summaryrefslogtreecommitdiffstats
path: root/sys/kern/kern_timeout.c
blob: ec540c56824eb95ef93e2045ef8863cfa1356987 (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
/*-
 * Copyright (c) 1982, 1986, 1991, 1993
 *	The Regents of the University of California.  All rights reserved.
 * (c) UNIX System Laboratories, Inc.
 * All or some portions of this file are derived from material licensed
 * to the University of California by American Telephone and Telegraph
 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
 * the permission of UNIX System Laboratories, Inc.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	From: @(#)kern_clock.c	8.5 (Berkeley) 1/21/94
 */

#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <sys/callout.h>
#include <sys/condvar.h>
#include <sys/interrupt.h>
#include <sys/kernel.h>
#include <sys/ktr.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/sleepqueue.h>
#include <sys/sysctl.h>
#include <sys/smp.h>

static int avg_depth;
SYSCTL_INT(_debug, OID_AUTO, to_avg_depth, CTLFLAG_RD, &avg_depth, 0,
    "Average number of items examined per softclock call. Units = 1/1000");
static int avg_gcalls;
SYSCTL_INT(_debug, OID_AUTO, to_avg_gcalls, CTLFLAG_RD, &avg_gcalls, 0,
    "Average number of Giant callouts made per softclock call. Units = 1/1000");
static int avg_lockcalls;
SYSCTL_INT(_debug, OID_AUTO, to_avg_lockcalls, CTLFLAG_RD, &avg_lockcalls, 0,
    "Average number of lock callouts made per softclock call. Units = 1/1000");
static int avg_mpcalls;
SYSCTL_INT(_debug, OID_AUTO, to_avg_mpcalls, CTLFLAG_RD, &avg_mpcalls, 0,
    "Average number of MP callouts made per softclock call. Units = 1/1000");
/*
 * TODO:
 *	allocate more timeout table slots when table overflows.
 */
int callwheelsize, callwheelbits, callwheelmask;

struct callout_cpu {
	struct mtx		cc_lock;
	struct callout		*cc_callout;
	struct callout_tailq	*cc_callwheel;
	struct callout_list	cc_callfree;
	struct callout		*cc_next;
	struct callout		*cc_curr;
	void			*cc_cookie;
	int 			cc_softticks;
	int			cc_cancel;
	int			cc_waiting;
};

#ifdef SMP
struct callout_cpu cc_cpu[MAXCPU];
#define	CC_CPU(cpu)	(&cc_cpu[(cpu)])
#define	CC_SELF()	CC_CPU(PCPU_GET(cpuid))
#else
struct callout_cpu cc_cpu;
#define	CC_CPU(cpu)	&cc_cpu
#define	CC_SELF()	&cc_cpu
#endif
#define	CC_LOCK(cc)	mtx_lock_spin(&(cc)->cc_lock)
#define	CC_UNLOCK(cc)	mtx_unlock_spin(&(cc)->cc_lock)

static int timeout_cpu;

MALLOC_DEFINE(M_CALLOUT, "callout", "Callout datastructures");

/**
 * Locked by cc_lock:
 *   cc_curr         - If a callout is in progress, it is curr_callout.
 *                     If curr_callout is non-NULL, threads waiting in
 *                     callout_drain() will be woken up as soon as the
 *                     relevant callout completes.
 *   cc_cancel       - Changing to 1 with both callout_lock and c_lock held
 *                     guarantees that the current callout will not run.
 *                     The softclock() function sets this to 0 before it
 *                     drops callout_lock to acquire c_lock, and it calls
 *                     the handler only if curr_cancelled is still 0 after
 *                     c_lock is successfully acquired.
 *   cc_waiting      - If a thread is waiting in callout_drain(), then
 *                     callout_wait is nonzero.  Set only when
 *                     curr_callout is non-NULL.
 */

/*
 * kern_timeout_callwheel_alloc() - kernel low level callwheel initialization 
 *
 *	This code is called very early in the kernel initialization sequence,
 *	and may be called more then once.
 */
caddr_t
kern_timeout_callwheel_alloc(caddr_t v)
{
	struct callout_cpu *cc;

	timeout_cpu = PCPU_GET(cpuid);
	cc = CC_CPU(timeout_cpu);
	/*
	 * Calculate callout wheel size
	 */
	for (callwheelsize = 1, callwheelbits = 0;
	     callwheelsize < ncallout;
	     callwheelsize <<= 1, ++callwheelbits)
		;
	callwheelmask = callwheelsize - 1;

	cc->cc_callout = (struct callout *)v;
	v = (caddr_t)(cc->cc_callout + ncallout);
	cc->cc_callwheel = (struct callout_tailq *)v;
	v = (caddr_t)(cc->cc_callwheel + callwheelsize);
	return(v);
}

static void
callout_cpu_init(struct callout_cpu *cc)
{
	struct callout *c;
	int i;

	mtx_init(&cc->cc_lock, "callout", NULL, MTX_SPIN | MTX_RECURSE);
	SLIST_INIT(&cc->cc_callfree);
	for (i = 0; i < callwheelsize; i++) {
		TAILQ_INIT(&cc->cc_callwheel[i]);
	}
	if (cc->cc_callout == NULL)
		return;
	for (i = 0; i < ncallout; i++) {
		c = &cc->cc_callout[i];
		callout_init(c, 0);
		c->c_flags = CALLOUT_LOCAL_ALLOC;
		SLIST_INSERT_HEAD(&cc->cc_callfree, c, c_links.sle);
	}
}

/*
 * kern_timeout_callwheel_init() - initialize previously reserved callwheel
 *				   space.
 *
 *	This code is called just once, after the space reserved for the
 *	callout wheel has been finalized.
 */
void
kern_timeout_callwheel_init(void)
{
	callout_cpu_init(CC_CPU(timeout_cpu));
}

/*
 * Start standard softclock thread.
 */
void    *softclock_ih;

static void
start_softclock(void *dummy)
{
	struct callout_cpu *cc;
#ifdef SMP
	int cpu;
#endif

	cc = CC_CPU(timeout_cpu);
	if (swi_add(&clk_intr_event, "clock", softclock, cc, SWI_CLOCK,
	    INTR_MPSAFE, &softclock_ih))
		panic("died while creating standard software ithreads");
	cc->cc_cookie = softclock_ih;
#ifdef SMP
	for (cpu = 0; cpu <= mp_maxid; cpu++) {
		if (cpu == timeout_cpu)
			continue;
		if (CPU_ABSENT(cpu))
			continue;
		cc = CC_CPU(cpu);
		if (swi_add(NULL, "clock", softclock, cc, SWI_CLOCK,
		    INTR_MPSAFE, &cc->cc_cookie))
			panic("died while creating standard software ithreads");
		cc->cc_callout = NULL;	/* Only cpu0 handles timeout(). */
		cc->cc_callwheel = malloc(
		    sizeof(struct callout_tailq) * callwheelsize, M_CALLOUT,
		    M_WAITOK);
		callout_cpu_init(cc);
	}
#endif
}

SYSINIT(start_softclock, SI_SUB_SOFTINTR, SI_ORDER_FIRST, start_softclock, NULL);

void
callout_tick(void)
{
	struct callout_cpu *cc;
	int need_softclock;
	int bucket;

	/*
	 * Process callouts at a very low cpu priority, so we don't keep the
	 * relatively high clock interrupt priority any longer than necessary.
	 */
	need_softclock = 0;
	cc = CC_SELF();
	mtx_lock_spin_flags(&cc->cc_lock, MTX_QUIET);
	for (; cc->cc_softticks < ticks; cc->cc_softticks++) {
		bucket = cc->cc_softticks & callwheelmask;
		if (!TAILQ_EMPTY(&cc->cc_callwheel[bucket])) {
			need_softclock = 1;
			break;
		}
	}
	mtx_unlock_spin_flags(&cc->cc_lock, MTX_QUIET);
	/*
	 * swi_sched acquires the thread lock, so we don't want to call it
	 * with cc_lock held; incorrect locking order.
	 */
	if (need_softclock)
		swi_sched(cc->cc_cookie, 0);
}

static struct callout_cpu *
callout_lock(struct callout *c)
{
	struct callout_cpu *cc;
	int cpu;

	for (;;) {
		cpu = c->c_cpu;
		cc = CC_CPU(cpu);
		CC_LOCK(cc);
		if (cpu == c->c_cpu)
			break;
		CC_UNLOCK(cc);
	}
	return (cc);
}

/*
 * The callout mechanism is based on the work of Adam M. Costello and 
 * George Varghese, published in a technical report entitled "Redesigning
 * the BSD Callout and Timer Facilities" and modified slightly for inclusion
 * in FreeBSD by Justin T. Gibbs.  The original work on the data structures
 * used in this implementation was published by G. Varghese and T. Lauck in
 * the paper "Hashed and Hierarchical Timing Wheels: Data Structures for
 * the Efficient Implementation of a Timer Facility" in the Proceedings of
 * the 11th ACM Annual Symposium on Operating Systems Principles,
 * Austin, Texas Nov 1987.
 */

/*
 * Software (low priority) clock interrupt.
 * Run periodic events from timeout queue.
 */
void
softclock(void *arg)
{
	struct callout_cpu *cc;
	struct callout *c;
	struct callout_tailq *bucket;
	int curticks;
	int steps;	/* #steps since we last allowed interrupts */
	int depth;
	int mpcalls;
	int lockcalls;
	int gcalls;
#ifdef DIAGNOSTIC
	struct bintime bt1, bt2;
	struct timespec ts2;
	static uint64_t maxdt = 36893488147419102LL;	/* 2 msec */
	static timeout_t *lastfunc;
#endif

#ifndef MAX_SOFTCLOCK_STEPS
#define MAX_SOFTCLOCK_STEPS 100 /* Maximum allowed value of steps. */
#endif /* MAX_SOFTCLOCK_STEPS */

	mpcalls = 0;
	lockcalls = 0;
	gcalls = 0;
	depth = 0;
	steps = 0;
	cc = (struct callout_cpu *)arg;
	CC_LOCK(cc);
	while (cc->cc_softticks != ticks) {
		/*
		 * cc_softticks may be modified by hard clock, so cache
		 * it while we work on a given bucket.
		 */
		curticks = cc->cc_softticks;
		cc->cc_softticks++;
		bucket = &cc->cc_callwheel[curticks & callwheelmask];
		c = TAILQ_FIRST(bucket);
		while (c) {
			depth++;
			if (c->c_time != curticks) {
				c = TAILQ_NEXT(c, c_links.tqe);
				++steps;
				if (steps >= MAX_SOFTCLOCK_STEPS) {
					cc->cc_next = c;
					/* Give interrupts a chance. */
					CC_UNLOCK(cc);
					;	/* nothing */
					CC_LOCK(cc);
					c = cc->cc_next;
					steps = 0;
				}
			} else {
				void (*c_func)(void *);
				void *c_arg;
				struct lock_class *class;
				struct lock_object *c_lock;
				int c_flags, sharedlock;

				cc->cc_next = TAILQ_NEXT(c, c_links.tqe);
				TAILQ_REMOVE(bucket, c, c_links.tqe);
				class = (c->c_lock != NULL) ?
				    LOCK_CLASS(c->c_lock) : NULL;
				sharedlock = (c->c_flags & CALLOUT_SHAREDLOCK) ?
				    0 : 1;
				c_lock = c->c_lock;
				c_func = c->c_func;
				c_arg = c->c_arg;
				c_flags = c->c_flags;
				if (c->c_flags & CALLOUT_LOCAL_ALLOC) {
					c->c_flags = CALLOUT_LOCAL_ALLOC;
				} else {
					c->c_flags =
					    (c->c_flags & ~CALLOUT_PENDING);
				}
				cc->cc_curr = c;
				cc->cc_cancel = 0;
				CC_UNLOCK(cc);
				if (c_lock != NULL) {
					class->lc_lock(c_lock, sharedlock);
					/*
					 * The callout may have been cancelled
					 * while we switched locks.
					 */
					if (cc->cc_cancel) {
						class->lc_unlock(c_lock);
						goto skip;
					}
					/* The callout cannot be stopped now. */
					cc->cc_cancel = 1;

					if (c_lock == &Giant.lock_object) {
						gcalls++;
						CTR3(KTR_CALLOUT,
						    "callout %p func %p arg %p",
						    c, c_func, c_arg);
					} else {
						lockcalls++;
						CTR3(KTR_CALLOUT, "callout lock"
						    " %p func %p arg %p",
						    c, c_func, c_arg);
					}
				} else {
					mpcalls++;
					CTR3(KTR_CALLOUT,
					    "callout mpsafe %p func %p arg %p",
					    c, c_func, c_arg);
				}
#ifdef DIAGNOSTIC
				binuptime(&bt1);
#endif
				THREAD_NO_SLEEPING();
				c_func(c_arg);
				THREAD_SLEEPING_OK();
#ifdef DIAGNOSTIC
				binuptime(&bt2);
				bintime_sub(&bt2, &bt1);
				if (bt2.frac > maxdt) {
					if (lastfunc != c_func ||
					    bt2.frac > maxdt * 2) {
						bintime2timespec(&bt2, &ts2);
						printf(
			"Expensive timeout(9) function: %p(%p) %jd.%09ld s\n",
						    c_func, c_arg,
						    (intmax_t)ts2.tv_sec,
						    ts2.tv_nsec);
					}
					maxdt = bt2.frac;
					lastfunc = c_func;
				}
#endif
				if ((c_flags & CALLOUT_RETURNUNLOCKED) == 0)
					class->lc_unlock(c_lock);
			skip:
				CC_LOCK(cc);
				/*
				 * If the current callout is locally
				 * allocated (from timeout(9))
				 * then put it on the freelist.
				 *
				 * Note: we need to check the cached
				 * copy of c_flags because if it was not
				 * local, then it's not safe to deref the
				 * callout pointer.
				 */
				if (c_flags & CALLOUT_LOCAL_ALLOC) {
					KASSERT(c->c_flags ==
					    CALLOUT_LOCAL_ALLOC,
					    ("corrupted callout"));
					c->c_func = NULL;
					SLIST_INSERT_HEAD(&cc->cc_callfree, c,
					    c_links.sle);
				}
				cc->cc_curr = NULL;
				if (cc->cc_waiting) {
					/*
					 * There is someone waiting
					 * for the callout to complete.
					 */
					cc->cc_waiting = 0;
					CC_UNLOCK(cc);
					wakeup(&cc->cc_waiting);
					CC_LOCK(cc);
				}
				steps = 0;
				c = cc->cc_next;
			}
		}
	}
	avg_depth += (depth * 1000 - avg_depth) >> 8;
	avg_mpcalls += (mpcalls * 1000 - avg_mpcalls) >> 8;
	avg_lockcalls += (lockcalls * 1000 - avg_lockcalls) >> 8;
	avg_gcalls += (gcalls * 1000 - avg_gcalls) >> 8;
	cc->cc_next = NULL;
	CC_UNLOCK(cc);
}

/*
 * timeout --
 *	Execute a function after a specified length of time.
 *
 * untimeout --
 *	Cancel previous timeout function call.
 *
 * callout_handle_init --
 *	Initialize a handle so that using it with untimeout is benign.
 *
 *	See AT&T BCI Driver Reference Manual for specification.  This
 *	implementation differs from that one in that although an 
 *	identification value is returned from timeout, the original
 *	arguments to timeout as well as the identifier are used to
 *	identify entries for untimeout.
 */
struct callout_handle
timeout(ftn, arg, to_ticks)
	timeout_t *ftn;
	void *arg;
	int to_ticks;
{
	struct callout_cpu *cc;
	struct callout *new;
	struct callout_handle handle;

	cc = CC_CPU(timeout_cpu);
	CC_LOCK(cc);
	/* Fill in the next free callout structure. */
	new = SLIST_FIRST(&cc->cc_callfree);
	if (new == NULL)
		/* XXX Attempt to malloc first */
		panic("timeout table full");
	SLIST_REMOVE_HEAD(&cc->cc_callfree, c_links.sle);
	callout_reset(new, to_ticks, ftn, arg);
	handle.callout = new;
	CC_UNLOCK(cc);

	return (handle);
}

void
untimeout(ftn, arg, handle)
	timeout_t *ftn;
	void *arg;
	struct callout_handle handle;
{
	struct callout_cpu *cc;

	/*
	 * Check for a handle that was initialized
	 * by callout_handle_init, but never used
	 * for a real timeout.
	 */
	if (handle.callout == NULL)
		return;

	cc = callout_lock(handle.callout);
	if (handle.callout->c_func == ftn && handle.callout->c_arg == arg)
		callout_stop(handle.callout);
	CC_UNLOCK(cc);
}

void
callout_handle_init(struct callout_handle *handle)
{
	handle->callout = NULL;
}

/*
 * New interface; clients allocate their own callout structures.
 *
 * callout_reset() - establish or change a timeout
 * callout_stop() - disestablish a timeout
 * callout_init() - initialize a callout structure so that it can
 *	safely be passed to callout_reset() and callout_stop()
 *
 * <sys/callout.h> defines three convenience macros:
 *
 * callout_active() - returns truth if callout has not been stopped,
 *	drained, or deactivated since the last time the callout was
 *	reset.
 * callout_pending() - returns truth if callout is still waiting for timeout
 * callout_deactivate() - marks the callout as having been serviced
 */
int
callout_reset_on(struct callout *c, int to_ticks, void (*ftn)(void *),
    void *arg, int cpu)
{
	struct callout_cpu *cc;
	int cancelled = 0;

	/*
	 * Don't allow migration of pre-allocated callouts lest they
	 * become unbalanced.
	 */
	if (c->c_flags & CALLOUT_LOCAL_ALLOC)
		cpu = c->c_cpu;
retry:
	cc = callout_lock(c);
	if (cc->cc_curr == c) {
		/*
		 * We're being asked to reschedule a callout which is
		 * currently in progress.  If there is a lock then we
		 * can cancel the callout if it has not really started.
		 */
		if (c->c_lock != NULL && !cc->cc_cancel)
			cancelled = cc->cc_cancel = 1;
		if (cc->cc_waiting) {
			/*
			 * Someone has called callout_drain to kill this
			 * callout.  Don't reschedule.
			 */
			CTR4(KTR_CALLOUT, "%s %p func %p arg %p",
			    cancelled ? "cancelled" : "failed to cancel",
			    c, c->c_func, c->c_arg);
			CC_UNLOCK(cc);
			return (cancelled);
		}
	}
	if (c->c_flags & CALLOUT_PENDING) {
		if (cc->cc_next == c) {
			cc->cc_next = TAILQ_NEXT(c, c_links.tqe);
		}
		TAILQ_REMOVE(&cc->cc_callwheel[c->c_time & callwheelmask], c,
		    c_links.tqe);

		cancelled = 1;
		c->c_flags &= ~(CALLOUT_ACTIVE | CALLOUT_PENDING);
	}
	/*
	 * If the lock must migrate we have to check the state again as
	 * we can't hold both the new and old locks simultaneously.
	 */
	if (c->c_cpu != cpu) {
		c->c_cpu = cpu;
		CC_UNLOCK(cc);
		goto retry;
	}

	if (to_ticks <= 0)
		to_ticks = 1;

	c->c_arg = arg;
	c->c_flags |= (CALLOUT_ACTIVE | CALLOUT_PENDING);
	c->c_func = ftn;
	c->c_time = ticks + to_ticks;
	TAILQ_INSERT_TAIL(&cc->cc_callwheel[c->c_time & callwheelmask], 
			  c, c_links.tqe);
	CTR5(KTR_CALLOUT, "%sscheduled %p func %p arg %p in %d",
	    cancelled ? "re" : "", c, c->c_func, c->c_arg, to_ticks);
	CC_UNLOCK(cc);

	return (cancelled);
}

int
_callout_stop_safe(c, safe)
	struct	callout *c;
	int	safe;
{
	struct callout_cpu *cc;
	struct lock_class *class;
	int use_lock, sq_locked;

	/*
	 * Some old subsystems don't hold Giant while running a callout_stop(),
	 * so just discard this check for the moment.
	 */
	if (!safe && c->c_lock != NULL) {
		if (c->c_lock == &Giant.lock_object)
			use_lock = mtx_owned(&Giant);
		else {
			use_lock = 1;
			class = LOCK_CLASS(c->c_lock);
			class->lc_assert(c->c_lock, LA_XLOCKED);
		}
	} else
		use_lock = 0;

	sq_locked = 0;
again:
	cc = callout_lock(c);
	/*
	 * If the callout isn't pending, it's not on the queue, so
	 * don't attempt to remove it from the queue.  We can try to
	 * stop it by other means however.
	 */
	if (!(c->c_flags & CALLOUT_PENDING)) {
		c->c_flags &= ~CALLOUT_ACTIVE;

		/*
		 * If it wasn't on the queue and it isn't the current
		 * callout, then we can't stop it, so just bail.
		 */
		if (cc->cc_curr != c) {
			CTR3(KTR_CALLOUT, "failed to stop %p func %p arg %p",
			    c, c->c_func, c->c_arg);
			CC_UNLOCK(cc);
			if (sq_locked)
				sleepq_release(&cc->cc_waiting);
			return (0);
		}

		if (safe) {
			/*
			 * The current callout is running (or just
			 * about to run) and blocking is allowed, so
			 * just wait for the current invocation to
			 * finish.
			 */
			while (cc->cc_curr == c) {

				/*
				 * Use direct calls to sleepqueue interface
				 * instead of cv/msleep in order to avoid
				 * a LOR between cc_lock and sleepqueue
				 * chain spinlocks.  This piece of code
				 * emulates a msleep_spin() call actually.
				 *
				 * If we already have the sleepqueue chain
				 * locked, then we can safely block.  If we
				 * don't already have it locked, however,
				 * we have to drop the cc_lock to lock
				 * it.  This opens several races, so we
				 * restart at the beginning once we have
				 * both locks.  If nothing has changed, then
				 * we will end up back here with sq_locked
				 * set.
				 */
				if (!sq_locked) {
					CC_UNLOCK(cc);
					sleepq_lock(&cc->cc_waiting);
					sq_locked = 1;
					goto again;
				}
				cc->cc_waiting = 1;
				DROP_GIANT();
				CC_UNLOCK(cc);
				sleepq_add(&cc->cc_waiting,
				    &cc->cc_lock.lock_object, "codrain",
				    SLEEPQ_SLEEP, 0);
				sleepq_wait(&cc->cc_waiting, 0);
				sq_locked = 0;

				/* Reacquire locks previously released. */
				PICKUP_GIANT();
				CC_LOCK(cc);
			}
		} else if (use_lock && !cc->cc_cancel) {
			/*
			 * The current callout is waiting for its
			 * lock which we hold.  Cancel the callout
			 * and return.  After our caller drops the
			 * lock, the callout will be skipped in
			 * softclock().
			 */
			cc->cc_cancel = 1;
			CTR3(KTR_CALLOUT, "cancelled %p func %p arg %p",
			    c, c->c_func, c->c_arg);
			CC_UNLOCK(cc);
			KASSERT(!sq_locked, ("sleepqueue chain locked"));
			return (1);
		}
		CTR3(KTR_CALLOUT, "failed to stop %p func %p arg %p",
		    c, c->c_func, c->c_arg);
		CC_UNLOCK(cc);
		KASSERT(!sq_locked, ("sleepqueue chain still locked"));
		return (0);
	}
	if (sq_locked)
		sleepq_release(&cc->cc_waiting);

	c->c_flags &= ~(CALLOUT_ACTIVE | CALLOUT_PENDING);

	if (cc->cc_next == c) {
		cc->cc_next = TAILQ_NEXT(c, c_links.tqe);
	}
	TAILQ_REMOVE(&cc->cc_callwheel[c->c_time & callwheelmask], c,
	    c_links.tqe);

	CTR3(KTR_CALLOUT, "cancelled %p func %p arg %p",
	    c, c->c_func, c->c_arg);

	if (c->c_flags & CALLOUT_LOCAL_ALLOC) {
		c->c_func = NULL;
		SLIST_INSERT_HEAD(&cc->cc_callfree, c, c_links.sle);
	}
	CC_UNLOCK(cc);
	return (1);
}

void
callout_init(c, mpsafe)
	struct	callout *c;
	int mpsafe;
{
	bzero(c, sizeof *c);
	if (mpsafe) {
		c->c_lock = NULL;
		c->c_flags = CALLOUT_RETURNUNLOCKED;
	} else {
		c->c_lock = &Giant.lock_object;
		c->c_flags = 0;
	}
	c->c_cpu = timeout_cpu;
}

void
_callout_init_lock(c, lock, flags)
	struct	callout *c;
	struct	lock_object *lock;
	int flags;
{
	bzero(c, sizeof *c);
	c->c_lock = lock;
	KASSERT((flags & ~(CALLOUT_RETURNUNLOCKED | CALLOUT_SHAREDLOCK)) == 0,
	    ("callout_init_lock: bad flags %d", flags));
	KASSERT(lock != NULL || (flags & CALLOUT_RETURNUNLOCKED) == 0,
	    ("callout_init_lock: CALLOUT_RETURNUNLOCKED with no lock"));
	KASSERT(lock == NULL || !(LOCK_CLASS(lock)->lc_flags &
	    (LC_SPINLOCK | LC_SLEEPABLE)), ("%s: invalid lock class",
	    __func__));
	c->c_flags = flags & (CALLOUT_RETURNUNLOCKED | CALLOUT_SHAREDLOCK);
	c->c_cpu = timeout_cpu;
}

#ifdef APM_FIXUP_CALLTODO
/* 
 * Adjust the kernel calltodo timeout list.  This routine is used after 
 * an APM resume to recalculate the calltodo timer list values with the 
 * number of hz's we have been sleeping.  The next hardclock() will detect 
 * that there are fired timers and run softclock() to execute them.
 *
 * Please note, I have not done an exhaustive analysis of what code this
 * might break.  I am motivated to have my select()'s and alarm()'s that
 * have expired during suspend firing upon resume so that the applications
 * which set the timer can do the maintanence the timer was for as close
 * as possible to the originally intended time.  Testing this code for a 
 * week showed that resuming from a suspend resulted in 22 to 25 timers 
 * firing, which seemed independant on whether the suspend was 2 hours or
 * 2 days.  Your milage may vary.   - Ken Key <key@cs.utk.edu>
 */
void
adjust_timeout_calltodo(time_change)
    struct timeval *time_change;
{
	register struct callout *p;
	unsigned long delta_ticks;

	/* 
	 * How many ticks were we asleep?
	 * (stolen from tvtohz()).
	 */

	/* Don't do anything */
	if (time_change->tv_sec < 0)
		return;
	else if (time_change->tv_sec <= LONG_MAX / 1000000)
		delta_ticks = (time_change->tv_sec * 1000000 +
			       time_change->tv_usec + (tick - 1)) / tick + 1;
	else if (time_change->tv_sec <= LONG_MAX / hz)
		delta_ticks = time_change->tv_sec * hz +
			      (time_change->tv_usec + (tick - 1)) / tick + 1;
	else
		delta_ticks = LONG_MAX;

	if (delta_ticks > INT_MAX)
		delta_ticks = INT_MAX;

	/* 
	 * Now rip through the timer calltodo list looking for timers
	 * to expire.
	 */

	/* don't collide with softclock() */
	CC_LOCK(cc);
	for (p = calltodo.c_next; p != NULL; p = p->c_next) {
		p->c_time -= delta_ticks;

		/* Break if the timer had more time on it than delta_ticks */
		if (p->c_time > 0)
			break;

		/* take back the ticks the timer didn't use (p->c_time <= 0) */
		delta_ticks = -p->c_time;
	}
	CC_UNLOCK(cc);

	return;
}
#endif /* APM_FIXUP_CALLTODO */
OpenPOWER on IntegriCloud