Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 5 Mar 2020 09:30:30 -0800
From:   Kees Cook <keescook@chromium.org>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     cl@rock-chips.com, heiko@sntech.de, mingo@redhat.com,
        juri.lelli@redhat.com, vincent.guittot@linaro.org,
        dietmar.eggemann@arm.com, rostedt@goodmis.org, bsegall@google.com,
        mgorman@suse.de, akpm@linux-foundation.org, tglx@linutronix.de,
        mpe@ellerman.id.au, surenb@google.com, ben.dooks@codethink.co.uk,
        anshuman.khandual@arm.com, catalin.marinas@arm.com,
        will@kernel.org, luto@amacapital.net, wad@chromium.org,
        mark.rutland@arm.com, geert+renesas@glider.be,
        george_davis@mentor.com, sudeep.holla@arm.com,
        linux@armlinux.org.uk, gregkh@linuxfoundation.org, info@metux.net,
        kstewart@linuxfoundation.org, allison@lohutok.net,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        huangtao@rock-chips.com
Subject: Re: [PATCH v1 1/1] sched/fair: do not preempt current task if it is
 going to call schedule()
Message-ID: <202003050929.DD4DB3529@keescook>
References: <20200305081611.29323-1-cl@rock-chips.com>
 <20200305081611.29323-2-cl@rock-chips.com>
 <20200305095803.GW2596@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200305095803.GW2596@hirez.programming.kicks-ass.net>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Mar 05, 2020 at 10:58:03AM +0100, Peter Zijlstra wrote:
> On Thu, Mar 05, 2020 at 04:16:11PM +0800, cl@rock-chips.com wrote:
> > From: Liang Chen <cl@rock-chips.com>
> > 
> > when we create a kthread with ktrhead_create_on_cpu(),the child thread
> > entry is ktread.c:ktrhead() which will be preempted by the parent after
> > call complete(done) while schedule() is not called yet,then the parent
> > will call wait_task_inactive(child) but the child is still on the runqueue,
> > so the parent will schedule_hrtimeout() for 1 jiffy,it will waste a lot of
> > time,especially on startup.
> > 
> >   parent                             child
> > ktrhead_create_on_cpu()
> >   wait_fo_completion(&done) -----> ktread.c:ktrhead()
> >                              |----- complete(done);--wakeup and preempted by parent
> >  kthread_bind() <------------|  |-> schedule();--dequeue here
> >   wait_task_inactive(child)     |
> >    schedule_hrtimeout(1 jiffy) -|
> > 
> > So we hope the child just wakeup parent but not preempted by parent, and the
> > child is going to call schedule() soon,then the parent will not call
> > schedule_hrtimeout(1 jiffy) as the child is already dequeue.
> > 
> > The same issue for ktrhead_park()&&kthread_parkme().
> > This patch can save 120ms on rk312x startup with CONFIG_HZ=300.
> 
> > diff --git a/kernel/kthread.c b/kernel/kthread.c
> > index b262f47046ca..8a4e4c9cdc22 100644
> > --- a/kernel/kthread.c
> > +++ b/kernel/kthread.c
> > @@ -199,8 +199,10 @@ static void __kthread_parkme(struct kthread *self)
> >  		if (!test_bit(KTHREAD_SHOULD_PARK, &self->flags))
> >  			break;
> >  
> > +		set_tsk_going_to_sched(current);
> >  		complete(&self->parked);
> >  		schedule();
> > +		clear_tsk_going_to_sched(current);
> >  	}
> >  	__set_current_state(TASK_RUNNING);
> >  }
> > @@ -245,8 +247,10 @@ static int kthread(void *_create)
> >  	/* OK, tell user we're spawned, wait for stop or wakeup */
> >  	__set_current_state(TASK_UNINTERRUPTIBLE);
> >  	create->result = current;
> > +	set_tsk_going_to_sched(current);
> >  	complete(done);
> >  	schedule();
> > +	clear_tsk_going_to_sched(current);
> >  
> >  	ret = -EINTR;
> >  	if (!test_bit(KTHREAD_SHOULD_STOP, &self->flags)) {
> 
> Were you looking for this? I think it does the same without having
> fallen from the ugly tree...
> 
> diff --git a/kernel/kthread.c b/kernel/kthread.c
> index b262f47046ca..62699ff414f4 100644
> --- a/kernel/kthread.c
> +++ b/kernel/kthread.c
> @@ -199,8 +199,10 @@ static void __kthread_parkme(struct kthread *self)
>  		if (!test_bit(KTHREAD_SHOULD_PARK, &self->flags))
>  			break;
>  
> +		preempt_disable()
>  		complete(&self->parked);
> -		schedule();
> +		schedule_preempt_disabled();
> +		preempt_enable();
>  	}
>  	__set_current_state(TASK_RUNNING);
>  }
> @@ -245,8 +247,10 @@ static int kthread(void *_create)
>  	/* OK, tell user we're spawned, wait for stop or wakeup */
>  	__set_current_state(TASK_UNINTERRUPTIBLE);
>  	create->result = current;
> +	preempt_disable()
>  	complete(done);
> -	schedule();
> +	schedule_preempt_disabled();
> +	preempt_enable();
>  
>  	ret = -EINTR;
>  	if (!test_bit(KTHREAD_SHOULD_STOP, &self->flags)) {

That's much nicer, yes! :) As I said, I don't know much about the
scheduler. ;)

-- 
Kees Cook