[Mesa-dev] [PATCH v2 5/5] i965: Enabled the OES_copy_image extension on Gen 7 GPUs

[Ilia Mirkin]

On Thu, Feb 7, 2019 at 2:49 AM Eleni Maria Stea <estea at igalia.com> wrote:
>
> On Wed, 6 Feb 2019 12:12:27 -0800
> Nanley Chery <nanleychery at gmail.com> wrote:
>
> > > +       * For now, we can't enable OES_texture_view on Gen 7
> > > because of
> > > +       * some piglit failures coming from
> > > +       * piglit/tests/spec/arb_texture_view/rendering-formats.c
> > > that need
> > > +       * investigation.
> > >         */
> >
> > What kind of failures are you seeing? I'd imagine texture views to
> > work with this version of your series.
> >
>
> Hi Nanley,
>
> If you run the piglit test: arb_texture_view-rendering-format, and grep
> for failures on HSW:
>
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_R32F" : "fail"}}
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_RG16F" : "fail"}}
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_RG16I" : "fail"}}
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_RG16_SNORM" : "fail"}}
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_RGBA8I" : "fail"}}
> PIGLIT: {"subtest": {"clear GL_RGB10_A2 as GL_RGBA8_SNORM" : "fail"}}
>
> I remember seeing similar errors on Ivy too. They must be irrelevant to
> the ETC support but as this test passes on BDW where the extension is
> enabled, I didn't enable it on Gen 7 for the moment. I think I had
> discussed about these failures with Kenneth before I disabled them, but
> I didn't investigated them further after that.

Do you also see the failures with desktop GL (and ARB_texture_view)?
If not, that'd be very surprising.

Note that the piglit test arb_texture_view-rendering-formats is the
desktop GL test. arb_texture_view-rendering-formats_gles3 is the ES
version.

  -ilia