From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ted Phelps <phelps@gnusto.com>
Subject: Re: [PATCH 1/2] drm/i915: add module parameter compiler
	hints
Date: Thu, 07 Jul 2011 10:48:41 +1000
Message-ID: <8844.1309999721@orpheus.gnusto.com>
References: <1309990493-27160-1-git-send-email-ben@bwidawsk.net>
	<yunr563vwp9.fsf@aiko.keithp.com>
	<20110706231601.GA1556@seagal.home>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org>
Received: from laika.gnusto.com (laika.gnusto.com [207.44.178.75])
	by gabe.freedesktop.org (Postfix) with ESMTP id 7A2239E74F
	for <intel-gfx@lists.freedesktop.org>;
	Wed,  6 Jul 2011 17:48:50 -0700 (PDT)
In-reply-to: Your message of Wed, 06 Jul 2011 16:16:01 MST.
	<20110706231601.GA1556@seagal.home>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/intel-gfx>
List-Post: <mailto:intel-gfx@lists.freedesktop.org>
List-Help: <mailto:intel-gfx-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=subscribe>
Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
To: Ben Widawsky <ben@bwidawsk.net>, Keith Packard <keithp@keithp.com>
Cc: intel-gfx@lists.freedesktop.org, Julien Cristau <jcristau@debian.org>
List-Id: intel-gfx@lists.freedesktop.org

Ben Widawsky writes:
> On Wed, Jul 06, 2011 at 04:03:14PM -0700, Keith Packard wrote:
> > On Wed,  6 Jul 2011 15:14:52 -0700, Ben Widawsky <ben@bwidawsk.net> wrote:
> >
> > > -static int i915_modeset =3D -1;
> > > +static int i915_modeset __read_mostly =3D -1;
> >
> > What effect does this have? Performance? Code size? More warnings?
> 
> __read_mostly attempts to put variables in a cachelines to make it more
> SMP friendly. I'm not sure if there is a performance impact as a result,
> but made this patch based on feedback from Chris. I've seen no
> regressions on Nexuiz.

A link to an informative LKML discussion on this topic is here:

    http://lkml.org/lkml/2007/12/17/43

-Ted