Re: [Discuss-gnuradio] RF recorder

[Adam Sampson]

Justin Zygmont <[EMAIL PROTECTED]> writes:

> I was looking for a program that will record input from the soundcard
> only when there is activity.

Would this do what you're after?
  

-- 
Adam Sampson <[EMAIL PROTECTED]>

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

[Discuss-gnuradio] Pipe Tricks

[cswiger]

Trying

  pipe_fd = open("/pipe","r")
  gr.file_descriptor_source(gr.sizeof_char,pipe_fd.fileno())

starts out but as soon as I start the source to fill the pipe
I get:

  file_descriptor_source[read]: Bad file descriptor

and the pipe filling script reports IOError: [Errno 32] Broken pipe

BTW, opening and reading the pipe manually in python works ok:
>>> fd=open("/pipe","r")
>>> a=fd.read(5)
>>> a
'\xaa\xaa\xaa\xaa\xaa'


where \xaa is the expected idle character read w/o having to close
the pipe at the source. Again, using gr.file_source(pipe) works
fine but only after the source closes the pipe. Oh, and yes there
is fd.flush() after ever fd.write(chr(c)) in the source script.

--Chuck


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Pipe Tricks

[James Cooley]

I had a similar thing I was using sockets, but then got it to work.
Are you sure the fd is staying in scope for the duration of the run?
-jamie

cswiger wrote:
Trying
 pipe_fd = open("/pipe","r")
 gr.file_descriptor_source(gr.sizeof_char,pipe_fd.fileno())
starts out but as soon as I start the source to fill the pipe
I get:
 file_descriptor_source[read]: Bad file descriptor
and the pipe filling script reports IOError: [Errno 32] Broken pipe
BTW, opening and reading the pipe manually in python works ok:
 

fd=open("/pipe","r")
a=fd.read(5)
a
   

'\xaa\xaa\xaa\xaa\xaa'
where \xaa is the expected idle character read w/o having to close
the pipe at the source. Again, using gr.file_source(pipe) works
fine but only after the source closes the pipe. Oh, and yes there
is fd.flush() after ever fd.write(chr(c)) in the source script.
--Chuck
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
 

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Pipe Tricks

[cswiger]

On Wed, 6 Apr 2005, cswiger wrote:

> Trying
>
>   pipe_fd = open("/pipe","r")
>   gr.file_descriptor_source(gr.sizeof_char,pipe_fd.fileno())
>
> starts out but as soon as I start the source to fill the pipe
> I get:
>
>   file_descriptor_source[read]: Bad file descriptor
>

Ok, turns out you have to use

  import os
  source_fd = os.open("/pipe", os.O_RDONLY)
  src = gr.file_descriptor_source(gr.sizeof_char,source_fd)

low level io.

--Chuck


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] CVS check out and then failed to build... (Further failures...)

[John Clark]

Eric Blossom wrote:
On Mon, Apr 04, 2005 at 05:39:35PM -0700, John Clark wrote:
 

John Clark wrote:
   

I just checked out the most recent CVS sources, using the procedure:
 

I then updated my version of 'autoconfig', and 'automake' to be the most
recent on ftp.gnu.org, and got the following: (truncated for brevity 
sake...),
but ultimately ending with a message to the effect of needing define
AC_PROG_LIBTOOL and re-run.
   

Reinstall libtool also.
 

Actually there seems to be a problem there as well. From another post:
LRK wrote:
On Mon, Apr 04, 2005 at 05:39:35PM -0700, John Clark wrote:
 

I just checked out the most recent CVS sources, using the procedure:
 

I then updated my version of 'autoconfig', and 'automake' to be the most
recent on ftp.gnu.org, and got the following: (truncated for brevity 
sake...),
but ultimately ending with a message to the effect of needing define
AC_PROG_LIBTOOL and re-run.
   

I have been trying to get the CVS compile working on FreeBSD and ran into
that one.
Aclocal19 looks into /usr/local/share/aclocal19/ for libtool.m4
and libtool puts it in /usr/local/share/aclocal/libtool15.m4.
So I put a link in there where aclocal can find the file.
 

After some set of upgrades and not-obvious-links I got the 
'for-all-dirs' to get to the point
of the CPPUNIT testing phase, and got this (python version is 2.3.3 ).

>>> gr_fir_fff: using SSE
.
--
Ran 1 test in 0.065s
OK
.E
==
ERROR: test_gru_import (__main__.test_head)
--
Traceback (most recent call last):
File "./qa_kludged_imports.py", line 39, in test_gru_import
from gnuradio import gru
File
"/home/GNURadio/gnuradio-core/src/python/gnuradio/gru/__init__.py",
line 37, in ?
exec "from gnuradio.gruimpl.%s import *" % (f,)
File "", line 1, in ?
File
"/home/GNURadio/gnuradio-core/src/python/gnuradio/gruimpl/freqz.py",
line 57, in ?
import Numeric
ImportError: No module named Numeric
--
Ran 2 tests in 0.111s
FAILED (errors=1)
.
--
Ran 5 tests in 0.009s
OK
...
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] CVS check out and then failed to build... (Further failures...)

[LRK]

On Wed, Apr 06, 2005 at 09:19:47AM -0700, John Clark wrote:
> 
> After some set of upgrades and not-obvious-links I got the 
> 'for-all-dirs' to get to the point
> of the CPPUNIT testing phase, and got this (python version is 2.3.3 ).
> 
> >>> gr_fir_fff: using SSE
> .
> --
> Ran 1 test in 0.065s
> 
> OK
> .E
> ==
> ERROR: test_gru_import (__main__.test_head)
> --
> Traceback (most recent call last):
> File "./qa_kludged_imports.py", line 39, in test_gru_import
> from gnuradio import gru
> File
> "/home/GNURadio/gnuradio-core/src/python/gnuradio/gru/__init__.py",
> line 37, in ?
> exec "from gnuradio.gruimpl.%s import *" % (f,)
> File "", line 1, in ?
> File
> "/home/GNURadio/gnuradio-core/src/python/gnuradio/gruimpl/freqz.py",
> line 57, in ?
> import Numeric
> ImportError: No module named Numeric
> 
> --
> Ran 2 tests in 0.111s
> 
> FAILED (errors=1)
> .
> --
> Ran 5 tests in 0.009s
> 
> OK
> ...


Glad to hear you are up to this point. I get the same error. :)


That was introduced in the 2.5 tarballs as well, wasn't in the 2.4 version.


-- 
LRK
[EMAIL PROTECTED]

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] Pipe Tricks

[Eric Blossom]

On Wed, Apr 06, 2005 at 11:28:18AM -0400, James Cooley wrote:
> I had a similar thing I was using sockets, but then got it to work.
> 
> Are you sure the fd is staying in scope for the duration of the run?
> 
> -jamie
> 

Chuck,

Jamie has it right.  Most likely the pipf_fd was going out of scope,
and then was being closed.  That's why in my orignal message I used
something like self.pipe_fd = open("/pipe", "r").  You need to hold a
reference to the return value of open for the duration of the run.

Eric

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] CVS check out and then failed to build... (Further failures...)

[Eric Blossom]

On Wed, Apr 06, 2005 at 09:19:47AM -0700, John Clark wrote:
> Eric Blossom wrote:
> 
> 
> After some set of upgrades and not-obvious-links I got the 
> 'for-all-dirs' to get to the point
> of the CPPUNIT testing phase, and got this (python version is 2.3.3 ).
> 
> >>> gr_fir_fff: using SSE
> .
> --
> Ran 1 test in 0.065s
> 
> OK
> .E
> ==
> ERROR: test_gru_import (__main__.test_head)
> --
> Traceback (most recent call last):
> File "./qa_kludged_imports.py", line 39, in test_gru_import
> from gnuradio import gru
> File
> "/home/GNURadio/gnuradio-core/src/python/gnuradio/gru/__init__.py",
> line 37, in ?
> exec "from gnuradio.gruimpl.%s import *" % (f,)
> File "", line 1, in ?
> File
> "/home/GNURadio/gnuradio-core/src/python/gnuradio/gruimpl/freqz.py",
> line 57, in ?
> import Numeric
> ImportError: No module named Numeric
> 
> --
> Ran 2 tests in 0.111s
> 
> FAILED (errors=1)
> .
> --
> Ran 5 tests in 0.009s
> 
> OK
> ...

Sorry about that.  I'll update the README.

We are now requiring the Numeric python library for gnuradio-core.
FYI, it was required for gr-wxgui in previous releases.

See:

  http://numeric.scipy.org/
  http://sourceforge.net/project/showfiles.php?group_id=1369&package_id=1351

Eric

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

[Discuss-gnuradio] Working PSK31

[cswiger]

Announcing a working audio-out psk31 program suitable for
driving a ssb transmitter. It doesn't have many creature
comforts like many  avaiable programs (linpsk, gpsk31, etc)
but it does work and will shortly drive a USRP directly,
which is the point of the operation.
By tweaking the number of taps in the root-raised-cosine
pulse shaping we could get the IMD down to -42 to -50db.
Someday it may run in wxpython but for now it's two terminal
windows.

http://webpages.charter.net/cswiger/psk_experiment.html

--Chuck


___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] usrper fails in usb_claim_interface

[Larry Doolittle]

On Tue, Apr 05, 2005 at 02:18:19PM -0700, Eric Blossom wrote:
> On Tue, Apr 05, 2005 at 12:03:00PM -0700, Larry Doolittle wrote:
> > 
> > I have an Avnet Virtex 4 evaluation board, intended for
> > projects similar in concept to GNU Radio.  It uses the
> > same CY7C68013 chip as USRP.
> > 
> > I used USRP as a software reference design, and started
> > splicing in some code of my own. [chop]
> > 2005-04-05 11:20:16.264596500 <6>usbtest 5-1:1.0: FX2 device
> > 2005-04-05 11:20:16.264640500 <6>usbtest 5-1:1.0: high speed {control 
> > bulk-in bulk-out} tests (+alt)
> --^^^
> 
> You have the usbtest module loaded.
> rmmod it.

Eric is right, when I rmmod usbtest this problem goes away.
Someplace in the bowels of debian sid this module is configured
to reload every time I plug in the device, and I then have to
rmmod it again by hand.  "find /etc -type f | xargs grep usbtest"
didn't find the magic.

So, I'm on to bigger and better problems!

Somehow, when I hacked up the Makefile for my custom FX2 build,
I left off
MEMOPTS = --code-loc 0x --code-size 0x1800 --xram-loc 0x1800 \
 --xram-size 0x0800 -Wl '-b USBDESCSEG = 0xE000'

Deadly!  Once I fixed that:

# usrper -x load_firmware ${FOO}/usrp_main.ihx 
lt-usrper: found unconfigured FX2; needs firmware.
# lsusb
Bus 005 Device 051: ID fffe:0001  
Bus 005 Device 001: ID :  
Bus 004 Device 001: ID :  
Bus 003 Device 001: ID :  
Bus 002 Device 001: ID :  
Bus 001 Device 001: ID :  
# usrper i2c_read 0 16

# 

Cool!  It looks like my code is in there, attempting
to read the JTAG ID chain, and getting all 1's.  Not
the right answer, of course, but I'm on my way.

 - Larry


signature.asc
Description: Digital signature
___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] usrper fails in usb_claim_interface

[Matt Ettus]

Quoting Larry Doolittle <[EMAIL PROTECTED]>:

> On Tue, Apr 05, 2005 at 02:18:19PM -0700, Eric Blossom wrote:
> > On Tue, Apr 05, 2005 at 12:03:00PM -0700, Larry Doolittle wrote:
> > >
> > > I have an Avnet Virtex 4 evaluation board, intended for
> > > projects similar in concept to GNU Radio.  It uses the
> > > same CY7C68013 chip as USRP.
> > >
> > > I used USRP as a software reference design, and started
> > > splicing in some code of my own. [chop]
> > > 2005-04-05 11:20:16.264596500 <6>usbtest 5-1:1.0: FX2 device
> > > 2005-04-05 11:20:16.264640500 <6>usbtest 5-1:1.0: high speed {control
> bulk-in bulk-out} tests (+alt)
> > --^^^
> >
> > You have the usbtest module loaded.
> > rmmod it.
>
> Eric is right, when I rmmod usbtest this problem goes away.
> Someplace in the bowels of debian sid this module is configured
> to reload every time I plug in the device, and I then have to
> rmmod it again by hand.  "find /etc -type f | xargs grep usbtest"
> didn't find the magic.


Once you reprogram the EEPROM to something other than the default device,
usbtest stops loading itself.

Matt

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Re: [Discuss-gnuradio] CVS check out and then failed to build... (Further failures...)

[LRK]

On Wed, Apr 06, 2005 at 12:02:00PM -0700, Eric Blossom wrote:
> On Wed, Apr 06, 2005 at 09:19:47AM -0700, John Clark wrote:
> >
> > import Numeric
> > ImportError: No module named Numeric
> 
> Sorry about that.  I'll update the README.
> 
> We are now requiring the Numeric python library for gnuradio-core.
> FYI, it was required for gr-wxgui in previous releases.

Ok. Now it builds.



Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython installed but "import wx" fails. Seems like there
was mention of that problem recently but I can't find it.

Need another clue.


Ok. Now it builds.


I think I have wxPython-2.5.4.1 installed correctly but "import wx" fails. 
Seems like there was mention of that problem recently but I can't find it.

Need another clue.


-- 
LRK
[EMAIL PROTECTED]

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

[Discuss-gnuradio] Trouble building GnuRadioPython.cc

[Will Ware]

I had this problem:
<
In file included from ../../../src/gnu/lib/gr/GrHilbert.h:29,
from GnuRadioPython.cc:1482:
../../../src/gnu/lib/gr/gr_fir.h:25:2: warning: #warning "gr_fir.h is
deprecated.  Please transition to gr_fir_???.h"
GnuRadioPython.cc: In function `PyObject*
  _wrap_new_sdr_1000_base__SWIG_1(PyObject*, PyObject*)':
GnuRadioPython.cc:8747: error: no matching function for call to `sdr_1000_base
  ::sdr_1000_base()'
../../../src/gnu/lib/grio/sdr_1000.h:31: error: candidates are:
  sdr_1000_base::sdr_1000_base(const sdr_1000_base&)
../../../src/gnu/lib/grio/sdr_1000.h:37: error:
  sdr_1000_base::sdr_1000_base(int)
../../../src/pspectra/lib/vr/VrSource.h: At top level:
GnuRadioPython.cc:231: warning: `swig_type_info*
  SWIG_TypeDynamicCast(swig_type_info*, void**)' defined but not used
GnuRadioPython.cc:419: warning: `const char* SWIG_UnpackDataName(const char*,
  void*, unsigned int, const char*)' defined but not used
GnuRadioPython.cc:499: warning: `void SWIG_PropagateClientData(swig_type_info*)
  ' defined but not used
GnuRadioPython.cc:1198: warning: `void* SWIG_Python_MustGetPtr(PyObject*,
  swig_type_info*, int, int)' defined but not used
GnuRadioPython.cc:1212: warning: `int SWIG_Python_ConvertPacked(PyObject*,
  void*, unsigned int, swig_type_info*, int)' defined but not used
>

I checked the mailing list archive and found a slightly similar
looking problem discussed on 25 July 2003, but on more careful
inspection it looks like a different problem. This almost seems to be
misbehavior on the part of SWIG. My g++ version is 3.3.3 and my SWIG
version is 1.3.24. This is on a Fedora Core 2 box with a KDE desktop.
Any suggestions?

Thanks much
Will Ware

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio