[USRP-users] Problem with error_code_late_command using two x310's

[Wolsieffer, Carl L. ERDC-RDE-CRL-NH CIV via USRP-users]

Hello all,

I have 2 x310's, one is a transmitter and the other is a receiver. I have some 
code I adapted from a usrp example awhile back and in testing the bounds of my 
system, I am getting a ERROR_CODE_LATE_COMMAND warning after a few seconds when 
trying to receive every few ms and think I may be misinterpreting at least 
partially what my code is doing. I can run the code for up to 5,000ish loops so 
I'm guessing something is eventually losing ground somewhere and fails

It all operates out of one .cpp program and manages a send_from_file and 
recv_to_file thread.

I'm guessing the problem is on the receive side, and essentially the 
recv_to_file function goes something like this:

stream_cmd.num_sampes = num_requested_samples;
stream_cmd.stream_now = false;
stream_cmd.time_spec = uhd::time_spec_t( recv_time );
rx_stream->issue_stream_cmd( stream_cmd );

while (not stop_signal_called) {
        
        num_samples = rx_stream->recv( buffer_ptrs, sampls_per_buff, meta_data, 
timeout );

        total_num_samps = total_num_samps + num_samps;

        ///// Receive for 20 us //////
        ///// basically it's listening for a 20us long signal sent from 
transmitter (sent at "recv_time" as well)
        ///// writes to outfile ////////
        ///// check for errors ////////
        
        // end of transmitted file presumably reached
        If ( num_total_samples >= num_requested_samples) {
        
                num_total_samples = 0;
                
                recv_time = recv_time + delay; // increment delay by 1 ms have 
tried as high as 10 ms with no luck after about 10,000 loops or so
                stream_cmd.time_spec = uhd::tune_spec_t( recv_time );
                rx_stream->issue_stream_cmd( stream_cmd );
                
                for (size_t i=0; i< buffs.size(); i++) {
                        buff_ptrs.push_back(&buffs[i].front());
                }
        }
}


I'm thinking it may just be an error in how I've interpreted the example code 
and repurposed it. I don't have a strong understanding of the difference 
between issue_stream_cmd() and recv() which could be part of the problem as 
well. This is a simplified version of the code to explain what's happening so 
please let me know if you need additional details


Thanks
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com