From 18dd5c972d03942ded6fd9b608f8ed3f2325d603 Mon Sep 17 00:00:00 2001 From: YC Yen-Ching Tseng Date: Tue, 17 Mar 2026 14:53:33 +0800 Subject: [PATCH] [AMD] Fix AMD CI : stage-c-large-8-gpu-amd-mi35x (#20521) Co-authored-by: billisyahao --- .github/workflows/pr-test-amd-rocm720.yml | 4 ++-- .github/workflows/pr-test-amd.yml | 4 ++-- test/registered/amd/test_kimi_k25_mxfp4.py | 4 +--- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/.github/workflows/pr-test-amd-rocm720.yml b/.github/workflows/pr-test-amd-rocm720.yml index 4489ac74a..66f546b3d 100644 --- a/.github/workflows/pr-test-amd-rocm720.yml +++ b/.github/workflows/pr-test-amd-rocm720.yml @@ -839,7 +839,7 @@ jobs: fail-fast: false matrix: runner: [linux-mi35x-gpu-8] - part: [0] + part: [0, 1] runs-on: ${{matrix.runner}} steps: - name: Checkout code @@ -860,7 +860,7 @@ jobs: - name: Run test timeout-minutes: 60 run: | - bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-c-test-large-8-gpu-amd-mi35x --auto-partition-id ${{ matrix.part }} --auto-partition-size 1 --timeout-per-file 3600 ${{ inputs.continue_on_error && '--continue-on-error' || '' }} + bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-c-test-large-8-gpu-amd-mi35x --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 --timeout-per-file 3600 ${{ inputs.continue_on_error && '--continue-on-error' || '' }} # =============================================== Disaggregation ==================================================== stage-b-test-large-8-gpu-35x-disaggregation-amd: diff --git a/.github/workflows/pr-test-amd.yml b/.github/workflows/pr-test-amd.yml index 4c60dba61..cceb529ed 100644 --- a/.github/workflows/pr-test-amd.yml +++ b/.github/workflows/pr-test-amd.yml @@ -748,7 +748,7 @@ jobs: fail-fast: false matrix: runner: [linux-mi35x-gpu-8] - part: [0] + part: [0, 1] runs-on: ${{matrix.runner}} steps: - name: Checkout code @@ -770,7 +770,7 @@ jobs: - name: Run test timeout-minutes: 60 run: | - bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-c-test-large-8-gpu-amd-mi35x --auto-partition-id ${{ matrix.part }} --auto-partition-size 1 --timeout-per-file 3600 ${{ inputs.continue_on_error && '--continue-on-error' || '' }} + bash scripts/ci/amd/amd_ci_exec.sh -w "/sglang-checkout/test" python3 run_suite.py --hw amd --suite stage-c-test-large-8-gpu-amd-mi35x --auto-partition-id ${{ matrix.part }} --auto-partition-size 2 --timeout-per-file 3600 ${{ inputs.continue_on_error && '--continue-on-error' || '' }} # =============================================== Disaggregation ==================================================== stage-b-test-large-8-gpu-35x-disaggregation-amd: diff --git a/test/registered/amd/test_kimi_k25_mxfp4.py b/test/registered/amd/test_kimi_k25_mxfp4.py index 9c24df457..196ee18e3 100644 --- a/test/registered/amd/test_kimi_k25_mxfp4.py +++ b/test/registered/amd/test_kimi_k25_mxfp4.py @@ -32,9 +32,7 @@ from sglang.test.test_utils import ( register_amd_ci(est_time=3600, suite="stage-c-test-large-8-gpu-amd-mi35x") -KIMI_K25_MXFP4_MODEL_PATH = os.environ.get( - "KIMI_K25_MXFP4_MODEL_PATH", "/data/models/amd/Kimi-K2.5-MXFP4" -) +KIMI_K25_MXFP4_MODEL_PATH = "amd/Kimi-K2.5-MXFP4" SERVER_LAUNCH_TIMEOUT = 3600