cputime.c 22.3 KB
Newer Older
1 2 3 4 5
#include <linux/export.h>
#include <linux/sched.h>
#include <linux/tsacct_kern.h>
#include <linux/kernel_stat.h>
#include <linux/static_key.h>
6
#include <linux/context_tracking.h>
7 8 9 10 11 12 13
#include "sched.h"


#ifdef CONFIG_IRQ_TIME_ACCOUNTING

/*
 * There are no locks covering percpu hardirq/softirq time.
14
 * They are only modified in vtime_account, on corresponding CPU
15 16 17
 * with interrupts disabled. So, writes are safe.
 * They are read and saved off onto struct rq in update_rq_clock().
 * This may result in other CPU reading this CPU's irq time and can
18
 * race with irq/vtime_account on this CPU. We would either get old
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
 * or new value with a side effect of accounting a slice of irq time to wrong
 * task when irq is in progress while we read rq->clock. That is a worthy
 * compromise in place of having locks on each irq in account_system_time.
 */
DEFINE_PER_CPU(u64, cpu_hardirq_time);
DEFINE_PER_CPU(u64, cpu_softirq_time);

static DEFINE_PER_CPU(u64, irq_start_time);
static int sched_clock_irqtime;

void enable_sched_clock_irqtime(void)
{
	sched_clock_irqtime = 1;
}

void disable_sched_clock_irqtime(void)
{
	sched_clock_irqtime = 0;
}

#ifndef CONFIG_64BIT
DEFINE_PER_CPU(seqcount_t, irq_time_seq);
#endif /* CONFIG_64BIT */

/*
 * Called before incrementing preempt_count on {soft,}irq_enter
 * and before decrementing preempt_count on {soft,}irq_exit.
 */
47
void irqtime_account_irq(struct task_struct *curr)
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
{
	unsigned long flags;
	s64 delta;
	int cpu;

	if (!sched_clock_irqtime)
		return;

	local_irq_save(flags);

	cpu = smp_processor_id();
	delta = sched_clock_cpu(cpu) - __this_cpu_read(irq_start_time);
	__this_cpu_add(irq_start_time, delta);

	irq_time_write_begin();
	/*
	 * We do not account for softirq time from ksoftirqd here.
	 * We want to continue accounting softirq time to ksoftirqd thread
	 * in that case, so as not to confuse scheduler with a special task
	 * that do not consume any time, but still wants to run.
	 */
	if (hardirq_count())
		__this_cpu_add(cpu_hardirq_time, delta);
	else if (in_serving_softirq() && curr != this_cpu_ksoftirqd())
		__this_cpu_add(cpu_softirq_time, delta);

	irq_time_write_end();
	local_irq_restore(flags);
}
77
EXPORT_SYMBOL_GPL(irqtime_account_irq);
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

static int irqtime_account_hi_update(void)
{
	u64 *cpustat = kcpustat_this_cpu->cpustat;
	unsigned long flags;
	u64 latest_ns;
	int ret = 0;

	local_irq_save(flags);
	latest_ns = this_cpu_read(cpu_hardirq_time);
	if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_IRQ])
		ret = 1;
	local_irq_restore(flags);
	return ret;
}

static int irqtime_account_si_update(void)
{
	u64 *cpustat = kcpustat_this_cpu->cpustat;
	unsigned long flags;
	u64 latest_ns;
	int ret = 0;

	local_irq_save(flags);
	latest_ns = this_cpu_read(cpu_softirq_time);
	if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_SOFTIRQ])
		ret = 1;
	local_irq_restore(flags);
	return ret;
}

#else /* CONFIG_IRQ_TIME_ACCOUNTING */

#define sched_clock_irqtime	(0)

#endif /* !CONFIG_IRQ_TIME_ACCOUNTING */

static inline void task_group_account_field(struct task_struct *p, int index,
					    u64 tmp)
{
	/*
	 * Since all updates are sure to touch the root cgroup, we
	 * get ourselves ahead and touch it first. If the root cgroup
	 * is the only cgroup, then nothing else should be necessary.
	 *
	 */
124
	__this_cpu_add(kernel_cpustat.cpustat[index], tmp);
125

126
	cpuacct_account_field(p, index, tmp);
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
}

/*
 * Account user cpu time to a process.
 * @p: the process that the cpu time gets accounted to
 * @cputime: the cpu time spent in user space since the last update
 * @cputime_scaled: cputime scaled by cpu frequency
 */
void account_user_time(struct task_struct *p, cputime_t cputime,
		       cputime_t cputime_scaled)
{
	int index;

	/* Add user time to process. */
	p->utime += cputime;
	p->utimescaled += cputime_scaled;
	account_group_user_time(p, cputime);

145
	index = (task_nice(p) > 0) ? CPUTIME_NICE : CPUTIME_USER;
146 147 148 149 150

	/* Add user time to cpustat. */
	task_group_account_field(p, index, (__force u64) cputime);

	/* Account for user time used */
151
	acct_account_cputime(p);
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
}

/*
 * Account guest cpu time to a process.
 * @p: the process that the cpu time gets accounted to
 * @cputime: the cpu time spent in virtual machine since the last update
 * @cputime_scaled: cputime scaled by cpu frequency
 */
static void account_guest_time(struct task_struct *p, cputime_t cputime,
			       cputime_t cputime_scaled)
{
	u64 *cpustat = kcpustat_this_cpu->cpustat;

	/* Add guest time to process. */
	p->utime += cputime;
	p->utimescaled += cputime_scaled;
	account_group_user_time(p, cputime);
	p->gtime += cputime;

	/* Add guest time to cpustat. */
172
	if (task_nice(p) > 0) {
173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
		cpustat[CPUTIME_NICE] += (__force u64) cputime;
		cpustat[CPUTIME_GUEST_NICE] += (__force u64) cputime;
	} else {
		cpustat[CPUTIME_USER] += (__force u64) cputime;
		cpustat[CPUTIME_GUEST] += (__force u64) cputime;
	}
}

/*
 * Account system cpu time to a process and desired cpustat field
 * @p: the process that the cpu time gets accounted to
 * @cputime: the cpu time spent in kernel space since the last update
 * @cputime_scaled: cputime scaled by cpu frequency
 * @target_cputime64: pointer to cpustat field that has to be updated
 */
static inline
void __account_system_time(struct task_struct *p, cputime_t cputime,
			cputime_t cputime_scaled, int index)
{
	/* Add system time to process. */
	p->stime += cputime;
	p->stimescaled += cputime_scaled;
	account_group_system_time(p, cputime);

	/* Add system time to cpustat. */
	task_group_account_field(p, index, (__force u64) cputime);

	/* Account for system time used */
201
	acct_account_cputime(p);
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
}

/*
 * Account system cpu time to a process.
 * @p: the process that the cpu time gets accounted to
 * @hardirq_offset: the offset to subtract from hardirq_count()
 * @cputime: the cpu time spent in kernel space since the last update
 * @cputime_scaled: cputime scaled by cpu frequency
 */
void account_system_time(struct task_struct *p, int hardirq_offset,
			 cputime_t cputime, cputime_t cputime_scaled)
{
	int index;

	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
		account_guest_time(p, cputime, cputime_scaled);
		return;
	}

	if (hardirq_count() - hardirq_offset)
		index = CPUTIME_IRQ;
	else if (in_serving_softirq())
		index = CPUTIME_SOFTIRQ;
	else
		index = CPUTIME_SYSTEM;

	__account_system_time(p, cputime, cputime_scaled, index);
}

/*
 * Account for involuntary wait time.
 * @cputime: the cpu time spent in involuntary wait
 */
void account_steal_time(cputime_t cputime)
{
	u64 *cpustat = kcpustat_this_cpu->cpustat;

	cpustat[CPUTIME_STEAL] += (__force u64) cputime;
}

/*
 * Account for idle time.
 * @cputime: the cpu time spent in idle wait
 */
void account_idle_time(cputime_t cputime)
{
	u64 *cpustat = kcpustat_this_cpu->cpustat;
	struct rq *rq = this_rq();

	if (atomic_read(&rq->nr_iowait) > 0)
		cpustat[CPUTIME_IOWAIT] += (__force u64) cputime;
	else
		cpustat[CPUTIME_IDLE] += (__force u64) cputime;
}

static __always_inline bool steal_account_process_tick(void)
{
#ifdef CONFIG_PARAVIRT
	if (static_key_false(&paravirt_steal_enabled)) {
261 262
		u64 steal;
		cputime_t steal_ct;
263 264 265 266

		steal = paravirt_steal_clock(smp_processor_id());
		steal -= this_rq()->prev_steal_time;

267 268 269 270 271 272 273
		/*
		 * cputime_t may be less precise than nsecs (eg: if it's
		 * based on jiffies). Lets cast the result to cputime
		 * granularity and account the rest on the next rounds.
		 */
		steal_ct = nsecs_to_cputime(steal);
		this_rq()->prev_steal_time += cputime_to_nsecs(steal_ct);
274

275 276
		account_steal_time(steal_ct);
		return steal_ct;
277 278 279 280 281
	}
#endif
	return false;
}

282 283 284 285 286 287 288
/*
 * Accumulate raw cputime values of dead tasks (sig->[us]time) and live
 * tasks (sum on group iteration) belonging to @tsk's group.
 */
void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
{
	struct signal_struct *sig = tsk->signal;
289
	cputime_t utime, stime;
290
	struct task_struct *t;
291
	unsigned int seq, nextseq;
292
	unsigned long flags;
293 294

	rcu_read_lock();
295 296 297 298
	/* Attempt a lockless read on the first round. */
	nextseq = 0;
	do {
		seq = nextseq;
299
		flags = read_seqbegin_or_lock_irqsave(&sig->stats_lock, &seq);
300 301 302 303 304 305 306 307 308 309 310 311 312
		times->utime = sig->utime;
		times->stime = sig->stime;
		times->sum_exec_runtime = sig->sum_sched_runtime;

		for_each_thread(tsk, t) {
			task_cputime(t, &utime, &stime);
			times->utime += utime;
			times->stime += stime;
			times->sum_exec_runtime += task_sched_runtime(t);
		}
		/* If lockless access failed, take the lock. */
		nextseq = 1;
	} while (need_seqretry(&sig->stats_lock, seq));
313
	done_seqretry_irqrestore(&sig->stats_lock, seq, flags);
314 315 316
	rcu_read_unlock();
}

317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
#ifdef CONFIG_IRQ_TIME_ACCOUNTING
/*
 * Account a tick to a process and cpustat
 * @p: the process that the cpu time gets accounted to
 * @user_tick: is the tick from userspace
 * @rq: the pointer to rq
 *
 * Tick demultiplexing follows the order
 * - pending hardirq update
 * - pending softirq update
 * - user_time
 * - idle_time
 * - system time
 *   - check for guest_time
 *   - else account as system_time
 *
 * Check for hardirq is done both for system and user time as there is
 * no timer going off while we are on hardirq and hence we may never get an
 * opportunity to update it solely in system time.
 * p->stime and friends are only updated on system time and not on irq
 * softirq as those do not count in task exec_runtime any more.
 */
static void irqtime_account_process_tick(struct task_struct *p, int user_tick,
340
					 struct rq *rq, int ticks)
341
{
342 343
	cputime_t scaled = cputime_to_scaled(cputime_one_jiffy);
	u64 cputime = (__force u64) cputime_one_jiffy;
344 345 346 347 348
	u64 *cpustat = kcpustat_this_cpu->cpustat;

	if (steal_account_process_tick())
		return;

349 350 351
	cputime *= ticks;
	scaled *= ticks;

352
	if (irqtime_account_hi_update()) {
353
		cpustat[CPUTIME_IRQ] += cputime;
354
	} else if (irqtime_account_si_update()) {
355
		cpustat[CPUTIME_SOFTIRQ] += cputime;
356 357 358 359 360 361
	} else if (this_cpu_ksoftirqd() == p) {
		/*
		 * ksoftirqd time do not get accounted in cpu_softirq_time.
		 * So, we have to handle it separately here.
		 * Also, p->stime needs to be updated for ksoftirqd.
		 */
362
		__account_system_time(p, cputime, scaled, CPUTIME_SOFTIRQ);
363
	} else if (user_tick) {
364
		account_user_time(p, cputime, scaled);
365
	} else if (p == rq->idle) {
366
		account_idle_time(cputime);
367
	} else if (p->flags & PF_VCPU) { /* System time or guest time */
368
		account_guest_time(p, cputime, scaled);
369
	} else {
370
		__account_system_time(p, cputime, scaled,	CPUTIME_SYSTEM);
371 372 373 374 375 376 377
	}
}

static void irqtime_account_idle_ticks(int ticks)
{
	struct rq *rq = this_rq();

378
	irqtime_account_process_tick(current, 0, rq, ticks);
379 380
}
#else /* CONFIG_IRQ_TIME_ACCOUNTING */
381 382
static inline void irqtime_account_idle_ticks(int ticks) {}
static inline void irqtime_account_process_tick(struct task_struct *p, int user_tick,
383
						struct rq *rq, int nr_ticks) {}
384 385 386 387 388 389
#endif /* CONFIG_IRQ_TIME_ACCOUNTING */

/*
 * Use precise platform statistics if available:
 */
#ifdef CONFIG_VIRT_CPU_ACCOUNTING
390

391
#ifndef __ARCH_HAS_VTIME_TASK_SWITCH
392
void vtime_common_task_switch(struct task_struct *prev)
393 394 395 396 397 398
{
	if (is_idle_task(prev))
		vtime_account_idle(prev);
	else
		vtime_account_system(prev);

399
#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
400
	vtime_account_user(prev);
401
#endif
402 403 404
	arch_vtime_task_switch(prev);
}
#endif
405

406 407 408
/*
 * Archs that account the whole time spent in the idle task
 * (outside irq) as idle time can rely on this and just implement
409
 * vtime_account_system() and vtime_account_idle(). Archs that
410 411 412 413 414
 * have other meaning of the idle time (s390 only includes the
 * time spent by the CPU when it's in low power mode) must override
 * vtime_account().
 */
#ifndef __ARCH_HAS_VTIME_ACCOUNT
415
void vtime_common_account_irq_enter(struct task_struct *tsk)
416
{
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
	if (!in_interrupt()) {
		/*
		 * If we interrupted user, context_tracking_in_user()
		 * is 1 because the context tracking don't hook
		 * on irq entry/exit. This way we know if
		 * we need to flush user time on kernel entry.
		 */
		if (context_tracking_in_user()) {
			vtime_account_user(tsk);
			return;
		}

		if (is_idle_task(tsk)) {
			vtime_account_idle(tsk);
			return;
		}
	}
	vtime_account_system(tsk);
435
}
436
EXPORT_SYMBOL_GPL(vtime_common_account_irq_enter);
437
#endif /* __ARCH_HAS_VTIME_ACCOUNT */
438 439 440 441 442 443 444 445 446
#endif /* CONFIG_VIRT_CPU_ACCOUNTING */


#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
{
	*ut = p->utime;
	*st = p->stime;
}
447

448 449 450
void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
{
	struct task_cputime cputime;
451

452 453 454 455 456 457 458 459 460 461 462 463
	thread_group_cputime(p, &cputime);

	*ut = cputime.utime;
	*st = cputime.stime;
}
#else /* !CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
/*
 * Account a single tick of cpu time.
 * @p: the process that the cpu time gets accounted to
 * @user_tick: indicates if the tick is a user or a system tick
 */
void account_process_tick(struct task_struct *p, int user_tick)
464
{
465 466
	cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy);
	struct rq *rq = this_rq();
467

468 469 470 471
	if (vtime_accounting_enabled())
		return;

	if (sched_clock_irqtime) {
472
		irqtime_account_process_tick(p, user_tick, rq, 1);
473 474 475 476 477
		return;
	}

	if (steal_account_process_tick())
		return;
478

479 480 481 482 483
	if (user_tick)
		account_user_time(p, cputime_one_jiffy, one_jiffy_scaled);
	else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET))
		account_system_time(p, HARDIRQ_OFFSET, cputime_one_jiffy,
				    one_jiffy_scaled);
484
	else
485 486
		account_idle_time(cputime_one_jiffy);
}
487

488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511
/*
 * Account multiple ticks of steal time.
 * @p: the process from which the cpu time has been stolen
 * @ticks: number of stolen ticks
 */
void account_steal_ticks(unsigned long ticks)
{
	account_steal_time(jiffies_to_cputime(ticks));
}

/*
 * Account multiple ticks of idle time.
 * @ticks: number of stolen ticks
 */
void account_idle_ticks(unsigned long ticks)
{

	if (sched_clock_irqtime) {
		irqtime_account_idle_ticks(ticks);
		return;
	}

	account_idle_time(jiffies_to_cputime(ticks));
}
512

513
/*
514 515
 * Perform (stime * rtime) / total, but avoid multiplication overflow by
 * loosing precision when the numbers are big.
516 517
 */
static cputime_t scale_stime(u64 stime, u64 rtime, u64 total)
518
{
519
	u64 scaled;
520

521 522
	for (;;) {
		/* Make sure "rtime" is the bigger of stime/rtime */
523 524
		if (stime > rtime)
			swap(rtime, stime);
525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546

		/* Make sure 'total' fits in 32 bits */
		if (total >> 32)
			goto drop_precision;

		/* Does rtime (and thus stime) fit in 32 bits? */
		if (!(rtime >> 32))
			break;

		/* Can we just balance rtime/stime rather than dropping bits? */
		if (stime >> 31)
			goto drop_precision;

		/* We can grow stime and shrink rtime and try to make them both fit */
		stime <<= 1;
		rtime >>= 1;
		continue;

drop_precision:
		/* We drop from rtime, it has more bits than stime */
		rtime >>= 1;
		total >>= 1;
547
	}
548

549 550 551 552 553
	/*
	 * Make sure gcc understands that this is a 32x32->64 multiply,
	 * followed by a 64/32->64 divide.
	 */
	scaled = div_u64((u64) (u32) stime * (u64) (u32) rtime, (u32)total);
554
	return (__force cputime_t) scaled;
555 556
}

557
/*
558 559
 * Adjust tick based cputime random precision against scheduler runtime
 * accounting.
560
 *
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
 * Tick based cputime accounting depend on random scheduling timeslices of a
 * task to be interrupted or not by the timer.  Depending on these
 * circumstances, the number of these interrupts may be over or
 * under-optimistic, matching the real user and system cputime with a variable
 * precision.
 *
 * Fix this by scaling these tick based values against the total runtime
 * accounted by the CFS scheduler.
 *
 * This code provides the following guarantees:
 *
 *   stime + utime == rtime
 *   stime_i+1 >= stime_i, utime_i+1 >= utime_i
 *
 * Assuming that rtime_i+1 >= rtime_i.
576
 */
577
static void cputime_adjust(struct task_cputime *curr,
578
			   struct prev_cputime *prev,
579
			   cputime_t *ut, cputime_t *st)
580
{
581
	cputime_t rtime, stime, utime;
582
	unsigned long flags;
583

584 585
	/* Serialize concurrent callers such that we can honour our guarantees */
	raw_spin_lock_irqsave(&prev->lock, flags);
586
	rtime = nsecs_to_cputime(curr->sum_exec_runtime);
587

588
	/*
589 590 591 592 593 594
	 * This is possible under two circumstances:
	 *  - rtime isn't monotonic after all (a bug);
	 *  - we got reordered by the lock.
	 *
	 * In both cases this acts as a filter such that the rest of the code
	 * can assume it is monotonic regardless of anything else.
595 596 597 598
	 */
	if (prev->stime + prev->utime >= rtime)
		goto out;

599 600 601 602 603
	stime = curr->stime;
	utime = curr->utime;

	if (utime == 0) {
		stime = rtime;
604 605
		goto update;
	}
606

607 608 609
	if (stime == 0) {
		utime = rtime;
		goto update;
610
	}
611

612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635
	stime = scale_stime((__force u64)stime, (__force u64)rtime,
			    (__force u64)(stime + utime));

	/*
	 * Make sure stime doesn't go backwards; this preserves monotonicity
	 * for utime because rtime is monotonic.
	 *
	 *  utime_i+1 = rtime_i+1 - stime_i
	 *            = rtime_i+1 - (rtime_i - utime_i)
	 *            = (rtime_i+1 - rtime_i) + utime_i
	 *            >= utime_i
	 */
	if (stime < prev->stime)
		stime = prev->stime;
	utime = rtime - stime;

	/*
	 * Make sure utime doesn't go backwards; this still preserves
	 * monotonicity for stime, analogous argument to above.
	 */
	if (utime < prev->utime) {
		utime = prev->utime;
		stime = rtime - utime;
	}
636

637 638 639
update:
	prev->stime = stime;
	prev->utime = utime;
640
out:
641 642
	*ut = prev->utime;
	*st = prev->stime;
643
	raw_spin_unlock_irqrestore(&prev->lock, flags);
644
}
645

646 647 648 649 650 651
void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
{
	struct task_cputime cputime = {
		.sum_exec_runtime = p->se.sum_exec_runtime,
	};

652
	task_cputime(p, &cputime.utime, &cputime.stime);
653
	cputime_adjust(&cputime, &p->prev_cputime, ut, st);
654 655
}

656
void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st)
657 658 659 660
{
	struct task_cputime cputime;

	thread_group_cputime(p, &cputime);
661
	cputime_adjust(&cputime, &p->signal->prev_cputime, ut, st);
662
}
663
#endif /* !CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */
664 665

#ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
666 667 668 669
static unsigned long long vtime_delta(struct task_struct *tsk)
{
	unsigned long long clock;

670
	clock = local_clock();
671 672
	if (clock < tsk->vtime_snap)
		return 0;
673

674 675 676 677
	return clock - tsk->vtime_snap;
}

static cputime_t get_vtime_delta(struct task_struct *tsk)
678
{
679
	unsigned long long delta = vtime_delta(tsk);
680

681 682
	WARN_ON_ONCE(tsk->vtime_snap_whence == VTIME_SLEEPING);
	tsk->vtime_snap += delta;
683 684 685 686 687

	/* CHECKME: always safe to convert nsecs to cputime? */
	return nsecs_to_cputime(delta);
}

688 689 690 691 692 693 694
static void __vtime_account_system(struct task_struct *tsk)
{
	cputime_t delta_cpu = get_vtime_delta(tsk);

	account_system_time(tsk, irq_count(), delta_cpu, cputime_to_scaled(delta_cpu));
}

695 696
void vtime_account_system(struct task_struct *tsk)
{
697 698 699 700
	write_seqlock(&tsk->vtime_seqlock);
	__vtime_account_system(tsk);
	write_sequnlock(&tsk->vtime_seqlock);
}
701

702
void vtime_gen_account_irq_exit(struct task_struct *tsk)
703 704
{
	write_seqlock(&tsk->vtime_seqlock);
705
	__vtime_account_system(tsk);
706 707 708
	if (context_tracking_in_user())
		tsk->vtime_snap_whence = VTIME_USER;
	write_sequnlock(&tsk->vtime_seqlock);
709 710 711 712
}

void vtime_account_user(struct task_struct *tsk)
{
713 714
	cputime_t delta_cpu;

715
	write_seqlock(&tsk->vtime_seqlock);
716
	delta_cpu = get_vtime_delta(tsk);
717
	tsk->vtime_snap_whence = VTIME_SYS;
718
	account_user_time(tsk, delta_cpu, cputime_to_scaled(delta_cpu));
719 720 721 722 723 724 725
	write_sequnlock(&tsk->vtime_seqlock);
}

void vtime_user_enter(struct task_struct *tsk)
{
	write_seqlock(&tsk->vtime_seqlock);
	__vtime_account_system(tsk);
726
	tsk->vtime_snap_whence = VTIME_USER;
727 728 729 730 731
	write_sequnlock(&tsk->vtime_seqlock);
}

void vtime_guest_enter(struct task_struct *tsk)
{
732 733 734 735 736 737 738
	/*
	 * The flags must be updated under the lock with
	 * the vtime_snap flush and update.
	 * That enforces a right ordering and update sequence
	 * synchronization against the reader (task_gtime())
	 * that can thus safely catch up with a tickless delta.
	 */
739 740 741 742 743
	write_seqlock(&tsk->vtime_seqlock);
	__vtime_account_system(tsk);
	current->flags |= PF_VCPU;
	write_sequnlock(&tsk->vtime_seqlock);
}
744
EXPORT_SYMBOL_GPL(vtime_guest_enter);
745 746 747 748 749 750 751

void vtime_guest_exit(struct task_struct *tsk)
{
	write_seqlock(&tsk->vtime_seqlock);
	__vtime_account_system(tsk);
	current->flags &= ~PF_VCPU;
	write_sequnlock(&tsk->vtime_seqlock);
752
}
753
EXPORT_SYMBOL_GPL(vtime_guest_exit);
754 755 756

void vtime_account_idle(struct task_struct *tsk)
{
757
	cputime_t delta_cpu = get_vtime_delta(tsk);
758 759 760

	account_idle_time(delta_cpu);
}
761

762 763 764 765 766 767 768 769
void arch_vtime_task_switch(struct task_struct *prev)
{
	write_seqlock(&prev->vtime_seqlock);
	prev->vtime_snap_whence = VTIME_SLEEPING;
	write_sequnlock(&prev->vtime_seqlock);

	write_seqlock(&current->vtime_seqlock);
	current->vtime_snap_whence = VTIME_SYS;
770
	current->vtime_snap = sched_clock_cpu(smp_processor_id());
771 772 773
	write_sequnlock(&current->vtime_seqlock);
}

774
void vtime_init_idle(struct task_struct *t, int cpu)
775 776 777 778 779
{
	unsigned long flags;

	write_seqlock_irqsave(&t->vtime_seqlock, flags);
	t->vtime_snap_whence = VTIME_SYS;
780
	t->vtime_snap = sched_clock_cpu(cpu);
781 782 783 784 785 786 787 788 789
	write_sequnlock_irqrestore(&t->vtime_seqlock, flags);
}

cputime_t task_gtime(struct task_struct *t)
{
	unsigned int seq;
	cputime_t gtime;

	do {
790
		seq = read_seqbegin(&t->vtime_seqlock);
791 792 793 794 795

		gtime = t->gtime;
		if (t->flags & PF_VCPU)
			gtime += vtime_delta(t);

796
	} while (read_seqretry(&t->vtime_seqlock, seq));
797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818

	return gtime;
}

/*
 * Fetch cputime raw values from fields of task_struct and
 * add up the pending nohz execution time since the last
 * cputime snapshot.
 */
static void
fetch_task_cputime(struct task_struct *t,
		   cputime_t *u_dst, cputime_t *s_dst,
		   cputime_t *u_src, cputime_t *s_src,
		   cputime_t *udelta, cputime_t *sdelta)
{
	unsigned int seq;
	unsigned long long delta;

	do {
		*udelta = 0;
		*sdelta = 0;

819
		seq = read_seqbegin(&t->vtime_seqlock);
820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842

		if (u_dst)
			*u_dst = *u_src;
		if (s_dst)
			*s_dst = *s_src;

		/* Task is sleeping, nothing to add */
		if (t->vtime_snap_whence == VTIME_SLEEPING ||
		    is_idle_task(t))
			continue;

		delta = vtime_delta(t);

		/*
		 * Task runs either in user or kernel space, add pending nohz time to
		 * the right place.
		 */
		if (t->vtime_snap_whence == VTIME_USER || t->flags & PF_VCPU) {
			*udelta = delta;
		} else {
			if (t->vtime_snap_whence == VTIME_SYS)
				*sdelta = delta;
		}
843
	} while (read_seqretry(&t->vtime_seqlock, seq));
844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
}


void task_cputime(struct task_struct *t, cputime_t *utime, cputime_t *stime)
{
	cputime_t udelta, sdelta;

	fetch_task_cputime(t, utime, stime, &t->utime,
			   &t->stime, &udelta, &sdelta);
	if (utime)
		*utime += udelta;
	if (stime)
		*stime += sdelta;
}

void task_cputime_scaled(struct task_struct *t,
			 cputime_t *utimescaled, cputime_t *stimescaled)
{
	cputime_t udelta, sdelta;

	fetch_task_cputime(t, utimescaled, stimescaled,
			   &t->utimescaled, &t->stimescaled, &udelta, &sdelta);
	if (utimescaled)
		*utimescaled += cputime_to_scaled(udelta);
	if (stimescaled)
		*stimescaled += cputime_to_scaled(sdelta);
}
871
#endif /* CONFIG_VIRT_CPU_ACCOUNTING_GEN */