segment.h 26.1 KB
Newer Older
C
Chao Yu 已提交
1
// SPDX-License-Identifier: GPL-2.0
J
Jaegeuk Kim 已提交
2
/*
3 4 5 6 7
 * fs/f2fs/segment.h
 *
 * Copyright (c) 2012 Samsung Electronics Co., Ltd.
 *             http://www.samsung.com/
 */
8
#include <linux/blkdev.h>
9
#include <linux/backing-dev.h>
10

11 12
/* constant macro */
#define NULL_SEGNO			((unsigned int)(~0))
13
#define NULL_SECNO			((unsigned int)(~0))
14

15
#define DEF_RECLAIM_PREFREE_SEGMENTS	5	/* 5% over total segments */
J
Jaegeuk Kim 已提交
16
#define DEF_MAX_RECLAIM_PREFREE_SEGMENTS	4096	/* 8GB in maximum */
17

J
Jaegeuk Kim 已提交
18 19
#define F2FS_MIN_SEGMENTS	9 /* SB + 2 (CP + SIT + NAT) + SSA + MAIN */

N
Namjae Jeon 已提交
20
/* L: Logical segment # in volume, R: Relative segment # in main area */
21 22
#define GET_L2R_SEGNO(free_i, segno)	((segno) - (free_i)->start_segno)
#define GET_R2L_SEGNO(free_i, segno)	((segno) + (free_i)->start_segno)
23

24 25
#define IS_DATASEG(t)	((t) <= CURSEG_COLD_DATA)
#define IS_NODESEG(t)	((t) >= CURSEG_HOT_NODE)
26

J
Jaegeuk Kim 已提交
27 28 29 30
#define IS_HOT(t)	((t) == CURSEG_HOT_NODE || (t) == CURSEG_HOT_DATA)
#define IS_WARM(t)	((t) == CURSEG_WARM_NODE || (t) == CURSEG_WARM_DATA)
#define IS_COLD(t)	((t) == CURSEG_COLD_NODE || (t) == CURSEG_COLD_DATA)

31
#define IS_CURSEG(sbi, seg)						\
32 33 34 35 36 37
	(((seg) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno) ||	\
	 ((seg) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno) ||	\
	 ((seg) == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno) ||	\
	 ((seg) == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno) ||	\
	 ((seg) == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno) ||	\
	 ((seg) == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno))
38 39

#define IS_CURSEC(sbi, secno)						\
40 41 42 43 44 45 46 47 48 49 50 51
	(((secno) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno /		\
	  (sbi)->segs_per_sec) ||	\
	 ((secno) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno /		\
	  (sbi)->segs_per_sec) ||	\
	 ((secno) == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno /		\
	  (sbi)->segs_per_sec) ||	\
	 ((secno) == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno /		\
	  (sbi)->segs_per_sec) ||	\
	 ((secno) == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno /		\
	  (sbi)->segs_per_sec) ||	\
	 ((secno) == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno /		\
	  (sbi)->segs_per_sec))	\
52

53 54 55 56 57 58
#define MAIN_BLKADDR(sbi)						\
	(SM_I(sbi) ? SM_I(sbi)->main_blkaddr : 				\
		le32_to_cpu(F2FS_RAW_SUPER(sbi)->main_blkaddr))
#define SEG0_BLKADDR(sbi)						\
	(SM_I(sbi) ? SM_I(sbi)->seg0_blkaddr : 				\
		le32_to_cpu(F2FS_RAW_SUPER(sbi)->segment0_blkaddr))
59 60

#define MAIN_SEGS(sbi)	(SM_I(sbi)->main_segments)
61
#define MAIN_SECS(sbi)	((sbi)->total_sections)
62

63 64 65
#define TOTAL_SEGS(sbi)							\
	(SM_I(sbi) ? SM_I(sbi)->segment_count : 				\
		le32_to_cpu(F2FS_RAW_SUPER(sbi)->segment_count))
66
#define TOTAL_BLKS(sbi)	(TOTAL_SEGS(sbi) << (sbi)->log_blocks_per_seg)
67 68

#define MAX_BLKADDR(sbi)	(SEG0_BLKADDR(sbi) + TOTAL_BLKS(sbi))
69 70
#define SEGMENT_SIZE(sbi)	(1ULL << ((sbi)->log_blocksize +	\
					(sbi)->log_blocks_per_seg))
71 72

#define START_BLOCK(sbi, segno)	(SEG0_BLKADDR(sbi) +			\
73
	 (GET_R2L_SEGNO(FREE_I(sbi), segno) << (sbi)->log_blocks_per_seg))
74

75
#define NEXT_FREE_BLKADDR(sbi, curseg)					\
76
	(START_BLOCK(sbi, (curseg)->segno) + (curseg)->next_blkoff)
77

78
#define GET_SEGOFF_FROM_SEG0(sbi, blk_addr)	((blk_addr) - SEG0_BLKADDR(sbi))
79
#define GET_SEGNO_FROM_SEG0(sbi, blk_addr)				\
80
	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) >> (sbi)->log_blocks_per_seg)
J
Jaegeuk Kim 已提交
81
#define GET_BLKOFF_FROM_SEG0(sbi, blk_addr)				\
82
	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) & ((sbi)->blocks_per_seg - 1))
J
Jaegeuk Kim 已提交
83

84
#define GET_SEGNO(sbi, blk_addr)					\
C
Chao Yu 已提交
85
	((!__is_valid_data_blkaddr(blk_addr)) ?			\
86 87
	NULL_SEGNO : GET_L2R_SEGNO(FREE_I(sbi),			\
		GET_SEGNO_FROM_SEG0(sbi, blk_addr)))
88 89 90
#define BLKS_PER_SEC(sbi)					\
	((sbi)->segs_per_sec * (sbi)->blocks_per_seg)
#define GET_SEC_FROM_SEG(sbi, segno)				\
91
	((segno) / (sbi)->segs_per_sec)
92
#define GET_SEG_FROM_SEC(sbi, secno)				\
93
	((secno) * (sbi)->segs_per_sec)
94 95 96 97
#define GET_ZONE_FROM_SEC(sbi, secno)				\
	((secno) / (sbi)->secs_per_zone)
#define GET_ZONE_FROM_SEG(sbi, segno)				\
	GET_ZONE_FROM_SEC(sbi, GET_SEC_FROM_SEG(sbi, segno))
98 99

#define GET_SUM_BLOCK(sbi, segno)				\
100
	((sbi)->sm_info->ssa_blkaddr + (segno))
101 102

#define GET_SUM_TYPE(footer) ((footer)->entry_type)
103
#define SET_SUM_TYPE(footer, type) ((footer)->entry_type = (type))
104 105

#define SIT_ENTRY_OFFSET(sit_i, segno)					\
106
	((segno) % (sit_i)->sents_per_block)
107
#define SIT_BLOCK_OFFSET(segno)					\
108
	((segno) / SIT_ENTRY_PER_BLOCK)
109 110
#define	START_SEGNO(segno)		\
	(SIT_BLOCK_OFFSET(segno) * SIT_ENTRY_PER_BLOCK)
111
#define SIT_BLK_CNT(sbi)			\
112
	DIV_ROUND_UP(MAIN_SEGS(sbi), SIT_ENTRY_PER_BLOCK)
113 114 115
#define f2fs_bitmap_size(nr)			\
	(BITS_TO_LONGS(nr) * sizeof(unsigned long))

C
Chao Yu 已提交
116 117 118
#define SECTOR_FROM_BLOCK(blk_addr)					\
	(((sector_t)blk_addr) << F2FS_LOG_SECTORS_PER_BLOCK)
#define SECTOR_TO_BLOCK(sectors)					\
119
	((sectors) >> F2FS_LOG_SECTORS_PER_BLOCK)
120

121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
/*
 * indicate a block allocation direction: RIGHT and LEFT.
 * RIGHT means allocating new sections towards the end of volume.
 * LEFT means the opposite direction.
 */
enum {
	ALLOC_RIGHT = 0,
	ALLOC_LEFT
};

/*
 * In the victim_sel_policy->alloc_mode, there are two block allocation modes.
 * LFS writes data sequentially with cleaning operations.
 * SSR (Slack Space Recycle) reuses obsolete space without cleaning operations.
 */
enum {
	LFS = 0,
	SSR
};

/*
 * In the victim_sel_policy->gc_mode, there are two gc, aka cleaning, modes.
 * GC_CB is based on cost-benefit algorithm.
 * GC_GREEDY is based on greedy algorithm.
 */
enum {
	GC_CB = 0,
148 149 150 151
	GC_GREEDY,
	ALLOC_NEXT,
	FLUSH_DEVICE,
	MAX_GC_POLICY,
152 153 154 155 156
};

/*
 * BG_GC means the background cleaning job.
 * FG_GC means the on-demand cleaning job.
157
 * FORCE_FG_GC means on-demand cleaning job in background.
158 159 160
 */
enum {
	BG_GC = 0,
161 162
	FG_GC,
	FORCE_FG_GC,
163 164 165 166 167 168 169
};

/* for a function parameter to select a victim segment */
struct victim_sel_policy {
	int alloc_mode;			/* LFS or SSR */
	int gc_mode;			/* GC_CB or GC_GREEDY */
	unsigned long *dirty_segmap;	/* dirty segment bitmap */
170
	unsigned int max_search;	/* maximum # of segments to search */
171 172 173 174 175 176 177
	unsigned int offset;		/* last scanned bitmap offset */
	unsigned int ofs_unit;		/* bitmap search unit */
	unsigned int min_cost;		/* minimum cost */
	unsigned int min_segno;		/* segment # having min. cost */
};

struct seg_entry {
C
Chao Yu 已提交
178 179 180 181
	unsigned int type:6;		/* segment type like CURSEG_XXX_TYPE */
	unsigned int valid_blocks:10;	/* # of valid blocks */
	unsigned int ckpt_valid_blocks:10;	/* # of valid blocks last cp */
	unsigned int padding:6;		/* padding */
182
	unsigned char *cur_valid_map;	/* validity bitmap of blocks */
C
Chao Yu 已提交
183 184 185
#ifdef CONFIG_F2FS_CHECK_FS
	unsigned char *cur_valid_map_mir;	/* mirror of current valid bitmap */
#endif
186 187 188 189
	/*
	 * # of valid blocks and the validity bitmap stored in the the last
	 * checkpoint pack. This information is used by the SSR mode.
	 */
C
Chao Yu 已提交
190
	unsigned char *ckpt_valid_map;	/* validity bitmap of blocks last cp */
191
	unsigned char *discard_map;
192 193 194 195 196 197 198 199 200 201 202
	unsigned long long mtime;	/* modification time of the segment */
};

struct sec_entry {
	unsigned int valid_blocks;	/* # of valid blocks in a section */
};

struct segment_allocation {
	void (*allocate_segment)(struct f2fs_sb_info *, int, bool);
};

C
Chao Yu 已提交
203 204 205 206
/*
 * this value is set in page as a private data which indicate that
 * the page is atomically written, and it is in inmem_pages list.
 */
207
#define ATOMIC_WRITTEN_PAGE		((unsigned long)-1)
208
#define DUMMY_WRITTEN_PAGE		((unsigned long)-2)
C
Chao Yu 已提交
209 210 211

#define IS_ATOMIC_WRITTEN_PAGE(page)			\
		(page_private(page) == (unsigned long)ATOMIC_WRITTEN_PAGE)
212 213
#define IS_DUMMY_WRITTEN_PAGE(page)			\
		(page_private(page) == (unsigned long)DUMMY_WRITTEN_PAGE)
C
Chao Yu 已提交
214

215
#define MAX_SKIP_GC_COUNT			16
216

J
Jaegeuk Kim 已提交
217 218 219
struct inmem_pages {
	struct list_head list;
	struct page *page;
220
	block_t old_addr;		/* for revoking when fail to commit */
J
Jaegeuk Kim 已提交
221 222
};

223 224 225 226 227 228
struct sit_info {
	const struct segment_allocation *s_ops;

	block_t sit_base_addr;		/* start block address of SIT area */
	block_t sit_blocks;		/* # of blocks used by SIT area */
	block_t written_valid_blocks;	/* # of valid blocks in main area */
229
	char *bitmap;			/* all bitmaps pointer */
230
	char *sit_bitmap;		/* SIT bitmap pointer */
231 232
#ifdef CONFIG_F2FS_CHECK_FS
	char *sit_bitmap_mir;		/* SIT bitmap mirror */
233 234 235

	/* bitmap of segments to be ignored by GC in case of errors */
	unsigned long *invalid_segmap;
236
#endif
237 238
	unsigned int bitmap_size;	/* SIT bitmap size */

J
Jaegeuk Kim 已提交
239
	unsigned long *tmp_map;			/* bitmap for temporal use */
240 241 242
	unsigned long *dirty_sentries_bitmap;	/* bitmap for dirty sentries */
	unsigned int dirty_sentries;		/* # of dirty sentries */
	unsigned int sents_per_block;		/* # of SIT entries per block */
243
	struct rw_semaphore sentry_lock;	/* to protect SIT cache */
244 245 246 247 248 249 250 251
	struct seg_entry *sentries;		/* SIT segment-level cache */
	struct sec_entry *sec_entries;		/* SIT section-level cache */

	/* for cost-benefit algorithm in cleaning procedure */
	unsigned long long elapsed_time;	/* elapsed time after mount */
	unsigned long long mounted_time;	/* mount time */
	unsigned long long min_mtime;		/* min. modification time */
	unsigned long long max_mtime;		/* max. modification time */
252 253

	unsigned int last_victim[MAX_GC_POLICY]; /* last victim segment # */
254 255 256 257 258 259
};

struct free_segmap_info {
	unsigned int start_segno;	/* start segment number logically */
	unsigned int free_segments;	/* # of free segments */
	unsigned int free_sections;	/* # of free sections */
260
	spinlock_t segmap_lock;		/* free segmap lock */
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
	unsigned long *free_segmap;	/* free segment bitmap */
	unsigned long *free_secmap;	/* free section bitmap */
};

/* Notice: The order of dirty type is same with CURSEG_XXX in f2fs.h */
enum dirty_type {
	DIRTY_HOT_DATA,		/* dirty segments assigned as hot data logs */
	DIRTY_WARM_DATA,	/* dirty segments assigned as warm data logs */
	DIRTY_COLD_DATA,	/* dirty segments assigned as cold data logs */
	DIRTY_HOT_NODE,		/* dirty segments assigned as hot node logs */
	DIRTY_WARM_NODE,	/* dirty segments assigned as warm node logs */
	DIRTY_COLD_NODE,	/* dirty segments assigned as cold node logs */
	DIRTY,			/* to count # of dirty segments */
	PRE,			/* to count # of entirely obsolete segments */
	NR_DIRTY_TYPE
};

struct dirty_seglist_info {
	const struct victim_selection *v_ops;	/* victim selction operation */
	unsigned long *dirty_segmap[NR_DIRTY_TYPE];
	struct mutex seglist_lock;		/* lock for segment bitmaps */
	int nr_dirty[NR_DIRTY_TYPE];		/* # of dirty segments */
283
	unsigned long *victim_secmap;		/* background GC victims */
284 285 286 287 288 289 290 291 292 293 294 295
};

/* victim selection function for cleaning and SSR */
struct victim_selection {
	int (*get_victim)(struct f2fs_sb_info *, unsigned int *,
							int, int, char);
};

/* for active log information */
struct curseg_info {
	struct mutex curseg_mutex;		/* lock for consistency */
	struct f2fs_summary_block *sum_blk;	/* cached summary block */
296 297
	struct rw_semaphore journal_rwsem;	/* protect journal area */
	struct f2fs_journal *journal;		/* cached journal info */
298 299 300 301 302 303 304
	unsigned char alloc_type;		/* current allocation type */
	unsigned int segno;			/* current segment number */
	unsigned short next_blkoff;		/* next block offset to write */
	unsigned int zone;			/* current zone number */
	unsigned int next_segno;		/* preallocated segment */
};

305 306 307 308 309 310
struct sit_entry_set {
	struct list_head set_list;	/* link with all sit sets */
	unsigned int start_segno;	/* start segno of sits in set */
	unsigned int entry_cnt;		/* the # of sit entries in set */
};

311 312 313 314 315
/*
 * inline functions
 */
static inline struct curseg_info *CURSEG_I(struct f2fs_sb_info *sbi, int type)
{
J
Jaegeuk Kim 已提交
316 317
	if (type == CURSEG_COLD_DATA_PINNED)
		type = CURSEG_COLD_DATA;
318 319 320 321 322 323 324 325 326 327 328 329 330 331
	return (struct curseg_info *)(SM_I(sbi)->curseg_array + type);
}

static inline struct seg_entry *get_seg_entry(struct f2fs_sb_info *sbi,
						unsigned int segno)
{
	struct sit_info *sit_i = SIT_I(sbi);
	return &sit_i->sentries[segno];
}

static inline struct sec_entry *get_sec_entry(struct f2fs_sb_info *sbi,
						unsigned int segno)
{
	struct sit_info *sit_i = SIT_I(sbi);
332
	return &sit_i->sec_entries[GET_SEC_FROM_SEG(sbi, segno)];
333 334 335
}

static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi,
336
				unsigned int segno, bool use_section)
337 338 339 340 341
{
	/*
	 * In order to get # of valid blocks in a section instantly from many
	 * segments, f2fs manages two counting structures separately.
	 */
342
	if (use_section && __is_large_section(sbi))
343 344 345 346 347
		return get_sec_entry(sbi, segno)->valid_blocks;
	else
		return get_seg_entry(sbi, segno)->valid_blocks;
}

D
Daniel Rosenberg 已提交
348 349 350 351 352 353
static inline unsigned int get_ckpt_valid_blocks(struct f2fs_sb_info *sbi,
				unsigned int segno)
{
	return get_seg_entry(sbi, segno)->ckpt_valid_blocks;
}

354 355 356 357 358 359 360
static inline void seg_info_from_raw_sit(struct seg_entry *se,
					struct f2fs_sit_entry *rs)
{
	se->valid_blocks = GET_SIT_VBLOCKS(rs);
	se->ckpt_valid_blocks = GET_SIT_VBLOCKS(rs);
	memcpy(se->cur_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
	memcpy(se->ckpt_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
C
Chao Yu 已提交
361 362 363
#ifdef CONFIG_F2FS_CHECK_FS
	memcpy(se->cur_valid_map_mir, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
#endif
364 365 366 367
	se->type = GET_SIT_TYPE(rs);
	se->mtime = le64_to_cpu(rs->mtime);
}

368
static inline void __seg_info_to_raw_sit(struct seg_entry *se,
369 370 371 372 373 374
					struct f2fs_sit_entry *rs)
{
	unsigned short raw_vblocks = (se->type << SIT_VBLOCKS_SHIFT) |
					se->valid_blocks;
	rs->vblocks = cpu_to_le16(raw_vblocks);
	memcpy(rs->valid_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
375 376 377 378 379 380 381 382 383 384 385 386 387 388
	rs->mtime = cpu_to_le64(se->mtime);
}

static inline void seg_info_to_sit_page(struct f2fs_sb_info *sbi,
				struct page *page, unsigned int start)
{
	struct f2fs_sit_block *raw_sit;
	struct seg_entry *se;
	struct f2fs_sit_entry *rs;
	unsigned int end = min(start + SIT_ENTRY_PER_BLOCK,
					(unsigned long)MAIN_SEGS(sbi));
	int i;

	raw_sit = (struct f2fs_sit_block *)page_address(page);
389
	memset(raw_sit, 0, PAGE_SIZE);
390 391 392 393 394 395 396 397 398 399 400 401
	for (i = 0; i < end - start; i++) {
		rs = &raw_sit->entries[i];
		se = get_seg_entry(sbi, start + i);
		__seg_info_to_raw_sit(se, rs);
	}
}

static inline void seg_info_to_raw_sit(struct seg_entry *se,
					struct f2fs_sit_entry *rs)
{
	__seg_info_to_raw_sit(se, rs);

402 403 404 405 406 407 408 409
	memcpy(se->ckpt_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
	se->ckpt_valid_blocks = se->valid_blocks;
}

static inline unsigned int find_next_inuse(struct free_segmap_info *free_i,
		unsigned int max, unsigned int segno)
{
	unsigned int ret;
410
	spin_lock(&free_i->segmap_lock);
411
	ret = find_next_bit(free_i->free_segmap, max, segno);
412
	spin_unlock(&free_i->segmap_lock);
413 414 415 416 417 418
	return ret;
}

static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
419 420
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
421 422
	unsigned int next;

423
	spin_lock(&free_i->segmap_lock);
424 425 426
	clear_bit(segno, free_i->free_segmap);
	free_i->free_segments++;

427 428
	next = find_next_bit(free_i->free_segmap,
			start_segno + sbi->segs_per_sec, start_segno);
429 430 431 432
	if (next >= start_segno + sbi->segs_per_sec) {
		clear_bit(secno, free_i->free_secmap);
		free_i->free_sections++;
	}
433
	spin_unlock(&free_i->segmap_lock);
434 435 436 437 438 439
}

static inline void __set_inuse(struct f2fs_sb_info *sbi,
		unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
440 441
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);

442 443 444 445 446 447 448 449 450 451
	set_bit(segno, free_i->free_segmap);
	free_i->free_segments--;
	if (!test_and_set_bit(secno, free_i->free_secmap))
		free_i->free_sections--;
}

static inline void __set_test_and_free(struct f2fs_sb_info *sbi,
		unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
452 453
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
454 455
	unsigned int next;

456
	spin_lock(&free_i->segmap_lock);
457 458 459
	if (test_and_clear_bit(segno, free_i->free_segmap)) {
		free_i->free_segments++;

460 461
		if (IS_CURSEC(sbi, secno))
			goto skip_free;
462 463
		next = find_next_bit(free_i->free_segmap,
				start_segno + sbi->segs_per_sec, start_segno);
464 465 466 467 468
		if (next >= start_segno + sbi->segs_per_sec) {
			if (test_and_clear_bit(secno, free_i->free_secmap))
				free_i->free_sections++;
		}
	}
469
skip_free:
470
	spin_unlock(&free_i->segmap_lock);
471 472 473 474 475 476
}

static inline void __set_test_and_inuse(struct f2fs_sb_info *sbi,
		unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
477 478
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);

479
	spin_lock(&free_i->segmap_lock);
480 481 482 483 484
	if (!test_and_set_bit(segno, free_i->free_segmap)) {
		free_i->free_segments--;
		if (!test_and_set_bit(secno, free_i->free_secmap))
			free_i->free_sections--;
	}
485
	spin_unlock(&free_i->segmap_lock);
486 487 488 489 490 491
}

static inline void get_sit_bitmap(struct f2fs_sb_info *sbi,
		void *dst_addr)
{
	struct sit_info *sit_i = SIT_I(sbi);
492 493 494 495 496 497

#ifdef CONFIG_F2FS_CHECK_FS
	if (memcmp(sit_i->sit_bitmap, sit_i->sit_bitmap_mir,
						sit_i->bitmap_size))
		f2fs_bug_on(sbi, 1);
#endif
498 499 500 501 502
	memcpy(dst_addr, sit_i->sit_bitmap, sit_i->bitmap_size);
}

static inline block_t written_block_count(struct f2fs_sb_info *sbi)
{
503
	return SIT_I(sbi)->written_valid_blocks;
504 505 506 507
}

static inline unsigned int free_segments(struct f2fs_sb_info *sbi)
{
508
	return FREE_I(sbi)->free_segments;
509 510 511 512 513 514 515 516 517
}

static inline int reserved_segments(struct f2fs_sb_info *sbi)
{
	return SM_I(sbi)->reserved_segments;
}

static inline unsigned int free_sections(struct f2fs_sb_info *sbi)
{
518
	return FREE_I(sbi)->free_sections;
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542
}

static inline unsigned int prefree_segments(struct f2fs_sb_info *sbi)
{
	return DIRTY_I(sbi)->nr_dirty[PRE];
}

static inline unsigned int dirty_segments(struct f2fs_sb_info *sbi)
{
	return DIRTY_I(sbi)->nr_dirty[DIRTY_HOT_DATA] +
		DIRTY_I(sbi)->nr_dirty[DIRTY_WARM_DATA] +
		DIRTY_I(sbi)->nr_dirty[DIRTY_COLD_DATA] +
		DIRTY_I(sbi)->nr_dirty[DIRTY_HOT_NODE] +
		DIRTY_I(sbi)->nr_dirty[DIRTY_WARM_NODE] +
		DIRTY_I(sbi)->nr_dirty[DIRTY_COLD_NODE];
}

static inline int overprovision_segments(struct f2fs_sb_info *sbi)
{
	return SM_I(sbi)->ovp_segments;
}

static inline int reserved_sections(struct f2fs_sb_info *sbi)
{
543
	return GET_SEC_FROM_SEG(sbi, (unsigned int)reserved_segments(sbi));
544 545
}

546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
static inline bool has_curseg_enough_space(struct f2fs_sb_info *sbi)
{
	unsigned int node_blocks = get_pages(sbi, F2FS_DIRTY_NODES) +
					get_pages(sbi, F2FS_DIRTY_DENTS);
	unsigned int dent_blocks = get_pages(sbi, F2FS_DIRTY_DENTS);
	unsigned int segno, left_blocks;
	int i;

	/* check current node segment */
	for (i = CURSEG_HOT_NODE; i <= CURSEG_COLD_NODE; i++) {
		segno = CURSEG_I(sbi, i)->segno;
		left_blocks = sbi->blocks_per_seg -
			get_seg_entry(sbi, segno)->ckpt_valid_blocks;

		if (node_blocks > left_blocks)
			return false;
	}

	/* check current data segment */
	segno = CURSEG_I(sbi, CURSEG_HOT_DATA)->segno;
	left_blocks = sbi->blocks_per_seg -
			get_seg_entry(sbi, segno)->ckpt_valid_blocks;
	if (dent_blocks > left_blocks)
		return false;
	return true;
}

573 574
static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
					int freed, int needed)
575
{
576 577
	int node_secs = get_blocktype_secs(sbi, F2FS_DIRTY_NODES);
	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
578
	int imeta_secs = get_blocktype_secs(sbi, F2FS_DIRTY_IMETA);
579

580
	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
581 582
		return false;

583 584 585
	if (free_sections(sbi) + freed == reserved_sections(sbi) + needed &&
			has_curseg_enough_space(sbi))
		return false;
586
	return (free_sections(sbi) + freed) <=
587 588
		(node_secs + 2 * dent_secs + imeta_secs +
		reserved_sections(sbi) + needed);
589 590
}

591
static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
D
Daniel Rosenberg 已提交
592 593
{
	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
594
		return true;
D
Daniel Rosenberg 已提交
595
	if (likely(!has_not_enough_free_secs(sbi, 0, 0)))
596 597
		return true;
	return false;
D
Daniel Rosenberg 已提交
598 599
}

600 601
static inline bool excess_prefree_segs(struct f2fs_sb_info *sbi)
{
C
Chris Fries 已提交
602
	return prefree_segments(sbi) > SM_I(sbi)->rec_prefree_segments;
603 604
}

605 606
static inline int utilization(struct f2fs_sb_info *sbi)
{
C
Chris Fries 已提交
607 608
	return div_u64((u64)valid_user_blocks(sbi) * 100,
					sbi->user_block_count);
609 610 611 612
}

/*
 * Sometimes f2fs may be better to drop out-of-place update policy.
613 614 615 616 617 618 619
 * And, users can control the policy through sysfs entries.
 * There are five policies with triggering conditions as follows.
 * F2FS_IPU_FORCE - all the time,
 * F2FS_IPU_SSR - if SSR mode is activated,
 * F2FS_IPU_UTIL - if FS utilization is over threashold,
 * F2FS_IPU_SSR_UTIL - if SSR mode is activated and FS utilization is over
 *                     threashold,
620 621 622
 * F2FS_IPU_FSYNC - activated in fsync path only for high performance flash
 *                     storages. IPU will be triggered only if the # of dirty
 *                     pages over min_fsync_blocks.
623
 * F2FS_IPUT_DISABLE - disable IPU. (=default option)
624
 */
625
#define DEF_MIN_IPU_UTIL	70
626
#define DEF_MIN_FSYNC_BLOCKS	8
627
#define DEF_MIN_HOT_BLOCKS	16
628

629 630
#define SMALL_VOLUME_SEGMENTS	(16 * 512)	/* 16GB */

631 632 633 634 635
enum {
	F2FS_IPU_FORCE,
	F2FS_IPU_SSR,
	F2FS_IPU_UTIL,
	F2FS_IPU_SSR_UTIL,
636
	F2FS_IPU_FSYNC,
H
Hou Pengyang 已提交
637
	F2FS_IPU_ASYNC,
638 639
};

640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
static inline unsigned int curseg_segno(struct f2fs_sb_info *sbi,
		int type)
{
	struct curseg_info *curseg = CURSEG_I(sbi, type);
	return curseg->segno;
}

static inline unsigned char curseg_alloc_type(struct f2fs_sb_info *sbi,
		int type)
{
	struct curseg_info *curseg = CURSEG_I(sbi, type);
	return curseg->alloc_type;
}

static inline unsigned short curseg_blkoff(struct f2fs_sb_info *sbi, int type)
{
	struct curseg_info *curseg = CURSEG_I(sbi, type);
	return curseg->next_blkoff;
}

static inline void check_seg_range(struct f2fs_sb_info *sbi, unsigned int segno)
{
662
	f2fs_bug_on(sbi, segno > TOTAL_SEGS(sbi) - 1);
663 664
}

C
Chao Yu 已提交
665
static inline void verify_fio_blkaddr(struct f2fs_io_info *fio)
666
{
667 668
	struct f2fs_sb_info *sbi = fio->sbi;

C
Chao Yu 已提交
669 670 671 672 673
	if (__is_valid_data_blkaddr(fio->old_blkaddr))
		verify_blkaddr(sbi, fio->old_blkaddr, __is_meta_io(fio) ?
					META_GENERIC : DATA_GENERIC);
	verify_blkaddr(sbi, fio->new_blkaddr, __is_meta_io(fio) ?
					META_GENERIC : DATA_GENERIC_ENHANCE);
674 675 676
}

/*
A
arter97 已提交
677
 * Summary block is always treated as an invalid block
678
 */
679
static inline int check_block_count(struct f2fs_sb_info *sbi,
680 681
		int segno, struct f2fs_sit_entry *raw_sit)
{
682
	bool is_valid  = test_bit_le(0, raw_sit->valid_map) ? true : false;
683
	int valid_blocks = 0;
684
	int cur_pos = 0, next_pos;
685 686

	/* check bitmap with valid block count */
687 688 689 690 691 692 693 694 695 696 697 698 699
	do {
		if (is_valid) {
			next_pos = find_next_zero_bit_le(&raw_sit->valid_map,
					sbi->blocks_per_seg,
					cur_pos);
			valid_blocks += next_pos - cur_pos;
		} else
			next_pos = find_next_bit_le(&raw_sit->valid_map,
					sbi->blocks_per_seg,
					cur_pos);
		cur_pos = next_pos;
		is_valid = !is_valid;
	} while (cur_pos < sbi->blocks_per_seg);
700 701

	if (unlikely(GET_SIT_VBLOCKS(raw_sit) != valid_blocks)) {
702 703
		f2fs_err(sbi, "Mismatch valid blocks %d vs. %d",
			 GET_SIT_VBLOCKS(raw_sit), valid_blocks);
704
		set_sbi_flag(sbi, SBI_NEED_FSCK);
705
		return -EFSCORRUPTED;
706
	}
707

J
Jaegeuk Kim 已提交
708
	/* check segment usage, and check boundary of a given segment number */
709 710
	if (unlikely(GET_SIT_VBLOCKS(raw_sit) > sbi->blocks_per_seg
					|| segno > TOTAL_SEGS(sbi) - 1)) {
711 712
		f2fs_err(sbi, "Wrong valid blocks %d or segno %u",
			 GET_SIT_VBLOCKS(raw_sit), segno);
713
		set_sbi_flag(sbi, SBI_NEED_FSCK);
714
		return -EFSCORRUPTED;
715 716
	}
	return 0;
717
}
718 719 720 721 722

static inline pgoff_t current_sit_addr(struct f2fs_sb_info *sbi,
						unsigned int start)
{
	struct sit_info *sit_i = SIT_I(sbi);
723
	unsigned int offset = SIT_BLOCK_OFFSET(start);
724 725 726 727
	block_t blk_addr = sit_i->sit_base_addr + offset;

	check_seg_range(sbi, start);

728 729 730 731 732 733
#ifdef CONFIG_F2FS_CHECK_FS
	if (f2fs_test_bit(offset, sit_i->sit_bitmap) !=
			f2fs_test_bit(offset, sit_i->sit_bitmap_mir))
		f2fs_bug_on(sbi, 1);
#endif

734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
	/* calculate sit block address */
	if (f2fs_test_bit(offset, sit_i->sit_bitmap))
		blk_addr += sit_i->sit_blocks;

	return blk_addr;
}

static inline pgoff_t next_sit_addr(struct f2fs_sb_info *sbi,
						pgoff_t block_addr)
{
	struct sit_info *sit_i = SIT_I(sbi);
	block_addr -= sit_i->sit_base_addr;
	if (block_addr < sit_i->sit_blocks)
		block_addr += sit_i->sit_blocks;
	else
		block_addr -= sit_i->sit_blocks;

	return block_addr + sit_i->sit_base_addr;
}

static inline void set_to_next_sit(struct sit_info *sit_i, unsigned int start)
{
756
	unsigned int block_off = SIT_BLOCK_OFFSET(start);
757

758
	f2fs_change_bit(block_off, sit_i->sit_bitmap);
759 760 761
#ifdef CONFIG_F2FS_CHECK_FS
	f2fs_change_bit(block_off, sit_i->sit_bitmap_mir);
#endif
762 763
}

C
Chao Yu 已提交
764 765
static inline unsigned long long get_mtime(struct f2fs_sb_info *sbi,
						bool base_time)
766 767
{
	struct sit_info *sit_i = SIT_I(sbi);
C
Chao Yu 已提交
768
	time64_t diff, now = ktime_get_real_seconds();
769

C
Chao Yu 已提交
770 771 772 773 774 775 776 777 778 779 780
	if (now >= sit_i->mounted_time)
		return sit_i->elapsed_time + now - sit_i->mounted_time;

	/* system time is set to the past */
	if (!base_time) {
		diff = sit_i->mounted_time - now;
		if (sit_i->elapsed_time >= diff)
			return sit_i->elapsed_time - diff;
		return 0;
	}
	return sit_i->elapsed_time;
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802
}

static inline void set_summary(struct f2fs_summary *sum, nid_t nid,
			unsigned int ofs_in_node, unsigned char version)
{
	sum->nid = cpu_to_le32(nid);
	sum->ofs_in_node = cpu_to_le16(ofs_in_node);
	sum->version = version;
}

static inline block_t start_sum_block(struct f2fs_sb_info *sbi)
{
	return __start_cp_addr(sbi) +
		le32_to_cpu(F2FS_CKPT(sbi)->cp_pack_start_sum);
}

static inline block_t sum_blk_addr(struct f2fs_sb_info *sbi, int base, int type)
{
	return __start_cp_addr(sbi) +
		le32_to_cpu(F2FS_CKPT(sbi)->cp_pack_total_block_count)
				- (base + 1) + type;
}
803 804 805 806 807 808 809

static inline bool sec_usage_check(struct f2fs_sb_info *sbi, unsigned int secno)
{
	if (IS_CURSEC(sbi, secno) || (sbi->cur_victim_sec == secno))
		return true;
	return false;
}
810

811 812 813 814
/*
 * It is very important to gather dirty pages and write at once, so that we can
 * submit a big bio without interfering other data writes.
 * By default, 512 pages for directory data,
815 816
 * 512 pages (2MB) * 8 for nodes, and
 * 256 pages * 8 for meta are set.
817 818 819
 */
static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type)
{
820
	if (sbi->sb->s_bdi->wb.dirty_exceeded)
821 822
		return 0;

823 824 825
	if (type == DATA)
		return sbi->blocks_per_seg;
	else if (type == NODE)
826
		return 8 * sbi->blocks_per_seg;
827
	else if (type == META)
828
		return 8 * BIO_MAX_PAGES;
829 830 831
	else
		return 0;
}
832 833 834 835 836 837 838 839 840 841 842 843 844

/*
 * When writing pages, it'd better align nr_to_write for segment size.
 */
static inline long nr_pages_to_write(struct f2fs_sb_info *sbi, int type,
					struct writeback_control *wbc)
{
	long nr_to_write, desired;

	if (wbc->sync_mode != WB_SYNC_NONE)
		return 0;

	nr_to_write = wbc->nr_to_write;
845
	desired = BIO_MAX_PAGES;
846
	if (type == NODE)
847
		desired <<= 1;
848 849 850 851

	wbc->nr_to_write = desired;
	return desired - nr_to_write;
}
852 853 854 855 856 857 858 859 860 861 862

static inline void wake_up_discard_thread(struct f2fs_sb_info *sbi, bool force)
{
	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
	bool wakeup = false;
	int i;

	if (force)
		goto wake_up;

	mutex_lock(&dcc->cmd_lock);
C
Chao Yu 已提交
863 864 865
	for (i = MAX_PLIST_NUM - 1; i >= 0; i--) {
		if (i + 1 < dcc->discard_granularity)
			break;
866 867 868 869 870 871
		if (!list_empty(&dcc->pend_list[i])) {
			wakeup = true;
			break;
		}
	}
	mutex_unlock(&dcc->cmd_lock);
872
	if (!wakeup || !is_idle(sbi, DISCARD_TIME))
873 874 875 876 877
		return;
wake_up:
	dcc->discard_wake = 1;
	wake_up_interruptible_all(&dcc->discard_wait_queue);
}