[PATCH] sched/fair: Optimize dequeue_task_fair()

[PATCH] sched/fair: Optimize dequeue_task_fair()

[Peng Wang]

While looking at enqueue_task_fair and dequeue_task_fair, it occurred
to me that dequeue_task_fair can also be optimized as Vincent described
in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").

dequeue_throttle label can ensure se not to be NULL, and se is
always NULL when reaching root level.

Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
---
 kernel/sched/fair.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index cbcb2f7..e6be952 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5614,10 +5614,10 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 
 	}
 
-dequeue_throttle:
-	if (!se)
-		sub_nr_running(rq, 1);
+	/* At this point se is NULL and we are at root level*/
+	sub_nr_running(rq, 1);
 
+dequeue_throttle:
 	/* balance early to pull high priority tasks */
 	if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
 		rq->next_balance = jiffies;
-- 
2.9.5

Re: [PATCH] sched/fair: Optimize dequeue_task_fair()

[Vincent Guittot]

On Mon, 15 Jun 2020 at 16:20, Peng Wang <rocking@linux.alibaba.com> wrote:
>
> While looking at enqueue_task_fair and dequeue_task_fair, it occurred
> to me that dequeue_task_fair can also be optimized as Vincent described
> in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").

good point

>
> dequeue_throttle label can ensure se not to be NULL, and se is
> always NULL when reaching root level.
>
> Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
> ---
>  kernel/sched/fair.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index cbcb2f7..e6be952 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5614,10 +5614,10 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
>
>         }
>
> -dequeue_throttle:
> -       if (!se)
> -               sub_nr_running(rq, 1);
> +       /* At this point se is NULL and we are at root level*/
> +       sub_nr_running(rq, 1);
>
> +dequeue_throttle:
>         /* balance early to pull high priority tasks */
>         if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
>                 rq->next_balance = jiffies;

You can move the label here because sched_idle_rq() uses
rq->nr-running and rq->cfs.idle_h_nr_running so they will not change
if we jump to the label before reaching root level

> --
> 2.9.5
>

[PATCH v2] sched/fair: Optimize dequeue_task_fair()

[Peng Wang]

While looking at enqueue_task_fair and dequeue_task_fair, it occurred
to me that dequeue_task_fair can also be optimized as Vincent described
in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").

When encountering throttled cfs_rq, dequeue_throttle label can ensure
se not to be NULL, and rq->nr_running remains unchanged, so we can also
skip the early balance check.

Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
---
 kernel/sched/fair.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index cbcb2f7..05242b7 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5614,14 +5614,14 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 
 	}
 
-dequeue_throttle:
-	if (!se)
-		sub_nr_running(rq, 1);
+	/* At this point se is NULL and we are at root level*/
+	sub_nr_running(rq, 1);
 
 	/* balance early to pull high priority tasks */
 	if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
 		rq->next_balance = jiffies;
 
+dequeue_throttle:
 	util_est_dequeue(&rq->cfs, p, task_sleep);
 	hrtick_update(rq);
 }
-- 
2.9.5

Re: [PATCH] sched/fair: Optimize dequeue_task_fair()

[Peng Wang]

On 6/15/20 11:09 PM, Vincent Guittot wrote:
> On Mon, 15 Jun 2020 at 16:20, Peng Wang <rocking@linux.alibaba.com> wrote:
>>
>> While looking at enqueue_task_fair and dequeue_task_fair, it occurred
>> to me that dequeue_task_fair can also be optimized as Vincent described
>> in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").
> 
> good point
> 
>>
>> dequeue_throttle label can ensure se not to be NULL, and se is
>> always NULL when reaching root level.
>>
>> Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
>> ---
>>   kernel/sched/fair.c | 6 +++---
>>   1 file changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index cbcb2f7..e6be952 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -5614,10 +5614,10 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
>>
>>          }
>>
>> -dequeue_throttle:
>> -       if (!se)
>> -               sub_nr_running(rq, 1);
>> +       /* At this point se is NULL and we are at root level*/
>> +       sub_nr_running(rq, 1);
>>
>> +dequeue_throttle:
>>          /* balance early to pull high priority tasks */
>>          if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
>>                  rq->next_balance = jiffies;
> 
> You can move the label here because sched_idle_rq() uses
> rq->nr-running and rq->cfs.idle_h_nr_running so they will not change
> if we jump to the label before reaching root level

Yes, then we can also skip the early balance check.

> 
>> --
>> 2.9.5
>>

Re: [PATCH v2] sched/fair: Optimize dequeue_task_fair()

[Vincent Guittot]

On Tue, 16 Jun 2020 at 08:05, Peng Wang <rocking@linux.alibaba.com> wrote:
>
> While looking at enqueue_task_fair and dequeue_task_fair, it occurred
> to me that dequeue_task_fair can also be optimized as Vincent described
> in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").
>
> When encountering throttled cfs_rq, dequeue_throttle label can ensure
> se not to be NULL, and rq->nr_running remains unchanged, so we can also
> skip the early balance check.
>
> Signed-off-by: Peng Wang <rocking@linux.alibaba.com>

Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>

> ---
>  kernel/sched/fair.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index cbcb2f7..05242b7 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5614,14 +5614,14 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
>
>         }
>
> -dequeue_throttle:
> -       if (!se)
> -               sub_nr_running(rq, 1);
> +       /* At this point se is NULL and we are at root level*/
> +       sub_nr_running(rq, 1);
>
>         /* balance early to pull high priority tasks */
>         if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
>                 rq->next_balance = jiffies;
>
> +dequeue_throttle:
>         util_est_dequeue(&rq->cfs, p, task_sleep);
>         hrtick_update(rq);
>  }
> --
> 2.9.5
>

Re: [PATCH v2] sched/fair: Optimize dequeue_task_fair()

[Peter Zijlstra]

On Tue, Jun 16, 2020 at 02:31:15PM +0200, Vincent Guittot wrote:
> On Tue, 16 Jun 2020 at 08:05, Peng Wang <rocking@linux.alibaba.com> wrote:
> >
> > While looking at enqueue_task_fair and dequeue_task_fair, it occurred
> > to me that dequeue_task_fair can also be optimized as Vincent described
> > in commit 7d148be69e3a ("sched/fair: Optimize enqueue_task_fair()").
> >
> > When encountering throttled cfs_rq, dequeue_throttle label can ensure
> > se not to be NULL, and rq->nr_running remains unchanged, so we can also
> > skip the early balance check.
> >
> > Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
> 
> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>

Thanks!