Convert users of net_buf_put() and net_buf_get() functions to use
non-wrapped putters and getters k_fifo_put() and k_fifo_get().
Special handling of net_bufs in k_fifos is no longer needed after commit
3d306c181f, since these actions are now
atomic regardless of any net_buf fragments.
Signed-off-by: Henrik Brix Andersen <henrik@brixandersen.dk>
The driver can run in either native_posix or native_sim.
Let's make sure to set the dependencies acordingly.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
gcc 13 produces a build warning (see below), as it seems to
believe the number of read bytes may overflow the frame_size
type.
Let's increase the frame_size bitwidth to avoid this.
(Any 32bit type, signed or unsigned, avoids this warning)
The build warning:
```
In file included from /usr/include/features.h:502,
from bits/libc-header-start.h:33,
from /usr/include/stdint.h:26,
from include/stdint.h:9,
from zephyr/include/zephyr/types.h:11,
from zephyr/include/zephyr/kernel_includes.h:21,
from zephyr/include/zephyr/kernel.h:17,
from zephyr/drivers/bluetooth/hci/userchan.c:9:
In function ‘read’,
inlined from ‘rx_thread’ at drivers/bluetooth/hci/userchan.c:201:9:
/usr/include/i386-linux-gnu/bits/unistd.h:28:10: warning: ‘__read_alias’
specified size between 4294902273 and 4294967295 exceeds maximum object
size 2147483647
[-Wstringop-overflow=]
28 | return __glibc_fortify (read, __nbytes, sizeof (char),
| ^~~~~~~~~~~~~~~
drivers/bluetooth/hci/userchan.c: In function ‘rx_thread’:
drivers/bluetooth/hci/userchan.c:187:32: note: destination object allocated
here
187 | static uint8_t frame[512];
| ^~~~~
/usr/include/i386-linux-gnu/bits/unistd-decl.h:29:16: note: in a call to
function
‘__read_alias’ declared with attribute ‘access (write_only, 2, 3)’
29 | extern ssize_t __REDIRECT_FORTIFY (__read_alias, (int __fd, void
| ^~~~~~~~~~~~~~~~~~
```
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Remove the logic from the board level about overriding the
flash load size and instead just select to use the DT code
partition if app is not expected to be able to occupy the
whole flash space because of the firmware image being
loaded separately.
Signed-off-by: Declan Snyder <declan.snyder@nxp.com>
Enabling BT_HCI_SETUP for STM32WB55 to have a
correct and proper initialization procedure to fix
#75318 issue
Signed-off-by: Alessandro Manganaro <alessandro.manganaro@st.com>
Implementing HCI setup function to have a correct and proper
initialization procedure to fix#75318 issue
Signed-off-by: Alessandro Manganaro <alessandro.manganaro@st.com>
Added force unblock psoc6 bless rx thread to process controller
events (at the end of psoc6_bless_send function)
Internal case: SWINTEGRAT-1767
Signed-off-by: Nazar Palamar <nazar.palamar@infineon.com>
Problem:
re-plug USB (e.g cy8cproto_062_4343w kit) can cause that
CY43xx device become unresponsive (cy8cproto_062_4343w).
As result we catch ASSERTION FAIL (timeout) on HCI_Reset command.
Fix: added delay (BT_POWER_CBUCK_DISCHARGE_TIME_MS) to be sure
that BT CBUCK regulator to discharge.
Tested on:
cy8cproto_062_4343w, cy8ckit_062s2_ai and cy8ckit_062s2_43012 kits.
Signed-off-by: Nazar Palamar <nazar.palamar@infineon.com>
Add Kconfig.nxp to support NXP Bluetooth Chipset.
Current only NXP IW612 Chipset (BT_NXP_NW612) has
been supported.
Add modules/hal_nxp/bt_controller/CMakeLists.txt to
determine whether any firmware is selected, and
check whether the firmware exists.
If the firmware exists, copy the firmware to the
temporary folder ${ZEPHYR_BINARY_DIR}/include/
generated/bt_nxp_ctlr_fw.h. OR, raise a fatal error.
In file hci_nxp_setup.c, includes the temporary file
bt_nxp_ctlr_fw.h.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
Implement UART firmware download driver for NXP
BT module.
Only support Murata 2EL M.2 module on RT1170EVKB.
And only one instance can be supported now.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
Convert the slz_hci.c HCI driver to use the new HCI driver API. This also
fixes the HCI bus type to correctly indicate VIRTUAL instead of UART.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Convert the hci_nxp.c HCI driver to use the new HCI driver API. Also move
the driver binding under dts/bindings/bluetooth, like all other HCI driver
bindings.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Convert the hci_stm32wba.c driver to the new HCI API. Unlike in most cases,
the devicetree node is already enabled on the SoC level (rather than board
level). This is in order to mirror how the Kconfig option was originally
enabled, i.e. on the SoC level.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Convert the spi.c and hci_spi_st.c drivers to use the new HCI driver API.
Both drivers are converted in one go since one derives from the other's
devicetree binding.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
With the new HCI driver model it's not always critical if a single
instance fails to initialize. This is especially true for many test
applications which provide their own HCI drivers. Instead of causing a
complete failure, simply fail to initialize the driver instance.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
The drivers should be independent after the move to the new HCI driver
API. Having them as a choice also has unexpected consequences with some
drivers being unexpectedly enabled.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Update the native controller to the new HCI driver API. The devicetree
node is placed under existing `radio` nodes, which seemed like the most
intuitive option.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Convert the H4 driver to the new HCI driver API. This includes updating
also any boards that use the driver, i.e. adding the appropriate
devicetree node and chosen property to them.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
Add API to perform hardware reset optionally entering firmware updater
mode.
Remove redundant declaration for bt_spi_send_aci_config; otherwise, we will
have compiler warning if CONFIG_BT_BLUENRG_ACI is not set.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Some vendor specific setup was done inside
the open() HCI function, those should be
inside setup() function instead.
Signed-off-by: Yassine El Aissaoui <yassine.elaissaoui@nxp.com>
Before sending packet to controller the host needs to poll the status of
controller to know it's ready, or before reading packets from controller
the host needs to get the payload size of coming packets by sending
specific command and putting the status or size to the rx buffer, the CS
should be held at this moment to continue to send or receive packets.
This change is needed for the based SPI driver update.
Signed-off-by: Aaron Ye <aye@ambiq.com>
Keep reading from the HCI socket when a packet is incomplete. The other
end may not write entire packets, or TCP could fragment in the middle of a
packet.
Also fix a potential infinite loop by advancing to the next packet before
any continue statements.
Signed-off-by: Patrick Stewart <patrick@rfcreations.com>
The driver isn't currently buildable due to "west blobs" support never
having been added for hal_telink. Furthermore, even if the blob
dependency is manually made available it turns out the code has
bitrotten to the point where it doesn't build anymore. This situation
has continued for several years without anyone taking action, so I think
it's safe to assume this is unmaintained and should be removed.
Signed-off-by: Johan Hedberg <johan.hedberg@gmail.com>
The controller may be unavailable to receive packets because it is busy
on processing something or have packets to send to host. Need to free the
SPI bus and wait some moment to try again.
Signed-off-by: Aaron Ye <aye@ambiq.com>
The BLE controller of some Ambiq Apollox Blue SOC may have issue to
report the expected supported features bitmask successfully, thought the
features are actually supportive. Need to correct them before going to
the host stack.
Signed-off-by: Aaron Ye <aye@ambiq.com>
This commit add the SPI-based HCI support for the Ambiq Apollo3 Blue
SOC (e.g. Apollo3 Blue Plus, Apollo3 Blue) support.
Also correct the dependency of necessary peripheral.
Signed-off-by: Aaron Ye <aye@ambiq.com>
Add HCI driver generic to NXP platforms.
Update west.yml to have ble support for rw61x
Signed-off-by: Axel Le Bourhis <axel.lebourhis@nxp.com>
Signed-off-by: Yassine El Aissaoui <yassine.elaissaoui@nxp.com>
If rx buffer allocation has to be deferred to rx_thread, we need to stop
isr reading from mailbox as otherwise rx_thread won't be able to process
other buffers.
Since CMAC2SYS irq is cleared before data is read from mailbox, in case
rx buffer allocation was deferred we also need to trigger irq manually
to make sure all pending data is processed.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
This adds HCI driver which enables communication with CMAC core on
Renesas SmartBond DA1469x series. The CMAC core is running an Apache
NimBLE controller binary and uses shared memory for communcation via
mailboxes.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
The `BT_DEBUG_*` Kconfig symbols have been deprecated for more than 2
versions, remove them.
Update code that was still using them.
Remove the Bluetooth specific `Kconfig.template.log_config_bt` and use
`Kconfig.template.log_config_inherit` from the logging subsystem
instead, now that the legacy symbols can be removed.
Signed-off-by: Théo Battrel <theo.battrel@nordicsemi.no>
Introduced a unified definition for HCI packet type indicators in
'bluetooth/hci_types.h. This change streamlines the code in
'drivers/bluetooth/hci/', reducing redundancy.
Enhances maintainability and consistency across all HCI drivers.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
`bt_recv` is invoked from the BT long work queue, which is preemptible.
The host uses cooperative scheduling to ensure thread safety.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Provide support for reading events longer than 127 bytes in
ST SPI protocol v1 by splitting them into several SPI transactions.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
This config selects a variant of the HCI driver interface that spills
out host internals unto the drivers and even the Zephyr controller. It
will now be removed in favor of driver interfaces that hide the
internals of the host.
The new default is `CONFIG_BT_RECV_WORKQ_BT`.
Any references to the removed kconfig are refactored out.
Any out-of-tree driver using the removed interface can be easily adapted
by copying the following implementations into the driver as private
functions:
- `hci_driver.h:BT_HCI_EVT_FLAG_RECV_PRIO`
- `hci_driver.h:BT_HCI_EVT_FLAG_RECV`
- `hci_driver.h:bt_hci_evt_get_flags`
- `hci_raw.c:bt_recv_prio`
In combination these symbols function as a interface adapter. These
symbols will be removed in this PR in subsequent commits.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Use SPI_LOCK_ON to lock configuration for multiple transaction while CS is
kept low.
Change control of CS line from direct GPIO manipulation to SPI driver API.
Signed-off-by: Michele Sardo <msmttchr@gmail.com>
Use the nrf_reset_network_force_off() function to release and set the
force-off signal.This ensures that the workaround for errata 161 is
applied.
Signed-off-by: Vidar Berg <vidar.berg@nordicsemi.no>
Exclude devices based on ST SPI protocol v2 from special handling of
hci_reset opcode as it is redundant.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Add raw mode in ST HCI SPI BT driver to support host-less configuration.
Remove compilation dependency for BT_QUIRK_NO_RESET as it applies to all
configurations.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
In the case that there are no more ISO buffers left,
"No available ISO buffers" is logged. However, given the
nature of ISO where we are (very) likely to get additional ISO
very soon after (typically every 7.5 or 10ms for audio),
this will get logged a lot, and the logging may in some
cases actually prevent the application from handling and
freeing existng buffers due to the immense logging,
which may make this (minor) issue into a blocking issue.
This is fixed by reducing the logging to the first
occurence, and then only every 100 afterwards, which has
shown to reduce the risk of this effectively blocking
the application.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
A payload size of 0xFF is valid and should not be a reason to keep
looping for more headers. The complete reqiurement is now:
```
header_slave[STATUS_HEADER_READY] == READY_NOW &&
header_slave[STATUS_HEADER_TOREAD] > 0
```
This fixes events being dropped when the payload size is 255.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Provide multiple write transaction support for BlueNRG-MS, st_hci_spi_v1
protocol. Since by default, BlueNRG-MS write buffer supports up to 127
bytes; however, it is possible to have consecutive write transactions
so as to send data more than 127 bytes.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
This change renames the RX message buffer from g_hciRxMsg to rxmsg,
as it does not follow the Zephyr coding style.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
This change relocates the tx/rx message buffer from the .bss section
to the .noinit section. This adjustment is aimed at reducing boot time by
lowering the size of the .bss section, without impacting the operational
functionality of the buffer.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
Fix issue #66806, caused by PR #65172.
Return -EINVAL from bt_spi_get_header if op is neither SPI_READ nor
SPI_WRITE.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
This commits create the dts binding for Ambiq BT HCI instance.
And create the SPI based common HCI driver for Ambiq Apollox
Blue SoC and the extended soc driver for HCI.
Signed-off-by: Aaron Ye <aye@ambiq.com>
Remove BT_SPI_BLUENRG Kconfig parameter as it is redundant according to
the new changes introduced by ST SPI protocol V1 and V2.
Remove "config BT_SPI_BLUENRG" from the boards that were using it.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Introduce STMicroelectronics SPI protocol V2 which is used in BlueNRG-1
and successor devices.
Change the size of the variable "size" to 16 bits as it is necessary for ST
BlueNRG-1 and successor devices.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Replace the header acquisition scope in bt_spi_send with bt_spi_get_header.
Add WRITE_DATA_CONDITION in order not to retransmit data without reading
the header as it is meaningless for ST BlueNRG devices.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Update structure of spi.c to have a better and cleaner separation
between STMicroelectronics and Zephyr SPI protocol.
Introduce bt_spi_get_header to separate algorithms well for
header acquisition in different protocols.
Remove header acquisition from bt_spi_rx_thread to make it simpler.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
Enforcing the peer's behavior is not strictly necessary. All the host
should do is make sure it is resilient to a spec-violating peer.
Moreover, a growing number of platforms were disabling the check, as the
spec allows "batching" HCI num complete packets events, stalling ATT RX.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Co-authored-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
If no public address was set, it should just do nothing instead of
erroring. The function should also be static and there's no need to copy
the address struct.
Signed-off-by: Armin Brauns <armin.brauns@embedded-solutions.at>
STM32CubeWB v1.18.0 replaces Master with Central & Slave with Peripheral
in the file app_conf.h (modified in the commit updating lib/stm32wb)
Signed-off-by: Abderrahmane Jarmouni <abderrahmane.jarmouni-ext@st.com>
Make sure that events flagged as high priority are handled when
CONFIG_BT_RECV_BLOCKING is not defined.
Fix for #65892.
Signed-off-by: Michele Sardo <msmttchr@gmail.com>
The zephyr bluetooth stack expects the controller to know its public
address, if any. At least for BlueNRG, the public address is forgotten with
every reset/power cycle, so there needs to be a way to set it from within
zephyr. This is accomplished using the `Aci_Hal_Write_Config_Data` HCI
command, as described in PM0237.
Signed-off-by: Armin Brauns <armin.brauns@embedded-solutions.at>
"LL only" is not the only config option of potential interest, e.g. the
public address is also important.
Signed-off-by: Armin Brauns <armin.brauns@embedded-solutions.at>
This allows HCI drivers to expose vendor-specific functions to set the
public address.
Signed-off-by: Armin Brauns <armin.brauns@embedded-solutions.at>
Increase the SPI RX driver stack size by 128 bytes. Overflows have
previously been observed on real hardware at the default stack size of
512.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
The HCI receive path has a delay between reading the header and payload
from the controller to give the controller time to setup the SPI
peripheral for the next transaction. Add the same delay on the transmit
path for the same reasons.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Add log output when the HCI interface is forced to retransmit a packet
because the controller is not ready.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Modify SPI configuration to match the features introduced in PR #63437.
Set the property "controller-data-delay-us" to zero for boards
which are using BlueNRG-MS.
Fix Chip Select configuration for stm32l562e_dk board.
Signed-off-by: Ali Hozhabri <ali.hozhabri@st.com>
