2 .\" Copyright 1993, 1994, 1995 by Theodore Ts'o. All Rights Reserved.
3 .\" This file may be copied under the terms of the GNU Public License.
5 .TH EXT4 5 "@E2FSPROGS_MONTH@ @E2FSPROGS_YEAR@" "E2fsprogs version @E2FSPROGS_VERSION@"
7 ext2 \- the second extended file system
9 ext2 \- the third extended file system
11 ext4 \- the fourth extended file system
13 The second, third, and fourth extended file systems, or ext2, ext3, and
14 ext4 as they are commonly known, are Linux file systems that have
15 historically been the default file system for many Linux distributions.
16 They are general purpose file systems that have been designed for
17 extensibility and backwards compatibility. In particular, file systems
18 previously intended for use with the ext2 and ext3 file systems can be
19 mounted using the ext4 file system driver, and indeed in many modern
20 Linux distributions, the ext4 file system driver has been configured
21 handle mount requests for ext2 and ext3 file systems.
22 .SH FILE SYSTEM FEATURES
23 A file system formated for ext2, ext3, or ext4 can be have some
24 collection of the follow file system feature flags enabled. Some of
25 these features are not supported by all implementations of the ext2,
26 ext3, and ext4 file system drivers, depending on Linux kernel version in
27 use. On other operating systems, such as the GNU/HURD or FreeBSD, only
28 a very restrictive set of file system features may be supported in their
29 implementations of ext2.
34 Enables the file system to be larger than 2^32 blocks. This feature is set
35 automatically, as needed, but it can be useful to specify this feature
36 explicitly if the file system might need to be resized larger than 2^32
37 blocks, even if it was smaller than that threshold when it was
38 originally created. Note that some older kernels and older versions
39 of e2fsprogs will not support file systems with this ext4 feature enabled.
43 This ext4 feature enables clustered block allocation, so that the unit of
44 allocation is a power of two number of blocks. That is, each bit in the
45 what had traditionally been known as the block allocation bitmap now
46 indicates whether a cluster is in use or not, where a cluster is by
47 default composed of 16 blocks. This feature can decrease the time
48 spent on doing block allocation and brings smaller fragmentation, especially
49 for large files. The size can be specified using the
53 The bigalloc feature is still under development, and may not be fully
54 supported with your kernel or may have various bugs. Please see the web
55 page http://ext4.wiki.kernel.org/index.php/Bigalloc for details.
56 May clash with delayed allocation (see
57 .BR nodelalloc mount option).
59 This feature requires that the
65 Use hashed b-trees to speed up name lookups in large directories. This
66 feature is supported by ext3 and ext4 file systems, and is ignored by
71 This ext4 feature allows more than 65000 subdirectories per directory.
75 This ext4 feature allows the mapping of logical block numbers for a
76 particular inode to physical blocks on the storage device to be stored
77 using an extent tree, which is a more efficient data structure than the
78 traditional indirect block scheme used by the ext2 and ext3 file
79 systems. The use of the extent tree decreases metadata block overhead,
80 improves file system performance, and decreases the needed to run
87 are accepted as valid names for this feature for
88 historical/backwards compatibility reasons.)
92 This ext4 feature reserves a specific amount of space in each inode for
93 extended metadata such as nanosecond timestamps and file creation time,
94 even if the current kernel does not current need to reserve this much
95 space. Without this feature, the kernel will reserve the amount of
96 space for features currently it currently needs, and the rest may be
97 consumed by extended attributes.
99 For this feature to be useful the inode size must be 256 bytes in size
104 This feature enables the use of extended attributes. This feature is
105 supported by ext2, ext3, and ext4.
109 This feature enables the storage file type information in directory
110 entries. This feature is supported by ext2, ext3, and ext4.
114 This ext4 feature allows the per-block group metadata (allocation
117 to be placed anywhere on the storage media. In addition,
119 will place the per-block group metadata together starting at the first
120 block group of each "flex_bg group". The size of the flex_bg group
121 can be specified using the
127 Create a journal to ensure filesystem consistency even across unclean
128 shutdowns. Setting the filesystem feature is equivalent to using the
130 option. This feature is supported by ext3 and ext4, and ignored by the
131 ext2 file system driver.
135 This ext4 feature allows files to be larger than 2 terabytes in size.
139 This feature is enabled on the superblock found on an external journal
140 device. The block size for the external journal must be the same as the
141 file system which uses it.
143 The external journal device can be used by a file system by specifying
146 .BR device= <external-device>
153 Allow data to be stored in the inode and extended attribute area
157 This feature flag is set automatically by modern kernels when a file
158 larger than 2 gigabytes is created. Very old kernels could not
159 handle large files, so this feature flag was used to prohibit those
160 kernels from mounting file systems that they could not understand.
164 .\" This ext4 feature enables metadata checksumming. This feature stores
165 .\" checksums for all of the filesystem metadata (superblock, group
166 .\" descriptor blocks, inode and block bitmaps, directories, and
167 .\" extent tree blocks). The checksum algorithm used for the metadata
168 .\" blocks is different than the one used for group descriptors with the
170 .\" feature, these two features are incompatible and
172 .\" will be used preferentially instead of
175 .\" .B Future feature, available in e2fsprogs 1.43-WIP
179 This feature indicates that there will only at most two backup
180 superblock and block group descriptors. The block groups used to store
181 the backup superblock and blockgroup descriptors are stored in the
182 superblock, but typically, one will be located at the beginning of block
183 group #1, and one in the last block group in the file system. This is
184 feature is essentially a more extreme version of sparse_super and is
185 designed to allow the a much larger percentage of the disk to have
186 contiguous blocks available for data files.
190 This ext4 feature allows file systems to be resized on-line without explicitly
191 needing to reserve space for growth in the size of the block group
192 descriptors. This scheme is also used to resize file systems which are
193 larger than 2^32 blocks. It is not recommended that this feature be set
194 when a file system is created, since this alternate method of storing
195 the block group descriptor will slow down the time needed to mount the
196 file system, and newer kernels can automatically set this feature as
197 necessary when doing an online resize and no more reserved space is
198 available in the resize inode.
202 This ext4 feature provides multiple mount protection (MMP). MMP helps to
203 protect the filesystem from being multiply mounted and is useful in
204 shared storage environments.
205 @QUOTA_MAN_COMMENT@.TP
206 @QUOTA_MAN_COMMENT@.B quota
207 @QUOTA_MAN_COMMENT@.br
208 @QUOTA_MAN_COMMENT@Create quota inodes (inode #3 for userquota and inode
209 @QUOTA_MAN_COMMENT@#4 for group quota) and set them in the superblock.
210 @QUOTA_MAN_COMMENT@With this feature, the quotas will be enabled
211 @QUOTA_MAN_COMMENT@automatically when the filesystem is mounted.
212 @QUOTA_MAN_COMMENT@.IP
213 @QUOTA_MAN_COMMENT@Causes the quota files (i.e., user.quota and
214 @QUOTA_MAN_COMMENT@group.quota which existed
215 @QUOTA_MAN_COMMENT@in the older quota design) to be hidden inodes.
219 This file system feature indicates that space has been reserved so
220 the block group descriptor table can be extended by the file system is
221 resized while the file system is mounted. The online resize operation
222 is carried out by the kernel, triggered, by
226 will attempt to reserve enough space so that the
227 filesystem may grow to 1024 times its initial size. This can be changed
232 This feature requires that the
238 This file system feature is set on all modern ext2, ext3, and ext4 file
239 system. It indicates that backup copies of the superblock and block
240 group descriptors be present only on a few block groups, and not all of
245 This ext4 file system feature indicates that the block group descriptors
246 will be protected using checksums, making it safe for
248 to create a file system without initializing all of the block groups.
249 The kernel will keep a high watermark of unused inodes, and initialize
250 inode tables and block lazily. This feature speeds up the time to check
251 the file system using
253 and it also speeds up the time required for
255 to create the file system.
258 This section describes mount options which are specific to ext2, ext3,
259 and ext4. Other generic mount options may be used as well; see
262 .SH "Mount options for ext2"
263 The `ext2' filesystem is the standard Linux filesystem.
264 Since Linux 2.5.46, for most mount options the default
265 is determined by the filesystem superblock. Set them with
269 Support POSIX Access Control Lists (or not).
272 Set the behavior for the
276 behavior is to return in the
278 field the total number of blocks of the filesystem, while the
280 behavior (which is the default) is to subtract the overhead blocks
281 used by the ext2 filesystem and not available for file storage. Thus
283 % mount /k \-o minixdf; df /k; umount /k
288 Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
289 /dev/sda6#2630655#86954#2412169#3%#/k
292 % mount /k \-o bsddf; df /k; umount /k
297 Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
298 /dev/sda6#2543714#13#2412169#0%#/k
301 (Note that this example shows that one can add command line options
302 to the options given in
305 .BR check=none " or " nocheck
306 No checking is done at mount time. This is the default. This is fast.
309 every now and then, e.g.\& at boot time. The non-default behavior is unsupported
310 (check=normal and check=strict options have been removed). Note that these mount options
311 don't have to be supported if ext4 kernel driver is used for ext2 and ext3 filesystems.
314 Print debugging info upon each (re)mount.
316 .BR errors= { continue | remount-ro | panic }
317 Define the behavior when an error is encountered.
318 (Either ignore errors and just mark the filesystem erroneous and continue,
319 or remount the filesystem read-only, or panic and halt the system.)
320 The default is set in the filesystem superblock, and can be
324 .BR grpid | bsdgroups " and " nogrpid | sysvgroups
325 These options define what group id a newly created file gets.
328 is set, it takes the group id of the directory in which it is created;
329 otherwise (the default) it takes the fsgid of the current process, unless
330 the directory has the setgid bit set, in which case it takes the gid
331 from the parent directory, and also gets the setgid bit set
332 if it is a directory itself.
334 .BR grpquota | noquota | quota | usrquota
335 The usrquota (same as quota) mount option enables user quota support on the
336 filesystem. grpquota enables group quotas support. You need the quota utilities
337 to actually enable and manage the quota system.
340 Disables 32-bit UIDs and GIDs. This is for interoperability with older
341 kernels which only store and expect 16-bit values.
343 .BR oldalloc " or " orlov
344 Use old allocator or Orlov allocator for new inodes. Orlov is default.
346 \fBresgid=\fP\,\fIn\fP and \fBresuid=\fP\,\fIn\fP
347 The ext2 filesystem reserves a certain percentage of the available
348 space (by default 5%, see
352 These options determine who can use the reserved blocks.
353 (Roughly: whoever has the specified uid, or belongs to the specified group.)
356 Instead of block 1, use block
358 as superblock. This could be useful when the filesystem has been damaged.
359 (Earlier, copies of the superblock would be made every 8192 blocks: in
360 block 1, 8193, 16385, \&...\& (and one got thousands of copies on
361 a big filesystem). Since version 1.08,
363 has a \-s (sparse superblock) option to reduce the number of backup
364 superblocks, and since version 1.15 this is the default. Note
365 that this may mean that ext2 filesystems created by a recent
367 cannot be mounted r/w under Linux 2.0.*.)
368 The block number here uses 1\ k units. Thus, if you want to use logical
369 block 32768 on a filesystem with 4\ k blocks, use "sb=131072".
371 .BR user_xattr | nouser_xattr
372 Support "user." extended attributes (or not).
375 .SH "Mount options for ext3"
376 The ext3 filesystem is a version of the ext2 filesystem which has been
377 enhanced with journaling. It supports the same options as ext2 as
378 well as the following additions:
381 Update the ext3 filesystem's journal to the current format.
384 When a journal already exists, this option is ignored. Otherwise, it
385 specifies the number of the inode which will represent the ext3 filesystem's
386 journal file; ext3 will create a new journal, overwriting the old contents
387 of the file whose inode number is
390 .BR journal_dev=devnum / journal_path=path
391 When the external journal device's major/minor numbers
392 have changed, these options allow the user to specify
393 the new journal location. The journal device is
394 identified either through its new major/minor numbers encoded
395 in devnum, or via a path to the device.
397 .BR norecovery / noload
398 Don't load the journal on mounting. Note that
399 if the filesystem was not unmounted cleanly,
400 skipping the journal replay will lead to the
401 filesystem containing inconsistencies that can
402 lead to any number of problems.
404 .BR data= { journal | ordered | writeback }
405 Specifies the journaling mode for file data. Metadata is always journaled.
406 To use modes other than
408 on the root filesystem, pass the mode to the kernel as boot parameter, e.g.\&
409 .IR rootflags=data=journal .
413 All data is committed into the journal prior to being written into the
417 This is the default mode. All data is forced directly out to the main file
418 system prior to its metadata being committed to the journal.
421 Data ordering is not preserved \(en data may be written into the main
422 filesystem after its metadata has been committed to the journal.
423 This is rumoured to be the highest-throughput option. It guarantees
424 internal filesystem integrity, however it can allow old data to appear
425 in files after a crash and journal recovery.
429 Just print an error message if an error occurs in a file data buffer in
433 Abort the journal if an error occurs in a file data buffer in ordered mode.
435 .BR barrier=0 " / " barrier=1 "
436 This disables / enables the use of write barriers in the jbd code. barrier=0
437 disables, barrier=1 enables (default). This also requires an IO stack which can
438 support barriers, and if jbd gets an error on a barrier write, it will disable
439 barriers again with a warning. Write barriers enforce proper on-disk ordering
440 of journal commits, making volatile disk write caches safe to use, at some
441 performance penalty. If your disks are battery-backed in one way or another,
442 disabling barriers may safely improve performance.
445 Sync all data and metadata every
447 seconds. The default value is 5 seconds. Zero means default.
450 Enable Extended User Attributes. See the
455 Enable POSIX Access Control Lists. See the
459 .BR usrjquota=aquota.user | grpjquota=aquota.group | jqfmt=vfsv0
460 Apart from the old quota system (as in ext2, jqfmt=vfsold aka version 1 quota)
461 ext3 also supports journaled quotas (version 2 quota). jqfmt=vfsv0
462 enables journaled quotas. For journaled quotas the mount options
463 usrjquota=aquota.user and grpjquota=aquota.group are required to tell the
464 quota system which quota database files to use. Journaled quotas have the
465 advantage that even after a crash no quota check is required.
467 .SH "Mount options for ext4"
468 The ext4 filesystem is an advanced level of the ext3 filesystem which
469 incorporates scalability and reliability enhancements for supporting large
473 .B journal_dev, norecovery, noload, data, commit, orlov, oldalloc, [no]user_xattr
474 .B [no]acl, bsddf, minixdf, debug, errors, data_err, grpid, bsdgroups, nogrpid
475 .B sysvgroups, resgid, resuid, sb, quota, noquota, grpquota, usrquota
476 .B usrjquota, grpjquota and jqfmt
477 are backwardly compatible with ext3 or ext2.
480 Enable checksumming of the journal transactions. This will allow the recovery
481 code in e2fsck and the kernel to detect corruption in the kernel. It is a
482 compatible change and will be ignored by older kernels.
484 .B journal_async_commit
485 Commit block can be written to disk without waiting for descriptor blocks. If
486 enabled older kernels cannot mount the device.
487 This will enable 'journal_checksum' internally.
489 .BR barrier=0 " / " barrier=1 " / " barrier " / " nobarrier
490 These mount options have the same effect as in ext3. The mount options
491 "barrier" and "nobarrier" are added for consistency with other ext4 mount
494 The ext4 filesystem enables write barriers by default.
496 .BI inode_readahead_blks= n
497 This tuning parameter controls the maximum number of inode table blocks that
498 ext4's inode table readahead algorithm will pre-read into the buffer cache.
499 The value must be a power of 2. The default value is 32 blocks.
502 Number of filesystem blocks that mballoc will try to use for allocation size
503 and alignment. For RAID5/6 systems this should be the number of data disks *
504 RAID chunk size in filesystem blocks.
507 Deferring block allocation until write-out time.
510 Disable delayed allocation. Blocks are allocated when data is copied from user
513 .BI max_batch_time= usec
514 Maximum amount of time ext4 should wait for additional filesystem operations to
515 be batch together with a synchronous write operation. Since a synchronous
516 write operation is going to force a commit and then a wait for the I/O
517 complete, it doesn't cost much, and can be a huge throughput win, we wait for a
518 small amount of time to see if any other transactions can piggyback on the
519 synchronous write. The algorithm used is designed to automatically tune for
520 the speed of the disk, by measuring the amount of time (on average) that it
521 takes to finish committing a transaction. Call this time the "commit time".
522 If the time that the transaction has been running is less than the commit time,
523 ext4 will try sleeping for the commit time to see if other operations will join
524 the transaction. The commit time is capped by the max_batch_time, which
525 defaults to 15000\ \[mc]s (15\ ms). This optimization can be turned off entirely by
526 setting max_batch_time to 0.
528 .BI min_batch_time= usec
529 This parameter sets the commit time (as described above) to be at least
530 min_batch_time. It defaults to zero microseconds. Increasing this parameter
531 may improve the throughput of multi-threaded, synchronous workloads on very
532 fast disks, at the cost of increasing latency.
534 .BI journal_ioprio= prio
535 The I/O priority (from 0 to 7, where 0 is the highest priority) which should be
536 used for I/O operations submitted by kjournald2 during a commit operation.
537 This defaults to 3, which is a slightly higher priority than the default I/O
541 Simulate the effects of calling ext4_abort() for
542 debugging purposes. This is normally used while
543 remounting a filesystem which is already mounted.
545 .BR auto_da_alloc | noauto_da_alloc
546 Many broken applications don't use fsync() when
547 replacing existing files via patterns such as
549 fd = open("foo.new")/write(fd,...)/close(fd)/ rename("foo.new", "foo")
553 fd = open("foo", O_TRUNC)/write(fd,...)/close(fd).
555 If auto_da_alloc is enabled, ext4 will detect the replace-via-rename and
556 replace-via-truncate patterns and force that any delayed allocation blocks are
557 allocated such that at the next journal commit, in the default data=ordered
558 mode, the data blocks of the new file are forced to disk before the rename()
559 operation is committed. This provides roughly the same level of guarantees as
560 ext3, and avoids the "zero-length" problem that can happen when a system
561 crashes before the delayed allocation blocks are forced to disk.
564 Do not initialize any uninitialized inode table blocks in the background. This
565 feature may be used by installation CD's so that the install process can
566 complete as quickly as possible; the inode table initialization process would
567 then be deferred until the next time the filesystem is mounted.
570 The lazy itable init code will wait n times the number of milliseconds it took
571 to zero out the previous block group's inode table. This minimizes the impact on
572 system performance while the filesystem's inode table is being initialized.
574 .BR discard / nodiscard
575 Controls whether ext4 should issue discard/TRIM commands to the underlying
576 block device when blocks are freed. This is useful for SSD devices and
577 sparse/thinly-provisioned LUNs, but it is off by default until sufficient
578 testing has been done.
581 Disables 32-bit UIDs and GIDs. This is for
582 interoperability with older kernels which only
583 store and expect 16-bit values.
585 .BR block_validity / noblock_validity
586 This options allows to enables/disables the in-kernel facility for tracking
587 filesystem metadata blocks within internal data structures. This allows multi-\c
588 block allocator and other routines to quickly locate extents which might
589 overlap with filesystem metadata blocks. This option is intended for debugging
590 purposes and since it negatively affects the performance, it is off by default.
592 .BR dioread_lock / dioread_nolock
593 Controls whether or not ext4 should use the DIO read locking. If the
594 dioread_nolock option is specified ext4 will allocate uninitialized extent
595 before buffer write and convert the extent to initialized after IO completes.
596 This approach allows ext4 code to avoid using inode mutex, which improves
597 scalability on high speed storages. However this does not work with data
598 journaling and dioread_nolock option will be ignored with kernel warning.
599 Note that dioread_nolock code path is only used for extent-based files.
600 Because of the restrictions this options comprises it is off by default
601 (e.g.\& dioread_lock).
604 This limits the size of the directories so that any attempt to expand them
605 beyond the specified limit in kilobytes will cause an ENOSPC error. This is
606 useful in memory-constrained environments, where a very large directory can
607 cause severe performance problems or even provoke the Out Of Memory killer. (For
608 example, if there is only 512\ MB memory available, a 176\ MB directory may
609 seriously cramp the system's style.)
612 Enable 64-bit inode version support. This option is off by default.
615 The ext2, ext3, and ext4 filesystems support setting the following file
616 attributes on Linux systems using the
620 .BR a " - append only"
622 .BR A " - no atime updates"
626 .BR D " - synchronous directory updates"
630 .BR S " - synchronous updates"
632 .BR u " - undeletable"
634 In addition, the ext3 and ext4 filesystems support the following flag:
636 .BR j " - data journaling"
638 Finally, the ext4 filesystem also supports the following flag:
640 .BR e " - extents format"
642 For descriptions of these attribute flags, please refer to the
projects / thirdparty / e2fsprogs.git / blob