[TEST] Update runtime skiplist (#4714)

pass rate: 98.91%->98.93%

Signed-off-by: Whitney Tsang <[email protected]>

Author: whitneywhtsang

python/test/unit/runtime/test_autotuner.py

@@ -173,8 +173,8 @@ def _kernel(dst, src, N, BLOCK_SIZE: tl.constexpr):
         assert records['capture_named_args']
 
 
-@pytest.mark.skipif(not is_cuda() or torch.cuda.get_device_capability()[0] < 9,
-                    reason="Requires compute capability >= 9 for NV")
+@pytest.mark.xfail(not is_cuda() or torch.cuda.get_device_capability()[0] < 9,
+                   reason="Requires compute capability >= 9 for NV")
 def test_override_ttir(device):
     N = 1024
     src = torch.randn(N, device=device)
@@ -222,8 +222,8 @@ def _kernel(dst, src, N, BLOCK_SIZE: tl.constexpr):
     torch.testing.assert_close(src * 10, dst)
 
 
-@pytest.mark.skipif(not is_cuda() or torch.cuda.get_device_capability()[0] < 9,
-                    reason="Requires compute capability >= 9 for NV")
+@pytest.mark.xfail(not is_cuda() or torch.cuda.get_device_capability()[0] < 9,
+                   reason="Requires compute capability >= 9 for NV")
 def test_override_ttgir(device):
     N = 1024
     src = torch.randn(N, device=device)
@@ -272,8 +272,8 @@ def _kernel(dst, src, N, BLOCK_SIZE: tl.constexpr):
     torch.testing.assert_close(src * 10, dst)
 
 
-@pytest.mark.skipif(not is_cuda() or torch.cuda.get_device_capability()[0] != 9,
-                    reason="PTX file in this unit test is only for SM90")
+@pytest.mark.xfail(not is_cuda() or torch.cuda.get_device_capability()[0] != 9,
+                   reason="PTX file in this unit test is only for SM90")
 def test_override_ptx(device):
     N = 1024
     src = torch.randn(N, device=device)
@@ -372,7 +372,7 @@ def _kernel(dst, src, N, BLOCK_SIZE: tl.constexpr):
 
 def test_exceed_tmem(device):
     if not torch.cuda.is_available() or not torch.cuda.get_device_capability()[0] == 10:
-        pytest.skip("Test requires tensor memory.")
+        pytest.xfail("Test requires tensor memory.")
     N = 512
     dst = torch.empty((N, ), device=device, dtype=torch.float32)
     configs = [triton.Config(kwargs={'BLOCK_SIZE': 128}), triton.Config(kwargs={'BLOCK_SIZE': 32})]
@@ -411,7 +411,7 @@ def dot_kernel(dst, BLOCK_SIZE: tl.constexpr):
 
 def test_exceed_threads(device):
     if not torch.cuda.is_available():
-        pytest.skip("CUDA is not available")
+        pytest.xfail("CUDA is not available")
     x = torch.empty(1024, device=device, dtype=torch.float32)
     y = torch.empty_like(x)
     output = torch.empty_like(x)