Re: [PATCH v2 3/6] sched/fair: cleanup sched_use_asym_prio

[Ricardo Neri]

On Tue, Jan 30, 2024 at 09:17:05PM +0800, alexs@xxxxxxxxxx wrote:
> From: Alex Shi <alexs@xxxxxxxxxx>
> 
> And simplify the one line code. No function change.
> 
> Signed-off-by: Alex Shi <alexs@xxxxxxxxxx>
> To: Ricardo Neri <ricardo.neri-calderon@xxxxxxxxxxxxxxx>
> To: Valentin Schneider <vschneid@xxxxxxxxxx>
> To: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> To: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> To: Ingo Molnar <mingo@xxxxxxxxxx>
> ---
>  kernel/sched/fair.c | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 8d70417f5125..ebd659af2d78 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9741,10 +9741,8 @@ group_type group_classify(unsigned int imbalance_pct,
>   */
>  static bool sched_use_asym_prio(struct sched_domain *sd, int cpu)
>  {
> -	if (!sched_smt_active())
> -		return true;
> -
> -	return sd->flags & SD_SHARE_CPUCAPACITY || is_core_idle(cpu);
> +	return (!sched_smt_active()) ||
> +		(sd->flags & SD_SHARE_CPUCAPACITY) || is_core_idle(cpu);

I think that compressing the two conditions into one hurts readability.
As implemented, it is clear that no further checks are required if there
is no SMT.

Also, please see my comment in patch 6/6.