When building an LLEXT-enabled kernel, 62b19ef65c added weak aliases
of all syscall implementation functions to a pointer to NULL, with the
assumption that LLEXT would check the required symbols at link time and
fail if any of them were found.
This check, however, is ineffective in the current implementation: the
actual address that is exported is the rather normal-looking location of
the variable containing the NULL pointer. This defeats the NULL symbol
validity checks in llext_link.c and causes the extension to crash at
runtime by jumping to a location containing a few zeroes in read-only
data memory.
This commit makes sure the alias target is actually placed at address 0
using the llext-sections.ld linker fragment, so that undefined syscall
implementations are exported as NULLs and as such properly flagged at
link time.
The test for this functionality is also updated to reflect the change.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
There are many places where source_id is retrieved and it
depends on runtime filtering being enabled. So far it was
all exposed but lets encapsulate that into a helper function.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
The host-based adv auto-resume function has both a problematic
implementation and disagreement in the community around how it should
behave. See the issue linked resolved below for details.
This patch makes the deprecation visible to the user. The user will be
better served by a auto-resume tailored their applications use case,
based on more primitive host API like `conn_cb.recycled`, which has
obvious behavior that is unlikely to change.
Resolves: https://github.com/zephyrproject-rtos/zephyr/issues/72567
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Provide a generic flash_copy() algorithm that is capable of
copying from one flash device to another or within different
regions of the same flash device.
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
Adding the macros `RA_CGC_CLK_SRC` and `RA_CGC_CLK_DIV` that derive
the BSP clock settings from the DeviceTree node settings.
I also define some aliases to fill in the gaps with the BSP
naming conventions.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
Like some other string properties, I will add a derived form
to FULL_NAME to make it easier to reference from macros.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
Extend the `coap_transmission_parameters` struct with the field
`ack_random_percent`. This was the last remaining CoAP transmission
parameter that was not configurable at runtime.
Signed-off-by: Adrian Friedli <adrian.friedli@husqvarnagroup.com>
Correct several inconsistent parameter names in the following functions:
- stm32_gpio_intc_select_line_trigger: rename `trigger` to `trg`
to match the header file.
- stm32_gpio_intc_set_irq_callback: rename the callback argument to `user`
to match the `stm32_gpio_irq_cb_t` type.
- stm32_exti_get_line_src_port: rename `pin` to `line` to align with
the Doxygen comment and implementation.
Signed-off-by: Pisit Sawangvonganan <pisit@ndrsolution.com>
In case WPA supplicant disabled DPP, we need to compile out the
corresponding DPP code in Wi-Fi shell too.
Signed-off-by: Chaitanya Tata <Chaitanya.Tata@nordicsemi.no>
PR #78645 replaced uses of the GNU keyword `__thread` with a new macro
`Z_THREAD_LOCAL` which expands to correspond C/C++ standard keyword if
applicable, else it falls back to `__thread`.
This PR addresses some missed replacements in headers files for #78645.
Signed-off-by: Daniel Flodin <daniel.flodin@iar.com>
Introduce support for 'uint64_t' type, so that unsigned numbers
can be serialized into JSON payloads.
Signed-off-by: Mykhailo Lohvynenko <Mykhailo_Lohvynenko@epam.com>
When CONFIG_LLEXT_EXPORT_DEVICES is not enabled, the DEVICE_DT_DEFINE
macro ends with the contents of the Z_DEVICE_DEFINE macro, which always
includes a terminating semicolon.
When CONFIG_LLEXT_EXPORT_DEVICES is enabled, the macro ends with
Z_DEVICE_EXPORT, which does _not_ include a terminating semicolon. This
leads to a syntax error in places where the DEVICE_DT_DEFINE macro is
used without it.
Fix this by adjusting the added code when CONFIG_LLEXT_EXPORT_DEVICES is
enabled to include a semicolon after the Z_DEVICE_EXPORT macro; also use
the opportunity to remove a stray backslash.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
Makes it easier to make sure the CS test code is removed if desired.
It should hopefully allow for a clean split in general.
Signed-off-by: Olivier Lesage <olivier.lesage@nordicsemi.no>
Adds support for:
- LE CS Subevent Result event
- LE CS Test End Complete event
For now, recombination of subevent results with more steps than could
fit within a single HCI event is not supported, and such events
are discarded.
Signed-off-by: Olivier Lesage <olivier.lesage@nordicsemi.no>
The QoS structure is not related to a codec, but rather a
stream, and should thus not use the "Codec" name.
The BAP and ASCS specs refer to the QoS as
"QoS configuration" several places, so it is an obvious
choice for a name.
Since the structure is defined and used by BAP, the prefix
was changed from bt_audio to bt_bap.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This commit introduces architecture-specific ELF relocations for RISC-V,
in accordance with the RISC-V PSABI specification:
https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-elf.adoc
Also, the necessary compiler configurations for compiling LLEXT
extensions on RISC-V are added, and the llext tests are executed on
RISC-V targets.
Calling llext extensions from user threads in RISC-V is still
unsupported as of this commit.
Signed-off-by: Eric Ackermann <eric.ackermann@cispa.de>
Use security mode (PSK or X509) to detect if we should
set the socket option to verify hostname.
PSK security mode cannot verify hostnames as this information
is coming in the certificate, so don't set the options.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
In case NET_NATIVE is disabled, certain network stack components do not
need to be compiled. Otherwise, they could throw errors if
--no-gc-sections compiler options is enabled.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Add SYS_SEM_LOCK() and SYS_SEM_LOCK_BREAK which are mostly the
same as K_SPINLOCK() and K_SPINLOCK_BREAK.
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
We already have `DT_HAS_COMPAT_STATUS_OKAY` and
`DT_NUM_INST_STATUS_OKAY`, it seems intuitive to assume that
`DT_NODE_HAS_STATUS_OKAY` exists, so much so that it was used
before it's implemented.
This patch implements `DT_NODE_HAS_STATUS_OKAY`, which is
equivalent to: `DT_NODE_HAS_STATUS(<node_id>, okay)`
Added test for it in `tests/lib/devicetree/api`
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
- Implement irq-set-affinity in RISCV PLIC.
- Added new affinity shell command to get/set the irq(s)
affinity in runtime, when `0` is sent as the `local_irq`, it
means set/get all IRQs affinity.
- Some minor optimizations
Updated the build_all test to build this new configuration.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Removes duplicate declaration of the bridging direction, keeping the
one in the public header file.
Signed-off-by: Håvard Reierstad <haavard.reierstad@nordicsemi.no>
Adds the `bt_mesh_brg_cfg` prefix to the public Subnet Bridge API, and
aligns the function and callback naming with the rest of the Bluetooth
Mesh API.
Signed-off-by: Håvard Reierstad <haavard.reierstad@nordicsemi.no>
This patch renames the identifiers for the structs in the LLEXT symbol
tables to have a prefix of "__llext_sym_". This is done so existing
scripts like gen_device_deps.py do not confuse them with the object
they are referring to.
Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
`DT_..._REG_ADDR` macros do not always return an unsigned value. This
commit adds and unsigned 32bit prefix to ensure the value is always
unsigned.
Signed-off-by: Jeppe Odgaard <jeppe.odgaard@prevas.dk>
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
As this is a header, let's explicitly initialize all members of the
struct to avoid a "missing initializer for member" warning in case the
file is built with -Wmissing-field-initializers, or worse, that warning
set to be an error.
Signed-off-by: Alberto Escolar Piedras <alberto.escolar.piedras@nordicsemi.no>
Unlike a normal IRQ, a multilevel IRQ can't be incremented by
simply `irq++`, as that would always increment the L1 of a IRQ,
regardless of its level. A function that understands the level
for which the IRQ operates in is required.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Use a union struct to partition the bits in a
multilevel-encoded interrupt number instead of manipulating
the bits manually, which is tedious and error prone.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
When `CONFIG_MULTI_LEVEL_INTERRUPTS` is enabled, bits in a
`uint32_t` is always partitioned into 3 levels, we can safely
remove the `CONFIG_*_LEVEL_INTERRUPTS` checks so that the
APIs always works.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
The IRQ for level 1 and above is incremented by 1 when encoded
with Zephyr's multilevel IRQ scheme, so it should be
decremented by 1.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
The `resolving` is used to flag the status that the stack is requesting
remote device name.
It is better to use the type `bool` instead of the original type
`uint8_t`.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
Move `struct discovery_priv` to classic.h
Rename `struct discovery_priv` to `struct bt_br_discovery_priv`.
Modify the structure `struct bt_br_discovery_priv` with `@private`.
Change field `_priv` of `struct bt_br_discovery_result` from
`uint8_t _priv[4]` to `struct bt_br_discovery_priv _priv`.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
Remove BR discovery callback from bt_br_discovery_start.
All discovery results will be notified through callback
registered by bt_br_discovery_cb_register.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
The results of inquiry and extended inquiry
are only reported from application level
after the inquiry complete event notified.
While the event of inquiry result is notified
by controller in real time.
It is not a good user experience.
Just like scanning of LE, add a listener
cb for discovery.
When the event of inquiry result, extended
inquiry result, or remote name request
complete notified, call listener `recv`
cb to notify the upper layer.
When the event of inquiry complete
notified or no pending of remote name
request , call listener `timeout`
cb to notify the upper layer.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
A libmetal "generic" device is created, then opened and used for its IO
regions. Since we are adding these IO regions in the first place there
is no reason for this layer of indirection.
Signed-off-by: Andrew Davis <afd@ti.com>
Update the documented value range that `uart_irq_tx_ready` returns in
order to provide more information to the callers about the number of
bytes that could be writted with `uart_fifo_fill` without fragmentation.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Move the `Z_THREAD_LOCAL` macro from
`include/zephyr/toolchain/common.h` to `include/zephyr/types.h`.
This is needed since some `syscall.h` files are using TLS but aren't
including `zephyr/toolchain.h` (but they are including
`zephyr/types.h`).
Signed-off-by: Daniel Flodin <daniel.flodin@iar.com>
