Re: [PATCH] sched/fair: Remove the cost of a redundant cpumask_next_wrap in select_idle_cpu

[Peter Zijlstra <peterz@infradead.org>]

On Tue, Nov 23, 2021 at 07:22:29PM +0800, Barry Song wrote:
> From: Barry Song <song.bao.hua@hisilicon.com>
> 
> This patch keeps the same scanning amount, but drops a redundant loop
> of cpumask_next_wrap.
> The original code did for_each_cpu_wrap(cpu, cpus, target + 1), then
> checked --nr; this patch does --nr before doing the next loop, thus,
> it can remove a cpumask_next_wrap() which costs a little bit.
> 
> Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
> ---
>  kernel/sched/fair.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index ff69f24..e2fb3e0 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6298,9 +6298,9 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool
>  
>  		span_avg = sd->span_weight * avg_idle;
>  		if (span_avg > 4*avg_cost)
> -			nr = div_u64(span_avg, avg_cost);
> +			nr = div_u64(span_avg, avg_cost) - 1;
>  		else
> -			nr = 4;
> +			nr = 3;
>  
>  		time = cpu_clock(this);
>  	}
> @@ -6312,11 +6312,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool
>  				return i;
>  
>  		} else {
> -			if (!--nr)
> -				return -1;
>  			idle_cpu = __select_idle_cpu(cpu, p);
>  			if ((unsigned int)idle_cpu < nr_cpumask_bits)
>  				break;
> +			if (!--nr)
> +				return -1;
>  		}
>  	}

That's just confusing code. Isn't it much clearer to write the whole
thing like so ?

	nr--;
	for_each_cpu_wrap(cpu, cpus, target+1) {
		...
		if (!nr--)
			return -1;
	}