[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Discuss-gnuradio] USRP2 abnormal function with spectrum_sense code
From: |
Eric Blossom |
Subject: |
Re: [Discuss-gnuradio] USRP2 abnormal function with spectrum_sense code |
Date: |
Sat, 3 Jan 2009 10:50:32 -0800 |
User-agent: |
Mutt/1.5.18 (2008-05-17) |
On Sat, Jan 03, 2009 at 03:45:22PM +0900, ILKYOUNG KWOUN wrote:
> Hi, folks. Happy New Year!
>
> With the nice post of Firas' 'usrp_spectrum_sense.py explanation'
> post, I slightly modified the code to work with my USRP2 box.
>
> 1. When I interrupt the execution of the code with ^C, I the following
> error message
> ------------------------------------------------------------------
> thread-per-block[4]: <gr_block bin_statistics_f (5)>]: caught
> unrecognized exception
> ------------------------------------------------------------------
> I guess this is because I forced to terminate the flow graph in
> the middle of its execution. I wonder if there is any 'nice way' to
> stop the execution normally so that I don't get the error above and
> the problem #2 below.
That's for posting the problem and the code that reproduces it.
We've got some control-C handling problems that need attention.
Because of the incorrect shutdown, the USRP2 isn't getting shutdown
properly, and is probably still streaming samples.
That's http://gnuradio.org/trac/ticket/276
> 3. I also experimented the code with pre-saved sample file. When I run
> the code with 'usrp2_spectrum_sense.py -i SomePreSavedSamples.dat' and
> terminate the execution with ^C, I get 'Segmentation fault' and heavy
> core file dumped.
> I wonder what is the most popular way to debug the core file to
> track down the problem? Is the gdb best as usual? Is there anybody to
> share the debugging tricks or tips ?
Use gdb. See these GNU Radio specific instructions on connecting to
the python process that has all the GR stuff loaded:
http://www.gnu.org/software/gnuradio/doc/howto-write-a-block.html#debugging
Once the problem occurs try:
> bt
or
> thread apply all bt
Eric