Skip to content

Revert "skip HPU graphs for long prefills"#850

Merged
adobrzyn merged 6 commits into
mainfrom
revert-780-skip_long_prefill_graph
Jan 26, 2026
Merged

Revert "skip HPU graphs for long prefills"#850
adobrzyn merged 6 commits into
mainfrom
revert-780-skip_long_prefill_graph

Conversation

@adobrzyn
Copy link
Copy Markdown
Collaborator

@adobrzyn adobrzyn commented Jan 21, 2026

Reverts #780

Copilot AI review requested due to automatic review settings January 21, 2026 12:36
Copilot AI reviewed Jan 21, 2026
View reviewed changes
Copy link
Copy Markdown
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR reverts a previous change that skipped HPU graphs for long prefill operations, restoring the simpler graph skipping logic based solely on max_cudagraph_capture_size.

Changes:

  • Simplified the condition for skipping HPU graphs from a complex formula considering sequence length and block size to a simpler check against max_cudagraph_capture_size
  • Moved max_cudagraph_capture_size assignment earlier in initialization and removed the max_graph_capture_tokens variable
  • Reformatted max_num_batched_tokens assignment for better readability

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

kamil-kaczor
kamil-kaczor approved these changes Jan 21, 2026
View reviewed changes
Copy link
Copy Markdown
Collaborator

@kamil-kaczor kamil-kaczor left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@github-actions github-actions Bot mentioned this pull request Jan 21, 2026
Open
@xuechendi xuechendi closed this Jan 21, 2026
@xuechendi xuechendi reopened this Jan 21, 2026
@github-actions
Copy link
Copy Markdown

github-actions Bot commented Jan 21, 2026

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

@github-actions
Copy link
Copy Markdown

github-actions Bot commented Jan 22, 2026

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

@github-actions
Copy link
Copy Markdown

github-actions Bot commented Jan 26, 2026

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

adobrzyn and others added 2 commits January 26, 2026 07:57
@adobrzyn
Merge branch 'main' into revert-780-skip_long_prefill_graph
4baf930
@adobrzyn
Adjust tests to avoid OOM
7f3c06d 
Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
@github-actions
Copy link
Copy Markdown

github-actions Bot commented Jan 26, 2026

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

@github-actions
Copy link
Copy Markdown

github-actions Bot commented Jan 26, 2026

✅ CI Passed

All checks passed successfully against the following vllm commit:
6218034dd7f9a56596e4fd8c8c8fc1d8011ed9c2

@adobrzyn adobrzyn merged commit 4c0e6ff into main Jan 26, 2026
53 checks passed
hlahkar pushed a commit to hlahkar/vllm-gaudi that referenced this pull request Jan 27, 2026
@adobrzyn @xuechendi
Revert "skip HPU graphs for long prefills" (vllm-project#850)
8a68c7d 
Reverts vllm-project#780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
adobrzyn added a commit that referenced this pull request Jan 27, 2026
@adobrzyn @xuechendi
Revert "skip HPU graphs for long prefills" (#850)
102e4a7 
Reverts #780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
mgawarkiewicz-intel pushed a commit that referenced this pull request Jan 28, 2026
@adobrzyn @xuechendi
Revert "skip HPU graphs for long prefills" (#850) (#888)
c66a038 
Reverts #780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
testdig pushed a commit to testdig/vllm-gaudi-fork that referenced this pull request Jan 29, 2026
@adobrzyn @xuechendi @testdig
Revert "skip HPU graphs for long prefills" (vllm-project#850)
756eaf6 
Reverts vllm-project#780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
Signed-off-by: Wang, Zheng W <zheng.w.wang@intel.com>
slokesha pushed a commit to libinta/vllm-gaudi that referenced this pull request Jan 29, 2026
@adobrzyn @xuechendi @slokesha
Revert "skip HPU graphs for long prefills" (vllm-project#850) (vllm-p…
a8def02 
…roject#888)

Reverts vllm-project#780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
Signed-off-by: slokesha <slokeshappa@habana.ai>
@yangulei
Copy link
Copy Markdown
Collaborator

yangulei commented Jan 30, 2026

@adobrzyn What is the motivation of this PR?

slokesha pushed a commit to libinta/vllm-gaudi that referenced this pull request Feb 9, 2026
@adobrzyn @xuechendi @slokesha
Revert "skip HPU graphs for long prefills" (vllm-project#850)
4a68c1f 
Reverts vllm-project#780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
Signed-off-by: slokesha <slokeshappa@habana.ai>
czhu15 added a commit that referenced this pull request Feb 10, 2026
@czhu15
Revert "Revert "skip HPU graphs for long prefills" (#850) (#888)"
c31c582 
This reverts commit c66a038.
yangulei added a commit that referenced this pull request Feb 24, 2026
@yangulei
Revert "Revert "skip HPU graphs for long prefills" (#850) (#888)"
4294027 
This reverts commit c66a038.
adobrzyn added a commit that referenced this pull request Mar 31, 2026
@adobrzyn @xuechendi
Revert "skip HPU graphs for long prefills" (#850)
31cf158 
Reverts #780

---------

Signed-off-by: Agata Dobrzyniewicz <adobrzyniewicz@habana.ai>
Co-authored-by: Chendi.Xue <chendi.xue@intel.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

Copilot code review Copilot Copilot left review comments

@kamil-kaczor kamil-kaczor kamil-kaczor approved these changes

@iboiko-habana iboiko-habana iboiko-habana approved these changes

@kzawora-intel kzawora-intel Awaiting requested review from kzawora-intel

@xuechendi xuechendi Awaiting requested review from xuechendi xuechendi is a code owner

@mgawarkiewicz-intel mgawarkiewicz-intel Awaiting requested review from mgawarkiewicz-intel mgawarkiewicz-intel is a code owner

@afierka-intel afierka-intel Awaiting requested review from afierka-intel afierka-intel is a code owner

@michalkuligowski michalkuligowski Awaiting requested review from michalkuligowski michalkuligowski is a code owner

@ksmusz ksmusz Awaiting requested review from ksmusz ksmusz is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@adobrzyn @yangulei @kamil-kaczor @iboiko-habana @xuechendi