtual + * addresses are the same, so override these defines which are ultimately + * used by the headers in misc.h. + */ +#define __pa(x) ((unsigned long)(x)) +#define __va(x) ((void *)((unsigned long)(x))) + +/* + * Special hack: we have to be careful, because no indirections are + * allowed here, and paravirt_ops is a kind of one. As it will only run in + * baremetal anyway, we just keep it from happening. (This list needs to + * be extended when new paravirt and debugging variants are added.) + */ +#undef CONFIG_PARAVIRT +#undef CONFIG_PARAVIRT_SPINLOCKS + +#include #include #include @@ -73,116 +91,83 @@ static void __init sme_clear_pgd(struct sme_populate_pgd_data *ppd) memset(pgd_p, 0, pgd_size); } -static pmd_t __init *sme_prepare_pgd(struct sme_populate_pgd_data *ppd) +static pud_t __init *sme_prepare_pgd(struct sme_populate_pgd_data *ppd) { - pgd_t *pgd_p; - p4d_t *p4d_p; - pud_t *pud_p; - pmd_t *pmd_p; - - pgd_p = ppd->pgd + pgd_index(ppd->vaddr); - if (native_pgd_val(*pgd_p)) { - if (IS_ENABLED(CONFIG_X86_5LEVEL)) - p4d_p = (p4d_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); - else - pud_p = (pud_t *)(native_pgd_val(*pgd_p) & ~PTE_FLAGS_MASK); - } else { - pgd_t pgd; - - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_p = ppd->pgtable_area; - memset(p4d_p, 0, sizeof(*p4d_p) * PTRS_PER_P4D); - ppd->pgtable_area += sizeof(*p4d_p) * PTRS_PER_P4D; - - pgd = native_make_pgd((pgdval_t)p4d_p + PGD_FLAGS); - } else { - pud_p = ppd->pgtable_area; - memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); - ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; - - pgd = native_make_pgd((pgdval_t)pud_p + PGD_FLAGS); - } - native_set_pgd(pgd_p, pgd); + pgd_t *pgd; + p4d_t *p4d; + pud_t *pud; + pmd_t *pmd; + + pgd = ppd->pgd + pgd_index(ppd->vaddr); + if (pgd_none(*pgd)) { + p4d = ppd->pgtable_area; + memset(p4d, 0, sizeof(*p4d) * PTRS_PER_P4D); + ppd->pgtable_area += sizeof(*p4d) * PTRS_PER_P4D; + set_pgd(pgd, __pgd(PGD_FLAGS | __pa(p4d))); } - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_p += p4d_index(ppd->vaddr); - if (native_p4d_val(*p4d_p)) { - pud_p = (pud_t *)(native_p4d_val(*p4d_p) & ~PTE_FLAGS_MASK); - } else { - p4d_t p4d; - - pud_p = ppd->pgtable_area; - memset(pud_p, 0, sizeof(*pud_p) * PTRS_PER_PUD); - ppd->pgtable_area += sizeof(*pud_p) * PTRS_PER_PUD; - - p4d = native_make_p4d((pudval_t)pud_p + P4D_FLAGS); - native_set_p4d(p4d_p, p4d); - } + p4d = p4d_offset(pgd, ppd->vaddr); + if (p4d_none(*p4d)) { + pud = ppd->pgtable_area; + memset(pud, 0, sizeof(*pud) * PTRS_PER_PUD); + ppd->pgtable_area += sizeof(*pud) * PTRS_PER_PUD; + set_p4d(p4d, __p4d(P4D_FLAGS | __pa(pud))); } - pud_p += pud_index(ppd->vaddr); - if (native_pud_val(*pud_p)) { - if (native_pud_val(*pud_p) & _PAGE_PSE) - return NULL; - - pmd_p = (pmd_t *)(native_pud_val(*pud_p) & ~PTE_FLAGS_MASK); - } else { - pud_t pud; - - pmd_p = ppd->pgtable_area; - memset(pmd_p, 0, sizeof(*pmd_p) * PTRS_PER_PMD); - ppd->pgtable_area += sizeof(*pmd_p) * PTRS_PER_PMD; - - pud = native_make_pud((pmdval_t)pmd_p + PUD_FLAGS); - native_set_pud(pud_p, pud); + pud = pud_offset(p4d, ppd->vaddr); + if (pud_none(*pud)) { + pmd = ppd->pgtable_area; + memset(pmd, 0, sizeof(*pmd) * PTRS_PER_PMD); + ppd->pgtable_area += sizeof(*pmd) * PTRS_PER_PMD; + set_pud(pud, __pud(PUD_FLAGS | __pa(pmd))); } - return pmd_p; + if (pud_large(*pud)) + return NULL; + + return pud; } static void __init sme_populate_pgd_large(struct sme_populate_pgd_data *ppd) { - pmd_t *pmd_p; + pud_t *pud; + pmd_t *pmd; - pmd_p = sme_prepare_pgd(ppd); - if (!pmd_p) + pud = sme_prepare_pgd(ppd); + if (!pud) return; - pmd_p += pmd_index(ppd->vaddr); - if (!native_pmd_val(*pmd_p) || !(native_pmd_val(*pmd_p) & _PAGE_PSE)) - native_set_pmd(pmd_p, native_make_pmd(ppd->paddr | ppd->pmd_flags)); + pmd = pmd_offset(pud, ppd->vaddr); + if (pmd_large(*pmd)) + return; + + set_pmd(pmd, __pmd(ppd->paddr | ppd->pmd_flags)); } static void __init sme_populate_pgd(struct sme_populate_pgd_data *ppd) { - pmd_t *pmd_p; - pte_t *pte_p; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; - pmd_p = sme_prepare_pgd(ppd); - if (!pmd_p) + pud = sme_prepare_pgd(ppd); + if (!pud) return; - pmd_p += pmd_index(ppd->vaddr); - if (native_pmd_val(*pmd_p)) { - if (native_pmd_val(*pmd_p) & _PAGE_PSE) - return; - - pte_p = (pte_t *)(native_pmd_val(*pmd_p) & ~PTE_FLAGS_MASK); - } else { - pmd_t pmd; - - pte_p = ppd->pgtable_area; - memset(pte_p, 0, sizeof(*pte_p) * PTRS_PER_PTE); - ppd->pgtable_area += sizeof(*pte_p) * PTRS_PER_PTE; - - pmd = native_make_pmd((pteval_t)pte_p + PMD_FLAGS); - native_set_pmd(pmd_p, pmd); + pmd = pmd_offset(pud, ppd->vaddr); + if (pmd_none(*pmd)) { + pte = ppd->pgtable_area; + memset(pte, 0, sizeof(pte) * PTRS_PER_PTE); + ppd->pgtable_area += sizeof(pte) * PTRS_PER_PTE; + set_pmd(pmd, __pmd(PMD_FLAGS | __pa(pte))); } - pte_p += pte_index(ppd->vaddr); - if (!native_pte_val(*pte_p)) - native_set_pte(pte_p, native_make_pte(ppd->paddr | ppd->pte_flags)); + if (pmd_large(*pmd)) + return; + + pte = pte_offset_map(pmd, ppd->vaddr); + if (pte_none(*pte)) + set_pte(pte, __pte(ppd->paddr | ppd->pte_flags)); } static void __init __sme_map_range_pmd(struct sme_populate_pgd_data *ppd) -- cgit v1.2.3-70-g09d2 From 1070730c1ad2bd49b0d11112728f1f4390137728 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 31 Jan 2018 16:54:04 +0300 Subject: x86/mm/encrypt: Simplify sme_pgtable_calc() sme_pgtable_calc() is unnecessary complex. It can be re-written in a more stream-lined way. As a side effect, we would get the code ready to boot-time switching between paging modes. Tested-by: Tom Lendacky Signed-off-by: Kirill A. Shutemov Reviewed-by: Tom Lendacky Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180131135404.40692-4-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/mm/mem_encrypt_identity.c | 42 +++++++++++--------------------------- 1 file changed, 12 insertions(+), 30 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/mem_encrypt_identity.c b/arch/x86/mm/mem_encrypt_identity.c index 4b6a2e3098c5..b4139c5ab972 100644 --- a/arch/x86/mm/mem_encrypt_identity.c +++ b/arch/x86/mm/mem_encrypt_identity.c @@ -231,8 +231,7 @@ static void __init sme_map_range_decrypted_wp(struct sme_populate_pgd_data *ppd) static unsigned long __init sme_pgtable_calc(unsigned long len) { - unsigned long p4d_size, pud_size, pmd_size, pte_size; - unsigned long total; + unsigned long entries = 0, tables = 0; /* * Perform a relatively simplistic calculation of the pagetable @@ -246,42 +245,25 @@ static unsigned long __init sme_pgtable_calc(unsigned long len) * Incrementing the count for each covers the case where the addresses * cross entries. */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; - p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; - pud_size = (ALIGN(len, P4D_SIZE) / P4D_SIZE) + 1; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } else { - p4d_size = 0; - pud_size = (ALIGN(len, PGDIR_SIZE) / PGDIR_SIZE) + 1; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } - pmd_size = (ALIGN(len, PUD_SIZE) / PUD_SIZE) + 1; - pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; - pte_size = 2 * sizeof(pte_t) * PTRS_PER_PTE; - total = p4d_size + pud_size + pmd_size + pte_size; + /* PGDIR_SIZE is equal to P4D_SIZE on 4-level machine. */ + if (PTRS_PER_P4D > 1) + entries += (DIV_ROUND_UP(len, PGDIR_SIZE) + 1) * sizeof(p4d_t) * PTRS_PER_P4D; + entries += (DIV_ROUND_UP(len, P4D_SIZE) + 1) * sizeof(pud_t) * PTRS_PER_PUD; + entries += (DIV_ROUND_UP(len, PUD_SIZE) + 1) * sizeof(pmd_t) * PTRS_PER_PMD; + entries += 2 * sizeof(pte_t) * PTRS_PER_PTE; /* * Now calculate the added pagetable structures needed to populate * the new pagetables. */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { - p4d_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; - p4d_size *= sizeof(p4d_t) * PTRS_PER_P4D; - pud_size = ALIGN(total, P4D_SIZE) / P4D_SIZE; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } else { - p4d_size = 0; - pud_size = ALIGN(total, PGDIR_SIZE) / PGDIR_SIZE; - pud_size *= sizeof(pud_t) * PTRS_PER_PUD; - } - pmd_size = ALIGN(total, PUD_SIZE) / PUD_SIZE; - pmd_size *= sizeof(pmd_t) * PTRS_PER_PMD; - total += p4d_size + pud_size + pmd_size; + if (PTRS_PER_P4D > 1) + tables += DIV_ROUND_UP(entries, PGDIR_SIZE) * sizeof(p4d_t) * PTRS_PER_P4D; + tables += DIV_ROUND_UP(entries, P4D_SIZE) * sizeof(pud_t) * PTRS_PER_PUD; + tables += DIV_ROUND_UP(entries, PUD_SIZE) * sizeof(pmd_t) * PTRS_PER_PMD; - return total; + return entries + tables; } void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) -- cgit v1.2.3-70-g09d2 From 116fef6408599dd6ff6996235c50aa692e9b5631 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Wed, 31 Jan 2018 07:56:22 -0800 Subject: x86/mm/dump_pagetables: Add the EFI pagetable to the debugfs 'page_tables' directory EFI is complicated enough that being able to view its pagetables is quite helpful. Rather than requiring users to fish it out of dmesg on an appropriately configured kernel, let users view it in debugfs as well. Signed-off-by: Andy Lutomirski Acked-by: Ard Biesheuvel Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-efi@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/ba158a93f3250e6fca752cff2cfb1fcdd9f2b50c.1517414050.git.luto@kernel.org [ Fixed trivial whitespace damage and fixed missing export. ] Signed-off-by: Ingo Molnar --- arch/x86/mm/debug_pagetables.c | 32 ++++++++++++++++++++++++++++++++ arch/x86/platform/efi/efi_64.c | 4 +++- 2 files changed, 35 insertions(+), 1 deletion(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/debug_pagetables.c b/arch/x86/mm/debug_pagetables.c index 421f2664ffa0..51a6f92da2bf 100644 --- a/arch/x86/mm/debug_pagetables.c +++ b/arch/x86/mm/debug_pagetables.c @@ -72,6 +72,31 @@ static const struct file_operations ptdump_curusr_fops = { }; #endif +#if defined(CONFIG_EFI) && defined(CONFIG_X86_64) +extern pgd_t *efi_pgd; +static struct dentry *pe_efi; + +static int ptdump_show_efi(struct seq_file *m, void *v) +{ + if (efi_pgd) + ptdump_walk_pgd_level_debugfs(m, efi_pgd, false); + return 0; +} + +static int ptdump_open_efi(struct inode *inode, struct file *filp) +{ + return single_open(filp, ptdump_show_efi, NULL); +} + +static const struct file_operations ptdump_efi_fops = { + .owner = THIS_MODULE, + .open = ptdump_open_efi, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + static struct dentry *dir, *pe_knl, *pe_curknl; static int __init pt_dump_debug_init(void) @@ -96,6 +121,13 @@ static int __init pt_dump_debug_init(void) if (!pe_curusr) goto err; #endif + +#if defined(CONFIG_EFI) && defined(CONFIG_X86_64) + pe_efi = debugfs_create_file("efi", 0400, dir, NULL, &ptdump_efi_fops); + if (!pe_efi) + goto err; +#endif + return 0; err: debugfs_remove_recursive(dir); diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index c310a8284358..780460aa5ea5 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -190,7 +191,8 @@ void __init efi_call_phys_epilog(pgd_t *save_pgd) early_code_mapping_set_exec(0); } -static pgd_t *efi_pgd; +pgd_t *efi_pgd; +EXPORT_SYMBOL_GPL(efi_pgd); /* * We need our own copy of the higher levels of the page tables -- cgit v1.2.3-70-g09d2 From eedb92abb9bb03ef21442614a6f5867eaac6e77f Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 14:16:50 +0300 Subject: x86/mm: Make virtual memory layout dynamic for CONFIG_X86_5LEVEL=y We need to be able to adjust virtual memory layout at runtime to be able to switch between 4- and 5-level paging at boot-time. KASLR already has movable __VMALLOC_BASE, __VMEMMAP_BASE and __PAGE_OFFSET. Let's re-use it. Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214111656.88514-4-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 8 ++++++++ arch/x86/include/asm/kaslr.h | 4 ---- arch/x86/include/asm/page_64.h | 4 ++++ arch/x86/include/asm/page_64_types.h | 4 ++-- arch/x86/include/asm/pgtable_64_types.h | 4 ++-- arch/x86/kernel/head64.c | 9 +++++++++ arch/x86/mm/kaslr.c | 8 -------- 7 files changed, 25 insertions(+), 16 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 63bf349b2b24..92256489b8a4 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1430,6 +1430,7 @@ config X86_PAE config X86_5LEVEL bool "Enable 5-level page tables support" + select DYNAMIC_MEMORY_LAYOUT depends on X86_64 ---help--- 5-level paging enables access to larger address space: @@ -2143,10 +2144,17 @@ config PHYSICAL_ALIGN Don't change this unless you know what you are doing. +config DYNAMIC_MEMORY_LAYOUT + bool + ---help--- + This option makes base addresses of vmalloc and vmemmap as well as + __PAGE_OFFSET movable during boot. + config RANDOMIZE_MEMORY bool "Randomize the kernel memory sections" depends on X86_64 depends on RANDOMIZE_BASE + select DYNAMIC_MEMORY_LAYOUT default RANDOMIZE_BASE ---help--- Randomizes the base virtual address of kernel memory sections diff --git a/arch/x86/include/asm/kaslr.h b/arch/x86/include/asm/kaslr.h index 460991e3b529..db7ba2feb947 100644 --- a/arch/x86/include/asm/kaslr.h +++ b/arch/x86/include/asm/kaslr.h @@ -5,10 +5,6 @@ unsigned long kaslr_get_random_long(const char *purpose); #ifdef CONFIG_RANDOMIZE_MEMORY -extern unsigned long page_offset_base; -extern unsigned long vmalloc_base; -extern unsigned long vmemmap_base; - void kernel_randomize_memory(void); #else static inline void kernel_randomize_memory(void) { } diff --git a/arch/x86/include/asm/page_64.h b/arch/x86/include/asm/page_64.h index 4baa6bceb232..096378650142 100644 --- a/arch/x86/include/asm/page_64.h +++ b/arch/x86/include/asm/page_64.h @@ -11,6 +11,10 @@ extern unsigned long max_pfn; extern unsigned long phys_base; +extern unsigned long page_offset_base; +extern unsigned long vmalloc_base; +extern unsigned long vmemmap_base; + static inline unsigned long __phys_addr_nodebug(unsigned long x) { unsigned long y = x - __START_KERNEL_map; diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index f68e6526891d..d54a3d5b5b3b 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -43,11 +43,11 @@ #define __PAGE_OFFSET_BASE _AC(0xffff880000000000, UL) #endif -#ifdef CONFIG_RANDOMIZE_MEMORY +#ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT #define __PAGE_OFFSET page_offset_base #else #define __PAGE_OFFSET __PAGE_OFFSET_BASE -#endif /* CONFIG_RANDOMIZE_MEMORY */ +#endif /* CONFIG_DYNAMIC_MEMORY_LAYOUT */ #define __START_KERNEL_map _AC(0xffffffff80000000, UL) diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index 7168de7d34eb..a0db91ab63b8 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -100,13 +100,13 @@ typedef struct { pteval_t pte; } pte_t; # define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT) #endif -#ifdef CONFIG_RANDOMIZE_MEMORY +#ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT # define VMALLOC_START vmalloc_base # define VMEMMAP_START vmemmap_base #else # define VMALLOC_START __VMALLOC_BASE # define VMEMMAP_START __VMEMMAP_BASE -#endif /* CONFIG_RANDOMIZE_MEMORY */ +#endif /* CONFIG_DYNAMIC_MEMORY_LAYOUT */ #define VMALLOC_END (VMALLOC_START + _AC((VMALLOC_SIZE_TB << 40) - 1, UL)) diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 7ba5d819ebe3..bf5c9ba63ba1 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -39,6 +39,15 @@ extern pmd_t early_dynamic_pgts[EARLY_DYNAMIC_PAGE_TABLES][PTRS_PER_PMD]; static unsigned int __initdata next_early_pgt; pmdval_t early_pmd_flags = __PAGE_KERNEL_LARGE & ~(_PAGE_GLOBAL | _PAGE_NX); +#ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT +unsigned long page_offset_base __ro_after_init = __PAGE_OFFSET_BASE; +EXPORT_SYMBOL(page_offset_base); +unsigned long vmalloc_base __ro_after_init = __VMALLOC_BASE; +EXPORT_SYMBOL(vmalloc_base); +unsigned long vmemmap_base __ro_after_init = __VMEMMAP_BASE; +EXPORT_SYMBOL(vmemmap_base); +#endif + #define __head __section(.head.text) static void __head *fixup_pointer(void *ptr, unsigned long physaddr) diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index aedebd2ebf1e..515b98a8ccee 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -43,14 +43,6 @@ static const unsigned long vaddr_start = __PAGE_OFFSET_BASE; static const unsigned long vaddr_end = CPU_ENTRY_AREA_BASE; -/* Default values */ -unsigned long page_offset_base = __PAGE_OFFSET_BASE; -EXPORT_SYMBOL(page_offset_base); -unsigned long vmalloc_base = __VMALLOC_BASE; -EXPORT_SYMBOL(vmalloc_base); -unsigned long vmemmap_base = __VMEMMAP_BASE; -EXPORT_SYMBOL(vmemmap_base); - /* * Memory regions randomized by KASLR (except modules that use a separate logic * earlier during boot). The list is ordered based on virtual addresses. This -- cgit v1.2.3-70-g09d2 From 5c7919bb1994f8dc7fed219a5db09e6bb9d473a5 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 14:16:52 +0300 Subject: x86/mm: Make LDT_BASE_ADDR dynamic LDT_BASE_ADDR has different value in 4- and 5-level paging configurations. We need to make it dynamic in preparation for boot-time switching between paging modes. Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214111656.88514-6-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable_64_types.h | 9 +++++---- arch/x86/mm/dump_pagetables.c | 5 ++++- 2 files changed, 9 insertions(+), 5 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index 5e2d724f8f47..903e4d054bcb 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -92,18 +92,19 @@ extern unsigned int pgtable_l5_enabled; */ #define MAXMEM _AC(__AC(1, UL) << MAX_PHYSMEM_BITS, UL) +#define LDT_PGD_ENTRY_L4 -3UL +#define LDT_PGD_ENTRY_L5 -112UL +#define LDT_PGD_ENTRY (pgtable_l5_enabled ? LDT_PGD_ENTRY_L5 : LDT_PGD_ENTRY_L4) +#define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT) + #ifdef CONFIG_X86_5LEVEL # define VMALLOC_SIZE_TB _AC(12800, UL) # define __VMALLOC_BASE _AC(0xffa0000000000000, UL) # define __VMEMMAP_BASE _AC(0xffd4000000000000, UL) -# define LDT_PGD_ENTRY _AC(-112, UL) -# define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT) #else # define VMALLOC_SIZE_TB _AC(32, UL) # define __VMALLOC_BASE _AC(0xffffc90000000000, UL) # define __VMEMMAP_BASE _AC(0xffffea0000000000, UL) -# define LDT_PGD_ENTRY _AC(-3, UL) -# define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT) #endif #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index 2a4849e92831..a89f2dbc3531 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -89,7 +89,7 @@ static struct addr_marker address_markers[] = { [KASAN_SHADOW_END_NR] = { KASAN_SHADOW_END, "KASAN shadow end" }, #endif #ifdef CONFIG_MODIFY_LDT_SYSCALL - [LDT_NR] = { LDT_BASE_ADDR, "LDT remap" }, + [LDT_NR] = { 0UL, "LDT remap" }, #endif [CPU_ENTRY_AREA_NR] = { CPU_ENTRY_AREA_BASE,"CPU entry Area" }, #ifdef CONFIG_X86_ESPFIX64 @@ -570,6 +570,9 @@ static int __init pt_dump_init(void) address_markers[LOW_KERNEL_NR].start_address = PAGE_OFFSET; address_markers[VMALLOC_START_NR].start_address = VMALLOC_START; address_markers[VMEMMAP_START_NR].start_address = VMEMMAP_START; +#ifdef CONFIG_MODIFY_LDT_SYSCALL + address_markers[LDT_NR].start_address = LDT_BASE_ADDR; +#endif #endif #ifdef CONFIG_X86_32 address_markers[VMALLOC_START_NR].start_address = VMALLOC_START; -- cgit v1.2.3-70-g09d2 From c65e774fb3f6af212641538694b9778ff9ab4300 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 14:16:53 +0300 Subject: x86/mm: Make PGDIR_SHIFT and PTRS_PER_P4D variable For boot-time switching between 4- and 5-level paging we need to be able to fold p4d page table level at runtime. It requires variable PGDIR_SHIFT and PTRS_PER_P4D. The change doesn't affect the kernel image size much: text data bss dec hex filename 8628091 4734304 1368064 14730459 e0c4db vmlinux.before 8628393 4734340 1368064 14730797 e0c62d vmlinux.after Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214111656.88514-7-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/boot/compressed/kaslr.c | 2 ++ arch/x86/include/asm/pgtable_32.h | 2 ++ arch/x86/include/asm/pgtable_64_types.h | 19 ++++++++++++------- arch/x86/kernel/cpu/mcheck/mce.c | 18 ++++++------------ arch/x86/kernel/head64.c | 6 +++++- arch/x86/mm/dump_pagetables.c | 12 +++++------- arch/x86/mm/init_64.c | 2 +- arch/x86/mm/kasan_init_64.c | 2 +- arch/x86/platform/efi/efi_64.c | 4 ++-- include/asm-generic/5level-fixup.h | 1 + include/asm-generic/pgtable-nop4d.h | 9 +++++---- include/linux/kasan.h | 2 +- mm/kasan/kasan_init.c | 2 +- 13 files changed, 44 insertions(+), 37 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c index bd69e1830fbe..b18e8f9512de 100644 --- a/arch/x86/boot/compressed/kaslr.c +++ b/arch/x86/boot/compressed/kaslr.c @@ -48,6 +48,8 @@ #ifdef CONFIG_X86_5LEVEL unsigned int pgtable_l5_enabled __ro_after_init = 1; +unsigned int pgdir_shift __ro_after_init = 48; +unsigned int ptrs_per_p4d __ro_after_init = 512; #endif extern unsigned long get_cmd_line_ptr(void); diff --git a/arch/x86/include/asm/pgtable_32.h b/arch/x86/include/asm/pgtable_32.h index e67c0620aec2..d829360e26bd 100644 --- a/arch/x86/include/asm/pgtable_32.h +++ b/arch/x86/include/asm/pgtable_32.h @@ -33,6 +33,8 @@ static inline void pgtable_cache_init(void) { } static inline void check_pgt_cache(void) { } void paging_init(void); +static inline int pgd_large(pgd_t pgd) { return 0; } + /* * Define this if things work differently on an i386 and an i486: * it will (on an i486) warn about kernel memory accesses that are diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index 903e4d054bcb..0c48d80e11d4 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -26,6 +26,9 @@ extern unsigned int pgtable_l5_enabled; #define pgtable_l5_enabled 0 #endif +extern unsigned int pgdir_shift; +extern unsigned int ptrs_per_p4d; + #endif /* !__ASSEMBLY__ */ #define SHARED_KERNEL_PMD 0 @@ -35,16 +38,17 @@ extern unsigned int pgtable_l5_enabled; /* * PGDIR_SHIFT determines what a top-level page table entry can map */ -#define PGDIR_SHIFT 48 +#define PGDIR_SHIFT pgdir_shift #define PTRS_PER_PGD 512 /* * 4th level page in 5-level paging case */ -#define P4D_SHIFT 39 -#define PTRS_PER_P4D 512 -#define P4D_SIZE (_AC(1, UL) << P4D_SHIFT) -#define P4D_MASK (~(P4D_SIZE - 1)) +#define P4D_SHIFT 39 +#define MAX_PTRS_PER_P4D 512 +#define PTRS_PER_P4D ptrs_per_p4d +#define P4D_SIZE (_AC(1, UL) << P4D_SHIFT) +#define P4D_MASK (~(P4D_SIZE - 1)) #define MAX_POSSIBLE_PHYSMEM_BITS 52 @@ -53,8 +57,9 @@ extern unsigned int pgtable_l5_enabled; /* * PGDIR_SHIFT determines what a top-level page table entry can map */ -#define PGDIR_SHIFT 39 -#define PTRS_PER_PGD 512 +#define PGDIR_SHIFT 39 +#define PTRS_PER_PGD 512 +#define MAX_PTRS_PER_P4D 1 #endif /* CONFIG_X86_5LEVEL */ diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 3a8e88a611eb..cbb3af721291 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1082,19 +1082,7 @@ void arch_unmap_kpfn(unsigned long pfn) * a legal address. */ -/* - * Build time check to see if we have a spare virtual bit. Don't want - * to leave this until run time because most developers don't have a - * system that can exercise this code path. This will only become a - * problem if/when we move beyond 5-level page tables. - * - * Hard code "9" here because cpp doesn't grok ilog2(PTRS_PER_PGD) - */ -#if PGDIR_SHIFT + 9 < 63 decoy_addr = (pfn << PAGE_SHIFT) + (PAGE_OFFSET ^ BIT(63)); -#else -#error "no unused virtual bit available" -#endif if (set_memory_np(decoy_addr, 1)) pr_warn("Could not invalidate pfn=0x%lx from 1:1 map\n", pfn); @@ -2328,6 +2316,12 @@ static __init int mcheck_init_device(void) { int err; + /* + * Check if we have a spare virtual bit. This will only become + * a problem if/when we move beyond 5-level page tables. + */ + MAYBE_BUILD_BUG_ON(__VIRTUAL_MASK_SHIFT >= 63); + if (!mce_available(&boot_cpu_data)) { err = -EIO; goto err_out; diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 17d00d1886de..98b0ff49b220 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -42,6 +42,10 @@ pmdval_t early_pmd_flags = __PAGE_KERNEL_LARGE & ~(_PAGE_GLOBAL | _PAGE_NX); #ifdef CONFIG_X86_5LEVEL unsigned int pgtable_l5_enabled __ro_after_init = 1; EXPORT_SYMBOL(pgtable_l5_enabled); +unsigned int pgdir_shift __ro_after_init = 48; +EXPORT_SYMBOL(pgdir_shift); +unsigned int ptrs_per_p4d __ro_after_init = 512; +EXPORT_SYMBOL(ptrs_per_p4d); #endif #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT @@ -336,7 +340,7 @@ asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data) BUILD_BUG_ON((__START_KERNEL_map & ~PMD_MASK) != 0); BUILD_BUG_ON((MODULES_VADDR & ~PMD_MASK) != 0); BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL)); - BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) == + MAYBE_BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) == (__START_KERNEL & PGDIR_MASK))); BUILD_BUG_ON(__fix_to_virt(__end_of_fixed_addresses) <= MODULES_END); diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index a89f2dbc3531..420058b05d39 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -428,14 +428,15 @@ static void walk_pud_level(struct seq_file *m, struct pg_state *st, p4d_t addr, #define p4d_none(a) pud_none(__pud(p4d_val(a))) #endif -#if PTRS_PER_P4D > 1 - static void walk_p4d_level(struct seq_file *m, struct pg_state *st, pgd_t addr, unsigned long P) { int i; p4d_t *start, *p4d_start; pgprotval_t prot; + if (PTRS_PER_P4D == 1) + return walk_pud_level(m, st, __p4d(pgd_val(addr)), P); + p4d_start = start = (p4d_t *)pgd_page_vaddr(addr); for (i = 0; i < PTRS_PER_P4D; i++) { @@ -455,11 +456,8 @@ static void walk_p4d_level(struct seq_file *m, struct pg_state *st, pgd_t addr, } } -#else -#define walk_p4d_level(m,s,a,p) walk_pud_level(m,s,__p4d(pgd_val(a)),p) -#define pgd_large(a) p4d_large(__p4d(pgd_val(a))) -#define pgd_none(a) p4d_none(__p4d(pgd_val(a))) -#endif +#define pgd_large(a) (pgtable_l5_enabled ? pgd_large(a) : p4d_large(__p4d(pgd_val(a)))) +#define pgd_none(a) (pgtable_l5_enabled ? pgd_none(a) : p4d_none(__p4d(pgd_val(a)))) static inline bool is_hypervisor_range(int idx) { diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 1ab42c852069..6a4b20bc7527 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -143,7 +143,7 @@ void sync_global_pgds(unsigned long start, unsigned long end) * With folded p4d, pgd_none() is always false, we need to * handle synchonization on p4d level. */ - BUILD_BUG_ON(pgd_none(*pgd_ref)); + MAYBE_BUILD_BUG_ON(pgd_none(*pgd_ref)); p4d_ref = p4d_offset(pgd_ref, addr); if (p4d_none(*p4d_ref)) diff --git a/arch/x86/mm/kasan_init_64.c b/arch/x86/mm/kasan_init_64.c index af6f2f9c6a26..12ec90f62457 100644 --- a/arch/x86/mm/kasan_init_64.c +++ b/arch/x86/mm/kasan_init_64.c @@ -19,7 +19,7 @@ extern struct range pfn_mapped[E820_MAX_ENTRIES]; -static p4d_t tmp_p4d_table[PTRS_PER_P4D] __initdata __aligned(PAGE_SIZE); +static p4d_t tmp_p4d_table[MAX_PTRS_PER_P4D] __initdata __aligned(PAGE_SIZE); static __init void *early_alloc(size_t size, int nid, bool panic) { diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 780460aa5ea5..d52aaa7dc088 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -257,8 +257,8 @@ void efi_sync_low_kernel_mappings(void) * only span a single PGD entry and that the entry also maps * other important kernel regions. */ - BUILD_BUG_ON(pgd_index(EFI_VA_END) != pgd_index(MODULES_END)); - BUILD_BUG_ON((EFI_VA_START & PGDIR_MASK) != + MAYBE_BUILD_BUG_ON(pgd_index(EFI_VA_END) != pgd_index(MODULES_END)); + MAYBE_BUILD_BUG_ON((EFI_VA_START & PGDIR_MASK) != (EFI_VA_END & PGDIR_MASK)); pgd_efi = efi_pgd + pgd_index(PAGE_OFFSET); diff --git a/include/asm-generic/5level-fixup.h b/include/asm-generic/5level-fixup.h index dfbd9d990637..9c2e0708eb82 100644 --- a/include/asm-generic/5level-fixup.h +++ b/include/asm-generic/5level-fixup.h @@ -8,6 +8,7 @@ #define P4D_SHIFT PGDIR_SHIFT #define P4D_SIZE PGDIR_SIZE #define P4D_MASK PGDIR_MASK +#define MAX_PTRS_PER_P4D 1 #define PTRS_PER_P4D 1 #define p4d_t pgd_t diff --git a/include/asm-generic/pgtable-nop4d.h b/include/asm-generic/pgtable-nop4d.h index 8f22f55de17a..1a29b2a0282b 100644 --- a/include/asm-generic/pgtable-nop4d.h +++ b/include/asm-generic/pgtable-nop4d.h @@ -8,10 +8,11 @@ typedef struct { pgd_t pgd; } p4d_t; -#define P4D_SHIFT PGDIR_SHIFT -#define PTRS_PER_P4D 1 -#define P4D_SIZE (1UL << P4D_SHIFT) -#define P4D_MASK (~(P4D_SIZE-1)) +#define P4D_SHIFT PGDIR_SHIFT +#define MAX_PTRS_PER_P4D 1 +#define PTRS_PER_P4D 1 +#define P4D_SIZE (1UL << P4D_SHIFT) +#define P4D_MASK (~(P4D_SIZE-1)) /* * The "pgd_xxx()" functions here are trivial for a folded two-level diff --git a/include/linux/kasan.h b/include/linux/kasan.h index adc13474a53b..d6459bd1376d 100644 --- a/include/linux/kasan.h +++ b/include/linux/kasan.h @@ -18,7 +18,7 @@ extern unsigned char kasan_zero_page[PAGE_SIZE]; extern pte_t kasan_zero_pte[PTRS_PER_PTE]; extern pmd_t kasan_zero_pmd[PTRS_PER_PMD]; extern pud_t kasan_zero_pud[PTRS_PER_PUD]; -extern p4d_t kasan_zero_p4d[PTRS_PER_P4D]; +extern p4d_t kasan_zero_p4d[MAX_PTRS_PER_P4D]; void kasan_populate_zero_shadow(const void *shadow_start, const void *shadow_end); diff --git a/mm/kasan/kasan_init.c b/mm/kasan/kasan_init.c index 554e4c0f23a2..f436246ccc79 100644 --- a/mm/kasan/kasan_init.c +++ b/mm/kasan/kasan_init.c @@ -31,7 +31,7 @@ unsigned char kasan_zero_page[PAGE_SIZE] __page_aligned_bss; #if CONFIG_PGTABLE_LEVELS > 4 -p4d_t kasan_zero_p4d[PTRS_PER_P4D] __page_aligned_bss; +p4d_t kasan_zero_p4d[MAX_PTRS_PER_P4D] __page_aligned_bss; #endif #if CONFIG_PGTABLE_LEVELS > 3 pud_t kasan_zero_pud[PTRS_PER_PUD] __page_aligned_bss; -- cgit v1.2.3-70-g09d2 From 09e61a779e7f171c50325e6d7108a593afb2e5d4 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 14:16:55 +0300 Subject: x86/mm: Make __VIRTUAL_MASK_SHIFT dynamic For boot-time switching between paging modes, we need to be able to adjust virtual mask shifts. The change doesn't affect the kernel image size much: text data bss dec hex filename 8628892 4734340 1368064 14731296 e0c820 vmlinux.before 8628966 4734340 1368064 14731370 e0c86a vmlinux.after Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214111656.88514-9-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/entry/entry_64.S | 12 ++++++++++++ arch/x86/include/asm/page_64_types.h | 2 +- arch/x86/mm/dump_pagetables.c | 12 ++++++++++-- arch/x86/mm/kaslr.c | 4 +++- 4 files changed, 26 insertions(+), 4 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 30c8c5344c4a..2c06348b7807 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -274,8 +274,20 @@ GLOBAL(entry_SYSCALL_64_after_hwframe) * Change top bits to match most significant bit (47th or 56th bit * depending on paging mode) in the address. */ +#ifdef CONFIG_X86_5LEVEL + testl $1, pgtable_l5_enabled(%rip) + jz 1f + shl $(64 - 57), %rcx + sar $(64 - 57), %rcx + jmp 2f +1: + shl $(64 - 48), %rcx + sar $(64 - 48), %rcx +2: +#else shl $(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx sar $(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx +#endif /* If this changed %rcx, it was not canonical */ cmpq %rcx, %r11 diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index d54a3d5b5b3b..fa7dc7cd8c19 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -56,7 +56,7 @@ #define __PHYSICAL_MASK_SHIFT 52 #ifdef CONFIG_X86_5LEVEL -#define __VIRTUAL_MASK_SHIFT 56 +#define __VIRTUAL_MASK_SHIFT (pgtable_l5_enabled ? 56 : 47) #else #define __VIRTUAL_MASK_SHIFT 47 #endif diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index 420058b05d39..9efee6f464ab 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -85,8 +85,12 @@ static struct addr_marker address_markers[] = { [VMALLOC_START_NR] = { 0UL, "vmalloc() Area" }, [VMEMMAP_START_NR] = { 0UL, "Vmemmap" }, #ifdef CONFIG_KASAN - [KASAN_SHADOW_START_NR] = { KASAN_SHADOW_START, "KASAN shadow" }, - [KASAN_SHADOW_END_NR] = { KASAN_SHADOW_END, "KASAN shadow end" }, + /* + * These fields get initialized with the (dynamic) + * KASAN_SHADOW_{START,END} values in pt_dump_init(). + */ + [KASAN_SHADOW_START_NR] = { 0UL, "KASAN shadow" }, + [KASAN_SHADOW_END_NR] = { 0UL, "KASAN shadow end" }, #endif #ifdef CONFIG_MODIFY_LDT_SYSCALL [LDT_NR] = { 0UL, "LDT remap" }, @@ -571,6 +575,10 @@ static int __init pt_dump_init(void) #ifdef CONFIG_MODIFY_LDT_SYSCALL address_markers[LDT_NR].start_address = LDT_BASE_ADDR; #endif +#ifdef CONFIG_KASAN + address_markers[KASAN_SHADOW_START_NR].start_address = KASAN_SHADOW_START; + address_markers[KASAN_SHADOW_END_NR].start_address = KASAN_SHADOW_END; +#endif #endif #ifdef CONFIG_X86_32 address_markers[VMALLOC_START_NR].start_address = VMALLOC_START; diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index 515b98a8ccee..d079878c6cbc 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -52,7 +52,7 @@ static __initdata struct kaslr_memory_region { unsigned long *base; unsigned long size_tb; } kaslr_regions[] = { - { &page_offset_base, 1 << (__PHYSICAL_MASK_SHIFT - TB_SHIFT) /* Maximum */ }, + { &page_offset_base, 0 }, { &vmalloc_base, VMALLOC_SIZE_TB }, { &vmemmap_base, 1 }, }; @@ -93,6 +93,8 @@ void __init kernel_randomize_memory(void) if (!kaslr_memory_enabled()) return; + kaslr_regions[0].size_tb = 1 << (__PHYSICAL_MASK_SHIFT - TB_SHIFT); + /* * Update Physical memory mapping to available and * add padding if needed (especially for memory hotplug support). -- cgit v1.2.3-70-g09d2 From 4fa5662b6b49611f11856db8be346710217473ef Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 21:25:36 +0300 Subject: x86/mm: Initialize 'page_offset_base' at boot-time For 4- and 5-level paging we have different 'page_offset_base'. Let's initialize it at boot-time accordingly to machine capability. We also have to split __PAGE_OFFSET_BASE into two constants -- for 4- and 5-level paging. Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Arjan van de Ven Cc: Borislav Petkov Cc: Dan Williams Cc: Dave Hansen Cc: David Woodhouse Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214182542.69302-4-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/page_64_types.h | 9 +++------ arch/x86/kernel/head64.c | 13 +++++++++---- arch/x86/kernel/head_64.S | 2 +- arch/x86/mm/kaslr.c | 8 ++++---- 4 files changed, 17 insertions(+), 15 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index fa7dc7cd8c19..2c5a966dc222 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -37,16 +37,13 @@ * hypervisor to fit. Choosing 16 slots here is arbitrary, but it's * what Xen requires. */ -#ifdef CONFIG_X86_5LEVEL -#define __PAGE_OFFSET_BASE _AC(0xff10000000000000, UL) -#else -#define __PAGE_OFFSET_BASE _AC(0xffff880000000000, UL) -#endif +#define __PAGE_OFFSET_BASE_L5 _AC(0xff10000000000000, UL) +#define __PAGE_OFFSET_BASE_L4 _AC(0xffff880000000000, UL) #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT #define __PAGE_OFFSET page_offset_base #else -#define __PAGE_OFFSET __PAGE_OFFSET_BASE +#define __PAGE_OFFSET __PAGE_OFFSET_BASE_L4 #endif /* CONFIG_DYNAMIC_MEMORY_LAYOUT */ #define __START_KERNEL_map _AC(0xffffffff80000000, UL) diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 8a0a485524da..876d3bf2b23a 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -49,7 +49,7 @@ EXPORT_SYMBOL(ptrs_per_p4d); #endif #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT -unsigned long page_offset_base __ro_after_init = __PAGE_OFFSET_BASE; +unsigned long page_offset_base __ro_after_init = __PAGE_OFFSET_BASE_L4; EXPORT_SYMBOL(page_offset_base); unsigned long vmalloc_base __ro_after_init = __VMALLOC_BASE; EXPORT_SYMBOL(vmalloc_base); @@ -64,6 +64,11 @@ static void __head *fixup_pointer(void *ptr, unsigned long physaddr) return ptr - (void *)_text + (void *)physaddr; } +static unsigned long __head *fixup_long(void *ptr, unsigned long physaddr) +{ + return fixup_pointer(ptr, physaddr); +} + #ifdef CONFIG_X86_5LEVEL static unsigned int __head *fixup_int(void *ptr, unsigned long physaddr) { @@ -81,6 +86,7 @@ static void __head check_la57_support(unsigned long physaddr) *fixup_int(&pgtable_l5_enabled, physaddr) = 1; *fixup_int(&pgdir_shift, physaddr) = 48; *fixup_int(&ptrs_per_p4d, physaddr) = 512; + *fixup_long(&page_offset_base, physaddr) = __PAGE_OFFSET_BASE_L5; } #else static void __head check_la57_support(unsigned long physaddr) {} @@ -89,7 +95,7 @@ static void __head check_la57_support(unsigned long physaddr) {} unsigned long __head __startup_64(unsigned long physaddr, struct boot_params *bp) { - unsigned long load_delta, *p; + unsigned long load_delta; unsigned long pgtable_flags; pgdval_t *pgd; p4dval_t *p4d; @@ -196,8 +202,7 @@ unsigned long __head __startup_64(unsigned long physaddr, * Fixup phys_base - remove the memory encryption mask to obtain * the true physical address. */ - p = fixup_pointer(&phys_base, physaddr); - *p += load_delta - sme_get_me_mask(); + *fixup_long(&phys_base, physaddr) += load_delta - sme_get_me_mask(); /* Encrypt the kernel and related (if SME is active) */ sme_encrypt_kernel(bp); diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 04a625f0fcda..d3f8b43d541a 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -41,7 +41,7 @@ #define pud_index(x) (((x) >> PUD_SHIFT) & (PTRS_PER_PUD-1)) #if defined(CONFIG_XEN_PV) || defined(CONFIG_XEN_PVH) -PGD_PAGE_OFFSET = pgd_index(__PAGE_OFFSET_BASE) +PGD_PAGE_OFFSET = pgd_index(__PAGE_OFFSET_BASE_L4) PGD_START_KERNEL = pgd_index(__START_KERNEL_map) #endif L3_START_KERNEL = pud_index(__START_KERNEL_map) diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index d079878c6cbc..7828a7ca3bba 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -34,13 +34,10 @@ #define TB_SHIFT 40 /* - * Virtual address start and end range for randomization. - * * The end address could depend on more configuration options to make the * highest amount of space for randomization available, but that's too hard * to keep straight and caused issues already. */ -static const unsigned long vaddr_start = __PAGE_OFFSET_BASE; static const unsigned long vaddr_end = CPU_ENTRY_AREA_BASE; /* @@ -76,11 +73,14 @@ static inline bool kaslr_memory_enabled(void) void __init kernel_randomize_memory(void) { size_t i; - unsigned long vaddr = vaddr_start; + unsigned long vaddr_start, vaddr; unsigned long rand, memory_tb; struct rnd_state rand_state; unsigned long remain_entropy; + vaddr_start = pgtable_l5_enabled ? __PAGE_OFFSET_BASE_L5 : __PAGE_OFFSET_BASE_L4; + vaddr = vaddr_start; + /* * These BUILD_BUG_ON checks ensure the memory layout is consistent * with the vaddr_start/vaddr_end variables. These checks are very -- cgit v1.2.3-70-g09d2 From a7412546d8cb5ad578805060b4006f2a021b5868 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 21:25:37 +0300 Subject: x86/mm: Adjust vmalloc base and size at boot-time vmalloc area has different placement and size depending on paging mode. Let's adjust it during early boot accodring to machine capability. Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Arjan van de Ven Cc: Borislav Petkov Cc: Dan Williams Cc: Dave Hansen Cc: David Woodhouse Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214182542.69302-5-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable_64_types.h | 16 ++++++++++------ arch/x86/kernel/head64.c | 3 ++- arch/x86/mm/kaslr.c | 3 ++- 3 files changed, 14 insertions(+), 8 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index 59d971c85de5..686329994ade 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -102,25 +102,29 @@ extern unsigned int ptrs_per_p4d; #define LDT_PGD_ENTRY (pgtable_l5_enabled ? LDT_PGD_ENTRY_L5 : LDT_PGD_ENTRY_L4) #define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT) +#define __VMALLOC_BASE_L4 0xffffc90000000000 +#define __VMALLOC_BASE_L5 0xffa0000000000000 + +#define VMALLOC_SIZE_TB_L4 32UL +#define VMALLOC_SIZE_TB_L5 12800UL + #ifdef CONFIG_X86_5LEVEL -# define VMALLOC_SIZE_TB _AC(12800, UL) -# define __VMALLOC_BASE _AC(0xffa0000000000000, UL) # define __VMEMMAP_BASE _AC(0xffd4000000000000, UL) #else -# define VMALLOC_SIZE_TB _AC(32, UL) -# define __VMALLOC_BASE _AC(0xffffc90000000000, UL) # define __VMEMMAP_BASE _AC(0xffffea0000000000, UL) #endif #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT # define VMALLOC_START vmalloc_base +# define VMALLOC_SIZE_TB (pgtable_l5_enabled ? VMALLOC_SIZE_TB_L5 : VMALLOC_SIZE_TB_L4) # define VMEMMAP_START vmemmap_base #else -# define VMALLOC_START __VMALLOC_BASE +# define VMALLOC_START __VMALLOC_BASE_L4 +# define VMALLOC_SIZE_TB VMALLOC_SIZE_TB_L4 # define VMEMMAP_START __VMEMMAP_BASE #endif /* CONFIG_DYNAMIC_MEMORY_LAYOUT */ -#define VMALLOC_END (VMALLOC_START + _AC((VMALLOC_SIZE_TB << 40) - 1, UL)) +#define VMALLOC_END (VMALLOC_START + (VMALLOC_SIZE_TB << 40) - 1) #define MODULES_VADDR (__START_KERNEL_map + KERNEL_IMAGE_SIZE) /* The module sections ends with the start of the fixmap */ diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 876d3bf2b23a..22bf2015254c 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -51,7 +51,7 @@ EXPORT_SYMBOL(ptrs_per_p4d); #ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT unsigned long page_offset_base __ro_after_init = __PAGE_OFFSET_BASE_L4; EXPORT_SYMBOL(page_offset_base); -unsigned long vmalloc_base __ro_after_init = __VMALLOC_BASE; +unsigned long vmalloc_base __ro_after_init = __VMALLOC_BASE_L4; EXPORT_SYMBOL(vmalloc_base); unsigned long vmemmap_base __ro_after_init = __VMEMMAP_BASE; EXPORT_SYMBOL(vmemmap_base); @@ -87,6 +87,7 @@ static void __head check_la57_support(unsigned long physaddr) *fixup_int(&pgdir_shift, physaddr) = 48; *fixup_int(&ptrs_per_p4d, physaddr) = 512; *fixup_long(&page_offset_base, physaddr) = __PAGE_OFFSET_BASE_L5; + *fixup_long(&vmalloc_base, physaddr) = __VMALLOC_BASE_L5; } #else static void __head check_la57_support(unsigned long physaddr) {} diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index 7828a7ca3bba..641169d38184 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -50,7 +50,7 @@ static __initdata struct kaslr_memory_region { unsigned long size_tb; } kaslr_regions[] = { { &page_offset_base, 0 }, - { &vmalloc_base, VMALLOC_SIZE_TB }, + { &vmalloc_base, 0 }, { &vmemmap_base, 1 }, }; @@ -94,6 +94,7 @@ void __init kernel_randomize_memory(void) return; kaslr_regions[0].size_tb = 1 << (__PHYSICAL_MASK_SHIFT - TB_SHIFT); + kaslr_regions[1].size_tb = VMALLOC_SIZE_TB; /* * Update Physical memory mapping to available and -- cgit v1.2.3-70-g09d2 From 91f606a8fa68264224cbc76888fa8649cdbe9990 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 14 Feb 2018 21:25:41 +0300 Subject: x86/mm: Replace compile-time checks for 5-level paging with runtime-time checks This patch converts the of CONFIG_X86_5LEVEL check to runtime checks for p4d folding. Signed-off-by: Kirill A. Shutemov Cc: Andy Lutomirski Cc: Arjan van de Ven Cc: Borislav Petkov Cc: Dan Williams Cc: Dave Hansen Cc: David Woodhouse Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180214182542.69302-9-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pgtable_64.h | 23 ++++++++++------------- arch/x86/mm/dump_pagetables.c | 4 +--- arch/x86/mm/fault.c | 4 ++-- arch/x86/mm/ident_map.c | 2 +- arch/x86/mm/init_64.c | 30 ++++++++++++++++++------------ arch/x86/mm/kasan_init_64.c | 12 ++++++------ arch/x86/mm/kaslr.c | 6 +++--- arch/x86/mm/tlb.c | 2 +- arch/x86/platform/efi/efi_64.c | 2 +- arch/x86/power/hibernate_64.c | 6 +++--- 10 files changed, 46 insertions(+), 45 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index 81462e9a34f6..81dda8d1d0bd 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -217,29 +217,26 @@ static inline pgd_t pti_set_user_pgd(pgd_t *pgdp, pgd_t pgd) static inline void native_set_p4d(p4d_t *p4dp, p4d_t p4d) { -#if defined(CONFIG_PAGE_TABLE_ISOLATION) && !defined(CONFIG_X86_5LEVEL) - p4dp->pgd = pti_set_user_pgd(&p4dp->pgd, p4d.pgd); -#else - *p4dp = p4d; -#endif + pgd_t pgd; + + if (pgtable_l5_enabled || !IS_ENABLED(CONFIG_PAGE_TABLE_ISOLATION)) { + *p4dp = p4d; + return; + } + + pgd = native_make_pgd(p4d_val(p4d)); + pgd = pti_set_user_pgd((pgd_t *)p4dp, pgd); + *p4dp = native_make_p4d(pgd_val(pgd)); } static inline void native_p4d_clear(p4d_t *p4d) { -#ifdef CONFIG_X86_5LEVEL native_set_p4d(p4d, native_make_p4d(0)); -#else - native_set_p4d(p4d, (p4d_t) { .pgd = native_make_pgd(0)}); -#endif } static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd) { -#ifdef CONFIG_PAGE_TABLE_ISOLATION *pgdp = pti_set_user_pgd(pgdp, pgd); -#else - *pgdp = pgd; -#endif } static inline void native_pgd_clear(pgd_t *pgd) diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index 9efee6f464ab..0d6d67d18ad6 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -348,9 +348,7 @@ static inline bool kasan_page_table(struct seq_file *m, struct pg_state *st, void *pt) { if (__pa(pt) == __pa(kasan_zero_pmd) || -#ifdef CONFIG_X86_5LEVEL - __pa(pt) == __pa(kasan_zero_p4d) || -#endif + (pgtable_l5_enabled && __pa(pt) == __pa(kasan_zero_p4d)) || __pa(pt) == __pa(kasan_zero_pud)) { pgprotval_t prot = pte_flags(kasan_zero_pte[0]); note_page(m, st, __pgprot(prot), 5); diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 800de815519c..321b78060e93 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -439,7 +439,7 @@ static noinline int vmalloc_fault(unsigned long address) if (pgd_none(*pgd_ref)) return -1; - if (CONFIG_PGTABLE_LEVELS > 4) { + if (pgtable_l5_enabled) { if (pgd_none(*pgd)) { set_pgd(pgd, *pgd_ref); arch_flush_lazy_mmu_mode(); @@ -454,7 +454,7 @@ static noinline int vmalloc_fault(unsigned long address) if (p4d_none(*p4d_ref)) return -1; - if (p4d_none(*p4d) && CONFIG_PGTABLE_LEVELS == 4) { + if (p4d_none(*p4d) && !pgtable_l5_enabled) { set_p4d(p4d, *p4d_ref); arch_flush_lazy_mmu_mode(); } else { diff --git a/arch/x86/mm/ident_map.c b/arch/x86/mm/ident_map.c index ab33a32df2a8..9aa22be8331e 100644 --- a/arch/x86/mm/ident_map.c +++ b/arch/x86/mm/ident_map.c @@ -120,7 +120,7 @@ int kernel_ident_mapping_init(struct x86_mapping_info *info, pgd_t *pgd_page, result = ident_p4d_init(info, p4d, addr, next); if (result) return result; - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + if (pgtable_l5_enabled) { set_pgd(pgd, __pgd(__pa(p4d) | info->kernpg_flag)); } else { /* diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 6a4b20bc7527..3186e6836036 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -88,12 +88,7 @@ static int __init nonx32_setup(char *str) } __setup("noexec32=", nonx32_setup); -/* - * When memory was added make sure all the processes MM have - * suitable PGD entries in the local PGD level page. - */ -#ifdef CONFIG_X86_5LEVEL -void sync_global_pgds(unsigned long start, unsigned long end) +static void sync_global_pgds_l5(unsigned long start, unsigned long end) { unsigned long addr; @@ -129,8 +124,8 @@ void sync_global_pgds(unsigned long start, unsigned long end) spin_unlock(&pgd_lock); } } -#else -void sync_global_pgds(unsigned long start, unsigned long end) + +static void sync_global_pgds_l4(unsigned long start, unsigned long end) { unsigned long addr; @@ -173,7 +168,18 @@ void sync_global_pgds(unsigned long start, unsigned long end) spin_unlock(&pgd_lock); } } -#endif + +/* + * When memory was added make sure all the processes MM have + * suitable PGD entries in the local PGD level page. + */ +void sync_global_pgds(unsigned long start, unsigned long end) +{ + if (pgtable_l5_enabled) + sync_global_pgds_l5(start, end); + else + sync_global_pgds_l4(start, end); +} /* * NOTE: This function is marked __ref because it calls __init function @@ -632,7 +638,7 @@ phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end, unsigned long vaddr = (unsigned long)__va(paddr); int i = p4d_index(vaddr); - if (!IS_ENABLED(CONFIG_X86_5LEVEL)) + if (!pgtable_l5_enabled) return phys_pud_init((pud_t *) p4d_page, paddr, paddr_end, page_size_mask); for (; i < PTRS_PER_P4D; i++, paddr = paddr_next) { @@ -712,7 +718,7 @@ kernel_physical_mapping_init(unsigned long paddr_start, page_size_mask); spin_lock(&init_mm.page_table_lock); - if (IS_ENABLED(CONFIG_X86_5LEVEL)) + if (pgtable_l5_enabled) pgd_populate(&init_mm, pgd, p4d); else p4d_populate(&init_mm, p4d_offset(pgd, vaddr), (pud_t *) p4d); @@ -1093,7 +1099,7 @@ remove_p4d_table(p4d_t *p4d_start, unsigned long addr, unsigned long end, * 5-level case we should free them. This code will have to change * to adapt for boot-time switching between 4 and 5 level page tables. */ - if (CONFIG_PGTABLE_LEVELS == 5) + if (pgtable_l5_enabled) free_pud_table(pud_base, p4d, altmap); } diff --git a/arch/x86/mm/kasan_init_64.c b/arch/x86/mm/kasan_init_64.c index 12ec90f62457..0df0dd13a71d 100644 --- a/arch/x86/mm/kasan_init_64.c +++ b/arch/x86/mm/kasan_init_64.c @@ -176,10 +176,10 @@ static void __init clear_pgds(unsigned long start, * With folded p4d, pgd_clear() is nop, use p4d_clear() * instead. */ - if (CONFIG_PGTABLE_LEVELS < 5) - p4d_clear(p4d_offset(pgd, start)); - else + if (pgtable_l5_enabled) pgd_clear(pgd); + else + p4d_clear(p4d_offset(pgd, start)); } pgd = pgd_offset_k(start); @@ -191,7 +191,7 @@ static inline p4d_t *early_p4d_offset(pgd_t *pgd, unsigned long addr) { unsigned long p4d; - if (!IS_ENABLED(CONFIG_X86_5LEVEL)) + if (!pgtable_l5_enabled) return (p4d_t *)pgd; p4d = __pa_nodebug(pgd_val(*pgd)) & PTE_PFN_MASK; @@ -272,7 +272,7 @@ void __init kasan_early_init(void) for (i = 0; i < PTRS_PER_PUD; i++) kasan_zero_pud[i] = __pud(pud_val); - for (i = 0; IS_ENABLED(CONFIG_X86_5LEVEL) && i < PTRS_PER_P4D; i++) + for (i = 0; pgtable_l5_enabled && i < PTRS_PER_P4D; i++) kasan_zero_p4d[i] = __p4d(p4d_val); kasan_map_early_shadow(early_top_pgt); @@ -303,7 +303,7 @@ void __init kasan_init(void) * bunch of things like kernel code, modules, EFI mapping, etc. * We need to take extra steps to not overwrite them. */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + if (pgtable_l5_enabled) { void *ptr; ptr = (void *)pgd_page_vaddr(*pgd_offset_k(KASAN_SHADOW_END)); diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c index 641169d38184..615cc03ced84 100644 --- a/arch/x86/mm/kaslr.c +++ b/arch/x86/mm/kaslr.c @@ -124,7 +124,7 @@ void __init kernel_randomize_memory(void) */ entropy = remain_entropy / (ARRAY_SIZE(kaslr_regions) - i); prandom_bytes_state(&rand_state, &rand, sizeof(rand)); - if (IS_ENABLED(CONFIG_X86_5LEVEL)) + if (pgtable_l5_enabled) entropy = (rand % (entropy + 1)) & P4D_MASK; else entropy = (rand % (entropy + 1)) & PUD_MASK; @@ -136,7 +136,7 @@ void __init kernel_randomize_memory(void) * randomization alignment. */ vaddr += get_padding(&kaslr_regions[i]); - if (IS_ENABLED(CONFIG_X86_5LEVEL)) + if (pgtable_l5_enabled) vaddr = round_up(vaddr + 1, P4D_SIZE); else vaddr = round_up(vaddr + 1, PUD_SIZE); @@ -212,7 +212,7 @@ void __meminit init_trampoline(void) return; } - if (IS_ENABLED(CONFIG_X86_5LEVEL)) + if (pgtable_l5_enabled) init_trampoline_p4d(); else init_trampoline_pud(); diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 6550d37d0f06..92cb8a901c36 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -157,7 +157,7 @@ static void sync_current_stack_to_mm(struct mm_struct *mm) unsigned long sp = current_stack_pointer; pgd_t *pgd = pgd_offset(mm, sp); - if (CONFIG_PGTABLE_LEVELS > 4) { + if (pgtable_l5_enabled) { if (unlikely(pgd_none(*pgd))) { pgd_t *pgd_ref = pgd_offset_k(sp); diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index d52aaa7dc088..4845871a2006 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -227,7 +227,7 @@ int __init efi_alloc_page_tables(void) pud = pud_alloc(&init_mm, p4d, EFI_VA_END); if (!pud) { - if (CONFIG_PGTABLE_LEVELS > 4) + if (pgtable_l5_enabled) free_page((unsigned long) pgd_page_vaddr(*pgd)); free_page((unsigned long)efi_pgd); return -ENOMEM; diff --git a/arch/x86/power/hibernate_64.c b/arch/x86/power/hibernate_64.c index 0ef5e5204968..74a532989308 100644 --- a/arch/x86/power/hibernate_64.c +++ b/arch/x86/power/hibernate_64.c @@ -50,7 +50,7 @@ static int set_up_temporary_text_mapping(pgd_t *pgd) { pmd_t *pmd; pud_t *pud; - p4d_t *p4d; + p4d_t *p4d = NULL; /* * The new mapping only has to cover the page containing the image @@ -66,7 +66,7 @@ static int set_up_temporary_text_mapping(pgd_t *pgd) * tables used by the image kernel. */ - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + if (pgtable_l5_enabled) { p4d = (p4d_t *)get_safe_page(GFP_ATOMIC); if (!p4d) return -ENOMEM; @@ -84,7 +84,7 @@ static int set_up_temporary_text_mapping(pgd_t *pgd) __pmd((jump_address_phys & PMD_MASK) | __PAGE_KERNEL_LARGE_EXEC)); set_pud(pud + pud_index(restore_jump_address), __pud(__pa(pmd) | _KERNPG_TABLE)); - if (IS_ENABLED(CONFIG_X86_5LEVEL)) { + if (p4d) { set_p4d(p4d + p4d_index(restore_jump_address), __p4d(__pa(pud) | _KERNPG_TABLE)); set_pgd(pgd + pgd_index(restore_jump_address), __pgd(__pa(p4d) | _KERNPG_TABLE)); } else { -- cgit v1.2.3-70-g09d2 From 39b9552281abfcdfc54162897018890dafe7ffef Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 16 Feb 2018 14:49:48 +0300 Subject: x86/mm: Optimize boot-time paging mode switching cost By this point we have functioning boot-time switching between 4- and 5-level paging mode. But naive approach comes with cost. Numbers below are for kernel build, allmodconfig, 5 times. CONFIG_X86_5LEVEL=n: Performance counter stats for 'sh -c make -j100 -B -k >/dev/null' (5 runs): 17308719.892691 task-clock:u (msec) # 26.772 CPUs utilized ( +- 0.11% ) 0 context-switches:u # 0.000 K/sec 0 cpu-migrations:u # 0.000 K/sec 331,993,164 page-faults:u # 0.019 M/sec ( +- 0.01% ) 43,614,978,867,455 cycles:u # 2.520 GHz ( +- 0.01% ) 39,371,534,575,126 stalled-cycles-frontend:u # 90.27% frontend cycles idle ( +- 0.09% ) 28,363,350,152,428 instructions:u # 0.65 insn per cycle # 1.39 stalled cycles per insn ( +- 0.00% ) 6,316,784,066,413 branches:u # 364.948 M/sec ( +- 0.00% ) 250,808,144,781 branch-misses:u # 3.97% of all branches ( +- 0.01% ) 646.531974142 seconds time elapsed ( +- 1.15% ) CONFIG_X86_5LEVEL=y: Performance counter stats for 'sh -c make -j100 -B -k >/dev/null' (5 runs): 17411536.780625 task-clock:u (msec) # 26.426 CPUs utilized ( +- 0.10% ) 0 context-switches:u # 0.000 K/sec 0 cpu-migrations:u # 0.000 K/sec 331,868,663 page-faults:u # 0.019 M/sec ( +- 0.01% ) 43,865,909,056,301 cycles:u # 2.519 GHz ( +- 0.01% ) 39,740,130,365,581 stalled-cycles-frontend:u # 90.59% frontend cycles idle ( +- 0.05% ) 28,363,358,997,959 instructions:u # 0.65 insn per cycle # 1.40 stalled cycles per insn ( +- 0.00% ) 6,316,784,937,460 branches:u # 362.793 M/sec ( +- 0.00% ) 251,531,919,485 branch-misses:u # 3.98% of all branches ( +- 0.00% ) 658.886307752 seconds time elapsed ( +- 0.92% ) The patch tries to fix the performance regression by using cpu_feature_enabled(X86_FEATURE_LA57) instead of pgtable_l5_enabled in all hot code paths. These will statically patch the target code for additional performance. CONFIG_X86_5LEVEL=y + the patch: Performance counter stats for 'sh -c make -j100 -B -k >/dev/null' (5 runs): 17381990.268506 task-clock:u (msec) # 26.907 CPUs utilized ( +- 0.19% ) 0 context-switches:u # 0.000 K/sec 0 cpu-migrations:u # 0.000 K/sec 331,862,625 page-faults:u # 0.019 M/sec ( +- 0.01% ) 43,697,726,320,051 cycles:u # 2.514 GHz ( +- 0.03% ) 39,480,408,690,401 stalled-cycles-frontend:u # 90.35% frontend cycles idle ( +- 0.05% ) 28,363,394,221,388 instructions:u # 0.65 insn per cycle # 1.39 stalled cycles per insn ( +- 0.00% ) 6,316,794,985,573 branches:u # 363.410 M/sec ( +- 0.00% ) 251,013,232,547 branch-misses:u # 3.97% of all branches ( +- 0.01% ) 645.991174661 seconds time elapsed ( +- 1.19% ) Unfortunately, this approach doesn't help with text size: vmlinux.before .text size: 8190319 vmlinux.after .text size: 8200623 The .text section is increased by about 4k. Not sure if we can do anything about this. Signed-off-by: Kirill A. Shuemov Cc: Andy Lutomirski Cc: Andy Lutomirski Cc: Arjan van de Ven Cc: Borislav Petkov Cc: Borislav Petkov Cc: Dan Williams Cc: Dave Hansen Cc: David Woodhouse Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20180216114948.68868-4-kirill.shutemov@linux.intel.com Signed-off-by: Ingo Molnar --- arch/x86/boot/compressed/misc.h | 5 +++++ arch/x86/entry/entry_64.S | 11 ++--------- arch/x86/include/asm/pgtable_64_types.h | 5 ++++- arch/x86/kernel/head64.c | 9 +++++++-- arch/x86/kernel/head_64.S | 2 +- arch/x86/mm/kasan_init_64.c | 6 ++++++ 6 files changed, 25 insertions(+), 13 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h index 9d323dc6b159..4d369c308ed7 100644 --- a/arch/x86/boot/compressed/misc.h +++ b/arch/x86/boot/compressed/misc.h @@ -12,6 +12,11 @@ #undef CONFIG_PARAVIRT_SPINLOCKS #undef CONFIG_KASAN +#ifdef CONFIG_X86_5LEVEL +/* cpu_feature_enabled() cannot be used that early */ +#define pgtable_l5_enabled __pgtable_l5_enabled +#endif + #include #include #include diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S index 2c06348b7807..b18acdff9c3f 100644 --- a/arch/x86/entry/entry_64.S +++ b/arch/x86/entry/entry_64.S @@ -275,15 +275,8 @@ GLOBAL(entry_SYSCALL_64_after_hwframe) * depending on paging mode) in the address. */ #ifdef CONFIG_X86_5LEVEL - testl $1, pgtable_l5_enabled(%rip) - jz 1f - shl $(64 - 57), %rcx - sar $(64 - 57), %rcx - jmp 2f -1: - shl $(64 - 48), %rcx - sar $(64 - 48), %rcx -2: + ALTERNATIVE "shl $(64 - 48), %rcx; sar $(64 - 48), %rcx", \ + "shl $(64 - 57), %rcx; sar $(64 - 57), %rcx", X86_FEATURE_LA57 #else shl $(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx sar $(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h index 68909a68e5b9..d5c21a382475 100644 --- a/arch/x86/include/asm/pgtable_64_types.h +++ b/arch/x86/include/asm/pgtable_64_types.h @@ -21,7 +21,10 @@ typedef unsigned long pgprotval_t; typedef struct { pteval_t pte; } pte_t; #ifdef CONFIG_X86_5LEVEL -extern unsigned int pgtable_l5_enabled; +extern unsigned int __pgtable_l5_enabled; +#ifndef pgtable_l5_enabled +#define pgtable_l5_enabled cpu_feature_enabled(X86_FEATURE_LA57) +#endif #else #define pgtable_l5_enabled 0 #endif diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 8161e719a20f..0c855deee165 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -32,6 +32,11 @@ #include #include +#ifdef CONFIG_X86_5LEVEL +#undef pgtable_l5_enabled +#define pgtable_l5_enabled __pgtable_l5_enabled +#endif + /* * Manage page tables very early on. */ @@ -40,8 +45,8 @@ static unsigned int __initdata next_early_pgt; pmdval_t early_pmd_flags = __PAGE_KERNEL_LARGE & ~(_PAGE_GLOBAL | _PAGE_NX); #ifdef CONFIG_X86_5LEVEL -unsigned int pgtable_l5_enabled __ro_after_init; -EXPORT_SYMBOL(pgtable_l5_enabled); +unsigned int __pgtable_l5_enabled __ro_after_init; +EXPORT_SYMBOL(__pgtable_l5_enabled); unsigned int pgdir_shift __ro_after_init = 39; EXPORT_SYMBOL(pgdir_shift); unsigned int ptrs_per_p4d __ro_after_init = 1; diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 3e9de0fc97de..326c63129417 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -124,7 +124,7 @@ ENTRY(secondary_startup_64) /* Enable PAE mode, PGE and LA57 */ movl $(X86_CR4_PAE | X86_CR4_PGE), %ecx #ifdef CONFIG_X86_5LEVEL - testl $1, pgtable_l5_enabled(%rip) + testl $1, __pgtable_l5_enabled(%rip) jz 1f orl $X86_CR4_LA57, %ecx 1: diff --git a/arch/x86/mm/kasan_init_64.c b/arch/x86/mm/kasan_init_64.c index 0df0dd13a71d..d8ff013ea9d0 100644 --- a/arch/x86/mm/kasan_init_64.c +++ b/arch/x86/mm/kasan_init_64.c @@ -1,6 +1,12 @@ // SPDX-License-Identifier: GPL-2.0 #define DISABLE_BRANCH_PROFILING #define pr_fmt(fmt) "kasan: " fmt + +#ifdef CONFIG_X86_5LEVEL +/* Too early to use cpu_feature_enabled() */ +#define pgtable_l5_enabled __pgtable_l5_enabled +#endif + #include #include #include -- cgit v1.2.3-70-g09d2 From 672c0ae09b33a11d8f31fc61526632e96301164c Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Fri, 23 Feb 2018 01:27:37 -0700 Subject: x86/mm: Consider effective protection attributes in W+X check Using just the leaf page table entry flags would cause a false warning in case _PAGE_RW is clear or _PAGE_NX is set in a higher level entry. Hand through both the current entry's flags as well as the accumulated effective value (the latter as pgprotval_t instead of pgprot_t, as it's not an actual entry's value). This in particular eliminates the false W+X warning when running under Xen, as commit: 2cc42bac1c ("x86-64/Xen: eliminate W+X mappings") had to make the necessary adjustment in L2 rather than L1 (the reason is explained there). I.e. _PAGE_RW is clear there in L1, but _PAGE_NX is set in L2. Signed-off-by: Jan Beulich Reviewed-by: Juergen Gross Cc: Alexander Potapenko Cc: Andrey Ryabinin Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: Dmitry Vyukov Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/5A8FDE8902000078001AABBB@prv-mh.provo.novell.com Signed-off-by: Ingo Molnar --- arch/x86/mm/dump_pagetables.c | 94 ++++++++++++++++++++++++++----------------- 1 file changed, 58 insertions(+), 36 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c index 0d6d67d18ad6..62a7e9f65dec 100644 --- a/arch/x86/mm/dump_pagetables.c +++ b/arch/x86/mm/dump_pagetables.c @@ -29,6 +29,7 @@ struct pg_state { int level; pgprot_t current_prot; + pgprotval_t effective_prot; unsigned long start_address; unsigned long current_address; const struct addr_marker *marker; @@ -235,9 +236,9 @@ static unsigned long normalize_addr(unsigned long u) * print what we collected so far. */ static void note_page(struct seq_file *m, struct pg_state *st, - pgprot_t new_prot, int level) + pgprot_t new_prot, pgprotval_t new_eff, int level) { - pgprotval_t prot, cur; + pgprotval_t prot, cur, eff; static const char units[] = "BKMGTPE"; /* @@ -247,23 +248,24 @@ static void note_page(struct seq_file *m, struct pg_state *st, */ prot = pgprot_val(new_prot); cur = pgprot_val(st->current_prot); + eff = st->effective_prot; if (!st->level) { /* First entry */ st->current_prot = new_prot; + st->effective_prot = new_eff; st->level = level; st->marker = address_markers; st->lines = 0; pt_dump_seq_printf(m, st->to_dmesg, "---[ %s ]---\n", st->marker->name); - } else if (prot != cur || level != st->level || + } else if (prot != cur || new_eff != eff || level != st->level || st->current_address >= st->marker[1].start_address) { const char *unit = units; unsigned long delta; int width = sizeof(unsigned long) * 2; - pgprotval_t pr = pgprot_val(st->current_prot); - if (st->check_wx && (pr & _PAGE_RW) && !(pr & _PAGE_NX)) { + if (st->check_wx && (eff & _PAGE_RW) && !(eff & _PAGE_NX)) { WARN_ONCE(1, "x86/mm: Found insecure W+X mapping at address %p/%pS\n", (void *)st->start_address, @@ -317,21 +319,30 @@ static void note_page(struct seq_file *m, struct pg_state *st, st->start_address = st->current_address; st->current_prot = new_prot; + st->effective_prot = new_eff; st->level = level; } } -static void walk_pte_level(struct seq_file *m, struct pg_state *st, pmd_t addr, unsigned long P) +static inline pgprotval_t effective_prot(pgprotval_t prot1, pgprotval_t prot2) +{ + return (prot1 & prot2 & (_PAGE_USER | _PAGE_RW)) | + ((prot1 | prot2) & _PAGE_NX); +} + +static void walk_pte_level(struct seq_file *m, struct pg_state *st, pmd_t addr, + pgprotval_t eff_in, unsigned long P) { int i; pte_t *start; - pgprotval_t prot; + pgprotval_t prot, eff; start = (pte_t *)pmd_page_vaddr(addr); for (i = 0; i < PTRS_PER_PTE; i++) { prot = pte_flags(*start); + eff = effective_prot(eff_in, prot); st->current_address = normalize_addr(P + i * PTE_LEVEL_MULT); - note_page(m, st, __pgprot(prot), 5); + note_page(m, st, __pgprot(prot), eff, 5); start++; } } @@ -351,7 +362,7 @@ static inline bool kasan_page_table(struct seq_file *m, struct pg_state *st, (pgtable_l5_enabled && __pa(pt) == __pa(kasan_zero_p4d)) || __pa(pt) == __pa(kasan_zero_pud)) { pgprotval_t prot = pte_flags(kasan_zero_pte[0]); - note_page(m, st, __pgprot(prot), 5); + note_page(m, st, __pgprot(prot), 0, 5); return true; } return false; @@ -366,42 +377,45 @@ static inline bool kasan_page_table(struct seq_file *m, struct pg_state *st, #if PTRS_PER_PMD > 1 -static void walk_pmd_level(struct seq_file *m, struct pg_state *st, pud_t addr, unsigned long P) +static void walk_pmd_level(struct seq_file *m, struct pg_state *st, pud_t addr, + pgprotval_t eff_in, unsigned long P) { int i; pmd_t *start, *pmd_start; - pgprotval_t prot; + pgprotval_t prot, eff; pmd_start = start = (pmd_t *)pud_page_vaddr(addr); for (i = 0; i < PTRS_PER_PMD; i++) { st->current_address = normalize_addr(P + i * PMD_LEVEL_MULT); if (!pmd_none(*start)) { + prot = pmd_flags(*start); + eff = effective_prot(eff_in, prot); if (pmd_large(*start) || !pmd_present(*start)) { - prot = pmd_flags(*start); - note_page(m, st, __pgprot(prot), 4); + note_page(m, st, __pgprot(prot), eff, 4); } else if (!kasan_page_table(m, st, pmd_start)) { - walk_pte_level(m, st, *start, + walk_pte_level(m, st, *start, eff, P + i * PMD_LEVEL_MULT); } } else - note_page(m, st, __pgprot(0), 4); + note_page(m, st, __pgprot(0), 0, 4); start++; } } #else -#define walk_pmd_level(m,s,a,p) walk_pte_level(m,s,__pmd(pud_val(a)),p) +#define walk_pmd_level(m,s,a,e,p) walk_pte_level(m,s,__pmd(pud_val(a)),e,p) #define pud_large(a) pmd_large(__pmd(pud_val(a))) #define pud_none(a) pmd_none(__pmd(pud_val(a))) #endif #if PTRS_PER_PUD > 1 -static void walk_pud_level(struct seq_file *m, struct pg_state *st, p4d_t addr, unsigned long P) +static void walk_pud_level(struct seq_file *m, struct pg_state *st, p4d_t addr, + pgprotval_t eff_in, unsigned long P) { int i; pud_t *start, *pud_start; - pgprotval_t prot; + pgprotval_t prot, eff; pud_t *prev_pud = NULL; pud_start = start = (pud_t *)p4d_page_vaddr(addr); @@ -409,15 +423,16 @@ static void walk_pud_level(struct seq_file *m, struct pg_state *st, p4d_t addr, for (i = 0; i < PTRS_PER_PUD; i++) { st->current_address = normalize_addr(P + i * PUD_LEVEL_MULT); if (!pud_none(*start)) { + prot = pud_flags(*start); + eff = effective_prot(eff_in, prot); if (pud_large(*start) || !pud_present(*start)) { - prot = pud_flags(*start); - note_page(m, st, __pgprot(prot), 3); + note_page(m, st, __pgprot(prot), eff, 3); } else if (!kasan_page_table(m, st, pud_start)) { - walk_pmd_level(m, st, *start, + walk_pmd_level(m, st, *start, eff, P + i * PUD_LEVEL_MULT); } } else - note_page(m, st, __pgprot(0), 3); + note_page(m, st, __pgprot(0), 0, 3); prev_pud = start; start++; @@ -425,34 +440,36 @@ static void walk_pud_level(struct seq_file *m, struct pg_state *st, p4d_t addr, } #else -#define walk_pud_level(m,s,a,p) walk_pmd_level(m,s,__pud(p4d_val(a)),p) +#define walk_pud_level(m,s,a,e,p) walk_pmd_level(m,s,__pud(p4d_val(a)),e,p) #define p4d_large(a) pud_large(__pud(p4d_val(a))) #define p4d_none(a) pud_none(__pud(p4d_val(a))) #endif -static void walk_p4d_level(struct seq_file *m, struct pg_state *st, pgd_t addr, unsigned long P) +static void walk_p4d_level(struct seq_file *m, struct pg_state *st, pgd_t addr, + pgprotval_t eff_in, unsigned long P) { int i; p4d_t *start, *p4d_start; - pgprotval_t prot; + pgprotval_t prot, eff; if (PTRS_PER_P4D == 1) - return walk_pud_level(m, st, __p4d(pgd_val(addr)), P); + return walk_pud_level(m, st, __p4d(pgd_val(addr)), eff_in, P); p4d_start = start = (p4d_t *)pgd_page_vaddr(addr); for (i = 0; i < PTRS_PER_P4D; i++) { st->current_address = normalize_addr(P + i * P4D_LEVEL_MULT); if (!p4d_none(*start)) { + prot = p4d_flags(*start); + eff = effective_prot(eff_in, prot); if (p4d_large(*start) || !p4d_present(*start)) { - prot = p4d_flags(*start); - note_page(m, st, __pgprot(prot), 2); + note_page(m, st, __pgprot(prot), eff, 2); } else if (!kasan_page_table(m, st, p4d_start)) { - walk_pud_level(m, st, *start, + walk_pud_level(m, st, *start, eff, P + i * P4D_LEVEL_MULT); } } else - note_page(m, st, __pgprot(0), 2); + note_page(m, st, __pgprot(0), 0, 2); start++; } @@ -483,7 +500,7 @@ static void ptdump_walk_pgd_level_core(struct seq_file *m, pgd_t *pgd, #else pgd_t *start = swapper_pg_dir; #endif - pgprotval_t prot; + pgprotval_t prot, eff; int i; struct pg_state st = {}; @@ -499,15 +516,20 @@ static void ptdump_walk_pgd_level_core(struct seq_file *m, pgd_t *pgd, for (i = 0; i < PTRS_PER_PGD; i++) { st.current_address = normalize_addr(i * PGD_LEVEL_MULT); if (!pgd_none(*start) && !is_hypervisor_range(i)) { + prot = pgd_flags(*start); +#ifdef CONFIG_X86_PAE + eff = _PAGE_USER | _PAGE_RW; +#else + eff = prot; +#endif if (pgd_large(*start) || !pgd_present(*start)) { - prot = pgd_flags(*start); - note_page(m, &st, __pgprot(prot), 1); + note_page(m, &st, __pgprot(prot), eff, 1); } else { - walk_p4d_level(m, &st, *start, + walk_p4d_level(m, &st, *start, eff, i * PGD_LEVEL_MULT); } } else - note_page(m, &st, __pgprot(0), 1); + note_page(m, &st, __pgprot(0), 0, 1); cond_resched(); start++; @@ -515,7 +537,7 @@ static void ptdump_walk_pgd_level_core(struct seq_file *m, pgd_t *pgd, /* Flush out the last page */ st.current_address = normalize_addr(PTRS_PER_PGD*PGD_LEVEL_MULT); - note_page(m, &st, __pgprot(0), 0); + note_page(m, &st, __pgprot(0), 0, 0); if (!checkwx) return; if (st.wx_pages) -- cgit v1.2.3-70-g09d2 From ae8d1d0061ad7996c2c5e769e809a593544fa145 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Mon, 26 Feb 2018 17:25:54 -0600 Subject: x86/mm/sme: Disable stack protection for mem_encrypt_identity.c Stack protection is not compatible with early boot code. All of the early SME boot code is now isolated in a separate file, mem_encrypt_identity.c, so arch/x86/mm/Makefile can be updated to turn off stack protection for the entire file. This eliminates the need to worry about other functions within the file being instrumented with stack protection (as was seen when a newer version of GCC instrumented sme_encrypt_kernel() where an older version hadn't). It also allows removal of the __nostackprotector attribute from individual functions. Signed-off-by: Tom Lendacky Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Acked-by: Kirill A. Shutemov Cc: Linus Torvalds Cc: Borislav Petkov Link: https://lkml.kernel.org/r/20180226232554.14108.16881.stgit@tlendack-t1.amdoffice.net --- arch/x86/mm/Makefile | 1 + arch/x86/mm/mem_encrypt_identity.c | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 03c6c8561623..4b101dd6e52f 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -19,6 +19,7 @@ obj-y := init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \ nostackp := $(call cc-option, -fno-stack-protector) CFLAGS_physaddr.o := $(nostackp) CFLAGS_setup_nx.o := $(nostackp) +CFLAGS_mem_encrypt_identity.o := $(nostackp) CFLAGS_fault.o := -I$(src)/../include/asm/trace diff --git a/arch/x86/mm/mem_encrypt_identity.c b/arch/x86/mm/mem_encrypt_identity.c index b4139c5ab972..1b2197d13832 100644 --- a/arch/x86/mm/mem_encrypt_identity.c +++ b/arch/x86/mm/mem_encrypt_identity.c @@ -266,7 +266,7 @@ static unsigned long __init sme_pgtable_calc(unsigned long len) return entries + tables; } -void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) +void __init sme_encrypt_kernel(struct boot_params *bp) { unsigned long workarea_start, workarea_end, workarea_len; unsigned long execute_start, execute_end, execute_len; @@ -468,7 +468,7 @@ void __init __nostackprotector sme_encrypt_kernel(struct boot_params *bp) native_write_cr3(__native_read_cr3()); } -void __init __nostackprotector sme_enable(struct boot_params *bp) +void __init sme_enable(struct boot_params *bp) { const char *cmdline_ptr, *cmdline_arg, *cmdline_on, *cmdline_off; unsigned int eax, ebx, ecx, edx; -- cgit v1.2.3-70-g09d2 From 565977a3d929fc4427769117a8ac976ec16776d5 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Wed, 14 Mar 2018 14:59:32 -0600 Subject: x86/mm: Remove pointless checks in vmalloc_fault vmalloc_fault() sets user's pgd or p4d from the kernel page table. Once it's set, all tables underneath are identical. There is no point of following the same page table with two separate pointers and make sure they see the same with BUG(). Remove the pointless checks in vmalloc_fault(). Also rename the kernel pgd/p4d pointers to pgd_k/p4d_k so that their names are consistent in the file. Suggested-by: Andy Lutomirski Signed-off-by: Toshi Kani Signed-off-by: Thomas Gleixner Cc: linux-mm@kvack.org Cc: Borislav Petkov Cc: Gratian Crisan Link: https://lkml.kernel.org/r/20180314205932.7193-1-toshi.kani@hpe.com --- arch/x86/mm/fault.c | 56 ++++++++++++++++------------------------------------- 1 file changed, 17 insertions(+), 39 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 84d702a71afe..70c3b1c43676 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -417,11 +417,11 @@ void vmalloc_sync_all(void) */ static noinline int vmalloc_fault(unsigned long address) { - pgd_t *pgd, *pgd_ref; - p4d_t *p4d, *p4d_ref; - pud_t *pud, *pud_ref; - pmd_t *pmd, *pmd_ref; - pte_t *pte, *pte_ref; + pgd_t *pgd, *pgd_k; + p4d_t *p4d, *p4d_k; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; /* Make sure we are in vmalloc area: */ if (!(address >= VMALLOC_START && address < VMALLOC_END)) @@ -435,73 +435,51 @@ static noinline int vmalloc_fault(unsigned long address) * case just flush: */ pgd = (pgd_t *)__va(read_cr3_pa()) + pgd_index(address); - pgd_ref = pgd_offset_k(address); - if (pgd_none(*pgd_ref)) + pgd_k = pgd_offset_k(address); + if (pgd_none(*pgd_k)) return -1; if (pgtable_l5_enabled) { if (pgd_none(*pgd)) { - set_pgd(pgd, *pgd_ref); + set_pgd(pgd, *pgd_k); arch_flush_lazy_mmu_mode(); } else { - BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref)); + BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_k)); } } /* With 4-level paging, copying happens on the p4d level. */ p4d = p4d_offset(pgd, address); - p4d_ref = p4d_offset(pgd_ref, address); - if (p4d_none(*p4d_ref)) + p4d_k = p4d_offset(pgd_k, address); + if (p4d_none(*p4d_k)) return -1; if (p4d_none(*p4d) && !pgtable_l5_enabled) { - set_p4d(p4d, *p4d_ref); + set_p4d(p4d, *p4d_k); arch_flush_lazy_mmu_mode(); } else { - BUG_ON(p4d_pfn(*p4d) != p4d_pfn(*p4d_ref)); + BUG_ON(p4d_pfn(*p4d) != p4d_pfn(*p4d_k)); } - /* - * Below here mismatches are bugs because these lower tables - * are shared: - */ BUILD_BUG_ON(CONFIG_PGTABLE_LEVELS < 4); pud = pud_offset(p4d, address); - pud_ref = pud_offset(p4d_ref, address); - if (pud_none(*pud_ref)) + if (pud_none(*pud)) return -1; - if (pud_none(*pud) || pud_pfn(*pud) != pud_pfn(*pud_ref)) - BUG(); - if (pud_large(*pud)) return 0; pmd = pmd_offset(pud, address); - pmd_ref = pmd_offset(pud_ref, address); - if (pmd_none(*pmd_ref)) + if (pmd_none(*pmd)) return -1; - if (pmd_none(*pmd) || pmd_pfn(*pmd) != pmd_pfn(*pmd_ref)) - BUG(); - if (pmd_large(*pmd)) return 0; - pte_ref = pte_offset_kernel(pmd_ref, address); - if (!pte_present(*pte_ref)) - return -1; - pte = pte_offset_kernel(pmd, address); - - /* - * Don't use pte_page here, because the mappings can point - * outside mem_map, and the NUMA hash lookup cannot handle - * that: - */ - if (!pte_present(*pte) || pte_pfn(*pte) != pte_pfn(*pte_ref)) - BUG(); + if (!pte_present(*pte)) + return -1; return 0; } -- cgit v1.2.3-70-g09d2 From fc5d1073cae299de4517755a910df4f12a6a438f Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 26 Mar 2018 23:27:21 -0700 Subject: x86/mm/32: Remove unused node_memmap_size_bytes() & CONFIG_NEED_NODE_MEMMAP_SIZE logic node_memmap_size_bytes() has been unused since the v3.9 kernel, so remove it. Signed-off-by: David Rientjes Cc: Dave Hansen Cc: Laura Abbott Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-mm@kvack.org Fixes: f03574f2d5b2 ("x86-32, mm: Rip out x86_32 NUMA remapping code") Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1803262325540.256524@chino.kir.corp.google.com Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 4 ---- arch/x86/mm/numa_32.c | 11 ----------- include/linux/mmzone.h | 5 ----- mm/sparse.c | 22 ---------------------- 4 files changed, 42 deletions(-) (limited to 'arch/x86/mm') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 18233e459bff..739aff253d17 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1608,10 +1608,6 @@ config ARCH_HAVE_MEMORY_PRESENT def_bool y depends on X86_32 && DISCONTIGMEM -config NEED_NODE_MEMMAP_SIZE - def_bool y - depends on X86_32 && (DISCONTIGMEM || SPARSEMEM) - config ARCH_FLATMEM_ENABLE def_bool y depends on X86_32 && !NUMA diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index aca6295350f3..e8a4a09e20f1 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -60,17 +60,6 @@ void memory_present(int nid, unsigned long start, unsigned long end) } printk(KERN_CONT "\n"); } - -unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn, - unsigned long end_pfn) -{ - unsigned long nr_pages = end_pfn - start_pfn; - - if (!nr_pages) - return 0; - - return (nr_pages + 1) * sizeof(struct page); -} #endif extern unsigned long highend_pfn, highstart_pfn; diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 7522a6987595..a2db4576e499 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -816,10 +816,6 @@ int local_memory_node(int node_id); static inline int local_memory_node(int node_id) { return node_id; }; #endif -#ifdef CONFIG_NEED_NODE_MEMMAP_SIZE -unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); -#endif - /* * zone_idx() returns 0 for the ZONE_DMA zone, 1 for the ZONE_NORMAL zone, etc. */ @@ -1289,7 +1285,6 @@ struct mminit_pfnnid_cache { #endif void memory_present(int nid, unsigned long start, unsigned long end); -unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); /* * If it is possible to have holes within a MAX_ORDER_NR_PAGES, then we diff --git a/mm/sparse.c b/mm/sparse.c index 7af5e7a92528..79b26f98d793 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -235,28 +235,6 @@ void __init memory_present(int nid, unsigned long start, unsigned long end) } } -/* - * Only used by the i386 NUMA architecures, but relatively - * generic code. - */ -unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn, - unsigned long end_pfn) -{ - unsigned long pfn; - unsigned long nr_pages = 0; - - mminit_validate_memmodel_limits(&start_pfn, &end_pfn); - for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) { - if (nid != early_pfn_to_nid(pfn)) - continue; - - if (pfn_present(pfn)) - nr_pages += PAGES_PER_SECTION; - } - - return nr_pages * sizeof(struct page); -} - /* * Subtle, we encode the real pfn into the mem_map such that * the identity pfn - section_mem_map will return the actual -- cgit v1.2.3-70-g09d2