expr does not appear to have a "match" operator in OSX. This has been
replaced with the ":" syntax for compatibility.
Additionally, a "--" argument has been added to avoid issues with
"$line" evaluating to an argument.
This fixes#278. Thank you to Mathieu Rene (@mrene) for reporting this
issue and providing these fixes.
A majority of functionality presented by bladeRF-flash is now duplicated
in the bladeRF-cli utility.
Now that the bladeRF USB interfaces have stabilized more and the
instructions for bootloader-based recovery are available, this program
is generally no longer needed. As such, it is being removed to avoid
future neglect and bitrot.
This program performs some simple operations with the various libbladeRF
"control" functions. It is mainly intended to only check for the
presence of grossly erroneous code in these functions, that would cause
crashes or bizarre results.
This error from the FX3 is likely harmless, and I'll investigate further
soon. It only appears to occur when the firmware loopback mode times
out, and does not hinder further use of the device.
Given the lack of apparent severity, this band-aid is intended to keep
libbladeRF from overreacting until some time can be spend fixing the
underlying issue.
Added the ability to read back the state of FW loopback from devices
running FW v1.7.1. While it is possible to enable FW loopback mode in
v1.7.0 firmware, libbladeRF will now enforce the v1.7.1 requirement so
it can correctly read back and report the FW mode.
bladerf_set_loopback() has also been updated to ensure that the
FW loopback is disabled when switching to an LMS loopback mode.
Stale RX and TX samples have been reported (issue #255). This occurs
because the CyU3PResetEp() call does not reset DMA channels and their
underlying DMA buffers, nor does it clear the DMA buffers associated
with an endpoint.
Use of the existing ClearDMAChannel() as is would have required some
ridiculous shoehorning, so some cleanup was done:
- Endpoint halt & ACK operations have been removed from
ClearDMAChannel(). These operations are outside of the scope
of clearing DMA buffers, and have been moved to their respective
locations in handling set/clear halt endpoint requests.
- reset_endpoint() has been added to the NuandApplication function
table. Internally, this operation is now used in performing
the halt_endpoint() operation.
- Misc return value checks added. Granted, these don't help much
as there's no accessible debug log.
Corrected a samplerate equation calculation which helps determine how
many transfers should be outstanding over any timeout period at a
specific samplerate. The samplerate should always be greater than the
calculated number, not less than.
Removed double check for the number of arguments in the cli poke
command. Also changed the error message to properly reflect an
invalid value instead of a number of addresses.
Since bladerf_set_frequency() does not return an actual value set
and may clamp the requested frequency, cause the CLI to query the
currently set frequency and print that out to avoid confusion.
While clamping the minimum values for the frequency selection, the
lowest value was being set but not checked against.
Also modify the range table to use the enumerations for lowest and
highest frequency for consistency.
This commit introduces 1dB and 3dB rolloff point based filterbank selection.
bladerf_set_frequency() will set the appropriate filter based on frequency.