(self)
| 175 | """The CpAsyncBulkTensor instruction submodule.""" |
| 176 | |
| 177 | def __init__(self): |
| 178 | self.g2c = _op_wrapper(_cuda_op.ptx_cp_async_bulk_tensor_global_to_cluster) |
| 179 | self.g2c_tile_gather4 = _op_wrapper( |
| 180 | _cuda_op.ptx_cp_async_bulk_tensor_tile_gather4_global_to_cluster |
| 181 | ) |
| 182 | self.s2g = _op_wrapper(_cuda_op.ptx_cp_async_bulk_tensor_shared_to_global) |
| 183 | self.s2g_reduce = _op_wrapper(_cuda_op.ptx_cp_async_bulk_tensor_shared_to_global_reduce) |
| 184 | self.g2c_prefetch = _op_wrapper( |
| 185 | _cuda_op.ptx_cp_async_bulk_tensor_global_to_cluster_prefetch |
| 186 | ) |
| 187 | |
| 188 | @staticmethod |
| 189 | def g2c_bar_addr( |
nothing calls this directly
no test coverage detected