Skip to content

[TRTLLM-12893][infra] Parallelize post stages: Rerun Report, Test Coverage, and AI Failure Analysis - #14528

Merged
ZhanruiSunCh merged 2 commits into
NVIDIA:mainfrom
ZhanruiSunCh:user/zhanruis/parallelize_post_stages
Jun 5, 2026
Merged

[TRTLLM-12893][infra] Parallelize post stages: Rerun Report, Test Coverage, and AI Failure Analysis#14528
ZhanruiSunCh merged 2 commits into
NVIDIA:mainfrom
ZhanruiSunCh:user/zhanruis/parallelize_post_stages

Conversation

@ZhanruiSunCh

@ZhanruiSunCh ZhanruiSunCh commented May 25, 2026

Copy link
Copy Markdown
Collaborator

Summary by CodeRabbit

  • Refactor
    • Restructured build pipeline test result processing workflow for improved efficiency.
    • Implemented parallel execution of test coverage reporting, test rerun analysis, and failure diagnostics for faster feedback cycles.
    • Pre-installed shared dependencies to optimize build performance.
    • Added build information artifact for enhanced build diagnostics and troubleshooting.

Review Change Stack

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

  • PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

  • PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

  • Test cases are provided for new code paths (see test instructions)

  • If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

  • Any new dependencies have been scanned for license and vulnerabilities

  • CODEOWNERS updated if ownership changes

  • Documentation updated as needed

  • Update tava architecture diagram if there is a significant design change in PR.

  • The reviewers assigned automatically/manually are appropriate for the PR.

  • Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

@ZhanruiSunCh
ZhanruiSunCh requested a review from a team May 25, 2026 09:29
@ZhanruiSunCh
ZhanruiSunCh requested a review from a team as a code owner May 25, 2026 09:29
@ZhanruiSunCh

Copy link
Copy Markdown
Collaborator Author

/bot run

@coderabbitai

coderabbitai Bot commented May 25, 2026

Copy link
Copy Markdown
Contributor
📝 Walkthrough

Walkthrough

The pipeline refactors test result collection by updating collectTestResults to accept globalVars, reorganizing test handling to parallelize report generation, conditionally include AI failure analysis on build failures, and update the caller to pass the new parameter through post.always.

Changes

Test Result Collection and Processing Refactoring

Layer / File(s) Summary
collectTestResults function signature and globalVars parameter
jenkins/L0_MergeRequest.groovy
collectTestResults signature updated to accept globalVars as a third parameter.
Test result handling and parallel task orchestration
jenkins/L0_MergeRequest.groovy
Test result processing is refactored to pre-install dependencies, upload build info, parallelize rerun-report and test-coverage tasks, and conditionally execute AI failure analysis when the build has failed.
Pipeline post.always integration with updated collectTestResults
jenkins/L0_MergeRequest.groovy
The pipeline's post.always block is updated to call collectTestResults(this, testFilter, globalVars) with the new parameter signature.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

Suggested reviewers

  • tburt-nv
  • mzweilz
  • yuanjingx87
🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description body is empty—only the template structure remains without any content in the Description or Test Coverage sections. Add a clear description explaining the issue, solution, and test coverage. At minimum, describe why parallelization is needed and what tests validate the changes.
✅ Passed checks (4 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly identifies the JIRA ticket, type, and main change: parallelizing post stages (Rerun Report, Test Coverage, AI Failure Analysis) in the pipeline.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

Comment @coderabbitai help to get the list of available commands and usage tips.

@coderabbitai coderabbitai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

🧹 Nitpick comments (1)
jenkins/L0_MergeRequest.groovy (1)

1257-1259: 💤 Low value

Consider logging the exception for debuggability.

The empty catch block silently swallows all exceptions. While the best-effort approach is appropriate, adding minimal logging would help diagnose issues when they occur.

Suggested improvement
                     } catch (Exception e) {
-                        // Analysis is best-effort; do not fail the pipeline
+                        // Analysis is best-effort; do not fail the pipeline
+                        echo "AI Failure Analysis skipped due to error: ${e.toString()}"
                     }
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@jenkins/L0_MergeRequest.groovy` around lines 1257 - 1259, The empty catch
(Exception e) block silently swallows failures from the analysis try/catch;
update that catch to log the exception (message and stacktrace) for
debuggability. Inside the catch referencing the existing catch variable e, call
the pipeline logger if available (e.g., logger.error(...) or
currentBuild/steps.echo) or fallback to println, and include
e.toString()/e.getMessage() plus the stacktrace (e.printStackTrace() or logger
with throwable) so the analysis failure is visible without failing the pipeline.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@jenkins/L0_MergeRequest.groovy`:
- Around line 1257-1259: The empty catch (Exception e) block silently swallows
failures from the analysis try/catch; update that catch to log the exception
(message and stacktrace) for debuggability. Inside the catch referencing the
existing catch variable e, call the pipeline logger if available (e.g.,
logger.error(...) or currentBuild/steps.echo) or fallback to println, and
include e.toString()/e.getMessage() plus the stacktrace (e.printStackTrace() or
logger with throwable) so the analysis failure is visible without failing the
pipeline.

ℹ️ Review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: e856f02e-cf40-4698-aea7-a11a1aa0c116

📥 Commits

Reviewing files that changed from the base of the PR and between 2e3a75c and 31f3ab0.

📒 Files selected for processing (1)
  • jenkins/L0_MergeRequest.groovy

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #50198 [ run ] triggered by Bot. Commit: 31f3ab0 Link to invocation

@ZhanruiSunCh
ZhanruiSunCh force-pushed the user/zhanruis/parallelize_post_stages branch from 31f3ab0 to 9559d3f Compare May 25, 2026 09:37
@ZhanruiSunCh

Copy link
Copy Markdown
Collaborator Author

/bot run

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #50201 [ run ] triggered by Bot. Commit: 9559d3f Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #50198 [ run ] completed with state ABORTED. Commit: 31f3ab0

Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #50201 [ run ] completed with state SUCCESS. Commit: 9559d3f
/LLM/main/L0_MergeRequest_PR pipeline #39740 completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

CI Report

Link to invocation

Comment thread jenkins/L0_MergeRequest.groovy
@ZhanruiSunCh
ZhanruiSunCh force-pushed the user/zhanruis/parallelize_post_stages branch from 9559d3f to 5478a96 Compare June 3, 2026 08:09
ZhanruiSunCh and others added 2 commits June 3, 2026 01:22
…erage, and AI Failure Analysis

Restructure the post block in L0_MergeRequest to run Rerun Report,
Test Coverage, and AI Failure Analysis in parallel after Collect Test
Result and Upload Build Info complete serially. This reduces post stage
wall time by ~3.5min on failed builds with coverage data.

- Refactor collectTestResults to run parallel tasks inside the pod
- Move Upload Build Info into collectTestResults (runs 2nd, serial)
- Move AI Failure Analysis from failure block into always block,
  conditionally triggered via currentBuild.currentResult
- Remove container("alpine") wrapper (already in alpine pod)
- Pre-install shared dependencies (py3-pip) in serial phase to
  avoid apk concurrent install conflicts

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
Add 10-min timeout to each parallel branch (Rerun Report, Test
Coverage, AI Failure Analysis) with outer try-catch to ensure
timeout or failure in any branch does not affect the pipeline result.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
@ZhanruiSunCh
ZhanruiSunCh force-pushed the user/zhanruis/parallelize_post_stages branch from 5478a96 to fb4427e Compare June 3, 2026 08:28
@ZhanruiSunCh
ZhanruiSunCh enabled auto-merge (squash) June 3, 2026 08:30
@ZhanruiSunCh

Copy link
Copy Markdown
Collaborator Author

/bot run

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #51805 [ run ] triggered by Bot. Commit: fb4427e Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #51805 [ run ] completed with state SUCCESS. Commit: fb4427e
/LLM/main/L0_MergeRequest_PR pipeline #41168 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@ZhanruiSunCh

Copy link
Copy Markdown
Collaborator Author

/bot run

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #51944 [ run ] triggered by Bot. Commit: fb4427e Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #51944 [ run ] completed with state FAILURE. Commit: fb4427e
/LLM/main/L0_MergeRequest_PR pipeline #41295 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

Link to invocation

@ZhanruiSunCh

Copy link
Copy Markdown
Collaborator Author

/bot skip --comment "Failed not related this PR change."

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #52331 [ skip ] triggered by Bot. Commit: fb4427e Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #52331 [ skip ] completed with state SUCCESS. Commit: fb4427e
Skipping testing for commit fb4427e

Link to invocation

@ZhanruiSunCh
ZhanruiSunCh merged commit 6387eac into NVIDIA:main Jun 5, 2026
8 checks passed
fbxai pushed a commit to fbxai/TensorRT-LLM that referenced this pull request Jun 5, 2026
…erage, and AI Failure Analysis (NVIDIA#14528)

Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: NVFB <186336021+NVFB@users.noreply.github.com>
2ez4bz pushed a commit to 2ez4bz/TensorRT-LLM that referenced this pull request Jun 8, 2026
…erage, and AI Failure Analysis (NVIDIA#14528)

Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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.

3 participants