migrate.c 25.6 KB
Newer Older
C
Christoph Lameter 已提交
1 2 3 4 5 6 7 8 9 10 11
/*
 * Memory Migration functionality - linux/mm/migration.c
 *
 * Copyright (C) 2006 Silicon Graphics, Inc., Christoph Lameter
 *
 * Page migration was first developed in the context of the memory hotplug
 * project. The main authors of the migration code are:
 *
 * IWAMOTO Toshihiro <iwamoto@valinux.co.jp>
 * Hirokazu Takahashi <taka@valinux.co.jp>
 * Dave Hansen <haveblue@us.ibm.com>
C
Christoph Lameter 已提交
12
 * Christoph Lameter
C
Christoph Lameter 已提交
13 14 15 16 17
 */

#include <linux/migrate.h>
#include <linux/module.h>
#include <linux/swap.h>
18
#include <linux/swapops.h>
C
Christoph Lameter 已提交
19
#include <linux/pagemap.h>
20
#include <linux/buffer_head.h>
C
Christoph Lameter 已提交
21
#include <linux/mm_inline.h>
22
#include <linux/nsproxy.h>
C
Christoph Lameter 已提交
23 24 25 26 27
#include <linux/pagevec.h>
#include <linux/rmap.h>
#include <linux/topology.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
28
#include <linux/writeback.h>
29 30
#include <linux/mempolicy.h>
#include <linux/vmalloc.h>
31
#include <linux/security.h>
32
#include <linux/memcontrol.h>
33
#include <linux/syscalls.h>
C
Christoph Lameter 已提交
34 35 36 37 38 39

#include "internal.h"

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

/*
40 41
 * migrate_prep() needs to be called before we start compiling a list of pages
 * to be migrated using isolate_lru_page().
C
Christoph Lameter 已提交
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
 */
int migrate_prep(void)
{
	/*
	 * Clear the LRU lists so pages can be isolated.
	 * Note that pages may be moved off the LRU after we have
	 * drained them. Those pages will fail to migrate like other
	 * pages that may be busy.
	 */
	lru_add_drain_all();

	return 0;
}

/*
L
Lee Schermerhorn 已提交
57 58
 * Add isolated pages on the list back to the LRU under page lock
 * to avoid leaking evictable pages back onto unevictable list.
C
Christoph Lameter 已提交
59 60 61 62 63 64 65 66 67 68
 *
 * returns the number of pages put back.
 */
int putback_lru_pages(struct list_head *l)
{
	struct page *page;
	struct page *page2;
	int count = 0;

	list_for_each_entry_safe(page, page2, l, lru) {
69
		list_del(&page->lru);
L
Lee Schermerhorn 已提交
70
		putback_lru_page(page);
C
Christoph Lameter 已提交
71 72 73 74 75
		count++;
	}
	return count;
}

76 77 78
/*
 * Restore a potential migration pte to a working pte entry
 */
79
static void remove_migration_pte(struct vm_area_struct *vma,
80 81 82 83 84 85 86 87 88
		struct page *old, struct page *new)
{
	struct mm_struct *mm = vma->vm_mm;
	swp_entry_t entry;
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
	pte_t *ptep, pte;
 	spinlock_t *ptl;
89 90 91 92
	unsigned long addr = page_address_in_vma(new, vma);

	if (addr == -EFAULT)
		return;
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

 	pgd = pgd_offset(mm, addr);
	if (!pgd_present(*pgd))
                return;

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

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

	ptep = pte_offset_map(pmd, addr);

	if (!is_swap_pte(*ptep)) {
		pte_unmap(ptep);
 		return;
 	}

 	ptl = pte_lockptr(mm, pmd);
 	spin_lock(ptl);
	pte = *ptep;
	if (!is_swap_pte(pte))
		goto out;

	entry = pte_to_swp_entry(pte);

	if (!is_migration_entry(entry) || migration_entry_to_page(entry) != old)
		goto out;

124 125 126 127 128 129 130 131 132 133 134 135 136 137
	/*
	 * Yes, ignore the return value from a GFP_ATOMIC mem_cgroup_charge.
	 * Failure is not an option here: we're now expected to remove every
	 * migration pte, and will cause crashes otherwise.  Normally this
	 * is not an issue: mem_cgroup_prepare_migration bumped up the old
	 * page_cgroup count for safety, that's now attached to the new page,
	 * so this charge should just be another incrementation of the count,
	 * to keep in balance with rmap.c's mem_cgroup_uncharging.  But if
	 * there's been a force_empty, those reference counts may no longer
	 * be reliable, and this charge can actually fail: oh well, we don't
	 * make the situation any worse by proceeding as if it had succeeded.
	 */
	mem_cgroup_charge(new, mm, GFP_ATOMIC);

138 139 140 141
	get_page(new);
	pte = pte_mkold(mk_pte(new, vma->vm_page_prot));
	if (is_write_migration_entry(entry))
		pte = pte_mkwrite(pte);
142
	flush_cache_page(vma, addr, pte_pfn(pte));
143
	set_pte_at(mm, addr, ptep, pte);
144 145 146 147 148 149 150 151 152

	if (PageAnon(new))
		page_add_anon_rmap(new, vma, addr);
	else
		page_add_file_rmap(new);

	/* No need to invalidate - it was non-present before */
	update_mmu_cache(vma, addr, pte);

153 154 155 156 157
out:
	pte_unmap_unlock(ptep, ptl);
}

/*
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179
 * Note that remove_file_migration_ptes will only work on regular mappings,
 * Nonlinear mappings do not use migration entries.
 */
static void remove_file_migration_ptes(struct page *old, struct page *new)
{
	struct vm_area_struct *vma;
	struct address_space *mapping = page_mapping(new);
	struct prio_tree_iter iter;
	pgoff_t pgoff = new->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);

	if (!mapping)
		return;

	spin_lock(&mapping->i_mmap_lock);

	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff)
		remove_migration_pte(vma, old, new);

	spin_unlock(&mapping->i_mmap_lock);
}

/*
180 181 182
 * Must hold mmap_sem lock on at least one of the vmas containing
 * the page so that the anon_vma cannot vanish.
 */
183
static void remove_anon_migration_ptes(struct page *old, struct page *new)
184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
{
	struct anon_vma *anon_vma;
	struct vm_area_struct *vma;
	unsigned long mapping;

	mapping = (unsigned long)new->mapping;

	if (!mapping || (mapping & PAGE_MAPPING_ANON) == 0)
		return;

	/*
	 * We hold the mmap_sem lock. So no need to call page_lock_anon_vma.
	 */
	anon_vma = (struct anon_vma *) (mapping - PAGE_MAPPING_ANON);
	spin_lock(&anon_vma->lock);

	list_for_each_entry(vma, &anon_vma->head, anon_vma_node)
201
		remove_migration_pte(vma, old, new);
202 203 204 205

	spin_unlock(&anon_vma->lock);
}

206 207 208 209 210 211 212 213 214 215 216 217
/*
 * Get rid of all migration entries and replace them by
 * references to the indicated page.
 */
static void remove_migration_ptes(struct page *old, struct page *new)
{
	if (PageAnon(new))
		remove_anon_migration_ptes(old, new);
	else
		remove_file_migration_ptes(old, new);
}

218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
/*
 * Something used the pte of a page under migration. We need to
 * get to the page and wait until migration is finished.
 * When we return from this function the fault will be retried.
 *
 * This function is called from do_swap_page().
 */
void migration_entry_wait(struct mm_struct *mm, pmd_t *pmd,
				unsigned long address)
{
	pte_t *ptep, pte;
	spinlock_t *ptl;
	swp_entry_t entry;
	struct page *page;

	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
	pte = *ptep;
	if (!is_swap_pte(pte))
		goto out;

	entry = pte_to_swp_entry(pte);
	if (!is_migration_entry(entry))
		goto out;

	page = migration_entry_to_page(entry);

N
Nick Piggin 已提交
244 245 246 247 248 249 250 251 252
	/*
	 * Once radix-tree replacement of page migration started, page_count
	 * *must* be zero. And, we don't want to call wait_on_page_locked()
	 * against a page without get_page().
	 * So, we use get_page_unless_zero(), here. Even failed, page fault
	 * will occur again.
	 */
	if (!get_page_unless_zero(page))
		goto out;
253 254 255 256 257 258 259 260
	pte_unmap_unlock(ptep, ptl);
	wait_on_page_locked(page);
	put_page(page);
	return;
out:
	pte_unmap_unlock(ptep, ptl);
}

C
Christoph Lameter 已提交
261
/*
262
 * Replace the page in the mapping.
263 264 265 266 267
 *
 * The number of remaining references must be:
 * 1 for anonymous pages without a mapping
 * 2 for pages with a mapping
 * 3 for pages with a mapping and PagePrivate set.
C
Christoph Lameter 已提交
268
 */
269 270
static int migrate_page_move_mapping(struct address_space *mapping,
		struct page *newpage, struct page *page)
C
Christoph Lameter 已提交
271
{
N
Nick Piggin 已提交
272
	int expected_count;
273
	void **pslot;
C
Christoph Lameter 已提交
274

275
	if (!mapping) {
276
		/* Anonymous page without mapping */
277 278 279 280 281
		if (page_count(page) != 1)
			return -EAGAIN;
		return 0;
	}

N
Nick Piggin 已提交
282
	spin_lock_irq(&mapping->tree_lock);
C
Christoph Lameter 已提交
283

284 285
	pslot = radix_tree_lookup_slot(&mapping->page_tree,
 					page_index(page));
C
Christoph Lameter 已提交
286

N
Nick Piggin 已提交
287 288
	expected_count = 2 + !!PagePrivate(page);
	if (page_count(page) != expected_count ||
289
			(struct page *)radix_tree_deref_slot(pslot) != page) {
N
Nick Piggin 已提交
290
		spin_unlock_irq(&mapping->tree_lock);
291
		return -EAGAIN;
C
Christoph Lameter 已提交
292 293
	}

N
Nick Piggin 已提交
294
	if (!page_freeze_refs(page, expected_count)) {
N
Nick Piggin 已提交
295
		spin_unlock_irq(&mapping->tree_lock);
N
Nick Piggin 已提交
296 297 298
		return -EAGAIN;
	}

C
Christoph Lameter 已提交
299 300 301
	/*
	 * Now we know that no one else is looking at the page.
	 */
302
	get_page(newpage);	/* add cache reference */
303
#ifdef CONFIG_SWAP
C
Christoph Lameter 已提交
304 305 306 307
	if (PageSwapCache(page)) {
		SetPageSwapCache(newpage);
		set_page_private(newpage, page_private(page));
	}
308
#endif
C
Christoph Lameter 已提交
309

310 311
	radix_tree_replace_slot(pslot, newpage);

N
Nick Piggin 已提交
312
	page_unfreeze_refs(page, expected_count);
313 314 315 316
	/*
	 * Drop cache reference from old page.
	 * We know this isn't the last reference.
	 */
C
Christoph Lameter 已提交
317
	__put_page(page);
318

319 320 321 322 323 324 325 326 327 328 329 330 331
	/*
	 * If moved to a different zone then also account
	 * the page for that zone. Other VM counters will be
	 * taken care of when we establish references to the
	 * new page and drop references to the old page.
	 *
	 * Note that anonymous pages are accounted for
	 * via NR_FILE_PAGES and NR_ANON_PAGES if they
	 * are mapped to swap space.
	 */
	__dec_zone_page_state(page, NR_FILE_PAGES);
	__inc_zone_page_state(newpage, NR_FILE_PAGES);

N
Nick Piggin 已提交
332
	spin_unlock_irq(&mapping->tree_lock);
C
Christoph Lameter 已提交
333 334 335 336 337 338 339

	return 0;
}

/*
 * Copy the page to its new location
 */
340
static void migrate_page_copy(struct page *newpage, struct page *page)
C
Christoph Lameter 已提交
341
{
342 343
	int anon;

C
Christoph Lameter 已提交
344 345 346 347 348 349 350 351
	copy_highpage(newpage, page);

	if (PageError(page))
		SetPageError(newpage);
	if (PageReferenced(page))
		SetPageReferenced(newpage);
	if (PageUptodate(page))
		SetPageUptodate(newpage);
L
Lee Schermerhorn 已提交
352 353
	if (TestClearPageActive(page)) {
		VM_BUG_ON(PageUnevictable(page));
C
Christoph Lameter 已提交
354
		SetPageActive(newpage);
L
Lee Schermerhorn 已提交
355 356
	} else
		unevictable_migrate_page(newpage, page);
C
Christoph Lameter 已提交
357 358 359 360 361 362 363
	if (PageChecked(page))
		SetPageChecked(newpage);
	if (PageMappedToDisk(page))
		SetPageMappedToDisk(newpage);

	if (PageDirty(page)) {
		clear_page_dirty_for_io(page);
N
Nick Piggin 已提交
364 365 366 367 368 369 370 371
		/*
		 * Want to mark the page and the radix tree as dirty, and
		 * redo the accounting that clear_page_dirty_for_io undid,
		 * but we can't use set_page_dirty because that function
		 * is actually a signal that all of the page has become dirty.
		 * Wheras only part of our page may be dirty.
		 */
		__set_page_dirty_nobuffers(newpage);
C
Christoph Lameter 已提交
372 373
 	}

N
Nick Piggin 已提交
374 375
	mlock_migrate_page(newpage, page);

376
#ifdef CONFIG_SWAP
C
Christoph Lameter 已提交
377
	ClearPageSwapCache(page);
378
#endif
C
Christoph Lameter 已提交
379 380
	ClearPagePrivate(page);
	set_page_private(page, 0);
381 382
	/* page->mapping contains a flag for PageAnon() */
	anon = PageAnon(page);
C
Christoph Lameter 已提交
383 384
	page->mapping = NULL;

385 386 387
	if (!anon) /* This page was removed from radix-tree. */
		mem_cgroup_uncharge_cache_page(page);

C
Christoph Lameter 已提交
388 389 390 391 392 393 394 395
	/*
	 * If any waiters have accumulated on the new page then
	 * wake them up.
	 */
	if (PageWriteback(newpage))
		end_page_writeback(newpage);
}

396 397 398 399 400
/************************************************************
 *                    Migration functions
 ***********************************************************/

/* Always fail migration. Used for mappings that are not movable */
401 402
int fail_migrate_page(struct address_space *mapping,
			struct page *newpage, struct page *page)
403 404 405 406 407
{
	return -EIO;
}
EXPORT_SYMBOL(fail_migrate_page);

C
Christoph Lameter 已提交
408 409 410 411 412 413
/*
 * Common logic to directly migrate a single page suitable for
 * pages that do not use PagePrivate.
 *
 * Pages are locked upon entry and exit.
 */
414 415
int migrate_page(struct address_space *mapping,
		struct page *newpage, struct page *page)
C
Christoph Lameter 已提交
416 417 418 419 420
{
	int rc;

	BUG_ON(PageWriteback(page));	/* Writeback must be complete */

421
	rc = migrate_page_move_mapping(mapping, newpage, page);
C
Christoph Lameter 已提交
422 423 424 425 426 427 428 429 430

	if (rc)
		return rc;

	migrate_page_copy(newpage, page);
	return 0;
}
EXPORT_SYMBOL(migrate_page);

431
#ifdef CONFIG_BLOCK
432 433 434 435 436
/*
 * Migration function for pages with buffers. This function can only be used
 * if the underlying filesystem guarantees that no other references to "page"
 * exist.
 */
437 438
int buffer_migrate_page(struct address_space *mapping,
		struct page *newpage, struct page *page)
439 440 441 442 443
{
	struct buffer_head *bh, *head;
	int rc;

	if (!page_has_buffers(page))
444
		return migrate_page(mapping, newpage, page);
445 446 447

	head = page_buffers(page);

448
	rc = migrate_page_move_mapping(mapping, newpage, page);
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488

	if (rc)
		return rc;

	bh = head;
	do {
		get_bh(bh);
		lock_buffer(bh);
		bh = bh->b_this_page;

	} while (bh != head);

	ClearPagePrivate(page);
	set_page_private(newpage, page_private(page));
	set_page_private(page, 0);
	put_page(page);
	get_page(newpage);

	bh = head;
	do {
		set_bh_page(bh, newpage, bh_offset(bh));
		bh = bh->b_this_page;

	} while (bh != head);

	SetPagePrivate(newpage);

	migrate_page_copy(newpage, page);

	bh = head;
	do {
		unlock_buffer(bh);
 		put_bh(bh);
		bh = bh->b_this_page;

	} while (bh != head);

	return 0;
}
EXPORT_SYMBOL(buffer_migrate_page);
489
#endif
490

491 492 493 494
/*
 * Writeback a page to clean the dirty state
 */
static int writeout(struct address_space *mapping, struct page *page)
495
{
496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513
	struct writeback_control wbc = {
		.sync_mode = WB_SYNC_NONE,
		.nr_to_write = 1,
		.range_start = 0,
		.range_end = LLONG_MAX,
		.nonblocking = 1,
		.for_reclaim = 1
	};
	int rc;

	if (!mapping->a_ops->writepage)
		/* No write method for the address space */
		return -EINVAL;

	if (!clear_page_dirty_for_io(page))
		/* Someone else already triggered a write */
		return -EAGAIN;

514
	/*
515 516 517 518 519 520
	 * A dirty page may imply that the underlying filesystem has
	 * the page on some queue. So the page must be clean for
	 * migration. Writeout may mean we loose the lock and the
	 * page state is no longer what we checked for earlier.
	 * At this point we know that the migration attempt cannot
	 * be successful.
521
	 */
522
	remove_migration_ptes(page, page);
523

524 525 526 527
	rc = mapping->a_ops->writepage(page, &wbc);
	if (rc < 0)
		/* I/O Error writing */
		return -EIO;
528

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543
	if (rc != AOP_WRITEPAGE_ACTIVATE)
		/* unlocked. Relock */
		lock_page(page);

	return -EAGAIN;
}

/*
 * Default handling if a filesystem does not provide a migration function.
 */
static int fallback_migrate_page(struct address_space *mapping,
	struct page *newpage, struct page *page)
{
	if (PageDirty(page))
		return writeout(mapping, page);
544 545 546 547 548

	/*
	 * Buffers may be managed in a filesystem specific way.
	 * We must have no buffers or drop them.
	 */
549
	if (PagePrivate(page) &&
550 551 552 553 554 555
	    !try_to_release_page(page, GFP_KERNEL))
		return -EAGAIN;

	return migrate_page(mapping, newpage, page);
}

556 557 558 559 560 561
/*
 * Move a page to a newly allocated page
 * The page is locked and all ptes have been successfully removed.
 *
 * The new page will have replaced the old page if this function
 * is successful.
L
Lee Schermerhorn 已提交
562 563 564 565
 *
 * Return value:
 *   < 0 - error code
 *  == 0 - success
566 567 568 569 570 571 572 573 574 575 576
 */
static int move_to_new_page(struct page *newpage, struct page *page)
{
	struct address_space *mapping;
	int rc;

	/*
	 * Block others from accessing the page when we get around to
	 * establishing additional references. We are the only one
	 * holding a reference to the new page at this point.
	 */
N
Nick Piggin 已提交
577
	if (!trylock_page(newpage))
578 579 580 581 582
		BUG();

	/* Prepare mapping for the new page.*/
	newpage->index = page->index;
	newpage->mapping = page->mapping;
R
Rik van Riel 已提交
583 584
	if (PageSwapBacked(page))
		SetPageSwapBacked(newpage);
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601

	mapping = page_mapping(page);
	if (!mapping)
		rc = migrate_page(mapping, newpage, page);
	else if (mapping->a_ops->migratepage)
		/*
		 * Most pages have a mapping and most filesystems
		 * should provide a migration function. Anonymous
		 * pages are part of swap space which also has its
		 * own migration function. This is the most common
		 * path for page migration.
		 */
		rc = mapping->a_ops->migratepage(mapping,
						newpage, page);
	else
		rc = fallback_migrate_page(mapping, newpage, page);

602
	if (!rc) {
603
		remove_migration_ptes(page, newpage);
604
	} else
605 606 607 608 609 610 611 612 613 614 615
		newpage->mapping = NULL;

	unlock_page(newpage);

	return rc;
}

/*
 * Obtain the lock on page, remove all ptes and migrate the page
 * to the newly allocated page in newpage.
 */
616 617
static int unmap_and_move(new_page_t get_new_page, unsigned long private,
			struct page *page, int force)
618 619
{
	int rc = 0;
620 621
	int *result = NULL;
	struct page *newpage = get_new_page(page, private, &result);
622
	int rcu_locked = 0;
623
	int charge = 0;
624 625 626

	if (!newpage)
		return -ENOMEM;
627

L
Lee Schermerhorn 已提交
628
	if (page_count(page) == 1) {
629
		/* page was freed from under us. So we are done. */
630
		goto move_newpage;
L
Lee Schermerhorn 已提交
631
	}
632

633 634 635 636 637 638 639 640
	charge = mem_cgroup_prepare_migration(page, newpage);
	if (charge == -ENOMEM) {
		rc = -ENOMEM;
		goto move_newpage;
	}
	/* prepare cgroup just returns 0 or -ENOMEM */
	BUG_ON(charge);

641
	rc = -EAGAIN;
N
Nick Piggin 已提交
642
	if (!trylock_page(page)) {
643
		if (!force)
644
			goto move_newpage;
645 646 647 648 649 650 651 652 653
		lock_page(page);
	}

	if (PageWriteback(page)) {
		if (!force)
			goto unlock;
		wait_on_page_writeback(page);
	}
	/*
654 655 656 657
	 * By try_to_unmap(), page->mapcount goes down to 0 here. In this case,
	 * we cannot notice that anon_vma is freed while we migrates a page.
	 * This rcu_read_lock() delays freeing anon_vma pointer until the end
	 * of migration. File cache pages are no problem because of page_lock()
658 659
	 * File Caches may use write_page() or lock_page() in migration, then,
	 * just care Anon page here.
660
	 */
661 662 663 664
	if (PageAnon(page)) {
		rcu_read_lock();
		rcu_locked = 1;
	}
665

666
	/*
667 668 669 670 671 672 673 674 675 676
	 * Corner case handling:
	 * 1. When a new swap-cache page is read into, it is added to the LRU
	 * and treated as swapcache but it has no rmap yet.
	 * Calling try_to_unmap() against a page->mapping==NULL page will
	 * trigger a BUG.  So handle it here.
	 * 2. An orphaned page (see truncate_complete_page) might have
	 * fs-private metadata. The page can be picked up due to memory
	 * offlining.  Everywhere else except page reclaim, the page is
	 * invisible to the vm, so the page can not be migrated.  So try to
	 * free the metadata, so the page can be freed.
677
	 */
678 679 680 681 682 683 684 685 686 687 688
	if (!page->mapping) {
		if (!PageAnon(page) && PagePrivate(page)) {
			/*
			 * Go direct to try_to_free_buffers() here because
			 * a) that's what try_to_release_page() would do anyway
			 * b) we may be under rcu_read_lock() here, so we can't
			 *    use GFP_KERNEL which is what try_to_release_page()
			 *    needs to be effective.
			 */
			try_to_free_buffers(page);
		}
689
		goto rcu_unlock;
690 691
	}

692
	/* Establish migration ptes or remove ptes */
693
	try_to_unmap(page, 1);
694

695 696
	if (!page_mapped(page))
		rc = move_to_new_page(newpage, page);
697

698
	if (rc)
699
		remove_migration_ptes(page, page);
700
rcu_unlock:
701 702
	if (rcu_locked)
		rcu_read_unlock();
703

704 705
unlock:
	unlock_page(page);
706

707
	if (rc != -EAGAIN) {
708 709 710 711 712 713 714
 		/*
 		 * A page that has been migrated has all references
 		 * removed and will be freed. A page that has not been
 		 * migrated will have kepts its references and be
 		 * restored.
 		 */
 		list_del(&page->lru);
L
Lee Schermerhorn 已提交
715
		putback_lru_page(page);
716
	}
717 718

move_newpage:
719 720
	if (!charge)
		mem_cgroup_end_migration(newpage);
L
Lee Schermerhorn 已提交
721

722 723 724 725
	/*
	 * Move the new page to the LRU. If migration was not successful
	 * then this will free the page.
	 */
L
Lee Schermerhorn 已提交
726 727
	putback_lru_page(newpage);

728 729 730 731 732 733
	if (result) {
		if (rc)
			*result = rc;
		else
			*result = page_to_nid(newpage);
	}
734 735 736
	return rc;
}

C
Christoph Lameter 已提交
737 738 739
/*
 * migrate_pages
 *
740 741 742
 * The function takes one list of pages to migrate and a function
 * that determines from the page to be migrated and the private data
 * the target of the move and allocates the page.
C
Christoph Lameter 已提交
743 744 745
 *
 * The function returns after 10 attempts or if no pages
 * are movable anymore because to has become empty
746
 * or no retryable pages exist anymore. All pages will be
G
Gabriel Craciunescu 已提交
747
 * returned to the LRU or freed.
C
Christoph Lameter 已提交
748
 *
749
 * Return: Number of pages not migrated or error code.
C
Christoph Lameter 已提交
750
 */
751 752
int migrate_pages(struct list_head *from,
		new_page_t get_new_page, unsigned long private)
C
Christoph Lameter 已提交
753
{
754
	int retry = 1;
C
Christoph Lameter 已提交
755 756 757 758 759 760 761 762 763 764
	int nr_failed = 0;
	int pass = 0;
	struct page *page;
	struct page *page2;
	int swapwrite = current->flags & PF_SWAPWRITE;
	int rc;

	if (!swapwrite)
		current->flags |= PF_SWAPWRITE;

765 766
	for(pass = 0; pass < 10 && retry; pass++) {
		retry = 0;
C
Christoph Lameter 已提交
767

768 769
		list_for_each_entry_safe(page, page2, from, lru) {
			cond_resched();
770

771 772
			rc = unmap_and_move(get_new_page, private,
						page, pass > 2);
773

774
			switch(rc) {
775 776
			case -ENOMEM:
				goto out;
777
			case -EAGAIN:
778
				retry++;
779 780 781 782
				break;
			case 0:
				break;
			default:
783 784
				/* Permanent failure */
				nr_failed++;
785
				break;
786
			}
C
Christoph Lameter 已提交
787 788
		}
	}
789 790
	rc = 0;
out:
C
Christoph Lameter 已提交
791 792 793
	if (!swapwrite)
		current->flags &= ~PF_SWAPWRITE;

794
	putback_lru_pages(from);
C
Christoph Lameter 已提交
795

796 797
	if (rc)
		return rc;
C
Christoph Lameter 已提交
798

799
	return nr_failed + retry;
C
Christoph Lameter 已提交
800
}
801

802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825
#ifdef CONFIG_NUMA
/*
 * Move a list of individual pages
 */
struct page_to_node {
	unsigned long addr;
	struct page *page;
	int node;
	int status;
};

static struct page *new_page_node(struct page *p, unsigned long private,
		int **result)
{
	struct page_to_node *pm = (struct page_to_node *)private;

	while (pm->node != MAX_NUMNODES && pm->page != p)
		pm++;

	if (pm->node == MAX_NUMNODES)
		return NULL;

	*result = &pm->status;

826 827
	return alloc_pages_node(pm->node,
				GFP_HIGHUSER_MOVABLE | GFP_THISNODE, 0);
828 829 830 831 832 833
}

/*
 * Move a set of pages as indicated in the pm array. The addr
 * field must be set to the virtual address of the page to be moved
 * and the node number must contain a valid target node.
834
 * The pm array ends with node = MAX_NUMNODES.
835
 */
836 837 838
static int do_move_page_to_node_array(struct mm_struct *mm,
				      struct page_to_node *pm,
				      int migrate_all)
839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861
{
	int err;
	struct page_to_node *pp;
	LIST_HEAD(pagelist);

	down_read(&mm->mmap_sem);

	/*
	 * Build a list of pages to migrate
	 */
	migrate_prep();
	for (pp = pm; pp->node != MAX_NUMNODES; pp++) {
		struct vm_area_struct *vma;
		struct page *page;

		/*
		 * A valid page pointer that will not match any of the
		 * pages that will be moved.
		 */
		pp->page = ZERO_PAGE(0);

		err = -EFAULT;
		vma = find_vma(mm, pp->addr);
862
		if (!vma || !vma_migratable(vma))
863 864 865
			goto set_status;

		page = follow_page(vma, pp->addr, FOLL_GET);
866 867 868 869 870

		err = PTR_ERR(page);
		if (IS_ERR(page))
			goto set_status;

871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891
		err = -ENOENT;
		if (!page)
			goto set_status;

		if (PageReserved(page))		/* Check for zero page */
			goto put_and_set;

		pp->page = page;
		err = page_to_nid(page);

		if (err == pp->node)
			/*
			 * Node already in the right place
			 */
			goto put_and_set;

		err = -EACCES;
		if (page_mapcount(page) > 1 &&
				!migrate_all)
			goto put_and_set;

892 893 894
		err = isolate_lru_page(page);
		if (!err)
			list_add_tail(&page->lru, &pagelist);
895 896 897 898 899 900 901 902 903 904 905
put_and_set:
		/*
		 * Either remove the duplicate refcount from
		 * isolate_lru_page() or drop the page ref if it was
		 * not isolated.
		 */
		put_page(page);
set_status:
		pp->status = err;
	}

906
	err = 0;
907 908 909 910 911 912 913 914
	if (!list_empty(&pagelist))
		err = migrate_pages(&pagelist, new_page_node,
				(unsigned long)pm);

	up_read(&mm->mmap_sem);
	return err;
}

915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
/*
 * Migrate an array of page address onto an array of nodes and fill
 * the corresponding array of status.
 */
static int do_pages_move(struct mm_struct *mm, struct task_struct *task,
			 unsigned long nr_pages,
			 const void __user * __user *pages,
			 const int __user *nodes,
			 int __user *status, int flags)
{
	struct page_to_node *pm = NULL;
	nodemask_t task_nodes;
	int err = 0;
	int i;

	task_nodes = cpuset_mems_allowed(task);

	/* Limit nr_pages so that the multiplication may not overflow */
	if (nr_pages >= ULONG_MAX / sizeof(struct page_to_node) - 1) {
		err = -E2BIG;
		goto out;
	}

	pm = vmalloc((nr_pages + 1) * sizeof(struct page_to_node));
	if (!pm) {
		err = -ENOMEM;
		goto out;
	}

	/*
	 * Get parameters from user space and initialize the pm
	 * array. Return various errors if the user did something wrong.
	 */
	for (i = 0; i < nr_pages; i++) {
		const void __user *p;

		err = -EFAULT;
		if (get_user(p, pages + i))
			goto out_pm;

		pm[i].addr = (unsigned long)p;
		if (nodes) {
			int node;

			if (get_user(node, nodes + i))
				goto out_pm;

			err = -ENODEV;
			if (!node_state(node, N_HIGH_MEMORY))
				goto out_pm;

			err = -EACCES;
			if (!node_isset(node, task_nodes))
				goto out_pm;

			pm[i].node = node;
		} else
			pm[i].node = 0;	/* anything to not match MAX_NUMNODES */
	}
	/* End marker */
	pm[nr_pages].node = MAX_NUMNODES;

	err = do_move_page_to_node_array(mm, pm, flags & MPOL_MF_MOVE_ALL);
	if (err >= 0)
		/* Return status information */
		for (i = 0; i < nr_pages; i++)
			if (put_user(pm[i].status, status + i))
				err = -EFAULT;

out_pm:
	vfree(pm);
out:
	return err;
}

990
/*
991
 * Determine the nodes of an array of pages and store it in an array of status.
992
 */
993 994 995
static int do_pages_stat(struct mm_struct *mm, unsigned long nr_pages,
			 const void __user * __user *pages,
			 int __user *status)
996
{
997 998 999
	unsigned long i;
	int err;

1000 1001
	down_read(&mm->mmap_sem);

1002 1003 1004
	for (i = 0; i < nr_pages; i++) {
		const void __user *p;
		unsigned long addr;
1005 1006 1007 1008
		struct vm_area_struct *vma;
		struct page *page;

		err = -EFAULT;
1009 1010 1011 1012 1013
		if (get_user(p, pages+i))
			goto out;
		addr = (unsigned long) p;

		vma = find_vma(mm, addr);
1014 1015 1016
		if (!vma)
			goto set_status;

1017
		page = follow_page(vma, addr, 0);
1018 1019 1020 1021 1022

		err = PTR_ERR(page);
		if (IS_ERR(page))
			goto set_status;

1023 1024 1025 1026 1027 1028 1029
		err = -ENOENT;
		/* Use PageReserved to check for zero page */
		if (!page || PageReserved(page))
			goto set_status;

		err = page_to_nid(page);
set_status:
1030
		put_user(err, status+i);
1031
	}
1032
	err = 0;
1033

1034
out:
1035
	up_read(&mm->mmap_sem);
1036
	return err;
1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047
}

/*
 * Move a list of pages in the address space of the currently executing
 * process.
 */
asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages,
			const void __user * __user *pages,
			const int __user *nodes,
			int __user *status, int flags)
{
1048
	struct cred *cred, *tcred;
1049 1050
	struct task_struct *task;
	struct mm_struct *mm;
1051
	int err;
1052 1053 1054 1055 1056 1057 1058 1059 1060 1061

	/* Check flags */
	if (flags & ~(MPOL_MF_MOVE|MPOL_MF_MOVE_ALL))
		return -EINVAL;

	if ((flags & MPOL_MF_MOVE_ALL) && !capable(CAP_SYS_NICE))
		return -EPERM;

	/* Find the mm_struct */
	read_lock(&tasklist_lock);
1062
	task = pid ? find_task_by_vpid(pid) : current;
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078
	if (!task) {
		read_unlock(&tasklist_lock);
		return -ESRCH;
	}
	mm = get_task_mm(task);
	read_unlock(&tasklist_lock);

	if (!mm)
		return -EINVAL;

	/*
	 * Check if this process has the right to modify the specified
	 * process. The right exists if the process has administrative
	 * capabilities, superuser privileges or the same
	 * userid as the target process.
	 */
1079 1080 1081 1082
	cred = current->cred;
	tcred = task->cred;
	if (cred->euid != tcred->suid && cred->euid != tcred->uid &&
	    cred->uid  != tcred->suid && cred->uid  != tcred->uid &&
1083 1084
	    !capable(CAP_SYS_NICE)) {
		err = -EPERM;
1085
		goto out;
1086 1087
	}

1088 1089
 	err = security_task_movememory(task);
 	if (err)
1090
		goto out;
1091

1092 1093 1094 1095
	if (nodes) {
		err = do_pages_move(mm, task, nr_pages, pages, nodes, status,
				    flags);
	} else {
1096
		err = do_pages_stat(mm, nr_pages, pages, status);
1097 1098 1099 1100 1101 1102 1103
	}

out:
	mmput(mm);
	return err;
}

1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123
/*
 * Call migration functions in the vma_ops that may prepare
 * memory in a vm for migration. migration functions may perform
 * the migration for vmas that do not have an underlying page struct.
 */
int migrate_vmas(struct mm_struct *mm, const nodemask_t *to,
	const nodemask_t *from, unsigned long flags)
{
 	struct vm_area_struct *vma;
 	int err = 0;

 	for(vma = mm->mmap; vma->vm_next && !err; vma = vma->vm_next) {
 		if (vma->vm_ops && vma->vm_ops->migrate) {
 			err = vma->vm_ops->migrate(vma, to, from, flags);
 			if (err)
 				break;
 		}
 	}
 	return err;
}
1124
#endif