Provide symbols for the creation of dynamic memory pool.
Fix the loader ROM buffers start address.
Fix static allocation size check.
Signed-off-by: Marek Matej <marek.matej@espressif.com>
Add the `CONFIG_ESP_RUNTIME_HEAP` kconfig.
This allows the memory pool to be created starting
at `z_mapped_end` ending at `_heap_sentry`.
Added choice symbol ESP_WIFI_HEAP_* to select which
heap to use in the ESP WiFi adapter module.
Add file heap.c with code to initialize the runtime heap.
Size of the pool is checked during the runtime.
Signed-off-by: Marek Matej <marek.matej@espressif.com>
This lets the SoC to select the correct kconfigs to show that
it supports coredump, and with the ability to dump privilege
stack.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Modify the test to spawn a user thread if it is running
with userspace enabled. This allows testing coredump when
the crashing thread is a user thread.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds the bits to call into architecture code to dump
the privileged stack for user threads.
The weak implementation is simply there as a stub until
all architectures have implemented the associated function.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds a interface to allow coredump to dump privileged
stack which is defined in architecture specific way.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Changed the scan option type from uint32_t to
uint8_t.
There are 2 reasons for this:
1) This reduces the size of the struct bt_le_scan_param.
Since we are now storing two copies of scan parameters
statically in the host, this is not insignficant.
2) This fixes a "hole" in the struct. There are no longer
3 empty octets between the `type` and the `options`, which
caused valgrind warnings when using `memcpy` and `memcmp`
of the struct.
Currently we only need 8 bits for the options available.
If additional options are added later, the field need
to be increased. For the above reasons some additional
refactoring my be required to avoid significant size
increases and the valgrind issue.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Add cargo's bin to the path, and print out the versioning of the tool to
make it easy to ensure the right version has been installed.
Signed-off-by: David Brown <david.brown@linaro.org>
The Buffer data was being stored in cacheable memory for
the MIMXRT1010_EVK, the caching is not handle in these
test and cause mimxrt1010 to fail, this commit moves the
memory region to DTCM and in some cases lowers the size
of the DATA Buffer since it won't fit in DTCM for MIMXRT1010_EVK
Signed-off-by: Emilio Benavente <emilio.benavente@nxp.com>
Add VGA and QVGA resolution support to enable
usage of gc2145 sensor by smaller ram capable devices.
Signed-off-by: Felipe Neves <ryukokki.felipe@gmail.com>
Demultiplexer was not ready to handle case when log message was
incomplete which was followed by other log messages. Such scenario
could occur if there was a fault that happen during logging of
a message. In that case incomplete message was followed by valid
messages (fault report) and this fault report was not handled
because processing was blocked waiting for completion of a
message which preceeded fault report.
Since it is expected that some messages may be incomplete a
garbage collection mechanism is added. When start of a message is
received timestamp is logged and list of incomplete messages
is checked for 'old' messages which persist in incomplete state
for long. When message timeouts it is closed and marked as
invalid. It unblocks processing of following messages.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
write_data function which was writing to STMESP data registers was
starting by writing words and tail was written using byte access.
However, RISCV core does not support unaligned access and on Cortex-M33
even if supported it is faster to do aligned access. Reworked
write_data to start first by writing data using byte or half word
access until data pointer is word aligned, then word access is used
and finally tail is written using byte or half word access.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
This fixes the following warning:
> unit address and first address in 'reg' (0x88000) don't match for
> /soc/flash-controller@4001e000/flash@0/partitions/partition@87000
Signed-off-by: Reto Schneider <reto.schneider@husqvarnagroup.com>
This fixes the following warning:
> unit address and first address in 'reg' (0x6a) don't match for
> /soc/sercom@42001800/atecc608a@15
Signed-off-by: Reto Schneider <reto.schneider@husqvarnagroup.com>
This fixes the following warning:
> unit address and first address in 'reg' (0x1000) don't match for
> /soc/spi@40003800/spi_nor@0/partitions/partition@10000
Signed-off-by: Reto Schneider <reto.schneider@husqvarnagroup.com>
Introduce GPIO-based driver for MIPI DBI class that allows MIPI DBI
type A and B displays to be used on general platforms.
Since each data pin GPIO can be selected individually, the bus pins are
set in a loop, which has a significant negative impact on performance.
When using 8-bit mode and all the data GPIO pins are on the same port,
a look-up table is generated to set the whole port at once as a
performance optimization. This creates a ROM overhead of about 1 kiB.
Tested 8-bit 8080 mode with ILI9486 display on nRF52840-DK board.
Signed-off-by: Stefan Gloor <code@stefan-gloor.ch>
The MIPI DBI API supports MIPI DBI controllers type A, B, and C
(except with 16 write clocks). Update the documentation accordingly.
Signed-off-by: Stefan Gloor <code@stefan-gloor.ch>
Instead of always using the SPI MIPI DBI mode (type C), look up the
mipi-mode from the device tree and only set the mode to
MIPI_DBI_MODE_SPI_4WIRE as a fallback in case the property is not given.
Signed-off-by: Stefan Gloor <code@stefan-gloor.ch>
Currently the RAM allocated for the bootloader is not
enough to use MCUBoot with crypto signatures.
This commit bumps the #defines accordingly to fix
compile errors with ecdsa_p256 and RSA.
Signed-off-by: Brandon Allen <brandon.allen@exacttechnology.com>
The Linux binding DO's and DONT's about designing DT bindings
pretty much all apply to Zephyr as well, and a lot of these issues
come up in reviews in Zephyr, which is the reason that Linux has
this page in the first place, to list common binding review topics.
Signed-off-by: Declan Snyder <declan.snyder@nxp.com>
