- Linux-stable-mirror - lists.linaro.org

[Linux-stable-mirror] [PATCH] x86/mm: Encrypt the initrd earlier for BSP microcode update

by Tom Lendacky

commit 107cd2532181b96c549e8f224cdcca8631c3076b upstream to be applied to stable 4.14 This patch resolves minor conflicts associated with backporting the upstream commit. Currently the BSP microcode update code examines the initrd very early in the boot process. If SME is active, the initrd is treated as being encrypted but it has not been encrypted (in place) yet. Update the early boot code that encrypts the kernel to also encrypt the initrd so that early BSP microcode updates work. Tested-by: Gabriel Craciunescu <nix.or.die(a)gmail.com> Signed-off-by: Tom Lendacky <thomas.lendacky(a)amd.com> --- arch/x86/include/asm/mem_encrypt.h | 4 +- arch/x86/kernel/head64.c | 4 +- arch/x86/kernel/setup.c | 8 ---- arch/x86/mm/mem_encrypt.c | 66 ++++++++++++++++++++++++++++++++---- arch/x86/mm/mem_encrypt_boot.S | 46 +++++++++++++------------ 5 files changed, 85 insertions(+), 43 deletions(-) diff --git a/arch/x86/include/asm/mem_encrypt.h b/arch/x86/include/asm/mem_encrypt.h index 6a77c63..e7d96c0 100644 --- a/arch/x86/include/asm/mem_encrypt.h +++ b/arch/x86/include/asm/mem_encrypt.h @@ -39,7 +39,7 @@ void __init sme_early_decrypt(resource_size_t paddr, void __init sme_early_init(void); -void __init sme_encrypt_kernel(void); +void __init sme_encrypt_kernel(struct boot_params *bp); void __init sme_enable(struct boot_params *bp); /* Architecture __weak replacement functions */ @@ -61,7 +61,7 @@ static inline void __init sme_unmap_bootdata(char *real_mode_data) { } static inline void __init sme_early_init(void) { } -static inline void __init sme_encrypt_kernel(void) { } +static inline void __init sme_encrypt_kernel(struct boot_params *bp) { } static inline void __init sme_enable(struct boot_params *bp) { } #endif /* CONFIG_AMD_MEM_ENCRYPT */ diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 6a5d757..7ba5d81 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -157,8 +157,8 @@ unsigned long __head __startup_64(unsigned long physaddr, p = fixup_pointer(&phys_base, physaddr); *p += load_delta - sme_get_me_mask(); - /* Encrypt the kernel (if SME is active) */ - sme_encrypt_kernel(); + /* Encrypt the kernel and related (if SME is active) */ + sme_encrypt_kernel(bp); /* * Return the SME encryption mask (if SME is active) to be used as a diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 0957dd7..e84cb4c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -376,14 +376,6 @@ static void __init reserve_initrd(void) !ramdisk_image || !ramdisk_size) return; /* No initrd provided by bootloader */ - /* - * If SME is active, this memory will be marked encrypted by the - * kernel when it is accessed (including relocation). However, the - * ramdisk image was loaded decrypted by the bootloader, so make - * sure that it is encrypted before accessing it. - */ - sme_early_encrypt(ramdisk_image, ramdisk_end - ramdisk_image); - initrd_start = 0; mapped_size = memblock_mem_size(max_pfn_mapped); diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c index b4f15fb..e110fab 100644 --- a/arch/x86/mm/mem_encrypt.c +++ b/arch/x86/mm/mem_encrypt.c @@ -487,11 +487,12 @@ static unsigned long __init sme_pgtable_calc(unsigned long len) return total; } -void __init sme_encrypt_kernel(void) +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; 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; @@ -500,14 +501,15 @@ void __init sme_encrypt_kernel(void) return; /* - * Prepare for encrypting the kernel by building new pagetables with - * the necessary attributes needed to encrypt the kernel in place. + * 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 as encrypted. + * by the kernel and initrd as encrypted. * * Another range of virtual addresses will map the memory occupied - * by the kernel as decrypted and write-protected. + * 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. @@ -518,6 +520,20 @@ void __init sme_encrypt_kernel(void) 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; @@ -540,6 +556,8 @@ void __init sme_encrypt_kernel(void) */ 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); @@ -578,9 +596,9 @@ void __init sme_encrypt_kernel(void) /* * A new pagetable structure is being built to allow for the kernel - * 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. + * 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); @@ -593,6 +611,12 @@ void __init sme_encrypt_kernel(void) * 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 */ @@ -607,6 +631,21 @@ void __init sme_encrypt_kernel(void) 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; @@ -622,6 +661,11 @@ void __init sme_encrypt_kernel(void) 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 @@ -631,6 +675,12 @@ void __init sme_encrypt_kernel(void) 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); diff --git a/arch/x86/mm/mem_encrypt_boot.S b/arch/x86/mm/mem_encrypt_boot.S index 23a8a9e..01f682c 100644 --- a/arch/x86/mm/mem_encrypt_boot.S +++ b/arch/x86/mm/mem_encrypt_boot.S @@ -22,9 +22,9 @@ ENTRY(sme_encrypt_execute) /* * Entry parameters: - * RDI - virtual address for the encrypted kernel mapping - * RSI - virtual address for the decrypted kernel mapping - * RDX - length of kernel + * RDI - virtual address for the encrypted mapping + * RSI - virtual address for the decrypted mapping + * RDX - length to encrypt * RCX - virtual address of the encryption workarea, including: * - stack page (PAGE_SIZE) * - encryption routine page (PAGE_SIZE) @@ -41,9 +41,9 @@ ENTRY(sme_encrypt_execute) addq $PAGE_SIZE, %rax /* Workarea encryption routine */ push %r12 - movq %rdi, %r10 /* Encrypted kernel */ - movq %rsi, %r11 /* Decrypted kernel */ - movq %rdx, %r12 /* Kernel length */ + movq %rdi, %r10 /* Encrypted area */ + movq %rsi, %r11 /* Decrypted area */ + movq %rdx, %r12 /* Area length */ /* Copy encryption routine into the workarea */ movq %rax, %rdi /* Workarea encryption routine */ @@ -52,10 +52,10 @@ ENTRY(sme_encrypt_execute) rep movsb /* Setup registers for call */ - movq %r10, %rdi /* Encrypted kernel */ - movq %r11, %rsi /* Decrypted kernel */ + movq %r10, %rdi /* Encrypted area */ + movq %r11, %rsi /* Decrypted area */ movq %r8, %rdx /* Pagetables used for encryption */ - movq %r12, %rcx /* Kernel length */ + movq %r12, %rcx /* Area length */ movq %rax, %r8 /* Workarea encryption routine */ addq $PAGE_SIZE, %r8 /* Workarea intermediate copy buffer */ @@ -71,7 +71,7 @@ ENDPROC(sme_encrypt_execute) ENTRY(__enc_copy) /* - * Routine used to encrypt kernel. + * Routine used to encrypt memory in place. * This routine must be run outside of the kernel proper since * the kernel will be encrypted during the process. So this * routine is defined here and then copied to an area outside @@ -79,19 +79,19 @@ ENTRY(__enc_copy) * during execution. * * On entry the registers must be: - * RDI - virtual address for the encrypted kernel mapping - * RSI - virtual address for the decrypted kernel mapping + * RDI - virtual address for the encrypted mapping + * RSI - virtual address for the decrypted mapping * RDX - address of the pagetables to use for encryption - * RCX - length of kernel + * RCX - length of area * R8 - intermediate copy buffer * * RAX - points to this routine * - * The kernel will be encrypted by copying from the non-encrypted - * kernel space to an intermediate buffer and then copying from the - * intermediate buffer back to the encrypted kernel space. The physical - * addresses of the two kernel space mappings are the same which - * results in the kernel being encrypted "in place". + * The area will be encrypted by copying from the non-encrypted + * memory space to an intermediate buffer and then copying from the + * intermediate buffer back to the encrypted memory space. The physical + * addresses of the two mappings are the same which results in the area + * being encrypted "in place". */ /* Enable the new page tables */ mov %rdx, %cr3 @@ -106,9 +106,9 @@ ENTRY(__enc_copy) push %r15 push %r12 - movq %rcx, %r9 /* Save kernel length */ - movq %rdi, %r10 /* Save encrypted kernel address */ - movq %rsi, %r11 /* Save decrypted kernel address */ + movq %rcx, %r9 /* Save area length */ + movq %rdi, %r10 /* Save encrypted area address */ + movq %rsi, %r11 /* Save decrypted area address */ /* Set the PAT register PA5 entry to write-protect */ movl $MSR_IA32_CR_PAT, %ecx @@ -128,13 +128,13 @@ ENTRY(__enc_copy) movq %r9, %r12 2: - movq %r11, %rsi /* Source - decrypted kernel */ + movq %r11, %rsi /* Source - decrypted area */ movq %r8, %rdi /* Dest - intermediate copy buffer */ movq %r12, %rcx rep movsb movq %r8, %rsi /* Source - intermediate copy buffer */ - movq %r10, %rdi /* Dest - encrypted kernel */ + movq %r10, %rdi /* Dest - encrypted area */ movq %r12, %rcx rep movsb

7 years, 1 month

2
1
0 0

[Linux-stable-mirror] Patch "x86/mm: Encrypt the initrd earlier for BSP microcode update" has been added to the 4.14-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled x86/mm: Encrypt the initrd earlier for BSP microcode update to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: x86-mm-encrypt-the-initrd-earlier-for-bsp-microcode-update.patch and it can be found in the queue-4.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 107cd2532181b96c549e8f224cdcca8631c3076b Mon Sep 17 00:00:00 2001 From: Tom Lendacky <thomas.lendacky(a)amd.com> Date: Wed, 10 Jan 2018 13:26:34 -0600 Subject: x86/mm: Encrypt the initrd earlier for BSP microcode update From: Tom Lendacky <thomas.lendacky(a)amd.com> commit 107cd2532181b96c549e8f224cdcca8631c3076b upstream. Currently the BSP microcode update code examines the initrd very early in the boot process. If SME is active, the initrd is treated as being encrypted but it has not been encrypted (in place) yet. Update the early boot code that encrypts the kernel to also encrypt the initrd so that early BSP microcode updates work. Tested-by: Gabriel Craciunescu <nix.or.die(a)gmail.com> Signed-off-by: Tom Lendacky <thomas.lendacky(a)amd.com> Reviewed-by: Borislav Petkov <bp(a)suse.de> Cc: Borislav Petkov <bp(a)alien8.de> Cc: Brijesh Singh <brijesh.singh(a)amd.com> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Thomas Gleixner <tglx(a)linutronix.de> Link: http://lkml.kernel.org/r/20180110192634.6026.10452.stgit@tlendack-t1.amdoff… Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- arch/x86/include/asm/mem_encrypt.h | 4 +- arch/x86/kernel/head64.c | 4 +- arch/x86/kernel/setup.c | 8 ---- arch/x86/mm/mem_encrypt.c | 66 ++++++++++++++++++++++++++++++++----- arch/x86/mm/mem_encrypt_boot.S | 46 ++++++++++++------------- 5 files changed, 85 insertions(+), 43 deletions(-) --- a/arch/x86/include/asm/mem_encrypt.h +++ b/arch/x86/include/asm/mem_encrypt.h @@ -39,7 +39,7 @@ void __init sme_unmap_bootdata(char *rea void __init sme_early_init(void); -void __init sme_encrypt_kernel(void); +void __init sme_encrypt_kernel(struct boot_params *bp); void __init sme_enable(struct boot_params *bp); /* Architecture __weak replacement functions */ @@ -61,7 +61,7 @@ static inline void __init sme_unmap_boot static inline void __init sme_early_init(void) { } -static inline void __init sme_encrypt_kernel(void) { } +static inline void __init sme_encrypt_kernel(struct boot_params *bp) { } static inline void __init sme_enable(struct boot_params *bp) { } #endif /* CONFIG_AMD_MEM_ENCRYPT */ --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -157,8 +157,8 @@ unsigned long __head __startup_64(unsign p = fixup_pointer(&phys_base, physaddr); *p += load_delta - sme_get_me_mask(); - /* Encrypt the kernel (if SME is active) */ - sme_encrypt_kernel(); + /* Encrypt the kernel and related (if SME is active) */ + sme_encrypt_kernel(bp); /* * Return the SME encryption mask (if SME is active) to be used as a --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -376,14 +376,6 @@ static void __init reserve_initrd(void) !ramdisk_image || !ramdisk_size) return; /* No initrd provided by bootloader */ - /* - * If SME is active, this memory will be marked encrypted by the - * kernel when it is accessed (including relocation). However, the - * ramdisk image was loaded decrypted by the bootloader, so make - * sure that it is encrypted before accessing it. - */ - sme_early_encrypt(ramdisk_image, ramdisk_end - ramdisk_image); - initrd_start = 0; mapped_size = memblock_mem_size(max_pfn_mapped); --- a/arch/x86/mm/mem_encrypt.c +++ b/arch/x86/mm/mem_encrypt.c @@ -487,11 +487,12 @@ static unsigned long __init sme_pgtable_ return total; } -void __init sme_encrypt_kernel(void) +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; 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; @@ -500,14 +501,15 @@ void __init sme_encrypt_kernel(void) return; /* - * Prepare for encrypting the kernel by building new pagetables with - * the necessary attributes needed to encrypt the kernel in place. + * 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 as encrypted. + * by the kernel and initrd as encrypted. * * Another range of virtual addresses will map the memory occupied - * by the kernel as decrypted and write-protected. + * 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. @@ -518,6 +520,20 @@ void __init sme_encrypt_kernel(void) 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; @@ -540,6 +556,8 @@ void __init sme_encrypt_kernel(void) */ 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); @@ -578,9 +596,9 @@ void __init sme_encrypt_kernel(void) /* * A new pagetable structure is being built to allow for the kernel - * 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. + * 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); @@ -593,6 +611,12 @@ void __init sme_encrypt_kernel(void) * 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 */ @@ -607,6 +631,21 @@ void __init sme_encrypt_kernel(void) 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; @@ -622,6 +661,11 @@ void __init sme_encrypt_kernel(void) 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 @@ -631,6 +675,12 @@ void __init sme_encrypt_kernel(void) 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); --- a/arch/x86/mm/mem_encrypt_boot.S +++ b/arch/x86/mm/mem_encrypt_boot.S @@ -22,9 +22,9 @@ ENTRY(sme_encrypt_execute) /* * Entry parameters: - * RDI - virtual address for the encrypted kernel mapping - * RSI - virtual address for the decrypted kernel mapping - * RDX - length of kernel + * RDI - virtual address for the encrypted mapping + * RSI - virtual address for the decrypted mapping + * RDX - length to encrypt * RCX - virtual address of the encryption workarea, including: * - stack page (PAGE_SIZE) * - encryption routine page (PAGE_SIZE) @@ -41,9 +41,9 @@ ENTRY(sme_encrypt_execute) addq $PAGE_SIZE, %rax /* Workarea encryption routine */ push %r12 - movq %rdi, %r10 /* Encrypted kernel */ - movq %rsi, %r11 /* Decrypted kernel */ - movq %rdx, %r12 /* Kernel length */ + movq %rdi, %r10 /* Encrypted area */ + movq %rsi, %r11 /* Decrypted area */ + movq %rdx, %r12 /* Area length */ /* Copy encryption routine into the workarea */ movq %rax, %rdi /* Workarea encryption routine */ @@ -52,10 +52,10 @@ ENTRY(sme_encrypt_execute) rep movsb /* Setup registers for call */ - movq %r10, %rdi /* Encrypted kernel */ - movq %r11, %rsi /* Decrypted kernel */ + movq %r10, %rdi /* Encrypted area */ + movq %r11, %rsi /* Decrypted area */ movq %r8, %rdx /* Pagetables used for encryption */ - movq %r12, %rcx /* Kernel length */ + movq %r12, %rcx /* Area length */ movq %rax, %r8 /* Workarea encryption routine */ addq $PAGE_SIZE, %r8 /* Workarea intermediate copy buffer */ @@ -71,7 +71,7 @@ ENDPROC(sme_encrypt_execute) ENTRY(__enc_copy) /* - * Routine used to encrypt kernel. + * Routine used to encrypt memory in place. * This routine must be run outside of the kernel proper since * the kernel will be encrypted during the process. So this * routine is defined here and then copied to an area outside @@ -79,19 +79,19 @@ ENTRY(__enc_copy) * during execution. * * On entry the registers must be: - * RDI - virtual address for the encrypted kernel mapping - * RSI - virtual address for the decrypted kernel mapping + * RDI - virtual address for the encrypted mapping + * RSI - virtual address for the decrypted mapping * RDX - address of the pagetables to use for encryption - * RCX - length of kernel + * RCX - length of area * R8 - intermediate copy buffer * * RAX - points to this routine * - * The kernel will be encrypted by copying from the non-encrypted - * kernel space to an intermediate buffer and then copying from the - * intermediate buffer back to the encrypted kernel space. The physical - * addresses of the two kernel space mappings are the same which - * results in the kernel being encrypted "in place". + * The area will be encrypted by copying from the non-encrypted + * memory space to an intermediate buffer and then copying from the + * intermediate buffer back to the encrypted memory space. The physical + * addresses of the two mappings are the same which results in the area + * being encrypted "in place". */ /* Enable the new page tables */ mov %rdx, %cr3 @@ -106,9 +106,9 @@ ENTRY(__enc_copy) push %r15 push %r12 - movq %rcx, %r9 /* Save kernel length */ - movq %rdi, %r10 /* Save encrypted kernel address */ - movq %rsi, %r11 /* Save decrypted kernel address */ + movq %rcx, %r9 /* Save area length */ + movq %rdi, %r10 /* Save encrypted area address */ + movq %rsi, %r11 /* Save decrypted area address */ /* Set the PAT register PA5 entry to write-protect */ movl $MSR_IA32_CR_PAT, %ecx @@ -128,13 +128,13 @@ ENTRY(__enc_copy) movq %r9, %r12 2: - movq %r11, %rsi /* Source - decrypted kernel */ + movq %r11, %rsi /* Source - decrypted area */ movq %r8, %rdi /* Dest - intermediate copy buffer */ movq %r12, %rcx rep movsb movq %r8, %rsi /* Source - intermediate copy buffer */ - movq %r10, %rdi /* Dest - encrypted kernel */ + movq %r10, %rdi /* Dest - encrypted area */ movq %r12, %rcx rep movsb Patches currently in stable-queue which might be from thomas.lendacky(a)amd.com are queue-4.14/x86-mm-clean-up-register-saving-in-the-__enc_copy-assembly-code.patch queue-4.14/x86-mm-use-a-struct-to-reduce-parameters-for-sme-pgd-mapping.patch queue-4.14/x86-mm-centralize-pmd-flags-in-sme_encrypt_kernel.patch queue-4.14/x86-retpoline-fill-rsb-on-context-switch-for-affected-cpus.patch queue-4.14/x86-mm-prepare-sme_encrypt_kernel-for-page-aligned-encryption.patch queue-4.14/x86-retpoline-add-lfence-to-the-retpoline-rsb-filling-rsb-macros.patch queue-4.14/x86-mm-encrypt-the-initrd-earlier-for-bsp-microcode-update.patch

7 years, 1 month

1
0
0 0

[Linux-stable-mirror] ARM: OMAP3: hwmod_data: add missing module_offs for MMC3

by Adam Ford

Please port 3c4d296e58a2 ("ARM: OMAP3: hwmod_data: add missing module_offs for MMC3") to Linux 4.14.y branch. This corrects an issue where a WiFi device connected to MMC3 fails to operate correctly. Fixes: 6c0afb503937 ("clk: ti: convert to use proper register definition for all accesses") adam

7 years, 1 month

2
1
0 0

[Linux-stable-mirror] Patch "ARM: OMAP3: hwmod_data: add missing module_offs for MMC3" has been added to the 4.14-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled ARM: OMAP3: hwmod_data: add missing module_offs for MMC3 to the 4.14-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: arm-omap3-hwmod_data-add-missing-module_offs-for-mmc3.patch and it can be found in the queue-4.14 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From 3c4d296e58a23687f2076d8ad531e6ae2b725846 Mon Sep 17 00:00:00 2001 From: Tero Kristo <t-kristo(a)ti.com> Date: Mon, 30 Oct 2017 11:11:03 +0200 Subject: ARM: OMAP3: hwmod_data: add missing module_offs for MMC3 From: Tero Kristo <t-kristo(a)ti.com> commit 3c4d296e58a23687f2076d8ad531e6ae2b725846 upstream. MMC3 hwmod data is missing the module_offs definition. MMC3 belongs under core, so add CORE_MOD for it. Signed-off-by: Tero Kristo <t-kristo(a)ti.com> Signed-off-by: Tony Lindgren <tony(a)atomide.com> Cc: Adam Ford <aford173(a)gmail.com> Fixes: 6c0afb503937 ("clk: ti: convert to use proper register definition for all accesses") Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 1 + 1 file changed, 1 insertion(+) --- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c @@ -1656,6 +1656,7 @@ static struct omap_hwmod omap3xxx_mmc3_h .main_clk = "mmchs3_fck", .prcm = { .omap2 = { + .module_offs = CORE_MOD, .prcm_reg_id = 1, .module_bit = OMAP3430_EN_MMC3_SHIFT, .idlest_reg_id = 1, Patches currently in stable-queue which might be from t-kristo(a)ti.com are queue-4.14/arm-omap3-hwmod_data-add-missing-module_offs-for-mmc3.patch

7 years, 1 month

1
0
0 0

Re: [Linux-stable-mirror] Build errors in v4.4-stable-queue (i386)

by Guenter Roeck

On Fri, Jan 19, 2018 at 02:08:35PM +0100, David Woodhouse wrote: > On Fri, 2018-01-19 at 13:26 +0100, Greg Kroah-Hartman wrote: > > > > I don't have a way to test this, I'll merge it into the existing patch > > and push out a new tree to see how 0-day and Guenter's build-farm handle > > it. > > It seems to work (and boot in qemu) here. Same here. Guenter

7 years, 1 month

2
1
0 0

[Linux-stable-mirror] [PATCH] ARM: sunxi_defconfig: Enable CMA

by Maxime Ripard

The DRM driver most notably, but also out of tree drivers (for now) like the VPU or GPU drivers, are quite big consumers of large, contiguous memory buffers. However, the sunxi_defconfig doesn't enable CMA in order to mitigate that, which makes them almost unusable. Enable it to make sure it somewhat works. Cc: <stable(a)vger.kernel.org> Signed-off-by: Maxime Ripard <maxime.ripard(a)free-electrons.com> --- Hi Arnd, Olof, Could you queue that patch as a fix? Thanks! Maxime arch/arm/configs/sunxi_defconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/configs/sunxi_defconfig b/arch/arm/configs/sunxi_defconfig index 5caaf971fb50..df433abfcb02 100644 --- a/arch/arm/configs/sunxi_defconfig +++ b/arch/arm/configs/sunxi_defconfig @@ -10,6 +10,7 @@ CONFIG_SMP=y CONFIG_NR_CPUS=8 CONFIG_AEABI=y CONFIG_HIGHMEM=y +CONFIG_CMA=y CONFIG_ARM_APPENDED_DTB=y CONFIG_ARM_ATAG_DTB_COMPAT=y CONFIG_CPU_FREQ=y @@ -33,6 +34,7 @@ CONFIG_CAN_SUN4I=y # CONFIG_WIRELESS is not set CONFIG_DEVTMPFS=y CONFIG_DEVTMPFS_MOUNT=y +CONFIG_DMA_CMA=y CONFIG_BLK_DEV_SD=y CONFIG_ATA=y CONFIG_AHCI_SUNXI=y -- 2.14.3

7 years, 1 month

3
2
0 0

[Linux-stable-mirror] [PATCH] NFS: reject request for id_legacy key without auxdata

by Eric Biggers

From: Eric Biggers <ebiggers(a)google.com> nfs_idmap_legacy_upcall() is supposed to be called with 'aux' pointing to a 'struct idmap', via the call to request_key_with_auxdata() in nfs_idmap_request_key(). However it can also be reached via the request_key() system call in which case 'aux' will be NULL, causing a NULL pointer dereference in nfs_idmap_prepare_pipe_upcall(), assuming that the key description is valid enough to get that far. Fix this by making nfs_idmap_legacy_upcall() negate the key if no auxdata is provided. As usual, this bug was found by syzkaller. A simple reproducer using the command-line keyctl program is: keyctl request2 id_legacy uid:0 '' @s Fixes: 57e62324e469 ("NFS: Store the legacy idmapper result in the keyring") Reported-by: syzbot+5dfdbcf7b3eb5912abbb(a)syzkaller.appspotmail.com Cc: <stable(a)vger.kernel.org> # v3.4+ Signed-off-by: Eric Biggers <ebiggers(a)google.com> --- fs/nfs/nfs4idmap.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c index 30426c1a1bbd..22dc30a679a0 100644 --- a/fs/nfs/nfs4idmap.c +++ b/fs/nfs/nfs4idmap.c @@ -568,9 +568,13 @@ static int nfs_idmap_legacy_upcall(struct key_construction *cons, struct idmap_msg *im; struct idmap *idmap = (struct idmap *)aux; struct key *key = cons->key; - int ret = -ENOMEM; + int ret = -ENOKEY; + + if (!aux) + goto out1; /* msg and im are freed in idmap_pipe_destroy_msg */ + ret = -ENOMEM; data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) goto out1; -- 2.16.0.rc1.238.g530d649a79-goog

7 years, 1 month

1
0
0 0

[Linux-stable-mirror] + lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang.patch added to -mm tree

by akpm＠linux-foundation.org

The patch titled Subject: lib/ubsan: add type mismatch handler for new GCC/Clang has been added to the -mm tree. Its filename is lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang.patch This patch should soon appear at http://ozlabs.org/~akpm/mmots/broken-out/lib-ubsan-add-type-mismatch-handle… and later at http://ozlabs.org/~akpm/mmotm/broken-out/lib-ubsan-add-type-mismatch-handle… Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** The -mm tree is included into linux-next and is updated there every 3-4 working days ------------------------------------------------------ From: Andrey Ryabinin <aryabinin(a)virtuozzo.com> Subject: lib/ubsan: add type mismatch handler for new GCC/Clang UBSAN=y fails to build with new GCC/clang: arch/x86/kernel/head64.o: In function `sanitize_boot_params': arch/x86/include/asm/bootparam_utils.h:37: undefined reference to `__ubsan_handle_type_mismatch_v1' because Clang and GCC 8 slightly changed ABI for 'type mismatch' errors. Compiler now uses new __ubsan_handle_type_mismatch_v1() function with slightly modified 'struct type_mismatch_data'. Let's add new 'struct type_mismatch_data_common' which is independent from compiler's layout of 'struct type_mismatch_data'. And make __ubsan_handle_type_mismatch[_v1]() functions transform compiler-dependent type mismatch data to our internal representation. This way, we can support both old and new compilers with minimal amount of change. Link: http://lkml.kernel.org/r/20180119152853.16806-1-aryabinin@virtuozzo.com Signed-off-by: Andrey Ryabinin <aryabinin(a)virtuozzo.com> Reported-by: Sodagudi Prasad <psodagud(a)codeaurora.org> Cc: <stable(a)vger.kernel.org> [4.5+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- lib/ubsan.c | 48 ++++++++++++++++++++++++++++++++++++++---------- lib/ubsan.h | 14 ++++++++++++++ 2 files changed, 52 insertions(+), 10 deletions(-) diff -puN lib/ubsan.c~lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang lib/ubsan.c --- a/lib/ubsan.c~lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang +++ a/lib/ubsan.c @@ -265,14 +265,14 @@ void __ubsan_handle_divrem_overflow(stru } EXPORT_SYMBOL(__ubsan_handle_divrem_overflow); -static void handle_null_ptr_deref(struct type_mismatch_data *data) +static void handle_null_ptr_deref(struct type_mismatch_data_common *data) { unsigned long flags; - if (suppress_report(&data->location)) + if (suppress_report(data->location)) return; - ubsan_prologue(&data->location, &flags); + ubsan_prologue(data->location, &flags); pr_err("%s null pointer of type %s\n", type_check_kinds[data->type_check_kind], @@ -281,15 +281,15 @@ static void handle_null_ptr_deref(struct ubsan_epilogue(&flags); } -static void handle_misaligned_access(struct type_mismatch_data *data, +static void handle_misaligned_access(struct type_mismatch_data_common *data, unsigned long ptr) { unsigned long flags; - if (suppress_report(&data->location)) + if (suppress_report(data->location)) return; - ubsan_prologue(&data->location, &flags); + ubsan_prologue(data->location, &flags); pr_err("%s misaligned address %p for type %s\n", type_check_kinds[data->type_check_kind], @@ -299,15 +299,15 @@ static void handle_misaligned_access(str ubsan_epilogue(&flags); } -static void handle_object_size_mismatch(struct type_mismatch_data *data, +static void handle_object_size_mismatch(struct type_mismatch_data_common *data, unsigned long ptr) { unsigned long flags; - if (suppress_report(&data->location)) + if (suppress_report(data->location)) return; - ubsan_prologue(&data->location, &flags); + ubsan_prologue(data->location, &flags); pr_err("%s address %p with insufficient space\n", type_check_kinds[data->type_check_kind], (void *) ptr); @@ -315,7 +315,7 @@ static void handle_object_size_mismatch( ubsan_epilogue(&flags); } -void __ubsan_handle_type_mismatch(struct type_mismatch_data *data, +static void ubsan_type_mismatch_common(struct type_mismatch_data_common *data, unsigned long ptr) { @@ -326,8 +326,36 @@ void __ubsan_handle_type_mismatch(struct else handle_object_size_mismatch(data, ptr); } + +void __ubsan_handle_type_mismatch(struct type_mismatch_data *data, + unsigned long ptr) +{ + struct type_mismatch_data_common common_data = { + .location = &data->location, + .type = data->type, + .alignment = data->alignment, + .type_check_kind = data->type_check_kind + }; + + ubsan_type_mismatch_common(&common_data, ptr); +} EXPORT_SYMBOL(__ubsan_handle_type_mismatch); +void __ubsan_handle_type_mismatch_v1(struct type_mismatch_data_v1 *data, + unsigned long ptr) +{ + + struct type_mismatch_data_common common_data = { + .location = &data->location, + .type = data->type, + .alignment = 1UL << data->log_alignment, + .type_check_kind = data->type_check_kind + }; + + ubsan_type_mismatch_common(&common_data, ptr); +} +EXPORT_SYMBOL(__ubsan_handle_type_mismatch_v1); + void __ubsan_handle_nonnull_return(struct nonnull_return_data *data) { unsigned long flags; diff -puN lib/ubsan.h~lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang lib/ubsan.h --- a/lib/ubsan.h~lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang +++ a/lib/ubsan.h @@ -37,6 +37,20 @@ struct type_mismatch_data { unsigned char type_check_kind; }; +struct type_mismatch_data_v1 { + struct source_location location; + struct type_descriptor *type; + unsigned char log_alignment; + unsigned char type_check_kind; +}; + +struct type_mismatch_data_common { + struct source_location *location; + struct type_descriptor *type; + unsigned long alignment; + unsigned char type_check_kind; +}; + struct nonnull_arg_data { struct source_location location; struct source_location attr_location; _ Patches currently in -mm which might be from aryabinin(a)virtuozzo.com are mm-memcg-try-harder-to-decrease-limit_in_bytes.patch kasan-makefile-support-llvm-style-asan-parameters.patch lib-strscpy-remove-word-at-a-time-optimization.patch lib-ubsan-add-type-mismatch-handler-for-new-gcc-clang.patch lib-ubsan-remove-returns-nonnull-attribute-checks.patch

7 years, 1 month

1
0
0 0

[Linux-stable-mirror] [PATCH 2/3] Input: synaptics_rmi4 - unmask F03 interrupts when port is opened

by Dmitry Torokhov

Currently we register the pass-through serio port when we probe the F03 RMI function, and then, in sensor configure phase, we unmask interrupts. Unfortunately this is too late, as other drivers are free probe devices attached to the serio port as soon as it is probed. Because interrupts are masked, the IO times out, which may result in not being able to detect trackpoints on the pass-through port. To fix the issue we implement open() and close() methods for the pass-through serio port and unmask interrupts from there. We also move creation of the pass-through port form probe to configure stage, as RMI driver does not enable transport interrupt until all functions are probed (we should change this, but this is a separate topic). We also try to clear the pending data before unmasking interrupts, because some devices like to spam the system with multiple 0xaa 0x00 announcements, which may interfere with us trying to query ID of the device. Fixes: c5e8848fc98e ("Input: synaptics-rmi4 - add support for F03") Cc: stable(a)vger.kernel.org Signed-off-by: Dmitry Torokhov <dmitry.torokhov(a)gmail.com> --- drivers/input/rmi4/rmi_f03.c | 64 +++++++++++++++++++++++++++++++++++++------- 1 file changed, 54 insertions(+), 10 deletions(-) diff --git a/drivers/input/rmi4/rmi_f03.c b/drivers/input/rmi4/rmi_f03.c index ad71a5e768dc4..7ccbb370a9a81 100644 --- a/drivers/input/rmi4/rmi_f03.c +++ b/drivers/input/rmi4/rmi_f03.c @@ -32,6 +32,7 @@ struct f03_data { struct rmi_function *fn; struct serio *serio; + bool serio_registered; unsigned int overwrite_buttons; @@ -138,6 +139,37 @@ static int rmi_f03_initialize(struct f03_data *f03) return 0; } +static int rmi_f03_pt_open(struct serio *serio) +{ + struct f03_data *f03 = serio->port_data; + struct rmi_function *fn = f03->fn; + const u8 ob_len = f03->rx_queue_length * RMI_F03_OB_SIZE; + const u16 data_addr = fn->fd.data_base_addr + RMI_F03_OB_OFFSET; + u8 obs[RMI_F03_QUEUE_LENGTH * RMI_F03_OB_SIZE]; + int error; + + /* + * Consume any pending data. Some devices like to spam with + * 0xaa 0x00 announcements which may confuse us as we try to + * probe the device. + */ + error = rmi_read_block(fn->rmi_dev, data_addr, &obs, ob_len); + if (!error) + rmi_dbg(RMI_DEBUG_FN, &fn->dev, + "%s: Consumed %*ph (%d) from PS2 guest\n", + __func__, ob_len, obs, ob_len); + + return fn->rmi_dev->driver->set_irq_bits(fn->rmi_dev, fn->irq_mask); +} + +static void rmi_f03_pt_close(struct serio *serio) +{ + struct f03_data *f03 = serio->port_data; + struct rmi_function *fn = f03->fn; + + fn->rmi_dev->driver->clear_irq_bits(fn->rmi_dev, fn->irq_mask); +} + static int rmi_f03_register_pt(struct f03_data *f03) { struct serio *serio; @@ -148,6 +180,8 @@ static int rmi_f03_register_pt(struct f03_data *f03) serio->id.type = SERIO_PS_PSTHRU; serio->write = rmi_f03_pt_write; + serio->open = rmi_f03_pt_open; + serio->close = rmi_f03_pt_close; serio->port_data = f03; strlcpy(serio->name, "Synaptics RMI4 PS/2 pass-through", @@ -184,17 +218,27 @@ static int rmi_f03_probe(struct rmi_function *fn) f03->device_count); dev_set_drvdata(dev, f03); - - error = rmi_f03_register_pt(f03); - if (error) - return error; - return 0; } static int rmi_f03_config(struct rmi_function *fn) { - fn->rmi_dev->driver->set_irq_bits(fn->rmi_dev, fn->irq_mask); + struct f03_data *f03 = dev_get_drvdata(&fn->dev); + int error; + + if (!f03->serio_registered) { + error = rmi_f03_register_pt(f03); + if (error) + return error; + + f03->serio_registered = true; + } else { + /* + * We must be re-configuring the sensor, just enable + * interrupts for this function. + */ + fn->rmi_dev->driver->set_irq_bits(fn->rmi_dev, fn->irq_mask); + } return 0; } @@ -204,7 +248,7 @@ static int rmi_f03_attention(struct rmi_function *fn, unsigned long *irq_bits) struct rmi_device *rmi_dev = fn->rmi_dev; struct rmi_driver_data *drvdata = dev_get_drvdata(&rmi_dev->dev); struct f03_data *f03 = dev_get_drvdata(&fn->dev); - u16 data_addr = fn->fd.data_base_addr; + const u16 data_addr = fn->fd.data_base_addr + RMI_F03_OB_OFFSET; const u8 ob_len = f03->rx_queue_length * RMI_F03_OB_SIZE; u8 obs[RMI_F03_QUEUE_LENGTH * RMI_F03_OB_SIZE]; u8 ob_status; @@ -226,8 +270,7 @@ static int rmi_f03_attention(struct rmi_function *fn, unsigned long *irq_bits) drvdata->attn_data.size -= ob_len; } else { /* Grab all of the data registers, and check them for data */ - error = rmi_read_block(fn->rmi_dev, data_addr + RMI_F03_OB_OFFSET, - &obs, ob_len); + error = rmi_read_block(fn->rmi_dev, data_addr, &obs, ob_len); if (error) { dev_err(&fn->dev, "%s: Failed to read F03 output buffers: %d\n", @@ -266,7 +309,8 @@ static void rmi_f03_remove(struct rmi_function *fn) { struct f03_data *f03 = dev_get_drvdata(&fn->dev); - serio_unregister_port(f03->serio); + if (f03->serio_registered) + serio_unregister_port(f03->serio); } struct rmi_function_handler rmi_f03_handler = { -- 2.16.0.rc1.238.g530d649a79-goog

7 years, 1 month

2
1
0 0

[Linux-stable-mirror] [PATCHv2] mm, page_vma_mapped: Drop faulty pointer arithmetics in check_pte()

by Kirill A. Shutemov

Tetsuo reported random crashes under memory pressure on 32-bit x86 system and tracked down to change that introduced page_vma_mapped_walk(). The root cause of the issue is the faulty pointer math in check_pte(). As ->pte may point to an arbitrary page we have to check that they are belong to the section before doing math. Otherwise it may lead to weird results. It wasn't noticed until now as mem_map[] is virtually contiguous on flatmem or vmemmap sparsemem. Pointer arithmetic just works against all 'struct page' pointers. But with classic sparsemem, it doesn't because each section memap is allocated separately and so consecutive pfns crossing two sections might have struct pages at completely unrelated addresses. Let's restructure code a bit and replace pointer arithmetic with operations on pfns. Signed-off-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Reported-by: Tetsuo Handa <penguin-kernel(a)i-love.sakura.ne.jp> Fixes: ace71a19cec5 ("mm: introduce page_vma_mapped_walk()") Cc: stable(a)vger.kernel.org --- v2: - Do not use uninitialized 'pfn' for !MIGRATION case (Michal) --- include/linux/swapops.h | 21 +++++++++++++++++ mm/page_vma_mapped.c | 63 +++++++++++++++++++++++++++++-------------------- 2 files changed, 59 insertions(+), 25 deletions(-) diff --git a/include/linux/swapops.h b/include/linux/swapops.h index 9c5a2628d6ce..1d3877c39a00 100644 --- a/include/linux/swapops.h +++ b/include/linux/swapops.h @@ -124,6 +124,11 @@ static inline bool is_write_device_private_entry(swp_entry_t entry) return unlikely(swp_type(entry) == SWP_DEVICE_WRITE); } +static inline unsigned long device_private_entry_to_pfn(swp_entry_t entry) +{ + return swp_offset(entry); +} + static inline struct page *device_private_entry_to_page(swp_entry_t entry) { return pfn_to_page(swp_offset(entry)); @@ -154,6 +159,11 @@ static inline bool is_write_device_private_entry(swp_entry_t entry) return false; } +static inline unsigned long device_private_entry_to_pfn(swp_entry_t entry) +{ + return 0; +} + static inline struct page *device_private_entry_to_page(swp_entry_t entry) { return NULL; @@ -189,6 +199,11 @@ static inline int is_write_migration_entry(swp_entry_t entry) return unlikely(swp_type(entry) == SWP_MIGRATION_WRITE); } +static inline unsigned long migration_entry_to_pfn(swp_entry_t entry) +{ + return swp_offset(entry); +} + static inline struct page *migration_entry_to_page(swp_entry_t entry) { struct page *p = pfn_to_page(swp_offset(entry)); @@ -218,6 +233,12 @@ static inline int is_migration_entry(swp_entry_t swp) { return 0; } + +static inline unsigned long migration_entry_to_pfn(swp_entry_t entry) +{ + return 0; +} + static inline struct page *migration_entry_to_page(swp_entry_t entry) { return NULL; diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c index d22b84310f6d..956015614395 100644 --- a/mm/page_vma_mapped.c +++ b/mm/page_vma_mapped.c @@ -30,10 +30,29 @@ static bool map_pte(struct page_vma_mapped_walk *pvmw) return true; } +/** + * check_pte - check if @pvmw->page is mapped at the @pvmw->pte + * + * page_vma_mapped_walk() found a place where @pvmw->page is *potentially* + * mapped. check_pte() has to validate this. + * + * @pvmw->pte may point to empty PTE, swap PTE or PTE pointing to arbitrary + * page. + * + * If PVMW_MIGRATION flag is set, returns true if @pvmw->pte contains migration + * entry that points to @pvmw->page or any subpage in case of THP. + * + * If PVMW_MIGRATION flag is not set, returns true if @pvmw->pte points to + * @pvmw->page or any subpage in case of THP. + * + * Otherwise, return false. + * + */ static bool check_pte(struct page_vma_mapped_walk *pvmw) { + unsigned long pfn; + if (pvmw->flags & PVMW_MIGRATION) { -#ifdef CONFIG_MIGRATION swp_entry_t entry; if (!is_swap_pte(*pvmw->pte)) return false; @@ -41,37 +60,31 @@ static bool check_pte(struct page_vma_mapped_walk *pvmw) if (!is_migration_entry(entry)) return false; - if (migration_entry_to_page(entry) - pvmw->page >= - hpage_nr_pages(pvmw->page)) { - return false; - } - if (migration_entry_to_page(entry) < pvmw->page) - return false; -#else - WARN_ON_ONCE(1); -#endif - } else { - if (is_swap_pte(*pvmw->pte)) { - swp_entry_t entry; - entry = pte_to_swp_entry(*pvmw->pte); - if (is_device_private_entry(entry) && - device_private_entry_to_page(entry) == pvmw->page) - return true; - } + pfn = migration_entry_to_pfn(entry); + } else if (is_swap_pte(*pvmw->pte)) { + swp_entry_t entry; - if (!pte_present(*pvmw->pte)) + /* Handle un-addressable ZONE_DEVICE memory */ + entry = pte_to_swp_entry(*pvmw->pte); + if (!is_device_private_entry(entry)) return false; - /* THP can be referenced by any subpage */ - if (pte_page(*pvmw->pte) - pvmw->page >= - hpage_nr_pages(pvmw->page)) { - return false; - } - if (pte_page(*pvmw->pte) < pvmw->page) + pfn = device_private_entry_to_pfn(entry); + } else { + if (!pte_present(*pvmw->pte)) return false; + + pfn = pte_pfn(*pvmw->pte); } + if (pfn < page_to_pfn(pvmw->page)) + return false; + + /* THP can be referenced by any subpage */ + if (pfn - page_to_pfn(pvmw->page) >= hpage_nr_pages(pvmw->page)) + return false; + return true; } -- 2.15.1

7 years, 1 month

3
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror