Add point about moving maximum allocation size from fixed buffer pool's
internal structure to a common one.
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
This commit introduces two changes:
- veryfing fixed buffer's size and length after allocation to existing
test case
- new test veryfing appending bytes to a network buffer from a fixed
size pool
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
Previously, there was no way to determine maximum number of bytes
that can be allocated using only net_buf structure. This commit
introduces such field.
Moreover, this commit fixes an issue where allocation of less than
maximum number of bytes from a fixed buffer pool would set buffer's
size to this number instead of the whole buffer size.
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
From change 47eb592c28b9e7dfbdd25fedbf07a528ad240084 net_buf structures
allocated from pools defined with NET_BUF_POOL_FIXED_DEFINE() will keep
their `size` member set to the actual fixed size and not requested
size like before.
For this set of tests NET_BUF_POOL_FIXED_DEFINE() for several test
cases. These tests check the several values based on a fact that the
network buffer's size was set to the requested size.
This commit changes definition of the buffer pool to
NET_BUF_POOL_FIXED_DEFINE() in order to satisfy these expectations.
This change does not impact the tests themselves.
Signed-off-by: Konrad Derda <konrad.derda@nordicsemi.no>
Whenever UART_RX_DISABLED event is received module attempts to
re-enable receiver since in interrupt driven API receiver is
always on. However, it is possible that there is no free buffers
and in that case attempt to enable the receiver will fail with
-EBUSY. That scenario shall be accepted since the receiver will
be re-enabled when at least one buffer will be freed.
Given that, -EBUSY return shall be accepted (no assert) and
number of pending RX buffer requests shall be reset only on
successful enabling.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
When CONFIG_LOG_USE_TAGGED_ARGUMENTS is enabled, and
CONFIG_CBPRINTF_COMPLETE is also enabled, we should not be
blindly skipping tags when processing the tagged package
for output. The issue is that if there is a "%%" in
the format string, the specifier is considered invalid but
the code blindly skips ahead in the argument list as if
it is a valid specifier (think "%s"), which resulting in
the next valid specifier using incorrect argument in
the list. So fix it by skipping ahead if and only if
the specifier is not invalid.
Fixes#68271
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Move the check of PM_DEVICE_FLAG_RUNTIME_ENABLED to the beginning of
the function. With this we avoid taking/release the semaphore and also
we no longer need check it in runtime_enable_sync(), because it was
already checked in pm_device_runtime_enable().
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move around SYS_PORT_FUNC_ENTER and change a early return
in a way that when this function is called it will generate
tracing for entering / leaving this function.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move the check of PM_DEVICE_FLAG_RUNTIME_ENABLED to the beginning of
the function. With this we avoid taking/release the semaphore and also
we no longer need check it in runtime_disable_sync(), because it was
already checked in pm_device_runtime_disable().
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move around SYS_PORT_FUNC_ENTER and change a early return
in a way that when this function is called it will generate
tracing for entering / leaving this function.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Since we may go from enabling to qos-configured state if the
call to bt_iso_big_create fails, that is a valid transition.
If bt_iso_big_create currently fails, then the source ends in a
state where it cannot be recovered.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
When CONFIG_BT_ISO_TEST_PARAMS is enabled then the
bt_bap_broadcast_source_param in
bt_cap_initiator_broadcast_audio_create had uninitialized values.
A general and future proof solution for this is to simply initialize
the entire struct to 0.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Include the full version in the output of the "kernel version" shell
subcommand. Previously, EXTRAVERSION was not included, causing
e.g. "3.6.0-rc1" to be printed as "3.6.0".
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
return value of pwm_rpi_get_clkdiv contains implicit conversions
from 'float' to 'double', triggered by floating-point operations
involving mixed data types.
Signed-off-by: Jannis Ruellmann <j.ruellmann@kunbus.com>
Modifies two log statements so that it is clear whether it
is failing to validate broadcast or unicast PDU sizes.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Some applications turn logging off. It makes impossible to get information
about exceptions if it occures.
This PR restores correct behavior of Error_Dump messages, and
exception dump is printed anyway independent on CONFIG_LOG.
ARC_EXCEPTION_DEBUG added to default boards configs in addition to nsim.
CONFIG_FAULT_DUMP removed from ARC branch as it has been unused
since v1.8.0
Signed-off-by: Nikolay Agishev <agishev@synopsys.com>
Update cmsis-dsp revision to contain the fixes for:
- Buffer overflow in certain implementation of arm_conv_q31, q15 and q7
- Name conflicts with Zephyr macros SQ and ROUND_UP
Signed-off-by: Wilfried Chauveau <wilfried.chauveau@arm.com>
Define the reg and size property for the stm32 disco kits
which have an octospi instance
Refer to the dts/bindings/flash_controller/st,stm32-ospi-nor.yaml.
Also remove the <size> property for the stm362h750 disco kit.
Signed-off-by: Francois Ramu <francois.ramu@st.com>
This commit adds documentation about added functionality
for now interrupt tables parser and Link Time Optimisation.
Signed-off-by: Radosław Koppel <radoslaw.koppel@nordicsemi.no>
Improves the broadcast assistant shell documentation
which updates the existing documentation and adds
documentation to commands not yet documented.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Modify the broadcast assistant commands that take the PA sync argument
so that it is treated as a bool. This allows users to use
"true", "on", "1" or "0x01" for true values and "false", "off", "0"
for false values.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Advertising data has to be truncated at a PDU boundary
Similar to earlier fix for extended advertising reports
Found in several EBQ tests, including LL/DDI/SCN/BV-25-C,
LL/DDI/SCN/BV-21-C and LL/DDI/SCN/BV-46-C
Signed-off-by: Troels Nilsson <trnn@demant.com>
The SCO cannot be created if the eSCO is unsupported.
It is caused by the packet type cannot be set
correctly.
Set valid bitmap for SCO and eSCO packet type in
command "Accept Synchronous Connection Request".
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
In the update_recv_state_base function, we should update the
BASE with any new metadata that we have received (as that may
change while streaming). However, this function did not properly
keep the existing bis_sync values for the
subgroups, causing the bis_sync for all subgroups to be 0.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The receive state may be added by the broadcast sink if not added
by the application, but even in that case when the broadcast sink
is deleted, we should not remove the receive state, as the receive
state may still container information about the PA sync that has a
lifetime not coupled with the broadcast sink.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
If the receive state is locally removed, then we should not
request permission from the application, as that is implicit
when removing the source locally.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The test is configured to replace the config rather than extending,
resulting in a build with CONFIG_DISPLAY=n and the driver not being
actually built.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add a short migration guide for the LC3 macro renames.
Only a few examples are mentioned, as there are far too
many changes to describe them all.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The assigned numbers are not specific to LC3, so they should be
defined in the more general audio.h header file.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The defines from the Bluetooth Audio assigned numbers are not
specific for LC3. This commit removes the LC3 infix and
the reference to LC3 in the documentation for them.
It also cleans up some of the documentation and
names of the enums.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The BT_HCI_CODING_FORMAT_LC3 is defined in hci_types.h and
applications should use that value.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This change mitigates the following cmake warning
No SOURCES given to Zephyr library: lib__posix__options
without needing to have a separate interface library
(as we do not need private headers exposed).
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
Commit b2eaa6448076 ("drivers: dma: intel-adsp-hda: add delay to stop
host dma") added a wait on GBUSY state to host DMA stop.
This is problematic as in some case (like SOF chain-DMA usage),
the host DMA side RUN bit is not cleared when intel_adsp_hda_dma_stop()
is called. It is not possible to wait on GBUSY bit as there are
valid cases where it can remain set.
Address the original problem described in SOF bug #8686 and add a
polling check for intel_adsp_hda_is_enabled(). As per the bug
description, in some cases the GEN/FIFORDY bits are not cleared
immediately and if a new call to intel_adsp_hda_dma_stop() is made, the
PM refcounting will go haywire.
Link: https://github.com/thesofproject/sof/issues/8686
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
The CST816S chip ID have an alternative value. It seems that this
field represents in fact a version number of controller. Fix by adding
the new chip ID.
Signed-off-by: Joel Guittet <joelguittet@gmail.com>
With `CONFIG_XIP=y`, this linker script would derive the ROM region from
the chosen `zephyr,flash` DT node with "soc-nv-flash" or "jedec,spi-nor"
as its compatible. If the node was absent or had a different compatible,
then linking would fail with:
undefined symbol `ROM_BASE' referenced in expression
Fix this by using `CONFIG_FLASH_BASE_ADDRESS` and `CONFIG_FLASH_SIZE`
for ROM base and size respectively. The existing DT logic is preserved
for compatibility with out-of-tree boards, so the flash Kconfigs serve
as a mere fallback.
In addition, use `CONFIG_FLASH_LOAD_OFFSET` and `CONFIG_FLASH_LOAD_SIZE`
if defined, to align with some other architectures' linker scripts. For
the existing in-tree RISC-V boards, this should not make a difference.
The alternative would've been making sure that all boards and SoCs have
the relevant Kconfigs set, and only using those in the linker script.
The downside is that `CONFIG_FLASH_SIZE` is given in units of 1 KiB,
while some existing boards - hifive1_revb, sparkfun_red_v_things_plus -
have more granular flash sizes, which would've been rounded down.
Signed-off-by: Grzegorz Swiderski <grzegorz.swiderski@nordicsemi.no>
Fix coverity integer handling issue (CWE-188).
Modifying a variable through a pointer of an incompatible type (other
than unsigned char) can lead to unpredictable results.
Fix: #67965
Coverity-CID: 248434
Signed-off-by: Armando Visconti <armando.visconti@st.com>
Some places VCP accessed AICS and VOCS when it was not supported.
Also modify existing guards to be consistent with new guards.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The VCP and MICP instances should use their respective
AICS and VOCS Kconfig options, rather than the overall
Kconfig options.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
This fix removes the zephyr/ prefix from linker included files.
With this prefix the build works only for Ninja and not for
other build tools.
Linking in Zephyr / CMake:
- Ninja invokes linking directly from <build>.
- Make invokes linking form <build>/zephyr.
The linker default uses cwd for looking up INCLUDE directives if not found
in list of includes.
Zephyr always adds <build>/zephyr as link include using CMake,
and this is passed to ld as -L<build>/zephyr therefore using
INCLUDE isr_tables_swi.ld ensures it will be correctly found in all cases.
Signed-off-by: Radosław Koppel <radoslaw.koppel@nordicsemi.no>
This commit removes the need of swi_tables.ld file if the
ISR table generator is not configured to use it.
Signed-off-by: Radosław Koppel <radoslaw.koppel@nordicsemi.no>
