Blame view
mm/swapfile.c
65.6 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 |
/* * linux/mm/swapfile.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds * Swap reorganised 29.12.95, Stephen Tweedie */ |
1da177e4c Linux-2.6.12-rc2 |
7 8 9 10 11 12 13 14 15 |
#include <linux/mm.h> #include <linux/hugetlb.h> #include <linux/mman.h> #include <linux/slab.h> #include <linux/kernel_stat.h> #include <linux/swap.h> #include <linux/vmalloc.h> #include <linux/pagemap.h> #include <linux/namei.h> |
072441e21 mm: move shmem pr... |
16 |
#include <linux/shmem_fs.h> |
1da177e4c Linux-2.6.12-rc2 |
17 |
#include <linux/blkdev.h> |
20137a490 swapfile: swapon ... |
18 |
#include <linux/random.h> |
1da177e4c Linux-2.6.12-rc2 |
19 20 21 22 |
#include <linux/writeback.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/init.h> |
5ad646880 ksm: let shared p... |
23 |
#include <linux/ksm.h> |
1da177e4c Linux-2.6.12-rc2 |
24 25 26 |
#include <linux/rmap.h> #include <linux/security.h> #include <linux/backing-dev.h> |
fc0abb145 [PATCH] sem2mutex... |
27 |
#include <linux/mutex.h> |
c59ede7b7 [PATCH] move capa... |
28 |
#include <linux/capability.h> |
1da177e4c Linux-2.6.12-rc2 |
29 |
#include <linux/syscalls.h> |
8a9f3ccd2 Memory controller... |
30 |
#include <linux/memcontrol.h> |
66d7dd518 /proc/swaps: supp... |
31 |
#include <linux/poll.h> |
72788c385 oom: replace PF_O... |
32 |
#include <linux/oom.h> |
38b5faf4b mm: frontswap: co... |
33 34 |
#include <linux/frontswap.h> #include <linux/swapfile.h> |
f981c5950 mm: methods for t... |
35 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
36 37 38 39 |
#include <asm/pgtable.h> #include <asm/tlbflush.h> #include <linux/swapops.h> |
27a7faa07 memcg: swap cgrou... |
40 |
#include <linux/page_cgroup.h> |
1da177e4c Linux-2.6.12-rc2 |
41 |
|
570a335b8 swap_info: swap c... |
42 43 44 |
static bool swap_count_continued(struct swap_info_struct *, pgoff_t, unsigned char); static void free_swap_count_continuations(struct swap_info_struct *); |
d4906e1aa swap: rework map_... |
45 |
static sector_t map_swap_entry(swp_entry_t, struct block_device**); |
570a335b8 swap_info: swap c... |
46 |
|
38b5faf4b mm: frontswap: co... |
47 |
DEFINE_SPINLOCK(swap_lock); |
7c363b8c6 mm/swapfile.c: ma... |
48 |
static unsigned int nr_swapfiles; |
ec8acf20a swap: add per-par... |
49 50 |
atomic_long_t nr_swap_pages; /* protected with swap_lock. reading in vm_swap_full() doesn't need lock */ |
1da177e4c Linux-2.6.12-rc2 |
51 |
long total_swap_pages; |
78ecba081 mm: fix ever-decr... |
52 |
static int least_priority; |
ec8acf20a swap: add per-par... |
53 |
static atomic_t highest_priority_index = ATOMIC_INIT(-1); |
1da177e4c Linux-2.6.12-rc2 |
54 |
|
1da177e4c Linux-2.6.12-rc2 |
55 56 57 58 |
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 "; |
38b5faf4b mm: frontswap: co... |
59 |
struct swap_list_t swap_list = {-1, -1}; |
1da177e4c Linux-2.6.12-rc2 |
60 |
|
38b5faf4b mm: frontswap: co... |
61 |
struct swap_info_struct *swap_info[MAX_SWAPFILES]; |
1da177e4c Linux-2.6.12-rc2 |
62 |
|
fc0abb145 [PATCH] sem2mutex... |
63 |
static DEFINE_MUTEX(swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
64 |
|
66d7dd518 /proc/swaps: supp... |
65 66 67 |
static DECLARE_WAIT_QUEUE_HEAD(proc_poll_wait); /* Activity counter to indicate that a swapon or swapoff has occurred */ static atomic_t proc_poll_event = ATOMIC_INIT(0); |
8d69aaee8 swap_info: swap_m... |
68 |
static inline unsigned char swap_count(unsigned char ent) |
355cfa73d mm: modify swap_m... |
69 |
{ |
570a335b8 swap_info: swap c... |
70 |
return ent & ~SWAP_HAS_CACHE; /* may include SWAP_HAS_CONT flag */ |
355cfa73d mm: modify swap_m... |
71 |
} |
efa90a981 swap_info: change... |
72 |
/* returns 1 if swap entry is freed */ |
c9e444103 mm: reuse unused ... |
73 74 75 |
static int __try_to_reclaim_swap(struct swap_info_struct *si, unsigned long offset) { |
efa90a981 swap_info: change... |
76 |
swp_entry_t entry = swp_entry(si->type, offset); |
c9e444103 mm: reuse unused ... |
77 78 |
struct page *page; int ret = 0; |
33806f06d swap: make each s... |
79 |
page = find_get_page(swap_address_space(entry), entry.val); |
c9e444103 mm: reuse unused ... |
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 |
if (!page) return 0; /* * This function is called from scan_swap_map() and it's called * by vmscan.c at reclaiming pages. So, we hold a lock on a page, here. * We have to use trylock for avoiding deadlock. This is a special * case and you should use try_to_free_swap() with explicit lock_page() * in usual operations. */ if (trylock_page(page)) { ret = try_to_free_swap(page); unlock_page(page); } page_cache_release(page); return ret; } |
355cfa73d mm: modify swap_m... |
96 |
|
1da177e4c Linux-2.6.12-rc2 |
97 |
/* |
6a6ba8317 swapfile: swapon ... |
98 99 100 101 102 103 |
* 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... |
104 105 |
sector_t start_block; sector_t nr_blocks; |
6a6ba8317 swapfile: swapon ... |
106 |
int err = 0; |
9625a5f28 swap_info: includ... |
107 108 109 110 111 112 |
/* 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... |
113 |
nr_blocks, GFP_KERNEL, 0); |
9625a5f28 swap_info: includ... |
114 115 116 117 |
if (err) return err; cond_resched(); } |
6a6ba8317 swapfile: swapon ... |
118 |
|
9625a5f28 swap_info: includ... |
119 120 121 |
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 ... |
122 123 |
err = blkdev_issue_discard(si->bdev, start_block, |
dd3932edd block: remove BLK... |
124 |
nr_blocks, GFP_KERNEL, 0); |
6a6ba8317 swapfile: swapon ... |
125 126 127 128 129 130 131 |
if (err) break; cond_resched(); } return err; /* That will often be -EOPNOTSUPP */ } |
7992fde72 swapfile: swap al... |
132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 |
/* * swap allocation tell device that a cluster of swap can now be discarded, * to allow the swap device to optimize its wear-levelling. */ static void discard_swap_cluster(struct swap_info_struct *si, pgoff_t start_page, pgoff_t nr_pages) { struct swap_extent *se = si->curr_swap_extent; int found_extent = 0; while (nr_pages) { struct list_head *lh; if (se->start_page <= start_page && start_page < se->start_page + se->nr_pages) { pgoff_t offset = start_page - se->start_page; sector_t start_block = se->start_block + offset; |
858a29900 swapfile: change ... |
149 |
sector_t nr_blocks = se->nr_pages - offset; |
7992fde72 swapfile: swap al... |
150 151 152 153 154 155 156 157 158 159 160 161 |
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... |
162 |
nr_blocks, GFP_NOIO, 0)) |
7992fde72 swapfile: swap al... |
163 164 165 166 |
break; } lh = se->list.next; |
7992fde72 swapfile: swap al... |
167 168 169 170 171 172 173 174 175 |
se = list_entry(lh, struct swap_extent, list); } } static int wait_for_discard(void *word) { schedule(); return 0; } |
048c27fd7 [PATCH] swap: sca... |
176 177 |
#define SWAPFILE_CLUSTER 256 #define LATENCY_LIMIT 256 |
24b8ff7c2 mm: remove inline... |
178 179 |
static unsigned long scan_swap_map(struct swap_info_struct *si, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
180 |
{ |
ebebbbe90 swapfile: rearran... |
181 |
unsigned long offset; |
c60aa176c swapfile: swap al... |
182 |
unsigned long scan_base; |
7992fde72 swapfile: swap al... |
183 |
unsigned long last_in_cluster = 0; |
048c27fd7 [PATCH] swap: sca... |
184 |
int latency_ration = LATENCY_LIMIT; |
7992fde72 swapfile: swap al... |
185 |
int found_free_cluster = 0; |
7dfad4183 [PATCH] swap: sca... |
186 |
|
886bb7e9c swapfile: remove ... |
187 |
/* |
7dfad4183 [PATCH] swap: sca... |
188 189 190 191 192 193 194 |
* 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... |
195 |
* And we let swap pages go all over an SSD partition. Hugh |
7dfad4183 [PATCH] swap: sca... |
196 |
*/ |
52b7efdbe [PATCH] swap: sca... |
197 |
si->flags += SWP_SCANNING; |
c60aa176c swapfile: swap al... |
198 |
scan_base = offset = si->cluster_next; |
ebebbbe90 swapfile: rearran... |
199 200 201 202 203 204 |
if (unlikely(!si->cluster_nr--)) { if (si->pages - si->inuse_pages < SWAPFILE_CLUSTER) { si->cluster_nr = SWAPFILE_CLUSTER - 1; goto checks; } |
7992fde72 swapfile: swap al... |
205 206 207 208 209 210 211 212 213 214 215 216 217 |
if (si->flags & SWP_DISCARDABLE) { /* * Start range check on racing allocations, in case * they overlap the cluster we eventually decide on * (we scan without swap_lock to allow preemption). * It's hardly conceivable that cluster_nr could be * wrapped during our scan, but don't depend on it. */ if (si->lowest_alloc) goto checks; si->lowest_alloc = si->max; si->highest_alloc = 0; } |
ec8acf20a swap: add per-par... |
218 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
219 |
|
c60aa176c swapfile: swap al... |
220 221 222 223 224 225 226 227 228 229 |
/* * If seek is expensive, start searching for new cluster from * start of partition, to minimize the span of allocated swap. * But if seek is cheap, search from our current position, so * that swap is allocated from all over the partition: if the * Flash Translation Layer only remaps within limited zones, * we don't want to wear out the first zone too quickly. */ if (!(si->flags & SWP_SOLIDSTATE)) scan_base = offset = si->lowest_bit; |
7dfad4183 [PATCH] swap: sca... |
230 231 232 233 |
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 |
234 |
if (si->swap_map[offset]) |
7dfad4183 [PATCH] swap: sca... |
235 236 |
last_in_cluster = offset + SWAPFILE_CLUSTER; else if (offset == last_in_cluster) { |
ec8acf20a swap: add per-par... |
237 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
238 239 240 |
offset -= SWAPFILE_CLUSTER - 1; si->cluster_next = offset; si->cluster_nr = SWAPFILE_CLUSTER - 1; |
7992fde72 swapfile: swap al... |
241 |
found_free_cluster = 1; |
ebebbbe90 swapfile: rearran... |
242 |
goto checks; |
1da177e4c Linux-2.6.12-rc2 |
243 |
} |
048c27fd7 [PATCH] swap: sca... |
244 245 246 247 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
7dfad4183 [PATCH] swap: sca... |
248 |
} |
ebebbbe90 swapfile: rearran... |
249 250 |
offset = si->lowest_bit; |
c60aa176c swapfile: swap al... |
251 252 253 254 255 256 257 |
last_in_cluster = offset + SWAPFILE_CLUSTER - 1; /* Locate the first empty (unaligned) cluster */ for (; last_in_cluster < scan_base; offset++) { if (si->swap_map[offset]) last_in_cluster = offset + SWAPFILE_CLUSTER; else if (offset == last_in_cluster) { |
ec8acf20a swap: add per-par... |
258 |
spin_lock(&si->lock); |
c60aa176c swapfile: swap al... |
259 260 261 262 263 264 265 266 267 268 269 270 271 |
offset -= SWAPFILE_CLUSTER - 1; si->cluster_next = offset; si->cluster_nr = SWAPFILE_CLUSTER - 1; found_free_cluster = 1; goto checks; } if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } } offset = scan_base; |
ec8acf20a swap: add per-par... |
272 |
spin_lock(&si->lock); |
ebebbbe90 swapfile: rearran... |
273 |
si->cluster_nr = SWAPFILE_CLUSTER - 1; |
7992fde72 swapfile: swap al... |
274 |
si->lowest_alloc = 0; |
1da177e4c Linux-2.6.12-rc2 |
275 |
} |
7dfad4183 [PATCH] swap: sca... |
276 |
|
ebebbbe90 swapfile: rearran... |
277 278 |
checks: if (!(si->flags & SWP_WRITEOK)) |
52b7efdbe [PATCH] swap: sca... |
279 |
goto no_page; |
7dfad4183 [PATCH] swap: sca... |
280 281 |
if (!si->highest_bit) goto no_page; |
ebebbbe90 swapfile: rearran... |
282 |
if (offset > si->highest_bit) |
c60aa176c swapfile: swap al... |
283 |
scan_base = offset = si->lowest_bit; |
c9e444103 mm: reuse unused ... |
284 |
|
b73d7fcec swap: prevent reu... |
285 286 |
/* 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 ... |
287 |
int swap_was_freed; |
ec8acf20a swap: add per-par... |
288 |
spin_unlock(&si->lock); |
c9e444103 mm: reuse unused ... |
289 |
swap_was_freed = __try_to_reclaim_swap(si, offset); |
ec8acf20a swap: add per-par... |
290 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
291 292 293 294 295 |
/* entry was freed successfully, try to use this again */ if (swap_was_freed) goto checks; goto scan; /* check next one */ } |
ebebbbe90 swapfile: rearran... |
296 297 298 299 300 301 302 303 304 305 306 |
if (si->swap_map[offset]) goto scan; if (offset == si->lowest_bit) si->lowest_bit++; if (offset == si->highest_bit) si->highest_bit--; si->inuse_pages++; if (si->inuse_pages == si->pages) { si->lowest_bit = si->max; si->highest_bit = 0; |
1da177e4c Linux-2.6.12-rc2 |
307 |
} |
253d553ba swap_info: SWAP_H... |
308 |
si->swap_map[offset] = usage; |
ebebbbe90 swapfile: rearran... |
309 310 |
si->cluster_next = offset + 1; si->flags -= SWP_SCANNING; |
7992fde72 swapfile: swap al... |
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 |
if (si->lowest_alloc) { /* * Only set when SWP_DISCARDABLE, and there's a scan * for a free cluster in progress or just completed. */ if (found_free_cluster) { /* * To optimize wear-levelling, discard the * old data of the cluster, taking care not to * discard any of its pages that have already * been allocated by racing tasks (offset has * already stepped over any at the beginning). */ if (offset < si->highest_alloc && si->lowest_alloc <= last_in_cluster) last_in_cluster = si->lowest_alloc - 1; si->flags |= SWP_DISCARDING; |
ec8acf20a swap: add per-par... |
329 |
spin_unlock(&si->lock); |
7992fde72 swapfile: swap al... |
330 331 332 333 |
if (offset < last_in_cluster) discard_swap_cluster(si, offset, last_in_cluster - offset + 1); |
ec8acf20a swap: add per-par... |
334 |
spin_lock(&si->lock); |
7992fde72 swapfile: swap al... |
335 336 337 338 339 340 341 342 343 344 345 346 347 |
si->lowest_alloc = 0; si->flags &= ~SWP_DISCARDING; smp_mb(); /* wake_up_bit advises this */ wake_up_bit(&si->flags, ilog2(SWP_DISCARDING)); } else if (si->flags & SWP_DISCARDING) { /* * Delay using pages allocated by racing tasks * until the whole discard has been issued. We * could defer that delay until swap_writepage, * but it's easier to keep this self-contained. */ |
ec8acf20a swap: add per-par... |
348 |
spin_unlock(&si->lock); |
7992fde72 swapfile: swap al... |
349 350 |
wait_on_bit(&si->flags, ilog2(SWP_DISCARDING), wait_for_discard, TASK_UNINTERRUPTIBLE); |
ec8acf20a swap: add per-par... |
351 |
spin_lock(&si->lock); |
7992fde72 swapfile: swap al... |
352 353 354 355 356 357 358 359 360 361 362 363 |
} else { /* * Note pages allocated by racing tasks while * scan for a free cluster is in progress, so * that its final discard can exclude them. */ if (offset < si->lowest_alloc) si->lowest_alloc = offset; if (offset > si->highest_alloc) si->highest_alloc = offset; } } |
ebebbbe90 swapfile: rearran... |
364 |
return offset; |
7dfad4183 [PATCH] swap: sca... |
365 |
|
ebebbbe90 swapfile: rearran... |
366 |
scan: |
ec8acf20a swap: add per-par... |
367 |
spin_unlock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
368 |
while (++offset <= si->highest_bit) { |
52b7efdbe [PATCH] swap: sca... |
369 |
if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
370 |
spin_lock(&si->lock); |
52b7efdbe [PATCH] swap: sca... |
371 372 |
goto checks; } |
c9e444103 mm: reuse unused ... |
373 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
374 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
375 376 |
goto checks; } |
048c27fd7 [PATCH] swap: sca... |
377 378 379 380 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } |
7dfad4183 [PATCH] swap: sca... |
381 |
} |
c60aa176c swapfile: swap al... |
382 383 384 |
offset = si->lowest_bit; while (++offset < scan_base) { if (!si->swap_map[offset]) { |
ec8acf20a swap: add per-par... |
385 |
spin_lock(&si->lock); |
c60aa176c swapfile: swap al... |
386 387 |
goto checks; } |
c9e444103 mm: reuse unused ... |
388 |
if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { |
ec8acf20a swap: add per-par... |
389 |
spin_lock(&si->lock); |
c9e444103 mm: reuse unused ... |
390 391 |
goto checks; } |
c60aa176c swapfile: swap al... |
392 393 394 395 396 |
if (unlikely(--latency_ration < 0)) { cond_resched(); latency_ration = LATENCY_LIMIT; } } |
ec8acf20a swap: add per-par... |
397 |
spin_lock(&si->lock); |
7dfad4183 [PATCH] swap: sca... |
398 399 |
no_page: |
52b7efdbe [PATCH] swap: sca... |
400 |
si->flags -= SWP_SCANNING; |
1da177e4c Linux-2.6.12-rc2 |
401 402 403 404 405 |
return 0; } swp_entry_t get_swap_page(void) { |
fb4f88dca [PATCH] swap: get... |
406 407 408 409 |
struct swap_info_struct *si; pgoff_t offset; int type, next; int wrapped = 0; |
ec8acf20a swap: add per-par... |
410 |
int hp_index; |
1da177e4c Linux-2.6.12-rc2 |
411 |
|
5d337b919 [PATCH] swap: swa... |
412 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
413 |
if (atomic_long_read(&nr_swap_pages) <= 0) |
fb4f88dca [PATCH] swap: get... |
414 |
goto noswap; |
ec8acf20a swap: add per-par... |
415 |
atomic_long_dec(&nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
416 417 |
for (type = swap_list.next; type >= 0 && wrapped < 2; type = next) { |
ec8acf20a swap: add per-par... |
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 |
hp_index = atomic_xchg(&highest_priority_index, -1); /* * highest_priority_index records current highest priority swap * type which just frees swap entries. If its priority is * higher than that of swap_list.next swap type, we use it. It * isn't protected by swap_lock, so it can be an invalid value * if the corresponding swap type is swapoff. We double check * the flags here. It's even possible the swap type is swapoff * and swapon again and its priority is changed. In such rare * case, low prority swap type might be used, but eventually * high priority swap will be used after several rounds of * swap. */ if (hp_index != -1 && hp_index != type && swap_info[type]->prio < swap_info[hp_index]->prio && (swap_info[hp_index]->flags & SWP_WRITEOK)) { type = hp_index; swap_list.next = type; } |
efa90a981 swap_info: change... |
437 |
si = swap_info[type]; |
fb4f88dca [PATCH] swap: get... |
438 439 |
next = si->next; if (next < 0 || |
efa90a981 swap_info: change... |
440 |
(!wrapped && si->prio != swap_info[next]->prio)) { |
fb4f88dca [PATCH] swap: get... |
441 442 |
next = swap_list.head; wrapped++; |
1da177e4c Linux-2.6.12-rc2 |
443 |
} |
fb4f88dca [PATCH] swap: get... |
444 |
|
ec8acf20a swap: add per-par... |
445 446 447 |
spin_lock(&si->lock); if (!si->highest_bit) { spin_unlock(&si->lock); |
fb4f88dca [PATCH] swap: get... |
448 |
continue; |
ec8acf20a swap: add per-par... |
449 450 451 |
} if (!(si->flags & SWP_WRITEOK)) { spin_unlock(&si->lock); |
fb4f88dca [PATCH] swap: get... |
452 |
continue; |
ec8acf20a swap: add per-par... |
453 |
} |
fb4f88dca [PATCH] swap: get... |
454 455 |
swap_list.next = next; |
ec8acf20a swap: add per-par... |
456 457 |
spin_unlock(&swap_lock); |
355cfa73d mm: modify swap_m... |
458 |
/* This is called for allocating swap entry for cache */ |
253d553ba swap_info: SWAP_H... |
459 |
offset = scan_swap_map(si, SWAP_HAS_CACHE); |
ec8acf20a swap: add per-par... |
460 461 |
spin_unlock(&si->lock); if (offset) |
fb4f88dca [PATCH] swap: get... |
462 |
return swp_entry(type, offset); |
ec8acf20a swap: add per-par... |
463 |
spin_lock(&swap_lock); |
fb4f88dca [PATCH] swap: get... |
464 |
next = swap_list.next; |
1da177e4c Linux-2.6.12-rc2 |
465 |
} |
fb4f88dca [PATCH] swap: get... |
466 |
|
ec8acf20a swap: add per-par... |
467 |
atomic_long_inc(&nr_swap_pages); |
fb4f88dca [PATCH] swap: get... |
468 |
noswap: |
5d337b919 [PATCH] swap: swa... |
469 |
spin_unlock(&swap_lock); |
fb4f88dca [PATCH] swap: get... |
470 |
return (swp_entry_t) {0}; |
1da177e4c Linux-2.6.12-rc2 |
471 |
} |
910321ea8 swap: revert spec... |
472 473 474 475 476 |
/* The only caller of this function is now susupend routine */ swp_entry_t get_swap_page_of_type(int type) { struct swap_info_struct *si; pgoff_t offset; |
910321ea8 swap: revert spec... |
477 |
si = swap_info[type]; |
ec8acf20a swap: add per-par... |
478 |
spin_lock(&si->lock); |
910321ea8 swap: revert spec... |
479 |
if (si && (si->flags & SWP_WRITEOK)) { |
ec8acf20a swap: add per-par... |
480 |
atomic_long_dec(&nr_swap_pages); |
910321ea8 swap: revert spec... |
481 482 483 |
/* This is called for allocating swap entry, not cache */ offset = scan_swap_map(si, 1); if (offset) { |
ec8acf20a swap: add per-par... |
484 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
485 486 |
return swp_entry(type, offset); } |
ec8acf20a swap: add per-par... |
487 |
atomic_long_inc(&nr_swap_pages); |
910321ea8 swap: revert spec... |
488 |
} |
ec8acf20a swap: add per-par... |
489 |
spin_unlock(&si->lock); |
910321ea8 swap: revert spec... |
490 491 |
return (swp_entry_t) {0}; } |
73c34b6ac swap_info: miscel... |
492 |
static struct swap_info_struct *swap_info_get(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
493 |
{ |
73c34b6ac swap_info: miscel... |
494 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
495 496 497 498 499 500 501 |
unsigned long offset, type; if (!entry.val) goto out; type = swp_type(entry); if (type >= nr_swapfiles) goto bad_nofile; |
efa90a981 swap_info: change... |
502 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
503 504 505 506 507 508 509 |
if (!(p->flags & SWP_USED)) goto bad_device; offset = swp_offset(entry); if (offset >= p->max) goto bad_offset; if (!p->swap_map[offset]) goto bad_free; |
ec8acf20a swap: add per-par... |
510 |
spin_lock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 |
return p; bad_free: printk(KERN_ERR "swap_free: %s%08lx ", Unused_offset, entry.val); goto out; bad_offset: printk(KERN_ERR "swap_free: %s%08lx ", Bad_offset, entry.val); goto out; bad_device: printk(KERN_ERR "swap_free: %s%08lx ", Unused_file, entry.val); goto out; bad_nofile: printk(KERN_ERR "swap_free: %s%08lx ", Bad_file, entry.val); out: return NULL; |
886bb7e9c swapfile: remove ... |
530 |
} |
1da177e4c Linux-2.6.12-rc2 |
531 |
|
ec8acf20a swap: add per-par... |
532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 |
/* * This swap type frees swap entry, check if it is the highest priority swap * type which just frees swap entry. get_swap_page() uses * highest_priority_index to search highest priority swap type. The * swap_info_struct.lock can't protect us if there are multiple swap types * active, so we use atomic_cmpxchg. */ static void set_highest_priority_index(int type) { int old_hp_index, new_hp_index; do { old_hp_index = atomic_read(&highest_priority_index); if (old_hp_index != -1 && swap_info[old_hp_index]->prio >= swap_info[type]->prio) break; new_hp_index = type; } while (atomic_cmpxchg(&highest_priority_index, old_hp_index, new_hp_index) != old_hp_index); } |
8d69aaee8 swap_info: swap_m... |
552 553 |
static unsigned char swap_entry_free(struct swap_info_struct *p, swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
554 |
{ |
253d553ba swap_info: SWAP_H... |
555 |
unsigned long offset = swp_offset(entry); |
8d69aaee8 swap_info: swap_m... |
556 557 |
unsigned char count; unsigned char has_cache; |
355cfa73d mm: modify swap_m... |
558 |
|
253d553ba swap_info: SWAP_H... |
559 560 561 |
count = p->swap_map[offset]; has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; |
355cfa73d mm: modify swap_m... |
562 |
|
253d553ba swap_info: SWAP_H... |
563 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
564 |
VM_BUG_ON(!has_cache); |
253d553ba swap_info: SWAP_H... |
565 |
has_cache = 0; |
aaa468653 swap_info: note S... |
566 567 568 569 570 571 |
} 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... |
572 573 574 575 576 577 578 579 580 |
} 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... |
581 582 583 584 585 586 |
if (!count) mem_cgroup_uncharge_swap(entry); usage = count | has_cache; p->swap_map[offset] = usage; |
355cfa73d mm: modify swap_m... |
587 |
|
355cfa73d mm: modify swap_m... |
588 |
/* free if no reference */ |
253d553ba swap_info: SWAP_H... |
589 |
if (!usage) { |
355cfa73d mm: modify swap_m... |
590 591 592 593 |
if (offset < p->lowest_bit) p->lowest_bit = offset; if (offset > p->highest_bit) p->highest_bit = offset; |
ec8acf20a swap: add per-par... |
594 595 |
set_highest_priority_index(p->type); atomic_long_inc(&nr_swap_pages); |
355cfa73d mm: modify swap_m... |
596 |
p->inuse_pages--; |
38b5faf4b mm: frontswap: co... |
597 |
frontswap_invalidate_page(p->type, offset); |
737449236 swapfile: avoid d... |
598 599 600 601 602 603 |
if (p->flags & SWP_BLKDEV) { struct gendisk *disk = p->bdev->bd_disk; if (disk->fops->swap_slot_free_notify) disk->fops->swap_slot_free_notify(p->bdev, offset); } |
1da177e4c Linux-2.6.12-rc2 |
604 |
} |
253d553ba swap_info: SWAP_H... |
605 606 |
return usage; |
1da177e4c Linux-2.6.12-rc2 |
607 608 609 610 611 612 613 614 |
} /* * Caller has made sure that the swapdevice corresponding to entry * is still around or has not been recycled. */ void swap_free(swp_entry_t entry) { |
73c34b6ac swap_info: miscel... |
615 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
616 617 618 |
p = swap_info_get(entry); if (p) { |
253d553ba swap_info: SWAP_H... |
619 |
swap_entry_free(p, entry, 1); |
ec8acf20a swap: add per-par... |
620 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
621 622 623 624 |
} } /* |
cb4b86ba4 mm: add swap cach... |
625 626 627 628 |
* Called after dropping swapcache to decrease refcnt to swap entries. */ void swapcache_free(swp_entry_t entry, struct page *page) { |
355cfa73d mm: modify swap_m... |
629 |
struct swap_info_struct *p; |
8d69aaee8 swap_info: swap_m... |
630 |
unsigned char count; |
355cfa73d mm: modify swap_m... |
631 |
|
355cfa73d mm: modify swap_m... |
632 633 |
p = swap_info_get(entry); if (p) { |
253d553ba swap_info: SWAP_H... |
634 635 636 |
count = swap_entry_free(p, entry, SWAP_HAS_CACHE); if (page) mem_cgroup_uncharge_swapcache(page, entry, count != 0); |
ec8acf20a swap: add per-par... |
637 |
spin_unlock(&p->lock); |
355cfa73d mm: modify swap_m... |
638 |
} |
cb4b86ba4 mm: add swap cach... |
639 640 641 |
} /* |
c475a8ab6 [PATCH] can_share... |
642 |
* How many references to page are currently swapped out? |
570a335b8 swap_info: swap c... |
643 644 |
* 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 |
645 |
*/ |
bde05d1cc shmem: replace pa... |
646 |
int page_swapcount(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
647 |
{ |
c475a8ab6 [PATCH] can_share... |
648 649 |
int count = 0; struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
650 |
swp_entry_t entry; |
4c21e2f24 [PATCH] mm: split... |
651 |
entry.val = page_private(page); |
1da177e4c Linux-2.6.12-rc2 |
652 653 |
p = swap_info_get(entry); if (p) { |
355cfa73d mm: modify swap_m... |
654 |
count = swap_count(p->swap_map[swp_offset(entry)]); |
ec8acf20a swap: add per-par... |
655 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
656 |
} |
c475a8ab6 [PATCH] can_share... |
657 |
return count; |
1da177e4c Linux-2.6.12-rc2 |
658 659 660 |
} /* |
7b1fe5979 mm: reuse_swap_pa... |
661 662 663 664 |
* We can write to an anon page without COW if there are no other references * to it. And as a side-effect, free up its swap: because the old content * on disk will never be read, and seeking back there to write new content * later would only waste time away from clustering. |
1da177e4c Linux-2.6.12-rc2 |
665 |
*/ |
7b1fe5979 mm: reuse_swap_pa... |
666 |
int reuse_swap_page(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
667 |
{ |
c475a8ab6 [PATCH] can_share... |
668 |
int count; |
51726b122 mm: replace some ... |
669 |
VM_BUG_ON(!PageLocked(page)); |
5ad646880 ksm: let shared p... |
670 671 |
if (unlikely(PageKsm(page))) return 0; |
c475a8ab6 [PATCH] can_share... |
672 |
count = page_mapcount(page); |
7b1fe5979 mm: reuse_swap_pa... |
673 |
if (count <= 1 && PageSwapCache(page)) { |
c475a8ab6 [PATCH] can_share... |
674 |
count += page_swapcount(page); |
7b1fe5979 mm: reuse_swap_pa... |
675 676 677 678 679 |
if (count == 1 && !PageWriteback(page)) { delete_from_swap_cache(page); SetPageDirty(page); } } |
5ad646880 ksm: let shared p... |
680 |
return count <= 1; |
1da177e4c Linux-2.6.12-rc2 |
681 682 683 |
} /* |
a2c43eed8 mm: try_to_free_s... |
684 685 |
* 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 |
686 |
*/ |
a2c43eed8 mm: try_to_free_s... |
687 |
int try_to_free_swap(struct page *page) |
1da177e4c Linux-2.6.12-rc2 |
688 |
{ |
51726b122 mm: replace some ... |
689 |
VM_BUG_ON(!PageLocked(page)); |
1da177e4c Linux-2.6.12-rc2 |
690 691 692 693 694 |
if (!PageSwapCache(page)) return 0; if (PageWriteback(page)) return 0; |
a2c43eed8 mm: try_to_free_s... |
695 |
if (page_swapcount(page)) |
1da177e4c Linux-2.6.12-rc2 |
696 |
return 0; |
b73d7fcec swap: prevent reu... |
697 698 699 700 701 702 703 704 705 706 707 708 |
/* * 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. * |
f90ac3982 mm: avoid liveloc... |
709 710 |
* Hibration suspends storage while it is writing the image * to disk so check that here. |
b73d7fcec swap: prevent reu... |
711 |
*/ |
f90ac3982 mm: avoid liveloc... |
712 |
if (pm_suspended_storage()) |
b73d7fcec swap: prevent reu... |
713 |
return 0; |
a2c43eed8 mm: try_to_free_s... |
714 715 716 |
delete_from_swap_cache(page); SetPageDirty(page); return 1; |
68a22394c vmscan: free swap... |
717 718 719 |
} /* |
1da177e4c Linux-2.6.12-rc2 |
720 721 722 |
* 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... |
723 |
int free_swap_and_cache(swp_entry_t entry) |
1da177e4c Linux-2.6.12-rc2 |
724 |
{ |
2509ef26d badpage: zap prin... |
725 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
726 |
struct page *page = NULL; |
a7420aa54 HWPOISON: Add sup... |
727 |
if (non_swap_entry(entry)) |
2509ef26d badpage: zap prin... |
728 |
return 1; |
0697212a4 [PATCH] Swapless ... |
729 |
|
1da177e4c Linux-2.6.12-rc2 |
730 731 |
p = swap_info_get(entry); if (p) { |
253d553ba swap_info: SWAP_H... |
732 |
if (swap_entry_free(p, entry, 1) == SWAP_HAS_CACHE) { |
33806f06d swap: make each s... |
733 734 |
page = find_get_page(swap_address_space(entry), entry.val); |
8413ac9d8 mm: page lock use... |
735 |
if (page && !trylock_page(page)) { |
93fac7041 [PATCH] mm: sched... |
736 737 738 739 |
page_cache_release(page); page = NULL; } } |
ec8acf20a swap: add per-par... |
740 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
741 742 |
} if (page) { |
a2c43eed8 mm: try_to_free_s... |
743 744 745 746 |
/* * Not mapped elsewhere, or swap space full? Free it! * Also recheck PageSwapCache now page is locked (above). */ |
93fac7041 [PATCH] mm: sched... |
747 |
if (PageSwapCache(page) && !PageWriteback(page) && |
a2c43eed8 mm: try_to_free_s... |
748 |
(!page_mapped(page) || vm_swap_full())) { |
1da177e4c Linux-2.6.12-rc2 |
749 750 751 752 753 754 |
delete_from_swap_cache(page); SetPageDirty(page); } unlock_page(page); page_cache_release(page); } |
2509ef26d badpage: zap prin... |
755 |
return p != NULL; |
1da177e4c Linux-2.6.12-rc2 |
756 |
} |
b0cb1a19d Replace CONFIG_SO... |
757 |
#ifdef CONFIG_HIBERNATION |
f577eb30a [PATCH] swsusp: l... |
758 |
/* |
915bae9eb [PATCH] swsusp: u... |
759 |
* Find the swap type that corresponds to given device (if any). |
f577eb30a [PATCH] swsusp: l... |
760 |
* |
915bae9eb [PATCH] swsusp: u... |
761 762 763 764 |
* @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... |
765 |
*/ |
7bf236874 [PATCH] swsusp: D... |
766 |
int swap_type_of(dev_t device, sector_t offset, struct block_device **bdev_p) |
f577eb30a [PATCH] swsusp: l... |
767 |
{ |
915bae9eb [PATCH] swsusp: u... |
768 |
struct block_device *bdev = NULL; |
efa90a981 swap_info: change... |
769 |
int type; |
f577eb30a [PATCH] swsusp: l... |
770 |
|
915bae9eb [PATCH] swsusp: u... |
771 772 |
if (device) bdev = bdget(device); |
f577eb30a [PATCH] swsusp: l... |
773 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
774 775 |
for (type = 0; type < nr_swapfiles; type++) { struct swap_info_struct *sis = swap_info[type]; |
f577eb30a [PATCH] swsusp: l... |
776 |
|
915bae9eb [PATCH] swsusp: u... |
777 |
if (!(sis->flags & SWP_WRITEOK)) |
f577eb30a [PATCH] swsusp: l... |
778 |
continue; |
b6b5bce35 [PATCH] swsusp: F... |
779 |
|
915bae9eb [PATCH] swsusp: u... |
780 |
if (!bdev) { |
7bf236874 [PATCH] swsusp: D... |
781 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
782 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
783 |
|
6e1819d61 [PATCH] swsusp: u... |
784 |
spin_unlock(&swap_lock); |
efa90a981 swap_info: change... |
785 |
return type; |
6e1819d61 [PATCH] swsusp: u... |
786 |
} |
915bae9eb [PATCH] swsusp: u... |
787 |
if (bdev == sis->bdev) { |
9625a5f28 swap_info: includ... |
788 |
struct swap_extent *se = &sis->first_swap_extent; |
915bae9eb [PATCH] swsusp: u... |
789 |
|
915bae9eb [PATCH] swsusp: u... |
790 |
if (se->start_block == offset) { |
7bf236874 [PATCH] swsusp: D... |
791 |
if (bdev_p) |
dddac6a7b PM / Hibernate: R... |
792 |
*bdev_p = bdgrab(sis->bdev); |
7bf236874 [PATCH] swsusp: D... |
793 |
|
915bae9eb [PATCH] swsusp: u... |
794 795 |
spin_unlock(&swap_lock); bdput(bdev); |
efa90a981 swap_info: change... |
796 |
return type; |
915bae9eb [PATCH] swsusp: u... |
797 |
} |
f577eb30a [PATCH] swsusp: l... |
798 799 800 |
} } spin_unlock(&swap_lock); |
915bae9eb [PATCH] swsusp: u... |
801 802 |
if (bdev) bdput(bdev); |
f577eb30a [PATCH] swsusp: l... |
803 804 805 806 |
return -ENODEV; } /* |
73c34b6ac swap_info: miscel... |
807 808 809 810 811 812 813 814 815 816 817 |
* 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_... |
818 |
return map_swap_entry(swp_entry(type, offset), &bdev); |
73c34b6ac swap_info: miscel... |
819 820 821 |
} /* |
f577eb30a [PATCH] swsusp: l... |
822 823 824 825 826 827 828 829 |
* 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... |
830 831 832 |
spin_lock(&swap_lock); if ((unsigned int)type < nr_swapfiles) { struct swap_info_struct *sis = swap_info[type]; |
ec8acf20a swap: add per-par... |
833 |
spin_lock(&sis->lock); |
efa90a981 swap_info: change... |
834 835 |
if (sis->flags & SWP_WRITEOK) { n = sis->pages; |
f577eb30a [PATCH] swsusp: l... |
836 |
if (free) |
efa90a981 swap_info: change... |
837 |
n -= sis->inuse_pages; |
f577eb30a [PATCH] swsusp: l... |
838 |
} |
ec8acf20a swap: add per-par... |
839 |
spin_unlock(&sis->lock); |
f577eb30a [PATCH] swsusp: l... |
840 |
} |
efa90a981 swap_info: change... |
841 |
spin_unlock(&swap_lock); |
f577eb30a [PATCH] swsusp: l... |
842 843 |
return n; } |
73c34b6ac swap_info: miscel... |
844 |
#endif /* CONFIG_HIBERNATION */ |
f577eb30a [PATCH] swsusp: l... |
845 |
|
1da177e4c Linux-2.6.12-rc2 |
846 |
/* |
72866f6f2 [PATCH] mm: anon ... |
847 848 849 |
* 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 |
850 |
*/ |
044d66c1d memcgroup: reinst... |
851 |
static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd, |
1da177e4c Linux-2.6.12-rc2 |
852 853 |
unsigned long addr, swp_entry_t entry, struct page *page) { |
9e16b7fb1 mm,ksm: swapoff m... |
854 |
struct page *swapcache; |
72835c86c mm: unify remaini... |
855 |
struct mem_cgroup *memcg; |
044d66c1d memcgroup: reinst... |
856 857 858 |
spinlock_t *ptl; pte_t *pte; int ret = 1; |
9e16b7fb1 mm,ksm: swapoff m... |
859 860 861 862 |
swapcache = page; page = ksm_might_need_to_copy(page, vma, addr); if (unlikely(!page)) return -ENOMEM; |
72835c86c mm: unify remaini... |
863 864 |
if (mem_cgroup_try_charge_swapin(vma->vm_mm, page, GFP_KERNEL, &memcg)) { |
044d66c1d memcgroup: reinst... |
865 |
ret = -ENOMEM; |
85d9fc89f memcg: fix refcnt... |
866 867 |
goto out_nolock; } |
044d66c1d memcgroup: reinst... |
868 869 870 |
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); if (unlikely(!pte_same(*pte, swp_entry_to_pte(entry)))) { |
5d84c7766 mm: swapfile: cle... |
871 |
mem_cgroup_cancel_charge_swapin(memcg); |
044d66c1d memcgroup: reinst... |
872 873 874 |
ret = 0; goto out; } |
8a9f3ccd2 Memory controller... |
875 |
|
b084d4353 mm: count swap usage |
876 |
dec_mm_counter(vma->vm_mm, MM_SWAPENTS); |
d559db086 mm: clean up mm_c... |
877 |
inc_mm_counter(vma->vm_mm, MM_ANONPAGES); |
1da177e4c Linux-2.6.12-rc2 |
878 879 880 |
get_page(page); set_pte_at(vma->vm_mm, addr, pte, pte_mkold(mk_pte(page, vma->vm_page_prot))); |
9e16b7fb1 mm,ksm: swapoff m... |
881 882 883 884 |
if (page == swapcache) page_add_anon_rmap(page, vma, addr); else /* ksm created a completely new copy */ page_add_new_anon_rmap(page, vma, addr); |
72835c86c mm: unify remaini... |
885 |
mem_cgroup_commit_charge_swapin(page, memcg); |
1da177e4c Linux-2.6.12-rc2 |
886 887 888 889 890 891 |
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... |
892 893 |
out: pte_unmap_unlock(pte, ptl); |
85d9fc89f memcg: fix refcnt... |
894 |
out_nolock: |
9e16b7fb1 mm,ksm: swapoff m... |
895 896 897 898 |
if (page != swapcache) { unlock_page(page); put_page(page); } |
044d66c1d memcgroup: reinst... |
899 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
900 901 902 903 904 905 |
} 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 |
906 |
pte_t swp_pte = swp_entry_to_pte(entry); |
705e87c0c [PATCH] mm: pte_o... |
907 |
pte_t *pte; |
8a9f3ccd2 Memory controller... |
908 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
909 |
|
044d66c1d memcgroup: reinst... |
910 911 912 913 914 915 916 917 918 919 |
/* * 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 * preemptible whenever CONFIG_PREEMPT but not CONFIG_HIGHPTE. */ pte = pte_offset_map(pmd, addr); |
1da177e4c Linux-2.6.12-rc2 |
920 921 922 923 924 925 |
do { /* * swapoff spends a _lot_ of time in this loop! * Test inline before going to call unuse_pte. */ if (unlikely(pte_same(*pte, swp_pte))) { |
044d66c1d memcgroup: reinst... |
926 927 928 929 930 |
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 |
931 932 |
} } while (pte++, addr += PAGE_SIZE, addr != end); |
044d66c1d memcgroup: reinst... |
933 934 |
pte_unmap(pte - 1); out: |
8a9f3ccd2 Memory controller... |
935 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
936 937 938 939 940 941 942 943 |
} 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... |
944 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
945 946 947 948 |
pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); |
1a5a9906d mm: thp: fix pmd_... |
949 |
if (pmd_none_or_trans_huge_or_clear_bad(pmd)) |
1da177e4c Linux-2.6.12-rc2 |
950 |
continue; |
8a9f3ccd2 Memory controller... |
951 952 953 |
ret = unuse_pte_range(vma, pmd, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
954 955 956 957 958 959 960 961 962 963 |
} while (pmd++, addr = next, addr != end); return 0; } static inline int unuse_pud_range(struct vm_area_struct *vma, pgd_t *pgd, unsigned long addr, unsigned long end, swp_entry_t entry, struct page *page) { pud_t *pud; unsigned long next; |
8a9f3ccd2 Memory controller... |
964 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
965 966 967 968 969 970 |
pud = pud_offset(pgd, addr); do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) continue; |
8a9f3ccd2 Memory controller... |
971 972 973 |
ret = unuse_pmd_range(vma, pud, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
974 975 976 977 978 979 980 981 982 |
} while (pud++, addr = next, addr != end); return 0; } static int unuse_vma(struct vm_area_struct *vma, swp_entry_t entry, struct page *page) { pgd_t *pgd; unsigned long addr, end, next; |
8a9f3ccd2 Memory controller... |
983 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
984 |
|
3ca7b3c5b mm: define PAGE_M... |
985 |
if (page_anon_vma(page)) { |
1da177e4c Linux-2.6.12-rc2 |
986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 |
addr = page_address_in_vma(page, vma); if (addr == -EFAULT) return 0; else end = addr + PAGE_SIZE; } else { addr = vma->vm_start; end = vma->vm_end; } pgd = pgd_offset(vma->vm_mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; |
8a9f3ccd2 Memory controller... |
1001 1002 1003 |
ret = unuse_pud_range(vma, pgd, addr, next, entry, page); if (ret) return ret; |
1da177e4c Linux-2.6.12-rc2 |
1004 1005 1006 1007 1008 1009 1010 1011 |
} 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... |
1012 |
int ret = 0; |
1da177e4c Linux-2.6.12-rc2 |
1013 1014 1015 |
if (!down_read_trylock(&mm->mmap_sem)) { /* |
7d03431cf swapfile/vmscan: ... |
1016 1017 |
* 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 |
1018 |
*/ |
c475a8ab6 [PATCH] can_share... |
1019 |
activate_page(page); |
1da177e4c Linux-2.6.12-rc2 |
1020 1021 1022 1023 |
unlock_page(page); down_read(&mm->mmap_sem); lock_page(page); } |
1da177e4c Linux-2.6.12-rc2 |
1024 |
for (vma = mm->mmap; vma; vma = vma->vm_next) { |
8a9f3ccd2 Memory controller... |
1025 |
if (vma->anon_vma && (ret = unuse_vma(vma, entry, page))) |
1da177e4c Linux-2.6.12-rc2 |
1026 1027 |
break; } |
1da177e4c Linux-2.6.12-rc2 |
1028 |
up_read(&mm->mmap_sem); |
8a9f3ccd2 Memory controller... |
1029 |
return (ret < 0)? ret: 0; |
1da177e4c Linux-2.6.12-rc2 |
1030 1031 1032 |
} /* |
38b5faf4b mm: frontswap: co... |
1033 1034 |
* 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 |
1035 1036 |
* Recycle to start on reaching the end, returning 0 when empty. */ |
6eb396dc4 [PATCH] swap: swa... |
1037 |
static unsigned int find_next_to_unuse(struct swap_info_struct *si, |
38b5faf4b mm: frontswap: co... |
1038 |
unsigned int prev, bool frontswap) |
1da177e4c Linux-2.6.12-rc2 |
1039 |
{ |
6eb396dc4 [PATCH] swap: swa... |
1040 1041 |
unsigned int max = si->max; unsigned int i = prev; |
8d69aaee8 swap_info: swap_m... |
1042 |
unsigned char count; |
1da177e4c Linux-2.6.12-rc2 |
1043 1044 |
/* |
5d337b919 [PATCH] swap: swa... |
1045 |
* No need for swap_lock here: we're just looking |
1da177e4c Linux-2.6.12-rc2 |
1046 1047 |
* 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... |
1048 |
* allocations from this area (while holding swap_lock). |
1da177e4c Linux-2.6.12-rc2 |
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 |
*/ for (;;) { if (++i >= max) { if (!prev) { i = 0; break; } /* * No entries in use at top of swap_map, * loop back to start and recheck there. */ max = prev + 1; prev = 0; i = 1; } |
38b5faf4b mm: frontswap: co... |
1064 1065 1066 1067 1068 1069 |
if (frontswap) { if (frontswap_test(si, i)) break; else continue; } |
1da177e4c Linux-2.6.12-rc2 |
1070 |
count = si->swap_map[i]; |
355cfa73d mm: modify swap_m... |
1071 |
if (count && swap_count(count) != SWAP_MAP_BAD) |
1da177e4c Linux-2.6.12-rc2 |
1072 1073 1074 1075 1076 1077 1078 1079 1080 |
break; } return i; } /* * We completely avoid races by reading each swap page in advance, * and then search for the process using it. All the necessary * page table adjustments can then be made atomically. |
38b5faf4b mm: frontswap: co... |
1081 1082 1083 |
* * 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 |
1084 |
*/ |
38b5faf4b mm: frontswap: co... |
1085 1086 |
int try_to_unuse(unsigned int type, bool frontswap, unsigned long pages_to_unuse) |
1da177e4c Linux-2.6.12-rc2 |
1087 |
{ |
efa90a981 swap_info: change... |
1088 |
struct swap_info_struct *si = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
1089 |
struct mm_struct *start_mm; |
8d69aaee8 swap_info: swap_m... |
1090 1091 |
unsigned char *swap_map; unsigned char swcount; |
1da177e4c Linux-2.6.12-rc2 |
1092 1093 |
struct page *page; swp_entry_t entry; |
6eb396dc4 [PATCH] swap: swa... |
1094 |
unsigned int i = 0; |
1da177e4c Linux-2.6.12-rc2 |
1095 |
int retval = 0; |
1da177e4c Linux-2.6.12-rc2 |
1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 |
/* * 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... |
1109 |
* that. |
1da177e4c Linux-2.6.12-rc2 |
1110 1111 1112 1113 1114 1115 1116 1117 1118 |
*/ start_mm = &init_mm; atomic_inc(&init_mm.mm_users); /* * Keep on scanning until all entries have gone. Usually, * one pass through swap_map is enough, but not necessarily: * there are races when an instance of an entry might be missed. */ |
38b5faf4b mm: frontswap: co... |
1119 |
while ((i = find_next_to_unuse(si, i, frontswap)) != 0) { |
1da177e4c Linux-2.6.12-rc2 |
1120 1121 1122 1123 |
if (signal_pending(current)) { retval = -EINTR; break; } |
886bb7e9c swapfile: remove ... |
1124 |
/* |
1da177e4c Linux-2.6.12-rc2 |
1125 1126 |
* Get a page for the entry, using the existing swap * cache page if there is one. Otherwise, get a clean |
886bb7e9c swapfile: remove ... |
1127 |
* page and read the swap into it. |
1da177e4c Linux-2.6.12-rc2 |
1128 1129 1130 |
*/ swap_map = &si->swap_map[i]; entry = swp_entry(type, i); |
02098feaa swapin needs gfp_... |
1131 1132 |
page = read_swap_cache_async(entry, GFP_HIGHUSER_MOVABLE, NULL, 0); |
1da177e4c Linux-2.6.12-rc2 |
1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 |
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. */ if (!*swap_map) continue; retval = -ENOMEM; break; } /* * Don't hold on to start_mm if it looks like exiting. */ if (atomic_read(&start_mm->mm_users) == 1) { mmput(start_mm); start_mm = &init_mm; atomic_inc(&init_mm.mm_users); } /* * Wait for and lock page. When do_swap_page races with * try_to_unuse, do_swap_page can handle the fault much * faster than try_to_unuse can locate the entry. This * apparently redundant "wait_on_page_locked" lets try_to_unuse * defer to do_swap_page in such a case - in some tests, * do_swap_page and try_to_unuse repeatedly compete. */ wait_on_page_locked(page); wait_on_page_writeback(page); lock_page(page); wait_on_page_writeback(page); /* * Remove all references to entry. |
1da177e4c Linux-2.6.12-rc2 |
1170 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
1171 |
swcount = *swap_map; |
aaa468653 swap_info: note S... |
1172 1173 1174 1175 1176 1177 |
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 |
1178 |
} |
aaa468653 swap_info: note S... |
1179 1180 |
if (swap_count(swcount) && start_mm != &init_mm) retval = unuse_mm(start_mm, entry, page); |
355cfa73d mm: modify swap_m... |
1181 |
if (swap_count(*swap_map)) { |
1da177e4c Linux-2.6.12-rc2 |
1182 1183 1184 1185 1186 1187 1188 1189 1190 |
int set_start_mm = (*swap_map >= swcount); struct list_head *p = &start_mm->mmlist; struct mm_struct *new_start_mm = start_mm; struct mm_struct *prev_mm = start_mm; struct mm_struct *mm; atomic_inc(&new_start_mm->mm_users); atomic_inc(&prev_mm->mm_users); spin_lock(&mmlist_lock); |
aaa468653 swap_info: note S... |
1191 |
while (swap_count(*swap_map) && !retval && |
1da177e4c Linux-2.6.12-rc2 |
1192 1193 |
(p = p->next) != &start_mm->mmlist) { mm = list_entry(p, struct mm_struct, mmlist); |
70af7c5c6 [PATCH] swapoff: ... |
1194 |
if (!atomic_inc_not_zero(&mm->mm_users)) |
1da177e4c Linux-2.6.12-rc2 |
1195 |
continue; |
1da177e4c Linux-2.6.12-rc2 |
1196 1197 1198 1199 1200 1201 1202 |
spin_unlock(&mmlist_lock); mmput(prev_mm); prev_mm = mm; cond_resched(); swcount = *swap_map; |
355cfa73d mm: modify swap_m... |
1203 |
if (!swap_count(swcount)) /* any usage ? */ |
1da177e4c Linux-2.6.12-rc2 |
1204 |
; |
aaa468653 swap_info: note S... |
1205 |
else if (mm == &init_mm) |
1da177e4c Linux-2.6.12-rc2 |
1206 |
set_start_mm = 1; |
aaa468653 swap_info: note S... |
1207 |
else |
1da177e4c Linux-2.6.12-rc2 |
1208 |
retval = unuse_mm(mm, entry, page); |
355cfa73d mm: modify swap_m... |
1209 |
|
32c5fc10e mm: remove incorr... |
1210 |
if (set_start_mm && *swap_map < swcount) { |
1da177e4c Linux-2.6.12-rc2 |
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 |
mmput(new_start_mm); atomic_inc(&mm->mm_users); new_start_mm = mm; set_start_mm = 0; } spin_lock(&mmlist_lock); } spin_unlock(&mmlist_lock); mmput(prev_mm); mmput(start_mm); start_mm = new_start_mm; } if (retval) { unlock_page(page); page_cache_release(page); break; } /* |
1da177e4c Linux-2.6.12-rc2 |
1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 |
* 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... |
1241 1242 1243 1244 1245 1246 |
* * 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 |
1247 |
*/ |
355cfa73d mm: modify swap_m... |
1248 1249 |
if (swap_count(*swap_map) && PageDirty(page) && PageSwapCache(page)) { |
1da177e4c Linux-2.6.12-rc2 |
1250 1251 1252 1253 1254 1255 1256 1257 |
struct writeback_control wbc = { .sync_mode = WB_SYNC_NONE, }; swap_writepage(page, &wbc); lock_page(page); wait_on_page_writeback(page); } |
68bdc8d64 mm: try_to_unuse ... |
1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 |
/* * It is conceivable that a racing task removed this page from * swap cache just before we acquired the page lock at the top, * or while we dropped it in unuse_mm(). The page might even * be back in swap cache on another swap area: that we must not * delete, since it may not have been written out to swap yet. */ if (PageSwapCache(page) && likely(page_private(page) == entry.val)) |
2e0e26c76 tmpfs: open a win... |
1268 |
delete_from_swap_cache(page); |
1da177e4c Linux-2.6.12-rc2 |
1269 1270 1271 1272 |
/* * 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... |
1273 |
* mark page dirty so shrink_page_list will preserve it. |
1da177e4c Linux-2.6.12-rc2 |
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 |
*/ SetPageDirty(page); unlock_page(page); page_cache_release(page); /* * Make sure that we aren't completely killing * interactive performance. */ cond_resched(); |
38b5faf4b mm: frontswap: co... |
1284 1285 1286 1287 |
if (frontswap && pages_to_unuse > 0) { if (!--pages_to_unuse) break; } |
1da177e4c Linux-2.6.12-rc2 |
1288 1289 1290 |
} mmput(start_mm); |
1da177e4c Linux-2.6.12-rc2 |
1291 1292 1293 1294 |
return retval; } /* |
5d337b919 [PATCH] swap: swa... |
1295 1296 1297 |
* 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 |
1298 1299 1300 1301 1302 |
* 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... |
1303 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
1304 |
|
efa90a981 swap_info: change... |
1305 1306 |
for (type = 0; type < nr_swapfiles; type++) if (swap_info[type]->inuse_pages) |
1da177e4c Linux-2.6.12-rc2 |
1307 1308 1309 1310 1311 1312 1313 1314 1315 |
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_... |
1316 1317 1318 |
* 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 |
1319 |
*/ |
d4906e1aa swap: rework map_... |
1320 |
static sector_t map_swap_entry(swp_entry_t entry, struct block_device **bdev) |
1da177e4c Linux-2.6.12-rc2 |
1321 |
{ |
f29ad6a99 swap_info: privat... |
1322 1323 1324 1325 |
struct swap_info_struct *sis; struct swap_extent *start_se; struct swap_extent *se; pgoff_t offset; |
efa90a981 swap_info: change... |
1326 |
sis = swap_info[swp_type(entry)]; |
f29ad6a99 swap_info: privat... |
1327 1328 1329 1330 1331 |
*bdev = sis->bdev; offset = swp_offset(entry); start_se = sis->curr_swap_extent; se = start_se; |
1da177e4c Linux-2.6.12-rc2 |
1332 1333 1334 1335 1336 1337 1338 1339 |
for ( ; ; ) { struct list_head *lh; if (se->start_page <= offset && offset < (se->start_page + se->nr_pages)) { return se->start_block + (offset - se->start_page); } |
11d31886d [PATCH] swap: swa... |
1340 |
lh = se->list.next; |
1da177e4c Linux-2.6.12-rc2 |
1341 1342 1343 1344 1345 1346 1347 |
se = list_entry(lh, struct swap_extent, list); sis->curr_swap_extent = se; BUG_ON(se == start_se); /* It *must* be present */ } } /* |
d4906e1aa swap: rework map_... |
1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 |
* 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 |
1358 1359 1360 1361 |
* Free all of a swapdev's extent information */ static void destroy_swap_extents(struct swap_info_struct *sis) { |
9625a5f28 swap_info: includ... |
1362 |
while (!list_empty(&sis->first_swap_extent.list)) { |
1da177e4c Linux-2.6.12-rc2 |
1363 |
struct swap_extent *se; |
9625a5f28 swap_info: includ... |
1364 |
se = list_entry(sis->first_swap_extent.list.next, |
1da177e4c Linux-2.6.12-rc2 |
1365 1366 1367 1368 |
struct swap_extent, list); list_del(&se->list); kfree(se); } |
62c230bc1 mm: add support f... |
1369 1370 1371 1372 1373 1374 1375 1376 |
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 |
1377 1378 1379 1380 |
} /* * Add a block range (and the corresponding page range) into this swapdev's |
11d31886d [PATCH] swap: swa... |
1381 |
* extent list. The extent list is kept sorted in page order. |
1da177e4c Linux-2.6.12-rc2 |
1382 |
* |
11d31886d [PATCH] swap: swa... |
1383 |
* This function rather assumes that it is called in ascending page order. |
1da177e4c Linux-2.6.12-rc2 |
1384 |
*/ |
a509bc1a9 mm: swap: impleme... |
1385 |
int |
1da177e4c Linux-2.6.12-rc2 |
1386 1387 1388 1389 1390 1391 |
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... |
1392 1393 1394 1395 1396 1397 1398 1399 1400 |
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 |
1401 |
se = list_entry(lh, struct swap_extent, list); |
11d31886d [PATCH] swap: swa... |
1402 1403 |
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 |
1404 1405 1406 1407 |
/* Merge it */ se->nr_pages += nr_pages; return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 |
} /* * 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... |
1419 |
list_add_tail(&new_se->list, &sis->first_swap_extent.list); |
53092a740 [PATCH] swap: sho... |
1420 |
return 1; |
1da177e4c Linux-2.6.12-rc2 |
1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 |
} /* * 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... |
1443 |
* For S_ISREG swapfiles we set S_SWAPFILE across the life of the swapon. This |
1da177e4c Linux-2.6.12-rc2 |
1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 |
* 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... |
1454 |
static int setup_swap_extents(struct swap_info_struct *sis, sector_t *span) |
1da177e4c Linux-2.6.12-rc2 |
1455 |
{ |
62c230bc1 mm: add support f... |
1456 1457 1458 |
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 |
1459 |
int ret; |
1da177e4c Linux-2.6.12-rc2 |
1460 1461 |
if (S_ISBLK(inode->i_mode)) { ret = add_swap_extent(sis, 0, sis->max, 0); |
53092a740 [PATCH] swap: sho... |
1462 |
*span = sis->pages; |
a509bc1a9 mm: swap: impleme... |
1463 |
return ret; |
1da177e4c Linux-2.6.12-rc2 |
1464 |
} |
62c230bc1 mm: add support f... |
1465 |
if (mapping->a_ops->swap_activate) { |
a509bc1a9 mm: swap: impleme... |
1466 |
ret = mapping->a_ops->swap_activate(sis, swap_file, span); |
62c230bc1 mm: add support f... |
1467 1468 1469 1470 1471 |
if (!ret) { sis->flags |= SWP_FILE; ret = add_swap_extent(sis, 0, sis->max, 0); *span = sis->pages; } |
a509bc1a9 mm: swap: impleme... |
1472 |
return ret; |
62c230bc1 mm: add support f... |
1473 |
} |
a509bc1a9 mm: swap: impleme... |
1474 |
return generic_swapfile_activate(sis, swap_file, span); |
1da177e4c Linux-2.6.12-rc2 |
1475 |
} |
cf0cac0a0 mm: refactor rein... |
1476 |
static void _enable_swap_info(struct swap_info_struct *p, int prio, |
38b5faf4b mm: frontswap: co... |
1477 1478 |
unsigned char *swap_map, unsigned long *frontswap_map) |
40531542e sys_swapon: separ... |
1479 1480 |
{ int i, prev; |
40531542e sys_swapon: separ... |
1481 1482 1483 1484 1485 |
if (prio >= 0) p->prio = prio; else p->prio = --least_priority; p->swap_map = swap_map; |
38b5faf4b mm: frontswap: co... |
1486 |
frontswap_map_set(p, frontswap_map); |
40531542e sys_swapon: separ... |
1487 |
p->flags |= SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
1488 |
atomic_long_add(p->pages, &nr_swap_pages); |
40531542e sys_swapon: separ... |
1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 |
total_swap_pages += p->pages; /* insert swap space into swap_list: */ prev = -1; for (i = swap_list.head; i >= 0; i = swap_info[i]->next) { if (p->prio >= swap_info[i]->prio) break; prev = i; } p->next = i; if (prev < 0) swap_list.head = swap_list.next = p->type; else swap_info[prev]->next = p->type; |
cf0cac0a0 mm: refactor rein... |
1503 1504 1505 1506 1507 1508 1509 |
} static void enable_swap_info(struct swap_info_struct *p, int prio, unsigned char *swap_map, unsigned long *frontswap_map) { spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1510 |
spin_lock(&p->lock); |
cf0cac0a0 mm: refactor rein... |
1511 |
_enable_swap_info(p, prio, swap_map, frontswap_map); |
6555bc035 mm: do not call f... |
1512 |
frontswap_init(p->type); |
ec8acf20a swap: add per-par... |
1513 |
spin_unlock(&p->lock); |
cf0cac0a0 mm: refactor rein... |
1514 1515 1516 1517 1518 1519 |
spin_unlock(&swap_lock); } static void reinsert_swap_info(struct swap_info_struct *p) { spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1520 |
spin_lock(&p->lock); |
cf0cac0a0 mm: refactor rein... |
1521 |
_enable_swap_info(p, p->prio, p->swap_map, frontswap_map_get(p)); |
ec8acf20a swap: add per-par... |
1522 |
spin_unlock(&p->lock); |
40531542e sys_swapon: separ... |
1523 1524 |
spin_unlock(&swap_lock); } |
c4ea37c26 [CVE-2009-0029] S... |
1525 |
SYSCALL_DEFINE1(swapoff, const char __user *, specialfile) |
1da177e4c Linux-2.6.12-rc2 |
1526 |
{ |
73c34b6ac swap_info: miscel... |
1527 |
struct swap_info_struct *p = NULL; |
8d69aaee8 swap_info: swap_m... |
1528 |
unsigned char *swap_map; |
1da177e4c Linux-2.6.12-rc2 |
1529 1530 1531 |
struct file *swap_file, *victim; struct address_space *mapping; struct inode *inode; |
91a27b2a7 vfs: define struc... |
1532 |
struct filename *pathname; |
1da177e4c Linux-2.6.12-rc2 |
1533 1534 |
int i, type, prev; int err; |
886bb7e9c swapfile: remove ... |
1535 |
|
1da177e4c Linux-2.6.12-rc2 |
1536 1537 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; |
191c54244 mm: collapse secu... |
1538 |
BUG_ON(!current->mm); |
1da177e4c Linux-2.6.12-rc2 |
1539 |
pathname = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
1540 |
if (IS_ERR(pathname)) |
f58b59c1d swapfile: fix nam... |
1541 |
return PTR_ERR(pathname); |
1da177e4c Linux-2.6.12-rc2 |
1542 |
|
669abf4e5 vfs: make path_op... |
1543 |
victim = file_open_name(pathname, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
1544 1545 1546 1547 1548 1549 |
err = PTR_ERR(victim); if (IS_ERR(victim)) goto out; mapping = victim->f_mapping; prev = -1; |
5d337b919 [PATCH] swap: swa... |
1550 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
1551 1552 |
for (type = swap_list.head; type >= 0; type = swap_info[type]->next) { p = swap_info[type]; |
22c6f8fdb swapfile: remove ... |
1553 |
if (p->flags & SWP_WRITEOK) { |
1da177e4c Linux-2.6.12-rc2 |
1554 1555 1556 1557 1558 1559 1560 |
if (p->swap_file->f_mapping == mapping) break; } prev = type; } if (type < 0) { err = -EINVAL; |
5d337b919 [PATCH] swap: swa... |
1561 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
1562 1563 |
goto out_dput; } |
191c54244 mm: collapse secu... |
1564 |
if (!security_vm_enough_memory_mm(current->mm, p->pages)) |
1da177e4c Linux-2.6.12-rc2 |
1565 1566 1567 |
vm_unacct_memory(p->pages); else { err = -ENOMEM; |
5d337b919 [PATCH] swap: swa... |
1568 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
1569 1570 |
goto out_dput; } |
efa90a981 swap_info: change... |
1571 |
if (prev < 0) |
1da177e4c Linux-2.6.12-rc2 |
1572 |
swap_list.head = p->next; |
efa90a981 swap_info: change... |
1573 1574 |
else swap_info[prev]->next = p->next; |
1da177e4c Linux-2.6.12-rc2 |
1575 1576 1577 1578 |
if (type == swap_list.next) { /* just pick something that's safe... */ swap_list.next = swap_list.head; } |
ec8acf20a swap: add per-par... |
1579 |
spin_lock(&p->lock); |
78ecba081 mm: fix ever-decr... |
1580 |
if (p->prio < 0) { |
efa90a981 swap_info: change... |
1581 1582 |
for (i = p->next; i >= 0; i = swap_info[i]->next) swap_info[i]->prio = p->prio--; |
78ecba081 mm: fix ever-decr... |
1583 1584 |
least_priority++; } |
ec8acf20a swap: add per-par... |
1585 |
atomic_long_sub(p->pages, &nr_swap_pages); |
1da177e4c Linux-2.6.12-rc2 |
1586 1587 |
total_swap_pages -= p->pages; p->flags &= ~SWP_WRITEOK; |
ec8acf20a swap: add per-par... |
1588 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1589 |
spin_unlock(&swap_lock); |
fb4f88dca [PATCH] swap: get... |
1590 |
|
e1e12d2f3 mm, oom: fix race... |
1591 |
set_current_oom_origin(); |
38b5faf4b mm: frontswap: co... |
1592 |
err = try_to_unuse(type, false, 0); /* force all pages to be unused */ |
e1e12d2f3 mm, oom: fix race... |
1593 |
clear_current_oom_origin(); |
1da177e4c Linux-2.6.12-rc2 |
1594 |
|
1da177e4c Linux-2.6.12-rc2 |
1595 1596 |
if (err) { /* re-insert swap space back into swap_list */ |
cf0cac0a0 mm: refactor rein... |
1597 |
reinsert_swap_info(p); |
1da177e4c Linux-2.6.12-rc2 |
1598 1599 |
goto out_dput; } |
52b7efdbe [PATCH] swap: sca... |
1600 |
|
5d337b919 [PATCH] swap: swa... |
1601 |
destroy_swap_extents(p); |
570a335b8 swap_info: swap c... |
1602 1603 |
if (p->flags & SWP_CONTINUED) free_swap_count_continuations(p); |
fc0abb145 [PATCH] sem2mutex... |
1604 |
mutex_lock(&swapon_mutex); |
5d337b919 [PATCH] swap: swa... |
1605 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1606 |
spin_lock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1607 |
drain_mmlist(); |
52b7efdbe [PATCH] swap: sca... |
1608 |
/* wait for anyone still in scan_swap_map */ |
52b7efdbe [PATCH] swap: sca... |
1609 1610 |
p->highest_bit = 0; /* cuts scans short */ while (p->flags >= SWP_SCANNING) { |
ec8acf20a swap: add per-par... |
1611 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1612 |
spin_unlock(&swap_lock); |
13e4b57f6 [PATCH] mm: fix-u... |
1613 |
schedule_timeout_uninterruptible(1); |
5d337b919 [PATCH] swap: swa... |
1614 |
spin_lock(&swap_lock); |
ec8acf20a swap: add per-par... |
1615 |
spin_lock(&p->lock); |
52b7efdbe [PATCH] swap: sca... |
1616 |
} |
52b7efdbe [PATCH] swap: sca... |
1617 |
|
1da177e4c Linux-2.6.12-rc2 |
1618 1619 1620 1621 1622 1623 |
swap_file = p->swap_file; p->swap_file = NULL; p->max = 0; swap_map = p->swap_map; p->swap_map = NULL; p->flags = 0; |
38b5faf4b mm: frontswap: co... |
1624 |
frontswap_invalidate_area(type); |
ec8acf20a swap: add per-par... |
1625 |
spin_unlock(&p->lock); |
5d337b919 [PATCH] swap: swa... |
1626 |
spin_unlock(&swap_lock); |
fc0abb145 [PATCH] sem2mutex... |
1627 |
mutex_unlock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1628 |
vfree(swap_map); |
38b5faf4b mm: frontswap: co... |
1629 |
vfree(frontswap_map_get(p)); |
27a7faa07 memcg: swap cgrou... |
1630 1631 |
/* Destroy swap account informatin */ swap_cgroup_swapoff(type); |
1da177e4c Linux-2.6.12-rc2 |
1632 1633 1634 1635 |
inode = mapping->host; if (S_ISBLK(inode->i_mode)) { struct block_device *bdev = I_BDEV(inode); set_blocksize(bdev, p->old_block_size); |
e525fd89d block: make blkde... |
1636 |
blkdev_put(bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
1637 |
} else { |
1b1dcc1b5 [PATCH] mutex sub... |
1638 |
mutex_lock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1639 |
inode->i_flags &= ~S_SWAPFILE; |
1b1dcc1b5 [PATCH] mutex sub... |
1640 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1641 1642 1643 |
} filp_close(swap_file, NULL); err = 0; |
66d7dd518 /proc/swaps: supp... |
1644 1645 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
1da177e4c Linux-2.6.12-rc2 |
1646 1647 1648 1649 |
out_dput: filp_close(victim, NULL); out: |
f58b59c1d swapfile: fix nam... |
1650 |
putname(pathname); |
1da177e4c Linux-2.6.12-rc2 |
1651 1652 1653 1654 |
return err; } #ifdef CONFIG_PROC_FS |
66d7dd518 /proc/swaps: supp... |
1655 1656 |
static unsigned swaps_poll(struct file *file, poll_table *wait) { |
f15146380 fs: seq_file - ad... |
1657 |
struct seq_file *seq = file->private_data; |
66d7dd518 /proc/swaps: supp... |
1658 1659 |
poll_wait(file, &proc_poll_wait, wait); |
f15146380 fs: seq_file - ad... |
1660 1661 |
if (seq->poll_event != atomic_read(&proc_poll_event)) { seq->poll_event = atomic_read(&proc_poll_event); |
66d7dd518 /proc/swaps: supp... |
1662 1663 1664 1665 1666 |
return POLLIN | POLLRDNORM | POLLERR | POLLPRI; } return POLLIN | POLLRDNORM; } |
1da177e4c Linux-2.6.12-rc2 |
1667 1668 1669 |
/* iterator */ static void *swap_start(struct seq_file *swap, loff_t *pos) { |
efa90a981 swap_info: change... |
1670 1671 |
struct swap_info_struct *si; int type; |
1da177e4c Linux-2.6.12-rc2 |
1672 |
loff_t l = *pos; |
fc0abb145 [PATCH] sem2mutex... |
1673 |
mutex_lock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1674 |
|
881e4aabe [PATCH] Always pr... |
1675 1676 |
if (!l) return SEQ_START_TOKEN; |
efa90a981 swap_info: change... |
1677 1678 1679 1680 |
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 |
1681 |
continue; |
881e4aabe [PATCH] Always pr... |
1682 |
if (!--l) |
efa90a981 swap_info: change... |
1683 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
1684 1685 1686 1687 1688 1689 1690 |
} return NULL; } static void *swap_next(struct seq_file *swap, void *v, loff_t *pos) { |
efa90a981 swap_info: change... |
1691 1692 |
struct swap_info_struct *si = v; int type; |
1da177e4c Linux-2.6.12-rc2 |
1693 |
|
881e4aabe [PATCH] Always pr... |
1694 |
if (v == SEQ_START_TOKEN) |
efa90a981 swap_info: change... |
1695 1696 1697 |
type = 0; else type = si->type + 1; |
881e4aabe [PATCH] Always pr... |
1698 |
|
efa90a981 swap_info: change... |
1699 1700 1701 1702 |
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 |
1703 1704 |
continue; ++*pos; |
efa90a981 swap_info: change... |
1705 |
return si; |
1da177e4c Linux-2.6.12-rc2 |
1706 1707 1708 1709 1710 1711 1712 |
} return NULL; } static void swap_stop(struct seq_file *swap, void *v) { |
fc0abb145 [PATCH] sem2mutex... |
1713 |
mutex_unlock(&swapon_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1714 1715 1716 1717 |
} static int swap_show(struct seq_file *swap, void *v) { |
efa90a981 swap_info: change... |
1718 |
struct swap_info_struct *si = v; |
1da177e4c Linux-2.6.12-rc2 |
1719 1720 |
struct file *file; int len; |
efa90a981 swap_info: change... |
1721 |
if (si == SEQ_START_TOKEN) { |
881e4aabe [PATCH] Always pr... |
1722 1723 1724 1725 |
seq_puts(swap,"Filename\t\t\t\tType\t\tSize\tUsed\tPriority "); return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1726 |
|
efa90a981 swap_info: change... |
1727 |
file = si->swap_file; |
c32c2f63a d_path: Make seq_... |
1728 1729 |
len = seq_path(swap, &file->f_path, " \t \\"); |
6eb396dc4 [PATCH] swap: swa... |
1730 1731 |
seq_printf(swap, "%*s%s\t%u\t%u\t%d ", |
886bb7e9c swapfile: remove ... |
1732 |
len < 40 ? 40 - len : 1, " ", |
496ad9aa8 new helper: file_... |
1733 |
S_ISBLK(file_inode(file)->i_mode) ? |
1da177e4c Linux-2.6.12-rc2 |
1734 |
"partition" : "file\t", |
efa90a981 swap_info: change... |
1735 1736 1737 |
si->pages << (PAGE_SHIFT - 10), si->inuse_pages << (PAGE_SHIFT - 10), si->prio); |
1da177e4c Linux-2.6.12-rc2 |
1738 1739 |
return 0; } |
15ad7cdcf [PATCH] struct se... |
1740 |
static const struct seq_operations swaps_op = { |
1da177e4c Linux-2.6.12-rc2 |
1741 1742 1743 1744 1745 1746 1747 1748 |
.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... |
1749 |
struct seq_file *seq; |
66d7dd518 /proc/swaps: supp... |
1750 |
int ret; |
66d7dd518 /proc/swaps: supp... |
1751 |
ret = seq_open(file, &swaps_op); |
f15146380 fs: seq_file - ad... |
1752 |
if (ret) |
66d7dd518 /proc/swaps: supp... |
1753 |
return ret; |
66d7dd518 /proc/swaps: supp... |
1754 |
|
f15146380 fs: seq_file - ad... |
1755 1756 1757 |
seq = file->private_data; seq->poll_event = atomic_read(&proc_poll_event); return 0; |
1da177e4c Linux-2.6.12-rc2 |
1758 |
} |
15ad7cdcf [PATCH] struct se... |
1759 |
static const struct file_operations proc_swaps_operations = { |
1da177e4c Linux-2.6.12-rc2 |
1760 1761 1762 1763 |
.open = swaps_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, |
66d7dd518 /proc/swaps: supp... |
1764 |
.poll = swaps_poll, |
1da177e4c Linux-2.6.12-rc2 |
1765 1766 1767 1768 |
}; static int __init procswaps_init(void) { |
3d71f86f4 mm: use non-racy ... |
1769 |
proc_create("swaps", 0, NULL, &proc_swaps_operations); |
1da177e4c Linux-2.6.12-rc2 |
1770 1771 1772 1773 |
return 0; } __initcall(procswaps_init); #endif /* CONFIG_PROC_FS */ |
1796316a8 x86: consolidate ... |
1774 1775 1776 1777 1778 1779 1780 1781 |
#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... |
1782 |
static struct swap_info_struct *alloc_swap_info(void) |
1da177e4c Linux-2.6.12-rc2 |
1783 |
{ |
73c34b6ac swap_info: miscel... |
1784 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
1785 |
unsigned int type; |
efa90a981 swap_info: change... |
1786 1787 1788 |
p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) |
53cbb2435 sys_swapon: separ... |
1789 |
return ERR_PTR(-ENOMEM); |
efa90a981 swap_info: change... |
1790 |
|
5d337b919 [PATCH] swap: swa... |
1791 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
1792 1793 |
for (type = 0; type < nr_swapfiles; type++) { if (!(swap_info[type]->flags & SWP_USED)) |
1da177e4c Linux-2.6.12-rc2 |
1794 |
break; |
efa90a981 swap_info: change... |
1795 |
} |
0697212a4 [PATCH] Swapless ... |
1796 |
if (type >= MAX_SWAPFILES) { |
5d337b919 [PATCH] swap: swa... |
1797 |
spin_unlock(&swap_lock); |
efa90a981 swap_info: change... |
1798 |
kfree(p); |
730c0581c sys_swapon: simpl... |
1799 |
return ERR_PTR(-EPERM); |
1da177e4c Linux-2.6.12-rc2 |
1800 |
} |
efa90a981 swap_info: change... |
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 |
if (type >= nr_swapfiles) { p->type = type; swap_info[type] = p; /* * Write swap_info[type] before nr_swapfiles, in case a * racing procfs swap_start() or swap_next() is reading them. * (We never shrink nr_swapfiles, we never free this entry.) */ smp_wmb(); nr_swapfiles++; } else { kfree(p); p = swap_info[type]; /* * Do not memset this entry: a racing procfs swap_next() * would be relying on p->type to remain valid. */ } |
9625a5f28 swap_info: includ... |
1819 |
INIT_LIST_HEAD(&p->first_swap_extent.list); |
1da177e4c Linux-2.6.12-rc2 |
1820 |
p->flags = SWP_USED; |
1da177e4c Linux-2.6.12-rc2 |
1821 |
p->next = -1; |
5d337b919 [PATCH] swap: swa... |
1822 |
spin_unlock(&swap_lock); |
ec8acf20a swap: add per-par... |
1823 |
spin_lock_init(&p->lock); |
efa90a981 swap_info: change... |
1824 |
|
53cbb2435 sys_swapon: separ... |
1825 |
return p; |
53cbb2435 sys_swapon: separ... |
1826 |
} |
4d0e1e107 sys_swapon: separ... |
1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 |
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, FMODE_READ | FMODE_WRITE | FMODE_EXCL, sys_swapon); if (error < 0) { p->bdev = NULL; |
87ade72a7 sys_swapon: simpl... |
1838 |
return -EINVAL; |
4d0e1e107 sys_swapon: separ... |
1839 1840 1841 1842 |
} p->old_block_size = block_size(p->bdev); error = set_blocksize(p->bdev, PAGE_SIZE); if (error < 0) |
87ade72a7 sys_swapon: simpl... |
1843 |
return error; |
4d0e1e107 sys_swapon: separ... |
1844 1845 1846 1847 |
p->flags |= SWP_BLKDEV; } else if (S_ISREG(inode->i_mode)) { p->bdev = inode->i_sb->s_bdev; mutex_lock(&inode->i_mutex); |
87ade72a7 sys_swapon: simpl... |
1848 1849 1850 1851 |
if (IS_SWAPFILE(inode)) return -EBUSY; } else return -EINVAL; |
4d0e1e107 sys_swapon: separ... |
1852 1853 |
return 0; |
4d0e1e107 sys_swapon: separ... |
1854 |
} |
ca8bd38bf sys_swapon: separ... |
1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 |
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; if (memcmp("SWAPSPACE2", swap_header->magic.magic, 10)) { printk(KERN_ERR "Unable to find swap-space signature "); |
387190253 sys_swapon: simpl... |
1866 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 |
} /* swap partition endianess hack... */ if (swab32(swap_header->info.version) == 1) { swab32s(&swap_header->info.version); swab32s(&swap_header->info.last_page); swab32s(&swap_header->info.nr_badpages); for (i = 0; i < swap_header->info.nr_badpages; i++) swab32s(&swap_header->info.badpages[i]); } /* Check the swap header's sub-version */ if (swap_header->info.version != 1) { printk(KERN_WARNING "Unable to handle swap header version %d ", swap_header->info.version); |
387190253 sys_swapon: simpl... |
1883 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1884 1885 1886 1887 1888 1889 1890 1891 |
} p->lowest_bit = 1; p->cluster_next = 1; p->cluster_nr = 0; /* * Find out how many pages are allowed for a single swap |
9b15b817f swap: fix shmem s... |
1892 |
* device. There are two limiting factors: 1) the number |
a2c16d6cb mm: let swap use ... |
1893 1894 |
* of bits for the swap offset in the swp_entry_t type, and * 2) the number of bits in the swap pte as defined by the |
9b15b817f swap: fix shmem s... |
1895 |
* different architectures. In order to find the |
a2c16d6cb mm: let swap use ... |
1896 |
* largest possible bit mask, a swap entry with swap type 0 |
ca8bd38bf sys_swapon: separ... |
1897 |
* and swap offset ~0UL is created, encoded to a swap pte, |
a2c16d6cb mm: let swap use ... |
1898 |
* decoded to a swp_entry_t again, and finally the swap |
ca8bd38bf sys_swapon: separ... |
1899 1900 1901 |
* offset is extracted. This will mask all the bits from * the initial ~0UL mask that can't be encoded in either * the swp_entry_t or the architecture definition of a |
9b15b817f swap: fix shmem s... |
1902 |
* swap pte. |
ca8bd38bf sys_swapon: separ... |
1903 1904 |
*/ maxpages = swp_offset(pte_to_swp_entry( |
9b15b817f swap: fix shmem s... |
1905 |
swp_entry_to_pte(swp_entry(0, ~0UL)))) + 1; |
ca8bd38bf sys_swapon: separ... |
1906 1907 1908 1909 1910 1911 1912 1913 1914 |
if (maxpages > swap_header->info.last_page) { maxpages = swap_header->info.last_page + 1; /* 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... |
1915 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1916 1917 1918 1919 1920 |
swapfilepages = i_size_read(inode) >> PAGE_SHIFT; if (swapfilepages && maxpages > swapfilepages) { printk(KERN_WARNING "Swap area shorter than signature indicates "); |
387190253 sys_swapon: simpl... |
1921 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1922 1923 |
} if (swap_header->info.nr_badpages && S_ISREG(inode->i_mode)) |
387190253 sys_swapon: simpl... |
1924 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1925 |
if (swap_header->info.nr_badpages > MAX_SWAP_BADPAGES) |
387190253 sys_swapon: simpl... |
1926 |
return 0; |
ca8bd38bf sys_swapon: separ... |
1927 1928 |
return maxpages; |
ca8bd38bf sys_swapon: separ... |
1929 |
} |
915d4d7bc sys_swapon: separ... |
1930 1931 1932 1933 1934 1935 1936 |
static int setup_swap_map_and_extents(struct swap_info_struct *p, union swap_header *swap_header, unsigned char *swap_map, unsigned long maxpages, sector_t *span) { int i; |
915d4d7bc sys_swapon: separ... |
1937 1938 1939 1940 1941 1942 1943 |
unsigned int nr_good_pages; int nr_extents; nr_good_pages = maxpages - 1; /* omit header page */ for (i = 0; i < swap_header->info.nr_badpages; i++) { unsigned int page_nr = swap_header->info.badpages[i]; |
bdb8e3f68 sys_swapon: simpl... |
1944 1945 |
if (page_nr == 0 || page_nr > swap_header->info.last_page) return -EINVAL; |
915d4d7bc sys_swapon: separ... |
1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 |
if (page_nr < maxpages) { swap_map[page_nr] = SWAP_MAP_BAD; nr_good_pages--; } } if (nr_good_pages) { swap_map[0] = SWAP_MAP_BAD; p->max = maxpages; p->pages = nr_good_pages; nr_extents = setup_swap_extents(p, span); |
bdb8e3f68 sys_swapon: simpl... |
1957 1958 |
if (nr_extents < 0) return nr_extents; |
915d4d7bc sys_swapon: separ... |
1959 1960 1961 1962 1963 |
nr_good_pages = p->pages; } if (!nr_good_pages) { printk(KERN_WARNING "Empty swap-file "); |
bdb8e3f68 sys_swapon: simpl... |
1964 |
return -EINVAL; |
915d4d7bc sys_swapon: separ... |
1965 1966 1967 |
} return nr_extents; |
915d4d7bc sys_swapon: separ... |
1968 |
} |
53cbb2435 sys_swapon: separ... |
1969 1970 1971 |
SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) { struct swap_info_struct *p; |
91a27b2a7 vfs: define struc... |
1972 |
struct filename *name; |
53cbb2435 sys_swapon: separ... |
1973 1974 |
struct file *swap_file = NULL; struct address_space *mapping; |
40531542e sys_swapon: separ... |
1975 1976 |
int i; int prio; |
53cbb2435 sys_swapon: separ... |
1977 1978 |
int error; union swap_header *swap_header; |
915d4d7bc sys_swapon: separ... |
1979 |
int nr_extents; |
53cbb2435 sys_swapon: separ... |
1980 1981 |
sector_t span; unsigned long maxpages; |
53cbb2435 sys_swapon: separ... |
1982 |
unsigned char *swap_map = NULL; |
38b5faf4b mm: frontswap: co... |
1983 |
unsigned long *frontswap_map = NULL; |
53cbb2435 sys_swapon: separ... |
1984 1985 |
struct page *page = NULL; struct inode *inode = NULL; |
53cbb2435 sys_swapon: separ... |
1986 |
|
d15cab975 swapon: check val... |
1987 1988 |
if (swap_flags & ~SWAP_FLAGS_VALID) return -EINVAL; |
53cbb2435 sys_swapon: separ... |
1989 1990 1991 1992 |
if (!capable(CAP_SYS_ADMIN)) return -EPERM; p = alloc_swap_info(); |
2542e5134 sys_swapon: simpl... |
1993 1994 |
if (IS_ERR(p)) return PTR_ERR(p); |
53cbb2435 sys_swapon: separ... |
1995 |
|
1da177e4c Linux-2.6.12-rc2 |
1996 |
name = getname(specialfile); |
1da177e4c Linux-2.6.12-rc2 |
1997 |
if (IS_ERR(name)) { |
7de7fb6b3 sys_swapon: move ... |
1998 |
error = PTR_ERR(name); |
1da177e4c Linux-2.6.12-rc2 |
1999 |
name = NULL; |
bd69010b0 sys_swapon: use a... |
2000 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2001 |
} |
669abf4e5 vfs: make path_op... |
2002 |
swap_file = file_open_name(name, O_RDWR|O_LARGEFILE, 0); |
1da177e4c Linux-2.6.12-rc2 |
2003 |
if (IS_ERR(swap_file)) { |
7de7fb6b3 sys_swapon: move ... |
2004 |
error = PTR_ERR(swap_file); |
1da177e4c Linux-2.6.12-rc2 |
2005 |
swap_file = NULL; |
bd69010b0 sys_swapon: use a... |
2006 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2007 2008 2009 2010 |
} p->swap_file = swap_file; mapping = swap_file->f_mapping; |
1da177e4c Linux-2.6.12-rc2 |
2011 |
|
1da177e4c Linux-2.6.12-rc2 |
2012 |
for (i = 0; i < nr_swapfiles; i++) { |
efa90a981 swap_info: change... |
2013 |
struct swap_info_struct *q = swap_info[i]; |
1da177e4c Linux-2.6.12-rc2 |
2014 |
|
e8e6c2ec4 sys_swapon: do no... |
2015 |
if (q == p || !q->swap_file) |
1da177e4c Linux-2.6.12-rc2 |
2016 |
continue; |
7de7fb6b3 sys_swapon: move ... |
2017 2018 |
if (mapping == q->swap_file->f_mapping) { error = -EBUSY; |
1da177e4c Linux-2.6.12-rc2 |
2019 |
goto bad_swap; |
7de7fb6b3 sys_swapon: move ... |
2020 |
} |
1da177e4c Linux-2.6.12-rc2 |
2021 |
} |
2130781e2 sys_swapon: fix i... |
2022 2023 |
inode = mapping->host; /* If S_ISREG(inode->i_mode) will do mutex_lock(&inode->i_mutex); */ |
4d0e1e107 sys_swapon: separ... |
2024 2025 |
error = claim_swapfile(p, inode); if (unlikely(error)) |
1da177e4c Linux-2.6.12-rc2 |
2026 |
goto bad_swap; |
1da177e4c Linux-2.6.12-rc2 |
2027 |
|
1da177e4c Linux-2.6.12-rc2 |
2028 2029 2030 2031 2032 2033 2034 |
/* * Read the swap header. */ if (!mapping->a_ops->readpage) { error = -EINVAL; goto bad_swap; } |
090d2b185 [PATCH] read_mapp... |
2035 |
page = read_mapping_page(mapping, 0, swap_file); |
1da177e4c Linux-2.6.12-rc2 |
2036 2037 2038 2039 |
if (IS_ERR(page)) { error = PTR_ERR(page); goto bad_swap; } |
81e339712 swapfile: remove ... |
2040 |
swap_header = kmap(page); |
1da177e4c Linux-2.6.12-rc2 |
2041 |
|
ca8bd38bf sys_swapon: separ... |
2042 2043 |
maxpages = read_swap_header(p, swap_header, inode); if (unlikely(!maxpages)) { |
1da177e4c Linux-2.6.12-rc2 |
2044 2045 2046 |
error = -EINVAL; goto bad_swap; } |
886bb7e9c swapfile: remove ... |
2047 |
|
81e339712 swapfile: remove ... |
2048 |
/* OK, set up the swap map and apply the bad block list */ |
803d0c835 sys_swapon: use v... |
2049 |
swap_map = vzalloc(maxpages); |
81e339712 swapfile: remove ... |
2050 2051 2052 2053 |
if (!swap_map) { error = -ENOMEM; goto bad_swap; } |
1da177e4c Linux-2.6.12-rc2 |
2054 |
|
1421ef3cd sys_swapon: call ... |
2055 2056 2057 |
error = swap_cgroup_swapon(p->type, maxpages); if (error) goto bad_swap; |
915d4d7bc sys_swapon: separ... |
2058 2059 2060 2061 |
nr_extents = setup_swap_map_and_extents(p, swap_header, swap_map, maxpages, &span); if (unlikely(nr_extents < 0)) { error = nr_extents; |
1da177e4c Linux-2.6.12-rc2 |
2062 2063 |
goto bad_swap; } |
38b5faf4b mm: frontswap: co... |
2064 2065 2066 |
/* frontswap enabled? set up bit-per-page map for frontswap */ if (frontswap_enabled) frontswap_map = vzalloc(maxpages / sizeof(long)); |
1da177e4c Linux-2.6.12-rc2 |
2067 |
|
3bd0f0c76 swapfile: avoid N... |
2068 2069 2070 2071 2072 |
if (p->bdev) { if (blk_queue_nonrot(bdev_get_queue(p->bdev))) { p->flags |= SWP_SOLIDSTATE; p->cluster_next = 1 + (random32() % p->highest_bit); } |
052b1987f swap: don't do di... |
2073 |
if ((swap_flags & SWAP_FLAG_DISCARD) && discard_swap(p) == 0) |
3bd0f0c76 swapfile: avoid N... |
2074 |
p->flags |= SWP_DISCARDABLE; |
20137a490 swapfile: swapon ... |
2075 |
} |
6a6ba8317 swapfile: swapon ... |
2076 |
|
fc0abb145 [PATCH] sem2mutex... |
2077 |
mutex_lock(&swapon_mutex); |
40531542e sys_swapon: separ... |
2078 |
prio = -1; |
78ecba081 mm: fix ever-decr... |
2079 |
if (swap_flags & SWAP_FLAG_PREFER) |
40531542e sys_swapon: separ... |
2080 |
prio = |
78ecba081 mm: fix ever-decr... |
2081 |
(swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT; |
38b5faf4b mm: frontswap: co... |
2082 |
enable_swap_info(p, prio, swap_map, frontswap_map); |
c69dbfb84 sys_swapon: move ... |
2083 2084 |
printk(KERN_INFO "Adding %uk swap on %s. " |
38b5faf4b mm: frontswap: co... |
2085 2086 |
"Priority:%d extents:%d across:%lluk %s%s%s ", |
91a27b2a7 vfs: define struc... |
2087 |
p->pages<<(PAGE_SHIFT-10), name->name, p->prio, |
c69dbfb84 sys_swapon: move ... |
2088 2089 |
nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10), (p->flags & SWP_SOLIDSTATE) ? "SS" : "", |
38b5faf4b mm: frontswap: co... |
2090 2091 |
(p->flags & SWP_DISCARDABLE) ? "D" : "", (frontswap_map) ? "FS" : ""); |
c69dbfb84 sys_swapon: move ... |
2092 |
|
fc0abb145 [PATCH] sem2mutex... |
2093 |
mutex_unlock(&swapon_mutex); |
66d7dd518 /proc/swaps: supp... |
2094 2095 |
atomic_inc(&proc_poll_event); wake_up_interruptible(&proc_poll_wait); |
9b01c350a sys_swapon: do on... |
2096 2097 |
if (S_ISREG(inode->i_mode)) inode->i_flags |= S_SWAPFILE; |
1da177e4c Linux-2.6.12-rc2 |
2098 2099 2100 |
error = 0; goto out; bad_swap: |
bd69010b0 sys_swapon: use a... |
2101 |
if (inode && S_ISBLK(inode->i_mode) && p->bdev) { |
f2090d2df sys_swapon: remov... |
2102 2103 |
set_blocksize(p->bdev, p->old_block_size); blkdev_put(p->bdev, FMODE_READ | FMODE_WRITE | FMODE_EXCL); |
1da177e4c Linux-2.6.12-rc2 |
2104 |
} |
4cd3bb10f [PATCH] swap: mov... |
2105 |
destroy_swap_extents(p); |
e8e6c2ec4 sys_swapon: do no... |
2106 |
swap_cgroup_swapoff(p->type); |
5d337b919 [PATCH] swap: swa... |
2107 |
spin_lock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2108 |
p->swap_file = NULL; |
1da177e4c Linux-2.6.12-rc2 |
2109 |
p->flags = 0; |
5d337b919 [PATCH] swap: swa... |
2110 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2111 |
vfree(swap_map); |
52c50567d mm: swap: unlock ... |
2112 |
if (swap_file) { |
2130781e2 sys_swapon: fix i... |
2113 |
if (inode && S_ISREG(inode->i_mode)) { |
52c50567d mm: swap: unlock ... |
2114 |
mutex_unlock(&inode->i_mutex); |
2130781e2 sys_swapon: fix i... |
2115 2116 |
inode = NULL; } |
1da177e4c Linux-2.6.12-rc2 |
2117 |
filp_close(swap_file, NULL); |
52c50567d mm: swap: unlock ... |
2118 |
} |
1da177e4c Linux-2.6.12-rc2 |
2119 2120 2121 2122 2123 2124 2125 |
out: if (page && !IS_ERR(page)) { kunmap(page); page_cache_release(page); } if (name) putname(name); |
9b01c350a sys_swapon: do on... |
2126 |
if (inode && S_ISREG(inode->i_mode)) |
1b1dcc1b5 [PATCH] mutex sub... |
2127 |
mutex_unlock(&inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2128 2129 2130 2131 2132 |
return error; } void si_swapinfo(struct sysinfo *val) { |
efa90a981 swap_info: change... |
2133 |
unsigned int type; |
1da177e4c Linux-2.6.12-rc2 |
2134 |
unsigned long nr_to_be_unused = 0; |
5d337b919 [PATCH] swap: swa... |
2135 |
spin_lock(&swap_lock); |
efa90a981 swap_info: change... |
2136 2137 2138 2139 2140 |
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 |
2141 |
} |
ec8acf20a swap: add per-par... |
2142 |
val->freeswap = atomic_long_read(&nr_swap_pages) + nr_to_be_unused; |
1da177e4c Linux-2.6.12-rc2 |
2143 |
val->totalswap = total_swap_pages + nr_to_be_unused; |
5d337b919 [PATCH] swap: swa... |
2144 |
spin_unlock(&swap_lock); |
1da177e4c Linux-2.6.12-rc2 |
2145 2146 2147 2148 2149 |
} /* * Verify that a swap entry is valid and increment its swap map count. * |
355cfa73d mm: modify swap_m... |
2150 2151 2152 2153 2154 2155 |
* 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... |
2156 |
* - swap-mapped reference requested but needs continued swap count. -> ENOMEM |
1da177e4c Linux-2.6.12-rc2 |
2157 |
*/ |
8d69aaee8 swap_info: swap_m... |
2158 |
static int __swap_duplicate(swp_entry_t entry, unsigned char usage) |
1da177e4c Linux-2.6.12-rc2 |
2159 |
{ |
73c34b6ac swap_info: miscel... |
2160 |
struct swap_info_struct *p; |
1da177e4c Linux-2.6.12-rc2 |
2161 |
unsigned long offset, type; |
8d69aaee8 swap_info: swap_m... |
2162 2163 |
unsigned char count; unsigned char has_cache; |
253d553ba swap_info: SWAP_H... |
2164 |
int err = -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
2165 |
|
a7420aa54 HWPOISON: Add sup... |
2166 |
if (non_swap_entry(entry)) |
253d553ba swap_info: SWAP_H... |
2167 |
goto out; |
0697212a4 [PATCH] Swapless ... |
2168 |
|
1da177e4c Linux-2.6.12-rc2 |
2169 2170 2171 |
type = swp_type(entry); if (type >= nr_swapfiles) goto bad_file; |
efa90a981 swap_info: change... |
2172 |
p = swap_info[type]; |
1da177e4c Linux-2.6.12-rc2 |
2173 |
offset = swp_offset(entry); |
ec8acf20a swap: add per-par... |
2174 |
spin_lock(&p->lock); |
355cfa73d mm: modify swap_m... |
2175 2176 |
if (unlikely(offset >= p->max)) goto unlock_out; |
253d553ba swap_info: SWAP_H... |
2177 2178 2179 2180 |
count = p->swap_map[offset]; has_cache = count & SWAP_HAS_CACHE; count &= ~SWAP_HAS_CACHE; err = 0; |
355cfa73d mm: modify swap_m... |
2181 |
|
253d553ba swap_info: SWAP_H... |
2182 |
if (usage == SWAP_HAS_CACHE) { |
355cfa73d mm: modify swap_m... |
2183 2184 |
/* set SWAP_HAS_CACHE if there is no cache and entry is used */ |
253d553ba swap_info: SWAP_H... |
2185 2186 2187 2188 2189 2190 |
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... |
2191 2192 |
} else if (count || has_cache) { |
253d553ba swap_info: SWAP_H... |
2193 |
|
570a335b8 swap_info: swap c... |
2194 2195 2196 |
if ((count & ~COUNT_CONTINUED) < SWAP_MAP_MAX) count += usage; else if ((count & ~COUNT_CONTINUED) > SWAP_MAP_MAX) |
253d553ba swap_info: SWAP_H... |
2197 |
err = -EINVAL; |
570a335b8 swap_info: swap c... |
2198 2199 2200 2201 |
else if (swap_count_continued(p, offset, count)) count = COUNT_CONTINUED; else err = -ENOMEM; |
355cfa73d mm: modify swap_m... |
2202 |
} else |
253d553ba swap_info: SWAP_H... |
2203 2204 2205 |
err = -ENOENT; /* unused swap entry */ p->swap_map[offset] = count | has_cache; |
355cfa73d mm: modify swap_m... |
2206 |
unlock_out: |
ec8acf20a swap: add per-par... |
2207 |
spin_unlock(&p->lock); |
1da177e4c Linux-2.6.12-rc2 |
2208 |
out: |
253d553ba swap_info: SWAP_H... |
2209 |
return err; |
1da177e4c Linux-2.6.12-rc2 |
2210 2211 2212 2213 2214 2215 |
bad_file: printk(KERN_ERR "swap_dup: %s%08lx ", Bad_file, entry.val); goto out; } |
253d553ba swap_info: SWAP_H... |
2216 |
|
355cfa73d mm: modify swap_m... |
2217 |
/* |
aaa468653 swap_info: note S... |
2218 2219 2220 2221 2222 2223 2224 2225 2226 |
* 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... |
2227 2228 2229 2230 2231 |
* 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... |
2232 |
*/ |
570a335b8 swap_info: swap c... |
2233 |
int swap_duplicate(swp_entry_t entry) |
355cfa73d mm: modify swap_m... |
2234 |
{ |
570a335b8 swap_info: swap c... |
2235 2236 2237 2238 2239 |
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... |
2240 |
} |
1da177e4c Linux-2.6.12-rc2 |
2241 |
|
cb4b86ba4 mm: add swap cach... |
2242 |
/* |
355cfa73d mm: modify swap_m... |
2243 2244 |
* @entry: swap entry for which we allocate swap cache. * |
73c34b6ac swap_info: miscel... |
2245 |
* Called when allocating swap cache for existing swap entry, |
355cfa73d mm: modify swap_m... |
2246 2247 2248 |
* 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... |
2249 2250 2251 |
*/ int swapcache_prepare(swp_entry_t entry) { |
253d553ba swap_info: SWAP_H... |
2252 |
return __swap_duplicate(entry, SWAP_HAS_CACHE); |
cb4b86ba4 mm: add swap cach... |
2253 |
} |
f981c5950 mm: methods for t... |
2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 |
struct swap_info_struct *page_swap_info(struct page *page) { swp_entry_t swap = { .val = page_private(page) }; BUG_ON(!PageSwapCache(page)); return swap_info[swp_type(swap)]; } /* * out-of-line __page_file_ methods to avoid include hell. */ struct address_space *__page_file_mapping(struct page *page) { VM_BUG_ON(!PageSwapCache(page)); 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) }; VM_BUG_ON(!PageSwapCache(page)); return swp_offset(swap); } EXPORT_SYMBOL_GPL(__page_file_index); |
1da177e4c Linux-2.6.12-rc2 |
2278 |
/* |
570a335b8 swap_info: swap c... |
2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 |
* add_swap_count_continuation - called when a swap count is duplicated * beyond SWAP_MAP_MAX, it allocates a new page and links that to the entry's * page of the original vmalloc'ed swap_map, to hold the continuation count * (for that entry and for its neighbouring PAGE_SIZE swap entries). Called * again when count is duplicated beyond SWAP_MAP_MAX * SWAP_CONT_MAX, etc. * * These continuation pages are seldom referenced: the common paths all work * on the original swap_map, only referring to a continuation page when the * low "digit" of a count is incremented or decremented through SWAP_MAP_MAX. * * add_swap_count_continuation(, GFP_ATOMIC) can be called while holding * page table locks; if it fails, add_swap_count_continuation(, GFP_KERNEL) * can be called after dropping locks. */ int add_swap_count_continuation(swp_entry_t entry, gfp_t gfp_mask) { struct swap_info_struct *si; struct page *head; struct page *page; struct page *list_page; pgoff_t offset; unsigned char count; /* * When debugging, it's easier to use __GFP_ZERO here; but it's better * for latency not to zero a page while GFP_ATOMIC and holding locks. */ page = alloc_page(gfp_mask | __GFP_HIGHMEM); si = swap_info_get(entry); if (!si) { /* * An acceptable race has occurred since the failing * __swap_duplicate(): the swap entry has been freed, * perhaps even the whole swap_map cleared for swapoff. */ goto outer; } offset = swp_offset(entry); count = si->swap_map[offset] & ~SWAP_HAS_CACHE; if ((count & ~COUNT_CONTINUED) != SWAP_MAP_MAX) { /* * The higher the swap count, the more likely it is that tasks * will race to add swap count continuation: we need to avoid * over-provisioning. */ goto out; } if (!page) { |
ec8acf20a swap: add per-par... |
2331 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 |
return -ENOMEM; } /* * We are fortunate that although vmalloc_to_page uses pte_offset_map, * no architecture is using highmem pages for kernel pagetables: so it * will not corrupt the GFP_ATOMIC caller's atomic pagetable kmaps. */ head = vmalloc_to_page(si->swap_map + offset); offset &= ~PAGE_MASK; /* * Page allocation does not initialize the page's lru field, * but it does always reset its private field. */ if (!page_private(head)) { BUG_ON(count & COUNT_CONTINUED); INIT_LIST_HEAD(&head->lru); set_page_private(head, SWP_CONTINUED); si->flags |= SWP_CONTINUED; } list_for_each_entry(list_page, &head->lru, lru) { unsigned char *map; /* * If the previous map said no continuation, but we've found * a continuation page, free our allocation and use this one. */ if (!(count & COUNT_CONTINUED)) goto out; |
9b04c5fec mm: remove the se... |
2363 |
map = kmap_atomic(list_page) + offset; |
570a335b8 swap_info: swap c... |
2364 |
count = *map; |
9b04c5fec mm: remove the se... |
2365 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 |
/* * If this continuation count now has some space in it, * free our allocation and use this one. */ if ((count & ~COUNT_CONTINUED) != SWAP_CONT_MAX) goto out; } list_add_tail(&page->lru, &head->lru); page = NULL; /* now it's attached, don't free it */ out: |
ec8acf20a swap: add per-par... |
2378 |
spin_unlock(&si->lock); |
570a335b8 swap_info: swap c... |
2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 |
outer: if (page) __free_page(page); return 0; } /* * swap_count_continued - when the original swap_map count is incremented * from SWAP_MAP_MAX, check if there is already a continuation page to carry * into, carry if so, or else fail until a new continuation page is allocated; * when the original swap_map count is decremented from 0 with continuation, * borrow from the continuation and report whether it still holds more. * Called while __swap_duplicate() or swap_entry_free() holds swap_lock. */ static bool swap_count_continued(struct swap_info_struct *si, pgoff_t offset, unsigned char count) { struct page *head; struct page *page; unsigned char *map; head = vmalloc_to_page(si->swap_map + offset); if (page_private(head) != SWP_CONTINUED) { BUG_ON(count & COUNT_CONTINUED); return false; /* need to add count continuation */ } offset &= ~PAGE_MASK; page = list_entry(head->lru.next, struct page, lru); |
9b04c5fec mm: remove the se... |
2408 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2409 2410 2411 2412 2413 2414 2415 2416 2417 |
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... |
2418 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2419 2420 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
2421 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2422 2423 |
} if (*map == SWAP_CONT_MAX) { |
9b04c5fec mm: remove the se... |
2424 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2425 2426 2427 |
page = list_entry(page->lru.next, struct page, lru); if (page == head) return false; /* add count continuation */ |
9b04c5fec mm: remove the se... |
2428 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2429 2430 2431 |
init_map: *map = 0; /* we didn't zero the page */ } *map += 1; |
9b04c5fec mm: remove the se... |
2432 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2433 2434 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
2435 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2436 |
*map = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
2437 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 |
page = list_entry(page->lru.prev, struct page, lru); } return true; /* incremented */ } else { /* decrementing */ /* * Think of how you subtract 1 from 1000 */ BUG_ON(count != COUNT_CONTINUED); while (*map == COUNT_CONTINUED) { |
9b04c5fec mm: remove the se... |
2448 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2449 2450 |
page = list_entry(page->lru.next, struct page, lru); BUG_ON(page == head); |
9b04c5fec mm: remove the se... |
2451 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2452 2453 2454 2455 2456 |
} BUG_ON(*map == 0); *map -= 1; if (*map == 0) count = 0; |
9b04c5fec mm: remove the se... |
2457 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2458 2459 |
page = list_entry(page->lru.prev, struct page, lru); while (page != head) { |
9b04c5fec mm: remove the se... |
2460 |
map = kmap_atomic(page) + offset; |
570a335b8 swap_info: swap c... |
2461 2462 |
*map = SWAP_CONT_MAX | count; count = COUNT_CONTINUED; |
9b04c5fec mm: remove the se... |
2463 |
kunmap_atomic(map); |
570a335b8 swap_info: swap c... |
2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 |
page = list_entry(page->lru.prev, struct page, lru); } return count == COUNT_CONTINUED; } } /* * free_swap_count_continuations - swapoff free all the continuation pages * appended to the swap_map, after swap_map is quiesced, before vfree'ing it. */ static void free_swap_count_continuations(struct swap_info_struct *si) { pgoff_t offset; for (offset = 0; offset < si->max; offset += PAGE_SIZE) { struct page *head; head = vmalloc_to_page(si->swap_map + offset); if (page_private(head)) { struct list_head *this, *next; list_for_each_safe(this, next, &head->lru) { struct page *page; page = list_entry(this, struct page, lru); list_del(this); __free_page(page); } } } } |