[Intel-gfx] [PATCH 2/3] drm/i915: Record pid/comm of hanging task

[Chris Wilson]

On Wed, Mar 05, 2014 at 02:34:48PM +0100, Daniel Vetter wrote:
> On Wed, Mar 05, 2014 at 01:08:15PM +0000, Chris Wilson wrote:
> > On Wed, Feb 12, 2014 at 11:32:00AM -0800, Ben Widawsky wrote:
> > > On Wed, Feb 12, 2014 at 07:18:19PM +0000, Chris Wilson wrote:
> > > > On Wed, Feb 12, 2014 at 10:55:07AM -0800, Ben Widawsky wrote:
> > > > > On Wed, Feb 12, 2014 at 08:15:58AM +0000, Chris Wilson wrote:
> > > > > Are you opposed to doing anything at context creation? pid reference
> > > > > works for me too, or hold on to it with the request - I don't care.
> > > > 
> > > > I'm just don't feel that the issue is severe enough for yet another pid
> > > > reference floating around inside drm. I could well be wrong though, but
> > > > I would like to be sure we have tried our best to use the available
> > > > information we have already.
> > > > -Chris
> > > > 
> > > > -- 
> > > > Chris Wilson, Intel Open Source Technology Centre
> > > 
> > > Well, the ref wouldn't be held if you just copied it at context create,
> > > and it avoids having to stick in *file to actually do the proper look up
> > > (not that that in itself is so horrible).
> > > 
> > > For the sake of making progress, I disagree with Chris in tbut think it's
> > > better than what is currently there, so:
> > > Acked-by: Ben Widawsky <ben at bwidawsk.net>
> > 
> > Daniel?
> 
> I've thought I pick this all up when Mika resends the revised versions for
> 1&3. Or is everything ready already (it didn't quite look like that ...)?

Oh, I just forgot where this was taken up. It just made it to the top of
my wishlist with a couple of bizarre error-states containing gibberish
for which I want to find the culprit.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre