mm/thp: do not wait for lock_page() in deferred_split_scan()
authorKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Thu, 22 Mar 2018 23:17:31 +0000 (16:17 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 23 Mar 2018 00:07:01 +0000 (17:07 -0700)
deferred_split_scan() gets called from reclaim path.  Waiting for page
lock may lead to deadlock there.

Replace lock_page() with trylock_page() and skip the page if we failed
to lock it.  We will get to the page on the next scan.

Link: http://lkml.kernel.org/r/20180315150747.31945-1-kirill.shutemov@linux.intel.com
Fixes: 9a982250f773 ("thp: introduce deferred_split_huge_page()")
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/huge_memory.c

index 87ab9b8f56b53dae6875e19533b5e1e171d2533c..529cf36b7edb6d404068c6872ba1693f14b988cd 100644 (file)
@@ -2783,11 +2783,13 @@ static unsigned long deferred_split_scan(struct shrinker *shrink,
 
        list_for_each_safe(pos, next, &list) {
                page = list_entry((void *)pos, struct page, mapping);
-               lock_page(page);
+               if (!trylock_page(page))
+                       goto next;
                /* split_huge_page() removes page from list on success */
                if (!split_huge_page(page))
                        split++;
                unlock_page(page);
+next:
                put_page(page);
        }