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 truned off.
 106                        Default value for this option is on. So garbage
 107                        collection is on by default.
 108 disable_roll_forward   Disable the roll-forward recovery routine
 109 norecovery             Disable the roll-forward recovery routine, mounted read-
 110                        only (i.e., -o ro,disable_roll_forward)
 111 discard                Issue discard/TRIM commands when a segment is cleaned.
 112 no_heap                Disable heap-style segment allocation which finds free
 113                        segments for data from the beginning of main area, while
 114                        for node from the end of main area.
 115 nouser_xattr           Disable Extended User Attributes. Note: xattr is enabled
 116                        by default if CONFIG_F2FS_FS_XATTR is selected.
 117 noacl                  Disable POSIX Access Control List. Note: acl is enabled
 118                        by default if CONFIG_F2FS_FS_POSIX_ACL is selected.
 119 active_logs=%u         Support configuring the number of active logs. In the
 120                        current design, f2fs supports only 2, 4, and 6 logs.
 121                        Default number is 6.
 122 disable_ext_identify   Disable the extension list configured by mkfs, so f2fs
 123                        does not aware of cold files such as media files.
 124 inline_xattr           Enable the inline xattrs feature.
 125 inline_data            Enable the inline data feature: New created small(<~3.4k)
 126                        files can be written into inode block.
 127 inline_dentry          Enable the inline dir feature: data in new created
 128                        directory entries can be written into inode block. The
 129                        space of inode block which is used to store inline
 130                        dentries is limited to ~3.4k.
 131 flush_merge            Merge concurrent cache_flush commands as much as possible
 132                        to eliminate redundant command issues. If the underlying
 133                        device handles the cache_flush command relatively slowly,
 134                        recommend to enable this option.
 135 nobarrier              This option can be used if underlying storage guarantees
 136                        its cached data should be written to the novolatile area.
 137                        If this option is set, no cache_flush commands are issued
 138                        but f2fs still guarantees the write ordering of all the
 139                        data writes.
 140 fastboot               This option is used when a system wants to reduce mount
 141                        time as much as possible, even though normal performance
 142                        can be sacrificed.
 143
 144 ================================================================================
 145 DEBUGFS ENTRIES
 146 ================================================================================
 147
 148 /sys/kernel/debug/f2fs/ contains information about all the partitions mounted as
 149 f2fs. Each file shows the whole f2fs information.
 150
 151 /sys/kernel/debug/f2fs/status includes:
 152  - major file system information managed by f2fs currently
 153  - average SIT information about whole segments
 154  - current memory footprint consumed by f2fs.
 155
 156 ================================================================================
 157 SYSFS ENTRIES
 158 ================================================================================
 159
 160 Information about mounted f2f2 file systems can be found in
 161 /sys/fs/f2fs.  Each mounted filesystem will have a directory in
 162 /sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda).
 163 The files in each per-device directory are shown in table below.
 164
 165 Files in /sys/fs/f2fs/<devname>
 166 (see also Documentation/ABI/testing/sysfs-fs-f2fs)
 167 ..............................................................................
 168  File                         Content
 169
 170  gc_max_sleep_time            This tuning parameter controls the maximum sleep
 171                               time for the garbage collection thread. Time is
 172                               in milliseconds.
 173
 174  gc_min_sleep_time            This tuning parameter controls the minimum sleep
 175                               time for the garbage collection thread. Time is
 176                               in milliseconds.
 177
 178  gc_no_gc_sleep_time          This tuning parameter controls the default sleep
 179                               time for the garbage collection thread. Time is
 180                               in milliseconds.
 181
 182  gc_idle                      This parameter controls the selection of victim
 183                               policy for garbage collection. Setting gc_idle = 0
 184                               (default) will disable this option. Setting
 185                               gc_idle = 1 will select the Cost Benefit approach
 186                               & setting gc_idle = 2 will select the greedy aproach.
 187
 188  reclaim_segments             This parameter controls the number of prefree
 189                               segments to be reclaimed. If the number of prefree
 190                               segments is larger than the number of segments
 191                               in the proportion to the percentage over total
 192                               volume size, f2fs tries to conduct checkpoint to
 193                               reclaim the prefree segments to free segments.
 194                               By default, 5% over total # of segments.
 195
 196  max_small_discards           This parameter controls the number of discard
 197                               commands that consist small blocks less than 2MB.
 198                               The candidates to be discarded are cached until
 199                               checkpoint is triggered, and issued during the
 200                               checkpoint. By default, it is disabled with 0.
 201
 202  trim_sections                This parameter controls the number of sections
 203                               to be trimmed out in batch mode when FITRIM
 204                               conducts. 32 sections is set by default.
 205
 206  ipu_policy                   This parameter controls the policy of in-place
 207                               updates in f2fs. There are five policies:
 208                                0x01: F2FS_IPU_FORCE, 0x02: F2FS_IPU_SSR,
 209                                0x04: F2FS_IPU_UTIL,  0x08: F2FS_IPU_SSR_UTIL,
 210                                0x10: F2FS_IPU_FSYNC.
 211
 212  min_ipu_util                 This parameter controls the threshold to trigger
 213                               in-place-updates. The number indicates percentage
 214                               of the filesystem utilization, and used by
 215                               F2FS_IPU_UTIL and F2FS_IPU_SSR_UTIL policies.
 216
 217  min_fsync_blocks             This parameter controls the threshold to trigger
 218                               in-place-updates when F2FS_IPU_FSYNC mode is set.
 219                               The number indicates the number of dirty pages
 220                               when fsync needs to flush on its call path. If
 221                               the number is less than this value, it triggers
 222                               in-place-updates.
 223
 224  max_victim_search            This parameter controls the number of trials to
 225                               find a victim segment when conducting SSR and
 226                               cleaning operations. The default value is 4096
 227                               which covers 8GB block address range.
 228
 229  dir_level                    This parameter controls the directory level to
 230                               support large directory. If a directory has a
 231                               number of files, it can reduce the file lookup
 232                               latency by increasing this dir_level value.
 233                               Otherwise, it needs to decrease this value to
 234                               reduce the space overhead. The default value is 0.
 235
 236  ram_thresh                   This parameter controls the memory footprint used
 237                               by free nids and cached nat entries. By default,
 238                               10 is set, which indicates 10 MB / 1 GB RAM.
 239
 240 ================================================================================
 241 USAGE
 242 ================================================================================
 243
 244 1. Download userland tools and compile them.
 245
 246 2. Skip, if f2fs was compiled statically inside kernel.
 247    Otherwise, insert the f2fs.ko module.
 248  # insmod f2fs.ko
 249
 250 3. Create a directory trying to mount
 251  # mkdir /mnt/f2fs
 252
 253 4. Format the block device, and then mount as f2fs
 254  # mkfs.f2fs -l label /dev/block_device
 255  # mount -t f2fs /dev/block_device /mnt/f2fs
 256
 257 mkfs.f2fs
 258 ---------
 259 The mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem,
 260 which builds a basic on-disk layout.
 261
 262 The options consist of:
 263 -l [label]   : Give a volume label, up to 512 unicode name.
 264 -a [0 or 1]  : Split start location of each area for heap-based allocation.
 265                1 is set by default, which performs this.
 266 -o [int]     : Set overprovision ratio in percent over volume size.
 267                5 is set by default.
 268 -s [int]     : Set the number of segments per section.
 269                1 is set by default.
 270 -z [int]     : Set the number of sections per zone.
 271                1 is set by default.
 272 -e [str]     : Set basic extension list. e.g. "mp3,gif,mov"
 273 -t [0 or 1]  : Disable discard command or not.
 274                1 is set by default, which conducts discard.
 275
 276 fsck.f2fs
 277 ---------
 278 The fsck.f2fs is a tool to check the consistency of an f2fs-formatted
 279 partition, which examines whether the filesystem metadata and user-made data
 280 are cross-referenced correctly or not.
 281 Note that, initial version of the tool does not fix any inconsistency.
 282
 283 The options consist of:
 284   -d debug level [default:0]
 285
 286 dump.f2fs
 287 ---------
 288 The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
 289 file. Each file is dump_ssa and dump_sit.
 290
 291 The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
 292 It shows on-disk inode information reconized by a given inode number, and is
 293 able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
 294 ./dump_sit respectively.
 295
 296 The options consist of:
 297   -d debug level [default:0]
 298   -i inode no (hex)
 299   -s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
 300   -a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
 301
 302 Examples:
 303 # dump.f2fs -i [ino] /dev/sdx
 304 # dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
 305 # dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
 306
 307 ================================================================================
 308 DESIGN
 309 ================================================================================
 310
 311 On-disk Layout
 312 --------------
 313
 314 F2FS divides the whole volume into a number of segments, each of which is fixed
 315 to 2MB in size. A section is composed of consecutive segments, and a zone
 316 consists of a set of sections. By default, section and zone sizes are set to one
 317 segment size identically, but users can easily modify the sizes by mkfs.
 318
 319 F2FS splits the entire volume into six areas, and all the areas except superblock
 320 consists of multiple segments as described below.
 321
 322                                             align with the zone size <-|
 323                  |-> align with the segment size
 324      _________________________________________________________________________
 325     |            |            |   Segment   |    Node     |   Segment  |      |
 326     | Superblock | Checkpoint |    Info.    |   Address   |   Summary  | Main |
 327     |    (SB)    |   (CP)     | Table (SIT) | Table (NAT) | Area (SSA) |      |
 328     |____________|_____2______|______N______|______N______|______N_____|__N___|
 329                                                                        .      .
 330                                                              .                .
 331                                                  .                            .
 332                                     ._________________________________________.
 333                                     |_Segment_|_..._|_Segment_|_..._|_Segment_|
 334                                     .           .
 335                                     ._________._________
 336                                     |_section_|__...__|_
 337                                     .            .
 338                                     .________.
 339                                     |__zone__|
 340
 341 - Superblock (SB)
 342  : It is located at the beginning of the partition, and there exist two copies
 343    to avoid file system crash. It contains basic partition information and some
 344    default parameters of f2fs.
 345
 346 - Checkpoint (CP)
 347  : It contains file system information, bitmaps for valid NAT/SIT sets, orphan
 348    inode lists, and summary entries of current active segments.
 349
 350 - Segment Information Table (SIT)
 351  : It contains segment information such as valid block count and bitmap for the
 352    validity of all the blocks.
 353
 354 - Node Address Table (NAT)
 355  : It is composed of a block address table for all the node blocks stored in
 356    Main area.
 357
 358 - Segment Summary Area (SSA)
 359  : It contains summary entries which contains the owner information of all the
 360    data and node blocks stored in Main area.
 361
 362 - Main Area
 363  : It contains file and directory data including their indices.
 364
 365 In order to avoid misalignment between file system and flash-based storage, F2FS
 366 aligns the start block address of CP with the segment size. Also, it aligns the
 367 start block address of Main area with the zone size by reserving some segments
 368 in SSA area.
 369
 370 Reference the following survey for additional technical details.
 371 https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey
 372
 373 File System Metadata Structure
 374 ------------------------------
 375
 376 F2FS adopts the checkpointing scheme to maintain file system consistency. At
 377 mount time, F2FS first tries to find the last valid checkpoint data by scanning
 378 CP area. In order to reduce the scanning time, F2FS uses only two copies of CP.
 379 One of them always indicates the last valid data, which is called as shadow copy
 380 mechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism.
 381
 382 For file system consistency, each CP points to which NAT and SIT copies are
 383 valid, as shown as below.
 384
 385   +--------+----------+---------+
 386   |   CP   |    SIT   |   NAT   |
 387   +--------+----------+---------+
 388   .         .          .          .
 389   .            .              .              .
 390   .               .                 .                 .
 391   +-------+-------+--------+--------+--------+--------+
 392   | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 |
 393   +-------+-------+--------+--------+--------+--------+
 394      |             ^                          ^
 395      |             |                          |
 396      `----------------------------------------'
 397
 398 Index Structure
 399 ---------------
 400
 401 The key data structure to manage the data locations is a "node". Similar to
 402 traditional file structures, F2FS has three types of node: inode, direct node,
 403 indirect node. F2FS assigns 4KB to an inode block which contains 923 data block
 404 indices, two direct node pointers, two indirect node pointers, and one double
 405 indirect node pointer as described below. One direct node block contains 1018
 406 data blocks, and one indirect node block contains also 1018 node blocks. Thus,
 407 one inode block (i.e., a file) covers:
 408
 409   4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB.
 410
 411    Inode block (4KB)
 412      |- data (923)
 413      |- direct node (2)
 414      |          `- data (1018)
 415      |- indirect node (2)
 416      |            `- direct node (1018)
 417      |                       `- data (1018)
 418      `- double indirect node (1)
 419                          `- indirect node (1018)
 420                                       `- direct node (1018)
 421                                                  `- data (1018)
 422
 423 Note that, all the node blocks are mapped by NAT which means the location of
 424 each node is translated by the NAT table. In the consideration of the wandering
 425 tree problem, F2FS is able to cut off the propagation of node updates caused by
 426 leaf data writes.
 427
 428 Directory Structure
 429 -------------------
 430
 431 A directory entry occupies 11 bytes, which consists of the following attributes.
 432
 433 - hash          hash value of the file name
 434 - ino           inode number
 435 - len           the length of file name
 436 - type          file type such as directory, symlink, etc
 437
 438 A dentry block consists of 214 dentry slots and file names. Therein a bitmap is
 439 used to represent whether each dentry is valid or not. A dentry block occupies
 440 4KB with the following composition.
 441
 442   Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) +
 443                       dentries(11 * 214 bytes) + file name (8 * 214 bytes)
 444
 445                          [Bucket]
 446              +--------------------------------+
 447              |dentry block 1 | dentry block 2 |
 448              +--------------------------------+
 449              .               .
 450        .                             .
 451   .       [Dentry Block Structure: 4KB]       .
 452   +--------+----------+----------+------------+
 453   | bitmap | reserved | dentries | file names |
 454   +--------+----------+----------+------------+
 455   [Dentry Block: 4KB] .   .
 456                  .               .
 457             .                          .
 458             +------+------+-----+------+
 459             | hash | ino  | len | type |
 460             +------+------+-----+------+
 461             [Dentry Structure: 11 bytes]
 462
 463 F2FS implements multi-level hash tables for directory structure. Each level has
 464 a hash table with dedicated number of hash buckets as shown below. Note that
 465 "A(2B)" means a bucket includes 2 data blocks.
 466
 467 ----------------------
 468 A : bucket
 469 B : block
 470 N : MAX_DIR_HASH_DEPTH
 471 ----------------------
 472
 473 level #0   | A(2B)
 474            |
 475 level #1   | A(2B) - A(2B)
 476            |
 477 level #2   | A(2B) - A(2B) - A(2B) - A(2B)
 478      .     |   .       .       .       .
 479 level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
 480      .     |   .       .       .       .
 481 level #N   | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
 482
 483 The number of blocks and buckets are determined by,
 484
 485                             ,- 2, if n < MAX_DIR_HASH_DEPTH / 2,
 486   # of blocks in level #n = |
 487                             `- 4, Otherwise
 488
 489                              ,- 2^(n + dir_level),
 490                              |        if n + dir_level < MAX_DIR_HASH_DEPTH / 2,
 491   # of buckets in level #n = |
 492                              `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1),
 493                                       Otherwise
 494
 495 When F2FS finds a file name in a directory, at first a hash value of the file
 496 name is calculated. Then, F2FS scans the hash table in level #0 to find the
 497 dentry consisting of the file name and its inode number. If not found, F2FS
 498 scans the next hash table in level #1. In this way, F2FS scans hash tables in
 499 each levels incrementally from 1 to N. In each levels F2FS needs to scan only
 500 one bucket determined by the following equation, which shows O(log(# of files))
 501 complexity.
 502
 503   bucket number to scan in level #n = (hash value) % (# of buckets in level #n)
 504
 505 In the case of file creation, F2FS finds empty consecutive slots that cover the
 506 file name. F2FS searches the empty slots in the hash tables of whole levels from
 507 1 to N in the same way as the lookup operation.
 508
 509 The following figure shows an example of two cases holding children.
 510        --------------> Dir <--------------
 511        |                                 |
 512     child                             child
 513
 514     child - child                     [hole] - child
 515
 516     child - child - child             [hole] - [hole] - child
 517
 518    Case 1:                           Case 2:
 519    Number of children = 6,           Number of children = 3,
 520    File size = 7                     File size = 7
 521
 522 Default Block Allocation
 523 ------------------------
 524
 525 At runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node
 526 and Hot/Warm/Cold data.
 527
 528 - Hot node      contains direct node blocks of directories.
 529 - Warm node     contains direct node blocks except hot node blocks.
 530 - Cold node     contains indirect node blocks
 531 - Hot data      contains dentry blocks
 532 - Warm data     contains data blocks except hot and cold data blocks
 533 - Cold data     contains multimedia data or migrated data blocks
 534
 535 LFS has two schemes for free space management: threaded log and copy-and-compac-
 536 tion. The copy-and-compaction scheme which is known as cleaning, is well-suited
 537 for devices showing very good sequential write performance, since free segments
 538 are served all the time for writing new data. However, it suffers from cleaning
 539 overhead under high utilization. Contrarily, the threaded log scheme suffers
 540 from random writes, but no cleaning process is needed. F2FS adopts a hybrid
 541 scheme where the copy-and-compaction scheme is adopted by default, but the
 542 policy is dynamically changed to the threaded log scheme according to the file
 543 system status.
 544
 545 In order to align F2FS with underlying flash-based storage, F2FS allocates a
 546 segment in a unit of section. F2FS expects that the section size would be the
 547 same as the unit size of garbage collection in FTL. Furthermore, with respect
 548 to the mapping granularity in FTL, F2FS allocates each section of the active
 549 logs from different zones as much as possible, since FTL can write the data in
 550 the active logs into one allocation unit according to its mapping granularity.
 551
 552 Cleaning process
 553 ----------------
 554
 555 F2FS does cleaning both on demand and in the background. On-demand cleaning is
 556 triggered when there are not enough free segments to serve VFS calls. Background
 557 cleaner is operated by a kernel thread, and triggers the cleaning job when the
 558 system is idle.
 559
 560 F2FS supports two victim selection policies: greedy and cost-benefit algorithms.
 561 In the greedy algorithm, F2FS selects a victim segment having the smallest number
 562 of valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment
 563 according to the segment age and the number of valid blocks in order to address
 564 log block thrashing problem in the greedy algorithm. F2FS adopts the greedy
 565 algorithm for on-demand cleaner, while background cleaner adopts cost-benefit
 566 algorithm.
 567
 568 In order to identify whether the data in the victim segment are valid or not,
 569 F2FS manages a bitmap. Each bit represents the validity of a block, and the
 570 bitmap is composed of a bit stream covering whole blocks in main area.