Re: [RFC 2/3] sched/fair: use util_est in LB

2017-09-04 Thread Pavan Kondeti
On Mon, Sep 4, 2017 at 7:48 PM, Patrick Bellasi wrote: > On 29-Aug 10:15, Pavan Kondeti wrote: >> On Fri, Aug 25, 2017 at 3:50 PM, Patrick Bellasi >> wrote: >> > When the scheduler looks at the CPU utlization, the current PELT value >> > for a CPU is returned straight away. In certain scenarios t

Re: [RFC 2/3] sched/fair: use util_est in LB

2017-09-04 Thread Patrick Bellasi
On 29-Aug 10:15, Pavan Kondeti wrote: > On Fri, Aug 25, 2017 at 3:50 PM, Patrick Bellasi > wrote: > > When the scheduler looks at the CPU utlization, the current PELT value > > for a CPU is returned straight away. In certain scenarios this can have > > undesired side effects on task placement. > >

Re: [RFC 2/3] sched/fair: use util_est in LB

2017-08-28 Thread Pavan Kondeti
On Fri, Aug 25, 2017 at 3:50 PM, Patrick Bellasi wrote: > When the scheduler looks at the CPU utlization, the current PELT value > for a CPU is returned straight away. In certain scenarios this can have > undesired side effects on task placement. > > +/** > + * cpu_util_est: estimated utilizati

[RFC 2/3] sched/fair: use util_est in LB

2017-08-25 Thread Patrick Bellasi
When the scheduler looks at the CPU utlization, the current PELT value for a CPU is returned straight away. In certain scenarios this can have undesired side effects on task placement. For example, since the task utilization is decayed at wakeup time, when a long sleeping big task is enqueued it d