From: Tejun Heo <tj@kernel.org>
To: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>,
pjt@google.com, paul.mckenney@linaro.org, tglx@linutronix.de,
suresh.b.siddha@intel.com, venki@google.com, mingo@redhat.com,
peterz@infradead.org, rostedt@goodmis.org,
Arvind.Chauhan@arm.com, linaro-dev@lists.linaro.org,
patches@linaro.org, pdsw-power-team@arm.com,
linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org
Subject: Re: [PATCH V2 Resend 3/4] workqueue: Schedule work on non-idle cpu instead of current one
Date: Fri, 4 Jan 2013 10:09:28 -0500 [thread overview]
Message-ID: <20130104150928.GC15633@mtj.dyndns.org> (raw)
In-Reply-To: <CAKohpokBMhto6xzp3MPGRJqstEkkC=kzgrKNhVXzi1Y8GejkXg@mail.gmail.com>
Hello, Viresh.
On Fri, Jan 04, 2013 at 04:41:47PM +0530, Viresh Kumar wrote:
> I got a list of files where cpu/processor_id strings are found, which
> may break with
> this patch (still can't guarantee as i don't have knowledge of these drivers)...
...
> I am not sure what to do now :) , can you assist ?
I don't know either. Changing behavior subtly like this is hard. I
usually try to spot some problem cases and try to identify patterns
there. Once you identify a few of them, understanding and detecting
other problem cases get a lot easier. In this case, maybe there are
too many places to audit and the problems are too subtle, and, if we
*have* to do it, the only thing we can do is implementing a debug
option to make such problems more visible - say, always schedule to a
different cpu on queue_work().
That said, at this point, the patchset doesn't seem all that
convincing to me and if I'm comprehending responses from others
correctly that seems to be the consensus. It might be a better
approach to identify the specific offending workqueue users and make
them handle the situation more intelligently than trying to impose the
behavior on all workqueue users. At any rate, we need way more data
showing this actually helps and if so why.
Thanks.
--
tejun
next prev parent reply other threads:[~2013-01-04 15:09 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-06 10:34 [PATCH V2 Resend 0/4] Create sched_select_cpu() and use it for workqueues and timers Viresh Kumar
2012-11-06 10:38 ` [PATCH V2 Resend 3/4] workqueue: Schedule work on non-idle cpu instead of current one Viresh Kumar
[not found] ` <ad25c731a6ca1bd1269555245952f05c856a9759.1352196505.git.viresh.kumar-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
2012-11-26 17:15 ` Tejun Heo
2012-11-27 5:19 ` Viresh Kumar
2012-11-27 12:54 ` Vincent Guittot
[not found] ` <CAKohpom9=K976NmyFWtXx2UQyRcLAqsoGw-5AaqXRpSL6uxP_g-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-01-04 11:11 ` Viresh Kumar
2013-01-04 15:09 ` Tejun Heo [this message]
2013-01-07 9:58 ` Viresh Kumar
2013-01-07 13:28 ` Steven Rostedt
2013-01-07 17:59 ` Viresh Kumar
2013-01-07 22:29 ` Steven Rostedt
2013-01-08 4:03 ` Viresh Kumar
2013-01-07 15:04 ` Tejun Heo
2013-01-07 15:40 ` Amit Kucheria
2013-01-07 18:07 ` Viresh Kumar
2013-01-09 18:49 ` Tejun Heo
2013-01-10 5:04 ` Viresh Kumar
2012-11-27 13:26 ` Steven Rostedt
2012-11-27 13:48 ` Viresh Kumar
2012-11-27 13:59 ` Steven Rostedt
2012-11-27 14:55 ` Vincent Guittot
2012-11-27 15:04 ` Steven Rostedt
2012-11-27 15:35 ` Vincent Guittot
[not found] ` <cover.1352196505.git.viresh.kumar-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
2012-11-06 10:38 ` [PATCH V2 Resend 1/4] sched: Create sched_select_cpu() to give preferred CPU for power saving Viresh Kumar
2012-11-06 10:38 ` [PATCH V2 Resend 2/4] timer: hrtimer: Don't check idle_cpu() before calling get_nohz_timer_target() Viresh Kumar
2012-11-06 10:38 ` [PATCH V2 Resend 4/4] timer: Migrate running timer Viresh Kumar
2012-11-27 13:47 ` Steven Rostedt
2013-03-20 15:13 ` Viresh Kumar
2013-04-09 14:52 ` Viresh Kumar
2013-04-24 11:22 ` Viresh Kumar
2013-05-13 9:19 ` Viresh Kumar
2013-05-13 10:35 ` Thomas Gleixner
2013-05-22 8:34 ` Viresh Kumar
2013-05-22 9:06 ` Peter Zijlstra
2013-05-22 9:23 ` Viresh Kumar
2013-05-31 10:49 ` Viresh Kumar
2013-06-18 4:51 ` Viresh Kumar
2013-07-24 9:17 ` Viresh Kumar
2013-08-07 9:55 ` Viresh Kumar
2013-10-04 12:39 ` Sebastian Andrzej Siewior
2013-10-23 5:55 ` Viresh Kumar
2012-11-26 15:00 ` [PATCH V2 Resend 0/4] Create sched_select_cpu() and use it for workqueues and timers Viresh Kumar
2012-11-26 16:40 ` Steven Rostedt
[not found] ` <1353948027.6276.38.camel-f9ZlEuEWxVcJvu8Pb33WZ0EMvNT87kid@public.gmane.org>
2012-11-26 17:03 ` Paul E. McKenney
2012-11-26 17:35 ` Steven Rostedt
[not found] ` <1353951352.6276.43.camel-f9ZlEuEWxVcJvu8Pb33WZ0EMvNT87kid@public.gmane.org>
2012-11-26 19:03 ` Paul E. McKenney
2012-11-26 19:17 ` Steven Rostedt
2012-11-27 6:25 ` 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=20130104150928.GC15633@mtj.dyndns.org \
--to=tj@kernel.org \
--cc=Arvind.Chauhan@arm.com \
--cc=linaro-dev@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=patches@linaro.org \
--cc=paul.mckenney@linaro.org \
--cc=pdsw-power-team@arm.com \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=rostedt@goodmis.org \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
--cc=venki@google.com \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@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).