In current implementation, the bonding information of classic is
queried by calling the function `bt_foreach_bond()`. And the bonding
information of classic is deleted by calling `bt_unpair()`.
There are two issues if the LE and classic are bonded at same time for
dual mode peer device.
Issue 1, for the function `bt_foreach_bond()`, there are two bonding
information will be found. But there is no way to find which bonding
information belongs to LE or classic.
Issue 2, For the function `bt_unpair()`, all bonding information will
be deleted if the passed address is the public address. But there is
no way to delete the bonding information of LE or classic.
Remove the calling of function `bt_foreach_bond_br()` from the
function `bt_foreach_bond()`. And rename it to `bt_br_foreach_bond()`,
and public it as an API to query the bonding information of classic.
Remove the calling of function `bt_keys_link_key_clear_addr()` from
the function `bt_unpair()`. Add an API `bt_br_unpair()` to delete the
bonding information of classic.
Add a `br_bond_deleted` to structure `bt_conn_auth_info_cb` for
classic. The callback will be triggered if the bonding information has
been deleted by the function `bt_br_unpair()`.
Add a `clear` shell command for classic to delete the bonding
information.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
The i2c driver is missing the option to check if a I2C_IODEV is ready for
use. The proposed function works the same as the check for a SPI_IODDEV
to keep the use consistent across different bus systems.
Signed-off-by: Patrik Sieverding <sieverdp@uni-bremen.de>
Added support for differentail or single ended setup.
Reading resolution form dt - adc spec.
Signed-off-by: Dimitrije Lilic <dimitrije.lilic@orioninc.com>
In current implementation, when receiving the L2CAP frame
`L2CAP_COMMAND_REJECT_RSP` if the L2CAP is working as client role, the
L2CAP frame `L2CAP_COMMAND_REJECT_RSP` will be sent by the L2CAP
client incorrectly.
When the command response `L2CAP_COMMAND_REJECT_RSP` is received, find
the channel of the command request according to the received `ident`.
If the channel is found, try to disconnect the channel. If fail to
disconnect the channel, remove and delete the channel.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
In current implementation, the `ident` of the command response is not
checked whether it is same as the `ident` sent by the L2CAP command.
Keep the `ident` of the command. If the received `ident` of the
command response is not aligned with the `ident` of the command,
discard the response silently. Or, clear the saved `ident`.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
If more than one CIS in the same CIG was shutdown at the same time
(e.g. due to supervision timeout), the mayfly calling cis_tx_lll_flush
would be busy and the assert would be triggered
cis_tx_lll_flush now operates similar to cis_disabled_cb; Loop through
all CISes for a CIG and flush any marked as LLL_CIS_FLUSH_PENDING
Signed-off-by: Troels Nilsson <trnn@demant.com>
Allows erasing secondary slot which is marked for test or confirmed.
This is safe as bootloader doesn't make any action on boot-setup yet.
Erase of such pending image might considered like the case when it
was never downloaded as well.
This allow user to not stuck with pending irremovable image.
Signed-off-by: Andrzej Puzdrowski <andrzej.puzdrowski@nordicsemi.no>
When sys_heap_realloc() is expressed in terms of sys_heap_aligned_realloc()
it invokes a longer aligned allocation code path with an extra runtime
overhead even though no alignment is necessary.
Let's reference and invoke the aligned allocation code path only when an
actual aligned allocation is requested. This opens the possibility for
the linker to garbage-collect the aligning code otherwise.
Improve realloc documentation while at it.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
When k_malloc() is expressed in terms of k_aligned_alloc() it invokes a
longer aligned allocation code path with an extra runtime overhead even
though no alignment is necessary.
Let's reference and invoke the aligned allocation code path only when an
actual aligned allocation is requested. This opens the possibility for
the linker to garbage-collect the aligning code otherwise.
Also bypass k_heap_malloc() and friends given they're invoked with
K_NO_WAIT. Go directly to sys_heap_*() instead to cut some more unneeded
overhead.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
In .../modules/debug/percepio/TraceRecorder/kernelports/Zephyr/include/\
tracing_tracerecorder.h there is a concealed non-parameterized direct
reference to a local variable that is no longer in scope. Provide a dummy
stub for compilation to succeed until that module's layering violation is
fixed, after which this could be reverted.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
When k_heap_alloc() is expressed in terms of k_heap_aligned_alloc()
it invokes a longer aligned allocation code path with an extra runtime
overhead even though no alignment is necessary.
Let's reference and invoke the aligned allocation code path only when an
actual aligned allocation is requested. This opens the possibility for
the linker to garbage-collect the aligning code otherwise.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Let's avoid division and modulus operations as they're costly... and even
more so when they're unnecessary as in this case the main constraint is
about the alignment being a power of 2 which is a very small subset of
sizeof(void *) multiples.
Then move the assertion to common code for wider coverage.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Moved the grove light sensor into a generic light sensor sample.
Replaced board specific node name for light sensor with an alias.
Refactored project name to reflect generic light sensor.
Signed-off-by: Dipak Shetty <shetty.dipak@gmx.com>
The vendor prefix has been changed from 'ltr' to 'liteon'.
This change is made to better reflect the actual name of
the vendor, which is LiteOn. The previous prefix 'ltr'
was referring to the LiteOn Optical Sensor family, which
could lead to misunderstandings.
Signed-off-by: Konrad Sikora <kontakt@konradsikora.pl>
RAM power configuration is preserved through soft reset,
meaning that there is a risk of accessing powered off RAM blocks
when booting in different application (i.e. bootloader).
Add option to force all RAM blocks to be powered on
before triggering soft reset to prevent this from happening.
Signed-off-by: Nikodem Kastelik <nikodem.kastelik@nordicsemi.no>
We can't do synchronous HCI command sending in any settings commit()
callback, since we don't know what context the callback is being called
from. One particular deadlock can happen if settings_load() is called from
the preemptible main thread:
main()
|--> settings_load() (aquire settings_lock mutex)
|-->commit callback A that defers to system wq
|-->commit callback B that calls bt_hci_cmd_send_sync()
system wq from the previous deferral from within settings_load():
|--> work item
|--> settings_save_one()
|--> attempt to aquire settings_lock mutex
In the above scenario, the bt_hci_cmd_send_sync() call from the main thread
depends on the system workqueue being processed (since that's what does HCI
command processing by default), while at the same time holding the settings
subsystem's mutex. At the same time, a system wq item tries to store
something into settings, however it deadlocks waiting for the settings
mutex.
The actual scenario that we have in the Bluetooth subsystem is where
"commit callback A" is commit_settings() in host/settings.c, and "commit
callback B" is keys_commit() in host/keys.c.
The solution to the deadlock is to take advantage of deferred bt_id_add()
handling which already exists, i.e. set a flag and deferre the actual
adding to the system workqueue.
Signed-off-by: Johan Hedberg <johan.hedberg@silabs.com>
Sending an HCI message requires to wake up the controller
core if it's asleep. Platform controllers may send responses using
non wakeable interrupts and those can be lost during standby usage.
Blocking standby usage until the HCI message is sent.
Signed-off-by: Yassine El Aissaoui <yassine.elaissaoui@nxp.com>
Add some extra ifdefs around MQTT 5.0 specific structures to prevent
compiler errors due to undefined symbols in certain configurations.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The CS complete callbacks provide both status and params.
In the case of errors, NULL pointer is passed to the params of callbacks.
Signed-off-by: Ryan Chu <ryan.chu@nordicsemi.no>
If the HCI status of a complete event is not BT_HCI_ERR_SUCCESS,
the remaining parameters could be invalid.
In this case, the params is passed as NULL pointer to the callbacks.
- LE CS Read Remote Supported Capabilities Complete event
- LE CS Read Remote FAE Table Complete event
- LE CS Config Complete event
- LE CS Security Enable Complete event
- LE CS Procedure Enable Complete event
This change avoids forwarding the invalid fileds to the applications.
Signed-off-by: Ryan Chu <ryan.chu@nordicsemi.no>
String location information should only be sent for core which
do not append strings to the log message (PPR, FLPR). Without
this, cpurad was also sending that information and that was
redundant.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When decoding logs from a remote core with memory that APP can
access, wrong address of an array with string addresses was used.
Log message contains index of a string and APP strings array was
used instead of remote core. Extend STMESP logging so that address
of string array of a remote core is send during startup to the APP
and APP is using this array to decode strings from remote cores.
Bug applies only to PPR and FLPR as APP has no access to RAD memory.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
As there is no currently known board that has the cadence spi,
add a test to just build it under the qemu cortex m3.
Signed-off-by: Ryan McClelland <ryanmcclelland@meta.com>
Replace manually authored hardware features table with the new Zephyr
board supported hardware directive which automatically generates an
up-to-date table based on the boards' Devicetree.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
This is to fix ci failure caused by commit d5d4d57
In the above commit, it's wrongly assumed that ROM_ADDR and RAM_ADDR
are same **only** when CONFIG_XIP is not defined.
Since aarch64 was taken as reference in the review,
we missed to see that CONFIG_FLASH_SIZE is also part of the condition
to define ROM_ADDR as RAM_ADDR.
CONFIG_XIP is not defined for qemu_cortex_a9/xc7z007s however,
CONFIG_FLASH_SIZE is not 0 so the assumption fails
and the required reset to RAM_ADDR is not applied leading to the test
failure.
Signed-off-by: Sudan Landge <sudan.landge@arm.com>
By default the thread priority is still the lowest application thread
priority. Allow the priority to be increase by specifying both, the
`override` option as well as the desired `priority`. This pattern was taken
from the shell thread.
Signed-off-by: Cla Mattia Galliard <cla-mattia.galliard@zuehlke.com>
Add an example of how to initialize gbias. The real application should
restore latest valid values taken from non-volatile memory.
Signed-off-by: Armando Visconti <armando.visconti@st.com>
Use SENSOR_ATTR_OFFSET attribute to initialize the Sensor Fusion GBIAS.
This is useful when the device is not sill (e.g. hold in hand) and so
the gyroscope calibration does not run.
Signed-off-by: Armando Visconti <armando.visconti@st.com>
Move gyro_freq setting inside lsm6dsv16x_gyro_set_odr_raw() routine,
so that it is properly saved in all cases (i.e. from DT configuration
and from runtime ATTR set case).
Signed-off-by: Armando Visconti <armando.visconti@st.com>
When building with clang, it warns:
tests/subsys/usb/bos/src/test_bos.c:24:22: error: variable
'dummy_descriptor' is not needed and will not be emitted
[-Werror,-Wunneeded-internal-declaration]
static const uint8_t dummy_descriptor[] = {
^
Only the size of dummy_descriptor is used, not the contents. Replace the
struct with a constant for the size.
Signed-off-by: Tom Hughes <tomhughes@chromium.org>
The least significant bit of the first octet of a MAC address is a
unicast/multicast bit. The bit should be cleared when generating a
random link address for a virtual interface. Any frames from such
interface/address will be dropped by the network as invalid if
the multicast bit is set.
Signed-off-by: Jack Chistyakov <jack.chistyakov@outlook.com>
