Re: [Patch v9 5/8] sched/fair: update cpu_capacity to reflect thermal pressure

[Amit Kucheria]

On Wed, Jan 29, 2020 at 4:06 AM Thara Gopinath
<thara.gopinath@xxxxxxxxxx> wrote:
>
> cpu_capacity initially reflects the maximum possible capacity of a cpu.
> Thermal pressure on a cpu means this maximum possible capacity is
> unavailable due to thermal events. This patch subtracts the average thermal
> pressure for a cpu from its maximum possible capacity so that cpu_capacity
> reflects the actual maximum currently available capacity.

"actual maximum currently available capacity" is quite a mouthful. :-)

"Remaining capacity" or "Effective capacity" anyone?

IIUC, this remaining capacity is NOT the same as the capped/decreased
capacity referred to in patches 1 and 3. The delta capacity (aka
thermal pressure) there refers to the difference between HW max
capacity and thermally throttled capacity.
Here, we also subtract RT/DL utilisation. Is that accurate?




>
> Signed-off-by: Thara Gopinath <thara.gopinath@xxxxxxxxxx>
> ---
>
> v8->v9:
>         - Use thermal_load_avg to read rq->avg_thermal.load_avg.
>
>  kernel/sched/fair.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 5f58c03..d879077 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -7753,8 +7753,15 @@ static unsigned long scale_rt_capacity(struct sched_domain *sd, int cpu)
>         if (unlikely(irq >= max))
>                 return 1;
>
> +       /*
> +        * avg_rt.util avg and avg_dl.util track binary signals
> +        * (running and not running) with weights 0 and 1024 respectively.
> +        * avg_thermal.load_avg tracks thermal pressure and the weighted
> +        * average uses the actual delta max capacity(load).
> +        */
>         used = READ_ONCE(rq->avg_rt.util_avg);
>         used += READ_ONCE(rq->avg_dl.util_avg);
> +       used += thermal_load_avg(rq);
>
>         if (unlikely(used >= max))
>                 return 1;
> --
> 2.1.4
>