diff options
| author | Kirill A. Shutemov <kirill.shutemov@linux.intel.com> | 2017-04-14 00:56:20 +0300 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2017-12-16 12:33:50 +0300 |
| commit | 47c53ccca4a0984d45b4157447b8eeca47d9a51f (patch) | |
| tree | a442e154492f1310edbab38d15054af4a8b2db1a /mm/huge_memory.c | |
| parent | d4443b53732e5e98220ec8777d11e3c9eb87f5fd (diff) | |
| download | linux-47c53ccca4a0984d45b4157447b8eeca47d9a51f.tar.xz | |
thp: fix MADV_DONTNEED vs. numa balancing race
commit ced108037c2aa542b3ed8b7afd1576064ad1362a upstream.
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>
[jwang: adjust context for 4.4]
Signed-off-by: Jack Wang <jinpu.wang@profitbricks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'mm/huge_memory.c')
| -rw-r--r-- | mm/huge_memory.c | 34 |
1 files changed, 33 insertions, 1 deletions
diff --git a/mm/huge_memory.c b/mm/huge_memory.c index ea013cb2c3f2..0127b788272f 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1588,7 +1588,39 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, if (prot_numa && pmd_protnone(*pmd)) goto unlock; - entry = pmdp_huge_get_and_clear_notify(mm, addr, pmd); + /* + * 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. + * + * pmdp_invalidate() is required to make sure we don't miss + * dirty/young flags set by hardware. + */ + entry = *pmd; + pmdp_invalidate(vma, addr, pmd); + + /* + * Recover dirty/young flags. It relies on pmdp_invalidate to not + * corrupt them. + */ + if (pmd_dirty(*pmd)) + entry = pmd_mkdirty(entry); + if (pmd_young(*pmd)) + entry = pmd_mkyoung(entry); + entry = pmd_modify(entry, newprot); if (preserve_write) entry = pmd_mkwrite(entry); |