]> git.ipfire.org Git - thirdparty/man-pages.git/blob - man2/flock.2
grantpt.3: SYNOPSIS: Explicitly show #define _XOPEN_SOURCE requirement
[thirdparty/man-pages.git] / man2 / flock.2
1 .\" Copyright 1993 Rickard E. Faith (faith@cs.unc.edu) and
2 .\" and Copyright 2002 Michael Kerrisk
3 .\"
4 .\" %%%LICENSE_START(VERBATIM)
5 .\" Permission is granted to make and distribute verbatim copies of this
6 .\" manual provided the copyright notice and this permission notice are
7 .\" preserved on all copies.
8 .\"
9 .\" Permission is granted to copy and distribute modified versions of this
10 .\" manual under the conditions for verbatim copying, provided that the
11 .\" entire resulting derived work is distributed under the terms of a
12 .\" permission notice identical to this one.
13 .\"
14 .\" Since the Linux kernel and libraries are constantly changing, this
15 .\" manual page may be incorrect or out-of-date. The author(s) assume no
16 .\" responsibility for errors or omissions, or for damages resulting from
17 .\" the use of the information contained herein. The author(s) may not
18 .\" have taken the same level of care in the production of this manual,
19 .\" which is licensed free of charge, as they might when working
20 .\" professionally.
21 .\"
22 .\" Formatted or processed versions of this manual, if unaccompanied by
23 .\" the source, must acknowledge the copyright and authors of this work.
24 .\" %%%LICENSE_END
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@gmail.com>
29 .\" Substantial rewrites and additions
30 .\" 2005-05-10 mtk, noted that lock conversions are not atomic.
31 .\"
32 .\" FIXME Maybe document LOCK_MAND, LOCK_RW, LOCK_READ, LOCK_WRITE
33 .\" which only have effect for SAMBA.
34 .\"
35 .TH FLOCK 2 2017-09-15 "Linux" "Linux Programmer's Manual"
36 .SH NAME
37 flock \- apply or remove an advisory lock on an open file
38 .SH SYNOPSIS
39 .B #include <sys/file.h>
40 .PP
41 .BI "int flock(int " fd ", int " operation );
42 .SH DESCRIPTION
43 Apply or remove an advisory lock on the open file specified by
44 .IR fd .
45 The argument
46 .I operation
47 is one of the following:
48 .RS 4
49 .TP 9
50 .B LOCK_SH
51 Place a shared lock.
52 More than one process may hold a shared lock for a given file
53 at a given time.
54 .TP
55 .B LOCK_EX
56 Place an exclusive lock.
57 Only one process may hold an exclusive lock for a given
58 file at a given time.
59 .TP
60 .B LOCK_UN
61 Remove an existing lock held by this process.
62 .RE
63 .PP
64 A call to
65 .BR flock ()
66 may block if an incompatible lock is held by another process.
67 To make a nonblocking request, include
68 .B LOCK_NB
69 (by ORing)
70 with any of the above operations.
71 .PP
72 A single file may not simultaneously have both shared and exclusive locks.
73 .PP
74 Locks created by
75 .BR flock ()
76 are associated with an open file description (see
77 .BR open (2)).
78 This means that duplicate file descriptors (created by, for example,
79 .BR fork (2)
80 or
81 .BR dup (2))
82 refer to the same lock, and this lock may be modified
83 or released using any of these file descriptors.
84 Furthermore, the lock is released either by an explicit
85 .B LOCK_UN
86 operation on any of these duplicate file descriptors, or when all
87 such file descriptors have been closed.
88 .PP
89 If a process uses
90 .BR open (2)
91 (or similar) to obtain more than one file descriptor for the same file,
92 these file descriptors are treated independently by
93 .BR flock ().
94 An attempt to lock the file using one of these file descriptors
95 may be denied by a lock that the calling process has
96 already placed via another file descriptor.
97 .PP
98 A process may hold only one type of lock (shared or exclusive)
99 on a file.
100 Subsequent
101 .BR flock ()
102 calls on an already locked file will convert an existing lock to the new
103 lock mode.
104 .PP
105 Locks created by
106 .BR flock ()
107 are preserved across an
108 .BR execve (2).
109 .PP
110 A shared or exclusive lock can be placed on a file regardless of the
111 mode in which the file was opened.
112 .SH RETURN VALUE
113 On success, zero is returned.
114 On error, \-1 is returned, and
115 .I errno
116 is set appropriately.
117 .SH ERRORS
118 .TP
119 .B EBADF
120 .I fd
121 is not an open file descriptor.
122 .TP
123 .B EINTR
124 While waiting to acquire a lock, the call was interrupted by
125 delivery of a signal caught by a handler; see
126 .BR signal (7).
127 .TP
128 .B EINVAL
129 .I operation
130 is invalid.
131 .TP
132 .B ENOLCK
133 The kernel ran out of memory for allocating lock records.
134 .TP
135 .B EWOULDBLOCK
136 The file is locked and the
137 .B LOCK_NB
138 flag was selected.
139 .SH CONFORMING TO
140 4.4BSD (the
141 .BR flock ()
142 call first appeared in 4.2BSD).
143 A version of
144 .BR flock (),
145 possibly implemented in terms of
146 .BR fcntl (2),
147 appears on most UNIX systems.
148 .SH NOTES
149 Since kernel 2.0,
150 .BR flock ()
151 is implemented as a system call in its own right rather
152 than being emulated in the GNU C library as a call to
153 .BR fcntl (2).
154 With this implementation,
155 there is no interaction between the types of lock
156 placed by
157 .BR flock ()
158 and
159 .BR fcntl (2),
160 and
161 .BR flock ()
162 does not detect deadlock.
163 (Note, however, that on some systems, such as the modern BSDs,
164 .\" E.g., according to the flock(2) man page, FreeBSD since at least 5.3
165 .BR flock ()
166 and
167 .BR fcntl (2)
168 locks
169 .I do
170 interact with one another.)
171 .PP
172 .BR flock ()
173 places advisory locks only; given suitable permissions on a file,
174 a process is free to ignore the use of
175 .BR flock ()
176 and perform I/O on the file.
177 .PP
178 .BR flock ()
179 and
180 .BR fcntl (2)
181 locks have different semantics with respect to forked processes and
182 .BR dup (2).
183 On systems that implement
184 .BR flock ()
185 using
186 .BR fcntl (2),
187 the semantics of
188 .BR flock ()
189 will be different from those described in this manual page.
190 .PP
191 Converting a lock
192 (shared to exclusive, or vice versa) is not guaranteed to be atomic:
193 the existing lock is first removed, and then a new lock is established.
194 Between these two steps,
195 a pending lock request by another process may be granted,
196 with the result that the conversion either blocks, or fails if
197 .B LOCK_NB
198 was specified.
199 (This is the original BSD behavior,
200 and occurs on many other implementations.)
201 .\" Kernel 2.5.21 changed things a little: during lock conversion
202 .\" it is now the highest priority process that will get the lock -- mtk
203 .SS NFS details
204 In Linux kernels up to 2.6.11,
205 .BR flock ()
206 does not lock files over NFS
207 (i.e., the scope of locks was limited to the local system).
208 Instead, one could use
209 .BR fcntl (2)
210 byte-range locking, which does work over NFS,
211 given a sufficiently recent version of
212 Linux and a server which supports locking.
213 .PP
214 Since Linux 2.6.12, NFS clients support
215 .BR flock ()
216 locks by emulating them as
217 .BR fcntl (2)
218 byte-range locks on the entire file.
219 This means that
220 .BR fcntl (2)
221 and
222 .BR flock ()
223 locks
224 .I do
225 interact with one another over NFS.
226 It also means that in order to place an exclusive lock,
227 the file must be opened for writing.
228 .PP
229 Since Linux 2.6.37,
230 .\" commit 5eebde23223aeb0ad2d9e3be6590ff8bbfab0fc2
231 the kernel supports a compatibility mode that allows
232 .BR flock ()
233 locks (and also
234 .BR fcntl (2)
235 byte region locks) to be treated as local;
236 see the discussion of the
237 .I "local_lock"
238 option in
239 .BR nfs (5).
240 .SH SEE ALSO
241 .BR flock (1),
242 .BR close (2),
243 .BR dup (2),
244 .BR execve (2),
245 .BR fcntl (2),
246 .BR fork (2),
247 .BR open (2),
248 .BR lockf (3),
249 .BR lslocks (8)
250 .PP
251 .I Documentation/filesystems/locks.txt
252 in the Linux kernel source tree
253 .RI ( Documentation/locks.txt
254 in older kernels)