This commit fixes issue with undefined value during position calculations.
Shift for positions may be >32, so we need to explicitly use unsigned long
casting to prevent unexpected behavior during event handling.
Thanks @jgrall for suggestion.
Signed-off-by: Dmytro Firsov <dmytro_firsov@epam.com>
This commit fixes barrier usage in Xen event channel driver. Previously
compiler_barrier() was used and it did not prevent processor from
re-ordering of the write operations, that is needed for correct event
handling. It is now changed to data memory barrier (dmb()), which will
work as expected in this situation.
Thanks @jgrall for suggestion.
Signed-off-by: Dmytro Firsov <dmytro_firsov@epam.com>
This patch modifies the ADXL362 driver to use the new
SENSOR_TRIG_MOTION trigger for activity detection
and SENSOR_TRIG_STATIONARY for inactivity detection.
Signed-off-by: Maximilian Deubel <maximilian.deubel@nordicsemi.no>
This patch adds the new trigger types SENSOR_TRIG_MOTION
and SENSOR_TRIG_STATIONARY to distinguish these common
motion detection events.
Signed-off-by: Maximilian Deubel <maximilian.deubel@nordicsemi.no>
The interrupt names are now defined in the HW models.
As a result, all source files in the nrf52_bsim folder are now
free of nrf52832 specific references. Therefore these can in
theory be used for other models as well.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
Mostly includes minor maintainability updates:
- ci: Add building the HW models as a github workflow
- irq_sources: Remove unneeded irq_sources.h
- irq_ctrl: Define IRQ names inside HW models
- HW_models: NRF_RADIO: Set missing NRF_RADIO states
- HW_models: NRF_RADIO: Disable radio state while RXDISABLE
- HW_Models: NRF_PPI: Enable ppi task and event for TIMER0 CH3
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
The iso_rx_qos is only used for unicast audio, as
we do not set the RX QOS for the broadcast sink, nor
can a broadcast source set RX QOS.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Tagged arguments feature fails to correctly tag a char array in
C++ configuration. Temporarily casting to char pointer to make
the test pass.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Logging v1 has been removed and log_strdup wrapper function is no
longer needed. Removing the function and its use in the tree.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Pull in a fix in the nrfx_qspi driver for a problem with incorrect
status returned by `nrfx_qspi_mem_busy_check()`.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
Cocinelle complains about "shell" being used as a parameter name:
"Violation to rule 5.7 (Tag name should be unique) tag: shell"
Therefore rename the parameter throughout the sample to "sh".
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Rewrite the TCP/UDP download part of the zperf sample to use socket API.
For UDP, performance impact is negligible (< 1 Mbps), for TCP it's
noticable, but still throughputs can be considered satisfactory (up to
~75 Mbps).
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Rewrite TCP/UDP upload part of the zperf sample to use socket API
instead of net_context. This has a negligible impact on the upload
throughputs (< 1 Mbps).
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
So far, TCP cloned a packet with data on an RX path for the application,
leaving the original packet intact. This isn't really needed, as the
original packet is unconditionally freed later anyway, so the TCP can as
well simply queue the original packet for the application, while
informing the network processing core, that the packet was consumed by
the TCP layer.
This allows to improve the download throughput even further, since the
CPU don't waste time on needles packet copying.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
The current default of 250ms appears to be too low in case connection
teardown takes place in lossy networks - in case of FIN packet
retransmission, the connection on the Zephyr side could have already
been dismissed due to low TIME_WAIT state delay, resulting in ICMP
Destination Unreachable replies.
Increase the default value to 1500ms - this is still pretty low, but at
least gives the peer some time to retransmit the FIN packet.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In case TCP stack enters TIMED_WAIT state (after receiving FIN/ACK reply
from peer), it should stil be ready to reply with ACK for any
consecutive FIN attempts. Othewise, in case the final ACK from Zephyr
side is lost, the connection is not properly closed on the other end,
and peer keeps retransmitting the final FIN packet.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
Implement a mechanism, according to RFC 813, which allows to prevent so
called "Silly Window Syndrome" - a scenario where the TCP receiver keeps
reporting small window sizes in the acknowledgments, effectively
limiting the connection throughput. This allows to improve performance
in low-buffer configurations, where the maximum window size is small,
and the issue was hitting quite often.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In case NET_MGMT_EVENT module was enabled but w/o NET_MGMT_EVENT_INFO,
the OpenThread integration layer failed to build as the "info" field in
the net mgmt callback structure is not available then.
Fix this by conditionally enabling code processing the event only if
NET_MGMT_EVENT_INFO is enabled. Otherwise, print a warning, as the event
is not really useful if no address information is provided.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
This runs the existing kernel common tests using picolibc as some of those
depend on libc functionality.
Signed-off-by: Keith Packard <keithp@keithp.com>
For targets without thread local storage, we need to use the builtin
per-thread errno support provided by Zephyr as the multi-thread errno
support provided in picolibc relies on TLS.
Signed-off-by: Keith Packard <keithp@keithp.com>
log_Switch_format test has incomplete handler logs on frdm_k64f
and also a false positive for logging.log_switch_format.immediate
testcase due to harness console which does not wait for
completion of ztest testcases to indicate pass/fail status to twister.
Fixes#46368
Signed-off-by: Aastha Grover <aastha.grover@intel.com>
Option --runtime-artifact-cleanup/-M allows to clean build artifacts
during runtime to save space usage. However, if one wanted to combine
it with --build-only and then --test-only, they would get an error.
This commit expands the list of files stored with -M with files
required to made --test-only also to work afterwards. Such change can
be useful for setups where building is done on one node and then
minimal amount of artifacts are transfered to another one.
Signed-off-by: Maciej Perkowski <Maciej.Perkowski@nordicsemi.no>
Prevent local "addrlen_copy" variable from being used uninitialized in
accept() userspace verification function.
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
This adds a table to set the dma channel priority from the DTS
The config bits are converted through the table_priority
to match the DMA_Priority_level of the stm32cube HAL.
Fixes#46725
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Return the hardware mpid value directly. Guard against values that
wouldn't fit in the returned uint32_t. If that ever happens then
reverting to the Zephyr sequential CPU number should be considered.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Some STM32 based boards where missing the alternate node label zephyr_udc0,
leading to build issues when compiling usb samples.
Fixes#46626
Signed-off-by: Erwan Gouriou <erwan.gouriou@linaro.org>
A recent change to hal_rpi_pico's cmake, intended to fix the
compilation via Unix Makefiles, ended up not linking the BL
to the final ELF. This commit fixes both problems.
Signed-off-by: Yonatan Schachter <yonatan.schachter@gmail.com>
Whenever MBEDTLS_BUILTIN is selected then
CONFIG_MBEDTLS_CFG_FILE="config-tls-generic.h" is set as a default value.
It is even impossible to change it, as without CUSTOM_MBEDTLS_CFG_FILE it
is just a hidden (not configurable by user) Kconfig option.
Remove explicit configuration from prj.conf.
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Whenever MBEDTLS_BUILTIN is selected then
CONFIG_MBEDTLS_CFG_FILE="config-tls-generic.h" is set as a default value.
Remove explicit configuration from prj_mtls_shim.conf.
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Whenever MBEDTLS_BUILTIN is selected then
CONFIG_MBEDTLS_CFG_FILE="config-tls-generic.h" is set as a default value.
Remove explicit configuration from prj.conf.
Signed-off-by: Marcin Niestroj <m.niestroj@emb.dev>
Modified the samples such that the client will attempt to
setup 2 sink streams and 1 source stream, basically
representing a headset use case.
This also excercises both setting up a unidirectional
and a bidirectional ISO channel using the 3 audio streams.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The sample used `net_buf_ref` to send the same data multiple
times on multiple streams, but the send procedure, which
ends up calling bt_conn_send_cb, does not work
with multiple refs. We thus replace the refs with net_buf_clones.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
In the sent callbacks we used CONTAINER_OF to get the bt_audio_ep,
but that no longer has the ISO channel, causing these
CONTAINER_OF to return a wrong pointer.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
On X86 platforms, the interrupt trigger method has been
changed to use APIC IPI, we don't use INT command to trigger
interrupt, so remove this unused macro.
Signed-off-by: Chen Peng1 <peng1.chen@intel.com>
change the test interrupt line number to a bigger one, because
the low number usually will be used by other devices.
Signed-off-by: Chen Peng1 <peng1.chen@intel.com>
On X86 platforms, the interrupt trigger method has been changed
from using INT command to using APIC IPI, we need to make sure
the IPI interrupt is handled before do our check, so add some
nop operations.
Signed-off-by: Chen Peng1 <peng1.chen@intel.com>
`bt_conn_index` has been changed to take a `const bt_conn`
which allows our functions that use this to also take a
`const` parameter.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
the CSIS client now has a ordered access procedure function.
The function implements the procedure as per the CSIP spec,
where it will verify that 1 or more members are unlocked,
and the execture a procedure on each member in ascending
order of rank.
The procedure can be anything (even a non-BT procedure),
and it will be up to the applicaiton to implement
what to do in that.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Instead of getting doing look ups for each member
for the CSIS client procedures, we store the active
members ordered by rank.
This is slightly less efficient in cases of error,
but for some procedures as the upcoming Ordered
access procedure where we need to access each
member twice, ordering them once is more
efficient.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Having the members array being const causes a lot of
limitations in the implementatation, and was
occasionally even disregarded. Removed to make the
implementation more flexible.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
