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: <20230526145519.2282062-1-vineeth@bitbyteword.org>
 <20230526172959.5440b9dd@nowhere> <CAO7JXPgLFgXANqXY7L-YLbijxJCOgUKmd8nUQizb62JorOXGhA@mail.gmail.com>
In-Reply-To: <CAO7JXPgLFgXANqXY7L-YLbijxJCOgUKmd8nUQizb62JorOXGhA@mail.gmail.com>
From:   Vineeth Remanan Pillai <vineeth@bitbyteword.org>
Date:   Fri, 26 May 2023 11:58:50 -0400
Message-ID: <CAO7JXPit6snFM2oW-uSs-mtbmKr9wpbX2exqAUF0iAiaAO6hOg@mail.gmail.com>
Subject: Re: [PATCH v4 1/2] sched/deadline: Fix bandwidth reclaim equation in GRUB
To:     luca abeni <luca.abeni@santannapisa.it>
Cc:     Juri Lelli <juri.lelli@redhat.com>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Steven Rostedt <rostedt@goodmis.org>,
        Joel Fernandes <joel@joelfernandes.org>,
        youssefesmat@google.com,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Valentin Schneider <vschneid@redhat.com>,
        Jonathan Corbet <corbet@lwn.net>, linux-kernel@vger.kernel.org,
        linux-doc@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi Luca,

>
> On Fri, May 26, 2023 at 11:30=E2=80=AFAM luca abeni <luca.abeni@santannap=
isa.it> wrote:
> > [...]
> > > + * but as "dq =3D -(max{u, (1 - Uinact - Uextra)} / Umax) dt",
> >
> > ...But I think this is wrong (should be "Umax - ...", not "1 - ...").
> > I think patch 2/2 has the same issue.
> >
> Oops sorry, I missed this. Will send the fixed patch as a
> reply to the original v4.
>
On looking again, I think the description is correct. That line
mentions the actual m-GRUB equation from the paper. And then the
comment explains why we use Umax(because of limiting the bandwidth
to RT capacity).

Does it seem correct to you?

Thanks,
Vineeth