sched/fair: Update blocked load from NEWIDLE

Since we already iterate CPUs looking for work on NEWIDLE, use this
iteration to age the blocked load. If the domain for which this is
done completely spand the idle set, we can push the ILB based aging
forward.

Suggested-by: Brendan Jackman <brendan.jackman@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

1 files changed, 43 insertions, 6 deletions

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d8693fa9e7c5..85232dad89c9 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5376,6 +5376,14 @@ decay_load_missed(unsigned long load, unsigned long missed_updates, int idx)
 	}
 	return load;
 }
+
+static struct {
+	cpumask_var_t idle_cpus_mask;
+	atomic_t nr_cpus;
+	unsigned long next_balance;     /* in jiffy units */
+	unsigned long next_stats;
+} nohz ____cacheline_aligned;
+
 #endif /* CONFIG_NO_HZ_COMMON */
 
 /**
@@ -7022,6 +7030,7 @@ enum fbq_type { regular, remote, all };
 #define LBF_NEED_BREAK	0x02
 #define LBF_DST_PINNED  0x04
 #define LBF_SOME_PINNED	0x08
+#define LBF_NOHZ_STATS	0x10
 
 struct lb_env {
 	struct sched_domain	*sd;
@@ -7460,6 +7469,10 @@ static void update_blocked_averages(int cpu)
 		if (cfs_rq_is_decayed(cfs_rq))
 			list_del_leaf_cfs_rq(cfs_rq);
 	}
+
+#ifdef CONFIG_NO_HZ_COMMON
+	rq->last_blocked_load_update_tick = jiffies;
+#endif
 	rq_unlock_irqrestore(rq, &rf);
 }
 
@@ -7519,6 +7532,9 @@ static inline void update_blocked_averages(int cpu)
 	rq_lock_irqsave(rq, &rf);
 	update_rq_clock(rq);
 	update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq);
+#ifdef CONFIG_NO_HZ_COMMON
+	rq->last_blocked_load_update_tick = jiffies;
+#endif
 	rq_unlock_irqrestore(rq, &rf);
 }
 
@@ -7853,6 +7869,21 @@ group_type group_classify(struct sched_group *group,
 	return group_other;
 }
 
+static void update_nohz_stats(struct rq *rq)
+{
+#ifdef CONFIG_NO_HZ_COMMON
+	unsigned int cpu = rq->cpu;
+
+	if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask))
+		return;
+
+	if (!time_after(jiffies, rq->last_blocked_load_update_tick))
+		return;
+
+	update_blocked_averages(cpu);
+#endif
+}
+
 /**
  * update_sg_lb_stats - Update sched_group's statistics for load balancing.
  * @env: The load balancing environment.
@@ -7875,6 +7906,9 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 	for_each_cpu_and(i, sched_group_span(group), env->cpus) {
 		struct rq *rq = cpu_rq(i);
 
+		if (env->flags & LBF_NOHZ_STATS)
+			update_nohz_stats(rq);
+
 		/* Bias balancing toward CPUs of our domain: */
 		if (local_group)
 			load = target_load(i, load_idx);
@@ -8030,6 +8064,15 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 	if (child && child->flags & SD_PREFER_SIBLING)
 		prefer_sibling = 1;
 
+#ifdef CONFIG_NO_HZ_COMMON
+	if (env->idle == CPU_NEWLY_IDLE) {
+		env->flags |= LBF_NOHZ_STATS;
+
+		if (cpumask_subset(nohz.idle_cpus_mask, sched_domain_span(env->sd)))
+			nohz.next_stats = jiffies + msecs_to_jiffies(LOAD_AVG_PERIOD);
+	}
+#endif
+
 	load_idx = get_sd_load_idx(env->sd, env->idle);
 
 	do {
@@ -9049,12 +9092,6 @@ static inline int on_null_domain(struct rq *rq)
  *   needed, they will kick the idle load balancer, which then does idle
  *   load balancing for all the idle CPUs.
  */
-static struct {
-	cpumask_var_t idle_cpus_mask;
-	atomic_t nr_cpus;
-	unsigned long next_balance;     /* in jiffy units */
-	unsigned long next_stats;
-} nohz ____cacheline_aligned;
 
 static inline int find_new_ilb(void)
 {