timer.c 49.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
/*
 *  linux/kernel/timer.c
 *
 *  Kernel internal timers, kernel timekeeping, basic process system calls
 *
 *  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>
L
Linus Torvalds 已提交
37 38 39 40 41 42 43

#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
44 45 46 47
u64 jiffies_64 __cacheline_aligned_in_smp = INITIAL_JIFFIES;

EXPORT_SYMBOL(jiffies_64);

L
Linus Torvalds 已提交
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
/*
 * 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 {
67 68
	spinlock_t lock;
	struct timer_list *running_timer;
L
Linus Torvalds 已提交
69 70 71 72 73 74 75 76 77
	unsigned long timer_jiffies;
	tvec_root_t tv1;
	tvec_t tv2;
	tvec_t tv3;
	tvec_t tv4;
	tvec_t tv5;
} ____cacheline_aligned_in_smp;

typedef struct tvec_t_base_s tvec_base_t;
A
Andrew Morton 已提交
78

79 80
tvec_base_t boot_tvec_bases;
EXPORT_SYMBOL(boot_tvec_bases);
81
static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases;
L
Linus Torvalds 已提交
82

83 84 85 86 87
/**
 * __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
 *
88
 * __round_jiffies() rounds an absolute time in the future (in jiffies)
89 90 91 92 93 94 95 96 97 98 99 100
 * 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.
 *
101
 * The return value is the rounded version of the @j parameter.
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
 */
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
 *
145
 * __round_jiffies_relative() rounds a time delta  in the future (in jiffies)
146 147 148 149 150 151 152 153 154 155 156 157
 * 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.
 *
158
 * The return value is the rounded version of the @j parameter.
159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
 */
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
 *
176
 * round_jiffies() rounds an absolute time in the future (in jiffies)
177 178 179 180 181 182 183 184
 * 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.
 *
185
 * The return value is the rounded version of the @j parameter.
186 187 188 189 190 191 192 193 194 195 196
 */
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
 *
197
 * round_jiffies_relative() rounds a time delta  in the future (in jiffies)
198 199 200 201 202 203 204 205
 * 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.
 *
206
 * The return value is the rounded version of the @j parameter.
207 208 209 210 211 212 213 214
 */
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
215 216 217 218
static inline void set_running_timer(tvec_base_t *base,
					struct timer_list *timer)
{
#ifdef CONFIG_SMP
219
	base->running_timer = timer;
L
Linus Torvalds 已提交
220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
#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);
}

265
/**
266 267 268 269 270 271 272 273 274
 * 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;
275
	timer->base = __raw_get_cpu_var(tvec_bases);
276 277 278 279 280 281 282 283 284 285 286 287 288 289 290
}
EXPORT_SYMBOL(init_timer);

static inline void detach_timer(struct timer_list *timer,
					int clear_pending)
{
	struct list_head *entry = &timer->entry;

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

/*
291
 * We are using hashed locking: holding per_cpu(tvec_bases).lock
292 293 294 295 296 297 298 299 300 301
 * 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.
 */
302
static tvec_base_t *lock_timer_base(struct timer_list *timer,
303
					unsigned long *flags)
304
	__acquires(timer->base->lock)
305
{
306
	tvec_base_t *base;
307 308 309 310 311 312 313 314 315 316 317 318 319 320

	for (;;) {
		base = timer->base;
		if (likely(base != NULL)) {
			spin_lock_irqsave(&base->lock, *flags);
			if (likely(base == timer->base))
				return base;
			/* The timer has migrated to another CPU */
			spin_unlock_irqrestore(&base->lock, *flags);
		}
		cpu_relax();
	}
}

L
Linus Torvalds 已提交
321 322
int __mod_timer(struct timer_list *timer, unsigned long expires)
{
323
	tvec_base_t *base, *new_base;
L
Linus Torvalds 已提交
324 325 326 327 328
	unsigned long flags;
	int ret = 0;

	BUG_ON(!timer->function);

329 330 331 332 333 334 335
	base = lock_timer_base(timer, &flags);

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

336
	new_base = __get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
337

338
	if (base != new_base) {
L
Linus Torvalds 已提交
339
		/*
340 341 342 343 344
		 * 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
345
		 */
346
		if (likely(base->running_timer != timer)) {
347 348 349
			/* See the comment in lock_timer_base() */
			timer->base = NULL;
			spin_unlock(&base->lock);
350 351 352
			base = new_base;
			spin_lock(&base->lock);
			timer->base = base;
L
Linus Torvalds 已提交
353 354 355 356
		}
	}

	timer->expires = expires;
357 358
	internal_add_timer(base, timer);
	spin_unlock_irqrestore(&base->lock, flags);
L
Linus Torvalds 已提交
359 360 361 362 363 364

	return ret;
}

EXPORT_SYMBOL(__mod_timer);

365
/**
L
Linus Torvalds 已提交
366 367 368 369 370 371 372 373
 * 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)
{
374
	tvec_base_t *base = per_cpu(tvec_bases, cpu);
L
Linus Torvalds 已提交
375
  	unsigned long flags;
376

L
Linus Torvalds 已提交
377
  	BUG_ON(timer_pending(timer) || !timer->function);
378 379
	spin_lock_irqsave(&base->lock, flags);
	timer->base = base;
L
Linus Torvalds 已提交
380
	internal_add_timer(base, timer);
381
	spin_unlock_irqrestore(&base->lock, flags);
L
Linus Torvalds 已提交
382 383 384
}


385
/**
L
Linus Torvalds 已提交
386 387
 * mod_timer - modify a timer's timeout
 * @timer: the timer to be modified
388
 * @expires: new timeout in jiffies
L
Linus Torvalds 已提交
389
 *
390
 * mod_timer() is a more efficient way to update the expire field of an
L
Linus Torvalds 已提交
391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
 * 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);

	/*
	 * 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);

422
/**
L
Linus Torvalds 已提交
423 424 425 426 427 428 429 430 431 432 433 434
 * 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)
{
435
	tvec_base_t *base;
L
Linus Torvalds 已提交
436
	unsigned long flags;
437
	int ret = 0;
L
Linus Torvalds 已提交
438

439 440 441 442 443 444
	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
445 446 447
		spin_unlock_irqrestore(&base->lock, flags);
	}

448
	return ret;
L
Linus Torvalds 已提交
449 450 451 452 453
}

EXPORT_SYMBOL(del_timer);

#ifdef CONFIG_SMP
454 455 456 457
/**
 * try_to_del_timer_sync - Try to deactivate a timer
 * @timer: timer do del
 *
458 459 460 461 462 463 464
 * 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)
{
465
	tvec_base_t *base;
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484
	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;
}

485
/**
L
Linus Torvalds 已提交
486 487 488 489 490 491 492
 * 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.
 *
493
 * Synchronization rules: Callers must prevent restarting of the timer,
L
Linus Torvalds 已提交
494 495
 * otherwise this function is meaningless. It must not be called from
 * interrupt contexts. The caller must not hold locks which would prevent
496 497 498
 * 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
499 500 501 502 503
 *
 * The function returns whether it has deactivated a pending timer or not.
 */
int del_timer_sync(struct timer_list *timer)
{
504 505 506 507
	for (;;) {
		int ret = try_to_del_timer_sync(timer);
		if (ret >= 0)
			return ret;
508
		cpu_relax();
509
	}
L
Linus Torvalds 已提交
510 511
}

512
EXPORT_SYMBOL(del_timer_sync);
L
Linus Torvalds 已提交
513 514 515 516 517
#endif

static int cascade(tvec_base_t *base, tvec_t *tv, int index)
{
	/* cascade all the timers from tv up one level */
518 519 520 521
	struct timer_list *timer, *tmp;
	struct list_head tv_list;

	list_replace_init(tv->vec + index, &tv_list);
L
Linus Torvalds 已提交
522 523

	/*
524 525
	 * We are removing _all_ timers from the list, so we
	 * don't have to detach them individually.
L
Linus Torvalds 已提交
526
	 */
527 528 529
	list_for_each_entry_safe(timer, tmp, &tv_list, entry) {
		BUG_ON(timer->base != base);
		internal_add_timer(base, timer);
L
Linus Torvalds 已提交
530 531 532 533 534
	}

	return index;
}

535 536 537
#define INDEX(N) ((base->timer_jiffies >> (TVR_BITS + (N) * TVN_BITS)) & TVN_MASK)

/**
L
Linus Torvalds 已提交
538 539 540 541 542 543 544 545 546 547
 * __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;

548
	spin_lock_irq(&base->lock);
L
Linus Torvalds 已提交
549
	while (time_after_eq(jiffies, base->timer_jiffies)) {
550
		struct list_head work_list;
L
Linus Torvalds 已提交
551 552
		struct list_head *head = &work_list;
 		int index = base->timer_jiffies & TVR_MASK;
553

L
Linus Torvalds 已提交
554 555 556 557 558 559 560 561
		/*
		 * 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));
562 563
		++base->timer_jiffies;
		list_replace_init(base->tv1.vec + index, &work_list);
564
		while (!list_empty(head)) {
L
Linus Torvalds 已提交
565 566 567 568 569 570 571 572
			void (*fn)(unsigned long);
			unsigned long data;

			timer = list_entry(head->next,struct timer_list,entry);
 			fn = timer->function;
 			data = timer->data;

			set_running_timer(base, timer);
573
			detach_timer(timer, 1);
574
			spin_unlock_irq(&base->lock);
L
Linus Torvalds 已提交
575
			{
576
				int preempt_count = preempt_count();
L
Linus Torvalds 已提交
577 578
				fn(data);
				if (preempt_count != preempt_count()) {
579 580 581 582 583
					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
584 585 586
					BUG();
				}
			}
587
			spin_lock_irq(&base->lock);
L
Linus Torvalds 已提交
588 589 590
		}
	}
	set_running_timer(base, NULL);
591
	spin_unlock_irq(&base->lock);
L
Linus Torvalds 已提交
592 593 594 595 596 597 598 599 600 601 602 603 604 605
}

#ifdef CONFIG_NO_IDLE_HZ
/*
 * 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.
 */
unsigned long next_timer_interrupt(void)
{
	tvec_base_t *base;
	struct list_head *list;
	struct timer_list *nte;
	unsigned long expires;
606 607
	unsigned long hr_expires = MAX_JIFFY_OFFSET;
	ktime_t hr_delta;
L
Linus Torvalds 已提交
608 609 610
	tvec_t *varray[4];
	int i, j;

611 612 613 614 615 616 617 618 619 620
	hr_delta = hrtimer_get_next_event();
	if (hr_delta.tv64 != KTIME_MAX) {
		struct timespec tsdelta;
		tsdelta = ktime_to_timespec(hr_delta);
		hr_expires = timespec_to_jiffies(&tsdelta);
		if (hr_expires < 3)
			return hr_expires + jiffies;
	}
	hr_expires += jiffies;

621
	base = __get_cpu_var(tvec_bases);
622
	spin_lock(&base->lock);
L
Linus Torvalds 已提交
623
	expires = base->timer_jiffies + (LONG_MAX >> 1);
A
Al Viro 已提交
624
	list = NULL;
L
Linus Torvalds 已提交
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669

	/* Look for timer events in tv1. */
	j = base->timer_jiffies & TVR_MASK;
	do {
		list_for_each_entry(nte, base->tv1.vec + j, entry) {
			expires = nte->expires;
			if (j < (base->timer_jiffies & TVR_MASK))
				list = base->tv2.vec + (INDEX(0));
			goto found;
		}
		j = (j + 1) & TVR_MASK;
	} while (j != (base->timer_jiffies & TVR_MASK));

	/* Check tv2-tv5. */
	varray[0] = &base->tv2;
	varray[1] = &base->tv3;
	varray[2] = &base->tv4;
	varray[3] = &base->tv5;
	for (i = 0; i < 4; i++) {
		j = INDEX(i);
		do {
			if (list_empty(varray[i]->vec + j)) {
				j = (j + 1) & TVN_MASK;
				continue;
			}
			list_for_each_entry(nte, varray[i]->vec + j, entry)
				if (time_before(nte->expires, expires))
					expires = nte->expires;
			if (j < (INDEX(i)) && i < 3)
				list = varray[i + 1]->vec + (INDEX(i + 1));
			goto found;
		} while (j != (INDEX(i)));
	}
found:
	if (list) {
		/*
		 * The search wrapped. We need to look at the next list
		 * from next tv element that would cascade into tv element
		 * where we found the timer element.
		 */
		list_for_each_entry(nte, list, entry) {
			if (time_before(nte->expires, expires))
				expires = nte->expires;
		}
	}
670
	spin_unlock(&base->lock);
671

672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687
	/*
	 * It can happen that other CPUs service timer IRQs and increment
	 * jiffies, but we have not yet got a local timer tick to process
	 * the timer wheels.  In that case, the expiry time can be before
	 * jiffies, but since the high-resolution timer here is relative to
	 * jiffies, the default expression when high-resolution timers are
	 * not active,
	 *
	 *   time_before(MAX_JIFFY_OFFSET + jiffies, expires)
	 *
	 * would falsely evaluate to true.  If that is the case, just
	 * return jiffies so that we can immediately fire the local timer
	 */
	if (time_before(expires, jiffies))
		return jiffies;

688 689 690
	if (time_before(hr_expires, expires))
		return hr_expires;

L
Linus Torvalds 已提交
691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709
	return expires;
}
#endif

/******************************************************************/

/* 
 * The current time 
 * wall_to_monotonic is what we need to add to xtime (or xtime corrected 
 * for sub jiffie times) to get to monotonic time.  Monotonic is pegged
 * at zero at system boot time, so wall_to_monotonic will be negative,
 * however, we will ALWAYS keep the tv_nsec part positive so we can use
 * the usual normalization.
 */
struct timespec xtime __attribute__ ((aligned (16)));
struct timespec wall_to_monotonic __attribute__ ((aligned (16)));

EXPORT_SYMBOL(xtime);

710

711 712 713
/* XXX - all of this timekeeping code should be later moved to time.c */
#include <linux/clocksource.h>
static struct clocksource *clock; /* pointer to current clocksource */
714 715 716 717 718 719 720 721 722 723 724 725 726 727 728

#ifdef CONFIG_GENERIC_TIME
/**
 * __get_nsec_offset - Returns nanoseconds since last call to periodic_hook
 *
 * private function, must hold xtime_lock lock when being
 * called. Returns the number of nanoseconds since the
 * last call to update_wall_time() (adjusted by NTP scaling)
 */
static inline s64 __get_nsec_offset(void)
{
	cycle_t cycle_now, cycle_delta;
	s64 ns_offset;

	/* read clocksource: */
729
	cycle_now = clocksource_read(clock);
730 731

	/* calculate the delta since the last update_wall_time: */
732
	cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763

	/* convert to nanoseconds: */
	ns_offset = cyc2ns(clock, cycle_delta);

	return ns_offset;
}

/**
 * __get_realtime_clock_ts - Returns the time of day in a timespec
 * @ts:		pointer to the timespec to be set
 *
 * Returns the time of day in a timespec. Used by
 * do_gettimeofday() and get_realtime_clock_ts().
 */
static inline void __get_realtime_clock_ts(struct timespec *ts)
{
	unsigned long seq;
	s64 nsecs;

	do {
		seq = read_seqbegin(&xtime_lock);

		*ts = xtime;
		nsecs = __get_nsec_offset();

	} while (read_seqretry(&xtime_lock, seq));

	timespec_add_ns(ts, nsecs);
}

/**
764
 * getnstimeofday - Returns the time of day in a timespec
765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
 * @ts:		pointer to the timespec to be set
 *
 * Returns the time of day in a timespec.
 */
void getnstimeofday(struct timespec *ts)
{
	__get_realtime_clock_ts(ts);
}

EXPORT_SYMBOL(getnstimeofday);

/**
 * do_gettimeofday - Returns the time of day in a timeval
 * @tv:		pointer to the timeval to be set
 *
 * NOTE: Users should be converted to using get_realtime_clock_ts()
 */
void do_gettimeofday(struct timeval *tv)
{
	struct timespec now;

	__get_realtime_clock_ts(&now);
	tv->tv_sec = now.tv_sec;
	tv->tv_usec = now.tv_nsec/1000;
}

EXPORT_SYMBOL(do_gettimeofday);
/**
 * do_settimeofday - Sets the time of day
 * @tv:		pointer to the timespec variable containing the new time
 *
 * Sets the time of day to the new time and update NTP and notify hrtimers
 */
int do_settimeofday(struct timespec *tv)
{
	unsigned long flags;
	time_t wtm_sec, sec = tv->tv_sec;
	long wtm_nsec, nsec = tv->tv_nsec;

	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
		return -EINVAL;

	write_seqlock_irqsave(&xtime_lock, flags);

	nsec -= __get_nsec_offset();

	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);

	set_normalized_timespec(&xtime, sec, nsec);
	set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);

817
	clock->error = 0;
818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839
	ntp_clear();

	write_sequnlock_irqrestore(&xtime_lock, flags);

	/* signal hrtimers about time change */
	clock_was_set();

	return 0;
}

EXPORT_SYMBOL(do_settimeofday);

/**
 * change_clocksource - Swaps clocksources if a new one is available
 *
 * Accumulates current time interval and initializes new clocksource
 */
static int change_clocksource(void)
{
	struct clocksource *new;
	cycle_t now;
	u64 nsec;
840
	new = clocksource_get_next();
841
	if (clock != new) {
842
		now = clocksource_read(new);
843 844 845 846
		nsec =  __get_nsec_offset();
		timespec_add_ns(&xtime, nsec);

		clock = new;
847
		clock->cycle_last = now;
848
		printk(KERN_INFO "Time: %s clocksource has been installed.\n",
D
Daniel Walker 已提交
849
		       clock->name);
850 851 852 853 854
		return 1;
	}
	return 0;
}
#else
D
Daniel Walker 已提交
855 856 857 858
static inline int change_clocksource(void)
{
	return 0;
}
859 860 861 862 863 864 865 866 867 868 869 870 871
#endif

/**
 * timeofday_is_continuous - check to see if timekeeping is free running
 */
int timekeeping_is_continuous(void)
{
	unsigned long seq;
	int ret;

	do {
		seq = read_seqbegin(&xtime_lock);

872
		ret = clock->flags & CLOCK_SOURCE_IS_CONTINUOUS;
873 874 875 876 877 878

	} while (read_seqretry(&xtime_lock, seq));

	return ret;
}

879 880 881 882 883 884 885 886 887 888 889 890 891 892
/**
 * read_persistent_clock -  Return time in seconds from the persistent clock.
 *
 * Weak dummy function for arches that do not yet support it.
 * Returns seconds from epoch using the battery backed persistent clock.
 * Returns zero if unsupported.
 *
 *  XXX - Do be sure to remove it once all arches implement it.
 */
unsigned long __attribute__((weak)) read_persistent_clock(void)
{
	return 0;
}

L
Linus Torvalds 已提交
893
/*
894
 * timekeeping_init - Initializes the clocksource and common timekeeping values
L
Linus Torvalds 已提交
895
 */
896
void __init timekeeping_init(void)
L
Linus Torvalds 已提交
897
{
898
	unsigned long flags;
899
	unsigned long sec = read_persistent_clock();
900 901

	write_seqlock_irqsave(&xtime_lock, flags);
902 903 904

	ntp_clear();

905
	clock = clocksource_get_next();
J
john stultz 已提交
906
	clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH);
907
	clock->cycle_last = clocksource_read(clock);
908

909 910 911 912 913
	xtime.tv_sec = sec;
	xtime.tv_nsec = 0;
	set_normalized_timespec(&wall_to_monotonic,
		-xtime.tv_sec, -xtime.tv_nsec);

914 915 916 917
	write_sequnlock_irqrestore(&xtime_lock, flags);
}


918
/* flag for if timekeeping is suspended */
919
static int timekeeping_suspended;
920 921 922
/* time in seconds when suspend began */
static unsigned long timekeeping_suspend_time;

923
/**
924 925 926 927
 * timekeeping_resume - Resumes the generic timekeeping subsystem.
 * @dev:	unused
 *
 * This is for the generic clocksource timekeeping.
A
Atsushi Nemoto 已提交
928
 * xtime/wall_to_monotonic/jiffies/etc are
929 930 931 932 933
 * still managed by arch specific suspend/resume code.
 */
static int timekeeping_resume(struct sys_device *dev)
{
	unsigned long flags;
934
	unsigned long now = read_persistent_clock();
935 936

	write_seqlock_irqsave(&xtime_lock, flags);
937 938 939 940 941 942 943 944

	if (now && (now > timekeeping_suspend_time)) {
		unsigned long sleep_length = now - timekeeping_suspend_time;

		xtime.tv_sec += sleep_length;
		wall_to_monotonic.tv_sec -= sleep_length;
	}
	/* re-base the last cycle value */
945
	clock->cycle_last = clocksource_read(clock);
946 947 948
	clock->error = 0;
	timekeeping_suspended = 0;
	write_sequnlock_irqrestore(&xtime_lock, flags);
949 950 951 952

	touch_softlockup_watchdog();
	hrtimer_notify_resume();

953 954 955 956 957 958 959 960 961
	return 0;
}

static int timekeeping_suspend(struct sys_device *dev, pm_message_t state)
{
	unsigned long flags;

	write_seqlock_irqsave(&xtime_lock, flags);
	timekeeping_suspended = 1;
962
	timekeeping_suspend_time = read_persistent_clock();
963 964 965 966 967 968 969
	write_sequnlock_irqrestore(&xtime_lock, flags);
	return 0;
}

/* sysfs resume/suspend bits for timekeeping */
static struct sysdev_class timekeeping_sysclass = {
	.resume		= timekeeping_resume,
970
	.suspend	= timekeeping_suspend,
971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988
	set_kset_name("timekeeping"),
};

static struct sys_device device_timer = {
	.id		= 0,
	.cls		= &timekeeping_sysclass,
};

static int __init timekeeping_init_device(void)
{
	int error = sysdev_class_register(&timekeeping_sysclass);
	if (!error)
		error = sysdev_register(&device_timer);
	return error;
}

device_initcall(timekeeping_init_device);

989
/*
990
 * If the error is already larger, we look ahead even further
991 992
 * to compensate for late or lost adjustments.
 */
D
Daniel Walker 已提交
993 994
static __always_inline int clocksource_bigadjust(s64 error, s64 *interval,
						 s64 *offset)
995
{
996 997 998
	s64 tick_error, i;
	u32 look_ahead, adj;
	s32 error2, mult;
999 1000

	/*
1001 1002 1003 1004 1005 1006 1007
	 * Use the current error value to determine how much to look ahead.
	 * The larger the error the slower we adjust for it to avoid problems
	 * with losing too many ticks, otherwise we would overadjust and
	 * produce an even larger error.  The smaller the adjustment the
	 * faster we try to adjust for it, as lost ticks can do less harm
	 * here.  This is tuned so that an error of about 1 msec is adusted
	 * within about 1 sec (or 2^20 nsec in 2^SHIFT_HZ ticks).
1008
	 */
1009 1010 1011 1012
	error2 = clock->error >> (TICK_LENGTH_SHIFT + 22 - 2 * SHIFT_HZ);
	error2 = abs(error2);
	for (look_ahead = 0; error2 > 0; look_ahead++)
		error2 >>= 2;
1013 1014

	/*
1015 1016
	 * Now calculate the error in (1 << look_ahead) ticks, but first
	 * remove the single look ahead already included in the error.
1017
	 */
D
Daniel Walker 已提交
1018 1019
	tick_error = current_tick_length() >>
		(TICK_LENGTH_SHIFT - clock->shift + 1);
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
	tick_error -= clock->xtime_interval >> 1;
	error = ((error - tick_error) >> look_ahead) + tick_error;

	/* Finally calculate the adjustment shift value.  */
	i = *interval;
	mult = 1;
	if (error < 0) {
		error = -error;
		*interval = -*interval;
		*offset = -*offset;
		mult = -1;
1031
	}
1032 1033
	for (adj = 0; error > i; adj++)
		error >>= 1;
1034 1035 1036

	*interval <<= adj;
	*offset <<= adj;
1037
	return mult << adj;
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051
}

/*
 * Adjust the multiplier to reduce the error value,
 * this is optimized for the most common adjustments of -1,0,1,
 * for other values we can do a bit more work.
 */
static void clocksource_adjust(struct clocksource *clock, s64 offset)
{
	s64 error, interval = clock->cycle_interval;
	int adj;

	error = clock->error >> (TICK_LENGTH_SHIFT - clock->shift - 1);
	if (error > interval) {
1052 1053 1054 1055 1056
		error >>= 2;
		if (likely(error <= interval))
			adj = 1;
		else
			adj = clocksource_bigadjust(error, &interval, &offset);
1057
	} else if (error < -interval) {
1058 1059 1060 1061 1062 1063 1064
		error >>= 2;
		if (likely(error >= -interval)) {
			adj = -1;
			interval = -interval;
			offset = -offset;
		} else
			adj = clocksource_bigadjust(error, &interval, &offset);
1065 1066 1067 1068 1069 1070
	} else
		return;

	clock->mult += adj;
	clock->xtime_interval += interval;
	clock->xtime_nsec -= offset;
D
Daniel Walker 已提交
1071 1072
	clock->error -= (interval - offset) <<
			(TICK_LENGTH_SHIFT - clock->shift);
1073 1074
}

1075
/**
1076 1077 1078 1079 1080 1081
 * update_wall_time - Uses the current clocksource to increment the wall time
 *
 * Called from the timer interrupt, must hold a write on xtime_lock.
 */
static void update_wall_time(void)
{
1082
	cycle_t offset;
1083

1084 1085 1086
	/* Make sure we're fully resumed: */
	if (unlikely(timekeeping_suspended))
		return;
1087

1088 1089 1090 1091 1092
#ifdef CONFIG_GENERIC_TIME
	offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask;
#else
	offset = clock->cycle_interval;
#endif
1093
	clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift;
1094 1095 1096 1097

	/* normally this loop will run just once, however in the
	 * case of lost or late ticks, it will accumulate correctly.
	 */
1098
	while (offset >= clock->cycle_interval) {
1099
		/* accumulate one interval */
1100 1101 1102 1103 1104 1105 1106 1107 1108
		clock->xtime_nsec += clock->xtime_interval;
		clock->cycle_last += clock->cycle_interval;
		offset -= clock->cycle_interval;

		if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) {
			clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift;
			xtime.tv_sec++;
			second_overflow();
		}
1109

1110
		/* interpolator bits */
1111
		time_interpolator_update(clock->xtime_interval
1112 1113 1114
						>> clock->shift);

		/* accumulate error between NTP and clock interval */
1115 1116 1117
		clock->error += current_tick_length();
		clock->error -= clock->xtime_interval << (TICK_LENGTH_SHIFT - clock->shift);
	}
1118

1119 1120
	/* correct the clock when NTP error is too big */
	clocksource_adjust(clock, offset);
1121 1122

	/* store full nanoseconds into xtime */
1123
	xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift;
1124
	clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift;
1125 1126 1127

	/* check to see if there is a new clocksource to use */
	if (change_clocksource()) {
1128 1129
		clock->error = 0;
		clock->xtime_nsec = 0;
J
john stultz 已提交
1130
		clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH);
1131
	}
L
Linus Torvalds 已提交
1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159
}

/*
 * 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)
{
1160
	return nr_active() * FIXED_1;
L
Linus Torvalds 已提交
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
}

/*
 * 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
1184 1185 1186 1187 1188 1189 1190 1191 1192
	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
1193 1194 1195 1196 1197 1198 1199
	}
}

/*
 * This read-write spinlock protects us from races in SMP while
 * playing with xtime and avenrun.
 */
1200
__attribute__((weak)) __cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock);
L
Linus Torvalds 已提交
1201 1202 1203 1204 1205 1206 1207 1208

EXPORT_SYMBOL(xtime_lock);

/*
 * This function runs timers and the timer-tq in bottom half context.
 */
static void run_timer_softirq(struct softirq_action *h)
{
1209
	tvec_base_t *base = __get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
1210

1211
 	hrtimer_run_queues();
L
Linus Torvalds 已提交
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221
	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);
1222
	softlockup_tick();
L
Linus Torvalds 已提交
1223 1224 1225 1226 1227 1228
}

/*
 * Called by the timer interrupt. xtime_lock must already be taken
 * by the timer IRQ!
 */
1229
static inline void update_times(unsigned long ticks)
L
Linus Torvalds 已提交
1230
{
1231
	update_wall_time();
L
Linus Torvalds 已提交
1232 1233 1234 1235 1236 1237 1238 1239 1240
	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...
 */

1241
void do_timer(unsigned long ticks)
L
Linus Torvalds 已提交
1242
{
1243 1244
	jiffies_64 += ticks;
	update_times(ticks);
L
Linus Torvalds 已提交
1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
}

#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)
{
1255
	return alarm_setitimer(seconds);
L
Linus Torvalds 已提交
1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281
}

#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;
}

/*
1282 1283 1284 1285
 * 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
1286 1287 1288 1289 1290
 */
asmlinkage long sys_getppid(void)
{
	int pid;

1291 1292 1293
	rcu_read_lock();
	pid = rcu_dereference(current->real_parent)->tgid;
	rcu_read_unlock();
L
Linus Torvalds 已提交
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325

	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)
{
1326
	wake_up_process((struct task_struct *)__data);
L
Linus Torvalds 已提交
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379
}

/**
 * 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.
		 */
1380
		if (timeout < 0) {
L
Linus Torvalds 已提交
1381
			printk(KERN_ERR "schedule_timeout: wrong timeout "
1382 1383
				"value %lx\n", timeout);
			dump_stack();
L
Linus Torvalds 已提交
1384 1385 1386 1387 1388 1389 1390
			current->state = TASK_RUNNING;
			goto out;
		}
	}

	expire = timeout + jiffies;

1391 1392
	setup_timer(&timer, process_timeout, (unsigned long)current);
	__mod_timer(&timer, expire);
L
Linus Torvalds 已提交
1393 1394 1395 1396 1397 1398 1399 1400 1401 1402
	schedule();
	del_singleshot_timer_sync(&timer);

	timeout = expire - jiffies;

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

1403 1404 1405 1406
/*
 * We can use __set_current_state() here because schedule_timeout() calls
 * schedule() unconditionally.
 */
1407 1408
signed long __sched schedule_timeout_interruptible(signed long timeout)
{
A
Andrew Morton 已提交
1409 1410
	__set_current_state(TASK_INTERRUPTIBLE);
	return schedule_timeout(timeout);
1411 1412 1413 1414 1415
}
EXPORT_SYMBOL(schedule_timeout_interruptible);

signed long __sched schedule_timeout_uninterruptible(signed long timeout)
{
A
Andrew Morton 已提交
1416 1417
	__set_current_state(TASK_UNINTERRUPTIBLE);
	return schedule_timeout(timeout);
1418 1419 1420
}
EXPORT_SYMBOL(schedule_timeout_uninterruptible);

L
Linus Torvalds 已提交
1421 1422 1423 1424 1425 1426
/* Thread ID - the internal kernel "pid" */
asmlinkage long sys_gettid(void)
{
	return current->pid;
}

1427
/**
1428
 * do_sysinfo - fill in sysinfo struct
1429
 * @info: pointer to buffer to fill
L
Linus Torvalds 已提交
1430
 */ 
1431
int do_sysinfo(struct sysinfo *info)
L
Linus Torvalds 已提交
1432 1433 1434 1435 1436
{
	unsigned long mem_total, sav_total;
	unsigned int mem_unit, bitcount;
	unsigned long seq;

1437
	memset(info, 0, sizeof(struct sysinfo));
L
Linus Torvalds 已提交
1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456

	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++;
		}
1457
		info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0);
L
Linus Torvalds 已提交
1458

1459 1460 1461
		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
1462

1463
		info->procs = nr_threads;
L
Linus Torvalds 已提交
1464 1465
	} while (read_seqretry(&xtime_lock, seq));

1466 1467
	si_meminfo(info);
	si_swapinfo(info);
L
Linus Torvalds 已提交
1468 1469 1470 1471 1472 1473 1474 1475 1476 1477

	/*
	 * 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>
	 */

1478 1479
	mem_total = info->totalram + info->totalswap;
	if (mem_total < info->totalram || mem_total < info->totalswap)
L
Linus Torvalds 已提交
1480 1481
		goto out;
	bitcount = 0;
1482
	mem_unit = info->mem_unit;
L
Linus Torvalds 已提交
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493
	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
1494
	 * info->mem_unit and set it to 1.  This leaves things compatible
L
Linus Torvalds 已提交
1495 1496 1497 1498
	 * with 2.2.x, and also retains compatibility with earlier 2.4.x
	 * kernels...
	 */

1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517
	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
1518 1519 1520 1521 1522 1523 1524

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

	return 0;
}

1525 1526 1527 1528 1529 1530 1531
/*
 * 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];

1532
static int __devinit init_timers_cpu(int cpu)
L
Linus Torvalds 已提交
1533 1534 1535
{
	int j;
	tvec_base_t *base;
A
Andrew Morton 已提交
1536
	static char __devinitdata tvec_base_done[NR_CPUS];
1537

A
Andrew Morton 已提交
1538
	if (!tvec_base_done[cpu]) {
1539 1540 1541
		static char boot_done;

		if (boot_done) {
A
Andrew Morton 已提交
1542 1543 1544
			/*
			 * The APs use this path later in boot
			 */
1545 1546 1547 1548 1549
			base = kmalloc_node(sizeof(*base), GFP_KERNEL,
						cpu_to_node(cpu));
			if (!base)
				return -ENOMEM;
			memset(base, 0, sizeof(*base));
A
Andrew Morton 已提交
1550
			per_cpu(tvec_bases, cpu) = base;
1551
		} else {
A
Andrew Morton 已提交
1552 1553 1554 1555 1556 1557
			/*
			 * 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.
			 */
1558
			boot_done = 1;
A
Andrew Morton 已提交
1559
			base = &boot_tvec_bases;
1560
		}
A
Andrew Morton 已提交
1561 1562 1563
		tvec_base_done[cpu] = 1;
	} else {
		base = per_cpu(tvec_bases, cpu);
1564
	}
A
Andrew Morton 已提交
1565

1566
	spin_lock_init(&base->lock);
1567 1568
	lockdep_set_class(&base->lock, base_lock_keys + cpu);

L
Linus Torvalds 已提交
1569 1570 1571 1572 1573 1574 1575 1576 1577 1578
	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;
1579
	return 0;
L
Linus Torvalds 已提交
1580 1581 1582
}

#ifdef CONFIG_HOTPLUG_CPU
1583
static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
L
Linus Torvalds 已提交
1584 1585 1586 1587 1588
{
	struct timer_list *timer;

	while (!list_empty(head)) {
		timer = list_entry(head->next, struct timer_list, entry);
1589
		detach_timer(timer, 0);
1590
		timer->base = new_base;
L
Linus Torvalds 已提交
1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601
		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));
1602 1603
	old_base = per_cpu(tvec_bases, cpu);
	new_base = get_cpu_var(tvec_bases);
L
Linus Torvalds 已提交
1604 1605

	local_irq_disable();
1606 1607 1608 1609
	spin_lock(&new_base->lock);
	spin_lock(&old_base->lock);

	BUG_ON(old_base->running_timer);
L
Linus Torvalds 已提交
1610 1611

	for (i = 0; i < TVR_SIZE; i++)
1612 1613 1614 1615 1616 1617 1618 1619
		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);
	}

1620 1621
	spin_unlock(&old_base->lock);
	spin_unlock(&new_base->lock);
L
Linus Torvalds 已提交
1622 1623 1624 1625 1626
	local_irq_enable();
	put_cpu_var(tvec_bases);
}
#endif /* CONFIG_HOTPLUG_CPU */

1627
static int __cpuinit timer_cpu_notify(struct notifier_block *self,
L
Linus Torvalds 已提交
1628 1629 1630 1631 1632
				unsigned long action, void *hcpu)
{
	long cpu = (long)hcpu;
	switch(action) {
	case CPU_UP_PREPARE:
1633 1634
		if (init_timers_cpu(cpu) < 0)
			return NOTIFY_BAD;
L
Linus Torvalds 已提交
1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646
		break;
#ifdef CONFIG_HOTPLUG_CPU
	case CPU_DEAD:
		migrate_timers(cpu);
		break;
#endif
	default:
		break;
	}
	return NOTIFY_OK;
}

1647
static struct notifier_block __cpuinitdata timers_nb = {
L
Linus Torvalds 已提交
1648 1649 1650 1651 1652 1653
	.notifier_call	= timer_cpu_notify,
};


void __init init_timers(void)
{
1654
	int err = timer_cpu_notify(&timers_nb, (unsigned long)CPU_UP_PREPARE,
L
Linus Torvalds 已提交
1655
				(void *)(long)smp_processor_id());
1656 1657

	BUG_ON(err == NOTIFY_BAD);
L
Linus Torvalds 已提交
1658 1659 1660 1661 1662 1663
	register_cpu_notifier(&timers_nb);
	open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL);
}

#ifdef CONFIG_TIME_INTERPOLATION

1664 1665
struct time_interpolator *time_interpolator __read_mostly;
static struct time_interpolator *time_interpolator_list __read_mostly;
L
Linus Torvalds 已提交
1666 1667
static DEFINE_SPINLOCK(time_interpolator_lock);

1668
static inline cycles_t time_interpolator_get_cycles(unsigned int src)
L
Linus Torvalds 已提交
1669 1670 1671 1672 1673 1674 1675 1676 1677 1678
{
	unsigned long (*x)(void);

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

		case TIME_SOURCE_MMIO64	:
1679
			return readq_relaxed((void __iomem *)time_interpolator->addr);
L
Linus Torvalds 已提交
1680 1681

		case TIME_SOURCE_MMIO32	:
1682
			return readl_relaxed((void __iomem *)time_interpolator->addr);
L
Linus Torvalds 已提交
1683 1684 1685 1686 1687

		default: return get_cycles();
	}
}

1688
static inline u64 time_interpolator_get_counter(int writelock)
L
Linus Torvalds 已提交
1689 1690 1691 1692 1693
{
	unsigned int src = time_interpolator->source;

	if (time_interpolator->jitter)
	{
1694 1695
		cycles_t lcycle;
		cycles_t now;
L
Linus Torvalds 已提交
1696 1697 1698 1699 1700 1701

		do {
			lcycle = time_interpolator->last_cycle;
			now = time_interpolator_get_cycles(src);
			if (lcycle && time_after(lcycle, now))
				return lcycle;
1702 1703 1704 1705 1706 1707 1708 1709 1710

			/* 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
1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723
			/* 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;
1724
	time_interpolator->last_counter = time_interpolator_get_counter(1);
L
Linus Torvalds 已提交
1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
}

#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 +
1736
		GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator);
L
Linus Torvalds 已提交
1737 1738 1739 1740 1741
}

#define INTERPOLATOR_ADJUST 65536
#define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST

1742
void time_interpolator_update(long delta_nsec)
L
Linus Torvalds 已提交
1743 1744 1745 1746 1747 1748 1749 1750
{
	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
1751 1752 1753 1754 1755 1756 1757 1758
	/*
	 * 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
1759

1760
	counter = time_interpolator_get_counter(1);
A
Andrew Morton 已提交
1761 1762
	offset = time_interpolator->offset +
			GET_TI_NSECS(counter, time_interpolator);
L
Linus Torvalds 已提交
1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778

	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)
	{
1779
		if (time_interpolator->skips == 0 && time_interpolator->offset > tick_nsec)
L
Linus Torvalds 已提交
1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802
			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 */
1803
	BUG_ON(ti->frequency == 0 || ti->mask == 0);
L
Linus Torvalds 已提交
1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857

	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;

1858 1859
	while (timeout)
		timeout = schedule_timeout_uninterruptible(timeout);
L
Linus Torvalds 已提交
1860 1861 1862 1863 1864
}

EXPORT_SYMBOL(msleep);

/**
1865
 * msleep_interruptible - sleep waiting for signals
L
Linus Torvalds 已提交
1866 1867 1868 1869 1870 1871
 * @msecs: Time in milliseconds to sleep for
 */
unsigned long msleep_interruptible(unsigned int msecs)
{
	unsigned long timeout = msecs_to_jiffies(msecs) + 1;

1872 1873
	while (timeout && !signal_pending(current))
		timeout = schedule_timeout_interruptible(timeout);
L
Linus Torvalds 已提交
1874 1875 1876 1877
	return jiffies_to_msecs(timeout);
}

EXPORT_SYMBOL(msleep_interruptible);