Re: [PATCH] sched/fair: Update rq clock before updating nohz cpu load

[Wanpeng Li]

2016-05-04 3:46 GMT+08:00 Matt Fleming <matt@xxxxxxxxxxxxxxxxxxx>:
> If we're accessing rq_clock() (e.g. in sched_avg_update()) we should
> update the rq clock before calling cpu_load_update(), otherwise any
> time calculations will be stale.
>
> All other paths currently call update_rq_clock().
>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Mike Galbraith <umgwanakikbuti@xxxxxxxxx>
> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: Frederic Weisbecker <fweisbec@xxxxxxxxx>
> Cc: Rik van Riel <riel@xxxxxxxxxx>
> Signed-off-by: Matt Fleming <matt@xxxxxxxxxxxxxxxxxxx>

Reviewed-by: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>

> ---
>  kernel/sched/fair.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index b8a33abce650..aa9ba82f0d7c 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -4723,6 +4723,7 @@ void cpu_load_update_nohz_stop(void)
>
>         load = weighted_cpuload(cpu_of(this_rq));
>         raw_spin_lock(&this_rq->lock);
> +       update_rq_clock(this_rq);
>         cpu_load_update_nohz(this_rq, curr_jiffies, load);
>         raw_spin_unlock(&this_rq->lock);
>  }
> --
> 2.7.3
>



-- 
Regards,
Wanpeng Li