linux-kernel - Re: [PATCH v1 2/7] drivers: cpufreq: Add thermal_cooling_device pointer to struct cpufreq

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, 9 Jan 2019 17:01:57 -0800
From:   Matthias Kaehlcke <mka@...omium.org>
To:     Amit Kucheria <amit.kucheria@...aro.org>
Cc:     linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org,
        bjorn.andersson@...aro.org, viresh.kumar@...aro.org,
        edubezval@...il.com, andy.gross@...aro.org, tdas@...eaurora.org,
        swboyd@...omium.org, dianders@...omium.org,
        "Rafael J. Wysocki" <rjw@...ysocki.net>,
        "open list:CPU FREQUENCY DRIVERS" <linux-pm@...r.kernel.org>
Subject: Re: [PATCH v1 2/7] drivers: cpufreq: Add thermal_cooling_device
 pointer to struct cpufreq_policy

On Thu, Jan 10, 2019 at 05:30:51AM +0530, Amit Kucheria wrote:
> Several cpufreq drivers register themselves as thermal cooling devices.
> Adding a pointer to struct cpufreq_policy removes the need for them to
> store this pointer in a private data structure.
> 
> Signed-off-by: Amit Kucheria <amit.kucheria@...aro.org>
> ---
>  include/linux/cpufreq.h | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index c86d6d8bdfed..2496549d7573 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -95,6 +95,8 @@ struct cpufreq_policy {
>  	struct cpufreq_frequency_table	*freq_table;
>  	enum cpufreq_table_sorting freq_table_sorted;
>  
> +	struct thermal_cooling_device *cooldev; /* Pointer to the cooling
> +						 * device if used for thermal mitigation */
>  	struct list_head        policy_list;
>  	struct kobject		kobj;
>  	struct completion	kobj_unregister;

I've mixed feelings about this. It's definitely desirable to avoid
code duplication and tying the cooling device to the cpufreq_policy is
a convenient way to achieve that. However semantically it seems a bit
odd that a CPU cooling device is part of the cpufreq policy.

Anyway, unless there are better ideas we probably want to be pragmatic
here, so if Viresh is fine with it who am I to complain ;-)

Cheers

Matthias