Re: [PATCH] mm: memcontrol: switch to rcu protection in drain_all_stock()

[Roman Gushchin <guro@fb.com>]

On Fri, Aug 02, 2019 at 07:14:51PM +0200, Michal Hocko wrote:
> On Fri 02-08-19 17:00:34, Roman Gushchin wrote:
> > On Fri, Aug 02, 2019 at 10:59:47AM +0200, Michal Hocko wrote:
> > > On Fri 02-08-19 10:04:22, Michal Hocko wrote:
> > > > On Thu 01-08-19 16:35:13, Roman Gushchin wrote:
> > > > > Commit 72f0184c8a00 ("mm, memcg: remove hotplug locking from try_charge")
> > > > > introduced css_tryget()/css_put() calls in drain_all_stock(),
> > > > > which are supposed to protect the target memory cgroup from being
> > > > > released during the mem_cgroup_is_descendant() call.
> > > > > 
> > > > > However, it's not completely safe. In theory, memcg can go away
> > > > > between reading stock->cached pointer and calling css_tryget().
> > > > 
> > > > I have to remember how is this whole thing supposed to work, it's been
> > > > some time since I've looked into that.
> > > 
> > > OK, I guess I remember now and I do not see how the race is possible.
> > > Stock cache is keeping its memcg alive because it elevates the reference
> > > counting for each cached charge. And that should keep the whole chain up
> > > to the root (of draining) alive, no? Or do I miss something, could you
> > > generate a sequence of events that would lead to use-after-free?
> > 
> > Right, but it's true when you reading a local percpu stock.
> > But here we read a remote stock->cached pointer, which can be cleared
> > by a remote concurrent drain_local_stock() execution.
> 
> OK, I can see how refill_stock can race with drain_all_stock. I am not
> sure I see drain_local_stock race because that should be triggered only
> from drain_all_stock and only one cpu is allowed to do that. Maybe we
> might have scheduled a work from the previous run?

Exactly. Previously executed drain_all_stock() -> schedule_work ->
drain_local_stock() on a remote cpu races with checking memcg pointer
from drain_all_stock.

> 
> In any case, please document the race in the changelog please. This code
> is indeed tricky and a comment would help as well.

Sure, will send out v2 soon.

Thanks!