[Discuss-gnuradio] Issue with current RBF?

[Michael Dickens]

We recently gave some quite successful demos for a sponsor.  In order  
to get one of the demos to work, we had to revert the RBF files from  
the latest SVN trunk ... we chose r4287 < http://www.gnuradio.org/trac/changeset/4287 
 > because that's when the check in to fix the issue of turning off  
the TX immediately after using it (for all practical purposes).  There  
might be more recent changesets that would work, but we didn't have  
time to test them out.  Our program is a variant of benchmark_tx and  
_rx in the digital directory - admittedly before the transition from  
gr.flow_graph to gr.top_block - but a quick update seems to have done  
no harm.

The issue we're seeing is that when the USRP_TX is turned off in  
Python, the USRP will still TX what was there before.  We can very  
clearly see this with out demo as well as a spectrum analyzer.  This  
happens using the latest trunk RBF, but not the RBF from r4287.

We're working on Ubuntu 7.10, latest updates, but it happens using OSX  
10.4 and 10.5 as well - so I think it's the RBF or benchmark_XX code  
at issue.

Anyone have any ideas that we might try to further track down the  
issue, or thoughts as to where to correct this issue in the benchmark*  
code?  Thanks in advance. - MLD