From: Viresh Kumar <viresh.kumar@linaro.org>
To: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>,
Linux PM list <linux-pm@vger.kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
Juri Lelli <juri.lelli@arm.com>,
Steve Muckle <steve.muckle@linaro.org>,
Saravana Kannan <skannan@codeaurora.org>
Subject: Re: [PATCH 2/11] cpufreq: governor: Use common mutex for dbs_data protection
Date: Fri, 5 Feb 2016 08:29:48 +0530 [thread overview]
Message-ID: <20160205025948.GE3068@vireshk> (raw)
In-Reply-To: <CAJZ5v0hWj4JgCh76Fg48ex3Svv7BUHR=9+RoA728R9aYU+12Pw@mail.gmail.com>
On 04-02-16, 17:46, Rafael J. Wysocki wrote:
> On Thu, Feb 4, 2016 at 6:09 AM, Viresh Kumar <viresh.kumar@linaro.org> wrote:
> > On 04-02-16, 00:16, Rafael J. Wysocki wrote:
> >> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >>
> >> Every governor relying on the common code in cpufreq_governor.c
> >> has to provide its own mutex in struct common_dbs_data. However,
> >> those mutexes are never used at the same time
> >
> > Why do you think so? I thought they can always be used in parallel.
> >
> > Consider 2 or more policies, one can have ondemand as the governor,
> > whereas other one can have conservative.
> >
> > If CPUs go online/offline or if governors are switching in parallel,
> > then cpufreq_governor_dbs() can very much run in parallel for ondemand
> > and conservative.
> >
> > Or am I missing something here ?
>
> Well, so perhaps the changelog is inaccurate.
>
> However, what's wrong with using a single mutex then?
You are killing the possibility of running the code faster. Consider
this:
- A 16 policy system with N CPUs in every policy (IBM has something
similar only :) )..
- 4 policies using ondemand, 4 using conservative, 4 using powersave
and 4 with performance.
- Now if we try to change governors for all of them in parallel, only
one will be done at a time and others have to wait for this
BIG-kernel lock.
- Ideally the lock shouldn't have been in cdata itself, but dbs_data
only. But there was a specific race because of which we were
required to move it to a higher level, i.e. cdata. And so we killed
the possibility of parallelism of multiple governors of same type
(ofcourse only of update-sampling-rate and cpufreq_governor_dbs()..
So, it makes thing much slower..
--
viresh
next prev parent reply other threads:[~2016-02-05 2:59 UTC|newest]
Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-03 23:12 [PATCH 0/11] cpufreq: governor: ondemand/conservative data structures rework Rafael J. Wysocki
2016-02-03 23:14 ` [PATCH 1/11] cpufreq: Clean up default and fallback governor setup Rafael J. Wysocki
2016-02-04 0:55 ` Saravana Kannan
2016-02-04 5:03 ` Viresh Kumar
2016-02-03 23:16 ` [PATCH 2/11] cpufreq: governor: Use common mutex for dbs_data protection Rafael J. Wysocki
2016-02-04 0:59 ` Saravana Kannan
2016-02-04 5:09 ` Viresh Kumar
2016-02-04 16:46 ` Rafael J. Wysocki
2016-02-05 2:59 ` Viresh Kumar [this message]
2016-02-05 3:06 ` Rafael J. Wysocki
2016-02-05 3:15 ` Rafael J. Wysocki
2016-02-05 3:17 ` Rafael J. Wysocki
2016-02-05 3:24 ` Viresh Kumar
2016-02-05 3:33 ` Rafael J. Wysocki
2016-02-05 3:22 ` Viresh Kumar
2016-02-03 23:22 ` [PATCH 3/11] cpufreq: governor: Use common global_dbs_data pointer Rafael J. Wysocki
2016-02-04 1:11 ` Saravana Kannan
2016-02-04 1:25 ` Rafael J. Wysocki
2016-02-04 1:40 ` Saravana Kannan
2016-02-04 5:38 ` Viresh Kumar
2016-02-04 1:47 ` Saravana Kannan
2016-02-04 5:36 ` Viresh Kumar
[not found] ` <CAHZ_5WxJSDtFyFdCc-D2=HSaPON=3rzUxpxPYsCyZvrV1Nv3qw@mail.gmail.com>
2016-02-04 8:25 ` Viresh Kumar
2016-02-04 11:31 ` Gautham R Shenoy
2016-02-04 11:35 ` Viresh Kumar
2016-02-04 16:52 ` Rafael J. Wysocki
2016-02-05 3:02 ` Viresh Kumar
2016-02-05 3:10 ` Rafael J. Wysocki
2016-02-03 23:29 ` [PATCH 4/11] cpufreq: governor: Avoid passing dbs_data pointers around unnecessarily Rafael J. Wysocki
2016-02-04 1:37 ` Saravana Kannan
2016-02-03 23:31 ` [PATCH 5/11] cpufreq: governor: Put governor structure into common_dbs_data Rafael J. Wysocki
2016-02-04 1:57 ` Saravana Kannan
2016-02-03 23:32 ` [PATCH 6/11] cpufreq: governor: Rename some data types and variables Rafael J. Wysocki
2016-02-03 23:33 ` [PATCH 7/11] cpufreq: governor: Rework cpufreq_governor_dbs() Rafael J. Wysocki
2016-02-04 2:03 ` Saravana Kannan
2016-02-03 23:35 ` [PATCH 8/11] cpufreq: governor: Drop the gov pointer from struct dbs_data Rafael J. Wysocki
2016-02-03 23:35 ` [PATCH 9/11] cpufreq: governor: Rename cpu_common_dbs_info to policy_dbs_info Rafael J. Wysocki
2016-02-03 23:37 ` [PATCH 10/11] cpufreq: governor: Rearrange governor data structures Rafael J. Wysocki
2016-02-03 23:38 ` [PATCH 11/11] cpufreq: governor: Drop cpu argument from dbs_check_cpu() Rafael J. Wysocki
2016-02-04 5:40 ` [PATCH 0/11] cpufreq: governor: ondemand/conservative data structures rework Viresh Kumar
2016-02-04 17:22 ` Rafael J. Wysocki
2016-02-05 2:07 ` [PATCH v2 0/10] " Rafael J. Wysocki
2016-02-05 2:11 ` [PATCH v2 1/10] cpufreq: Clean up default and fallback governor setup Rafael J. Wysocki
2016-02-10 5:15 ` Gautham R Shenoy
2016-02-10 5:48 ` Gautham R Shenoy
2016-02-05 2:14 ` [PATCH v2 2/10] cpufreq: governor: Use common mutex for dbs_data protection Rafael J. Wysocki
2016-02-05 6:53 ` Viresh Kumar
2016-02-05 22:59 ` Rafael J. Wysocki
2016-02-07 9:31 ` Viresh Kumar
2016-02-07 14:33 ` Rafael J. Wysocki
2016-02-05 2:15 ` [PATCH v2 3/10] cpufreq: governor: Avoid passing dbs_data pointers around unnecessarily Rafael J. Wysocki
2016-02-05 7:09 ` Viresh Kumar
2016-02-05 2:16 ` [PATCH v2 4/10] cpufreq: governor: Put governor structure into common_dbs_data Rafael J. Wysocki
2016-02-05 7:13 ` Viresh Kumar
2016-02-05 2:17 ` [PATCH v2 5/10] cpufreq: governor: Rename some data types and variables Rafael J. Wysocki
2016-02-05 7:17 ` Viresh Kumar
2016-02-05 2:18 ` [PATCH v2 6/10] cpufreq: governor: Rework cpufreq_governor_dbs() Rafael J. Wysocki
2016-02-05 8:14 ` Viresh Kumar
2016-02-05 2:19 ` [PATCH v2 7/10] cpufreq: governor: Drop the gov pointer from struct dbs_data Rafael J. Wysocki
2016-02-05 8:28 ` Viresh Kumar
2016-02-05 2:20 ` [PATCH v2 8/10] cpufreq: governor: Rename cpu_common_dbs_info to policy_dbs_info Rafael J. Wysocki
2016-02-05 8:34 ` Viresh Kumar
2016-02-05 22:50 ` Rafael J. Wysocki
2016-02-06 12:48 ` [PATCH v3 " Rafael J. Wysocki
2016-02-07 9:31 ` Viresh Kumar
2016-02-05 2:21 ` [PATCH v2 9/10] cpufreq: governor: Rearrange governor data structures Rafael J. Wysocki
2016-02-05 9:13 ` Viresh Kumar
2016-02-05 22:47 ` Rafael J. Wysocki
2016-02-07 9:29 ` Viresh Kumar
2016-02-07 14:34 ` Rafael J. Wysocki
2016-02-05 2:21 ` [PATCH v2 10/10] cpufreq: governor: Drop cpu argument from dbs_check_cpu() Rafael J. Wysocki
2016-02-05 9:15 ` Viresh Kumar
2016-02-06 12:50 ` [PATCH v3 " Rafael J. Wysocki
2016-02-07 9:32 ` Viresh Kumar
2016-02-06 12:44 ` [PATCH v2 0/10] cpufreq: governor: ondemand/conservative data structures rework Rafael J. Wysocki
2016-02-07 15:22 ` [PATCH 0/3] cpufreq: governor: Data structure rearrangement Rafael J. Wysocki
2016-02-07 15:23 ` [PATCH 1/3] cpufreq: governor: Simplify cpufreq_governor_limits() Rafael J. Wysocki
2016-02-07 15:40 ` Viresh Kumar
2016-02-08 0:59 ` Rafael J. Wysocki
2016-02-07 15:24 ` [PATCH 2/3] cpufreq: governor: Rearrange governor data structures Rafael J. Wysocki
2016-02-07 15:45 ` Viresh Kumar
2016-02-07 15:54 ` Viresh Kumar
2016-02-07 15:55 ` Viresh Kumar
2016-02-07 15:25 ` [PATCH 3/3] cpufreq: governor: Symmetrize cpu_dbs_info initialization and cleanup Rafael J. Wysocki
2016-02-07 15:52 ` Viresh Kumar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160205025948.GE3068@vireshk \
--to=viresh.kumar@linaro.org \
--cc=juri.lelli@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=rafael@kernel.org \
--cc=rjw@rjwysocki.net \
--cc=skannan@codeaurora.org \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=steve.muckle@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).