[Piglit] random results with texture-integer-glsl130

Vinson Lee vlee at freedesktop.org
Thu Sep 6 22:35:13 PDT 2012


On Fri, Aug 31, 2012 at 10:13 AM, Vadim Girlin <vadimgirlin at gmail.com> wrote:
>
> texture-integer-glsl130 is randomly failing/passing for me with r600g.
> AFAICS it usually fails due to alpha component mismatch while testing
> the formats that don't have alpha. Here is the example:
>
> texture-integer: failure with format GL_RG8I:
>   texture color = 98, 98, 17, 0
>   expected color = 0.25, 0.5, 0, 0
>   result color = 0.25098, 0.501961, 0, 1
> PIGLIT: {'result': 'fail' }
>
> If I'm not missing something, then the result in fact is correct and
> expected alpha should be 1 according to GL/EXT_texture_integer specs:
>
>> Final Expansion to RGBA
>>
>>     This step is performed only for non-depth component groups. Each
>>     group is converted to a group of 4 elements as follows: if a group
>>     does not contain an A element, then A is added and set to 1 for
>>     integer components or 1.0 for floating-point components.
>
> Probably there is also a problem in the r600g driver, because it should
> never pass the test as far as I understand, but I'd like to find out
> exactly what is the correct expected alpha value before trying to fix
> the driver.
>
> So I think the test should set expected alpha to 1 for the formats
> without alpha component, is it right?
>
> Vadim
>

texture-integer-glsl130 fails randomly on softpipe as well.


More information about the Piglit mailing list