Re: [PATCH 4/4] sched/fair: Merge select_idle_core/cpu()

[Vincent Guittot]

On Wed, 27 Jan 2021 at 14:21, Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Wed, Jan 27, 2021 at 02:07:50PM +0100, Vincent Guittot wrote:
> > > @@ -6162,18 +6162,16 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
> > >
> > >         for_each_cpu_wrap(cpu, cpus, target) {
> > >                 if (smt) {
> > > -                       i = select_idle_core(p, cpu, cpus, &idle_cpu);
> > > -                       if ((unsigned int)i < nr_cpumask_bits)
> > > -                               return i;
> > > +                       idle_cpu = select_idle_core(p, cpu, cpus, &idle_cpu);
> >
> > but how do you differentiate idle core (return value) and an idle cpu
> > in the core set in &idle_cpu
> >
> > You will return as soon as a cpu is idle and before testing all cores
> >
>
> Bah, I'm sorry, I was context switching between multiple tasks and failed
> to engage brain. I'll apply your hunk and resend. I don't think this

That's also happen to me

> merits retesting as the saving of avoiding the intermeriate is marginal.

I agree. You can add my Reviewed-by

>
> --
> Mel Gorman
> SUSE Labs