fs/mbcache.c: make count_objects() more robust

When running ltp stress test for 7*24 hours, vmscan occasionally emits
the following warning continuously:

  mb_cache_scan+0x0/0x3f0 negative objects to delete
  nr=-9232265467809300450
  ...

Tracing shows the freeable(mb_cache_count returns) is -1, which causes
the continuous accumulation and overflow of total_scan.

This patch makes sure that mb_cache_count() cannot return a negative
value, which makes the mbcache shrinker more robust.

Link: http://lkml.kernel.org/r/1511753419-52328-1-git-send-email-jiang.biao2@zte.com.cn
Signed-off-by: Jiang Biao <jiang.biao2@zte.com.cn>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: <zhong.weidong@zte.com.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/fs/mbcache.c b/fs/mbcache.c
index d818fd23678700bf8435af2fa64ed62dfdba4d2c..b8b8b9ced9f81c47a1f76da3ef61c2f9d4645c78 100644 (file)
--- a/fs/mbcache.c
+++ b/fs/mbcache.c
@@ -269,6 +269,9 @@ static unsigned long mb_cache_count(struct shrinker *shrink,
        struct mb_cache *cache = container_of(shrink, struct mb_cache,
                                              c_shrink);
 
+       /* Unlikely, but not impossible */
+       if (unlikely(cache->c_entry_count < 0))
+               return 0;
        return cache->c_entry_count;
 }