linux-kernel - Re: [PATCH v5] cpuidle: menu: Handle stopped tick more aggressively

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] [day] [month] [year] [list]

Message-ID: <20180820110410.GY2494@hirez.programming.kicks-ass.net>
Date:   Mon, 20 Aug 2018 13:04:10 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     leo.yan@...aro.org
Cc:     "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Linux PM <linux-pm@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Frederic Weisbecker <frederic@...nel.org>
Subject: Re: [PATCH v5] cpuidle: menu: Handle stopped tick more aggressively

On Tue, Aug 14, 2018 at 11:44:41PM +0800, leo.yan@...aro.org wrote:
> I agree this patch can resolve the issue you mentioned in the commit
> log, but this patch will be fragile for below case, so it will select
> state1 but not state2, how about you think for this case?
> 
> Idle_state0::
> target_residency   TICK_USEC   Prediction
>     |                 |       /
>     V                 V      /
> -----------------------------------------------------> Time length
>                                 ^                 ^
>                                 |                 |
>                             Idle_state1:       Idle_state2:
>                             target_residency   target_residency

If someone can demonstate this is an actual problem; then we need to fix
it I suppose. But it would be very good to have a real world
need/reproducer for it first.