[Usage]: Is there any fp8_blockscale_gemm performance comparison data between nvcc and nvrtc?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 14.7k
- Forks
- 2.8k
- Avg merge
- 2d 23h
- Merged PRs (30d)
- 489
Description
Is there any fp8_blockscale_gemm performance comparison data between nvcc and nvrtc?
"Note that there is some perf drop when using NVRTC due to a known bug of NVRTC which leads to extra instructions (but in the m=4096,n=2112,k=7168 case, NVRTC version was faster, which was a bit strange)" From Deepgemm。
Has this bug been fixed？
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.
Research direction
Start with the fp8_blockscale_gemm performance comparison between nvcc and NVRTC, using the DeepGEMM note in the issue as the initial reference. Check whether the cited NVRTC extra-instruction bug has been fixed and whether results differ for the m=4096,n=2112,k=7168 case; done means documenting the comparison data and fix status.
Written by the indexing model from the issue text.
Assessment
- Domain
- compilers, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100