Hi
> > >
> > > In fair.c:select_task_rq_fair(), if feec() returns a error (< 0), then
> > > prev_cpu is selected. I think it's better to still let feec() signal
> > > that something happened and let select_task_rq_fair() select prev_cpu by
> > > itself.
> > In fair.c:select_task_rq_fair(), when
Hi
On Tue, Apr 13, 2021 at 1:15 AM Pierre Gondois wrote:
>
> Hi
> > > > >
> > > > > This patch-set is not significantly improving the execution time of
> > > > > feec(). The results we have so far are an improvement of 5-10% in
> > > > > execution time, with feec() being executed in < 10us. So
Hi
>
> Hi,
> > >
> > > This patch-set is not significantly improving the execution time of
> > > feec(). The results we have so far are an improvement of 5-10% in
> > > execution time, with feec() being executed in < 10us. So the gain is not
> > > spectacular.
> >
> > well, I meaned to cache all
Hi
>
> Hi,
> > Hi
> >
> > On Wed, Apr 7, 2021 at 10:11 PM Pierre wrote:
> > >
> > > Hi,
> > > > I test the patch, but the overflow still exists.
> > > > In the "sched/fair: Use pd_cache to speed up
> > find_energy_efficient_cpu()"
> > > > I wonder why recompute the cpu util when cpu==dst_cpu in
>
Hi
On Wed, Apr 7, 2021 at 10:11 PM Pierre wrote:
>
> Hi,
> > I test the patch, but the overflow still exists.
> > In the "sched/fair: Use pd_cache to speed up find_energy_efficient_cpu()"
> > I wonder why recompute the cpu util when cpu==dst_cpu in compute_energy(),
> > when the dst_cpu's util
Hi Dietmar
On Fri, Apr 2, 2021 at 2:07 AM Dietmar Eggemann
wrote:
>
> +cc: Pierre Gondois
>
> On 30/03/2021 11:45, Quentin Perret wrote:
> > Hi,
> >
> > On Tuesday 30 Mar 2021 at 13:21:54 (+0800), Xuewen Yan wrote:
> >> From: Xuewen Yan
> >
From: Xuewen Yan
now the energy delta compute as follow:
base_energy_pd = compute_energy(p, -1, pd);
--->Traverse all CPUs in pd
--->em_pd_energy()
- \
search for the max_sapre_c
From: Xuewen Yan
When the policy->related_cpus are all offline, and then
bring up one cpu, this time, if the ->online is NULL,
the code would update the ->related_cpus with ->cpus,
and now ->cpus is only one online cpu, as a result, the
->related_cpus is different from the orig
The following commit has been merged into the sched/core branch of tip:
Commit-ID: 8c1f560c1ea3f19e22ba356f62680d9d449c9ec2
Gitweb:
https://git.kernel.org/tip/8c1f560c1ea3f19e22ba356f62680d9d449c9ec2
Author:Xuewen Yan
AuthorDate:Fri, 18 Dec 2020 17:27:52 +08:00
Committer
From: Xuewen Yan
CPU (root cfs_rq) estimated utilization (util_est) is currently used in
dequeue_task_fair() to drive frequency selection before it is updated.
with:
CPU_util: rq->cfs.avg.util_avg
CPU_util_est: rq->cfs.avg.util_est
CPU_utilization : max(CPU_util, CPU_ut
From: Xuewen Yan
CPU (root cfs_rq) estimated utilization (util_est) is currently used in
dequeue_task_fair() to drive frequency selection before it is updated.
with:
CPU_util: rq->cfs.avg.util_avg
CPU_util_est: rq->cfs.avg.util_est
CPU_utilization : max(CPU_util, CPU_ut
From: Xuewen Yan
when a task dequeued, it would update it's util and cfs_rq's util,
the following calling relationship exists here:
update_load_avg() -> cfs_rq_util_change() -> cpufreq_update_util()->
sugov_update_[shared\|single] -> sugov_get_util() -> cpu_util_cfs();
the
load is updated, the cpu may always maintain a high frequency.
if keep the util_est_dequeue as it is, are there other concerns,
or this patch would affect other places of system?
> >
> > If I would run with your patch cpu_util_cfs() would chose between 597 and 0
> >
a result, cfs_rq_util_change may change freq unreasonablely.
separate the util_est_dequeue() into util_est_dequeue() and
util_est_update(), and dequeue the _task_util_est(p) before update util.
Signed-off-by: Xuewen Yan
---
kernel/sched/fair.c | 24 +++-
1 file changed, 19 i
gt;avg.util_avg
so the must be extremely small, the
cfs_rq->avg.load_avg
judgment condition "sa->util_avg < cap" could be established.
It's not fair for those tasks who has smaller nice value.
Signed-off-by: Xuewen Yan
---
changes since V2:
*ke
se->load.weight.
Thanks!
发件人: Vincent Guittot
发送时间: 2020年11月5日 20:30
收件人: 闫学文 (Xuewen Yan)
抄送: Ingo Molnar; Peter Zijlstra; Juri Lelli; Dietmar Eggemann; Steven Rostedt;
Ben Segall; Mel Gorman; Daniel Bristot de Oliveira; linux-kernel;
xuewen.ya...@gmail.co
gt;avg.util_avg
so the must be extremely small, the
cfs_rq->avg.load_avg
judgment condition "sa->util_avg < cap" could be established.
It's not fair for those tasks who has smaller nice value.
Signed-off-by: Xuewen Yan
---
kernel/sched/fair.c | 6 +-
1 file
flag;
In DIE domain : containing SD_ASYM_CPUCAPACITY flag;
the "highest_flag_domain(cpu, SD_ASYM_CPUCAPACITY)" will return NULL.
Signed-off-by: Xuewen Yan
---
kernel/sched/sched.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/sched/sched.h b/kernel/sch
18 matches
Mail list logo