timer.c 47.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
/*
 *  linux/kernel/timer.c
 *
4
 *  Kernel internal timers
L
Linus Torvalds 已提交
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  1997-01-28  Modified by Finn Arne Gangstad to make timers scale better.
 *
 *  1997-09-10  Updated NTP code according to technical memorandum Jan '96
 *              "A Kernel Model for Precision Timekeeping" by Dave Mills
 *  1998-12-24  Fixed a xtime SMP race (we need the xtime_lock rw spinlock to
 *              serialize accesses to xtime/lost_ticks).
 *                              Copyright (C) 1998  Andrea Arcangeli
 *  1999-03-10  Improved NTP compatibility by Ulrich Windl
 *  2002-05-31	Move sys_sysinfo here and make its locking sane, Robert Love
 *  2000-10-05  Implemented scalable SMP per-CPU timer handling.
 *                              Copyright (C) 2000, 2001, 2002  Ingo Molnar
 *              Designed by David S. Miller, Alexey Kuznetsov and Ingo Molnar
 */

#include <linux/kernel_stat.h>
23
#include <linux/export.h>
L
Linus Torvalds 已提交
24 25 26 27 28
#include <linux/interrupt.h>
#include <linux/percpu.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/swap.h>
29
#include <linux/pid_namespace.h>
L
Linus Torvalds 已提交
30 31 32 33 34 35 36
#include <linux/notifier.h>
#include <linux/thread_info.h>
#include <linux/time.h>
#include <linux/jiffies.h>
#include <linux/posix-timers.h>
#include <linux/cpu.h>
#include <linux/syscalls.h>
A
Adrian Bunk 已提交
37
#include <linux/delay.h>
38
#include <linux/tick.h>
39
#include <linux/kallsyms.h>
40
#include <linux/irq_work.h>
41
#include <linux/sched.h>
42
#include <linux/sched/sysctl.h>
43
#include <linux/slab.h>
44
#include <linux/compat.h>
L
Linus Torvalds 已提交
45 46 47 48 49 50 51

#include <asm/uaccess.h>
#include <asm/unistd.h>
#include <asm/div64.h>
#include <asm/timex.h>
#include <asm/io.h>

52 53 54
#define CREATE_TRACE_POINTS
#include <trace/events/timer.h>

A
Andi Kleen 已提交
55
__visible u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES;
T
Thomas Gleixner 已提交
56 57 58

EXPORT_SYMBOL(jiffies_64);

L
Linus Torvalds 已提交
59 60 61 62 63 64 65 66 67
/*
 * per-CPU timer vector definitions:
 */
#define TVN_BITS (CONFIG_BASE_SMALL ? 4 : 6)
#define TVR_BITS (CONFIG_BASE_SMALL ? 6 : 8)
#define TVN_SIZE (1 << TVN_BITS)
#define TVR_SIZE (1 << TVR_BITS)
#define TVN_MASK (TVN_SIZE - 1)
#define TVR_MASK (TVR_SIZE - 1)
68
#define MAX_TVAL ((unsigned long)((1ULL << (TVR_BITS + 4*TVN_BITS)) - 1))
L
Linus Torvalds 已提交
69

70
struct tvec {
L
Linus Torvalds 已提交
71
	struct list_head vec[TVN_SIZE];
72
};
L
Linus Torvalds 已提交
73

74
struct tvec_root {
L
Linus Torvalds 已提交
75
	struct list_head vec[TVR_SIZE];
76
};
L
Linus Torvalds 已提交
77

78
struct tvec_base {
79 80
	spinlock_t lock;
	struct timer_list *running_timer;
L
Linus Torvalds 已提交
81
	unsigned long timer_jiffies;
82
	unsigned long next_timer;
83
	unsigned long active_timers;
84
	unsigned long all_timers;
85
	int cpu;
86 87 88 89 90
	struct tvec_root tv1;
	struct tvec tv2;
	struct tvec tv3;
	struct tvec tv4;
	struct tvec tv5;
91
} ____cacheline_aligned;
L
Linus Torvalds 已提交
92

93
struct tvec_base boot_tvec_bases;
94
EXPORT_SYMBOL(boot_tvec_bases);
95
static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases;
L
Linus Torvalds 已提交
96

97
/* Functions below help us manage 'deferrable' flag */
98
static inline unsigned int tbase_get_deferrable(struct tvec_base *base)
99
{
100
	return ((unsigned int)(unsigned long)base & TIMER_DEFERRABLE);
101 102
}

T
Tejun Heo 已提交
103 104 105 106 107
static inline unsigned int tbase_get_irqsafe(struct tvec_base *base)
{
	return ((unsigned int)(unsigned long)base & TIMER_IRQSAFE);
}

108
static inline struct tvec_base *tbase_get_base(struct tvec_base *base)
109
{
110
	return ((struct tvec_base *)((unsigned long)base & ~TIMER_FLAG_MASK));
111 112 113
}

static inline void
114
timer_set_base(struct timer_list *timer, struct tvec_base *new_base)
115
{
116 117 118
	unsigned long flags = (unsigned long)timer->base & TIMER_FLAG_MASK;

	timer->base = (struct tvec_base *)((unsigned long)(new_base) | flags);
119 120
}

121 122
static unsigned long round_jiffies_common(unsigned long j, int cpu,
		bool force_up)
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143
{
	int rem;
	unsigned long original = j;

	/*
	 * We don't want all cpus firing their timers at once hitting the
	 * same lock or cachelines, so we skew each extra cpu with an extra
	 * 3 jiffies. This 3 jiffies came originally from the mm/ code which
	 * already did this.
	 * The skew is done by adding 3*cpunr, then round, then subtract this
	 * extra offset again.
	 */
	j += cpu * 3;

	rem = j % HZ;

	/*
	 * If the target jiffie is just after a whole second (which can happen
	 * due to delays of the timer irq, long irq off times etc etc) then
	 * we should round down to the whole second, not up. Use 1/4th second
	 * as cutoff for this rounding as an extreme upper bound for this.
144
	 * But never round down if @force_up is set.
145
	 */
146
	if (rem < HZ/4 && !force_up) /* round down */
147 148 149 150 151 152 153
		j = j - rem;
	else /* round up */
		j = j - rem + HZ;

	/* now that we have rounded, subtract the extra skew again */
	j -= cpu * 3;

154 155 156 157 158
	/*
	 * Make sure j is still in the future. Otherwise return the
	 * unmodified value.
	 */
	return time_is_after_jiffies(j) ? j : original;
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

/**
 * __round_jiffies - function to round jiffies to a full second
 * @j: the time in (absolute) jiffies that should be rounded
 * @cpu: the processor number on which the timeout will happen
 *
 * __round_jiffies() rounds an absolute time in the future (in jiffies)
 * up or down to (approximately) full seconds. This is useful for timers
 * for which the exact time they fire does not matter too much, as long as
 * they fire approximately every X seconds.
 *
 * By rounding these timers to whole seconds, all such timers will fire
 * at the same time, rather than at various times spread out. The goal
 * of this is to have the CPU wake up less, which saves power.
 *
 * The exact rounding is skewed for each processor to avoid all
 * processors firing at the exact same time, which could lead
 * to lock contention or spurious cache line bouncing.
 *
 * The return value is the rounded version of the @j parameter.
 */
unsigned long __round_jiffies(unsigned long j, int cpu)
{
	return round_jiffies_common(j, cpu, false);
}
185 186 187 188 189 190 191
EXPORT_SYMBOL_GPL(__round_jiffies);

/**
 * __round_jiffies_relative - function to round jiffies to a full second
 * @j: the time in (relative) jiffies that should be rounded
 * @cpu: the processor number on which the timeout will happen
 *
192
 * __round_jiffies_relative() rounds a time delta  in the future (in jiffies)
193 194 195 196 197 198 199 200 201 202 203 204
 * up or down to (approximately) full seconds. This is useful for timers
 * for which the exact time they fire does not matter too much, as long as
 * they fire approximately every X seconds.
 *
 * By rounding these timers to whole seconds, all such timers will fire
 * at the same time, rather than at various times spread out. The goal
 * of this is to have the CPU wake up less, which saves power.
 *
 * The exact rounding is skewed for each processor to avoid all
 * processors firing at the exact same time, which could lead
 * to lock contention or spurious cache line bouncing.
 *
205
 * The return value is the rounded version of the @j parameter.
206 207 208
 */
unsigned long __round_jiffies_relative(unsigned long j, int cpu)
{
209 210 211 212
	unsigned long j0 = jiffies;

	/* Use j0 because jiffies might change while we run */
	return round_jiffies_common(j + j0, cpu, false) - j0;
213 214 215 216 217 218 219
}
EXPORT_SYMBOL_GPL(__round_jiffies_relative);

/**
 * round_jiffies - function to round jiffies to a full second
 * @j: the time in (absolute) jiffies that should be rounded
 *
220
 * round_jiffies() rounds an absolute time in the future (in jiffies)
221 222 223 224 225 226 227 228
 * up or down to (approximately) full seconds. This is useful for timers
 * for which the exact time they fire does not matter too much, as long as
 * they fire approximately every X seconds.
 *
 * By rounding these timers to whole seconds, all such timers will fire
 * at the same time, rather than at various times spread out. The goal
 * of this is to have the CPU wake up less, which saves power.
 *
229
 * The return value is the rounded version of the @j parameter.
230 231 232
 */
unsigned long round_jiffies(unsigned long j)
{
233
	return round_jiffies_common(j, raw_smp_processor_id(), false);
234 235 236 237 238 239 240
}
EXPORT_SYMBOL_GPL(round_jiffies);

/**
 * round_jiffies_relative - function to round jiffies to a full second
 * @j: the time in (relative) jiffies that should be rounded
 *
241
 * round_jiffies_relative() rounds a time delta  in the future (in jiffies)
242 243 244 245 246 247 248 249
 * up or down to (approximately) full seconds. This is useful for timers
 * for which the exact time they fire does not matter too much, as long as
 * they fire approximately every X seconds.
 *
 * By rounding these timers to whole seconds, all such timers will fire
 * at the same time, rather than at various times spread out. The goal
 * of this is to have the CPU wake up less, which saves power.
 *
250
 * The return value is the rounded version of the @j parameter.
251 252 253 254 255 256 257
 */
unsigned long round_jiffies_relative(unsigned long j)
{
	return __round_jiffies_relative(j, raw_smp_processor_id());
}
EXPORT_SYMBOL_GPL(round_jiffies_relative);

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
/**
 * __round_jiffies_up - function to round jiffies up to a full second
 * @j: the time in (absolute) jiffies that should be rounded
 * @cpu: the processor number on which the timeout will happen
 *
 * This is the same as __round_jiffies() except that it will never
 * round down.  This is useful for timeouts for which the exact time
 * of firing does not matter too much, as long as they don't fire too
 * early.
 */
unsigned long __round_jiffies_up(unsigned long j, int cpu)
{
	return round_jiffies_common(j, cpu, true);
}
EXPORT_SYMBOL_GPL(__round_jiffies_up);

/**
 * __round_jiffies_up_relative - function to round jiffies up to a full second
 * @j: the time in (relative) jiffies that should be rounded
 * @cpu: the processor number on which the timeout will happen
 *
 * This is the same as __round_jiffies_relative() except that it will never
 * round down.  This is useful for timeouts for which the exact time
 * of firing does not matter too much, as long as they don't fire too
 * early.
 */
unsigned long __round_jiffies_up_relative(unsigned long j, int cpu)
{
	unsigned long j0 = jiffies;

	/* Use j0 because jiffies might change while we run */
	return round_jiffies_common(j + j0, cpu, true) - j0;
}
EXPORT_SYMBOL_GPL(__round_jiffies_up_relative);

/**
 * round_jiffies_up - function to round jiffies up to a full second
 * @j: the time in (absolute) jiffies that should be rounded
 *
 * This is the same as round_jiffies() except that it will never
 * round down.  This is useful for timeouts for which the exact time
 * of firing does not matter too much, as long as they don't fire too
 * early.
 */
unsigned long round_jiffies_up(unsigned long j)
{
	return round_jiffies_common(j, raw_smp_processor_id(), true);
}
EXPORT_SYMBOL_GPL(round_jiffies_up);

/**
 * round_jiffies_up_relative - function to round jiffies up to a full second
 * @j: the time in (relative) jiffies that should be rounded
 *
 * This is the same as round_jiffies_relative() except that it will never
 * round down.  This is useful for timeouts for which the exact time
 * of firing does not matter too much, as long as they don't fire too
 * early.
 */
unsigned long round_jiffies_up_relative(unsigned long j)
{
	return __round_jiffies_up_relative(j, raw_smp_processor_id());
}
EXPORT_SYMBOL_GPL(round_jiffies_up_relative);

323 324
/**
 * set_timer_slack - set the allowed slack for a timer
325
 * @timer: the timer to be modified
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
 * @slack_hz: the amount of time (in jiffies) allowed for rounding
 *
 * Set the amount of time, in jiffies, that a certain timer has
 * in terms of slack. By setting this value, the timer subsystem
 * will schedule the actual timer somewhere between
 * the time mod_timer() asks for, and that time plus the slack.
 *
 * By setting the slack to -1, a percentage of the delay is used
 * instead.
 */
void set_timer_slack(struct timer_list *timer, int slack_hz)
{
	timer->slack = slack_hz;
}
EXPORT_SYMBOL_GPL(set_timer_slack);

342 343 344 345 346 347 348 349 350 351 352 353 354 355
/*
 * If the list is empty, catch up ->timer_jiffies to the current time.
 * The caller must hold the tvec_base lock.  Returns true if the list
 * was empty and therefore ->timer_jiffies was updated.
 */
static bool catchup_timer_jiffies(struct tvec_base *base)
{
	if (!base->all_timers) {
		base->timer_jiffies = jiffies;
		return true;
	}
	return false;
}

356 357
static void
__internal_add_timer(struct tvec_base *base, struct timer_list *timer)
L
Linus Torvalds 已提交
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
{
	unsigned long expires = timer->expires;
	unsigned long idx = expires - base->timer_jiffies;
	struct list_head *vec;

	if (idx < TVR_SIZE) {
		int i = expires & TVR_MASK;
		vec = base->tv1.vec + i;
	} else if (idx < 1 << (TVR_BITS + TVN_BITS)) {
		int i = (expires >> TVR_BITS) & TVN_MASK;
		vec = base->tv2.vec + i;
	} else if (idx < 1 << (TVR_BITS + 2 * TVN_BITS)) {
		int i = (expires >> (TVR_BITS + TVN_BITS)) & TVN_MASK;
		vec = base->tv3.vec + i;
	} else if (idx < 1 << (TVR_BITS + 3 * TVN_BITS)) {
		int i = (expires >> (TVR_BITS + 2 * TVN_BITS)) & TVN_MASK;
		vec = base->tv4.vec + i;
	} else if ((signed long) idx < 0) {
		/*
		 * Can happen if you add a timer with expires == jiffies,
		 * or you set a timer to go off in the past
		 */
		vec = base->tv1.vec + (base->timer_jiffies & TVR_MASK);
	} else {
		int i;
383 384 385
		/* If the timeout is larger than MAX_TVAL (on 64-bit
		 * architectures or with CONFIG_BASE_SMALL=1) then we
		 * use the maximum timeout.
L
Linus Torvalds 已提交
386
		 */
387 388
		if (idx > MAX_TVAL) {
			idx = MAX_TVAL;
L
Linus Torvalds 已提交
389 390 391 392 393 394 395 396 397 398 399
			expires = idx + base->timer_jiffies;
		}
		i = (expires >> (TVR_BITS + 3 * TVN_BITS)) & TVN_MASK;
		vec = base->tv5.vec + i;
	}
	/*
	 * Timers are FIFO:
	 */
	list_add_tail(&timer->entry, vec);
}

400 401
static void internal_add_timer(struct tvec_base *base, struct timer_list *timer)
{
402
	(void)catchup_timer_jiffies(base);
403 404
	__internal_add_timer(base, timer);
	/*
405
	 * Update base->active_timers and base->next_timer
406
	 */
407
	if (!tbase_get_deferrable(timer->base)) {
408 409
		if (!base->active_timers++ ||
		    time_before(timer->expires, base->next_timer))
410 411
			base->next_timer = timer->expires;
	}
412
	base->all_timers++;
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428

	/*
	 * Check whether the other CPU is in dynticks mode and needs
	 * to be triggered to reevaluate the timer wheel.
	 * We are protected against the other CPU fiddling
	 * with the timer by holding the timer base lock. This also
	 * makes sure that a CPU on the way to stop its tick can not
	 * evaluate the timer wheel.
	 *
	 * Spare the IPI for deferrable timers on idle targets though.
	 * The next busy ticks will take care of it. Except full dynticks
	 * require special care against races with idle_cpu(), lets deal
	 * with that later.
	 */
	if (!tbase_get_deferrable(base) || tick_nohz_full_cpu(base->cpu))
		wake_up_nohz_cpu(base->cpu);
429 430
}

431 432 433 434 435 436 437 438 439 440
#ifdef CONFIG_TIMER_STATS
void __timer_stats_timer_set_start_info(struct timer_list *timer, void *addr)
{
	if (timer->start_site)
		return;

	timer->start_site = addr;
	memcpy(timer->start_comm, current->comm, TASK_COMM_LEN);
	timer->start_pid = current->pid;
}
441 442 443 444 445

static void timer_stats_account_timer(struct timer_list *timer)
{
	unsigned int flag = 0;

446 447
	if (likely(!timer->start_site))
		return;
448 449 450 451 452 453 454 455 456
	if (unlikely(tbase_get_deferrable(timer->base)))
		flag |= TIMER_STATS_FLAG_DEFERRABLE;

	timer_stats_update_stats(timer, timer->start_pid, timer->start_site,
				 timer->function, timer->start_comm, flag);
}

#else
static void timer_stats_account_timer(struct timer_list *timer) {}
457 458
#endif

459 460 461 462
#ifdef CONFIG_DEBUG_OBJECTS_TIMERS

static struct debug_obj_descr timer_debug_descr;

463 464 465 466 467
static void *timer_debug_hint(void *addr)
{
	return ((struct timer_list *) addr)->function;
}

468 469 470
/*
 * fixup_init is called when:
 * - an active object is initialized
471
 */
472 473 474 475 476 477 478 479 480 481 482 483 484 485
static int timer_fixup_init(void *addr, enum debug_obj_state state)
{
	struct timer_list *timer = addr;

	switch (state) {
	case ODEBUG_STATE_ACTIVE:
		del_timer_sync(timer);
		debug_object_init(timer, &timer_debug_descr);
		return 1;
	default:
		return 0;
	}
}

486 487 488 489 490 491
/* Stub timer callback for improperly used timers. */
static void stub_timer(unsigned long data)
{
	WARN_ON(1);
}

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514
/*
 * fixup_activate is called when:
 * - an active object is activated
 * - an unknown object is activated (might be a statically initialized object)
 */
static int timer_fixup_activate(void *addr, enum debug_obj_state state)
{
	struct timer_list *timer = addr;

	switch (state) {

	case ODEBUG_STATE_NOTAVAILABLE:
		/*
		 * This is not really a fixup. The timer was
		 * statically initialized. We just make sure that it
		 * is tracked in the object tracker.
		 */
		if (timer->entry.next == NULL &&
		    timer->entry.prev == TIMER_ENTRY_STATIC) {
			debug_object_init(timer, &timer_debug_descr);
			debug_object_activate(timer, &timer_debug_descr);
			return 0;
		} else {
515 516
			setup_timer(timer, stub_timer, 0);
			return 1;
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
		}
		return 0;

	case ODEBUG_STATE_ACTIVE:
		WARN_ON(1);

	default:
		return 0;
	}
}

/*
 * fixup_free is called when:
 * - an active object is freed
 */
static int timer_fixup_free(void *addr, enum debug_obj_state state)
{
	struct timer_list *timer = addr;

	switch (state) {
	case ODEBUG_STATE_ACTIVE:
		del_timer_sync(timer);
		debug_object_free(timer, &timer_debug_descr);
		return 1;
	default:
		return 0;
	}
}

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
/*
 * fixup_assert_init is called when:
 * - an untracked/uninit-ed object is found
 */
static int timer_fixup_assert_init(void *addr, enum debug_obj_state state)
{
	struct timer_list *timer = addr;

	switch (state) {
	case ODEBUG_STATE_NOTAVAILABLE:
		if (timer->entry.prev == TIMER_ENTRY_STATIC) {
			/*
			 * This is not really a fixup. The timer was
			 * statically initialized. We just make sure that it
			 * is tracked in the object tracker.
			 */
			debug_object_init(timer, &timer_debug_descr);
			return 0;
		} else {
			setup_timer(timer, stub_timer, 0);
			return 1;
		}
	default:
		return 0;
	}
}

573
static struct debug_obj_descr timer_debug_descr = {
574 575 576 577 578 579
	.name			= "timer_list",
	.debug_hint		= timer_debug_hint,
	.fixup_init		= timer_fixup_init,
	.fixup_activate		= timer_fixup_activate,
	.fixup_free		= timer_fixup_free,
	.fixup_assert_init	= timer_fixup_assert_init,
580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
};

static inline void debug_timer_init(struct timer_list *timer)
{
	debug_object_init(timer, &timer_debug_descr);
}

static inline void debug_timer_activate(struct timer_list *timer)
{
	debug_object_activate(timer, &timer_debug_descr);
}

static inline void debug_timer_deactivate(struct timer_list *timer)
{
	debug_object_deactivate(timer, &timer_debug_descr);
}

static inline void debug_timer_free(struct timer_list *timer)
{
	debug_object_free(timer, &timer_debug_descr);
}

602 603 604 605 606
static inline void debug_timer_assert_init(struct timer_list *timer)
{
	debug_object_assert_init(timer, &timer_debug_descr);
}

T
Tejun Heo 已提交
607 608
static void do_init_timer(struct timer_list *timer, unsigned int flags,
			  const char *name, struct lock_class_key *key);
609

T
Tejun Heo 已提交
610 611
void init_timer_on_stack_key(struct timer_list *timer, unsigned int flags,
			     const char *name, struct lock_class_key *key)
612 613
{
	debug_object_init_on_stack(timer, &timer_debug_descr);
T
Tejun Heo 已提交
614
	do_init_timer(timer, flags, name, key);
615
}
616
EXPORT_SYMBOL_GPL(init_timer_on_stack_key);
617 618 619 620 621 622 623 624 625 626 627

void destroy_timer_on_stack(struct timer_list *timer)
{
	debug_object_free(timer, &timer_debug_descr);
}
EXPORT_SYMBOL_GPL(destroy_timer_on_stack);

#else
static inline void debug_timer_init(struct timer_list *timer) { }
static inline void debug_timer_activate(struct timer_list *timer) { }
static inline void debug_timer_deactivate(struct timer_list *timer) { }
628
static inline void debug_timer_assert_init(struct timer_list *timer) { }
629 630
#endif

631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
static inline void debug_init(struct timer_list *timer)
{
	debug_timer_init(timer);
	trace_timer_init(timer);
}

static inline void
debug_activate(struct timer_list *timer, unsigned long expires)
{
	debug_timer_activate(timer);
	trace_timer_start(timer, expires);
}

static inline void debug_deactivate(struct timer_list *timer)
{
	debug_timer_deactivate(timer);
	trace_timer_cancel(timer);
}

650 651 652 653 654
static inline void debug_assert_init(struct timer_list *timer)
{
	debug_timer_assert_init(timer);
}

T
Tejun Heo 已提交
655 656
static void do_init_timer(struct timer_list *timer, unsigned int flags,
			  const char *name, struct lock_class_key *key)
657
{
658
	struct tvec_base *base = raw_cpu_read(tvec_bases);
T
Tejun Heo 已提交
659

660
	timer->entry.next = NULL;
T
Tejun Heo 已提交
661
	timer->base = (void *)((unsigned long)base | flags);
662
	timer->slack = -1;
663 664 665 666 667
#ifdef CONFIG_TIMER_STATS
	timer->start_site = NULL;
	timer->start_pid = -1;
	memset(timer->start_comm, 0, TASK_COMM_LEN);
#endif
668
	lockdep_init_map(&timer->lockdep_map, name, key, 0);
669
}
670 671

/**
R
Randy Dunlap 已提交
672
 * init_timer_key - initialize a timer
673
 * @timer: the timer to be initialized
T
Tejun Heo 已提交
674
 * @flags: timer flags
R
Randy Dunlap 已提交
675 676 677
 * @name: name of the timer
 * @key: lockdep class key of the fake lock used for tracking timer
 *       sync lock dependencies
678
 *
R
Randy Dunlap 已提交
679
 * init_timer_key() must be done to a timer prior calling *any* of the
680 681
 * other timer functions.
 */
T
Tejun Heo 已提交
682 683
void init_timer_key(struct timer_list *timer, unsigned int flags,
		    const char *name, struct lock_class_key *key)
684
{
685
	debug_init(timer);
T
Tejun Heo 已提交
686
	do_init_timer(timer, flags, name, key);
687
}
688
EXPORT_SYMBOL(init_timer_key);
689

690
static inline void detach_timer(struct timer_list *timer, bool clear_pending)
691 692 693
{
	struct list_head *entry = &timer->entry;

694
	debug_deactivate(timer);
695

696 697 698 699 700 701
	__list_del(entry->prev, entry->next);
	if (clear_pending)
		entry->next = NULL;
	entry->prev = LIST_POISON2;
}

702 703 704 705 706
static inline void
detach_expired_timer(struct timer_list *timer, struct tvec_base *base)
{
	detach_timer(timer, true);
	if (!tbase_get_deferrable(timer->base))
707
		base->active_timers--;
708
	base->all_timers--;
709
	(void)catchup_timer_jiffies(base);
710 711
}

712 713 714 715 716 717 718
static int detach_if_pending(struct timer_list *timer, struct tvec_base *base,
			     bool clear_pending)
{
	if (!timer_pending(timer))
		return 0;

	detach_timer(timer, clear_pending);
719
	if (!tbase_get_deferrable(timer->base)) {
720
		base->active_timers--;
721 722 723
		if (timer->expires == base->next_timer)
			base->next_timer = base->timer_jiffies;
	}
724
	base->all_timers--;
725
	(void)catchup_timer_jiffies(base);
726 727 728
	return 1;
}

729
/*
730
 * We are using hashed locking: holding per_cpu(tvec_bases).lock
731 732 733 734 735 736 737 738 739 740
 * means that all timers which are tied to this base via timer->base are
 * locked, and the base itself is locked too.
 *
 * So __run_timers/migrate_timers can safely modify all timers which could
 * be found on ->tvX lists.
 *
 * When the timer's base is locked, and the timer removed from list, it is
 * possible to set timer->base = NULL and drop the lock: the timer remains
 * locked.
 */
741
static struct tvec_base *lock_timer_base(struct timer_list *timer,
742
					unsigned long *flags)
743
	__acquires(timer->base->lock)
744
{
745
	struct tvec_base *base;
746 747

	for (;;) {
748
		struct tvec_base *prelock_base = timer->base;
749
		base = tbase_get_base(prelock_base);
750 751
		if (likely(base != NULL)) {
			spin_lock_irqsave(&base->lock, *flags);
752
			if (likely(prelock_base == timer->base))
753 754 755 756 757 758 759 760
				return base;
			/* The timer has migrated to another CPU */
			spin_unlock_irqrestore(&base->lock, *flags);
		}
		cpu_relax();
	}
}

I
Ingo Molnar 已提交
761
static inline int
762 763
__mod_timer(struct timer_list *timer, unsigned long expires,
						bool pending_only, int pinned)
L
Linus Torvalds 已提交
764
{
765
	struct tvec_base *base, *new_base;
L
Linus Torvalds 已提交
766
	unsigned long flags;
767
	int ret = 0 , cpu;
L
Linus Torvalds 已提交
768

769
	timer_stats_timer_set_start_info(timer);
L
Linus Torvalds 已提交
770 771
	BUG_ON(!timer->function);

772 773
	base = lock_timer_base(timer, &flags);

774 775 776
	ret = detach_if_pending(timer, base, false);
	if (!ret && pending_only)
		goto out_unlock;
777

778
	debug_activate(timer, expires);
779

780
	cpu = get_nohz_timer_target(pinned);
781 782
	new_base = per_cpu(tvec_bases, cpu);

783
	if (base != new_base) {
L
Linus Torvalds 已提交
784
		/*
785 786 787 788 789
		 * We are trying to schedule the timer on the local CPU.
		 * However we can't change timer's base while it is running,
		 * otherwise del_timer_sync() can't detect that the timer's
		 * handler yet has not finished. This also guarantees that
		 * the timer is serialized wrt itself.
L
Linus Torvalds 已提交
790
		 */
791
		if (likely(base->running_timer != timer)) {
792
			/* See the comment in lock_timer_base() */
793
			timer_set_base(timer, NULL);
794
			spin_unlock(&base->lock);
795 796
			base = new_base;
			spin_lock(&base->lock);
797
			timer_set_base(timer, base);
L
Linus Torvalds 已提交
798 799 800 801
		}
	}

	timer->expires = expires;
802
	internal_add_timer(base, timer);
I
Ingo Molnar 已提交
803 804

out_unlock:
805
	spin_unlock_irqrestore(&base->lock, flags);
L
Linus Torvalds 已提交
806 807 808 809

	return ret;
}

810
/**
I
Ingo Molnar 已提交
811 812 813
 * mod_timer_pending - modify a pending timer's timeout
 * @timer: the pending timer to be modified
 * @expires: new timeout in jiffies
L
Linus Torvalds 已提交
814
 *
I
Ingo Molnar 已提交
815 816 817 818
 * mod_timer_pending() is the same for pending timers as mod_timer(),
 * but will not re-activate and modify already deleted timers.
 *
 * It is useful for unserialized use of timers.
L
Linus Torvalds 已提交
819
 */
I
Ingo Molnar 已提交
820
int mod_timer_pending(struct timer_list *timer, unsigned long expires)
L
Linus Torvalds 已提交
821
{
822
	return __mod_timer(timer, expires, true, TIMER_NOT_PINNED);
L
Linus Torvalds 已提交
823
}
I
Ingo Molnar 已提交
824
EXPORT_SYMBOL(mod_timer_pending);
L
Linus Torvalds 已提交
825

826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841
/*
 * Decide where to put the timer while taking the slack into account
 *
 * Algorithm:
 *   1) calculate the maximum (absolute) time
 *   2) calculate the highest bit where the expires and new max are different
 *   3) use this bit to make a mask
 *   4) use the bitmask to round down the maximum time, so that all last
 *      bits are zeros
 */
static inline
unsigned long apply_slack(struct timer_list *timer, unsigned long expires)
{
	unsigned long expires_limit, mask;
	int bit;

842
	if (timer->slack >= 0) {
843
		expires_limit = expires + timer->slack;
844
	} else {
845 846 847 848
		long delta = expires - jiffies;

		if (delta < 256)
			return expires;
849

850
		expires_limit = expires + delta / 256;
851
	}
852 853 854 855 856 857
	mask = expires ^ expires_limit;
	if (mask == 0)
		return expires;

	bit = find_last_bit(&mask, BITS_PER_LONG);

858
	mask = (1UL << bit) - 1;
859 860 861 862 863 864

	expires_limit = expires_limit & ~(mask);

	return expires_limit;
}

865
/**
L
Linus Torvalds 已提交
866 867
 * mod_timer - modify a timer's timeout
 * @timer: the timer to be modified
868
 * @expires: new timeout in jiffies
L
Linus Torvalds 已提交
869
 *
870
 * mod_timer() is a more efficient way to update the expire field of an
L
Linus Torvalds 已提交
871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886
 * active timer (if the timer is inactive it will be activated)
 *
 * mod_timer(timer, expires) is equivalent to:
 *
 *     del_timer(timer); timer->expires = expires; add_timer(timer);
 *
 * Note that if there are multiple unserialized concurrent users of the
 * same timer, then mod_timer() is the only safe way to modify the timeout,
 * since add_timer() cannot modify an already running timer.
 *
 * The function returns whether it has modified a pending timer or not.
 * (ie. mod_timer() of an inactive timer returns 0, mod_timer() of an
 * active timer returns 1.)
 */
int mod_timer(struct timer_list *timer, unsigned long expires)
{
887 888
	expires = apply_slack(timer, expires);

L
Linus Torvalds 已提交
889 890 891 892 893
	/*
	 * This is a common optimization triggered by the
	 * networking code - if the timer is re-modified
	 * to be the same thing then just return:
	 */
894
	if (timer_pending(timer) && timer->expires == expires)
L
Linus Torvalds 已提交
895 896
		return 1;

897
	return __mod_timer(timer, expires, false, TIMER_NOT_PINNED);
L
Linus Torvalds 已提交
898 899 900
}
EXPORT_SYMBOL(mod_timer);

901 902 903 904 905 906 907
/**
 * mod_timer_pinned - modify a timer's timeout
 * @timer: the timer to be modified
 * @expires: new timeout in jiffies
 *
 * mod_timer_pinned() is a way to update the expire field of an
 * active timer (if the timer is inactive it will be activated)
908 909 910 911 912 913 914
 * and to ensure that the timer is scheduled on the current CPU.
 *
 * Note that this does not prevent the timer from being migrated
 * when the current CPU goes offline.  If this is a problem for
 * you, use CPU-hotplug notifiers to handle it correctly, for
 * example, cancelling the timer when the corresponding CPU goes
 * offline.
915 916 917 918 919 920 921 922 923 924 925 926 927 928
 *
 * mod_timer_pinned(timer, expires) is equivalent to:
 *
 *     del_timer(timer); timer->expires = expires; add_timer(timer);
 */
int mod_timer_pinned(struct timer_list *timer, unsigned long expires)
{
	if (timer->expires == expires && timer_pending(timer))
		return 1;

	return __mod_timer(timer, expires, false, TIMER_PINNED);
}
EXPORT_SYMBOL(mod_timer_pinned);

I
Ingo Molnar 已提交
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
/**
 * add_timer - start a timer
 * @timer: the timer to be added
 *
 * The kernel will do a ->function(->data) callback from the
 * timer interrupt at the ->expires point in the future. The
 * current time is 'jiffies'.
 *
 * The timer's ->expires, ->function (and if the handler uses it, ->data)
 * fields must be set prior calling this function.
 *
 * Timers with an ->expires field in the past will be executed in the next
 * timer tick.
 */
void add_timer(struct timer_list *timer)
{
	BUG_ON(timer_pending(timer));
	mod_timer(timer, timer->expires);
}
EXPORT_SYMBOL(add_timer);

/**
 * add_timer_on - start a timer on a particular CPU
 * @timer: the timer to be added
 * @cpu: the CPU to start it on
 *
 * This is not very scalable on SMP. Double adds are not possible.
 */
void add_timer_on(struct timer_list *timer, int cpu)
{
	struct tvec_base *base = per_cpu(tvec_bases, cpu);
	unsigned long flags;

	timer_stats_timer_set_start_info(timer);
	BUG_ON(timer_pending(timer) || !timer->function);
	spin_lock_irqsave(&base->lock, flags);
	timer_set_base(timer, base);
966
	debug_activate(timer, timer->expires);
I
Ingo Molnar 已提交
967 968 969
	internal_add_timer(base, timer);
	spin_unlock_irqrestore(&base->lock, flags);
}
A
Andi Kleen 已提交
970
EXPORT_SYMBOL_GPL(add_timer_on);
I
Ingo Molnar 已提交
971

972
/**
L
Linus Torvalds 已提交
973 974 975 976 977 978 979 980 981 982 983 984
 * del_timer - deactive a timer.
 * @timer: the timer to be deactivated
 *
 * del_timer() deactivates a timer - this works on both active and inactive
 * timers.
 *
 * The function returns whether it has deactivated a pending timer or not.
 * (ie. del_timer() of an inactive timer returns 0, del_timer() of an
 * active timer returns 1.)
 */
int del_timer(struct timer_list *timer)
{
985
	struct tvec_base *base;
L
Linus Torvalds 已提交
986
	unsigned long flags;
987
	int ret = 0;
L
Linus Torvalds 已提交
988

989 990
	debug_assert_init(timer);

991
	timer_stats_timer_clear_start_info(timer);
992 993
	if (timer_pending(timer)) {
		base = lock_timer_base(timer, &flags);
994
		ret = detach_if_pending(timer, base, true);
L
Linus Torvalds 已提交
995 996 997
		spin_unlock_irqrestore(&base->lock, flags);
	}

998
	return ret;
L
Linus Torvalds 已提交
999 1000 1001
}
EXPORT_SYMBOL(del_timer);

1002 1003 1004 1005
/**
 * try_to_del_timer_sync - Try to deactivate a timer
 * @timer: timer do del
 *
1006 1007 1008 1009 1010
 * This function tries to deactivate a timer. Upon successful (ret >= 0)
 * exit the timer is not queued and the handler is not running on any CPU.
 */
int try_to_del_timer_sync(struct timer_list *timer)
{
1011
	struct tvec_base *base;
1012 1013 1014
	unsigned long flags;
	int ret = -1;

1015 1016
	debug_assert_init(timer);

1017 1018
	base = lock_timer_base(timer, &flags);

1019 1020 1021
	if (base->running_timer != timer) {
		timer_stats_timer_clear_start_info(timer);
		ret = detach_if_pending(timer, base, true);
1022 1023 1024 1025 1026
	}
	spin_unlock_irqrestore(&base->lock, flags);

	return ret;
}
1027 1028
EXPORT_SYMBOL(try_to_del_timer_sync);

1029
#ifdef CONFIG_SMP
1030
/**
L
Linus Torvalds 已提交
1031 1032 1033 1034 1035 1036 1037
 * del_timer_sync - deactivate a timer and wait for the handler to finish.
 * @timer: the timer to be deactivated
 *
 * This function only differs from del_timer() on SMP: besides deactivating
 * the timer it also makes sure the handler has finished executing on other
 * CPUs.
 *
1038
 * Synchronization rules: Callers must prevent restarting of the timer,
L
Linus Torvalds 已提交
1039
 * otherwise this function is meaningless. It must not be called from
T
Tejun Heo 已提交
1040 1041 1042 1043
 * interrupt contexts unless the timer is an irqsafe one. The caller must
 * not hold locks which would prevent completion of the timer's
 * handler. The timer's handler must not call add_timer_on(). Upon exit the
 * timer is not queued and the handler is not running on any CPU.
L
Linus Torvalds 已提交
1044
 *
T
Tejun Heo 已提交
1045 1046 1047
 * Note: For !irqsafe timers, you must not hold locks that are held in
 *   interrupt context while calling this function. Even if the lock has
 *   nothing to do with the timer in question.  Here's why:
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
 *
 *    CPU0                             CPU1
 *    ----                             ----
 *                                   <SOFTIRQ>
 *                                   call_timer_fn();
 *                                     base->running_timer = mytimer;
 *  spin_lock_irq(somelock);
 *                                     <IRQ>
 *                                        spin_lock(somelock);
 *  del_timer_sync(mytimer);
 *   while (base->running_timer == mytimer);
 *
 * Now del_timer_sync() will never return and never release somelock.
 * The interrupt on the other CPU is waiting to grab somelock but
 * it has interrupted the softirq that CPU0 is waiting to finish.
 *
L
Linus Torvalds 已提交
1064 1065 1066 1067
 * The function returns whether it has deactivated a pending timer or not.
 */
int del_timer_sync(struct timer_list *timer)
{
1068
#ifdef CONFIG_LOCKDEP
1069 1070
	unsigned long flags;

1071 1072 1073 1074
	/*
	 * If lockdep gives a backtrace here, please reference
	 * the synchronization rules above.
	 */
1075
	local_irq_save(flags);
1076 1077
	lock_map_acquire(&timer->lockdep_map);
	lock_map_release(&timer->lockdep_map);
1078
	local_irq_restore(flags);
1079
#endif
1080 1081 1082 1083
	/*
	 * don't use it in hardirq context, because it
	 * could lead to deadlock.
	 */
T
Tejun Heo 已提交
1084
	WARN_ON(in_irq() && !tbase_get_irqsafe(timer->base));
1085 1086 1087 1088
	for (;;) {
		int ret = try_to_del_timer_sync(timer);
		if (ret >= 0)
			return ret;
1089
		cpu_relax();
1090
	}
L
Linus Torvalds 已提交
1091
}
1092
EXPORT_SYMBOL(del_timer_sync);
L
Linus Torvalds 已提交
1093 1094
#endif

1095
static int cascade(struct tvec_base *base, struct tvec *tv, int index)
L
Linus Torvalds 已提交
1096 1097
{
	/* cascade all the timers from tv up one level */
1098 1099 1100 1101
	struct timer_list *timer, *tmp;
	struct list_head tv_list;

	list_replace_init(tv->vec + index, &tv_list);
L
Linus Torvalds 已提交
1102 1103

	/*
1104 1105
	 * We are removing _all_ timers from the list, so we
	 * don't have to detach them individually.
L
Linus Torvalds 已提交
1106
	 */
1107
	list_for_each_entry_safe(timer, tmp, &tv_list, entry) {
1108
		BUG_ON(tbase_get_base(timer->base) != base);
1109 1110
		/* No accounting, while moving them */
		__internal_add_timer(base, timer);
L
Linus Torvalds 已提交
1111 1112 1113 1114 1115
	}

	return index;
}

1116 1117 1118
static void call_timer_fn(struct timer_list *timer, void (*fn)(unsigned long),
			  unsigned long data)
{
1119
	int count = preempt_count();
1120 1121 1122 1123 1124 1125 1126 1127 1128

#ifdef CONFIG_LOCKDEP
	/*
	 * It is permissible to free the timer from inside the
	 * function that is called from it, this we need to take into
	 * account for lockdep too. To avoid bogus "held lock freed"
	 * warnings as well as problems when looking into
	 * timer->lockdep_map, make a copy and use that here.
	 */
1129 1130 1131
	struct lockdep_map lockdep_map;

	lockdep_copy_map(&lockdep_map, &timer->lockdep_map);
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145
#endif
	/*
	 * Couple the lock chain with the lock chain at
	 * del_timer_sync() by acquiring the lock_map around the fn()
	 * call here and in del_timer_sync().
	 */
	lock_map_acquire(&lockdep_map);

	trace_timer_expire_entry(timer);
	fn(data);
	trace_timer_expire_exit(timer);

	lock_map_release(&lockdep_map);

1146
	if (count != preempt_count()) {
1147
		WARN_ONCE(1, "timer: %pF preempt leak: %08x -> %08x\n",
1148
			  fn, count, preempt_count());
1149 1150 1151 1152 1153 1154
		/*
		 * Restore the preempt count. That gives us a decent
		 * chance to survive and extract information. If the
		 * callback kept a lock held, bad luck, but not worse
		 * than the BUG() we had.
		 */
1155
		preempt_count_set(count);
1156 1157 1158
	}
}

1159 1160 1161
#define INDEX(N) ((base->timer_jiffies >> (TVR_BITS + (N) * TVN_BITS)) & TVN_MASK)

/**
L
Linus Torvalds 已提交
1162 1163 1164 1165 1166 1167
 * __run_timers - run all expired timers (if any) on this CPU.
 * @base: the timer vector to be processed.
 *
 * This function cascades all vectors and executes all expired timer
 * vectors.
 */
1168
static inline void __run_timers(struct tvec_base *base)
L
Linus Torvalds 已提交
1169 1170 1171
{
	struct timer_list *timer;

1172
	spin_lock_irq(&base->lock);
1173 1174 1175 1176
	if (catchup_timer_jiffies(base)) {
		spin_unlock_irq(&base->lock);
		return;
	}
L
Linus Torvalds 已提交
1177
	while (time_after_eq(jiffies, base->timer_jiffies)) {
1178
		struct list_head work_list;
L
Linus Torvalds 已提交
1179
		struct list_head *head = &work_list;
1180
		int index = base->timer_jiffies & TVR_MASK;
1181

L
Linus Torvalds 已提交
1182 1183 1184 1185 1186 1187 1188 1189
		/*
		 * Cascade timers:
		 */
		if (!index &&
			(!cascade(base, &base->tv2, INDEX(0))) &&
				(!cascade(base, &base->tv3, INDEX(1))) &&
					!cascade(base, &base->tv4, INDEX(2)))
			cascade(base, &base->tv5, INDEX(3));
1190
		++base->timer_jiffies;
1191
		list_replace_init(base->tv1.vec + index, head);
1192
		while (!list_empty(head)) {
L
Linus Torvalds 已提交
1193 1194
			void (*fn)(unsigned long);
			unsigned long data;
T
Tejun Heo 已提交
1195
			bool irqsafe;
L
Linus Torvalds 已提交
1196

1197
			timer = list_first_entry(head, struct timer_list,entry);
1198 1199
			fn = timer->function;
			data = timer->data;
T
Tejun Heo 已提交
1200
			irqsafe = tbase_get_irqsafe(timer->base);
L
Linus Torvalds 已提交
1201

1202 1203
			timer_stats_account_timer(timer);

1204
			base->running_timer = timer;
1205
			detach_expired_timer(timer, base);
1206

T
Tejun Heo 已提交
1207 1208 1209 1210 1211 1212 1213 1214 1215
			if (irqsafe) {
				spin_unlock(&base->lock);
				call_timer_fn(timer, fn, data);
				spin_lock(&base->lock);
			} else {
				spin_unlock_irq(&base->lock);
				call_timer_fn(timer, fn, data);
				spin_lock_irq(&base->lock);
			}
L
Linus Torvalds 已提交
1216 1217
		}
	}
1218
	base->running_timer = NULL;
1219
	spin_unlock_irq(&base->lock);
L
Linus Torvalds 已提交
1220 1221
}

1222
#ifdef CONFIG_NO_HZ_COMMON
L
Linus Torvalds 已提交
1223 1224
/*
 * Find out when the next timer event is due to happen. This
R
Randy Dunlap 已提交
1225 1226
 * is used on S/390 to stop all activity when a CPU is idle.
 * This function needs to be called with interrupts disabled.
L
Linus Torvalds 已提交
1227
 */
1228
static unsigned long __next_timer_interrupt(struct tvec_base *base)
L
Linus Torvalds 已提交
1229
{
1230
	unsigned long timer_jiffies = base->timer_jiffies;
1231
	unsigned long expires = timer_jiffies + NEXT_TIMER_MAX_DELTA;
1232
	int index, slot, array, found = 0;
L
Linus Torvalds 已提交
1233
	struct timer_list *nte;
1234
	struct tvec *varray[4];
L
Linus Torvalds 已提交
1235 1236

	/* Look for timer events in tv1. */
1237
	index = slot = timer_jiffies & TVR_MASK;
L
Linus Torvalds 已提交
1238
	do {
1239
		list_for_each_entry(nte, base->tv1.vec + slot, entry) {
1240 1241
			if (tbase_get_deferrable(nte->base))
				continue;
1242

1243
			found = 1;
L
Linus Torvalds 已提交
1244
			expires = nte->expires;
1245 1246 1247 1248
			/* Look at the cascade bucket(s)? */
			if (!index || slot < index)
				goto cascade;
			return expires;
L
Linus Torvalds 已提交
1249
		}
1250 1251 1252 1253 1254 1255 1256 1257
		slot = (slot + 1) & TVR_MASK;
	} while (slot != index);

cascade:
	/* Calculate the next cascade event */
	if (index)
		timer_jiffies += TVR_SIZE - index;
	timer_jiffies >>= TVR_BITS;
L
Linus Torvalds 已提交
1258 1259 1260 1261 1262 1263

	/* Check tv2-tv5. */
	varray[0] = &base->tv2;
	varray[1] = &base->tv3;
	varray[2] = &base->tv4;
	varray[3] = &base->tv5;
1264 1265

	for (array = 0; array < 4; array++) {
1266
		struct tvec *varp = varray[array];
1267 1268

		index = slot = timer_jiffies & TVN_MASK;
L
Linus Torvalds 已提交
1269
		do {
1270
			list_for_each_entry(nte, varp->vec + slot, entry) {
1271 1272 1273
				if (tbase_get_deferrable(nte->base))
					continue;

1274
				found = 1;
L
Linus Torvalds 已提交
1275 1276
				if (time_before(nte->expires, expires))
					expires = nte->expires;
1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293
			}
			/*
			 * Do we still search for the first timer or are
			 * we looking up the cascade buckets ?
			 */
			if (found) {
				/* Look at the cascade bucket(s)? */
				if (!index || slot < index)
					break;
				return expires;
			}
			slot = (slot + 1) & TVN_MASK;
		} while (slot != index);

		if (index)
			timer_jiffies += TVN_SIZE - index;
		timer_jiffies >>= TVN_BITS;
L
Linus Torvalds 已提交
1294
	}
1295 1296
	return expires;
}
1297

1298 1299 1300 1301 1302 1303 1304 1305 1306
/*
 * Check, if the next hrtimer event is before the next timer wheel
 * event:
 */
static unsigned long cmp_next_hrtimer_event(unsigned long now,
					    unsigned long expires)
{
	ktime_t hr_delta = hrtimer_get_next_event();
	struct timespec tsdelta;
1307
	unsigned long delta;
1308 1309 1310

	if (hr_delta.tv64 == KTIME_MAX)
		return expires;
1311

1312 1313 1314 1315 1316
	/*
	 * Expired timer available, let it expire in the next tick
	 */
	if (hr_delta.tv64 <= 0)
		return now + 1;
1317

1318
	tsdelta = ktime_to_timespec(hr_delta);
1319
	delta = timespec_to_jiffies(&tsdelta);
1320 1321 1322 1323 1324 1325 1326 1327

	/*
	 * Limit the delta to the max value, which is checked in
	 * tick_nohz_stop_sched_tick():
	 */
	if (delta > NEXT_TIMER_MAX_DELTA)
		delta = NEXT_TIMER_MAX_DELTA;

1328 1329 1330 1331 1332 1333 1334 1335 1336
	/*
	 * Take rounding errors in to account and make sure, that it
	 * expires in the next tick. Otherwise we go into an endless
	 * ping pong due to tick_nohz_stop_sched_tick() retriggering
	 * the timer softirq
	 */
	if (delta < 1)
		delta = 1;
	now += delta;
1337 1338
	if (time_before(now, expires))
		return now;
L
Linus Torvalds 已提交
1339 1340
	return expires;
}
1341 1342

/**
1343
 * get_next_timer_interrupt - return the jiffy of the next pending timer
1344
 * @now: current time (in jiffies)
1345
 */
1346
unsigned long get_next_timer_interrupt(unsigned long now)
1347
{
C
Christoph Lameter 已提交
1348
	struct tvec_base *base = __this_cpu_read(tvec_bases);
1349
	unsigned long expires = now + NEXT_TIMER_MAX_DELTA;
1350

1351 1352 1353 1354 1355
	/*
	 * Pretend that there is no timer pending if the cpu is offline.
	 * Possible pending timers will be migrated later to an active cpu.
	 */
	if (cpu_is_offline(smp_processor_id()))
1356 1357
		return expires;

1358
	spin_lock(&base->lock);
1359 1360 1361 1362 1363
	if (base->active_timers) {
		if (time_before_eq(base->next_timer, base->timer_jiffies))
			base->next_timer = __next_timer_interrupt(base);
		expires = base->next_timer;
	}
1364 1365 1366 1367 1368 1369 1370
	spin_unlock(&base->lock);

	if (time_before_eq(expires, now))
		return now;

	return cmp_next_hrtimer_event(now, expires);
}
L
Linus Torvalds 已提交
1371 1372 1373
#endif

/*
D
Daniel Walker 已提交
1374
 * Called from the timer interrupt handler to charge one tick to the current
L
Linus Torvalds 已提交
1375 1376 1377 1378 1379 1380 1381 1382
 * process.  user_tick is 1 if the tick is user time, 0 for system.
 */
void update_process_times(int user_tick)
{
	struct task_struct *p = current;
	int cpu = smp_processor_id();

	/* Note: this timer irq context must be accounted for as well. */
1383
	account_process_tick(p, user_tick);
L
Linus Torvalds 已提交
1384
	run_local_timers();
1385
	rcu_check_callbacks(cpu, user_tick);
1386 1387
#ifdef CONFIG_IRQ_WORK
	if (in_irq())
1388
		irq_work_tick();
1389
#endif
L
Linus Torvalds 已提交
1390
	scheduler_tick();
1391
	run_posix_cpu_timers(p);
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396 1397 1398
}

/*
 * This function runs timers and the timer-tq in bottom half context.
 */
static void run_timer_softirq(struct softirq_action *h)
{
C
Christoph Lameter 已提交
1399
	struct tvec_base *base = __this_cpu_read(tvec_bases);
L
Linus Torvalds 已提交
1400

1401
	hrtimer_run_pending();
1402

L
Linus Torvalds 已提交
1403 1404 1405 1406 1407 1408 1409 1410 1411
	if (time_after_eq(jiffies, base->timer_jiffies))
		__run_timers(base);
}

/*
 * Called by the local, per-CPU timer interrupt on SMP.
 */
void run_local_timers(void)
{
1412
	hrtimer_run_queues();
L
Linus Torvalds 已提交
1413 1414 1415 1416 1417 1418 1419 1420 1421
	raise_softirq(TIMER_SOFTIRQ);
}

#ifdef __ARCH_WANT_SYS_ALARM

/*
 * For backwards compatibility?  This can be done in libc so Alpha
 * and all newer ports shouldn't need it.
 */
1422
SYSCALL_DEFINE1(alarm, unsigned int, seconds)
L
Linus Torvalds 已提交
1423
{
1424
	return alarm_setitimer(seconds);
L
Linus Torvalds 已提交
1425 1426 1427 1428 1429 1430
}

#endif

static void process_timeout(unsigned long __data)
{
1431
	wake_up_process((struct task_struct *)__data);
L
Linus Torvalds 已提交
1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459
}

/**
 * schedule_timeout - sleep until timeout
 * @timeout: timeout value in jiffies
 *
 * Make the current task sleep until @timeout jiffies have
 * elapsed. The routine will return immediately unless
 * the current task state has been set (see set_current_state()).
 *
 * You can set the task state as follows -
 *
 * %TASK_UNINTERRUPTIBLE - at least @timeout jiffies are guaranteed to
 * pass before the routine returns. The routine will return 0
 *
 * %TASK_INTERRUPTIBLE - the routine may return early if a signal is
 * delivered to the current task. In this case the remaining time
 * in jiffies will be returned, or 0 if the timer expired in time
 *
 * The current task state is guaranteed to be TASK_RUNNING when this
 * routine returns.
 *
 * Specifying a @timeout value of %MAX_SCHEDULE_TIMEOUT will schedule
 * the CPU away without a bound on the timeout. In this case the return
 * value will be %MAX_SCHEDULE_TIMEOUT.
 *
 * In all cases the return value is guaranteed to be non-negative.
 */
1460
signed long __sched schedule_timeout(signed long timeout)
L
Linus Torvalds 已提交
1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484
{
	struct timer_list timer;
	unsigned long expire;

	switch (timeout)
	{
	case MAX_SCHEDULE_TIMEOUT:
		/*
		 * These two special cases are useful to be comfortable
		 * in the caller. Nothing more. We could take
		 * MAX_SCHEDULE_TIMEOUT from one of the negative value
		 * but I' d like to return a valid offset (>=0) to allow
		 * the caller to do everything it want with the retval.
		 */
		schedule();
		goto out;
	default:
		/*
		 * Another bit of PARANOID. Note that the retval will be
		 * 0 since no piece of kernel is supposed to do a check
		 * for a negative retval of schedule_timeout() (since it
		 * should never happens anyway). You just have the printk()
		 * that will tell you if something is gone wrong and where.
		 */
1485
		if (timeout < 0) {
L
Linus Torvalds 已提交
1486
			printk(KERN_ERR "schedule_timeout: wrong timeout "
1487 1488
				"value %lx\n", timeout);
			dump_stack();
L
Linus Torvalds 已提交
1489 1490 1491 1492 1493 1494 1495
			current->state = TASK_RUNNING;
			goto out;
		}
	}

	expire = timeout + jiffies;

1496
	setup_timer_on_stack(&timer, process_timeout, (unsigned long)current);
1497
	__mod_timer(&timer, expire, false, TIMER_NOT_PINNED);
L
Linus Torvalds 已提交
1498 1499 1500
	schedule();
	del_singleshot_timer_sync(&timer);

1501 1502 1503
	/* Remove the timer from the object tracker */
	destroy_timer_on_stack(&timer);

L
Linus Torvalds 已提交
1504 1505 1506 1507 1508 1509 1510
	timeout = expire - jiffies;

 out:
	return timeout < 0 ? 0 : timeout;
}
EXPORT_SYMBOL(schedule_timeout);

1511 1512 1513 1514
/*
 * We can use __set_current_state() here because schedule_timeout() calls
 * schedule() unconditionally.
 */
1515 1516
signed long __sched schedule_timeout_interruptible(signed long timeout)
{
A
Andrew Morton 已提交
1517 1518
	__set_current_state(TASK_INTERRUPTIBLE);
	return schedule_timeout(timeout);
1519 1520 1521
}
EXPORT_SYMBOL(schedule_timeout_interruptible);

M
Matthew Wilcox 已提交
1522 1523 1524 1525 1526 1527 1528
signed long __sched schedule_timeout_killable(signed long timeout)
{
	__set_current_state(TASK_KILLABLE);
	return schedule_timeout(timeout);
}
EXPORT_SYMBOL(schedule_timeout_killable);

1529 1530
signed long __sched schedule_timeout_uninterruptible(signed long timeout)
{
A
Andrew Morton 已提交
1531 1532
	__set_current_state(TASK_UNINTERRUPTIBLE);
	return schedule_timeout(timeout);
1533 1534 1535
}
EXPORT_SYMBOL(schedule_timeout_uninterruptible);

1536
static int init_timers_cpu(int cpu)
L
Linus Torvalds 已提交
1537 1538
{
	int j;
1539
	struct tvec_base *base;
1540
	static char tvec_base_done[NR_CPUS];
1541

A
Andrew Morton 已提交
1542
	if (!tvec_base_done[cpu]) {
1543 1544 1545
		static char boot_done;

		if (boot_done) {
A
Andrew Morton 已提交
1546 1547 1548
			/*
			 * The APs use this path later in boot
			 */
1549 1550
			base = kzalloc_node(sizeof(*base), GFP_KERNEL,
					    cpu_to_node(cpu));
1551 1552
			if (!base)
				return -ENOMEM;
1553

1554 1555
			/* Make sure tvec_base has TIMER_FLAG_MASK bits free */
			if (WARN_ON(base != tbase_get_base(base))) {
1556 1557 1558
				kfree(base);
				return -ENOMEM;
			}
A
Andrew Morton 已提交
1559
			per_cpu(tvec_bases, cpu) = base;
1560
		} else {
A
Andrew Morton 已提交
1561 1562 1563 1564 1565 1566
			/*
			 * This is for the boot CPU - we use compile-time
			 * static initialisation because per-cpu memory isn't
			 * ready yet and because the memory allocators are not
			 * initialised either.
			 */
1567
			boot_done = 1;
A
Andrew Morton 已提交
1568
			base = &boot_tvec_bases;
1569
		}
1570
		spin_lock_init(&base->lock);
A
Andrew Morton 已提交
1571
		tvec_base_done[cpu] = 1;
1572
		base->cpu = cpu;
A
Andrew Morton 已提交
1573 1574
	} else {
		base = per_cpu(tvec_bases, cpu);
1575
	}
A
Andrew Morton 已提交
1576

1577

L
Linus Torvalds 已提交
1578 1579 1580 1581 1582 1583 1584 1585 1586 1587
	for (j = 0; j < TVN_SIZE; j++) {
		INIT_LIST_HEAD(base->tv5.vec + j);
		INIT_LIST_HEAD(base->tv4.vec + j);
		INIT_LIST_HEAD(base->tv3.vec + j);
		INIT_LIST_HEAD(base->tv2.vec + j);
	}
	for (j = 0; j < TVR_SIZE; j++)
		INIT_LIST_HEAD(base->tv1.vec + j);

	base->timer_jiffies = jiffies;
1588
	base->next_timer = base->timer_jiffies;
1589
	base->active_timers = 0;
1590
	base->all_timers = 0;
1591
	return 0;
L
Linus Torvalds 已提交
1592 1593 1594
}

#ifdef CONFIG_HOTPLUG_CPU
1595
static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head)
L
Linus Torvalds 已提交
1596 1597 1598 1599
{
	struct timer_list *timer;

	while (!list_empty(head)) {
1600
		timer = list_first_entry(head, struct timer_list, entry);
1601
		/* We ignore the accounting on the dying cpu */
1602
		detach_timer(timer, false);
1603
		timer_set_base(timer, new_base);
L
Linus Torvalds 已提交
1604 1605 1606 1607
		internal_add_timer(new_base, timer);
	}
}

1608
static void migrate_timers(int cpu)
L
Linus Torvalds 已提交
1609
{
1610 1611
	struct tvec_base *old_base;
	struct tvec_base *new_base;
L
Linus Torvalds 已提交
1612 1613 1614
	int i;

	BUG_ON(cpu_online(cpu));
1615 1616
	old_base = per_cpu(tvec_bases, cpu);
	new_base = get_cpu_var(tvec_bases);
1617 1618 1619 1620 1621
	/*
	 * The caller is globally serialized and nobody else
	 * takes two locks at once, deadlock is not possible.
	 */
	spin_lock_irq(&new_base->lock);
1622
	spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
1623 1624

	BUG_ON(old_base->running_timer);
L
Linus Torvalds 已提交
1625 1626

	for (i = 0; i < TVR_SIZE; i++)
1627 1628 1629 1630 1631 1632 1633 1634
		migrate_timer_list(new_base, old_base->tv1.vec + i);
	for (i = 0; i < TVN_SIZE; i++) {
		migrate_timer_list(new_base, old_base->tv2.vec + i);
		migrate_timer_list(new_base, old_base->tv3.vec + i);
		migrate_timer_list(new_base, old_base->tv4.vec + i);
		migrate_timer_list(new_base, old_base->tv5.vec + i);
	}

1635
	spin_unlock(&old_base->lock);
1636
	spin_unlock_irq(&new_base->lock);
L
Linus Torvalds 已提交
1637 1638 1639 1640
	put_cpu_var(tvec_bases);
}
#endif /* CONFIG_HOTPLUG_CPU */

1641
static int timer_cpu_notify(struct notifier_block *self,
L
Linus Torvalds 已提交
1642 1643 1644
				unsigned long action, void *hcpu)
{
	long cpu = (long)hcpu;
1645 1646
	int err;

L
Linus Torvalds 已提交
1647 1648
	switch(action) {
	case CPU_UP_PREPARE:
1649
	case CPU_UP_PREPARE_FROZEN:
1650 1651 1652
		err = init_timers_cpu(cpu);
		if (err < 0)
			return notifier_from_errno(err);
L
Linus Torvalds 已提交
1653 1654 1655
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_DEAD:
1656
	case CPU_DEAD_FROZEN:
L
Linus Torvalds 已提交
1657 1658 1659 1660 1661 1662 1663 1664 1665
		migrate_timers(cpu);
		break;
#endif
	default:
		break;
	}
	return NOTIFY_OK;
}

1666
static struct notifier_block timers_nb = {
L
Linus Torvalds 已提交
1667 1668 1669 1670 1671 1672
	.notifier_call	= timer_cpu_notify,
};


void __init init_timers(void)
{
1673 1674 1675 1676
	int err;

	/* ensure there are enough low bits for flags in timer->base pointer */
	BUILD_BUG_ON(__alignof__(struct tvec_base) & TIMER_FLAG_MASK);
1677

1678 1679
	err = timer_cpu_notify(&timers_nb, (unsigned long)CPU_UP_PREPARE,
			       (void *)(long)smp_processor_id());
1680
	BUG_ON(err != NOTIFY_OK);
1681 1682

	init_timer_stats();
L
Linus Torvalds 已提交
1683
	register_cpu_notifier(&timers_nb);
1684
	open_softirq(TIMER_SOFTIRQ, run_timer_softirq);
L
Linus Torvalds 已提交
1685 1686 1687 1688 1689 1690 1691 1692 1693 1694
}

/**
 * msleep - sleep safely even with waitqueue interruptions
 * @msecs: Time in milliseconds to sleep for
 */
void msleep(unsigned int msecs)
{
	unsigned long timeout = msecs_to_jiffies(msecs) + 1;

1695 1696
	while (timeout)
		timeout = schedule_timeout_uninterruptible(timeout);
L
Linus Torvalds 已提交
1697 1698 1699 1700 1701
}

EXPORT_SYMBOL(msleep);

/**
1702
 * msleep_interruptible - sleep waiting for signals
L
Linus Torvalds 已提交
1703 1704 1705 1706 1707 1708
 * @msecs: Time in milliseconds to sleep for
 */
unsigned long msleep_interruptible(unsigned int msecs)
{
	unsigned long timeout = msecs_to_jiffies(msecs) + 1;

1709 1710
	while (timeout && !signal_pending(current))
		timeout = schedule_timeout_interruptible(timeout);
L
Linus Torvalds 已提交
1711 1712 1713 1714
	return jiffies_to_msecs(timeout);
}

EXPORT_SYMBOL(msleep_interruptible);
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736

static int __sched do_usleep_range(unsigned long min, unsigned long max)
{
	ktime_t kmin;
	unsigned long delta;

	kmin = ktime_set(0, min * NSEC_PER_USEC);
	delta = (max - min) * NSEC_PER_USEC;
	return schedule_hrtimeout_range(&kmin, delta, HRTIMER_MODE_REL);
}

/**
 * usleep_range - Drop in replacement for udelay where wakeup is flexible
 * @min: Minimum time in usecs to sleep
 * @max: Maximum time in usecs to sleep
 */
void usleep_range(unsigned long min, unsigned long max)
{
	__set_current_state(TASK_UNINTERRUPTIBLE);
	do_usleep_range(min, max);
}
EXPORT_SYMBOL(usleep_range);