From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>, Ingo Molnar <mingo@elte.hu>,
Andrew Morton <akpm@osdl.org>,
dipankar@in.ibm.com, Gautham Shenoy <ego@linux.vnet.ibm.com>,
linux-kernel@vger.kernel.org
Subject: Re: Fw: Re: [mm PATCH 4/6] RCU: (now) CPU hotplug
Date: Sat, 3 Feb 2007 16:31:22 -0800 [thread overview]
Message-ID: <20070204003122.GD5647@linux.vnet.ibm.com> (raw)
In-Reply-To: <200702032327.25163.rjw@sisk.pl>
On Sat, Feb 03, 2007 at 11:27:24PM +0100, Rafael J. Wysocki wrote:
> On Saturday, 3 February 2007 01:01, Pavel Machek wrote:
> > Hi!
> >
> > > > > > static int _cpu_down(unsigned int cpu)
> > > > > > {
> > > > > > int err;
> > > > > > struct task_struct *p;
> > > > > > cpumask_t old_allowed, tmp;
> > > > > >
> > > > > > if (num_online_cpus() == 1)
> > > > > > return -EBUSY;
> > > > > >
> > > > > > if (!cpu_online(cpu))
> > > > > > return -EINVAL;
> > > > > >
> > > > > > if (freeze_processes()) {
> > > > > > err = -EBUSY;
> > > > > > goto out_freeze_notify_failed;
> > > > > > }
> > > > > > err = raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE,
> > > > > > (void *)(long)cpu);
> > > > >
> > > > > yeah. This all looks so nice that i almost cannot believe it's true :-)
> > > >
> > > > Well, it turns out that maybe it is in fact untrue. :-/
> > > >
> > > > I need to look at all uses of PF_NOFREEZE -- as I understand the
> > > > code, processes marked PF_NOFREEZE will continue running, potentially
> > > > interfering with the hotplug operation. :-(
> > > >
> > > > I will pass my findings on to this list.
> > >
> > > Well, I did it some time ago, although not very thoroughly.
> > >
> > > AFAICS there are not so many, but one that stands out is the worker threads.
> > > We needed two of them to actually go to sleep, so now it's possible to create
> > > a "freezeable workqueue" the worker thread of which will not set PF_NOFREEZE,
> > > but currently this is only used by XFS.
> >
> > We should slowly move as workqueues to freezeable ones... Having too
> > much stuff NOFREEZE is evil, even for swsusp.
>
> On the other hand, some of the workqueues may be necessary for saving the image
> (still, I have no examples ;-)).
In any case, getting CPU hotplug working is probably first priority.
That said, I am testing a patch that (hopefully) gets rid of the
RCU-boost and rcutorture NOFREEZE tasks. Every little bit will help.
Thanx, Paul
next prev parent reply other threads:[~2007-02-04 0:31 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20070124011519.GG1613@linux.vnet.ibm.com>
[not found] ` <20070124090111.GC27221@in.ibm.com>
[not found] ` <20070124161559.GA1762@linux.vnet.ibm.com>
[not found] ` <20070124210645.GA19650@in.ibm.com>
2007-01-26 19:11 ` Fw: Re: [mm PATCH 4/6] RCU: (now) CPU hotplug Dipankar Sarma
2007-01-26 19:28 ` Andrew Morton
2007-01-26 19:46 ` Dipankar Sarma
2007-01-26 20:17 ` Andrew Morton
2007-01-26 20:44 ` Dipankar Sarma
2007-01-26 21:29 ` Andrew Morton
2007-01-28 22:47 ` Paul E. McKenney
2007-01-28 23:30 ` Andrew Morton
2007-01-29 2:40 ` Paul E. McKenney
2007-01-29 19:12 ` Ingo Molnar
2007-01-30 2:45 ` Paul E. McKenney
2007-01-30 7:33 ` Ingo Molnar
2007-01-30 16:02 ` Paul E. McKenney
2007-01-30 16:44 ` Rafael J. Wysocki
2007-01-30 18:27 ` Andrew Morton
2007-01-30 19:49 ` Paul E. McKenney
2007-01-31 23:10 ` Paul E. McKenney
2007-02-03 0:17 ` Pavel Machek
2007-02-04 4:39 ` Paul E. McKenney
2007-02-04 11:08 ` Rafael J. Wysocki
2007-02-04 12:53 ` Pavel Machek
2007-02-04 13:46 ` Rafael J. Wysocki
2007-02-04 13:50 ` Pavel Machek
2007-02-04 13:59 ` Rafael J. Wysocki
2007-02-03 0:01 ` Pavel Machek
2007-02-03 22:27 ` Rafael J. Wysocki
2007-02-04 0:31 ` Paul E. McKenney [this message]
2007-01-30 14:02 ` Gautham R Shenoy
2007-01-30 16:47 ` Paul E. McKenney
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=20070204003122.GD5647@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=akpm@osdl.org \
--cc=dipankar@in.ibm.com \
--cc=ego@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=pavel@ucw.cz \
--cc=rjw@sisk.pl \
/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