Blame view
mm/swapfile.c
94.9 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 |
#include <linux/mm.h> |
6e84f3152 sched/headers: Pr... |
8 |
#include <linux/sched/mm.h> |
299300258 sched/headers: Pr... |
9 |
#include <linux/sched/task.h> |
1da177e4c Linux-2.6.12-rc2 |
10 11 12 13 14 15 16 17 |
#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... |
18 |
#include <linux/shmem_fs.h> |
1da177e4c Linux-2.6.12-rc2 |
19 |
#include <linux/blkdev.h> |
20137a490 swapfile: swapon ... |
20 |
#include <linux/random.h> |
1da177e4c Linux-2.6.12-rc2 |
21 22 23 24 |
#include <linux/writeback.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/init.h> |
5ad646880 ksm: let shared p... |
25 |
#include <linux/ksm.h> |
1da177e4c Linux-2.6.12-rc2 |
26 27 28 |
#include <linux/rmap.h> #include <linux/security.h> #include <linux/backing-dev.h> |
fc0abb145 [PATCH] sem2mutex... |
29 |
#include <linux/mutex.h> |
c59ede7b7 [PATCH] move capa... |
30 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
31 |
#include <linux/syscalls.h> |
8a9f3ccd2 Memory controller... |
32 |
#include <linux/memcontrol.h> |
66d7dd518 /proc/swaps: supp... |
33 |
#include <linux/poll.h> |
72788c385 oom: replace PF_O... |
34 |
#include <linux/oom.h> |
38b5faf4b mm: frontswap: co... |
35 36 |
#include <linux/frontswap.h> #include <linux/swapfile.h> |
f981c5950 mm: methods for t... |
37 |
#include <linux/export.h> |
67afa38e0 mm/swap: add cach... |
38 |
#include <linux/swap_slots.h> |
155b5f88e mm/swapfile.c: so... |
39 |
#include <linux/sort.h> |
1da177e4c Linux-2.6.12-rc2 |
40 41 42 43 |
#include <asm/pgtable.h> #include <asm/tlbflush.h> #include <linux/swapops.h> |
5d1ea48bd mm: page_cgroup: ... |
44 |
#include <linux/swap_cgroup.h> |
1da177e4c Linux-2.6.12-rc2 |
45 |
|
570a335b8 swap_info: swap c... |
46 47 48 |
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_... |
49 |
static sector_t map_swap_entry(swp_entry_t, struct block_device**); |
570a335b8 swap_info: swap c... |
50 |
|
38b5faf4b mm: frontswap: co... |
51 |
DEFINE_SPINLOCK(swap_lock); |
7c363b8c6 mm/swapfile.c: ma... |
52 |
static unsigned int nr_swapfiles; |
ec8acf20a swap: add per-par... |
53 |
atomic_long_t nr_swap_pages; |
fb0fec501 mm: Export nr_swa... |
54 55 56 57 58 59 |
/* * Some modules use swappable objects and may try to swap them out under * memory pressure (via the shrinker). Before doing so, they may wish to * check to see if any swap space is available. */ EXPORT_SYMBOL_GPL(nr_swap_pages); |
ec8acf20a swap: add per-par... |
60 |
/* protected with swap_lock. reading in vm_swap_full() doesn't need lock */ |
1da177e4c Linux-2.6.12-rc2 |
61 |
long total_swap_pages; |
a2468cc9b swap: choose swap... |
62 |
static int least_priority = -1; |
1da177e4c Linux-2.6.12-rc2 |
63 |
|
1da177e4c Linux-2.6.12-rc2 |
64 65 66 67 |
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... |
68 69 70 71 |
/* * all active swap_info_structs * protected with swap_lock, and ordered by priority. */ |
18ab4d4ce swap: change swap... |
72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
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. */ |
a2468cc9b swap: choose swap... |
86 |
struct plist_head *swap_avail_heads; |
18ab4d4ce swap: change swap... |
87 |
static DEFINE_SPINLOCK(swap_avail_lock); |
1da177e4c Linux-2.6.12-rc2 |
88 |
|
38b5faf4b mm: frontswap: co... |
89 |
struct swap_info_struct *swap_info[MAX_SWAPFILES]; |
1da177e4c Linux-2.6.12-rc2 |
90 |
|
fc0abb145 [PATCH] sem2mutex... |
91 |
static DEFINE_MUTEX(swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
92 |
|
66d7dd518 /proc/swaps: supp... |
93 94 95 |
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); |
81a0298bd mm, swap: don't u... |
96 |
atomic_t nr_rotate_swap = ATOMIC_INIT(0); |
8d69aaee8 swap_info: swap_m... |
97 |
static inline unsigned char swap_count(unsigned char ent) |
355cfa73d mm: modify swap_m... |
98 |
{ |
570a335b8 swap_info: swap c... |
99 |
return ent & ~SWAP_HAS_CACHE; /* may include SWAP_HAS_CONT flag */ |
355cfa73d mm: modify swap_m... |
100 |
} |
efa90a981 swap_info: change... |
101 |
/* returns 1 if swap entry is freed */ |
c9e444103 mm: reuse unused ... |
102 103 104 |
static int __try_to_reclaim_swap(struct swap_info_struct *si, unsigned long offset) { |
efa90a981 swap_info: change... |
105 |
swp_entry_t entry = swp_entry(si->type, offset); |
c9e444103 mm: reuse unused ... |
106 107 |
struct page *page; int ret = 0; |
f6ab1f7f6 mm, swap: use off... |
108 |
page = find_get_page(swap_address_space(entry), swp_offset(entry)); |
c9e444103 mm: reuse unused ... |
109 110 111 112 113 114 115 116 117 118 119 120 121 |
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); } |
09cbfeaf1 mm, fs: get rid o... |
122 |
put_page(page); |
c9e444103 mm: reuse unused ... |
123 124 |
return ret; } |
355cfa73d mm: modify swap_m... |
125 |
|
1da177e4c Linux-2.6.12-rc2 |
126 |
/* |
6a6ba8317 swapfile: swapon ... |
127 128 129 130 131 132 |
* 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... |
133 134 |
sector_t start_block; sector_t nr_blocks; |
6a6ba8317 swapfile: swapon ... |
135 |
int err = 0; |
9625a5f28 swap_info: includ... |
136 137 138 139 140 141 |
/* 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... |
142 |
nr_blocks, GFP_KERNEL, 0); |
9625a5f28 swap_info: includ... |
143 144 145 146 |
if (err) return err; cond_resched(); } |
6a6ba8317 swapfile: swapon ... |
147 |
|
9625a5f28 swap_info: includ... |
148 149 150 |
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 ... |
151 152 |
err = blkdev_issue_discard(si->bdev, start_block, |
dd3932edd block: remove BLK... |
153 |
nr_blocks, GFP_KERNEL, 0); |
6a6ba8317 swapfile: swapon ... |
154 155 156 157 158 159 160 |
if (err) break; cond_resched(); } return err; /* That will often be -EOPNOTSUPP */ } |
7992fde72 swapfile: swap al... |
161 162 163 164 165 166 167 168 169 170 171 |
/* * 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) { |
7992fde72 swapfile: swap al... |
172 173 174 175 |
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 ... |
176 |
sector_t nr_blocks = se->nr_pages - offset; |
7992fde72 swapfile: swap al... |
177 178 179 180 181 182 183 184 185 186 187 188 |
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... |
189 |
nr_blocks, GFP_NOIO, 0)) |
7992fde72 swapfile: swap al... |
190 191 |
break; } |
a8ae49917 mm/swapfile.c: us... |
192 |
se = list_next_entry(se, list); |
7992fde72 swapfile: swap al... |
193 194 |
} } |
38d8b4e6b mm, THP, swap: de... |
195 196 197 |
#ifdef CONFIG_THP_SWAP #define SWAPFILE_CLUSTER HPAGE_PMD_NR #else |
048c27fd7 [PATCH] swap: sca... |
198 |
#define SWAPFILE_CLUSTER 256 |
38d8b4e6b mm, THP, swap: de... |
199 |
#endif |
048c27fd7 [PATCH] swap: sca... |
200 |
#define LATENCY_LIMIT 256 |
2a8f94493 swap: change bloc... |
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 |
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; } |
e07098294 mm, THP, swap: su... |
258 259 260 261 262 263 264 265 266 |
static inline bool cluster_is_huge(struct swap_cluster_info *info) { return info->flags & CLUSTER_FLAG_HUGE; } static inline void cluster_clear_huge(struct swap_cluster_info *info) { info->flags &= ~CLUSTER_FLAG_HUGE; } |
235b62176 mm/swap: add clus... |
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 |
static inline struct swap_cluster_info *lock_cluster(struct swap_info_struct *si, unsigned long offset) { struct swap_cluster_info *ci; ci = si->cluster_info; if (ci) { ci += offset / SWAPFILE_CLUSTER; spin_lock(&ci->lock); } return ci; } static inline void unlock_cluster(struct swap_cluster_info *ci) { if (ci) spin_unlock(&ci->lock); } static inline struct swap_cluster_info *lock_cluster_or_swap_info( struct swap_info_struct *si, unsigned long offset) { struct swap_cluster_info *ci; ci = lock_cluster(si, offset); if (!ci) spin_lock(&si->lock); return ci; } static inline void unlock_cluster_or_swap_info(struct swap_info_struct *si, struct swap_cluster_info *ci) { if (ci) unlock_cluster(ci); else spin_unlock(&si->lock); } |
6b5349159 mm, swap: add swa... |
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 |
static inline bool cluster_list_empty(struct swap_cluster_list *list) { return cluster_is_null(&list->head); } static inline unsigned int cluster_list_first(struct swap_cluster_list *list) { return cluster_next(&list->head); } static void cluster_list_init(struct swap_cluster_list *list) { cluster_set_null(&list->head); cluster_set_null(&list->tail); } static void cluster_list_add_tail(struct swap_cluster_list *list, struct swap_cluster_info *ci, unsigned int idx) { if (cluster_list_empty(list)) { cluster_set_next_flag(&list->head, idx, 0); cluster_set_next_flag(&list->tail, idx, 0); } else { |
235b62176 mm/swap: add clus... |
331 |
struct swap_cluster_info *ci_tail; |
6b5349159 mm, swap: add swa... |
332 |
unsigned int tail = cluster_next(&list->tail); |
235b62176 mm/swap: add clus... |
333 334 335 336 337 338 339 |
/* * Nested cluster lock, but both cluster locks are * only acquired when we held swap_info_struct->lock */ ci_tail = ci + tail; spin_lock_nested(&ci_tail->lock, SINGLE_DEPTH_NESTING); cluster_set_next(ci_tail, idx); |
0ef017d11 mm, swap: improve... |
340 |
spin_unlock(&ci_tail->lock); |
6b5349159 mm, swap: add swa... |
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 |
cluster_set_next_flag(&list->tail, idx, 0); } } static unsigned int cluster_list_del_first(struct swap_cluster_list *list, struct swap_cluster_info *ci) { unsigned int idx; idx = cluster_next(&list->head); if (cluster_next(&list->tail) == idx) { cluster_set_null(&list->head); cluster_set_null(&list->tail); } else cluster_set_next_flag(&list->head, cluster_next(&ci[idx]), 0); return idx; } |
815c2c543 swap: make swap d... |
360 361 362 363 364 365 366 367 368 369 370 371 |
/* 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); |
6b5349159 mm, swap: add swa... |
372 |
cluster_list_add_tail(&si->discard_clusters, si->cluster_info, idx); |
815c2c543 swap: make swap d... |
373 374 375 |
schedule_work(&si->discard_work); } |
38d8b4e6b mm, THP, swap: de... |
376 377 378 379 380 381 382 |
static void __free_cluster(struct swap_info_struct *si, unsigned long idx) { struct swap_cluster_info *ci = si->cluster_info; cluster_set_flag(ci + idx, CLUSTER_FLAG_FREE); cluster_list_add_tail(&si->free_clusters, ci, idx); } |
815c2c543 swap: make swap d... |
383 384 385 386 387 388 |
/* * 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) { |
235b62176 mm/swap: add clus... |
389 |
struct swap_cluster_info *info, *ci; |
815c2c543 swap: make swap d... |
390 391 392 |
unsigned int idx; info = si->cluster_info; |
6b5349159 mm, swap: add swa... |
393 394 |
while (!cluster_list_empty(&si->discard_clusters)) { idx = cluster_list_del_first(&si->discard_clusters, info); |
815c2c543 swap: make swap d... |
395 396 397 398 399 400 |
spin_unlock(&si->lock); discard_swap_cluster(si, idx * SWAPFILE_CLUSTER, SWAPFILE_CLUSTER); spin_lock(&si->lock); |
235b62176 mm/swap: add clus... |
401 |
ci = lock_cluster(si, idx * SWAPFILE_CLUSTER); |
38d8b4e6b mm, THP, swap: de... |
402 |
__free_cluster(si, idx); |
815c2c543 swap: make swap d... |
403 404 |
memset(si->swap_map + idx * SWAPFILE_CLUSTER, 0, SWAPFILE_CLUSTER); |
235b62176 mm/swap: add clus... |
405 |
unlock_cluster(ci); |
815c2c543 swap: make swap d... |
406 407 408 409 410 411 412 413 414 415 416 417 418 |
} } 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); } |
38d8b4e6b mm, THP, swap: de... |
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 |
static void alloc_cluster(struct swap_info_struct *si, unsigned long idx) { struct swap_cluster_info *ci = si->cluster_info; VM_BUG_ON(cluster_list_first(&si->free_clusters) != idx); cluster_list_del_first(&si->free_clusters, ci); cluster_set_count_flag(ci + idx, 0, 0); } static void free_cluster(struct swap_info_struct *si, unsigned long idx) { struct swap_cluster_info *ci = si->cluster_info + idx; VM_BUG_ON(cluster_count(ci) != 0); /* * If the swap is discardable, prepare discard the cluster * instead of free it immediately. The cluster will be freed * after discard. */ if ((si->flags & (SWP_WRITEOK | SWP_PAGE_DISCARD)) == (SWP_WRITEOK | SWP_PAGE_DISCARD)) { swap_cluster_schedule_discard(si, idx); return; } __free_cluster(si, idx); } |
2a8f94493 swap: change bloc... |
446 447 448 449 450 451 452 453 454 455 456 |
/* * 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; |
38d8b4e6b mm, THP, swap: de... |
457 458 |
if (cluster_is_free(&cluster_info[idx])) alloc_cluster(p, idx); |
2a8f94493 swap: change bloc... |
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 |
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); |
38d8b4e6b mm, THP, swap: de... |
481 482 |
if (cluster_count(&cluster_info[idx]) == 0) free_cluster(p, idx); |
2a8f94493 swap: change bloc... |
483 484 485 486 487 488 |
} /* * 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... |
489 490 |
static bool scan_swap_map_ssd_cluster_conflict(struct swap_info_struct *si, |
2a8f94493 swap: change bloc... |
491 492 |
unsigned long offset) { |
ebc2a1a69 swap: make cluste... |
493 494 |
struct percpu_cluster *percpu_cluster; bool conflict; |
2a8f94493 swap: change bloc... |
495 |
offset /= SWAPFILE_CLUSTER; |
6b5349159 mm, swap: add swa... |
496 497 |
conflict = !cluster_list_empty(&si->free_clusters) && offset != cluster_list_first(&si->free_clusters) && |
2a8f94493 swap: change bloc... |
498 |
cluster_is_free(&si->cluster_info[offset]); |
ebc2a1a69 swap: make cluste... |
499 500 501 502 503 504 505 506 507 508 509 510 511 |
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. */ |
36005bae2 mm/swap: allocate... |
512 |
static bool scan_swap_map_try_ssd_cluster(struct swap_info_struct *si, |
ebc2a1a69 swap: make cluste... |
513 514 515 |
unsigned long *offset, unsigned long *scan_base) { struct percpu_cluster *cluster; |
235b62176 mm/swap: add clus... |
516 |
struct swap_cluster_info *ci; |
ebc2a1a69 swap: make cluste... |
517 |
bool found_free; |
235b62176 mm/swap: add clus... |
518 |
unsigned long tmp, max; |
ebc2a1a69 swap: make cluste... |
519 520 521 522 |
new_cluster: cluster = this_cpu_ptr(si->percpu_cluster); if (cluster_is_null(&cluster->index)) { |
6b5349159 mm, swap: add swa... |
523 524 |
if (!cluster_list_empty(&si->free_clusters)) { cluster->index = si->free_clusters.head; |
ebc2a1a69 swap: make cluste... |
525 526 |
cluster->next = cluster_next(&cluster->index) * SWAPFILE_CLUSTER; |
6b5349159 mm, swap: add swa... |
527 |
} else if (!cluster_list_empty(&si->discard_clusters)) { |
ebc2a1a69 swap: make cluste... |
528 529 530 531 532 533 534 535 |
/* * 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 |
36005bae2 mm/swap: allocate... |
536 |
return false; |
ebc2a1a69 swap: make cluste... |
537 538 539 540 541 542 543 544 545 |
} 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; |
235b62176 mm/swap: add clus... |
546 547 548 549 550 551 552 553 |
max = min_t(unsigned long, si->max, (cluster_next(&cluster->index) + 1) * SWAPFILE_CLUSTER); if (tmp >= max) { cluster_set_null(&cluster->index); goto new_cluster; } ci = lock_cluster(si, tmp); while (tmp < max) { |
ebc2a1a69 swap: make cluste... |
554 555 556 557 558 559 |
if (!si->swap_map[tmp]) { found_free = true; break; } tmp++; } |
235b62176 mm/swap: add clus... |
560 |
unlock_cluster(ci); |
ebc2a1a69 swap: make cluste... |
561 562 563 564 565 566 567 |
if (!found_free) { cluster_set_null(&cluster->index); goto new_cluster; } cluster->next = tmp + 1; *offset = tmp; *scan_base = tmp; |
36005bae2 mm/swap: allocate... |
568 |
return found_free; |
2a8f94493 swap: change bloc... |
569 |
} |
a2468cc9b swap: choose swap... |
570 571 572 573 574 575 576 577 578 579 580 581 582 583 |
static void __del_from_avail_list(struct swap_info_struct *p) { int nid; for_each_node(nid) plist_del(&p->avail_lists[nid], &swap_avail_heads[nid]); } static void del_from_avail_list(struct swap_info_struct *p) { spin_lock(&swap_avail_lock); __del_from_avail_list(p); spin_unlock(&swap_avail_lock); } |
38d8b4e6b mm, THP, swap: de... |
584 585 586 587 588 589 590 591 592 593 594 595 596 |
static void swap_range_alloc(struct swap_info_struct *si, unsigned long offset, unsigned int nr_entries) { unsigned int end = offset + nr_entries - 1; if (offset == si->lowest_bit) si->lowest_bit += nr_entries; if (end == si->highest_bit) si->highest_bit -= nr_entries; si->inuse_pages += nr_entries; if (si->inuse_pages == si->pages) { si->lowest_bit = si->max; si->highest_bit = 0; |
a2468cc9b swap: choose swap... |
597 |
del_from_avail_list(si); |
38d8b4e6b mm, THP, swap: de... |
598 599 |
} } |
a2468cc9b swap: choose swap... |
600 601 602 603 604 605 606 607 608 609 610 |
static void add_to_avail_list(struct swap_info_struct *p) { int nid; spin_lock(&swap_avail_lock); for_each_node(nid) { WARN_ON(!plist_node_empty(&p->avail_lists[nid])); plist_add(&p->avail_lists[nid], &swap_avail_heads[nid]); } spin_unlock(&swap_avail_lock); } |
38d8b4e6b mm, THP, swap: de... |
611 612 613 614 615 616 617 618 619 620 621 622 |
static void swap_range_free(struct swap_info_struct *si, unsigned long offset, unsigned int nr_entries) { unsigned long end = offset + nr_entries - 1; void (*swap_slot_free_notify)(struct block_device *, unsigned long); if (offset < si->lowest_bit) si->lowest_bit = offset; if (end > si->highest_bit) { bool was_full = !si->highest_bit; si->highest_bit = end; |
a2468cc9b swap: choose swap... |
623 624 |
if (was_full && (si->flags & SWP_WRITEOK)) add_to_avail_list(si); |
38d8b4e6b mm, THP, swap: de... |
625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 |
} atomic_long_add(nr_entries, &nr_swap_pages); si->inuse_pages -= nr_entries; if (si->flags & SWP_BLKDEV) swap_slot_free_notify = si->bdev->bd_disk->fops->swap_slot_free_notify; else swap_slot_free_notify = NULL; while (offset <= end) { frontswap_invalidate_page(si->type, offset); if (swap_slot_free_notify) swap_slot_free_notify(si->bdev, offset); offset++; } } |
36005bae2 mm/swap: allocate... |
640 641 642 |
static int scan_swap_map_slots(struct swap_info_struct *si, unsigned char usage, int nr, swp_entry_t slots[]) |
1da177e4c Linux-2.6.12-rc2 |
643 |
{ |
235b62176 mm/swap: add clus... |
644 |
struct swap_cluster_info *ci; |
ebebbbe90 swapfile: rearran... |
645 |
unsigned long offset; |
c60aa176c swapfile: swap al... |
646 |
unsigned long scan_base; |
7992fde72 swapfile: swap al... |
647 |
unsigned long last_in_cluster = 0; |
048c27fd7 [PATCH] swap: sca... |
648 |
int latency_ration = LATENCY_LIMIT; |
36005bae2 mm/swap: allocate... |
649 650 651 652 |
int n_ret = 0; if (nr > SWAP_BATCH) nr = SWAP_BATCH; |
7dfad4183 [PATCH] swap: sca... |
653 |
|
886bb7e9c swapfile: remove ... |
654 |
/* |
7dfad4183 [PATCH] swap: sca... |
655 656 657 658 659 660 661 |
* 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... |
662 |
* And we let swap pages go all over an SSD partition. Hugh |
7dfad4183 [PATCH] swap: sca... |
663 |
*/ |
52b7efdbe [PATCH] swap: sca... |
664 |
si->flags += SWP_SCANNING; |
c60aa176c swapfile: swap al... |
665 |
scan_base = offset = si->cluster_next; |
ebebbbe90 swapfile: rearran... |
666 |
|
ebc2a1a69 swap: make cluste... |
667 668 |
/* SSD algorithm */ if (si->cluster_info) { |
36005bae2 mm/swap: allocate... |
669 670 671 672 |
if (scan_swap_map_try_ssd_cluster(si, &offset, &scan_base)) goto checks; else goto scan; |
ebc2a1a69 swap: make cluste... |
673 |
} |
ebebbbe90 swapfile: rearran... |
674 675 676 677 678 |
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... |
679 |
|
ec8acf20a swap: add per-par... |
680 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
681 |
|
c60aa176c swapfile: swap al... |
682 683 684 |
/* * 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... |
685 686 |
* 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... |
687 |
*/ |
50088c440 mm/swapfile.c: de... |
688 |
scan_base = offset = si->lowest_bit; |
7dfad4183 [PATCH] swap: sca... |
689 690 691 692 |
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 |
693 |
if (si->swap_map[offset]) |
7dfad4183 [PATCH] swap: sca... |
694 695 |
last_in_cluster = offset + SWAPFILE_CLUSTER; else if (offset == last_in_cluster) { |
ec8acf20a swap: add per-par... |
696 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
697 698 699 |
offset -= SWAPFILE_CLUSTER - 1; si->cluster_next = offset; si->cluster_nr = SWAPFILE_CLUSTER - 1; |
c60aa176c swapfile: swap al... |
700 701 702 703 704 705 706 707 708 |
goto checks; } if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } } offset = scan_base; |
ec8acf20a swap: add per-par... |
709 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
710 |
si->cluster_nr = SWAPFILE_CLUSTER - 1; |
1da177e4c Linux-2.6.12-rc2 |
711 |
} |
7dfad4183 [PATCH] swap: sca... |
712 |
|
ebebbbe90 swapfile: rearran... |
713 |
checks: |
ebc2a1a69 swap: make cluste... |
714 |
if (si->cluster_info) { |
36005bae2 mm/swap: allocate... |
715 716 717 718 719 720 721 722 |
while (scan_swap_map_ssd_cluster_conflict(si, offset)) { /* take a break if we already got some slots */ if (n_ret) goto done; if (!scan_swap_map_try_ssd_cluster(si, &offset, &scan_base)) goto scan; } |
ebc2a1a69 swap: make cluste... |
723 |
} |
ebebbbe90 swapfile: rearran... |
724 |
if (!(si->flags & SWP_WRITEOK)) |
52b7efdbe [PATCH] swap: sca... |
725 |
goto no_page; |
7dfad4183 [PATCH] swap: sca... |
726 727 |
if (!si->highest_bit) goto no_page; |
ebebbbe90 swapfile: rearran... |
728 |
if (offset > si->highest_bit) |
c60aa176c swapfile: swap al... |
729 |
scan_base = offset = si->lowest_bit; |
c9e444103 mm: reuse unused ... |
730 |
|
235b62176 mm/swap: add clus... |
731 |
ci = lock_cluster(si, offset); |
b73d7fcec swap: prevent reu... |
732 733 |
/* 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 ... |
734 |
int swap_was_freed; |
235b62176 mm/swap: add clus... |
735 |
unlock_cluster(ci); |
ec8acf20a swap: add per-par... |
736 |
spin_unlock(&si->lock); |
c9e444103 mm: reuse unused ... |
737 |
swap_was_freed = __try_to_reclaim_swap(si, offset); |
ec8acf20a swap: add per-par... |
738 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
739 740 741 742 743 |
/* entry was freed successfully, try to use this again */ if (swap_was_freed) goto checks; goto scan; /* check next one */ } |
235b62176 mm/swap: add clus... |
744 745 |
if (si->swap_map[offset]) { unlock_cluster(ci); |
36005bae2 mm/swap: allocate... |
746 747 748 749 |
if (!n_ret) goto scan; else goto done; |
235b62176 mm/swap: add clus... |
750 |
} |
2872bb2d0 mm, swap: avoid l... |
751 752 753 |
si->swap_map[offset] = usage; inc_cluster_info_page(si, si->cluster_info, offset); unlock_cluster(ci); |
ebebbbe90 swapfile: rearran... |
754 |
|
38d8b4e6b mm, THP, swap: de... |
755 |
swap_range_alloc(si, offset, 1); |
ebebbbe90 swapfile: rearran... |
756 |
si->cluster_next = offset + 1; |
36005bae2 mm/swap: allocate... |
757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 |
slots[n_ret++] = swp_entry(si->type, offset); /* got enough slots or reach max slots? */ if ((n_ret == nr) || (offset >= si->highest_bit)) goto done; /* search for next available slot */ /* time to take a break? */ if (unlikely(--latency_ration < 0)) { if (n_ret) goto done; spin_unlock(&si->lock); cond_resched(); spin_lock(&si->lock); latency_ration = LATENCY_LIMIT; } /* try to get more slots in cluster */ if (si->cluster_info) { if (scan_swap_map_try_ssd_cluster(si, &offset, &scan_base)) goto checks; else goto done; } /* non-ssd case */ ++offset; /* non-ssd case, still more slots in cluster? */ if (si->cluster_nr && !si->swap_map[offset]) { --si->cluster_nr; goto checks; } |
7992fde72 swapfile: swap al... |
790 |
|
36005bae2 mm/swap: allocate... |
791 792 793 |
done: si->flags -= SWP_SCANNING; return n_ret; |
7dfad4183 [PATCH] swap: sca... |
794 |
|
ebebbbe90 swapfile: rearran... |
795 |
scan: |
ec8acf20a swap: add per-par... |
796 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
797 |
while (++offset <= si->highest_bit) { |
52b7efdbe [PATCH] swap: sca... |
798 |
if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
799 |
spin_lock(&si->lock); |
52b7efdbe [PATCH] swap: sca... |
800 801 |
goto checks; } |
c9e444103 mm: reuse unused ... |
802 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
803 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
804 805 |
goto checks; } |
048c27fd7 [PATCH] swap: sca... |
806 807 808 809 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
7dfad4183 [PATCH] swap: sca... |
810 |
} |
c60aa176c swapfile: swap al... |
811 |
offset = si->lowest_bit; |
a5998061d mm/swapfile.c: do... |
812 |
while (offset < scan_base) { |
c60aa176c swapfile: swap al... |
813 |
if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
814 |
spin_lock(&si->lock); |
c60aa176c swapfile: swap al... |
815 816 |
goto checks; } |
c9e444103 mm: reuse unused ... |
817 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
818 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
819 820 |
goto checks; } |
c60aa176c swapfile: swap al... |
821 822 823 824 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
a5998061d mm/swapfile.c: do... |
825 |
offset++; |
c60aa176c swapfile: swap al... |
826 |
} |
ec8acf20a swap: add per-par... |
827 |
spin_lock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
828 829 |
no_page: |
52b7efdbe [PATCH] swap: sca... |
830 |
si->flags -= SWP_SCANNING; |
36005bae2 mm/swap: allocate... |
831 |
return n_ret; |
1da177e4c Linux-2.6.12-rc2 |
832 |
} |
38d8b4e6b mm, THP, swap: de... |
833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 |
#ifdef CONFIG_THP_SWAP static int swap_alloc_cluster(struct swap_info_struct *si, swp_entry_t *slot) { unsigned long idx; struct swap_cluster_info *ci; unsigned long offset, i; unsigned char *map; if (cluster_list_empty(&si->free_clusters)) return 0; idx = cluster_list_first(&si->free_clusters); offset = idx * SWAPFILE_CLUSTER; ci = lock_cluster(si, offset); alloc_cluster(si, idx); |
e07098294 mm, THP, swap: su... |
848 |
cluster_set_count_flag(ci, SWAPFILE_CLUSTER, CLUSTER_FLAG_HUGE); |
38d8b4e6b mm, THP, swap: de... |
849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 |
map = si->swap_map + offset; for (i = 0; i < SWAPFILE_CLUSTER; i++) map[i] = SWAP_HAS_CACHE; unlock_cluster(ci); swap_range_alloc(si, offset, SWAPFILE_CLUSTER); *slot = swp_entry(si->type, offset); return 1; } static void swap_free_cluster(struct swap_info_struct *si, unsigned long idx) { unsigned long offset = idx * SWAPFILE_CLUSTER; struct swap_cluster_info *ci; ci = lock_cluster(si, offset); cluster_set_count_flag(ci, 0, 0); free_cluster(si, idx); unlock_cluster(ci); swap_range_free(si, offset, SWAPFILE_CLUSTER); } #else static int swap_alloc_cluster(struct swap_info_struct *si, swp_entry_t *slot) { VM_WARN_ON_ONCE(1); return 0; } #endif /* CONFIG_THP_SWAP */ |
36005bae2 mm/swap: allocate... |
878 879 880 881 882 883 884 885 886 887 888 889 890 891 |
static unsigned long scan_swap_map(struct swap_info_struct *si, unsigned char usage) { swp_entry_t entry; int n_ret; n_ret = scan_swap_map_slots(si, usage, 1, &entry); if (n_ret) return swp_offset(entry); else return 0; } |
38d8b4e6b mm, THP, swap: de... |
892 |
int get_swap_pages(int n_goal, bool cluster, swp_entry_t swp_entries[]) |
1da177e4c Linux-2.6.12-rc2 |
893 |
{ |
38d8b4e6b mm, THP, swap: de... |
894 |
unsigned long nr_pages = cluster ? SWAPFILE_CLUSTER : 1; |
adfab836f swap: change swap... |
895 |
struct swap_info_struct *si, *next; |
36005bae2 mm/swap: allocate... |
896 897 |
long avail_pgs; int n_ret = 0; |
a2468cc9b swap: choose swap... |
898 |
int node; |
1da177e4c Linux-2.6.12-rc2 |
899 |
|
38d8b4e6b mm, THP, swap: de... |
900 901 902 903 |
/* Only single cluster request supported */ WARN_ON_ONCE(n_goal > 1 && cluster); avail_pgs = atomic_long_read(&nr_swap_pages) / nr_pages; |
36005bae2 mm/swap: allocate... |
904 |
if (avail_pgs <= 0) |
fb4f88dca [PATCH] swap: get... |
905 |
goto noswap; |
36005bae2 mm/swap: allocate... |
906 907 908 909 910 911 |
if (n_goal > SWAP_BATCH) n_goal = SWAP_BATCH; if (n_goal > avail_pgs) n_goal = avail_pgs; |
38d8b4e6b mm, THP, swap: de... |
912 |
atomic_long_sub(n_goal * nr_pages, &nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
913 |
|
18ab4d4ce swap: change swap... |
914 915 916 |
spin_lock(&swap_avail_lock); start_over: |
a2468cc9b swap: choose swap... |
917 918 |
node = numa_node_id(); plist_for_each_entry_safe(si, next, &swap_avail_heads[node], avail_lists[node]) { |
18ab4d4ce swap: change swap... |
919 |
/* requeue si to after same-priority siblings */ |
a2468cc9b swap: choose swap... |
920 |
plist_requeue(&si->avail_lists[node], &swap_avail_heads[node]); |
18ab4d4ce swap: change swap... |
921 |
spin_unlock(&swap_avail_lock); |
ec8acf20a swap: add per-par... |
922 |
spin_lock(&si->lock); |
adfab836f swap: change swap... |
923 |
if (!si->highest_bit || !(si->flags & SWP_WRITEOK)) { |
18ab4d4ce swap: change swap... |
924 |
spin_lock(&swap_avail_lock); |
a2468cc9b swap: choose swap... |
925 |
if (plist_node_empty(&si->avail_lists[node])) { |
18ab4d4ce swap: change swap... |
926 927 928 929 930 931 932 933 934 935 936 |
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); |
a2468cc9b swap: choose swap... |
937 |
__del_from_avail_list(si); |
ec8acf20a swap: add per-par... |
938 |
spin_unlock(&si->lock); |
18ab4d4ce swap: change swap... |
939 |
goto nextsi; |
ec8acf20a swap: add per-par... |
940 |
} |
f0eea189e mm, THP, swap: do... |
941 942 943 944 |
if (cluster) { if (!(si->flags & SWP_FILE)) n_ret = swap_alloc_cluster(si, swp_entries); } else |
38d8b4e6b mm, THP, swap: de... |
945 946 |
n_ret = scan_swap_map_slots(si, SWAP_HAS_CACHE, n_goal, swp_entries); |
ec8acf20a swap: add per-par... |
947 |
spin_unlock(&si->lock); |
38d8b4e6b mm, THP, swap: de... |
948 |
if (n_ret || cluster) |
36005bae2 mm/swap: allocate... |
949 |
goto check_out; |
18ab4d4ce swap: change swap... |
950 951 |
pr_debug("scan_swap_map of si %d failed to find offset ", |
36005bae2 mm/swap: allocate... |
952 |
si->type); |
18ab4d4ce swap: change swap... |
953 954 |
spin_lock(&swap_avail_lock); nextsi: |
adfab836f swap: change swap... |
955 956 957 958 |
/* * 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... |
959 960 961 962 |
* 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 |
36005bae2 mm/swap: allocate... |
963 964 |
* swap_avail_head list then try it, otherwise start over * if we have not gotten any slots. |
adfab836f swap: change swap... |
965 |
*/ |
a2468cc9b swap: choose swap... |
966 |
if (plist_node_empty(&next->avail_lists[node])) |
18ab4d4ce swap: change swap... |
967 |
goto start_over; |
1da177e4c Linux-2.6.12-rc2 |
968 |
} |
fb4f88dca [PATCH] swap: get... |
969 |
|
18ab4d4ce swap: change swap... |
970 |
spin_unlock(&swap_avail_lock); |
36005bae2 mm/swap: allocate... |
971 972 |
check_out: if (n_ret < n_goal) |
38d8b4e6b mm, THP, swap: de... |
973 974 |
atomic_long_add((long)(n_goal - n_ret) * nr_pages, &nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
975 |
noswap: |
36005bae2 mm/swap: allocate... |
976 977 |
return n_ret; } |
2de1a7e40 mm/swapfile.c: fi... |
978 |
/* The only caller of this function is now suspend routine */ |
910321ea8 swap: revert spec... |
979 980 981 982 |
swp_entry_t get_swap_page_of_type(int type) { struct swap_info_struct *si; pgoff_t offset; |
910321ea8 swap: revert spec... |
983 |
si = swap_info[type]; |
ec8acf20a swap: add per-par... |
984 |
spin_lock(&si->lock); |
910321ea8 swap: revert spec... |
985 |
if (si && (si->flags & SWP_WRITEOK)) { |
ec8acf20a swap: add per-par... |
986 |
atomic_long_dec(&nr_swap_pages); |
910321ea8 swap: revert spec... |
987 988 989 |
/* This is called for allocating swap entry, not cache */ offset = scan_swap_map(si, 1); if (offset) { |
ec8acf20a swap: add per-par... |
990 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
991 992 |
return swp_entry(type, offset); } |
ec8acf20a swap: add per-par... |
993 |
atomic_long_inc(&nr_swap_pages); |
910321ea8 swap: revert spec... |
994 |
} |
ec8acf20a swap: add per-par... |
995 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
996 997 |
return (swp_entry_t) {0}; } |
e8c26ab60 mm/swap: skip rea... |
998 |
static struct swap_info_struct *__swap_info_get(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
999 |
{ |
73c34b6ac swap_info: miscel... |
1000 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
1001 1002 1003 1004 1005 1006 1007 |
unsigned long offset, type; if (!entry.val) goto out; type = swp_type(entry); if (type >= nr_swapfiles) goto bad_nofile; |
efa90a981 swap_info: change... |
1008 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
1009 1010 1011 1012 1013 |
if (!(p->flags & SWP_USED)) goto bad_device; offset = swp_offset(entry); if (offset >= p->max) goto bad_offset; |
1da177e4c Linux-2.6.12-rc2 |
1014 |
return p; |
1da177e4c Linux-2.6.12-rc2 |
1015 |
bad_offset: |
6a991fc72 mm/swap: fix kern... |
1016 1017 |
pr_err("swap_info_get: %s%08lx ", Bad_offset, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
1018 1019 |
goto out; bad_device: |
6a991fc72 mm/swap: fix kern... |
1020 1021 |
pr_err("swap_info_get: %s%08lx ", Unused_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
1022 1023 |
goto out; bad_nofile: |
6a991fc72 mm/swap: fix kern... |
1024 1025 |
pr_err("swap_info_get: %s%08lx ", Bad_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
1026 1027 |
out: return NULL; |
886bb7e9c swapfile: remove ... |
1028 |
} |
1da177e4c Linux-2.6.12-rc2 |
1029 |
|
e8c26ab60 mm/swap: skip rea... |
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 |
static struct swap_info_struct *_swap_info_get(swp_entry_t entry) { struct swap_info_struct *p; p = __swap_info_get(entry); if (!p) goto out; if (!p->swap_map[swp_offset(entry)]) goto bad_free; return p; bad_free: pr_err("swap_info_get: %s%08lx ", Unused_offset, entry.val); goto out; out: return NULL; } |
235b62176 mm/swap: add clus... |
1048 1049 1050 1051 1052 1053 1054 1055 1056 |
static struct swap_info_struct *swap_info_get(swp_entry_t entry) { struct swap_info_struct *p; p = _swap_info_get(entry); if (p) spin_lock(&p->lock); return p; } |
7c00bafee mm/swap: free swa... |
1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 |
static struct swap_info_struct *swap_info_get_cont(swp_entry_t entry, struct swap_info_struct *q) { struct swap_info_struct *p; p = _swap_info_get(entry); if (p != q) { if (q != NULL) spin_unlock(&q->lock); if (p != NULL) spin_lock(&p->lock); } return p; } static unsigned char __swap_entry_free(struct swap_info_struct *p, swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
1075 |
{ |
235b62176 mm/swap: add clus... |
1076 |
struct swap_cluster_info *ci; |
253d553ba swap_info: SWAP_H... |
1077 |
unsigned long offset = swp_offset(entry); |
8d69aaee8 swap_info: swap_m... |
1078 1079 |
unsigned char count; unsigned char has_cache; |
235b62176 mm/swap: add clus... |
1080 |
|
7c00bafee mm/swap: free swa... |
1081 |
ci = lock_cluster_or_swap_info(p, offset); |
355cfa73d mm: modify swap_m... |
1082 |
|
253d553ba swap_info: SWAP_H... |
1083 |
count = p->swap_map[offset]; |
235b62176 mm/swap: add clus... |
1084 |
|
253d553ba swap_info: SWAP_H... |
1085 1086 |
has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; |
355cfa73d mm: modify swap_m... |
1087 |
|
253d553ba swap_info: SWAP_H... |
1088 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
1089 |
VM_BUG_ON(!has_cache); |
253d553ba swap_info: SWAP_H... |
1090 |
has_cache = 0; |
aaa468653 swap_info: note S... |
1091 1092 1093 1094 1095 1096 |
} 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... |
1097 1098 1099 1100 1101 1102 1103 1104 1105 |
} 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... |
1106 |
|
253d553ba swap_info: SWAP_H... |
1107 |
usage = count | has_cache; |
7c00bafee mm/swap: free swa... |
1108 1109 1110 1111 1112 1113 |
p->swap_map[offset] = usage ? : SWAP_HAS_CACHE; unlock_cluster_or_swap_info(p, ci); return usage; } |
355cfa73d mm: modify swap_m... |
1114 |
|
7c00bafee mm/swap: free swa... |
1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 |
static void swap_entry_free(struct swap_info_struct *p, swp_entry_t entry) { struct swap_cluster_info *ci; unsigned long offset = swp_offset(entry); unsigned char count; ci = lock_cluster(p, offset); count = p->swap_map[offset]; VM_BUG_ON(count != SWAP_HAS_CACHE); p->swap_map[offset] = 0; dec_cluster_info_page(p, p->cluster_info, offset); |
235b62176 mm/swap: add clus... |
1126 |
unlock_cluster(ci); |
38d8b4e6b mm, THP, swap: de... |
1127 1128 |
mem_cgroup_uncharge_swap(entry, 1); swap_range_free(p, offset, 1); |
1da177e4c Linux-2.6.12-rc2 |
1129 1130 1131 |
} /* |
2de1a7e40 mm/swapfile.c: fi... |
1132 |
* Caller has made sure that the swap device corresponding to entry |
1da177e4c Linux-2.6.12-rc2 |
1133 1134 1135 1136 |
* is still around or has not been recycled. */ void swap_free(swp_entry_t entry) { |
73c34b6ac swap_info: miscel... |
1137 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
1138 |
|
235b62176 mm/swap: add clus... |
1139 |
p = _swap_info_get(entry); |
7c00bafee mm/swap: free swa... |
1140 1141 |
if (p) { if (!__swap_entry_free(p, entry, 1)) |
67afa38e0 mm/swap: add cach... |
1142 |
free_swap_slot(entry); |
7c00bafee mm/swap: free swa... |
1143 |
} |
1da177e4c Linux-2.6.12-rc2 |
1144 1145 1146 |
} /* |
cb4b86ba4 mm: add swap cach... |
1147 1148 |
* Called after dropping swapcache to decrease refcnt to swap entries. */ |
75f6d6d29 mm, THP, swap: un... |
1149 |
static void swapcache_free(swp_entry_t entry) |
cb4b86ba4 mm: add swap cach... |
1150 |
{ |
355cfa73d mm: modify swap_m... |
1151 |
struct swap_info_struct *p; |
235b62176 mm/swap: add clus... |
1152 |
p = _swap_info_get(entry); |
7c00bafee mm/swap: free swa... |
1153 1154 |
if (p) { if (!__swap_entry_free(p, entry, SWAP_HAS_CACHE)) |
67afa38e0 mm/swap: add cach... |
1155 |
free_swap_slot(entry); |
7c00bafee mm/swap: free swa... |
1156 1157 |
} } |
38d8b4e6b mm, THP, swap: de... |
1158 |
#ifdef CONFIG_THP_SWAP |
75f6d6d29 mm, THP, swap: un... |
1159 |
static void swapcache_free_cluster(swp_entry_t entry) |
38d8b4e6b mm, THP, swap: de... |
1160 1161 1162 1163 1164 1165 |
{ unsigned long offset = swp_offset(entry); unsigned long idx = offset / SWAPFILE_CLUSTER; struct swap_cluster_info *ci; struct swap_info_struct *si; unsigned char *map; |
a3aea839e mm, THP, swap: su... |
1166 1167 |
unsigned int i, free_entries = 0; unsigned char val; |
38d8b4e6b mm, THP, swap: de... |
1168 |
|
a3aea839e mm, THP, swap: su... |
1169 |
si = _swap_info_get(entry); |
38d8b4e6b mm, THP, swap: de... |
1170 1171 1172 1173 |
if (!si) return; ci = lock_cluster(si, offset); |
e07098294 mm, THP, swap: su... |
1174 |
VM_BUG_ON(!cluster_is_huge(ci)); |
38d8b4e6b mm, THP, swap: de... |
1175 1176 |
map = si->swap_map + offset; for (i = 0; i < SWAPFILE_CLUSTER; i++) { |
a3aea839e mm, THP, swap: su... |
1177 1178 1179 1180 1181 1182 1183 1184 |
val = map[i]; VM_BUG_ON(!(val & SWAP_HAS_CACHE)); if (val == SWAP_HAS_CACHE) free_entries++; } if (!free_entries) { for (i = 0; i < SWAPFILE_CLUSTER; i++) map[i] &= ~SWAP_HAS_CACHE; |
38d8b4e6b mm, THP, swap: de... |
1185 |
} |
e07098294 mm, THP, swap: su... |
1186 |
cluster_clear_huge(ci); |
38d8b4e6b mm, THP, swap: de... |
1187 |
unlock_cluster(ci); |
a3aea839e mm, THP, swap: su... |
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 |
if (free_entries == SWAPFILE_CLUSTER) { spin_lock(&si->lock); ci = lock_cluster(si, offset); memset(map, 0, SWAPFILE_CLUSTER); unlock_cluster(ci); mem_cgroup_uncharge_swap(entry, SWAPFILE_CLUSTER); swap_free_cluster(si, idx); spin_unlock(&si->lock); } else if (free_entries) { for (i = 0; i < SWAPFILE_CLUSTER; i++, entry.val++) { if (!__swap_entry_free(si, entry, SWAP_HAS_CACHE)) free_swap_slot(entry); } } |
38d8b4e6b mm, THP, swap: de... |
1202 |
} |
59807685a mm, THP, swap: su... |
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 |
int split_swap_cluster(swp_entry_t entry) { struct swap_info_struct *si; struct swap_cluster_info *ci; unsigned long offset = swp_offset(entry); si = _swap_info_get(entry); if (!si) return -EBUSY; ci = lock_cluster(si, offset); cluster_clear_huge(ci); unlock_cluster(ci); return 0; } |
75f6d6d29 mm, THP, swap: un... |
1218 1219 1220 1221 |
#else static inline void swapcache_free_cluster(swp_entry_t entry) { } |
38d8b4e6b mm, THP, swap: de... |
1222 |
#endif /* CONFIG_THP_SWAP */ |
75f6d6d29 mm, THP, swap: un... |
1223 1224 1225 1226 1227 1228 1229 |
void put_swap_page(struct page *page, swp_entry_t entry) { if (!PageTransHuge(page)) swapcache_free(entry); else swapcache_free_cluster(entry); } |
155b5f88e mm/swapfile.c: so... |
1230 1231 1232 1233 1234 1235 |
static int swp_entry_cmp(const void *ent1, const void *ent2) { const swp_entry_t *e1 = ent1, *e2 = ent2; return (int)swp_type(*e1) - (int)swp_type(*e2); } |
7c00bafee mm/swap: free swa... |
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 |
void swapcache_free_entries(swp_entry_t *entries, int n) { struct swap_info_struct *p, *prev; int i; if (n <= 0) return; prev = NULL; p = NULL; |
155b5f88e mm/swapfile.c: so... |
1246 1247 1248 1249 1250 1251 1252 1253 |
/* * Sort swap entries by swap device, so each lock is only taken once. * nr_swapfiles isn't absolutely correct, but the overhead of sort() is * so low that it isn't necessary to optimize further. */ if (nr_swapfiles > 1) sort(entries, n, sizeof(entries[0]), swp_entry_cmp, NULL); |
7c00bafee mm/swap: free swa... |
1254 1255 1256 1257 |
for (i = 0; i < n; ++i) { p = swap_info_get_cont(entries[i], prev); if (p) swap_entry_free(p, entries[i]); |
7c00bafee mm/swap: free swa... |
1258 1259 |
prev = p; } |
235b62176 mm/swap: add clus... |
1260 |
if (p) |
7c00bafee mm/swap: free swa... |
1261 |
spin_unlock(&p->lock); |
cb4b86ba4 mm: add swap cach... |
1262 1263 1264 |
} /* |
c475a8ab6 [PATCH] can_share... |
1265 |
* How many references to page are currently swapped out? |
570a335b8 swap_info: swap c... |
1266 1267 |
* 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 |
1268 |
*/ |
bde05d1cc shmem: replace pa... |
1269 |
int page_swapcount(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
1270 |
{ |
c475a8ab6 [PATCH] can_share... |
1271 1272 |
int count = 0; struct swap_info_struct *p; |
235b62176 mm/swap: add clus... |
1273 |
struct swap_cluster_info *ci; |
1da177e4c Linux-2.6.12-rc2 |
1274 |
swp_entry_t entry; |
235b62176 mm/swap: add clus... |
1275 |
unsigned long offset; |
1da177e4c Linux-2.6.12-rc2 |
1276 |
|
4c21e2f24 [PATCH] mm: split... |
1277 |
entry.val = page_private(page); |
235b62176 mm/swap: add clus... |
1278 |
p = _swap_info_get(entry); |
1da177e4c Linux-2.6.12-rc2 |
1279 |
if (p) { |
235b62176 mm/swap: add clus... |
1280 1281 1282 1283 |
offset = swp_offset(entry); ci = lock_cluster_or_swap_info(p, offset); count = swap_count(p->swap_map[offset]); unlock_cluster_or_swap_info(p, ci); |
1da177e4c Linux-2.6.12-rc2 |
1284 |
} |
c475a8ab6 [PATCH] can_share... |
1285 |
return count; |
1da177e4c Linux-2.6.12-rc2 |
1286 |
} |
322b8afe4 mm, swap: Fix a r... |
1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 |
static int swap_swapcount(struct swap_info_struct *si, swp_entry_t entry) { int count = 0; pgoff_t offset = swp_offset(entry); struct swap_cluster_info *ci; ci = lock_cluster_or_swap_info(si, offset); count = swap_count(si->swap_map[offset]); unlock_cluster_or_swap_info(si, ci); return count; } |
1da177e4c Linux-2.6.12-rc2 |
1298 |
/* |
8334b9622 mm: /proc/pid/sma... |
1299 |
* How many references to @entry are currently swapped out? |
e8c26ab60 mm/swap: skip rea... |
1300 1301 1302 1303 1304 1305 |
* This does not give an exact answer when swap count is continued, * but does include the high COUNT_CONTINUED flag to allow for that. */ int __swp_swapcount(swp_entry_t entry) { int count = 0; |
e8c26ab60 mm/swap: skip rea... |
1306 |
struct swap_info_struct *si; |
e8c26ab60 mm/swap: skip rea... |
1307 1308 |
si = __swap_info_get(entry); |
322b8afe4 mm, swap: Fix a r... |
1309 1310 |
if (si) count = swap_swapcount(si, entry); |
e8c26ab60 mm/swap: skip rea... |
1311 1312 1313 1314 1315 |
return count; } /* * How many references to @entry are currently swapped out? |
8334b9622 mm: /proc/pid/sma... |
1316 1317 1318 1319 1320 1321 |
* 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; |
235b62176 mm/swap: add clus... |
1322 |
struct swap_cluster_info *ci; |
8334b9622 mm: /proc/pid/sma... |
1323 1324 1325 |
struct page *page; pgoff_t offset; unsigned char *map; |
235b62176 mm/swap: add clus... |
1326 |
p = _swap_info_get(entry); |
8334b9622 mm: /proc/pid/sma... |
1327 1328 |
if (!p) return 0; |
235b62176 mm/swap: add clus... |
1329 1330 1331 1332 1333 |
offset = swp_offset(entry); ci = lock_cluster_or_swap_info(p, offset); count = swap_count(p->swap_map[offset]); |
8334b9622 mm: /proc/pid/sma... |
1334 1335 1336 1337 1338 |
if (!(count & COUNT_CONTINUED)) goto out; count &= ~COUNT_CONTINUED; n = SWAP_MAP_MAX + 1; |
8334b9622 mm: /proc/pid/sma... |
1339 1340 1341 1342 1343 |
page = vmalloc_to_page(p->swap_map + offset); offset &= ~PAGE_MASK; VM_BUG_ON(page_private(page) != SWP_CONTINUED); do { |
a8ae49917 mm/swapfile.c: us... |
1344 |
page = list_next_entry(page, lru); |
8334b9622 mm: /proc/pid/sma... |
1345 1346 1347 1348 1349 1350 1351 1352 |
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: |
235b62176 mm/swap: add clus... |
1353 |
unlock_cluster_or_swap_info(p, ci); |
8334b9622 mm: /proc/pid/sma... |
1354 1355 |
return count; } |
e07098294 mm, THP, swap: su... |
1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 |
#ifdef CONFIG_THP_SWAP static bool swap_page_trans_huge_swapped(struct swap_info_struct *si, swp_entry_t entry) { struct swap_cluster_info *ci; unsigned char *map = si->swap_map; unsigned long roffset = swp_offset(entry); unsigned long offset = round_down(roffset, SWAPFILE_CLUSTER); int i; bool ret = false; ci = lock_cluster_or_swap_info(si, offset); if (!ci || !cluster_is_huge(ci)) { if (map[roffset] != SWAP_HAS_CACHE) ret = true; goto unlock_out; } for (i = 0; i < SWAPFILE_CLUSTER; i++) { if (map[offset + i] != SWAP_HAS_CACHE) { ret = true; break; } } unlock_out: unlock_cluster_or_swap_info(si, ci); return ret; } static bool page_swapped(struct page *page) { swp_entry_t entry; struct swap_info_struct *si; if (likely(!PageTransCompound(page))) return page_swapcount(page) != 0; page = compound_head(page); entry.val = page_private(page); si = _swap_info_get(entry); if (si) return swap_page_trans_huge_swapped(si, entry); return false; } |
ba3c4ce6d mm, THP, swap: ma... |
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 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 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 |
static int page_trans_huge_map_swapcount(struct page *page, int *total_mapcount, int *total_swapcount) { int i, map_swapcount, _total_mapcount, _total_swapcount; unsigned long offset = 0; struct swap_info_struct *si; struct swap_cluster_info *ci = NULL; unsigned char *map = NULL; int mapcount, swapcount = 0; /* hugetlbfs shouldn't call it */ VM_BUG_ON_PAGE(PageHuge(page), page); if (likely(!PageTransCompound(page))) { mapcount = atomic_read(&page->_mapcount) + 1; if (total_mapcount) *total_mapcount = mapcount; if (PageSwapCache(page)) swapcount = page_swapcount(page); if (total_swapcount) *total_swapcount = swapcount; return mapcount + swapcount; } page = compound_head(page); _total_mapcount = _total_swapcount = map_swapcount = 0; if (PageSwapCache(page)) { swp_entry_t entry; entry.val = page_private(page); si = _swap_info_get(entry); if (si) { map = si->swap_map; offset = swp_offset(entry); } } if (map) ci = lock_cluster(si, offset); for (i = 0; i < HPAGE_PMD_NR; i++) { mapcount = atomic_read(&page[i]._mapcount) + 1; _total_mapcount += mapcount; if (map) { swapcount = swap_count(map[offset + i]); _total_swapcount += swapcount; } map_swapcount = max(map_swapcount, mapcount + swapcount); } unlock_cluster(ci); if (PageDoubleMap(page)) { map_swapcount -= 1; _total_mapcount -= HPAGE_PMD_NR; } mapcount = compound_mapcount(page); map_swapcount += mapcount; _total_mapcount += mapcount; if (total_mapcount) *total_mapcount = _total_mapcount; if (total_swapcount) *total_swapcount = _total_swapcount; return map_swapcount; } |
e07098294 mm, THP, swap: su... |
1463 1464 1465 |
#else #define swap_page_trans_huge_swapped(si, entry) swap_swapcount(si, entry) #define page_swapped(page) (page_swapcount(page) != 0) |
ba3c4ce6d mm, THP, swap: ma... |
1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 |
static int page_trans_huge_map_swapcount(struct page *page, int *total_mapcount, int *total_swapcount) { int mapcount, swapcount = 0; /* hugetlbfs shouldn't call it */ VM_BUG_ON_PAGE(PageHuge(page), page); mapcount = page_trans_huge_mapcount(page, total_mapcount); if (PageSwapCache(page)) swapcount = page_swapcount(page); if (total_swapcount) *total_swapcount = swapcount; return mapcount + swapcount; } |
e07098294 mm, THP, swap: su... |
1482 |
#endif |
8334b9622 mm: /proc/pid/sma... |
1483 |
/* |
7b1fe5979 mm: reuse_swap_pa... |
1484 1485 1486 1487 |
* 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. |
6d0a07edd mm: thp: calculat... |
1488 |
* |
ba3c4ce6d mm, THP, swap: ma... |
1489 |
* NOTE: total_map_swapcount should not be relied upon by the caller if |
6d0a07edd mm: thp: calculat... |
1490 1491 |
* reuse_swap_page() returns false, but it may be always overwritten * (see the other implementation for CONFIG_SWAP=n). |
1da177e4c Linux-2.6.12-rc2 |
1492 |
*/ |
ba3c4ce6d mm, THP, swap: ma... |
1493 |
bool reuse_swap_page(struct page *page, int *total_map_swapcount) |
1da177e4c Linux-2.6.12-rc2 |
1494 |
{ |
ba3c4ce6d mm, THP, swap: ma... |
1495 |
int count, total_mapcount, total_swapcount; |
c475a8ab6 [PATCH] can_share... |
1496 |
|
309381fea mm: dump page whe... |
1497 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
5ad646880 ksm: let shared p... |
1498 |
if (unlikely(PageKsm(page))) |
6d0a07edd mm: thp: calculat... |
1499 |
return false; |
ba3c4ce6d mm, THP, swap: ma... |
1500 1501 1502 1503 1504 1505 1506 1507 |
count = page_trans_huge_map_swapcount(page, &total_mapcount, &total_swapcount); if (total_map_swapcount) *total_map_swapcount = total_mapcount + total_swapcount; if (count == 1 && PageSwapCache(page) && (likely(!PageTransCompound(page)) || /* The remaining swap count will be freed soon */ total_swapcount == page_swapcount(page))) { |
f05714293 mm: support anony... |
1508 |
if (!PageWriteback(page)) { |
ba3c4ce6d mm, THP, swap: ma... |
1509 |
page = compound_head(page); |
7b1fe5979 mm: reuse_swap_pa... |
1510 1511 |
delete_from_swap_cache(page); SetPageDirty(page); |
f05714293 mm: support anony... |
1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 |
} else { swp_entry_t entry; struct swap_info_struct *p; entry.val = page_private(page); p = swap_info_get(entry); if (p->flags & SWP_STABLE_WRITES) { spin_unlock(&p->lock); return false; } spin_unlock(&p->lock); |
7b1fe5979 mm: reuse_swap_pa... |
1523 1524 |
} } |
ba3c4ce6d mm, THP, swap: ma... |
1525 |
|
5ad646880 ksm: let shared p... |
1526 |
return count <= 1; |
1da177e4c Linux-2.6.12-rc2 |
1527 1528 1529 |
} /* |
a2c43eed8 mm: try_to_free_s... |
1530 1531 |
* 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 |
1532 |
*/ |
a2c43eed8 mm: try_to_free_s... |
1533 |
int try_to_free_swap(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
1534 |
{ |
309381fea mm: dump page whe... |
1535 |
VM_BUG_ON_PAGE(!PageLocked(page), page); |
1da177e4c Linux-2.6.12-rc2 |
1536 1537 1538 1539 1540 |
if (!PageSwapCache(page)) return 0; if (PageWriteback(page)) return 0; |
e07098294 mm, THP, swap: su... |
1541 |
if (page_swapped(page)) |
1da177e4c Linux-2.6.12-rc2 |
1542 |
return 0; |
b73d7fcec swap: prevent reu... |
1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 |
/* * 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... |
1555 |
* Hibernation suspends storage while it is writing the image |
f90ac3982 mm: avoid liveloc... |
1556 |
* to disk so check that here. |
b73d7fcec swap: prevent reu... |
1557 |
*/ |
f90ac3982 mm: avoid liveloc... |
1558 |
if (pm_suspended_storage()) |
b73d7fcec swap: prevent reu... |
1559 |
return 0; |
e07098294 mm, THP, swap: su... |
1560 |
page = compound_head(page); |
a2c43eed8 mm: try_to_free_s... |
1561 1562 1563 |
delete_from_swap_cache(page); SetPageDirty(page); return 1; |
68a22394c vmscan: free swap... |
1564 1565 1566 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
1567 1568 1569 |
* 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... |
1570 |
int free_swap_and_cache(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
1571 |
{ |
2509ef26d badpage: zap prin... |
1572 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
1573 |
struct page *page = NULL; |
7c00bafee mm/swap: free swa... |
1574 |
unsigned char count; |
1da177e4c Linux-2.6.12-rc2 |
1575 |
|
a7420aa54 HWPOISON: Add sup... |
1576 |
if (non_swap_entry(entry)) |
2509ef26d badpage: zap prin... |
1577 |
return 1; |
0697212a4 [PATCH] Swapless ... |
1578 |
|
7c00bafee mm/swap: free swa... |
1579 |
p = _swap_info_get(entry); |
1da177e4c Linux-2.6.12-rc2 |
1580 |
if (p) { |
7c00bafee mm/swap: free swa... |
1581 |
count = __swap_entry_free(p, entry, 1); |
e07098294 mm, THP, swap: su... |
1582 1583 |
if (count == SWAP_HAS_CACHE && !swap_page_trans_huge_swapped(p, entry)) { |
33806f06d swap: make each s... |
1584 |
page = find_get_page(swap_address_space(entry), |
f6ab1f7f6 mm, swap: use off... |
1585 |
swp_offset(entry)); |
8413ac9d8 mm: page lock use... |
1586 |
if (page && !trylock_page(page)) { |
09cbfeaf1 mm, fs: get rid o... |
1587 |
put_page(page); |
93fac7041 [PATCH] mm: sched... |
1588 1589 |
page = NULL; } |
7c00bafee mm/swap: free swa... |
1590 |
} else if (!count) |
67afa38e0 mm/swap: add cach... |
1591 |
free_swap_slot(entry); |
1da177e4c Linux-2.6.12-rc2 |
1592 1593 |
} if (page) { |
a2c43eed8 mm: try_to_free_s... |
1594 1595 1596 1597 |
/* * Not mapped elsewhere, or swap space full? Free it! * Also recheck PageSwapCache now page is locked (above). */ |
93fac7041 [PATCH] mm: sched... |
1598 |
if (PageSwapCache(page) && !PageWriteback(page) && |
322b8afe4 mm, swap: Fix a r... |
1599 |
(!page_mapped(page) || mem_cgroup_swap_full(page)) && |
e07098294 mm, THP, swap: su... |
1600 1601 |
!swap_page_trans_huge_swapped(p, entry)) { page = compound_head(page); |
1da177e4c Linux-2.6.12-rc2 |
1602 1603 1604 1605 |
delete_from_swap_cache(page); SetPageDirty(page); } unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
1606 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1607 |
} |
2509ef26d badpage: zap prin... |
1608 |
return p != NULL; |
1da177e4c Linux-2.6.12-rc2 |
1609 |
} |
b0cb1a19d Replace CONFIG_SO... |
1610 |
#ifdef CONFIG_HIBERNATION |
f577eb30a [PATCH] swsusp: l... |
1611 |
/* |
915bae9eb [PATCH] swsusp: u... |
1612 |
* Find the swap type that corresponds to given device (if any). |
f577eb30a [PATCH] swsusp: l... |
1613 |
* |
915bae9eb [PATCH] swsusp: u... |
1614 1615 1616 1617 |
* @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... |
1618 |
*/ |
7bf236874 [PATCH] swsusp: D... |
1619 |
int swap_type_of(dev_t device, sector_t offset, struct block_device **bdev_p) |
f577eb30a [PATCH] swsusp: l... |
1620 |
{ |
915bae9eb [PATCH] swsusp: u... |
1621 |
struct block_device *bdev = NULL; |
efa90a981 swap_info: change... |
1622 |
int type; |
f577eb30a [PATCH] swsusp: l... |
1623 |
|
915bae9eb [PATCH] swsusp: u... |
1624 1625 |
if (device) bdev = bdget(device); |
f577eb30a [PATCH] swsusp: l... |
1626 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
1627 1628 |
for (type = 0; type < nr_swapfiles; type++) { struct swap_info_struct *sis = swap_info[type]; |
f577eb30a [PATCH] swsusp: l... |
1629 |
|
915bae9eb [PATCH] swsusp: u... |
1630 |
if (!(sis->flags & SWP_WRITEOK)) |
f577eb30a [PATCH] swsusp: l... |
1631 |
continue; |
b6b5bce35 [PATCH] swsusp: F... |
1632 |
|
915bae9eb [PATCH] swsusp: u... |
1633 |
if (!bdev) { |
7bf236874 [PATCH] swsusp: D... |
1634 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
1635 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
1636 |
|
6e1819d61 [PATCH] swsusp: u... |
1637 |
spin_unlock(&swap_lock); |
efa90a981 swap_info: change... |
1638 |
return type; |
6e1819d61 [PATCH] swsusp: u... |
1639 |
} |
915bae9eb [PATCH] swsusp: u... |
1640 |
if (bdev == sis->bdev) { |
9625a5f28 swap_info: includ... |
1641 |
struct swap_extent *se = &sis->first_swap_extent; |
915bae9eb [PATCH] swsusp: u... |
1642 |
|
915bae9eb [PATCH] swsusp: u... |
1643 |
if (se->start_block == offset) { |
7bf236874 [PATCH] swsusp: D... |
1644 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
1645 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
1646 |
|
915bae9eb [PATCH] swsusp: u... |
1647 1648 |
spin_unlock(&swap_lock); bdput(bdev); |
efa90a981 swap_info: change... |
1649 |
return type; |
915bae9eb [PATCH] swsusp: u... |
1650 |
} |
f577eb30a [PATCH] swsusp: l... |
1651 1652 1653 |
} } spin_unlock(&swap_lock); |
915bae9eb [PATCH] swsusp: u... |
1654 1655 |
if (bdev) bdput(bdev); |
f577eb30a [PATCH] swsusp: l... |
1656 1657 1658 1659 |
return -ENODEV; } /* |
73c34b6ac swap_info: miscel... |
1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 |
* 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_... |
1671 |
return map_swap_entry(swp_entry(type, offset), &bdev); |
73c34b6ac swap_info: miscel... |
1672 1673 1674 |
} /* |
f577eb30a [PATCH] swsusp: l... |
1675 1676 1677 1678 1679 1680 1681 1682 |
* 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... |
1683 1684 1685 |
spin_lock(&swap_lock); if ((unsigned int)type < nr_swapfiles) { struct swap_info_struct *sis = swap_info[type]; |
ec8acf20a swap: add per-par... |
1686 |
spin_lock(&sis->lock); |
efa90a981 swap_info: change... |
1687 1688 |
if (sis->flags & SWP_WRITEOK) { n = sis->pages; |
f577eb30a [PATCH] swsusp: l... |
1689 |
if (free) |
efa90a981 swap_info: change... |
1690 |
n -= sis->inuse_pages; |
f577eb30a [PATCH] swsusp: l... |
1691 |
} |
ec8acf20a swap: add per-par... |
1692 |
spin_unlock(&sis->lock); |
f577eb30a [PATCH] swsusp: l... |
1693 |
} |
efa90a981 swap_info: change... |
1694 |
spin_unlock(&swap_lock); |
f577eb30a [PATCH] swsusp: l... |
1695 1696 |
return n; } |
73c34b6ac swap_info: miscel... |
1697 |
#endif /* CONFIG_HIBERNATION */ |
f577eb30a [PATCH] swsusp: l... |
1698 |
|
9f8bdb3f3 mm: make swapoff ... |
1699 |
static inline int pte_same_as_swp(pte_t pte, pte_t swp_pte) |
179ef71cb mm: save soft-dir... |
1700 |
{ |
9f8bdb3f3 mm: make swapoff ... |
1701 |
return pte_same(pte_swp_clear_soft_dirty(pte), swp_pte); |
179ef71cb mm: save soft-dir... |
1702 |
} |
1da177e4c Linux-2.6.12-rc2 |
1703 |
/* |
72866f6f2 [PATCH] mm: anon ... |
1704 1705 1706 |
* 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 |
1707 |
*/ |
044d66c1d memcgroup: reinst... |
1708 |
static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd, |
1da177e4c Linux-2.6.12-rc2 |
1709 1710 |
unsigned long addr, swp_entry_t entry, struct page *page) { |
9e16b7fb1 mm,ksm: swapoff m... |
1711 |
struct page *swapcache; |
72835c86c mm: unify remaini... |
1712 |
struct mem_cgroup *memcg; |
044d66c1d memcgroup: reinst... |
1713 1714 1715 |
spinlock_t *ptl; pte_t *pte; int ret = 1; |
9e16b7fb1 mm,ksm: swapoff m... |
1716 1717 1718 1719 |
swapcache = page; page = ksm_might_need_to_copy(page, vma, addr); if (unlikely(!page)) return -ENOMEM; |
f627c2f53 memcg: adjust to ... |
1720 1721 |
if (mem_cgroup_try_charge(page, vma->vm_mm, GFP_KERNEL, &memcg, false)) { |
044d66c1d memcgroup: reinst... |
1722 |
ret = -ENOMEM; |
85d9fc89f memcg: fix refcnt... |
1723 1724 |
goto out_nolock; } |
044d66c1d memcgroup: reinst... |
1725 1726 |
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); |
9f8bdb3f3 mm: make swapoff ... |
1727 |
if (unlikely(!pte_same_as_swp(*pte, swp_entry_to_pte(entry)))) { |
f627c2f53 memcg: adjust to ... |
1728 |
mem_cgroup_cancel_charge(page, memcg, false); |
044d66c1d memcgroup: reinst... |
1729 1730 1731 |
ret = 0; goto out; } |
8a9f3ccd2 Memory controller... |
1732 |
|
b084d4353 mm: count swap usage |
1733 |
dec_mm_counter(vma->vm_mm, MM_SWAPENTS); |
d559db086 mm: clean up mm_c... |
1734 |
inc_mm_counter(vma->vm_mm, MM_ANONPAGES); |
1da177e4c Linux-2.6.12-rc2 |
1735 1736 1737 |
get_page(page); set_pte_at(vma->vm_mm, addr, pte, pte_mkold(mk_pte(page, vma->vm_page_prot))); |
00501b531 mm: memcontrol: r... |
1738 |
if (page == swapcache) { |
d281ee614 rmap: add argumen... |
1739 |
page_add_anon_rmap(page, vma, addr, false); |
f627c2f53 memcg: adjust to ... |
1740 |
mem_cgroup_commit_charge(page, memcg, true, false); |
00501b531 mm: memcontrol: r... |
1741 |
} else { /* ksm created a completely new copy */ |
d281ee614 rmap: add argumen... |
1742 |
page_add_new_anon_rmap(page, vma, addr, false); |
f627c2f53 memcg: adjust to ... |
1743 |
mem_cgroup_commit_charge(page, memcg, false, false); |
00501b531 mm: memcontrol: r... |
1744 1745 |
lru_cache_add_active_or_unevictable(page, vma); } |
1da177e4c Linux-2.6.12-rc2 |
1746 1747 1748 1749 1750 1751 |
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... |
1752 1753 |
out: pte_unmap_unlock(pte, ptl); |
85d9fc89f memcg: fix refcnt... |
1754 |
out_nolock: |
9e16b7fb1 mm,ksm: swapoff m... |
1755 1756 1757 1758 |
if (page != swapcache) { unlock_page(page); put_page(page); } |
044d66c1d memcgroup: reinst... |
1759 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1760 1761 1762 1763 1764 1765 |
} 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 |
1766 |
pte_t swp_pte = swp_entry_to_pte(entry); |
705e87c0c [PATCH] mm: pte_o... |
1767 |
pte_t *pte; |
8a9f3ccd2 Memory controller... |
1768 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1769 |
|
044d66c1d memcgroup: reinst... |
1770 1771 1772 1773 1774 1775 1776 |
/* * 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... |
1777 |
* preemptable whenever CONFIG_PREEMPT but not CONFIG_HIGHPTE. |
044d66c1d memcgroup: reinst... |
1778 1779 |
*/ pte = pte_offset_map(pmd, addr); |
1da177e4c Linux-2.6.12-rc2 |
1780 1781 1782 1783 1784 |
do { /* * swapoff spends a _lot_ of time in this loop! * Test inline before going to call unuse_pte. */ |
9f8bdb3f3 mm: make swapoff ... |
1785 |
if (unlikely(pte_same_as_swp(*pte, swp_pte))) { |
044d66c1d memcgroup: reinst... |
1786 1787 1788 1789 1790 |
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 |
1791 1792 |
} } while (pte++, addr += PAGE_SIZE, addr != end); |
044d66c1d memcgroup: reinst... |
1793 1794 |
pte_unmap(pte - 1); out: |
8a9f3ccd2 Memory controller... |
1795 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1796 1797 1798 1799 1800 1801 1802 1803 |
} 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... |
1804 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1805 1806 1807 |
pmd = pmd_offset(pud, addr); do { |
dc644a073 mm: add three mor... |
1808 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1809 |
next = pmd_addr_end(addr, end); |
1a5a9906d mm: thp: fix pmd_... |
1810 |
if (pmd_none_or_trans_huge_or_clear_bad(pmd)) |
1da177e4c Linux-2.6.12-rc2 |
1811 |
continue; |
8a9f3ccd2 Memory controller... |
1812 1813 1814 |
ret = unuse_pte_range(vma, pmd, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1815 1816 1817 |
} while (pmd++, addr = next, addr != end); return 0; } |
c2febafc6 mm: convert gener... |
1818 |
static inline int unuse_pud_range(struct vm_area_struct *vma, p4d_t *p4d, |
1da177e4c Linux-2.6.12-rc2 |
1819 1820 1821 1822 1823 |
unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { pud_t *pud; unsigned long next; |
8a9f3ccd2 Memory controller... |
1824 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1825 |
|
c2febafc6 mm: convert gener... |
1826 |
pud = pud_offset(p4d, addr); |
1da177e4c Linux-2.6.12-rc2 |
1827 1828 1829 1830 |
do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) continue; |
8a9f3ccd2 Memory controller... |
1831 1832 1833 |
ret = unuse_pmd_range(vma, pud, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1834 1835 1836 |
} while (pud++, addr = next, addr != end); return 0; } |
c2febafc6 mm: convert gener... |
1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 |
static inline int unuse_p4d_range(struct vm_area_struct *vma, pgd_t *pgd, unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { p4d_t *p4d; unsigned long next; int ret; p4d = p4d_offset(pgd, addr); do { next = p4d_addr_end(addr, end); if (p4d_none_or_clear_bad(p4d)) continue; ret = unuse_pud_range(vma, p4d, addr, next, entry, page); if (ret) return ret; } while (p4d++, addr = next, addr != end); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1856 1857 1858 1859 1860 |
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... |
1861 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1862 |
|
3ca7b3c5b mm: define PAGE_M... |
1863 |
if (page_anon_vma(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 |
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; |
c2febafc6 mm: convert gener... |
1879 |
ret = unuse_p4d_range(vma, pgd, addr, next, entry, page); |
8a9f3ccd2 Memory controller... |
1880 1881 |
if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1882 1883 1884 1885 1886 1887 1888 1889 |
} 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... |
1890 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1891 1892 1893 |
if (!down_read_trylock(&mm->mmap_sem)) { /* |
7d03431cf swapfile/vmscan: ... |
1894 1895 |
* 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 |
1896 |
*/ |
c475a8ab6 [PATCH] can_share... |
1897 |
activate_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1898 1899 1900 1901 |
unlock_page(page); down_read(&mm->mmap_sem); lock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1902 |
for (vma = mm->mmap; vma; vma = vma->vm_next) { |
8a9f3ccd2 Memory controller... |
1903 |
if (vma->anon_vma && (ret = unuse_vma(vma, entry, page))) |
1da177e4c Linux-2.6.12-rc2 |
1904 |
break; |
dc644a073 mm: add three mor... |
1905 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1906 |
} |
1da177e4c Linux-2.6.12-rc2 |
1907 |
up_read(&mm->mmap_sem); |
8a9f3ccd2 Memory controller... |
1908 |
return (ret < 0)? ret: 0; |
1da177e4c Linux-2.6.12-rc2 |
1909 1910 1911 |
} /* |
38b5faf4b mm: frontswap: co... |
1912 1913 |
* 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 |
1914 1915 |
* Recycle to start on reaching the end, returning 0 when empty. */ |
6eb396dc4 [PATCH] swap: swa... |
1916 |
static unsigned int find_next_to_unuse(struct swap_info_struct *si, |
38b5faf4b mm: frontswap: co... |
1917 |
unsigned int prev, bool frontswap) |
1da177e4c Linux-2.6.12-rc2 |
1918 |
{ |
6eb396dc4 [PATCH] swap: swa... |
1919 1920 |
unsigned int max = si->max; unsigned int i = prev; |
8d69aaee8 swap_info: swap_m... |
1921 |
unsigned char count; |
1da177e4c Linux-2.6.12-rc2 |
1922 1923 |
/* |
5d337b919 [PATCH] swap: swa... |
1924 |
* No need for swap_lock here: we're just looking |
1da177e4c Linux-2.6.12-rc2 |
1925 1926 |
* 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... |
1927 |
* allocations from this area (while holding swap_lock). |
1da177e4c Linux-2.6.12-rc2 |
1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 |
*/ 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; } |
4db0c3c29 mm: remove rest o... |
1943 |
count = READ_ONCE(si->swap_map[i]); |
355cfa73d mm: modify swap_m... |
1944 |
if (count && swap_count(count) != SWAP_MAP_BAD) |
dc644a073 mm: add three mor... |
1945 1946 1947 1948 |
if (!frontswap || frontswap_test(si, i)) break; if ((i % LATENCY_LIMIT) == 0) cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
1949 1950 1951 1952 1953 1954 1955 1956 |
} 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... |
1957 1958 1959 |
* * 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 |
1960 |
*/ |
38b5faf4b mm: frontswap: co... |
1961 1962 |
int try_to_unuse(unsigned int type, bool frontswap, unsigned long pages_to_unuse) |
1da177e4c Linux-2.6.12-rc2 |
1963 |
{ |
efa90a981 swap_info: change... |
1964 |
struct swap_info_struct *si = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
1965 |
struct mm_struct *start_mm; |
edfe23dac swap: fix races e... |
1966 1967 1968 1969 1970 |
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... |
1971 |
unsigned char swcount; |
1da177e4c Linux-2.6.12-rc2 |
1972 1973 |
struct page *page; swp_entry_t entry; |
6eb396dc4 [PATCH] swap: swa... |
1974 |
unsigned int i = 0; |
1da177e4c Linux-2.6.12-rc2 |
1975 |
int retval = 0; |
1da177e4c Linux-2.6.12-rc2 |
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 |
/* * 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... |
1989 |
* that. |
1da177e4c Linux-2.6.12-rc2 |
1990 1991 |
*/ start_mm = &init_mm; |
3fce371bf mm: add new mmget... |
1992 |
mmget(&init_mm); |
1da177e4c Linux-2.6.12-rc2 |
1993 1994 1995 1996 1997 1998 |
/* * 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... |
1999 |
while ((i = find_next_to_unuse(si, i, frontswap)) != 0) { |
1da177e4c Linux-2.6.12-rc2 |
2000 2001 2002 2003 |
if (signal_pending(current)) { retval = -EINTR; break; } |
886bb7e9c swapfile: remove ... |
2004 |
/* |
1da177e4c Linux-2.6.12-rc2 |
2005 2006 |
* Get a page for the entry, using the existing swap * cache page if there is one. Otherwise, get a clean |
886bb7e9c swapfile: remove ... |
2007 |
* page and read the swap into it. |
1da177e4c Linux-2.6.12-rc2 |
2008 2009 2010 |
*/ swap_map = &si->swap_map[i]; entry = swp_entry(type, i); |
02098feaa swapin needs gfp_... |
2011 |
page = read_swap_cache_async(entry, |
23955622f swap: add block i... |
2012 |
GFP_HIGHUSER_MOVABLE, NULL, 0, false); |
1da177e4c Linux-2.6.12-rc2 |
2013 2014 2015 2016 2017 2018 2019 |
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... |
2020 2021 2022 2023 2024 2025 2026 2027 2028 |
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 |
2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 |
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; |
3fce371bf mm: add new mmget... |
2040 |
mmget(&init_mm); |
1da177e4c Linux-2.6.12-rc2 |
2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 |
} /* * 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 |
2058 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
2059 |
swcount = *swap_map; |
aaa468653 swap_info: note S... |
2060 2061 2062 2063 2064 2065 |
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 |
2066 |
} |
aaa468653 swap_info: note S... |
2067 2068 |
if (swap_count(swcount) && start_mm != &init_mm) retval = unuse_mm(start_mm, entry, page); |
355cfa73d mm: modify swap_m... |
2069 |
if (swap_count(*swap_map)) { |
1da177e4c Linux-2.6.12-rc2 |
2070 2071 2072 2073 2074 |
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; |
3fce371bf mm: add new mmget... |
2075 2076 |
mmget(new_start_mm); mmget(prev_mm); |
1da177e4c Linux-2.6.12-rc2 |
2077 |
spin_lock(&mmlist_lock); |
aaa468653 swap_info: note S... |
2078 |
while (swap_count(*swap_map) && !retval && |
1da177e4c Linux-2.6.12-rc2 |
2079 2080 |
(p = p->next) != &start_mm->mmlist) { mm = list_entry(p, struct mm_struct, mmlist); |
388f79345 mm: use mmget_not... |
2081 |
if (!mmget_not_zero(mm)) |
1da177e4c Linux-2.6.12-rc2 |
2082 |
continue; |
1da177e4c Linux-2.6.12-rc2 |
2083 2084 2085 2086 2087 2088 2089 |
spin_unlock(&mmlist_lock); mmput(prev_mm); prev_mm = mm; cond_resched(); swcount = *swap_map; |
355cfa73d mm: modify swap_m... |
2090 |
if (!swap_count(swcount)) /* any usage ? */ |
1da177e4c Linux-2.6.12-rc2 |
2091 |
; |
aaa468653 swap_info: note S... |
2092 |
else if (mm == &init_mm) |
1da177e4c Linux-2.6.12-rc2 |
2093 |
set_start_mm = 1; |
aaa468653 swap_info: note S... |
2094 |
else |
1da177e4c Linux-2.6.12-rc2 |
2095 |
retval = unuse_mm(mm, entry, page); |
355cfa73d mm: modify swap_m... |
2096 |
|
32c5fc10e mm: remove incorr... |
2097 |
if (set_start_mm && *swap_map < swcount) { |
1da177e4c Linux-2.6.12-rc2 |
2098 |
mmput(new_start_mm); |
3fce371bf mm: add new mmget... |
2099 |
mmget(mm); |
1da177e4c Linux-2.6.12-rc2 |
2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 |
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); |
09cbfeaf1 mm, fs: get rid o... |
2112 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2113 2114 2115 2116 |
break; } /* |
1da177e4c Linux-2.6.12-rc2 |
2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 |
* 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... |
2128 2129 2130 2131 2132 2133 |
* * 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 |
2134 |
*/ |
355cfa73d mm: modify swap_m... |
2135 2136 |
if (swap_count(*swap_map) && PageDirty(page) && PageSwapCache(page)) { |
1da177e4c Linux-2.6.12-rc2 |
2137 2138 2139 |
struct writeback_control wbc = { .sync_mode = WB_SYNC_NONE, }; |
e07098294 mm, THP, swap: su... |
2140 |
swap_writepage(compound_head(page), &wbc); |
1da177e4c Linux-2.6.12-rc2 |
2141 2142 2143 |
lock_page(page); wait_on_page_writeback(page); } |
68bdc8d64 mm: try_to_unuse ... |
2144 2145 2146 2147 2148 2149 2150 2151 2152 |
/* * 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) && |
e07098294 mm, THP, swap: su... |
2153 |
likely(page_private(page) == entry.val) && |
89b038770 mm, swap: fix swa... |
2154 2155 |
(!PageTransCompound(page) || !swap_page_trans_huge_swapped(si, entry))) |
e07098294 mm, THP, swap: su... |
2156 |
delete_from_swap_cache(compound_head(page)); |
1da177e4c Linux-2.6.12-rc2 |
2157 2158 2159 2160 |
/* * 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... |
2161 |
* mark page dirty so shrink_page_list will preserve it. |
1da177e4c Linux-2.6.12-rc2 |
2162 2163 2164 |
*/ SetPageDirty(page); unlock_page(page); |
09cbfeaf1 mm, fs: get rid o... |
2165 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
2166 2167 2168 2169 2170 2171 |
/* * Make sure that we aren't completely killing * interactive performance. */ cond_resched(); |
38b5faf4b mm: frontswap: co... |
2172 2173 2174 2175 |
if (frontswap && pages_to_unuse > 0) { if (!--pages_to_unuse) break; } |
1da177e4c Linux-2.6.12-rc2 |
2176 2177 2178 |
} mmput(start_mm); |
1da177e4c Linux-2.6.12-rc2 |
2179 2180 2181 2182 |
return retval; } /* |
5d337b919 [PATCH] swap: swa... |
2183 2184 2185 |
* 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 |
2186 2187 2188 2189 2190 |
* 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... |
2191 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
2192 |
|
efa90a981 swap_info: change... |
2193 2194 |
for (type = 0; type < nr_swapfiles; type++) if (swap_info[type]->inuse_pages) |
1da177e4c Linux-2.6.12-rc2 |
2195 2196 2197 2198 2199 2200 2201 2202 2203 |
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_... |
2204 2205 2206 |
* 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 |
2207 |
*/ |
d4906e1aa swap: rework map_... |
2208 |
static sector_t map_swap_entry(swp_entry_t entry, struct block_device **bdev) |
1da177e4c Linux-2.6.12-rc2 |
2209 |
{ |
f29ad6a99 swap_info: privat... |
2210 2211 2212 2213 |
struct swap_info_struct *sis; struct swap_extent *start_se; struct swap_extent *se; pgoff_t offset; |
efa90a981 swap_info: change... |
2214 |
sis = swap_info[swp_type(entry)]; |
f29ad6a99 swap_info: privat... |
2215 2216 2217 2218 2219 |
*bdev = sis->bdev; offset = swp_offset(entry); start_se = sis->curr_swap_extent; se = start_se; |
1da177e4c Linux-2.6.12-rc2 |
2220 2221 |
for ( ; ; ) { |
1da177e4c Linux-2.6.12-rc2 |
2222 2223 2224 2225 |
if (se->start_page <= offset && offset < (se->start_page + se->nr_pages)) { return se->start_block + (offset - se->start_page); } |
a8ae49917 mm/swapfile.c: us... |
2226 |
se = list_next_entry(se, list); |
1da177e4c Linux-2.6.12-rc2 |
2227 2228 2229 2230 2231 2232 |
sis->curr_swap_extent = se; BUG_ON(se == start_se); /* It *must* be present */ } } /* |
d4906e1aa swap: rework map_... |
2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 |
* 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 |
2243 2244 2245 2246 |
* Free all of a swapdev's extent information */ static void destroy_swap_extents(struct swap_info_struct *sis) { |
9625a5f28 swap_info: includ... |
2247 |
while (!list_empty(&sis->first_swap_extent.list)) { |
1da177e4c Linux-2.6.12-rc2 |
2248 |
struct swap_extent *se; |
a8ae49917 mm/swapfile.c: us... |
2249 |
se = list_first_entry(&sis->first_swap_extent.list, |
1da177e4c Linux-2.6.12-rc2 |
2250 2251 2252 2253 |
struct swap_extent, list); list_del(&se->list); kfree(se); } |
62c230bc1 mm: add support f... |
2254 2255 2256 2257 2258 2259 2260 2261 |
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 |
2262 2263 2264 2265 |
} /* * Add a block range (and the corresponding page range) into this swapdev's |
11d31886d [PATCH] swap: swa... |
2266 |
* extent list. The extent list is kept sorted in page order. |
1da177e4c Linux-2.6.12-rc2 |
2267 |
* |
11d31886d [PATCH] swap: swa... |
2268 |
* This function rather assumes that it is called in ascending page order. |
1da177e4c Linux-2.6.12-rc2 |
2269 |
*/ |
a509bc1a9 mm: swap: impleme... |
2270 |
int |
1da177e4c Linux-2.6.12-rc2 |
2271 2272 2273 2274 2275 2276 |
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... |
2277 2278 2279 2280 2281 2282 2283 2284 2285 |
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 |
2286 |
se = list_entry(lh, struct swap_extent, list); |
11d31886d [PATCH] swap: swa... |
2287 2288 |
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 |
2289 2290 2291 2292 |
/* Merge it */ se->nr_pages += nr_pages; return 0; } |
1da177e4c Linux-2.6.12-rc2 |
2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 |
} /* * 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... |
2304 |
list_add_tail(&new_se->list, &sis->first_swap_extent.list); |
53092a740 [PATCH] swap: sho... |
2305 |
return 1; |
1da177e4c Linux-2.6.12-rc2 |
2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 |
} /* * 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... |
2328 |
* For S_ISREG swapfiles we set S_SWAPFILE across the life of the swapon. This |
1da177e4c Linux-2.6.12-rc2 |
2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 |
* 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... |
2339 |
static int setup_swap_extents(struct swap_info_struct *sis, sector_t *span) |
1da177e4c Linux-2.6.12-rc2 |
2340 |
{ |
62c230bc1 mm: add support f... |
2341 2342 2343 |
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 |
2344 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
2345 2346 |
if (S_ISBLK(inode->i_mode)) { ret = add_swap_extent(sis, 0, sis->max, 0); |
53092a740 [PATCH] swap: sho... |
2347 |
*span = sis->pages; |
a509bc1a9 mm: swap: impleme... |
2348 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
2349 |
} |
62c230bc1 mm: add support f... |
2350 |
if (mapping->a_ops->swap_activate) { |
a509bc1a9 mm: swap: impleme... |
2351 |
ret = mapping->a_ops->swap_activate(sis, swap_file, span); |
62c230bc1 mm: add support f... |
2352 2353 2354 2355 2356 |
if (!ret) { sis->flags |= SWP_FILE; ret = add_swap_extent(sis, 0, sis->max, 0); *span = sis->pages; } |
a509bc1a9 mm: swap: impleme... |
2357 |
return ret; |
62c230bc1 mm: add support f... |
2358 |
} |
a509bc1a9 mm: swap: impleme... |
2359 |
return generic_swapfile_activate(sis, swap_file, span); |
1da177e4c Linux-2.6.12-rc2 |
2360 |
} |
a2468cc9b swap: choose swap... |
2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 |
static int swap_node(struct swap_info_struct *p) { struct block_device *bdev; if (p->bdev) bdev = p->bdev; else bdev = p->swap_file->f_inode->i_sb->s_bdev; return bdev ? bdev->bd_disk->node_id : NUMA_NO_NODE; } |
cf0cac0a0 mm: refactor rein... |
2372 |
static void _enable_swap_info(struct swap_info_struct *p, int prio, |
2a8f94493 swap: change bloc... |
2373 2374 |
unsigned char *swap_map, struct swap_cluster_info *cluster_info) |
40531542e sys_swapon: separ... |
2375 |
{ |
a2468cc9b swap: choose swap... |
2376 |
int i; |
40531542e sys_swapon: separ... |
2377 2378 2379 2380 |
if (prio >= 0) p->prio = prio; else p->prio = --least_priority; |
18ab4d4ce swap: change swap... |
2381 2382 2383 2384 2385 |
/* * the plist prio is negated because plist ordering is * low-to-high, while swap ordering is high-to-low */ p->list.prio = -p->prio; |
a2468cc9b swap: choose swap... |
2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 |
for_each_node(i) { if (p->prio >= 0) p->avail_lists[i].prio = -p->prio; else { if (swap_node(p) == i) p->avail_lists[i].prio = 1; else p->avail_lists[i].prio = -p->prio; } } |
40531542e sys_swapon: separ... |
2396 |
p->swap_map = swap_map; |
2a8f94493 swap: change bloc... |
2397 |
p->cluster_info = cluster_info; |
40531542e sys_swapon: separ... |
2398 |
p->flags |= SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
2399 |
atomic_long_add(p->pages, &nr_swap_pages); |
40531542e sys_swapon: separ... |
2400 |
total_swap_pages += p->pages; |
adfab836f swap: change swap... |
2401 |
assert_spin_locked(&swap_lock); |
adfab836f swap: change swap... |
2402 |
/* |
18ab4d4ce swap: change swap... |
2403 2404 2405 2406 2407 2408 2409 2410 |
* 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... |
2411 |
*/ |
18ab4d4ce swap: change swap... |
2412 |
plist_add(&p->list, &swap_active_head); |
a2468cc9b swap: choose swap... |
2413 |
add_to_avail_list(p); |
cf0cac0a0 mm: refactor rein... |
2414 2415 2416 2417 |
} static void enable_swap_info(struct swap_info_struct *p, int prio, unsigned char *swap_map, |
2a8f94493 swap: change bloc... |
2418 |
struct swap_cluster_info *cluster_info, |
cf0cac0a0 mm: refactor rein... |
2419 2420 |
unsigned long *frontswap_map) { |
4f89849da frontswap: get ri... |
2421 |
frontswap_init(p->type, frontswap_map); |
cf0cac0a0 mm: refactor rein... |
2422 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
2423 |
spin_lock(&p->lock); |
2a8f94493 swap: change bloc... |
2424 |
_enable_swap_info(p, prio, swap_map, cluster_info); |
ec8acf20a swap: add per-par... |
2425 |
spin_unlock(&p->lock); |
cf0cac0a0 mm: refactor rein... |
2426 2427 2428 2429 2430 2431 |
spin_unlock(&swap_lock); } static void reinsert_swap_info(struct swap_info_struct *p) { spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
2432 |
spin_lock(&p->lock); |
2a8f94493 swap: change bloc... |
2433 |
_enable_swap_info(p, p->prio, p->swap_map, p->cluster_info); |
ec8acf20a swap: add per-par... |
2434 |
spin_unlock(&p->lock); |
40531542e sys_swapon: separ... |
2435 2436 |
spin_unlock(&swap_lock); } |
67afa38e0 mm/swap: add cach... |
2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 |
bool has_usable_swap(void) { bool ret = true; spin_lock(&swap_lock); if (plist_head_empty(&swap_active_head)) ret = false; spin_unlock(&swap_lock); return ret; } |
c4ea37c26 [CVE-2009-0029] S... |
2447 |
SYSCALL_DEFINE1(swapoff, const char __user *, specialfile) |
1da177e4c Linux-2.6.12-rc2 |
2448 |
{ |
73c34b6ac swap_info: miscel... |
2449 |
struct swap_info_struct *p = NULL; |
8d69aaee8 swap_info: swap_m... |
2450 |
unsigned char *swap_map; |
2a8f94493 swap: change bloc... |
2451 |
struct swap_cluster_info *cluster_info; |
4f89849da frontswap: get ri... |
2452 |
unsigned long *frontswap_map; |
1da177e4c Linux-2.6.12-rc2 |
2453 2454 2455 |
struct file *swap_file, *victim; struct address_space *mapping; struct inode *inode; |
91a27b2a7 vfs: define struc... |
2456 |
struct filename *pathname; |
adfab836f swap: change swap... |
2457 |
int err, found = 0; |
5b808a230 swap: fix set_blo... |
2458 |
unsigned int old_block_size; |
886bb7e9c swapfile: remove ... |
2459 |
|
1da177e4c Linux-2.6.12-rc2 |
2460 2461 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
191c54244 mm: collapse secu... |
2462 |
BUG_ON(!current->mm); |
1da177e4c Linux-2.6.12-rc2 |
2463 |
pathname = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
2464 |
if (IS_ERR(pathname)) |
f58b59c1d swapfile: fix nam... |
2465 |
return PTR_ERR(pathname); |
1da177e4c Linux-2.6.12-rc2 |
2466 |
|
669abf4e5 vfs: make path_op... |
2467 |
victim = file_open_name(pathname, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
2468 2469 2470 2471 2472 |
err = PTR_ERR(victim); if (IS_ERR(victim)) goto out; mapping = victim->f_mapping; |
5d337b919 [PATCH] swap: swa... |
2473 |
spin_lock(&swap_lock); |
18ab4d4ce swap: change swap... |
2474 |
plist_for_each_entry(p, &swap_active_head, list) { |
22c6f8fdb swapfile: remove ... |
2475 |
if (p->flags & SWP_WRITEOK) { |
adfab836f swap: change swap... |
2476 2477 |
if (p->swap_file->f_mapping == mapping) { found = 1; |
1da177e4c Linux-2.6.12-rc2 |
2478 |
break; |
adfab836f swap: change swap... |
2479 |
} |
1da177e4c Linux-2.6.12-rc2 |
2480 |
} |
1da177e4c Linux-2.6.12-rc2 |
2481 |
} |
adfab836f swap: change swap... |
2482 |
if (!found) { |
1da177e4c Linux-2.6.12-rc2 |
2483 |
err = -EINVAL; |
5d337b919 [PATCH] swap: swa... |
2484 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2485 2486 |
goto out_dput; } |
191c54244 mm: collapse secu... |
2487 |
if (!security_vm_enough_memory_mm(current->mm, p->pages)) |
1da177e4c Linux-2.6.12-rc2 |
2488 2489 2490 |
vm_unacct_memory(p->pages); else { err = -ENOMEM; |
5d337b919 [PATCH] swap: swa... |
2491 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2492 2493 |
goto out_dput; } |
a2468cc9b swap: choose swap... |
2494 |
del_from_avail_list(p); |
ec8acf20a swap: add per-par... |
2495 |
spin_lock(&p->lock); |
78ecba081 mm: fix ever-decr... |
2496 |
if (p->prio < 0) { |
adfab836f swap: change swap... |
2497 |
struct swap_info_struct *si = p; |
a2468cc9b swap: choose swap... |
2498 |
int nid; |
adfab836f swap: change swap... |
2499 |
|
18ab4d4ce swap: change swap... |
2500 |
plist_for_each_entry_continue(si, &swap_active_head, list) { |
adfab836f swap: change swap... |
2501 |
si->prio++; |
18ab4d4ce swap: change swap... |
2502 |
si->list.prio--; |
a2468cc9b swap: choose swap... |
2503 2504 2505 2506 |
for_each_node(nid) { if (si->avail_lists[nid].prio != 1) si->avail_lists[nid].prio--; } |
adfab836f swap: change swap... |
2507 |
} |
78ecba081 mm: fix ever-decr... |
2508 2509 |
least_priority++; } |
18ab4d4ce swap: change swap... |
2510 |
plist_del(&p->list, &swap_active_head); |
ec8acf20a swap: add per-par... |
2511 |
atomic_long_sub(p->pages, &nr_swap_pages); |
1da177e4c Linux-2.6.12-rc2 |
2512 2513 |
total_swap_pages -= p->pages; p->flags &= ~SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
2514 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
2515 |
spin_unlock(&swap_lock); |
fb4f88dca [PATCH] swap: get... |
2516 |
|
039939a65 mm/swap: enable s... |
2517 |
disable_swap_slots_cache_lock(); |
e1e12d2f3 mm, oom: fix race... |
2518 |
set_current_oom_origin(); |
adfab836f swap: change swap... |
2519 |
err = try_to_unuse(p->type, false, 0); /* force unuse all pages */ |
e1e12d2f3 mm, oom: fix race... |
2520 |
clear_current_oom_origin(); |
1da177e4c Linux-2.6.12-rc2 |
2521 |
|
1da177e4c Linux-2.6.12-rc2 |
2522 2523 |
if (err) { /* re-insert swap space back into swap_list */ |
cf0cac0a0 mm: refactor rein... |
2524 |
reinsert_swap_info(p); |
039939a65 mm/swap: enable s... |
2525 |
reenable_swap_slots_cache_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
2526 2527 |
goto out_dput; } |
52b7efdbe [PATCH] swap: sca... |
2528 |
|
039939a65 mm/swap: enable s... |
2529 |
reenable_swap_slots_cache_unlock(); |
815c2c543 swap: make swap d... |
2530 |
flush_work(&p->discard_work); |
5d337b919 [PATCH] swap: swa... |
2531 |
destroy_swap_extents(p); |
570a335b8 swap_info: swap c... |
2532 2533 |
if (p->flags & SWP_CONTINUED) free_swap_count_continuations(p); |
81a0298bd mm, swap: don't u... |
2534 2535 |
if (!p->bdev || !blk_queue_nonrot(bdev_get_queue(p->bdev))) atomic_dec(&nr_rotate_swap); |
fc0abb145 [PATCH] sem2mutex... |
2536 |
mutex_lock(&swapon_mutex); |
5d337b919 [PATCH] swap: swa... |
2537 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
2538 |
spin_lock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
2539 |
drain_mmlist(); |
52b7efdbe [PATCH] swap: sca... |
2540 |
/* wait for anyone still in scan_swap_map */ |
52b7efdbe [PATCH] swap: sca... |
2541 2542 |
p->highest_bit = 0; /* cuts scans short */ while (p->flags >= SWP_SCANNING) { |
ec8acf20a swap: add per-par... |
2543 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
2544 |
spin_unlock(&swap_lock); |
13e4b57f6 [PATCH] mm: fix-u... |
2545 |
schedule_timeout_uninterruptible(1); |
5d337b919 [PATCH] swap: swa... |
2546 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
2547 |
spin_lock(&p->lock); |
52b7efdbe [PATCH] swap: sca... |
2548 |
} |
52b7efdbe [PATCH] swap: sca... |
2549 |
|
1da177e4c Linux-2.6.12-rc2 |
2550 |
swap_file = p->swap_file; |
5b808a230 swap: fix set_blo... |
2551 |
old_block_size = p->old_block_size; |
1da177e4c Linux-2.6.12-rc2 |
2552 2553 2554 2555 |
p->swap_file = NULL; p->max = 0; swap_map = p->swap_map; p->swap_map = NULL; |
2a8f94493 swap: change bloc... |
2556 2557 |
cluster_info = p->cluster_info; p->cluster_info = NULL; |
4f89849da frontswap: get ri... |
2558 |
frontswap_map = frontswap_map_get(p); |
ec8acf20a swap: add per-par... |
2559 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
2560 |
spin_unlock(&swap_lock); |
adfab836f swap: change swap... |
2561 |
frontswap_invalidate_area(p->type); |
58e97ba6b frontswap: enable... |
2562 |
frontswap_map_set(p, NULL); |
fc0abb145 [PATCH] sem2mutex... |
2563 |
mutex_unlock(&swapon_mutex); |
ebc2a1a69 swap: make cluste... |
2564 2565 |
free_percpu(p->percpu_cluster); p->percpu_cluster = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2566 |
vfree(swap_map); |
54f180d3c mm, swap: use kvz... |
2567 2568 |
kvfree(cluster_info); kvfree(frontswap_map); |
2de1a7e40 mm/swapfile.c: fi... |
2569 |
/* Destroy swap account information */ |
adfab836f swap: change swap... |
2570 |
swap_cgroup_swapoff(p->type); |
4b3ef9daa mm/swap: split sw... |
2571 |
exit_swap_address_space(p->type); |
27a7faa07 memcg: swap cgrou... |
2572 |
|
1da177e4c Linux-2.6.12-rc2 |
2573 2574 2575 |
inode = mapping->host; if (S_ISBLK(inode->i_mode)) { struct block_device *bdev = I_BDEV(inode); |
5b808a230 swap: fix set_blo... |
2576 |
set_blocksize(bdev, old_block_size); |
e525fd89d block: make blkde... |
2577 |
blkdev_put(bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
2578 |
} else { |
5955102c9 wrappers for ->i_... |
2579 |
inode_lock(inode); |
1da177e4c Linux-2.6.12-rc2 |
2580 |
inode->i_flags &= ~S_SWAPFILE; |
5955102c9 wrappers for ->i_... |
2581 |
inode_unlock(inode); |
1da177e4c Linux-2.6.12-rc2 |
2582 2583 |
} filp_close(swap_file, NULL); |
f893ab41e mm/swap: fix race... |
2584 2585 2586 2587 2588 2589 2590 2591 2592 |
/* * 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 |
2593 |
err = 0; |
66d7dd518 /proc/swaps: supp... |
2594 2595 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
1da177e4c Linux-2.6.12-rc2 |
2596 2597 2598 2599 |
out_dput: filp_close(victim, NULL); out: |
f58b59c1d swapfile: fix nam... |
2600 |
putname(pathname); |
1da177e4c Linux-2.6.12-rc2 |
2601 2602 2603 2604 |
return err; } #ifdef CONFIG_PROC_FS |
66d7dd518 /proc/swaps: supp... |
2605 2606 |
static unsigned swaps_poll(struct file *file, poll_table *wait) { |
f15146380 fs: seq_file - ad... |
2607 |
struct seq_file *seq = file->private_data; |
66d7dd518 /proc/swaps: supp... |
2608 2609 |
poll_wait(file, &proc_poll_wait, wait); |
f15146380 fs: seq_file - ad... |
2610 2611 |
if (seq->poll_event != atomic_read(&proc_poll_event)) { seq->poll_event = atomic_read(&proc_poll_event); |
66d7dd518 /proc/swaps: supp... |
2612 2613 2614 2615 2616 |
return POLLIN | POLLRDNORM | POLLERR | POLLPRI; } return POLLIN | POLLRDNORM; } |
1da177e4c Linux-2.6.12-rc2 |
2617 2618 2619 |
/* iterator */ static void *swap_start(struct seq_file *swap, loff_t *pos) { |
efa90a981 swap_info: change... |
2620 2621 |
struct swap_info_struct *si; int type; |
1da177e4c Linux-2.6.12-rc2 |
2622 |
loff_t l = *pos; |
fc0abb145 [PATCH] sem2mutex... |
2623 |
mutex_lock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2624 |
|
881e4aabe [PATCH] Always pr... |
2625 2626 |
if (!l) return SEQ_START_TOKEN; |
efa90a981 swap_info: change... |
2627 2628 2629 2630 |
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 |
2631 |
continue; |
881e4aabe [PATCH] Always pr... |
2632 |
if (!--l) |
efa90a981 swap_info: change... |
2633 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
2634 2635 2636 2637 2638 2639 2640 |
} return NULL; } static void *swap_next(struct seq_file *swap, void *v, loff_t *pos) { |
efa90a981 swap_info: change... |
2641 2642 |
struct swap_info_struct *si = v; int type; |
1da177e4c Linux-2.6.12-rc2 |
2643 |
|
881e4aabe [PATCH] Always pr... |
2644 |
if (v == SEQ_START_TOKEN) |
efa90a981 swap_info: change... |
2645 2646 2647 |
type = 0; else type = si->type + 1; |
881e4aabe [PATCH] Always pr... |
2648 |
|
efa90a981 swap_info: change... |
2649 2650 2651 2652 |
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 |
2653 2654 |
continue; ++*pos; |
efa90a981 swap_info: change... |
2655 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
2656 2657 2658 2659 2660 2661 2662 |
} return NULL; } static void swap_stop(struct seq_file *swap, void *v) { |
fc0abb145 [PATCH] sem2mutex... |
2663 |
mutex_unlock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2664 2665 2666 2667 |
} static int swap_show(struct seq_file *swap, void *v) { |
efa90a981 swap_info: change... |
2668 |
struct swap_info_struct *si = v; |
1da177e4c Linux-2.6.12-rc2 |
2669 2670 |
struct file *file; int len; |
efa90a981 swap_info: change... |
2671 |
if (si == SEQ_START_TOKEN) { |
881e4aabe [PATCH] Always pr... |
2672 2673 2674 2675 |
seq_puts(swap,"Filename\t\t\t\tType\t\tSize\tUsed\tPriority "); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
2676 |
|
efa90a981 swap_info: change... |
2677 |
file = si->swap_file; |
2726d5662 vfs: add seq_file... |
2678 2679 |
len = seq_file_path(swap, file, " \t \\"); |
6eb396dc4 [PATCH] swap: swa... |
2680 2681 |
seq_printf(swap, "%*s%s\t%u\t%u\t%d ", |
886bb7e9c swapfile: remove ... |
2682 |
len < 40 ? 40 - len : 1, " ", |
496ad9aa8 new helper: file_... |
2683 |
S_ISBLK(file_inode(file)->i_mode) ? |
1da177e4c Linux-2.6.12-rc2 |
2684 |
"partition" : "file\t", |
efa90a981 swap_info: change... |
2685 2686 2687 |
si->pages << (PAGE_SHIFT - 10), si->inuse_pages << (PAGE_SHIFT - 10), si->prio); |
1da177e4c Linux-2.6.12-rc2 |
2688 2689 |
return 0; } |
15ad7cdcf [PATCH] struct se... |
2690 |
static const struct seq_operations swaps_op = { |
1da177e4c Linux-2.6.12-rc2 |
2691 2692 2693 2694 2695 2696 2697 2698 |
.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... |
2699 |
struct seq_file *seq; |
66d7dd518 /proc/swaps: supp... |
2700 |
int ret; |
66d7dd518 /proc/swaps: supp... |
2701 |
ret = seq_open(file, &swaps_op); |
f15146380 fs: seq_file - ad... |
2702 |
if (ret) |
66d7dd518 /proc/swaps: supp... |
2703 |
return ret; |
66d7dd518 /proc/swaps: supp... |
2704 |
|
f15146380 fs: seq_file - ad... |
2705 2706 2707 |
seq = file->private_data; seq->poll_event = atomic_read(&proc_poll_event); return 0; |
1da177e4c Linux-2.6.12-rc2 |
2708 |
} |
15ad7cdcf [PATCH] struct se... |
2709 |
static const struct file_operations proc_swaps_operations = { |
1da177e4c Linux-2.6.12-rc2 |
2710 2711 2712 2713 |
.open = swaps_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, |
66d7dd518 /proc/swaps: supp... |
2714 |
.poll = swaps_poll, |
1da177e4c Linux-2.6.12-rc2 |
2715 2716 2717 2718 |
}; static int __init procswaps_init(void) { |
3d71f86f4 mm: use non-racy ... |
2719 |
proc_create("swaps", 0, NULL, &proc_swaps_operations); |
1da177e4c Linux-2.6.12-rc2 |
2720 2721 2722 2723 |
return 0; } __initcall(procswaps_init); #endif /* CONFIG_PROC_FS */ |
1796316a8 x86: consolidate ... |
2724 2725 2726 2727 2728 2729 2730 2731 |
#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... |
2732 |
static struct swap_info_struct *alloc_swap_info(void) |
1da177e4c Linux-2.6.12-rc2 |
2733 |
{ |
73c34b6ac swap_info: miscel... |
2734 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
2735 |
unsigned int type; |
a2468cc9b swap: choose swap... |
2736 |
int i; |
4fb12a087 mm/swap: use nr_n... |
2737 |
int size = sizeof(*p) + nr_node_ids * sizeof(struct plist_node); |
efa90a981 swap_info: change... |
2738 |
|
4fb12a087 mm/swap: use nr_n... |
2739 |
p = kvzalloc(size, GFP_KERNEL); |
efa90a981 swap_info: change... |
2740 |
if (!p) |
53cbb2435 sys_swapon: separ... |
2741 |
return ERR_PTR(-ENOMEM); |
efa90a981 swap_info: change... |
2742 |
|
5d337b919 [PATCH] swap: swa... |
2743 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
2744 2745 |
for (type = 0; type < nr_swapfiles; type++) { if (!(swap_info[type]->flags & SWP_USED)) |
1da177e4c Linux-2.6.12-rc2 |
2746 |
break; |
efa90a981 swap_info: change... |
2747 |
} |
0697212a4 [PATCH] Swapless ... |
2748 |
if (type >= MAX_SWAPFILES) { |
5d337b919 [PATCH] swap: swa... |
2749 |
spin_unlock(&swap_lock); |
d85699259 mm/swapfile.c: us... |
2750 |
kvfree(p); |
730c0581c sys_swapon: simpl... |
2751 |
return ERR_PTR(-EPERM); |
1da177e4c Linux-2.6.12-rc2 |
2752 |
} |
efa90a981 swap_info: change... |
2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 |
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 { |
d85699259 mm/swapfile.c: us... |
2764 |
kvfree(p); |
efa90a981 swap_info: change... |
2765 2766 2767 2768 2769 2770 |
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... |
2771 |
INIT_LIST_HEAD(&p->first_swap_extent.list); |
18ab4d4ce swap: change swap... |
2772 |
plist_node_init(&p->list, 0); |
a2468cc9b swap: choose swap... |
2773 2774 |
for_each_node(i) plist_node_init(&p->avail_lists[i], 0); |
1da177e4c Linux-2.6.12-rc2 |
2775 |
p->flags = SWP_USED; |
5d337b919 [PATCH] swap: swa... |
2776 |
spin_unlock(&swap_lock); |
ec8acf20a swap: add per-par... |
2777 |
spin_lock_init(&p->lock); |
2628bd6fc mm, swap: fix rac... |
2778 |
spin_lock_init(&p->cont_lock); |
efa90a981 swap_info: change... |
2779 |
|
53cbb2435 sys_swapon: separ... |
2780 |
return p; |
53cbb2435 sys_swapon: separ... |
2781 |
} |
4d0e1e107 sys_swapon: separ... |
2782 2783 2784 2785 2786 2787 2788 |
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... |
2789 |
FMODE_READ | FMODE_WRITE | FMODE_EXCL, p); |
4d0e1e107 sys_swapon: separ... |
2790 2791 |
if (error < 0) { p->bdev = NULL; |
6f179af88 mm: fix potential... |
2792 |
return error; |
4d0e1e107 sys_swapon: separ... |
2793 2794 2795 2796 |
} p->old_block_size = block_size(p->bdev); error = set_blocksize(p->bdev, PAGE_SIZE); if (error < 0) |
87ade72a7 sys_swapon: simpl... |
2797 |
return error; |
4d0e1e107 sys_swapon: separ... |
2798 2799 2800 |
p->flags |= SWP_BLKDEV; } else if (S_ISREG(inode->i_mode)) { p->bdev = inode->i_sb->s_bdev; |
5955102c9 wrappers for ->i_... |
2801 |
inode_lock(inode); |
87ade72a7 sys_swapon: simpl... |
2802 2803 2804 2805 |
if (IS_SWAPFILE(inode)) return -EBUSY; } else return -EINVAL; |
4d0e1e107 sys_swapon: separ... |
2806 2807 |
return 0; |
4d0e1e107 sys_swapon: separ... |
2808 |
} |
4bb1a8d8f x86/speculation/l... |
2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 |
/* * Find out how many pages are allowed for a single swap device. There * are two limiting factors: * 1) the number 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 different * architectures. * * In order to find the largest possible bit mask, a swap entry with * swap type 0 and swap offset ~0UL is created, encoded to a swap pte, * decoded to a swp_entry_t again, and finally the swap 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 swap pte. */ unsigned long generic_max_swapfile_size(void) { return swp_offset(pte_to_swp_entry( swp_entry_to_pte(swp_entry(0, ~0UL)))) + 1; } /* Can be overridden by an architecture for additional checks. */ __weak unsigned long max_swapfile_size(void) { return generic_max_swapfile_size(); } |
ca8bd38bf sys_swapon: separ... |
2837 2838 2839 2840 2841 2842 2843 |
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... |
2844 |
unsigned long last_page; |
ca8bd38bf sys_swapon: separ... |
2845 2846 |
if (memcmp("SWAPSPACE2", swap_header->magic.magic, 10)) { |
465c47fd8 mm/swapfile.c: co... |
2847 2848 |
pr_err("Unable to find swap-space signature "); |
387190253 sys_swapon: simpl... |
2849 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2850 2851 2852 2853 2854 2855 2856 |
} /* 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); |
dd111be69 swapfile: fix mem... |
2857 2858 |
if (swap_header->info.nr_badpages > MAX_SWAP_BADPAGES) return 0; |
ca8bd38bf sys_swapon: separ... |
2859 2860 2861 2862 2863 |
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... |
2864 2865 2866 |
pr_warn("Unable to handle swap header version %d ", swap_header->info.version); |
387190253 sys_swapon: simpl... |
2867 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2868 2869 2870 2871 2872 |
} p->lowest_bit = 1; p->cluster_next = 1; p->cluster_nr = 0; |
4bb1a8d8f x86/speculation/l... |
2873 |
maxpages = max_swapfile_size(); |
d6bbbd29b swap: warn when a... |
2874 |
last_page = swap_header->info.last_page; |
2ab773810 swap: divide-by-z... |
2875 2876 2877 2878 2879 |
if (!last_page) { pr_warn("Empty swap-file "); return 0; } |
d6bbbd29b swap: warn when a... |
2880 |
if (last_page > maxpages) { |
465c47fd8 mm/swapfile.c: co... |
2881 2882 |
pr_warn("Truncating oversized swap area, only using %luk out of %luk ", |
d6bbbd29b swap: warn when a... |
2883 2884 2885 2886 2887 |
maxpages << (PAGE_SHIFT - 10), last_page << (PAGE_SHIFT - 10)); } if (maxpages > last_page) { maxpages = last_page + 1; |
ca8bd38bf sys_swapon: separ... |
2888 2889 2890 2891 2892 2893 2894 |
/* 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... |
2895 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2896 2897 |
swapfilepages = i_size_read(inode) >> PAGE_SHIFT; if (swapfilepages && maxpages > swapfilepages) { |
465c47fd8 mm/swapfile.c: co... |
2898 2899 |
pr_warn("Swap area shorter than signature indicates "); |
387190253 sys_swapon: simpl... |
2900 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2901 2902 |
} if (swap_header->info.nr_badpages && S_ISREG(inode->i_mode)) |
387190253 sys_swapon: simpl... |
2903 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2904 |
if (swap_header->info.nr_badpages > MAX_SWAP_BADPAGES) |
387190253 sys_swapon: simpl... |
2905 |
return 0; |
ca8bd38bf sys_swapon: separ... |
2906 2907 |
return maxpages; |
ca8bd38bf sys_swapon: separ... |
2908 |
} |
4b3ef9daa mm/swap: split sw... |
2909 |
#define SWAP_CLUSTER_INFO_COLS \ |
235b62176 mm/swap: add clus... |
2910 |
DIV_ROUND_UP(L1_CACHE_BYTES, sizeof(struct swap_cluster_info)) |
4b3ef9daa mm/swap: split sw... |
2911 2912 2913 2914 |
#define SWAP_CLUSTER_SPACE_COLS \ DIV_ROUND_UP(SWAP_ADDRESS_SPACE_PAGES, SWAPFILE_CLUSTER) #define SWAP_CLUSTER_COLS \ max_t(unsigned int, SWAP_CLUSTER_INFO_COLS, SWAP_CLUSTER_SPACE_COLS) |
235b62176 mm/swap: add clus... |
2915 |
|
915d4d7bc sys_swapon: separ... |
2916 2917 2918 |
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... |
2919 |
struct swap_cluster_info *cluster_info, |
915d4d7bc sys_swapon: separ... |
2920 2921 2922 |
unsigned long maxpages, sector_t *span) { |
235b62176 mm/swap: add clus... |
2923 |
unsigned int j, k; |
915d4d7bc sys_swapon: separ... |
2924 2925 |
unsigned int nr_good_pages; int nr_extents; |
2a8f94493 swap: change bloc... |
2926 |
unsigned long nr_clusters = DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); |
235b62176 mm/swap: add clus... |
2927 2928 |
unsigned long col = p->cluster_next / SWAPFILE_CLUSTER % SWAP_CLUSTER_COLS; unsigned long i, idx; |
915d4d7bc sys_swapon: separ... |
2929 2930 |
nr_good_pages = maxpages - 1; /* omit header page */ |
6b5349159 mm, swap: add swa... |
2931 2932 |
cluster_list_init(&p->free_clusters); cluster_list_init(&p->discard_clusters); |
2a8f94493 swap: change bloc... |
2933 |
|
915d4d7bc sys_swapon: separ... |
2934 2935 |
for (i = 0; i < swap_header->info.nr_badpages; i++) { unsigned int page_nr = swap_header->info.badpages[i]; |
bdb8e3f68 sys_swapon: simpl... |
2936 2937 |
if (page_nr == 0 || page_nr > swap_header->info.last_page) return -EINVAL; |
915d4d7bc sys_swapon: separ... |
2938 2939 2940 |
if (page_nr < maxpages) { swap_map[page_nr] = SWAP_MAP_BAD; nr_good_pages--; |
2a8f94493 swap: change bloc... |
2941 2942 2943 2944 2945 |
/* * Haven't marked the cluster free yet, no list * operation involved */ inc_cluster_info_page(p, cluster_info, page_nr); |
915d4d7bc sys_swapon: separ... |
2946 2947 |
} } |
2a8f94493 swap: change bloc... |
2948 2949 2950 |
/* 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... |
2951 2952 |
if (nr_good_pages) { swap_map[0] = SWAP_MAP_BAD; |
2a8f94493 swap: change bloc... |
2953 2954 2955 2956 2957 |
/* * Not mark the cluster free yet, no list * operation involved */ inc_cluster_info_page(p, cluster_info, 0); |
915d4d7bc sys_swapon: separ... |
2958 2959 2960 |
p->max = maxpages; p->pages = nr_good_pages; nr_extents = setup_swap_extents(p, span); |
bdb8e3f68 sys_swapon: simpl... |
2961 2962 |
if (nr_extents < 0) return nr_extents; |
915d4d7bc sys_swapon: separ... |
2963 2964 2965 |
nr_good_pages = p->pages; } if (!nr_good_pages) { |
465c47fd8 mm/swapfile.c: co... |
2966 2967 |
pr_warn("Empty swap-file "); |
bdb8e3f68 sys_swapon: simpl... |
2968 |
return -EINVAL; |
915d4d7bc sys_swapon: separ... |
2969 |
} |
2a8f94493 swap: change bloc... |
2970 2971 |
if (!cluster_info) return nr_extents; |
235b62176 mm/swap: add clus... |
2972 |
|
4b3ef9daa mm/swap: split sw... |
2973 2974 2975 2976 |
/* * Reduce false cache line sharing between cluster_info and * sharing same address space. */ |
235b62176 mm/swap: add clus... |
2977 2978 2979 2980 2981 2982 2983 2984 |
for (k = 0; k < SWAP_CLUSTER_COLS; k++) { j = (k + col) % SWAP_CLUSTER_COLS; for (i = 0; i < DIV_ROUND_UP(nr_clusters, SWAP_CLUSTER_COLS); i++) { idx = i * SWAP_CLUSTER_COLS + j; if (idx >= nr_clusters) continue; if (cluster_count(&cluster_info[idx])) continue; |
2a8f94493 swap: change bloc... |
2985 |
cluster_set_flag(&cluster_info[idx], CLUSTER_FLAG_FREE); |
6b5349159 mm, swap: add swa... |
2986 2987 |
cluster_list_add_tail(&p->free_clusters, cluster_info, idx); |
2a8f94493 swap: change bloc... |
2988 |
} |
2a8f94493 swap: change bloc... |
2989 |
} |
915d4d7bc sys_swapon: separ... |
2990 |
return nr_extents; |
915d4d7bc sys_swapon: separ... |
2991 |
} |
dcf6b7ddd swap: discard whi... |
2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 |
/* * 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... |
3005 3006 3007 |
SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) { struct swap_info_struct *p; |
91a27b2a7 vfs: define struc... |
3008 |
struct filename *name; |
53cbb2435 sys_swapon: separ... |
3009 3010 |
struct file *swap_file = NULL; struct address_space *mapping; |
40531542e sys_swapon: separ... |
3011 |
int prio; |
53cbb2435 sys_swapon: separ... |
3012 3013 |
int error; union swap_header *swap_header; |
915d4d7bc sys_swapon: separ... |
3014 |
int nr_extents; |
53cbb2435 sys_swapon: separ... |
3015 3016 |
sector_t span; unsigned long maxpages; |
53cbb2435 sys_swapon: separ... |
3017 |
unsigned char *swap_map = NULL; |
2a8f94493 swap: change bloc... |
3018 |
struct swap_cluster_info *cluster_info = NULL; |
38b5faf4b mm: frontswap: co... |
3019 |
unsigned long *frontswap_map = NULL; |
53cbb2435 sys_swapon: separ... |
3020 3021 |
struct page *page = NULL; struct inode *inode = NULL; |
53cbb2435 sys_swapon: separ... |
3022 |
|
d15cab975 swapon: check val... |
3023 3024 |
if (swap_flags & ~SWAP_FLAGS_VALID) return -EINVAL; |
53cbb2435 sys_swapon: separ... |
3025 3026 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
a2468cc9b swap: choose swap... |
3027 3028 |
if (!swap_avail_heads) return -ENOMEM; |
53cbb2435 sys_swapon: separ... |
3029 |
p = alloc_swap_info(); |
2542e5134 sys_swapon: simpl... |
3030 3031 |
if (IS_ERR(p)) return PTR_ERR(p); |
53cbb2435 sys_swapon: separ... |
3032 |
|
815c2c543 swap: make swap d... |
3033 |
INIT_WORK(&p->discard_work, swap_discard_work); |
1da177e4c Linux-2.6.12-rc2 |
3034 |
name = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
3035 |
if (IS_ERR(name)) { |
7de7fb6b3 sys_swapon: move ... |
3036 |
error = PTR_ERR(name); |
1da177e4c Linux-2.6.12-rc2 |
3037 |
name = NULL; |
bd69010b0 sys_swapon: use a... |
3038 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
3039 |
} |
669abf4e5 vfs: make path_op... |
3040 |
swap_file = file_open_name(name, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
3041 |
if (IS_ERR(swap_file)) { |
7de7fb6b3 sys_swapon: move ... |
3042 |
error = PTR_ERR(swap_file); |
1da177e4c Linux-2.6.12-rc2 |
3043 |
swap_file = NULL; |
bd69010b0 sys_swapon: use a... |
3044 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
3045 3046 3047 3048 |
} p->swap_file = swap_file; mapping = swap_file->f_mapping; |
2130781e2 sys_swapon: fix i... |
3049 |
inode = mapping->host; |
6f179af88 mm: fix potential... |
3050 |
|
5955102c9 wrappers for ->i_... |
3051 |
/* If S_ISREG(inode->i_mode) will do inode_lock(inode); */ |
4d0e1e107 sys_swapon: separ... |
3052 3053 |
error = claim_swapfile(p, inode); if (unlikely(error)) |
1da177e4c Linux-2.6.12-rc2 |
3054 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
3055 |
|
1da177e4c Linux-2.6.12-rc2 |
3056 3057 3058 3059 3060 3061 3062 |
/* * Read the swap header. */ if (!mapping->a_ops->readpage) { error = -EINVAL; goto bad_swap; } |
090d2b185 [PATCH] read_mapp... |
3063 |
page = read_mapping_page(mapping, 0, swap_file); |
1da177e4c Linux-2.6.12-rc2 |
3064 3065 3066 3067 |
if (IS_ERR(page)) { error = PTR_ERR(page); goto bad_swap; } |
81e339712 swapfile: remove ... |
3068 |
swap_header = kmap(page); |
1da177e4c Linux-2.6.12-rc2 |
3069 |
|
ca8bd38bf sys_swapon: separ... |
3070 3071 |
maxpages = read_swap_header(p, swap_header, inode); if (unlikely(!maxpages)) { |
1da177e4c Linux-2.6.12-rc2 |
3072 3073 3074 |
error = -EINVAL; goto bad_swap; } |
886bb7e9c swapfile: remove ... |
3075 |
|
81e339712 swapfile: remove ... |
3076 |
/* OK, set up the swap map and apply the bad block list */ |
803d0c835 sys_swapon: use v... |
3077 |
swap_map = vzalloc(maxpages); |
81e339712 swapfile: remove ... |
3078 3079 3080 3081 |
if (!swap_map) { error = -ENOMEM; goto bad_swap; } |
f05714293 mm: support anony... |
3082 3083 3084 |
if (bdi_cap_stable_pages_required(inode_to_bdi(inode))) p->flags |= SWP_STABLE_WRITES; |
2a8f94493 swap: change bloc... |
3085 |
if (p->bdev && blk_queue_nonrot(bdev_get_queue(p->bdev))) { |
6f179af88 mm: fix potential... |
3086 |
int cpu; |
235b62176 mm/swap: add clus... |
3087 |
unsigned long ci, nr_cluster; |
6f179af88 mm: fix potential... |
3088 |
|
2a8f94493 swap: change bloc... |
3089 3090 3091 3092 3093 3094 |
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); |
235b62176 mm/swap: add clus... |
3095 |
nr_cluster = DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); |
2a8f94493 swap: change bloc... |
3096 |
|
54f180d3c mm, swap: use kvz... |
3097 3098 |
cluster_info = kvzalloc(nr_cluster * sizeof(*cluster_info), GFP_KERNEL); |
2a8f94493 swap: change bloc... |
3099 3100 3101 3102 |
if (!cluster_info) { error = -ENOMEM; goto bad_swap; } |
235b62176 mm/swap: add clus... |
3103 3104 3105 |
for (ci = 0; ci < nr_cluster; ci++) spin_lock_init(&((cluster_info + ci)->lock)); |
ebc2a1a69 swap: make cluste... |
3106 3107 3108 3109 3110 |
p->percpu_cluster = alloc_percpu(struct percpu_cluster); if (!p->percpu_cluster) { error = -ENOMEM; goto bad_swap; } |
6f179af88 mm: fix potential... |
3111 |
for_each_possible_cpu(cpu) { |
ebc2a1a69 swap: make cluste... |
3112 |
struct percpu_cluster *cluster; |
6f179af88 mm: fix potential... |
3113 |
cluster = per_cpu_ptr(p->percpu_cluster, cpu); |
ebc2a1a69 swap: make cluste... |
3114 3115 |
cluster_set_null(&cluster->index); } |
81a0298bd mm, swap: don't u... |
3116 3117 |
} else atomic_inc(&nr_rotate_swap); |
1da177e4c Linux-2.6.12-rc2 |
3118 |
|
1421ef3cd sys_swapon: call ... |
3119 3120 3121 |
error = swap_cgroup_swapon(p->type, maxpages); if (error) goto bad_swap; |
915d4d7bc sys_swapon: separ... |
3122 |
nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map, |
2a8f94493 swap: change bloc... |
3123 |
cluster_info, maxpages, &span); |
915d4d7bc sys_swapon: separ... |
3124 3125 |
if (unlikely(nr_extents < 0)) { error = nr_extents; |
1da177e4c Linux-2.6.12-rc2 |
3126 3127 |
goto bad_swap; } |
38b5faf4b mm: frontswap: co... |
3128 |
/* frontswap enabled? set up bit-per-page map for frontswap */ |
8ea1d2a19 mm, frontswap: co... |
3129 |
if (IS_ENABLED(CONFIG_FRONTSWAP)) |
54f180d3c mm, swap: use kvz... |
3130 3131 |
frontswap_map = kvzalloc(BITS_TO_LONGS(maxpages) * sizeof(long), GFP_KERNEL); |
1da177e4c Linux-2.6.12-rc2 |
3132 |
|
2a8f94493 swap: change bloc... |
3133 3134 3135 3136 3137 3138 3139 3140 3141 |
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... |
3142 |
|
2a8f94493 swap: change bloc... |
3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 |
/* * 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... |
3161 |
} |
20137a490 swapfile: swapon ... |
3162 |
} |
6a6ba8317 swapfile: swapon ... |
3163 |
|
4b3ef9daa mm/swap: split sw... |
3164 3165 3166 |
error = init_swap_address_space(p->type, maxpages); if (error) goto bad_swap; |
fc0abb145 [PATCH] sem2mutex... |
3167 |
mutex_lock(&swapon_mutex); |
40531542e sys_swapon: separ... |
3168 |
prio = -1; |
78ecba081 mm: fix ever-decr... |
3169 |
if (swap_flags & SWAP_FLAG_PREFER) |
40531542e sys_swapon: separ... |
3170 |
prio = |
78ecba081 mm: fix ever-decr... |
3171 |
(swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT; |
2a8f94493 swap: change bloc... |
3172 |
enable_swap_info(p, prio, swap_map, cluster_info, frontswap_map); |
c69dbfb84 sys_swapon: move ... |
3173 |
|
756a025f0 mm: coalesce spli... |
3174 3175 |
pr_info("Adding %uk swap on %s. Priority:%d extents:%d across:%lluk %s%s%s%s%s ", |
91a27b2a7 vfs: define struc... |
3176 |
p->pages<<(PAGE_SHIFT-10), name->name, p->prio, |
c69dbfb84 sys_swapon: move ... |
3177 3178 |
nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10), (p->flags & SWP_SOLIDSTATE) ? "SS" : "", |
38b5faf4b mm: frontswap: co... |
3179 |
(p->flags & SWP_DISCARDABLE) ? "D" : "", |
dcf6b7ddd swap: discard whi... |
3180 3181 |
(p->flags & SWP_AREA_DISCARD) ? "s" : "", (p->flags & SWP_PAGE_DISCARD) ? "c" : "", |
38b5faf4b mm: frontswap: co... |
3182 |
(frontswap_map) ? "FS" : ""); |
c69dbfb84 sys_swapon: move ... |
3183 |
|
fc0abb145 [PATCH] sem2mutex... |
3184 |
mutex_unlock(&swapon_mutex); |
66d7dd518 /proc/swaps: supp... |
3185 3186 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
9b01c350a sys_swapon: do on... |
3187 3188 |
if (S_ISREG(inode->i_mode)) inode->i_flags |= S_SWAPFILE; |
1da177e4c Linux-2.6.12-rc2 |
3189 3190 3191 |
error = 0; goto out; bad_swap: |
ebc2a1a69 swap: make cluste... |
3192 3193 |
free_percpu(p->percpu_cluster); p->percpu_cluster = NULL; |
bd69010b0 sys_swapon: use a... |
3194 |
if (inode && S_ISBLK(inode->i_mode) && p->bdev) { |
f2090d2df sys_swapon: remov... |
3195 3196 |
set_blocksize(p->bdev, p->old_block_size); blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
3197 |
} |
4cd3bb10f [PATCH] swap: mov... |
3198 |
destroy_swap_extents(p); |
e8e6c2ec4 sys_swapon: do no... |
3199 |
swap_cgroup_swapoff(p->type); |
5d337b919 [PATCH] swap: swa... |
3200 |
spin_lock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
3201 |
p->swap_file = NULL; |
1da177e4c Linux-2.6.12-rc2 |
3202 |
p->flags = 0; |
5d337b919 [PATCH] swap: swa... |
3203 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
3204 |
vfree(swap_map); |
8606a1a94 mm: kvfree the sw... |
3205 |
kvfree(cluster_info); |
b6b1fd2a6 mm/swapfile.c: fi... |
3206 |
kvfree(frontswap_map); |
52c50567d mm: swap: unlock ... |
3207 |
if (swap_file) { |
2130781e2 sys_swapon: fix i... |
3208 |
if (inode && S_ISREG(inode->i_mode)) { |
5955102c9 wrappers for ->i_... |
3209 |
inode_unlock(inode); |
2130781e2 sys_swapon: fix i... |
3210 3211 |
inode = NULL; } |
1da177e4c Linux-2.6.12-rc2 |
3212 |
filp_close(swap_file, NULL); |
52c50567d mm: swap: unlock ... |
3213 |
} |
1da177e4c Linux-2.6.12-rc2 |
3214 3215 3216 |
out: if (page && !IS_ERR(page)) { kunmap(page); |
09cbfeaf1 mm, fs: get rid o... |
3217 |
put_page(page); |
1da177e4c Linux-2.6.12-rc2 |
3218 3219 3220 |
} if (name) putname(name); |
9b01c350a sys_swapon: do on... |
3221 |
if (inode && S_ISREG(inode->i_mode)) |
5955102c9 wrappers for ->i_... |
3222 |
inode_unlock(inode); |
039939a65 mm/swap: enable s... |
3223 3224 |
if (!error) enable_swap_slots_cache(); |
1da177e4c Linux-2.6.12-rc2 |
3225 3226 3227 3228 3229 |
return error; } void si_swapinfo(struct sysinfo *val) { |
efa90a981 swap_info: change... |
3230 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
3231 |
unsigned long nr_to_be_unused = 0; |
5d337b919 [PATCH] swap: swa... |
3232 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
3233 3234 3235 3236 3237 |
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 |
3238 |
} |
ec8acf20a swap: add per-par... |
3239 |
val->freeswap = atomic_long_read(&nr_swap_pages) + nr_to_be_unused; |
1da177e4c Linux-2.6.12-rc2 |
3240 |
val->totalswap = total_swap_pages + nr_to_be_unused; |
5d337b919 [PATCH] swap: swa... |
3241 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
3242 3243 3244 3245 3246 |
} /* * Verify that a swap entry is valid and increment its swap map count. * |
355cfa73d mm: modify swap_m... |
3247 3248 3249 3250 3251 3252 |
* 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... |
3253 |
* - swap-mapped reference requested but needs continued swap count. -> ENOMEM |
1da177e4c Linux-2.6.12-rc2 |
3254 |
*/ |
8d69aaee8 swap_info: swap_m... |
3255 |
static int __swap_duplicate(swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
3256 |
{ |
73c34b6ac swap_info: miscel... |
3257 |
struct swap_info_struct *p; |
235b62176 mm/swap: add clus... |
3258 |
struct swap_cluster_info *ci; |
1da177e4c Linux-2.6.12-rc2 |
3259 |
unsigned long offset, type; |
8d69aaee8 swap_info: swap_m... |
3260 3261 |
unsigned char count; unsigned char has_cache; |
253d553ba swap_info: SWAP_H... |
3262 |
int err = -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
3263 |
|
a7420aa54 HWPOISON: Add sup... |
3264 |
if (non_swap_entry(entry)) |
253d553ba swap_info: SWAP_H... |
3265 |
goto out; |
0697212a4 [PATCH] Swapless ... |
3266 |
|
1da177e4c Linux-2.6.12-rc2 |
3267 3268 3269 |
type = swp_type(entry); if (type >= nr_swapfiles) goto bad_file; |
efa90a981 swap_info: change... |
3270 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
3271 |
offset = swp_offset(entry); |
355cfa73d mm: modify swap_m... |
3272 |
if (unlikely(offset >= p->max)) |
235b62176 mm/swap: add clus... |
3273 3274 3275 |
goto out; ci = lock_cluster_or_swap_info(p, offset); |
355cfa73d mm: modify swap_m... |
3276 |
|
253d553ba swap_info: SWAP_H... |
3277 |
count = p->swap_map[offset]; |
edfe23dac swap: fix races e... |
3278 3279 3280 3281 3282 3283 3284 3285 3286 |
/* * 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... |
3287 3288 3289 |
has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; err = 0; |
355cfa73d mm: modify swap_m... |
3290 |
|
253d553ba swap_info: SWAP_H... |
3291 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
3292 3293 |
/* set SWAP_HAS_CACHE if there is no cache and entry is used */ |
253d553ba swap_info: SWAP_H... |
3294 3295 3296 3297 3298 3299 |
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... |
3300 3301 |
} else if (count || has_cache) { |
253d553ba swap_info: SWAP_H... |
3302 |
|
570a335b8 swap_info: swap c... |
3303 3304 3305 |
if ((count & ~COUNT_CONTINUED) < SWAP_MAP_MAX) count += usage; else if ((count & ~COUNT_CONTINUED) > SWAP_MAP_MAX) |
253d553ba swap_info: SWAP_H... |
3306 |
err = -EINVAL; |
570a335b8 swap_info: swap c... |
3307 3308 3309 3310 |
else if (swap_count_continued(p, offset, count)) count = COUNT_CONTINUED; else err = -ENOMEM; |
355cfa73d mm: modify swap_m... |
3311 |
} else |
253d553ba swap_info: SWAP_H... |
3312 3313 3314 |
err = -ENOENT; /* unused swap entry */ p->swap_map[offset] = count | has_cache; |
355cfa73d mm: modify swap_m... |
3315 |
unlock_out: |
235b62176 mm/swap: add clus... |
3316 |
unlock_cluster_or_swap_info(p, ci); |
1da177e4c Linux-2.6.12-rc2 |
3317 |
out: |
253d553ba swap_info: SWAP_H... |
3318 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
3319 3320 |
bad_file: |
465c47fd8 mm/swapfile.c: co... |
3321 3322 |
pr_err("swap_dup: %s%08lx ", Bad_file, entry.val); |
1da177e4c Linux-2.6.12-rc2 |
3323 3324 |
goto out; } |
253d553ba swap_info: SWAP_H... |
3325 |
|
355cfa73d mm: modify swap_m... |
3326 |
/* |
aaa468653 swap_info: note S... |
3327 3328 3329 3330 3331 3332 3333 3334 3335 |
* 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... |
3336 3337 3338 3339 3340 |
* 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... |
3341 |
*/ |
570a335b8 swap_info: swap c... |
3342 |
int swap_duplicate(swp_entry_t entry) |
355cfa73d mm: modify swap_m... |
3343 |
{ |
570a335b8 swap_info: swap c... |
3344 3345 3346 3347 3348 |
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... |
3349 |
} |
1da177e4c Linux-2.6.12-rc2 |
3350 |
|
cb4b86ba4 mm: add swap cach... |
3351 |
/* |
355cfa73d mm: modify swap_m... |
3352 3353 |
* @entry: swap entry for which we allocate swap cache. * |
73c34b6ac swap_info: miscel... |
3354 |
* Called when allocating swap cache for existing swap entry, |
355cfa73d mm: modify swap_m... |
3355 3356 3357 |
* 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... |
3358 3359 3360 |
*/ int swapcache_prepare(swp_entry_t entry) { |
253d553ba swap_info: SWAP_H... |
3361 |
return __swap_duplicate(entry, SWAP_HAS_CACHE); |
cb4b86ba4 mm: add swap cach... |
3362 |
} |
f981c5950 mm: methods for t... |
3363 3364 3365 |
struct swap_info_struct *page_swap_info(struct page *page) { swp_entry_t swap = { .val = page_private(page) }; |
f981c5950 mm: methods for t... |
3366 3367 3368 3369 3370 3371 3372 3373 |
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... |
3374 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
f981c5950 mm: methods for t... |
3375 3376 3377 3378 3379 3380 3381 |
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... |
3382 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
f981c5950 mm: methods for t... |
3383 3384 3385 |
return swp_offset(swap); } EXPORT_SYMBOL_GPL(__page_file_index); |
1da177e4c Linux-2.6.12-rc2 |
3386 |
/* |
570a335b8 swap_info: swap c... |
3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 |
* 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; |
235b62176 mm/swap: add clus... |
3404 |
struct swap_cluster_info *ci; |
570a335b8 swap_info: swap c... |
3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 |
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); |
235b62176 mm/swap: add clus... |
3428 3429 |
ci = lock_cluster(si, offset); |
570a335b8 swap_info: swap c... |
3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 |
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) { |
235b62176 mm/swap: add clus... |
3442 |
unlock_cluster(ci); |
ec8acf20a swap: add per-par... |
3443 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
3444 3445 3446 3447 3448 |
return -ENOMEM; } /* * We are fortunate that although vmalloc_to_page uses pte_offset_map, |
2de1a7e40 mm/swapfile.c: fi... |
3449 3450 |
* 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... |
3451 3452 3453 |
*/ head = vmalloc_to_page(si->swap_map + offset); offset &= ~PAGE_MASK; |
2628bd6fc mm, swap: fix rac... |
3454 |
spin_lock(&si->cont_lock); |
570a335b8 swap_info: swap c... |
3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 |
/* * 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)) |
2628bd6fc mm, swap: fix rac... |
3474 |
goto out_unlock_cont; |
570a335b8 swap_info: swap c... |
3475 |
|
9b04c5fec mm: remove the se... |
3476 |
map = kmap_atomic(list_page) + offset; |
570a335b8 swap_info: swap c... |
3477 |
count = *map; |
9b04c5fec mm: remove the se... |
3478 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3479 3480 3481 3482 3483 3484 |
/* * If this continuation count now has some space in it, * free our allocation and use this one. */ if ((count & ~COUNT_CONTINUED) != SWAP_CONT_MAX) |
2628bd6fc mm, swap: fix rac... |
3485 |
goto out_unlock_cont; |
570a335b8 swap_info: swap c... |
3486 3487 3488 3489 |
} list_add_tail(&page->lru, &head->lru); page = NULL; /* now it's attached, don't free it */ |
2628bd6fc mm, swap: fix rac... |
3490 3491 |
out_unlock_cont: spin_unlock(&si->cont_lock); |
570a335b8 swap_info: swap c... |
3492 |
out: |
235b62176 mm/swap: add clus... |
3493 |
unlock_cluster(ci); |
ec8acf20a swap: add per-par... |
3494 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 |
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. |
235b62176 mm/swap: add clus... |
3507 3508 |
* Called while __swap_duplicate() or swap_entry_free() holds swap or cluster * lock. |
570a335b8 swap_info: swap c... |
3509 3510 3511 3512 3513 3514 3515 |
*/ 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; |
2628bd6fc mm, swap: fix rac... |
3516 |
bool ret; |
570a335b8 swap_info: swap c... |
3517 3518 3519 3520 3521 3522 |
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 */ } |
2628bd6fc mm, swap: fix rac... |
3523 |
spin_lock(&si->cont_lock); |
570a335b8 swap_info: swap c... |
3524 3525 |
offset &= ~PAGE_MASK; page = list_entry(head->lru.next, struct page, lru); |
9b04c5fec mm: remove the se... |
3526 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3527 3528 3529 3530 3531 3532 3533 3534 3535 |
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... |
3536 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3537 3538 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
3539 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3540 3541 |
} if (*map == SWAP_CONT_MAX) { |
9b04c5fec mm: remove the se... |
3542 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3543 |
page = list_entry(page->lru.next, struct page, lru); |
2628bd6fc mm, swap: fix rac... |
3544 3545 3546 3547 |
if (page == head) { ret = false; /* add count continuation */ goto out; } |
9b04c5fec mm: remove the se... |
3548 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3549 3550 3551 |
init_map: *map = 0; /* we didn't zero the page */ } *map += 1; |
9b04c5fec mm: remove the se... |
3552 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3553 3554 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
3555 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3556 |
*map = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
3557 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3558 3559 |
page = list_entry(page->lru.prev, struct page, lru); } |
2628bd6fc mm, swap: fix rac... |
3560 |
ret = true; /* incremented */ |
570a335b8 swap_info: swap c... |
3561 3562 3563 3564 3565 3566 3567 |
} else { /* decrementing */ /* * Think of how you subtract 1 from 1000 */ BUG_ON(count != COUNT_CONTINUED); while (*map == COUNT_CONTINUED) { |
9b04c5fec mm: remove the se... |
3568 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3569 3570 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
3571 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3572 3573 3574 3575 3576 |
} BUG_ON(*map == 0); *map -= 1; if (*map == 0) count = 0; |
9b04c5fec mm: remove the se... |
3577 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3578 3579 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
3580 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
3581 3582 |
*map = SWAP_CONT_MAX | count; count = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
3583 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
3584 3585 |
page = list_entry(page->lru.prev, struct page, lru); } |
2628bd6fc mm, swap: fix rac... |
3586 |
ret = count == COUNT_CONTINUED; |
570a335b8 swap_info: swap c... |
3587 |
} |
2628bd6fc mm, swap: fix rac... |
3588 3589 3590 |
out: spin_unlock(&si->cont_lock); return ret; |
570a335b8 swap_info: swap c... |
3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 |
} /* * 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)) { |
0d576d20c mm/swapfile.c: us... |
3605 3606 3607 3608 |
struct page *page, *next; list_for_each_entry_safe(page, next, &head->lru, lru) { list_del(&page->lru); |
570a335b8 swap_info: swap c... |
3609 3610 3611 3612 3613 |
__free_page(page); } } } } |
a2468cc9b swap: choose swap... |
3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 |
static int __init swapfile_init(void) { int nid; swap_avail_heads = kmalloc_array(nr_node_ids, sizeof(struct plist_head), GFP_KERNEL); if (!swap_avail_heads) { pr_emerg("Not enough memory for swap heads, swap is disabled "); return -ENOMEM; } for_each_node(nid) plist_head_init(&swap_avail_heads[nid]); return 0; } subsys_initcall(swapfile_init); |