bucket: add query len 1 to prefill bucket #645
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
xinyu-intel
requested review from
adobrzyn,
afierka-intel,
iboiko-habana,
kamil-kaczor,
ksmusz,
kzawora-intel,
mgawarkiewicz-intel,
michalkuligowski,
vivekgoe and
xuechendi
as code owners
Contributor
There was a problem hiding this comment.
Pull request overview
This PR modifies the bucketing configuration to support query length 1 for prefill operations, preventing unnecessary padding to block size in prefix-decode (PD) scenarios.
- Changes the minimum query bucket size from block_size to 1
- Updates dummy prefill batch generation to use query_len=1 and context_len=127
- Adds support for bucket value 1 in the exponential bucketing warmup logic
Reviewed changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| vllm_gaudi/extension/bucketing/linear.py | Sets minimum prompt query bucket to 1 instead of block_size |
| vllm_gaudi/extension/bucketing/exponential.py | Updates exponential bucketing to start from 1 and adds logic to handle bucket value 1 |
| vllm_gaudi/v1/worker/hpu_model_runner.py | Adjusts dummy prefill batch to use query_len=1 and context_len=127 |
| tests/unit_tests/test_bucketing.py | Adds test case for warmup_range starting with 1 |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
xinyu-intel
force-pushed
the
dev/xinyu/prefill-bucket-one
branch
from
500c177 to
37b3e7d
Compare
✅ CI PassedAll checks passed successfully against the following vllm commit: |
xinyu-intel
force-pushed
the
dev/xinyu/prefill-bucket-one
branch
2 times, most recently
from
32c4ddc to
01a12b7
Compare
Signed-off-by: Xinyu Chen <[email protected]>
Signed-off-by: Wuxun Zhang <[email protected]>
✅ CI PassedAll checks passed successfully against the following vllm commit: |
wuxun-zhang
reviewed
Nov 28, 2025
Signed-off-by: Xinyu Chen <[email protected]>
Co-authored-by: Wuxun Zhang <[email protected]> Signed-off-by: Xinyu Chen <[email protected]>
✅ CI PassedAll checks passed successfully against the following vllm commit: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Avoid the query length(1) of the prefix prefill on the decode side to be padded to the block size under PD+DP scenario.
Use case:
VLLM_EXPONENTIAL_BUCKETING=false/true VLLM_PROMPT_QUERY_BUCKET_MIN=1 on the decode side.