UT cases failed for CUDA graph:
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_False_xpu_float32
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_max_use_cuda_graph_True_w_empty_False_xpu_float64
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_True_xpu_float32
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_True_xpu_float64
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_False_xpu_float32
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_norm_use_cuda_graph_True_w_empty_False_xpu_float64
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_True_xpu_float32
test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_True_xpu_float64
UT cases failed for CUDA graph: test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_False_xpu_float32 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_max_use_cuda_graph_True_w_empty_False_xpu_float64 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_True_xpu_float32 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_max_use_cuda_graph_True_w_empty_True_xpu_float64 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_False_xpu_float32 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensorsforeach_norm_use_cuda_graph_True_w_empty_False_xpu_float64 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_True_xpu_float32 test_foreach_xpu.py::TestForeachXPU::test_big_num_tensors__foreach_norm_use_cuda_graph_True_w_empty_True_xpu_float64