[P/D] force with_prefill true after allreduce in kv producer #3768
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
Signed-off-by: liziyu <[email protected]>
liziyu179
changed the title
![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 moves the logic for forcing with_prefill = True on KV producer nodes to be executed before the _sync_metadata_across_dp call. This change is crucial as it ensures that the with_prefill state is consistently synchronized across all data-parallel ranks. By doing so, it prevents potential deadlocks or correctness issues that could arise from inconsistent states during distributed dummy runs. The modification is a clean and effective fix for a critical distributed logic issue.
|
👋 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. |
Signed-off-by: liziyu <[email protected]>
yiz-liu
approved these changes
Oct 28, 2025
Collaborator
yiz-liu
left a comment
yiz-liu
left a comment
There was a problem hiding this comment.
We should remove this with_prefill = True once attn_state is refactored or removed.
linfeng-yuan
approved these changes
Oct 29, 2025
MengqingCao
approved these changes
Oct 29, 2025
Collaborator
|
LGTM. It doesn't affect torchair scenario since we always expect |
MengqingCao
pushed a commit
that referenced
this pull request
Oct 29, 2025
#3835) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer. This is a backport of #3768 and #3849 --------- Signed-off-by: liziyu <[email protected]>
luolun
pushed a commit
to luolun/vllm-ascend
that referenced
this pull request
Nov 19, 2025
…oject#3768) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer - vLLM version: v0.11.0rc3 - vLLM main: vllm-project/vllm@c9461e0 --------- Signed-off-by: liziyu <[email protected]> Signed-off-by: luolun <[email protected]>
hwhaokun
pushed a commit
to hwhaokun/vllm-ascend
that referenced
this pull request
Nov 19, 2025
…oject#3768) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer - vLLM version: v0.11.0rc3 - vLLM main: vllm-project/vllm@c9461e0 --------- Signed-off-by: liziyu <[email protected]> Signed-off-by: hwhaokun <[email protected]>
NSDie
pushed a commit
to NSDie/vllm-ascend
that referenced
this pull request
Nov 24, 2025
…oject#3768) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer - vLLM version: v0.11.0rc3 - vLLM main: vllm-project/vllm@c9461e0 --------- Signed-off-by: liziyu <[email protected]> Signed-off-by: nsdie <[email protected]>
Clorist33
pushed a commit
to Clorist33/vllm-ascend
that referenced
this pull request
Dec 9, 2025
vllm-project#3835) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer. This is a backport of vllm-project#3768 and vllm-project#3849 --------- Signed-off-by: liziyu <[email protected]>
Clorist33
pushed a commit
to Clorist33/vllm-ascend
that referenced
this pull request
Dec 10, 2025
…oject#3768) ### What this PR does / why we need it? force with_prefill true after allreduce in kv producer - vLLM version: v0.11.0rc3 - vLLM main: vllm-project/vllm@c9461e0 --------- Signed-off-by: liziyu <[email protected]>
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?
force with_prefill true after allreduce in kv producer
Does this PR introduce any user-facing change?
How was this patch tested?