-
The SPIR-V generator currently uses `unsigned int` for all its internal representation. However there is no guarantee for this type to be 32 bits (one SPIR-V word) wide.
There are two solutions to th…
-
Today, variables (usually arrays) placed into groupshared memory are simply marked with 'groupshared' and the compiler sees fit where to place those variables in the available groupshared memory.
How…
-
Follow up from #6747 :
```
void bar (inout float p) { }
[numthreads(1, 1, 1)]
void main() {
int3 a = 0;
bar(a);
}
```
```
dxc -T cs_6_6 -E main -spirv -fcgl repro.hlsl
dxc: /h…
-
The notes say: VUDA does not provide any support for compiling CUDA C kernels directly to SPIR-V (yet)
Is there a roadmap or technical approach you're planning to support that? Maybe we can help impl…
-
> I found it odd that we need to add a 'Targets' folder in Basic when no other Target seems to require this.
Other targets have this value pulled from the backend via the `TargetTransformInfo` (whi…
-
We encountered the following error with one of our shaders when using `-fspv-debug=vulkan-with-source` to debug the shader in RenderDoc:
```
fatal error: generated SPIR-V is invalid: NonSemantic.S…
-
At the moment, the current SPIR-V back-end uses an exta pass, which results in increased and (probably) unnecessary memory.
After everything is parsed, it then rearranges all the data in the correc…
-
- [ ] Implement `GetRenderTargetSampleCount` clang builtin,
- [ ] Link `GetRenderTargetSampleCount` clang builtin with `hlsl_intrinsics.h`
- [ ] Add sema checks for `GetRenderTargetSampleCount` to `Ch…
-
HLSL has buffer, texture, and sampler types that need to be lowered to resource representations when lowered to DXIL and SPIR-V.
## Milestones
### Compile a runnable shader from clang (#7)
- [x] ll…
-
Current issues:
Issue | Total failures | First passes to fail | Phabricator reviews/Pull Requests
-- | -- | -- | --
Illegal virtual register for instruction | 102 | Post-RA pseudo instructi…