[ubuntu-x] Intel GPU hangs and batchbuffer dumps

[Geir Ove Myhr]

On Fri, Feb 19, 2010 at 3:55 AM, Bryce Harrington <bryce at canonical.com> wrote:
> I synthetically created an X freeze to test the apport freeze collector,

How can I synthetically create an X freeze? This would be useful for
testing troubleshooting procedures.

> I guess we need more than just the gpu dump, but this at least gives a
> starting point.

We used to collect /sys/kernel/debug/dri/0/i915* as well. I don't know
what the different files really means, but I think i915_regs contains
the same information as the output of intel_gpu_dump and has been
removed in recent kernels since reading it from debugfs sometimes
caused GPU hangs [1].

With the patch from Chris Wilson, it should be sufficient to capture
only the file i915_error_state, but I guess we have to get the timing
right. The udev rule is only triggered when the kernel notices that
the GPU is hung, right? At that time the GPU is reset and this is
probably also the time that i915_error_state shows up. So I'm
wondering if we currently end up with recording a GPU dump of a
reinitialized GPU, which is not very useful. Maybe this would have
been obvious to me if I knew how to read the output of
intel_gpu_dump...

[1]: http://lists.freedesktop.org/archives/intel-gfx/2009-December/005250.html