linux-kernel - Re: [PATCH v4] cpufreq: stats: Add 'load_table' debugfs file to show accumulated data of CPUs

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: <CAKohpokZP=zJFoMGmQ7RY7OrttmbG+o2Hf+6ybJdx9PFqKmNPg@mail.gmail.com>
Date:	Wed, 3 Jul 2013 12:11:14 +0530
From:	Viresh Kumar <viresh.kumar@...aro.org>
To:	Chanwoo Choi <cw00.choi@...sung.com>
Cc:	rjw@...k.pl, linux-kernel@...r.kernel.org,
	linux-pm@...r.kernel.org, cpufreq@...r.kernel.org,
	kyungmin.park@...sung.com, myungjoo.ham@...sung.com
Subject: Re: [PATCH v4] cpufreq: stats: Add 'load_table' debugfs file to show
 accumulated data of CPUs

On 2 July 2013 16:19, Chanwoo Choi <cw00.choi@...sung.com> wrote:
> Previously, I sent two reply about your reply. But, Please ignore previous reply.
> Those have wrong function flow about creating sysfs file and poor wrong opinion.
>
> I'm so sorry if you're confused.

Its okay.

> OK. I'll create link for CPU[1-3] if multiple core use one cpufreq policy.
> In result, we can check below directory structure.
>
> sh-4.1# ls -al /sys/kernel/debug/cpufreq/
> total 0
> drwxr-xr-x  3 root root 0 Jan  1 09:00 .
> drwx------ 26 root root 0 Jan  1 09:00 ..
> drwxr-xr-x  2 root root 0 Jan  1 09:00 cpu0
> lrwxrwxrwx  1 root root 0 Jan  1 09:00 cpu1 -> ./cpu0
> lrwxrwxrwx  1 root root 0 Jan  1 09:00 cpu2 -> ./cpu0
> lrwxrwxrwx  1 root root 0 Jan  1 09:00 cpu3 -> ./cpu0
>
> And then I will create load_table debugfs file below of /sys/kernel/debug/cpufreq/cpu0
> in cpufreq_stats.c

Looks good.

> I understand your opinion. But I have a suggestion for using load_table debugfs file
> if cpufreq governor is performance/powersave.
>
> So, I suggest that performance/powersave governor may need to executes dbs_check_cpu()
> to calculate CPUx load. Sometimes, we need CPUs load on performance/powersave
> govenor because we could get different power-consumption according to CPUs load
> on same cpu frequency when we estimate power-consumption on specific test case.

How will these two call dbs_check_cpu()? We don't have a timer for
those two governors :)

Okay do one thing. Create debugfs and debug/cpufreq/cpuX directory always.
Let load_table contain zero when we have irrelevant governors set for it.

We will see if we want some smart code in place for this or not.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/