| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
- Fixes a case where the e320 would be unable to lock to an external 10 MHz
reference
- Previously, calling set_time_source would set the reference clock source to
internal as a side effect
|
|
|
|
|
|
|
|
|
|
|
| |
Prior commit disabled flow control for all lossless links,
including links between blocks on the same crossbar. That
caused packets to jam up the crossbar leading to control
packets failing to ACK under certain conditions. This
reverts it so flow control is enabled and the crossbar is
kept clear.
Signed-off-by: Michael West <michael.west@ettus.com>
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Adding advisory in the X300 manual page regarding issuing multiple
timed commands causing device lockups.
|
|
|
|
|
|
|
|
|
| |
To check if a device is fully reachable through the
chdr address i.e. SFP port, an RPC request is sent
to "second_addr" and "addr" of that device. This commit
fixes the case where if "second_addr" is not connected to
the host, the function returns earlier and does not
check for "addr".
|
|
|
|
|
|
|
| |
- Fixes an issue where a multi-device device3 object would cause connections to
be generated across devices in legacy_compat
- The connect_blocks function previously assumed that find_blocks would always
return blocks in the same device number order
|
|
|
|
|
|
|
| |
Updated b2xx_fx3_utils to add a --query_bootloader option to check if
the bootloader is loaded or not.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
|
|
|
| |
This pulls out a lot of code from x300_impl and puts it into its own
compilation units:
- EEPROM code goes to x300_mb_eeprom.*
- Claim code goes to x300_claim.*
- PCIe code goes to uhd::usrp::x300::pcie_manager
- Ethernet code goes to uhd::usrp::x300::eth_manager
|
|
|
|
|
| |
This puts all functions that identify motherboards, distinguishes
X310/X300/NI2974, checks MB revs etc. into its own compilation unit.
|
|
|
|
|
| |
GPSd was used by the E310 prior to its MPM-ification, GPSd is therefore
no longer required by UHD.
|
| |
|
|
|
|
|
| |
This is because FILE is not actually a type, and we don't care about an
existing file either. It's just a setting for the build.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
With the keywords "rx_gain_profile", "tx_gain_profile" and the new
values "default_bypass_always_off" and "default_bypass_always_on" the
customer can control whether the filter bypass shall be applied as
defined in the gain-table or whether to switch the bypass to ON or OFF
for all gains.
With the keywords "rx_band_map" and "tx_band_map" the user can define
custom frequency borders for the band mapping.
These new keywords can be used in the uhd.con, e.g.:
[type=n3xx]
rx_gain_profile=default_bypass_always_off
tx_gain_profile=default_bypass_always_on
rx_band_map=431e6;601e6;1051e6;1601e6;2101e6;2701e6
tx_band_map=723.18e6;1623.18e6;3323.18e6
|
|
|
|
| |
See: https://gitlab.com/gpsd/gpsd/commit/c711d531a6faab93df68ec625a076631569a729d
|
| |
|
| |
|
|
|
|
|
|
|
| |
When trying to create a converter with an illegal string, it will now
look like this:
[convert] Cannot find an item size for: `???'
|
| |
|
|
|
|
|
| |
- Remove some spurious \n
- Remove some usage of boost::format
|
|
|
|
|
|
|
|
|
|
| |
The latest changes to the get_*x_stream() functions to calculate the MTU for
the channel caused default frame size values to be ignored. This change fixes
that by changing the key from "send/recv_frame_size" to "mtu" and then changing
the implementations of make_transport() constrain the frame size values based
on the "mtu" value as well as any device and/or transport-specific limits.
Signed-off-by: Michael West <michael.west@ettus.com>
|
|
|
|
| |
Signed-off-by: Michael West <michael.west@ettus.com>
|
| |
|
|
|
|
|
|
|
| |
The SBX tracks the LO lock state via on-board LEDs. However, querying
the LO lock status spawns an update to the entire ATR registers. To
reduce the number of register reads/writes, the LO lock status LED is
now only updated if it changed.
|
|
|
|
|
| |
Removing unnecessary CMAKE_INSTALL_PREFIX to destination for usrp mpm
python dir
|
| |
|
|
|
|
|
|
| |
Blocks that change scaling, tick rate, or sampling rate can now notify
the graph to update streamers. Before, this was handled only by
mult_usrp, and only for DDC and DUC blocks.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Applying clang-format to files used in upcoming changes.
clang-format -i --style=file host/include/uhd/rfnoc/node_ctrl_base.hpp
clang-format -i --style=file host/lib/rfnoc/ddc_block_ctrl_impl.cpp \
host/lib/rfnoc/duc_block_ctrl_impl.cpp \
host/lib/rfnoc/legacy_compat.cpp \
host/lib/usrp/device3/device3_impl.cpp \
host/lib/usrp/device3/device3_impl.hpp \
host/lib/usrp/device3/device3_io_impl.cpp \
host/lib/usrp/x300/x300_impl.cpp
|
| |
|
| |
|
|
|
|
|
|
| |
In the PACKET_INLINE_MESSAGE case, we need to extract the error code
from the packet buffer. But the buffer was being released before that
happens, resulting in garbage values for metadata.error_code.
|
|
|
|
| |
access or data as bytes instead of str was not possible due to unconditional assert check for str
|
| |
|
| |
|
|
|
|
|
| |
- Split tuning log into 3 lines
- Remove duplicated MAX287X
|
|
|
|
|
|
|
| |
- This mode defaults to "disabled" (no change from current behavior)
- Enabling this mode fixes some instances where the MAX287x would fail to lock
when in integer-n mode
- Enabling this mode increases the maximum lock time to up to 100ms
|
|
|
|
| |
Using UHD_HEX() requires <iomanip>, which is now part of log.hpp.
|
|
|
|
| |
Should use value initialization for non-trivial classes.
|
| |
|
|
|
|
| |
Removes warnings due to -Wdeprecated-copy
|
| |
|
|
|
|
| |
- This change improves spur performance for the Rev C TwinRX daughterboard
|
|
|
|
| |
Signed-off-by: Matthew Crymble <matthew.crymble@ni.com>
|
|
|
|
|
|
| |
- Makes use of more modern find_package(Python2/3) if available
- Moves almost all Python-related code to UHDPython.cmake
- ENABLE_PYTHON3 is no longer necessary
|
|
|
|
| |
This will include updated E310 FPGA images.
|
| |
|
| |
|
|
|
|
|
| |
Widening and narrowing of bitwidths causes errors on this platform
without explicit casts.
|