From: Andrea Arcangeli Date: Fri, 26 Feb 2016 23:19:28 +0000 (-0800) Subject: mm: thp: fix SMP race condition between THP page fault and MADV_DONTNEED X-Git-Tag: v4.1.12-92~150^2~152 X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=af0f6483b690b063ad6c0ca49de5e96eab6c7627;p=users%2Fjedix%2Flinux-maple.git mm: thp: fix SMP race condition between THP page fault and MADV_DONTNEED Orabug: stable_rc4 [ Upstream commit ad33bb04b2a6cee6c1f99fabb15cddbf93ff0433 ] pmd_trans_unstable()/pmd_none_or_trans_huge_or_clear_bad() were introduced to locklessy (but atomically) detect when a pmd is a regular (stable) pmd or when the pmd is unstable and can infinitely transition from pmd_none() and pmd_trans_huge() from under us, while only holding the mmap_sem for reading (for writing not). While holding the mmap_sem only for reading, MADV_DONTNEED can run from under us and so before we can assume the pmd to be a regular stable pmd we need to compare it against pmd_none() and pmd_trans_huge() in an atomic way, with pmd_trans_unstable(). The old pmd_trans_huge() left a tiny window for a race. Useful applications are unlikely to notice the difference as doing MADV_DONTNEED concurrently with a page fault would lead to undefined behavior. [akpm@linux-foundation.org: tidy up comment grammar/layout] Signed-off-by: Andrea Arcangeli Reported-by: Kirill A. Shutemov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Sasha Levin (cherry picked from commit d347d0e9ae617bd44ca7679786ebf11a06d50372) Signed-off-by: Dan Duval --- diff --git a/mm/memory.c b/mm/memory.c index 2a9e09870c20..701d9ad45c46 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3363,8 +3363,18 @@ static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, if (unlikely(pmd_none(*pmd)) && unlikely(__pte_alloc(mm, vma, pmd, address))) return VM_FAULT_OOM; - /* if an huge pmd materialized from under us just retry later */ - if (unlikely(pmd_trans_huge(*pmd))) + /* + * If a huge pmd materialized under us just retry later. Use + * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd + * didn't become pmd_trans_huge under us and then back to pmd_none, as + * a result of MADV_DONTNEED running immediately after a huge pmd fault + * in a different thread of this mm, in turn leading to a misleading + * pmd_trans_huge() retval. All we have to ensure is that it is a + * regular pmd that we can walk with pte_offset_map() and we can do that + * through an atomic read in C, which is what pmd_trans_unstable() + * provides. + */ + if (unlikely(pmd_trans_unstable(pmd))) return 0; /* * A regular pmd is established and it can't morph into a huge pmd