[NPU] Add PR test cases (#32392)
Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: sglang-npu-bot <sglangnpu@163.com>
This commit is contained in:
Sugar920
co-authored by
Claude
sglang-npu-bot
Claude
sglang-npu-bot
parent
27b15349e5
commit
f8e62a9224
@@ -236,10 +236,10 @@ jobs:
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-2-npu-a2 --auto-partition-id ${{ matrix.part }} --auto-partition-size 2
|
||||
|
||||
stage-b-test-4-npu-a3:
|
||||
stage-b-test-1-npu-a3:
|
||||
needs: [check-changes, pr-gate, set-image-config]
|
||||
if: needs.check-changes.outputs.main_package == 'true'
|
||||
runs-on: linux-aarch64-a3-4
|
||||
runs-on: linux-aarch64-a3-2
|
||||
container:
|
||||
image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
|
||||
steps:
|
||||
@@ -283,13 +283,13 @@ jobs:
|
||||
STREAMS_PER_DEVICE: 32
|
||||
run: |
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-4-npu-a3 --timeout-per-file 3600
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-1-npu-a3 --timeout-per-file 3600
|
||||
|
||||
|
||||
stage-b-test-16-npu-a3:
|
||||
stage-b-test-2-npu-a3:
|
||||
needs: [check-changes, pr-gate, set-image-config]
|
||||
if: needs.check-changes.outputs.main_package == 'true'
|
||||
runs-on: linux-aarch64-a3-16
|
||||
runs-on: linux-aarch64-a3-2
|
||||
container:
|
||||
image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
|
||||
steps:
|
||||
@@ -332,6 +332,161 @@ jobs:
|
||||
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
|
||||
STREAMS_PER_DEVICE: 32
|
||||
run: |
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-2-npu-a3 --timeout-per-file 3600
|
||||
|
||||
|
||||
stage-b-test-4-npu-a3:
|
||||
needs: [check-changes, pr-gate, set-image-config]
|
||||
if: needs.check-changes.outputs.main_package == 'true'
|
||||
runs-on: linux-aarch64-a3-4
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
part: [ 0, 1 ]
|
||||
container:
|
||||
image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
|
||||
steps:
|
||||
- name: Checkout code
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
ref: ${{ inputs.ref || github.ref }}
|
||||
|
||||
- name: Mark repository safe
|
||||
run: |
|
||||
git config --system --add safe.directory ${GITHUB_WORKSPACE}
|
||||
|
||||
- name: Install dependencies
|
||||
env:
|
||||
TORCH_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/whl/cpu"
|
||||
PYPI_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
UV_INDEX_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
GITHUB_PROXY_URL: "https://gh-proxy.test.osinfra.cn/"
|
||||
RUSTUP_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local:8082"
|
||||
run: |
|
||||
# speed up by using infra cache services
|
||||
CACHING_URL="cache-service.nginx-pypi-cache.svc.cluster.local"
|
||||
sed -Ei "s@(ports|archive).ubuntu.com@${CACHING_URL}:8081@g" /etc/apt/sources.list
|
||||
pip config set global.index-url http://${CACHING_URL}/pypi/simple
|
||||
pip config set global.trusted-host "${CACHING_URL}"
|
||||
|
||||
bash scripts/ci/npu/npu_ci_install_dependency.sh a3
|
||||
# copy required file from our daily cache
|
||||
cp ~/.cache/modelscope/hub/datasets/otavia/ShareGPT_Vicuna_unfiltered/ShareGPT_V3_unfiltered_cleaned_split.json /tmp
|
||||
# copy gsm8k dataset
|
||||
cp ~/.cache/modelscope/hub/datasets/tmp/test.jsonl /tmp
|
||||
|
||||
- name: Run test
|
||||
timeout-minutes: 120
|
||||
env:
|
||||
SGLANG_USE_MODELSCOPE: true
|
||||
SGLANG_IS_IN_CI: true
|
||||
HF_ENDPOINT: https://hf-mirror.com
|
||||
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
|
||||
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
|
||||
STREAMS_PER_DEVICE: 32
|
||||
run: |
|
||||
pip install sglang_router
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-4-npu-a3 --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 --timeout-per-file 3600
|
||||
|
||||
stage-b-test-8-npu-a3:
|
||||
needs: [check-changes, pr-gate, set-image-config]
|
||||
if: needs.check-changes.outputs.main_package == 'true'
|
||||
runs-on: linux-aarch64-a3-8
|
||||
container:
|
||||
image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
|
||||
steps:
|
||||
- name: Checkout code
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
ref: ${{ inputs.ref || github.ref }}
|
||||
|
||||
- name: Mark repository safe
|
||||
run: |
|
||||
git config --system --add safe.directory ${GITHUB_WORKSPACE}
|
||||
|
||||
- name: Install dependencies
|
||||
env:
|
||||
TORCH_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/whl/cpu"
|
||||
PYPI_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
UV_INDEX_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
GITHUB_PROXY_URL: "https://gh-proxy.test.osinfra.cn/"
|
||||
RUSTUP_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local:8082"
|
||||
run: |
|
||||
# speed up by using infra cache services
|
||||
CACHING_URL="cache-service.nginx-pypi-cache.svc.cluster.local"
|
||||
sed -Ei "s@(ports|archive).ubuntu.com@${CACHING_URL}:8081@g" /etc/apt/sources.list
|
||||
pip config set global.index-url http://${CACHING_URL}/pypi/simple
|
||||
pip config set global.trusted-host "${CACHING_URL}"
|
||||
|
||||
bash scripts/ci/npu/npu_ci_install_dependency.sh a3
|
||||
# copy required file from our daily cache
|
||||
cp ~/.cache/modelscope/hub/datasets/otavia/ShareGPT_Vicuna_unfiltered/ShareGPT_V3_unfiltered_cleaned_split.json /tmp
|
||||
# copy gsm8k dataset
|
||||
cp ~/.cache/modelscope/hub/datasets/tmp/test.jsonl /tmp
|
||||
|
||||
- name: Run test
|
||||
timeout-minutes: 60
|
||||
env:
|
||||
SGLANG_USE_MODELSCOPE: true
|
||||
SGLANG_IS_IN_CI: true
|
||||
HF_ENDPOINT: https://hf-mirror.com
|
||||
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
|
||||
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
|
||||
STREAMS_PER_DEVICE: 32
|
||||
run: |
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-8-npu-a3 --timeout-per-file 3600
|
||||
|
||||
stage-b-test-16-npu-a3:
|
||||
needs: [check-changes, pr-gate, set-image-config]
|
||||
if: needs.check-changes.outputs.main_package == 'true'
|
||||
runs-on: linux-aarch64-a3-16
|
||||
container:
|
||||
image: ${{ needs.set-image-config.outputs.CANN_image_a3 }}
|
||||
steps:
|
||||
- name: Checkout code
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
ref: ${{ inputs.ref || github.ref }}
|
||||
|
||||
- name: Mark repository safe
|
||||
run: |
|
||||
git config --system --add safe.directory ${GITHUB_WORKSPACE}
|
||||
|
||||
- name: Install dependencies
|
||||
env:
|
||||
TORCH_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/whl/cpu"
|
||||
PYPI_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
UV_INDEX_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local/pypi/simple"
|
||||
GITHUB_PROXY_URL: "https://gh-proxy.test.osinfra.cn/"
|
||||
RUSTUP_CACHE_URL: "http://cache-service.nginx-pypi-cache.svc.cluster.local:8082"
|
||||
run: |
|
||||
# speed up by using infra cache services
|
||||
CACHING_URL="cache-service.nginx-pypi-cache.svc.cluster.local"
|
||||
sed -Ei "s@(ports|archive).ubuntu.com@${CACHING_URL}:8081@g" /etc/apt/sources.list
|
||||
pip config set global.index-url http://${CACHING_URL}/pypi/simple
|
||||
pip config set global.trusted-host "${CACHING_URL}"
|
||||
|
||||
bash scripts/ci/npu/npu_ci_install_dependency.sh a3
|
||||
# copy required file from our daily cache
|
||||
cp ~/.cache/modelscope/hub/datasets/otavia/ShareGPT_Vicuna_unfiltered/ShareGPT_V3_unfiltered_cleaned_split.json /tmp
|
||||
# copy gsm8k dataset
|
||||
cp ~/.cache/modelscope/hub/datasets/tmp/test.jsonl /tmp
|
||||
|
||||
- name: Run test
|
||||
timeout-minutes: 120
|
||||
env:
|
||||
SGLANG_USE_MODELSCOPE: true
|
||||
SGLANG_IS_IN_CI: true
|
||||
HF_ENDPOINT: https://hf-mirror.com
|
||||
TORCH_EXTENSIONS_DIR: /tmp/torch_extensions
|
||||
PYTORCH_NPU_ALLOC_CONF: "expandable_segments:True"
|
||||
STREAMS_PER_DEVICE: 32
|
||||
run: |
|
||||
apt-get install -y libssl-dev
|
||||
pip install sglang_router
|
||||
cd test
|
||||
python3 run_suite.py --hw npu --suite stage-b-test-16-npu-a3 --timeout-per-file 3600
|
||||
|
||||
@@ -537,27 +692,27 @@ jobs:
|
||||
# NPU accuracy tests
|
||||
- name: glm4_7_flash_1p_gsm8k
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/glm4_7_flash/test_npu_glm4_7_flash_1p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/glm4_7_flash/test_npu_glm4_7_flash_1p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
- name: qwen3_vl_30b_a3b_bf16_2p_gsm8k
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/accuracy/qwen3_vl_30b_a3b/test_npu_qwen3_vl_30b_a3b_bf16_2p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_vl_30b_a3b/test_npu_qwen3_vl_30b_a3b_bf16_2p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
- name: glm5_top64_pruned_bf16_8p_gsm8k
|
||||
runner: linux-aarch64-a3-16
|
||||
test_case: test/registered/ascend/accuracy/glm5_top64_pruned/test_npu_glm5_top64_pruned_bf16_8p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/glm5_top64_pruned/test_npu_glm5_top64_pruned_bf16_8p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
- name: moonshotai_moonlight_16b_a3b_bf16_1p_gsm8k
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/moonshotai_moonlight_16b_a3b/test_npu_moonlight_16b_a3b_bf16_1p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/moonshotai_moonlight_16b_a3b/test_npu_moonlight_16b_a3b_bf16_1p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
- name: qwen3_5_9b_bf16_1p_gsm8k
|
||||
runner: linux-aarch64-a3-2
|
||||
test_case: test/registered/ascend/accuracy/qwen3_5_9b/test_npu_qwen3_5_9b_bf16_1p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_5_9b/test_npu_qwen3_5_9b_bf16_1p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
- name: qwen3_vl_8b_bf16_2p_gsm8k
|
||||
runner: linux-aarch64-a3-4
|
||||
test_case: test/registered/ascend/accuracy/qwen3_vl_8b/test_npu_qwen3_vl_8b_bf16_2p_gsm8k.py
|
||||
test_case: test/registered/npu/accuracy/qwen3_vl_8b/test_npu_qwen3_vl_8b_bf16_2p_gsm8k.py
|
||||
test_type: 'accuracy'
|
||||
uses: ./.github/workflows/nightly-test-npu-e2e-single-node.yml
|
||||
with:
|
||||
@@ -578,8 +733,11 @@ jobs:
|
||||
|
||||
stage-a-unit-test-npu,
|
||||
stage-b-test-1-npu-a2,
|
||||
stage-b-test-1-npu-a3,
|
||||
stage-b-test-2-npu-a2,
|
||||
stage-b-test-2-npu-a3,
|
||||
stage-b-test-4-npu-a3,
|
||||
stage-b-test-8-npu-a3,
|
||||
stage-b-test-16-npu-a3,
|
||||
|
||||
multimodal-gen-test-1-npu-a3,
|
||||
|
||||
Reference in New Issue
Block a user