Closed kusma closed 3 months ago
What about the XML? I guess the only change is that this:
<enum value="0x93A1" name="GL_BGRA8_EXT"/>
Should be:
<enum value="0x93A1" name="GL_BGRA8_EXT" group="PixelFormat"/>
@SunSerega you are right that it should be part of the PixelFormat group, but this fact is not changed by this PR and therefore is effectively out of scope. What would be in scope for this PR is adding the InternalFormat group as well, but given that neither group is present currently I suggest one of the XML editors fix this in a subsequent PR.
By the way, I think I've come to the realization that the change in this MR needs to actually be an interaction with OpenGL ES 3.0, because in OpenGL ES 1.x and 2.x it's not really legal to use different formats and internalformats... This basically depends on sized internalformat support.
By the way, I think I've come to the realization that the change in this MR needs to actually be an interaction with OpenGL ES 3.0, because in OpenGL ES 1.x and 2.x it's not really legal to use different formats and internalformats... This basically depends on sized internalformat support.
Actually, GLES 2 has some support for sized internalformats... But not to the full extent, so it seems better to limit this to OpenGL ES 3.0 and later... That's also where we see the problems in current implementations.
CTS tests added here: https://github.com/KhronosGroup/VK-GL-CTS/pull/449
lgtm, thankyou!
WG has approved the latest version.
WG has approved the latest version.
@zmike could you set the Signed Off milestone if that's what this means?
WG has approved the latest version.
@zmike could you set the Signed Off milestone if that's what this means?
The comment was for the author. I'll set the milestone when this is ready to merge.
The CTS tests landed in the Khronos GitLab repo. Is it time to merge this PR?
As agreed on today's call, let's make GL_BGRA8_EXT an allowed internalformat as well.