Skip to content

Comments

Run single and multi GPU tests in parallel and upload combined results#257

Open
psanal35 wants to merge 4 commits intomasterfrom
ci-nightly-parallel-single-multi
Open

Run single and multi GPU tests in parallel and upload combined results#257
psanal35 wants to merge 4 commits intomasterfrom
ci-nightly-parallel-single-multi

Conversation

@psanal35
Copy link
Contributor

Previously, each uploaded log represented a full test suite run (single+multi GPU). To preserve this model while enabling parallel execution, single and multi GPU tests are now run as separate parallel jobs and their results are combined before uploading.

docker pull "$IMAGE"
- name: Run multi GPU tests
env:
GPU_COUNT: "8"
Copy link
Contributor

Choose a reason for hiding this comment

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

shouldn't this be 4? Anyway, I think a previous patch updated run_single_gpu.py to detect the number of GPUs with rocinfo. So is there another reason to pass GPU_COUNT?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

This env variable doesn't seem to have any effect, so I'm not changing for now. There is another one in ci.yml that should be updated.

@psanal35 psanal35 force-pushed the ci-nightly-parallel-single-multi branch 2 times, most recently from 79c25c2 to 9692a0b Compare January 23, 2026 01:47
@psanal35 psanal35 force-pushed the ci-nightly-parallel-single-multi branch from 864887b to 165e75a Compare January 23, 2026 20:01
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.

2 participants