Re: [PATCH] sched/fair: improve spreading of utilization

[Valentin Schneider <valentin.schneider@arm.com>]

On Thu, Mar 12 2020, Vincent Guittot wrote:
>  kernel/sched/fair.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 3c8a379c357e..97a0307312d9 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9025,6 +9025,14 @@ static struct rq *find_busiest_queue(struct lb_env *env,
>               case migrate_util:
>                       util = cpu_util(cpu_of(rq));
>
> +			/*
> +			 * Don't try to pull utilization from a CPU with one
> +			 * running task. Whatever its utilization, we will fail
> +			 * detach the task.
> +			 */
> +			if (nr_running <= 1)
> +				continue;
> +

Doesn't this break misfit? If the busiest group is group_misfit_task, it
is totally valid for the runqueues to have a single running task -
that's the CPU-bound task we want to upmigrate.

If the busiest rq has only a single running task, we'll skip the
detach_tasks() block and go straight to the active balance bits.
Misfit balancing totally relies on this, and IMO ASYM_PACKING does
too. Looking at voluntary_active_balance(), it seems your change also
goes against the one added by
  1aaf90a4b88a ("sched: Move CFS tasks to CPUs with higher capacity")

The bandaid here would be gate this 'continue' with checks against the
busiest_group_type, but that's only a loose link wrt
voluntary_active_balance().

>                       if (busiest_util < util) {
>                               busiest_util = util;
>                               busiest = rq;