Yikes Kevin, you always seem to find the tricky issues. As you say, the Python API is just a wrapper of the C++ API, so I can't think of a good reason why it would behave differently. Is it possible that the Python app does a different setup of the block than the C++ app, leading to a state where zero is expected?
If you're really pulling at straws, you could capture a session with Wireshark, install this dissector: https://github.com/EttusResearch/uhd/blob/master/tools/dissectors/lua/rfnoc.lua, and then see if the response packet to the peek is really carrying a zero payload or not. But I admit it's a bit of an elaborate setup. --M On Thu, May 8, 2025 at 11:30 AM Kevin Williams <kevin.willi...@vastech.co.za> wrote: > Hi, > > > > I’ve got a strange issue reading rfnoc block registers in python. > > > > I have a register populated by windowed statistics in my rfnoc block. > > > > In my C++ app I can read this register and see it is being updated at > exactly the right time intervals, but in python this register is mostly > read back (using my block driver methods – i.e. the same C++ methods but > now automatically wrapped for python by the normal compile) as just zero. > > > > The update interval is relatively long at 200 ms. > > > > Is there any possible reason why this might be? > > > > Kind regards, Kevin > > > _______________________________________________ > USRP-users mailing list -- usrp-users@lists.ettus.com > To unsubscribe send an email to usrp-users-le...@lists.ettus.com >
_______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com
