On 20/12/2018 07:55, Vincent Guittot wrote:
> When check_asym_packing() is triggered, the imbalance is set to :
> busiest_stat.avg_load * busiest_stat.group_capacity / SCHED_CAPACITY_SCALE
> But busiest_stat.avg_load equals
> sgs->group_load *SCHED_CAPACITY_SCALE / sgs->group_capacity
> These divisions can generate a rounding that will make imbalance slightly
> lower than the weighted load of the cfs_rq.
> But this is enough to skip the rq in find_busiest_queue and prevents asym
> migration to happen.
>
> Directly set imbalance to sgs->group_load to remove the rounding.
^^^^^^^^^^^^^^^
I see where that's coming from, but using 'sgs' here is tad confusing since
'sds->busiest_stat' is what's actually used.
Maybe just something like 'the busiest's sgs->group_load' would be good
enough to make things explicit.
>
> Signed-off-by: Vincent Guittot <vincent.guit...@linaro.org>
> ---
> kernel/sched/fair.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index ca46964..9b31247 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -8476,9 +8476,7 @@ static int check_asym_packing(struct lb_env *env,
> struct sd_lb_stats *sds)
> if (sched_asym_prefer(busiest_cpu, env->dst_cpu))
> return 0;
>
> - env->imbalance = DIV_ROUND_CLOSEST(
> - sds->busiest_stat.avg_load * sds->busiest_stat.group_capacity,
> - SCHED_CAPACITY_SCALE);
> + env->imbalance = sds->busiest_stat.avg_load;
That should be group_load, not avg_load. With that fixed:
Reviewed-by: Valentin Schneider <valentin.schnei...@arm.com>
>
> return 1;
> }
>
