On 11/03/2020 02:19 PM, Luke Stutters wrote:
I have only succeeded in running a B210 on a Raspberry Pi at lower
data rates (closer to 12MS) so your experience is consistent with mine.
In certain very-simple DSP flows, I've achieved 14Msps on an Odroid
XU4--which is spec-similar to the Rpi4 B.
CPU/Memory/IO performance really matters. Just because the system has a
USB3 interface does NOT mean that it can
sustain high rates. Even just moving samples through your system,
without doing anything to them (as in the benchmark_rate
example) requires code-paths that are at least several hundred
instructions long. Multi-core doesn't necessarily help much with
this because there's no performant way to effectively parallelize the
simple process of pulling samples off of USB and getting them
into the user layer. The SMP aspects of most modern systems only
really start to "shine" when you have a DSP work-flow with
lots of steps that can each benefit from running in their own
thread. But you *still* have a rate-limiting step of getting the samples
out of the device and into the work-flow. In that portion, system
details matter A LOT.
On Tue, 3 Nov 2020 at 17:20, Marcus D. Leech via USRP-users
<usrp-users@lists.ettus.com <mailto:usrp-users@lists.ettus.com>> wrote:
On 11/03/2020 10:49 AM, Alvaro Pendas Recondo via USRP-users wrote:
> Hello,
>
> I am using a USRP B200mini with a sampling rate of 40MS that works
> perfectly fine connected to a laptop with USB 3. However, when I
> connect it to a Raspberry Pi 4 (which has two USB 3 ports) and I
run
> the example benchmark_rate with the same sampling rate I get the
> output that I copy below. It seems that even if it is also
operating
> over USB 3, this connection cannot meet the expected performance
> anymore. If I reduce the sampling rate (down to 12 MS approx)
> everything works fine. Any ideas about what might be causing
this problem?
>
>
> By the way, I already followed all the instructions explained at
>
https://kb.ettus.com/USRP_Host_Performance_Tuning_Tips_and_Tricks#:~:text=Thread%20Priority%20Scheduling,-When%20UHD%20spawns&text=To%20address%20this%20issue%2C%20non,%2Fetc%2Fsecurity%2Flimits
<https://kb.ettus.com/USRP_Host_Performance_Tuning_Tips_and_Tricks#:%7E:text=Thread%20Priority%20Scheduling,-When%20UHD%20spawns&text=To%20address%20this%20issue%2C%20non,%2Fetc%2Fsecurity%2Flimits>
>
<https://kb.ettus.com/USRP_Host_Performance_Tuning_Tips_and_Tricks#:%7E:text=Thread%20Priority%20Scheduling,-When%20UHD%20spawns&text=To%20address%20this%20issue%2C%20non,%2Fetc%2Fsecurity%2Flimits>.
>
>
>
> ./benchmark_rate --rx_rate 40e6 --tx_rate 40e6
>
> [INFO] [UHD] linux; GNU C++ version 8.3.0; Boost_106700;
> UHD_3.15.0.HEAD-0-gaea0e2de
> [INFO] [B200] Loading firmware image:
> /usr/local/share/uhd/images/usrp_b200_fw.hex...
> [00:00:00.000044] Creating the usrp device with: ...
> [INFO] [B200] Detected Device: B200mini
> [INFO] [B200] Loading FPGA image:
> /usr/local/share/uhd/images/usrp_b200mini_fpga.bin...
> [INFO] [B200] Operating over USB 3.
> [INFO] [B200] Initialize CODEC control...
> [INFO] [B200] Initialize Radio control...
> [INFO] [B200] Performing register loopback test...
> [INFO] [B200] Register loopback test passed
> [INFO] [B200] Setting master clock rate selection to 'automatic'.
> [INFO] [B200] Asking for clock rate 16.000000 MHz...
> [INFO] [B200] Actually got clock rate 16.000000 MHz.
> Using Device: Single USRP:
> Device: B-Series Device
> Mboard 0: B200mini
> RX Channel: 0
> RX DSP: 0
> RX Dboard: A
> RX Subdev: FE-RX1
> TX Channel: 0
> TX DSP: 0
> TX Dboard: A
> TX Subdev: FE-TX1
>
> [00:00:11.448560] Setting device timestamp to 0...
> [INFO] [B200] Asking for clock rate 40.000000 MHz...
> [INFO] [B200] Actually got clock rate 40.000000 MHz.
> [WARNING] [MULTI_USRP] The total sum of rates (40.000000 MSps on 1
> channels) exceeds the maximum capacity of the connection (overflows
> (O) MSps).
> This can cause 22.7428.
> [00:00:11.766752] Testing receive rate 40.000000 Msps on 1 channels
> [WARNING] [MULTI_USRP] The total sum of rates (40.000000 MSps on 1
> channels) exceeds the maximum capacity of the connection (underruns
> (U) MSps).
> This can cause 22.7428.
> [00:00:11.790580] Testing transmit rate 40.000000 Msps on 1 channels
> [00:00:11.891995] Tx timeouts: 1
> UUUUUUUUUUUUUUUO[00:00:12.078147] Timestamp after overrun recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUUUUO[00:00:12.092404] Timestamp after overrun
> recovery ahead of error timestamp! Unable to calculate number of
> dropped samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUO[00:00:12.108355] Timestamp after overrun recovery
ahead
> of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUUUUUUUUU[OU00:00:12.123737] Timestamp after
overrun
> recovery ahead of error timestamp! Unable to calculate number of
> dropped samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUUUUOU[00:00:12.132437] Timestamp after overrun
> recovery ahead of error timestamp! Unable to calculate number of
> dropped samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUUUUUUOU[00:00:12.142422] Timestamp after overrun
> recovery ahead of error timestamp! Unable to calculate number of
> dropped samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUO[00:00:12.155257] Timestamp after overrun recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUO[00:00:12.168528] Timestamp after overrun recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUU[O00:00:12.178400] Timestamp after overrun recovery
ahead
> of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUU[00:00:12.193251] Timestamp after overrun recovery
ahead of
> error timestamp! Unable to calculate number of dropped
samples.(Delta:
> -3170 ticks)
> OUUUUUUUUUUUUUUUUUO[00:00:12.204356] Timestamp after overrun
recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUO[00:00:12.224770] Timestamp after overrun recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUO[00:00:12.235145] Timestamp after overrun
recovery
> ahead of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUU[O00:00:12.247517] Timestamp after overrun recovery
ahead
> of error timestamp! Unable to calculate number of dropped
> samples.(Delta: -3170 ticks)
> UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU[00:00:12.475571] Receiver error:
> ERROR_CODE_TIMEOUT, continuing...
> [00:00:12.575910] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> [00:00:12.676171] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> [00:00:12.776414] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> [00:00:12.876663] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> [00:00:12.976968] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> [00:00:13.077257] Receiver error: ERROR_CODE_TIMEOUT, continuing...
> terminate called after throwing an instance of 'uhd::io_error'
> what(): EnvironmentError: IOError: usb tx2 transfer status:
> LIBUSB_TRANSFER_NO_DEVICE[
> 00:00:13.083166] Caught an IO exception.
> EnvironmentError: IOError: usb rx6 transfer status:
> LIBUSB_TRANSFER_NO_DEVICE
>
Well, the main reason is that your typical laptop compute
environment,
based on x86 processor technology, is going to be more powerful
than the ARM on a Raspberry Pi4.
Now, you *may* be able to improve things slightly by adjusting the
USB
transport parameters, as described here:
https://files.ettus.com/manual/page_transport.html#transport_usb
But once you actually start "doing stuff" on the Raspberry Pi, you'll
find that it can't process as many samples per second as on an x86
system--whether a laptop or desktop machine. There's a reason
that a
Raspberry Pi is $50, and a typical low-end laptop is 10 times that.
_______________________________________________
USRP-users mailing list
USRP-users@lists.ettus.com <mailto: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
