vmscan.c 43.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
/*
 *  linux/mm/vmscan.c
 *
 *  Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
 *
 *  Swap reorganised 29.12.95, Stephen Tweedie.
 *  kswapd added: 7.1.96  sct
 *  Removed kswapd_ctl limits, and swap out as many pages as needed
 *  to bring the system back to freepages.high: 2.4.97, Rik van Riel.
 *  Zone aware kswapd started 02/00, Kanoj Sarcar (kanoj@sgi.com).
 *  Multiqueue VM started 5.8.00, Rik van Riel.
 */

#include <linux/mm.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/kernel_stat.h>
#include <linux/swap.h>
#include <linux/pagemap.h>
#include <linux/init.h>
#include <linux/highmem.h>
#include <linux/file.h>
#include <linux/writeback.h>
#include <linux/blkdev.h>
#include <linux/buffer_head.h>	/* for try_to_release_page(),
					buffer_heads_over_limit */
#include <linux/mm_inline.h>
#include <linux/pagevec.h>
#include <linux/backing-dev.h>
#include <linux/rmap.h>
#include <linux/topology.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
#include <linux/notifier.h>
#include <linux/rwsem.h>
36
#include <linux/delay.h>
L
Linus Torvalds 已提交
37 38 39 40 41 42

#include <asm/tlbflush.h>
#include <asm/div64.h>

#include <linux/swapops.h>

43 44
#include "internal.h"

L
Linus Torvalds 已提交
45 46 47 48 49 50 51
struct scan_control {
	/* Incremented by the number of inactive pages that were scanned */
	unsigned long nr_scanned;

	unsigned long nr_mapped;	/* From page_state */

	/* This context's GFP mask */
A
Al Viro 已提交
52
	gfp_t gfp_mask;
L
Linus Torvalds 已提交
53 54 55

	int may_writepage;

56 57 58
	/* Can pages be swapped as part of reclaim? */
	int may_swap;

L
Linus Torvalds 已提交
59 60 61 62 63
	/* This context's SWAP_CLUSTER_MAX. If freeing memory for
	 * suspend, we effectively ignore SWAP_CLUSTER_MAX.
	 * In this context, it doesn't matter that we scan the
	 * whole list at once. */
	int swap_cluster_max;
64 65

	int swappiness;
L
Linus Torvalds 已提交
66 67 68 69 70 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
};

/*
 * The list of shrinker callbacks used by to apply pressure to
 * ageable caches.
 */
struct shrinker {
	shrinker_t		shrinker;
	struct list_head	list;
	int			seeks;	/* seeks to recreate an obj */
	long			nr;	/* objs pending delete */
};

#define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru))

#ifdef ARCH_HAS_PREFETCH
#define prefetch_prev_lru_page(_page, _base, _field)			\
	do {								\
		if ((_page)->lru.prev != _base) {			\
			struct page *prev;				\
									\
			prev = lru_to_page(&(_page->lru));		\
			prefetch(&prev->_field);			\
		}							\
	} while (0)
#else
#define prefetch_prev_lru_page(_page, _base, _field) do { } while (0)
#endif

#ifdef ARCH_HAS_PREFETCHW
#define prefetchw_prev_lru_page(_page, _base, _field)			\
	do {								\
		if ((_page)->lru.prev != _base) {			\
			struct page *prev;				\
									\
			prev = lru_to_page(&(_page->lru));		\
			prefetchw(&prev->_field);			\
		}							\
	} while (0)
#else
#define prefetchw_prev_lru_page(_page, _base, _field) do { } while (0)
#endif

/*
 * From 0 .. 100.  Higher means more swappy.
 */
int vm_swappiness = 60;
static long total_memory;

static LIST_HEAD(shrinker_list);
static DECLARE_RWSEM(shrinker_rwsem);

/*
 * Add a shrinker callback to be called from the vm
 */
struct shrinker *set_shrinker(int seeks, shrinker_t theshrinker)
{
        struct shrinker *shrinker;

        shrinker = kmalloc(sizeof(*shrinker), GFP_KERNEL);
        if (shrinker) {
	        shrinker->shrinker = theshrinker;
	        shrinker->seeks = seeks;
	        shrinker->nr = 0;
	        down_write(&shrinker_rwsem);
	        list_add_tail(&shrinker->list, &shrinker_list);
	        up_write(&shrinker_rwsem);
	}
	return shrinker;
}
EXPORT_SYMBOL(set_shrinker);

/*
 * Remove one
 */
void remove_shrinker(struct shrinker *shrinker)
{
	down_write(&shrinker_rwsem);
	list_del(&shrinker->list);
	up_write(&shrinker_rwsem);
	kfree(shrinker);
}
EXPORT_SYMBOL(remove_shrinker);

#define SHRINK_BATCH 128
/*
 * Call the shrink functions to age shrinkable caches
 *
 * Here we assume it costs one seek to replace a lru page and that it also
 * takes a seek to recreate a cache object.  With this in mind we age equal
 * percentages of the lru and ageable caches.  This should balance the seeks
 * generated by these structures.
 *
 * If the vm encounted mapped pages on the LRU it increase the pressure on
 * slab to avoid swapping.
 *
 * We do weird things to avoid (scanned*seeks*entries) overflowing 32 bits.
 *
 * `lru_pages' represents the number of on-LRU pages in all the zones which
 * are eligible for the caller's allocation attempt.  It is used for balancing
 * slab reclaim versus page reclaim.
167 168
 *
 * Returns the number of slab objects which we shrunk.
L
Linus Torvalds 已提交
169
 */
170 171
unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask,
			unsigned long lru_pages)
L
Linus Torvalds 已提交
172 173
{
	struct shrinker *shrinker;
174
	unsigned long ret = 0;
L
Linus Torvalds 已提交
175 176 177 178 179

	if (scanned == 0)
		scanned = SWAP_CLUSTER_MAX;

	if (!down_read_trylock(&shrinker_rwsem))
180
		return 1;	/* Assume we'll be able to shrink next time */
L
Linus Torvalds 已提交
181 182 183 184

	list_for_each_entry(shrinker, &shrinker_list, list) {
		unsigned long long delta;
		unsigned long total_scan;
185
		unsigned long max_pass = (*shrinker->shrinker)(0, gfp_mask);
L
Linus Torvalds 已提交
186 187

		delta = (4 * scanned) / shrinker->seeks;
188
		delta *= max_pass;
L
Linus Torvalds 已提交
189 190
		do_div(delta, lru_pages + 1);
		shrinker->nr += delta;
191 192 193 194 195 196 197 198 199 200 201 202 203
		if (shrinker->nr < 0) {
			printk(KERN_ERR "%s: nr=%ld\n",
					__FUNCTION__, shrinker->nr);
			shrinker->nr = max_pass;
		}

		/*
		 * Avoid risking looping forever due to too large nr value:
		 * never try to free more than twice the estimate number of
		 * freeable entries.
		 */
		if (shrinker->nr > max_pass * 2)
			shrinker->nr = max_pass * 2;
L
Linus Torvalds 已提交
204 205 206 207 208 209 210

		total_scan = shrinker->nr;
		shrinker->nr = 0;

		while (total_scan >= SHRINK_BATCH) {
			long this_scan = SHRINK_BATCH;
			int shrink_ret;
211
			int nr_before;
L
Linus Torvalds 已提交
212

213
			nr_before = (*shrinker->shrinker)(0, gfp_mask);
L
Linus Torvalds 已提交
214 215 216
			shrink_ret = (*shrinker->shrinker)(this_scan, gfp_mask);
			if (shrink_ret == -1)
				break;
217 218
			if (shrink_ret < nr_before)
				ret += nr_before - shrink_ret;
L
Linus Torvalds 已提交
219 220 221 222 223 224 225 226 227
			mod_page_state(slabs_scanned, this_scan);
			total_scan -= this_scan;

			cond_resched();
		}

		shrinker->nr += total_scan;
	}
	up_read(&shrinker_rwsem);
228
	return ret;
L
Linus Torvalds 已提交
229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
}

/* Called without lock on whether page is mapped, so answer is unstable */
static inline int page_mapping_inuse(struct page *page)
{
	struct address_space *mapping;

	/* Page is in somebody's page tables. */
	if (page_mapped(page))
		return 1;

	/* Be more reluctant to reclaim swapcache than pagecache */
	if (PageSwapCache(page))
		return 1;

	mapping = page_mapping(page);
	if (!mapping)
		return 0;

	/* File is mmap'd by somebody? */
	return mapping_mapped(mapping);
}

static inline int is_page_cache_freeable(struct page *page)
{
	return page_count(page) - !!PagePrivate(page) == 2;
}

static int may_write_to_queue(struct backing_dev_info *bdi)
{
259
	if (current->flags & PF_SWAPWRITE)
L
Linus Torvalds 已提交
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293
		return 1;
	if (!bdi_write_congested(bdi))
		return 1;
	if (bdi == current->backing_dev_info)
		return 1;
	return 0;
}

/*
 * We detected a synchronous write error writing a page out.  Probably
 * -ENOSPC.  We need to propagate that into the address_space for a subsequent
 * fsync(), msync() or close().
 *
 * The tricky part is that after writepage we cannot touch the mapping: nothing
 * prevents it from being freed up.  But we have a ref on the page and once
 * that page is locked, the mapping is pinned.
 *
 * We're allowed to run sleeping lock_page() here because we know the caller has
 * __GFP_FS.
 */
static void handle_write_error(struct address_space *mapping,
				struct page *page, int error)
{
	lock_page(page);
	if (page_mapping(page) == mapping) {
		if (error == -ENOSPC)
			set_bit(AS_ENOSPC, &mapping->flags);
		else
			set_bit(AS_EIO, &mapping->flags);
	}
	unlock_page(page);
}

/*
A
Andrew Morton 已提交
294 295
 * pageout is called by shrink_page_list() for each dirty page.
 * Calls ->writepage().
L
Linus Torvalds 已提交
296
 */
C
Christoph Lameter 已提交
297
pageout_t pageout(struct page *page, struct address_space *mapping)
L
Linus Torvalds 已提交
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
{
	/*
	 * If the page is dirty, only perform writeback if that write
	 * will be non-blocking.  To prevent this allocation from being
	 * stalled by pagecache activity.  But note that there may be
	 * stalls if we need to run get_block().  We could test
	 * PagePrivate for that.
	 *
	 * If this process is currently in generic_file_write() against
	 * this page's queue, we can perform writeback even if that
	 * will block.
	 *
	 * If the page is swapcache, write it back even if that would
	 * block, for some throttling. This happens by accident, because
	 * swap_backing_dev_info is bust: it doesn't reflect the
	 * congestion state of the swapdevs.  Easy to fix, if needed.
	 * See swapfile.c:page_queue_congested().
	 */
	if (!is_page_cache_freeable(page))
		return PAGE_KEEP;
	if (!mapping) {
		/*
		 * Some data journaling orphaned pages can have
		 * page->mapping == NULL while being dirty with clean buffers.
		 */
323
		if (PagePrivate(page)) {
L
Linus Torvalds 已提交
324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349
			if (try_to_free_buffers(page)) {
				ClearPageDirty(page);
				printk("%s: orphaned page\n", __FUNCTION__);
				return PAGE_CLEAN;
			}
		}
		return PAGE_KEEP;
	}
	if (mapping->a_ops->writepage == NULL)
		return PAGE_ACTIVATE;
	if (!may_write_to_queue(mapping->backing_dev_info))
		return PAGE_KEEP;

	if (clear_page_dirty_for_io(page)) {
		int res;
		struct writeback_control wbc = {
			.sync_mode = WB_SYNC_NONE,
			.nr_to_write = SWAP_CLUSTER_MAX,
			.nonblocking = 1,
			.for_reclaim = 1,
		};

		SetPageReclaim(page);
		res = mapping->a_ops->writepage(page, &wbc);
		if (res < 0)
			handle_write_error(mapping, page, res);
350
		if (res == AOP_WRITEPAGE_ACTIVATE) {
L
Linus Torvalds 已提交
351 352 353 354 355 356 357 358 359 360 361 362 363 364
			ClearPageReclaim(page);
			return PAGE_ACTIVATE;
		}
		if (!PageWriteback(page)) {
			/* synchronous write or broken a_ops? */
			ClearPageReclaim(page);
		}

		return PAGE_SUCCESS;
	}

	return PAGE_CLEAN;
}

C
Christoph Lameter 已提交
365
int remove_mapping(struct address_space *mapping, struct page *page)
366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
{
	if (!mapping)
		return 0;		/* truncate got there first */

	write_lock_irq(&mapping->tree_lock);

	/*
	 * The non-racy check for busy page.  It is critical to check
	 * PageDirty _after_ making sure that the page is freeable and
	 * not in use by anybody. 	(pagecache + us == 2)
	 */
	if (unlikely(page_count(page) != 2))
		goto cannot_free;
	smp_rmb();
	if (unlikely(PageDirty(page)))
		goto cannot_free;

	if (PageSwapCache(page)) {
		swp_entry_t swap = { .val = page_private(page) };
		__delete_from_swap_cache(page);
		write_unlock_irq(&mapping->tree_lock);
		swap_free(swap);
		__put_page(page);	/* The pagecache ref */
		return 1;
	}

	__remove_from_page_cache(page);
	write_unlock_irq(&mapping->tree_lock);
	__put_page(page);
	return 1;

cannot_free:
	write_unlock_irq(&mapping->tree_lock);
	return 0;
}

L
Linus Torvalds 已提交
402
/*
A
Andrew Morton 已提交
403
 * shrink_page_list() returns the number of reclaimed pages
L
Linus Torvalds 已提交
404
 */
A
Andrew Morton 已提交
405 406
static unsigned long shrink_page_list(struct list_head *page_list,
					struct scan_control *sc)
L
Linus Torvalds 已提交
407 408 409 410
{
	LIST_HEAD(ret_pages);
	struct pagevec freed_pvec;
	int pgactivate = 0;
411
	unsigned long nr_reclaimed = 0;
L
Linus Torvalds 已提交
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

	cond_resched();

	pagevec_init(&freed_pvec, 1);
	while (!list_empty(page_list)) {
		struct address_space *mapping;
		struct page *page;
		int may_enter_fs;
		int referenced;

		cond_resched();

		page = lru_to_page(page_list);
		list_del(&page->lru);

		if (TestSetPageLocked(page))
			goto keep;

		BUG_ON(PageActive(page));

		sc->nr_scanned++;
433 434 435 436

		if (!sc->may_swap && page_mapped(page))
			goto keep_locked;

L
Linus Torvalds 已提交
437 438 439 440 441 442 443
		/* Double the slab pressure for mapped and swapcache pages */
		if (page_mapped(page) || PageSwapCache(page))
			sc->nr_scanned++;

		if (PageWriteback(page))
			goto keep_locked;

444
		referenced = page_referenced(page, 1);
L
Linus Torvalds 已提交
445 446 447 448 449 450 451 452 453
		/* In active use or really unfreeable?  Activate it. */
		if (referenced && page_mapping_inuse(page))
			goto activate_locked;

#ifdef CONFIG_SWAP
		/*
		 * Anonymous process memory has backing store?
		 * Try to allocate it some swap space here.
		 */
454
		if (PageAnon(page) && !PageSwapCache(page))
455
			if (!add_to_swap(page, GFP_ATOMIC))
L
Linus Torvalds 已提交
456 457 458 459 460 461 462 463 464 465 466 467
				goto activate_locked;
#endif /* CONFIG_SWAP */

		mapping = page_mapping(page);
		may_enter_fs = (sc->gfp_mask & __GFP_FS) ||
			(PageSwapCache(page) && (sc->gfp_mask & __GFP_IO));

		/*
		 * The page is mapped into the page tables of one or more
		 * processes. Try to unmap it here.
		 */
		if (page_mapped(page) && mapping) {
468
			switch (try_to_unmap(page, 0)) {
L
Linus Torvalds 已提交
469 470 471 472 473 474 475 476 477 478 479 480 481 482
			case SWAP_FAIL:
				goto activate_locked;
			case SWAP_AGAIN:
				goto keep_locked;
			case SWAP_SUCCESS:
				; /* try to free the page below */
			}
		}

		if (PageDirty(page)) {
			if (referenced)
				goto keep_locked;
			if (!may_enter_fs)
				goto keep_locked;
483
			if (!sc->may_writepage)
L
Linus Torvalds 已提交
484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
				goto keep_locked;

			/* Page is dirty, try to write it out here */
			switch(pageout(page, mapping)) {
			case PAGE_KEEP:
				goto keep_locked;
			case PAGE_ACTIVATE:
				goto activate_locked;
			case PAGE_SUCCESS:
				if (PageWriteback(page) || PageDirty(page))
					goto keep;
				/*
				 * A synchronous write - probably a ramdisk.  Go
				 * ahead and try to reclaim the page.
				 */
				if (TestSetPageLocked(page))
					goto keep;
				if (PageDirty(page) || PageWriteback(page))
					goto keep_locked;
				mapping = page_mapping(page);
			case PAGE_CLEAN:
				; /* try to free the page below */
			}
		}

		/*
		 * If the page has buffers, try to free the buffer mappings
		 * associated with this page. If we succeed we try to free
		 * the page as well.
		 *
		 * We do this even if the page is PageDirty().
		 * try_to_release_page() does not perform I/O, but it is
		 * possible for a page to have PageDirty set, but it is actually
		 * clean (all its buffers are clean).  This happens if the
		 * buffers were written out directly, with submit_bh(). ext3
		 * will do this, as well as the blockdev mapping. 
		 * try_to_release_page() will discover that cleanness and will
		 * drop the buffers and mark the page clean - it can be freed.
		 *
		 * Rarely, pages can have buffers and no ->mapping.  These are
		 * the pages which were not successfully invalidated in
		 * truncate_complete_page().  We try to drop those buffers here
		 * and if that worked, and the page is no longer mapped into
		 * process address space (page_count == 1) it can be freed.
		 * Otherwise, leave the page on the LRU so it is swappable.
		 */
		if (PagePrivate(page)) {
			if (!try_to_release_page(page, sc->gfp_mask))
				goto activate_locked;
			if (!mapping && page_count(page) == 1)
				goto free_it;
		}

537 538
		if (!remove_mapping(mapping, page))
			goto keep_locked;
L
Linus Torvalds 已提交
539 540 541

free_it:
		unlock_page(page);
542
		nr_reclaimed++;
L
Linus Torvalds 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559
		if (!pagevec_add(&freed_pvec, page))
			__pagevec_release_nonlru(&freed_pvec);
		continue;

activate_locked:
		SetPageActive(page);
		pgactivate++;
keep_locked:
		unlock_page(page);
keep:
		list_add(&page->lru, &ret_pages);
		BUG_ON(PageLRU(page));
	}
	list_splice(&ret_pages, page_list);
	if (pagevec_count(&freed_pvec))
		__pagevec_release_nonlru(&freed_pvec);
	mod_page_state(pgactivate, pgactivate);
560
	return nr_reclaimed;
L
Linus Torvalds 已提交
561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
}

/*
 * zone->lru_lock is heavily contended.  Some of the functions that
 * shrink the lists perform better by taking out a batch of pages
 * and working on them outside the LRU lock.
 *
 * For pagecache intensive workloads, this function is the hottest
 * spot in the kernel (apart from copy_*_user functions).
 *
 * Appropriate locks must be held before calling this function.
 *
 * @nr_to_scan:	The number of pages to look through on the list.
 * @src:	The LRU list to pull pages off.
 * @dst:	The temp list to put pages on to.
 * @scanned:	The number of pages that were scanned.
 *
 * returns how many pages were moved onto *@dst.
 */
580 581 582
static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
		struct list_head *src, struct list_head *dst,
		unsigned long *scanned)
L
Linus Torvalds 已提交
583
{
584
	unsigned long nr_taken = 0;
L
Linus Torvalds 已提交
585
	struct page *page;
586
	unsigned long scan;
L
Linus Torvalds 已提交
587

588
	for (scan = 0; scan < nr_to_scan && !list_empty(src); scan++) {
589
		struct list_head *target;
L
Linus Torvalds 已提交
590 591 592
		page = lru_to_page(src);
		prefetchw_prev_lru_page(page, src, flags);

N
Nick Piggin 已提交
593 594
		BUG_ON(!PageLRU(page));

595
		list_del(&page->lru);
596 597
		target = src;
		if (likely(get_page_unless_zero(page))) {
598
			/*
599 600 601
			 * Be careful not to clear PageLRU until after we're
			 * sure the page is not being freed elsewhere -- the
			 * page release code relies on it.
602
			 */
603 604 605 606
			ClearPageLRU(page);
			target = dst;
			nr_taken++;
		} /* else it is being freed elsewhere */
607

608
		list_add(&page->lru, target);
L
Linus Torvalds 已提交
609 610 611 612 613 614 615
	}

	*scanned = scan;
	return nr_taken;
}

/*
A
Andrew Morton 已提交
616 617
 * shrink_inactive_list() is a helper for shrink_zone().  It returns the number
 * of reclaimed pages
L
Linus Torvalds 已提交
618
 */
A
Andrew Morton 已提交
619 620
static unsigned long shrink_inactive_list(unsigned long max_scan,
				struct zone *zone, struct scan_control *sc)
L
Linus Torvalds 已提交
621 622 623
{
	LIST_HEAD(page_list);
	struct pagevec pvec;
624
	unsigned long nr_scanned = 0;
625
	unsigned long nr_reclaimed = 0;
L
Linus Torvalds 已提交
626 627 628 629 630

	pagevec_init(&pvec, 1);

	lru_add_drain();
	spin_lock_irq(&zone->lru_lock);
631
	do {
L
Linus Torvalds 已提交
632
		struct page *page;
633 634 635
		unsigned long nr_taken;
		unsigned long nr_scan;
		unsigned long nr_freed;
L
Linus Torvalds 已提交
636 637 638 639 640 641 642 643

		nr_taken = isolate_lru_pages(sc->swap_cluster_max,
					     &zone->inactive_list,
					     &page_list, &nr_scan);
		zone->nr_inactive -= nr_taken;
		zone->pages_scanned += nr_scan;
		spin_unlock_irq(&zone->lru_lock);

644
		nr_scanned += nr_scan;
A
Andrew Morton 已提交
645
		nr_freed = shrink_page_list(&page_list, sc);
646
		nr_reclaimed += nr_freed;
N
Nick Piggin 已提交
647 648 649 650 651 652 653 654
		local_irq_disable();
		if (current_is_kswapd()) {
			__mod_page_state_zone(zone, pgscan_kswapd, nr_scan);
			__mod_page_state(kswapd_steal, nr_freed);
		} else
			__mod_page_state_zone(zone, pgscan_direct, nr_scan);
		__mod_page_state_zone(zone, pgsteal, nr_freed);

655 656 657
		if (nr_taken == 0)
			goto done;

N
Nick Piggin 已提交
658
		spin_lock(&zone->lru_lock);
L
Linus Torvalds 已提交
659 660 661 662 663
		/*
		 * Put back any unfreeable pages.
		 */
		while (!list_empty(&page_list)) {
			page = lru_to_page(&page_list);
N
Nick Piggin 已提交
664 665
			BUG_ON(PageLRU(page));
			SetPageLRU(page);
L
Linus Torvalds 已提交
666 667 668 669 670 671 672 673 674 675 676
			list_del(&page->lru);
			if (PageActive(page))
				add_page_to_active_list(zone, page);
			else
				add_page_to_inactive_list(zone, page);
			if (!pagevec_add(&pvec, page)) {
				spin_unlock_irq(&zone->lru_lock);
				__pagevec_release(&pvec);
				spin_lock_irq(&zone->lru_lock);
			}
		}
677
  	} while (nr_scanned < max_scan);
678
	spin_unlock(&zone->lru_lock);
L
Linus Torvalds 已提交
679
done:
680
	local_irq_enable();
L
Linus Torvalds 已提交
681
	pagevec_release(&pvec);
682
	return nr_reclaimed;
L
Linus Torvalds 已提交
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701
}

/*
 * This moves pages from the active list to the inactive list.
 *
 * We move them the other way if the page is referenced by one or more
 * processes, from rmap.
 *
 * If the pages are mostly unmapped, the processing is fast and it is
 * appropriate to hold zone->lru_lock across the whole operation.  But if
 * the pages are mapped, the processing is slow (page_referenced()) so we
 * should drop zone->lru_lock around each page.  It's impossible to balance
 * this, so instead we remove the pages from the LRU while processing them.
 * It is safe to rely on PG_active against the non-LRU pages in here because
 * nobody will play with that bit on a non-LRU page.
 *
 * The downside is that we have to touch page->_count against each page.
 * But we had to alter page->flags anyway.
 */
A
Andrew Morton 已提交
702 703
static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
				struct scan_control *sc)
L
Linus Torvalds 已提交
704
{
705
	unsigned long pgmoved;
L
Linus Torvalds 已提交
706
	int pgdeactivate = 0;
707
	unsigned long pgscanned;
L
Linus Torvalds 已提交
708 709 710 711 712 713
	LIST_HEAD(l_hold);	/* The pages which were snipped off */
	LIST_HEAD(l_inactive);	/* Pages to go onto the inactive_list */
	LIST_HEAD(l_active);	/* Pages to go onto the active_list */
	struct page *page;
	struct pagevec pvec;
	int reclaim_mapped = 0;
714

715
	if (sc->may_swap) {
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745
		long mapped_ratio;
		long distress;
		long swap_tendency;

		/*
		 * `distress' is a measure of how much trouble we're having
		 * reclaiming pages.  0 -> no problems.  100 -> great trouble.
		 */
		distress = 100 >> zone->prev_priority;

		/*
		 * The point of this algorithm is to decide when to start
		 * reclaiming mapped memory instead of just pagecache.  Work out
		 * how much memory
		 * is mapped.
		 */
		mapped_ratio = (sc->nr_mapped * 100) / total_memory;

		/*
		 * Now decide how much we really want to unmap some pages.  The
		 * mapped ratio is downgraded - just because there's a lot of
		 * mapped memory doesn't necessarily mean that page reclaim
		 * isn't succeeding.
		 *
		 * The distress ratio is important - we don't want to start
		 * going oom.
		 *
		 * A 100% value of vm_swappiness overrides this algorithm
		 * altogether.
		 */
746
		swap_tendency = mapped_ratio / 2 + distress + sc->swappiness;
747 748 749 750 751 752 753 754

		/*
		 * Now use this metric to decide whether to start moving mapped
		 * memory onto the inactive list.
		 */
		if (swap_tendency >= 100)
			reclaim_mapped = 1;
	}
L
Linus Torvalds 已提交
755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770

	lru_add_drain();
	spin_lock_irq(&zone->lru_lock);
	pgmoved = isolate_lru_pages(nr_pages, &zone->active_list,
				    &l_hold, &pgscanned);
	zone->pages_scanned += pgscanned;
	zone->nr_active -= pgmoved;
	spin_unlock_irq(&zone->lru_lock);

	while (!list_empty(&l_hold)) {
		cond_resched();
		page = lru_to_page(&l_hold);
		list_del(&page->lru);
		if (page_mapped(page)) {
			if (!reclaim_mapped ||
			    (total_swap_pages == 0 && PageAnon(page)) ||
771
			    page_referenced(page, 0)) {
L
Linus Torvalds 已提交
772 773 774 775 776 777 778 779 780 781 782 783 784
				list_add(&page->lru, &l_active);
				continue;
			}
		}
		list_add(&page->lru, &l_inactive);
	}

	pagevec_init(&pvec, 1);
	pgmoved = 0;
	spin_lock_irq(&zone->lru_lock);
	while (!list_empty(&l_inactive)) {
		page = lru_to_page(&l_inactive);
		prefetchw_prev_lru_page(page, &l_inactive, flags);
N
Nick Piggin 已提交
785 786
		BUG_ON(PageLRU(page));
		SetPageLRU(page);
N
Nick Piggin 已提交
787 788 789
		BUG_ON(!PageActive(page));
		ClearPageActive(page);

L
Linus Torvalds 已提交
790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814
		list_move(&page->lru, &zone->inactive_list);
		pgmoved++;
		if (!pagevec_add(&pvec, page)) {
			zone->nr_inactive += pgmoved;
			spin_unlock_irq(&zone->lru_lock);
			pgdeactivate += pgmoved;
			pgmoved = 0;
			if (buffer_heads_over_limit)
				pagevec_strip(&pvec);
			__pagevec_release(&pvec);
			spin_lock_irq(&zone->lru_lock);
		}
	}
	zone->nr_inactive += pgmoved;
	pgdeactivate += pgmoved;
	if (buffer_heads_over_limit) {
		spin_unlock_irq(&zone->lru_lock);
		pagevec_strip(&pvec);
		spin_lock_irq(&zone->lru_lock);
	}

	pgmoved = 0;
	while (!list_empty(&l_active)) {
		page = lru_to_page(&l_active);
		prefetchw_prev_lru_page(page, &l_active, flags);
N
Nick Piggin 已提交
815 816
		BUG_ON(PageLRU(page));
		SetPageLRU(page);
L
Linus Torvalds 已提交
817 818 819 820 821 822 823 824 825 826 827 828
		BUG_ON(!PageActive(page));
		list_move(&page->lru, &zone->active_list);
		pgmoved++;
		if (!pagevec_add(&pvec, page)) {
			zone->nr_active += pgmoved;
			pgmoved = 0;
			spin_unlock_irq(&zone->lru_lock);
			__pagevec_release(&pvec);
			spin_lock_irq(&zone->lru_lock);
		}
	}
	zone->nr_active += pgmoved;
N
Nick Piggin 已提交
829 830 831 832 833
	spin_unlock(&zone->lru_lock);

	__mod_page_state_zone(zone, pgrefill, pgscanned);
	__mod_page_state(pgdeactivate, pgdeactivate);
	local_irq_enable();
L
Linus Torvalds 已提交
834

N
Nick Piggin 已提交
835
	pagevec_release(&pvec);
L
Linus Torvalds 已提交
836 837 838 839 840
}

/*
 * This is a basic per-zone page freer.  Used by both kswapd and direct reclaim.
 */
841 842
static unsigned long shrink_zone(int priority, struct zone *zone,
				struct scan_control *sc)
L
Linus Torvalds 已提交
843 844 845
{
	unsigned long nr_active;
	unsigned long nr_inactive;
846
	unsigned long nr_to_scan;
847
	unsigned long nr_reclaimed = 0;
L
Linus Torvalds 已提交
848

849 850
	atomic_inc(&zone->reclaim_in_progress);

L
Linus Torvalds 已提交
851 852 853 854
	/*
	 * Add one to `nr_to_scan' just to make sure that the kernel will
	 * slowly sift through the active list.
	 */
855
	zone->nr_scan_active += (zone->nr_active >> priority) + 1;
L
Linus Torvalds 已提交
856 857 858 859 860 861
	nr_active = zone->nr_scan_active;
	if (nr_active >= sc->swap_cluster_max)
		zone->nr_scan_active = 0;
	else
		nr_active = 0;

862
	zone->nr_scan_inactive += (zone->nr_inactive >> priority) + 1;
L
Linus Torvalds 已提交
863 864 865 866 867 868 869 870
	nr_inactive = zone->nr_scan_inactive;
	if (nr_inactive >= sc->swap_cluster_max)
		zone->nr_scan_inactive = 0;
	else
		nr_inactive = 0;

	while (nr_active || nr_inactive) {
		if (nr_active) {
871
			nr_to_scan = min(nr_active,
L
Linus Torvalds 已提交
872
					(unsigned long)sc->swap_cluster_max);
873
			nr_active -= nr_to_scan;
A
Andrew Morton 已提交
874
			shrink_active_list(nr_to_scan, zone, sc);
L
Linus Torvalds 已提交
875 876 877
		}

		if (nr_inactive) {
878
			nr_to_scan = min(nr_inactive,
L
Linus Torvalds 已提交
879
					(unsigned long)sc->swap_cluster_max);
880
			nr_inactive -= nr_to_scan;
A
Andrew Morton 已提交
881 882
			nr_reclaimed += shrink_inactive_list(nr_to_scan, zone,
								sc);
L
Linus Torvalds 已提交
883 884 885 886
		}
	}

	throttle_vm_writeout();
887 888

	atomic_dec(&zone->reclaim_in_progress);
889
	return nr_reclaimed;
L
Linus Torvalds 已提交
890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907
}

/*
 * This is the direct reclaim path, for page-allocating processes.  We only
 * try to reclaim pages from zones which will satisfy the caller's allocation
 * request.
 *
 * We reclaim from a zone even if that zone is over pages_high.  Because:
 * a) The caller may be trying to free *extra* pages to satisfy a higher-order
 *    allocation or
 * b) The zones may be over pages_high but they must go *over* pages_high to
 *    satisfy the `incremental min' zone defense algorithm.
 *
 * Returns the number of reclaimed pages.
 *
 * If a zone is deemed to be full of pinned pages then just give it a light
 * scan then give up on it.
 */
A
Andrew Morton 已提交
908
static unsigned long shrink_zones(int priority, struct zone **zones,
909
					struct scan_control *sc)
L
Linus Torvalds 已提交
910
{
911
	unsigned long nr_reclaimed = 0;
L
Linus Torvalds 已提交
912 913 914 915 916
	int i;

	for (i = 0; zones[i] != NULL; i++) {
		struct zone *zone = zones[i];

917
		if (!populated_zone(zone))
L
Linus Torvalds 已提交
918 919
			continue;

920
		if (!cpuset_zone_allowed(zone, __GFP_HARDWALL))
L
Linus Torvalds 已提交
921 922
			continue;

923 924 925
		zone->temp_priority = priority;
		if (zone->prev_priority > priority)
			zone->prev_priority = priority;
L
Linus Torvalds 已提交
926

927
		if (zone->all_unreclaimable && priority != DEF_PRIORITY)
L
Linus Torvalds 已提交
928 929
			continue;	/* Let kswapd poll it */

930
		nr_reclaimed += shrink_zone(priority, zone, sc);
L
Linus Torvalds 已提交
931
	}
932
	return nr_reclaimed;
L
Linus Torvalds 已提交
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947
}
 
/*
 * This is the main entry point to direct page reclaim.
 *
 * If a full scan of the inactive list fails to free enough memory then we
 * are "out of memory" and something needs to be killed.
 *
 * If the caller is !__GFP_FS then the probability of a failure is reasonably
 * high - the zone may be full of dirty or under-writeback pages, which this
 * caller can't do much about.  We kick pdflush and take explicit naps in the
 * hope that some of these pages can be written.  But if the allocating task
 * holds filesystem locks which prevent writeout this might not work, and the
 * allocation attempt will fail.
 */
948
unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask)
L
Linus Torvalds 已提交
949 950 951
{
	int priority;
	int ret = 0;
952
	unsigned long total_scanned = 0;
953
	unsigned long nr_reclaimed = 0;
L
Linus Torvalds 已提交
954 955 956
	struct reclaim_state *reclaim_state = current->reclaim_state;
	unsigned long lru_pages = 0;
	int i;
957 958 959 960 961
	struct scan_control sc = {
		.gfp_mask = gfp_mask,
		.may_writepage = !laptop_mode,
		.swap_cluster_max = SWAP_CLUSTER_MAX,
		.may_swap = 1,
962
		.swappiness = vm_swappiness,
963
	};
L
Linus Torvalds 已提交
964 965 966 967 968 969

	inc_page_state(allocstall);

	for (i = 0; zones[i] != NULL; i++) {
		struct zone *zone = zones[i];

970
		if (!cpuset_zone_allowed(zone, __GFP_HARDWALL))
L
Linus Torvalds 已提交
971 972 973 974 975 976 977 978 979
			continue;

		zone->temp_priority = DEF_PRIORITY;
		lru_pages += zone->nr_active + zone->nr_inactive;
	}

	for (priority = DEF_PRIORITY; priority >= 0; priority--) {
		sc.nr_mapped = read_page_state(nr_mapped);
		sc.nr_scanned = 0;
980 981
		if (!priority)
			disable_swap_token();
A
Andrew Morton 已提交
982
		nr_reclaimed += shrink_zones(priority, zones, &sc);
L
Linus Torvalds 已提交
983 984
		shrink_slab(sc.nr_scanned, gfp_mask, lru_pages);
		if (reclaim_state) {
985
			nr_reclaimed += reclaim_state->reclaimed_slab;
L
Linus Torvalds 已提交
986 987 988
			reclaim_state->reclaimed_slab = 0;
		}
		total_scanned += sc.nr_scanned;
989
		if (nr_reclaimed >= sc.swap_cluster_max) {
L
Linus Torvalds 已提交
990 991 992 993 994 995 996 997 998 999 1000
			ret = 1;
			goto out;
		}

		/*
		 * Try to write back as many pages as we just scanned.  This
		 * tends to cause slow streaming writers to write data to the
		 * disk smoothly, at the dirtying rate, which is nice.   But
		 * that's undesirable in laptop mode, where we *want* lumpy
		 * writeout.  So in laptop mode, write out the whole world.
		 */
1001 1002
		if (total_scanned > sc.swap_cluster_max +
					sc.swap_cluster_max / 2) {
1003
			wakeup_pdflush(laptop_mode ? 0 : total_scanned);
L
Linus Torvalds 已提交
1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014
			sc.may_writepage = 1;
		}

		/* Take a nap, wait for some writeback to complete */
		if (sc.nr_scanned && priority < DEF_PRIORITY - 2)
			blk_congestion_wait(WRITE, HZ/10);
	}
out:
	for (i = 0; zones[i] != 0; i++) {
		struct zone *zone = zones[i];

1015
		if (!cpuset_zone_allowed(zone, __GFP_HARDWALL))
L
Linus Torvalds 已提交
1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
			continue;

		zone->prev_priority = zone->temp_priority;
	}
	return ret;
}

/*
 * For kswapd, balance_pgdat() will work across all this node's zones until
 * they are all at pages_high.
 *
 * Returns the number of pages which were actually freed.
 *
 * There is special handling here for zones which are full of pinned pages.
 * This can happen if the pages are all mlocked, or if they are all used by
 * device drivers (say, ZONE_DMA).  Or if they are all in use by hugetlb.
 * What we do is to detect the case where all pages in the zone have been
 * scanned twice and there has been zero successful reclaim.  Mark the zone as
 * dead and from now on, only perform a short scan.  Basically we're polling
 * the zone for when the problem goes away.
 *
 * kswapd scans the zones in the highmem->normal->dma direction.  It skips
 * zones which have free_pages > pages_high, but once a zone is found to have
 * free_pages <= pages_high, we scan that zone and the lower zones regardless
 * of the number of free pages in the lower zones.  This interoperates with
 * the page allocator fallback scheme to ensure that aging of pages is balanced
 * across the zones.
 */
1044
static unsigned long balance_pgdat(pg_data_t *pgdat, int order)
L
Linus Torvalds 已提交
1045 1046 1047 1048
{
	int all_zones_ok;
	int priority;
	int i;
1049
	unsigned long total_scanned;
1050
	unsigned long nr_reclaimed;
L
Linus Torvalds 已提交
1051
	struct reclaim_state *reclaim_state = current->reclaim_state;
1052 1053 1054
	struct scan_control sc = {
		.gfp_mask = GFP_KERNEL,
		.may_swap = 1,
1055 1056
		.swap_cluster_max = SWAP_CLUSTER_MAX,
		.swappiness = vm_swappiness,
1057
	};
L
Linus Torvalds 已提交
1058 1059 1060

loop_again:
	total_scanned = 0;
1061
	nr_reclaimed = 0;
C
Christoph Lameter 已提交
1062
	sc.may_writepage = !laptop_mode;
L
Linus Torvalds 已提交
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076
	sc.nr_mapped = read_page_state(nr_mapped);

	inc_page_state(pageoutrun);

	for (i = 0; i < pgdat->nr_zones; i++) {
		struct zone *zone = pgdat->node_zones + i;

		zone->temp_priority = DEF_PRIORITY;
	}

	for (priority = DEF_PRIORITY; priority >= 0; priority--) {
		int end_zone = 0;	/* Inclusive.  0 = ZONE_DMA */
		unsigned long lru_pages = 0;

1077 1078 1079 1080
		/* The swap token gets in the way of swapout... */
		if (!priority)
			disable_swap_token();

L
Linus Torvalds 已提交
1081 1082
		all_zones_ok = 1;

1083 1084 1085 1086 1087 1088
		/*
		 * Scan in the highmem->dma direction for the highest
		 * zone which needs scanning
		 */
		for (i = pgdat->nr_zones - 1; i >= 0; i--) {
			struct zone *zone = pgdat->node_zones + i;
L
Linus Torvalds 已提交
1089

1090 1091
			if (!populated_zone(zone))
				continue;
L
Linus Torvalds 已提交
1092

1093 1094
			if (zone->all_unreclaimable && priority != DEF_PRIORITY)
				continue;
L
Linus Torvalds 已提交
1095

1096 1097 1098 1099
			if (!zone_watermark_ok(zone, order, zone->pages_high,
					       0, 0)) {
				end_zone = i;
				goto scan;
L
Linus Torvalds 已提交
1100 1101
			}
		}
1102
		goto out;
L
Linus Torvalds 已提交
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
scan:
		for (i = 0; i <= end_zone; i++) {
			struct zone *zone = pgdat->node_zones + i;

			lru_pages += zone->nr_active + zone->nr_inactive;
		}

		/*
		 * Now scan the zone in the dma->highmem direction, stopping
		 * at the last zone which needs scanning.
		 *
		 * We do this because the page allocator works in the opposite
		 * direction.  This prevents the page allocator from allocating
		 * pages behind kswapd's direction of progress, which would
		 * cause too much scanning of the lower zones.
		 */
		for (i = 0; i <= end_zone; i++) {
			struct zone *zone = pgdat->node_zones + i;
1121
			int nr_slab;
L
Linus Torvalds 已提交
1122

1123
			if (!populated_zone(zone))
L
Linus Torvalds 已提交
1124 1125 1126 1127 1128
				continue;

			if (zone->all_unreclaimable && priority != DEF_PRIORITY)
				continue;

1129 1130 1131
			if (!zone_watermark_ok(zone, order, zone->pages_high,
					       end_zone, 0))
				all_zones_ok = 0;
L
Linus Torvalds 已提交
1132 1133 1134 1135
			zone->temp_priority = priority;
			if (zone->prev_priority > priority)
				zone->prev_priority = priority;
			sc.nr_scanned = 0;
1136
			nr_reclaimed += shrink_zone(priority, zone, &sc);
L
Linus Torvalds 已提交
1137
			reclaim_state->reclaimed_slab = 0;
1138 1139
			nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL,
						lru_pages);
1140
			nr_reclaimed += reclaim_state->reclaimed_slab;
L
Linus Torvalds 已提交
1141 1142 1143
			total_scanned += sc.nr_scanned;
			if (zone->all_unreclaimable)
				continue;
1144 1145
			if (nr_slab == 0 && zone->pages_scanned >=
				    (zone->nr_active + zone->nr_inactive) * 4)
L
Linus Torvalds 已提交
1146 1147 1148 1149 1150 1151 1152
				zone->all_unreclaimable = 1;
			/*
			 * If we've done a decent amount of scanning and
			 * the reclaim ratio is low, start doing writepage
			 * even in laptop mode
			 */
			if (total_scanned > SWAP_CLUSTER_MAX * 2 &&
1153
			    total_scanned > nr_reclaimed + nr_reclaimed / 2)
L
Linus Torvalds 已提交
1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
				sc.may_writepage = 1;
		}
		if (all_zones_ok)
			break;		/* kswapd: all done */
		/*
		 * OK, kswapd is getting into trouble.  Take a nap, then take
		 * another pass across the zones.
		 */
		if (total_scanned && priority < DEF_PRIORITY - 2)
			blk_congestion_wait(WRITE, HZ/10);

		/*
		 * We do this so kswapd doesn't build up large priorities for
		 * example when it is freeing in parallel with allocators. It
		 * matches the direct reclaim path behaviour in terms of impact
		 * on zone->*_priority.
		 */
1171
		if (nr_reclaimed >= SWAP_CLUSTER_MAX)
L
Linus Torvalds 已提交
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184
			break;
	}
out:
	for (i = 0; i < pgdat->nr_zones; i++) {
		struct zone *zone = pgdat->node_zones + i;

		zone->prev_priority = zone->temp_priority;
	}
	if (!all_zones_ok) {
		cond_resched();
		goto loop_again;
	}

1185
	return nr_reclaimed;
L
Linus Torvalds 已提交
1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
}

/*
 * The background pageout daemon, started as a kernel thread
 * from the init process. 
 *
 * This basically trickles out pages so that we have _some_
 * free memory available even if there is no other activity
 * that frees anything up. This is needed for things like routing
 * etc, where we otherwise might have all activity going on in
 * asynchronous contexts that cannot page things out.
 *
 * If there are applications that are active memory-allocators
 * (most normal use), this basically shouldn't matter.
 */
static int kswapd(void *p)
{
	unsigned long order;
	pg_data_t *pgdat = (pg_data_t*)p;
	struct task_struct *tsk = current;
	DEFINE_WAIT(wait);
	struct reclaim_state reclaim_state = {
		.reclaimed_slab = 0,
	};
	cpumask_t cpumask;

	daemonize("kswapd%d", pgdat->node_id);
	cpumask = node_to_cpumask(pgdat->node_id);
	if (!cpus_empty(cpumask))
		set_cpus_allowed(tsk, cpumask);
	current->reclaim_state = &reclaim_state;

	/*
	 * Tell the memory management that we're a "memory allocator",
	 * and that if we need more memory we should get access to it
	 * regardless (see "__alloc_pages()"). "kswapd" should
	 * never get caught in the normal page freeing logic.
	 *
	 * (Kswapd normally doesn't need memory anyway, but sometimes
	 * you need a small amount of memory in order to be able to
	 * page out something else, and this flag essentially protects
	 * us from recursively trying to free more memory as we're
	 * trying to free the first piece of memory in the first place).
	 */
1230
	tsk->flags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD;
L
Linus Torvalds 已提交
1231 1232 1233 1234

	order = 0;
	for ( ; ; ) {
		unsigned long new_order;
1235 1236

		try_to_freeze();
L
Linus Torvalds 已提交
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252

		prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE);
		new_order = pgdat->kswapd_max_order;
		pgdat->kswapd_max_order = 0;
		if (order < new_order) {
			/*
			 * Don't sleep if someone wants a larger 'order'
			 * allocation
			 */
			order = new_order;
		} else {
			schedule();
			order = pgdat->kswapd_max_order;
		}
		finish_wait(&pgdat->kswapd_wait, &wait);

1253
		balance_pgdat(pgdat, order);
L
Linus Torvalds 已提交
1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264
	}
	return 0;
}

/*
 * A zone is low on free memory, so wake its kswapd task to service it.
 */
void wakeup_kswapd(struct zone *zone, int order)
{
	pg_data_t *pgdat;

1265
	if (!populated_zone(zone))
L
Linus Torvalds 已提交
1266 1267 1268
		return;

	pgdat = zone->zone_pgdat;
R
Rohit Seth 已提交
1269
	if (zone_watermark_ok(zone, order, zone->pages_low, 0, 0))
L
Linus Torvalds 已提交
1270 1271 1272
		return;
	if (pgdat->kswapd_max_order < order)
		pgdat->kswapd_max_order = order;
1273
	if (!cpuset_zone_allowed(zone, __GFP_HARDWALL))
L
Linus Torvalds 已提交
1274
		return;
1275
	if (!waitqueue_active(&pgdat->kswapd_wait))
L
Linus Torvalds 已提交
1276
		return;
1277
	wake_up_interruptible(&pgdat->kswapd_wait);
L
Linus Torvalds 已提交
1278 1279 1280 1281
}

#ifdef CONFIG_PM
/*
1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331
 * Helper function for shrink_all_memory().  Tries to reclaim 'nr_pages' pages
 * from LRU lists system-wide, for given pass and priority, and returns the
 * number of reclaimed pages
 *
 * For pass > 3 we also try to shrink the LRU lists that contain a few pages
 */
static unsigned long shrink_all_zones(unsigned long nr_pages, int pass,
				      int prio, struct scan_control *sc)
{
	struct zone *zone;
	unsigned long nr_to_scan, ret = 0;

	for_each_zone(zone) {

		if (!populated_zone(zone))
			continue;

		if (zone->all_unreclaimable && prio != DEF_PRIORITY)
			continue;

		/* For pass = 0 we don't shrink the active list */
		if (pass > 0) {
			zone->nr_scan_active += (zone->nr_active >> prio) + 1;
			if (zone->nr_scan_active >= nr_pages || pass > 3) {
				zone->nr_scan_active = 0;
				nr_to_scan = min(nr_pages, zone->nr_active);
				shrink_active_list(nr_to_scan, zone, sc);
			}
		}

		zone->nr_scan_inactive += (zone->nr_inactive >> prio) + 1;
		if (zone->nr_scan_inactive >= nr_pages || pass > 3) {
			zone->nr_scan_inactive = 0;
			nr_to_scan = min(nr_pages, zone->nr_inactive);
			ret += shrink_inactive_list(nr_to_scan, zone, sc);
			if (ret >= nr_pages)
				return ret;
		}
	}

	return ret;
}

/*
 * Try to free `nr_pages' of memory, system-wide, and return the number of
 * freed pages.
 *
 * Rather than trying to age LRUs the aim is to preserve the overall
 * LRU order by reclaiming preferentially
 * inactive > active > active referenced > active mapped
L
Linus Torvalds 已提交
1332
 */
1333
unsigned long shrink_all_memory(unsigned long nr_pages)
L
Linus Torvalds 已提交
1334
{
1335
	unsigned long lru_pages, nr_slab;
1336
	unsigned long ret = 0;
1337 1338 1339 1340 1341 1342 1343 1344 1345
	int pass;
	struct reclaim_state reclaim_state;
	struct zone *zone;
	struct scan_control sc = {
		.gfp_mask = GFP_KERNEL,
		.may_swap = 0,
		.swap_cluster_max = nr_pages,
		.may_writepage = 1,
		.swappiness = vm_swappiness,
L
Linus Torvalds 已提交
1346 1347 1348
	};

	current->reclaim_state = &reclaim_state;
1349

1350 1351 1352 1353 1354 1355 1356 1357 1358 1359
	lru_pages = 0;
	for_each_zone(zone)
		lru_pages += zone->nr_active + zone->nr_inactive;

	nr_slab = read_page_state(nr_slab);
	/* If slab caches are huge, it's better to hit them first */
	while (nr_slab >= lru_pages) {
		reclaim_state.reclaimed_slab = 0;
		shrink_slab(nr_pages, sc.gfp_mask, lru_pages);
		if (!reclaim_state.reclaimed_slab)
L
Linus Torvalds 已提交
1360
			break;
1361 1362 1363 1364 1365 1366

		ret += reclaim_state.reclaimed_slab;
		if (ret >= nr_pages)
			goto out;

		nr_slab -= reclaim_state.reclaimed_slab;
L
Linus Torvalds 已提交
1367
	}
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413

	/*
	 * We try to shrink LRUs in 5 passes:
	 * 0 = Reclaim from inactive_list only
	 * 1 = Reclaim from active list but don't reclaim mapped
	 * 2 = 2nd pass of type 1
	 * 3 = Reclaim mapped (normal reclaim)
	 * 4 = 2nd pass of type 3
	 */
	for (pass = 0; pass < 5; pass++) {
		int prio;

		/* Needed for shrinking slab caches later on */
		if (!lru_pages)
			for_each_zone(zone) {
				lru_pages += zone->nr_active;
				lru_pages += zone->nr_inactive;
			}

		/* Force reclaiming mapped pages in the passes #3 and #4 */
		if (pass > 2) {
			sc.may_swap = 1;
			sc.swappiness = 100;
		}

		for (prio = DEF_PRIORITY; prio >= 0; prio--) {
			unsigned long nr_to_scan = nr_pages - ret;

			sc.nr_mapped = read_page_state(nr_mapped);
			sc.nr_scanned = 0;

			ret += shrink_all_zones(nr_to_scan, prio, pass, &sc);
			if (ret >= nr_pages)
				goto out;

			reclaim_state.reclaimed_slab = 0;
			shrink_slab(sc.nr_scanned, sc.gfp_mask, lru_pages);
			ret += reclaim_state.reclaimed_slab;
			if (ret >= nr_pages)
				goto out;

			if (sc.nr_scanned && prio < DEF_PRIORITY - 2)
				blk_congestion_wait(WRITE, HZ / 10);
		}

		lru_pages = 0;
1414
	}
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427

	/*
	 * If ret = 0, we could not shrink LRUs, but there may be something
	 * in slab caches
	 */
	if (!ret)
		do {
			reclaim_state.reclaimed_slab = 0;
			shrink_slab(nr_pages, sc.gfp_mask, lru_pages);
			ret += reclaim_state.reclaimed_slab;
		} while (ret < nr_pages && reclaim_state.reclaimed_slab > 0);

out:
L
Linus Torvalds 已提交
1428
	current->reclaim_state = NULL;
1429

L
Linus Torvalds 已提交
1430 1431 1432 1433 1434 1435 1436 1437 1438
	return ret;
}
#endif

#ifdef CONFIG_HOTPLUG_CPU
/* It's optimal to keep kswapds on the same CPUs as their memory, but
   not required for correctness.  So if the last cpu in a node goes
   away, we get changed to run anywhere: as the first one comes back,
   restore their cpu bindings. */
1439
static int cpu_callback(struct notifier_block *nfb,
1440
				  unsigned long action, void *hcpu)
L
Linus Torvalds 已提交
1441 1442 1443 1444 1445
{
	pg_data_t *pgdat;
	cpumask_t mask;

	if (action == CPU_ONLINE) {
1446
		for_each_online_pgdat(pgdat) {
L
Linus Torvalds 已提交
1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459
			mask = node_to_cpumask(pgdat->node_id);
			if (any_online_cpu(mask) != NR_CPUS)
				/* One of our CPUs online: restore mask */
				set_cpus_allowed(pgdat->kswapd, mask);
		}
	}
	return NOTIFY_OK;
}
#endif /* CONFIG_HOTPLUG_CPU */

static int __init kswapd_init(void)
{
	pg_data_t *pgdat;
1460

L
Linus Torvalds 已提交
1461
	swap_setup();
1462
	for_each_online_pgdat(pgdat) {
1463 1464 1465 1466
		pid_t pid;

		pid = kernel_thread(kswapd, pgdat, CLONE_KERNEL);
		BUG_ON(pid < 0);
1467
		read_lock(&tasklist_lock);
1468
		pgdat->kswapd = find_task_by_pid(pid);
1469
		read_unlock(&tasklist_lock);
1470
	}
L
Linus Torvalds 已提交
1471 1472 1473 1474 1475 1476
	total_memory = nr_free_pagecache_pages();
	hotcpu_notifier(cpu_callback, 0);
	return 0;
}

module_init(kswapd_init)
1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490

#ifdef CONFIG_NUMA
/*
 * Zone reclaim mode
 *
 * If non-zero call zone_reclaim when the number of free pages falls below
 * the watermarks.
 *
 * In the future we may add flags to the mode. However, the page allocator
 * should only have to check that zone_reclaim_mode != 0 before calling
 * zone_reclaim().
 */
int zone_reclaim_mode __read_mostly;

1491 1492 1493 1494
#define RECLAIM_OFF 0
#define RECLAIM_ZONE (1<<0)	/* Run shrink_cache on the zone */
#define RECLAIM_WRITE (1<<1)	/* Writeout pages during reclaim */
#define RECLAIM_SWAP (1<<2)	/* Swap pages out during reclaim */
1495
#define RECLAIM_SLAB (1<<3)	/* Do a global slab shrink if the zone is out of memory */
1496

1497 1498 1499
/*
 * Mininum time between zone reclaim scans
 */
1500
int zone_reclaim_interval __read_mostly = 30*HZ;
1501 1502 1503 1504 1505 1506 1507 1508

/*
 * Priority for ZONE_RECLAIM. This determines the fraction of pages
 * of a node considered for each zone_reclaim. 4 scans 1/16th of
 * a zone.
 */
#define ZONE_RECLAIM_PRIORITY 4

1509 1510 1511
/*
 * Try to free up some pages from this zone through reclaim.
 */
1512
static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
1513
{
1514
	/* Minimum pages needed in order to stay on node */
1515
	const unsigned long nr_pages = 1 << order;
1516 1517
	struct task_struct *p = current;
	struct reclaim_state reclaim_state;
1518
	int priority;
1519
	unsigned long nr_reclaimed = 0;
1520 1521 1522 1523
	struct scan_control sc = {
		.may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE),
		.may_swap = !!(zone_reclaim_mode & RECLAIM_SWAP),
		.nr_mapped = read_page_state(nr_mapped),
1524 1525
		.swap_cluster_max = max_t(unsigned long, nr_pages,
					SWAP_CLUSTER_MAX),
1526
		.gfp_mask = gfp_mask,
1527
		.swappiness = vm_swappiness,
1528
	};
1529 1530 1531

	disable_swap_token();
	cond_resched();
1532 1533 1534 1535 1536 1537
	/*
	 * We need to be able to allocate from the reserves for RECLAIM_SWAP
	 * and we also need to be able to write out pages for RECLAIM_WRITE
	 * and RECLAIM_SWAP.
	 */
	p->flags |= PF_MEMALLOC | PF_SWAPWRITE;
1538 1539
	reclaim_state.reclaimed_slab = 0;
	p->reclaim_state = &reclaim_state;
1540

1541 1542 1543 1544
	/*
	 * Free memory by calling shrink zone with increasing priorities
	 * until we have enough memory freed.
	 */
1545
	priority = ZONE_RECLAIM_PRIORITY;
1546
	do {
1547
		nr_reclaimed += shrink_zone(priority, zone, &sc);
1548
		priority--;
1549
	} while (priority >= 0 && nr_reclaimed < nr_pages);
1550

1551
	if (nr_reclaimed < nr_pages && (zone_reclaim_mode & RECLAIM_SLAB)) {
1552
		/*
1553 1554 1555 1556 1557 1558
		 * shrink_slab() does not currently allow us to determine how
		 * many pages were freed in this zone. So we just shake the slab
		 * a bit and then go off node for this particular allocation
		 * despite possibly having freed enough memory to allocate in
		 * this zone.  If we freed local memory then the next
		 * allocations will be local again.
1559 1560 1561 1562 1563 1564 1565
		 *
		 * shrink_slab will free memory on all zones and may take
		 * a long time.
		 */
		shrink_slab(sc.nr_scanned, gfp_mask, order);
	}

1566
	p->reclaim_state = NULL;
1567
	current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
1568

1569 1570 1571 1572 1573 1574
	if (nr_reclaimed == 0) {
		/*
		 * We were unable to reclaim enough pages to stay on node.  We
		 * now allow off node accesses for a certain time period before
		 * trying again to reclaim pages from the local zone.
		 */
1575
		zone->last_unsuccessful_zone_reclaim = jiffies;
1576
	}
1577

1578
	return nr_reclaimed >= nr_pages;
1579
}
1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618

int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
{
	cpumask_t mask;
	int node_id;

	/*
	 * Do not reclaim if there was a recent unsuccessful attempt at zone
	 * reclaim.  In that case we let allocations go off node for the
	 * zone_reclaim_interval.  Otherwise we would scan for each off-node
	 * page allocation.
	 */
	if (time_before(jiffies,
		zone->last_unsuccessful_zone_reclaim + zone_reclaim_interval))
			return 0;

	/*
	 * Avoid concurrent zone reclaims, do not reclaim in a zone that does
	 * not have reclaimable pages and if we should not delay the allocation
	 * then do not scan.
	 */
	if (!(gfp_mask & __GFP_WAIT) ||
		zone->all_unreclaimable ||
		atomic_read(&zone->reclaim_in_progress) > 0 ||
		(current->flags & PF_MEMALLOC))
			return 0;

	/*
	 * Only run zone reclaim on the local zone or on zones that do not
	 * have associated processors. This will favor the local processor
	 * over remote processors and spread off node memory allocations
	 * as wide as possible.
	 */
	node_id = zone->zone_pgdat->node_id;
	mask = node_to_cpumask(node_id);
	if (!cpus_empty(mask) && node_id != numa_node_id())
		return 0;
	return __zone_reclaim(zone, gfp_mask, order);
}
1619
#endif