these list of patches add support the Starfive Visionfive2
SBC based on the JH7110 SoC. Board identifier is `visionfive2`.
Signed-off-by: Kanak Shilledar <kanakshilledar111@protonmail.com>
Signed-off-by: Pratik Farkase <pratik.farkase@wsisweden.com>
These list of files add basic support for StarFive
JH7110 SOC Device Tree includes for VisionFive2
board.
Signed-off-by: Pratik Farkase <pratik.farkase@wsisweden.com>
See https://github.com/zephyrproject-rtos/zephyr/discussions/69997
The driver thinks that when there are zero touches, the screen is
pressed. This changes it so zero touches mean the screen is not
pressed.
Also, multi-touch is now considered a single touch, and the
TOUCH_ID is checked so that only valid touches are registered.
Signed-off-by: Glenn Andrews <glenn.andrews.42@gmail.com>
On some content-types, it might be impossible to
send strings with size of zero.
Therefore empty URI should also allow cases
where strlen() is zero.
Signed-off-by: Seppo Takalo <seppo.takalo@nordicsemi.no>
Adding Fin Maaß <f.maass@vogl-electronic.com> (maass-hamburg) as
a collaborator to the DHCPv4.
Related PR #69992
Signed-off-by: Jukka Rissanen <jukka.rissanen@nordicsemi.no>
Previously, CONFIG_MAX_TIMER_COUNT and CONFIG_MSG_COUNT_MAX were
used without checking if CONFIG_TIMER and CONFIG_POSIX_MQUEUE
were enabled.
Use Z_SC_VAL_IFDEF() to conditionally define those constants
based on their gating Kconfig values, otherwise, they are defined
to -1.
Signed-off-by: Christopher Friedt <cfriedt@meta.com>
When the AUX_ADV_IND and AUX_SYNC_IND are close to each
other, the duration between them is not sufficient to
schedule a new instance of ticker to establish
synchronization. This processing time introduces latencies
detected by the prepare callback.
When the sync offset is low, schedule the start of the
reception to next periodic interval.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix missing PPI to timer start compensation and missing
inclusion of range delay in the calculation of packet
header receive timeout value.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
Fix Extended Advertising Parameter check to exclude legacy
high duty cycle directed advertising from validating the
intervals as advertising interval parameter values are not
applicable for high duty cycle directed advertising.
Signed-off-by: Vinayak Kariappa Chettimada <vich@nordicsemi.no>
This makes .config, autoconf.h, and configs.c deterministic again.
Directory listing is not deterministic, it must always be sorted.
https://reproducible-builds.org/docs/stable-inputs/
Fixes commit 61bbfb5ba25f ("scripts: introduce list_hardware.py for
listing of architectures and SoCs") in collab-hwm branch which was
squashed in mega HWMv2 commit 8dc3f85622 ("hwmv2: Introduce Hardware
model version 2 and convert devices")
SOF CI builds with both Windows and Linux and compares the outputs. This
catches practically 100% of build reproducibility issues and caught this
one too:
https://github.com/thesofproject/sof/actions/runs/8241692987/job/22539664560
HWMv2 was a "big bang" integration on both the Zephyr and SOF sides. So
this `rglob()` was a needle in a haystack but with hindsight, this issue
is really trivial to reproduce and verify:
```
apt-get install disorderfs
mkdir disorderedWorkspace/
disorderfs --shuffle-dirents=yes workspace/ disorderedWorkspace/
```
... then just build `samples/hello_world/` twice in disorderedWorkspace/
with any --board and compare the build directories.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
If no callbacks have been registered then callback_list
is NULL, in which case we can skip searching.
This also fixes a NULL-pointer acces in the while loop
as previous_callback would be NULL when it gets
dereferenced as previous_callback->_next.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
With the transition to HWMv2, `CONFIG_CPU_HAS_DCACHE` is no
longer selected. This causes issues with Sound Open Firmware
since this configuration allows the usage of DCACHE-related
cache management operations. As such, to fix said issues,
select `CONFIG_CPU_HAS_DCACHE` on all NXP ADSP SOCs.
Signed-off-by: Laurentiu Mihalcea <laurentiu.mihalcea@nxp.com>
The BIGinfo PDU bitsets were not accessed in a portable maner for
endianness independence, and need to be handled as bit-manipulations.
Add utility functions to access PDU bitset variables independent of
endianness. Use macros to abstract the offsets and widths.
Conversion of BIG_Handle to LE in HCI layer incorrectly use
sys_cpu_to_le16, which fail because target type is 8-bit.
Signed-off-by: Morten Priess <mtpr@oticon.com>
Configure a new Kconfig (`CONFIG_RISCV_HART_MASK`) so that it
is possible to mask the `mhartid` of a processor. This is
helpful in the cases when the hart id starts from non-zero
value.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Default pyocd configuration doesn't allow debugging on this board.
Now that Zephyr openocd version support stm32g0 series, use it instead
of pyocd.
This commit provides working openocd configuration for this board
an update documentation.
Signed-off-by: Erwan Gouriou <erwan.gouriou@st.com>
Check the driver configuration to disable
the USB Type-C dead battery, only if
- a USB PD system is in place (the UCPD node is enabled)
- or the user does not require USB PHY anyway
Signed-off-by: Francois Ramu <francois.ramu@st.com>
Do not use assert_called_once_with, we can some functions more than once
and test should not prevent that.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add parentheses missing around $BOARD_REVISION.
The syntax without parentheses has been deprecated for at least 5 years:
374f488734
That same documentation states "Using the old syntax with an undefined
environment variable keeps the string as is." This what actually happens
on Windows where `build/zephyr/.config` looks like this:
CONFIG_BOARD_REVISION="$BOARD_REVISION"
I found this because the behavior differs on Linux where the same,
"old" syntax produces this instead:
CONFIG_BOARD_REVISION=""
This could be because environment variables work differently?
(BOARD_REVISION is passed from kconfig.cmake to kconfiglib.py thanks to
cmake -E env)
Let's not try to debug this and just drop the deprecated syntax. It
aligns both Windows and Linux on the same, expected, empty string
behavior.
Note these are the only two `def_string` found across all Kconfig files
right now.
Fixes commit c11b7852d1 ("Kconfig: add CONFIG_BOARD_REVISION")
Fixes commit e2ff2a88ba ("sysbuild: include HWMv2 Kconfig in sysbuild")
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Set total priorities to 71.
This setting already exceeds original limit(32) and exceeds 64 to test
secondary bitmap.
Signed-off-by: TaiJu Wu <tjwu1217@gmail.com>
Modified bitmask to bitmask array, it can make multilevel queue remove
32 bit prioriry limit.
We can scan bitmask array to find which queue have ready thread.
Only need the number of queues as priority because the priority
is checked on create_thread.
Signed-off-by: TaiJu Wu <tjwu1217@gmail.com>
Split the common keyboard scanning code out of the XEC specific driver
and use the generic code instead.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Convert the XEC keyboard scanning driver from kscan to input, add the
corresponding kscan compatibility node to the current board, build test
only.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
Add a blank define for the case where DMA channels are defined in ADC node
of the dts but STM32_ADC_DMA is not enabled. Otherwise compilation fails.
Also fix the way the DMA channel is configured by using a standard DT
macro, otherwise it doesn't work for dma-v2bis DMA types.
Signed-off-by: Guillaume Gautier <guillaume.gautier-ext@st.com>
Fixes commit a9a82d557c ("llext: use elf_rela_t instead of elf_rel_t")
Also switch sign of (unused?) `r_addend` to unsigned.
https://refspecs.linuxfoundation.org/elf/gabi4+/ch4.reloc.html
Issue found thanks to the following warnings when compiling in 64bits:
```
/__w/zephyr/zephyr/include/zephyr/llext/elf.h:349:29: error:
right shift count >= width of type [-Werror=shift-count-overflow]
349 | #define ELF64_R_SYM(i) ((i) >> 32)
```
The name `elf64_word` was admittedly confusing.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Note `elf64_addr` and friends are defined as uint64_t which is defined
as `long long unsigned` which is for some reason different from
`size_t`. So they all require a cast.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
pinmux.c is inaptly named since it has nothing to do with pinmux.
Also, remove inclusion of iocon.h since this file does not use it,
as this file has nothing to do with pinmux.
Signed-off-by: Declan Snyder <declan.snyder@nxp.com>
With recently introduced DNS retransmission mechanism, a certain bug
could occur when calculating query timeout.
If the time until the final DNS timeout (as indicated by
CONFIG_NET_SOCKETS_DNS_TIMEOUT) was less than 1 millisecond, the actual
millisecond timeout value was rounded down, resulting in 0 ms timeout.
This in order was interpreted as invalid argument by dns_get_addr_info()
function, so in result, instead of reporting query timeout, the function
reported invalid argument error.
Fix this by rounding the millisecond timeout up, instead of down, so
that in any case, if the final timeout is not due, we always provide
non-zero timeout to dns_get_addr_info().
Signed-off-by: Robert Lubos <robert.lubos@nordicsemi.no>
In case the core is not power gated, waiti will restore intlevel. In
this case we lock interruption after it.
In the bug scenario, the host starts streaming and via SOF APIs, keeps a
lock to prevent Zephyr from entering PM_STATE_RUNTIME_IDLE. During the
test case, host removes this block and core0 is allowed to enter IDLE
state.
When core0 enters power gated state, interrrupts are left enabled (so
the core can be woken up when something happens). This leaves a race
where suitably timed interrupt will actually block entry to power gated
state and k_cpu_idle() in power_gate_entry() will return. This is rare,
but happens often enough that the relatively short test plan run on SOF
pull-requests will trigger this case.
Fixes#69807
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
As seen in the PR #68125 discussion, commit 19a33c7884 ("init: adjust
the SYS_INIT dev field init to play nice with older compilers") entirely
threw away designated initializers in SYS_INIT_NAMED() to avoid
compatibility issues across toolchains.
One key aspect that was probably missed at the time: C and C++ are two
different languages and this is especially true with respect to
designated initializers.
Designated initializers provide safer and more readable code, especially
in their much stricter C++ version. So use an #ifdef to restore them in
SYS_INIT_NAMED() thanks to a small braces difference between C and C++.
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Taking a spinlock will result in interrupts being blocked in the MIPI
DBI driver, which is not desired behavior while issuing SPI transfers,
since the driver may use interrupts to drive the transfer
Fixes#68815
Signed-off-by: Daniel DeGrasse <daniel.degrasse@nxp.com>
