Hi all , i deployed the vta.bit on my ultra96 board, when i tried to run test_benchmark_topi_conv2d.py, the PC side will stop and report as the picture below:
And the ultra96 also gave some log info as below:
Traceback (most recent call last): File “/usr/lib/python3.6/multiprocessing/process.py”, line 258, in _bootstrap self.run() File “/usr/lib/python3.6/multiprocessing/process.py”, line 93, in run self._target(*self._args, **self._kwargs) File “/home/xilinx/jupyter_notebooks/tvm_vta/tvm/python/tvm/rpc/server.py”, line 140, in _serve_loop _ffi_api.ServerLoop(sock.fileno()) File “/home/xilinx/jupyter_notebooks/tvm_vta/tvm/python/tvm/_ffi/_ctypes/packed_func.py”, line 239, in call raise_last_ffi_error() File “/home/xilinx/jupyter_notebooks/tvm_vta/tvm/python/tvm/_ffi/base.py”, line 466, in raise_last_ffi_error tb = _filter_traceback_frames(tb, filter_funcs) File “/home/xilinx/jupyter_notebooks/tvm_vta/tvm/python/tvm/_ffi/base.py”, line 399, in _filter_traceback_frames new_tb = functools.reduce(_append_frame, reversed(filtered_frames)) File “/home/xilinx/jupyter_notebooks/tvm_vta/tvm/python/tvm/_ffi/base.py”, line 396, in _append_frame tb, next_tb_frame.tb_frame, next_tb_frame.tb_lasti, next_tb_frame.tb_lineno TypeError: cannot create ‘traceback’ instances
It seems that there’s something wrong with tvm.nd.array(), when I set the ctx as the second parm, it’s always going wrong(ps. ctx has been set as remote.ext_dev(0) before). If I leave the second param of tvm.nd.array() as empy, the func works well and I can see that the data was allocated on cpu(0). Can anybody shoot this problem, thanks a lot !