[FIX_FOR_VLLM_CUSTOM=a78b842d0e85d287176031334f4721cd96b6e47d] Fix offloading_connector test flush assertion for load transfers#1468
Merged
iboiko-habana merged 1 commit intoMay 21, 2026
Conversation
pawel-olejniczak
requested review from
PatrykWo,
adobrzyn,
afierka-intel,
iboiko-habana,
jbyczkow,
kamil-kaczor,
ksmusz,
mgawarkiewicz-intel,
michalkuligowski and
xuechendi
as code owners
🚧 CI BlockedThe main CI workflow was not started for the following reason:
|
Contributor
There was a problem hiding this comment.
Pull request overview
Note
Copilot was unable to run its full agentic suite in this review.
Updates KV offload test utilities and scheduler tests to correctly account for “flush” behavior in both store and load transfer paths, especially under async scheduling.
Changes:
- Extend transfer parsing to record flushed GPU blocks for both store-flush and load-flush scenarios.
- Adjust scheduler tests’ expected flushed GPU block indexes depending on sync vs async scheduling.
- Add an assertion to ensure pending jobs are drained after request completion.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| tests/unit_tests/kv_offload/offloading_connector/utils.py | Updates _parse_transfers to handle load-flush transfers in addition to store-flush. |
| tests/unit_tests/kv_offload/offloading_connector/test_scheduler.py | Updates expected flush behavior for async scheduling and adds a pending-jobs drain assertion. |
pawel-olejniczak
force-pushed
the
fix/offloading-connector-flush-assertion
branch
from
88f030e to
0d5d656
Compare
🚧 CI BlockedThe main CI workflow was not started for the following reason:
|
pawel-olejniczak
force-pushed
the
fix/offloading-connector-flush-assertion
branch
from
0d5d656 to
e8f321d
Compare
pawel-olejniczak
force-pushed
the
fix/offloading-connector-flush-assertion
branch
from
e8f321d to
2850195
Compare
Upstream PR #42611 changed get_flushed_transfers() to return both store and load flushes. The vllm-gaudi test assumed only store flushes, causing isInstance assertion failures. Mirror the upstream fix: handle both flush types in _parse_transfers and update test expectations. Signed-off-by: Paweł Olejniczak <pawelx.olejniczak@intel.com>
pawel-olejniczak
force-pushed
the
fix/offloading-connector-flush-assertion
branch
from
2850195 to
98e5637
Compare
✅ CI PassedAll checks passed successfully against the following vllm commit: |
iboiko-habana
approved these changes
May 21, 2026
12010486
pushed a commit
to 12010486/vllm-gaudi
that referenced
this pull request
May 21, 2026
…floading_connector test flush assertion for load transfers (vllm-project#1468) Upstream vLLM PR vllm-project/vllm#42611 ("Flush all pending jobs on last step") changed \`get_flushed_transfers()\` to return both store and load flushes. The vllm-gaudi copy of the offloading_connector unit tests assumed only store flushes, causing: 1. \`AssertionError\` in \`utils.py\` \`_parse_transfers\` (\`isinstance(src_spec, GPULoadStoreSpec)\` assert fails on load flushes) 2. \`flushed_gpu_block_indexes\` mismatch in \`test_scheduler\` tests **Fix**: Mirror the upstream change — replace the assert with an \`if/else\` handling both store and load flush types, and add \`expected_flushed_gpu_block_indexes\` to affected tests. Signed-off-by: Paweł Olejniczak <pawelx.olejniczak@intel.com> Signed-off-by: 12010486 <silvia.colabrese@intel.com>
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
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.
Upstream vLLM PR vllm-project/vllm#42611 ("Flush all pending jobs on last step") changed `get_flushed_transfers()` to return both store and load flushes. The vllm-gaudi copy of the offloading_connector unit tests assumed only store flushes, causing:
Fix: Mirror the upstream change — replace the assert with an `if/else` handling both store and load flush types, and add `expected_flushed_gpu_block_indexes` to affected tests.