[QST] cute::abs_fn of cute::C fails to build on Windows with CUDA 11.8
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 10.5k
- Forks
- 2.1k
- Avg merge
- 3d 11h
- Merged PRs (30d)
- 7
Description
On Windows, this statement
std::cout << cute::abs_fn{}(cute::C<16> {}).value << std::endl;
does not build with CUTLASS 3.5.1 with CUDA 11.8. It does build with CUDA 12.2. With CUDA 11.8, the following static_assert statement triggers.
static_assert(std::is_arithmetic_v<cute::_1> == true);
The issue arose when @toothache was trying to build VLLM. Unary operations like abs_fn fail to build. Link to @toothache 's original post: https://github.com/NVIDIA/cutlass/issues/1126#issuecomment-2259463959
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the minimal cute::abs_fn{}(cute::C<16>{}) statement on Windows with CUTLASS 3.5.1 and CUDA 11.8, then compare the static_assert result and build behavior with CUDA 12.2. Read the cute::abs_fn and cute::C entry points involved in unary operations; done means the reported expression builds under CUDA 11.8 without breaking the CUDA 12.2 behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- build-system
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100