linux-kernel - Re: [RFT][PATCH v4 3/7] sched: idle: Do not stop the tick before cpuidle_idle

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]

Message-ID: <CAJZ5v0jGi_ewnJXxwuY1-y6Ti64Y601A4DQuDhtMJ+riYPJpzA@mail.gmail.com>
Date:   Thu, 15 Mar 2018 21:41:10 +0100
From:   "Rafael J. Wysocki" <rafael@...nel.org>
To:     Frederic Weisbecker <frederic@...nel.org>
Cc:     "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Peter Zijlstra <peterz@...radead.org>,
        Linux PM <linux-pm@...r.kernel.org>,
        Frederic Weisbecker <fweisbec@...il.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Paul McKenney <paulmck@...ux.vnet.ibm.com>,
        Thomas Ilsche <thomas.ilsche@...dresden.de>,
        Doug Smythies <dsmythies@...us.net>,
        Rik van Riel <riel@...riel.com>,
        Aubrey Li <aubrey.li@...ux.intel.com>,
        Mike Galbraith <mgalbraith@...e.de>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [RFT][PATCH v4 3/7] sched: idle: Do not stop the tick before cpuidle_idle_call()

On Thu, Mar 15, 2018 at 7:19 PM, Frederic Weisbecker
<frederic@...nel.org> wrote:
> On Mon, Mar 12, 2018 at 10:53:25AM +0100, Rafael J. Wysocki wrote:
>> From: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
>>
>> Make cpuidle_idle_call() decide whether or not to stop the tick.
>>
>> First, the cpuidle_enter_s2idle() path deals with the tick (and with
>> the entire timekeeping for that matter) by itself and it doesn't need
>> the tick to be stopped beforehand.
>
> Not sure you meant timekeeping either :)

Yeah, I meant nohz.

>>       if (idle_should_enter_s2idle() || dev->use_deepest_state) {
>>               if (idle_should_enter_s2idle()) {
>> +                     rcu_idle_enter();
>> +
>>                       entered_state = cpuidle_enter_s2idle(drv, dev);
>>                       if (entered_state > 0) {
>>                               local_irq_enable();
>>                               goto exit_idle;
>>                       }
>> +
>> +                     rcu_idle_exit();
>>               }
>
> I'm not sure how the tick is stopped on suspend to idle. Perhaps through
> hrtimer (tick_cancel_sched_timer()) or clockevents code.

The latter.

It does clockevents_shutdown() down the road, eventually.

IOW, it couldn't care less. :-)

> But we may have a similar problem than with idle_poll() called right after
> call_cpuidle(). Ie: we arrive in cpuidle_enter_s2idle() with a tick that
> should be reprogrammed while it is not. No idea if that can hurt somehow.
>
> I guess it depends what happens to the tick on s2idle, I'm not clear with that.

No problem there, AFAICS.