cgroup: Add memory barriers to plug cgroup_rstat_updated() race window

cgroup_rstat_updated() has a small race window where an updated
signaling can race with flush and could be lost till the next update.
This wasn't a problem for the existing usages, but we plan to use
rstat to track counters which need to be accurate.

This patch plugs the race window by synchronizing
cgroup_rstat_updated() and flush path with memory barriers around
cgroup_rstat_cpu->updated_next pointer.

Signed-off-by: Tejun Heo <tj@kernel.org>
This commit is contained in:
Tejun Heo 2018-04-26 14:29:05 -07:00
parent 8f53470bab
commit 9a9e97b2f1

View file

@ -28,9 +28,12 @@ void cgroup_rstat_updated(struct cgroup *cgrp, int cpu)
unsigned long flags; unsigned long flags;
/* /*
* Speculative already-on-list test. This may race leading to * Paired with the one in cgroup_rstat_cpu_pop_upated(). Either we
* temporary inaccuracies, which is fine. * see NULL updated_next or they see our updated stat.
* */
smp_mb();
/*
* Because @parent's updated_children is terminated with @parent * Because @parent's updated_children is terminated with @parent
* instead of NULL, we can tell whether @cgrp is on the list by * instead of NULL, we can tell whether @cgrp is on the list by
* testing the next pointer for NULL. * testing the next pointer for NULL.
@ -126,6 +129,13 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
*nextp = rstatc->updated_next; *nextp = rstatc->updated_next;
rstatc->updated_next = NULL; rstatc->updated_next = NULL;
/*
* Paired with the one in cgroup_rstat_cpu_updated().
* Either they see NULL updated_next or we see their
* updated stat.
*/
smp_mb();
} }
return pos; return pos;