mm: vmscan: respect NUMA policy mask when shrinking slab on direct reclaim
authorVladimir Davydov <vdavydov@parallels.com>
Thu, 3 Apr 2014 21:47:19 +0000 (14:47 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 3 Apr 2014 23:20:58 +0000 (16:20 -0700)
When direct reclaim is executed by a process bound to a set of NUMA
nodes, we should scan only those nodes when possible, but currently we
will scan kmem from all online nodes even if the kmem shrinker is NUMA
aware.  That said, binding a process to a particular NUMA node won't
prevent it from shrinking inode/dentry caches from other nodes, which is
not good.  Fix this.

Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Dave Chinner <dchinner@redhat.com>
Cc: Glauber Costa <glommer@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmscan.c

index a9c74b4..4acee0f 100644 (file)
@@ -2425,8 +2425,8 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
                        unsigned long lru_pages = 0;
 
                        nodes_clear(shrink->nodes_to_scan);
-                       for_each_zone_zonelist(zone, z, zonelist,
-                                       gfp_zone(sc->gfp_mask)) {
+                       for_each_zone_zonelist_nodemask(zone, z, zonelist,
+                                       gfp_zone(sc->gfp_mask), sc->nodemask) {
                                if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
                                        continue;