Increase stack size for test thread. With no optimization
default stack size may not be enough.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Increase timer timeout. 5ms timeout was too short for some
devices (e.g. with slow and blocking uart output).
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
This is a bug fix. A pointer to the spi configuration is not saved when
the spi driver is configured for slave operation and it can lead to
runtime errors.
Signed-off-by: Bryce Wilkins <bryce.wilkins@gmail.com>
() Fix the compatible string from "intel,dai,dmic" to
"intel,dai-dmic".
() Also rename the yaml file to have vendor name first.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Make all series consistent with regard to where code is located and put
the power init code in power.c alongside other power related code.
Reduce soc.c per series to an empty file that can be removed.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
initializing the SoC is the same across series, so put everything in one
place in the common code base.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Al is part of the core team working on testing & emulation at Google.
He should be added to reviews for changes pertaining to these.
Signed-off-by: Yuval Peress <peress@google.com>
Changing mkfs options from FM_FAT to FM_ANY so that on larger storage
FAT32 instead of FAT16 is automatically created.
Signed-off-by: Nahal Farhi <nahal@whisper.ai>
Sensor value unit is [m/s^2].
Currently calcs as [g/1000] unit.
Correcting calculation to convert [m/s^2].
Store the integer part to val1, and the fractional part to val2.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@fujitsu.com>
Fix build error since the I2C controller is not enabled in the devicetree
for the bus that the ccs811 sensor is on.
Signed-off-by: Kumar Gala <galak@kernel.org>
Now that counter drivers are enabled based on devicetree
we need to remove any cases of them getting enabled by
Kconfig.defconfig* files as this can lead to errors.
Signed-off-by: Kumar Gala <galak@kernel.org>
Header file generation from device tree was failing in pulling mode.
Restrict the interrupt node usage to interrupt mode only.
Signed-off-by: Romain Mahoux <romain@mahoux.fr>
When there is no external pull-up on the CSN line and the line
is put into the low-power mode (input direction and disconnected
buffer), after some time its level falls to low and some flash
chips, like for example MX25R64, exit the Deep Power-down mode
when that happens. To prevent this, activate the GPIO internal
pull-up for this line in boards that do not have the external
pull-up on it.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
Channels are not unique across bands, they need to be interpreted along
with frequency band. Add a separate band parameter that supports 2.4, 5
and 6 GHz bands.
Also, 6GHz has maximum channel of 233 (as of now), so, change the
maximum channel to that, this is a global maximum, per-band channel
validation is left to the offload driver as that depends on regulatory
rules (not a simple maximum check).
Signed-off-by: Krishna T <krishna.t@nordicsemi.no>
MFP (Management frame protection) a.k.a PMF (protected management
frames) is now a mandatory feature for certification esp. for WPA3.
Support is added for both scan and connect similar to security type.
Signed-off-by: Krishna T <krishna.t@nordicsemi.no>
SAE password is similar to WPA2 password but without any length
restrictions, so, add a separate optional field, if this is not given we
can fallback to WPA2 password (psk).
Signed-off-by: Krishna T <krishna.t@nordicsemi.no>
This is a variant of standard PSK which uses KDF instead of PRF for
key generation, see section Table 9-151 in [1].
[1] - https://standards.ieee.org/ieee/802.11/7028/
Signed-off-by: Krishna T <krishna.t@nordicsemi.no>
This commit enables the assert test mode (`CONFIG_ASSERT_TEST`) for the
ARM interrupt test because it relies on the assert function to return
without aborting in the "Assert occurring inside kernel panic" test.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit enables the assert test mode (`CONFIG_ASSERT_TEST`) for the
ARM interrupt test because it relies on the assert function to return
without aborting in the in-ISR "Intentional assert" test.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit enables the assert test mode (`CONFIG_ASSERT_TEST`) for the
ztest error hook test because it implements a custom post assert fail
hook (`ztest_post_assert_fail_hook`) that returns without aborting to
faciliate the testing of the assert functions.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit adds the `CODE_UNREACHABLE` hint at the end of the
assertion failure branch so that the compiler takes note of the assert
function not returning when an assertion fails.
This prevents the compiler from generating misguided warnings assuming
the asserted execution paths.
It also introduces the `ASSERT_TEST` Kconfig symbol, which indicates
that the "assert test mode" is enabled. This symbol may be selected by
the tests that require the assert post action function to return
without aborting so that the test can proceed.
Note that the `CODE_UNREACHABLE` hint is specified only when the assert
test mode is disabled in order to prevent the tests from crashing when
the assert post action function returns.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
The regulator voltage should be set before the clocks are enabled.
This is especially the case when the MSIS at 48MHz is selected as
SYSCLK.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
machine-timer node name changed in #48429.
but Kcondif.defconfig.gd32vf103 not followed it.
It makes misconfigures the RISCV_MACHINE_TIMER_SYSTEM_CLOCK_DIVIDER.
So the machine-timer did not count the actual time.
Correcting node name to fix this problem.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
IRQ7 is placed on the second element of interrupt definition.
Select it by DT_INST_IRQ_BY_IDX() explicitly.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
Add the hearing aid sample to the footprint tracking for the
nrf52840dk_nrf52840 and nrf5340dk_nrf5340_cpuapp.
The sample should represent a simple hearing aid implementation,
and is thus a very good candidate for tracking the footprint
of LE Audio in Zephyr.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Change introduces authentication callbacks used only for specified
BLE connection. The feature can be used by devices that require
reporting specific bonding capabilities only when pairing using
vendor-specific procedures.
If per-connection authentication callbacks are defined for given
connection, they are used instead of global authentication
callbacks. SMP latches authentication callbacks during the first
access to prevent updating the callbacks while pairing.
Fixes: #38336
Signed-off-by: Marek Pieta <Marek.Pieta@nordicsemi.no>
The Babblesim test for eatt for reconfiguration of MTU is not
foolproof. Central actually generates two mtu update events.
This is fixed by only setting the reconfigured flag when the MTU actually
changes to the new values
In the Babblesim test for eatt for collision the peripheral
needs to sync with central to ensure collisions occur.
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
Add `OPENTHREAD_CONFIG_MESSAGE_BUFFER_SIZE` to Kconfig.
Also set the number of children to minumum possible for MTD builds
in order to save some resources (~512B of RAM).
Signed-off-by: Eduardo Montoya <eduardo.montoya@nordicsemi.no>
Devicetree aliases, accel0 and magn0, were recently introduced to enable
sensor sample applications to support multiple sensor drivers of the
same type and thereby reduce the number of driver-specific sensor sample
applications.
This was implemented by adding devicetree nodelabels to accelerometer
and magnetometer sensor nodes, which unfortunately can suffer from
naming conflicts when using shield boards. Fix the problem by appending
the I2C bus address or SPI chip select to the devicetree nodelabel for
sensor nodes on shield boards only. Base boards are left alone to keep
their nodelabels as short as possible.
Signed-off-by: Maureen Helm <maureen.helm@intel.com>
This commit adds an initialisation for the `bias_dims` variable, which
is given as an input to the `arm_depthwise_conv_s8` function.
Note that the `bias_dims` parameter is currently unused by the function
implementation.
This fixes the following warning reported by the GCC 12:
error: 'bias_dims' may be used uninitialized
Werror=maybe-uninitialized]
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
