| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
| |
On ARM Thumb we may have function addresses ending with 0x7f, if we are
lucky.
|
|
|
|
|
| |
Initialize the thread queue context with invalid data in debug
configurations to catch missing set up steps.
|
|
|
|
|
|
|
|
|
| |
Move thread state for _Thread_queue_Enqueue() to the thread queue
context. This reduces the parameter count of _Thread_queue_Enqueue()
from five to four (ARM for example has only four function parameter
registers). Since the thread state is used after several function calls
inside _Thread_queue_Enqueue() this parameter was saved on the stack
previously.
|
|
|
|
|
|
|
|
|
|
| |
Callers of _Thread_Do_dispatch() must have a valid
Per_CPU_Control::Stats::thread_dispatch_disabled_instant.
Call _Profiling_Outer_most_interrupt_entry_and_exit() with the interrupt
stack to not exceed Per_CPU_Control::Interrupt_frame.
Update #2751.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Turn pthread_spinlock_t into a self-contained object. On uni-processor
configurations, interrupts are disabled in the lock/trylock operations
and the previous interrupt status is restored in the corresponding
unlock operations. On SMP configurations, a ticket lock is a acquired
and released in addition.
The self-contained pthread_spinlock_t object is defined by Newlib in
<sys/_pthreadtypes.h>.
typedef struct {
struct _Ticket_lock_Control _lock;
__uint32_t _interrupt_state;
} pthread_spinlock_t;
This implementation is simple and efficient. However, this test case of
the Linux Test Project would fail due to call of printf() and sleep()
during spin lock ownership:
https://github.com/linux-test-project/ltp/blob/master/testcases/open_posix_testsuite/conformance/interfaces/pthread_spin_lock/1-2.c
There is only limited support for profiling on SMP configurations.
Delete CORE spinlock implementation.
Update #2674.
|
|
|
|
|
| |
Delete unused _Thread_queue_Enqueue() and rename
_Thread_queue_Enqueue_critical() to _Thread_queue_Enqueue().
|
|
|
|
|
|
|
| |
Replace the expected thread dispatch disable level with a thread queue
enqueue callout. This enables the use of _Thread_Dispatch_direct() in
the thread queue enqueue procedure. This avoids impossible exection
paths, e.g. Per_CPU_Control::dispatch_necessary is always true.
|
|
|
|
|
|
|
|
|
|
|
|
| |
On SMP configurations, it is a fatal error to call blocking operating
system with interrupts disabled, since this prevents delivery of
inter-processor interrupts. This could lead to executing threads which
are not allowed to execute resulting in undefined behaviour.
The ARM Cortex-M port has a similar problem, since the interrupt state
is not a part of the thread context.
Update #2811.
|
| |
|
|
|
|
| |
Avoid use of the stack for the hot paths.
|
|
|
|
| |
Update #2811.
|
|
|
|
|
|
|
|
|
| |
Use a processor-specific interrupt frame during context switches in case
the executing thread is longer executes on the processor and the heir
thread is about to start execution. During this period we must not use
a thread stack for interrupt processing.
Update #2809.
|
|
|
|
| |
Update #2809.
|
|
|
|
| |
Update #2751.
|
|
|
|
|
|
|
|
|
|
| |
This function is useful for operations which synchronously block, e.g.
self restart, self deletion, yield, sleep. It helps to detect if these
operations are called in the wrong context. Since the thread dispatch
necessary indicator is not used, this is more robust in some SMP
situations.
Update #2751.
|
|
|
|
| |
Update #2797.
|
|
|
|
| |
Update #2797.
|
|
|
|
|
|
|
|
| |
Rename _Scheduler_Assignments into _Scheduler_Initial_assignments to
make it clear that they may not reflect the run-time scheduler
assignment.
Update #2797.
|
| |
|
|
|
|
|
|
|
| |
This makes it possible to adjust the scheduler of a processor at
run-time.
Update #2797.
|
|
|
|
| |
Avoid use of processor index 0 which may have no scheduler assigned.
|
|
|
|
|
|
|
|
|
| |
The SMP ticket lock release turned out to be suitable for inlining, e.g.
a hand full of instructions, no branches.
The changes in the screen files do not reflect the changes due to this
commit. However, they are now up to date. Obtained on a T4240 running
at 1.5GHz using GCC 7.0.0 20161108..
|
|
|
|
|
| |
These kernel space header files must be provided for Newlib
172e2050d95b41861db858dd9bc43a3fb4a28987.
|
| |
|
| |
|
|
|
|
|
|
| |
This reduces the code size and helps to reduce the amount of testing.
Hot paths can use the _Thread_queue_Queue_acquire_critical() and
_Thread_queue_Queue_release_critical() functions which are still inline.
|
|
|
|
|
| |
Use non-inline SMP lock acquire and release operations by default.
Provide inline variants for the hot spots, e.g. mutex acquire/release.
|
|
|
|
|
| |
Maintain the thread resource count only in debug configurations. This
is a performance optimization for non-debug configurations.
|
|
|
|
|
| |
This makes it easier to conditionally enable/disable the thread resource
count usage.
|
|
|
|
| |
Update #2556.
|
|
|
|
|
|
|
|
| |
Replace Thread_Scheduler_control::control and
Thread_Scheduler_control::own_control with new
Thread_Scheduler_control::home.
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
| |
|
|
|
|
| |
Remove unused return status.
|
|
|
|
|
|
| |
Delete Thread_Control::Resource_node.
Update #2556.
|
|
|
|
|
|
|
| |
Delete _Scheduler_Thread_change_resource_root() and
_Scheduler_Thread_change_help_state().
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
|
|
|
| |
Add the ability to add/remove scheduler nodes to/from the set of
scheduler nodes available to the schedulers for a particular thread.
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
| |
Update #2556.
|
|
|
|
|
|
| |
Changed for consistency with other scheduler operations.
Update #2556.
|
|
|
|
|
|
| |
Changed for consistency with other scheduler operations.
Update #2556.
|