Blame view
mm/page_io.c
10.4 KB
b24413180
|
1 |
// SPDX-License-Identifier: GPL-2.0 |
1da177e4c
|
2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
/* * linux/mm/page_io.c * * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds * * Swap reorganised 29.12.95, * Asynchronous swapping added 30.12.95. Stephen Tweedie * Removed race in async swapping. 14.4.1996. Bruno Haible * Add swap of shared pages through the page cache. 20.2.1998. Stephen Tweedie * Always use brw_page, life becomes simpler. 12 May 1998 Eric Biederman */ #include <linux/mm.h> #include <linux/kernel_stat.h> |
5a0e3ad6a
|
16 |
#include <linux/gfp.h> |
1da177e4c
|
17 18 19 20 |
#include <linux/pagemap.h> #include <linux/swap.h> #include <linux/bio.h> #include <linux/swapops.h> |
62c230bc1
|
21 |
#include <linux/buffer_head.h> |
1da177e4c
|
22 |
#include <linux/writeback.h> |
38b5faf4b
|
23 |
#include <linux/frontswap.h> |
b430e9d1c
|
24 |
#include <linux/blkdev.h> |
e2e40f2c1
|
25 |
#include <linux/uio.h> |
b0ba2d0fa
|
26 |
#include <linux/sched/task.h> |
1da177e4c
|
27 |
#include <asm/pgtable.h> |
f29ad6a99
|
28 |
static struct bio *get_swap_bio(gfp_t gfp_flags, |
1da177e4c
|
29 30 |
struct page *page, bio_end_io_t end_io) { |
225311a46
|
31 |
int i, nr = hpage_nr_pages(page); |
1da177e4c
|
32 |
struct bio *bio; |
225311a46
|
33 |
bio = bio_alloc(gfp_flags, nr); |
1da177e4c
|
34 |
if (bio) { |
74d46992e
|
35 36 37 38 |
struct block_device *bdev; bio->bi_iter.bi_sector = map_swap_page(page, &bdev); bio_set_dev(bio, bdev); |
4f024f379
|
39 |
bio->bi_iter.bi_sector <<= PAGE_SHIFT - 9; |
1da177e4c
|
40 |
bio->bi_end_io = end_io; |
6cf66b4ca
|
41 |
|
225311a46
|
42 43 44 |
for (i = 0; i < nr; i++) bio_add_page(bio, page + i, PAGE_SIZE, 0); VM_BUG_ON(bio->bi_iter.bi_size != PAGE_SIZE * nr); |
1da177e4c
|
45 46 47 |
} return bio; } |
4246a0b63
|
48 |
void end_swap_bio_write(struct bio *bio) |
1da177e4c
|
49 |
{ |
263663cd3
|
50 |
struct page *page = bio_first_page_all(bio); |
1da177e4c
|
51 |
|
4e4cbee93
|
52 |
if (bio->bi_status) { |
1da177e4c
|
53 |
SetPageError(page); |
6ddab3b9e
|
54 55 56 57 58 59 60 61 62 |
/* * We failed to write the page out to swap-space. * Re-dirty the page in order to avoid it being reclaimed. * Also print a dire warning that things will go BAD (tm) * very quickly. * * Also clear PG_reclaim to avoid rotate_reclaimable_page() */ set_page_dirty(page); |
1170532bb
|
63 64 |
pr_alert("Write-error on swap-device (%u:%u:%llu) ", |
74d46992e
|
65 |
MAJOR(bio_dev(bio)), MINOR(bio_dev(bio)), |
1170532bb
|
66 |
(unsigned long long)bio->bi_iter.bi_sector); |
6ddab3b9e
|
67 68 |
ClearPageReclaim(page); } |
1da177e4c
|
69 70 |
end_page_writeback(page); bio_put(bio); |
1da177e4c
|
71 |
} |
3f2b1a04f
|
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 |
static void swap_slot_free_notify(struct page *page) { struct swap_info_struct *sis; struct gendisk *disk; /* * There is no guarantee that the page is in swap cache - the software * suspend code (at least) uses end_swap_bio_read() against a non- * swapcache page. So we must check PG_swapcache before proceeding with * this optimization. */ if (unlikely(!PageSwapCache(page))) return; sis = page_swap_info(page); if (!(sis->flags & SWP_BLKDEV)) return; /* * The swap subsystem performs lazy swap slot freeing, * expecting that the page will be swapped out again. * So we can avoid an unnecessary write if the page * isn't redirtied. * This is good for real swap storage because we can * reduce unnecessary I/O and enhance wear-leveling * if an SSD is used as the as swap device. * But if in-memory swap device (eg zram) is used, * this causes a duplicated copy between uncompressed * data in VM-owned memory and compressed data in * zram-owned memory. So let's free zram-owned memory * and make the VM-owned decompressed page *dirty*, * so the page should be swapped out somewhere again if * we again wish to reclaim it. */ disk = sis->bdev->bd_disk; if (disk->fops->swap_slot_free_notify) { swp_entry_t entry; unsigned long offset; entry.val = page_private(page); offset = swp_offset(entry); SetPageDirty(page); disk->fops->swap_slot_free_notify(sis->bdev, offset); } } |
4246a0b63
|
119 |
static void end_swap_bio_read(struct bio *bio) |
1da177e4c
|
120 |
{ |
263663cd3
|
121 |
struct page *page = bio_first_page_all(bio); |
23955622f
|
122 |
struct task_struct *waiter = bio->bi_private; |
1da177e4c
|
123 |
|
4e4cbee93
|
124 |
if (bio->bi_status) { |
1da177e4c
|
125 126 |
SetPageError(page); ClearPageUptodate(page); |
1170532bb
|
127 128 |
pr_alert("Read-error on swap-device (%u:%u:%llu) ", |
74d46992e
|
129 |
MAJOR(bio_dev(bio)), MINOR(bio_dev(bio)), |
1170532bb
|
130 |
(unsigned long long)bio->bi_iter.bi_sector); |
b430e9d1c
|
131 |
goto out; |
1da177e4c
|
132 |
} |
b430e9d1c
|
133 134 |
SetPageUptodate(page); |
3f2b1a04f
|
135 |
swap_slot_free_notify(page); |
b430e9d1c
|
136 |
out: |
1da177e4c
|
137 |
unlock_page(page); |
23955622f
|
138 |
WRITE_ONCE(bio->bi_private, NULL); |
1da177e4c
|
139 |
bio_put(bio); |
23955622f
|
140 |
wake_up_process(waiter); |
b0ba2d0fa
|
141 |
put_task_struct(waiter); |
1da177e4c
|
142 |
} |
a509bc1a9
|
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 |
int generic_swapfile_activate(struct swap_info_struct *sis, struct file *swap_file, sector_t *span) { struct address_space *mapping = swap_file->f_mapping; struct inode *inode = mapping->host; unsigned blocks_per_page; unsigned long page_no; unsigned blkbits; sector_t probe_block; sector_t last_block; sector_t lowest_block = -1; sector_t highest_block = 0; int nr_extents = 0; int ret; blkbits = inode->i_blkbits; blocks_per_page = PAGE_SIZE >> blkbits; /* * Map all the blocks into the extent list. This code doesn't try * to be very smart. */ probe_block = 0; page_no = 0; last_block = i_size_read(inode) >> blkbits; while ((probe_block + blocks_per_page) <= last_block && page_no < sis->max) { unsigned block_in_page; sector_t first_block; |
7e4411bfe
|
173 |
cond_resched(); |
a509bc1a9
|
174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 |
first_block = bmap(inode, probe_block); if (first_block == 0) goto bad_bmap; /* * It must be PAGE_SIZE aligned on-disk */ if (first_block & (blocks_per_page - 1)) { probe_block++; goto reprobe; } for (block_in_page = 1; block_in_page < blocks_per_page; block_in_page++) { sector_t block; block = bmap(inode, probe_block + block_in_page); if (block == 0) goto bad_bmap; if (block != first_block + block_in_page) { /* Discontiguity */ probe_block++; goto reprobe; } } first_block >>= (PAGE_SHIFT - blkbits); if (page_no) { /* exclude the header page */ if (first_block < lowest_block) lowest_block = first_block; if (first_block > highest_block) highest_block = first_block; } /* * We found a PAGE_SIZE-length, PAGE_SIZE-aligned run of blocks */ ret = add_swap_extent(sis, page_no, 1, first_block); if (ret < 0) goto out; nr_extents += ret; page_no++; probe_block += blocks_per_page; reprobe: continue; } ret = nr_extents; *span = 1 + highest_block - lowest_block; if (page_no == 0) page_no = 1; /* force Empty message */ sis->max = page_no; sis->pages = page_no - 1; sis->highest_bit = page_no - 1; out: return ret; bad_bmap: |
1170532bb
|
230 231 |
pr_err("swapon: swapfile has holes "); |
a509bc1a9
|
232 233 234 |
ret = -EINVAL; goto out; } |
1da177e4c
|
235 236 237 238 239 240 |
/* * We may have stale swap cache pages in memory: notice * them here and get rid of the unnecessary final write. */ int swap_writepage(struct page *page, struct writeback_control *wbc) { |
2f772e6ca
|
241 |
int ret = 0; |
1da177e4c
|
242 |
|
a2c43eed8
|
243 |
if (try_to_free_swap(page)) { |
1da177e4c
|
244 245 246 |
unlock_page(page); goto out; } |
165c8aed5
|
247 |
if (frontswap_store(page) == 0) { |
38b5faf4b
|
248 249 250 251 252 |
set_page_writeback(page); unlock_page(page); end_page_writeback(page); goto out; } |
1eec6702a
|
253 |
ret = __swap_writepage(page, wbc, end_swap_bio_write); |
2f772e6ca
|
254 255 256 |
out: return ret; } |
dd6bd0d9c
|
257 258 |
static sector_t swap_page_sector(struct page *page) { |
09cbfeaf1
|
259 |
return (sector_t)__page_file_index(page) << (PAGE_SHIFT - 9); |
dd6bd0d9c
|
260 |
} |
225311a46
|
261 262 263 264 265 266 267 268 |
static inline void count_swpout_vm_event(struct page *page) { #ifdef CONFIG_TRANSPARENT_HUGEPAGE if (unlikely(PageTransHuge(page))) count_vm_event(THP_SWPOUT); #endif count_vm_events(PSWPOUT, hpage_nr_pages(page)); } |
1eec6702a
|
269 |
int __swap_writepage(struct page *page, struct writeback_control *wbc, |
4246a0b63
|
270 |
bio_end_io_t end_write_func) |
2f772e6ca
|
271 272 |
{ struct bio *bio; |
4e49ea4a3
|
273 |
int ret; |
2f772e6ca
|
274 |
struct swap_info_struct *sis = page_swap_info(page); |
62c230bc1
|
275 |
|
cc30c5d64
|
276 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
62c230bc1
|
277 278 279 280 |
if (sis->flags & SWP_FILE) { struct kiocb kiocb; struct file *swap_file = sis->swap_file; struct address_space *mapping = swap_file->f_mapping; |
62a8067a7
|
281 282 283 284 285 |
struct bio_vec bv = { .bv_page = page, .bv_len = PAGE_SIZE, .bv_offset = 0 }; |
05afcb77e
|
286 |
struct iov_iter from; |
62c230bc1
|
287 |
|
05afcb77e
|
288 |
iov_iter_bvec(&from, ITER_BVEC | WRITE, &bv, 1, PAGE_SIZE); |
62c230bc1
|
289 290 |
init_sync_kiocb(&kiocb, swap_file); kiocb.ki_pos = page_file_offset(page); |
62c230bc1
|
291 |
|
0cdc444a6
|
292 |
set_page_writeback(page); |
62c230bc1
|
293 |
unlock_page(page); |
c8b8e32d7
|
294 |
ret = mapping->a_ops->direct_IO(&kiocb, &from); |
62c230bc1
|
295 296 297 |
if (ret == PAGE_SIZE) { count_vm_event(PSWPOUT); ret = 0; |
2d30d31ea
|
298 |
} else { |
0cdc444a6
|
299 300 301 302 303 304 305 306 307 308 |
/* * In the case of swap-over-nfs, this can be a * temporary failure if the system has limited * memory for allocating transmit buffers. * Mark the page dirty and avoid * rotate_reclaimable_page but rate-limit the * messages but do not flag PageError like * the normal direct-to-bio case as it could * be temporary. */ |
2d30d31ea
|
309 |
set_page_dirty(page); |
0cdc444a6
|
310 |
ClearPageReclaim(page); |
1170532bb
|
311 312 313 |
pr_err_ratelimited("Write error on dio swapfile (%llu) ", page_file_offset(page)); |
62c230bc1
|
314 |
} |
0cdc444a6
|
315 |
end_page_writeback(page); |
62c230bc1
|
316 317 |
return ret; } |
dd6bd0d9c
|
318 319 |
ret = bdev_write_page(sis->bdev, swap_page_sector(page), page, wbc); if (!ret) { |
225311a46
|
320 |
count_swpout_vm_event(page); |
dd6bd0d9c
|
321 322 323 324 |
return 0; } ret = 0; |
1eec6702a
|
325 |
bio = get_swap_bio(GFP_NOIO, page, end_write_func); |
1da177e4c
|
326 327 328 329 330 331 |
if (bio == NULL) { set_page_dirty(page); unlock_page(page); ret = -ENOMEM; goto out; } |
7637241e6
|
332 |
bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc); |
225311a46
|
333 |
count_swpout_vm_event(page); |
1da177e4c
|
334 335 |
set_page_writeback(page); unlock_page(page); |
4e49ea4a3
|
336 |
submit_bio(bio); |
1da177e4c
|
337 338 339 |
out: return ret; } |
0bcac06f2
|
340 |
int swap_readpage(struct page *page, bool synchronous) |
1da177e4c
|
341 342 343 |
{ struct bio *bio; int ret = 0; |
62c230bc1
|
344 |
struct swap_info_struct *sis = page_swap_info(page); |
23955622f
|
345 |
blk_qc_t qc; |
74d46992e
|
346 |
struct gendisk *disk; |
1da177e4c
|
347 |
|
0bcac06f2
|
348 |
VM_BUG_ON_PAGE(!PageSwapCache(page) && !synchronous, page); |
309381fea
|
349 350 |
VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageUptodate(page), page); |
165c8aed5
|
351 |
if (frontswap_load(page) == 0) { |
38b5faf4b
|
352 353 354 355 |
SetPageUptodate(page); unlock_page(page); goto out; } |
62c230bc1
|
356 357 358 359 360 361 362 363 364 365 |
if (sis->flags & SWP_FILE) { struct file *swap_file = sis->swap_file; struct address_space *mapping = swap_file->f_mapping; ret = mapping->a_ops->readpage(swap_file, page); if (!ret) count_vm_event(PSWPIN); return ret; } |
dd6bd0d9c
|
366 367 |
ret = bdev_read_page(sis->bdev, swap_page_sector(page), page); if (!ret) { |
b06bad17c
|
368 369 370 371 |
if (trylock_page(page)) { swap_slot_free_notify(page); unlock_page(page); } |
dd6bd0d9c
|
372 373 374 375 376 |
count_vm_event(PSWPIN); return 0; } ret = 0; |
f29ad6a99
|
377 |
bio = get_swap_bio(GFP_KERNEL, page, end_swap_bio_read); |
1da177e4c
|
378 379 380 381 382 |
if (bio == NULL) { unlock_page(page); ret = -ENOMEM; goto out; } |
74d46992e
|
383 |
disk = bio->bi_disk; |
b0ba2d0fa
|
384 385 386 387 388 |
/* * Keep this task valid during swap readpage because the oom killer may * attempt to access it in the page fault retry time check. */ get_task_struct(current); |
23955622f
|
389 |
bio->bi_private = current; |
95fe6c1a2
|
390 |
bio_set_op_attrs(bio, REQ_OP_READ, 0); |
f8891e5e1
|
391 |
count_vm_event(PSWPIN); |
23955622f
|
392 393 |
bio_get(bio); qc = submit_bio(bio); |
0bcac06f2
|
394 |
while (synchronous) { |
23955622f
|
395 396 397 |
set_current_state(TASK_UNINTERRUPTIBLE); if (!READ_ONCE(bio->bi_private)) break; |
ea435e1b9
|
398 |
if (!blk_poll(disk->queue, qc)) |
23955622f
|
399 400 401 402 |
break; } __set_current_state(TASK_RUNNING); bio_put(bio); |
1da177e4c
|
403 404 405 |
out: return ret; } |
62c230bc1
|
406 407 408 409 410 411 412 |
int swap_set_page_dirty(struct page *page) { struct swap_info_struct *sis = page_swap_info(page); if (sis->flags & SWP_FILE) { struct address_space *mapping = sis->swap_file->f_mapping; |
cc30c5d64
|
413 414 |
VM_BUG_ON_PAGE(!PageSwapCache(page), page); |
62c230bc1
|
415 416 417 418 419 |
return mapping->a_ops->set_page_dirty(page); } else { return __set_page_dirty_no_writeback(page); } } |