Can the pythn code still work to download bitstream if the Jupyter is turned off?

Hi, dear PYNQ developers,

If we turn off the Jupyter to offload the CPU on ZCU-104, can our Python code still work as well which loads the FPGA bitstream image and perform register configuration of the hardware accelerator on the ZCU-104? It is assumed that we use command mode to execute the Pyhon code on PYNQ ZCU-104.

xilinx@pynq_sdu:~ $ ps aux | grep jupyter
xilinx 4283 0.0 0.0 5668 668 pts/3 S+ 10:31 0:00 grep --color=auto jupyter
Tue 22. Dec 10:31:34 AM UTC 2020 :heavy_check_mark:

BTW, is it related to the following problem?

xilinx@pynq_sdu:~/pynq/prj_tst/int_add $ sudo python3.6
[sudo] password for xilinx:
/usr/local/lib/python3.6/dist-packages/pynq/ UserWarning: Users will not get PARAMETERS / REGISTERS information through TCL files. HWH file is recommended.
warnings.warn(message, UserWarning)
Traceback (most recent call last):
File “”, line 4, in
ol = Overlay("/home/xilinx/pynq/overlays/int_add/int_add.bit")
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 315, in init
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 379, in download
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 1938, in download
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 1458, in shutdown
ip = cls.ip_dict
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 1297, in ip_dict
File “/usr/local/lib/python3.6/dist-packages/pynq/”, line 1416, in client_request
cls._remote = Client(address, family=‘AF_UNIX’, authkey=key)
File “/usr/lib/python3.6/multiprocessing/”, line 487, in Client
c = SocketClient(address)
File “/usr/lib/python3.6/multiprocessing/”, line 614, in SocketClient
ConnectionRefusedError: [Errno 111] Connection refused
Tue 22. Dec 09:36:53 AM UTC 2020 ✘ ❬1❭

Thank you very much
All the best,

Yes you can run your Python code with PYNQ without Jupyter - from the command line, or from a script etc.


1 Like

Thank you very much for the swift reply.

4 posts were split to a new topic: How to use PYNQ without Jupyter from the command line