1LOCKF(3P) POSIX Programmer's Manual LOCKF(3P)
2
6 This manual page is part of the POSIX Programmer's Manual. The Linux
7 implementation of this interface may differ (consult the corresponding
8 Linux manual page for details of Linux behavior), or the interface may
9 not be implemented on Linux.
10
12 lockf - record locking on files
13
15 #include <unistd.h>
16 int lockf(int fildes, int function, off_t size);
18
21 The lockf() function shall lock sections of a file with advisory-mode
22 locks. Calls to lockf() from other threads which attempt to lock the
23 locked file section shall either return an error value or block until
24 the section becomes unlocked. All the locks for a process are removed
25 when the process terminates. Record locking with lockf() shall be sup‐
26 ported for regular files and may be supported for other files.
27 The fildes argument is an open file descriptor. To establish a lock
29 with this function, the file descriptor shall be opened with write-only
30 permission (O_WRONLY) or with read/write permission (O_RDWR).
31 The function argument is a control value which specifies the action to
33 be taken. The permissible values for function are defined in <unistd.h>
34 as follows:
35 Function Description
37 F_ULOCK Unlock locked sections.
38 F_LOCK Lock a section for exclusive use.
39 F_TLOCK Test and lock a section for exclusive use.
40 F_TEST Test a section for locks by other processes.
41 F_TEST shall detect if a lock by another process is present on the
43 specified section.
44 F_LOCK and F_TLOCK shall both lock a section of a file if the section
46 is available.
47 F_ULOCK shall remove locks from a section of the file.
49 The size argument is the number of contiguous bytes to be locked or
51 unlocked. The section to be locked or unlocked starts at the current
52 offset in the file and extends forward for a positive size or backward
53 for a negative size (the preceding bytes up to but not including the
54 current offset). If size is 0, the section from the current offset
55 through the largest possible file offset shall be locked (that is, from
56 the current offset through the present or any future end-of-file). An
57 area need not be allocated to the file to be locked because locks may
58 exist past the end-of-file.
59 The sections locked with F_LOCK or F_TLOCK may, in whole or in part,
61 contain or be contained by a previously locked section for the same
62 process. When this occurs, or if adjacent locked sections would occur,
63 the sections shall be combined into a single locked section. If the
64 request would cause the number of locks to exceed a system-imposed
65 limit, the request shall fail.
66 F_LOCK and F_TLOCK requests differ only by the action taken if the sec‐
68 tion is not available. F_LOCK shall block the calling thread until the
69 section is available. F_TLOCK shall cause the function to fail if the
70 section is already locked by another process.
71 File locks shall be released on first close by the locking process of
73 any file descriptor for the file.
74 F_ULOCK requests may release (wholly or in part) one or more locked
76 sections controlled by the process. Locked sections shall be unlocked
77 starting at the current file offset through size bytes or to the end-
78 of-file if size is (off_t)0. When all of a locked section is not
79 released (that is, when the beginning or end of the area to be unlocked
80 falls within a locked section), the remaining portions of that section
81 shall remain locked by the process. Releasing the center portion of a
82 locked section shall cause the remaining locked beginning and end por‐
83 tions to become two separate locked sections. If the request would
84 cause the number of locks in the system to exceed a system-imposed
85 limit, the request shall fail.
86 A potential for deadlock occurs if the threads of a process controlling
88 a locked section are blocked by accessing another process' locked sec‐
89 tion. If the system detects that deadlock would occur, lockf() shall
90 fail with an [EDEADLK] error.
91 The interaction between fcntl() and lockf() locks is unspecified.
93 Blocking on a section shall be interrupted by any signal.
95 An F_ULOCK request in which size is non-zero and the offset of the last
97 byte of the requested section is the maximum value for an object of
98 type off_t, when the process has an existing lock in which size is 0
99 and which includes the last byte of the requested section, shall be
100 treated as a request to unlock from the start of the requested section
101 with a size equal to 0. Otherwise, an F_ULOCK request shall attempt to
102 unlock only the requested section.
103 Attempting to lock a section of a file that is associated with a
105 buffered stream produces unspecified results.
106
108 Upon successful completion, lockf() shall return 0. Otherwise, it shall
109 return -1, set errno to indicate an error, and existing locks shall not
110 be changed.
111
113 The lockf() function shall fail if:
114 EBADF The fildes argument is not a valid open file descriptor; or
116 function is F_LOCK or F_TLOCK and fildes is not a valid file
117 descriptor open for writing.
118 EACCES or EAGAIN
120 The function argument is F_TLOCK or F_TEST and the section is
122 already locked by another process.
123 EDEADLK
125 The function argument is F_LOCK and a deadlock is detected.
126 EINTR A signal was caught during execution of the function.
128 EINVAL The function argument is not one of F_LOCK, F_TLOCK, F_TEST, or
130 F_ULOCK; or size plus the current file offset is less than 0.
131 EOVERFLOW
133 The offset of the first, or if size is not 0 then the last, byte
134 in the requested section cannot be represented correctly in an
135 object of type off_t.
136 The lockf() function may fail if:
139 EAGAIN The function argument is F_LOCK or F_TLOCK and the file is
141 mapped with mmap().
142 EDEADLK or ENOLCK
144 The function argument is F_LOCK, F_TLOCK, or F_ULOCK, and the
146 request would cause the number of locks to exceed a system-
147 imposed limit.
148 EOPNOTSUPP or EINVAL
150 The implementation does not support the locking of files of the
152 type indicated by the fildes argument.
153 The following sections are informative.
156
158 Locking a Portion of a File
159 In the following example, a file named /home/cnd/mod1 is being modi‐
160 fied. Other processes that use locking are prevented from changing it
161 during this process. Only the first 10000 bytes are locked, and the
162 lock call fails if another process has any part of this area locked
163 already.
164 #include <fcntl.h>
167 #include <unistd.h>
168 int fildes;
171 int status;
172 ...
173 fildes = open("/home/cnd/mod1", O_RDWR);
174 status = lockf(fildes, F_TLOCK, (off_t)10000);
175
177 Record-locking should not be used in combination with the fopen(),
178 fread(), fwrite(), and other stdio functions. Instead, the more primi‐
179 tive, non-buffered functions (such as open()) should be used. Unex‐
180 pected results may occur in processes that do buffering in the user
181 address space. The process may later read/write data which is/was
182 locked. The stdio functions are the most common source of unexpected
183 buffering.
184 The alarm() function may be used to provide a timeout facility in
186 applications requiring it.
187
189 None.
190
192 None.
193
195 alarm(), chmod(), close(), creat(), fcntl(), fopen(), mmap(), open(),
196 read(), write(), the Base Definitions volume of IEEE Std 1003.1-2001,
197 <unistd.h>
198
200 Portions of this text are reprinted and reproduced in electronic form
201 from IEEE Std 1003.1, 2003 Edition, Standard for Information Technology
202 -- Portable Operating System Interface (POSIX), The Open Group Base
203 Specifications Issue 6, Copyright (C) 2001-2003 by the Institute of
204 Electrical and Electronics Engineers, Inc and The Open Group. In the
205 event of any discrepancy between this version and the original IEEE and
206 The Open Group Standard, the original IEEE and The Open Group Standard
207 is the referee document. The original Standard can be obtained online
208 at http://www.opengroup.org/unix/online.html .
209IEEE/The Open Group 2003 LOCKF(3P)