Skip to content

[PyTorch][Test-Fix] Scale test_multi_process_ep outer timeout with launcher pass count - #3415

Merged
fheinecke merged 3 commits into
NVIDIA:mainfrom
phu0ngng:fix/ep-test-outer-timeout
Aug 22, 2026
Merged

[PyTorch][Test-Fix] Scale test_multi_process_ep outer timeout with launcher pass count#3415
fheinecke merged 3 commits into
NVIDIA:mainfrom
phu0ngng:fix/ep-test-outer-timeout

Conversation

@phu0ngng

Copy link
Copy Markdown
Collaborator

Description

Please include a brief summary of the changes, relevant motivation and context.

Fixes # (issue)

Type of change

  • Documentation change (change only to the documentation, either a fix or a new content)
  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • Infra/Build change
  • Code refactoring

Changes

Please list the changes introduced in this PR:

  • Change A
  • Change B

Checklist:

  • I have read and followed the contributing guidelines
  • The functionality is complete
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • My changes generate no new warnings
  • I have added tests that prove my fix is effective or that my feature works
  • New and existing unit tests pass locally with my changes

…s count
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>
@phu0ngng

Copy link
Copy Markdown
CollaboratorAuthor

/te-ci L1 pytorch

@greptile-apps

greptile-appsBot commented Aug 21, 2026

Copy link
Copy Markdown
Contributor

Greptile Summary

The PR scales the PyTorch expert-parallel test’s outer subprocess timeout according to the number of launcher passes.

  • Adds a helper that counts the launcher’s seven run_pass invocations.
  • Retains the configured timeout as a per-pass budget.
  • Adds a fixed allowance for launcher overhead.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains.

Important Files Changed

FilenameOverview
tests/pytorch/distributed/test_ep.pyUpdates the EP test harness so its outer timeout reflects the launcher’s sequential pass count; no follow-up-eligible blocking issue was identified.

Reviews (2): Last reviewed commit: "Merge branch 'main' into fix/ep-test-out..." | Re-trigger Greptile

@fheinecke

Copy link
Copy Markdown
Collaborator

/te-ci L1 pytorch

@fheinecke
fheinecke merged commit d0067d0 into NVIDIA:mainAug 22, 2026
12 of 18 checks passed
fheinecke added a commit that referenced this pull request Aug 22, 2026
…uncher pass count (#3415)
[PyTorch] Scale test_multi_process_ep outer timeout with launcher pass count
Signed-off-by: Phuong Nguyen <phuonguyen@nvidia.com>
Co-authored-by: fheinecke <23390735+fheinecke@users.noreply.github.com>
(cherry picked from commit d0067d0)
Sign up for freeto join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

@phu0ngng@fheinecke