Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-03 Thread Alex Shi
On 04/03/2013 01:08 PM, Namhyung Kim wrote: >> > - for_each_cpu(i, sched_group_cpus(group)) >> > - sgs->group_util += max_rq_util(i); >> > + for_each_cpu(i, sched_group_cpus(group)) { >> > + struct rq *rq = cpu_rq(i); >> > + >> > + if (burst && rq->nr_running > 1) >> >

Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-03 Thread Alex Shi
On 04/03/2013 01:08 PM, Namhyung Kim wrote: - for_each_cpu(i, sched_group_cpus(group)) - sgs-group_util += max_rq_util(i); + for_each_cpu(i, sched_group_cpus(group)) { + struct rq *rq = cpu_rq(i); + + if (burst rq-nr_running 1) + /* use

Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-02 Thread Alex Shi
struct rq *rq = cpu_rq(i); >> + >> +if (burst && rq->nr_running > 1) >> +/* use nr_running as instant utilization */ >> +sgs->group_util += rq->nr_running; > > I guess multiplying FULL_UTIL to rq->nr_running here will remove >

Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-02 Thread Namhyung Kim
Hi Alex, On Sat, 30 Mar 2013 22:35:00 +0800, Alex Shi wrote: > Sleeping task has no utiliation, when they were bursty waked up, the > zero utilization make scheduler out of balance, like aim7 benchmark. > > rq->avg_idle is 'to used to accommodate bursty loads in a dirt simple > dirt cheap manner'

Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-02 Thread Namhyung Kim
Hi Alex, On Sat, 30 Mar 2013 22:35:00 +0800, Alex Shi wrote: Sleeping task has no utiliation, when they were bursty waked up, the zero utilization make scheduler out of balance, like aim7 benchmark. rq-avg_idle is 'to used to accommodate bursty loads in a dirt simple dirt cheap manner' --

Re: [patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-04-02 Thread Alex Shi
struct rq *rq = cpu_rq(i); + +if (burst rq-nr_running 1) +/* use nr_running as instant utilization */ +sgs-group_util += rq-nr_running; I guess multiplying FULL_UTIL to rq-nr_running here will remove special-casing the burst in

[patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-03-30 Thread Alex Shi
Sleeping task has no utiliation, when they were bursty waked up, the zero utilization make scheduler out of balance, like aim7 benchmark. rq->avg_idle is 'to used to accommodate bursty loads in a dirt simple dirt cheap manner' -- Mike Galbraith. With this cheap and smart bursty indicator, we can

[patch v6 13/21] sched: using avg_idle to detect bursty wakeup

2013-03-30 Thread Alex Shi
Sleeping task has no utiliation, when they were bursty waked up, the zero utilization make scheduler out of balance, like aim7 benchmark. rq-avg_idle is 'to used to accommodate bursty loads in a dirt simple dirt cheap manner' -- Mike Galbraith. With this cheap and smart bursty indicator, we can