sched: Weaken SD_POWERSAVINGS_BALANCE

One of the problems of power-saving balancing is that under certain
scenarios it is too slow and allows tons of real work to pile up.

Avoid this by ignoring the powersave stuff when there's real work
to be done.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

2 files changed, 38 insertions, 23 deletions

diff --git a/kernel/sched.c b/kernel/sched.c
index 6c819f338b1..f0ccb8b926c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1538,6 +1538,26 @@ static unsigned long target_load(int cpu, int type)
 	return max(rq->cpu_load[type-1], total);
 }
 
+static struct sched_group *group_of(int cpu)
+{
+	struct sched_domain *sd = rcu_dereference(cpu_rq(cpu)->sd);
+
+	if (!sd)
+		return NULL;
+
+	return sd->groups;
+}
+
+static unsigned long power_of(int cpu)
+{
+	struct sched_group *group = group_of(cpu);
+
+	if (!group)
+		return SCHED_LOAD_SCALE;
+
+	return group->cpu_power;
+}
+
 static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
 
 static unsigned long cpu_avg_load_per_task(int cpu)
@@ -3982,26 +4002,6 @@ ret:
 	return NULL;
 }
 
-static struct sched_group *group_of(int cpu)
-{
-	struct sched_domain *sd = rcu_dereference(cpu_rq(cpu)->sd);
-
-	if (!sd)
-		return NULL;
-
-	return sd->groups;
-}
-
-static unsigned long power_of(int cpu)
-{
-	struct sched_group *group = group_of(cpu);
-
-	if (!group)
-		return SCHED_LOAD_SCALE;
-
-	return group->cpu_power;
-}
-
 /*
  * find_busiest_queue - find the busiest runqueue among the cpus in group.
  */
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 09d19f77eb3..eaa00014b49 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1333,10 +1333,25 @@ static int select_task_rq_fair(struct task_struct *p, int flag, int sync)
 
 	for_each_domain(cpu, tmp) {
 		/*
-		 * If power savings logic is enabled for a domain, stop there.
+		 * If power savings logic is enabled for a domain, see if we
+		 * are not overloaded, if so, don't balance wider.
 		 */
-		if (tmp->flags & SD_POWERSAVINGS_BALANCE)
-			break;
+		if (tmp->flags & SD_POWERSAVINGS_BALANCE) {
+			unsigned long power = 0;
+			unsigned long nr_running = 0;
+			unsigned long capacity;
+			int i;
+
+			for_each_cpu(i, sched_domain_span(tmp)) {
+				power += power_of(i);
+				nr_running += cpu_rq(i)->cfs.nr_running;
+			}
+
+			capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE);
+
+			if (nr_running/2 < capacity)
+				break;
+		}
 
 		switch (flag) {
 		case SD_BALANCE_WAKE: