Messages in this thread | | | From | John Stultz <> | Date | Thu, 14 Jul 2016 10:51:37 -0700 | Subject | Re: Severe performance regression w/ 4.4+ on Android due to cgroup locking changes |
| |
On Thu, Jul 14, 2016 at 10:41 AM, Oleg Nesterov <oleg@redhat.com> wrote: > On 07/14, John Stultz wrote: >> >> I'm not supposed to be applying this on-top of >> Paul's change, right? > > Right, unless I am totally confused, > >> > Just in case, could you try the patch below? Of course, without other >> > optimizations from Peter, this change makes cgroup_threadgroup_rwsem >> > much worse than a plain rw_semaphore. >> > >> > Oleg. >> > >> > --- x/kernel/cgroup.c >> > +++ x/kernel/cgroup.c >> > @@ -5605,6 +5605,8 @@ int __init cgroup_init(void) >> > BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files)); >> > BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files)); >> > >> > + rcu_sync_enter(&cgroup_threadgroup_rwsem.rss); >> > + >> >> >> So adding this does make a huge difference ontop of Peter's patch. > > Ah, sorry for confusion. I meant, you could try this one-liner without > any other changes.
So the one-liner without other changes helps at a similar level as Paul's change. From some simple testing I've got a 3.5ms spike, but otherwise the values are under 200us.
> But we will need the "slow mode optimization" part from Peter's patch > anyway, otherwise percpu_rw_semaphore simply makes no sense for > cgroup_threadgroup_rwsem.
Yea, with Peter's patch it is further improved.
thanks -john
| |