1MOUNT(2) Linux Programmer's Manual MOUNT(2)
2
6 mount, umount - mount and unmount filesystems
7
9 #include <sys/mount.h>
10 int mount(const char *source, const char *target,
12 const char *filesystemtype, unsigned long mountflags,
13 const void *data);
14 int umount(const char *target);
16 int umount2(const char *target, int flags);
18
20 mount() attaches the filesystem specified by source (which is often a
21 device name, but can also be a directory name or a dummy) to the direc‐
22 tory specified by target.
23 umount() and umount2() remove the attachment of the (topmost) filesys‐
25 tem mounted on target.
26 Appropriate privilege (Linux: the CAP_SYS_ADMIN capability) is required
28 to mount and unmount filesystems.
29 Since Linux 2.4 a single filesystem can be visible at multiple mount
31 points, and multiple mounts can be stacked on the same mount point.
32 Values for the filesystemtype argument supported by the kernel are
34 listed in /proc/filesystems (like "minix", "ext2", "msdos", "proc",
35 "nfs", "iso9660" etc.). Further types may become available when the
36 appropriate modules are loaded.
37 The mountflags argument may have the magic number 0xC0ED (MS_MGC_VAL)
39 in the top 16 bits (this was required in kernel versions prior to 2.4,
40 but is no longer required and ignored if specified), and various mount
41 flags (as defined in <linux/fs.h> for libc4 and libc5 and in
42 <sys/mount.h> for glibc2) in the low order 16 bits:
43 MS_BIND
45 (Linux 2.4 onwards) Perform a bind mount, making a file or a
46 directory subtree visible at another point within a file system.
47 Bind mounts may cross file system boundaries and span chroot(2)
48 jails. The filesystemtype, mountflags, and data arguments are
49 ignored.
50 MS_DIRSYNC (since Linux 2.5.19)
52 Make directory changes on this file system synchronous. (This
53 property can be obtained for individual directories or subtrees
54 using chattr(8).)
55 MS_MANDLOCK
57 Permit mandatory locking on files in this file system. (Manda‐
58 tory locking must still be enabled on a per-file basis, as
59 described in fcntl(2).)
60 MS_MOVE
62 Move a subtree. source specifies an existing mount point and
63 target specifies the new location. The move is atomic: at no
64 point is the subtree unmounted. The filesystemtype, mountflags,
65 and data arguments are ignored.
66 MS_NOATIME
68 Do not update access times for (all types of) files on this file
69 system.
70 MS_NODEV
72 Do not allow access to devices (special files) on this file sys‐
73 tem.
74 MS_NODIRATIME
76 Do not update access times for directories on this file system.
77 MS_NOEXEC
79 Do not allow programs to be executed from this file system.
80 MS_NOSUID
82 Do not honour set-user-ID and set-group-ID bits when executing
83 programs from this file system.
84 MS_RDONLY
86 Mount file system read-only.
87 MS_RELATIME(Since Linux 2.6.20)
89 When a file on this file system is accessed, only update the
90 file's last accessed time (atime) if the current value of atime
91 is less than or equal to the file's last modified (mtime) or
92 last status change time (ctime). This option is useful for pro‐
93 grams, such as mutt(1), that need to know when a file has been
94 read since it was last modified.
95 MS_REMOUNT
97 Remount an existing mount. This is allows you to change the
98 mountflags and data of an existing mount without having to
99 unmount and remount the file system. source and target should
100 be the same values specified in the initial mount() call;
101 filesystemtype is ignored.
102 The following mountflags can be changed: MS_RDONLY, MS_SYNCHRO‐
104 NOUS, MS_MANDLOCK; before kernel 2.6.16, the following could
105 also be changed: MS_NOATIME and MS_NODIRATIME; and, addition‐
106 ally, before kernel 2.4, the following could also be changed:
107 MS_NOSUID, MS_NODEV, MS_NOEXEC.
108 MS_SYNCHRONOUS
110 Make writes on this file system synchronous (as though the
111 O_SYNC flag to open(2) was specified for all file opens to this
112 file system).
113 From Linux 2.4 onwards, the MS_NODEV, MS_NOEXEC, and MS_NOSUID flags
115 are settable on a per-mount-point basis. From kernel 2.6.16 onwards,
116 MS_NOATIME and MS_NODIRATIME are also settable on a per-mount-point
117 basis. The MS_RELATIME flag is also settable on a per-mount-point
118 basis.
119 The data argument is interpreted by the different file systems. Typi‐
121 cally it is a string of comma-separated options understood by this file
122 system. See mount(8) for details of the options available for each
123 filesystem type.
124 Linux 2.1.116 added the umount2() system call, which, like umount(),
126 unmounts a target, but allows additional flags controlling the behav‐
127 iour of the operation. Now there is only one possible flag:
128 MNT_FORCE (since Linux 2.1.116)
130 Force unmount even if busy. This can cause data loss. (Only
131 for NFS mounts.)
132 In previous version of kernel there were two more possible
134 flags: MNT_DETACH (since Linux 2.4.11) Perform a lazy unmount:
135 make the mount point unavailable for new accesses, and actually
136 perform the unmount when the mount point ceases to be busy.
137 MNT_EXPIRE (since Linux 2.6.8)
139 Mark the mount point as expired. If a mount point is not cur‐
140 rently in use, then an initial call to umount2() with this flag
141 fails with the error EAGAIN, but marks the mount point as
142 expired. The mount point remains expired as long as it isn't
143 accessed by any process. A second umount2() call specifying
144 MNT_EXPIRE unmounts an expired mount point. This flag cannot be
145 specified with either MNT_FORCE or MNT_DETACH.
146
148 On success, zero is returned. On error, -1 is returned, and errno is
149 set appropriately.
150
152 The error values given below result from filesystem type independent
153 errors. Each filesystem type may have its own special errors and its
154 own special behavior. See the kernel source code for details.
155 EACCES A component of a path was not searchable. (See also path_resolu‐
158 tion(2).) Or, mounting a read-only filesystem was attempted
159 without giving the MS_RDONLY flag. Or, the block device source
160 is located on a filesystem mounted with the MS_NODEV option.
161 EAGAIN A call to umount2() specifying MNT_EXPIRE successfully marked an
163 unbusy file system as expired.
164 EBUSY source is already mounted. Or, it cannot be remounted read-only,
166 because it still holds files open for writing. Or, it cannot be
167 mounted on target because target is still busy (it is the work‐
168 ing directory of some task, the mount point of another device,
169 has open files, etc.). Or, it could not be unmounted because it
170 is busy.
171 EFAULT One of the pointer arguments points outside the user address
173 space.
174 EINVAL source had an invalid superblock. Or, a remount (MS_REMOUNT)
176 was attempted, but source was not already mounted on target.
177 Or, a move (MS_MOVE) was attempted, but source was not a mount
178 point, or was '/'. Or, an unmount was attempted, but target was
179 not a mount point. Or, umount2() was called with MNT_EXPIRE and
180 either MNT_DETACH or MNT_FORCE.
181 ELOOP Too many link encountered during pathname resolution. Or, a
183 move was attempted, while target is a descendant of source.
184 EMFILE (In case no block device is required:) Table of dummy devices is
186 full.
187 ENAMETOOLONG
189 A pathname was longer than MAXPATHLEN.
190 ENODEV filesystemtype not configured in the kernel.
192 ENOENT A pathname was empty or had a nonexistent component.
194 ENOMEM The kernel could not allocate a free page to copy filenames or
196 data into.
197 ENOTBLK
199 source is not a block device (and a device was required).
200 ENOTDIR
202 The second argument, or a prefix of the first argument, is not a
203 directory.
204 ENXIO The major number of the block device source is out of range.
206 EPERM The caller does not have the required privileges.
208
210 These functions are Linux specific and should not be used in programs
211 intended to be portable.
212
214 The original umount() function was called as umount(device) and would
215 return ENOTBLK when called with something other than a block device.
216 In Linux 0.98p4 a call umount(dir) was added, in order to support
217 anonymous devices. In Linux 2.3.99-pre7 the call umount(device) was
218 removed, leaving only umount(dir) (since now devices can be mounted in
219 more than one place, so specifying the device does not suffice).
220 The original MS_SYNC flag was renamed MS_SYNCHRONOUS in 1.1.69 when a
222 different MS_SYNC was added to <mman.h>.
223 Before Linux 2.4 an attempt to execute a set-user-ID or set-group-ID
225 program on a filesystem mounted with MS_NOSUID would fail with EPERM.
226 Since Linux 2.4 the set-user-ID and set-group-ID bits are just silently
227 ignored in this case.
228
230 path_resolution(2), mount(8), umount(8)
231Linux 2.6.12 2004-05-18 MOUNT(2)