| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
FreeBSD assumes that u_int is a 32-bit integer type. This is wrong for
some 16-bit targets supported by RTEMS.
Update #2271.
|
|
|
|
| |
Update #2271.
|
|
|
|
|
|
|
| |
Thread dispatching is disabled in case interrupts are disabled. To get
an accurate thread dispatch disabled time it is important to use the
interrupt disabled instant in case a transition from an interrupt
disabled section to a thread dispatch level section happens.
|
| |
|
|
|
|
| |
This reduces the code size drastically.
|
|
|
|
|
| |
This function is superfluous due to the introduction of fine grained
locking.
|
|
|
|
| |
Update #2273.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move the writes to Thread_Control::current_priority and
Thread_Control::real_priority into _Thread_Change_priority() under the
protection of the thread lock. Add a filter function to
_Thread_Change_priority() to enable specialized variants.
Avoid race conditions during a thread priority restore with the new
Thread_Control::priority_restore_hint for an important average case
optimizations used by priority inheritance mutexes.
Update #2273.
|
|
|
|
| |
Update #2273.
|
| |
|
| |
|
|
|
|
| |
Rename _CORE_semaphore_Seize_isr_disable() to _CORE_semaphore_Seize().
|
|
|
|
| |
Update #2273.
|
|
|
|
|
|
|
|
|
|
|
| |
Aggregate several critical sections into a bigger one. Sending and
receiving messages is now protected by an ISR lock. Thread dispatching
is only disabled in case a blocking operation is necessary. The message
copy procedure is done inside the critical section (interrupts
disabled). Thus this change may have a negative impact on the interrupt
latency in case very large messages are transferred.
Update #2273.
|
|
|
|
|
| |
Check the number of pending messages in _CORE_message_queue_Flush() to
avoid race conditions.
|
|
|
|
|
| |
Use a parameter for _Thread_queue_Enqueue() instead to reduce memory
usage.
|
|
|
|
|
|
|
|
|
| |
Use thread wait flags for synchronization. The enqueue operation is now
part of the initial critical section. This is the key change and
enables fine grained locking on SMP for objects using a thread queue
like semaphores and message queues.
Update #2273.
|
|
|
|
|
|
|
|
|
| |
Move thread queue discipline specific operations into
Thread_queue_Operations. Use a separate node in the thread control
block for the thread queue to make it independent of the scheduler data
structures.
Update #2273.
|
|
|
|
|
|
|
|
| |
Replace the Thread_Priority_control with more general
Thread_queue_Operations which will be used for generic priority change,
timeout, signal and wait queue operations in the future.
Update #2273.
|
|
|
|
|
|
|
|
|
|
|
| |
Move the complete thread queue enqueue procedure into
_Thread_queue_Enqueue_critical(). It is possible to use the thread
queue lock to protect state of the object embedding the thread queue.
This enables per object fine grained locking in the future.
Delete _Thread_queue_Enter_critical_section().
Update #2273.
|
|
|
|
|
|
|
| |
Add a thread wait timeout code. Replace _Event_Timeout() with a general
purpose _Thread_Timeout() watchdog handler.
Update #2273.
|
|
|
|
|
|
|
|
|
|
| |
Merge THREAD_WAIT_STATE_SATISFIED, THREAD_WAIT_STATE_TIMEOUT,
THREAD_WAIT_STATE_INTERRUPT_SATISFIED, and
THREAD_WAIT_STATE_INTERRUPT_TIMEOUT into one state
THREAD_WAIT_STATE_READY_AGAIN. This helps to write generic routines to
block a thread.
Update #2273.
|
|
|
|
|
|
|
|
| |
Use mostly the standard watchdog operations. Use a system event for
synchronization. This implementation is simpler and offers better SMP
performance.
Close #2131.
|
|
|
|
|
|
|
| |
Make internal function _Watchdog_Remove_it() static to avoid accidental
usage.
Update #2307.
|
|
|
|
|
|
|
|
|
|
| |
Rewrite the _Watchdog_Insert(), _Watchdog_Remove() and
_Watchdog_Tickle() functions to use iterator items to synchronize
concurrent operations. This makes it possible to get rid of the global
variables _Watchdog_Sync_level and _Watchdog_Sync_count which are a
blocking point for scalable SMP solutions.
Update #2307.
|
|
|
|
| |
Update #2307.
|
|
|
|
|
|
|
|
| |
Add watchdog header parameter to _Watchdog_Remove() to be in line with
the other operations. Add _Watchdog_Remove_ticks() and
_Watchdog_Remove_seconds() for convenience.
Update #2307.
|
|
|
|
|
|
| |
Avoid the usage of the current thread state in
_Thread_queue_Extract_with_return_code() since thread queues should not
know anything about thread states.
|
|
|
|
|
| |
Remove thread queue parameter from _Thread_queue_Extract() since the
current thread queue is stored in the thread control block.
|
| |
|
|
|
|
|
|
| |
Account for priority changes of threads executing in a foreign
partition. Exchange idle threads in case a victim node uses an idle
thread and the new scheduled node needs an idle thread.
|
|
|
|
|
| |
Fix layout of the common block of Thread_Control and
Thread_Proxy_control. Ensure that the offsets match.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
A thread join is twofold. There is one thread that exists and an
arbitrary number of threads that wait for the thread exit (one-to-many
relation). The exiting thread may want to wait for a thread that wants
to join its exit (STATES_WAITING_FOR_JOIN_AT_EXIT in
_POSIX_Thread_Exit()). On the other side we need a thread queue for all
the threads that wait for the exit of one particular thread
(STATES_WAITING_FOR_JOIN in pthread_join()).
Update #2035.
|
| |
|
|
|
|
|
| |
Move some code into _CORE_mutex_Seize_interrupt_blocking() so that the
thread queue handling is in one place.
|
|
|
|
|
| |
Use a parameter for _Thread_queue_Enqueue() instead to reduce memory
usage.
|
|
|
|
|
|
|
|
| |
Move the linear search into a critical section to avoid corruption due
to higher priority interrupts. The interrupt disable time depends now
on the count of pending messages.
Close #2328.
|
|
|
|
|
| |
This function was identical to _Thread_queue_Timeout(). This makes
_Thread_queue_Enqueue_with_handler() obsolete.
|
|
|
|
|
| |
It makes no sense to use this indirection since the type for timeout
values is Watchdog_Interval.
|
|
|
|
|
|
|
| |
The Objects_Control::Lock was a software layer violation. It worked
only for the threads since they are somewhat special.
Update #2273.
|
|
|
|
|
|
|
|
|
|
| |
Remove _Thread_Acquire() and _Thread_Acquire_for_executing(). Add
utility functions for the default thread lock. Use the default thread
lock for the RTEMS events. There is no need to disable thread
dispatching and a Giant acquire in _Event_Timeout() since this was
already done by the caller.
Update #2273.
|
|
|
|
|
| |
Return the current processor to be in line with
_Thread_Disable_dispatch().
|
|
|
|
|
|
|
|
| |
Do not disable thread dispatching and do not acquire the Giant lock.
This makes it possible to use this object get variant for fine grained
locking.
Update #2273.
|
|
|
|
|
|
|
| |
Use ISR_lock_Context instead of ISR_Level to allow use of ISR locks for
low-level locking.
Update #2273.
|
| |
|
| |
|
| |
|
| |
|
| |
|