• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1================================================================================
2WHAT IS Flash-Friendly File System (F2FS)?
3================================================================================
4
5NAND flash memory-based storage devices, such as SSD, eMMC, and SD cards, have
6been equipped on a variety systems ranging from mobile to server systems. Since
7they are known to have different characteristics from the conventional rotating
8disks, a file system, an upper layer to the storage device, should adapt to the
9changes from the sketch in the design level.
10
11F2FS is a file system exploiting NAND flash memory-based storage devices, which
12is based on Log-structured File System (LFS). The design has been focused on
13addressing the fundamental issues in LFS, which are snowball effect of wandering
14tree and high cleaning overhead.
15
16Since a NAND flash memory-based storage device shows different characteristic
17according to its internal geometry or flash memory management scheme, namely FTL,
18F2FS and its tools support various parameters not only for configuring on-disk
19layout, but also for selecting allocation and cleaning algorithms.
20
21The following git tree provides the file system formatting tool (mkfs.f2fs),
22a 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
25For reporting bugs and sending patches, please use the following mailing list:
26>> linux-f2fs-devel@lists.sourceforge.net
27
28================================================================================
29BACKGROUND AND DESIGN ISSUES
30================================================================================
31
32Log-structured File System (LFS)
33--------------------------------
34"A log-structured file system writes all modifications to disk sequentially in
35a log-like structure, thereby speeding up  both file writing and crash recovery.
36The log is the only structure on disk; it contains indexing information so that
37files can be read back from the log efficiently. In order to maintain large free
38areas on disk for fast writing, we divide  the log into segments and use a
39segment cleaner to compress the live information from heavily fragmented
40segments." from Rosenblum, M. and Ousterhout, J. K., 1992, "The design and
41implementation of a log-structured file system", ACM Trans. Computer Systems
4210, 1, 26–52.
43
44Wandering Tree Problem
45----------------------
46In LFS, when a file data is updated and written to the end of log, its direct
47pointer block is updated due to the changed location. Then the indirect pointer
48block is also updated due to the direct pointer block update. In this manner,
49the upper index structures such as inode, inode map, and checkpoint block are
50also updated recursively. This problem is called as wandering tree problem [1],
51and in order to enhance the performance, it should eliminate or relax the update
52propagation as much as possible.
53
54[1] Bityutskiy, A. 2005. JFFS3 design issues. http://www.linux-mtd.infradead.org/
55
56Cleaning Overhead
57-----------------
58Since LFS is based on out-of-place writes, it produces so many obsolete blocks
59scattered across the whole storage. In order to serve new empty log space, it
60needs to reclaim these obsolete blocks seamlessly to users. This job is called
61as a cleaning process.
62
63The process consists of three operations as follows.
641. A victim segment is selected through referencing segment usage table.
652. It loads parent index structures of all the data in the victim identified by
66   segment summary blocks.
673. It checks the cross-reference between the data and its parent index structure.
684. It moves valid data selectively.
69
70This cleaning job may cause unexpected long delays, so the most important goal
71is to hide the latencies to users. And also definitely, it should reduce the
72amount of valid data to be moved, and move them quickly as well.
73
74================================================================================
75KEY FEATURES
76================================================================================
77
78Flash 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
84Wandering 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
90Cleaning 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================================================================================
98MOUNT OPTIONS
99================================================================================
100
101background_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.
109disable_roll_forward   Disable the roll-forward recovery routine
110norecovery             Disable the roll-forward recovery routine, mounted read-
111                       only (i.e., -o ro,disable_roll_forward)
112discard/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.
115no_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.
118nouser_xattr           Disable Extended User Attributes. Note: xattr is enabled
119                       by default if CONFIG_F2FS_FS_XATTR is selected.
120noacl                  Disable POSIX Access Control List. Note: acl is enabled
121                       by default if CONFIG_F2FS_FS_POSIX_ACL is selected.
122active_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.
125disable_ext_identify   Disable the extension list configured by mkfs, so f2fs
126                       does not aware of cold files such as media files.
127inline_xattr           Enable the inline xattrs feature.
128inline_data            Enable the inline data feature: New created small(<~3.4k)
129                       files can be written into inode block.
130inline_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.
134noinline_dentry        Diable the inline dentry feature.
135flush_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.
139nobarrier              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.
144fastboot               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.
147extent_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.
151noextent_cache         Disable an extent cache based on rb-tree explicitly, see
152                       the above extent_cache mount option.
153noinline_data          Disable the inline data feature, inline data feature is
154                       enabled by default.
155data_flush             Enable data flushing before checkpoint in order to
156                       persist data of regular and symlink.
157mode=%s                Control block allocation mode which supports "adaptive"
158                       and "lfs". In "lfs" mode, there should be no random
159                       writes towards main area.
160io_bits=%u             Set the bit size of write IO requests. It should be set
161                       with "mode=lfs".
162usrquota               Enable plain user disk quota accounting.
163grpquota               Enable plain group disk quota accounting.
164prjquota               Enable plain project quota accounting.
165usrjquota=<file>       Appoint specified file and type during mount, so that quota
166grpjquota=<file>       information can be properly updated during recovery flow,
167prjjquota=<file>       <quota file>: must be in root directory;
168jqfmt=<quota type>     <quota type>: [vfsold,vfsv0,vfsv1].
169offusrjquota           Turn off user journelled quota.
170offgrpjquota           Turn off group journelled quota.
171offprjjquota           Turn off project journelled quota.
172quota                  Enable plain user disk quota accounting.
173noquota                Disable all plain disk quota option.
174whint_mode=%s          Control which write hints are passed down to block
175                       layer. This supports "off", "user-based", and
176                       "fs-based".  In "off" mode (default), f2fs does not pass
177                       down hints. In "user-based" mode, f2fs tries to pass
178                       down hints given by users. And in "fs-based" mode, f2fs
179                       passes down hints with its policy.
180alloc_mode=%s          Adjust block allocation policy, which supports "reuse"
181                       and "default".
182fsync_mode=%s          Control the policy of fsync. Currently supports "posix"
183                       and "strict". In "posix" mode, which is default, fsync
184                       will follow POSIX semantics and does a light operation
185                       to improve the filesystem performance. In "strict" mode,
186                       fsync will be heavy and behaves in line with xfs, ext4
187                       and btrfs, where xfstest generic/342 will pass, but the
188                       performance will regress.
189test_dummy_encryption  Enable dummy encryption, which provides a fake fscrypt
190                       context. The fake fscrypt context is used by xfstests.
191
192================================================================================
193DEBUGFS ENTRIES
194================================================================================
195
196/sys/kernel/debug/f2fs/ contains information about all the partitions mounted as
197f2fs. Each file shows the whole f2fs information.
198
199/sys/kernel/debug/f2fs/status includes:
200 - major file system information managed by f2fs currently
201 - average SIT information about whole segments
202 - current memory footprint consumed by f2fs.
203
204================================================================================
205SYSFS ENTRIES
206================================================================================
207
208Information about mounted f2f2 file systems can be found in
209/sys/fs/f2fs.  Each mounted filesystem will have a directory in
210/sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda).
211The files in each per-device directory are shown in table below.
212
213Files in /sys/fs/f2fs/<devname>
214(see also Documentation/ABI/testing/sysfs-fs-f2fs)
215..............................................................................
216 File                         Content
217
218 gc_max_sleep_time            This tuning parameter controls the maximum sleep
219                              time for the garbage collection thread. Time is
220                              in milliseconds.
221
222 gc_min_sleep_time            This tuning parameter controls the minimum sleep
223                              time for the garbage collection thread. Time is
224                              in milliseconds.
225
226 gc_no_gc_sleep_time          This tuning parameter controls the default sleep
227                              time for the garbage collection thread. Time is
228                              in milliseconds.
229
230 gc_idle                      This parameter controls the selection of victim
231                              policy for garbage collection. Setting gc_idle = 0
232                              (default) will disable this option. Setting
233                              gc_idle = 1 will select the Cost Benefit approach
234                              & setting gc_idle = 2 will select the greedy approach.
235
236 gc_urgent                    This parameter controls triggering background GCs
237                              urgently or not. Setting gc_urgent = 0 [default]
238                              makes back to default behavior, while if it is set
239                              to 1, background thread starts to do GC by given
240                              gc_urgent_sleep_time interval.
241
242 gc_urgent_sleep_time         This parameter controls sleep time for gc_urgent.
243                              500 ms is set by default. See above gc_urgent.
244
245 reclaim_segments             This parameter controls the number of prefree
246                              segments to be reclaimed. If the number of prefree
247			      segments is larger than the number of segments
248			      in the proportion to the percentage over total
249			      volume size, f2fs tries to conduct checkpoint to
250			      reclaim the prefree segments to free segments.
251			      By default, 5% over total # of segments.
252
253 max_small_discards	      This parameter controls the number of discard
254			      commands that consist small blocks less than 2MB.
255			      The candidates to be discarded are cached until
256			      checkpoint is triggered, and issued during the
257			      checkpoint. By default, it is disabled with 0.
258
259 trim_sections                This parameter controls the number of sections
260                              to be trimmed out in batch mode when FITRIM
261                              conducts. 32 sections is set by default.
262
263 ipu_policy                   This parameter controls the policy of in-place
264                              updates in f2fs. There are five policies:
265                               0x01: F2FS_IPU_FORCE, 0x02: F2FS_IPU_SSR,
266                               0x04: F2FS_IPU_UTIL,  0x08: F2FS_IPU_SSR_UTIL,
267                               0x10: F2FS_IPU_FSYNC.
268
269 min_ipu_util                 This parameter controls the threshold to trigger
270                              in-place-updates. The number indicates percentage
271                              of the filesystem utilization, and used by
272                              F2FS_IPU_UTIL and F2FS_IPU_SSR_UTIL policies.
273
274 min_fsync_blocks             This parameter controls the threshold to trigger
275                              in-place-updates when F2FS_IPU_FSYNC mode is set.
276			      The number indicates the number of dirty pages
277			      when fsync needs to flush on its call path. If
278			      the number is less than this value, it triggers
279			      in-place-updates.
280
281 max_victim_search	      This parameter controls the number of trials to
282			      find a victim segment when conducting SSR and
283			      cleaning operations. The default value is 4096
284			      which covers 8GB block address range.
285
286 dir_level                    This parameter controls the directory level to
287			      support large directory. If a directory has a
288			      number of files, it can reduce the file lookup
289			      latency by increasing this dir_level value.
290			      Otherwise, it needs to decrease this value to
291			      reduce the space overhead. The default value is 0.
292
293 ram_thresh                   This parameter controls the memory footprint used
294			      by free nids and cached nat entries. By default,
295			      10 is set, which indicates 10 MB / 1 GB RAM.
296
297================================================================================
298USAGE
299================================================================================
300
3011. Download userland tools and compile them.
302
3032. Skip, if f2fs was compiled statically inside kernel.
304   Otherwise, insert the f2fs.ko module.
305 # insmod f2fs.ko
306
3073. Create a directory trying to mount
308 # mkdir /mnt/f2fs
309
3104. Format the block device, and then mount as f2fs
311 # mkfs.f2fs -l label /dev/block_device
312 # mount -t f2fs /dev/block_device /mnt/f2fs
313
314mkfs.f2fs
315---------
316The mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem,
317which builds a basic on-disk layout.
318
319The options consist of:
320-l [label]   : Give a volume label, up to 512 unicode name.
321-a [0 or 1]  : Split start location of each area for heap-based allocation.
322               1 is set by default, which performs this.
323-o [int]     : Set overprovision ratio in percent over volume size.
324               5 is set by default.
325-s [int]     : Set the number of segments per section.
326               1 is set by default.
327-z [int]     : Set the number of sections per zone.
328               1 is set by default.
329-e [str]     : Set basic extension list. e.g. "mp3,gif,mov"
330-t [0 or 1]  : Disable discard command or not.
331               1 is set by default, which conducts discard.
332
333fsck.f2fs
334---------
335The fsck.f2fs is a tool to check the consistency of an f2fs-formatted
336partition, which examines whether the filesystem metadata and user-made data
337are cross-referenced correctly or not.
338Note that, initial version of the tool does not fix any inconsistency.
339
340The options consist of:
341  -d debug level [default:0]
342
343dump.f2fs
344---------
345The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
346file. Each file is dump_ssa and dump_sit.
347
348The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
349It shows on-disk inode information recognized by a given inode number, and is
350able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
351./dump_sit respectively.
352
353The options consist of:
354  -d debug level [default:0]
355  -i inode no (hex)
356  -s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
357  -a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
358
359Examples:
360# dump.f2fs -i [ino] /dev/sdx
361# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
362# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
363
364================================================================================
365DESIGN
366================================================================================
367
368On-disk Layout
369--------------
370
371F2FS divides the whole volume into a number of segments, each of which is fixed
372to 2MB in size. A section is composed of consecutive segments, and a zone
373consists of a set of sections. By default, section and zone sizes are set to one
374segment size identically, but users can easily modify the sizes by mkfs.
375
376F2FS splits the entire volume into six areas, and all the areas except superblock
377consists of multiple segments as described below.
378
379                                            align with the zone size <-|
380                 |-> align with the segment size
381     _________________________________________________________________________
382    |            |            |   Segment   |    Node     |   Segment  |      |
383    | Superblock | Checkpoint |    Info.    |   Address   |   Summary  | Main |
384    |    (SB)    |   (CP)     | Table (SIT) | Table (NAT) | Area (SSA) |      |
385    |____________|_____2______|______N______|______N______|______N_____|__N___|
386                                                                       .      .
387                                                             .                .
388                                                 .                            .
389                                    ._________________________________________.
390                                    |_Segment_|_..._|_Segment_|_..._|_Segment_|
391                                    .           .
392                                    ._________._________
393                                    |_section_|__...__|_
394                                    .            .
395		                    .________.
396	                            |__zone__|
397
398- Superblock (SB)
399 : It is located at the beginning of the partition, and there exist two copies
400   to avoid file system crash. It contains basic partition information and some
401   default parameters of f2fs.
402
403- Checkpoint (CP)
404 : It contains file system information, bitmaps for valid NAT/SIT sets, orphan
405   inode lists, and summary entries of current active segments.
406
407- Segment Information Table (SIT)
408 : It contains segment information such as valid block count and bitmap for the
409   validity of all the blocks.
410
411- Node Address Table (NAT)
412 : It is composed of a block address table for all the node blocks stored in
413   Main area.
414
415- Segment Summary Area (SSA)
416 : It contains summary entries which contains the owner information of all the
417   data and node blocks stored in Main area.
418
419- Main Area
420 : It contains file and directory data including their indices.
421
422In order to avoid misalignment between file system and flash-based storage, F2FS
423aligns the start block address of CP with the segment size. Also, it aligns the
424start block address of Main area with the zone size by reserving some segments
425in SSA area.
426
427Reference the following survey for additional technical details.
428https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey
429
430File System Metadata Structure
431------------------------------
432
433F2FS adopts the checkpointing scheme to maintain file system consistency. At
434mount time, F2FS first tries to find the last valid checkpoint data by scanning
435CP area. In order to reduce the scanning time, F2FS uses only two copies of CP.
436One of them always indicates the last valid data, which is called as shadow copy
437mechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism.
438
439For file system consistency, each CP points to which NAT and SIT copies are
440valid, as shown as below.
441
442  +--------+----------+---------+
443  |   CP   |    SIT   |   NAT   |
444  +--------+----------+---------+
445  .         .          .          .
446  .            .              .              .
447  .               .                 .                 .
448  +-------+-------+--------+--------+--------+--------+
449  | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 |
450  +-------+-------+--------+--------+--------+--------+
451     |             ^                          ^
452     |             |                          |
453     `----------------------------------------'
454
455Index Structure
456---------------
457
458The key data structure to manage the data locations is a "node". Similar to
459traditional file structures, F2FS has three types of node: inode, direct node,
460indirect node. F2FS assigns 4KB to an inode block which contains 923 data block
461indices, two direct node pointers, two indirect node pointers, and one double
462indirect node pointer as described below. One direct node block contains 1018
463data blocks, and one indirect node block contains also 1018 node blocks. Thus,
464one inode block (i.e., a file) covers:
465
466  4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB.
467
468   Inode block (4KB)
469     |- data (923)
470     |- direct node (2)
471     |          `- data (1018)
472     |- indirect node (2)
473     |            `- direct node (1018)
474     |                       `- data (1018)
475     `- double indirect node (1)
476                         `- indirect node (1018)
477			              `- direct node (1018)
478	                                         `- data (1018)
479
480Note that, all the node blocks are mapped by NAT which means the location of
481each node is translated by the NAT table. In the consideration of the wandering
482tree problem, F2FS is able to cut off the propagation of node updates caused by
483leaf data writes.
484
485Directory Structure
486-------------------
487
488A directory entry occupies 11 bytes, which consists of the following attributes.
489
490- hash		hash value of the file name
491- ino		inode number
492- len		the length of file name
493- type		file type such as directory, symlink, etc
494
495A dentry block consists of 214 dentry slots and file names. Therein a bitmap is
496used to represent whether each dentry is valid or not. A dentry block occupies
4974KB with the following composition.
498
499  Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) +
500	              dentries(11 * 214 bytes) + file name (8 * 214 bytes)
501
502                         [Bucket]
503             +--------------------------------+
504             |dentry block 1 | dentry block 2 |
505             +--------------------------------+
506             .               .
507       .                             .
508  .       [Dentry Block Structure: 4KB]       .
509  +--------+----------+----------+------------+
510  | bitmap | reserved | dentries | file names |
511  +--------+----------+----------+------------+
512  [Dentry Block: 4KB] .   .
513		 .               .
514            .                          .
515            +------+------+-----+------+
516            | hash | ino  | len | type |
517            +------+------+-----+------+
518            [Dentry Structure: 11 bytes]
519
520F2FS implements multi-level hash tables for directory structure. Each level has
521a hash table with dedicated number of hash buckets as shown below. Note that
522"A(2B)" means a bucket includes 2 data blocks.
523
524----------------------
525A : bucket
526B : block
527N : MAX_DIR_HASH_DEPTH
528----------------------
529
530level #0   | A(2B)
531           |
532level #1   | A(2B) - A(2B)
533           |
534level #2   | A(2B) - A(2B) - A(2B) - A(2B)
535     .     |   .       .       .       .
536level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
537     .     |   .       .       .       .
538level #N   | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
539
540The number of blocks and buckets are determined by,
541
542                            ,- 2, if n < MAX_DIR_HASH_DEPTH / 2,
543  # of blocks in level #n = |
544                            `- 4, Otherwise
545
546                             ,- 2^(n + dir_level),
547			     |        if n + dir_level < MAX_DIR_HASH_DEPTH / 2,
548  # of buckets in level #n = |
549                             `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1),
550			              Otherwise
551
552When F2FS finds a file name in a directory, at first a hash value of the file
553name is calculated. Then, F2FS scans the hash table in level #0 to find the
554dentry consisting of the file name and its inode number. If not found, F2FS
555scans the next hash table in level #1. In this way, F2FS scans hash tables in
556each levels incrementally from 1 to N. In each levels F2FS needs to scan only
557one bucket determined by the following equation, which shows O(log(# of files))
558complexity.
559
560  bucket number to scan in level #n = (hash value) % (# of buckets in level #n)
561
562In the case of file creation, F2FS finds empty consecutive slots that cover the
563file name. F2FS searches the empty slots in the hash tables of whole levels from
5641 to N in the same way as the lookup operation.
565
566The following figure shows an example of two cases holding children.
567       --------------> Dir <--------------
568       |                                 |
569    child                             child
570
571    child - child                     [hole] - child
572
573    child - child - child             [hole] - [hole] - child
574
575   Case 1:                           Case 2:
576   Number of children = 6,           Number of children = 3,
577   File size = 7                     File size = 7
578
579Default Block Allocation
580------------------------
581
582At runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node
583and Hot/Warm/Cold data.
584
585- Hot node	contains direct node blocks of directories.
586- Warm node	contains direct node blocks except hot node blocks.
587- Cold node	contains indirect node blocks
588- Hot data	contains dentry blocks
589- Warm data	contains data blocks except hot and cold data blocks
590- Cold data	contains multimedia data or migrated data blocks
591
592LFS has two schemes for free space management: threaded log and copy-and-compac-
593tion. The copy-and-compaction scheme which is known as cleaning, is well-suited
594for devices showing very good sequential write performance, since free segments
595are served all the time for writing new data. However, it suffers from cleaning
596overhead under high utilization. Contrarily, the threaded log scheme suffers
597from random writes, but no cleaning process is needed. F2FS adopts a hybrid
598scheme where the copy-and-compaction scheme is adopted by default, but the
599policy is dynamically changed to the threaded log scheme according to the file
600system status.
601
602In order to align F2FS with underlying flash-based storage, F2FS allocates a
603segment in a unit of section. F2FS expects that the section size would be the
604same as the unit size of garbage collection in FTL. Furthermore, with respect
605to the mapping granularity in FTL, F2FS allocates each section of the active
606logs from different zones as much as possible, since FTL can write the data in
607the active logs into one allocation unit according to its mapping granularity.
608
609Cleaning process
610----------------
611
612F2FS does cleaning both on demand and in the background. On-demand cleaning is
613triggered when there are not enough free segments to serve VFS calls. Background
614cleaner is operated by a kernel thread, and triggers the cleaning job when the
615system is idle.
616
617F2FS supports two victim selection policies: greedy and cost-benefit algorithms.
618In the greedy algorithm, F2FS selects a victim segment having the smallest number
619of valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment
620according to the segment age and the number of valid blocks in order to address
621log block thrashing problem in the greedy algorithm. F2FS adopts the greedy
622algorithm for on-demand cleaner, while background cleaner adopts cost-benefit
623algorithm.
624
625In order to identify whether the data in the victim segment are valid or not,
626F2FS manages a bitmap. Each bit represents the validity of a block, and the
627bitmap is composed of a bit stream covering whole blocks in main area.
628