Hi Rob, Thank you for the suggestion. The ID is “TxStreamer#0”. I also found that disconnecting a streamer does not fully restore its status so I need to create new streamers every time. This works for me now. I opened an issue<https://github.com/EttusResearch/uhd/issues/903> though.
Regards, Nan From: Rob Kossler <[email protected]> Sent: Tuesday, January 27, 2026 7:02 AM To: Yang, Nan <[email protected]> Cc: [email protected] Subject: [EXTERNAL] Re: [USRP-users] How to disconnect streamers on RFNoC graph I'm not sure but I seem to recall that the streamer ID could not be queried but that it used a naming convention (something like "tx_streamer_1") that you could guess and use to successfully disconnect the graph. Perhaps if you log DEBUG or TRACE level messages (or inspect Ettus code), you can see what the stream ID is. I know that this is not the right answer, but it might get you by if the API is missing the needed functions to query the ID. Rob On Mon, Jan 26, 2026 at 5:52 PM Yang, Nan via USRP-users <[email protected]<mailto:[email protected]>> wrote: Hello UHD folks, I’m trying to buffer all 4 channels tx/rx data on DRAM on X410. So my application does 1. Connect tx_streamer to replay block and write samples to DRAM 2. Disconnect tx_streamer 3. Connect replay block to radio block 4. Do tx and rx 5. Disconnect radio blocks 6. Connect replay block to rx_streamer and read samples I don’t find the correct API for step 2. The uhd::rfnoc::rfnoc_graph::disconnect() function takes std::string &streamer_id, however, I can’t find a public API that returns a streamer_id. The connect() function accepts uhd::tx_streamer::sptr and uhd::rx_streamer::sptr, I wonder if disconnect() should accept same argument types. If I skip step 2, I get an error in step 3. Regards, Nan _______________________________________________ USRP-users mailing list -- [email protected]<mailto:[email protected]> To unsubscribe send an email to [email protected]<mailto:[email protected]>
_______________________________________________ USRP-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
