189272ca1SMauro Carvalho Chehab.. SPDX-License-Identifier: GPL-2.0 289272ca1SMauro Carvalho Chehab 389272ca1SMauro Carvalho Chehab========================================== 489272ca1SMauro Carvalho ChehabWHAT IS Flash-Friendly File System (F2FS)? 589272ca1SMauro Carvalho Chehab========================================== 689272ca1SMauro Carvalho Chehab 789272ca1SMauro Carvalho ChehabNAND flash memory-based storage devices, such as SSD, eMMC, and SD cards, have 889272ca1SMauro Carvalho Chehabbeen equipped on a variety systems ranging from mobile to server systems. Since 989272ca1SMauro Carvalho Chehabthey are known to have different characteristics from the conventional rotating 1089272ca1SMauro Carvalho Chehabdisks, a file system, an upper layer to the storage device, should adapt to the 1189272ca1SMauro Carvalho Chehabchanges from the sketch in the design level. 1289272ca1SMauro Carvalho Chehab 1389272ca1SMauro Carvalho ChehabF2FS is a file system exploiting NAND flash memory-based storage devices, which 1489272ca1SMauro Carvalho Chehabis based on Log-structured File System (LFS). The design has been focused on 1589272ca1SMauro Carvalho Chehabaddressing the fundamental issues in LFS, which are snowball effect of wandering 1689272ca1SMauro Carvalho Chehabtree and high cleaning overhead. 1789272ca1SMauro Carvalho Chehab 1889272ca1SMauro Carvalho ChehabSince a NAND flash memory-based storage device shows different characteristic 1989272ca1SMauro Carvalho Chehabaccording to its internal geometry or flash memory management scheme, namely FTL, 2089272ca1SMauro Carvalho ChehabF2FS and its tools support various parameters not only for configuring on-disk 2189272ca1SMauro Carvalho Chehablayout, but also for selecting allocation and cleaning algorithms. 2289272ca1SMauro Carvalho Chehab 2389272ca1SMauro Carvalho ChehabThe following git tree provides the file system formatting tool (mkfs.f2fs), 2489272ca1SMauro Carvalho Chehaba consistency checking tool (fsck.f2fs), and a debugging tool (dump.f2fs). 2589272ca1SMauro Carvalho Chehab 2689272ca1SMauro Carvalho Chehab- git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs-tools.git 2789272ca1SMauro Carvalho Chehab 2889272ca1SMauro Carvalho ChehabFor reporting bugs and sending patches, please use the following mailing list: 2989272ca1SMauro Carvalho Chehab 3089272ca1SMauro Carvalho Chehab- linux-f2fs-devel@lists.sourceforge.net 3189272ca1SMauro Carvalho Chehab 3289272ca1SMauro Carvalho ChehabBackground and Design issues 3389272ca1SMauro Carvalho Chehab============================ 3489272ca1SMauro Carvalho Chehab 3589272ca1SMauro Carvalho ChehabLog-structured File System (LFS) 3689272ca1SMauro Carvalho Chehab-------------------------------- 3789272ca1SMauro Carvalho Chehab"A log-structured file system writes all modifications to disk sequentially in 3889272ca1SMauro Carvalho Chehaba log-like structure, thereby speeding up both file writing and crash recovery. 3989272ca1SMauro Carvalho ChehabThe log is the only structure on disk; it contains indexing information so that 4089272ca1SMauro Carvalho Chehabfiles can be read back from the log efficiently. In order to maintain large free 4189272ca1SMauro Carvalho Chehabareas on disk for fast writing, we divide the log into segments and use a 4289272ca1SMauro Carvalho Chehabsegment cleaner to compress the live information from heavily fragmented 4389272ca1SMauro Carvalho Chehabsegments." from Rosenblum, M. and Ousterhout, J. K., 1992, "The design and 4489272ca1SMauro Carvalho Chehabimplementation of a log-structured file system", ACM Trans. Computer Systems 4589272ca1SMauro Carvalho Chehab10, 1, 26–52. 4689272ca1SMauro Carvalho Chehab 4789272ca1SMauro Carvalho ChehabWandering Tree Problem 4889272ca1SMauro Carvalho Chehab---------------------- 4989272ca1SMauro Carvalho ChehabIn LFS, when a file data is updated and written to the end of log, its direct 5089272ca1SMauro Carvalho Chehabpointer block is updated due to the changed location. Then the indirect pointer 5189272ca1SMauro Carvalho Chehabblock is also updated due to the direct pointer block update. In this manner, 5289272ca1SMauro Carvalho Chehabthe upper index structures such as inode, inode map, and checkpoint block are 5389272ca1SMauro Carvalho Chehabalso updated recursively. This problem is called as wandering tree problem [1], 5489272ca1SMauro Carvalho Chehaband in order to enhance the performance, it should eliminate or relax the update 5589272ca1SMauro Carvalho Chehabpropagation as much as possible. 5689272ca1SMauro Carvalho Chehab 5789272ca1SMauro Carvalho Chehab[1] Bityutskiy, A. 2005. JFFS3 design issues. http://www.linux-mtd.infradead.org/ 5889272ca1SMauro Carvalho Chehab 5989272ca1SMauro Carvalho ChehabCleaning Overhead 6089272ca1SMauro Carvalho Chehab----------------- 6189272ca1SMauro Carvalho ChehabSince LFS is based on out-of-place writes, it produces so many obsolete blocks 6289272ca1SMauro Carvalho Chehabscattered across the whole storage. In order to serve new empty log space, it 6389272ca1SMauro Carvalho Chehabneeds to reclaim these obsolete blocks seamlessly to users. This job is called 6489272ca1SMauro Carvalho Chehabas a cleaning process. 6589272ca1SMauro Carvalho Chehab 6689272ca1SMauro Carvalho ChehabThe process consists of three operations as follows. 6789272ca1SMauro Carvalho Chehab 6889272ca1SMauro Carvalho Chehab1. A victim segment is selected through referencing segment usage table. 6989272ca1SMauro Carvalho Chehab2. It loads parent index structures of all the data in the victim identified by 7089272ca1SMauro Carvalho Chehab segment summary blocks. 7189272ca1SMauro Carvalho Chehab3. It checks the cross-reference between the data and its parent index structure. 7289272ca1SMauro Carvalho Chehab4. It moves valid data selectively. 7389272ca1SMauro Carvalho Chehab 7489272ca1SMauro Carvalho ChehabThis cleaning job may cause unexpected long delays, so the most important goal 7589272ca1SMauro Carvalho Chehabis to hide the latencies to users. And also definitely, it should reduce the 7689272ca1SMauro Carvalho Chehabamount of valid data to be moved, and move them quickly as well. 7789272ca1SMauro Carvalho Chehab 7889272ca1SMauro Carvalho ChehabKey Features 7989272ca1SMauro Carvalho Chehab============ 8089272ca1SMauro Carvalho Chehab 8189272ca1SMauro Carvalho ChehabFlash Awareness 8289272ca1SMauro Carvalho Chehab--------------- 8389272ca1SMauro Carvalho Chehab- Enlarge the random write area for better performance, but provide the high 8489272ca1SMauro Carvalho Chehab spatial locality 8589272ca1SMauro Carvalho Chehab- Align FS data structures to the operational units in FTL as best efforts 8689272ca1SMauro Carvalho Chehab 8789272ca1SMauro Carvalho ChehabWandering Tree Problem 8889272ca1SMauro Carvalho Chehab---------------------- 8989272ca1SMauro Carvalho Chehab- Use a term, “node”, that represents inodes as well as various pointer blocks 9089272ca1SMauro Carvalho Chehab- Introduce Node Address Table (NAT) containing the locations of all the “node” 9189272ca1SMauro Carvalho Chehab blocks; this will cut off the update propagation. 9289272ca1SMauro Carvalho Chehab 9389272ca1SMauro Carvalho ChehabCleaning Overhead 9489272ca1SMauro Carvalho Chehab----------------- 9589272ca1SMauro Carvalho Chehab- Support a background cleaning process 9689272ca1SMauro Carvalho Chehab- Support greedy and cost-benefit algorithms for victim selection policies 9789272ca1SMauro Carvalho Chehab- Support multi-head logs for static/dynamic hot and cold data separation 9889272ca1SMauro Carvalho Chehab- Introduce adaptive logging for efficient block allocation 9989272ca1SMauro Carvalho Chehab 10089272ca1SMauro Carvalho ChehabMount Options 10189272ca1SMauro Carvalho Chehab============= 10289272ca1SMauro Carvalho Chehab 10389272ca1SMauro Carvalho Chehab 1049aa1ccb4SJonathan Corbet======================== ============================================================ 10589272ca1SMauro Carvalho Chehabbackground_gc=%s Turn on/off cleaning operations, namely garbage 10689272ca1SMauro Carvalho Chehab collection, triggered in background when I/O subsystem is 10789272ca1SMauro Carvalho Chehab idle. If background_gc=on, it will turn on the garbage 10889272ca1SMauro Carvalho Chehab collection and if background_gc=off, garbage collection 10989272ca1SMauro Carvalho Chehab will be turned off. If background_gc=sync, it will turn 11089272ca1SMauro Carvalho Chehab on synchronous garbage collection running in background. 11189272ca1SMauro Carvalho Chehab Default value for this option is on. So garbage 11289272ca1SMauro Carvalho Chehab collection is on by default. 1135911d2d1SChao Yugc_merge When background_gc is on, this option can be enabled to 1145911d2d1SChao Yu let background GC thread to handle foreground GC requests, 1155911d2d1SChao Yu it can eliminate the sluggish issue caused by slow foreground 1165911d2d1SChao Yu GC operation when GC is triggered from a process with limited 1175911d2d1SChao Yu I/O and CPU resources. 1185911d2d1SChao Yunogc_merge Disable GC merge feature. 11989272ca1SMauro Carvalho Chehabdisable_roll_forward Disable the roll-forward recovery routine 12089272ca1SMauro Carvalho Chehabnorecovery Disable the roll-forward recovery routine, mounted read- 12189272ca1SMauro Carvalho Chehab only (i.e., -o ro,disable_roll_forward) 12289272ca1SMauro Carvalho Chehabdiscard/nodiscard Enable/disable real-time discard in f2fs, if discard is 12389272ca1SMauro Carvalho Chehab enabled, f2fs will issue discard/TRIM commands when a 12489272ca1SMauro Carvalho Chehab segment is cleaned. 12589272ca1SMauro Carvalho Chehabno_heap Disable heap-style segment allocation which finds free 12689272ca1SMauro Carvalho Chehab segments for data from the beginning of main area, while 12789272ca1SMauro Carvalho Chehab for node from the end of main area. 12889272ca1SMauro Carvalho Chehabnouser_xattr Disable Extended User Attributes. Note: xattr is enabled 12989272ca1SMauro Carvalho Chehab by default if CONFIG_F2FS_FS_XATTR is selected. 13089272ca1SMauro Carvalho Chehabnoacl Disable POSIX Access Control List. Note: acl is enabled 13189272ca1SMauro Carvalho Chehab by default if CONFIG_F2FS_FS_POSIX_ACL is selected. 13289272ca1SMauro Carvalho Chehabactive_logs=%u Support configuring the number of active logs. In the 13389272ca1SMauro Carvalho Chehab current design, f2fs supports only 2, 4, and 6 logs. 13489272ca1SMauro Carvalho Chehab Default number is 6. 13589272ca1SMauro Carvalho Chehabdisable_ext_identify Disable the extension list configured by mkfs, so f2fs 136ca313c82SRandy Dunlap is not aware of cold files such as media files. 13789272ca1SMauro Carvalho Chehabinline_xattr Enable the inline xattrs feature. 13889272ca1SMauro Carvalho Chehabnoinline_xattr Disable the inline xattrs feature. 13989272ca1SMauro Carvalho Chehabinline_xattr_size=%u Support configuring inline xattr size, it depends on 14089272ca1SMauro Carvalho Chehab flexible inline xattr feature. 141ca313c82SRandy Dunlapinline_data Enable the inline data feature: Newly created small (<~3.4k) 14289272ca1SMauro Carvalho Chehab files can be written into inode block. 143ca313c82SRandy Dunlapinline_dentry Enable the inline dir feature: data in newly created 14489272ca1SMauro Carvalho Chehab directory entries can be written into inode block. The 14589272ca1SMauro Carvalho Chehab space of inode block which is used to store inline 14689272ca1SMauro Carvalho Chehab dentries is limited to ~3.4k. 14789272ca1SMauro Carvalho Chehabnoinline_dentry Disable the inline dentry feature. 14889272ca1SMauro Carvalho Chehabflush_merge Merge concurrent cache_flush commands as much as possible 14989272ca1SMauro Carvalho Chehab to eliminate redundant command issues. If the underlying 15089272ca1SMauro Carvalho Chehab device handles the cache_flush command relatively slowly, 15189272ca1SMauro Carvalho Chehab recommend to enable this option. 15289272ca1SMauro Carvalho Chehabnobarrier This option can be used if underlying storage guarantees 15389272ca1SMauro Carvalho Chehab its cached data should be written to the novolatile area. 15489272ca1SMauro Carvalho Chehab If this option is set, no cache_flush commands are issued 15589272ca1SMauro Carvalho Chehab but f2fs still guarantees the write ordering of all the 15689272ca1SMauro Carvalho Chehab data writes. 15789272ca1SMauro Carvalho Chehabfastboot This option is used when a system wants to reduce mount 15889272ca1SMauro Carvalho Chehab time as much as possible, even though normal performance 15989272ca1SMauro Carvalho Chehab can be sacrificed. 16089272ca1SMauro Carvalho Chehabextent_cache Enable an extent cache based on rb-tree, it can cache 16189272ca1SMauro Carvalho Chehab as many as extent which map between contiguous logical 16289272ca1SMauro Carvalho Chehab address and physical address per inode, resulting in 16389272ca1SMauro Carvalho Chehab increasing the cache hit ratio. Set by default. 16489272ca1SMauro Carvalho Chehabnoextent_cache Disable an extent cache based on rb-tree explicitly, see 16589272ca1SMauro Carvalho Chehab the above extent_cache mount option. 16689272ca1SMauro Carvalho Chehabnoinline_data Disable the inline data feature, inline data feature is 16789272ca1SMauro Carvalho Chehab enabled by default. 16889272ca1SMauro Carvalho Chehabdata_flush Enable data flushing before checkpoint in order to 16989272ca1SMauro Carvalho Chehab persist data of regular and symlink. 17089272ca1SMauro Carvalho Chehabreserve_root=%d Support configuring reserved space which is used for 17189272ca1SMauro Carvalho Chehab allocation from a privileged user with specified uid or 17289272ca1SMauro Carvalho Chehab gid, unit: 4KB, the default limit is 0.2% of user blocks. 17389272ca1SMauro Carvalho Chehabresuid=%d The user ID which may use the reserved blocks. 17489272ca1SMauro Carvalho Chehabresgid=%d The group ID which may use the reserved blocks. 17589272ca1SMauro Carvalho Chehabfault_injection=%d Enable fault injection in all supported types with 17689272ca1SMauro Carvalho Chehab specified injection rate. 17789272ca1SMauro Carvalho Chehabfault_type=%d Support configuring fault injection type, should be 17889272ca1SMauro Carvalho Chehab enabled with fault_injection option, fault type value 17989272ca1SMauro Carvalho Chehab is shown below, it supports single or combined type. 18089272ca1SMauro Carvalho Chehab 18189272ca1SMauro Carvalho Chehab =================== =========== 18289272ca1SMauro Carvalho Chehab Type_Name Type_Value 18389272ca1SMauro Carvalho Chehab =================== =========== 18489272ca1SMauro Carvalho Chehab FAULT_KMALLOC 0x000000001 18589272ca1SMauro Carvalho Chehab FAULT_KVMALLOC 0x000000002 18689272ca1SMauro Carvalho Chehab FAULT_PAGE_ALLOC 0x000000004 18789272ca1SMauro Carvalho Chehab FAULT_PAGE_GET 0x000000008 18889272ca1SMauro Carvalho Chehab FAULT_ALLOC_NID 0x000000020 18989272ca1SMauro Carvalho Chehab FAULT_ORPHAN 0x000000040 19089272ca1SMauro Carvalho Chehab FAULT_BLOCK 0x000000080 19189272ca1SMauro Carvalho Chehab FAULT_DIR_DEPTH 0x000000100 19289272ca1SMauro Carvalho Chehab FAULT_EVICT_INODE 0x000000200 19389272ca1SMauro Carvalho Chehab FAULT_TRUNCATE 0x000000400 19489272ca1SMauro Carvalho Chehab FAULT_READ_IO 0x000000800 19589272ca1SMauro Carvalho Chehab FAULT_CHECKPOINT 0x000001000 19689272ca1SMauro Carvalho Chehab FAULT_DISCARD 0x000002000 19789272ca1SMauro Carvalho Chehab FAULT_WRITE_IO 0x000004000 19889272ca1SMauro Carvalho Chehab =================== =========== 19989272ca1SMauro Carvalho Chehabmode=%s Control block allocation mode which supports "adaptive" 20089272ca1SMauro Carvalho Chehab and "lfs". In "lfs" mode, there should be no random 20189272ca1SMauro Carvalho Chehab writes towards main area. 20289272ca1SMauro Carvalho Chehabio_bits=%u Set the bit size of write IO requests. It should be set 20389272ca1SMauro Carvalho Chehab with "mode=lfs". 20489272ca1SMauro Carvalho Chehabusrquota Enable plain user disk quota accounting. 20589272ca1SMauro Carvalho Chehabgrpquota Enable plain group disk quota accounting. 20689272ca1SMauro Carvalho Chehabprjquota Enable plain project quota accounting. 20789272ca1SMauro Carvalho Chehabusrjquota=<file> Appoint specified file and type during mount, so that quota 20889272ca1SMauro Carvalho Chehabgrpjquota=<file> information can be properly updated during recovery flow, 20989272ca1SMauro Carvalho Chehabprjjquota=<file> <quota file>: must be in root directory; 21089272ca1SMauro Carvalho Chehabjqfmt=<quota type> <quota type>: [vfsold,vfsv0,vfsv1]. 211ca313c82SRandy Dunlapoffusrjquota Turn off user journalled quota. 212ca313c82SRandy Dunlapoffgrpjquota Turn off group journalled quota. 213ca313c82SRandy Dunlapoffprjjquota Turn off project journalled quota. 21489272ca1SMauro Carvalho Chehabquota Enable plain user disk quota accounting. 21589272ca1SMauro Carvalho Chehabnoquota Disable all plain disk quota option. 21689272ca1SMauro Carvalho Chehabwhint_mode=%s Control which write hints are passed down to block 21789272ca1SMauro Carvalho Chehab layer. This supports "off", "user-based", and 21889272ca1SMauro Carvalho Chehab "fs-based". In "off" mode (default), f2fs does not pass 21989272ca1SMauro Carvalho Chehab down hints. In "user-based" mode, f2fs tries to pass 22089272ca1SMauro Carvalho Chehab down hints given by users. And in "fs-based" mode, f2fs 22189272ca1SMauro Carvalho Chehab passes down hints with its policy. 22289272ca1SMauro Carvalho Chehaballoc_mode=%s Adjust block allocation policy, which supports "reuse" 22389272ca1SMauro Carvalho Chehab and "default". 22489272ca1SMauro Carvalho Chehabfsync_mode=%s Control the policy of fsync. Currently supports "posix", 22589272ca1SMauro Carvalho Chehab "strict", and "nobarrier". In "posix" mode, which is 22689272ca1SMauro Carvalho Chehab default, fsync will follow POSIX semantics and does a 22789272ca1SMauro Carvalho Chehab light operation to improve the filesystem performance. 22889272ca1SMauro Carvalho Chehab In "strict" mode, fsync will be heavy and behaves in line 22989272ca1SMauro Carvalho Chehab with xfs, ext4 and btrfs, where xfstest generic/342 will 23089272ca1SMauro Carvalho Chehab pass, but the performance will regress. "nobarrier" is 23189272ca1SMauro Carvalho Chehab based on "posix", but doesn't issue flush command for 23289272ca1SMauro Carvalho Chehab non-atomic files likewise "nobarrier" mount option. 233ed318a6cSEric Biggerstest_dummy_encryption 234ed318a6cSEric Biggerstest_dummy_encryption=%s 235ed318a6cSEric Biggers Enable dummy encryption, which provides a fake fscrypt 23689272ca1SMauro Carvalho Chehab context. The fake fscrypt context is used by xfstests. 237ed318a6cSEric Biggers The argument may be either "v1" or "v2", in order to 238ed318a6cSEric Biggers select the corresponding fscrypt policy version. 23989272ca1SMauro Carvalho Chehabcheckpoint=%s[:%u[%]] Set to "disable" to turn off checkpointing. Set to "enable" 24089272ca1SMauro Carvalho Chehab to reenable checkpointing. Is enabled by default. While 24189272ca1SMauro Carvalho Chehab disabled, any unmounting or unexpected shutdowns will cause 24289272ca1SMauro Carvalho Chehab the filesystem contents to appear as they did when the 24389272ca1SMauro Carvalho Chehab filesystem was mounted with that option. 24489272ca1SMauro Carvalho Chehab While mounting with checkpoint=disabled, the filesystem must 24589272ca1SMauro Carvalho Chehab run garbage collection to ensure that all available space can 24689272ca1SMauro Carvalho Chehab be used. If this takes too much time, the mount may return 24789272ca1SMauro Carvalho Chehab EAGAIN. You may optionally add a value to indicate how much 24889272ca1SMauro Carvalho Chehab of the disk you would be willing to temporarily give up to 24989272ca1SMauro Carvalho Chehab avoid additional garbage collection. This can be given as a 25089272ca1SMauro Carvalho Chehab number of blocks, or as a percent. For instance, mounting 25189272ca1SMauro Carvalho Chehab with checkpoint=disable:100% would always succeed, but it may 25289272ca1SMauro Carvalho Chehab hide up to all remaining free space. The actual space that 25389272ca1SMauro Carvalho Chehab would be unusable can be viewed at /sys/fs/f2fs/<disk>/unusable 25489272ca1SMauro Carvalho Chehab This space is reclaimed once checkpoint=enable. 255261eeb9cSDaeho Jeongcheckpoint_merge When checkpoint is enabled, this can be used to create a kernel 256261eeb9cSDaeho Jeong daemon and make it to merge concurrent checkpoint requests as 257261eeb9cSDaeho Jeong much as possible to eliminate redundant checkpoint issues. Plus, 258261eeb9cSDaeho Jeong we can eliminate the sluggish issue caused by slow checkpoint 259261eeb9cSDaeho Jeong operation when the checkpoint is done in a process context in 260261eeb9cSDaeho Jeong a cgroup having low i/o budget and cpu shares. To make this 261261eeb9cSDaeho Jeong do better, we set the default i/o priority of the kernel daemon 262261eeb9cSDaeho Jeong to "3", to give one higher priority than other kernel threads. 263261eeb9cSDaeho Jeong This is the same way to give a I/O priority to the jbd2 264261eeb9cSDaeho Jeong journaling thread of ext4 filesystem. 265261eeb9cSDaeho Jeongnocheckpoint_merge Disable checkpoint merge feature. 266f40f31caSLinus Torvaldscompress_algorithm=%s Control compress algorithm, currently f2fs supports "lzo", 2676d92b201SChao Yu "lz4", "zstd" and "lzo-rle" algorithm. 2683fde13f8SChao Yucompress_algorithm=%s:%d Control compress algorithm and its compress level, now, only 2693fde13f8SChao Yu "lz4" and "zstd" support compress level config. 2703fde13f8SChao Yu algorithm level range 2713fde13f8SChao Yu lz4 3 - 16 2723fde13f8SChao Yu zstd 1 - 22 27389272ca1SMauro Carvalho Chehabcompress_log_size=%u Support configuring compress cluster size, the size will 27489272ca1SMauro Carvalho Chehab be 4KB * (1 << %u), 16KB is minimum size, also it's 27589272ca1SMauro Carvalho Chehab default size. 27689272ca1SMauro Carvalho Chehabcompress_extension=%s Support adding specified extension, so that f2fs can enable 27789272ca1SMauro Carvalho Chehab compression on those corresponding files, e.g. if all files 27889272ca1SMauro Carvalho Chehab with '.ext' has high compression rate, we can set the '.ext' 27989272ca1SMauro Carvalho Chehab on compression extension list and enable compression on 28089272ca1SMauro Carvalho Chehab these file by default rather than to enable it via ioctl. 28189272ca1SMauro Carvalho Chehab For other files, we can still enable compression via ioctl. 282fa679555SChao Yu Note that, there is one reserved special extension '*', it 283fa679555SChao Yu can be set to enable compression for all files. 284b28f047bSChao Yucompress_chksum Support verifying chksum of raw data in compressed cluster. 285602a16d5SDaeho Jeongcompress_mode=%s Control file compression mode. This supports "fs" and "user" 286602a16d5SDaeho Jeong modes. In "fs" mode (default), f2fs does automatic compression 287602a16d5SDaeho Jeong on the compression enabled files. In "user" mode, f2fs disables 288602a16d5SDaeho Jeong the automaic compression and gives the user discretion of 289602a16d5SDaeho Jeong choosing the target file and the timing. The user can do manual 290602a16d5SDaeho Jeong compression/decompression on the compression enabled files using 291602a16d5SDaeho Jeong ioctls. 2922324d50dSLinus Torvaldsinlinecrypt When possible, encrypt/decrypt the contents of encrypted 29327aacd28SSatya Tangirala files using the blk-crypto framework rather than 29427aacd28SSatya Tangirala filesystem-layer encryption. This allows the use of 29527aacd28SSatya Tangirala inline encryption hardware. The on-disk format is 29627aacd28SSatya Tangirala unaffected. For more details, see 29727aacd28SSatya Tangirala Documentation/block/inline-encryption.rst. 298093749e2SChao Yuatgc Enable age-threshold garbage collection, it provides high 299093749e2SChao Yu effectiveness and efficiency on background GC. 3009aa1ccb4SJonathan Corbet======================== ============================================================ 30189272ca1SMauro Carvalho Chehab 30289272ca1SMauro Carvalho ChehabDebugfs Entries 30389272ca1SMauro Carvalho Chehab=============== 30489272ca1SMauro Carvalho Chehab 30589272ca1SMauro Carvalho Chehab/sys/kernel/debug/f2fs/ contains information about all the partitions mounted as 30689272ca1SMauro Carvalho Chehabf2fs. Each file shows the whole f2fs information. 30789272ca1SMauro Carvalho Chehab 30889272ca1SMauro Carvalho Chehab/sys/kernel/debug/f2fs/status includes: 30989272ca1SMauro Carvalho Chehab 31089272ca1SMauro Carvalho Chehab - major file system information managed by f2fs currently 31189272ca1SMauro Carvalho Chehab - average SIT information about whole segments 31289272ca1SMauro Carvalho Chehab - current memory footprint consumed by f2fs. 31389272ca1SMauro Carvalho Chehab 31489272ca1SMauro Carvalho ChehabSysfs Entries 31589272ca1SMauro Carvalho Chehab============= 31689272ca1SMauro Carvalho Chehab 31789272ca1SMauro Carvalho ChehabInformation about mounted f2fs file systems can be found in 31889272ca1SMauro Carvalho Chehab/sys/fs/f2fs. Each mounted filesystem will have a directory in 31989272ca1SMauro Carvalho Chehab/sys/fs/f2fs based on its device name (i.e., /sys/fs/f2fs/sda). 32089272ca1SMauro Carvalho ChehabThe files in each per-device directory are shown in table below. 32189272ca1SMauro Carvalho Chehab 32289272ca1SMauro Carvalho ChehabFiles in /sys/fs/f2fs/<devname> 32389272ca1SMauro Carvalho Chehab(see also Documentation/ABI/testing/sysfs-fs-f2fs) 32489272ca1SMauro Carvalho Chehab 32589272ca1SMauro Carvalho ChehabUsage 32689272ca1SMauro Carvalho Chehab===== 32789272ca1SMauro Carvalho Chehab 32889272ca1SMauro Carvalho Chehab1. Download userland tools and compile them. 32989272ca1SMauro Carvalho Chehab 33089272ca1SMauro Carvalho Chehab2. Skip, if f2fs was compiled statically inside kernel. 33189272ca1SMauro Carvalho Chehab Otherwise, insert the f2fs.ko module:: 33289272ca1SMauro Carvalho Chehab 33389272ca1SMauro Carvalho Chehab # insmod f2fs.ko 33489272ca1SMauro Carvalho Chehab 335ca313c82SRandy Dunlap3. Create a directory to use when mounting:: 33689272ca1SMauro Carvalho Chehab 33789272ca1SMauro Carvalho Chehab # mkdir /mnt/f2fs 33889272ca1SMauro Carvalho Chehab 33989272ca1SMauro Carvalho Chehab4. Format the block device, and then mount as f2fs:: 34089272ca1SMauro Carvalho Chehab 34189272ca1SMauro Carvalho Chehab # mkfs.f2fs -l label /dev/block_device 34289272ca1SMauro Carvalho Chehab # mount -t f2fs /dev/block_device /mnt/f2fs 34389272ca1SMauro Carvalho Chehab 34489272ca1SMauro Carvalho Chehabmkfs.f2fs 34589272ca1SMauro Carvalho Chehab--------- 34689272ca1SMauro Carvalho ChehabThe mkfs.f2fs is for the use of formatting a partition as the f2fs filesystem, 34789272ca1SMauro Carvalho Chehabwhich builds a basic on-disk layout. 34889272ca1SMauro Carvalho Chehab 349568d2a1eSJaegeuk KimThe quick options consist of: 35089272ca1SMauro Carvalho Chehab 35189272ca1SMauro Carvalho Chehab=============== =========================================================== 35289272ca1SMauro Carvalho Chehab``-l [label]`` Give a volume label, up to 512 unicode name. 35389272ca1SMauro Carvalho Chehab``-a [0 or 1]`` Split start location of each area for heap-based allocation. 35489272ca1SMauro Carvalho Chehab 35589272ca1SMauro Carvalho Chehab 1 is set by default, which performs this. 35689272ca1SMauro Carvalho Chehab``-o [int]`` Set overprovision ratio in percent over volume size. 35789272ca1SMauro Carvalho Chehab 35889272ca1SMauro Carvalho Chehab 5 is set by default. 35989272ca1SMauro Carvalho Chehab``-s [int]`` Set the number of segments per section. 36089272ca1SMauro Carvalho Chehab 36189272ca1SMauro Carvalho Chehab 1 is set by default. 36289272ca1SMauro Carvalho Chehab``-z [int]`` Set the number of sections per zone. 36389272ca1SMauro Carvalho Chehab 36489272ca1SMauro Carvalho Chehab 1 is set by default. 36589272ca1SMauro Carvalho Chehab``-e [str]`` Set basic extension list. e.g. "mp3,gif,mov" 36689272ca1SMauro Carvalho Chehab``-t [0 or 1]`` Disable discard command or not. 36789272ca1SMauro Carvalho Chehab 36889272ca1SMauro Carvalho Chehab 1 is set by default, which conducts discard. 36989272ca1SMauro Carvalho Chehab=============== =========================================================== 37089272ca1SMauro Carvalho Chehab 371ca313c82SRandy DunlapNote: please refer to the manpage of mkfs.f2fs(8) to get full option list. 372568d2a1eSJaegeuk Kim 37389272ca1SMauro Carvalho Chehabfsck.f2fs 37489272ca1SMauro Carvalho Chehab--------- 37589272ca1SMauro Carvalho ChehabThe fsck.f2fs is a tool to check the consistency of an f2fs-formatted 37689272ca1SMauro Carvalho Chehabpartition, which examines whether the filesystem metadata and user-made data 37789272ca1SMauro Carvalho Chehabare cross-referenced correctly or not. 37889272ca1SMauro Carvalho ChehabNote that, initial version of the tool does not fix any inconsistency. 37989272ca1SMauro Carvalho Chehab 380568d2a1eSJaegeuk KimThe quick options consist of:: 38189272ca1SMauro Carvalho Chehab 38289272ca1SMauro Carvalho Chehab -d debug level [default:0] 38389272ca1SMauro Carvalho Chehab 384ca313c82SRandy DunlapNote: please refer to the manpage of fsck.f2fs(8) to get full option list. 385568d2a1eSJaegeuk Kim 38689272ca1SMauro Carvalho Chehabdump.f2fs 38789272ca1SMauro Carvalho Chehab--------- 38889272ca1SMauro Carvalho ChehabThe dump.f2fs shows the information of specific inode and dumps SSA and SIT to 38989272ca1SMauro Carvalho Chehabfile. Each file is dump_ssa and dump_sit. 39089272ca1SMauro Carvalho Chehab 39189272ca1SMauro Carvalho ChehabThe dump.f2fs is used to debug on-disk data structures of the f2fs filesystem. 39289272ca1SMauro Carvalho ChehabIt shows on-disk inode information recognized by a given inode number, and is 39389272ca1SMauro Carvalho Chehabable to dump all the SSA and SIT entries into predefined files, ./dump_ssa and 39489272ca1SMauro Carvalho Chehab./dump_sit respectively. 39589272ca1SMauro Carvalho Chehab 39689272ca1SMauro Carvalho ChehabThe options consist of:: 39789272ca1SMauro Carvalho Chehab 39889272ca1SMauro Carvalho Chehab -d debug level [default:0] 39989272ca1SMauro Carvalho Chehab -i inode no (hex) 40089272ca1SMauro Carvalho Chehab -s [SIT dump segno from #1~#2 (decimal), for all 0~-1] 40189272ca1SMauro Carvalho Chehab -a [SSA dump segno from #1~#2 (decimal), for all 0~-1] 40289272ca1SMauro Carvalho Chehab 40389272ca1SMauro Carvalho ChehabExamples:: 40489272ca1SMauro Carvalho Chehab 40589272ca1SMauro Carvalho Chehab # dump.f2fs -i [ino] /dev/sdx 40689272ca1SMauro Carvalho Chehab # dump.f2fs -s 0~-1 /dev/sdx (SIT dump) 40789272ca1SMauro Carvalho Chehab # dump.f2fs -a 0~-1 /dev/sdx (SSA dump) 40889272ca1SMauro Carvalho Chehab 409ca313c82SRandy DunlapNote: please refer to the manpage of dump.f2fs(8) to get full option list. 410568d2a1eSJaegeuk Kim 411568d2a1eSJaegeuk Kimsload.f2fs 412568d2a1eSJaegeuk Kim---------- 413568d2a1eSJaegeuk KimThe sload.f2fs gives a way to insert files and directories in the exisiting disk 414568d2a1eSJaegeuk Kimimage. This tool is useful when building f2fs images given compiled files. 415568d2a1eSJaegeuk Kim 416ca313c82SRandy DunlapNote: please refer to the manpage of sload.f2fs(8) to get full option list. 417568d2a1eSJaegeuk Kim 418568d2a1eSJaegeuk Kimresize.f2fs 419568d2a1eSJaegeuk Kim----------- 420ca313c82SRandy DunlapThe resize.f2fs lets a user resize the f2fs-formatted disk image, while preserving 421568d2a1eSJaegeuk Kimall the files and directories stored in the image. 422568d2a1eSJaegeuk Kim 423ca313c82SRandy DunlapNote: please refer to the manpage of resize.f2fs(8) to get full option list. 424568d2a1eSJaegeuk Kim 425568d2a1eSJaegeuk Kimdefrag.f2fs 426568d2a1eSJaegeuk Kim----------- 427ca313c82SRandy DunlapThe defrag.f2fs can be used to defragment scattered written data as well as 428568d2a1eSJaegeuk Kimfilesystem metadata across the disk. This can improve the write speed by giving 429568d2a1eSJaegeuk Kimmore free consecutive space. 430568d2a1eSJaegeuk Kim 431ca313c82SRandy DunlapNote: please refer to the manpage of defrag.f2fs(8) to get full option list. 432568d2a1eSJaegeuk Kim 433568d2a1eSJaegeuk Kimf2fs_io 434568d2a1eSJaegeuk Kim------- 435568d2a1eSJaegeuk KimThe f2fs_io is a simple tool to issue various filesystem APIs as well as 436568d2a1eSJaegeuk Kimf2fs-specific ones, which is very useful for QA tests. 437568d2a1eSJaegeuk Kim 438ca313c82SRandy DunlapNote: please refer to the manpage of f2fs_io(8) to get full option list. 439568d2a1eSJaegeuk Kim 44089272ca1SMauro Carvalho ChehabDesign 44189272ca1SMauro Carvalho Chehab====== 44289272ca1SMauro Carvalho Chehab 44389272ca1SMauro Carvalho ChehabOn-disk Layout 44489272ca1SMauro Carvalho Chehab-------------- 44589272ca1SMauro Carvalho Chehab 44689272ca1SMauro Carvalho ChehabF2FS divides the whole volume into a number of segments, each of which is fixed 44789272ca1SMauro Carvalho Chehabto 2MB in size. A section is composed of consecutive segments, and a zone 44889272ca1SMauro Carvalho Chehabconsists of a set of sections. By default, section and zone sizes are set to one 44989272ca1SMauro Carvalho Chehabsegment size identically, but users can easily modify the sizes by mkfs. 45089272ca1SMauro Carvalho Chehab 45189272ca1SMauro Carvalho ChehabF2FS splits the entire volume into six areas, and all the areas except superblock 452ca313c82SRandy Dunlapconsist of multiple segments as described below:: 45389272ca1SMauro Carvalho Chehab 45489272ca1SMauro Carvalho Chehab align with the zone size <-| 45589272ca1SMauro Carvalho Chehab |-> align with the segment size 45689272ca1SMauro Carvalho Chehab _________________________________________________________________________ 45789272ca1SMauro Carvalho Chehab | | | Segment | Node | Segment | | 45889272ca1SMauro Carvalho Chehab | Superblock | Checkpoint | Info. | Address | Summary | Main | 45989272ca1SMauro Carvalho Chehab | (SB) | (CP) | Table (SIT) | Table (NAT) | Area (SSA) | | 46089272ca1SMauro Carvalho Chehab |____________|_____2______|______N______|______N______|______N_____|__N___| 46189272ca1SMauro Carvalho Chehab . . 46289272ca1SMauro Carvalho Chehab . . 46389272ca1SMauro Carvalho Chehab . . 46489272ca1SMauro Carvalho Chehab ._________________________________________. 46589272ca1SMauro Carvalho Chehab |_Segment_|_..._|_Segment_|_..._|_Segment_| 46689272ca1SMauro Carvalho Chehab . . 46789272ca1SMauro Carvalho Chehab ._________._________ 46889272ca1SMauro Carvalho Chehab |_section_|__...__|_ 46989272ca1SMauro Carvalho Chehab . . 47089272ca1SMauro Carvalho Chehab .________. 47189272ca1SMauro Carvalho Chehab |__zone__| 47289272ca1SMauro Carvalho Chehab 47389272ca1SMauro Carvalho Chehab- Superblock (SB) 47489272ca1SMauro Carvalho Chehab It is located at the beginning of the partition, and there exist two copies 47589272ca1SMauro Carvalho Chehab to avoid file system crash. It contains basic partition information and some 47689272ca1SMauro Carvalho Chehab default parameters of f2fs. 47789272ca1SMauro Carvalho Chehab 47889272ca1SMauro Carvalho Chehab- Checkpoint (CP) 47989272ca1SMauro Carvalho Chehab It contains file system information, bitmaps for valid NAT/SIT sets, orphan 48089272ca1SMauro Carvalho Chehab inode lists, and summary entries of current active segments. 48189272ca1SMauro Carvalho Chehab 48289272ca1SMauro Carvalho Chehab- Segment Information Table (SIT) 48389272ca1SMauro Carvalho Chehab It contains segment information such as valid block count and bitmap for the 48489272ca1SMauro Carvalho Chehab validity of all the blocks. 48589272ca1SMauro Carvalho Chehab 48689272ca1SMauro Carvalho Chehab- Node Address Table (NAT) 48789272ca1SMauro Carvalho Chehab It is composed of a block address table for all the node blocks stored in 48889272ca1SMauro Carvalho Chehab Main area. 48989272ca1SMauro Carvalho Chehab 49089272ca1SMauro Carvalho Chehab- Segment Summary Area (SSA) 49189272ca1SMauro Carvalho Chehab It contains summary entries which contains the owner information of all the 49289272ca1SMauro Carvalho Chehab data and node blocks stored in Main area. 49389272ca1SMauro Carvalho Chehab 49489272ca1SMauro Carvalho Chehab- Main Area 49589272ca1SMauro Carvalho Chehab It contains file and directory data including their indices. 49689272ca1SMauro Carvalho Chehab 49789272ca1SMauro Carvalho ChehabIn order to avoid misalignment between file system and flash-based storage, F2FS 49889272ca1SMauro Carvalho Chehabaligns the start block address of CP with the segment size. Also, it aligns the 49989272ca1SMauro Carvalho Chehabstart block address of Main area with the zone size by reserving some segments 50089272ca1SMauro Carvalho Chehabin SSA area. 50189272ca1SMauro Carvalho Chehab 50289272ca1SMauro Carvalho ChehabReference the following survey for additional technical details. 50389272ca1SMauro Carvalho Chehabhttps://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey 50489272ca1SMauro Carvalho Chehab 50589272ca1SMauro Carvalho ChehabFile System Metadata Structure 50689272ca1SMauro Carvalho Chehab------------------------------ 50789272ca1SMauro Carvalho Chehab 50889272ca1SMauro Carvalho ChehabF2FS adopts the checkpointing scheme to maintain file system consistency. At 50989272ca1SMauro Carvalho Chehabmount time, F2FS first tries to find the last valid checkpoint data by scanning 51089272ca1SMauro Carvalho ChehabCP area. In order to reduce the scanning time, F2FS uses only two copies of CP. 51189272ca1SMauro Carvalho ChehabOne of them always indicates the last valid data, which is called as shadow copy 51289272ca1SMauro Carvalho Chehabmechanism. In addition to CP, NAT and SIT also adopt the shadow copy mechanism. 51389272ca1SMauro Carvalho Chehab 51489272ca1SMauro Carvalho ChehabFor file system consistency, each CP points to which NAT and SIT copies are 51589272ca1SMauro Carvalho Chehabvalid, as shown as below:: 51689272ca1SMauro Carvalho Chehab 51789272ca1SMauro Carvalho Chehab +--------+----------+---------+ 51889272ca1SMauro Carvalho Chehab | CP | SIT | NAT | 51989272ca1SMauro Carvalho Chehab +--------+----------+---------+ 52089272ca1SMauro Carvalho Chehab . . . . 52189272ca1SMauro Carvalho Chehab . . . . 52289272ca1SMauro Carvalho Chehab . . . . 52389272ca1SMauro Carvalho Chehab +-------+-------+--------+--------+--------+--------+ 52489272ca1SMauro Carvalho Chehab | CP #0 | CP #1 | SIT #0 | SIT #1 | NAT #0 | NAT #1 | 52589272ca1SMauro Carvalho Chehab +-------+-------+--------+--------+--------+--------+ 52689272ca1SMauro Carvalho Chehab | ^ ^ 52789272ca1SMauro Carvalho Chehab | | | 52889272ca1SMauro Carvalho Chehab `----------------------------------------' 52989272ca1SMauro Carvalho Chehab 53089272ca1SMauro Carvalho ChehabIndex Structure 53189272ca1SMauro Carvalho Chehab--------------- 53289272ca1SMauro Carvalho Chehab 53389272ca1SMauro Carvalho ChehabThe key data structure to manage the data locations is a "node". Similar to 53489272ca1SMauro Carvalho Chehabtraditional file structures, F2FS has three types of node: inode, direct node, 53589272ca1SMauro Carvalho Chehabindirect node. F2FS assigns 4KB to an inode block which contains 923 data block 53689272ca1SMauro Carvalho Chehabindices, two direct node pointers, two indirect node pointers, and one double 53789272ca1SMauro Carvalho Chehabindirect node pointer as described below. One direct node block contains 1018 53889272ca1SMauro Carvalho Chehabdata blocks, and one indirect node block contains also 1018 node blocks. Thus, 53989272ca1SMauro Carvalho Chehabone inode block (i.e., a file) covers:: 54089272ca1SMauro Carvalho Chehab 54189272ca1SMauro Carvalho Chehab 4KB * (923 + 2 * 1018 + 2 * 1018 * 1018 + 1018 * 1018 * 1018) := 3.94TB. 54289272ca1SMauro Carvalho Chehab 54389272ca1SMauro Carvalho Chehab Inode block (4KB) 54489272ca1SMauro Carvalho Chehab |- data (923) 54589272ca1SMauro Carvalho Chehab |- direct node (2) 54689272ca1SMauro Carvalho Chehab | `- data (1018) 54789272ca1SMauro Carvalho Chehab |- indirect node (2) 54889272ca1SMauro Carvalho Chehab | `- direct node (1018) 54989272ca1SMauro Carvalho Chehab | `- data (1018) 55089272ca1SMauro Carvalho Chehab `- double indirect node (1) 55189272ca1SMauro Carvalho Chehab `- indirect node (1018) 55289272ca1SMauro Carvalho Chehab `- direct node (1018) 55389272ca1SMauro Carvalho Chehab `- data (1018) 55489272ca1SMauro Carvalho Chehab 555ca313c82SRandy DunlapNote that all the node blocks are mapped by NAT which means the location of 55689272ca1SMauro Carvalho Chehabeach node is translated by the NAT table. In the consideration of the wandering 55789272ca1SMauro Carvalho Chehabtree problem, F2FS is able to cut off the propagation of node updates caused by 55889272ca1SMauro Carvalho Chehableaf data writes. 55989272ca1SMauro Carvalho Chehab 56089272ca1SMauro Carvalho ChehabDirectory Structure 56189272ca1SMauro Carvalho Chehab------------------- 56289272ca1SMauro Carvalho Chehab 56389272ca1SMauro Carvalho ChehabA directory entry occupies 11 bytes, which consists of the following attributes. 56489272ca1SMauro Carvalho Chehab 56589272ca1SMauro Carvalho Chehab- hash hash value of the file name 56689272ca1SMauro Carvalho Chehab- ino inode number 56789272ca1SMauro Carvalho Chehab- len the length of file name 56889272ca1SMauro Carvalho Chehab- type file type such as directory, symlink, etc 56989272ca1SMauro Carvalho Chehab 57089272ca1SMauro Carvalho ChehabA dentry block consists of 214 dentry slots and file names. Therein a bitmap is 57189272ca1SMauro Carvalho Chehabused to represent whether each dentry is valid or not. A dentry block occupies 57289272ca1SMauro Carvalho Chehab4KB with the following composition. 57389272ca1SMauro Carvalho Chehab 57489272ca1SMauro Carvalho Chehab:: 57589272ca1SMauro Carvalho Chehab 57689272ca1SMauro Carvalho Chehab Dentry Block(4 K) = bitmap (27 bytes) + reserved (3 bytes) + 57789272ca1SMauro Carvalho Chehab dentries(11 * 214 bytes) + file name (8 * 214 bytes) 57889272ca1SMauro Carvalho Chehab 57989272ca1SMauro Carvalho Chehab [Bucket] 58089272ca1SMauro Carvalho Chehab +--------------------------------+ 58189272ca1SMauro Carvalho Chehab |dentry block 1 | dentry block 2 | 58289272ca1SMauro Carvalho Chehab +--------------------------------+ 58389272ca1SMauro Carvalho Chehab . . 58489272ca1SMauro Carvalho Chehab . . 58589272ca1SMauro Carvalho Chehab . [Dentry Block Structure: 4KB] . 58689272ca1SMauro Carvalho Chehab +--------+----------+----------+------------+ 58789272ca1SMauro Carvalho Chehab | bitmap | reserved | dentries | file names | 58889272ca1SMauro Carvalho Chehab +--------+----------+----------+------------+ 58989272ca1SMauro Carvalho Chehab [Dentry Block: 4KB] . . 59089272ca1SMauro Carvalho Chehab . . 59189272ca1SMauro Carvalho Chehab . . 59289272ca1SMauro Carvalho Chehab +------+------+-----+------+ 59389272ca1SMauro Carvalho Chehab | hash | ino | len | type | 59489272ca1SMauro Carvalho Chehab +------+------+-----+------+ 59589272ca1SMauro Carvalho Chehab [Dentry Structure: 11 bytes] 59689272ca1SMauro Carvalho Chehab 59789272ca1SMauro Carvalho ChehabF2FS implements multi-level hash tables for directory structure. Each level has 59889272ca1SMauro Carvalho Chehaba hash table with dedicated number of hash buckets as shown below. Note that 59989272ca1SMauro Carvalho Chehab"A(2B)" means a bucket includes 2 data blocks. 60089272ca1SMauro Carvalho Chehab 60189272ca1SMauro Carvalho Chehab:: 60289272ca1SMauro Carvalho Chehab 60389272ca1SMauro Carvalho Chehab ---------------------- 60489272ca1SMauro Carvalho Chehab A : bucket 60589272ca1SMauro Carvalho Chehab B : block 60689272ca1SMauro Carvalho Chehab N : MAX_DIR_HASH_DEPTH 60789272ca1SMauro Carvalho Chehab ---------------------- 60889272ca1SMauro Carvalho Chehab 60989272ca1SMauro Carvalho Chehab level #0 | A(2B) 61089272ca1SMauro Carvalho Chehab | 61189272ca1SMauro Carvalho Chehab level #1 | A(2B) - A(2B) 61289272ca1SMauro Carvalho Chehab | 61389272ca1SMauro Carvalho Chehab level #2 | A(2B) - A(2B) - A(2B) - A(2B) 61489272ca1SMauro Carvalho Chehab . | . . . . 61589272ca1SMauro Carvalho Chehab level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B) 61689272ca1SMauro Carvalho Chehab . | . . . . 61789272ca1SMauro Carvalho Chehab level #N | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B) 61889272ca1SMauro Carvalho Chehab 61989272ca1SMauro Carvalho ChehabThe number of blocks and buckets are determined by:: 62089272ca1SMauro Carvalho Chehab 62189272ca1SMauro Carvalho Chehab ,- 2, if n < MAX_DIR_HASH_DEPTH / 2, 62289272ca1SMauro Carvalho Chehab # of blocks in level #n = | 62389272ca1SMauro Carvalho Chehab `- 4, Otherwise 62489272ca1SMauro Carvalho Chehab 62589272ca1SMauro Carvalho Chehab ,- 2^(n + dir_level), 62689272ca1SMauro Carvalho Chehab | if n + dir_level < MAX_DIR_HASH_DEPTH / 2, 62789272ca1SMauro Carvalho Chehab # of buckets in level #n = | 62889272ca1SMauro Carvalho Chehab `- 2^((MAX_DIR_HASH_DEPTH / 2) - 1), 62989272ca1SMauro Carvalho Chehab Otherwise 63089272ca1SMauro Carvalho Chehab 63189272ca1SMauro Carvalho ChehabWhen F2FS finds a file name in a directory, at first a hash value of the file 63289272ca1SMauro Carvalho Chehabname is calculated. Then, F2FS scans the hash table in level #0 to find the 63389272ca1SMauro Carvalho Chehabdentry consisting of the file name and its inode number. If not found, F2FS 63489272ca1SMauro Carvalho Chehabscans the next hash table in level #1. In this way, F2FS scans hash tables in 635ca313c82SRandy Dunlapeach levels incrementally from 1 to N. In each level F2FS needs to scan only 63689272ca1SMauro Carvalho Chehabone bucket determined by the following equation, which shows O(log(# of files)) 63789272ca1SMauro Carvalho Chehabcomplexity:: 63889272ca1SMauro Carvalho Chehab 63989272ca1SMauro Carvalho Chehab bucket number to scan in level #n = (hash value) % (# of buckets in level #n) 64089272ca1SMauro Carvalho Chehab 64189272ca1SMauro Carvalho ChehabIn the case of file creation, F2FS finds empty consecutive slots that cover the 64289272ca1SMauro Carvalho Chehabfile name. F2FS searches the empty slots in the hash tables of whole levels from 64389272ca1SMauro Carvalho Chehab1 to N in the same way as the lookup operation. 64489272ca1SMauro Carvalho Chehab 64589272ca1SMauro Carvalho ChehabThe following figure shows an example of two cases holding children:: 64689272ca1SMauro Carvalho Chehab 64789272ca1SMauro Carvalho Chehab --------------> Dir <-------------- 64889272ca1SMauro Carvalho Chehab | | 64989272ca1SMauro Carvalho Chehab child child 65089272ca1SMauro Carvalho Chehab 65189272ca1SMauro Carvalho Chehab child - child [hole] - child 65289272ca1SMauro Carvalho Chehab 65389272ca1SMauro Carvalho Chehab child - child - child [hole] - [hole] - child 65489272ca1SMauro Carvalho Chehab 65589272ca1SMauro Carvalho Chehab Case 1: Case 2: 65689272ca1SMauro Carvalho Chehab Number of children = 6, Number of children = 3, 65789272ca1SMauro Carvalho Chehab File size = 7 File size = 7 65889272ca1SMauro Carvalho Chehab 65989272ca1SMauro Carvalho ChehabDefault Block Allocation 66089272ca1SMauro Carvalho Chehab------------------------ 66189272ca1SMauro Carvalho Chehab 66289272ca1SMauro Carvalho ChehabAt runtime, F2FS manages six active logs inside "Main" area: Hot/Warm/Cold node 66389272ca1SMauro Carvalho Chehaband Hot/Warm/Cold data. 66489272ca1SMauro Carvalho Chehab 66589272ca1SMauro Carvalho Chehab- Hot node contains direct node blocks of directories. 66689272ca1SMauro Carvalho Chehab- Warm node contains direct node blocks except hot node blocks. 66789272ca1SMauro Carvalho Chehab- Cold node contains indirect node blocks 66889272ca1SMauro Carvalho Chehab- Hot data contains dentry blocks 66989272ca1SMauro Carvalho Chehab- Warm data contains data blocks except hot and cold data blocks 67089272ca1SMauro Carvalho Chehab- Cold data contains multimedia data or migrated data blocks 67189272ca1SMauro Carvalho Chehab 67289272ca1SMauro Carvalho ChehabLFS has two schemes for free space management: threaded log and copy-and-compac- 67389272ca1SMauro Carvalho Chehabtion. The copy-and-compaction scheme which is known as cleaning, is well-suited 67489272ca1SMauro Carvalho Chehabfor devices showing very good sequential write performance, since free segments 67589272ca1SMauro Carvalho Chehabare served all the time for writing new data. However, it suffers from cleaning 67689272ca1SMauro Carvalho Chehaboverhead under high utilization. Contrarily, the threaded log scheme suffers 67789272ca1SMauro Carvalho Chehabfrom random writes, but no cleaning process is needed. F2FS adopts a hybrid 67889272ca1SMauro Carvalho Chehabscheme where the copy-and-compaction scheme is adopted by default, but the 67989272ca1SMauro Carvalho Chehabpolicy is dynamically changed to the threaded log scheme according to the file 68089272ca1SMauro Carvalho Chehabsystem status. 68189272ca1SMauro Carvalho Chehab 68289272ca1SMauro Carvalho ChehabIn order to align F2FS with underlying flash-based storage, F2FS allocates a 68389272ca1SMauro Carvalho Chehabsegment in a unit of section. F2FS expects that the section size would be the 68489272ca1SMauro Carvalho Chehabsame as the unit size of garbage collection in FTL. Furthermore, with respect 68589272ca1SMauro Carvalho Chehabto the mapping granularity in FTL, F2FS allocates each section of the active 68689272ca1SMauro Carvalho Chehablogs from different zones as much as possible, since FTL can write the data in 68789272ca1SMauro Carvalho Chehabthe active logs into one allocation unit according to its mapping granularity. 68889272ca1SMauro Carvalho Chehab 68989272ca1SMauro Carvalho ChehabCleaning process 69089272ca1SMauro Carvalho Chehab---------------- 69189272ca1SMauro Carvalho Chehab 69289272ca1SMauro Carvalho ChehabF2FS does cleaning both on demand and in the background. On-demand cleaning is 69389272ca1SMauro Carvalho Chehabtriggered when there are not enough free segments to serve VFS calls. Background 69489272ca1SMauro Carvalho Chehabcleaner is operated by a kernel thread, and triggers the cleaning job when the 69589272ca1SMauro Carvalho Chehabsystem is idle. 69689272ca1SMauro Carvalho Chehab 69789272ca1SMauro Carvalho ChehabF2FS supports two victim selection policies: greedy and cost-benefit algorithms. 69889272ca1SMauro Carvalho ChehabIn the greedy algorithm, F2FS selects a victim segment having the smallest number 69989272ca1SMauro Carvalho Chehabof valid blocks. In the cost-benefit algorithm, F2FS selects a victim segment 70089272ca1SMauro Carvalho Chehabaccording to the segment age and the number of valid blocks in order to address 70189272ca1SMauro Carvalho Chehablog block thrashing problem in the greedy algorithm. F2FS adopts the greedy 70289272ca1SMauro Carvalho Chehabalgorithm for on-demand cleaner, while background cleaner adopts cost-benefit 70389272ca1SMauro Carvalho Chehabalgorithm. 70489272ca1SMauro Carvalho Chehab 70589272ca1SMauro Carvalho ChehabIn order to identify whether the data in the victim segment are valid or not, 70689272ca1SMauro Carvalho ChehabF2FS manages a bitmap. Each bit represents the validity of a block, and the 70789272ca1SMauro Carvalho Chehabbitmap is composed of a bit stream covering whole blocks in main area. 70889272ca1SMauro Carvalho Chehab 70989272ca1SMauro Carvalho ChehabWrite-hint Policy 71089272ca1SMauro Carvalho Chehab----------------- 71189272ca1SMauro Carvalho Chehab 71289272ca1SMauro Carvalho Chehab1) whint_mode=off. F2FS only passes down WRITE_LIFE_NOT_SET. 71389272ca1SMauro Carvalho Chehab 71489272ca1SMauro Carvalho Chehab2) whint_mode=user-based. F2FS tries to pass down hints given by 71589272ca1SMauro Carvalho Chehabusers. 71689272ca1SMauro Carvalho Chehab 71789272ca1SMauro Carvalho Chehab===================== ======================== =================== 71889272ca1SMauro Carvalho ChehabUser F2FS Block 71989272ca1SMauro Carvalho Chehab===================== ======================== =================== 72089272ca1SMauro Carvalho Chehab META WRITE_LIFE_NOT_SET 72189272ca1SMauro Carvalho Chehab HOT_NODE " 72289272ca1SMauro Carvalho Chehab WARM_NODE " 72389272ca1SMauro Carvalho Chehab COLD_NODE " 72489272ca1SMauro Carvalho Chehabioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME 72589272ca1SMauro Carvalho Chehabextension list " " 72689272ca1SMauro Carvalho Chehab 72789272ca1SMauro Carvalho Chehab-- buffered io 72889272ca1SMauro Carvalho ChehabWRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME 72989272ca1SMauro Carvalho ChehabWRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT 73089272ca1SMauro Carvalho ChehabWRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET 73189272ca1SMauro Carvalho ChehabWRITE_LIFE_NONE " " 73289272ca1SMauro Carvalho ChehabWRITE_LIFE_MEDIUM " " 73389272ca1SMauro Carvalho ChehabWRITE_LIFE_LONG " " 73489272ca1SMauro Carvalho Chehab 73589272ca1SMauro Carvalho Chehab-- direct io 73689272ca1SMauro Carvalho ChehabWRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME 73789272ca1SMauro Carvalho ChehabWRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT 73889272ca1SMauro Carvalho ChehabWRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET 73989272ca1SMauro Carvalho ChehabWRITE_LIFE_NONE " WRITE_LIFE_NONE 74089272ca1SMauro Carvalho ChehabWRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM 74189272ca1SMauro Carvalho ChehabWRITE_LIFE_LONG " WRITE_LIFE_LONG 74289272ca1SMauro Carvalho Chehab===================== ======================== =================== 74389272ca1SMauro Carvalho Chehab 74489272ca1SMauro Carvalho Chehab3) whint_mode=fs-based. F2FS passes down hints with its policy. 74589272ca1SMauro Carvalho Chehab 74689272ca1SMauro Carvalho Chehab===================== ======================== =================== 74789272ca1SMauro Carvalho ChehabUser F2FS Block 74889272ca1SMauro Carvalho Chehab===================== ======================== =================== 74989272ca1SMauro Carvalho Chehab META WRITE_LIFE_MEDIUM; 75089272ca1SMauro Carvalho Chehab HOT_NODE WRITE_LIFE_NOT_SET 75189272ca1SMauro Carvalho Chehab WARM_NODE " 75289272ca1SMauro Carvalho Chehab COLD_NODE WRITE_LIFE_NONE 75389272ca1SMauro Carvalho Chehabioctl(COLD) COLD_DATA WRITE_LIFE_EXTREME 75489272ca1SMauro Carvalho Chehabextension list " " 75589272ca1SMauro Carvalho Chehab 75689272ca1SMauro Carvalho Chehab-- buffered io 75789272ca1SMauro Carvalho ChehabWRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME 75889272ca1SMauro Carvalho ChehabWRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT 75989272ca1SMauro Carvalho ChehabWRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_LONG 76089272ca1SMauro Carvalho ChehabWRITE_LIFE_NONE " " 76189272ca1SMauro Carvalho ChehabWRITE_LIFE_MEDIUM " " 76289272ca1SMauro Carvalho ChehabWRITE_LIFE_LONG " " 76389272ca1SMauro Carvalho Chehab 76489272ca1SMauro Carvalho Chehab-- direct io 76589272ca1SMauro Carvalho ChehabWRITE_LIFE_EXTREME COLD_DATA WRITE_LIFE_EXTREME 76689272ca1SMauro Carvalho ChehabWRITE_LIFE_SHORT HOT_DATA WRITE_LIFE_SHORT 76789272ca1SMauro Carvalho ChehabWRITE_LIFE_NOT_SET WARM_DATA WRITE_LIFE_NOT_SET 76889272ca1SMauro Carvalho ChehabWRITE_LIFE_NONE " WRITE_LIFE_NONE 76989272ca1SMauro Carvalho ChehabWRITE_LIFE_MEDIUM " WRITE_LIFE_MEDIUM 77089272ca1SMauro Carvalho ChehabWRITE_LIFE_LONG " WRITE_LIFE_LONG 77189272ca1SMauro Carvalho Chehab===================== ======================== =================== 77289272ca1SMauro Carvalho Chehab 77389272ca1SMauro Carvalho ChehabFallocate(2) Policy 77489272ca1SMauro Carvalho Chehab------------------- 77589272ca1SMauro Carvalho Chehab 776ca313c82SRandy DunlapThe default policy follows the below POSIX rule. 77789272ca1SMauro Carvalho Chehab 77889272ca1SMauro Carvalho ChehabAllocating disk space 77989272ca1SMauro Carvalho Chehab The default operation (i.e., mode is zero) of fallocate() allocates 78089272ca1SMauro Carvalho Chehab the disk space within the range specified by offset and len. The 78189272ca1SMauro Carvalho Chehab file size (as reported by stat(2)) will be changed if offset+len is 78289272ca1SMauro Carvalho Chehab greater than the file size. Any subregion within the range specified 78389272ca1SMauro Carvalho Chehab by offset and len that did not contain data before the call will be 78489272ca1SMauro Carvalho Chehab initialized to zero. This default behavior closely resembles the 78589272ca1SMauro Carvalho Chehab behavior of the posix_fallocate(3) library function, and is intended 78689272ca1SMauro Carvalho Chehab as a method of optimally implementing that function. 78789272ca1SMauro Carvalho Chehab 78889272ca1SMauro Carvalho ChehabHowever, once F2FS receives ioctl(fd, F2FS_IOC_SET_PIN_FILE) in prior to 789ca313c82SRandy Dunlapfallocate(fd, DEFAULT_MODE), it allocates on-disk block addressess having 79089272ca1SMauro Carvalho Chehabzero or random data, which is useful to the below scenario where: 79189272ca1SMauro Carvalho Chehab 79289272ca1SMauro Carvalho Chehab 1. create(fd) 79389272ca1SMauro Carvalho Chehab 2. ioctl(fd, F2FS_IOC_SET_PIN_FILE) 79489272ca1SMauro Carvalho Chehab 3. fallocate(fd, 0, 0, size) 79589272ca1SMauro Carvalho Chehab 4. address = fibmap(fd, offset) 79689272ca1SMauro Carvalho Chehab 5. open(blkdev) 79789272ca1SMauro Carvalho Chehab 6. write(blkdev, address) 79889272ca1SMauro Carvalho Chehab 79989272ca1SMauro Carvalho ChehabCompression implementation 80089272ca1SMauro Carvalho Chehab-------------------------- 80189272ca1SMauro Carvalho Chehab 80289272ca1SMauro Carvalho Chehab- New term named cluster is defined as basic unit of compression, file can 80389272ca1SMauro Carvalho Chehab be divided into multiple clusters logically. One cluster includes 4 << n 80489272ca1SMauro Carvalho Chehab (n >= 0) logical pages, compression size is also cluster size, each of 80589272ca1SMauro Carvalho Chehab cluster can be compressed or not. 80689272ca1SMauro Carvalho Chehab 80789272ca1SMauro Carvalho Chehab- In cluster metadata layout, one special block address is used to indicate 808ca313c82SRandy Dunlap a cluster is a compressed one or normal one; for compressed cluster, following 80989272ca1SMauro Carvalho Chehab metadata maps cluster to [1, 4 << n - 1] physical blocks, in where f2fs 81089272ca1SMauro Carvalho Chehab stores data including compress header and compressed data. 81189272ca1SMauro Carvalho Chehab 81289272ca1SMauro Carvalho Chehab- In order to eliminate write amplification during overwrite, F2FS only 81389272ca1SMauro Carvalho Chehab support compression on write-once file, data can be compressed only when 8144fc781a3SChao Yu all logical blocks in cluster contain valid data and compress ratio of 8154fc781a3SChao Yu cluster data is lower than specified threshold. 81689272ca1SMauro Carvalho Chehab 81789272ca1SMauro Carvalho Chehab- To enable compression on regular inode, there are three ways: 81889272ca1SMauro Carvalho Chehab 81989272ca1SMauro Carvalho Chehab * chattr +c file 82089272ca1SMauro Carvalho Chehab * chattr +c dir; touch dir/file 82189272ca1SMauro Carvalho Chehab * mount w/ -o compress_extension=ext; touch file.ext 822*38740707SChao Yu * mount w/ -o compress_extension=*; touch any_file 823*38740707SChao Yu 824*38740707SChao Yu- At this point, compression feature doesn't expose compressed space to user 825*38740707SChao Yu directly in order to guarantee potential data updates later to the space. 826*38740707SChao Yu Instead, the main goal is to reduce data writes to flash disk as much as 827*38740707SChao Yu possible, resulting in extending disk life time as well as relaxing IO 828*38740707SChao Yu congestion. Alternatively, we've added ioctl interface to reclaim compressed 829*38740707SChao Yu space and show it to user after putting the immutable bit. 83089272ca1SMauro Carvalho Chehab 83189272ca1SMauro Carvalho ChehabCompress metadata layout:: 83289272ca1SMauro Carvalho Chehab 83389272ca1SMauro Carvalho Chehab [Dnode Structure] 83489272ca1SMauro Carvalho Chehab +-----------------------------------------------+ 83589272ca1SMauro Carvalho Chehab | cluster 1 | cluster 2 | ......... | cluster N | 83689272ca1SMauro Carvalho Chehab +-----------------------------------------------+ 83789272ca1SMauro Carvalho Chehab . . . . 83889272ca1SMauro Carvalho Chehab . . . . 83989272ca1SMauro Carvalho Chehab . Compressed Cluster . . Normal Cluster . 84089272ca1SMauro Carvalho Chehab +----------+---------+---------+---------+ +---------+---------+---------+---------+ 84189272ca1SMauro Carvalho Chehab |compr flag| block 1 | block 2 | block 3 | | block 1 | block 2 | block 3 | block 4 | 84289272ca1SMauro Carvalho Chehab +----------+---------+---------+---------+ +---------+---------+---------+---------+ 84389272ca1SMauro Carvalho Chehab . . 84489272ca1SMauro Carvalho Chehab . . 84589272ca1SMauro Carvalho Chehab . . 84689272ca1SMauro Carvalho Chehab +-------------+-------------+----------+----------------------------+ 84789272ca1SMauro Carvalho Chehab | data length | data chksum | reserved | compressed data | 84889272ca1SMauro Carvalho Chehab +-------------+-------------+----------+----------------------------+ 849de881df9SAravind Ramesh 850602a16d5SDaeho JeongCompression mode 851602a16d5SDaeho Jeong-------------------------- 852602a16d5SDaeho Jeong 853602a16d5SDaeho Jeongf2fs supports "fs" and "user" compression modes with "compression_mode" mount option. 854602a16d5SDaeho JeongWith this option, f2fs provides a choice to select the way how to compress the 855602a16d5SDaeho Jeongcompression enabled files (refer to "Compression implementation" section for how to 856602a16d5SDaeho Jeongenable compression on a regular inode). 857602a16d5SDaeho Jeong 858602a16d5SDaeho Jeong1) compress_mode=fs 859602a16d5SDaeho JeongThis is the default option. f2fs does automatic compression in the writeback of the 860602a16d5SDaeho Jeongcompression enabled files. 861602a16d5SDaeho Jeong 862602a16d5SDaeho Jeong2) compress_mode=user 863092af2ebSEd TsaiThis disables the automatic compression and gives the user discretion of choosing the 864602a16d5SDaeho Jeongtarget file and the timing. The user can do manual compression/decompression on the 865602a16d5SDaeho Jeongcompression enabled files using F2FS_IOC_DECOMPRESS_FILE and F2FS_IOC_COMPRESS_FILE 866602a16d5SDaeho Jeongioctls like the below. 867602a16d5SDaeho Jeong 868602a16d5SDaeho JeongTo decompress a file, 869602a16d5SDaeho Jeong 870602a16d5SDaeho Jeongfd = open(filename, O_WRONLY, 0); 871602a16d5SDaeho Jeongret = ioctl(fd, F2FS_IOC_DECOMPRESS_FILE); 872602a16d5SDaeho Jeong 873602a16d5SDaeho JeongTo compress a file, 874602a16d5SDaeho Jeong 875602a16d5SDaeho Jeongfd = open(filename, O_WRONLY, 0); 876602a16d5SDaeho Jeongret = ioctl(fd, F2FS_IOC_COMPRESS_FILE); 877602a16d5SDaeho Jeong 878de881df9SAravind RameshNVMe Zoned Namespace devices 879de881df9SAravind Ramesh---------------------------- 880de881df9SAravind Ramesh 881de881df9SAravind Ramesh- ZNS defines a per-zone capacity which can be equal or less than the 882de881df9SAravind Ramesh zone-size. Zone-capacity is the number of usable blocks in the zone. 883ca313c82SRandy Dunlap F2FS checks if zone-capacity is less than zone-size, if it is, then any 884de881df9SAravind Ramesh segment which starts after the zone-capacity is marked as not-free in 885de881df9SAravind Ramesh the free segment bitmap at initial mount time. These segments are marked 886de881df9SAravind Ramesh as permanently used so they are not allocated for writes and 887de881df9SAravind Ramesh consequently are not needed to be garbage collected. In case the 888de881df9SAravind Ramesh zone-capacity is not aligned to default segment size(2MB), then a segment 889de881df9SAravind Ramesh can start before the zone-capacity and span across zone-capacity boundary. 890de881df9SAravind Ramesh Such spanning segments are also considered as usable segments. All blocks 891de881df9SAravind Ramesh past the zone-capacity are considered unusable in these segments. 892