summaryrefslogtreecommitdiff
path: root/net/sunrpc/sched.c
blob: b91fd9c597b45158cc9ce88a8fb4caa5130faf96 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
/*
 * linux/net/sunrpc/sched.c
 *
 * Scheduling for synchronous and asynchronous RPC requests.
 *
 * Copyright (C) 1996 Olaf Kirch, <okir@monad.swb.de>
 *
 * TCP NFS related read + write fixes
 * (C) 1999 Dave Airlie, University of Limerick, Ireland <airlied@linux.ie>
 */

#include <linux/module.h>

#include <linux/sched.h>
#include <linux/interrupt.h>
#include <linux/slab.h>
#include <linux/mempool.h>
#include <linux/smp.h>
#include <linux/spinlock.h>
#include <linux/mutex.h>
#include <linux/freezer.h>

#include <linux/sunrpc/clnt.h>

#include "sunrpc.h"

#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
#define RPCDBG_FACILITY		RPCDBG_SCHED
#endif

#define CREATE_TRACE_POINTS
#include <trace/events/sunrpc.h>

/*
 * RPC slabs and memory pools
 */
#define RPC_BUFFER_MAXSIZE	(2048)
#define RPC_BUFFER_POOLSIZE	(8)
#define RPC_TASK_POOLSIZE	(8)
static struct kmem_cache	*rpc_task_slabp __read_mostly;
static struct kmem_cache	*rpc_buffer_slabp __read_mostly;
static mempool_t	*rpc_task_mempool __read_mostly;
static mempool_t	*rpc_buffer_mempool __read_mostly;

static void			rpc_async_schedule(struct work_struct *);
static void			 rpc_release_task(struct rpc_task *task);
static void __rpc_queue_timer_fn(unsigned long ptr);

/*
 * RPC tasks sit here while waiting for conditions to improve.
 */
static struct rpc_wait_queue delay_queue;

/*
 * rpciod-related stuff
 */
struct workqueue_struct *rpciod_workqueue;

/*
 * Disable the timer for a given RPC task. Should be called with
 * queue->lock and bh_disabled in order to avoid races within
 * rpc_run_timer().
 */
static void
__rpc_disable_timer(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	if (task->tk_timeout == 0)
		return;
	dprintk("RPC: %5u disabling timer\n", task->tk_pid);
	task->tk_timeout = 0;
	list_del(&task->u.tk_wait.timer_list);
	if (list_empty(&queue->timer_list.list))
		del_timer(&queue->timer_list.timer);
}

static void
rpc_set_queue_timer(struct rpc_wait_queue *queue, unsigned long expires)
{
	queue->timer_list.expires = expires;
	mod_timer(&queue->timer_list.timer, expires);
}

/*
 * Set up a timer for the current task.
 */
static void
__rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	if (!task->tk_timeout)
		return;

	dprintk("RPC: %5u setting alarm for %lu ms\n",
			task->tk_pid, task->tk_timeout * 1000 / HZ);

	task->u.tk_wait.expires = jiffies + task->tk_timeout;
	if (list_empty(&queue->timer_list.list) || time_before(task->u.tk_wait.expires, queue->timer_list.expires))
		rpc_set_queue_timer(queue, task->u.tk_wait.expires);
	list_add(&task->u.tk_wait.timer_list, &queue->timer_list.list);
}

static void rpc_rotate_queue_owner(struct rpc_wait_queue *queue)
{
	struct list_head *q = &queue->tasks[queue->priority];
	struct rpc_task *task;

	if (!list_empty(q)) {
		task = list_first_entry(q, struct rpc_task, u.tk_wait.list);
		if (task->tk_owner == queue->owner)
			list_move_tail(&task->u.tk_wait.list, q);
	}
}

static void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority)
{
	if (queue->priority != priority) {
		/* Fairness: rotate the list when changing priority */
		rpc_rotate_queue_owner(queue);
		queue->priority = priority;
	}
}

static void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid)
{
	queue->owner = pid;
	queue->nr = RPC_BATCH_COUNT;
}

static void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue)
{
	rpc_set_waitqueue_priority(queue, queue->maxpriority);
	rpc_set_waitqueue_owner(queue, 0);
}

/*
 * Add new request to a priority queue.
 */
static void __rpc_add_wait_queue_priority(struct rpc_wait_queue *queue,
		struct rpc_task *task,
		unsigned char queue_priority)
{
	struct list_head *q;
	struct rpc_task *t;

	INIT_LIST_HEAD(&task->u.tk_wait.links);
	if (unlikely(queue_priority > queue->maxpriority))
		queue_priority = queue->maxpriority;
	if (queue_priority > queue->priority)
		rpc_set_waitqueue_priority(queue, queue_priority);
	q = &queue->tasks[queue_priority];
	list_for_each_entry(t, q, u.tk_wait.list) {
		if (t->tk_owner == task->tk_owner) {
			list_add_tail(&task->u.tk_wait.list, &t->u.tk_wait.links);
			return;
		}
	}
	list_add_tail(&task->u.tk_wait.list, q);
}

/*
 * Add new request to wait queue.
 *
 * Swapper tasks always get inserted at the head of the queue.
 * This should avoid many nasty memory deadlocks and hopefully
 * improve overall performance.
 * Everyone else gets appended to the queue to ensure proper FIFO behavior.
 */
static void __rpc_add_wait_queue(struct rpc_wait_queue *queue,
		struct rpc_task *task,
		unsigned char queue_priority)
{
	WARN_ON_ONCE(RPC_IS_QUEUED(task));
	if (RPC_IS_QUEUED(task))
		return;

	if (RPC_IS_PRIORITY(queue))
		__rpc_add_wait_queue_priority(queue, task, queue_priority);
	else if (RPC_IS_SWAPPER(task))
		list_add(&task->u.tk_wait.list, &queue->tasks[0]);
	else
		list_add_tail(&task->u.tk_wait.list, &queue->tasks[0]);
	task->tk_waitqueue = queue;
	queue->qlen++;
	/* barrier matches the read in rpc_wake_up_task_queue_locked() */
	smp_wmb();
	rpc_set_queued(task);

	dprintk("RPC: %5u added to queue %p \"%s\"\n",
			task->tk_pid, queue, rpc_qname(queue));
}

/*
 * Remove request from a priority queue.
 */
static void __rpc_remove_wait_queue_priority(struct rpc_task *task)
{
	struct rpc_task *t;

	if (!list_empty(&task->u.tk_wait.links)) {
		t = list_entry(task->u.tk_wait.links.next, struct rpc_task, u.tk_wait.list);
		list_move(&t->u.tk_wait.list, &task->u.tk_wait.list);
		list_splice_init(&task->u.tk_wait.links, &t->u.tk_wait.links);
	}
}

/*
 * Remove request from queue.
 * Note: must be called with spin lock held.
 */
static void __rpc_remove_wait_queue(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	__rpc_disable_timer(queue, task);
	if (RPC_IS_PRIORITY(queue))
		__rpc_remove_wait_queue_priority(task);
	list_del(&task->u.tk_wait.list);
	queue->qlen--;
	dprintk("RPC: %5u removed from queue %p \"%s\"\n",
			task->tk_pid, queue, rpc_qname(queue));
}

static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, unsigned char nr_queues)
{
	int i;

	spin_lock_init(&queue->lock);
	for (i = 0; i < ARRAY_SIZE(queue->tasks); i++)
		INIT_LIST_HEAD(&queue->tasks[i]);
	queue->maxpriority = nr_queues - 1;
	rpc_reset_waitqueue_priority(queue);
	queue->qlen = 0;
	setup_timer(&queue->timer_list.timer, __rpc_queue_timer_fn, (unsigned long)queue);
	INIT_LIST_HEAD(&queue->timer_list.list);
	rpc_assign_waitqueue_name(queue, qname);
}

void rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname)
{
	__rpc_init_priority_wait_queue(queue, qname, RPC_NR_PRIORITY);
}
EXPORT_SYMBOL_GPL(rpc_init_priority_wait_queue);

void rpc_init_wait_queue(struct rpc_wait_queue *queue, const char *qname)
{
	__rpc_init_priority_wait_queue(queue, qname, 1);
}
EXPORT_SYMBOL_GPL(rpc_init_wait_queue);

void rpc_destroy_wait_queue(struct rpc_wait_queue *queue)
{
	del_timer_sync(&queue->timer_list.timer);
}
EXPORT_SYMBOL_GPL(rpc_destroy_wait_queue);

static int rpc_wait_bit_killable(struct wait_bit_key *key)
{
	if (fatal_signal_pending(current))
		return -ERESTARTSYS;
	freezable_schedule_unsafe();
	return 0;
}

#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || IS_ENABLED(CONFIG_TRACEPOINTS)
static void rpc_task_set_debuginfo(struct rpc_task *task)
{
	static atomic_t rpc_pid;

	task->tk_pid = atomic_inc_return(&rpc_pid);
}
#else
static inline void rpc_task_set_debuginfo(struct rpc_task *task)
{
}
#endif

static void rpc_set_active(struct rpc_task *task)
{
	trace_rpc_task_begin(task->tk_client, task, NULL);

	rpc_task_set_debuginfo(task);
	set_bit(RPC_TASK_ACTIVE, &task->tk_runstate);
}

/*
 * Mark an RPC call as having completed by clearing the 'active' bit
 * and then waking up all tasks that were sleeping.
 */
static int rpc_complete_task(struct rpc_task *task)
{
	void *m = &task->tk_runstate;
	wait_queue_head_t *wq = bit_waitqueue(m, RPC_TASK_ACTIVE);
	struct wait_bit_key k = __WAIT_BIT_KEY_INITIALIZER(m, RPC_TASK_ACTIVE);
	unsigned long flags;
	int ret;

	trace_rpc_task_complete(task->tk_client, task, NULL);

	spin_lock_irqsave(&wq->lock, flags);
	clear_bit(RPC_TASK_ACTIVE, &task->tk_runstate);
	ret = atomic_dec_and_test(&task->tk_count);
	if (waitqueue_active(wq))
		__wake_up_locked_key(wq, TASK_NORMAL, &k);
	spin_unlock_irqrestore(&wq->lock, flags);
	return ret;
}

/*
 * Allow callers to wait for completion of an RPC call
 *
 * Note the use of out_of_line_wait_on_bit() rather than wait_on_bit()
 * to enforce taking of the wq->lock and hence avoid races with
 * rpc_complete_task().
 */
int __rpc_wait_for_completion_task(struct rpc_task *task, wait_bit_action_f *action)
{
	if (action == NULL)
		action = rpc_wait_bit_killable;
	return out_of_line_wait_on_bit(&task->tk_runstate, RPC_TASK_ACTIVE,
			action, TASK_KILLABLE);
}
EXPORT_SYMBOL_GPL(__rpc_wait_for_completion_task);

/*
 * Make an RPC task runnable.
 *
 * Note: If the task is ASYNC, and is being made runnable after sitting on an
 * rpc_wait_queue, this must be called with the queue spinlock held to protect
 * the wait queue operation.
 * Note the ordering of rpc_test_and_set_running() and rpc_clear_queued(),
 * which is needed to ensure that __rpc_execute() doesn't loop (due to the
 * lockless RPC_IS_QUEUED() test) before we've had a chance to test
 * the RPC_TASK_RUNNING flag.
 */
static void rpc_make_runnable(struct rpc_task *task)
{
	bool need_wakeup = !rpc_test_and_set_running(task);

	rpc_clear_queued(task);
	if (!need_wakeup)
		return;
	if (RPC_IS_ASYNC(task)) {
		INIT_WORK(&task->u.tk_work, rpc_async_schedule);
		queue_work(rpciod_workqueue, &task->u.tk_work);
	} else
		wake_up_bit(&task->tk_runstate, RPC_TASK_QUEUED);
}

/*
 * Prepare for sleeping on a wait queue.
 * By always appending tasks to the list we ensure FIFO behavior.
 * NB: An RPC task will only receive interrupt-driven events as long
 * as it's on a wait queue.
 */
static void __rpc_sleep_on_priority(struct rpc_wait_queue *q,
		struct rpc_task *task,
		rpc_action action,
		unsigned char queue_priority)
{
	dprintk("RPC: %5u sleep_on(queue \"%s\" time %lu)\n",
			task->tk_pid, rpc_qname(q), jiffies);

	trace_rpc_task_sleep(task->tk_client, task, q);

	__rpc_add_wait_queue(q, task, queue_priority);

	WARN_ON_ONCE(task->tk_callback != NULL);
	task->tk_callback = action;
	__rpc_add_timer(q, task);
}

void rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task,
				rpc_action action)
{
	/* We shouldn't ever put an inactive task to sleep */
	WARN_ON_ONCE(!RPC_IS_ACTIVATED(task));
	if (!RPC_IS_ACTIVATED(task)) {
		task->tk_status = -EIO;
		rpc_put_task_async(task);
		return;
	}

	/*
	 * Protect the queue operations.
	 */
	spin_lock_bh(&q->lock);
	__rpc_sleep_on_priority(q, task, action, task->tk_priority);
	spin_unlock_bh(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on);

void rpc_sleep_on_priority(struct rpc_wait_queue *q, struct rpc_task *task,
		rpc_action action, int priority)
{
	/* We shouldn't ever put an inactive task to sleep */
	WARN_ON_ONCE(!RPC_IS_ACTIVATED(task));
	if (!RPC_IS_ACTIVATED(task)) {
		task->tk_status = -EIO;
		rpc_put_task_async(task);
		return;
	}

	/*
	 * Protect the queue operations.
	 */
	spin_lock_bh(&q->lock);
	__rpc_sleep_on_priority(q, task, action, priority - RPC_PRIORITY_LOW);
	spin_unlock_bh(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on_priority);

/**
 * __rpc_do_wake_up_task - wake up a single rpc_task
 * @queue: wait queue
 * @task: task to be woken up
 *
 * Caller must hold queue->lock, and have cleared the task queued flag.
 */
static void __rpc_do_wake_up_task(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	dprintk("RPC: %5u __rpc_wake_up_task (now %lu)\n",
			task->tk_pid, jiffies);

	/* Has the task been executed yet? If not, we cannot wake it up! */
	if (!RPC_IS_ACTIVATED(task)) {
		printk(KERN_ERR "RPC: Inactive task (%p) being woken up!\n", task);
		return;
	}

	trace_rpc_task_wakeup(task->tk_client, task, queue);

	__rpc_remove_wait_queue(queue, task);

	rpc_make_runnable(task);

	dprintk("RPC:       __rpc_wake_up_task done\n");
}

/*
 * Wake up a queued task while the queue lock is being held
 */
static void rpc_wake_up_task_queue_locked(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	if (RPC_IS_QUEUED(task)) {
		smp_rmb();
		if (task->tk_waitqueue == queue)
			__rpc_do_wake_up_task(queue, task);
	}
}

/*
 * Wake up a task on a specific queue
 */
void rpc_wake_up_queued_task(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	spin_lock_bh(&queue->lock);
	rpc_wake_up_task_queue_locked(queue, task);
	spin_unlock_bh(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up_queued_task);

/*
 * Wake up the next task on a priority queue.
 */
static struct rpc_task *__rpc_find_next_queued_priority(struct rpc_wait_queue *queue)
{
	struct list_head *q;
	struct rpc_task *task;

	/*
	 * Service a batch of tasks from a single owner.
	 */
	q = &queue->tasks[queue->priority];
	if (!list_empty(q)) {
		task = list_entry(q->next, struct rpc_task, u.tk_wait.list);
		if (queue->owner == task->tk_owner) {
			if (--queue->nr)
				goto out;
			list_move_tail(&task->u.tk_wait.list, q);
		}
		/*
		 * Check if we need to switch queues.
		 */
		goto new_owner;
	}

	/*
	 * Service the next queue.
	 */
	do {
		if (q == &queue->tasks[0])
			q = &queue->tasks[queue->maxpriority];
		else
			q = q - 1;
		if (!list_empty(q)) {
			task = list_entry(q->next, struct rpc_task, u.tk_wait.list);
			goto new_queue;
		}
	} while (q != &queue->tasks[queue->priority]);

	rpc_reset_waitqueue_priority(queue);
	return NULL;

new_queue:
	rpc_set_waitqueue_priority(queue, (unsigned int)(q - &queue->tasks[0]));
new_owner:
	rpc_set_waitqueue_owner(queue, task->tk_owner);
out:
	return task;
}

static struct rpc_task *__rpc_find_next_queued(struct rpc_wait_queue *queue)
{
	if (RPC_IS_PRIORITY(queue))
		return __rpc_find_next_queued_priority(queue);
	if (!list_empty(&queue->tasks[0]))
		return list_first_entry(&queue->tasks[0], struct rpc_task, u.tk_wait.list);
	return NULL;
}

/*
 * Wake up the first task on the wait queue.
 */
struct rpc_task *rpc_wake_up_first(struct rpc_wait_queue *queue,
		bool (*func)(struct rpc_task *, void *), void *data)
{
	struct rpc_task	*task = NULL;

	dprintk("RPC:       wake_up_first(%p \"%s\")\n",
			queue, rpc_qname(queue));
	spin_lock_bh(&queue->lock);
	task = __rpc_find_next_queued(queue);
	if (task != NULL) {
		if (func(task, data))
			rpc_wake_up_task_queue_locked(queue, task);
		else
			task = NULL;
	}
	spin_unlock_bh(&queue->lock);

	return task;
}
EXPORT_SYMBOL_GPL(rpc_wake_up_first);

static bool rpc_wake_up_next_func(struct rpc_task *task, void *data)
{
	return true;
}

/*
 * Wake up the next task on the wait queue.
*/
struct rpc_task *rpc_wake_up_next(struct rpc_wait_queue *queue)
{
	return rpc_wake_up_first(queue, rpc_wake_up_next_func, NULL);
}
EXPORT_SYMBOL_GPL(rpc_wake_up_next);

/**
 * rpc_wake_up - wake up all rpc_tasks
 * @queue: rpc_wait_queue on which the tasks are sleeping
 *
 * Grabs queue->lock
 */
void rpc_wake_up(struct rpc_wait_queue *queue)
{
	struct list_head *head;

	spin_lock_bh(&queue->lock);
	head = &queue->tasks[queue->maxpriority];
	for (;;) {
		while (!list_empty(head)) {
			struct rpc_task *task;
			task = list_first_entry(head,
					struct rpc_task,
					u.tk_wait.list);
			rpc_wake_up_task_queue_locked(queue, task);
		}
		if (head == &queue->tasks[0])
			break;
		head--;
	}
	spin_unlock_bh(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up);

/**
 * rpc_wake_up_status - wake up all rpc_tasks and set their status value.
 * @queue: rpc_wait_queue on which the tasks are sleeping
 * @status: status value to set
 *
 * Grabs queue->lock
 */
void rpc_wake_up_status(struct rpc_wait_queue *queue, int status)
{
	struct list_head *head;

	spin_lock_bh(&queue->lock);
	head = &queue->tasks[queue->maxpriority];
	for (;;) {
		while (!list_empty(head)) {
			struct rpc_task *task;
			task = list_first_entry(head,
					struct rpc_task,
					u.tk_wait.list);
			task->tk_status = status;
			rpc_wake_up_task_queue_locked(queue, task);
		}
		if (head == &queue->tasks[0])
			break;
		head--;
	}
	spin_unlock_bh(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up_status);

static void __rpc_queue_timer_fn(unsigned long ptr)
{
	struct rpc_wait_queue *queue = (struct rpc_wait_queue *)ptr;
	struct rpc_task *task, *n;
	unsigned long expires, now, timeo;

	spin_lock(&queue->lock);
	expires = now = jiffies;
	list_for_each_entry_safe(task, n, &queue->timer_list.list, u.tk_wait.timer_list) {
		timeo = task->u.tk_wait.expires;
		if (time_after_eq(now, timeo)) {
			dprintk("RPC: %5u timeout\n", task->tk_pid);
			task->tk_status = -ETIMEDOUT;
			rpc_wake_up_task_queue_locked(queue, task);
			continue;
		}
		if (expires == now || time_after(expires, timeo))
			expires = timeo;
	}
	if (!list_empty(&queue->timer_list.list))
		rpc_set_queue_timer(queue, expires);
	spin_unlock(&queue->lock);
}

static void __rpc_atrun(struct rpc_task *task)
{
	if (task->tk_status == -ETIMEDOUT)
		task->tk_status = 0;
}

/*
 * Run a task at a later time
 */
void rpc_delay(struct rpc_task *task, unsigned long delay)
{
	task->tk_timeout = delay;
	rpc_sleep_on(&delay_queue, task, __rpc_atrun);
}
EXPORT_SYMBOL_GPL(rpc_delay);

/*
 * Helper to call task->tk_ops->rpc_call_prepare
 */
void rpc_prepare_task(struct rpc_task *task)
{
	task->tk_ops->rpc_call_prepare(task, task->tk_calldata);
}

static void
rpc_init_task_statistics(struct rpc_task *task)
{
	/* Initialize retry counters */
	task->tk_garb_retry = 2;
	task->tk_cred_retry = 2;
	task->tk_rebind_retry = 2;

	/* starting timestamp */
	task->tk_start = ktime_get();
}

static void
rpc_reset_task_statistics(struct rpc_task *task)
{
	task->tk_timeouts = 0;
	task->tk_flags &= ~(RPC_CALL_MAJORSEEN|RPC_TASK_KILLED|RPC_TASK_SENT);

	rpc_init_task_statistics(task);
}

/*
 * Helper that calls task->tk_ops->rpc_call_done if it exists
 */
void rpc_exit_task(struct rpc_task *task)
{
	task->tk_action = NULL;
	if (task->tk_ops->rpc_call_done != NULL) {
		task->tk_ops->rpc_call_done(task, task->tk_calldata);
		if (task->tk_action != NULL) {
			WARN_ON(RPC_ASSASSINATED(task));
			/* Always release the RPC slot and buffer memory */
			xprt_release(task);
			rpc_reset_task_statistics(task);
		}
	}
}

void rpc_exit(struct rpc_task *task, int status)
{
	task->tk_status = status;
	task->tk_action = rpc_exit_task;
	if (RPC_IS_QUEUED(task))
		rpc_wake_up_queued_task(task->tk_waitqueue, task);
}
EXPORT_SYMBOL_GPL(rpc_exit);

void rpc_release_calldata(const struct rpc_call_ops *ops, void *calldata)
{
	if (ops->rpc_release != NULL)
		ops->rpc_release(calldata);
}

/*
 * This is the RPC `scheduler' (or rather, the finite state machine).
 */
static void __rpc_execute(struct rpc_task *task)
{
	struct rpc_wait_queue *queue;
	int task_is_async = RPC_IS_ASYNC(task);
	int status = 0;

	dprintk("RPC: %5u __rpc_execute flags=0x%x\n",
			task->tk_pid, task->tk_flags);

	WARN_ON_ONCE(RPC_IS_QUEUED(task));
	if (RPC_IS_QUEUED(task))
		return;

	for (;;) {
		void (*do_action)(struct rpc_task *);

		/*
		 * Execute any pending callback first.
		 */
		do_action = task->tk_callback;
		task->tk_callback = NULL;
		if (do_action == NULL) {
			/*
			 * Perform the next FSM step.
			 * tk_action may be NULL if the task has been killed.
			 * In particular, note that rpc_killall_tasks may
			 * do this at any time, so beware when dereferencing.
			 */
			do_action = task->tk_action;
			if (do_action == NULL)
				break;
		}
		trace_rpc_task_run_action(task->tk_client, task, task->tk_action);
		do_action(task);

		/*
		 * Lockless check for whether task is sleeping or not.
		 */
		if (!RPC_IS_QUEUED(task))
			continue;
		/*
		 * The queue->lock protects against races with
		 * rpc_make_runnable().
		 *
		 * Note that once we clear RPC_TASK_RUNNING on an asynchronous
		 * rpc_task, rpc_make_runnable() can assign it to a
		 * different workqueue. We therefore cannot assume that the
		 * rpc_task pointer may still be dereferenced.
		 */
		queue = task->tk_waitqueue;
		spin_lock_bh(&queue->lock);
		if (!RPC_IS_QUEUED(task)) {
			spin_unlock_bh(&queue->lock);
			continue;
		}
		rpc_clear_running(task);
		spin_unlock_bh(&queue->lock);
		if (task_is_async)
			return;

		/* sync task: sleep here */
		dprintk("RPC: %5u sync task going to sleep\n", task->tk_pid);
		status = out_of_line_wait_on_bit(&task->tk_runstate,
				RPC_TASK_QUEUED, rpc_wait_bit_killable,
				TASK_KILLABLE);
		if (status == -ERESTARTSYS) {
			/*
			 * When a sync task receives a signal, it exits with
			 * -ERESTARTSYS. In order to catch any callbacks that
			 * clean up after sleeping on some queue, we don't
			 * break the loop here, but go around once more.
			 */
			dprintk("RPC: %5u got signal\n", task->tk_pid);
			task->tk_flags |= RPC_TASK_KILLED;
			rpc_exit(task, -ERESTARTSYS);
		}
		dprintk("RPC: %5u sync task resuming\n", task->tk_pid);
	}

	dprintk("RPC: %5u return %d, status %d\n", task->tk_pid, status,
			task->tk_status);
	/* Release all resources associated with the task */
	rpc_release_task(task);
}

/*
 * User-visible entry point to the scheduler.
 *
 * This may be called recursively if e.g. an async NFS task updates
 * the attributes and finds that dirty pages must be flushed.
 * NOTE: Upon exit of this function the task is guaranteed to be
 *	 released. In particular note that tk_release() will have
 *	 been called, so your task memory may have been freed.
 */
void rpc_execute(struct rpc_task *task)
{
	bool is_async = RPC_IS_ASYNC(task);

	rpc_set_active(task);
	rpc_make_runnable(task);
	if (!is_async)
		__rpc_execute(task);
}

static void rpc_async_schedule(struct work_struct *work)
{
	__rpc_execute(container_of(work, struct rpc_task, u.tk_work));
}

/**
 * rpc_malloc - allocate an RPC buffer
 * @task: RPC task that will use this buffer
 * @size: requested byte size
 *
 * To prevent rpciod from hanging, this allocator never sleeps,
 * returning NULL and suppressing warning if the request cannot be serviced
 * immediately.
 * The caller can arrange to sleep in a way that is safe for rpciod.
 *
 * Most requests are 'small' (under 2KiB) and can be serviced from a
 * mempool, ensuring that NFS reads and writes can always proceed,
 * and that there is good locality of reference for these buffers.
 *
 * In order to avoid memory starvation triggering more writebacks of
 * NFS requests, we avoid using GFP_KERNEL.
 */
void *rpc_malloc(struct rpc_task *task, size_t size)
{
	struct rpc_buffer *buf;
	gfp_t gfp = GFP_NOIO | __GFP_NOWARN;

	if (RPC_IS_SWAPPER(task))
		gfp = __GFP_MEMALLOC | GFP_NOWAIT | __GFP_NOWARN;

	size += sizeof(struct rpc_buffer);
	if (size <= RPC_BUFFER_MAXSIZE)
		buf = mempool_alloc(rpc_buffer_mempool, gfp);
	else
		buf = kmalloc(size, gfp);

	if (!buf)
		return NULL;

	buf->len = size;
	dprintk("RPC: %5u allocated buffer of size %zu at %p\n",
			task->tk_pid, size, buf);
	return &buf->data;
}
EXPORT_SYMBOL_GPL(rpc_malloc);

/**
 * rpc_free - free buffer allocated via rpc_malloc
 * @buffer: buffer to free
 *
 */
void rpc_free(void *buffer)
{
	size_t size;
	struct rpc_buffer *buf;

	if (!buffer)
		return;

	buf = container_of(buffer, struct rpc_buffer, data);
	size = buf->len;

	dprintk("RPC:       freeing buffer of size %zu at %p\n",
			size, buf);

	if (size <= RPC_BUFFER_MAXSIZE)
		mempool_free(buf, rpc_buffer_mempool);
	else
		kfree(buf);
}
EXPORT_SYMBOL_GPL(rpc_free);

/*
 * Creation and deletion of RPC task structures
 */
static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *task_setup_data)
{
	memset(task, 0, sizeof(*task));
	atomic_set(&task->tk_count, 1);
	task->tk_flags  = task_setup_data->flags;
	task->tk_ops = task_setup_data->callback_ops;
	task->tk_calldata = task_setup_data->callback_data;
	INIT_LIST_HEAD(&task->tk_task);

	task->tk_priority = task_setup_data->priority - RPC_PRIORITY_LOW;
	task->tk_owner = current->tgid;

	/* Initialize workqueue for async tasks */
	task->tk_workqueue = task_setup_data->workqueue;

	if (task->tk_ops->rpc_call_prepare != NULL)
		task->tk_action = rpc_prepare_task;

	rpc_init_task_statistics(task);

	dprintk("RPC:       new task initialized, procpid %u\n",
				task_pid_nr(current));
}

static struct rpc_task *
rpc_alloc_task(void)
{
	return (struct rpc_task *)mempool_alloc(rpc_task_mempool, GFP_NOIO);
}

/*
 * Create a new task for the specified client.
 */
struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data)
{
	struct rpc_task	*task = setup_data->task;
	unsigned short flags = 0;

	if (task == NULL) {
		task = rpc_alloc_task();
		if (task == NULL) {
			rpc_release_calldata(setup_data->callback_ops,
					setup_data->callback_data);
			return ERR_PTR(-ENOMEM);
		}
		flags = RPC_TASK_DYNAMIC;
	}

	rpc_init_task(task, setup_data);
	task->tk_flags |= flags;
	dprintk("RPC:       allocated task %p\n", task);
	return task;
}

/*
 * rpc_free_task - release rpc task and perform cleanups
 *
 * Note that we free up the rpc_task _after_ rpc_release_calldata()
 * in order to work around a workqueue dependency issue.
 *
 * Tejun Heo states:
 * "Workqueue currently considers two work items to be the same if they're
 * on the same address and won't execute them concurrently - ie. it
 * makes a work item which is queued again while being executed wait
 * for the previous execution to complete.
 *
 * If a work function frees the work item, and then waits for an event
 * which should be performed by another work item and *that* work item
 * recycles the freed work item, it can create a false dependency loop.
 * There really is no reliable way to detect this short of verifying
 * every memory free."
 *
 */
static void rpc_free_task(struct rpc_task *task)
{
	unsigned short tk_flags = task->tk_flags;

	rpc_release_calldata(task->tk_ops, task->tk_calldata);

	if (tk_flags & RPC_TASK_DYNAMIC) {
		dprintk("RPC: %5u freeing task\n", task->tk_pid);
		mempool_free(task, rpc_task_mempool);
	}
}

static void rpc_async_release(struct work_struct *work)
{
	rpc_free_task(container_of(work, struct rpc_task, u.tk_work));
}

static void rpc_release_resources_task(struct rpc_task *task)
{
	xprt_release(task);
	if (task->tk_msg.rpc_cred) {
		put_rpccred(task->tk_msg.rpc_cred);
		task->tk_msg.rpc_cred = NULL;
	}
	rpc_task_release_client(task);
}

static void rpc_final_put_task(struct rpc_task *task,
		struct workqueue_struct *q)
{
	if (q != NULL) {
		INIT_WORK(&task->u.tk_work, rpc_async_release);
		queue_work(q, &task->u.tk_work);
	} else
		rpc_free_task(task);
}

static void rpc_do_put_task(struct rpc_task *task, struct workqueue_struct *q)
{
	if (atomic_dec_and_test(&task->tk_count)) {
		rpc_release_resources_task(task);
		rpc_final_put_task(task, q);
	}
}

void rpc_put_task(struct rpc_task *task)
{
	rpc_do_put_task(task, NULL);
}
EXPORT_SYMBOL_GPL(rpc_put_task);

void rpc_put_task_async(struct rpc_task *task)
{
	rpc_do_put_task(task, task->tk_workqueue);
}
EXPORT_SYMBOL_GPL(rpc_put_task_async);

static void rpc_release_task(struct rpc_task *task)
{
	dprintk("RPC: %5u release task\n", task->tk_pid);

	WARN_ON_ONCE(RPC_IS_QUEUED(task));

	rpc_release_resources_task(task);

	/*
	 * Note: at this point we have been removed from rpc_clnt->cl_tasks,
	 * so it should be safe to use task->tk_count as a test for whether
	 * or not any other processes still hold references to our rpc_task.
	 */
	if (atomic_read(&task->tk_count) != 1 + !RPC_IS_ASYNC(task)) {
		/* Wake up anyone who may be waiting for task completion */
		if (!rpc_complete_task(task))
			return;
	} else {
		if (!atomic_dec_and_test(&task->tk_count))
			return;
	}
	rpc_final_put_task(task, task->tk_workqueue);
}

int rpciod_up(void)
{
	return try_module_get(THIS_MODULE) ? 0 : -EINVAL;
}

void rpciod_down(void)
{
	module_put(THIS_MODULE);
}

/*
 * Start up the rpciod workqueue.
 */
static int rpciod_start(void)
{
	struct workqueue_struct *wq;

	/*
	 * Create the rpciod thread and wait for it to start.
	 */
	dprintk("RPC:       creating workqueue rpciod\n");
	/* Note: highpri because network receive is latency sensitive */
	wq = alloc_workqueue("rpciod", WQ_MEM_RECLAIM | WQ_HIGHPRI, 0);
	rpciod_workqueue = wq;
	return rpciod_workqueue != NULL;
}

static void rpciod_stop(void)
{
	struct workqueue_struct *wq = NULL;

	if (rpciod_workqueue == NULL)
		return;
	dprintk("RPC:       destroying workqueue rpciod\n");

	wq = rpciod_workqueue;
	rpciod_workqueue = NULL;
	destroy_workqueue(wq);
}

void
rpc_destroy_mempool(void)
{
	rpciod_stop();
	if (rpc_buffer_mempool)
		mempool_destroy(rpc_buffer_mempool);
	if (rpc_task_mempool)
		mempool_destroy(rpc_task_mempool);
	if (rpc_task_slabp)
		kmem_cache_destroy(rpc_task_slabp);
	if (rpc_buffer_slabp)
		kmem_cache_destroy(rpc_buffer_slabp);
	rpc_destroy_wait_queue(&delay_queue);
}

int
rpc_init_mempool(void)
{
	/*
	 * The following is not strictly a mempool initialisation,
	 * but there is no harm in doing it here
	 */
	rpc_init_wait_queue(&delay_queue, "delayq");
	if (!rpciod_start())
		goto err_nomem;

	rpc_task_slabp = kmem_cache_create("rpc_tasks",
					     sizeof(struct rpc_task),
					     0, SLAB_HWCACHE_ALIGN,
					     NULL);
	if (!rpc_task_slabp)
		goto err_nomem;
	rpc_buffer_slabp = kmem_cache_create("rpc_buffers",
					     RPC_BUFFER_MAXSIZE,
					     0, SLAB_HWCACHE_ALIGN,
					     NULL);
	if (!rpc_buffer_slabp)
		goto err_nomem;
	rpc_task_mempool = mempool_create_slab_pool(RPC_TASK_POOLSIZE,
						    rpc_task_slabp);
	if (!rpc_task_mempool)
		goto err_nomem;
	rpc_buffer_mempool = mempool_create_slab_pool(RPC_BUFFER_POOLSIZE,
						      rpc_buffer_slabp);
	if (!rpc_buffer_mempool)
		goto err_nomem;
	return 0;
err_nomem:
	rpc_destroy_mempool();
	return -ENOMEM;
}