Introduce three subcategories for Silabs boards: dev_kits, radio_boards
and starter_kits. The dev_kits category contains small form-factor
standalone boards branded as "Dev Kit" or "Thunderboard". The
radio_boards category contains radio boards to be used with a wireless
mainboard, and the starter_kits category contains larger form-factor
standalone boards branded as "Starter Kit".
Structure the Sphinx documentation the same way as the directory tree.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
This commit adds support for configuring SPI frequency per transaction.
The "clock-frequency" devicetree property is used as the default
frequency unless the peripheral being communicated with has a lower
"max-spi-frequency" (passed to the driver as spi_config::frequency).
An error is returned if the requested frequency is higher than the
hardware can support. This limit is half the peripheral clock (PCLK on
Series 2, HFPERCLK on Series 0/1) for SPI in controller mode.
Previously, the driver hard-coded 1 MHz operation. For backwards
compatibility, default to 1 MHz if the "clock-frequency" property is not
set in devicetree.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
This commit adds support for devices such as EFR32xG24 in the SPI
driver, as these devices only have a single USART peripheral.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
View buffers are now also a limited resource. Acquire them before
attempting to pull data. `CONFIG_BT_CONN_FRAG_COUNT` should be tuned on
a per-application basis to avoid this.
A possible optimization, that was present before, is to not create a
frag when the original buffer fits the controller's HCI size.
I prefer deferring this optimization to a future patchset.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
We can get rid of the view pool for SDU segments :)
We have to make the code slightly more complex :'(
The basic idea is always giving the original SDU buffer to `conn.c` for it
to pull ACL fragments from.
In order to do this, we need to add the PDU headers just-in-time.
`bt_l2cap_send_pdu()` does not add them before putting the PDU on the queue
anymore. They are added by `l2cap_data_pull()` right before the data leaves
`l2cap.c` for `conn.c`.
We also have to inform `conn.c` "out of band" of the real L2CAP PDU size so
it doesn't fragment across segment boundaries. This oob is the new `length`
parameter to the `.pull()` method.
This is the added complexity mentioned above.
Since SDU segmentation concerns only LE-L2CAP, ISO and Classic L2CAP don't
need this extra logic.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
We don't need the TX thread anymore.
Generalizing the pull-based architecture (ie. `tx_processor`) to HCI
commands makes it possible to run the whole TX path from the the system
workqueue, or any workqueue really.
There is an edge-case, where we call `bt_hci_cmd_send_sync()` from the
syswq, stalling the system. The proposed mitigation is to attempt to drain
the command queue from within `bt_hci_cmd_send_sync()`.
My spidey sense tingles however, and it would be better to just remove the
capability of calling this fn from the syswq. But doing this requires
refactoring a bunch of synchronous procedures in the stack (e.g. stack
init, connection establishment, address setting etc), dragging in more
work. I will do it, but in a later patch.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
This API replaces `bt_l2cap_send()` and `bt_l2cap_send_cb()`.
The difference is that it takes the `struct bt_l2cap_le_chan` object
directly instead of a connection + CID.
We need the channel object in order to put the PDU on the TX queue. It
is inefficient to do a search for every PDU when the caller knows the
channel object's address and can just pass it down.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
The current TX pattern in the host is to try to push a buffer through all
the layers up until it is ingested by the controller.
Since sending can fail at any layer, we need error-handling and separate
retry logic on pretty much all layers. That logic obscures the "happy path"
for people trying ot understand the code.
This commit inverts the control, in a way that doesn't require changing the
host or HCI driver API (yet):
Layers don't send buffers synchronously, they instead put their buffer in a
private queue of their own and raise a TX flag on the lower layer. Think of
it as a `READY` interrupt line that has to be serviced by the lower layer.
Sending is now non-blocking, rate depends on the size of buffer pools.
There is a single TX processing function. This can be thought as the
Interrupt Service Routine that will handle the `READY` interrupt from the
layers above.
That `tx_processor()` will then attempt to allocate enough resources in
order to send the buffer through to the controller. This allocation logic
does not block.
After acquiring all the resources, the TX processor will attempt to pull
data from the upper layer. The upper layer has to figure out which buffer
to pass to the controller. This is a good spot to put scheduling or QoS
logic in the upper layer.
Notes:
- user-facing API for tuning QoS will be implemented in a future patch
- this scheme could (and probably will) be extended to upper layers (e.g.
ATT, L2CAP CoC segmentation).
- this patch removes the `pending_no_cb()` memory optimization for
clarity/correctness. It might get re-implemented after a stabilization
period. Hopefully with more documentation.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Co-authored-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Instead of allocating segments/fragments and copying data into them, we
allocate segments as "views" (or slices) into the original buffer.
The view also gives access to the headroom of the original buffer, allowing
lower layers to push their headers.
We choose not to allow multiple views into the same buffer as the headroom
of a view would overlap with the data of the previous view.
We mark a buffer as locked (or "in-view") by temporarily setting its
headroom to zero. This effectively stops create_view because the requested
headroom is not available.
Each layer that does some kind of fragmentation and wants to use views for
that needs to maintain a buffer pool (bufsize 0, count = max views) and a
metadata array (size = max views) for the view mechanism to work.
Maximum number of views: number of parallel buffers from the upper layer,
e.g. number of L2CAP channels for L2CAP segmentation or number of ACL
connections for HCI fragmentation.
Reason for the change:
1. prevent deadlocks or (ATT/SMP) requests timing out
2. save time (zero-copy)
3. save memory (gets rid of frag pools)
L2CAP CoC: would either allocate from the `alloc_seg` application callback,
or worse _steal_ from the same pool, or allocate from the global ACL pool.
Conn/HCI: would either allocate from `frag_pool` or the global ACL pool.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Co-authored-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
`bt_conn_send_cb` used to allocate a TX context (K_FOREVER).
Instead, we now put the context in the userdata of the buffer.
This means that now this fn will never block and always succeed since the
tx_queue is a FIFO (infinite size). It just puts the buf on the queue.
The metadata is stored safely in there until we have acquired all the
necessary resources to send it to the controller without failing: TX
context and controller buffer.
I.e. when `bt_conn_process_tx` is called, that's when a TX context is
try-allocated and the contents of `buf->userdata` is moved into it.
The buffer is now ready to be sent to the lower layer.
`bt_conn_process_tx` will return -EWOULDBLOCK if it's not able to acquire a
TX context, this PR modifies `bt_conn_prepare_events` to respond to this by
also waiting on the TX context pool.
Unfortunately, this increases the required userdata size for any buffers
handed to `bt_conn_send_cb`. This will be fixed in a later commit.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Fix compilation error when reset-gpios is enabled.
Undefined reference to dev_cfg variable.
Reset gpio duration needs to be defined, but is not in binding file.
Signed-off-by: Joakim Andersson <joerchan@gmail.com>
In case we want to immediately send empty Ack to server,
we should bypass all send queues.
This is required when we try to send Ack from callbacks
that happen from socket-loop context. On those cases
the Ack would have not been send because the callback
might be blocking the socket-loop while processing
a request (like write callbacks).
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
Instead of blindly using the module names provided by the user via
command-line arguments, check if those actually exist in the current
manifest, and error out if any of them does not.
Fixes#73901.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Use some new kernel features to make the experience of finding and
dealing with GPIO devices much more ergonomic by allowing the use of
devicetree node labels to identify GPIO devices by default.
Users who wish to avoid the associated footprint penalty can set
CONFIG_DEVICE_DT_METADATA=n by hand, but I think the convenience is
worth the price as a default. If we're running a shell, then we've
already paid a heavy footprint penalty.
Example output for qemu_cortex_m3:
uart:~$ gpio devices
Device Other names
gpio@40004000 gpio0
gpio@40005000 gpio1
gpio@40006000 gpio2
gpio@40007000 gpio3
gpio@40024000 gpio4
gpio@40025000 gpio5
gpio@40026000 gpio6
Signed-off-by: Martí Bolívar <mbolivar@amperecomputing.com>
This option allows you to look up a struct device from any of the
node labels that were attached to the devicetree node used to create
the device, etc.
This is helpful because node labels are a much more human-friendly set
of unique identifiers than the node names we are currently relying on
for use with device_get_binding(). Adding this infrastructure in the
device core allows anyone to make use of it without having to
replicate node label storage and search functions in various places in
the tree. The main use case, however, is for looking up devices by
node label in the shell.
Since there is a footprint penalty associated with storing the node
label metadata, leave this option disabled by default.
Signed-off-by: Martí Bolívar <mbolivar@amperecomputing.com>
Add the following new macros:
- DT_FOREACH_NODELABEL
- DT_FOREACH_NODELABEL_VARGS
- DT_INST_FOREACH_NODELABEL
- DT_INST_FOREACH_NODELABEL_VARGS
These are for-each helpers for iterating over the node labels of a
devicetree node. Since node labels are unique in the entire
devicetree, their token representations can be useful as unique IDs in
code as well.
As a first user of these, add:
- DT_NODELABEL_STRING_ARRAY
- DT_INST_NODELABEL_STRING_ARRAY
The motivating use case for these macros is to allow looking up a
struct device by devicetree node label in Zephyr shell utilities.
The work on the shells themselves is deferred to other patches.
To make working with the string array helpers easier, add:
- DT_NUM_NODELABELS
- DT_INST_NUM_NODELABELS
Signed-off-by: Martí Bolívar <mbolivar@amperecomputing.com>
The `LOG_BACKEND_FORMAT_TIMESTAMP` Kconfig currently depends on
a list of hardcoded backends.
Let's modify it to depend on an intermediary Kconfig
`LOG_BACKEND_SUPPORTS_FORMAT_TIMESTAMP` instead, which can be
selected by a OOT log backend.
Updated all exisitng supported backends to select this new
Kconfig.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Release buffer allocated in dwc2_ctrl_feed_dout() on endpoint deactivate
to prevent the buffer from leaking on USB stack disable.
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Set the bootloader choice for the
sample.modules.canopennode.program_download test case in the sample.yaml
file instead of sysbuild.conf. The reason for this change is to avoid
building MCUboot by default when using the --sysbuild option, as it may
not be compatible with all platforms. This is currently causing issues
on platforms like nRF54H20 where sysbuild is always used.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
This change marks each instance of the `gnss_xxx_config` and
`gnss_driver_api` as `const`.
By using `const`, we ensure immutability, leading to usage of only
`.rodata` and a reduction in the `.data` area.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
This sample reports the I2C configuration of the selected I2C node
from the board DTS : bitrate property of the I2C node
In case of I2C V2, it also gives the TIMINGS which has to be written
in the DTS <timings> property of the I2C V2 node.
Uses the i2c_get_config function with CONFIG_I2C_STM32_V2_TIMING.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Change the get_config API for the stm32 I2C V2 driver.
It will also return the value of the content of TIMING register
for the I2C V2 bus.
This is hold by a i2c_config_timing structure of the device data
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Define Macro to fix the compliance check warning :
"DEEP_INDENTATION: Too many leading tabs - consider code refactoring
in the i2c_compute_scll_sclh() function
in the i2c_compute_presc_scldel_sdadel() function
Signed-off-by: Francois Ramu <francois.ramu@st.com>
This PR is implementing a new formula to calculate the I2C timing
value from the I2C clock source and the bit rate.
This is done under flag CONFIG_I2C_STM32_V2_TIMING.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
This adds a basic example demonstrating how to enable and use the CMSIS-DSP
module in Zephyr.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
`irq0-channels` defines even-numbered channels as the default value.
Since 6 was dropped from this definition, it is added.
Also, since the maximum number of channels is 12,
remove the ones that are exceeded.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
Restructures the latency_measure benchmark configurations. Most of them
do not need to be executed by default by CI. However, we still want to
make it easy to run the benchmark with various configurations so that
they can be easily compared against the default.
The updated documentation shows how this can be done.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
Add an option to reset gcov counters before running tests.
This ensures that only code lines triggered by test itself are counted in
the coverage report and all the board initialization code and pre-test
bootstrap is not counted. This is useful when, for example, you are
testing code that is also executed during bootup
Test Plan:
west build -p always -b qemu_x86_64 tests/ztest/base/ -- \
-DCONFIG_COVERAGE=y -DCONFIG_COVERAGE_GCOV=y -DCONFIG_COVERAGE_DUMP=y \
-DCONFIG_ZTEST_COVERAGE_RESET_BEFORE_TESTS=y
ninja -Cbuild run | tee log.log
Signed-off-by: Roman Studenikin <srv@meta.com>
Ability to reset gcov counters allows better isolation of tests
coverage. Specifically it allows to:
1. Not include counters for any code that was executed prior running the
test, which includes all the code executed during the board boot.
2. Run multiple tests without firmware reload by resetting counters
between each.
Signed-off-by: Roman Studenikin <srv@meta.com>
If the length of the string literal reserved for the serial number
descriptor is odd, the string is not used because its length is always
even and therefore one character longer. Fix this by using the shortest
length for the copy.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
