- Add support for four new ports, UART2-UART5.
- Add build options to allow console device configuration.
- Segregate device-specific definitions from the device family files. X macros
are used to maintain a single source of truth and have the configuration
done at compile-time
- Add raspberrypi_uart_init() to make it convenient for users to install uart
devices
Close#5130
Co-authored-by: Ning Yang <yangn0@qq.com>
* It is not granted that we will always access to natually aligned
access. So, before handling endianess do byte-a-byte load using
the appropriate libfdt function to avoid unaligned access issues
Signed-off-by: Francescodario Cuzzocrea <bosconovic@gmail.com>
This removes the carriage return printed before the shell prompt as it
can cause other output to be overwritten unexpectedly. This was
discovered when using the "cat" shell command on a file whose content
contained no trailing \n\r and was short enough to be entirely swallowed
by the prompt.
The rtime decompression routine does not fully check bounds during the
entirety of the decompression pass and can corrupt memory outside the
decompression buffer if the compressed data is corrupted. This adds the
required check to prevent this failure mode.
Updates #5072
This reverts commit c1aa06dbd2.
Extending the rtems_stack_checker_info like this breaks the existing
visitor API used by rtems_stack_checker_iterate(). After review, using
an information structure contradicts the intention of the reporter to
allow a minimal reporting. Filling up the information structure with
all the details would pull in several dependencies, like
_Thread_Get_name(). The API should be improved to provide the user with
the necessary information from the TCB without having to access members
directly.
This removes the xilinx and lp64 identifiers from the ZynqMP BSPs and
updates those names throughout the build system. The xilinx identifier
is being removed because it is implied by zynqmp and lp64 is being
removed because it is the default for AArch64.
This also renames grp_zu3eg.yml to grp_hardware.yml to more accurately
reflect its purpose.
Updates #5119
Now that minimum.exe is minimum.norun.exe, the tester will not attempt
to run this sample test and will therefore not categorize it as a
failure since it includes no output. BSP-level test configurations for
this test can now be removed since those configurations only existed to
prevent it from being run.
On other targets, the rtems_cache_flush_entire_data() also performs a
clean and invalidate.
This is a workaround for "Answer 68874 - Zynq UltraScale+ MPSoC, APU -
An Eviction Might Overtake A Cache Clean Operation".
Let the caller provide the translation table base and the used page
tables to ease testing.
Simplify the error handling by using early returns.
Return RTEMS_TOO_MANY instead of RTEMS_NO_MEMORY if the page tables are
all used.
There is no need to wait on secondary processors for the GIC distributor
enable since the BSPs for real targets start the secondary processors
in _CPU_SMP_Start_processor().
Whether a floating-point implementation supports trapping of
floating-point exceptions is implementation defined. At least Qemu,
Cortex-A53, and Cortex-A72 do not support them.
In SMP configurations, check that we run on a configured processor. If not,
then there is not much that can be done since we do not have a stack available
for this processor. Just loop forever in this case. Do this in assemlby to
ensure that no stack memory is used.
