[Intel-gfx] [01/15] drm/i915: Copy user requested buffers into the error state

[Chris Wilson]

On Tue, Mar 21, 2017 at 10:53:53AM -0700, Ben Widawsky wrote:
> On 17-03-21 16:23:05, Tahvanainen, Jari wrote:
> >See below [Jari]...
> >
> >-----Original Message-----
> >From: Ben Widawsky [mailto:ben at bwidawsk.net]
> >Sent: Tuesday, March 21, 2017 5:38 PM
> >To: Tahvanainen, Jari <jari.tahvanainen at intel.com>
> >Cc: Chris Wilson <chris at chris-wilson.co.uk>; intel-gfx at lists.freedesktop.org
> >Subject: Re: [01/15] drm/i915: Copy user requested buffers into the error state
> >
> >On 17-03-21 11:30:36, Tahvanainen, Jari wrote:
> >>Note that this is for all the patches in series, replied only on [1/15].
> >>
> >>See also https://bugs.freedesktop.org/show_bug.cgi?id=94001#c45
> >>
> >
> >Jari, did you test this patch specifically? It would involve introspection of the error state.
> >
> >[Jari]  like said I tested the patch series including this patch
> >	" Note that this is for all the patches in series, replied only on [1/15]"
> >             Tested-by " for https://patchwork.freedesktop.org/series/21377"
> >          	
> >	If this is not the way to do it then I need to stop.
> >          	And since being tester (not programmer) you need to tell more what do you mean with " would involve introspection of the error state".
> >	What should be outcome? What skill shall have for it, etc.? If I cannot do it then assumable tested-by is not the thing that I will do in future.
> 
> Well there is tested-by "this doesn't regress anything" and there is tested-by
> "this new feature works properly". I've no doubt you asserted the first, but my
> concern was around the second. For this patch specifically, it's a new feature
> and there is no igt test for it AFAIK.

There is/will be, gem_exec_capture; sent alongside the original
patch.
-Chris