Re: [Discuss-gnuradio] usrp & ppc-linux

[Eric Blossom]

On Thu, Feb 10, 2005 at 01:32:21AM -0500, kilian wrote:
> 
> Hi,
> 
> I received the usrp a short time ago and was so far able to at least
> compile the basic toolset comming with the usrp-0.7 tarball. The host is
> a dual address@hidden running yellowdog-linux-4.0 . I had no luck compiling
> gnuradio for it though. This is acceptable, since I was just using the
> Mac for convenience (I can switch to a PC). This article:
> 
> http://lists.gnu.org/archive/html/discuss-gnuradio/2003-12/msg00016.html
> 
> mentioned USB performance and now I would like to add another data point
> to it. After I found out that I had to upload the firmware first, using
> 'burn-usrp2-eeprom', I ran test_usrp0 and got the following output:

First off do you have a rev0 or rev2 board?  The rev0 board doesn't
have any daughterboards, just SMA connectors on the board.  The rev2
board has connectors for daughterboards and is 160 x 160 mm.

If you used burn-usrp2-eeprom and you have a usrp0 board,  you just
fried the board.  In 99.9999% of the cases there is no reason to use
burn-usrp2-eeprom.  The motherboards come with the boot eeprom already
burned.  Putting the wrong image in there can permanently kill your
board.  [Part of what's in there is the information that tells us what
kind of board you have, so we can be sure to load the right firmware
and fpga configuration.]

# ./test_usrp0 test usrp0 boards

test_usrp0 tests only rev0 usrp boards.  No harm done, it just won't
give good results.  I have updated CVS so that it bails on non-rev0 
boards.


Try test_usrp_standard_tx and test_usrp_standard_rx.

With test_usrp_standard_tx mess around with the -I <interp> option to
control that data rate across the USB.  The sample rate across the USB
on transmit is 128e6 / <interp>.  The data rate in bytes/sec across
the USB is sample_rate * 4.  You should see a sine wave out I & Q.

Likewise, try test_usrp_standard_rx (use the CVS version, the tarball
version's got a problem).  Mess with the -D <decim> option to
control the sample rate across the USB.  The sample rate across the
USB on receiver is 64e6 / <decim>.  The data rate in bytes/sec across
the USB is sample_rate * 4. 

> Well, this not too great, although better than the previous article
> suggests some NEC chipsets are. I get no overruns on the PC. I'm a bit
> puzzled about the fact that I get no underruns while transmitting (maybe
> the test doesn't include that but the target should signal an underrun
> somehow). 

Try with tst_usrp_standard_{rx,tx} and let us know how that goes.

Also, grep the top level Makefile for FUSB_TECH.  Does it say
FUSB_TECH = linux


> In case you are wondering what kind of hardware I'm using, 'cat
> /proc/bus/usb/devices'  shows in part:
> 
> ...
> 
> T:  Bus=01 Lev=00 Prnt=00 Port=00 Cnt=00 Dev#=  1 Spd=480 MxCh= 5
> B:  Alloc=  0/800 us ( 0%), #Int=  0, #Iso=  0
> D:  Ver= 2.00 Cls=09(hub  ) Sub=00 Prot=01 MxPS= 8 #Cfgs=  1
> P:  Vendor=0000 ProdID=0000 Rev= 2.06
> S:  Manufacturer=Linux 2.6.8-1.ydl.7g5-smp ehci_hcd
> S:  Product=NEC Corporation USB 2.0
> S:  SerialNumber=0001:02:0b.2
> C:* #Ifs= 1 Cfg#= 1 Atr=e0 MxPwr=  0mA
> I:  If#= 0 Alt= 0 #EPs= 1 Cls=09(hub  ) Sub=00 Prot=00 Driver=hub
> E:  Ad=81(I) Atr=03(Int.) MxPS=   2 Ivl=256ms
> 
> T:  Bus=01 Lev=01 Prnt=01 Port=02 Cnt=01 Dev#= 12 Spd=480 MxCh= 0
> D:  Ver= 2.00 Cls=ff(vend.) Sub=ff Prot=ff MxPS=64 #Cfgs=  1
> P:  Vendor=fffe ProdID=0002 Rev= 1.02
> S:  Manufacturer=Free Software Folks
> S:  Product=USRP Rev 2
> S:  SerialNumber=3.141593
> C:* #Ifs= 3 Cfg#= 1 Atr=c0 MxPwr=  0mA
> I:  If#= 0 Alt= 0 #EPs= 0 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
> I:  If#= 1 Alt= 0 #EPs= 1 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
> E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
> I:  If#= 2 Alt= 0 #EPs= 1 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
> E:  Ad=86(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
> 
> I'm wondering whether this is a valid test for USB performance. If
> anybody has found a good PCI based USB adapter please let me know.  
> 
> I mainly got the board for educational purposes and possibly to build a
> receiver (to cover 1-30MHz as a first step). Now that the possibilities
> (4RX/4TX channels) are starting to sink in, I'm getting all kinds of
> ideas though (i.e. using the DACs for DDS, ...). 
> 
> I'm curious what the unpopulated J2001 can be used for.

They're all grounds. See the schematics at ettus.com


> I have yet to test the board with some actual input, but I'm pretty
> happy with what I got so far. I've taken part in more difficult
> projects, one annoying thing there being tightlipped hardware vendors.
> The fact that one can easily get all the source code is awesome.
> 
> Thanks a lot,
> 
> Kilian