Skip to contentSkip to conversation

Sched/fair: Improve the accuracy of sched_stat_wait statistics

PATCH In review [email protected] 3 participants 3 messages Oct 9, 2020 – Oct 15, 2020 Scheduler

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Sched/fair: Improve the accuracy of sched_stat_wait statistics

From: jun qian <[email protected]>

When the sched_schedstat changes from 0 to 1, some sched se maybe already in the runqueue, the se->statistics.wait_start will be 0. So it will let the (rq_of(cfs_rq)) - se->statistics.wait_start) wrong. We need to avoid this scenario.

---
 kernel/sched/fair.c | 9 +++++++++
 1 file changed, 9 insertions(+)
kernel/sched/fair.c +9 -0
index 1a68a05..6f8ca0c 100644
@@ -906,6 +906,15 @@ static void update_curr_fair(struct rq *rq)
 	if (!schedstat_enabled())
 		return;
 
+	/*
+	 * When the sched_schedstat changes from 0 to 1, some sched se
+	 * maybe already in the runqueue, the se->statistics.wait_start
+	 * will be 0.So it will let the delta wrong. We need to avoid this
+	 * scenario.
+	 */
+	if (unlikely(!schedstat_val(se->statistics.wait_start)))
+		return;
+
 	delta = rq_clock(rq_of(cfs_rq)) - schedstat_val(se->statistics.wait_start);
 
 	if (entity_is_task(se)) {
Signature · 1 line
1.8.3.1
maintainer
#1 lore raw
Quoted 9 lines · On Fri, Oct 09, 2020 at 05:25:30PM +0800, [email protected] wrote:

From: jun qian <[email protected]>

When the sched_schedstat changes from 0 to 1, some sched se maybe already in the runqueue, the se->statistics.wait_start will be 0. So it will let the (rq_of(cfs_rq)) - se->statistics.wait_start) wrong. We need to avoid this scenario.

This SoB chain isn't valid. Did Yafang's tag need to a reviewed-by or something?

Quoted 27 lines
---
 kernel/sched/fair.c | 9 +++++++++
 1 file changed, 9 insertions(+)
kernel/sched/fair.c +9 -0
index 1a68a05..6f8ca0c 100644
@@ -906,6 +906,15 @@ static void update_curr_fair(struct rq *rq)
 	if (!schedstat_enabled())
 		return;
 
+	/*
+	 * When the sched_schedstat changes from 0 to 1, some sched se
+	 * maybe already in the runqueue, the se->statistics.wait_start
+	 * will be 0.So it will let the delta wrong. We need to avoid this
+	 * scenario.
+	 */
+	if (unlikely(!schedstat_val(se->statistics.wait_start)))
+		return;
+
 	delta = rq_clock(rq_of(cfs_rq)) - schedstat_val(se->statistics.wait_start);
 
 	if (entity_is_task(se)) {
Signature · 1 line
1.8.3.1
Reviewed-by
#2 lore raw
Quoted 9 lines · Peter Zijlstra
Quoted 7 lines · On Fri, Oct 09, 2020 at 05:25:30PM +0800, [email protected] wrote:

From: jun qian <[email protected]>

When the sched_schedstat changes from 0 to 1, some sched se maybe already in the runqueue, the se->statistics.wait_start will be 0. So it will let the (rq_of(cfs_rq)) - se->statistics.wait_start) wrong. We need to avoid this scenario.

Quoted 6 lines
Quoted 2 lines

This SoB chain isn't valid. Did Yafang's tag need to a reviewed-by or something?

This patch improves the behavior when sched_schedstat is changed from 0 to 1, so it looks good to me.

Quoted 27 lines
Quoted 27 lines
---
 kernel/sched/fair.c | 9 +++++++++
 1 file changed, 9 insertions(+)
kernel/sched/fair.c +9 -1
index 1a68a05..6f8ca0c 100644
@@ -906,6 +906,15 @@ static void update_curr_fair(struct rq *rq)
      if (!schedstat_enabled())
              return;
 
+     /*
+      * When the sched_schedstat changes from 0 to 1, some sched se
+      * maybe already in the runqueue, the se->statistics.wait_start
+      * will be 0.So it will let the delta wrong. We need to avoid this
+      * scenario.
+      */
+     if (unlikely(!schedstat_val(se->statistics.wait_start)))
+             return;
+
      delta = rq_clock(rq_of(cfs_rq)) - schedstat_val(se->statistics.wait_start);
 
      if (entity_is_task(se)) {
--

1.8.3.1

Signature · 2 lines
Thanks
Yafang