| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <BYAPR02MB44888C0FE86220485A38363694819@BYAPR02MB4488.namprd02.prod.outlook.com>
Date: Tue, 5 Jul 2022 23:49:50 +0000
From: David Chen <david.chen@...anix.com>
To: Zhang Qiao <zhangqiao22@...wei.com>,
Vincent Guittot <vincent.guittot@...aro.org>
CC: "mingo@...hat.com" <mingo@...hat.com>,
"peterz@...radead.org" <peterz@...radead.org>,
"juri.lelli@...hat.com" <juri.lelli@...hat.com>,
"dietmar.eggemann@....com" <dietmar.eggemann@....com>,
"rostedt@...dmis.org" <rostedt@...dmis.org>,
"bsegall@...gle.com" <bsegall@...gle.com>,
"mgorman@...e.de" <mgorman@...e.de>,
"bristot@...hat.com" <bristot@...hat.com>,
"vschneid@...hat.com" <vschneid@...hat.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH] sched/fair: fix case with reduced capacity CPU
> -----Original Message-----
> From: Zhang Qiao <zhangqiao22@...wei.com>
> Sent: Tuesday, July 5, 2022 1:23 AM
> To: Vincent Guittot <vincent.guittot@...aro.org>
> Cc: mingo@...hat.com; peterz@...radead.org; juri.lelli@...hat.com; dietmar.eggemann@....com; rostedt@...dmis.org;
> bsegall@...gle.com; mgorman@...e.de; bristot@...hat.com; vschneid@...hat.com; linux-kernel@...r.kernel.org; David Chen
> <david.chen@...anix.com>
> Subject: Re: [PATCH] sched/fair: fix case with reduced capacity CPU
>
>
>
> 在 2022/7/2 12:52, Vincent Guittot 写道:
> > The capacity of the CPU available for CFS tasks can be reduced because of
> > other activities running on the latter. In such case, it's worth trying to
> > move CFS tasks on a CPU with more available capacity.
> >
> > The rework of the load balance has filterd the case when the CPU is
> > classified to be fully busy but its capacity is reduced.
> >
> > Check if CPU's capacity is reduced while gathering load balance statistics
> > and classify it group_misfit_task instead of group_fully_busy so we can
> > try to move the load on another CPU.
> >
> > Reported-by: David Chen <david.chen@...anix.com>
> > Reported-by: Zhang Qiao <zhangqiao22@...wei.com>
> > Signed-off-by: Vincent Guittot <vincent.guittot@...aro.org>
> > ---
> >
> > David, Zhang,
> >
> > I haven't put your tested-by because I have reworked and cleaned the patch to
> > cover more cases.
> >
> > Could you run some tests with this version ?
>
> I tested with this version, it is ok.
>
> Tested-by: Zhang Qiao <zhangqiao22@...wei.com>
>
> Thanks
This version works fine with me.
Tested-by: David Chen <david.chen@...anix.com>
Thanks
>
> >
> > Thanks
> >
> > kernel/sched/fair.c | 50 ++++++++++++++++++++++++++++++++++++---------
> > 1 file changed, 40 insertions(+), 10 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index a78d2e3b9d49..126b82ef4279 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -8798,6 +8798,19 @@ sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs
> > return sched_asym_prefer(env->dst_cpu, group->asym_prefer_cpu);
> > }
> >
> > +static inline bool
> > +sched_reduced_capacity(struct rq *rq, struct sched_domain *sd)
> > +{
> > + /*
> > + * When there is more than 1 task, the group_overloaded case already
> > + * takes care of cpu with reduced capacity
> > + */
> > + if (rq->cfs.h_nr_running != 1)
> > + return false;
> > +
> > + return check_cpu_capacity(rq, sd);
> > +}
> > +
> > /**
> > * update_sg_lb_stats - Update sched_group's statistics for load balancing.
> > * @env: The load balancing environment.
> > @@ -8820,8 +8833,9 @@ static inline void update_sg_lb_stats(struct lb_env *env,
> >
> > for_each_cpu_and(i, sched_group_span(group), env->cpus) {
> > struct rq *rq = cpu_rq(i);
> > + unsigned long load = cpu_load(rq);
> >
> > - sgs->group_load += cpu_load(rq);
> > + sgs->group_load += load;
> > sgs->group_util += cpu_util_cfs(i);
> > sgs->group_runnable += cpu_runnable(rq);
> > sgs->sum_h_nr_running += rq->cfs.h_nr_running;
> > @@ -8851,11 +8865,17 @@ static inline void update_sg_lb_stats(struct lb_env *env,
> > if (local_group)
> > continue;
> >
> > - /* Check for a misfit task on the cpu */
> > - if (env->sd->flags & SD_ASYM_CPUCAPACITY &&
> > - sgs->group_misfit_task_load < rq->misfit_task_load) {
> > - sgs->group_misfit_task_load = rq->misfit_task_load;
> > - *sg_status |= SG_OVERLOAD;
> > + if (env->sd->flags & SD_ASYM_CPUCAPACITY) {
> > + /* Check for a misfit task on the cpu */
> > + if (sgs->group_misfit_task_load < rq->misfit_task_load) {
> > + sgs->group_misfit_task_load = rq->misfit_task_load;
> > + *sg_status |= SG_OVERLOAD;
> > + }
> > + } else if ((env->idle != CPU_NOT_IDLE) &&
> > + sched_reduced_capacity(rq, env->sd) &&
> > + (sgs->group_misfit_task_load < load)) {
> > + /* Check for a task running on a CPU with reduced capacity */
> > + sgs->group_misfit_task_load = load;
> > }
> > }
> >
> > @@ -8908,7 +8928,8 @@ static bool update_sd_pick_busiest(struct lb_env *env,
> > * CPUs in the group should either be possible to resolve
> > * internally or be covered by avg_load imbalance (eventually).
> > */
> > - if (sgs->group_type == group_misfit_task &&
> > + if ((env->sd->flags & SD_ASYM_CPUCAPACITY) &&
> > + (sgs->group_type == group_misfit_task) &&
> > (!capacity_greater(capacity_of(env->dst_cpu), sg->sgc->max_capacity) ||
> > sds->local_stat.group_type != group_has_spare))
> > return false;
> > @@ -9517,9 +9538,18 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
> > busiest = &sds->busiest_stat;
> >
> > if (busiest->group_type == group_misfit_task) {
> > - /* Set imbalance to allow misfit tasks to be balanced. */
> > - env->migration_type = migrate_misfit;
> > - env->imbalance = 1;
> > + if (env->sd->flags & SD_ASYM_CPUCAPACITY) {
> > + /* Set imbalance to allow misfit tasks to be balanced. */
> > + env->migration_type = migrate_misfit;
> > + env->imbalance = 1;
> > + } else {
> > + /*
> > + * Set load imbalance to allow moving task from cpu
> > + * with reduced capacity
> > + */
> > + env->migration_type = migrate_load;
> > + env->imbalance = busiest->group_misfit_task_load;
> > + }
> > return;
> > }
> >
> >
Powered by blists - more mailing lists