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:   Wed, 12 Apr 2017 08:35:57 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Xunlei Pang <xpang@...hat.com>
Cc:     xlpang@...hat.com, Luca Abeni <luca.abeni@...tannapisa.it>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        linux-kernel@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>,
        Juri Lelli <juri.lelli@....com>,
        Ingo Molnar <mingo@...hat.com>,
        Tommaso Cucinotta <tommaso.cucinotta@...up.it>,
        RĂ´mulo Silva de Oliveira 
        <romulo.deoliveira@...c.br>,
        Mathieu Poirier <mathieu.poirier@...aro.org>
Subject: Re: [PATCH] sched/deadline: Throttle a constrained task activated
 if overflow

On Wed, 12 Apr 2017 20:30:04 +0800
Xunlei Pang <xpang@...hat.com> wrote:

> Image this deadline task was preempted after running 0.1ms by another one with an
> earlier absolute deadline for 1.9ms, after scheduled back it will have the same status
> (1.9ms remaining runtime, 2ms remaining deadline) under the current implementation.
> 
> I can hardly figure out the difference, why is wake-up(will check dl_entity_overflow())
> so special?

The difference between a wake-up and being preempted is that one is
voluntarily done by the application, the other is done by the
scheduler. The preemption done by the schedule must very well be in
line with the algorithms of the scheduler. But for cases where the task
voluntarily sleeps and then wakes up at some arbitrary time, the
scheduler has no idea what its intent was. Not to mention, it must
still guarantee the bandwidth for other tasks.

-- Steve

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ