Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-05 Thread Daniel Vetter
On Sat, Nov 05, 2011 at 03:07:45PM -0700, Ben Widawsky wrote: > On Wed, 2 Nov 2011 12:46:36 +0100 > Daniel Vetter wrote: > > > GPU reset is a very important piece of our infrastructure. > > Unfortunately we only really test it by actually hanging the gpu, > > which often has bad side-effects for

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-05 Thread Ben Widawsky
On Wed, 2 Nov 2011 12:46:36 +0100 Daniel Vetter wrote: > GPU reset is a very important piece of our infrastructure. > Unfortunately we only really test it by actually hanging the gpu, > which often has bad side-effects for the entire system. And the gpu > hang handling code is one of the rather

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-04 Thread Eugeni Dodonov
On Fri, Nov 4, 2011 at 14:31, Chris Wilson wrote: > On Fri, 4 Nov 2011 14:23:19 -0200, Eugeni Dodonov > wrote: > > Yep, it works for me. I can disable and enable rings at will by echo > [01] > > > /sys/kernel/debug/dri/0/i915_ring_stop > > > > I was using it for working further on that i915_erro

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-04 Thread Chris Wilson
On Fri, 4 Nov 2011 14:23:19 -0200, Eugeni Dodonov wrote: > Yep, it works for me. I can disable and enable rings at will by echo [01] > > /sys/kernel/debug/dri/0/i915_ring_stop > > I was using it for working further on that i915_error_state heuristics > idea, outside of intel_decode. It is *very*

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-04 Thread Eugeni Dodonov
On Thu, Nov 3, 2011 at 14:10, Daniel Vetter wrote: > On Thu, Nov 03, 2011 at 01:56:11PM -0200, Eugeni Dodonov wrote: > > Could be handy for debugging, I like it. > > Have you tried actually blowing up your gpu with it? That's where the fun > starts ;-) My gpu-hanger branch contains the latest ite

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-03 Thread Daniel Vetter
On Thu, Nov 03, 2011 at 01:56:11PM -0200, Eugeni Dodonov wrote: > Could be handy for debugging, I like it. Have you tried actually blowing up your gpu with it? That's where the fun starts ;-) My gpu-hanger branch contains the latest iteration of this idea: http://cgit.freedesktop.org/~danvet/drm/

Re: [Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-03 Thread Eugeni Dodonov
On Wed, Nov 2, 2011 at 09:46, Daniel Vetter wrote: > GPU reset is a very important piece of our infrastructure. > Unfortunately we only really test it by actually hanging the gpu, > which often has bad side-effects for the entire system. And the gpu > hang handling code is one of the rather compl

[Intel-gfx] [PATCH 3/3] [RFC] drm/i915: add interface to simulate gpu hangs

2011-11-02 Thread Daniel Vetter
GPU reset is a very important piece of our infrastructure. Unfortunately we only really test it by actually hanging the gpu, which often has bad side-effects for the entire system. And the gpu hang handling code is one of the rather complicated pieces of code we have, consisting of - hang detection