Loading...
Note: File does not exist in v6.13.7.
1
2Ext4 Filesystem
3===============
4
5Ext4 is an an advanced level of the ext3 filesystem which incorporates
6scalability and reliability enhancements for supporting large filesystems
7(64 bit) in keeping with increasing disk capacities and state-of-the-art
8feature requirements.
9
10Mailing list: linux-ext4@vger.kernel.org
11Web site: http://ext4.wiki.kernel.org
12
13
141. Quick usage instructions:
15===========================
16
17Note: More extensive information for getting started with ext4 can be
18 found at the ext4 wiki site at the URL:
19 http://ext4.wiki.kernel.org/index.php/Ext4_Howto
20
21 - Compile and install the latest version of e2fsprogs (as of this
22 writing version 1.41.3) from:
23
24 http://sourceforge.net/project/showfiles.php?group_id=2406
25
26 or
27
28 ftp://ftp.kernel.org/pub/linux/kernel/people/tytso/e2fsprogs/
29
30 or grab the latest git repository from:
31
32 git://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git
33
34 - Note that it is highly important to install the mke2fs.conf file
35 that comes with the e2fsprogs 1.41.x sources in /etc/mke2fs.conf. If
36 you have edited the /etc/mke2fs.conf file installed on your system,
37 you will need to merge your changes with the version from e2fsprogs
38 1.41.x.
39
40 - Create a new filesystem using the ext4 filesystem type:
41
42 # mke2fs -t ext4 /dev/hda1
43
44 Or to configure an existing ext3 filesystem to support extents:
45
46 # tune2fs -O extents /dev/hda1
47
48 If the filesystem was created with 128 byte inodes, it can be
49 converted to use 256 byte for greater efficiency via:
50
51 # tune2fs -I 256 /dev/hda1
52
53 (Note: we currently do not have tools to convert an ext4
54 filesystem back to ext3; so please do not do try this on production
55 filesystems.)
56
57 - Mounting:
58
59 # mount -t ext4 /dev/hda1 /wherever
60
61 - When comparing performance with other filesystems, it's always
62 important to try multiple workloads; very often a subtle change in a
63 workload parameter can completely change the ranking of which
64 filesystems do well compared to others. When comparing versus ext3,
65 note that ext4 enables write barriers by default, while ext3 does
66 not enable write barriers by default. So it is useful to use
67 explicitly specify whether barriers are enabled or not when via the
68 '-o barriers=[0|1]' mount option for both ext3 and ext4 filesystems
69 for a fair comparison. When tuning ext3 for best benchmark numbers,
70 it is often worthwhile to try changing the data journaling mode; '-o
71 data=writeback' can be faster for some workloads. (Note however that
72 running mounted with data=writeback can potentially leave stale data
73 exposed in recently written files in case of an unclean shutdown,
74 which could be a security exposure in some situations.) Configuring
75 the filesystem with a large journal can also be helpful for
76 metadata-intensive workloads.
77
782. Features
79===========
80
812.1 Currently available
82
83* ability to use filesystems > 16TB (e2fsprogs support not available yet)
84* extent format reduces metadata overhead (RAM, IO for access, transactions)
85* extent format more robust in face of on-disk corruption due to magics,
86* internal redundancy in tree
87* improved file allocation (multi-block alloc)
88* lift 32000 subdirectory limit imposed by i_links_count[1]
89* nsec timestamps for mtime, atime, ctime, create time
90* inode version field on disk (NFSv4, Lustre)
91* reduced e2fsck time via uninit_bg feature
92* journal checksumming for robustness, performance
93* persistent file preallocation (e.g for streaming media, databases)
94* ability to pack bitmaps and inode tables into larger virtual groups via the
95 flex_bg feature
96* large file support
97* Inode allocation using large virtual block groups via flex_bg
98* delayed allocation
99* large block (up to pagesize) support
100* efficient new ordered mode in JBD2 and ext4(avoid using buffer head to force
101 the ordering)
102
103[1] Filesystems with a block size of 1k may see a limit imposed by the
104directory hash tree having a maximum depth of two.
105
1062.2 Candidate features for future inclusion
107
108* Online defrag (patches available but not well tested)
109* reduced mke2fs time via lazy itable initialization in conjunction with
110 the uninit_bg feature (capability to do this is available in e2fsprogs
111 but a kernel thread to do lazy zeroing of unused inode table blocks
112 after filesystem is first mounted is required for safety)
113
114There are several others under discussion, whether they all make it in is
115partly a function of how much time everyone has to work on them. Features like
116metadata checksumming have been discussed and planned for a bit but no patches
117exist yet so I'm not sure they're in the near-term roadmap.
118
119The big performance win will come with mballoc, delalloc and flex_bg
120grouping of bitmaps and inode tables. Some test results available here:
121
122 - http://www.bullopensource.org/ext4/20080818-ffsb/ffsb-write-2.6.27-rc1.html
123 - http://www.bullopensource.org/ext4/20080818-ffsb/ffsb-readwrite-2.6.27-rc1.html
124
1253. Options
126==========
127
128When mounting an ext4 filesystem, the following option are accepted:
129(*) == default
130
131ro Mount filesystem read only. Note that ext4 will
132 replay the journal (and thus write to the
133 partition) even when mounted "read only". The
134 mount options "ro,noload" can be used to prevent
135 writes to the filesystem.
136
137journal_checksum Enable checksumming of the journal transactions.
138 This will allow the recovery code in e2fsck and the
139 kernel to detect corruption in the kernel. It is a
140 compatible change and will be ignored by older kernels.
141
142journal_async_commit Commit block can be written to disk without waiting
143 for descriptor blocks. If enabled older kernels cannot
144 mount the device. This will enable 'journal_checksum'
145 internally.
146
147journal=update Update the ext4 file system's journal to the current
148 format.
149
150journal_dev=devnum When the external journal device's major/minor numbers
151 have changed, this option allows the user to specify
152 the new journal location. The journal device is
153 identified through its new major/minor numbers encoded
154 in devnum.
155
156norecovery Don't load the journal on mounting. Note that
157noload if the filesystem was not unmounted cleanly,
158 skipping the journal replay will lead to the
159 filesystem containing inconsistencies that can
160 lead to any number of problems.
161
162data=journal All data are committed into the journal prior to being
163 written into the main file system.
164
165data=ordered (*) All data are forced directly out to the main file
166 system prior to its metadata being committed to the
167 journal.
168
169data=writeback Data ordering is not preserved, data may be written
170 into the main file system after its metadata has been
171 committed to the journal.
172
173commit=nrsec (*) Ext4 can be told to sync all its data and metadata
174 every 'nrsec' seconds. The default value is 5 seconds.
175 This means that if you lose your power, you will lose
176 as much as the latest 5 seconds of work (your
177 filesystem will not be damaged though, thanks to the
178 journaling). This default value (or any low value)
179 will hurt performance, but it's good for data-safety.
180 Setting it to 0 will have the same effect as leaving
181 it at the default (5 seconds).
182 Setting it to very large values will improve
183 performance.
184
185barrier=<0|1(*)> This enables/disables the use of write barriers in
186barrier(*) the jbd code. barrier=0 disables, barrier=1 enables.
187nobarrier This also requires an IO stack which can support
188 barriers, and if jbd gets an error on a barrier
189 write, it will disable again with a warning.
190 Write barriers enforce proper on-disk ordering
191 of journal commits, making volatile disk write caches
192 safe to use, at some performance penalty. If
193 your disks are battery-backed in one way or another,
194 disabling barriers may safely improve performance.
195 The mount options "barrier" and "nobarrier" can
196 also be used to enable or disable barriers, for
197 consistency with other ext4 mount options.
198
199inode_readahead_blks=n This tuning parameter controls the maximum
200 number of inode table blocks that ext4's inode
201 table readahead algorithm will pre-read into
202 the buffer cache. The default value is 32 blocks.
203
204orlov (*) This enables the new Orlov block allocator. It is
205 enabled by default.
206
207oldalloc This disables the Orlov block allocator and enables
208 the old block allocator. Orlov should have better
209 performance - we'd like to get some feedback if it's
210 the contrary for you.
211
212user_xattr Enables Extended User Attributes. Additionally, you
213 need to have extended attribute support enabled in the
214 kernel configuration (CONFIG_EXT4_FS_XATTR). See the
215 attr(5) manual page and http://acl.bestbits.at/ to
216 learn more about extended attributes.
217
218nouser_xattr Disables Extended User Attributes.
219
220acl Enables POSIX Access Control Lists support.
221 Additionally, you need to have ACL support enabled in
222 the kernel configuration (CONFIG_EXT4_FS_POSIX_ACL).
223 See the acl(5) manual page and http://acl.bestbits.at/
224 for more information.
225
226noacl This option disables POSIX Access Control List
227 support.
228
229bsddf (*) Make 'df' act like BSD.
230minixdf Make 'df' act like Minix.
231
232debug Extra debugging information is sent to syslog.
233
234abort Simulate the effects of calling ext4_abort() for
235 debugging purposes. This is normally used while
236 remounting a filesystem which is already mounted.
237
238errors=remount-ro Remount the filesystem read-only on an error.
239errors=continue Keep going on a filesystem error.
240errors=panic Panic and halt the machine if an error occurs.
241 (These mount options override the errors behavior
242 specified in the superblock, which can be configured
243 using tune2fs)
244
245data_err=ignore(*) Just print an error message if an error occurs
246 in a file data buffer in ordered mode.
247data_err=abort Abort the journal if an error occurs in a file
248 data buffer in ordered mode.
249
250grpid Give objects the same group ID as their creator.
251bsdgroups
252
253nogrpid (*) New objects have the group ID of their creator.
254sysvgroups
255
256resgid=n The group ID which may use the reserved blocks.
257
258resuid=n The user ID which may use the reserved blocks.
259
260sb=n Use alternate superblock at this location.
261
262quota These options are ignored by the filesystem. They
263noquota are used only by quota tools to recognize volumes
264grpquota where quota should be turned on. See documentation
265usrquota in the quota-tools package for more details
266 (http://sourceforge.net/projects/linuxquota).
267
268jqfmt=<quota type> These options tell filesystem details about quota
269usrjquota=<file> so that quota information can be properly updated
270grpjquota=<file> during journal replay. They replace the above
271 quota options. See documentation in the quota-tools
272 package for more details
273 (http://sourceforge.net/projects/linuxquota).
274
275stripe=n Number of filesystem blocks that mballoc will try
276 to use for allocation size and alignment. For RAID5/6
277 systems this should be the number of data
278 disks * RAID chunk size in file system blocks.
279
280delalloc (*) Defer block allocation until just before ext4
281 writes out the block(s) in question. This
282 allows ext4 to better allocation decisions
283 more efficiently.
284nodelalloc Disable delayed allocation. Blocks are allocated
285 when the data is copied from userspace to the
286 page cache, either via the write(2) system call
287 or when an mmap'ed page which was previously
288 unallocated is written for the first time.
289
290max_batch_time=usec Maximum amount of time ext4 should wait for
291 additional filesystem operations to be batch
292 together with a synchronous write operation.
293 Since a synchronous write operation is going to
294 force a commit and then a wait for the I/O
295 complete, it doesn't cost much, and can be a
296 huge throughput win, we wait for a small amount
297 of time to see if any other transactions can
298 piggyback on the synchronous write. The
299 algorithm used is designed to automatically tune
300 for the speed of the disk, by measuring the
301 amount of time (on average) that it takes to
302 finish committing a transaction. Call this time
303 the "commit time". If the time that the
304 transaction has been running is less than the
305 commit time, ext4 will try sleeping for the
306 commit time to see if other operations will join
307 the transaction. The commit time is capped by
308 the max_batch_time, which defaults to 15000us
309 (15ms). This optimization can be turned off
310 entirely by setting max_batch_time to 0.
311
312min_batch_time=usec This parameter sets the commit time (as
313 described above) to be at least min_batch_time.
314 It defaults to zero microseconds. Increasing
315 this parameter may improve the throughput of
316 multi-threaded, synchronous workloads on very
317 fast disks, at the cost of increasing latency.
318
319journal_ioprio=prio The I/O priority (from 0 to 7, where 0 is the
320 highest priorty) which should be used for I/O
321 operations submitted by kjournald2 during a
322 commit operation. This defaults to 3, which is
323 a slightly higher priority than the default I/O
324 priority.
325
326auto_da_alloc(*) Many broken applications don't use fsync() when
327noauto_da_alloc replacing existing files via patterns such as
328 fd = open("foo.new")/write(fd,..)/close(fd)/
329 rename("foo.new", "foo"), or worse yet,
330 fd = open("foo", O_TRUNC)/write(fd,..)/close(fd).
331 If auto_da_alloc is enabled, ext4 will detect
332 the replace-via-rename and replace-via-truncate
333 patterns and force that any delayed allocation
334 blocks are allocated such that at the next
335 journal commit, in the default data=ordered
336 mode, the data blocks of the new file are forced
337 to disk before the rename() operation is
338 committed. This provides roughly the same level
339 of guarantees as ext3, and avoids the
340 "zero-length" problem that can happen when a
341 system crashes before the delayed allocation
342 blocks are forced to disk.
343
344noinit_itable Do not initialize any uninitialized inode table
345 blocks in the background. This feature may be
346 used by installation CD's so that the install
347 process can complete as quickly as possible; the
348 inode table initialization process would then be
349 deferred until the next time the file system
350 is unmounted.
351
352init_itable=n The lazy itable init code will wait n times the
353 number of milliseconds it took to zero out the
354 previous block group's inode table. This
355 minimizes the impact on the systme performance
356 while file system's inode table is being initialized.
357
358discard Controls whether ext4 should issue discard/TRIM
359nodiscard(*) commands to the underlying block device when
360 blocks are freed. This is useful for SSD devices
361 and sparse/thinly-provisioned LUNs, but it is off
362 by default until sufficient testing has been done.
363
364nouid32 Disables 32-bit UIDs and GIDs. This is for
365 interoperability with older kernels which only
366 store and expect 16-bit values.
367
368resize Allows to resize filesystem to the end of the last
369 existing block group, further resize has to be done
370 with resize2fs either online, or offline. It can be
371 used only with conjunction with remount.
372
373block_validity This options allows to enables/disables the in-kernel
374noblock_validity facility for tracking filesystem metadata blocks
375 within internal data structures. This allows multi-
376 block allocator and other routines to quickly locate
377 extents which might overlap with filesystem metadata
378 blocks. This option is intended for debugging
379 purposes and since it negatively affects the
380 performance, it is off by default.
381
382dioread_lock Controls whether or not ext4 should use the DIO read
383dioread_nolock locking. If the dioread_nolock option is specified
384 ext4 will allocate uninitialized extent before buffer
385 write and convert the extent to initialized after IO
386 completes. This approach allows ext4 code to avoid
387 using inode mutex, which improves scalability on high
388 speed storages. However this does not work with
389 data journaling and dioread_nolock option will be
390 ignored with kernel warning. Note that dioread_nolock
391 code path is only used for extent-based files.
392 Because of the restrictions this options comprises
393 it is off by default (e.g. dioread_lock).
394
395i_version Enable 64-bit inode version support. This option is
396 off by default.
397
398Data Mode
399=========
400There are 3 different data modes:
401
402* writeback mode
403In data=writeback mode, ext4 does not journal data at all. This mode provides
404a similar level of journaling as that of XFS, JFS, and ReiserFS in its default
405mode - metadata journaling. A crash+recovery can cause incorrect data to
406appear in files which were written shortly before the crash. This mode will
407typically provide the best ext4 performance.
408
409* ordered mode
410In data=ordered mode, ext4 only officially journals metadata, but it logically
411groups metadata information related to data changes with the data blocks into a
412single unit called a transaction. When it's time to write the new metadata
413out to disk, the associated data blocks are written first. In general,
414this mode performs slightly slower than writeback but significantly faster than journal mode.
415
416* journal mode
417data=journal mode provides full data and metadata journaling. All new data is
418written to the journal first, and then to its final location.
419In the event of a crash, the journal can be replayed, bringing both data and
420metadata into a consistent state. This mode is the slowest except when data
421needs to be read from and written to disk at the same time where it
422outperforms all others modes. Currently ext4 does not have delayed
423allocation support if this data journalling mode is selected.
424
425/proc entries
426=============
427
428Information about mounted ext4 file systems can be found in
429/proc/fs/ext4. Each mounted filesystem will have a directory in
430/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
431/proc/fs/ext4/dm-0). The files in each per-device directory are shown
432in table below.
433
434Files in /proc/fs/ext4/<devname>
435..............................................................................
436 File Content
437 mb_groups details of multiblock allocator buddy cache of free blocks
438..............................................................................
439
440/sys entries
441============
442
443Information about mounted ext4 file systems can be found in
444/sys/fs/ext4. Each mounted filesystem will have a directory in
445/sys/fs/ext4 based on its device name (i.e., /sys/fs/ext4/hdc or
446/sys/fs/ext4/dm-0). The files in each per-device directory are shown
447in table below.
448
449Files in /sys/fs/ext4/<devname>
450(see also Documentation/ABI/testing/sysfs-fs-ext4)
451..............................................................................
452 File Content
453
454 delayed_allocation_blocks This file is read-only and shows the number of
455 blocks that are dirty in the page cache, but
456 which do not have their location in the
457 filesystem allocated yet.
458
459 inode_goal Tuning parameter which (if non-zero) controls
460 the goal inode used by the inode allocator in
461 preference to all other allocation heuristics.
462 This is intended for debugging use only, and
463 should be 0 on production systems.
464
465 inode_readahead_blks Tuning parameter which controls the maximum
466 number of inode table blocks that ext4's inode
467 table readahead algorithm will pre-read into
468 the buffer cache
469
470 lifetime_write_kbytes This file is read-only and shows the number of
471 kilobytes of data that have been written to this
472 filesystem since it was created.
473
474 max_writeback_mb_bump The maximum number of megabytes the writeback
475 code will try to write out before move on to
476 another inode.
477
478 mb_group_prealloc The multiblock allocator will round up allocation
479 requests to a multiple of this tuning parameter if
480 the stripe size is not set in the ext4 superblock
481
482 mb_max_to_scan The maximum number of extents the multiblock
483 allocator will search to find the best extent
484
485 mb_min_to_scan The minimum number of extents the multiblock
486 allocator will search to find the best extent
487
488 mb_order2_req Tuning parameter which controls the minimum size
489 for requests (as a power of 2) where the buddy
490 cache is used
491
492 mb_stats Controls whether the multiblock allocator should
493 collect statistics, which are shown during the
494 unmount. 1 means to collect statistics, 0 means
495 not to collect statistics
496
497 mb_stream_req Files which have fewer blocks than this tunable
498 parameter will have their blocks allocated out
499 of a block group specific preallocation pool, so
500 that small files are packed closely together.
501 Each large file will have its blocks allocated
502 out of its own unique preallocation pool.
503
504 session_write_kbytes This file is read-only and shows the number of
505 kilobytes of data that have been written to this
506 filesystem since it was mounted.
507..............................................................................
508
509Ioctls
510======
511
512There is some Ext4 specific functionality which can be accessed by applications
513through the system call interfaces. The list of all Ext4 specific ioctls are
514shown in the table below.
515
516Table of Ext4 specific ioctls
517..............................................................................
518 Ioctl Description
519 EXT4_IOC_GETFLAGS Get additional attributes associated with inode.
520 The ioctl argument is an integer bitfield, with
521 bit values described in ext4.h. This ioctl is an
522 alias for FS_IOC_GETFLAGS.
523
524 EXT4_IOC_SETFLAGS Set additional attributes associated with inode.
525 The ioctl argument is an integer bitfield, with
526 bit values described in ext4.h. This ioctl is an
527 alias for FS_IOC_SETFLAGS.
528
529 EXT4_IOC_GETVERSION
530 EXT4_IOC_GETVERSION_OLD
531 Get the inode i_generation number stored for
532 each inode. The i_generation number is normally
533 changed only when new inode is created and it is
534 particularly useful for network filesystems. The
535 '_OLD' version of this ioctl is an alias for
536 FS_IOC_GETVERSION.
537
538 EXT4_IOC_SETVERSION
539 EXT4_IOC_SETVERSION_OLD
540 Set the inode i_generation number stored for
541 each inode. The '_OLD' version of this ioctl
542 is an alias for FS_IOC_SETVERSION.
543
544 EXT4_IOC_GROUP_EXTEND This ioctl has the same purpose as the resize
545 mount option. It allows to resize filesystem
546 to the end of the last existing block group,
547 further resize has to be done with resize2fs,
548 either online, or offline. The argument points
549 to the unsigned logn number representing the
550 filesystem new block count.
551
552 EXT4_IOC_MOVE_EXT Move the block extents from orig_fd (the one
553 this ioctl is pointing to) to the donor_fd (the
554 one specified in move_extent structure passed
555 as an argument to this ioctl). Then, exchange
556 inode metadata between orig_fd and donor_fd.
557 This is especially useful for online
558 defragmentation, because the allocator has the
559 opportunity to allocate moved blocks better,
560 ideally into one contiguous extent.
561
562 EXT4_IOC_GROUP_ADD Add a new group descriptor to an existing or
563 new group descriptor block. The new group
564 descriptor is described by ext4_new_group_input
565 structure, which is passed as an argument to
566 this ioctl. This is especially useful in
567 conjunction with EXT4_IOC_GROUP_EXTEND,
568 which allows online resize of the filesystem
569 to the end of the last existing block group.
570 Those two ioctls combined is used in userspace
571 online resize tool (e.g. resize2fs).
572
573 EXT4_IOC_MIGRATE This ioctl operates on the filesystem itself.
574 It converts (migrates) ext3 indirect block mapped
575 inode to ext4 extent mapped inode by walking
576 through indirect block mapping of the original
577 inode and converting contiguous block ranges
578 into ext4 extents of the temporary inode. Then,
579 inodes are swapped. This ioctl might help, when
580 migrating from ext3 to ext4 filesystem, however
581 suggestion is to create fresh ext4 filesystem
582 and copy data from the backup. Note, that
583 filesystem has to support extents for this ioctl
584 to work.
585
586 EXT4_IOC_ALLOC_DA_BLKS Force all of the delay allocated blocks to be
587 allocated to preserve application-expected ext3
588 behaviour. Note that this will also start
589 triggering a write of the data blocks, but this
590 behaviour may change in the future as it is
591 not necessary and has been done this way only
592 for sake of simplicity.
593..............................................................................
594
595References
596==========
597
598kernel source: <file:fs/ext4/>
599 <file:fs/jbd2/>
600
601programs: http://e2fsprogs.sourceforge.net/
602
603useful links: http://fedoraproject.org/wiki/ext3-devel
604 http://www.bullopensource.org/ext4/
605 http://ext4.wiki.kernel.org/index.php/Main_Page
606 http://fedoraproject.org/wiki/Features/Ext4