[Intel-gfx] [PATCH i-g-t 3/6] lib/igt_fb: Add igt_cairo_image_surface_create_from_png()

[Petri Latvala]

On Fri, Sep 22, 2017 at 03:05:44PM +0300, Ville Syrjälä wrote:
> On Fri, Sep 22, 2017 at 12:52:59PM +0300, Petri Latvala wrote:
> > On Thu, Sep 21, 2017 at 05:39:30PM +0300, Ville Syrjala wrote:
> > > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > > 
> > > Raw usage of cairo_image_surface_create_from_png() doesn't work
> > > since it doesn't know about IGT_DATADIR and IGT_SRCDIR. Let's extract
> > > the helper from igt_paint_image() that uses igt_fopen_data() +
> > > cairo_image_surface_create_from_png_stream() and call it
> > > igt_cairo_image_surface_create_from_png_file().
> > 
> > 
> > s/from_png_file/from_png/
> 
> Argh, still one left. My brain kept subconsciously adding the _file()
> (to match the _stream() I suppose). In fact at one point I was scratching
> my head for quite a while wondering why it wasn't compiling...
> 
> > 
> > 
> > > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > Thanks for fixing this.
> > 
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92248
> > (Not sure if this patch fixes it yet or if CI switch to make install -deployment is also needed)
> 
> Hmm. The asserts I see there seem to be about display commits failing.


Oh, gah. The bug report has changed meaning. I was looking at the
original description that had

Failed assertion: cairo_surface_status(image) == CAIRO_STATUS_SUCCESS



> 
> > kms_plane_fitting doesn't seem to have a separate bug report.
> 
> I'm pretty sure such a bug report did exist in the past. But I
> wasn't able to find it either. I guess it was closed for some reason.

(For those following along, I did search for kms_panel_fitting too,
not just that brainfart of a misname.)

I suppose kms_panel_fitting failure is bundled up in #92248 along with
kms_plane_scaling.


-- 
Petri Latvala