The previously added overlay can now be used. Wire up SPI tests for the
Nucleo h745zi-q board.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Connect D12(PA6) & D22(PB5) pins on the board to pass the test.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Reviewed-by: Vasilij Strassheim <vasilij.strassheim@linutronix.de>
The board is capable to serve SPI on the Arduino extention header.
Configure and enable SPI to work on the extention.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
The nucleo_h7X3zi overlays use memory attributes and therfore the
dedicated include file. This is unnessesary, since the memory attribute
include is already included from the underlaying CPU base file
"stm32h7.dtsi". Remove the unnessesary includes.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Test that SQE's created with `rtio_sqe_prep_callback_no_cqe` run, but
don't create a completion queue event.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add a helper function that constructs a callback operation that doesn't
generate a completion queue event.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add Doxygen tooltips in the main documentation.
The tooltips are triggered when hovering over C symbols in the
documentation and display a preview of the Doxygen documentation for the
symbol.
Some shadow DOM magic is needed to style the tooltips correctly, without
affecting the rest of the page.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Every High-Speed capable device must be able to enumerate at Full-Speed.
The functionality at different speeds can be different. Sometimes it is
possible to support exactly the same functionality on both High-Speed
and Full-Speed, but sometimes it is not. The problem is particurarly
relevant for UAC2 because it utilizes isochronous endpoints which means
that the available bandwidth is drastically different between High-Speed
and Full-Speed.
Full-Speed isochronous endpoint can support up to 1023 bytes per frame.
Typical 48 kHz 16-bit stereo stream consumes 48 * 2 * 2 = 192 bytes per
frame. Zephyr UAC2 instances with such streams are fine to work both at
Full-Speed and High-Speed.
An example stream that is too much for Full-Speed is the sometimes used
192 kHz 24-bit stereo (whether or not it is useful is out-of-scope here,
because it should be up to application developer) which would require
192 * 3 * 2 = 1152 bytes per frame.
Because the bandwidth required for audio stream depends on three
different parameters (sample rate, bit resolution and number of
channels), the UAC2 implementation should not automatically limit
available parameters to fit bandwidth requirements.
Adding explicit full-speed and high-speed boolean options to zephyr,uac2
compatible seems to be not only the easiest solution to the problem, but
also the most flexible one. Depending on the use case, the application
developer can then decide for example:
* to not support High-Speed at all - by having one zephyr,uac2 instance
with full-speed property
* to not support Full-Speed at all - by having one zephyr,uac2 instance
with high-speed property
* to support limited number of channels at Full-Speed and all channels
at High-Speed - by having one zephyr,uac2 instance with full-speed
property and separate instance with high-speed property
* to have exactly the same functionality at both speeds - by having one
zephyr,uac2 instance with both full-speed and high-speed properties
Signed-off-by: Tomasz Moń <tomasz.mon@nordicsemi.no>
Add a composite fuel-gauge device to the Thingy53 to enable converting
the raw voltage to a battery percentage.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Support additional properties from the linux `battery.yaml`.
The `ocv-capacity-table-0` definition differs from the Linux version as
we do not have a good way to generate variable length arrays in our
config structs, or a good way to separate out the percentage vs voltage
in the mapping.
We reduce the flexibility by enforcing a step size of 10%, which removes
the need for the percentage to be in the array and solves the variable
length problem.
To simplify the effort of defining these voltage curves, default curves
for a variety of chemistries have been added, extracted from datasheet
graphs of discharge curves from reputable manufacturers.
Signed-off-by: Jordan Yates <jordan@embeint.com>
- HRS control point was added to server but no write callback is
defined, this causes issues if control point characterstic is
written by the client
Signed-off-by: Anuj Pathak <anuj@croxel.com>
Modify the function "i2s_stm32_set_clock" to handle the case where pclk_len
it is not greater then 1 and to ensure that the freq_in
variable is correctly set in both cases.
Signed-off-by: IBEN EL HADJ MESSAOUD Marwa <marwa.ibenelhadjmessaoud-ext@st.com>
RT1011 expects it's flash configuration block at a different offset than
the rest of the RT10xx series. Add default to fix the platform not
booting.
Signed-off-by: Declan Snyder <declan.snyder@nxp.com>
I2S driver was merged after interrupt .dtsi was changed,
causing CI to fail. This updates it accordingly.
Signed-off-by: Sylvio Alves <sylvio.alves@espressif.com>
Extensions could have used logging, when log processing is
deferred, the logging thread can run after the extension has
been unloaded and thereby access invalid memory addresses.
Make sure to flush all logs before unloading extensions.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Zephyr already has various FlexIO drivers but imxrt11xx didn't enable
the base flexio peripheral.
Signed-off-by: Peter van der Perk <peter.vanderperk@nxp.com>
Due to issues with running the watchdog test in a fully
automated manner on the IT8XXX2_EVB, it was previously removed.
However, a fix has now been merged, and it should be operational
on this board once again.
Signed-off-by: Patryk Kuniecki <patryk.kuniecki@intel.com>
Flash partition was defining a scratch area which was implying the use
of mcuboot BOOT_SWAP_USING_SCRATCH f/w update method.
By removing this partition, mcuboot will automatically use
BOOT_PREFER_SWAP_MOVE method which should be prefered as it levels
flash wear. Incidentally increase slot1 by one sector (as this is
preferred for the swap method).
Signed-off-by: Erwan Gouriou <erwan.gouriou@st.com>
1) After the wifi driver successfully connects to the AP,
the dormant state should be set to OFF.
Once a deauth/disassoc frame is received or link lost occurs,
the dormant state should be set to ON.
2) Enable CONFIG_NET_TC_SKIP_FOR_HIGH_PRIO and CONFIG_NET_CONTEXT_PRIORITY
to push high priority packets directly to network driver.
Signed-off-by: Gang Li <gang.li_1@nxp.com>
The STM32H5x adc has a special option register that
needs to be set when using channel 0 on adc1.
fixes: #77618
Signed-off-by: Lars Jeppesen <lje@foss.dk>
The gen_kobject_list.py script expects the __subystem declaration to end
with _driver_api. Adjust the stepper api and existing driver implementation
accordingly.
Signed-off-by: Fabian Blatz <fabianblatz@gmail.com>
Update the EDTT module to:
b9ca3c7030518f07b7937dacf970d37a47865a76
Including the following:
* b9ca3c7 Fix for for python >= 3.11
* fe9b1d1 Corrected ISO interval for LL/CIS/PER/BV-05-C
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Some hardware configuration require rely on the ALERT/RDY
pin to know when and ADC conversion is completed.
The polling thread is left as fallback, when the pin
is not defined.
Signed-off-by: Efrain Calderon <efrain.calderon.estrada@gmail.com>
Add a new macro, DT_ANY_COMPAT_HAS_PROP_STATUS_OKAY
name, evaluates to 1 if any enabled instance of `compat` has the
property or to zero if it hasn't.
This macro can be useful in drivers for a family of devices.
Signed-off-by: Efrain Calderon <efrain.calderon.estrada@gmail.com>
This example demonstrates I2S audio playback combined with the codec
API. It can use the dmic API for audio input.
Signed-off-by: Hake Huang <hake.huang@oss.nxp.com>
Signed-off-by: Vit Stanicek <vit.stanicek@nxp.com>
