Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:5 as permitted sender) client-ip=2620:137:e000::3:5;
Sender: Ingo Molnar <mingo.kernel.org@gmail.com>
Date:   Thu, 28 Sep 2023 23:21:42 +0200
From:   Ingo Molnar <mingo@kernel.org>
To:     Valentin Schneider <vschneid@redhat.com>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        linux-kernel@vger.kernel.org, linux-tip-commits@vger.kernel.org,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        x86@kernel.org
Subject: Re: [tip: sched/core] sched/rt: Make rt_rq->pushable_tasks updates
 drive rto_mask
Message-ID: <ZRXuZmjU2tP8ETP5@gmail.com>
References: <20230811112044.3302588-1-vschneid@redhat.com>
 <169563211069.27769.17070510461354463740.tip-bot2@tip-bot2>
 <20230925101127.GB31921@noisy.programming.kicks-ass.net>
 <xhsmh1qemv3x7.mognet@vschneid.remote.csb>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <xhsmh1qemv3x7.mognet@vschneid.remote.csb>
Precedence: bulk


* Valentin Schneider <vschneid@redhat.com> wrote:

> > sched/deadline.c has something very similar, does that need updating 
> > too?
> 
> Hm I think so yes:
> - push_dl_task() is an obvious noop if the pushable tree is empty
> - pull_dl_task() can be kicked if !rq->dl.overloaded, which similarly to rt
>   is driven by nr_migratory but could be boiled down to having pushable
>   tasks (due to the nr_cpus_allowed > 1 constraint).
> 
> Lemme poke at it.

For reference, the matching DL change now lives in tip:sched/core as:

   177f608a4c82 ("sched/deadline: Make dl_rq->pushable_dl_tasks update drive dl_rq->overloaded")

Thanks,

	Ingo