Blame view
mm/vmscan.c
45.4 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
/* * linux/mm/vmscan.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds * * Swap reorganised 29.12.95, Stephen Tweedie. * kswapd added: 7.1.96 sct * Removed kswapd_ctl limits, and swap out as many pages as needed * to bring the system back to freepages.high: 2.4.97, Rik van Riel. * Zone aware kswapd started 02/00, Kanoj Sarcar (kanoj@sgi.com). * Multiqueue VM started 5.8.00, Rik van Riel. */ #include <linux/mm.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/kernel_stat.h> #include <linux/swap.h> #include <linux/pagemap.h> #include <linux/init.h> #include <linux/highmem.h> |
e129b5c23 [PATCH] vm: add p... |
22 |
#include <linux/vmstat.h> |
1da177e4c Linux-2.6.12-rc2 |
23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
#include <linux/file.h> #include <linux/writeback.h> #include <linux/blkdev.h> #include <linux/buffer_head.h> /* for try_to_release_page(), buffer_heads_over_limit */ #include <linux/mm_inline.h> #include <linux/pagevec.h> #include <linux/backing-dev.h> #include <linux/rmap.h> #include <linux/topology.h> #include <linux/cpu.h> #include <linux/cpuset.h> #include <linux/notifier.h> #include <linux/rwsem.h> |
248a0301e [PATCH] mm: make ... |
37 |
#include <linux/delay.h> |
3218ae14b [PATCH] pgdat all... |
38 |
#include <linux/kthread.h> |
1da177e4c Linux-2.6.12-rc2 |
39 40 41 42 43 |
#include <asm/tlbflush.h> #include <asm/div64.h> #include <linux/swapops.h> |
0f8053a50 [PATCH] mm: make ... |
44 |
#include "internal.h" |
1da177e4c Linux-2.6.12-rc2 |
45 |
struct scan_control { |
1da177e4c Linux-2.6.12-rc2 |
46 47 |
/* Incremented by the number of inactive pages that were scanned */ unsigned long nr_scanned; |
1da177e4c Linux-2.6.12-rc2 |
48 |
/* This context's GFP mask */ |
6daa0e286 [PATCH] gfp_t: mm... |
49 |
gfp_t gfp_mask; |
1da177e4c Linux-2.6.12-rc2 |
50 51 |
int may_writepage; |
f1fd1067e [PATCH] Zone recl... |
52 53 |
/* Can pages be swapped as part of reclaim? */ int may_swap; |
1da177e4c Linux-2.6.12-rc2 |
54 55 56 57 58 |
/* This context's SWAP_CLUSTER_MAX. If freeing memory for * suspend, we effectively ignore SWAP_CLUSTER_MAX. * In this context, it doesn't matter that we scan the * whole list at once. */ int swap_cluster_max; |
d6277db4a [PATCH] swsusp: r... |
59 60 |
int swappiness; |
408d85441 [PATCH] oom: use ... |
61 62 |
int all_unreclaimable; |
1da177e4c Linux-2.6.12-rc2 |
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 |
}; /* * The list of shrinker callbacks used by to apply pressure to * ageable caches. */ struct shrinker { shrinker_t shrinker; struct list_head list; int seeks; /* seeks to recreate an obj */ long nr; /* objs pending delete */ }; #define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru)) #ifdef ARCH_HAS_PREFETCH #define prefetch_prev_lru_page(_page, _base, _field) \ do { \ if ((_page)->lru.prev != _base) { \ struct page *prev; \ \ prev = lru_to_page(&(_page->lru)); \ prefetch(&prev->_field); \ } \ } while (0) #else #define prefetch_prev_lru_page(_page, _base, _field) do { } while (0) #endif #ifdef ARCH_HAS_PREFETCHW #define prefetchw_prev_lru_page(_page, _base, _field) \ do { \ if ((_page)->lru.prev != _base) { \ struct page *prev; \ \ prev = lru_to_page(&(_page->lru)); \ prefetchw(&prev->_field); \ } \ } while (0) #else #define prefetchw_prev_lru_page(_page, _base, _field) do { } while (0) #endif /* * From 0 .. 100. Higher means more swappy. */ int vm_swappiness = 60; |
bd1e22b8e [PATCH] initialis... |
110 |
long vm_total_pages; /* The total number of pages which the VM controls */ |
1da177e4c Linux-2.6.12-rc2 |
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 |
static LIST_HEAD(shrinker_list); static DECLARE_RWSEM(shrinker_rwsem); /* * Add a shrinker callback to be called from the vm */ struct shrinker *set_shrinker(int seeks, shrinker_t theshrinker) { struct shrinker *shrinker; shrinker = kmalloc(sizeof(*shrinker), GFP_KERNEL); if (shrinker) { shrinker->shrinker = theshrinker; shrinker->seeks = seeks; shrinker->nr = 0; down_write(&shrinker_rwsem); list_add_tail(&shrinker->list, &shrinker_list); up_write(&shrinker_rwsem); } return shrinker; } EXPORT_SYMBOL(set_shrinker); /* * Remove one */ void remove_shrinker(struct shrinker *shrinker) { down_write(&shrinker_rwsem); list_del(&shrinker->list); up_write(&shrinker_rwsem); kfree(shrinker); } EXPORT_SYMBOL(remove_shrinker); #define SHRINK_BATCH 128 /* * Call the shrink functions to age shrinkable caches * * Here we assume it costs one seek to replace a lru page and that it also * takes a seek to recreate a cache object. With this in mind we age equal * percentages of the lru and ageable caches. This should balance the seeks * generated by these structures. * * If the vm encounted mapped pages on the LRU it increase the pressure on * slab to avoid swapping. * * We do weird things to avoid (scanned*seeks*entries) overflowing 32 bits. * * `lru_pages' represents the number of on-LRU pages in all the zones which * are eligible for the caller's allocation attempt. It is used for balancing * slab reclaim versus page reclaim. |
b15e0905f [PATCH] vmscan: n... |
164 165 |
* * Returns the number of slab objects which we shrunk. |
1da177e4c Linux-2.6.12-rc2 |
166 |
*/ |
69e05944a [PATCH] vmscan: u... |
167 168 |
unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask, unsigned long lru_pages) |
1da177e4c Linux-2.6.12-rc2 |
169 170 |
{ struct shrinker *shrinker; |
69e05944a [PATCH] vmscan: u... |
171 |
unsigned long ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
172 173 174 175 176 |
if (scanned == 0) scanned = SWAP_CLUSTER_MAX; if (!down_read_trylock(&shrinker_rwsem)) |
b15e0905f [PATCH] vmscan: n... |
177 |
return 1; /* Assume we'll be able to shrink next time */ |
1da177e4c Linux-2.6.12-rc2 |
178 179 180 181 |
list_for_each_entry(shrinker, &shrinker_list, list) { unsigned long long delta; unsigned long total_scan; |
ea164d73a [PATCH] shrinker-... |
182 |
unsigned long max_pass = (*shrinker->shrinker)(0, gfp_mask); |
1da177e4c Linux-2.6.12-rc2 |
183 184 |
delta = (4 * scanned) / shrinker->seeks; |
ea164d73a [PATCH] shrinker-... |
185 |
delta *= max_pass; |
1da177e4c Linux-2.6.12-rc2 |
186 187 |
do_div(delta, lru_pages + 1); shrinker->nr += delta; |
ea164d73a [PATCH] shrinker-... |
188 189 190 191 192 193 194 195 196 197 198 199 200 201 |
if (shrinker->nr < 0) { printk(KERN_ERR "%s: nr=%ld ", __FUNCTION__, shrinker->nr); shrinker->nr = max_pass; } /* * Avoid risking looping forever due to too large nr value: * never try to free more than twice the estimate number of * freeable entries. */ if (shrinker->nr > max_pass * 2) shrinker->nr = max_pass * 2; |
1da177e4c Linux-2.6.12-rc2 |
202 203 204 205 206 207 208 |
total_scan = shrinker->nr; shrinker->nr = 0; while (total_scan >= SHRINK_BATCH) { long this_scan = SHRINK_BATCH; int shrink_ret; |
b15e0905f [PATCH] vmscan: n... |
209 |
int nr_before; |
1da177e4c Linux-2.6.12-rc2 |
210 |
|
b15e0905f [PATCH] vmscan: n... |
211 |
nr_before = (*shrinker->shrinker)(0, gfp_mask); |
1da177e4c Linux-2.6.12-rc2 |
212 213 214 |
shrink_ret = (*shrinker->shrinker)(this_scan, gfp_mask); if (shrink_ret == -1) break; |
b15e0905f [PATCH] vmscan: n... |
215 216 |
if (shrink_ret < nr_before) ret += nr_before - shrink_ret; |
f8891e5e1 [PATCH] Light wei... |
217 |
count_vm_events(SLABS_SCANNED, this_scan); |
1da177e4c Linux-2.6.12-rc2 |
218 219 220 221 222 223 224 225 |
total_scan -= this_scan; cond_resched(); } shrinker->nr += total_scan; } up_read(&shrinker_rwsem); |
b15e0905f [PATCH] vmscan: n... |
226 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 |
} /* Called without lock on whether page is mapped, so answer is unstable */ static inline int page_mapping_inuse(struct page *page) { struct address_space *mapping; /* Page is in somebody's page tables. */ if (page_mapped(page)) return 1; /* Be more reluctant to reclaim swapcache than pagecache */ if (PageSwapCache(page)) return 1; mapping = page_mapping(page); if (!mapping) return 0; /* File is mmap'd by somebody? */ return mapping_mapped(mapping); } static inline int is_page_cache_freeable(struct page *page) { return page_count(page) - !!PagePrivate(page) == 2; } static int may_write_to_queue(struct backing_dev_info *bdi) { |
930d91525 [PATCH] Swap Migr... |
257 |
if (current->flags & PF_SWAPWRITE) |
1da177e4c Linux-2.6.12-rc2 |
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 |
return 1; if (!bdi_write_congested(bdi)) return 1; if (bdi == current->backing_dev_info) return 1; return 0; } /* * We detected a synchronous write error writing a page out. Probably * -ENOSPC. We need to propagate that into the address_space for a subsequent * fsync(), msync() or close(). * * The tricky part is that after writepage we cannot touch the mapping: nothing * prevents it from being freed up. But we have a ref on the page and once * that page is locked, the mapping is pinned. * * We're allowed to run sleeping lock_page() here because we know the caller has * __GFP_FS. */ static void handle_write_error(struct address_space *mapping, struct page *page, int error) { lock_page(page); if (page_mapping(page) == mapping) { if (error == -ENOSPC) set_bit(AS_ENOSPC, &mapping->flags); else set_bit(AS_EIO, &mapping->flags); } unlock_page(page); } |
04e62a29b [PATCH] More page... |
290 291 292 293 294 295 296 297 298 299 300 |
/* possible outcome of pageout() */ typedef enum { /* failed to write page out, page is locked */ PAGE_KEEP, /* move page to the active list, page is locked */ PAGE_ACTIVATE, /* page has been sent to the disk successfully, page is unlocked */ PAGE_SUCCESS, /* page is clean and locked */ PAGE_CLEAN, } pageout_t; |
1da177e4c Linux-2.6.12-rc2 |
301 |
/* |
1742f19fa [PATCH] vmscan: r... |
302 303 |
* pageout is called by shrink_page_list() for each dirty page. * Calls ->writepage(). |
1da177e4c Linux-2.6.12-rc2 |
304 |
*/ |
04e62a29b [PATCH] More page... |
305 |
static pageout_t pageout(struct page *page, struct address_space *mapping) |
1da177e4c Linux-2.6.12-rc2 |
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 |
{ /* * If the page is dirty, only perform writeback if that write * will be non-blocking. To prevent this allocation from being * stalled by pagecache activity. But note that there may be * stalls if we need to run get_block(). We could test * PagePrivate for that. * * If this process is currently in generic_file_write() against * this page's queue, we can perform writeback even if that * will block. * * If the page is swapcache, write it back even if that would * block, for some throttling. This happens by accident, because * swap_backing_dev_info is bust: it doesn't reflect the * congestion state of the swapdevs. Easy to fix, if needed. * See swapfile.c:page_queue_congested(). */ if (!is_page_cache_freeable(page)) return PAGE_KEEP; if (!mapping) { /* * Some data journaling orphaned pages can have * page->mapping == NULL while being dirty with clean buffers. */ |
323aca6c0 [PATCH] vmscan: p... |
331 |
if (PagePrivate(page)) { |
1da177e4c Linux-2.6.12-rc2 |
332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 |
if (try_to_free_buffers(page)) { ClearPageDirty(page); printk("%s: orphaned page ", __FUNCTION__); return PAGE_CLEAN; } } return PAGE_KEEP; } if (mapping->a_ops->writepage == NULL) return PAGE_ACTIVATE; if (!may_write_to_queue(mapping->backing_dev_info)) return PAGE_KEEP; if (clear_page_dirty_for_io(page)) { int res; struct writeback_control wbc = { .sync_mode = WB_SYNC_NONE, .nr_to_write = SWAP_CLUSTER_MAX, |
111ebb6e6 [PATCH] writeback... |
351 352 |
.range_start = 0, .range_end = LLONG_MAX, |
1da177e4c Linux-2.6.12-rc2 |
353 354 355 356 357 358 359 360 |
.nonblocking = 1, .for_reclaim = 1, }; SetPageReclaim(page); res = mapping->a_ops->writepage(page, &wbc); if (res < 0) handle_write_error(mapping, page, res); |
994fc28c7 [PATCH] add AOP_T... |
361 |
if (res == AOP_WRITEPAGE_ACTIVATE) { |
1da177e4c Linux-2.6.12-rc2 |
362 363 364 365 366 367 368 |
ClearPageReclaim(page); return PAGE_ACTIVATE; } if (!PageWriteback(page)) { /* synchronous write or broken a_ops? */ ClearPageReclaim(page); } |
e129b5c23 [PATCH] vm: add p... |
369 |
inc_zone_page_state(page, NR_VMSCAN_WRITE); |
1da177e4c Linux-2.6.12-rc2 |
370 371 372 373 374 |
return PAGE_SUCCESS; } return PAGE_CLEAN; } |
b20a35035 [PATCH] page migr... |
375 |
int remove_mapping(struct address_space *mapping, struct page *page) |
49d2e9cc4 [PATCH] Swap Migr... |
376 |
{ |
28e4d965e [PATCH] mm: remov... |
377 378 |
BUG_ON(!PageLocked(page)); BUG_ON(mapping != page_mapping(page)); |
49d2e9cc4 [PATCH] Swap Migr... |
379 380 |
write_lock_irq(&mapping->tree_lock); |
49d2e9cc4 [PATCH] Swap Migr... |
381 |
/* |
0fd0e6b05 [PATCH] page inva... |
382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 |
* The non racy check for a busy page. * * Must be careful with the order of the tests. When someone has * a ref to the page, it may be possible that they dirty it then * drop the reference. So if PageDirty is tested before page_count * here, then the following race may occur: * * get_user_pages(&page); * [user mapping goes away] * write_to(page); * !PageDirty(page) [good] * SetPageDirty(page); * put_page(page); * !page_count(page) [good, discard it] * * [oops, our write_to data is lost] * * Reversing the order of the tests ensures such a situation cannot * escape unnoticed. The smp_rmb is needed to ensure the page->flags * load is not satisfied before that of page->_count. * * Note that if SetPageDirty is always performed via set_page_dirty, * and thus under tree_lock, then this ordering is not required. |
49d2e9cc4 [PATCH] Swap Migr... |
405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 |
*/ if (unlikely(page_count(page) != 2)) goto cannot_free; smp_rmb(); if (unlikely(PageDirty(page))) goto cannot_free; if (PageSwapCache(page)) { swp_entry_t swap = { .val = page_private(page) }; __delete_from_swap_cache(page); write_unlock_irq(&mapping->tree_lock); swap_free(swap); __put_page(page); /* The pagecache ref */ return 1; } __remove_from_page_cache(page); write_unlock_irq(&mapping->tree_lock); __put_page(page); return 1; cannot_free: write_unlock_irq(&mapping->tree_lock); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
430 |
/* |
1742f19fa [PATCH] vmscan: r... |
431 |
* shrink_page_list() returns the number of reclaimed pages |
1da177e4c Linux-2.6.12-rc2 |
432 |
*/ |
1742f19fa [PATCH] vmscan: r... |
433 434 |
static unsigned long shrink_page_list(struct list_head *page_list, struct scan_control *sc) |
1da177e4c Linux-2.6.12-rc2 |
435 436 437 438 |
{ LIST_HEAD(ret_pages); struct pagevec freed_pvec; int pgactivate = 0; |
05ff51376 [PATCH] vmscan re... |
439 |
unsigned long nr_reclaimed = 0; |
1da177e4c Linux-2.6.12-rc2 |
440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 |
cond_resched(); pagevec_init(&freed_pvec, 1); while (!list_empty(page_list)) { struct address_space *mapping; struct page *page; int may_enter_fs; int referenced; cond_resched(); page = lru_to_page(page_list); list_del(&page->lru); if (TestSetPageLocked(page)) goto keep; |
725d704ec [PATCH] mm: VM_BU... |
457 |
VM_BUG_ON(PageActive(page)); |
1da177e4c Linux-2.6.12-rc2 |
458 459 |
sc->nr_scanned++; |
80e434260 [PATCH] zone recl... |
460 461 462 |
if (!sc->may_swap && page_mapped(page)) goto keep_locked; |
1da177e4c Linux-2.6.12-rc2 |
463 464 465 466 467 468 |
/* Double the slab pressure for mapped and swapcache pages */ if (page_mapped(page) || PageSwapCache(page)) sc->nr_scanned++; if (PageWriteback(page)) goto keep_locked; |
f7b7fd8f3 [PATCH] temporari... |
469 |
referenced = page_referenced(page, 1); |
1da177e4c Linux-2.6.12-rc2 |
470 471 472 473 474 475 476 477 478 |
/* In active use or really unfreeable? Activate it. */ if (referenced && page_mapping_inuse(page)) goto activate_locked; #ifdef CONFIG_SWAP /* * Anonymous process memory has backing store? * Try to allocate it some swap space here. */ |
6e5ef1a96 [PATCH] vmscan: e... |
479 |
if (PageAnon(page) && !PageSwapCache(page)) |
1480a540c [PATCH] SwapMig: ... |
480 |
if (!add_to_swap(page, GFP_ATOMIC)) |
1da177e4c Linux-2.6.12-rc2 |
481 |
goto activate_locked; |
1da177e4c Linux-2.6.12-rc2 |
482 483 484 485 486 487 488 489 490 491 492 |
#endif /* CONFIG_SWAP */ mapping = page_mapping(page); may_enter_fs = (sc->gfp_mask & __GFP_FS) || (PageSwapCache(page) && (sc->gfp_mask & __GFP_IO)); /* * The page is mapped into the page tables of one or more * processes. Try to unmap it here. */ if (page_mapped(page) && mapping) { |
a48d07afd [PATCH] Direct Mi... |
493 |
switch (try_to_unmap(page, 0)) { |
1da177e4c Linux-2.6.12-rc2 |
494 495 496 497 498 499 500 501 502 503 504 505 506 507 |
case SWAP_FAIL: goto activate_locked; case SWAP_AGAIN: goto keep_locked; case SWAP_SUCCESS: ; /* try to free the page below */ } } if (PageDirty(page)) { if (referenced) goto keep_locked; if (!may_enter_fs) goto keep_locked; |
52a8363ea [PATCH] mm: impro... |
508 |
if (!sc->may_writepage) |
1da177e4c Linux-2.6.12-rc2 |
509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 |
goto keep_locked; /* Page is dirty, try to write it out here */ switch(pageout(page, mapping)) { case PAGE_KEEP: goto keep_locked; case PAGE_ACTIVATE: goto activate_locked; case PAGE_SUCCESS: if (PageWriteback(page) || PageDirty(page)) goto keep; /* * A synchronous write - probably a ramdisk. Go * ahead and try to reclaim the page. */ if (TestSetPageLocked(page)) goto keep; if (PageDirty(page) || PageWriteback(page)) goto keep_locked; mapping = page_mapping(page); case PAGE_CLEAN: ; /* try to free the page below */ } } /* * If the page has buffers, try to free the buffer mappings * associated with this page. If we succeed we try to free * the page as well. * * We do this even if the page is PageDirty(). * try_to_release_page() does not perform I/O, but it is * possible for a page to have PageDirty set, but it is actually * clean (all its buffers are clean). This happens if the * buffers were written out directly, with submit_bh(). ext3 * will do this, as well as the blockdev mapping. * try_to_release_page() will discover that cleanness and will * drop the buffers and mark the page clean - it can be freed. * * Rarely, pages can have buffers and no ->mapping. These are * the pages which were not successfully invalidated in * truncate_complete_page(). We try to drop those buffers here * and if that worked, and the page is no longer mapped into * process address space (page_count == 1) it can be freed. * Otherwise, leave the page on the LRU so it is swappable. */ if (PagePrivate(page)) { if (!try_to_release_page(page, sc->gfp_mask)) goto activate_locked; if (!mapping && page_count(page) == 1) goto free_it; } |
28e4d965e [PATCH] mm: remov... |
561 |
if (!mapping || !remove_mapping(mapping, page)) |
49d2e9cc4 [PATCH] Swap Migr... |
562 |
goto keep_locked; |
1da177e4c Linux-2.6.12-rc2 |
563 564 565 |
free_it: unlock_page(page); |
05ff51376 [PATCH] vmscan re... |
566 |
nr_reclaimed++; |
1da177e4c Linux-2.6.12-rc2 |
567 568 569 570 571 572 573 574 575 576 577 |
if (!pagevec_add(&freed_pvec, page)) __pagevec_release_nonlru(&freed_pvec); continue; activate_locked: SetPageActive(page); pgactivate++; keep_locked: unlock_page(page); keep: list_add(&page->lru, &ret_pages); |
725d704ec [PATCH] mm: VM_BU... |
578 |
VM_BUG_ON(PageLRU(page)); |
1da177e4c Linux-2.6.12-rc2 |
579 580 581 582 |
} list_splice(&ret_pages, page_list); if (pagevec_count(&freed_pvec)) __pagevec_release_nonlru(&freed_pvec); |
f8891e5e1 [PATCH] Light wei... |
583 |
count_vm_events(PGACTIVATE, pgactivate); |
05ff51376 [PATCH] vmscan re... |
584 |
return nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
585 |
} |
49d2e9cc4 [PATCH] Swap Migr... |
586 |
/* |
1da177e4c Linux-2.6.12-rc2 |
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 |
* zone->lru_lock is heavily contended. Some of the functions that * shrink the lists perform better by taking out a batch of pages * and working on them outside the LRU lock. * * For pagecache intensive workloads, this function is the hottest * spot in the kernel (apart from copy_*_user functions). * * Appropriate locks must be held before calling this function. * * @nr_to_scan: The number of pages to look through on the list. * @src: The LRU list to pull pages off. * @dst: The temp list to put pages on to. * @scanned: The number of pages that were scanned. * * returns how many pages were moved onto *@dst. */ |
69e05944a [PATCH] vmscan: u... |
603 604 605 |
static unsigned long isolate_lru_pages(unsigned long nr_to_scan, struct list_head *src, struct list_head *dst, unsigned long *scanned) |
1da177e4c Linux-2.6.12-rc2 |
606 |
{ |
69e05944a [PATCH] vmscan: u... |
607 |
unsigned long nr_taken = 0; |
1da177e4c Linux-2.6.12-rc2 |
608 |
struct page *page; |
c9b02d970 [PATCH] mm: isola... |
609 |
unsigned long scan; |
1da177e4c Linux-2.6.12-rc2 |
610 |
|
c9b02d970 [PATCH] mm: isola... |
611 |
for (scan = 0; scan < nr_to_scan && !list_empty(src); scan++) { |
7c8ee9a86 [PATCH] mm: simpl... |
612 |
struct list_head *target; |
1da177e4c Linux-2.6.12-rc2 |
613 614 |
page = lru_to_page(src); prefetchw_prev_lru_page(page, src, flags); |
725d704ec [PATCH] mm: VM_BU... |
615 |
VM_BUG_ON(!PageLRU(page)); |
8d438f96d [PATCH] mm: PageL... |
616 |
|
053837fce [PATCH] mm: migra... |
617 |
list_del(&page->lru); |
7c8ee9a86 [PATCH] mm: simpl... |
618 619 |
target = src; if (likely(get_page_unless_zero(page))) { |
053837fce [PATCH] mm: migra... |
620 |
/* |
7c8ee9a86 [PATCH] mm: simpl... |
621 622 623 |
* Be careful not to clear PageLRU until after we're * sure the page is not being freed elsewhere -- the * page release code relies on it. |
053837fce [PATCH] mm: migra... |
624 |
*/ |
7c8ee9a86 [PATCH] mm: simpl... |
625 626 627 628 |
ClearPageLRU(page); target = dst; nr_taken++; } /* else it is being freed elsewhere */ |
46453a6e1 [PATCH] mm: never... |
629 |
|
7c8ee9a86 [PATCH] mm: simpl... |
630 |
list_add(&page->lru, target); |
1da177e4c Linux-2.6.12-rc2 |
631 632 633 634 635 636 637 |
} *scanned = scan; return nr_taken; } /* |
1742f19fa [PATCH] vmscan: r... |
638 639 |
* shrink_inactive_list() is a helper for shrink_zone(). It returns the number * of reclaimed pages |
1da177e4c Linux-2.6.12-rc2 |
640 |
*/ |
1742f19fa [PATCH] vmscan: r... |
641 642 |
static unsigned long shrink_inactive_list(unsigned long max_scan, struct zone *zone, struct scan_control *sc) |
1da177e4c Linux-2.6.12-rc2 |
643 644 645 |
{ LIST_HEAD(page_list); struct pagevec pvec; |
69e05944a [PATCH] vmscan: u... |
646 |
unsigned long nr_scanned = 0; |
05ff51376 [PATCH] vmscan re... |
647 |
unsigned long nr_reclaimed = 0; |
1da177e4c Linux-2.6.12-rc2 |
648 649 650 651 652 |
pagevec_init(&pvec, 1); lru_add_drain(); spin_lock_irq(&zone->lru_lock); |
69e05944a [PATCH] vmscan: u... |
653 |
do { |
1da177e4c Linux-2.6.12-rc2 |
654 |
struct page *page; |
69e05944a [PATCH] vmscan: u... |
655 656 657 |
unsigned long nr_taken; unsigned long nr_scan; unsigned long nr_freed; |
1da177e4c Linux-2.6.12-rc2 |
658 659 660 661 662 663 664 |
nr_taken = isolate_lru_pages(sc->swap_cluster_max, &zone->inactive_list, &page_list, &nr_scan); zone->nr_inactive -= nr_taken; zone->pages_scanned += nr_scan; spin_unlock_irq(&zone->lru_lock); |
69e05944a [PATCH] vmscan: u... |
665 |
nr_scanned += nr_scan; |
1742f19fa [PATCH] vmscan: r... |
666 |
nr_freed = shrink_page_list(&page_list, sc); |
05ff51376 [PATCH] vmscan re... |
667 |
nr_reclaimed += nr_freed; |
a74609faf [PATCH] mm: page_... |
668 669 |
local_irq_disable(); if (current_is_kswapd()) { |
f8891e5e1 [PATCH] Light wei... |
670 671 |
__count_zone_vm_events(PGSCAN_KSWAPD, zone, nr_scan); __count_vm_events(KSWAPD_STEAL, nr_freed); |
a74609faf [PATCH] mm: page_... |
672 |
} else |
f8891e5e1 [PATCH] Light wei... |
673 674 |
__count_zone_vm_events(PGSCAN_DIRECT, zone, nr_scan); __count_vm_events(PGACTIVATE, nr_freed); |
a74609faf [PATCH] mm: page_... |
675 |
|
fb8d14e17 [PATCH] mm: shrin... |
676 677 |
if (nr_taken == 0) goto done; |
a74609faf [PATCH] mm: page_... |
678 |
spin_lock(&zone->lru_lock); |
1da177e4c Linux-2.6.12-rc2 |
679 680 681 682 683 |
/* * Put back any unfreeable pages. */ while (!list_empty(&page_list)) { page = lru_to_page(&page_list); |
725d704ec [PATCH] mm: VM_BU... |
684 |
VM_BUG_ON(PageLRU(page)); |
8d438f96d [PATCH] mm: PageL... |
685 |
SetPageLRU(page); |
1da177e4c Linux-2.6.12-rc2 |
686 687 688 689 690 691 692 693 694 695 696 |
list_del(&page->lru); if (PageActive(page)) add_page_to_active_list(zone, page); else add_page_to_inactive_list(zone, page); if (!pagevec_add(&pvec, page)) { spin_unlock_irq(&zone->lru_lock); __pagevec_release(&pvec); spin_lock_irq(&zone->lru_lock); } } |
69e05944a [PATCH] vmscan: u... |
697 |
} while (nr_scanned < max_scan); |
fb8d14e17 [PATCH] mm: shrin... |
698 |
spin_unlock(&zone->lru_lock); |
1da177e4c Linux-2.6.12-rc2 |
699 |
done: |
fb8d14e17 [PATCH] mm: shrin... |
700 |
local_irq_enable(); |
1da177e4c Linux-2.6.12-rc2 |
701 |
pagevec_release(&pvec); |
05ff51376 [PATCH] vmscan re... |
702 |
return nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
703 |
} |
4ff1ffb48 [PATCH] oom: recl... |
704 705 706 707 |
static inline int zone_is_near_oom(struct zone *zone) { return zone->pages_scanned >= (zone->nr_active + zone->nr_inactive)*3; } |
1da177e4c Linux-2.6.12-rc2 |
708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 |
/* * This moves pages from the active list to the inactive list. * * We move them the other way if the page is referenced by one or more * processes, from rmap. * * If the pages are mostly unmapped, the processing is fast and it is * appropriate to hold zone->lru_lock across the whole operation. But if * the pages are mapped, the processing is slow (page_referenced()) so we * should drop zone->lru_lock around each page. It's impossible to balance * this, so instead we remove the pages from the LRU while processing them. * It is safe to rely on PG_active against the non-LRU pages in here because * nobody will play with that bit on a non-LRU page. * * The downside is that we have to touch page->_count against each page. * But we had to alter page->flags anyway. */ |
1742f19fa [PATCH] vmscan: r... |
725 726 |
static void shrink_active_list(unsigned long nr_pages, struct zone *zone, struct scan_control *sc) |
1da177e4c Linux-2.6.12-rc2 |
727 |
{ |
69e05944a [PATCH] vmscan: u... |
728 |
unsigned long pgmoved; |
1da177e4c Linux-2.6.12-rc2 |
729 |
int pgdeactivate = 0; |
69e05944a [PATCH] vmscan: u... |
730 |
unsigned long pgscanned; |
1da177e4c Linux-2.6.12-rc2 |
731 732 733 734 735 736 |
LIST_HEAD(l_hold); /* The pages which were snipped off */ LIST_HEAD(l_inactive); /* Pages to go onto the inactive_list */ LIST_HEAD(l_active); /* Pages to go onto the active_list */ struct page *page; struct pagevec pvec; int reclaim_mapped = 0; |
2903fb169 [PATCH] vmscan: s... |
737 |
|
6e5ef1a96 [PATCH] vmscan: e... |
738 |
if (sc->may_swap) { |
2903fb169 [PATCH] vmscan: s... |
739 740 741 |
long mapped_ratio; long distress; long swap_tendency; |
4ff1ffb48 [PATCH] oom: recl... |
742 743 |
if (zone_is_near_oom(zone)) goto force_reclaim_mapped; |
2903fb169 [PATCH] vmscan: s... |
744 745 746 747 748 749 750 751 752 753 754 755 |
/* * `distress' is a measure of how much trouble we're having * reclaiming pages. 0 -> no problems. 100 -> great trouble. */ distress = 100 >> zone->prev_priority; /* * The point of this algorithm is to decide when to start * reclaiming mapped memory instead of just pagecache. Work out * how much memory * is mapped. */ |
f3dbd3446 [PATCH] zoned vm ... |
756 757 |
mapped_ratio = ((global_page_state(NR_FILE_MAPPED) + global_page_state(NR_ANON_PAGES)) * 100) / |
bf02cf4b6 [PATCH] zoned vm ... |
758 |
vm_total_pages; |
2903fb169 [PATCH] vmscan: s... |
759 760 761 762 763 764 765 766 767 768 769 770 771 |
/* * Now decide how much we really want to unmap some pages. The * mapped ratio is downgraded - just because there's a lot of * mapped memory doesn't necessarily mean that page reclaim * isn't succeeding. * * The distress ratio is important - we don't want to start * going oom. * * A 100% value of vm_swappiness overrides this algorithm * altogether. */ |
d6277db4a [PATCH] swsusp: r... |
772 |
swap_tendency = mapped_ratio / 2 + distress + sc->swappiness; |
2903fb169 [PATCH] vmscan: s... |
773 774 775 776 777 778 |
/* * Now use this metric to decide whether to start moving mapped * memory onto the inactive list. */ if (swap_tendency >= 100) |
4ff1ffb48 [PATCH] oom: recl... |
779 |
force_reclaim_mapped: |
2903fb169 [PATCH] vmscan: s... |
780 781 |
reclaim_mapped = 1; } |
1da177e4c Linux-2.6.12-rc2 |
782 783 784 785 786 787 788 789 |
lru_add_drain(); spin_lock_irq(&zone->lru_lock); pgmoved = isolate_lru_pages(nr_pages, &zone->active_list, &l_hold, &pgscanned); zone->pages_scanned += pgscanned; zone->nr_active -= pgmoved; spin_unlock_irq(&zone->lru_lock); |
1da177e4c Linux-2.6.12-rc2 |
790 791 792 793 794 795 796 |
while (!list_empty(&l_hold)) { cond_resched(); page = lru_to_page(&l_hold); list_del(&page->lru); if (page_mapped(page)) { if (!reclaim_mapped || (total_swap_pages == 0 && PageAnon(page)) || |
f7b7fd8f3 [PATCH] temporari... |
797 |
page_referenced(page, 0)) { |
1da177e4c Linux-2.6.12-rc2 |
798 799 800 801 802 803 804 805 806 807 808 809 810 |
list_add(&page->lru, &l_active); continue; } } list_add(&page->lru, &l_inactive); } pagevec_init(&pvec, 1); pgmoved = 0; spin_lock_irq(&zone->lru_lock); while (!list_empty(&l_inactive)) { page = lru_to_page(&l_inactive); prefetchw_prev_lru_page(page, &l_inactive, flags); |
725d704ec [PATCH] mm: VM_BU... |
811 |
VM_BUG_ON(PageLRU(page)); |
8d438f96d [PATCH] mm: PageL... |
812 |
SetPageLRU(page); |
725d704ec [PATCH] mm: VM_BU... |
813 |
VM_BUG_ON(!PageActive(page)); |
4c84cacfa [PATCH] mm: PageA... |
814 |
ClearPageActive(page); |
1da177e4c Linux-2.6.12-rc2 |
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 |
list_move(&page->lru, &zone->inactive_list); pgmoved++; if (!pagevec_add(&pvec, page)) { zone->nr_inactive += pgmoved; spin_unlock_irq(&zone->lru_lock); pgdeactivate += pgmoved; pgmoved = 0; if (buffer_heads_over_limit) pagevec_strip(&pvec); __pagevec_release(&pvec); spin_lock_irq(&zone->lru_lock); } } zone->nr_inactive += pgmoved; pgdeactivate += pgmoved; if (buffer_heads_over_limit) { spin_unlock_irq(&zone->lru_lock); pagevec_strip(&pvec); spin_lock_irq(&zone->lru_lock); } pgmoved = 0; while (!list_empty(&l_active)) { page = lru_to_page(&l_active); prefetchw_prev_lru_page(page, &l_active, flags); |
725d704ec [PATCH] mm: VM_BU... |
840 |
VM_BUG_ON(PageLRU(page)); |
8d438f96d [PATCH] mm: PageL... |
841 |
SetPageLRU(page); |
725d704ec [PATCH] mm: VM_BU... |
842 |
VM_BUG_ON(!PageActive(page)); |
1da177e4c Linux-2.6.12-rc2 |
843 844 845 846 847 848 849 850 851 852 853 |
list_move(&page->lru, &zone->active_list); pgmoved++; if (!pagevec_add(&pvec, page)) { zone->nr_active += pgmoved; pgmoved = 0; spin_unlock_irq(&zone->lru_lock); __pagevec_release(&pvec); spin_lock_irq(&zone->lru_lock); } } zone->nr_active += pgmoved; |
a74609faf [PATCH] mm: page_... |
854 |
|
f8891e5e1 [PATCH] Light wei... |
855 856 857 |
__count_zone_vm_events(PGREFILL, zone, pgscanned); __count_vm_events(PGDEACTIVATE, pgdeactivate); spin_unlock_irq(&zone->lru_lock); |
1da177e4c Linux-2.6.12-rc2 |
858 |
|
a74609faf [PATCH] mm: page_... |
859 |
pagevec_release(&pvec); |
1da177e4c Linux-2.6.12-rc2 |
860 861 862 863 864 |
} /* * This is a basic per-zone page freer. Used by both kswapd and direct reclaim. */ |
05ff51376 [PATCH] vmscan re... |
865 866 |
static unsigned long shrink_zone(int priority, struct zone *zone, struct scan_control *sc) |
1da177e4c Linux-2.6.12-rc2 |
867 868 869 |
{ unsigned long nr_active; unsigned long nr_inactive; |
8695949a1 [PATCH] Thin out ... |
870 |
unsigned long nr_to_scan; |
05ff51376 [PATCH] vmscan re... |
871 |
unsigned long nr_reclaimed = 0; |
1da177e4c Linux-2.6.12-rc2 |
872 |
|
53e9a6159 [PATCH] VM: zone ... |
873 |
atomic_inc(&zone->reclaim_in_progress); |
1da177e4c Linux-2.6.12-rc2 |
874 875 876 877 |
/* * Add one to `nr_to_scan' just to make sure that the kernel will * slowly sift through the active list. */ |
8695949a1 [PATCH] Thin out ... |
878 |
zone->nr_scan_active += (zone->nr_active >> priority) + 1; |
1da177e4c Linux-2.6.12-rc2 |
879 880 881 882 883 |
nr_active = zone->nr_scan_active; if (nr_active >= sc->swap_cluster_max) zone->nr_scan_active = 0; else nr_active = 0; |
8695949a1 [PATCH] Thin out ... |
884 |
zone->nr_scan_inactive += (zone->nr_inactive >> priority) + 1; |
1da177e4c Linux-2.6.12-rc2 |
885 886 887 888 889 |
nr_inactive = zone->nr_scan_inactive; if (nr_inactive >= sc->swap_cluster_max) zone->nr_scan_inactive = 0; else nr_inactive = 0; |
1da177e4c Linux-2.6.12-rc2 |
890 891 |
while (nr_active || nr_inactive) { if (nr_active) { |
8695949a1 [PATCH] Thin out ... |
892 |
nr_to_scan = min(nr_active, |
1da177e4c Linux-2.6.12-rc2 |
893 |
(unsigned long)sc->swap_cluster_max); |
8695949a1 [PATCH] Thin out ... |
894 |
nr_active -= nr_to_scan; |
1742f19fa [PATCH] vmscan: r... |
895 |
shrink_active_list(nr_to_scan, zone, sc); |
1da177e4c Linux-2.6.12-rc2 |
896 897 898 |
} if (nr_inactive) { |
8695949a1 [PATCH] Thin out ... |
899 |
nr_to_scan = min(nr_inactive, |
1da177e4c Linux-2.6.12-rc2 |
900 |
(unsigned long)sc->swap_cluster_max); |
8695949a1 [PATCH] Thin out ... |
901 |
nr_inactive -= nr_to_scan; |
1742f19fa [PATCH] vmscan: r... |
902 903 |
nr_reclaimed += shrink_inactive_list(nr_to_scan, zone, sc); |
1da177e4c Linux-2.6.12-rc2 |
904 905 906 907 |
} } throttle_vm_writeout(); |
53e9a6159 [PATCH] VM: zone ... |
908 909 |
atomic_dec(&zone->reclaim_in_progress); |
05ff51376 [PATCH] vmscan re... |
910 |
return nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 |
} /* * This is the direct reclaim path, for page-allocating processes. We only * try to reclaim pages from zones which will satisfy the caller's allocation * request. * * We reclaim from a zone even if that zone is over pages_high. Because: * a) The caller may be trying to free *extra* pages to satisfy a higher-order * allocation or * b) The zones may be over pages_high but they must go *over* pages_high to * satisfy the `incremental min' zone defense algorithm. * * Returns the number of reclaimed pages. * * If a zone is deemed to be full of pinned pages then just give it a light * scan then give up on it. */ |
1742f19fa [PATCH] vmscan: r... |
929 |
static unsigned long shrink_zones(int priority, struct zone **zones, |
05ff51376 [PATCH] vmscan re... |
930 |
struct scan_control *sc) |
1da177e4c Linux-2.6.12-rc2 |
931 |
{ |
05ff51376 [PATCH] vmscan re... |
932 |
unsigned long nr_reclaimed = 0; |
1da177e4c Linux-2.6.12-rc2 |
933 |
int i; |
408d85441 [PATCH] oom: use ... |
934 |
sc->all_unreclaimable = 1; |
1da177e4c Linux-2.6.12-rc2 |
935 936 |
for (i = 0; zones[i] != NULL; i++) { struct zone *zone = zones[i]; |
f3fe65122 [PATCH] mm: add p... |
937 |
if (!populated_zone(zone)) |
1da177e4c Linux-2.6.12-rc2 |
938 |
continue; |
9bf2229f8 [PATCH] cpusets: ... |
939 |
if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) |
1da177e4c Linux-2.6.12-rc2 |
940 |
continue; |
8695949a1 [PATCH] Thin out ... |
941 942 943 |
zone->temp_priority = priority; if (zone->prev_priority > priority) zone->prev_priority = priority; |
1da177e4c Linux-2.6.12-rc2 |
944 |
|
8695949a1 [PATCH] Thin out ... |
945 |
if (zone->all_unreclaimable && priority != DEF_PRIORITY) |
1da177e4c Linux-2.6.12-rc2 |
946 |
continue; /* Let kswapd poll it */ |
408d85441 [PATCH] oom: use ... |
947 |
sc->all_unreclaimable = 0; |
05ff51376 [PATCH] vmscan re... |
948 |
nr_reclaimed += shrink_zone(priority, zone, sc); |
1da177e4c Linux-2.6.12-rc2 |
949 |
} |
05ff51376 [PATCH] vmscan re... |
950 |
return nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 |
} /* * This is the main entry point to direct page reclaim. * * If a full scan of the inactive list fails to free enough memory then we * are "out of memory" and something needs to be killed. * * If the caller is !__GFP_FS then the probability of a failure is reasonably * high - the zone may be full of dirty or under-writeback pages, which this * caller can't do much about. We kick pdflush and take explicit naps in the * hope that some of these pages can be written. But if the allocating task * holds filesystem locks which prevent writeout this might not work, and the * allocation attempt will fail. */ |
69e05944a [PATCH] vmscan: u... |
966 |
unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
967 968 969 |
{ int priority; int ret = 0; |
69e05944a [PATCH] vmscan: u... |
970 |
unsigned long total_scanned = 0; |
05ff51376 [PATCH] vmscan re... |
971 |
unsigned long nr_reclaimed = 0; |
1da177e4c Linux-2.6.12-rc2 |
972 |
struct reclaim_state *reclaim_state = current->reclaim_state; |
1da177e4c Linux-2.6.12-rc2 |
973 974 |
unsigned long lru_pages = 0; int i; |
179e96395 [PATCH] vmscan: s... |
975 976 977 978 979 |
struct scan_control sc = { .gfp_mask = gfp_mask, .may_writepage = !laptop_mode, .swap_cluster_max = SWAP_CLUSTER_MAX, .may_swap = 1, |
d6277db4a [PATCH] swsusp: r... |
980 |
.swappiness = vm_swappiness, |
179e96395 [PATCH] vmscan: s... |
981 |
}; |
1da177e4c Linux-2.6.12-rc2 |
982 |
|
f8891e5e1 [PATCH] Light wei... |
983 |
count_vm_event(ALLOCSTALL); |
1da177e4c Linux-2.6.12-rc2 |
984 985 986 |
for (i = 0; zones[i] != NULL; i++) { struct zone *zone = zones[i]; |
9bf2229f8 [PATCH] cpusets: ... |
987 |
if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) |
1da177e4c Linux-2.6.12-rc2 |
988 989 990 991 992 993 994 |
continue; zone->temp_priority = DEF_PRIORITY; lru_pages += zone->nr_active + zone->nr_inactive; } for (priority = DEF_PRIORITY; priority >= 0; priority--) { |
1da177e4c Linux-2.6.12-rc2 |
995 |
sc.nr_scanned = 0; |
f7b7fd8f3 [PATCH] temporari... |
996 997 |
if (!priority) disable_swap_token(); |
1742f19fa [PATCH] vmscan: r... |
998 |
nr_reclaimed += shrink_zones(priority, zones, &sc); |
1da177e4c Linux-2.6.12-rc2 |
999 1000 |
shrink_slab(sc.nr_scanned, gfp_mask, lru_pages); if (reclaim_state) { |
05ff51376 [PATCH] vmscan re... |
1001 |
nr_reclaimed += reclaim_state->reclaimed_slab; |
1da177e4c Linux-2.6.12-rc2 |
1002 1003 1004 |
reclaim_state->reclaimed_slab = 0; } total_scanned += sc.nr_scanned; |
05ff51376 [PATCH] vmscan re... |
1005 |
if (nr_reclaimed >= sc.swap_cluster_max) { |
1da177e4c Linux-2.6.12-rc2 |
1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 |
ret = 1; goto out; } /* * Try to write back as many pages as we just scanned. This * tends to cause slow streaming writers to write data to the * disk smoothly, at the dirtying rate, which is nice. But * that's undesirable in laptop mode, where we *want* lumpy * writeout. So in laptop mode, write out the whole world. */ |
179e96395 [PATCH] vmscan: s... |
1017 1018 |
if (total_scanned > sc.swap_cluster_max + sc.swap_cluster_max / 2) { |
687a21cee [PATCH] rename wa... |
1019 |
wakeup_pdflush(laptop_mode ? 0 : total_scanned); |
1da177e4c Linux-2.6.12-rc2 |
1020 1021 1022 1023 1024 1025 1026 |
sc.may_writepage = 1; } /* Take a nap, wait for some writeback to complete */ if (sc.nr_scanned && priority < DEF_PRIORITY - 2) blk_congestion_wait(WRITE, HZ/10); } |
408d85441 [PATCH] oom: use ... |
1027 1028 1029 |
/* top priority shrink_caches still had more to do? don't OOM, then */ if (!sc.all_unreclaimable) ret = 1; |
1da177e4c Linux-2.6.12-rc2 |
1030 1031 1032 |
out: for (i = 0; zones[i] != 0; i++) { struct zone *zone = zones[i]; |
9bf2229f8 [PATCH] cpusets: ... |
1033 |
if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) |
1da177e4c Linux-2.6.12-rc2 |
1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 |
continue; zone->prev_priority = zone->temp_priority; } return ret; } /* * For kswapd, balance_pgdat() will work across all this node's zones until * they are all at pages_high. * |
1da177e4c Linux-2.6.12-rc2 |
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 |
* Returns the number of pages which were actually freed. * * There is special handling here for zones which are full of pinned pages. * This can happen if the pages are all mlocked, or if they are all used by * device drivers (say, ZONE_DMA). Or if they are all in use by hugetlb. * What we do is to detect the case where all pages in the zone have been * scanned twice and there has been zero successful reclaim. Mark the zone as * dead and from now on, only perform a short scan. Basically we're polling * the zone for when the problem goes away. * * kswapd scans the zones in the highmem->normal->dma direction. It skips * zones which have free_pages > pages_high, but once a zone is found to have * free_pages <= pages_high, we scan that zone and the lower zones regardless * of the number of free pages in the lower zones. This interoperates with * the page allocator fallback scheme to ensure that aging of pages is balanced * across the zones. */ |
d6277db4a [PATCH] swsusp: r... |
1062 |
static unsigned long balance_pgdat(pg_data_t *pgdat, int order) |
1da177e4c Linux-2.6.12-rc2 |
1063 |
{ |
1da177e4c Linux-2.6.12-rc2 |
1064 1065 1066 |
int all_zones_ok; int priority; int i; |
69e05944a [PATCH] vmscan: u... |
1067 |
unsigned long total_scanned; |
05ff51376 [PATCH] vmscan re... |
1068 |
unsigned long nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
1069 |
struct reclaim_state *reclaim_state = current->reclaim_state; |
179e96395 [PATCH] vmscan: s... |
1070 1071 1072 |
struct scan_control sc = { .gfp_mask = GFP_KERNEL, .may_swap = 1, |
d6277db4a [PATCH] swsusp: r... |
1073 1074 |
.swap_cluster_max = SWAP_CLUSTER_MAX, .swappiness = vm_swappiness, |
179e96395 [PATCH] vmscan: s... |
1075 |
}; |
1da177e4c Linux-2.6.12-rc2 |
1076 1077 1078 |
loop_again: total_scanned = 0; |
05ff51376 [PATCH] vmscan re... |
1079 |
nr_reclaimed = 0; |
c0bbbc73d [PATCH] typo in v... |
1080 |
sc.may_writepage = !laptop_mode; |
f8891e5e1 [PATCH] Light wei... |
1081 |
count_vm_event(PAGEOUTRUN); |
1da177e4c Linux-2.6.12-rc2 |
1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 |
for (i = 0; i < pgdat->nr_zones; i++) { struct zone *zone = pgdat->node_zones + i; zone->temp_priority = DEF_PRIORITY; } for (priority = DEF_PRIORITY; priority >= 0; priority--) { int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ unsigned long lru_pages = 0; |
f7b7fd8f3 [PATCH] temporari... |
1092 1093 1094 |
/* The swap token gets in the way of swapout... */ if (!priority) disable_swap_token(); |
1da177e4c Linux-2.6.12-rc2 |
1095 |
all_zones_ok = 1; |
d6277db4a [PATCH] swsusp: r... |
1096 1097 1098 1099 1100 1101 |
/* * Scan in the highmem->dma direction for the highest * zone which needs scanning */ for (i = pgdat->nr_zones - 1; i >= 0; i--) { struct zone *zone = pgdat->node_zones + i; |
1da177e4c Linux-2.6.12-rc2 |
1102 |
|
d6277db4a [PATCH] swsusp: r... |
1103 1104 |
if (!populated_zone(zone)) continue; |
1da177e4c Linux-2.6.12-rc2 |
1105 |
|
d6277db4a [PATCH] swsusp: r... |
1106 1107 |
if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; |
1da177e4c Linux-2.6.12-rc2 |
1108 |
|
d6277db4a [PATCH] swsusp: r... |
1109 1110 1111 1112 |
if (!zone_watermark_ok(zone, order, zone->pages_high, 0, 0)) { end_zone = i; goto scan; |
1da177e4c Linux-2.6.12-rc2 |
1113 |
} |
1da177e4c Linux-2.6.12-rc2 |
1114 |
} |
d6277db4a [PATCH] swsusp: r... |
1115 |
goto out; |
1da177e4c Linux-2.6.12-rc2 |
1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 |
scan: for (i = 0; i <= end_zone; i++) { struct zone *zone = pgdat->node_zones + i; lru_pages += zone->nr_active + zone->nr_inactive; } /* * Now scan the zone in the dma->highmem direction, stopping * at the last zone which needs scanning. * * We do this because the page allocator works in the opposite * direction. This prevents the page allocator from allocating * pages behind kswapd's direction of progress, which would * cause too much scanning of the lower zones. */ for (i = 0; i <= end_zone; i++) { struct zone *zone = pgdat->node_zones + i; |
b15e0905f [PATCH] vmscan: n... |
1134 |
int nr_slab; |
1da177e4c Linux-2.6.12-rc2 |
1135 |
|
f3fe65122 [PATCH] mm: add p... |
1136 |
if (!populated_zone(zone)) |
1da177e4c Linux-2.6.12-rc2 |
1137 1138 1139 1140 |
continue; if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; |
d6277db4a [PATCH] swsusp: r... |
1141 1142 1143 |
if (!zone_watermark_ok(zone, order, zone->pages_high, end_zone, 0)) all_zones_ok = 0; |
1da177e4c Linux-2.6.12-rc2 |
1144 1145 1146 1147 |
zone->temp_priority = priority; if (zone->prev_priority > priority) zone->prev_priority = priority; sc.nr_scanned = 0; |
05ff51376 [PATCH] vmscan re... |
1148 |
nr_reclaimed += shrink_zone(priority, zone, &sc); |
1da177e4c Linux-2.6.12-rc2 |
1149 |
reclaim_state->reclaimed_slab = 0; |
b15e0905f [PATCH] vmscan: n... |
1150 1151 |
nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL, lru_pages); |
05ff51376 [PATCH] vmscan re... |
1152 |
nr_reclaimed += reclaim_state->reclaimed_slab; |
1da177e4c Linux-2.6.12-rc2 |
1153 1154 1155 |
total_scanned += sc.nr_scanned; if (zone->all_unreclaimable) continue; |
b15e0905f [PATCH] vmscan: n... |
1156 |
if (nr_slab == 0 && zone->pages_scanned >= |
4ff1ffb48 [PATCH] oom: recl... |
1157 |
(zone->nr_active + zone->nr_inactive) * 6) |
1da177e4c Linux-2.6.12-rc2 |
1158 1159 1160 1161 1162 1163 1164 |
zone->all_unreclaimable = 1; /* * If we've done a decent amount of scanning and * the reclaim ratio is low, start doing writepage * even in laptop mode */ if (total_scanned > SWAP_CLUSTER_MAX * 2 && |
05ff51376 [PATCH] vmscan re... |
1165 |
total_scanned > nr_reclaimed + nr_reclaimed / 2) |
1da177e4c Linux-2.6.12-rc2 |
1166 1167 |
sc.may_writepage = 1; } |
1da177e4c Linux-2.6.12-rc2 |
1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 |
if (all_zones_ok) break; /* kswapd: all done */ /* * OK, kswapd is getting into trouble. Take a nap, then take * another pass across the zones. */ if (total_scanned && priority < DEF_PRIORITY - 2) blk_congestion_wait(WRITE, HZ/10); /* * We do this so kswapd doesn't build up large priorities for * example when it is freeing in parallel with allocators. It * matches the direct reclaim path behaviour in terms of impact * on zone->*_priority. */ |
d6277db4a [PATCH] swsusp: r... |
1183 |
if (nr_reclaimed >= SWAP_CLUSTER_MAX) |
1da177e4c Linux-2.6.12-rc2 |
1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 |
break; } out: for (i = 0; i < pgdat->nr_zones; i++) { struct zone *zone = pgdat->node_zones + i; zone->prev_priority = zone->temp_priority; } if (!all_zones_ok) { cond_resched(); goto loop_again; } |
05ff51376 [PATCH] vmscan re... |
1196 |
return nr_reclaimed; |
1da177e4c Linux-2.6.12-rc2 |
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 |
} /* * The background pageout daemon, started as a kernel thread * from the init process. * * This basically trickles out pages so that we have _some_ * free memory available even if there is no other activity * that frees anything up. This is needed for things like routing * etc, where we otherwise might have all activity going on in * asynchronous contexts that cannot page things out. * * If there are applications that are active memory-allocators * (most normal use), this basically shouldn't matter. */ static int kswapd(void *p) { unsigned long order; pg_data_t *pgdat = (pg_data_t*)p; struct task_struct *tsk = current; DEFINE_WAIT(wait); struct reclaim_state reclaim_state = { .reclaimed_slab = 0, }; cpumask_t cpumask; |
1da177e4c Linux-2.6.12-rc2 |
1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 |
cpumask = node_to_cpumask(pgdat->node_id); if (!cpus_empty(cpumask)) set_cpus_allowed(tsk, cpumask); current->reclaim_state = &reclaim_state; /* * Tell the memory management that we're a "memory allocator", * and that if we need more memory we should get access to it * regardless (see "__alloc_pages()"). "kswapd" should * never get caught in the normal page freeing logic. * * (Kswapd normally doesn't need memory anyway, but sometimes * you need a small amount of memory in order to be able to * page out something else, and this flag essentially protects * us from recursively trying to free more memory as we're * trying to free the first piece of memory in the first place). */ |
930d91525 [PATCH] Swap Migr... |
1239 |
tsk->flags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD; |
1da177e4c Linux-2.6.12-rc2 |
1240 1241 1242 1243 |
order = 0; for ( ; ; ) { unsigned long new_order; |
3e1d1d28d [PATCH] Cleanup p... |
1244 1245 |
try_to_freeze(); |
1da177e4c Linux-2.6.12-rc2 |
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 |
prepare_to_wait(&pgdat->kswapd_wait, &wait, TASK_INTERRUPTIBLE); new_order = pgdat->kswapd_max_order; pgdat->kswapd_max_order = 0; if (order < new_order) { /* * Don't sleep if someone wants a larger 'order' * allocation */ order = new_order; } else { schedule(); order = pgdat->kswapd_max_order; } finish_wait(&pgdat->kswapd_wait, &wait); |
d6277db4a [PATCH] swsusp: r... |
1261 |
balance_pgdat(pgdat, order); |
1da177e4c Linux-2.6.12-rc2 |
1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 |
} return 0; } /* * A zone is low on free memory, so wake its kswapd task to service it. */ void wakeup_kswapd(struct zone *zone, int order) { pg_data_t *pgdat; |
f3fe65122 [PATCH] mm: add p... |
1272 |
if (!populated_zone(zone)) |
1da177e4c Linux-2.6.12-rc2 |
1273 1274 1275 |
return; pgdat = zone->zone_pgdat; |
7fb1d9fca [PATCH] mm: __all... |
1276 |
if (zone_watermark_ok(zone, order, zone->pages_low, 0, 0)) |
1da177e4c Linux-2.6.12-rc2 |
1277 1278 1279 |
return; if (pgdat->kswapd_max_order < order) pgdat->kswapd_max_order = order; |
9bf2229f8 [PATCH] cpusets: ... |
1280 |
if (!cpuset_zone_allowed(zone, __GFP_HARDWALL)) |
1da177e4c Linux-2.6.12-rc2 |
1281 |
return; |
8d0986e28 [PATCH] vm: kswap... |
1282 |
if (!waitqueue_active(&pgdat->kswapd_wait)) |
1da177e4c Linux-2.6.12-rc2 |
1283 |
return; |
8d0986e28 [PATCH] vm: kswap... |
1284 |
wake_up_interruptible(&pgdat->kswapd_wait); |
1da177e4c Linux-2.6.12-rc2 |
1285 1286 1287 1288 |
} #ifdef CONFIG_PM /* |
d6277db4a [PATCH] swsusp: r... |
1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 |
* Helper function for shrink_all_memory(). Tries to reclaim 'nr_pages' pages * from LRU lists system-wide, for given pass and priority, and returns the * number of reclaimed pages * * For pass > 3 we also try to shrink the LRU lists that contain a few pages */ static unsigned long shrink_all_zones(unsigned long nr_pages, int pass, int prio, struct scan_control *sc) { struct zone *zone; unsigned long nr_to_scan, ret = 0; for_each_zone(zone) { if (!populated_zone(zone)) continue; if (zone->all_unreclaimable && prio != DEF_PRIORITY) continue; /* For pass = 0 we don't shrink the active list */ if (pass > 0) { zone->nr_scan_active += (zone->nr_active >> prio) + 1; if (zone->nr_scan_active >= nr_pages || pass > 3) { zone->nr_scan_active = 0; nr_to_scan = min(nr_pages, zone->nr_active); shrink_active_list(nr_to_scan, zone, sc); } } zone->nr_scan_inactive += (zone->nr_inactive >> prio) + 1; if (zone->nr_scan_inactive >= nr_pages || pass > 3) { zone->nr_scan_inactive = 0; nr_to_scan = min(nr_pages, zone->nr_inactive); ret += shrink_inactive_list(nr_to_scan, zone, sc); if (ret >= nr_pages) return ret; } } return ret; } /* * Try to free `nr_pages' of memory, system-wide, and return the number of * freed pages. * * Rather than trying to age LRUs the aim is to preserve the overall * LRU order by reclaiming preferentially * inactive > active > active referenced > active mapped |
1da177e4c Linux-2.6.12-rc2 |
1339 |
*/ |
69e05944a [PATCH] vmscan: u... |
1340 |
unsigned long shrink_all_memory(unsigned long nr_pages) |
1da177e4c Linux-2.6.12-rc2 |
1341 |
{ |
d6277db4a [PATCH] swsusp: r... |
1342 |
unsigned long lru_pages, nr_slab; |
69e05944a [PATCH] vmscan: u... |
1343 |
unsigned long ret = 0; |
d6277db4a [PATCH] swsusp: r... |
1344 1345 1346 1347 1348 1349 1350 1351 1352 |
int pass; struct reclaim_state reclaim_state; struct zone *zone; struct scan_control sc = { .gfp_mask = GFP_KERNEL, .may_swap = 0, .swap_cluster_max = nr_pages, .may_writepage = 1, .swappiness = vm_swappiness, |
1da177e4c Linux-2.6.12-rc2 |
1353 1354 1355 |
}; current->reclaim_state = &reclaim_state; |
69e05944a [PATCH] vmscan: u... |
1356 |
|
d6277db4a [PATCH] swsusp: r... |
1357 1358 1359 |
lru_pages = 0; for_each_zone(zone) lru_pages += zone->nr_active + zone->nr_inactive; |
972d1a7b1 [PATCH] ZVC: Supp... |
1360 |
nr_slab = global_page_state(NR_SLAB_RECLAIMABLE); |
d6277db4a [PATCH] swsusp: r... |
1361 1362 1363 1364 1365 |
/* If slab caches are huge, it's better to hit them first */ while (nr_slab >= lru_pages) { reclaim_state.reclaimed_slab = 0; shrink_slab(nr_pages, sc.gfp_mask, lru_pages); if (!reclaim_state.reclaimed_slab) |
1da177e4c Linux-2.6.12-rc2 |
1366 |
break; |
d6277db4a [PATCH] swsusp: r... |
1367 1368 1369 1370 1371 1372 |
ret += reclaim_state.reclaimed_slab; if (ret >= nr_pages) goto out; nr_slab -= reclaim_state.reclaimed_slab; |
1da177e4c Linux-2.6.12-rc2 |
1373 |
} |
d6277db4a [PATCH] swsusp: r... |
1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 |
/* * We try to shrink LRUs in 5 passes: * 0 = Reclaim from inactive_list only * 1 = Reclaim from active list but don't reclaim mapped * 2 = 2nd pass of type 1 * 3 = Reclaim mapped (normal reclaim) * 4 = 2nd pass of type 3 */ for (pass = 0; pass < 5; pass++) { int prio; /* Needed for shrinking slab caches later on */ if (!lru_pages) for_each_zone(zone) { lru_pages += zone->nr_active; lru_pages += zone->nr_inactive; } /* Force reclaiming mapped pages in the passes #3 and #4 */ if (pass > 2) { sc.may_swap = 1; sc.swappiness = 100; } for (prio = DEF_PRIORITY; prio >= 0; prio--) { unsigned long nr_to_scan = nr_pages - ret; |
d6277db4a [PATCH] swsusp: r... |
1401 |
sc.nr_scanned = 0; |
d6277db4a [PATCH] swsusp: r... |
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 |
ret += shrink_all_zones(nr_to_scan, prio, pass, &sc); if (ret >= nr_pages) goto out; reclaim_state.reclaimed_slab = 0; shrink_slab(sc.nr_scanned, sc.gfp_mask, lru_pages); ret += reclaim_state.reclaimed_slab; if (ret >= nr_pages) goto out; if (sc.nr_scanned && prio < DEF_PRIORITY - 2) blk_congestion_wait(WRITE, HZ / 10); } lru_pages = 0; |
248a0301e [PATCH] mm: make ... |
1417 |
} |
d6277db4a [PATCH] swsusp: r... |
1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 |
/* * If ret = 0, we could not shrink LRUs, but there may be something * in slab caches */ if (!ret) do { reclaim_state.reclaimed_slab = 0; shrink_slab(nr_pages, sc.gfp_mask, lru_pages); ret += reclaim_state.reclaimed_slab; } while (ret < nr_pages && reclaim_state.reclaimed_slab > 0); out: |
1da177e4c Linux-2.6.12-rc2 |
1431 |
current->reclaim_state = NULL; |
d6277db4a [PATCH] swsusp: r... |
1432 |
|
1da177e4c Linux-2.6.12-rc2 |
1433 1434 1435 1436 1437 1438 1439 1440 1441 |
return ret; } #endif #ifdef CONFIG_HOTPLUG_CPU /* It's optimal to keep kswapds on the same CPUs as their memory, but not required for correctness. So if the last cpu in a node goes away, we get changed to run anywhere: as the first one comes back, restore their cpu bindings. */ |
9c7b216d2 [PATCH] cpu hotpl... |
1442 |
static int __devinit cpu_callback(struct notifier_block *nfb, |
69e05944a [PATCH] vmscan: u... |
1443 |
unsigned long action, void *hcpu) |
1da177e4c Linux-2.6.12-rc2 |
1444 1445 1446 1447 1448 |
{ pg_data_t *pgdat; cpumask_t mask; if (action == CPU_ONLINE) { |
ec936fc56 [PATCH] for_each_... |
1449 |
for_each_online_pgdat(pgdat) { |
1da177e4c Linux-2.6.12-rc2 |
1450 1451 1452 1453 1454 1455 1456 1457 1458 |
mask = node_to_cpumask(pgdat->node_id); if (any_online_cpu(mask) != NR_CPUS) /* One of our CPUs online: restore mask */ set_cpus_allowed(pgdat->kswapd, mask); } } return NOTIFY_OK; } #endif /* CONFIG_HOTPLUG_CPU */ |
3218ae14b [PATCH] pgdat all... |
1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 |
/* * This kswapd start function will be called by init and node-hot-add. * On node-hot-add, kswapd will moved to proper cpus if cpus are hot-added. */ int kswapd_run(int nid) { pg_data_t *pgdat = NODE_DATA(nid); int ret = 0; if (pgdat->kswapd) return 0; pgdat->kswapd = kthread_run(kswapd, pgdat, "kswapd%d", nid); if (IS_ERR(pgdat->kswapd)) { /* failure at boot is fatal */ BUG_ON(system_state == SYSTEM_BOOTING); printk("Failed to start kswapd on node %d ",nid); ret = -1; } return ret; } |
1da177e4c Linux-2.6.12-rc2 |
1481 1482 |
static int __init kswapd_init(void) { |
3218ae14b [PATCH] pgdat all... |
1483 |
int nid; |
69e05944a [PATCH] vmscan: u... |
1484 |
|
1da177e4c Linux-2.6.12-rc2 |
1485 |
swap_setup(); |
3218ae14b [PATCH] pgdat all... |
1486 1487 |
for_each_online_node(nid) kswapd_run(nid); |
1da177e4c Linux-2.6.12-rc2 |
1488 1489 1490 1491 1492 |
hotcpu_notifier(cpu_callback, 0); return 0; } module_init(kswapd_init) |
9eeff2395 [PATCH] Zone recl... |
1493 1494 1495 1496 1497 1498 1499 |
#ifdef CONFIG_NUMA /* * Zone reclaim mode * * If non-zero call zone_reclaim when the number of free pages falls below * the watermarks. |
9eeff2395 [PATCH] Zone recl... |
1500 1501 |
*/ int zone_reclaim_mode __read_mostly; |
1b2ffb789 [PATCH] Zone recl... |
1502 1503 1504 1505 |
#define RECLAIM_OFF 0 #define RECLAIM_ZONE (1<<0) /* Run shrink_cache on the zone */ #define RECLAIM_WRITE (1<<1) /* Writeout pages during reclaim */ #define RECLAIM_SWAP (1<<2) /* Swap pages out during reclaim */ |
9eeff2395 [PATCH] Zone recl... |
1506 |
/* |
a92f71263 [PATCH] zone_recl... |
1507 1508 1509 1510 1511 |
* Priority for ZONE_RECLAIM. This determines the fraction of pages * of a node considered for each zone_reclaim. 4 scans 1/16th of * a zone. */ #define ZONE_RECLAIM_PRIORITY 4 |
9eeff2395 [PATCH] Zone recl... |
1512 |
/* |
9614634fe [PATCH] ZVC/zone_... |
1513 1514 1515 1516 1517 1518 |
* Percentage of pages in a zone that must be unmapped for zone_reclaim to * occur. */ int sysctl_min_unmapped_ratio = 1; /* |
0ff38490c [PATCH] zone_recl... |
1519 1520 1521 1522 1523 1524 |
* If the number of slab pages in a zone grows beyond this percentage then * slab reclaim needs to occur. */ int sysctl_min_slab_ratio = 5; /* |
9eeff2395 [PATCH] Zone recl... |
1525 1526 |
* Try to free up some pages from this zone through reclaim. */ |
179e96395 [PATCH] vmscan: s... |
1527 |
static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) |
9eeff2395 [PATCH] Zone recl... |
1528 |
{ |
7fb2d46d3 [PATCH] zone_recl... |
1529 |
/* Minimum pages needed in order to stay on node */ |
69e05944a [PATCH] vmscan: u... |
1530 |
const unsigned long nr_pages = 1 << order; |
9eeff2395 [PATCH] Zone recl... |
1531 1532 |
struct task_struct *p = current; struct reclaim_state reclaim_state; |
8695949a1 [PATCH] Thin out ... |
1533 |
int priority; |
05ff51376 [PATCH] vmscan re... |
1534 |
unsigned long nr_reclaimed = 0; |
179e96395 [PATCH] vmscan: s... |
1535 1536 1537 |
struct scan_control sc = { .may_writepage = !!(zone_reclaim_mode & RECLAIM_WRITE), .may_swap = !!(zone_reclaim_mode & RECLAIM_SWAP), |
69e05944a [PATCH] vmscan: u... |
1538 1539 |
.swap_cluster_max = max_t(unsigned long, nr_pages, SWAP_CLUSTER_MAX), |
179e96395 [PATCH] vmscan: s... |
1540 |
.gfp_mask = gfp_mask, |
d6277db4a [PATCH] swsusp: r... |
1541 |
.swappiness = vm_swappiness, |
179e96395 [PATCH] vmscan: s... |
1542 |
}; |
83e33a471 [PATCH] zone recl... |
1543 |
unsigned long slab_reclaimable; |
9eeff2395 [PATCH] Zone recl... |
1544 1545 |
disable_swap_token(); |
9eeff2395 [PATCH] Zone recl... |
1546 |
cond_resched(); |
d4f7796e9 [PATCH] vmscan: f... |
1547 1548 1549 1550 1551 1552 |
/* * We need to be able to allocate from the reserves for RECLAIM_SWAP * and we also need to be able to write out pages for RECLAIM_WRITE * and RECLAIM_SWAP. */ p->flags |= PF_MEMALLOC | PF_SWAPWRITE; |
9eeff2395 [PATCH] Zone recl... |
1553 1554 |
reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; |
c84db23c6 [PATCH] zone_recl... |
1555 |
|
0ff38490c [PATCH] zone_recl... |
1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 |
if (zone_page_state(zone, NR_FILE_PAGES) - zone_page_state(zone, NR_FILE_MAPPED) > zone->min_unmapped_pages) { /* * Free memory by calling shrink zone with increasing * priorities until we have enough memory freed. */ priority = ZONE_RECLAIM_PRIORITY; do { nr_reclaimed += shrink_zone(priority, zone, &sc); priority--; } while (priority >= 0 && nr_reclaimed < nr_pages); } |
c84db23c6 [PATCH] zone_recl... |
1569 |
|
83e33a471 [PATCH] zone recl... |
1570 1571 |
slab_reclaimable = zone_page_state(zone, NR_SLAB_RECLAIMABLE); if (slab_reclaimable > zone->min_slab_pages) { |
2a16e3f4b [PATCH] Reclaim s... |
1572 |
/* |
7fb2d46d3 [PATCH] zone_recl... |
1573 |
* shrink_slab() does not currently allow us to determine how |
0ff38490c [PATCH] zone_recl... |
1574 1575 1576 1577 |
* many pages were freed in this zone. So we take the current * number of slab pages and shake the slab until it is reduced * by the same nr_pages that we used for reclaiming unmapped * pages. |
2a16e3f4b [PATCH] Reclaim s... |
1578 |
* |
0ff38490c [PATCH] zone_recl... |
1579 1580 |
* Note that shrink_slab will free memory on all zones and may * take a long time. |
2a16e3f4b [PATCH] Reclaim s... |
1581 |
*/ |
0ff38490c [PATCH] zone_recl... |
1582 |
while (shrink_slab(sc.nr_scanned, gfp_mask, order) && |
83e33a471 [PATCH] zone recl... |
1583 1584 |
zone_page_state(zone, NR_SLAB_RECLAIMABLE) > slab_reclaimable - nr_pages) |
0ff38490c [PATCH] zone_recl... |
1585 |
; |
83e33a471 [PATCH] zone recl... |
1586 1587 1588 1589 1590 1591 1592 |
/* * Update nr_reclaimed by the number of slab pages we * reclaimed from this zone. */ nr_reclaimed += slab_reclaimable - zone_page_state(zone, NR_SLAB_RECLAIMABLE); |
2a16e3f4b [PATCH] Reclaim s... |
1593 |
} |
9eeff2395 [PATCH] Zone recl... |
1594 |
p->reclaim_state = NULL; |
d4f7796e9 [PATCH] vmscan: f... |
1595 |
current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE); |
05ff51376 [PATCH] vmscan re... |
1596 |
return nr_reclaimed >= nr_pages; |
9eeff2395 [PATCH] Zone recl... |
1597 |
} |
179e96395 [PATCH] vmscan: s... |
1598 1599 1600 1601 1602 1603 1604 |
int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) { cpumask_t mask; int node_id; /* |
0ff38490c [PATCH] zone_recl... |
1605 1606 |
* Zone reclaim reclaims unmapped file backed pages and * slab pages if we are over the defined limits. |
34aa1330f [PATCH] zoned vm ... |
1607 |
* |
9614634fe [PATCH] ZVC/zone_... |
1608 1609 1610 1611 1612 |
* A small portion of unmapped file backed pages is needed for * file I/O otherwise pages read by file I/O will be immediately * thrown out if the zone is overallocated. So we do not reclaim * if less than a specified percentage of the zone is used by * unmapped file backed pages. |
179e96395 [PATCH] vmscan: s... |
1613 |
*/ |
34aa1330f [PATCH] zoned vm ... |
1614 |
if (zone_page_state(zone, NR_FILE_PAGES) - |
0ff38490c [PATCH] zone_recl... |
1615 1616 1617 |
zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages && zone_page_state(zone, NR_SLAB_RECLAIMABLE) <= zone->min_slab_pages) |
9614634fe [PATCH] ZVC/zone_... |
1618 |
return 0; |
179e96395 [PATCH] vmscan: s... |
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 |
/* * Avoid concurrent zone reclaims, do not reclaim in a zone that does * not have reclaimable pages and if we should not delay the allocation * then do not scan. */ if (!(gfp_mask & __GFP_WAIT) || zone->all_unreclaimable || atomic_read(&zone->reclaim_in_progress) > 0 || (current->flags & PF_MEMALLOC)) return 0; /* * Only run zone reclaim on the local zone or on zones that do not * have associated processors. This will favor the local processor * over remote processors and spread off node memory allocations * as wide as possible. */ |
89fa30242 [PATCH] NUMA: Add... |
1637 |
node_id = zone_to_nid(zone); |
179e96395 [PATCH] vmscan: s... |
1638 1639 1640 1641 1642 |
mask = node_to_cpumask(node_id); if (!cpus_empty(mask) && node_id != numa_node_id()) return 0; return __zone_reclaim(zone, gfp_mask, order); } |
9eeff2395 [PATCH] Zone recl... |
1643 |
#endif |