From: Valentin Schneider <valentin.schneider@arm.com>
To: Peter Zijlstra <peterz@infradead.org>,
mingo@kernel.org, vincent.guittot@linaro.org,
dietmar.eggemann@arm.com, juri.lelli@redhat.com,
rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de
Cc: linux-kernel@vger.kernel.org, qperret@google.com,
qais.yousef@arm.com, ktkhai@virtuozzo.com
Subject: Re: [PATCH 1/7] sched: Fix pick_next_task() vs change pattern race
Date: Fri, 8 Nov 2019 16:05:25 +0000 [thread overview]
Message-ID: <e19c566b-dc14-a5aa-de4f-c67cdb17620c@arm.com> (raw)
In-Reply-To: <20191108131909.428842459@infradead.org>
On 08/11/2019 13:15, Peter Zijlstra wrote:
> Fixes: 67692435c411 ("sched: Rework pick_next_task() slow-path")
> Reported-by: Quentin Perret <qperret@google.com>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
I've been running the same reproducer as Quentin's for a similar length of
time (~3h) and it's still going strong, so FWIW:
Tested-by: Valentin Schneider <valentin.schneider@arm.com>
> ---
> kernel/sched/core.c | 21 +++++++++++++++------
> kernel/sched/deadline.c | 40 ++++++++++++++++++++--------------------
> kernel/sched/fair.c | 15 ++++++++++++---
> kernel/sched/idle.c | 9 ++++++++-
> kernel/sched/rt.c | 37 +++++++++++++++++++------------------
> kernel/sched/sched.h | 30 +++++++++++++++++++++++++++---
> kernel/sched/stop_task.c | 18 +++++++++++-------
> 7 files changed, 112 insertions(+), 58 deletions(-)
>
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -3929,13 +3929,22 @@ pick_next_task(struct rq *rq, struct tas
> }
>
> restart:
> +#ifdef CONFIG_SMP
I suppose we could dump that in a core balance_prev_task() to avoid the
inline #ifdeffery, but eh...
> /*
> - * Ensure that we put DL/RT tasks before the pick loop, such that they
> - * can PULL higher prio tasks when we lower the RQ 'priority'.
> + * We must do the balancing pass before put_next_task(), such
> + * that when we release the rq->lock the task is in the same
> + * state as before we took rq->lock.
> + *
> + * We can terminate the balance pass as soon as we know there is
> + * a runnable task of @class priority or higher.
> */
> - prev->sched_class->put_prev_task(rq, prev, rf);
> - if (!rq->nr_running)
> - newidle_balance(rq, rf);
> + for_class_range(class, prev->sched_class, &idle_sched_class) {
> + if (class->balance(rq, prev, rf))
> + break;
> + }
> +#endif
> +
> + put_prev_task(rq, prev);
>
> for_each_class(class) {
> p = class->pick_next_task(rq, NULL, NULL);
> @@ -1469,6 +1469,22 @@ static void check_preempt_equal_prio(str
> resched_curr(rq);
> }
>
> +static int balance_rt(struct rq *rq, struct task_struct *p, struct rq_flags *rf)
> +{
> + if (!on_rt_rq(&p->rt) && need_pull_rt_task(rq, p)) {
> + /*
> + * This is OK, because current is on_cpu, which avoids it being
> + * picked for load-balance and preemption/IRQs are still
> + * disabled avoiding further scheduler activity on it and we've
> + * not yet started the picking loop.
> + */
> + rq_unpin_lock(rq, rf);
> + pull_rt_task(rq);
> + rq_repin_lock(rq, rf);
> + }
> +
> + return sched_stop_runnable(rq) || sched_dl_runnable(rq) || sched_rt_runnable(rq);
So we already have some dependencies on the class ordering (e.g. fair->idle),
but I'm wondering if would it make sense to have these runnable functions be
defined as sched_class callbacks?
e.g.
rt_sched_class.runnable = rt_runnable
w/ rt_runnable() just being a non-inlined sched_rt_runnable() you define
further down the patch (or a wrapper to it). The balance return pattern could
then become:
for_class_range(class, sched_class_highest, rt_sched_class->next)
if (class->runnable(rq))
return true;
return false;
(and replace rt_sched_class by whatever class' balance callback this is)
It's a bit neater, but I'm pretty sure it's going to run worse :/
The only unaffected one would be fair, since newidle_balance() already does
that "for free".
> +}
> #endif /* CONFIG_SMP */
>
> /*
next prev parent reply other threads:[~2019-11-08 16:05 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-08 13:15 [PATCH 0/7] scheduler patches Peter Zijlstra
2019-11-08 13:15 ` [PATCH 1/7] sched: Fix pick_next_task() vs change pattern race Peter Zijlstra
2019-11-08 14:28 ` Quentin Perret
2019-11-08 16:05 ` Valentin Schneider [this message]
2019-11-08 20:49 ` Peter Zijlstra
2019-11-08 17:03 ` Qais Yousef
2019-11-08 13:15 ` [PATCH 2/7] sched/fair: Better document newidle_balance() Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2019-11-08 13:15 ` [PATCH 3/7] sched: Make pick_next_task_idle() more consistent Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] sched/core: " tip-bot2 for Peter Zijlstra
2019-11-08 13:15 ` [PATCH 4/7] sched: Optimize pick_next_task() Peter Zijlstra
2019-11-08 14:33 ` Quentin Perret
2019-11-08 16:46 ` Vincent Guittot
2019-11-08 20:46 ` Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] sched/core: " tip-bot2 for Peter Zijlstra
2019-11-08 13:15 ` [PATCH 5/7] sched: Simplify sched_class::pick_next_task() Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] sched/core: " tip-bot2 for Peter Zijlstra
2019-11-08 13:15 ` [PATCH 6/7] sched/fair: Use mul_u32_u32() Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2019-11-08 13:16 ` [PATCH 7/7] sched: Further clarify sched_class::set_next_task() Peter Zijlstra
2019-11-11 9:32 ` [tip: sched/core] sched/core: " tip-bot2 for Peter Zijlstra
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=e19c566b-dc14-a5aa-de4f-c67cdb17620c@arm.com \
--to=valentin.schneider@arm.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=juri.lelli@redhat.com \
--cc=ktkhai@virtuozzo.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=qais.yousef@arm.com \
--cc=qperret@google.com \
--cc=rostedt@goodmis.org \
--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).