extent_io.h 17.2 KB
Newer Older
1 2 3 4
#ifndef __EXTENTIO__
#define __EXTENTIO__

#include <linux/rbtree.h>
5
#include "ulist.h"
6 7

/* bits for the extent state */
8 9 10 11 12 13 14 15 16 17 18 19 20 21
#define EXTENT_DIRTY		(1U << 0)
#define EXTENT_WRITEBACK	(1U << 1)
#define EXTENT_UPTODATE		(1U << 2)
#define EXTENT_LOCKED		(1U << 3)
#define EXTENT_NEW		(1U << 4)
#define EXTENT_DELALLOC		(1U << 5)
#define EXTENT_DEFRAG		(1U << 6)
#define EXTENT_BOUNDARY		(1U << 9)
#define EXTENT_NODATASUM	(1U << 10)
#define EXTENT_DO_ACCOUNTING	(1U << 11)
#define EXTENT_FIRST_DELALLOC	(1U << 12)
#define EXTENT_NEED_WAIT	(1U << 13)
#define EXTENT_DAMAGED		(1U << 14)
#define EXTENT_NORESERVE	(1U << 15)
22
#define EXTENT_QGROUP_RESERVED	(1U << 16)
23
#define EXTENT_CLEAR_DATA_RESV	(1U << 17)
24 25
#define EXTENT_IOBITS		(EXTENT_LOCKED | EXTENT_WRITEBACK)
#define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
26

27 28 29 30
/*
 * flags for bio submission. The high bits indicate the compression
 * type for this bio
 */
C
Chris Mason 已提交
31
#define EXTENT_BIO_COMPRESSED 1
32
#define EXTENT_BIO_TREE_LOG 2
33
#define EXTENT_BIO_FLAG_SHIFT 16
C
Chris Mason 已提交
34

35 36
/* these are bit numbers for test/set bit */
#define EXTENT_BUFFER_UPTODATE 0
37
#define EXTENT_BUFFER_DIRTY 2
38
#define EXTENT_BUFFER_CORRUPT 3
39
#define EXTENT_BUFFER_READAHEAD 4	/* this got triggered by readahead */
40 41
#define EXTENT_BUFFER_TREE_REF 5
#define EXTENT_BUFFER_STALE 6
42
#define EXTENT_BUFFER_WRITEBACK 7
43
#define EXTENT_BUFFER_READ_ERR 8        /* read IO error */
44
#define EXTENT_BUFFER_DUMMY 9
45
#define EXTENT_BUFFER_IN_TREE 10
46
#define EXTENT_BUFFER_WRITE_ERR 11    /* write IO error */
47

48
/* these are flags for __process_pages_contig */
49 50 51 52 53
#define PAGE_UNLOCK		(1 << 0)
#define PAGE_CLEAR_DIRTY	(1 << 1)
#define PAGE_SET_WRITEBACK	(1 << 2)
#define PAGE_END_WRITEBACK	(1 << 3)
#define PAGE_SET_PRIVATE2	(1 << 4)
54
#define PAGE_SET_ERROR		(1 << 5)
55
#define PAGE_LOCK		(1 << 6)
56

57 58 59 60 61 62
/*
 * page->private values.  Every page that is controlled by the extent
 * map has page->private set to one.
 */
#define EXTENT_PAGE_PRIVATE 1

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
/*
 * The extent buffer bitmap operations are done with byte granularity instead of
 * word granularity for two reasons:
 * 1. The bitmaps must be little-endian on disk.
 * 2. Bitmap items are not guaranteed to be aligned to a word and therefore a
 *    single word in a bitmap may straddle two pages in the extent buffer.
 */
#define BIT_BYTE(nr) ((nr) / BITS_PER_BYTE)
#define BYTE_MASK ((1 << BITS_PER_BYTE) - 1)
#define BITMAP_FIRST_BYTE_MASK(start) \
	((BYTE_MASK << ((start) & (BITS_PER_BYTE - 1))) & BYTE_MASK)
#define BITMAP_LAST_BYTE_MASK(nbits) \
	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))

static inline int le_test_bit(int nr, const u8 *addr)
{
	return 1U & (addr[BIT_BYTE(nr)] >> (nr & (BITS_PER_BYTE-1)));
}

extern void le_bitmap_set(u8 *map, unsigned int start, int len);
extern void le_bitmap_clear(u8 *map, unsigned int start, int len);

85
struct extent_state;
86
struct btrfs_root;
87
struct btrfs_io_bio;
88
struct io_failure_record;
89

90 91 92
typedef	int (extent_submit_bio_hook_t)(struct inode *inode, struct bio *bio,
				       int mirror_num, unsigned long bio_flags,
				       u64 bio_offset);
93
struct extent_io_ops {
C
Chris Mason 已提交
94
	int (*fill_delalloc)(struct inode *inode, struct page *locked_page,
95 96
			     u64 start, u64 end, int *page_started,
			     unsigned long *nr_written);
97
	int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
98
	extent_submit_bio_hook_t *submit_bio_hook;
99
	int (*merge_bio_hook)(struct page *page, unsigned long offset,
C
Chris Mason 已提交
100 101
			      size_t size, struct bio *bio,
			      unsigned long bio_flags);
102
	int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
103 104 105
	int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
				    struct page *page, u64 start, u64 end,
				    int mirror);
106
	int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
107
				      struct extent_state *state, int uptodate);
108
	void (*set_bit_hook)(struct inode *inode, struct extent_state *state,
109
			     unsigned *bits);
110
	void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
111
			       unsigned *bits);
112 113 114 115 116
	void (*merge_extent_hook)(struct inode *inode,
				  struct extent_state *new,
				  struct extent_state *other);
	void (*split_extent_hook)(struct inode *inode,
				  struct extent_state *orig, u64 split);
117 118 119 120 121 122
};

struct extent_io_tree {
	struct rb_root state;
	struct address_space *mapping;
	u64 dirty_bytes;
123
	int track_uptodate;
124
	spinlock_t lock;
125
	const struct extent_io_ops *ops;
126 127 128 129 130 131
};

struct extent_state {
	u64 start;
	u64 end; /* inclusive */
	struct rb_node rb_node;
J
Josef Bacik 已提交
132 133

	/* ADD NEW ELEMENTS AFTER THIS */
134 135
	wait_queue_head_t wq;
	atomic_t refs;
136
	unsigned state;
137

138
	struct io_failure_record *failrec;
139

140
#ifdef CONFIG_BTRFS_DEBUG
141
	struct list_head leak_list;
142
#endif
143 144
};

145
#define INLINE_EXTENT_BUFFER_PAGES 16
146
#define MAX_INLINE_EXTENT_BUFFER_SIZE (INLINE_EXTENT_BUFFER_PAGES * PAGE_SIZE)
147 148 149
struct extent_buffer {
	u64 start;
	unsigned long len;
150
	unsigned long bflags;
151
	struct btrfs_fs_info *fs_info;
152
	spinlock_t refs_lock;
153
	atomic_t refs;
154
	atomic_t io_pages;
155
	int read_mirror;
156
	struct rcu_head rcu_head;
157
	pid_t lock_owner;
158

159 160 161 162 163 164 165
	/* count of read lock holders on the extent buffer */
	atomic_t write_locks;
	atomic_t read_locks;
	atomic_t blocking_writers;
	atomic_t blocking_readers;
	atomic_t spinning_readers;
	atomic_t spinning_writers;
166 167 168
	short lock_nested;
	/* >= 0 if eb belongs to a log tree, -1 otherwise */
	short log_index;
169 170 171 172 173 174 175 176

	/* protects write locks */
	rwlock_t lock;

	/* readers use lock_wq while they wait for the write
	 * lock holders to unlock
	 */
	wait_queue_head_t write_lock_wq;
177

178 179
	/* writers use read_lock_wq while they wait for readers
	 * to unlock
180
	 */
181
	wait_queue_head_t read_lock_wq;
182
	struct page *pages[INLINE_EXTENT_BUFFER_PAGES];
183 184 185
#ifdef CONFIG_BTRFS_DEBUG
	struct list_head leak_list;
#endif
186 187
};

188 189 190 191 192 193 194 195 196 197 198
/*
 * Structure to record how many bytes and which ranges are set/cleared
 */
struct extent_changeset {
	/* How many bytes are set/cleared in this operation */
	u64 bytes_changed;

	/* Changed ranges */
	struct ulist *range_changed;
};

199 200 201 202 203 204 205 206 207 208 209
static inline void extent_set_compress_type(unsigned long *bio_flags,
					    int compress_type)
{
	*bio_flags |= compress_type << EXTENT_BIO_FLAG_SHIFT;
}

static inline int extent_compress_type(unsigned long bio_flags)
{
	return bio_flags >> EXTENT_BIO_FLAG_SHIFT;
}

210 211 212 213
struct extent_map_tree;

typedef struct extent_map *(get_extent_t)(struct inode *inode,
					  struct page *page,
214
					  size_t pg_offset,
215 216 217 218
					  u64 start, u64 len,
					  int create);

void extent_io_tree_init(struct extent_io_tree *tree,
219
			 struct address_space *mapping);
220
int try_release_extent_mapping(struct extent_map_tree *map,
221 222
			       struct extent_io_tree *tree, struct page *page,
			       gfp_t mask);
223
int try_release_extent_buffer(struct page *page);
224
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
225
		     struct extent_state **cached);
226 227 228 229 230 231

static inline int lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
{
	return lock_extent_bits(tree, start, end, NULL);
}

232
int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
233
int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
234
			  get_extent_t *get_extent, int mirror_num);
235 236 237 238 239
int __init extent_io_init(void);
void extent_io_exit(void);

u64 count_range_bits(struct extent_io_tree *tree,
		     u64 *start, u64 search_end,
240
		     u64 max_bytes, unsigned bits, int contig);
241

242
void free_extent_state(struct extent_state *state);
243
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
244
		   unsigned bits, int filled,
245
		   struct extent_state *cached_state);
246
int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
247
		unsigned bits, struct extent_changeset *changeset);
248
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
249
		     unsigned bits, int wake, int delete,
250
		     struct extent_state **cached, gfp_t mask);
251

252 253 254 255 256 257 258 259 260 261 262 263 264 265
static inline int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
{
	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL,
				GFP_NOFS);
}

static inline int unlock_extent_cached(struct extent_io_tree *tree, u64 start,
		u64 end, struct extent_state **cached, gfp_t mask)
{
	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
				mask);
}

static inline int clear_extent_bits(struct extent_io_tree *tree, u64 start,
266
		u64 end, unsigned bits)
267 268 269 270 271 272
{
	int wake = 0;

	if (bits & EXTENT_LOCKED)
		wake = 1;

273 274
	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL,
			GFP_NOFS);
275 276
}

277
int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
278
			   unsigned bits, struct extent_changeset *changeset);
279
int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
280
		   unsigned bits, u64 *failed_start,
281
		   struct extent_state **cached_state, gfp_t mask);
282 283

static inline int set_extent_bits(struct extent_io_tree *tree, u64 start,
284
		u64 end, unsigned bits)
285
{
286
	return set_extent_bit(tree, start, end, bits, NULL, NULL, GFP_NOFS);
287 288
}

289 290 291 292 293 294
static inline int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
		u64 end, struct extent_state **cached_state, gfp_t mask)
{
	return clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
				cached_state, mask);
}
295 296 297 298 299 300 301 302

static inline int set_extent_dirty(struct extent_io_tree *tree, u64 start,
		u64 end, gfp_t mask)
{
	return set_extent_bit(tree, start, end, EXTENT_DIRTY, NULL,
			      NULL, mask);
}

303
static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
304
		u64 end)
305 306 307
{
	return clear_extent_bit(tree, start, end,
				EXTENT_DIRTY | EXTENT_DELALLOC |
308
				EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS);
309 310
}

J
Josef Bacik 已提交
311
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
312
		       unsigned bits, unsigned clear_bits,
313
		       struct extent_state **cached_state);
314 315

static inline int set_extent_delalloc(struct extent_io_tree *tree, u64 start,
316
		u64 end, struct extent_state **cached_state)
317 318 319
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE,
320
			      NULL, cached_state, GFP_NOFS);
321 322 323
}

static inline int set_extent_defrag(struct extent_io_tree *tree, u64 start,
324
		u64 end, struct extent_state **cached_state)
325 326 327
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
328
			      NULL, cached_state, GFP_NOFS);
329 330 331
}

static inline int set_extent_new(struct extent_io_tree *tree, u64 start,
332
		u64 end)
333
{
334 335
	return set_extent_bit(tree, start, end, EXTENT_NEW, NULL, NULL,
			GFP_NOFS);
336 337 338 339 340 341 342 343 344
}

static inline int set_extent_uptodate(struct extent_io_tree *tree, u64 start,
		u64 end, struct extent_state **cached_state, gfp_t mask)
{
	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, NULL,
			      cached_state, mask);
}

345
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
346
			  u64 *start_ret, u64 *end_ret, unsigned bits,
347
			  struct extent_state **cached_state);
348 349 350 351 352
int extent_invalidatepage(struct extent_io_tree *tree,
			  struct page *page, unsigned long offset);
int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
			  get_extent_t *get_extent,
			  struct writeback_control *wbc);
353 354 355
int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
			      u64 start, u64 end, get_extent_t *get_extent,
			      int mode);
356 357 358 359
int extent_writepages(struct extent_io_tree *tree,
		      struct address_space *mapping,
		      get_extent_t *get_extent,
		      struct writeback_control *wbc);
360 361
int btree_write_cache_pages(struct address_space *mapping,
			    struct writeback_control *wbc);
362 363 364 365
int extent_readpages(struct extent_io_tree *tree,
		     struct address_space *mapping,
		     struct list_head *pages, unsigned nr_pages,
		     get_extent_t get_extent);
Y
Yehuda Sadeh 已提交
366 367
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
		__u64 start, __u64 len, get_extent_t *get_extent);
368 369
void set_page_extent_mapped(struct page *page);

370
struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
371
					  u64 start);
372 373
struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
						  u64 start, unsigned long len);
374
struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
375
						u64 start);
376
struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
377
struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
378
					 u64 start);
379
void free_extent_buffer(struct extent_buffer *eb);
380
void free_extent_buffer_stale(struct extent_buffer *eb);
381 382 383
#define WAIT_NONE	0
#define WAIT_COMPLETE	1
#define WAIT_PAGE_LOCK	2
384
int read_extent_buffer_pages(struct extent_io_tree *tree,
385
			     struct extent_buffer *eb, int wait,
386
			     get_extent_t *get_extent, int mirror_num);
387
void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
388 389 390

static inline unsigned long num_extent_pages(u64 start, u64 len)
{
391 392
	return ((start + len + PAGE_SIZE - 1) >> PAGE_SHIFT) -
		(start >> PAGE_SHIFT);
393 394
}

395 396 397 398 399 400 401 402 403 404 405
static inline void extent_buffer_get(struct extent_buffer *eb)
{
	atomic_inc(&eb->refs);
}

int memcmp_extent_buffer(struct extent_buffer *eb, const void *ptrv,
			  unsigned long start,
			  unsigned long len);
void read_extent_buffer(struct extent_buffer *eb, void *dst,
			unsigned long start,
			unsigned long len);
406 407 408
int read_extent_buffer_to_user(struct extent_buffer *eb, void __user *dst,
			       unsigned long start,
			       unsigned long len);
409 410 411
void write_extent_buffer_fsid(struct extent_buffer *eb, const void *src);
void write_extent_buffer_chunk_tree_uuid(struct extent_buffer *eb,
		const void *src);
412 413
void write_extent_buffer(struct extent_buffer *eb, const void *src,
			 unsigned long start, unsigned long len);
414 415
void copy_extent_buffer_full(struct extent_buffer *dst,
			     struct extent_buffer *src);
416 417 418 419 420 421 422
void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src,
			unsigned long dst_offset, unsigned long src_offset,
			unsigned long len);
void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
			   unsigned long src_offset, unsigned long len);
void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
			   unsigned long src_offset, unsigned long len);
423 424
void memzero_extent_buffer(struct extent_buffer *eb, unsigned long start,
			   unsigned long len);
425 426 427 428 429 430
int extent_buffer_test_bit(struct extent_buffer *eb, unsigned long start,
			   unsigned long pos);
void extent_buffer_bitmap_set(struct extent_buffer *eb, unsigned long start,
			      unsigned long pos, unsigned long len);
void extent_buffer_bitmap_clear(struct extent_buffer *eb, unsigned long start,
				unsigned long pos, unsigned long len);
431
void clear_extent_buffer_dirty(struct extent_buffer *eb);
432
int set_extent_buffer_dirty(struct extent_buffer *eb);
433
void set_extent_buffer_uptodate(struct extent_buffer *eb);
434
void clear_extent_buffer_uptodate(struct extent_buffer *eb);
435
int extent_buffer_uptodate(struct extent_buffer *eb);
436
int extent_buffer_under_io(struct extent_buffer *eb);
437
int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
438
		      unsigned long min_len, char **map,
439
		      unsigned long *map_start,
440
		      unsigned long *map_len);
441
void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
442
void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
443
void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
444
				 u64 delalloc_end, struct page *locked_page,
445
				 unsigned bits_to_clear,
446
				 unsigned long page_ops);
447 448 449
struct bio *
btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
		gfp_t gfp_flags);
450 451
struct bio *btrfs_io_bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs);
struct bio *btrfs_bio_clone(struct bio *bio, gfp_t gfp_mask);
452

453
struct btrfs_fs_info;
454

455 456 457
int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
		      struct page *page, unsigned int pg_offset,
		      int mirror_num);
458 459
int clean_io_failure(struct inode *inode, u64 start, struct page *page,
		     unsigned int pg_offset);
460
void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
461 462
int repair_eb_io_failure(struct btrfs_fs_info *fs_info,
			 struct extent_buffer *eb, int mirror_num);
463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482

/*
 * When IO fails, either with EIO or csum verification fails, we
 * try other mirrors that might have a good copy of the data.  This
 * io_failure_record is used to record state as we go through all the
 * mirrors.  If another mirror has good data, the page is set up to date
 * and things continue.  If a good mirror can't be found, the original
 * bio end_io callback is called to indicate things have failed.
 */
struct io_failure_record {
	struct page *page;
	u64 start;
	u64 len;
	u64 logical;
	unsigned long bio_flags;
	int this_mirror;
	int failed_mirror;
	int in_validation;
};

483
void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end);
484 485 486 487 488 489 490
int btrfs_get_io_failure_record(struct inode *inode, u64 start, u64 end,
				struct io_failure_record **failrec_ret);
int btrfs_check_repairable(struct inode *inode, struct bio *failed_bio,
			   struct io_failure_record *failrec, int fail_mirror);
struct bio *btrfs_create_repair_bio(struct inode *inode, struct bio *failed_bio,
				    struct io_failure_record *failrec,
				    struct page *page, int pg_offset, int icsum,
491 492
				    bio_end_io_t *endio_func, void *data);
int free_io_failure(struct inode *inode, struct io_failure_record *rec);
493 494 495 496 497
#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
noinline u64 find_lock_delalloc_range(struct inode *inode,
				      struct extent_io_tree *tree,
				      struct page *locked_page, u64 *start,
				      u64 *end, u64 max_bytes);
498
#endif
499
struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
500
					       u64 start);
501
#endif