| blob | dae15c96e659e2bdaaf24cf760912cec855f079e |
1 .. SPDX-License-Identifier: GPL-2.0
3 ==========================================
4 WHAT IS Flash-Friendly File System (F2FS)?
5 ==========================================
7 NAND flash memory-based storage devices, such as SSD, eMMC, and SD cards, have
8 been equipped on a variety systems ranging from mobile to server systems. Since
9 they are known to have different characteristics from the conventional rotating
10 disks, a file system, an upper layer to the storage device, should adapt to the
11 changes from the sketch in the design level.
13 F2FS is a file system exploiting NAND flash memory-based storage devices, which
14 is based on Log-structured File System (LFS). The design has been focused on
15 addressing the fundamental issues in LFS, which are snowball effect of wandering
16 tree and high cleaning overhead.
18 Since a NAND flash memory-based storage device shows different characteristic
19 according to its internal geometry or flash memory management scheme, namely FTL,
20 F2FS and its tools support various parameters not only for configuring on-disk
21 layout, but also for selecting allocation and cleaning algorithms.
23 The following git tree provides the file system formatting tool (mkfs.f2fs),
24 a consistency checking tool (fsck.f2fs), and a debugging tool (dump.f2fs).
26 - git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git
28 For reporting bugs and sending patches, please use the following mailing list:
30 - linux-f2fs-devel@lists.sourceforge.net
32 Background and Design issues
33 ============================
35 Log-structured File System (LFS)
36 --------------------------------
37 "A log-structured file system writes all modifications to disk sequentially in
38 a log-like structure, thereby speeding up both file writing and crash recovery.
39 The log is the only structure on disk; it contains indexing information so that
40 files can be read back from the log efficiently. In order to maintain large free
41 areas on disk for fast writing, we divide the log into segments and use a
42 segment cleaner to compress the live information from heavily fragmented
43 segments." from Rosenblum, M. and Ousterhout, J. K., 1992, "The design and
44 implementation of a log-structured file system", ACM Trans. Computer Systems
45 10, 1, 26–52.
47 Wandering Tree Problem
48 ----------------------
49 In LFS, when a file data is updated and written to the end of log, its direct
50 pointer block is updated due to the changed location. Then the indirect pointer
51 block is also updated due to the direct pointer block update. In this manner,
52 the upper index structures such as inode, inode map, and checkpoint block are
53 also updated recursively. This problem is called as wandering tree problem [1],
54 and in order to enhance the performance, it should eliminate or relax the update
55 propagation as much as possible.
57 [1] Bityutskiy, A. 2005. JFFS3 design issues. http://www.linux-mtd.infradead.org/
59 Cleaning Overhead
60 -----------------
61 Since LFS is based on out-of-place writes, it produces so many obsolete blocks
62 scattered across the whole storage. In order to serve new empty log space, it
63 needs to reclaim these obsolete blocks seamlessly to users. This job is called
64 as a cleaning process.
66 The process consists of three operations as follows.
68 1. A victim segment is selected through referencing segment usage table.
69 2. It loads parent index structures of all the data in the victim identified by
70 segment summary blocks.
71 3. It checks the cross-reference between the data and its parent index structure.
72 4. It moves valid data selectively.
74 This cleaning job may cause unexpected long delays, so the most important goal
75 is to hide the latencies to users. And also definitely, it should reduce the
76 amount of valid data to be moved, and move them quickly as well.
78 Key Features
79 ============
81 Flash Awareness
82 ---------------
83 - Enlarge the random write area for better performance, but provide the high
84 spatial locality
85 - Align FS data structures to the operational units in FTL as best efforts
87 Wandering Tree Problem
88 ----------------------
89 - Use a term, “node”, that represents inodes as well as various pointer blocks
90 - Introduce Node Address Table (NAT) containing the locations of all the “node”
91 blocks; this will cut off the update propagation.
93 Cleaning Overhead
94 -----------------
95 - Support a background cleaning process
96 - Support greedy and cost-benefit algorithms for victim selection policies
97 - Support multi-head logs for static/dynamic hot and cold data separation
98 - Introduce adaptive logging for efficient block allocation
100 Mount Options
101 =============
104 ======================== ============================================================
105 background_gc=%s Turn on/off cleaning operations, namely garbage
106 collection, triggered in background when I/O subsystem is
107 idle. If background_gc=on, it will turn on the garbage
108 collection and if background_gc=off, garbage collection
109 will be turned off. If background_gc=sync, it will turn
110 on synchronous garbage collection running in background.
111 Default value for this option is on. So garbage
112 collection is on by default.
113 disable_roll_forward Disable the roll-forward recovery routine
114 norecovery Disable the roll-forward recovery routine, mounted read-
115 only (i.e., -o ro,disable_roll_forward)
116 discard/nodiscard Enable/disable real-time discard in f2fs, if discard is
117 enabled, f2fs will issue discard/TRIM commands when a
118 segment is cleaned.
119 no_heap Disable heap-style segment allocation which finds free
120 segments for data from the beginning of main area, while
121 for node from the end of main area.
122 nouser_xattr Disable Extended User Attributes. Note: xattr is enabled
123 by default if CONFIG_F2FS_FS_XATTR is selected.
124 noacl Disable POSIX Access Control List. Note: acl is enabled
125 by default if CONFIG_F2FS_FS_POSIX_ACL is selected.
126 active_logs=%u Support configuring the number of active logs. In the
127 current design, f2fs supports only 2, 4, and 6 logs.
128 Default number is 6.
129 disable_ext_identify Disable the extension list configured by mkfs, so f2fs
130 is not aware of cold files such as media files.
131 inline_xattr Enable the inline xattrs feature.
132 noinline_xattr Disable the inline xattrs feature.
133 inline_xattr_size=%u Support configuring inline xattr size, it depends on
134 flexible inline xattr feature.
135 inline_data Enable the inline data feature: Newly created small (<~3.4k)
136 files can be written into inode block.
137 inline_dentry Enable the inline dir feature: data in newly created
138 directory entries can be written into inode block. The
139 space of inode block which is used to store inline
140 dentries is limited to ~3.4k.
141 noinline_dentry Disable the inline dentry feature.
142 flush_merge Merge concurrent cache_flush commands as much as possible
143 to eliminate redundant command issues. If the underlying
144 device handles the cache_flush command relatively slowly,
145 recommend to enable this option.
146 nobarrier This option can be used if underlying storage guarantees
147 its cached data should be written to the novolatile area.
148 If this option is set, no cache_flush commands are issued
149 but f2fs still guarantees the write ordering of all the
150 data writes.
151 fastboot This option is used when a system wants to reduce mount
152 time as much as possible, even though normal performance
153 can be sacrificed.
154 extent_cache Enable an extent cache based on rb-tree, it can cache
155 as many as extent which map between contiguous logical
156 address and physical address per inode, resulting in
157 increasing the cache hit ratio. Set by default.
158 noextent_cache Disable an extent cache based on rb-tree explicitly, see
159 the above extent_cache mount option.
160 noinline_data Disable the inline data feature, inline data feature is
161 enabled by default.
162 data_flush Enable data flushing before checkpoint in order to
163 persist data of regular and symlink.
164 reserve_root=%d Support configuring reserved space which is used for
165 allocation from a privileged user with specified uid or
166 gid, unit: 4KB, the default limit is 0.2% of user blocks.
167 resuid=%d The user ID which may use the reserved blocks.
168 resgid=%d The group ID which may use the reserved blocks.
169 fault_injection=%d Enable fault injection in all supported types with
170 specified injection rate.
171 fault_type=%d Support configuring fault injection type, should be
172 enabled with fault_injection option, fault type value
173 is shown below, it supports single or combined type.
175 =================== ===========
176 Type_Name Type_Value
177 =================== ===========
178 FAULT_KMALLOC 0x000000001
179 FAULT_KVMALLOC 0x000000002
180 FAULT_PAGE_ALLOC 0x000000004
181 FAULT_PAGE_GET 0x000000008
182 FAULT_ALLOC_BIO 0x000000010
183 FAULT_ALLOC_NID 0x000000020
184 FAULT_ORPHAN 0x000000040
185 FAULT_BLOCK 0x000000080
186 FAULT_DIR_DEPTH 0x000000100
187 FAULT_EVICT_INODE 0x000000200
188 FAULT_TRUNCATE 0x000000400
189 FAULT_READ_IO 0x000000800
190 FAULT_CHECKPOINT 0x000001000
191 FAULT_DISCARD 0x000002000
192 FAULT_WRITE_IO 0x000004000
193 =================== ===========
194 mode=%s Control block allocation mode which supports "adaptive"
195 and "lfs". In "lfs" mode, there should be no random
196 writes towards main area.
197 io_bits=%u Set the bit size of write IO requests. It should be set
198 with "mode=lfs".
199 usrquota Enable plain user disk quota accounting.
200 grpquota Enable plain group disk quota accounting.
201 prjquota Enable plain project quota accounting.
202 usrjquota=<file> Appoint specified file and type during mount, so that quota
203 grpjquota=<file> information can be properly updated during recovery flow,
204 prjjquota=<file> <quota file>: must be in root directory;
205 jqfmt=<quota type> <quota type>: [vfsold,vfsv0,vfsv1].
206 offusrjquota Turn off user journalled quota.
207 offgrpjquota Turn off group journalled quota.
208 offprjjquota Turn off project journalled quota.
209 quota Enable plain user disk quota accounting.
210 noquota Disable all plain disk quota option.
211 whint_mode=%s Control which write hints are passed down to block
212 layer. This supports "off", "user-based", and
213 "fs-based". In "off" mode (default), f2fs does not pass
214 down hints. In "user-based" mode, f2fs tries to pass
215 down hints given by users. And in "fs-based" mode, f2fs
216 passes down hints with its policy.
217 alloc_mode=%s Adjust block allocation policy, which supports "reuse"
218 and "default".
219 fsync_mode=%s Control the policy of fsync. Currently supports "posix",
220 "strict", and "nobarrier". In "posix" mode, which is
221 default, fsync will follow POSIX semantics and does a
222 light operation to improve the filesystem performance.
223 In "strict" mode, fsync will be heavy and behaves in line
224 with xfs, ext4 and btrfs, where xfstest generic/342 will
225 pass, but the performance will regress. "nobarrier" is
226 based on "posix", but doesn't issue flush command for
227 non-atomic files likewise "nobarrier" mount option.
228 test_dummy_encryption
229 test_dummy_encryption=%s
230 Enable dummy encryption, which provides a fake fscrypt
231 context. The fake fscrypt context is used by xfstests.
232 The argument may be either "v1" or "v2", in order to
233 select the corresponding fscrypt policy version.
234 checkpoint=%s[:%u[%]] Set to "disable" to turn off checkpointing. Set to "enable"
235 to reenable checkpointing. Is enabled by default. While
236 disabled, any unmounting or unexpected shutdowns will cause
237 the filesystem contents to appear as they did when the
238 filesystem was mounted with that option.
239 While mounting with checkpoint=disabled, the filesystem must
240 run garbage collection to ensure that all available space can
241 be used. If this takes too much time, the mount may return
242 EAGAIN. You may optionally add a value to indicate how much
243 of the disk you would be willing to temporarily give up to
244 avoid additional garbage collection. This can be given as a
245 number of blocks, or as a percent. For instance, mounting
246 with checkpoint=disable:100% would always succeed, but it may
247 hide up to all remaining free space. The actual space that
248 would be unusable can be viewed at /sys/fs/f2fs/<disk>/unusable
249 This space is reclaimed once checkpoint=enable.
250 compress_algorithm=%s Control compress algorithm, currently f2fs supports "lzo",
251 "lz4", "zstd" and "lzo-rle" algorithm.
252 compress_log_size=%u Support configuring compress cluster size, the size will
253 be 4KB * (1 << %u), 16KB is minimum size, also it's
254 default size.
255 compress_extension=%s Support adding specified extension, so that f2fs can enable
256 compression on those corresponding files, e.g. if all files
257 with '.ext' has high compression rate, we can set the '.ext'
258 on compression extension list and enable compression on
259 these file by default rather than to enable it via ioctl.
260 For other files, we can still enable compression via ioctl.
261 Note that, there is one reserved special extension '*', it
262 can be set to enable compression for all files.
263 compress_chksum Support verifying chksum of raw data in compressed cluster.
264 compress_mode=%s Control file compression mode. This supports "fs" and "user"
265 modes. In "fs" mode (default), f2fs does automatic compression
266 on the compression enabled files. In "user" mode, f2fs disables
267 the automaic compression and gives the user discretion of
268 choosing the target file and the timing. The user can do manual
269 compression/decompression on the compression enabled files using
270 ioctls.
271 inlinecrypt When possible, encrypt/decrypt the contents of encrypted
272 files using the blk-crypto framework rather than
273 filesystem-layer encryption. This allows the use of
274 inline encryption hardware. The on-disk format is
275 unaffected. For more details, see
276 Documentation/block/inline-encryption.rst.
277 atgc Enable age-threshold garbage collection, it provides high
278 effectiveness and efficiency on background GC.
279 ======================== ============================================================
281 Debugfs Entries
282 ===============
284 /sys/kernel/debug/f2fs/ contains information about all the partitions mounted as
285 f2fs. Each file shows the whole f2fs information.
287 /sys/kernel/debug/f2fs/status includes:
289 - major file system information managed by f2fs currently
290 - average SIT information about whole segments
291 - current memory footprint consumed by f2fs.
293 Sysfs Entries
294 =============
296 Information about mounted f2fs file systems can be found in
297 /sys/fs/f2fs. Each mounted filesystem will have a directory in
298 /sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda).
299 The files in each per-device directory are shown in table below.
301 Files in /sys/fs/f2fs/<devname>
302 (see also Documentation/ABI/testing/sysfs-fs-f2fs)
304 Usage
305 =====
307 1. Download userland tools and compile them.
309 2. Skip, if f2fs was compiled statically inside kernel.
310 Otherwise, insert the f2fs.ko module::
312 # insmod f2fs.ko
314 3. Create a directory to use when mounting::
316 # mkdir /mnt/f2fs
318 4. Format the block device, and then mount as f2fs::
320 # mkfs.f2fs -l label /dev/block_device
321 # mount -t f2fs /dev/block_device /mnt/f2fs
323 mkfs.f2fs
324 ---------
325 The mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem,
326 which builds a basic on-disk layout.
328 The quick options consist of:
330 =============== ===========================================================
331 ``-l [label]`` Give a volume label, up to 512 unicode name.
332 ``-a [0 or 1]`` Split start location of each area for heap-based allocation.
334 1 is set by default, which performs this.
335 ``-o [int]`` Set overprovision ratio in percent over volume size.
337 5 is set by default.
338 ``-s [int]`` Set the number of segments per section.
340 1 is set by default.
341 ``-z [int]`` Set the number of sections per zone.
343 1 is set by default.
344 ``-e [str]`` Set basic extension list. e.g. "mp3,gif,mov"
345 ``-t [0 or 1]`` Disable discard command or not.
347 1 is set by default, which conducts discard.
348 =============== ===========================================================
350 Note: please refer to the manpage of mkfs.f2fs(8) to get full option list.
352 fsck.f2fs
353 ---------
354 The fsck.f2fs is a tool to check the consistency of an f2fs-formatted
355 partition, which examines whether the filesystem metadata and user-made data
356 are cross-referenced correctly or not.
357 Note that, initial version of the tool does not fix any inconsistency.
359 The quick options consist of::
361 -d debug level [default:0]
363 Note: please refer to the manpage of fsck.f2fs(8) to get full option list.
365 dump.f2fs
366 ---------
367 The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
368 file. Each file is dump_ssa and dump_sit.
370 The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
371 It shows on-disk inode information recognized by a given inode number, and is
372 able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
373 ./dump_sit respectively.
375 The options consist of::
377 -d debug level [default:0]
378 -i inode no (hex)
379 -s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
380 -a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
382 Examples::
384 # dump.f2fs -i [ino] /dev/sdx
385 # dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
386 # dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
388 Note: please refer to the manpage of dump.f2fs(8) to get full option list.
390 sload.f2fs
391 ----------
392 The sload.f2fs gives a way to insert files and directories in the exisiting disk
393 image. This tool is useful when building f2fs images given compiled files.
395 Note: please refer to the manpage of sload.f2fs(8) to get full option list.
397 resize.f2fs
398 -----------
399 The resize.f2fs lets a user resize the f2fs-formatted disk image, while preserving
400 all the files and directories stored in the image.
402 Note: please refer to the manpage of resize.f2fs(8) to get full option list.
404 defrag.f2fs
405 -----------
406 The defrag.f2fs can be used to defragment scattered written data as well as
407 filesystem metadata across the disk. This can improve the write speed by giving
408 more free consecutive space.
410 Note: please refer to the manpage of defrag.f2fs(8) to get full option list.
412 f2fs_io
413 -------
414 The f2fs_io is a simple tool to issue various filesystem APIs as well as
415 f2fs-specific ones, which is very useful for QA tests.
417 Note: please refer to the manpage of f2fs_io(8) to get full option list.
419 Design
420 ======
422 On-disk Layout
423 --------------
425 F2FS divides the whole volume into a number of segments, each of which is fixed
426 to 2MB in size. A section is composed of consecutive segments, and a zone
427 consists of a set of sections. By default, section and zone sizes are set to one
428 segment size identically, but users can easily modify the sizes by mkfs.
430 F2FS splits the entire volume into six areas, and all the areas except superblock
431 consist of multiple segments as described below::
433 align with the zone size <-|
434 |-> align with the segment size
435 _________________________________________________________________________
436 | | | Segment | Node | Segment | |
437 | Superblock | Checkpoint | Info. | Address | Summary | Main |
438 | (SB) | (CP) | Table (SIT) | Table (NAT) | Area (SSA) | |
439 |____________|_____2______|______N______|______N______|______N_____|__N___|
440 . .
441 . .
442 . .
443 ._________________________________________.
444 |_Segment_|_..._|_Segment_|_..._|_Segment_|
445 . .
446 ._________._________
447 |_section_|__...__|_
448 . .
449 .________.
450 |__zone__|
452 - Superblock (SB)
453 It is located at the beginning of the partition, and there exist two copies
454 to avoid file system crash. It contains basic partition information and some
455 default parameters of f2fs.
457 - Checkpoint (CP)
458 It contains file system information, bitmaps for valid NAT/SIT sets, orphan
459 inode lists, and summary entries of current active segments.
461 - Segment Information Table (SIT)
462 It contains segment information such as valid block count and bitmap for the
463 validity of all the blocks.
465 - Node Address Table (NAT)
466 It is composed of a block address table for all the node blocks stored in
467 Main area.
469 - Segment Summary Area (SSA)
470 It contains summary entries which contains the owner information of all the
471 data and node blocks stored in Main area.
473 - Main Area
474 It contains file and directory data including their indices.
476 In order to avoid misalignment between file system and flash-based storage, F2FS
477 aligns the start block address of CP with the segment size. Also, it aligns the
478 start block address of Main area with the zone size by reserving some segments
479 in SSA area.
481 Reference the following survey for additional technical details.
482 https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey
484 File System Metadata Structure
485 ------------------------------
487 F2FS adopts the checkpointing scheme to maintain file system consistency. At
488 mount time, F2FS first tries to find the last valid checkpoint data by scanning
489 CP area. In order to reduce the scanning time, F2FS uses only two copies of CP.
490 One of them always indicates the last valid data, which is called as shadow copy
491 mechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism.
493 For file system consistency, each CP points to which NAT and SIT copies are
494 valid, as shown as below::
496 +--------+----------+---------+
497 | CP | SIT | NAT |
498 +--------+----------+---------+
499 . . . .
500 . . . .
501 . . . .
502 +-------+-------+--------+--------+--------+--------+
503 | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 |
504 +-------+-------+--------+--------+--------+--------+
505 | ^ ^
506 | | |
507 `----------------------------------------'
509 Index Structure
510 ---------------
512 The key data structure to manage the data locations is a "node". Similar to
513 traditional file structures, F2FS has three types of node: inode, direct node,
514 indirect node. F2FS assigns 4KB to an inode block which contains 923 data block
515 indices, two direct node pointers, two indirect node pointers, and one double
516 indirect node pointer as described below. One direct node block contains 1018
517 data blocks, and one indirect node block contains also 1018 node blocks. Thus,
518 one inode block (i.e., a file) covers::
520 4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB.
522 Inode block (4KB)
523 |- data (923)
524 |- direct node (2)
525 | `- data (1018)
526 |- indirect node (2)
527 | `- direct node (1018)
528 | `- data (1018)
529 `- double indirect node (1)
530 `- indirect node (1018)
531 `- direct node (1018)
532 `- data (1018)
534 Note that all the node blocks are mapped by NAT which means the location of
535 each node is translated by the NAT table. In the consideration of the wandering
536 tree problem, F2FS is able to cut off the propagation of node updates caused by
537 leaf data writes.
539 Directory Structure
540 -------------------
542 A directory entry occupies 11 bytes, which consists of the following attributes.
544 - hash hash value of the file name
545 - ino inode number
546 - len the length of file name
547 - type file type such as directory, symlink, etc
549 A dentry block consists of 214 dentry slots and file names. Therein a bitmap is
550 used to represent whether each dentry is valid or not. A dentry block occupies
551 4KB with the following composition.
553 ::
555 Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) +
556 dentries(11 * 214 bytes) + file name (8 * 214 bytes)
558 [Bucket]
559 +--------------------------------+
560 |dentry block 1 | dentry block 2 |
561 +--------------------------------+
562 . .
563 . .
564 . [Dentry Block Structure: 4KB] .
565 +--------+----------+----------+------------+
566 | bitmap | reserved | dentries | file names |
567 +--------+----------+----------+------------+
568 [Dentry Block: 4KB] . .
569 . .
570 . .
571 +------+------+-----+------+
572 | hash | ino | len | type |
573 +------+------+-----+------+
574 [Dentry Structure: 11 bytes]
576 F2FS implements multi-level hash tables for directory structure. Each level has
577 a hash table with dedicated number of hash buckets as shown below. Note that
578 "A(2B)" means a bucket includes 2 data blocks.
580 ::
582 ----------------------
583 A : bucket
584 B : block
585 N : MAX_DIR_HASH_DEPTH
586 ----------------------
588 level #0 | A(2B)
589 |
590 level #1 | A(2B) - A(2B)
591 |
592 level #2 | A(2B) - A(2B) - A(2B) - A(2B)
593 . | . . . .
594 level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
595 . | . . . .
596 level #N | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
598 The number of blocks and buckets are determined by::
600 ,- 2, if n < MAX_DIR_HASH_DEPTH / 2,
601 # of blocks in level #n = |
602 `- 4, Otherwise
604 ,- 2^(n + dir_level),
605 | if n + dir_level < MAX_DIR_HASH_DEPTH / 2,
606 # of buckets in level #n = |
607 `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1),
608 Otherwise
610 When F2FS finds a file name in a directory, at first a hash value of the file
611 name is calculated. Then, F2FS scans the hash table in level #0 to find the
612 dentry consisting of the file name and its inode number. If not found, F2FS
613 scans the next hash table in level #1. In this way, F2FS scans hash tables in
614 each levels incrementally from 1 to N. In each level F2FS needs to scan only
615 one bucket determined by the following equation, which shows O(log(# of files))
616 complexity::
618 bucket number to scan in level #n = (hash value) % (# of buckets in level #n)
620 In the case of file creation, F2FS finds empty consecutive slots that cover the
621 file name. F2FS searches the empty slots in the hash tables of whole levels from
622 1 to N in the same way as the lookup operation.
624 The following figure shows an example of two cases holding children::
626 --------------> Dir <--------------
627 | |
628 child child
630 child - child [hole] - child
632 child - child - child [hole] - [hole] - child
634 Case 1: Case 2:
635 Number of children = 6, Number of children = 3,
636 File size = 7 File size = 7
638 Default Block Allocation
639 ------------------------
641 At runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node
642 and Hot/Warm/Cold data.
644 - Hot node contains direct node blocks of directories.
645 - Warm node contains direct node blocks except hot node blocks.
646 - Cold node contains indirect node blocks
647 - Hot data contains dentry blocks
648 - Warm data contains data blocks except hot and cold data blocks
649 - Cold data contains multimedia data or migrated data blocks
651 LFS has two schemes for free space management: threaded log and copy-and-compac-
652 tion. The copy-and-compaction scheme which is known as cleaning, is well-suited
653 for devices showing very good sequential write performance, since free segments
654 are served all the time for writing new data. However, it suffers from cleaning
655 overhead under high utilization. Contrarily, the threaded log scheme suffers
656 from random writes, but no cleaning process is needed. F2FS adopts a hybrid
657 scheme where the copy-and-compaction scheme is adopted by default, but the
658 policy is dynamically changed to the threaded log scheme according to the file
659 system status.
661 In order to align F2FS with underlying flash-based storage, F2FS allocates a
662 segment in a unit of section. F2FS expects that the section size would be the
663 same as the unit size of garbage collection in FTL. Furthermore, with respect
664 to the mapping granularity in FTL, F2FS allocates each section of the active
665 logs from different zones as much as possible, since FTL can write the data in
666 the active logs into one allocation unit according to its mapping granularity.
668 Cleaning process
669 ----------------
671 F2FS does cleaning both on demand and in the background. On-demand cleaning is
672 triggered when there are not enough free segments to serve VFS calls. Background
673 cleaner is operated by a kernel thread, and triggers the cleaning job when the
674 system is idle.
676 F2FS supports two victim selection policies: greedy and cost-benefit algorithms.
677 In the greedy algorithm, F2FS selects a victim segment having the smallest number
678 of valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment
679 according to the segment age and the number of valid blocks in order to address
680 log block thrashing problem in the greedy algorithm. F2FS adopts the greedy
681 algorithm for on-demand cleaner, while background cleaner adopts cost-benefit
682 algorithm.
684 In order to identify whether the data in the victim segment are valid or not,
685 F2FS manages a bitmap. Each bit represents the validity of a block, and the
686 bitmap is composed of a bit stream covering whole blocks in main area.
688 Write-hint Policy
689 -----------------
691 1) whint_mode=off. F2FS only passes down WRITE_LIFE_NOT_SET.
693 2) whint_mode=user-based. F2FS tries to pass down hints given by
694 users.
696 ===================== ======================== ===================
697 User F2FS Block
698 ===================== ======================== ===================
699 META WRITE_LIFE_NOT_SET
700 HOT_NODE "
701 WARM_NODE "
702 COLD_NODE "
703 ioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME
704 extension list " "
706 -- buffered io
707 WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME
708 WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT
709 WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET
710 WRITE_LIFE_NONE " "
711 WRITE_LIFE_MEDIUM " "
712 WRITE_LIFE_LONG " "
714 -- direct io
715 WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME
716 WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT
717 WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET
718 WRITE_LIFE_NONE " WRITE_LIFE_NONE
719 WRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM
720 WRITE_LIFE_LONG " WRITE_LIFE_LONG
721 ===================== ======================== ===================
723 3) whint_mode=fs-based. F2FS passes down hints with its policy.
725 ===================== ======================== ===================
726 User F2FS Block
727 ===================== ======================== ===================
728 META WRITE_LIFE_MEDIUM;
729 HOT_NODE WRITE_LIFE_NOT_SET
730 WARM_NODE "
731 COLD_NODE WRITE_LIFE_NONE
732 ioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME
733 extension list " "
735 -- buffered io
736 WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME
737 WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT
738 WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_LONG
739 WRITE_LIFE_NONE " "
740 WRITE_LIFE_MEDIUM " "
741 WRITE_LIFE_LONG " "
743 -- direct io
744 WRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME
745 WRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT
746 WRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET
747 WRITE_LIFE_NONE " WRITE_LIFE_NONE
748 WRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM
749 WRITE_LIFE_LONG " WRITE_LIFE_LONG
750 ===================== ======================== ===================
752 Fallocate(2) Policy
753 -------------------
755 The default policy follows the below POSIX rule.
757 Allocating disk space
758 The default operation (i.e., mode is zero) of fallocate() allocates
759 the disk space within the range specified by offset and len. The
760 file size (as reported by stat(2)) will be changed if offset+len is
761 greater than the file size. Any subregion within the range specified
762 by offset and len that did not contain data before the call will be
763 initialized to zero. This default behavior closely resembles the
764 behavior of the posix_fallocate(3) library function, and is intended
765 as a method of optimally implementing that function.
767 However, once F2FS receives ioctl(fd, F2FS_IOC_SET_PIN_FILE) in prior to
768 fallocate(fd, DEFAULT_MODE), it allocates on-disk block addressess having
769 zero or random data, which is useful to the below scenario where:
771 1. create(fd)
772 2. ioctl(fd, F2FS_IOC_SET_PIN_FILE)
773 3. fallocate(fd, 0, 0, size)
774 4. address = fibmap(fd, offset)
775 5. open(blkdev)
776 6. write(blkdev, address)
778 Compression implementation
779 --------------------------
781 - New term named cluster is defined as basic unit of compression, file can
782 be divided into multiple clusters logically. One cluster includes 4 << n
783 (n >= 0) logical pages, compression size is also cluster size, each of
784 cluster can be compressed or not.
786 - In cluster metadata layout, one special block address is used to indicate
787 a cluster is a compressed one or normal one; for compressed cluster, following
788 metadata maps cluster to [1, 4 << n - 1] physical blocks, in where f2fs
789 stores data including compress header and compressed data.
791 - In order to eliminate write amplification during overwrite, F2FS only
792 support compression on write-once file, data can be compressed only when
793 all logical blocks in cluster contain valid data and compress ratio of
794 cluster data is lower than specified threshold.
796 - To enable compression on regular inode, there are three ways:
798 * chattr +c file
799 * chattr +c dir; touch dir/file
800 * mount w/ -o compress_extension=ext; touch file.ext
802 Compress metadata layout::
804 [Dnode Structure]
805 +-----------------------------------------------+
806 | cluster 1 | cluster 2 | ......... | cluster N |
807 +-----------------------------------------------+
808 . . . .
809 . . . .
810 . Compressed Cluster . . Normal Cluster .
811 +----------+---------+---------+---------+ +---------+---------+---------+---------+
812 |compr flag| block 1 | block 2 | block 3 | | block 1 | block 2 | block 3 | block 4 |
813 +----------+---------+---------+---------+ +---------+---------+---------+---------+
814 . .
815 . .
816 . .
817 +-------------+-------------+----------+----------------------------+
818 | data length | data chksum | reserved | compressed data |
819 +-------------+-------------+----------+----------------------------+
821 Compression mode
822 --------------------------
824 f2fs supports "fs" and "user" compression modes with "compression_mode" mount option.
825 With this option, f2fs provides a choice to select the way how to compress the
826 compression enabled files (refer to "Compression implementation" section for how to
827 enable compression on a regular inode).
829 1) compress_mode=fs
830 This is the default option. f2fs does automatic compression in the writeback of the
831 compression enabled files.
833 2) compress_mode=user
834 This disables the automaic compression and gives the user discretion of choosing the
835 target file and the timing. The user can do manual compression/decompression on the
836 compression enabled files using F2FS_IOC_DECOMPRESS_FILE and F2FS_IOC_COMPRESS_FILE
837 ioctls like the below.
839 To decompress a file,
841 fd = open(filename, O_WRONLY, 0);
842 ret = ioctl(fd, F2FS_IOC_DECOMPRESS_FILE);
844 To compress a file,
846 fd = open(filename, O_WRONLY, 0);
847 ret = ioctl(fd, F2FS_IOC_COMPRESS_FILE);
849 NVMe Zoned Namespace devices
850 ----------------------------
852 - ZNS defines a per-zone capacity which can be equal or less than the
853 zone-size. Zone-capacity is the number of usable blocks in the zone.
854 F2FS checks if zone-capacity is less than zone-size, if it is, then any
855 segment which starts after the zone-capacity is marked as not-free in
856 the free segment bitmap at initial mount time. These segments are marked
857 as permanently used so they are not allocated for writes and
858 consequently are not needed to be garbage collected. In case the
859 zone-capacity is not aligned to default segment size(2MB), then a segment
860 can start before the zone-capacity and span across zone-capacity boundary.
861 Such spanning segments are also considered as usable segments. All blocks
862 past the zone-capacity are considered unusable in these segments.