GitLab

Rebase fixup for the changes introdused in the
commit 42c1f75f


("usb: Rename usb_cfg_data to usb_class_data.")
Signed-off-by: Johann Fischer <j.fischer@phytec.de>

Apart from the fixed standard request handler,
custom, class, and vendor handlers are handled
on the basis of class/functions.
Handlers array and its initialization is no longer necessary.
Signed-off-by: Johann Fischer <j.fischer@phytec.de>

The name 'usb_cfg_data' is legacy of initial implementation
of the USB device stack. This structure was intended to be used to
configure the device in run time by calling usb_set_config().

Because of how things grown this was used to represent the data
of classes rather than the device as a whole. One device may consist of
many classes (composite device). When the stack was initially
developed this was not considered.

Alongside with renaming usb_cfg_data -> usb_class_data, all relevant
functions/macros/variables are updated accordingly.
Signed-off-by: Emil Obalski <emil.obalski@nordicsemi.no>

Rework the way how the interfaces are handled by the USB core stack.
Until now USB stack was aware of only the first interface associated
with each class. This led to a problem when not first interface was
addressed by the request.

This patch allow to filter out the request at core level.

The usb_cfg_data struct is legacy of initial implementation
of USB device stack. Initially the USB device stack was
implemented to support single class devices with
no composite support. This structure was intended to be used to
configure the device in run time by calling usb_set_config().

Because of how things grow the structure was used to represent
the classes instead of devices.

This patch removes device configuration descriptor pointer.
Adds pointer to table of USB interface containers associated with
he function it represents. Each container consist of pointer to
interface descriptor, its alternate setting (if exists) and currently
selected alternate setting. This help USB core stack to properly address
SET/GET interface requests.

This patch should be followed by renaming the structure to something
closer to 'class' specific data as described in USB 2.0 spec.

This patch is first step to rework the USB core stack and is not meant
to be considered as final form of USB core.
Signed-off-by: Emil Obalski <emil.obalski@nordicsemi.no>

This patch cleans webusb class. WebUSB class were using custom
request handler structure which is not necessary. This one
is removed and API struct is used instead.
Signed-off-by: Emil Obalski <emil.obalski@nordicsemi.no>

Rename USB usb_cfg_data structures to better reflect its meaning.

interface -> request_handlers
    Name interface was used here as a communication interface
    between USB core stack and Class implementations.
    This naming could be missleading as USB specification defines
    interface as a part of the USB device descriptor.

endpoint -> endpoints
    This change was made to reflect that there might be more than one
    endpoint.
Signed-off-by: Emil Obalski <emil.obalski@nordicsemi.no>

usb_cfg_data data structure is handling usb funciton specifig data.
USB device descriptor is single instance located in special RAM section.

This deletion is a step forward to make core stack aware of classes
it handles.
Signed-off-by: Emil Obalski <emil.obalski@nordicsemi.no>

Add for various flash runners (pyocd, nrfjprog, and dfu-util) to do a
full chip erase.  This is needed for the sample to pass in sanitycheck
on real hardware.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>

Currently '--coverage-tool gcovr' results in using lcov. Fix that to use
gcovr, as requested by user.

Fixes: f6462a3a

 ("sanitycheck: get rid of global VERBOSE")
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>

For platforms using non-native stacks, net_tcp_init() should be
compiled out, similar to how it is done in tcp_internal.h.

Fixes #27463
Signed-off-by: Vincent Wan <vwan@ti.com>

Added shell support to create and delete a periodic adv sync
object.
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>

Added support for syncing to periodic advetisements.
The API and usage is heavily inspired by the extended advertisement
and connection APIs.
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>

Added the ext adv report interval to bt_le_scan_recv_info which
denotes whether or not there is periodic advertisement data
associated with the advertisement report.
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>

Added shell support for periodic advertisement params,
data and enable.
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>

Added API to set periodic adv parameters, periodic adv data
and to enable/disable periodic advertisement.
Signed-off-by: Emil Gydesen <emil_gydesen@bose.com>

When FAT FS option is enabled, then the FAT FS sub-menu will appear
next to it, instead of appearing far down the list.
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>

New priority value check was faultly done on zephyr priority instead
of posix one.
Signed-off-by: Jan Pohanka <xhpohanka@gmail.com>

Add tests that will make sure that refcounting works as expected.

Also fixed two tests that were using the object values before
the fdtable contained proper values. After reserving a fd by
z_reserve_fd(), we can only call z_finalize_fd() or z_free_fd(),
and only after those calls can fetch obj and vtable values.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>

Use proper refcounting instead of magic value in obj field
when checking whether the fd is still in use. This will make
sure that if fd is shared between two threads, we do not
release it too soon.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>

Because the H1 heading was not set in mikroe_eth_click index.rst
file, the shields listing page was listing H2 headings for this
shield.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>

This file is setting Kconfig options even when it is not the chosen
SoC. I noticed this because without this patch, CONFIG_SOC_GECKO_EMU=y
when building for an unrelated board with SYS_POWER_MANAGEMENT=y.

Hide any subtrees in this file when the EXX32 family isn't selected.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>

This patch enables the rtc so that the testcase
tests/drivers/counter/counter_basic_api
can run on this nucleo_l152re board
also when running sanity checks patch set
Signed-off-by: Francois Ramu <francois.ramu@st.com>

The rtc counter is also built for the stm32l1 series
from STMicroelectronics.
The EXTI line 17 is mapped on RTC alarm pin
Signed-off-by: Francois Ramu <francois.ramu@st.com>

nucleo_l4r5zi add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

nucleo_g474re add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

stm32f3_disco add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

nucleo_f103rb add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

nucleo_f746zg add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

nucleo_f429zi add spi loopback support
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

Add fixture spi_loopback.
Signed-off-by: Alexandre Bourdiol <alexandre.bourdiol@st.com>

It does not auto-initiate the DLE procedure if the default data length
parameters are not equal to the initial parameters.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>

Some controllers do not follow the recommendation of auto-initiating
a data length update if the host writes default data length parameters.

If the controller follows this recommendation, calling
LE Write Suggested Default Data Length command is sufficient
to ensure that the LL will auto-initiate the DLE procedure. Performing
a second procedure is unecessary.

In Core v5.2, Vol 4, Part E, Section 7.8.35:
The HCI_LE_Write_Suggested_Default_Data_Length command allows the
Host to specify its suggested values for the Controller's maximum
transmission number of payload octets and maximum packet
transmission time for packets containing LL Data PDUs to be used
for new connections. The Controller may use smaller or larger values
for connInitialMaxTxOctets and connInitialMaxTxTime based on local
information.

In short, this command sets connInitialMaxTxOctets
and connInitialMaxTxTime.

In Core v5.2, Vol 6, Part B, Section 4.5.10:

For a new connection: connMaxTxOctets shall be set to
connInitialMaxTxOctets and connMaxRxOctets shall be chosen by the
Controller. If either value is not 27 then the Controller should
initiate the Data Length Update Procedure at the earliest practical
opportunity.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>

Document the behavior of bt_le_ext_adv_set_data ignoring either
advertising or scan response data when the configuration does
not support this type of data.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>

This commit updates the release notes with the changes introduced
in #26715.

It also informs readers that existing use of `$(SOC_DIR)` in Kconfig
must be updated, for example to use `rsource` instead of
`source $(SOC_DIR)` in order to adopt to latest changes.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>

This commit updates the description of SOC_ROOT according to the new
multiple SOC_ROOT enhancement.
Signed-off-by: Torsten Rasmussen <Torsten.Rasmussen@nordicsemi.no>

Fix issue with bt_gatt_service_unregister not clearing CCC information.

If the service is unregistered while an unbonded peer is connected and
subscribed then this CCC would be restored again when registering the
service even if the peer is now disconnected.

If the service is unregistered with bonded peers CCCs stored in settings
then this value would never be cleared. If the service is registered
again then the peer would receive a service changed for this service,
but it's CCC value would still be enabled.

Fixes: #26924
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>

Extract enqueue of GATT CCC when CONFIG_BT_SETTINGS_CCC_STORE_ON_WRITE
is enabled to a helper function gatt_ccc_conn_enqueue so that it can be
reused.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>

Document need to register services that are always registered as dynamic
services with CCCs loaded from settings.
These services must be registered before calling settings_load.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>

Move the clear_ccc_cfg function further up to be re-used without a
forward declaration in bt_gatt_service_unregister.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>

This configuration is used to describe what board should
be associated with the CPUNET domain.
Signed-off-by: Håkon Øye Amundsen <haakon.amundsen@nordicsemi.no>