| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Replay block would previously fail output on low mtu links because it
was creating packets based on the default contruction mtu of 8192.
This change will update the packet size when mtu gets updated.
This also removes a warning for large packet sizes because the replay
block uses a maximum packet size property. The block should be
free to coerce down based on mtu limitations.
Signed-off-by: Steven Koo <steven.koo@ni.com>
|
|
|
|
|
| |
This gives the source generator some valid initial values in the Null
Source/Sink RFNoC block.
|
|
|
|
|
|
| |
Increases the supported memory sizes in software to 2^32 and beyond.
Signed-off-by: mattprost <matt.prost@ni.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
The previously added APIs for getting/setting power reference levels was
missing an option to read back the currently available power levels
(minimum and maximum power levels).
This adds getters for TX and RX power ranges to multi_usrp and
radio_control. The power API is thus now more similar to the gain API,
which always had getters for gain ranges.
|
|
|
|
| |
The .1 second timeout fails on macOS. Expand this timeout to 1 second.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The max FFT size was 4096, but we don't currently have any devices that
can do that without modification. This is because, currently, the FFT
size must be the same as the packet size, and the largest packet
size supported by most devices is about 8000 bytes, or 2000
sc16 samples. Therefore, the largest FFT size supported without
modifying other code is 1024 samples.
This change frees up about 21% of the LUTs and 36% of the BRAM used by
axi_fft and makes the software block controller and the IP agree on the
maximum FFT size.
|
|
|
|
|
|
| |
This commit resolves an issue where the disconnect call would fail
if the streamer was created, but not connected. If disconnect gets
called on a streamer that isn't on the graph, ignore it.
|
|
|
|
|
|
|
|
|
| |
This commit resolves a segfault that occurs during teardown. Since
teardown isn't detereministic in python, there were cases where the
graph would destruct before the streamers or links got chance to
cleanup. This would result in a segfault. This change gives the lambda
a shared pointer to the object the callback method is calling on, so
that it won't destruct before being called.
|
|
|
|
|
|
|
| |
Add Python bindings for recently added rfnoc_graph::disconnect()
methods.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
|
| |
Reverting inadvertent change in num_chans argument during
tx/rx_streamer_impl construction.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
| |
M_PI may not exist if _USE_MATH_DEFINES isn't defined before the
first include of math.h or cmath on Windows. This changes avoids the
issue all together by defining our own PI.
|
| |
|
| |
|
|
|
|
|
|
| |
Change message from warning to debug when spp is greater than MTU.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
|
|
|
|
|
| |
- Added rfnoc_graph method to disconnect a connection.
- Added rfnoc_graph method to disconnect a streamer.
- Added rfnoc_graph method to disconnect a port on a streamer.
- Added disconnect callback to rfnoc_rx_streamer and rfnoc_tx_streamer.
- Registered disconnect callback functions to streamers returned by
get_rx_streamer and get_tx_streamer methods.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
|
|
|
|
| |
- Added method to disconnect an edge
- Added method to remove a node
- Fixed algorithm to check edges during connect. Previous code was
checking some edges twice and allowing duplicate edges to be created
for existing edges.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Transports were not disconnecting their links from the I/O service upon
destruction, leaving behind inaccessible send and recv links used by
nothing. This led to I/O errors after creating several transports.
Added callbacks to transports to automatically disconnect their links
from the I/O service when the transport is destroyed. Updated all
callers to supply a disconnect callback.
Signed-off-by: michael-west <michael.west@ettus.com>
|
|
|
|
| |
Signed-off-by: mattprost <matt.prost@ni.com>
|
|
|
|
| |
Signed-off-by: mattprost <matt.prost@ni.com>
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds an enhancement to node_t::set_properties() in which
the instance argument provided to the function (which normally applies
to all properties in the key/value list) can be overridden on a
per-property basis using a special syntax.
If the key consists of the property name followed by a colon (':') and
then a number, the number following the colon is used to determine which
instance of the property this set pertains to, and the value passed via
the instance parameter is ignored for that property. For example, in the
following call:
node->set_properties("dog=10,cat:2=5,bird:0=0.5", 1)
instance 1 of node's 'dog' property is set to 10, the 1 coming from the
instance parameter, instance 2 of the node's 'cat' property is set to 5
due to the override syntax provided in the string, and instance 0 of the
node's 'bird' property is set to 0.5 due to its override.
If the name/instance pair is malformed, e.g. 'value:=10' or
'value:foobar=10', a runtime error is thrown.
|
|
|
|
| |
Signed-off-by: mattprost <matt.prost@ni.com>
|
|
|
|
| |
Signed-off-by: mattprost <matt.prost@ni.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit alters the ctrl_payload#serialize function to assert the
serialized length of the packet fits inside the buffer before writing.
Originally, the function writes the data and then asserts that the final
pointer position isn't further than the end of the buffer. While this
works, if it were to fail, that means we have already caused undefined
behavior by writing to some random memory location, which isn't good
practice.
This commit is possible now that the get_length method exists.
Signed-off-by: Samuel O'Brien <sam.obrien@ni.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit introduces a new public api in uhd::utils which allows serializing
and deserializing chdr packets.
As far as testing, this commit adds the chdr_parse_test test. It uses a
wireshark trace located in rfnoc_packets_*.cpp as well as hand coded
packets from hardcoded_packets.cpp to test the serialization and
deserialization process
Signed-off-by: Samuel O'Brien <sam.obrien@ni.com>
|
|
|
|
|
|
|
|
|
|
| |
This commit exposes uhdlib/rfnoc/chdr_types.hpp in the public includes.
Additionally, it takes some types from uhdlib/rfnoc/rfnoc_common.hpp and
exposes them publicly in uhd/rfnoc/rfnoc_types.hpp.
Finally, one constant is moved from uhdlib/rfnoc/rfnoc_common.hpp to
uhd/rfnoc/constants.hpp
Signed-off-by: robot-rover <sam.obrien@ni.com>
|
|
|
|
|
|
|
|
|
| |
It would be confusing to have two classes named chdr_packet. As it makes
more sense to name the new public chdr parser class chdr_packet, the
internal uhd::rfnoc::chdr::chdr_packet class is being renamed to
chdr_packet_writer to better represent its functionality.
Signed-off-by: Samuel O'Brien <sam.obrien@ni.com>
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Support management payloads on busses over 64 bits
Automatically set CHDR width for mpmd_link_if_ctrl_udp
|
| |
|
|
|
|
|
|
|
| |
This commit augments the existing FFT RFNoC block controller with
C++ functions through which the block can be configured, as well as
adding range checking to the various properties that sit atop the FFT
RFNoC block registers.
|
|
|
|
|
| |
This avoids a dynamic linker error by copying the ALL_CHANS value into
the the Python bindings before using it.
|
|
|
|
|
|
|
|
| |
The various implementations for the reference power APIs are always the
same, assuming the existence of a pwr_cal_mgr object. We therefore store
references to power cal managers in radio_control_impl, which radios can
choose to populate. The APIs then don't have to be reimplemented in the
various radio classes, unless they want to for whatever reason.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds a __repr__ function to the noc_block_base bindings,
which helpfully displays the block's unique ID, e.g.:
>>> block = g.get_block('0/VectorIIR#0')
>>> block
<NocBlock for block ID '0/VectorIIR#0'>
Also added are get_property_ids and set_properties functions, so Python
clients can set block properties by string if desired, e.g.:
>>> block.get_property_ids()
['alpha', 'beta', 'delay', 'max_delay']
>>> block.set_properties('alpha=0.45,beta=0.77,delay=41')
>>> viir = uhd.rfnoc.VectorIirBlockControl(block)
>>> viir.get_alpha(0)
0.45
>>> viir.get_beta(0)
0.77
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds a utility class for use with the Python RFNoC block
controller PyBind bindings which facilitates constructing instances of a
specific block controller type from its noc_block_base base class. This
allows Python code to create and configure specific block controller
instances by calling get_block on a uhd.rfnoc.RfnocGraph object with the
block ID of the block in question and then passing the result into the
constructor method of the block controller, e.g.:
graph = uhd.rfnoc.RfnocGraph("addr=...")
block = graph.get_block(uhd.rfnoc.BlockID("0/DDC#0"))
ddc = uhd.rfnoc.DdcBlockControl(block)
ddc.set_input_rate(10e6, 0)
|