All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dave Jones <davej@redhat.com>
To: Suresh Siddha <suresh.b.siddha@intel.com>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Linux Kernel <linux-kernel@vger.kernel.org>,
	Ingo Molnar <mingo@elte.hu>
Subject: Re: sched domains oddness.
Date: Fri, 31 Oct 2008 13:37:37 -0400	[thread overview]
Message-ID: <20081031173737.GA11443@redhat.com> (raw)
In-Reply-To: <20081031171721.GA8837@redhat.com>

On Fri, Oct 31, 2008 at 01:17:21PM -0400, Dave Jones wrote:
 > On Fri, Oct 31, 2008 at 10:09:29AM -0700, Suresh Siddha wrote:
 > 
 >  > > looks like someone is triggering rebuild_sched_domains(), is something
 >  > > poking cpusetfs files or flipping between sched_mc settings?
 >  > 
 >  > I remember someone mentioning that some distro's started setting
 >  > sched_mc_power_savings to '1' by default during boot. On a dual-core
 >  > laptop, this will not give any advantage.
 >  > 
 >  > I have to fix the code to not export this tunable, when we have only
 >  > socket in the system.
 >  > 
 >  > Dave, Is your distro also setting this tunable blindly during boot :(
 > 
 > (13:15:25:davej@vaio:~)$ cat /sys/devices/system/cpu/sched_mc_power_savings
 > 0
 > 
 > So no, unless something set it to 1, and then back to 0.
 > 
 > A grep of etc shows up nothing in initscripts.  Does hal or something
 > play with this?

Even curiouser.. When I unplug AC and replug it, it happens again, but slightly
differently..

CPU0 attaching NULL sched-domain.
CPU1 attaching NULL sched-domain.
CPU0 attaching sched-domain:
 domain 0: span 0-1 level MC
  groups: 0 1
  domain 1: span 0-1 level CPU
   groups: 0-1
   domain 2: span 0-1 level NODE
    groups: 0-1
CPU1 attaching sched-domain:
 domain 0: span 0-1 level MC
  groups: 1 0
  domain 1: span 0-1 level CPU
   groups: 0-1
   domain 2: span 0-1 level NODE
    groups: 0-1
CPU0 attaching NULL sched-domain.
CPU1 attaching NULL sched-domain.
CPU0 attaching sched-domain:
 domain 0: span 0-1 level MC
  groups: 0 1
  domain 1: span 0-1 level NODE
   groups: 0-1
CPU1 attaching sched-domain:
 domain 0: span 0-1 level MC
  groups: 1 0
  domain 1: span 0-1 level NODE
   groups: 0-1


Note how the CPU level doesn't show up in the 2nd case.

This still doesn't explain the flip-flop I saw just from booting,
as that was on AC the whole time.

I grepped hal and gnome-power-manager, and didn't see anything
touching sched_mc

	Dave

-- 
http://www.codemonkey.org.uk

  reply	other threads:[~2008-10-31 17:38 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-10-31 16:24 sched domains oddness Dave Jones
2008-10-31 16:51 ` Peter Zijlstra
2008-10-31 17:09   ` Suresh Siddha
2008-10-31 17:17     ` Dave Jones
2008-10-31 17:37       ` Dave Jones [this message]
2008-10-31 17:58         ` Suresh Siddha
2008-10-31 19:08     ` Dave Jones
2008-10-31 17:19   ` Dave Jones

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=20081031173737.GA11443@redhat.com \
    --to=davej@redhat.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=suresh.b.siddha@intel.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.