fix: resolve tuning fused moe issue (#9587)

benchmark/kernels/fused_moe_triton/tuning_fused_moe_triton.py

@@ -22,7 +22,7 @@ from sglang.srt.layers.moe.fused_moe_triton.fused_moe import (
 )
 from sglang.srt.layers.moe.moe_runner import MoeRunnerConfig
 from sglang.srt.layers.moe.topk import TopKConfig, select_experts
-from sglang.srt.utils import is_hip, is_rocm
+from sglang.srt.utils import is_hip
 
 _is_hip = is_hip()
 
@@ -287,7 +287,7 @@ class BenchmarkWorker:
             )
         else:
             config = op_config[min(op_config.keys(), key=lambda x: abs(x - num_tokens))]
-        with torch.cuda.device(self.device_id) if is_rocm() else nullcontext():
+        with torch.cuda.device(self.device_id) if is_hip() else nullcontext():
             kernel_time = benchmark_config(
                 config,
                 num_tokens,
@@ -319,7 +319,7 @@ class BenchmarkWorker:
     ) -> Dict[str, int]:
         best_config = None
         best_time = float("inf")
-        with torch.cuda.device(self.device_id) if is_rocm() else nullcontext():
+        with torch.cuda.device(self.device_id) if is_hip() else nullcontext():
             for config in tqdm(search_space):
                 try:
                     kernel_time = benchmark_config(