linux-pm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Parth Shah <parth@linux.ibm.com>
To: linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org
Cc: peterz@infradead.org, mingo@redhat.com,
	vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
	patrick.bellasi@matbug.net, valentin.schneider@arm.com,
	pavel@ucw.cz, dsmythies@telus.net, quentin.perret@arm.com,
	rafael.j.wysocki@intel.com, tim.c.chen@linux.intel.com,
	daniel.lezcano@linaro.org
Subject: [RFC v5 5/6] sched/fair: Provide arch hook to find domain for non idle core search scan
Date: Mon,  7 Oct 2019 14:00:50 +0530	[thread overview]
Message-ID: <20191007083051.4820-6-parth@linux.ibm.com> (raw)
In-Reply-To: <20191007083051.4820-1-parth@linux.ibm.com>

Specify the method which returns cpumask within which to limit the
search for a non idle core. By default, limit the search in LLC domain
which usually includes few/all the cores in the processor chip.

The select_non_idle_core searches for the non idle cores in the LLC domain.
But in the systems with multiple NUMA domains, the Turbo frequency can be
sustained within the NUMA domain without being affected from other
NUMA. For such case, arch_turbo_domain can be tuned to change domain for
non idle core search.

Signed-off-by: Parth Shah <parth@linux.ibm.com>
---
 kernel/sched/fair.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d4a1b6474338..731d1aaa6bc0 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5967,6 +5967,14 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
 }
 
 #ifdef CONFIG_SCHED_SMT
+
+#ifndef arch_turbo_domain
+static __always_inline struct cpumask *arch_turbo_domain(int cpu)
+{
+	return sched_domain_span(rcu_dereference(per_cpu(sd_llc, cpu)));
+}
+#endif
+
 static inline bool is_background_task(struct task_struct *p)
 {
 	if (p->flags & PF_CAN_BE_PACKED)
@@ -5991,7 +5999,8 @@ static int select_non_idle_core(struct task_struct *p, int prev_cpu, int target)
 	struct cpumask *cpus = this_cpu_cpumask_var_ptr(turbo_sched_mask);
 	int iter_cpu, sibling;
 
-	cpumask_and(cpus, cpu_online_mask, p->cpus_ptr);
+	cpumask_and(cpus, cpu_online_mask, arch_turbo_domain(prev_cpu));
+	cpumask_and(cpus, cpus, p->cpus_ptr);
 
 	for_each_cpu_wrap(iter_cpu, cpus, prev_cpu) {
 		int idle_cpu_count = 0, non_idle_cpu_count = 0;
-- 
2.17.1


  parent reply	other threads:[~2019-10-07  8:31 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-07  8:30 [RFC v5 0/6] TurboSched: A scheduler for sustaining Turbo Frequencies for longer durations Parth Shah
2019-10-07  8:30 ` [RFC v5 1/6] sched/core: Add manual background task classification using sched_setattr syscall Parth Shah
2019-10-07  8:30 ` [RFC v5 2/6] sched: Introduce switch to enable TurboSched for task packing Parth Shah
2019-10-07  8:30 ` [RFC v5 3/6] sched/core: Update turbo_sched count only when required Parth Shah
2019-10-07  8:30 ` [RFC v5 4/6] sched/fair: Tune task wake-up logic to pack small background tasks on fewer cores Parth Shah
2019-10-07 12:19   ` Vincent Guittot
2019-10-07 16:53     ` Parth Shah
2019-10-08 16:20       ` Vincent Guittot
2019-10-09  8:46         ` Parth Shah
2019-10-08 16:52       ` Dietmar Eggemann
2019-10-09  8:57         ` Parth Shah
2019-10-09 14:26           ` Dietmar Eggemann
2019-10-09 17:02             ` Parth Shah
2019-10-10 14:53               ` Dietmar Eggemann
2019-10-07  8:30 ` Parth Shah [this message]
2019-10-07  8:30 ` [RFC v5 6/6] powerpc: Set turbo domain to NUMA node for task packing Parth Shah
     [not found] ` <20191008132842.6612-1-hdanton@sina.com>
2019-10-09  9:22   ` [RFC v5 4/6] sched/fair: Tune task wake-up logic to pack small background tasks on fewer cores Parth Shah
2019-10-09 11:34     ` Vincent Guittot
2019-10-09 16:55       ` Parth Shah

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=20191007083051.4820-6-parth@linux.ibm.com \
    --to=parth@linux.ibm.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=dietmar.eggemann@arm.com \
    --cc=dsmythies@telus.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=patrick.bellasi@matbug.net \
    --cc=pavel@ucw.cz \
    --cc=peterz@infradead.org \
    --cc=quentin.perret@arm.com \
    --cc=rafael.j.wysocki@intel.com \
    --cc=tim.c.chen@linux.intel.com \
    --cc=valentin.schneider@arm.com \
    --cc=vincent.guittot@linaro.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).