Do you mind trying an update?
I have experienced the same issue with B100 and UHD:
linux; GNU C++ version 4.6.3; Boost_104900;
UHD_003.004.002-137-g49d4f8e4
the SID message appears in first overrun and after this the B100
device needs to power down otherwise the SID message not disappears.
I tried various combinations of versions from git repo (main, master,
next) with the respective images and the problem persist. I use Gentoo
Linux 64bit, kernel 3.4.0, glibc 2.15, gcc 4.6.3
Sumit Kr.
Research Assistant
Communication Research center
IIIT Hyderabad
India
I think the problem is related to the host code, because i have an
UBUNTU installations where the SID message appears less times and b100
hardware no need a reboot to working correctly.
In GENTOO i have this version of the UHD:
linux; GNU C++ version 4.6.3; Boost_104900;
UHD_003.004.002-137-g49d4f8e4
I have my B100 with SBX about one week and in GENTOO freezes
constantly but in UBUNTU this happens less times.
I will try to compile UHD host code with the same “BOOST” version of
ubuntu in my gentoo to view if the problem persisting and then inform
the result.
Alright, I am getting out my big hammer. Not only will this code be
fixed, but a new feature will be added!
-josh
Hi Josh,
Has the big hammer been acquired? I still get these control errors all
the
time, requiring manual intervention (ie, un-powering and re-powering the
radio) to recover. This means I cant use USRP remotely which will
severely
limit my ability to use Ettus hardware…
After compiling with different versions of the library BOOST
(1.46.1,1.47.0,1.48.0,1.49.0) the problem remained, so it is not the
library BOOST the blame of the SID message problem.
But in my case the problem has become much less evident after
recompile the library LIBUSB, what I think happened was that the
library will have built with gcc4.5.3 (x86_64-pc-linux-gnu-4.5.3) and
compiled after with gcc-4.6.3 (x86_64-pc-linux-gnu-4.6.3) the SID
message problem rarely manifests itself.
I currently have libusb-1.0.9 and rarely see the SID messages, t think
the problem is related to libusb in my case
Linux
uname -a
Linux radio 3.4.4-gentoo #1 SMP PREEMPT Mon Jun 25 10:58:07 WEST 2012
x86_64 Intel(R) Core™ i7 CPU 950 @ 3.07GHz GenuineIntel GNU/Linux
gcc-4.6.3 (x86_64-pc-linux-gnu-4.6.3)
glibc-2.15
UHD
linux; GNU C++ version 4.6.3; Boost_104900;
UHD_003.004.002-162-g485fe587
I was getting both, however moving to a physical platform (instead of
VM)
seems to have resolved a few isues, the control problems are still an
issue. I’ll see if I can pull the new branch and try the new driver.
Has the big hammer been acquired? I still get these control errors all the
time, requiring manual intervention (ie, un-powering and re-powering the
radio) to recover. This means I cant use USRP remotely which will severely
limit my ability to use Ettus hardware…
I was waiting to announce the work until I get the E100 work squashed.
But you can grab the next branch, it has B100 bus work and timed
commands capability. Its worth trying.
But I’m not sure that this was related to your USB issue. I have a vague
recollection, and I recall the USB driver was crashing and unrecoverable
after the process was killed.
FWIW, as the code stands now, on the latest release and up, the FX2 is
reset and remunerated if its uncommunicable for some reason (so long as
the control endpoint to the FX2 remains up). But is that not working for
you?
-josh
This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.