Because they are needed in SoC power.c, clock control init functions
for all other STM32 series has been exported as global.
This commit exports stm32_clock_control_init function as global for
STM32H7 series as well, making custom power management
implementations for STM32H7 series possible.
Signed-off-by: Tomislav Milkovic <tomislav.milkovic95@gmail.com>
Add a variation of the basic Kconfig check that runs with no modules,
catches symbols that are used in the main repository but are defined
only in modules, which are potentially problematic if a downstream
project is not using the specific module.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
These are referenced by the "x module not available" message in
modules/Kconfig, but were only defined in the module, were generating an
undefined symbol warning when running CI compliance with no modules.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
NO_CACHE cannot be selected for certain cores.
Use ARCH_HAS_NOCACHE_MEMORY_SUPPORT as the condtion
to select NO_CACHE config
Signed-off-by: Mahesh Mahadevan <mahesh.mahadevan@nxp.com>
Split flash into partitions (4x256KB). There is no way to make partitions
smaller, because of block size. This feature enables to switch between
2 applications.
Signed-off-by: Andrej Valek <andy@skyrain.eu>
PR instantiates cdb and core keys to prevent scenario
when key is changed over cdb API and gets new ID but
core still operates with old ID(potentially invalid).
Known issue the keys will be desynchronized
during\after key refresh procedure.
Signed-off-by: Aleksandr Khromykh <aleksandr.khromykh@nordicsemi.no>
This commits eliminates the old way of
pacs notifications, replacing k_work with
simple bt_gatt_notify.
Signed-off-by: Szymon Czapracki <szymon.czapracki@codecoup.pl>
Fixes a minor issue with the documentation where there are
conflicting messages about the required validity of the parameters.
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
The blink command should blink until the user presses a key, however on
some systems a pending keypress is present when entering the loop
causing it to immediate exit before getting a chance to blink.
Add a dummy read to clear the buffer before starting the blink loop.
Signed-off-by: Eric Holmberg <eric.holmberg@northriversystems.co.nz>
Introduce a new area of maintenance for the IEEE 802.15.4 native stack
as distinguished from both, IEEE 802.15.4 drivers and OpenThread and
more specialized than generic network subsystem maintainership.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
When doing the bt_ots_dir_list_reset_anchor we now explicitly ignore
the return value of bt_gatt_ots_obj_manager_first_obj_get
to fix a coverity issue.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Some node items in Kconfig can be kconfiglib.MENU or kconfiglib.COMMENT.
Those are integers and thus do not contain a node.item.name field.
Handle those separately to avoid hitting the followig exception:
Traceback (most recent call last):
File "/home/runner/work/zephyr/zephyr/./scripts/ci/check_compliance.py",\
line 1307, in main
n_fails = _main(args)
File "/home/runner/work/zephyr/zephyr/./scripts/ci/check_compliance.py",\
line 1242, in _main
test.run()
File "/home/runner/work/zephyr/zephyr/./scripts/ci/check_compliance.py",\
line 277, in run
self.check_no_redefined_in_defconfig(kconf)
File "/home/runner/work/zephyr/zephyr/./scripts/ci/check_compliance.py",\
line 445, in check_no_redefined_in_defconfig
Kconfig node '{node.item.name}' found with prompt or help in\
{node.filename}.
AttributeError: 'int' object has no attribute 'name'
Seen in #58454.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Set the PWM clock prescaler values to 64 for controlling standard servo
50Hz carrier frequency.
Signed-off-by: Benjamin Perseghetti <bperseghetti@rudislabs.com>
This patch sets ENET2 ref clock to be generated by External OSC
ENET2 ref clock direction as output
ENET2 ref clk frequency to 50MHz
Signed-off-by: Sumit Batra <sumit.batra@nxp.com>
used for gps and telemtry.
This patch configures the correct pinmux settings for
lpuart2, lpuart3, lpuart5 ports used for gps and telemetry modules.
It also adds software pull up for the UART console port lpuart7.
Signed-off-by: Sumit Batra <sumit.batra@nxp.com>
Co-authored-by: Benjamin Perseghetti <bperseghetti@rudislabs.com>
Co-authored-by: Iain Galloway <iaintravelgalloway@gmail.com>
Cast ticker_stop() calls to void where return value is not checked.
This is to satisfy coverity and indicate that return value is not
important.
Signed-off-by: Mateusz Kusiak <mateusz.kusiak@intel.com>
This change replaces one letter variable names with considerable
variable spans by much more readable alternatives that ease code
maintenance and help human reviewers to catch semantic errors.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
The f_ctx variable renamed in this change has a considerable variable
span. This makes the code hard to read as the variable name is neither
defined in the IEEE 802.15.4 standard nor can it be deduced from the
variable name.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
The 'ar' abbreviation has a well defined meaning in IEEE 802.15.4 which
is NOT the meaning of the local "ar" variable being renamed in this
change (see the comparison to the actual MHR ar field in this change
set).
To avoid confusion, a non-abbreviated variable name is introduced
instead.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
The 'af' abbreviation is nowhere standardized in IEEE 802.15.4 and makes
the source code unnecessarily hard to read. It is replaced by a readable
long name.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
Several distinct naming conventions existed within the IEEE 802.15.4
stack wrt header length. This change converges to a single naming
convention, the one that is less ambiguous and already most used.
The change also makes the distinction between L2 (link layer/LL) header
length and 6LoWPAN fragmentation header length to avoid confusion.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
This change is in preparation for the newer frame version's more
complex compression algorithm that may compress both, the source
and the destination PAN independently.
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
This change renames buf to pkt_buf in one very specific instance to
distinguish the packet buffer from the frame buffer which is kept in the
same local scope. In all other instances the meaning of "buf" should be
obvious from context, not so here, though
Signed-off-by: Florian Grandel <fgrandel@code-for-humans.de>
The display controller supports software reset and the driver already
implements it. Therefore it's not necessary to require a reset gpio in
device tree.
Signed-off-by: Caspar Friedrich <c.s.w.friedrich@gmail.com>
The device passed to the DMA callback is the DMA device. Instead use
the ADC device available in the private data.
Signed-off-by: Brett Witherspoon <brett@witherspoon.engineering>
Add fixed-clock clock control driver. This is a first step towards
making fixed-clocks a first-class citizen in the clock control
framework.
Since the change is hidden behind a Kconfig enable this is opt-in
for now.
Signed-off-by: Moritz Fischer <moritzf@google.com>
Remove `clocks` property for fixed-clock binding.
A fixed-clock should not have an input clock, since by
definition it's an always on fixed-rate clock.
Signed-off-by: Moritz Fischer <moritzf@google.com>
This allows us to identify any patterns that we can remove or that
we didn't realise are no longer in use. This might happen if issues
within doxygen or docleaf are resolved.
This allows us to remove the pattern:
.*Duplicate C declaration.*\n.*'\.\. c:.*:: uint16_t id'.*
which does not match anything in the current set up.
We also split the filter patterns in known-warnings out into different
sections depending on their cause.
Also extend the pattern parser to ignore empty lines so that we can
have some formatting in the known-warnings file.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
We should probably clear out the old warnings and see what still
applies for docleaf now that we have that instead of breathe.
However for this moment, these are new duplication declarations that
we're seeing errors for and so we'd like to silence them.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
So that we see all the potential errors in CI and not just the first
and can assess how much work there is to do.
This was introduced for debuggin but gmarull has suggested keeping it
in.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
There are times when the log message is an exception object and that
causes the regex to fail with a TypeError. Here we work around that
by converting the input to a string.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
We make sure that we have the 'root' and 'xml' defined for the project
which is necessary at the moment for docleaf though the 'root' is only
really used for the 'linkcode' integration.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
Instead of breathe.
I'm not sure of the best version comparison to use. There is no
intention to break the API for docleaf but it is also still young
in some ways and there might be adjustments to the configuration values
that would need to be reflected in the version number as it is still
pre-1.0 that would only be a minor version change.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
This switches the Sphinx conf.py file over to using the
'docleaf.doxygen' module and the associated docleaf configuration
entries, replacing the breathe module and config.
Signed-off-by: Michael Jones <m.pricejones@gmail.com>
