vmstat.c 47.6 KB
Newer Older
1 2 3 4 5
/*
 *  linux/mm/vmstat.c
 *
 *  Manages VM statistics
 *  Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
6 7 8 9
 *
 *  zoned VM statistics
 *  Copyright (C) 2006 Silicon Graphics, Inc.,
 *		Christoph Lameter <christoph@lameter.com>
10
 *  Copyright (C) 2008-2014 Christoph Lameter
11
 */
12
#include <linux/fs.h>
13
#include <linux/mm.h>
Alexey Dobriyan's avatar
Alexey Dobriyan committed
14
#include <linux/err.h>
15
#include <linux/module.h>
16
#include <linux/slab.h>
17
#include <linux/cpu.h>
18
#include <linux/cpumask.h>
Adrian Bunk's avatar
Adrian Bunk committed
19
#include <linux/vmstat.h>
20 21 22
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/debugfs.h>
Alexey Dobriyan's avatar
Alexey Dobriyan committed
23
#include <linux/sched.h>
24
#include <linux/math64.h>
25
#include <linux/writeback.h>
26
#include <linux/compaction.h>
27
#include <linux/mm_inline.h>
28 29
#include <linux/page_ext.h>
#include <linux/page_owner.h>
30 31

#include "internal.h"
32

33 34 35 36
#ifdef CONFIG_VM_EVENT_COUNTERS
DEFINE_PER_CPU(struct vm_event_state, vm_event_states) = {{0}};
EXPORT_PER_CPU_SYMBOL(vm_event_states);

37
static void sum_vm_events(unsigned long *ret)
38
{
Christoph Lameter's avatar
Christoph Lameter committed
39
	int cpu;
40 41 42 43
	int i;

	memset(ret, 0, NR_VM_EVENT_ITEMS * sizeof(unsigned long));

44
	for_each_online_cpu(cpu) {
45 46 47 48 49 50 51 52 53 54 55 56 57 58
		struct vm_event_state *this = &per_cpu(vm_event_states, cpu);

		for (i = 0; i < NR_VM_EVENT_ITEMS; i++)
			ret[i] += this->event[i];
	}
}

/*
 * Accumulate the vm event counters across all CPUs.
 * The result is unavoidably approximate - it can change
 * during and after execution of this function.
*/
void all_vm_events(unsigned long *ret)
{
KOSAKI Motohiro's avatar
KOSAKI Motohiro committed
59
	get_online_cpus();
60
	sum_vm_events(ret);
KOSAKI Motohiro's avatar
KOSAKI Motohiro committed
61
	put_online_cpus();
62
}
63
EXPORT_SYMBOL_GPL(all_vm_events);
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

/*
 * Fold the foreign cpu events into our own.
 *
 * This is adding to the events on one processor
 * but keeps the global counts constant.
 */
void vm_events_fold_cpu(int cpu)
{
	struct vm_event_state *fold_state = &per_cpu(vm_event_states, cpu);
	int i;

	for (i = 0; i < NR_VM_EVENT_ITEMS; i++) {
		count_vm_events(i, fold_state->event[i]);
		fold_state->event[i] = 0;
	}
}

#endif /* CONFIG_VM_EVENT_COUNTERS */

84 85 86 87 88
/*
 * Manage combined zone based / global counters
 *
 * vm_stat contains the global counters
 */
89 90 91 92
atomic_long_t vm_zone_stat[NR_VM_ZONE_STAT_ITEMS] __cacheline_aligned_in_smp;
atomic_long_t vm_node_stat[NR_VM_NODE_STAT_ITEMS] __cacheline_aligned_in_smp;
EXPORT_SYMBOL(vm_zone_stat);
EXPORT_SYMBOL(vm_node_stat);
93 94 95

#ifdef CONFIG_SMP

96
int calculate_pressure_threshold(struct zone *zone)
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
{
	int threshold;
	int watermark_distance;

	/*
	 * As vmstats are not up to date, there is drift between the estimated
	 * and real values. For high thresholds and a high number of CPUs, it
	 * is possible for the min watermark to be breached while the estimated
	 * value looks fine. The pressure threshold is a reduced value such
	 * that even the maximum amount of drift will not accidentally breach
	 * the min watermark
	 */
	watermark_distance = low_wmark_pages(zone) - min_wmark_pages(zone);
	threshold = max(1, (int)(watermark_distance / num_online_cpus()));

	/*
	 * Maximum threshold is 125
	 */
	threshold = min(125, threshold);

	return threshold;
}

120
int calculate_normal_threshold(struct zone *zone)
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154
{
	int threshold;
	int mem;	/* memory in 128 MB units */

	/*
	 * The threshold scales with the number of processors and the amount
	 * of memory per zone. More memory means that we can defer updates for
	 * longer, more processors could lead to more contention.
 	 * fls() is used to have a cheap way of logarithmic scaling.
	 *
	 * Some sample thresholds:
	 *
	 * Threshold	Processors	(fls)	Zonesize	fls(mem+1)
	 * ------------------------------------------------------------------
	 * 8		1		1	0.9-1 GB	4
	 * 16		2		2	0.9-1 GB	4
	 * 20 		2		2	1-2 GB		5
	 * 24		2		2	2-4 GB		6
	 * 28		2		2	4-8 GB		7
	 * 32		2		2	8-16 GB		8
	 * 4		2		2	<128M		1
	 * 30		4		3	2-4 GB		5
	 * 48		4		3	8-16 GB		8
	 * 32		8		4	1-2 GB		4
	 * 32		8		4	0.9-1GB		4
	 * 10		16		5	<128M		1
	 * 40		16		5	900M		4
	 * 70		64		7	2-4 GB		5
	 * 84		64		7	4-8 GB		6
	 * 108		512		9	4-8 GB		6
	 * 125		1024		10	8-16 GB		8
	 * 125		1024		10	16-32 GB	9
	 */

155
	mem = zone->managed_pages >> (27 - PAGE_SHIFT);
156 157 158 159 160 161 162 163 164 165

	threshold = 2 * fls(num_online_cpus()) * (1 + fls(mem));

	/*
	 * Maximum threshold is 125
	 */
	threshold = min(125, threshold);

	return threshold;
}
166 167

/*
168
 * Refresh the thresholds for each zone.
169
 */
170
void refresh_zone_stat_thresholds(void)
171
{
172
	struct pglist_data *pgdat;
173 174 175 176
	struct zone *zone;
	int cpu;
	int threshold;

177 178 179 180 181 182 183
	/* Zero current pgdat thresholds */
	for_each_online_pgdat(pgdat) {
		for_each_online_cpu(cpu) {
			per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold = 0;
		}
	}

184
	for_each_populated_zone(zone) {
185
		struct pglist_data *pgdat = zone->zone_pgdat;
186 187
		unsigned long max_drift, tolerate_drift;

188
		threshold = calculate_normal_threshold(zone);
189

190 191 192
		for_each_online_cpu(cpu) {
			int pgdat_threshold;

193 194
			per_cpu_ptr(zone->pageset, cpu)->stat_threshold
							= threshold;
195

196 197 198 199 200 201
			/* Base nodestat threshold on the largest populated zone. */
			pgdat_threshold = per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold;
			per_cpu_ptr(pgdat->per_cpu_nodestats, cpu)->stat_threshold
				= max(threshold, pgdat_threshold);
		}

202 203 204 205 206 207 208 209 210 211
		/*
		 * Only set percpu_drift_mark if there is a danger that
		 * NR_FREE_PAGES reports the low watermark is ok when in fact
		 * the min watermark could be breached by an allocation
		 */
		tolerate_drift = low_wmark_pages(zone) - min_wmark_pages(zone);
		max_drift = num_online_cpus() * threshold;
		if (max_drift > tolerate_drift)
			zone->percpu_drift_mark = high_wmark_pages(zone) +
					max_drift;
212
	}
213 214
}

215 216
void set_pgdat_percpu_threshold(pg_data_t *pgdat,
				int (*calculate_pressure)(struct zone *))
217 218 219 220 221 222 223 224 225 226 227
{
	struct zone *zone;
	int cpu;
	int threshold;
	int i;

	for (i = 0; i < pgdat->nr_zones; i++) {
		zone = &pgdat->node_zones[i];
		if (!zone->percpu_drift_mark)
			continue;

228
		threshold = (*calculate_pressure)(zone);
229
		for_each_online_cpu(cpu)
230 231 232 233 234
			per_cpu_ptr(zone->pageset, cpu)->stat_threshold
							= threshold;
	}
}

235
/*
236 237 238
 * For use when we know that interrupts are disabled,
 * or when we know that preemption is disabled and that
 * particular counter cannot be updated from interrupt context.
239 240
 */
void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
241
			   long delta)
242
{
243 244
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
245
	long x;
246 247 248
	long t;

	x = delta + __this_cpu_read(*p);
249

250
	t = __this_cpu_read(pcp->stat_threshold);
251

252
	if (unlikely(x > t || x < -t)) {
253 254 255
		zone_page_state_add(x, zone, item);
		x = 0;
	}
256
	__this_cpu_write(*p, x);
257 258 259
}
EXPORT_SYMBOL(__mod_zone_page_state);

260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
				long delta)
{
	struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats;
	s8 __percpu *p = pcp->vm_node_stat_diff + item;
	long x;
	long t;

	x = delta + __this_cpu_read(*p);

	t = __this_cpu_read(pcp->stat_threshold);

	if (unlikely(x > t || x < -t)) {
		node_page_state_add(x, pgdat, item);
		x = 0;
	}
	__this_cpu_write(*p, x);
}
EXPORT_SYMBOL(__mod_node_page_state);

280 281 282 283 284 285 286 287 288 289 290 291 292
/*
 * Optimized increment and decrement functions.
 *
 * These are only for a single page and therefore can take a struct page *
 * argument instead of struct zone *. This allows the inclusion of the code
 * generated for page_zone(page) into the optimized functions.
 *
 * No overflow check is necessary and therefore the differential can be
 * incremented or decremented in place which may allow the compilers to
 * generate better code.
 * The increment or decrement is known and therefore one boundary check can
 * be omitted.
 *
293 294 295
 * NOTE: These functions are very performance sensitive. Change only
 * with care.
 *
296 297 298 299 300 301 302
 * Some processors have inc/dec instructions that are atomic vs an interrupt.
 * However, the code must first determine the differential location in a zone
 * based on the processor number and then inc/dec the counter. There is no
 * guarantee without disabling preemption that the processor will not change
 * in between and therefore the atomicity vs. interrupt cannot be exploited
 * in a useful way here.
 */
303
void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
304
{
305 306 307
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
	s8 v, t;
308

309
	v = __this_cpu_inc_return(*p);
310 311 312
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v > t)) {
		s8 overstep = t >> 1;
313

314 315
		zone_page_state_add(v + overstep, zone, item);
		__this_cpu_write(*p, -overstep);
316 317
	}
}
318

319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
{
	struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats;
	s8 __percpu *p = pcp->vm_node_stat_diff + item;
	s8 v, t;

	v = __this_cpu_inc_return(*p);
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v > t)) {
		s8 overstep = t >> 1;

		node_page_state_add(v + overstep, pgdat, item);
		__this_cpu_write(*p, -overstep);
	}
}

335 336 337 338
void __inc_zone_page_state(struct page *page, enum zone_stat_item item)
{
	__inc_zone_state(page_zone(page), item);
}
339 340
EXPORT_SYMBOL(__inc_zone_page_state);

341 342 343 344 345 346
void __inc_node_page_state(struct page *page, enum node_stat_item item)
{
	__inc_node_state(page_pgdat(page), item);
}
EXPORT_SYMBOL(__inc_node_page_state);

347
void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
348
{
349 350 351
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
	s8 v, t;
352

353
	v = __this_cpu_dec_return(*p);
354 355 356
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v < - t)) {
		s8 overstep = t >> 1;
357

358 359
		zone_page_state_add(v - overstep, zone, item);
		__this_cpu_write(*p, overstep);
360 361
	}
}
362

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item)
{
	struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats;
	s8 __percpu *p = pcp->vm_node_stat_diff + item;
	s8 v, t;

	v = __this_cpu_dec_return(*p);
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v < - t)) {
		s8 overstep = t >> 1;

		node_page_state_add(v - overstep, pgdat, item);
		__this_cpu_write(*p, overstep);
	}
}

379 380 381 382
void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
{
	__dec_zone_state(page_zone(page), item);
}
383 384
EXPORT_SYMBOL(__dec_zone_page_state);

385 386 387 388 389 390
void __dec_node_page_state(struct page *page, enum node_stat_item item)
{
	__dec_node_state(page_pgdat(page), item);
}
EXPORT_SYMBOL(__dec_node_page_state);

391
#ifdef CONFIG_HAVE_CMPXCHG_LOCAL
392 393 394 395 396 397 398 399 400 401 402 403
/*
 * If we have cmpxchg_local support then we do not need to incur the overhead
 * that comes with local_irq_save/restore if we use this_cpu_cmpxchg.
 *
 * mod_state() modifies the zone counter state through atomic per cpu
 * operations.
 *
 * Overstep mode specifies how overstep should handled:
 *     0       No overstepping
 *     1       Overstepping half of threshold
 *     -1      Overstepping minus half of threshold
*/
404 405
static inline void mod_zone_state(struct zone *zone,
       enum zone_stat_item item, long delta, int overstep_mode)
406 407 408 409 410 411 412 413 414 415 416
{
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
	long o, n, t, z;

	do {
		z = 0;  /* overflow to zone counters */

		/*
		 * The fetching of the stat_threshold is racy. We may apply
		 * a counter threshold to the wrong the cpu if we get
417 418 419 420 421 422
		 * rescheduled while executing here. However, the next
		 * counter update will apply the threshold again and
		 * therefore bring the counter under the threshold again.
		 *
		 * Most of the time the thresholds are the same anyways
		 * for all cpus in a zone.
423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
		 */
		t = this_cpu_read(pcp->stat_threshold);

		o = this_cpu_read(*p);
		n = delta + o;

		if (n > t || n < -t) {
			int os = overstep_mode * (t >> 1) ;

			/* Overflow must be added to zone counters */
			z = n + os;
			n = -os;
		}
	} while (this_cpu_cmpxchg(*p, o, n) != o);

	if (z)
		zone_page_state_add(z, zone, item);
}

void mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
443
			 long delta)
444
{
445
	mod_zone_state(zone, item, delta, 0);
446 447 448 449 450
}
EXPORT_SYMBOL(mod_zone_page_state);

void inc_zone_page_state(struct page *page, enum zone_stat_item item)
{
451
	mod_zone_state(page_zone(page), item, 1, 1);
452 453 454 455 456
}
EXPORT_SYMBOL(inc_zone_page_state);

void dec_zone_page_state(struct page *page, enum zone_stat_item item)
{
457
	mod_zone_state(page_zone(page), item, -1, -1);
458 459
}
EXPORT_SYMBOL(dec_zone_page_state);
460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521

static inline void mod_node_state(struct pglist_data *pgdat,
       enum node_stat_item item, int delta, int overstep_mode)
{
	struct per_cpu_nodestat __percpu *pcp = pgdat->per_cpu_nodestats;
	s8 __percpu *p = pcp->vm_node_stat_diff + item;
	long o, n, t, z;

	do {
		z = 0;  /* overflow to node counters */

		/*
		 * The fetching of the stat_threshold is racy. We may apply
		 * a counter threshold to the wrong the cpu if we get
		 * rescheduled while executing here. However, the next
		 * counter update will apply the threshold again and
		 * therefore bring the counter under the threshold again.
		 *
		 * Most of the time the thresholds are the same anyways
		 * for all cpus in a node.
		 */
		t = this_cpu_read(pcp->stat_threshold);

		o = this_cpu_read(*p);
		n = delta + o;

		if (n > t || n < -t) {
			int os = overstep_mode * (t >> 1) ;

			/* Overflow must be added to node counters */
			z = n + os;
			n = -os;
		}
	} while (this_cpu_cmpxchg(*p, o, n) != o);

	if (z)
		node_page_state_add(z, pgdat, item);
}

void mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
					long delta)
{
	mod_node_state(pgdat, item, delta, 0);
}
EXPORT_SYMBOL(mod_node_page_state);

void inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
{
	mod_node_state(pgdat, item, 1, 1);
}

void inc_node_page_state(struct page *page, enum node_stat_item item)
{
	mod_node_state(page_pgdat(page), item, 1, 1);
}
EXPORT_SYMBOL(inc_node_page_state);

void dec_node_page_state(struct page *page, enum node_stat_item item)
{
	mod_node_state(page_pgdat(page), item, -1, -1);
}
EXPORT_SYMBOL(dec_node_page_state);
522 523 524 525 526
#else
/*
 * Use interrupt disable to serialize counter updates
 */
void mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
527
			 long delta)
528 529 530 531 532 533 534 535 536
{
	unsigned long flags;

	local_irq_save(flags);
	__mod_zone_page_state(zone, item, delta);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(mod_zone_page_state);

537 538 539 540 541 542 543
void inc_zone_page_state(struct page *page, enum zone_stat_item item)
{
	unsigned long flags;
	struct zone *zone;

	zone = page_zone(page);
	local_irq_save(flags);
544
	__inc_zone_state(zone, item);
545 546 547 548 549 550 551 552 553
	local_irq_restore(flags);
}
EXPORT_SYMBOL(inc_zone_page_state);

void dec_zone_page_state(struct page *page, enum zone_stat_item item)
{
	unsigned long flags;

	local_irq_save(flags);
554
	__dec_zone_page_state(page, item);
555 556 557 558
	local_irq_restore(flags);
}
EXPORT_SYMBOL(dec_zone_page_state);

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
void inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
{
	unsigned long flags;

	local_irq_save(flags);
	__inc_node_state(pgdat, item);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(inc_node_state);

void mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
					long delta)
{
	unsigned long flags;

	local_irq_save(flags);
	__mod_node_page_state(pgdat, item, delta);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(mod_node_page_state);

void inc_node_page_state(struct page *page, enum node_stat_item item)
{
	unsigned long flags;
	struct pglist_data *pgdat;

	pgdat = page_pgdat(page);
	local_irq_save(flags);
	__inc_node_state(pgdat, item);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(inc_node_page_state);

void dec_node_page_state(struct page *page, enum node_stat_item item)
{
	unsigned long flags;

	local_irq_save(flags);
	__dec_node_page_state(page, item);
	local_irq_restore(flags);
}
EXPORT_SYMBOL(dec_node_page_state);
#endif
602 603 604 605 606

/*
 * Fold a differential into the global counters.
 * Returns the number of counters updated.
 */
607
static int fold_diff(int *zone_diff, int *node_diff)
Christoph Lameter's avatar
Christoph Lameter committed
608 609
{
	int i;
610
	int changes = 0;
Christoph Lameter's avatar
Christoph Lameter committed
611 612

	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
613 614 615 616 617 618 619 620
		if (zone_diff[i]) {
			atomic_long_add(zone_diff[i], &vm_zone_stat[i]);
			changes++;
	}

	for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
		if (node_diff[i]) {
			atomic_long_add(node_diff[i], &vm_node_stat[i]);
621 622 623
			changes++;
	}
	return changes;
Christoph Lameter's avatar
Christoph Lameter committed
624 625
}

626
/*
627
 * Update the zone counters for the current cpu.
628
 *
629 630 631 632 633 634 635 636 637 638
 * Note that refresh_cpu_vm_stats strives to only access
 * node local memory. The per cpu pagesets on remote zones are placed
 * in the memory local to the processor using that pageset. So the
 * loop over all zones will access a series of cachelines local to
 * the processor.
 *
 * The call to zone_page_state_add updates the cachelines with the
 * statistics in the remote zone struct as well as the global cachelines
 * with the global counters. These could cause remote node cache line
 * bouncing and will have to be only done when necessary.
639 640
 *
 * The function returns the number of global counters updated.
641
 */
642
static int refresh_cpu_vm_stats(bool do_pagesets)
643
{
644
	struct pglist_data *pgdat;
645 646
	struct zone *zone;
	int i;
647 648
	int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, };
	int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, };
649
	int changes = 0;
650

651
	for_each_populated_zone(zone) {
652
		struct per_cpu_pageset __percpu *p = zone->pageset;
653

654 655
		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) {
			int v;
656

657 658
			v = this_cpu_xchg(p->vm_stat_diff[i], 0);
			if (v) {
659 660

				atomic_long_add(v, &zone->vm_stat[i]);
661
				global_zone_diff[i] += v;
662 663
#ifdef CONFIG_NUMA
				/* 3 seconds idle till flush */
664
				__this_cpu_write(p->expire, 3);
665
#endif
666
			}
667
		}
668
#ifdef CONFIG_NUMA
669 670 671 672 673 674 675 676 677 678
		if (do_pagesets) {
			cond_resched();
			/*
			 * Deal with draining the remote pageset of this
			 * processor
			 *
			 * Check if there are pages remaining in this pageset
			 * if not then there is nothing to expire.
			 */
			if (!__this_cpu_read(p->expire) ||
679
			       !__this_cpu_read(p->pcp.count))
680
				continue;
681

682 683 684 685 686 687 688
			/*
			 * We never drain zones local to this processor.
			 */
			if (zone_to_nid(zone) == numa_node_id()) {
				__this_cpu_write(p->expire, 0);
				continue;
			}
689

690 691
			if (__this_cpu_dec_return(p->expire))
				continue;
692

693 694 695 696
			if (__this_cpu_read(p->pcp.count)) {
				drain_zone_pages(zone, this_cpu_ptr(&p->pcp));
				changes++;
			}
697
		}
698
#endif
699
	}
700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715

	for_each_online_pgdat(pgdat) {
		struct per_cpu_nodestat __percpu *p = pgdat->per_cpu_nodestats;

		for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
			int v;

			v = this_cpu_xchg(p->vm_node_stat_diff[i], 0);
			if (v) {
				atomic_long_add(v, &pgdat->vm_stat[i]);
				global_node_diff[i] += v;
			}
		}
	}

	changes += fold_diff(global_zone_diff, global_node_diff);
716
	return changes;
717 718
}

719 720 721 722 723 724 725
/*
 * Fold the data for an offline cpu into the global array.
 * There cannot be any access by the offline cpu and therefore
 * synchronization is simplified.
 */
void cpu_vm_stats_fold(int cpu)
{
726
	struct pglist_data *pgdat;
727 728
	struct zone *zone;
	int i;
729 730
	int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, };
	int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, };
731 732 733 734 735 736 737 738 739 740 741 742 743

	for_each_populated_zone(zone) {
		struct per_cpu_pageset *p;

		p = per_cpu_ptr(zone->pageset, cpu);

		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
			if (p->vm_stat_diff[i]) {
				int v;

				v = p->vm_stat_diff[i];
				p->vm_stat_diff[i] = 0;
				atomic_long_add(v, &zone->vm_stat[i]);
744
				global_zone_diff[i] += v;
745 746 747
			}
	}

748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764
	for_each_online_pgdat(pgdat) {
		struct per_cpu_nodestat *p;

		p = per_cpu_ptr(pgdat->per_cpu_nodestats, cpu);

		for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
			if (p->vm_node_stat_diff[i]) {
				int v;

				v = p->vm_node_stat_diff[i];
				p->vm_node_stat_diff[i] = 0;
				atomic_long_add(v, &pgdat->vm_stat[i]);
				global_node_diff[i] += v;
			}
	}

	fold_diff(global_zone_diff, global_node_diff);
765 766
}

767 768 769 770
/*
 * this is only called if !populated_zone(zone), which implies no other users of
 * pset->vm_stat_diff[] exsist.
 */
771 772 773 774 775 776 777 778 779
void drain_zonestat(struct zone *zone, struct per_cpu_pageset *pset)
{
	int i;

	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
		if (pset->vm_stat_diff[i]) {
			int v = pset->vm_stat_diff[i];
			pset->vm_stat_diff[i] = 0;
			atomic_long_add(v, &zone->vm_stat[i]);
780
			atomic_long_add(v, &vm_zone_stat[i]);
781 782
		}
}
783 784
#endif

785
#ifdef CONFIG_NUMA
786
/*
787 788 789
 * Determine the per node value of a stat item. This function
 * is called frequently in a NUMA machine, so try to be as
 * frugal as possible.
790
 */
791 792
unsigned long sum_zone_node_page_state(int node,
				 enum zone_stat_item item)
793 794
{
	struct zone *zones = NODE_DATA(node)->node_zones;
795 796
	int i;
	unsigned long count = 0;
797

798 799 800 801
	for (i = 0; i < MAX_NR_ZONES; i++)
		count += zone_page_state(zones + i, item);

	return count;
802 803
}

804 805 806 807 808 809 810 811 812 813 814 815 816
/*
 * Determine the per node value of a stat item.
 */
unsigned long node_page_state(struct pglist_data *pgdat,
				enum node_stat_item item)
{
	long x = atomic_long_read(&pgdat->vm_stat[item]);
#ifdef CONFIG_SMP
	if (x < 0)
		x = 0;
#endif
	return x;
}
817 818
#endif

819
#ifdef CONFIG_COMPACTION
820

821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860
struct contig_page_info {
	unsigned long free_pages;
	unsigned long free_blocks_total;
	unsigned long free_blocks_suitable;
};

/*
 * Calculate the number of free pages in a zone, how many contiguous
 * pages are free and how many are large enough to satisfy an allocation of
 * the target size. Note that this function makes no attempt to estimate
 * how many suitable free blocks there *might* be if MOVABLE pages were
 * migrated. Calculating that is possible, but expensive and can be
 * figured out from userspace
 */
static void fill_contig_page_info(struct zone *zone,
				unsigned int suitable_order,
				struct contig_page_info *info)
{
	unsigned int order;

	info->free_pages = 0;
	info->free_blocks_total = 0;
	info->free_blocks_suitable = 0;

	for (order = 0; order < MAX_ORDER; order++) {
		unsigned long blocks;

		/* Count number of free blocks */
		blocks = zone->free_area[order].nr_free;
		info->free_blocks_total += blocks;

		/* Count free base pages */
		info->free_pages += blocks << order;

		/* Count the suitable free blocks */
		if (order >= suitable_order)
			info->free_blocks_suitable += blocks <<
						(order - suitable_order);
	}
}
861 862 863 864 865 866 867 868

/*
 * A fragmentation index only makes sense if an allocation of a requested
 * size would fail. If that is true, the fragmentation index indicates
 * whether external fragmentation or a lack of memory was the problem.
 * The value can be used to determine if page reclaim or compaction
 * should be used
 */
869
static int __fragmentation_index(unsigned int order, struct contig_page_info *info)
870 871 872
{
	unsigned long requested = 1UL << order;

873 874 875
	if (WARN_ON_ONCE(order >= MAX_ORDER))
		return 0;

876 877 878 879 880 881 882 883 884 885 886 887 888 889 890
	if (!info->free_blocks_total)
		return 0;

	/* Fragmentation index only makes sense when a request would fail */
	if (info->free_blocks_suitable)
		return -1000;

	/*
	 * Index is between 0 and 1 so return within 3 decimal places
	 *
	 * 0 => allocation would fail due to lack of memory
	 * 1 => allocation would fail due to fragmentation
	 */
	return 1000 - div_u64( (1000+(div_u64(info->free_pages * 1000ULL, requested))), info->free_blocks_total);
}
891 892 893 894 895 896 897 898 899

/* Same as __fragmentation index but allocs contig_page_info on stack */
int fragmentation_index(struct zone *zone, unsigned int order)
{
	struct contig_page_info info;

	fill_contig_page_info(zone, order, &info);
	return __fragmentation_index(order, &info);
}
900 901
#endif

902
#if defined(CONFIG_PROC_FS) || defined(CONFIG_SYSFS) || defined(CONFIG_NUMA)
903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
#ifdef CONFIG_ZONE_DMA
#define TEXT_FOR_DMA(xx) xx "_dma",
#else
#define TEXT_FOR_DMA(xx)
#endif

#ifdef CONFIG_ZONE_DMA32
#define TEXT_FOR_DMA32(xx) xx "_dma32",
#else
#define TEXT_FOR_DMA32(xx)
#endif

#ifdef CONFIG_HIGHMEM
#define TEXT_FOR_HIGHMEM(xx) xx "_high",
#else
#define TEXT_FOR_HIGHMEM(xx)
#endif

#define TEXTS_FOR_ZONES(xx) TEXT_FOR_DMA(xx) TEXT_FOR_DMA32(xx) xx "_normal", \
					TEXT_FOR_HIGHMEM(xx) xx "_movable",

const char * const vmstat_text[] = {
925
	/* enum zone_stat_item countes */
926
	"nr_free_pages",
Minchan Kim's avatar
Minchan Kim committed
927 928 929 930 931
	"nr_zone_inactive_anon",
	"nr_zone_active_anon",
	"nr_zone_inactive_file",
	"nr_zone_active_file",
	"nr_zone_unevictable",
932
	"nr_zone_write_pending",
933 934 935 936
	"nr_mlock",
	"nr_page_table_pages",
	"nr_kernel_stack",
	"nr_bounce",
Minchan Kim's avatar
Minchan Kim committed
937 938 939
#if IS_ENABLED(CONFIG_ZSMALLOC)
	"nr_zspages",
#endif
940 941 942 943 944 945 946 947
#ifdef CONFIG_NUMA
	"numa_hit",
	"numa_miss",
	"numa_foreign",
	"numa_interleave",
	"numa_local",
	"numa_other",
#endif
948
	"nr_free_cma",
949

950 951 952 953 954 955
	/* Node-based counters */
	"nr_inactive_anon",
	"nr_active_anon",
	"nr_inactive_file",
	"nr_active_file",
	"nr_unevictable",
956 957
	"nr_slab_reclaimable",
	"nr_slab_unreclaimable",
958 959
	"nr_isolated_anon",
	"nr_isolated_file",
960 961 962
	"workingset_refault",
	"workingset_activate",
	"workingset_nodereclaim",
963 964
	"nr_anon_pages",
	"nr_mapped",
965 966 967 968 969 970 971 972 973
	"nr_file_pages",
	"nr_dirty",
	"nr_writeback",
	"nr_writeback_temp",
	"nr_shmem",
	"nr_shmem_hugepages",
	"nr_shmem_pmdmapped",
	"nr_anon_transparent_hugepages",
	"nr_unstable",
974 975 976 977
	"nr_vmscan_write",
	"nr_vmscan_immediate_reclaim",
	"nr_dirtied",
	"nr_written",
978

979
	/* enum writeback_stat_item counters */
980 981 982 983
	"nr_dirty_threshold",
	"nr_dirty_background_threshold",

#ifdef CONFIG_VM_EVENT_COUNTERS
984
	/* enum vm_event_item counters */
985 986 987 988 989 990
	"pgpgin",
	"pgpgout",
	"pswpin",
	"pswpout",

	TEXTS_FOR_ZONES("pgalloc")
991 992
	TEXTS_FOR_ZONES("allocstall")
	TEXTS_FOR_ZONES("pgskip")
993 994 995 996

	"pgfree",
	"pgactivate",
	"pgdeactivate",
997
	"pglazyfree",
998 999 1000

	"pgfault",
	"pgmajfault",
Minchan Kim's avatar
Minchan Kim committed
1001
	"pglazyfreed",
1002

1003 1004 1005 1006 1007
	"pgrefill",
	"pgsteal_kswapd",
	"pgsteal_direct",
	"pgscan_kswapd",
	"pgscan_direct",
1008
	"pgscan_direct_throttle",
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021

#ifdef CONFIG_NUMA
	"zone_reclaim_failed",
#endif
	"pginodesteal",
	"slabs_scanned",
	"kswapd_inodesteal",
	"kswapd_low_wmark_hit_quickly",
	"kswapd_high_wmark_hit_quickly",
	"pageoutrun",

	"pgrotated",

1022 1023
	"drop_pagecache",
	"drop_slab",
1024
	"oom_kill",
1025

1026 1027
#ifdef CONFIG_NUMA_BALANCING
	"numa_pte_updates",
1028
	"numa_huge_pte_updates",
1029 1030 1031 1032
	"numa_hint_faults",
	"numa_hint_faults_local",
	"numa_pages_migrated",
#endif
1033 1034 1035 1036
#ifdef CONFIG_MIGRATION
	"pgmigrate_success",
	"pgmigrate_fail",
#endif
1037
#ifdef CONFIG_COMPACTION
1038 1039 1040
	"compact_migrate_scanned",
	"compact_free_scanned",
	"compact_isolated",
1041 1042 1043
	"compact_stall",
	"compact_fail",
	"compact_success",
1044
	"compact_daemon_wake",
1045 1046
	"compact_daemon_migrate_scanned",
	"compact_daemon_free_scanned",
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065
#endif

#ifdef CONFIG_HUGETLB_PAGE
	"htlb_buddy_alloc_success",
	"htlb_buddy_alloc_fail",
#endif
	"unevictable_pgs_culled",
	"unevictable_pgs_scanned",
	"unevictable_pgs_rescued",
	"unevictable_pgs_mlocked",
	"unevictable_pgs_munlocked",
	"unevictable_pgs_cleared",
	"unevictable_pgs_stranded",

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	"thp_fault_alloc",
	"thp_fault_fallback",
	"thp_collapse_alloc",
	"thp_collapse_alloc_failed",
1066 1067
	"thp_file_alloc",
	"thp_file_mapped",
1068 1069
	"thp_split_page",
	"thp_split_page_failed",
1070
	"thp_deferred_split_page",
1071
	"thp_split_pmd",
1072 1073 1074
#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
	"thp_split_pud",
#endif
1075 1076
	"thp_zero_page_alloc",
	"thp_zero_page_alloc_failed",
1077
	"thp_swpout",
1078
	"thp_swpout_fallback",
1079
#endif
1080 1081 1082 1083 1084 1085 1086
#ifdef CONFIG_MEMORY_BALLOON
	"balloon_inflate",
	"balloon_deflate",
#ifdef CONFIG_BALLOON_COMPACTION
	"balloon_migrate",
#endif
#endif /* CONFIG_MEMORY_BALLOON */
1087
#ifdef CONFIG_DEBUG_TLBFLUSH
1088
#ifdef CONFIG_SMP
Dave Hansen's avatar
Dave Hansen committed
1089 1090
	"nr_tlb_remote_flush",
	"nr_tlb_remote_flush_received",
1091
#endif /* CONFIG_SMP */
Dave Hansen's avatar
Dave Hansen committed
1092 1093
	"nr_tlb_local_flush_all",
	"nr_tlb_local_flush_one",
1094
#endif /* CONFIG_DEBUG_TLBFLUSH */
1095

Davidlohr Bueso's avatar
Davidlohr Bueso committed
1096 1097 1098
#ifdef CONFIG_DEBUG_VM_VMACACHE
	"vmacache_find_calls",
	"vmacache_find_hits",
1099
	"vmacache_full_flushes",
Davidlohr Bueso's avatar
Davidlohr Bueso committed
1100
#endif
1101 1102
#endif /* CONFIG_VM_EVENTS_COUNTERS */
};
1103
#endif /* CONFIG_PROC_FS || CONFIG_SYSFS || CONFIG_NUMA */
1104 1105


1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
#if (defined(CONFIG_DEBUG_FS) && defined(CONFIG_COMPACTION)) || \
     defined(CONFIG_PROC_FS)
static void *frag_start(struct seq_file *m, loff_t *pos)
{
	pg_data_t *pgdat;
	loff_t node = *pos;

	for (pgdat = first_online_pgdat();
	     pgdat && node;
	     pgdat = next_online_pgdat(pgdat))
		--node;

	return pgdat;
}

static void *frag_next(struct seq_file *m, void *arg, loff_t *pos)
{
	pg_data_t *pgdat = (pg_data_t *)arg;

	(*pos)++;
	return next_online_pgdat(pgdat);
}

static void frag_stop(struct seq_file *m, void *arg)
{
}

1133 1134 1135 1136
/*
 * Walk zones in a node and print using a callback.
 * If @assert_populated is true, only use callback for zones that are populated.
 */
1137
static void walk_zones_in_node(struct seq_file *m, pg_data_t *pgdat,
1138
		bool assert_populated, bool nolock,
1139 1140 1141 1142 1143 1144 1145
		void (*print)(struct seq_file *m, pg_data_t *, struct zone *))
{
	struct zone *zone;
	struct zone *node_zones = pgdat->node_zones;
	unsigned long flags;

	for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
1146
		if (assert_populated && !populated_zone(zone))
1147 1148
			continue;

1149 1150
		if (!nolock)
			spin_lock_irqsave(&zone->lock, flags);
1151
		print(m, pgdat, zone);
1152 1153
		if (!nolock)
			spin_unlock_irqrestore(&zone->lock, flags);
1154 1155 1156 1157
	}
}
#endif

1158
#ifdef CONFIG_PROC_FS
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
static void frag_show_print(struct seq_file *m, pg_data_t *pgdat,
						struct zone *zone)
{
	int order;

	seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name);
	for (order = 0; order < MAX_ORDER; ++order)
		seq_printf(m, "%6lu ", zone->free_area[order].nr_free);
	seq_putc(m, '\n');
}

/*
 * This walks the free areas for each zone.
 */
static int frag_show(struct seq_file *m, void *arg)
{
	pg_data_t *pgdat = (pg_data_t *)arg;
1176
	walk_zones_in_node(m, pgdat, true, false, frag_show_print);
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
	return 0;
}

static void pagetypeinfo_showfree_print(struct seq_file *m,
					pg_data_t *pgdat, struct zone *zone)
{
	int order, mtype;

	for (mtype = 0; mtype < MIGRATE_TYPES; mtype++) {
		seq_printf(m, "Node %4d, zone %8s, type %12s ",
					pgdat->node_id,
					zone->name,
					migratetype_names[mtype]);
		for (order = 0; order < MAX_ORDER; ++order) {
			unsigned long freecount = 0;
			struct free_area *area;
			struct list_head *curr;

			area = &(zone->free_area[order]);

			list_for_each(curr, &area->free_list[mtype])
				freecount++;
			seq_printf(m, "%6lu ", freecount);
		}
1201 1202
		seq_putc(m, '\n');
	}
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
}

/* Print out the free pages at each order for each migatetype */
static int pagetypeinfo_showfree(struct seq_file *m, void *arg)
{
	int order;
	pg_data_t *pgdat = (pg_data_t *)arg;

	/* Print header */
	seq_printf(m, "%-43s ", "Free pages count per migrate type at order");
	for (order = 0; order < MAX_ORDER; ++order)
		seq_printf(m, "%6d ", order);
	seq_putc(m, '\n');

1217
	walk_zones_in_node(m, pgdat, true, false, pagetypeinfo_showfree_print);
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227

	return 0;
}

static void pagetypeinfo_showblockcount_print(struct seq_file *m,
					pg_data_t *pgdat, struct zone *zone)
{
	int mtype;
	unsigned long pfn;
	unsigned long start_pfn = zone->zone_start_pfn;
1228
	unsigned long end_pfn = zone_end_pfn(zone);
1229 1230 1231 1232 1233
	unsigned long count[MIGRATE_TYPES] = { 0, };

	for (pfn = start_pfn; pfn < end_pfn; pfn += pageblock_nr_pages) {
		struct page *page;

1234 1235
		page = pfn_to_online_page(pfn);
		if (!page)
1236 1237
			continue;

1238 1239
		/* Watch for unexpected holes punched in the memmap */
		if (!memmap_valid_within(pfn, page, zone))
1240
			continue;
1241

1242 1243 1244
		if (page_zone(page) != zone)
			continue;

1245 1246
		mtype = get_pageblock_migratetype(page);

1247 1248
		if (mtype < MIGRATE_TYPES)
			count[mtype]++;
1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
	}

	/* Print counts */
	seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name);
	for (mtype = 0; mtype < MIGRATE_TYPES; mtype++)
		seq_printf(m, "%12lu ", count[mtype]);
	seq_putc(m, '\n');
}

/* Print out the free pages at each order for each migratetype */
static int pagetypeinfo_showblockcount(struct seq_file *m, void *arg)
{
	int mtype;
	pg_data_t *pgdat = (pg_data_t *)arg;

	seq_printf(m, "\n%-23s", "Number of blocks type ");
	for (mtype = 0; mtype < MIGRATE_TYPES; mtype++)
		seq_printf(m, "%12s ", migratetype_names[mtype]);
	seq_putc(m, '\n');
1268 1269
	walk_zones_in_node(m, pgdat, true, false,
		pagetypeinfo_showblockcount_print);
1270 1271 1272 1273

	return 0;
}

1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284
/*
 * Print out the number of pageblocks for each migratetype that contain pages
 * of other types. This gives an indication of how well fallbacks are being
 * contained by rmqueue_fallback(). It requires information from PAGE_OWNER
 * to determine what is going on
 */
static void pagetypeinfo_showmixedcount(struct seq_file *m, pg_data_t *pgdat)
{
#ifdef CONFIG_PAGE_OWNER
	int mtype;

1285
	if (!static_branch_unlikely(&page_owner_inited))
1286 1287 1288 1289 1290 1291 1292 1293 1294
		return;

	drain_all_pages(NULL);

	seq_printf(m, "\n%-23s", "Number of mixed blocks ");
	for (mtype = 0; mtype < MIGRATE_TYPES; mtype++)
		seq_printf(m, "%12s ", migratetype_names[mtype]);
	seq_putc(m, '\n');

1295 1296
	walk_zones_in_node(m, pgdat, true, true,
		pagetypeinfo_showmixedcount_print);
1297 1298 1299
#endif /* CONFIG_PAGE_OWNER */
}

1300 1301 1302 1303 1304 1305 1306 1307
/*
 * This prints out statistics in relation to grouping pages by mobility.
 * It is expensive to collect so do not constantly read the file.
 */
static int pagetypeinfo_show(struct seq_file *m, void *arg)
{
	pg_data_t *pgdat = (pg_data_t *)arg;

1308
	/* check memoryless node */
1309
	if (!node_state(pgdat->node_id, N_MEMORY))