All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: LKML <linux-kernel@vger.kernel.org>, Ingo Molnar <mingo@elte.hu>,
	ktkhai@parallels.com
Subject: Re: sched_setscheduler() vs idle_balance() race
Date: Mon, 1 Jun 2015 10:14:29 +0200	[thread overview]
Message-ID: <20150601081429.GP19282@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <1432925336.8727.29.camel@gmail.com>

On Fri, May 29, 2015 at 08:48:56PM +0200, Mike Galbraith wrote:
> 
> P.S. intel_idle is not all that wonderful on this box.
> 
> -   78.31%  [kernel]       [k] _raw_spin_lock                                                                                                                                                                                              ▒
>    - _raw_spin_lock                                                                                                                                                                                                                        ▒
>       - 94.91% tick_broadcast_oneshot_control                                                                                                                                                                                              ▒

Your DL980 G7 has E7-4800 parts in, right? Which if Wikipedia is
correct, resolves to a Nehalem-EX.

Now the NHM-EX has a fun 'feature' that for (some?) idle states the
local timer stops, so we have to revert back to a global broadcast
timer.

Now go count the number of cpus on your box and then imagine a global
spinlock, oh wait, you already found it ^ :-)

  reply	other threads:[~2015-06-01  8:14 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-28  7:43 sched_setscheduler() vs idle_balance() race Mike Galbraith
2015-05-28 11:51 ` Peter Zijlstra
2015-05-28 12:04   ` Mike Galbraith
2015-05-28 12:06     ` Peter Zijlstra
2015-05-28 12:32       ` Mike Galbraith
2015-05-28 13:53 ` Peter Zijlstra
2015-05-28 14:54   ` Mike Galbraith
2015-05-28 15:24     ` Peter Zijlstra
2015-05-29 18:30       ` Mike Galbraith
2015-05-29 18:48         ` Mike Galbraith
2015-06-01  8:14           ` Peter Zijlstra [this message]
2015-06-01  8:16         ` Peter Zijlstra
2015-06-01 10:00           ` Mike Galbraith
2015-05-28 16:59   ` Kirill Tkhai
2015-05-30 13:08   ` Mike Galbraith
2015-05-31  6:39     ` Mike Galbraith
2015-06-01  8:24       ` Peter Zijlstra
2015-06-01  8:19     ` Peter Zijlstra

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=20150601081429.GP19282@twins.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=ktkhai@parallels.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=umgwanakikbuti@gmail.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.