Re: [PATCH v3] sched/fair: prefer prev cpu in asymmetric wakeup path

2020-11-10 Thread Peter Zijlstra
On Thu, Oct 29, 2020 at 05:18:24PM +0100, Vincent Guittot wrote: > During fast wakeup path, scheduler always check whether local or prev cpus > are good candidates for the task before looking for other cpus in the > domain. With > commit b7a331615d25 ("sched/fair: Add asymmetric CPU capacity

Re: [PATCH v3] sched/fair: prefer prev cpu in asymmetric wakeup path

2020-11-03 Thread Dietmar Eggemann
On 29/10/2020 17:18, Vincent Guittot wrote: [...] > - On hikey960 with performance governor (EAS disable) > > ./perf bench sched pipe -T -l 5 > mainline w/ patch > # migrations 999364 0 > ops/sec149313(+/-0.28%) 182587(+/- 0.40) +22% > >

[PATCH v3] sched/fair: prefer prev cpu in asymmetric wakeup path

2020-10-29 Thread Vincent Guittot
During fast wakeup path, scheduler always check whether local or prev cpus are good candidates for the task before looking for other cpus in the domain. With commit b7a331615d25 ("sched/fair: Add asymmetric CPU capacity wakeup scan") the heterogenous system gains a dedicated path but doesn't try