sched: Pre-compute cpumask_weight(sched_domain_span(sd))

Dave reported that his large SPARC machines spend lots of time in
hweight64(), try and optimize some of those needless cpumask_weight()
invocations (esp. with the large offstack cpumasks these are very
expensive indeed).
Reported-by: David Miller <davem@davemloft.net>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

include/linux/sched.h

@@ -960,6 +960,7 @@ struct sched_domain {
 	char *name;
 #endif
 
+	unsigned int span_weight;
 	/*
 	 * Span of all CPUs in this domain.
 	 *

kernel/sched.c

@@ -6271,6 +6271,9 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu)
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
 
+	for (tmp = sd; tmp; tmp = tmp->parent)
+		tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
+
 	/* Remove the sched domains which do not contribute to scheduling. */
 	for (tmp = sd; tmp; ) {
 		struct sched_domain *parent = tmp->parent;

kernel/sched_fair.c

@@ -1508,9 +1508,7 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_
 		 * Pick the largest domain to update shares over
 		 */
 		tmp = sd;
-		if (affine_sd && (!tmp ||
-				  cpumask_weight(sched_domain_span(affine_sd)) >
-				  cpumask_weight(sched_domain_span(sd))))
+		if (affine_sd && (!tmp || affine_sd->span_weight > sd->span_weight))
 			tmp = affine_sd;
 
 		if (tmp) {
@@ -1554,10 +1552,10 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_
 
 		/* Now try balancing at a lower domain level of new_cpu */
 		cpu = new_cpu;
-		weight = cpumask_weight(sched_domain_span(sd));
+		weight = sd->span_weight;
 		sd = NULL;
 		for_each_domain(cpu, tmp) {
-			if (weight <= cpumask_weight(sched_domain_span(tmp)))
+			if (weight <= tmp->span_weight)
 				break;
 			if (tmp->flags & sd_flag)
 				sd = tmp;
@@ -2243,7 +2241,7 @@ unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu)
 
 unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu)
 {
-	unsigned long weight = cpumask_weight(sched_domain_span(sd));
+	unsigned long weight = sd->span_weight;
 	unsigned long smt_gain = sd->smt_gain;
 
 	smt_gain /= weight;
@@ -2276,7 +2274,7 @@ unsigned long scale_rt_power(int cpu)
 
 static void update_cpu_power(struct sched_domain *sd, int cpu)
 {
-	unsigned long weight = cpumask_weight(sched_domain_span(sd));
+	unsigned long weight = sd->span_weight;
 	unsigned long power = SCHED_LOAD_SCALE;
 	struct sched_group *sdg = sd->groups;