Statically initialize the interrupt stack area
(_Configuration_Interrupt_stack_area_begin,
_Configuration_Interrupt_stack_area_end, and
_Configuration_Interrupt_stack_size) via <rtems/confdefs.h>. Place the
interrupt stack area in a special section ".rtemsstack.interrupt". Let
BSPs define the optimal placement of this section in their linker
command files (e.g. in a fast on-chip memory).
This change makes makes the CPU_HAS_SOFTWARE_INTERRUPT_STACK and
CPU_HAS_HARDWARE_INTERRUPT_STACK CPU port defines superfluous, since the
low level initialization code has all information available via global
symbols.
This change makes the CPU_ALLOCATE_INTERRUPT_STACK CPU port define
superfluous, since the interrupt stacks are allocated by confdefs.h for
all architectures. There is no need for BSP-specific linker command
file magic (except the section placement), see previous ARM linker
command file as a bad example.
Remove _CPU_Install_interrupt_stack(). Initialize the hardware
interrupt stack in _CPU_Initialize() if necessary (e.g.
m68k_install_interrupt_stack()).
The optional _CPU_Interrupt_stack_setup() is still useful to customize
the registration of the interrupt stack area in the per-CPU information.
The initialization stack can reuse the interrupt stack, since
* interrupts are disabled during the sequential system initialization,
and
* the boot_card() function does not return.
This stack resuse saves memory.
Changes per architecture:
arm:
* Mostly replace the linker symbol based configuration of stacks with
the standard <rtems/confdefs.h> configuration via
CONFIGURE_INTERRUPT_STACK_SIZE. The size of the FIQ, ABT and UND
mode stack is still defined via linker symbols. These modes are
rarely used in applications and the default values provided by the
BSP should be sufficient in most cases.
* Remove the bsp_processor_count linker symbol hack used for the SMP
support. This is possible since the interrupt stack area is now
allocated by the linker and not allocated from the heap. This makes
some configure.ac stuff obsolete. Remove the now superfluous BSP
variants altcycv_devkit_smp and realview_pbx_a9_qemu_smp.
bfin:
* Remove unused magic linker command file allocation of initialization
stack. Maybe a previous linker command file copy and paste problem?
In the start.S the initialization stack is set to a hard coded value.
lm32, m32c, mips, nios2, riscv, sh, v850:
* Remove magic linker command file allocation of initialization stack.
Reuse interrupt stack for initialization stack.
m68k:
* Remove magic linker command file allocation of initialization stack.
Reuse interrupt stack for initialization stack.
powerpc:
* Remove magic linker command file allocation of initialization stack.
Reuse interrupt stack for initialization stack.
* Used dedicated memory region (REGION_RTEMSSTACK) for the interrupt
stack on BSPs using the shared linkcmds.base (replacement for
REGION_RWEXTRA).
sparc:
* Remove the hard coded initialization stack. Use the interrupt stack
for the initialization stack on the boot processor. This saves
16KiB of RAM.
Update #3459.
The aim of this clock driver hook was to stop clock tick interrupts at
some late point in the exit() procedure.
The use of atexit() pulls in malloc() which pulls in errno. It is
incompatible with the intention of the
CONFIGURE_DISABLE_NEWLIB_REENTRANCY configuration option.
The exit() function must be called from thread context, so accompanied
clock tick interrupts should cause no harm. On the contrary, someone
may assume a normal operating system operation, e.g. working timeouts.
Remove the Clock_driver_support_shutdown_hardware() clock driver hook.
Close#3436.
Use the standard ARMv7-M systick module for the ARMv7-M CPU counter
instead of DWT counter since the DWT counter is affected by power saving
states.
Use an inline function for _CPU_Counter_difference() for all ARM BSPs.
Update #3456.
Add rtems_counter_frequency() API function. Use it to initialize the
counter value converter via the new system initialization step
(RTEMS_SYSINIT_CPU_COUNTER). This decouples the counter implementation
and the counter converter. It avoids an unnecessary pull in of the
64-bit integer division from libgcc.
Update #3456.
All tools should be removed from the RTEMS source repository at some
point in time. Tools with a BSD-style license will be moved to the
RTEMS tools repository. Unfortunately, this tool has no license
information.
Remove all uses of this tool from the code base. Users of HEX files
should consider to use ELF instead.
Close#3379.
All tools should be removed from the RTEMS source repository at some
point in time. Tools with a BSD-style license will be moved to the RTEMS
tools repository. Unfortunately, the shgen tool is GPL licensed.
Remove all uses of this tool from the code base. Replace generated files
with stub functions. If users of this BSP still exist, they can
reimplement the functionality using a BSD-style license.
Close#3443.
The ppc405 define must be checked before the ppc403 define. The ppc405
define is provided by <bspopts.h>. The ppc403 define is provided by GCC
as a built-in define if no ppc* or mpc* define is set via the
command line (see GCC sources "gcc/config/rs6000/rtems.h").
Update #3425.
Quite a number of GMAC_... functions had a side effect of enabling the
transmit and receive. During the initialization of the ethernet driver,
that lead to the situation that the receive has been already enabled
while the DMA isn't set up correctly. Therefore the DMA interpreted the
values at address 0 as an DMA descriptor. When a packet is received
during that time, the behaviour is quite undefined.
This allows a user to overwrite some MPU settings. That can for example
be usefull for devices with special requirements that are connected to
the external bus interface.
This reverts commit 40c623a883.
The use of postfix wildcards, e.g. of the form "*.x" is dangerous since
it circumvents the standard matching rules for sections.
Unknown input sections should be added explicitly to the desired output
section via "x.*" wildcards.
Update #3307.
