Do not use SYS_INIT for initializing irq_offload when enabled, instead
using a new interface that is called during the boot process for all
architectures.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add a new call for initializing cache on architectures that need that.
Avoid using SYS_INIT for this and instead call the hook in a fixed place
and run if implemented.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Fill the memory of all CPU's IRQ stack with 0xAA on init, so
that `z_stack_space_get` can calculate the remaining space
correctly.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
The callee-saved-registers can be helpful to debug the state of
a core upon an exception, however, currently there's no way to
access that information in user-implemented
`k_sys_fatal_error_handler()`, even though the csf is already stored
in the stack.
This patch conditionally add a `csf` member in the `arch_esf` when
`CONFIG_EXTRA_EXCEPTION_INFO=y`*, which the `_isr_wrapper` would update
when a fatal error occurs before invoking `z_riscv_fatal_error_csf()`.
Functions such as `k_sys_fatal_error_handler()` would then be able
to access the callee-saved-registers at the time of exception via
`esf->csf`.
* For SoCs that select `RISCV_SOC_HAS_ISR_STACKING`, the
`SOC_ISR_STACKING_ESF_DECLARE` has to include the `csf` member,
otherwise the build would fail.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Get rid of the switch statement and use an string array
for the cause instead. This saves about ~600 bytes.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Rename `cause_str` to `z_riscv_mcause_str` and make it non-static,
so that it can be used in user-implemented `k_sys_fatal_error_handler`.
Also, this function should return a constant string, so add `const`
to it.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Relocate the logging of mcause & mtval from `_Fault` to
`z_riscv_fatal_error_csf` so that they are always printed
upon exception.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Introduce soc and board hooks to replace arch specific code
and replace usages of SYS_INIT for platform initialization.
include/zephyr/platform/hooks.h introduces the hooks to be implemented
by boards and SoCs.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add the possibility to disable fault handling in spurious
interrupt handler on RISCs and replacce it with an infinite loop.
Signed-off-by: Magdalena Pastula <magdalena.pastula@nordicsemi.no>
The change of alignment check in #76045 could be wrong and
isn't unnecessary to fix the stack traces output, undo it for
now.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
This implements arch_thread_priv_stack_space_get() so this can
be used to figure out how much privileged stack space is used.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds the bits to initialize the privileged stack when
a thread is transitioning to user mode. This prevents
information leaking if the stack is reused, and also aids
in calculating stack space usage during system calls.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Having `CONFIG_EXCEPTION_STACK_TRACE_SYMTAB` to select the
`CONFIG_SYMTAB` or to explicitly not print the symbol name
during exception stack unwind seems unnecessary, as the extra
code to print the symbol name is negligible when compared with
the symbol table, so just remove it.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
This commit introduces a new ARCH_STACKWALK Kconfig which
determines if the `arch_stack_walk()` is available should the
arch supports it.
Starting from RISCV, this will be able to converge the exception
stack trace implementation & stack walking features. Existing
exception stack trace implementation will be updated later.
Eventually we will end up with the following:
1. If an arch implements `arch_stack_walk()`
`ARCH_HAS_STACKWALK` should be selected.
2. If the above is enabled, `ARCH_SUPPORTS_STACKWALK` indicates
if the dependencies are met for arch to enable stack walking.
This Kconfig replaces `<arch>_EXCEPTION_STACK_TRACE`
2. If the above is enabled, then, `ARCH_STACKWALK` determines
if `arch_stack_walk()` should be compiled.
3. `EXCEPTION_STACK_TRACE` should build on top of the
`ARCH_STACKWALK`, stack traces will be printed when it
is enabled.
4. `ARCH_STACKWALK_MAX_FRAMES` will be removed as it is
replaced by `ARCH_STACKWALK_MAX_FRAMES`
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Relocate stack unwind backends from `arch/` to perf's
`backends/` folder, just like logging/shell/..
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
Implement stack trace function for riscv arch, that get required
thread register values and unwind stack with it.
Signed-off-by: Mikhail Kushnerov <m.kushnerov@yadro.com>
custom arch_cpu_idle and arch_cpu_atomic_idle implementation was done
differently on different architectures. riscv implemented those as weak
symbols, xtensa used a kconfig and all other architectures did not
really care, but this was a global kconfig that should apply to all
architectures.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Account for the scenario when we are doing `esf`-based
unwinding from a function which doesn't have any callee.
In this case the `ra` is not saved on the stack and the
second function from the top of the frame could be missing.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Signed-off-by: Yong Cong Sin <yongcong.sin@gmail.com>
According to the riscv's `arch.h`:
+------------+ <- thread.arch.priv_stack_start
| Guard | } Z_RISCV_STACK_GUARD_SIZE
+------------+
| Priv Stack | } CONFIG_PRIVILEGED_STACK_SIZE
+------------+ <- thread.arch.priv_stack_start +
CONFIG_PRIVILEGED_STACK_SIZE +
Z_RISCV_STACK_GUARD_SIZE
The start of the privilege stack should be:
`thread.arch.priv_stack_start + Z_RISCV_STACK_GUARD_SIZE`
Instead of
`thread.arch.priv_stack_start - CONFIG_PRIVILEGED_STACK_SIZE`
For the `end`, use the same equation of `top_of_priv_stack` in
the `arch_user_mode_enter()`
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
When RISCV_ALWAYS_SWITCH_THROUGH_ECALL is enabled, do_swap() enables PMP
checking in is_kernel_syscall.
If the PMP stack guard is triggered and do_swap() is called from the
fault handler, a PMP error occurs because the stack usage violates the
previous PMP setting.
Remove the stack guard setting during a stack overflow handler to allow
enabling PMP checking safely in fault handler.
Signed-off-by: Jimmy Zheng <jimmyzhe@andestech.com>
When RISCV_ALWAYS_SWITCH_THROUGH_ECALL is enabled, do_swap() enables PMP
checking in is_kernel_syscall.
If a user thread violates memory protection and do_swap() is called from
the fault handler, a PMP error occurs because the thread is in privileged
mode but still using the old user mode PMP setting.
Update the PMP setting to privileged mode for fault handler.
This also enables the stack guard for user thread's privileged stack in
fault handler.
Signed-off-by: Jimmy Zheng <jimmyzhe@andestech.com>
When 'arch_switch()' switches though Ecall, 'exception_depth' is
incorrectly added to the next thread because the current thread is updated
before arch_switch().
Add 'exception_depth' back to the previous thread when Ecall is called from
'arch_switch()'.
Signed-off-by: Jimmy Zheng <jimmyzhe@andestech.com>
Before this, stack protection would be effective only after switching to
the first thread.
Even before the first thread is created, the kernel init code uses the
IRQ stack to set things up. Let's make sure this is safeguarded as well.
This also fixes the incompatibility between CONFIG_RISCV_PMP and
CONFIG_RISCV_ALWAYS_SWITCH_THROUGH_ECALL, the later needing an exception
call to switch to the first thread and exception code assuming stack
guard is already set up in the PMP.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Update the description of the `INCLUDE_RESET_VECTOR` Kconfig so
that it is more clear to the user what it does.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Zephyr's thread stack size is not fixed, in most cases we would
need the `thread` argument to obtain the `stack_info`, unless
we are unwinding the irq stack, since that is fixed.
Otherwise we can only safely print the current `mepc` register,
unwinding the esf without the stack info of a thread can
result in undefined behavior.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Pass the current thread to `walk_stackframe()`, so that we do
not need to hardcode `_current` in `in_fatal_stack_bound()`,
which will allow it to reuse the `in_stack_bound()`
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Check if an address is in the thread stack only when
`CONFIG_THREAD_STACK_INFO` is enabled, since otherwise the
`stack_info` will not be available.
This fixes compilation error when `CONFIG_THREAD_STACK_INFO`
is explicitly disabled.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
`isr.S` depends on `CONFIG_GEN_SW_ISR_TABLE`.
Do not build it if SW ISR table is not present.
Signed-off-by: Marcin Szymczyk <marcin.szymczyk@nordicsemi.no>
Now that the unwind starts from mepc already, the symbol
name at the mepc reg is kinda redundant, so just remove it.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Created the `arch_stack_walk()` function out from the original
`z_riscv_unwind_stack()`, it's been updated to support
unwinding any thread.
Updated the stack_unwind test case accordingly.
Increased the delay in `test_fatal_on_smp`, to wait
for the the fatal thread to be terminated, as stacktrace can
take a bit more time.
Doubled the kernel/smp testcase timeout from 60 (default) to
120s, as some of the tests can take a little bit more than 60s
to finish.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Move this to a call in the init process. arch_* calls are no services
and should be called consistently during initialization.
Place it between PRE_KERNEL_1 and PRE_KERNEL_2 as some drivers
initialized in PRE_KERNEL_2 might depend on SMP being setup.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Platforms that support IPIs allow them to be broadcast via the
new arch_sched_broadcast_ipi() routine (replacing arch_sched_ipi()).
Those that also allow IPIs to be directed to specific CPUs may
use arch_sched_directed_ipi() to do so.
As the kernel has the capability to track which CPUs may need an IPI
(see CONFIG_IPI_OPTIMIZE), this commit updates the signalling of
tracked IPIs to use the directed version if supported; otherwise
they continue to use the broadcast version.
Platforms that allow directed IPIs may see a significant reduction
in the number of IPI related ISRs when CONFIG_IPI_OPTIMIZE is
enabled and the number of CPUs increases. These platforms can be
identified by the Kconfig option CONFIG_ARCH_HAS_DIRECTED_IPIS.
Signed-off-by: Peter Mitsis <peter.mitsis@intel.com>
Created `GEN_OFFSET_STRUCT` & `GEN_NAMED_OFFSET_STRUCT` that
works for `struct`, and remove the use of `z_arch_esf_t`
completely.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Make `struct arch_esf` compulsory for all architectures by
declaring it in the `arch_interface.h` header.
After this commit, the named struct `z_arch_esf_t` is only used
internally to generate offsets, and is slated to be removed
from the `arch_interface.h` header in the future.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
The CLIC requires that mtvec.base is aligned to 64 bytes.
_isr_wrapper is used as mtvec.base, so align it to 64 bytes.
Signed-off-by: Piotr Wojnarowski <pwojnarowski@antmicro.com>
Fix the dependencies of `CONFIG_EXCEPTION_STACK_TRACE`:
- Architecture-specific Kconfig, i.e.
`X86_EXCEPTION_STACK_TRACE`, will be enabled automatically
when all the dependencies are met.
- `EXCEPTION_STACK_TRACE` depends on architecture-specific
Kconfig to be enabled.
- The stack trace implementations should be compiled only if
user enables `CONFIG_EXCEPTION_STACK_TRACE`.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
For SoC with `CONFIG_RISCV_RESERVED_IRQ_ISR_TABLES_OFFSET`,
it should be taken into consideration when disconnecting IRQ.
Signed-off-by: Marcin Szymczyk <marcin.szymczyk@nordicsemi.no>
Change the type of `cpu_id` to `uint8_t` since that is the type
of `arch_curr_cpu()->id`.
Instead of using precompiler switch (`#ifdef CONFIG_SMP`), use
if-else shorthand instead (`IS_ENABLED(CONFIG_SMP)`).
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Namespaced the generated headers with `zephyr` to prevent
potential conflict with other headers.
Introduce a temporary Kconfig `LEGACY_GENERATED_INCLUDE_PATH`
that is enabled by default. This allows the developers to
continue the use of the old include paths for the time being
until it is deprecated and eventually removed. The Kconfig will
generate a build-time warning message, similar to the
`CONFIG_TIMER_RANDOM_GENERATOR`.
Updated the includes path of in-tree sources accordingly.
Most of the changes here are scripted, check the PR for more
info.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Make sure that esf is not NULL before calling
z_riscv_unwind_stack to prevent NULL pointer dereferencing.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Current on x86 & risc-v that implement stack trace, the
maximum depth of the stack trace is defined by a macro.
Introduce a new Kconfig:EXCEPTION_STACK_TRACE_MAX_FRAMES
so that this is configurable in software.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
The mepc register is the address of the instruction that was
interrupted, it will make debugging easier if we know the
name of the symbol, so print it if `CONFIG_SYMTAB` is enabled.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
Selecting `CONFIG_EXCEPTION_STACK_TRACE_SYMTAB` will
enable the symtab generation which will be used in the
stack trace to print the function name of the return
address.
Updated the `stack_unwind` test to test the symbols in a
stack trace.
Signed-off-by: Yong Cong Sin <ycsin@meta.com>
