Custom Blocks and FPGA builds

GNU Radio
1

Hi

I just wanted clarification on somethings:

  1. For creating customs blocks, I wasn’t sure on the
    compiling process. Once you’ve generated the three
    files needed for your custom block, and placed them in
    the correct place, does one simply compile them in
    their folder using the ./bootstrap → ./configure →
    etc. process? Is it only the new files that are
    compiled and integrated? Or does this process
    re-compile and install the whole gnuradio setup?

  2. For custom FPGA builds. So far I’ve been creating
    and compiling custom FPGA builds for the rev 2 USRP
    that I’m using. My fellow students are all using rev 4
    boards. I seem to vaguely remember Eric mentioning
    that one should be cautious about where custom .rbfs
    are placed for rev 4 boards, and how they are called
    from python.

If the rev 4 users would like to use the rbf’s I’ve
made, what is the safe way for them to load it from
python? Do I need to recompile my rbf’s differently to
work with rev 4 boards?

Regards

Lance

  ____________________________________________________________________________________

Be a better friend, newshound, and
know-it-all with Yahoo! Mobile. Try it now.
http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

2

On Wed, Jan 16, 2008 at 04:29:28AM -0800, seph 004 wrote:

compiled and integrated? Or does this process
re-compile and install the whole gnuradio setup?

The easiest way to create new blocks is to build them “outside the
tree”, as shown in gr-howto-write-a-block. That way they’re available
to you, and your Makefiles don’t get hosed everytime we make a change
on the trunk.

python? Do I need to recompile my rbf’s differently to
work with rev 4 boards?

Currently we run the same RBF’s for rev 2 and rev 4.
We do have infrastructure in place that loads them from different
paths. E.g., /usr/local/share/usrp/rev2 vs rev4.
Copy your .rbf into both.

Eric