Setting a large number of UHD buffers can help quite a bit. In the GRC
UHD Sink block device address:
"num_send_frames=256"
In Python:
self.uhd_usrp_sink_0_0 = uhd.usrp_sink(
",".join(("num_send_frames=256", "")),
uhd.stream_args(
cpu_format="fc32",
channels=range(1),
),
)
Ron
On 5/2/19 05:18, Matthias Brändli via USRP-users wrote:
Dear all,
I'm maintaining ODR-DabMod[1], a Digital Audio Broadcasting modulator
which uses UHD to drive a USRP at 2048ksps, in most applications a B200.
ODR-DabMod runs the modulator and the UHD output in separate threads, to
ensure that a few modulated frames are always ready when the USRP needs
one[2]. It requests realtime scheduling[3]. The system runs headless (no
X11 running), and generally doesn't have anything else running.
On some systems, we observe sporadic underruns. They occur in irregular
intervals, something like once a minute, sometimes rarer. Seen with both
USB2 and USB3 hosts, over several UHD versions. Until now we've mostly
been able to avoid the issue by selecting machines that don't show the
issue, but we hope we can find a mitigation in software.
Both the data source the modulator connects to and the USRP have a
disciplined time reference, implying there is no rate drift (plus, that
would trigger regular underruns, not sporadic ones).
There are some parameters described in USB Transport Parameters[4], does
it make sense to change those?
Could CPU frequency scaling lead to interruptions?
Are there other knobs to turn that I'm not aware of? Device, stream or
linux kernel settings?
Is the approach of using a separate thread for UHD output beneficial?
Anything else?
Sorry for the vague description of the issue and the many questions, but
I'm a bit out of ideas about how to approach that. If anybody had
similar problems I'd be curious to hear your experiences.
Cheers
mpb
[1]
https://github.com/Opendigitalradio/ODR-DabMod
[2]
Look for m_queue in
https://github.com/Opendigitalradio/ODR-DabMod/blob/master/src/output/SDR.cpp
[3]
For all threads involved in DSP and the SDR output thread. Grep for
`set_realtime_prio`.
[4]
https://files.ettus.com/manual/page_transport.html#transport_usb
Default send_frame_size seems to be on line 85 of
https://files.ettus.com/manual/b200__impl_8hpp_source.html
_______________________________________________
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
_______________________________________________
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
