1MOUNT(2) Linux Programmer's Manual MOUNT(2)
2
6 mount - mount file system
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
16 mount() attaches the file system specified by source (which is often a
17 device name, but can also be a directory name or a dummy) to the direc‐
18 tory specified by target.
19 Appropriate privilege (Linux: the CAP_SYS_ADMIN capability) is required
21 to mount file systems.
22 Since Linux 2.4 a single file system can be visible at multiple mount
24 points, and multiple mounts can be stacked on the same mount point.
25 Values for the filesystemtype argument supported by the kernel are
27 listed in /proc/filesystems (like "minix", "ext2", "ext3", "jfs",
28 "xfs", "reiserfs", "msdos", "proc", "nfs", "iso9660" etc.). Further
29 types may become available when the appropriate modules are loaded.
30 The mountflags argument may have the magic number 0xC0ED (MS_MGC_VAL)
32 in the top 16 bits (this was required in kernel versions prior to 2.4,
33 but is no longer required and ignored if specified), and various mount
34 flags (as defined in <linux/fs.h> for libc4 and libc5 and in
35 <sys/mount.h> for glibc2) in the low order 16 bits:
36 MS_BIND (Linux 2.4 onwards)
38 Perform a bind mount, making a file or a directory subtree visi‐
39 ble at another point within a file system. Bind mounts may
40 cross file system boundaries and span chroot(2) jails. The
41 filesystemtype and data arguments are ignored. Up until Linux
42 2.6.26, mountflags was also ignored (the bind mount has the same
43 mount options as the underlying mount point). Since Linux
44 2.6.26, the MS_RDONLY flag is honored when making a bind mount.
45 MS_DIRSYNC (since Linux 2.5.19)
47 Make directory changes on this file system synchronous. (This
48 property can be obtained for individual directories or subtrees
49 using chattr(1).)
50 MS_MANDLOCK
52 Permit mandatory locking on files in this file system. (Manda‐
53 tory locking must still be enabled on a per-file basis, as
54 described in fcntl(2).)
55 MS_MOVE
57 Move a subtree. source specifies an existing mount point and
58 target specifies the new location. The move is atomic: at no
59 point is the subtree unmounted. The filesystemtype, mountflags,
60 and data arguments are ignored.
61 MS_NOATIME
63 Do not update access times for (all types of) files on this file
64 system.
65 MS_NODEV
67 Do not allow access to devices (special files) on this file sys‐
68 tem.
69 MS_NODIRATIME
71 Do not update access times for directories on this file system.
72 This flag provides a subset of the functionality provided by
73 MS_NOATIME; that is, MS_NOATIME implies MS_NODIRATIME.
74 MS_NOEXEC
76 Do not allow programs to be executed from this file system.
77 MS_NOSUID
79 Do not honor set-user-ID and set-group-ID bits when executing
80 programs from this file system.
81 MS_RDONLY
83 Mount file system read-only.
84 MS_RELATIME (Since Linux 2.6.20)
86 When a file on this file system is accessed, only update the
87 file's last access time (atime) if the current value of atime is
88 less than or equal to the file's last modification time (mtime)
89 or last status change time (ctime). This option is useful for
90 programs, such as mutt(1), that need to know when a file has
91 been read since it was last modified. Since Linux 2.6.30, the
92 kernel defaults to the behavior provided by this flag (unless
93 MS_NOATIME was specified), and the MS_STRICTATIME flag is
94 required to obtain traditional semantics. In addition, since
95 Linux 2.6.30, the file's last access time is always updated if
96 it is more than 1 day old.
97 MS_REMOUNT
99 Remount an existing mount. This allows you to change the mount‐
100 flags and data of an existing mount without having to unmount
101 and remount the file system. source and target should be the
102 same values specified in the initial mount() call; filesystem‐
103 type is ignored.
104 The following mountflags can be changed: MS_RDONLY, MS_SYNCHRO‐
106 NOUS, MS_MANDLOCK; before kernel 2.6.16, the following could
107 also be changed: MS_NOATIME and MS_NODIRATIME; and, addition‐
108 ally, before kernel 2.4.10, the following could also be changed:
109 MS_NOSUID, MS_NODEV, MS_NOEXEC.
110 MS_SILENT (since Linux 2.6.17)
112 Suppress the display of certain (printk()) warning messages in
113 the kernel log. This flag supersedes the misnamed and obsolete
114 MS_VERBOSE flag (available since Linux 2.4.12), which has the
115 same meaning.
116 MS_STRICTATIME (Since Linux 2.6.30)
118 Always update the last access time (atime) when files on this
119 file system are accessed. (This was the default behavior before
120 Linux 2.6.30.) Specifying this flag overrides the effect of
121 setting the MS_NOATIME and MS_RELATIME flags.
122 MS_SYNCHRONOUS
124 Make writes on this file system synchronous (as though the
125 O_SYNC flag to open(2) was specified for all file opens to this
126 file system).
127 From Linux 2.4 onwards, the MS_NODEV, MS_NOEXEC, and MS_NOSUID flags
129 are settable on a per-mount-point basis. From kernel 2.6.16 onwards,
130 MS_NOATIME and MS_NODIRATIME are also settable on a per-mount-point
131 basis. The MS_RELATIME flag is also settable on a per-mount-point
132 basis.
133 The data argument is interpreted by the different file systems. Typi‐
135 cally it is a string of comma-separated options understood by this file
136 system. See mount(8) for details of the options available for each
137 filesystem type.
138
140 On success, zero is returned. On error, -1 is returned, and errno is
141 set appropriately.
142
144 The error values given below result from filesystem type independent
145 errors. Each filesystem type may have its own special errors and its
146 own special behavior. See the kernel source code for details.
147 EACCES A component of a path was not searchable. (See also path_reso‐
149 lution(7).) Or, mounting a read-only filesystem was attempted
150 without giving the MS_RDONLY flag. Or, the block device source
151 is located on a filesystem mounted with the MS_NODEV option.
152 EBUSY source is already mounted. Or, it cannot be remounted read-
154 only, because it still holds files open for writing. Or, it
155 cannot be mounted on target because target is still busy (it is
156 the working directory of some task, the mount point of another
157 device, has open files, etc.).
158 EFAULT One of the pointer arguments points outside the user address
160 space.
161 EINVAL source had an invalid superblock. Or, a remount (MS_REMOUNT)
163 was attempted, but source was not already mounted on target.
164 Or, a move (MS_MOVE) was attempted, but source was not a mount
165 point, or was '/'.
166 ELOOP Too many links encountered during pathname resolution. Or, a
168 move was attempted, while target is a descendant of source.
169 EMFILE (In case no block device is required:) Table of dummy devices is
171 full.
172 ENAMETOOLONG
174 A pathname was longer than MAXPATHLEN.
175 ENODEV filesystemtype not configured in the kernel.
177 ENOENT A pathname was empty or had a nonexistent component.
179 ENOMEM The kernel could not allocate a free page to copy filenames or
181 data into.
182 ENOTBLK
184 source is not a block device (and a device was required).
185 ENOTDIR
187 target, or a prefix of source, is not a directory.
188 ENXIO The major number of the block device source is out of range.
190 EPERM The caller does not have the required privileges.
192
194 This function is Linux-specific and should not be used in programs
195 intended to be portable.
196
198 The original MS_SYNC flag was renamed MS_SYNCHRONOUS in 1.1.69 when a
199 different MS_SYNC was added to <mman.h>.
200 Before Linux 2.4 an attempt to execute a set-user-ID or set-group-ID
202 program on a filesystem mounted with MS_NOSUID would fail with EPERM.
203 Since Linux 2.4 the set-user-ID and set-group-ID bits are just silently
204 ignored in this case.
205 Per-process Namespaces
207 Starting with kernel 2.4.19, Linux provides per-process mount names‐
208 paces. A mount namespace is the set of file system mounts that are
209 visible to a process. Mount-point namespaces can be (and usually are)
210 shared between multiple processes, and changes to the namespace (i.e.,
211 mounts and unmounts) by one process are visible to all other processes
212 sharing the same namespace. (The pre-2.4.19 Linux situation can be
213 considered as one in which there was a single namespace was shared by
214 every process on the system.)
215 A child process created by fork(2) shares its parent's mount namespace;
217 the mount namespace is preserved across an execve(2).
218 A process can obtain a private mount namespace if: it was created using
220 the clone() CLONE_NEWNS flag, in which case its new namespace is ini‐
221 tialized to be a copy of the namespace of the process that called
222 clone(); or it calls unshare(2) with the CLONE_NEWNS flag, which causes
223 the caller's mount namespace to obtain a private copy of the namespace
224 that it was previously sharing with other processes, so that future
225 mounts and unmounts by the caller are invisible to other processes
226 (except child processes that the caller subsequently creates) and vice
227 versa.
228 The Linux-specific /proc/PID/self file exposes the list of mount points
230 in the mount namespace of the process with the specified ID; see
231 proc(5) for details.
232
234 umount(2), path_resolution(7), mount(8), umount(8)
235
237 This page is part of release 3.22 of the Linux man-pages project. A
238 description of the project, and information about reporting bugs, can
239 be found at http://www.kernel.org/doc/man-pages/.
240Linux 2009-06-26 MOUNT(2)