linux-kernel - Re: [PATCH v3 2/5] sched/deadline: Fix reclaim inaccuracy with SMP

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 23 May 2023 22:11:23 -0400
From:   Vineeth Remanan Pillai <vineeth@...byteword.org>
To:     luca abeni <luca.abeni@...tannapisa.it>
Cc:     Juri Lelli <juri.lelli@...hat.com>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Joel Fernandes <joel@...lfernandes.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        Valentin Schneider <vschneid@...hat.com>,
        Jonathan Corbet <corbet@....net>, linux-kernel@...r.kernel.org,
        linux-doc@...r.kernel.org
Subject: Re: [PATCH v3 2/5] sched/deadline: Fix reclaim inaccuracy with SMP

Hi Luca,

> >
> > I think it probably makes sense to get the fix for the equation to go
> > in as a first step and then we can investigate more about the second
> > issue (less reclaiming with less load and different bandwidth) and
> > fix it separately. What do you think?
>
> I fully agree. If you split this change in a first patch, IMHO it
> can be applied.
>
Thanks, I shall have it as the first patch in the next iteration.

> BTW, I tried changing the equation without introducing div64, and it
> seems to me that it works well... So, if removing the bw_ratio
> approximation is needed, I think you can do it in a second patch (so,
> the first patch changes the reclaiming equation, and the second one
> introduces div64)
>
There was one more reason for removing the bw_ratio. The second patch
fixing the issue with mix of normal and SCHED_FLAG_RECLAIM tasks,
do not have a constant Umax, so we cannot calculate the bw_ratio in
advance and has to be calculated in grub_reclaim().

As you suggested, I shall remove the bw_ratio as the second patch and
then have the third patch to fix the case of SCHED_FLAG_RECLAIM and
normal deadline tasks. For the rest of the issues, I shall work on it as
a separate patch series once this goes in.

Will send it out tomorrow after a bit more testing.

Thanks again,
Vineeth