-
Notifications
You must be signed in to change notification settings - Fork 691
fix: add deepep and pplx pinning commit in vllm repo checkout #3168
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Signed-off-by: Anant Sharma <anants@nvidia.com>
WalkthroughAdds a nvshmem cherry-pick step to the VLLM install script. Introduces early VLLM_REF variable definition. Configures git committer identity, checks out the specified VLLM ref, cherry-picks commit 906e461ed6ddccd3cc7b68fa72048d2d3fcbd72c, and notes TODO to remove this when upgrading VLLM. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant S as Install Script
participant G as Git
participant R as VLLM Repo
participant B as Build/Install
S->>S: Define VLLM_REF
S->>G: Configure committer name/email
S->>G: Clone/Fetch VLLM
S->>G: Checkout VLLM_REF
Note over S,G: Temporary step for nvshmem fix
S->>G: Cherry-pick commit 906e461…bd72c
alt Cherry-pick succeeds
S->>B: Proceed with build/install
B-->>S: Installation complete
else Cherry-pick fails
S-->>S: Exit with error (manual resolution)
end
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Poem
Pre-merge checks❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
Tip 👮 Agentic pre-merge checks are now available in preview!Pro plan users can now enable pre-merge checks in their settings to enforce checklists before merging PRs.
Please see the documentation for more information. Example: reviews:
pre_merge_checks:
custom_checks:
- name: "Undocumented Breaking Changes"
mode: "warning"
instructions: |
Pass/fail criteria: All breaking changes to public APIs, CLI flags, environment variables, configuration keys, database schemas, or HTTP/GraphQL endpoints must be documented in the "Breaking Change" section of the PR description and in CHANGELOG.md. Exclude purely internal or private changes (e.g., code not exported from package entry points or explicitly marked as internal).Please share your feedback with us on this Discord post. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Actionable comments posted: 0
🧹 Nitpick comments (2)
container/deps/vllm/install_vllm.sh (2)
23-23: Fix misleading TODO label (“nvshmem”)This cherry-pick is about pinning pplx (and deepep per PR title), not nvshmem. Update the comment to avoid confusion later.
Apply:
-# REMOVE nvshmem cherry-pick when moving to next version of vllm +# REMOVE pplx pin cherry-pick when moving to next version of vllm
132-136: Make the cherry-pick idempotent and saferGuard the cherry-pick so repeated builds don't fail if the commit is already present; set committer via git -c and keep -x for traceability. Verified: commit 906e461ed6ddccd3cc7b68fa72048d2d3fcbd72c is absent from REF 1da94e673c257373280026f75ceb4effac80e892.
File: container/deps/vllm/install_vllm.sh (lines 132-136)
-# nvshmem fix - cherry-pick commit pinning pplx version -# https://github.com/ai-dynamo/dynamo/actions/runs/17907241473/job/50910654042?pr=2969#step:8:280 -# remove when moving to next version of vllm -# Configure git user for cherry-pick operation -GIT_COMMITTER_NAME="Container Build" GIT_COMMITTER_EMAIL="container@buildkitsandbox.local" git cherry-pick 906e461ed6ddccd3cc7b68fa72048d2d3fcbd72c +# pplx pin: cherry-pick commit that pins pplx; remove on next vLLM bump +# https://github.com/ai-dynamo/dynamo/actions/runs/17907241473/job/50910654042?pr=2969#step:8:280 +PPLX_PIN_COMMIT="906e461ed6ddccd3cc7b68fa72048d2d3fcbd72c" +# Skip if already included in the checked-out ref +if ! git merge-base --is-ancestor "${PPLX_PIN_COMMIT}" HEAD; then + git -c user.name="Container Build" -c user.email="container@buildkitsandbox.local" \ + cherry-pick -x "${PPLX_PIN_COMMIT}" +else + echo "pplx pin already present in ${VLLM_REF}; skipping cherry-pick" +fi
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
container/deps/vllm/install_vllm.sh(2 hunks)
🧰 Additional context used
🧠 Learnings (3)
📚 Learning: 2025-08-18T16:52:15.659Z
Learnt from: nnshah1
PR: ai-dynamo/dynamo#2489
File: container/deps/vllm/install_vllm.sh:151-152
Timestamp: 2025-08-18T16:52:15.659Z
Learning: The VLLM_PRECOMPILED_WHEEL_LOCATION environment variable, when exported, automatically triggers vLLM's build system to use the precompiled wheel instead of building from source, even when using standard `uv pip install .` commands in container/deps/vllm/install_vllm.sh.
Applied to files:
container/deps/vllm/install_vllm.sh
📚 Learning: 2025-08-18T16:52:15.659Z
Learnt from: nnshah1
PR: ai-dynamo/dynamo#2489
File: container/deps/vllm/install_vllm.sh:151-152
Timestamp: 2025-08-18T16:52:15.659Z
Learning: The VLLM_PRECOMPILED_WHEEL_LOCATION environment variable is an official vLLM environment variable that, when exported, automatically triggers vLLM's build system to use the specified precompiled wheel instead of building from source. This works even with standard `uv pip install .` commands without requiring explicit reference to the variable in the install command. The vLLM build system internally detects and uses this environment variable.
Applied to files:
container/deps/vllm/install_vllm.sh
📚 Learning: 2025-07-21T00:10:56.947Z
Learnt from: zaristei
PR: ai-dynamo/dynamo#2020
File: container/deps/vllm/install_vllm.sh:115-118
Timestamp: 2025-07-21T00:10:56.947Z
Learning: Graceful fallback for PyTorch wheel installation is broken on ARM architecture, so immediate exit on pinned version failure is preferred over fallback mechanisms in container/deps/vllm/install_vllm.sh for ARM64.
Applied to files:
container/deps/vllm/install_vllm.sh
dmitrytokarev
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm
Signed-off-by: Anant Sharma <anants@nvidia.com> Signed-off-by: Jason Zhou <jasonzho@nvidia.com>
Signed-off-by: Anant Sharma <anants@nvidia.com>
Signed-off-by: Anant Sharma <anants@nvidia.com> Signed-off-by: Jason Zhou <jasonzho@nvidia.com>
Signed-off-by: Anant Sharma <anants@nvidia.com> Signed-off-by: Kyle H <kylhuang@nvidia.com>
Overview:
Failure - https://github.com/ai-dynamo/dynamo/actions/runs/17907241473/job/50910654042?pr=2969#step:8:280
seems like an issue in vLLM repo scripts dynamo uses this script to install ep-kernels - https://github.com/vllm-project/vllm/blob/v0.10.1.1/tools/ep_kernels/install_python_libraries.sh#L100
and they are not pinning pplx-kernels repo to any commit or tag, so likely this change in upstream repo is causing failures - perplexityai/pplx-kernels@2bd6e30
vLLM fixed pinning issue in later versions but not in the one dynamo uses - vllm-project/vllm@906e461
closes: OPS-1190