Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230411042511.1606592-1-jstultz@google.com> <20230411042511.1606592-10-jstultz@google.com>
 <20230422101427.GD1214746@hirez.programming.kicks-ass.net>
In-Reply-To: <20230422101427.GD1214746@hirez.programming.kicks-ass.net>
From:   John Stultz <jstultz@google.com>
Date:   Tue, 25 Apr 2023 15:52:31 +0100
Message-ID: <CANDhNCr3acrEpBYd2LVkY3At=HCDZxGWqbMMwzVJ-Mn--dv3DA@mail.gmail.com>
Subject: Re: [PATCH v3 09/14] sched: Split scheduler execution context
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Joel Fernandes <joelaf@google.com>,
        Qais Yousef <qyousef@google.com>,
        Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Valentin Schneider <vschneid@redhat.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>,
        Zimuzo Ezeozue <zezeozue@google.com>,
        Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Will Deacon <will@kernel.org>,
        Waiman Long <longman@redhat.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        "Paul E . McKenney" <paulmck@kernel.org>,
        "Connor O'Brien" <connoro@google.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Sat, Apr 22, 2023 at 11:14=E2=80=AFAM Peter Zijlstra <peterz@infradead.o=
rg> wrote:
>
> On Tue, Apr 11, 2023 at 04:25:06AM +0000, John Stultz wrote:
> > From: Peter Zijlstra <peterz@infradead.org>
> >
> > Lets define the scheduling context as all the scheduler state in
> > task_struct and the execution context as all state required to run the
> > task.
> >
> > Currently both are intertwined in task_struct. We want to logically
> > split these such that we can run the execution context of one task
> > with the scheduling context of another.
> >
> > To this purpose introduce rq_selected() to point to the task_struct
> > used for scheduler state and preserve rq_curr() to denote the execution
> > context.
>
> I can't say I like the rq_selected() naming :/

So, I'm not married to any particular naming, but the earlier use of
"rq->proxy" in the earlier patches really made very little sense, at
least from my perspective of the word (especially as it had logical
knots  where it would set rq->proxy to what pick_next_task() returned,
and then go and set curr to the results of proxy()).  So it seemed
cleanest to separate it out and use a different term.

We could do "rq_picked()", or "rq_next", or ...  suggestions are welcome.

Thanks again for taking a look here and providing feedback! I really
appreciate it!
-john