Move the `REQUIRES_FULL_LIBC` dependency from `config LORA` to
`config LORAWAN`. The commit that added the `select` (f590d4fa) mentions
that this is required by `loramac-node`, which is only used by LoRaWAN,
not the base LoRa code.
This results in small FLASH savings when compiling the samples, but can
result in larger savings in more complex applications:
```
// With REQUIRES_FULL_LIBC
west build -b 96b_wistrio zephyr/samples/drivers/lora/send/
[162/162] Linking C executable zephyr/zephyr.elf
Memory region Used Size Region Size %age Used
FLASH: 37708 B 128 KB 28.77%
SRAM: 8832 B 32 KB 26.95%
IDT_LIST: 0 GB 2 KB 0.00%
// Without REQUIRES_FULL_LIBC
[181/181] Linking C executable zephyr/zephyr.elf
Memory region Used Size Region Size %age Used
FLASH: 37444 B 128 KB 28.57%
SRAM: 8832 B 32 KB 26.95%
IDT_LIST: 0 GB 2 KB 0.00%
```
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Add early returns from functions which are not used when
there is only one frontend in the system (no backends). This
allows to significantly reduce logging code size in that
configuration.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
Fix radio_tmr_start_now from returning delayed start time
under race conditions where the capture time has elapsed but
the compare has triggered at the same time.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This adds sending Active Index notification and Preset Changed
notifications on connection to previously bonded client.
Fixes: HAS/SR/CP/BV-03-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The implementtaion supports dynamic presets i.e. changing it's name,
registration and unregistration of the presets, thus Dynamic Presets
feature flag shall be set.
Fixes: HAS/SR/FEAT/BV-05-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
This handles Write Preset Name operation that is used to change the
preset name. This covers as well changing the preset name by server.
Fixes: HAS/SR/CP/BV-04-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
If preset synchronization is not supported, Preset Sync
Not Supported (0x82) shall be sent in error response.
Fixes: HAS/SR/SPE/BI-04-C, HAS/SR/SPE/BI-05-C, HAS/SR/SPE/BI-06-C
Signed-off-by: Mariusz Skamra <mariusz.skamra@codecoup.pl>
The malloc partition may actually be empty, if so, don't try to add it to
the domain as that will generate an error.
Signed-off-by: Keith Packard <keithp@keithp.com>
Send long-running tasks to a dedicated low-priority workqueue.
This shouldn't increase memory usage since by doing this, we get rid of the
ECC processing thread.
This should fix issues like #43811, since the system workqueue runs at a
cooperative priority, and the new dedicated one runs at a pre-emptible
priority.
Fixes#43811
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Workaround for nRF52 anomalies 102, 106, and 107 was applied also for
SoCs that were not affected with those, namely nRF52811 and nRF52840.
Since the side effect of this workaround is reduction of sensitivity,
this was highly undesirable.
This commit uses dedicated functions provided by MDK for checking if
a given anomaly applies to the used SoC (and its actual revision) so
that the workaround is applied only when it is really needed.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
Path list was replaced accidentally with free list during refactoring.
This change does fix the issue.
Signed-off-by: Veijo Pesonen <veijo.pesonen@nordicsemi.no>
Picolibc doesn't have the non-standard printf support required for tagged
arguments in cbprintf. Disable this and use the format string parsing code
instead.
Signed-off-by: Keith Packard <keithp@keithp.com>
Some errors can occur in the sending process that have to be handled
in a polling fasion instead of blocking using semaphores. In this case
apply an exponentially growing backoff time. This will allow for fast
reactions in most situations and prevents high system loads in case
resolving the situation takes a little longer.
Signed-off-by: Sjors Hettinga <s.a.hettinga@gmail.com>
The change in this commit is required to avoid regression errors
on EBQ test for the PHY update procedure
When in the peripheral role transmission of data must be resumed
while waiting for the PHY IND response from peer.
In other words: in the LP_PU_STATE_WAIT_TX_ACK_PHY_REQ state
data transmission must resume when acting as peripheral,
but not when in the central role
Following tests are effected
LL/CON/PER/BV-49-C
LL/CON/PER/BV-50-C
LL/CON/PER/BV-52-C
LL/CON/PER/BV-53-C
LL/CON/PER/BV-54-C
LL/CON/PER/BV-55-C
LL/CON/PER/BV-56-C
LL/CON/PER/BV-58-C
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
This PR fixes the PHY update procedures for conformance tests when
being a Central
The problem was that data was in the LLL tx queue and was still being
queued before the PHY IND was queued (with a given instant).
As a result by the time the PHY IND was transmitted over the air the
instant was in the past.
The fix is to ensure that the LLL tx queue is empty, and to stop
queueing new data before queueing the PHY IND
Following tests are fixed:
LL/CON/CEN/BV-49-C
LL/CON/CEN/BV-50-C
LL/CON/CEN/BV-53-C
LL/CON/CEN/BV-54-C
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
The BT_CODEC_LC3_CONFIG_* macros had a hardcoded location
and context, which makes them a lot less usable.
Updates the macro, and the macros that used them.
This also removes the BT_CODEC_LC3_CONFIG and instead
just uses the BT_CODEC_LC3_CONFIG_N macro, which has
then been renamed to BT_CODEC_LC3_CONFIG.
As an addition, the macros and their input has
also been better documented.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Enable pin control support for SWO log backend, by creating a new
ITM node for the ARM instrumentation trace macrocell. Add pin control
properties under this node, and refactor the swo-req-freq property to be
defined within this node.
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
Schedules I/O chains in the same order as they arrive providing a fixed
amount of concurrency. The low memory cost comes at the cost of some
computational cost that is likely to be acceptable with small amounts
of concurrency.
The code cost is about 4x higher than the simple linear executor
which isn't entirely unexpected as the logic requirements are quite a bit
more than doing the next thing in the queue.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
A DMA friendly Stream API for zephyr. Based on ideas from io_uring
and iio, a queue based API for I/O operations.
Provides a pair of fixed length ringbuffer backed queues for submitting
I/O requests and recieving I/O completions. The requests may be chained
together to ensure the next operation does not start until the current
one is complete.
Requests target an abstract rtio_iodev which is expected to wrap all
the hardware particulars of how to perform the operation. For example
with a SPI bus device, a description of what a read, and write mean
can be decided by the iodev wrapping a particular device
hanging off of a SPI controller.
The queue pair are submitted to an executor which may be a simple
inplace looping executor done in the callers execution context
(thread/stack) but other executors are expected. A threadpool executor
might for example allow for concurrent request chains to execute in
parallel. A DMA executor, in conjunction with DMA aware iodevs
would allow for hardware offloading of operations going so far as to
schedule with priority using hardware arbitration.
Both the iodev and executor are definable by a particular
SoC, meaning they can work in conjuction to perform IO operations
using a particular DMA controller or methodology if desired.
The application decides entirely how large the queues are, where
the buffers to read/write come from (some executors
may have particular demands!), and which executor to submit
requests to.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Add additional user tracing calls for thread states, names
and priorities.
add the following user tracing calls:
sys_trace_thread_create_user
sys_trace_thread_abort_user
sys_trace_thread_suspend_user
sys_trace_thread_resume_user
sys_trace_thread_name_set_user
sys_trace_thread_info_user
sys_trace_thread_priority_set_user
sys_trace_thread_sched_ready_user
sys_trace_thread_pend_user
Signed-off-by: Eugene Cohen <quic_egmc@quicinc.com>
In the function tx_work_handler, a pointer to ring buffer data is given
to usb_transfer then the ring_buf_get_finish is called. So, the data is
mark as read (by ring_buf_get_finish) while the data are not yet
transferred to usb. If later a user send data, the pointer hold by usb
stack could be rewrite.
Signed-off-by: Julien D'Ascenzio <julien.dascenzio@paratronic.fr>
Fix duplicate ID for systemview. Remove k_timer_user_data_set as it is
not being tracing and uses a duplicate ID like k_sem_reset.
Fixes#46541
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Complete the remote initiated version exchange if a LL_VERSION_IND is
received while already having responded in an earlier version exchange
procedure.
Clarify comment regarding how to handle this invalid behaviour.
This has been seen when running the LL/CON/CEN/BI-12-C test on EBQ.
Signed-off-by: Thomas Ebert Hansen <thoh@oticon.com>
Calculation of the DLE related parameters (rx/tx octets and time) depend
on the actual phy in use. For this reason the PHY settings must be
initialised before doing the DLE parameter calculations
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
On remote terminate on central the conn clean-up would happen before ack
of terminate ind was sent to peer.
Now clean-up is 'postponed' until subsequent event.
Also now data tx is paused on rx of terminate ind to ensure no data is
tx'ed after rx of terminate ind
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
On disconnect with refactored LLCP, if data tx is paused,
possibly 'waiting' tx nodes would not get released.
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
Only apply change to effective DLE times if current max times are too
small to accommodate. Similar to legacy implementation
Update unit tests to new DLE ntf behavior
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
When an operation on the socket is not supported by the implementation,
which is the case for some drivers, set errno to a value that reflects
this situation rather than signalling an error with the file descriptor.
Signed-off-by: Ulf Lilleengen <lulf@redhat.com>
The new interface contract guarantees that OpenThread stack uses
otPlatSettingsSet only for aKey which has at most one value at time.
This implies the simplification for key name used by settings subsystem
and decreases the count of records written each time when the value
for specific key is updated. In result non-volatile memory can be used
more efficiently.
It relates to zephyrproject-rtos/openthread commit: ed665e9 .
We still need to make sure that old entries are being removed for the
case with DFU.
Signed-off-by: Lukasz Maciejonczyk <lukasz.maciejonczyk@nordicsemi.no>
Use the newly added dma buffer alignment device tree property and macro
helper to set the buffer alignment.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
pb gatt server starts fast advertising for the first 60 seconds.
After that, server goes into slow mode.
When pb gatt and pb adv work in parallel pd gatt changes fast to
slow modes after the first tx frame in pd adv (about 110ms)
since they both are handled in the same thread.
Actually, pb gatt never worked in fast mode
if pb adv was enabled (in most configuration cases).
Signed-off-by: Aleksandr Khromykh <Aleksandr.Khromykh@nordicsemi.no>
