memcg: avoid oom-killing innocent task in case of use_hierarchy
authorDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Wed, 16 Dec 2009 00:47:12 +0000 (16:47 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 16 Dec 2009 15:20:07 +0000 (07:20 -0800)
commitd31f56dbf8bafaacb0c617f9a6f137498d5c7aed
tree88d095c2208d27362e58ff7431407040ead9d848
parent57f9fd7d25ac9a0d7e3a4ced580e780ab4524e3b
memcg: avoid oom-killing innocent task in case of use_hierarchy

task_in_mem_cgroup(), which is called by select_bad_process() to check
whether a task can be a candidate for being oom-killed from memcg's limit,
checks "curr->use_hierarchy"("curr" is the mem_cgroup the task belongs
to).

But this check return true(it's false positive) when:

<some path>/aa use_hierarchy == 0 <- hitting limit
  <some path>/aa/00 use_hierarchy == 1 <- the task belongs to

This leads to killing an innocent task in aa/00.  This patch is a fix for
this bug.  And this patch also fixes the arg for
mem_cgroup_print_oom_info().  We should print information of mem_cgroup
which the task being killed, not current, belongs to.

Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/memcontrol.c
mm/oom_kill.c