rmap.c 22.4 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
/*
 * mm/rmap.c - physical to virtual reverse mappings
 *
 * Copyright 2001, Rik van Riel <riel@conectiva.com.br>
 * Released under the General Public License (GPL).
 *
 * Simple, low overhead reverse mapping scheme.
 * Please try to keep this thing as modular as possible.
 *
 * Provides methods for unmapping each kind of mapped page:
 * the anon methods track anonymous pages, and
 * the file methods track pages belonging to an inode.
 *
 * Original design by Rik van Riel <riel@conectiva.com.br> 2001
 * File methods by Dave McCracken <dmccr@us.ibm.com> 2003, 2004
 * Anonymous methods by Andrea Arcangeli <andrea@suse.de> 2004
 * Contributions by Hugh Dickins <hugh@veritas.com> 2003, 2004
 */

/*
 * Lock ordering in mm:
 *
23
 * inode->i_mutex	(while writing or truncating, not reading or faulting)
L
Linus Torvalds 已提交
24 25 26
 *   inode->i_alloc_sem
 *
 * When a page fault occurs in writing from user to file, down_read
27 28 29
 * of mmap_sem nests within i_mutex; in sys_msync, i_mutex nests within
 * down_read of mmap_sem; i_mutex and down_write of mmap_sem are never
 * taken together; in truncation, i_mutex is taken outermost.
L
Linus Torvalds 已提交
30 31 32 33 34
 *
 * mm->mmap_sem
 *   page->flags PG_locked (lock_page)
 *     mapping->i_mmap_lock
 *       anon_vma->lock
35
 *         mm->page_table_lock or pte_lock
L
Linus Torvalds 已提交
36
 *           zone->lru_lock (in mark_page_accessed)
37
 *           swap_lock (in swap_duplicate, swap_info_get)
L
Linus Torvalds 已提交
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227
 *             mmlist_lock (in mmput, drain_mmlist and others)
 *             mapping->private_lock (in __set_page_dirty_buffers)
 *             inode_lock (in set_page_dirty's __mark_inode_dirty)
 *               sb_lock (within inode_lock in fs/fs-writeback.c)
 *               mapping->tree_lock (widely used, in set_page_dirty,
 *                         in arch-dependent flush_dcache_mmap_lock,
 *                         within inode_lock in __sync_single_inode)
 */

#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/swap.h>
#include <linux/swapops.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/rmap.h>
#include <linux/rcupdate.h>

#include <asm/tlbflush.h>

//#define RMAP_DEBUG /* can be enabled only for debugging */

kmem_cache_t *anon_vma_cachep;

static inline void validate_anon_vma(struct vm_area_struct *find_vma)
{
#ifdef RMAP_DEBUG
	struct anon_vma *anon_vma = find_vma->anon_vma;
	struct vm_area_struct *vma;
	unsigned int mapcount = 0;
	int found = 0;

	list_for_each_entry(vma, &anon_vma->head, anon_vma_node) {
		mapcount++;
		BUG_ON(mapcount > 100000);
		if (vma == find_vma)
			found = 1;
	}
	BUG_ON(!found);
#endif
}

/* This must be called under the mmap_sem. */
int anon_vma_prepare(struct vm_area_struct *vma)
{
	struct anon_vma *anon_vma = vma->anon_vma;

	might_sleep();
	if (unlikely(!anon_vma)) {
		struct mm_struct *mm = vma->vm_mm;
		struct anon_vma *allocated, *locked;

		anon_vma = find_mergeable_anon_vma(vma);
		if (anon_vma) {
			allocated = NULL;
			locked = anon_vma;
			spin_lock(&locked->lock);
		} else {
			anon_vma = anon_vma_alloc();
			if (unlikely(!anon_vma))
				return -ENOMEM;
			allocated = anon_vma;
			locked = NULL;
		}

		/* page_table_lock to protect against threads */
		spin_lock(&mm->page_table_lock);
		if (likely(!vma->anon_vma)) {
			vma->anon_vma = anon_vma;
			list_add(&vma->anon_vma_node, &anon_vma->head);
			allocated = NULL;
		}
		spin_unlock(&mm->page_table_lock);

		if (locked)
			spin_unlock(&locked->lock);
		if (unlikely(allocated))
			anon_vma_free(allocated);
	}
	return 0;
}

void __anon_vma_merge(struct vm_area_struct *vma, struct vm_area_struct *next)
{
	BUG_ON(vma->anon_vma != next->anon_vma);
	list_del(&next->anon_vma_node);
}

void __anon_vma_link(struct vm_area_struct *vma)
{
	struct anon_vma *anon_vma = vma->anon_vma;

	if (anon_vma) {
		list_add(&vma->anon_vma_node, &anon_vma->head);
		validate_anon_vma(vma);
	}
}

void anon_vma_link(struct vm_area_struct *vma)
{
	struct anon_vma *anon_vma = vma->anon_vma;

	if (anon_vma) {
		spin_lock(&anon_vma->lock);
		list_add(&vma->anon_vma_node, &anon_vma->head);
		validate_anon_vma(vma);
		spin_unlock(&anon_vma->lock);
	}
}

void anon_vma_unlink(struct vm_area_struct *vma)
{
	struct anon_vma *anon_vma = vma->anon_vma;
	int empty;

	if (!anon_vma)
		return;

	spin_lock(&anon_vma->lock);
	validate_anon_vma(vma);
	list_del(&vma->anon_vma_node);

	/* We must garbage collect the anon_vma if it's empty */
	empty = list_empty(&anon_vma->head);
	spin_unlock(&anon_vma->lock);

	if (empty)
		anon_vma_free(anon_vma);
}

static void anon_vma_ctor(void *data, kmem_cache_t *cachep, unsigned long flags)
{
	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
						SLAB_CTOR_CONSTRUCTOR) {
		struct anon_vma *anon_vma = data;

		spin_lock_init(&anon_vma->lock);
		INIT_LIST_HEAD(&anon_vma->head);
	}
}

void __init anon_vma_init(void)
{
	anon_vma_cachep = kmem_cache_create("anon_vma", sizeof(struct anon_vma),
			0, SLAB_DESTROY_BY_RCU|SLAB_PANIC, anon_vma_ctor, NULL);
}

/*
 * Getting a lock on a stable anon_vma from a page off the LRU is
 * tricky: page_lock_anon_vma rely on RCU to guard against the races.
 */
static struct anon_vma *page_lock_anon_vma(struct page *page)
{
	struct anon_vma *anon_vma = NULL;
	unsigned long anon_mapping;

	rcu_read_lock();
	anon_mapping = (unsigned long) page->mapping;
	if (!(anon_mapping & PAGE_MAPPING_ANON))
		goto out;
	if (!page_mapped(page))
		goto out;

	anon_vma = (struct anon_vma *) (anon_mapping - PAGE_MAPPING_ANON);
	spin_lock(&anon_vma->lock);
out:
	rcu_read_unlock();
	return anon_vma;
}

/*
 * At what user virtual address is page expected in vma?
 */
static inline unsigned long
vma_address(struct page *page, struct vm_area_struct *vma)
{
	pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
	unsigned long address;

	address = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
	if (unlikely(address < vma->vm_start || address >= vma->vm_end)) {
		/* page should be within any vma from prio_tree_next */
		BUG_ON(!PageAnon(page));
		return -EFAULT;
	}
	return address;
}

/*
 * At what user virtual address is page expected in vma? checking that the
H
Hugh Dickins 已提交
228
 * page matches the vma: currently only used on anon pages, by unuse_vma;
L
Linus Torvalds 已提交
229 230 231 232 233 234 235 236
 */
unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
{
	if (PageAnon(page)) {
		if ((void *)vma->anon_vma !=
		    (void *)page->mapping - PAGE_MAPPING_ANON)
			return -EFAULT;
	} else if (page->mapping && !(vma->vm_flags & VM_NONLINEAR)) {
H
Hugh Dickins 已提交
237 238
		if (!vma->vm_file ||
		    vma->vm_file->f_mapping != page->mapping)
L
Linus Torvalds 已提交
239 240 241 242 243 244
			return -EFAULT;
	} else
		return -EFAULT;
	return vma_address(page, vma);
}

N
Nikita Danilov 已提交
245 246 247
/*
 * Check that @page is mapped at @address into @mm.
 *
248
 * On success returns with pte mapped and locked.
N
Nikita Danilov 已提交
249
 */
250
pte_t *page_check_address(struct page *page, struct mm_struct *mm,
H
Hugh Dickins 已提交
251
			  unsigned long address, spinlock_t **ptlp)
N
Nikita Danilov 已提交
252 253 254 255 256
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
H
Hugh Dickins 已提交
257
	spinlock_t *ptl;
N
Nikita Danilov 已提交
258 259

	pgd = pgd_offset(mm, address);
H
Hugh Dickins 已提交
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277
	if (!pgd_present(*pgd))
		return NULL;

	pud = pud_offset(pgd, address);
	if (!pud_present(*pud))
		return NULL;

	pmd = pmd_offset(pud, address);
	if (!pmd_present(*pmd))
		return NULL;

	pte = pte_offset_map(pmd, address);
	/* Make a quick check before getting the lock */
	if (!pte_present(*pte)) {
		pte_unmap(pte);
		return NULL;
	}

H
Hugh Dickins 已提交
278
	ptl = pte_lockptr(mm, pmd);
H
Hugh Dickins 已提交
279 280 281 282
	spin_lock(ptl);
	if (pte_present(*pte) && page_to_pfn(page) == pte_pfn(*pte)) {
		*ptlp = ptl;
		return pte;
N
Nikita Danilov 已提交
283
	}
H
Hugh Dickins 已提交
284 285
	pte_unmap_unlock(pte, ptl);
	return NULL;
N
Nikita Danilov 已提交
286 287
}

L
Linus Torvalds 已提交
288 289 290 291 292
/*
 * Subfunctions of page_referenced: page_referenced_one called
 * repeatedly from either page_referenced_anon or page_referenced_file.
 */
static int page_referenced_one(struct page *page,
293
	struct vm_area_struct *vma, unsigned int *mapcount)
L
Linus Torvalds 已提交
294 295 296 297
{
	struct mm_struct *mm = vma->vm_mm;
	unsigned long address;
	pte_t *pte;
H
Hugh Dickins 已提交
298
	spinlock_t *ptl;
L
Linus Torvalds 已提交
299 300 301 302 303 304
	int referenced = 0;

	address = vma_address(page, vma);
	if (address == -EFAULT)
		goto out;

H
Hugh Dickins 已提交
305 306 307
	pte = page_check_address(page, mm, address, &ptl);
	if (!pte)
		goto out;
L
Linus Torvalds 已提交
308

H
Hugh Dickins 已提交
309 310
	if (ptep_clear_flush_young(vma, address, pte))
		referenced++;
L
Linus Torvalds 已提交
311

H
Hugh Dickins 已提交
312 313
	/* Pretend the page is referenced if the task has the
	   swap token and is in the middle of a page fault. */
314
	if (mm != current->mm && has_swap_token(mm) &&
H
Hugh Dickins 已提交
315 316 317 318 319
			rwsem_is_locked(&mm->mmap_sem))
		referenced++;

	(*mapcount)--;
	pte_unmap_unlock(pte, ptl);
L
Linus Torvalds 已提交
320 321 322 323
out:
	return referenced;
}

324
static int page_referenced_anon(struct page *page)
L
Linus Torvalds 已提交
325 326 327 328 329 330 331 332 333 334 335 336
{
	unsigned int mapcount;
	struct anon_vma *anon_vma;
	struct vm_area_struct *vma;
	int referenced = 0;

	anon_vma = page_lock_anon_vma(page);
	if (!anon_vma)
		return referenced;

	mapcount = page_mapcount(page);
	list_for_each_entry(vma, &anon_vma->head, anon_vma_node) {
337
		referenced += page_referenced_one(page, vma, &mapcount);
L
Linus Torvalds 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355
		if (!mapcount)
			break;
	}
	spin_unlock(&anon_vma->lock);
	return referenced;
}

/**
 * page_referenced_file - referenced check for object-based rmap
 * @page: the page we're checking references on.
 *
 * For an object-based mapped page, find all the places it is mapped and
 * check/clear the referenced flag.  This is done by following the page->mapping
 * pointer, then walking the chain of vmas it holds.  It returns the number
 * of references it found.
 *
 * This function is only called from page_referenced for object-based pages.
 */
356
static int page_referenced_file(struct page *page)
L
Linus Torvalds 已提交
357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
{
	unsigned int mapcount;
	struct address_space *mapping = page->mapping;
	pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
	struct vm_area_struct *vma;
	struct prio_tree_iter iter;
	int referenced = 0;

	/*
	 * The caller's checks on page->mapping and !PageAnon have made
	 * sure that this is a file page: the check for page->mapping
	 * excludes the case just before it gets set on an anon page.
	 */
	BUG_ON(PageAnon(page));

	/*
	 * The page lock not only makes sure that page->mapping cannot
	 * suddenly be NULLified by truncation, it makes sure that the
	 * structure at mapping cannot be freed and reused yet,
	 * so we can safely take mapping->i_mmap_lock.
	 */
	BUG_ON(!PageLocked(page));

	spin_lock(&mapping->i_mmap_lock);

	/*
	 * i_mmap_lock does not stabilize mapcount at all, but mapcount
	 * is more likely to be accurate if we note it after spinning.
	 */
	mapcount = page_mapcount(page);

	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
		if ((vma->vm_flags & (VM_LOCKED|VM_MAYSHARE))
				  == (VM_LOCKED|VM_MAYSHARE)) {
			referenced++;
			break;
		}
394
		referenced += page_referenced_one(page, vma, &mapcount);
L
Linus Torvalds 已提交
395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
		if (!mapcount)
			break;
	}

	spin_unlock(&mapping->i_mmap_lock);
	return referenced;
}

/**
 * page_referenced - test if the page was referenced
 * @page: the page to test
 * @is_locked: caller holds lock on the page
 *
 * Quick test_and_clear_referenced for all mappings to a page,
 * returns the number of ptes which referenced the page.
 */
411
int page_referenced(struct page *page, int is_locked)
L
Linus Torvalds 已提交
412 413 414 415 416 417 418 419 420 421 422
{
	int referenced = 0;

	if (page_test_and_clear_young(page))
		referenced++;

	if (TestClearPageReferenced(page))
		referenced++;

	if (page_mapped(page) && page->mapping) {
		if (PageAnon(page))
423
			referenced += page_referenced_anon(page);
L
Linus Torvalds 已提交
424
		else if (is_locked)
425
			referenced += page_referenced_file(page);
L
Linus Torvalds 已提交
426 427 428 429
		else if (TestSetPageLocked(page))
			referenced++;
		else {
			if (page->mapping)
430
				referenced += page_referenced_file(page);
L
Linus Torvalds 已提交
431 432 433 434 435 436
			unlock_page(page);
		}
	}
	return referenced;
}

N
Nick Piggin 已提交
437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453
/**
 * page_set_anon_rmap - setup new anonymous rmap
 * @page:	the page to add the mapping to
 * @vma:	the vm area in which the mapping is added
 * @address:	the user virtual address mapped
 */
static void __page_set_anon_rmap(struct page *page,
	struct vm_area_struct *vma, unsigned long address)
{
	struct anon_vma *anon_vma = vma->anon_vma;

	BUG_ON(!anon_vma);
	anon_vma = (void *) anon_vma + PAGE_MAPPING_ANON;
	page->mapping = (struct address_space *) anon_vma;

	page->index = linear_page_index(vma, address);

N
Nick Piggin 已提交
454 455 456 457 458
	/*
	 * nr_mapped state can be updated without turning off
	 * interrupts because it is not modified via interrupt.
	 */
	__inc_page_state(nr_mapped);
N
Nick Piggin 已提交
459 460
}

L
Linus Torvalds 已提交
461 462 463 464 465 466
/**
 * page_add_anon_rmap - add pte mapping to an anonymous page
 * @page:	the page to add the mapping to
 * @vma:	the vm area in which the mapping is added
 * @address:	the user virtual address mapped
 *
467
 * The caller needs to hold the pte lock.
L
Linus Torvalds 已提交
468 469 470 471
 */
void page_add_anon_rmap(struct page *page,
	struct vm_area_struct *vma, unsigned long address)
{
N
Nick Piggin 已提交
472 473
	if (atomic_inc_and_test(&page->_mapcount))
		__page_set_anon_rmap(page, vma, address);
L
Linus Torvalds 已提交
474 475 476
	/* else checking page index and mapping is racy */
}

N
Nick Piggin 已提交
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492
/*
 * page_add_new_anon_rmap - add pte mapping to a new anonymous page
 * @page:	the page to add the mapping to
 * @vma:	the vm area in which the mapping is added
 * @address:	the user virtual address mapped
 *
 * Same as page_add_anon_rmap but must only be called on *new* pages.
 * This means the inc-and-test can be bypassed.
 */
void page_add_new_anon_rmap(struct page *page,
	struct vm_area_struct *vma, unsigned long address)
{
	atomic_set(&page->_mapcount, 0); /* elevate count by 1 (starts at -1) */
	__page_set_anon_rmap(page, vma, address);
}

L
Linus Torvalds 已提交
493 494 495 496
/**
 * page_add_file_rmap - add pte mapping to a file page
 * @page: the page to add the mapping to
 *
497
 * The caller needs to hold the pte lock.
L
Linus Torvalds 已提交
498 499 500 501
 */
void page_add_file_rmap(struct page *page)
{
	BUG_ON(PageAnon(page));
N
Nick Piggin 已提交
502
	BUG_ON(!pfn_valid(page_to_pfn(page)));
L
Linus Torvalds 已提交
503 504

	if (atomic_inc_and_test(&page->_mapcount))
N
Nick Piggin 已提交
505
		__inc_page_state(nr_mapped);
L
Linus Torvalds 已提交
506 507 508 509 510 511
}

/**
 * page_remove_rmap - take down pte mapping from a page
 * @page: page to remove mapping from
 *
512
 * The caller needs to hold the pte lock.
L
Linus Torvalds 已提交
513 514 515 516
 */
void page_remove_rmap(struct page *page)
{
	if (atomic_add_negative(-1, &page->_mapcount)) {
517 518 519 520 521 522 523
		if (page_mapcount(page) < 0) {
			printk (KERN_EMERG "Eeek! page_mapcount(page) went negative! (%d)\n", page_mapcount(page));
			printk (KERN_EMERG "  page->flags = %lx\n", page->flags);
			printk (KERN_EMERG "  page->count = %x\n", page_count(page));
			printk (KERN_EMERG "  page->mapping = %p\n", page->mapping);
		}

L
Linus Torvalds 已提交
524 525 526 527 528 529 530 531 532 533 534 535
		BUG_ON(page_mapcount(page) < 0);
		/*
		 * It would be tidy to reset the PageAnon mapping here,
		 * but that might overwrite a racing page_add_anon_rmap
		 * which increments mapcount after us but sets mapping
		 * before us: so leave the reset to free_hot_cold_page,
		 * and remember that it's only reliable while mapped.
		 * Leaving it set also helps swapoff to reinstate ptes
		 * faster for those pages still in swapcache.
		 */
		if (page_test_and_clear_dirty(page))
			set_page_dirty(page);
N
Nick Piggin 已提交
536
		__dec_page_state(nr_mapped);
L
Linus Torvalds 已提交
537 538 539 540 541 542 543 544 545 546 547 548 549
	}
}

/*
 * Subfunctions of try_to_unmap: try_to_unmap_one called
 * repeatedly from either try_to_unmap_anon or try_to_unmap_file.
 */
static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma)
{
	struct mm_struct *mm = vma->vm_mm;
	unsigned long address;
	pte_t *pte;
	pte_t pteval;
H
Hugh Dickins 已提交
550
	spinlock_t *ptl;
L
Linus Torvalds 已提交
551 552 553 554 555 556
	int ret = SWAP_AGAIN;

	address = vma_address(page, vma);
	if (address == -EFAULT)
		goto out;

H
Hugh Dickins 已提交
557 558
	pte = page_check_address(page, mm, address, &ptl);
	if (!pte)
N
Nikita Danilov 已提交
559
		goto out;
L
Linus Torvalds 已提交
560 561 562 563 564 565

	/*
	 * If the page is mlock()d, we cannot swap it out.
	 * If it's recently referenced (perhaps page_referenced
	 * skipped over this mm) then we should reactivate it.
	 */
566
	if ((vma->vm_flags & VM_LOCKED) ||
L
Linus Torvalds 已提交
567 568 569 570 571 572 573 574 575 576 577 578 579
			ptep_clear_flush_young(vma, address, pte)) {
		ret = SWAP_FAIL;
		goto out_unmap;
	}

	/* Nuke the page table entry. */
	flush_cache_page(vma, address, page_to_pfn(page));
	pteval = ptep_clear_flush(vma, address, pte);

	/* Move the dirty bit to the physical page now the pte is gone. */
	if (pte_dirty(pteval))
		set_page_dirty(page);

580 581 582
	/* Update high watermark before we lower rss */
	update_hiwater_rss(mm);

L
Linus Torvalds 已提交
583
	if (PageAnon(page)) {
H
Hugh Dickins 已提交
584
		swp_entry_t entry = { .val = page_private(page) };
L
Linus Torvalds 已提交
585 586 587 588 589 590 591 592
		/*
		 * Store the swap location in the pte.
		 * See handle_pte_fault() ...
		 */
		BUG_ON(!PageSwapCache(page));
		swap_duplicate(entry);
		if (list_empty(&mm->mmlist)) {
			spin_lock(&mmlist_lock);
593 594
			if (list_empty(&mm->mmlist))
				list_add(&mm->mmlist, &init_mm.mmlist);
L
Linus Torvalds 已提交
595 596 597 598 599
			spin_unlock(&mmlist_lock);
		}
		set_pte_at(mm, address, pte, swp_entry_to_pte(entry));
		BUG_ON(pte_file(*pte));
		dec_mm_counter(mm, anon_rss);
600 601
	} else
		dec_mm_counter(mm, file_rss);
L
Linus Torvalds 已提交
602 603 604 605 606

	page_remove_rmap(page);
	page_cache_release(page);

out_unmap:
H
Hugh Dickins 已提交
607
	pte_unmap_unlock(pte, ptl);
L
Linus Torvalds 已提交
608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
out:
	return ret;
}

/*
 * objrmap doesn't work for nonlinear VMAs because the assumption that
 * offset-into-file correlates with offset-into-virtual-addresses does not hold.
 * Consequently, given a particular page and its ->index, we cannot locate the
 * ptes which are mapping that page without an exhaustive linear search.
 *
 * So what this code does is a mini "virtual scan" of each nonlinear VMA which
 * maps the file to which the target page belongs.  The ->vm_private_data field
 * holds the current cursor into that scan.  Successive searches will circulate
 * around the vma's virtual address space.
 *
 * So as more replacement pressure is applied to the pages in a nonlinear VMA,
 * more scanning pressure is placed against them as well.   Eventually pages
 * will become fully unmapped and are eligible for eviction.
 *
 * For very sparsely populated VMAs this is a little inefficient - chances are
 * there there won't be many ptes located within the scan cluster.  In this case
 * maybe we could scan further - to the end of the pte page, perhaps.
 */
#define CLUSTER_SIZE	min(32*PAGE_SIZE, PMD_SIZE)
#define CLUSTER_MASK	(~(CLUSTER_SIZE - 1))

static void try_to_unmap_cluster(unsigned long cursor,
	unsigned int *mapcount, struct vm_area_struct *vma)
{
	struct mm_struct *mm = vma->vm_mm;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
H
Hugh Dickins 已提交
641
	pte_t *pte;
L
Linus Torvalds 已提交
642
	pte_t pteval;
H
Hugh Dickins 已提交
643
	spinlock_t *ptl;
L
Linus Torvalds 已提交
644 645 646 647 648 649 650 651 652 653 654 655 656
	struct page *page;
	unsigned long address;
	unsigned long end;

	address = (vma->vm_start + cursor) & CLUSTER_MASK;
	end = address + CLUSTER_SIZE;
	if (address < vma->vm_start)
		address = vma->vm_start;
	if (end > vma->vm_end)
		end = vma->vm_end;

	pgd = pgd_offset(mm, address);
	if (!pgd_present(*pgd))
H
Hugh Dickins 已提交
657
		return;
L
Linus Torvalds 已提交
658 659 660

	pud = pud_offset(pgd, address);
	if (!pud_present(*pud))
H
Hugh Dickins 已提交
661
		return;
L
Linus Torvalds 已提交
662 663 664

	pmd = pmd_offset(pud, address);
	if (!pmd_present(*pmd))
H
Hugh Dickins 已提交
665 666 667
		return;

	pte = pte_offset_map_lock(mm, pmd, address, &ptl);
L
Linus Torvalds 已提交
668

669 670 671
	/* Update high watermark before we lower rss */
	update_hiwater_rss(mm);

H
Hugh Dickins 已提交
672
	for (; address < end; pte++, address += PAGE_SIZE) {
L
Linus Torvalds 已提交
673 674
		if (!pte_present(*pte))
			continue;
675 676
		page = vm_normal_page(vma, address, *pte);
		BUG_ON(!page || PageAnon(page));
L
Linus Torvalds 已提交
677 678 679 680 681

		if (ptep_clear_flush_young(vma, address, pte))
			continue;

		/* Nuke the page table entry. */
682
		flush_cache_page(vma, address, pte_pfn(*pte));
L
Linus Torvalds 已提交
683 684 685 686 687 688 689 690 691 692 693 694
		pteval = ptep_clear_flush(vma, address, pte);

		/* If nonlinear, store the file page offset in the pte. */
		if (page->index != linear_page_index(vma, address))
			set_pte_at(mm, address, pte, pgoff_to_pte(page->index));

		/* Move the dirty bit to the physical page now the pte is gone. */
		if (pte_dirty(pteval))
			set_page_dirty(page);

		page_remove_rmap(page);
		page_cache_release(page);
695
		dec_mm_counter(mm, file_rss);
L
Linus Torvalds 已提交
696 697
		(*mapcount)--;
	}
H
Hugh Dickins 已提交
698
	pte_unmap_unlock(pte - 1, ptl);
L
Linus Torvalds 已提交
699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752
}

static int try_to_unmap_anon(struct page *page)
{
	struct anon_vma *anon_vma;
	struct vm_area_struct *vma;
	int ret = SWAP_AGAIN;

	anon_vma = page_lock_anon_vma(page);
	if (!anon_vma)
		return ret;

	list_for_each_entry(vma, &anon_vma->head, anon_vma_node) {
		ret = try_to_unmap_one(page, vma);
		if (ret == SWAP_FAIL || !page_mapped(page))
			break;
	}
	spin_unlock(&anon_vma->lock);
	return ret;
}

/**
 * try_to_unmap_file - unmap file page using the object-based rmap method
 * @page: the page to unmap
 *
 * Find all the mappings of a page using the mapping pointer and the vma chains
 * contained in the address_space struct it points to.
 *
 * This function is only called from try_to_unmap for object-based pages.
 */
static int try_to_unmap_file(struct page *page)
{
	struct address_space *mapping = page->mapping;
	pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
	struct vm_area_struct *vma;
	struct prio_tree_iter iter;
	int ret = SWAP_AGAIN;
	unsigned long cursor;
	unsigned long max_nl_cursor = 0;
	unsigned long max_nl_size = 0;
	unsigned int mapcount;

	spin_lock(&mapping->i_mmap_lock);
	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
		ret = try_to_unmap_one(page, vma);
		if (ret == SWAP_FAIL || !page_mapped(page))
			goto out;
	}

	if (list_empty(&mapping->i_mmap_nonlinear))
		goto out;

	list_for_each_entry(vma, &mapping->i_mmap_nonlinear,
						shared.vm_set.list) {
753
		if (vma->vm_flags & VM_LOCKED)
L
Linus Torvalds 已提交
754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786
			continue;
		cursor = (unsigned long) vma->vm_private_data;
		if (cursor > max_nl_cursor)
			max_nl_cursor = cursor;
		cursor = vma->vm_end - vma->vm_start;
		if (cursor > max_nl_size)
			max_nl_size = cursor;
	}

	if (max_nl_size == 0) {	/* any nonlinears locked or reserved */
		ret = SWAP_FAIL;
		goto out;
	}

	/*
	 * We don't try to search for this page in the nonlinear vmas,
	 * and page_referenced wouldn't have found it anyway.  Instead
	 * just walk the nonlinear vmas trying to age and unmap some.
	 * The mapcount of the page we came in with is irrelevant,
	 * but even so use it as a guide to how hard we should try?
	 */
	mapcount = page_mapcount(page);
	if (!mapcount)
		goto out;
	cond_resched_lock(&mapping->i_mmap_lock);

	max_nl_size = (max_nl_size + CLUSTER_SIZE - 1) & CLUSTER_MASK;
	if (max_nl_cursor == 0)
		max_nl_cursor = CLUSTER_SIZE;

	do {
		list_for_each_entry(vma, &mapping->i_mmap_nonlinear,
						shared.vm_set.list) {
787
			if (vma->vm_flags & VM_LOCKED)
L
Linus Torvalds 已提交
788 789
				continue;
			cursor = (unsigned long) vma->vm_private_data;
H
Hugh Dickins 已提交
790
			while ( cursor < max_nl_cursor &&
L
Linus Torvalds 已提交
791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808
				cursor < vma->vm_end - vma->vm_start) {
				try_to_unmap_cluster(cursor, &mapcount, vma);
				cursor += CLUSTER_SIZE;
				vma->vm_private_data = (void *) cursor;
				if ((int)mapcount <= 0)
					goto out;
			}
			vma->vm_private_data = (void *) max_nl_cursor;
		}
		cond_resched_lock(&mapping->i_mmap_lock);
		max_nl_cursor += CLUSTER_SIZE;
	} while (max_nl_cursor <= max_nl_size);

	/*
	 * Don't loop forever (perhaps all the remaining pages are
	 * in locked vmas).  Reset cursor on all unreserved nonlinear
	 * vmas, now forgetting on which ones it had fallen behind.
	 */
809 810
	list_for_each_entry(vma, &mapping->i_mmap_nonlinear, shared.vm_set.list)
		vma->vm_private_data = NULL;
L
Linus Torvalds 已提交
811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
out:
	spin_unlock(&mapping->i_mmap_lock);
	return ret;
}

/**
 * try_to_unmap - try to remove all page table mappings to a page
 * @page: the page to get unmapped
 *
 * Tries to remove all the page table entries which are mapping this
 * page, used in the pageout path.  Caller must hold the page lock.
 * Return values are:
 *
 * SWAP_SUCCESS	- we succeeded in removing all mappings
 * SWAP_AGAIN	- we missed a mapping, try again later
 * SWAP_FAIL	- the page is unswappable
 */
int try_to_unmap(struct page *page)
{
	int ret;

	BUG_ON(!PageLocked(page));

	if (PageAnon(page))
		ret = try_to_unmap_anon(page);
	else
		ret = try_to_unmap_file(page);

	if (!page_mapped(page))
		ret = SWAP_SUCCESS;
	return ret;
}
N
Nikita Danilov 已提交
843