John Kessenich
John Kessenich
On further internal discussion, it is better to keep this as an **in**, at least for Vukan, because we don't want to make the ISV change the `.conf` file per...
I see the value in that, but suspect it should be achieved by a different change.
> This is also identical problem as for the other physical device limits, no? Technically, but in practice there is a different emphasis on knowing this compared to limits: This...
Vulkan has `VK_PIPELINE_SHADER_STAGE_CREATE_ALLOW_VARYING_SUBGROUP_SIZE_BIT_EXT` which says whether or not the SPIR-V `SubgroupSize` can vary.
>It seems like we should allow format qualifiers on function parameters, so that loads can be supported. I think that's the right answer, and possibly to warn (or error) at...
Glslang is updated now to give an error if formats do not match between calling argument and formal parameter (unless the formal parameter is writeonly and at least one of...
Are you asking to add a statement like the following? > Passing bits that represent a denormalized value might cause the result to be flushed to 0. I haven't looked...
It's probably the return type of the entry point.
It turns out this was only happening in non-normative SPIR-V mapping sections. Still fixing it, but there are no normative implications.
The derived variables are based on the workgroup size that was fixed before the shader started running, so there is no problem using them early, or with multiple compilation units....