Documentation/filesystems/f2fs.txt

   1 ================================================================================
   2 WHAT IS Flash-Friendly File System (F2FS)?
   3 ================================================================================
   4
   5 NAND flash memory-based storage devices, such as SSD, eMMC, and SD cards, have
   6 been equipped on a variety systems ranging from mobile to server systems. Since
   7 they are known to have different characteristics from the conventional rotating
   8 disks, a file system, an upper layer to the storage device, should adapt to the
   9 changes from the sketch in the design level.
  10
  11 F2FS is a file system exploiting NAND flash memory-based storage devices, which
  12 is based on Log-structured File System (LFS). The design has been focused on
  13 addressing the fundamental issues in LFS, which are snowball effect of wandering
  14 tree and high cleaning overhead.
  15
  16 Since a NAND flash memory-based storage device shows different characteristic
  17 according to its internal geometry or flash memory management scheme, namely FTL,
  18 F2FS and its tools support various parameters not only for configuring on-disk
  19 layout, but also for selecting allocation and cleaning algorithms.
  20
  21 The following git tree provides the file system formatting tool (mkfs.f2fs),
  22 a consistency checking tool (fsck.f2fs), and a debugging tool (dump.f2fs).
  23 >> git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git
  24
  25 For reporting bugs and sending patches, please use the following mailing list:
  26 >> linux-f2fs-devel@lists.sourceforge.net
  27
  28 ================================================================================
  29 BACKGROUND AND DESIGN ISSUES
  30 ================================================================================
  31
  32 Log-structured File System (LFS)
  33 --------------------------------
  34 "A log-structured file system writes all modifications to disk sequentially in
  35 a log-like structure, thereby speeding up  both file writing and crash recovery.
  36 The log is the only structure on disk; it contains indexing information so that
  37 files can be read back from the log efficiently. In order to maintain large free
  38 areas on disk for fast writing, we divide  the log into segments and use a
  39 segment cleaner to compress the live information from heavily fragmented
  40 segments." from Rosenblum, M. and Ousterhout, J. K., 1992, "The design and
  41 implementation of a log-structured file system", ACM Trans. Computer Systems
  42 10, 1, 26–52.
  43
  44 Wandering Tree Problem
  45 ----------------------
  46 In LFS, when a file data is updated and written to the end of log, its direct
  47 pointer block is updated due to the changed location. Then the indirect pointer
  48 block is also updated due to the direct pointer block update. In this manner,
  49 the upper index structures such as inode, inode map, and checkpoint block are
  50 also updated recursively. This problem is called as wandering tree problem [1],
  51 and in order to enhance the performance, it should eliminate or relax the update
  52 propagation as much as possible.
  53
  54 [1] Bityutskiy, A. 2005. JFFS3 design issues. http://www.linux-mtd.infradead.org/
  55
  56 Cleaning Overhead
  57 -----------------
  58 Since LFS is based on out-of-place writes, it produces so many obsolete blocks
  59 scattered across the whole storage. In order to serve new empty log space, it
  60 needs to reclaim these obsolete blocks seamlessly to users. This job is called
  61 as a cleaning process.
  62
  63 The process consists of three operations as follows.
  64 1. A victim segment is selected through referencing segment usage table.
  65 2. It loads parent index structures of all the data in the victim identified by
  66    segment summary blocks.
  67 3. It checks the cross-reference between the data and its parent index structure.
  68 4. It moves valid data selectively.
  69
  70 This cleaning job may cause unexpected long delays, so the most important goal
  71 is to hide the latencies to users. And also definitely, it should reduce the
  72 amount of valid data to be moved, and move them quickly as well.
  73
  74 ================================================================================
  75 KEY FEATURES
  76 ================================================================================
  77
  78 Flash Awareness
  79 ---------------
  80 - Enlarge the random write area for better performance, but provide the high
  81   spatial locality
  82 - Align FS data structures to the operational units in FTL as best efforts
  83
  84 Wandering Tree Problem
  85 ----------------------
  86 - Use a term, “node”, that represents inodes as well as various pointer blocks
  87 - Introduce Node Address Table (NAT) containing the locations of all the “node”
  88   blocks; this will cut off the update propagation.
  89
  90 Cleaning Overhead
  91 -----------------
  92 - Support a background cleaning process
  93 - Support greedy and cost-benefit algorithms for victim selection policies
  94 - Support multi-head logs for static/dynamic hot and cold data separation
  95 - Introduce adaptive logging for efficient block allocation
  96
  97 ================================================================================
  98 MOUNT OPTIONS
  99 ================================================================================
 100
 101 background_gc=%s       Turn on/off cleaning operations, namely garbage
 102                        collection, triggered in background when I/O subsystem is
 103                        idle. If background_gc=on, it will turn on the garbage
 104                        collection and if background_gc=off, garbage collection
 105                        will be turned off. If background_gc=sync, it will turn
 106                        on synchronous garbage collection running in background.
 107                        Default value for this option is on. So garbage
 108                        collection is on by default.
 109 disable_roll_forward   Disable the roll-forward recovery routine
 110 norecovery             Disable the roll-forward recovery routine, mounted read-
 111                        only (i.e., -o ro,disable_roll_forward)
 112 discard/nodiscard      Enable/disable real-time discard in f2fs, if discard is
 113                        enabled, f2fs will issue discard/TRIM commands when a
 114                        segment is cleaned.
 115 no_heap                Disable heap-style segment allocation which finds free
 116                        segments for data from the beginning of main area, while
 117                        for node from the end of main area.
 118 nouser_xattr           Disable Extended User Attributes. Note: xattr is enabled
 119                        by default if CONFIG_F2FS_FS_XATTR is selected.
 120 noacl                  Disable POSIX Access Control List. Note: acl is enabled
 121                        by default if CONFIG_F2FS_FS_POSIX_ACL is selected.
 122 active_logs=%u         Support configuring the number of active logs. In the
 123                        current design, f2fs supports only 2, 4, and 6 logs.
 124                        Default number is 6.
 125 disable_ext_identify   Disable the extension list configured by mkfs, so f2fs
 126                        does not aware of cold files such as media files.
 127 inline_xattr           Enable the inline xattrs feature.
 128 inline_data            Enable the inline data feature: New created small(<~3.4k)
 129                        files can be written into inode block.
 130 inline_dentry          Enable the inline dir feature: data in new created
 131                        directory entries can be written into inode block. The
 132                        space of inode block which is used to store inline
 133                        dentries is limited to ~3.4k.
 134 noinline_dentry        Diable the inline dentry feature.
 135 flush_merge            Merge concurrent cache_flush commands as much as possible
 136                        to eliminate redundant command issues. If the underlying
 137                        device handles the cache_flush command relatively slowly,
 138                        recommend to enable this option.
 139 nobarrier              This option can be used if underlying storage guarantees
 140                        its cached data should be written to the novolatile area.
 141                        If this option is set, no cache_flush commands are issued
 142                        but f2fs still guarantees the write ordering of all the
 143                        data writes.
 144 fastboot               This option is used when a system wants to reduce mount
 145                        time as much as possible, even though normal performance
 146                        can be sacrificed.
 147 extent_cache           Enable an extent cache based on rb-tree, it can cache
 148                        as many as extent which map between contiguous logical
 149                        address and physical address per inode, resulting in
 150                        increasing the cache hit ratio. Set by default.
 151 noextent_cache         Disable an extent cache based on rb-tree explicitly, see
 152                        the above extent_cache mount option.
 153 noinline_data          Disable the inline data feature, inline data feature is
 154                        enabled by default.
 155 data_flush             Enable data flushing before checkpoint in order to
 156                        persist data of regular and symlink.
 157 mode=%s                Control block allocation mode which supports "adaptive"
 158                        and "lfs". In "lfs" mode, there should be no random
 159                        writes towards main area.
 160 io_bits=%u             Set the bit size of write IO requests. It should be set
 161                        with "mode=lfs".
 162
 163 ================================================================================
 164 DEBUGFS ENTRIES
 165 ================================================================================
 166
 167 /sys/kernel/debug/f2fs/ contains information about all the partitions mounted as
 168 f2fs. Each file shows the whole f2fs information.
 169
 170 /sys/kernel/debug/f2fs/status includes:
 171  - major file system information managed by f2fs currently
 172  - average SIT information about whole segments
 173  - current memory footprint consumed by f2fs.
 174
 175 ================================================================================
 176 SYSFS ENTRIES
 177 ================================================================================
 178
 179 Information about mounted f2f2 file systems can be found in
 180 /sys/fs/f2fs.  Each mounted filesystem will have a directory in
 181 /sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda).
 182 The files in each per-device directory are shown in table below.
 183
 184 Files in /sys/fs/f2fs/<devname>
 185 (see also Documentation/ABI/testing/sysfs-fs-f2fs)
 186 ..............................................................................
 187  File                         Content
 188
 189  gc_max_sleep_time            This tuning parameter controls the maximum sleep
 190                               time for the garbage collection thread. Time is
 191                               in milliseconds.
 192
 193  gc_min_sleep_time            This tuning parameter controls the minimum sleep
 194                               time for the garbage collection thread. Time is
 195                               in milliseconds.
 196
 197  gc_no_gc_sleep_time          This tuning parameter controls the default sleep
 198                               time for the garbage collection thread. Time is
 199                               in milliseconds.
 200
 201  gc_idle                      This parameter controls the selection of victim
 202                               policy for garbage collection. Setting gc_idle = 0
 203                               (default) will disable this option. Setting
 204                               gc_idle = 1 will select the Cost Benefit approach
 205                               & setting gc_idle = 2 will select the greedy approach.
 206
 207  reclaim_segments             This parameter controls the number of prefree
 208                               segments to be reclaimed. If the number of prefree
 209                               segments is larger than the number of segments
 210                               in the proportion to the percentage over total
 211                               volume size, f2fs tries to conduct checkpoint to
 212                               reclaim the prefree segments to free segments.
 213                               By default, 5% over total # of segments.
 214
 215  max_small_discards           This parameter controls the number of discard
 216                               commands that consist small blocks less than 2MB.
 217                               The candidates to be discarded are cached until
 218                               checkpoint is triggered, and issued during the
 219                               checkpoint. By default, it is disabled with 0.
 220
 221  trim_sections                This parameter controls the number of sections
 222                               to be trimmed out in batch mode when FITRIM
 223                               conducts. 32 sections is set by default.
 224
 225  ipu_policy                   This parameter controls the policy of in-place
 226                               updates in f2fs. There are five policies:
 227                                0x01: F2FS_IPU_FORCE, 0x02: F2FS_IPU_SSR,
 228                                0x04: F2FS_IPU_UTIL,  0x08: F2FS_IPU_SSR_UTIL,
 229                                0x10: F2FS_IPU_FSYNC.
 230
 231  min_ipu_util                 This parameter controls the threshold to trigger
 232                               in-place-updates. The number indicates percentage
 233                               of the filesystem utilization, and used by
 234                               F2FS_IPU_UTIL and F2FS_IPU_SSR_UTIL policies.
 235
 236  min_fsync_blocks             This parameter controls the threshold to trigger
 237                               in-place-updates when F2FS_IPU_FSYNC mode is set.
 238                               The number indicates the number of dirty pages
 239                               when fsync needs to flush on its call path. If
 240                               the number is less than this value, it triggers
 241                               in-place-updates.
 242
 243  max_victim_search            This parameter controls the number of trials to
 244                               find a victim segment when conducting SSR and
 245                               cleaning operations. The default value is 4096
 246                               which covers 8GB block address range.
 247
 248  dir_level                    This parameter controls the directory level to
 249                               support large directory. If a directory has a
 250                               number of files, it can reduce the file lookup
 251                               latency by increasing this dir_level value.
 252                               Otherwise, it needs to decrease this value to
 253                               reduce the space overhead. The default value is 0.
 254
 255  ram_thresh                   This parameter controls the memory footprint used
 256                               by free nids and cached nat entries. By default,
 257                               10 is set, which indicates 10 MB / 1 GB RAM.
 258
 259 ================================================================================
 260 USAGE
 261 ================================================================================
 262
 263 1. Download userland tools and compile them.
 264
 265 2. Skip, if f2fs was compiled statically inside kernel.
 266    Otherwise, insert the f2fs.ko module.
 267  # insmod f2fs.ko
 268
 269 3. Create a directory trying to mount
 270  # mkdir /mnt/f2fs
 271
 272 4. Format the block device, and then mount as f2fs
 273  # mkfs.f2fs -l label /dev/block_device
 274  # mount -t f2fs /dev/block_device /mnt/f2fs
 275
 276 mkfs.f2fs
 277 ---------
 278 The mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem,
 279 which builds a basic on-disk layout.
 280
 281 The options consist of:
 282 -l [label]   : Give a volume label, up to 512 unicode name.
 283 -a [0 or 1]  : Split start location of each area for heap-based allocation.
 284                1 is set by default, which performs this.
 285 -o [int]     : Set overprovision ratio in percent over volume size.
 286                5 is set by default.
 287 -s [int]     : Set the number of segments per section.
 288                1 is set by default.
 289 -z [int]     : Set the number of sections per zone.
 290                1 is set by default.
 291 -e [str]     : Set basic extension list. e.g. "mp3,gif,mov"
 292 -t [0 or 1]  : Disable discard command or not.
 293                1 is set by default, which conducts discard.
 294
 295 fsck.f2fs
 296 ---------
 297 The fsck.f2fs is a tool to check the consistency of an f2fs-formatted
 298 partition, which examines whether the filesystem metadata and user-made data
 299 are cross-referenced correctly or not.
 300 Note that, initial version of the tool does not fix any inconsistency.
 301
 302 The options consist of:
 303   -d debug level [default:0]
 304
 305 dump.f2fs
 306 ---------
 307 The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
 308 file. Each file is dump_ssa and dump_sit.
 309
 310 The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
 311 It shows on-disk inode information recognized by a given inode number, and is
 312 able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
 313 ./dump_sit respectively.
 314
 315 The options consist of:
 316   -d debug level [default:0]
 317   -i inode no (hex)
 318   -s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
 319   -a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
 320
 321 Examples:
 322 # dump.f2fs -i [ino] /dev/sdx
 323 # dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
 324 # dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
 325
 326 ================================================================================
 327 DESIGN
 328 ================================================================================
 329
 330 On-disk Layout
 331 --------------
 332
 333 F2FS divides the whole volume into a number of segments, each of which is fixed
 334 to 2MB in size. A section is composed of consecutive segments, and a zone
 335 consists of a set of sections. By default, section and zone sizes are set to one
 336 segment size identically, but users can easily modify the sizes by mkfs.
 337
 338 F2FS splits the entire volume into six areas, and all the areas except superblock
 339 consists of multiple segments as described below.
 340
 341                                             align with the zone size <-|
 342                  |-> align with the segment size
 343      _________________________________________________________________________
 344     |            |            |   Segment   |    Node     |   Segment  |      |
 345     | Superblock | Checkpoint |    Info.    |   Address   |   Summary  | Main |
 346     |    (SB)    |   (CP)     | Table (SIT) | Table (NAT) | Area (SSA) |      |
 347     |____________|_____2______|______N______|______N______|______N_____|__N___|
 348                                                                        .      .
 349                                                              .                .
 350                                                  .                            .
 351                                     ._________________________________________.
 352                                     |_Segment_|_..._|_Segment_|_..._|_Segment_|
 353                                     .           .
 354                                     ._________._________
 355                                     |_section_|__...__|_
 356                                     .            .
 357                                     .________.
 358                                     |__zone__|
 359
 360 - Superblock (SB)
 361  : It is located at the beginning of the partition, and there exist two copies
 362    to avoid file system crash. It contains basic partition information and some
 363    default parameters of f2fs.
 364
 365 - Checkpoint (CP)
 366  : It contains file system information, bitmaps for valid NAT/SIT sets, orphan
 367    inode lists, and summary entries of current active segments.
 368
 369 - Segment Information Table (SIT)
 370  : It contains segment information such as valid block count and bitmap for the
 371    validity of all the blocks.
 372
 373 - Node Address Table (NAT)
 374  : It is composed of a block address table for all the node blocks stored in
 375    Main area.
 376
 377 - Segment Summary Area (SSA)
 378  : It contains summary entries which contains the owner information of all the
 379    data and node blocks stored in Main area.
 380
 381 - Main Area
 382  : It contains file and directory data including their indices.
 383
 384 In order to avoid misalignment between file system and flash-based storage, F2FS
 385 aligns the start block address of CP with the segment size. Also, it aligns the
 386 start block address of Main area with the zone size by reserving some segments
 387 in SSA area.
 388
 389 Reference the following survey for additional technical details.
 390 https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey
 391
 392 File System Metadata Structure
 393 ------------------------------
 394
 395 F2FS adopts the checkpointing scheme to maintain file system consistency. At
 396 mount time, F2FS first tries to find the last valid checkpoint data by scanning
 397 CP area. In order to reduce the scanning time, F2FS uses only two copies of CP.
 398 One of them always indicates the last valid data, which is called as shadow copy
 399 mechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism.
 400
 401 For file system consistency, each CP points to which NAT and SIT copies are
 402 valid, as shown as below.
 403
 404   +--------+----------+---------+
 405   |   CP   |    SIT   |   NAT   |
 406   +--------+----------+---------+
 407   .         .          .          .
 408   .            .              .              .
 409   .               .                 .                 .
 410   +-------+-------+--------+--------+--------+--------+
 411   | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 |
 412   +-------+-------+--------+--------+--------+--------+
 413      |             ^                          ^
 414      |             |                          |
 415      `----------------------------------------'
 416
 417 Index Structure
 418 ---------------
 419
 420 The key data structure to manage the data locations is a "node". Similar to
 421 traditional file structures, F2FS has three types of node: inode, direct node,
 422 indirect node. F2FS assigns 4KB to an inode block which contains 923 data block
 423 indices, two direct node pointers, two indirect node pointers, and one double
 424 indirect node pointer as described below. One direct node block contains 1018
 425 data blocks, and one indirect node block contains also 1018 node blocks. Thus,
 426 one inode block (i.e., a file) covers:
 427
 428   4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB.
 429
 430    Inode block (4KB)
 431      |- data (923)
 432      |- direct node (2)
 433      |          `- data (1018)
 434      |- indirect node (2)
 435      |            `- direct node (1018)
 436      |                       `- data (1018)
 437      `- double indirect node (1)
 438                          `- indirect node (1018)
 439                                       `- direct node (1018)
 440                                                  `- data (1018)
 441
 442 Note that, all the node blocks are mapped by NAT which means the location of
 443 each node is translated by the NAT table. In the consideration of the wandering
 444 tree problem, F2FS is able to cut off the propagation of node updates caused by
 445 leaf data writes.
 446
 447 Directory Structure
 448 -------------------
 449
 450 A directory entry occupies 11 bytes, which consists of the following attributes.
 451
 452 - hash          hash value of the file name
 453 - ino           inode number
 454 - len           the length of file name
 455 - type          file type such as directory, symlink, etc
 456
 457 A dentry block consists of 214 dentry slots and file names. Therein a bitmap is
 458 used to represent whether each dentry is valid or not. A dentry block occupies
 459 4KB with the following composition.
 460
 461   Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) +
 462                       dentries(11 * 214 bytes) + file name (8 * 214 bytes)
 463
 464                          [Bucket]
 465              +--------------------------------+
 466              |dentry block 1 | dentry block 2 |
 467              +--------------------------------+
 468              .               .
 469        .                             .
 470   .       [Dentry Block Structure: 4KB]       .
 471   +--------+----------+----------+------------+
 472   | bitmap | reserved | dentries | file names |
 473   +--------+----------+----------+------------+
 474   [Dentry Block: 4KB] .   .
 475                  .               .
 476             .                          .
 477             +------+------+-----+------+
 478             | hash | ino  | len | type |
 479             +------+------+-----+------+
 480             [Dentry Structure: 11 bytes]
 481
 482 F2FS implements multi-level hash tables for directory structure. Each level has
 483 a hash table with dedicated number of hash buckets as shown below. Note that
 484 "A(2B)" means a bucket includes 2 data blocks.
 485
 486 ----------------------
 487 A : bucket
 488 B : block
 489 N : MAX_DIR_HASH_DEPTH
 490 ----------------------
 491
 492 level #0   | A(2B)
 493            |
 494 level #1   | A(2B) - A(2B)
 495            |
 496 level #2   | A(2B) - A(2B) - A(2B) - A(2B)
 497      .     |   .       .       .       .
 498 level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
 499      .     |   .       .       .       .
 500 level #N   | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
 501
 502 The number of blocks and buckets are determined by,
 503
 504                             ,- 2, if n < MAX_DIR_HASH_DEPTH / 2,
 505   # of blocks in level #n = |
 506                             `- 4, Otherwise
 507
 508                              ,- 2^(n + dir_level),
 509                              |        if n + dir_level < MAX_DIR_HASH_DEPTH / 2,
 510   # of buckets in level #n = |
 511                              `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1),
 512                                       Otherwise
 513
 514 When F2FS finds a file name in a directory, at first a hash value of the file
 515 name is calculated. Then, F2FS scans the hash table in level #0 to find the
 516 dentry consisting of the file name and its inode number. If not found, F2FS
 517 scans the next hash table in level #1. In this way, F2FS scans hash tables in
 518 each levels incrementally from 1 to N. In each levels F2FS needs to scan only
 519 one bucket determined by the following equation, which shows O(log(# of files))
 520 complexity.
 521
 522   bucket number to scan in level #n = (hash value) % (# of buckets in level #n)
 523
 524 In the case of file creation, F2FS finds empty consecutive slots that cover the
 525 file name. F2FS searches the empty slots in the hash tables of whole levels from
 526 1 to N in the same way as the lookup operation.
 527
 528 The following figure shows an example of two cases holding children.
 529        --------------> Dir <--------------
 530        |                                 |
 531     child                             child
 532
 533     child - child                     [hole] - child
 534
 535     child - child - child             [hole] - [hole] - child
 536
 537    Case 1:                           Case 2:
 538    Number of children = 6,           Number of children = 3,
 539    File size = 7                     File size = 7
 540
 541 Default Block Allocation
 542 ------------------------
 543
 544 At runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node
 545 and Hot/Warm/Cold data.
 546
 547 - Hot node      contains direct node blocks of directories.
 548 - Warm node     contains direct node blocks except hot node blocks.
 549 - Cold node     contains indirect node blocks
 550 - Hot data      contains dentry blocks
 551 - Warm data     contains data blocks except hot and cold data blocks
 552 - Cold data     contains multimedia data or migrated data blocks
 553
 554 LFS has two schemes for free space management: threaded log and copy-and-compac-
 555 tion. The copy-and-compaction scheme which is known as cleaning, is well-suited
 556 for devices showing very good sequential write performance, since free segments
 557 are served all the time for writing new data. However, it suffers from cleaning
 558 overhead under high utilization. Contrarily, the threaded log scheme suffers
 559 from random writes, but no cleaning process is needed. F2FS adopts a hybrid
 560 scheme where the copy-and-compaction scheme is adopted by default, but the
 561 policy is dynamically changed to the threaded log scheme according to the file
 562 system status.
 563
 564 In order to align F2FS with underlying flash-based storage, F2FS allocates a
 565 segment in a unit of section. F2FS expects that the section size would be the
 566 same as the unit size of garbage collection in FTL. Furthermore, with respect
 567 to the mapping granularity in FTL, F2FS allocates each section of the active
 568 logs from different zones as much as possible, since FTL can write the data in
 569 the active logs into one allocation unit according to its mapping granularity.
 570
 571 Cleaning process
 572 ----------------
 573
 574 F2FS does cleaning both on demand and in the background. On-demand cleaning is
 575 triggered when there are not enough free segments to serve VFS calls. Background
 576 cleaner is operated by a kernel thread, and triggers the cleaning job when the
 577 system is idle.
 578
 579 F2FS supports two victim selection policies: greedy and cost-benefit algorithms.
 580 In the greedy algorithm, F2FS selects a victim segment having the smallest number
 581 of valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment
 582 according to the segment age and the number of valid blocks in order to address
 583 log block thrashing problem in the greedy algorithm. F2FS adopts the greedy
 584 algorithm for on-demand cleaner, while background cleaner adopts cost-benefit
 585 algorithm.
 586
 587 In order to identify whether the data in the victim segment are valid or not,
 588 F2FS manages a bitmap. Each bit represents the validity of a block, and the
 589 bitmap is composed of a bit stream covering whole blocks in main area.