[Triton-MLIR][BACKEND] Make mmav1 works on basic cases (#944)

TODO:

- Add more cases
- Currently, we just set vec to 4 to make the basic cases pass

Issue:

- the vec in shared layout is different compared to master branch
- when vec=1, it encounters CUDA misalignment error, it doesn't work in
master branch as well
- when setting vec to the value identical to master branch, the MMA
works

.github/workflows/integration-tests.yml

@@ -17,7 +17,7 @@ jobs:
         id: set-matrix
         run: |
           if [ x"${{ github.repository }}" == x"openai/triton" ]; then
-            echo '::set-output name=matrix::[["self-hosted", "A10"], "macos-10.15"]'
+            echo '::set-output name=matrix::[["self-hosted", "A10"], ["self-hosted", "V100"], "macos-10.15"]'
           else
             echo '::set-output name=matrix::["ubuntu-latest", "macos-10.15"]'
           fi
@@ -79,11 +79,18 @@ jobs:
           lit -v "$LIT_TEST_DIR"
 
       - name: Run python tests
-        if: ${{matrix.runner[0] == 'self-hosted'}}
+        if: ${{matrix.runner[0] == 'self-hosted' && matrix.runner[1] == 'A10'}}
         run: |
           cd python/tests
           pytest
 
+      # TODO[Superjomn] Enable all the tests on V100 if available
+      - name: Run python tests on V100
+        if: ${{matrix.runner[0] == 'self-hosted' && matrix.runner[1] == 'V100'}}
+        run: |
+          cd python/tests
+          pytest test_gemm.py::test_gemm_no_scf_for_mmav1
+
       - name: Run CXX unittests
         run: |
           cd python/