Blame view
fs/block_dev.c
41.4 KB
1da177e4c
|
1 2 3 4 5 6 |
/* * linux/fs/block_dev.c * * Copyright (C) 1991, 1992 Linus Torvalds * Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE */ |
1da177e4c
|
7 8 9 10 11 12 |
#include <linux/init.h> #include <linux/mm.h> #include <linux/fcntl.h> #include <linux/slab.h> #include <linux/kmod.h> #include <linux/major.h> |
7db9cfd38
|
13 |
#include <linux/device_cgroup.h> |
1da177e4c
|
14 15 16 17 18 |
#include <linux/highmem.h> #include <linux/blkdev.h> #include <linux/module.h> #include <linux/blkpg.h> #include <linux/buffer_head.h> |
ff01bb483
|
19 |
#include <linux/swap.h> |
585d3bc06
|
20 |
#include <linux/pagevec.h> |
811d736f9
|
21 |
#include <linux/writeback.h> |
1da177e4c
|
22 23 24 25 |
#include <linux/mpage.h> #include <linux/mount.h> #include <linux/uio.h> #include <linux/namei.h> |
1368c4f24
|
26 |
#include <linux/log2.h> |
ff01bb483
|
27 |
#include <linux/cleancache.h> |
1da177e4c
|
28 |
#include <asm/uaccess.h> |
07f3f05c1
|
29 |
#include "internal.h" |
1da177e4c
|
30 31 32 33 34 |
struct bdev_inode { struct block_device bdev; struct inode vfs_inode; }; |
4c54ac62d
|
35 |
static const struct address_space_operations def_blk_aops; |
1da177e4c
|
36 37 38 39 40 41 42 43 44 |
static inline struct bdev_inode *BDEV_I(struct inode *inode) { return container_of(inode, struct bdev_inode, vfs_inode); } inline struct block_device *I_BDEV(struct inode *inode) { return &BDEV_I(inode)->bdev; } |
1da177e4c
|
45 |
EXPORT_SYMBOL(I_BDEV); |
a5491e0c7
|
46 |
/* |
f758eeabe
|
47 48 49 |
* Move the inode from its current bdi to a new bdi. If the inode is dirty we * need to move it onto the dirty list of @dst so that the inode is always on * the right list. |
a5491e0c7
|
50 51 52 53 |
*/ static void bdev_inode_switch_bdi(struct inode *inode, struct backing_dev_info *dst) { |
f758eeabe
|
54 55 56 57 58 |
struct backing_dev_info *old = inode->i_data.backing_dev_info; if (unlikely(dst == old)) /* deadlock avoidance */ return; bdi_lock_two(&old->wb, &dst->wb); |
250df6ed2
|
59 |
spin_lock(&inode->i_lock); |
a5491e0c7
|
60 61 |
inode->i_data.backing_dev_info = dst; if (inode->i_state & I_DIRTY) |
7ccf19a80
|
62 |
list_move(&inode->i_wb_list, &dst->wb.b_dirty); |
250df6ed2
|
63 |
spin_unlock(&inode->i_lock); |
f758eeabe
|
64 65 |
spin_unlock(&old->wb.list_lock); spin_unlock(&dst->wb.list_lock); |
a5491e0c7
|
66 |
} |
1da177e4c
|
67 68 69 70 71 72 73 74 75 76 77 78 |
static sector_t max_block(struct block_device *bdev) { sector_t retval = ~((sector_t)0); loff_t sz = i_size_read(bdev->bd_inode); if (sz) { unsigned int size = block_size(bdev); unsigned int sizebits = blksize_bits(size); retval = (sz >> sizebits); } return retval; } |
f9a14399a
|
79 |
/* Kill _all_ buffers and pagecache , dirty or not.. */ |
ff01bb483
|
80 |
void kill_bdev(struct block_device *bdev) |
1da177e4c
|
81 |
{ |
ff01bb483
|
82 83 84 |
struct address_space *mapping = bdev->bd_inode->i_mapping; if (mapping->nrpages == 0) |
f9a14399a
|
85 |
return; |
ff01bb483
|
86 |
|
f9a14399a
|
87 |
invalidate_bh_lrus(); |
ff01bb483
|
88 |
truncate_inode_pages(mapping, 0); |
1da177e4c
|
89 |
} |
ff01bb483
|
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 |
EXPORT_SYMBOL(kill_bdev); /* Invalidate clean unused buffers and pagecache. */ void invalidate_bdev(struct block_device *bdev) { struct address_space *mapping = bdev->bd_inode->i_mapping; if (mapping->nrpages == 0) return; invalidate_bh_lrus(); lru_add_drain_all(); /* make sure all lru add caches are flushed */ invalidate_mapping_pages(mapping, 0, -1); /* 99% of the time, we don't need to flush the cleancache on the bdev. * But, for the strange corners, lets be cautious */ cleancache_flush_inode(mapping); } EXPORT_SYMBOL(invalidate_bdev); |
1da177e4c
|
109 110 111 112 |
int set_blocksize(struct block_device *bdev, int size) { /* Size must be a power of two, and between 512 and PAGE_SIZE */ |
1368c4f24
|
113 |
if (size > PAGE_SIZE || size < 512 || !is_power_of_2(size)) |
1da177e4c
|
114 115 116 |
return -EINVAL; /* Size cannot be smaller than the size supported by the device */ |
e1defc4ff
|
117 |
if (size < bdev_logical_block_size(bdev)) |
1da177e4c
|
118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 |
return -EINVAL; /* Don't change the size if it is same as current */ if (bdev->bd_block_size != size) { sync_blockdev(bdev); bdev->bd_block_size = size; bdev->bd_inode->i_blkbits = blksize_bits(size); kill_bdev(bdev); } return 0; } EXPORT_SYMBOL(set_blocksize); int sb_set_blocksize(struct super_block *sb, int size) { |
1da177e4c
|
134 135 136 137 138 |
if (set_blocksize(sb->s_bdev, size)) return 0; /* If we get here, we know size is power of two * and it's value is between 512 and PAGE_SIZE */ sb->s_blocksize = size; |
38885bd4c
|
139 |
sb->s_blocksize_bits = blksize_bits(size); |
1da177e4c
|
140 141 142 143 144 145 146 |
return sb->s_blocksize; } EXPORT_SYMBOL(sb_set_blocksize); int sb_min_blocksize(struct super_block *sb, int size) { |
e1defc4ff
|
147 |
int minsize = bdev_logical_block_size(sb->s_bdev); |
1da177e4c
|
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 |
if (size < minsize) size = minsize; return sb_set_blocksize(sb, size); } EXPORT_SYMBOL(sb_min_blocksize); static int blkdev_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh, int create) { if (iblock >= max_block(I_BDEV(inode))) { if (create) return -EIO; /* * for reads, we're just trying to fill a partial page. * return a hole, they will have to call get_block again * before they can fill it, and they will get -EIO at that * time */ return 0; } bh->b_bdev = I_BDEV(inode); bh->b_blocknr = iblock; set_buffer_mapped(bh); return 0; } |
b2e895dbd
|
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 |
static int blkdev_get_blocks(struct inode *inode, sector_t iblock, struct buffer_head *bh, int create) { sector_t end_block = max_block(I_BDEV(inode)); unsigned long max_blocks = bh->b_size >> inode->i_blkbits; if ((iblock + max_blocks) > end_block) { max_blocks = end_block - iblock; if ((long)max_blocks <= 0) { if (create) return -EIO; /* write fully beyond EOF */ /* * It is a read which is fully beyond EOF. We return * a !buffer_mapped buffer */ max_blocks = 0; } } bh->b_bdev = I_BDEV(inode); bh->b_blocknr = iblock; bh->b_size = max_blocks << inode->i_blkbits; if (max_blocks) set_buffer_mapped(bh); return 0; } static ssize_t blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; |
eafdc7d19
|
210 211 |
return __blockdev_direct_IO(rw, iocb, inode, I_BDEV(inode), iov, offset, nr_segs, blkdev_get_blocks, NULL, NULL, 0); |
b2e895dbd
|
212 |
} |
5cee5815d
|
213 214 215 216 217 218 219 220 |
int __sync_blockdev(struct block_device *bdev, int wait) { if (!bdev) return 0; if (!wait) return filemap_flush(bdev->bd_inode->i_mapping); return filemap_write_and_wait(bdev->bd_inode->i_mapping); } |
585d3bc06
|
221 222 223 224 225 226 |
/* * Write out and wait upon all the dirty data associated with a block * device via its mapping. Does not take the superblock lock. */ int sync_blockdev(struct block_device *bdev) { |
5cee5815d
|
227 |
return __sync_blockdev(bdev, 1); |
585d3bc06
|
228 229 230 231 232 233 234 235 236 237 238 239 |
} EXPORT_SYMBOL(sync_blockdev); /* * Write out and wait upon all dirty data associated with this * device. Filesystem data as well as the underlying block * device. Takes the superblock lock. */ int fsync_bdev(struct block_device *bdev) { struct super_block *sb = get_super(bdev); if (sb) { |
60b0680fa
|
240 |
int res = sync_filesystem(sb); |
585d3bc06
|
241 242 243 244 245 |
drop_super(sb); return res; } return sync_blockdev(bdev); } |
47e4491b4
|
246 |
EXPORT_SYMBOL(fsync_bdev); |
585d3bc06
|
247 248 249 250 251 |
/** * freeze_bdev -- lock a filesystem and force it into a consistent state * @bdev: blockdevice to lock * |
585d3bc06
|
252 253 254 255 256 257 258 259 260 261 262 263 264 265 |
* If a superblock is found on this device, we take the s_umount semaphore * on it to make sure nobody unmounts until the snapshot creation is done. * The reference counter (bd_fsfreeze_count) guarantees that only the last * unfreeze process can unfreeze the frozen filesystem actually when multiple * freeze requests arrive simultaneously. It counts up in freeze_bdev() and * count down in thaw_bdev(). When it becomes 0, thaw_bdev() will unfreeze * actually. */ struct super_block *freeze_bdev(struct block_device *bdev) { struct super_block *sb; int error = 0; mutex_lock(&bdev->bd_fsfreeze_mutex); |
4504230a7
|
266 267 268 269 270 271 |
if (++bdev->bd_fsfreeze_count > 1) { /* * We don't even need to grab a reference - the first call * to freeze_bdev grab an active reference and only the last * thaw_bdev drops it. */ |
585d3bc06
|
272 |
sb = get_super(bdev); |
4504230a7
|
273 274 275 276 277 278 279 280 |
drop_super(sb); mutex_unlock(&bdev->bd_fsfreeze_mutex); return sb; } sb = get_active_super(bdev); if (!sb) goto out; |
18e9e5104
|
281 282 283 284 |
error = freeze_super(sb); if (error) { deactivate_super(sb); bdev->bd_fsfreeze_count--; |
585d3bc06
|
285 |
mutex_unlock(&bdev->bd_fsfreeze_mutex); |
18e9e5104
|
286 |
return ERR_PTR(error); |
585d3bc06
|
287 |
} |
18e9e5104
|
288 |
deactivate_super(sb); |
4504230a7
|
289 |
out: |
585d3bc06
|
290 291 |
sync_blockdev(bdev); mutex_unlock(&bdev->bd_fsfreeze_mutex); |
4fadd7bb2
|
292 |
return sb; /* thaw_bdev releases s->s_umount */ |
585d3bc06
|
293 294 295 296 297 298 299 300 301 302 303 304 |
} EXPORT_SYMBOL(freeze_bdev); /** * thaw_bdev -- unlock filesystem * @bdev: blockdevice to unlock * @sb: associated superblock * * Unlocks the filesystem and marks it writeable again after freeze_bdev(). */ int thaw_bdev(struct block_device *bdev, struct super_block *sb) { |
4504230a7
|
305 |
int error = -EINVAL; |
585d3bc06
|
306 307 |
mutex_lock(&bdev->bd_fsfreeze_mutex); |
4504230a7
|
308 |
if (!bdev->bd_fsfreeze_count) |
18e9e5104
|
309 |
goto out; |
4504230a7
|
310 311 312 |
error = 0; if (--bdev->bd_fsfreeze_count > 0) |
18e9e5104
|
313 |
goto out; |
4504230a7
|
314 315 |
if (!sb) |
18e9e5104
|
316 |
goto out; |
4504230a7
|
317 |
|
18e9e5104
|
318 319 320 321 322 323 324 |
error = thaw_super(sb); if (error) { bdev->bd_fsfreeze_count++; mutex_unlock(&bdev->bd_fsfreeze_mutex); return error; } out: |
585d3bc06
|
325 326 327 328 |
mutex_unlock(&bdev->bd_fsfreeze_mutex); return 0; } EXPORT_SYMBOL(thaw_bdev); |
1da177e4c
|
329 330 331 332 333 334 335 336 337 |
static int blkdev_writepage(struct page *page, struct writeback_control *wbc) { return block_write_full_page(page, blkdev_get_block, wbc); } static int blkdev_readpage(struct file * file, struct page * page) { return block_read_full_page(page, blkdev_get_block); } |
6272b5a58
|
338 339 340 |
static int blkdev_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) |
1da177e4c
|
341 |
{ |
155130a4f
|
342 343 |
return block_write_begin(mapping, pos, len, flags, pagep, blkdev_get_block); |
1da177e4c
|
344 |
} |
6272b5a58
|
345 346 347 |
static int blkdev_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata) |
1da177e4c
|
348 |
{ |
6272b5a58
|
349 350 351 352 353 354 355 |
int ret; ret = block_write_end(file, mapping, pos, len, copied, page, fsdata); unlock_page(page); page_cache_release(page); return ret; |
1da177e4c
|
356 357 358 359 |
} /* * private llseek: |
0f7fc9e4d
|
360 |
* for a block special file file->f_path.dentry->d_inode->i_size is zero |
1da177e4c
|
361 362 363 364 365 366 367 |
* so we compute the size by hand (just as in block_read/write above) */ static loff_t block_llseek(struct file *file, loff_t offset, int origin) { struct inode *bd_inode = file->f_mapping->host; loff_t size; loff_t retval; |
1b1dcc1b5
|
368 |
mutex_lock(&bd_inode->i_mutex); |
1da177e4c
|
369 |
size = i_size_read(bd_inode); |
06222e491
|
370 |
retval = -EINVAL; |
1da177e4c
|
371 |
switch (origin) { |
06222e491
|
372 |
case SEEK_END: |
1da177e4c
|
373 374 |
offset += size; break; |
06222e491
|
375 |
case SEEK_CUR: |
1da177e4c
|
376 |
offset += file->f_pos; |
06222e491
|
377 378 379 380 |
case SEEK_SET: break; default: goto out; |
1da177e4c
|
381 |
} |
1da177e4c
|
382 383 384 385 386 387 |
if (offset >= 0 && offset <= size) { if (offset != file->f_pos) { file->f_pos = offset; } retval = offset; } |
06222e491
|
388 |
out: |
1b1dcc1b5
|
389 |
mutex_unlock(&bd_inode->i_mutex); |
1da177e4c
|
390 391 392 |
return retval; } |
02c24a821
|
393 |
int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync) |
1da177e4c
|
394 |
{ |
b8af67e26
|
395 396 |
struct inode *bd_inode = filp->f_mapping->host; struct block_device *bdev = I_BDEV(bd_inode); |
ab0a9735e
|
397 |
int error; |
da5aa861b
|
398 399 400 401 |
error = filemap_write_and_wait_range(filp->f_mapping, start, end); if (error) return error; |
ab0a9735e
|
402 |
|
b8af67e26
|
403 404 405 406 407 |
/* * There is no need to serialise calls to blkdev_issue_flush with * i_mutex and doing so causes performance issues with concurrent * O_SYNC writers to a block device. */ |
dd3932edd
|
408 |
error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL); |
ab0a9735e
|
409 410 |
if (error == -EOPNOTSUPP) error = 0; |
b8af67e26
|
411 |
|
ab0a9735e
|
412 |
return error; |
1da177e4c
|
413 |
} |
b1dd3b284
|
414 |
EXPORT_SYMBOL(blkdev_fsync); |
1da177e4c
|
415 416 417 418 419 420 |
/* * pseudo-fs */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(bdev_lock); |
e18b890bb
|
421 |
static struct kmem_cache * bdev_cachep __read_mostly; |
1da177e4c
|
422 423 424 |
static struct inode *bdev_alloc_inode(struct super_block *sb) { |
e94b17660
|
425 |
struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL); |
1da177e4c
|
426 427 428 429 |
if (!ei) return NULL; return &ei->vfs_inode; } |
fa0d7e3de
|
430 |
static void bdev_i_callback(struct rcu_head *head) |
1da177e4c
|
431 |
{ |
fa0d7e3de
|
432 |
struct inode *inode = container_of(head, struct inode, i_rcu); |
1da177e4c
|
433 |
struct bdev_inode *bdi = BDEV_I(inode); |
1da177e4c
|
434 435 |
kmem_cache_free(bdev_cachep, bdi); } |
fa0d7e3de
|
436 437 438 439 |
static void bdev_destroy_inode(struct inode *inode) { call_rcu(&inode->i_rcu, bdev_i_callback); } |
51cc50685
|
440 |
static void init_once(void *foo) |
1da177e4c
|
441 442 443 |
{ struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; |
a35afb830
|
444 445 |
memset(bdev, 0, sizeof(*bdev)); mutex_init(&bdev->bd_mutex); |
a35afb830
|
446 447 |
INIT_LIST_HEAD(&bdev->bd_inodes); INIT_LIST_HEAD(&bdev->bd_list); |
49731baa4
|
448 449 450 |
#ifdef CONFIG_SYSFS INIT_LIST_HEAD(&bdev->bd_holder_disks); #endif |
a35afb830
|
451 |
inode_init_once(&ei->vfs_inode); |
fcccf5025
|
452 453 |
/* Initialize mutex for freeze. */ mutex_init(&bdev->bd_fsfreeze_mutex); |
1da177e4c
|
454 455 456 457 458 459 460 461 |
} static inline void __bd_forget(struct inode *inode) { list_del_init(&inode->i_devices); inode->i_bdev = NULL; inode->i_mapping = &inode->i_data; } |
b57922d97
|
462 |
static void bdev_evict_inode(struct inode *inode) |
1da177e4c
|
463 464 465 |
{ struct block_device *bdev = &BDEV_I(inode)->bdev; struct list_head *p; |
b57922d97
|
466 467 468 |
truncate_inode_pages(&inode->i_data, 0); invalidate_inode_buffers(inode); /* is it needed here? */ end_writeback(inode); |
1da177e4c
|
469 470 471 472 473 474 475 |
spin_lock(&bdev_lock); while ( (p = bdev->bd_inodes.next) != &bdev->bd_inodes ) { __bd_forget(list_entry(p, struct inode, i_devices)); } list_del_init(&bdev->bd_list); spin_unlock(&bdev_lock); } |
ee9b6d61a
|
476 |
static const struct super_operations bdev_sops = { |
1da177e4c
|
477 478 479 480 |
.statfs = simple_statfs, .alloc_inode = bdev_alloc_inode, .destroy_inode = bdev_destroy_inode, .drop_inode = generic_delete_inode, |
b57922d97
|
481 |
.evict_inode = bdev_evict_inode, |
1da177e4c
|
482 |
}; |
51139adac
|
483 484 |
static struct dentry *bd_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) |
1da177e4c
|
485 |
{ |
c74a1cbb3
|
486 |
return mount_pseudo(fs_type, "bdev:", &bdev_sops, NULL, 0x62646576); |
1da177e4c
|
487 488 489 490 |
} static struct file_system_type bd_type = { .name = "bdev", |
51139adac
|
491 |
.mount = bd_mount, |
1da177e4c
|
492 493 |
.kill_sb = kill_anon_super, }; |
f47ec3f28
|
494 |
static struct super_block *blockdev_superblock __read_mostly; |
1da177e4c
|
495 496 497 498 |
void __init bdev_cache_init(void) { int err; |
ace8577ae
|
499 |
static struct vfsmount *bd_mnt; |
c2acf7b90
|
500 |
|
1da177e4c
|
501 |
bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode), |
fffb60f93
|
502 503 |
0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT| SLAB_MEM_SPREAD|SLAB_PANIC), |
20c2df83d
|
504 |
init_once); |
1da177e4c
|
505 506 507 508 |
err = register_filesystem(&bd_type); if (err) panic("Cannot register bdev pseudo-fs"); bd_mnt = kern_mount(&bd_type); |
1da177e4c
|
509 510 |
if (IS_ERR(bd_mnt)) panic("Cannot create bdev pseudo-fs"); |
ace8577ae
|
511 |
blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */ |
1da177e4c
|
512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 |
} /* * Most likely _very_ bad one - but then it's hardly critical for small * /dev and can be fixed when somebody will need really large one. * Keep in mind that it will be fed through icache hash function too. */ static inline unsigned long hash(dev_t dev) { return MAJOR(dev)+MINOR(dev); } static int bdev_test(struct inode *inode, void *data) { return BDEV_I(inode)->bdev.bd_dev == *(dev_t *)data; } static int bdev_set(struct inode *inode, void *data) { BDEV_I(inode)->bdev.bd_dev = *(dev_t *)data; return 0; } static LIST_HEAD(all_bdevs); struct block_device *bdget(dev_t dev) { struct block_device *bdev; struct inode *inode; |
c2acf7b90
|
541 |
inode = iget5_locked(blockdev_superblock, hash(dev), |
1da177e4c
|
542 543 544 545 546 547 548 549 550 |
bdev_test, bdev_set, &dev); if (!inode) return NULL; bdev = &BDEV_I(inode)->bdev; if (inode->i_state & I_NEW) { bdev->bd_contains = NULL; |
782b94cdf
|
551 |
bdev->bd_super = NULL; |
1da177e4c
|
552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 |
bdev->bd_inode = inode; bdev->bd_block_size = (1 << inode->i_blkbits); bdev->bd_part_count = 0; bdev->bd_invalidated = 0; inode->i_mode = S_IFBLK; inode->i_rdev = dev; inode->i_bdev = bdev; inode->i_data.a_ops = &def_blk_aops; mapping_set_gfp_mask(&inode->i_data, GFP_USER); inode->i_data.backing_dev_info = &default_backing_dev_info; spin_lock(&bdev_lock); list_add(&bdev->bd_list, &all_bdevs); spin_unlock(&bdev_lock); unlock_new_inode(inode); } return bdev; } EXPORT_SYMBOL(bdget); |
dddac6a7b
|
571 572 573 574 575 576 |
/** * bdgrab -- Grab a reference to an already referenced block device * @bdev: Block device to grab a reference to. */ struct block_device *bdgrab(struct block_device *bdev) { |
7de9c6ee3
|
577 |
ihold(bdev->bd_inode); |
dddac6a7b
|
578 579 |
return bdev; } |
1da177e4c
|
580 581 |
long nr_blockdev_pages(void) { |
203a2935c
|
582 |
struct block_device *bdev; |
1da177e4c
|
583 584 |
long ret = 0; spin_lock(&bdev_lock); |
203a2935c
|
585 |
list_for_each_entry(bdev, &all_bdevs, bd_list) { |
1da177e4c
|
586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 |
ret += bdev->bd_inode->i_mapping->nrpages; } spin_unlock(&bdev_lock); return ret; } void bdput(struct block_device *bdev) { iput(bdev->bd_inode); } EXPORT_SYMBOL(bdput); static struct block_device *bd_acquire(struct inode *inode) { struct block_device *bdev; |
09d967c6f
|
602 |
|
1da177e4c
|
603 604 |
spin_lock(&bdev_lock); bdev = inode->i_bdev; |
09d967c6f
|
605 |
if (bdev) { |
7de9c6ee3
|
606 |
ihold(bdev->bd_inode); |
1da177e4c
|
607 608 609 610 |
spin_unlock(&bdev_lock); return bdev; } spin_unlock(&bdev_lock); |
09d967c6f
|
611 |
|
1da177e4c
|
612 613 614 |
bdev = bdget(inode->i_rdev); if (bdev) { spin_lock(&bdev_lock); |
09d967c6f
|
615 616 |
if (!inode->i_bdev) { /* |
7de9c6ee3
|
617 |
* We take an additional reference to bd_inode, |
09d967c6f
|
618 619 620 621 |
* and it's released in clear_inode() of inode. * So, we can access it via ->i_mapping always * without igrab(). */ |
7de9c6ee3
|
622 |
ihold(bdev->bd_inode); |
09d967c6f
|
623 624 625 626 |
inode->i_bdev = bdev; inode->i_mapping = bdev->bd_inode->i_mapping; list_add(&inode->i_devices, &bdev->bd_inodes); } |
1da177e4c
|
627 628 629 630 |
spin_unlock(&bdev_lock); } return bdev; } |
f47ec3f28
|
631 632 633 634 |
static inline int sb_is_blkdev_sb(struct super_block *sb) { return sb == blockdev_superblock; } |
1da177e4c
|
635 636 637 638 |
/* Call when you free inode */ void bd_forget(struct inode *inode) { |
09d967c6f
|
639 |
struct block_device *bdev = NULL; |
1da177e4c
|
640 |
spin_lock(&bdev_lock); |
09d967c6f
|
641 |
if (inode->i_bdev) { |
c2acf7b90
|
642 |
if (!sb_is_blkdev_sb(inode->i_sb)) |
09d967c6f
|
643 |
bdev = inode->i_bdev; |
1da177e4c
|
644 |
__bd_forget(inode); |
09d967c6f
|
645 |
} |
1da177e4c
|
646 |
spin_unlock(&bdev_lock); |
09d967c6f
|
647 648 649 |
if (bdev) iput(bdev->bd_inode); |
1da177e4c
|
650 |
} |
1a3cbbc5a
|
651 652 653 654 655 656 |
/** * bd_may_claim - test whether a block device can be claimed * @bdev: block device of interest * @whole: whole block device containing @bdev, may equal @bdev * @holder: holder trying to claim @bdev * |
25985edce
|
657 |
* Test whether @bdev can be claimed by @holder. |
1a3cbbc5a
|
658 659 660 661 662 663 664 665 666 |
* * CONTEXT: * spin_lock(&bdev_lock). * * RETURNS: * %true if @bdev can be claimed, %false otherwise. */ static bool bd_may_claim(struct block_device *bdev, struct block_device *whole, void *holder) |
1da177e4c
|
667 |
{ |
1da177e4c
|
668 |
if (bdev->bd_holder == holder) |
1a3cbbc5a
|
669 |
return true; /* already a holder */ |
1da177e4c
|
670 |
else if (bdev->bd_holder != NULL) |
1a3cbbc5a
|
671 |
return false; /* held by someone else */ |
1da177e4c
|
672 |
else if (bdev->bd_contains == bdev) |
1a3cbbc5a
|
673 |
return true; /* is a whole device which isn't held */ |
1da177e4c
|
674 |
|
e525fd89d
|
675 |
else if (whole->bd_holder == bd_may_claim) |
1a3cbbc5a
|
676 677 678 |
return true; /* is a partition of a device that is being partitioned */ else if (whole->bd_holder != NULL) return false; /* is a partition of a held device */ |
1da177e4c
|
679 |
else |
1a3cbbc5a
|
680 681 682 683 |
return true; /* is a partition of an un-held device */ } /** |
6b4517a79
|
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 |
* bd_prepare_to_claim - prepare to claim a block device * @bdev: block device of interest * @whole: the whole device containing @bdev, may equal @bdev * @holder: holder trying to claim @bdev * * Prepare to claim @bdev. This function fails if @bdev is already * claimed by another holder and waits if another claiming is in * progress. This function doesn't actually claim. On successful * return, the caller has ownership of bd_claiming and bd_holder[s]. * * CONTEXT: * spin_lock(&bdev_lock). Might release bdev_lock, sleep and regrab * it multiple times. * * RETURNS: * 0 if @bdev can be claimed, -EBUSY otherwise. */ static int bd_prepare_to_claim(struct block_device *bdev, struct block_device *whole, void *holder) { retry: /* if someone else claimed, fail */ if (!bd_may_claim(bdev, whole, holder)) return -EBUSY; |
e75aa8589
|
708 709 |
/* if claiming is already in progress, wait for it to finish */ if (whole->bd_claiming) { |
6b4517a79
|
710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 |
wait_queue_head_t *wq = bit_waitqueue(&whole->bd_claiming, 0); DEFINE_WAIT(wait); prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE); spin_unlock(&bdev_lock); schedule(); finish_wait(wq, &wait); spin_lock(&bdev_lock); goto retry; } /* yay, all mine */ return 0; } /** * bd_start_claiming - start claiming a block device * @bdev: block device of interest * @holder: holder trying to claim @bdev * * @bdev is about to be opened exclusively. Check @bdev can be opened * exclusively and mark that an exclusive open is in progress. Each * successful call to this function must be matched with a call to |
b0018361c
|
733 734 735 736 737 738 739 |
* either bd_finish_claiming() or bd_abort_claiming() (which do not * fail). * * This function is used to gain exclusive access to the block device * without actually causing other exclusive open attempts to fail. It * should be used when the open sequence itself requires exclusive * access but may subsequently fail. |
6b4517a79
|
740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 |
* * CONTEXT: * Might sleep. * * RETURNS: * Pointer to the block device containing @bdev on success, ERR_PTR() * value on failure. */ static struct block_device *bd_start_claiming(struct block_device *bdev, void *holder) { struct gendisk *disk; struct block_device *whole; int partno, err; might_sleep(); /* * @bdev might not have been initialized properly yet, look up * and grab the outer block device the hard way. */ disk = get_gendisk(bdev->bd_dev, &partno); if (!disk) return ERR_PTR(-ENXIO); |
d4c208b86
|
764 765 766 767 768 769 770 771 772 773 774 775 |
/* * Normally, @bdev should equal what's returned from bdget_disk() * if partno is 0; however, some drivers (floppy) use multiple * bdev's for the same physical device and @bdev may be one of the * aliases. Keep @bdev if partno is 0. This means claimer * tracking is broken for those devices but it has always been that * way. */ if (partno) whole = bdget_disk(disk, 0); else whole = bdgrab(bdev); |
cf3425707
|
776 |
module_put(disk->fops->owner); |
6b4517a79
|
777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 |
put_disk(disk); if (!whole) return ERR_PTR(-ENOMEM); /* prepare to claim, if successful, mark claiming in progress */ spin_lock(&bdev_lock); err = bd_prepare_to_claim(bdev, whole, holder); if (err == 0) { whole->bd_claiming = holder; spin_unlock(&bdev_lock); return whole; } else { spin_unlock(&bdev_lock); bdput(whole); return ERR_PTR(err); } } |
641dc636b
|
795 |
#ifdef CONFIG_SYSFS |
49731baa4
|
796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 |
struct bd_holder_disk { struct list_head list; struct gendisk *disk; int refcnt; }; static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev, struct gendisk *disk) { struct bd_holder_disk *holder; list_for_each_entry(holder, &bdev->bd_holder_disks, list) if (holder->disk == disk) return holder; return NULL; } |
4d7dd8fd9
|
812 |
static int add_symlink(struct kobject *from, struct kobject *to) |
641dc636b
|
813 |
{ |
4d7dd8fd9
|
814 |
return sysfs_create_link(from, to, kobject_name(to)); |
641dc636b
|
815 816 817 818 |
} static void del_symlink(struct kobject *from, struct kobject *to) { |
641dc636b
|
819 820 |
sysfs_remove_link(from, kobject_name(to)); } |
641dc636b
|
821 |
/** |
e09b457bd
|
822 823 824 |
* bd_link_disk_holder - create symlinks between holding disk and slave bdev * @bdev: the claimed slave bdev * @disk: the holding disk |
df6c0cd9a
|
825 |
* |
49731baa4
|
826 827 |
* DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. * |
e09b457bd
|
828 |
* This functions creates the following sysfs symlinks. |
641dc636b
|
829 |
* |
e09b457bd
|
830 831 |
* - from "slaves" directory of the holder @disk to the claimed @bdev * - from "holders" directory of the @bdev to the holder @disk |
641dc636b
|
832 |
* |
e09b457bd
|
833 834 |
* For example, if /dev/dm-0 maps to /dev/sda and disk for dm-0 is * passed to bd_link_disk_holder(), then: |
641dc636b
|
835 |
* |
e09b457bd
|
836 837 |
* /sys/block/dm-0/slaves/sda --> /sys/block/sda * /sys/block/sda/holders/dm-0 --> /sys/block/dm-0 |
641dc636b
|
838 |
* |
e09b457bd
|
839 840 841 |
* The caller must have claimed @bdev before calling this function and * ensure that both @bdev and @disk are valid during the creation and * lifetime of these symlinks. |
641dc636b
|
842 |
* |
e09b457bd
|
843 844 |
* CONTEXT: * Might sleep. |
641dc636b
|
845 |
* |
e09b457bd
|
846 847 |
* RETURNS: * 0 on success, -errno on failure. |
641dc636b
|
848 |
*/ |
e09b457bd
|
849 |
int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) |
641dc636b
|
850 |
{ |
49731baa4
|
851 |
struct bd_holder_disk *holder; |
e09b457bd
|
852 |
int ret = 0; |
641dc636b
|
853 |
|
2e7b651df
|
854 |
mutex_lock(&bdev->bd_mutex); |
df6c0cd9a
|
855 |
|
49731baa4
|
856 |
WARN_ON_ONCE(!bdev->bd_holder); |
4e91672c7
|
857 |
|
e09b457bd
|
858 859 860 |
/* FIXME: remove the following once add_disk() handles errors */ if (WARN_ON(!disk->slave_dir || !bdev->bd_part->holder_dir)) goto out_unlock; |
4e91672c7
|
861 |
|
49731baa4
|
862 863 864 |
holder = bd_find_holder_disk(bdev, disk); if (holder) { holder->refcnt++; |
e09b457bd
|
865 |
goto out_unlock; |
49731baa4
|
866 |
} |
641dc636b
|
867 |
|
49731baa4
|
868 869 870 |
holder = kzalloc(sizeof(*holder), GFP_KERNEL); if (!holder) { ret = -ENOMEM; |
e09b457bd
|
871 872 |
goto out_unlock; } |
641dc636b
|
873 |
|
49731baa4
|
874 875 876 877 878 879 880 881 882 883 884 |
INIT_LIST_HEAD(&holder->list); holder->disk = disk; holder->refcnt = 1; ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); if (ret) goto out_free; ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); if (ret) goto out_del; |
e7407d161
|
885 886 887 888 889 |
/* * bdev could be deleted beneath us which would implicitly destroy * the holder directory. Hold on to it. */ kobject_get(bdev->bd_part->holder_dir); |
49731baa4
|
890 891 892 893 894 895 896 897 |
list_add(&holder->list, &bdev->bd_holder_disks); goto out_unlock; out_del: del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); out_free: kfree(holder); |
e09b457bd
|
898 |
out_unlock: |
b4cf1b72e
|
899 |
mutex_unlock(&bdev->bd_mutex); |
e09b457bd
|
900 |
return ret; |
641dc636b
|
901 |
} |
e09b457bd
|
902 |
EXPORT_SYMBOL_GPL(bd_link_disk_holder); |
641dc636b
|
903 |
|
49731baa4
|
904 905 906 907 908 909 910 911 912 913 914 |
/** * bd_unlink_disk_holder - destroy symlinks created by bd_link_disk_holder() * @bdev: the calimed slave bdev * @disk: the holding disk * * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. * * CONTEXT: * Might sleep. */ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk) |
641dc636b
|
915 |
{ |
49731baa4
|
916 |
struct bd_holder_disk *holder; |
641dc636b
|
917 |
|
49731baa4
|
918 |
mutex_lock(&bdev->bd_mutex); |
641dc636b
|
919 |
|
49731baa4
|
920 921 922 923 924 925 |
holder = bd_find_holder_disk(bdev, disk); if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) { del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); del_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); |
e7407d161
|
926 |
kobject_put(bdev->bd_part->holder_dir); |
49731baa4
|
927 928 929 930 931 |
list_del_init(&holder->list); kfree(holder); } mutex_unlock(&bdev->bd_mutex); |
1da177e4c
|
932 |
} |
49731baa4
|
933 |
EXPORT_SYMBOL_GPL(bd_unlink_disk_holder); |
641dc636b
|
934 |
#endif |
1da177e4c
|
935 |
|
0c002c2f7
|
936 |
/** |
56ade44b4
|
937 938 939 |
* flush_disk - invalidates all buffer-cache entries on a disk * * @bdev: struct block device to be flushed |
e6eb5ce1b
|
940 |
* @kill_dirty: flag to guide handling of dirty inodes |
56ade44b4
|
941 942 943 944 945 |
* * Invalidates all buffer-cache entries on a disk. It should be called * when a disk has been changed -- either by a media change or online * resize. */ |
93b270f76
|
946 |
static void flush_disk(struct block_device *bdev, bool kill_dirty) |
56ade44b4
|
947 |
{ |
93b270f76
|
948 |
if (__invalidate_device(bdev, kill_dirty)) { |
56ade44b4
|
949 950 951 952 953 954 955 956 957 958 959 |
char name[BDEVNAME_SIZE] = ""; if (bdev->bd_disk) disk_name(bdev->bd_disk, 0, name); printk(KERN_WARNING "VFS: busy inodes on changed media or " "resized disk %s ", name); } if (!bdev->bd_disk) return; |
d27769ec3
|
960 |
if (disk_part_scan_enabled(bdev->bd_disk)) |
56ade44b4
|
961 962 963 964 |
bdev->bd_invalidated = 1; } /** |
57d1b5366
|
965 |
* check_disk_size_change - checks for disk size change and adjusts bdev size. |
c3279d145
|
966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 |
* @disk: struct gendisk to check * @bdev: struct bdev to adjust. * * This routine checks to see if the bdev size does not match the disk size * and adjusts it if it differs. */ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) { loff_t disk_size, bdev_size; disk_size = (loff_t)get_capacity(disk) << 9; bdev_size = i_size_read(bdev->bd_inode); if (disk_size != bdev_size) { char name[BDEVNAME_SIZE]; disk_name(disk, 0, name); printk(KERN_INFO "%s: detected capacity change from %lld to %lld ", name, bdev_size, disk_size); i_size_write(bdev->bd_inode, disk_size); |
93b270f76
|
987 |
flush_disk(bdev, false); |
c3279d145
|
988 989 990 991 992 |
} } EXPORT_SYMBOL(check_disk_size_change); /** |
57d1b5366
|
993 |
* revalidate_disk - wrapper for lower-level driver's revalidate_disk call-back |
0c002c2f7
|
994 995 996 997 998 999 1000 1001 |
* @disk: struct gendisk to be revalidated * * This routine is a wrapper for lower-level driver's revalidate_disk * call-backs. It is used to do common pre and post operations needed * for all revalidate_disk operations. */ int revalidate_disk(struct gendisk *disk) { |
c3279d145
|
1002 |
struct block_device *bdev; |
0c002c2f7
|
1003 1004 1005 1006 |
int ret = 0; if (disk->fops->revalidate_disk) ret = disk->fops->revalidate_disk(disk); |
c3279d145
|
1007 1008 1009 1010 1011 1012 1013 1014 |
bdev = bdget_disk(disk, 0); if (!bdev) return ret; mutex_lock(&bdev->bd_mutex); check_disk_size_change(disk, bdev); mutex_unlock(&bdev->bd_mutex); bdput(bdev); |
0c002c2f7
|
1015 1016 1017 |
return ret; } EXPORT_SYMBOL(revalidate_disk); |
1da177e4c
|
1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 |
/* * This routine checks whether a removable media has been changed, * and invalidates all buffer-cache-entries in that case. This * is a relatively slow routine, so we have to try to minimize using * it. Thus it is called only upon a 'mount' or 'open'. This * is the best way of combining speed and utility, I think. * People changing diskettes in the middle of an operation deserve * to lose :-) */ int check_disk_change(struct block_device *bdev) { struct gendisk *disk = bdev->bd_disk; |
83d5cde47
|
1030 |
const struct block_device_operations *bdops = disk->fops; |
77ea887e4
|
1031 |
unsigned int events; |
1da177e4c
|
1032 |
|
77ea887e4
|
1033 1034 1035 |
events = disk_clear_events(disk, DISK_EVENT_MEDIA_CHANGE | DISK_EVENT_EJECT_REQUEST); if (!(events & DISK_EVENT_MEDIA_CHANGE)) |
1da177e4c
|
1036 |
return 0; |
93b270f76
|
1037 |
flush_disk(bdev, true); |
1da177e4c
|
1038 1039 |
if (bdops->revalidate_disk) bdops->revalidate_disk(bdev->bd_disk); |
1da177e4c
|
1040 1041 1042 1043 1044 1045 1046 |
return 1; } EXPORT_SYMBOL(check_disk_change); void bd_set_size(struct block_device *bdev, loff_t size) { |
e1defc4ff
|
1047 |
unsigned bsize = bdev_logical_block_size(bdev); |
1da177e4c
|
1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 |
bdev->bd_inode->i_size = size; while (bsize < PAGE_CACHE_SIZE) { if (size & bsize) break; bsize <<= 1; } bdev->bd_block_size = bsize; bdev->bd_inode->i_blkbits = blksize_bits(bsize); } EXPORT_SYMBOL(bd_set_size); |
9a1c35427
|
1059 |
static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part); |
37be41241
|
1060 |
|
6d740cd5b
|
1061 1062 1063 1064 1065 1066 |
/* * bd_mutex locking: * * mutex_lock(part->bd_mutex) * mutex_lock_nested(whole->bd_mutex, 1) */ |
572c48921
|
1067 |
static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) |
1da177e4c
|
1068 |
{ |
1da177e4c
|
1069 |
struct gendisk *disk; |
523e1d399
|
1070 |
struct module *owner; |
7db9cfd38
|
1071 |
int ret; |
cf771cb5a
|
1072 |
int partno; |
fe6e9c1f2
|
1073 |
int perm = 0; |
572c48921
|
1074 |
if (mode & FMODE_READ) |
fe6e9c1f2
|
1075 |
perm |= MAY_READ; |
572c48921
|
1076 |
if (mode & FMODE_WRITE) |
fe6e9c1f2
|
1077 1078 1079 1080 |
perm |= MAY_WRITE; /* * hooks: /n/, see "layering violations". */ |
b7300b78d
|
1081 1082 1083 1084 1085 1086 |
if (!for_part) { ret = devcgroup_inode_permission(bdev->bd_inode, perm); if (ret != 0) { bdput(bdev); return ret; } |
826660203
|
1087 |
} |
7db9cfd38
|
1088 |
|
d3374825c
|
1089 |
restart: |
0762b8bde
|
1090 |
|
89f97496e
|
1091 |
ret = -ENXIO; |
cf771cb5a
|
1092 |
disk = get_gendisk(bdev->bd_dev, &partno); |
0762b8bde
|
1093 |
if (!disk) |
6e9624b8c
|
1094 |
goto out; |
523e1d399
|
1095 |
owner = disk->fops->owner; |
1da177e4c
|
1096 |
|
69e02c59a
|
1097 |
disk_block_events(disk); |
6796bf54a
|
1098 |
mutex_lock_nested(&bdev->bd_mutex, for_part); |
1da177e4c
|
1099 1100 1101 |
if (!bdev->bd_openers) { bdev->bd_disk = disk; bdev->bd_contains = bdev; |
cf771cb5a
|
1102 |
if (!partno) { |
1da177e4c
|
1103 |
struct backing_dev_info *bdi; |
89f97496e
|
1104 1105 1106 1107 1108 |
ret = -ENXIO; bdev->bd_part = disk_get_part(disk, partno); if (!bdev->bd_part) goto out_clear; |
1196f8b81
|
1109 |
ret = 0; |
1da177e4c
|
1110 |
if (disk->fops->open) { |
572c48921
|
1111 |
ret = disk->fops->open(bdev, mode); |
d3374825c
|
1112 1113 1114 1115 1116 1117 1118 |
if (ret == -ERESTARTSYS) { /* Lost a race with 'disk' being * deleted, try again. * See md.c */ disk_put_part(bdev->bd_part); bdev->bd_part = NULL; |
d3374825c
|
1119 1120 |
bdev->bd_disk = NULL; mutex_unlock(&bdev->bd_mutex); |
69e02c59a
|
1121 |
disk_unblock_events(disk); |
69e02c59a
|
1122 |
put_disk(disk); |
523e1d399
|
1123 |
module_put(owner); |
d3374825c
|
1124 1125 |
goto restart; } |
1da177e4c
|
1126 |
} |
7e69723fe
|
1127 1128 1129 1130 1131 1132 1133 1134 |
if (!ret && !bdev->bd_openers) { bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); bdi = blk_get_backing_dev_info(bdev); if (bdi == NULL) bdi = &default_backing_dev_info; bdev_inode_switch_bdi(bdev->bd_inode, bdi); } |
1196f8b81
|
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 |
/* * If the device is invalidated, rescan partition * if open succeeded or failed with -ENOMEDIUM. * The latter is necessary to prevent ghost * partitions on a removed medium. */ if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) rescan_partitions(disk, bdev); if (ret) goto out_clear; |
1da177e4c
|
1145 |
} else { |
1da177e4c
|
1146 1147 1148 1149 |
struct block_device *whole; whole = bdget_disk(disk, 0); ret = -ENOMEM; if (!whole) |
0762b8bde
|
1150 |
goto out_clear; |
37be41241
|
1151 |
BUG_ON(for_part); |
572c48921
|
1152 |
ret = __blkdev_get(whole, mode, 1); |
1da177e4c
|
1153 |
if (ret) |
0762b8bde
|
1154 |
goto out_clear; |
1da177e4c
|
1155 |
bdev->bd_contains = whole; |
a5491e0c7
|
1156 1157 |
bdev_inode_switch_bdi(bdev->bd_inode, whole->bd_inode->i_data.backing_dev_info); |
89f97496e
|
1158 |
bdev->bd_part = disk_get_part(disk, partno); |
e71bf0d0e
|
1159 |
if (!(disk->flags & GENHD_FL_UP) || |
89f97496e
|
1160 |
!bdev->bd_part || !bdev->bd_part->nr_sects) { |
1da177e4c
|
1161 |
ret = -ENXIO; |
0762b8bde
|
1162 |
goto out_clear; |
1da177e4c
|
1163 |
} |
89f97496e
|
1164 |
bd_set_size(bdev, (loff_t)bdev->bd_part->nr_sects << 9); |
1da177e4c
|
1165 1166 |
} } else { |
1da177e4c
|
1167 |
if (bdev->bd_contains == bdev) { |
1196f8b81
|
1168 1169 |
ret = 0; if (bdev->bd_disk->fops->open) |
572c48921
|
1170 |
ret = bdev->bd_disk->fops->open(bdev, mode); |
1196f8b81
|
1171 1172 |
/* the same as first opener case, read comment there */ if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) |
1da177e4c
|
1173 |
rescan_partitions(bdev->bd_disk, bdev); |
1196f8b81
|
1174 1175 |
if (ret) goto out_unlock_bdev; |
1da177e4c
|
1176 |
} |
69e02c59a
|
1177 |
/* only one opener holds refs to the module and disk */ |
69e02c59a
|
1178 |
put_disk(disk); |
523e1d399
|
1179 |
module_put(owner); |
1da177e4c
|
1180 1181 |
} bdev->bd_openers++; |
37be41241
|
1182 1183 |
if (for_part) bdev->bd_part_count++; |
c039e3134
|
1184 |
mutex_unlock(&bdev->bd_mutex); |
69e02c59a
|
1185 |
disk_unblock_events(disk); |
1da177e4c
|
1186 |
return 0; |
0762b8bde
|
1187 |
out_clear: |
89f97496e
|
1188 |
disk_put_part(bdev->bd_part); |
1da177e4c
|
1189 |
bdev->bd_disk = NULL; |
0762b8bde
|
1190 |
bdev->bd_part = NULL; |
a5491e0c7
|
1191 |
bdev_inode_switch_bdi(bdev->bd_inode, &default_backing_dev_info); |
1da177e4c
|
1192 |
if (bdev != bdev->bd_contains) |
572c48921
|
1193 |
__blkdev_put(bdev->bd_contains, mode, 1); |
1da177e4c
|
1194 |
bdev->bd_contains = NULL; |
0762b8bde
|
1195 |
out_unlock_bdev: |
c039e3134
|
1196 |
mutex_unlock(&bdev->bd_mutex); |
69e02c59a
|
1197 |
disk_unblock_events(disk); |
0762b8bde
|
1198 |
put_disk(disk); |
523e1d399
|
1199 |
module_put(owner); |
4345caba3
|
1200 |
out: |
0762b8bde
|
1201 |
bdput(bdev); |
1da177e4c
|
1202 1203 |
return ret; } |
d4d776299
|
1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 |
/** * blkdev_get - open a block device * @bdev: block_device to open * @mode: FMODE_* mask * @holder: exclusive holder identifier * * Open @bdev with @mode. If @mode includes %FMODE_EXCL, @bdev is * open with exclusive access. Specifying %FMODE_EXCL with %NULL * @holder is invalid. Exclusive opens may nest for the same @holder. * * On success, the reference count of @bdev is unchanged. On failure, * @bdev is put. * * CONTEXT: * Might sleep. * * RETURNS: * 0 on success, -errno on failure. */ |
e525fd89d
|
1223 |
int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) |
1da177e4c
|
1224 |
{ |
e525fd89d
|
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 |
struct block_device *whole = NULL; int res; WARN_ON_ONCE((mode & FMODE_EXCL) && !holder); if ((mode & FMODE_EXCL) && holder) { whole = bd_start_claiming(bdev, holder); if (IS_ERR(whole)) { bdput(bdev); return PTR_ERR(whole); } } res = __blkdev_get(bdev, mode, 0); if (whole) { |
d4dc210f6
|
1241 |
struct gendisk *disk = whole->bd_disk; |
6a027eff6
|
1242 |
/* finish claiming */ |
77ea887e4
|
1243 |
mutex_lock(&bdev->bd_mutex); |
6a027eff6
|
1244 |
spin_lock(&bdev_lock); |
77ea887e4
|
1245 |
if (!res) { |
6a027eff6
|
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 |
BUG_ON(!bd_may_claim(bdev, whole, holder)); /* * Note that for a whole device bd_holders * will be incremented twice, and bd_holder * will be set to bd_may_claim before being * set to holder */ whole->bd_holders++; whole->bd_holder = bd_may_claim; bdev->bd_holders++; bdev->bd_holder = holder; } /* tell others that we're done */ BUG_ON(whole->bd_claiming != holder); whole->bd_claiming = NULL; wake_up_bit(&whole->bd_claiming, 0); spin_unlock(&bdev_lock); |
77ea887e4
|
1265 1266 |
/* |
d4dc210f6
|
1267 1268 1269 1270 1271 |
* Block event polling for write claims if requested. Any * write holder makes the write_holder state stick until * all are released. This is good enough and tracking * individual writeable reference is too fragile given the * way @mode is used in blkdev_get/put(). |
77ea887e4
|
1272 |
*/ |
4c49ff3fe
|
1273 1274 |
if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder && (disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE)) { |
77ea887e4
|
1275 |
bdev->bd_write_holder = true; |
d4dc210f6
|
1276 |
disk_block_events(disk); |
77ea887e4
|
1277 1278 1279 |
} mutex_unlock(&bdev->bd_mutex); |
6a027eff6
|
1280 |
bdput(whole); |
e525fd89d
|
1281 1282 1283 |
} return res; |
37be41241
|
1284 |
} |
1da177e4c
|
1285 |
EXPORT_SYMBOL(blkdev_get); |
d4d776299
|
1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 |
/** * blkdev_get_by_path - open a block device by name * @path: path to the block device to open * @mode: FMODE_* mask * @holder: exclusive holder identifier * * Open the blockdevice described by the device file at @path. @mode * and @holder are identical to blkdev_get(). * * On success, the returned block_device has reference count of one. * * CONTEXT: * Might sleep. * * RETURNS: * Pointer to block_device on success, ERR_PTR(-errno) on failure. */ struct block_device *blkdev_get_by_path(const char *path, fmode_t mode, void *holder) { struct block_device *bdev; int err; bdev = lookup_bdev(path); if (IS_ERR(bdev)) return bdev; err = blkdev_get(bdev, mode, holder); if (err) return ERR_PTR(err); |
e51900f7d
|
1316 1317 1318 1319 |
if ((mode & FMODE_WRITE) && bdev_read_only(bdev)) { blkdev_put(bdev, mode); return ERR_PTR(-EACCES); } |
d4d776299
|
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 |
return bdev; } EXPORT_SYMBOL(blkdev_get_by_path); /** * blkdev_get_by_dev - open a block device by device number * @dev: device number of block device to open * @mode: FMODE_* mask * @holder: exclusive holder identifier * * Open the blockdevice described by device number @dev. @mode and * @holder are identical to blkdev_get(). * * Use it ONLY if you really do not have anything better - i.e. when * you are behind a truly sucky interface and all you are given is a * device number. _Never_ to be used for internal purposes. If you * ever need it - reconsider your API. * * On success, the returned block_device has reference count of one. * * CONTEXT: * Might sleep. * * RETURNS: * Pointer to block_device on success, ERR_PTR(-errno) on failure. */ struct block_device *blkdev_get_by_dev(dev_t dev, fmode_t mode, void *holder) { struct block_device *bdev; int err; bdev = bdget(dev); if (!bdev) return ERR_PTR(-ENOMEM); err = blkdev_get(bdev, mode, holder); if (err) return ERR_PTR(err); return bdev; } EXPORT_SYMBOL(blkdev_get_by_dev); |
1da177e4c
|
1362 1363 1364 |
static int blkdev_open(struct inode * inode, struct file * filp) { struct block_device *bdev; |
1da177e4c
|
1365 1366 1367 1368 1369 1370 1371 1372 |
/* * Preserve backwards compatibility and allow large file access * even if userspace doesn't ask for it explicitly. Some mkfs * binary needs it. We might want to drop this workaround * during an unstable branch. */ filp->f_flags |= O_LARGEFILE; |
572c48921
|
1373 1374 1375 1376 1377 1378 |
if (filp->f_flags & O_NDELAY) filp->f_mode |= FMODE_NDELAY; if (filp->f_flags & O_EXCL) filp->f_mode |= FMODE_EXCL; if ((filp->f_flags & O_ACCMODE) == 3) filp->f_mode |= FMODE_WRITE_IOCTL; |
1da177e4c
|
1379 |
bdev = bd_acquire(inode); |
6a2aae06c
|
1380 1381 |
if (bdev == NULL) return -ENOMEM; |
1da177e4c
|
1382 |
|
572c48921
|
1383 |
filp->f_mapping = bdev->bd_inode->i_mapping; |
e525fd89d
|
1384 |
return blkdev_get(bdev, filp->f_mode, filp); |
1da177e4c
|
1385 |
} |
9a1c35427
|
1386 |
static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) |
2e7b651df
|
1387 1388 |
{ int ret = 0; |
2e7b651df
|
1389 |
struct gendisk *disk = bdev->bd_disk; |
37be41241
|
1390 |
struct block_device *victim = NULL; |
2e7b651df
|
1391 |
|
6796bf54a
|
1392 |
mutex_lock_nested(&bdev->bd_mutex, for_part); |
37be41241
|
1393 1394 |
if (for_part) bdev->bd_part_count--; |
2e7b651df
|
1395 |
if (!--bdev->bd_openers) { |
6a027eff6
|
1396 |
WARN_ON_ONCE(bdev->bd_holders); |
2e7b651df
|
1397 1398 |
sync_blockdev(bdev); kill_bdev(bdev); |
94007751b
|
1399 1400 1401 1402 1403 |
/* ->release can cause the old bdi to disappear, * so must switch it out first */ bdev_inode_switch_bdi(bdev->bd_inode, &default_backing_dev_info); |
2e7b651df
|
1404 1405 1406 |
} if (bdev->bd_contains == bdev) { if (disk->fops->release) |
9a1c35427
|
1407 |
ret = disk->fops->release(disk, mode); |
2e7b651df
|
1408 1409 1410 |
} if (!bdev->bd_openers) { struct module *owner = disk->fops->owner; |
0762b8bde
|
1411 1412 |
disk_put_part(bdev->bd_part); bdev->bd_part = NULL; |
2e7b651df
|
1413 |
bdev->bd_disk = NULL; |
37be41241
|
1414 1415 |
if (bdev != bdev->bd_contains) victim = bdev->bd_contains; |
2e7b651df
|
1416 |
bdev->bd_contains = NULL; |
523e1d399
|
1417 1418 1419 |
put_disk(disk); module_put(owner); |
2e7b651df
|
1420 |
} |
2e7b651df
|
1421 1422 |
mutex_unlock(&bdev->bd_mutex); bdput(bdev); |
37be41241
|
1423 |
if (victim) |
9a1c35427
|
1424 |
__blkdev_put(victim, mode, 1); |
2e7b651df
|
1425 1426 |
return ret; } |
9a1c35427
|
1427 |
int blkdev_put(struct block_device *bdev, fmode_t mode) |
37be41241
|
1428 |
{ |
85ef06d1d
|
1429 |
mutex_lock(&bdev->bd_mutex); |
e525fd89d
|
1430 |
if (mode & FMODE_EXCL) { |
6a027eff6
|
1431 1432 1433 1434 1435 1436 1437 |
bool bdev_free; /* * Release a claim on the device. The holder fields * are protected with bdev_lock. bd_mutex is to * synchronize disk_holder unlinking. */ |
6a027eff6
|
1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 |
spin_lock(&bdev_lock); WARN_ON_ONCE(--bdev->bd_holders < 0); WARN_ON_ONCE(--bdev->bd_contains->bd_holders < 0); /* bd_contains might point to self, check in a separate step */ if ((bdev_free = !bdev->bd_holders)) bdev->bd_holder = NULL; if (!bdev->bd_contains->bd_holders) bdev->bd_contains->bd_holder = NULL; spin_unlock(&bdev_lock); |
77ea887e4
|
1450 1451 1452 1453 |
/* * If this was the last claim, remove holder link and * unblock evpoll if it was a write holder. */ |
85ef06d1d
|
1454 1455 1456 |
if (bdev_free && bdev->bd_write_holder) { disk_unblock_events(bdev->bd_disk); bdev->bd_write_holder = false; |
77ea887e4
|
1457 |
} |
6936217cc
|
1458 |
} |
77ea887e4
|
1459 |
|
85ef06d1d
|
1460 1461 1462 1463 1464 1465 1466 1467 |
/* * Trigger event checking and tell drivers to flush MEDIA_CHANGE * event. This is to ensure detection of media removal commanded * from userland - e.g. eject(1). */ disk_flush_events(bdev->bd_disk, DISK_EVENT_MEDIA_CHANGE); mutex_unlock(&bdev->bd_mutex); |
9a1c35427
|
1468 |
return __blkdev_put(bdev, mode, 0); |
37be41241
|
1469 |
} |
2e7b651df
|
1470 |
EXPORT_SYMBOL(blkdev_put); |
1da177e4c
|
1471 1472 1473 |
static int blkdev_close(struct inode * inode, struct file * filp) { struct block_device *bdev = I_BDEV(filp->f_mapping->host); |
e525fd89d
|
1474 |
|
9a1c35427
|
1475 |
return blkdev_put(bdev, filp->f_mode); |
1da177e4c
|
1476 |
} |
bb93e3a52
|
1477 |
static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) |
1da177e4c
|
1478 |
{ |
56b26add0
|
1479 1480 |
struct block_device *bdev = I_BDEV(file->f_mapping->host); fmode_t mode = file->f_mode; |
fd4ce1acd
|
1481 1482 1483 1484 1485 |
/* * O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have * to updated it before every ioctl. */ |
56b26add0
|
1486 |
if (file->f_flags & O_NDELAY) |
fd4ce1acd
|
1487 1488 1489 |
mode |= FMODE_NDELAY; else mode &= ~FMODE_NDELAY; |
56b26add0
|
1490 |
return blkdev_ioctl(bdev, mode, cmd, arg); |
1da177e4c
|
1491 |
} |
87d8fe1ee
|
1492 |
/* |
eef993806
|
1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 |
* Write data to the block device. Only intended for the block device itself * and the raw driver which basically is a fake block device. * * Does not take i_mutex for the write and thus is not for general purpose * use. */ ssize_t blkdev_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct file *file = iocb->ki_filp; ssize_t ret; BUG_ON(iocb->ki_pos != pos); ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); if (ret > 0 || ret == -EIOCBQUEUED) { ssize_t err; err = generic_write_sync(file, pos, ret); if (err < 0 && ret > 0) ret = err; } return ret; } EXPORT_SYMBOL_GPL(blkdev_aio_write); /* |
87d8fe1ee
|
1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 |
* Try to release a page associated with block device when the system * is under memory pressure. */ static int blkdev_releasepage(struct page *page, gfp_t wait) { struct super_block *super = BDEV_I(page->mapping->host)->bdev.bd_super; if (super && super->s_op->bdev_try_to_free_page) return super->s_op->bdev_try_to_free_page(super, page, wait); return try_to_free_buffers(page); } |
4c54ac62d
|
1532 |
static const struct address_space_operations def_blk_aops = { |
1da177e4c
|
1533 1534 |
.readpage = blkdev_readpage, .writepage = blkdev_writepage, |
6272b5a58
|
1535 1536 |
.write_begin = blkdev_write_begin, .write_end = blkdev_write_end, |
1da177e4c
|
1537 |
.writepages = generic_writepages, |
87d8fe1ee
|
1538 |
.releasepage = blkdev_releasepage, |
1da177e4c
|
1539 1540 |
.direct_IO = blkdev_direct_IO, }; |
4b6f5d20b
|
1541 |
const struct file_operations def_blk_fops = { |
1da177e4c
|
1542 1543 1544 |
.open = blkdev_open, .release = blkdev_close, .llseek = block_llseek, |
543ade1fc
|
1545 1546 |
.read = do_sync_read, .write = do_sync_write, |
1da177e4c
|
1547 |
.aio_read = generic_file_aio_read, |
eef993806
|
1548 |
.aio_write = blkdev_aio_write, |
1da177e4c
|
1549 |
.mmap = generic_file_mmap, |
b1dd3b284
|
1550 |
.fsync = blkdev_fsync, |
bb93e3a52
|
1551 |
.unlocked_ioctl = block_ioctl, |
1da177e4c
|
1552 1553 1554 |
#ifdef CONFIG_COMPAT .compat_ioctl = compat_blkdev_ioctl, #endif |
7f9c51f0d
|
1555 1556 |
.splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, |
1da177e4c
|
1557 1558 1559 1560 1561 1562 1563 |
}; int ioctl_by_bdev(struct block_device *bdev, unsigned cmd, unsigned long arg) { int res; mm_segment_t old_fs = get_fs(); set_fs(KERNEL_DS); |
56b26add0
|
1564 |
res = blkdev_ioctl(bdev, 0, cmd, arg); |
1da177e4c
|
1565 1566 1567 1568 1569 1570 1571 1572 |
set_fs(old_fs); return res; } EXPORT_SYMBOL(ioctl_by_bdev); /** * lookup_bdev - lookup a struct block_device by name |
94e2959e7
|
1573 |
* @pathname: special file representing the block device |
1da177e4c
|
1574 |
* |
57d1b5366
|
1575 |
* Get a reference to the blockdevice at @pathname in the current |
1da177e4c
|
1576 1577 1578 |
* namespace if possible and return it. Return ERR_PTR(error) * otherwise. */ |
421748ecd
|
1579 |
struct block_device *lookup_bdev(const char *pathname) |
1da177e4c
|
1580 1581 1582 |
{ struct block_device *bdev; struct inode *inode; |
421748ecd
|
1583 |
struct path path; |
1da177e4c
|
1584 |
int error; |
421748ecd
|
1585 |
if (!pathname || !*pathname) |
1da177e4c
|
1586 |
return ERR_PTR(-EINVAL); |
421748ecd
|
1587 |
error = kern_path(pathname, LOOKUP_FOLLOW, &path); |
1da177e4c
|
1588 1589 |
if (error) return ERR_PTR(error); |
421748ecd
|
1590 |
inode = path.dentry->d_inode; |
1da177e4c
|
1591 1592 1593 1594 |
error = -ENOTBLK; if (!S_ISBLK(inode->i_mode)) goto fail; error = -EACCES; |
421748ecd
|
1595 |
if (path.mnt->mnt_flags & MNT_NODEV) |
1da177e4c
|
1596 1597 1598 1599 1600 1601 |
goto fail; error = -ENOMEM; bdev = bd_acquire(inode); if (!bdev) goto fail; out: |
421748ecd
|
1602 |
path_put(&path); |
1da177e4c
|
1603 1604 1605 1606 1607 |
return bdev; fail: bdev = ERR_PTR(error); goto out; } |
d5686b444
|
1608 |
EXPORT_SYMBOL(lookup_bdev); |
1da177e4c
|
1609 |
|
93b270f76
|
1610 |
int __invalidate_device(struct block_device *bdev, bool kill_dirty) |
b71e8a4ce
|
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 |
{ struct super_block *sb = get_super(bdev); int res = 0; if (sb) { /* * no need to lock the super, get_super holds the * read mutex so the filesystem cannot go away * under us (->put_super runs with the write lock * hold). */ shrink_dcache_sb(sb); |
93b270f76
|
1623 |
res = invalidate_inodes(sb, kill_dirty); |
b71e8a4ce
|
1624 1625 |
drop_super(sb); } |
f98393a64
|
1626 |
invalidate_bdev(bdev); |
b71e8a4ce
|
1627 1628 1629 |
return res; } EXPORT_SYMBOL(__invalidate_device); |