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 ext3 \- 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 to handle mount requests for ext2 and ext3 file systems.
22 .SH FILE SYSTEM FEATURES
23 A file system formatted for ext2, ext3, or ext4 can have some
24 collection of the following 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.
33 Enables the file system to be larger than 2^32 blocks. This feature is set
34 automatically, as needed, but it can be useful to specify this feature
35 explicitly if the file system might need to be resized larger than 2^32
36 blocks, even if it was smaller than that threshold when it was
37 originally created. Note that some older kernels and older versions
38 of e2fsprogs will not support file systems with this ext4 feature enabled.
42 This ext4 feature enables clustered block allocation, so that the unit of
43 allocation is a power of two number of blocks. That is, each bit in the
44 what had traditionally been known as the block allocation bitmap now
45 indicates whether a cluster is in use or not, where a cluster is by
46 default composed of 16 blocks. This feature can decrease the time
47 spent on doing block allocation and brings smaller fragmentation, especially
48 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
60 This feature requires that the
66 Use hashed b-trees to speed up name lookups in large directories. This
67 feature is supported by ext3 and ext4 file systems, and is ignored by
72 Normally, ext4 allows an inode to have no more than 65,000 hard links.
73 This applies to regular files as well as directories, which means that
74 there can be no more than 64,998 subdirectories in a directory (because
75 each of the '.' and '..' entries, as well as the directory entry for the
76 directory in its parent directory counts as a hard link). This feature
77 lifts this limit by causing ext4 to use a link count of 1 to indicate
78 that the number of hard links to a directory is not known when the link
79 count might exceed the maximum count limit.
83 This ext4 feature provides file-system level encryption of data blocks
84 and file names. The inode metadata (timestamps, file size, user/group
89 This feature is most useful on file systems with multiple users, or
90 where not all files should be encrypted. In many use cases, especially
91 on single-user systems, encryption at the block device layer using
92 dm-crypt may provide much better security.
96 This feature enables the use of extended attributes. This feature is
97 supported by ext2, ext3, and ext4.
101 This ext4 feature allows the mapping of logical block numbers for a
102 particular inode to physical blocks on the storage device to be stored
103 using an extent tree, which is a more efficient data structure than the
104 traditional indirect block scheme used by the ext2 and ext3 file
105 systems. The use of the extent tree decreases metadata block overhead,
106 improves file system performance, and decreases the needed to run
113 are accepted as valid names for this feature for
114 historical/backwards compatibility reasons.)
118 This ext4 feature reserves a specific amount of space in each inode for
119 extended metadata such as nanosecond timestamps and file creation time,
120 even if the current kernel does not currently need to reserve this much
121 space. Without this feature, the kernel will reserve the amount of
122 space for features it currently needs, and the rest may be
123 consumed by extended attributes.
125 For this feature to be useful the inode size must be 256 bytes in size
130 This feature enables the storage of file type information in directory
131 entries. This feature is supported by ext2, ext3, and ext4.
135 This ext4 feature allows the per-block group metadata (allocation
138 to be placed anywhere on the storage media. In addition,
140 will place the per-block group metadata together starting at the first
141 block group of each "flex_bg group". The size of the flex_bg group
142 can be specified using the
148 Create a journal to ensure filesystem consistency even across unclean
149 shutdowns. Setting the filesystem feature is equivalent to using the
152 .BR mke2fs " or " tune2fs.
153 This feature is supported by ext3 and ext4, and ignored by the
154 ext2 file system driver.
158 This ext4 feature allows files to be larger than 2 terabytes in size.
161 Allow data to be stored in the inode and extended attribute area.
165 This feature is enabled on the superblock found on an external journal
166 device. The block size for the external journal must be the same as the
167 file system which uses it.
169 The external journal device can be used by a file system by specifying
172 .BR device= <external-device>
180 This feature flag is set automatically by modern kernels when a file
181 larger than 2 gigabytes is created. Very old kernels could not
182 handle large files, so this feature flag was used to prohibit those
183 kernels from mounting file systems that they could not understand.
187 This ext4 feature enables metadata checksumming. This feature stores
188 checksums for all of the filesystem metadata (superblock, group
189 descriptor blocks, inode and block bitmaps, directories, and
190 extent tree blocks). The checksum algorithm used for the metadata
191 blocks is different than the one used for group descriptors with the
193 feature. These two features are incompatible and
195 will be used preferentially instead of
198 .B metadata_csum_seed
200 This feature allows the filesystem to store the metadata checksum seed in the
201 superblock, which allows the administrator to change the UUID of a filesystem
204 feature while it is mounted.
208 This ext4 feature allows file systems to be resized on-line without explicitly
209 needing to reserve space for growth in the size of the block group
210 descriptors. This scheme is also used to resize file systems which are
211 larger than 2^32 blocks. It is not recommended that this feature be set
212 when a file system is created, since this alternate method of storing
213 the block group descriptors will slow down the time needed to mount the
214 file system, and newer kernels can automatically set this feature as
215 necessary when doing an online resize and no more reserved space is
216 available in the resize inode.
220 This ext4 feature provides multiple mount protection (MMP). MMP helps to
221 protect the filesystem from being multiply mounted and is useful in
222 shared storage environments.
226 This ext4 feature provides project quota support. With this feature,
227 the project ID of inode will be managed when the filesystem is mounted.
231 Create quota inodes (inode #3 for userquota and inode
232 #4 for group quota) and set them in the superblock.
233 With this feature, the quotas will be enabled
234 automatically when the filesystem is mounted.
236 Causes the quota files (i.e., user.quota and
237 group.quota which existed
238 in the older quota design) to be hidden inodes.
242 This file system feature indicates that space has been reserved so that
243 the block group descriptor table can be extended while resizing a mounted
244 file system. The online resize operation
245 is carried out by the kernel, triggered by
249 will attempt to reserve enough space so that the
250 filesystem may grow to 1024 times its initial size. This can be changed
255 This feature requires that the
263 This file system feature is set on all modern ext2, ext3, and ext4 file
264 systems. It indicates that backup copies of the superblock and block
265 group descriptors are present only in a few block groups, not all of
270 This feature indicates that there will only be at most two backup
271 superblocks and block group descriptors. The block groups used to store
272 the backup superblock(s) and blockgroup descriptor(s) are stored in the
273 superblock, but typically, one will be located at the beginning of block
274 group #1, and one in the last block group in the file system. This
275 feature is essentially a more extreme version of sparse_super and is
276 designed to allow a much larger percentage of the disk to have
277 contiguous blocks available for data files.
281 This ext4 file system feature indicates that the block group descriptors
282 will be protected using checksums, making it safe for
284 to create a file system without initializing all of the block groups.
285 The kernel will keep a high watermark of unused inodes, and initialize
286 inode tables and blocks lazily. This feature speeds up the time to check
287 the file system using
289 and it also speeds up the time required for
291 to create the file system.
293 This section describes mount options which are specific to ext2, ext3,
294 and ext4. Other generic mount options may be used as well; see
297 .SH "Mount options for ext2"
298 The `ext2' filesystem is the standard Linux filesystem.
299 Since Linux 2.5.46, for most mount options the default
300 is determined by the filesystem superblock. Set them with
304 Support POSIX Access Control Lists (or not). See the
309 Set the behavior for the
313 behavior is to return in the
315 field the total number of blocks of the filesystem, while the
317 behavior (which is the default) is to subtract the overhead blocks
318 used by the ext2 filesystem and not available for file storage. Thus
320 % mount /k \-o minixdf; df /k; umount /k
325 Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
326 /dev/sda6#2630655#86954#2412169#3%#/k
329 % mount /k \-o bsddf; df /k; umount /k
334 Filesystem#1024-blocks#Used#Available#Capacity#Mounted on
335 /dev/sda6#2543714#13#2412169#0%#/k
338 (Note that this example shows that one can add command line options
339 to the options given in
342 .BR check=none " or " nocheck
343 No checking is done at mount time. This is the default. This is fast.
346 every now and then, e.g.\& at boot time. The non-default behavior is unsupported
347 (check=normal and check=strict options have been removed). Note that these mount options
348 don't have to be supported if ext4 kernel driver is used for ext2 and ext3 filesystems.
351 Print debugging info upon each (re)mount.
353 .BR errors= { continue | remount-ro | panic }
354 Define the behavior when an error is encountered.
355 (Either ignore errors and just mark the filesystem erroneous and continue,
356 or remount the filesystem read-only, or panic and halt the system.)
357 The default is set in the filesystem superblock, and can be
361 .BR grpid | bsdgroups " and " nogrpid | sysvgroups
362 These options define what group id a newly created file gets.
365 is set, it takes the group id of the directory in which it is created;
366 otherwise (the default) it takes the fsgid of the current process, unless
367 the directory has the setgid bit set, in which case it takes the gid
368 from the parent directory, and also gets the setgid bit set
369 if it is a directory itself.
371 .BR grpquota | noquota | quota | usrquota
372 The usrquota (same as quota) mount option enables user quota support on the
373 filesystem. grpquota enables group quotas support. You need the quota utilities
374 to actually enable and manage the quota system.
377 Disables 32-bit UIDs and GIDs. This is for interoperability with older
378 kernels which only store and expect 16-bit values.
380 .BR oldalloc " or " orlov
381 Use old allocator or Orlov allocator for new inodes. Orlov is default.
383 \fBresgid=\fP\,\fIn\fP and \fBresuid=\fP\,\fIn\fP
384 The ext2 filesystem reserves a certain percentage of the available
385 space (by default 5%, see
389 These options determine who can use the reserved blocks.
390 (Roughly: whoever has the specified uid, or belongs to the specified group.)
393 Instead of using the normal superblock, use an alternative superblock
396 This option is normally used when the primary superblock has been
397 corrupted. The location of backup superblocks is dependent on the
398 filesystem's blocksize, the number of blocks per group, and features
402 Additional backup superblocks can be determined by using the
406 option to print out where the superblocks exist, supposing
408 is supplied with arguments that are consistent with the filesystem's layout
409 (e.g. blocksize, blocks per group,
413 The block number here uses 1\ k units. Thus, if you want to use logical
414 block 32768 on a filesystem with 4\ k blocks, use "sb=131072".
416 .BR user_xattr | nouser_xattr
417 Support "user." extended attributes (or not).
420 .SH "Mount options for ext3"
421 The ext3 filesystem is a version of the ext2 filesystem which has been
422 enhanced with journaling. It supports the same options as ext2 as
423 well as the following additions:
425 .BR journal_dev=devnum / journal_path=path
426 When the external journal device's major/minor numbers
427 have changed, these options allow the user to specify
428 the new journal location. The journal device is
429 identified either through its new major/minor numbers encoded
430 in devnum, or via a path to the device.
432 .BR norecovery / noload
433 Don't load the journal on mounting. Note that
434 if the filesystem was not unmounted cleanly,
435 skipping the journal replay will lead to the
436 filesystem containing inconsistencies that can
437 lead to any number of problems.
439 .BR data= { journal | ordered | writeback }
440 Specifies the journaling mode for file data. Metadata is always journaled.
441 To use modes other than
443 on the root filesystem, pass the mode to the kernel as boot parameter, e.g.\&
444 .IR rootflags=data=journal .
448 All data is committed into the journal prior to being written into the
452 This is the default mode. All data is forced directly out to the main file
453 system prior to its metadata being committed to the journal.
456 Data ordering is not preserved \(en data may be written into the main
457 filesystem after its metadata has been committed to the journal.
458 This is rumoured to be the highest-throughput option. It guarantees
459 internal filesystem integrity, however it can allow old data to appear
460 in files after a crash and journal recovery.
464 Just print an error message if an error occurs in a file data buffer in
468 Abort the journal if an error occurs in a file data buffer in ordered mode.
470 .BR barrier=0 " / " barrier=1 "
471 This disables / enables the use of write barriers in the jbd code. barrier=0
472 disables, barrier=1 enables (default). This also requires an IO stack which can
473 support barriers, and if jbd gets an error on a barrier write, it will disable
474 barriers again with a warning. Write barriers enforce proper on-disk ordering
475 of journal commits, making volatile disk write caches safe to use, at some
476 performance penalty. If your disks are battery-backed in one way or another,
477 disabling barriers may safely improve performance.
480 Start a journal commit every
482 seconds. The default value is 5 seconds. Zero means default.
485 Enable Extended User Attributes. See the
489 .BR jqfmt= { vfsold | vfsv0 | vfsv1 }
490 Apart from the old quota system (as in ext2, jqfmt=vfsold aka version 1 quota)
491 ext3 also supports journaled quotas (version 2 quota). jqfmt=vfsv0 or
492 jqfmt=vfsv1 enables journaled quotas. Journaled quotas have the advantage that
493 even after a crash no quota check is required. When the
495 filesystem feature is enabled, journaled quotas are used automatically, and
496 this mount option is ignored.
498 .BR usrjquota=aquota.user | grpjquota=aquota.group
499 For journaled quotas (jqfmt=vfsv0 or jqfmt=vfsv1), the mount options
500 usrjquota=aquota.user and grpjquota=aquota.group are required to tell the
501 quota system which quota database files to use. When the
503 filesystem feature is enabled, journaled quotas are used automatically, and
504 this mount option is ignored.
506 .SH "Mount options for ext4"
507 The ext4 filesystem is an advanced level of the ext3 filesystem which
508 incorporates scalability and reliability enhancements for supporting large
512 .B journal_dev, journal_path, norecovery, noload, data, commit, orlov,
513 .B oldalloc, [no]user_xattr, [no]acl, bsddf, minixdf, debug, errors,
514 .B data_err, grpid, bsdgroups, nogrpid, sysvgroups, resgid, resuid, sb,
515 .B quota, noquota, nouid32, grpquota, usrquota, usrjquota, grpjquota,
516 .B and jqfmt are backwardly compatible with ext3 or ext2.
518 .B journal_checksum | nojournal_checksum
519 The journal_checksum option enables checksumming of the journal transactions.
520 This will allow the recovery code in e2fsck and the kernel to detect corruption
521 in the kernel. It is a compatible change and will be ignored by older kernels.
523 .B journal_async_commit
524 Commit block can be written to disk without waiting for descriptor blocks. If
525 enabled older kernels cannot mount the device.
526 This will enable 'journal_checksum' internally.
528 .BR barrier=0 " / " barrier=1 " / " barrier " / " nobarrier
529 These mount options have the same effect as in ext3. The mount options
530 "barrier" and "nobarrier" are added for consistency with other ext4 mount
533 The ext4 filesystem enables write barriers by default.
535 .BI inode_readahead_blks= n
536 This tuning parameter controls the maximum number of inode table blocks that
537 ext4's inode table readahead algorithm will pre-read into the buffer cache.
538 The value must be a power of 2. The default value is 32 blocks.
541 Number of filesystem blocks that mballoc will try to use for allocation size
542 and alignment. For RAID5/6 systems this should be the number of data disks *
543 RAID chunk size in filesystem blocks.
546 Deferring block allocation until write-out time.
549 Disable delayed allocation. Blocks are allocated when data is copied from user
552 .BI max_batch_time= usec
553 Maximum amount of time ext4 should wait for additional filesystem operations to
554 be batch together with a synchronous write operation. Since a synchronous
555 write operation is going to force a commit and then a wait for the I/O
556 complete, it doesn't cost much, and can be a huge throughput win, we wait for a
557 small amount of time to see if any other transactions can piggyback on the
558 synchronous write. The algorithm used is designed to automatically tune for
559 the speed of the disk, by measuring the amount of time (on average) that it
560 takes to finish committing a transaction. Call this time the "commit time".
561 If the time that the transaction has been running is less than the commit time,
562 ext4 will try sleeping for the commit time to see if other operations will join
563 the transaction. The commit time is capped by the max_batch_time, which
564 defaults to 15000\ \[mc]s (15\ ms). This optimization can be turned off entirely by
565 setting max_batch_time to 0.
567 .BI min_batch_time= usec
568 This parameter sets the commit time (as described above) to be at least
569 min_batch_time. It defaults to zero microseconds. Increasing this parameter
570 may improve the throughput of multi-threaded, synchronous workloads on very
571 fast disks, at the cost of increasing latency.
573 .BI journal_ioprio= prio
574 The I/O priority (from 0 to 7, where 0 is the highest priority) which should be
575 used for I/O operations submitted by kjournald2 during a commit operation.
576 This defaults to 3, which is a slightly higher priority than the default I/O
580 Simulate the effects of calling ext4_abort() for
581 debugging purposes. This is normally used while
582 remounting a filesystem which is already mounted.
584 .BR auto_da_alloc | noauto_da_alloc
585 Many broken applications don't use fsync() when
586 replacing existing files via patterns such as
588 fd = open("foo.new")/write(fd,...)/close(fd)/ rename("foo.new", "foo")
592 fd = open("foo", O_TRUNC)/write(fd,...)/close(fd).
594 If auto_da_alloc is enabled, ext4 will detect the replace-via-rename and
595 replace-via-truncate patterns and force that any delayed allocation blocks are
596 allocated such that at the next journal commit, in the default data=ordered
597 mode, the data blocks of the new file are forced to disk before the rename()
598 operation is committed. This provides roughly the same level of guarantees as
599 ext3, and avoids the "zero-length" problem that can happen when a system
600 crashes before the delayed allocation blocks are forced to disk.
603 Do not initialize any uninitialized inode table blocks in the background. This
604 feature may be used by installation CD's so that the install process can
605 complete as quickly as possible; the inode table initialization process would
606 then be deferred until the next time the filesystem is mounted.
609 The lazy itable init code will wait n times the number of milliseconds it took
610 to zero out the previous block group's inode table. This minimizes the impact on
611 system performance while the filesystem's inode table is being initialized.
613 .BR discard / nodiscard
614 Controls whether ext4 should issue discard/TRIM commands to the underlying
615 block device when blocks are freed. This is useful for SSD devices and
616 sparse/thinly-provisioned LUNs, but it is off by default until sufficient
617 testing has been done.
619 .BR block_validity / noblock_validity
620 This option enables/disables the in-kernel facility for tracking
621 filesystem metadata blocks within internal data structures. This allows multi-\c
622 block allocator and other routines to quickly locate extents which might
623 overlap with filesystem metadata blocks. This option is intended for debugging
624 purposes and since it negatively affects the performance, it is off by default.
626 .BR dioread_lock / dioread_nolock
627 Controls whether or not ext4 should use the DIO read locking. If the
628 dioread_nolock option is specified ext4 will allocate uninitialized extent
629 before buffer write and convert the extent to initialized after IO completes.
630 This approach allows ext4 code to avoid using inode mutex, which improves
631 scalability on high speed storages. However this does not work with data
632 journaling and dioread_nolock option will be ignored with kernel warning.
633 Note that dioread_nolock code path is only used for extent-based files.
634 Because of the restrictions this options comprises it is off by default
635 (e.g.\& dioread_lock).
638 This limits the size of the directories so that any attempt to expand them
639 beyond the specified limit in kilobytes will cause an ENOSPC error. This is
640 useful in memory-constrained environments, where a very large directory can
641 cause severe performance problems or even provoke the Out Of Memory killer. (For
642 example, if there is only 512\ MB memory available, a 176\ MB directory may
643 seriously cramp the system's style.)
646 Enable 64-bit inode version support. This option is off by default.
649 This option disables use of mbcache for extended attribute deduplication. On
650 systems where extended attributes are rarely or never shared between files,
651 use of mbcache for deduplication adds unnecessary computational overhead.
654 The prjquota mount option enables project quota support on the filesystem.
655 You need the quota utilities to actually enable and manage the quota system.
656 This mount option requires the
661 The ext2, ext3, and ext4 filesystems support setting the following file
662 attributes on Linux systems using the
666 .BR a " - append only"
668 .BR A " - no atime updates"
672 .BR D " - synchronous directory updates"
676 .BR S " - synchronous updates"
678 .BR u " - undeletable"
680 In addition, the ext3 and ext4 filesystems support the following flag:
682 .BR j " - data journaling"
684 Finally, the ext4 filesystem also supports the following flag:
686 .BR e " - extents format"
688 For descriptions of these attribute flags, please refer to the
692 This section lists the file system driver (e.g., ext2, ext3, ext4) and
693 upstream kernel version where a particular file system feature was
694 supported. Note that in some cases the feature was present in earlier
695 kernel versions, but there were known, serious bugs. In other cases the
696 feature may still be considered in an experimental state. Finally, note
697 that some distributions may have backported features into older kernels;
698 in particular the kernel versions in certain "enterprise distributions"
699 can be extremely misleading.
700 .IP "\fBfiletype\fR" 2in
702 .IP "\fBsparse_super\fR" 2in
704 .IP "\fBlarge_file\fR" 2in
706 .IP "\fBhas_journal\fR" 2in
708 .IP "\fBext_attr\fR" 2in
710 .IP "\fBdir_index\fR" 2in
712 .IP "\fBresize_inode\fR" 2in
713 ext3, 2.6.10 (online resizing)
714 .IP "\fB64bit\fR" 2in
716 .IP "\fBdir_nlink\fR" 2in
718 .IP "\fBextent\fR" 2in
720 .IP "\fBextra_isize\fR" 2in
722 .IP "\fBflex_bg\fR" 2in
724 .IP "\fBhuge_file\fR" 2in
726 .IP "\fBmeta_bg\fR" 2in
728 .IP "\fBuninit_bg\fR" 2in
732 .IP "\fBbigalloc\fR" 2in
734 .IP "\fBquota\fR" 2in
736 .IP "\fBinline_data\fR" 2in
738 .IP "\fBsparse_super2\fR" 2in
740 .IP "\fBmetadata_csum\fR" 2in
742 .IP "\fBencrypt\fR" 2in
744 .IP "\fBmetadata_csum_seed\fR" 2in
746 .IP "\fBproject\fR" 2in
748 .IP "\fBea_inode\fR" 2in
750 .IP "\fBlarge_dir\fR" 2in
git://git.whamcloud.com / tools / e2fsprogs.git / blob