Re: [Mesa-dev] [PATCH] mesa/st/nir: fix driver_location for arrays of image/sampler

2018-06-11 Thread Rob Clark
fwiw, I didn't notice yet any case where this matters for textures, but for images, you could try (from deqp gles31 module) dEQP-GLES31.functional.layout_binding.image.image2d.fragment_binding_array as an example (perhaps I should have mentioned that in the commit msg) I'm not 100% sure if we

Re: [Mesa-dev] [PATCH] mesa/st/nir: fix driver_location for arrays of image/sampler

2018-06-11 Thread Timothy Arceri
Looks right to me. I guess we have no piglit tests for it, I ran the CTS a few times on radeonsi but I was more focused on getting piglit working. Reviewed-by: Timothy Arceri On 12/06/18 04:56, Rob Clark wrote: We can have arrays of images or samplers. But I forgot to handle that case long

[Mesa-dev] [PATCH] mesa/st/nir: fix driver_location for arrays of image/sampler

2018-06-11 Thread Rob Clark
We can have arrays of images or samplers. But I forgot to handle that case long ago. Suprised no one complained yet. Signed-off-by: Rob Clark --- src/mesa/state_tracker/st_glsl_to_nir.cpp | 16 +--- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git