In current implementation, the flag `local_dist` will be cleared when
the distributed key frame is performed if the local is the SMP
initiator. After the distributed key is sent out, the function
`smp_pairing_br_complete()` will be called if all bits of `local_dist`
are cleared.
It causes the function `smp_pairing_br_complete()` will be called
multiple times.
Add a flag `local_distributed` to flag the all sent keys. Add only the
flag `local_distributed` is not set, preform the key distribution
frame.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
The derived LE keys are not saved to NVM. And the IRK is not added to
controller resolving list. It causes two issues,
Issue 1, the LE connection connection cannot be established if the adv
address of peer is RPA.
Issue 2, the LE keys are missing after the power reset.
For issue 1, add a function `smp_br_id_add_replace` to add LE keys.
For issue 2, check the BR bondable flag `BT_CONN_BR_NOBOND` instead of
`SMP_FLAG_BOND`.
Signed-off-by: Lyle Zhu <lyle.zhu@nxp.com>
This adds new Arm MHUv3 mailbox driver.
The Arm MHUv3 (Message Handling Unit version 3) is a hardware
block designed for inter-processor communication in SoCs.
The ARM MHUv3 can optionally support various extensions to
enable different kinds of transport to be used for communication.
At the moment only the doorbell extension is supported in the
driver.
For more information refer to Arm Message Handling Unit
Architecture
https://documentation-service.arm.com/static/65f01fbab5e3c10fe1335edf
Signed-off-by: Ziad Elhanafy <ziad.elhanafy@arm.com>
Speed things up by using GH runners, we do not need the power of our own
runners to create the test plan. This also simplifies the workflow
considerably.
PRs now will get queued immediately, ie. no need to wait for a Zephyr
hosted runner to create the plan and then wait again for another round
to get more runners for execution.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Updated the stm32f3_disco board user button logic from
active low to active high as per datasheet.
Signed-off-by: Muhammad Waleed Badar <walid.badar@gmail.com>
Change default pwm to pwm0 on ek_ra6m3 in test app pwm_loopback
to avoid pin conflict with ethernet phy.
Signed-off-by: Ta Minh Nhat <nhat-minh.ta.yn@bp.renesas.com>
Change the pwm default node to pwm0 on ek_ra6m3 to avoid
pinsel conflict with ethernet phy.
Signed-off-by: Ta Minh Nhat <nhat-minh.ta.yn@bp.renesas.com>
Use the `arduino,uno-adc` mapping to change the settings
to make it easier to understand.
Add available A4 and A5 pins.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
udc_ambiq was using USB test mode definition in dwc2 header, which
the include was removed, hence compilation now fails. This commit
fixes the compilation failure.
Signed-off-by: Chew Zeh Yang <zeon.chew@ambiq.com>
This patch allows checking for any custom ARM interrupt controller to get a
non-enabled IRQ line instead of assuming that GIC is enabled
Signed-off-by: Amneesh Singh <a-singh7@ti.com>
As above, the test verify the ability to wake up CPU
from the lowest power state. The test has been
prepared for stm32l562e_dk board.
Signed-off-by: Arkadiusz Cholewinski <arkadiuszx.cholewinski@intel.com>
This test evaluates the residency time of different power states
on the stm32l562e_dk board and calculates the
RMS (Root Mean Square) current for every detected state.
Signed-off-by: Arkadiusz Cholewinski <arkadiuszx.cholewinski@intel.com>
During the test, pytest detects each step and calculates
the RMS (Root Mean Square) current for every detected state.
The test then compares the calculated RMS values
against the expected values.
Signed-off-by: Arkadiusz Cholewinski <arkadiuszx.cholewinski@intel.com>
Add support for 8-channel configurable DMA controller. The driver
supports the following features:
- memory to peripheral (ch0 to ch5)
- peripheral to memory (ch0 to ch5)
- memory to memory (ch6 and ch7)
Each DMA channel is multiplexed between two or more trigger sources:
- ch0 -> SPI0_TX or UART0_RX
- ch1 -> SPI0_RX or UART0_TX
- ch2 -> LRFD or UART0_TX
- ch3 -> ADC0 or UART0_RX
- ch4 -> AES_A or LRFD
- ch5 -> AES_B or ADC0
- ch6 -> Software Event Channel 0
- ch7 -> Software Event Channel 1
Signed-off-by: Julien Panis <jpanis@baylibre.com>
This test fails on older python versions (3.10) and only on CI.
Disabling it while we investigate. The test itself verifies inline logs
options, so the functionality test is not impacted.
Tracked in #87769
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Pin python dependencies to hashes and cleanup/unify python setup steps in
various workflows.
We now have one dependency file containing all requirements for github
actions that is managed centrally with hashes. No direct pip installs
are needed in workflow files and everything shall go via the
requirements file.
Pinning to specific version and hashes helps with preventing supply
chain attacks.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Providing the right config options and a new test scenario,
coredump_backends test will run under the in-memory backend.
Signed-off-by: Tomasz Bursztyka <tobu@bang-olufsen.dk>
For testing purposes only. Testing the fatal error handler or
a coredump backend. Or just out of curiosity.
Signed-off-by: Tomasz Bursztyka <tobu@bang-olufsen.dk>
Only meant to be used on platforms that would not wipe the RAM on reboot.
This won't work if the platform is halted and restarted as RAM is
volatile, obviously. A relevant fatal error handler needs to be provided
to reboot.
Signed-off-by: Tomasz Bursztyka <tobu@bang-olufsen.dk>
This might come in handy for other coredump backends too, not only flash
one. Obviously: changing the logic to use the coredump generic API instead
of the flash backend's internal functions.
Adding a better output, following how the coredump is actually created:
parsing and printing headers, and their respective content.
Signed-off-by: Tomasz Bursztyka <tobu@bang-olufsen.dk>
Reintroduced by commit 8c9af42a63
("portenta_h7: enable onboard regulator") and becomes annoying to
remove again and again. There is also no test feature usb_cdc.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
get_maintainer.py does not support recursive glob.
In the future, get_maintainer.pl could take advantage of
glob.translate()[1]. Unfortunately, glob.translate() is only supported
from Python 3.14 which is not yet widely adopted.
Meanwhile, just avoid to use recursive globs.
[1]: https://docs.python.org/3/library/glob.html#glob.translate
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
The `tests/subsys/logging/log_stack` test no longer exists.
Remove a reference to the test from the document.
Signed-off-by: TOKITA Hiroshi <tokita.hiroshi@gmail.com>
Add missing entry for DAI_NXP_MICFIL, which was accidentally removed
during the transition to SDK-NG.
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
The new SDK-NG is missing some EDMA-related headers/drivers because of
which the 'dma_nxp_edma' driver no longer compiles with it. Mark the new
SDK as unsupported for this platform until this is fixed.
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
Fix incorrect payload count at CIS Establish due to existing
CIG event overlapping the ACL event at the instant when the
CIS gets the active flag set.
The overlapping CIG event picked up the new CIS that had its
active flag set in the current CIG event instead of at the
actual CIS offset which is in the next CIG event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix CIS offset calculation due to use of decremented ACL
event counter, where as the CIS offset is inquired in the
next ACL event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix incorrect CIS offset in use if instant is picked from
the peer sent CIS RSP PDU. Instead, keep the instant that
was sent in the CIS REQ PDU as the instant to send in the
CIS IND PDU.
This fixes CIS failed to be established when dissimilar
ACL and ISO intervals are in use.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix incorrect payload count at CIS Establish due to existing
CIG event overlapping the ACL event at the instant when the
CIS gets the active flag set.
The overlapping CIG event picked up the new CIS that had its
active flag set in the current CIG event instead of at the
actual CIS offset which is in the next CIG event.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix Peripheral CIS sorted by CIG implemenation to use CIS
offset stored in LLL context which is the correct offset
from the CIG anchor point. CIS offset in the ULL context
is the offset from the ACL anchor point at the time of
the CIS establishment.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix CIS event_count_prepare use missed as part of fixes
related to commit be91cfedfb ("Bluetooth: Controller: Fix
incorrect event_count when CIG overlaps").
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
- Sets LinkServer as the default runner for lpcxpresso55sXX boards.
- Unifies the "Configuring a Debug Probe" chapter
for lpcxpresso55sXX boards.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
- Fixes declared memory region sizes in .yaml and .rst.
- Adds SRAMX linker region.
- The lpcxpresso55s36 linker memory regions are:
FLASH =246KB, RAM=112KB, SRAMX=16KB.
Signed-off-by: Andrej Butok <andrey.butok@nxp.com>
