Instead of passing the crt1 _start function as the entry code for
auxiliary CPUs, use a tiny assembly stub instead which can avoid the
runtime testing needed to skip the work in _start. All the crt1 code
was doing was clearing BSS (which must not happen on a second CPU) and
setting the stack pointer (which is wrong on the second CPU).
This allows us to clean out the SMP code in crt1.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The kernel passes the CPU's interrupt stack expected that it will
start on that, so do it. Pass the initial stack pointer from the SOC
layer in the variable "z_mp_stack_top" and set it in the assembly
startup before calling z_mp_entry().
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The xtensa atomics layer was written with hand-coded assembly that had
to be called as functions. That's needlessly slow, given that the low
level primitives are a two-instruction sequence. Ideally the compiler
should see this as an inline to permit it to better optimize around
the needed barriers.
There was also a bug with the atomic_cas function, which had a loop
internally instead of returning the old value synchronously on a
failed swap. That's benign right now because our existing spin lock
does nothing but retry it in a tight loop anyway, but it's incorrect
per spec and would have caused a contention hang with more elaborate
algorithms (for example a spinlock with backoff semantics).
Remove the old implementation and replace with a much smaller inline C
one based on just two assembly primitives.
This patch also contains a little bit of refactoring to address the
scheme has been split out into a separate header for each, and the
ATOMIC_OPERATIONS_CUSTOM kconfig has been renamed to
ATOMIC_OPERATIONS_ARCH to better capture what it means.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
There was a bunch of dead historical cruft floating around in the
arch/xtensa tree, left over from older code versions. It's time to do
a cleanup pass. This is entirely refactoring and size optimization,
no behavior changes on any in-tree devices should be present.
Among the more notable changes:
+ xtensa_context.h offered an elaborate API to deal with a stack frame
and context layout that we no longer use.
+ xtensa_rtos.h was entirely dead code
+ xtensa_timer.h was a parallel abstraction layer implementing in the
architecture layer what we're already doing in our timer driver.
+ The architecture thread structs (_callee_saved and _thread_arch)
aren't used by current code, and had dead fields that were removed.
Unfortunately for standards compliance and C++ compatibility it's
not possible to leave an empty struct here, so they have a single
byte field.
+ xtensa_api.h was really just some interrupt management inlines used
by irq.h, so fold that code into the outer header.
+ Remove the stale assembly offsets. This architecture doesn't use
that facility.
All told, more than a thousand lines have been removed. Not bad.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
With _kernel_offset_to_nested, we only able to access the nested counter
of the first cpu. Since we are going to support SMP, we need accessing
nested from per cpu.
To get the current cpu, introduce z_arm64_curr_cpu for asm usage,
because arch_curr_cpu could not be compiled in asm code.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
This patch adds weak sys_arch_reboot() function to avoid build error
with CONFIG_REBOOT=y. Some SoC has already had own reboot function
but others (Ex. qemu boards) faced buld error.
- openisa_rv32m1: Not change
- riscv-ite: Do nothing, remove and use arch/riscv function
Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>
There is no strict reason to use assembly for the reset routine. Move as
much code as possible to C code using the proper helpers.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
The name for registers and bit-field in the cpu.h file is incoherent and
messy. Refactor the whole file using the proper suffixes for bits,
shifts and masks.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
For some unknown reason, the pagetable address for _df_tss.cr3
did not get translated from virtual to physical. However,
the translation is done if the pointer to pagetable is obtained
through reference to the first array element (instead of simply
through the name of array). Without CR3 pointing to the page
table via physical address, double fault does not work. So
fixing this by being explicit with the page table pointer.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When adding a new thread to memory domain, there is a NULL check
to figure out if a thread is being migrated to another memory
domain. However, the NULL check is AFTER physical-to-virtual
address translation which means (NULL + offset) != NULL anymore.
This results in calling reset_region() with an invalid page table
pointer. Fix this by doing the NULL check before address
translation.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When linking in virtual address space, we still need physical
addresses in SRAM to be mapped so platform can boot from physical
memory and to access structure necessary for boot (e.g. GDT and
IDT). So we need to enlarge the reserved space for page table
to accommodate this.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
We have been having the assumption that the physical memory
is identity-mapped to virtual address space. However, with
the ability to set CONFIG_KERNEL_VM_BASE separately from
CONFIG_SRAM_BASE_ADDRESS, the assumption is no longer valid.
This changes the boot code in x86 32-bit, so that once
the page table is loaded, we can proceed with executing in
the virtual address space. So do a long jump to virtual
address just before calling z_x86_prep_c. From this point on,
code execution is in virtual address space.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When linking in virtual address space, we still need physical
addressed in SRAM to be mapped so platform can boot from physical
memory and to access structure necessary for boot (e.g. GDT and
IDT). So identity maps the kernel in SRAM.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When the kernel is mapped into virtual address space
that is different than the physical address space,
the dynamic GDT generation uses the virtual addresses.
However, the GDT table is required at boot before
page table is loaded where the virtual addresses are
invalid. So make sure GDT generation is using
physical addresses.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
There is an assumption made in the page table generation code
that the kernel would occupy the same physical and virtual
addresses. However, we may want to map the kernel into
a virtual address space which differs from kernel's physical
address space. For example, with demand paging enabled on
kernel code and data, we can accommodate kernel that is
larger than physical memory size, and may want to utilize
a bigger virtual address space. So add address translation
in the gen_mmu.py script for this.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds virtual address translation to a few variables
used in crt0.S. This is needed as they are linked at
virtual addresses but before page table is loaded,
they are not available at virtual addresses and must be
referred via physical addresses.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
When feeding &z_shared_kernel_page_start directly to
Z_X86_PHYS_ADDR(), the compiler would complain array subscript
out of bound if linking in virtual address space. So cast it
into uintptr_t first before translation.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Each vector slot has room for 32 instructions. The exception context
saving needs 15 instructions already. Rather than duplicating those
instructions in each out-of-line exception routines, let's store
them directly in the vector table. That vector space is otherwise
wasted anyway. Move the z_arm64_enter_exc macro into vector_table.S
as this is the only place where it should be used.
To further reduce code size, let's make z_arm64_exit_exc into a
function of its own to avoid code duplication again. It is put in
vector_table.S as this is the most logical location to go with its
z_arm64_enter_exc counterpart.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
a0 is used as scratch register. Restore value of a0 (return address)
from stack frame before spilling registers on stack
Signed-off-by: Shubham Kulkarni <shubham.kulkarni@espressif.com>
Assert if the null pointer de-referencing detection (via DWT) is
enabled when the processor is in debug mode, because the debug
monitor exception can not be triggered in debug mode (i.e. the
behavior is unpredictable). Add a note in the Kconfig definition
of the null-pointer detection implementation via DWT, stressing
that the solution requires the core be in normal mode.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
We introduce build time asserts for
CONFIG_CORTEX_M_DEBUG_NULL_POINTER_EXCEPTION_PAGE_SIZE
to catch that the user-supplied value has, as requested
by the Kconfig symbol specification, a power of 2 value.
For the MPU-based implementation of null-pointer detection
we can use an existing macro for the build time assert,
since the region for catching null-pointer exceptions
is a regular MPU region, with different restrictions,
depending on the MPU architecture. For the DWT-based
implementation, we introduce a custom build-time assert.
We add also a run-time ASSERT for the MPU-based
implementation in ARMv8-M platforms, which require
that the null pointer exception detection page is
already mapped by the MPU.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
By design, the DebugMonitor exception is only employed
for null-pointer dereferencing detection, and enabling
that feature is not supported in Non-Secure builds. So
when enabling the DebugMonitor exception, assert that
it is not targeting the Non Secure domain.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Enable the null-pointer dereferencing detection by default
throughout the test-suite. Explicitly disable this for the
gen_isr_table test which needs to perform vector table reads.
Disable null-pointer exception detection on qemu_cortex_m3
board, as DWT it is not emulated by QEMU on this platform.
Additionally, disable null-pointer exception detection on
mps2_an521 (QEMU target), as DWT is not present and the MPU
based solution won't work, since the target does not have
the area 0x0 - 0x400 mapped, but the QEMU still permits
read access.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Implementation for null pointer exception detection feature
using the MPU on Cortex-M. Null-pointer detection is implemented
by programming an MPU to guard a limited area starting at
address 0x0. on non ARMv8-M we program an MPU region with
No-access policy. On ARMv8-M we program a region with any
permissions, assuming the region will overlap with fixed
FLASH0 region. We add a compile-time message to warn the
user if the MPU-based null-pointer exception solution can
not be used (ARMv8-M only).
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Padding inserted after the (first-stage) vector table,
so that the Zephyr image does not attempt to use the
area which we reserve to detect null pointer dereferencing
(0x0 - <size>). If the end of the vector table section is
higher than the upper end of the reserved area, no padding
will be added. Note also that the padding will be added
only once, to the first stage vector table, even if the current
snipped is included multiple times (this is for a corner case,
when we want to use this feature together with SW Vector Relaying
on MCUs without VTOR but with an MPU present).
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Additions to the null-pointer exception detection mechanism
for ARMv8-M Mainline MCUs.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Implement the functionality to detect null pointer dereference
exceptions via the DWT unit in the ARMv7-M Mainline MCUs.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
When we enable the null pointer exceptino feature (using DWT)
we include debug.c in the build. debug.c contains the functions
to configure and enable null pointer detection using the Data
Watchdog and Trace unit.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Extend the debug monitor exception handler to
- return recoverable faults when the debug monitor
is enabled but we do not get an expected DWT event,
- call a debug monitor routine to check for null pointer
exceptions.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Move the DWT utility functions, present in timing.c
in an internal cortex-m header.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Introduce the required Kconfig symbol framework for the
Cortex-M-specific null pointer dereferencing detection
feature. There are two implementations (based on DWT and
MPU) so we introduce the corresponding choice symbols,
including a choice symbol to signify that the feature
is to be disabled.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
The most common secure monitor firmware in the ARM world is TF-A. The
current release allows up to 8 64-bit values to be returned from a
SMC64 call from AArch64 state.
Extend the number of possible return values from 4 to 8.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
Instead of relying on hardcoded offset in the assembly code, introduce
the offset macros to make the code more clear.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
The current code is assuming that the SMC/HVC helpers can only be used
by the PSCI driver. This is wrong because a mechanism to call into the
secure monitor should be made available regardless of using PSCI or not.
For example several SoCs relies on SMC calls to read/write e-fuses,
retrieve the chip ID, control power domains, etc...
This patch introduces a new CONFIG_HAS_ARM_SMCCC symbol to enable the
SMC/HVC helpers support and export that to drivers that require it.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
This is fundamental enough that it better be initialized ASAP.
Many other things get initialized soon afterwards assuming the MMU
is already operational.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Location of __kernel_ram_start is too far and _app_smem .bss areas
are not covered. Use _image_ram_start instead.
Location of __kernel_ram_end is also way too far. We should stop at
_image_ram_end where the expected unmapped area starts.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
This is easier to cover multiple segments this way. Especially since
not all boundary symbols from the linker script come with a size
derrivative.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
The MT_OVERWRITE case is much more common. Redefine that flag as
MT_NO_OVERWRITE instead for those fewer cases where it is needed.
One such case is platform provided mappings. Apply them after the
common kernel mappings and use the MT_NO_OVERWRITE on them.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
