On Wed 20-04-16 17:29:22, Tejun Heo wrote: > Hello, Michal. > > On Sun, Apr 17, 2016 at 08:07:48AM -0400, Michal Hocko wrote: [...] > > I liked your proposal when mem_cgroup_move_charge would be called from a > > context which doesn't hold the problematic rwsem much more. Would that > > be too intrusive for the stable backport? > > Yeah, I'm working on the fix but let's plug this one first as it seems > really easy to trigger. I got a couple off-list reports (in and > outside fb) of this triggering.
Sure, I think we can use this for an immediate workaround. Nevertheless we want to have a full fix and do not rely on this being the only problem. I would eventually like to reintroduce the draining later when we have a better fix because even though this is not a correctness problem I think we should try hard to not leave anything behind. That being said: Acked-by: Michal Hocko <mho...@suse.com> -- Michal Hocko SUSE Labs
