Re: [PATCH v2] sched/fair: unlink misfit task from cpu overutilized

[Vincent Guittot]

On Mon, 16 Jan 2023 at 10:21, Dietmar Eggemann <dietmar.eggemann@xxxxxxx> wrote:
>
> On 14/01/2023 22:18, Qais Yousef wrote:
> > On 01/13/23 15:28, Vincent Guittot wrote:
> >> Hi Kajetan,
> >>
> >> On Fri, 13 Jan 2023 at 14:50, Kajetan Puchalski
> >> <kajetan.puchalski@xxxxxxx> wrote:
>
> [...]
>
> >>> sched/uclamp: Fix a uninitialized variable warnings
> >>> (baseline_ufc) sched/fair: Check if prev_cpu has highest spare cap in feec()
> >>> sched/uclamp: Cater for uclamp in find_energy_efficient_cpu()'s early exit condition
> >>> sched/uclamp: Make cpu_overutilized() use util_fits_cpu()
> >>> sched/uclamp: Make asym_fits_capacity() use util_fits_cpu()
> >>> sched/uclamp: Make select_idle_capacity() use util_fits_cpu()
> >>> sched/uclamp: Fix fits_capacity() check in feec()
> >>> sched/uclamp: Make task_fits_capacity() use util_fits_cpu()
> >>> sched/uclamp: Fix relationship between uclamp and migration margin
> >>> (previous 'baseline' was here)
> >>>
> >>> I omitted the 3 patches relating directly to capacity_inversion but in
> >
> > This could lead to confusion. Was there a specific reason for this omission?
> > Did you hit some problem?
>
> I thought Vincent's 'split MF from CPU OU' should replace 'CapInv'?

Current patch v3 applies on top of tip/sched/core which includes
'CapInv'. The end goal being that CapInv becomes useless and can be
removed

>
> https://lkml.kernel.org/r/20221209164739.GA24368@vingu-book
>
> ... I have reverted patches:
> Revert "sched/fair: Detect capacity inversion"
> Revert "sched/fair: Consider capacity inversion in util_fits_cpu()"
> Revert "sched/uclamp: Make cpu_overutilized() use util_fits_cpu()" ...
>
> You also mentioned this in
> https://lkml.kernel.org/r/<20230115001906.v7uq4ddodrbvye7d@airbuntu
>
> [...]