Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 7 Aug 2020 12:46:18 +0200
From:   peterz@infradead.org
To:     Juri Lelli <juri.lelli@redhat.com>
Cc:     mingo@redhat.com, rostedt@goodmis.org, tglx@linutronix.de,
        linux-kernel@vger.kernel.org, luca.abeni@santannapisa.it,
        tommaso.cucinotta@santannapisa.it, alessio.balsini@gmail.com,
        bristot@redhat.com, dietmar.eggemann@arm.com,
        linux-rt-users@vger.kernel.org, mtosatti@redhat.com,
        williams@redhat.com, valentin.schneider@arm.com
Subject: Re: [RFC PATCH v2 6/6] sched/fair: Implement starvation monitor
Message-ID: <20200807104618.GH2674@hirez.programming.kicks-ass.net>
References: <20200807095051.385985-1-juri.lelli@redhat.com>
 <20200807095604.GO42956@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200807095604.GO42956@localhost.localdomain>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Aug 07, 2020 at 11:56:04AM +0200, Juri Lelli wrote:
> Starting deadline server for lower priority classes right away when
> first task is enqueued might break guarantees, as tasks belonging to
> intermediate priority classes could be uselessly preempted. E.g., a well
> behaving (non hog) FIFO task can be preempted by NORMAL tasks even if
> there are still CPU cycles available for NORMAL tasks to run, as they'll
> be running inside the fair deadline server for some period of time.
> 
> To prevent this issue, implement a starvation monitor mechanism that
> starts the deadline server only if a (fair in this case) task hasn't
> been scheduled for some interval of time after it has been enqueued.
> Use pick/put functions to manage starvation monitor status.

One thing I considerd was scheduling this as a least-laxity entity --
such that it runs late, not early -- and start the server when
rq->nr_running != rq->cfs.h_nr_running, IOW when there's !fair tasks
around.

Not saying we should do it like that, but that's perhaps more
deterministic than this.