Blame view
mm/migrate.c
26.3 KB
b20a35035
|
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> |
cde535359
|
12 |
* Christoph Lameter |
b20a35035
|
13 14 15 16 17 |
*/ #include <linux/migrate.h> #include <linux/module.h> #include <linux/swap.h> |
0697212a4
|
18 |
#include <linux/swapops.h> |
b20a35035
|
19 |
#include <linux/pagemap.h> |
e23ca00bf
|
20 |
#include <linux/buffer_head.h> |
b20a35035
|
21 |
#include <linux/mm_inline.h> |
b488893a3
|
22 |
#include <linux/nsproxy.h> |
b20a35035
|
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> |
04e62a29b
|
28 |
#include <linux/writeback.h> |
742755a1d
|
29 30 |
#include <linux/mempolicy.h> #include <linux/vmalloc.h> |
86c3a7645
|
31 |
#include <linux/security.h> |
8a9f3ccd2
|
32 |
#include <linux/memcontrol.h> |
4f5ca2657
|
33 |
#include <linux/syscalls.h> |
b20a35035
|
34 35 |
#include "internal.h" |
b20a35035
|
36 37 38 |
#define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru)) /* |
742755a1d
|
39 40 |
* migrate_prep() needs to be called before we start compiling a list of pages * to be migrated using isolate_lru_page(). |
b20a35035
|
41 42 43 |
*/ int migrate_prep(void) { |
b20a35035
|
44 45 46 47 48 49 50 51 52 53 |
/* * 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; } |
b20a35035
|
54 |
/* |
894bc3104
|
55 56 |
* Add isolated pages on the list back to the LRU under page lock * to avoid leaking evictable pages back onto unevictable list. |
b20a35035
|
57 58 59 60 61 62 63 64 65 66 |
* * 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) { |
e24f0b8f7
|
67 |
list_del(&page->lru); |
a731286de
|
68 |
dec_zone_page_state(page, NR_ISOLATED_ANON + |
6c0b13519
|
69 |
page_is_file_cache(page)); |
894bc3104
|
70 |
putback_lru_page(page); |
b20a35035
|
71 72 73 74 |
count++; } return count; } |
0697212a4
|
75 76 77 |
/* * Restore a potential migration pte to a working pte entry */ |
04e62a29b
|
78 |
static void remove_migration_pte(struct vm_area_struct *vma, |
0697212a4
|
79 80 81 82 83 84 85 86 87 |
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; |
04e62a29b
|
88 89 90 91 |
unsigned long addr = page_address_in_vma(new, vma); if (addr == -EFAULT) return; |
0697212a4
|
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 |
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; |
0697212a4
|
122 123 124 125 |
get_page(new); pte = pte_mkold(mk_pte(new, vma->vm_page_prot)); if (is_write_migration_entry(entry)) pte = pte_mkwrite(pte); |
97ee05246
|
126 |
flush_cache_page(vma, addr, pte_pfn(pte)); |
0697212a4
|
127 |
set_pte_at(mm, addr, ptep, pte); |
04e62a29b
|
128 129 130 131 132 133 134 135 |
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); |
04e62a29b
|
136 |
|
0697212a4
|
137 138 139 140 141 |
out: pte_unmap_unlock(ptep, ptl); } /* |
04e62a29b
|
142 143 144 145 146 147 |
* 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; |
abfc34881
|
148 |
struct address_space *mapping = new->mapping; |
04e62a29b
|
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 |
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); } /* |
0697212a4
|
164 165 166 |
* Must hold mmap_sem lock on at least one of the vmas containing * the page so that the anon_vma cannot vanish. */ |
04e62a29b
|
167 |
static void remove_anon_migration_ptes(struct page *old, struct page *new) |
0697212a4
|
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 |
{ 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) |
04e62a29b
|
185 |
remove_migration_pte(vma, old, new); |
0697212a4
|
186 187 188 189 190 |
spin_unlock(&anon_vma->lock); } /* |
04e62a29b
|
191 192 193 194 195 196 197 198 199 200 201 202 |
* 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); } /* |
0697212a4
|
203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 |
* 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); |
e286781d5
|
227 228 229 230 231 232 233 234 235 |
/* * 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; |
0697212a4
|
236 237 238 239 240 241 242 |
pte_unmap_unlock(ptep, ptl); wait_on_page_locked(page); put_page(page); return; out: pte_unmap_unlock(ptep, ptl); } |
b20a35035
|
243 |
/* |
c3fcf8a5d
|
244 |
* Replace the page in the mapping. |
5b5c7120e
|
245 246 247 248 |
* * The number of remaining references must be: * 1 for anonymous pages without a mapping * 2 for pages with a mapping |
266cf658e
|
249 |
* 3 for pages with a mapping and PagePrivate/PagePrivate2 set. |
b20a35035
|
250 |
*/ |
2d1db3b11
|
251 252 |
static int migrate_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page) |
b20a35035
|
253 |
{ |
e286781d5
|
254 |
int expected_count; |
7cf9c2c76
|
255 |
void **pslot; |
b20a35035
|
256 |
|
6c5240ae7
|
257 |
if (!mapping) { |
0e8c7d0fd
|
258 |
/* Anonymous page without mapping */ |
6c5240ae7
|
259 260 261 262 |
if (page_count(page) != 1) return -EAGAIN; return 0; } |
19fd62312
|
263 |
spin_lock_irq(&mapping->tree_lock); |
b20a35035
|
264 |
|
7cf9c2c76
|
265 266 |
pslot = radix_tree_lookup_slot(&mapping->page_tree, page_index(page)); |
b20a35035
|
267 |
|
edcf4748c
|
268 |
expected_count = 2 + page_has_private(page); |
e286781d5
|
269 |
if (page_count(page) != expected_count || |
7cf9c2c76
|
270 |
(struct page *)radix_tree_deref_slot(pslot) != page) { |
19fd62312
|
271 |
spin_unlock_irq(&mapping->tree_lock); |
e23ca00bf
|
272 |
return -EAGAIN; |
b20a35035
|
273 |
} |
e286781d5
|
274 |
if (!page_freeze_refs(page, expected_count)) { |
19fd62312
|
275 |
spin_unlock_irq(&mapping->tree_lock); |
e286781d5
|
276 277 |
return -EAGAIN; } |
b20a35035
|
278 279 |
/* * Now we know that no one else is looking at the page. |
b20a35035
|
280 |
*/ |
7cf9c2c76
|
281 |
get_page(newpage); /* add cache reference */ |
b20a35035
|
282 283 284 285 |
if (PageSwapCache(page)) { SetPageSwapCache(newpage); set_page_private(newpage, page_private(page)); } |
7cf9c2c76
|
286 |
radix_tree_replace_slot(pslot, newpage); |
e286781d5
|
287 |
page_unfreeze_refs(page, expected_count); |
7cf9c2c76
|
288 289 290 291 |
/* * Drop cache reference from old page. * We know this isn't the last reference. */ |
b20a35035
|
292 |
__put_page(page); |
7cf9c2c76
|
293 |
|
0e8c7d0fd
|
294 295 296 297 298 299 300 301 302 303 304 305 |
/* * 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); |
4b02108ac
|
306 307 308 309 |
if (PageSwapBacked(page)) { __dec_zone_page_state(page, NR_SHMEM); __inc_zone_page_state(newpage, NR_SHMEM); } |
19fd62312
|
310 |
spin_unlock_irq(&mapping->tree_lock); |
b20a35035
|
311 312 313 |
return 0; } |
b20a35035
|
314 315 316 317 |
/* * Copy the page to its new location */ |
e7340f733
|
318 |
static void migrate_page_copy(struct page *newpage, struct page *page) |
b20a35035
|
319 |
{ |
b7abea963
|
320 |
int anon; |
b20a35035
|
321 322 323 324 325 326 327 328 |
copy_highpage(newpage, page); if (PageError(page)) SetPageError(newpage); if (PageReferenced(page)) SetPageReferenced(newpage); if (PageUptodate(page)) SetPageUptodate(newpage); |
894bc3104
|
329 330 |
if (TestClearPageActive(page)) { VM_BUG_ON(PageUnevictable(page)); |
b20a35035
|
331 |
SetPageActive(newpage); |
894bc3104
|
332 333 |
} else unevictable_migrate_page(newpage, page); |
b20a35035
|
334 335 336 337 338 339 340 |
if (PageChecked(page)) SetPageChecked(newpage); if (PageMappedToDisk(page)) SetPageMappedToDisk(newpage); if (PageDirty(page)) { clear_page_dirty_for_io(page); |
3a902c5f6
|
341 342 343 344 345 346 347 348 |
/* * 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); |
b20a35035
|
349 |
} |
b291f0003
|
350 |
mlock_migrate_page(newpage, page); |
b20a35035
|
351 |
ClearPageSwapCache(page); |
b20a35035
|
352 353 |
ClearPagePrivate(page); set_page_private(page, 0); |
b7abea963
|
354 355 |
/* page->mapping contains a flag for PageAnon() */ anon = PageAnon(page); |
b20a35035
|
356 357 358 359 360 361 362 363 364 |
page->mapping = NULL; /* * If any waiters have accumulated on the new page then * wake them up. */ if (PageWriteback(newpage)) end_page_writeback(newpage); } |
b20a35035
|
365 |
|
1d8b85ccf
|
366 367 368 369 370 |
/************************************************************ * Migration functions ***********************************************************/ /* Always fail migration. Used for mappings that are not movable */ |
2d1db3b11
|
371 372 |
int fail_migrate_page(struct address_space *mapping, struct page *newpage, struct page *page) |
1d8b85ccf
|
373 374 375 376 |
{ return -EIO; } EXPORT_SYMBOL(fail_migrate_page); |
b20a35035
|
377 378 |
/* * Common logic to directly migrate a single page suitable for |
266cf658e
|
379 |
* pages that do not use PagePrivate/PagePrivate2. |
b20a35035
|
380 381 382 |
* * Pages are locked upon entry and exit. */ |
2d1db3b11
|
383 384 |
int migrate_page(struct address_space *mapping, struct page *newpage, struct page *page) |
b20a35035
|
385 386 387 388 |
{ int rc; BUG_ON(PageWriteback(page)); /* Writeback must be complete */ |
2d1db3b11
|
389 |
rc = migrate_page_move_mapping(mapping, newpage, page); |
b20a35035
|
390 391 392 393 394 |
if (rc) return rc; migrate_page_copy(newpage, page); |
b20a35035
|
395 396 397 |
return 0; } EXPORT_SYMBOL(migrate_page); |
9361401eb
|
398 |
#ifdef CONFIG_BLOCK |
b20a35035
|
399 |
/* |
1d8b85ccf
|
400 401 402 403 |
* Migration function for pages with buffers. This function can only be used * if the underlying filesystem guarantees that no other references to "page" * exist. */ |
2d1db3b11
|
404 405 |
int buffer_migrate_page(struct address_space *mapping, struct page *newpage, struct page *page) |
1d8b85ccf
|
406 |
{ |
1d8b85ccf
|
407 408 |
struct buffer_head *bh, *head; int rc; |
1d8b85ccf
|
409 |
if (!page_has_buffers(page)) |
2d1db3b11
|
410 |
return migrate_page(mapping, newpage, page); |
1d8b85ccf
|
411 412 |
head = page_buffers(page); |
2d1db3b11
|
413 |
rc = migrate_page_move_mapping(mapping, newpage, page); |
1d8b85ccf
|
414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 |
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); |
9361401eb
|
454 |
#endif |
1d8b85ccf
|
455 |
|
04e62a29b
|
456 457 458 459 |
/* * Writeback a page to clean the dirty state */ static int writeout(struct address_space *mapping, struct page *page) |
8351a6e47
|
460 |
{ |
04e62a29b
|
461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 |
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; |
8351a6e47
|
478 |
/* |
04e62a29b
|
479 480 481 482 483 484 |
* 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. |
8351a6e47
|
485 |
*/ |
04e62a29b
|
486 |
remove_migration_ptes(page, page); |
8351a6e47
|
487 |
|
04e62a29b
|
488 |
rc = mapping->a_ops->writepage(page, &wbc); |
8351a6e47
|
489 |
|
04e62a29b
|
490 491 492 |
if (rc != AOP_WRITEPAGE_ACTIVATE) /* unlocked. Relock */ lock_page(page); |
bda8550de
|
493 |
return (rc < 0) ? -EIO : -EAGAIN; |
04e62a29b
|
494 495 496 497 498 499 500 501 502 503 |
} /* * 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); |
8351a6e47
|
504 505 506 507 508 |
/* * Buffers may be managed in a filesystem specific way. * We must have no buffers or drop them. */ |
266cf658e
|
509 |
if (page_has_private(page) && |
8351a6e47
|
510 511 512 513 514 |
!try_to_release_page(page, GFP_KERNEL)) return -EAGAIN; return migrate_page(mapping, newpage, page); } |
1d8b85ccf
|
515 |
/* |
e24f0b8f7
|
516 517 518 519 520 |
* 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. |
894bc3104
|
521 522 523 524 |
* * Return value: * < 0 - error code * == 0 - success |
e24f0b8f7
|
525 526 527 528 529 530 531 532 533 534 535 |
*/ 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. */ |
529ae9aaa
|
536 |
if (!trylock_page(newpage)) |
e24f0b8f7
|
537 538 539 540 541 |
BUG(); /* Prepare mapping for the new page.*/ newpage->index = page->index; newpage->mapping = page->mapping; |
b2e185384
|
542 543 |
if (PageSwapBacked(page)) SetPageSwapBacked(newpage); |
e24f0b8f7
|
544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 |
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); |
ae41be374
|
560 |
if (!rc) { |
e24f0b8f7
|
561 |
remove_migration_ptes(page, newpage); |
ae41be374
|
562 |
} else |
e24f0b8f7
|
563 564 565 566 567 568 569 570 571 572 573 |
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. */ |
95a402c38
|
574 575 |
static int unmap_and_move(new_page_t get_new_page, unsigned long private, struct page *page, int force) |
e24f0b8f7
|
576 577 |
{ int rc = 0; |
742755a1d
|
578 579 |
int *result = NULL; struct page *newpage = get_new_page(page, private, &result); |
989f89c57
|
580 |
int rcu_locked = 0; |
ae41be374
|
581 |
int charge = 0; |
01b1ae63c
|
582 |
struct mem_cgroup *mem; |
95a402c38
|
583 584 585 |
if (!newpage) return -ENOMEM; |
e24f0b8f7
|
586 |
|
894bc3104
|
587 |
if (page_count(page) == 1) { |
e24f0b8f7
|
588 |
/* page was freed from under us. So we are done. */ |
95a402c38
|
589 |
goto move_newpage; |
894bc3104
|
590 |
} |
e24f0b8f7
|
591 |
|
e8589cc18
|
592 |
/* prepare cgroup just returns 0 or -ENOMEM */ |
e24f0b8f7
|
593 |
rc = -EAGAIN; |
01b1ae63c
|
594 |
|
529ae9aaa
|
595 |
if (!trylock_page(page)) { |
e24f0b8f7
|
596 |
if (!force) |
95a402c38
|
597 |
goto move_newpage; |
e24f0b8f7
|
598 599 |
lock_page(page); } |
01b1ae63c
|
600 601 602 603 604 605 606 |
/* charge against new page */ charge = mem_cgroup_prepare_migration(page, &mem); if (charge == -ENOMEM) { rc = -ENOMEM; goto unlock; } BUG_ON(charge); |
e24f0b8f7
|
607 608 |
if (PageWriteback(page)) { if (!force) |
01b1ae63c
|
609 |
goto uncharge; |
e24f0b8f7
|
610 611 |
wait_on_page_writeback(page); } |
e24f0b8f7
|
612 |
/* |
dc386d4d1
|
613 614 615 616 |
* 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() |
989f89c57
|
617 618 |
* File Caches may use write_page() or lock_page() in migration, then, * just care Anon page here. |
dc386d4d1
|
619 |
*/ |
989f89c57
|
620 621 622 623 |
if (PageAnon(page)) { rcu_read_lock(); rcu_locked = 1; } |
62e1c5530
|
624 |
|
dc386d4d1
|
625 |
/* |
62e1c5530
|
626 627 628 629 630 631 632 633 634 635 |
* 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. |
e24f0b8f7
|
636 |
*/ |
62e1c5530
|
637 |
if (!page->mapping) { |
266cf658e
|
638 |
if (!PageAnon(page) && page_has_private(page)) { |
62e1c5530
|
639 640 641 642 643 644 645 646 |
/* * 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); |
abfc34881
|
647 |
goto rcu_unlock; |
62e1c5530
|
648 |
} |
abfc34881
|
649 |
goto skip_unmap; |
62e1c5530
|
650 |
} |
dc386d4d1
|
651 |
/* Establish migration ptes or remove ptes */ |
14fa31b89
|
652 |
try_to_unmap(page, TTU_MIGRATION|TTU_IGNORE_MLOCK|TTU_IGNORE_ACCESS); |
dc386d4d1
|
653 |
|
abfc34881
|
654 |
skip_unmap: |
e6a1530d6
|
655 656 |
if (!page_mapped(page)) rc = move_to_new_page(newpage, page); |
e24f0b8f7
|
657 |
|
e8589cc18
|
658 |
if (rc) |
e24f0b8f7
|
659 |
remove_migration_ptes(page, page); |
dc386d4d1
|
660 |
rcu_unlock: |
989f89c57
|
661 662 |
if (rcu_locked) rcu_read_unlock(); |
01b1ae63c
|
663 664 665 |
uncharge: if (!charge) mem_cgroup_end_migration(mem, page, newpage); |
e24f0b8f7
|
666 667 |
unlock: unlock_page(page); |
95a402c38
|
668 |
|
e24f0b8f7
|
669 |
if (rc != -EAGAIN) { |
aaa994b30
|
670 671 672 673 674 675 676 |
/* * 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); |
a731286de
|
677 |
dec_zone_page_state(page, NR_ISOLATED_ANON + |
6c0b13519
|
678 |
page_is_file_cache(page)); |
894bc3104
|
679 |
putback_lru_page(page); |
e24f0b8f7
|
680 |
} |
95a402c38
|
681 682 |
move_newpage: |
894bc3104
|
683 |
|
95a402c38
|
684 685 686 687 |
/* * Move the new page to the LRU. If migration was not successful * then this will free the page. */ |
894bc3104
|
688 |
putback_lru_page(newpage); |
742755a1d
|
689 690 691 692 693 694 |
if (result) { if (rc) *result = rc; else *result = page_to_nid(newpage); } |
e24f0b8f7
|
695 696 697 698 |
return rc; } /* |
b20a35035
|
699 700 |
* migrate_pages * |
95a402c38
|
701 702 703 |
* 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. |
b20a35035
|
704 705 706 |
* * The function returns after 10 attempts or if no pages * are movable anymore because to has become empty |
aaa994b30
|
707 |
* or no retryable pages exist anymore. All pages will be |
e9534b3fd
|
708 |
* returned to the LRU or freed. |
b20a35035
|
709 |
* |
95a402c38
|
710 |
* Return: Number of pages not migrated or error code. |
b20a35035
|
711 |
*/ |
95a402c38
|
712 713 |
int migrate_pages(struct list_head *from, new_page_t get_new_page, unsigned long private) |
b20a35035
|
714 |
{ |
e24f0b8f7
|
715 |
int retry = 1; |
b20a35035
|
716 717 718 719 720 721 |
int nr_failed = 0; int pass = 0; struct page *page; struct page *page2; int swapwrite = current->flags & PF_SWAPWRITE; int rc; |
a731286de
|
722 723 724 725 726 |
unsigned long flags; local_irq_save(flags); list_for_each_entry(page, from, lru) __inc_zone_page_state(page, NR_ISOLATED_ANON + |
6c0b13519
|
727 |
page_is_file_cache(page)); |
a731286de
|
728 |
local_irq_restore(flags); |
b20a35035
|
729 730 731 |
if (!swapwrite) current->flags |= PF_SWAPWRITE; |
e24f0b8f7
|
732 733 |
for(pass = 0; pass < 10 && retry; pass++) { retry = 0; |
b20a35035
|
734 |
|
e24f0b8f7
|
735 |
list_for_each_entry_safe(page, page2, from, lru) { |
e24f0b8f7
|
736 |
cond_resched(); |
2d1db3b11
|
737 |
|
95a402c38
|
738 739 |
rc = unmap_and_move(get_new_page, private, page, pass > 2); |
2d1db3b11
|
740 |
|
e24f0b8f7
|
741 |
switch(rc) { |
95a402c38
|
742 743 |
case -ENOMEM: goto out; |
e24f0b8f7
|
744 |
case -EAGAIN: |
2d1db3b11
|
745 |
retry++; |
e24f0b8f7
|
746 747 |
break; case 0: |
e24f0b8f7
|
748 749 |
break; default: |
2d1db3b11
|
750 |
/* Permanent failure */ |
2d1db3b11
|
751 |
nr_failed++; |
e24f0b8f7
|
752 |
break; |
2d1db3b11
|
753 |
} |
b20a35035
|
754 755 |
} } |
95a402c38
|
756 757 |
rc = 0; out: |
b20a35035
|
758 759 |
if (!swapwrite) current->flags &= ~PF_SWAPWRITE; |
aaa994b30
|
760 |
putback_lru_pages(from); |
b20a35035
|
761 |
|
95a402c38
|
762 763 |
if (rc) return rc; |
b20a35035
|
764 |
|
95a402c38
|
765 |
return nr_failed + retry; |
b20a35035
|
766 |
} |
95a402c38
|
767 |
|
742755a1d
|
768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 |
#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; |
6484eb3e2
|
791 |
return alloc_pages_exact_node(pm->node, |
769848c03
|
792 |
GFP_HIGHUSER_MOVABLE | GFP_THISNODE, 0); |
742755a1d
|
793 794 795 796 797 798 |
} /* * 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. |
5e9a0f023
|
799 |
* The pm array ends with node = MAX_NUMNODES. |
742755a1d
|
800 |
*/ |
5e9a0f023
|
801 802 803 |
static int do_move_page_to_node_array(struct mm_struct *mm, struct page_to_node *pm, int migrate_all) |
742755a1d
|
804 805 806 807 808 809 810 811 812 813 |
{ int err; struct page_to_node *pp; LIST_HEAD(pagelist); down_read(&mm->mmap_sem); /* * Build a list of pages to migrate */ |
742755a1d
|
814 815 816 |
for (pp = pm; pp->node != MAX_NUMNODES; pp++) { struct vm_area_struct *vma; struct page *page; |
742755a1d
|
817 818 |
err = -EFAULT; vma = find_vma(mm, pp->addr); |
0dc952dc3
|
819 |
if (!vma || !vma_migratable(vma)) |
742755a1d
|
820 821 822 |
goto set_status; page = follow_page(vma, pp->addr, FOLL_GET); |
89f5b7da2
|
823 824 825 826 |
err = PTR_ERR(page); if (IS_ERR(page)) goto set_status; |
742755a1d
|
827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 |
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; |
62695a84e
|
847 848 849 |
err = isolate_lru_page(page); if (!err) list_add_tail(&page->lru, &pagelist); |
742755a1d
|
850 851 852 853 854 855 856 857 858 859 |
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; } |
e78bbfa82
|
860 |
err = 0; |
742755a1d
|
861 862 863 |
if (!list_empty(&pagelist)) err = migrate_pages(&pagelist, new_page_node, (unsigned long)pm); |
742755a1d
|
864 865 866 867 868 869 |
up_read(&mm->mmap_sem); return err; } /* |
5e9a0f023
|
870 871 872 873 874 875 876 877 878 |
* 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) { |
3140a2273
|
879 |
struct page_to_node *pm; |
5e9a0f023
|
880 |
nodemask_t task_nodes; |
3140a2273
|
881 882 883 |
unsigned long chunk_nr_pages; unsigned long chunk_start; int err; |
5e9a0f023
|
884 885 |
task_nodes = cpuset_mems_allowed(task); |
3140a2273
|
886 887 888 |
err = -ENOMEM; pm = (struct page_to_node *)__get_free_page(GFP_KERNEL); if (!pm) |
5e9a0f023
|
889 |
goto out; |
35282a2de
|
890 891 |
migrate_prep(); |
5e9a0f023
|
892 |
/* |
3140a2273
|
893 894 |
* Store a chunk of page_to_node array in a page, * but keep the last one as a marker |
5e9a0f023
|
895 |
*/ |
3140a2273
|
896 |
chunk_nr_pages = (PAGE_SIZE / sizeof(struct page_to_node)) - 1; |
5e9a0f023
|
897 |
|
3140a2273
|
898 899 900 901 |
for (chunk_start = 0; chunk_start < nr_pages; chunk_start += chunk_nr_pages) { int j; |
5e9a0f023
|
902 |
|
3140a2273
|
903 904 905 906 907 908 |
if (chunk_start + chunk_nr_pages > nr_pages) chunk_nr_pages = nr_pages - chunk_start; /* fill the chunk pm with addrs and nodes from user-space */ for (j = 0; j < chunk_nr_pages; j++) { const void __user *p; |
5e9a0f023
|
909 |
int node; |
3140a2273
|
910 911 912 913 914 915 |
err = -EFAULT; if (get_user(p, pages + j + chunk_start)) goto out_pm; pm[j].addr = (unsigned long) p; if (get_user(node, nodes + j + chunk_start)) |
5e9a0f023
|
916 917 918 919 920 921 922 923 924 |
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; |
3140a2273
|
925 926 927 928 929 930 931 932 933 934 935 |
pm[j].node = node; } /* End marker for this chunk */ pm[chunk_nr_pages].node = MAX_NUMNODES; /* Migrate this chunk */ err = do_move_page_to_node_array(mm, pm, flags & MPOL_MF_MOVE_ALL); if (err < 0) goto out_pm; |
5e9a0f023
|
936 |
|
5e9a0f023
|
937 |
/* Return status information */ |
3140a2273
|
938 939 |
for (j = 0; j < chunk_nr_pages; j++) if (put_user(pm[j].status, status + j + chunk_start)) { |
5e9a0f023
|
940 |
err = -EFAULT; |
3140a2273
|
941 942 943 944 |
goto out_pm; } } err = 0; |
5e9a0f023
|
945 946 |
out_pm: |
3140a2273
|
947 |
free_page((unsigned long)pm); |
5e9a0f023
|
948 949 950 951 952 |
out: return err; } /* |
2f007e74b
|
953 |
* Determine the nodes of an array of pages and store it in an array of status. |
742755a1d
|
954 |
*/ |
80bba1290
|
955 956 |
static void do_pages_stat_array(struct mm_struct *mm, unsigned long nr_pages, const void __user **pages, int *status) |
742755a1d
|
957 |
{ |
2f007e74b
|
958 |
unsigned long i; |
2f007e74b
|
959 |
|
742755a1d
|
960 |
down_read(&mm->mmap_sem); |
2f007e74b
|
961 |
for (i = 0; i < nr_pages; i++) { |
80bba1290
|
962 |
unsigned long addr = (unsigned long)(*pages); |
742755a1d
|
963 964 |
struct vm_area_struct *vma; struct page *page; |
c095adbc2
|
965 |
int err = -EFAULT; |
2f007e74b
|
966 967 |
vma = find_vma(mm, addr); |
742755a1d
|
968 969 |
if (!vma) goto set_status; |
2f007e74b
|
970 |
page = follow_page(vma, addr, 0); |
89f5b7da2
|
971 972 973 974 |
err = PTR_ERR(page); if (IS_ERR(page)) goto set_status; |
742755a1d
|
975 976 977 978 979 980 981 |
err = -ENOENT; /* Use PageReserved to check for zero page */ if (!page || PageReserved(page)) goto set_status; err = page_to_nid(page); set_status: |
80bba1290
|
982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 |
*status = err; pages++; status++; } up_read(&mm->mmap_sem); } /* * Determine the nodes of a user array of pages and store it in * a user array of status. */ static int do_pages_stat(struct mm_struct *mm, unsigned long nr_pages, const void __user * __user *pages, int __user *status) { #define DO_PAGES_STAT_CHUNK_NR 16 const void __user *chunk_pages[DO_PAGES_STAT_CHUNK_NR]; int chunk_status[DO_PAGES_STAT_CHUNK_NR]; unsigned long i, chunk_nr = DO_PAGES_STAT_CHUNK_NR; int err; for (i = 0; i < nr_pages; i += chunk_nr) { if (chunk_nr + i > nr_pages) chunk_nr = nr_pages - i; err = copy_from_user(chunk_pages, &pages[i], chunk_nr * sizeof(*chunk_pages)); if (err) { err = -EFAULT; goto out; } do_pages_stat_array(mm, chunk_nr, chunk_pages, chunk_status); err = copy_to_user(&status[i], chunk_status, chunk_nr * sizeof(*chunk_status)); if (err) { err = -EFAULT; goto out; } |
742755a1d
|
1024 |
} |
2f007e74b
|
1025 |
err = 0; |
742755a1d
|
1026 |
|
2f007e74b
|
1027 |
out: |
2f007e74b
|
1028 |
return err; |
742755a1d
|
1029 1030 1031 1032 1033 1034 |
} /* * Move a list of pages in the address space of the currently executing * process. */ |
938bb9f5e
|
1035 1036 1037 1038 |
SYSCALL_DEFINE6(move_pages, pid_t, pid, unsigned long, nr_pages, const void __user * __user *, pages, const int __user *, nodes, int __user *, status, int, flags) |
742755a1d
|
1039 |
{ |
c69e8d9c0
|
1040 |
const struct cred *cred = current_cred(), *tcred; |
742755a1d
|
1041 |
struct task_struct *task; |
742755a1d
|
1042 |
struct mm_struct *mm; |
5e9a0f023
|
1043 |
int err; |
742755a1d
|
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 |
/* 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); |
228ebcbe6
|
1054 |
task = pid ? find_task_by_vpid(pid) : current; |
742755a1d
|
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 |
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. */ |
c69e8d9c0
|
1071 1072 |
rcu_read_lock(); tcred = __task_cred(task); |
b6dff3ec5
|
1073 1074 |
if (cred->euid != tcred->suid && cred->euid != tcred->uid && cred->uid != tcred->suid && cred->uid != tcred->uid && |
742755a1d
|
1075 |
!capable(CAP_SYS_NICE)) { |
c69e8d9c0
|
1076 |
rcu_read_unlock(); |
742755a1d
|
1077 |
err = -EPERM; |
5e9a0f023
|
1078 |
goto out; |
742755a1d
|
1079 |
} |
c69e8d9c0
|
1080 |
rcu_read_unlock(); |
742755a1d
|
1081 |
|
86c3a7645
|
1082 1083 |
err = security_task_movememory(task); if (err) |
5e9a0f023
|
1084 |
goto out; |
86c3a7645
|
1085 |
|
5e9a0f023
|
1086 1087 1088 1089 |
if (nodes) { err = do_pages_move(mm, task, nr_pages, pages, nodes, status, flags); } else { |
2f007e74b
|
1090 |
err = do_pages_stat(mm, nr_pages, pages, status); |
742755a1d
|
1091 |
} |
742755a1d
|
1092 |
out: |
742755a1d
|
1093 1094 1095 |
mmput(mm); return err; } |
742755a1d
|
1096 |
|
7b2259b3e
|
1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 |
/* * 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; |
1001c9fb8
|
1107 |
for (vma = mm->mmap; vma && !err; vma = vma->vm_next) { |
7b2259b3e
|
1108 1109 1110 1111 1112 1113 1114 1115 |
if (vma->vm_ops && vma->vm_ops->migrate) { err = vma->vm_ops->migrate(vma, to, from, flags); if (err) break; } } return err; } |
83d1674a9
|
1116 |
#endif |