Make the assembly codes compatible with both GNU
and Metaware toolchain.
* replace ".balign" with ".align"
".align" assembler directive is supposed by all
ARC toolchains and it is implemented in a same
way across ARC toolchains.
* replace "mov_s __certain_reg" with "mov __certain_reg"
Even though GCC encodes those mnemonics and even real
HW executes them according to PRM these are restricted
ones for mov_s and CCAC rightfully refuses to accept
such mnemonics. So for compatibility and clarity sake
we switch to 32-bit mov instruction which allows use
of all those instructions.
* Add "%%" prefix while accessing registers from inline
ASM as it is required by MWDT.
* Drop "@" prefix while accessing symbols (defined in C
code) from ASM code as it is required by MWDT.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
/#
* update comments to match latest codes
* add extra comments for some assembly, macros
* use macro to replace duplcated codes
* remove unused codes, lables, symobols
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
overhaul the thread switch code in epilogue of irq and
exception handling:
* add z_arch_get_next_switch_handle to call z_get_next_switch_handle,
let the scheduler to decide the switch thread. This will also cover
the case of SMP.
* put lots of common codes in macros for thread switch to improve
the maintainablity, readability.
* clean up of some lables to make codes easier to understand
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
blt is signed comparsion, if r6 is a negative number created by
malicious code, it will pass the check, bring a secure risk.
use blo (unsinged comparison) to do the check.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* for COOP_SCHED case, i.e., PREEMPT_ENABLED is not enabled, the
idle thread will block other threads which is not correct.
* remove the check of PREEMPT_ENABLED in the epilogue of irq and
exception handling. Let the scheduler (should_preempt()) decide
whether the thread should be preempted.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* remove irq lock/unlock which is not needed because of
the protection of offload_sem in irq_offload
* simplify the assembly codes related irq_offload, remove
the thread switch logic
* the old codes may do thread switch in the epilogue of
irq_offload handling with int locked, this is not correct
may cause irq_offload related codes crash.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
These are not part of the generic kernel to
architecture interface, rename appropriately to
reflect they are ARC-specific.
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
after recent changes in zephyr's fault handling, e.g. use log
to repace printk, it requires more stack to exception handling, or
the stack overflow may happen and crash the system.
this commit adds a kconfig option for exception stack size with
a larger default size.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* the old codes may not save the caller saved regs correctly,
e.g. r7- r12. Because the sys call entry is called in the form
of static inline function. The compiler optimizations may not save
all the caller saved regs.
* new codes use the irq stack frame as the sys call frame and gurantee
all the called saved regs are pushed and popped correctly.
* the side effect of new codes are more stack operations and a little
overhead.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
For the old codes, if nest interrupts come out after _isr_wrapper
and before _check_nest_int_by_irq_act, then multi-bits in irq_act
will be set, this will result irq stack will not be switched in
correctly
As a fix, it's still need to use nest interrupt counter to do
interrupt stack switch as before
The difference is in the past exc_nest_count is used, but here
_kernel.nested/_kernel.cpus[cpu_id].nested is used.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
For arc processor equiped with secureshield, SEC_STAT.IRM
bit should be recorded, it determins which mode irq should
return
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* it's based on ARC SecureShield
* add basic secure service in arch/arc/core/secureshield
* necesssary changes in arch level
* thread switch
* irq/exception handling
* initialization
* add secure time support
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
exception, different with irq offload, may be raised interrupt
handling, e.g.
* z_check_stack_sentinel
* wrong code
we need to add specific handling of this case in exception handling
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
after appling the new "_get_curr_cpu_irq_stack" in _exc_entry,
the caculation of exception stack is wrong, this will
cause stack overflow, make the exception handling corrupt.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* modify the reset flow for SMP
* add smp related initialization
* implement ipi related functions
* implement thread switch in isr/exception
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* use IRQ_ACT to check nest interrupt
* implement an asm macro for nest interrupt check
* no need to use exc_nest_count, remove it
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* as ilink has a copy in ERET, it can be reused as a gp
* use ilink to do the job of arc_exc_saved_sp to save 4 bytes
and save some cycles because no load/store of memory
* it will make code scalable, e.g. for SMP, no need to
define variables for each core
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* here use new style z_arch_switch,i.e. CONFIG_USE_SWITCH
to replace old swap mechnism.
* it's also required by SMP support
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
Rename reserved function names in arch/ subdirectory. The Python
script gen_priv_stacks.py was updated to follow the 'z_' prefix
naming.
Signed-off-by: Patrik Flykt <patrik.flykt@intel.com>
* mpu stack guard exception is a kind of mpu violation exception
* use SW way to distingusih it from other mpu vioation exception
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
stack check bit of status32/sec_stat will be cleared
automically in exception entry.
so remove the redundent codes
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
There exist two symbols that became equivalent when PR #9383 was
merged; _SYSCALL_LIMIT and K_SYSCALL_LIMIT. This patch deprecates the
redundant _SYSCALL_LIMIT symbol.
_SYSCALL_LIMIT was initally introduced because before PR #9383 was
merged K_SYSCALL_LIMIT was an enum, which couldn't be included into
assembly files. PR #9383 converted it into a define, which can be
included into assembly files, making _SYSCALL_LIMIT redundant.
Likewise for _SYSCALL_BAD.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
This code path for returning from an exception wasn't
updating ERET with ESF->pc, resulting in any updates to
the PC by the fault handler being ignored.
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
* re-use top of isr stack as exception stack
* bug fixes in irq offload's implementation
* improve kernel oops implementation
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* use a separate stack for exception handling, this
will gurantee the exception handling always work, not
affected by some speical cases, e.g., stack check exception
or mem. protection exception at the exception entry.
* this commit can fix the second case of #8092
* note: the thread switch is still possible in exception
handling, e.g, caused by thread_abort. But the switched out
thread cannot be recovered, as the thread context is not
setup.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* blink, fp should not be always cleared
* clean up code format
* use a better way to save and restore ER_SEC_STAT
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* current codes use a faked interrupt return to do a thread switch
in exception return.
* so the different between exception return and interrupt return
should be carefully considered.
* when secure is enabled, the sec_stat should also be pushed in
exception entry.
* when secure is enabled, SEC_STAT.IRM must be configured corrrectly
for interrupt return.
* For the case of faked interrupt return in exception return, the
correct value of SEC_STAT.IRM comes from ER_SEC_STAT.ERM.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
The original exception handling has space to optimize and
and some bugs need to be fixed.
* define NANO_ESF
* add the definition of NANO_ESF which is an irq_stack_frame
* add the corresponding codes in exception entry and handler
* remove _default_esf
* implement the _ARCH_EXCEPT
* use trap exception to raise exception by kernel
* add corresponding trap exception entry
* add _do_kernel_oops to handle the exception raised by
_ARCH_EXCEPT.
* add the thread context switch in exception return
* case: kernel oops may raise thread context switch
* case: some tests will re-implement SysFatalHandler to raise
thread context switch.
* as the exception and isr are handled in kernel isr stack, so
the thread context switch must be in the return of exception/isr
, and the exception handler must return, should not be decorated
with FUNC_NORETURN
* for arc, _is_in_isr should consider the case of exception
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
when USERSPACE is enabled, exception is handled in the privilege
stack of thread. This make thread context switch is possible in the
exception handler. For some case,e.g. tests, this is useful.
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* add the implementation of syscall
* based on 'trap_s' intruction, id = 3
* add the privilege stack
* the privilege stack is allocted with thread stack
* for the kernel thread, the privilege stack is also a
part of thread stack, the start of stack can be configured
as stack guard
* for the user thread, no stack guard, when the user stack is
overflow, it will fall into kernel memory area which requires
kernel privilege, privilege violation will be raised
* modify the linker template and add MPU_ADDR_ALIGN
* add user space corresponding codes in mpu
* the user sp aux reg will be part of thread context
* When user thread is interruptted for the 1st time, the context is
saved in user stack (U bit of IRQ_CTLR is set to 1). When nest
interrupt comes, the context is saved in thread's privilege stack
* the arc_mpu_regions.c is moved to board folder, as it's board
specific
* the above codes have been tested through tests/kernel/mem_protect/
userspace for MPU version 2
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
* add nested interrupt support for interrupts
+ use a varibale exc_nest_count to trace nest interrupt and exception
+ regular interrupts can be nested by regular interrupts and fast
interrupts
+ fast interrupt's priority is the highest, cannot be nested
* remove the firq stack and exception stack
+ remove the coressponding kconfig option
+ all interrupts (normal and fast) and exceptions will be handled
in the same stack (_interrupt stack)
+ the pros are, smaller memory footprint (no firq stack), simpler
stack management, simpler codes, etc.. The cons are, possible
10-15 instructions overhead for the case where fast irq nests
regular irq
* add the case of ARC in test/kernel/gen_isr_table
Signed-off-by: Wayne Ren <wei.ren@synopsys.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The .balign directives were not working correctly in their
previous positions as the directive was applying to the section
before the variable's section, causing in some builds the
variables to be misaligned, and accesses to them causing faults.
With the alignments after the section declaration, the variables
will now be aligned as specified. Any future variable declarations
should use this form instead to ensure proper alignment.
Signed-off-by: Michael R Rosen <michael.r.rosen@intel.com>
This avoids asm files from having to explicitly define the _ASMLANGUAGE
symbol themselves.
Change-Id: I71f5a169f75d7443a58a0365a41c55b20dae3029
Signed-off-by: Benjamin Walsh <walsh.benj@gmail.com>
Replace the existing Apache 2.0 boilerplate header with an SPDX tag
throughout the zephyr code tree. This patch was generated via a
script run over the master branch.
Also updated doc/porting/application.rst that had a dependency on
line numbers in a literal include.
Manually updated subsys/logging/sys_log.c that had a malformed
header in the original file. Also cleanup several cases that already
had a SPDX tag and we either got a duplicate or missed updating.
Jira: ZEP-1457
Change-Id: I6131a1d4ee0e58f5b938300c2d2fc77d2e69572c
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Force-align all variables defined via asm .word to ensure 4-byte
alignment.
The straddled_tick_on_idle_enter variable was a bool, which resolved in
an one-byte quantity. Changing it to a 32-bit integer. It would have
occupied 4 bytes anyway with alignment.
Fixes ZEP-1549.
Change-Id: If5e0aa1a75dbc73d896b44616f059d221fe191c6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
These two fields in the thread structure control the preemptibility of a
thread.
sched_locked is decremented when the scheduler gets locked, which means
that the scheduler is locked for values 0xff to 0x01, since it can be
locked recursively. A thread is coop if its priority is negative, thus
if the prio field value is 0x80 to 0xff when looked at as an unsigned
value.
By putting them end-to-end, this means that a thread is non-preemptible
if the bundled value is greater than or equal to 0x0080. This is the
only thing the interrupt exit code has to check to decide to try a
reschedule or not.
Change-Id: I902d36c14859d0d7a951a6aa1bea164613821aca
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Some thread fields were 32-bit wide, when they are not even close to
using that full range of values. They are instead changed to 8-bit fields.
- prio can fit in one byte, limiting the priorities range to -128 to 127
- recursive scheduler locking can be limited to 255; a rollover results
most probably from a logic error
- flags are split into execution flags and thread states; 8 bits is
enough for each of them currently, with at worst two states and four
flags to spare (on x86, on other archs, there are six flags to spare)
Doing this saves 8 bytes per stack. It also sets up an incoming
enhancement when checking if the current thread is preemptible on
interrupt exit.
Change-Id: Ieb5321a5b99f99173b0605dd4a193c3bc7ddabf4
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
This will allow for an enhancement when checking if the thread is
preemptible when exiting an interrupt.
Change-Id: If93ccd1916eacb5e02a4d15b259fb74f9800d6f4
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
The way the ready thread cache was implemented caused it to not always
be "hot", i.e. there could be some misses, which happened when the
cached thread was taken out of the ready queue. When that happened, it
was not replaced immediately, since doing so could mean that the
replacement might not run because the flow could be interrupted and
another thread could take its place. This was the more conservative
approach that insured that moving a thread to the cache would never be
wasted.
However, this caused two problems:
1. The cache could not be refilled until another thread context-switched
in, since there was no thread in the cache to compare priorities
against.
2. Interrupt exit code would always have to call into C to find what
thread to run when the current thread was not coop and did not have the
scheduler locked. Furthermore, it was possible for this code path to
encounter a cold cache and then it had to find out what thread to run
the long way.
To fix this, filling the cache is now more aggressive, i.e. the next
thread to put in the cache is found even in the case the current cached
thread is context-switched out. This ensures the interrupt exit code is
much faster on the slow path. In addition, since finding the next thread
to run is now always "get it from the cache", which is a simple fetch
from memory (_kernel.ready_q.cache), there is no need to call the more
complex C code.
On the ARM FRDM K64F board, this improvement is seen:
Before:
1- Measure time to switch from ISR back to interrupted task
switching time is 215 tcs = 1791 nsec
2- Measure time from ISR to executing a different task (rescheduled)
switch time is 315 tcs = 2625 nsec
After:
1- Measure time to switch from ISR back to interrupted task
switching time is 130 tcs = 1083 nsec
2- Measure time from ISR to executing a different task (rescheduled)
switch time is 225 tcs = 1875 nsec
These are the most dramatic improvements, but most of the numbers
generated by the latency_measure test are improved.
Fixes ZEP-1401.
Change-Id: I2eaac147048b1ec71a93bd0a285e743a39533973
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
When going into DEEP_SLEEP mode, the ARC core now saves
its context. This includes:
- All core registers
- Stack pointer
- Program counter (restored by jumping to the restore code)
The arc reset code now checks if the GPS0 bit 2 is set.
This is similar to the behavior of the x86 core done by
the QMSI bootloader which is setting GPS0 bit 1 in order
to call the restore path instead of cold boot path.
The sample has been adapted in order to support the ARC.
Jira: ZEP-1222
Change-Id: I375f03b16b8a5fd1f07ead55cf7e4947d6290c9f
Signed-off-by: Julien Delayen <julien.delayen@intel.com>
There was a lot of duplication between architectures for the definition
of threads and the "nanokernel" guts. These have been consolidated.
Now, a common file kernel/unified/include/kernel_structs.h holds the
common definitions. Architectures provide two files to complement it:
kernel_arch_data.h and kernel_arch_func.h. The first one contains at
least the struct _thread_arch and struct _kernel_arch data structures,
as well as the struct _callee_saved and struct _caller_saved register
layouts. The second file contains anything that needs what is provided
by the common stuff in kernel_structs.h. Those two files are only meant
to be included in kernel_structs.h in very specific locations.
The thread data structure has been separated into three major parts:
common struct _thread_base and struct k_thread, and arch-specific struct
_thread_arch. The first and third ones are included in the second.
The struct s_NANO data structure has been split into two: common struct
_kernel and arch-specific struct _kernel_arch. The latter is included in
the former.
Offsets files have also changed: nano_offsets.h has been renamed
kernel_offsets.h and is still included by the arch-specific offsets.c.
Also, since the thread and kernel data structures are now made of
sub-structures, offsets have to be added to make up the full offset.
Some of these additions have been consolidated in shorter symbols,
available from kernel/unified/include/offsets_short.h, which includes an
arch-specific offsets_arch_short.h. Most of the code include
offsets_short.h now instead of offsets.h.
Change-Id: I084645cb7e6db8db69aeaaf162963fe157045d5a
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Regardless of the number of interrupt priorities
supported, the exit of the trap handler can check
AUX_IRQ_ACT to see if it interrupted a FIRQ or IRQ.
Change-Id: I0b84d8298d3e6e437c934c01db4535fa8fe29458
Signed-off-by: Chuck Jordan <cjordan@synopsys.com>
- the interrupt (both regular and fast) now does not do rescheduling
if the current thread is a coop thread or if the scheduler is not locked
- the _nanokernel.flags cache of _current.flags is not used anymore
(could be a source of bugs) and is not needed in the scheduling algo
- there is no 'task' field in the _nanokernel anymore: scheduling routines
call _get_next_ready_thread instead
- the _nanokernel.fiber field is replaced by a more sophisticated
ready_q, based on the microkernel's priority-bitmap-based one
- thread initialization initializes new fields in the tcs, and does not
initialize obsolete ones
- nano_private includes nano_internal.h from the unified directory
- The FIBER, TASK and PREEMPTIBLE flags do not exist anymore: the thread
priority drives the behaviour
- the tcs uses a dlist for queuing in both ready and wait queues instead
of a custom singly-linked list
- other new fields in the tcs include a schedule-lock count, a
back-pointer to init data (when the task is static) and a pointer to
swap data, needed when a thread pending on _Swap() must be passed more
then just one value (e.g. k_stack_pop() needs an error code and data)
- the 'fiber' and 'task' fields of _nanokernel are replaced with an O(1)
ready queue (taken from the microkernel)
- fiberRtnValueSet() is aliased to _set_thread_return_value since it
also operates on preempt threads now
- _set_thread_return_value_with_data() sets the swap_data field in
addition to a return value from _Swap()
- convenience aliases are created for shorter names:
- _current is defined as _nanokernel.current
- _ready_q is defined as _nanokernel.ready_q
- _Swap() sets the threads's return code to -EAGAIN before swapping out
to prevent timeouts to have to set it (solves hard issues in some
kernel objects).
Change-Id: Ib9690173cbc36c36a9ec67e65590b40d758673de
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
