Blame view
mm/swap.c
26.1 KB
1da177e4c
|
1 2 3 4 5 6 7 |
/* * linux/mm/swap.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds */ /* |
183ff22bb
|
8 |
* This file contains the default values for the operation of the |
1da177e4c
|
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
* Linux VM subsystem. Fine-tuning documentation can be found in * Documentation/sysctl/vm.txt. * Started 18.12.91 * Swap aging added 23.2.95, Stephen Tweedie. * Buffermem limits added 12.3.98, Rik van Riel. */ #include <linux/mm.h> #include <linux/sched.h> #include <linux/kernel_stat.h> #include <linux/swap.h> #include <linux/mman.h> #include <linux/pagemap.h> #include <linux/pagevec.h> #include <linux/init.h> |
b95f1b31b
|
24 |
#include <linux/export.h> |
1da177e4c
|
25 |
#include <linux/mm_inline.h> |
1da177e4c
|
26 |
#include <linux/percpu_counter.h> |
3565fce3a
|
27 |
#include <linux/memremap.h> |
1da177e4c
|
28 29 30 |
#include <linux/percpu.h> #include <linux/cpu.h> #include <linux/notifier.h> |
e0bf68dde
|
31 |
#include <linux/backing-dev.h> |
66e1707bc
|
32 |
#include <linux/memcontrol.h> |
5a0e3ad6a
|
33 |
#include <linux/gfp.h> |
a27bb332c
|
34 |
#include <linux/uio.h> |
822fc6136
|
35 |
#include <linux/hugetlb.h> |
33c3fc71c
|
36 |
#include <linux/page_idle.h> |
1da177e4c
|
37 |
|
64d6519dd
|
38 |
#include "internal.h" |
c6286c983
|
39 40 |
#define CREATE_TRACE_POINTS #include <trace/events/pagemap.h> |
1da177e4c
|
41 42 |
/* How many pages do we try to swap or page in/out together? */ int page_cluster; |
13f7f7898
|
43 |
static DEFINE_PER_CPU(struct pagevec, lru_add_pvec); |
f84f9504b
|
44 |
static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs); |
cc5993bd7
|
45 |
static DEFINE_PER_CPU(struct pagevec, lru_deactivate_file_pvecs); |
10853a039
|
46 |
static DEFINE_PER_CPU(struct pagevec, lru_deactivate_pvecs); |
902aaed0d
|
47 |
|
b221385bc
|
48 49 50 51 |
/* * This path almost never happens for VM activity - pages are normally * freed via pagevecs. But it gets used by networking. */ |
920c7a5d0
|
52 |
static void __page_cache_release(struct page *page) |
b221385bc
|
53 54 |
{ if (PageLRU(page)) { |
b221385bc
|
55 |
struct zone *zone = page_zone(page); |
fa9add641
|
56 57 |
struct lruvec *lruvec; unsigned long flags; |
b221385bc
|
58 59 |
spin_lock_irqsave(&zone->lru_lock, flags); |
fa9add641
|
60 |
lruvec = mem_cgroup_page_lruvec(page, zone); |
309381fea
|
61 |
VM_BUG_ON_PAGE(!PageLRU(page), page); |
b221385bc
|
62 |
__ClearPageLRU(page); |
fa9add641
|
63 |
del_page_from_lru_list(page, lruvec, page_off_lru(page)); |
b221385bc
|
64 65 |
spin_unlock_irqrestore(&zone->lru_lock, flags); } |
0a31bc97c
|
66 |
mem_cgroup_uncharge(page); |
918070634
|
67 68 69 70 71 |
} static void __put_single_page(struct page *page) { __page_cache_release(page); |
b745bc85f
|
72 |
free_hot_cold_page(page, false); |
b221385bc
|
73 |
} |
918070634
|
74 |
static void __put_compound_page(struct page *page) |
1da177e4c
|
75 |
{ |
918070634
|
76 |
compound_page_dtor *dtor; |
1da177e4c
|
77 |
|
822fc6136
|
78 79 80 81 82 83 84 85 |
/* * __page_cache_release() is supposed to be called for thp, not for * hugetlb. This is because hugetlb page does never have PageLRU set * (it's never listed to any LRU lists) and no memcg routines should * be called for hugetlb (it has a separate hugetlb_cgroup.) */ if (!PageHuge(page)) __page_cache_release(page); |
918070634
|
86 87 88 |
dtor = get_compound_page_dtor(page); (*dtor)(page); } |
ddc58f27f
|
89 |
void __put_page(struct page *page) |
8519fb30e
|
90 91 |
{ if (unlikely(PageCompound(page))) |
ddc58f27f
|
92 93 |
__put_compound_page(page); else |
918070634
|
94 |
__put_single_page(page); |
1da177e4c
|
95 |
} |
ddc58f27f
|
96 |
EXPORT_SYMBOL(__put_page); |
70b50f94f
|
97 |
|
1d7ea7324
|
98 |
/** |
7682486b3
|
99 100 |
* put_pages_list() - release a list of pages * @pages: list of pages threaded on page->lru |
1d7ea7324
|
101 102 103 |
* * Release a list of pages which are strung together on page.lru. Currently * used by read_cache_pages() and related error recovery code. |
1d7ea7324
|
104 105 106 107 108 109 110 111 |
*/ void put_pages_list(struct list_head *pages) { while (!list_empty(pages)) { struct page *victim; victim = list_entry(pages->prev, struct page, lru); list_del(&victim->lru); |
09cbfeaf1
|
112 |
put_page(victim); |
1d7ea7324
|
113 114 115 |
} } EXPORT_SYMBOL(put_pages_list); |
18022c5d8
|
116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 |
/* * get_kernel_pages() - pin kernel pages in memory * @kiov: An array of struct kvec structures * @nr_segs: number of segments to pin * @write: pinning for read/write, currently ignored * @pages: array that receives pointers to the pages pinned. * Should be at least nr_segs long. * * Returns number of pages pinned. This may be fewer than the number * requested. If nr_pages is 0 or negative, returns 0. If no pages * were pinned, returns -errno. Each page returned must be released * with a put_page() call when it is finished with. */ int get_kernel_pages(const struct kvec *kiov, int nr_segs, int write, struct page **pages) { int seg; for (seg = 0; seg < nr_segs; seg++) { if (WARN_ON(kiov[seg].iov_len != PAGE_SIZE)) return seg; |
5a178119b
|
137 |
pages[seg] = kmap_to_page(kiov[seg].iov_base); |
09cbfeaf1
|
138 |
get_page(pages[seg]); |
18022c5d8
|
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 |
} return seg; } EXPORT_SYMBOL_GPL(get_kernel_pages); /* * get_kernel_page() - pin a kernel page in memory * @start: starting kernel address * @write: pinning for read/write, currently ignored * @pages: array that receives pointer to the page pinned. * Must be at least nr_segs long. * * Returns 1 if page is pinned. If the page was not pinned, returns * -errno. The page returned must be released with a put_page() call * when it is finished with. */ int get_kernel_page(unsigned long start, int write, struct page **pages) { const struct kvec kiov = { .iov_base = (void *)start, .iov_len = PAGE_SIZE }; return get_kernel_pages(&kiov, 1, write, pages); } EXPORT_SYMBOL_GPL(get_kernel_page); |
3dd7ae8ec
|
166 |
static void pagevec_lru_move_fn(struct pagevec *pvec, |
fa9add641
|
167 168 |
void (*move_fn)(struct page *page, struct lruvec *lruvec, void *arg), void *arg) |
902aaed0d
|
169 170 |
{ int i; |
902aaed0d
|
171 |
struct zone *zone = NULL; |
fa9add641
|
172 |
struct lruvec *lruvec; |
3dd7ae8ec
|
173 |
unsigned long flags = 0; |
902aaed0d
|
174 175 176 177 178 179 180 |
for (i = 0; i < pagevec_count(pvec); i++) { struct page *page = pvec->pages[i]; struct zone *pagezone = page_zone(page); if (pagezone != zone) { if (zone) |
3dd7ae8ec
|
181 |
spin_unlock_irqrestore(&zone->lru_lock, flags); |
902aaed0d
|
182 |
zone = pagezone; |
3dd7ae8ec
|
183 |
spin_lock_irqsave(&zone->lru_lock, flags); |
902aaed0d
|
184 |
} |
3dd7ae8ec
|
185 |
|
fa9add641
|
186 187 |
lruvec = mem_cgroup_page_lruvec(page, zone); (*move_fn)(page, lruvec, arg); |
902aaed0d
|
188 189 |
} if (zone) |
3dd7ae8ec
|
190 |
spin_unlock_irqrestore(&zone->lru_lock, flags); |
83896fb5e
|
191 192 |
release_pages(pvec->pages, pvec->nr, pvec->cold); pagevec_reinit(pvec); |
d8505dee1
|
193 |
} |
fa9add641
|
194 195 |
static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruvec, void *arg) |
3dd7ae8ec
|
196 197 |
{ int *pgmoved = arg; |
3dd7ae8ec
|
198 199 200 |
if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) { enum lru_list lru = page_lru_base_type(page); |
925b7673c
|
201 |
list_move_tail(&page->lru, &lruvec->lists[lru]); |
3dd7ae8ec
|
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 |
(*pgmoved)++; } } /* * pagevec_move_tail() must be called with IRQ disabled. * Otherwise this may cause nasty races. */ static void pagevec_move_tail(struct pagevec *pvec) { int pgmoved = 0; pagevec_lru_move_fn(pvec, pagevec_move_tail_fn, &pgmoved); __count_vm_events(PGROTATED, pgmoved); } |
902aaed0d
|
217 |
/* |
1da177e4c
|
218 219 |
* Writeback is about to end against a page which has been marked for immediate * reclaim. If it still appears to be reclaimable, move it to the tail of the |
902aaed0d
|
220 |
* inactive list. |
1da177e4c
|
221 |
*/ |
3dd7ae8ec
|
222 |
void rotate_reclaimable_page(struct page *page) |
1da177e4c
|
223 |
{ |
ac6aadb24
|
224 |
if (!PageLocked(page) && !PageDirty(page) && !PageActive(page) && |
894bc3104
|
225 |
!PageUnevictable(page) && PageLRU(page)) { |
ac6aadb24
|
226 227 |
struct pagevec *pvec; unsigned long flags; |
09cbfeaf1
|
228 |
get_page(page); |
ac6aadb24
|
229 |
local_irq_save(flags); |
7c8e0181e
|
230 |
pvec = this_cpu_ptr(&lru_rotate_pvecs); |
ac6aadb24
|
231 232 233 234 |
if (!pagevec_add(pvec, page)) pagevec_move_tail(pvec); local_irq_restore(flags); } |
1da177e4c
|
235 |
} |
fa9add641
|
236 |
static void update_page_reclaim_stat(struct lruvec *lruvec, |
3e2f41f1f
|
237 238 |
int file, int rotated) { |
fa9add641
|
239 |
struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat; |
3e2f41f1f
|
240 241 242 243 |
reclaim_stat->recent_scanned[file]++; if (rotated) reclaim_stat->recent_rotated[file]++; |
3e2f41f1f
|
244 |
} |
fa9add641
|
245 246 |
static void __activate_page(struct page *page, struct lruvec *lruvec, void *arg) |
1da177e4c
|
247 |
{ |
744ed1442
|
248 |
if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) { |
7a608572a
|
249 250 |
int file = page_is_file_cache(page); int lru = page_lru_base_type(page); |
744ed1442
|
251 |
|
fa9add641
|
252 |
del_page_from_lru_list(page, lruvec, lru); |
7a608572a
|
253 254 |
SetPageActive(page); lru += LRU_ACTIVE; |
fa9add641
|
255 |
add_page_to_lru_list(page, lruvec, lru); |
24b7e5819
|
256 |
trace_mm_lru_activate(page); |
4f98a2fee
|
257 |
|
fa9add641
|
258 259 |
__count_vm_event(PGACTIVATE); update_page_reclaim_stat(lruvec, file, 1); |
1da177e4c
|
260 |
} |
eb709b0d0
|
261 262 263 264 265 266 267 268 269 270 271 272 |
} #ifdef CONFIG_SMP static DEFINE_PER_CPU(struct pagevec, activate_page_pvecs); static void activate_page_drain(int cpu) { struct pagevec *pvec = &per_cpu(activate_page_pvecs, cpu); if (pagevec_count(pvec)) pagevec_lru_move_fn(pvec, __activate_page, NULL); } |
5fbc46163
|
273 274 275 276 |
static bool need_activate_page_drain(int cpu) { return pagevec_count(&per_cpu(activate_page_pvecs, cpu)) != 0; } |
eb709b0d0
|
277 278 279 280 |
void activate_page(struct page *page) { if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) { struct pagevec *pvec = &get_cpu_var(activate_page_pvecs); |
09cbfeaf1
|
281 |
get_page(page); |
eb709b0d0
|
282 283 284 285 286 287 288 289 290 291 |
if (!pagevec_add(pvec, page)) pagevec_lru_move_fn(pvec, __activate_page, NULL); put_cpu_var(activate_page_pvecs); } } #else static inline void activate_page_drain(int cpu) { } |
5fbc46163
|
292 293 294 295 |
static bool need_activate_page_drain(int cpu) { return false; } |
eb709b0d0
|
296 297 298 299 300 |
void activate_page(struct page *page) { struct zone *zone = page_zone(page); spin_lock_irq(&zone->lru_lock); |
fa9add641
|
301 |
__activate_page(page, mem_cgroup_page_lruvec(page, zone), NULL); |
1da177e4c
|
302 303 |
spin_unlock_irq(&zone->lru_lock); } |
eb709b0d0
|
304 |
#endif |
1da177e4c
|
305 |
|
059285a25
|
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 |
static void __lru_cache_activate_page(struct page *page) { struct pagevec *pvec = &get_cpu_var(lru_add_pvec); int i; /* * Search backwards on the optimistic assumption that the page being * activated has just been added to this pagevec. Note that only * the local pagevec is examined as a !PageLRU page could be in the * process of being released, reclaimed, migrated or on a remote * pagevec that is currently being drained. Furthermore, marking * a remote pagevec's page PageActive potentially hits a race where * a page is marked PageActive just after it is added to the inactive * list causing accounting errors and BUG_ON checks to trigger. */ for (i = pagevec_count(pvec) - 1; i >= 0; i--) { struct page *pagevec_page = pvec->pages[i]; if (pagevec_page == page) { SetPageActive(page); break; } } put_cpu_var(lru_add_pvec); } |
1da177e4c
|
332 333 334 335 336 337 |
/* * Mark a page as having seen activity. * * inactive,unreferenced -> inactive,referenced * inactive,referenced -> active,unreferenced * active,unreferenced -> active,referenced |
eb39d618f
|
338 339 340 |
* * When a newly allocated page is not yet visible, so safe for non-atomic ops, * __SetPageReferenced(page) may be substituted for mark_page_accessed(page). |
1da177e4c
|
341 |
*/ |
920c7a5d0
|
342 |
void mark_page_accessed(struct page *page) |
1da177e4c
|
343 |
{ |
e90309c9f
|
344 |
page = compound_head(page); |
894bc3104
|
345 |
if (!PageActive(page) && !PageUnevictable(page) && |
059285a25
|
346 347 348 349 350 351 352 353 354 355 356 357 |
PageReferenced(page)) { /* * If the page is on the LRU, queue it for activation via * activate_page_pvecs. Otherwise, assume the page is on a * pagevec, mark it active and it'll be moved to the active * LRU on the next drain. */ if (PageLRU(page)) activate_page(page); else __lru_cache_activate_page(page); |
1da177e4c
|
358 |
ClearPageReferenced(page); |
a528910e1
|
359 360 |
if (page_is_file_cache(page)) workingset_activation(page); |
1da177e4c
|
361 362 363 |
} else if (!PageReferenced(page)) { SetPageReferenced(page); } |
33c3fc71c
|
364 365 |
if (page_is_idle(page)) clear_page_idle(page); |
1da177e4c
|
366 |
} |
1da177e4c
|
367 |
EXPORT_SYMBOL(mark_page_accessed); |
2329d3751
|
368 |
static void __lru_cache_add(struct page *page) |
1da177e4c
|
369 |
{ |
13f7f7898
|
370 |
struct pagevec *pvec = &get_cpu_var(lru_add_pvec); |
09cbfeaf1
|
371 |
get_page(page); |
d741c9cde
|
372 |
if (!pagevec_space(pvec)) |
a0b8cab3b
|
373 |
__pagevec_lru_add(pvec); |
d741c9cde
|
374 |
pagevec_add(pvec, page); |
13f7f7898
|
375 |
put_cpu_var(lru_add_pvec); |
1da177e4c
|
376 |
} |
2329d3751
|
377 378 379 380 381 382 383 |
/** * lru_cache_add: add a page to the page lists * @page: the page to add */ void lru_cache_add_anon(struct page *page) { |
6fb81a17d
|
384 385 |
if (PageActive(page)) ClearPageActive(page); |
2329d3751
|
386 387 388 389 390 |
__lru_cache_add(page); } void lru_cache_add_file(struct page *page) { |
6fb81a17d
|
391 392 |
if (PageActive(page)) ClearPageActive(page); |
2329d3751
|
393 394 395 |
__lru_cache_add(page); } EXPORT_SYMBOL(lru_cache_add_file); |
1da177e4c
|
396 |
|
f04e9ebbe
|
397 |
/** |
c53954a09
|
398 |
* lru_cache_add - add a page to a page list |
f04e9ebbe
|
399 |
* @page: the page to be added to the LRU. |
2329d3751
|
400 401 402 403 404 |
* * Queue the page for addition to the LRU via pagevec. The decision on whether * to add the page to the [in]active [file|anon] list is deferred until the * pagevec is drained. This gives a chance for the caller of lru_cache_add() * have the page added to the active list using mark_page_accessed(). |
f04e9ebbe
|
405 |
*/ |
c53954a09
|
406 |
void lru_cache_add(struct page *page) |
1da177e4c
|
407 |
{ |
309381fea
|
408 409 |
VM_BUG_ON_PAGE(PageActive(page) && PageUnevictable(page), page); VM_BUG_ON_PAGE(PageLRU(page), page); |
c53954a09
|
410 |
__lru_cache_add(page); |
1da177e4c
|
411 |
} |
894bc3104
|
412 413 414 415 416 417 418 419 420 421 422 423 424 |
/** * add_page_to_unevictable_list - add a page to the unevictable list * @page: the page to be added to the unevictable list * * Add page directly to its zone's unevictable list. To avoid races with * tasks that might be making the page evictable, through eg. munlock, * munmap or exit, while it's not on the lru, we want to add the page * while it's locked or otherwise "invisible" to other tasks. This is * difficult to do when using the pagevec cache, so bypass that. */ void add_page_to_unevictable_list(struct page *page) { struct zone *zone = page_zone(page); |
fa9add641
|
425 |
struct lruvec *lruvec; |
894bc3104
|
426 427 |
spin_lock_irq(&zone->lru_lock); |
fa9add641
|
428 |
lruvec = mem_cgroup_page_lruvec(page, zone); |
ef2a2cbdd
|
429 |
ClearPageActive(page); |
894bc3104
|
430 431 |
SetPageUnevictable(page); SetPageLRU(page); |
fa9add641
|
432 |
add_page_to_lru_list(page, lruvec, LRU_UNEVICTABLE); |
894bc3104
|
433 434 |
spin_unlock_irq(&zone->lru_lock); } |
00501b531
|
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 |
/** * lru_cache_add_active_or_unevictable * @page: the page to be added to LRU * @vma: vma in which page is mapped for determining reclaimability * * Place @page on the active or unevictable LRU list, depending on its * evictability. Note that if the page is not evictable, it goes * directly back onto it's zone's unevictable list, it does NOT use a * per cpu pagevec. */ void lru_cache_add_active_or_unevictable(struct page *page, struct vm_area_struct *vma) { VM_BUG_ON_PAGE(PageLRU(page), page); if (likely((vma->vm_flags & (VM_LOCKED | VM_SPECIAL)) != VM_LOCKED)) { SetPageActive(page); lru_cache_add(page); return; } if (!TestSetPageMlocked(page)) { /* * We use the irq-unsafe __mod_zone_page_stat because this * counter is not modified from interrupt context, and the pte * lock is held(spinlock), which implies preemption disabled. */ __mod_zone_page_state(page_zone(page), NR_MLOCK, hpage_nr_pages(page)); count_vm_event(UNEVICTABLE_PGMLOCKED); } add_page_to_unevictable_list(page); } |
902aaed0d
|
468 |
/* |
315601809
|
469 470 471 472 473 |
* If the page can not be invalidated, it is moved to the * inactive list to speed up its reclaim. It is moved to the * head of the list, rather than the tail, to give the flusher * threads some time to write it out, as this is much more * effective than the single-page writeout from reclaim. |
278df9f45
|
474 475 476 477 478 479 480 481 482 483 484 485 486 487 |
* * If the page isn't page_mapped and dirty/writeback, the page * could reclaim asap using PG_reclaim. * * 1. active, mapped page -> none * 2. active, dirty/writeback page -> inactive, head, PG_reclaim * 3. inactive, mapped page -> none * 4. inactive, dirty/writeback page -> inactive, head, PG_reclaim * 5. inactive, clean -> inactive, tail * 6. Others -> none * * In 4, why it moves inactive's head, the VM expects the page would * be write it out by flusher threads as this is much more effective * than the single-page writeout from reclaim. |
315601809
|
488 |
*/ |
cc5993bd7
|
489 |
static void lru_deactivate_file_fn(struct page *page, struct lruvec *lruvec, |
fa9add641
|
490 |
void *arg) |
315601809
|
491 492 |
{ int lru, file; |
278df9f45
|
493 |
bool active; |
315601809
|
494 |
|
278df9f45
|
495 |
if (!PageLRU(page)) |
315601809
|
496 |
return; |
bad49d9c8
|
497 498 |
if (PageUnevictable(page)) return; |
315601809
|
499 500 501 |
/* Some processes are using the page */ if (page_mapped(page)) return; |
278df9f45
|
502 |
active = PageActive(page); |
315601809
|
503 504 |
file = page_is_file_cache(page); lru = page_lru_base_type(page); |
fa9add641
|
505 506 |
del_page_from_lru_list(page, lruvec, lru + active); |
315601809
|
507 508 |
ClearPageActive(page); ClearPageReferenced(page); |
fa9add641
|
509 |
add_page_to_lru_list(page, lruvec, lru); |
315601809
|
510 |
|
278df9f45
|
511 512 513 514 515 516 517 518 519 520 521 522 |
if (PageWriteback(page) || PageDirty(page)) { /* * PG_reclaim could be raced with end_page_writeback * It can make readahead confusing. But race window * is _really_ small and it's non-critical problem. */ SetPageReclaim(page); } else { /* * The page's writeback ends up during pagevec * We moves tha page into tail of inactive. */ |
925b7673c
|
523 |
list_move_tail(&page->lru, &lruvec->lists[lru]); |
278df9f45
|
524 525 526 527 528 |
__count_vm_event(PGROTATED); } if (active) __count_vm_event(PGDEACTIVATE); |
fa9add641
|
529 |
update_page_reclaim_stat(lruvec, file, 0); |
315601809
|
530 |
} |
10853a039
|
531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 |
static void lru_deactivate_fn(struct page *page, struct lruvec *lruvec, void *arg) { if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) { int file = page_is_file_cache(page); int lru = page_lru_base_type(page); del_page_from_lru_list(page, lruvec, lru + LRU_ACTIVE); ClearPageActive(page); ClearPageReferenced(page); add_page_to_lru_list(page, lruvec, lru); __count_vm_event(PGDEACTIVATE); update_page_reclaim_stat(lruvec, file, 0); } } |
315601809
|
548 |
/* |
902aaed0d
|
549 550 551 552 |
* Drain pages out of the cpu's pagevecs. * Either "cpu" is the current CPU, and preemption has already been * disabled; or "cpu" is being hot-unplugged, and is already dead. */ |
f0cb3c76a
|
553 |
void lru_add_drain_cpu(int cpu) |
1da177e4c
|
554 |
{ |
13f7f7898
|
555 |
struct pagevec *pvec = &per_cpu(lru_add_pvec, cpu); |
1da177e4c
|
556 |
|
13f7f7898
|
557 |
if (pagevec_count(pvec)) |
a0b8cab3b
|
558 |
__pagevec_lru_add(pvec); |
902aaed0d
|
559 560 561 562 563 564 565 566 567 568 |
pvec = &per_cpu(lru_rotate_pvecs, cpu); if (pagevec_count(pvec)) { unsigned long flags; /* No harm done if a racing interrupt already did this */ local_irq_save(flags); pagevec_move_tail(pvec); local_irq_restore(flags); } |
315601809
|
569 |
|
cc5993bd7
|
570 |
pvec = &per_cpu(lru_deactivate_file_pvecs, cpu); |
315601809
|
571 |
if (pagevec_count(pvec)) |
cc5993bd7
|
572 |
pagevec_lru_move_fn(pvec, lru_deactivate_file_fn, NULL); |
eb709b0d0
|
573 |
|
10853a039
|
574 575 576 |
pvec = &per_cpu(lru_deactivate_pvecs, cpu); if (pagevec_count(pvec)) pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL); |
eb709b0d0
|
577 |
activate_page_drain(cpu); |
315601809
|
578 579 580 |
} /** |
cc5993bd7
|
581 |
* deactivate_file_page - forcefully deactivate a file page |
315601809
|
582 583 584 585 586 587 |
* @page: page to deactivate * * This function hints the VM that @page is a good reclaim candidate, * for example if its invalidation fails due to the page being dirty * or under writeback. */ |
cc5993bd7
|
588 |
void deactivate_file_page(struct page *page) |
315601809
|
589 |
{ |
821ed6bbe
|
590 |
/* |
cc5993bd7
|
591 592 |
* In a workload with many unevictable page such as mprotect, * unevictable page deactivation for accelerating reclaim is pointless. |
821ed6bbe
|
593 594 595 |
*/ if (PageUnevictable(page)) return; |
315601809
|
596 |
if (likely(get_page_unless_zero(page))) { |
cc5993bd7
|
597 |
struct pagevec *pvec = &get_cpu_var(lru_deactivate_file_pvecs); |
315601809
|
598 599 |
if (!pagevec_add(pvec, page)) |
cc5993bd7
|
600 601 |
pagevec_lru_move_fn(pvec, lru_deactivate_file_fn, NULL); put_cpu_var(lru_deactivate_file_pvecs); |
315601809
|
602 |
} |
80bfed904
|
603 |
} |
10853a039
|
604 605 606 607 608 609 610 611 612 613 614 615 |
/** * deactivate_page - deactivate a page * @page: page to deactivate * * deactivate_page() moves @page to the inactive list if @page was on the active * list and was not an unevictable page. This is done to accelerate the reclaim * of @page. */ void deactivate_page(struct page *page) { if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) { struct pagevec *pvec = &get_cpu_var(lru_deactivate_pvecs); |
09cbfeaf1
|
616 |
get_page(page); |
10853a039
|
617 618 619 620 621 |
if (!pagevec_add(pvec, page)) pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL); put_cpu_var(lru_deactivate_pvecs); } } |
80bfed904
|
622 623 |
void lru_add_drain(void) { |
f0cb3c76a
|
624 |
lru_add_drain_cpu(get_cpu()); |
80bfed904
|
625 |
put_cpu(); |
1da177e4c
|
626 |
} |
c4028958b
|
627 |
static void lru_add_drain_per_cpu(struct work_struct *dummy) |
053837fce
|
628 629 630 |
{ lru_add_drain(); } |
5fbc46163
|
631 632 633 |
static DEFINE_PER_CPU(struct work_struct, lru_add_drain_work); void lru_add_drain_all(void) |
053837fce
|
634 |
{ |
5fbc46163
|
635 636 637 638 639 640 641 642 643 644 645 646 647 |
static DEFINE_MUTEX(lock); static struct cpumask has_work; int cpu; mutex_lock(&lock); get_online_cpus(); cpumask_clear(&has_work); for_each_online_cpu(cpu) { struct work_struct *work = &per_cpu(lru_add_drain_work, cpu); if (pagevec_count(&per_cpu(lru_add_pvec, cpu)) || pagevec_count(&per_cpu(lru_rotate_pvecs, cpu)) || |
cc5993bd7
|
648 |
pagevec_count(&per_cpu(lru_deactivate_file_pvecs, cpu)) || |
10853a039
|
649 |
pagevec_count(&per_cpu(lru_deactivate_pvecs, cpu)) || |
5fbc46163
|
650 651 652 653 654 655 656 657 658 659 660 661 |
need_activate_page_drain(cpu)) { INIT_WORK(work, lru_add_drain_per_cpu); schedule_work_on(cpu, work); cpumask_set_cpu(cpu, &has_work); } } for_each_cpu(cpu, &has_work) flush_work(&per_cpu(lru_add_drain_work, cpu)); put_online_cpus(); mutex_unlock(&lock); |
053837fce
|
662 |
} |
aabfb5729
|
663 |
/** |
ea1754a08
|
664 |
* release_pages - batched put_page() |
aabfb5729
|
665 666 667 |
* @pages: array of pages to release * @nr: number of pages * @cold: whether the pages are cache cold |
1da177e4c
|
668 |
* |
aabfb5729
|
669 670 |
* Decrement the reference count on all the pages in @pages. If it * fell to zero, remove the page from the LRU and free it. |
1da177e4c
|
671 |
*/ |
b745bc85f
|
672 |
void release_pages(struct page **pages, int nr, bool cold) |
1da177e4c
|
673 674 |
{ int i; |
cc59850ef
|
675 |
LIST_HEAD(pages_to_free); |
1da177e4c
|
676 |
struct zone *zone = NULL; |
fa9add641
|
677 |
struct lruvec *lruvec; |
902aaed0d
|
678 |
unsigned long uninitialized_var(flags); |
aabfb5729
|
679 |
unsigned int uninitialized_var(lock_batch); |
1da177e4c
|
680 |
|
1da177e4c
|
681 682 |
for (i = 0; i < nr; i++) { struct page *page = pages[i]; |
1da177e4c
|
683 |
|
aabfb5729
|
684 685 686 687 688 689 690 691 692 |
/* * Make sure the IRQ-safe lock-holding time does not get * excessive with a continuous string of pages from the * same zone. The lock is held only if zone != NULL. */ if (zone && ++lock_batch == SWAP_CLUSTER_MAX) { spin_unlock_irqrestore(&zone->lru_lock, flags); zone = NULL; } |
ddc58f27f
|
693 |
page = compound_head(page); |
b5810039a
|
694 |
if (!put_page_testzero(page)) |
1da177e4c
|
695 |
continue; |
ddc58f27f
|
696 697 698 699 700 701 702 703 |
if (PageCompound(page)) { if (zone) { spin_unlock_irqrestore(&zone->lru_lock, flags); zone = NULL; } __put_compound_page(page); continue; } |
46453a6e1
|
704 705 |
if (PageLRU(page)) { struct zone *pagezone = page_zone(page); |
894bc3104
|
706 |
|
46453a6e1
|
707 708 |
if (pagezone != zone) { if (zone) |
902aaed0d
|
709 710 |
spin_unlock_irqrestore(&zone->lru_lock, flags); |
aabfb5729
|
711 |
lock_batch = 0; |
46453a6e1
|
712 |
zone = pagezone; |
902aaed0d
|
713 |
spin_lock_irqsave(&zone->lru_lock, flags); |
46453a6e1
|
714 |
} |
fa9add641
|
715 716 |
lruvec = mem_cgroup_page_lruvec(page, zone); |
309381fea
|
717 |
VM_BUG_ON_PAGE(!PageLRU(page), page); |
674539115
|
718 |
__ClearPageLRU(page); |
fa9add641
|
719 |
del_page_from_lru_list(page, lruvec, page_off_lru(page)); |
46453a6e1
|
720 |
} |
c53954a09
|
721 |
/* Clear Active bit in case of parallel mark_page_accessed */ |
e3741b506
|
722 |
__ClearPageActive(page); |
c53954a09
|
723 |
|
cc59850ef
|
724 |
list_add(&page->lru, &pages_to_free); |
1da177e4c
|
725 726 |
} if (zone) |
902aaed0d
|
727 |
spin_unlock_irqrestore(&zone->lru_lock, flags); |
1da177e4c
|
728 |
|
747db954c
|
729 |
mem_cgroup_uncharge_list(&pages_to_free); |
cc59850ef
|
730 |
free_hot_cold_page_list(&pages_to_free, cold); |
1da177e4c
|
731 |
} |
0be8557bc
|
732 |
EXPORT_SYMBOL(release_pages); |
1da177e4c
|
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 |
/* * The pages which we're about to release may be in the deferred lru-addition * queues. That would prevent them from really being freed right now. That's * OK from a correctness point of view but is inefficient - those pages may be * cache-warm and we want to give them back to the page allocator ASAP. * * So __pagevec_release() will drain those queues here. __pagevec_lru_add() * and __pagevec_lru_add_active() call release_pages() directly to avoid * mutual recursion. */ void __pagevec_release(struct pagevec *pvec) { lru_add_drain(); release_pages(pvec->pages, pagevec_count(pvec), pvec->cold); pagevec_reinit(pvec); } |
7f2857018
|
750 |
EXPORT_SYMBOL(__pagevec_release); |
12d271078
|
751 |
#ifdef CONFIG_TRANSPARENT_HUGEPAGE |
71e3aac07
|
752 |
/* used by __split_huge_page_refcount() */ |
fa9add641
|
753 |
void lru_add_page_tail(struct page *page, struct page *page_tail, |
5bc7b8aca
|
754 |
struct lruvec *lruvec, struct list_head *list) |
71e3aac07
|
755 |
{ |
71e3aac07
|
756 |
const int file = 0; |
71e3aac07
|
757 |
|
309381fea
|
758 759 760 |
VM_BUG_ON_PAGE(!PageHead(page), page); VM_BUG_ON_PAGE(PageCompound(page_tail), page); VM_BUG_ON_PAGE(PageLRU(page_tail), page); |
fa9add641
|
761 762 |
VM_BUG_ON(NR_CPUS != 1 && !spin_is_locked(&lruvec_zone(lruvec)->lru_lock)); |
71e3aac07
|
763 |
|
5bc7b8aca
|
764 765 |
if (!list) SetPageLRU(page_tail); |
71e3aac07
|
766 |
|
12d271078
|
767 768 |
if (likely(PageLRU(page))) list_add_tail(&page_tail->lru, &page->lru); |
5bc7b8aca
|
769 770 771 772 773 |
else if (list) { /* page reclaim is reclaiming a huge page */ get_page(page_tail); list_add_tail(&page_tail->lru, list); } else { |
12d271078
|
774 775 776 777 778 779 780 781 |
struct list_head *list_head; /* * Head page has not yet been counted, as an hpage, * so we must account for each subpage individually. * * Use the standard add function to put page_tail on the list, * but then correct its position so they all end up in order. */ |
e180cf806
|
782 |
add_page_to_lru_list(page_tail, lruvec, page_lru(page_tail)); |
12d271078
|
783 784 |
list_head = page_tail->lru.prev; list_move_tail(&page_tail->lru, list_head); |
71e3aac07
|
785 |
} |
7512102cf
|
786 787 |
if (!PageUnevictable(page)) |
e180cf806
|
788 |
update_page_reclaim_stat(lruvec, file, PageActive(page_tail)); |
71e3aac07
|
789 |
} |
12d271078
|
790 |
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ |
71e3aac07
|
791 |
|
fa9add641
|
792 793 |
static void __pagevec_lru_add_fn(struct page *page, struct lruvec *lruvec, void *arg) |
3dd7ae8ec
|
794 |
{ |
13f7f7898
|
795 796 797 |
int file = page_is_file_cache(page); int active = PageActive(page); enum lru_list lru = page_lru(page); |
3dd7ae8ec
|
798 |
|
309381fea
|
799 |
VM_BUG_ON_PAGE(PageLRU(page), page); |
3dd7ae8ec
|
800 801 |
SetPageLRU(page); |
fa9add641
|
802 803 |
add_page_to_lru_list(page, lruvec, lru); update_page_reclaim_stat(lruvec, file, active); |
24b7e5819
|
804 |
trace_mm_lru_insertion(page, lru); |
3dd7ae8ec
|
805 |
} |
1da177e4c
|
806 |
/* |
1da177e4c
|
807 808 809 |
* Add the passed pages to the LRU, then drop the caller's refcount * on them. Reinitialises the caller's pagevec. */ |
a0b8cab3b
|
810 |
void __pagevec_lru_add(struct pagevec *pvec) |
1da177e4c
|
811 |
{ |
a0b8cab3b
|
812 |
pagevec_lru_move_fn(pvec, __pagevec_lru_add_fn, NULL); |
1da177e4c
|
813 |
} |
5095ae837
|
814 |
EXPORT_SYMBOL(__pagevec_lru_add); |
1da177e4c
|
815 |
|
1da177e4c
|
816 |
/** |
0cd6144aa
|
817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 |
* pagevec_lookup_entries - gang pagecache lookup * @pvec: Where the resulting entries are placed * @mapping: The address_space to search * @start: The starting entry index * @nr_entries: The maximum number of entries * @indices: The cache indices corresponding to the entries in @pvec * * pagevec_lookup_entries() will search for and return a group of up * to @nr_entries pages and shadow entries in the mapping. All * entries are placed in @pvec. pagevec_lookup_entries() takes a * reference against actual pages in @pvec. * * The search returns a group of mapping-contiguous entries with * ascending indexes. There may be holes in the indices due to * not-present entries. * * pagevec_lookup_entries() returns the number of entries which were * found. */ unsigned pagevec_lookup_entries(struct pagevec *pvec, struct address_space *mapping, pgoff_t start, unsigned nr_pages, pgoff_t *indices) { pvec->nr = find_get_entries(mapping, start, nr_pages, pvec->pages, indices); return pagevec_count(pvec); } /** * pagevec_remove_exceptionals - pagevec exceptionals pruning * @pvec: The pagevec to prune * * pagevec_lookup_entries() fills both pages and exceptional radix * tree entries into the pagevec. This function prunes all * exceptionals from @pvec without leaving holes, so that it can be * passed on to page-only pagevec operations. */ void pagevec_remove_exceptionals(struct pagevec *pvec) { int i, j; for (i = 0, j = 0; i < pagevec_count(pvec); i++) { struct page *page = pvec->pages[i]; if (!radix_tree_exceptional_entry(page)) pvec->pages[j++] = page; } pvec->nr = j; } /** |
1da177e4c
|
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 |
* pagevec_lookup - gang pagecache lookup * @pvec: Where the resulting pages are placed * @mapping: The address_space to search * @start: The starting page index * @nr_pages: The maximum number of pages * * pagevec_lookup() will search for and return a group of up to @nr_pages pages * in the mapping. The pages are placed in @pvec. pagevec_lookup() takes a * reference against the pages in @pvec. * * The search returns a group of mapping-contiguous pages with ascending * indexes. There may be holes in the indices due to not-present pages. * * pagevec_lookup() returns the number of pages which were found. */ unsigned pagevec_lookup(struct pagevec *pvec, struct address_space *mapping, pgoff_t start, unsigned nr_pages) { pvec->nr = find_get_pages(mapping, start, nr_pages, pvec->pages); return pagevec_count(pvec); } |
78539fdfa
|
889 |
EXPORT_SYMBOL(pagevec_lookup); |
1da177e4c
|
890 891 892 893 894 895 896 |
unsigned pagevec_lookup_tag(struct pagevec *pvec, struct address_space *mapping, pgoff_t *index, int tag, unsigned nr_pages) { pvec->nr = find_get_pages_tag(mapping, index, tag, nr_pages, pvec->pages); return pagevec_count(pvec); } |
7f2857018
|
897 |
EXPORT_SYMBOL(pagevec_lookup_tag); |
1da177e4c
|
898 |
|
1da177e4c
|
899 900 901 902 903 |
/* * Perform any setup for the swap system */ void __init swap_setup(void) { |
4481374ce
|
904 |
unsigned long megs = totalram_pages >> (20 - PAGE_SHIFT); |
e0bf68dde
|
905 |
#ifdef CONFIG_SWAP |
33806f06d
|
906 |
int i; |
27ba0644e
|
907 |
for (i = 0; i < MAX_SWAPFILES; i++) |
33806f06d
|
908 |
spin_lock_init(&swapper_spaces[i].tree_lock); |
e0bf68dde
|
909 |
#endif |
1da177e4c
|
910 911 912 913 914 915 916 917 918 |
/* Use a smaller cluster for small-memory machines */ if (megs < 16) page_cluster = 2; else page_cluster = 3; /* * Right now other parts of the system means that we * _really_ don't want to cluster much more */ |
1da177e4c
|
919 |
} |