memcg: skip scanning active lists based on individual size
authorJohannes Weiner <jweiner@redhat.com>
Wed, 2 Nov 2011 20:38:23 +0000 (13:38 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 2 Nov 2011 23:07:00 +0000 (16:07 -0700)
commit9b272977e3b99a8699361d214b51f98c8a9e0e7b
tree2113cee95a42ea893aa6eddb01b14e563153fabb
parent0a619e58703b86d53d07e938eade9a91a4a863c6
memcg: skip scanning active lists based on individual size

Reclaim decides to skip scanning an active list when the corresponding
inactive list is above a certain size in comparison to leave the assumed
working set alone while there are still enough reclaim candidates around.

The memcg implementation of comparing those lists instead reports whether
the whole memcg is low on the requested type of inactive pages,
considering all nodes and zones.

This can lead to an oversized active list not being scanned because of the
state of the other lists in the memcg, as well as an active list being
scanned while its corresponding inactive list has enough pages.

Not only is this wrong, it's also a scalability hazard, because the global
memory state over all nodes and zones has to be gathered for each memcg
and zone scanned.

Make these calculations purely based on the size of the two LRU lists
that are actually affected by the outcome of the decision.

Signed-off-by: Johannes Weiner <jweiner@redhat.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <bsingharora@gmail.com>
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>
Reviewed-by: Ying Han <yinghan@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Documentation/cgroups/memory.txt
include/linux/memcontrol.h
mm/memcontrol.c
mm/vmscan.c