Patch "thp: fix MADV_DONTNEED vs. numa balancing race" has been added to the 4.4-stable tree

This is a note to let you know that I've just added the patch titled

thp: fix MADV_DONTNEED vs. numa balancing race

to the 4.4-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
thp-fix-madv_dontneed-vs.-numa-balancing-race.patch
and it can be found in the queue-4.4 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

From: "Kirill A. Shutemov" <[email protected]>
Date: Thu, 13 Apr 2017 14:56:20 -0700
Subject: thp: fix MADV_DONTNEED vs. numa balancing race

From: Kirill A. Shutemov <[email protected]>

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.

---
 mm/huge_memory.c |   34 +++++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

@@ -1588,7 +1588,39 @@ int change_huge_pmd(struct vm_area_struc
 	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);

Patches currently in stable-queue which might be from [email protected] are

queue-4.4/mm-drop-unused-pmdp_huge_get_and_clear_notify.patch queue-4.4/thp-fix-madv_dontneed-vs.-numa-balancing-race.patch queue-4.4/thp-reduce-indentation-level-in-change_huge_pmd.patch