Subject: Re: volano ~30% regression with 2.6.33-rc1 & -rc2
From: Peter Zijlstra <peterz@infradead.org>
To: Mike Galbraith <efault@gmx.de>
Cc: Arjan van de Ven <arjan@infradead.org>, Lin Ming <ming.m.lin@intel.com>,
       lkml <linux-kernel@vger.kernel.org>,
       "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
In-Reply-To: <1262610927.9734.64.camel@marge.simson.net>
References: <1262592958.22471.104.camel@minggr.sh.intel.com>
	 <20100104044024.5390ff1c@infradead.org>
	 <1262609854.9734.56.camel@marge.simson.net>
	 <1262610123.6408.120.camel@laptop>
	 <1262610927.9734.64.camel@marge.simson.net>
Content-Type: text/plain; charset="UTF-8"
Date: Mon, 04 Jan 2010 14:26:48 +0100
Message-ID: <1262611608.6408.126.camel@laptop>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1816
Lines: 38

On Mon, 2010-01-04 at 14:15 +0100, Mike Galbraith wrote:
> On Mon, 2010-01-04 at 14:02 +0100, Peter Zijlstra wrote:
> > On Mon, 2010-01-04 at 13:57 +0100, Mike Galbraith wrote:
> > > On Mon, 2010-01-04 at 04:40 -0800, Arjan van de Ven wrote:
> > > > On Mon, 04 Jan 2010 16:15:58 +0800
> > > > Lin Ming <ming.m.lin@intel.com> wrote:
> > > > 
> > > > > Mike & Peter,
> > > > > 
> > > > > Compared with 2.6.32, volano has ~30% regression with 2.6.33-rc1 &
> > > > > -rc2. Testing machine: Tigerton Xeon, 16cpus(4P/4Core), 16G memory
> > > > 
> > > > did this show up only on this cpu?
> > > > (since this is a multi-core-without-shared-cache cpu, it could be that
> > > > we get the topology wrong and think cores share cache where they don't)
> > > 
> > > My fault for using PREFER_SIBLING I guess.  However, I do wonder why in
> > > the heck we set that at the CPU domain level.  Siblings lie northward.
> > 
> > Ah, PREFER_SIBLING means prefer sibling domain, not sibling thread. Its
> > set at the CPU (really socket) level so make tasks spread over sockets
> > first, so that there is no competition for the socket wide resources.
> 
> WRT the regression, would you prefer only the sched_fair.c hunk, and
> maybe plunking the topology hunk in sched_devel, or both lines in one
> patch, since ramp-up gain remains unrealized half of the time on Nehalem
> and ilk.

Both bits seem sane I guess, you change SD_SIBLING_INIT(), right?
Threads really do share package resources so it makes sense to set it.

I guess its back to poking at nehalem to see what makes it tick..

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/