[AMD] Split qwen3.5 triton DCP test into its own nightly job (#29409)

This commit is contained in:
YC Yen-Ching Tseng
2026-07-01 00:39:39 -07:00
committed by GitHub
parent 00e12cebb4
commit 308d89e042
3 changed files with 75 additions and 1 deletions
@@ -79,6 +79,7 @@ on:
# 8-GPU Qwen 3.5 (MI30x + MI35x)
- nightly-8-gpu-qwen35-rocm720
- nightly-8-gpu-mi35x-qwen35-rocm720
- nightly-8-gpu-mi35x-qwen35-triton-dcp-rocm720
# 8-GPU GLM-5.1 (MI30x + MI35x)
- nightly-8-gpu-glm51-rocm720
- nightly-8-gpu-mi35x-glm51-rocm720
@@ -1660,6 +1661,40 @@ jobs:
echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
exit ${TEST_EXIT_CODE:-0}
nightly-8-gpu-mi35x-qwen35-triton-dcp-rocm720:
if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (!(inputs.job_filter || inputs.job_select) || (inputs.job_filter || inputs.job_select) == 'all' || contains(format(',{0},', inputs.job_filter || inputs.job_select), ',nightly-8-gpu-mi35x-qwen35-triton-dcp-rocm720,'))
runs-on: linux-mi35x-gpu-8
steps:
- name: Checkout code
uses: actions/checkout@v4
with:
ref: ${{ inputs.ref || github.sha }}
- name: Ensure VRAM is clear
run: bash scripts/ci/amd/ensure_vram_clear.sh rocm
- name: Setup docker (ROCm 7.2)
run: |
touch github_summary.md
bash scripts/ci/amd/amd_ci_start_container.sh --rocm-version rocm720
env:
GITHUB_WORKSPACE: ${{ github.workspace }}
- name: Install dependencies
run: |
bash scripts/ci/amd/amd_ci_install_dependency.sh --skip-aiter-build --skip-test-time-deps
bash scripts/ci/amd/amd_ci_exec.sh pip install tabulate
- name: Accuracy Test MI35x ROCm 7.2 (8-GPU Qwen 3.5 Triton DCP)
timeout-minutes: 120
run: |
> github_summary.md # Clear summary file
bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
-e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-mi35x-qwen35-triton-dcp --nightly --timeout-per-file 7200 ${{ (github.event_name == 'schedule' || inputs.continue_on_error) && '--continue-on-error' || '' }} || TEST_EXIT_CODE=$?
echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
exit ${TEST_EXIT_CODE:-0}
# ==============================================================================
# 8-GPU GLM-5.1 (MI30x + MI35x)
# ==============================================================================
@@ -1991,6 +2026,7 @@ jobs:
# 8-GPU Qwen 3.5 (MI30x + MI35x)
- nightly-8-gpu-qwen35-rocm720
- nightly-8-gpu-mi35x-qwen35-rocm720
- nightly-8-gpu-mi35x-qwen35-triton-dcp-rocm720
# 8-GPU GLM-5.1 (MI30x + MI35x)
- nightly-8-gpu-glm51-rocm720
- nightly-8-gpu-mi35x-glm51-rocm720
+36
View File
@@ -75,6 +75,7 @@ on:
# 8-GPU Qwen 3.5 (MI30x + MI35x)
- nightly-8-gpu-qwen35
- nightly-8-gpu-mi35x-qwen35
- nightly-8-gpu-mi35x-qwen35-triton-dcp
# 8-GPU GLM-5.1 (MI30x + MI35x)
- nightly-8-gpu-glm51
- nightly-8-gpu-mi35x-glm51
@@ -1514,6 +1515,40 @@ jobs:
echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
exit ${TEST_EXIT_CODE:-0}
nightly-8-gpu-mi35x-qwen35-triton-dcp:
if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') && (!(inputs.job_filter || inputs.job_select) || (inputs.job_filter || inputs.job_select) == 'all' || contains(format(',{0},', inputs.job_filter || inputs.job_select), ',nightly-8-gpu-mi35x-qwen35-triton-dcp,'))
runs-on: linux-mi35x-gpu-8
steps:
- name: Checkout code
uses: actions/checkout@v4
with:
ref: ${{ inputs.ref || github.sha }}
- name: Ensure VRAM is clear
run: bash scripts/ci/amd/ensure_vram_clear.sh rocm
- name: Setup docker
run: |
touch github_summary.md
bash scripts/ci/amd/amd_ci_start_container.sh
env:
GITHUB_WORKSPACE: ${{ github.workspace }}
- name: Install dependencies
run: |
bash scripts/ci/amd/amd_ci_install_dependency.sh
bash scripts/ci/amd/amd_ci_exec.sh pip install tabulate
- name: Accuracy Test MI35x (8-GPU Qwen 3.5 Triton DCP)
timeout-minutes: 120
run: |
> github_summary.md # Clear summary file
bash scripts/ci/amd/amd_ci_exec.sh -w /sglang-checkout/test \
-e GITHUB_STEP_SUMMARY="/sglang-checkout/github_summary.md" \
python3 run_suite.py --hw amd --suite nightly-amd-accuracy-8-gpu-mi35x-qwen35-triton-dcp --nightly --timeout-per-file 7200 ${{ (github.event_name == 'schedule' || inputs.continue_on_error) && '--continue-on-error' || '' }} || TEST_EXIT_CODE=$?
echo "$(<github_summary.md )" >> $GITHUB_STEP_SUMMARY || true
exit ${TEST_EXIT_CODE:-0}
# ==============================================================================
# 8-GPU GLM-5.1 (MI30x + MI35x)
# ==============================================================================
@@ -1841,6 +1876,7 @@ jobs:
# 8-GPU Qwen 3.5 (MI30x + MI35x)
- nightly-8-gpu-qwen35
- nightly-8-gpu-mi35x-qwen35
- nightly-8-gpu-mi35x-qwen35-triton-dcp
# 8-GPU GLM-5.1 (MI30x + MI35x)
- nightly-8-gpu-glm51
- nightly-8-gpu-mi35x-glm51