Test GPU #20

	# Copyright 2024 Advanced Micro Devices, Inc.
	#
	# Licensed under the Apache License v2.0 with LLVM Exceptions.
	# See https://llvm.org/LICENSE.txt for license information.
	# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception

	name: Test GPU
	on:
	pull_request:
	workflow_dispatch:

	concurrency:
	# A PR number if a pull request and otherwise the commit hash. This cancels
	# queued and in-progress runs for the same PR (presubmit) or commit
	# (postsubmit). The workflow name is prepended to avoid conflicts between
	# different workflows.
	group: ${{ github.workflow }}-${{ github.event.number \|\| github.sha }}
	cancel-in-progress: true

	jobs:
	test-gpu:
	name: "Test AMD GPU Cluster"
	runs-on: linux-mi300
	steps:
	- name: "Checking out repository"
	uses: actions/checkout@v4

	- name: "Setting up Python"
	uses: actions/setup-python@75f3110429a8c05be0e1bf360334e4cced2b63fa # v2.3.3
	with:
	python-version: 3.11

	# Test rocm
	- name: "rocm test"
	run: \|
	ls -a /home/runner
	ls -a /etc/podinfo
	cat /etc/podinfo/gha-render-devices
	if [ -f "/etc/podinfo/gha-render-devices" ]; then
	DEVICE_FLAG=$(cat /etc/podinfo/gha-render-devices)
	else
	DEVICE_FLAG="--device /dev/dri"
	fi
	echo $DEVICE_FLAG
	docker pull rocm/dev-ubuntu-22.04:6.3
	docker run -dt --network=host --device=/dev/kfd $DEVICE_FLAG \
	--group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined \
	--name ci_container_test \
	rocm/dev-ubuntu-22.04:6.3

	# Test rocm commands
	- name: "Test rocm commands"
	run: \|
	docker exec ci_container_test rocm-smi
	docker exec ci_container_test rocminfo

Provide feedback