summaryrefslogtreecommitdiff
path: root/mm
diff options
context:
space:
mode:
authorAndrea Arcangeli <aarcange@redhat.com>2016-02-26 15:19:28 -0800
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>2016-03-03 15:07:10 -0800
commit915d02457e74344bcd99fe64b159de2f6074b2c6 (patch)
tree4b14f1a5fb21fb4a8fb28b6ed9d94baeff69c7cb /mm
parentc88edc09363243dd84d6e8ac7a36773213b26ac8 (diff)
downloadlwn-915d02457e74344bcd99fe64b159de2f6074b2c6.tar.gz
lwn-915d02457e74344bcd99fe64b159de2f6074b2c6.zip
mm: thp: fix SMP race condition between THP page fault and MADV_DONTNEED
commit ad33bb04b2a6cee6c1f99fabb15cddbf93ff0433 upstream. 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 <aarcange@redhat.com> Reported-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'mm')
-rw-r--r--mm/memory.c14
1 files changed, 12 insertions, 2 deletions
diff --git a/mm/memory.c b/mm/memory.c
index c387430f06c3..b80bf4746b67 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3399,8 +3399,18 @@ 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