C++ subsystem defines exception throwing operator new.
Though the implementation never throws an exception.
If used by an application it should verify if requested
memory was allocated. This is against C++ standard.
If an application does not support exceptions or does not
want to call throwing new operator, it should use a
specialization of a new operator that makes sure it
never throws bad_alloc exception. The cpp subsystem
does not provide this specialization.
The commit adds missing operator new specializations.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Adds Kconfig option for the Mesh Shell Health server.
The shell implementation of Health Server test commands is
dependent on the user adding and initializing the shell Health
Server instance in their application to make the commands functional.
To prevent any confussion about how these commands work, a Kconfig
option has been added to allow conditional compilation of these
commands. In this way the user must actively enable this feature, making
the conditional requirements more clear.
Signed-off-by: Anders Storrø <anders.storro@nordicsemi.no>
This function allows the application to close an ipc_service instance.
It is intended for use when the remote agent it is communicating with
has gone down and allows for cleanup.
It first checks that the endpoints have been separately deregistered
with the instance using the deregister_endpoint function before
calling into previously added functions in ipc_static_vrings.c
and ipc_rpmsg.c to close those instances.
Signed-off-by: Jackson Cooper-Driver <jackson.cooper---driver@amd.com>
This function deinitialises the static vrings used by the ipc_service
instance. It acheieves this by tearing down the initialised
virtqueues before closing the various aspects of libmetal which
are in use.
Signed-off-by: Jackson Cooper-Driver <jackson.cooper---driver@amd.com>
This function does the opposite of the ipc_rpmsg_init function and
allows the rpmsg instance to be torndown in the case that the application
wishes to do so. It is intended to be used by the ipc_service subsystem.
Signed-off-by: Jackson Cooper-Driver <jackson.cooper---driver@amd.com>
LE Write Suggested Default Data Length and LE Set Data Length commands
are suggestions from host and should be validated only as per HCI
specification regarding internal setting of LLCP.
LLCP is allowed to use other values if needed.
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Due to the diverse coding styles and lack of preprocessing when
scanning for test cases, there were many unintended combinations
of yaml test scenarios and C test functions with the regex-based
test case discovery, which caused an inaccurate test plan and test
result stats.
As the new ztest fx is used, the test cases of a test instance can
be accurately determined via zephyr.symbols file.
Because the zephyr.symbols file is not available until after build,
test cases determination is placed right after the build stage of
the pipeline and before the runtime test collection.
For those test instances that don't go through the build stage,
such as those with "Skip filtered" (statically filtered) reason
before entering the pipeline, they are not affected.
This patch also adjust the stats logic because the zephyr.symbols
file is generated after build. So ExecutionCounter update is split
and some must be postponed until test execution pipeline is completed.
Some concepts:
A test instance = a yaml scenario + a platform
"Test instance" and "test config" are synonyms in twister, unfortunately
excessive IMHO...
A test instance can be filtered or skipped in 3 ways.
Let's define "runtime" as "after entering the execution pipeline".
1) statically filtered (before runtime)
Such test instance is filtered by analyzing the yaml and never
enters the execution pipeline.
2) cmake filtered (runtime)
Such test instance enters pipeline and is filtered at cmake stage.
3) build skipped (also runtime)
Such test instance enters pipeline and is skipped at build stage.
All the test instances that enter the execution pipeline will go
through the report stage, where ExecutionCounter is further updated.
The meaning of the fields of ExecutionCounter are:
.skipped_configs = static filtered + cmake filtered + build skipped
.skipped_runtime = cmake filtered + build skipped
.skipped_filter = static filtered
.done = instances that enter the execution pipeline
.passed = instances that actually executed and passed
Definition of the overall stats:
total_complete = .done + .skipped_filter
total = yaml test scenario * applicable platforms
complete percentage = total_complete / total
pass rate = passed / (total - skipped_configs)
Signed-off-by: Ming Shao <ming.shao@intel.com>
This is mainly to reduce clutter in `subsys/logging`, but also to make
backend management slightly easier.
Signed-off-by: Christopher Friedt <cfriedt@fb.com>
The memory used for Tx buffers and procedure context must
be aligned on a word boundary, otherwise it is possible
to get unaligned memory access
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
This change fixes a bug where the second use of shell command `gatt
exchange-mtu` would trigger an assertion failure in the stack. The BT
API `bt_gatt_exchange_mtu` asserts `params->func != NULL`.
Signed-off-by: Aleksander Wasaznik <aleksander.wasaznik@nordicsemi.no>
Fix radio_tmr_start_us to be soft realtime, if the requested
start in microseconds is in the past, setup the next
earliest expiry. This is to prevent radio not being started
and not let the LLL hanging. When setup to receive auxiliary
PDU in LLL scheduling, is_aux_sched is set and if radio does
not start and the window is requested to be pre-empted, then
it is blocked from being preempted due to the is_aux_sched
flag remaining to be set. This stalls the scanning
infinitely until explicit scan disable.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix missing is_aux_sched flag being reset before resuming to
scan for new advertising PDU in the same scan window after
having received an earlier primary channel PDU and auxiliary
channel PDU using LLL scheduling.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix receiving spurious overlapping ADV_IND and SCAN_IND PDU
when LLL state expects to receive SCAN_RSP PDU. Explicitly
check that these PDUs are not processed when in active
scanning and waiting for SCAN_RSP PDUs.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
If scan is being disabled, rx could already be enqueued
before coming to ull_scan_aux_rx_flush. Check if node rx is
not the last in the list of received PDUs then add it to the
list, else do not add it to the list to avoid duplicate
report generation, release and probable infinite loop
processing of the list. It is possible that node rx has not
be already added in case of last PDU in the chain, under
this case when scan is being stopped add it to the list
before auxiliary context is flushed.
Add the node rx to aux context list of node rx if not
already added when coming to ull_scan_aux_rx_flush. This is
handling a race condition wherein the last PDU in chain is
received and at the same time scan is being disabled.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix scan aux context leak on Periodic Sync Terminate. On
Periodic Sync Terminate the `is_stop` flag is set to prevent
the ULL from assigning an aux context to the sync context,
but the acquired aux context was neither assigned to sync
context nor was being released back to the aux context pool.
Fixed by releasing the aux context if not assigned to the
sync context when `is_stop` flag is set.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
If a device sends an MTU that is bigger than our maximum tx buffer size,
that could cause assertion failures down the line.
This PR limits it to the maximum we support (CONFIG_BT_L2CAP_TX_MTU).
The issue has been observed with a gatt discovery procedure, but is likely
present in other places in att.c.
To reproduce it, we need two zephyr shell devices, with one having a larger
MTU than the other:
- connect
- do data length update to the bigger MTU
- set security to 2, EATT channels get connected
- launch a gatt discovery from the device with the larger MTU
- observe kernel panic on the other device when it attempts to add too much
memory to a net buf.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Log the LE Secure Connection (SC) LTK in a pairing procedure without
bonding when CONFIG_BT_LOG_SNIFFER_INFO is enabled.
Fixes: #50691
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
With introduction of refactored LLCP was also introduced an error
re. calculation of initial slot reservation in connection setups.
Due to this error a bigger slot was reserved, leading to failure due to
connection overlaps in multiple.sh test.
Also the removal of this 'overhead' in slot reservations calls for
bigger time-allowance for l2pcap_stress test.
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
The unicast client should never need to set the state of the
endpoints directly, without an notification from the server.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Remove the support for caching ASEs on release.
This is mainly due to the 2-stage design in Zephyr: We
have the bt_audio_stream and the bt_audio_ep, where the
the latter represents an ASE in most cases.
However, once the endpoint goes into the codec configured
state, then a stream must be bound to it. This would be fine
(although a stream is technically not established at this point),
except that a stream is also, for unicast, specified for a
specific ACL. Once the ACL disconnects, the stream technically
becomes invalid, and if the stream is invalid, so is the endpoint.
Based on that, there's no reason (or even option) to keep the
endpoint in a codec configured state, as it would not
have a valid stream associated with it.
If we want to support an ASE going into the codec
configured state on ASE release (or ACL disconnect),
the relationsship between ASEs and streams needs to
be redefined.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Fixed unstable thread state read for detetect engine thread
state.
Fixed engine missing socket conrext add.
lwm2m resume now do update also from network error state
Signed-off-by: Juha Heiskanen <juha.heiskanen@nordicsemi.no>
When getting a channel timeout, l2cap_chan_destroy is called from the
rtx_work work item.
In that function we attempted to cancel the current work item, and sync on
it being cancelled. The kernel API says that this will block until the work
item completes execution, hence a deadlock.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Fix implementation to not reset the ticks_slot_id_previous
when ticker yields its reserved slot, only reset if ticker
was stopped.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix ticker worker from checking the ticks_slot_id_previous,
only ticks_slot_previous be checked as previous slot id
would be assigned to TICKER_NULL if the ticker was stopped.
The ticks_slot_previous value should be used to check that
a slot is reserved even if the ticker has been stopped.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
If LwM2M engine was paused at ENGINE_UPDATE_SENT it was not
able to resume lwm2m engine properly. There was also missing
RD client Pending message free which could also affect a issue.
Signed-off-by: Juha Heiskanen <juha.heiskanen@nordicsemi.no>
RD client was not closing socket socket when Update registration
timeout. It fail new connection open. There is added a flag for
detect a case when Registration state need to be re-open a socket
for new connection.
Signed-off-by: Juha Heiskanen <juha.heiskanen@nordicsemi.no>
Some testsuites dump lots of output very fast where some systems are not
able to capture the complete output from the tests. Add a slight delay
between each test in the suite.
New kconfig CONFIG_ZTEST_TEST_DELAY_MS is added to ztest.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add in Kconfig, select DEPRECATED, to warn at build time
that the LLCP legacy implementation is deprecated and the
refactored LLCP enabled by BT_LL_SW_LLCP Kconfig option be
used instead (which is the default now).
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This callback allows use-cases where the SDU is much larger than the l2cap
MPS. The stack will then try to allocate using this callback if specified,
and fall-back on using the buffer's pool (previous behavior).
This way one can define two buffer pools, one with a very large buffer
size, and one with a buffer size >= MPS, and the stack will allocate from
that instead.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
See the code comments.
SDUs might enter a state where they will be blocked forever, as a
workaround, we nudge them when another SDU has been sent.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
There was an edge-case where we were sending back too much credits, add a
check so we can't do that anymore.
Signed-off-by: Jonathan Rico <jonathan.rico@nordicsemi.no>
Fix BIS max PDU size used in calc of sub_interval, use the
minimum of supported maximum BIS PDU size and the requested
SDU size. This will avoid wasted air time between subevents
when SDU size is less than maximum supported PDU size, where
only 150 us tIFS is mandatory between subevents.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When calculating the offsets between primary advertising
PDU, auxiliary PDU, Periodic Advertising PDU, and BIS PDU,
the values used as anchor points for starting the periodic
interval for auxiliary, periodic and BIG events, should be
ensured to have the minimum auxiliary frame spacing T_MAFS
between the PDUs.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Use a local variable to assert check on auxiliary context's
release.
Add missing volatile qualifier to the parent field of the
auxiliary scan context.
Under race condition a released aux context can be allocated
for reception of chain PDU of a periodic sync role, hence
fix the assertion check to consider that the released
context can be allocated to other roles.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Increase the BIS PDU access address receive timeout, use the
+/- 2 us jitter for each subevent.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix inifinite loop in ll_rx_get() when both ISO Broadcast
and ISO Receive features are enabled in a build. Missing
conditional compile prevented the ack_last index that
maintains a relation between number of packets complete and
the received data packets was unassigned.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
According to Bluetooth 5.3 Vol 4, Part E section 7.7.65.21 the
Controller should send an IQ report with Packet_Status set to 0xFF
and Sample_Count to 0x0, to inform Host that it is not able to
store IQ samples for all received Constant Tone Extensions and has
failed to sample at least once.
This commit adds handling of such situation for direction finding
in connectionless mode in lower link layer. It includes:
- lack of node_rx_iq_report to store IQ samples,
- lack of free node_rx to report received advertising data (generates
an advertising data report with indication "incomplete, no more to
come" and cancels reception of chained PDUs,
- cancellation of chained PDUs reception due to other scheduling
reasons.
In all of these situations CTE reception must be enabled and number of
received CTEs is smaller than requested by Host.
If there is not sufficient number of node_rx_iq_report nodes and
report was generated, the CTE will not be samples in a given
periodic advertising synchronization event. The CTE will be
sampled again in next event if there are node_rx_iq_report available.
End user applications may check periodic advertising sync event
counter when use of the reports to be sure about sync event
the received report is connected with.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The img_mgmt_erase could return MGMT_ERR_EOK even when
img_mgmt_impl_erase_slot failed.
Fixes#50522
Signed-off-by: Dominik Ermel <dominik.ermel@nordicsemi.no>
The GCC/GCOV version over 12 has slight format change of the gcno
and gcda. Make some adaption in the gcov dump function to fix the
code coverage report.
Mainly two places change:
1. Added the checksum in the struct gcov_info. This fix the crash
in qemu_x86, and mps2_an385 when run with --coverage.
2. Adjust the GCOV_TAG_FUNCTION_LENGTH accroding to gcov-io.h. It's
length unit is caculated by bytes now.
Fixes#50255.
Fixes#50257.
Signed-off-by: Enjia Mai <enjia.mai@intel.com>
