Kirill A. Shutemov ced108037c thp: fix MADV_DONTNEED vs. numa balancing race
In case prot_numa, we are under down_read(mmap_sem).  It's critical to
not clear pmd intermittently to avoid race with MADV_DONTNEED which is
also under down_read(mmap_sem):

	CPU0:				CPU1:
				change_huge_pmd(prot_numa=1)
				 pmdp_huge_get_and_clear_notify()
madvise_dontneed()
 zap_pmd_range()
  pmd_trans_huge(*pmd) == 0 (without ptl)
  // skip the pmd
				 set_pmd_at();
				 // pmd is re-established

The race makes MADV_DONTNEED miss the huge pmd and don't clear it
which may break userspace.

Found by code analysis, never saw triggered.

Link: http://lkml.kernel.org/r/20170302151034.27829-3-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-04-13 18:24:20 -07:00
..
2017-02-27 18:43:47 -08:00
2017-03-13 08:58:09 -07:00
2016-12-12 18:55:08 -08:00
2017-03-09 11:48:48 -08:00
2015-09-08 15:35:28 -07:00
2017-03-10 08:59:07 -08:00
2015-11-05 19:34:48 -08:00
2016-11-02 10:24:03 -06:00
2016-03-17 15:09:34 -07:00
2017-02-27 18:43:46 -08:00