Do not delete the thread from within the thread callback function,
secure thread_delete against trying to delete a terminated thread.
Signed-off-by: Fabian Blatz <fabianblatz@gmail.com>
Fix a bug in the test which caused it to fail (depending on compiler,
platform and optimization level).
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Change direction for raw data file from $ZEPHYR_BASE
to <path_to>twister-out/../<test>/
Signed-off-by: Arkadiusz Cholewinski <arkadiuszx.cholewinski@intel.com>
Fix broken copyright parsing. Since REUSE 6.0.0 the copyright_lines
attribute has been removed. The new attribute copyright_notices is used.
Additionally, the python package REUSE should now be at least v0.6.0.
This will require an update of REUSE for users that were using a
version earlier than 0.6.
Fixes#98378
Signed-off-by: Simone Orru <simone.orru@secomind.com>
As of Core v6.2, the passkey entry pairing method for legacy pairing
does no longer grant authenticated MITM protection. This commit
updates `smp.c` accordingly to not grant the authenticated states when
using legacy passkey entry pairing.
Adds a check to make sure that bonds that have been stored persistently
adheres to these changes. Bonds that have been generated using the
legacy passkey entry pairing method will thus be downgraded from
authenticated to unauthenticated when restored from storage.
Signed-off-by: Håvard Reierstad <haavard.reierstad@nordicsemi.no>
Update top-level comment of `include/zephyr/crypto/crypto.h` as the API is
no longer experimental, as indicated by the @version 1.0.0 tag in Doxygen.
Signed-off-by: Mathieu Choplain <mathieu.choplain-ext@st.com>
MAX32657 NS does not have access to the flash peripheral, so ensure the
placeholder flash controller node is disabled, and update the "storage"
node in the evkit board definition to properly document its use for TFM.
Signed-off-by: Pete Johanson <pete.johanson@analog.com>
remove mdio_bus_enable/mdio_bus_disable, as they are no
longer needed and make the mdio api simpler.
Signed-off-by: Fin Maaß <f.maass@vogl-electronic.com>
Removes differentiation between pdl and hal based drivers from device
tree. This differentiation is now done as a Kconfig option.
Signed-off-by: John Batch <john.batch@infineon.com>
Fixing build errors related to reference to an incorrect define.
Adds missing headers to pinctrl_soc.h for Edge platform.
Signed-off-by: John Batch <john.batch@infineon.com>
Add checks for the values in btp_ascs_preconfigure_qos
to determine if they are valid. If they are invalid
we reject the command.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
If both the TX and RX SDUs are 0, then we reject
the call as that is an invalid configuration
(minimum SDU size is 1, and 0 indicates that
no data in that direction is being configured).
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Fixes the max32 can driver to follow the convention for drivers to
select pin control if they need it.
Signed-off-by: Maureen Helm <maureen.helm@analog.com>
SDIO is now fully supported on STM32 targets, and this enabled the use
of Wi-Fi functionality on Arduino H7 boards. Remove the note in the
documentation that states only Bluetooth is supported.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
The tests are failing in architectures that support SMP. Filtering them to
make CI happy, as done in #97827. Issue #98217 was created to address that
separately for all tests.
Signed-off-by: Rodrigo Peixoto <rodrigopex@gmail.com>
Add a new type of observer: Async Listeners. They are executed within the
system work queue context, offering distinct advantages:
- Asychrouness: obviously it gains the ability of run some urgent/light
work load in a separate context.
- Prioritization: Async Listeners typically execute before other
application observers, as they run within the system work queue context
which is cooperative.
- Reliability: Ensures no message loss during execution. They use the same
mechanism as Message Subscribers.
Signed-off-by: Rodrigo Peixoto <rodrigopex@gmail.com>
This commit deinitializes platform modules when external network
interface is brought down. It also delets the multicast routes that are
added for OpenThread interface.
Signed-off-by: Cristian Bulacu <cristian.bulacu@nxp.com>
In this commit, `net_socket_service_register` is called when a platform
module that has sockets is deinitialized, and it's socket/sockets are
closed.
This commit also handles a case when a module tries to join a multicast
group and a subscription, from another module, is already present.
Also, covered network namespace changes done in #99169
Signed-off-by: Cristian Bulacu <cristian.bulacu@nxp.com>
OpenThread interface is initialized beform Wi-Fi interface.
`otPlatTrelEnable` is called by OpenThread stack when it's interface is
being initialized. Given this scenario, socket operation, like `bind`,
will fail. There is also no mean to get a valid pointer to backbone
interface. This is why, `trel_plat_init` was declared and called when
backbone interface reported connectivity.
This commit handles the `ot trel disable/ot trel enable` scenario that
can be initialized from CLI. `otPlatTrelEnable` will be called, but
`trel_plat_init` will not be called anymore, leaving trel socket without
any options set, and `net_socket_service_register` is not called
anymore, to handle incoming traffic.
Now, when `otPlatTrelEnable` is called, it will verify if Wi-Fi
interface is connected and will call `trel_plat_init` if needed.
Signed-off-by: Cristian Bulacu <cristian.bulacu@nxp.com>
Zephyr subsystems' headers should not duplicate C library prototypes
(in this case strdup(), which is either ISO C23 or a POSIX extension to
the C library <string.h>).
Instead they should request those prototypes from the C library.
By now Zephyr only requires ISO C17, but many C libraries will have
strdup() and expose it also when the POSIX extensions prototypes are
requested, so let's request these prototype from the C library by
setting the feature test macro _POSIX_C_SOURCE to version 2008.09
which includes it.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Fix a missing unref of a bt_conn reference, leading to a ref count
mismatch, and causing the following warning to be printed:
bt_conn: Found valid connection ... in disconnected state.
Signed-off-by: Aleksandar Stanoev <aleksandar.stanoev@nordicsemi.no>
USBD_MAX_UDC_MSG configures the number of events coming from the UDC
driver that the stack can keep. This can be filled very quickly if there
would be multiple bus events for some reason, or function handlers of
those events are blocked for long time. As a consequence, subsequent
events could be dropped, and completed transfers not handled. To avoid
it, we can store completed transfer requests in a slist and check it on
every event.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
SOC_RESET_HOOK is already too late because the code touches ram already
instead use the newer SOC_EARLY_RESET_HOOK so we can do ECC
initialization before using the memory
Signed-off-by: Peter van der Perk <peter.vanderperk@nxp.com>
Verify that dns_unpack_name() generates a valid DNS name when unpacking
records and that it returns an error in case of overflow.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
As the loop unpacking the DNS name from records checks the current
label length on each iteration, it's also needed to update the remaining
buffer length on each iteration, otherwise the buffer length checks
doesn't work as expected.
Additionally, the remaining buffer checks while technically worked, they
were conceptually wrong and unintuitive. The buf->data pointer doesn't
move, so comparing against this pointer when adding new labels doesn't
make sense. It's more intuitive to simply compare the label size vs
the remaining buffer space.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The rewrite by ruff seems to wrongly pass a single argument to filter(),
as seen in the logs:
```
f.process()
File "/home/runner/work/zephyr/zephyr/zephyr/./scripts/ci/test_plan.py",
line 145, in process
self.find_boards()
File "/home/runner/work/zephyr/zephyr/zephyr/./scripts/ci/test_plan.py",
line 292, in find_boards
self.resolved_files.extend(list(filter(resolved_filter)))
^^^^^^^^^^^^^^^^^^^^^^^
TypeError: filter expected 2 arguments, got 1
```
Split things up properly to avoid this.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Verify that if the application triggers asynchronous connect with
non-blocking sockets, and starts to monitor the socket with poll()
immediately but with POLLIN only set, the poll() is still able to
report an error if the connection fails. As no POLLOUT is monitored, the
application won't know when the connection is done in case of success,
but it should still be notified in case of errors.
To achieve this, modify the existing test case to allow to specify what
events should be monitored by poll(). Additionally monitor the time
spent in poll() - an error should cause poll() to exit immediately,
not after the specified timeout.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
If a thread was sleeping in poll(), monitoring POLLIN only, it should
still be notified if an asynchronous connection, triggered by
non-blocking connect(), failed. Currently that's not the case.
In result the application would never know whether the connection
was successful or not and would be stuck with a disconnected socket
that would not report anything with poll().
This was not an issue if POLLOUT was monitored as well, because POLLOUT
sets up a connect semaphore that was reset in case of errors. POLLIN
however only monitors the recv fifo, which was not waken up in such
case.
Fix this by canceling any pending waits on recv fifo and recv
condition variable.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The function prototype for `z_riscv_pmp_read_config()` was previously
declared in `tests/arch/riscv/pmp/clear-pmp-unlocked-entries/src/main.c`.
This local declaration is no longer necessary as the prototype is now
defined within the centralized PMP header, `include/arch/riscv/pmp.h`,
which is already included via `kernel_internal.h`. The prototype
is guarded by the `CONFIG_ZTEST` Kconfig option.
Signed-off-by: Firas Sammoura <fsammoura@google.com>
- BASS mandates that a remove_source operation is not done if
we are synced to PA or BIS. This PR fixes that
- Update BabbleSim tests to reflect this behavior
- Fixes BASS/SR/SPE/BI-05-C
Signed-off-by: Alexander Svensen <alexander.svensen@nordicsemi.no>
Change the zephyr,memory-attr of the ext_memory / ext_flash_mem node to
ATTR_MPU_FLASH because ATTR_MPU_IO does not allow unaligned memory access.
Signed-off-by: Thomas Decker <decker@jb-lighting.de>
Remove the ext_memory: memory@70000000 node as the external flash memory
is part of the board design. The nucleo_h7s3l8 and stm32h7s78_dk boards
already have this node.
Signed-off-by: Thomas Decker <decker@jb-lighting.de>
Until now the code was using the path to the .hex file to select the
directory where the generated JSON file required for nrfutil would be
generated. But this has a problem: if the .hex file in a sysbuild
project is located in the tree as a precompiled binary, the runner would
place a file in there and make the tree dirty. Instead use the build_dir
folder which always corresponds (in both sysbuild and regular builds)
to the current target build directory.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
This cannot possibly work, if POSIX is disabled, then the POSIX
symbols are also not found so remove the "without_posix_api" tests.
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
west: update hal_nxp to integrate mcux: middleware: ieee_802_15_4 MAC: Add
CONFIG_NXP_IEEE802154_MAC
zephyr: blobs: add MCXW71 & MCXW72 BLE controller and Host libraries
(ieee-802.15.4-mac) to MCUXSDK 25.09.00 release
This new version of the BLE controller is for ZigBee support and
implements ieee-802.15.4 MAC interface.
Update connectivity framework for MCXW71 & MCXW72 BLE controller
(ieee-802.15.4 MAC) from MCUXSDK 25.09 release.
Signed-off-by: Guillaume Legoupil <guillaume.legoupil@nxp.com>
STM32H7 platforms now enable cache management by default.
To ensure data coherency after flash writes, invalidate cache
lines to the region written.
This prevents stale data and ensures proper memory visibility.
Signed-off-by: Fabrice DJIATSA <fabrice.djiatsa-ext@st.com>
