From: Ingo Molnar <mingo@elte.hu>
To: Gautham R Shenoy <ego@in.ibm.com>
Cc: akpm@linux-foundation.org, paulmck@us.ibm.com,
torvalds@linux-foundation.org, linux-kernel@vger.kernel.org,
vatsa@in.ibm.com, Oleg Nesterov <oleg@tv-sign.ru>,
"Rafael J. Wysocki" <rjw@sisk.pl>,
dipankar@in.ibm.com, dino@in.ibm.com,
masami.hiramatsu.pt@hitachi.com
Subject: Re: [RFC] Cpu-hotplug: Using the Process Freezer (try2)
Date: Mon, 2 Apr 2007 08:16:12 +0200 [thread overview]
Message-ID: <20070402061612.GA7072@elte.hu> (raw)
In-Reply-To: <20070402053457.GA9076@in.ibm.com>
* Gautham R Shenoy <ego@in.ibm.com> wrote:
> Hello Everybody,
>
> This is another attempt towards process-freezer based cpu-hotplug.
> This patchset covers just about everything that was discussed on the
> LKML with respect to the freezer-based cpu-hotplug.
wow - you have made really nice progress!
> I believe that the reasons for freezer failing as N increases are :
> - 'make -jN' keeps forking new tasks every now and then, thereby resulting
> in a never-ending catching up game in the do_while loop inside
> try_to_freeze_tasks (kernel/power/process.c)
hm, shouldnt the make be frozen immediately?
doesnt the 'please freeze ASAP' flag get propagated to all tasks,
immediately? After that point any cloning activity should duplicate that
flag too, resulting in any new child freezing immediately too.
> Instead of waiting for all the tasks to call try_to_freeze in the
> above mentioned do_while loop, I wonder if we can put some hooks in
> sched.c so asto not schedule the task marked PF_FREEZING/PF_FROZEN.
we could definitely do that - but i think it should be unnecessary: if
we mark all tasks as PF_FREEZING atomically, that should result in
_every_ task immediately dropping dead (once they get back from
TASK_UNINTERRUPTIBLE). No excuses. If there's some longer delay then
that can only be explained by some new cloned task/thread slipping
through the net somehow. (i.e. the PF_FREEZING flag not being duplicated
across fork?)
i'm wondering about how TASK_UNINTERRUPTIBLE tasks are handled by the
freezer: are they assumed frozen immediately, or do we wait until they
notice their PF_FREEZING and go into try_to_freeze()? I'd expect
TASK_UNINTERRUPTIBLE to be the largest source of latency. (and hence be
the primary source for freezing 'failures')
Ingo
next prev parent reply other threads:[~2007-04-02 6:17 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-02 5:34 [RFC] Cpu-hotplug: Using the Process Freezer (try2) Gautham R Shenoy
2007-04-02 5:37 ` [PATCH 1/8] Enhance process freezer interface for usage beyond software suspend Gautham R Shenoy
2007-04-02 13:56 ` Pavel Machek
2007-04-02 20:48 ` Rafael J. Wysocki
2007-04-02 20:51 ` Pavel Machek
2007-04-06 14:34 ` Rafael J. Wysocki
2007-04-06 22:20 ` Nigel Cunningham
2007-04-07 9:33 ` Rafael J. Wysocki
2007-04-07 9:47 ` Nigel Cunningham
2007-04-09 3:04 ` Gautham R Shenoy
2007-04-03 7:59 ` Gautham R Shenoy
2007-04-05 9:46 ` Oleg Nesterov
2007-04-05 10:59 ` Gautham R Shenoy
2007-04-05 11:30 ` Oleg Nesterov
2007-04-02 5:37 ` [PATCH 2/8] Make process freezer reentrant Gautham R Shenoy
2007-04-05 9:53 ` Oleg Nesterov
2007-04-05 10:19 ` Gautham R Shenoy
2007-04-02 5:38 ` [PATCH 3/8] Use process freezer for cpu-hotplug Gautham R Shenoy
2007-04-05 10:53 ` Oleg Nesterov
2007-04-05 12:14 ` Gautham R Shenoy
2007-04-05 13:34 ` Oleg Nesterov
2007-04-06 17:27 ` Nathan Lynch
2007-04-06 17:34 ` Ingo Molnar
2007-04-06 17:47 ` Nathan Lynch
2007-04-06 22:22 ` Nigel Cunningham
2007-04-14 18:48 ` Pavel Machek
2007-04-02 5:39 ` [PATCH 4/8] Rip out lock_cpu_hotplug() Gautham R Shenoy
2007-04-02 5:40 ` [PATCH 5/8] __cpu_up: use singlethreaded workqueue Gautham R Shenoy
2007-04-05 12:08 ` Oleg Nesterov
2007-04-02 5:41 ` [PATCH 6/8] Make non-singlethreaded workqueues freezeable by default Gautham R Shenoy
2007-04-05 11:57 ` Oleg Nesterov
2007-04-05 20:06 ` Andrew Morton
2007-04-02 5:42 ` [PATCH 7/8] Clean up workqueue.c with respect to the freezer based cpu-hotplug Gautham R Shenoy
2007-04-03 11:47 ` Oleg Nesterov
2007-04-03 13:59 ` Srivatsa Vaddagiri
2007-04-03 15:03 ` Oleg Nesterov
2007-04-03 17:18 ` Srivatsa Vaddagiri
2007-04-04 15:28 ` Oleg Nesterov
2007-04-04 17:49 ` Srivatsa Vaddagiri
2007-04-05 12:20 ` Oleg Nesterov
2007-04-12 2:22 ` Srivatsa Vaddagiri
2007-04-12 10:01 ` Gautham R Shenoy
2007-04-12 16:00 ` Oleg Nesterov
2007-04-13 9:46 ` Gautham R Shenoy
2007-04-02 5:42 ` [PATCH 8/8] Make kernel threads freezeable for cpu-hotplug Gautham R Shenoy
2007-04-02 6:16 ` Ingo Molnar [this message]
2007-04-02 9:28 ` [RFC] Cpu-hotplug: Using the Process Freezer (try2) Srivatsa Vaddagiri
2007-04-02 11:18 ` Ingo Molnar
2007-04-02 12:42 ` Srivatsa Vaddagiri
2007-04-02 14:16 ` Gautham R Shenoy
2007-04-02 18:56 ` Ingo Molnar
2007-04-03 12:56 ` Srivatsa Vaddagiri
2007-04-03 14:15 ` Gautham R Shenoy
2007-04-03 19:25 ` Rafael J. Wysocki
2007-04-04 3:15 ` Srivatsa Vaddagiri
2007-04-04 10:04 ` Ingo Molnar
2007-04-04 10:41 ` Gautham R Shenoy
2007-04-04 11:49 ` Ingo Molnar
2007-04-04 12:24 ` Gautham R Shenoy
2007-04-02 11:19 ` Gautham R Shenoy
2007-04-02 11:27 ` Ingo Molnar
2007-04-02 22:12 ` Rafael J. Wysocki
2007-04-02 13:22 ` Pavel Machek
2007-04-03 12:01 ` Gautham R Shenoy
2007-04-03 19:34 ` Rafael J. Wysocki
2007-04-03 20:24 ` Andrew Morton
2007-04-04 10:06 ` utrace merge Ingo Molnar
2007-04-04 10:36 ` Christoph Hellwig
2007-04-04 18:41 ` Andrew Morton
2007-04-03 14:01 ` [RFC] Cpu-hotplug: Using the Process Freezer (try2) Gautham R Shenoy
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=20070402061612.GA7072@elte.hu \
--to=mingo@elte.hu \
--cc=akpm@linux-foundation.org \
--cc=dino@in.ibm.com \
--cc=dipankar@in.ibm.com \
--cc=ego@in.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=oleg@tv-sign.ru \
--cc=paulmck@us.ibm.com \
--cc=rjw@sisk.pl \
--cc=torvalds@linux-foundation.org \
--cc=vatsa@in.ibm.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.