linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Song Bao Hua (Barry Song)" <song.bao.hua@hisilicon.com>
To: Mel Gorman <mgorman@techsingularity.net>,
	LKML <linux-kernel@vger.kernel.org>
Cc: Ingo Molnar <mingo@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Valentin Schneider <valentin.schneider@arm.com>,
	Aubrey Li <aubrey.li@linux.intel.com>,
	yangyicong <yangyicong@huawei.com>
Subject: RE: [PATCH 8/9] sched/fair: select idle cpu from idle cpumask for task wakeup
Date: Mon, 2 Aug 2021 10:41:13 +0000	[thread overview]
Message-ID: <9dde989df06b483790cc24dc7670a919@hisilicon.com> (raw)
In-Reply-To: <20210726102247.21437-9-mgorman@techsingularity.net>



> -----Original Message-----
> From: Mel Gorman [mailto:mgorman@techsingularity.net]
> Sent: Monday, July 26, 2021 10:23 PM
> To: LKML <linux-kernel@vger.kernel.org>
> Cc: Ingo Molnar <mingo@kernel.org>; Peter Zijlstra <peterz@infradead.org>;
> Vincent Guittot <vincent.guittot@linaro.org>; Valentin Schneider
> <valentin.schneider@arm.com>; Aubrey Li <aubrey.li@linux.intel.com>; Mel
> Gorman <mgorman@techsingularity.net>
> Subject: [PATCH 8/9] sched/fair: select idle cpu from idle cpumask for task
> wakeup
> 
> From: Aubrey Li <aubrey.li@linux.intel.com>
> 
> Add idle cpumask to track idle cpus in sched domain. Every time
> a CPU enters idle, the CPU is set in idle cpumask to be a wakeup
> target. And if the CPU is not in idle, the CPU is cleared in idle
> cpumask during scheduler tick to ratelimit idle cpumask update.
> 
> When a task wakes up to select an idle cpu, scanning idle cpumask
> has lower cost than scanning all the cpus in last level cache domain,
> especially when the system is heavily loaded.
> 
> v10-v11:
> - Forward port to 5.13-rc5 based kernel
> 
> v9->v10:
> - Update scan cost only when the idle cpumask is scanned, i.e, the
>   idle cpumask is not empty
> 
> v8->v9:
> - rebase on top of tip/sched/core, no code change
> 
> v7->v8:
> - refine update_idle_cpumask, no functionality change
> - fix a suspicious RCU usage warning with CONFIG_PROVE_RCU=y
> 
> v6->v7:
> - place the whole idle cpumask mechanism under CONFIG_SMP
> 
> v5->v6:
> - decouple idle cpumask update from stop_tick signal, set idle CPU
>   in idle cpumask every time the CPU enters idle
> 
> v4->v5:
> - add update_idle_cpumask for s2idle case
> - keep the same ordering of tick_nohz_idle_stop_tick() and update_
>   idle_cpumask() everywhere
> 
> v3->v4:
> - change setting idle cpumask from every idle entry to tickless idle
>   if cpu driver is available
> - move clearing idle cpumask to scheduler_tick to decouple nohz mode
> 
> v2->v3:
> - change setting idle cpumask to every idle entry, otherwise schbench
>   has a regression of 99th percentile latency
> - change clearing idle cpumask to nohz_balancer_kick(), so updating
>   idle cpumask is ratelimited in the idle exiting path
> - set SCHED_IDLE cpu in idle cpumask to allow it as a wakeup target
> 
> v1->v2:
> - idle cpumask is updated in the nohz routines, by initializing idle
>   cpumask with sched_domain_span(sd), nohz=off case remains the original
>   behavior
> 
> [mgorman@techsingularity.net: RCU protection in update_idle_cpumask]
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Mel Gorman <mgorman@suse.de>
> Cc: Vincent Guittot <vincent.guittot@linaro.org>
> Cc: Qais Yousef <qais.yousef@arm.com>
> Cc: Valentin Schneider <valentin.schneider@arm.com>
> Cc: Jiang Biao <benbjiang@gmail.com>
> Cc: Tim Chen <tim.c.chen@linux.intel.com>
> Signed-off-by: Aubrey Li <aubrey.li@linux.intel.com>
> Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
> ---
>  include/linux/sched/topology.h | 13 ++++++++++
>  kernel/sched/core.c            |  2 ++
>  kernel/sched/fair.c            | 46 +++++++++++++++++++++++++++++++++-
>  kernel/sched/idle.c            |  5 ++++
>  kernel/sched/sched.h           |  4 +++
>  kernel/sched/topology.c        |  3 ++-
>  6 files changed, 71 insertions(+), 2 deletions(-)
> 

Hi Mel, Aubrey,
A similar thing Yicong and me has discussed is having a mask or a count for
idle cores. And then we can only scan idle cores in this mask in
select_idle_cpu().

A obvious problem is that has_idle_cores is a bool, it can seriously lag
from the real status. I mean, after system enters the status without idle
cores, has_idle_cores could be still true.

Right now, we are setting has_idle_cores to true while cpu enters idle
and its smt sibling is also idle. But we are setting has_idle_cores to
false only after we scan all cores in a llc.

So we have thought for a while to provide an idle core mask. But never
really made a workable patch.

Mel's patch7/9 limits the number of cores which will be scanned in
select_idle_cpu(), it might somehow alleviate the problem we redundantly
scan all cores while we actually have no idle core even has_idle_cores
is true.

However, if we can get idle core mask, it could be also good to
select_idle_core()? Maybe after that, we don't have to enforce
proportional scan limits while scanning for an idle core?

> diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h
> index 8f0f778b7c91..905e382ece1a 100644
> --- a/include/linux/sched/topology.h
> +++ b/include/linux/sched/topology.h
> @@ -74,8 +74,21 @@ struct sched_domain_shared {
>  	atomic_t	ref;
>  	atomic_t	nr_busy_cpus;
>  	int		has_idle_cores;
> +	/*
> +	 * Span of all idle CPUs in this domain.
> +	 *
> +	 * NOTE: this field is variable length. (Allocated dynamically
> +	 * by attaching extra space to the end of the structure,
> +	 * depending on how many CPUs the kernel has booted up with)
> +	 */
> +	unsigned long	idle_cpus_span[];
>  };
> 
> +static inline struct cpumask *sds_idle_cpus(struct sched_domain_shared *sds)
> +{
> +	return to_cpumask(sds->idle_cpus_span);
> +}
> +
>  struct sched_domain {
>  	/* These fields must be setup */
>  	struct sched_domain __rcu *parent;	/* top domain must be null terminated
> */
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 7c073b67f1ea..2751614ce0cb 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -4965,6 +4965,7 @@ void scheduler_tick(void)
> 
>  #ifdef CONFIG_SMP
>  	rq->idle_balance = idle_cpu(cpu);
> +	update_idle_cpumask(cpu, rq->idle_balance);
>  	trigger_load_balance(rq);
>  #endif
>  }
> @@ -9031,6 +9032,7 @@ void __init sched_init(void)
>  		rq->wake_stamp = jiffies;
>  		rq->wake_avg_idle = rq->avg_idle;
>  		rq->max_idle_balance_cost = sysctl_sched_migration_cost;
> +		rq->last_idle_state = 1;
> 
>  		INIT_LIST_HEAD(&rq->cfs_tasks);
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index b180205e6b25..fe87af2ccc80 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6230,7 +6230,12 @@ static int select_idle_cpu(struct task_struct *p, struct
> sched_domain *sd, bool
>  	if (!this_sd)
>  		return -1;
> 
> -	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
> +	/*
> +	 * sched_domain_shared is set only at shared cache level,
> +	 * this works only because select_idle_cpu is called with
> +	 * sd_llc.
> +	 */
> +	cpumask_and(cpus, sds_idle_cpus(sd->shared), p->cpus_ptr);
> 
>  	if (sched_feat(SIS_PROP)) {
>  		u64 avg_cost, avg_idle, span_avg;
> @@ -7018,6 +7023,45 @@ balance_fair(struct rq *rq, struct task_struct *prev,
> struct rq_flags *rf)
> 
>  	return newidle_balance(rq, rf) != 0;
>  }
> +
> +/*
> + * Update cpu idle state and record this information
> + * in sd_llc_shared->idle_cpus_span.
> + *
> + * This function is called with interrupts disabled.
> + */
> +void update_idle_cpumask(int cpu, bool idle)
> +{
> +	struct sched_domain *sd;
> +	struct rq *rq = cpu_rq(cpu);
> +	int idle_state;
> +
> +	/*
> +	 * Also set SCHED_IDLE cpu in idle cpumask to
> +	 * allow SCHED_IDLE cpu as a wakeup target.
> +	 */
> +	idle_state = idle || sched_idle_cpu(cpu);
> +	/*
> +	 * No need to update idle cpumask if the state
> +	 * does not change.
> +	 */
> +	if (rq->last_idle_state == idle_state)
> +		return;
> +
> +	rcu_read_lock();
> +	sd = per_cpu(sd_llc, cpu);
> +	if (unlikely(!sd))
> +		goto unlock;
> +
> +	if (idle_state)
> +		cpumask_set_cpu(cpu, sds_idle_cpus(sd->shared));
> +	else
> +		cpumask_clear_cpu(cpu, sds_idle_cpus(sd->shared));
> +
> +	rq->last_idle_state = idle_state;
> +unlock:
> +	rcu_read_unlock();
> +}
>  #endif /* CONFIG_SMP */
> 
>  static unsigned long wakeup_gran(struct sched_entity *se)
> diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
> index 912b47aa99d8..86bfe81cc280 100644
> --- a/kernel/sched/idle.c
> +++ b/kernel/sched/idle.c
> @@ -289,6 +289,11 @@ static void do_idle(void)
>  			cpuhp_report_idle_dead();
>  			arch_cpu_idle_dead();
>  		}
> +		/*
> +		 * The CPU is about to go idle, set it in idle cpumask
> +		 * to be a wake up target.
> +		 */
> +		update_idle_cpumask(cpu, true);
> 
>  		arch_cpu_idle_enter();
>  		rcu_nocb_flush_deferred_wakeup();
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 4d47a0969710..2d6456fa15cb 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -998,6 +998,7 @@ struct rq {
> 
>  	unsigned char		nohz_idle_balance;
>  	unsigned char		idle_balance;
> +	unsigned char		last_idle_state;
> 
>  	unsigned long		misfit_task_load;
> 
> @@ -1846,6 +1847,8 @@ static inline unsigned int group_first_cpu(struct
> sched_group *group)
> 
>  extern int group_balance_cpu(struct sched_group *sg);
> 
> +void update_idle_cpumask(int cpu, bool idle);
> +
>  #ifdef CONFIG_SCHED_DEBUG
>  void update_sched_domain_debugfs(void);
>  void dirty_sched_domain_sysctl(int cpu);
> @@ -1864,6 +1867,7 @@ extern void flush_smp_call_function_from_idle(void);
> 
>  #else /* !CONFIG_SMP: */
>  static inline void flush_smp_call_function_from_idle(void) { }
> +static inline void update_idle_cpumask(int cpu, bool idle) { }
>  #endif
> 
>  #include "stats.h"
> diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
> index b77ad49dc14f..2075bc417b90 100644
> --- a/kernel/sched/topology.c
> +++ b/kernel/sched/topology.c
> @@ -1611,6 +1611,7 @@ sd_init(struct sched_domain_topology_level *tl,
>  		sd->shared = *per_cpu_ptr(sdd->sds, sd_id);
>  		atomic_inc(&sd->shared->ref);
>  		atomic_set(&sd->shared->nr_busy_cpus, sd_weight);
> +		cpumask_copy(sds_idle_cpus(sd->shared), sched_domain_span(sd));
>  	}
> 
>  	sd->private = sdd;
> @@ -1970,7 +1971,7 @@ static int __sdt_alloc(const struct cpumask *cpu_map)
> 
>  			*per_cpu_ptr(sdd->sd, j) = sd;
> 
> -			sds = kzalloc_node(sizeof(struct sched_domain_shared),
> +			sds = kzalloc_node(sizeof(struct sched_domain_shared) +
> cpumask_size(),
>  					GFP_KERNEL, cpu_to_node(j));
>  			if (!sds)
>  				return -ENOMEM;
> --
> 2.26.2


Thanks
Barry


  reply	other threads:[~2021-08-02 10:41 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-26 10:22 [RFC PATCH 0/9] Modify and/or delete SIS_PROP Mel Gorman
2021-07-26 10:22 ` [PATCH 1/9] sched/fair: Track efficiency of select_idle_sibling Mel Gorman
2021-07-26 10:22 ` [PATCH 2/9] sched/fair: Track efficiency of task recent_used_cpu Mel Gorman
2021-07-26 10:22 ` [PATCH 3/9] sched/fair: Track efficiency of select_idle_core Mel Gorman
2021-07-26 10:22 ` [PATCH 4/9] sched/fair: Use prev instead of new target as recent_used_cpu Mel Gorman
2021-07-26 10:22 ` [PATCH 5/9] sched/fair: Avoid a second scan of target in select_idle_cpu Mel Gorman
2021-07-26 10:22 ` [PATCH 6/9] sched/fair: Make select_idle_cpu() proportional to cores Mel Gorman
2021-07-26 10:22 ` [PATCH 7/9] sched/fair: Enforce proportional scan limits when scanning for an idle core Mel Gorman
2021-08-02 10:52   ` Song Bao Hua (Barry Song)
2021-08-04 10:22     ` Mel Gorman
2021-07-26 10:22 ` [PATCH 8/9] sched/fair: select idle cpu from idle cpumask for task wakeup Mel Gorman
2021-08-02 10:41   ` Song Bao Hua (Barry Song) [this message]
2021-08-04 10:26     ` Mel Gorman
2021-08-05  0:23       ` Aubrey Li
2021-09-17  3:44         ` Barry Song
2021-09-17  4:15         ` Barry Song
2021-09-17  9:11           ` Aubrey Li
2021-09-17 13:35             ` Mel Gorman
2021-07-26 10:22 ` [PATCH 9/9] sched/core: Delete SIS_PROP and rely on the idle cpu mask Mel Gorman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=9dde989df06b483790cc24dc7670a919@hisilicon.com \
    --to=song.bao.hua@hisilicon.com \
    --cc=aubrey.li@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@techsingularity.net \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=valentin.schneider@arm.com \
    --cc=vincent.guittot@linaro.org \
    --cc=yangyicong@huawei.com \
    --subject='RE: [PATCH 8/9] sched/fair: select idle cpu from idle cpumask for task wakeup' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

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).