History

mm/hugetlb: only drop uffd-wp special pte if required

As with shmem uffd-wp special ptes, only drop the uffd-wp special swap pte if unmapping an entire vma or synchronized such that faults can not race with the unmap operation. This requires passing zap_flags all the way to the lowest level hugetlb unmap routine: __unmap_hugepage_range. In general, unmap calls originated in hugetlbfs code will pass the ZAP_FLAG_DROP_MARKER flag as synchronization is in place to prevent faults. The exception is hole punch which will first unmap without any synchronization. Later when hole punch actually removes the page from the file, it will check to see if there was a subsequent fault and if so take the hugetlb fault mutex while unmapping again. This second unmap will pass in ZAP_FLAG_DROP_MARKER. The justification of "whether to apply ZAP_FLAG_DROP_MARKER flag when unmap a hugetlb range" is (IMHO): we should never reach a state when a page fault could errornously fault in a page-cache page that was wr-protected to be writable, even in an extremely short period. That could happen if e.g. we pass ZAP_FLAG_DROP_MARKER when hugetlbfs_punch_hole() calls hugetlb_vmdelete_list(), because if a page faults after that call and before remove_inode_hugepages() is executed, the page cache can be mapped writable again in the small racy window, that can cause unexpected data overwritten. [peterx@redhat.com: fix sparse warning] Link: https://lkml.kernel.org/r/Ylcdw8I1L5iAoWhb@xz-m1.local [akpm@linux-foundation.org: move zap_flags_t from mm.h to mm_types.h to fix build issues] Link: https://lkml.kernel.org/r/20220405014915.14873-1-peterx@redhat.com Signed-off-by: Peter Xu <peterx@redhat.com> Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com> Cc: Alistair Popple <apopple@nvidia.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Axel Rasmussen <axelrasmussen@google.com> Cc: David Hildenbrand <david@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jerome Glisse <jglisse@redhat.com> Cc: "Kirill A . Shutemov" <kirill@shutemov.name> Cc: Matthew Wilcox <willy@infradead.org> Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Cc: Nadav Amit <nadav.amit@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Peter Xu committed 4 years ago

Commit 05e90bd1 parent 60dfaad6 files · +45 −20Browse files

Showing the changes to include (3 of 6 files). Show all

Unified · Split

3 files
  1. include/linux/hugetlb.h +5 −3
  2. include/linux/mm.h +0 −2
  3. include/linux/mm_types.h +2 −0
include/linux/hugetlb.h+5−3
@@ -143,11 +143,12 @@ long follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, · 142 unmodified lines
143 unsigned long *, unsigned long *, long, unsigned int,143 unsigned long *, unsigned long *, long, unsigned int,
144 int *);144 int *);
145void unmap_hugepage_range(struct vm_area_struct *,145void unmap_hugepage_range(struct vm_area_struct *,
146 unsigned long, unsigned long, struct page *);146 unsigned long, unsigned long, struct page *,
147 zap_flags_t);
147void __unmap_hugepage_range_final(struct mmu_gather *tlb,148void __unmap_hugepage_range_final(struct mmu_gather *tlb,
148 struct vm_area_struct *vma,149 struct vm_area_struct *vma,
149 unsigned long start, unsigned long end,150 unsigned long start, unsigned long end,
150 struct page *ref_page);151 struct page *ref_page, zap_flags_t zap_flags);
151void hugetlb_report_meminfo(struct seq_file *);152void hugetlb_report_meminfo(struct seq_file *);
152int hugetlb_report_node_meminfo(char *buf, int len, int nid);153int hugetlb_report_node_meminfo(char *buf, int len, int nid);
153void hugetlb_show_meminfo(void);154void hugetlb_show_meminfo(void);
@@ -406,7 +407,8 @@ static inline unsigned long hugetlb_change_protection( · 252 unmodified lines
406 407
407static inline void __unmap_hugepage_range_final(struct mmu_gather *tlb,408static inline void __unmap_hugepage_range_final(struct mmu_gather *tlb,
408 struct vm_area_struct *vma, unsigned long start,409 struct vm_area_struct *vma, unsigned long start,
409 unsigned long end, struct page *ref_page)410 unsigned long end, struct page *ref_page,
411 zap_flags_t zap_flags)
410{412{
411 BUG();413 BUG();
412}414}
include/linux/mm.h+0−2
@@ -3428,8 +3428,6 @@ madvise_set_anon_name(struct mm_struct *mm, unsigned long start, · 3427 unmodified lines
3428}3428}
3429#endif3429#endif
3430 3430
3431typedef unsigned int __bitwise zap_flags_t;
3432
3433/*3431/*
3434 * Whether to drop the pte markers, for example, the uffd-wp information for3432 * Whether to drop the pte markers, for example, the uffd-wp information for
3435 * file-backed memory. This should only be specified when we will completely3433 * file-backed memory. This should only be specified when we will completely
include/linux/mm_types.h+2−0
@@ -863,4 +863,6 @@ enum fault_flag { · 862 unmodified lines
863 FAULT_FLAG_ORIG_PTE_VALID = 1 << 11,863 FAULT_FLAG_ORIG_PTE_VALID = 1 << 11,
864};864};
865 865
866typedef unsigned int __bitwise zap_flags_t;
867
866#endif /* _LINUX_MM_TYPES_H */868#endif /* _LINUX_MM_TYPES_H */