extent_io.h 16.1 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 24
#define EXTENT_IOBITS		(EXTENT_LOCKED | EXTENT_WRITEBACK)
#define EXTENT_CTLBITS		(EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC)
25

26 27 28 29
/*
 * flags for bio submission. The high bits indicate the compression
 * type for this bio
 */
C
Chris Mason 已提交
30
#define EXTENT_BIO_COMPRESSED 1
31
#define EXTENT_BIO_TREE_LOG 2
32
#define EXTENT_BIO_FLAG_SHIFT 16
C
Chris Mason 已提交
33

34 35
/* these are bit numbers for test/set bit */
#define EXTENT_BUFFER_UPTODATE 0
36
#define EXTENT_BUFFER_DIRTY 2
37
#define EXTENT_BUFFER_CORRUPT 3
38
#define EXTENT_BUFFER_READAHEAD 4	/* this got triggered by readahead */
39 40
#define EXTENT_BUFFER_TREE_REF 5
#define EXTENT_BUFFER_STALE 6
41
#define EXTENT_BUFFER_WRITEBACK 7
42
#define EXTENT_BUFFER_READ_ERR 8        /* read IO error */
43
#define EXTENT_BUFFER_DUMMY 9
44
#define EXTENT_BUFFER_IN_TREE 10
45
#define EXTENT_BUFFER_WRITE_ERR 11    /* write IO error */
46

47
/* these are flags for extent_clear_unlock_delalloc */
48 49 50 51 52
#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)
53
#define PAGE_SET_ERROR		(1 << 5)
54

55 56 57 58 59 60
/*
 * page->private values.  Every page that is controlled by the extent
 * map has page->private set to one.
 */
#define EXTENT_PAGE_PRIVATE 1

61
struct extent_state;
62
struct btrfs_root;
63
struct btrfs_io_bio;
64
struct io_failure_record;
65

66
typedef	int (extent_submit_bio_hook_t)(struct inode *inode, int rw,
C
Chris Mason 已提交
67
				       struct bio *bio, int mirror_num,
68
				       unsigned long bio_flags, u64 bio_offset);
69
struct extent_io_ops {
C
Chris Mason 已提交
70
	int (*fill_delalloc)(struct inode *inode, struct page *locked_page,
71 72
			     u64 start, u64 end, int *page_started,
			     unsigned long *nr_written);
73
	int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
74
	extent_submit_bio_hook_t *submit_bio_hook;
75
	int (*merge_bio_hook)(int rw, struct page *page, unsigned long offset,
C
Chris Mason 已提交
76 77
			      size_t size, struct bio *bio,
			      unsigned long bio_flags);
78
	int (*readpage_io_failed_hook)(struct page *page, int failed_mirror);
79 80 81
	int (*readpage_end_io_hook)(struct btrfs_io_bio *io_bio, u64 phy_offset,
				    struct page *page, u64 start, u64 end,
				    int mirror);
82
	int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end,
83
				      struct extent_state *state, int uptodate);
84
	void (*set_bit_hook)(struct inode *inode, struct extent_state *state,
85
			     unsigned *bits);
86
	void (*clear_bit_hook)(struct inode *inode, struct extent_state *state,
87
			       unsigned *bits);
88 89 90 91 92
	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);
93 94 95 96 97 98
};

struct extent_io_tree {
	struct rb_root state;
	struct address_space *mapping;
	u64 dirty_bytes;
99
	int track_uptodate;
100
	spinlock_t lock;
101
	const struct extent_io_ops *ops;
102 103 104 105 106 107
};

struct extent_state {
	u64 start;
	u64 end; /* inclusive */
	struct rb_node rb_node;
J
Josef Bacik 已提交
108 109

	/* ADD NEW ELEMENTS AFTER THIS */
110 111
	wait_queue_head_t wq;
	atomic_t refs;
112
	unsigned state;
113

114
	struct io_failure_record *failrec;
115

116
#ifdef CONFIG_BTRFS_DEBUG
117
	struct list_head leak_list;
118
#endif
119 120
};

121
#define INLINE_EXTENT_BUFFER_PAGES 16
122
#define MAX_INLINE_EXTENT_BUFFER_SIZE (INLINE_EXTENT_BUFFER_PAGES * PAGE_SIZE)
123 124 125
struct extent_buffer {
	u64 start;
	unsigned long len;
126
	unsigned long bflags;
127
	struct btrfs_fs_info *fs_info;
128
	spinlock_t refs_lock;
129
	atomic_t refs;
130
	atomic_t io_pages;
131
	int read_mirror;
132
	struct rcu_head rcu_head;
133
	pid_t lock_owner;
134

135 136 137 138 139 140 141
	/* 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;
142 143 144
	short lock_nested;
	/* >= 0 if eb belongs to a log tree, -1 otherwise */
	short log_index;
145 146 147 148 149 150 151 152

	/* 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;
153

154 155
	/* writers use read_lock_wq while they wait for readers
	 * to unlock
156
	 */
157
	wait_queue_head_t read_lock_wq;
158
	struct page *pages[INLINE_EXTENT_BUFFER_PAGES];
159 160 161
#ifdef CONFIG_BTRFS_DEBUG
	struct list_head leak_list;
#endif
162 163
};

164 165 166 167 168 169 170 171 172 173 174
/*
 * 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;
};

175 176 177 178 179 180 181 182 183 184 185
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;
}

186 187 188 189
struct extent_map_tree;

typedef struct extent_map *(get_extent_t)(struct inode *inode,
					  struct page *page,
190
					  size_t pg_offset,
191 192 193 194
					  u64 start, u64 len,
					  int create);

void extent_io_tree_init(struct extent_io_tree *tree,
195
			 struct address_space *mapping);
196
int try_release_extent_mapping(struct extent_map_tree *map,
197 198
			       struct extent_io_tree *tree, struct page *page,
			       gfp_t mask);
199
int try_release_extent_buffer(struct page *page);
200
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
201
		     struct extent_state **cached);
202 203 204 205 206 207

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

208
int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
209
int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
210
			  get_extent_t *get_extent, int mirror_num);
211 212 213 214 215
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,
216
		     u64 max_bytes, unsigned bits, int contig);
217

218
void free_extent_state(struct extent_state *state);
219
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
220
		   unsigned bits, int filled,
221
		   struct extent_state *cached_state);
222
int clear_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
223
		unsigned bits, struct extent_changeset *changeset);
224
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
225
		     unsigned bits, int wake, int delete,
226
		     struct extent_state **cached, gfp_t mask);
227

228 229 230 231 232 233 234 235 236 237 238 239 240 241
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,
242
		u64 end, unsigned bits)
243 244 245 246 247 248
{
	int wake = 0;

	if (bits & EXTENT_LOCKED)
		wake = 1;

249 250
	return clear_extent_bit(tree, start, end, bits, wake, 0, NULL,
			GFP_NOFS);
251 252
}

253
int set_record_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
254
			   unsigned bits, struct extent_changeset *changeset);
255
int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
256
		   unsigned bits, u64 *failed_start,
257
		   struct extent_state **cached_state, gfp_t mask);
258 259

static inline int set_extent_bits(struct extent_io_tree *tree, u64 start,
260
		u64 end, unsigned bits)
261
{
262
	return set_extent_bit(tree, start, end, bits, NULL, NULL, GFP_NOFS);
263 264
}

265 266 267 268 269 270
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);
}
271 272 273 274 275 276 277 278

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);
}

279
static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
280
		u64 end)
281 282 283
{
	return clear_extent_bit(tree, start, end,
				EXTENT_DIRTY | EXTENT_DELALLOC |
284
				EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS);
285 286
}

J
Josef Bacik 已提交
287
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
288
		       unsigned bits, unsigned clear_bits,
289
		       struct extent_state **cached_state);
290 291

static inline int set_extent_delalloc(struct extent_io_tree *tree, u64 start,
292
		u64 end, struct extent_state **cached_state)
293 294 295
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE,
296
			      NULL, cached_state, GFP_NOFS);
297 298 299
}

static inline int set_extent_defrag(struct extent_io_tree *tree, u64 start,
300
		u64 end, struct extent_state **cached_state)
301 302 303
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
304
			      NULL, cached_state, GFP_NOFS);
305 306 307
}

static inline int set_extent_new(struct extent_io_tree *tree, u64 start,
308
		u64 end)
309
{
310 311
	return set_extent_bit(tree, start, end, EXTENT_NEW, NULL, NULL,
			GFP_NOFS);
312 313 314 315 316 317 318 319 320
}

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);
}

321
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
322
			  u64 *start_ret, u64 *end_ret, unsigned bits,
323
			  struct extent_state **cached_state);
324 325 326 327 328
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);
329 330 331
int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
			      u64 start, u64 end, get_extent_t *get_extent,
			      int mode);
332 333 334 335
int extent_writepages(struct extent_io_tree *tree,
		      struct address_space *mapping,
		      get_extent_t *get_extent,
		      struct writeback_control *wbc);
336 337
int btree_write_cache_pages(struct address_space *mapping,
			    struct writeback_control *wbc);
338 339 340 341
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
342 343
int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
		__u64 start, __u64 len, get_extent_t *get_extent);
344 345
void set_page_extent_mapped(struct page *page);

346
struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
347
					  u64 start);
348 349
struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
						  u64 start, unsigned long len);
350
struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
351
						u64 start);
352
struct extent_buffer *btrfs_clone_extent_buffer(struct extent_buffer *src);
353
struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
354
					 u64 start);
355
void free_extent_buffer(struct extent_buffer *eb);
356
void free_extent_buffer_stale(struct extent_buffer *eb);
357 358 359
#define WAIT_NONE	0
#define WAIT_COMPLETE	1
#define WAIT_PAGE_LOCK	2
360
int read_extent_buffer_pages(struct extent_io_tree *tree,
361
			     struct extent_buffer *eb, u64 start, int wait,
362
			     get_extent_t *get_extent, int mirror_num);
363
void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
364 365 366

static inline unsigned long num_extent_pages(u64 start, u64 len)
{
367 368
	return ((start + len + PAGE_SIZE - 1) >> PAGE_SHIFT) -
		(start >> PAGE_SHIFT);
369 370
}

371 372 373 374 375 376 377 378 379 380 381
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);
382 383 384
int read_extent_buffer_to_user(struct extent_buffer *eb, void __user *dst,
			       unsigned long start,
			       unsigned long len);
385 386 387 388 389 390 391 392 393 394 395
void write_extent_buffer(struct extent_buffer *eb, const void *src,
			 unsigned long start, unsigned long len);
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);
void memset_extent_buffer(struct extent_buffer *eb, char c,
			  unsigned long start, unsigned long len);
396 397 398 399 400 401
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);
402
void clear_extent_buffer_dirty(struct extent_buffer *eb);
403
int set_extent_buffer_dirty(struct extent_buffer *eb);
404
void set_extent_buffer_uptodate(struct extent_buffer *eb);
405
void clear_extent_buffer_uptodate(struct extent_buffer *eb);
406
int extent_buffer_uptodate(struct extent_buffer *eb);
407
int extent_buffer_under_io(struct extent_buffer *eb);
408
int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset,
409
		      unsigned long min_len, char **map,
410
		      unsigned long *map_start,
411
		      unsigned long *map_len);
412
void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
413
void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
414
void extent_clear_unlock_delalloc(struct inode *inode, u64 start, u64 end,
415
				 struct page *locked_page,
416
				 unsigned bits_to_clear,
417
				 unsigned long page_ops);
418 419 420
struct bio *
btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs,
		gfp_t gfp_flags);
421 422
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);
423

424
struct btrfs_fs_info;
425

426 427 428
int repair_io_failure(struct inode *inode, u64 start, u64 length, u64 logical,
		      struct page *page, unsigned int pg_offset,
		      int mirror_num);
429 430
int clean_io_failure(struct inode *inode, u64 start, struct page *page,
		     unsigned int pg_offset);
431
void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
432 433
int repair_eb_io_failure(struct btrfs_root *root, struct extent_buffer *eb,
			 int mirror_num);
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453

/*
 * 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;
};

454
void btrfs_free_io_failure_record(struct inode *inode, u64 start, u64 end);
455 456 457 458 459 460 461
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,
462 463
				    bio_end_io_t *endio_func, void *data);
int free_io_failure(struct inode *inode, struct io_failure_record *rec);
464 465 466 467 468
#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);
469
#endif
470
struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
471
					       u64 start);
472
#endif