TensorRT-LLM Engine integration #5577

	name: CI GPU

	on:
	workflow_dispatch:
	push:
	branches:
	- master
	pull_request:
	branches:
	- master
	merge_group:

	concurrency:
	group: ci-gpu-${{ github.workflow }}-${{ github.ref == 'refs/heads/master' && github.run_number \|\| github.ref }}
	cancel-in-progress: true

	jobs:
	ci-gpu:
	runs-on: [self-hosted, ci-gpu]
	steps:
	- name: Clean up previous run
	run: \|
	echo "Cleaning up previous run"
	ls -la ./
	# deletes all files in the current directory
	sudo rm -rf ./* \|\| true
	# deletes all hidden files in a directory
	sudo rm -rf ./.??* \|\| true
	ls -la ./
	- name: Setup Python 3.9
	uses: actions/setup-python@v5
	with:
	python-version: 3.9
	architecture: x64
	- name: Setup Java 17
	uses: actions/setup-java@v3
	with:
	distribution: 'zulu'
	java-version: '17'
	- name: Checkout TorchServe
	uses: actions/checkout@v3
	with:
	submodules: recursive
	- name: Install dependencies
	run: \|
	python ts_scripts/install_dependencies.py --environment=dev --cuda=cu121
	- name: Torchserve Sanity
	uses: nick-fields/retry@v3
	with:
	timeout_minutes: 60
	retry_on: error
	max_attempts: 3
	command: \|
	python torchserve_sanity.py

	# Any coverage.xml will be picked up by this step
	# Just make sure each coverage.xml is in a different folder
	- name: Upload codecov
	run : \|
	curl -Os https://uploader.codecov.io/latest/linux/codecov
	chmod +x codecov
	./codecov

Provide feedback