man2/flock.2

   1 .\" Hey Emacs! This file is -*- nroff -*- source.
   2 .\"
   3 .\" Copyright 1993 Rickard E. Faith (faith@cs.unc.edu) and
   4 .\" and Copyright 2002 Michael Kerrisk
   5 .\"
   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  16 .\" manual page may be incorrect or out-of-date.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 .\"
  26 .\" Modified Fri Jan 31 16:26:07 1997 by Eric S. Raymond <esr@thyrsus.com>
  27 .\" Modified Fri Dec 11 17:57:27 1998 by Jamie Lokier <jamie@imbolc.ucc.ie>
  28 .\" Modified 24 Apr 2002 by Michael Kerrisk <mtk-manpages@gmx.net>
  29 .\"     Substantial rewrites and additions
  30 .\" 2005-05-10 mtk, noted that lock conversions are not atomic.
  31 .\"
  32 .TH FLOCK 2 2002-04-24 "Linux" "Linux Programmer's Manual"
  33 .SH NAME
  34 flock \- apply or remove an advisory lock on an open file
  35 .SH SYNOPSIS
  36 .B #include <sys/file.h>
  37 .sp
  38 .BI "int flock(int " fd ", int " operation );
  39 .SH DESCRIPTION
  40 Apply or remove an advisory lock on the open file specified by
  41 .IR fd .
  42 The parameter
  43 .I operation
  44 is one of the following:
  45 .RS
  46 .TP 1.0i
  47 .B LOCK_SH
  48 Place a shared lock.
  49 More than one process may hold a shared lock for a given file
  50 at a given time.
  51 .TP
  52 .B LOCK_EX
  53 Place an exclusive lock.
  54 Only one process may hold an exclusive lock for a given
  55 file at a given time.
  56 .TP
  57 .B LOCK_UN
  58 Remove an existing lock held by this process.
  59 .sp
  60 .RE
  61 A call to
  62 .BR flock ()
  63 may block if an incompatible lock is held by another process.
  64 To make a non-blocking request, include
  65 .B LOCK_NB
  66 (by
  67 .IR OR ing)
  68 with any of the above operations.
  69
  70 A single file may not simultaneously have both shared and exclusive locks.
  71
  72 Locks created by
  73 .BR flock ()
  74 are associated with an open file table entry.
  75 This means that duplicate file descriptors (created by, for example,
  76 .BR fork (2)
  77 or
  78 .BR dup (2))
  79 refer to the same lock, and this lock may be modified
  80 or released using any of these descriptors.
  81 Furthermore, the lock is released either by an explicit
  82 .B LOCK_UN
  83 operation on any of these duplicate descriptors, or when all
  84 such descriptors have been closed.
  85
  86 If a process uses
  87 .BR open (2)
  88 (or similar) to obtain more than one descriptor for the same file,
  89 these descriptors are treated independently by
  90 .BR flock ().
  91 An attempt to lock the file using one of these file descriptors
  92 may be denied by a lock that the calling process has
  93 already placed via another descriptor.
  94
  95 A process may only hold one type of lock (shared or exclusive)
  96 on a file.
  97 Subsequent
  98 .BR flock ()
  99 calls on an already locked file will convert an existing lock to the new
 100 lock mode.
 101
 102 Locks created by
 103 .BR flock ()
 104 are preserved across an
 105 .BR execve (2).
 106
 107 A shared or exclusive lock can be placed on a file regardless of the
 108 mode in which the file was opened.
 109 .SH "RETURN VALUE"
 110 On success, zero is returned.
 111 On error, \-1 is returned, and
 112 .I errno
 113 is set appropriately.
 114 .SH ERRORS
 115 .TP
 116 .B EBADF
 117 .I fd
 118 is not a not an open file descriptor.
 119 .TP
 120 .B EINTR
 121 While waiting to acquire a lock, the call was interrupted by
 122 delivery of a signal caught by a handler.
 123 .TP
 124 .B EINVAL
 125 .I operation
 126 is invalid.
 127 .TP
 128 .B ENOLCK
 129 The kernel ran out of memory for allocating lock records.
 130 .TP
 131 .B EWOULDBLOCK
 132 The file is locked and the
 133 .B LOCK_NB
 134 flag was selected.
 135 .SH "CONFORMING TO"
 136 4.4BSD (the
 137 .BR flock (2)
 138 call first appeared in 4.2BSD).
 139 A version of
 140 .BR flock (2),
 141 possibly implemented in terms of
 142 .BR fcntl (2),
 143 appears on most Unix systems.
 144 .SH NOTES
 145 .BR flock (2)
 146 does not lock files over NFS.
 147 Use
 148 .BR fcntl (2)
 149 instead: that does work over NFS, given a sufficiently recent version of
 150 Linux and a server which supports locking.
 151 .PP
 152 Since kernel 2.0,
 153 .BR flock (2)
 154 is implemented as a system call in its own right rather
 155 than being emulated in the GNU C library as a call to
 156 .BR fcntl (2).
 157 This yields true BSD semantics:
 158 there is no interaction between the types of lock
 159 placed by
 160 .BR flock (2)
 161 and
 162 .BR fcntl (2),
 163 and
 164 .BR flock (2)
 165 does not detect deadlock.
 166 .PP
 167 .BR flock (2)
 168 places advisory locks only; given suitable permissions on a file,
 169 a process is free to ignore the use of
 170 .BR flock (2)
 171 and perform I/O on the file.
 172 .PP
 173 .BR flock (2)
 174 and
 175 .BR fcntl (2)
 176 locks have different semantics with respect to forked processes and
 177 .BR dup (2).
 178 On systems that implement
 179 .BR flock ()
 180 using
 181 .BR fcntl (2),
 182 the semantics of
 183 .BR flock ()
 184 will be different from those described in this manual page.
 185 .PP
 186 Converting a lock
 187 (shared to exclusive, or vice versa) is not guaranteed to be atomic:
 188 the existing lock is first removed, and then a new lock is established.
 189 Between these two steps,
 190 a pending lock request by another process may be granted,
 191 with the result that the conversion either blocks, or fails if
 192 .B LOCK_NB
 193 was specified.
 194 (This is the original BSD behavior,
 195 and occurs on many other implementations.)
 196 .\" Kernel 2.5.21 changed things a little: during lock conversion
 197 .\" it is now the highest priority process that will get the lock -- mtk
 198 .SH "SEE ALSO"
 199 .BR close (2),
 200 .BR dup (2),
 201 .BR execve (2),
 202 .BR fcntl (2),
 203 .BR fork (2),
 204 .BR open (2),
 205 .BR lockf (3)
 206
 207 See also
 208 .I Documentation/locks.txt
 209 and
 210 .I Documentation/mandatory.txt
 211 in the kernel source.