`TestInstance.get_elf_file()` does not include the binary name used for
unit testing (`testbinary`) in its search, causing unit tests to not
complete successfully. This does not cause tests to fail and thus goes
unnoticed.
Update the glob expressions used in `get_elf_file()` to also look for
files called `testbinary`
Signed-off-by: Tristan Honscheid <honscheid@google.com>
This applies the coccinelle script to another set of files:
samples/bluetooth/bthome_sensor_template/src/main.c
samples/boards/stm32/power_mgmt/standby_shutdown/src/main.c
samples/drivers/smbus/src/main.c
samples/drivers/virtualization/ivshmem/doorbell/src/ivshmem.c
samples/fuel_gauge/max17048/src/main.c
samples/hello_world/src/main.c
samples/sensor/proximity_polling/src/main.c
samples/subsys/logging/ble_backend/src/main.c
tests/drivers/build_all/mfd/src/main.c
Signed-off-by: Keith Packard <keithp@keithp.com>
Add a paragraph explaining the change from void main(void) to int
main(void) and the requirement for main functions to return zero.
Signed-off-by: Keith Packard <keithp@keithp.com>
Remove RETURN_FROM_MAIN macro which performed control flow, violating
compliance tests. Replace with explicit use of posix_exit_main or return as
appropriate.
Signed-off-by: Keith Packard <keithp@keithp.com>
These flags were added to avoid warnings when main was declared to return
void. Now that main returns int, those warnings will flag errors.
Signed-off-by: Keith Packard <keithp@keithp.com>
Looks like switching the main return value to int means that stack
frame persists and increases stack usage by a few bytes. Increase the
main stack size to avoid overflows.
Signed-off-by: Keith Packard <keithp@keithp.com>
These changes were automated using coccinelle with the following
script:
@@
@@
- void
+ int
main(...) {
...
- return;
+ return 0;
...
}
Signed-off-by: Keith Packard <keithp@keithp.com>
As both C and C++ standards require applications running under an OS to
return 'int', adapt that for Zephyr to align with those standard. This also
eliminates errors when building with clang when not using -ffreestanding,
and reduces the need for compiler flags to silence warnings for both clang
and gcc.
Most of these changes were automated using coccinelle with the following
script:
@@
@@
- void
+ int
main(...) {
...
- return;
+ return 0;
...
}
Approximately 40 files had to be edited by hand as coccinelle was unable to
fix them.
Signed-off-by: Keith Packard <keithp@keithp.com>
As both C and C++ standards require applications running under an OS to
return 'int', adapt that for Zephyr to align with those standard. This also
eliminates errors when building with clang when not using -ffreestanding,
and reduces the need for compiler flags to silence warnings for both clang
and gcc
Signed-off-by: Keith Packard <keithp@keithp.com>
As both C and C++ standards require applications running under an OS to
return 'int', adapt that for Zephyr to align with those standard. This also
eliminates errors when building with clang when not using -ffreestanding,
and reduces the need for compiler flags to silence warnings for both clang
and gcc.
Signed-off-by: Keith Packard <keithp@keithp.com>
[PR#51217](https://github.com/zephyrproject-rtos/zephyr/pull/51217)
changed the `init_entry` structure, but `Z_INIT_ENTRY_NAME` (called
by `SYS_INIT` or `SYS_INIT_NAMED`) does not initialize all of the
members of the struct, leading to errors when building with
`-Werror=missing-field-initializers`.
Change the macro to initialize the `dev` member to `NULL` so that
all members of the struct are initialized.
Signed-off-by: Paul Fagerburg <pfagerburg@google.com>
This commit updates the CI workflows to use the CI image v0.26.1, which
is based on Ubuntu 22.04 and and includes LLVM 16.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit updates the CI workflows to use the GitHub Ubuntu 22.04
virtual environment for running jobs.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This adds USB-HS support for LPC55S16, much in the same way that
LPC55S28 support was added previously.
Signed-off-by: Maxime Vincent <maxime@veemax.be>
Now that there is a generic die_temp_polling sample, this specific STM32
sample is redundant and can be removed.
Signed-off-by: Guillaume Gautier <guillaume.gautier-ext@st.com>
Add die_temp0 alias for all the STM32 boards that already enable the
temperature sensor so that the die_temp_polling samples can be used.
Also enable die_temp for two boards that are supported by the
stm32_temp_sensor sample.
Signed-off-by: Guillaume Gautier <guillaume.gautier-ext@st.com>
The audio code utilizes BT_CONN so ensure its only built if
CONFIG_BT_CONN is set. This is to fix a build issue we see
with arm-clang for:
sample.bluetooth.broadcast_audio_source
which fails to link since bt_conn_get_info is missing.
Signed-off-by: Kumar Gala <kumar.gala@intel.com>
Test that if all targets are lost because of failing each DFU Transfer
step `target_lost` callback is called on every one, and `end` callback
is called as procedure aborts.
Signed-off-by: Krzysztof Kopyściński <krzysztof.kopyscinski@codecoup.pl>
Fix window offset calculation which is missing the radio
rx ready delay timing be subtracted because the connection
offset from the start of the radio for scan window to the
end of the ADV_IND PDU includes the radio rx ready delay
timing, where as the determined window offset by advanced
scheduling calculates the offset from the ticks_at_expire
of the scan window to an active central connection's
ticks_at_expire.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Update Advanced Scheduling implementation to use remainder
value in precisely calculating the connection request window
offset in microseconds.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Move the use of rx chain delay value used to calculate the
window offset into lll_scan_prepare_connect_req() instead of
current two places, one for legacy and other for extended
connection setup.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Find and match only Central role ticker when applying the
BT_CTLR_CENTRAL_SPACING value.
Use the EVENT_TICKER_RES_MARGIN_US value as +/- jitter
margin when ticker advanced scheduling is used to group
radio events.
When using CONFIG_BT_CTLR_CENTRAL_SPACING, do not add the
+/- jitter to the ticks_slot.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add display to default configuration.
Update board file to work with changes to graphics library.
Enable compatibility with new version of LVGL.
Prevent hard fault when running LVGL sample.
Add alias for touch controller.
Signed-off-by: Andrew Hedin <andrew.hedin@lairdconnect.com>
Add bt-hci-rpmsg-ipc to DTS.
Move common items to their own files.
Align debug setup with nrf5340 DK.
Remove scratch partition.
Signed-off-by: Andrew Hedin <andrew.hedin@lairdconnect.com>
Add support for negative temperature coefficient for single calibration
chips in the STM32 temperature driver.
That feature is necessary for STM32F0x0.
Signed-off-by: Guillaume Gautier <guillaume.gautier-ext@st.com>
The temperature sensor of the STM32F0x0 is similar to C0 (with one value
for calibration) but uses a negative coefficient, so we add it to the
bindings, just like in st,stm32-temp.yaml
Signed-off-by: Guillaume Gautier <guillaume.gautier-ext@st.com>
-- Unit test to check that the correct payload number is selected based
on the SDU packet number and the time stamp when fragmenting SDUs
into PDUs.
-- Ran clang-format on isoal_test_tx and isoal_test_rx
Signed-off-by: Nirosharn Amarasinghe <niag@demant.com>
Included consideration for the packet sequence number and time stamps
together with the SDU interval in matching the event for fragmentation
of SDUs to unframed PDUs.
Implemented:
-- Increase in kconfig limit of BT_CTLR_ISO_TX_BUFFERS to allow full 32
buffers to be available for LL/CIS/PER/BV-39C
-- ISO-AL microsecond time wrapping exposed to be used externally to
adjust incoming Group Reference Points and Time-stamps where required
-- Function that handles adjustment of payload number according to
incoming SDU exposed for external use to allow next payload number to
be accessed by ISO transmit tests
-- Changed internal fragmentation source identification parameter from
the source structure pointer to the source handle so that only the
source handle needs to be used for external calls
Signed-off-by: Nirosharn Amarasinghe <niag@demant.com>
The platform nrf52_bsim was specified as an integration platform,
but was missing from the allow list.
This commit adds this platform to the allow list.
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
__noasan attribute was accidentally removed by
a5fd0d184a. Add it back as it is causing
some address sanitizer issues when using the LLVM toolchain.
Details on why this is needed can be found in
74cc534758.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The native_tasks definition was directly in the soc_inf soc.h
header. But soc.h pulls a lot of other headers.
Some of those could cause conflicts, say with application
headers, for users who only wanted the be able to register
native tasks in a module.
Let's refactor the native tasks definitions into their own header
and include that header from soc_inf's soc.h.
In this way users who need only need to register a native tasks
can just include posix_native_tasks.h, and all previous users
see no change.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Fix broadcast audio source and sink samples' conf file to
use Broadcast ISO feature in Zephyr Bluetooth Controller.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This test is going away when #55976 is merged.
Add a sleep to work around this sequence of events:
- central establishes channels
- peripheral tries to establish channels, fails (max num reached)
- central (still thinking chans are valid) triggers reconfigure
- collision mitigation on the peripheral, triggering a disconnect
of both sides
- channels are established again, but the reconfigure isn't
triggered again
- devices wait forever for the reconfigure cb
Fixes#55912
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
This new configuration option, SHELL_BACKEND_RTT_BUFFER, allows selecting
an alternative buffer for the Shell's RTT backend. By default buffer 0 is
used, which conflicts with the default logging subsystem RTT backend
buffer.
This option is the counterpart to the logger's LOG_BACKEND_RTT_BUFFER.
Signed-off-by: Maurits Fassaert <maurits.fassaert@sensorfy.ai>
Some functions such as the ISH power management function needs to
reserve GDT entry place holders which are filled during runtime.
Add config option to define the number of GDT entry place holders
reserved and change the ia32 linker script accordingly.
Signed-off-by: Leifu Zhao <leifu.zhao@intel.com>
add support for async call and repeat sample test
1. change the DMA req to 2 byte each
2. increase the buffer pre-dma
3. add protection on invalid buffer
depends on: PR #56104
fixing: issue #56070
Signed-off-by: Hake Huang <hake.huang@oss.nxp.com>
Private Beacons are mesh-1.1 feature and without mesh-1.1 spec enabled,
the node shall not process such beacons.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Private Beacons and Secure Network Beacons are not mutually exclusive by
spec. This means both beacons can be sent simultaneously. For both
beacons spec defines their own observation intervals and related
parameters, which means we need to monitor both beacons types
separately.
Signed-off-by: Pavel Vasilyev <pavel.vasilyev@nordicsemi.no>
Switch the NXP i.MX RT1060 evaluation kit from using FlexCAN2 to FlexCAN3.
FlexCAN1 and FlexCAN2 are standard FlexCAN controllers supporting CAN 2.0b,
whereas FlexCAN3 supports both CAN 2.0b and CAN-FD. Switching to FlexCAN3
by default allows testing the CAN-FD capabilities of this board.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
