[PATCH 3/3 v5] sched: Remove task and group entity load_avg when they are dead

[Yuyang Du <yuyang.du@intel.com>]

When task exits or group is destroyed, the entity's load_avg should be
removed from its parent cfs_rq's load_avg. Otherwise, it will take time
for the parent cfs_rq to decay the dead entity's load_avg to 0, which
is not desired.

Signed-off-by: Yuyang Du <yuyang.du@intel.com>
---
 kernel/sched/fair.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index fa4af54..04dfd77 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4400,6 +4400,11 @@ static void migrate_task_rq_fair(struct task_struct *p, int next_cpu)
 	/* We have migrated, no longer consider this task hot */
 	p->se.exec_start = 0;
 }
+
+static void task_dead_fair(struct task_struct *p)
+{
+	remove_entity_load_avg(&p->se);
+}
 #endif /* CONFIG_SMP */
 
 static unsigned long
@@ -7367,8 +7372,11 @@ void free_fair_sched_group(struct task_group *tg)
 	for_each_possible_cpu(i) {
 		if (tg->cfs_rq)
 			kfree(tg->cfs_rq[i]);
-		if (tg->se)
+		if (tg->se) {
+			if (tg->se[i])
+				remove_entity_load_avg(tg->se[i]);
 			kfree(tg->se[i]);
+		}
 	}
 
 	kfree(tg->cfs_rq);
@@ -7554,6 +7562,7 @@ const struct sched_class fair_sched_class = {
 	.rq_offline		= rq_offline_fair,
 
 	.task_waking		= task_waking_fair,
+	.task_dead			= task_dead_fair,
 #endif
 
 	.set_curr_task          = set_curr_task_fair,
-- 
1.7.9.5