diff options
| author | Michal Hocko <mhocko@suse.cz> | 2014-01-24 00:53:35 +0100 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2014-01-24 01:36:53 +0100 |
| commit | ecc736fc3c71c411a9d201d8588c9e7e049e5d8c (patch) | |
| tree | bc150baf2dbb0a8374c69e7c2c2d53c1d0d36777 /mm | |
| parent | mm, oom: prefer thread group leaders for display purposes (diff) | |
| download | linux-ecc736fc3c71c411a9d201d8588c9e7e049e5d8c.tar.xz linux-ecc736fc3c71c411a9d201d8588c9e7e049e5d8c.zip | |
memcg: fix endless loop caused by mem_cgroup_iter
Hugh has reported an endless loop when the hardlimit reclaim sees the
same group all the time. This might happen when the reclaim races with
the memcg removal.
shrink_zone
[rmdir root]
mem_cgroup_iter(root, NULL, reclaim)
// prev = NULL
rcu_read_lock()
mem_cgroup_iter_load
last_visited = iter->last_visited // gets root || NULL
css_tryget(last_visited) // failed
last_visited = NULL [1]
memcg = root = __mem_cgroup_iter_next(root, NULL)
mem_cgroup_iter_update
iter->last_visited = root;
reclaim->generation = iter->generation
mem_cgroup_iter(root, root, reclaim)
// prev = root
rcu_read_lock
mem_cgroup_iter_load
last_visited = iter->last_visited // gets root
css_tryget(last_visited) // failed
[1]
The issue seemed to be introduced by commit 5f5781619718 ("memcg: relax
memcg iter caching") which has replaced unconditional css_get/css_put by
css_tryget/css_put for the cached iterator.
This patch fixes the issue by skipping css_tryget on the root of the
tree walk in mem_cgroup_iter_load and symmetrically doesn't release it
in mem_cgroup_iter_update.
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Reported-by: Hugh Dickins <hughd@google.com>
Tested-by: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Greg Thelen <gthelen@google.com>
Cc: <stable@vger.kernel.org> [3.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm')
| -rw-r--r-- | mm/memcontrol.c | 17 |
1 files changed, 14 insertions, 3 deletions
diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c8336e8f8df0..da07784dde87 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1158,7 +1158,15 @@ mem_cgroup_iter_load(struct mem_cgroup_reclaim_iter *iter, if (iter->last_dead_count == *sequence) { smp_rmb(); position = iter->last_visited; - if (position && !css_tryget(&position->css)) + + /* + * We cannot take a reference to root because we might race + * with root removal and returning NULL would end up in + * an endless loop on the iterator user level when root + * would be returned all the time. + */ + if (position && position != root && + !css_tryget(&position->css)) position = NULL; } return position; @@ -1167,9 +1175,11 @@ mem_cgroup_iter_load(struct mem_cgroup_reclaim_iter *iter, static void mem_cgroup_iter_update(struct mem_cgroup_reclaim_iter *iter, struct mem_cgroup *last_visited, struct mem_cgroup *new_position, + struct mem_cgroup *root, int sequence) { - if (last_visited) + /* root reference counting symmetric to mem_cgroup_iter_load */ + if (last_visited && last_visited != root) css_put(&last_visited->css); /* * We store the sequence count from the time @last_visited was @@ -1244,7 +1254,8 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root, memcg = __mem_cgroup_iter_next(root, last_visited); if (reclaim) { - mem_cgroup_iter_update(iter, last_visited, memcg, seq); + mem_cgroup_iter_update(iter, last_visited, memcg, root, + seq); if (!memcg) iter->generation++; |