Add Force MD bit feature wherein connection events can be
extended to match the rate at which applications provide
new Tx data. MD bit in Tx PDUs is forced on if there has
been CONFIG_BT_CTLR_TX_BUFFERS count number times Tx/Rx
happened in a connection event. The assumption here is,
if controller's tx buffers are full, it is probable that
the application would send more Tx data during the
connection event, and keeping the connection event alive
will help improve overall throughput.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add force Md bit feature wherein connection events can be
extended to match the rate at which applications provide
new Tx data.
Fixes#27981.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix the check that decides to close a connection event,
which was missing a check on MD bit being set for empty PDU
to be Tx-ed out.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Optimize the Tx PDU preparation, empty PDU only needs MD bit
to be modified, other fields be initialised only at power up.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix premature connection event close due to the new Tx
buffers not being de-multiplexed and routed to connection's
Lower Link Layer context when they arrive while being
inside the connection's radio event.
Also, fix master prepare to demux and enqueue two Tx buffers
so that MD bit can be set correctly.
Relates to #27981.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Refactor ull_conn_tx_ack function as it no longer needs to
return the connection context back to caller.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix the redundant PDU transmission when the new Tx PDU
buffer is enqueued after MD bit value of zero was
transmitted previously in a connection event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
In case where ull_adv::disable() is disallowed, disable_mark is
erroneously re-mark instead of un-marked
Signed-off-by: Erik Brockhoff <erbr@oticon.com>
Fix for handling REJECT_IND PDU received for PHY Update,
Connection Parameter Request and Data Length Update control
procedures.
If a link layer control procedure collision occurs, example
with local initiated PHY Update Procedure, and peer sends a
REJECT_IND PDU, then the PHY Update Procedure is stalled.
Fixes#28282.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix Kconfig dependencies for enabling the controller PHY options
for LE 2M Phy and LE Coded Phy. These options should be selectable
without connections enabled when extended advertising has been enabled.
The issue affects out-of-tree controllers that wish to use the
BT_CTLR_PHY_CODED and BT_CTLR_PHY_2M options.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Select the controller feature support for data length and LE 2M PHY
based on the SoC hardware capabilities instead of relying on SoC
family.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Now that CONFIG_BT_CTLR_RX_PRIO_STACK_SIZE defines a stack size, allow
for it to be configurable from a .conf file just like the rest of the
threads in Bluetooth.
Fixes#27768.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
When controller is built with extended advertising support but without
LL Privacy build fails due to issue with code that sets TxAdd in aux
PDU:
.../ll_sw/ull_adv_aux.c: In function 'ull_adv_aux_hdr_set_clear':
.../ll_sw/ull_adv_aux.c:502:25: error: 'struct ll_adv_set' has no
member named 'own_addr_type'
502 | sec_pdu->tx_addr = adv->own_addr_type & 0x1;
|
We need own_addr_type value if an advertising set with aux but without
AD is created since we cannot get TxAdd value from other PDU.
ull_adv_aux_hdr_set_clear() can now be called with a flag to explicitly
set AdvA field and pass own_addr_type in 'value' parameter. This allows
to set proper TxAdd when creating new advertising set.
On aux updates, we use TxAdd from other PDU:
- if primary PDU has AdvA, we use TxAdd from that PDU
- if previous secundary PDU has AdvA, we use TxAdd from that PDU
In other case we just return an unpsecified error since this should not
happen in any valid scenario.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
Added a Kconfig option to enable scanner with unreserved
scan window when in continuous scan mode. This will permit
scanner to be always pre-empted by other roles, making it
the lowest in priority when it comes to using the radio.
Fixes#27414.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Remove the BT_LL_SW_SPLIT dependency for
BT_CTLR_DEBUG_PINS_CPUAPP when building CPUAPP application
that is using Zephyr BLE HCI controller on the CPUNET core.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add a name to the Kconfig choice that selects the controller TX power.
This allows board `*.defconfig` files to change the default TX power
as their certifications warrant. For example:
```
choice BT_CTLR_TX_PWR
default BT_CTLR_TX_PWR_PLUS_4
endchoice
```
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Preserve the TX power set with the `BT_HCI_OP_VS_WRITE_TX_POWER_LEVEL`
command for advertising and scanning procedures. As the TX power is no
longer overwritten on procedure start, the TX powers can now be set
before starting the procedure. This eliminates the chance for the first
transmissions to use the default TX power instead of the request TX
power.
Similar changes are not made to the connection advertising powers as
changing the TX power for a connection to device A should not change
the TX power for a future connection to device B using the same
`struct lll_adv` instance.
Fixes#27921.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
ADI field is optional in AUX_SCAN_RSP so by default it's disabled, but
now we'll have an option to include it.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
When aux is upated, DID is changed and unlike SID it's only stored in
aux PDU. To allow quick access to new DID value, let's just optionally
return current ADI value after aux was updated. It will be useful e.g.
when updating scan response data if we want to put ADI there.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
This enables processing of AUX_SCAN_REQ and sending back AUX_SCAN_RSP
on non-connectabe scannable advertising instances. Similar path will
be used for AUX_CONNECT_REQ so there are few references for connectable
as well, but this is not supported as for now.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
This adds restrictions for setting scan response data as specified by
Core spec:
* on legacy, only complete data can be set
* on legacy, data cannot be longer than 31 octets
* on ext, if non-scannable, data can only be discarded
* if ext enabled, only complete data can be set
* if ext enabled, data cannot be discarded
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
This allows to set scan response data using ext adv commands.
New AUX_SCAN_RSP can be created without analyzing previous PDU since
the only value that need to be copied from previous PDU is AdvA which
is placed at fixed location. TxPower is omitted since, if enabled, it
is already present in AUX_ADV_IND and does not need to be presed in
other PDUs. ADI support is to be added later.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
AdvA in aux is mandatory for any connectable and scannable set so make
sure it's always added.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
Connectable and scannable instances always have aux so we need to force
AuxPtr to be included in primary PDU and have aux allocated even if no
data are set.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
Connectable and scannable instances always have AuxPtr so we can return
error immediately if controller is configured with no aux.
Signed-off-by: Andrzej Kaczmarek <andrzej.kaczmarek@codecoup.pl>
Prevent redundant collision resolves and potential incorrect and harmful
increment of lazy_current.
During collision scenarios with high CPU load, the ticker_worker may be
called a second time before the ticker_job gets to run. This will cause
ticker operations on the previosly expired node (expected), and in some
cases increment lazy_current, even though the node was not sceduled to
execute via the req/ack mechanism.
By moving the request check before collision resolve, CPU time is saved,
and lazy_current will not incorrectly be incremented if node is in
collision.
The problem may be seen as a connection suddenly not receiving packets,
or MIC error on master, because the event counter goes out of sync.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Use a helper config BT_CTLR_CONN_RSSI_SUPPORT so that it is only
enabled when supported by the controller.
Signed-off-by: Rubin Gerritsen <rubin.gerritsen@nordicsemi.no>
Now that device_api attribute is unmodified at runtime, as well as all
the other attributes, it is possible to switch all device driver
instance to be constant.
A coccinelle rule is used for this:
@r_const_dev_1
disable optional_qualifier
@
@@
-struct device *
+const struct device *
@r_const_dev_2
disable optional_qualifier
@
@@
-struct device * const
+const struct device *
Fixes#27399
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
A build assert in dummy.c lists the following requirement:
[...] receive thread priority shall be higher than the Bluetooth
Host's Tx and the Controller's receive thread priority.
This is required in order to dispatch Number of Completed Packets
event before any new data arrives on a connection to the Host threads.
The drivers uses a priority that is equal to the Host TX thread,
and since they don't use the CONFIG define that is only available
to the controller then this BUILD_ASSERT will not catch the
requirement.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Add the nRF52805-specific file with radio related defintions.
Use also specific configurations for certain PPI channels used by
the controller, as this SoC has limited number of programmable PPI
channels.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
Change hard-coded (D)PPI channel bitfield values to the BIT() macro
calls with the corresponding channel base value supplied as parameter,
to make it possible to actually change this base value if needed.
Remove unused *_EXCLUDE macros.
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
nRF5340 does have SWI peripheral, hence use it instead of
the single available EGU peripheral. Use of SWI will allow
controller's LLL, ULL_HIGH and ULL_LOW execution context to
be independently configured to different interrupt priority
levels.
When ULL_HIGH priority equals ULL_LOW priority, only SWI2
is used by controller. Otherwise, SWI3 is used for ULL_LOW.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Detect and handle AD data set race condition between thread
and ISR context.
Fixes#27637.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Add cpu_dsb() hal interface so that data synchronization
barrier be used in ARM Cortex M4 and M33 architecture
implementation.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Set the AD data double buffer first index as volatile
because it is modified in LLL ISR context.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
