memcg: reparent charges of children before processing parent

Sometimes the cleanup after memcg hierarchy testing gets stuck in
mem_cgroup_reparent_charges(), unable to bring non-kmem usage down to 0.

There may turn out to be several causes, but a major cause is this: the
workitem to offline parent can get run before workitem to offline child;
parent's mem_cgroup_reparent_charges() circles around waiting for the
child's pages to be reparented to its lrus, but it's holding
cgroup_mutex which prevents the child from reaching its
mem_cgroup_reparent_charges().

Further testing showed that an ordered workqueue for cgroup_destroy_wq
is not always good enough: percpu_ref_kill_and_confirm's call_rcu_sched
stage on the way can mess up the order before reaching the workqueue.

Instead, when offlining a memcg, call mem_cgroup_reparent_charges() on
all its children (and grandchildren, in the correct order) to have their
charges reparented first.

Fixes: e5fca243abae ("cgroup: use a dedicated workqueue for cgroup destruction")
Signed-off-by: Filipe Brandenburger <filbranden@google.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Tejun Heo <tj@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org>	[v3.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9d17310..5b6b003 100644 (file)
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -6595,6 +6595,7 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css)
 {
        struct mem_cgroup *memcg = mem_cgroup_from_css(css);
        struct mem_cgroup_event *event, *tmp;
+       struct cgroup_subsys_state *iter;
 
        /*
         * Unregister events and notify userspace.
@@ -6611,7 +6612,14 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css)
        kmem_cgroup_css_offline(memcg);
 
        mem_cgroup_invalidate_reclaim_iterators(memcg);
-       mem_cgroup_reparent_charges(memcg);
+
+       /*
+        * This requires that offlining is serialized.  Right now that is
+        * guaranteed because css_killed_work_fn() holds the cgroup_mutex.
+        */
+       css_for_each_descendant_post(iter, css)
+               mem_cgroup_reparent_charges(mem_cgroup_from_css(iter));
+
        mem_cgroup_destroy_all_caches(memcg);
        vmpressure_cleanup(&memcg->vmpressure);
 }