Refactor JIT kernel CI to use run_suite.py registration system (#21239)

.github/workflows/nightly-test-nvidia.yml

@@ -25,6 +25,7 @@ on:
           - 'nightly-test-multimodal-server-2-gpu'
           - 'nightly-test-perf-4-gpu-b200'
           - 'nightly-test-perf-8-gpu-b200'
+          - 'nightly-test-kernel-1-gpu-h100'
   workflow_call:
     inputs:
       ref:
@@ -76,6 +77,42 @@ jobs:
       - uses: ./.github/actions/upload-cuda-coredumps
         if: always()
 
+  # JIT kernel full unit tests (expanded parameter ranges via SGLANG_JIT_KERNEL_RUN_FULL_TESTS)
+  nightly-test-kernel-1-gpu-h100:
+    if: github.repository == 'sgl-project/sglang' && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-test-kernel-1-gpu-h100')
+    runs-on: 1-gpu-h100
+    timeout-minutes: 240
+    env:
+      # Full jit_kernel test grids (see sglang.jit_kernel.utils.should_run_full_tests)
+      SGLANG_JIT_KERNEL_RUN_FULL_TESTS: "1"
+      # Match pr-test-jit-kernel workflow for consistent JIT warmup behavior
+      SGLANG_JIT_DEEPGEMM_FAST_WARMUP: true
+      # Allow maintenance bypass on default branch (same semantics as PR JIT workflow)
+      SGLANG_PR_TEST_BYPASS_MAINTENANCE_ON_MAIN: ${{ github.ref == 'refs/heads/main' && 'true' || 'false' }}
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ inputs.ref || github.ref }}
+
+      - uses: ./.github/actions/check-maintenance
+        with:
+          github-token: ${{ github.token }}
+
+      - name: Install dependencies
+        timeout-minutes: 20
+        run: |
+          bash scripts/ci/cuda/ci_install_dependency.sh
+
+      - name: Run jit kernel nightly suite
+        timeout-minutes: 60
+        run: |
+          cd test
+          python3 run_suite.py --hw cuda --suite nightly-kernel-1-gpu --nightly --continue-on-error
+
+      - uses: ./.github/actions/upload-cuda-coredumps
+        if: always()
+
   # General tests - 4 GPU H100
   nightly-test-general-4-gpu-h100:
     if: github.repository == 'sgl-project/sglang' && (inputs.job_filter == '' || inputs.job_filter == 'all' || inputs.job_filter == 'nightly-test-general-4-gpu-h100')

.github/workflows/pr-test-jit-kernel.yml

@@ -56,36 +56,8 @@ jobs:
       - name: Run test
         timeout-minutes: 30
         run: |
-          cd python/sglang/jit_kernel
-          pytest tests/
-
-  jit-kernel-unit-test-nightly:
-    if: |
-      github.event_name == 'schedule' &&
-      inputs.jit_kernel == 'true'
-    runs-on: 1-gpu-h100
-    timeout-minutes: 240
-    env:
-      SGLANG_JIT_KERNEL_RUN_FULL_TESTS: "1"
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          ref: ${{ inputs.pr_head_sha || inputs.git_ref || github.sha }}
-
-      - uses: ./.github/actions/check-maintenance
-        with:
-          github-token: ${{ github.token }}
-
-      - name: Install dependencies
-        timeout-minutes: 20
-        run: |
-          bash scripts/ci/cuda/ci_install_dependency.sh
-
-      - name: Run full nightly test
-        timeout-minutes: 60
-        run: |
-          cd python/sglang/jit_kernel
-          pytest tests/
+          cd test/
+          python3 run_suite.py --hw cuda --suite stage-b-kernel-unit-1-gpu-large
 
   jit-kernel-benchmark-test:
     if: |
@@ -111,23 +83,5 @@ jobs:
       - name: Run benchmark tests
         timeout-minutes: 45
         run: |
-          cd python/sglang/jit_kernel/benchmark
-          echo "Running jit-kernel benchmark tests in CI mode..."
-
-          failures=()
-
-          for bench_file in bench_*.py; do
-            echo "Testing $bench_file..."
-            if ! timeout 120 python3 "$bench_file"; then
-              failures+=("$bench_file")
-            fi
-            echo "Completed $bench_file"
-            echo "---"
-          done
-
-          if [ ${#failures[@]} -ne 0 ]; then
-            echo "The following benchmark tests failed: ${failures[*]}"
-            exit 1
-          fi
-
-          echo "All jit-kernel benchmark tests completed successfully!"
+          cd test/
+          python3 run_suite.py --hw cuda --suite stage-b-kernel-benchmark-1-gpu-large