1CLOSE(3P) POSIX Programmer's Manual CLOSE(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 close — close a file descriptor
13
15 #include <unistd.h>
16 int close(int fildes);
18
20 The close() function shall deallocate the file descriptor indicated by
21 fildes. To deallocate means to make the file descriptor available for
22 return by subsequent calls to open() or other functions that allocate
23 file descriptors. All outstanding record locks owned by the process on
24 the file associated with the file descriptor shall be removed (that is,
25 unlocked).
26 If close() is interrupted by a signal that is to be caught, it shall
28 return -1 with errno set to [EINTR] and the state of fildes is unspeci‐
29 fied. If an I/O error occurred while reading from or writing to the
30 file system during close(), it may return -1 with errno set to [EIO];
31 if this error is returned, the state of fildes is unspecified.
32 When all file descriptors associated with a pipe or FIFO special file
34 are closed, any data remaining in the pipe or FIFO shall be discarded.
35 When all file descriptors associated with an open file description have
37 been closed, the open file description shall be freed.
38 If the link count of the file is 0, when all file descriptors associ‐
40 ated with the file are closed, the space occupied by the file shall be
41 freed and the file shall no longer be accessible.
42 If a STREAMS-based fildes is closed and the calling process was previ‐
44 ously registered to receive a SIGPOLL signal for events associated with
45 that STREAM, the calling process shall be unregistered for events asso‐
46 ciated with the STREAM. The last close() for a STREAM shall cause the
47 STREAM associated with fildes to be dismantled. If O_NONBLOCK is not
48 set and there have been no signals posted for the STREAM, and if there
49 is data on the module's write queue, close() shall wait for an unspeci‐
50 fied time (for each module and driver) for any output to drain before
51 dismantling the STREAM. The time delay can be changed via an I_SET‐
52 CLTIME ioctl() request. If the O_NONBLOCK flag is set, or if there are
53 any pending signals, close() shall not wait for output to drain, and
54 shall dismantle the STREAM immediately.
55 If the implementation supports STREAMS-based pipes, and fildes is asso‐
57 ciated with one end of a pipe, the last close() shall cause a hangup to
58 occur on the other end of the pipe. In addition, if the other end of
59 the pipe has been named by fattach(), then the last close() shall force
60 the named end to be detached by fdetach(). If the named end has no
61 open file descriptors associated with it and gets detached, the STREAM
62 associated with that end shall also be dismantled.
63 If fildes refers to the master side of a pseudo-terminal, and this is
65 the last close, a SIGHUP signal shall be sent to the controlling
66 process, if any, for which the slave side of the pseudo-terminal is the
67 controlling terminal. It is unspecified whether closing the master side
68 of the pseudo-terminal flushes all queued input and output.
69 If fildes refers to the slave side of a STREAMS-based pseudo-terminal,
71 a zero-length message may be sent to the master.
72 When there is an outstanding cancelable asynchronous I/O operation
74 against fildes when close() is called, that I/O operation may be can‐
75 celed. An I/O operation that is not canceled completes as if the
76 close() operation had not yet occurred. All operations that are not
77 canceled shall complete as if the close() blocked until the operations
78 completed. The close() operation itself need not block awaiting such
79 I/O completion. Whether any I/O operation is canceled, and which I/O
80 operation may be canceled upon close(), is implementation-defined.
81 If a memory mapped file or a shared memory object remains referenced at
83 the last close (that is, a process has it mapped), then the entire con‐
84 tents of the memory object shall persist until the memory object
85 becomes unreferenced. If this is the last close of a memory mapped
86 file or a shared memory object and the close results in the memory
87 object becoming unreferenced, and the memory object has been unlinked,
88 then the memory object shall be removed.
89 If fildes refers to a socket, close() shall cause the socket to be
91 destroyed. If the socket is in connection-mode, and the SO_LINGER
92 option is set for the socket with non-zero linger time, and the socket
93 has untransmitted data, then close() shall block for up to the current
94 linger interval until all data is transmitted.
95
97 Upon successful completion, 0 shall be returned; otherwise, -1 shall be
98 returned and errno set to indicate the error.
99
101 The close() function shall fail if:
102 EBADF The fildes argument is not a open file descriptor.
104 EINTR The close() function was interrupted by a signal.
106 The close() function may fail if:
108 EIO An I/O error occurred while reading from or writing to the file
110 system.
111 The following sections are informative.
113
115 Reassigning a File Descriptor
116 The following example closes the file descriptor associated with stan‐
117 dard output for the current process, re-assigns standard output to a
118 new file descriptor, and closes the original file descriptor to clean
119 up. This example assumes that the file descriptor 0 (which is the
120 descriptor for standard input) is not closed.
121 #include <unistd.h>
124 ...
125 int pfd;
126 ...
127 close(1);
128 dup(pfd);
129 close(pfd);
130 ...
131 Incidentally, this is exactly what could be achieved using:
133 dup2(pfd, 1);
136 close(pfd);
137 Closing a File Descriptor
139 In the following example, close() is used to close a file descriptor
140 after an unsuccessful attempt is made to associate that file descriptor
141 with a stream.
142 #include <stdio.h>
145 #include <unistd.h>
146 #include <stdlib.h>
147 #define LOCKFILE "/etc/ptmp"
149 ...
150 int pfd;
151 FILE *fpfd;
152 ...
153 if ((fpfd = fdopen (pfd, "w")) == NULL) {
154 close(pfd);
155 unlink(LOCKFILE);
156 exit(1);
157 }
158 ...
159
161 An application that had used the stdio routine fopen() to open a file
162 should use the corresponding fclose() routine rather than close().
163 Once a file is closed, the file descriptor no longer exists, since the
164 integer corresponding to it no longer refers to a file.
165 Implementations may use file descriptors that must be inherited into
167 child processes for the child process to remain conforming, such as for
168 message catalog or tracing purposes. Therefore, an application that
169 calls close() on an arbitrary integer risks non-conforming behavior,
170 and close() can only portably be used on file descriptor values that
171 the application has obtained through explicit actions, as well as the
172 three file descriptors corresponding to the standard file streams. In
173 multi-threaded parent applications, the practice of calling close() in
174 a loop after fork() and before an exec call in order to avoid a race
175 condition of leaking an unintended file descriptor into a child
176 process, is therefore unsafe, and the race should instead be combatted
177 by opening all file descriptors with the FD_CLOEXEC bit set unless the
178 file descriptor is intended to be inherited across exec.
179 Usage of close() on file descriptors STDIN_FILENO, STDOUT_FILENO, or
181 STDERR_FILENO should immediately be followed by an operation to reopen
182 these file descriptors. Unexpected behavior will result if any of these
183 file descriptors is left in a closed state (for example, an [EBADF]
184 error from perror()) or if an unrelated open() or similar call later in
185 the application accidentally allocates a file to one of these well-
186 known file descriptors. Furthermore, a close() followed by a reopen
187 operation (e.g., open(), dup(), etc.) is not atomic; dup2() should be
188 used to change standard file descriptors.
189
191 The use of interruptible device close routines should be discouraged to
192 avoid problems with the implicit closes of file descriptors by exec and
193 exit(). This volume of POSIX.1‐2017 only intends to permit such behav‐
194 ior by specifying the [EINTR] error condition.
195 Note that the requirement for close() on a socket to block for up to
197 the current linger interval is not conditional on the O_NONBLOCK set‐
198 ting.
199 The standard developers rejected a proposal to add closefrom() to the
201 standard. Because the standard permits implementations to use inherited
202 file descriptors as a means of providing a conforming environment for
203 the child process, it is not possible to standardize an interface that
204 closes arbitrary file descriptors above a certain value while still
205 guaranteeing a conforming environment.
206
208 None.
209
211 Section 2.6, STREAMS, dup(), exec, exit(), fattach(), fclose(), fde‐
212 tach(), fopen(), fork(), ioctl(), open(), perror(), unlink()
213 The Base Definitions volume of POSIX.1‐2017, <unistd.h>
215
217 Portions of this text are reprinted and reproduced in electronic form
218 from IEEE Std 1003.1-2017, Standard for Information Technology -- Por‐
219 table Operating System Interface (POSIX), The Open Group Base Specifi‐
220 cations Issue 7, 2018 Edition, Copyright (C) 2018 by the Institute of
221 Electrical and Electronics Engineers, Inc and The Open Group. In the
222 event of any discrepancy between this version and the original IEEE and
223 The Open Group Standard, the original IEEE and The Open Group Standard
224 is the referee document. The original Standard can be obtained online
225 at http://www.opengroup.org/unix/online.html .
226 Any typographical or formatting errors that appear in this page are
228 most likely to have been introduced during the conversion of the source
229 files to man page format. To report such errors, see https://www.ker‐
230 nel.org/doc/man-pages/reporting_bugs.html .
231IEEE/The Open Group 2017 CLOSE(3P)