gnl21

> I am going to ping the GLSL spec editor @gnl21 for confirmation. I don't have any official confirmation to give, but I agree that this would be desirable and...

Support for these builtins in SPIR-V was discussed at Khronos here: spirv/SPIR-V#296. It looks like these functions were not exposed in GLSL on 64-bit types except in the explicit types...

> FYI, these functions are not explicitly mentioned in GL_EXT_shader_explicit_arithmetic_types, but ended up included in the extension by default since they are defined in the core spec with genIType and...

I think that the glslang behaviour does follow the current spec [1], although the spec itself is a bit weird. The GL working group will resolve what the spec should...

This seems like a generalisation of the device profiles that glslang already supports. I believe that you can currently supply a file with device limits to use during compilation. If...

I don't think this was ever intended to be allowed in GLSL, but I can't find anything specifically disallowing it in the spec either. Could someone open a GLSL issue...

> You may also wish to open a request to update the GLSL spec to specify that #if expressions must be signed integer. This is not how I had interpreted...

This was discussed in the OpenGL working group and the clarification from Vulkan is intended to apply in GLSL as well. It will be added in a future revision of...

> I believe type typo is on the last line. Yes, this looks clearly incorrect. I'll fix it in the next version of the specification. Thanks. > I'm not sure...

Sorry, looks like I had left my last comments unpublished. Just let me catch back up with what I had written ...