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>
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>
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>
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>
Encryption request is enqueued in thread context from the Tx
buffer pool, so that it is serialized alongwith the already
enqueued data buffers ensuring they are transmitted out to
peer before encryption is setup. Allocate additional Tx
buffers to accommodate simultaneous encryption setup across
active connections.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
- Prevent NULL-pointer dereferencing if datapath is created late
- Support SDU fragment complete-counting for framed case
Signed-off-by: Morten Priess <mtpr@oticon.com>
Implemented incrementing TX SDU fragment count such that it indicates
the number of completed SDU fragments in the PDU being emitted for both
unframed and framed transmission.
Signed-off-by: Nirosharn Amarasinghe <niag@demant.com>
HCI:
-- Discarded data from HCI ISO Data packets from controller to host if
data has been lost
ISO-AL:
-- Corrected iso_interval in latency calculations
-- Updated handling and release of SDUs for error conditions / padding
at the last PDU for the SDU
-- Updated prioritisation of error status in released PDUs
-- Included error spooling exit on based on payload number to SDU
mapping for unframed reassembly
-- Updated sequence number handling for framed recombination
Signed-off-by: Nirosharn Amarasinghe <niag@demant.com>
Do not use ull_ticker_stop_with_mark for ISO one-shot resume ticker, but
stop without checking result. If active this will stop it, otherwise it
is ignored. Also, this prevents calling lll_disable twice.
Signed-off-by: Morten Priess <mtpr@oticon.com>
- Prioritize CIS_REQ handling in (old) LLCP
- Reject if CIS_REQ uses exisiting CIS ID
These fixes prevent assertions in /LL/CIS/PER/BV-38-C.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Implementing proper validation of PHY selection for PHY UPDATE procedure
Implement connection termination on PHY UPDATE with instant in the past
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
Replace k_sem_take loop used for consuming the remaining
sem give counts with k_sem_reset.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix assertion due to multiple mayfly_enqueue calls used
under ZLI when pdu_free_sem_give is invoked from the LLL.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
The Configure Data Path command must be included when the read Local
Supported Codecs [v2] command is supported. The Configure Data Path
command is needed when vendor-specific ISO data paths are supported,
but not otherwise tied to ISO. The respective weak function is
therefore moved to ull.c.
Signed-off-by: Wolfgang Puffitsch <wopu@demant.com>
When supporting chain PDUs, after HCI reset the first PDU
allocated did not initialize its next pointer causing use
of unallocated PDUs for chaining while a duplicate of such
PDU buffer being allocated for other states/roles. This
causes in certain cases both Extended and Periodic PDUs
having same data, the Extended Advertising PDU being
transmitted at the instant of the Periodic Advertising.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix BT_CTLR_RX_ENQUEUE_HOLD implementation that stalled
generating the CONN_UPDATE and PHY_UPDATE when there were
no Rx data or Tx acknowledgements to be processed when
the controller was built with BT_CTLR_LOW_LAT_ULL.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add Kconfig option to conditionally enable tickers with slot
window to yield to normal tickers and be placed at the end
of their slot window if possible.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
The open source controller did not expose support for shutting it down
via the close() API in the HCI driver. Add support for it, which
completes the support for disabling and enabling the Bluetooth stack in
Zephyr.
Closes#3192.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Allow the controller to be deinitialized, adding the whole chain of
calls:
- ll_deinit()
- lll_deinit()
- lll_clock_deinit()
in order to be able to turn everything off, including the controller's
refcount of the LF clock.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
Fix Data Length Request CMD parameter checking to not
restrict invalid maximum Tx time values. This addresses
conformance test cases that provide maximum Tx time values
of 2128 us when not supporting Coded PHY in the Controller
implementation.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
If there is a PHY change on a connection it may happen that effective
RX and TX time changes also. That change is applied after an instant.
Implemented handling of effective time calculation is based on the
maximum PDU length, new PHY and local (default) maximum TX or RX time.
The maximum TX value is set to default one that corresponds to PHY 1M
during the Controller initialization. It can be updated by host to other
value. By default Zephyr Host updates it to max possible TX time for all
supported PHYs. If PHY CODED is enabled, it is the longest possible TX
duration 17040 us.
The maximum RX value is set to default during connection creation.
In case of use of legacy advertising, the value is also related with
PHY 1M. It can be updated by data length extension procedure.
If the maximum RX value is set to some value and there is a change
of a PHY to one that requires more time to send a PDU with the same
length, then the maximum RX value is wrongly calculated.
Function pu_calc_eff_time returns a value that is the default_time
argument. The problem is that the default_time should be adjusted
to new maximum RX time required for a new PHY.
To solve that there should be an evaluation of a new maximum RX and
TX time based on new PHY.
The commit adds missing evaluation.
The problem occurred in DF tests that check collision mitigation
between PHY update control procedure and CTE request control procedure.
There was missing CONFIG_BT_CTLR_PHY_CODED option in CTE request
unit tests. The code was working because the ULL implementation of
PHY change control procedure does not verify if PHY CODED is supported.
When missing support was enabled, tests showed wrong evaluation of
maximum RX time. It also unveiled error in CTE request unit tests
implementation. The default_tx_time was set to wrong value 2120 us
as if PHY CODED was not supported. To fix it, the value was changed
to PDU_DC_PAYLOAD_TIME_MAX_CODED.
There was also added a mock for a feature exchange procedure done
during unit tests setup step. That allows to correctly calculate
maximum TX time by ull_dle_max_time_get function.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Fix filter accept list and privacy feature Kconfig default
based on whether host has them enabled.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix missing BT_CTLR_PRIVACY conditional compilations that
included redundant HAL code when Controller privacy is
disabled.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix PHY update procedure to handle unsupported PHY requested
by peer central device. PHY update complete will not be
generated to Host, connection is maintained on the old
PHY and the Controller will not respond to PDUs received on
the unsupported PHY.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fixes issue re. sending packets after sending TERMINATE_IND.
Fixes issue re. erroneous error code on 'timed out' termination (ie
when peer does not ack terminate_ind)
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
On a completion of feature exchange the max DLE times needs to be
updated if Coded PHY has become supported.
For this now also keep a separate store of the default_tx_octets/time
for the sake of re-calculation of local versions of DLE values
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
Fix prepare pipeline from not resuming non-resume events
in pipeline when more than one resume events are present
before a non-resume event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When host issue HCI Reset controller should clear LE Static Random
address. Otherwise controller may incorrectly use one from before
HCI reset insread of rejecting HCI commands.
This was affecting following qualification test cases:
HCI/CCO/BI-51-C
HCI/CCO/BI-53-C
HCI/CCO/BI-54-C
HCI/CCO/BI-56-C
HCI/DDI/BI-06-C
HCI/DDI/BI-07-C
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
At least one supported PHY bit shall be set and none of
non-supported or RFU bits should be set.
This was affecting HCI/DDI/BI-05-C qualification test case.
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Removes label experimental for direction finding feature in
Bluetooth Controller.
The feature stays disabled by default because it requires additional
dedicated hardware to be used. Besides that it significantly enlarges
amount of memory used by the Controller.
End user has to enable the feature explicitly by use of configuration
options.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Fix preempt timeout scheduling to not skip first prepare in
pipeline when there are no previous preempt timeout
scheduled already.
Fix related to commit 27b8beaa22 ("Bluetooth: Controller:
Fix to handle relative short preempt timeout").
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Maximum number of CTE to be send in a periodic advertising chain is
limited by BT 5.3 Vol 4 Part E, section 7.8.80 and is 16.
Maximum number of samples CTE in a periodic advertising chain is
limited by BT 5.3 Vol 4 Part E, section 7.8.82 and is also 16.
Both values are limited by amount of memory reserved in Controller
by Kcofigs, respectfully: CONFIG_BT_CTLR_DF_PER_ADV_CTE_NUM_MAX and
CONFIG_BT_CTLR_DF_PER_SCAN_CTE_NUM_MAX.
Functions ll_df_set_cl_cte_tx_params and ll_df_set_cl_iq_sampling_enable
check limits against Bluetooth specification provided values.
That can end with memory overwrite if number of requested CTEs is
greater that number provided in configuration.
The commit changes mentioned functions to validate HCI commands
parameters against configuration values.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
In case there are pending two control procedures: local and remote,
and remote device sends LL_REJECT_IND PDU then assertion happened.
ull_cp_rx function triggers assertion because LL_REJECT_IND is
considered as expected packet for all local and remote control
procedures.
That is not correct. It does not allow local procedure to handle
a response and stops a device on assert.
A remote device should not send LL_REJECT_IND PDU for locally initiated
control procedures, hence local device should not expect to receive
LL_REJECT_IND PDU to be expected one for pending remote control
procedure.
Remote device is allowed to send LL_REJECT_EXT_IND PDU for a locally
initiated control procedures. The LL_REJECT_EXT_IND PDU has a Reject-
Opcode that identifies rejected control procedure, hence is may be
expected by local device.
The commit provides changes that prevent the Controller assertion
in described case.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
CCM during on-the-fly decryption of a received packet starts decryption
when Radio triggers EVETNS_ADDRESS. In case there is possibility a
packet may include Constant Tone Extension, on-the-fly parsing of a
received packet for CTEInfo is enabled.
If there is a PHY 1M enabled the Radio stores received bits with a
delay, that is equal to time required to receive 3 bits. CCM TASKS_CRYPT
related with packet decryption should be delayed by the time the Radio
needs to store received data.
The commit provides changes required to delay start of the CCM
TASKS_CRYPT. It uses NRF_RADIO Bit counter feature. The Bit counter is
configured to trigger NRF_RADIO->EVENTS_BCMATCH on reception of 3rd bit.
The event is connected through PPI with CCM TASKS_CRYPT.
The PPI used is shared with Radio Rate override. That is possible because
direction finding feature is not allowed on PHY Coded and CCM needs a delay
only when used PHY 1M.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
lll_conn_rx_pkt_set function is called before one of radio_switch-
_complete_XXX functions calls in lll_peripheral.c. That prevents
lll_conn_rx_pkt to call any Radio setup related function that touches
NRF_RADIO->SHORTS. NRF_RADIO->SHORTS are assigned in radio_switch_-
complete_XXX functions, hence any change operation on the register done
before is overwritten.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Add initial support for Periodic Advertising Data fragment
operation support.
This commit is limited to detection of partial data if used
before Periodic Advertising is enabled, which will lead to
HCI error reason of command disallowed.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
When resolving collision if ticks_slot_window is set for
either of the ticker then skip and such ticker be
rescheduled outside the collision within the
ticks_slot_window.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
[140/235] Building C object zephyr/subsys/bluetooth/controller/
CMakeFiles/subsys__bluetooth__controller.dir/ll_sw/ull_sched.c.obj
In file included from zephyr/subsys/bluetooth/controller/ll_sw/
ull_sched.c:28:
zephyr/subsys/bluetooth/controller/ll_sw/nordic/lll/lll_adv_pdu.h:
In function 'lll_adv_pdu_linked_next_get':
zephyr/subsys/bluetooth/controller/ll_sw/nordic/lll/lll_adv_pdu.h:8:32:
warning: implicit declaration of function 'MROUND'
[-Wimplicit-function-declaration]
8 | #define PDU_ADV_MEM_SIZE MROUND(PDU_AC_LL_HEADER_SIZE + \
| ^~~~~~
zephyr/subsys/bluetooth/controller/ll_sw/nordic/lll/lll_adv_pdu.h:12:11:
note: in expansion of macro 'PDU_ADV_MEM_SIZE'
12 | PDU_ADV_MEM_SIZE - \
| ^~~~~~~~~~~~~~~~
zephyr/subsys/bluetooth/controller/ll_sw/nordic/lll/lll_adv_pdu.h:197:9:
note: in expansion of macro 'PDU_ADV_NEXT_PTR'
197 | return PDU_ADV_NEXT_PTR(pdu);
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Rework instant based procedure complete event generation to be
held until after the on-air instant has elapsed, to have
conditional compilation around the code where the event
generation be held or immediately dispatched so that it
improves readability.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix instant based procedure complete event generation to be
held until after the on-air instant has elapsed.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
If peer sents invalid value in DLE request or response just ignore
those and keep using old values when calculating effective DLE.
This was affecting following qualification test cases:
LL/CON/PER/BI-10-C
LL/CON/PER/BI-11-C
LL/CON/PER/BI-12-C
LL/CON/CEN/BI-07-C
LL/CON/CEN/BI-08-C
LL/CON/CEN/BI-09-C
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Terminate connection when Connection Update with instant in the past
Fix type re. win_offset calc in conn update ind pdu decoder
Remove feature bit on unsupported in peer
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
Upon collision either the error code for different procedure
collision or same procedure collision must be transmitted,
which is fixed in this PR. Previously always the error code
for same procedure collision was sent
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
Add Extended Advertising Type validation when associated
Periodic Advertising is enable, and Extended Advertising
set is re-configured to other advertising types.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
There is an error. CTE request control procedure can be run in single
shot or periodic mode. In case of run in single show, it is not
disabled after completion.
The code responsible for the disable was deleted by commit:
ac7d0506f88508e7440288c1ec4319a9f7d0ad44.
The cte_req.is_enabled should be set to zero if the CTE request
completes and cte_req.req_interval is zero.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There was missing code responsible for handling of unexpected response
for CTE request. The commit adds code that will terminate connection
in case a peer device reposnes with unexpected control PDU that is not
a remote procedure reques.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The CTE request procedure should be enabled only if a peer supports
CTE response feature. That information can be obtained by feature
exchange procedure. If there were no feature exchange then CTE
request feature may be disabled if a peer responses with LL_UNKNOWN_RSP
for a CTE request.
The implementation of ll_df_set_conn_cte_req_enable was checking if
CTE response feature is supported only when there was feature exchange.
There was missing possibility to stop CTE request if a peer responded
with LL_UNKNOWN_RSP for an earlier CTE request.
The commit changes the implementation of ll_df_set_conn_cte_req_enable.
The CTE response feature check is moved to ull_cp_cte_req function,
because it belongs more to control procedure than to function that
handles Host request to start the procedure.
Second change is related with use of conn->llcp.fex.features_used.
It stores information about features supported by peer. It does
not depend on execution of the feature exchange control procedure.
By the way, there were removed else statement in ll_df_set_conn_cte_-
req_enable because it was not needed.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There was missing handling of LL_UNKNOWN_RSP in CTE request control
procedure.In case there is a pending CTE request and peer responses
with LL_UNKNOWN_RSP then Host should be notified with HCI_LE_CTE_-
Request_Failed event. The pending CTE request procedure should be
completed.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Fix handling of AD Data set with operation type of
unchanged data with respect to invalid parameter and
state of Periodic Advertising.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix nRF51x series advertising channel PDU S1 byte
configuration regression, that causes corrupted advertising
PDUs being reported to Host, by adding the missing bits_s1
local variable initialization.
Regression introduced in commit 1dcbe73cc8 ("Bluetooth:
controller: radio: add setting of S1 byte in radio packet").
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Current implementation of ll_sync_create_cancel does not allow to stop
synchronization after ull_sync_setup is called. When that is done,
sync->timeout_reload is not zero and the ll_sync_create_cancel will
return BT_HCI_ERR_CMD_DISALLOWED. That means the Controller is able to
cancel periodic advertising synchronization only in period between
call to ll_sync_create and reception of AUX_ADV_IND that has SyncInfo
field.
The Controller should be able to cancell synchronization until first
AUX_SYNC_IND PDU is received and host notified about synchronization
established.
Complete information about synchronization status is provdied by two
ll_sync_set members: node_rx_sync_established and timeout_reload.
These two members of the structure were used in ll_sync_create_cancel
function to do a proper cancel and cleanup.
The node_rx_sync_established member was not cleared when sync was
established or expired. That was required to get a proper information
about synchronization state.
Besides that, to avoid race condition between ll_sync_create_cancel
and ull_sync_established_report, the latter function was extended
to check if cancel operation or sync lost has happened.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Remove experimental label from Advertising Extensions and
Periodic Advertising features in the Controller.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This PR sets the refactored LLCP as the default instead of legacy
so that more people will use this, giving us more feedback
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
Direction finding functionality does antenna switching during CTE
reception in AoA mode and CTE transmission in AoD mode. Antennas are
switched according to user provided antenna switch pattern. If a CTE
length is enough to exhaust all antenna ids in a switch pattern then
radio should loopback to reference antenna and continue from switching
from that. Current implementation loops back to antenna that is just
after reference antenna in the switch pattern.
The commit fixes that by insertion of the reference antenna
at the end of switch pattern. Radio will operate as it was before,
it will restart switching from the same index in a switch pattern.
At the same time reference antenna will be inserted into the loop
if switching pattern is exhausted. That also means, the maximum
number of antenna ids in a switch pattern has to be decreased by
one.
The commit also fixes lower bound in the BT_CTLR_DF_MAX_ANT_SW_PATTERN-
_LEN that should be equal two. It was set to three, because in the
past the antenna at index zero was used for reception of a PDU.
Now this antenna ID is provided by device tree configuration.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The SYNC_STAT_READY_OR_CONT_SCAN state into two: SYNC_STAT_READY
and SYNC_STAT_CONT_SCAN. The connected state was used for two
cases:
- synchronization with periodic advertising is already established,
- CTE type is wrong but periodic advertising list is used to filter
synchronized devices.
In the past states were connected because there was only one difference
between them. In case of sync already established, Controller generated
sync report notification to host.The sync was not terminated when
Controller was about to continue search for other periodic advertiser.
The state was split because now Controller terminates sync ticker and
continues search for other periodic advertised. The split improves
readability and makes code easier to understand.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
In case of use of filtering based on: periodic advertising list and
CTE type, the synchronization can hang. That is possible if a periodic
advertiser uses wrong CTE type. In such situation the sync is not
released in ull_sync_done call. What more the sync->timeout_reload
is not cleared and Host is not able to cancel the synchronization.
The periodic advertising is in a semi-sync-established state.
There are no reports send to Host. Host can't use the sync set to
synchronize with other device. It is only able to terminate the
sync (call to ll_sync_terminate).
To fix the issue following changes should be applied:
- isr_rx_adv_sync_estab should call isr_rx_done_cleanup
with sync_term parameter in case the sync_ok isn't SYNC_STAT_ALLOWED.
In any case the CTE type is wrong, no matter is the periodic
advertising list filtering is enabled or not.
- ull_sync_established_report should set sync->is_term to true
in case the CTE type is not allowed. That change is required for devices
that do not support Direction Finding Extension. For those devices CTE
type based filtering is done in ULL by ull_sync_established_report
function. The sync->is_term should be set unconditionally, hence is
moved up in the function.
With these two changes done, ull_sync_done function will execute
sync_ticker_release in case the CTE has wrong type. ULL, depending on
notifications prepared by ull_sync_established_report, will follow up
on sync termination if required.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
The periodic CTE_REQ disable command, requested by Host, may be locked
until connection is dropped due to missing CTE_RSP from peer device.
That is caused by implementation of CTE_REQ disable and CTE_REQ
control procedure handling.
The procedure is marked as active when CTE request was send to peer
device. It is marked as inactive after completion of the procedure.
That caused locking of CTE disable on a semaphore.
The BT 5.3 Core Spec, Vol 4, Part E, section 7.8.85 says the HCI_LE_-
Connection_CTE_Request_Enable should be considered active on a conne-
ction from when Host successfully issues the command with Enable=0x1
until a command is issued with Enable=0x0 or single LLCP CTE request
has finished (CTE_Request_Interval=0x0). Also there is a clarification
from BT SIG that the command with Enable=0x0 does not affect any
initiated LLCP CTE request. That means Controller is allowed to finish
already started procedure and it is not allowed to start new LLCP CTE
request procedure after completion of the command with Enable=0x0.
Taking that into account, there is no need to synchronize ULL and LLL
in regard of disable the LLCP CTE request while the procedure is
pending. Controller is free to complete the procedure or terminate it.
The change removes all code related with cte_req.is_active, disable
callback and waiting of ULL for LLL to finish the LLCP CTE request.
The ULL will complete the HCI_LE_Connection_Request_Enable with
Enable=0x0 immediately. In case the procedure is disabled in before
the response arrives, then further processing of the response is
dropped and the procedure context released.
The context is not released by the code responsible for disable
handling, to have single place where it is done.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
There is an error in the condition that checks if new CTE_REQ should
be started after end of connection event. The condition verifies if
counter req_expire is set to zero. Then new CTE_REQ is started
irrespectively to CTE_REQ being disabled.
req_interval is used to store information if the CTE_REQ is:
- periodic, then value doesn't equal zero,
- single shot or disabled, then value equals zero.
The condition should verify if the req_interval is not zero and
req_expire is not zero. The second part of the if condition is
required to avoid starting next CTE_REQ until last one has been
completed.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Some of control procedures has set ctx->rx_opcode in rp_comm_tx
function to values that state there is an expected PDU.
This is not always true. In case of procedures that do not expect
any response from remote device, the ctx->rx_opcode should be set
to PDU_DATA_LLCTRL_TYPE_UNUSED.
In the worst cases scenario, when the same control procedure
is executed locally and remotely, correct response PDU may
cause an assertion in ull_cp_rx. It could happen because
of wrong ctx->rx_opcode value. A packet with opcode that is
set in remote and local control procedure context rx_opcode
will be treated as expected value for both. That is a situation
that cannot happen.
The commit changes the assignments of rx_opcode to fix the problem.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
This commit fixes the build errors for the testcases in
the tests/bluetooth/init folder
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
The tests/bluetooth/ctrl_user_ext test fails due to
compilation errors, which are fixed with this PR
Signed-off-by: Andries Kruithof <andries.kruithof@nordicsemi.no>
In order to bring consistency in-tree, migrate all subsystems code to
the new prefix <zephyr/...>. Note that the conversion has been scripted,
refer to zephyrproject-rtos#45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Strictly restrict AD Data length to BT_CTLR_ADV_DATA_LEN_MAX
when there can be free bytes in Advertising PDU with common
extended header format of less that the maximum 64 bytes.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
In a function pu_prepare_instant there is a condition that
checks if there is actual change of a PHY. That condition
was based on ctx->data.pu.tx and ctx->data.pu.rx.
These members store PHY that is or will be used, hence
the condition is wrong. Even there is no actual change in
the PHY, values could be not equal to zero. In such case
the instant value would be set to wrong value.
What more the condition be an 'or' not an 'and' because
one of the values must be different than zero to have the
PHY change and instant different than zero.
After update of the condition, the function call places
must be changed. The ctx->data.pu.c_to_p_phy and
ctx->data.pu.p_to_c_phy are set in pu_prepare_update_ind
function, hence pu_prepare_instant should be called after
that.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
- ISO TX data path for HCI and support for vendor path
- ISO-AL segmentation of framed PDUs
- Insertion of segment headers
- Reconstruction and storing of CIG reference point in ULL
- Calculation and insertion of of Time-Offset
- Exit error spooling in ISO-AL on detecting start
- ISO-AL TX unframed fragmentation
Signed-off-by: Nirosharn Amarasinghe <niag@demant.com>
Signed-off-by: Morten Priess <mtpr@oticon.com>
Add implementation to generate node rx for missing ISO PDUs
and set the status as invalid. This is required for ISOAL to
correctly track the sequence numbers for every SDU interval.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add Kconfig to limit the Primary Advertising Interval and
Periodic Advertising Interval maximum supported values in
the Controller implementation.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
None of the common DF configs should depend on using BT_LL_SW_SPLIT.
Added dependencies to it where this was the case.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
These configurations are tighlty coupled to the implementation,
so these should be hidden when not using BT_LL_SW_SPLIT.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
Some controllers may support only TXing or RXing CTE.
As all DF features are guarded by BT_CTLR_DF, we need to ensure
that it is possible to select those when only TX or RX is available.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
In case a connection is encrypted received PDU is decrypted by CCM.
CCM does not encrypt/decrypt S1 byte that stores CTEInfo.
In case of reception of a PDU by encrypted connection there is missing
CTEInfo in a memory where CCM stores decrypted PDU.
The CTEInfo data must be copied from scratch packet.
The commit adds code responsible for copying of the CTEInfo into
target PDU memory.
Signed-off-by: Piotr Pryga <piotr.pryga@nordicsemi.no>
Terminate connection with a MIC failure if an unexpected control PDU
is received during the Encryption Start procedure.
Add a greedy option to pdu_is_expected() to make sure the procedure
processes all unexpected control PDU in all cases.
Add unit test inspired by Bluetooth Qualification test
LL/SEC/CEN/BV-14-C,
Central Receiving unexpected PDU during encryption start
Signed-off-by: Thomas Ebert Hansen <thoh@oticon.com>
Add support for both LL_REJECT_IND and LL_REJECT_EXT_IND when waiting
for the the response to the LL_ENC_REQ and LL_START_ENC_REQ.
Add unit test to test both LL_REJECT_IND and LL_REJECT_EXT_IND as
responses to LL_ENC_REQ.
Signed-off-by: Thomas Ebert Hansen <thoh@oticon.com>
Chose correct rejection PDU based on features supported on remote peer
when rejecting due to missing LTK.
Update unit test setup with faking that a feature exchange procedure
has run.
Signed-off-by: Thomas Ebert Hansen <thoh@oticon.com>
Terminate connection with a MIC failure if an unexpected control PDU
is received during the Encryption Start procedure.
Add unit test inspired by Bluetooth Qualification test
LL/SEC/PER/BI-05-C,
Peripheral Receiving unexpected PDU during encryption start
Signed-off-by: Thomas Ebert Hansen <thoh@oticon.com>
This adds suport for validation of received PDUs before they are
pass further to LLCP state machines. If PDU size is invalid it is
rejected with LL_UNKNOWN_RSP.
This was affecting following qualification test cases:
LL/PAC/PER/BI-01-C
LL/PAC/CEN/BI-01-C
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
Instead of waiting one connection interval, as a result of the state
change, perform the check for instant already on reception of
CONNECTION_UPDATE_IND
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
