Re: [PATCH] sched/rt: Add a new sysctl to control uclamp_util_min

From: Patrick Bellasi <patrick.bellasi@matbug.net>
To: Quentin Perret <qperret@google.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Qais Yousef <qais.yousef@arm.com>, Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Luis Chamberlain <mcgrof@kernel.org>,
	Kees Cook <keescook@chromium.org>,
	Iurii Zaikin <yzaikin@google.com>,
	Juri Lelli <juri.lelli@redhat.com>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
	valentin.schneider@arm.com, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, kernel-team@android.com
Subject: Re: [PATCH] sched/rt: Add a new sysctl to control uclamp_util_min
Date: Wed, 8 Jan 2020 20:16:00 +0100	[thread overview]
Message-ID: <20200108191600.GC9635@darkstar> (raw)
In-Reply-To: <20200108095108.GA153171@google.com>

On 08-Jan 09:51, Quentin Perret wrote:
> On Tuesday 07 Jan 2020 at 20:30:36 (+0100), Dietmar Eggemann wrote:
> > On 07/01/2020 14:42, Quentin Perret wrote:
> > > Hi Qais,
> > > 
> > > On Friday 20 Dec 2019 at 16:48:38 (+0000), Qais Yousef wrote:
> > 
> > [...]
> > 
> > >> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> > >> index e591d40fd645..19572dfc175b 100644
> > >> --- a/kernel/sched/rt.c
> > >> +++ b/kernel/sched/rt.c
> > >> @@ -2147,6 +2147,12 @@ static void pull_rt_task(struct rq *this_rq)
> > >>   */
> > >>  static void task_woken_rt(struct rq *rq, struct task_struct *p)
> > >>  {
> > >> +	/*
> > >> +	 * When sysctl_sched_rt_uclamp_util_min value is changed by the user,
> > >> +	 * we apply any new value on the next wakeup, which is here.
> > >> +	 */
> > >> +	uclamp_rt_sync_default_util_min(p);
> > > 
> > > The task has already been enqueued and sugov has been called by then I
> > > think, so this is a bit late. You could do that in uclamp_rq_inc() maybe?

No, the above sync should neven be done.

For CFS tasks is currenly working by computing the effective clamp
from core.c::enqueue_task() before calling into the scheduling class
callback.

We should use a similar approach for RT tasks thus avoiding rt.c
specific code and ensuring correct effective values are _aggregated_
out of task's _requests_ and system-wide's _constraints_ before
calling sugov.

> > That's probably better.
> > Just to be sure ...we want this feature (an existing rt task gets its
> > UCLAMP_MIN value set when the sysctl changes)

No, that's not a feature. That's an hack I would avoid.

> > because there could be rt tasks running before the sysctl is set?
> 
> Yeah, I was wondering the same thing, but I'd expect sysadmin to want
> this. We could change the min clamp of existing RT tasks in userspace
> instead, but given how simple Qais' lazy update code is, the in-kernel
> looks reasonable to me. No strong opinion, though.
> 
> Thanks,
> Quentin

-- 
#include <best/regards.h>

Patrick Bellasi