man2/mount.2

   1 .\" Copyright (C) 1993 Rickard E. Faith <faith@cs.unc.edu>
   2 .\" and Copyright (C) 1994 Andries E. Brouwer <aeb@cwi.nl>
   3 .\" and Copyright (C) 2002, 2005, 2016 Michael Kerrisk <mtk.manpages@gmail.com>
   4 .\"
   5 .\" %%%LICENSE_START(VERBATIM)
   6 .\" Permission is granted to make and distribute verbatim copies of this
   7 .\" manual provided the copyright notice and this permission notice are
   8 .\" preserved on all copies.
   9 .\"
  10 .\" Permission is granted to copy and distribute modified versions of this
  11 .\" manual under the conditions for verbatim copying, provided that the
  12 .\" entire resulting derived work is distributed under the terms of a
  13 .\" permission notice identical to this one.
  14 .\"
  15 .\" Since the Linux kernel and libraries are constantly changing, this
  16 .\" manual page may be incorrect or out-of-date.  The author(s) assume no
  17 .\" responsibility for errors or omissions, or for damages resulting from
  18 .\" the use of the information contained herein.  The author(s) may not
  19 .\" have taken the same level of care in the production of this manual,
  20 .\" which is licensed free of charge, as they might when working
  21 .\" professionally.
  22 .\"
  23 .\" Formatted or processed versions of this manual, if unaccompanied by
  24 .\" the source, must acknowledge the copyright and authors of this work.
  25 .\" %%%LICENSE_END
  26 .\"
  27 .\" Modified 1996-11-04 by Eric S. Raymond <esr@thyrsus.com>
  28 .\" Modified 2001-10-13 by Michael Kerrisk <mtk.manpages@gmail.com>
  29 .\"     Added note on historical behavior of MS_NOSUID
  30 .\" Modified 2002-05-16 by Michael Kerrisk <mtk.manpages@gmail.com>
  31 .\"     Extensive changes and additions
  32 .\" Modified 2002-05-27 by aeb
  33 .\" Modified 2002-06-11 by Michael Kerrisk <mtk.manpages@gmail.com>
  34 .\"     Enhanced descriptions of MS_MOVE, MS_BIND, and MS_REMOUNT
  35 .\" Modified 2004-06-17 by Michael Kerrisk <mtk.manpages@gmail.com>
  36 .\" 2005-05-18, mtk, Added MNT_EXPIRE, plus a few other tidy-ups.
  37 .\" 2008-10-06, mtk: move umount*() material into separate umount.2 page.
  38 .\" 2008-10-06, mtk: Add discussion of namespaces.
  39 .\"
  40 .TH MOUNT 2 2016-12-12 "Linux" "Linux Programmer's Manual"
  41 .SH NAME
  42 mount \- mount filesystem
  43 .SH SYNOPSIS
  44 .nf
  45 .B "#include <sys/mount.h>"
  46 .sp
  47 .BI "int mount(const char *" source ", const char *" target ,
  48 .BI "          const char *" filesystemtype ", unsigned long " mountflags ,
  49 .BI "          const void *" data );
  50 .fi
  51 .SH DESCRIPTION
  52 .BR mount ()
  53 attaches the filesystem specified by
  54 .I source
  55 (which is often a pathname referring to a device,
  56 but can also be the pathname of a directory or file,
  57 or a dummy string) to the location (a directory or file)
  58 specified by the pathname in
  59 .IR target .
  60
  61 Appropriate privilege (Linux: the
  62 .B CAP_SYS_ADMIN
  63 capability) is required to mount filesystems.
  64
  65 Values for the
  66 .I filesystemtype
  67 argument supported by the kernel are listed in
  68 .I /proc/filesystems
  69 (e.g., "btrfs", "ext4", "jfs", "xfs", "vfat", "fuse",
  70 "tmpfs", "cgroup", "proc", "mqueue", "nfs", "cifs", "iso9660").
  71 Further types may become available when the appropriate modules
  72 are loaded.
  73
  74 The
  75 .I data
  76 argument is interpreted by the different filesystems.
  77 Typically it is a string of comma-separated options
  78 understood by this filesystem.
  79 See
  80 .BR mount (8)
  81 for details of the options available for each filesystem type.
  82
  83 A call to
  84 .BR mount ()
  85 performs one of a number of general types of operation,
  86 depending on the bits specified in
  87 .IR mountflags .
  88 The choice of which operation to perform is determined by
  89 testing the bits set in
  90 .IR mountflags ,
  91 with the tests being conducted in the order listed here:
  92 .IP * 3
  93 Remount an existing mount:
  94 .IR mountflags
  95 includes
  96 .BR MS_REMOUNT .
  97 .IP *
  98 Create a bind mount:
  99 .IR mountflags
 100 includes
 101 .BR MS_BIND .
 102 .IP *
 103 Change the propagation type of an existing mount:
 104 .IR mountflags
 105 includes one of
 106 .BR MS_SHARED ,
 107 .BR MS_PRIVATE ,
 108 .BR MS_SLAVE ,
 109 or
 110 .BR MS_UNBINDABLE .
 111 .IP *
 112 Move an existing mount to a new location:
 113 .IR mountflags
 114 includes
 115 .BR MS_MOVE .
 116 .IP *
 117 Create a new mount:
 118 .IR mountflags
 119 includes none of the above flags.
 120 .PP
 121 Each of these operations is detailed later in this page.
 122 Further flags may be specified in
 123 .IR mountflags
 124 to modify the behavior of
 125 .BR mount (),
 126 as described below.
 127 .\"
 128 .SS Additional mount flags
 129 The list below describes the additional flags that can be specified in
 130 .IR mountflags .
 131 Note that some operation types ignore some or all of these flags,
 132 as described later in this page.
 133 .\"
 134 .\" FIXME 2.6.25 Added MS_I_VERSION, which needs to be documented.
 135 .\"
 136 .TP
 137 .BR MS_DIRSYNC " (since Linux 2.5.19)"
 138 Make directory changes on this filesystem synchronous.
 139 (This property can be obtained for individual directories
 140 or subtrees using
 141 .BR chattr (1).)
 142 .TP
 143 .BR MS_LAZYTIME " (since Linux 4.0)"
 144 .\" commit 0ae45f63d4ef8d8eeec49c7d8b44a1775fff13e8
 145 .\" commit fe032c422c5ba562ba9c2d316f55e258e03259c6
 146 .\" commit a26f49926da938f47561f386be56a83dd37a496d
 147 Reduce on-disk updates of inode timestamps (atime, mtime, ctime)
 148 by maintaining these changes only in memory.
 149 The on-disk timestamps are updated only when:
 150 .RS
 151 .IP (a) 5
 152 the inode needs to be updated for some change unrelated to file timestamps;
 153 .IP (b)
 154 the application employs
 155 .BR fsync (2),
 156 .BR syncfs (2),
 157 or
 158 .BR sync (2);
 159 .IP (c)
 160 an undeleted inode is evicted from memory; or
 161 .IP (d)
 162 more than 24 hours have passed since the inode was written to disk.
 163 .RE
 164 .IP
 165 This mount option significantly reduces writes
 166 needed to update the inode's timestamps, especially mtime and atime.
 167 However, in the event of a system crash, the atime and mtime fields
 168 on disk might be out of date by up to 24 hours.
 169
 170 Examples of workloads where this option could be of significant benefit
 171 include frequent random writes to preallocated files,
 172 as well as cases where the
 173 .B MS_STRICTATIME
 174 mount option is also enabled.
 175 (The advantage of combining
 176 .BR MS_STRICTATIME
 177 and
 178 .BR MS_LAZYTIME
 179 is that
 180 .BR stat (2)
 181 will return the correctly updated atime, but the atime updates
 182 will be flushed to disk only in the cases listed above.)
 183 .TP
 184 .B MS_MANDLOCK
 185 Permit mandatory locking on files in this filesystem.
 186 (Mandatory locking must still be enabled on a per-file basis,
 187 as described in
 188 .BR fcntl (2).)
 189 Since Linux 4.5,
 190 .\" commit 95ace75414f312f9a7b93d873f386987b92a5301
 191 this mount option requires the
 192 .B CAP_SYS_ADMIN
 193 capability.
 194 .TP
 195 .B MS_NOATIME
 196 Do not update access times for (all types of) files on this filesystem.
 197 .TP
 198 .B MS_NODEV
 199 Do not allow access to devices (special files) on this filesystem.
 200 .TP
 201 .B MS_NODIRATIME
 202 Do not update access times for directories on this filesystem.
 203 This flag provides a subset of the functionality provided by
 204 .BR MS_NOATIME ;
 205 that is,
 206 .BR MS_NOATIME
 207 implies
 208 .BR MS_NODIRATIME .
 209 .TP
 210 .B MS_NOEXEC
 211 Do not allow programs to be executed from this filesystem.
 212 .\" (Possibly useful for a filesystem that contains non-Linux executables.
 213 .\" Often used as a security feature, e.g., to make sure that restricted
 214 .\" users cannot execute files uploaded using ftp or so.)
 215 .TP
 216 .B MS_NOSUID
 217 Do not honor set-user-ID and set-group-ID bits or file capabilities
 218 when executing programs from this filesystem.
 219 .\" (This is a security feature to prevent users executing set-user-ID and
 220 .\" set-group-ID programs from removable disk devices.)
 221 .TP
 222 .B MS_RDONLY
 223 Mount filesystem read-only.
 224 .TP
 225 .BR MS_REC " (since Linux 2.4.11)"
 226 Used in conjunction with
 227 .BR MS_BIND
 228 to create a recursive bind mount,
 229 and in conjunction with the propagation type flags to recursively change
 230 the propagation type of all of the mounts in a subtree.
 231 See below for further details.
 232 .TP
 233 .BR MS_RELATIME " (since Linux 2.6.20)"
 234 When a file on this filesystem is accessed,
 235 update the file's last access time (atime) only if the current value
 236 of atime is less than or equal to the file's last modification time (mtime)
 237 or last status change time (ctime).
 238 This option is useful for programs, such as
 239 .BR mutt (1),
 240 that need to know when a file has been read since it was last modified.
 241 Since Linux 2.6.30, the kernel defaults to the behavior provided
 242 by this flag (unless
 243 .BR MS_NOATIME
 244 was specified), and the
 245 .B MS_STRICTATIME
 246 flag is required to obtain traditional semantics.
 247 In addition, since Linux 2.6.30,
 248 the file's last access time is always updated if it
 249 is more than 1 day old.
 250 .\" Matthew Garrett notes in the patch that added this behavior
 251 .\" that this lets utilities such as tmpreaper (which deletes
 252 .\" files based on last access time) work correctly.
 253 .TP
 254 .BR MS_SILENT " (since Linux 2.6.17)"
 255 Suppress the display of certain
 256 .RI ( printk ())
 257 warning messages in the kernel log.
 258 This flag supersedes the misnamed and obsolete
 259 .BR MS_VERBOSE
 260 flag (available since Linux 2.4.12), which has the same meaning.
 261 .TP
 262 .BR MS_STRICTATIME " (since Linux 2.6.30)"
 263 Always update the last access time (atime) when files on this
 264 filesystem are accessed.
 265 (This was the default behavior before Linux 2.6.30.)
 266 Specifying this flag overrides the effect of setting the
 267 .BR MS_NOATIME
 268 and
 269 .BR MS_RELATIME
 270 flags.
 271 .TP
 272 .B MS_SYNCHRONOUS
 273 Make writes on this filesystem synchronous (as though
 274 the
 275 .B O_SYNC
 276 flag to
 277 .BR open (2)
 278 was specified for all file opens to this filesystem).
 279 .PP
 280 From Linux 2.4 onward, the
 281 .BR MS_NODEV ", " MS_NOEXEC ", and " MS_NOSUID
 282 flags are settable on a per-mount-point basis.
 283 From kernel 2.6.16 onward,
 284 .B MS_NOATIME
 285 and
 286 .B MS_NODIRATIME
 287 are also settable on a per-mount-point basis.
 288 The
 289 .B MS_RELATIME
 290 flag is also settable on a per-mount-point basis.
 291 Since Linux 2.6.16,
 292 .B MS_RDONLY
 293 can be set or cleared on a per-mount-point basis as well as on
 294 the underlying filesystem.
 295 The mounted filesystem will be writable only if neither the filesystem
 296 nor the mountpoint are flagged as read-only.
 297 .\"
 298 .SS Remounting an existing mount
 299 An existing mount may be remounted by specifying
 300 .B MS_REMOUNT
 301 in
 302 .IR mountflags .
 303 This allows you to change the
 304 .I mountflags
 305 and
 306 .I data
 307 of an existing mount without having to unmount and remount the filesystem.
 308 .I target
 309 should be the same value specified in the initial
 310 .BR mount ()
 311 call.
 312
 313 The
 314 .I source
 315 and
 316 .I filesystemtype
 317 arguments are ignored.
 318
 319 The
 320 .I mountflags
 321 and
 322 .I data
 323 arguments should match the values used in the original
 324 .BR mount ()
 325 call, except for those parameters that are being deliberately changed.
 326 Another exception is that
 327 .B MS_BIND
 328 has a different meaning for remount, and it should be included only if
 329 explicitly desired.
 330 .PP
 331 The following
 332 .I mountflags
 333 can be changed:
 334 .BR MS_LAZYTIME ,
 335 .BR MS_MANDLOCK ,
 336 .BR MS_NOATIME ,
 337 .BR MS_NODEV ,
 338 .BR MS_NODIRATIME ,
 339 .BR MS_NOEXEC ,
 340 .BR MS_NOSUID ,
 341 .BR MS_RELATIME ,
 342 .BR MS_RDONLY ,
 343 and
 344 .BR MS_SYNCHRONOUS .
 345 Attempts to change the setting of the
 346 .\" See the definition of MS_RMT_MASK in include/uapi/linux/fs.h
 347 .BR MS_DIRSYNC
 348 flag during a remount are silently ignored.
 349
 350 Since Linux 3.17,
 351 .\" commit ffbc6f0ead47fa5a1dc9642b0331cb75c20a640e
 352 if none of
 353 .BR MS_NOATIME ,
 354 .BR MS_NODIRATIME ,
 355 .BR MS_RELATIME ,
 356 or
 357 .BR MS_STRICTATIME
 358 is specified in
 359 .IR mountflags ,
 360 then the remount operation preserves the existing values of these flags
 361 (rather than defaulting to
 362 .BR MS_RELATIME ).
 363 .PP
 364 Since Linux 2.6.26, this flag can be used with
 365 .B MS_BIND
 366 to modify only the per-mount-point flags.
 367 .\" See https://lwn.net/Articles/281157/
 368 This is particularly useful for setting or clearing the "read-only"
 369 flag on a mount point without changing the underlying filesystem.
 370 Specifying
 371 .IR mountflags
 372 as:
 373 .PP
 374     MS_REMOUNT | MS_BIND | MS_RDONLY
 375 .PP
 376 will make access through this mountpoint read-only, without affecting
 377 other mount points.
 378 .\"
 379 .SS Creating a bind mount
 380 If
 381 .I mountflags
 382 includes
 383 .BR MS_BIND
 384 (available since Linux 2.4),
 385 .\" since 2.4.0-test9
 386 then perform a bind mount.
 387 A bind mount makes a file or a directory subtree visible at
 388 another point within the single directory hierarchy.
 389 Bind mounts may cross filesystem boundaries and span
 390 .BR chroot (2)
 391 jails.
 392
 393 The
 394 .IR filesystemtype
 395 and
 396 .IR data
 397 arguments are ignored.
 398
 399 The remaining bits in the
 400 .I mountflags
 401 argument are also ignored, with the exception of
 402 .BR MS_REC .
 403 (The bind mount has the same mount options as
 404 the underlying mount point.)
 405 However, see the discussion of remounting above,
 406 for a method of making an existing bind mount read-only.
 407
 408 By default, when a directory is bind mounted,
 409 only that directory is mounted;
 410 if there are any submounts under the directory tree,
 411 they are not bind mounted.
 412 If the
 413 .BR MS_REC
 414 flag is also specified, then a recursive bind mount operation is performed:
 415 all submounts under the
 416 .I source
 417 subtree (other than unbindable mounts)
 418 are also bind mounted at the corresponding location in the
 419 .I target
 420 subtree.
 421 .\"
 422 .SS Changing the propagation type of an existing mount
 423 If
 424 .IR mountflags
 425 includes one of
 426 .BR MS_SHARED ,
 427 .BR MS_PRIVATE ,
 428 .BR MS_SLAVE ,
 429 or
 430 .BR MS_UNBINDABLE
 431 (all available since Linux 2.6.15),
 432 then the propagation type of an existing mount is changed.
 433 If more than one of these flags is specified, an error results.
 434
 435 The only flags that can be used with changing the propagation type are
 436 .BR MS_REC
 437 and
 438 .BR MS_SILENT .
 439
 440 The
 441 .IR source ,
 442 .IR filesystemtype ,
 443 and
 444 .IR data
 445 arguments are ignored.
 446
 447 The meanings of the propagation type flags are as follows:
 448 .TP
 449 .BR MS_SHARED
 450 Make this mount point shared.
 451 Mount and unmount events immediately under this mount point will propagate
 452 to the other mount points that are members of this mount's peer group.
 453 Propagation here means that the same mount or unmount will automatically
 454 occur under all of the other mount points in the peer group.
 455 Conversely, mount and unmount events that take place under
 456 peer mount points will propagate to this mount point.
 457 .TP
 458 .BR MS_PRIVATE
 459 Make this mount point private.
 460 Mount and unmount events do not propagate into or out of this mount point.
 461 .TP
 462 .BR MS_SLAVE
 463 If this is a shared mount point that is a member of a peer group
 464 that contains other members, convert it to a slave mount.
 465 If this is a shared mount point that is a member of a peer group
 466 that contains no other members, convert it to a private mount.
 467 Otherwise, the propagation type of the mount point is left unchanged.
 468
 469 When a mount point is a slave,
 470 mount and unmount events propagate into this mount point from
 471 the (master) shared peer group of which it was formerly a member.
 472 Mount and unmount events under this mount point do not propagate to any peer.
 473
 474 A mount point can be the slave of another peer group
 475 while at the same time sharing mount and unmount events
 476 with a peer group of which it is a member.
 477 .TP
 478 .BR MS_UNBINDABLE
 479 Make this mount unbindable.
 480 This is like a private mount,
 481 and in addition this mount can't be bind mounted.
 482 When a recursive bind mount
 483 .RB ( mount ()
 484 with the
 485 .BR MS_BIND
 486 and
 487 .BR MS_REC
 488 flags) is performed on a directory subtree,
 489 any bind mounts within the subtree are automatically pruned
 490 (i.e., not replicated)
 491 when replicating that subtree to produce the target subtree.
 492 .PP
 493 By default, changing the propagation type affects only the
 494 .I target
 495 mount point.
 496 If the
 497 .B MS_REC
 498 flag is also specified in
 499 .IR mountflags ,
 500 then the propagation type of all mount points under
 501 .IR target
 502 is also changed.
 503
 504 For further details regarding mount propagation types
 505 (including the default propagation type assigned to new mounts), see
 506 .BR mount_namespaces (7).
 507 .\"
 508 .SS Moving a mount
 509 If
 510 .I mountflags
 511 contains the flag
 512 .BR MS_MOVE
 513 (available since Linux 2.4.18),
 514 then move a subtree:
 515 .I source
 516 specifies an existing mount point and
 517 .I target
 518 specifies the new location to which that mount point is to be relocated.
 519 The move is atomic: at no point is the subtree unmounted.
 520
 521 The remaining bits in the
 522 .IR mountflags
 523 argument are ignored, as are the
 524 .IR filesystemtype
 525 and
 526 .IR data
 527 arguments.
 528 .\"
 529 .SS Creating a new mount point
 530 If none of
 531 .BR MS_REMOUNT ,
 532 .BR MS_BIND ,
 533 .BR MS_MOVE ,
 534 .BR MS_SHARED ,
 535 .BR MS_PRIVATE ,
 536 .BR MS_SLAVE ,
 537 or
 538 .BR MS_UNBINDABLE
 539 is specified in
 540 .IR mountflags ,
 541 then
 542 .BR mount ()
 543 performs its default action: creating a new mount point.
 544 .IR source
 545 specifies the source for the new mount point, and
 546 .IR target
 547 specifies the directory at which to create the mount point.
 548
 549 The
 550 .I filesystemtype
 551 and
 552 .I data
 553 arguments are employed, and further bits may be specified in
 554 .IR mountflags
 555 to modify the behavior of the call.
 556 .\"
 557 .SH RETURN VALUE
 558 On success, zero is returned.
 559 On error, \-1 is returned, and
 560 .I errno
 561 is set appropriately.
 562 .SH ERRORS
 563 The error values given below result from filesystem type independent
 564 errors.
 565 Each filesystem type may have its own special errors and its
 566 own special behavior.
 567 See the Linux kernel source code for details.
 568 .TP
 569 .B EACCES
 570 A component of a path was not searchable.
 571 (See also
 572 .BR path_resolution (7).)
 573 .TP
 574 .B EACCES
 575 Mounting a read-only filesystem was attempted without giving the
 576 .B MS_RDONLY
 577 flag.
 578 .TP
 579 .B EACCES
 580 The block device
 581 .I source
 582 is located on a filesystem mounted with the
 583 .B MS_NODEV
 584 option.
 585 .\" mtk: Probably: write permission is required for MS_BIND, with
 586 .\" the error EPERM if not present; CAP_DAC_OVERRIDE is required.
 587 .TP
 588 .B EBUSY
 589 .I source
 590 is already mounted.
 591 .TP
 592 .B EBUSY
 593 .I source
 594 cannot be remounted read-only,
 595 because it still holds files open for writing.
 596 .TP
 597 .B EBUSY
 598 .I source
 599 cannot be mounted on
 600 .I target
 601 because
 602 .I target
 603 is still busy (it is the working directory of some thread,
 604 the mount point of another device, has open files, etc.).
 605 .TP
 606 .B EFAULT
 607 One of the pointer arguments points outside the user address space.
 608 .TP
 609 .B EINVAL
 610 .I source
 611 had an invalid superblock.
 612 .TP
 613 .B EINVAL
 614 A remount operation
 615 .RB ( MS_REMOUNT )
 616 was attempted, but
 617 .I source
 618 was not already mounted on
 619 .IR target .
 620 .TP
 621 .B EINVAL
 622 A move operation
 623 .RB ( MS_MOVE )
 624 was attempted, but
 625 .I source
 626 was not a mount point, or was \(aq/\(aq.
 627 .TP
 628 .B EINVAL
 629 .I mountflags
 630 includes more than one of
 631 .BR MS_SHARED ,
 632 .BR MS_PRIVATE ,
 633 .BR MS_SLAVE ,
 634 or
 635 .BR MS_UNBINDABLE .
 636 .TP
 637 .B EINVAL
 638 .I mountflags
 639 includes
 640 .BR MS_SHARED ,
 641 .BR MS_PRIVATE ,
 642 .BR MS_SLAVE ,
 643 or
 644 .BR MS_UNBINDABLE
 645 and also includes a flag other than
 646 .BR MS_REC
 647 or
 648 .BR MS_SILENT .
 649 .TP
 650 .BR EINVAL
 651 An attempt was made to bind mount an unbindable mount.
 652 .TP
 653 .B ELOOP
 654 Too many links encountered during pathname resolution.
 655 .TP
 656 .B ELOOP
 657 A move operation was attempted, and
 658 .I target
 659 is a descendant of
 660 .IR source .
 661 .TP
 662 .B EMFILE
 663 (In case no block device is required:)
 664 Table of dummy devices is full.
 665 .TP
 666 .B ENAMETOOLONG
 667 A pathname was longer than
 668 .BR MAXPATHLEN .
 669 .TP
 670 .B ENODEV
 671 .I filesystemtype
 672 not configured in the kernel.
 673 .TP
 674 .B ENOENT
 675 A pathname was empty or had a nonexistent component.
 676 .TP
 677 .B ENOMEM
 678 The kernel could not allocate a free page to copy filenames or data into.
 679 .TP
 680 .B ENOTBLK
 681 .I source
 682 is not a block device (and a device was required).
 683 .TP
 684 .B ENOTDIR
 685 .IR target ,
 686 or a prefix of
 687 .IR source ,
 688 is not a directory.
 689 .TP
 690 .B ENXIO
 691 The major number of the block device
 692 .I source
 693 is out of range.
 694 .TP
 695 .B EPERM
 696 The caller does not have the required privileges.
 697 .SH VERSIONS
 698 The definitions of
 699 .BR MS_DIRSYNC ,
 700 .BR MS_MOVE ,
 701 .BR MS_PRIVATE ,
 702 .BR MS_REC ,
 703 .BR MS_RELATIME ,
 704 .BR MS_SHARED ,
 705 .BR MS_SLAVE ,
 706 .BR MS_STRICTATIME
 707 and
 708 .BR MS_UNBINDABLE
 709 were added to glibc headers in version 2.12.
 710 .\"
 711 .SH CONFORMING TO
 712 This function is Linux-specific and should not be used in
 713 programs intended to be portable.
 714 .SH NOTES
 715 Since Linux 2.4 a single filesystem can be mounted at
 716 multiple mount points, and multiple mounts can be stacked
 717 on the same mount point.
 718 .\" Multiple mounts on same mount point: since 2.3.99pre7.
 719
 720 The
 721 .I mountflags
 722 argument may have the magic number 0xC0ED (\fBMS_MGC_VAL\fP)
 723 in the top 16 bits.
 724 (All of the other flags discussed in DESCRIPTION
 725 occupy the low order 16 bits of
 726 .IR mountflags .)
 727 Specifying
 728 .BR MS_MGC_VAL
 729 was required in kernel versions prior to 2.4,
 730 but since Linux 2.4 is no longer required and is ignored if specified.
 731
 732 The original
 733 .B MS_SYNC
 734 flag was renamed
 735 .B MS_SYNCHRONOUS
 736 in 1.1.69
 737 when a different
 738 .B MS_SYNC
 739 was added to \fI<mman.h>\fP.
 740 .LP
 741 Before Linux 2.4 an attempt to execute a set-user-ID or set-group-ID program
 742 on a filesystem mounted with
 743 .B MS_NOSUID
 744 would fail with
 745 .BR EPERM .
 746 Since Linux 2.4 the set-user-ID and set-group-ID bits are
 747 just silently ignored in this case.
 748 .\" The change is in patch-2.4.0-prerelease.
 749 .\"
 750
 751 .SS Per-process namespaces
 752 Starting with kernel 2.4.19, Linux provides
 753 per-process mount namespaces.
 754 A mount namespace is the set of filesystem mounts that
 755 are visible to a process.
 756 Mount-point namespaces can be (and usually are)
 757 shared between multiple processes,
 758 and changes to the namespace (i.e., mounts and unmounts) by one process
 759 are visible to all other processes sharing the same namespace.
 760 (The pre-2.4.19 Linux situation can be considered as one in which
 761 a single namespace was shared by every process on the system.)
 762
 763 A child process created by
 764 .BR fork (2)
 765 shares its parent's mount namespace;
 766 the mount namespace is preserved across an
 767 .BR execve (2).
 768
 769 A process can obtain a private mount namespace if:
 770 it was created using the
 771 .BR clone (2)
 772 .BR CLONE_NEWNS
 773 flag,
 774 in which case its new namespace is initialized to be a
 775 .I copy
 776 of the namespace of the process that called
 777 .BR clone (2);
 778 or it calls
 779 .BR unshare (2)
 780 with the
 781 .BR CLONE_NEWNS
 782 flag,
 783 which causes the caller's mount namespace to obtain a private copy
 784 of the namespace that it was previously sharing with other processes,
 785 so that future mounts and unmounts by the caller are invisible
 786 to other processes (except child processes that the caller
 787 subsequently creates) and vice versa.
 788
 789 The Linux-specific
 790 .I /proc/[pid]/mounts
 791 file exposes the list of mount points in the mount
 792 namespace of the process with the specified ID; see
 793 .BR proc (5)
 794 for details.
 795 .SH SEE ALSO
 796 .BR mountpoint (1),
 797 .BR umount (2),
 798 .BR mount_namespaces (7),
 799 .BR path_resolution (7),
 800 .BR findmnt (8),
 801 .BR lsblk (8),
 802 .BR mount (8),
 803 .BR umount (8)