[lkml]   [2016]   [Oct]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[PATCH 3.10 12/16] mm: thp: fix SMP race condition between THP page fault and MADV_DONTNEED
From: Andrea Arcangeli <>

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

[js] 3.12 backport: no pmd_devmap in 3.12 yet.

[ tidy up comment grammar/layout]
Signed-off-by: Andrea Arcangeli <>
Reported-by: Kirill A. Shutemov <>
Signed-off-by: Andrew Morton <>
Signed-off-by: Linus Torvalds <>
Cc: Vlastimil Babka <>
Signed-off-by: Jiri Slaby <>
Signed-off-by: Greg Kroah-Hartman <>
Signed-off-by: Willy Tarreau <>
mm/memory.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index 30bf9cc..10cdade 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3834,8 +3834,18 @@ retry:
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
 \ /
  Last update: 2016-10-20 00:52    [W:0.072 / U:3.524 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site