diff options
Diffstat (limited to 'fs/f2fs/segment.h')
-rw-r--r-- | fs/f2fs/segment.h | 265 |
1 files changed, 177 insertions, 88 deletions
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 8129be788bd5..db619fd2f51a 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -18,6 +18,8 @@ #define F2FS_MIN_SEGMENTS 9 /* SB + 2 (CP + SIT + NAT) + SSA + MAIN */ #define F2FS_MIN_META_SEGMENTS 8 /* SB + 2 (CP + SIT + NAT) + SSA */ +#define INVALID_MTIME ULLONG_MAX /* no valid blocks in a segment/section */ + /* L: Logical segment # in volume, R: Relative segment # in main area */ #define GET_L2R_SEGNO(free_i, segno) ((segno) - (free_i)->start_segno) #define GET_R2L_SEGNO(free_i, segno) ((segno) + (free_i)->start_segno) @@ -32,10 +34,6 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi, f2fs_bug_on(sbi, seg_type >= NR_PERSISTENT_LOG); } -#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) - #define IS_CURSEG(sbi, seg) \ (((seg) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno) || \ ((seg) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno) || \ @@ -48,21 +46,21 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi, #define IS_CURSEC(sbi, secno) \ (((secno) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_COLD_DATA_PINNED)->segno / \ - (sbi)->segs_per_sec) || \ + SEGS_PER_SEC(sbi)) || \ ((secno) == CURSEG_I(sbi, CURSEG_ALL_DATA_ATGC)->segno / \ - (sbi)->segs_per_sec)) + SEGS_PER_SEC(sbi))) #define MAIN_BLKADDR(sbi) \ (SM_I(sbi) ? SM_I(sbi)->main_blkaddr : \ @@ -77,40 +75,39 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi, #define TOTAL_SEGS(sbi) \ (SM_I(sbi) ? SM_I(sbi)->segment_count : \ le32_to_cpu(F2FS_RAW_SUPER(sbi)->segment_count)) -#define TOTAL_BLKS(sbi) (TOTAL_SEGS(sbi) << (sbi)->log_blocks_per_seg) +#define TOTAL_BLKS(sbi) (SEGS_TO_BLKS(sbi, TOTAL_SEGS(sbi))) #define MAX_BLKADDR(sbi) (SEG0_BLKADDR(sbi) + TOTAL_BLKS(sbi)) #define SEGMENT_SIZE(sbi) (1ULL << ((sbi)->log_blocksize + \ (sbi)->log_blocks_per_seg)) #define START_BLOCK(sbi, segno) (SEG0_BLKADDR(sbi) + \ - (GET_R2L_SEGNO(FREE_I(sbi), segno) << (sbi)->log_blocks_per_seg)) + (SEGS_TO_BLKS(sbi, GET_R2L_SEGNO(FREE_I(sbi), segno)))) #define NEXT_FREE_BLKADDR(sbi, curseg) \ (START_BLOCK(sbi, (curseg)->segno) + (curseg)->next_blkoff) #define GET_SEGOFF_FROM_SEG0(sbi, blk_addr) ((blk_addr) - SEG0_BLKADDR(sbi)) #define GET_SEGNO_FROM_SEG0(sbi, blk_addr) \ - (GET_SEGOFF_FROM_SEG0(sbi, blk_addr) >> (sbi)->log_blocks_per_seg) + (BLKS_TO_SEGS(sbi, GET_SEGOFF_FROM_SEG0(sbi, blk_addr))) #define GET_BLKOFF_FROM_SEG0(sbi, blk_addr) \ - (GET_SEGOFF_FROM_SEG0(sbi, blk_addr) & ((sbi)->blocks_per_seg - 1)) + (GET_SEGOFF_FROM_SEG0(sbi, blk_addr) & (BLKS_PER_SEG(sbi) - 1)) #define GET_SEGNO(sbi, blk_addr) \ ((!__is_valid_data_blkaddr(blk_addr)) ? \ NULL_SEGNO : GET_L2R_SEGNO(FREE_I(sbi), \ GET_SEGNO_FROM_SEG0(sbi, blk_addr))) -#define BLKS_PER_SEC(sbi) \ - ((sbi)->segs_per_sec * (sbi)->blocks_per_seg) #define CAP_BLKS_PER_SEC(sbi) \ - ((sbi)->segs_per_sec * (sbi)->blocks_per_seg - \ - (sbi)->unusable_blocks_per_sec) + (BLKS_PER_SEC(sbi) - (sbi)->unusable_blocks_per_sec) #define CAP_SEGS_PER_SEC(sbi) \ - ((sbi)->segs_per_sec - ((sbi)->unusable_blocks_per_sec >>\ - (sbi)->log_blocks_per_seg)) + (SEGS_PER_SEC(sbi) - \ + BLKS_TO_SEGS(sbi, (sbi)->unusable_blocks_per_sec)) +#define GET_START_SEG_FROM_SEC(sbi, segno) \ + (rounddown(segno, SEGS_PER_SEC(sbi))) #define GET_SEC_FROM_SEG(sbi, segno) \ - (((segno) == -1) ? -1 : (segno) / (sbi)->segs_per_sec) + (((segno) == -1) ? -1 : (segno) / SEGS_PER_SEC(sbi)) #define GET_SEG_FROM_SEC(sbi, secno) \ - ((secno) * (sbi)->segs_per_sec) + ((secno) * SEGS_PER_SEC(sbi)) #define GET_ZONE_FROM_SEC(sbi, secno) \ (((secno) == -1) ? -1 : (secno) / (sbi)->secs_per_zone) #define GET_ZONE_FROM_SEG(sbi, segno) \ @@ -139,16 +136,6 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi, ((sectors) >> F2FS_LOG_SECTORS_PER_BLOCK) /* - * 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 three block allocation modes. * LFS writes data sequentially with cleaning operations. * SSR (Slack Space Recycle) reuses obsolete space without cleaning operations. @@ -201,6 +188,7 @@ struct victim_sel_policy { unsigned int min_segno; /* segment # having min. cost */ unsigned long long age; /* mtime of GCed section*/ unsigned long long age_threshold;/* age threshold */ + bool one_time_gc; /* one time GC */ }; struct seg_entry { @@ -223,6 +211,7 @@ struct seg_entry { struct sec_entry { unsigned int valid_blocks; /* # of valid blocks in a section */ + unsigned int ckpt_valid_blocks; /* # of valid blocks last cp in a section */ }; #define MAX_SKIP_GC_COUNT 16 @@ -359,21 +348,57 @@ static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi, static inline unsigned int get_ckpt_valid_blocks(struct f2fs_sb_info *sbi, unsigned int segno, bool use_section) { - if (use_section && __is_large_section(sbi)) { - unsigned int start_segno = START_SEGNO(segno); - unsigned int blocks = 0; - int i; + if (use_section && __is_large_section(sbi)) + return get_sec_entry(sbi, segno)->ckpt_valid_blocks; + else + return get_seg_entry(sbi, segno)->ckpt_valid_blocks; +} - for (i = 0; i < sbi->segs_per_sec; i++, start_segno++) { - struct seg_entry *se = get_seg_entry(sbi, start_segno); +static inline void set_ckpt_valid_blocks(struct f2fs_sb_info *sbi, + unsigned int segno) +{ + unsigned int secno = GET_SEC_FROM_SEG(sbi, segno); + unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno); + unsigned int blocks = 0; + int i; - blocks += se->ckpt_valid_blocks; - } - return blocks; + for (i = 0; i < SEGS_PER_SEC(sbi); i++, start_segno++) { + struct seg_entry *se = get_seg_entry(sbi, start_segno); + + blocks += se->ckpt_valid_blocks; } - return get_seg_entry(sbi, segno)->ckpt_valid_blocks; + get_sec_entry(sbi, segno)->ckpt_valid_blocks = blocks; } +#ifdef CONFIG_F2FS_CHECK_FS +static inline void sanity_check_valid_blocks(struct f2fs_sb_info *sbi, + unsigned int segno) +{ + unsigned int secno = GET_SEC_FROM_SEG(sbi, segno); + unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno); + unsigned int blocks = 0; + int i; + + for (i = 0; i < SEGS_PER_SEC(sbi); i++, start_segno++) { + struct seg_entry *se = get_seg_entry(sbi, start_segno); + + blocks += se->ckpt_valid_blocks; + } + + if (blocks != get_sec_entry(sbi, segno)->ckpt_valid_blocks) { + f2fs_err(sbi, + "Inconsistent ckpt valid blocks: " + "seg entry(%d) vs sec entry(%d) at secno %d", + blocks, get_sec_entry(sbi, segno)->ckpt_valid_blocks, secno); + f2fs_bug_on(sbi, 1); + } +} +#else +static inline void sanity_check_valid_blocks(struct f2fs_sb_info *sbi, + unsigned int segno) +{ +} +#endif static inline void seg_info_from_raw_sit(struct seg_entry *se, struct f2fs_sit_entry *rs) { @@ -398,8 +423,8 @@ static inline void __seg_info_to_raw_sit(struct seg_entry *se, 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) +static inline void seg_info_to_sit_folio(struct f2fs_sb_info *sbi, + struct folio *folio, unsigned int start) { struct f2fs_sit_block *raw_sit; struct seg_entry *se; @@ -408,7 +433,7 @@ static inline void seg_info_to_sit_page(struct f2fs_sb_info *sbi, (unsigned long)MAIN_SEGS(sbi)); int i; - raw_sit = (struct f2fs_sit_block *)page_address(page); + raw_sit = folio_address(folio); memset(raw_sit, 0, PAGE_SIZE); for (i = 0; i < end - start; i++) { rs = &raw_sit->entries[i]; @@ -442,15 +467,14 @@ static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno) unsigned int secno = GET_SEC_FROM_SEG(sbi, segno); unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno); unsigned int next; - unsigned int usable_segs = f2fs_usable_segs_in_sec(sbi, segno); spin_lock(&free_i->segmap_lock); clear_bit(segno, free_i->free_segmap); free_i->free_segments++; next = find_next_bit(free_i->free_segmap, - start_segno + sbi->segs_per_sec, start_segno); - if (next >= start_segno + usable_segs) { + start_segno + SEGS_PER_SEC(sbi), start_segno); + if (next >= start_segno + f2fs_usable_segs_in_sec(sbi)) { clear_bit(secno, free_i->free_secmap); free_i->free_sections++; } @@ -476,22 +500,36 @@ static inline void __set_test_and_free(struct f2fs_sb_info *sbi, unsigned int secno = GET_SEC_FROM_SEG(sbi, segno); unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno); unsigned int next; - unsigned int usable_segs = f2fs_usable_segs_in_sec(sbi, segno); + bool ret; spin_lock(&free_i->segmap_lock); - if (test_and_clear_bit(segno, free_i->free_segmap)) { - free_i->free_segments++; - - if (!inmem && IS_CURSEC(sbi, secno)) - goto skip_free; - next = find_next_bit(free_i->free_segmap, - start_segno + sbi->segs_per_sec, start_segno); - if (next >= start_segno + usable_segs) { - if (test_and_clear_bit(secno, free_i->free_secmap)) - free_i->free_sections++; - } - } -skip_free: + ret = test_and_clear_bit(segno, free_i->free_segmap); + if (!ret) + goto unlock_out; + + free_i->free_segments++; + + if (!inmem && IS_CURSEC(sbi, secno)) + goto unlock_out; + + /* check large section */ + next = find_next_bit(free_i->free_segmap, + start_segno + SEGS_PER_SEC(sbi), start_segno); + if (next < start_segno + f2fs_usable_segs_in_sec(sbi)) + goto unlock_out; + + ret = test_and_clear_bit(secno, free_i->free_secmap); + if (!ret) + goto unlock_out; + + free_i->free_sections++; + + if (GET_SEC_FROM_SEG(sbi, sbi->next_victim_seg[BG_GC]) == secno) + sbi->next_victim_seg[BG_GC] = NULL_SEGNO; + if (GET_SEC_FROM_SEG(sbi, sbi->next_victim_seg[FG_GC]) == secno) + sbi->next_victim_seg[FG_GC] = NULL_SEGNO; + +unlock_out: spin_unlock(&free_i->segmap_lock); } @@ -535,8 +573,7 @@ static inline unsigned int free_segments(struct f2fs_sb_info *sbi) static inline unsigned int reserved_segments(struct f2fs_sb_info *sbi) { - return SM_I(sbi)->reserved_segments + - SM_I(sbi)->additional_reserved_segments; + return SM_I(sbi)->reserved_segments; } static inline unsigned int free_sections(struct f2fs_sb_info *sbi) @@ -570,34 +607,57 @@ static inline int reserved_sections(struct f2fs_sb_info *sbi) } static inline bool has_curseg_enough_space(struct f2fs_sb_info *sbi, - unsigned int node_blocks, unsigned int dent_blocks) + unsigned int node_blocks, unsigned int data_blocks, + unsigned int dent_blocks) { - - unsigned int segno, left_blocks; + unsigned int segno, left_blocks, blocks; int i; - /* check current node segment */ - for (i = CURSEG_HOT_NODE; i <= CURSEG_COLD_NODE; i++) { + /* check current data/node sections in the worst case. */ + for (i = CURSEG_HOT_DATA; i < NR_PERSISTENT_LOG; i++) { segno = CURSEG_I(sbi, i)->segno; - left_blocks = f2fs_usable_blks_in_seg(sbi, segno) - - get_seg_entry(sbi, segno)->ckpt_valid_blocks; - if (node_blocks > left_blocks) + if (unlikely(segno == NULL_SEGNO)) + return false; + + if (f2fs_lfs_mode(sbi) && __is_large_section(sbi)) { + left_blocks = CAP_BLKS_PER_SEC(sbi) - + SEGS_TO_BLKS(sbi, (segno - GET_START_SEG_FROM_SEC(sbi, segno))) - + CURSEG_I(sbi, i)->next_blkoff; + } else { + left_blocks = CAP_BLKS_PER_SEC(sbi) - + get_ckpt_valid_blocks(sbi, segno, true); + } + + blocks = i <= CURSEG_COLD_DATA ? data_blocks : node_blocks; + if (blocks > left_blocks) return false; } - /* check current data segment */ + /* check current data section for dentry blocks. */ segno = CURSEG_I(sbi, CURSEG_HOT_DATA)->segno; - left_blocks = f2fs_usable_blks_in_seg(sbi, segno) - - get_seg_entry(sbi, segno)->ckpt_valid_blocks; + + if (unlikely(segno == NULL_SEGNO)) + return false; + + if (f2fs_lfs_mode(sbi) && __is_large_section(sbi)) { + left_blocks = CAP_BLKS_PER_SEC(sbi) - + SEGS_TO_BLKS(sbi, (segno - GET_START_SEG_FROM_SEC(sbi, segno))) - + CURSEG_I(sbi, CURSEG_HOT_DATA)->next_blkoff; + } else { + left_blocks = CAP_BLKS_PER_SEC(sbi) - + get_ckpt_valid_blocks(sbi, segno, true); + } + if (dent_blocks > left_blocks) return false; return true; } /* - * calculate needed sections for dirty node/dentry - * and call has_curseg_enough_space + * calculate needed sections for dirty node/dentry and call + * has_curseg_enough_space, please note that, it needs to account + * dirty data as well in lfs mode when checkpoint is disabled. */ static inline void __get_secs_required(struct f2fs_sb_info *sbi, unsigned int *lower_p, unsigned int *upper_p, bool *curseg_p) @@ -606,19 +666,30 @@ static inline void __get_secs_required(struct f2fs_sb_info *sbi, get_pages(sbi, F2FS_DIRTY_DENTS) + get_pages(sbi, F2FS_DIRTY_IMETA); unsigned int total_dent_blocks = get_pages(sbi, F2FS_DIRTY_DENTS); + unsigned int total_data_blocks = 0; unsigned int node_secs = total_node_blocks / CAP_BLKS_PER_SEC(sbi); unsigned int dent_secs = total_dent_blocks / CAP_BLKS_PER_SEC(sbi); + unsigned int data_secs = 0; unsigned int node_blocks = total_node_blocks % CAP_BLKS_PER_SEC(sbi); unsigned int dent_blocks = total_dent_blocks % CAP_BLKS_PER_SEC(sbi); + unsigned int data_blocks = 0; + + if (f2fs_lfs_mode(sbi) && + unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) { + total_data_blocks = get_pages(sbi, F2FS_DIRTY_DATA); + data_secs = total_data_blocks / CAP_BLKS_PER_SEC(sbi); + data_blocks = total_data_blocks % CAP_BLKS_PER_SEC(sbi); + } if (lower_p) - *lower_p = node_secs + dent_secs; + *lower_p = node_secs + dent_secs + data_secs; if (upper_p) *upper_p = node_secs + dent_secs + - (node_blocks ? 1 : 0) + (dent_blocks ? 1 : 0); + (node_blocks ? 1 : 0) + (dent_blocks ? 1 : 0) + + (data_blocks ? 1 : 0); if (curseg_p) *curseg_p = has_curseg_enough_space(sbi, - node_blocks, dent_blocks); + node_blocks, data_blocks, dent_blocks); } static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi, @@ -638,7 +709,7 @@ static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi, if (free_secs > upper_secs) return false; - else if (free_secs <= lower_secs) + if (free_secs <= lower_secs) return true; return !curseg_space; } @@ -649,12 +720,30 @@ static inline bool has_enough_free_secs(struct f2fs_sb_info *sbi, return !has_not_enough_free_secs(sbi, freed, needed); } +static inline bool has_enough_free_blks(struct f2fs_sb_info *sbi) +{ + unsigned int total_free_blocks = 0; + unsigned int avail_user_block_count; + + spin_lock(&sbi->stat_lock); + + avail_user_block_count = get_available_block_count(sbi, NULL, true); + total_free_blocks = avail_user_block_count - (unsigned int)valid_user_blocks(sbi); + + spin_unlock(&sbi->stat_lock); + + return total_free_blocks > 0; +} + static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi) { if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED))) return true; if (likely(has_enough_free_secs(sbi, 0, 0))) return true; + if (!f2fs_lfs_mode(sbi) && + likely(has_enough_free_blks(sbi))) + return true; return false; } @@ -793,10 +882,10 @@ static inline int check_block_count(struct f2fs_sb_info *sbi, return -EFSCORRUPTED; } - if (usable_blks_per_seg < sbi->blocks_per_seg) + if (usable_blks_per_seg < BLKS_PER_SEG(sbi)) f2fs_bug_on(sbi, find_next_bit_le(&raw_sit->valid_map, - sbi->blocks_per_seg, - usable_blks_per_seg) != sbi->blocks_per_seg); + BLKS_PER_SEG(sbi), + usable_blks_per_seg) != BLKS_PER_SEG(sbi)); /* check segment usage, and check boundary of a given segment number */ if (unlikely(GET_SIT_VBLOCKS(raw_sit) > usable_blks_per_seg @@ -915,9 +1004,9 @@ static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type) return 0; if (type == DATA) - return sbi->blocks_per_seg; + return BLKS_PER_SEG(sbi); else if (type == NODE) - return 8 * sbi->blocks_per_seg; + return SEGS_TO_BLKS(sbi, 8); else if (type == META) return 8 * BIO_MAX_VECS; else |