1 | Scheduling Concepts |
---|
2 | ################### |
---|
3 | |
---|
4 | .. index:: scheduling |
---|
5 | .. index:: task scheduling |
---|
6 | |
---|
7 | Introduction |
---|
8 | ============ |
---|
9 | |
---|
10 | The concept of scheduling in real-time systems dictates the ability to |
---|
11 | provide immediate response to specific external events, particularly |
---|
12 | the necessity of scheduling tasks to run within a specified time limit |
---|
13 | after the occurrence of an event. For example, software embedded in |
---|
14 | life-support systems used to monitor hospital patients must take instant |
---|
15 | action if a change in the patientâs status is detected. |
---|
16 | |
---|
17 | The component of RTEMS responsible for providing this capability is |
---|
18 | appropriately called the scheduler. The schedulerâs sole purpose is |
---|
19 | to allocate the all important resource of processor time to the various |
---|
20 | tasks competing for attention. |
---|
21 | |
---|
22 | Scheduling Algorithms |
---|
23 | ===================== |
---|
24 | |
---|
25 | .. index:: scheduling algorithms |
---|
26 | |
---|
27 | RTEMS provides a plugin framework which allows it to support |
---|
28 | multiple scheduling algorithms. RTEMS now includes multiple |
---|
29 | scheduling algorithms in the SuperCore and the user can select which |
---|
30 | of these they wish to use in their application. In addition, |
---|
31 | the user can implement their own scheduling algorithm and |
---|
32 | configure RTEMS to use it. |
---|
33 | |
---|
34 | Supporting multiple scheduling algorithms gives the end user the |
---|
35 | option to select the algorithm which is most appropriate to their use |
---|
36 | case. Most real-time operating systems schedule tasks using a priority |
---|
37 | based algorithm, possibly with preemption control. The classic |
---|
38 | RTEMS scheduling algorithm which was the only algorithm available |
---|
39 | in RTEMS 4.10 and earlier, is a priority based scheduling algorithm. |
---|
40 | This scheduling algoritm is suitable for single core (e.g. non-SMP) |
---|
41 | systems and is now known as the *Deterministic Priority Scheduler*. |
---|
42 | Unless the user configures another scheduling algorithm, RTEMS will use |
---|
43 | this on single core systems. |
---|
44 | |
---|
45 | Priority Scheduling |
---|
46 | ------------------- |
---|
47 | .. index:: priority scheduling |
---|
48 | |
---|
49 | When using priority based scheduling, RTEMS allocates the processor using |
---|
50 | a priority-based, preemptive algorithm augmented to provide round-robin |
---|
51 | characteristics within individual priority groups. The goal of this |
---|
52 | algorithm is to guarantee that the task which is executing on the |
---|
53 | processor at any point in time is the one with the highest priority |
---|
54 | among all tasks in the ready state. |
---|
55 | |
---|
56 | When a task is added to the ready chain, it is placed behind all other |
---|
57 | tasks of the same priority. This rule provides a round-robin within |
---|
58 | priority group scheduling characteristic. This means that in a group of |
---|
59 | equal priority tasks, tasks will execute in the order they become ready |
---|
60 | or FIFO order. Even though there are ways to manipulate and adjust task |
---|
61 | priorities, the most important rule to remember is: |
---|
62 | |
---|
63 | - *Priority based scheduling algorithms will always select the |
---|
64 | highest priority task that is ready to run when allocating the processor |
---|
65 | to a task.* |
---|
66 | |
---|
67 | Priority scheduling is the most commonly used scheduling algorithm. |
---|
68 | It should be used by applications in which multiple tasks contend for |
---|
69 | CPU time or other resources and there is a need to ensure certain tasks |
---|
70 | are given priority over other tasks. |
---|
71 | |
---|
72 | There are a few common methods of accomplishing the mechanics of this |
---|
73 | algorithm. These ways involve a list or chain of tasks in the ready state. |
---|
74 | |
---|
75 | - The least efficient method is to randomly place tasks in the ready |
---|
76 | chain forcing the scheduler to scan the entire chain to determine which |
---|
77 | task receives the processor. |
---|
78 | |
---|
79 | - A more efficient method is to schedule the task by placing it |
---|
80 | in the proper place on the ready chain based on the designated scheduling |
---|
81 | criteria at the time it enters the ready state. Thus, when the processor |
---|
82 | is free, the first task on the ready chain is allocated the processor. |
---|
83 | |
---|
84 | - Another mechanism is to maintain a list of FIFOs per priority. |
---|
85 | When a task is readied, it is placed on the rear of the FIFO for its |
---|
86 | priority. This method is often used with a bitmap to assist in locating |
---|
87 | which FIFOs have ready tasks on them. |
---|
88 | |
---|
89 | RTEMS currently includes multiple priority based scheduling algorithms |
---|
90 | as well as other algorithms which incorporate deadline. Each algorithm |
---|
91 | is discussed in the following sections. |
---|
92 | |
---|
93 | Deterministic Priority Scheduler |
---|
94 | -------------------------------- |
---|
95 | |
---|
96 | This is the scheduler implementation which has always been in RTEMS. |
---|
97 | After the 4.10 release series, it was factored into pluggable scheduler |
---|
98 | selection. It schedules tasks using a priority based algorithm which |
---|
99 | takes into account preemption. It is implemented using an array of FIFOs |
---|
100 | with a FIFO per priority. It maintains a bitmap which is used to track |
---|
101 | which priorities have ready tasks. |
---|
102 | |
---|
103 | This algorithm is deterministic (e.g. predictable and fixed) in execution |
---|
104 | time. This comes at the cost of using slightly over three (3) kilobytes |
---|
105 | of RAM on a system configured to support 256 priority levels. |
---|
106 | |
---|
107 | This scheduler is only aware of a single core. |
---|
108 | |
---|
109 | Simple Priority Scheduler |
---|
110 | ------------------------- |
---|
111 | |
---|
112 | This scheduler implementation has the same behaviour as the Deterministic |
---|
113 | Priority Scheduler but uses only one linked list to manage all ready |
---|
114 | tasks. When a task is readied, a linear search of that linked list is |
---|
115 | performed to determine where to insert the newly readied task. |
---|
116 | |
---|
117 | This algorithm uses much less RAM than the Deterministic Priority |
---|
118 | Scheduler but is *O(n)* where *n* is the number of ready tasks. |
---|
119 | In a small system with a small number of tasks, this will not be a |
---|
120 | performance issue. Reducing RAM consumption is often critical in small |
---|
121 | systems which are incapable of supporting a large number of tasks. |
---|
122 | |
---|
123 | This scheduler is only aware of a single core. |
---|
124 | |
---|
125 | Simple SMP Priority Scheduler |
---|
126 | ----------------------------- |
---|
127 | |
---|
128 | This scheduler is based upon the Simple Priority Scheduler and is designed |
---|
129 | to have the same behaviour on a single core system. But this scheduler |
---|
130 | is capable of scheduling threads across multiple cores in an SMP system. |
---|
131 | When given a choice of replacing one of two threads at equal priority |
---|
132 | on different cores, this algorithm favors replacing threads which are |
---|
133 | preemptible and have executed the longest. |
---|
134 | |
---|
135 | This algorithm is non-deterministic. When scheduling, it must consider |
---|
136 | which tasks are to be executed on each core while avoiding superfluous |
---|
137 | task migrations. |
---|
138 | |
---|
139 | Earliest Deadline First Scheduler |
---|
140 | --------------------------------- |
---|
141 | .. index:: earliest deadline first scheduling |
---|
142 | |
---|
143 | This is an alternative scheduler in RTEMS for single core applications. |
---|
144 | The primary EDF advantage is high total CPU utilization (theoretically |
---|
145 | up to 100%). It assumes that tasks have priorities equal to deadlines. |
---|
146 | |
---|
147 | This EDF is initially preemptive, however, individual tasks may be declared |
---|
148 | not-preemptive. Deadlines are declared using only Rate Monotonic manager which |
---|
149 | goal is to handle periodic behavior. Period is always equal to deadline. All |
---|
150 | ready tasks reside in a single ready queue implemented using a red-black tree. |
---|
151 | |
---|
152 | This implementation of EDF schedules two different types of task |
---|
153 | priority types while each task may switch between the two types within |
---|
154 | its execution. If a task does have a deadline declared using the Rate |
---|
155 | Monotonic manager, the task is deadline-driven and its priority is equal |
---|
156 | to deadline. On the contrary if a task does not have any deadline or |
---|
157 | the deadline is cancelled using the Rate Monotonic manager, the task is |
---|
158 | considered a background task with priority equal to that assigned |
---|
159 | upon initialization in the same manner as for priority scheduler. Each |
---|
160 | background task is of a lower importance than each deadline-driven one |
---|
161 | and is scheduled when no deadline-driven task and no higher priority |
---|
162 | background task is ready to run. |
---|
163 | |
---|
164 | Every deadline-driven scheduling algorithm requires means for tasks |
---|
165 | to claim a deadline. The Rate Monotonic Manager is responsible for |
---|
166 | handling periodic execution. In RTEMS periods are equal to deadlines, |
---|
167 | thus if a task announces a period, it has to be finished until the |
---|
168 | end of this period. The call of ``rtems_rate_monotonic_period`` |
---|
169 | passes the scheduler the length of oncoming deadline. Moreover, the``rtems_rate_monotonic_cancel`` and ``rtems_rate_monotonic_delete`` |
---|
170 | calls clear the deadlines assigned to the task. |
---|
171 | |
---|
172 | Constant Bandwidth Server Scheduling (CBS) |
---|
173 | ------------------------------------------ |
---|
174 | .. index:: constant bandwidth server scheduling |
---|
175 | |
---|
176 | This is an alternative scheduler in RTEMS for single core applications. |
---|
177 | The CBS is a budget aware extension of EDF scheduler. The main goal of this |
---|
178 | scheduler is to ensure temporal isolation of tasks meaning that a taskâs |
---|
179 | execution in terms of meeting deadlines must not be influenced by other |
---|
180 | tasks as if they were run on multiple independent processors. |
---|
181 | |
---|
182 | Each task can be assigned a server (current implementation supports only |
---|
183 | one task per server). The server is characterized by period (deadline) |
---|
184 | and computation time (budget). The ratio budget/period yields bandwidth, |
---|
185 | which is the fraction of CPU to be reserved by the scheduler for each |
---|
186 | subsequent period. |
---|
187 | |
---|
188 | The CBS is equipped with a set of rules applied to tasks attached to servers |
---|
189 | ensuring that deadline miss because of another task cannot occur. |
---|
190 | In case a task breaks one of the rules, its priority is pulled to background |
---|
191 | until the end of its period and then restored again. The rules are: |
---|
192 | |
---|
193 | - Task cannot exceed its registered budget, |
---|
194 | |
---|
195 | - Task cannot be |
---|
196 | unblocked when a ratio between remaining budget and remaining deadline |
---|
197 | is higher than declared bandwidth. |
---|
198 | |
---|
199 | The CBS provides an extensive API. Unlike EDF, the``rtems_rate_monotonic_period`` does not declare a deadline because |
---|
200 | it is carried out using CBS API. This call only announces next period. |
---|
201 | |
---|
202 | Scheduling Modification Mechanisms |
---|
203 | ================================== |
---|
204 | |
---|
205 | .. index:: scheduling mechanisms |
---|
206 | |
---|
207 | RTEMS provides four mechanisms which allow the user to alter the task |
---|
208 | scheduling decisions: |
---|
209 | |
---|
210 | - user-selectable task priority level |
---|
211 | |
---|
212 | - task preemption control |
---|
213 | |
---|
214 | - task timeslicing control |
---|
215 | |
---|
216 | - manual round-robin selection |
---|
217 | |
---|
218 | Each of these methods provides a powerful capability to customize sets |
---|
219 | of tasks to satisfy the unique and particular requirements encountered |
---|
220 | in custom real-time applications. Although each mechanism operates |
---|
221 | independently, there is a precedence relationship which governs the |
---|
222 | effects of scheduling modifications. The evaluation order for scheduling |
---|
223 | characteristics is always priority, preemption mode, and timeslicing. |
---|
224 | When reading the descriptions of timeslicing and manual round-robin |
---|
225 | it is important to keep in mind that preemption (if enabled) of a task |
---|
226 | by higher priority tasks will occur as required, overriding the other |
---|
227 | factors presented in the description. |
---|
228 | |
---|
229 | Task Priority and Scheduling |
---|
230 | ---------------------------- |
---|
231 | .. index:: task priority |
---|
232 | |
---|
233 | The most significant task scheduling modification mechanism is the ability |
---|
234 | for the user to assign a priority level to each individual task when it |
---|
235 | is created and to alter a taskâs priority at run-time. RTEMS supports |
---|
236 | up to 255 priority levels. Level 255 is the lowest priority and level |
---|
237 | 1 is the highest. |
---|
238 | |
---|
239 | Preemption |
---|
240 | ----------.. index:: preemption |
---|
241 | |
---|
242 | Another way the user can alter the basic scheduling algorithm is by |
---|
243 | manipulating the preemption mode flag (``RTEMS_PREEMPT_MASK``) |
---|
244 | of individual tasks. If preemption is disabled for a task |
---|
245 | (``RTEMS_NO_PREEMPT``), then the task will not relinquish |
---|
246 | control of the processor until it terminates, blocks, or re-enables |
---|
247 | preemption. Even tasks which become ready to run and possess higher |
---|
248 | priority levels will not be allowed to execute. Note that the preemption |
---|
249 | setting has no effect on the manner in which a task is scheduled. |
---|
250 | It only applies once a task has control of the processor. |
---|
251 | |
---|
252 | Timeslicing |
---|
253 | -----------.. index:: timeslicing |
---|
254 | .. index:: round robin scheduling |
---|
255 | |
---|
256 | Timeslicing or round-robin scheduling is an additional method which |
---|
257 | can be used to alter the basic scheduling algorithm. Like preemption, |
---|
258 | timeslicing is specified on a task by task basis using the timeslicing |
---|
259 | mode flag (``RTEMS_TIMESLICE_MASK``). If timeslicing is |
---|
260 | enabled for a task (``RTEMS_TIMESLICE``), then RTEMS will |
---|
261 | limit the amount of time the task can execute before the processor is |
---|
262 | allocated to another task. Each tick of the real-time clock reduces |
---|
263 | the currently running taskâs timeslice. When the execution time equals |
---|
264 | the timeslice, RTEMS will dispatch another task of the same priority |
---|
265 | to execute. If there are no other tasks of the same priority ready to |
---|
266 | execute, then the current task is allocated an additional timeslice and |
---|
267 | continues to run. Remember that a higher priority task will preempt |
---|
268 | the task (unless preemption is disabled) as soon as it is ready to run, |
---|
269 | even if the task has not used up its entire timeslice. |
---|
270 | |
---|
271 | Manual Round-Robin |
---|
272 | ------------------.. index:: manual round robin |
---|
273 | |
---|
274 | The final mechanism for altering the RTEMS scheduling algorithm is |
---|
275 | called manual round-robin. Manual round-robin is invoked by using the``rtems_task_wake_after`` directive with a time interval |
---|
276 | of ``RTEMS_YIELD_PROCESSOR``. This allows a task to give |
---|
277 | up the processor and be immediately returned to the ready chain at the |
---|
278 | end of its priority group. If no other tasks of the same priority are |
---|
279 | ready to run, then the task does not lose control of the processor. |
---|
280 | |
---|
281 | Dispatching Tasks |
---|
282 | =================.. index:: dispatching |
---|
283 | |
---|
284 | The dispatcher is the RTEMS component responsible for |
---|
285 | allocating the processor to a ready task. In order to allocate |
---|
286 | the processor to one task, it must be deallocated or retrieved |
---|
287 | from the task currently using it. This involves a concept |
---|
288 | called a context switch. To perform a context switch, the |
---|
289 | dispatcher saves the context of the current task and restores |
---|
290 | the context of the task which has been allocated to the |
---|
291 | processor. Saving and restoring a taskâs context is the |
---|
292 | storing/loading of all the essential information about a task to |
---|
293 | enable it to continue execution without any effects of the |
---|
294 | interruption. For example, the contents of a taskâs register |
---|
295 | set must be the same when it is given the processor as they were |
---|
296 | when it was taken away. All of the information that must be |
---|
297 | saved or restored for a context switch is located either in the |
---|
298 | TCB or on the taskâs stacks. |
---|
299 | |
---|
300 | Tasks that utilize a numeric coprocessor and are created with the``RTEMS_FLOATING_POINT`` attribute require additional |
---|
301 | operations during a context switch. These additional operations |
---|
302 | are necessary to save and restore the floating point context of``RTEMS_FLOATING_POINT`` tasks. To avoid unnecessary save |
---|
303 | and restore operations, the state of the numeric coprocessor is only |
---|
304 | saved when a ``RTEMS_FLOATING_POINT`` task is dispatched |
---|
305 | and that task was not the last task to utilize the coprocessor. |
---|
306 | |
---|
307 | Task State Transitions |
---|
308 | ======================.. index:: task state transitions |
---|
309 | |
---|
310 | Tasks in an RTEMS system must always be in one of the |
---|
311 | five allowable task states. These states are: executing, ready, |
---|
312 | blocked, dormant, and non-existent. |
---|
313 | |
---|
314 | A task occupies the non-existent state before |
---|
315 | a ``rtems_task_create`` has been issued on its behalf. |
---|
316 | A task enters the non-existent state from any other state in the system |
---|
317 | when it is deleted with the ``rtems_task_delete`` directive. |
---|
318 | While a task occupies this state it does not have a TCB or a task ID |
---|
319 | assigned to it; therefore, no other tasks in the system may reference |
---|
320 | this task. |
---|
321 | |
---|
322 | When a task is created via the ``rtems_task_create`` |
---|
323 | directive it enters the dormant state. This state is not entered through |
---|
324 | any other means. Although the task exists in the system, it cannot |
---|
325 | actively compete for system resources. It will remain in the dormant |
---|
326 | state until it is started via the ``rtems_task_start`` |
---|
327 | directive, at which time it enters the ready state. The task is now |
---|
328 | permitted to be scheduled for the processor and to compete for other |
---|
329 | system resources. |
---|
330 | |
---|
331 | .. code:: c |
---|
332 | |
---|
333 | +-------------------------------------------------------------+ |
---|
334 | | Non-existent | |
---|
335 | | +-------------------------------------------------------+ | |
---|
336 | | | | | |
---|
337 | | | | | |
---|
338 | | | Creating +---------+ Deleting | | |
---|
339 | | | -------------------> | Dormant | -------------------> | | |
---|
340 | | | +---------+ | | |
---|
341 | | | | | | |
---|
342 | | | Starting | | | |
---|
343 | | | | | | |
---|
344 | | | V Deleting | | |
---|
345 | | | +-------> +-------+ -------------------> | | |
---|
346 | | | Yielding / +----- | Ready | ------+ | | |
---|
347 | | | / / +-------+ <--+ \\ | | |
---|
348 | | | / / \\ \\ Blocking | | |
---|
349 | | | / / Dispatching Readying \\ \\ | | |
---|
350 | | | / V \\ V | | |
---|
351 | | | +-----------+ Blocking +---------+ | | |
---|
352 | | | | Executing | --------------> | Blocked | | | |
---|
353 | | | +-----------+ +---------+ | | |
---|
354 | | | | | |
---|
355 | | | | | |
---|
356 | | +-------------------------------------------------------+ | |
---|
357 | | Non-existent | |
---|
358 | +-------------------------------------------------------------+ |
---|
359 | |
---|
360 | A task occupies the blocked state whenever it is unable to be scheduled |
---|
361 | to run. A running task may block itself or be blocked by other tasks in |
---|
362 | the system. The running task blocks itself through voluntary operations |
---|
363 | that cause the task to wait. The only way a task can block a task other |
---|
364 | than itself is with the ``rtems_task_suspend`` directive. |
---|
365 | A task enters the blocked state due to any of the following conditions: |
---|
366 | |
---|
367 | - A task issues a ``rtems_task_suspend`` directive |
---|
368 | which blocks either itself or another task in the system. |
---|
369 | |
---|
370 | - The running task issues a ``rtems_barrier_wait`` |
---|
371 | directive. |
---|
372 | |
---|
373 | - The running task issues a ``rtems_message_queue_receive`` |
---|
374 | directive with the wait option and the message queue is empty. |
---|
375 | |
---|
376 | - The running task issues an ``rtems_event_receive`` |
---|
377 | directive with the wait option and the currently pending events do not |
---|
378 | satisfy the request. |
---|
379 | |
---|
380 | - The running task issues a ``rtems_semaphore_obtain`` |
---|
381 | directive with the wait option and the requested semaphore is unavailable. |
---|
382 | |
---|
383 | - The running task issues a ``rtems_task_wake_after`` |
---|
384 | directive which blocks the task for the given time interval. If the time |
---|
385 | interval specified is zero, the task yields the processor and remains |
---|
386 | in the ready state. |
---|
387 | |
---|
388 | - The running task issues a ``rtems_task_wake_when`` |
---|
389 | directive which blocks the task until the requested date and time arrives. |
---|
390 | |
---|
391 | - The running task issues a ``rtems_rate_monotonic_period`` |
---|
392 | directive and must wait for the specified rate monotonic period |
---|
393 | to conclude. |
---|
394 | |
---|
395 | - The running task issues a ``rtems_region_get_segment`` |
---|
396 | directive with the wait option and there is not an available segment large |
---|
397 | enough to satisfy the taskâs request. |
---|
398 | |
---|
399 | A blocked task may also be suspended. Therefore, both the suspension |
---|
400 | and the blocking condition must be removed before the task becomes ready |
---|
401 | to run again. |
---|
402 | |
---|
403 | A task occupies the ready state when it is able to be scheduled to run, |
---|
404 | but currently does not have control of the processor. Tasks of the same |
---|
405 | or higher priority will yield the processor by either becoming blocked, |
---|
406 | completing their timeslice, or being deleted. All tasks with the same |
---|
407 | priority will execute in FIFO order. A task enters the ready state due |
---|
408 | to any of the following conditions: |
---|
409 | |
---|
410 | - A running task issues a ``rtems_task_resume`` |
---|
411 | directive for a task that is suspended and the task is not blocked |
---|
412 | waiting on any resource. |
---|
413 | |
---|
414 | - A running task issues a ``rtems_message_queue_send``,``rtems_message_queue_broadcast``, or a``rtems_message_queue_urgent`` directive |
---|
415 | which posts a message to the queue on which the blocked task is |
---|
416 | waiting. |
---|
417 | |
---|
418 | - A running task issues an ``rtems_event_send`` |
---|
419 | directive which sends an event condition to a task which is blocked |
---|
420 | waiting on that event condition. |
---|
421 | |
---|
422 | - A running task issues a ``rtems_semaphore_release`` |
---|
423 | directive which releases the semaphore on which the blocked task is |
---|
424 | waiting. |
---|
425 | |
---|
426 | - A timeout interval expires for a task which was blocked |
---|
427 | by a call to the ``rtems_task_wake_after`` directive. |
---|
428 | |
---|
429 | - A timeout period expires for a task which blocked by a |
---|
430 | call to the ``rtems_task_wake_when`` directive. |
---|
431 | |
---|
432 | - A running task issues a ``rtems_region_return_segment`` |
---|
433 | directive which releases a segment to the region on which the blocked task |
---|
434 | is waiting and a resulting segment is large enough to satisfy |
---|
435 | the taskâs request. |
---|
436 | |
---|
437 | - A rate monotonic period expires for a task which blocked |
---|
438 | by a call to the ``rtems_rate_monotonic_period`` directive. |
---|
439 | |
---|
440 | - A timeout interval expires for a task which was blocked |
---|
441 | waiting on a message, event, semaphore, or segment with a |
---|
442 | timeout specified. |
---|
443 | |
---|
444 | - A running task issues a directive which deletes a |
---|
445 | message queue, a semaphore, or a region on which the blocked |
---|
446 | task is waiting. |
---|
447 | |
---|
448 | - A running task issues a ``rtems_task_restart`` |
---|
449 | directive for the blocked task. |
---|
450 | |
---|
451 | - The running task, with its preemption mode enabled, may |
---|
452 | be made ready by issuing any of the directives that may unblock |
---|
453 | a task with a higher priority. This directive may be issued |
---|
454 | from the running task itself or from an ISR. |
---|
455 | A ready task occupies the executing state when it has |
---|
456 | control of the CPU. A task enters the executing state due to |
---|
457 | any of the following conditions: |
---|
458 | |
---|
459 | - The task is the highest priority ready task in the |
---|
460 | system. |
---|
461 | |
---|
462 | - The running task blocks and the task is next in the |
---|
463 | scheduling queue. The task may be of equal priority as in |
---|
464 | round-robin scheduling or the task may possess the highest |
---|
465 | priority of the remaining ready tasks. |
---|
466 | |
---|
467 | - The running task may reenable its preemption mode and a |
---|
468 | task exists in the ready queue that has a higher priority than |
---|
469 | the running task. |
---|
470 | |
---|
471 | - The running task lowers its own priority and another |
---|
472 | task is of higher priority as a result. |
---|
473 | |
---|
474 | - The running task raises the priority of a task above its |
---|
475 | own and the running task is in preemption mode. |
---|
476 | |
---|
477 | .. COMMENT: COPYRIGHT (c) 1988-2008. |
---|
478 | |
---|
479 | .. COMMENT: On-Line Applications Research Corporation (OAR). |
---|
480 | |
---|
481 | .. COMMENT: All rights reserved. |
---|
482 | |
---|