From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Subject: Re: [PATCH] drm/i915: Use PIPE_CONTROL for flushing on
 gen6+.
Date: Mon, 3 Oct 2011 16:16:18 -0700
Message-ID: <20111003161618.0c5fa610@jbarnes-desktop>
References: <1317063563-1526-1-git-send-email-kenneth@whitecape.org>
	<20111003130016.61bcc8c0@jbarnes-desktop>
	<87aa9hk7fy.fsf@eliezer.anholt.net>
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 oproxy4-pub.bluehost.com (oproxy4-pub.bluehost.com
	[69.89.21.11])
	by gabe.freedesktop.org (Postfix) with SMTP id F3A65A1086
	for <intel-gfx@lists.freedesktop.org>;
	Mon,  3 Oct 2011 16:16:29 -0700 (PDT)
In-Reply-To: <87aa9hk7fy.fsf@eliezer.anholt.net>
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: Eric Anholt <eric@anholt.net>
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On Mon, 03 Oct 2011 15:59:29 -0700
Eric Anholt <eric@anholt.net> wrote:

> On Mon, 3 Oct 2011 13:00:16 -0700, Jesse Barnes <jbarnes@virtuousgeek.org> wrote:
> > On Mon, 26 Sep 2011 11:59:23 -0700
> > Kenneth Graunke <kenneth@whitecape.org> wrote:
> > 
> > > +	/* Just flush everything for now */
> > > +	flags |= PIPE_CONTROL_WC_FLUSH;
> > > +	flags |= PIPE_CONTROL_IS_FLUSH;
> > > +	flags |= PIPE_CONTROL_TC_FLUSH;
> > > +	flags |= PIPE_CONTROL_DEPTH_FLUSH;
> > > +	flags |= PIPE_CONTROL_VFC;
> > 
> > This is the only bit I'd like to see changed.  While we still have the
> > domain tracking code we may as well try to honor it and limit our
> > flushing here like we do with MI_FLUSH.
> > 
> > Unless someone has a "remove all domain tracking" patch already posted
> > that is. :)
> 
> I don't think we "might as well try to honor it".  Working out the
> workarounds for various combinations is difficult to do even for a fixed
> set of bits.  Let's not make the workarounds more complicated by varying
> them, when experiments showed no evidence for removing bits improving
> performance.

Oh if you have experiments showing that the individual bits provide no
benefit that's fine.

-- 
Jesse Barnes, Intel Open Source Technology Center