name: E2E-upstream
on:
schedule:
- cron: '0 8 * * 0'
defaults:
run:
shell: bash -el {0}
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: true
env:
UV_INDEX_URL: http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
UV_EXTRA_INDEX_URL: https://repo.huaweicloud.com/ascend/repos/pypi
UV_INDEX_STRATEGY: unsafe-best-match
UV_NO_CACHE: 1
UV_SYSTEM_PYTHON: 1
jobs:
e2e-upstream_singlecard:
runs-on: linux-aarch64-a2b3-1
strategy:
fail-fast: false
matrix:
part: [0, 1, 2, 3]
vllm: [v0.20.2]
container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:9.0.0-910b-ubuntu22.04-py3.11
env:
VLLM_LOGGING_LEVEL: ERROR
HF_HUB_OFFLINE: 1
steps:
- name: Checkout vllm-project/vllm-ascend repo
uses: actions/checkout@v6
- name: Check npu and CANN info
run: |
npu-smi info
cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
- name: Config mirrors
run: |
sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
apt-get update -y
apt install git -y
- name: Install system dependencies
run: |
apt-get -y install `cat packages.txt`
apt-get -y install gcc g++ cmake libnuma-dev clang-15
git config --global --add safe.directory /__w/vllm-ascend/vllm-ascend
update-alternatives --install /usr/bin/clang clang /usr/bin/clang-15 20
update-alternatives --install /usr/bin/clang++ clang++ /usr/bin/clang++-15 20
- name: Checkout vllm-project/vllm repo
uses: actions/checkout@v6
with:
repository: vllm-project/vllm
ref: ${{ matrix.vllm }}
path: ./vllm-empty
fetch-depth: 1
- name: Install vllm-project/vllm from source
working-directory: ./vllm-empty
run: |
VLLM_TARGET_DEVICE=empty pip install -e .
pip install -e tests/plugins/bge_m3_sparse_plugin
pip install -e tests/plugins/prithvi_io_processor_plugin
sed -i '5i\import vllm_ascend.patch.platform\nimport vllm_ascend.patch.worker' tests/conftest.py
- name: Install vllm-project/vllm-ascend
env:
PIP_EXTRA_INDEX_URL: "https://repo.huaweicloud.com/ascend/repos/pypi https://triton-ascend.osinfra.cn/pypi/simple https://download.pytorch.org/whl/cpu/"
run: |
pip install uc-manager
pip install -r requirements-dev.txt
pip install -e .
pip install tblib
pip install runai_model_streamer pqdm timm schemathesis==3.39.15 "vllm[helion]" "runai-model-streamer[s3,gcs]"
pip install "mteb[bm25s]" open_clip_torch terratorch==1.2.2 imagehash
pip install "lm-eval[api]" num2words "vllm[grpc]" "vllm[audio]"
pip uninstall -y triton
pip uninstall -y triton-ascend
pip install triton-ascend==3.2.1 --extra-index-url https://triton-ascend.osinfra.cn/pypi/simple
- name: Run vllm-project/vllm test
working-directory: ./vllm-empty
env:
PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
VLLM_WORKER_MULTIPROC_METHOD: spawn
run: |
mkdir -p tiktoken_encodings
curl -L -o tiktoken_encodings/o200k_base.tiktoken "https://openaipublic.blob.core.windows.net/encodings/o200k_base.tiktoken"
curl -L -o tiktoken_encodings/cl100k_base.tiktoken "https://openaipublic.blob.core.windows.net/encodings/cl100k_base.tiktoken"
export TIKTOKEN_ENCODINGS_BASE=${PWD}/tiktoken_encodings
python3 ../.github/workflows/scripts/run_suite.py \
--suite e2e-upstream_singlecard \
--auto-partition-id "${{ matrix.part }}" \
--auto-partition-size 4 \
--continue-on-error
e2e-upstream_a2_2:
runs-on: linux-aarch64-a2b3-2
strategy:
fail-fast: false
matrix:
part: [0]
vllm: [v0.20.2]
container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:9.0.0-910b-ubuntu22.04-py3.11
env:
VLLM_LOGGING_LEVEL: ERROR
HF_HUB_OFFLINE: 1
steps:
- name: Checkout vllm-project/vllm-ascend repo
uses: actions/checkout@v6
- name: Check npu and CANN info
run: |
npu-smi info
cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
- name: Config mirrors
run: |
sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
apt-get update -y
apt install git -y
- name: Install system dependencies
run: |
apt-get -y install `cat packages.txt`
apt-get -y install gcc g++ cmake libnuma-dev clang-15
git config --global --add safe.directory /__w/vllm-ascend/vllm-ascend
update-alternatives --install /usr/bin/clang clang /usr/bin/clang-15 20
update-alternatives --install /usr/bin/clang++ clang++ /usr/bin/clang++-15 20
- name: Checkout vllm-project/vllm repo
uses: actions/checkout@v6
with:
repository: vllm-project/vllm
ref: ${{ matrix.vllm }}
path: ./vllm-empty
fetch-depth: 1
- name: Install vllm-project/vllm from source
working-directory: ./vllm-empty
run: |
VLLM_TARGET_DEVICE=empty pip install -e .
sed -i '5i\import vllm_ascend.patch.platform\nimport vllm_ascend.patch.worker' tests/conftest.py
- name: Install vllm-project/vllm-ascend
env:
PIP_EXTRA_INDEX_URL: "https://repo.huaweicloud.com/ascend/repos/pypi https://triton-ascend.osinfra.cn/pypi/simple https://download.pytorch.org/whl/cpu/"
run: |
pip install uc-manager
pip install -r requirements-dev.txt
pip install -e .
pip install tblib
pip uninstall -y triton-ascend
pip uninstall -y triton
pip install triton-ascend==3.2.1 --extra-index-url https://triton-ascend.osinfra.cn/pypi/simple
- name: Run vllm-project/vllm test
working-directory: ./vllm-empty
env:
PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
VLLM_WORKER_MULTIPROC_METHOD: spawn
run: |
python3 ../.github/workflows/scripts/run_suite.py \
--suite e2e-upstream_a2_2 \
--auto-partition-id "${{ matrix.part }}" \
--auto-partition-size 1 \
--continue-on-error
e2e-upstream_a2_4:
runs-on: linux-aarch64-a2b3-4
strategy:
fail-fast: false
matrix:
part: [0]
vllm: [v0.20.2]
container:
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:9.0.0-910b-ubuntu22.04-py3.11
env:
VLLM_LOGGING_LEVEL: ERROR
HF_HUB_OFFLINE: 1
steps:
- name: Checkout vllm-project/vllm-ascend repo
uses: actions/checkout@v6
- name: Check npu and CANN info
run: |
npu-smi info
cat /usr/local/Ascend/ascend-toolkit/latest/"$(uname -i)"-linux/ascend_toolkit_install.info
- name: Config mirrors
run: |
sed -Ei 's@(ports|archive).ubuntu.com@cache-service.nginx-pypi-cache.svc.cluster.local:8081@g' /etc/apt/sources.list
pip config set global.index-url http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple
pip config set global.trusted-host cache-service.nginx-pypi-cache.svc.cluster.local
apt-get update -y
apt install git -y
- name: Install system dependencies
run: |
apt-get -y install `cat packages.txt`
apt-get -y install gcc g++ cmake libnuma-dev clang-15
git config --global --add safe.directory /__w/vllm-ascend/vllm-ascend
update-alternatives --install /usr/bin/clang clang /usr/bin/clang-15 20
update-alternatives --install /usr/bin/clang++ clang++ /usr/bin/clang++-15 20
- name: Checkout vllm-project/vllm repo
uses: actions/checkout@v6
with:
repository: vllm-project/vllm
ref: ${{ matrix.vllm }}
path: ./vllm-empty
fetch-depth: 1
- name: Install vllm-project/vllm from source
working-directory: ./vllm-empty
run: |
VLLM_TARGET_DEVICE=empty pip install -e .
sed -i '5i\import vllm_ascend.patch.platform\nimport vllm_ascend.patch.worker' tests/conftest.py
- name: Install vllm-project/vllm-ascend
env:
PIP_EXTRA_INDEX_URL: "https://repo.huaweicloud.com/ascend/repos/pypi https://triton-ascend.osinfra.cn/pypi/simple https://download.pytorch.org/whl/cpu/"
run: |
pip install uc-manager
pip install -r requirements-dev.txt
pip install -e .
pip install tblib
pip uninstall -y triton-ascend
pip uninstall -y triton
pip install triton-ascend==3.2.1 --extra-index-url https://triton-ascend.osinfra.cn/pypi/simple
- name: Run vllm-project/vllm test
working-directory: ./vllm-empty
env:
PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
VLLM_WORKER_MULTIPROC_METHOD: spawn
run: |
python3 ../.github/workflows/scripts/run_suite.py \
--suite e2e-upstream_a2_4 \
--auto-partition-id "${{ matrix.part }}" \
--auto-partition-size 1 \
--continue-on-error