From dc0e36581eb2da1aa3c63ceeff0f10ef1e899b2a Mon Sep 17 00:00:00 2001 From: Dirk Mueller Date: Wed, 25 Jul 2018 13:10:28 +0200 Subject: arm64: Check for errata before evaluating cpu features Since commit d3aec8a28be3b8 ("arm64: capabilities: Restrict KPTI detection to boot-time CPUs") we rely on errata flags being already populated during feature enumeration. The order of errata and features was flipped as part of commit ed478b3f9e4a ("arm64: capabilities: Group handling of features and errata workarounds"). Return to the orginal order of errata and feature evaluation to ensure errata flags are present during feature evaluation. Fixes: ed478b3f9e4a ("arm64: capabilities: Group handling of features and errata workarounds") CC: Suzuki K Poulose CC: Marc Zyngier Signed-off-by: Dirk Mueller Reviewed-by: Suzuki K Poulose Signed-off-by: Will Deacon --- arch/arm64/kernel/cpufeature.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm64') diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c index f24892a40d2c..c6d80743f4ed 100644 --- a/arch/arm64/kernel/cpufeature.c +++ b/arch/arm64/kernel/cpufeature.c @@ -1351,9 +1351,9 @@ static void __update_cpu_capabilities(const struct arm64_cpu_capabilities *caps, static void update_cpu_capabilities(u16 scope_mask) { - __update_cpu_capabilities(arm64_features, scope_mask, "detected:"); __update_cpu_capabilities(arm64_errata, scope_mask, "enabling workaround for"); + __update_cpu_capabilities(arm64_features, scope_mask, "detected:"); } static int __enable_cpu_capability(void *arg) @@ -1408,8 +1408,8 @@ __enable_cpu_capabilities(const struct arm64_cpu_capabilities *caps, static void __init enable_cpu_capabilities(u16 scope_mask) { - __enable_cpu_capabilities(arm64_features, scope_mask); __enable_cpu_capabilities(arm64_errata, scope_mask); + __enable_cpu_capabilities(arm64_features, scope_mask); } /* -- cgit v1.2.3 From 7b0eb6b41a08fa1fa0d04b1c53becd62b5fbfaee Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Mon, 23 Jul 2018 10:18:23 -0400 Subject: arm64: fix vmemmap BUILD_BUG_ON() triggering on !vmemmap setups Arnd reports the following arm64 randconfig build error with the PSI patches that add another page flag: /git/arm-soc/arch/arm64/mm/init.c: In function 'mem_init': /git/arm-soc/include/linux/compiler.h:357:38: error: call to '__compiletime_assert_618' declared with attribute error: BUILD_BUG_ON failed: sizeof(struct page) > (1 << STRUCT_PAGE_MAX_SHIFT) The additional page flag causes other information stored in page->flags to get bumped into their own struct page member: #if SECTIONS_WIDTH+ZONES_WIDTH+NODES_SHIFT+LAST_CPUPID_SHIFT <= BITS_PER_LONG - NR_PAGEFLAGS #define LAST_CPUPID_WIDTH LAST_CPUPID_SHIFT #else #define LAST_CPUPID_WIDTH 0 #endif #if defined(CONFIG_NUMA_BALANCING) && LAST_CPUPID_WIDTH == 0 #define LAST_CPUPID_NOT_IN_PAGE_FLAGS #endif which in turn causes the struct page size to exceed the size set in STRUCT_PAGE_MAX_SHIFT. This value is an an estimate used to size the VMEMMAP page array according to address space and struct page size. However, the check is performed - and triggers here - on a !VMEMMAP config, which consumes an additional 22 page bits for the sparse section id. When VMEMMAP is enabled, those bits are returned, cpupid doesn't need its own member, and the page passes the VMEMMAP check. Restrict that check to the situation it was meant to check: that we are sizing the VMEMMAP page array correctly. Says Arnd: Further experiments show that the build error already existed before, but was only triggered with larger values of CONFIG_NR_CPU and/or CONFIG_NODES_SHIFT that might be used in actual configurations but not in randconfig builds. With longer CPU and node masks, I could recreate the problem with kernels as old as linux-4.7 when arm64 NUMA support got added. Reported-by: Arnd Bergmann Tested-by: Arnd Bergmann Cc: stable@vger.kernel.org Fixes: 1a2db300348b ("arm64, numa: Add NUMA support for arm64 platforms.") Fixes: 3e1907d5bf5a ("arm64: mm: move vmemmap region right below the linear region") Signed-off-by: Johannes Weiner Signed-off-by: Will Deacon --- arch/arm64/mm/init.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/arm64') diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 325cfb3b858a..9abf8a1e7b25 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -611,11 +611,13 @@ void __init mem_init(void) BUILD_BUG_ON(TASK_SIZE_32 > TASK_SIZE_64); #endif +#ifdef CONFIG_SPARSEMEM_VMEMMAP /* * Make sure we chose the upper bound of sizeof(struct page) - * correctly. + * correctly when sizing the VMEMMAP array. */ BUILD_BUG_ON(sizeof(struct page) > (1 << STRUCT_PAGE_MAX_SHIFT)); +#endif if (PAGE_SIZE >= 16384 && get_num_physpages() <= 128) { extern int sysctl_overcommit_memory; -- cgit v1.2.3 From 2c4541e24c55e2847bede93e33d749280edd429a Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Thu, 26 Jul 2018 16:37:30 -0700 Subject: mm: use vma_init() to initialize VMAs on stack and data segments Make sure to initialize all VMAs properly, not only those which come from vm_area_cachep. Link: http://lkml.kernel.org/r/20180724121139.62570-3-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Acked-by: Linus Torvalds Reviewed-by: Andrew Morton Cc: Dmitry Vyukov Cc: Oleg Nesterov Cc: Andrea Arcangeli Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/process.c | 1 + arch/arm/mach-rpc/ecard.c | 2 +- arch/arm64/include/asm/tlb.h | 4 +++- arch/arm64/mm/hugetlbpage.c | 7 +++++-- arch/ia64/include/asm/tlb.h | 2 +- arch/ia64/mm/init.c | 2 +- arch/x86/um/mem_32.c | 2 +- fs/hugetlbfs/inode.c | 2 ++ mm/mempolicy.c | 1 + mm/shmem.c | 1 + 10 files changed, 17 insertions(+), 7 deletions(-) (limited to 'arch/arm64') diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 225d1c58d2de..d9c299133111 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -338,6 +338,7 @@ static struct vm_area_struct gate_vma = { static int __init gate_vma_init(void) { + vma_init(&gate_vma, NULL); gate_vma.vm_page_prot = PAGE_READONLY_EXEC; return 0; } diff --git a/arch/arm/mach-rpc/ecard.c b/arch/arm/mach-rpc/ecard.c index 39aef4876ed4..8db62cc54a6a 100644 --- a/arch/arm/mach-rpc/ecard.c +++ b/arch/arm/mach-rpc/ecard.c @@ -237,8 +237,8 @@ static void ecard_init_pgtables(struct mm_struct *mm) memcpy(dst_pgd, src_pgd, sizeof(pgd_t) * (EASI_SIZE / PGDIR_SIZE)); + vma_init(&vma, mm); vma.vm_flags = VM_EXEC; - vma.vm_mm = mm; flush_tlb_range(&vma, IO_START, IO_START + IO_SIZE); flush_tlb_range(&vma, EASI_START, EASI_START + EASI_SIZE); diff --git a/arch/arm64/include/asm/tlb.h b/arch/arm64/include/asm/tlb.h index ffdaea7954bb..d87f2d646caa 100644 --- a/arch/arm64/include/asm/tlb.h +++ b/arch/arm64/include/asm/tlb.h @@ -37,7 +37,9 @@ static inline void __tlb_remove_table(void *_table) static inline void tlb_flush(struct mmu_gather *tlb) { - struct vm_area_struct vma = { .vm_mm = tlb->mm, }; + struct vm_area_struct vma; + + vma_init(&vma, tlb->mm); /* * The ASID allocator will either invalidate the ASID or mark diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c index ecc6818191df..1854e49aa18a 100644 --- a/arch/arm64/mm/hugetlbpage.c +++ b/arch/arm64/mm/hugetlbpage.c @@ -108,11 +108,13 @@ static pte_t get_clear_flush(struct mm_struct *mm, unsigned long pgsize, unsigned long ncontig) { - struct vm_area_struct vma = { .vm_mm = mm }; + struct vm_area_struct vma; pte_t orig_pte = huge_ptep_get(ptep); bool valid = pte_valid(orig_pte); unsigned long i, saddr = addr; + vma_init(&vma, mm); + for (i = 0; i < ncontig; i++, addr += pgsize, ptep++) { pte_t pte = ptep_get_and_clear(mm, addr, ptep); @@ -145,9 +147,10 @@ static void clear_flush(struct mm_struct *mm, unsigned long pgsize, unsigned long ncontig) { - struct vm_area_struct vma = { .vm_mm = mm }; + struct vm_area_struct vma; unsigned long i, saddr = addr; + vma_init(&vma, mm); for (i = 0; i < ncontig; i++, addr += pgsize, ptep++) pte_clear(mm, addr, ptep); diff --git a/arch/ia64/include/asm/tlb.h b/arch/ia64/include/asm/tlb.h index 44f0ac0df308..db89e7306081 100644 --- a/arch/ia64/include/asm/tlb.h +++ b/arch/ia64/include/asm/tlb.h @@ -120,7 +120,7 @@ ia64_tlb_flush_mmu_tlbonly(struct mmu_gather *tlb, unsigned long start, unsigned */ struct vm_area_struct vma; - vma.vm_mm = tlb->mm; + vma_init(&vma, tlb->mm); /* flush the address range from the tlb: */ flush_tlb_range(&vma, start, end); /* now flush the virt. page-table area mapping the address range: */ diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index bdb14a369137..e6c6dfd98de2 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -273,7 +273,7 @@ static struct vm_area_struct gate_vma; static int __init gate_vma_init(void) { - gate_vma.vm_mm = NULL; + vma_init(&gate_vma, NULL); gate_vma.vm_start = FIXADDR_USER_START; gate_vma.vm_end = FIXADDR_USER_END; gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; diff --git a/arch/x86/um/mem_32.c b/arch/x86/um/mem_32.c index 744afdc18cf3..56c44d865f7b 100644 --- a/arch/x86/um/mem_32.c +++ b/arch/x86/um/mem_32.c @@ -16,7 +16,7 @@ static int __init gate_vma_init(void) if (!FIXADDR_USER_START) return 0; - gate_vma.vm_mm = NULL; + vma_init(&gate_vma, NULL); gate_vma.vm_start = FIXADDR_USER_START; gate_vma.vm_end = FIXADDR_USER_END; gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index d508c7844681..40d4c66c7751 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -411,6 +411,7 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart, bool truncate_op = (lend == LLONG_MAX); memset(&pseudo_vma, 0, sizeof(struct vm_area_struct)); + vma_init(&pseudo_vma, current->mm); pseudo_vma.vm_flags = (VM_HUGETLB | VM_MAYSHARE | VM_SHARED); pagevec_init(&pvec); next = start; @@ -595,6 +596,7 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, * as input to create an allocation policy. */ memset(&pseudo_vma, 0, sizeof(struct vm_area_struct)); + vma_init(&pseudo_vma, mm); pseudo_vma.vm_flags = (VM_HUGETLB | VM_MAYSHARE | VM_SHARED); pseudo_vma.vm_file = file; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 9ac49ef17b4e..01f1a14facc4 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -2505,6 +2505,7 @@ void mpol_shared_policy_init(struct shared_policy *sp, struct mempolicy *mpol) /* Create pseudo-vma that contains just the policy */ memset(&pvma, 0, sizeof(struct vm_area_struct)); + vma_init(&pvma, NULL); pvma.vm_end = TASK_SIZE; /* policy covers entire file */ mpol_set_shared_policy(sp, &pvma, new); /* adds ref */ diff --git a/mm/shmem.c b/mm/shmem.c index 2cab84403055..41b9bbf24e16 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1421,6 +1421,7 @@ static void shmem_pseudo_vma_init(struct vm_area_struct *vma, { /* Create a pseudo vma that just contains the policy */ memset(vma, 0, sizeof(*vma)); + vma_init(vma, NULL); /* Bias interleave by inode number to distribute better across nodes */ vma->vm_pgoff = index + info->vfs_inode.i_ino; vma->vm_policy = mpol_shared_policy_lookup(&info->policy, index); -- cgit v1.2.3 From 8b11ec1b5ffb54f71cb5a5e5c8c4d36e5d113085 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 1 Aug 2018 13:43:38 -0700 Subject: mm: do not initialize TLB stack vma's with vma_init() Commit 2c4541e24c55 ("mm: use vma_init() to initialize VMAs on stack and data segments") tried to initialize various left-over ad-hoc vma's "properly", but actually made things worse for the temporary vma's used for TLB flushing. vma_init() doesn't actually initialize all of the vma, just a few fields, so doing something like - struct vm_area_struct vma = { .vm_mm = tlb->mm, }; + struct vm_area_struct vma; + + vma_init(&vma, tlb->mm); was actually very bad: instead of having a nicely initialized vma with every field but "vm_mm" zeroed, you'd have an entirely uninitialized vma with only a couple of fields initialized. And they weren't even fields that the code in question mostly cared about. The flush_tlb_range() function takes a "struct vma" rather than a "struct mm_struct", because a few architectures actually care about what kind of range it is - being able to only do an ITLB flush if it's a range that doesn't have data accesses enabled, for example. And all the normal users already have the vma for doing the range invalidation. But a few people want to call flush_tlb_range() with a range they just made up, so they also end up using a made-up vma. x86 just has a special "flush_tlb_mm_range()" function for this, but other architectures (arm and ia64) do the "use fake vma" thing instead, and thus got caught up in the vma_init() changes. At the same time, the TLB flushing code really doesn't care about most other fields in the vma, so vma_init() is just unnecessary and pointless. This fixes things by having an explicit "this is just an initializer for the TLB flush" initializer macro, which is used by the arm/arm64/ia64 people who mis-use this interface with just a dummy vma. Fixes: 2c4541e24c55 ("mm: use vma_init() to initialize VMAs on stack and data segments") Cc: Dmitry Vyukov Cc: Oleg Nesterov Cc: Andrea Arcangeli Cc: Kirill Shutemov Cc: Andrew Morton Cc: John Stultz Cc: Hugh Dickins Signed-off-by: Linus Torvalds --- arch/arm/mach-rpc/ecard.c | 5 +---- arch/arm64/include/asm/tlb.h | 4 +--- arch/arm64/mm/hugetlbpage.c | 10 ++++------ arch/ia64/include/asm/tlb.h | 7 +++---- include/linux/mm.h | 3 +++ 5 files changed, 12 insertions(+), 17 deletions(-) (limited to 'arch/arm64') diff --git a/arch/arm/mach-rpc/ecard.c b/arch/arm/mach-rpc/ecard.c index 8db62cc54a6a..04b2f22c2739 100644 --- a/arch/arm/mach-rpc/ecard.c +++ b/arch/arm/mach-rpc/ecard.c @@ -212,7 +212,7 @@ static DEFINE_MUTEX(ecard_mutex); */ static void ecard_init_pgtables(struct mm_struct *mm) { - struct vm_area_struct vma; + struct vm_area_struct vma = TLB_FLUSH_VMA(mm, VM_EXEC); /* We want to set up the page tables for the following mapping: * Virtual Physical @@ -237,9 +237,6 @@ static void ecard_init_pgtables(struct mm_struct *mm) memcpy(dst_pgd, src_pgd, sizeof(pgd_t) * (EASI_SIZE / PGDIR_SIZE)); - vma_init(&vma, mm); - vma.vm_flags = VM_EXEC; - flush_tlb_range(&vma, IO_START, IO_START + IO_SIZE); flush_tlb_range(&vma, EASI_START, EASI_START + EASI_SIZE); } diff --git a/arch/arm64/include/asm/tlb.h b/arch/arm64/include/asm/tlb.h index d87f2d646caa..0ad1cf233470 100644 --- a/arch/arm64/include/asm/tlb.h +++ b/arch/arm64/include/asm/tlb.h @@ -37,9 +37,7 @@ static inline void __tlb_remove_table(void *_table) static inline void tlb_flush(struct mmu_gather *tlb) { - struct vm_area_struct vma; - - vma_init(&vma, tlb->mm); + struct vm_area_struct vma = TLB_FLUSH_VMA(tlb->mm, 0); /* * The ASID allocator will either invalidate the ASID or mark diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c index 1854e49aa18a..192b3ba07075 100644 --- a/arch/arm64/mm/hugetlbpage.c +++ b/arch/arm64/mm/hugetlbpage.c @@ -108,13 +108,10 @@ static pte_t get_clear_flush(struct mm_struct *mm, unsigned long pgsize, unsigned long ncontig) { - struct vm_area_struct vma; pte_t orig_pte = huge_ptep_get(ptep); bool valid = pte_valid(orig_pte); unsigned long i, saddr = addr; - vma_init(&vma, mm); - for (i = 0; i < ncontig; i++, addr += pgsize, ptep++) { pte_t pte = ptep_get_and_clear(mm, addr, ptep); @@ -127,8 +124,10 @@ static pte_t get_clear_flush(struct mm_struct *mm, orig_pte = pte_mkdirty(orig_pte); } - if (valid) + if (valid) { + struct vm_area_struct vma = TLB_FLUSH_VMA(mm, 0); flush_tlb_range(&vma, saddr, addr); + } return orig_pte; } @@ -147,10 +146,9 @@ static void clear_flush(struct mm_struct *mm, unsigned long pgsize, unsigned long ncontig) { - struct vm_area_struct vma; + struct vm_area_struct vma = TLB_FLUSH_VMA(mm, 0); unsigned long i, saddr = addr; - vma_init(&vma, mm); for (i = 0; i < ncontig; i++, addr += pgsize, ptep++) pte_clear(mm, addr, ptep); diff --git a/arch/ia64/include/asm/tlb.h b/arch/ia64/include/asm/tlb.h index db89e7306081..516355a774bf 100644 --- a/arch/ia64/include/asm/tlb.h +++ b/arch/ia64/include/asm/tlb.h @@ -115,12 +115,11 @@ ia64_tlb_flush_mmu_tlbonly(struct mmu_gather *tlb, unsigned long start, unsigned flush_tlb_all(); } else { /* - * XXX fix me: flush_tlb_range() should take an mm pointer instead of a - * vma pointer. + * flush_tlb_range() takes a vma instead of a mm pointer because + * some architectures want the vm_flags for ITLB/DTLB flush. */ - struct vm_area_struct vma; + struct vm_area_struct vma = TLB_FLUSH_VMA(tlb->mm, 0); - vma_init(&vma, tlb->mm); /* flush the address range from the tlb: */ flush_tlb_range(&vma, start, end); /* now flush the virt. page-table area mapping the address range: */ diff --git a/include/linux/mm.h b/include/linux/mm.h index 7ba6d356d18f..68a5121694ef 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -466,6 +466,9 @@ static inline void vma_set_anonymous(struct vm_area_struct *vma) vma->vm_ops = NULL; } +/* flush_tlb_range() takes a vma, not a mm, and can care about flags */ +#define TLB_FLUSH_VMA(mm,flags) { .vm_mm = (mm), .vm_flags = (flags) } + struct mmu_gather; struct inode; -- cgit v1.2.3