Re: [PATCH v3 08/10] sched/fair: use utilization to select misfit task

[Valentin Schneider]

On 19/09/2019 08:33, Vincent Guittot wrote:
> utilization is used to detect a misfit task but the load is then used to
> select the task on the CPU which can lead to select a small task with
> high weight instead of the task that triggered the misfit migration.
> 
> Signed-off-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> Acked-by: Valentin Schneider <valentin.schneider@xxxxxxx>
> ---
>  kernel/sched/fair.c | 10 ++--------
>  1 file changed, 2 insertions(+), 8 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index a7c8ee6..acca869 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -7429,14 +7429,8 @@ static int detach_tasks(struct lb_env *env)
>  			break;
>  
>  		case migrate_misfit:
> -			load = task_h_load(p);
> -
> -			/*
> -			 * utilization of misfit task might decrease a bit
> -			 * since it has been recorded. Be conservative in the
> -			 * condition.
> -			 */
> -			if (load < env->imbalance)
> +			/* This is not a misfit task */
> +			if (task_fits_capacity(p, capacity_of(env->src_cpu)))
>  				goto next;
>  
>  			env->imbalance = 0;
> 

Following my comment in [1], if you can't squash that in patch 04 then
perhaps you could add that to this change:

---
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 1fac444a4831..d09ce304161d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8343,7 +8343,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
 	if (busiest->group_type == group_misfit_task) {
 		/* Set imbalance to allow misfit task to be balanced. */
 		env->balance_type = migrate_misfit;
-		env->imbalance = busiest->group_misfit_task_load;
+		env->imbalance = 1;
 		return;
 	}
 
---

Reason being we don't care about the load (anymore), we just want a nonzero
imbalance value, so might as well assign something static.

[1]: https://lore.kernel.org/r/74bb33d7-3ba4-aabe-c7a2-3865d5759281@xxxxxxx