NVIDIA / NVIDIA/TransformerEngine
Questions about test_float8_blockwise_gemm_exact.py
Open
Nobody has claimed this yet.
question
- Dominant language
- Python
- Stars
- 3.5k
- Forks
- 831
- Avg merge
- 3d 11h
- Merged PRs (30d)
- 65
Description
Hello ：
We notice that the ref function in float8_blockwise_gemm is torch._scaled_mm, however torch._scaled_mm may call cublasLtMatmul , which is same to what TE blockwise gemm calls.
So I'm confused that this utest will pass even when the blas gemm returns wrong result. What is the purpose of this utest?
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
Read test_float8_blockwise_gemm_exact.py and trace the float8_blockwise_gemm reference through torch._scaled_mm. Compare that path with cublasLtMatmul and the TE blockwise GEMM call, then document whether the test independently detects an incorrect BLAS GEMM result and clarify the test's purpose.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- testing-qa
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100