[Bugfix] use module-level import for patched function in Qwen3Next #4354
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
|
👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:
If CI fails, you can run linting and testing checks locally according Contributing and Testing. |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request correctly addresses a bug related to function patching in specific execution environments. By changing the import of chunk_gated_delta_rule from a direct function import to a module-level import, the code ensures that the function call is resolved at runtime. This allows for monkey-patching to work as intended, preventing the model from holding a reference to an old, unpatched function. The change is minimal, targeted, and effectively solves the described problem. The implementation is sound and follows Python best practices for creating patchable code. The changes look good and I have no further comments.
zjchenn
force-pushed
the
fix/module-level-import-for-qwen3next
branch
from
0db2bd3 to
caeb8cf
Compare
Contributor
Author
|
@wangxiyuan hi, please help me have a review |
zjchenn
changed the title
zjchenn
force-pushed
the
fix/module-level-import-for-qwen3next
branch
2 times, most recently
from
58e92af to
d0f9b3b
Compare
Contributor
Author
zjchenn
changed the title
weijinqian0
added
ready
Signed-off-by: zjchenn <zjchenn@gmail.com>
zjchenn
force-pushed
the
fix/module-level-import-for-qwen3next
branch
from
456e709 to
112061f
Compare
yiz-liu
approved these changes
Nov 25, 2025
MengqingCao
approved these changes
Nov 25, 2025
Collaborator
MengqingCao
left a comment
MengqingCao
left a comment
There was a problem hiding this comment.
LGTM, thx for this fix!
845473182
pushed a commit
to 845473182/vllm-ascend
that referenced
this pull request
Nov 29, 2025
…llm-project#4354) ### What this PR does / why we need it? **Problem**: The Qwen3Next model implementation currently imports chunk_gated_delta_rule directly using `from ... import ...` In frameworks like `verl`, the model file is often imported before `vllm-ascend` initializes and applies its patches. This causes the model to permanently hold a reference to the original (unpatched) vLLM kernel, resulting in execution errors on Ascend devices even if the patch runs later. **Solution**: Changed the import style to `from vllm...ops import chunk` and call `chunk.chunk_gated_delta_rule().` This ensures that the function lookup happens at runtime (dynamic dispatch), allowing the model to correctly pick up the patched function regardless of import order. - vLLM version: v0.11.0 - vLLM main: vllm-project/vllm@2918c1b Signed-off-by: zjchenn <zjchenn@gmail.com>
Meihan-chen
pushed a commit
to Meihan-chen/vllm-ascend
that referenced
this pull request
Dec 5, 2025
…llm-project#4354) ### What this PR does / why we need it? **Problem**: The Qwen3Next model implementation currently imports chunk_gated_delta_rule directly using `from ... import ...` In frameworks like `verl`, the model file is often imported before `vllm-ascend` initializes and applies its patches. This causes the model to permanently hold a reference to the original (unpatched) vLLM kernel, resulting in execution errors on Ascend devices even if the patch runs later. **Solution**: Changed the import style to `from vllm...ops import chunk` and call `chunk.chunk_gated_delta_rule().` This ensures that the function lookup happens at runtime (dynamic dispatch), allowing the model to correctly pick up the patched function regardless of import order. - vLLM version: v0.11.0 - vLLM main: vllm-project/vllm@2918c1b Signed-off-by: zjchenn <zjchenn@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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?
Problem: The Qwen3Next model implementation currently imports chunk_gated_delta_rule directly using
from ... import ...In frameworks like
verl, the model file is often imported beforevllm-ascendinitializes and applies its patches. This causes the model to permanently hold a reference to the original (unpatched) vLLM kernel, resulting in execution errors on Ascend devices even if the patch runs later.Solution: Changed the import style to
from vllm...ops import chunkand callchunk.chunk_gated_delta_rule().This ensures that the function lookup happens at runtime (dynamic dispatch), allowing the model to correctly pick up the patched function regardless of import order.
Does this PR introduce any user-facing change?
No. This is an internal bug fix to resolve import reference issues.
How was this patch tested?