segment.h 25.8 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)					\
85
	((!is_valid_data_blkaddr(sbi, 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
	((MAIN_SEGS(sbi) + SIT_ENTRY_PER_BLOCK - 1) / 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
};

223 224 225 226 227 228 229
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 */
	char *sit_bitmap;		/* SIT bitmap pointer */
230 231 232
#ifdef CONFIG_F2FS_CHECK_FS
	char *sit_bitmap_mir;		/* SIT bitmap mirror */
#endif
233 234
	unsigned int bitmap_size;	/* SIT bitmap size */

J
Jaegeuk Kim 已提交
235
	unsigned long *tmp_map;			/* bitmap for temporal use */
236 237 238
	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 */
239
	struct rw_semaphore sentry_lock;	/* to protect SIT cache */
240 241 242 243 244 245 246 247
	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 */
248 249

	unsigned int last_victim[MAX_GC_POLICY]; /* last victim segment # */
250 251 252 253 254 255
};

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 */
256
	spinlock_t segmap_lock;		/* free segmap lock */
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278
	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 */
279
	unsigned long *victim_secmap;		/* background GC victims */
280 281 282 283 284 285 286 287 288 289 290 291
};

/* 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 */
292 293
	struct rw_semaphore journal_rwsem;	/* protect journal area */
	struct f2fs_journal *journal;		/* cached journal info */
294 295 296 297 298 299 300
	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 */
};

301 302 303 304 305 306
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 */
};

307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
/*
 * inline functions
 */
static inline struct curseg_info *CURSEG_I(struct f2fs_sb_info *sbi, int type)
{
	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);
326
	return &sit_i->sec_entries[GET_SEC_FROM_SEG(sbi, segno)];
327 328 329
}

static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi,
330
				unsigned int segno, bool use_section)
331 332 333 334 335
{
	/*
	 * In order to get # of valid blocks in a section instantly from many
	 * segments, f2fs manages two counting structures separately.
	 */
336
	if (use_section && __is_large_section(sbi))
337 338 339 340 341
		return get_sec_entry(sbi, segno)->valid_blocks;
	else
		return get_seg_entry(sbi, segno)->valid_blocks;
}

D
Daniel Rosenberg 已提交
342 343 344 345 346 347
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;
}

348 349 350 351 352 353 354
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
355 356 357
#ifdef CONFIG_F2FS_CHECK_FS
	memcpy(se->cur_valid_map_mir, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
#endif
358 359 360 361
	se->type = GET_SIT_TYPE(rs);
	se->mtime = le64_to_cpu(rs->mtime);
}

362
static inline void __seg_info_to_raw_sit(struct seg_entry *se,
363 364 365 366 367 368
					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);
369 370 371 372 373 374 375 376 377 378 379 380 381 382
	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);
383
	memset(raw_sit, 0, PAGE_SIZE);
384 385 386 387 388 389 390 391 392 393 394 395
	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);

396 397 398 399 400 401 402 403
	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;
404
	spin_lock(&free_i->segmap_lock);
405
	ret = find_next_bit(free_i->free_segmap, max, segno);
406
	spin_unlock(&free_i->segmap_lock);
407 408 409 410 411 412
	return ret;
}

static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
413 414
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
415 416
	unsigned int next;

417
	spin_lock(&free_i->segmap_lock);
418 419 420
	clear_bit(segno, free_i->free_segmap);
	free_i->free_segments++;

421 422
	next = find_next_bit(free_i->free_segmap,
			start_segno + sbi->segs_per_sec, start_segno);
423 424 425 426
	if (next >= start_segno + sbi->segs_per_sec) {
		clear_bit(secno, free_i->free_secmap);
		free_i->free_sections++;
	}
427
	spin_unlock(&free_i->segmap_lock);
428 429 430 431 432 433
}

static inline void __set_inuse(struct f2fs_sb_info *sbi,
		unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
434 435
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);

436 437 438 439 440 441 442 443 444 445
	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);
446 447
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
448 449
	unsigned int next;

450
	spin_lock(&free_i->segmap_lock);
451 452 453
	if (test_and_clear_bit(segno, free_i->free_segmap)) {
		free_i->free_segments++;

454 455
		if (IS_CURSEC(sbi, secno))
			goto skip_free;
456 457
		next = find_next_bit(free_i->free_segmap,
				start_segno + sbi->segs_per_sec, start_segno);
458 459 460 461 462
		if (next >= start_segno + sbi->segs_per_sec) {
			if (test_and_clear_bit(secno, free_i->free_secmap))
				free_i->free_sections++;
		}
	}
463
skip_free:
464
	spin_unlock(&free_i->segmap_lock);
465 466 467 468 469 470
}

static inline void __set_test_and_inuse(struct f2fs_sb_info *sbi,
		unsigned int segno)
{
	struct free_segmap_info *free_i = FREE_I(sbi);
471 472
	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);

473
	spin_lock(&free_i->segmap_lock);
474 475 476 477 478
	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--;
	}
479
	spin_unlock(&free_i->segmap_lock);
480 481 482 483 484 485
}

static inline void get_sit_bitmap(struct f2fs_sb_info *sbi,
		void *dst_addr)
{
	struct sit_info *sit_i = SIT_I(sbi);
486 487 488 489 490 491

#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
492 493 494 495 496
	memcpy(dst_addr, sit_i->sit_bitmap, sit_i->bitmap_size);
}

static inline block_t written_block_count(struct f2fs_sb_info *sbi)
{
497
	return SIT_I(sbi)->written_valid_blocks;
498 499 500 501
}

static inline unsigned int free_segments(struct f2fs_sb_info *sbi)
{
502
	return FREE_I(sbi)->free_segments;
503 504 505 506 507 508 509 510 511
}

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)
{
512
	return FREE_I(sbi)->free_sections;
513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
}

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)
{
537
	return GET_SEC_FROM_SEG(sbi, (unsigned int)reserved_segments(sbi));
538 539
}

540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
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;
}

567 568
static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
					int freed, int needed)
569
{
570 571
	int node_secs = get_blocktype_secs(sbi, F2FS_DIRTY_NODES);
	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
572
	int imeta_secs = get_blocktype_secs(sbi, F2FS_DIRTY_IMETA);
573

574
	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
575 576
		return false;

577 578 579
	if (free_sections(sbi) + freed == reserved_sections(sbi) + needed &&
			has_curseg_enough_space(sbi))
		return false;
580
	return (free_sections(sbi) + freed) <=
581 582
		(node_secs + 2 * dent_secs + imeta_secs +
		reserved_sections(sbi) + needed);
583 584
}

D
Daniel Rosenberg 已提交
585 586 587 588 589 590 591 592 593
static inline int f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
{
	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
		return 0;
	if (likely(!has_not_enough_free_secs(sbi, 0, 0)))
		return 0;
	return -ENOSPC;
}

594 595
static inline bool excess_prefree_segs(struct f2fs_sb_info *sbi)
{
C
Chris Fries 已提交
596
	return prefree_segments(sbi) > SM_I(sbi)->rec_prefree_segments;
597 598
}

599 600
static inline int utilization(struct f2fs_sb_info *sbi)
{
C
Chris Fries 已提交
601 602
	return div_u64((u64)valid_user_blocks(sbi) * 100,
					sbi->user_block_count);
603 604 605 606
}

/*
 * Sometimes f2fs may be better to drop out-of-place update policy.
607 608 609 610 611 612 613
 * 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,
614 615 616
 * 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.
617
 * F2FS_IPUT_DISABLE - disable IPU. (=default option)
618
 */
619
#define DEF_MIN_IPU_UTIL	70
620
#define DEF_MIN_FSYNC_BLOCKS	8
621
#define DEF_MIN_HOT_BLOCKS	16
622

623 624
#define SMALL_VOLUME_SEGMENTS	(16 * 512)	/* 16GB */

625 626 627 628 629
enum {
	F2FS_IPU_FORCE,
	F2FS_IPU_SSR,
	F2FS_IPU_UTIL,
	F2FS_IPU_SSR_UTIL,
630
	F2FS_IPU_FSYNC,
H
Hou Pengyang 已提交
631
	F2FS_IPU_ASYNC,
632 633
};

634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655
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)
{
656
	f2fs_bug_on(sbi, segno > TOTAL_SEGS(sbi) - 1);
657 658
}

659
static inline void verify_block_addr(struct f2fs_io_info *fio, block_t blk_addr)
660
{
661 662
	struct f2fs_sb_info *sbi = fio->sbi;

663
	if (__is_meta_io(fio))
664
		verify_blkaddr(sbi, blk_addr, META_GENERIC);
665
	else
666
		verify_blkaddr(sbi, blk_addr, DATA_GENERIC);
667 668 669
}

/*
A
arter97 已提交
670
 * Summary block is always treated as an invalid block
671
 */
672
static inline int check_block_count(struct f2fs_sb_info *sbi,
673 674
		int segno, struct f2fs_sit_entry *raw_sit)
{
J
Jaegeuk Kim 已提交
675
#ifdef CONFIG_F2FS_CHECK_FS
676
	bool is_valid  = test_bit_le(0, raw_sit->valid_map) ? true : false;
677
	int valid_blocks = 0;
678
	int cur_pos = 0, next_pos;
679 680

	/* check bitmap with valid block count */
681 682 683 684 685 686 687 688 689 690 691 692 693
	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);
694 695 696 697 698 699 700 701

	if (unlikely(GET_SIT_VBLOCKS(raw_sit) != valid_blocks)) {
		f2fs_msg(sbi->sb, KERN_ERR,
				"Mismatch valid blocks %d vs. %d",
					GET_SIT_VBLOCKS(raw_sit), valid_blocks);
		set_sbi_flag(sbi, SBI_NEED_FSCK);
		return -EINVAL;
	}
702
#endif
J
Jaegeuk Kim 已提交
703
	/* check segment usage, and check boundary of a given segment number */
704 705 706 707 708 709 710 711 712
	if (unlikely(GET_SIT_VBLOCKS(raw_sit) > sbi->blocks_per_seg
					|| segno > TOTAL_SEGS(sbi) - 1)) {
		f2fs_msg(sbi->sb, KERN_ERR,
				"Wrong valid blocks %d or segno %u",
					GET_SIT_VBLOCKS(raw_sit), segno);
		set_sbi_flag(sbi, SBI_NEED_FSCK);
		return -EINVAL;
	}
	return 0;
713
}
714 715 716 717 718

static inline pgoff_t current_sit_addr(struct f2fs_sb_info *sbi,
						unsigned int start)
{
	struct sit_info *sit_i = SIT_I(sbi);
719
	unsigned int offset = SIT_BLOCK_OFFSET(start);
720 721 722 723
	block_t blk_addr = sit_i->sit_base_addr + offset;

	check_seg_range(sbi, start);

724 725 726 727 728 729
#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

730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751
	/* 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)
{
752
	unsigned int block_off = SIT_BLOCK_OFFSET(start);
753

754
	f2fs_change_bit(block_off, sit_i->sit_bitmap);
755 756 757
#ifdef CONFIG_F2FS_CHECK_FS
	f2fs_change_bit(block_off, sit_i->sit_bitmap_mir);
#endif
758 759
}

C
Chao Yu 已提交
760 761
static inline unsigned long long get_mtime(struct f2fs_sb_info *sbi,
						bool base_time)
762 763
{
	struct sit_info *sit_i = SIT_I(sbi);
C
Chao Yu 已提交
764
	time64_t diff, now = ktime_get_real_seconds();
765

C
Chao Yu 已提交
766 767 768 769 770 771 772 773 774 775 776
	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;
777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
}

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;
}
799 800 801 802 803 804 805

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;
}
806

807 808 809 810
/*
 * 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,
811 812
 * 512 pages (2MB) * 8 for nodes, and
 * 256 pages * 8 for meta are set.
813 814 815
 */
static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type)
{
816
	if (sbi->sb->s_bdi->wb.dirty_exceeded)
817 818
		return 0;

819 820 821
	if (type == DATA)
		return sbi->blocks_per_seg;
	else if (type == NODE)
822
		return 8 * sbi->blocks_per_seg;
823
	else if (type == META)
824
		return 8 * BIO_MAX_PAGES;
825 826 827
	else
		return 0;
}
828 829 830 831 832 833 834 835 836 837 838 839 840

/*
 * 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;
841
	desired = BIO_MAX_PAGES;
842
	if (type == NODE)
843
		desired <<= 1;
844 845 846 847

	wbc->nr_to_write = desired;
	return desired - nr_to_write;
}
848 849 850 851 852 853 854 855 856 857 858

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
859 860 861
	for (i = MAX_PLIST_NUM - 1; i >= 0; i--) {
		if (i + 1 < dcc->discard_granularity)
			break;
862 863 864 865 866 867 868 869 870 871 872 873
		if (!list_empty(&dcc->pend_list[i])) {
			wakeup = true;
			break;
		}
	}
	mutex_unlock(&dcc->cmd_lock);
	if (!wakeup)
		return;
wake_up:
	dcc->discard_wake = 1;
	wake_up_interruptible_all(&dcc->discard_wait_queue);
}