1 ================================================================================
2 WHAT IS Flash-Friendly File System (F2FS)?
3 ================================================================================
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.
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.
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.
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
25 For reporting bugs and sending patches, please use the following mailing list:
26 >> linux-f2fs-devel@lists.sourceforge.net
28 ================================================================================
29 BACKGROUND AND DESIGN ISSUES
30 ================================================================================
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
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.
54 [1] Bityutskiy, A. 2005. JFFS3 design issues. http://www.linux-mtd.infradead.org/
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.
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.
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.
74 ================================================================================
76 ================================================================================
80 - Enlarge the random write area for better performance, but provide the high
82 - Align FS data structures to the operational units in FTL as best efforts
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.
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
97 ================================================================================
99 ================================================================================
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
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 discard Issue discard/TRIM commands when a segment is cleaned.
110 no_heap Disable heap-style segment allocation which finds free
111 segments for data from the beginning of main area, while
112 for node from the end of main area.
113 nouser_xattr Disable Extended User Attributes. Note: xattr is enabled
114 by default if CONFIG_F2FS_FS_XATTR is selected.
115 noacl Disable POSIX Access Control List. Note: acl is enabled
116 by default if CONFIG_F2FS_FS_POSIX_ACL is selected.
117 active_logs=%u Support configuring the number of active logs. In the
118 current design, f2fs supports only 2, 4, and 6 logs.
120 disable_ext_identify Disable the extension list configured by mkfs, so f2fs
121 does not aware of cold files such as media files.
122 inline_xattr Enable the inline xattrs feature.
124 ================================================================================
126 ================================================================================
128 /sys/kernel/debug/f2fs/ contains information about all the partitions mounted as
129 f2fs. Each file shows the whole f2fs information.
131 /sys/kernel/debug/f2fs/status includes:
132 - major file system information managed by f2fs currently
133 - average SIT information about whole segments
134 - current memory footprint consumed by f2fs.
136 ================================================================================
138 ================================================================================
140 Information about mounted f2f2 file systems can be found in
141 /sys/fs/f2fs. Each mounted filesystem will have a directory in
142 /sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda).
143 The files in each per-device directory are shown in table below.
145 Files in /sys/fs/f2fs/<devname>
146 (see also Documentation/ABI/testing/sysfs-fs-f2fs)
147 ..............................................................................
150 gc_max_sleep_time This tuning parameter controls the maximum sleep
151 time for the garbage collection thread. Time is
154 gc_min_sleep_time This tuning parameter controls the minimum sleep
155 time for the garbage collection thread. Time is
158 gc_no_gc_sleep_time This tuning parameter controls the default sleep
159 time for the garbage collection thread. Time is
162 gc_idle This parameter controls the selection of victim
163 policy for garbage collection. Setting gc_idle = 0
164 (default) will disable this option. Setting
165 gc_idle = 1 will select the Cost Benefit approach
166 & setting gc_idle = 2 will select the greedy aproach.
168 reclaim_segments This parameter controls the number of prefree
169 segments to be reclaimed. If the number of prefree
170 segments is larger than this number, f2fs tries to
171 conduct checkpoint to reclaim the prefree segments
172 to free segments. By default, 100 segments, 200MB.
174 ================================================================================
176 ================================================================================
178 1. Download userland tools and compile them.
180 2. Skip, if f2fs was compiled statically inside kernel.
181 Otherwise, insert the f2fs.ko module.
184 3. Create a directory trying to mount
187 4. Format the block device, and then mount as f2fs
188 # mkfs.f2fs -l label /dev/block_device
189 # mount -t f2fs /dev/block_device /mnt/f2fs
193 The mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem,
194 which builds a basic on-disk layout.
196 The options consist of:
197 -l [label] : Give a volume label, up to 512 unicode name.
198 -a [0 or 1] : Split start location of each area for heap-based allocation.
199 1 is set by default, which performs this.
200 -o [int] : Set overprovision ratio in percent over volume size.
202 -s [int] : Set the number of segments per section.
204 -z [int] : Set the number of sections per zone.
206 -e [str] : Set basic extension list. e.g. "mp3,gif,mov"
207 -t [0 or 1] : Disable discard command or not.
208 1 is set by default, which conducts discard.
212 The fsck.f2fs is a tool to check the consistency of an f2fs-formatted
213 partition, which examines whether the filesystem metadata and user-made data
214 are cross-referenced correctly or not.
215 Note that, initial version of the tool does not fix any inconsistency.
217 The options consist of:
218 -d debug level [default:0]
222 The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
223 file. Each file is dump_ssa and dump_sit.
225 The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
226 It shows on-disk inode information reconized by a given inode number, and is
227 able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
228 ./dump_sit respectively.
230 The options consist of:
231 -d debug level [default:0]
233 -s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
234 -a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
237 # dump.f2fs -i [ino] /dev/sdx
238 # dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
239 # dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
241 ================================================================================
243 ================================================================================
248 F2FS divides the whole volume into a number of segments, each of which is fixed
249 to 2MB in size. A section is composed of consecutive segments, and a zone
250 consists of a set of sections. By default, section and zone sizes are set to one
251 segment size identically, but users can easily modify the sizes by mkfs.
253 F2FS splits the entire volume into six areas, and all the areas except superblock
254 consists of multiple segments as described below.
256 align with the zone size <-|
257 |-> align with the segment size
258 _________________________________________________________________________
259 | | | Segment | Node | Segment | |
260 | Superblock | Checkpoint | Info. | Address | Summary | Main |
261 | (SB) | (CP) | Table (SIT) | Table (NAT) | Area (SSA) | |
262 |____________|_____2______|______N______|______N______|______N_____|__N___|
266 ._________________________________________.
267 |_Segment_|_..._|_Segment_|_..._|_Segment_|
276 : It is located at the beginning of the partition, and there exist two copies
277 to avoid file system crash. It contains basic partition information and some
278 default parameters of f2fs.
281 : It contains file system information, bitmaps for valid NAT/SIT sets, orphan
282 inode lists, and summary entries of current active segments.
284 - Segment Information Table (SIT)
285 : It contains segment information such as valid block count and bitmap for the
286 validity of all the blocks.
288 - Node Address Table (NAT)
289 : It is composed of a block address table for all the node blocks stored in
292 - Segment Summary Area (SSA)
293 : It contains summary entries which contains the owner information of all the
294 data and node blocks stored in Main area.
297 : It contains file and directory data including their indices.
299 In order to avoid misalignment between file system and flash-based storage, F2FS
300 aligns the start block address of CP with the segment size. Also, it aligns the
301 start block address of Main area with the zone size by reserving some segments
304 Reference the following survey for additional technical details.
305 https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey
307 File System Metadata Structure
308 ------------------------------
310 F2FS adopts the checkpointing scheme to maintain file system consistency. At
311 mount time, F2FS first tries to find the last valid checkpoint data by scanning
312 CP area. In order to reduce the scanning time, F2FS uses only two copies of CP.
313 One of them always indicates the last valid data, which is called as shadow copy
314 mechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism.
316 For file system consistency, each CP points to which NAT and SIT copies are
317 valid, as shown as below.
319 +--------+----------+---------+
321 +--------+----------+---------+
325 +-------+-------+--------+--------+--------+--------+
326 | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 |
327 +-------+-------+--------+--------+--------+--------+
330 `----------------------------------------'
335 The key data structure to manage the data locations is a "node". Similar to
336 traditional file structures, F2FS has three types of node: inode, direct node,
337 indirect node. F2FS assigns 4KB to an inode block which contains 923 data block
338 indices, two direct node pointers, two indirect node pointers, and one double
339 indirect node pointer as described below. One direct node block contains 1018
340 data blocks, and one indirect node block contains also 1018 node blocks. Thus,
341 one inode block (i.e., a file) covers:
343 4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB.
350 | `- direct node (1018)
352 `- double indirect node (1)
353 `- indirect node (1018)
354 `- direct node (1018)
357 Note that, all the node blocks are mapped by NAT which means the location of
358 each node is translated by the NAT table. In the consideration of the wandering
359 tree problem, F2FS is able to cut off the propagation of node updates caused by
365 A directory entry occupies 11 bytes, which consists of the following attributes.
367 - hash hash value of the file name
369 - len the length of file name
370 - type file type such as directory, symlink, etc
372 A dentry block consists of 214 dentry slots and file names. Therein a bitmap is
373 used to represent whether each dentry is valid or not. A dentry block occupies
374 4KB with the following composition.
376 Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) +
377 dentries(11 * 214 bytes) + file name (8 * 214 bytes)
380 +--------------------------------+
381 |dentry block 1 | dentry block 2 |
382 +--------------------------------+
385 . [Dentry Block Structure: 4KB] .
386 +--------+----------+----------+------------+
387 | bitmap | reserved | dentries | file names |
388 +--------+----------+----------+------------+
389 [Dentry Block: 4KB] . .
392 +------+------+-----+------+
393 | hash | ino | len | type |
394 +------+------+-----+------+
395 [Dentry Structure: 11 bytes]
397 F2FS implements multi-level hash tables for directory structure. Each level has
398 a hash table with dedicated number of hash buckets as shown below. Note that
399 "A(2B)" means a bucket includes 2 data blocks.
401 ----------------------
404 N : MAX_DIR_HASH_DEPTH
405 ----------------------
409 level #1 | A(2B) - A(2B)
411 level #2 | A(2B) - A(2B) - A(2B) - A(2B)
413 level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
415 level #N | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
417 The number of blocks and buckets are determined by,
419 ,- 2, if n < MAX_DIR_HASH_DEPTH / 2,
420 # of blocks in level #n = |
423 ,- 2^n, if n < MAX_DIR_HASH_DEPTH / 2,
424 # of buckets in level #n = |
425 `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1), Otherwise
427 When F2FS finds a file name in a directory, at first a hash value of the file
428 name is calculated. Then, F2FS scans the hash table in level #0 to find the
429 dentry consisting of the file name and its inode number. If not found, F2FS
430 scans the next hash table in level #1. In this way, F2FS scans hash tables in
431 each levels incrementally from 1 to N. In each levels F2FS needs to scan only
432 one bucket determined by the following equation, which shows O(log(# of files))
435 bucket number to scan in level #n = (hash value) % (# of buckets in level #n)
437 In the case of file creation, F2FS finds empty consecutive slots that cover the
438 file name. F2FS searches the empty slots in the hash tables of whole levels from
439 1 to N in the same way as the lookup operation.
441 The following figure shows an example of two cases holding children.
442 --------------> Dir <--------------
446 child - child [hole] - child
448 child - child - child [hole] - [hole] - child
451 Number of children = 6, Number of children = 3,
452 File size = 7 File size = 7
454 Default Block Allocation
455 ------------------------
457 At runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node
458 and Hot/Warm/Cold data.
460 - Hot node contains direct node blocks of directories.
461 - Warm node contains direct node blocks except hot node blocks.
462 - Cold node contains indirect node blocks
463 - Hot data contains dentry blocks
464 - Warm data contains data blocks except hot and cold data blocks
465 - Cold data contains multimedia data or migrated data blocks
467 LFS has two schemes for free space management: threaded log and copy-and-compac-
468 tion. The copy-and-compaction scheme which is known as cleaning, is well-suited
469 for devices showing very good sequential write performance, since free segments
470 are served all the time for writing new data. However, it suffers from cleaning
471 overhead under high utilization. Contrarily, the threaded log scheme suffers
472 from random writes, but no cleaning process is needed. F2FS adopts a hybrid
473 scheme where the copy-and-compaction scheme is adopted by default, but the
474 policy is dynamically changed to the threaded log scheme according to the file
477 In order to align F2FS with underlying flash-based storage, F2FS allocates a
478 segment in a unit of section. F2FS expects that the section size would be the
479 same as the unit size of garbage collection in FTL. Furthermore, with respect
480 to the mapping granularity in FTL, F2FS allocates each section of the active
481 logs from different zones as much as possible, since FTL can write the data in
482 the active logs into one allocation unit according to its mapping granularity.
487 F2FS does cleaning both on demand and in the background. On-demand cleaning is
488 triggered when there are not enough free segments to serve VFS calls. Background
489 cleaner is operated by a kernel thread, and triggers the cleaning job when the
492 F2FS supports two victim selection policies: greedy and cost-benefit algorithms.
493 In the greedy algorithm, F2FS selects a victim segment having the smallest number
494 of valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment
495 according to the segment age and the number of valid blocks in order to address
496 log block thrashing problem in the greedy algorithm. F2FS adopts the greedy
497 algorithm for on-demand cleaner, while background cleaner adopts cost-benefit
500 In order to identify whether the data in the victim segment are valid or not,
501 F2FS manages a bitmap. Each bit represents the validity of a block, and the
502 bitmap is composed of a bit stream covering whole blocks in main area.