memcg: add the soft_limit reclaim in global direct reclaim.
Ying Han [Thu, 26 May 2011 23:25:27 +0000 (16:25 -0700)]
We recently added the change in global background reclaim which counts the
return value of soft_limit reclaim.  Now this patch adds the similar logic
on global direct reclaim.

We should skip scanning global LRU on shrink_zone if soft_limit reclaim
does enough work.  This is the first step where we start with counting the
nr_scanned and nr_reclaimed from soft_limit reclaim into global
scan_control.

Signed-off-by: Ying Han <yinghan@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/vmscan.c

index 9ce6ec8..884ae08 100644 (file)
@@ -1977,11 +1977,14 @@ restart:
  * If a zone is deemed to be full of pinned pages then just give it a light
  * scan then give up on it.
  */
-static void shrink_zones(int priority, struct zonelist *zonelist,
+static unsigned long shrink_zones(int priority, struct zonelist *zonelist,
                                        struct scan_control *sc)
 {
        struct zoneref *z;
        struct zone *zone;
+       unsigned long nr_soft_reclaimed;
+       unsigned long nr_soft_scanned;
+       unsigned long total_scanned = 0;
 
        for_each_zone_zonelist_nodemask(zone, z, zonelist,
                                        gfp_zone(sc->gfp_mask), sc->nodemask) {
@@ -1998,8 +2001,17 @@ static void shrink_zones(int priority, struct zonelist *zonelist,
                                continue;       /* Let kswapd poll it */
                }
 
+               nr_soft_scanned = 0;
+               nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone,
+                                                       sc->order, sc->gfp_mask,
+                                                       &nr_soft_scanned);
+               sc->nr_reclaimed += nr_soft_reclaimed;
+               total_scanned += nr_soft_scanned;
+
                shrink_zone(priority, zone, sc);
        }
+
+       return total_scanned;
 }
 
 static bool zone_reclaimable(struct zone *zone)
@@ -2064,7 +2076,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
                sc->nr_scanned = 0;
                if (!priority)
                        disable_swap_token();
-               shrink_zones(priority, zonelist, sc);
+               total_scanned += shrink_zones(priority, zonelist, sc);
                /*
                 * Don't shrink slabs when reclaiming memory from
                 * over limit cgroups