From: Vlastimil Babka <vbabka@xxxxxxx>
A BUG_ON(!PageLocked) was triggered in mlock_vma_page() by Sasha Levin fuzzing
with trinity. The call site try_to_unmap_cluster() does not lock the pages
other than its check_page parameter (which is already locked).
The BUG_ON in mlock_vma_page() is not documented and its purpose is somewhat
unclear, but apparently it serializes against page migration, which could
otherwise fail to transfer the PG_mlocked flag. This would not be fatal, as the
page would be eventually encountered again, but NR_MLOCK accounting would
become distorted nevertheless. This patch adds a comment to the BUG_ON in
mlock_vma_page() and munlock_vma_page() to that effect.
The call site try_to_unmap_cluster() is fixed so that for page != check_page,
trylock_page() is attempted (to avoid possible deadlocks as we already have
check_page locked) and mlock_vma_page() is performed only upon success. If the
page lock cannot be obtained, the page is left without PG_mlocked, which is
again not a problem in the whole unevictable memory design.
Reported-by: Sasha Levin <sasha.levin@xxxxxxxxxx>
Cc: Wanpeng Li <liwanp@xxxxxxxxxxxxxxxxxx>
Cc: Michel Lespinasse <walken@xxxxxxxxxx>
Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Vlastimil Babka <vbabka@xxxxxxx>
Signed-off-by: Bob Liu <bob.liu@xxxxxxxxxx>
---
mm/mlock.c | 2 ++
mm/rmap.c | 14 ++++++++++++--
2 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/mm/mlock.c b/mm/mlock.c
index 4e1a6816..b1eb536 100644
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -79,6 +79,7 @@ void clear_page_mlock(struct page *page)
*/
void mlock_vma_page(struct page *page)
{
+ /* Serialize with page migration */
BUG_ON(!PageLocked(page));
if (!TestSetPageMlocked(page)) {
@@ -174,6 +175,7 @@ unsigned int munlock_vma_page(struct page *page)
unsigned int nr_pages;
struct zone *zone = page_zone(page);
+ /* For try_to_munlock() and to serialize with page migration */
BUG_ON(!PageLocked(page));
/*
diff --git a/mm/rmap.c b/mm/rmap.c
index d9d4231..43d429b 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1322,9 +1322,19 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount,
BUG_ON(!page || PageAnon(page));
if (locked_vma) {
- mlock_vma_page(page); /* no-op if already mlocked */
- if (page == check_page)
+ if (page == check_page) {
+ /* we know we have check_page locked */
+ mlock_vma_page(page);
ret = SWAP_MLOCK;
+ } else if (trylock_page(page)) {
+ /*
+ * If we can lock the page, perform mlock.
+ * Otherwise leave the page alone, it will be
+ * eventually encountered again later.
+ */
+ mlock_vma_page(page);
+ unlock_page(page);
+ }
continue; /* don't unmap */
}
--
1.7.10.4
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html