1 /* SPDX-License-Identifier: GPL-2.0 */ 2 /* 3 * fs/f2fs/f2fs.h 4 * 5 * Copyright (c) 2012 Samsung Electronics Co., Ltd. 6 * http://www.samsung.com/ 7 */ 8 #ifndef _LINUX_F2FS_H 9 #define _LINUX_F2FS_H 10 11 #include <linux/uio.h> 12 #include <linux/types.h> 13 #include <linux/page-flags.h> 14 #include <linux/buffer_head.h> 15 #include <linux/slab.h> 16 #include <linux/crc32.h> 17 #include <linux/magic.h> 18 #include <linux/kobject.h> 19 #include <linux/sched.h> 20 #include <linux/cred.h> 21 #include <linux/sched/mm.h> 22 #include <linux/vmalloc.h> 23 #include <linux/bio.h> 24 #include <linux/blkdev.h> 25 #include <linux/quotaops.h> 26 #include <linux/part_stat.h> 27 #include <crypto/hash.h> 28 29 #include <linux/fscrypt.h> 30 #include <linux/fsverity.h> 31 32 struct pagevec; 33 34 #ifdef CONFIG_F2FS_CHECK_FS 35 #define f2fs_bug_on(sbi, condition) BUG_ON(condition) 36 #else 37 #define f2fs_bug_on(sbi, condition) \ 38 do { \ 39 if (WARN_ON(condition)) \ 40 set_sbi_flag(sbi, SBI_NEED_FSCK); \ 41 } while (0) 42 #endif 43 44 enum { 45 FAULT_KMALLOC, 46 FAULT_KVMALLOC, 47 FAULT_PAGE_ALLOC, 48 FAULT_PAGE_GET, 49 FAULT_ALLOC_BIO, /* it's obsolete due to bio_alloc() will never fail */ 50 FAULT_ALLOC_NID, 51 FAULT_ORPHAN, 52 FAULT_BLOCK, 53 FAULT_DIR_DEPTH, 54 FAULT_EVICT_INODE, 55 FAULT_TRUNCATE, 56 FAULT_READ_IO, 57 FAULT_CHECKPOINT, 58 FAULT_DISCARD, 59 FAULT_WRITE_IO, 60 FAULT_SLAB_ALLOC, 61 FAULT_DQUOT_INIT, 62 FAULT_LOCK_OP, 63 FAULT_BLKADDR, 64 FAULT_MAX, 65 }; 66 67 #ifdef CONFIG_F2FS_FAULT_INJECTION 68 #define F2FS_ALL_FAULT_TYPE (GENMASK(FAULT_MAX - 1, 0)) 69 70 struct f2fs_fault_info { 71 atomic_t inject_ops; 72 unsigned int inject_rate; 73 unsigned int inject_type; 74 }; 75 76 extern const char *f2fs_fault_name[FAULT_MAX]; 77 #define IS_FAULT_SET(fi, type) ((fi)->inject_type & BIT(type)) 78 #endif 79 80 /* 81 * For mount options 82 */ 83 #define F2FS_MOUNT_DISABLE_ROLL_FORWARD 0x00000002 84 #define F2FS_MOUNT_DISCARD 0x00000004 85 #define F2FS_MOUNT_NOHEAP 0x00000008 86 #define F2FS_MOUNT_XATTR_USER 0x00000010 87 #define F2FS_MOUNT_POSIX_ACL 0x00000020 88 #define F2FS_MOUNT_DISABLE_EXT_IDENTIFY 0x00000040 89 #define F2FS_MOUNT_INLINE_XATTR 0x00000080 90 #define F2FS_MOUNT_INLINE_DATA 0x00000100 91 #define F2FS_MOUNT_INLINE_DENTRY 0x00000200 92 #define F2FS_MOUNT_FLUSH_MERGE 0x00000400 93 #define F2FS_MOUNT_NOBARRIER 0x00000800 94 #define F2FS_MOUNT_FASTBOOT 0x00001000 95 #define F2FS_MOUNT_READ_EXTENT_CACHE 0x00002000 96 #define F2FS_MOUNT_DATA_FLUSH 0x00008000 97 #define F2FS_MOUNT_FAULT_INJECTION 0x00010000 98 #define F2FS_MOUNT_USRQUOTA 0x00080000 99 #define F2FS_MOUNT_GRPQUOTA 0x00100000 100 #define F2FS_MOUNT_PRJQUOTA 0x00200000 101 #define F2FS_MOUNT_QUOTA 0x00400000 102 #define F2FS_MOUNT_INLINE_XATTR_SIZE 0x00800000 103 #define F2FS_MOUNT_RESERVE_ROOT 0x01000000 104 #define F2FS_MOUNT_DISABLE_CHECKPOINT 0x02000000 105 #define F2FS_MOUNT_NORECOVERY 0x04000000 106 #define F2FS_MOUNT_ATGC 0x08000000 107 #define F2FS_MOUNT_MERGE_CHECKPOINT 0x10000000 108 #define F2FS_MOUNT_GC_MERGE 0x20000000 109 #define F2FS_MOUNT_COMPRESS_CACHE 0x40000000 110 #define F2FS_MOUNT_AGE_EXTENT_CACHE 0x80000000 111 112 #define F2FS_OPTION(sbi) ((sbi)->mount_opt) 113 #define clear_opt(sbi, option) (F2FS_OPTION(sbi).opt &= ~F2FS_MOUNT_##option) 114 #define set_opt(sbi, option) (F2FS_OPTION(sbi).opt |= F2FS_MOUNT_##option) 115 #define test_opt(sbi, option) (F2FS_OPTION(sbi).opt & F2FS_MOUNT_##option) 116 117 #define ver_after(a, b) (typecheck(unsigned long long, a) && \ 118 typecheck(unsigned long long, b) && \ 119 ((long long)((a) - (b)) > 0)) 120 121 typedef u32 block_t; /* 122 * should not change u32, since it is the on-disk block 123 * address format, __le32. 124 */ 125 typedef u32 nid_t; 126 127 #define COMPRESS_EXT_NUM 16 128 129 /* 130 * An implementation of an rwsem that is explicitly unfair to readers. This 131 * prevents priority inversion when a low-priority reader acquires the read lock 132 * while sleeping on the write lock but the write lock is needed by 133 * higher-priority clients. 134 */ 135 136 struct f2fs_rwsem { 137 struct rw_semaphore internal_rwsem; 138 #ifdef CONFIG_F2FS_UNFAIR_RWSEM 139 wait_queue_head_t read_waiters; 140 #endif 141 }; 142 143 struct f2fs_mount_info { 144 unsigned int opt; 145 int write_io_size_bits; /* Write IO size bits */ 146 block_t root_reserved_blocks; /* root reserved blocks */ 147 kuid_t s_resuid; /* reserved blocks for uid */ 148 kgid_t s_resgid; /* reserved blocks for gid */ 149 int active_logs; /* # of active logs */ 150 int inline_xattr_size; /* inline xattr size */ 151 #ifdef CONFIG_F2FS_FAULT_INJECTION 152 struct f2fs_fault_info fault_info; /* For fault injection */ 153 #endif 154 #ifdef CONFIG_QUOTA 155 /* Names of quota files with journalled quota */ 156 char *s_qf_names[MAXQUOTAS]; 157 int s_jquota_fmt; /* Format of quota to use */ 158 #endif 159 /* For which write hints are passed down to block layer */ 160 int alloc_mode; /* segment allocation policy */ 161 int fsync_mode; /* fsync policy */ 162 int fs_mode; /* fs mode: LFS or ADAPTIVE */ 163 int bggc_mode; /* bggc mode: off, on or sync */ 164 int memory_mode; /* memory mode */ 165 int discard_unit; /* 166 * discard command's offset/size should 167 * be aligned to this unit: block, 168 * segment or section 169 */ 170 struct fscrypt_dummy_policy dummy_enc_policy; /* test dummy encryption */ 171 block_t unusable_cap_perc; /* percentage for cap */ 172 block_t unusable_cap; /* Amount of space allowed to be 173 * unusable when disabling checkpoint 174 */ 175 176 /* For compression */ 177 unsigned char compress_algorithm; /* algorithm type */ 178 unsigned char compress_log_size; /* cluster log size */ 179 unsigned char compress_level; /* compress level */ 180 bool compress_chksum; /* compressed data chksum */ 181 unsigned char compress_ext_cnt; /* extension count */ 182 unsigned char nocompress_ext_cnt; /* nocompress extension count */ 183 int compress_mode; /* compression mode */ 184 unsigned char extensions[COMPRESS_EXT_NUM][F2FS_EXTENSION_LEN]; /* extensions */ 185 unsigned char noextensions[COMPRESS_EXT_NUM][F2FS_EXTENSION_LEN]; /* extensions */ 186 }; 187 188 #define F2FS_FEATURE_ENCRYPT 0x0001 189 #define F2FS_FEATURE_BLKZONED 0x0002 190 #define F2FS_FEATURE_ATOMIC_WRITE 0x0004 191 #define F2FS_FEATURE_EXTRA_ATTR 0x0008 192 #define F2FS_FEATURE_PRJQUOTA 0x0010 193 #define F2FS_FEATURE_INODE_CHKSUM 0x0020 194 #define F2FS_FEATURE_FLEXIBLE_INLINE_XATTR 0x0040 195 #define F2FS_FEATURE_QUOTA_INO 0x0080 196 #define F2FS_FEATURE_INODE_CRTIME 0x0100 197 #define F2FS_FEATURE_LOST_FOUND 0x0200 198 #define F2FS_FEATURE_VERITY 0x0400 199 #define F2FS_FEATURE_SB_CHKSUM 0x0800 200 #define F2FS_FEATURE_CASEFOLD 0x1000 201 #define F2FS_FEATURE_COMPRESSION 0x2000 202 #define F2FS_FEATURE_RO 0x4000 203 204 #define __F2FS_HAS_FEATURE(raw_super, mask) \ 205 ((raw_super->feature & cpu_to_le32(mask)) != 0) 206 #define F2FS_HAS_FEATURE(sbi, mask) __F2FS_HAS_FEATURE(sbi->raw_super, mask) 207 208 /* 209 * Default values for user and/or group using reserved blocks 210 */ 211 #define F2FS_DEF_RESUID 0 212 #define F2FS_DEF_RESGID 0 213 214 /* 215 * For checkpoint manager 216 */ 217 enum { 218 NAT_BITMAP, 219 SIT_BITMAP 220 }; 221 222 #define CP_UMOUNT 0x00000001 223 #define CP_FASTBOOT 0x00000002 224 #define CP_SYNC 0x00000004 225 #define CP_RECOVERY 0x00000008 226 #define CP_DISCARD 0x00000010 227 #define CP_TRIMMED 0x00000020 228 #define CP_PAUSE 0x00000040 229 #define CP_RESIZE 0x00000080 230 231 #define DEF_MAX_DISCARD_REQUEST 8 /* issue 8 discards per round */ 232 #define DEF_MIN_DISCARD_ISSUE_TIME 50 /* 50 ms, if exists */ 233 #define DEF_MID_DISCARD_ISSUE_TIME 500 /* 500 ms, if device busy */ 234 #define DEF_MAX_DISCARD_ISSUE_TIME 60000 /* 60 s, if no candidates */ 235 #define DEF_DISCARD_URGENT_UTIL 80 /* do more discard over 80% */ 236 #define DEF_CP_INTERVAL 60 /* 60 secs */ 237 #define DEF_IDLE_INTERVAL 5 /* 5 secs */ 238 #define DEF_DISABLE_INTERVAL 5 /* 5 secs */ 239 #define DEF_DISABLE_QUICK_INTERVAL 1 /* 1 secs */ 240 #define DEF_UMOUNT_DISCARD_TIMEOUT 5 /* 5 secs */ 241 242 struct cp_control { 243 int reason; 244 __u64 trim_start; 245 __u64 trim_end; 246 __u64 trim_minlen; 247 }; 248 249 /* 250 * indicate meta/data type 251 */ 252 enum { 253 META_CP, 254 META_NAT, 255 META_SIT, 256 META_SSA, 257 META_MAX, 258 META_POR, 259 DATA_GENERIC, /* check range only */ 260 DATA_GENERIC_ENHANCE, /* strong check on range and segment bitmap */ 261 DATA_GENERIC_ENHANCE_READ, /* 262 * strong check on range and segment 263 * bitmap but no warning due to race 264 * condition of read on truncated area 265 * by extent_cache 266 */ 267 DATA_GENERIC_ENHANCE_UPDATE, /* 268 * strong check on range and segment 269 * bitmap for update case 270 */ 271 META_GENERIC, 272 }; 273 274 /* for the list of ino */ 275 enum { 276 ORPHAN_INO, /* for orphan ino list */ 277 APPEND_INO, /* for append ino list */ 278 UPDATE_INO, /* for update ino list */ 279 TRANS_DIR_INO, /* for transactions dir ino list */ 280 FLUSH_INO, /* for multiple device flushing */ 281 MAX_INO_ENTRY, /* max. list */ 282 }; 283 284 struct ino_entry { 285 struct list_head list; /* list head */ 286 nid_t ino; /* inode number */ 287 unsigned int dirty_device; /* dirty device bitmap */ 288 }; 289 290 /* for the list of inodes to be GCed */ 291 struct inode_entry { 292 struct list_head list; /* list head */ 293 struct inode *inode; /* vfs inode pointer */ 294 }; 295 296 struct fsync_node_entry { 297 struct list_head list; /* list head */ 298 struct page *page; /* warm node page pointer */ 299 unsigned int seq_id; /* sequence id */ 300 }; 301 302 struct ckpt_req { 303 struct completion wait; /* completion for checkpoint done */ 304 struct llist_node llnode; /* llist_node to be linked in wait queue */ 305 int ret; /* return code of checkpoint */ 306 ktime_t queue_time; /* request queued time */ 307 }; 308 309 struct ckpt_req_control { 310 struct task_struct *f2fs_issue_ckpt; /* checkpoint task */ 311 int ckpt_thread_ioprio; /* checkpoint merge thread ioprio */ 312 wait_queue_head_t ckpt_wait_queue; /* waiting queue for wake-up */ 313 atomic_t issued_ckpt; /* # of actually issued ckpts */ 314 atomic_t total_ckpt; /* # of total ckpts */ 315 atomic_t queued_ckpt; /* # of queued ckpts */ 316 struct llist_head issue_list; /* list for command issue */ 317 spinlock_t stat_lock; /* lock for below checkpoint time stats */ 318 unsigned int cur_time; /* cur wait time in msec for currently issued checkpoint */ 319 unsigned int peak_time; /* peak wait time in msec until now */ 320 }; 321 322 /* for the bitmap indicate blocks to be discarded */ 323 struct discard_entry { 324 struct list_head list; /* list head */ 325 block_t start_blkaddr; /* start blockaddr of current segment */ 326 unsigned char discard_map[SIT_VBLOCK_MAP_SIZE]; /* segment discard bitmap */ 327 }; 328 329 /* minimum discard granularity, unit: block count */ 330 #define MIN_DISCARD_GRANULARITY 1 331 /* default discard granularity of inner discard thread, unit: block count */ 332 #define DEFAULT_DISCARD_GRANULARITY 16 333 /* default maximum discard granularity of ordered discard, unit: block count */ 334 #define DEFAULT_MAX_ORDERED_DISCARD_GRANULARITY 16 335 336 /* max discard pend list number */ 337 #define MAX_PLIST_NUM 512 338 #define plist_idx(blk_num) ((blk_num) >= MAX_PLIST_NUM ? \ 339 (MAX_PLIST_NUM - 1) : ((blk_num) - 1)) 340 341 enum { 342 D_PREP, /* initial */ 343 D_PARTIAL, /* partially submitted */ 344 D_SUBMIT, /* all submitted */ 345 D_DONE, /* finished */ 346 }; 347 348 struct discard_info { 349 block_t lstart; /* logical start address */ 350 block_t len; /* length */ 351 block_t start; /* actual start address in dev */ 352 }; 353 354 struct discard_cmd { 355 struct rb_node rb_node; /* rb node located in rb-tree */ 356 struct discard_info di; /* discard info */ 357 struct list_head list; /* command list */ 358 struct completion wait; /* compleation */ 359 struct block_device *bdev; /* bdev */ 360 unsigned short ref; /* reference count */ 361 unsigned char state; /* state */ 362 unsigned char queued; /* queued discard */ 363 int error; /* bio error */ 364 spinlock_t lock; /* for state/bio_ref updating */ 365 unsigned short bio_ref; /* bio reference count */ 366 }; 367 368 enum { 369 DPOLICY_BG, 370 DPOLICY_FORCE, 371 DPOLICY_FSTRIM, 372 DPOLICY_UMOUNT, 373 MAX_DPOLICY, 374 }; 375 376 struct discard_policy { 377 int type; /* type of discard */ 378 unsigned int min_interval; /* used for candidates exist */ 379 unsigned int mid_interval; /* used for device busy */ 380 unsigned int max_interval; /* used for candidates not exist */ 381 unsigned int max_requests; /* # of discards issued per round */ 382 unsigned int io_aware_gran; /* minimum granularity discard not be aware of I/O */ 383 bool io_aware; /* issue discard in idle time */ 384 bool sync; /* submit discard with REQ_SYNC flag */ 385 bool ordered; /* issue discard by lba order */ 386 bool timeout; /* discard timeout for put_super */ 387 unsigned int granularity; /* discard granularity */ 388 }; 389 390 struct discard_cmd_control { 391 struct task_struct *f2fs_issue_discard; /* discard thread */ 392 struct list_head entry_list; /* 4KB discard entry list */ 393 struct list_head pend_list[MAX_PLIST_NUM];/* store pending entries */ 394 struct list_head wait_list; /* store on-flushing entries */ 395 struct list_head fstrim_list; /* in-flight discard from fstrim */ 396 wait_queue_head_t discard_wait_queue; /* waiting queue for wake-up */ 397 struct mutex cmd_lock; 398 unsigned int nr_discards; /* # of discards in the list */ 399 unsigned int max_discards; /* max. discards to be issued */ 400 unsigned int max_discard_request; /* max. discard request per round */ 401 unsigned int min_discard_issue_time; /* min. interval between discard issue */ 402 unsigned int mid_discard_issue_time; /* mid. interval between discard issue */ 403 unsigned int max_discard_issue_time; /* max. interval between discard issue */ 404 unsigned int discard_io_aware_gran; /* minimum discard granularity not be aware of I/O */ 405 unsigned int discard_urgent_util; /* utilization which issue discard proactively */ 406 unsigned int discard_granularity; /* discard granularity */ 407 unsigned int max_ordered_discard; /* maximum discard granularity issued by lba order */ 408 unsigned int undiscard_blks; /* # of undiscard blocks */ 409 unsigned int next_pos; /* next discard position */ 410 atomic_t issued_discard; /* # of issued discard */ 411 atomic_t queued_discard; /* # of queued discard */ 412 atomic_t discard_cmd_cnt; /* # of cached cmd count */ 413 struct rb_root_cached root; /* root of discard rb-tree */ 414 bool rbtree_check; /* config for consistence check */ 415 bool discard_wake; /* to wake up discard thread */ 416 }; 417 418 /* for the list of fsync inodes, used only during recovery */ 419 struct fsync_inode_entry { 420 struct list_head list; /* list head */ 421 struct inode *inode; /* vfs inode pointer */ 422 block_t blkaddr; /* block address locating the last fsync */ 423 block_t last_dentry; /* block address locating the last dentry */ 424 }; 425 426 #define nats_in_cursum(jnl) (le16_to_cpu((jnl)->n_nats)) 427 #define sits_in_cursum(jnl) (le16_to_cpu((jnl)->n_sits)) 428 429 #define nat_in_journal(jnl, i) ((jnl)->nat_j.entries[i].ne) 430 #define nid_in_journal(jnl, i) ((jnl)->nat_j.entries[i].nid) 431 #define sit_in_journal(jnl, i) ((jnl)->sit_j.entries[i].se) 432 #define segno_in_journal(jnl, i) ((jnl)->sit_j.entries[i].segno) 433 434 #define MAX_NAT_JENTRIES(jnl) (NAT_JOURNAL_ENTRIES - nats_in_cursum(jnl)) 435 #define MAX_SIT_JENTRIES(jnl) (SIT_JOURNAL_ENTRIES - sits_in_cursum(jnl)) 436 437 static inline int update_nats_in_cursum(struct f2fs_journal *journal, int i) 438 { 439 int before = nats_in_cursum(journal); 440 441 journal->n_nats = cpu_to_le16(before + i); 442 return before; 443 } 444 445 static inline int update_sits_in_cursum(struct f2fs_journal *journal, int i) 446 { 447 int before = sits_in_cursum(journal); 448 449 journal->n_sits = cpu_to_le16(before + i); 450 return before; 451 } 452 453 static inline bool __has_cursum_space(struct f2fs_journal *journal, 454 int size, int type) 455 { 456 if (type == NAT_JOURNAL) 457 return size <= MAX_NAT_JENTRIES(journal); 458 return size <= MAX_SIT_JENTRIES(journal); 459 } 460 461 /* for inline stuff */ 462 #define DEF_INLINE_RESERVED_SIZE 1 463 static inline int get_extra_isize(struct inode *inode); 464 static inline int get_inline_xattr_addrs(struct inode *inode); 465 #define MAX_INLINE_DATA(inode) (sizeof(__le32) * \ 466 (CUR_ADDRS_PER_INODE(inode) - \ 467 get_inline_xattr_addrs(inode) - \ 468 DEF_INLINE_RESERVED_SIZE)) 469 470 /* for inline dir */ 471 #define NR_INLINE_DENTRY(inode) (MAX_INLINE_DATA(inode) * BITS_PER_BYTE / \ 472 ((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \ 473 BITS_PER_BYTE + 1)) 474 #define INLINE_DENTRY_BITMAP_SIZE(inode) \ 475 DIV_ROUND_UP(NR_INLINE_DENTRY(inode), BITS_PER_BYTE) 476 #define INLINE_RESERVED_SIZE(inode) (MAX_INLINE_DATA(inode) - \ 477 ((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \ 478 NR_INLINE_DENTRY(inode) + \ 479 INLINE_DENTRY_BITMAP_SIZE(inode))) 480 481 /* 482 * For INODE and NODE manager 483 */ 484 /* for directory operations */ 485 486 struct f2fs_filename { 487 /* 488 * The filename the user specified. This is NULL for some 489 * filesystem-internal operations, e.g. converting an inline directory 490 * to a non-inline one, or roll-forward recovering an encrypted dentry. 491 */ 492 const struct qstr *usr_fname; 493 494 /* 495 * The on-disk filename. For encrypted directories, this is encrypted. 496 * This may be NULL for lookups in an encrypted dir without the key. 497 */ 498 struct fscrypt_str disk_name; 499 500 /* The dirhash of this filename */ 501 f2fs_hash_t hash; 502 503 #ifdef CONFIG_FS_ENCRYPTION 504 /* 505 * For lookups in encrypted directories: either the buffer backing 506 * disk_name, or a buffer that holds the decoded no-key name. 507 */ 508 struct fscrypt_str crypto_buf; 509 #endif 510 #if IS_ENABLED(CONFIG_UNICODE) 511 /* 512 * For casefolded directories: the casefolded name, but it's left NULL 513 * if the original name is not valid Unicode, if the original name is 514 * "." or "..", if the directory is both casefolded and encrypted and 515 * its encryption key is unavailable, or if the filesystem is doing an 516 * internal operation where usr_fname is also NULL. In all these cases 517 * we fall back to treating the name as an opaque byte sequence. 518 */ 519 struct fscrypt_str cf_name; 520 #endif 521 }; 522 523 struct f2fs_dentry_ptr { 524 struct inode *inode; 525 void *bitmap; 526 struct f2fs_dir_entry *dentry; 527 __u8 (*filename)[F2FS_SLOT_LEN]; 528 int max; 529 int nr_bitmap; 530 }; 531 532 static inline void make_dentry_ptr_block(struct inode *inode, 533 struct f2fs_dentry_ptr *d, struct f2fs_dentry_block *t) 534 { 535 d->inode = inode; 536 d->max = NR_DENTRY_IN_BLOCK; 537 d->nr_bitmap = SIZE_OF_DENTRY_BITMAP; 538 d->bitmap = t->dentry_bitmap; 539 d->dentry = t->dentry; 540 d->filename = t->filename; 541 } 542 543 static inline void make_dentry_ptr_inline(struct inode *inode, 544 struct f2fs_dentry_ptr *d, void *t) 545 { 546 int entry_cnt = NR_INLINE_DENTRY(inode); 547 int bitmap_size = INLINE_DENTRY_BITMAP_SIZE(inode); 548 int reserved_size = INLINE_RESERVED_SIZE(inode); 549 550 d->inode = inode; 551 d->max = entry_cnt; 552 d->nr_bitmap = bitmap_size; 553 d->bitmap = t; 554 d->dentry = t + bitmap_size + reserved_size; 555 d->filename = t + bitmap_size + reserved_size + 556 SIZE_OF_DIR_ENTRY * entry_cnt; 557 } 558 559 /* 560 * XATTR_NODE_OFFSET stores xattrs to one node block per file keeping -1 561 * as its node offset to distinguish from index node blocks. 562 * But some bits are used to mark the node block. 563 */ 564 #define XATTR_NODE_OFFSET ((((unsigned int)-1) << OFFSET_BIT_SHIFT) \ 565 >> OFFSET_BIT_SHIFT) 566 enum { 567 ALLOC_NODE, /* allocate a new node page if needed */ 568 LOOKUP_NODE, /* look up a node without readahead */ 569 LOOKUP_NODE_RA, /* 570 * look up a node with readahead called 571 * by get_data_block. 572 */ 573 }; 574 575 #define DEFAULT_RETRY_IO_COUNT 8 /* maximum retry read IO or flush count */ 576 577 /* congestion wait timeout value, default: 20ms */ 578 #define DEFAULT_IO_TIMEOUT (msecs_to_jiffies(20)) 579 580 /* maximum retry quota flush count */ 581 #define DEFAULT_RETRY_QUOTA_FLUSH_COUNT 8 582 583 /* maximum retry of EIO'ed page */ 584 #define MAX_RETRY_PAGE_EIO 100 585 586 #define F2FS_LINK_MAX 0xffffffff /* maximum link count per file */ 587 588 #define MAX_DIR_RA_PAGES 4 /* maximum ra pages of dir */ 589 590 /* dirty segments threshold for triggering CP */ 591 #define DEFAULT_DIRTY_THRESHOLD 4 592 593 #define RECOVERY_MAX_RA_BLOCKS BIO_MAX_VECS 594 #define RECOVERY_MIN_RA_BLOCKS 1 595 596 #define F2FS_ONSTACK_PAGES 16 /* nr of onstack pages */ 597 598 /* for in-memory extent cache entry */ 599 #define F2FS_MIN_EXTENT_LEN 64 /* minimum extent length */ 600 601 /* number of extent info in extent cache we try to shrink */ 602 #define READ_EXTENT_CACHE_SHRINK_NUMBER 128 603 604 /* number of age extent info in extent cache we try to shrink */ 605 #define AGE_EXTENT_CACHE_SHRINK_NUMBER 128 606 #define LAST_AGE_WEIGHT 30 607 #define SAME_AGE_REGION 1024 608 609 /* 610 * Define data block with age less than 1GB as hot data 611 * define data block with age less than 10GB but more than 1GB as warm data 612 */ 613 #define DEF_HOT_DATA_AGE_THRESHOLD 262144 614 #define DEF_WARM_DATA_AGE_THRESHOLD 2621440 615 616 /* extent cache type */ 617 enum extent_type { 618 EX_READ, 619 EX_BLOCK_AGE, 620 NR_EXTENT_CACHES, 621 }; 622 623 struct extent_info { 624 unsigned int fofs; /* start offset in a file */ 625 unsigned int len; /* length of the extent */ 626 union { 627 /* read extent_cache */ 628 struct { 629 /* start block address of the extent */ 630 block_t blk; 631 #ifdef CONFIG_F2FS_FS_COMPRESSION 632 /* physical extent length of compressed blocks */ 633 unsigned int c_len; 634 #endif 635 }; 636 /* block age extent_cache */ 637 struct { 638 /* block age of the extent */ 639 unsigned long long age; 640 /* last total blocks allocated */ 641 unsigned long long last_blocks; 642 }; 643 }; 644 }; 645 646 struct extent_node { 647 struct rb_node rb_node; /* rb node located in rb-tree */ 648 struct extent_info ei; /* extent info */ 649 struct list_head list; /* node in global extent list of sbi */ 650 struct extent_tree *et; /* extent tree pointer */ 651 }; 652 653 struct extent_tree { 654 nid_t ino; /* inode number */ 655 enum extent_type type; /* keep the extent tree type */ 656 struct rb_root_cached root; /* root of extent info rb-tree */ 657 struct extent_node *cached_en; /* recently accessed extent node */ 658 struct list_head list; /* to be used by sbi->zombie_list */ 659 rwlock_t lock; /* protect extent info rb-tree */ 660 atomic_t node_cnt; /* # of extent node in rb-tree*/ 661 bool largest_updated; /* largest extent updated */ 662 struct extent_info largest; /* largest cached extent for EX_READ */ 663 }; 664 665 struct extent_tree_info { 666 struct radix_tree_root extent_tree_root;/* cache extent cache entries */ 667 struct mutex extent_tree_lock; /* locking extent radix tree */ 668 struct list_head extent_list; /* lru list for shrinker */ 669 spinlock_t extent_lock; /* locking extent lru list */ 670 atomic_t total_ext_tree; /* extent tree count */ 671 struct list_head zombie_list; /* extent zombie tree list */ 672 atomic_t total_zombie_tree; /* extent zombie tree count */ 673 atomic_t total_ext_node; /* extent info count */ 674 }; 675 676 /* 677 * State of block returned by f2fs_map_blocks. 678 */ 679 #define F2FS_MAP_NEW (1U << 0) 680 #define F2FS_MAP_MAPPED (1U << 1) 681 #define F2FS_MAP_DELALLOC (1U << 2) 682 #define F2FS_MAP_FLAGS (F2FS_MAP_NEW | F2FS_MAP_MAPPED |\ 683 F2FS_MAP_DELALLOC) 684 685 struct f2fs_map_blocks { 686 struct block_device *m_bdev; /* for multi-device dio */ 687 block_t m_pblk; 688 block_t m_lblk; 689 unsigned int m_len; 690 unsigned int m_flags; 691 pgoff_t *m_next_pgofs; /* point next possible non-hole pgofs */ 692 pgoff_t *m_next_extent; /* point to next possible extent */ 693 int m_seg_type; 694 bool m_may_create; /* indicate it is from write path */ 695 bool m_multidev_dio; /* indicate it allows multi-device dio */ 696 }; 697 698 /* for flag in get_data_block */ 699 enum { 700 F2FS_GET_BLOCK_DEFAULT, 701 F2FS_GET_BLOCK_FIEMAP, 702 F2FS_GET_BLOCK_BMAP, 703 F2FS_GET_BLOCK_DIO, 704 F2FS_GET_BLOCK_PRE_DIO, 705 F2FS_GET_BLOCK_PRE_AIO, 706 F2FS_GET_BLOCK_PRECACHE, 707 }; 708 709 /* 710 * i_advise uses FADVISE_XXX_BIT. We can add additional hints later. 711 */ 712 #define FADVISE_COLD_BIT 0x01 713 #define FADVISE_LOST_PINO_BIT 0x02 714 #define FADVISE_ENCRYPT_BIT 0x04 715 #define FADVISE_ENC_NAME_BIT 0x08 716 #define FADVISE_KEEP_SIZE_BIT 0x10 717 #define FADVISE_HOT_BIT 0x20 718 #define FADVISE_VERITY_BIT 0x40 719 #define FADVISE_TRUNC_BIT 0x80 720 721 #define FADVISE_MODIFIABLE_BITS (FADVISE_COLD_BIT | FADVISE_HOT_BIT) 722 723 #define file_is_cold(inode) is_file(inode, FADVISE_COLD_BIT) 724 #define file_set_cold(inode) set_file(inode, FADVISE_COLD_BIT) 725 #define file_clear_cold(inode) clear_file(inode, FADVISE_COLD_BIT) 726 727 #define file_wrong_pino(inode) is_file(inode, FADVISE_LOST_PINO_BIT) 728 #define file_lost_pino(inode) set_file(inode, FADVISE_LOST_PINO_BIT) 729 #define file_got_pino(inode) clear_file(inode, FADVISE_LOST_PINO_BIT) 730 731 #define file_is_encrypt(inode) is_file(inode, FADVISE_ENCRYPT_BIT) 732 #define file_set_encrypt(inode) set_file(inode, FADVISE_ENCRYPT_BIT) 733 734 #define file_enc_name(inode) is_file(inode, FADVISE_ENC_NAME_BIT) 735 #define file_set_enc_name(inode) set_file(inode, FADVISE_ENC_NAME_BIT) 736 737 #define file_keep_isize(inode) is_file(inode, FADVISE_KEEP_SIZE_BIT) 738 #define file_set_keep_isize(inode) set_file(inode, FADVISE_KEEP_SIZE_BIT) 739 740 #define file_is_hot(inode) is_file(inode, FADVISE_HOT_BIT) 741 #define file_set_hot(inode) set_file(inode, FADVISE_HOT_BIT) 742 #define file_clear_hot(inode) clear_file(inode, FADVISE_HOT_BIT) 743 744 #define file_is_verity(inode) is_file(inode, FADVISE_VERITY_BIT) 745 #define file_set_verity(inode) set_file(inode, FADVISE_VERITY_BIT) 746 747 #define file_should_truncate(inode) is_file(inode, FADVISE_TRUNC_BIT) 748 #define file_need_truncate(inode) set_file(inode, FADVISE_TRUNC_BIT) 749 #define file_dont_truncate(inode) clear_file(inode, FADVISE_TRUNC_BIT) 750 751 #define DEF_DIR_LEVEL 0 752 753 enum { 754 GC_FAILURE_PIN, 755 MAX_GC_FAILURE 756 }; 757 758 /* used for f2fs_inode_info->flags */ 759 enum { 760 FI_NEW_INODE, /* indicate newly allocated inode */ 761 FI_DIRTY_INODE, /* indicate inode is dirty or not */ 762 FI_AUTO_RECOVER, /* indicate inode is recoverable */ 763 FI_DIRTY_DIR, /* indicate directory has dirty pages */ 764 FI_INC_LINK, /* need to increment i_nlink */ 765 FI_ACL_MODE, /* indicate acl mode */ 766 FI_NO_ALLOC, /* should not allocate any blocks */ 767 FI_FREE_NID, /* free allocated nide */ 768 FI_NO_EXTENT, /* not to use the extent cache */ 769 FI_INLINE_XATTR, /* used for inline xattr */ 770 FI_INLINE_DATA, /* used for inline data*/ 771 FI_INLINE_DENTRY, /* used for inline dentry */ 772 FI_APPEND_WRITE, /* inode has appended data */ 773 FI_UPDATE_WRITE, /* inode has in-place-update data */ 774 FI_NEED_IPU, /* used for ipu per file */ 775 FI_ATOMIC_FILE, /* indicate atomic file */ 776 FI_FIRST_BLOCK_WRITTEN, /* indicate #0 data block was written */ 777 FI_DROP_CACHE, /* drop dirty page cache */ 778 FI_DATA_EXIST, /* indicate data exists */ 779 FI_INLINE_DOTS, /* indicate inline dot dentries */ 780 FI_SKIP_WRITES, /* should skip data page writeback */ 781 FI_OPU_WRITE, /* used for opu per file */ 782 FI_DIRTY_FILE, /* indicate regular/symlink has dirty pages */ 783 FI_PREALLOCATED_ALL, /* all blocks for write were preallocated */ 784 FI_HOT_DATA, /* indicate file is hot */ 785 FI_EXTRA_ATTR, /* indicate file has extra attribute */ 786 FI_PROJ_INHERIT, /* indicate file inherits projectid */ 787 FI_PIN_FILE, /* indicate file should not be gced */ 788 FI_VERITY_IN_PROGRESS, /* building fs-verity Merkle tree */ 789 FI_COMPRESSED_FILE, /* indicate file's data can be compressed */ 790 FI_COMPRESS_CORRUPT, /* indicate compressed cluster is corrupted */ 791 FI_MMAP_FILE, /* indicate file was mmapped */ 792 FI_ENABLE_COMPRESS, /* enable compression in "user" compression mode */ 793 FI_COMPRESS_RELEASED, /* compressed blocks were released */ 794 FI_ALIGNED_WRITE, /* enable aligned write */ 795 FI_COW_FILE, /* indicate COW file */ 796 FI_ATOMIC_COMMITTED, /* indicate atomic commit completed except disk sync */ 797 FI_ATOMIC_REPLACE, /* indicate atomic replace */ 798 FI_MAX, /* max flag, never be used */ 799 }; 800 801 struct f2fs_inode_info { 802 struct inode vfs_inode; /* serve a vfs inode */ 803 unsigned long i_flags; /* keep an inode flags for ioctl */ 804 unsigned char i_advise; /* use to give file attribute hints */ 805 unsigned char i_dir_level; /* use for dentry level for large dir */ 806 unsigned int i_current_depth; /* only for directory depth */ 807 /* for gc failure statistic */ 808 unsigned int i_gc_failures[MAX_GC_FAILURE]; 809 unsigned int i_pino; /* parent inode number */ 810 umode_t i_acl_mode; /* keep file acl mode temporarily */ 811 812 /* Use below internally in f2fs*/ 813 unsigned long flags[BITS_TO_LONGS(FI_MAX)]; /* use to pass per-file flags */ 814 struct f2fs_rwsem i_sem; /* protect fi info */ 815 atomic_t dirty_pages; /* # of dirty pages */ 816 f2fs_hash_t chash; /* hash value of given file name */ 817 unsigned int clevel; /* maximum level of given file name */ 818 struct task_struct *task; /* lookup and create consistency */ 819 struct task_struct *cp_task; /* separate cp/wb IO stats*/ 820 struct task_struct *wb_task; /* indicate inode is in context of writeback */ 821 nid_t i_xattr_nid; /* node id that contains xattrs */ 822 loff_t last_disk_size; /* lastly written file size */ 823 spinlock_t i_size_lock; /* protect last_disk_size */ 824 825 #ifdef CONFIG_QUOTA 826 struct dquot *i_dquot[MAXQUOTAS]; 827 828 /* quota space reservation, managed internally by quota code */ 829 qsize_t i_reserved_quota; 830 #endif 831 struct list_head dirty_list; /* dirty list for dirs and files */ 832 struct list_head gdirty_list; /* linked in global dirty list */ 833 struct task_struct *atomic_write_task; /* store atomic write task */ 834 struct extent_tree *extent_tree[NR_EXTENT_CACHES]; 835 /* cached extent_tree entry */ 836 struct inode *cow_inode; /* copy-on-write inode for atomic write */ 837 838 /* avoid racing between foreground op and gc */ 839 struct f2fs_rwsem i_gc_rwsem[2]; 840 struct f2fs_rwsem i_xattr_sem; /* avoid racing between reading and changing EAs */ 841 842 int i_extra_isize; /* size of extra space located in i_addr */ 843 kprojid_t i_projid; /* id for project quota */ 844 int i_inline_xattr_size; /* inline xattr size */ 845 struct timespec64 i_crtime; /* inode creation time */ 846 struct timespec64 i_disk_time[3];/* inode disk times */ 847 848 /* for file compress */ 849 atomic_t i_compr_blocks; /* # of compressed blocks */ 850 unsigned char i_compress_algorithm; /* algorithm type */ 851 unsigned char i_log_cluster_size; /* log of cluster size */ 852 unsigned char i_compress_level; /* compress level (lz4hc,zstd) */ 853 unsigned char i_compress_flag; /* compress flag */ 854 unsigned int i_cluster_size; /* cluster size */ 855 856 unsigned int atomic_write_cnt; 857 loff_t original_i_size; /* original i_size before atomic write */ 858 }; 859 860 static inline void get_read_extent_info(struct extent_info *ext, 861 struct f2fs_extent *i_ext) 862 { 863 ext->fofs = le32_to_cpu(i_ext->fofs); 864 ext->blk = le32_to_cpu(i_ext->blk); 865 ext->len = le32_to_cpu(i_ext->len); 866 } 867 868 static inline void set_raw_read_extent(struct extent_info *ext, 869 struct f2fs_extent *i_ext) 870 { 871 i_ext->fofs = cpu_to_le32(ext->fofs); 872 i_ext->blk = cpu_to_le32(ext->blk); 873 i_ext->len = cpu_to_le32(ext->len); 874 } 875 876 static inline bool __is_discard_mergeable(struct discard_info *back, 877 struct discard_info *front, unsigned int max_len) 878 { 879 return (back->lstart + back->len == front->lstart) && 880 (back->len + front->len <= max_len); 881 } 882 883 static inline bool __is_discard_back_mergeable(struct discard_info *cur, 884 struct discard_info *back, unsigned int max_len) 885 { 886 return __is_discard_mergeable(back, cur, max_len); 887 } 888 889 static inline bool __is_discard_front_mergeable(struct discard_info *cur, 890 struct discard_info *front, unsigned int max_len) 891 { 892 return __is_discard_mergeable(cur, front, max_len); 893 } 894 895 /* 896 * For free nid management 897 */ 898 enum nid_state { 899 FREE_NID, /* newly added to free nid list */ 900 PREALLOC_NID, /* it is preallocated */ 901 MAX_NID_STATE, 902 }; 903 904 enum nat_state { 905 TOTAL_NAT, 906 DIRTY_NAT, 907 RECLAIMABLE_NAT, 908 MAX_NAT_STATE, 909 }; 910 911 struct f2fs_nm_info { 912 block_t nat_blkaddr; /* base disk address of NAT */ 913 nid_t max_nid; /* maximum possible node ids */ 914 nid_t available_nids; /* # of available node ids */ 915 nid_t next_scan_nid; /* the next nid to be scanned */ 916 nid_t max_rf_node_blocks; /* max # of nodes for recovery */ 917 unsigned int ram_thresh; /* control the memory footprint */ 918 unsigned int ra_nid_pages; /* # of nid pages to be readaheaded */ 919 unsigned int dirty_nats_ratio; /* control dirty nats ratio threshold */ 920 921 /* NAT cache management */ 922 struct radix_tree_root nat_root;/* root of the nat entry cache */ 923 struct radix_tree_root nat_set_root;/* root of the nat set cache */ 924 struct f2fs_rwsem nat_tree_lock; /* protect nat entry tree */ 925 struct list_head nat_entries; /* cached nat entry list (clean) */ 926 spinlock_t nat_list_lock; /* protect clean nat entry list */ 927 unsigned int nat_cnt[MAX_NAT_STATE]; /* the # of cached nat entries */ 928 unsigned int nat_blocks; /* # of nat blocks */ 929 930 /* free node ids management */ 931 struct radix_tree_root free_nid_root;/* root of the free_nid cache */ 932 struct list_head free_nid_list; /* list for free nids excluding preallocated nids */ 933 unsigned int nid_cnt[MAX_NID_STATE]; /* the number of free node id */ 934 spinlock_t nid_list_lock; /* protect nid lists ops */ 935 struct mutex build_lock; /* lock for build free nids */ 936 unsigned char **free_nid_bitmap; 937 unsigned char *nat_block_bitmap; 938 unsigned short *free_nid_count; /* free nid count of NAT block */ 939 940 /* for checkpoint */ 941 char *nat_bitmap; /* NAT bitmap pointer */ 942 943 unsigned int nat_bits_blocks; /* # of nat bits blocks */ 944 unsigned char *nat_bits; /* NAT bits blocks */ 945 unsigned char *full_nat_bits; /* full NAT pages */ 946 unsigned char *empty_nat_bits; /* empty NAT pages */ 947 #ifdef CONFIG_F2FS_CHECK_FS 948 char *nat_bitmap_mir; /* NAT bitmap mirror */ 949 #endif 950 int bitmap_size; /* bitmap size */ 951 }; 952 953 /* 954 * this structure is used as one of function parameters. 955 * all the information are dedicated to a given direct node block determined 956 * by the data offset in a file. 957 */ 958 struct dnode_of_data { 959 struct inode *inode; /* vfs inode pointer */ 960 struct page *inode_page; /* its inode page, NULL is possible */ 961 struct page *node_page; /* cached direct node page */ 962 nid_t nid; /* node id of the direct node block */ 963 unsigned int ofs_in_node; /* data offset in the node page */ 964 bool inode_page_locked; /* inode page is locked or not */ 965 bool node_changed; /* is node block changed */ 966 char cur_level; /* level of hole node page */ 967 char max_level; /* level of current page located */ 968 block_t data_blkaddr; /* block address of the node block */ 969 }; 970 971 static inline void set_new_dnode(struct dnode_of_data *dn, struct inode *inode, 972 struct page *ipage, struct page *npage, nid_t nid) 973 { 974 memset(dn, 0, sizeof(*dn)); 975 dn->inode = inode; 976 dn->inode_page = ipage; 977 dn->node_page = npage; 978 dn->nid = nid; 979 } 980 981 /* 982 * For SIT manager 983 * 984 * By default, there are 6 active log areas across the whole main area. 985 * When considering hot and cold data separation to reduce cleaning overhead, 986 * we split 3 for data logs and 3 for node logs as hot, warm, and cold types, 987 * respectively. 988 * In the current design, you should not change the numbers intentionally. 989 * Instead, as a mount option such as active_logs=x, you can use 2, 4, and 6 990 * logs individually according to the underlying devices. (default: 6) 991 * Just in case, on-disk layout covers maximum 16 logs that consist of 8 for 992 * data and 8 for node logs. 993 */ 994 #define NR_CURSEG_DATA_TYPE (3) 995 #define NR_CURSEG_NODE_TYPE (3) 996 #define NR_CURSEG_INMEM_TYPE (2) 997 #define NR_CURSEG_RO_TYPE (2) 998 #define NR_CURSEG_PERSIST_TYPE (NR_CURSEG_DATA_TYPE + NR_CURSEG_NODE_TYPE) 999 #define NR_CURSEG_TYPE (NR_CURSEG_INMEM_TYPE + NR_CURSEG_PERSIST_TYPE) 1000 1001 enum { 1002 CURSEG_HOT_DATA = 0, /* directory entry blocks */ 1003 CURSEG_WARM_DATA, /* data blocks */ 1004 CURSEG_COLD_DATA, /* multimedia or GCed data blocks */ 1005 CURSEG_HOT_NODE, /* direct node blocks of directory files */ 1006 CURSEG_WARM_NODE, /* direct node blocks of normal files */ 1007 CURSEG_COLD_NODE, /* indirect node blocks */ 1008 NR_PERSISTENT_LOG, /* number of persistent log */ 1009 CURSEG_COLD_DATA_PINNED = NR_PERSISTENT_LOG, 1010 /* pinned file that needs consecutive block address */ 1011 CURSEG_ALL_DATA_ATGC, /* SSR alloctor in hot/warm/cold data area */ 1012 NO_CHECK_TYPE, /* number of persistent & inmem log */ 1013 }; 1014 1015 struct flush_cmd { 1016 struct completion wait; 1017 struct llist_node llnode; 1018 nid_t ino; 1019 int ret; 1020 }; 1021 1022 struct flush_cmd_control { 1023 struct task_struct *f2fs_issue_flush; /* flush thread */ 1024 wait_queue_head_t flush_wait_queue; /* waiting queue for wake-up */ 1025 atomic_t issued_flush; /* # of issued flushes */ 1026 atomic_t queued_flush; /* # of queued flushes */ 1027 struct llist_head issue_list; /* list for command issue */ 1028 struct llist_node *dispatch_list; /* list for command dispatch */ 1029 }; 1030 1031 struct f2fs_sm_info { 1032 struct sit_info *sit_info; /* whole segment information */ 1033 struct free_segmap_info *free_info; /* free segment information */ 1034 struct dirty_seglist_info *dirty_info; /* dirty segment information */ 1035 struct curseg_info *curseg_array; /* active segment information */ 1036 1037 struct f2fs_rwsem curseg_lock; /* for preventing curseg change */ 1038 1039 block_t seg0_blkaddr; /* block address of 0'th segment */ 1040 block_t main_blkaddr; /* start block address of main area */ 1041 block_t ssa_blkaddr; /* start block address of SSA area */ 1042 1043 unsigned int segment_count; /* total # of segments */ 1044 unsigned int main_segments; /* # of segments in main area */ 1045 unsigned int reserved_segments; /* # of reserved segments */ 1046 unsigned int additional_reserved_segments;/* reserved segs for IO align feature */ 1047 unsigned int ovp_segments; /* # of overprovision segments */ 1048 1049 /* a threshold to reclaim prefree segments */ 1050 unsigned int rec_prefree_segments; 1051 1052 struct list_head sit_entry_set; /* sit entry set list */ 1053 1054 unsigned int ipu_policy; /* in-place-update policy */ 1055 unsigned int min_ipu_util; /* in-place-update threshold */ 1056 unsigned int min_fsync_blocks; /* threshold for fsync */ 1057 unsigned int min_seq_blocks; /* threshold for sequential blocks */ 1058 unsigned int min_hot_blocks; /* threshold for hot block allocation */ 1059 unsigned int min_ssr_sections; /* threshold to trigger SSR allocation */ 1060 1061 /* for flush command control */ 1062 struct flush_cmd_control *fcc_info; 1063 1064 /* for discard command control */ 1065 struct discard_cmd_control *dcc_info; 1066 }; 1067 1068 /* 1069 * For superblock 1070 */ 1071 /* 1072 * COUNT_TYPE for monitoring 1073 * 1074 * f2fs monitors the number of several block types such as on-writeback, 1075 * dirty dentry blocks, dirty node blocks, and dirty meta blocks. 1076 */ 1077 #define WB_DATA_TYPE(p) (__is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA) 1078 enum count_type { 1079 F2FS_DIRTY_DENTS, 1080 F2FS_DIRTY_DATA, 1081 F2FS_DIRTY_QDATA, 1082 F2FS_DIRTY_NODES, 1083 F2FS_DIRTY_META, 1084 F2FS_DIRTY_IMETA, 1085 F2FS_WB_CP_DATA, 1086 F2FS_WB_DATA, 1087 F2FS_RD_DATA, 1088 F2FS_RD_NODE, 1089 F2FS_RD_META, 1090 F2FS_DIO_WRITE, 1091 F2FS_DIO_READ, 1092 NR_COUNT_TYPE, 1093 }; 1094 1095 /* 1096 * The below are the page types of bios used in submit_bio(). 1097 * The available types are: 1098 * DATA User data pages. It operates as async mode. 1099 * NODE Node pages. It operates as async mode. 1100 * META FS metadata pages such as SIT, NAT, CP. 1101 * NR_PAGE_TYPE The number of page types. 1102 * META_FLUSH Make sure the previous pages are written 1103 * with waiting the bio's completion 1104 * ... Only can be used with META. 1105 */ 1106 #define PAGE_TYPE_OF_BIO(type) ((type) > META ? META : (type)) 1107 enum page_type { 1108 DATA = 0, 1109 NODE = 1, /* should not change this */ 1110 META, 1111 NR_PAGE_TYPE, 1112 META_FLUSH, 1113 IPU, /* the below types are used by tracepoints only. */ 1114 OPU, 1115 }; 1116 1117 enum temp_type { 1118 HOT = 0, /* must be zero for meta bio */ 1119 WARM, 1120 COLD, 1121 NR_TEMP_TYPE, 1122 }; 1123 1124 enum need_lock_type { 1125 LOCK_REQ = 0, 1126 LOCK_DONE, 1127 LOCK_RETRY, 1128 }; 1129 1130 enum cp_reason_type { 1131 CP_NO_NEEDED, 1132 CP_NON_REGULAR, 1133 CP_COMPRESSED, 1134 CP_HARDLINK, 1135 CP_SB_NEED_CP, 1136 CP_WRONG_PINO, 1137 CP_NO_SPC_ROLL, 1138 CP_NODE_NEED_CP, 1139 CP_FASTBOOT_MODE, 1140 CP_SPEC_LOG_NUM, 1141 CP_RECOVER_DIR, 1142 }; 1143 1144 enum iostat_type { 1145 /* WRITE IO */ 1146 APP_DIRECT_IO, /* app direct write IOs */ 1147 APP_BUFFERED_IO, /* app buffered write IOs */ 1148 APP_WRITE_IO, /* app write IOs */ 1149 APP_MAPPED_IO, /* app mapped IOs */ 1150 APP_BUFFERED_CDATA_IO, /* app buffered write IOs on compressed file */ 1151 APP_MAPPED_CDATA_IO, /* app mapped write IOs on compressed file */ 1152 FS_DATA_IO, /* data IOs from kworker/fsync/reclaimer */ 1153 FS_CDATA_IO, /* data IOs from kworker/fsync/reclaimer on compressed file */ 1154 FS_NODE_IO, /* node IOs from kworker/fsync/reclaimer */ 1155 FS_META_IO, /* meta IOs from kworker/reclaimer */ 1156 FS_GC_DATA_IO, /* data IOs from forground gc */ 1157 FS_GC_NODE_IO, /* node IOs from forground gc */ 1158 FS_CP_DATA_IO, /* data IOs from checkpoint */ 1159 FS_CP_NODE_IO, /* node IOs from checkpoint */ 1160 FS_CP_META_IO, /* meta IOs from checkpoint */ 1161 1162 /* READ IO */ 1163 APP_DIRECT_READ_IO, /* app direct read IOs */ 1164 APP_BUFFERED_READ_IO, /* app buffered read IOs */ 1165 APP_READ_IO, /* app read IOs */ 1166 APP_MAPPED_READ_IO, /* app mapped read IOs */ 1167 APP_BUFFERED_CDATA_READ_IO, /* app buffered read IOs on compressed file */ 1168 APP_MAPPED_CDATA_READ_IO, /* app mapped read IOs on compressed file */ 1169 FS_DATA_READ_IO, /* data read IOs */ 1170 FS_GDATA_READ_IO, /* data read IOs from background gc */ 1171 FS_CDATA_READ_IO, /* compressed data read IOs */ 1172 FS_NODE_READ_IO, /* node read IOs */ 1173 FS_META_READ_IO, /* meta read IOs */ 1174 1175 /* other */ 1176 FS_DISCARD_IO, /* discard */ 1177 FS_FLUSH_IO, /* flush */ 1178 NR_IO_TYPE, 1179 }; 1180 1181 struct f2fs_io_info { 1182 struct f2fs_sb_info *sbi; /* f2fs_sb_info pointer */ 1183 nid_t ino; /* inode number */ 1184 enum page_type type; /* contains DATA/NODE/META/META_FLUSH */ 1185 enum temp_type temp; /* contains HOT/WARM/COLD */ 1186 enum req_op op; /* contains REQ_OP_ */ 1187 blk_opf_t op_flags; /* req_flag_bits */ 1188 block_t new_blkaddr; /* new block address to be written */ 1189 block_t old_blkaddr; /* old block address before Cow */ 1190 struct page *page; /* page to be written */ 1191 struct page *encrypted_page; /* encrypted page */ 1192 struct page *compressed_page; /* compressed page */ 1193 struct list_head list; /* serialize IOs */ 1194 unsigned int compr_blocks; /* # of compressed block addresses */ 1195 unsigned int need_lock:8; /* indicate we need to lock cp_rwsem */ 1196 unsigned int version:8; /* version of the node */ 1197 unsigned int submitted:1; /* indicate IO submission */ 1198 unsigned int in_list:1; /* indicate fio is in io_list */ 1199 unsigned int is_por:1; /* indicate IO is from recovery or not */ 1200 unsigned int retry:1; /* need to reallocate block address */ 1201 unsigned int encrypted:1; /* indicate file is encrypted */ 1202 unsigned int post_read:1; /* require post read */ 1203 enum iostat_type io_type; /* io type */ 1204 struct writeback_control *io_wbc; /* writeback control */ 1205 struct bio **bio; /* bio for ipu */ 1206 sector_t *last_block; /* last block number in bio */ 1207 }; 1208 1209 struct bio_entry { 1210 struct bio *bio; 1211 struct list_head list; 1212 }; 1213 1214 #define is_read_io(rw) ((rw) == READ) 1215 struct f2fs_bio_info { 1216 struct f2fs_sb_info *sbi; /* f2fs superblock */ 1217 struct bio *bio; /* bios to merge */ 1218 sector_t last_block_in_bio; /* last block number */ 1219 struct f2fs_io_info fio; /* store buffered io info. */ 1220 struct f2fs_rwsem io_rwsem; /* blocking op for bio */ 1221 spinlock_t io_lock; /* serialize DATA/NODE IOs */ 1222 struct list_head io_list; /* track fios */ 1223 struct list_head bio_list; /* bio entry list head */ 1224 struct f2fs_rwsem bio_list_lock; /* lock to protect bio entry list */ 1225 }; 1226 1227 #define FDEV(i) (sbi->devs[i]) 1228 #define RDEV(i) (raw_super->devs[i]) 1229 struct f2fs_dev_info { 1230 struct block_device *bdev; 1231 char path[MAX_PATH_LEN]; 1232 unsigned int total_segments; 1233 block_t start_blk; 1234 block_t end_blk; 1235 #ifdef CONFIG_BLK_DEV_ZONED 1236 unsigned int nr_blkz; /* Total number of zones */ 1237 unsigned long *blkz_seq; /* Bitmap indicating sequential zones */ 1238 #endif 1239 }; 1240 1241 enum inode_type { 1242 DIR_INODE, /* for dirty dir inode */ 1243 FILE_INODE, /* for dirty regular/symlink inode */ 1244 DIRTY_META, /* for all dirtied inode metadata */ 1245 NR_INODE_TYPE, 1246 }; 1247 1248 /* for inner inode cache management */ 1249 struct inode_management { 1250 struct radix_tree_root ino_root; /* ino entry array */ 1251 spinlock_t ino_lock; /* for ino entry lock */ 1252 struct list_head ino_list; /* inode list head */ 1253 unsigned long ino_num; /* number of entries */ 1254 }; 1255 1256 /* for GC_AT */ 1257 struct atgc_management { 1258 bool atgc_enabled; /* ATGC is enabled or not */ 1259 struct rb_root_cached root; /* root of victim rb-tree */ 1260 struct list_head victim_list; /* linked with all victim entries */ 1261 unsigned int victim_count; /* victim count in rb-tree */ 1262 unsigned int candidate_ratio; /* candidate ratio */ 1263 unsigned int max_candidate_count; /* max candidate count */ 1264 unsigned int age_weight; /* age weight, vblock_weight = 100 - age_weight */ 1265 unsigned long long age_threshold; /* age threshold */ 1266 }; 1267 1268 struct f2fs_gc_control { 1269 unsigned int victim_segno; /* target victim segment number */ 1270 int init_gc_type; /* FG_GC or BG_GC */ 1271 bool no_bg_gc; /* check the space and stop bg_gc */ 1272 bool should_migrate_blocks; /* should migrate blocks */ 1273 bool err_gc_skipped; /* return EAGAIN if GC skipped */ 1274 unsigned int nr_free_secs; /* # of free sections to do GC */ 1275 }; 1276 1277 /* 1278 * For s_flag in struct f2fs_sb_info 1279 * Modification on enum should be synchronized with s_flag array 1280 */ 1281 enum { 1282 SBI_IS_DIRTY, /* dirty flag for checkpoint */ 1283 SBI_IS_CLOSE, /* specify unmounting */ 1284 SBI_NEED_FSCK, /* need fsck.f2fs to fix */ 1285 SBI_POR_DOING, /* recovery is doing or not */ 1286 SBI_NEED_SB_WRITE, /* need to recover superblock */ 1287 SBI_NEED_CP, /* need to checkpoint */ 1288 SBI_IS_SHUTDOWN, /* shutdown by ioctl */ 1289 SBI_IS_RECOVERED, /* recovered orphan/data */ 1290 SBI_CP_DISABLED, /* CP was disabled last mount */ 1291 SBI_CP_DISABLED_QUICK, /* CP was disabled quickly */ 1292 SBI_QUOTA_NEED_FLUSH, /* need to flush quota info in CP */ 1293 SBI_QUOTA_SKIP_FLUSH, /* skip flushing quota in current CP */ 1294 SBI_QUOTA_NEED_REPAIR, /* quota file may be corrupted */ 1295 SBI_IS_RESIZEFS, /* resizefs is in process */ 1296 SBI_IS_FREEZING, /* freezefs is in process */ 1297 SBI_IS_WRITABLE, /* remove ro mountoption transiently */ 1298 MAX_SBI_FLAG, 1299 }; 1300 1301 enum { 1302 CP_TIME, 1303 REQ_TIME, 1304 DISCARD_TIME, 1305 GC_TIME, 1306 DISABLE_TIME, 1307 UMOUNT_DISCARD_TIMEOUT, 1308 MAX_TIME, 1309 }; 1310 1311 /* Note that you need to keep synchronization with this gc_mode_names array */ 1312 enum { 1313 GC_NORMAL, 1314 GC_IDLE_CB, 1315 GC_IDLE_GREEDY, 1316 GC_IDLE_AT, 1317 GC_URGENT_HIGH, 1318 GC_URGENT_LOW, 1319 GC_URGENT_MID, 1320 MAX_GC_MODE, 1321 }; 1322 1323 enum { 1324 BGGC_MODE_ON, /* background gc is on */ 1325 BGGC_MODE_OFF, /* background gc is off */ 1326 BGGC_MODE_SYNC, /* 1327 * background gc is on, migrating blocks 1328 * like foreground gc 1329 */ 1330 }; 1331 1332 enum { 1333 FS_MODE_ADAPTIVE, /* use both lfs/ssr allocation */ 1334 FS_MODE_LFS, /* use lfs allocation only */ 1335 FS_MODE_FRAGMENT_SEG, /* segment fragmentation mode */ 1336 FS_MODE_FRAGMENT_BLK, /* block fragmentation mode */ 1337 }; 1338 1339 enum { 1340 ALLOC_MODE_DEFAULT, /* stay default */ 1341 ALLOC_MODE_REUSE, /* reuse segments as much as possible */ 1342 }; 1343 1344 enum fsync_mode { 1345 FSYNC_MODE_POSIX, /* fsync follows posix semantics */ 1346 FSYNC_MODE_STRICT, /* fsync behaves in line with ext4 */ 1347 FSYNC_MODE_NOBARRIER, /* fsync behaves nobarrier based on posix */ 1348 }; 1349 1350 enum { 1351 COMPR_MODE_FS, /* 1352 * automatically compress compression 1353 * enabled files 1354 */ 1355 COMPR_MODE_USER, /* 1356 * automatical compression is disabled. 1357 * user can control the file compression 1358 * using ioctls 1359 */ 1360 }; 1361 1362 enum { 1363 DISCARD_UNIT_BLOCK, /* basic discard unit is block */ 1364 DISCARD_UNIT_SEGMENT, /* basic discard unit is segment */ 1365 DISCARD_UNIT_SECTION, /* basic discard unit is section */ 1366 }; 1367 1368 enum { 1369 MEMORY_MODE_NORMAL, /* memory mode for normal devices */ 1370 MEMORY_MODE_LOW, /* memory mode for low memry devices */ 1371 }; 1372 1373 static inline int f2fs_test_bit(unsigned int nr, char *addr); 1374 static inline void f2fs_set_bit(unsigned int nr, char *addr); 1375 static inline void f2fs_clear_bit(unsigned int nr, char *addr); 1376 1377 /* 1378 * Layout of f2fs page.private: 1379 * 1380 * Layout A: lowest bit should be 1 1381 * | bit0 = 1 | bit1 | bit2 | ... | bit MAX | private data .... | 1382 * bit 0 PAGE_PRIVATE_NOT_POINTER 1383 * bit 1 PAGE_PRIVATE_DUMMY_WRITE 1384 * bit 2 PAGE_PRIVATE_ONGOING_MIGRATION 1385 * bit 3 PAGE_PRIVATE_INLINE_INODE 1386 * bit 4 PAGE_PRIVATE_REF_RESOURCE 1387 * bit 5- f2fs private data 1388 * 1389 * Layout B: lowest bit should be 0 1390 * page.private is a wrapped pointer. 1391 */ 1392 enum { 1393 PAGE_PRIVATE_NOT_POINTER, /* private contains non-pointer data */ 1394 PAGE_PRIVATE_DUMMY_WRITE, /* data page for padding aligned IO */ 1395 PAGE_PRIVATE_ONGOING_MIGRATION, /* data page which is on-going migrating */ 1396 PAGE_PRIVATE_INLINE_INODE, /* inode page contains inline data */ 1397 PAGE_PRIVATE_REF_RESOURCE, /* dirty page has referenced resources */ 1398 PAGE_PRIVATE_MAX 1399 }; 1400 1401 /* For compression */ 1402 enum compress_algorithm_type { 1403 COMPRESS_LZO, 1404 COMPRESS_LZ4, 1405 COMPRESS_ZSTD, 1406 COMPRESS_LZORLE, 1407 COMPRESS_MAX, 1408 }; 1409 1410 enum compress_flag { 1411 COMPRESS_CHKSUM, 1412 COMPRESS_MAX_FLAG, 1413 }; 1414 1415 #define COMPRESS_WATERMARK 20 1416 #define COMPRESS_PERCENT 20 1417 1418 #define COMPRESS_DATA_RESERVED_SIZE 4 1419 struct compress_data { 1420 __le32 clen; /* compressed data size */ 1421 __le32 chksum; /* compressed data chksum */ 1422 __le32 reserved[COMPRESS_DATA_RESERVED_SIZE]; /* reserved */ 1423 u8 cdata[]; /* compressed data */ 1424 }; 1425 1426 #define COMPRESS_HEADER_SIZE (sizeof(struct compress_data)) 1427 1428 #define F2FS_COMPRESSED_PAGE_MAGIC 0xF5F2C000 1429 1430 #define COMPRESS_LEVEL_OFFSET 8 1431 1432 /* compress context */ 1433 struct compress_ctx { 1434 struct inode *inode; /* inode the context belong to */ 1435 pgoff_t cluster_idx; /* cluster index number */ 1436 unsigned int cluster_size; /* page count in cluster */ 1437 unsigned int log_cluster_size; /* log of cluster size */ 1438 struct page **rpages; /* pages store raw data in cluster */ 1439 unsigned int nr_rpages; /* total page number in rpages */ 1440 struct page **cpages; /* pages store compressed data in cluster */ 1441 unsigned int nr_cpages; /* total page number in cpages */ 1442 unsigned int valid_nr_cpages; /* valid page number in cpages */ 1443 void *rbuf; /* virtual mapped address on rpages */ 1444 struct compress_data *cbuf; /* virtual mapped address on cpages */ 1445 size_t rlen; /* valid data length in rbuf */ 1446 size_t clen; /* valid data length in cbuf */ 1447 void *private; /* payload buffer for specified compression algorithm */ 1448 void *private2; /* extra payload buffer */ 1449 }; 1450 1451 /* compress context for write IO path */ 1452 struct compress_io_ctx { 1453 u32 magic; /* magic number to indicate page is compressed */ 1454 struct inode *inode; /* inode the context belong to */ 1455 struct page **rpages; /* pages store raw data in cluster */ 1456 unsigned int nr_rpages; /* total page number in rpages */ 1457 atomic_t pending_pages; /* in-flight compressed page count */ 1458 }; 1459 1460 /* Context for decompressing one cluster on the read IO path */ 1461 struct decompress_io_ctx { 1462 u32 magic; /* magic number to indicate page is compressed */ 1463 struct inode *inode; /* inode the context belong to */ 1464 pgoff_t cluster_idx; /* cluster index number */ 1465 unsigned int cluster_size; /* page count in cluster */ 1466 unsigned int log_cluster_size; /* log of cluster size */ 1467 struct page **rpages; /* pages store raw data in cluster */ 1468 unsigned int nr_rpages; /* total page number in rpages */ 1469 struct page **cpages; /* pages store compressed data in cluster */ 1470 unsigned int nr_cpages; /* total page number in cpages */ 1471 struct page **tpages; /* temp pages to pad holes in cluster */ 1472 void *rbuf; /* virtual mapped address on rpages */ 1473 struct compress_data *cbuf; /* virtual mapped address on cpages */ 1474 size_t rlen; /* valid data length in rbuf */ 1475 size_t clen; /* valid data length in cbuf */ 1476 1477 /* 1478 * The number of compressed pages remaining to be read in this cluster. 1479 * This is initially nr_cpages. It is decremented by 1 each time a page 1480 * has been read (or failed to be read). When it reaches 0, the cluster 1481 * is decompressed (or an error is reported). 1482 * 1483 * If an error occurs before all the pages have been submitted for I/O, 1484 * then this will never reach 0. In this case the I/O submitter is 1485 * responsible for calling f2fs_decompress_end_io() instead. 1486 */ 1487 atomic_t remaining_pages; 1488 1489 /* 1490 * Number of references to this decompress_io_ctx. 1491 * 1492 * One reference is held for I/O completion. This reference is dropped 1493 * after the pagecache pages are updated and unlocked -- either after 1494 * decompression (and verity if enabled), or after an error. 1495 * 1496 * In addition, each compressed page holds a reference while it is in a 1497 * bio. These references are necessary prevent compressed pages from 1498 * being freed while they are still in a bio. 1499 */ 1500 refcount_t refcnt; 1501 1502 bool failed; /* IO error occurred before decompression? */ 1503 bool need_verity; /* need fs-verity verification after decompression? */ 1504 void *private; /* payload buffer for specified decompression algorithm */ 1505 void *private2; /* extra payload buffer */ 1506 struct work_struct verity_work; /* work to verify the decompressed pages */ 1507 struct work_struct free_work; /* work for late free this structure itself */ 1508 }; 1509 1510 #define NULL_CLUSTER ((unsigned int)(~0)) 1511 #define MIN_COMPRESS_LOG_SIZE 2 1512 #define MAX_COMPRESS_LOG_SIZE 8 1513 #define MAX_COMPRESS_WINDOW_SIZE(log_size) ((PAGE_SIZE) << (log_size)) 1514 1515 struct f2fs_sb_info { 1516 struct super_block *sb; /* pointer to VFS super block */ 1517 struct proc_dir_entry *s_proc; /* proc entry */ 1518 struct f2fs_super_block *raw_super; /* raw super block pointer */ 1519 struct f2fs_rwsem sb_lock; /* lock for raw super block */ 1520 int valid_super_block; /* valid super block no */ 1521 unsigned long s_flag; /* flags for sbi */ 1522 struct mutex writepages; /* mutex for writepages() */ 1523 1524 #ifdef CONFIG_BLK_DEV_ZONED 1525 unsigned int blocks_per_blkz; /* F2FS blocks per zone */ 1526 unsigned int log_blocks_per_blkz; /* log2 F2FS blocks per zone */ 1527 #endif 1528 1529 /* for node-related operations */ 1530 struct f2fs_nm_info *nm_info; /* node manager */ 1531 struct inode *node_inode; /* cache node blocks */ 1532 1533 /* for segment-related operations */ 1534 struct f2fs_sm_info *sm_info; /* segment manager */ 1535 1536 /* for bio operations */ 1537 struct f2fs_bio_info *write_io[NR_PAGE_TYPE]; /* for write bios */ 1538 /* keep migration IO order for LFS mode */ 1539 struct f2fs_rwsem io_order_lock; 1540 mempool_t *write_io_dummy; /* Dummy pages */ 1541 pgoff_t page_eio_ofs[NR_PAGE_TYPE]; /* EIO page offset */ 1542 int page_eio_cnt[NR_PAGE_TYPE]; /* EIO count */ 1543 1544 /* for checkpoint */ 1545 struct f2fs_checkpoint *ckpt; /* raw checkpoint pointer */ 1546 int cur_cp_pack; /* remain current cp pack */ 1547 spinlock_t cp_lock; /* for flag in ckpt */ 1548 struct inode *meta_inode; /* cache meta blocks */ 1549 struct f2fs_rwsem cp_global_sem; /* checkpoint procedure lock */ 1550 struct f2fs_rwsem cp_rwsem; /* blocking FS operations */ 1551 struct f2fs_rwsem node_write; /* locking node writes */ 1552 struct f2fs_rwsem node_change; /* locking node change */ 1553 wait_queue_head_t cp_wait; 1554 unsigned long last_time[MAX_TIME]; /* to store time in jiffies */ 1555 long interval_time[MAX_TIME]; /* to store thresholds */ 1556 struct ckpt_req_control cprc_info; /* for checkpoint request control */ 1557 1558 struct inode_management im[MAX_INO_ENTRY]; /* manage inode cache */ 1559 1560 spinlock_t fsync_node_lock; /* for node entry lock */ 1561 struct list_head fsync_node_list; /* node list head */ 1562 unsigned int fsync_seg_id; /* sequence id */ 1563 unsigned int fsync_node_num; /* number of node entries */ 1564 1565 /* for orphan inode, use 0'th array */ 1566 unsigned int max_orphans; /* max orphan inodes */ 1567 1568 /* for inode management */ 1569 struct list_head inode_list[NR_INODE_TYPE]; /* dirty inode list */ 1570 spinlock_t inode_lock[NR_INODE_TYPE]; /* for dirty inode list lock */ 1571 struct mutex flush_lock; /* for flush exclusion */ 1572 1573 /* for extent tree cache */ 1574 struct extent_tree_info extent_tree[NR_EXTENT_CACHES]; 1575 atomic64_t allocated_data_blocks; /* for block age extent_cache */ 1576 1577 /* The threshold used for hot and warm data seperation*/ 1578 unsigned int hot_data_age_threshold; 1579 unsigned int warm_data_age_threshold; 1580 unsigned int last_age_weight; 1581 1582 /* basic filesystem units */ 1583 unsigned int log_sectors_per_block; /* log2 sectors per block */ 1584 unsigned int log_blocksize; /* log2 block size */ 1585 unsigned int blocksize; /* block size */ 1586 unsigned int root_ino_num; /* root inode number*/ 1587 unsigned int node_ino_num; /* node inode number*/ 1588 unsigned int meta_ino_num; /* meta inode number*/ 1589 unsigned int log_blocks_per_seg; /* log2 blocks per segment */ 1590 unsigned int blocks_per_seg; /* blocks per segment */ 1591 unsigned int unusable_blocks_per_sec; /* unusable blocks per section */ 1592 unsigned int segs_per_sec; /* segments per section */ 1593 unsigned int secs_per_zone; /* sections per zone */ 1594 unsigned int total_sections; /* total section count */ 1595 unsigned int total_node_count; /* total node block count */ 1596 unsigned int total_valid_node_count; /* valid node block count */ 1597 int dir_level; /* directory level */ 1598 bool readdir_ra; /* readahead inode in readdir */ 1599 u64 max_io_bytes; /* max io bytes to merge IOs */ 1600 1601 block_t user_block_count; /* # of user blocks */ 1602 block_t total_valid_block_count; /* # of valid blocks */ 1603 block_t discard_blks; /* discard command candidats */ 1604 block_t last_valid_block_count; /* for recovery */ 1605 block_t reserved_blocks; /* configurable reserved blocks */ 1606 block_t current_reserved_blocks; /* current reserved blocks */ 1607 1608 /* Additional tracking for no checkpoint mode */ 1609 block_t unusable_block_count; /* # of blocks saved by last cp */ 1610 1611 unsigned int nquota_files; /* # of quota sysfile */ 1612 struct f2fs_rwsem quota_sem; /* blocking cp for flags */ 1613 1614 /* # of pages, see count_type */ 1615 atomic_t nr_pages[NR_COUNT_TYPE]; 1616 /* # of allocated blocks */ 1617 struct percpu_counter alloc_valid_block_count; 1618 /* # of node block writes as roll forward recovery */ 1619 struct percpu_counter rf_node_block_count; 1620 1621 /* writeback control */ 1622 atomic_t wb_sync_req[META]; /* count # of WB_SYNC threads */ 1623 1624 /* valid inode count */ 1625 struct percpu_counter total_valid_inode_count; 1626 1627 struct f2fs_mount_info mount_opt; /* mount options */ 1628 1629 /* for cleaning operations */ 1630 struct f2fs_rwsem gc_lock; /* 1631 * semaphore for GC, avoid 1632 * race between GC and GC or CP 1633 */ 1634 struct f2fs_gc_kthread *gc_thread; /* GC thread */ 1635 struct atgc_management am; /* atgc management */ 1636 unsigned int cur_victim_sec; /* current victim section num */ 1637 unsigned int gc_mode; /* current GC state */ 1638 unsigned int next_victim_seg[2]; /* next segment in victim section */ 1639 spinlock_t gc_remaining_trials_lock; 1640 /* remaining trial count for GC_URGENT_* and GC_IDLE_* */ 1641 unsigned int gc_remaining_trials; 1642 1643 /* for skip statistic */ 1644 unsigned long long skipped_gc_rwsem; /* FG_GC only */ 1645 1646 /* threshold for gc trials on pinned files */ 1647 u64 gc_pin_file_threshold; 1648 struct f2fs_rwsem pin_sem; 1649 1650 /* maximum # of trials to find a victim segment for SSR and GC */ 1651 unsigned int max_victim_search; 1652 /* migration granularity of garbage collection, unit: segment */ 1653 unsigned int migration_granularity; 1654 1655 /* 1656 * for stat information. 1657 * one is for the LFS mode, and the other is for the SSR mode. 1658 */ 1659 #ifdef CONFIG_F2FS_STAT_FS 1660 struct f2fs_stat_info *stat_info; /* FS status information */ 1661 atomic_t meta_count[META_MAX]; /* # of meta blocks */ 1662 unsigned int segment_count[2]; /* # of allocated segments */ 1663 unsigned int block_count[2]; /* # of allocated blocks */ 1664 atomic_t inplace_count; /* # of inplace update */ 1665 /* # of lookup extent cache */ 1666 atomic64_t total_hit_ext[NR_EXTENT_CACHES]; 1667 /* # of hit rbtree extent node */ 1668 atomic64_t read_hit_rbtree[NR_EXTENT_CACHES]; 1669 /* # of hit cached extent node */ 1670 atomic64_t read_hit_cached[NR_EXTENT_CACHES]; 1671 /* # of hit largest extent node in read extent cache */ 1672 atomic64_t read_hit_largest; 1673 atomic_t inline_xattr; /* # of inline_xattr inodes */ 1674 atomic_t inline_inode; /* # of inline_data inodes */ 1675 atomic_t inline_dir; /* # of inline_dentry inodes */ 1676 atomic_t compr_inode; /* # of compressed inodes */ 1677 atomic64_t compr_blocks; /* # of compressed blocks */ 1678 atomic_t swapfile_inode; /* # of swapfile inodes */ 1679 atomic_t atomic_files; /* # of opened atomic file */ 1680 atomic_t max_aw_cnt; /* max # of atomic writes */ 1681 unsigned int io_skip_bggc; /* skip background gc for in-flight IO */ 1682 unsigned int other_skip_bggc; /* skip background gc for other reasons */ 1683 unsigned int ndirty_inode[NR_INODE_TYPE]; /* # of dirty inodes */ 1684 #endif 1685 spinlock_t stat_lock; /* lock for stat operations */ 1686 1687 /* to attach REQ_META|REQ_FUA flags */ 1688 unsigned int data_io_flag; 1689 unsigned int node_io_flag; 1690 1691 /* For sysfs support */ 1692 struct kobject s_kobj; /* /sys/fs/f2fs/<devname> */ 1693 struct completion s_kobj_unregister; 1694 1695 struct kobject s_stat_kobj; /* /sys/fs/f2fs/<devname>/stat */ 1696 struct completion s_stat_kobj_unregister; 1697 1698 struct kobject s_feature_list_kobj; /* /sys/fs/f2fs/<devname>/feature_list */ 1699 struct completion s_feature_list_kobj_unregister; 1700 1701 /* For shrinker support */ 1702 struct list_head s_list; 1703 struct mutex umount_mutex; 1704 unsigned int shrinker_run_no; 1705 1706 /* For multi devices */ 1707 int s_ndevs; /* number of devices */ 1708 struct f2fs_dev_info *devs; /* for device list */ 1709 unsigned int dirty_device; /* for checkpoint data flush */ 1710 spinlock_t dev_lock; /* protect dirty_device */ 1711 bool aligned_blksize; /* all devices has the same logical blksize */ 1712 1713 /* For write statistics */ 1714 u64 sectors_written_start; 1715 u64 kbytes_written; 1716 1717 /* Reference to checksum algorithm driver via cryptoapi */ 1718 struct crypto_shash *s_chksum_driver; 1719 1720 /* Precomputed FS UUID checksum for seeding other checksums */ 1721 __u32 s_chksum_seed; 1722 1723 struct workqueue_struct *post_read_wq; /* post read workqueue */ 1724 1725 unsigned char errors[MAX_F2FS_ERRORS]; /* error flags */ 1726 spinlock_t error_lock; /* protect errors array */ 1727 bool error_dirty; /* errors of sb is dirty */ 1728 1729 struct kmem_cache *inline_xattr_slab; /* inline xattr entry */ 1730 unsigned int inline_xattr_slab_size; /* default inline xattr slab size */ 1731 1732 /* For reclaimed segs statistics per each GC mode */ 1733 unsigned int gc_segment_mode; /* GC state for reclaimed segments */ 1734 unsigned int gc_reclaimed_segs[MAX_GC_MODE]; /* Reclaimed segs for each mode */ 1735 1736 unsigned long seq_file_ra_mul; /* multiplier for ra_pages of seq. files in fadvise */ 1737 1738 int max_fragment_chunk; /* max chunk size for block fragmentation mode */ 1739 int max_fragment_hole; /* max hole size for block fragmentation mode */ 1740 1741 /* For atomic write statistics */ 1742 atomic64_t current_atomic_write; 1743 s64 peak_atomic_write; 1744 u64 committed_atomic_block; 1745 u64 revoked_atomic_block; 1746 1747 #ifdef CONFIG_F2FS_FS_COMPRESSION 1748 struct kmem_cache *page_array_slab; /* page array entry */ 1749 unsigned int page_array_slab_size; /* default page array slab size */ 1750 1751 /* For runtime compression statistics */ 1752 u64 compr_written_block; 1753 u64 compr_saved_block; 1754 u32 compr_new_inode; 1755 1756 /* For compressed block cache */ 1757 struct inode *compress_inode; /* cache compressed blocks */ 1758 unsigned int compress_percent; /* cache page percentage */ 1759 unsigned int compress_watermark; /* cache page watermark */ 1760 atomic_t compress_page_hit; /* cache hit count */ 1761 #endif 1762 1763 #ifdef CONFIG_F2FS_IOSTAT 1764 /* For app/fs IO statistics */ 1765 spinlock_t iostat_lock; 1766 unsigned long long iostat_count[NR_IO_TYPE]; 1767 unsigned long long iostat_bytes[NR_IO_TYPE]; 1768 unsigned long long prev_iostat_bytes[NR_IO_TYPE]; 1769 bool iostat_enable; 1770 unsigned long iostat_next_period; 1771 unsigned int iostat_period_ms; 1772 1773 /* For io latency related statistics info in one iostat period */ 1774 spinlock_t iostat_lat_lock; 1775 struct iostat_lat_info *iostat_io_lat; 1776 #endif 1777 }; 1778 1779 #ifdef CONFIG_F2FS_FAULT_INJECTION 1780 #define time_to_inject(sbi, type) __time_to_inject(sbi, type, __func__, \ 1781 __builtin_return_address(0)) 1782 static inline bool __time_to_inject(struct f2fs_sb_info *sbi, int type, 1783 const char *func, const char *parent_func) 1784 { 1785 struct f2fs_fault_info *ffi = &F2FS_OPTION(sbi).fault_info; 1786 1787 if (!ffi->inject_rate) 1788 return false; 1789 1790 if (!IS_FAULT_SET(ffi, type)) 1791 return false; 1792 1793 atomic_inc(&ffi->inject_ops); 1794 if (atomic_read(&ffi->inject_ops) >= ffi->inject_rate) { 1795 atomic_set(&ffi->inject_ops, 0); 1796 printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", 1797 KERN_INFO, sbi->sb->s_id, f2fs_fault_name[type], 1798 func, parent_func); 1799 return true; 1800 } 1801 return false; 1802 } 1803 #else 1804 static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type) 1805 { 1806 return false; 1807 } 1808 #endif 1809 1810 /* 1811 * Test if the mounted volume is a multi-device volume. 1812 * - For a single regular disk volume, sbi->s_ndevs is 0. 1813 * - For a single zoned disk volume, sbi->s_ndevs is 1. 1814 * - For a multi-device volume, sbi->s_ndevs is always 2 or more. 1815 */ 1816 static inline bool f2fs_is_multi_device(struct f2fs_sb_info *sbi) 1817 { 1818 return sbi->s_ndevs > 1; 1819 } 1820 1821 static inline void f2fs_update_time(struct f2fs_sb_info *sbi, int type) 1822 { 1823 unsigned long now = jiffies; 1824 1825 sbi->last_time[type] = now; 1826 1827 /* DISCARD_TIME and GC_TIME are based on REQ_TIME */ 1828 if (type == REQ_TIME) { 1829 sbi->last_time[DISCARD_TIME] = now; 1830 sbi->last_time[GC_TIME] = now; 1831 } 1832 } 1833 1834 static inline bool f2fs_time_over(struct f2fs_sb_info *sbi, int type) 1835 { 1836 unsigned long interval = sbi->interval_time[type] * HZ; 1837 1838 return time_after(jiffies, sbi->last_time[type] + interval); 1839 } 1840 1841 static inline unsigned int f2fs_time_to_wait(struct f2fs_sb_info *sbi, 1842 int type) 1843 { 1844 unsigned long interval = sbi->interval_time[type] * HZ; 1845 unsigned int wait_ms = 0; 1846 long delta; 1847 1848 delta = (sbi->last_time[type] + interval) - jiffies; 1849 if (delta > 0) 1850 wait_ms = jiffies_to_msecs(delta); 1851 1852 return wait_ms; 1853 } 1854 1855 /* 1856 * Inline functions 1857 */ 1858 static inline u32 __f2fs_crc32(struct f2fs_sb_info *sbi, u32 crc, 1859 const void *address, unsigned int length) 1860 { 1861 struct { 1862 struct shash_desc shash; 1863 char ctx[4]; 1864 } desc; 1865 int err; 1866 1867 BUG_ON(crypto_shash_descsize(sbi->s_chksum_driver) != sizeof(desc.ctx)); 1868 1869 desc.shash.tfm = sbi->s_chksum_driver; 1870 *(u32 *)desc.ctx = crc; 1871 1872 err = crypto_shash_update(&desc.shash, address, length); 1873 BUG_ON(err); 1874 1875 return *(u32 *)desc.ctx; 1876 } 1877 1878 static inline u32 f2fs_crc32(struct f2fs_sb_info *sbi, const void *address, 1879 unsigned int length) 1880 { 1881 return __f2fs_crc32(sbi, F2FS_SUPER_MAGIC, address, length); 1882 } 1883 1884 static inline bool f2fs_crc_valid(struct f2fs_sb_info *sbi, __u32 blk_crc, 1885 void *buf, size_t buf_size) 1886 { 1887 return f2fs_crc32(sbi, buf, buf_size) == blk_crc; 1888 } 1889 1890 static inline u32 f2fs_chksum(struct f2fs_sb_info *sbi, u32 crc, 1891 const void *address, unsigned int length) 1892 { 1893 return __f2fs_crc32(sbi, crc, address, length); 1894 } 1895 1896 static inline struct f2fs_inode_info *F2FS_I(struct inode *inode) 1897 { 1898 return container_of(inode, struct f2fs_inode_info, vfs_inode); 1899 } 1900 1901 static inline struct f2fs_sb_info *F2FS_SB(struct super_block *sb) 1902 { 1903 return sb->s_fs_info; 1904 } 1905 1906 static inline struct f2fs_sb_info *F2FS_I_SB(struct inode *inode) 1907 { 1908 return F2FS_SB(inode->i_sb); 1909 } 1910 1911 static inline struct f2fs_sb_info *F2FS_M_SB(struct address_space *mapping) 1912 { 1913 return F2FS_I_SB(mapping->host); 1914 } 1915 1916 static inline struct f2fs_sb_info *F2FS_P_SB(struct page *page) 1917 { 1918 return F2FS_M_SB(page_file_mapping(page)); 1919 } 1920 1921 static inline struct f2fs_super_block *F2FS_RAW_SUPER(struct f2fs_sb_info *sbi) 1922 { 1923 return (struct f2fs_super_block *)(sbi->raw_super); 1924 } 1925 1926 static inline struct f2fs_checkpoint *F2FS_CKPT(struct f2fs_sb_info *sbi) 1927 { 1928 return (struct f2fs_checkpoint *)(sbi->ckpt); 1929 } 1930 1931 static inline struct f2fs_node *F2FS_NODE(struct page *page) 1932 { 1933 return (struct f2fs_node *)page_address(page); 1934 } 1935 1936 static inline struct f2fs_inode *F2FS_INODE(struct page *page) 1937 { 1938 return &((struct f2fs_node *)page_address(page))->i; 1939 } 1940 1941 static inline struct f2fs_nm_info *NM_I(struct f2fs_sb_info *sbi) 1942 { 1943 return (struct f2fs_nm_info *)(sbi->nm_info); 1944 } 1945 1946 static inline struct f2fs_sm_info *SM_I(struct f2fs_sb_info *sbi) 1947 { 1948 return (struct f2fs_sm_info *)(sbi->sm_info); 1949 } 1950 1951 static inline struct sit_info *SIT_I(struct f2fs_sb_info *sbi) 1952 { 1953 return (struct sit_info *)(SM_I(sbi)->sit_info); 1954 } 1955 1956 static inline struct free_segmap_info *FREE_I(struct f2fs_sb_info *sbi) 1957 { 1958 return (struct free_segmap_info *)(SM_I(sbi)->free_info); 1959 } 1960 1961 static inline struct dirty_seglist_info *DIRTY_I(struct f2fs_sb_info *sbi) 1962 { 1963 return (struct dirty_seglist_info *)(SM_I(sbi)->dirty_info); 1964 } 1965 1966 static inline struct address_space *META_MAPPING(struct f2fs_sb_info *sbi) 1967 { 1968 return sbi->meta_inode->i_mapping; 1969 } 1970 1971 static inline struct address_space *NODE_MAPPING(struct f2fs_sb_info *sbi) 1972 { 1973 return sbi->node_inode->i_mapping; 1974 } 1975 1976 static inline bool is_sbi_flag_set(struct f2fs_sb_info *sbi, unsigned int type) 1977 { 1978 return test_bit(type, &sbi->s_flag); 1979 } 1980 1981 static inline void set_sbi_flag(struct f2fs_sb_info *sbi, unsigned int type) 1982 { 1983 set_bit(type, &sbi->s_flag); 1984 } 1985 1986 static inline void clear_sbi_flag(struct f2fs_sb_info *sbi, unsigned int type) 1987 { 1988 clear_bit(type, &sbi->s_flag); 1989 } 1990 1991 static inline unsigned long long cur_cp_version(struct f2fs_checkpoint *cp) 1992 { 1993 return le64_to_cpu(cp->checkpoint_ver); 1994 } 1995 1996 static inline unsigned long f2fs_qf_ino(struct super_block *sb, int type) 1997 { 1998 if (type < F2FS_MAX_QUOTAS) 1999 return le32_to_cpu(F2FS_SB(sb)->raw_super->qf_ino[type]); 2000 return 0; 2001 } 2002 2003 static inline __u64 cur_cp_crc(struct f2fs_checkpoint *cp) 2004 { 2005 size_t crc_offset = le32_to_cpu(cp->checksum_offset); 2006 return le32_to_cpu(*((__le32 *)((unsigned char *)cp + crc_offset))); 2007 } 2008 2009 static inline bool __is_set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f) 2010 { 2011 unsigned int ckpt_flags = le32_to_cpu(cp->ckpt_flags); 2012 2013 return ckpt_flags & f; 2014 } 2015 2016 static inline bool is_set_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f) 2017 { 2018 return __is_set_ckpt_flags(F2FS_CKPT(sbi), f); 2019 } 2020 2021 static inline void __set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f) 2022 { 2023 unsigned int ckpt_flags; 2024 2025 ckpt_flags = le32_to_cpu(cp->ckpt_flags); 2026 ckpt_flags |= f; 2027 cp->ckpt_flags = cpu_to_le32(ckpt_flags); 2028 } 2029 2030 static inline void set_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f) 2031 { 2032 unsigned long flags; 2033 2034 spin_lock_irqsave(&sbi->cp_lock, flags); 2035 __set_ckpt_flags(F2FS_CKPT(sbi), f); 2036 spin_unlock_irqrestore(&sbi->cp_lock, flags); 2037 } 2038 2039 static inline void __clear_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f) 2040 { 2041 unsigned int ckpt_flags; 2042 2043 ckpt_flags = le32_to_cpu(cp->ckpt_flags); 2044 ckpt_flags &= (~f); 2045 cp->ckpt_flags = cpu_to_le32(ckpt_flags); 2046 } 2047 2048 static inline void clear_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f) 2049 { 2050 unsigned long flags; 2051 2052 spin_lock_irqsave(&sbi->cp_lock, flags); 2053 __clear_ckpt_flags(F2FS_CKPT(sbi), f); 2054 spin_unlock_irqrestore(&sbi->cp_lock, flags); 2055 } 2056 2057 #define init_f2fs_rwsem(sem) \ 2058 do { \ 2059 static struct lock_class_key __key; \ 2060 \ 2061 __init_f2fs_rwsem((sem), #sem, &__key); \ 2062 } while (0) 2063 2064 static inline void __init_f2fs_rwsem(struct f2fs_rwsem *sem, 2065 const char *sem_name, struct lock_class_key *key) 2066 { 2067 __init_rwsem(&sem->internal_rwsem, sem_name, key); 2068 #ifdef CONFIG_F2FS_UNFAIR_RWSEM 2069 init_waitqueue_head(&sem->read_waiters); 2070 #endif 2071 } 2072 2073 static inline int f2fs_rwsem_is_locked(struct f2fs_rwsem *sem) 2074 { 2075 return rwsem_is_locked(&sem->internal_rwsem); 2076 } 2077 2078 static inline int f2fs_rwsem_is_contended(struct f2fs_rwsem *sem) 2079 { 2080 return rwsem_is_contended(&sem->internal_rwsem); 2081 } 2082 2083 static inline void f2fs_down_read(struct f2fs_rwsem *sem) 2084 { 2085 #ifdef CONFIG_F2FS_UNFAIR_RWSEM 2086 wait_event(sem->read_waiters, down_read_trylock(&sem->internal_rwsem)); 2087 #else 2088 down_read(&sem->internal_rwsem); 2089 #endif 2090 } 2091 2092 static inline int f2fs_down_read_trylock(struct f2fs_rwsem *sem) 2093 { 2094 return down_read_trylock(&sem->internal_rwsem); 2095 } 2096 2097 #ifdef CONFIG_DEBUG_LOCK_ALLOC 2098 static inline void f2fs_down_read_nested(struct f2fs_rwsem *sem, int subclass) 2099 { 2100 down_read_nested(&sem->internal_rwsem, subclass); 2101 } 2102 #else 2103 #define f2fs_down_read_nested(sem, subclass) f2fs_down_read(sem) 2104 #endif 2105 2106 static inline void f2fs_up_read(struct f2fs_rwsem *sem) 2107 { 2108 up_read(&sem->internal_rwsem); 2109 } 2110 2111 static inline void f2fs_down_write(struct f2fs_rwsem *sem) 2112 { 2113 down_write(&sem->internal_rwsem); 2114 } 2115 2116 static inline int f2fs_down_write_trylock(struct f2fs_rwsem *sem) 2117 { 2118 return down_write_trylock(&sem->internal_rwsem); 2119 } 2120 2121 static inline void f2fs_up_write(struct f2fs_rwsem *sem) 2122 { 2123 up_write(&sem->internal_rwsem); 2124 #ifdef CONFIG_F2FS_UNFAIR_RWSEM 2125 wake_up_all(&sem->read_waiters); 2126 #endif 2127 } 2128 2129 static inline void f2fs_lock_op(struct f2fs_sb_info *sbi) 2130 { 2131 f2fs_down_read(&sbi->cp_rwsem); 2132 } 2133 2134 static inline int f2fs_trylock_op(struct f2fs_sb_info *sbi) 2135 { 2136 if (time_to_inject(sbi, FAULT_LOCK_OP)) 2137 return 0; 2138 return f2fs_down_read_trylock(&sbi->cp_rwsem); 2139 } 2140 2141 static inline void f2fs_unlock_op(struct f2fs_sb_info *sbi) 2142 { 2143 f2fs_up_read(&sbi->cp_rwsem); 2144 } 2145 2146 static inline void f2fs_lock_all(struct f2fs_sb_info *sbi) 2147 { 2148 f2fs_down_write(&sbi->cp_rwsem); 2149 } 2150 2151 static inline void f2fs_unlock_all(struct f2fs_sb_info *sbi) 2152 { 2153 f2fs_up_write(&sbi->cp_rwsem); 2154 } 2155 2156 static inline int __get_cp_reason(struct f2fs_sb_info *sbi) 2157 { 2158 int reason = CP_SYNC; 2159 2160 if (test_opt(sbi, FASTBOOT)) 2161 reason = CP_FASTBOOT; 2162 if (is_sbi_flag_set(sbi, SBI_IS_CLOSE)) 2163 reason = CP_UMOUNT; 2164 return reason; 2165 } 2166 2167 static inline bool __remain_node_summaries(int reason) 2168 { 2169 return (reason & (CP_UMOUNT | CP_FASTBOOT)); 2170 } 2171 2172 static inline bool __exist_node_summaries(struct f2fs_sb_info *sbi) 2173 { 2174 return (is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG) || 2175 is_set_ckpt_flags(sbi, CP_FASTBOOT_FLAG)); 2176 } 2177 2178 /* 2179 * Check whether the inode has blocks or not 2180 */ 2181 static inline int F2FS_HAS_BLOCKS(struct inode *inode) 2182 { 2183 block_t xattr_block = F2FS_I(inode)->i_xattr_nid ? 1 : 0; 2184 2185 return (inode->i_blocks >> F2FS_LOG_SECTORS_PER_BLOCK) > xattr_block; 2186 } 2187 2188 static inline bool f2fs_has_xattr_block(unsigned int ofs) 2189 { 2190 return ofs == XATTR_NODE_OFFSET; 2191 } 2192 2193 static inline bool __allow_reserved_blocks(struct f2fs_sb_info *sbi, 2194 struct inode *inode, bool cap) 2195 { 2196 if (!inode) 2197 return true; 2198 if (!test_opt(sbi, RESERVE_ROOT)) 2199 return false; 2200 if (IS_NOQUOTA(inode)) 2201 return true; 2202 if (uid_eq(F2FS_OPTION(sbi).s_resuid, current_fsuid())) 2203 return true; 2204 if (!gid_eq(F2FS_OPTION(sbi).s_resgid, GLOBAL_ROOT_GID) && 2205 in_group_p(F2FS_OPTION(sbi).s_resgid)) 2206 return true; 2207 if (cap && capable(CAP_SYS_RESOURCE)) 2208 return true; 2209 return false; 2210 } 2211 2212 static inline void f2fs_i_blocks_write(struct inode *, block_t, bool, bool); 2213 static inline int inc_valid_block_count(struct f2fs_sb_info *sbi, 2214 struct inode *inode, blkcnt_t *count) 2215 { 2216 blkcnt_t diff = 0, release = 0; 2217 block_t avail_user_block_count; 2218 int ret; 2219 2220 ret = dquot_reserve_block(inode, *count); 2221 if (ret) 2222 return ret; 2223 2224 if (time_to_inject(sbi, FAULT_BLOCK)) { 2225 release = *count; 2226 goto release_quota; 2227 } 2228 2229 /* 2230 * let's increase this in prior to actual block count change in order 2231 * for f2fs_sync_file to avoid data races when deciding checkpoint. 2232 */ 2233 percpu_counter_add(&sbi->alloc_valid_block_count, (*count)); 2234 2235 spin_lock(&sbi->stat_lock); 2236 sbi->total_valid_block_count += (block_t)(*count); 2237 avail_user_block_count = sbi->user_block_count - 2238 sbi->current_reserved_blocks; 2239 2240 if (!__allow_reserved_blocks(sbi, inode, true)) 2241 avail_user_block_count -= F2FS_OPTION(sbi).root_reserved_blocks; 2242 2243 if (F2FS_IO_ALIGNED(sbi)) 2244 avail_user_block_count -= sbi->blocks_per_seg * 2245 SM_I(sbi)->additional_reserved_segments; 2246 2247 if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) { 2248 if (avail_user_block_count > sbi->unusable_block_count) 2249 avail_user_block_count -= sbi->unusable_block_count; 2250 else 2251 avail_user_block_count = 0; 2252 } 2253 if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) { 2254 diff = sbi->total_valid_block_count - avail_user_block_count; 2255 if (diff > *count) 2256 diff = *count; 2257 *count -= diff; 2258 release = diff; 2259 sbi->total_valid_block_count -= diff; 2260 if (!*count) { 2261 spin_unlock(&sbi->stat_lock); 2262 goto enospc; 2263 } 2264 } 2265 spin_unlock(&sbi->stat_lock); 2266 2267 if (unlikely(release)) { 2268 percpu_counter_sub(&sbi->alloc_valid_block_count, release); 2269 dquot_release_reservation_block(inode, release); 2270 } 2271 f2fs_i_blocks_write(inode, *count, true, true); 2272 return 0; 2273 2274 enospc: 2275 percpu_counter_sub(&sbi->alloc_valid_block_count, release); 2276 release_quota: 2277 dquot_release_reservation_block(inode, release); 2278 return -ENOSPC; 2279 } 2280 2281 __printf(2, 3) 2282 void f2fs_printk(struct f2fs_sb_info *sbi, const char *fmt, ...); 2283 2284 #define f2fs_err(sbi, fmt, ...) \ 2285 f2fs_printk(sbi, KERN_ERR fmt, ##__VA_ARGS__) 2286 #define f2fs_warn(sbi, fmt, ...) \ 2287 f2fs_printk(sbi, KERN_WARNING fmt, ##__VA_ARGS__) 2288 #define f2fs_notice(sbi, fmt, ...) \ 2289 f2fs_printk(sbi, KERN_NOTICE fmt, ##__VA_ARGS__) 2290 #define f2fs_info(sbi, fmt, ...) \ 2291 f2fs_printk(sbi, KERN_INFO fmt, ##__VA_ARGS__) 2292 #define f2fs_debug(sbi, fmt, ...) \ 2293 f2fs_printk(sbi, KERN_DEBUG fmt, ##__VA_ARGS__) 2294 2295 #define PAGE_PRIVATE_GET_FUNC(name, flagname) \ 2296 static inline bool page_private_##name(struct page *page) \ 2297 { \ 2298 return PagePrivate(page) && \ 2299 test_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)) && \ 2300 test_bit(PAGE_PRIVATE_##flagname, &page_private(page)); \ 2301 } 2302 2303 #define PAGE_PRIVATE_SET_FUNC(name, flagname) \ 2304 static inline void set_page_private_##name(struct page *page) \ 2305 { \ 2306 if (!PagePrivate(page)) \ 2307 attach_page_private(page, (void *)0); \ 2308 set_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)); \ 2309 set_bit(PAGE_PRIVATE_##flagname, &page_private(page)); \ 2310 } 2311 2312 #define PAGE_PRIVATE_CLEAR_FUNC(name, flagname) \ 2313 static inline void clear_page_private_##name(struct page *page) \ 2314 { \ 2315 clear_bit(PAGE_PRIVATE_##flagname, &page_private(page)); \ 2316 if (page_private(page) == BIT(PAGE_PRIVATE_NOT_POINTER)) \ 2317 detach_page_private(page); \ 2318 } 2319 2320 PAGE_PRIVATE_GET_FUNC(nonpointer, NOT_POINTER); 2321 PAGE_PRIVATE_GET_FUNC(inline, INLINE_INODE); 2322 PAGE_PRIVATE_GET_FUNC(gcing, ONGOING_MIGRATION); 2323 PAGE_PRIVATE_GET_FUNC(dummy, DUMMY_WRITE); 2324 2325 PAGE_PRIVATE_SET_FUNC(reference, REF_RESOURCE); 2326 PAGE_PRIVATE_SET_FUNC(inline, INLINE_INODE); 2327 PAGE_PRIVATE_SET_FUNC(gcing, ONGOING_MIGRATION); 2328 PAGE_PRIVATE_SET_FUNC(dummy, DUMMY_WRITE); 2329 2330 PAGE_PRIVATE_CLEAR_FUNC(reference, REF_RESOURCE); 2331 PAGE_PRIVATE_CLEAR_FUNC(inline, INLINE_INODE); 2332 PAGE_PRIVATE_CLEAR_FUNC(gcing, ONGOING_MIGRATION); 2333 PAGE_PRIVATE_CLEAR_FUNC(dummy, DUMMY_WRITE); 2334 2335 static inline unsigned long get_page_private_data(struct page *page) 2336 { 2337 unsigned long data = page_private(page); 2338 2339 if (!test_bit(PAGE_PRIVATE_NOT_POINTER, &data)) 2340 return 0; 2341 return data >> PAGE_PRIVATE_MAX; 2342 } 2343 2344 static inline void set_page_private_data(struct page *page, unsigned long data) 2345 { 2346 if (!PagePrivate(page)) 2347 attach_page_private(page, (void *)0); 2348 set_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)); 2349 page_private(page) |= data << PAGE_PRIVATE_MAX; 2350 } 2351 2352 static inline void clear_page_private_data(struct page *page) 2353 { 2354 page_private(page) &= GENMASK(PAGE_PRIVATE_MAX - 1, 0); 2355 if (page_private(page) == BIT(PAGE_PRIVATE_NOT_POINTER)) 2356 detach_page_private(page); 2357 } 2358 2359 static inline void clear_page_private_all(struct page *page) 2360 { 2361 clear_page_private_data(page); 2362 clear_page_private_reference(page); 2363 clear_page_private_gcing(page); 2364 clear_page_private_inline(page); 2365 2366 f2fs_bug_on(F2FS_P_SB(page), page_private(page)); 2367 } 2368 2369 static inline void dec_valid_block_count(struct f2fs_sb_info *sbi, 2370 struct inode *inode, 2371 block_t count) 2372 { 2373 blkcnt_t sectors = count << F2FS_LOG_SECTORS_PER_BLOCK; 2374 2375 spin_lock(&sbi->stat_lock); 2376 f2fs_bug_on(sbi, sbi->total_valid_block_count < (block_t) count); 2377 sbi->total_valid_block_count -= (block_t)count; 2378 if (sbi->reserved_blocks && 2379 sbi->current_reserved_blocks < sbi->reserved_blocks) 2380 sbi->current_reserved_blocks = min(sbi->reserved_blocks, 2381 sbi->current_reserved_blocks + count); 2382 spin_unlock(&sbi->stat_lock); 2383 if (unlikely(inode->i_blocks < sectors)) { 2384 f2fs_warn(sbi, "Inconsistent i_blocks, ino:%lu, iblocks:%llu, sectors:%llu", 2385 inode->i_ino, 2386 (unsigned long long)inode->i_blocks, 2387 (unsigned long long)sectors); 2388 set_sbi_flag(sbi, SBI_NEED_FSCK); 2389 return; 2390 } 2391 f2fs_i_blocks_write(inode, count, false, true); 2392 } 2393 2394 static inline void inc_page_count(struct f2fs_sb_info *sbi, int count_type) 2395 { 2396 atomic_inc(&sbi->nr_pages[count_type]); 2397 2398 if (count_type == F2FS_DIRTY_DENTS || 2399 count_type == F2FS_DIRTY_NODES || 2400 count_type == F2FS_DIRTY_META || 2401 count_type == F2FS_DIRTY_QDATA || 2402 count_type == F2FS_DIRTY_IMETA) 2403 set_sbi_flag(sbi, SBI_IS_DIRTY); 2404 } 2405 2406 static inline void inode_inc_dirty_pages(struct inode *inode) 2407 { 2408 atomic_inc(&F2FS_I(inode)->dirty_pages); 2409 inc_page_count(F2FS_I_SB(inode), S_ISDIR(inode->i_mode) ? 2410 F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA); 2411 if (IS_NOQUOTA(inode)) 2412 inc_page_count(F2FS_I_SB(inode), F2FS_DIRTY_QDATA); 2413 } 2414 2415 static inline void dec_page_count(struct f2fs_sb_info *sbi, int count_type) 2416 { 2417 atomic_dec(&sbi->nr_pages[count_type]); 2418 } 2419 2420 static inline void inode_dec_dirty_pages(struct inode *inode) 2421 { 2422 if (!S_ISDIR(inode->i_mode) && !S_ISREG(inode->i_mode) && 2423 !S_ISLNK(inode->i_mode)) 2424 return; 2425 2426 atomic_dec(&F2FS_I(inode)->dirty_pages); 2427 dec_page_count(F2FS_I_SB(inode), S_ISDIR(inode->i_mode) ? 2428 F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA); 2429 if (IS_NOQUOTA(inode)) 2430 dec_page_count(F2FS_I_SB(inode), F2FS_DIRTY_QDATA); 2431 } 2432 2433 static inline void inc_atomic_write_cnt(struct inode *inode) 2434 { 2435 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); 2436 struct f2fs_inode_info *fi = F2FS_I(inode); 2437 u64 current_write; 2438 2439 fi->atomic_write_cnt++; 2440 atomic64_inc(&sbi->current_atomic_write); 2441 current_write = atomic64_read(&sbi->current_atomic_write); 2442 if (current_write > sbi->peak_atomic_write) 2443 sbi->peak_atomic_write = current_write; 2444 } 2445 2446 static inline void release_atomic_write_cnt(struct inode *inode) 2447 { 2448 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); 2449 struct f2fs_inode_info *fi = F2FS_I(inode); 2450 2451 atomic64_sub(fi->atomic_write_cnt, &sbi->current_atomic_write); 2452 fi->atomic_write_cnt = 0; 2453 } 2454 2455 static inline s64 get_pages(struct f2fs_sb_info *sbi, int count_type) 2456 { 2457 return atomic_read(&sbi->nr_pages[count_type]); 2458 } 2459 2460 static inline int get_dirty_pages(struct inode *inode) 2461 { 2462 return atomic_read(&F2FS_I(inode)->dirty_pages); 2463 } 2464 2465 static inline int get_blocktype_secs(struct f2fs_sb_info *sbi, int block_type) 2466 { 2467 unsigned int pages_per_sec = sbi->segs_per_sec * sbi->blocks_per_seg; 2468 unsigned int segs = (get_pages(sbi, block_type) + pages_per_sec - 1) >> 2469 sbi->log_blocks_per_seg; 2470 2471 return segs / sbi->segs_per_sec; 2472 } 2473 2474 static inline block_t valid_user_blocks(struct f2fs_sb_info *sbi) 2475 { 2476 return sbi->total_valid_block_count; 2477 } 2478 2479 static inline block_t discard_blocks(struct f2fs_sb_info *sbi) 2480 { 2481 return sbi->discard_blks; 2482 } 2483 2484 static inline unsigned long __bitmap_size(struct f2fs_sb_info *sbi, int flag) 2485 { 2486 struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi); 2487 2488 /* return NAT or SIT bitmap */ 2489 if (flag == NAT_BITMAP) 2490 return le32_to_cpu(ckpt->nat_ver_bitmap_bytesize); 2491 else if (flag == SIT_BITMAP) 2492 return le32_to_cpu(ckpt->sit_ver_bitmap_bytesize); 2493 2494 return 0; 2495 } 2496 2497 static inline block_t __cp_payload(struct f2fs_sb_info *sbi) 2498 { 2499 return le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_payload); 2500 } 2501 2502 static inline void *__bitmap_ptr(struct f2fs_sb_info *sbi, int flag) 2503 { 2504 struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi); 2505 void *tmp_ptr = &ckpt->sit_nat_version_bitmap; 2506 int offset; 2507 2508 if (is_set_ckpt_flags(sbi, CP_LARGE_NAT_BITMAP_FLAG)) { 2509 offset = (flag == SIT_BITMAP) ? 2510 le32_to_cpu(ckpt->nat_ver_bitmap_bytesize) : 0; 2511 /* 2512 * if large_nat_bitmap feature is enabled, leave checksum 2513 * protection for all nat/sit bitmaps. 2514 */ 2515 return tmp_ptr + offset + sizeof(__le32); 2516 } 2517 2518 if (__cp_payload(sbi) > 0) { 2519 if (flag == NAT_BITMAP) 2520 return tmp_ptr; 2521 else 2522 return (unsigned char *)ckpt + F2FS_BLKSIZE; 2523 } else { 2524 offset = (flag == NAT_BITMAP) ? 2525 le32_to_cpu(ckpt->sit_ver_bitmap_bytesize) : 0; 2526 return tmp_ptr + offset; 2527 } 2528 } 2529 2530 static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi) 2531 { 2532 block_t start_addr = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_blkaddr); 2533 2534 if (sbi->cur_cp_pack == 2) 2535 start_addr += sbi->blocks_per_seg; 2536 return start_addr; 2537 } 2538 2539 static inline block_t __start_cp_next_addr(struct f2fs_sb_info *sbi) 2540 { 2541 block_t start_addr = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_blkaddr); 2542 2543 if (sbi->cur_cp_pack == 1) 2544 start_addr += sbi->blocks_per_seg; 2545 return start_addr; 2546 } 2547 2548 static inline void __set_cp_next_pack(struct f2fs_sb_info *sbi) 2549 { 2550 sbi->cur_cp_pack = (sbi->cur_cp_pack == 1) ? 2 : 1; 2551 } 2552 2553 static inline block_t __start_sum_addr(struct f2fs_sb_info *sbi) 2554 { 2555 return le32_to_cpu(F2FS_CKPT(sbi)->cp_pack_start_sum); 2556 } 2557 2558 extern void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync); 2559 static inline int inc_valid_node_count(struct f2fs_sb_info *sbi, 2560 struct inode *inode, bool is_inode) 2561 { 2562 block_t valid_block_count; 2563 unsigned int valid_node_count, user_block_count; 2564 int err; 2565 2566 if (is_inode) { 2567 if (inode) { 2568 err = dquot_alloc_inode(inode); 2569 if (err) 2570 return err; 2571 } 2572 } else { 2573 err = dquot_reserve_block(inode, 1); 2574 if (err) 2575 return err; 2576 } 2577 2578 if (time_to_inject(sbi, FAULT_BLOCK)) 2579 goto enospc; 2580 2581 spin_lock(&sbi->stat_lock); 2582 2583 valid_block_count = sbi->total_valid_block_count + 2584 sbi->current_reserved_blocks + 1; 2585 2586 if (!__allow_reserved_blocks(sbi, inode, false)) 2587 valid_block_count += F2FS_OPTION(sbi).root_reserved_blocks; 2588 2589 if (F2FS_IO_ALIGNED(sbi)) 2590 valid_block_count += sbi->blocks_per_seg * 2591 SM_I(sbi)->additional_reserved_segments; 2592 2593 user_block_count = sbi->user_block_count; 2594 if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) 2595 user_block_count -= sbi->unusable_block_count; 2596 2597 if (unlikely(valid_block_count > user_block_count)) { 2598 spin_unlock(&sbi->stat_lock); 2599 goto enospc; 2600 } 2601 2602 valid_node_count = sbi->total_valid_node_count + 1; 2603 if (unlikely(valid_node_count > sbi->total_node_count)) { 2604 spin_unlock(&sbi->stat_lock); 2605 goto enospc; 2606 } 2607 2608 sbi->total_valid_node_count++; 2609 sbi->total_valid_block_count++; 2610 spin_unlock(&sbi->stat_lock); 2611 2612 if (inode) { 2613 if (is_inode) 2614 f2fs_mark_inode_dirty_sync(inode, true); 2615 else 2616 f2fs_i_blocks_write(inode, 1, true, true); 2617 } 2618 2619 percpu_counter_inc(&sbi->alloc_valid_block_count); 2620 return 0; 2621 2622 enospc: 2623 if (is_inode) { 2624 if (inode) 2625 dquot_free_inode(inode); 2626 } else { 2627 dquot_release_reservation_block(inode, 1); 2628 } 2629 return -ENOSPC; 2630 } 2631 2632 static inline void dec_valid_node_count(struct f2fs_sb_info *sbi, 2633 struct inode *inode, bool is_inode) 2634 { 2635 spin_lock(&sbi->stat_lock); 2636 2637 if (unlikely(!sbi->total_valid_block_count || 2638 !sbi->total_valid_node_count)) { 2639 f2fs_warn(sbi, "dec_valid_node_count: inconsistent block counts, total_valid_block:%u, total_valid_node:%u", 2640 sbi->total_valid_block_count, 2641 sbi->total_valid_node_count); 2642 set_sbi_flag(sbi, SBI_NEED_FSCK); 2643 } else { 2644 sbi->total_valid_block_count--; 2645 sbi->total_valid_node_count--; 2646 } 2647 2648 if (sbi->reserved_blocks && 2649 sbi->current_reserved_blocks < sbi->reserved_blocks) 2650 sbi->current_reserved_blocks++; 2651 2652 spin_unlock(&sbi->stat_lock); 2653 2654 if (is_inode) { 2655 dquot_free_inode(inode); 2656 } else { 2657 if (unlikely(inode->i_blocks == 0)) { 2658 f2fs_warn(sbi, "dec_valid_node_count: inconsistent i_blocks, ino:%lu, iblocks:%llu", 2659 inode->i_ino, 2660 (unsigned long long)inode->i_blocks); 2661 set_sbi_flag(sbi, SBI_NEED_FSCK); 2662 return; 2663 } 2664 f2fs_i_blocks_write(inode, 1, false, true); 2665 } 2666 } 2667 2668 static inline unsigned int valid_node_count(struct f2fs_sb_info *sbi) 2669 { 2670 return sbi->total_valid_node_count; 2671 } 2672 2673 static inline void inc_valid_inode_count(struct f2fs_sb_info *sbi) 2674 { 2675 percpu_counter_inc(&sbi->total_valid_inode_count); 2676 } 2677 2678 static inline void dec_valid_inode_count(struct f2fs_sb_info *sbi) 2679 { 2680 percpu_counter_dec(&sbi->total_valid_inode_count); 2681 } 2682 2683 static inline s64 valid_inode_count(struct f2fs_sb_info *sbi) 2684 { 2685 return percpu_counter_sum_positive(&sbi->total_valid_inode_count); 2686 } 2687 2688 static inline struct page *f2fs_grab_cache_page(struct address_space *mapping, 2689 pgoff_t index, bool for_write) 2690 { 2691 struct page *page; 2692 unsigned int flags; 2693 2694 if (IS_ENABLED(CONFIG_F2FS_FAULT_INJECTION)) { 2695 if (!for_write) 2696 page = find_get_page_flags(mapping, index, 2697 FGP_LOCK | FGP_ACCESSED); 2698 else 2699 page = find_lock_page(mapping, index); 2700 if (page) 2701 return page; 2702 2703 if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_ALLOC)) 2704 return NULL; 2705 } 2706 2707 if (!for_write) 2708 return grab_cache_page(mapping, index); 2709 2710 flags = memalloc_nofs_save(); 2711 page = grab_cache_page_write_begin(mapping, index); 2712 memalloc_nofs_restore(flags); 2713 2714 return page; 2715 } 2716 2717 static inline struct page *f2fs_pagecache_get_page( 2718 struct address_space *mapping, pgoff_t index, 2719 int fgp_flags, gfp_t gfp_mask) 2720 { 2721 if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_GET)) 2722 return NULL; 2723 2724 return pagecache_get_page(mapping, index, fgp_flags, gfp_mask); 2725 } 2726 2727 static inline void f2fs_put_page(struct page *page, int unlock) 2728 { 2729 if (!page) 2730 return; 2731 2732 if (unlock) { 2733 f2fs_bug_on(F2FS_P_SB(page), !PageLocked(page)); 2734 unlock_page(page); 2735 } 2736 put_page(page); 2737 } 2738 2739 static inline void f2fs_put_dnode(struct dnode_of_data *dn) 2740 { 2741 if (dn->node_page) 2742 f2fs_put_page(dn->node_page, 1); 2743 if (dn->inode_page && dn->node_page != dn->inode_page) 2744 f2fs_put_page(dn->inode_page, 0); 2745 dn->node_page = NULL; 2746 dn->inode_page = NULL; 2747 } 2748 2749 static inline struct kmem_cache *f2fs_kmem_cache_create(const char *name, 2750 size_t size) 2751 { 2752 return kmem_cache_create(name, size, 0, SLAB_RECLAIM_ACCOUNT, NULL); 2753 } 2754 2755 static inline void *f2fs_kmem_cache_alloc_nofail(struct kmem_cache *cachep, 2756 gfp_t flags) 2757 { 2758 void *entry; 2759 2760 entry = kmem_cache_alloc(cachep, flags); 2761 if (!entry) 2762 entry = kmem_cache_alloc(cachep, flags | __GFP_NOFAIL); 2763 return entry; 2764 } 2765 2766 static inline void *f2fs_kmem_cache_alloc(struct kmem_cache *cachep, 2767 gfp_t flags, bool nofail, struct f2fs_sb_info *sbi) 2768 { 2769 if (nofail) 2770 return f2fs_kmem_cache_alloc_nofail(cachep, flags); 2771 2772 if (time_to_inject(sbi, FAULT_SLAB_ALLOC)) 2773 return NULL; 2774 2775 return kmem_cache_alloc(cachep, flags); 2776 } 2777 2778 static inline bool is_inflight_io(struct f2fs_sb_info *sbi, int type) 2779 { 2780 if (get_pages(sbi, F2FS_RD_DATA) || get_pages(sbi, F2FS_RD_NODE) || 2781 get_pages(sbi, F2FS_RD_META) || get_pages(sbi, F2FS_WB_DATA) || 2782 get_pages(sbi, F2FS_WB_CP_DATA) || 2783 get_pages(sbi, F2FS_DIO_READ) || 2784 get_pages(sbi, F2FS_DIO_WRITE)) 2785 return true; 2786 2787 if (type != DISCARD_TIME && SM_I(sbi) && SM_I(sbi)->dcc_info && 2788 atomic_read(&SM_I(sbi)->dcc_info->queued_discard)) 2789 return true; 2790 2791 if (SM_I(sbi) && SM_I(sbi)->fcc_info && 2792 atomic_read(&SM_I(sbi)->fcc_info->queued_flush)) 2793 return true; 2794 return false; 2795 } 2796 2797 static inline bool is_idle(struct f2fs_sb_info *sbi, int type) 2798 { 2799 if (sbi->gc_mode == GC_URGENT_HIGH) 2800 return true; 2801 2802 if (is_inflight_io(sbi, type)) 2803 return false; 2804 2805 if (sbi->gc_mode == GC_URGENT_MID) 2806 return true; 2807 2808 if (sbi->gc_mode == GC_URGENT_LOW && 2809 (type == DISCARD_TIME || type == GC_TIME)) 2810 return true; 2811 2812 return f2fs_time_over(sbi, type); 2813 } 2814 2815 static inline void f2fs_radix_tree_insert(struct radix_tree_root *root, 2816 unsigned long index, void *item) 2817 { 2818 while (radix_tree_insert(root, index, item)) 2819 cond_resched(); 2820 } 2821 2822 #define RAW_IS_INODE(p) ((p)->footer.nid == (p)->footer.ino) 2823 2824 static inline bool IS_INODE(struct page *page) 2825 { 2826 struct f2fs_node *p = F2FS_NODE(page); 2827 2828 return RAW_IS_INODE(p); 2829 } 2830 2831 static inline int offset_in_addr(struct f2fs_inode *i) 2832 { 2833 return (i->i_inline & F2FS_EXTRA_ATTR) ? 2834 (le16_to_cpu(i->i_extra_isize) / sizeof(__le32)) : 0; 2835 } 2836 2837 static inline __le32 *blkaddr_in_node(struct f2fs_node *node) 2838 { 2839 return RAW_IS_INODE(node) ? node->i.i_addr : node->dn.addr; 2840 } 2841 2842 static inline int f2fs_has_extra_attr(struct inode *inode); 2843 static inline block_t data_blkaddr(struct inode *inode, 2844 struct page *node_page, unsigned int offset) 2845 { 2846 struct f2fs_node *raw_node; 2847 __le32 *addr_array; 2848 int base = 0; 2849 bool is_inode = IS_INODE(node_page); 2850 2851 raw_node = F2FS_NODE(node_page); 2852 2853 if (is_inode) { 2854 if (!inode) 2855 /* from GC path only */ 2856 base = offset_in_addr(&raw_node->i); 2857 else if (f2fs_has_extra_attr(inode)) 2858 base = get_extra_isize(inode); 2859 } 2860 2861 addr_array = blkaddr_in_node(raw_node); 2862 return le32_to_cpu(addr_array[base + offset]); 2863 } 2864 2865 static inline block_t f2fs_data_blkaddr(struct dnode_of_data *dn) 2866 { 2867 return data_blkaddr(dn->inode, dn->node_page, dn->ofs_in_node); 2868 } 2869 2870 static inline int f2fs_test_bit(unsigned int nr, char *addr) 2871 { 2872 int mask; 2873 2874 addr += (nr >> 3); 2875 mask = BIT(7 - (nr & 0x07)); 2876 return mask & *addr; 2877 } 2878 2879 static inline void f2fs_set_bit(unsigned int nr, char *addr) 2880 { 2881 int mask; 2882 2883 addr += (nr >> 3); 2884 mask = BIT(7 - (nr & 0x07)); 2885 *addr |= mask; 2886 } 2887 2888 static inline void f2fs_clear_bit(unsigned int nr, char *addr) 2889 { 2890 int mask; 2891 2892 addr += (nr >> 3); 2893 mask = BIT(7 - (nr & 0x07)); 2894 *addr &= ~mask; 2895 } 2896 2897 static inline int f2fs_test_and_set_bit(unsigned int nr, char *addr) 2898 { 2899 int mask; 2900 int ret; 2901 2902 addr += (nr >> 3); 2903 mask = BIT(7 - (nr & 0x07)); 2904 ret = mask & *addr; 2905 *addr |= mask; 2906 return ret; 2907 } 2908 2909 static inline int f2fs_test_and_clear_bit(unsigned int nr, char *addr) 2910 { 2911 int mask; 2912 int ret; 2913 2914 addr += (nr >> 3); 2915 mask = BIT(7 - (nr & 0x07)); 2916 ret = mask & *addr; 2917 *addr &= ~mask; 2918 return ret; 2919 } 2920 2921 static inline void f2fs_change_bit(unsigned int nr, char *addr) 2922 { 2923 int mask; 2924 2925 addr += (nr >> 3); 2926 mask = BIT(7 - (nr & 0x07)); 2927 *addr ^= mask; 2928 } 2929 2930 /* 2931 * On-disk inode flags (f2fs_inode::i_flags) 2932 */ 2933 #define F2FS_COMPR_FL 0x00000004 /* Compress file */ 2934 #define F2FS_SYNC_FL 0x00000008 /* Synchronous updates */ 2935 #define F2FS_IMMUTABLE_FL 0x00000010 /* Immutable file */ 2936 #define F2FS_APPEND_FL 0x00000020 /* writes to file may only append */ 2937 #define F2FS_NODUMP_FL 0x00000040 /* do not dump file */ 2938 #define F2FS_NOATIME_FL 0x00000080 /* do not update atime */ 2939 #define F2FS_NOCOMP_FL 0x00000400 /* Don't compress */ 2940 #define F2FS_INDEX_FL 0x00001000 /* hash-indexed directory */ 2941 #define F2FS_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */ 2942 #define F2FS_PROJINHERIT_FL 0x20000000 /* Create with parents projid */ 2943 #define F2FS_CASEFOLD_FL 0x40000000 /* Casefolded file */ 2944 2945 /* Flags that should be inherited by new inodes from their parent. */ 2946 #define F2FS_FL_INHERITED (F2FS_SYNC_FL | F2FS_NODUMP_FL | F2FS_NOATIME_FL | \ 2947 F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \ 2948 F2FS_CASEFOLD_FL) 2949 2950 /* Flags that are appropriate for regular files (all but dir-specific ones). */ 2951 #define F2FS_REG_FLMASK (~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \ 2952 F2FS_CASEFOLD_FL)) 2953 2954 /* Flags that are appropriate for non-directories/regular files. */ 2955 #define F2FS_OTHER_FLMASK (F2FS_NODUMP_FL | F2FS_NOATIME_FL) 2956 2957 static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags) 2958 { 2959 if (S_ISDIR(mode)) 2960 return flags; 2961 else if (S_ISREG(mode)) 2962 return flags & F2FS_REG_FLMASK; 2963 else 2964 return flags & F2FS_OTHER_FLMASK; 2965 } 2966 2967 static inline void __mark_inode_dirty_flag(struct inode *inode, 2968 int flag, bool set) 2969 { 2970 switch (flag) { 2971 case FI_INLINE_XATTR: 2972 case FI_INLINE_DATA: 2973 case FI_INLINE_DENTRY: 2974 case FI_NEW_INODE: 2975 if (set) 2976 return; 2977 fallthrough; 2978 case FI_DATA_EXIST: 2979 case FI_INLINE_DOTS: 2980 case FI_PIN_FILE: 2981 case FI_COMPRESS_RELEASED: 2982 f2fs_mark_inode_dirty_sync(inode, true); 2983 } 2984 } 2985 2986 static inline void set_inode_flag(struct inode *inode, int flag) 2987 { 2988 set_bit(flag, F2FS_I(inode)->flags); 2989 __mark_inode_dirty_flag(inode, flag, true); 2990 } 2991 2992 static inline int is_inode_flag_set(struct inode *inode, int flag) 2993 { 2994 return test_bit(flag, F2FS_I(inode)->flags); 2995 } 2996 2997 static inline void clear_inode_flag(struct inode *inode, int flag) 2998 { 2999 clear_bit(flag, F2FS_I(inode)->flags); 3000 __mark_inode_dirty_flag(inode, flag, false); 3001 } 3002 3003 static inline bool f2fs_verity_in_progress(struct inode *inode) 3004 { 3005 return IS_ENABLED(CONFIG_FS_VERITY) && 3006 is_inode_flag_set(inode, FI_VERITY_IN_PROGRESS); 3007 } 3008 3009 static inline void set_acl_inode(struct inode *inode, umode_t mode) 3010 { 3011 F2FS_I(inode)->i_acl_mode = mode; 3012 set_inode_flag(inode, FI_ACL_MODE); 3013 f2fs_mark_inode_dirty_sync(inode, false); 3014 } 3015 3016 static inline void f2fs_i_links_write(struct inode *inode, bool inc) 3017 { 3018 if (inc) 3019 inc_nlink(inode); 3020 else 3021 drop_nlink(inode); 3022 f2fs_mark_inode_dirty_sync(inode, true); 3023 } 3024 3025 static inline void f2fs_i_blocks_write(struct inode *inode, 3026 block_t diff, bool add, bool claim) 3027 { 3028 bool clean = !is_inode_flag_set(inode, FI_DIRTY_INODE); 3029 bool recover = is_inode_flag_set(inode, FI_AUTO_RECOVER); 3030 3031 /* add = 1, claim = 1 should be dquot_reserve_block in pair */ 3032 if (add) { 3033 if (claim) 3034 dquot_claim_block(inode, diff); 3035 else 3036 dquot_alloc_block_nofail(inode, diff); 3037 } else { 3038 dquot_free_block(inode, diff); 3039 } 3040 3041 f2fs_mark_inode_dirty_sync(inode, true); 3042 if (clean || recover) 3043 set_inode_flag(inode, FI_AUTO_RECOVER); 3044 } 3045 3046 static inline bool f2fs_is_atomic_file(struct inode *inode); 3047 3048 static inline void f2fs_i_size_write(struct inode *inode, loff_t i_size) 3049 { 3050 bool clean = !is_inode_flag_set(inode, FI_DIRTY_INODE); 3051 bool recover = is_inode_flag_set(inode, FI_AUTO_RECOVER); 3052 3053 if (i_size_read(inode) == i_size) 3054 return; 3055 3056 i_size_write(inode, i_size); 3057 3058 if (f2fs_is_atomic_file(inode)) 3059 return; 3060 3061 f2fs_mark_inode_dirty_sync(inode, true); 3062 if (clean || recover) 3063 set_inode_flag(inode, FI_AUTO_RECOVER); 3064 } 3065 3066 static inline void f2fs_i_depth_write(struct inode *inode, unsigned int depth) 3067 { 3068 F2FS_I(inode)->i_current_depth = depth; 3069 f2fs_mark_inode_dirty_sync(inode, true); 3070 } 3071 3072 static inline void f2fs_i_gc_failures_write(struct inode *inode, 3073 unsigned int count) 3074 { 3075 F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN] = count; 3076 f2fs_mark_inode_dirty_sync(inode, true); 3077 } 3078 3079 static inline void f2fs_i_xnid_write(struct inode *inode, nid_t xnid) 3080 { 3081 F2FS_I(inode)->i_xattr_nid = xnid; 3082 f2fs_mark_inode_dirty_sync(inode, true); 3083 } 3084 3085 static inline void f2fs_i_pino_write(struct inode *inode, nid_t pino) 3086 { 3087 F2FS_I(inode)->i_pino = pino; 3088 f2fs_mark_inode_dirty_sync(inode, true); 3089 } 3090 3091 static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri) 3092 { 3093 struct f2fs_inode_info *fi = F2FS_I(inode); 3094 3095 if (ri->i_inline & F2FS_INLINE_XATTR) 3096 set_bit(FI_INLINE_XATTR, fi->flags); 3097 if (ri->i_inline & F2FS_INLINE_DATA) 3098 set_bit(FI_INLINE_DATA, fi->flags); 3099 if (ri->i_inline & F2FS_INLINE_DENTRY) 3100 set_bit(FI_INLINE_DENTRY, fi->flags); 3101 if (ri->i_inline & F2FS_DATA_EXIST) 3102 set_bit(FI_DATA_EXIST, fi->flags); 3103 if (ri->i_inline & F2FS_INLINE_DOTS) 3104 set_bit(FI_INLINE_DOTS, fi->flags); 3105 if (ri->i_inline & F2FS_EXTRA_ATTR) 3106 set_bit(FI_EXTRA_ATTR, fi->flags); 3107 if (ri->i_inline & F2FS_PIN_FILE) 3108 set_bit(FI_PIN_FILE, fi->flags); 3109 if (ri->i_inline & F2FS_COMPRESS_RELEASED) 3110 set_bit(FI_COMPRESS_RELEASED, fi->flags); 3111 } 3112 3113 static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri) 3114 { 3115 ri->i_inline = 0; 3116 3117 if (is_inode_flag_set(inode, FI_INLINE_XATTR)) 3118 ri->i_inline |= F2FS_INLINE_XATTR; 3119 if (is_inode_flag_set(inode, FI_INLINE_DATA)) 3120 ri->i_inline |= F2FS_INLINE_DATA; 3121 if (is_inode_flag_set(inode, FI_INLINE_DENTRY)) 3122 ri->i_inline |= F2FS_INLINE_DENTRY; 3123 if (is_inode_flag_set(inode, FI_DATA_EXIST)) 3124 ri->i_inline |= F2FS_DATA_EXIST; 3125 if (is_inode_flag_set(inode, FI_INLINE_DOTS)) 3126 ri->i_inline |= F2FS_INLINE_DOTS; 3127 if (is_inode_flag_set(inode, FI_EXTRA_ATTR)) 3128 ri->i_inline |= F2FS_EXTRA_ATTR; 3129 if (is_inode_flag_set(inode, FI_PIN_FILE)) 3130 ri->i_inline |= F2FS_PIN_FILE; 3131 if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) 3132 ri->i_inline |= F2FS_COMPRESS_RELEASED; 3133 } 3134 3135 static inline int f2fs_has_extra_attr(struct inode *inode) 3136 { 3137 return is_inode_flag_set(inode, FI_EXTRA_ATTR); 3138 } 3139 3140 static inline int f2fs_has_inline_xattr(struct inode *inode) 3141 { 3142 return is_inode_flag_set(inode, FI_INLINE_XATTR); 3143 } 3144 3145 static inline int f2fs_compressed_file(struct inode *inode) 3146 { 3147 return S_ISREG(inode->i_mode) && 3148 is_inode_flag_set(inode, FI_COMPRESSED_FILE); 3149 } 3150 3151 static inline bool f2fs_need_compress_data(struct inode *inode) 3152 { 3153 int compress_mode = F2FS_OPTION(F2FS_I_SB(inode)).compress_mode; 3154 3155 if (!f2fs_compressed_file(inode)) 3156 return false; 3157 3158 if (compress_mode == COMPR_MODE_FS) 3159 return true; 3160 else if (compress_mode == COMPR_MODE_USER && 3161 is_inode_flag_set(inode, FI_ENABLE_COMPRESS)) 3162 return true; 3163 3164 return false; 3165 } 3166 3167 static inline unsigned int addrs_per_inode(struct inode *inode) 3168 { 3169 unsigned int addrs = CUR_ADDRS_PER_INODE(inode) - 3170 get_inline_xattr_addrs(inode); 3171 3172 if (!f2fs_compressed_file(inode)) 3173 return addrs; 3174 return ALIGN_DOWN(addrs, F2FS_I(inode)->i_cluster_size); 3175 } 3176 3177 static inline unsigned int addrs_per_block(struct inode *inode) 3178 { 3179 if (!f2fs_compressed_file(inode)) 3180 return DEF_ADDRS_PER_BLOCK; 3181 return ALIGN_DOWN(DEF_ADDRS_PER_BLOCK, F2FS_I(inode)->i_cluster_size); 3182 } 3183 3184 static inline void *inline_xattr_addr(struct inode *inode, struct page *page) 3185 { 3186 struct f2fs_inode *ri = F2FS_INODE(page); 3187 3188 return (void *)&(ri->i_addr[DEF_ADDRS_PER_INODE - 3189 get_inline_xattr_addrs(inode)]); 3190 } 3191 3192 static inline int inline_xattr_size(struct inode *inode) 3193 { 3194 if (f2fs_has_inline_xattr(inode)) 3195 return get_inline_xattr_addrs(inode) * sizeof(__le32); 3196 return 0; 3197 } 3198 3199 /* 3200 * Notice: check inline_data flag without inode page lock is unsafe. 3201 * It could change at any time by f2fs_convert_inline_page(). 3202 */ 3203 static inline int f2fs_has_inline_data(struct inode *inode) 3204 { 3205 return is_inode_flag_set(inode, FI_INLINE_DATA); 3206 } 3207 3208 static inline int f2fs_exist_data(struct inode *inode) 3209 { 3210 return is_inode_flag_set(inode, FI_DATA_EXIST); 3211 } 3212 3213 static inline int f2fs_has_inline_dots(struct inode *inode) 3214 { 3215 return is_inode_flag_set(inode, FI_INLINE_DOTS); 3216 } 3217 3218 static inline int f2fs_is_mmap_file(struct inode *inode) 3219 { 3220 return is_inode_flag_set(inode, FI_MMAP_FILE); 3221 } 3222 3223 static inline bool f2fs_is_pinned_file(struct inode *inode) 3224 { 3225 return is_inode_flag_set(inode, FI_PIN_FILE); 3226 } 3227 3228 static inline bool f2fs_is_atomic_file(struct inode *inode) 3229 { 3230 return is_inode_flag_set(inode, FI_ATOMIC_FILE); 3231 } 3232 3233 static inline bool f2fs_is_cow_file(struct inode *inode) 3234 { 3235 return is_inode_flag_set(inode, FI_COW_FILE); 3236 } 3237 3238 static inline bool f2fs_is_first_block_written(struct inode *inode) 3239 { 3240 return is_inode_flag_set(inode, FI_FIRST_BLOCK_WRITTEN); 3241 } 3242 3243 static inline bool f2fs_is_drop_cache(struct inode *inode) 3244 { 3245 return is_inode_flag_set(inode, FI_DROP_CACHE); 3246 } 3247 3248 static inline void *inline_data_addr(struct inode *inode, struct page *page) 3249 { 3250 struct f2fs_inode *ri = F2FS_INODE(page); 3251 int extra_size = get_extra_isize(inode); 3252 3253 return (void *)&(ri->i_addr[extra_size + DEF_INLINE_RESERVED_SIZE]); 3254 } 3255 3256 static inline int f2fs_has_inline_dentry(struct inode *inode) 3257 { 3258 return is_inode_flag_set(inode, FI_INLINE_DENTRY); 3259 } 3260 3261 static inline int is_file(struct inode *inode, int type) 3262 { 3263 return F2FS_I(inode)->i_advise & type; 3264 } 3265 3266 static inline void set_file(struct inode *inode, int type) 3267 { 3268 if (is_file(inode, type)) 3269 return; 3270 F2FS_I(inode)->i_advise |= type; 3271 f2fs_mark_inode_dirty_sync(inode, true); 3272 } 3273 3274 static inline void clear_file(struct inode *inode, int type) 3275 { 3276 if (!is_file(inode, type)) 3277 return; 3278 F2FS_I(inode)->i_advise &= ~type; 3279 f2fs_mark_inode_dirty_sync(inode, true); 3280 } 3281 3282 static inline bool f2fs_is_time_consistent(struct inode *inode) 3283 { 3284 if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime)) 3285 return false; 3286 if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime)) 3287 return false; 3288 if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime)) 3289 return false; 3290 return true; 3291 } 3292 3293 static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync) 3294 { 3295 bool ret; 3296 3297 if (dsync) { 3298 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); 3299 3300 spin_lock(&sbi->inode_lock[DIRTY_META]); 3301 ret = list_empty(&F2FS_I(inode)->gdirty_list); 3302 spin_unlock(&sbi->inode_lock[DIRTY_META]); 3303 return ret; 3304 } 3305 if (!is_inode_flag_set(inode, FI_AUTO_RECOVER) || 3306 file_keep_isize(inode) || 3307 i_size_read(inode) & ~PAGE_MASK) 3308 return false; 3309 3310 if (!f2fs_is_time_consistent(inode)) 3311 return false; 3312 3313 spin_lock(&F2FS_I(inode)->i_size_lock); 3314 ret = F2FS_I(inode)->last_disk_size == i_size_read(inode); 3315 spin_unlock(&F2FS_I(inode)->i_size_lock); 3316 3317 return ret; 3318 } 3319 3320 static inline bool f2fs_readonly(struct super_block *sb) 3321 { 3322 return sb_rdonly(sb); 3323 } 3324 3325 static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi) 3326 { 3327 return is_set_ckpt_flags(sbi, CP_ERROR_FLAG); 3328 } 3329 3330 static inline bool is_dot_dotdot(const u8 *name, size_t len) 3331 { 3332 if (len == 1 && name[0] == '.') 3333 return true; 3334 3335 if (len == 2 && name[0] == '.' && name[1] == '.') 3336 return true; 3337 3338 return false; 3339 } 3340 3341 static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi, 3342 size_t size, gfp_t flags) 3343 { 3344 if (time_to_inject(sbi, FAULT_KMALLOC)) 3345 return NULL; 3346 3347 return kmalloc(size, flags); 3348 } 3349 3350 static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi, 3351 size_t size, gfp_t flags) 3352 { 3353 return f2fs_kmalloc(sbi, size, flags | __GFP_ZERO); 3354 } 3355 3356 static inline void *f2fs_kvmalloc(struct f2fs_sb_info *sbi, 3357 size_t size, gfp_t flags) 3358 { 3359 if (time_to_inject(sbi, FAULT_KVMALLOC)) 3360 return NULL; 3361 3362 return kvmalloc(size, flags); 3363 } 3364 3365 static inline void *f2fs_kvzalloc(struct f2fs_sb_info *sbi, 3366 size_t size, gfp_t flags) 3367 { 3368 return f2fs_kvmalloc(sbi, size, flags | __GFP_ZERO); 3369 } 3370 3371 static inline int get_extra_isize(struct inode *inode) 3372 { 3373 return F2FS_I(inode)->i_extra_isize / sizeof(__le32); 3374 } 3375 3376 static inline int get_inline_xattr_addrs(struct inode *inode) 3377 { 3378 return F2FS_I(inode)->i_inline_xattr_size; 3379 } 3380 3381 #define f2fs_get_inode_mode(i) \ 3382 ((is_inode_flag_set(i, FI_ACL_MODE)) ? \ 3383 (F2FS_I(i)->i_acl_mode) : ((i)->i_mode)) 3384 3385 #define F2FS_TOTAL_EXTRA_ATTR_SIZE \ 3386 (offsetof(struct f2fs_inode, i_extra_end) - \ 3387 offsetof(struct f2fs_inode, i_extra_isize)) \ 3388 3389 #define F2FS_OLD_ATTRIBUTE_SIZE (offsetof(struct f2fs_inode, i_addr)) 3390 #define F2FS_FITS_IN_INODE(f2fs_inode, extra_isize, field) \ 3391 ((offsetof(typeof(*(f2fs_inode)), field) + \ 3392 sizeof((f2fs_inode)->field)) \ 3393 <= (F2FS_OLD_ATTRIBUTE_SIZE + (extra_isize))) \ 3394 3395 #define __is_large_section(sbi) ((sbi)->segs_per_sec > 1) 3396 3397 #define __is_meta_io(fio) (PAGE_TYPE_OF_BIO((fio)->type) == META) 3398 3399 bool f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi, 3400 block_t blkaddr, int type); 3401 static inline void verify_blkaddr(struct f2fs_sb_info *sbi, 3402 block_t blkaddr, int type) 3403 { 3404 if (!f2fs_is_valid_blkaddr(sbi, blkaddr, type)) { 3405 f2fs_err(sbi, "invalid blkaddr: %u, type: %d, run fsck to fix.", 3406 blkaddr, type); 3407 f2fs_bug_on(sbi, 1); 3408 } 3409 } 3410 3411 static inline bool __is_valid_data_blkaddr(block_t blkaddr) 3412 { 3413 if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR || 3414 blkaddr == COMPRESS_ADDR) 3415 return false; 3416 return true; 3417 } 3418 3419 /* 3420 * file.c 3421 */ 3422 int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync); 3423 void f2fs_truncate_data_blocks(struct dnode_of_data *dn); 3424 int f2fs_do_truncate_blocks(struct inode *inode, u64 from, bool lock); 3425 int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock); 3426 int f2fs_truncate(struct inode *inode); 3427 int f2fs_getattr(struct mnt_idmap *idmap, const struct path *path, 3428 struct kstat *stat, u32 request_mask, unsigned int flags); 3429 int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry, 3430 struct iattr *attr); 3431 int f2fs_truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end); 3432 void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count); 3433 int f2fs_precache_extents(struct inode *inode); 3434 int f2fs_fileattr_get(struct dentry *dentry, struct fileattr *fa); 3435 int f2fs_fileattr_set(struct mnt_idmap *idmap, 3436 struct dentry *dentry, struct fileattr *fa); 3437 long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); 3438 long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg); 3439 int f2fs_transfer_project_quota(struct inode *inode, kprojid_t kprojid); 3440 int f2fs_pin_file_control(struct inode *inode, bool inc); 3441 3442 /* 3443 * inode.c 3444 */ 3445 void f2fs_set_inode_flags(struct inode *inode); 3446 bool f2fs_inode_chksum_verify(struct f2fs_sb_info *sbi, struct page *page); 3447 void f2fs_inode_chksum_set(struct f2fs_sb_info *sbi, struct page *page); 3448 struct inode *f2fs_iget(struct super_block *sb, unsigned long ino); 3449 struct inode *f2fs_iget_retry(struct super_block *sb, unsigned long ino); 3450 int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink); 3451 void f2fs_update_inode(struct inode *inode, struct page *node_page); 3452 void f2fs_update_inode_page(struct inode *inode); 3453 int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc); 3454 void f2fs_evict_inode(struct inode *inode); 3455 void f2fs_handle_failed_inode(struct inode *inode); 3456 3457 /* 3458 * namei.c 3459 */ 3460 int f2fs_update_extension_list(struct f2fs_sb_info *sbi, const char *name, 3461 bool hot, bool set); 3462 struct dentry *f2fs_get_parent(struct dentry *child); 3463 int f2fs_get_tmpfile(struct mnt_idmap *idmap, struct inode *dir, 3464 struct inode **new_inode); 3465 3466 /* 3467 * dir.c 3468 */ 3469 int f2fs_init_casefolded_name(const struct inode *dir, 3470 struct f2fs_filename *fname); 3471 int f2fs_setup_filename(struct inode *dir, const struct qstr *iname, 3472 int lookup, struct f2fs_filename *fname); 3473 int f2fs_prepare_lookup(struct inode *dir, struct dentry *dentry, 3474 struct f2fs_filename *fname); 3475 void f2fs_free_filename(struct f2fs_filename *fname); 3476 struct f2fs_dir_entry *f2fs_find_target_dentry(const struct f2fs_dentry_ptr *d, 3477 const struct f2fs_filename *fname, int *max_slots); 3478 int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d, 3479 unsigned int start_pos, struct fscrypt_str *fstr); 3480 void f2fs_do_make_empty_dir(struct inode *inode, struct inode *parent, 3481 struct f2fs_dentry_ptr *d); 3482 struct page *f2fs_init_inode_metadata(struct inode *inode, struct inode *dir, 3483 const struct f2fs_filename *fname, struct page *dpage); 3484 void f2fs_update_parent_metadata(struct inode *dir, struct inode *inode, 3485 unsigned int current_depth); 3486 int f2fs_room_for_filename(const void *bitmap, int slots, int max_slots); 3487 void f2fs_drop_nlink(struct inode *dir, struct inode *inode); 3488 struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir, 3489 const struct f2fs_filename *fname, 3490 struct page **res_page); 3491 struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir, 3492 const struct qstr *child, struct page **res_page); 3493 struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p); 3494 ino_t f2fs_inode_by_name(struct inode *dir, const struct qstr *qstr, 3495 struct page **page); 3496 void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de, 3497 struct page *page, struct inode *inode); 3498 bool f2fs_has_enough_room(struct inode *dir, struct page *ipage, 3499 const struct f2fs_filename *fname); 3500 void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d, 3501 const struct fscrypt_str *name, f2fs_hash_t name_hash, 3502 unsigned int bit_pos); 3503 int f2fs_add_regular_entry(struct inode *dir, const struct f2fs_filename *fname, 3504 struct inode *inode, nid_t ino, umode_t mode); 3505 int f2fs_add_dentry(struct inode *dir, const struct f2fs_filename *fname, 3506 struct inode *inode, nid_t ino, umode_t mode); 3507 int f2fs_do_add_link(struct inode *dir, const struct qstr *name, 3508 struct inode *inode, nid_t ino, umode_t mode); 3509 void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, 3510 struct inode *dir, struct inode *inode); 3511 int f2fs_do_tmpfile(struct inode *inode, struct inode *dir); 3512 bool f2fs_empty_dir(struct inode *dir); 3513 3514 static inline int f2fs_add_link(struct dentry *dentry, struct inode *inode) 3515 { 3516 if (fscrypt_is_nokey_name(dentry)) 3517 return -ENOKEY; 3518 return f2fs_do_add_link(d_inode(dentry->d_parent), &dentry->d_name, 3519 inode, inode->i_ino, inode->i_mode); 3520 } 3521 3522 /* 3523 * super.c 3524 */ 3525 int f2fs_inode_dirtied(struct inode *inode, bool sync); 3526 void f2fs_inode_synced(struct inode *inode); 3527 int f2fs_dquot_initialize(struct inode *inode); 3528 int f2fs_enable_quota_files(struct f2fs_sb_info *sbi, bool rdonly); 3529 int f2fs_quota_sync(struct super_block *sb, int type); 3530 loff_t max_file_blocks(struct inode *inode); 3531 void f2fs_quota_off_umount(struct super_block *sb); 3532 void f2fs_handle_stop(struct f2fs_sb_info *sbi, unsigned char reason); 3533 void f2fs_save_errors(struct f2fs_sb_info *sbi, unsigned char flag); 3534 void f2fs_handle_error(struct f2fs_sb_info *sbi, unsigned char error); 3535 int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover); 3536 int f2fs_sync_fs(struct super_block *sb, int sync); 3537 int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi); 3538 3539 /* 3540 * hash.c 3541 */ 3542 void f2fs_hash_filename(const struct inode *dir, struct f2fs_filename *fname); 3543 3544 /* 3545 * node.c 3546 */ 3547 struct node_info; 3548 3549 int f2fs_check_nid_range(struct f2fs_sb_info *sbi, nid_t nid); 3550 bool f2fs_available_free_memory(struct f2fs_sb_info *sbi, int type); 3551 bool f2fs_in_warm_node_list(struct f2fs_sb_info *sbi, struct page *page); 3552 void f2fs_init_fsync_node_info(struct f2fs_sb_info *sbi); 3553 void f2fs_del_fsync_node_entry(struct f2fs_sb_info *sbi, struct page *page); 3554 void f2fs_reset_fsync_node_info(struct f2fs_sb_info *sbi); 3555 int f2fs_need_dentry_mark(struct f2fs_sb_info *sbi, nid_t nid); 3556 bool f2fs_is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid); 3557 bool f2fs_need_inode_block_update(struct f2fs_sb_info *sbi, nid_t ino); 3558 int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid, 3559 struct node_info *ni, bool checkpoint_context); 3560 pgoff_t f2fs_get_next_page_offset(struct dnode_of_data *dn, pgoff_t pgofs); 3561 int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode); 3562 int f2fs_truncate_inode_blocks(struct inode *inode, pgoff_t from); 3563 int f2fs_truncate_xattr_node(struct inode *inode); 3564 int f2fs_wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, 3565 unsigned int seq_id); 3566 bool f2fs_nat_bitmap_enabled(struct f2fs_sb_info *sbi); 3567 int f2fs_remove_inode_page(struct inode *inode); 3568 struct page *f2fs_new_inode_page(struct inode *inode); 3569 struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs); 3570 void f2fs_ra_node_page(struct f2fs_sb_info *sbi, nid_t nid); 3571 struct page *f2fs_get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid); 3572 struct page *f2fs_get_node_page_ra(struct page *parent, int start); 3573 int f2fs_move_node_page(struct page *node_page, int gc_type); 3574 void f2fs_flush_inline_data(struct f2fs_sb_info *sbi); 3575 int f2fs_fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode, 3576 struct writeback_control *wbc, bool atomic, 3577 unsigned int *seq_id); 3578 int f2fs_sync_node_pages(struct f2fs_sb_info *sbi, 3579 struct writeback_control *wbc, 3580 bool do_balance, enum iostat_type io_type); 3581 int f2fs_build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount); 3582 bool f2fs_alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid); 3583 void f2fs_alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid); 3584 void f2fs_alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid); 3585 int f2fs_try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink); 3586 int f2fs_recover_inline_xattr(struct inode *inode, struct page *page); 3587 int f2fs_recover_xattr_data(struct inode *inode, struct page *page); 3588 int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page); 3589 int f2fs_restore_node_summary(struct f2fs_sb_info *sbi, 3590 unsigned int segno, struct f2fs_summary_block *sum); 3591 void f2fs_enable_nat_bits(struct f2fs_sb_info *sbi); 3592 int f2fs_flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc); 3593 int f2fs_build_node_manager(struct f2fs_sb_info *sbi); 3594 void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi); 3595 int __init f2fs_create_node_manager_caches(void); 3596 void f2fs_destroy_node_manager_caches(void); 3597 3598 /* 3599 * segment.c 3600 */ 3601 bool f2fs_need_SSR(struct f2fs_sb_info *sbi); 3602 int f2fs_commit_atomic_write(struct inode *inode); 3603 void f2fs_abort_atomic_write(struct inode *inode, bool clean); 3604 void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need); 3605 void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi, bool from_bg); 3606 int f2fs_issue_flush(struct f2fs_sb_info *sbi, nid_t ino); 3607 int f2fs_create_flush_cmd_control(struct f2fs_sb_info *sbi); 3608 int f2fs_flush_device_cache(struct f2fs_sb_info *sbi); 3609 void f2fs_destroy_flush_cmd_control(struct f2fs_sb_info *sbi, bool free); 3610 void f2fs_invalidate_blocks(struct f2fs_sb_info *sbi, block_t addr); 3611 bool f2fs_is_checkpointed_data(struct f2fs_sb_info *sbi, block_t blkaddr); 3612 int f2fs_start_discard_thread(struct f2fs_sb_info *sbi); 3613 void f2fs_drop_discard_cmd(struct f2fs_sb_info *sbi); 3614 void f2fs_stop_discard_thread(struct f2fs_sb_info *sbi); 3615 bool f2fs_issue_discard_timeout(struct f2fs_sb_info *sbi); 3616 void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi, 3617 struct cp_control *cpc); 3618 void f2fs_dirty_to_prefree(struct f2fs_sb_info *sbi); 3619 block_t f2fs_get_unusable_blocks(struct f2fs_sb_info *sbi); 3620 int f2fs_disable_cp_again(struct f2fs_sb_info *sbi, block_t unusable); 3621 void f2fs_release_discard_addrs(struct f2fs_sb_info *sbi); 3622 int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra); 3623 bool f2fs_segment_has_free_slot(struct f2fs_sb_info *sbi, int segno); 3624 void f2fs_init_inmem_curseg(struct f2fs_sb_info *sbi); 3625 void f2fs_save_inmem_curseg(struct f2fs_sb_info *sbi); 3626 void f2fs_restore_inmem_curseg(struct f2fs_sb_info *sbi); 3627 void f2fs_get_new_segment(struct f2fs_sb_info *sbi, 3628 unsigned int *newseg, bool new_sec, int dir); 3629 void f2fs_allocate_segment_for_resize(struct f2fs_sb_info *sbi, int type, 3630 unsigned int start, unsigned int end); 3631 void f2fs_allocate_new_section(struct f2fs_sb_info *sbi, int type, bool force); 3632 void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi); 3633 int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range); 3634 bool f2fs_exist_trim_candidates(struct f2fs_sb_info *sbi, 3635 struct cp_control *cpc); 3636 struct page *f2fs_get_sum_page(struct f2fs_sb_info *sbi, unsigned int segno); 3637 void f2fs_update_meta_page(struct f2fs_sb_info *sbi, void *src, 3638 block_t blk_addr); 3639 void f2fs_do_write_meta_page(struct f2fs_sb_info *sbi, struct page *page, 3640 enum iostat_type io_type); 3641 void f2fs_do_write_node_page(unsigned int nid, struct f2fs_io_info *fio); 3642 void f2fs_outplace_write_data(struct dnode_of_data *dn, 3643 struct f2fs_io_info *fio); 3644 int f2fs_inplace_write_data(struct f2fs_io_info *fio); 3645 void f2fs_do_replace_block(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, 3646 block_t old_blkaddr, block_t new_blkaddr, 3647 bool recover_curseg, bool recover_newaddr, 3648 bool from_gc); 3649 void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn, 3650 block_t old_addr, block_t new_addr, 3651 unsigned char version, bool recover_curseg, 3652 bool recover_newaddr); 3653 void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page, 3654 block_t old_blkaddr, block_t *new_blkaddr, 3655 struct f2fs_summary *sum, int type, 3656 struct f2fs_io_info *fio); 3657 void f2fs_update_device_state(struct f2fs_sb_info *sbi, nid_t ino, 3658 block_t blkaddr, unsigned int blkcnt); 3659 void f2fs_wait_on_page_writeback(struct page *page, 3660 enum page_type type, bool ordered, bool locked); 3661 void f2fs_wait_on_block_writeback(struct inode *inode, block_t blkaddr); 3662 void f2fs_wait_on_block_writeback_range(struct inode *inode, block_t blkaddr, 3663 block_t len); 3664 void f2fs_write_data_summaries(struct f2fs_sb_info *sbi, block_t start_blk); 3665 void f2fs_write_node_summaries(struct f2fs_sb_info *sbi, block_t start_blk); 3666 int f2fs_lookup_journal_in_cursum(struct f2fs_journal *journal, int type, 3667 unsigned int val, int alloc); 3668 void f2fs_flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc); 3669 int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi); 3670 int f2fs_check_write_pointer(struct f2fs_sb_info *sbi); 3671 int f2fs_build_segment_manager(struct f2fs_sb_info *sbi); 3672 void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi); 3673 int __init f2fs_create_segment_manager_caches(void); 3674 void f2fs_destroy_segment_manager_caches(void); 3675 int f2fs_rw_hint_to_seg_type(enum rw_hint hint); 3676 unsigned int f2fs_usable_segs_in_sec(struct f2fs_sb_info *sbi, 3677 unsigned int segno); 3678 unsigned int f2fs_usable_blks_in_seg(struct f2fs_sb_info *sbi, 3679 unsigned int segno); 3680 3681 #define DEF_FRAGMENT_SIZE 4 3682 #define MIN_FRAGMENT_SIZE 1 3683 #define MAX_FRAGMENT_SIZE 512 3684 3685 static inline bool f2fs_need_rand_seg(struct f2fs_sb_info *sbi) 3686 { 3687 return F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_SEG || 3688 F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_BLK; 3689 } 3690 3691 /* 3692 * checkpoint.c 3693 */ 3694 void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi, bool end_io, 3695 unsigned char reason); 3696 void f2fs_flush_ckpt_thread(struct f2fs_sb_info *sbi); 3697 struct page *f2fs_grab_meta_page(struct f2fs_sb_info *sbi, pgoff_t index); 3698 struct page *f2fs_get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index); 3699 struct page *f2fs_get_meta_page_retry(struct f2fs_sb_info *sbi, pgoff_t index); 3700 struct page *f2fs_get_tmp_page(struct f2fs_sb_info *sbi, pgoff_t index); 3701 bool f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi, 3702 block_t blkaddr, int type); 3703 int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, 3704 int type, bool sync); 3705 void f2fs_ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index, 3706 unsigned int ra_blocks); 3707 long f2fs_sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type, 3708 long nr_to_write, enum iostat_type io_type); 3709 void f2fs_add_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type); 3710 void f2fs_remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type); 3711 void f2fs_release_ino_entry(struct f2fs_sb_info *sbi, bool all); 3712 bool f2fs_exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode); 3713 void f2fs_set_dirty_device(struct f2fs_sb_info *sbi, nid_t ino, 3714 unsigned int devidx, int type); 3715 bool f2fs_is_dirty_device(struct f2fs_sb_info *sbi, nid_t ino, 3716 unsigned int devidx, int type); 3717 int f2fs_acquire_orphan_inode(struct f2fs_sb_info *sbi); 3718 void f2fs_release_orphan_inode(struct f2fs_sb_info *sbi); 3719 void f2fs_add_orphan_inode(struct inode *inode); 3720 void f2fs_remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino); 3721 int f2fs_recover_orphan_inodes(struct f2fs_sb_info *sbi); 3722 int f2fs_get_valid_checkpoint(struct f2fs_sb_info *sbi); 3723 void f2fs_update_dirty_folio(struct inode *inode, struct folio *folio); 3724 void f2fs_remove_dirty_inode(struct inode *inode); 3725 int f2fs_sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type, 3726 bool from_cp); 3727 void f2fs_wait_on_all_pages(struct f2fs_sb_info *sbi, int type); 3728 u64 f2fs_get_sectors_written(struct f2fs_sb_info *sbi); 3729 int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc); 3730 void f2fs_init_ino_entry_info(struct f2fs_sb_info *sbi); 3731 int __init f2fs_create_checkpoint_caches(void); 3732 void f2fs_destroy_checkpoint_caches(void); 3733 int f2fs_issue_checkpoint(struct f2fs_sb_info *sbi); 3734 int f2fs_start_ckpt_thread(struct f2fs_sb_info *sbi); 3735 void f2fs_stop_ckpt_thread(struct f2fs_sb_info *sbi); 3736 void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi); 3737 3738 /* 3739 * data.c 3740 */ 3741 int __init f2fs_init_bioset(void); 3742 void f2fs_destroy_bioset(void); 3743 int f2fs_init_bio_entry_cache(void); 3744 void f2fs_destroy_bio_entry_cache(void); 3745 void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio, 3746 enum page_type type); 3747 int f2fs_init_write_merge_io(struct f2fs_sb_info *sbi); 3748 void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type); 3749 void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi, 3750 struct inode *inode, struct page *page, 3751 nid_t ino, enum page_type type); 3752 void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, 3753 struct bio **bio, struct page *page); 3754 void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi); 3755 int f2fs_submit_page_bio(struct f2fs_io_info *fio); 3756 int f2fs_merge_page_bio(struct f2fs_io_info *fio); 3757 void f2fs_submit_page_write(struct f2fs_io_info *fio); 3758 struct block_device *f2fs_target_device(struct f2fs_sb_info *sbi, 3759 block_t blk_addr, sector_t *sector); 3760 int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr); 3761 void f2fs_set_data_blkaddr(struct dnode_of_data *dn); 3762 void f2fs_update_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr); 3763 int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count); 3764 int f2fs_reserve_new_block(struct dnode_of_data *dn); 3765 int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index); 3766 int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index); 3767 struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, 3768 blk_opf_t op_flags, bool for_write, pgoff_t *next_pgofs); 3769 struct page *f2fs_find_data_page(struct inode *inode, pgoff_t index, 3770 pgoff_t *next_pgofs); 3771 struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index, 3772 bool for_write); 3773 struct page *f2fs_get_new_data_page(struct inode *inode, 3774 struct page *ipage, pgoff_t index, bool new_i_size); 3775 int f2fs_do_write_data_page(struct f2fs_io_info *fio); 3776 int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag); 3777 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, 3778 u64 start, u64 len); 3779 int f2fs_encrypt_one_page(struct f2fs_io_info *fio); 3780 bool f2fs_should_update_inplace(struct inode *inode, struct f2fs_io_info *fio); 3781 bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio); 3782 int f2fs_write_single_data_page(struct page *page, int *submitted, 3783 struct bio **bio, sector_t *last_block, 3784 struct writeback_control *wbc, 3785 enum iostat_type io_type, 3786 int compr_blocks, bool allow_balance); 3787 void f2fs_write_failed(struct inode *inode, loff_t to); 3788 void f2fs_invalidate_folio(struct folio *folio, size_t offset, size_t length); 3789 bool f2fs_release_folio(struct folio *folio, gfp_t wait); 3790 bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len); 3791 void f2fs_clear_page_cache_dirty_tag(struct page *page); 3792 int f2fs_init_post_read_processing(void); 3793 void f2fs_destroy_post_read_processing(void); 3794 int f2fs_init_post_read_wq(struct f2fs_sb_info *sbi); 3795 void f2fs_destroy_post_read_wq(struct f2fs_sb_info *sbi); 3796 extern const struct iomap_ops f2fs_iomap_ops; 3797 3798 /* 3799 * gc.c 3800 */ 3801 int f2fs_start_gc_thread(struct f2fs_sb_info *sbi); 3802 void f2fs_stop_gc_thread(struct f2fs_sb_info *sbi); 3803 block_t f2fs_start_bidx_of_node(unsigned int node_ofs, struct inode *inode); 3804 int f2fs_gc(struct f2fs_sb_info *sbi, struct f2fs_gc_control *gc_control); 3805 void f2fs_build_gc_manager(struct f2fs_sb_info *sbi); 3806 int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count); 3807 int __init f2fs_create_garbage_collection_cache(void); 3808 void f2fs_destroy_garbage_collection_cache(void); 3809 /* victim selection function for cleaning and SSR */ 3810 int f2fs_get_victim(struct f2fs_sb_info *sbi, unsigned int *result, 3811 int gc_type, int type, char alloc_mode, 3812 unsigned long long age); 3813 3814 /* 3815 * recovery.c 3816 */ 3817 int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only); 3818 bool f2fs_space_for_roll_forward(struct f2fs_sb_info *sbi); 3819 int __init f2fs_create_recovery_cache(void); 3820 void f2fs_destroy_recovery_cache(void); 3821 3822 /* 3823 * debug.c 3824 */ 3825 #ifdef CONFIG_F2FS_STAT_FS 3826 struct f2fs_stat_info { 3827 struct list_head stat_list; 3828 struct f2fs_sb_info *sbi; 3829 int all_area_segs, sit_area_segs, nat_area_segs, ssa_area_segs; 3830 int main_area_segs, main_area_sections, main_area_zones; 3831 unsigned long long hit_cached[NR_EXTENT_CACHES]; 3832 unsigned long long hit_rbtree[NR_EXTENT_CACHES]; 3833 unsigned long long total_ext[NR_EXTENT_CACHES]; 3834 unsigned long long hit_total[NR_EXTENT_CACHES]; 3835 int ext_tree[NR_EXTENT_CACHES]; 3836 int zombie_tree[NR_EXTENT_CACHES]; 3837 int ext_node[NR_EXTENT_CACHES]; 3838 /* to count memory footprint */ 3839 unsigned long long ext_mem[NR_EXTENT_CACHES]; 3840 /* for read extent cache */ 3841 unsigned long long hit_largest; 3842 /* for block age extent cache */ 3843 unsigned long long allocated_data_blocks; 3844 int ndirty_node, ndirty_dent, ndirty_meta, ndirty_imeta; 3845 int ndirty_data, ndirty_qdata; 3846 unsigned int ndirty_dirs, ndirty_files, nquota_files, ndirty_all; 3847 int nats, dirty_nats, sits, dirty_sits; 3848 int free_nids, avail_nids, alloc_nids; 3849 int total_count, utilization; 3850 int bg_gc, nr_wb_cp_data, nr_wb_data; 3851 int nr_rd_data, nr_rd_node, nr_rd_meta; 3852 int nr_dio_read, nr_dio_write; 3853 unsigned int io_skip_bggc, other_skip_bggc; 3854 int nr_flushing, nr_flushed, flush_list_empty; 3855 int nr_discarding, nr_discarded; 3856 int nr_discard_cmd; 3857 unsigned int undiscard_blks; 3858 int nr_issued_ckpt, nr_total_ckpt, nr_queued_ckpt; 3859 unsigned int cur_ckpt_time, peak_ckpt_time; 3860 int inline_xattr, inline_inode, inline_dir, append, update, orphans; 3861 int compr_inode, swapfile_inode; 3862 unsigned long long compr_blocks; 3863 int aw_cnt, max_aw_cnt; 3864 unsigned int valid_count, valid_node_count, valid_inode_count, discard_blks; 3865 unsigned int bimodal, avg_vblocks; 3866 int util_free, util_valid, util_invalid; 3867 int rsvd_segs, overp_segs; 3868 int dirty_count, node_pages, meta_pages, compress_pages; 3869 int compress_page_hit; 3870 int prefree_count, call_count, cp_count, bg_cp_count; 3871 int tot_segs, node_segs, data_segs, free_segs, free_secs; 3872 int bg_node_segs, bg_data_segs; 3873 int tot_blks, data_blks, node_blks; 3874 int bg_data_blks, bg_node_blks; 3875 int curseg[NR_CURSEG_TYPE]; 3876 int cursec[NR_CURSEG_TYPE]; 3877 int curzone[NR_CURSEG_TYPE]; 3878 unsigned int dirty_seg[NR_CURSEG_TYPE]; 3879 unsigned int full_seg[NR_CURSEG_TYPE]; 3880 unsigned int valid_blks[NR_CURSEG_TYPE]; 3881 3882 unsigned int meta_count[META_MAX]; 3883 unsigned int segment_count[2]; 3884 unsigned int block_count[2]; 3885 unsigned int inplace_count; 3886 unsigned long long base_mem, cache_mem, page_mem; 3887 }; 3888 3889 static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi) 3890 { 3891 return (struct f2fs_stat_info *)sbi->stat_info; 3892 } 3893 3894 #define stat_inc_cp_count(si) ((si)->cp_count++) 3895 #define stat_inc_bg_cp_count(si) ((si)->bg_cp_count++) 3896 #define stat_inc_call_count(si) ((si)->call_count++) 3897 #define stat_inc_bggc_count(si) ((si)->bg_gc++) 3898 #define stat_io_skip_bggc_count(sbi) ((sbi)->io_skip_bggc++) 3899 #define stat_other_skip_bggc_count(sbi) ((sbi)->other_skip_bggc++) 3900 #define stat_inc_dirty_inode(sbi, type) ((sbi)->ndirty_inode[type]++) 3901 #define stat_dec_dirty_inode(sbi, type) ((sbi)->ndirty_inode[type]--) 3902 #define stat_inc_total_hit(sbi, type) (atomic64_inc(&(sbi)->total_hit_ext[type])) 3903 #define stat_inc_rbtree_node_hit(sbi, type) (atomic64_inc(&(sbi)->read_hit_rbtree[type])) 3904 #define stat_inc_largest_node_hit(sbi) (atomic64_inc(&(sbi)->read_hit_largest)) 3905 #define stat_inc_cached_node_hit(sbi, type) (atomic64_inc(&(sbi)->read_hit_cached[type])) 3906 #define stat_inc_inline_xattr(inode) \ 3907 do { \ 3908 if (f2fs_has_inline_xattr(inode)) \ 3909 (atomic_inc(&F2FS_I_SB(inode)->inline_xattr)); \ 3910 } while (0) 3911 #define stat_dec_inline_xattr(inode) \ 3912 do { \ 3913 if (f2fs_has_inline_xattr(inode)) \ 3914 (atomic_dec(&F2FS_I_SB(inode)->inline_xattr)); \ 3915 } while (0) 3916 #define stat_inc_inline_inode(inode) \ 3917 do { \ 3918 if (f2fs_has_inline_data(inode)) \ 3919 (atomic_inc(&F2FS_I_SB(inode)->inline_inode)); \ 3920 } while (0) 3921 #define stat_dec_inline_inode(inode) \ 3922 do { \ 3923 if (f2fs_has_inline_data(inode)) \ 3924 (atomic_dec(&F2FS_I_SB(inode)->inline_inode)); \ 3925 } while (0) 3926 #define stat_inc_inline_dir(inode) \ 3927 do { \ 3928 if (f2fs_has_inline_dentry(inode)) \ 3929 (atomic_inc(&F2FS_I_SB(inode)->inline_dir)); \ 3930 } while (0) 3931 #define stat_dec_inline_dir(inode) \ 3932 do { \ 3933 if (f2fs_has_inline_dentry(inode)) \ 3934 (atomic_dec(&F2FS_I_SB(inode)->inline_dir)); \ 3935 } while (0) 3936 #define stat_inc_compr_inode(inode) \ 3937 do { \ 3938 if (f2fs_compressed_file(inode)) \ 3939 (atomic_inc(&F2FS_I_SB(inode)->compr_inode)); \ 3940 } while (0) 3941 #define stat_dec_compr_inode(inode) \ 3942 do { \ 3943 if (f2fs_compressed_file(inode)) \ 3944 (atomic_dec(&F2FS_I_SB(inode)->compr_inode)); \ 3945 } while (0) 3946 #define stat_add_compr_blocks(inode, blocks) \ 3947 (atomic64_add(blocks, &F2FS_I_SB(inode)->compr_blocks)) 3948 #define stat_sub_compr_blocks(inode, blocks) \ 3949 (atomic64_sub(blocks, &F2FS_I_SB(inode)->compr_blocks)) 3950 #define stat_inc_swapfile_inode(inode) \ 3951 (atomic_inc(&F2FS_I_SB(inode)->swapfile_inode)) 3952 #define stat_dec_swapfile_inode(inode) \ 3953 (atomic_dec(&F2FS_I_SB(inode)->swapfile_inode)) 3954 #define stat_inc_atomic_inode(inode) \ 3955 (atomic_inc(&F2FS_I_SB(inode)->atomic_files)) 3956 #define stat_dec_atomic_inode(inode) \ 3957 (atomic_dec(&F2FS_I_SB(inode)->atomic_files)) 3958 #define stat_inc_meta_count(sbi, blkaddr) \ 3959 do { \ 3960 if (blkaddr < SIT_I(sbi)->sit_base_addr) \ 3961 atomic_inc(&(sbi)->meta_count[META_CP]); \ 3962 else if (blkaddr < NM_I(sbi)->nat_blkaddr) \ 3963 atomic_inc(&(sbi)->meta_count[META_SIT]); \ 3964 else if (blkaddr < SM_I(sbi)->ssa_blkaddr) \ 3965 atomic_inc(&(sbi)->meta_count[META_NAT]); \ 3966 else if (blkaddr < SM_I(sbi)->main_blkaddr) \ 3967 atomic_inc(&(sbi)->meta_count[META_SSA]); \ 3968 } while (0) 3969 #define stat_inc_seg_type(sbi, curseg) \ 3970 ((sbi)->segment_count[(curseg)->alloc_type]++) 3971 #define stat_inc_block_count(sbi, curseg) \ 3972 ((sbi)->block_count[(curseg)->alloc_type]++) 3973 #define stat_inc_inplace_blocks(sbi) \ 3974 (atomic_inc(&(sbi)->inplace_count)) 3975 #define stat_update_max_atomic_write(inode) \ 3976 do { \ 3977 int cur = atomic_read(&F2FS_I_SB(inode)->atomic_files); \ 3978 int max = atomic_read(&F2FS_I_SB(inode)->max_aw_cnt); \ 3979 if (cur > max) \ 3980 atomic_set(&F2FS_I_SB(inode)->max_aw_cnt, cur); \ 3981 } while (0) 3982 #define stat_inc_seg_count(sbi, type, gc_type) \ 3983 do { \ 3984 struct f2fs_stat_info *si = F2FS_STAT(sbi); \ 3985 si->tot_segs++; \ 3986 if ((type) == SUM_TYPE_DATA) { \ 3987 si->data_segs++; \ 3988 si->bg_data_segs += (gc_type == BG_GC) ? 1 : 0; \ 3989 } else { \ 3990 si->node_segs++; \ 3991 si->bg_node_segs += (gc_type == BG_GC) ? 1 : 0; \ 3992 } \ 3993 } while (0) 3994 3995 #define stat_inc_tot_blk_count(si, blks) \ 3996 ((si)->tot_blks += (blks)) 3997 3998 #define stat_inc_data_blk_count(sbi, blks, gc_type) \ 3999 do { \ 4000 struct f2fs_stat_info *si = F2FS_STAT(sbi); \ 4001 stat_inc_tot_blk_count(si, blks); \ 4002 si->data_blks += (blks); \ 4003 si->bg_data_blks += ((gc_type) == BG_GC) ? (blks) : 0; \ 4004 } while (0) 4005 4006 #define stat_inc_node_blk_count(sbi, blks, gc_type) \ 4007 do { \ 4008 struct f2fs_stat_info *si = F2FS_STAT(sbi); \ 4009 stat_inc_tot_blk_count(si, blks); \ 4010 si->node_blks += (blks); \ 4011 si->bg_node_blks += ((gc_type) == BG_GC) ? (blks) : 0; \ 4012 } while (0) 4013 4014 int f2fs_build_stats(struct f2fs_sb_info *sbi); 4015 void f2fs_destroy_stats(struct f2fs_sb_info *sbi); 4016 void __init f2fs_create_root_stats(void); 4017 void f2fs_destroy_root_stats(void); 4018 void f2fs_update_sit_info(struct f2fs_sb_info *sbi); 4019 #else 4020 #define stat_inc_cp_count(si) do { } while (0) 4021 #define stat_inc_bg_cp_count(si) do { } while (0) 4022 #define stat_inc_call_count(si) do { } while (0) 4023 #define stat_inc_bggc_count(si) do { } while (0) 4024 #define stat_io_skip_bggc_count(sbi) do { } while (0) 4025 #define stat_other_skip_bggc_count(sbi) do { } while (0) 4026 #define stat_inc_dirty_inode(sbi, type) do { } while (0) 4027 #define stat_dec_dirty_inode(sbi, type) do { } while (0) 4028 #define stat_inc_total_hit(sbi, type) do { } while (0) 4029 #define stat_inc_rbtree_node_hit(sbi, type) do { } while (0) 4030 #define stat_inc_largest_node_hit(sbi) do { } while (0) 4031 #define stat_inc_cached_node_hit(sbi, type) do { } while (0) 4032 #define stat_inc_inline_xattr(inode) do { } while (0) 4033 #define stat_dec_inline_xattr(inode) do { } while (0) 4034 #define stat_inc_inline_inode(inode) do { } while (0) 4035 #define stat_dec_inline_inode(inode) do { } while (0) 4036 #define stat_inc_inline_dir(inode) do { } while (0) 4037 #define stat_dec_inline_dir(inode) do { } while (0) 4038 #define stat_inc_compr_inode(inode) do { } while (0) 4039 #define stat_dec_compr_inode(inode) do { } while (0) 4040 #define stat_add_compr_blocks(inode, blocks) do { } while (0) 4041 #define stat_sub_compr_blocks(inode, blocks) do { } while (0) 4042 #define stat_inc_swapfile_inode(inode) do { } while (0) 4043 #define stat_dec_swapfile_inode(inode) do { } while (0) 4044 #define stat_inc_atomic_inode(inode) do { } while (0) 4045 #define stat_dec_atomic_inode(inode) do { } while (0) 4046 #define stat_update_max_atomic_write(inode) do { } while (0) 4047 #define stat_inc_meta_count(sbi, blkaddr) do { } while (0) 4048 #define stat_inc_seg_type(sbi, curseg) do { } while (0) 4049 #define stat_inc_block_count(sbi, curseg) do { } while (0) 4050 #define stat_inc_inplace_blocks(sbi) do { } while (0) 4051 #define stat_inc_seg_count(sbi, type, gc_type) do { } while (0) 4052 #define stat_inc_tot_blk_count(si, blks) do { } while (0) 4053 #define stat_inc_data_blk_count(sbi, blks, gc_type) do { } while (0) 4054 #define stat_inc_node_blk_count(sbi, blks, gc_type) do { } while (0) 4055 4056 static inline int f2fs_build_stats(struct f2fs_sb_info *sbi) { return 0; } 4057 static inline void f2fs_destroy_stats(struct f2fs_sb_info *sbi) { } 4058 static inline void __init f2fs_create_root_stats(void) { } 4059 static inline void f2fs_destroy_root_stats(void) { } 4060 static inline void f2fs_update_sit_info(struct f2fs_sb_info *sbi) {} 4061 #endif 4062 4063 extern const struct file_operations f2fs_dir_operations; 4064 extern const struct file_operations f2fs_file_operations; 4065 extern const struct inode_operations f2fs_file_inode_operations; 4066 extern const struct address_space_operations f2fs_dblock_aops; 4067 extern const struct address_space_operations f2fs_node_aops; 4068 extern const struct address_space_operations f2fs_meta_aops; 4069 extern const struct inode_operations f2fs_dir_inode_operations; 4070 extern const struct inode_operations f2fs_symlink_inode_operations; 4071 extern const struct inode_operations f2fs_encrypted_symlink_inode_operations; 4072 extern const struct inode_operations f2fs_special_inode_operations; 4073 extern struct kmem_cache *f2fs_inode_entry_slab; 4074 4075 /* 4076 * inline.c 4077 */ 4078 bool f2fs_may_inline_data(struct inode *inode); 4079 bool f2fs_sanity_check_inline_data(struct inode *inode); 4080 bool f2fs_may_inline_dentry(struct inode *inode); 4081 void f2fs_do_read_inline_data(struct page *page, struct page *ipage); 4082 void f2fs_truncate_inline_inode(struct inode *inode, 4083 struct page *ipage, u64 from); 4084 int f2fs_read_inline_data(struct inode *inode, struct page *page); 4085 int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page); 4086 int f2fs_convert_inline_inode(struct inode *inode); 4087 int f2fs_try_convert_inline_dir(struct inode *dir, struct dentry *dentry); 4088 int f2fs_write_inline_data(struct inode *inode, struct page *page); 4089 int f2fs_recover_inline_data(struct inode *inode, struct page *npage); 4090 struct f2fs_dir_entry *f2fs_find_in_inline_dir(struct inode *dir, 4091 const struct f2fs_filename *fname, 4092 struct page **res_page); 4093 int f2fs_make_empty_inline_dir(struct inode *inode, struct inode *parent, 4094 struct page *ipage); 4095 int f2fs_add_inline_entry(struct inode *dir, const struct f2fs_filename *fname, 4096 struct inode *inode, nid_t ino, umode_t mode); 4097 void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, 4098 struct page *page, struct inode *dir, 4099 struct inode *inode); 4100 bool f2fs_empty_inline_dir(struct inode *dir); 4101 int f2fs_read_inline_dir(struct file *file, struct dir_context *ctx, 4102 struct fscrypt_str *fstr); 4103 int f2fs_inline_data_fiemap(struct inode *inode, 4104 struct fiemap_extent_info *fieinfo, 4105 __u64 start, __u64 len); 4106 4107 /* 4108 * shrinker.c 4109 */ 4110 unsigned long f2fs_shrink_count(struct shrinker *shrink, 4111 struct shrink_control *sc); 4112 unsigned long f2fs_shrink_scan(struct shrinker *shrink, 4113 struct shrink_control *sc); 4114 void f2fs_join_shrinker(struct f2fs_sb_info *sbi); 4115 void f2fs_leave_shrinker(struct f2fs_sb_info *sbi); 4116 4117 /* 4118 * extent_cache.c 4119 */ 4120 bool sanity_check_extent_cache(struct inode *inode); 4121 void f2fs_init_extent_tree(struct inode *inode); 4122 void f2fs_drop_extent_tree(struct inode *inode); 4123 void f2fs_destroy_extent_node(struct inode *inode); 4124 void f2fs_destroy_extent_tree(struct inode *inode); 4125 void f2fs_init_extent_cache_info(struct f2fs_sb_info *sbi); 4126 int __init f2fs_create_extent_cache(void); 4127 void f2fs_destroy_extent_cache(void); 4128 4129 /* read extent cache ops */ 4130 void f2fs_init_read_extent_tree(struct inode *inode, struct page *ipage); 4131 bool f2fs_lookup_read_extent_cache(struct inode *inode, pgoff_t pgofs, 4132 struct extent_info *ei); 4133 bool f2fs_lookup_read_extent_cache_block(struct inode *inode, pgoff_t index, 4134 block_t *blkaddr); 4135 void f2fs_update_read_extent_cache(struct dnode_of_data *dn); 4136 void f2fs_update_read_extent_cache_range(struct dnode_of_data *dn, 4137 pgoff_t fofs, block_t blkaddr, unsigned int len); 4138 unsigned int f2fs_shrink_read_extent_tree(struct f2fs_sb_info *sbi, 4139 int nr_shrink); 4140 4141 /* block age extent cache ops */ 4142 void f2fs_init_age_extent_tree(struct inode *inode); 4143 bool f2fs_lookup_age_extent_cache(struct inode *inode, pgoff_t pgofs, 4144 struct extent_info *ei); 4145 void f2fs_update_age_extent_cache(struct dnode_of_data *dn); 4146 void f2fs_update_age_extent_cache_range(struct dnode_of_data *dn, 4147 pgoff_t fofs, unsigned int len); 4148 unsigned int f2fs_shrink_age_extent_tree(struct f2fs_sb_info *sbi, 4149 int nr_shrink); 4150 4151 /* 4152 * sysfs.c 4153 */ 4154 #define MIN_RA_MUL 2 4155 #define MAX_RA_MUL 256 4156 4157 int __init f2fs_init_sysfs(void); 4158 void f2fs_exit_sysfs(void); 4159 int f2fs_register_sysfs(struct f2fs_sb_info *sbi); 4160 void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi); 4161 4162 /* verity.c */ 4163 extern const struct fsverity_operations f2fs_verityops; 4164 4165 /* 4166 * crypto support 4167 */ 4168 static inline bool f2fs_encrypted_file(struct inode *inode) 4169 { 4170 return IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode); 4171 } 4172 4173 static inline void f2fs_set_encrypted_inode(struct inode *inode) 4174 { 4175 #ifdef CONFIG_FS_ENCRYPTION 4176 file_set_encrypt(inode); 4177 f2fs_set_inode_flags(inode); 4178 #endif 4179 } 4180 4181 /* 4182 * Returns true if the reads of the inode's data need to undergo some 4183 * postprocessing step, like decryption or authenticity verification. 4184 */ 4185 static inline bool f2fs_post_read_required(struct inode *inode) 4186 { 4187 return f2fs_encrypted_file(inode) || fsverity_active(inode) || 4188 f2fs_compressed_file(inode); 4189 } 4190 4191 /* 4192 * compress.c 4193 */ 4194 #ifdef CONFIG_F2FS_FS_COMPRESSION 4195 bool f2fs_is_compressed_page(struct page *page); 4196 struct page *f2fs_compress_control_page(struct page *page); 4197 int f2fs_prepare_compress_overwrite(struct inode *inode, 4198 struct page **pagep, pgoff_t index, void **fsdata); 4199 bool f2fs_compress_write_end(struct inode *inode, void *fsdata, 4200 pgoff_t index, unsigned copied); 4201 int f2fs_truncate_partial_cluster(struct inode *inode, u64 from, bool lock); 4202 void f2fs_compress_write_end_io(struct bio *bio, struct page *page); 4203 bool f2fs_is_compress_backend_ready(struct inode *inode); 4204 int __init f2fs_init_compress_mempool(void); 4205 void f2fs_destroy_compress_mempool(void); 4206 void f2fs_decompress_cluster(struct decompress_io_ctx *dic, bool in_task); 4207 void f2fs_end_read_compressed_page(struct page *page, bool failed, 4208 block_t blkaddr, bool in_task); 4209 bool f2fs_cluster_is_empty(struct compress_ctx *cc); 4210 bool f2fs_cluster_can_merge_page(struct compress_ctx *cc, pgoff_t index); 4211 bool f2fs_all_cluster_page_ready(struct compress_ctx *cc, struct page **pages, 4212 int index, int nr_pages, bool uptodate); 4213 bool f2fs_sanity_check_cluster(struct dnode_of_data *dn); 4214 void f2fs_compress_ctx_add_page(struct compress_ctx *cc, struct page *page); 4215 int f2fs_write_multi_pages(struct compress_ctx *cc, 4216 int *submitted, 4217 struct writeback_control *wbc, 4218 enum iostat_type io_type); 4219 int f2fs_is_compressed_cluster(struct inode *inode, pgoff_t index); 4220 void f2fs_update_read_extent_tree_range_compressed(struct inode *inode, 4221 pgoff_t fofs, block_t blkaddr, 4222 unsigned int llen, unsigned int c_len); 4223 int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, 4224 unsigned nr_pages, sector_t *last_block_in_bio, 4225 bool is_readahead, bool for_write); 4226 struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc); 4227 void f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed, 4228 bool in_task); 4229 void f2fs_put_page_dic(struct page *page, bool in_task); 4230 unsigned int f2fs_cluster_blocks_are_contiguous(struct dnode_of_data *dn); 4231 int f2fs_init_compress_ctx(struct compress_ctx *cc); 4232 void f2fs_destroy_compress_ctx(struct compress_ctx *cc, bool reuse); 4233 void f2fs_init_compress_info(struct f2fs_sb_info *sbi); 4234 int f2fs_init_compress_inode(struct f2fs_sb_info *sbi); 4235 void f2fs_destroy_compress_inode(struct f2fs_sb_info *sbi); 4236 int f2fs_init_page_array_cache(struct f2fs_sb_info *sbi); 4237 void f2fs_destroy_page_array_cache(struct f2fs_sb_info *sbi); 4238 int __init f2fs_init_compress_cache(void); 4239 void f2fs_destroy_compress_cache(void); 4240 struct address_space *COMPRESS_MAPPING(struct f2fs_sb_info *sbi); 4241 void f2fs_invalidate_compress_page(struct f2fs_sb_info *sbi, block_t blkaddr); 4242 void f2fs_cache_compressed_page(struct f2fs_sb_info *sbi, struct page *page, 4243 nid_t ino, block_t blkaddr); 4244 bool f2fs_load_compressed_page(struct f2fs_sb_info *sbi, struct page *page, 4245 block_t blkaddr); 4246 void f2fs_invalidate_compress_pages(struct f2fs_sb_info *sbi, nid_t ino); 4247 #define inc_compr_inode_stat(inode) \ 4248 do { \ 4249 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); \ 4250 sbi->compr_new_inode++; \ 4251 } while (0) 4252 #define add_compr_block_stat(inode, blocks) \ 4253 do { \ 4254 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); \ 4255 int diff = F2FS_I(inode)->i_cluster_size - blocks; \ 4256 sbi->compr_written_block += blocks; \ 4257 sbi->compr_saved_block += diff; \ 4258 } while (0) 4259 #else 4260 static inline bool f2fs_is_compressed_page(struct page *page) { return false; } 4261 static inline bool f2fs_is_compress_backend_ready(struct inode *inode) 4262 { 4263 if (!f2fs_compressed_file(inode)) 4264 return true; 4265 /* not support compression */ 4266 return false; 4267 } 4268 static inline struct page *f2fs_compress_control_page(struct page *page) 4269 { 4270 WARN_ON_ONCE(1); 4271 return ERR_PTR(-EINVAL); 4272 } 4273 static inline int __init f2fs_init_compress_mempool(void) { return 0; } 4274 static inline void f2fs_destroy_compress_mempool(void) { } 4275 static inline void f2fs_decompress_cluster(struct decompress_io_ctx *dic, 4276 bool in_task) { } 4277 static inline void f2fs_end_read_compressed_page(struct page *page, 4278 bool failed, block_t blkaddr, bool in_task) 4279 { 4280 WARN_ON_ONCE(1); 4281 } 4282 static inline void f2fs_put_page_dic(struct page *page, bool in_task) 4283 { 4284 WARN_ON_ONCE(1); 4285 } 4286 static inline unsigned int f2fs_cluster_blocks_are_contiguous(struct dnode_of_data *dn) { return 0; } 4287 static inline bool f2fs_sanity_check_cluster(struct dnode_of_data *dn) { return false; } 4288 static inline int f2fs_init_compress_inode(struct f2fs_sb_info *sbi) { return 0; } 4289 static inline void f2fs_destroy_compress_inode(struct f2fs_sb_info *sbi) { } 4290 static inline int f2fs_init_page_array_cache(struct f2fs_sb_info *sbi) { return 0; } 4291 static inline void f2fs_destroy_page_array_cache(struct f2fs_sb_info *sbi) { } 4292 static inline int __init f2fs_init_compress_cache(void) { return 0; } 4293 static inline void f2fs_destroy_compress_cache(void) { } 4294 static inline void f2fs_invalidate_compress_page(struct f2fs_sb_info *sbi, 4295 block_t blkaddr) { } 4296 static inline void f2fs_cache_compressed_page(struct f2fs_sb_info *sbi, 4297 struct page *page, nid_t ino, block_t blkaddr) { } 4298 static inline bool f2fs_load_compressed_page(struct f2fs_sb_info *sbi, 4299 struct page *page, block_t blkaddr) { return false; } 4300 static inline void f2fs_invalidate_compress_pages(struct f2fs_sb_info *sbi, 4301 nid_t ino) { } 4302 #define inc_compr_inode_stat(inode) do { } while (0) 4303 static inline void f2fs_update_read_extent_tree_range_compressed( 4304 struct inode *inode, 4305 pgoff_t fofs, block_t blkaddr, 4306 unsigned int llen, unsigned int c_len) { } 4307 #endif 4308 4309 static inline int set_compress_context(struct inode *inode) 4310 { 4311 #ifdef CONFIG_F2FS_FS_COMPRESSION 4312 struct f2fs_sb_info *sbi = F2FS_I_SB(inode); 4313 4314 F2FS_I(inode)->i_compress_algorithm = 4315 F2FS_OPTION(sbi).compress_algorithm; 4316 F2FS_I(inode)->i_log_cluster_size = 4317 F2FS_OPTION(sbi).compress_log_size; 4318 F2FS_I(inode)->i_compress_flag = 4319 F2FS_OPTION(sbi).compress_chksum ? 4320 BIT(COMPRESS_CHKSUM) : 0; 4321 F2FS_I(inode)->i_cluster_size = 4322 BIT(F2FS_I(inode)->i_log_cluster_size); 4323 if ((F2FS_I(inode)->i_compress_algorithm == COMPRESS_LZ4 || 4324 F2FS_I(inode)->i_compress_algorithm == COMPRESS_ZSTD) && 4325 F2FS_OPTION(sbi).compress_level) 4326 F2FS_I(inode)->i_compress_level = 4327 F2FS_OPTION(sbi).compress_level; 4328 F2FS_I(inode)->i_flags |= F2FS_COMPR_FL; 4329 set_inode_flag(inode, FI_COMPRESSED_FILE); 4330 stat_inc_compr_inode(inode); 4331 inc_compr_inode_stat(inode); 4332 f2fs_mark_inode_dirty_sync(inode, true); 4333 return 0; 4334 #else 4335 return -EOPNOTSUPP; 4336 #endif 4337 } 4338 4339 static inline bool f2fs_disable_compressed_file(struct inode *inode) 4340 { 4341 struct f2fs_inode_info *fi = F2FS_I(inode); 4342 4343 if (!f2fs_compressed_file(inode)) 4344 return true; 4345 if (S_ISREG(inode->i_mode) && F2FS_HAS_BLOCKS(inode)) 4346 return false; 4347 4348 fi->i_flags &= ~F2FS_COMPR_FL; 4349 stat_dec_compr_inode(inode); 4350 clear_inode_flag(inode, FI_COMPRESSED_FILE); 4351 f2fs_mark_inode_dirty_sync(inode, true); 4352 return true; 4353 } 4354 4355 #define F2FS_FEATURE_FUNCS(name, flagname) \ 4356 static inline bool f2fs_sb_has_##name(struct f2fs_sb_info *sbi) \ 4357 { \ 4358 return F2FS_HAS_FEATURE(sbi, F2FS_FEATURE_##flagname); \ 4359 } 4360 4361 F2FS_FEATURE_FUNCS(encrypt, ENCRYPT); 4362 F2FS_FEATURE_FUNCS(blkzoned, BLKZONED); 4363 F2FS_FEATURE_FUNCS(extra_attr, EXTRA_ATTR); 4364 F2FS_FEATURE_FUNCS(project_quota, PRJQUOTA); 4365 F2FS_FEATURE_FUNCS(inode_chksum, INODE_CHKSUM); 4366 F2FS_FEATURE_FUNCS(flexible_inline_xattr, FLEXIBLE_INLINE_XATTR); 4367 F2FS_FEATURE_FUNCS(quota_ino, QUOTA_INO); 4368 F2FS_FEATURE_FUNCS(inode_crtime, INODE_CRTIME); 4369 F2FS_FEATURE_FUNCS(lost_found, LOST_FOUND); 4370 F2FS_FEATURE_FUNCS(verity, VERITY); 4371 F2FS_FEATURE_FUNCS(sb_chksum, SB_CHKSUM); 4372 F2FS_FEATURE_FUNCS(casefold, CASEFOLD); 4373 F2FS_FEATURE_FUNCS(compression, COMPRESSION); 4374 F2FS_FEATURE_FUNCS(readonly, RO); 4375 4376 #ifdef CONFIG_BLK_DEV_ZONED 4377 static inline bool f2fs_blkz_is_seq(struct f2fs_sb_info *sbi, int devi, 4378 block_t blkaddr) 4379 { 4380 unsigned int zno = blkaddr >> sbi->log_blocks_per_blkz; 4381 4382 return test_bit(zno, FDEV(devi).blkz_seq); 4383 } 4384 #endif 4385 4386 static inline bool f2fs_hw_should_discard(struct f2fs_sb_info *sbi) 4387 { 4388 return f2fs_sb_has_blkzoned(sbi); 4389 } 4390 4391 static inline bool f2fs_bdev_support_discard(struct block_device *bdev) 4392 { 4393 return bdev_max_discard_sectors(bdev) || bdev_is_zoned(bdev); 4394 } 4395 4396 static inline bool f2fs_hw_support_discard(struct f2fs_sb_info *sbi) 4397 { 4398 int i; 4399 4400 if (!f2fs_is_multi_device(sbi)) 4401 return f2fs_bdev_support_discard(sbi->sb->s_bdev); 4402 4403 for (i = 0; i < sbi->s_ndevs; i++) 4404 if (f2fs_bdev_support_discard(FDEV(i).bdev)) 4405 return true; 4406 return false; 4407 } 4408 4409 static inline bool f2fs_realtime_discard_enable(struct f2fs_sb_info *sbi) 4410 { 4411 return (test_opt(sbi, DISCARD) && f2fs_hw_support_discard(sbi)) || 4412 f2fs_hw_should_discard(sbi); 4413 } 4414 4415 static inline bool f2fs_hw_is_readonly(struct f2fs_sb_info *sbi) 4416 { 4417 int i; 4418 4419 if (!f2fs_is_multi_device(sbi)) 4420 return bdev_read_only(sbi->sb->s_bdev); 4421 4422 for (i = 0; i < sbi->s_ndevs; i++) 4423 if (bdev_read_only(FDEV(i).bdev)) 4424 return true; 4425 return false; 4426 } 4427 4428 static inline bool f2fs_dev_is_readonly(struct f2fs_sb_info *sbi) 4429 { 4430 return f2fs_sb_has_readonly(sbi) || f2fs_hw_is_readonly(sbi); 4431 } 4432 4433 static inline bool f2fs_lfs_mode(struct f2fs_sb_info *sbi) 4434 { 4435 return F2FS_OPTION(sbi).fs_mode == FS_MODE_LFS; 4436 } 4437 4438 static inline bool f2fs_low_mem_mode(struct f2fs_sb_info *sbi) 4439 { 4440 return F2FS_OPTION(sbi).memory_mode == MEMORY_MODE_LOW; 4441 } 4442 4443 static inline bool f2fs_may_compress(struct inode *inode) 4444 { 4445 if (IS_SWAPFILE(inode) || f2fs_is_pinned_file(inode) || 4446 f2fs_is_atomic_file(inode) || f2fs_has_inline_data(inode)) 4447 return false; 4448 return S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode); 4449 } 4450 4451 static inline void f2fs_i_compr_blocks_update(struct inode *inode, 4452 u64 blocks, bool add) 4453 { 4454 struct f2fs_inode_info *fi = F2FS_I(inode); 4455 int diff = fi->i_cluster_size - blocks; 4456 4457 /* don't update i_compr_blocks if saved blocks were released */ 4458 if (!add && !atomic_read(&fi->i_compr_blocks)) 4459 return; 4460 4461 if (add) { 4462 atomic_add(diff, &fi->i_compr_blocks); 4463 stat_add_compr_blocks(inode, diff); 4464 } else { 4465 atomic_sub(diff, &fi->i_compr_blocks); 4466 stat_sub_compr_blocks(inode, diff); 4467 } 4468 f2fs_mark_inode_dirty_sync(inode, true); 4469 } 4470 4471 static inline bool f2fs_allow_multi_device_dio(struct f2fs_sb_info *sbi, 4472 int flag) 4473 { 4474 if (!f2fs_is_multi_device(sbi)) 4475 return false; 4476 if (flag != F2FS_GET_BLOCK_DIO) 4477 return false; 4478 return sbi->aligned_blksize; 4479 } 4480 4481 static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx) 4482 { 4483 return fsverity_active(inode) && 4484 idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE); 4485 } 4486 4487 #ifdef CONFIG_F2FS_FAULT_INJECTION 4488 extern void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, 4489 unsigned int type); 4490 #else 4491 #define f2fs_build_fault_attr(sbi, rate, type) do { } while (0) 4492 #endif 4493 4494 static inline bool is_journalled_quota(struct f2fs_sb_info *sbi) 4495 { 4496 #ifdef CONFIG_QUOTA 4497 if (f2fs_sb_has_quota_ino(sbi)) 4498 return true; 4499 if (F2FS_OPTION(sbi).s_qf_names[USRQUOTA] || 4500 F2FS_OPTION(sbi).s_qf_names[GRPQUOTA] || 4501 F2FS_OPTION(sbi).s_qf_names[PRJQUOTA]) 4502 return true; 4503 #endif 4504 return false; 4505 } 4506 4507 static inline bool f2fs_block_unit_discard(struct f2fs_sb_info *sbi) 4508 { 4509 return F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_BLOCK; 4510 } 4511 4512 static inline void f2fs_io_schedule_timeout(long timeout) 4513 { 4514 set_current_state(TASK_UNINTERRUPTIBLE); 4515 io_schedule_timeout(timeout); 4516 } 4517 4518 static inline void f2fs_handle_page_eio(struct f2fs_sb_info *sbi, pgoff_t ofs, 4519 enum page_type type) 4520 { 4521 if (unlikely(f2fs_cp_error(sbi))) 4522 return; 4523 4524 if (ofs == sbi->page_eio_ofs[type]) { 4525 if (sbi->page_eio_cnt[type]++ == MAX_RETRY_PAGE_EIO) 4526 set_ckpt_flags(sbi, CP_ERROR_FLAG); 4527 } else { 4528 sbi->page_eio_ofs[type] = ofs; 4529 sbi->page_eio_cnt[type] = 0; 4530 } 4531 } 4532 4533 static inline bool f2fs_is_readonly(struct f2fs_sb_info *sbi) 4534 { 4535 return f2fs_sb_has_readonly(sbi) || f2fs_readonly(sbi->sb); 4536 } 4537 4538 #define EFSBADCRC EBADMSG /* Bad CRC detected */ 4539 #define EFSCORRUPTED EUCLEAN /* Filesystem is corrupted */ 4540 4541 #endif /* _LINUX_F2FS_H */ 4542