EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[CI] merge all ci tests into one file (#1289)

Browse Source
This commit is contained in:
Lianmin Zheng
2024-09-01 02:36:56 -07:00
committed by GitHub
parent 54772f784a
commit 761b2cebd6
7 changed files with 211 additions and 233 deletions
Download Patch File Download Diff File Expand all files Collapse all files

74
.github/workflows/accuracy-test.yml vendored
View File

@@ -1,74 +0,0 @@
name: Accuracy Test
on:
push:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
pull_request:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
workflow_dispatch:
concurrency:
group: accuracy-test-${{ github.ref }}
cancel-in-progress: true
jobs:
one-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
git clone https://github.com/merrymercy/human-eval.git
cd human-eval
pip install -e .
- name: Evaluate Accuracy
timeout-minutes: 20
run: |
cd test/srt
python3 test_eval_accuracy_large.py
two-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 2-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
git clone https://github.com/merrymercy/human-eval.git
cd human-eval
pip install -e .
- name: Evaluate Accuracy
timeout-minutes: 20
run: |
cd test/srt
python3 test_moe_eval_accuracy_large.py
finish:
needs: [one-gpu, two-gpu]
runs-on: ubuntu-latest
steps:
- name: Finish
run: echo "This is an empty step to ensure that all jobs are completed."

96
.github/workflows/e2e-test.yml vendored
View File

@@ -1,96 +0,0 @@
name: E2E Test
on:
push:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
pull_request:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
workflow_dispatch:
concurrency:
group: e2e-test-${{ github.ref }}
cancel-in-progress: true
jobs:
one-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Benchmark Serving Throughput
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default
- name: Benchmark Serving Latency
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_latency.TestServingLatency.test_default
- name: Benchmark Serving Throughput (w/o RadixAttention)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default_without_radix_cache
- name: Benchmark Serving Throughput (w/o ChunkedPrefill)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default_without_chunked_prefill
two-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 2-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Benchmark Serving Throughput (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default
- name: Benchmark Serving Latency (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_latency.TestServingLatency.test_default
- name: Benchmark Serving Throughput (w/o RadixAttention) (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default_without_radix_cache
finish:
needs: [one-gpu, two-gpu]
runs-on: ubuntu-latest
steps:
- name: Finish
run: echo "This is an empty step to ensure that all jobs are completed."

11
.github/workflows/lint.yml vendored
View File

@@ -1,19 +1,22 @@
name: Lint
on: [push, pull_request]
on: [pull_request]
jobs:
lint:
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Set up Python 3.8
- name: Set up Python 3.9
uses: actions/setup-python@v2
with:
python-version: 3.8
python-version: 3.9
- name: Install pre-commit hook
run: |
python -m pip install pre-commit
pre-commit install
- name: Linting
run: pre-commit run --all-files

201
.github/workflows/pr-test.yml vendored Normal file
View File

@@ -0,0 +1,201 @@
name: Pull Request Test
on:
push:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
pull_request:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
workflow_dispatch:
concurrency:
group: pr-test-${{ github.ref }}
cancel-in-progress: true
jobs:
unit-test-frontend:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[dev]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Run test
timeout-minutes: 20
run: |
cd test/lang
python3 run_suite.py --suite minimal
unit-test-backend-part-0:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[dev]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Run test
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite minimal --range-begin 0 --range-end 8
unit-test-backend-part-1:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[dev]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Run test
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite minimal --range-begin 8
performance-test-1-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Benchmark Serving Throughput
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default
- name: Benchmark Serving Latency
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_latency.TestServingLatency.test_default
- name: Benchmark Serving Throughput (w/o RadixAttention)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default_without_radix_cache
- name: Benchmark Serving Throughput (w/o ChunkedPrefill)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_serving_throughput.TestServingThroughput.test_default_without_chunked_prefill
performance-test-2-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 2-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Benchmark Serving Throughput (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default
- name: Benchmark Serving Latency (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_latency.TestServingLatency.test_default
- name: Benchmark Serving Throughput (w/o RadixAttention) (TP=2)
timeout-minutes: 10
run: |
cd test/srt
python3 -m unittest test_moe_serving_throughput.TestServingThroughput.test_default_without_radix_cache
accuracy-test-1-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
git clone https://github.com/merrymercy/human-eval.git
cd human-eval
pip install -e .
- name: Evaluate Accuracy
timeout-minutes: 20
run: |
cd test/srt
python3 test_eval_accuracy_large.py
accuracy-test-2-gpu:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 2-gpu-runner
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[all]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
git clone https://github.com/merrymercy/human-eval.git
cd human-eval
pip install -e .
- name: Evaluate Accuracy
timeout-minutes: 20
run: |
cd test/srt
python3 test_moe_eval_accuracy_large.py
finish:
needs: [
unit-test-frontend, unit-test-backend-part-0, unit-test-backend-part-1,
performance-test-1-gpu, performance-test-2-gpu,
accuracy-test-1-gpu, accuracy-test-2-gpu
]
runs-on: ubuntu-latest
steps:
- name: Finish
run: echo "This is an empty step to ensure that all jobs are completed."

56
.github/workflows/unit-test.yml vendored
View File

@@ -1,56 +0,0 @@
name: Unit Test
on:
push:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
pull_request:
branches: [ main ]
paths:
- "python/sglang/**"
- "test/**"
workflow_dispatch:
concurrency:
group: unit-test-${{ github.ref }}
cancel-in-progress: true
jobs:
run-test:
if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
runs-on: 1-gpu-runner
strategy:
matrix:
test_type: ['backend-0', 'backend-1', 'frontend']
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Install dependencies
run: |
pip install --upgrade pip
pip install -e "python[dev]"
pip install flashinfer -i https://flashinfer.ai/whl/cu121/torch2.4/ --force-reinstall
- name: Run test
timeout-minutes: 20
run: |
if [ "${{ matrix.test_type }}" = "frontend" ]; then
cd test/lang
python3 run_suite.py --suite minimal
elif [ "${{ matrix.test_type }}" = "backend-0" ]; then
cd test/srt
python3 run_suite.py --suite minimal --range-begin 0 --range-end 8
elif [ "${{ matrix.test_type }}" = "backend-1" ]; then
cd test/srt
python3 run_suite.py --suite minimal --range-begin 8
fi
finish:
needs: [run-test]
runs-on: ubuntu-latest
steps:
- name: Finish
run: echo "This is an empty step to ensure that all jobs are completed."

4
python/sglang/README.md
View File

@@ -2,8 +2,8 @@
- `lang`: The frontend language.
- `srt`: The backend engine for running local models. (SRT = SGLang Runtime).
- `test`: Test utilities.
- `api.py`: Public API.
- `test`: The test utilities.
- `api.py`: The public APIs.
- `bench_latency.py`: Benchmark a single static batch.
- `bench_serving.py`: Benchmark online serving with dynamic requests.
- `global_config.py`: The global configs and constants.

2
test/srt/test_moe_serving_throughput.py
View File

@@ -75,7 +75,7 @@ class TestServingThroughput(unittest.TestCase):
)
if os.getenv("SGLANG_IS_IN_CI", "false") == "true":
assert res["output_throughput"] > 1850
assert res["output_throughput"] > 1800
def test_default_without_radix_cache(self):
res = self.run_test(