Add devicetree binding for the NXP LPC CRC controller
(compatible "nxp,lpc-crc") with required reg property.
Signed-off-by: Holt Sun <holt.sun@nxp.com>
Minimal board definition for the Elan em32f967 SOC:
- Includes DTS, YAML, and Kconfig files
- Provides board.cmake with misc-flasher for CI/build
- Actual flashing is not included in this PR
Signed-off-by: Johnny Chuang <johnny.chuang@emc.com.tw>
Add initial support for the Elan em32f967 SoC based on ARM Cortex-M4.
Minimal SoC definition for upstream:
- soc.h, Kconfig, and CMakeLists.txt provided for completeness
- This PR does not implement drivers or UART functionality
Signed-off-by: Johnny Chuang <johnny.chuang@emc.com.tw>
Add minimal DT binding headers and matching YAML files for the
Elan em32f967 SoC. The new definitions are required to build the
sample board and follow the upstream style.
Signed-off-by: Johnny Chuang <johnny.chuang@emc.com.tw>
Add documentation for Greybus Zephyr implementation. The RFC [0] goes into
more detail regarding Greybus and it's current status.
[0]: https://github.com/zephyrproject-rtos/zephyr/issues/98259
Signed-off-by: Ayush Singh <ayush@beagleboard.org>
Minor improvements in this cmake:
Add a comment to clarify why we set one option, and move an if
into an else with a comment of what is doing.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
This variable is not set to anything.
This line has always been just bogus.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Fix unsafe handling of the clock control subsystem pointer.
The subsystem argument was cast before being validated, which could
lead to invalid memory access when NULL is passed.
Rework the on/off callbacks to validate the subsystem pointer before
casting it, drop unnecessary device checks, and mark the device
argument as unused.
Also simplify get_rate() validation by checking only the rate
pointer.
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
Remove a NULL check on the device pointer that occurs after the
pointer has already been dereferenced.
This check can never be true and is therefore misleading.
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
NO_POSIX_CHEATS was a macro used to avoid including the content of a
header (`posix_cheats.h`) which allowed building applications in the
POSIX architecture without the native simulator, avoiding collisions
between some embedded symbols and those from the host C library.
Support for this way of building, and this header and macro were
removed in e150ffb92c, but these users
were forgotten. This was harmless, but let's just clean it up now.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Add newlib to ignore_tags so twister will no longer run newlib
tests. These platforms have memory mappings that are not quite
what newlib is expecting so malloc may sometimes fail or even
returns invalid addresses. Fixing it is not worth the effort
because:
() Zephyr SDK provides picolibc now and not newlib, and malloc
works fine with picolibc.
() Downstream projects uses xcc/xt-clang as toolchains where
their provided newlib are not multi-thread safe, and cannot
be safely used under SMP environment.
So simply ignore any newlib tests as they are not really
providing any values on these platforms.
Fixes#100895
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Instead of directly calling the current thread-specific time slice
handler in z_time_slice(), we must call a saved copy of the handler
that was made when _sched_spinlock was still held. Otherwise there
is a small window of time where another CPU could change the handler
to NULL just before we call it.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
When k_sched_time_slice_set() is called, the current time slice
should not be reset if the current thread is using thread-grained
time slicing. This is to maintain consistency with the already
established idea that thread-grained time slicing takes precedence
over the system-wide time slice size `slice_ticks`.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
This fixes several minor items related to the priority or importance
of checks in determining whether the thread can be time sliced.
A thread that is prevented from running can not be time sliced
regardless of whether it was configured for thread-grained
time slicing or not. Nor can the idle thread be time sliced.
If the thread is configured for thread-grained time slicing, then
do not bother with the preemptible or priority threshhold checks.
This maintains the same behavior, and just optimizes the checks.
If the thread is sliceable, we may as well return the size of the
tick slice since we are checking that information anyway. Thus, a
return value of zero (0) means that the thread is not sliceable,
and a value greater than zero (0) means that it is sliceable.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
Within z_sched_ipi() there is no need for the thread_is_sliceable()
test as z_time_slice() performs that check. Since as a result of this
thread_is_sliceable() is now only used within timeslicing.c, the
'static' keyword is applied to it.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
When a driver requires the usage of a low level driver subsystem
(like SPI, I2C, DMA...), it shall enable it instead of depending on it.
Also, make DMA usage depend on its DT properties being defined.
Signed-off-by: Abderrahmane JARMOUNI <git@jarmouni.me>
When a driver requires the usage of a low level driver subsystem
(like SPI, I2C, DMA...), it shall enable it instead of depending on it.
Also, make DMA usage depend on its DT properties being defined.
Signed-off-by: Abderrahmane JARMOUNI <git@jarmouni.me>
Update broadcaster_multiple sample to start multiple
advertising sets of type legacy and extended advertising.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Tune the aux offset calculation overhead assertion to use
EVENT_OVERHEAD_START_US; this is the correct maximum
overhead causing ISR latency for start of a radio event that
occurs thereafter.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Enable similar Controller Kconfig features between nRF52 and
nRF5340 BabbleSIM LE Audio testing.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Disables the shell serial backend in this Kconfig fragment to
prevent conflicting with the serial one
Signed-off-by: Jamie McCrae <jamie.mccrae@nordicsemi.no>
Remove a redundant NULL check on the device pointer in the GPIO
interrupt handler.
The device pointer is dereferenced unconditionally before the check
and cannot be NULL in this context.
No functional change intended.
Signed-off-by: Gaetan Perrot <gaetan.perrot@spacecubics.com>
There are cases where the device can operate by polling
without using interrupts, so `interrupts` are not mark as required.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
- Adds sam_e54_xpro.overlay files for tcc.
- Adds sam_e54_xpro platform allow in testcase.yaml.
Signed-off-by: Farsin NASAR V A <Farsin.NASARVA@microchip.com>
- Add counter tag in the board yaml file seciton to allow CI run TCC
G1 counter tests on this board
Signed-off-by: Muhammed Asif <muhammed.asif@microchip.com>
The tolower() function takes an int parameter. LLVM compilers generate a
warning if a char is passed instead.
Signed-off-by: Keith Short <keithshort@google.com>
Add explicit casts from 'uint8_t *' to 'char *' in at_get_string()
and at_get_raw_string() return statements to resolve implicit cast
warnings. The data.start field is uint8_t *, but the functions
return char *.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
Add macro definitions to map generic ADC instance names to hardware
specific SARADC base addresses, enabling driver portability.
Signed-off-by: Ha Duong Quang <ha.duongquang@nxp.com>
