[refactor] Refactoring AscendFusedMoE #1169
Closed
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
zzzzwwjj
force-pushed
the
main
branch
6 times, most recently
from
a79efd7 to
20e3bb6
Compare
Yikun
added
long-term-test
|
I really like this change. Let's merge this first. @Yikun @ganyi1996ppo @jianzs please take this as the high priory. Thanks. |
wangxiyuan
reviewed
Jun 11, 2025
vllm_ascend/utils.py
Outdated
There was a problem hiding this comment.
We can add a sub module names fused_moe in ops. Then move FusedMoEState and get_fused_moe_state to that module's utils file.
wangxiyuan
reviewed
Jun 11, 2025
Any change in this PR related to this commit message? |
|
|
|
It's better to describe which communication kernel is chosen for different configurations. |
whx-sjtu
reviewed
Jun 12, 2025
There was a problem hiding this comment.
The current usage of MC2 kernel does not support non-uniform inputs, so padding is still required.
|
please rebase to main to make sure the torchair CI passed |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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Outdated suggestions cannot be applied.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.
What this PR does / why we need it?
This PR is used for resolved issue 1147
fused_moe.py.get_fused_moe_state.Does this PR introduce any user-facing change?
VLLM_ENABLE_MC2, because I think this env is useless, we can make judgments based on the current scenario without this env, it will only increase complexity.USING_LCCL_COM, because this env has already expired.additional_config.expert_tensor_parallel_sizehas already expired, and now we also use parameterenable_expert_parallel, consistent with the vLLM.How was this patch tested?