From: Peter Zijlstra <peterz@infradead.org>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: mingo@redhat.com, juri.lelli@redhat.com,
dietmar.eggemann@arm.com, rostedt@goodmis.org,
bsegall@google.com, mgorman@suse.de, bristot@redhat.com,
linux-kernel@vger.kernel.org, tim.c.chen@linux.intel.com,
joel@joelfernandes.org
Subject: Re: [PATCH 1/2] sched/fair: skip newidle update stats
Date: Fri, 12 Nov 2021 15:55:27 +0100 [thread overview]
Message-ID: <YY6AXx2lhJ27IBuJ@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <YY6ARaFrpuWYqege@hirez.programming.kicks-ass.net>
Subject: sched/fair: Reflow newidle_balance()
From: Peter Zijlstra <peterz@infradead.org>
Date: Fri Nov 12 15:46:08 CET 2021
The control flow in newidle_balance() is a little convoluted, attempt
simplification.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
kernel/sched/fair.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -10858,10 +10858,10 @@ static inline void nohz_newidle_balance(
static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
{
unsigned long next_balance = jiffies + HZ;
+ int pulled_task = 0, timeout = 0;
int this_cpu = this_rq->cpu;
u64 t0, t1, curr_cost = 0;
struct sched_domain *sd;
- int pulled_task = 0, early_stop = 0;
update_misfit_status(NULL, this_rq);
@@ -10889,17 +10889,9 @@ static int newidle_balance(struct rq *th
if (!READ_ONCE(this_rq->rd->overload) ||
(sd && this_rq->avg_idle < sd->max_newidle_lb_cost)) {
- if (sd) {
+ if (sd)
update_next_balance(sd, &next_balance);
- /*
- * We skip new idle LB because there is not enough
- * time before next wake up. Make sure that we will
- * not kick NOHZ_NEWILB_KICK
- */
- early_stop = 1;
- }
-
goto out;
}
@@ -10922,7 +10914,7 @@ static int newidle_balance(struct rq *th
update_next_balance(sd, &next_balance);
if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) {
- early_stop = 1;
+ timeout = 1;
break;
}
@@ -10967,6 +10959,11 @@ static int newidle_balance(struct rq *th
if (this_rq->nr_running != this_rq->cfs.h_nr_running)
pulled_task = -1;
+ if (pulled_task || timeout)
+ goto out;
+
+ nohz_newidle_balance(this_rq);
+
out:
/* Move the next balance forward */
if (time_after(this_rq->next_balance, next_balance))
@@ -10974,8 +10971,6 @@ static int newidle_balance(struct rq *th
if (pulled_task)
this_rq->idle_stamp = 0;
- else if (!early_stop)
- nohz_newidle_balance(this_rq);
return pulled_task;
}
next prev parent reply other threads:[~2021-11-12 14:55 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-12 9:58 [PATCH 0/2] avoid spurious blocked load update Vincent Guittot
2021-11-12 9:58 ` [PATCH 1/2] sched/fair: skip newidle update stats Vincent Guittot
2021-11-12 14:29 ` Peter Zijlstra
2021-11-12 14:47 ` Vincent Guittot
2021-11-12 15:29 ` Peter Zijlstra
2021-11-12 16:00 ` Vincent Guittot
2021-11-12 14:55 ` Peter Zijlstra
2021-11-12 14:55 ` Peter Zijlstra [this message]
2021-11-12 16:05 ` Vincent Guittot
2021-11-16 23:48 ` [PATCH 0/2] avoid spurious blocked load update Tim Chen
2021-11-18 14:40 ` Vincent Guittot
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=YY6AXx2lhJ27IBuJ@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=joel@joelfernandes.org \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=rostedt@goodmis.org \
--cc=tim.c.chen@linux.intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.