chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Fix CI by reverting incorrect metric check logic (#15004)

Browse Source
This commit is contained in:
Kangyan-Zhou
2025-12-12 10:07:38 -08:00
committed by GitHub
parent 526fd0082f
commit b243154614
4 changed files with 74 additions and 277 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
.github/workflows/pr-test.yml vendored
View File

@@ -27,7 +27,6 @@ concurrency:
env:
SGLANG_IS_IN_CI: true
SGLANG_CI_ENABLE_RETRY: true
jobs:
# =============================================== check changes ====================================================
@@ -551,10 +550,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh
- name: Run test
timeout-minutes: 40
timeout-minutes: 30
run: |
cd test/srt
python3 run_suite.py --suite quantization_test --enable-retry
python3 run_suite.py --suite quantization_test
unit-test-backend-1-gpu:
needs: [check-changes, call-gate, stage-a-test-1]
@@ -593,10 +592,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh
- name: Run test
timeout-minutes: 40
timeout-minutes: 30
run: |
cd test/srt
python3 run_suite.py --suite per-commit-1-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 15 --enable-retry
python3 run_suite.py --suite per-commit-1-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 15
unit-test-backend-2-gpu:
needs: [check-changes, call-gate, unit-test-backend-1-gpu]
@@ -634,10 +633,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh
- name: Run test
timeout-minutes: 40
timeout-minutes: 30
run: |
cd test/srt
python3 run_suite.py --suite per-commit-2-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 --enable-retry
python3 run_suite.py --suite per-commit-2-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 2
unit-test-backend-4-gpu:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -675,10 +674,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_dependency.sh
- name: Run test
timeout-minutes: 30
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite per-commit-4-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --enable-retry
python3 run_suite.py --suite per-commit-4-gpu --auto-partition-id ${{ matrix.part }} --auto-partition-size 3
unit-test-backend-8-gpu-h200:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -722,10 +721,10 @@ jobs:
python3 -m sglang.compile_deep_gemm --model deepseek-ai/DeepSeek-V3-0324 --tp 8 --trust-remote-code
- name: Run test
timeout-minutes: 30
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite per-commit-8-gpu-h200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --enable-retry
python3 run_suite.py --suite per-commit-8-gpu-h200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 3
unit-test-backend-8-gpu-h20:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -764,10 +763,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_deepep.sh
- name: Run test
timeout-minutes: 30
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite per-commit-8-gpu-h20 --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 --enable-retry
python3 run_suite.py --suite per-commit-8-gpu-h20 --auto-partition-id ${{ matrix.part }} --auto-partition-size 2
performance-test-1-gpu-part-1:
needs: [check-changes, call-gate, stage-a-test-1]
@@ -1056,12 +1055,10 @@ jobs:
pip install -e .
- name: Evaluate accuracy
timeout-minutes: 25
timeout-minutes: 20
run: |
cd test/srt
python3 test_eval_accuracy_large.py
env:
SGLANG_CI_ENABLE_RETRY: ${{ env.SGLANG_CI_ENABLE_RETRY }}
accuracy-test-2-gpu:
needs: [check-changes, call-gate, accuracy-test-1-gpu]
@@ -1098,12 +1095,10 @@ jobs:
pip install -e .
- name: Evaluate accuracy (TP=2)
timeout-minutes: 25
timeout-minutes: 20
run: |
cd test/srt
python3 test_moe_eval_accuracy_large.py
env:
SGLANG_CI_ENABLE_RETRY: ${{ env.SGLANG_CI_ENABLE_RETRY }}
unit-test-deepep-4-gpu:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -1137,10 +1132,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_deepep.sh
- name: Run test
timeout-minutes: 30
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite per-commit-4-gpu-deepep --enable-retry
python3 run_suite.py --suite per-commit-4-gpu-deepep
unit-test-deepep-8-gpu:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -1174,10 +1169,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} bash scripts/ci/ci_install_deepep.sh
- name: Run test
timeout-minutes: 30
timeout-minutes: 20
run: |
cd test/srt
python3 run_suite.py --suite per-commit-8-gpu-h200-deepep --enable-retry
python3 run_suite.py --suite per-commit-8-gpu-h200-deepep
unit-test-backend-4-gpu-b200:
needs: [check-changes, call-gate, unit-test-backend-2-gpu]
@@ -1216,10 +1211,10 @@ jobs:
CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} IS_BLACKWELL=1 bash scripts/ci/ci_install_dependency.sh
- name: Run test
timeout-minutes: 40
timeout-minutes: 30
run: |
cd test/srt
IS_BLACKWELL=1 python3 run_suite.py --suite per-commit-4-gpu-b200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --timeout-per-file 1800 --enable-retry
IS_BLACKWELL=1 python3 run_suite.py --suite per-commit-4-gpu-b200 --auto-partition-id ${{ matrix.part }} --auto-partition-size 3 --timeout-per-file 1800
# TODO: Add gb200 tests back after the ci runner is fixed
# unit-test-backend-4-gpu-gb200:
@@ -1256,10 +1251,10 @@ jobs:
# CUSTOM_BUILD_SGL_KERNEL=${{needs.check-changes.outputs.sgl_kernel}} IS_BLACKWELL=1 GRACE_BLACKWELL=1 bash scripts/ci/ci_install_deepep.sh
# - name: Run test
# timeout-minutes: 55
# timeout-minutes: 45
# run: |
# cd test/srt
# python3 run_suite.py --suite per-commit-4-gpu-gb200 --auto-partition-id 0 --auto-partition-size 1 --timeout-per-file 3600 --enable-retry
# python3 run_suite.py --suite per-commit-4-gpu-gb200 --auto-partition-id 0 --auto-partition-size 1 --timeout-per-file 3600
pr-test-finish:
needs: