Blame view
mm/swapfile.c
77.7 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 |
/* * linux/mm/swapfile.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds * Swap reorganised 29.12.95, Stephen Tweedie */ |
1da177e4c Linux-2.6.12-rc2 |
7 8 9 10 11 12 13 14 15 |
#include <linux/mm.h> #include <linux/hugetlb.h> #include <linux/mman.h> #include <linux/slab.h> #include <linux/kernel_stat.h> #include <linux/swap.h> #include <linux/vmalloc.h> #include <linux/pagemap.h> #include <linux/namei.h> |
072441e21 mm: move shmem pr... |
16 |
#include <linux/shmem_fs.h> |
1da177e4c Linux-2.6.12-rc2 |
17 |
#include <linux/blkdev.h> |
20137a490 swapfile: swapon ... |
18 |
#include <linux/random.h> |
1da177e4c Linux-2.6.12-rc2 |
19 20 21 22 |
#include <linux/writeback.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/init.h> |
5ad646880 ksm: let shared p... |
23 |
#include <linux/ksm.h> |
1da177e4c Linux-2.6.12-rc2 |
24 25 26 |
#include <linux/rmap.h> #include <linux/security.h> #include <linux/backing-dev.h> |
fc0abb145 [PATCH] sem2mutex... |
27 |
#include <linux/mutex.h> |
c59ede7b7 [PATCH] move capa... |
28 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
29 |
#include <linux/syscalls.h> |
8a9f3ccd2 Memory controller... |
30 |
#include <linux/memcontrol.h> |
66d7dd518 /proc/swaps: supp... |
31 |
#include <linux/poll.h> |
72788c385 oom: replace PF_O... |
32 |
#include <linux/oom.h> |
38b5faf4b mm: frontswap: co... |
33 34 |
#include <linux/frontswap.h> #include <linux/swapfile.h> |
f981c5950 mm: methods for t... |
35 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
36 37 38 39 |
#include <asm/pgtable.h> #include <asm/tlbflush.h> #include <linux/swapops.h> |
5d1ea48bd mm: page_cgroup: ... |
40 |
#include <linux/swap_cgroup.h> |
1da177e4c Linux-2.6.12-rc2 |
41 |
|
570a335b8 swap_info: swap c... |
42 43 44 |
static bool swap_count_continued(struct swap_info_struct *, pgoff_t, unsigned char); static void free_swap_count_continuations(struct swap_info_struct *); |
d4906e1aa swap: rework map_... |
45 |
static sector_t map_swap_entry(swp_entry_t, struct block_device**); |
570a335b8 swap_info: swap c... |
46 |
|
38b5faf4b mm: frontswap: co... |
47 |
DEFINE_SPINLOCK(swap_lock); |
7c363b8c6 mm/swapfile.c: ma... |
48 |
static unsigned int nr_swapfiles; |
ec8acf20a swap: add per-par... |
49 50 |
atomic_long_t nr_swap_pages; /* protected with swap_lock. reading in vm_swap_full() doesn't need lock */ |
1da177e4c Linux-2.6.12-rc2 |
51 |
long total_swap_pages; |
78ecba081 mm: fix ever-decr... |
52 |
static int least_priority; |
1da177e4c Linux-2.6.12-rc2 |
53 |
|
1da177e4c Linux-2.6.12-rc2 |
54 55 56 57 |
static const char Bad_file[] = "Bad swap file entry "; static const char Unused_file[] = "Unused swap file entry "; static const char Bad_offset[] = "Bad swap offset entry "; static const char Unused_offset[] = "Unused swap offset entry "; |
adfab836f swap: change swap... |
58 59 60 61 |
/* * all active swap_info_structs * protected with swap_lock, and ordered by priority. */ |
18ab4d4ce swap: change swap... |
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 |
PLIST_HEAD(swap_active_head); /* * all available (active, not full) swap_info_structs * protected with swap_avail_lock, ordered by priority. * This is used by get_swap_page() instead of swap_active_head * because swap_active_head includes all swap_info_structs, * but get_swap_page() doesn't need to look at full ones. * This uses its own lock instead of swap_lock because when a * swap_info_struct changes between not-full/full, it needs to * add/remove itself to/from this list, but the swap_info_struct->lock * is held and the locking order requires swap_lock to be taken * before any swap_info_struct->lock. */ static PLIST_HEAD(swap_avail_head); static DEFINE_SPINLOCK(swap_avail_lock); |
1da177e4c Linux-2.6.12-rc2 |
78 |
|
38b5faf4b mm: frontswap: co... |
79 |
struct swap_info_struct *swap_info[MAX_SWAPFILES]; |
1da177e4c Linux-2.6.12-rc2 |
80 |
|
fc0abb145 [PATCH] sem2mutex... |
81 |
static DEFINE_MUTEX(swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
82 |
|
66d7dd518 /proc/swaps: supp... |
83 84 85 |
static DECLARE_WAIT_QUEUE_HEAD(proc_poll_wait); /* Activity counter to indicate that a swapon or swapoff has occurred */ static atomic_t proc_poll_event = ATOMIC_INIT(0); |
8d69aaee8 swap_info: swap_m... |
86 |
static inline unsigned char swap_count(unsigned char ent) |
355cfa73d mm: modify swap_m... |
87 |
{ |
570a335b8 swap_info: swap c... |
88 |
return ent & ~SWAP_HAS_CACHE; /* may include SWAP_HAS_CONT flag */ |
355cfa73d mm: modify swap_m... |
89 |
} |
efa90a981 swap_info: change... |
90 |
/* returns 1 if swap entry is freed */ |
c9e444103 mm: reuse unused ... |
91 92 93 |
static int __try_to_reclaim_swap(struct swap_info_struct *si, unsigned long offset) { |
efa90a981 swap_info: change... |
94 |
swp_entry_t entry = swp_entry(si->type, offset); |
c9e444103 mm: reuse unused ... |
95 96 |
struct page *page; int ret = 0; |
33806f06d swap: make each s... |
97 |
page = find_get_page(swap_address_space(entry), entry.val); |
c9e444103 mm: reuse unused ... |
98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 |
if (!page) return 0; /* * This function is called from scan_swap_map() and it's called * by vmscan.c at reclaiming pages. So, we hold a lock on a page, here. * We have to use trylock for avoiding deadlock. This is a special * case and you should use try_to_free_swap() with explicit lock_page() * in usual operations. */ if (trylock_page(page)) { ret = try_to_free_swap(page); unlock_page(page); } page_cache_release(page); return ret; } |
355cfa73d mm: modify swap_m... |
114 |
|
1da177e4c Linux-2.6.12-rc2 |
115 |
/* |
6a6ba8317 swapfile: swapon ... |
116 117 118 119 120 121 |
* swapon tell device that all the old swap contents can be discarded, * to allow the swap device to optimize its wear-levelling. */ static int discard_swap(struct swap_info_struct *si) { struct swap_extent *se; |
9625a5f28 swap_info: includ... |
122 123 |
sector_t start_block; sector_t nr_blocks; |
6a6ba8317 swapfile: swapon ... |
124 |
int err = 0; |
9625a5f28 swap_info: includ... |
125 126 127 128 129 130 |
/* Do not discard the swap header page! */ se = &si->first_swap_extent; start_block = (se->start_block + 1) << (PAGE_SHIFT - 9); nr_blocks = ((sector_t)se->nr_pages - 1) << (PAGE_SHIFT - 9); if (nr_blocks) { err = blkdev_issue_discard(si->bdev, start_block, |
dd3932edd block: remove BLK... |
131 |
nr_blocks, GFP_KERNEL, 0); |
9625a5f28 swap_info: includ... |
132 133 134 135 |
if (err) return err; cond_resched(); } |
6a6ba8317 swapfile: swapon ... |
136 |
|
9625a5f28 swap_info: includ... |
137 138 139 |
list_for_each_entry(se, &si->first_swap_extent.list, list) { start_block = se->start_block << (PAGE_SHIFT - 9); nr_blocks = (sector_t)se->nr_pages << (PAGE_SHIFT - 9); |
6a6ba8317 swapfile: swapon ... |
140 141 |
err = blkdev_issue_discard(si->bdev, start_block, |
dd3932edd block: remove BLK... |
142 |
nr_blocks, GFP_KERNEL, 0); |
6a6ba8317 swapfile: swapon ... |
143 144 145 146 147 148 149 |
if (err) break; cond_resched(); } return err; /* That will often be -EOPNOTSUPP */ } |
7992fde72 swapfile: swap al... |
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 |
/* * swap allocation tell device that a cluster of swap can now be discarded, * to allow the swap device to optimize its wear-levelling. */ static void discard_swap_cluster(struct swap_info_struct *si, pgoff_t start_page, pgoff_t nr_pages) { struct swap_extent *se = si->curr_swap_extent; int found_extent = 0; while (nr_pages) { struct list_head *lh; if (se->start_page <= start_page && start_page < se->start_page + se->nr_pages) { pgoff_t offset = start_page - se->start_page; sector_t start_block = se->start_block + offset; |
858a29900 swapfile: change ... |
167 |
sector_t nr_blocks = se->nr_pages - offset; |
7992fde72 swapfile: swap al... |
168 169 170 171 172 173 174 175 176 177 178 179 |
if (nr_blocks > nr_pages) nr_blocks = nr_pages; start_page += nr_blocks; nr_pages -= nr_blocks; if (!found_extent++) si->curr_swap_extent = se; start_block <<= PAGE_SHIFT - 9; nr_blocks <<= PAGE_SHIFT - 9; if (blkdev_issue_discard(si->bdev, start_block, |
dd3932edd block: remove BLK... |
180 |
nr_blocks, GFP_NOIO, 0)) |
7992fde72 swapfile: swap al... |
181 182 183 184 |
break; } lh = se->list.next; |
7992fde72 swapfile: swap al... |
185 186 187 |
se = list_entry(lh, struct swap_extent, list); } } |
048c27fd7 [PATCH] swap: sca... |
188 189 |
#define SWAPFILE_CLUSTER 256 #define LATENCY_LIMIT 256 |
2a8f94493 swap: change bloc... |
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
static inline void cluster_set_flag(struct swap_cluster_info *info, unsigned int flag) { info->flags = flag; } static inline unsigned int cluster_count(struct swap_cluster_info *info) { return info->data; } static inline void cluster_set_count(struct swap_cluster_info *info, unsigned int c) { info->data = c; } static inline void cluster_set_count_flag(struct swap_cluster_info *info, unsigned int c, unsigned int f) { info->flags = f; info->data = c; } static inline unsigned int cluster_next(struct swap_cluster_info *info) { return info->data; } static inline void cluster_set_next(struct swap_cluster_info *info, unsigned int n) { info->data = n; } static inline void cluster_set_next_flag(struct swap_cluster_info *info, unsigned int n, unsigned int f) { info->flags = f; info->data = n; } static inline bool cluster_is_free(struct swap_cluster_info *info) { return info->flags & CLUSTER_FLAG_FREE; } static inline bool cluster_is_null(struct swap_cluster_info *info) { return info->flags & CLUSTER_FLAG_NEXT_NULL; } static inline void cluster_set_null(struct swap_cluster_info *info) { info->flags = CLUSTER_FLAG_NEXT_NULL; info->data = 0; } |
815c2c543 swap: make swap d... |
247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 |
/* Add a cluster to discard list and schedule it to do discard */ static void swap_cluster_schedule_discard(struct swap_info_struct *si, unsigned int idx) { /* * If scan_swap_map() can't find a free cluster, it will check * si->swap_map directly. To make sure the discarding cluster isn't * taken by scan_swap_map(), mark the swap entries bad (occupied). It * will be cleared after discard */ memset(si->swap_map + idx * SWAPFILE_CLUSTER, SWAP_MAP_BAD, SWAPFILE_CLUSTER); if (cluster_is_null(&si->discard_cluster_head)) { cluster_set_next_flag(&si->discard_cluster_head, idx, 0); cluster_set_next_flag(&si->discard_cluster_tail, idx, 0); } else { unsigned int tail = cluster_next(&si->discard_cluster_tail); cluster_set_next(&si->cluster_info[tail], idx); cluster_set_next_flag(&si->discard_cluster_tail, idx, 0); } schedule_work(&si->discard_work); } /* * Doing discard actually. After a cluster discard is finished, the cluster * will be added to free cluster list. caller should hold si->lock. */ static void swap_do_scheduled_discard(struct swap_info_struct *si) { struct swap_cluster_info *info; unsigned int idx; info = si->cluster_info; while (!cluster_is_null(&si->discard_cluster_head)) { idx = cluster_next(&si->discard_cluster_head); cluster_set_next_flag(&si->discard_cluster_head, cluster_next(&info[idx]), 0); if (cluster_next(&si->discard_cluster_tail) == idx) { cluster_set_null(&si->discard_cluster_head); cluster_set_null(&si->discard_cluster_tail); } spin_unlock(&si->lock); discard_swap_cluster(si, idx * SWAPFILE_CLUSTER, SWAPFILE_CLUSTER); spin_lock(&si->lock); cluster_set_flag(&info[idx], CLUSTER_FLAG_FREE); if (cluster_is_null(&si->free_cluster_head)) { cluster_set_next_flag(&si->free_cluster_head, idx, 0); cluster_set_next_flag(&si->free_cluster_tail, idx, 0); } else { unsigned int tail; tail = cluster_next(&si->free_cluster_tail); cluster_set_next(&info[tail], idx); cluster_set_next_flag(&si->free_cluster_tail, idx, 0); } memset(si->swap_map + idx * SWAPFILE_CLUSTER, 0, SWAPFILE_CLUSTER); } } static void swap_discard_work(struct work_struct *work) { struct swap_info_struct *si; si = container_of(work, struct swap_info_struct, discard_work); spin_lock(&si->lock); swap_do_scheduled_discard(si); spin_unlock(&si->lock); } |
2a8f94493 swap: change bloc... |
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 |
/* * The cluster corresponding to page_nr will be used. The cluster will be * removed from free cluster list and its usage counter will be increased. */ static void inc_cluster_info_page(struct swap_info_struct *p, struct swap_cluster_info *cluster_info, unsigned long page_nr) { unsigned long idx = page_nr / SWAPFILE_CLUSTER; if (!cluster_info) return; if (cluster_is_free(&cluster_info[idx])) { VM_BUG_ON(cluster_next(&p->free_cluster_head) != idx); cluster_set_next_flag(&p->free_cluster_head, cluster_next(&cluster_info[idx]), 0); if (cluster_next(&p->free_cluster_tail) == idx) { cluster_set_null(&p->free_cluster_tail); cluster_set_null(&p->free_cluster_head); } cluster_set_count_flag(&cluster_info[idx], 0, 0); } VM_BUG_ON(cluster_count(&cluster_info[idx]) >= SWAPFILE_CLUSTER); cluster_set_count(&cluster_info[idx], cluster_count(&cluster_info[idx]) + 1); } /* * The cluster corresponding to page_nr decreases one usage. If the usage * counter becomes 0, which means no page in the cluster is in using, we can * optionally discard the cluster and add it to free cluster list. */ static void dec_cluster_info_page(struct swap_info_struct *p, struct swap_cluster_info *cluster_info, unsigned long page_nr) { unsigned long idx = page_nr / SWAPFILE_CLUSTER; if (!cluster_info) return; VM_BUG_ON(cluster_count(&cluster_info[idx]) == 0); cluster_set_count(&cluster_info[idx], cluster_count(&cluster_info[idx]) - 1); if (cluster_count(&cluster_info[idx]) == 0) { |
815c2c543 swap: make swap d... |
375 376 377 378 379 |
/* * If the swap is discardable, prepare discard the cluster * instead of free it immediately. The cluster will be freed * after discard. */ |
edfe23dac swap: fix races e... |
380 381 |
if ((p->flags & (SWP_WRITEOK | SWP_PAGE_DISCARD)) == (SWP_WRITEOK | SWP_PAGE_DISCARD)) { |
815c2c543 swap: make swap d... |
382 383 384 |
swap_cluster_schedule_discard(p, idx); return; } |
2a8f94493 swap: change bloc... |
385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 |
cluster_set_flag(&cluster_info[idx], CLUSTER_FLAG_FREE); if (cluster_is_null(&p->free_cluster_head)) { cluster_set_next_flag(&p->free_cluster_head, idx, 0); cluster_set_next_flag(&p->free_cluster_tail, idx, 0); } else { unsigned int tail = cluster_next(&p->free_cluster_tail); cluster_set_next(&cluster_info[tail], idx); cluster_set_next_flag(&p->free_cluster_tail, idx, 0); } } } /* * It's possible scan_swap_map() uses a free cluster in the middle of free * cluster list. Avoiding such abuse to avoid list corruption. */ |
ebc2a1a69 swap: make cluste... |
401 402 |
static bool scan_swap_map_ssd_cluster_conflict(struct swap_info_struct *si, |
2a8f94493 swap: change bloc... |
403 404 |
unsigned long offset) { |
ebc2a1a69 swap: make cluste... |
405 406 |
struct percpu_cluster *percpu_cluster; bool conflict; |
2a8f94493 swap: change bloc... |
407 |
offset /= SWAPFILE_CLUSTER; |
ebc2a1a69 swap: make cluste... |
408 |
conflict = !cluster_is_null(&si->free_cluster_head) && |
2a8f94493 swap: change bloc... |
409 410 |
offset != cluster_next(&si->free_cluster_head) && cluster_is_free(&si->cluster_info[offset]); |
ebc2a1a69 swap: make cluste... |
411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 |
if (!conflict) return false; percpu_cluster = this_cpu_ptr(si->percpu_cluster); cluster_set_null(&percpu_cluster->index); return true; } /* * Try to get a swap entry from current cpu's swap entry pool (a cluster). This * might involve allocating a new cluster for current CPU too. */ static void scan_swap_map_try_ssd_cluster(struct swap_info_struct *si, unsigned long *offset, unsigned long *scan_base) { struct percpu_cluster *cluster; bool found_free; unsigned long tmp; new_cluster: cluster = this_cpu_ptr(si->percpu_cluster); if (cluster_is_null(&cluster->index)) { if (!cluster_is_null(&si->free_cluster_head)) { cluster->index = si->free_cluster_head; cluster->next = cluster_next(&cluster->index) * SWAPFILE_CLUSTER; } else if (!cluster_is_null(&si->discard_cluster_head)) { /* * we don't have free cluster but have some clusters in * discarding, do discard now and reclaim them */ swap_do_scheduled_discard(si); *scan_base = *offset = si->cluster_next; goto new_cluster; } else return; } found_free = false; /* * Other CPUs can use our cluster if they can't find a free cluster, * check if there is still free entry in the cluster */ tmp = cluster->next; while (tmp < si->max && tmp < (cluster_next(&cluster->index) + 1) * SWAPFILE_CLUSTER) { if (!si->swap_map[tmp]) { found_free = true; break; } tmp++; } if (!found_free) { cluster_set_null(&cluster->index); goto new_cluster; } cluster->next = tmp + 1; *offset = tmp; *scan_base = tmp; |
2a8f94493 swap: change bloc... |
472 |
} |
24b8ff7c2 mm: remove inline... |
473 474 |
static unsigned long scan_swap_map(struct swap_info_struct *si, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
475 |
{ |
ebebbbe90 swapfile: rearran... |
476 |
unsigned long offset; |
c60aa176c swapfile: swap al... |
477 |
unsigned long scan_base; |
7992fde72 swapfile: swap al... |
478 |
unsigned long last_in_cluster = 0; |
048c27fd7 [PATCH] swap: sca... |
479 |
int latency_ration = LATENCY_LIMIT; |
7dfad4183 [PATCH] swap: sca... |
480 |
|
886bb7e9c swapfile: remove ... |
481 |
/* |
7dfad4183 [PATCH] swap: sca... |
482 483 484 485 486 487 488 |
* We try to cluster swap pages by allocating them sequentially * in swap. Once we've allocated SWAPFILE_CLUSTER pages this * way, however, we resort to first-free allocation, starting * a new cluster. This prevents us from scattering swap pages * all over the entire swap partition, so that we reduce * overall disk seek times between swap pages. -- sct * But we do now try to find an empty cluster. -Andrea |
c60aa176c swapfile: swap al... |
489 |
* And we let swap pages go all over an SSD partition. Hugh |
7dfad4183 [PATCH] swap: sca... |
490 |
*/ |
52b7efdbe [PATCH] swap: sca... |
491 |
si->flags += SWP_SCANNING; |
c60aa176c swapfile: swap al... |
492 |
scan_base = offset = si->cluster_next; |
ebebbbe90 swapfile: rearran... |
493 |
|
ebc2a1a69 swap: make cluste... |
494 495 496 497 498 |
/* SSD algorithm */ if (si->cluster_info) { scan_swap_map_try_ssd_cluster(si, &offset, &scan_base); goto checks; } |
ebebbbe90 swapfile: rearran... |
499 500 501 502 503 |
if (unlikely(!si->cluster_nr--)) { if (si->pages - si->inuse_pages < SWAPFILE_CLUSTER) { si->cluster_nr = SWAPFILE_CLUSTER - 1; goto checks; } |
2a8f94493 swap: change bloc... |
504 |
|
ec8acf20a swap: add per-par... |
505 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
506 |
|
c60aa176c swapfile: swap al... |
507 508 509 |
/* * If seek is expensive, start searching for new cluster from * start of partition, to minimize the span of allocated swap. |
50088c440 mm/swapfile.c: de... |
510 511 |
* If seek is cheap, that is the SWP_SOLIDSTATE si->cluster_info * case, just handled by scan_swap_map_try_ssd_cluster() above. |
c60aa176c swapfile: swap al... |
512 |
*/ |
50088c440 mm/swapfile.c: de... |
513 |
scan_base = offset = si->lowest_bit; |
7dfad4183 [PATCH] swap: sca... |
514 515 516 517 |
last_in_cluster = offset + SWAPFILE_CLUSTER - 1; /* Locate the first empty (unaligned) cluster */ for (; last_in_cluster <= si->highest_bit; offset++) { |
1da177e4c Linux-2.6.12-rc2 |
518 |
if (si->swap_map[offset]) |
7dfad4183 [PATCH] swap: sca... |
519 520 |
last_in_cluster = offset + SWAPFILE_CLUSTER; else if (offset == last_in_cluster) { |
ec8acf20a swap: add per-par... |
521 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
522 523 524 |
offset -= SWAPFILE_CLUSTER - 1; si->cluster_next = offset; si->cluster_nr = SWAPFILE_CLUSTER - 1; |
c60aa176c swapfile: swap al... |
525 526 527 528 529 530 531 532 533 |
goto checks; } if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } } offset = scan_base; |
ec8acf20a swap: add per-par... |
534 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
535 |
si->cluster_nr = SWAPFILE_CLUSTER - 1; |
1da177e4c Linux-2.6.12-rc2 |
536 |
} |
7dfad4183 [PATCH] swap: sca... |
537 |
|
ebebbbe90 swapfile: rearran... |
538 |
checks: |
ebc2a1a69 swap: make cluste... |
539 540 541 542 |
if (si->cluster_info) { while (scan_swap_map_ssd_cluster_conflict(si, offset)) scan_swap_map_try_ssd_cluster(si, &offset, &scan_base); } |
ebebbbe90 swapfile: rearran... |
543 |
if (!(si->flags & SWP_WRITEOK)) |
52b7efdbe [PATCH] swap: sca... |
544 |
goto no_page; |
7dfad4183 [PATCH] swap: sca... |
545 546 |
if (!si->highest_bit) goto no_page; |
ebebbbe90 swapfile: rearran... |
547 |
if (offset > si->highest_bit) |
c60aa176c swapfile: swap al... |
548 |
scan_base = offset = si->lowest_bit; |
c9e444103 mm: reuse unused ... |
549 |
|
b73d7fcec swap: prevent reu... |
550 551 |
/* reuse swap entry of cache-only swap if not busy. */ if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
c9e444103 mm: reuse unused ... |
552 |
int swap_was_freed; |
ec8acf20a swap: add per-par... |
553 |
spin_unlock(&si->lock); |
c9e444103 mm: reuse unused ... |
554 |
swap_was_freed = __try_to_reclaim_swap(si, offset); |
ec8acf20a swap: add per-par... |
555 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
556 557 558 559 560 |
/* entry was freed successfully, try to use this again */ if (swap_was_freed) goto checks; goto scan; /* check next one */ } |
ebebbbe90 swapfile: rearran... |
561 562 563 564 565 566 567 568 569 570 571 |
if (si->swap_map[offset]) goto scan; if (offset == si->lowest_bit) si->lowest_bit++; if (offset == si->highest_bit) si->highest_bit--; si->inuse_pages++; if (si->inuse_pages == si->pages) { si->lowest_bit = si->max; si->highest_bit = 0; |
18ab4d4ce swap: change swap... |
572 573 574 |
spin_lock(&swap_avail_lock); plist_del(&si->avail_list, &swap_avail_head); spin_unlock(&swap_avail_lock); |
1da177e4c Linux-2.6.12-rc2 |
575 |
} |
253d553ba swap_info: SWAP_H... |
576 |
si->swap_map[offset] = usage; |
2a8f94493 swap: change bloc... |
577 |
inc_cluster_info_page(si, si->cluster_info, offset); |
ebebbbe90 swapfile: rearran... |
578 579 |
si->cluster_next = offset + 1; si->flags -= SWP_SCANNING; |
7992fde72 swapfile: swap al... |
580 |
|
ebebbbe90 swapfile: rearran... |
581 |
return offset; |
7dfad4183 [PATCH] swap: sca... |
582 |
|
ebebbbe90 swapfile: rearran... |
583 |
scan: |
ec8acf20a swap: add per-par... |
584 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
585 |
while (++offset <= si->highest_bit) { |
52b7efdbe [PATCH] swap: sca... |
586 |
if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
587 |
spin_lock(&si->lock); |
52b7efdbe [PATCH] swap: sca... |
588 589 |
goto checks; } |
c9e444103 mm: reuse unused ... |
590 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
591 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
592 593 |
goto checks; } |
048c27fd7 [PATCH] swap: sca... |
594 595 596 597 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
7dfad4183 [PATCH] swap: sca... |
598 |
} |
c60aa176c swapfile: swap al... |
599 |
offset = si->lowest_bit; |
a5998061d mm/swapfile.c: do... |
600 |
while (offset < scan_base) { |
c60aa176c swapfile: swap al... |
601 |
if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
602 |
spin_lock(&si->lock); |
c60aa176c swapfile: swap al... |
603 604 |
goto checks; } |
c9e444103 mm: reuse unused ... |
605 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
606 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
607 608 |
goto checks; } |
c60aa176c swapfile: swap al... |
609 610 611 612 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
a5998061d mm/swapfile.c: do... |
613 |
offset++; |
c60aa176c swapfile: swap al... |
614 |
} |
ec8acf20a swap: add per-par... |
615 |
spin_lock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
616 617 |
no_page: |
52b7efdbe [PATCH] swap: sca... |
618 |
si->flags -= SWP_SCANNING; |
1da177e4c Linux-2.6.12-rc2 |
619 620 621 622 623 |
return 0; } swp_entry_t get_swap_page(void) { |
adfab836f swap: change swap... |
624 |
struct swap_info_struct *si, *next; |
fb4f88dca [PATCH] swap: get... |
625 |
pgoff_t offset; |
1da177e4c Linux-2.6.12-rc2 |
626 |
|
ec8acf20a swap: add per-par... |
627 |
if (atomic_long_read(&nr_swap_pages) <= 0) |
fb4f88dca [PATCH] swap: get... |
628 |
goto noswap; |
ec8acf20a swap: add per-par... |
629 |
atomic_long_dec(&nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
630 |
|
18ab4d4ce swap: change swap... |
631 632 633 634 635 636 637 |
spin_lock(&swap_avail_lock); start_over: plist_for_each_entry_safe(si, next, &swap_avail_head, avail_list) { /* requeue si to after same-priority siblings */ plist_requeue(&si->avail_list, &swap_avail_head); spin_unlock(&swap_avail_lock); |
ec8acf20a swap: add per-par... |
638 |
spin_lock(&si->lock); |
adfab836f swap: change swap... |
639 |
if (!si->highest_bit || !(si->flags & SWP_WRITEOK)) { |
18ab4d4ce swap: change swap... |
640 641 642 643 644 645 646 647 648 649 650 651 652 653 |
spin_lock(&swap_avail_lock); if (plist_node_empty(&si->avail_list)) { spin_unlock(&si->lock); goto nextsi; } WARN(!si->highest_bit, "swap_info %d in list but !highest_bit ", si->type); WARN(!(si->flags & SWP_WRITEOK), "swap_info %d in list but !SWP_WRITEOK ", si->type); plist_del(&si->avail_list, &swap_avail_head); |
ec8acf20a swap: add per-par... |
654 |
spin_unlock(&si->lock); |
18ab4d4ce swap: change swap... |
655 |
goto nextsi; |
ec8acf20a swap: add per-par... |
656 |
} |
fb4f88dca [PATCH] swap: get... |
657 |
|
355cfa73d mm: modify swap_m... |
658 |
/* This is called for allocating swap entry for cache */ |
253d553ba swap_info: SWAP_H... |
659 |
offset = scan_swap_map(si, SWAP_HAS_CACHE); |
ec8acf20a swap: add per-par... |
660 661 |
spin_unlock(&si->lock); if (offset) |
adfab836f swap: change swap... |
662 |
return swp_entry(si->type, offset); |
18ab4d4ce swap: change swap... |
663 664 665 666 667 |
pr_debug("scan_swap_map of si %d failed to find offset ", si->type); spin_lock(&swap_avail_lock); nextsi: |
adfab836f swap: change swap... |
668 669 670 671 |
/* * if we got here, it's likely that si was almost full before, * and since scan_swap_map() can drop the si->lock, multiple * callers probably all tried to get a page from the same si |
18ab4d4ce swap: change swap... |
672 673 674 675 676 |
* and it filled up before we could get one; or, the si filled * up between us dropping swap_avail_lock and taking si->lock. * Since we dropped the swap_avail_lock, the swap_avail_head * list may have been modified; so if next is still in the * swap_avail_head list then try it, otherwise start over. |
adfab836f swap: change swap... |
677 |
*/ |
18ab4d4ce swap: change swap... |
678 679 |
if (plist_node_empty(&next->avail_list)) goto start_over; |
1da177e4c Linux-2.6.12-rc2 |
680 |
} |
fb4f88dca [PATCH] swap: get... |
681 |
|
18ab4d4ce swap: change swap... |
682 |
spin_unlock(&swap_avail_lock); |
ec8acf20a swap: add per-par... |
683 |
atomic_long_inc(&nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
684 |
noswap: |
fb4f88dca [PATCH] swap: get... |
685 |
return (swp_entry_t) {0}; |
1da177e4c Linux-2.6.12-rc2 |
686 |
} |
2de1a7e40 mm/swapfile.c: fi... |
687 |
/* The only caller of this function is now suspend routine */ |
910321ea8 swap: revert spec... |
688 689 690 691 |
swp_entry_t get_swap_page_of_type(int type) { struct swap_info_struct *si; pgoff_t offset; |
910321ea8 swap: revert spec... |
692 |
si = swap_info[type]; |
ec8acf20a swap: add per-par... |
693 |
spin_lock(&si->lock); |
910321ea8 swap: revert spec... |
694 |
if (si && (si->flags & SWP_WRITEOK)) { |
ec8acf20a swap: add per-par... |
695 |
atomic_long_dec(&nr_swap_pages); |
910321ea8 swap: revert spec... |
696 697 698 |
/* This is called for allocating swap entry, not cache */ offset = scan_swap_map(si, 1); if (offset) { |
ec8acf20a swap: add per-par... |
699 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
700 701 |
return swp_entry(type, offset); } |
ec8acf20a swap: add per-par... |
702 |
atomic_long_inc(&nr_swap_pages); |
910321ea8 swap: revert spec... |
703 |
} |
ec8acf20a swap: add per-par... |
704 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
705 706 |
return (swp_entry_t) {0}; } |
73c34b6ac swap_info: miscel... |
707 |
static struct swap_info_struct *swap_info_get(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
708 |
{ |
73c34b6ac swap_info: miscel... |
709 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
710 711 712 713 714 715 716 |
unsigned long offset, type; if (!entry.val) goto out; type = swp_type(entry); if (type >= nr_swapfiles) goto bad_nofile; |
efa90a981 swap_info: change... |
717 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
718 719 720 721 722 723 724 |
if (!(p->flags & SWP_USED)) goto bad_device; offset = swp_offset(entry); if (offset >= p->max) goto bad_offset; if (!p->swap_map[offset]) goto bad_free; |
ec8acf20a swap: add per-par... |
725 |
spin_lock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
726 727 728 |
return p; bad_free: |
465c47fd8 mm/swapfile.c: co... |
729 730 |
pr_err("swap_free: %s%08lx ", Unused_offset, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
731 732 |
goto out; bad_offset: |
465c47fd8 mm/swapfile.c: co... |
733 734 |
pr_err("swap_free: %s%08lx ", Bad_offset, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
735 736 |
goto out; bad_device: |
465c47fd8 mm/swapfile.c: co... |
737 738 |
pr_err("swap_free: %s%08lx ", Unused_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
739 740 |
goto out; bad_nofile: |
465c47fd8 mm/swapfile.c: co... |
741 742 |
pr_err("swap_free: %s%08lx ", Bad_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
743 744 |
out: return NULL; |
886bb7e9c swapfile: remove ... |
745 |
} |
1da177e4c Linux-2.6.12-rc2 |
746 |
|
8d69aaee8 swap_info: swap_m... |
747 748 |
static unsigned char swap_entry_free(struct swap_info_struct *p, swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
749 |
{ |
253d553ba swap_info: SWAP_H... |
750 |
unsigned long offset = swp_offset(entry); |
8d69aaee8 swap_info: swap_m... |
751 752 |
unsigned char count; unsigned char has_cache; |
355cfa73d mm: modify swap_m... |
753 |
|
253d553ba swap_info: SWAP_H... |
754 755 756 |
count = p->swap_map[offset]; has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; |
355cfa73d mm: modify swap_m... |
757 |
|
253d553ba swap_info: SWAP_H... |
758 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
759 |
VM_BUG_ON(!has_cache); |
253d553ba swap_info: SWAP_H... |
760 |
has_cache = 0; |
aaa468653 swap_info: note S... |
761 762 763 764 765 766 |
} else if (count == SWAP_MAP_SHMEM) { /* * Or we could insist on shmem.c using a special * swap_shmem_free() and free_shmem_swap_and_cache()... */ count = 0; |
570a335b8 swap_info: swap c... |
767 768 769 770 771 772 773 774 775 |
} else if ((count & ~COUNT_CONTINUED) <= SWAP_MAP_MAX) { if (count == COUNT_CONTINUED) { if (swap_count_continued(p, offset, count)) count = SWAP_MAP_MAX | COUNT_CONTINUED; else count = SWAP_MAP_MAX; } else count--; } |
253d553ba swap_info: SWAP_H... |
776 777 778 779 780 781 |
if (!count) mem_cgroup_uncharge_swap(entry); usage = count | has_cache; p->swap_map[offset] = usage; |
355cfa73d mm: modify swap_m... |
782 |
|
355cfa73d mm: modify swap_m... |
783 |
/* free if no reference */ |
253d553ba swap_info: SWAP_H... |
784 |
if (!usage) { |
2a8f94493 swap: change bloc... |
785 |
dec_cluster_info_page(p, p->cluster_info, offset); |
355cfa73d mm: modify swap_m... |
786 787 |
if (offset < p->lowest_bit) p->lowest_bit = offset; |
18ab4d4ce swap: change swap... |
788 789 |
if (offset > p->highest_bit) { bool was_full = !p->highest_bit; |
355cfa73d mm: modify swap_m... |
790 |
p->highest_bit = offset; |
18ab4d4ce swap: change swap... |
791 792 793 794 795 796 797 798 799 |
if (was_full && (p->flags & SWP_WRITEOK)) { spin_lock(&swap_avail_lock); WARN_ON(!plist_node_empty(&p->avail_list)); if (plist_node_empty(&p->avail_list)) plist_add(&p->avail_list, &swap_avail_head); spin_unlock(&swap_avail_lock); } } |
ec8acf20a swap: add per-par... |
800 |
atomic_long_inc(&nr_swap_pages); |
355cfa73d mm: modify swap_m... |
801 |
p->inuse_pages--; |
38b5faf4b mm: frontswap: co... |
802 |
frontswap_invalidate_page(p->type, offset); |
737449236 swapfile: avoid d... |
803 804 805 806 807 808 |
if (p->flags & SWP_BLKDEV) { struct gendisk *disk = p->bdev->bd_disk; if (disk->fops->swap_slot_free_notify) disk->fops->swap_slot_free_notify(p->bdev, offset); } |
1da177e4c Linux-2.6.12-rc2 |
809 |
} |
253d553ba swap_info: SWAP_H... |
810 811 |
return usage; |
1da177e4c Linux-2.6.12-rc2 |
812 813 814 |
} /* |
2de1a7e40 mm/swapfile.c: fi... |
815 |
* Caller has made sure that the swap device corresponding to entry |
1da177e4c Linux-2.6.12-rc2 |
816 817 818 819 |
* is still around or has not been recycled. */ void swap_free(swp_entry_t entry) { |
73c34b6ac swap_info: miscel... |
820 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
821 822 823 |
p = swap_info_get(entry); if (p) { |
253d553ba swap_info: SWAP_H... |
824 |
swap_entry_free(p, entry, 1); |
ec8acf20a swap: add per-par... |
825 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
826 827 828 829 |
} } /* |
cb4b86ba4 mm: add swap cach... |
830 831 |
* Called after dropping swapcache to decrease refcnt to swap entries. */ |
0a31bc97c mm: memcontrol: r... |
832 |
void swapcache_free(swp_entry_t entry) |
cb4b86ba4 mm: add swap cach... |
833 |
{ |
355cfa73d mm: modify swap_m... |
834 |
struct swap_info_struct *p; |
355cfa73d mm: modify swap_m... |
835 836 |
p = swap_info_get(entry); if (p) { |
0a31bc97c mm: memcontrol: r... |
837 |
swap_entry_free(p, entry, SWAP_HAS_CACHE); |
ec8acf20a swap: add per-par... |
838 |
spin_unlock(&p->lock); |
355cfa73d mm: modify swap_m... |
839 |
} |
cb4b86ba4 mm: add swap cach... |
840 841 842 |
} /* |
c475a8ab6 [PATCH] can_share... |
843 |
* How many references to page are currently swapped out? |
570a335b8 swap_info: swap c... |
844 845 |
* This does not give an exact answer when swap count is continued, * but does include the high COUNT_CONTINUED flag to allow for that. |
1da177e4c Linux-2.6.12-rc2 |
846 |
*/ |
bde05d1cc shmem: replace pa... |
847 |
int page_swapcount(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
848 |
{ |
c475a8ab6 [PATCH] can_share... |
849 850 |
int count = 0; struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
851 |
swp_entry_t entry; |
4c21e2f24 [PATCH] mm: split... |
852 |
entry.val = page_private(page); |
1da177e4c Linux-2.6.12-rc2 |
853 854 |
p = swap_info_get(entry); if (p) { |
355cfa73d mm: modify swap_m... |
855 |
count = swap_count(p->swap_map[swp_offset(entry)]); |
ec8acf20a swap: add per-par... |
856 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
857 |
} |
c475a8ab6 [PATCH] can_share... |
858 |
return count; |
1da177e4c Linux-2.6.12-rc2 |
859 860 861 |
} /* |
8334b9622 mm: /proc/pid/sma... |
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 |
* How many references to @entry are currently swapped out? * This considers COUNT_CONTINUED so it returns exact answer. */ int swp_swapcount(swp_entry_t entry) { int count, tmp_count, n; struct swap_info_struct *p; struct page *page; pgoff_t offset; unsigned char *map; p = swap_info_get(entry); if (!p) return 0; count = swap_count(p->swap_map[swp_offset(entry)]); if (!(count & COUNT_CONTINUED)) goto out; count &= ~COUNT_CONTINUED; n = SWAP_MAP_MAX + 1; offset = swp_offset(entry); page = vmalloc_to_page(p->swap_map + offset); offset &= ~PAGE_MASK; VM_BUG_ON(page_private(page) != SWP_CONTINUED); do { page = list_entry(page->lru.next, struct page, lru); map = kmap_atomic(page); tmp_count = map[offset]; kunmap_atomic(map); count += (tmp_count & ~COUNT_CONTINUED) * n; n *= (SWAP_CONT_MAX + 1); } while (tmp_count & COUNT_CONTINUED); out: spin_unlock(&p->lock); return count; } /* |
7b1fe5979 mm: reuse_swap_pa... |
904 905 906 907 |
* We can write to an anon page without COW if there are no other references * to it. And as a side-effect, free up its swap: because the old content * on disk will never be read, and seeking back there to write new content * later would only waste time away from clustering. |
1da177e4c Linux-2.6.12-rc2 |
908 |
*/ |
7b1fe5979 mm: reuse_swap_pa... |
909 |
int reuse_swap_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
910 |
{ |
c475a8ab6 [PATCH] can_share... |
911 |
int count; |
309381fea mm: dump page whe... |
912 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
5ad646880 ksm: let shared p... |
913 914 |
if (unlikely(PageKsm(page))) return 0; |
c475a8ab6 [PATCH] can_share... |
915 |
count = page_mapcount(page); |
7b1fe5979 mm: reuse_swap_pa... |
916 |
if (count <= 1 && PageSwapCache(page)) { |
c475a8ab6 [PATCH] can_share... |
917 |
count += page_swapcount(page); |
7b1fe5979 mm: reuse_swap_pa... |
918 919 920 921 922 |
if (count == 1 && !PageWriteback(page)) { delete_from_swap_cache(page); SetPageDirty(page); } } |
5ad646880 ksm: let shared p... |
923 |
return count <= 1; |
1da177e4c Linux-2.6.12-rc2 |
924 925 926 |
} /* |
a2c43eed8 mm: try_to_free_s... |
927 928 |
* If swap is getting full, or if there are no more mappings of this page, * then try_to_free_swap is called to free its swap space. |
1da177e4c Linux-2.6.12-rc2 |
929 |
*/ |
a2c43eed8 mm: try_to_free_s... |
930 |
int try_to_free_swap(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
931 |
{ |
309381fea mm: dump page whe... |
932 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
1da177e4c Linux-2.6.12-rc2 |
933 934 935 936 937 |
if (!PageSwapCache(page)) return 0; if (PageWriteback(page)) return 0; |
a2c43eed8 mm: try_to_free_s... |
938 |
if (page_swapcount(page)) |
1da177e4c Linux-2.6.12-rc2 |
939 |
return 0; |
b73d7fcec swap: prevent reu... |
940 941 942 943 944 945 946 947 948 949 950 951 |
/* * Once hibernation has begun to create its image of memory, * there's a danger that one of the calls to try_to_free_swap() * - most probably a call from __try_to_reclaim_swap() while * hibernation is allocating its own swap pages for the image, * but conceivably even a call from memory reclaim - will free * the swap from a page which has already been recorded in the * image as a clean swapcache page, and then reuse its swap for * another page of the image. On waking from hibernation, the * original page might be freed under memory pressure, then * later read back in from swap, now with the wrong data. * |
2de1a7e40 mm/swapfile.c: fi... |
952 |
* Hibernation suspends storage while it is writing the image |
f90ac3982 mm: avoid liveloc... |
953 |
* to disk so check that here. |
b73d7fcec swap: prevent reu... |
954 |
*/ |
f90ac3982 mm: avoid liveloc... |
955 |
if (pm_suspended_storage()) |
b73d7fcec swap: prevent reu... |
956 |
return 0; |
a2c43eed8 mm: try_to_free_s... |
957 958 959 |
delete_from_swap_cache(page); SetPageDirty(page); return 1; |
68a22394c vmscan: free swap... |
960 961 962 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
963 964 965 |
* Free the swap entry like above, but also try to * free the page cache entry if it is the last user. */ |
2509ef26d badpage: zap prin... |
966 |
int free_swap_and_cache(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
967 |
{ |
2509ef26d badpage: zap prin... |
968 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
969 |
struct page *page = NULL; |
a7420aa54 HWPOISON: Add sup... |
970 |
if (non_swap_entry(entry)) |
2509ef26d badpage: zap prin... |
971 |
return 1; |
0697212a4 [PATCH] Swapless ... |
972 |
|
1da177e4c Linux-2.6.12-rc2 |
973 974 |
p = swap_info_get(entry); if (p) { |
253d553ba swap_info: SWAP_H... |
975 |
if (swap_entry_free(p, entry, 1) == SWAP_HAS_CACHE) { |
33806f06d swap: make each s... |
976 977 |
page = find_get_page(swap_address_space(entry), entry.val); |
8413ac9d8 mm: page lock use... |
978 |
if (page && !trylock_page(page)) { |
93fac7041 [PATCH] mm: sched... |
979 980 981 982 |
page_cache_release(page); page = NULL; } } |
ec8acf20a swap: add per-par... |
983 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
984 985 |
} if (page) { |
a2c43eed8 mm: try_to_free_s... |
986 987 988 989 |
/* * Not mapped elsewhere, or swap space full? Free it! * Also recheck PageSwapCache now page is locked (above). */ |
93fac7041 [PATCH] mm: sched... |
990 |
if (PageSwapCache(page) && !PageWriteback(page) && |
a2c43eed8 mm: try_to_free_s... |
991 |
(!page_mapped(page) || vm_swap_full())) { |
1da177e4c Linux-2.6.12-rc2 |
992 993 994 995 996 997 |
delete_from_swap_cache(page); SetPageDirty(page); } unlock_page(page); page_cache_release(page); } |
2509ef26d badpage: zap prin... |
998 |
return p != NULL; |
1da177e4c Linux-2.6.12-rc2 |
999 |
} |
b0cb1a19d Replace CONFIG_SO... |
1000 |
#ifdef CONFIG_HIBERNATION |
f577eb30a [PATCH] swsusp: l... |
1001 |
/* |
915bae9eb [PATCH] swsusp: u... |
1002 |
* Find the swap type that corresponds to given device (if any). |
f577eb30a [PATCH] swsusp: l... |
1003 |
* |
915bae9eb [PATCH] swsusp: u... |
1004 1005 1006 1007 |
* @offset - number of the PAGE_SIZE-sized block of the device, starting * from 0, in which the swap header is expected to be located. * * This is needed for the suspend to disk (aka swsusp). |
f577eb30a [PATCH] swsusp: l... |
1008 |
*/ |
7bf236874 [PATCH] swsusp: D... |
1009 |
int swap_type_of(dev_t device, sector_t offset, struct block_device **bdev_p) |
f577eb30a [PATCH] swsusp: l... |
1010 |
{ |
915bae9eb [PATCH] swsusp: u... |
1011 |
struct block_device *bdev = NULL; |
efa90a981 swap_info: change... |
1012 |
int type; |
f577eb30a [PATCH] swsusp: l... |
1013 |
|
915bae9eb [PATCH] swsusp: u... |
1014 1015 |
if (device) bdev = bdget(device); |
f577eb30a [PATCH] swsusp: l... |
1016 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
1017 1018 |
for (type = 0; type < nr_swapfiles; type++) { struct swap_info_struct *sis = swap_info[type]; |
f577eb30a [PATCH] swsusp: l... |
1019 |
|
915bae9eb [PATCH] swsusp: u... |
1020 |
if (!(sis->flags & SWP_WRITEOK)) |
f577eb30a [PATCH] swsusp: l... |
1021 |
continue; |
b6b5bce35 [PATCH] swsusp: F... |
1022 |
|
915bae9eb [PATCH] swsusp: u... |
1023 |
if (!bdev) { |
7bf236874 [PATCH] swsusp: D... |
1024 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
1025 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
1026 |
|
6e1819d61 [PATCH] swsusp: u... |
1027 |
spin_unlock(&swap_lock); |
efa90a981 swap_info: change... |
1028 |
return type; |
6e1819d61 [PATCH] swsusp: u... |
1029 |
} |
915bae9eb [PATCH] swsusp: u... |
1030 |
if (bdev == sis->bdev) { |
9625a5f28 swap_info: includ... |
1031 |
struct swap_extent *se = &sis->first_swap_extent; |
915bae9eb [PATCH] swsusp: u... |
1032 |
|
915bae9eb [PATCH] swsusp: u... |
1033 |
if (se->start_block == offset) { |
7bf236874 [PATCH] swsusp: D... |
1034 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
1035 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
1036 |
|
915bae9eb [PATCH] swsusp: u... |
1037 1038 |
spin_unlock(&swap_lock); bdput(bdev); |
efa90a981 swap_info: change... |
1039 |
return type; |
915bae9eb [PATCH] swsusp: u... |
1040 |
} |
f577eb30a [PATCH] swsusp: l... |
1041 1042 1043 |
} } spin_unlock(&swap_lock); |
915bae9eb [PATCH] swsusp: u... |
1044 1045 |
if (bdev) bdput(bdev); |
f577eb30a [PATCH] swsusp: l... |
1046 1047 1048 1049 |
return -ENODEV; } /* |
73c34b6ac swap_info: miscel... |
1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 |
* Get the (PAGE_SIZE) block corresponding to given offset on the swapdev * corresponding to given index in swap_info (swap type). */ sector_t swapdev_block(int type, pgoff_t offset) { struct block_device *bdev; if ((unsigned int)type >= nr_swapfiles) return 0; if (!(swap_info[type]->flags & SWP_WRITEOK)) return 0; |
d4906e1aa swap: rework map_... |
1061 |
return map_swap_entry(swp_entry(type, offset), &bdev); |
73c34b6ac swap_info: miscel... |
1062 1063 1064 |
} /* |
f577eb30a [PATCH] swsusp: l... |
1065 1066 1067 1068 1069 1070 1071 1072 |
* Return either the total number of swap pages of given type, or the number * of free pages of that type (depending on @free) * * This is needed for software suspend */ unsigned int count_swap_pages(int type, int free) { unsigned int n = 0; |
efa90a981 swap_info: change... |
1073 1074 1075 |
spin_lock(&swap_lock); if ((unsigned int)type < nr_swapfiles) { struct swap_info_struct *sis = swap_info[type]; |
ec8acf20a swap: add per-par... |
1076 |
spin_lock(&sis->lock); |
efa90a981 swap_info: change... |
1077 1078 |
if (sis->flags & SWP_WRITEOK) { n = sis->pages; |
f577eb30a [PATCH] swsusp: l... |
1079 |
if (free) |
efa90a981 swap_info: change... |
1080 |
n -= sis->inuse_pages; |
f577eb30a [PATCH] swsusp: l... |
1081 |
} |
ec8acf20a swap: add per-par... |
1082 |
spin_unlock(&sis->lock); |
f577eb30a [PATCH] swsusp: l... |
1083 |
} |
efa90a981 swap_info: change... |
1084 |
spin_unlock(&swap_lock); |
f577eb30a [PATCH] swsusp: l... |
1085 1086 |
return n; } |
73c34b6ac swap_info: miscel... |
1087 |
#endif /* CONFIG_HIBERNATION */ |
f577eb30a [PATCH] swsusp: l... |
1088 |
|
179ef71cb mm: save soft-dir... |
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 |
static inline int maybe_same_pte(pte_t pte, pte_t swp_pte) { #ifdef CONFIG_MEM_SOFT_DIRTY /* * When pte keeps soft dirty bit the pte generated * from swap entry does not has it, still it's same * pte from logical point of view. */ pte_t swp_pte_dirty = pte_swp_mksoft_dirty(swp_pte); return pte_same(pte, swp_pte) || pte_same(pte, swp_pte_dirty); #else return pte_same(pte, swp_pte); #endif } |
1da177e4c Linux-2.6.12-rc2 |
1103 |
/* |
72866f6f2 [PATCH] mm: anon ... |
1104 1105 1106 |
* No need to decide whether this PTE shares the swap entry with others, * just let do_wp_page work it out if a write is requested later - to * force COW, vm_page_prot omits write permission from any private vma. |
1da177e4c Linux-2.6.12-rc2 |
1107 |
*/ |
044d66c1d memcgroup: reinst... |
1108 |
static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd, |
1da177e4c Linux-2.6.12-rc2 |
1109 1110 |
unsigned long addr, swp_entry_t entry, struct page *page) { |
9e16b7fb1 mm,ksm: swapoff m... |
1111 |
struct page *swapcache; |
72835c86c mm: unify remaini... |
1112 |
struct mem_cgroup *memcg; |
044d66c1d memcgroup: reinst... |
1113 1114 1115 |
spinlock_t *ptl; pte_t *pte; int ret = 1; |
9e16b7fb1 mm,ksm: swapoff m... |
1116 1117 1118 1119 |
swapcache = page; page = ksm_might_need_to_copy(page, vma, addr); if (unlikely(!page)) return -ENOMEM; |
00501b531 mm: memcontrol: r... |
1120 |
if (mem_cgroup_try_charge(page, vma->vm_mm, GFP_KERNEL, &memcg)) { |
044d66c1d memcgroup: reinst... |
1121 |
ret = -ENOMEM; |
85d9fc89f memcg: fix refcnt... |
1122 1123 |
goto out_nolock; } |
044d66c1d memcgroup: reinst... |
1124 1125 |
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); |
179ef71cb mm: save soft-dir... |
1126 |
if (unlikely(!maybe_same_pte(*pte, swp_entry_to_pte(entry)))) { |
00501b531 mm: memcontrol: r... |
1127 |
mem_cgroup_cancel_charge(page, memcg); |
044d66c1d memcgroup: reinst... |
1128 1129 1130 |
ret = 0; goto out; } |
8a9f3ccd2 Memory controller... |
1131 |
|
b084d4353 mm: count swap usage |
1132 |
dec_mm_counter(vma->vm_mm, MM_SWAPENTS); |
d559db086 mm: clean up mm_c... |
1133 |
inc_mm_counter(vma->vm_mm, MM_ANONPAGES); |
1da177e4c Linux-2.6.12-rc2 |
1134 1135 1136 |
get_page(page); set_pte_at(vma->vm_mm, addr, pte, pte_mkold(mk_pte(page, vma->vm_page_prot))); |
00501b531 mm: memcontrol: r... |
1137 |
if (page == swapcache) { |
9e16b7fb1 mm,ksm: swapoff m... |
1138 |
page_add_anon_rmap(page, vma, addr); |
00501b531 mm: memcontrol: r... |
1139 1140 |
mem_cgroup_commit_charge(page, memcg, true); } else { /* ksm created a completely new copy */ |
9e16b7fb1 mm,ksm: swapoff m... |
1141 |
page_add_new_anon_rmap(page, vma, addr); |
00501b531 mm: memcontrol: r... |
1142 1143 1144 |
mem_cgroup_commit_charge(page, memcg, false); lru_cache_add_active_or_unevictable(page, vma); } |
1da177e4c Linux-2.6.12-rc2 |
1145 1146 1147 1148 1149 1150 |
swap_free(entry); /* * Move the page to the active list so it is not * immediately swapped out again after swapon. */ activate_page(page); |
044d66c1d memcgroup: reinst... |
1151 1152 |
out: pte_unmap_unlock(pte, ptl); |
85d9fc89f memcg: fix refcnt... |
1153 |
out_nolock: |
9e16b7fb1 mm,ksm: swapoff m... |
1154 1155 1156 1157 |
if (page != swapcache) { unlock_page(page); put_page(page); } |
044d66c1d memcgroup: reinst... |
1158 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1159 1160 1161 1162 1163 1164 |
} static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd, unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { |
1da177e4c Linux-2.6.12-rc2 |
1165 |
pte_t swp_pte = swp_entry_to_pte(entry); |
705e87c0c [PATCH] mm: pte_o... |
1166 |
pte_t *pte; |
8a9f3ccd2 Memory controller... |
1167 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1168 |
|
044d66c1d memcgroup: reinst... |
1169 1170 1171 1172 1173 1174 1175 |
/* * We don't actually need pte lock while scanning for swp_pte: since * we hold page lock and mmap_sem, swp_pte cannot be inserted into the * page table while we're scanning; though it could get zapped, and on * some architectures (e.g. x86_32 with PAE) we might catch a glimpse * of unmatched parts which look like swp_pte, so unuse_pte must * recheck under pte lock. Scanning without pte lock lets it be |
2de1a7e40 mm/swapfile.c: fi... |
1176 |
* preemptable whenever CONFIG_PREEMPT but not CONFIG_HIGHPTE. |
044d66c1d memcgroup: reinst... |
1177 1178 |
*/ pte = pte_offset_map(pmd, addr); |
1da177e4c Linux-2.6.12-rc2 |
1179 1180 1181 1182 1183 |
do { /* * swapoff spends a _lot_ of time in this loop! * Test inline before going to call unuse_pte. */ |
179ef71cb mm: save soft-dir... |
1184 |
if (unlikely(maybe_same_pte(*pte, swp_pte))) { |
044d66c1d memcgroup: reinst... |
1185 1186 1187 1188 1189 |
pte_unmap(pte); ret = unuse_pte(vma, pmd, addr, entry, page); if (ret) goto out; pte = pte_offset_map(pmd, addr); |
1da177e4c Linux-2.6.12-rc2 |
1190 1191 |
} } while (pte++, addr += PAGE_SIZE, addr != end); |
044d66c1d memcgroup: reinst... |
1192 1193 |
pte_unmap(pte - 1); out: |
8a9f3ccd2 Memory controller... |
1194 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1195 1196 1197 1198 1199 1200 1201 1202 |
} static inline int unuse_pmd_range(struct vm_area_struct *vma, pud_t *pud, unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { pmd_t *pmd; unsigned long next; |
8a9f3ccd2 Memory controller... |
1203 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1204 1205 1206 1207 |
pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); |
1a5a9906d mm: thp: fix pmd_... |
1208 |
if (pmd_none_or_trans_huge_or_clear_bad(pmd)) |
1da177e4c Linux-2.6.12-rc2 |
1209 |
continue; |
8a9f3ccd2 Memory controller... |
1210 1211 1212 |
ret = unuse_pte_range(vma, pmd, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 |
} while (pmd++, addr = next, addr != end); return 0; } static inline int unuse_pud_range(struct vm_area_struct *vma, pgd_t *pgd, unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { pud_t *pud; unsigned long next; |
8a9f3ccd2 Memory controller... |
1223 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1224 1225 1226 1227 1228 1229 |
pud = pud_offset(pgd, addr); do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) continue; |
8a9f3ccd2 Memory controller... |
1230 1231 1232 |
ret = unuse_pmd_range(vma, pud, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1233 1234 1235 1236 1237 1238 1239 1240 1241 |
} while (pud++, addr = next, addr != end); return 0; } static int unuse_vma(struct vm_area_struct *vma, swp_entry_t entry, struct page *page) { pgd_t *pgd; unsigned long addr, end, next; |
8a9f3ccd2 Memory controller... |
1242 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1243 |
|
3ca7b3c5b mm: define PAGE_M... |
1244 |
if (page_anon_vma(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 |
addr = page_address_in_vma(page, vma); if (addr == -EFAULT) return 0; else end = addr + PAGE_SIZE; } else { addr = vma->vm_start; end = vma->vm_end; } pgd = pgd_offset(vma->vm_mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; |
8a9f3ccd2 Memory controller... |
1260 1261 1262 |
ret = unuse_pud_range(vma, pgd, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1263 1264 1265 1266 1267 1268 1269 1270 |
} while (pgd++, addr = next, addr != end); return 0; } static int unuse_mm(struct mm_struct *mm, swp_entry_t entry, struct page *page) { struct vm_area_struct *vma; |
8a9f3ccd2 Memory controller... |
1271 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1272 1273 1274 |
if (!down_read_trylock(&mm->mmap_sem)) { /* |
7d03431cf swapfile/vmscan: ... |
1275 1276 |
* Activate page so shrink_inactive_list is unlikely to unmap * its ptes while lock is dropped, so swapoff can make progress. |
1da177e4c Linux-2.6.12-rc2 |
1277 |
*/ |
c475a8ab6 [PATCH] can_share... |
1278 |
activate_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1279 1280 1281 1282 |
unlock_page(page); down_read(&mm->mmap_sem); lock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1283 |
for (vma = mm->mmap; vma; vma = vma->vm_next) { |
8a9f3ccd2 Memory controller... |
1284 |
if (vma->anon_vma && (ret = unuse_vma(vma, entry, page))) |
1da177e4c Linux-2.6.12-rc2 |
1285 1286 |
break; } |
1da177e4c Linux-2.6.12-rc2 |
1287 |
up_read(&mm->mmap_sem); |
8a9f3ccd2 Memory controller... |
1288 |
return (ret < 0)? ret: 0; |
1da177e4c Linux-2.6.12-rc2 |
1289 1290 1291 |
} /* |
38b5faf4b mm: frontswap: co... |
1292 1293 |
* Scan swap_map (or frontswap_map if frontswap parameter is true) * from current position to next entry still in use. |
1da177e4c Linux-2.6.12-rc2 |
1294 1295 |
* Recycle to start on reaching the end, returning 0 when empty. */ |
6eb396dc4 [PATCH] swap: swa... |
1296 |
static unsigned int find_next_to_unuse(struct swap_info_struct *si, |
38b5faf4b mm: frontswap: co... |
1297 |
unsigned int prev, bool frontswap) |
1da177e4c Linux-2.6.12-rc2 |
1298 |
{ |
6eb396dc4 [PATCH] swap: swa... |
1299 1300 |
unsigned int max = si->max; unsigned int i = prev; |
8d69aaee8 swap_info: swap_m... |
1301 |
unsigned char count; |
1da177e4c Linux-2.6.12-rc2 |
1302 1303 |
/* |
5d337b919 [PATCH] swap: swa... |
1304 |
* No need for swap_lock here: we're just looking |
1da177e4c Linux-2.6.12-rc2 |
1305 1306 |
* for whether an entry is in use, not modifying it; false * hits are okay, and sys_swapoff() has already prevented new |
5d337b919 [PATCH] swap: swa... |
1307 |
* allocations from this area (while holding swap_lock). |
1da177e4c Linux-2.6.12-rc2 |
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 |
*/ for (;;) { if (++i >= max) { if (!prev) { i = 0; break; } /* * No entries in use at top of swap_map, * loop back to start and recheck there. */ max = prev + 1; prev = 0; i = 1; } |
38b5faf4b mm: frontswap: co... |
1323 1324 1325 1326 1327 1328 |
if (frontswap) { if (frontswap_test(si, i)) break; else continue; } |
4db0c3c29 mm: remove rest o... |
1329 |
count = READ_ONCE(si->swap_map[i]); |
355cfa73d mm: modify swap_m... |
1330 |
if (count && swap_count(count) != SWAP_MAP_BAD) |
1da177e4c Linux-2.6.12-rc2 |
1331 1332 1333 1334 1335 1336 1337 1338 1339 |
break; } return i; } /* * We completely avoid races by reading each swap page in advance, * and then search for the process using it. All the necessary * page table adjustments can then be made atomically. |
38b5faf4b mm: frontswap: co... |
1340 1341 1342 |
* * if the boolean frontswap is true, only unuse pages_to_unuse pages; * pages_to_unuse==0 means all pages; ignored if frontswap is false |
1da177e4c Linux-2.6.12-rc2 |
1343 |
*/ |
38b5faf4b mm: frontswap: co... |
1344 1345 |
int try_to_unuse(unsigned int type, bool frontswap, unsigned long pages_to_unuse) |
1da177e4c Linux-2.6.12-rc2 |
1346 |
{ |
efa90a981 swap_info: change... |
1347 |
struct swap_info_struct *si = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
1348 |
struct mm_struct *start_mm; |
edfe23dac swap: fix races e... |
1349 1350 1351 1352 1353 |
volatile unsigned char *swap_map; /* swap_map is accessed without * locking. Mark it as volatile * to prevent compiler doing * something odd. */ |
8d69aaee8 swap_info: swap_m... |
1354 |
unsigned char swcount; |
1da177e4c Linux-2.6.12-rc2 |
1355 1356 |
struct page *page; swp_entry_t entry; |
6eb396dc4 [PATCH] swap: swa... |
1357 |
unsigned int i = 0; |
1da177e4c Linux-2.6.12-rc2 |
1358 |
int retval = 0; |
1da177e4c Linux-2.6.12-rc2 |
1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 |
/* * When searching mms for an entry, a good strategy is to * start at the first mm we freed the previous entry from * (though actually we don't notice whether we or coincidence * freed the entry). Initialize this start_mm with a hold. * * A simpler strategy would be to start at the last mm we * freed the previous entry from; but that would take less * advantage of mmlist ordering, which clusters forked mms * together, child after parent. If we race with dup_mmap(), we * prefer to resolve parent before child, lest we miss entries * duplicated after we scanned child: using last mm would invert |
570a335b8 swap_info: swap c... |
1372 |
* that. |
1da177e4c Linux-2.6.12-rc2 |
1373 1374 1375 1376 1377 1378 1379 1380 1381 |
*/ start_mm = &init_mm; atomic_inc(&init_mm.mm_users); /* * Keep on scanning until all entries have gone. Usually, * one pass through swap_map is enough, but not necessarily: * there are races when an instance of an entry might be missed. */ |
38b5faf4b mm: frontswap: co... |
1382 |
while ((i = find_next_to_unuse(si, i, frontswap)) != 0) { |
1da177e4c Linux-2.6.12-rc2 |
1383 1384 1385 1386 |
if (signal_pending(current)) { retval = -EINTR; break; } |
886bb7e9c swapfile: remove ... |
1387 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1388 1389 |
* Get a page for the entry, using the existing swap * cache page if there is one. Otherwise, get a clean |
886bb7e9c swapfile: remove ... |
1390 |
* page and read the swap into it. |
1da177e4c Linux-2.6.12-rc2 |
1391 1392 1393 |
*/ swap_map = &si->swap_map[i]; entry = swp_entry(type, i); |
02098feaa swapin needs gfp_... |
1394 1395 |
page = read_swap_cache_async(entry, GFP_HIGHUSER_MOVABLE, NULL, 0); |
1da177e4c Linux-2.6.12-rc2 |
1396 1397 1398 1399 1400 1401 1402 |
if (!page) { /* * Either swap_duplicate() failed because entry * has been freed independently, and will not be * reused since sys_swapoff() already disabled * allocation from here, or alloc_page() failed. */ |
edfe23dac swap: fix races e... |
1403 1404 1405 1406 1407 1408 1409 1410 1411 |
swcount = *swap_map; /* * We don't hold lock here, so the swap entry could be * SWAP_MAP_BAD (when the cluster is discarding). * Instead of fail out, We can just skip the swap * entry because swapoff will wait for discarding * finish anyway. */ if (!swcount || swcount == SWAP_MAP_BAD) |
1da177e4c Linux-2.6.12-rc2 |
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 |
continue; retval = -ENOMEM; break; } /* * Don't hold on to start_mm if it looks like exiting. */ if (atomic_read(&start_mm->mm_users) == 1) { mmput(start_mm); start_mm = &init_mm; atomic_inc(&init_mm.mm_users); } /* * Wait for and lock page. When do_swap_page races with * try_to_unuse, do_swap_page can handle the fault much * faster than try_to_unuse can locate the entry. This * apparently redundant "wait_on_page_locked" lets try_to_unuse * defer to do_swap_page in such a case - in some tests, * do_swap_page and try_to_unuse repeatedly compete. */ wait_on_page_locked(page); wait_on_page_writeback(page); lock_page(page); wait_on_page_writeback(page); /* * Remove all references to entry. |
1da177e4c Linux-2.6.12-rc2 |
1441 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
1442 |
swcount = *swap_map; |
aaa468653 swap_info: note S... |
1443 1444 1445 1446 1447 1448 |
if (swap_count(swcount) == SWAP_MAP_SHMEM) { retval = shmem_unuse(entry, page); /* page has already been unlocked and released */ if (retval < 0) break; continue; |
1da177e4c Linux-2.6.12-rc2 |
1449 |
} |
aaa468653 swap_info: note S... |
1450 1451 |
if (swap_count(swcount) && start_mm != &init_mm) retval = unuse_mm(start_mm, entry, page); |
355cfa73d mm: modify swap_m... |
1452 |
if (swap_count(*swap_map)) { |
1da177e4c Linux-2.6.12-rc2 |
1453 1454 1455 1456 1457 1458 1459 1460 1461 |
int set_start_mm = (*swap_map >= swcount); struct list_head *p = &start_mm->mmlist; struct mm_struct *new_start_mm = start_mm; struct mm_struct *prev_mm = start_mm; struct mm_struct *mm; atomic_inc(&new_start_mm->mm_users); atomic_inc(&prev_mm->mm_users); spin_lock(&mmlist_lock); |
aaa468653 swap_info: note S... |
1462 |
while (swap_count(*swap_map) && !retval && |
1da177e4c Linux-2.6.12-rc2 |
1463 1464 |
(p = p->next) != &start_mm->mmlist) { mm = list_entry(p, struct mm_struct, mmlist); |
70af7c5c6 [PATCH] swapoff: ... |
1465 |
if (!atomic_inc_not_zero(&mm->mm_users)) |
1da177e4c Linux-2.6.12-rc2 |
1466 |
continue; |
1da177e4c Linux-2.6.12-rc2 |
1467 1468 1469 1470 1471 1472 1473 |
spin_unlock(&mmlist_lock); mmput(prev_mm); prev_mm = mm; cond_resched(); swcount = *swap_map; |
355cfa73d mm: modify swap_m... |
1474 |
if (!swap_count(swcount)) /* any usage ? */ |
1da177e4c Linux-2.6.12-rc2 |
1475 |
; |
aaa468653 swap_info: note S... |
1476 |
else if (mm == &init_mm) |
1da177e4c Linux-2.6.12-rc2 |
1477 |
set_start_mm = 1; |
aaa468653 swap_info: note S... |
1478 |
else |
1da177e4c Linux-2.6.12-rc2 |
1479 |
retval = unuse_mm(mm, entry, page); |
355cfa73d mm: modify swap_m... |
1480 |
|
32c5fc10e mm: remove incorr... |
1481 |
if (set_start_mm && *swap_map < swcount) { |
1da177e4c Linux-2.6.12-rc2 |
1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 |
mmput(new_start_mm); atomic_inc(&mm->mm_users); new_start_mm = mm; set_start_mm = 0; } spin_lock(&mmlist_lock); } spin_unlock(&mmlist_lock); mmput(prev_mm); mmput(start_mm); start_mm = new_start_mm; } if (retval) { unlock_page(page); page_cache_release(page); break; } /* |
1da177e4c Linux-2.6.12-rc2 |
1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 |
* If a reference remains (rare), we would like to leave * the page in the swap cache; but try_to_unmap could * then re-duplicate the entry once we drop page lock, * so we might loop indefinitely; also, that page could * not be swapped out to other storage meanwhile. So: * delete from cache even if there's another reference, * after ensuring that the data has been saved to disk - * since if the reference remains (rarer), it will be * read from disk into another page. Splitting into two * pages would be incorrect if swap supported "shared * private" pages, but they are handled by tmpfs files. |
5ad646880 ksm: let shared p... |
1512 1513 1514 1515 1516 1517 |
* * Given how unuse_vma() targets one particular offset * in an anon_vma, once the anon_vma has been determined, * this splitting happens to be just what is needed to * handle where KSM pages have been swapped out: re-reading * is unnecessarily slow, but we can fix that later on. |
1da177e4c Linux-2.6.12-rc2 |
1518 |
*/ |
355cfa73d mm: modify swap_m... |
1519 1520 |
if (swap_count(*swap_map) && PageDirty(page) && PageSwapCache(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1521 1522 1523 1524 1525 1526 1527 1528 |
struct writeback_control wbc = { .sync_mode = WB_SYNC_NONE, }; swap_writepage(page, &wbc); lock_page(page); wait_on_page_writeback(page); } |
68bdc8d64 mm: try_to_unuse ... |
1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 |
/* * It is conceivable that a racing task removed this page from * swap cache just before we acquired the page lock at the top, * or while we dropped it in unuse_mm(). The page might even * be back in swap cache on another swap area: that we must not * delete, since it may not have been written out to swap yet. */ if (PageSwapCache(page) && likely(page_private(page) == entry.val)) |
2e0e26c76 tmpfs: open a win... |
1539 |
delete_from_swap_cache(page); |
1da177e4c Linux-2.6.12-rc2 |
1540 1541 1542 1543 |
/* * So we could skip searching mms once swap count went * to 1, we did not mark any present ptes as dirty: must |
2706a1b89 vmscan: fix comme... |
1544 |
* mark page dirty so shrink_page_list will preserve it. |
1da177e4c Linux-2.6.12-rc2 |
1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 |
*/ SetPageDirty(page); unlock_page(page); page_cache_release(page); /* * Make sure that we aren't completely killing * interactive performance. */ cond_resched(); |
38b5faf4b mm: frontswap: co... |
1555 1556 1557 1558 |
if (frontswap && pages_to_unuse > 0) { if (!--pages_to_unuse) break; } |
1da177e4c Linux-2.6.12-rc2 |
1559 1560 1561 |
} mmput(start_mm); |
1da177e4c Linux-2.6.12-rc2 |
1562 1563 1564 1565 |
return retval; } /* |
5d337b919 [PATCH] swap: swa... |
1566 1567 1568 |
* After a successful try_to_unuse, if no swap is now in use, we know * we can empty the mmlist. swap_lock must be held on entry and exit. * Note that mmlist_lock nests inside swap_lock, and an mm must be |
1da177e4c Linux-2.6.12-rc2 |
1569 1570 1571 1572 1573 |
* added to the mmlist just after page_duplicate - before would be racy. */ static void drain_mmlist(void) { struct list_head *p, *next; |
efa90a981 swap_info: change... |
1574 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
1575 |
|
efa90a981 swap_info: change... |
1576 1577 |
for (type = 0; type < nr_swapfiles; type++) if (swap_info[type]->inuse_pages) |
1da177e4c Linux-2.6.12-rc2 |
1578 1579 1580 1581 1582 1583 1584 1585 1586 |
return; spin_lock(&mmlist_lock); list_for_each_safe(p, next, &init_mm.mmlist) list_del_init(p); spin_unlock(&mmlist_lock); } /* * Use this swapdev's extent info to locate the (PAGE_SIZE) block which |
d4906e1aa swap: rework map_... |
1587 1588 1589 |
* corresponds to page offset for the specified swap entry. * Note that the type of this function is sector_t, but it returns page offset * into the bdev, not sector offset. |
1da177e4c Linux-2.6.12-rc2 |
1590 |
*/ |
d4906e1aa swap: rework map_... |
1591 |
static sector_t map_swap_entry(swp_entry_t entry, struct block_device **bdev) |
1da177e4c Linux-2.6.12-rc2 |
1592 |
{ |
f29ad6a99 swap_info: privat... |
1593 1594 1595 1596 |
struct swap_info_struct *sis; struct swap_extent *start_se; struct swap_extent *se; pgoff_t offset; |
efa90a981 swap_info: change... |
1597 |
sis = swap_info[swp_type(entry)]; |
f29ad6a99 swap_info: privat... |
1598 1599 1600 1601 1602 |
*bdev = sis->bdev; offset = swp_offset(entry); start_se = sis->curr_swap_extent; se = start_se; |
1da177e4c Linux-2.6.12-rc2 |
1603 1604 1605 1606 1607 1608 1609 1610 |
for ( ; ; ) { struct list_head *lh; if (se->start_page <= offset && offset < (se->start_page + se->nr_pages)) { return se->start_block + (offset - se->start_page); } |
11d31886d [PATCH] swap: swa... |
1611 |
lh = se->list.next; |
1da177e4c Linux-2.6.12-rc2 |
1612 1613 1614 1615 1616 1617 1618 |
se = list_entry(lh, struct swap_extent, list); sis->curr_swap_extent = se; BUG_ON(se == start_se); /* It *must* be present */ } } /* |
d4906e1aa swap: rework map_... |
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 |
* Returns the page offset into bdev for the specified page's swap entry. */ sector_t map_swap_page(struct page *page, struct block_device **bdev) { swp_entry_t entry; entry.val = page_private(page); return map_swap_entry(entry, bdev); } /* |
1da177e4c Linux-2.6.12-rc2 |
1629 1630 1631 1632 |
* Free all of a swapdev's extent information */ static void destroy_swap_extents(struct swap_info_struct *sis) { |
9625a5f28 swap_info: includ... |
1633 |
while (!list_empty(&sis->first_swap_extent.list)) { |
1da177e4c Linux-2.6.12-rc2 |
1634 |
struct swap_extent *se; |
9625a5f28 swap_info: includ... |
1635 |
se = list_entry(sis->first_swap_extent.list.next, |
1da177e4c Linux-2.6.12-rc2 |
1636 1637 1638 1639 |
struct swap_extent, list); list_del(&se->list); kfree(se); } |
62c230bc1 mm: add support f... |
1640 1641 1642 1643 1644 1645 1646 1647 |
if (sis->flags & SWP_FILE) { struct file *swap_file = sis->swap_file; struct address_space *mapping = swap_file->f_mapping; sis->flags &= ~SWP_FILE; mapping->a_ops->swap_deactivate(swap_file); } |
1da177e4c Linux-2.6.12-rc2 |
1648 1649 1650 1651 |
} /* * Add a block range (and the corresponding page range) into this swapdev's |
11d31886d [PATCH] swap: swa... |
1652 |
* extent list. The extent list is kept sorted in page order. |
1da177e4c Linux-2.6.12-rc2 |
1653 |
* |
11d31886d [PATCH] swap: swa... |
1654 |
* This function rather assumes that it is called in ascending page order. |
1da177e4c Linux-2.6.12-rc2 |
1655 |
*/ |
a509bc1a9 mm: swap: impleme... |
1656 |
int |
1da177e4c Linux-2.6.12-rc2 |
1657 1658 1659 1660 1661 1662 |
add_swap_extent(struct swap_info_struct *sis, unsigned long start_page, unsigned long nr_pages, sector_t start_block) { struct swap_extent *se; struct swap_extent *new_se; struct list_head *lh; |
9625a5f28 swap_info: includ... |
1663 1664 1665 1666 1667 1668 1669 1670 1671 |
if (start_page == 0) { se = &sis->first_swap_extent; sis->curr_swap_extent = se; se->start_page = 0; se->nr_pages = nr_pages; se->start_block = start_block; return 1; } else { lh = sis->first_swap_extent.list.prev; /* Highest extent */ |
1da177e4c Linux-2.6.12-rc2 |
1672 |
se = list_entry(lh, struct swap_extent, list); |
11d31886d [PATCH] swap: swa... |
1673 1674 |
BUG_ON(se->start_page + se->nr_pages != start_page); if (se->start_block + se->nr_pages == start_block) { |
1da177e4c Linux-2.6.12-rc2 |
1675 1676 1677 1678 |
/* Merge it */ se->nr_pages += nr_pages; return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 |
} /* * No merge. Insert a new extent, preserving ordering. */ new_se = kmalloc(sizeof(*se), GFP_KERNEL); if (new_se == NULL) return -ENOMEM; new_se->start_page = start_page; new_se->nr_pages = nr_pages; new_se->start_block = start_block; |
9625a5f28 swap_info: includ... |
1690 |
list_add_tail(&new_se->list, &sis->first_swap_extent.list); |
53092a740 [PATCH] swap: sho... |
1691 |
return 1; |
1da177e4c Linux-2.6.12-rc2 |
1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 |
} /* * A `swap extent' is a simple thing which maps a contiguous range of pages * onto a contiguous range of disk blocks. An ordered list of swap extents * is built at swapon time and is then used at swap_writepage/swap_readpage * time for locating where on disk a page belongs. * * If the swapfile is an S_ISBLK block device, a single extent is installed. * This is done so that the main operating code can treat S_ISBLK and S_ISREG * swap files identically. * * Whether the swapdev is an S_ISREG file or an S_ISBLK blockdev, the swap * extent list operates in PAGE_SIZE disk blocks. Both S_ISREG and S_ISBLK * swapfiles are handled *identically* after swapon time. * * For S_ISREG swapfiles, setup_swap_extents() will walk all the file's blocks * and will parse them into an ordered extent list, in PAGE_SIZE chunks. If * some stray blocks are found which do not fall within the PAGE_SIZE alignment * requirements, they are simply tossed out - we will never use those blocks * for swapping. * |
b0d9bcd4b [PATCH] swap: upd... |
1714 |
* For S_ISREG swapfiles we set S_SWAPFILE across the life of the swapon. This |
1da177e4c Linux-2.6.12-rc2 |
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 |
* prevents root from shooting her foot off by ftruncating an in-use swapfile, * which will scribble on the fs. * * The amount of disk space which a single swap extent represents varies. * Typically it is in the 1-4 megabyte range. So we can have hundreds of * extents in the list. To avoid much list walking, we cache the previous * search location in `curr_swap_extent', and start new searches from there. * This is extremely effective. The average number of iterations in * map_swap_page() has been measured at about 0.3 per page. - akpm. */ |
53092a740 [PATCH] swap: sho... |
1725 |
static int setup_swap_extents(struct swap_info_struct *sis, sector_t *span) |
1da177e4c Linux-2.6.12-rc2 |
1726 |
{ |
62c230bc1 mm: add support f... |
1727 1728 1729 |
struct file *swap_file = sis->swap_file; struct address_space *mapping = swap_file->f_mapping; struct inode *inode = mapping->host; |
1da177e4c Linux-2.6.12-rc2 |
1730 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1731 1732 |
if (S_ISBLK(inode->i_mode)) { ret = add_swap_extent(sis, 0, sis->max, 0); |
53092a740 [PATCH] swap: sho... |
1733 |
*span = sis->pages; |
a509bc1a9 mm: swap: impleme... |
1734 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1735 |
} |
62c230bc1 mm: add support f... |
1736 |
if (mapping->a_ops->swap_activate) { |
a509bc1a9 mm: swap: impleme... |
1737 |
ret = mapping->a_ops->swap_activate(sis, swap_file, span); |
62c230bc1 mm: add support f... |
1738 1739 1740 1741 1742 |
if (!ret) { sis->flags |= SWP_FILE; ret = add_swap_extent(sis, 0, sis->max, 0); *span = sis->pages; } |
a509bc1a9 mm: swap: impleme... |
1743 |
return ret; |
62c230bc1 mm: add support f... |
1744 |
} |
a509bc1a9 mm: swap: impleme... |
1745 |
return generic_swapfile_activate(sis, swap_file, span); |
1da177e4c Linux-2.6.12-rc2 |
1746 |
} |
cf0cac0a0 mm: refactor rein... |
1747 |
static void _enable_swap_info(struct swap_info_struct *p, int prio, |
2a8f94493 swap: change bloc... |
1748 1749 |
unsigned char *swap_map, struct swap_cluster_info *cluster_info) |
40531542e sys_swapon: separ... |
1750 |
{ |
40531542e sys_swapon: separ... |
1751 1752 1753 1754 |
if (prio >= 0) p->prio = prio; else p->prio = --least_priority; |
18ab4d4ce swap: change swap... |
1755 1756 1757 1758 1759 1760 |
/* * the plist prio is negated because plist ordering is * low-to-high, while swap ordering is high-to-low */ p->list.prio = -p->prio; p->avail_list.prio = -p->prio; |
40531542e sys_swapon: separ... |
1761 |
p->swap_map = swap_map; |
2a8f94493 swap: change bloc... |
1762 |
p->cluster_info = cluster_info; |
40531542e sys_swapon: separ... |
1763 |
p->flags |= SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
1764 |
atomic_long_add(p->pages, &nr_swap_pages); |
40531542e sys_swapon: separ... |
1765 |
total_swap_pages += p->pages; |
adfab836f swap: change swap... |
1766 |
assert_spin_locked(&swap_lock); |
adfab836f swap: change swap... |
1767 |
/* |
18ab4d4ce swap: change swap... |
1768 1769 1770 1771 1772 1773 1774 1775 |
* both lists are plists, and thus priority ordered. * swap_active_head needs to be priority ordered for swapoff(), * which on removal of any swap_info_struct with an auto-assigned * (i.e. negative) priority increments the auto-assigned priority * of any lower-priority swap_info_structs. * swap_avail_head needs to be priority ordered for get_swap_page(), * which allocates swap pages from the highest available priority * swap_info_struct. |
adfab836f swap: change swap... |
1776 |
*/ |
18ab4d4ce swap: change swap... |
1777 1778 1779 1780 |
plist_add(&p->list, &swap_active_head); spin_lock(&swap_avail_lock); plist_add(&p->avail_list, &swap_avail_head); spin_unlock(&swap_avail_lock); |
cf0cac0a0 mm: refactor rein... |
1781 1782 1783 1784 |
} static void enable_swap_info(struct swap_info_struct *p, int prio, unsigned char *swap_map, |
2a8f94493 swap: change bloc... |
1785 |
struct swap_cluster_info *cluster_info, |
cf0cac0a0 mm: refactor rein... |
1786 1787 |
unsigned long *frontswap_map) { |
4f89849da frontswap: get ri... |
1788 |
frontswap_init(p->type, frontswap_map); |
cf0cac0a0 mm: refactor rein... |
1789 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1790 |
spin_lock(&p->lock); |
2a8f94493 swap: change bloc... |
1791 |
_enable_swap_info(p, prio, swap_map, cluster_info); |
ec8acf20a swap: add per-par... |
1792 |
spin_unlock(&p->lock); |
cf0cac0a0 mm: refactor rein... |
1793 1794 1795 1796 1797 1798 |
spin_unlock(&swap_lock); } static void reinsert_swap_info(struct swap_info_struct *p) { spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1799 |
spin_lock(&p->lock); |
2a8f94493 swap: change bloc... |
1800 |
_enable_swap_info(p, p->prio, p->swap_map, p->cluster_info); |
ec8acf20a swap: add per-par... |
1801 |
spin_unlock(&p->lock); |
40531542e sys_swapon: separ... |
1802 1803 |
spin_unlock(&swap_lock); } |
c4ea37c26 [CVE-2009-0029] S... |
1804 |
SYSCALL_DEFINE1(swapoff, const char __user *, specialfile) |
1da177e4c Linux-2.6.12-rc2 |
1805 |
{ |
73c34b6ac swap_info: miscel... |
1806 |
struct swap_info_struct *p = NULL; |
8d69aaee8 swap_info: swap_m... |
1807 |
unsigned char *swap_map; |
2a8f94493 swap: change bloc... |
1808 |
struct swap_cluster_info *cluster_info; |
4f89849da frontswap: get ri... |
1809 |
unsigned long *frontswap_map; |
1da177e4c Linux-2.6.12-rc2 |
1810 1811 1812 |
struct file *swap_file, *victim; struct address_space *mapping; struct inode *inode; |
91a27b2a7 vfs: define struc... |
1813 |
struct filename *pathname; |
adfab836f swap: change swap... |
1814 |
int err, found = 0; |
5b808a230 swap: fix set_blo... |
1815 |
unsigned int old_block_size; |
886bb7e9c swapfile: remove ... |
1816 |
|
1da177e4c Linux-2.6.12-rc2 |
1817 1818 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
191c54244 mm: collapse secu... |
1819 |
BUG_ON(!current->mm); |
1da177e4c Linux-2.6.12-rc2 |
1820 |
pathname = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
1821 |
if (IS_ERR(pathname)) |
f58b59c1d swapfile: fix nam... |
1822 |
return PTR_ERR(pathname); |
1da177e4c Linux-2.6.12-rc2 |
1823 |
|
669abf4e5 vfs: make path_op... |
1824 |
victim = file_open_name(pathname, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
1825 1826 1827 1828 1829 |
err = PTR_ERR(victim); if (IS_ERR(victim)) goto out; mapping = victim->f_mapping; |
5d337b919 [PATCH] swap: swa... |
1830 |
spin_lock(&swap_lock); |
18ab4d4ce swap: change swap... |
1831 |
plist_for_each_entry(p, &swap_active_head, list) { |
22c6f8fdb swapfile: remove ... |
1832 |
if (p->flags & SWP_WRITEOK) { |
adfab836f swap: change swap... |
1833 1834 |
if (p->swap_file->f_mapping == mapping) { found = 1; |
1da177e4c Linux-2.6.12-rc2 |
1835 |
break; |
adfab836f swap: change swap... |
1836 |
} |
1da177e4c Linux-2.6.12-rc2 |
1837 |
} |
1da177e4c Linux-2.6.12-rc2 |
1838 |
} |
adfab836f swap: change swap... |
1839 |
if (!found) { |
1da177e4c Linux-2.6.12-rc2 |
1840 |
err = -EINVAL; |
5d337b919 [PATCH] swap: swa... |
1841 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
1842 1843 |
goto out_dput; } |
191c54244 mm: collapse secu... |
1844 |
if (!security_vm_enough_memory_mm(current->mm, p->pages)) |
1da177e4c Linux-2.6.12-rc2 |
1845 1846 1847 |
vm_unacct_memory(p->pages); else { err = -ENOMEM; |
5d337b919 [PATCH] swap: swa... |
1848 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
1849 1850 |
goto out_dput; } |
18ab4d4ce swap: change swap... |
1851 1852 1853 |
spin_lock(&swap_avail_lock); plist_del(&p->avail_list, &swap_avail_head); spin_unlock(&swap_avail_lock); |
ec8acf20a swap: add per-par... |
1854 |
spin_lock(&p->lock); |
78ecba081 mm: fix ever-decr... |
1855 |
if (p->prio < 0) { |
adfab836f swap: change swap... |
1856 |
struct swap_info_struct *si = p; |
18ab4d4ce swap: change swap... |
1857 |
plist_for_each_entry_continue(si, &swap_active_head, list) { |
adfab836f swap: change swap... |
1858 |
si->prio++; |
18ab4d4ce swap: change swap... |
1859 1860 |
si->list.prio--; si->avail_list.prio--; |
adfab836f swap: change swap... |
1861 |
} |
78ecba081 mm: fix ever-decr... |
1862 1863 |
least_priority++; } |
18ab4d4ce swap: change swap... |
1864 |
plist_del(&p->list, &swap_active_head); |
ec8acf20a swap: add per-par... |
1865 |
atomic_long_sub(p->pages, &nr_swap_pages); |
1da177e4c Linux-2.6.12-rc2 |
1866 1867 |
total_swap_pages -= p->pages; p->flags &= ~SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
1868 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1869 |
spin_unlock(&swap_lock); |
fb4f88dca [PATCH] swap: get... |
1870 |
|
e1e12d2f3 mm, oom: fix race... |
1871 |
set_current_oom_origin(); |
adfab836f swap: change swap... |
1872 |
err = try_to_unuse(p->type, false, 0); /* force unuse all pages */ |
e1e12d2f3 mm, oom: fix race... |
1873 |
clear_current_oom_origin(); |
1da177e4c Linux-2.6.12-rc2 |
1874 |
|
1da177e4c Linux-2.6.12-rc2 |
1875 1876 |
if (err) { /* re-insert swap space back into swap_list */ |
cf0cac0a0 mm: refactor rein... |
1877 |
reinsert_swap_info(p); |
1da177e4c Linux-2.6.12-rc2 |
1878 1879 |
goto out_dput; } |
52b7efdbe [PATCH] swap: sca... |
1880 |
|
815c2c543 swap: make swap d... |
1881 |
flush_work(&p->discard_work); |
5d337b919 [PATCH] swap: swa... |
1882 |
destroy_swap_extents(p); |
570a335b8 swap_info: swap c... |
1883 1884 |
if (p->flags & SWP_CONTINUED) free_swap_count_continuations(p); |
fc0abb145 [PATCH] sem2mutex... |
1885 |
mutex_lock(&swapon_mutex); |
5d337b919 [PATCH] swap: swa... |
1886 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1887 |
spin_lock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1888 |
drain_mmlist(); |
52b7efdbe [PATCH] swap: sca... |
1889 |
/* wait for anyone still in scan_swap_map */ |
52b7efdbe [PATCH] swap: sca... |
1890 1891 |
p->highest_bit = 0; /* cuts scans short */ while (p->flags >= SWP_SCANNING) { |
ec8acf20a swap: add per-par... |
1892 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1893 |
spin_unlock(&swap_lock); |
13e4b57f6 [PATCH] mm: fix-u... |
1894 |
schedule_timeout_uninterruptible(1); |
5d337b919 [PATCH] swap: swa... |
1895 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1896 |
spin_lock(&p->lock); |
52b7efdbe [PATCH] swap: sca... |
1897 |
} |
52b7efdbe [PATCH] swap: sca... |
1898 |
|
1da177e4c Linux-2.6.12-rc2 |
1899 |
swap_file = p->swap_file; |
5b808a230 swap: fix set_blo... |
1900 |
old_block_size = p->old_block_size; |
1da177e4c Linux-2.6.12-rc2 |
1901 1902 1903 1904 |
p->swap_file = NULL; p->max = 0; swap_map = p->swap_map; p->swap_map = NULL; |
2a8f94493 swap: change bloc... |
1905 1906 |
cluster_info = p->cluster_info; p->cluster_info = NULL; |
4f89849da frontswap: get ri... |
1907 |
frontswap_map = frontswap_map_get(p); |
ec8acf20a swap: add per-par... |
1908 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1909 |
spin_unlock(&swap_lock); |
adfab836f swap: change swap... |
1910 |
frontswap_invalidate_area(p->type); |
58e97ba6b frontswap: enable... |
1911 |
frontswap_map_set(p, NULL); |
fc0abb145 [PATCH] sem2mutex... |
1912 |
mutex_unlock(&swapon_mutex); |
ebc2a1a69 swap: make cluste... |
1913 1914 |
free_percpu(p->percpu_cluster); p->percpu_cluster = NULL; |
1da177e4c Linux-2.6.12-rc2 |
1915 |
vfree(swap_map); |
2a8f94493 swap: change bloc... |
1916 |
vfree(cluster_info); |
4f89849da frontswap: get ri... |
1917 |
vfree(frontswap_map); |
2de1a7e40 mm/swapfile.c: fi... |
1918 |
/* Destroy swap account information */ |
adfab836f swap: change swap... |
1919 |
swap_cgroup_swapoff(p->type); |
27a7faa07 memcg: swap cgrou... |
1920 |
|
1da177e4c Linux-2.6.12-rc2 |
1921 1922 1923 |
inode = mapping->host; if (S_ISBLK(inode->i_mode)) { struct block_device *bdev = I_BDEV(inode); |
5b808a230 swap: fix set_blo... |
1924 |
set_blocksize(bdev, old_block_size); |
e525fd89d block: make blkde... |
1925 |
blkdev_put(bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
1926 |
} else { |
1b1dcc1b5 [PATCH] mutex sub... |
1927 |
mutex_lock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1928 |
inode->i_flags &= ~S_SWAPFILE; |
1b1dcc1b5 [PATCH] mutex sub... |
1929 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1930 1931 |
} filp_close(swap_file, NULL); |
f893ab41e mm/swap: fix race... |
1932 1933 1934 1935 1936 1937 1938 1939 1940 |
/* * Clear the SWP_USED flag after all resources are freed so that swapon * can reuse this swap_info in alloc_swap_info() safely. It is ok to * not hold p->lock after we cleared its SWP_WRITEOK. */ spin_lock(&swap_lock); p->flags = 0; spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
1941 |
err = 0; |
66d7dd518 /proc/swaps: supp... |
1942 1943 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
1da177e4c Linux-2.6.12-rc2 |
1944 1945 1946 1947 |
out_dput: filp_close(victim, NULL); out: |
f58b59c1d swapfile: fix nam... |
1948 |
putname(pathname); |
1da177e4c Linux-2.6.12-rc2 |
1949 1950 1951 1952 |
return err; } #ifdef CONFIG_PROC_FS |
66d7dd518 /proc/swaps: supp... |
1953 1954 |
static unsigned swaps_poll(struct file *file, poll_table *wait) { |
f15146380 fs: seq_file - ad... |
1955 |
struct seq_file *seq = file->private_data; |
66d7dd518 /proc/swaps: supp... |
1956 1957 |
poll_wait(file, &proc_poll_wait, wait); |
f15146380 fs: seq_file - ad... |
1958 1959 |
if (seq->poll_event != atomic_read(&proc_poll_event)) { seq->poll_event = atomic_read(&proc_poll_event); |
66d7dd518 /proc/swaps: supp... |
1960 1961 1962 1963 1964 |
return POLLIN | POLLRDNORM | POLLERR | POLLPRI; } return POLLIN | POLLRDNORM; } |
1da177e4c Linux-2.6.12-rc2 |
1965 1966 1967 |
/* iterator */ static void *swap_start(struct seq_file *swap, loff_t *pos) { |
efa90a981 swap_info: change... |
1968 1969 |
struct swap_info_struct *si; int type; |
1da177e4c Linux-2.6.12-rc2 |
1970 |
loff_t l = *pos; |
fc0abb145 [PATCH] sem2mutex... |
1971 |
mutex_lock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1972 |
|
881e4aabe [PATCH] Always pr... |
1973 1974 |
if (!l) return SEQ_START_TOKEN; |
efa90a981 swap_info: change... |
1975 1976 1977 1978 |
for (type = 0; type < nr_swapfiles; type++) { smp_rmb(); /* read nr_swapfiles before swap_info[type] */ si = swap_info[type]; if (!(si->flags & SWP_USED) || !si->swap_map) |
1da177e4c Linux-2.6.12-rc2 |
1979 |
continue; |
881e4aabe [PATCH] Always pr... |
1980 |
if (!--l) |
efa90a981 swap_info: change... |
1981 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
1982 1983 1984 1985 1986 1987 1988 |
} return NULL; } static void *swap_next(struct seq_file *swap, void *v, loff_t *pos) { |
efa90a981 swap_info: change... |
1989 1990 |
struct swap_info_struct *si = v; int type; |
1da177e4c Linux-2.6.12-rc2 |
1991 |
|
881e4aabe [PATCH] Always pr... |
1992 |
if (v == SEQ_START_TOKEN) |
efa90a981 swap_info: change... |
1993 1994 1995 |
type = 0; else type = si->type + 1; |
881e4aabe [PATCH] Always pr... |
1996 |
|
efa90a981 swap_info: change... |
1997 1998 1999 2000 |
for (; type < nr_swapfiles; type++) { smp_rmb(); /* read nr_swapfiles before swap_info[type] */ si = swap_info[type]; if (!(si->flags & SWP_USED) || !si->swap_map) |
1da177e4c Linux-2.6.12-rc2 |
2001 2002 |
continue; ++*pos; |
efa90a981 swap_info: change... |
2003 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
2004 2005 2006 2007 2008 2009 2010 |
} return NULL; } static void swap_stop(struct seq_file *swap, void *v) { |
fc0abb145 [PATCH] sem2mutex... |
2011 |
mutex_unlock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2012 2013 2014 2015 |
} static int swap_show(struct seq_file *swap, void *v) { |
efa90a981 swap_info: change... |
2016 |
struct swap_info_struct *si = v; |
1da177e4c Linux-2.6.12-rc2 |
2017 2018 |
struct file *file; int len; |
efa90a981 swap_info: change... |
2019 |
if (si == SEQ_START_TOKEN) { |
881e4aabe [PATCH] Always pr... |
2020 2021 2022 2023 |
seq_puts(swap,"Filename\t\t\t\tType\t\tSize\tUsed\tPriority "); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
2024 |
|
efa90a981 swap_info: change... |
2025 |
file = si->swap_file; |
2726d5662 vfs: add seq_file... |
2026 2027 |
len = seq_file_path(swap, file, " \t \\"); |
6eb396dc4 [PATCH] swap: swa... |
2028 2029 |
seq_printf(swap, "%*s%s\t%u\t%u\t%d ", |
886bb7e9c swapfile: remove ... |
2030 |
len < 40 ? 40 - len : 1, " ", |
496ad9aa8 new helper: file_... |
2031 |
S_ISBLK(file_inode(file)->i_mode) ? |
1da177e4c Linux-2.6.12-rc2 |
2032 |
"partition" : "file\t", |
efa90a981 swap_info: change... |
2033 2034 2035 |
si->pages << (PAGE_SHIFT - 10), si->inuse_pages << (PAGE_SHIFT - 10), si->prio); |
1da177e4c Linux-2.6.12-rc2 |
2036 2037 |
return 0; } |
15ad7cdcf [PATCH] struct se... |
2038 |
static const struct seq_operations swaps_op = { |
1da177e4c Linux-2.6.12-rc2 |
2039 2040 2041 2042 2043 2044 2045 2046 |
.start = swap_start, .next = swap_next, .stop = swap_stop, .show = swap_show }; static int swaps_open(struct inode *inode, struct file *file) { |
f15146380 fs: seq_file - ad... |
2047 |
struct seq_file *seq; |
66d7dd518 /proc/swaps: supp... |
2048 |
int ret; |
66d7dd518 /proc/swaps: supp... |
2049 |
ret = seq_open(file, &swaps_op); |
f15146380 fs: seq_file - ad... |
2050 |
if (ret) |
66d7dd518 /proc/swaps: supp... |
2051 |
return ret; |
66d7dd518 /proc/swaps: supp... |
2052 |
|
f15146380 fs: seq_file - ad... |
2053 2054 2055 |
seq = file->private_data; seq->poll_event = atomic_read(&proc_poll_event); return 0; |
1da177e4c Linux-2.6.12-rc2 |
2056 |
} |
15ad7cdcf [PATCH] struct se... |
2057 |
static const struct file_operations proc_swaps_operations = { |
1da177e4c Linux-2.6.12-rc2 |
2058 2059 2060 2061 |
.open = swaps_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, |
66d7dd518 /proc/swaps: supp... |
2062 |
.poll = swaps_poll, |
1da177e4c Linux-2.6.12-rc2 |
2063 2064 2065 2066 |
}; static int __init procswaps_init(void) { |
3d71f86f4 mm: use non-racy ... |
2067 |
proc_create("swaps", 0, NULL, &proc_swaps_operations); |
1da177e4c Linux-2.6.12-rc2 |
2068 2069 2070 2071 |
return 0; } __initcall(procswaps_init); #endif /* CONFIG_PROC_FS */ |
1796316a8 x86: consolidate ... |
2072 2073 2074 2075 2076 2077 2078 2079 |
#ifdef MAX_SWAPFILES_CHECK static int __init max_swapfiles_check(void) { MAX_SWAPFILES_CHECK(); return 0; } late_initcall(max_swapfiles_check); #endif |
53cbb2435 sys_swapon: separ... |
2080 |
static struct swap_info_struct *alloc_swap_info(void) |
1da177e4c Linux-2.6.12-rc2 |
2081 |
{ |
73c34b6ac swap_info: miscel... |
2082 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
2083 |
unsigned int type; |
efa90a981 swap_info: change... |
2084 2085 2086 |
p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) |
53cbb2435 sys_swapon: separ... |
2087 |
return ERR_PTR(-ENOMEM); |
efa90a981 swap_info: change... |
2088 |
|
5d337b919 [PATCH] swap: swa... |
2089 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
2090 2091 |
for (type = 0; type < nr_swapfiles; type++) { if (!(swap_info[type]->flags & SWP_USED)) |
1da177e4c Linux-2.6.12-rc2 |
2092 |
break; |
efa90a981 swap_info: change... |
2093 |
} |
0697212a4 [PATCH] Swapless ... |
2094 |
if (type >= MAX_SWAPFILES) { |
5d337b919 [PATCH] swap: swa... |
2095 |
spin_unlock(&swap_lock); |
efa90a981 swap_info: change... |
2096 |
kfree(p); |
730c0581c sys_swapon: simpl... |
2097 |
return ERR_PTR(-EPERM); |
1da177e4c Linux-2.6.12-rc2 |
2098 |
} |
efa90a981 swap_info: change... |
2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 |
if (type >= nr_swapfiles) { p->type = type; swap_info[type] = p; /* * Write swap_info[type] before nr_swapfiles, in case a * racing procfs swap_start() or swap_next() is reading them. * (We never shrink nr_swapfiles, we never free this entry.) */ smp_wmb(); nr_swapfiles++; } else { kfree(p); p = swap_info[type]; /* * Do not memset this entry: a racing procfs swap_next() * would be relying on p->type to remain valid. */ } |
9625a5f28 swap_info: includ... |
2117 |
INIT_LIST_HEAD(&p->first_swap_extent.list); |
18ab4d4ce swap: change swap... |
2118 2119 |
plist_node_init(&p->list, 0); plist_node_init(&p->avail_list, 0); |
1da177e4c Linux-2.6.12-rc2 |
2120 |
p->flags = SWP_USED; |
5d337b919 [PATCH] swap: swa... |
2121 |
spin_unlock(&swap_lock); |
ec8acf20a swap: add per-par... |
2122 |
spin_lock_init(&p->lock); |
efa90a981 swap_info: change... |
2123 |
|
53cbb2435 sys_swapon: separ... |
2124 |
return p; |
53cbb2435 sys_swapon: separ... |
2125 |
} |
4d0e1e107 sys_swapon: separ... |
2126 2127 2128 2129 2130 2131 2132 |
static int claim_swapfile(struct swap_info_struct *p, struct inode *inode) { int error; if (S_ISBLK(inode->i_mode)) { p->bdev = bdgrab(I_BDEV(inode)); error = blkdev_get(p->bdev, |
6f179af88 mm: fix potential... |
2133 |
FMODE_READ | FMODE_WRITE | FMODE_EXCL, p); |
4d0e1e107 sys_swapon: separ... |
2134 2135 |
if (error < 0) { p->bdev = NULL; |
6f179af88 mm: fix potential... |
2136 |
return error; |
4d0e1e107 sys_swapon: separ... |
2137 2138 2139 2140 |
} p->old_block_size = block_size(p->bdev); error = set_blocksize(p->bdev, PAGE_SIZE); if (error < 0) |
87ade72a7 sys_swapon: simpl... |
2141 |
return error; |
4d0e1e107 sys_swapon: separ... |
2142 2143 2144 2145 |
p->flags |= SWP_BLKDEV; } else if (S_ISREG(inode->i_mode)) { p->bdev = inode->i_sb->s_bdev; mutex_lock(&inode->i_mutex); |
87ade72a7 sys_swapon: simpl... |
2146 2147 2148 2149 |
if (IS_SWAPFILE(inode)) return -EBUSY; } else return -EINVAL; |
4d0e1e107 sys_swapon: separ... |
2150 2151 |
return 0; |
4d0e1e107 sys_swapon: separ... |
2152 |
} |
ca8bd38bf sys_swapon: separ... |
2153 2154 2155 2156 2157 2158 2159 |
static unsigned long read_swap_header(struct swap_info_struct *p, union swap_header *swap_header, struct inode *inode) { int i; unsigned long maxpages; unsigned long swapfilepages; |
d6bbbd29b swap: warn when a... |
2160 |
unsigned long last_page; |
ca8bd38bf sys_swapon: separ... |
2161 2162 |
if (memcmp("SWAPSPACE2", swap_header->magic.magic, 10)) { |
465c47fd8 mm/swapfile.c: co... |
2163 2164 |
pr_err("Unable to find swap-space signature "); |
387190253 sys_swapon: simpl... |
2165 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 |
} /* swap partition endianess hack... */ if (swab32(swap_header->info.version) == 1) { swab32s(&swap_header->info.version); swab32s(&swap_header->info.last_page); swab32s(&swap_header->info.nr_badpages); for (i = 0; i < swap_header->info.nr_badpages; i++) swab32s(&swap_header->info.badpages[i]); } /* Check the swap header's sub-version */ if (swap_header->info.version != 1) { |
465c47fd8 mm/swapfile.c: co... |
2178 2179 2180 |
pr_warn("Unable to handle swap header version %d ", swap_header->info.version); |
387190253 sys_swapon: simpl... |
2181 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2182 2183 2184 2185 2186 2187 2188 2189 |
} p->lowest_bit = 1; p->cluster_next = 1; p->cluster_nr = 0; /* * Find out how many pages are allowed for a single swap |
9b15b817f swap: fix shmem s... |
2190 |
* device. There are two limiting factors: 1) the number |
a2c16d6cb mm: let swap use ... |
2191 2192 |
* of bits for the swap offset in the swp_entry_t type, and * 2) the number of bits in the swap pte as defined by the |
9b15b817f swap: fix shmem s... |
2193 |
* different architectures. In order to find the |
a2c16d6cb mm: let swap use ... |
2194 |
* largest possible bit mask, a swap entry with swap type 0 |
ca8bd38bf sys_swapon: separ... |
2195 |
* and swap offset ~0UL is created, encoded to a swap pte, |
a2c16d6cb mm: let swap use ... |
2196 |
* decoded to a swp_entry_t again, and finally the swap |
ca8bd38bf sys_swapon: separ... |
2197 2198 2199 |
* offset is extracted. This will mask all the bits from * the initial ~0UL mask that can't be encoded in either * the swp_entry_t or the architecture definition of a |
9b15b817f swap: fix shmem s... |
2200 |
* swap pte. |
ca8bd38bf sys_swapon: separ... |
2201 2202 |
*/ maxpages = swp_offset(pte_to_swp_entry( |
9b15b817f swap: fix shmem s... |
2203 |
swp_entry_to_pte(swp_entry(0, ~0UL)))) + 1; |
d6bbbd29b swap: warn when a... |
2204 2205 |
last_page = swap_header->info.last_page; if (last_page > maxpages) { |
465c47fd8 mm/swapfile.c: co... |
2206 2207 |
pr_warn("Truncating oversized swap area, only using %luk out of %luk ", |
d6bbbd29b swap: warn when a... |
2208 2209 2210 2211 2212 |
maxpages << (PAGE_SHIFT - 10), last_page << (PAGE_SHIFT - 10)); } if (maxpages > last_page) { maxpages = last_page + 1; |
ca8bd38bf sys_swapon: separ... |
2213 2214 2215 2216 2217 2218 2219 |
/* p->max is an unsigned int: don't overflow it */ if ((unsigned int)maxpages == 0) maxpages = UINT_MAX; } p->highest_bit = maxpages - 1; if (!maxpages) |
387190253 sys_swapon: simpl... |
2220 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2221 2222 |
swapfilepages = i_size_read(inode) >> PAGE_SHIFT; if (swapfilepages && maxpages > swapfilepages) { |
465c47fd8 mm/swapfile.c: co... |
2223 2224 |
pr_warn("Swap area shorter than signature indicates "); |
387190253 sys_swapon: simpl... |
2225 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2226 2227 |
} if (swap_header->info.nr_badpages && S_ISREG(inode->i_mode)) |
387190253 sys_swapon: simpl... |
2228 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2229 |
if (swap_header->info.nr_badpages > MAX_SWAP_BADPAGES) |
387190253 sys_swapon: simpl... |
2230 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2231 2232 |
return maxpages; |
ca8bd38bf sys_swapon: separ... |
2233 |
} |
915d4d7bc sys_swapon: separ... |
2234 2235 2236 |
static int setup_swap_map_and_extents(struct swap_info_struct *p, union swap_header *swap_header, unsigned char *swap_map, |
2a8f94493 swap: change bloc... |
2237 |
struct swap_cluster_info *cluster_info, |
915d4d7bc sys_swapon: separ... |
2238 2239 2240 2241 |
unsigned long maxpages, sector_t *span) { int i; |
915d4d7bc sys_swapon: separ... |
2242 2243 |
unsigned int nr_good_pages; int nr_extents; |
2a8f94493 swap: change bloc... |
2244 2245 |
unsigned long nr_clusters = DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); unsigned long idx = p->cluster_next / SWAPFILE_CLUSTER; |
915d4d7bc sys_swapon: separ... |
2246 2247 |
nr_good_pages = maxpages - 1; /* omit header page */ |
2a8f94493 swap: change bloc... |
2248 2249 |
cluster_set_null(&p->free_cluster_head); cluster_set_null(&p->free_cluster_tail); |
815c2c543 swap: make swap d... |
2250 2251 |
cluster_set_null(&p->discard_cluster_head); cluster_set_null(&p->discard_cluster_tail); |
2a8f94493 swap: change bloc... |
2252 |
|
915d4d7bc sys_swapon: separ... |
2253 2254 |
for (i = 0; i < swap_header->info.nr_badpages; i++) { unsigned int page_nr = swap_header->info.badpages[i]; |
bdb8e3f68 sys_swapon: simpl... |
2255 2256 |
if (page_nr == 0 || page_nr > swap_header->info.last_page) return -EINVAL; |
915d4d7bc sys_swapon: separ... |
2257 2258 2259 |
if (page_nr < maxpages) { swap_map[page_nr] = SWAP_MAP_BAD; nr_good_pages--; |
2a8f94493 swap: change bloc... |
2260 2261 2262 2263 2264 |
/* * Haven't marked the cluster free yet, no list * operation involved */ inc_cluster_info_page(p, cluster_info, page_nr); |
915d4d7bc sys_swapon: separ... |
2265 2266 |
} } |
2a8f94493 swap: change bloc... |
2267 2268 2269 |
/* Haven't marked the cluster free yet, no list operation involved */ for (i = maxpages; i < round_up(maxpages, SWAPFILE_CLUSTER); i++) inc_cluster_info_page(p, cluster_info, i); |
915d4d7bc sys_swapon: separ... |
2270 2271 |
if (nr_good_pages) { swap_map[0] = SWAP_MAP_BAD; |
2a8f94493 swap: change bloc... |
2272 2273 2274 2275 2276 |
/* * Not mark the cluster free yet, no list * operation involved */ inc_cluster_info_page(p, cluster_info, 0); |
915d4d7bc sys_swapon: separ... |
2277 2278 2279 |
p->max = maxpages; p->pages = nr_good_pages; nr_extents = setup_swap_extents(p, span); |
bdb8e3f68 sys_swapon: simpl... |
2280 2281 |
if (nr_extents < 0) return nr_extents; |
915d4d7bc sys_swapon: separ... |
2282 2283 2284 |
nr_good_pages = p->pages; } if (!nr_good_pages) { |
465c47fd8 mm/swapfile.c: co... |
2285 2286 |
pr_warn("Empty swap-file "); |
bdb8e3f68 sys_swapon: simpl... |
2287 |
return -EINVAL; |
915d4d7bc sys_swapon: separ... |
2288 |
} |
2a8f94493 swap: change bloc... |
2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 |
if (!cluster_info) return nr_extents; for (i = 0; i < nr_clusters; i++) { if (!cluster_count(&cluster_info[idx])) { cluster_set_flag(&cluster_info[idx], CLUSTER_FLAG_FREE); if (cluster_is_null(&p->free_cluster_head)) { cluster_set_next_flag(&p->free_cluster_head, idx, 0); cluster_set_next_flag(&p->free_cluster_tail, idx, 0); } else { unsigned int tail; tail = cluster_next(&p->free_cluster_tail); cluster_set_next(&cluster_info[tail], idx); cluster_set_next_flag(&p->free_cluster_tail, idx, 0); } } idx++; if (idx == nr_clusters) idx = 0; } |
915d4d7bc sys_swapon: separ... |
2313 |
return nr_extents; |
915d4d7bc sys_swapon: separ... |
2314 |
} |
dcf6b7ddd swap: discard whi... |
2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 |
/* * Helper to sys_swapon determining if a given swap * backing device queue supports DISCARD operations. */ static bool swap_discardable(struct swap_info_struct *si) { struct request_queue *q = bdev_get_queue(si->bdev); if (!q || !blk_queue_discard(q)) return false; return true; } |
53cbb2435 sys_swapon: separ... |
2328 2329 2330 |
SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) { struct swap_info_struct *p; |
91a27b2a7 vfs: define struc... |
2331 |
struct filename *name; |
53cbb2435 sys_swapon: separ... |
2332 2333 |
struct file *swap_file = NULL; struct address_space *mapping; |
40531542e sys_swapon: separ... |
2334 |
int prio; |
53cbb2435 sys_swapon: separ... |
2335 2336 |
int error; union swap_header *swap_header; |
915d4d7bc sys_swapon: separ... |
2337 |
int nr_extents; |
53cbb2435 sys_swapon: separ... |
2338 2339 |
sector_t span; unsigned long maxpages; |
53cbb2435 sys_swapon: separ... |
2340 |
unsigned char *swap_map = NULL; |
2a8f94493 swap: change bloc... |
2341 |
struct swap_cluster_info *cluster_info = NULL; |
38b5faf4b mm: frontswap: co... |
2342 |
unsigned long *frontswap_map = NULL; |
53cbb2435 sys_swapon: separ... |
2343 2344 |
struct page *page = NULL; struct inode *inode = NULL; |
53cbb2435 sys_swapon: separ... |
2345 |
|
d15cab975 swapon: check val... |
2346 2347 |
if (swap_flags & ~SWAP_FLAGS_VALID) return -EINVAL; |
53cbb2435 sys_swapon: separ... |
2348 2349 2350 2351 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; p = alloc_swap_info(); |
2542e5134 sys_swapon: simpl... |
2352 2353 |
if (IS_ERR(p)) return PTR_ERR(p); |
53cbb2435 sys_swapon: separ... |
2354 |
|
815c2c543 swap: make swap d... |
2355 |
INIT_WORK(&p->discard_work, swap_discard_work); |
1da177e4c Linux-2.6.12-rc2 |
2356 |
name = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
2357 |
if (IS_ERR(name)) { |
7de7fb6b3 sys_swapon: move ... |
2358 |
error = PTR_ERR(name); |
1da177e4c Linux-2.6.12-rc2 |
2359 |
name = NULL; |
bd69010b0 sys_swapon: use a... |
2360 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2361 |
} |
669abf4e5 vfs: make path_op... |
2362 |
swap_file = file_open_name(name, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
2363 |
if (IS_ERR(swap_file)) { |
7de7fb6b3 sys_swapon: move ... |
2364 |
error = PTR_ERR(swap_file); |
1da177e4c Linux-2.6.12-rc2 |
2365 |
swap_file = NULL; |
bd69010b0 sys_swapon: use a... |
2366 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2367 2368 2369 2370 |
} p->swap_file = swap_file; mapping = swap_file->f_mapping; |
2130781e2 sys_swapon: fix i... |
2371 |
inode = mapping->host; |
6f179af88 mm: fix potential... |
2372 |
|
2130781e2 sys_swapon: fix i... |
2373 |
/* If S_ISREG(inode->i_mode) will do mutex_lock(&inode->i_mutex); */ |
4d0e1e107 sys_swapon: separ... |
2374 2375 |
error = claim_swapfile(p, inode); if (unlikely(error)) |
1da177e4c Linux-2.6.12-rc2 |
2376 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2377 |
|
1da177e4c Linux-2.6.12-rc2 |
2378 2379 2380 2381 2382 2383 2384 |
/* * Read the swap header. */ if (!mapping->a_ops->readpage) { error = -EINVAL; goto bad_swap; } |
090d2b185 [PATCH] read_mapp... |
2385 |
page = read_mapping_page(mapping, 0, swap_file); |
1da177e4c Linux-2.6.12-rc2 |
2386 2387 2388 2389 |
if (IS_ERR(page)) { error = PTR_ERR(page); goto bad_swap; } |
81e339712 swapfile: remove ... |
2390 |
swap_header = kmap(page); |
1da177e4c Linux-2.6.12-rc2 |
2391 |
|
ca8bd38bf sys_swapon: separ... |
2392 2393 |
maxpages = read_swap_header(p, swap_header, inode); if (unlikely(!maxpages)) { |
1da177e4c Linux-2.6.12-rc2 |
2394 2395 2396 |
error = -EINVAL; goto bad_swap; } |
886bb7e9c swapfile: remove ... |
2397 |
|
81e339712 swapfile: remove ... |
2398 |
/* OK, set up the swap map and apply the bad block list */ |
803d0c835 sys_swapon: use v... |
2399 |
swap_map = vzalloc(maxpages); |
81e339712 swapfile: remove ... |
2400 2401 2402 2403 |
if (!swap_map) { error = -ENOMEM; goto bad_swap; } |
2a8f94493 swap: change bloc... |
2404 |
if (p->bdev && blk_queue_nonrot(bdev_get_queue(p->bdev))) { |
6f179af88 mm: fix potential... |
2405 |
int cpu; |
2a8f94493 swap: change bloc... |
2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 |
p->flags |= SWP_SOLIDSTATE; /* * select a random position to start with to help wear leveling * SSD */ p->cluster_next = 1 + (prandom_u32() % p->highest_bit); cluster_info = vzalloc(DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER) * sizeof(*cluster_info)); if (!cluster_info) { error = -ENOMEM; goto bad_swap; } |
ebc2a1a69 swap: make cluste... |
2419 2420 2421 2422 2423 |
p->percpu_cluster = alloc_percpu(struct percpu_cluster); if (!p->percpu_cluster) { error = -ENOMEM; goto bad_swap; } |
6f179af88 mm: fix potential... |
2424 |
for_each_possible_cpu(cpu) { |
ebc2a1a69 swap: make cluste... |
2425 |
struct percpu_cluster *cluster; |
6f179af88 mm: fix potential... |
2426 |
cluster = per_cpu_ptr(p->percpu_cluster, cpu); |
ebc2a1a69 swap: make cluste... |
2427 2428 |
cluster_set_null(&cluster->index); } |
2a8f94493 swap: change bloc... |
2429 |
} |
1da177e4c Linux-2.6.12-rc2 |
2430 |
|
1421ef3cd sys_swapon: call ... |
2431 2432 2433 |
error = swap_cgroup_swapon(p->type, maxpages); if (error) goto bad_swap; |
915d4d7bc sys_swapon: separ... |
2434 |
nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map, |
2a8f94493 swap: change bloc... |
2435 |
cluster_info, maxpages, &span); |
915d4d7bc sys_swapon: separ... |
2436 2437 |
if (unlikely(nr_extents < 0)) { error = nr_extents; |
1da177e4c Linux-2.6.12-rc2 |
2438 2439 |
goto bad_swap; } |
38b5faf4b mm: frontswap: co... |
2440 2441 |
/* frontswap enabled? set up bit-per-page map for frontswap */ if (frontswap_enabled) |
7b57976da frontswap: fix in... |
2442 |
frontswap_map = vzalloc(BITS_TO_LONGS(maxpages) * sizeof(long)); |
1da177e4c Linux-2.6.12-rc2 |
2443 |
|
2a8f94493 swap: change bloc... |
2444 2445 2446 2447 2448 2449 2450 2451 2452 |
if (p->bdev &&(swap_flags & SWAP_FLAG_DISCARD) && swap_discardable(p)) { /* * When discard is enabled for swap with no particular * policy flagged, we set all swap discard flags here in * order to sustain backward compatibility with older * swapon(8) releases. */ p->flags |= (SWP_DISCARDABLE | SWP_AREA_DISCARD | SWP_PAGE_DISCARD); |
dcf6b7ddd swap: discard whi... |
2453 |
|
2a8f94493 swap: change bloc... |
2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 |
/* * By flagging sys_swapon, a sysadmin can tell us to * either do single-time area discards only, or to just * perform discards for released swap page-clusters. * Now it's time to adjust the p->flags accordingly. */ if (swap_flags & SWAP_FLAG_DISCARD_ONCE) p->flags &= ~SWP_PAGE_DISCARD; else if (swap_flags & SWAP_FLAG_DISCARD_PAGES) p->flags &= ~SWP_AREA_DISCARD; /* issue a swapon-time discard if it's still required */ if (p->flags & SWP_AREA_DISCARD) { int err = discard_swap(p); if (unlikely(err)) pr_err("swapon: discard_swap(%p): %d ", p, err); |
dcf6b7ddd swap: discard whi... |
2472 |
} |
20137a490 swapfile: swapon ... |
2473 |
} |
6a6ba8317 swapfile: swapon ... |
2474 |
|
fc0abb145 [PATCH] sem2mutex... |
2475 |
mutex_lock(&swapon_mutex); |
40531542e sys_swapon: separ... |
2476 |
prio = -1; |
78ecba081 mm: fix ever-decr... |
2477 |
if (swap_flags & SWAP_FLAG_PREFER) |
40531542e sys_swapon: separ... |
2478 |
prio = |
78ecba081 mm: fix ever-decr... |
2479 |
(swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT; |
2a8f94493 swap: change bloc... |
2480 |
enable_swap_info(p, prio, swap_map, cluster_info, frontswap_map); |
c69dbfb84 sys_swapon: move ... |
2481 |
|
465c47fd8 mm/swapfile.c: co... |
2482 |
pr_info("Adding %uk swap on %s. " |
dcf6b7ddd swap: discard whi... |
2483 2484 |
"Priority:%d extents:%d across:%lluk %s%s%s%s%s ", |
91a27b2a7 vfs: define struc... |
2485 |
p->pages<<(PAGE_SHIFT-10), name->name, p->prio, |
c69dbfb84 sys_swapon: move ... |
2486 2487 |
nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10), (p->flags & SWP_SOLIDSTATE) ? "SS" : "", |
38b5faf4b mm: frontswap: co... |
2488 |
(p->flags & SWP_DISCARDABLE) ? "D" : "", |
dcf6b7ddd swap: discard whi... |
2489 2490 |
(p->flags & SWP_AREA_DISCARD) ? "s" : "", (p->flags & SWP_PAGE_DISCARD) ? "c" : "", |
38b5faf4b mm: frontswap: co... |
2491 |
(frontswap_map) ? "FS" : ""); |
c69dbfb84 sys_swapon: move ... |
2492 |
|
fc0abb145 [PATCH] sem2mutex... |
2493 |
mutex_unlock(&swapon_mutex); |
66d7dd518 /proc/swaps: supp... |
2494 2495 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
9b01c350a sys_swapon: do on... |
2496 2497 |
if (S_ISREG(inode->i_mode)) inode->i_flags |= S_SWAPFILE; |
1da177e4c Linux-2.6.12-rc2 |
2498 2499 2500 |
error = 0; goto out; bad_swap: |
ebc2a1a69 swap: make cluste... |
2501 2502 |
free_percpu(p->percpu_cluster); p->percpu_cluster = NULL; |
bd69010b0 sys_swapon: use a... |
2503 |
if (inode && S_ISBLK(inode->i_mode) && p->bdev) { |
f2090d2df sys_swapon: remov... |
2504 2505 |
set_blocksize(p->bdev, p->old_block_size); blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
2506 |
} |
4cd3bb10f [PATCH] swap: mov... |
2507 |
destroy_swap_extents(p); |
e8e6c2ec4 sys_swapon: do no... |
2508 |
swap_cgroup_swapoff(p->type); |
5d337b919 [PATCH] swap: swa... |
2509 |
spin_lock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2510 |
p->swap_file = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2511 |
p->flags = 0; |
5d337b919 [PATCH] swap: swa... |
2512 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2513 |
vfree(swap_map); |
2a8f94493 swap: change bloc... |
2514 |
vfree(cluster_info); |
52c50567d mm: swap: unlock ... |
2515 |
if (swap_file) { |
2130781e2 sys_swapon: fix i... |
2516 |
if (inode && S_ISREG(inode->i_mode)) { |
52c50567d mm: swap: unlock ... |
2517 |
mutex_unlock(&inode->i_mutex); |
2130781e2 sys_swapon: fix i... |
2518 2519 |
inode = NULL; } |
1da177e4c Linux-2.6.12-rc2 |
2520 |
filp_close(swap_file, NULL); |
52c50567d mm: swap: unlock ... |
2521 |
} |
1da177e4c Linux-2.6.12-rc2 |
2522 2523 2524 2525 2526 2527 2528 |
out: if (page && !IS_ERR(page)) { kunmap(page); page_cache_release(page); } if (name) putname(name); |
9b01c350a sys_swapon: do on... |
2529 |
if (inode && S_ISREG(inode->i_mode)) |
1b1dcc1b5 [PATCH] mutex sub... |
2530 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2531 2532 2533 2534 2535 |
return error; } void si_swapinfo(struct sysinfo *val) { |
efa90a981 swap_info: change... |
2536 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
2537 |
unsigned long nr_to_be_unused = 0; |
5d337b919 [PATCH] swap: swa... |
2538 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
2539 2540 2541 2542 2543 |
for (type = 0; type < nr_swapfiles; type++) { struct swap_info_struct *si = swap_info[type]; if ((si->flags & SWP_USED) && !(si->flags & SWP_WRITEOK)) nr_to_be_unused += si->inuse_pages; |
1da177e4c Linux-2.6.12-rc2 |
2544 |
} |
ec8acf20a swap: add per-par... |
2545 |
val->freeswap = atomic_long_read(&nr_swap_pages) + nr_to_be_unused; |
1da177e4c Linux-2.6.12-rc2 |
2546 |
val->totalswap = total_swap_pages + nr_to_be_unused; |
5d337b919 [PATCH] swap: swa... |
2547 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2548 2549 2550 2551 2552 |
} /* * Verify that a swap entry is valid and increment its swap map count. * |
355cfa73d mm: modify swap_m... |
2553 2554 2555 2556 2557 2558 |
* Returns error code in following case. * - success -> 0 * - swp_entry is invalid -> EINVAL * - swp_entry is migration entry -> EINVAL * - swap-cache reference is requested but there is already one. -> EEXIST * - swap-cache reference is requested but the entry is not used. -> ENOENT |
570a335b8 swap_info: swap c... |
2559 |
* - swap-mapped reference requested but needs continued swap count. -> ENOMEM |
1da177e4c Linux-2.6.12-rc2 |
2560 |
*/ |
8d69aaee8 swap_info: swap_m... |
2561 |
static int __swap_duplicate(swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
2562 |
{ |
73c34b6ac swap_info: miscel... |
2563 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
2564 |
unsigned long offset, type; |
8d69aaee8 swap_info: swap_m... |
2565 2566 |
unsigned char count; unsigned char has_cache; |
253d553ba swap_info: SWAP_H... |
2567 |
int err = -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2568 |
|
a7420aa54 HWPOISON: Add sup... |
2569 |
if (non_swap_entry(entry)) |
253d553ba swap_info: SWAP_H... |
2570 |
goto out; |
0697212a4 [PATCH] Swapless ... |
2571 |
|
1da177e4c Linux-2.6.12-rc2 |
2572 2573 2574 |
type = swp_type(entry); if (type >= nr_swapfiles) goto bad_file; |
efa90a981 swap_info: change... |
2575 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
2576 |
offset = swp_offset(entry); |
ec8acf20a swap: add per-par... |
2577 |
spin_lock(&p->lock); |
355cfa73d mm: modify swap_m... |
2578 2579 |
if (unlikely(offset >= p->max)) goto unlock_out; |
253d553ba swap_info: SWAP_H... |
2580 |
count = p->swap_map[offset]; |
edfe23dac swap: fix races e... |
2581 2582 2583 2584 2585 2586 2587 2588 2589 |
/* * swapin_readahead() doesn't check if a swap entry is valid, so the * swap entry could be SWAP_MAP_BAD. Check here with lock held. */ if (unlikely(swap_count(count) == SWAP_MAP_BAD)) { err = -ENOENT; goto unlock_out; } |
253d553ba swap_info: SWAP_H... |
2590 2591 2592 |
has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; err = 0; |
355cfa73d mm: modify swap_m... |
2593 |
|
253d553ba swap_info: SWAP_H... |
2594 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
2595 2596 |
/* set SWAP_HAS_CACHE if there is no cache and entry is used */ |
253d553ba swap_info: SWAP_H... |
2597 2598 2599 2600 2601 2602 |
if (!has_cache && count) has_cache = SWAP_HAS_CACHE; else if (has_cache) /* someone else added cache */ err = -EEXIST; else /* no users remaining */ err = -ENOENT; |
355cfa73d mm: modify swap_m... |
2603 2604 |
} else if (count || has_cache) { |
253d553ba swap_info: SWAP_H... |
2605 |
|
570a335b8 swap_info: swap c... |
2606 2607 2608 |
if ((count & ~COUNT_CONTINUED) < SWAP_MAP_MAX) count += usage; else if ((count & ~COUNT_CONTINUED) > SWAP_MAP_MAX) |
253d553ba swap_info: SWAP_H... |
2609 |
err = -EINVAL; |
570a335b8 swap_info: swap c... |
2610 2611 2612 2613 |
else if (swap_count_continued(p, offset, count)) count = COUNT_CONTINUED; else err = -ENOMEM; |
355cfa73d mm: modify swap_m... |
2614 |
} else |
253d553ba swap_info: SWAP_H... |
2615 2616 2617 |
err = -ENOENT; /* unused swap entry */ p->swap_map[offset] = count | has_cache; |
355cfa73d mm: modify swap_m... |
2618 |
unlock_out: |
ec8acf20a swap: add per-par... |
2619 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
2620 |
out: |
253d553ba swap_info: SWAP_H... |
2621 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
2622 2623 |
bad_file: |
465c47fd8 mm/swapfile.c: co... |
2624 2625 |
pr_err("swap_dup: %s%08lx ", Bad_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
2626 2627 |
goto out; } |
253d553ba swap_info: SWAP_H... |
2628 |
|
355cfa73d mm: modify swap_m... |
2629 |
/* |
aaa468653 swap_info: note S... |
2630 2631 2632 2633 2634 2635 2636 2637 2638 |
* Help swapoff by noting that swap entry belongs to shmem/tmpfs * (in which case its reference count is never incremented). */ void swap_shmem_alloc(swp_entry_t entry) { __swap_duplicate(entry, SWAP_MAP_SHMEM); } /* |
08259d58e mm: add comment o... |
2639 2640 2641 2642 2643 |
* Increase reference count of swap entry by 1. * Returns 0 for success, or -ENOMEM if a swap_count_continuation is required * but could not be atomically allocated. Returns 0, just as if it succeeded, * if __swap_duplicate() fails for another reason (-EINVAL or -ENOENT), which * might occur if a page table entry has got corrupted. |
355cfa73d mm: modify swap_m... |
2644 |
*/ |
570a335b8 swap_info: swap c... |
2645 |
int swap_duplicate(swp_entry_t entry) |
355cfa73d mm: modify swap_m... |
2646 |
{ |
570a335b8 swap_info: swap c... |
2647 2648 2649 2650 2651 |
int err = 0; while (!err && __swap_duplicate(entry, 1) == -ENOMEM) err = add_swap_count_continuation(entry, GFP_ATOMIC); return err; |
355cfa73d mm: modify swap_m... |
2652 |
} |
1da177e4c Linux-2.6.12-rc2 |
2653 |
|
cb4b86ba4 mm: add swap cach... |
2654 |
/* |
355cfa73d mm: modify swap_m... |
2655 2656 |
* @entry: swap entry for which we allocate swap cache. * |
73c34b6ac swap_info: miscel... |
2657 |
* Called when allocating swap cache for existing swap entry, |
355cfa73d mm: modify swap_m... |
2658 2659 2660 |
* This can return error codes. Returns 0 at success. * -EBUSY means there is a swap cache. * Note: return code is different from swap_duplicate(). |
cb4b86ba4 mm: add swap cach... |
2661 2662 2663 |
*/ int swapcache_prepare(swp_entry_t entry) { |
253d553ba swap_info: SWAP_H... |
2664 |
return __swap_duplicate(entry, SWAP_HAS_CACHE); |
cb4b86ba4 mm: add swap cach... |
2665 |
} |
f981c5950 mm: methods for t... |
2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 |
struct swap_info_struct *page_swap_info(struct page *page) { swp_entry_t swap = { .val = page_private(page) }; BUG_ON(!PageSwapCache(page)); return swap_info[swp_type(swap)]; } /* * out-of-line __page_file_ methods to avoid include hell. */ struct address_space *__page_file_mapping(struct page *page) { |
309381fea mm: dump page whe... |
2678 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
f981c5950 mm: methods for t... |
2679 2680 2681 2682 2683 2684 2685 |
return page_swap_info(page)->swap_file->f_mapping; } EXPORT_SYMBOL_GPL(__page_file_mapping); pgoff_t __page_file_index(struct page *page) { swp_entry_t swap = { .val = page_private(page) }; |
309381fea mm: dump page whe... |
2686 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
f981c5950 mm: methods for t... |
2687 2688 2689 |
return swp_offset(swap); } EXPORT_SYMBOL_GPL(__page_file_index); |
1da177e4c Linux-2.6.12-rc2 |
2690 |
/* |
570a335b8 swap_info: swap c... |
2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 |
* add_swap_count_continuation - called when a swap count is duplicated * beyond SWAP_MAP_MAX, it allocates a new page and links that to the entry's * page of the original vmalloc'ed swap_map, to hold the continuation count * (for that entry and for its neighbouring PAGE_SIZE swap entries). Called * again when count is duplicated beyond SWAP_MAP_MAX * SWAP_CONT_MAX, etc. * * These continuation pages are seldom referenced: the common paths all work * on the original swap_map, only referring to a continuation page when the * low "digit" of a count is incremented or decremented through SWAP_MAP_MAX. * * add_swap_count_continuation(, GFP_ATOMIC) can be called while holding * page table locks; if it fails, add_swap_count_continuation(, GFP_KERNEL) * can be called after dropping locks. */ int add_swap_count_continuation(swp_entry_t entry, gfp_t gfp_mask) { struct swap_info_struct *si; struct page *head; struct page *page; struct page *list_page; pgoff_t offset; unsigned char count; /* * When debugging, it's easier to use __GFP_ZERO here; but it's better * for latency not to zero a page while GFP_ATOMIC and holding locks. */ page = alloc_page(gfp_mask | __GFP_HIGHMEM); si = swap_info_get(entry); if (!si) { /* * An acceptable race has occurred since the failing * __swap_duplicate(): the swap entry has been freed, * perhaps even the whole swap_map cleared for swapoff. */ goto outer; } offset = swp_offset(entry); count = si->swap_map[offset] & ~SWAP_HAS_CACHE; if ((count & ~COUNT_CONTINUED) != SWAP_MAP_MAX) { /* * The higher the swap count, the more likely it is that tasks * will race to add swap count continuation: we need to avoid * over-provisioning. */ goto out; } if (!page) { |
ec8acf20a swap: add per-par... |
2743 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
2744 2745 2746 2747 2748 |
return -ENOMEM; } /* * We are fortunate that although vmalloc_to_page uses pte_offset_map, |
2de1a7e40 mm/swapfile.c: fi... |
2749 2750 |
* no architecture is using highmem pages for kernel page tables: so it * will not corrupt the GFP_ATOMIC caller's atomic page table kmaps. |
570a335b8 swap_info: swap c... |
2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 |
*/ head = vmalloc_to_page(si->swap_map + offset); offset &= ~PAGE_MASK; /* * Page allocation does not initialize the page's lru field, * but it does always reset its private field. */ if (!page_private(head)) { BUG_ON(count & COUNT_CONTINUED); INIT_LIST_HEAD(&head->lru); set_page_private(head, SWP_CONTINUED); si->flags |= SWP_CONTINUED; } list_for_each_entry(list_page, &head->lru, lru) { unsigned char *map; /* * If the previous map said no continuation, but we've found * a continuation page, free our allocation and use this one. */ if (!(count & COUNT_CONTINUED)) goto out; |
9b04c5fec mm: remove the se... |
2775 |
map = kmap_atomic(list_page) + offset; |
570a335b8 swap_info: swap c... |
2776 |
count = *map; |
9b04c5fec mm: remove the se... |
2777 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 |
/* * If this continuation count now has some space in it, * free our allocation and use this one. */ if ((count & ~COUNT_CONTINUED) != SWAP_CONT_MAX) goto out; } list_add_tail(&page->lru, &head->lru); page = NULL; /* now it's attached, don't free it */ out: |
ec8acf20a swap: add per-par... |
2790 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 |
outer: if (page) __free_page(page); return 0; } /* * swap_count_continued - when the original swap_map count is incremented * from SWAP_MAP_MAX, check if there is already a continuation page to carry * into, carry if so, or else fail until a new continuation page is allocated; * when the original swap_map count is decremented from 0 with continuation, * borrow from the continuation and report whether it still holds more. * Called while __swap_duplicate() or swap_entry_free() holds swap_lock. */ static bool swap_count_continued(struct swap_info_struct *si, pgoff_t offset, unsigned char count) { struct page *head; struct page *page; unsigned char *map; head = vmalloc_to_page(si->swap_map + offset); if (page_private(head) != SWP_CONTINUED) { BUG_ON(count & COUNT_CONTINUED); return false; /* need to add count continuation */ } offset &= ~PAGE_MASK; page = list_entry(head->lru.next, struct page, lru); |
9b04c5fec mm: remove the se... |
2820 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2821 2822 2823 2824 2825 2826 2827 2828 2829 |
if (count == SWAP_MAP_MAX) /* initial increment from swap_map */ goto init_map; /* jump over SWAP_CONT_MAX checks */ if (count == (SWAP_MAP_MAX | COUNT_CONTINUED)) { /* incrementing */ /* * Think of how you add 1 to 999 */ while (*map == (SWAP_CONT_MAX | COUNT_CONTINUED)) { |
9b04c5fec mm: remove the se... |
2830 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2831 2832 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
2833 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2834 2835 |
} if (*map == SWAP_CONT_MAX) { |
9b04c5fec mm: remove the se... |
2836 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2837 2838 2839 |
page = list_entry(page->lru.next, struct page, lru); if (page == head) return false; /* add count continuation */ |
9b04c5fec mm: remove the se... |
2840 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2841 2842 2843 |
init_map: *map = 0; /* we didn't zero the page */ } *map += 1; |
9b04c5fec mm: remove the se... |
2844 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2845 2846 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
2847 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2848 |
*map = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
2849 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 |
page = list_entry(page->lru.prev, struct page, lru); } return true; /* incremented */ } else { /* decrementing */ /* * Think of how you subtract 1 from 1000 */ BUG_ON(count != COUNT_CONTINUED); while (*map == COUNT_CONTINUED) { |
9b04c5fec mm: remove the se... |
2860 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2861 2862 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
2863 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2864 2865 2866 2867 2868 |
} BUG_ON(*map == 0); *map -= 1; if (*map == 0) count = 0; |
9b04c5fec mm: remove the se... |
2869 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2870 2871 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
2872 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2873 2874 |
*map = SWAP_CONT_MAX | count; count = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
2875 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 |
page = list_entry(page->lru.prev, struct page, lru); } return count == COUNT_CONTINUED; } } /* * free_swap_count_continuations - swapoff free all the continuation pages * appended to the swap_map, after swap_map is quiesced, before vfree'ing it. */ static void free_swap_count_continuations(struct swap_info_struct *si) { pgoff_t offset; for (offset = 0; offset < si->max; offset += PAGE_SIZE) { struct page *head; head = vmalloc_to_page(si->swap_map + offset); if (page_private(head)) { struct list_head *this, *next; list_for_each_safe(this, next, &head->lru) { struct page *page; page = list_entry(this, struct page, lru); list_del(this); __free_page(page); } } } } |