From: Peter Zijlstra <peterz@infradead.org>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: "Niklas Söderlund" <niklas.soderlund@ragnatech.se>,
"Heiner Kallweit" <hkallweit1@gmail.com>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
"Ingo Molnar" <mingo@redhat.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
linux-renesas-soc@vger.kernel.org
Subject: Re: Potential problem with 31e77c93e432dec7 ("sched/fair: Update blocked load when newly idle")
Date: Thu, 26 Apr 2018 13:48:35 +0200 [thread overview]
Message-ID: <20180426114835.GI4064@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20180426103133.GA6953@linaro.org>
On Thu, Apr 26, 2018 at 12:31:33PM +0200, Vincent Guittot wrote:
> From: Vincent Guittot <vincent.guittot@linaro.org>
> Date: Thu, 26 Apr 2018 12:19:32 +0200
> Subject: [PATCH] sched/fair: fix the update of blocked load when newly idle
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
>
> With commit 31e77c93e432 ("sched/fair: Update blocked load when newly idle"),
> we release the rq->lock when updating blocked load of idle CPUs. This open
> a time window during which another CPU can add a task to this CPU's cfs_rq.
> The check for newly added task of idle_balance() is not in the common path.
> Move the out label to include this check.
Ah quite so indeed. This could result in us running idle even though
there is a runnable task around -- which is bad.
> Fixes: 31e77c93e432 ("sched/fair: Update blocked load when newly idle")
> Reported-by: Heiner Kallweit <hkallweit1@gmail.com>
> Reported-by: Niklas Söderlund <niklas.soderlund@ragnatech.se>
> Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
> ---
> kernel/sched/fair.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 0951d1c..15a9f5e 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9847,6 +9847,7 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf)
> if (curr_cost > this_rq->max_idle_balance_cost)
> this_rq->max_idle_balance_cost = curr_cost;
>
> +out:
> /*
> * While browsing the domains, we released the rq lock, a task could
> * have been enqueued in the meantime. Since we're not going idle,
> @@ -9855,7 +9856,6 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf)
> if (this_rq->cfs.h_nr_running && !pulled_task)
> pulled_task = 1;
>
> -out:
> /* Move the next balance forward */
> if (time_after(this_rq->next_balance, next_balance))
> this_rq->next_balance = next_balance;
next prev parent reply other threads:[~2018-04-26 11:48 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-12 9:18 Niklas Söderlund
2018-04-12 10:33 ` Vincent Guittot
2018-04-12 11:15 ` Niklas Söderlund
[not found] ` <20180412133031.GA551@linaro.org>
2018-04-12 19:43 ` Heiner Kallweit
2018-04-14 11:21 ` Vincent Guittot
2018-04-12 22:39 ` Niklas Söderlund
2018-04-14 11:24 ` Vincent Guittot
2018-04-20 16:00 ` Vincent Guittot
2018-04-20 16:30 ` Joel Fernandes
2018-04-22 22:18 ` Niklas Söderlund
2018-04-23 9:54 ` Vincent Guittot
[not found] ` <20180425225603.GA26177@bigcity.dyn.berto.se>
2018-04-26 10:31 ` Vincent Guittot
2018-04-26 11:48 ` Peter Zijlstra [this message]
2018-04-26 14:41 ` Niklas Söderlund
2018-04-26 15:27 ` Vincent Guittot
2018-04-26 15:38 ` Niklas Söderlund
2018-05-02 13:40 ` Geert Uytterhoeven
2018-05-03 9:25 ` [tip:sched/urgent] sched/fair: Fix the update of blocked load when newly idle tip-bot for Vincent Guittot
2018-04-13 20:38 ` Potential problem with 31e77c93e432dec7 ("sched/fair: Update blocked load when newly idle") Niklas Söderlund
2018-04-14 11:26 ` Vincent Guittot
2018-04-12 22:06 ` Niklas Söderlund
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=20180426114835.GI4064@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=hkallweit1@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-renesas-soc@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=niklas.soderlund@ragnatech.se \
--cc=paulmck@linux.vnet.ibm.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
Powered by JetHome