1.correct timer register control flow
2.select timer interrupt rising edge trigger, instead of default
level trigger
Stress test: top timer fires interrupt every 300ms for 18 hours.
The result is that we don't lose any interrupts.
Signed-off-by: Ruibin Chang <Ruibin.Chang@ite.com.tw>
HID lookup table can live in Flash and save some RAM (quite a bit
actually, as it's quite large!).
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Clarify linking issues in PRs and Commit messages and address the issue
of ambiguity across multiple github repos.
Not allowing links in commit messages so that forks can be happy is not
great.
Forks on Github have no problem with linking, however,
disconnected forks seem to not link correctly, but then why does this
matter?
If someone is taking Zephyr and developing it as their own, they
can for sure do that, but why this should force the Zephyr project to
manage traceability differently and make it more difficult for
developers to add such information in the git metadata directly and not
relying on pull requests.
Most developers use git directly and rely less on pull request and
github UI when browsing changes in the code base.. We need to be able to
see by looking at git commits and git
history if a commit is associated with a bug, some PRs might fix
multiple issues, so the context is important and by looking at each
commit the trace can be established immesiately without going back to
github pull requests.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The optee suite was failing in test_suspend() after a call to
k_aligned_alloc() which returns -ENOMEM. The alignment requirement is
4096 bytes, which seems reasonable, but what is odd about that is
that the alignment fails in spite of
CONFIG_SRAM_SIZE=145131134582784
and
CONFIG_HEAP_MEM_POOL_SIZE=270044
So there is plenty of memory available for this test and alignment
considerations should not be an issue.
Removing `CONFIG_DEBUG=y` solved the problem for me.
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
Ensure that qemu_x86 (the integration platform) is on the allow list
for `sample.net.sockets.socketpair.s32k148_evb`.
```
INFO: Error found: sample.net.sockets.socketpair.s32k148_evb on
qemu_x86/atom (Not in testsuite platform allow list but is one of the
integration platforms)
```
Signed-off-by: Chris Friedt <cfriedt@tenstorrent.com>
This test fails sporadically on SMP, so make it run only on one CPU.
Related to zephyrproject-rtos/zephyr#91620
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
It seems there is no functional impact, but it makes sense to also update
the state of the interface when AP is started/stopped.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
The interface was properly set "dormant" on disconnect. However on startup,
it arrived in the system with "awake" status. Hence, some configuration
frames were sent before the interface was connected. Then, when the
interface was marked awake after the connection, the frames were not sent
again since the operational mode did not changed.
Therefore, Router Solicitations were not send, the Router Advertisements
were not received and the IPv6 address was not set.
Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Disable support to local ISR declaration on Espressif SoCs.
Code relocation is not yet supported, causing build fail.
Signed-off-by: Sylvio Alves <sylvio.alves@espressif.com>
gpio_rpi_isr() always addressed io_bank0->proc0_irq_ctrl, so any
interrupts taken while code was running on core 1 were invisible and
left pending.
Use get_core_num() to pick proc1_irq_ctrl when the ISR executes on core
1, ensuring callbacks fire from both cores.
Also fix stray `iobank0_hw` symbol for the correct `io_bank0_hw`.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Ensure gpio_set_dir() receives GPIO_IN or GPIO_OUT by mapping
INIT_{LOW,HIGH} flags explicitly, instead of passing raw bitmasks.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Fixed several occurrences of offset not being calculated in case
multiple GPIO ports are present.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Fix incorrect use of bitwise OR operator when checking pull-up and
pull-down resistor configuration.
Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
Core Specification 6.1 clarified exptected erro code in case peer
sending invalid Public Key. In case pairing is aborted during or
immediately after Public Key Exchange phase has completed (which is
the case here) expected reason code is set to "DHKey Check Failed".
This was affecting SM/CEN/KDU/BI-04-C and SM/PER/KDU/BI-04-C
qualification test cases.
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
It is not an error if the clock source is already enabled. This
may happen if a bootloader has used the peripheral and not
de-initialized it before booting the application.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
It is not an error if the clock source is already enabled. This
may happen if a bootloader has used the peripheral and not
de-initialized it before booting the application.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
It is not an error if the clock source for the UART is already
enabled. This may happen if a bootloader has used the UART and
not de-initialized it before booting the application.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
If a bootloader is used, it will configure the SMU and SAU.
The application cannot also configure it, since it can't know
what alias (S vs NS) to access the CMU through. Only connect
the interrupt to the application's vector table if a bootloader
is present.
Signed-off-by: Aksel Skauge Mellbye <aksel.mellbye@silabs.com>
The DTS for this board contained a few mistakes:
* The PHY address was incorrect
* This did not impair functionality because the RTL8211F
treats address 0 as "broadcast" and accepts commands
sent to it by default
* An incorrect PHY driver was used
* The generic "phy_mii" driver is not the most appropriate
for this PHY - use the HW-specific RTL8211F driver instead
* Also remove the PHY IRQ pin configuration from Ethernet
pinctrl to allow RTL8211F driver to receive IRQ via GPIO
instead
* The RGMII GTX CLK pinctrl must be configured at "medium-speed"
* This is a temporary solution. The proper solution would use
very-high-speed combined with a GPIO delay, but delays are
not supported yet in Zephyr.
Signed-off-by: Mathieu CHOPLAIN <mathieu.choplain@st.com>
Migrate the dma_nxp_edma driver to SDK-NG. This means:
1) Adding the CMAKE logic required for compiling the module
2) Updating the HAL_NXP manifest to pull in the patches which
handle the migration on the HAL side
3) Removing CMAKE logic from old mcux-sdk so that they are no
longer included in the build
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
This commit adds support for configuring the power/performance
profile for the siwx91x device using a generic
'power-profile' property in DeviceTree.
The property is available for NWP nodes,
allowing flexible selection of power management
profiles per application or board via overlay.
Signed-off-by: S Mohamed Fiaz <fiaz.mohamed@silabs.com>
The sample was not able to build because of a typo in
Kconfig when enabling CONFIG_CHARGER_DISCHARGE_CURRENT_NOTIFICATIONS
Signed-off-by: Matthias Alleman <matthias.alleman@basalte.be>
add timestamping on Tx to packets marked for timestamping
add timestamping on Rx to all packets for later use
fix race condidition on adding timestamp when sending delay_req
Signed-off-by: Adib Taraben <theadib@gmail.com>
