Skip to content

Enable linux_job.yml workflow for ROCm#6003

Closed
amdfaa wants to merge 0 commit intomainfrom
rocm_experiment
Closed

Enable linux_job.yml workflow for ROCm#6003
amdfaa wants to merge 0 commit intomainfrom
rocm_experiment

Conversation

@amdfaa
Copy link
Copy Markdown
Collaborator

@amdfaa amdfaa commented Dec 2, 2024

Relates to pytorch/ao#999

This PR has been closed in favor of: #6104

@vercel
Copy link
Copy Markdown

vercel bot commented Dec 2, 2024

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name Status Preview Updated (UTC)
torchci ❌ Failed (Inspect) Dec 20, 2024 5:07pm

@pytorch-bot
Copy link
Copy Markdown

pytorch-bot bot commented Dec 2, 2024

No ciflow labels are configured for this repo.
For information on how to enable CIFlow bot see this wiki

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Dec 2, 2024
@amdfaa amdfaa changed the title Rocm experiment Enable Torchao workflow Dec 2, 2024
@jithunnair-amd jithunnair-amd changed the title Enable Torchao workflow Enable linux_job.yml workflow for ROCm Dec 19, 2024
if: ${{ inputs.rocm == false }}

- name: Setup ROCm
uses: amdfaa/pytorch/.github/actions/setup-rocm@torchao_experiment
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Temporary change until pytorch/pytorch#143590 is merged

env_file = f"{ os.environ.get('RUNNER_TEMP', '') }/github_env_{ os.environ.get('GITHUB_RUN_ID', '') }"
if f"{ os.environ.get('ROCM', '') }" == "true":
env_file = f"/tmp/github_env_{ os.environ.get('GITHUB_RUN_ID', '') }"

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

This shouldn't be needed if we update setup-rocm.yml to export all the required env vars to ${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID} like in actions/setup-linux/action.yml

-v "{ os.environ.get('RUNNER_TEST_RESULTS_DIR', '') }:/test-results" \
-v "{ os.environ.get('RUNNER_TEMP', '') }/artifacts:/artifacts" \
-v "{ os.environ.get('RUNNER_TEMP', '') }/docs:/docs" \
-v "{ os.environ.get('RUNNER_TEMP', '') }/test-results:/test-results" \
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Let's define these RUNNER_ env vars in setup-rocm.yml similar to

RUNNER_ARTIFACT_DIR="${RUNNER_TEMP}/artifacts"
, so we don't have to change the code here, since it's commonly used by both ROCm and non-ROCm workflows.

docker-image:
description: Identifies the Docker image by name.
default: "pytorch/conda-builder"
default: "pytorch/almalinux-builder"
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Let's make changes to linux_job_v2.yml instead, which already has these almalinux-based changes, so the only updates would be for the ROCm conditionalization.

pytorchmergebot pushed a commit to pytorch/pytorch that referenced this pull request Dec 20, 2024
Needed for pytorch/test-infra#6003 and pytorch/ao#999

Pull Request resolved: #143590
Approved by: https://github.com/atalman

Co-authored-by: Jithun Nair <37884920+jithunnair-amd@users.noreply.github.com>
@amdfaa amdfaa closed this Dec 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/rocm CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. module: rocm

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants