From mboxrd@z Thu Jan  1 00:00:00 1970
From: Peter Clifton <pcjc2@cam.ac.uk>
Subject: Re: High-resolution intel_gpu_top output
Date: Sat, 30 Oct 2010 00:17:41 +0100
Message-ID: <1288394261.1156.17.camel@pcjc2lap>
References: <1288373504.27413.8.camel@pcjc2lap>
	<1288393688.1156.15.camel@pcjc2lap>
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 ppsw-51.csi.cam.ac.uk (ppsw-51.csi.cam.ac.uk [131.111.8.151])
	by gabe.freedesktop.org (Postfix) with ESMTP id D44E39E910
	for <intel-gfx@lists.freedesktop.org>;
	Fri, 29 Oct 2010 16:17:42 -0700 (PDT)
Received: from cpc1-cmbg11-0-0-cust238.5-4.cable.virginmedia.com
	([86.22.72.239]:50546 helo=[192.168.1.4])
	by ppsw-51.csi.cam.ac.uk (smtp.hermes.cam.ac.uk [131.111.8.158]:465)
	with esmtpsa (LOGIN:pcjc2) (SSLv3:DHE-RSA-CAMELLIA256-SHA:256)
	id 1PByCc-00057q-WL (Exim 4.72) for intel-gfx@lists.freedesktop.org
	(return-path <pcjc2@cam.ac.uk>); Sat, 30 Oct 2010 00:17:42 +0100
In-Reply-To: <1288393688.1156.15.camel@pcjc2lap>
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: "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org>
List-Id: intel-gfx@lists.freedesktop.org

On Sat, 2010-10-30 at 00:08 +0100, Peter Clifton wrote:

> It is pretty hard to make out, but I _think_ I'm seeing the GPU starved
> at certain points. It looks like the URB busy flag at 100% for these
> portions of the profile. I've highlighted them in RED as the bad stalls
> I'm worried about.

I misread my graphs... URB is the green portion very bottom trace.. it
is the RENDER CACHE which appears to stall things when it hits 100%
usage. I wish I had a better model of the hardware internals, and
exactly how these flags relate to it.

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)