Avoid deleting entire cache for missing shards (#14754 follow-up) (#14853)

This commit is contained in:
Alison Shao
2025-12-11 01:17:04 -08:00
committed by GitHub
parent e52cf30e81
commit f85460fb19
4 changed files with 240 additions and 24 deletions

View File

@@ -115,6 +115,25 @@ jobs:
cd test
python3 run_suite.py --hw cuda --suite nightly-8-gpu-h200 --nightly --continue-on-error
- name: Run MiniMax-M2 nightly performance test
timeout-minutes: 180
env:
TRACE_BASE_URL: https://raw.githubusercontent.com/sglang-bot/sglang-ci-data/main/traces/${{ github.run_id }}
PERFETTO_RELAY_URL: ${{ vars.PERFETTO_RELAY_URL }}
GPU_CONFIG: "8-gpu-h200"
run: |
rm -rf test/performance_profiles_minimax_m2/
cd test
python3 nightly/test_minimax_m2_perf.py
- name: Publish MiniMax-M2 traces to storage repo
env:
GITHUB_TOKEN: ${{ secrets.GH_PAT_FOR_NIGHTLY_CI_DATA }}
GITHUB_RUN_ID: ${{ github.run_id }}
GITHUB_RUN_NUMBER: ${{ github.run_number }}
run: |
python3 scripts/ci/publish_traces.py --traces-dir test/performance_profiles_minimax_m2
- name: Run Qwen3-235B nightly performance test
timeout-minutes: 180
env:
@@ -172,25 +191,6 @@ jobs:
run: |
python3 scripts/ci/publish_traces.py --traces-dir test/performance_profiles_glm_4_6
- name: Run MiniMax-M2 nightly performance test
timeout-minutes: 180
env:
TRACE_BASE_URL: https://raw.githubusercontent.com/sglang-bot/sglang-ci-data/main/traces/${{ github.run_id }}
PERFETTO_RELAY_URL: ${{ vars.PERFETTO_RELAY_URL }}
GPU_CONFIG: "8-gpu-h200"
run: |
rm -rf test/performance_profiles_minimax_m2/
cd test
python3 nightly/test_minimax_m2_perf.py
- name: Publish MiniMax-M2 traces to storage repo
env:
GITHUB_TOKEN: ${{ secrets.GH_PAT_FOR_NIGHTLY_CI_DATA }}
GITHUB_RUN_ID: ${{ github.run_id }}
GITHUB_RUN_NUMBER: ${{ github.run_number }}
run: |
python3 scripts/ci/publish_traces.py --traces-dir test/performance_profiles_minimax_m2
# General tests - 8 GPU H20
nightly-test-general-8-gpu-h20:
if: github.repository == 'sgl-project/sglang' && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-test-general-8-gpu-h20')