Skip to content

Conversation

@pcastonguay
Copy link
Collaborator

@pcastonguay pcastonguay commented Aug 21, 2025

Summary by CodeRabbit

  • Tests

    • Centralized MPI connection and termination logic across integration tests for consistency.
    • Improved error handling with clearer logging, try/except around MPI operations, and safer fallbacks.
    • Added real-time log flushing to aid debugging and visibility during test runs.
    • Unified termination flow to ensure cleaner shutdowns and fewer intermittent failures.
  • Refactor

    • Replaced scattered direct MPI calls with shared helper routines to reduce duplication and simplify maintenance.
  • Chores

    • Minor stability enhancements around worker startup and submission within test workflows.

Description

Test Coverage

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

Details

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

@pcastonguay pcastonguay requested a review from a team as a code owner August 21, 2025 20:13
@coderabbitai
Copy link
Contributor

coderabbitai bot commented Aug 21, 2025

📝 Walkthrough

Walkthrough

Added three MPI helper functions and replaced direct MPI port/accept/termination calls in the test with these helpers; added try/except logging and flushes; retained intercommunicator setup, request/response exchange, and termination semantics while centralizing connectivity and error handling.

Changes

Cohort / File(s) Summary
MPI helper centralization in tests
tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py
Introduced mpi_publish_name, mpi_initialize_intercomm, mpi_send_termination_request; replaced direct MPI.Open_port/MPI.Publish_name/COMM_SELF.Accept and manual termination sends with helpers; added MPI and generic exception handling, log flushing, and minor robustness tweaks around worker startup/submission.

Sequence Diagram(s)

sequenceDiagram
  autonumber
  participant T as Test Case
  participant H as MPI Helpers
  participant M as MPI Runtime

  rect rgb(240,248,255)
  note over T,H: Setup phase (centralized)
  T->>H: mpi_publish_name()
  H->>M: Open_port + Publish_name('my_port')
  M-->>H: port_name
  H-->>T: port_name
  end

  rect rgb(245,255,250)
  note over T,H: Connection phase
  T->>H: mpi_initialize_intercomm(port_name)
  H->>M: COMM_SELF.Accept(port_name)
  M-->>H: intercomm
  H-->>T: intercomm
  end

  rect rgb(255,250,240)
  note over T,M: Test exchanges (unchanged semantics)
  T->>M: Readiness + request/response via intercomm
  M-->>T: Responses
  end

  rect rgb(255,240,245)
  note over T,H: Termination phase (centralized)
  T->>H: mpi_send_termination_request(intercomm)
  H->>M: Send termination to ranks 0 and 1
  M-->>T: Acknowledged/close
  end

  note over T: Errors handled via try/except with logging and flush
Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Suggested reviewers

  • chzblych
  • Tabrizian
  • chuangz0
  • liji-nv

Tip

🔌 Remote MCP (Model Context Protocol) integration is now available!

Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats.


📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

💡 Knowledge Base configuration:

  • MCP integration is disabled by default for public repositories
  • Jira integration is disabled by default for public repositories
  • Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 29fc899 and b445bbf.

📒 Files selected for processing (1)
  • tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py (11 hunks)
🚧 Files skipped from review as they are similar to previous changes (1)
  • tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
  • GitHub Check: Pre-commit Check
✨ Finishing Touches
  • 📝 Generate Docstrings
🧪 Generate unit tests
  • Create PR with unit tests
  • Post copyable unit tests in a comment

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.

❤️ Share
🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

  • Review comments: Directly reply to a review comment made by CodeRabbit. Example:
    • I pushed a fix in commit <commit_id>, please review it.
    • Open a follow-up GitHub issue for this discussion.
  • Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query.
  • PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
    • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
    • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

CodeRabbit Commands (Invoked using PR/Issue comments)

Type @coderabbitai help to get the list of available commands.

Other keywords and placeholders

  • Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
  • Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
  • Add @coderabbitai or @coderabbitai title anywhere in the PR title to generate the title automatically.

Status, Documentation and Community

  • Visit our Status Page to check the current availability of CodeRabbit.
  • Visit our Documentation for detailed information on how to use CodeRabbit.
  • Join our Discord Community to get help, request features, and share feedback.
  • Follow us on X/Twitter for updates and announcements.

@pcastonguay
Copy link
Collaborator Author

/bot run --disable-fail-fast

@pcastonguay pcastonguay requested a review from Tabrizian August 21, 2025 20:13
@tensorrt-cicd
Copy link
Collaborator

PR_Github #16081 [ run ] triggered by Bot

Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

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

Actionable comments posted: 5

🧹 Nitpick comments (6)
tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py (6)

39-52: Add a constant for the MPI service name and lightweight docstring; fix prints/raise usage.

Hardcoding "my_port" in multiple places is brittle. Define a single constant (UPPER_SNAKE_CASE per guidelines) and reuse it across helpers and workers. Also, add a brief docstring and use flush/raise best practices.

Apply within this hunk:

-def mpi_publish_name():
+def mpi_publish_name():
+    """Open an MPI port and publish it under MPI_SERVICE_NAME.
+
+    Returns:
+        str: The MPI port name string.
+    """
     port_name = None
     try:
         port_name = MPI.Open_port()
-        MPI.Publish_name('my_port', port_name)
+        MPI.Publish_name(MPI_SERVICE_NAME, port_name)
     except MPI.Exception as e:
-        print(f"Error publishing port name: {e}")
-        raise e
+        print(f"Error publishing port name: {e}", flush=True)
+        raise
     except Exception as e:
-        print(f"Unexpected error publishing port name: {e}")
-        raise e
+        print(f"Unexpected error publishing port name: {e}", flush=True)
+        raise

And add the constant near the other MPI constants (outside this hunk):

# Single point of truth for the service name used by Publish/Lookup
MPI_SERVICE_NAME = "disagg_test_port"

Also consider adding the NVIDIA copyright header at the top of the file as per repository guidelines. Want me to generate it with the current year?


68-74: Generalize termination to N workers instead of hardcoding 0 and 1.

Use Get_remote_size() to avoid fragile assumptions and ease future scaling.

-def mpi_send_termination_request(intercomm):
-    if intercomm is not None:
-        # Send termination requests
-        intercomm.send(None, dest=0, tag=MPI_REQUEST)
-        intercomm.send(None, dest=1, tag=MPI_REQUEST)
-        print("Sent termination requests to the workers.")
+def mpi_send_termination_request(intercomm):
+    if intercomm is not None:
+        for dst in range(intercomm.Get_remote_size()):
+            intercomm.send(None, dest=dst, tag=MPI_REQUEST)
+        print("Sent termination requests to the workers.", flush=True)

88-97: Fix misleading error messages and use the service-name constant in Lookup.

The current messages say “publishing” while this block performs Lookup/Connect. Use MPI_SERVICE_NAME for consistency with Publish_name.

-    try:
-        port_name = MPI.Lookup_name('my_port')
-        intercomm = MPI.COMM_WORLD.Connect(port_name)
-    except MPI.Exception as e:
-        print(f"Error publishing port name: {e}")
-        raise e
-    except Exception as e:
-        print(f"Unexpected error publishing port name: {e}")
-        raise e
+    try:
+        port_name = MPI.Lookup_name(MPI_SERVICE_NAME)
+        intercomm = MPI.COMM_WORLD.Connect(port_name)
+    except MPI.Exception as e:
+        print(f"Error during Lookup/Connect: {e}", flush=True)
+        raise
+    except Exception as e:
+        print(f"Unexpected error during Lookup/Connect: {e}", flush=True)
+        raise

233-235: Use bare raise to preserve traceback context; include flush.

Minor, but improves debuggability.

-        except Exception as e:
-            print(f"Exception encountered: {e}", flush=True)
-            raise e
+        except Exception as e:
+            print(f"Exception encountered: {e}", flush=True)
+            raise

380-386: Improve MPI exception logging and ensure termination is robust.

Print the exception details and flush. Also relies on intercomm being defined—apply the earlier predefinition.

-        except MPI.Exception as e:
-            print(f"MPI Error")
-            raise e
+        except MPI.Exception as e:
+            print(f"MPI error: {e}", flush=True)
+            raise

After applying, please re-run the relevant tests to confirm no residual leaks or name-collision errors occur when running the whole test module.


481-486: Improve MPI exception logging and keep termination safe.

Same as above: print the exception object for context and use bare raise.

-        except MPI.Exception as e:
-            print(f"MPI Error")
-            raise e
+        except MPI.Exception as e:
+            print(f"MPI error: {e}", flush=True)
+            raise
📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

💡 Knowledge Base configuration:

  • MCP integration is disabled by default for public repositories
  • Jira integration is disabled by default for public repositories
  • Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 1e5a6be and 29fc899.

📒 Files selected for processing (1)
  • tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py (11 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py

📄 CodeRabbit inference engine (CODING_GUIDELINES.md)

**/*.py: Python code must target Python 3.8+
Python indentation: 4 spaces, no tabs
Maintain module namespace in imports (from package.subpackage import foo; then use foo.SomeClass())
Python file names use snake_case
Python class names use PascalCase
Python functions/methods and local variables use snake_case; variables starting with a number get k_ prefix (e.g., k_99th_percentile)
Global variables use G_ prefixed UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE in Python
Avoid shadowing variables from outer scopes in Python
Initialize all externally visible members of a Python class in init
Prefer docstrings for interfaces used outside a file; comments for local code
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes/variables inline with short docstrings
Avoid reflection when simple alternatives exist (e.g., prefer explicit parameters over dict(**locals()))
In try/except, catch the narrowest exceptions possible
For duck-typing with try/except, keep try body minimal and put logic in else

Files:

  • tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py
**/*.{cpp,cxx,cc,cu,h,hpp,hxx,hh,cuh,py}

📄 CodeRabbit inference engine (CODING_GUIDELINES.md)

Prepend NVIDIA copyright header (current year) to all source files

Files:

  • tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
  • GitHub Check: Pre-commit Check
🔇 Additional comments (4)
tests/integration/defs/disaggregated/test_disaggregated_single_gpu.py (4)

186-186: LGTM: helper used for publishing.

Using a helper here improves readability and centralizes error handling. No issues.


241-241: LGTM: explicit “waiting” log with flush improves CI readability.

No action needed.


329-329: LGTM: consistent use of mpi_publish_name().

No issues here.


431-431: LGTM: consistent use of mpi_publish_name().

No issues here.

Copy link
Member

@Tabrizian Tabrizian left a comment

Choose a reason for hiding this comment

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

Thanks for the refactor!

@tensorrt-cicd
Copy link
Collaborator

PR_Github #16081 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #254 completed with status: 'FAILURE'

@pcastonguay pcastonguay force-pushed the disagg_test_mpi_init_fix branch from 29fc899 to b445bbf Compare August 22, 2025 16:00
@pcastonguay
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #16200 [ run ] triggered by Bot

@pcastonguay
Copy link
Collaborator Author

Could someone from @NVIDIA/trt-llm-release-branch-approval please review? Thanks.

@MartinMarciniszyn MartinMarciniszyn enabled auto-merge (squash) August 22, 2025 16:42
@pcastonguay
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #16463 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #16463 [ run ] completed with state SUCCESS
/LLM/release-1.0/L0_MergeRequest_PR pipeline #300 completed with status: 'SUCCESS'

@MartinMarciniszyn MartinMarciniszyn merged commit 9d2b181 into NVIDIA:release/1.0 Aug 25, 2025
5 checks passed
yuanjingx87 pushed a commit that referenced this pull request Aug 28, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Sep 5, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Sep 5, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Sep 6, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Sep 6, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Sep 7, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants