timer.c 40.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3
/*
 *  linux/kernel/timer.c
 *
4
 *  Kernel internal timers, basic process system calls
L
Linus Torvalds 已提交
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
 *
 *  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>
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/percpu.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/swap.h>
#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
36
#include <linux/delay.h>
37
#include <linux/tick.h>
38
#include <linux/kallsyms.h>
L
Linus Torvalds 已提交
39 40 41 42 43 44 45

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

T
Thomas Gleixner 已提交
46 47 48 49
u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES;

EXPORT_SYMBOL(jiffies_64);

L
Linus Torvalds 已提交
50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
/*
 * 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)

typedef struct tvec_s {
	struct list_head vec[TVN_SIZE];
} tvec_t;

typedef struct tvec_root_s {
	struct list_head vec[TVR_SIZE];
} tvec_root_t;

struct tvec_t_base_s {
69 70
	spinlock_t lock;
	struct timer_list *running_timer;
L
Linus Torvalds 已提交
71 72 73 74 75 76
	unsigned long timer_jiffies;
	tvec_root_t tv1;
	tvec_t tv2;
	tvec_t tv3;
	tvec_t tv4;
	tvec_t tv5;
77
} ____cacheline_aligned;
L
Linus Torvalds 已提交
78 79

typedef struct tvec_t_base_s tvec_base_t;
A
Andrew Morton 已提交
80

81 82
tvec_base_t boot_tvec_bases;
EXPORT_SYMBOL(boot_tvec_bases);
83
static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases;
L
Linus Torvalds 已提交
84

85 86 87 88 89 90 91 92 93 94
/*
 * Note that all tvec_bases is 2 byte aligned and lower bit of
 * base in timer_list is guaranteed to be zero. Use the LSB for
 * the new flag to indicate whether the timer is deferrable
 */
#define TBASE_DEFERRABLE_FLAG		(0x1)

/* Functions below help us manage 'deferrable' flag */
static inline unsigned int tbase_get_deferrable(tvec_base_t *base)
{
95
	return (unsigned int)((unsigned long)base & TBASE_DEFERRABLE_FLAG);
96 97 98 99
}

static inline tvec_base_t *tbase_get_base(tvec_base_t *base)
{
100
	return (tvec_base_t *)((unsigned long)base & ~TBASE_DEFERRABLE_FLAG);
101 102 103 104
}

static inline void timer_set_deferrable(struct timer_list *timer)
{
105 106
	timer->base = (tvec_base_t *)((unsigned long)timer->base |
	                               TBASE_DEFERRABLE_FLAG);
107 108 109 110 111
}

static inline void
timer_set_base(struct timer_list *timer, tvec_base_t *new_base)
{
112
	timer->base = (tvec_base_t *)((unsigned long)new_base |
113 114 115
	                              tbase_get_deferrable(timer->base));
}

116 117 118 119 120
/**
 * __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
 *
121
 * __round_jiffies() rounds an absolute time in the future (in jiffies)
122 123 124 125 126 127 128 129 130 131 132 133
 * 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.
 *
134
 * The return value is the rounded version of the @j parameter.
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
 */
unsigned long __round_jiffies(unsigned long j, int cpu)
{
	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.
	 */
	if (rem < HZ/4) /* round down */
		j = j - rem;
	else /* round up */
		j = j - rem + HZ;

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

	if (j <= jiffies) /* rounding ate our timeout entirely; */
		return original;
	return j;
}
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
 *
178
 * __round_jiffies_relative() rounds a time delta  in the future (in jiffies)
179 180 181 182 183 184 185 186 187 188 189 190
 * 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.
 *
191
 * The return value is the rounded version of the @j parameter.
192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
 */
unsigned long __round_jiffies_relative(unsigned long j, int cpu)
{
	/*
	 * In theory the following code can skip a jiffy in case jiffies
	 * increments right between the addition and the later subtraction.
	 * However since the entire point of this function is to use approximate
	 * timeouts, it's entirely ok to not handle that.
	 */
	return  __round_jiffies(j + jiffies, cpu) - jiffies;
}
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
 *
209
 * round_jiffies() rounds an absolute time in the future (in jiffies)
210 211 212 213 214 215 216 217
 * 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.
 *
218
 * The return value is the rounded version of the @j parameter.
219 220 221 222 223 224 225 226 227 228 229
 */
unsigned long round_jiffies(unsigned long j)
{
	return __round_jiffies(j, raw_smp_processor_id());
}
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
 *
230
 * round_jiffies_relative() rounds a time delta  in the future (in jiffies)
231 232 233 234 235 236 237 238
 * 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.
 *
239
 * The return value is the rounded version of the @j parameter.
240 241 242 243 244 245 246 247
 */
unsigned long round_jiffies_relative(unsigned long j)
{
	return __round_jiffies_relative(j, raw_smp_processor_id());
}
EXPORT_SYMBOL_GPL(round_jiffies_relative);


L
Linus Torvalds 已提交
248 249 250 251
static inline void set_running_timer(tvec_base_t *base,
					struct timer_list *timer)
{
#ifdef CONFIG_SMP
252
	base->running_timer = timer;
L
Linus Torvalds 已提交
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
#endif
}

static void internal_add_timer(tvec_base_t *base, struct timer_list *timer)
{
	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;
		/* If the timeout is larger than 0xffffffff on 64-bit
		 * architectures then we use the maximum timeout:
		 */
		if (idx > 0xffffffffUL) {
			idx = 0xffffffffUL;
			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);
}

298 299 300 301 302 303 304 305 306 307 308 309
#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;
}
#endif

310
/**
311 312 313 314 315 316 317 318 319
 * init_timer - initialize a timer.
 * @timer: the timer to be initialized
 *
 * init_timer() must be done to a timer prior calling *any* of the
 * other timer functions.
 */
void fastcall init_timer(struct timer_list *timer)
{
	timer->entry.next = NULL;
320
	timer->base = __raw_get_cpu_var(tvec_bases);
321 322 323 324 325
#ifdef CONFIG_TIMER_STATS
	timer->start_site = NULL;
	timer->start_pid = -1;
	memset(timer->start_comm, 0, TASK_COMM_LEN);
#endif
326 327 328
}
EXPORT_SYMBOL(init_timer);

329 330 331 332 333 334 335
void fastcall init_timer_deferrable(struct timer_list *timer)
{
	init_timer(timer);
	timer_set_deferrable(timer);
}
EXPORT_SYMBOL(init_timer_deferrable);

336
static inline void detach_timer(struct timer_list *timer,
337
				int clear_pending)
338 339 340 341 342 343 344 345 346 347
{
	struct list_head *entry = &timer->entry;

	__list_del(entry->prev, entry->next);
	if (clear_pending)
		entry->next = NULL;
	entry->prev = LIST_POISON2;
}

/*
348
 * We are using hashed locking: holding per_cpu(tvec_bases).lock
349 350 351 352 353 354 355 356 357 358
 * 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.
 */
359
static tvec_base_t *lock_timer_base(struct timer_list *timer,
360
					unsigned long *flags)
361
	__acquires(timer->base->lock)
362
{
363
	tvec_base_t *base;
364 365

	for (;;) {
366 367
		tvec_base_t *prelock_base = timer->base;
		base = tbase_get_base(prelock_base);
368 369
		if (likely(base != NULL)) {
			spin_lock_irqsave(&base->lock, *flags);
370
			if (likely(prelock_base == timer->base))
371 372 373 374 375 376 377 378
				return base;
			/* The timer has migrated to another CPU */
			spin_unlock_irqrestore(&base->lock, *flags);
		}
		cpu_relax();
	}
}

L
Linus Torvalds 已提交
379 380
int __mod_timer(struct timer_list *timer, unsigned long expires)
{
381
	tvec_base_t *base, *new_base;
L
Linus Torvalds 已提交
382 383 384
	unsigned long flags;
	int ret = 0;

385
	timer_stats_timer_set_start_info(timer);
L
Linus Torvalds 已提交
386 387
	BUG_ON(!timer->function);

388 389 390 391 392 393 394
	base = lock_timer_base(timer, &flags);

	if (timer_pending(timer)) {
		detach_timer(timer, 0);
		ret = 1;
	}

395
	new_base = __get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
396

397
	if (base != new_base) {
L
Linus Torvalds 已提交
398
		/*
399 400 401 402 403
		 * 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
404
		 */
405
		if (likely(base->running_timer != timer)) {
406
			/* See the comment in lock_timer_base() */
407
			timer_set_base(timer, NULL);
408
			spin_unlock(&base->lock);
409 410
			base = new_base;
			spin_lock(&base->lock);
411
			timer_set_base(timer, base);
L
Linus Torvalds 已提交
412 413 414 415
		}
	}

	timer->expires = expires;
416 417
	internal_add_timer(base, timer);
	spin_unlock_irqrestore(&base->lock, flags);
L
Linus Torvalds 已提交
418 419 420 421 422 423

	return ret;
}

EXPORT_SYMBOL(__mod_timer);

424
/**
L
Linus Torvalds 已提交
425 426 427 428 429 430 431 432
 * 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)
{
433
	tvec_base_t *base = per_cpu(tvec_bases, cpu);
L
Linus Torvalds 已提交
434
  	unsigned long flags;
435

436
	timer_stats_timer_set_start_info(timer);
L
Linus Torvalds 已提交
437
  	BUG_ON(timer_pending(timer) || !timer->function);
438
	spin_lock_irqsave(&base->lock, flags);
439
	timer_set_base(timer, base);
L
Linus Torvalds 已提交
440
	internal_add_timer(base, timer);
441
	spin_unlock_irqrestore(&base->lock, flags);
L
Linus Torvalds 已提交
442 443 444
}


445
/**
L
Linus Torvalds 已提交
446 447
 * mod_timer - modify a timer's timeout
 * @timer: the timer to be modified
448
 * @expires: new timeout in jiffies
L
Linus Torvalds 已提交
449
 *
450
 * mod_timer() is a more efficient way to update the expire field of an
L
Linus Torvalds 已提交
451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468
 * 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)
{
	BUG_ON(!timer->function);

469
	timer_stats_timer_set_start_info(timer);
L
Linus Torvalds 已提交
470 471 472 473 474 475 476 477 478 479 480 481 482
	/*
	 * This is a common optimization triggered by the
	 * networking code - if the timer is re-modified
	 * to be the same thing then just return:
	 */
	if (timer->expires == expires && timer_pending(timer))
		return 1;

	return __mod_timer(timer, expires);
}

EXPORT_SYMBOL(mod_timer);

483
/**
L
Linus Torvalds 已提交
484 485 486 487 488 489 490 491 492 493 494 495
 * 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)
{
496
	tvec_base_t *base;
L
Linus Torvalds 已提交
497
	unsigned long flags;
498
	int ret = 0;
L
Linus Torvalds 已提交
499

500
	timer_stats_timer_clear_start_info(timer);
501 502 503 504 505 506
	if (timer_pending(timer)) {
		base = lock_timer_base(timer, &flags);
		if (timer_pending(timer)) {
			detach_timer(timer, 1);
			ret = 1;
		}
L
Linus Torvalds 已提交
507 508 509
		spin_unlock_irqrestore(&base->lock, flags);
	}

510
	return ret;
L
Linus Torvalds 已提交
511 512 513 514 515
}

EXPORT_SYMBOL(del_timer);

#ifdef CONFIG_SMP
516 517 518 519
/**
 * try_to_del_timer_sync - Try to deactivate a timer
 * @timer: timer do del
 *
520 521 522 523 524 525 526
 * 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.
 *
 * It must not be called from interrupt contexts.
 */
int try_to_del_timer_sync(struct timer_list *timer)
{
527
	tvec_base_t *base;
528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546
	unsigned long flags;
	int ret = -1;

	base = lock_timer_base(timer, &flags);

	if (base->running_timer == timer)
		goto out;

	ret = 0;
	if (timer_pending(timer)) {
		detach_timer(timer, 1);
		ret = 1;
	}
out:
	spin_unlock_irqrestore(&base->lock, flags);

	return ret;
}

547 548
EXPORT_SYMBOL(try_to_del_timer_sync);

549
/**
L
Linus Torvalds 已提交
550 551 552 553 554 555 556
 * 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.
 *
557
 * Synchronization rules: Callers must prevent restarting of the timer,
L
Linus Torvalds 已提交
558 559
 * otherwise this function is meaningless. It must not be called from
 * interrupt contexts. The caller must not hold locks which would prevent
560 561 562
 * 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
563 564 565 566 567
 *
 * The function returns whether it has deactivated a pending timer or not.
 */
int del_timer_sync(struct timer_list *timer)
{
568 569 570 571
	for (;;) {
		int ret = try_to_del_timer_sync(timer);
		if (ret >= 0)
			return ret;
572
		cpu_relax();
573
	}
L
Linus Torvalds 已提交
574 575
}

576
EXPORT_SYMBOL(del_timer_sync);
L
Linus Torvalds 已提交
577 578 579 580 581
#endif

static int cascade(tvec_base_t *base, tvec_t *tv, int index)
{
	/* cascade all the timers from tv up one level */
582 583 584 585
	struct timer_list *timer, *tmp;
	struct list_head tv_list;

	list_replace_init(tv->vec + index, &tv_list);
L
Linus Torvalds 已提交
586 587

	/*
588 589
	 * We are removing _all_ timers from the list, so we
	 * don't have to detach them individually.
L
Linus Torvalds 已提交
590
	 */
591
	list_for_each_entry_safe(timer, tmp, &tv_list, entry) {
592
		BUG_ON(tbase_get_base(timer->base) != base);
593
		internal_add_timer(base, timer);
L
Linus Torvalds 已提交
594 595 596 597 598
	}

	return index;
}

599 600 601
#define INDEX(N) ((base->timer_jiffies >> (TVR_BITS + (N) * TVN_BITS)) & TVN_MASK)

/**
L
Linus Torvalds 已提交
602 603 604 605 606 607 608 609 610 611
 * __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.
 */
static inline void __run_timers(tvec_base_t *base)
{
	struct timer_list *timer;

612
	spin_lock_irq(&base->lock);
L
Linus Torvalds 已提交
613
	while (time_after_eq(jiffies, base->timer_jiffies)) {
614
		struct list_head work_list;
L
Linus Torvalds 已提交
615 616
		struct list_head *head = &work_list;
 		int index = base->timer_jiffies & TVR_MASK;
617

L
Linus Torvalds 已提交
618 619 620 621 622 623 624 625
		/*
		 * 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));
626 627
		++base->timer_jiffies;
		list_replace_init(base->tv1.vec + index, &work_list);
628
		while (!list_empty(head)) {
L
Linus Torvalds 已提交
629 630 631
			void (*fn)(unsigned long);
			unsigned long data;

632
			timer = list_first_entry(head, struct timer_list,entry);
L
Linus Torvalds 已提交
633 634 635
 			fn = timer->function;
 			data = timer->data;

636 637
			timer_stats_account_timer(timer);

L
Linus Torvalds 已提交
638
			set_running_timer(base, timer);
639
			detach_timer(timer, 1);
640
			spin_unlock_irq(&base->lock);
L
Linus Torvalds 已提交
641
			{
642
				int preempt_count = preempt_count();
L
Linus Torvalds 已提交
643 644
				fn(data);
				if (preempt_count != preempt_count()) {
645 646 647 648 649
					printk(KERN_WARNING "huh, entered %p "
					       "with preempt_count %08x, exited"
					       " with %08x?\n",
					       fn, preempt_count,
					       preempt_count());
L
Linus Torvalds 已提交
650 651 652
					BUG();
				}
			}
653
			spin_lock_irq(&base->lock);
L
Linus Torvalds 已提交
654 655 656
		}
	}
	set_running_timer(base, NULL);
657
	spin_unlock_irq(&base->lock);
L
Linus Torvalds 已提交
658 659
}

660
#if defined(CONFIG_NO_IDLE_HZ) || defined(CONFIG_NO_HZ)
L
Linus Torvalds 已提交
661 662 663 664 665
/*
 * Find out when the next timer event is due to happen. This
 * is used on S/390 to stop all activity when a cpus is idle.
 * This functions needs to be called disabled.
 */
666
static unsigned long __next_timer_interrupt(tvec_base_t *base)
L
Linus Torvalds 已提交
667
{
668 669 670
	unsigned long timer_jiffies = base->timer_jiffies;
	unsigned long expires = timer_jiffies + (LONG_MAX >> 1);
	int index, slot, array, found = 0;
L
Linus Torvalds 已提交
671 672 673 674
	struct timer_list *nte;
	tvec_t *varray[4];

	/* Look for timer events in tv1. */
675
	index = slot = timer_jiffies & TVR_MASK;
L
Linus Torvalds 已提交
676
	do {
677
		list_for_each_entry(nte, base->tv1.vec + slot, entry) {
678 679 680
 			if (tbase_get_deferrable(nte->base))
 				continue;

681
			found = 1;
L
Linus Torvalds 已提交
682
			expires = nte->expires;
683 684 685 686
			/* Look at the cascade bucket(s)? */
			if (!index || slot < index)
				goto cascade;
			return expires;
L
Linus Torvalds 已提交
687
		}
688 689 690 691 692 693 694 695
		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
696 697 698 699 700 701

	/* Check tv2-tv5. */
	varray[0] = &base->tv2;
	varray[1] = &base->tv3;
	varray[2] = &base->tv4;
	varray[3] = &base->tv5;
702 703 704 705 706

	for (array = 0; array < 4; array++) {
		tvec_t *varp = varray[array];

		index = slot = timer_jiffies & TVN_MASK;
L
Linus Torvalds 已提交
707
		do {
708 709
			list_for_each_entry(nte, varp->vec + slot, entry) {
				found = 1;
L
Linus Torvalds 已提交
710 711
				if (time_before(nte->expires, expires))
					expires = nte->expires;
712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
			}
			/*
			 * 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
729
	}
730 731
	return expires;
}
732

733 734 735 736 737 738 739 740 741
/*
 * 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;
742
	unsigned long delta;
743 744 745

	if (hr_delta.tv64 == KTIME_MAX)
		return expires;
746

747 748 749 750 751
	/*
	 * Expired timer available, let it expire in the next tick
	 */
	if (hr_delta.tv64 <= 0)
		return now + 1;
752

753
	tsdelta = ktime_to_timespec(hr_delta);
754 755 756 757 758 759 760 761 762 763
	delta = timespec_to_jiffies(&tsdelta);
	/*
	 * 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;
764 765
	if (time_before(now, expires))
		return now;
L
Linus Torvalds 已提交
766 767
	return expires;
}
768 769 770

/**
 * next_timer_interrupt - return the jiffy of the next pending timer
771
 * @now: current time (in jiffies)
772
 */
773
unsigned long get_next_timer_interrupt(unsigned long now)
774 775
{
	tvec_base_t *base = __get_cpu_var(tvec_bases);
776
	unsigned long expires;
777 778 779 780 781 782 783 784 785 786

	spin_lock(&base->lock);
	expires = __next_timer_interrupt(base);
	spin_unlock(&base->lock);

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

	return cmp_next_hrtimer_event(now, expires);
}
787 788 789 790 791 792 793 794

#ifdef CONFIG_NO_IDLE_HZ
unsigned long next_timer_interrupt(void)
{
	return get_next_timer_interrupt(jiffies);
}
#endif

L
Linus Torvalds 已提交
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
#endif

/*
 * Called from the timer interrupt handler to charge one tick to the current 
 * 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. */
	if (user_tick)
		account_user_time(p, jiffies_to_cputime(1));
	else
		account_system_time(p, HARDIRQ_OFFSET, jiffies_to_cputime(1));
	run_local_timers();
	if (rcu_pending(cpu))
		rcu_check_callbacks(cpu, user_tick);
	scheduler_tick();
 	run_posix_cpu_timers(p);
}

/*
 * Nr of active tasks - counted in fixed-point numbers
 */
static unsigned long count_active_tasks(void)
{
823
	return nr_active() * FIXED_1;
L
Linus Torvalds 已提交
824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846
}

/*
 * Hmm.. Changed this, as the GNU make sources (load.c) seems to
 * imply that avenrun[] is the standard name for this kind of thing.
 * Nothing else seems to be standardized: the fractional size etc
 * all seem to differ on different machines.
 *
 * Requires xtime_lock to access.
 */
unsigned long avenrun[3];

EXPORT_SYMBOL(avenrun);

/*
 * calc_load - given tick count, update the avenrun load estimates.
 * This is called while holding a write_lock on xtime_lock.
 */
static inline void calc_load(unsigned long ticks)
{
	unsigned long active_tasks; /* fixed-point */
	static int count = LOAD_FREQ;

E
Eric Dumazet 已提交
847 848 849 850 851 852 853 854 855
	count -= ticks;
	if (unlikely(count < 0)) {
		active_tasks = count_active_tasks();
		do {
			CALC_LOAD(avenrun[0], EXP_1, active_tasks);
			CALC_LOAD(avenrun[1], EXP_5, active_tasks);
			CALC_LOAD(avenrun[2], EXP_15, active_tasks);
			count += LOAD_FREQ;
		} while (count < 0);
L
Linus Torvalds 已提交
856 857 858 859 860 861 862 863
	}
}

/*
 * This function runs timers and the timer-tq in bottom half context.
 */
static void run_timer_softirq(struct softirq_action *h)
{
864
	tvec_base_t *base = __get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
865

866 867
	hrtimer_run_queues();

L
Linus Torvalds 已提交
868 869 870 871 872 873 874 875 876 877
	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)
{
	raise_softirq(TIMER_SOFTIRQ);
878
	softlockup_tick();
L
Linus Torvalds 已提交
879 880 881 882 883 884
}

/*
 * Called by the timer interrupt. xtime_lock must already be taken
 * by the timer IRQ!
 */
885
static inline void update_times(unsigned long ticks)
L
Linus Torvalds 已提交
886
{
887
	update_wall_time();
L
Linus Torvalds 已提交
888 889 890 891 892 893 894 895 896
	calc_load(ticks);
}
  
/*
 * The 64-bit jiffies value is not atomic - you MUST NOT read it
 * without sampling the sequence number in xtime_lock.
 * jiffies is defined in the linker script...
 */

897
void do_timer(unsigned long ticks)
L
Linus Torvalds 已提交
898
{
899 900
	jiffies_64 += ticks;
	update_times(ticks);
L
Linus Torvalds 已提交
901 902 903 904 905 906 907 908 909 910
}

#ifdef __ARCH_WANT_SYS_ALARM

/*
 * For backwards compatibility?  This can be done in libc so Alpha
 * and all newer ports shouldn't need it.
 */
asmlinkage unsigned long sys_alarm(unsigned int seconds)
{
911
	return alarm_setitimer(seconds);
L
Linus Torvalds 已提交
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
}

#endif

#ifndef __alpha__

/*
 * The Alpha uses getxpid, getxuid, and getxgid instead.  Maybe this
 * should be moved into arch/i386 instead?
 */

/**
 * sys_getpid - return the thread group id of the current process
 *
 * Note, despite the name, this returns the tgid not the pid.  The tgid and
 * the pid are identical unless CLONE_THREAD was specified on clone() in
 * which case the tgid is the same in all threads of the same group.
 *
 * This is SMP safe as current->tgid does not change.
 */
asmlinkage long sys_getpid(void)
{
	return current->tgid;
}

/*
938 939 940 941
 * Accessing ->real_parent is not SMP-safe, it could
 * change from under us. However, we can use a stale
 * value of ->real_parent under rcu_read_lock(), see
 * release_task()->call_rcu(delayed_put_task_struct).
L
Linus Torvalds 已提交
942 943 944 945 946
 */
asmlinkage long sys_getppid(void)
{
	int pid;

947 948 949
	rcu_read_lock();
	pid = rcu_dereference(current->real_parent)->tgid;
	rcu_read_unlock();
L
Linus Torvalds 已提交
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

	return pid;
}

asmlinkage long sys_getuid(void)
{
	/* Only we change this so SMP safe */
	return current->uid;
}

asmlinkage long sys_geteuid(void)
{
	/* Only we change this so SMP safe */
	return current->euid;
}

asmlinkage long sys_getgid(void)
{
	/* Only we change this so SMP safe */
	return current->gid;
}

asmlinkage long sys_getegid(void)
{
	/* Only we change this so SMP safe */
	return  current->egid;
}

#endif

static void process_timeout(unsigned long __data)
{
982
	wake_up_process((struct task_struct *)__data);
L
Linus Torvalds 已提交
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
}

/**
 * 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.
 */
fastcall signed long __sched schedule_timeout(signed long timeout)
{
	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.
		 */
1036
		if (timeout < 0) {
L
Linus Torvalds 已提交
1037
			printk(KERN_ERR "schedule_timeout: wrong timeout "
1038 1039
				"value %lx\n", timeout);
			dump_stack();
L
Linus Torvalds 已提交
1040 1041 1042 1043 1044 1045 1046
			current->state = TASK_RUNNING;
			goto out;
		}
	}

	expire = timeout + jiffies;

1047 1048
	setup_timer(&timer, process_timeout, (unsigned long)current);
	__mod_timer(&timer, expire);
L
Linus Torvalds 已提交
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
	schedule();
	del_singleshot_timer_sync(&timer);

	timeout = expire - jiffies;

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

1059 1060 1061 1062
/*
 * We can use __set_current_state() here because schedule_timeout() calls
 * schedule() unconditionally.
 */
1063 1064
signed long __sched schedule_timeout_interruptible(signed long timeout)
{
A
Andrew Morton 已提交
1065 1066
	__set_current_state(TASK_INTERRUPTIBLE);
	return schedule_timeout(timeout);
1067 1068 1069 1070 1071
}
EXPORT_SYMBOL(schedule_timeout_interruptible);

signed long __sched schedule_timeout_uninterruptible(signed long timeout)
{
A
Andrew Morton 已提交
1072 1073
	__set_current_state(TASK_UNINTERRUPTIBLE);
	return schedule_timeout(timeout);
1074 1075 1076
}
EXPORT_SYMBOL(schedule_timeout_uninterruptible);

L
Linus Torvalds 已提交
1077 1078 1079 1080 1081 1082
/* Thread ID - the internal kernel "pid" */
asmlinkage long sys_gettid(void)
{
	return current->pid;
}

1083
/**
1084
 * do_sysinfo - fill in sysinfo struct
1085
 * @info: pointer to buffer to fill
L
Linus Torvalds 已提交
1086
 */ 
1087
int do_sysinfo(struct sysinfo *info)
L
Linus Torvalds 已提交
1088 1089 1090 1091 1092
{
	unsigned long mem_total, sav_total;
	unsigned int mem_unit, bitcount;
	unsigned long seq;

1093
	memset(info, 0, sizeof(struct sysinfo));
L
Linus Torvalds 已提交
1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112

	do {
		struct timespec tp;
		seq = read_seqbegin(&xtime_lock);

		/*
		 * This is annoying.  The below is the same thing
		 * posix_get_clock_monotonic() does, but it wants to
		 * take the lock which we want to cover the loads stuff
		 * too.
		 */

		getnstimeofday(&tp);
		tp.tv_sec += wall_to_monotonic.tv_sec;
		tp.tv_nsec += wall_to_monotonic.tv_nsec;
		if (tp.tv_nsec - NSEC_PER_SEC >= 0) {
			tp.tv_nsec = tp.tv_nsec - NSEC_PER_SEC;
			tp.tv_sec++;
		}
1113
		info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0);
L
Linus Torvalds 已提交
1114

1115 1116 1117
		info->loads[0] = avenrun[0] << (SI_LOAD_SHIFT - FSHIFT);
		info->loads[1] = avenrun[1] << (SI_LOAD_SHIFT - FSHIFT);
		info->loads[2] = avenrun[2] << (SI_LOAD_SHIFT - FSHIFT);
L
Linus Torvalds 已提交
1118

1119
		info->procs = nr_threads;
L
Linus Torvalds 已提交
1120 1121
	} while (read_seqretry(&xtime_lock, seq));

1122 1123
	si_meminfo(info);
	si_swapinfo(info);
L
Linus Torvalds 已提交
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133

	/*
	 * If the sum of all the available memory (i.e. ram + swap)
	 * is less than can be stored in a 32 bit unsigned long then
	 * we can be binary compatible with 2.2.x kernels.  If not,
	 * well, in that case 2.2.x was broken anyways...
	 *
	 *  -Erik Andersen <andersee@debian.org>
	 */

1134 1135
	mem_total = info->totalram + info->totalswap;
	if (mem_total < info->totalram || mem_total < info->totalswap)
L
Linus Torvalds 已提交
1136 1137
		goto out;
	bitcount = 0;
1138
	mem_unit = info->mem_unit;
L
Linus Torvalds 已提交
1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149
	while (mem_unit > 1) {
		bitcount++;
		mem_unit >>= 1;
		sav_total = mem_total;
		mem_total <<= 1;
		if (mem_total < sav_total)
			goto out;
	}

	/*
	 * If mem_total did not overflow, multiply all memory values by
1150
	 * info->mem_unit and set it to 1.  This leaves things compatible
L
Linus Torvalds 已提交
1151 1152 1153 1154
	 * with 2.2.x, and also retains compatibility with earlier 2.4.x
	 * kernels...
	 */

1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173
	info->mem_unit = 1;
	info->totalram <<= bitcount;
	info->freeram <<= bitcount;
	info->sharedram <<= bitcount;
	info->bufferram <<= bitcount;
	info->totalswap <<= bitcount;
	info->freeswap <<= bitcount;
	info->totalhigh <<= bitcount;
	info->freehigh <<= bitcount;

out:
	return 0;
}

asmlinkage long sys_sysinfo(struct sysinfo __user *info)
{
	struct sysinfo val;

	do_sysinfo(&val);
L
Linus Torvalds 已提交
1174 1175 1176 1177 1178 1179 1180

	if (copy_to_user(info, &val, sizeof(struct sysinfo)))
		return -EFAULT;

	return 0;
}

1181 1182 1183 1184 1185 1186 1187
/*
 * lockdep: we want to track each per-CPU base as a separate lock-class,
 * but timer-bases are kmalloc()-ed, so we need to attach separate
 * keys to them:
 */
static struct lock_class_key base_lock_keys[NR_CPUS];

1188
static int __devinit init_timers_cpu(int cpu)
L
Linus Torvalds 已提交
1189 1190 1191
{
	int j;
	tvec_base_t *base;
A
Andrew Morton 已提交
1192
	static char __devinitdata tvec_base_done[NR_CPUS];
1193

A
Andrew Morton 已提交
1194
	if (!tvec_base_done[cpu]) {
1195 1196 1197
		static char boot_done;

		if (boot_done) {
A
Andrew Morton 已提交
1198 1199 1200
			/*
			 * The APs use this path later in boot
			 */
1201 1202 1203 1204
			base = kmalloc_node(sizeof(*base), GFP_KERNEL,
						cpu_to_node(cpu));
			if (!base)
				return -ENOMEM;
1205 1206 1207 1208 1209 1210 1211

			/* Make sure that tvec_base is 2 byte aligned */
			if (tbase_get_deferrable(base)) {
				WARN_ON(1);
				kfree(base);
				return -ENOMEM;
			}
1212
			memset(base, 0, sizeof(*base));
A
Andrew Morton 已提交
1213
			per_cpu(tvec_bases, cpu) = base;
1214
		} else {
A
Andrew Morton 已提交
1215 1216 1217 1218 1219 1220
			/*
			 * 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.
			 */
1221
			boot_done = 1;
A
Andrew Morton 已提交
1222
			base = &boot_tvec_bases;
1223
		}
A
Andrew Morton 已提交
1224 1225 1226
		tvec_base_done[cpu] = 1;
	} else {
		base = per_cpu(tvec_bases, cpu);
1227
	}
A
Andrew Morton 已提交
1228

1229
	spin_lock_init(&base->lock);
1230 1231
	lockdep_set_class(&base->lock, base_lock_keys + cpu);

L
Linus Torvalds 已提交
1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
	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;
1242
	return 0;
L
Linus Torvalds 已提交
1243 1244 1245
}

#ifdef CONFIG_HOTPLUG_CPU
1246
static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
L
Linus Torvalds 已提交
1247 1248 1249 1250
{
	struct timer_list *timer;

	while (!list_empty(head)) {
1251
		timer = list_first_entry(head, struct timer_list, entry);
1252
		detach_timer(timer, 0);
1253
		timer_set_base(timer, new_base);
L
Linus Torvalds 已提交
1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
		internal_add_timer(new_base, timer);
	}
}

static void __devinit migrate_timers(int cpu)
{
	tvec_base_t *old_base;
	tvec_base_t *new_base;
	int i;

	BUG_ON(cpu_online(cpu));
1265 1266
	old_base = per_cpu(tvec_bases, cpu);
	new_base = get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
1267 1268

	local_irq_disable();
1269 1270
	double_spin_lock(&new_base->lock, &old_base->lock,
			 smp_processor_id() < cpu);
1271 1272

	BUG_ON(old_base->running_timer);
L
Linus Torvalds 已提交
1273 1274

	for (i = 0; i < TVR_SIZE; i++)
1275 1276 1277 1278 1279 1280 1281 1282
		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);
	}

1283 1284
	double_spin_unlock(&new_base->lock, &old_base->lock,
			   smp_processor_id() < cpu);
L
Linus Torvalds 已提交
1285 1286 1287 1288 1289
	local_irq_enable();
	put_cpu_var(tvec_bases);
}
#endif /* CONFIG_HOTPLUG_CPU */

1290
static int __cpuinit timer_cpu_notify(struct notifier_block *self,
L
Linus Torvalds 已提交
1291 1292 1293 1294 1295
				unsigned long action, void *hcpu)
{
	long cpu = (long)hcpu;
	switch(action) {
	case CPU_UP_PREPARE:
1296 1297
		if (init_timers_cpu(cpu) < 0)
			return NOTIFY_BAD;
L
Linus Torvalds 已提交
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_DEAD:
		migrate_timers(cpu);
		break;
#endif
	default:
		break;
	}
	return NOTIFY_OK;
}

1310
static struct notifier_block __cpuinitdata timers_nb = {
L
Linus Torvalds 已提交
1311 1312 1313 1314 1315 1316
	.notifier_call	= timer_cpu_notify,
};


void __init init_timers(void)
{
1317
	int err = timer_cpu_notify(&timers_nb, (unsigned long)CPU_UP_PREPARE,
L
Linus Torvalds 已提交
1318
				(void *)(long)smp_processor_id());
1319

1320 1321
	init_timer_stats();

1322
	BUG_ON(err == NOTIFY_BAD);
L
Linus Torvalds 已提交
1323 1324 1325 1326 1327 1328
	register_cpu_notifier(&timers_nb);
	open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL);
}

#ifdef CONFIG_TIME_INTERPOLATION

1329 1330
struct time_interpolator *time_interpolator __read_mostly;
static struct time_interpolator *time_interpolator_list __read_mostly;
L
Linus Torvalds 已提交
1331 1332
static DEFINE_SPINLOCK(time_interpolator_lock);

1333
static inline cycles_t time_interpolator_get_cycles(unsigned int src)
L
Linus Torvalds 已提交
1334 1335 1336 1337 1338 1339 1340 1341 1342 1343
{
	unsigned long (*x)(void);

	switch (src)
	{
		case TIME_SOURCE_FUNCTION:
			x = time_interpolator->addr;
			return x();

		case TIME_SOURCE_MMIO64	:
1344
			return readq_relaxed((void __iomem *)time_interpolator->addr);
L
Linus Torvalds 已提交
1345 1346

		case TIME_SOURCE_MMIO32	:
1347
			return readl_relaxed((void __iomem *)time_interpolator->addr);
L
Linus Torvalds 已提交
1348 1349 1350 1351 1352

		default: return get_cycles();
	}
}

1353
static inline u64 time_interpolator_get_counter(int writelock)
L
Linus Torvalds 已提交
1354 1355 1356 1357 1358
{
	unsigned int src = time_interpolator->source;

	if (time_interpolator->jitter)
	{
1359 1360
		cycles_t lcycle;
		cycles_t now;
L
Linus Torvalds 已提交
1361 1362 1363 1364 1365 1366

		do {
			lcycle = time_interpolator->last_cycle;
			now = time_interpolator_get_cycles(src);
			if (lcycle && time_after(lcycle, now))
				return lcycle;
1367 1368 1369 1370 1371 1372 1373 1374 1375

			/* When holding the xtime write lock, there's no need
			 * to add the overhead of the cmpxchg.  Readers are
			 * force to retry until the write lock is released.
			 */
			if (writelock) {
				time_interpolator->last_cycle = now;
				return now;
			}
L
Linus Torvalds 已提交
1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388
			/* Keep track of the last timer value returned. The use of cmpxchg here
			 * will cause contention in an SMP environment.
			 */
		} while (unlikely(cmpxchg(&time_interpolator->last_cycle, lcycle, now) != lcycle));
		return now;
	}
	else
		return time_interpolator_get_cycles(src);
}

void time_interpolator_reset(void)
{
	time_interpolator->offset = 0;
1389
	time_interpolator->last_counter = time_interpolator_get_counter(1);
L
Linus Torvalds 已提交
1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400
}

#define GET_TI_NSECS(count,i) (((((count) - i->last_counter) & (i)->mask) * (i)->nsec_per_cyc) >> (i)->shift)

unsigned long time_interpolator_get_offset(void)
{
	/* If we do not have a time interpolator set up then just return zero */
	if (!time_interpolator)
		return 0;

	return time_interpolator->offset +
1401
		GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator);
L
Linus Torvalds 已提交
1402 1403 1404 1405 1406
}

#define INTERPOLATOR_ADJUST 65536
#define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST

1407
void time_interpolator_update(long delta_nsec)
L
Linus Torvalds 已提交
1408 1409 1410 1411 1412 1413 1414 1415
{
	u64 counter;
	unsigned long offset;

	/* If there is no time interpolator set up then do nothing */
	if (!time_interpolator)
		return;

A
Andrew Morton 已提交
1416 1417 1418 1419 1420 1421 1422 1423
	/*
	 * The interpolator compensates for late ticks by accumulating the late
	 * time in time_interpolator->offset. A tick earlier than expected will
	 * lead to a reset of the offset and a corresponding jump of the clock
	 * forward. Again this only works if the interpolator clock is running
	 * slightly slower than the regular clock and the tuning logic insures
	 * that.
	 */
L
Linus Torvalds 已提交
1424

1425
	counter = time_interpolator_get_counter(1);
A
Andrew Morton 已提交
1426 1427
	offset = time_interpolator->offset +
			GET_TI_NSECS(counter, time_interpolator);
L
Linus Torvalds 已提交
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443

	if (delta_nsec < 0 || (unsigned long) delta_nsec < offset)
		time_interpolator->offset = offset - delta_nsec;
	else {
		time_interpolator->skips++;
		time_interpolator->ns_skipped += delta_nsec - offset;
		time_interpolator->offset = 0;
	}
	time_interpolator->last_counter = counter;

	/* Tuning logic for time interpolator invoked every minute or so.
	 * Decrease interpolator clock speed if no skips occurred and an offset is carried.
	 * Increase interpolator clock speed if we skip too much time.
	 */
	if (jiffies % INTERPOLATOR_ADJUST == 0)
	{
1444
		if (time_interpolator->skips == 0 && time_interpolator->offset > tick_nsec)
L
Linus Torvalds 已提交
1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
			time_interpolator->nsec_per_cyc--;
		if (time_interpolator->ns_skipped > INTERPOLATOR_MAX_SKIP && time_interpolator->offset == 0)
			time_interpolator->nsec_per_cyc++;
		time_interpolator->skips = 0;
		time_interpolator->ns_skipped = 0;
	}
}

static inline int
is_better_time_interpolator(struct time_interpolator *new)
{
	if (!time_interpolator)
		return 1;
	return new->frequency > 2*time_interpolator->frequency ||
	    (unsigned long)new->drift < (unsigned long)time_interpolator->drift;
}

void
register_time_interpolator(struct time_interpolator *ti)
{
	unsigned long flags;

	/* Sanity check */
1468
	BUG_ON(ti->frequency == 0 || ti->mask == 0);
L
Linus Torvalds 已提交
1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522

	ti->nsec_per_cyc = ((u64)NSEC_PER_SEC << ti->shift) / ti->frequency;
	spin_lock(&time_interpolator_lock);
	write_seqlock_irqsave(&xtime_lock, flags);
	if (is_better_time_interpolator(ti)) {
		time_interpolator = ti;
		time_interpolator_reset();
	}
	write_sequnlock_irqrestore(&xtime_lock, flags);

	ti->next = time_interpolator_list;
	time_interpolator_list = ti;
	spin_unlock(&time_interpolator_lock);
}

void
unregister_time_interpolator(struct time_interpolator *ti)
{
	struct time_interpolator *curr, **prev;
	unsigned long flags;

	spin_lock(&time_interpolator_lock);
	prev = &time_interpolator_list;
	for (curr = *prev; curr; curr = curr->next) {
		if (curr == ti) {
			*prev = curr->next;
			break;
		}
		prev = &curr->next;
	}

	write_seqlock_irqsave(&xtime_lock, flags);
	if (ti == time_interpolator) {
		/* we lost the best time-interpolator: */
		time_interpolator = NULL;
		/* find the next-best interpolator */
		for (curr = time_interpolator_list; curr; curr = curr->next)
			if (is_better_time_interpolator(curr))
				time_interpolator = curr;
		time_interpolator_reset();
	}
	write_sequnlock_irqrestore(&xtime_lock, flags);
	spin_unlock(&time_interpolator_lock);
}
#endif /* CONFIG_TIME_INTERPOLATION */

/**
 * 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;

1523 1524
	while (timeout)
		timeout = schedule_timeout_uninterruptible(timeout);
L
Linus Torvalds 已提交
1525 1526 1527 1528 1529
}

EXPORT_SYMBOL(msleep);

/**
1530
 * msleep_interruptible - sleep waiting for signals
L
Linus Torvalds 已提交
1531 1532 1533 1534 1535 1536
 * @msecs: Time in milliseconds to sleep for
 */
unsigned long msleep_interruptible(unsigned int msecs)
{
	unsigned long timeout = msecs_to_jiffies(msecs) + 1;

1537 1538
	while (timeout && !signal_pending(current))
		timeout = schedule_timeout_interruptible(timeout);
L
Linus Torvalds 已提交
1539 1540 1541 1542
	return jiffies_to_msecs(timeout);
}

EXPORT_SYMBOL(msleep_interruptible);