From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: vatsa@linux.vnet.ibm.com
Cc: Mike Galbraith <efault@gmx.de>,
"Zhang, Yanmin" <yanmin_zhang@linux.intel.com>,
Dhaval Giani <dhaval@linux.vnet.ibm.com>,
LKML <linux-kernel@vger.kernel.org>, Ingo Molnar <mingo@elte.hu>,
Aneesh Kumar KV <aneesh.kumar@linux.vnet.ibm.com>
Subject: Re: volanoMark regression with kernel 2.6.26-rc1
Date: Thu, 15 May 2008 10:41:38 +0200 [thread overview]
Message-ID: <1210840898.21616.27.camel@twins> (raw)
In-Reply-To: <20080515082033.GA14823@linux.vnet.ibm.com>
On Thu, 2008-05-15 at 13:50 +0530, Srivatsa Vaddagiri wrote:
> On Wed, May 14, 2008 at 05:12:47PM +0200, Peter Zijlstra wrote:
> > A few other things I found that make a significant difference:
>
> One more observation: access to aggregate()->rq_weight etc arent
> correctly synchronized i.e while a cpu is doing a aggregate_walk_tree()
> in a domain, and thus possibly modifying rq_weight, load etc, other cpus could
> be concurrently accessing the same data. As a result, its possible to
> see inconsistent rq_weight, load, task_weight combination?
Yes - and that should not be too big an issue as long as we can deal
with it.
Any number we'll put to it will be based on a snapshot of the state so
we're wrong no matter what we do. The trick is trying to keep sane.
My current stack on top of sched-devel:
http://programming.kicks-ass.net/kernel-patches/sched-smp-group-fixes/
I've found that:
http://programming.kicks-ass.net/kernel-patches/sched-smp-group-fixes/sched-agg-update-move_tasks.patch
was sufficient to deal with all the anomalities I've found so far.
next prev parent reply other threads:[~2008-05-15 8:41 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-05-06 2:06 volanoMark regression with kernel 2.6.26-rc1 Zhang, Yanmin
2008-05-06 5:41 ` Zhang, Yanmin
2008-05-06 11:52 ` Dhaval Giani
2008-05-07 17:33 ` Dhaval Giani
2008-05-08 5:18 ` Zhang, Yanmin
2008-05-08 5:32 ` Dhaval Giani
2008-05-08 5:40 ` Dhaval Giani
2008-05-08 5:53 ` Zhang, Yanmin
2008-05-08 6:04 ` Dhaval Giani
2008-05-08 6:11 ` Srivatsa Vaddagiri
2008-05-09 15:52 ` Srivatsa Vaddagiri
2008-05-09 15:54 ` Srivatsa Vaddagiri
2008-05-12 1:39 ` Zhang, Yanmin
2008-05-12 2:04 ` Dhaval Giani
2008-05-12 2:37 ` Srivatsa Vaddagiri
2008-05-12 3:33 ` Zhang, Yanmin
2008-05-12 4:52 ` Srivatsa Vaddagiri
2008-05-12 5:02 ` Zhang, Yanmin
2008-05-12 5:43 ` Zhang, Yanmin
2008-05-12 9:04 ` Mike Galbraith
2008-05-12 9:20 ` Peter Zijlstra
2008-05-14 9:22 ` Zhang, Yanmin
2008-05-14 13:44 ` Srivatsa Vaddagiri
2008-05-14 14:50 ` Mike Galbraith
2008-05-14 15:12 ` Peter Zijlstra
2008-05-15 8:20 ` Srivatsa Vaddagiri
2008-05-15 8:41 ` Peter Zijlstra [this message]
2008-05-15 17:10 ` Srivatsa Vaddagiri
2008-05-07 7:04 ` Andrew Morton
2008-05-07 9:17 ` Ingo Molnar
2008-05-07 9:33 ` Zhang, Yanmin
2008-05-07 17:34 ` Peter Zijlstra
2008-05-07 18:58 ` Peter Zijlstra
2008-05-08 6:07 ` Zhang, Yanmin
2008-05-08 5:20 ` Zhang, Yanmin
2008-05-08 5:34 ` Dhaval Giani
2008-05-08 6:43 ` Peter Zijlstra
2008-05-07 17:42 ` Dhaval Giani
2008-05-08 5:21 ` Zhang, Yanmin
2008-05-08 5:39 ` Dhaval Giani
2008-05-08 6:03 ` Zhang, Yanmin
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=1210840898.21616.27.camel@twins \
--to=a.p.zijlstra@chello.nl \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=dhaval@linux.vnet.ibm.com \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=vatsa@linux.vnet.ibm.com \
--cc=yanmin_zhang@linux.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox