1EXT4(5)                       File Formats Manual                      EXT4(5)
2
3
4

NAME

6       ext2 - the second extended file system
7       ext3 - the third extended file system
8       ext4 - the fourth extended file system
9

DESCRIPTION

11       The second, third, and fourth extended file systems, or ext2, ext3, and
12       ext4 as they are commonly known, are Linux file systems that have  his‐
13       torically  been  the  default file system for many Linux distributions.
14       They are general purpose file  systems  that  have  been  designed  for
15       extensibility and backwards compatibility.  In particular, file systems
16       previously intended for use with the ext2 and ext3 file systems can  be
17       mounted  using  the  ext4 file system driver, and indeed in many modern
18       Linux distributions, the ext4 file system driver has been configured to
19       handle mount requests for ext2 and ext3 file systems.
20

FILE SYSTEM FEATURES

22       A  file  system formatted for ext2, ext3, or ext4 can have some collec‐
23       tion of the following file system feature flags enabled.  Some of these
24       features  are  not  supported by all implementations of the ext2, ext3,
25       and ext4 file system drivers, depending on Linux kernel version in use.
26       On  other  operating  systems,  such as the GNU/HURD or FreeBSD, only a
27       very restrictive set of file system features may be supported in  their
28       implementations of ext2.
29
30       64bit
31              Enables  the  file  system  to be larger than 2^32 blocks.  This
32              feature is set automatically, as needed, but it can be useful to
33              specify this feature explicitly if the file system might need to
34              be resized larger than 2^32 blocks, even if it was smaller  than
35              that  threshold  when it was originally created.  Note that some
36              older kernels and older versions of e2fsprogs will  not  support
37              file systems with this ext4 feature enabled.
38
39       bigalloc
40              This  ext4  feature  enables clustered block allocation, so that
41              the unit of allocation is a power of two number of blocks.  That
42              is,  each  bit  in  the what had traditionally been known as the
43              block allocation bitmap now indicates whether a  cluster  is  in
44              use or not, where a cluster is by default composed of 16 blocks.
45              This feature can decrease the time spent on doing block  alloca‐
46              tion  and  brings  smaller  fragmentation,  especially for large
47              files.  The size can be specified using the mke2fs -C option.
48
49              Warning: The bigalloc feature is still  under  development,  and
50              may  not be fully supported with your kernel or may have various
51              bugs.    Please   see   the   web   page   http://ext4.wiki.ker
52              nel.org/index.php/Bigalloc  for details.  May clash with delayed
53              allocation (see nodelalloc mount option).
54
55              This feature requires that the extent feature be enabled.
56
57       dir_index
58              Use hashed b-trees to speed up name lookups  in  large  directo‐
59              ries.   This feature is supported by ext3 and ext4 file systems,
60              and is ignored by ext2 file systems.
61
62       dir_nlink
63              Normally, ext4 allows an inode to have no more than 65,000  hard
64              links.   This  applies  to regular files as well as directories,
65              which means that there can be no more than 64,998 subdirectories
66              in  a  directory  (because  each of the '.' and '..' entries, as
67              well as the directory entry for  the  directory  in  its  parent
68              directory counts as a hard link).  This feature lifts this limit
69              by causing ext4 to use a link count of 1 to  indicate  that  the
70              number  of  hard links to a directory is not known when the link
71              count might exceed the maximum count limit.
72
73       ea_inode
74              Normally, a file's extended attributes and  associated  metadata
75              must  fit  within  the  inode or the inode's associated extended
76              attribute block. This feature allows the value of each  extended
77              attribute to be placed in the data blocks of a separate inode if
78              necessary, increasing the  limit  on  the  size  and  number  of
79              extended attributes per file.
80
81       encrypt
82              This  ext4 feature provides file-system level encryption of data
83              blocks and file names.  The  inode  metadata  (timestamps,  file
84              size, user/group ownership, etc.) is not encrypted.
85
86              This feature is most useful on file systems with multiple users,
87              or where not all files should be encrypted.  In many use  cases,
88              especially  on  single-user  systems,  encryption  at  the block
89              device layer using dm-crypt may provide much better security.
90
91       ext_attr
92              This feature enables the use of extended attributes.  This  fea‐
93              ture is supported by ext2, ext3, and ext4.
94
95       extent
96              This  ext4  feature  allows the mapping of logical block numbers
97              for a particular inode to physical blocks on the storage  device
98              to  be  stored  using  an extent tree, which is a more efficient
99              data structure than the traditional indirect block  scheme  used
100              by  the  ext2 and ext3 file systems.  The use of the extent tree
101              decreases metadata block overhead, improves file system  perfor‐
102              mance,  and  decreases  the  needed to run e2fsck(8) on the file
103              system.  (Note: both extent and extents are  accepted  as  valid
104              names  for  this  feature for historical/backwards compatibility
105              reasons.)
106
107       extra_isize
108              This ext4 feature reserves a specific amount of  space  in  each
109              inode  for  extended  metadata such as nanosecond timestamps and
110              file creation time, even if the current  kernel  does  not  cur‐
111              rently  need  to reserve this much space.  Without this feature,
112              the kernel will reserve the amount of space for features it cur‐
113              rently   needs,  and  the  rest  may  be  consumed  by  extended
114              attributes.
115
116              For this feature to be useful the inode size must be  256  bytes
117              in size or larger.
118
119       filetype
120              This  feature  enables  the  storage of file type information in
121              directory entries.  This feature is supported by ext2, ext3, and
122              ext4.
123
124       flex_bg
125              This  ext4  feature allows the per-block group metadata (alloca‐
126              tion bitmaps and inode tables) to  be  placed  anywhere  on  the
127              storage  media.   In  addition,  mke2fs will place the per-block
128              group metadata together starting at the  first  block  group  of
129              each  "flex_bg  group".    The  size of the flex_bg group can be
130              specified using the -G option.
131
132       has_journal
133              Create a journal to ensure filesystem  consistency  even  across
134              unclean shutdowns.  Setting the filesystem feature is equivalent
135              to using the -j option with mke2fs or tune2fs.  This feature  is
136              supported  by ext3 and ext4, and ignored by the ext2 file system
137              driver.
138
139       huge_file
140              This ext4 feature allows files to be larger than 2 terabytes  in
141              size.
142
143       inline_data
144              Allow  data  to  be  stored  in the inode and extended attribute
145              area.
146
147       journal_dev
148              This feature is enabled on the superblock found on  an  external
149              journal device.  The block size for the external journal must be
150              the same as the file system which uses it.
151
152              The external journal device can be used  by  a  file  system  by
153              specifying  the  -J device=<external-device> option to mke2fs(8)
154              or tune2fs(8).
155
156       large_dir
157              This feature increases the limit on  the  number  of  files  per
158              directory  by  raising  the maximum size of directories and, for
159              hashed b-tree directories (see dir_index), the maximum height of
160              the hashed b-tree used to store the directory entries.
161
162       large_file
163              This  feature flag is set automatically by modern kernels when a
164              file larger than 2 gigabytes is created.  Very old kernels could
165              not  handle  large  files, so this feature flag was used to pro‐
166              hibit those kernels from mounting file systems that  they  could
167              not understand.
168
169       metadata_csum
170              This  ext4  feature enables metadata checksumming.  This feature
171              stores checksums for all of the filesystem metadata (superblock,
172              group  descriptor  blocks, inode and block bitmaps, directories,
173              and extent tree blocks).  The checksum algorithm  used  for  the
174              metadata  blocks  is  different  than  the  one  used  for group
175              descriptors with the uninit_bg feature.  These two features  are
176              incompatible  and  metadata_csum  will  be  used  preferentially
177              instead of uninit_bg.
178
179       metadata_csum_seed
180              This feature allows the filesystem to store the metadata  check‐
181              sum  seed  in  the superblock, which allows the administrator to
182              change the UUID of a filesystem using the metadata_csum  feature
183              while it is mounted.
184
185       meta_bg
186              This  ext4  feature  allows  file  systems to be resized on-line
187              without explicitly needing to reserve space for  growth  in  the
188              size  of  the block group descriptors.  This scheme is also used
189              to resize file systems which are larger than 2^32 blocks.  It is
190              not  recommended  that this feature be set when a file system is
191              created, since this alternate method of storing the block  group
192              descriptors  will  slow  down  the time needed to mount the file
193              system, and newer kernels can automatically set this feature  as
194              necessary when doing an online resize and no more reserved space
195              is available in the resize inode.
196
197       mmp
198              This ext4 feature provides multiple mount protection (MMP).  MMP
199              helps  to protect the filesystem from being multiply mounted and
200              is useful in shared storage environments.
201
202       project
203              This ext4 feature provides project quota support. With this fea‐
204              ture,  the project ID of inode will be managed when the filesys‐
205              tem is mounted.
206
207       quota
208              Create quota inodes (inode #3 for userquota  and  inode  #4  for
209              group quota) and set them in the superblock.  With this feature,
210              the quotas will be enabled automatically when the filesystem  is
211              mounted.
212
213              Causes  the  quota files (i.e., user.quota and group.quota which
214              existed in the older quota design) to be hidden inodes.
215
216       resize_inode
217              This file system feature indicates that space has been  reserved
218              so  that  the block group descriptor table can be extended while
219              resizing a mounted file system.  The online resize operation  is
220              carried  out  by  the  kernel,  triggered  by  resize2fs(8).  By
221              default mke2fs will attempt to reserve enough space so that  the
222              filesystem may grow to 1024 times its initial size.  This can be
223              changed using the resize extended option.
224
225              This feature requires that  the  sparse_super  or  sparse_super2
226              feature be enabled.
227
228       sparse_super
229              This  file  system  feature is set on all modern ext2, ext3, and
230              ext4 file systems.  It  indicates  that  backup  copies  of  the
231              superblock and block group descriptors are present only in a few
232              block groups, not all of them.
233
234       sparse_super2
235              This feature indicates that there  will  only  be  at  most  two
236              backup  superblocks  and  block  group  descriptors.   The block
237              groups used to store the  backup  superblock(s)  and  blockgroup
238              descriptor(s)  are  stored in the superblock, but typically, one
239              will be located at the beginning of block group #1, and  one  in
240              the last block group in the file system.  This feature is essen‐
241              tially a more extreme version of sparse_super and is designed to
242              allow  a  much  larger percentage of the disk to have contiguous
243              blocks available for data files.
244
245       uninit_bg
246              This ext4 file system feature indicates  that  the  block  group
247              descriptors  will  be  protected using checksums, making it safe
248              for mke2fs(8) to create a file system without  initializing  all
249              of  the  block groups.  The kernel will keep a high watermark of
250              unused inodes, and initialize inode tables  and  blocks  lazily.
251              This  feature  speeds up the time to check the file system using
252              e2fsck(8), and it also speeds up the time required for mke2fs(8)
253              to create the file system.
254

MOUNT OPTIONS

256       This  section describes mount options which are specific to ext2, ext3,
257       and ext4.  Other generic  mount  options  may  be  used  as  well;  see
258       mount(8) for details.
259

Mount options for ext2

261       The  `ext2'  filesystem  is the standard Linux filesystem.  Since Linux
262       2.5.46, for most  mount  options  the  default  is  determined  by  the
263       filesystem superblock. Set them with tune2fs(8).
264
265       acl|noacl
266              Support  POSIX  Access  Control  Lists (or not).  See the acl(5)
267              manual page.
268
269       bsddf|minixdf
270              Set the behavior for the statfs system call. The minixdf  behav‐
271              ior  is  to  return  in  the  f_blocks field the total number of
272              blocks of the filesystem, while the bsddf behavior (which is the
273              default)  is  to  subtract  the overhead blocks used by the ext2
274              filesystem and not available for file storage. Thus
275
276              % mount /k -o minixdf; df /k; umount /k
277
278              Filesystem  1024-blocks   Used  Available  Capacity  Mounted on
279              /dev/sda6     2630655    86954   2412169      3%     /k
280
281              % mount /k -o bsddf; df /k; umount /k
282
283              Filesystem  1024-blocks  Used  Available  Capacity  Mounted on
284              /dev/sda6     2543714      13   2412169      0%     /k
285
286              (Note that this example shows that  one  can  add  command  line
287              options to the options given in /etc/fstab.)
288
289       check=none or nocheck
290              No  checking is done at mount time. This is the default. This is
291              fast.  It is wise to invoke e2fsck(8) every now and  then,  e.g.
292              at   boot   time.   The   non-default  behavior  is  unsupported
293              (check=normal and check=strict options have been removed).  Note
294              that these mount options don't have to be supported if ext4 ker‐
295              nel driver is used for ext2 and ext3 filesystems.
296
297       debug  Print debugging info upon each (re)mount.
298
299       errors={continue|remount-ro|panic}
300              Define the behavior  when  an  error  is  encountered.   (Either
301              ignore  errors  and  just mark the filesystem erroneous and con‐
302              tinue, or remount the filesystem read-only, or  panic  and  halt
303              the  system.)   The default is set in the filesystem superblock,
304              and can be changed using tune2fs(8).
305
306       grpid|bsdgroups and nogrpid|sysvgroups
307              These options define what group id a newly  created  file  gets.
308              When  grpid  is  set,  it takes the group id of the directory in
309              which it is created; otherwise (the default) it takes the  fsgid
310              of  the current process, unless the directory has the setgid bit
311              set, in which case it takes the gid from the  parent  directory,
312              and also gets the setgid bit set if it is a directory itself.
313
314       grpquota|noquota|quota|usrquota
315              The  usrquota  (same  as  quota) mount option enables user quota
316              support on the filesystem. grpquota enables  group  quotas  sup‐
317              port. You need the quota utilities to actually enable and manage
318              the quota system.
319
320       nouid32
321              Disables 32-bit UIDs and GIDs.   This  is  for  interoperability
322              with older kernels which only store and expect 16-bit values.
323
324       oldalloc or orlov
325              Use  old  allocator  or Orlov allocator for new inodes. Orlov is
326              default.
327
328       resgid=n and resuid=n
329              The ext2 filesystem reserves a certain percentage of the  avail‐
330              able space (by default 5%, see mke2fs(8) and tune2fs(8)).  These
331              options determine who can use the  reserved  blocks.   (Roughly:
332              whoever  has  the  specified  uid,  or  belongs to the specified
333              group.)
334
335       sb=n   Instead of using  the  normal  superblock,  use  an  alternative
336              superblock  specified  by  n.  This option is normally used when
337              the primary superblock has  been  corrupted.   The  location  of
338              backup  superblocks  is dependent on the filesystem's blocksize,
339              the  number  of  blocks  per  group,  and   features   such   as
340              sparse_super.
341
342              Additional  backup  superblocks  can  be determined by using the
343              mke2fs program using the  -n  option  to  print  out  where  the
344              superblocks  exist,  supposing mke2fs is supplied with arguments
345              that are consistent with the filesystem's  layout  (e.g.  block‐
346              size, blocks per group, sparse_super, etc.).
347
348              The  block  number here uses 1 k units. Thus, if you want to use
349              logical block  32768  on  a  filesystem  with  4 k  blocks,  use
350              "sb=131072".
351
352       user_xattr|nouser_xattr
353              Support "user." extended attributes (or not).
354
355
356

Mount options for ext3

358       The  ext3 filesystem is a version of the ext2 filesystem which has been
359       enhanced with journaling.  It supports the same options as ext2 as well
360       as the following additions:
361
362       journal_dev=devnum/journal_path=path
363              When  the  external  journal  device's  major/minor numbers have
364              changed, these options allow the user to specify the new journal
365              location.   The  journal device is identified either through its
366              new major/minor numbers encoded in devnum, or via a path to  the
367              device.
368
369       norecovery/noload
370              Don't load the journal on mounting.  Note that if the filesystem
371              was not unmounted cleanly, skipping the journal replay will lead
372              to  the  filesystem  containing inconsistencies that can lead to
373              any number of problems.
374
375       data={journal|ordered|writeback}
376              Specifies the journaling mode for file data.  Metadata is always
377              journaled.  To use modes other than ordered on the root filesys‐
378              tem, pass the mode to the kernel as boot parameter,  e.g.  root‐
379              flags=data=journal.
380
381              journal
382                     All  data  is  committed  into the journal prior to being
383                     written into the main filesystem.
384
385              ordered
386                     This is the default mode.  All data  is  forced  directly
387                     out  to  the main file system prior to its metadata being
388                     committed to the journal.
389
390              writeback
391                     Data ordering is not preserved – data may be written into
392                     the main filesystem after its metadata has been committed
393                     to the journal.  This is  rumoured  to  be  the  highest-
394                     throughput  option.   It  guarantees  internal filesystem
395                     integrity, however it can allow old  data  to  appear  in
396                     files after a crash and journal recovery.
397
398       data_err=ignore
399              Just  print  an  error message if an error occurs in a file data
400              buffer in ordered mode.
401
402       data_err=abort
403              Abort the journal if an error occurs in a file  data  buffer  in
404              ordered mode.
405
406       barrier=0 / barrier=1
407              This  disables  /  enables  the use of write barriers in the jbd
408              code.  barrier=0 disables,  barrier=1  enables  (default).  This
409              also requires an IO stack which can support barriers, and if jbd
410              gets an error on a barrier write, it will disable barriers again
411              with  a warning.  Write barriers enforce proper on-disk ordering
412              of journal commits, making volatile disk write  caches  safe  to
413              use,  at  some  performance penalty.  If your disks are battery-
414              backed in one way or  another,  disabling  barriers  may  safely
415              improve performance.
416
417       commit=nrsec
418              Start  a  journal commit every nrsec seconds.  The default value
419              is 5 seconds.  Zero means default.
420
421       user_xattr
422              Enable Extended User Attributes. See the attr(5) manual page.
423
424       jqfmt={vfsold|vfsv0|vfsv1}
425              Apart from the old quota system (as in  ext2,  jqfmt=vfsold  aka
426              version  1 quota) ext3 also supports journaled quotas (version 2
427              quota). jqfmt=vfsv0 or  jqfmt=vfsv1  enables  journaled  quotas.
428              Journaled  quotas  have the advantage that even after a crash no
429              quota check is required. When the quota  filesystem  feature  is
430              enabled, journaled quotas are used automatically, and this mount
431              option is ignored.
432
433       usrjquota=aquota.user|grpjquota=aquota.group
434              For journaled quotas (jqfmt=vfsv0  or  jqfmt=vfsv1),  the  mount
435              options  usrjquota=aquota.user  and  grpjquota=aquota.group  are
436              required to tell the quota system which quota database files  to
437              use.  When  the  quota  filesystem feature is enabled, journaled
438              quotas are used automatically, and this mount option is ignored.
439
440

Mount options for ext4

442       The ext4 filesystem is an advanced level of the ext3  filesystem  which
443       incorporates  scalability  and  reliability enhancements for supporting
444       large filesystem.
445
446       The options journal_dev, journal_path, norecovery, noload,  data,  com‐
447       mit,  orlov,  oldalloc, [no]user_xattr, [no]acl, bsddf, minixdf, debug,
448       errors,  data_err,  grpid,  bsdgroups,  nogrpid,  sysvgroups,   resgid,
449       resuid,  sb,  quota,  noquota,  nouid32, grpquota, usrquota, usrjquota,
450       grpjquota, and jqfmt are backwardly compatible with ext3 or ext2.
451
452       journal_checksum | nojournal_checksum
453              The journal_checksum option enables checksumming of the  journal
454              transactions.   This  will allow the recovery code in e2fsck and
455              the kernel to detect corruption in the kernel. It is a  compati‐
456              ble change and will be ignored by older kernels.
457
458       journal_async_commit
459              Commit block can be written to disk without waiting for descrip‐
460              tor blocks. If enabled older kernels cannot  mount  the  device.
461              This will enable 'journal_checksum' internally.
462
463       barrier=0 / barrier=1 / barrier / nobarrier
464              These  mount options have the same effect as in ext3.  The mount
465              options "barrier" and "nobarrier" are added for consistency with
466              other ext4 mount options.
467
468              The ext4 filesystem enables write barriers by default.
469
470       inode_readahead_blks=n
471              This tuning parameter controls the maximum number of inode table
472              blocks that ext4's inode table readahead algorithm will pre-read
473              into  the  buffer  cache.   The  value must be a power of 2. The
474              default value is 32 blocks.
475
476       stripe=n
477              Number of filesystem blocks that mballoc will  try  to  use  for
478              allocation  size  and alignment. For RAID5/6 systems this should
479              be the number of data disks *  RAID  chunk  size  in  filesystem
480              blocks.
481
482       delalloc
483              Deferring block allocation until write-out time.
484
485       nodelalloc
486              Disable  delayed  allocation.  Blocks are allocated when data is
487              copied from user to page cache.
488
489       max_batch_time=usec
490              Maximum amount of time ext4 should wait for additional  filesys‐
491              tem  operations  to  be  batch together with a synchronous write
492              operation. Since a synchronous write operation is going to force
493              a  commit  and then a wait for the I/O complete, it doesn't cost
494              much, and can be a huge throughput win,  we  wait  for  a  small
495              amount of time to see if any other transactions can piggyback on
496              the synchronous write. The algorithm used is designed  to  auto‐
497              matically  tune  for  the  speed  of  the disk, by measuring the
498              amount of time (on average) that it takes to finish committing a
499              transaction. Call this time the "commit time".  If the time that
500              the transaction has been running is less than the  commit  time,
501              ext4 will try sleeping for the commit time to see if other oper‐
502              ations will join the transaction. The commit time is  capped  by
503              the  max_batch_time,  which  defaults  to 15000 µs (15 ms). This
504              optimization   can   be   turned   off   entirely   by   setting
505              max_batch_time to 0.
506
507       min_batch_time=usec
508              This  parameter  sets the commit time (as described above) to be
509              at least  min_batch_time.  It  defaults  to  zero  microseconds.
510              Increasing  this  parameter may improve the throughput of multi-
511              threaded, synchronous workloads on very fast disks, at the  cost
512              of increasing latency.
513
514       journal_ioprio=prio
515              The  I/O priority (from 0 to 7, where 0 is the highest priority)
516              which should be used for I/O operations submitted by  kjournald2
517              during  a  commit  operation.   This  defaults  to 3, which is a
518              slightly higher priority than the default I/O priority.
519
520       abort  Simulate the effects of calling ext4_abort() for debugging  pur‐
521              poses.   This  is  normally  used  while remounting a filesystem
522              which is already mounted.
523
524       auto_da_alloc|noauto_da_alloc
525              Many broken applications don't use fsync() when replacing exist‐
526              ing files via patterns such as
527
528              fd  = open("foo.new")/write(fd,...)/close(fd)/ rename("foo.new",
529              "foo")
530
531              or worse yet
532
533              fd = open("foo", O_TRUNC)/write(fd,...)/close(fd).
534
535              If auto_da_alloc is enabled, ext4 will detect  the  replace-via-
536              rename  and  replace-via-truncate  patterns  and  force that any
537              delayed allocation blocks are allocated such that  at  the  next
538              journal  commit,  in  the  default  data=ordered  mode, the data
539              blocks of the new file are forced to disk  before  the  rename()
540              operation is committed.  This provides roughly the same level of
541              guarantees as ext3, and avoids the  "zero-length"  problem  that
542              can  happen  when a system crashes before the delayed allocation
543              blocks are forced to disk.
544
545       noinit_itable
546              Do not initialize any uninitialized inode table  blocks  in  the
547              background.  This  feature  may  be used by installation CD's so
548              that the install process can complete as  quickly  as  possible;
549              the  inode  table  initialization process would then be deferred
550              until the next time the filesystem is mounted.
551
552       init_itable=n
553              The lazy itable init code will wait n times the number  of  mil‐
554              liseconds  it  took to zero out the previous block group's inode
555              table. This minimizes the impact on system performance while the
556              filesystem's inode table is being initialized.
557
558       discard/nodiscard
559              Controls  whether ext4 should issue discard/TRIM commands to the
560              underlying block device when blocks are freed.  This  is  useful
561              for  SSD  devices  and sparse/thinly-provisioned LUNs, but it is
562              off by default until sufficient testing has been done.
563
564       block_validity/noblock_validity
565              This option enables/disables the in-kernel facility for tracking
566              filesystem metadata blocks within internal data structures. This
567              allows multi-block  allocator  and  other  routines  to  quickly
568              locate  extents  which  might  overlap  with filesystem metadata
569              blocks. This option is intended for debugging purposes and since
570              it negatively affects the performance, it is off by default.
571
572       dioread_lock/dioread_nolock
573              Controls whether or not ext4 should use the DIO read locking. If
574              the dioread_nolock option is specified ext4 will allocate unini‐
575              tialized  extent  before  buffer write and convert the extent to
576              initialized after IO completes.  This approach allows ext4  code
577              to  avoid  using inode mutex, which improves scalability on high
578              speed storages. However this does not work with data  journaling
579              and  dioread_nolock  option will be ignored with kernel warning.
580              Note that dioread_nolock code path is only used for extent-based
581              files.  Because of the restrictions this options comprises it is
582              off by default (e.g. dioread_lock).
583
584       max_dir_size_kb=n
585              This limits the size of the directories so that any  attempt  to
586              expand  them  beyond the specified limit in kilobytes will cause
587              an ENOSPC error. This is useful in  memory-constrained  environ‐
588              ments, where a very large directory can cause severe performance
589              problems or even provoke the Out Of Memory killer. (For example,
590              if there is only 512 MB memory available, a 176 MB directory may
591              seriously cramp the system's style.)
592
593       i_version
594              Enable 64-bit inode version  support.  This  option  is  off  by
595              default.
596
597       nombcache
598              This option disables use of mbcache for extended attribute dedu‐
599              plication. On systems where extended attributes  are  rarely  or
600              never  shared  between  files,  use of mbcache for deduplication
601              adds unnecessary computational overhead.
602
603       prjquota
604              The prjquota mount option enables project quota support  on  the
605              filesystem.  You need the quota utilities to actually enable and
606              manage the quota system.  This mount option requires the project
607              filesystem feature.
608
609

FILE ATTRIBUTES

611       The ext2, ext3, and ext4 filesystems support setting the following file
612       attributes on Linux systems using the chattr(1) utility:
613
614       a - append only
615
616       A - no atime updates
617
618       d - no dump
619
620       D - synchronous directory updates
621
622       i - immutable
623
624       S - synchronous updates
625
626       u - undeletable
627
628       In addition, the ext3 and ext4 filesystems support the following flag:
629
630       j - data journaling
631
632       Finally, the ext4 filesystem also supports the following flag:
633
634       e - extents format
635
636       For  descriptions  of  these  attribute  flags,  please  refer  to  the
637       chattr(1) man page.
638

KERNEL SUPPORT

640       This  section lists the file system driver (e.g., ext2, ext3, ext4) and
641       upstream kernel version where a particular file system feature was sup‐
642       ported.   Note  that  in  some cases the feature was present in earlier
643       kernel versions, but there were known, serious bugs.   In  other  cases
644       the feature may still be considered in an experimental state.  Finally,
645       note that some distributions may have backported  features  into  older
646       kernels;  in particular the kernel versions in certain "enterprise dis‐
647       tributions" can be extremely misleading.
648
649       filetype            ext2, 2.2.0
650
651       sparse_super        ext2, 2.2.0
652
653       large_file          ext2, 2.2.0
654
655       has_journal         ext3, 2.4.15
656
657       ext_attr            ext2/ext3, 2.6.0
658
659       dir_index           ext3, 2.6.0
660
661       resize_inode        ext3, 2.6.10 (online resizing)
662
663       64bit               ext4, 2.6.28
664
665       dir_nlink           ext4, 2.6.28
666
667       extent              ext4, 2.6.28
668
669       extra_isize         ext4, 2.6.28
670
671       flex_bg             ext4, 2.6.28
672
673       huge_file           ext4, 2.6.28
674
675       meta_bg             ext4, 2.6.28
676
677       uninit_bg           ext4, 2.6.28
678
679       mmp                 ext4, 3.0
680
681       bigalloc            ext4, 3.2
682
683       quota               ext4, 3.6
684
685       inline_data         ext4, 3.8
686
687       sparse_super2       ext4, 3.16
688
689       metadata_csum       ext4, 3.18
690
691       encrypt             ext4, 4.1
692
693       metadata_csum_seed  ext4, 4.4
694
695       project             ext4, 4.5
696
697       ea_inode            ext4, 4.13
698
699       large_dir           ext4, 4.13
700

SEE ALSO

702       mke2fs(8),   mke2fs.conf(5),   e2fsck(8),   dumpe2fs(8),    tune2fs(8),
703       debugfs(8), mount(8), chattr(1)
704
705
706
707E2fsprogs version 1.44.6          March 2019                           EXT4(5)
Impressum