vmstat.c 51.2 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
atomic_long_t vm_zone_stat[NR_VM_ZONE_STAT_ITEMS] __cacheline_aligned_in_smp;
90
atomic_long_t vm_numa_stat[NR_VM_NUMA_STAT_ITEMS] __cacheline_aligned_in_smp;
91 92
atomic_long_t vm_node_stat[NR_VM_NODE_STAT_ITEMS] __cacheline_aligned_in_smp;
EXPORT_SYMBOL(vm_zone_stat);
93
EXPORT_SYMBOL(vm_numa_stat);
94
EXPORT_SYMBOL(vm_node_stat);
95 96 97

#ifdef CONFIG_SMP

98
int calculate_pressure_threshold(struct zone *zone)
99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
{
	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;
}

122
int calculate_normal_threshold(struct zone *zone)
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
{
	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
	 */

157
	mem = zone->managed_pages >> (27 - PAGE_SHIFT);
158 159 160 161 162 163 164 165 166 167

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

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

	return threshold;
}
168 169

/*
170
 * Refresh the thresholds for each zone.
171
 */
172
void refresh_zone_stat_thresholds(void)
173
{
174
	struct pglist_data *pgdat;
175 176 177 178
	struct zone *zone;
	int cpu;
	int threshold;

179 180 181 182 183 184 185
	/* 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;
		}
	}

186
	for_each_populated_zone(zone) {
187
		struct pglist_data *pgdat = zone->zone_pgdat;
188 189
		unsigned long max_drift, tolerate_drift;

190
		threshold = calculate_normal_threshold(zone);
191

192 193 194
		for_each_online_cpu(cpu) {
			int pgdat_threshold;

195 196
			per_cpu_ptr(zone->pageset, cpu)->stat_threshold
							= threshold;
197 198 199 200
#ifdef CONFIG_NUMA
			per_cpu_ptr(zone->pageset, cpu)->numa_stat_threshold
							= threshold;
#endif
201 202 203 204 205 206
			/* 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);
		}

207 208 209 210 211 212 213 214 215 216
		/*
		 * 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;
217
	}
218 219
}

220 221
void set_pgdat_percpu_threshold(pg_data_t *pgdat,
				int (*calculate_pressure)(struct zone *))
222 223 224 225 226 227 228 229 230 231 232
{
	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;

233
		threshold = (*calculate_pressure)(zone);
234
		for_each_online_cpu(cpu) {
235 236
			per_cpu_ptr(zone->pageset, cpu)->stat_threshold
							= threshold;
237 238 239 240 241
#ifdef CONFIG_NUMA
			per_cpu_ptr(zone->pageset, cpu)->numa_stat_threshold
							= threshold;
#endif
		}
242 243 244
	}
}

245
/*
246 247 248
 * 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.
249 250
 */
void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
251
			   long delta)
252
{
253 254
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
255
	long x;
256 257 258
	long t;

	x = delta + __this_cpu_read(*p);
259

260
	t = __this_cpu_read(pcp->stat_threshold);
261

262
	if (unlikely(x > t || x < -t)) {
263 264 265
		zone_page_state_add(x, zone, item);
		x = 0;
	}
266
	__this_cpu_write(*p, x);
267 268 269
}
EXPORT_SYMBOL(__mod_zone_page_state);

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
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);

290 291 292 293 294 295 296 297 298 299 300 301 302
/*
 * 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.
 *
303 304 305
 * NOTE: These functions are very performance sensitive. Change only
 * with care.
 *
306 307 308 309 310 311 312
 * 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.
 */
313
void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
314
{
315 316 317
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
	s8 v, t;
318

319
	v = __this_cpu_inc_return(*p);
320 321 322
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v > t)) {
		s8 overstep = t >> 1;
323

324 325
		zone_page_state_add(v + overstep, zone, item);
		__this_cpu_write(*p, -overstep);
326 327
	}
}
328

329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344
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);
	}
}

345 346 347 348
void __inc_zone_page_state(struct page *page, enum zone_stat_item item)
{
	__inc_zone_state(page_zone(page), item);
}
349 350
EXPORT_SYMBOL(__inc_zone_page_state);

351 352 353 354 355 356
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);

357
void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
358
{
359 360 361
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_stat_diff + item;
	s8 v, t;
362

363
	v = __this_cpu_dec_return(*p);
364 365 366
	t = __this_cpu_read(pcp->stat_threshold);
	if (unlikely(v < - t)) {
		s8 overstep = t >> 1;
367

368 369
		zone_page_state_add(v - overstep, zone, item);
		__this_cpu_write(*p, overstep);
370 371
	}
}
372

373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
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);
	}
}

389 390 391 392
void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
{
	__dec_zone_state(page_zone(page), item);
}
393 394
EXPORT_SYMBOL(__dec_zone_page_state);

395 396 397 398 399 400
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);

401
#ifdef CONFIG_HAVE_CMPXCHG_LOCAL
402 403 404 405 406 407 408 409 410 411 412 413
/*
 * 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
*/
414 415
static inline void mod_zone_state(struct zone *zone,
       enum zone_stat_item item, long delta, int overstep_mode)
416 417 418 419 420 421 422 423 424 425 426
{
	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
427 428 429 430 431 432
		 * 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.
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
		 */
		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,
453
			 long delta)
454
{
455
	mod_zone_state(zone, item, delta, 0);
456 457 458 459 460
}
EXPORT_SYMBOL(mod_zone_page_state);

void inc_zone_page_state(struct page *page, enum zone_stat_item item)
{
461
	mod_zone_state(page_zone(page), item, 1, 1);
462 463 464 465 466
}
EXPORT_SYMBOL(inc_zone_page_state);

void dec_zone_page_state(struct page *page, enum zone_stat_item item)
{
467
	mod_zone_state(page_zone(page), item, -1, -1);
468 469
}
EXPORT_SYMBOL(dec_zone_page_state);
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531

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);
532 533 534 535 536
#else
/*
 * Use interrupt disable to serialize counter updates
 */
void mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
537
			 long delta)
538 539 540 541 542 543 544 545 546
{
	unsigned long flags;

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

547 548 549 550 551 552 553
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);
554
	__inc_zone_state(zone, item);
555 556 557 558 559 560 561 562 563
	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);
564
	__dec_zone_page_state(page, item);
565 566 567 568
	local_irq_restore(flags);
}
EXPORT_SYMBOL(dec_zone_page_state);

569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611
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
612 613 614 615 616

/*
 * Fold a differential into the global counters.
 * Returns the number of counters updated.
 */
617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
#ifdef CONFIG_NUMA
static int fold_diff(int *zone_diff, int *numa_diff, int *node_diff)
{
	int i;
	int changes = 0;

	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
		if (zone_diff[i]) {
			atomic_long_add(zone_diff[i], &vm_zone_stat[i]);
			changes++;
	}

	for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++)
		if (numa_diff[i]) {
			atomic_long_add(numa_diff[i], &vm_numa_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]);
			changes++;
	}
	return changes;
}
#else
643
static int fold_diff(int *zone_diff, int *node_diff)
Christoph Lameter's avatar
Christoph Lameter committed
644 645
{
	int i;
646
	int changes = 0;
Christoph Lameter's avatar
Christoph Lameter committed
647 648

	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
649 650 651 652 653 654 655 656
		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]);
657 658 659
			changes++;
	}
	return changes;
Christoph Lameter's avatar
Christoph Lameter committed
660
}
661
#endif /* CONFIG_NUMA */
Christoph Lameter's avatar
Christoph Lameter committed
662

663
/*
664
 * Update the zone counters for the current cpu.
665
 *
666 667 668 669 670 671 672 673 674 675
 * 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.
676 677
 *
 * The function returns the number of global counters updated.
678
 */
679
static int refresh_cpu_vm_stats(bool do_pagesets)
680
{
681
	struct pglist_data *pgdat;
682 683
	struct zone *zone;
	int i;
684
	int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, };
685 686 687
#ifdef CONFIG_NUMA
	int global_numa_diff[NR_VM_NUMA_STAT_ITEMS] = { 0, };
#endif
688
	int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, };
689
	int changes = 0;
690

691
	for_each_populated_zone(zone) {
692
		struct per_cpu_pageset __percpu *p = zone->pageset;
693

694 695
		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) {
			int v;
696

697 698
			v = this_cpu_xchg(p->vm_stat_diff[i], 0);
			if (v) {
699 700

				atomic_long_add(v, &zone->vm_stat[i]);
701
				global_zone_diff[i] += v;
702 703
#ifdef CONFIG_NUMA
				/* 3 seconds idle till flush */
704
				__this_cpu_write(p->expire, 3);
705
#endif
706
			}
707
		}
708
#ifdef CONFIG_NUMA
709 710 711 712 713 714 715 716 717 718 719 720
		for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++) {
			int v;

			v = this_cpu_xchg(p->vm_numa_stat_diff[i], 0);
			if (v) {

				atomic_long_add(v, &zone->vm_numa_stat[i]);
				global_numa_diff[i] += v;
				__this_cpu_write(p->expire, 3);
			}
		}

721 722 723 724 725 726 727 728 729 730
		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) ||
731
			       !__this_cpu_read(p->pcp.count))
732
				continue;
733

734 735 736 737 738 739 740
			/*
			 * We never drain zones local to this processor.
			 */
			if (zone_to_nid(zone) == numa_node_id()) {
				__this_cpu_write(p->expire, 0);
				continue;
			}
741

742 743
			if (__this_cpu_dec_return(p->expire))
				continue;
744

745 746 747 748
			if (__this_cpu_read(p->pcp.count)) {
				drain_zone_pages(zone, this_cpu_ptr(&p->pcp));
				changes++;
			}
749
		}
750
#endif
751
	}
752 753 754 755 756 757 758 759 760 761 762 763 764 765 766

	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;
			}
		}
	}

767 768 769 770
#ifdef CONFIG_NUMA
	changes += fold_diff(global_zone_diff, global_numa_diff,
			     global_node_diff);
#else
771
	changes += fold_diff(global_zone_diff, global_node_diff);
772
#endif
773
	return changes;
774 775
}

776 777 778 779 780 781 782
/*
 * 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)
{
783
	struct pglist_data *pgdat;
784 785
	struct zone *zone;
	int i;
786
	int global_zone_diff[NR_VM_ZONE_STAT_ITEMS] = { 0, };
787 788 789
#ifdef CONFIG_NUMA
	int global_numa_diff[NR_VM_NUMA_STAT_ITEMS] = { 0, };
#endif
790
	int global_node_diff[NR_VM_NODE_STAT_ITEMS] = { 0, };
791 792 793 794 795 796 797 798 799 800 801 802 803

	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]);
804
				global_zone_diff[i] += v;
805
			}
806 807 808 809 810 811 812 813 814 815 816 817

#ifdef CONFIG_NUMA
		for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++)
			if (p->vm_numa_stat_diff[i]) {
				int v;

				v = p->vm_numa_stat_diff[i];
				p->vm_numa_stat_diff[i] = 0;
				atomic_long_add(v, &zone->vm_numa_stat[i]);
				global_numa_diff[i] += v;
			}
#endif
818 819
	}

820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835
	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;
			}
	}

836 837 838
#ifdef CONFIG_NUMA
	fold_diff(global_zone_diff, global_numa_diff, global_node_diff);
#else
839
	fold_diff(global_zone_diff, global_node_diff);
840
#endif
841 842
}

843 844 845 846
/*
 * this is only called if !populated_zone(zone), which implies no other users of
 * pset->vm_stat_diff[] exsist.
 */
847 848 849 850 851 852 853 854 855
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]);
856
			atomic_long_add(v, &vm_zone_stat[i]);
857
		}
858 859 860 861 862 863 864 865 866 867 868

#ifdef CONFIG_NUMA
	for (i = 0; i < NR_VM_NUMA_STAT_ITEMS; i++)
		if (pset->vm_numa_stat_diff[i]) {
			int v = pset->vm_numa_stat_diff[i];

			pset->vm_numa_stat_diff[i] = 0;
			atomic_long_add(v, &zone->vm_numa_stat[i]);
			atomic_long_add(v, &vm_numa_stat[i]);
		}
#endif
869
}
870 871
#endif

872
#ifdef CONFIG_NUMA
873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889
void __inc_numa_state(struct zone *zone,
				 enum numa_stat_item item)
{
	struct per_cpu_pageset __percpu *pcp = zone->pageset;
	s8 __percpu *p = pcp->vm_numa_stat_diff + item;
	s8 v, t;

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

		zone_numa_state_add(v + overstep, zone, item);
		__this_cpu_write(*p, -overstep);
	}
}

890
/*
891 892 893
 * 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.
894
 */
895 896
unsigned long sum_zone_node_page_state(int node,
				 enum zone_stat_item item)
897 898
{
	struct zone *zones = NODE_DATA(node)->node_zones;
899 900
	int i;
	unsigned long count = 0;
901

902 903 904 905
	for (i = 0; i < MAX_NR_ZONES; i++)
		count += zone_page_state(zones + i, item);

	return count;
906 907
}

908 909 910 911 912 913 914 915 916 917 918 919 920
unsigned long sum_zone_numa_state(int node,
				 enum numa_stat_item item)
{
	struct zone *zones = NODE_DATA(node)->node_zones;
	int i;
	unsigned long count = 0;

	for (i = 0; i < MAX_NR_ZONES; i++)
		count += zone_numa_state(zones + i, item);

	return count;
}

921 922 923 924 925 926 927 928 929 930 931 932 933
/*
 * 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;
}
934 935
#endif

936
#ifdef CONFIG_COMPACTION
937

938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977
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);
	}
}
978 979 980 981 982 983 984 985

/*
 * 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
 */
986
static int __fragmentation_index(unsigned int order, struct contig_page_info *info)
987 988 989
{
	unsigned long requested = 1UL << order;

990 991 992
	if (WARN_ON_ONCE(order >= MAX_ORDER))
		return 0;

993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
	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);
}
1008 1009 1010 1011 1012 1013 1014 1015 1016

/* 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);
}
1017 1018
#endif

1019
#if defined(CONFIG_PROC_FS) || defined(CONFIG_SYSFS) || defined(CONFIG_NUMA)
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041
#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[] = {
1042
	/* enum zone_stat_item countes */
1043
	"nr_free_pages",
Minchan Kim's avatar
Minchan Kim committed
1044 1045 1046 1047 1048
	"nr_zone_inactive_anon",
	"nr_zone_active_anon",
	"nr_zone_inactive_file",
	"nr_zone_active_file",
	"nr_zone_unevictable",
1049
	"nr_zone_write_pending",
1050 1051 1052 1053
	"nr_mlock",
	"nr_page_table_pages",
	"nr_kernel_stack",
	"nr_bounce",
Minchan Kim's avatar
Minchan Kim committed
1054 1055 1056
#if IS_ENABLED(CONFIG_ZSMALLOC)
	"nr_zspages",
#endif
1057 1058 1059
	"nr_free_cma",

	/* enum numa_stat_item counters */
1060 1061 1062 1063 1064 1065 1066 1067
#ifdef CONFIG_NUMA
	"numa_hit",
	"numa_miss",
	"numa_foreign",
	"numa_interleave",
	"numa_local",
	"numa_other",
#endif
1068

1069 1070 1071 1072 1073 1074
	/* Node-based counters */
	"nr_inactive_anon",
	"nr_active_anon",
	"nr_inactive_file",
	"nr_active_file",
	"nr_unevictable",
1075 1076
	"nr_slab_reclaimable",
	"nr_slab_unreclaimable",
1077 1078
	"nr_isolated_anon",
	"nr_isolated_file",
1079 1080 1081
	"workingset_refault",
	"workingset_activate",
	"workingset_nodereclaim",
1082 1083
	"nr_anon_pages",
	"nr_mapped",
1084 1085 1086 1087 1088 1089 1090 1091 1092
	"nr_file_pages",
	"nr_dirty",
	"nr_writeback",
	"nr_writeback_temp",
	"nr_shmem",
	"nr_shmem_hugepages",
	"nr_shmem_pmdmapped",
	"nr_anon_transparent_hugepages",
	"nr_unstable",
1093 1094 1095 1096
	"nr_vmscan_write",
	"nr_vmscan_immediate_reclaim",
	"nr_dirtied",
	"nr_written",
1097

1098
	/* enum writeback_stat_item counters */
1099 1100 1101 1102
	"nr_dirty_threshold",
	"nr_dirty_background_threshold",

#ifdef CONFIG_VM_EVENT_COUNTERS
1103
	/* enum vm_event_item counters */
1104 1105 1106 1107 1108 1109
	"pgpgin",
	"pgpgout",
	"pswpin",
	"pswpout",

	TEXTS_FOR_ZONES("pgalloc")
1110 1111
	TEXTS_FOR_ZONES("allocstall")
	TEXTS_FOR_ZONES("pgskip")
1112 1113 1114 1115

	"pgfree",
	"pgactivate",
	"pgdeactivate",
1116
	"pglazyfree",
1117 1118 1119

	"pgfault",
	"pgmajfault",
Minchan Kim's avatar
Minchan Kim committed
1120
	"pglazyfreed",
1121

1122 1123 1124 1125 1126
	"pgrefill",
	"pgsteal_kswapd",
	"pgsteal_direct",
	"pgscan_kswapd",
	"pgscan_direct",
1127
	"pgscan_direct_throttle",
1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

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

	"pgrotated",

1141 1142
	"drop_pagecache",
	"drop_slab",
1143
	"oom_kill",
1144

1145 1146
#ifdef CONFIG_NUMA_BALANCING
	"numa_pte_updates",
1147
	"numa_huge_pte_updates",
1148 1149 1150 1151
	"numa_hint_faults",
	"numa_hint_faults_local",
	"numa_pages_migrated",
#endif
1152 1153 1154 1155
#ifdef CONFIG_MIGRATION
	"pgmigrate_success",
	"pgmigrate_fail",
#endif
1156
#ifdef CONFIG_COMPACTION
1157 1158 1159
	"compact_migrate_scanned",
	"compact_free_scanned",
	"compact_isolated",
1160 1161 1162
	"compact_stall",
	"compact_fail",
	"compact_success",
1163
	"compact_daemon_wake",
1164 1165
	"compact_daemon_migrate_scanned",
	"compact_daemon_free_scanned",
1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184
#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",
1185 1186
	"thp_file_alloc",
	"thp_file_mapped",
1187 1188
	"thp_split_page",
	"thp_split_page_failed",
1189
	"thp_deferred_split_page",
1190
	"thp_split_pmd",
1191 1192 1193
#ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
	"thp_split_pud",
#endif
1194 1195
	"thp_zero_page_alloc",
	"thp_zero_page_alloc_failed",
1196
	"thp_swpout",
1197
	"thp_swpout_fallback",
1198
#endif
1199 1200 1201 1202 1203 1204 1205
#ifdef CONFIG_MEMORY_BALLOON
	"balloon_inflate",
	"balloon_deflate",
#ifdef CONFIG_BALLOON_COMPACTION
	"balloon_migrate",
#endif
#endif /* CONFIG_MEMORY_BALLOON */
1206
#ifdef CONFIG_DEBUG_TLBFLUSH
1207
#ifdef CONFIG_SMP
Dave Hansen's avatar
Dave Hansen committed
1208 1209
	"nr_tlb_remote_flush",
	"nr_tlb_remote_flush_received",
1210
#endif /* CONFIG_SMP */
Dave Hansen's avatar
Dave Hansen committed
1211 1212
	"nr_tlb_local_flush_all",
	"nr_tlb_local_flush_one",
1213
#endif /* CONFIG_DEBUG_TLBFLUSH */
1214

Davidlohr Bueso's avatar
Davidlohr Bueso committed
1215 1216 1217
#ifdef CONFIG_DEBUG_VM_VMACACHE
	"vmacache_find_calls",
	"vmacache_find_hits",
1218
	"vmacache_full_flushes",
Davidlohr Bueso's avatar
Davidlohr Bueso committed
1219
#endif
1220 1221 1222 1223
#ifdef CONFIG_SWAP
	"swap_ra",
	"swap_ra_hit",
#endif
1224 1225
#endif /* CONFIG_VM_EVENTS_COUNTERS */
};
1226
#endif /* CONFIG_PROC_FS || CONFIG_SYSFS || CONFIG_NUMA */
1227

1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254
#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)
{
}

1255 1256 1257 1258
/*
 * Walk zones in a node and print using a callback.
 * If @assert_populated is true, only use callback for zones that are populated.
 */
1259
static void walk_zones_in_node(struct seq_file *m, pg_data_t *pgdat,
1260
		bool assert_populated, bool nolock,
1261 1262 1263 1264 1265 1266 1267
		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) {
1268
		if (assert_populated && !populated_zone(zone))
1269 1270
			continue;

1271 1272
		if (!nolock)
			spin_lock_irqsave(&zone->lock, flags);
1273
		print(m, pgdat, zone);
1274 1275
		if (!nolock)
			spin_unlock_irqrestore(&zone->lock, flags);
1276 1277 1278 1279
	}
}
#endif

1280
#ifdef CONFIG_PROC_FS
1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297
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;
1298
	walk_zones_in_node(m, pgdat, true, false, frag_show_print);
1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311