From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ben Widawsky <ben@bwidawsk.net>
Subject: Re: [PATCH] drm/i915: hangcheck parameter
Date: Wed, 22 Jun 2011 10:55:33 -0700
Message-ID: <20110622175526.GD5157@bwgnt.jf.intel.com>
References: <1308763970-2989-1-git-send-email-ben@bwidawsk.net>
	<d08817$g5q74@azsmga001.ch.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org>
Received: from cloud01.chad-versace.us (184-106-247-128.static.cloud-ips.com
	[184.106.247.128])
	by gabe.freedesktop.org (Postfix) with ESMTP id 13D9C9E8F5
	for <intel-gfx@lists.freedesktop.org>;
	Wed, 22 Jun 2011 10:55:36 -0700 (PDT)
Content-Disposition: inline
In-Reply-To: <d08817$g5q74@azsmga001.ch.intel.com>
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: Chris Wilson <chris@chris-wilson.co.uk>
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On Wed, Jun 22, 2011 at 06:45:56PM +0100, Chris Wilson wrote:
> On Wed, 22 Jun 2011 10:32:50 -0700, Ben Widawsky <ben@bwidawsk.net> wrote:
> > The hangcheck is undesirable when doing shader debugging. The debugger
> > interacts with the EU, and these may cause the hangcheck to fire at most
> > unfortunate times.
> > 
> > This provides a way to let the user disable the hangcheck when they want
> > to do shader debugging.
> 
> Looks like a reasonable idea, I'm tempted to suggest to make it the period
> in ms rather a boolean. But for actual use, either a new ioctl for
> enable/disable or another wakelock? Just in case we have more than one
> program trying to debug at the same time. (Not so far fetched, as the user
> may be snooping on the results whilst debugging in another process?)
> 
> -Chris

Ah, Chris! You reminded me what I had planned to do a month ago, but it
got lost in all the other stuff floating around in my brain.

Yes, actually I will make a debugfs file, which notifies the kernel
we're debugging from userspace, and it should disable the hangcheck
timer. This will allow us to read/write potentially valuable information
in the future, and like you said acts as a lock to prevent concurrent
debuggers.

Thanks! Very helpful response/reminder.

I was planning a debugfs entry for it. Do you see any advantages to
using an IOCTL?

Ben