linux-kernel - Re: [PATCH 4/4] PM / devfreq: Handle monitor start/stop in the devfreq core

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20190214192855.GD117604@google.com>
Date:   Thu, 14 Feb 2019 11:28:55 -0800
From:   Matthias Kaehlcke <mka@...omium.org>
To:     Chanwoo Choi <cwchoi00@...il.com>
Cc:     MyungJoo Ham <myungjoo.ham@...sung.com>,
        Kyungmin Park <kyungmin.park@...sung.com>,
        Chanwoo Choi <cw00.choi@...sung.com>,
        Thierry Reding <thierry.reding@...il.com>,
        Jonathan Hunter <jonathanh@...dia.com>,
        Linux PM list <linux-pm@...r.kernel.org>,
        linux-kernel <linux-kernel@...r.kernel.org>,
        linux-tegra@...r.kernel.org,
        Lukasz Luba <l.luba@...tner.samsung.com>
Subject: Re: [PATCH 4/4] PM / devfreq: Handle monitor start/stop in the
 devfreq core

Hi Chanwoo,

On Thu, Feb 14, 2019 at 11:17:36PM +0900, Chanwoo Choi wrote:
> Hi Matthias,
> 
> As I commented on the first patch, it is not possible to call some codes
> according to the intention of each governor between 'devfreq_moniotr_*()'
> and some codes which are executed before or after 'devfreq_moniotr_*()'
> 
> For example, if some governor requires the following sequence,
> after this patch, it is not possible.
> 
> case DEVFREQ_GOV_xxx:
>         /* execute some code before devfreq_monitor_xxx() */
>         devfreq_monitor_xxx()
>         /* execute some code after devfreq_monitor_xxx() */

As for the suspend/resume case I agree that the patch introduces this
limitation, but I'm not convinced that this is an actual problem.

For governor_start(): why can't the governor execute the code
before polling started, does it make any difference to the governor
that a work is scheduled?

For governor_stop(): why would the governor require polling to be
active during stop? If it needs update_devfreq() to run (called by
devfreq_monitor()) it can call it directly, instead of waiting for the
monitor to run at some later time.

Cheers

Matthias