[PATCH] sched/fair: Fix util_avg of new tasks for asymmetric systems

[PATCH] sched/fair: Fix util_avg of new tasks for asymmetric systems

[Quentin Perret]

When a new task wakes-up for the first time, its initial utilization
is set to half of the spare capacity of its CPU. The current
implementation of post_init_entity_util_avg() uses SCHED_CAPACITY_SCALE
directly as a capacity reference. As a result, on a big.LITTLE system, a
new task waking up on an idle little CPU will be given ~512 of util_avg,
even if the CPU's capacity is significantly less than that.

Fix this by computing the spare capacity with arch_scale_cpu_capacity().

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Quentin Perret <quentin.perret@arm.com>
---
 kernel/sched/fair.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e497c05aab7f..f19432c17017 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -735,11 +735,12 @@ static void attach_entity_cfs_rq(struct sched_entity *se);
  * To solve this problem, we also cap the util_avg of successive tasks to
  * only 1/2 of the left utilization budget:
  *
- *   util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
+ *   util_avg_cap = (cpu_scale - cfs_rq->avg.util_avg) / 2^n
  *
- * where n denotes the nth task.
+ * where n denotes the nth task and cpu_scale the CPU capacity.
  *
- * For example, a simplest series from the beginning would be like:
+ * For example, for a CPU with 1024 of capacity, a simplest series from
+ * the beginning would be like:
  *
  *  task  util_avg: 512, 256, 128,  64,  32,   16,    8, ...
  * cfs_rq util_avg: 512, 768, 896, 960, 992, 1008, 1016, ...
@@ -751,7 +752,8 @@ void post_init_entity_util_avg(struct sched_entity *se)
 {
 	struct cfs_rq *cfs_rq = cfs_rq_of(se);
 	struct sched_avg *sa = &se->avg;
-	long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
+	long cpu_scale = arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq)));
+	long cap = (long)(cpu_scale - cfs_rq->avg.util_avg) / 2;
 
 	if (cap > 0) {
 		if (cfs_rq->avg.util_avg != 0) {
-- 
2.17.0

Re: [PATCH] sched/fair: Fix util_avg of new tasks for asymmetric systems

[Vincent Guittot]

On 4 June 2018 at 13:58, Quentin Perret <quentin.perret@arm.com> wrote:
> When a new task wakes-up for the first time, its initial utilization
> is set to half of the spare capacity of its CPU. The current
> implementation of post_init_entity_util_avg() uses SCHED_CAPACITY_SCALE
> directly as a capacity reference. As a result, on a big.LITTLE system, a
> new task waking up on an idle little CPU will be given ~512 of util_avg,
> even if the CPU's capacity is significantly less than that.
>
> Fix this by computing the spare capacity with arch_scale_cpu_capacity().
>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Signed-off-by: Quentin Perret <quentin.perret@arm.com>

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

> ---
>  kernel/sched/fair.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index e497c05aab7f..f19432c17017 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -735,11 +735,12 @@ static void attach_entity_cfs_rq(struct sched_entity *se);
>   * To solve this problem, we also cap the util_avg of successive tasks to
>   * only 1/2 of the left utilization budget:
>   *
> - *   util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
> + *   util_avg_cap = (cpu_scale - cfs_rq->avg.util_avg) / 2^n
>   *
> - * where n denotes the nth task.
> + * where n denotes the nth task and cpu_scale the CPU capacity.
>   *
> - * For example, a simplest series from the beginning would be like:
> + * For example, for a CPU with 1024 of capacity, a simplest series from
> + * the beginning would be like:
>   *
>   *  task  util_avg: 512, 256, 128,  64,  32,   16,    8, ...
>   * cfs_rq util_avg: 512, 768, 896, 960, 992, 1008, 1016, ...
> @@ -751,7 +752,8 @@ void post_init_entity_util_avg(struct sched_entity *se)
>  {
>         struct cfs_rq *cfs_rq = cfs_rq_of(se);
>         struct sched_avg *sa = &se->avg;
> -       long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
> +       long cpu_scale = arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq)));
> +       long cap = (long)(cpu_scale - cfs_rq->avg.util_avg) / 2;
>
>         if (cap > 0) {
>                 if (cfs_rq->avg.util_avg != 0) {
> --
> 2.17.0
>

Re: [PATCH] sched/fair: Fix util_avg of new tasks for asymmetric systems

[Quentin Perret]

On Monday 04 Jun 2018 at 14:23:42 (+0200), Vincent Guittot wrote:
> On 4 June 2018 at 13:58, Quentin Perret <quentin.perret@arm.com> wrote:
> > When a new task wakes-up for the first time, its initial utilization
> > is set to half of the spare capacity of its CPU. The current
> > implementation of post_init_entity_util_avg() uses SCHED_CAPACITY_SCALE
> > directly as a capacity reference. As a result, on a big.LITTLE system, a
> > new task waking up on an idle little CPU will be given ~512 of util_avg,
> > even if the CPU's capacity is significantly less than that.
> >
> > Fix this by computing the spare capacity with arch_scale_cpu_capacity().
> >
> > Cc: Ingo Molnar <mingo@redhat.com>
> > Cc: Peter Zijlstra <peterz@infradead.org>
> > Signed-off-by: Quentin Perret <quentin.perret@arm.com>
> 
> Acked-by: Vincent Guittot <vincent.guittot@linaro.org>

Thanks !

> 
> > ---
> >  kernel/sched/fair.c | 10 ++++++----
> >  1 file changed, 6 insertions(+), 4 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index e497c05aab7f..f19432c17017 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -735,11 +735,12 @@ static void attach_entity_cfs_rq(struct sched_entity *se);
> >   * To solve this problem, we also cap the util_avg of successive tasks to
> >   * only 1/2 of the left utilization budget:
> >   *
> > - *   util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
> > + *   util_avg_cap = (cpu_scale - cfs_rq->avg.util_avg) / 2^n
> >   *
> > - * where n denotes the nth task.
> > + * where n denotes the nth task and cpu_scale the CPU capacity.
> >   *
> > - * For example, a simplest series from the beginning would be like:
> > + * For example, for a CPU with 1024 of capacity, a simplest series from
> > + * the beginning would be like:
> >   *
> >   *  task  util_avg: 512, 256, 128,  64,  32,   16,    8, ...
> >   * cfs_rq util_avg: 512, 768, 896, 960, 992, 1008, 1016, ...
> > @@ -751,7 +752,8 @@ void post_init_entity_util_avg(struct sched_entity *se)
> >  {
> >         struct cfs_rq *cfs_rq = cfs_rq_of(se);
> >         struct sched_avg *sa = &se->avg;
> > -       long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
> > +       long cpu_scale = arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq)));
> > +       long cap = (long)(cpu_scale - cfs_rq->avg.util_avg) / 2;
> >
> >         if (cap > 0) {
> >                 if (cfs_rq->avg.util_avg != 0) {
> > --
> > 2.17.0
> >