From mboxrd@z Thu Jan  1 00:00:00 1970
From: Kenneth Graunke <kenneth@whitecape.org>
Subject: Re: [PATCH 1/2] drm/i915: clear GFX_MODE on IVB at init
 time
Date: Mon, 15 Aug 2011 16:59:57 -0700
Message-ID: <4E49B2FD.7010303@whitecape.org>
References: <1313186133-2724-1-git-send-email-jbarnes@virtuousgeek.org>
	<20110812151809.7fdab276@jbarnes-desktop>
	<20110812152832.6c922a17@jbarnes-desktop>
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 homiemail-a62.g.dreamhost.com (caiajhbdcaid.dreamhost.com
	[208.97.132.83])
	by gabe.freedesktop.org (Postfix) with ESMTP id DB4B09E761
	for <intel-gfx@lists.freedesktop.org>;
	Mon, 15 Aug 2011 16:59:29 -0700 (PDT)
In-Reply-To: <20110812152832.6c922a17@jbarnes-desktop>
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: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On 08/12/2011 03:28 PM, Jesse Barnes wrote:
> On Fri, 12 Aug 2011 15:18:09 -0700
> Jesse Barnes <jbarnes@virtuousgeek.org> wrote:
> 
>> On Fri, 12 Aug 2011 14:55:32 -0700
>> Jesse Barnes <jbarnes@virtuousgeek.org> wrote:
>>
>>> GFX_MODE controls important behavior like PPGTT, run lists, and TLB
>>> invalidate behavior.  On the SDV I'm using, the TLB invalidation mode
>>> was defaulting to "pipe control only" which meant regular MI_FLUSHes
>>> wouldn't actually flush the TLB, leading to all sorts of stale data
>>> getting used.
>>>
>>> So initialize it to 0 at ring buffer init time until we actually use
>>> PIPE_CONTROL for TLB invalidation.
>>
>> Ignore this one, see below for an updated patch that uses bit
>> definitions and makes sure the register gets reset at GPU reset time as
>> well.
> 
> Ignore the last one too.  Third time's the charm!

Tested-by: Kenneth Graunke <kenneth@whitecape.org>

With drm-intel-next, I've been seeing frequent but intermittent
rendering corruption in GNOME Terminal, rendercheck failures, and the
occasional kernel crash.

After applying the patch, I was able to run for an entire day without
any crashes or rendering corruption.  To double check, I reverted this
patch and was immediately able to reproduce my earlier problems.

With this patch and C0 hardware, my system seems quite stable.  Thanks
Jesse!