linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@kernel.org>,
	"Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Paul McKenney <paulmck@linux.vnet.ibm.com>,
	Mel Gorman <mgorman@suse.de>, Rik van Riel <riel@redhat.com>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Steven Rostedt <rostedt@goodmis.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/6] Optimize the cpu hotplug locking -v2
Date: Thu, 10 Oct 2013 18:54:12 +0200	[thread overview]
Message-ID: <20131010165412.GA16553@redhat.com> (raw)
In-Reply-To: <20131010090044.7f12ddaf.akpm@linux-foundation.org>

On 10/10, Andrew Morton wrote:
>
> On Thu, 10 Oct 2013 17:26:12 +0200 Oleg Nesterov <oleg@redhat.com> wrote:
>
> > On 10/10, Ingo Molnar wrote:
> > >
> > > So ... why not make it _really_ cheap, i.e. the read lock costing nothing,
> > > and tie CPU hotplug to freezing all tasks in the system?
> > >
> > > Actual CPU hot unplugging and repluggin is _ridiculously_ rare in a
> > > system, I don't understand how we tolerate _any_ overhead from this utter
> > > slowpath.
> >
> > Well, iirc Srivatsa (cc'ed) pointed out that some systems do cpu_down/up
> > quite often to save the power.
>
> cpu hotremove already uses stop_machine,

And Srivatsa wants to remove it from cpu_down().

> so such an approach shouldn't
> actually worsen things (a lot) for them?

this depends on what this "freezing all tasks" actually means.
I understood it as try_to_freeze_tasks/etc, looks too heavy...

But my only point was, I am not sure we can assume that cpu_down/up
is extremly rare and its cost does not matter.

> use stop_machine() on the add/remove
> (ie, "writer") side and nothing at all on the "reader" side.  Is there
> anything which fundamentally prevents cpu hotplug from doing the same?

Well, then we actually need to park all tasks in system, I guess.
IOW, freezer.

Oleg.


  parent reply	other threads:[~2013-10-10 17:01 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-08 10:25 [PATCH 0/6] Optimize the cpu hotplug locking -v2 Peter Zijlstra
2013-10-08 10:25 ` [PATCH 1/6] hotplug: Optimize {get,put}_online_cpus() Peter Zijlstra
2013-10-08 15:08   ` Rik van Riel
2013-10-10  5:47   ` Andrew Morton
2013-10-10 11:06     ` Oleg Nesterov
2013-10-10 14:55       ` Paul E. McKenney
2013-10-08 10:25 ` [PATCH 2/6] rcu: Create rcu_sync infrastructure Peter Zijlstra
2013-10-08 20:40   ` Jonathan Corbet
2013-10-09 19:52     ` Peter Zijlstra
2013-10-17  2:56   ` Lai Jiangshan
2013-10-17 10:36   ` Srikar Dronamraju
2013-10-08 10:25 ` [PATCH 3/6] hotplug: Optimize cpu_hotplug_{begin,done}() using rcu_sync Peter Zijlstra
2013-10-08 16:28   ` Paul E. McKenney
2013-10-08 10:25 ` [PATCH 4/6] rcusync: Introduce struct rcu_sync_ops Peter Zijlstra
2013-10-08 16:30   ` Paul E. McKenney
2013-10-17  2:07   ` Lai Jiangshan
     [not found]     ` <20131017154228.GL4553@linux.vnet.ibm.com>
2013-10-18  1:23       ` Lai Jiangshan
2013-10-18 12:10         ` Oleg Nesterov
2013-10-20 16:58           ` Paul E. McKenney
2013-10-08 10:25 ` [PATCH 5/6] rcusync: Add the CONFIG_PROVE_RCU checks Peter Zijlstra
2013-10-08 16:30   ` Paul E. McKenney
2013-10-08 10:25 ` [PATCH 6/6] rcusync: Introduce rcu_sync_dtor() Peter Zijlstra
2013-10-08 16:32   ` Paul E. McKenney
2013-10-08 15:27 ` [PATCH 0/6] Optimize the cpu hotplug locking -v2 Oleg Nesterov
2013-10-08 15:38   ` Peter Zijlstra
2013-10-10  5:50 ` Andrew Morton
2013-10-10  6:27   ` Ingo Molnar
2013-10-10  6:34     ` Andrew Morton
2013-10-10  7:27       ` Ingo Molnar
2013-10-10  7:33         ` Andrew Morton
2013-10-10  7:45           ` Ingo Molnar
2013-10-10 12:19   ` Peter Zijlstra
2013-10-10 14:57     ` Ingo Molnar
2013-10-10 15:21       ` Peter Zijlstra
2013-10-10 15:36         ` Oleg Nesterov
2013-10-10 16:50         ` Ingo Molnar
2013-10-10 17:13           ` Paul E. McKenney
2013-10-10 17:35             ` Ingo Molnar
2013-10-10 18:35             ` Peter Zijlstra
2013-10-10 15:26       ` Oleg Nesterov
2013-10-10 16:00         ` Andrew Morton
2013-10-10 16:36           ` Steven Rostedt
2013-10-10 16:43             ` Andrew Morton
2013-10-10 16:53               ` Peter Zijlstra
2013-10-10 17:13                 ` Steven Rostedt
2013-10-10 17:48                   ` Andrew Morton
2013-10-10 18:10                     ` Linus Torvalds
2013-10-10 18:43                       ` Steven Rostedt
2013-10-10 18:50                         ` Peter Zijlstra
2013-10-10 19:15                           ` Paul E. McKenney
2013-10-10 19:00                         ` Linus Torvalds
2013-10-10 18:46                       ` Peter Zijlstra
2013-10-10 18:34                     ` Peter Zijlstra
2013-10-10 18:49                       ` Linus Torvalds
2013-10-10 19:04                         ` Steven Rostedt
2013-10-10 19:16                           ` Linus Torvalds
2013-10-10 19:34                             ` Peter Zijlstra
2013-10-10 19:34                             ` Steven Rostedt
2013-10-11  6:09                             ` Ingo Molnar
2013-10-11 12:38                         ` Peter Zijlstra
2013-10-11 18:25                           ` Oleg Nesterov
2013-10-11 20:48                             ` Peter Zijlstra
2013-10-12 17:06                               ` Oleg Nesterov
2013-10-14  9:05                                 ` Peter Zijlstra
2013-10-14  9:23                                   ` Paul E. McKenney
2013-10-15  1:01                                     ` Paul E. McKenney
2013-10-17 16:49                           ` [tip:sched/core] sched: Remove get_online_cpus() usage tip-bot for Peter Zijlstra
2013-10-10 17:39                 ` [PATCH 0/6] Optimize the cpu hotplug locking -v2 Oleg Nesterov
2013-10-10 16:52           ` Ingo Molnar
2013-10-10 17:44             ` Paul E. McKenney
2013-10-10 16:54           ` Oleg Nesterov [this message]
2013-10-10 19:04             ` Srivatsa S. Bhat
2013-10-10 18:52         ` Srivatsa S. Bhat
2013-10-11 12:14 George Spelvin
2013-10-11 12:45 ` Steven Rostedt
2013-10-11 13:51   ` George Spelvin

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=20131010165412.GA16553@redhat.com \
    --to=oleg@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=riel@redhat.com \
    --cc=rostedt@goodmis.org \
    --cc=srikar@linux.vnet.ibm.com \
    --cc=srivatsa.bhat@linux.vnet.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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).