* [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics
@ 2020-10-09 9:25 qianjun.kernel
2020-10-14 13:19 ` Peter Zijlstra
0 siblings, 1 reply; 5+ messages in thread
From: qianjun.kernel @ 2020-10-09 9:25 UTC (permalink / raw)
To: mingo, peterz, juri.lelli, vincent.guittot
Cc: linux-kernel, jun qian, Yafang Shao
From: jun qian <qianjun.kernel@gmail.com>
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.
Signed-off-by: jun qian <qianjun.kernel@gmail.com>
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
---
kernel/sched/fair.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 1a68a05..6f8ca0c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -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
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics
2020-10-09 9:25 [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics qianjun.kernel
@ 2020-10-14 13:19 ` Peter Zijlstra
2020-10-15 6:06 ` Yafang Shao
0 siblings, 1 reply; 5+ messages in thread
From: Peter Zijlstra @ 2020-10-14 13:19 UTC (permalink / raw)
To: qianjun.kernel
Cc: mingo, juri.lelli, vincent.guittot, linux-kernel, Yafang Shao
On Fri, Oct 09, 2020 at 05:25:30PM +0800, qianjun.kernel@gmail.com wrote:
> From: jun qian <qianjun.kernel@gmail.com>
>
> 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.
>
> Signed-off-by: jun qian <qianjun.kernel@gmail.com>
> Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
This SoB chain isn't valid. Did Yafang's tag need to a reviewed-by or
something?
> ---
> kernel/sched/fair.c | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 1a68a05..6f8ca0c 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -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
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics
2020-10-14 13:19 ` Peter Zijlstra
@ 2020-10-15 6:06 ` Yafang Shao
0 siblings, 0 replies; 5+ messages in thread
From: Yafang Shao @ 2020-10-15 6:06 UTC (permalink / raw)
To: Peter Zijlstra; +Cc: jun qian, Ingo Molnar, juri.lelli, Vincent Guittot, LKML
On Wed, Oct 14, 2020 at 9:19 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Fri, Oct 09, 2020 at 05:25:30PM +0800, qianjun.kernel@gmail.com wrote:
> > From: jun qian <qianjun.kernel@gmail.com>
> >
> > 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.
> >
> > Signed-off-by: jun qian <qianjun.kernel@gmail.com>
> > Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
>
> 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.
Reviewed-by: Yafang Shao <laoar.shao@gmail.com>
> > ---
> > kernel/sched/fair.c | 9 +++++++++
> > 1 file changed, 9 insertions(+)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 1a68a05..6f8ca0c 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -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
> >
--
Thanks
Yafang
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics
2020-10-15 6:48 qianjun.kernel
@ 2020-10-15 11:02 ` Peter Zijlstra
0 siblings, 0 replies; 5+ messages in thread
From: Peter Zijlstra @ 2020-10-15 11:02 UTC (permalink / raw)
To: qianjun.kernel
Cc: mingo, juri.lelli, vincent.guittot, linux-kernel, Yafang Shao
On Thu, Oct 15, 2020 at 02:48:46PM +0800, qianjun.kernel@gmail.com wrote:
> From: jun qian <qianjun.kernel@gmail.com>
>
> 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.
>
> Signed-off-by: jun qian <qianjun.kernel@gmail.com>
> Reviewed-by: Yafang Shao <laoar.shao@gmail.com>
Thanks!
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics
@ 2020-10-15 6:48 qianjun.kernel
2020-10-15 11:02 ` Peter Zijlstra
0 siblings, 1 reply; 5+ messages in thread
From: qianjun.kernel @ 2020-10-15 6:48 UTC (permalink / raw)
To: mingo, peterz, juri.lelli, vincent.guittot, linux-kernel
Cc: jun qian, Yafang Shao
From: jun qian <qianjun.kernel@gmail.com>
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.
Signed-off-by: jun qian <qianjun.kernel@gmail.com>
Reviewed-by: Yafang Shao <laoar.shao@gmail.com>
---
kernel/sched/fair.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 1a68a05..6f8ca0c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -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
^ permalink raw reply related [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-10-15 11:02 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-10-09 9:25 [PATCH 1/1] Sched/fair: Improve the accuracy of sched_stat_wait statistics qianjun.kernel
2020-10-14 13:19 ` Peter Zijlstra
2020-10-15 6:06 ` Yafang Shao
2020-10-15 6:48 qianjun.kernel
2020-10-15 11:02 ` Peter Zijlstra
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).