Revert "mm: vmscan: fix misused nr_reclaimed in shrink_mem_cgroup_zone()"
Ying Han [Thu, 12 Apr 2012 19:49:16 +0000 (12:49 -0700)]
This reverts commit c38446cc65e1f2b3eb8630c53943b94c4f65f670.

Before the commit, the code makes senses to me but not after the commit.
The "nr_reclaimed" is the number of pages reclaimed by scanning through
the memcg's lru lists.  The "nr_to_reclaim" is the target value for the
whole function.  For example, we like to early break the reclaim if
reclaimed 32 pages under direct reclaim (not DEF_PRIORITY).

After the reverted commit, the target "nr_to_reclaim" is decremented each
time by "nr_reclaimed" but we still use it to compare the "nr_reclaimed".
It just doesn't make sense to me...

Signed-off-by: Ying Han <yinghan@google.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/vmscan.c

index 33c332b..1a51868 100644 (file)
@@ -2107,12 +2107,7 @@ restart:
                 * with multiple processes reclaiming pages, the total
                 * freeing target can get unreasonably large.
                 */
-               if (nr_reclaimed >= nr_to_reclaim)
-                       nr_to_reclaim = 0;
-               else
-                       nr_to_reclaim -= nr_reclaimed;
-
-               if (!nr_to_reclaim && priority < DEF_PRIORITY)
+               if (nr_reclaimed >= nr_to_reclaim && priority < DEF_PRIORITY)
                        break;
        }
        blk_finish_plug(&plug);