skip cutlass fwop and xformers cuda backend

tests/kernels/test_prefix_prefill.py

@@ -9,6 +9,7 @@ from xformers.ops.fmha.attn_bias import BlockDiagonalCausalFromBottomRightMask
 
 from vllm.attention.backends.xformers import _make_alibi_bias
 from vllm.attention.ops.prefix_prefill import context_attention_fwd
+from vllm.utils import  is_hip
 
 NUM_HEADS = [64]
 NUM_QUERIES_PER_KV = [1, 8, 64]
@@ -158,6 +159,7 @@ def test_contexted_kv_attention(
     end_time = time.time()
     print(f"triton Time: {(end_time - start_time)*1000:.2f} ms")
 
+    if not is_hip():
         scale = float(1.0 / (head_size**0.5))
 
         attn_op = xops.fmha.cutlass.FwOp()
@@ -373,6 +375,8 @@ def test_contexted_kv_attention_alibi(
     torch.cuda.synchronize()
     end_time = time.time()
     print(f"triton Time: {(end_time - start_time)*1000:.2f} ms")
+    
+    if not is_hip():
         scale = float(1.0 / (head_size**0.5))
 
         # NOTE(DefTruth): In order to reuse _make_alibi_bias function,