1MOUNT(8) Linux Programmer's Manual MOUNT(8)
2
6 mount - mount a filesystem
7
9 mount [-lhV]
10 mount -a [-fFnrsvw] [-t vfstype] [-O optlist]
12 mount [-fnrsvw] [-o option[,option]...] device|dir
14 mount [-fnrsvw] [-t vfstype] [-o options] device dir
16
18 All files accessible in a Unix system are arranged in one big tree, the
19 file hierarchy, rooted at /. These files can be spread out over sev‐
20 eral devices. The mount command serves to attach the filesystem found
21 on some device to the big file tree. Conversely, the umount(8) command
22 will detach it again.
23 The standard form of the mount command, is
25 mount -t type device dir
27 This tells the kernel to attach the filesystem found on device (which
29 is of type type) at the directory dir. The previous contents (if any)
30 and owner and mode of dir become invisible, and as long as this
31 filesystem remains mounted, the pathname dir refers to the root of the
32 filesystem on device.
33 The listing and help.
35 Three forms of invocation do not actually mount anything:
36 mount -h
38 prints a help message
39 mount -V
41 prints a version string
42 mount [-l] [-t type]
44 lists all mounted filesystems (of type type). The option
45 -l adds the labels in this listing. See below.
46 The device indication.
48 Most devices are indicated by a file name (of a block special
49 device), like /dev/sda1, but there are other possibilities. For
50 example, in the case of an NFS mount, device may look like
51 knuth.cwi.nl:/dir. It is possible to indicate a block special
52 device using its volume LABEL or UUID (see the -L and -U options
53 below).
54 The recommended setup is to use LABEL=<label> or UUID=<uuid>
56 tags rather than /dev/disk/by-{label,uuid} udev symlinks in the
57 /etc/fstab file. The tags are more readable, robust and porta‐
58 ble. The mount(8) command internally uses udev symlinks, so use
59 the symlinks in /etc/fstab is not advantage over LABEL=/UUID=.
60 For more details see libblkid(3).
61 The proc filesystem is not associated with a special device, and
63 when mounting it, an arbitrary keyword, such as proc can be used
64 instead of a device specification. (The customary choice none
65 is less fortunate: the error message `none busy' from umount can
66 be confusing.)
67 The /etc/fstab, /etc/mtab and /proc/mounts files.
69 The file /etc/fstab (see fstab(5)), may contain lines describing
70 what devices are usually mounted where, using which options.
71 The command
73 mount -a [-t type] [-O optlist]
75 (usually given in a bootscript) causes all filesystems mentioned
77 in fstab (of the proper type and/or having or not having the
78 proper options) to be mounted as indicated, except for those
79 whose line contains the noauto keyword. Adding the -F option
80 will make mount fork, so that the filesystems are mounted simul‐
81 taneously.
82 When mounting a filesystem mentioned in fstab or mtab, it suf‐
84 fices to give only the device, or only the mount point.
85 The programs mount and umount maintain a list of currently
88 mounted filesystems in the file /etc/mtab. If no arguments are
89 given to mount, this list is printed.
90 The mount program does not read the /etc/fstab file if device
92 (or LABEL/UUID) and dir are specified. For example:
93 mount /dev/foo /dir
95 If you want to override mount options from /etc/fstab you have
97 to use:
98 mount device|dir -o <options>
100 and then the mount options from command line will be appended to
102 the list of options from /etc/fstab. The usual behaviour is
103 that the last option wins if there is more duplicated options.
104 When the proc filesystem is mounted (say at /proc), the files
106 /etc/mtab and /proc/mounts have very similar contents. The for‐
107 mer has somewhat more information, such as the mount options
108 used, but is not necessarily up-to-date (cf. the -n option
109 below). It is possible to replace /etc/mtab by a symbolic link
110 to /proc/mounts, and especially when you have very large numbers
111 of mounts things will be much faster with that symlink, but some
112 information is lost that way, and in particular using the "user"
113 option will fail.
114 The non-superuser mounts.
116 Normally, only the superuser can mount filesystems. However,
117 when fstab contains the user option on a line, anybody can mount
118 the corresponding system.
119 Thus, given a line
121 /dev/cdrom /cd iso9660 ro,user,noauto,unhide
123 any user can mount the iso9660 filesystem found on his CDROM
125 using the command
126 mount /dev/cdrom
128 or
130 mount /cd
132 For more details, see fstab(5). Only the user that mounted a
134 filesystem can unmount it again. If any user should be able to
135 unmount, then use users instead of user in the fstab line. The
136 owner option is similar to the user option, with the restriction
137 that the user must be the owner of the special file. This may be
138 useful e.g. for /dev/fd if a login script makes the console user
139 owner of this device. The group option is similar, with the
140 restriction that the user must be member of the group of the
141 special file.
142 The bind mounts.
145 Since Linux 2.4.0 it is possible to remount part of the file
146 hierarchy somewhere else. The call is
147 mount --bind olddir newdir
148 or shortoption
149 mount -B olddir newdir
150 or fstab entry is:
151 /olddir /newdir none bind
152 After this call the same contents is accessible in two places.
154 One can also remount a single file (on a single file).
155 This call attaches only (part of) a single filesystem, not pos‐
157 sible submounts. The entire file hierarchy including submounts
158 is attached a second place using
159 mount --rbind olddir newdir
160 or shortoption
161 mount -R olddir newdir
162 Note that the filesystem mount options will remain the same as
164 those on the original mount point, and cannot be changed by
165 passing the -o option along with --bind/--rbind. The mount
166 options can be changed by a separate remount command, for exam‐
167 ple:
168 mount --bind olddir newdir
170 mount -o remount,ro newdir
171 Note that behavior of the remount operation depends on the
173 /etc/mtab file. The first command stores the 'bind' flag to the
174 /etc/mtab file and the second command reads the flag from the
175 file. If you have a system without the /etc/mtab file or if you
176 explicitly define source and target for the remount command
177 (then mount(8) does not read /etc/mtab), then you have to use
178 bind flag (or option) for the remount command too. For example:
179 mount --bind olddir newdir
181 mount -o remount,ro,bind olddir newdir
182 The move operation.
184 Since Linux 2.5.1 it is possible to atomically move a mounted
185 tree to another place. The call is
186 mount --move olddir newdir
187 or shortoption
188 mount -M olddir newdir
189 This will cause the contents which previously appeared under
190 olddir to be accessed under newdir. The physical location of
191 the files is not changed.
192 The shared subtrees operations.
194 Since Linux 2.6.15 it is possible to mark a mount and its sub‐
195 mounts as shared, private, slave or unbindable. A shared mount
196 provides ability to create mirrors of that mount such that
197 mounts and umounts within any of the mirrors propagate to the
198 other mirror. A slave mount receives propagation from its mas‐
199 ter, but any not vice-versa. A private mount carries no propa‐
200 gation abilities. A unbindable mount is a private mount which
201 cannot be cloned through a bind operation. Detailed semantics is
202 documented in Documentation/sharedsubtree.txt file in the kernel
203 source tree.
204 mount --make-shared mountpoint
206 mount --make-slave mountpoint
207 mount --make-private mountpoint
208 mount --make-unbindable mountpoint
209 The following commands allows one to recursively change the type
211 of all the mounts under a given mountpoint.
212 mount --make-rshared mountpoint
214 mount --make-rslave mountpoint
215 mount --make-rprivate mountpoint
216 mount --make-runbindable mountpoint
217
220 The full set of mount options used by an invocation of mount is deter‐
221 mined by first extracting the mount options for the filesystem from the
222 fstab table, then applying any options specified by the -o argument,
223 and finally applying a -r or -w option, when present.
224 Command line options available for the mount command:
226 -V, --version
228 Output version.
229 -h, --help
231 Print a help message.
232 -v, --verbose
234 Verbose mode.
235 -a, --all
237 Mount all filesystems (of the given types) mentioned in fstab.
238 -F, --fork
240 (Used in conjunction with -a.) Fork off a new incarnation of
241 mount for each device. This will do the mounts on different
242 devices or different NFS servers in parallel. This has the
243 advantage that it is faster; also NFS timeouts go in parallel. A
244 disadvantage is that the mounts are done in undefined order.
245 Thus, you cannot use this option if you want to mount both /usr
246 and /usr/spool.
247 -f, --fake
249 Causes everything to be done except for the actual system call;
250 if it's not obvious, this ``fakes'' mounting the filesystem.
251 This option is useful in conjunction with the -v flag to deter‐
252 mine what the mount command is trying to do. It can also be used
253 to add entries for devices that were mounted earlier with the -n
254 option. The -f option checks for existing record in /etc/mtab
255 and fails when the record already exists (with regular non-fake
256 mount, this check is done by kernel).
257 -i, --internal-only
259 Don't call the /sbin/mount.<filesystem> helper even if it
260 exists.
261 -l Add the labels in the mount output. Mount must have permission
263 to read the disk device (e.g. be suid root) for this to work.
264 One can set such a label for ext2, ext3 or ext4 using the
265 e2label(8) utility, or for XFS using xfs_admin(8), or for reis‐
266 erfs using reiserfstune(8).
267 -n, --no-mtab
269 Mount without writing in /etc/mtab. This is necessary for exam‐
270 ple when /etc is on a read-only filesystem.
271 --no-canonicalize
273 Don't canonicalize paths. The mount command canonicalizes all
274 paths (from command line or fstab) and stores canonicalized
275 paths to the /etc/mtab file. This option can be used together
276 with the -f flag for already canonicalized absolut paths.
277 -p, --pass-fd num
279 In case of a loop mount with encryption, read the passphrase
280 from file descriptor num instead of from the terminal.
281 -s Tolerate sloppy mount options rather than failing. This will
283 ignore mount options not supported by a filesystem type. Not all
284 filesystems support this option. This option exists for support
285 of the Linux autofs-based automounter.
286 -r, --read-only
288 Mount the filesystem read-only. A synonym is -o ro.
289 Note that, depending on the filesystem type, state and kernel
291 behavior, the system may still write to the device. For example,
292 Ext3 or ext4 will replay its journal if the filesystem is dirty.
293 To prevent this kind of write access, you may want to mount ext3
294 or ext4 filesystem with "ro,noload" mount options or set the
295 block device to read-only mode, see command blockdev(8).
296 -w, --rw
298 Mount the filesystem read/write. This is the default. A synonym
299 is -o rw.
300 -L label
302 Mount the partition that has the specified label.
303 -U uuid
305 Mount the partition that has the specified uuid. These two
306 options require the file /proc/partitions (present since Linux
307 2.1.116) to exist.
308 -t, --types vfstype
310 The argument following the -t is used to indicate the filesystem
311 type. The filesystem types which are currently supported
312 include: adfs, affs, autofs, cifs, coda, coherent, cramfs,
313 debugfs, devpts, efs, ext, ext2, ext3, ext4, hfs, hfsplus, hpfs,
314 iso9660, jfs, minix, msdos, ncpfs, nfs, nfs4, ntfs, proc, qnx4,
315 ramfs, reiserfs, romfs, squashfs, smbfs, sysv, tmpfs, ubifs,
316 udf, ufs, umsdos, usbfs, vfat, xenix, xfs, xiafs. Note that
317 coherent, sysv and xenix are equivalent and that xenix and
318 coherent will be removed at some point in the future — use sysv
319 instead. Since kernel version 2.1.21 the types ext and xiafs do
320 not exist anymore. Earlier, usbfs was known as usbdevfs. Note,
321 the real list of all supported filesystems depends on your ker‐
322 nel.
323 For most types all the mount program has to do is issue a simple
325 mount(2) system call, and no detailed knowledge of the filesys‐
326 tem type is required. For a few types however (like nfs, nfs4,
327 cifs, smbfs, ncpfs) ad hoc code is necessary. The nfs, nfs4,
328 cifs, smbfs, and ncpfs filesystems have a separate mount pro‐
329 gram. In order to make it possible to treat all types in a uni‐
330 form way, mount will execute the program /sbin/mount.TYPE (if
331 that exists) when called with type TYPE. Since various versions
332 of the smbmount program have different calling conventions,
333 /sbin/mount.smbfs may have to be a shell script that sets up the
334 desired call.
335 If no -t option is given, or if the auto type is specified,
337 mount will try to guess the desired type. Mount uses the blkid
338 or volume_id library for guessing the filesystem type; if that
339 does not turn up anything that looks familiar, mount will try to
340 read the file /etc/filesystems, or, if that does not exist,
341 /proc/filesystems. All of the filesystem types listed there
342 will be tried, except for those that are labeled "nodev" (e.g.,
343 devpts, proc and nfs). If /etc/filesystems ends in a line with
344 a single * only, mount will read /proc/filesystems afterwards.
345 The auto type may be useful for user-mounted floppies. Creating
347 a file /etc/filesystems can be useful to change the probe order
348 (e.g., to try vfat before msdos or ext3 before ext2) or if you
349 use a kernel module autoloader. Warning: the probing uses a
350 heuristic (the presence of appropriate `magic'), and could rec‐
351 ognize the wrong filesystem type, possibly with catastrophic
352 consequences. If your data is valuable, don't ask mount to
353 guess.
354 More than one type may be specified in a comma separated list.
356 The list of filesystem types can be prefixed with no to specify
357 the filesystem types on which no action should be taken. (This
358 can be meaningful with the -a option.) For example, the command:
359 mount -a -t nomsdos,ext
361 mounts all filesystems except those of type msdos and ext.
363 -O, --test-opts opts
365 Used in conjunction with -a, to limit the set of filesystems to
366 which the -a is applied. Like -t in this regard except that it
367 is useless except in the context of -a. For example, the com‐
368 mand:
369 mount -a -O no_netdev
371 mounts all filesystems except those which have the option _net‐
373 dev specified in the options field in the /etc/fstab file.
374 It is different from -t in that each option is matched exactly;
376 a leading no at the beginning of one option does not negate the
377 rest.
378 The -t and -O options are cumulative in effect; that is, the
380 command
381 mount -a -t ext2 -O _netdev
383 mounts all ext2 filesystems with the _netdev option, not all
385 filesystems that are either ext2 or have the _netdev option
386 specified.
387 -o, --options opts
389 Options are specified with a -o flag followed by a comma sepa‐
390 rated string of options. For example:
391 mount LABEL=mydisk -o noatime,nouser
393 For more details, see FILESYSTEM INDEPENDENT MOUNT OPTIONS and
396 FILESYSTEM SPECIFIC MOUNT OPTIONS sections.
397 -B, --bind
399 Remount a subtree somewhere else (so that its contents are
400 available in both places). See above.
401 -R, --rbind
403 Remount a subtree and all possible submounts somewhere else (so
404 that its contents are available in both places). See above.
405 -M, --move
407 Move a subtree to some other place. See above.
408
411 Some of these options are only useful when they appear in the
412 /etc/fstab file.
413 Some of these options could be enabled or disabled by default in the
415 system kernel. To check the current setting see the options in
416 /proc/mounts.
417 The following options apply to any filesystem that is being mounted
419 (but not every filesystem actually honors them - e.g., the sync option
420 today has effect only for ext2, ext3, fat, vfat and ufs):
421 async All I/O to the filesystem should be done asynchronously. (See
424 also the sync option.)
425 atime Update inode access time for each access. See also the stricta‐
427 time mount option.
428 noatime
430 Do not update inode access times on this filesystem (e.g, for
431 faster access on the news spool to speed up news servers).
432 auto Can be mounted with the -a option.
434 noauto Can only be mounted explicitly (i.e., the -a option will not
436 cause the filesystem to be mounted).
437 context=context, fscontext=context, defcontext=context and rootcon‐
439 text=context
440 The context= option is useful when mounting filesystems that do
441 not support extended attributes, such as a floppy or hard disk
442 formatted with VFAT, or systems that are not normally running
443 under SELinux, such as an ext3 formatted disk from a non-SELinux
444 workstation. You can also use context= on filesystems you do not
445 trust, such as a floppy. It also helps in compatibility with
446 xattr-supporting filesystems on earlier 2.4.<x> kernel versions.
447 Even where xattrs are supported, you can save time not having to
448 label every file by assigning the entire disk one security con‐
449 text.
450 A commonly used option for removable media is context=sys‐
452 tem_u:object_r:removable_t.
453 Two other options are fscontext= and defcontext=, both of which
455 are mutually exclusive of the context option. This means you can
456 use fscontext and defcontext with each other, but neither can be
457 used with context.
458 The fscontext= option works for all filesystems, regardless of
460 their xattr support. The fscontext option sets the overarching
461 filesystem label to a specific security context. This filesystem
462 label is separate from the individual labels on the files. It
463 represents the entire filesystem for certain kinds of permission
464 checks, such as during mount or file creation. Individual file
465 labels are still obtained from the xattrs on the files them‐
466 selves. The context option actually sets the aggregate context
467 that fscontext provides, in addition to supplying the same label
468 for individual files.
469 You can set the default security context for unlabeled files
471 using defcontext= option. This overrides the value set for unla‐
472 beled files in the policy and requires a filesystem that sup‐
473 ports xattr labeling.
474 The rootcontext= option allows you to explicitly label the root
476 inode of a FS being mounted before that FS or inode because vis‐
477 able to userspace. This was found to be useful for things like
478 stateless linux.
479 For more details, see selinux(8)
481 defaults
484 Use default options: rw, suid, dev, exec, auto, nouser, and
485 async.
486 dev Interpret character or block special devices on the filesystem.
488 nodev Do not interpret character or block special devices on the file
490 system.
491 diratime
493 Update directory inode access times on this filesystem. This is
494 the default.
495 nodiratime
497 Do not update directory inode access times on this filesystem.
498 dirsync
500 All directory updates within the filesystem should be done syn‐
501 chronously. This affects the following system calls: creat,
502 link, unlink, symlink, mkdir, rmdir, mknod and rename.
503 exec Permit execution of binaries.
505 noexec Do not allow direct execution of any binaries on the mounted
507 filesystem. (Until recently it was possible to run binaries
508 anyway using a command like /lib/ld*.so /mnt/binary. This trick
509 fails since Linux 2.4.25 / 2.6.0.)
510 group Allow an ordinary (i.e., non-root) user to mount the filesystem
512 if one of his groups matches the group of the device. This
513 option implies the options nosuid and nodev (unless overridden
514 by subsequent options, as in the option line group,dev,suid).
515 iversion
517 Every time the inode is modified, the i_version field will be
518 incremented.
519 noiversion
521 Do not increment the i_version inode field.
522 mand Allow mandatory locks on this filesystem. See fcntl(2).
524 nomand Do not allow mandatory locks on this filesystem.
526 _netdev
528 The filesystem resides on a device that requires network access
529 (used to prevent the system from attempting to mount these
530 filesystems until the network has been enabled on the system).
531 nofail Do not report errors for this device if it does not exist.
533 relatime
535 Update inode access times relative to modify or change time.
536 Access time is only updated if the previous access time was ear‐
537 lier than the current modify or change time. (Similar to noat‐
538 ime, but doesn't break mutt or other applications that need to
539 know if a file has been read since the last time it was modi‐
540 fied.)
541 norelatime
543 Do not use relatime feature. See also the strictatime mount
544 option.
545 strictatime
547 Allows to explicitly requesting full atime updates. This makes
548 it possible for kernel to defaults to relatime or noatime but
549 still allow userspace to override it. For more details about the
550 default system mount options see /proc/mounts.
551 nostrictatime
553 Use the kernel's default behaviour for inode access time
554 updates.
555 suid Allow set-user-identifier or set-group-identifier bits to take
557 effect.
558 nosuid Do not allow set-user-identifier or set-group-identifier bits to
560 take effect. (This seems safe, but is in fact rather unsafe if
561 you have suidperl(1) installed.)
562 owner Allow an ordinary (i.e., non-root) user to mount the filesystem
564 if he is the owner of the device. This option implies the
565 options nosuid and nodev (unless overridden by subsequent
566 options, as in the option line owner,dev,suid).
567 remount
569 Attempt to remount an already-mounted filesystem. This is com‐
570 monly used to change the mount flags for a filesystem, espe‐
571 cially to make a readonly filesystem writeable. It does not
572 change device or mount point.
573 The remount functionality follows the standard way how the mount
575 command works with options from fstab. It means the mount com‐
576 mand doesn't read fstab (or mtab) only when a device and dir are
577 fully specified.
578 mount -o remount,rw /dev/foo /dir
580 After this call all old mount options are replaced and arbitrary
582 stuff from fstab is ignored, except the loop= option which is
583 internally generated and maintained by the mount command.
584 mount -o remount,rw /dir
586 After this call mount reads fstab (or mtab) and merges these
588 options with options from command line ( -o ).
589 ro Mount the filesystem read-only.
591 rw Mount the filesystem read-write.
593 sync All I/O to the filesystem should be done synchronously. In case
595 of media with limited number of write cycles (e.g. some flash
596 drives) "sync" may cause life-cycle shortening.
597 user Allow an ordinary user to mount the filesystem. The name of the
599 mounting user is written to mtab so that he can unmount the
600 filesystem again. This option implies the options noexec,
601 nosuid, and nodev (unless overridden by subsequent options, as
602 in the option line user,exec,dev,suid).
603 nouser Forbid an ordinary (i.e., non-root) user to mount the filesys‐
605 tem. This is the default.
606 users Allow every user to mount and unmount the filesystem. This
608 option implies the options noexec, nosuid, and nodev (unless
609 overridden by subsequent options, as in the option line
610 users,exec,dev,suid).
611
614 The following options apply only to certain filesystems. We sort them
615 by filesystem. They all follow the -o flag.
616 What options are supported depends a bit on the running kernel. More
618 info may be found in the kernel source subdirectory Documenta‐
619 tion/filesystems.
620
623 uid=value and gid=value
624 Set the owner and group of the files in the filesystem (default:
625 uid=gid=0).
626 ownmask=value and othmask=value
628 Set the permission mask for ADFS 'owner' permissions and 'other'
629 permissions, respectively (default: 0700 and 0077, respec‐
630 tively). See also /usr/src/linux/Documentation/filesys‐
631 tems/adfs.txt.
632
634 uid=value and gid=value
635 Set the owner and group of the root of the filesystem (default:
636 uid=gid=0, but with option uid or gid without specified value,
637 the uid and gid of the current process are taken).
638 setuid=value and setgid=value
640 Set the owner and group of all files.
641 mode=value
643 Set the mode of all files to value & 0777 disregarding the orig‐
644 inal permissions. Add search permission to directories that
645 have read permission. The value is given in octal.
646 protect
648 Do not allow any changes to the protection bits on the filesys‐
649 tem.
650 usemp Set uid and gid of the root of the filesystem to the uid and gid
652 of the mount point upon the first sync or umount, and then clear
653 this option. Strange...
654 verbose
656 Print an informational message for each successful mount.
657 prefix=string
659 Prefix used before volume name, when following a link.
660 volume=string
662 Prefix (of length at most 30) used before '/' when following a
663 symbolic link.
664 reserved=value
666 (Default: 2.) Number of unused blocks at the start of the
667 device.
668 root=value
670 Give explicitly the location of the root block.
671 bs=value
673 Give blocksize. Allowed values are 512, 1024, 2048, 4096.
674 grpquota|noquota|quota|usrquota
676 These options are accepted but ignored. (However, quota utili‐
677 ties may react to such strings in /etc/fstab.)
678
681 See the options section of the mount.cifs(8) man page (samba-client
682 package must be installed).
683
686 None.
687
690 The debugfs filesystem is a pseudo filesystem, traditionally mounted on
691 /sys/kernel/debug. There are no mount options.
692
695 The devpts filesystem is a pseudo filesystem, traditionally mounted on
696 /dev/pts. In order to acquire a pseudo terminal, a process opens
697 /dev/ptmx; the number of the pseudo terminal is then made available to
698 the process and the pseudo terminal slave can be accessed as
699 /dev/pts/<number>.
700 uid=value and gid=value
702 This sets the owner or the group of newly created PTYs to the
703 specified values. When nothing is specified, they will be set to
704 the UID and GID of the creating process. For example, if there
705 is a tty group with GID 5, then gid=5 will cause newly created
706 PTYs to belong to the tty group.
707 mode=value
709 Set the mode of newly created PTYs to the specified value. The
710 default is 0600. A value of mode=620 and gid=5 makes "mesg y"
711 the default on newly created PTYs.
712 newinstance
714 Create a private instance of devpts filesystem, such that
715 indices of ptys allocated in this new instance are independent
716 of indices created in other instances of devpts.
717 All mounts of devpts without this newinstance option share the
719 same set of pty indices (i.e legacy mode). Each mount of devpts
720 with the newinstance option has a private set of pty indices.
721 This option is mainly used to support containers in the linux
723 kernel. It is implemented in linux kernel versions starting with
724 2.6.29. Further, this mount option is valid only if CON‐
725 FIG_DEVPTS_MULTIPLE_INSTANCES is enabled in the kernel configu‐
726 ration.
727 To use this option effectively, /dev/ptmx must be a symbolic
729 link to pts/ptmx. See Documentation/filesystems/devpts.txt in
730 the linux kernel source tree for details.
731 ptmxmode=value
733 Set the mode for the new ptmx device node in the devpts filesys‐
735 tem.
736 With the support for multiple instances of devpts (see newin‐
738 stance option above), each instance has a private ptmx node in
739 the root of the devpts filesystem (typically /dev/pts/ptmx).
740 For compatibility with older versions of the kernel, the default
742 mode of the new ptmx node is 0000. ptmxmode=value specifies a
743 more useful mode for the ptmx node and is highly recommended
744 when the newinstance option is specified.
745 This option is only implemented in linux kernel versions start‐
747 ing with 2.6.29. Further this option is valid only if CON‐
748 FIG_DEVPTS_MULTIPLE_INSTANCES is enabled in the kernel configu‐
749 ration.
750
753 None. Note that the `ext' filesystem is obsolete. Don't use it. Since
754 Linux version 2.1.21 extfs is no longer part of the kernel source.
755
758 The `ext2' filesystem is the standard Linux filesystem. Since Linux
759 2.5.46, for most mount options the default is determined by the
760 filesystem superblock. Set them with tune2fs(8).
761 acl|noacl
763 Support POSIX Access Control Lists (or not).
764 bsddf|minixdf
766 Set the behaviour for the statfs system call. The minixdf behav‐
767 iour is to return in the f_blocks field the total number of
768 blocks of the filesystem, while the bsddf behaviour (which is
769 the default) is to subtract the overhead blocks used by the ext2
770 filesystem and not available for file storage. Thus
771 % mount /k -o minixdf; df /k; umount /k
773 Filesystem 1024-blocks Used Available Capacity Mounted on
774 /dev/sda6 2630655 86954 2412169 3% /k
775 % mount /k -o bsddf; df /k; umount /k
776 Filesystem 1024-blocks Used Available Capacity Mounted on
777 /dev/sda6 2543714 13 2412169 0% /k
778 (Note that this example shows that one can add command line
780 options to the options given in /etc/fstab.)
781 check={none|nocheck}
784 No checking is done at mount time. This is the default. This is
785 fast. It is wise to invoke e2fsck(8) every now and then, e.g.
786 at boot time.
787 debug Print debugging info upon each (re)mount.
789 errors={continue|remount-ro|panic}
791 Define the behaviour when an error is encountered. (Either
792 ignore errors and just mark the filesystem erroneous and con‐
793 tinue, or remount the filesystem read-only, or panic and halt
794 the system.) The default is set in the filesystem superblock,
795 and can be changed using tune2fs(8).
796 grpid|bsdgroups and nogrpid|sysvgroups
798 These options define what group id a newly created file gets.
799 When grpid is set, it takes the group id of the directory in
800 which it is created; otherwise (the default) it takes the fsgid
801 of the current process, unless the directory has the setgid bit
802 set, in which case it takes the gid from the parent directory,
803 and also gets the setgid bit set if it is a directory itself.
804 grpquota|noquota|quota|usrquota
806 These options are accepted but ignored.
807 nobh Do not attach buffer_heads to file pagecache. (Since 2.5.49.)
809 nouid32
811 Disables 32-bit UIDs and GIDs. This is for interoperability
812 with older kernels which only store and expect 16-bit values.
813 oldalloc or orlov
815 Use old allocator or Orlov allocator for new inodes. Orlov is
816 default.
817 resgid=n and resuid=n
819 The ext2 filesystem reserves a certain percentage of the avail‐
820 able space (by default 5%, see mke2fs(8) and tune2fs(8)). These
821 options determine who can use the reserved blocks. (Roughly:
822 whoever has the specified uid, or belongs to the specified
823 group.)
824 sb=n Instead of block 1, use block n as superblock. This could be
826 useful when the filesystem has been damaged. (Earlier, copies
827 of the superblock would be made every 8192 blocks: in block 1,
828 8193, 16385, ... (and one got thousands of copies on a big
829 filesystem). Since version 1.08, mke2fs has a -s (sparse
830 superblock) option to reduce the number of backup superblocks,
831 and since version 1.15 this is the default. Note that this may
832 mean that ext2 filesystems created by a recent mke2fs cannot be
833 mounted r/w under Linux 2.0.*.) The block number here uses 1k
834 units. Thus, if you want to use logical block 32768 on a
835 filesystem with 4k blocks, use "sb=131072".
836 user_xattr|nouser_xattr
838 Support "user." extended attributes (or not).
839
843 The ext3 filesystem is a version of the ext2 filesystem which has been
844 enhanced with journalling. It supports the same options as ext2 as
845 well as the following additions:
846 journal=update
848 Update the ext3 filesystem's journal to the current format.
849 journal=inum
851 When a journal already exists, this option is ignored. Other‐
852 wise, it specifies the number of the inode which will represent
853 the ext3 filesystem's journal file; ext3 will create a new
854 journal, overwriting the old contents of the file whose inode
855 number is inum.
856 noload Do not load the ext3 filesystem's journal on mounting.
858 data={journal|ordered|writeback}
860 Specifies the journalling mode for file data. Metadata is
861 always journaled. To use modes other than ordered on the root
862 filesystem, pass the mode to the kernel as boot parameter, e.g.
863 rootflags=data=journal.
864 journal
866 All data is committed into the journal prior to being
867 written into the main filesystem.
868 ordered
870 This is the default mode. All data is forced directly
871 out to the main file system prior to its metadata being
872 committed to the journal.
873 writeback
875 Data ordering is not preserved - data may be written into
876 the main filesystem after its metadata has been committed
877 to the journal. This is rumoured to be the highest-
878 throughput option. It guarantees internal filesystem
879 integrity, however it can allow old data to appear in
880 files after a crash and journal recovery.
881 barrier=0 / barrier=1
883 This enables/disables barriers. barrier=0 disables it, bar‐
884 rier=1 enables it. Write barriers enforce proper on-disk order‐
885 ing of journal commits, making volatile disk write caches safe
886 to use, at some performance penalty. The ext3 filesystem does
887 not enable write barriers by default. Be sure to enable barri‐
888 ers unless your disks are battery-backed one way or another.
889 Otherwise you risk filesystem corruption in case of power fail‐
890 ure.
891 commit=nrsec
893 Sync all data and metadata every nrsec seconds. The default
894 value is 5 seconds. Zero means default.
895 user_xattr
897 Enable Extended User Attributes. See the attr(5) manual page.
898 acl Enable POSIX Access Control Lists. See the acl(5) manual page.
900
903 The ext4 filesystem is an an advanced level of the ext3 filesystem
904 which incorporates scalability and reliability enhancements for sup‐
905 porting large filesystem.
906 The options journal_dev, noload, data, commit, orlov, oldalloc,
908 [no]user_xattr [no]acl, bsddf, minixdf, debug, errors, data_err, grpid,
909 bsdgroups, nogrpid sysvgroups, resgid, resuid, sb, quota, noquota,
910 grpquota, usrquota and [no]bh are backwardly compatible with ext3 or
911 ext2.
912 journal_checksum
914 Enable checksumming of the journal transactions. This will
915 allow the recovery code in e2fsck and the kernel to detect cor‐
916 ruption in the kernel. It is a compatible change and will be
917 ignored by older kernels.
918 journal_async_commit
920 Commit block can be written to disk without waiting for descrip‐
921 tor blocks. If enabled older kernels cannot mount the device.
922 This will enable
923 journal=update
925 Update the ext4 filesystem's journal to the current format.
926 barrier=0 / barrier=1 / barrier / nobarrier
928 This enables/disables the use of write barriers in the jbd code.
929 barrier=0 disables, barrier=1 enables. This also requires an IO
930 stack which can support barriers, and if jbd gets an error on a
931 barrier write, it will disable again with a warning. Write bar‐
932 riers enforce proper on-disk ordering of journal commits, making
933 volatile disk write caches safe to use, at some performance
934 penalty. If your disks are battery-backed in one way or
935 another, disabling barriers may safely improve performance. The
936 mount options "barrier" and "nobarrier" can also be used to
937 enable or disable barriers, for consistency with other ext4
938 mount options.
939 The ext4 filesystem enables write barriers by default.
941 inode_readahead=n
943 This tuning parameter controls the maximum number of inode table
944 blocks that ext4's inode table readahead algorithm will pre-read
945 into the buffer cache. The default value is 32 blocks.
946 stripe=n
948 Number of filesystem blocks that mballoc will try to use for
949 allocation size and alignment. For RAID5/6 systems this should
950 be the number of data disks * RAID chunk size in filesystem
951 blocks.
952 delalloc
954 Deferring block allocation until write-out time.
955 nodelalloc
957 Disable delayed allocation. Blocks are allocation when data is
958 copied from user to page cache.
959 max_batch_time=usec
961 Maximum amount of time ext4 should wait for additional filesys‐
962 tem operations to be batch together with a synchronous write
963 operation. Since a synchronous write operation is going to force
964 a commit and then a wait for the I/O complete, it doesn't cost
965 much, and can be a huge throughput win, we wait for a small
966 amount of time to see if any other transactions can piggyback on
967 the synchronous write. The algorithm used is designed to auto‐
968 matically tune for the speed of the disk, by measuring the
969 amount of time (on average) that it takes to finish committing a
970 transaction. Call this time the "commit time". If the time that
971 the transaction has been running is less than the commit time,
972 ext4 will try sleeping for the commit time to see if other oper‐
973 ations will join the transaction. The commit time is capped by
974 the max_batch_time, which defaults to 15000us (15ms). This opti‐
975 mization can be turned off entirely by setting max_batch_time to
976 0.
977 min_batch_time=usec
979 This parameter sets the commit time (as described above) to be
980 at least min_batch_time. It defaults to zero microseconds.
981 Increasing this parameter may improve the throughput of multi-
982 threaded, synchronous workloads on very fast disks, at the cost
983 of increasing latency.
984 journal_ioprio=prio
986 The I/O priority (from 0 to 7, where 0 is the highest priorty)
987 which should be used for I/O operations submitted by kjournald2
988 during a commit operation. This defaults to 3, which is a
989 slightly higher priority than the default I/O priority.
990 auto_da_alloc|noauto_da_alloc
992 Many broken applications don't use fsync() when noauto_da_alloc
993 replacing existing files via patterns such as
994 fd = open("foo.new")/write(fd,..)/close(fd)/ rename("foo.new",
996 "foo")
997 or worse yet
999 fd = open("foo", O_TRUNC)/write(fd,..)/close(fd).
1001 If auto_da_alloc is enabled, ext4 will detect the replace-via-
1003 rename and replace-via-truncate patterns and force that any
1004 delayed allocation blocks are allocated such that at the next
1005 journal commit, in the default data=ordered mode, the data
1006 blocks of the new file are forced to disk before the rename()
1007 operation is committed. This provides roughly the same level of
1008 guarantees as ext3, and avoids the "zero-length" problem that
1009 can happen when a system crashes before the delayed allocation
1010 blocks are forced to disk.
1011
1014 (Note: fat is not a separate filesystem, but a common part of the
1015 msdos, umsdos and vfat filesystems.)
1016 blocksize={512|1024|2048}
1018 Set blocksize (default 512). This option is obsolete.
1019 uid=value and gid=value
1021 Set the owner and group of all files. (Default: the uid and gid
1022 of the current process.)
1023 umask=value
1025 Set the umask (the bitmask of the permissions that are not
1026 present). The default is the umask of the current process. The
1027 value is given in octal.
1028 dmask=value
1030 Set the umask applied to directories only. The default is the
1031 umask of the current process. The value is given in octal.
1032 fmask=value
1034 Set the umask applied to regular files only. The default is the
1035 umask of the current process. The value is given in octal.
1036 allow_utime=value
1038 This option controls the permission check of mtime/atime.
1039 20 If current process is in group of file's group ID, you
1041 can change timestamp.
1042 2 Other users can change timestamp.
1044 The default is set from `dmask' option. (If the directory is
1046 writable, utime(2) is also allowed. I.e. ~dmask & 022)
1047 Normally utime(2) checks current process is owner of the file,
1049 or it has CAP_FOWNER capability. But FAT filesystem doesn't
1050 have uid/gid on disk, so normal check is too unflexible. With
1051 this option you can relax it.
1052 check=value
1054 Three different levels of pickyness can be chosen:
1055 r[elaxed]
1057 Upper and lower case are accepted and equivalent, long
1058 name parts are truncated (e.g. verylongname.foobar
1059 becomes verylong.foo), leading and embedded spaces are
1060 accepted in each name part (name and extension).
1061 n[ormal]
1063 Like "relaxed", but many special characters (*, ?, <,
1064 spaces, etc.) are rejected. This is the default.
1065 s[trict]
1067 Like "normal", but names may not contain long parts and
1068 special characters that are sometimes used on Linux, but
1069 are not accepted by MS-DOS are rejected. (+, =, spaces,
1070 etc.)
1071 codepage=value
1073 Sets the codepage for converting to shortname characters on FAT
1074 and VFAT filesystems. By default, codepage 437 is used.
1075 conv={b[inary]|t[ext]|a[uto]}
1077 The fat filesystem can perform CRLF<-->NL (MS-DOS text format to
1078 UNIX text format) conversion in the kernel. The following con‐
1079 version modes are available:
1080 binary no translation is performed. This is the default.
1082 text CRLF<-->NL translation is performed on all files.
1084 auto CRLF<-->NL translation is performed on all files that
1086 don't have a "well-known binary" extension. The list of
1087 known extensions can be found at the beginning of
1088 fs/fat/misc.c (as of 2.0, the list is: exe, com, bin,
1089 app, sys, drv, ovl, ovr, obj, lib, dll, pif, arc, zip,
1090 lha, lzh, zoo, tar, z, arj, tz, taz, tzp, tpz, gz, tgz,
1091 deb, gif, bmp, tif, gl, jpg, pcx, tfm, vf, gf, pk, pxl,
1092 dvi).
1093 Programs that do computed lseeks won't like in-kernel text con‐
1095 version. Several people have had their data ruined by this
1096 translation. Beware!
1097 For filesystems mounted in binary mode, a conversion tool (from‐
1099 dos/todos) is available. This option is obsolete.
1100 cvf_format=module
1102 Forces the driver to use the CVF (Compressed Volume File) module
1103 cvf_module instead of auto-detection. If the kernel supports
1104 kmod, the cvf_format=xxx option also controls on-demand CVF mod‐
1105 ule loading. This option is obsolete.
1106 cvf_option=option
1108 Option passed to the CVF module. This option is obsolete.
1109 debug Turn on the debug flag. A version string and a list of filesys‐
1111 tem parameters will be printed (these data are also printed if
1112 the parameters appear to be inconsistent).
1113 fat={12|16|32}
1115 Specify a 12, 16 or 32 bit fat. This overrides the automatic
1116 FAT type detection routine. Use with caution!
1117 iocharset=value
1119 Character set to use for converting between 8 bit characters and
1120 16 bit Unicode characters. The default is iso8859-1. Long file‐
1121 names are stored on disk in Unicode format.
1122 tz=UTC This option disables the conversion of timestamps between local
1124 time (as used by Windows on FAT) and UTC (which Linux uses
1125 internally). This is particularly useful when mounting devices
1126 (like digital cameras) that are set to UTC in order to avoid the
1127 pitfalls of local time.
1128 quiet Turn on the quiet flag. Attempts to chown or chmod files do not
1130 return errors, although they fail. Use with caution!
1131 showexec
1133 If set, the execute permission bits of the file will be allowed
1134 only if the extension part of the name is .EXE, .COM, or .BAT.
1135 Not set by default.
1136 sys_immutable
1138 If set, ATTR_SYS attribute on FAT is handled as IMMUTABLE flag
1139 on Linux. Not set by default.
1140 flush If set, the filesystem will try to flush to disk more early than
1142 normal. Not set by default.
1143 usefree
1145 Use the "free clusters" value stored on FSINFO. It'll be used to
1146 determine number of free clusters without scanning disk. But
1147 it's not used by default, because recent Windows don't update it
1148 correctly in some case. If you are sure the "free clusters" on
1149 FSINFO is correct, by this option you can avoid scanning disk.
1150 dots, nodots, dotsOK=[yes|no]
1152 Various misguided attempts to force Unix or DOS conventions onto
1153 a FAT filesystem.
1154
1157 creator=cccc, type=cccc
1158 Set the creator/type values as shown by the MacOS finder used
1159 for creating new files. Default values: '????'.
1160 uid=n, gid=n
1162 Set the owner and group of all files. (Default: the uid and gid
1163 of the current process.)
1164 dir_umask=n, file_umask=n, umask=n
1166 Set the umask used for all directories, all regular files, or
1167 all files and directories. Defaults to the umask of the current
1168 process.
1169 session=n
1171 Select the CDROM session to mount. Defaults to leaving that
1172 decision to the CDROM driver. This option will fail with any‐
1173 thing but a CDROM as underlying device.
1174 part=n Select partition number n from the device. Only makes sense for
1176 CDROMS. Defaults to not parsing the partition table at all.
1177 quiet Don't complain about invalid mount options.
1179
1182 uid=value and gid=value
1183 Set the owner and group of all files. (Default: the uid and gid
1184 of the current process.)
1185 umask=value
1187 Set the umask (the bitmask of the permissions that are not
1188 present). The default is the umask of the current process. The
1189 value is given in octal.
1190 case={lower|asis}
1192 Convert all files names to lower case, or leave them. (Default:
1193 case=lower.)
1194 conv={binary|text|auto}
1196 For conv=text, delete some random CRs (in particular, all fol‐
1197 lowed by NL) when reading a file. For conv=auto, choose more or
1198 less at random between conv=binary and conv=text. For
1199 conv=binary, just read what is in the file. This is the default.
1200 nocheck
1202 Do not abort mounting when certain consistency checks fail.
1203
1206 ISO 9660 is a standard describing a filesystem structure to be used on
1207 CD-ROMs. (This filesystem type is also seen on some DVDs. See also the
1208 udf filesystem.)
1209 Normal iso9660 filenames appear in a 8.3 format (i.e., DOS-like
1211 restrictions on filename length), and in addition all characters are in
1212 upper case. Also there is no field for file ownership, protection,
1213 number of links, provision for block/character devices, etc.
1214 Rock Ridge is an extension to iso9660 that provides all of these unix
1216 like features. Basically there are extensions to each directory record
1217 that supply all of the additional information, and when Rock Ridge is
1218 in use, the filesystem is indistinguishable from a normal UNIX filesys‐
1219 tem (except that it is read-only, of course).
1220 norock Disable the use of Rock Ridge extensions, even if available. Cf.
1222 map.
1223 nojoliet
1225 Disable the use of Microsoft Joliet extensions, even if avail‐
1226 able. Cf. map.
1227 check={r[elaxed]|s[trict]}
1229 With check=relaxed, a filename is first converted to lower case
1230 before doing the lookup. This is probably only meaningful
1231 together with norock and map=normal. (Default: check=strict.)
1232 uid=value and gid=value
1234 Give all files in the filesystem the indicated user or group id,
1235 possibly overriding the information found in the Rock Ridge
1236 extensions. (Default: uid=0,gid=0.)
1237 map={n[ormal]|o[ff]|a[corn]}
1239 For non-Rock Ridge volumes, normal name translation maps upper
1240 to lower case ASCII, drops a trailing `;1', and converts `;' to
1241 `.'. With map=off no name translation is done. See norock.
1242 (Default: map=normal.) map=acorn is like map=normal but also
1243 apply Acorn extensions if present.
1244 mode=value
1246 For non-Rock Ridge volumes, give all files the indicated mode.
1247 (Default: read permission for everybody.) Since Linux 2.1.37
1248 one no longer needs to specify the mode in decimal. (Octal is
1249 indicated by a leading 0.)
1250 unhide Also show hidden and associated files. (If the ordinary files
1252 and the associated or hidden files have the same filenames, this
1253 may make the ordinary files inaccessible.)
1254 block={512|1024|2048}
1256 Set the block size to the indicated value. (Default:
1257 block=1024.)
1258 conv={a[uto]|b[inary]|m[text]|t[ext]}
1260 (Default: conv=binary.) Since Linux 1.3.54 this option has no
1261 effect anymore. (And non-binary settings used to be very dan‐
1262 gerous, possibly leading to silent data corruption.)
1263 cruft If the high byte of the file length contains other garbage, set
1265 this mount option to ignore the high order bits of the file
1266 length. This implies that a file cannot be larger than 16MB.
1267 session=x
1269 Select number of session on multisession CD. (Since 2.3.4.)
1270 sbsector=xxx
1272 Session begins from sector xxx. (Since 2.3.4.)
1273 The following options are the same as for vfat and specifying them only
1275 makes sense when using discs encoded using Microsoft's Joliet exten‐
1276 sions.
1277 iocharset=value
1279 Character set to use for converting 16 bit Unicode characters on
1280 CD to 8 bit characters. The default is iso8859-1.
1281 utf8 Convert 16 bit Unicode characters on CD to UTF-8.
1283
1286 iocharset=name
1287 Character set to use for converting from Unicode to ASCII. The
1288 default is to do no conversion. Use iocharset=utf8 for UTF8
1289 translations. This requires CONFIG_NLS_UTF8 to be set in the
1290 kernel .config file.
1291 resize=value
1293 Resize the volume to value blocks. JFS only supports growing a
1294 volume, not shrinking it. This option is only valid during a
1295 remount, when the volume is mounted read-write. The resize key‐
1296 word with no value will grow the volume to the full size of the
1297 partition.
1298 nointegrity
1300 Do not write to the journal. The primary use of this option is
1301 to allow for higher performance when restoring a volume from
1302 backup media. The integrity of the volume is not guaranteed if
1303 the system abnormally abends.
1304 integrity
1306 Default. Commit metadata changes to the journal. Use this
1307 option to remount a volume where the nointegrity option was pre‐
1308 viously specified in order to restore normal behavior.
1309 errors={continue|remount-ro|panic}
1311 Define the behaviour when an error is encountered. (Either
1312 ignore errors and just mark the filesystem erroneous and con‐
1313 tinue, or remount the filesystem read-only, or panic and halt
1314 the system.)
1315 noquota|quota|usrquota|grpquota
1317 These options are accepted but ignored.
1318
1321 None.
1322
1325 See mount options for fat. If the msdos filesystem detects an incon‐
1326 sistency, it reports an error and sets the file system read-only. The
1327 filesystem can be made writeable again by remounting it.
1328
1331 Just like nfs, the ncpfs implementation expects a binary argument (a
1332 struct ncp_mount_data) to the mount system call. This argument is con‐
1333 structed by ncpmount(8) and the current version of mount (2.12) does
1334 not know anything about ncpfs.
1335
1338 See the options section of the nfs(5) man page (nfs-utils package must
1339 be installed).
1340 The nfs and nfs4 implementation expects a binary argument (a struct
1342 nfs_mount_data) to the mount system call. This argument is constructed
1343 by mount.nfs(8) and the current version of mount (2.13) does not know
1344 anything about nfs and nfs4.
1345
1348 iocharset=name
1349 Character set to use when returning file names. Unlike VFAT,
1350 NTFS suppresses names that contain unconvertible characters.
1351 Deprecated.
1352 nls=name
1354 New name for the option earlier called iocharset.
1355 utf8 Use UTF-8 for converting file names.
1357 uni_xlate={0|1|2}
1359 For 0 (or `no' or `false'), do not use escape sequences for
1360 unknown Unicode characters. For 1 (or `yes' or `true') or 2,
1361 use vfat-style 4-byte escape sequences starting with ":". Here 2
1362 give a little-endian encoding and 1 a byteswapped bigendian
1363 encoding.
1364 posix=[0|1]
1366 If enabled (posix=1), the filesystem distinguishes between upper
1367 and lower case. The 8.3 alias names are presented as hard links
1368 instead of being suppressed. This option is obsolete.
1369 uid=value, gid=value and umask=value
1371 Set the file permission on the filesystem. The umask value is
1372 given in octal. By default, the files are owned by root and not
1373 readable by somebody else.
1374
1377 uid=value and gid=value
1378 These options are recognized, but have no effect as far as I can
1379 see.
1380
1383 Ramfs is a memory based filesystem. Mount it and you have it. Unmount
1384 it and it is gone. Present since Linux 2.3.99pre4. There are no mount
1385 options.
1386
1389 Reiserfs is a journaling filesystem.
1390 conv Instructs version 3.6 reiserfs software to mount a version 3.5
1392 filesystem, using the 3.6 format for newly created objects. This
1393 filesystem will no longer be compatible with reiserfs 3.5 tools.
1394 hash={rupasov|tea|r5|detect}
1396 Choose which hash function reiserfs will use to find files
1397 within directories.
1398 rupasov
1400 A hash invented by Yury Yu. Rupasov. It is fast and pre‐
1401 serves locality, mapping lexicographically close file
1402 names to close hash values. This option should not be
1403 used, as it causes a high probability of hash collisions.
1404 tea A Davis-Meyer function implemented by Jeremy
1406 Fitzhardinge. It uses hash permuting bits in the name.
1407 It gets high randomness and, therefore, low probability
1408 of hash collisions at some CPU cost. This may be used if
1409 EHASHCOLLISION errors are experienced with the r5 hash.
1410 r5 A modified version of the rupasov hash. It is used by
1412 default and is the best choice unless the filesystem has
1413 huge directories and unusual file-name patterns.
1414 detect Instructs mount to detect which hash function is in use
1416 by examining the filesystem being mounted, and to write
1417 this information into the reiserfs superblock. This is
1418 only useful on the first mount of an old format filesys‐
1419 tem.
1420 hashed_relocation
1422 Tunes the block allocator. This may provide performance improve‐
1423 ments in some situations.
1424 no_unhashed_relocation
1426 Tunes the block allocator. This may provide performance improve‐
1427 ments in some situations.
1428 noborder
1430 Disable the border allocator algorithm invented by Yury Yu.
1431 Rupasov. This may provide performance improvements in some sit‐
1432 uations.
1433 nolog Disable journalling. This will provide slight performance
1435 improvements in some situations at the cost of losing reiserfs's
1436 fast recovery from crashes. Even with this option turned on,
1437 reiserfs still performs all journalling operations, save for
1438 actual writes into its journalling area. Implementation of
1439 nolog is a work in progress.
1440 notail By default, reiserfs stores small files and `file tails'
1442 directly into its tree. This confuses some utilities such as
1443 LILO(8). This option is used to disable packing of files into
1444 the tree.
1445 replayonly
1447 Replay the transactions which are in the journal, but do not
1448 actually mount the filesystem. Mainly used by reiserfsck.
1449 resize=number
1451 A remount option which permits online expansion of reiserfs par‐
1452 titions. Instructs reiserfs to assume that the device has num‐
1453 ber blocks. This option is designed for use with devices which
1454 are under logical volume management (LVM). There is a special
1455 resizer utility which can be obtained from
1456 ftp://ftp.namesys.com/pub/reiserfsprogs.
1457 user_xattr
1459 Enable Extended User Attributes. See the attr(5) manual page.
1460 acl Enable POSIX Access Control Lists. See the acl(5) manual page.
1462 barrier=none / barrier=flush
1464 This enables/disables the use of write barriers in the journal‐
1465 ing code. barrier=none disables it, barrier=flush enables it.
1466 Write barriers enforce proper on-disk ordering of journal com‐
1467 mits, making volatile disk write caches safe to use, at some
1468 performance penalty. The reiserfs filesystem does not enable
1469 write barriers by default. Be sure to enable barriers unless
1470 your disks are battery-backed one way or another. Otherwise you
1471 risk filesystem corruption in case of power failure.
1472
1475 None.
1476
1479 None.
1480
1483 Just like nfs, the smbfs implementation expects a binary argument (a
1484 struct smb_mount_data) to the mount system call. This argument is con‐
1485 structed by smbmount(8) and the current version of mount (2.12) does
1486 not know anything about smbfs.
1487
1490 None.
1491
1494 size=nbytes
1495 Override default maximum size of the filesystem. The size is
1496 given in bytes, and rounded up to entire pages. The default is
1497 half of the memory. The size parameter also accepts a suffix %
1498 to limit this tmpfs instance to that percentage of your physical
1499 RAM: the default, when neither size nor nr_blocks is specified,
1500 is size=50%
1501 nr_blocks=
1503 The same as size, but in blocks of PAGE_CACHE_SIZE
1504 nr_inodes=
1506 The maximum number of inodes for this instance. The default is
1507 half of the number of your physical RAM pages, or (on a machine
1508 with highmem) the number of lowmem RAM pages, whichever is the
1509 lower.
1510 The tmpfs mount options for sizing ( size, nr_blocks, and nr_inodes)
1512 accept a suffix k, m or g for Ki, Mi, Gi (binary kilo, mega and giga)
1513 and can be changed on remount.
1514 mode= Set initial permissions of the root directory.
1517 uid= The user id.
1519 gid= The group id.
1521 mpol=[default|prefer:Node|bind:NodeList|interleave|interleave:NodeList]
1523 Set the NUMA memory allocation policy for all files in that
1524 instance (if the kernel CONFIG_NUMA is enabled) - which can be
1525 adjusted on the fly via 'mount -o remount ...'
1526 default
1528 prefers to allocate memory from the local node
1529 prefer:Node
1531 prefers to allocate memory from the given Node
1532 bind:NodeList
1534 allocates memory only from nodes in NodeList
1535 interleave
1537 prefers to allocate from each node in turn
1538 interleave:NodeList
1540 allocates from each node of NodeList in turn.
1541 The NodeList format is a comma-separated list of decimal numbers
1543 and ranges, a range being two hyphen-separated decimal numbers,
1544 the smallest and largest node numbers in the range. For exam‐
1545 ple, mpol=bind:0-3,5,7,9-15
1546 Note that trying to mount a tmpfs with an mpol option will fail
1548 if the running kernel does not support NUMA; and will fail if
1549 its nodelist specifies a node which is not online. If your sys‐
1550 tem relies on that tmpfs being mounted, but from time to time
1551 runs a kernel built without NUMA capability (perhaps a safe
1552 recovery kernel), or with fewer nodes online, then it is advis‐
1553 able to omit the mpol option from automatic mount options. It
1554 can be added later, when the tmpfs is already mounted on Mount‐
1555 Point, by 'mount -o remount,mpol=Policy:NodeList MountPoint'.
1556
1559 UBIFS is a flash file system which works on top of UBI volumes. Note
1560 that atime is not supported and is always turned off.
1561 The device name may be specified as
1563 ubiX_Y UBI device number X, volume number Y
1564 ubiY UBI device number 0, volume number Y
1566 ubiX:NAME
1568 UBI device number X, volume with name NAME
1569 ubi:NAME
1571 UBI device number 0, volume with name NAME
1572 Alternative ! separator may be used instead of :.
1573 The following mount options are available:
1575 bulk_read
1577 Enable bulk-read. VFS read-ahead is disabled because it slows
1578 down the file system. Bulk-Read is an internal optimization.
1579 Some flashes may read faster if the data are read at one go,
1580 rather than at several read requests. For example, OneNAND can
1581 do "read-while-load" if it reads more than one NAND page.
1582 no_bulk_read
1584 Do not bulk-read. This is the default.
1585 chk_data_crc
1587 Check data CRC-32 checksums. This is the default.
1588 no_chk_data_crc.
1590 Do not check data CRC-32 checksums. With this option, the
1591 filesystem does not check CRC-32 checksum for data, but it does
1592 check it for the internal indexing information. This option only
1593 affects reading, not writing. CRC-32 is always calculated when
1594 writing the data.
1595 compr={none|lzo|zlib}
1597 Select the default compressor which is used when new files are
1598 written. It is still possible to read compressed files if
1599 mounted with the none option.
1600
1603 udf is the "Universal Disk Format" filesystem defined by the Optical
1604 Storage Technology Association, and is often used for DVD-ROM. See
1605 also iso9660.
1606 gid= Set the default group.
1608 umask= Set the default umask. The value is given in octal.
1610 uid= Set the default user.
1612 unhide Show otherwise hidden files.
1614 undelete
1616 Show deleted files in lists.
1617 nostrict
1619 Unset strict conformance.
1620 iocharset
1622 Set the NLS character set.
1623 bs= Set the block size. (May not work unless 2048.)
1625 novrs Skip volume sequence recognition.
1627 session=
1629 Set the CDROM session counting from 0. Default: last session.
1630 anchor=
1632 Override standard anchor location. Default: 256.
1633 volume=
1635 Override the VolumeDesc location. (unused)
1636 partition=
1638 Override the PartitionDesc location. (unused)
1639 lastblock=
1641 Set the last block of the filesystem.
1642 fileset=
1644 Override the fileset block location. (unused)
1645 rootdir=
1647 Override the root directory location. (unused)
1648
1651 ufstype=value
1652 UFS is a filesystem widely used in different operating systems.
1653 The problem are differences among implementations. Features of
1654 some implementations are undocumented, so its hard to recognize
1655 the type of ufs automatically. That's why the user must specify
1656 the type of ufs by mount option. Possible values are:
1657 old Old format of ufs, this is the default, read only.
1659 (Don't forget to give the -r option.)
1660 44bsd For filesystems created by a BSD-like system (Net‐
1662 BSD,FreeBSD,OpenBSD).
1663 sun For filesystems created by SunOS or Solaris on Sparc.
1665 sunx86 For filesystems created by Solaris on x86.
1667 hp For filesystems created by HP-UX, read-only.
1669 nextstep
1671 For filesystems created by NeXTStep (on NeXT station)
1672 (currently read only).
1673 nextstep-cd
1675 For NextStep CDROMs (block_size == 2048), read-only.
1676 openstep
1678 For filesystems created by OpenStep (currently read
1679 only). The same filesystem type is also used by Mac OS
1680 X.
1681 onerror=value
1684 Set behaviour on error:
1685 panic If an error is encountered, cause a kernel panic.
1687 [lock|umount|repair]
1689 These mount options don't do anything at present; when an
1690 error is encountered only a console message is printed.
1691
1694 See mount options for msdos. The dotsOK option is explicitly killed by
1695 umsdos.
1696
1699 First of all, the mount options for fat are recognized. The dotsOK
1700 option is explicitly killed by vfat. Furthermore, there are
1701 uni_xlate
1703 Translate unhandled Unicode characters to special escaped
1704 sequences. This lets you backup and restore filenames that are
1705 created with any Unicode characters. Without this option, a '?'
1706 is used when no translation is possible. The escape character is
1707 ':' because it is otherwise illegal on the vfat filesystem. The
1708 escape sequence that gets used, where u is the unicode charac‐
1709 ter, is: ':', (u & 0x3f), ((u>>6) & 0x3f), (u>>12).
1710 posix Allow two files with names that only differ in case. This
1712 option is obsolete.
1713 nonumtail
1715 First try to make a short name without sequence number, before
1716 trying name~num.ext.
1717 utf8 UTF8 is the filesystem safe 8-bit encoding of Unicode that is
1719 used by the console. It can be be enabled for the filesystem
1720 with this option or disabled with utf8=0, utf8=no or utf8=false.
1721 If `uni_xlate' gets set, UTF8 gets disabled.
1722 shortname={lower|win95|winnt|mixed}
1724 Defines the behaviour for creation and display of filenames
1726 which fit into 8.3 characters. If a long name for a file exists,
1727 it will always be preferred display. There are four modes: :
1728 lower Force the short name to lower case upon display; store a
1730 long name when the short name is not all upper case. This
1731 mode is the default.
1732 win95 Force the short name to upper case upon display; store a
1734 long name when the short name is not all upper case.
1735 winnt Display the shortname as is; store a long name when the
1737 short name is not all lower case or all upper case.
1738 mixed Display the short name as is; store a long name when the
1740 short name is not all upper case.
1741
1745 devuid=uid and devgid=gid and devmode=mode
1746 Set the owner and group and mode of the device files in the
1747 usbfs filesystem (default: uid=gid=0, mode=0644). The mode is
1748 given in octal.
1749 busuid=uid and busgid=gid and busmode=mode
1751 Set the owner and group and mode of the bus directories in the
1752 usbfs filesystem (default: uid=gid=0, mode=0555). The mode is
1753 given in octal.
1754 listuid=uid and listgid=gid and listmode=mode
1756 Set the owner and group and mode of the file devices (default:
1757 uid=gid=0, mode=0444). The mode is given in octal.
1758
1761 None.
1762
1765 allocsize=size
1766 Sets the buffered I/O end-of-file preallocation size when doing
1767 delayed allocation writeout (default size is 64KiB). Valid val‐
1768 ues for this option are page size (typically 4KiB) through to
1769 1GiB, inclusive, in power-of-2 increments.
1770 attr2|noattr2
1772 The options enable/disable (default is enabled) an "opportunis‐
1773 tic" improvement to be made in the way inline extended
1774 attributes are stored on-disk. When the new form is used for
1775 the first time (by setting or removing extended attributes) the
1776 on-disk superblock feature bit field will be updated to reflect
1777 this format being in use.
1778 barrier
1780 Enables the use of block layer write barriers for writes into
1781 the journal and unwritten extent conversion. This allows for
1782 drive level write caching to be enabled, for devices that sup‐
1783 port write barriers.
1784 dmapi Enable the DMAPI (Data Management API) event callouts. Use with
1786 the mtpt option.
1787 grpid|bsdgroups and nogrpid|sysvgroups
1789 These options define what group ID a newly created file gets.
1790 When grpid is set, it takes the group ID of the directory in
1791 which it is created; otherwise (the default) it takes the fsgid
1792 of the current process, unless the directory has the setgid bit
1793 set, in which case it takes the gid from the parent directory,
1794 and also gets the setgid bit set if it is a directory itself.
1795 ihashsize=value
1797 Sets the number of hash buckets available for hashing the in-
1798 memory inodes of the specified mount point. If a value of zero
1799 is used, the value selected by the default algorithm will be
1800 displayed in /proc/mounts.
1801 ikeep|noikeep
1803 When inode clusters are emptied of inodes, keep them around on
1804 the disk (ikeep) - this is the traditional XFS behaviour and is
1805 still the default for now. Using the noikeep option, inode
1806 clusters are returned to the free space pool.
1807 inode64
1809 Indicates that XFS is allowed to create inodes at any location
1810 in the filesystem, including those which will result in inode
1811 numbers occupying more than 32 bits of significance. This is
1812 provided for backwards compatibility, but causes problems for
1813 backup applications that cannot handle large inode numbers.
1814 largeio|nolargeio
1816 If nolargeio is specified, the optimal I/O reported in st_blk‐
1817 size by stat(2) will be as small as possible to allow user
1818 applications to avoid inefficient read/modify/write I/O. If
1819 largeio is specified, a filesystem that has a swidth specified
1820 will return the swidth value (in bytes) in st_blksize. If the
1821 filesystem does not have a swidth specified but does specify an
1822 allocsize then allocsize (in bytes) will be returned instead.
1823 If neither of these two options are specified, then filesystem
1824 will behave as if nolargeio was specified.
1825 logbufs=value
1827 Set the number of in-memory log buffers. Valid numbers range
1828 from 2-8 inclusive. The default value is 8 buffers for any
1829 recent kernel.
1830 logbsize=value
1832 Set the size of each in-memory log buffer. Size may be speci‐
1833 fied in bytes, or in kilobytes with a "k" suffix. Valid sizes
1834 for version 1 and version 2 logs are 16384 (16k) and 32768
1835 (32k). Valid sizes for version 2 logs also include 65536 (64k),
1836 131072 (128k) and 262144 (256k). The default value for any
1837 recent kernel is 32768.
1838 logdev=device and rtdev=device
1840 Use an external log (metadata journal) and/or real-time device.
1841 An XFS filesystem has up to three parts: a data section, a log
1842 section, and a real-time section. The real-time section is
1843 optional, and the log section can be separate from the data sec‐
1844 tion or contained within it. Refer to xfs(5).
1845 mtpt=mountpoint
1847 Use with the dmapi option. The value specified here will be
1848 included in the DMAPI mount event, and should be the path of the
1849 actual mountpoint that is used.
1850 noalign
1852 Data allocations will not be aligned at stripe unit boundaries.
1853 noatime
1855 Access timestamps are not updated when a file is read.
1856 norecovery
1858 The filesystem will be mounted without running log recovery. If
1859 the filesystem was not cleanly unmounted, it is likely to be
1860 inconsistent when mounted in norecovery mode. Some files or
1861 directories may not be accessible because of this. Filesystems
1862 mounted norecovery must be mounted read-only or the mount will
1863 fail.
1864 nouuid Don't check for double mounted filesystems using the filesystem
1866 uuid. This is useful to mount LVM snapshot volumes.
1867 osyncisosync
1869 Make O_SYNC writes implement true O_SYNC. WITHOUT this option,
1870 Linux XFS behaves as if an osyncisdsync option is used, which
1871 will make writes to files opened with the O_SYNC flag set behave
1872 as if the O_DSYNC flag had been used instead. This can result
1873 in better performance without compromising data safety. However
1874 if this option is not in effect, timestamp updates from O_SYNC
1875 writes can be lost if the system crashes. If timestamp updates
1876 are critical, use the osyncisosync option.
1877 uquota|usrquota|uqnoenforce|quota
1879 User disk quota accounting enabled, and limits (optionally)
1880 enforced. Refer to xfs_quota(8) for further details.
1881 gquota|grpquota|gqnoenforce
1883 Group disk quota accounting enabled and limits (optionally)
1884 enforced. Refer to xfs_quota(8) for further details.
1885 pquota|prjquota|pqnoenforce
1887 Project disk quota accounting enabled and limits (optionally)
1888 enforced. Refer to xfs_quota(8) for further details.
1889 sunit=value and swidth=value
1891 Used to specify the stripe unit and width for a RAID device or a
1892 stripe volume. value must be specified in 512-byte block units.
1893 If this option is not specified and the filesystem was made on a
1894 stripe volume or the stripe width or unit were specified for the
1895 RAID device at mkfs time, then the mount system call will
1896 restore the value from the superblock. For filesystems that are
1897 made directly on RAID devices, these options can be used to
1898 override the information in the superblock if the underlying
1899 disk layout changes after the filesystem has been created. The
1900 swidth option is required if the sunit option has been speci‐
1901 fied, and must be a multiple of the sunit value.
1902 swalloc
1904 Data allocations will be rounded up to stripe width boundaries
1905 when the current end of file is being extended and the file size
1906 is larger than the stripe width size.
1907
1910 None. Although nothing is wrong with xiafs, it is not used much, and is
1911 not maintained. Probably one shouldn't use it. Since Linux version
1912 2.1.21 xiafs is no longer part of the kernel source.
1913
1916 One further possible type is a mount via the loop device. For example,
1917 the command
1918 mount /tmp/disk.img /mnt -t vfat -o loop=/dev/loop
1920 will set up the loop device /dev/loop3 to correspond to the file
1922 /tmp/disk.img, and then mount this device on /mnt.
1923 If no explicit loop device is mentioned (but just an option `-o loop'
1925 is given), then mount will try to find some unused loop device and use
1926 that, for example
1927 mount /tmp/disk.img /mnt -o loop
1929 The mount command automatically creates a loop device from a regular
1931 file if a filesystem type is not specified or the filesystem is known
1932 for libblkid, for example:
1933 mount /tmp/disk.img /mnt
1935 mount -t ext3 /tmp/disk.img /mnt
1937 This type of mount knows about four options, namely loop, offset, size‐
1939 limit and encryption, that are really options to losetup(8). (These
1940 options can be used in addition to those specific to the filesystem
1941 type.)
1942 Since Linux 2.6.25 is supported auto-destruction of loop devices and
1944 then any loop device allocated by mount will be freed by umount inde‐
1945 pendently on /etc/mtab.
1946 You can also free a loop device by hand, using `losetup -d' or `umount
1948 -d`.
1949
1952 mount has the following return codes (the bits can be ORed):
1953 0 success
1955 1 incorrect invocation or permissions
1957 2 system error (out of memory, cannot fork, no more loop devices)
1959 4 internal mount bug
1961 8 user interrupt
1963 16 problems writing or locking /etc/mtab
1965 32 mount failure
1967 64 some mount succeeded
1969
1972 The syntax of external mount helpers is:
1973 /sbin/mount.<suffix> spec dir [-sfnv] [-o options]
1975 where the <suffix> is filesystem type and -sfnvo options have same
1977 meaning like standard mount options.
1978
1981 /etc/fstab filesystem table
1982 /etc/mtab table of mounted filesystems
1984 /etc/mtab~ lock file
1986 /etc/mtab.tmp temporary file
1988 /etc/filesystems a list of filesystem types to try
1990
1993 mount(2), umount(2), fstab(5), umount(8), swapon(8), nfs(5), xfs(5),
1994 e2label(8), xfs_admin(8), mountd(8), nfsd(8), mke2fs(8), tune2fs(8),
1995 losetup(8)
1996
1998 It is possible for a corrupted filesystem to cause a crash.
1999 Some Linux filesystems don't support -o sync and -o dirsync (the ext2,
2001 ext3, fat and vfat filesystems do support synchronous updates (a la
2002 BSD) when mounted with the sync option).
2003 The -o remount may not be able to change mount parameters (all ext2fs-
2005 specific parameters, except sb, are changeable with a remount, for
2006 example, but you can't change gid or umask for the fatfs).
2007 Mount by label or uuid will work only if your devices have the names
2009 listed in /proc/partitions. In particular, it may well fail if the
2010 kernel was compiled with devfs but devfs is not mounted.
2011 It is possible that files /etc/mtab and /proc/mounts don't match. The
2013 first file is based only on the mount command options, but the content
2014 of the second file also depends on the kernel and others settings (e.g.
2015 remote NFS server. In particular case the mount command may reports
2016 unreliable information about a NFS mount point and the /proc/mounts
2017 file usually contains more reliable information.)
2018 Checking files on NFS filesystem referenced by file descriptors (i.e.
2020 the fcntl and ioctl families of functions) may lead to inconsistent
2021 result due to the lack of consistency check in kernel even if noac is
2022 used.
2023
2025 A mount command existed in Version 5 AT&T UNIX.
2026
2028 The mount command is part of the util-linux-ng package and is available
2029 from ftp://ftp.kernel.org/pub/linux/utils/util-linux-ng/.
2030Linux 2.6 2004-12-16 MOUNT(8)