Re: [PATCH v2] sched: smart wake-affine

[Peter Zijlstra <peterz@infradead.org>]

On Wed, Jul 03, 2013 at 02:10:32PM +0800, Michael Wang wrote:
> +static int wake_wide(struct task_struct *p)
> +{
> +	int factor = nr_cpus_node(cpu_to_node(smp_processor_id()));

That's still a cpumask_weight() in there... we should avoid that. How
about something like the below:

> +
> +	/*
> +	 * Yeah, it's the switching-frequency, could means many wakee or
> +	 * rapidly switch, use factor here will just help to automatically
> +	 * adjust the loose-degree, so bigger node will lead to more pull.
> +	 */
> +	if (p->nr_wakee_switch > factor) {
> +		/*
> +		 * wakee is somewhat hot, it needs certain amount of cpu
> +		 * resource, so if waker is far more hot, prefer to leave
> +		 * it alone.
> +		 */
> +		if (current->nr_wakee_switch > (factor * p->nr_wakee_switch))
> +			return 1;
> +	}
> +
> +	return 0;
> +}

---
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 9b1f2e5..166ab9b 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5081,18 +5083,23 @@ static void destroy_sched_domains(struct sched_domain *sd, int cpu)
  * two cpus are in the same cache domain, see cpus_share_cache().
  */
 DEFINE_PER_CPU(struct sched_domain *, sd_llc);
+DEFINE_PER_CPU(int, sd_llc_size);
 DEFINE_PER_CPU(int, sd_llc_id);
 
 static void update_top_cache_domain(int cpu)
 {
 	struct sched_domain *sd;
 	int id = cpu;
+	int size = 1;
 
 	sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES);
-	if (sd)
+	if (sd) {
 		id = cpumask_first(sched_domain_span(sd));
+		size = cpumask_weight(sched_domain_span(sd));
+	}
 
 	rcu_assign_pointer(per_cpu(sd_llc, cpu), sd);
+	per_cpu(sd_llc_size, cpu) = size;
 	per_cpu(sd_llc_id, cpu) = id;
 }
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index ef0a7b2..c992f58 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -595,6 +595,7 @@ static inline struct sched_domain *highest_flag_domain(int cpu, int flag)
 }
 
 DECLARE_PER_CPU(struct sched_domain *, sd_llc);
+DECLARE_PER_CPU(int, sd_llc_size);
 DECLARE_PER_CPU(int, sd_llc_id);
 
 struct sched_group_power {
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3175,7 +3175,7 @@ static inline unsigned long effective_lo
 
 static int wake_wide(struct task_struct *p)
 {
-	int factor = nr_cpus_node(cpu_to_node(smp_processor_id()));
+	int factor = this_cpu_read(sd_llc_size);
 
 	/*
 	 * Yeah, it's the switching-frequency, could means many wakee or