Skip to contentSkip to conversation

FAILED: patch "[PATCH] thp: fix MADV_DONTNEED vs. numa balancing race" failed to apply to 4.10-stable tree

PATCH [email protected] 1 participant 1 message Apr 18, 2017

Subsystems (3)
Stable branch Memory management THP
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 4.10-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

Quoted 1 line

From ced108037c2aa542b3ed8b7afd1576064ad1362a Mon Sep 17 00:00:00 2001

From: "Kirill A. Shutemov" <[email protected]>
Date: Thu, 13 Apr 2017 14:56:20 -0700
Subject: [PATCH] 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/[email protected] open here
Signed-off-by Kirill A. Shutemov <[email protected]>
Cc Andrea Arcangeli <[email protected]>
Cc Hillf Danton <[email protected]>
Cc <[email protected]>
Signed-off-by Andrew Morton <[email protected]>
Signed-off-by Linus Torvalds <[email protected]>
mm/huge_memory.c +33 -1
index a513861a9037..26769465af63 100644
@@ -1746,7 +1746,39 @@ 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_mk_savedwrite(entry);