Re: [PATCH v8 2/2] sched/fair: Scan cluster before scanning LLC in wake-up path

[Peter Zijlstra]

On Tue, May 30, 2023 at 03:02:53PM +0800, Yicong Yang wrote:

> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 373ff5f55884..b8c129ed8b47 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6994,6 +6994,30 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool
>  		}
>  	}
>  
> +	if (static_branch_unlikely(&sched_cluster_active)) {
> +		struct sched_domain *sdc = rcu_dereference(per_cpu(sd_cluster, target));
> +
> +		if (sdc) {
> +			for_each_cpu_wrap(cpu, sched_domain_span(sdc), target + 1) {
> +				if (!cpumask_test_cpu(cpu, cpus))
> +					continue;
> +
> +				if (has_idle_core) {
> +					i = select_idle_core(p, cpu, cpus, &idle_cpu);
> +					if ((unsigned int)i < nr_cpumask_bits)
> +						return i;
> +				} else {
> +					if (--nr <= 0)
> +						return -1;
> +					idle_cpu = __select_idle_cpu(cpu, p);
> +					if ((unsigned int)idle_cpu < nr_cpumask_bits)
> +						return idle_cpu;
> +				}
> +			}
> +			cpumask_andnot(cpus, cpus, sched_domain_span(sdc));
> +		}
> +	}

Would not this:

--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6994,6 +6994,29 @@ static int select_idle_cpu(struct task_s
 		}
 	}
 
+	if (static_branch_unlikely(&sched_cluster_active)) {
+		struct sched_group *sg = sd->groups;
+		if (sg->flags & SD_CLUSTER) {
+			for_each_cpu_wrap(cpu, sched_group_span(sg), target+1) {
+				if (!cpumask_test_cpu(cpu, cpus))
+					continue;
+
+				if (has_idle_core) {
+					i = select_idle_core(p, cpu, cpus, &idle_cpu);
+					if ((unsigned)i < nr_cpumask_bits)
+						return 1;
+				} else {
+					if (--nr <= 0)
+						return -1;
+					idle_cpu = __select_idle_cpu(cpu, p);
+					if ((unsigned)idle_cpu < nr_cpumask_bits)
+						return idle_cpu;
+				}
+			}
+			cpumask_andnot(cpus, cpus, sched_group_span(sg));
+		}
+	}
+
 	for_each_cpu_wrap(cpu, cpus, target + 1) {
 		if (has_idle_core) {
 			i = select_idle_core(p, cpu, cpus, &idle_cpu);

also work? Then we can avoid the extra sd_cluster per-cpu variable.