swap.c 19.1 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7
/*
 *  linux/mm/swap.c
 *
 *  Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
 */

/*
S
Simon Arlott 已提交
8
 * This file contains the default values for the operation of the
L
Linus Torvalds 已提交
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
 * Linux VM subsystem. Fine-tuning documentation can be found in
 * Documentation/sysctl/vm.txt.
 * Started 18.12.91
 * Swap aging added 23.2.95, Stephen Tweedie.
 * Buffermem limits added 12.3.98, Rik van Riel.
 */

#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/kernel_stat.h>
#include <linux/swap.h>
#include <linux/mman.h>
#include <linux/pagemap.h>
#include <linux/pagevec.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/mm_inline.h>
#include <linux/buffer_head.h>	/* for try_to_release_page() */
#include <linux/percpu_counter.h>
#include <linux/percpu.h>
#include <linux/cpu.h>
#include <linux/notifier.h>
P
Peter Zijlstra 已提交
31
#include <linux/backing-dev.h>
32
#include <linux/memcontrol.h>
33
#include <linux/gfp.h>
L
Linus Torvalds 已提交
34

35 36
#include "internal.h"

L
Linus Torvalds 已提交
37 38 39
/* How many pages do we try to swap or page in/out together? */
int page_cluster;

40
static DEFINE_PER_CPU(struct pagevec[NR_LRU_LISTS], lru_add_pvecs);
41
static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs);
M
Minchan Kim 已提交
42
static DEFINE_PER_CPU(struct pagevec, lru_deactivate_pvecs);
43

A
Adrian Bunk 已提交
44 45 46 47
/*
 * This path almost never happens for VM activity - pages are normally
 * freed via pagevecs.  But it gets used by networking.
 */
H
Harvey Harrison 已提交
48
static void __page_cache_release(struct page *page)
A
Adrian Bunk 已提交
49 50 51 52 53 54 55 56 57 58 59
{
	if (PageLRU(page)) {
		unsigned long flags;
		struct zone *zone = page_zone(page);

		spin_lock_irqsave(&zone->lru_lock, flags);
		VM_BUG_ON(!PageLRU(page));
		__ClearPageLRU(page);
		del_page_from_lru(zone, page);
		spin_unlock_irqrestore(&zone->lru_lock, flags);
	}
60 61 62 63 64
}

static void __put_single_page(struct page *page)
{
	__page_cache_release(page);
L
Li Hong 已提交
65
	free_hot_cold_page(page, 0);
A
Adrian Bunk 已提交
66 67
}

68
static void __put_compound_page(struct page *page)
L
Linus Torvalds 已提交
69
{
70
	compound_page_dtor *dtor;
L
Linus Torvalds 已提交
71

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
	__page_cache_release(page);
	dtor = get_compound_page_dtor(page);
	(*dtor)(page);
}

static void put_compound_page(struct page *page)
{
	if (unlikely(PageTail(page))) {
		/* __split_huge_page_refcount can run under us */
		struct page *page_head = page->first_page;
		smp_rmb();
		/*
		 * If PageTail is still set after smp_rmb() we can be sure
		 * that the page->first_page we read wasn't a dangling pointer.
		 * See __split_huge_page_refcount() smp_wmb().
		 */
		if (likely(PageTail(page) && get_page_unless_zero(page_head))) {
			unsigned long flags;
			/*
			 * Verify that our page_head wasn't converted
			 * to a a regular page before we got a
			 * reference on it.
			 */
			if (unlikely(!PageHead(page_head))) {
				/* PageHead is cleared after PageTail */
				smp_rmb();
				VM_BUG_ON(PageTail(page));
				goto out_put_head;
			}
			/*
			 * Only run compound_lock on a valid PageHead,
			 * after having it pinned with
			 * get_page_unless_zero() above.
			 */
			smp_mb();
			/* page_head wasn't a dangling pointer */
			flags = compound_lock_irqsave(page_head);
			if (unlikely(!PageTail(page))) {
				/* __split_huge_page_refcount run before us */
				compound_unlock_irqrestore(page_head, flags);
				VM_BUG_ON(PageHead(page_head));
			out_put_head:
				if (put_page_testzero(page_head))
					__put_single_page(page_head);
			out_put_single:
				if (put_page_testzero(page))
					__put_single_page(page);
				return;
			}
			VM_BUG_ON(page_head != page->first_page);
			/*
			 * We can release the refcount taken by
			 * get_page_unless_zero now that
			 * split_huge_page_refcount is blocked on the
			 * compound_lock.
			 */
			if (put_page_testzero(page_head))
				VM_BUG_ON(1);
			/* __split_huge_page_refcount will wait now */
			VM_BUG_ON(atomic_read(&page->_count) <= 0);
			atomic_dec(&page->_count);
			VM_BUG_ON(atomic_read(&page_head->_count) <= 0);
			compound_unlock_irqrestore(page_head, flags);
135 136 137 138 139 140
			if (put_page_testzero(page_head)) {
				if (PageHead(page_head))
					__put_compound_page(page_head);
				else
					__put_single_page(page_head);
			}
141 142 143 144 145 146 147 148 149 150
		} else {
			/* page_head is a dangling pointer */
			VM_BUG_ON(PageTail(page));
			goto out_put_single;
		}
	} else if (put_page_testzero(page)) {
		if (PageHead(page))
			__put_compound_page(page);
		else
			__put_single_page(page);
L
Linus Torvalds 已提交
151
	}
N
Nick Piggin 已提交
152 153 154 155 156 157 158
}

void put_page(struct page *page)
{
	if (unlikely(PageCompound(page)))
		put_compound_page(page);
	else if (put_page_testzero(page))
159
		__put_single_page(page);
L
Linus Torvalds 已提交
160 161 162
}
EXPORT_SYMBOL(put_page);

163
/**
164 165
 * put_pages_list() - release a list of pages
 * @pages: list of pages threaded on page->lru
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
 *
 * Release a list of pages which are strung together on page.lru.  Currently
 * used by read_cache_pages() and related error recovery code.
 */
void put_pages_list(struct list_head *pages)
{
	while (!list_empty(pages)) {
		struct page *victim;

		victim = list_entry(pages->prev, struct page, lru);
		list_del(&victim->lru);
		page_cache_release(victim);
	}
}
EXPORT_SYMBOL(put_pages_list);

S
Shaohua Li 已提交
182 183 184
static void pagevec_lru_move_fn(struct pagevec *pvec,
				void (*move_fn)(struct page *page, void *arg),
				void *arg)
185 186 187
{
	int i;
	struct zone *zone = NULL;
S
Shaohua Li 已提交
188
	unsigned long flags = 0;
189 190 191 192 193 194 195

	for (i = 0; i < pagevec_count(pvec); i++) {
		struct page *page = pvec->pages[i];
		struct zone *pagezone = page_zone(page);

		if (pagezone != zone) {
			if (zone)
S
Shaohua Li 已提交
196
				spin_unlock_irqrestore(&zone->lru_lock, flags);
197
			zone = pagezone;
S
Shaohua Li 已提交
198
			spin_lock_irqsave(&zone->lru_lock, flags);
199
		}
S
Shaohua Li 已提交
200 201

		(*move_fn)(page, arg);
202 203
	}
	if (zone)
S
Shaohua Li 已提交
204
		spin_unlock_irqrestore(&zone->lru_lock, flags);
205 206
	release_pages(pvec->pages, pvec->nr, pvec->cold);
	pagevec_reinit(pvec);
S
Shaohua Li 已提交
207 208
}

S
Shaohua Li 已提交
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
static void pagevec_move_tail_fn(struct page *page, void *arg)
{
	int *pgmoved = arg;
	struct zone *zone = page_zone(page);

	if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
		enum lru_list lru = page_lru_base_type(page);
		list_move_tail(&page->lru, &zone->lru[lru].list);
		mem_cgroup_rotate_reclaimable_page(page);
		(*pgmoved)++;
	}
}

/*
 * pagevec_move_tail() must be called with IRQ disabled.
 * Otherwise this may cause nasty races.
 */
static void pagevec_move_tail(struct pagevec *pvec)
{
	int pgmoved = 0;

	pagevec_lru_move_fn(pvec, pagevec_move_tail_fn, &pgmoved);
	__count_vm_events(PGROTATED, pgmoved);
}

L
Linus Torvalds 已提交
234 235 236
/*
 * Writeback is about to end against a page which has been marked for immediate
 * reclaim.  If it still appears to be reclaimable, move it to the tail of the
237
 * inactive list.
L
Linus Torvalds 已提交
238
 */
S
Shaohua Li 已提交
239
void rotate_reclaimable_page(struct page *page)
L
Linus Torvalds 已提交
240
{
241
	if (!PageLocked(page) && !PageDirty(page) && !PageActive(page) &&
L
Lee Schermerhorn 已提交
242
	    !PageUnevictable(page) && PageLRU(page)) {
243 244 245 246 247 248 249 250 251 252
		struct pagevec *pvec;
		unsigned long flags;

		page_cache_get(page);
		local_irq_save(flags);
		pvec = &__get_cpu_var(lru_rotate_pvecs);
		if (!pagevec_add(pvec, page))
			pagevec_move_tail(pvec);
		local_irq_restore(flags);
	}
L
Linus Torvalds 已提交
253 254
}

K
KOSAKI Motohiro 已提交
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274
static void update_page_reclaim_stat(struct zone *zone, struct page *page,
				     int file, int rotated)
{
	struct zone_reclaim_stat *reclaim_stat = &zone->reclaim_stat;
	struct zone_reclaim_stat *memcg_reclaim_stat;

	memcg_reclaim_stat = mem_cgroup_get_reclaim_stat_from_page(page);

	reclaim_stat->recent_scanned[file]++;
	if (rotated)
		reclaim_stat->recent_rotated[file]++;

	if (!memcg_reclaim_stat)
		return;

	memcg_reclaim_stat->recent_scanned[file]++;
	if (rotated)
		memcg_reclaim_stat->recent_rotated[file]++;
}

L
Linus Torvalds 已提交
275
/*
276
 * FIXME: speed this up?
L
Linus Torvalds 已提交
277
 */
278
void activate_page(struct page *page)
L
Linus Torvalds 已提交
279 280
{
	struct zone *zone = page_zone(page);
281

282
	spin_lock_irq(&zone->lru_lock);
283
	if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
284 285 286
		int file = page_is_file_cache(page);
		int lru = page_lru_base_type(page);
		del_page_from_lru_list(zone, page, lru);
287

288 289 290 291
		SetPageActive(page);
		lru += LRU_ACTIVE;
		add_page_to_lru_list(zone, page, lru);
		__count_vm_event(PGACTIVATE);
292

293
		update_page_reclaim_stat(zone, page, file, 1);
L
Linus Torvalds 已提交
294 295 296 297 298 299 300 301 302 303 304
	}
	spin_unlock_irq(&zone->lru_lock);
}

/*
 * Mark a page as having seen activity.
 *
 * inactive,unreferenced	->	inactive,referenced
 * inactive,referenced		->	active,unreferenced
 * active,unreferenced		->	active,referenced
 */
H
Harvey Harrison 已提交
305
void mark_page_accessed(struct page *page)
L
Linus Torvalds 已提交
306
{
L
Lee Schermerhorn 已提交
307 308
	if (!PageActive(page) && !PageUnevictable(page) &&
			PageReferenced(page) && PageLRU(page)) {
L
Linus Torvalds 已提交
309 310 311 312 313 314 315 316 317
		activate_page(page);
		ClearPageReferenced(page);
	} else if (!PageReferenced(page)) {
		SetPageReferenced(page);
	}
}

EXPORT_SYMBOL(mark_page_accessed);

318
void __lru_cache_add(struct page *page, enum lru_list lru)
L
Linus Torvalds 已提交
319
{
320
	struct pagevec *pvec = &get_cpu_var(lru_add_pvecs)[lru];
L
Linus Torvalds 已提交
321 322 323

	page_cache_get(page);
	if (!pagevec_add(pvec, page))
324
		____pagevec_lru_add(pvec, lru);
L
Linus Torvalds 已提交
325 326
	put_cpu_var(lru_add_pvecs);
}
327
EXPORT_SYMBOL(__lru_cache_add);
L
Linus Torvalds 已提交
328

329 330 331 332 333 334
/**
 * lru_cache_add_lru - add a page to a page list
 * @page: the page to be added to the LRU.
 * @lru: the LRU list to which the page is added.
 */
void lru_cache_add_lru(struct page *page, enum lru_list lru)
L
Linus Torvalds 已提交
335
{
336
	if (PageActive(page)) {
L
Lee Schermerhorn 已提交
337
		VM_BUG_ON(PageUnevictable(page));
338
		ClearPageActive(page);
L
Lee Schermerhorn 已提交
339 340 341
	} else if (PageUnevictable(page)) {
		VM_BUG_ON(PageActive(page));
		ClearPageUnevictable(page);
342
	}
L
Linus Torvalds 已提交
343

L
Lee Schermerhorn 已提交
344
	VM_BUG_ON(PageLRU(page) || PageActive(page) || PageUnevictable(page));
345
	__lru_cache_add(page, lru);
L
Linus Torvalds 已提交
346 347
}

L
Lee Schermerhorn 已提交
348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368
/**
 * add_page_to_unevictable_list - add a page to the unevictable list
 * @page:  the page to be added to the unevictable list
 *
 * Add page directly to its zone's unevictable list.  To avoid races with
 * tasks that might be making the page evictable, through eg. munlock,
 * munmap or exit, while it's not on the lru, we want to add the page
 * while it's locked or otherwise "invisible" to other tasks.  This is
 * difficult to do when using the pagevec cache, so bypass that.
 */
void add_page_to_unevictable_list(struct page *page)
{
	struct zone *zone = page_zone(page);

	spin_lock_irq(&zone->lru_lock);
	SetPageUnevictable(page);
	SetPageLRU(page);
	add_page_to_lru_list(zone, page, LRU_UNEVICTABLE);
	spin_unlock_irq(&zone->lru_lock);
}

M
Minchan Kim 已提交
369 370 371 372 373 374
/*
 * If the page can not be invalidated, it is moved to the
 * inactive list to speed up its reclaim.  It is moved to the
 * head of the list, rather than the tail, to give the flusher
 * threads some time to write it out, as this is much more
 * effective than the single-page writeout from reclaim.
M
Minchan Kim 已提交
375 376 377 378 379 380 381 382 383 384 385 386 387 388
 *
 * If the page isn't page_mapped and dirty/writeback, the page
 * could reclaim asap using PG_reclaim.
 *
 * 1. active, mapped page -> none
 * 2. active, dirty/writeback page -> inactive, head, PG_reclaim
 * 3. inactive, mapped page -> none
 * 4. inactive, dirty/writeback page -> inactive, head, PG_reclaim
 * 5. inactive, clean -> inactive, tail
 * 6. Others -> none
 *
 * In 4, why it moves inactive's head, the VM expects the page would
 * be write it out by flusher threads as this is much more effective
 * than the single-page writeout from reclaim.
M
Minchan Kim 已提交
389
 */
S
Shaohua Li 已提交
390
static void lru_deactivate_fn(struct page *page, void *arg)
M
Minchan Kim 已提交
391 392
{
	int lru, file;
M
Minchan Kim 已提交
393
	bool active;
S
Shaohua Li 已提交
394
	struct zone *zone = page_zone(page);
M
Minchan Kim 已提交
395

M
Minchan Kim 已提交
396
	if (!PageLRU(page))
M
Minchan Kim 已提交
397 398
		return;

399 400 401
	if (PageUnevictable(page))
		return;

M
Minchan Kim 已提交
402 403 404 405
	/* Some processes are using the page */
	if (page_mapped(page))
		return;

M
Minchan Kim 已提交
406 407
	active = PageActive(page);

M
Minchan Kim 已提交
408 409
	file = page_is_file_cache(page);
	lru = page_lru_base_type(page);
M
Minchan Kim 已提交
410
	del_page_from_lru_list(zone, page, lru + active);
M
Minchan Kim 已提交
411 412 413 414
	ClearPageActive(page);
	ClearPageReferenced(page);
	add_page_to_lru_list(zone, page, lru);

M
Minchan Kim 已提交
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433
	if (PageWriteback(page) || PageDirty(page)) {
		/*
		 * PG_reclaim could be raced with end_page_writeback
		 * It can make readahead confusing.  But race window
		 * is _really_ small and  it's non-critical problem.
		 */
		SetPageReclaim(page);
	} else {
		/*
		 * The page's writeback ends up during pagevec
		 * We moves tha page into tail of inactive.
		 */
		list_move_tail(&page->lru, &zone->lru[lru].list);
		mem_cgroup_rotate_reclaimable_page(page);
		__count_vm_event(PGROTATED);
	}

	if (active)
		__count_vm_event(PGDEACTIVATE);
M
Minchan Kim 已提交
434 435 436
	update_page_reclaim_stat(zone, page, file, 0);
}

437 438 439 440 441 442
/*
 * Drain pages out of the cpu's pagevecs.
 * Either "cpu" is the current CPU, and preemption has already been
 * disabled; or "cpu" is being hot-unplugged, and is already dead.
 */
static void drain_cpu_pagevecs(int cpu)
L
Linus Torvalds 已提交
443
{
444
	struct pagevec *pvecs = per_cpu(lru_add_pvecs, cpu);
445
	struct pagevec *pvec;
446
	int lru;
L
Linus Torvalds 已提交
447

448 449 450 451 452
	for_each_lru(lru) {
		pvec = &pvecs[lru - LRU_BASE];
		if (pagevec_count(pvec))
			____pagevec_lru_add(pvec, lru);
	}
453 454 455 456 457 458 459 460 461 462

	pvec = &per_cpu(lru_rotate_pvecs, cpu);
	if (pagevec_count(pvec)) {
		unsigned long flags;

		/* No harm done if a racing interrupt already did this */
		local_irq_save(flags);
		pagevec_move_tail(pvec);
		local_irq_restore(flags);
	}
M
Minchan Kim 已提交
463 464 465

	pvec = &per_cpu(lru_deactivate_pvecs, cpu);
	if (pagevec_count(pvec))
S
Shaohua Li 已提交
466
		pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL);
M
Minchan Kim 已提交
467 468 469 470 471 472 473 474 475 476 477 478
}

/**
 * deactivate_page - forcefully deactivate a page
 * @page: page to deactivate
 *
 * This function hints the VM that @page is a good reclaim candidate,
 * for example if its invalidation fails due to the page being dirty
 * or under writeback.
 */
void deactivate_page(struct page *page)
{
479 480 481 482 483 484 485
	/*
	 * In a workload with many unevictable page such as mprotect, unevictable
	 * page deactivation for accelerating reclaim is pointless.
	 */
	if (PageUnevictable(page))
		return;

M
Minchan Kim 已提交
486 487 488 489
	if (likely(get_page_unless_zero(page))) {
		struct pagevec *pvec = &get_cpu_var(lru_deactivate_pvecs);

		if (!pagevec_add(pvec, page))
S
Shaohua Li 已提交
490
			pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL);
M
Minchan Kim 已提交
491 492
		put_cpu_var(lru_deactivate_pvecs);
	}
493 494 495 496
}

void lru_add_drain(void)
{
497
	drain_cpu_pagevecs(get_cpu());
498
	put_cpu();
L
Linus Torvalds 已提交
499 500
}

D
David Howells 已提交
501
static void lru_add_drain_per_cpu(struct work_struct *dummy)
502 503 504 505 506 507 508 509 510
{
	lru_add_drain();
}

/*
 * Returns 0 for success
 */
int lru_add_drain_all(void)
{
D
David Howells 已提交
511
	return schedule_on_each_cpu(lru_add_drain_per_cpu);
512 513
}

L
Linus Torvalds 已提交
514 515 516 517 518 519 520 521
/*
 * Batched page_cache_release().  Decrement the reference count on all the
 * passed pages.  If it fell to zero then remove the page from the LRU and
 * free it.
 *
 * Avoid taking zone->lru_lock if possible, but if it is taken, retain it
 * for the remainder of the operation.
 *
522 523 524 525
 * The locking in this function is against shrink_inactive_list(): we recheck
 * the page count inside the lock to see whether shrink_inactive_list()
 * grabbed the page via the LRU.  If it did, give up: shrink_inactive_list()
 * will free it.
L
Linus Torvalds 已提交
526 527 528 529 530 531
 */
void release_pages(struct page **pages, int nr, int cold)
{
	int i;
	struct pagevec pages_to_free;
	struct zone *zone = NULL;
532
	unsigned long uninitialized_var(flags);
L
Linus Torvalds 已提交
533 534 535 536 537

	pagevec_init(&pages_to_free, cold);
	for (i = 0; i < nr; i++) {
		struct page *page = pages[i];

N
Nick Piggin 已提交
538 539
		if (unlikely(PageCompound(page))) {
			if (zone) {
540
				spin_unlock_irqrestore(&zone->lru_lock, flags);
N
Nick Piggin 已提交
541 542 543 544 545 546
				zone = NULL;
			}
			put_compound_page(page);
			continue;
		}

N
Nick Piggin 已提交
547
		if (!put_page_testzero(page))
L
Linus Torvalds 已提交
548 549
			continue;

550 551
		if (PageLRU(page)) {
			struct zone *pagezone = page_zone(page);
L
Lee Schermerhorn 已提交
552

553 554
			if (pagezone != zone) {
				if (zone)
555 556
					spin_unlock_irqrestore(&zone->lru_lock,
									flags);
557
				zone = pagezone;
558
				spin_lock_irqsave(&zone->lru_lock, flags);
559
			}
N
Nick Piggin 已提交
560
			VM_BUG_ON(!PageLRU(page));
N
Nick Piggin 已提交
561
			__ClearPageLRU(page);
L
Linus Torvalds 已提交
562
			del_page_from_lru(zone, page);
563 564 565 566
		}

		if (!pagevec_add(&pages_to_free, page)) {
			if (zone) {
567
				spin_unlock_irqrestore(&zone->lru_lock, flags);
568
				zone = NULL;
L
Linus Torvalds 已提交
569
			}
570 571 572
			__pagevec_free(&pages_to_free);
			pagevec_reinit(&pages_to_free);
  		}
L
Linus Torvalds 已提交
573 574
	}
	if (zone)
575
		spin_unlock_irqrestore(&zone->lru_lock, flags);
L
Linus Torvalds 已提交
576 577 578

	pagevec_free(&pages_to_free);
}
M
Miklos Szeredi 已提交
579
EXPORT_SYMBOL(release_pages);
L
Linus Torvalds 已提交
580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597

/*
 * The pages which we're about to release may be in the deferred lru-addition
 * queues.  That would prevent them from really being freed right now.  That's
 * OK from a correctness point of view but is inefficient - those pages may be
 * cache-warm and we want to give them back to the page allocator ASAP.
 *
 * So __pagevec_release() will drain those queues here.  __pagevec_lru_add()
 * and __pagevec_lru_add_active() call release_pages() directly to avoid
 * mutual recursion.
 */
void __pagevec_release(struct pagevec *pvec)
{
	lru_add_drain();
	release_pages(pvec->pages, pagevec_count(pvec), pvec->cold);
	pagevec_reinit(pvec);
}

598 599
EXPORT_SYMBOL(__pagevec_release);

600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636
/* used by __split_huge_page_refcount() */
void lru_add_page_tail(struct zone* zone,
		       struct page *page, struct page *page_tail)
{
	int active;
	enum lru_list lru;
	const int file = 0;
	struct list_head *head;

	VM_BUG_ON(!PageHead(page));
	VM_BUG_ON(PageCompound(page_tail));
	VM_BUG_ON(PageLRU(page_tail));
	VM_BUG_ON(!spin_is_locked(&zone->lru_lock));

	SetPageLRU(page_tail);

	if (page_evictable(page_tail, NULL)) {
		if (PageActive(page)) {
			SetPageActive(page_tail);
			active = 1;
			lru = LRU_ACTIVE_ANON;
		} else {
			active = 0;
			lru = LRU_INACTIVE_ANON;
		}
		update_page_reclaim_stat(zone, page_tail, file, active);
		if (likely(PageLRU(page)))
			head = page->lru.prev;
		else
			head = &zone->lru[lru].list;
		__add_page_to_lru_list(zone, page_tail, lru, head);
	} else {
		SetPageUnevictable(page_tail);
		add_page_to_lru_list(zone, page_tail, LRU_UNEVICTABLE);
	}
}

S
Shaohua Li 已提交
637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654
static void ____pagevec_lru_add_fn(struct page *page, void *arg)
{
	enum lru_list lru = (enum lru_list)arg;
	struct zone *zone = page_zone(page);
	int file = is_file_lru(lru);
	int active = is_active_lru(lru);

	VM_BUG_ON(PageActive(page));
	VM_BUG_ON(PageUnevictable(page));
	VM_BUG_ON(PageLRU(page));

	SetPageLRU(page);
	if (active)
		SetPageActive(page);
	update_page_reclaim_stat(zone, page, file, active);
	add_page_to_lru_list(zone, page, lru);
}

L
Linus Torvalds 已提交
655 656 657 658
/*
 * Add the passed pages to the LRU, then drop the caller's refcount
 * on them.  Reinitialises the caller's pagevec.
 */
659
void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
L
Linus Torvalds 已提交
660
{
L
Lee Schermerhorn 已提交
661
	VM_BUG_ON(is_unevictable_lru(lru));
L
Linus Torvalds 已提交
662

S
Shaohua Li 已提交
663
	pagevec_lru_move_fn(pvec, ____pagevec_lru_add_fn, (void *)lru);
L
Linus Torvalds 已提交
664 665
}

666
EXPORT_SYMBOL(____pagevec_lru_add);
L
Linus Torvalds 已提交
667 668 669 670 671 672 673 674 675 676 677

/*
 * Try to drop buffers from the pages in a pagevec
 */
void pagevec_strip(struct pagevec *pvec)
{
	int i;

	for (i = 0; i < pagevec_count(pvec); i++) {
		struct page *page = pvec->pages[i];

678 679
		if (page_has_private(page) && trylock_page(page)) {
			if (page_has_private(page))
680
				try_to_release_page(page, 0);
L
Linus Torvalds 已提交
681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
			unlock_page(page);
		}
	}
}

/**
 * pagevec_lookup - gang pagecache lookup
 * @pvec:	Where the resulting pages are placed
 * @mapping:	The address_space to search
 * @start:	The starting page index
 * @nr_pages:	The maximum number of pages
 *
 * pagevec_lookup() will search for and return a group of up to @nr_pages pages
 * in the mapping.  The pages are placed in @pvec.  pagevec_lookup() takes a
 * reference against the pages in @pvec.
 *
 * The search returns a group of mapping-contiguous pages with ascending
 * indexes.  There may be holes in the indices due to not-present pages.
 *
 * pagevec_lookup() returns the number of pages which were found.
 */
unsigned pagevec_lookup(struct pagevec *pvec, struct address_space *mapping,
		pgoff_t start, unsigned nr_pages)
{
	pvec->nr = find_get_pages(mapping, start, nr_pages, pvec->pages);
	return pagevec_count(pvec);
}

709 710
EXPORT_SYMBOL(pagevec_lookup);

L
Linus Torvalds 已提交
711 712 713 714 715 716 717 718
unsigned pagevec_lookup_tag(struct pagevec *pvec, struct address_space *mapping,
		pgoff_t *index, int tag, unsigned nr_pages)
{
	pvec->nr = find_get_pages_tag(mapping, index, tag,
					nr_pages, pvec->pages);
	return pagevec_count(pvec);
}

719
EXPORT_SYMBOL(pagevec_lookup_tag);
L
Linus Torvalds 已提交
720 721 722 723 724 725

/*
 * Perform any setup for the swap system
 */
void __init swap_setup(void)
{
726
	unsigned long megs = totalram_pages >> (20 - PAGE_SHIFT);
L
Linus Torvalds 已提交
727

P
Peter Zijlstra 已提交
728 729 730 731
#ifdef CONFIG_SWAP
	bdi_init(swapper_space.backing_dev_info);
#endif

L
Linus Torvalds 已提交
732 733 734 735 736 737 738 739 740 741
	/* Use a smaller cluster for small-memory machines */
	if (megs < 16)
		page_cluster = 2;
	else
		page_cluster = 3;
	/*
	 * Right now other parts of the system means that we
	 * _really_ don't want to cluster much more
	 */
}