diff options
Diffstat (limited to 'kernel/slow-work.c')
-rw-r--r-- | kernel/slow-work.c | 494 |
1 files changed, 451 insertions, 43 deletions
diff --git a/kernel/slow-work.c b/kernel/slow-work.c index 0d31135efbf4..da94f3c101af 100644 --- a/kernel/slow-work.c +++ b/kernel/slow-work.c | |||
@@ -16,11 +16,8 @@ | |||
16 | #include <linux/kthread.h> | 16 | #include <linux/kthread.h> |
17 | #include <linux/freezer.h> | 17 | #include <linux/freezer.h> |
18 | #include <linux/wait.h> | 18 | #include <linux/wait.h> |
19 | 19 | #include <linux/proc_fs.h> | |
20 | #define SLOW_WORK_CULL_TIMEOUT (5 * HZ) /* cull threads 5s after running out of | 20 | #include "slow-work.h" |
21 | * things to do */ | ||
22 | #define SLOW_WORK_OOM_TIMEOUT (5 * HZ) /* can't start new threads for 5s after | ||
23 | * OOM */ | ||
24 | 21 | ||
25 | static void slow_work_cull_timeout(unsigned long); | 22 | static void slow_work_cull_timeout(unsigned long); |
26 | static void slow_work_oom_timeout(unsigned long); | 23 | static void slow_work_oom_timeout(unsigned long); |
@@ -46,7 +43,7 @@ static unsigned vslow_work_proportion = 50; /* % of threads that may process | |||
46 | 43 | ||
47 | #ifdef CONFIG_SYSCTL | 44 | #ifdef CONFIG_SYSCTL |
48 | static const int slow_work_min_min_threads = 2; | 45 | static const int slow_work_min_min_threads = 2; |
49 | static int slow_work_max_max_threads = 255; | 46 | static int slow_work_max_max_threads = SLOW_WORK_THREAD_LIMIT; |
50 | static const int slow_work_min_vslow = 1; | 47 | static const int slow_work_min_vslow = 1; |
51 | static const int slow_work_max_vslow = 99; | 48 | static const int slow_work_max_vslow = 99; |
52 | 49 | ||
@@ -98,6 +95,32 @@ static DEFINE_TIMER(slow_work_oom_timer, slow_work_oom_timeout, 0, 0); | |||
98 | static struct slow_work slow_work_new_thread; /* new thread starter */ | 95 | static struct slow_work slow_work_new_thread; /* new thread starter */ |
99 | 96 | ||
100 | /* | 97 | /* |
98 | * slow work ID allocation (use slow_work_queue_lock) | ||
99 | */ | ||
100 | static DECLARE_BITMAP(slow_work_ids, SLOW_WORK_THREAD_LIMIT); | ||
101 | |||
102 | /* | ||
103 | * Unregistration tracking to prevent put_ref() from disappearing during module | ||
104 | * unload | ||
105 | */ | ||
106 | #ifdef CONFIG_MODULES | ||
107 | static struct module *slow_work_thread_processing[SLOW_WORK_THREAD_LIMIT]; | ||
108 | static struct module *slow_work_unreg_module; | ||
109 | static struct slow_work *slow_work_unreg_work_item; | ||
110 | static DECLARE_WAIT_QUEUE_HEAD(slow_work_unreg_wq); | ||
111 | static DEFINE_MUTEX(slow_work_unreg_sync_lock); | ||
112 | #endif | ||
113 | |||
114 | /* | ||
115 | * Data for tracking currently executing items for indication through /proc | ||
116 | */ | ||
117 | #ifdef CONFIG_SLOW_WORK_PROC | ||
118 | struct slow_work *slow_work_execs[SLOW_WORK_THREAD_LIMIT]; | ||
119 | pid_t slow_work_pids[SLOW_WORK_THREAD_LIMIT]; | ||
120 | DEFINE_RWLOCK(slow_work_execs_lock); | ||
121 | #endif | ||
122 | |||
123 | /* | ||
101 | * The queues of work items and the lock governing access to them. These are | 124 | * The queues of work items and the lock governing access to them. These are |
102 | * shared between all the CPUs. It doesn't make sense to have per-CPU queues | 125 | * shared between all the CPUs. It doesn't make sense to have per-CPU queues |
103 | * as the number of threads bears no relation to the number of CPUs. | 126 | * as the number of threads bears no relation to the number of CPUs. |
@@ -105,9 +128,18 @@ static struct slow_work slow_work_new_thread; /* new thread starter */ | |||
105 | * There are two queues of work items: one for slow work items, and one for | 128 | * There are two queues of work items: one for slow work items, and one for |
106 | * very slow work items. | 129 | * very slow work items. |
107 | */ | 130 | */ |
108 | static LIST_HEAD(slow_work_queue); | 131 | LIST_HEAD(slow_work_queue); |
109 | static LIST_HEAD(vslow_work_queue); | 132 | LIST_HEAD(vslow_work_queue); |
110 | static DEFINE_SPINLOCK(slow_work_queue_lock); | 133 | DEFINE_SPINLOCK(slow_work_queue_lock); |
134 | |||
135 | /* | ||
136 | * The following are two wait queues that get pinged when a work item is placed | ||
137 | * on an empty queue. These allow work items that are hogging a thread by | ||
138 | * sleeping in a way that could be deferred to yield their thread and enqueue | ||
139 | * themselves. | ||
140 | */ | ||
141 | static DECLARE_WAIT_QUEUE_HEAD(slow_work_queue_waits_for_occupation); | ||
142 | static DECLARE_WAIT_QUEUE_HEAD(vslow_work_queue_waits_for_occupation); | ||
111 | 143 | ||
112 | /* | 144 | /* |
113 | * The thread controls. A variable used to signal to the threads that they | 145 | * The thread controls. A variable used to signal to the threads that they |
@@ -126,6 +158,20 @@ static DECLARE_COMPLETION(slow_work_last_thread_exited); | |||
126 | static int slow_work_user_count; | 158 | static int slow_work_user_count; |
127 | static DEFINE_MUTEX(slow_work_user_lock); | 159 | static DEFINE_MUTEX(slow_work_user_lock); |
128 | 160 | ||
161 | static inline int slow_work_get_ref(struct slow_work *work) | ||
162 | { | ||
163 | if (work->ops->get_ref) | ||
164 | return work->ops->get_ref(work); | ||
165 | |||
166 | return 0; | ||
167 | } | ||
168 | |||
169 | static inline void slow_work_put_ref(struct slow_work *work) | ||
170 | { | ||
171 | if (work->ops->put_ref) | ||
172 | work->ops->put_ref(work); | ||
173 | } | ||
174 | |||
129 | /* | 175 | /* |
130 | * Calculate the maximum number of active threads in the pool that are | 176 | * Calculate the maximum number of active threads in the pool that are |
131 | * permitted to process very slow work items. | 177 | * permitted to process very slow work items. |
@@ -149,8 +195,11 @@ static unsigned slow_work_calc_vsmax(void) | |||
149 | * Attempt to execute stuff queued on a slow thread. Return true if we managed | 195 | * Attempt to execute stuff queued on a slow thread. Return true if we managed |
150 | * it, false if there was nothing to do. | 196 | * it, false if there was nothing to do. |
151 | */ | 197 | */ |
152 | static bool slow_work_execute(void) | 198 | static noinline bool slow_work_execute(int id) |
153 | { | 199 | { |
200 | #ifdef CONFIG_MODULES | ||
201 | struct module *module; | ||
202 | #endif | ||
154 | struct slow_work *work = NULL; | 203 | struct slow_work *work = NULL; |
155 | unsigned vsmax; | 204 | unsigned vsmax; |
156 | bool very_slow; | 205 | bool very_slow; |
@@ -186,6 +235,16 @@ static bool slow_work_execute(void) | |||
186 | } else { | 235 | } else { |
187 | very_slow = false; /* avoid the compiler warning */ | 236 | very_slow = false; /* avoid the compiler warning */ |
188 | } | 237 | } |
238 | |||
239 | #ifdef CONFIG_MODULES | ||
240 | if (work) | ||
241 | slow_work_thread_processing[id] = work->owner; | ||
242 | #endif | ||
243 | if (work) { | ||
244 | slow_work_mark_time(work); | ||
245 | slow_work_begin_exec(id, work); | ||
246 | } | ||
247 | |||
189 | spin_unlock_irq(&slow_work_queue_lock); | 248 | spin_unlock_irq(&slow_work_queue_lock); |
190 | 249 | ||
191 | if (!work) | 250 | if (!work) |
@@ -194,12 +253,19 @@ static bool slow_work_execute(void) | |||
194 | if (!test_and_clear_bit(SLOW_WORK_PENDING, &work->flags)) | 253 | if (!test_and_clear_bit(SLOW_WORK_PENDING, &work->flags)) |
195 | BUG(); | 254 | BUG(); |
196 | 255 | ||
197 | work->ops->execute(work); | 256 | /* don't execute if the work is in the process of being cancelled */ |
257 | if (!test_bit(SLOW_WORK_CANCELLING, &work->flags)) | ||
258 | work->ops->execute(work); | ||
198 | 259 | ||
199 | if (very_slow) | 260 | if (very_slow) |
200 | atomic_dec(&vslow_work_executing_count); | 261 | atomic_dec(&vslow_work_executing_count); |
201 | clear_bit_unlock(SLOW_WORK_EXECUTING, &work->flags); | 262 | clear_bit_unlock(SLOW_WORK_EXECUTING, &work->flags); |
202 | 263 | ||
264 | /* wake up anyone waiting for this work to be complete */ | ||
265 | wake_up_bit(&work->flags, SLOW_WORK_EXECUTING); | ||
266 | |||
267 | slow_work_end_exec(id, work); | ||
268 | |||
203 | /* if someone tried to enqueue the item whilst we were executing it, | 269 | /* if someone tried to enqueue the item whilst we were executing it, |
204 | * then it'll be left unenqueued to avoid multiple threads trying to | 270 | * then it'll be left unenqueued to avoid multiple threads trying to |
205 | * execute it simultaneously | 271 | * execute it simultaneously |
@@ -219,7 +285,18 @@ static bool slow_work_execute(void) | |||
219 | spin_unlock_irq(&slow_work_queue_lock); | 285 | spin_unlock_irq(&slow_work_queue_lock); |
220 | } | 286 | } |
221 | 287 | ||
222 | work->ops->put_ref(work); | 288 | /* sort out the race between module unloading and put_ref() */ |
289 | slow_work_put_ref(work); | ||
290 | |||
291 | #ifdef CONFIG_MODULES | ||
292 | module = slow_work_thread_processing[id]; | ||
293 | slow_work_thread_processing[id] = NULL; | ||
294 | smp_mb(); | ||
295 | if (slow_work_unreg_work_item == work || | ||
296 | slow_work_unreg_module == module) | ||
297 | wake_up_all(&slow_work_unreg_wq); | ||
298 | #endif | ||
299 | |||
223 | return true; | 300 | return true; |
224 | 301 | ||
225 | auto_requeue: | 302 | auto_requeue: |
@@ -227,15 +304,61 @@ auto_requeue: | |||
227 | * - we transfer our ref on the item back to the appropriate queue | 304 | * - we transfer our ref on the item back to the appropriate queue |
228 | * - don't wake another thread up as we're awake already | 305 | * - don't wake another thread up as we're awake already |
229 | */ | 306 | */ |
307 | slow_work_mark_time(work); | ||
230 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) | 308 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) |
231 | list_add_tail(&work->link, &vslow_work_queue); | 309 | list_add_tail(&work->link, &vslow_work_queue); |
232 | else | 310 | else |
233 | list_add_tail(&work->link, &slow_work_queue); | 311 | list_add_tail(&work->link, &slow_work_queue); |
234 | spin_unlock_irq(&slow_work_queue_lock); | 312 | spin_unlock_irq(&slow_work_queue_lock); |
313 | slow_work_thread_processing[id] = NULL; | ||
235 | return true; | 314 | return true; |
236 | } | 315 | } |
237 | 316 | ||
238 | /** | 317 | /** |
318 | * slow_work_sleep_till_thread_needed - Sleep till thread needed by other work | ||
319 | * work: The work item under execution that wants to sleep | ||
320 | * _timeout: Scheduler sleep timeout | ||
321 | * | ||
322 | * Allow a requeueable work item to sleep on a slow-work processor thread until | ||
323 | * that thread is needed to do some other work or the sleep is interrupted by | ||
324 | * some other event. | ||
325 | * | ||
326 | * The caller must set up a wake up event before calling this and must have set | ||
327 | * the appropriate sleep mode (such as TASK_UNINTERRUPTIBLE) and tested its own | ||
328 | * condition before calling this function as no test is made here. | ||
329 | * | ||
330 | * False is returned if there is nothing on the queue; true is returned if the | ||
331 | * work item should be requeued | ||
332 | */ | ||
333 | bool slow_work_sleep_till_thread_needed(struct slow_work *work, | ||
334 | signed long *_timeout) | ||
335 | { | ||
336 | wait_queue_head_t *wfo_wq; | ||
337 | struct list_head *queue; | ||
338 | |||
339 | DEFINE_WAIT(wait); | ||
340 | |||
341 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { | ||
342 | wfo_wq = &vslow_work_queue_waits_for_occupation; | ||
343 | queue = &vslow_work_queue; | ||
344 | } else { | ||
345 | wfo_wq = &slow_work_queue_waits_for_occupation; | ||
346 | queue = &slow_work_queue; | ||
347 | } | ||
348 | |||
349 | if (!list_empty(queue)) | ||
350 | return true; | ||
351 | |||
352 | add_wait_queue_exclusive(wfo_wq, &wait); | ||
353 | if (list_empty(queue)) | ||
354 | *_timeout = schedule_timeout(*_timeout); | ||
355 | finish_wait(wfo_wq, &wait); | ||
356 | |||
357 | return !list_empty(queue); | ||
358 | } | ||
359 | EXPORT_SYMBOL(slow_work_sleep_till_thread_needed); | ||
360 | |||
361 | /** | ||
239 | * slow_work_enqueue - Schedule a slow work item for processing | 362 | * slow_work_enqueue - Schedule a slow work item for processing |
240 | * @work: The work item to queue | 363 | * @work: The work item to queue |
241 | * | 364 | * |
@@ -260,16 +383,22 @@ auto_requeue: | |||
260 | * allowed to pick items to execute. This ensures that very slow items won't | 383 | * allowed to pick items to execute. This ensures that very slow items won't |
261 | * overly block ones that are just ordinarily slow. | 384 | * overly block ones that are just ordinarily slow. |
262 | * | 385 | * |
263 | * Returns 0 if successful, -EAGAIN if not. | 386 | * Returns 0 if successful, -EAGAIN if not (or -ECANCELED if cancelled work is |
387 | * attempted queued) | ||
264 | */ | 388 | */ |
265 | int slow_work_enqueue(struct slow_work *work) | 389 | int slow_work_enqueue(struct slow_work *work) |
266 | { | 390 | { |
391 | wait_queue_head_t *wfo_wq; | ||
392 | struct list_head *queue; | ||
267 | unsigned long flags; | 393 | unsigned long flags; |
394 | int ret; | ||
395 | |||
396 | if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) | ||
397 | return -ECANCELED; | ||
268 | 398 | ||
269 | BUG_ON(slow_work_user_count <= 0); | 399 | BUG_ON(slow_work_user_count <= 0); |
270 | BUG_ON(!work); | 400 | BUG_ON(!work); |
271 | BUG_ON(!work->ops); | 401 | BUG_ON(!work->ops); |
272 | BUG_ON(!work->ops->get_ref); | ||
273 | 402 | ||
274 | /* when honouring an enqueue request, we only promise that we will run | 403 | /* when honouring an enqueue request, we only promise that we will run |
275 | * the work function in the future; we do not promise to run it once | 404 | * the work function in the future; we do not promise to run it once |
@@ -280,8 +409,19 @@ int slow_work_enqueue(struct slow_work *work) | |||
280 | * maintaining our promise | 409 | * maintaining our promise |
281 | */ | 410 | */ |
282 | if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { | 411 | if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { |
412 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { | ||
413 | wfo_wq = &vslow_work_queue_waits_for_occupation; | ||
414 | queue = &vslow_work_queue; | ||
415 | } else { | ||
416 | wfo_wq = &slow_work_queue_waits_for_occupation; | ||
417 | queue = &slow_work_queue; | ||
418 | } | ||
419 | |||
283 | spin_lock_irqsave(&slow_work_queue_lock, flags); | 420 | spin_lock_irqsave(&slow_work_queue_lock, flags); |
284 | 421 | ||
422 | if (unlikely(test_bit(SLOW_WORK_CANCELLING, &work->flags))) | ||
423 | goto cancelled; | ||
424 | |||
285 | /* we promise that we will not attempt to execute the work | 425 | /* we promise that we will not attempt to execute the work |
286 | * function in more than one thread simultaneously | 426 | * function in more than one thread simultaneously |
287 | * | 427 | * |
@@ -299,25 +439,221 @@ int slow_work_enqueue(struct slow_work *work) | |||
299 | if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { | 439 | if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { |
300 | set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); | 440 | set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); |
301 | } else { | 441 | } else { |
302 | if (work->ops->get_ref(work) < 0) | 442 | ret = slow_work_get_ref(work); |
303 | goto cant_get_ref; | 443 | if (ret < 0) |
304 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) | 444 | goto failed; |
305 | list_add_tail(&work->link, &vslow_work_queue); | 445 | slow_work_mark_time(work); |
306 | else | 446 | list_add_tail(&work->link, queue); |
307 | list_add_tail(&work->link, &slow_work_queue); | ||
308 | wake_up(&slow_work_thread_wq); | 447 | wake_up(&slow_work_thread_wq); |
448 | |||
449 | /* if someone who could be requeued is sleeping on a | ||
450 | * thread, then ask them to yield their thread */ | ||
451 | if (work->link.prev == queue) | ||
452 | wake_up(wfo_wq); | ||
309 | } | 453 | } |
310 | 454 | ||
311 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); | 455 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); |
312 | } | 456 | } |
313 | return 0; | 457 | return 0; |
314 | 458 | ||
315 | cant_get_ref: | 459 | cancelled: |
460 | ret = -ECANCELED; | ||
461 | failed: | ||
316 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); | 462 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); |
317 | return -EAGAIN; | 463 | return ret; |
318 | } | 464 | } |
319 | EXPORT_SYMBOL(slow_work_enqueue); | 465 | EXPORT_SYMBOL(slow_work_enqueue); |
320 | 466 | ||
467 | static int slow_work_wait(void *word) | ||
468 | { | ||
469 | schedule(); | ||
470 | return 0; | ||
471 | } | ||
472 | |||
473 | /** | ||
474 | * slow_work_cancel - Cancel a slow work item | ||
475 | * @work: The work item to cancel | ||
476 | * | ||
477 | * This function will cancel a previously enqueued work item. If we cannot | ||
478 | * cancel the work item, it is guarenteed to have run when this function | ||
479 | * returns. | ||
480 | */ | ||
481 | void slow_work_cancel(struct slow_work *work) | ||
482 | { | ||
483 | bool wait = true, put = false; | ||
484 | |||
485 | set_bit(SLOW_WORK_CANCELLING, &work->flags); | ||
486 | smp_mb(); | ||
487 | |||
488 | /* if the work item is a delayed work item with an active timer, we | ||
489 | * need to wait for the timer to finish _before_ getting the spinlock, | ||
490 | * lest we deadlock against the timer routine | ||
491 | * | ||
492 | * the timer routine will leave DELAYED set if it notices the | ||
493 | * CANCELLING flag in time | ||
494 | */ | ||
495 | if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { | ||
496 | struct delayed_slow_work *dwork = | ||
497 | container_of(work, struct delayed_slow_work, work); | ||
498 | del_timer_sync(&dwork->timer); | ||
499 | } | ||
500 | |||
501 | spin_lock_irq(&slow_work_queue_lock); | ||
502 | |||
503 | if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { | ||
504 | /* the timer routine aborted or never happened, so we are left | ||
505 | * holding the timer's reference on the item and should just | ||
506 | * drop the pending flag and wait for any ongoing execution to | ||
507 | * finish */ | ||
508 | struct delayed_slow_work *dwork = | ||
509 | container_of(work, struct delayed_slow_work, work); | ||
510 | |||
511 | BUG_ON(timer_pending(&dwork->timer)); | ||
512 | BUG_ON(!list_empty(&work->link)); | ||
513 | |||
514 | clear_bit(SLOW_WORK_DELAYED, &work->flags); | ||
515 | put = true; | ||
516 | clear_bit(SLOW_WORK_PENDING, &work->flags); | ||
517 | |||
518 | } else if (test_bit(SLOW_WORK_PENDING, &work->flags) && | ||
519 | !list_empty(&work->link)) { | ||
520 | /* the link in the pending queue holds a reference on the item | ||
521 | * that we will need to release */ | ||
522 | list_del_init(&work->link); | ||
523 | wait = false; | ||
524 | put = true; | ||
525 | clear_bit(SLOW_WORK_PENDING, &work->flags); | ||
526 | |||
527 | } else if (test_and_clear_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags)) { | ||
528 | /* the executor is holding our only reference on the item, so | ||
529 | * we merely need to wait for it to finish executing */ | ||
530 | clear_bit(SLOW_WORK_PENDING, &work->flags); | ||
531 | } | ||
532 | |||
533 | spin_unlock_irq(&slow_work_queue_lock); | ||
534 | |||
535 | /* the EXECUTING flag is set by the executor whilst the spinlock is set | ||
536 | * and before the item is dequeued - so assuming the above doesn't | ||
537 | * actually dequeue it, simply waiting for the EXECUTING flag to be | ||
538 | * released here should be sufficient */ | ||
539 | if (wait) | ||
540 | wait_on_bit(&work->flags, SLOW_WORK_EXECUTING, slow_work_wait, | ||
541 | TASK_UNINTERRUPTIBLE); | ||
542 | |||
543 | clear_bit(SLOW_WORK_CANCELLING, &work->flags); | ||
544 | if (put) | ||
545 | slow_work_put_ref(work); | ||
546 | } | ||
547 | EXPORT_SYMBOL(slow_work_cancel); | ||
548 | |||
549 | /* | ||
550 | * Handle expiry of the delay timer, indicating that a delayed slow work item | ||
551 | * should now be queued if not cancelled | ||
552 | */ | ||
553 | static void delayed_slow_work_timer(unsigned long data) | ||
554 | { | ||
555 | wait_queue_head_t *wfo_wq; | ||
556 | struct list_head *queue; | ||
557 | struct slow_work *work = (struct slow_work *) data; | ||
558 | unsigned long flags; | ||
559 | bool queued = false, put = false, first = false; | ||
560 | |||
561 | if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { | ||
562 | wfo_wq = &vslow_work_queue_waits_for_occupation; | ||
563 | queue = &vslow_work_queue; | ||
564 | } else { | ||
565 | wfo_wq = &slow_work_queue_waits_for_occupation; | ||
566 | queue = &slow_work_queue; | ||
567 | } | ||
568 | |||
569 | spin_lock_irqsave(&slow_work_queue_lock, flags); | ||
570 | if (likely(!test_bit(SLOW_WORK_CANCELLING, &work->flags))) { | ||
571 | clear_bit(SLOW_WORK_DELAYED, &work->flags); | ||
572 | |||
573 | if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { | ||
574 | /* we discard the reference the timer was holding in | ||
575 | * favour of the one the executor holds */ | ||
576 | set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); | ||
577 | put = true; | ||
578 | } else { | ||
579 | slow_work_mark_time(work); | ||
580 | list_add_tail(&work->link, queue); | ||
581 | queued = true; | ||
582 | if (work->link.prev == queue) | ||
583 | first = true; | ||
584 | } | ||
585 | } | ||
586 | |||
587 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); | ||
588 | if (put) | ||
589 | slow_work_put_ref(work); | ||
590 | if (first) | ||
591 | wake_up(wfo_wq); | ||
592 | if (queued) | ||
593 | wake_up(&slow_work_thread_wq); | ||
594 | } | ||
595 | |||
596 | /** | ||
597 | * delayed_slow_work_enqueue - Schedule a delayed slow work item for processing | ||
598 | * @dwork: The delayed work item to queue | ||
599 | * @delay: When to start executing the work, in jiffies from now | ||
600 | * | ||
601 | * This is similar to slow_work_enqueue(), but it adds a delay before the work | ||
602 | * is actually queued for processing. | ||
603 | * | ||
604 | * The item can have delayed processing requested on it whilst it is being | ||
605 | * executed. The delay will begin immediately, and if it expires before the | ||
606 | * item finishes executing, the item will be placed back on the queue when it | ||
607 | * has done executing. | ||
608 | */ | ||
609 | int delayed_slow_work_enqueue(struct delayed_slow_work *dwork, | ||
610 | unsigned long delay) | ||
611 | { | ||
612 | struct slow_work *work = &dwork->work; | ||
613 | unsigned long flags; | ||
614 | int ret; | ||
615 | |||
616 | if (delay == 0) | ||
617 | return slow_work_enqueue(&dwork->work); | ||
618 | |||
619 | BUG_ON(slow_work_user_count <= 0); | ||
620 | BUG_ON(!work); | ||
621 | BUG_ON(!work->ops); | ||
622 | |||
623 | if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) | ||
624 | return -ECANCELED; | ||
625 | |||
626 | if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { | ||
627 | spin_lock_irqsave(&slow_work_queue_lock, flags); | ||
628 | |||
629 | if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) | ||
630 | goto cancelled; | ||
631 | |||
632 | /* the timer holds a reference whilst it is pending */ | ||
633 | ret = work->ops->get_ref(work); | ||
634 | if (ret < 0) | ||
635 | goto cant_get_ref; | ||
636 | |||
637 | if (test_and_set_bit(SLOW_WORK_DELAYED, &work->flags)) | ||
638 | BUG(); | ||
639 | dwork->timer.expires = jiffies + delay; | ||
640 | dwork->timer.data = (unsigned long) work; | ||
641 | dwork->timer.function = delayed_slow_work_timer; | ||
642 | add_timer(&dwork->timer); | ||
643 | |||
644 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); | ||
645 | } | ||
646 | |||
647 | return 0; | ||
648 | |||
649 | cancelled: | ||
650 | ret = -ECANCELED; | ||
651 | cant_get_ref: | ||
652 | spin_unlock_irqrestore(&slow_work_queue_lock, flags); | ||
653 | return ret; | ||
654 | } | ||
655 | EXPORT_SYMBOL(delayed_slow_work_enqueue); | ||
656 | |||
321 | /* | 657 | /* |
322 | * Schedule a cull of the thread pool at some time in the near future | 658 | * Schedule a cull of the thread pool at some time in the near future |
323 | */ | 659 | */ |
@@ -368,13 +704,23 @@ static inline bool slow_work_available(int vsmax) | |||
368 | */ | 704 | */ |
369 | static int slow_work_thread(void *_data) | 705 | static int slow_work_thread(void *_data) |
370 | { | 706 | { |
371 | int vsmax; | 707 | int vsmax, id; |
372 | 708 | ||
373 | DEFINE_WAIT(wait); | 709 | DEFINE_WAIT(wait); |
374 | 710 | ||
375 | set_freezable(); | 711 | set_freezable(); |
376 | set_user_nice(current, -5); | 712 | set_user_nice(current, -5); |
377 | 713 | ||
714 | /* allocate ourselves an ID */ | ||
715 | spin_lock_irq(&slow_work_queue_lock); | ||
716 | id = find_first_zero_bit(slow_work_ids, SLOW_WORK_THREAD_LIMIT); | ||
717 | BUG_ON(id < 0 || id >= SLOW_WORK_THREAD_LIMIT); | ||
718 | __set_bit(id, slow_work_ids); | ||
719 | slow_work_set_thread_pid(id, current->pid); | ||
720 | spin_unlock_irq(&slow_work_queue_lock); | ||
721 | |||
722 | sprintf(current->comm, "kslowd%03u", id); | ||
723 | |||
378 | for (;;) { | 724 | for (;;) { |
379 | vsmax = vslow_work_proportion; | 725 | vsmax = vslow_work_proportion; |
380 | vsmax *= atomic_read(&slow_work_thread_count); | 726 | vsmax *= atomic_read(&slow_work_thread_count); |
@@ -395,7 +741,7 @@ static int slow_work_thread(void *_data) | |||
395 | vsmax *= atomic_read(&slow_work_thread_count); | 741 | vsmax *= atomic_read(&slow_work_thread_count); |
396 | vsmax /= 100; | 742 | vsmax /= 100; |
397 | 743 | ||
398 | if (slow_work_available(vsmax) && slow_work_execute()) { | 744 | if (slow_work_available(vsmax) && slow_work_execute(id)) { |
399 | cond_resched(); | 745 | cond_resched(); |
400 | if (list_empty(&slow_work_queue) && | 746 | if (list_empty(&slow_work_queue) && |
401 | list_empty(&vslow_work_queue) && | 747 | list_empty(&vslow_work_queue) && |
@@ -412,6 +758,11 @@ static int slow_work_thread(void *_data) | |||
412 | break; | 758 | break; |
413 | } | 759 | } |
414 | 760 | ||
761 | spin_lock_irq(&slow_work_queue_lock); | ||
762 | slow_work_set_thread_pid(id, 0); | ||
763 | __clear_bit(id, slow_work_ids); | ||
764 | spin_unlock_irq(&slow_work_queue_lock); | ||
765 | |||
415 | if (atomic_dec_and_test(&slow_work_thread_count)) | 766 | if (atomic_dec_and_test(&slow_work_thread_count)) |
416 | complete_and_exit(&slow_work_last_thread_exited, 0); | 767 | complete_and_exit(&slow_work_last_thread_exited, 0); |
417 | return 0; | 768 | return 0; |
@@ -427,21 +778,6 @@ static void slow_work_cull_timeout(unsigned long data) | |||
427 | } | 778 | } |
428 | 779 | ||
429 | /* | 780 | /* |
430 | * Get a reference on slow work thread starter | ||
431 | */ | ||
432 | static int slow_work_new_thread_get_ref(struct slow_work *work) | ||
433 | { | ||
434 | return 0; | ||
435 | } | ||
436 | |||
437 | /* | ||
438 | * Drop a reference on slow work thread starter | ||
439 | */ | ||
440 | static void slow_work_new_thread_put_ref(struct slow_work *work) | ||
441 | { | ||
442 | } | ||
443 | |||
444 | /* | ||
445 | * Start a new slow work thread | 781 | * Start a new slow work thread |
446 | */ | 782 | */ |
447 | static void slow_work_new_thread_execute(struct slow_work *work) | 783 | static void slow_work_new_thread_execute(struct slow_work *work) |
@@ -475,9 +811,11 @@ static void slow_work_new_thread_execute(struct slow_work *work) | |||
475 | } | 811 | } |
476 | 812 | ||
477 | static const struct slow_work_ops slow_work_new_thread_ops = { | 813 | static const struct slow_work_ops slow_work_new_thread_ops = { |
478 | .get_ref = slow_work_new_thread_get_ref, | 814 | .owner = THIS_MODULE, |
479 | .put_ref = slow_work_new_thread_put_ref, | ||
480 | .execute = slow_work_new_thread_execute, | 815 | .execute = slow_work_new_thread_execute, |
816 | #ifdef CONFIG_SLOW_WORK_PROC | ||
817 | .desc = slow_work_new_thread_desc, | ||
818 | #endif | ||
481 | }; | 819 | }; |
482 | 820 | ||
483 | /* | 821 | /* |
@@ -546,12 +884,13 @@ static int slow_work_max_threads_sysctl(struct ctl_table *table, int write, | |||
546 | 884 | ||
547 | /** | 885 | /** |
548 | * slow_work_register_user - Register a user of the facility | 886 | * slow_work_register_user - Register a user of the facility |
887 | * @module: The module about to make use of the facility | ||
549 | * | 888 | * |
550 | * Register a user of the facility, starting up the initial threads if there | 889 | * Register a user of the facility, starting up the initial threads if there |
551 | * aren't any other users at this point. This will return 0 if successful, or | 890 | * aren't any other users at this point. This will return 0 if successful, or |
552 | * an error if not. | 891 | * an error if not. |
553 | */ | 892 | */ |
554 | int slow_work_register_user(void) | 893 | int slow_work_register_user(struct module *module) |
555 | { | 894 | { |
556 | struct task_struct *p; | 895 | struct task_struct *p; |
557 | int loop; | 896 | int loop; |
@@ -598,14 +937,79 @@ error: | |||
598 | } | 937 | } |
599 | EXPORT_SYMBOL(slow_work_register_user); | 938 | EXPORT_SYMBOL(slow_work_register_user); |
600 | 939 | ||
940 | /* | ||
941 | * wait for all outstanding items from the calling module to complete | ||
942 | * - note that more items may be queued whilst we're waiting | ||
943 | */ | ||
944 | static void slow_work_wait_for_items(struct module *module) | ||
945 | { | ||
946 | DECLARE_WAITQUEUE(myself, current); | ||
947 | struct slow_work *work; | ||
948 | int loop; | ||
949 | |||
950 | mutex_lock(&slow_work_unreg_sync_lock); | ||
951 | add_wait_queue(&slow_work_unreg_wq, &myself); | ||
952 | |||
953 | for (;;) { | ||
954 | spin_lock_irq(&slow_work_queue_lock); | ||
955 | |||
956 | /* first of all, we wait for the last queued item in each list | ||
957 | * to be processed */ | ||
958 | list_for_each_entry_reverse(work, &vslow_work_queue, link) { | ||
959 | if (work->owner == module) { | ||
960 | set_current_state(TASK_UNINTERRUPTIBLE); | ||
961 | slow_work_unreg_work_item = work; | ||
962 | goto do_wait; | ||
963 | } | ||
964 | } | ||
965 | list_for_each_entry_reverse(work, &slow_work_queue, link) { | ||
966 | if (work->owner == module) { | ||
967 | set_current_state(TASK_UNINTERRUPTIBLE); | ||
968 | slow_work_unreg_work_item = work; | ||
969 | goto do_wait; | ||
970 | } | ||
971 | } | ||
972 | |||
973 | /* then we wait for the items being processed to finish */ | ||
974 | slow_work_unreg_module = module; | ||
975 | smp_mb(); | ||
976 | for (loop = 0; loop < SLOW_WORK_THREAD_LIMIT; loop++) { | ||
977 | if (slow_work_thread_processing[loop] == module) | ||
978 | goto do_wait; | ||
979 | } | ||
980 | spin_unlock_irq(&slow_work_queue_lock); | ||
981 | break; /* okay, we're done */ | ||
982 | |||
983 | do_wait: | ||
984 | spin_unlock_irq(&slow_work_queue_lock); | ||
985 | schedule(); | ||
986 | slow_work_unreg_work_item = NULL; | ||
987 | slow_work_unreg_module = NULL; | ||
988 | } | ||
989 | |||
990 | remove_wait_queue(&slow_work_unreg_wq, &myself); | ||
991 | mutex_unlock(&slow_work_unreg_sync_lock); | ||
992 | } | ||
993 | |||
601 | /** | 994 | /** |
602 | * slow_work_unregister_user - Unregister a user of the facility | 995 | * slow_work_unregister_user - Unregister a user of the facility |
996 | * @module: The module whose items should be cleared | ||
603 | * | 997 | * |
604 | * Unregister a user of the facility, killing all the threads if this was the | 998 | * Unregister a user of the facility, killing all the threads if this was the |
605 | * last one. | 999 | * last one. |
1000 | * | ||
1001 | * This waits for all the work items belonging to the nominated module to go | ||
1002 | * away before proceeding. | ||
606 | */ | 1003 | */ |
607 | void slow_work_unregister_user(void) | 1004 | void slow_work_unregister_user(struct module *module) |
608 | { | 1005 | { |
1006 | /* first of all, wait for all outstanding items from the calling module | ||
1007 | * to complete */ | ||
1008 | if (module) | ||
1009 | slow_work_wait_for_items(module); | ||
1010 | |||
1011 | /* then we can actually go about shutting down the facility if need | ||
1012 | * be */ | ||
609 | mutex_lock(&slow_work_user_lock); | 1013 | mutex_lock(&slow_work_user_lock); |
610 | 1014 | ||
611 | BUG_ON(slow_work_user_count <= 0); | 1015 | BUG_ON(slow_work_user_count <= 0); |
@@ -639,6 +1043,10 @@ static int __init init_slow_work(void) | |||
639 | if (slow_work_max_max_threads < nr_cpus * 2) | 1043 | if (slow_work_max_max_threads < nr_cpus * 2) |
640 | slow_work_max_max_threads = nr_cpus * 2; | 1044 | slow_work_max_max_threads = nr_cpus * 2; |
641 | #endif | 1045 | #endif |
1046 | #ifdef CONFIG_SLOW_WORK_PROC | ||
1047 | proc_create("slow_work_rq", S_IFREG | 0400, NULL, | ||
1048 | &slow_work_runqueue_fops); | ||
1049 | #endif | ||
642 | return 0; | 1050 | return 0; |
643 | } | 1051 | } |
644 | 1052 | ||