Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 3d80636a authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Fix memory ordering bug in page reclaim 



As noticed by Nick Piggin, we need to make sure that we check the page
count before we check for PageDirty, since the dirty check is only valid
if the count implies that we're the only possible ones holding the page.

We always did do this, but the code needs a read-memory-barrier to make
sure that the orderign is also honored by the CPU.

(The writer side is ordered due to the atomic decrement and test on the
page count, see the discussion on linux-kernel)

Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
parent 688ce17b

mm/vmscan.c

+9 −4
Original line number Original line Diff line number Diff line
@@ -511,10 +511,11 @@ static int shrink_list(struct list_head *page_list, struct scan_control *sc) 
		 * PageDirty _after_ making sure that the page is freeable and

		 * PageDirty _after_ making sure that the page is freeable and

		 * not in use by anybody. 	(pagecache + us == 2)

		 * not in use by anybody. 	(pagecache + us == 2)

		 */

		 */

		if (page_count(page) != 2 || PageDirty(page)) {

		if (unlikely(page_count(page) != 2))

			write_unlock_irq(&mapping->tree_lock);

			goto cannot_free;

			goto keep_locked;

		smp_rmb();

		}

		if (unlikely(PageDirty(page)))

			goto cannot_free;





#ifdef CONFIG_SWAP

#ifdef CONFIG_SWAP

		if (PageSwapCache(page)) {

		if (PageSwapCache(page)) {
@@ -538,6 +539,10 @@ static int shrink_list(struct list_head *page_list, struct scan_control *sc) 
			__pagevec_release_nonlru(&freed_pvec);

			__pagevec_release_nonlru(&freed_pvec);

		continue;

		continue;





cannot_free:

		write_unlock_irq(&mapping->tree_lock);

		goto keep_locked;



activate_locked:

activate_locked:

		SetPageActive(page);

		SetPageActive(page);

		pgactivate++;

		pgactivate++;