Ensure CUDA kernels will run correctly regardless of calling function
Open
@pelesh is already working on this.
Since Oct 18, 2023.
- Dominant language
- C++
- Stars
- 84
- Forks
- 13
- PR merge metrics
- No merged PRs in 30d
Description
CUDA kernels are often written under assumption that grid size and block size will be selected so that the total number of threads is greater or equal to the number of parallel tasks. Consider writing kernels in a way that it will add thread blocks if needed to complete the computation correctly.
Contributor guide
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.
Assessment
This issue has not been assessed yet.