Add a unittest for fused_moe #2416
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
BBuf
commented
Dec 9, 2024
|
|
||
| def _test_case(self, m, n, k, e, topk, dtype, use_fp8_w8a8=False): | ||
| if use_fp8_w8a8: | ||
| # AssertionError: fp8e4nv data type is not supported on CUDA arch < 89 |
There was a problem hiding this comment.
If the GPU is Ampere architecture, we should fallback from fused_moe to either naive implementation or torch.compile implementation to prevent errors.
merrymercy
reviewed
Dec 9, 2024
timethink
pushed a commit
to timethink/sglang
that referenced
this pull request
Mar 9, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
fused_moe_tritonunittest for bf16, fp16 and fp8_w8a8.fused_moebenchmark readme.mdWhen I wan't to deploy qwen2-57b-a14b model in A800 with fp8, the error happens:
The reason is that in Triton, the Ampere architecture currently doesn't support the
fp8e4nvdtype. To detect this situation early, I added thefused_moetest mentioned above, and verify whether thefused_moeoperator can work properly on the current GPU by checking the GPU architecture information.