[PATCH] sched: Fast idling of CPU when system is partially loaded

[Tim Chen]

When a system is lightly loaded (i.e. no more than 1 job per cpu),
attempt to pull job to a cpu before putting it to idle is unnecessary and
can be skipped.  This patch adds an indicator so the scheduler can know
when there's no more than 1 active job is on any CPU in the system to
skip needless job pulls.

On a 4 socket machine with a request/response kind of workload from
clients, we saw about 0.13 msec delay when we go through a full load
balance to try pull job from all the other cpus.  While 0.1 msec was
spent on processing the request and generating a response, the 0.13 msec
load balance overhead was actually more than the actual work being done.
This overhead can be skipped much of the time for lightly loaded systems.

With this patch, we tested with a netperf request/response workload that
has the server busy with half the cpus in a 4 socket system.  We found
the patch eliminated 75% of the load balance attempts before idling a cpu.

The overhead of setting/clearing the indicator is low as
we already gather the necessary info while we call add_nr_running
and update_sd_lb_stats.

Signed-off-by: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx>
---
 kernel/sched/core.c  | 12 ++++++++----
 kernel/sched/fair.c  | 23 +++++++++++++++++++++--
 kernel/sched/sched.h | 10 ++++++++--
 3 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c6b9879..4f57221 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2630,7 +2630,7 @@ static inline struct task_struct *
 pick_next_task(struct rq *rq, struct task_struct *prev)
 {
 	const struct sched_class *class = &fair_sched_class;
-	struct task_struct *p;
+	struct task_struct *p = NULL;
 
 	/*
 	 * Optimization: we know that if all tasks are in
@@ -2638,9 +2638,13 @@ pick_next_task(struct rq *rq, struct task_struct *prev)
 	 */
 	if (likely(prev->sched_class == class &&
 		   rq->nr_running == rq->cfs.h_nr_running)) {
-		p = fair_sched_class.pick_next_task(rq, prev);
-		if (unlikely(p == RETRY_TASK))
-			goto again;
+
+		/* If no cpu has more than 1 task, skip */
+		if (rq->nr_running > 0 || rq->rd->overload) {
+			p = fair_sched_class.pick_next_task(rq, prev);
+			if (unlikely(p == RETRY_TASK))
+				goto again;
+		}
 
 		/* assumes fair_sched_class->next == idle_sched_class */
 		if (unlikely(!p))
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 9855e87..00ab38c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5863,7 +5863,8 @@ static inline int sg_capacity(struct lb_env *env, struct sched_group *group)
  */
 static inline void update_sg_lb_stats(struct lb_env *env,
 			struct sched_group *group, int load_idx,
-			int local_group, struct sg_lb_stats *sgs)
+			int local_group, struct sg_lb_stats *sgs,
+			bool *overload)
 {
 	unsigned long load;
 	int i;
@@ -5881,6 +5882,8 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 
 		sgs->group_load += load;
 		sgs->sum_nr_running += rq->nr_running;
+		if (overload && rq->nr_running > 1)
+			*overload = true;
 #ifdef CONFIG_NUMA_BALANCING
 		sgs->nr_numa_running += rq->nr_numa_running;
 		sgs->nr_preferred_running += rq->nr_preferred_running;
@@ -5991,6 +5994,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 	struct sched_group *sg = env->sd->groups;
 	struct sg_lb_stats tmp_sgs;
 	int load_idx, prefer_sibling = 0;
+	bool overload = false;
 
 	if (child && child->flags & SD_PREFER_SIBLING)
 		prefer_sibling = 1;
@@ -6011,7 +6015,13 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 				update_group_power(env->sd, env->dst_cpu);
 		}
 
-		update_sg_lb_stats(env, sg, load_idx, local_group, sgs);
+		if (env->sd->parent)
+			update_sg_lb_stats(env, sg, load_idx, local_group, sgs,
+						NULL);
+		else
+			/* gather overload info if we are at root domain */
+			update_sg_lb_stats(env, sg, load_idx, local_group, sgs,
+						&overload);
 
 		if (local_group)
 			goto next_group;
@@ -6045,6 +6055,15 @@ next_group:
 
 	if (env->sd->flags & SD_NUMA)
 		env->fbq_type = fbq_classify_group(&sds->busiest_stat);
+
+	if (!env->sd->parent) {
+		/* update overload indicator if we are at root domain */
+		int i = cpumask_first(sched_domain_span(env->sd));
+		struct rq *rq = cpu_rq(i);
+		if (rq->rd->overload != overload)
+			rq->rd->overload = overload;
+	}
+
 }
 
 /**
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index e47679b..a0cd5c1 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -477,6 +477,9 @@ struct root_domain {
 	cpumask_var_t span;
 	cpumask_var_t online;
 
+	/* Indicate more than one runnable task for any CPU */
+	bool overload;
+
 	/*
 	 * The bit corresponding to a CPU gets set here if such CPU has more
 	 * than one runnable -deadline task (as it is below for RT tasks).
@@ -1212,15 +1215,18 @@ static inline void add_nr_running(struct rq *rq, unsigned count)
 
 	rq->nr_running = prev_nr + count;
 
-#ifdef CONFIG_NO_HZ_FULL
 	if (prev_nr < 2 && rq->nr_running >= 2) {
+		if (!rq->rd->overload)
+			rq->rd->overload = true;
+
+#ifdef CONFIG_NO_HZ_FULL
 		if (tick_nohz_full_cpu(rq->cpu)) {
 			/* Order rq->nr_running write against the IPI */
 			smp_wmb();
 			smp_send_reschedule(rq->cpu);
 		}
-       }
 #endif
+       }
 }
 
 static inline void sub_nr_running(struct rq *rq, unsigned count)
-- 
1.7.11.7


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/