1OPEN(3P) POSIX Programmer's Manual OPEN(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 open - open a file
13
15 #include <sys/stat.h>
16 #include <fcntl.h>
17 int open(const char *path, int oflag, ... );
19
22 The open() function shall establish the connection between a file and a
23 file descriptor. It shall create an open file description that refers
24 to a file and a file descriptor that refers to that open file descrip‐
25 tion. The file descriptor is used by other I/O functions to refer to
26 that file. The path argument points to a pathname naming the file.
27 The open() function shall return a file descriptor for the named file
29 that is the lowest file descriptor not currently open for that process.
30 The open file description is new, and therefore the file descriptor
31 shall not share it with any other process in the system. The FD_CLOEXEC
32 file descriptor flag associated with the new file descriptor shall be
33 cleared.
34 The file offset used to mark the current position within the file shall
36 be set to the beginning of the file.
37 The file status flags and file access modes of the open file descrip‐
39 tion shall be set according to the value of oflag.
40 Values for oflag are constructed by a bitwise-inclusive OR of flags
42 from the following list, defined in <fcntl.h>. Applications shall spec‐
43 ify exactly one of the first three values (file access modes) below in
44 the value of oflag:
45 O_RDONLY
47 Open for reading only.
48 O_WRONLY
50 Open for writing only.
51 O_RDWR Open for reading and writing. The result is undefined if this
53 flag is applied to a FIFO.
54 Any combination of the following may be used:
57 O_APPEND
59 If set, the file offset shall be set to the end of the file
60 prior to each write.
61 O_CREAT
63 If the file exists, this flag has no effect except as noted
64 under O_EXCL below. Otherwise, the file shall be created; the
65 user ID of the file shall be set to the effective user ID of the
66 process; the group ID of the file shall be set to the group ID
67 of the file's parent directory or to the effective group ID of
68 the process; and the access permission bits (see <sys/stat.h>)
69 of the file mode shall be set to the value of the third argument
70 taken as type mode_t modified as follows: a bitwise AND is per‐
71 formed on the file-mode bits and the corresponding bits in the
72 complement of the process' file mode creation mask. Thus, all
73 bits in the file mode whose corresponding bit in the file mode
74 creation mask is set are cleared. When bits other than the file
75 permission bits are set, the effect is unspecified. The third
76 argument does not affect whether the file is open for reading,
77 writing, or for both. Implementations shall provide a way to
78 initialize the file's group ID to the group ID of the parent
79 directory. Implementations may, but need not, provide an imple‐
80 mentation-defined way to initialize the file's group ID to the
81 effective group ID of the calling process.
82 O_DSYNC
84 Write I/O operations on the file descriptor shall complete as
85 defined by synchronized I/O data integrity completion.
86 O_EXCL If O_CREAT and O_EXCL are set, open() shall fail if the file
88 exists. The check for the existence of the file and the creation
89 of the file if it does not exist shall be atomic with respect to
90 other threads executing open() naming the same filename in the
91 same directory with O_EXCL and O_CREAT set. If O_EXCL and
92 O_CREAT are set, and path names a symbolic link, open() shall
93 fail and set errno to [EEXIST], regardless of the contents of
94 the symbolic link. If O_EXCL is set and O_CREAT is not set, the
95 result is undefined.
96 O_NOCTTY
98 If set and path identifies a terminal device, open() shall not
99 cause the terminal device to become the controlling terminal for
100 the process.
101 O_NONBLOCK
103 When opening a FIFO with O_RDONLY or O_WRONLY set:
104 * If O_NONBLOCK is set, an open() for reading-only shall return
106 without delay. An open() for writing-only shall return an
107 error if no process currently has the file open for reading.
108 * If O_NONBLOCK is clear, an open() for reading-only shall
110 block the calling thread until a thread opens the file for
111 writing. An open() for writing-only shall block the calling
112 thread until a thread opens the file for reading.
113 When opening a block special or character special file that supports
115 non-blocking opens:
116 * If O_NONBLOCK is set, the open() function shall return with‐
118 out blocking for the device to be ready or available. Subse‐
119 quent behavior of the device is device-specific.
120 * If O_NONBLOCK is clear, the open() function shall block the
122 calling thread until the device is ready or available before
123 returning.
124 Otherwise, the behavior of O_NONBLOCK is unspecified.
126 O_RSYNC
128 Read I/O operations on the file descriptor shall complete at the
129 same level of integrity as specified by the O_DSYNC and O_SYNC
130 flags. If both O_DSYNC and O_RSYNC are set in oflag, all I/O
131 operations on the file descriptor shall complete as defined by
132 synchronized I/O data integrity completion. If both O_SYNC and
133 O_RSYNC are set in flags, all I/O operations on the file
134 descriptor shall complete as defined by synchronized I/O file
135 integrity completion.
136 O_SYNC Write I/O operations on the file descriptor shall complete as
138 defined by synchronized I/O file integrity completion.
139 O_TRUNC
141 If the file exists and is a regular file, and the file is suc‐
142 cessfully opened O_RDWR or O_WRONLY, its length shall be trun‐
143 cated to 0, and the mode and owner shall be unchanged. It shall
144 have no effect on FIFO special files or terminal device files.
145 Its effect on other file types is implementation-defined. The
146 result of using O_TRUNC with O_RDONLY is undefined.
147 If O_CREAT is set and the file did not previously exist, upon success‐
150 ful completion, open() shall mark for update the st_atime, st_ctime,
151 and st_mtime fields of the file and the st_ctime and st_mtime fields of
152 the parent directory.
153 If O_TRUNC is set and the file did previously exist, upon successful
155 completion, open() shall mark for update the st_ctime and st_mtime
156 fields of the file.
157 If both the O_SYNC and O_DSYNC flags are set, the effect is as if only
159 the O_SYNC flag was set.
160 If path refers to a STREAMS file, oflag may be constructed from O_NON‐
162 BLOCK OR'ed with either O_RDONLY, O_WRONLY, or O_RDWR. Other flag val‐
163 ues are not applicable to STREAMS devices and shall have no effect on
164 them. The value O_NONBLOCK affects the operation of STREAMS drivers and
165 certain functions applied to file descriptors associated with STREAMS
166 files. For STREAMS drivers, the implementation of O_NONBLOCK is device-
167 specific.
168 If path names the master side of a pseudo-terminal device, then it is
170 unspecified whether open() locks the slave side so that it cannot be
171 opened. Conforming applications shall call unlockpt() before opening
172 the slave side.
173 The largest value that can be represented correctly in an object of
175 type off_t shall be established as the offset maximum in the open file
176 description.
177
179 Upon successful completion, the function shall open the file and return
180 a non-negative integer representing the lowest numbered unused file
181 descriptor. Otherwise, -1 shall be returned and errno set to indicate
182 the error. No files shall be created or modified if the function
183 returns -1.
184
186 The open() function shall fail if:
187 EACCES Search permission is denied on a component of the path prefix,
189 or the file exists and the permissions specified by oflag are
190 denied, or the file does not exist and write permission is
191 denied for the parent directory of the file to be created, or
192 O_TRUNC is specified and write permission is denied.
193 EEXIST O_CREAT and O_EXCL are set, and the named file exists.
195 EINTR A signal was caught during open().
197 EINVAL The implementation does not support synchronized I/O for this
199 file.
200 EIO The path argument names a STREAMS file and a hangup or error
202 occurred during the open().
203 EISDIR The named file is a directory and oflag includes O_WRONLY or
205 O_RDWR.
206 ELOOP A loop exists in symbolic links encountered during resolution of
208 the path argument.
209 EMFILE {OPEN_MAX} file descriptors are currently open in the calling
211 process.
212 ENAMETOOLONG
214 The length of the path argument exceeds {PATH_MAX} or a pathname
215 component is longer than {NAME_MAX}.
216 ENFILE The maximum allowable number of files is currently open in the
218 system.
219 ENOENT O_CREAT is not set and the named file does not exist; or O_CREAT
221 is set and either the path prefix does not exist or the path
222 argument points to an empty string.
223 ENOSR The path argument names a STREAMS-based file and the system is
225 unable to allocate a STREAM.
226 ENOSPC The directory or file system that would contain the new file
228 cannot be expanded, the file does not exist, and O_CREAT is
229 specified.
230 ENOTDIR
232 A component of the path prefix is not a directory.
233 ENXIO O_NONBLOCK is set, the named file is a FIFO, O_WRONLY is set,
235 and no process has the file open for reading.
236 ENXIO The named file is a character special or block special file, and
238 the device associated with this special file does not exist.
239 EOVERFLOW
241 The named file is a regular file and the size of the file cannot
242 be represented correctly in an object of type off_t.
243 EROFS The named file resides on a read-only file system and either
245 O_WRONLY, O_RDWR, O_CREAT (if the file does not exist), or
246 O_TRUNC is set in the oflag argument.
247 The open() function may fail if:
250 EAGAIN The path argument names the slave side of a pseudo-terminal
252 device that is locked.
253 EINVAL The value of the oflag argument is not valid.
255 ELOOP More than {SYMLOOP_MAX} symbolic links were encountered during
257 resolution of the path argument.
258 ENAMETOOLONG
260 As a result of encountering a symbolic link in resolution of the
261 path argument, the length of the substituted pathname string
262 exceeded {PATH_MAX}.
263 ENOMEM The path argument names a STREAMS file and the system is unable
265 to allocate resources.
266 ETXTBSY
268 The file is a pure procedure (shared text) file that is being
269 executed and oflag is O_WRONLY or O_RDWR.
270 The following sections are informative.
273
275 Opening a File for Writing by the Owner
276 The following example opens the file /tmp/file, either by creating it
277 (if it does not already exist), or by truncating its length to 0 (if it
278 does exist). In the former case, if the call creates a new file, the
279 access permission bits in the file mode of the file are set to permit
280 reading and writing by the owner, and to permit reading only by group
281 members and others.
282 If the call to open() is successful, the file is opened for writing.
284 #include <fcntl.h>
287 ...
288 int fd;
289 mode_t mode = S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH;
290 char *filename = "/tmp/file";
291 ...
292 fd = open(filename, O_WRONLY | O_CREAT | O_TRUNC, mode);
293 ...
294 Opening a File Using an Existence Check
296 The following example uses the open() function to try to create the
297 LOCKFILE file and open it for writing. Since the open() function speci‐
298 fies the O_EXCL flag, the call fails if the file already exists. In
299 that case, the program assumes that someone else is updating the pass‐
300 word file and exits.
301 #include <fcntl.h>
304 #include <stdio.h>
305 #include <stdlib.h>
306 #define LOCKFILE "/etc/ptmp"
309 ...
310 int pfd; /* Integer for file descriptor returned by open() call. */
311 ...
312 if ((pfd = open(LOCKFILE, O_WRONLY | O_CREAT | O_EXCL,
313 S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH)) == -1)
314 {
315 fprintf(stderr, "Cannot open /etc/ptmp. Try again later.\n");
316 exit(1);
317 }
318 ...
319 Opening a File for Writing
321 The following example opens a file for writing, creating the file if it
322 does not already exist. If the file does exist, the system truncates
323 the file to zero bytes.
324 #include <fcntl.h>
327 #include <stdio.h>
328 #include <stdlib.h>
329 #define LOCKFILE "/etc/ptmp"
332 ...
333 int pfd;
334 char filename[PATH_MAX+1];
335 ...
336 if ((pfd = open(filename, O_WRONLY | O_CREAT | O_TRUNC,
337 S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH)) == -1)
338 {
339 perror("Cannot open output file\n"); exit(1);
340 }
341 ...
342
344 None.
345
347 Except as specified in this volume of IEEE Std 1003.1-2001, the flags
348 allowed in oflag are not mutually-exclusive and any number of them may
349 be used simultaneously.
350 Some implementations permit opening FIFOs with O_RDWR. Since FIFOs
352 could be implemented in other ways, and since two file descriptors can
353 be used to the same effect, this possibility is left as undefined.
354 See getgroups() about the group of a newly created file.
356 The use of open() to create a regular file is preferable to the use of
358 creat(), because the latter is redundant and included only for histori‐
359 cal reasons.
360 The use of the O_TRUNC flag on FIFOs and directories (pipes cannot be
362 open()-ed) must be permissible without unexpected side effects (for
363 example, creat() on a FIFO must not remove data). Since terminal spe‐
364 cial files might have type-ahead data stored in the buffer, O_TRUNC
365 should not affect their content, particularly if a program that nor‐
366 mally opens a regular file should open the current controlling terminal
367 instead. Other file types, particularly implementation-defined ones,
368 are left implementation-defined.
369 IEEE Std 1003.1-2001 permits [EACCES] to be returned for conditions
371 other than those explicitly listed.
372 The O_NOCTTY flag was added to allow applications to avoid unintention‐
374 ally acquiring a controlling terminal as a side effect of opening a
375 terminal file. This volume of IEEE Std 1003.1-2001 does not specify how
376 a controlling terminal is acquired, but it allows an implementation to
377 provide this on open() if the O_NOCTTY flag is not set and other condi‐
378 tions specified in the Base Definitions volume of IEEE Std 1003.1-2001,
379 Chapter 11, General Terminal Interface are met. The O_NOCTTY flag is an
380 effective no-op if the file being opened is not a terminal device.
381 In historical implementations the value of O_RDONLY is zero. Because of
383 that, it is not possible to detect the presence of O_RDONLY and another
384 option. Future implementations should encode O_RDONLY and O_WRONLY as
385 bit flags so that:
386 O_RDONLY | O_WRONLY == O_RDWR
389 In general, the open() function follows the symbolic link if path names
391 a symbolic link. However, the open() function, when called with O_CREAT
392 and O_EXCL, is required to fail with [EEXIST] if path names an existing
393 symbolic link, even if the symbolic link refers to a nonexistent file.
394 This behavior is required so that privileged applications can create a
395 new file in a known location without the possibility that a symbolic
396 link might cause the file to be created in a different location.
397 For example, a privileged application that must create a file with a
399 predictable name in a user-writable directory, such as the user's home
400 directory, could be compromised if the user creates a symbolic link
401 with that name that refers to a nonexistent file in a system directory.
402 If the user can influence the contents of a file, the user could com‐
403 promise the system by creating a new system configuration or spool file
404 that would then be interpreted by the system. The test for a symbolic
405 link which refers to a nonexisting file must be atomic with the cre‐
406 ation of a new file.
407 The POSIX.1-1990 standard required that the group ID of a newly created
409 file be set to the group ID of its parent directory or to the effective
410 group ID of the creating process. FIPS 151-2 required that implementa‐
411 tions provide a way to have the group ID be set to the group ID of the
412 containing directory, but did not prohibit implementations also sup‐
413 porting a way to set the group ID to the effective group ID of the cre‐
414 ating process. Conforming applications should not assume which group ID
415 will be used. If it matters, an application can use chown() to set the
416 group ID after the file is created, or determine under what conditions
417 the implementation will set the desired group ID.
418
420 None.
421
423 chmod(), close(), creat(), dup(), fcntl(), lseek(), read(), umask(),
424 unlockpt(), write(), the Base Definitions volume of
425 IEEE Std 1003.1-2001, <fcntl.h>, <sys/stat.h>, <sys/types.h>
426
428 Portions of this text are reprinted and reproduced in electronic form
429 from IEEE Std 1003.1, 2003 Edition, Standard for Information Technology
430 -- Portable Operating System Interface (POSIX), The Open Group Base
431 Specifications Issue 6, Copyright (C) 2001-2003 by the Institute of
432 Electrical and Electronics Engineers, Inc and The Open Group. In the
433 event of any discrepancy between this version and the original IEEE and
434 The Open Group Standard, the original IEEE and The Open Group Standard
435 is the referee document. The original Standard can be obtained online
436 at http://www.opengroup.org/unix/online.html .
437IEEE/The Open Group 2003 OPEN(3P)