[Mesa-dev] [PATCH] main/shaderimage: image unit invalid if texture is incomplete, independently of the level

[Alejandro Piñeiro]

On 27/07/16 13:47, Alejandro Piñeiro wrote:
> On 26/07/16 22:16, Francisco Jerez wrote:
>>>> -- However that spec change would have the side effect of making
>>>> the CTS test you're trying to fix non-compliant...
>>> This would not be the first case of a CTS test that is not correct due
>>> the complexities/ambiguities of the OpenGL spec. Or worse, CTS tests
>>> that are just
>>>
>> Heh, seems like some words were censored from your last paragraph. :P
> Totally unintentional and innocent, I promise ;) When writing the email
> I didn't find a suitable word when writing that paragraph, and then I
> forgot to got back. Let's go simple and just say "wrong".
>
>>> So what about this: I open a Khronos bug (public unless someone thinks
>>> that it would be better private), using your explanation as a base to
>>> the description (although tempted to just point this email), and we put
>>> the patch/bugfix on hold until we get an answer?
>>>
>> Sure, feel free to use my last post as starting point, and thanks for
>> filing the bug at Khronos. :)
> Thanks to you for the patience and the long and detailed explanation.
>
> You can find the bug here:
> https://www.khronos.org/bugzilla/show_bug.cgi?id=1654
>
> Is basically a edited form of your explanation (some sections are your
> explanation word by word).

FWIW, the list of CTS tests that would need to be tweaked is bigger. 8
CTS tests would need to be updated if the conclusion is that as soon as
the texture is incomplete the image unit is invalid. Full list on the
ping I just sent to the khronos bug:
https://www.khronos.org/bugzilla/show_bug.cgi?id=1654#c1

BR

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: OpenPGP digital signature
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20160803/c77b2fae/attachment-0001.sig>