From 00c30e0681bf5563c8670c0ab419886f56626430 Mon Sep 17 00:00:00 2001 From: Jonas Bonn Date: Thu, 28 Feb 2013 06:37:05 +0100 Subject: asm-generic: move cmpxchg*_local defs to cmpxchg.h asm/cmpxchg.h can be included on its own and needs to be self-consistent. The definitions for the cmpxchg*_local macros, as such, need to be part of this file. This fixes a build issue on OpenRISC since the system.h smashing patch 96f951edb1f1bdbbc99b0cd458f9808bb83d58ae that introdued the direct inclusion asm/cmpxchg.h into linux/llist.h. CC: David Howells Signed-off-by: Jonas Bonn Acked-by: Arnd Bergmann --- include/asm-generic/atomic.h | 6 ------ include/asm-generic/cmpxchg.h | 10 ++++++++++ 2 files changed, 10 insertions(+), 6 deletions(-) (limited to 'include/asm-generic') diff --git a/include/asm-generic/atomic.h b/include/asm-generic/atomic.h index 1ced6413ea03..33bd2de3bc1e 100644 --- a/include/asm-generic/atomic.h +++ b/include/asm-generic/atomic.h @@ -136,12 +136,6 @@ static inline void atomic_dec(atomic_t *v) #define atomic_xchg(ptr, v) (xchg(&(ptr)->counter, (v))) #define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new))) -#define cmpxchg_local(ptr, o, n) \ - ((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), (unsigned long)(o),\ - (unsigned long)(n), sizeof(*(ptr)))) - -#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) - static inline int __atomic_add_unless(atomic_t *v, int a, int u) { int c, old; diff --git a/include/asm-generic/cmpxchg.h b/include/asm-generic/cmpxchg.h index 14883026015d..811fb1e9b061 100644 --- a/include/asm-generic/cmpxchg.h +++ b/include/asm-generic/cmpxchg.h @@ -92,6 +92,16 @@ unsigned long __xchg(unsigned long x, volatile void *ptr, int size) */ #include +#ifndef cmpxchg_local +#define cmpxchg_local(ptr, o, n) \ + ((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), (unsigned long)(o),\ + (unsigned long)(n), sizeof(*(ptr)))) +#endif + +#ifndef cmpxchg64_local +#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) +#endif + #define cmpxchg(ptr, o, n) cmpxchg_local((ptr), (o), (n)) #define cmpxchg64(ptr, o, n) cmpxchg64_local((ptr), (o), (n)) -- cgit v1.2.3-73-gaa49b From 1de14c3c5cbc9bb17e9dcc648cda51c0c85d54b9 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 12 Apr 2013 16:23:54 -0700 Subject: x86-32: Fix possible incomplete TLB invalidate with PAE pagetables This patch attempts to fix: https://bugzilla.kernel.org/show_bug.cgi?id=56461 The symptom is a crash and messages like this: chrome: Corrupted page table at address 34a03000 *pdpt = 0000000000000000 *pde = 0000000000000000 Bad pagetable: 000f [#1] PREEMPT SMP Ingo guesses this got introduced by commit 611ae8e3f520 ("x86/tlb: enable tlb flush range support for x86") since that code started to free unused pagetables. On x86-32 PAE kernels, that new code has the potential to free an entire PMD page and will clear one of the four page-directory-pointer-table (aka pgd_t entries). The hardware aggressively "caches" these top-level entries and invlpg does not actually affect the CPU's copy. If we clear one we *HAVE* to do a full TLB flush, otherwise we might continue using a freed pmd page. (note, we do this properly on the population side in pud_populate()). This patch tracks whenever we clear one of these entries in the 'struct mmu_gather', and ensures that we follow up with a full tlb flush. BTW, I disassembled and checked that: if (tlb->fullmm == 0) and if (!tlb->fullmm && !tlb->need_flush_all) generate essentially the same code, so there should be zero impact there to the !PAE case. Signed-off-by: Dave Hansen Cc: Peter Anvin Cc: Ingo Molnar Cc: Artem S Tashkinov Signed-off-by: Linus Torvalds --- arch/x86/include/asm/tlb.h | 2 +- arch/x86/mm/pgtable.c | 7 +++++++ include/asm-generic/tlb.h | 7 ++++++- mm/memory.c | 1 + 4 files changed, 15 insertions(+), 2 deletions(-) (limited to 'include/asm-generic') diff --git a/arch/x86/include/asm/tlb.h b/arch/x86/include/asm/tlb.h index 4fef20773b8f..c7797307fc2b 100644 --- a/arch/x86/include/asm/tlb.h +++ b/arch/x86/include/asm/tlb.h @@ -7,7 +7,7 @@ #define tlb_flush(tlb) \ { \ - if (tlb->fullmm == 0) \ + if (!tlb->fullmm && !tlb->need_flush_all) \ flush_tlb_mm_range(tlb->mm, tlb->start, tlb->end, 0UL); \ else \ flush_tlb_mm_range(tlb->mm, 0UL, TLB_FLUSH_ALL, 0UL); \ diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 193350b51f90..17fda6a8b3c2 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -58,6 +58,13 @@ void ___pte_free_tlb(struct mmu_gather *tlb, struct page *pte) void ___pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd) { paravirt_release_pmd(__pa(pmd) >> PAGE_SHIFT); + /* + * NOTE! For PAE, any changes to the top page-directory-pointer-table + * entries need a full cr3 reload to flush. + */ +#ifdef CONFIG_X86_PAE + tlb->need_flush_all = 1; +#endif tlb_remove_page(tlb, virt_to_page(pmd)); } diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h index 25f01d0bc149..b1b1fa6ffffe 100644 --- a/include/asm-generic/tlb.h +++ b/include/asm-generic/tlb.h @@ -99,7 +99,12 @@ struct mmu_gather { unsigned int need_flush : 1, /* Did free PTEs */ fast_mode : 1; /* No batching */ - unsigned int fullmm; + /* we are in the middle of an operation to clear + * a full mm and can make some optimizations */ + unsigned int fullmm : 1, + /* we have performed an operation which + * requires a complete flush of the tlb */ + need_flush_all : 1; struct mmu_gather_batch *active; struct mmu_gather_batch local; diff --git a/mm/memory.c b/mm/memory.c index 494526ae024a..13cbc420fead 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -216,6 +216,7 @@ void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, bool fullmm) tlb->mm = mm; tlb->fullmm = fullmm; + tlb->need_flush_all = 0; tlb->start = -1UL; tlb->end = 0; tlb->need_flush = 0; -- cgit v1.2.3-73-gaa49b From 106c992a5ebef28193cf5958e49ceff5e4aebb04 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Mon, 29 Apr 2013 15:07:23 -0700 Subject: mm/hugetlb: add more arch-defined huge_pte functions Commit abf09bed3cce ("s390/mm: implement software dirty bits") introduced another difference in the pte layout vs. the pmd layout on s390, thoroughly breaking the s390 support for hugetlbfs. This requires replacing some more pte_xxx functions in mm/hugetlbfs.c with a huge_pte_xxx version. This patch introduces those huge_pte_xxx functions and their generic implementation in asm-generic/hugetlb.h, which will now be included on all architectures supporting hugetlbfs apart from s390. This change will be a no-op for those architectures. [akpm@linux-foundation.org: fix warning] Signed-off-by: Gerald Schaefer Cc: Mel Gorman Cc: Hugh Dickins Cc: Hillf Danton Acked-by: Michal Hocko [for !s390 parts] Cc: Tony Luck Cc: Fenghua Yu Cc: Ralf Baechle Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Paul Mundt Cc: "David S. Miller" Cc: Chris Metcalf Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/include/asm/hugetlb.h | 1 + arch/mips/include/asm/hugetlb.h | 1 + arch/powerpc/include/asm/hugetlb.h | 1 + arch/s390/include/asm/hugetlb.h | 56 +++++++++++++++++++++- arch/s390/include/asm/pgtable.h | 95 ++++++++++++++++---------------------- arch/s390/mm/hugetlbpage.c | 2 +- arch/sh/include/asm/hugetlb.h | 1 + arch/sparc/include/asm/hugetlb.h | 1 + arch/tile/include/asm/hugetlb.h | 1 + arch/x86/include/asm/hugetlb.h | 1 + include/asm-generic/hugetlb.h | 40 ++++++++++++++++ mm/hugetlb.c | 24 +++++----- 12 files changed, 156 insertions(+), 68 deletions(-) create mode 100644 include/asm-generic/hugetlb.h (limited to 'include/asm-generic') diff --git a/arch/ia64/include/asm/hugetlb.h b/arch/ia64/include/asm/hugetlb.h index 94eaa5bd5d0c..aa910054b8e7 100644 --- a/arch/ia64/include/asm/hugetlb.h +++ b/arch/ia64/include/asm/hugetlb.h @@ -2,6 +2,7 @@ #define _ASM_IA64_HUGETLB_H #include +#include void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, diff --git a/arch/mips/include/asm/hugetlb.h b/arch/mips/include/asm/hugetlb.h index ef99db994c2f..fe0d15d32660 100644 --- a/arch/mips/include/asm/hugetlb.h +++ b/arch/mips/include/asm/hugetlb.h @@ -10,6 +10,7 @@ #define __ASM_HUGETLB_H #include +#include static inline int is_hugepage_only_range(struct mm_struct *mm, diff --git a/arch/powerpc/include/asm/hugetlb.h b/arch/powerpc/include/asm/hugetlb.h index 62e11a32c4c2..4fcbd6b14a3a 100644 --- a/arch/powerpc/include/asm/hugetlb.h +++ b/arch/powerpc/include/asm/hugetlb.h @@ -3,6 +3,7 @@ #ifdef CONFIG_HUGETLB_PAGE #include +#include extern struct kmem_cache *hugepte_cache; diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h index 593753ee07f3..bd90359d6d22 100644 --- a/arch/s390/include/asm/hugetlb.h +++ b/arch/s390/include/asm/hugetlb.h @@ -114,7 +114,7 @@ static inline pte_t huge_ptep_get_and_clear(struct mm_struct *mm, #define huge_ptep_set_wrprotect(__mm, __addr, __ptep) \ ({ \ pte_t __pte = huge_ptep_get(__ptep); \ - if (pte_write(__pte)) { \ + if (huge_pte_write(__pte)) { \ huge_ptep_invalidate(__mm, __addr, __ptep); \ set_huge_pte_at(__mm, __addr, __ptep, \ huge_pte_wrprotect(__pte)); \ @@ -127,4 +127,58 @@ static inline void huge_ptep_clear_flush(struct vm_area_struct *vma, huge_ptep_invalidate(vma->vm_mm, address, ptep); } +static inline pte_t mk_huge_pte(struct page *page, pgprot_t pgprot) +{ + pte_t pte; + pmd_t pmd; + + pmd = mk_pmd_phys(page_to_phys(page), pgprot); + pte_val(pte) = pmd_val(pmd); + return pte; +} + +static inline int huge_pte_write(pte_t pte) +{ + pmd_t pmd; + + pmd_val(pmd) = pte_val(pte); + return pmd_write(pmd); +} + +static inline int huge_pte_dirty(pte_t pte) +{ + /* No dirty bit in the segment table entry. */ + return 0; +} + +static inline pte_t huge_pte_mkwrite(pte_t pte) +{ + pmd_t pmd; + + pmd_val(pmd) = pte_val(pte); + pte_val(pte) = pmd_val(pmd_mkwrite(pmd)); + return pte; +} + +static inline pte_t huge_pte_mkdirty(pte_t pte) +{ + /* No dirty bit in the segment table entry. */ + return pte; +} + +static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot) +{ + pmd_t pmd; + + pmd_val(pmd) = pte_val(pte); + pte_val(pte) = pmd_val(pmd_modify(pmd, newprot)); + return pte; +} + +static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pmd_clear((pmd_t *) ptep); +} + #endif /* _ASM_S390_HUGETLB_H */ diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 4a64c0e5428f..b4622915bd15 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -424,6 +424,13 @@ extern unsigned long MODULES_END; #define __S110 PAGE_RW #define __S111 PAGE_RW +/* + * Segment entry (large page) protection definitions. + */ +#define SEGMENT_NONE __pgprot(_HPAGE_TYPE_NONE) +#define SEGMENT_RO __pgprot(_HPAGE_TYPE_RO) +#define SEGMENT_RW __pgprot(_HPAGE_TYPE_RW) + static inline int mm_exclusive(struct mm_struct *mm) { return likely(mm == current->active_mm && @@ -914,26 +921,6 @@ static inline pte_t pte_mkspecial(pte_t pte) #ifdef CONFIG_HUGETLB_PAGE static inline pte_t pte_mkhuge(pte_t pte) { - /* - * PROT_NONE needs to be remapped from the pte type to the ste type. - * The HW invalid bit is also different for pte and ste. The pte - * invalid bit happens to be the same as the ste _SEGMENT_ENTRY_LARGE - * bit, so we don't have to clear it. - */ - if (pte_val(pte) & _PAGE_INVALID) { - if (pte_val(pte) & _PAGE_SWT) - pte_val(pte) |= _HPAGE_TYPE_NONE; - pte_val(pte) |= _SEGMENT_ENTRY_INV; - } - /* - * Clear SW pte bits, there are no SW bits in a segment table entry. - */ - pte_val(pte) &= ~(_PAGE_SWT | _PAGE_SWX | _PAGE_SWC | - _PAGE_SWR | _PAGE_SWW); - /* - * Also set the change-override bit because we don't need dirty bit - * tracking for hugetlbfs pages. - */ pte_val(pte) |= (_SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_CO); return pte; } @@ -1278,31 +1265,7 @@ static inline void __pmd_idte(unsigned long address, pmd_t *pmdp) } } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE - -#define SEGMENT_NONE __pgprot(_HPAGE_TYPE_NONE) -#define SEGMENT_RO __pgprot(_HPAGE_TYPE_RO) -#define SEGMENT_RW __pgprot(_HPAGE_TYPE_RW) - -#define __HAVE_ARCH_PGTABLE_DEPOSIT -extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pgtable_t pgtable); - -#define __HAVE_ARCH_PGTABLE_WITHDRAW -extern pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm); - -static inline int pmd_trans_splitting(pmd_t pmd) -{ - return pmd_val(pmd) & _SEGMENT_ENTRY_SPLIT; -} - -static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, - pmd_t *pmdp, pmd_t entry) -{ - if (!(pmd_val(entry) & _SEGMENT_ENTRY_INV) && MACHINE_HAS_EDAT1) - pmd_val(entry) |= _SEGMENT_ENTRY_CO; - *pmdp = entry; -} - +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLB_PAGE) static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot) { /* @@ -1323,10 +1286,11 @@ static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) return pmd; } -static inline pmd_t pmd_mkhuge(pmd_t pmd) +static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot) { - pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE; - return pmd; + pmd_t __pmd; + pmd_val(__pmd) = physpage + massage_pgprot_pmd(pgprot); + return __pmd; } static inline pmd_t pmd_mkwrite(pmd_t pmd) @@ -1336,6 +1300,34 @@ static inline pmd_t pmd_mkwrite(pmd_t pmd) pmd_val(pmd) &= ~_SEGMENT_ENTRY_RO; return pmd; } +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLB_PAGE */ + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + +#define __HAVE_ARCH_PGTABLE_DEPOSIT +extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pgtable_t pgtable); + +#define __HAVE_ARCH_PGTABLE_WITHDRAW +extern pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm); + +static inline int pmd_trans_splitting(pmd_t pmd) +{ + return pmd_val(pmd) & _SEGMENT_ENTRY_SPLIT; +} + +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t entry) +{ + if (!(pmd_val(entry) & _SEGMENT_ENTRY_INV) && MACHINE_HAS_EDAT1) + pmd_val(entry) |= _SEGMENT_ENTRY_CO; + *pmdp = entry; +} + +static inline pmd_t pmd_mkhuge(pmd_t pmd) +{ + pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE; + return pmd; +} static inline pmd_t pmd_wrprotect(pmd_t pmd) { @@ -1432,13 +1424,6 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm, } } -static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot) -{ - pmd_t __pmd; - pmd_val(__pmd) = physpage + massage_pgprot_pmd(pgprot); - return __pmd; -} - #define pfn_pmd(pfn, pgprot) mk_pmd_phys(__pa((pfn) << PAGE_SHIFT), (pgprot)) #define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 532525ec88c1..121089d57802 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -39,7 +39,7 @@ int arch_prepare_hugepage(struct page *page) if (!ptep) return -ENOMEM; - pte = mk_pte(page, PAGE_RW); + pte_val(pte) = addr; for (i = 0; i < PTRS_PER_PTE; i++) { set_pte_at(&init_mm, addr + i * PAGE_SIZE, ptep + i, pte); pte_val(pte) += PAGE_SIZE; diff --git a/arch/sh/include/asm/hugetlb.h b/arch/sh/include/asm/hugetlb.h index b3808c7d67b2..699255d6d1c6 100644 --- a/arch/sh/include/asm/hugetlb.h +++ b/arch/sh/include/asm/hugetlb.h @@ -3,6 +3,7 @@ #include #include +#include static inline int is_hugepage_only_range(struct mm_struct *mm, diff --git a/arch/sparc/include/asm/hugetlb.h b/arch/sparc/include/asm/hugetlb.h index 7eb57d245044..e4cab465b81f 100644 --- a/arch/sparc/include/asm/hugetlb.h +++ b/arch/sparc/include/asm/hugetlb.h @@ -2,6 +2,7 @@ #define _ASM_SPARC64_HUGETLB_H #include +#include void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, diff --git a/arch/tile/include/asm/hugetlb.h b/arch/tile/include/asm/hugetlb.h index 0f885af2b621..3257733003f8 100644 --- a/arch/tile/include/asm/hugetlb.h +++ b/arch/tile/include/asm/hugetlb.h @@ -16,6 +16,7 @@ #define _ASM_TILE_HUGETLB_H #include +#include static inline int is_hugepage_only_range(struct mm_struct *mm, diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h index bdd35dbd0605..a8091216963b 100644 --- a/arch/x86/include/asm/hugetlb.h +++ b/arch/x86/include/asm/hugetlb.h @@ -2,6 +2,7 @@ #define _ASM_X86_HUGETLB_H #include +#include static inline int is_hugepage_only_range(struct mm_struct *mm, diff --git a/include/asm-generic/hugetlb.h b/include/asm-generic/hugetlb.h new file mode 100644 index 000000000000..d06079c774a0 --- /dev/null +++ b/include/asm-generic/hugetlb.h @@ -0,0 +1,40 @@ +#ifndef _ASM_GENERIC_HUGETLB_H +#define _ASM_GENERIC_HUGETLB_H + +static inline pte_t mk_huge_pte(struct page *page, pgprot_t pgprot) +{ + return mk_pte(page, pgprot); +} + +static inline int huge_pte_write(pte_t pte) +{ + return pte_write(pte); +} + +static inline int huge_pte_dirty(pte_t pte) +{ + return pte_dirty(pte); +} + +static inline pte_t huge_pte_mkwrite(pte_t pte) +{ + return pte_mkwrite(pte); +} + +static inline pte_t huge_pte_mkdirty(pte_t pte) +{ + return pte_mkdirty(pte); +} + +static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot) +{ + return pte_modify(pte, newprot); +} + +static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pte_clear(mm, addr, ptep); +} + +#endif /* _ASM_GENERIC_HUGETLB_H */ diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 1a12f5b9a0ab..73b864a32017 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -2247,10 +2247,11 @@ static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, pte_t entry; if (writable) { - entry = - pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot))); + entry = huge_pte_mkwrite(huge_pte_mkdirty(mk_huge_pte(page, + vma->vm_page_prot))); } else { - entry = huge_pte_wrprotect(mk_pte(page, vma->vm_page_prot)); + entry = huge_pte_wrprotect(mk_huge_pte(page, + vma->vm_page_prot)); } entry = pte_mkyoung(entry); entry = pte_mkhuge(entry); @@ -2264,7 +2265,7 @@ static void set_huge_ptep_writable(struct vm_area_struct *vma, { pte_t entry; - entry = pte_mkwrite(pte_mkdirty(huge_ptep_get(ptep))); + entry = huge_pte_mkwrite(huge_pte_mkdirty(huge_ptep_get(ptep))); if (huge_ptep_set_access_flags(vma, address, ptep, entry, 1)) update_mmu_cache(vma, address, ptep); } @@ -2379,7 +2380,7 @@ again: * HWPoisoned hugepage is already unmapped and dropped reference */ if (unlikely(is_hugetlb_entry_hwpoisoned(pte))) { - pte_clear(mm, address, ptep); + huge_pte_clear(mm, address, ptep); continue; } @@ -2403,7 +2404,7 @@ again: pte = huge_ptep_get_and_clear(mm, address, ptep); tlb_remove_tlb_entry(tlb, ptep, address); - if (pte_dirty(pte)) + if (huge_pte_dirty(pte)) set_page_dirty(page); page_remove_rmap(page); @@ -2856,7 +2857,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, * page now as it is used to determine if a reservation has been * consumed. */ - if ((flags & FAULT_FLAG_WRITE) && !pte_write(entry)) { + if ((flags & FAULT_FLAG_WRITE) && !huge_pte_write(entry)) { if (vma_needs_reservation(h, vma, address) < 0) { ret = VM_FAULT_OOM; goto out_mutex; @@ -2886,12 +2887,12 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, if (flags & FAULT_FLAG_WRITE) { - if (!pte_write(entry)) { + if (!huge_pte_write(entry)) { ret = hugetlb_cow(mm, vma, address, ptep, entry, pagecache_page); goto out_page_table_lock; } - entry = pte_mkdirty(entry); + entry = huge_pte_mkdirty(entry); } entry = pte_mkyoung(entry); if (huge_ptep_set_access_flags(vma, address, ptep, entry, @@ -2972,7 +2973,8 @@ long follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, * directly from any kind of swap entries. */ if (absent || is_swap_pte(huge_ptep_get(pte)) || - ((flags & FOLL_WRITE) && !pte_write(huge_ptep_get(pte)))) { + ((flags & FOLL_WRITE) && + !huge_pte_write(huge_ptep_get(pte)))) { int ret; spin_unlock(&mm->page_table_lock); @@ -3042,7 +3044,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma, } if (!huge_pte_none(huge_ptep_get(ptep))) { pte = huge_ptep_get_and_clear(mm, address, ptep); - pte = pte_mkhuge(pte_modify(pte, newprot)); + pte = pte_mkhuge(huge_pte_modify(pte, newprot)); pte = arch_make_huge_pte(pte, vma, NULL, 0); set_huge_pte_at(mm, address, ptep, pte); pages++; -- cgit v1.2.3-73-gaa49b From 6ee8630e02be6dd89926ca0fbc21af68b23dc087 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 29 Apr 2013 15:07:44 -0700 Subject: mm: allow arch code to control the user page table ceiling On architectures where a pgd entry may be shared between user and kernel (e.g. ARM+LPAE), freeing page tables needs a ceiling other than 0. This patch introduces a generic USER_PGTABLES_CEILING that arch code can override. It is the responsibility of the arch code setting the ceiling to ensure the complete freeing of the page tables (usually in pgd_free()). [catalin.marinas@arm.com: commit log; shift_arg_pages(), asm-generic/pgtables.h changes] Signed-off-by: Hugh Dickins Signed-off-by: Catalin Marinas Cc: Russell King Cc: [3.3+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 4 ++-- include/asm-generic/pgtable.h | 10 ++++++++++ mm/mmap.c | 4 ++-- 3 files changed, 14 insertions(+), 4 deletions(-) (limited to 'include/asm-generic') diff --git a/fs/exec.c b/fs/exec.c index a96a4885bbbf..87e731f020fb 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -613,7 +613,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * when the old and new regions overlap clear from new_end. */ free_pgd_range(&tlb, new_end, old_end, new_end, - vma->vm_next ? vma->vm_next->vm_start : 0); + vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); } else { /* * otherwise, clean from old_start; this is done to not touch @@ -622,7 +622,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * for the others its just a little faster. */ free_pgd_range(&tlb, old_start, old_end, new_end, - vma->vm_next ? vma->vm_next->vm_start : 0); + vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); } tlb_finish_mmu(&tlb, new_end, old_end); diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index bfd87685fc1f..a59ff51b0166 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -7,6 +7,16 @@ #include #include +/* + * On almost all architectures and configurations, 0 can be used as the + * upper ceiling to free_pgtables(): on many architectures it has the same + * effect as using TASK_SIZE. However, there is one configuration which + * must impose a more careful limit, to avoid freeing kernel pgtables. + */ +#ifndef USER_PGTABLES_CEILING +#define USER_PGTABLES_CEILING 0UL +#endif + #ifndef __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS extern int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long address, pte_t *ptep, diff --git a/mm/mmap.c b/mm/mmap.c index b2c363f7ae54..288958f05f1c 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -2302,7 +2302,7 @@ static void unmap_region(struct mm_struct *mm, update_hiwater_rss(mm); unmap_vmas(&tlb, vma, start, end); free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS, - next ? next->vm_start : 0); + next ? next->vm_start : USER_PGTABLES_CEILING); tlb_finish_mmu(&tlb, start, end); } @@ -2682,7 +2682,7 @@ void exit_mmap(struct mm_struct *mm) /* Use -1 here to ensure all VMAs in the mm are unmapped */ unmap_vmas(&tlb, vma, 0, -1); - free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, 0); + free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, USER_PGTABLES_CEILING); tlb_finish_mmu(&tlb, 0, -1); /* -- cgit v1.2.3-73-gaa49b