Skip to content

Conversation

@chuanqi129
Copy link
Contributor

No description provided.

Copilot AI review requested due to automatic review settings January 26, 2026 10:15
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull request overview

This PR introduces a new GitHub Actions workflow for running XPU (Intel GPU) tests on the torchcomms repository. The workflow is designed to execute on pull requests and manual triggers, running tests in a containerized environment with XPU device access.

Changes:

  • Added comprehensive XPU testing workflow with container-based test execution
  • Implemented health checks for runner GPU availability and disk space
  • Configured Docker container setup with XPU device access and proper permissions

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

working-directory: pytorch
run: |
DOCKER_BUILD_DIR=".ci/docker"
DOCKER_IMAGE_NAME=${${env.DOCKER_IMAGE}-}$(git rev-parse HEAD:"${DOCKER_BUILD_DIR}")
Copy link

Copilot AI Jan 26, 2026

Choose a reason for hiding this comment

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

Invalid shell parameter expansion syntax. The expression ${${env.DOCKER_IMAGE}-} contains nested parameter expansions which is not valid in bash. This should be ${DOCKER_IMAGE:-} to provide an empty default value if DOCKER_IMAGE is unset.

Suggested change
DOCKER_IMAGE_NAME=${${env.DOCKER_IMAGE}-}$(git rev-parse HEAD:"${DOCKER_BUILD_DIR}")
DOCKER_IMAGE_NAME=${DOCKER_IMAGE:-}$(git rev-parse HEAD:"${DOCKER_BUILD_DIR}")

Copilot uses AI. Check for mistakes.
Comment on lines 86 to 87
with:
working-directory: pytorch
Copy link

Copilot AI Jan 26, 2026

Choose a reason for hiding this comment

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

The 'with' keyword is invalid for a run step. The 'with' keyword is used for action inputs, but this is a 'run' step executing shell commands. Remove the 'with' block or convert this to use an action that accepts working-directory as an input.

Suggested change
with:
working-directory: pytorch
working-directory: pytorch

Copilot uses AI. Check for mistakes.
--shm-size="8g" \
--tty \
--detach \
--name="${container_name}" \
Copy link

Copilot AI Jan 26, 2026

Choose a reason for hiding this comment

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

The variable 'container_name' is used before it is defined. The docker run command uses '--name="${container_name}"' but the output is assigned to container_name afterward. This will fail because the variable is empty when used. Remove the --name flag and let Docker generate a name, or generate the name before the docker run command.

Suggested change
--name="${container_name}" \

Copilot uses AI. Check for mistakes.
-e JOB_ID \
-e BRANCH \
-e SHA1 \
--user $(id -u):$(id -g) \
Copy link

Copilot AI Jan 26, 2026

Choose a reason for hiding this comment

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

User is specified twice in the docker run command. Line 131 sets '--user $(id -u):$(id -g)' and line 140 sets '--user jenkins'. This creates a conflict where the second specification overrides the first. Remove one of these --user flags.

Suggested change
--user $(id -u):$(id -g) \

Copilot uses AI. Check for mistakes.
@chuanqi129 chuanqi129 force-pushed the test_wf branch 5 times, most recently from 19d33da to 1eedc5d Compare January 26, 2026 10:25
@chuanqi129 chuanqi129 marked this pull request as draft January 26, 2026 10:35
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