Messages in this thread | | | From | Vincent Guittot <> | Date | Wed, 28 Nov 2018 14:35:00 +0100 | Subject | Re: [PATCH v7 2/2] sched/fair: update scale invariance of PELT |
| |
On Wed, 28 Nov 2018 at 14:33, Vincent Guittot <vincent.guittot@linaro.org> wrote: > > On Wed, 28 Nov 2018 at 12:53, Patrick Bellasi <patrick.bellasi@arm.com> wrote: > > > > On 28-Nov 11:02, Peter Zijlstra wrote: > > > On Wed, Nov 28, 2018 at 10:54:13AM +0100, Vincent Guittot wrote: > > > > > > > Is there anything else that I should do for these patches ? > > > > > > IIRC, Morten mention they break util_est; Patrick was going to explain. > > > > I guess the problem is that, once we cross the current capacity, > > strictly speaking util_avg does not represent anymore a utilization. > > > > With the new signal this could happen and we end up storing estimated > > utilization samples which will overestimate the task requirements. > > > > We will have a spike in estimated utilization at next wakeup, since we > > use MAX(util_avg@dequeue_time, ewma). Potentially we also inflate the EWMA in > > case we collect multiple samples above the current capacity. > > TBH I don't see how it's different from current implementation with a > task that was scheduled on big core and now wakes up on little core. > The util_est is overestimated as well. > > But I'm fine with adding your proposal on to on the patchset s/on to on/on top of/
> > > > > So, a possible fix could be to avoid storing util_est samples if we > > end up with a utilization above the current capacity. > > > > Something like: > > > > ----8<--- > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > > index ac855b2f4774..93e0cf5d8a76 100644 > > --- a/kernel/sched/fair.c > > +++ b/kernel/sched/fair.c > > @@ -3661,6 +3661,10 @@ util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) > > if (!task_sleep) > > return; > > > > + /* Skip samples which do not represent an actual utilization */ > > + if (unlikely(task_util(p) > capacity_of(task_cpu(p)))) > > + return; > > + > > /* > > * If the PELT values haven't changed since enqueue time, > > * skip the util_est update. > > ---8<--- > > > > Could that work ? > > > > Maybe using a new utility function to wrap the new check. > > > > -- > > #include <best/regards.h> > > > > Patrick Bellasi
| |