Fix patch_hf3fs_mock_client_for_cpu_only#1439
Merged
iboiko-habana merged 9 commits intoMay 19, 2026
Merged
Conversation
Signed-off-by: Harish Subramony <harish.subramony@intel.com>
hsubramony
requested review from
PatrykWo,
adobrzyn,
afierka-intel,
iboiko-habana,
jbyczkow,
kamil-kaczor,
ksmusz,
mgawarkiewicz-intel,
michalkuligowski and
xuechendi
as code owners
Contributor
There was a problem hiding this comment.
Pull request overview
Expands the set of supported HPU V1 attention kernel block sizes to allow a smaller 16-token option (useful for testing/saller models) while retaining the existing production sizes used for standard and Granite 4.0-H configurations.
Changes:
- Add
16to the supported kernel block size list for the HPU V1 attention backend. - Update the in-code comment to document when/why each block size is used (standard vs Granite 4.0-H, with/without prefix caching).
✅ CI PassedAll checks passed successfully against the following vllm commit: |
hsubramony
force-pushed
the
fix_kernel_block_sizes
branch
from
7e3bcb4 to
592b1c3
Compare
🚧 CI BlockedThe main CI workflow was not started for the following reason:
|
Signed-off-by: Harish Subramony <harish.subramony@intel.com>
hsubramony
force-pushed
the
fix_kernel_block_sizes
branch
from
592b1c3 to
1550f30
Compare
Signed-off-by: Harish Subramony <harish.subramony@intel.com>
kamil-kaczor
pushed a commit
that referenced
this pull request
May 19, 2026
iboiko-habana
approved these changes
May 19, 2026
iboiko-habana
changed the title
✅ CI PassedAll checks passed successfully against the following vllm commit: |
mgawarkiewicz-intel
pushed a commit
that referenced
this pull request
May 25, 2026
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.
added in fix kernel block size, port of #1439 #1453
16 is supported for testing/smaller models; 128 is the standard HPU
kernel block size; 528 is required for Granite 4.0-H
(granitemoehybrid) without prefix caching (16-token FA alignment),
768 with prefix caching (chunk-aligned).
_patch_hf3fs_mock_client_for_cpu_only
Upstream mock client unconditionally calls
torch.cuda.current_stream().wait_event(event)inbatch_write.In environments where PyTorch is not compiled with CUDA, that path throws
and the method returns
-1for writes, causing connector unit tests tofail. This patch keeps the same behavior but skips CUDA synchronization when
CUDA is unavailable.